JH2005

Kasetsart Journal : N

atural Science A

pril - June 2005 Volum

e 39 Num

ber 2

April - June 2005

Volume 39 Number 2http://www.rdi.ku.ac.th

The Kasetsart Journal

Advisor : Samakkee Boonyawat

Rangsit Suwanketnikom

Editor-in-Chief : Ed Sarobol

Associate Editors : Wanchai Chanprasert, Natural Science

Suparp Chatraphorn, Social Science

Editorial Board : Natural Sciences Social Sciences

Amara Thongpan Suwanna Thuvachote

Pornsri Chairatanayuth Pongpan Trimongkholkul

Onanong Naivikul Matrini Ruktanonchai

Praparat Hormchan Nongnuch Sriussadaporn

Korchoke Chantawarangul Patana Sukprasert

Aree Thunyakijjanukij

Overseas Members

G. Baker (Mississippi State University, USA.)

A. Bruce Bishop (Utah State University, USA.)

John Hampton (Lincoln University, New Zealand)

Helen H. Keenan (University of Stathclyde, Scotland)

Chitochi Miki (Tokyo Institute of Technology, Japan)

Eiji Nawata (Kyoto University, Japan)

Manager : Orawan Wongwanich

Assistant Managers : Surai Suwannarat

Business Office : Kasetsart University Research and Development Institute (KURDI)

Kasetsart University, Chatuchak, Bangkok 10900.

The Kasetsart Journal is a publication of Kasetsart University intended to make available the results

of technical work in the natural and the social sciences. Articles are contributed by Kasetsart University faculty

members as well as by those from other institutions. The Kasetsart Journal : Natural Sciences edition is issued

four times per year in March, June, September and December while The Kasetsart Journal : Social Sciences

edition is issued twice a year in June and December.

Exchange publications should be addressed to

The Librarian,

Main Library,

Kasetsart University,

Bangkok 10900, Thailand.

KASETSART JOURNALNATURAL SCIENCE

The publication of Kasetsart University

VOLUME 39 April - June 2005 NUMBER 2

Response of Weeds and Yield of Dry Direct Seeded Rice to Tillage and Weed Management

................................................ Jagat Devi Ranjit and Rungsit Suwanketnikom 165

Screening and Selection for Physiological Characters Contributing to Salinity Tolerance in Rice

................... Duangjai Suriya-arunroj, Nopporn Supapoj, Apichart Vanavichit

.................................................................................... and Theerayut Toojinda 174

Weed Control Measures and Moisture Conservation Practices Effects on Seedbank Composition

and Vertical Distribution in the Soil

............... Girma Woldetsadik, Sombat Chinawong, Rungsit Suwanketnikom,

.............................................................Sunanta Juntakool and Visoot Verasan 186

Genetic Diversity of Elite and Exotic Oilseed Meadowfoam Germplasm using AFLP Markers

....................................................... Sureeporn Katengam and Steven J. Knapp 194

Effects of Gramma Radiation on Azuki Bean Weevil, Callosobruchus chinensis (L.)

...................... Jakarpong Supawan, Praparat Hormchan, Manon Sutantawong

............................................................................... and Arunee Wongpiyasatid 206

Occurrence and Distribution of Major Seedborne Fungi Associated with Phaseolus Bean

Seeds in Ethiopia

....................................................... Mohammed Yesuf and Somsiri Sangchote 216

Short-Term Stressor Effects of Water Deprivation Prior to the Onset of Lay on Subsequent

Reproductive Performance of ISA Brown Pullets

............................................. Nirat Gongruttananun and Ratana Chotesangasa 226

Pharmacokinetics and Withdrawal Times of Enrofloxacin in Ducks

................... Natthasit Tansakul, Amnart Poapolathep, Naruamol Klangkaew,

.............................................. Napasorn Phaochoosak and Wanida Passudaruk 235

Antimicrobial Resistance of Campylobacter jejuni Isolated from Chicken in Nakhon Pathom

Province, Thailand

.............. Jananya Sukhapesna, Patamaporn Amavisit, Worawidh Wajjwalku,

.......................................Arinthip Thamchaipenet and Thavajchai Sukpuaram 240

Hematology, Cytochemistry and Ultrastructure of Blood Cells in Asiatic Black Bear

(Ursus thibetanus)

......................... Chaleow Salakij, Jarernsak Salakij, Nual-Anong Narkkong,

............................................ Ludda Trongwonsa and Rattapan Pattanarangsan 247

Probiotic Properties of Bacillus pumilus, Bacillus sphaericus and Bacillus subtilis

in Black Tiger Shrimp (Penaeus monodon Fabricius) Culture

.......... Watchariya Purivirojkul, Monchan Maketon and Nontawith Areechon 262

Extracts of Thai Indigenous Vegetables as Rancid Inhibitor in a Model System

......................................... Plernchai Tangkanakul, Gassinee Trakoontivakorn

....................................................................... and Chansuda Jariyavattanavijit 274

Screening and Characterization of Lactic Acid Bacteria Producing Antimicrobial Substance

against Staphylococcus aureus

........ Chatinan Ratanapibulsawat, Pumrussiri Kroujkaew, Ohmomo Sadahiro

................................................................................... and Sunee Nitisinprasert 284

Studies on Nham-Pla’s Processing by Using Rock Salt and Solar Salt

. Mathana Sangjindavong, Pranisa Chuapoehuk and Daungdoen Vareevanich 294

Product Development of Crocodile Jerky

........... Sinee Nongtaodum, Nongnuch Raksakulthai and Mayuree Chaiyawat 300

Utilization of Fish Flour in Canned Concentrated Seasoning Stock for Thai Foods Preparation

......... Plernchai Tangkanakul, Payom Auttaviboonkul, Patcharee Tungtrakul,

....................... Mantana Ruamrux Chidchom Hiraga, Kanjanarat Thaveesook

................................................................................... and Montatip Yunchalad 308

Lightning Surge Response of Concrete Pole due to Effect of the Electrical Properties

of Concrete based on the Electromagnetic Field Method

....................................................... Samroeng Hintamai and Jamnarn Hokierti 319

Kasetsart J. (Nat. Sci.) 39 : 165 - 173 (2005)

Response of Weeds and Yield of Dry Direct Seeded Rice to Tillageand Weed Management

Jagat Devi Ranjit1 and Rungsit Suwanketnikom2

ABSTRACT

The study was initiated to assess the performance of rice (Oryza sativa) under dry direct seeded

environment with two tillage systems of conventional tillage and minimum tillage and five weed

management treatments namely unweeded control, handweeding twice 25 and 45 days after seeding,

anilophos + one handweeding, bispyribac-sodium, and straw mulch + bispyribac-sodium as an alternate

method of transplanting in the mid-hill ecology. Both anilophos and bispyribac–sodium were found to

reduce narrowleaf and broadleaf weeds compared to unweeded control. However, anilophos reduced

Cyperus difformis, C. sanguinolentus, and C. iria 4 weeks after seeding (WAS) but not Ammania sp. and

Dopatrium junceum 8 WAS. Bispyribac–sodium and straw mulch + bispyribac-sodium reduced the

population of Alternanthera philoxeroides, Ammania sp., Commelina diffusa, C. difformis, C. iria, and

D. junceum 8 WAS. No phytotoxic effect on the rice plants was observed due to both herbicides. Yield

and yield attributes were not affected by the tillage systems. The weed managements were found to affect

the numbers of tiller per square meter and grain yield. The increasing number of weed did not affect the

plant height of rice (Khumal-4). The numbers of tiller and grain yield highly affected the increasing

number of weed population. Anilophos plus one handweeding, straw mulch plus bispyribac-sodium,

handweeded twice and bispyribac–sodium alone gave higher yield compared to unweeded control.

Promising grain yield could be achieved with the anilophos or bispyribac-sodium with additional physical

or mechanical control methods in dry direct seeded rice.

Key words: dry direct seeded rice, bispyribac-sodium, anilophos, tillage, weed flora

INTRODUCTION

Transplanting is the popular rice

establishment practice throughout Nepal with very

little in direct seeding in some pocket areas. But

with the ascending problem of labor and time,

alternate method of rice culture may be beneficial

in the future. However, direct seeding will be an

alternate option to transplanting. Puddling for rice

transplanting also makes land preparation difficult

for wheat crop in rice–wheat rotation resulting in

1 Agronomy Division, Nepal Agriculture Research Council, GPO 404, Kathmandu, Nepal.2 Department of Agronomy, Faculty of Agriculture, Kasetsart University, Bangkok 10900, Thailand.

Received date : 06/07/04 Accepted date : 01/03/05

cloddy soil structure, loss of soil moisture, delayed

and inadequate seed soil contact (Sharma and De

Datta, 1985). Weeds are one of the limiting factors

in direct seeded rice in reducing the yield. Weeds

account for 50-80% yield reduction in rainfed

uplands (Ranjit et al., 1989; Sinha et al., 1996).

Yield reduction in rice is even higher (97%) due to

competition of Echinochloa crusgalli (Kurchania

et al., 1991). However, Echinochloa spp. was

reported to be more competitive causing greater

loss in growth and yield of rice compared to C.

166 Kasetsart J. (Nat. Sci.) 39 (2)

difformis, Eclipta alba, Marsilia minuta, and

Paspalum distichum (Srinivasan and Palaniappon,

1994). The yield losses caused by different weeds

depends on the type of rice culture, weed infestation,

density and weed species prevalence.

Hand weeding is the most popular method

of weed management in Nepal as well as in many

parts of the world. Besides hand pulling and hand

weeding, a number of herbicides have been

developed and tested for the direct seeded rice

around the world. Herbicides such as butachlor,

thiobencarb, pendimethalin, oxyfluorfen, propanil,

quinclorac, ioxynil, 2,4-D, piperophos +

sulfonylurea, bentazone, molinate and bispyribac-

sodium have been tested in direct seeded rice in the

past research (Biswas et al., 1992; Chin, 1999;

Crawford and Jordan, 1995; Im, et al., 1999;

Ranjit et al., 1989). Many factors affected cause

the change of weed communities. Weed flora in

the rainfed ecosystem has been reported to be the

most complex compared to irrigated rice, but the

weed management is the most important and can

fill up at least 15% yield gap in different growing

conditions (Moody, 1982). This study aimed to

assess the responses of weed and yield attributes of

dry direct seeded rice to tillage and weed

management with bispyribac-sodium and

anilophos herbicide and straw mulch in the mid-

hill ecology.

MATERIALS AND METHODS

This experiment was conducted in the

lowland field at Agronomy farm, Khumaltar, Nepal

in a split plot design with RCBD replicated 4 times

during the summer season of 2002. The main plots

and sub plots were compiled of tillage and weed

management respectively. The plot size was 4m ¥5m (20m2) and row spacing 20 cm. The field was

located at an elevation of 1360 m above mean sea

level on 27∞ 40’N latitude and 85∞ 20’E longitude.

Land preparation was done with 2 ploughing

and 2 harrowing in case of conventional tillage

(CT). But, for minimum tillage (MT), about 5-7

cm deep ploughing (only one time) was undertaken

by the Chinese Seed Drill. Rice seeding was done

after wheat harvest.

Planting was carried out after making a line

with hand hoe for both tillage systems. Planting

and harvesting were conducted in June and October

respectively.

The variety used was Khumal-4. Seed rate

was 90 kg/ha. Chemical fertilizer was applied at

100 kg nitrogen, 50 kg phosphorus, and 30 kg

potash per hectare. Nitrogen was splitted in two

halves. The 1st half was given as basal dose during

planting and 2nd half during top dressing 45 days

after planting. Chopped rice straw at 4 t/ha was

used for the mulch treatment one day after rice

seeding.

Weed count was initiated from 0.50 m2

placing 50 cm by 50 cm quadrat at 2 places in each

plot. Weed count was performed 3 times first 4

weeks after rice seeding (WAS), the second one 8

WAS and the 3rd at milking stage of rice (MSR).

The first and second weed counts were carried out

from the same spots in each plot but the third count

was done from the different spot in each plot to see

the changes in weed flora during the reproductive

stage of rice. Individual weed species was counted.

Weeds were pulled during the second and third

counts and biomass was recorded after separating

and cutting the roots of the narrowleaf and broadleaf

weeds.

Chopped rice straw @ 4t/ha was used for

the mulch treatment one day after rice seeding.

There were 5 weed management treatments namely

unweeded control (W1), twice hand weeding 25

and 45 days after sowing (DAS) (W2),

preemergence application of anilophos [S[N(4-

chloro-phenyl-)-N-isopropyl-carbamoyl-methyl-]-

o, o-dimethyl-dithiophosphate, trade name =

Arozin“ 30EC] @ 0.4kg ai/ha (W3),

postemergence application of bispyribac-sodium

[ 2,6-bis{(4,6-dimethoxypyrimidin-2-yl)

oxy}benzoate, trade name = Nominee“10 EC] @

Kasetsart J. (Nat. Sci.) 39 (2) 167

50 g ai (W4), and rice straw mulch @ 4 t/ha +

postemergence application of bispyribac-sodium

@ 40 g ai /ha 40 days after seeding (DAS) (W5).

Bispyribac-sodium was applied after

mixing with 1/1 v/v surfactant. Anilophos was

applied one day after rice sowing. Aspee backpack

sprayer with 4 flat fan nozzles (8002) was used for

herbicide spray. The spray volume was 500 l/ha.

The weather during herbicide spray was sunny sky

with patches of cloud and mild wind.

Plant height (cm), tillers per square meter,

seeds per panicle, thousand seed weight (g) and

grain yield (kg/ha) were recorded. Plant height

was recorded from the averages of 5 plant in each

plot. Tillers were recorded from one square meter

in each plot. Harvesting was done from 9.60 square

meter (3m ¥ 3.20m). Grain yield was adjusted at

14 percent moisture contents.

The mean minimum temperature during

the rice crop ranged from 20.3∞C (June) to 12.8∞C(October) and the maximum temperature ranged

from 28.5∞C (June) to 24.7∞C (October). The total

rainfall was 993.5 mm from June to October. The

percent soil moisture during the rice crop was 40-

53.

RESULTS AND DISCUSSION

Weeds were categorized in narrowleaf

(grass and sedge), broadleaf (monocot and dicot)

and pteridophyte. The important species were A.

philoxeroides, C. diffusa, C. difformis, C. iria, C.

sanguinolentus, Ceratopteris thalictroides, E.

colona, F. miliacea, Lindernia procumbens, and

P. distichum (Table1).

Response of weeds and weed biomass to tillageand management

Both narrowleaf and broadleaf weeds and

their biomass were found not to be different due to

tillage in all counts. However, P. distichum was

noticed to be more in minimum tillage 8 WAS and

Table 1 Weed species recorded in the experimental field at different stages of direct seeded rice.

Weeds species Family Weeds species Family

Narrowleaf weeds : Alternanthera philoxeroides (Mart) Griseb. Amaranthaceae

Ammania baccifera L. Lythraceae

Cynodon dactylon L. Pers Poaceae Dopatrium junceum Hamilt. Scrophulariaceae

Cyperus difformis L. Cyperaceae Lindernia procumbens Philcox Scrophulariaceae

C. dilutus L Cyperaceae Polygonum hydropiper L. Polygonaceae

C. iria L Cyperaceae Rotola indica Koehne Lythraceae

C. sanguinolentus Vahl. Cyperaceae Rorrippa indica Brassicaceae

Echinochloa colona L. (Link) Poaceae Vandellia angustifolia Benth. Scrophulariaceae

E. crusgalli (L) P. Beauv. Poaceae

Eriocaulan sp. Eriocaulaceae Broadleaf weeds (Monocot) :

Eriocaulan sieboldtianum Sieb.et Zucc Eriocaulaceae

Fimbristylis miliacea Vahl. Cyperaceae Commelina diffusa Burm.f Commelinaceae

Paspalum distichum L. Poaceae Murdania sp. Commelinaceae

Panium sp. Poaceae Monochoria vaginalis Presl. Pontederiaceae

Scirpus juncoides Roxb. Cyperaceae Sagittaria guayanensis H.B.K Alismataceae

Broadleaf weeds (Dicot) : Pteridophyte :

Ageratum conyzoides L. Asteraceae Ceratopteris thalictroides (L) Brongn Parkeriaceae

Eclipta prostrata L. Asteraceae

Erigeron sp. Asteraceae


MSR although the population was not high. But

the number of C. sanguinolentus was less in

minimum tillage 8 WAS (Figure 1).

It has been reported that different tillage

systems have different rates of weed suppression.

Reduced tillage (one round tillage + leveling)

resulted in heavy infestation of F. miliacea. But

conventional tillage increased the amount of M.

vaginalis. A seeding rate of 100 kg/ha significantly

reduced sedges and broadleaf biomass 60 days

after planting but not E. crusgalli (Azmi and

Mortimer, 1999). However, the tillage did not

affect the total populations of weed in the study. It

might need a few years to see the change in weed

population and species (Table 2).

Total number of weed was different due to

weed management treatments in all counts except

broadleaf weeds 4 WAS. Preemergence application

of anilophos + handweeding, straw mulch +

bispyribac-sodium and bispyribac-sodium alone

(a) Cyperus sanguinolentus

0

5

10

15

20

25

4 WAS 8 WAS MSR

Plan

ts/0

.50

m2

CT

MT

a

b

(b) Paspalum distichum

0

5

10

15

4 WAS 8 WAS MSR

Plan

ts/0

.50

m2

CT

MTb

aa

b

Figure 1 Weed species response to conventional

tillage (CT) and minimum tillage (MT)

at 4 WAS, 8 WAS, and MSR. Values in

the bars with the same letters above are

not significantly different at the 0.05

level. Bars without letter are not signifi-

cantly different.

reduced the narrowleaf weeds in all counts (Table

2). Among the individual species, anilophos and

straw mulch suppressed more amount of C.

difformis and C. sanguinolentus than the unweeded

control 4 WAS (Figure 2). All weed control

treatments were found to suppress both narrowleaf

and broadleaf weeds 8 WAS, and MSR (Table 2).

Post-emergence application of bispyribac-

sodium alone and straw mulch + bispyribac-sodium

were found to suppress both types of weed namely

Ammania sp. and D. junceum, but both of them

were not suppressed by anilophos when compared

to the unweeded control 8 WAS (Figure 2). The

number of these weeds decreased during the

maturity stage of rice. Weeds like Cyperus spp., C.

diffusa, and D. junceum were suppressed by

bispyribac-sodium alone and in combination with

straw mulch (Figure 2).

Narrowleaf and broadleaf weed biomasses

were significantly different due to weed

management 8 WAS and MSR. Higher weed

biomass was recorded in the unweeded control.

The rest of the weed management treatments

lowered the weed biomass in the same range. The

herbicide application was also equally effective as

twice handweeding (Table 3).

Earlier researches also reported that

bispyribac–sodium controlled many narrowleaf

and broadleaf weeds such as C. diffusa, C. iria, E.

crusgalli, Fimbristyllis spp., Leersia oryzoides,

Murdania sp., P. distichum, Polygonum sp.,

Sagittaria spp., Scirpus spp., and Sphenoclea

zeylanica (Han, 2001; Kobayashi et al., 1995;

Shinohara et al., 1994; Tachikawa et al., 1997;

Yokohama et al., 1993). A. philoxiroides,

Aeschenomene indica, Ammania coccinea, and

Heteranthera limosa were also controlled by KIH

2023 (bispyribac–soduim) (Braverman and Jordan,

1996). Anilophos + ethoxysulfuron or anilophos

alone controlled the most dominant weed Cyperus

sp., F. miliacia and also Saccolepis interrupta

(Moorthy et al., 1999; Screedevi and Thomas,

1993).


(a) Cyperus difformis

0

50

100

150

4WAS 8WAS MSR

Plan

ts/0

.50m

2

W1

W2

W3

W4

W5

a a

b

ab

ba a

b a b

(b) Cyperus iria

0

5

10

15

20

4WAS 8WAS MSR

Plan

ts/0

.50m

2

W1

W2

W3

W4

W5

a

b bb

b ab b a

b

(c) Cyperus sanguinolentus

0

10

20

30

40

50

4WAS 8WAS MSR

Plan

ts/0

.502

W1

W2

W3

W4

W5a a ba

b

a

b

c

bc

bc

a

bb b b

(d) Commelina diffusa

0

5

10

15

20

25

4WAS 8WAS MSR

Plan

ts/0

.50m

2

W1

W2

W3

W4

W5

a

bb b b

a

b b b b

Table 2 Narrowleaf and broadleaf weeds as affected by tillage and weed management.

Treatment Narrowleaf weeds Broadleaf weeds

4 WAS 8 WAS MSR 4 WAS 8 WAS MSR

------------------------ (Plants/0.50 m2)------------------------

Tillage :

Conventional tillage (CT) 921 65 31 23 45 11

Minimum tillage (MT) 145 79 43 36 45 11

Weed management :

Unweeded control (W1) 172 a 117 a 58 a 27 25 c 30 a

Handweeding twice (W2) 161 a 90 ab 48 ab 32 56 b 7 b

Anilophos + handweeding one (W3) 63 b 59 bc 34 bc 28 107 a 7 b

Bisbyribac-sodium (W4) 127 ab 55 c 21 c 35 12 c 5 b

Straw mulch + bisbyribac-sodium (W5) 70 b 40 c 23 c 24 25 c 7 b

Tillage (T) NS2 NS NS NS NS NS

Weed management (W) ** ** ** NS ** **

T x W NS NS * NS NS NS

1 Means within the same column and grouping followed by the same letters are not different according to Fisher’s protected test

P=0.05.2 Treatment effects and interactions were significant at 5% (*), significant at 1% (**) or nonsignificant (NS).

Figure 2 Weed species responses to different weed managements of W1 (unweeded control), W2

(handweeded twice), W3 (anilophos + one weeding), W4 (bispyribac-sodium), and W5 (straw

mulch + bispyribac-sodium) at 4 WAS, 8 WAS, and MSR. Values in the bars with the same

letters above are not significantly different at 0.05 level. Bars without letters are not

significantly different.


Response of yield attributes of rice to tillageThere were no significant differences on

plant height, tillers per square meter, thousand

seed weight and grain yield due to tillage. It

showed that dry direct seeding rice in conventional

and minimum tillage did not affect the yield

attributes and could be planted in both tillage

systems (Table 4). Hobbs et al. (2002) also reported

that rice yield was in the same range in both

puddled and unpuddled rice cultures. This might

be due to the condition under the unpuddled rice

culture where the weeds were more properly

controlled, since, in general, weeds in unpuddled

rice culture were more serious problem than in

puddled rice culture.

Response of yields attributes to weedmanagement

Tillers per square meter, grain yield and

dry straw weight were significantly different due

to weed management, but not plant height, number

of seeds per panicle, and thousand seed weight. It

showed that higher numbers of weed did not affect

plant height because the plant height in other weed

management treatments was in the same range of

the unweeded control. Number of tillers per square

meter ranged from 205 in unweeded control to 335

in straw mulch + bispyribac-sodium. Higher yield

(6,708 kg/ha) was recorded in handweeding twice,

straw mulch + bispyribac-sodium (6,445 kg/ha),

and anilophos + one handweeding (6,416 kg/ha)

which were at par to each other. Bispyribac-sodium

alone yielded 5,469 kg/ha which was higher than

that in unweeded control (2,136 kg/ha) (Table 4).

All weed management treatments except unweeded

control in this experiment gave promising yields

up to 670 kg/ha. In the study, both herbicides did

not show any phytotoxic effect on rice plants.

However, the phytotoxic effect of these herbicides

on different agroecological rice cultivars needs to

Table 3 Effects of weed management on dry weed biomass at different stages of rice.

8 Weeks after sowing (WAS) Maturity stage of rice (MSR)

Treatments Narrowleaf Broadleaf Narrowleaf Broadleaf

--------------------------------(g/0.50 m2)------------------------------

Tillage :

Conventional tillage (CT) 35.61 5.8 30.5 13.4

Minimum tillage (MT) 48.7 8.2 43.9 13.7

Weed management :

Unweeded control (W1) 113.1 a 18.1 a 88.3a 50.8 a

Handweeding twice (W2) 20.9 b 4.6 b 11.1 b 2.4 b

Anilophos + handweeding one (W3) 14.5 b 4.5 b 24.4 b 3.7 b

Bisbyribac-sodium (W4) 38.4 b 2.3 b 35.5 b 5.3 b

Straw mulch + bisbyribac- 24.0 b 5.6 b 26.8 b 5.5 b

sodium (W5)

Tillage (T) NS2 NS NS NS

Weed management (W) ** ** ** **

T ¥ W NS NS NS NS




be assessed in the future. The rotational effect of

these herbicides to wheat herbicides should be

studied in depth in different agroecological

environments to find the effect on crop and weed

shifts in the future.

With the increasing number of narrowleaf

weed population, both tillers per square meter and

grain yield decreased (Table 2, 4, and Figure 3). In

this study Cyperus spp. were the found to be

dominant narrowleaf weed. Broadleaf weed like

D. junceum did not affect the rice yields (Figure 3).

Because the yield in anilophos treatment was

higher, even the broadleaf weeds was not

suppressed. The yield reduction might be depended

on weed species.

However, the low yield in bispyribac-

sodium alone compared to other treatments W2,

W3 and W5 was actually not known although it

suppressed both narrowleaf and broadleaf weeds

(Figure 3). This herbicide might need to be assessed

with regard to time, rate and the cultivar in different

agroecological environments for more seasons.

CONCLUSION

Most common weeds associated with dry

direct seeded rice were A. philoxiroides, C.

difformis, C. iria, C. sanguinolentus, C. diffusa, D.

junceum, E. colona, and Lindernia sp. Both

narrowleaf and broadleaf weeds were not

significantly different due to tillage, but was

significantly different due to weed management.

Both narrowleaf and broadleaf weeds were reduced

by bispyribac-sodium. Weeds like A. philoxiroides,

Cyperus spp., and D. junceum were significantly

reduced. However, Ammania sp. and D. junceum

were not suppressed by anilophos. No phytotoxic

effect on rice plants has been observed due to both

herbicides. This study showed that both herbicides

could be applied in dry direct seeded rice culture in

the mid hill ecology. The weed managements

showed significant impact on tillers and grain

Table 4 Effects of tillage and weed management on plant height, tillers, seeds/panicle, thousand seed

weight, grain yield and dry straw weight of dry direct seeded rice.

Treatments Plant Tiller Seed/panicle 1000 Grain Straw

height Filled Unfilled seed wt. yield biomass

(cm) (no./ m2) ----(no./panicle)---- (g) (kg/ha) (kg/ha)

Tillage :

Conventional tillage (CT) 127.61 281 146 12 18.7 5630 7395

Minimum tillage (MT) 127.8 256 126 11 19.4 5239 6432

Weed management :

Unweeded control (W1) 127.4 205c 106 11 19.1 2136c 3989b

Handweeding twice (W2) 127.2 258ab 138 11 19.3 6708a 7701a

Anilophos + handweeding one (W3) 128.5 277ab 170 11 18.9 6416a 7541a

Bisbyribac-sodium (W4) 127.5 270abc 128 10 19.2 5469b 7195a

Straw mulch + bisbyribac-sodium (W5) 128.0 335a 139 15 18.8 6445a 8140a

Tillage (T) NS2 NS NS NS NS NS NS

Weed management (W) NS NS * NS NS ** **

T ¥ W NS NS NS NS NS NS NS




Figure 3 Grain yields of rice as affected by narrow leaf (NL) and broadleaf weed (BL) under different

weed managements of W1 of (unweeded control), W2 (handweeded twice), W3 (anilophos +

1 weeding), W4 (bispyribac-sodium), and W5 (straw mulch + bispyribac-sodium) 8 WAS and

MSR.

8 WAS

0

2000

4000

6000

8000

W1 W2 W3 W4 W5

Yie

ld k

g/ha

0

50

100

150

Plan

ts/0

.50

m2

YL

NL

0.05

0.05

8 WAS

0

2000

4000

6000

8000

W1 W2 W3 W4 W5

Yie

ld k

g/ha

0

50

100

150

Plan

ts/0

.50

m2

YL

BL0.05

0.05

MSR

0

2000

4000

6000

8000

W1 W2 W3 W4 W5

Yie

ld k

g/ha

0

10

20

30

40

Plan

ts/0

.50

m2

YL

BL 0.05

0.05

LSD= 593

LSD= 9

MSR

0

2000

4000

6000

8000

W1 W2 W3 W4 W5

Yie

ld k

g/ha

0

20

40

60

80

Plan

ts/0

.50

m2

YL

NL

0.05

0.05

LSD= 593

LSD= 16

LSD= 593

LSD= 32

LSD= 593

LSD= 27

yield. With the increasing number of weed

population, the numbers of tiller and grain yields

decreased. All weed management gave comparable

yields to twice handweeding. With the proper

weed management 150-200 percent rice yield could

be increased so that the drudgery operation like

seedbed preparation and transplanting could be

avoided in dry direct seeded rice culture.

ACKNOWLEDGEMENTS

The authors would like to give their sincere

thanks to Chief and the staffs of the Agronomy

Division for providing the field and other facilities

to conduct this research. We are thankful to Dr. P.

R. Hobbs, former regional agronomist CIMMYT/

Nepal, Rice – Wheat Consortium New Delhi and

Kumiai Chemical Company, Japan for their

cooperation in providing the rice herbicides for

this research. Nepal Agricultural Research Council,

Nepal supported this program.

LITERATURE CITED

Azmi, M.and A.M. Mortimer. 1999. Effect of

tillage practices, seeding rates and herbicides

on weed infestations in direct seeded rice, pp.

199-204. In Proc. 17th Asian Pacific WeedScience Society Conference, Bangkok,

Thailand.

Biswas, J.C., S.A. Sattar, and S.B. Siddique. 1992.

Evaluation of herbicides in direct seeded rice

in Bangladesh. Bangladesh Rice Journal.2(1-2): 40-43.

Braverman, M.P. and D.L Jordan. 1996. Efficacy

of KIH-2023 in dry–and water seeded rice


(Oryza sativa). Weed Tech. 10: 876-882.Chin, D.V. 1999. Bispyribac-sodium, a new

selective rice herbicide in direct seeded rice inVietnam, pp. 443-446. In Proc. 17th AsianPacific Weed Science Society Conference.Bangkok, Thailand.

Crawford, S.H. and D.L. Jordan. 1995. Comparisonof single and multiple applications of propaniland residual herbicides in dry seeded rice(Oryza sativa). Weed Tech. 9(1): 153-157.

Han, F.C. 2001. Effect of Nominee (Bispyribac-sodium) on rice cutgrass (Leersia oryzoides)in water seeded rice in Heilongjiang, China,pp. 787-792. In Proc. of 18th Asian PacificWeed Science Society Conference, Beijing,China.

Hobbs, P.R., Y. Singh, G.S. Giri, J.G. Lauren, andJ.M. Duxbury. 2002. Direct seeding andreduced tillage options in the rice-wheatsystems of the Indo-Gangetic plains of SouthAsia, pp. 201-215. In Pandey, S., M. Mortimer,I. Wade, T.P. Tuong, K. Lopez and B. Hardy(eds). Direct seeding: research issues andopportunities. In Proc. of the InternationalWorkshop on Direct Seeding in Asian RiceSystems: Strategic Research Issues andOpportunities, Bangkok, Thailand,International Rice Research Institute, LosBanos, Philippines.

Im, I.B., C.K. Kang, S.S. Han, and S.Y. Cho, 1999.Weed control by weed emergence types in drydirect seeded rice fields. Korean Journal ofWeed Science. 19(1): 7-14.

Kobayashi, K., M. Yokohama, O. Watanabe, H.Sadohara, and N. Wada. 1995. KIH 2023 anew post emergence herbicide in rice (Oryzasativa), pp. 221-226. In Proc. 15th AsianPacific Weed Science Society Conference.Tsukuba, Japan.

Kurchania, S.P., J.P. Tiwari, and N.R. Paradkar.1991. Weed control in rice (Oryza sativa)–wheat (Triticum sativum) cropping system.Indian Jour. Agric. Sci. 61(10): 720-725.

Moody, K. 1982. The status of weed control in rice

in Asia. FAO Plant Protection Bulletin. 30:1-10.

Moorthy, B.T.S., S. Saha and S. Sanjoy. 1999.Relative efficacy of different herbicides forweed control in direct seeded rice in puddledsoil. Indian Jour. of Weed Sci. 31(3-4): 210-213.

Ranjit, J.D., K.P. Bhurer, K.P. Koirala, Y. Thakur,and D.N. Choudhary. 1989. Screening ofherbicides in upland and transplanted rice, pp.129-139. In Proc. 14th Summer cropsworkshop. Parwanipur, Nepal.

Screedevi, P. and C.G. Thomas. 1993. Control ofSaccolepis interrupta (Wild) Stapf in dryseeded rice in Kerela, pp.10-12. In Proc.Indian Society of Weed ScienceInternational Symposium. India.

Sharma, P.K. and S.K. De Datta. 1985. Effects ofpuddling on soil physical properties andprocesses. In Soil Physics and Rice. IRRIpublications.

Shinohara, T., M Yokoyama, O. Watanabe, K.Kawano, and S. Shigematsu. 1994. KIH 2023,A new post –emergence herbicide in rice, p 8.In Proc. Weed Science Society of America,Abstract.

Sinha, P.K., C.V. Singh, R.K. Mishra, D. Maiti,V.D. Shukla and M. Variar. 1996. Rainfedupland rice–future strategies. IndianFarming. 66(9): 25-29.

Srinivasan, G. and S. Palaniappon. 1994. Effect ofmajor weed species on growth and yield ofrice. Indian Jour. Agron. 69(1): 13-15.

Tachikawa, S., T. Miyazawa, and H. Sadohara.1997. Vegetation management by KIH-2023in rice levees and highways and railroad right-of-ways, pp. 114-117. In Proc. 16th AsianPacific Weed Science Society Conference.Kualalumpur, Malaysia.

Yokohama, M.,O. Watanabe, K. Kawano, S.Shigematsu and N. Wada. 1993. KIH-2023,A new post-emergence herbicide in rice pp.61-66. In Brighton Crop ProtectionConference-Weeds.


Screening and Selection for Physiological CharactersContributing to Salinity Tolerance in Rice

Duangjai Suriya-arunroj1 ,Nopporn Supapoj1, Apichart Vanavichit2

and Theerayut Toojinda3

ABSTRACT

Two screening techniques, one initial screening using nutrient solution culture at young seedling

stage and the other using soil in pots at vegetative stage, were used for identification of genotypes for

salinity tolerance. Sixteen rice lines and cultivars were screened initially at seedling stage. Germinated

seeds were placed on styrofoam plates floated on nutrient solution in a plastic container. After 14 days

of sowing, the seedlings were subjected to salinization. The results showed that there were 3 groups of

rice with different levels of response to salinity ; tolerant group consisted of Pokkali, FL496 and FL530,

moderately tolerant group consisted of FL358, FL367, FL411, FL416, FL434, FL443, FL478, FL523,

FL563, KMK and DDG, and susceptible group were KDML105 and RD6. To confirm the reliability of

this initial screening technique, the visual salt-injury symptom was compared with the mean performance

of salinity damage rating, Na+, K+ content and Na+/K+ absorption ratio in young leaves , old leaves and

stem to identify physiological characters contributing to salinity tolerance in rice in the vegetative

experiment. Eight lines/cultivars of rice were selected from the seedling screening to further investigate

tolerant ability at vegetative stage in soil medium. In this test, 21 day-old seedling were subjected to 3

levels of salinity, 4,8 and 12 dS /m. The results showed visual scores to match with the Na+/K+ ratio, the

cultivars with low Na+/K+ ratio had high tolerant ability and the susceptible one had high Na+/K+ ratio.

The two selected salt tolerant recombinant inbred lines were FL496 and FL478 and two landrace cultivars

selected were DDG and KMK which will be used as donors to introgress salt tolerant QTL into target

cultivars (KDML105 and RD6) in salt tolerance breeding program.

Key words: rice, salt tolerant rice, salt tolerance screening, Na+/K+ ratio

INTRODUCTION

Soil salinity is the single most widespread

soil toxicity problem in rice growing countries.

Thus, an increase in salinity resistance in rice is

necessary for further expansion of rice growing

area because good agricultural land is limited

1 Ubon Ratchathani Rice Research Center, Ubon Ratchathani 34000, Thailand.2 Rice Genome Project, Kasetsart University Khampaeng Saen Campus, Nakhon Pathom 74130, Thailand.3 BIOTEC, National Center for Genetic Engineering and Biotechnology. Kasetsart University, Khampaeng Saen Campus,

Nakhon Prathom 74130, Thailand.


(Toenniessen, 1984). Breeding for salinity

tolerance in rice requires reliable screening

techniques and must be rapid to keep pace with a

large amount of breeding materials. Salt sensitivity

of rice varies not only among genotypes but also

among developmental stages of the plant (Akbar

and Yabuno, 1974). According to Pearson and

Ayers (1960), rice is very tolerant to salt during

germination, but very sensitive during the early

seedling stages. Once panicles have developed in

the leaf sheaths, subsequent phases of development

are not sensitive to salt (Kaddah et al., 1973).

Two screening techniques were developed

(Gregorio et al., 1997) for use at seedling stage

and vegetative and reproductive stages. In this

study the second techniques used was only for

vegetative stage. The two screening techniques

adopted were 1. screening at seedling stage in

which salinity was imposed to medium solution 14

days after growing in the solution and 2. screening

at vegetative stage where the rice plants in pots

were salinized 21 days after sowing. The purpose

of the latter was to confirm the result of the first

screening and also to identify physiological traits

associated with salinity tolerance. Physiological

mechanisms confering Na+ exclusion and

selectivity for K+ and Na+ have been described for

salt tolerant ability of plant. Thus the visual salt-

injury symptoms were compared with Na+, K+

content and Na+/K+ ratio in young leaf, old leaf

and stem of selected lines and cultivars. Rice

plants maintain the leaf water content to avoid the

injury from drought stress which is the early phase

of salt stress (Munns,1993) or osmotic phase. K+

and organic solutes accumulate in the cytoplasm

and organelles to balance the osmotic pressure of

ions in the vacuoles (Munns and James, 2003).

Mechanisms of salt tolerance at the cellular level

involve keeping the salt out of cytoplasm and

sequestering it in the vacuoles of the cell. The

objectives of this study were to elucidate how

some rice cultivars had an ability to grow in

salinity conditions, with particular emphasis on

phenological development, leaf injury (salt tolerant

scoring), dry matter growth, plant height, Na+ and

K+ concentration in plant parts and relative water

content, to identify physiological traits for salinity

tolerance and to select salt tolerant cultivars/lines.

The information obtained from this work would

assist in identifying traits which could be used as

selection criteria for salt tolerance in rice.


1. Tested materialsAt seedling stage, 16 rice lines/cultivars

were used; IR66946-3R-58-1-1(FL358), IR66946-

3R-67-1-1(FL367), IR66946-3R-111-1-1

(FL411), IR66946-3R-116-1-1(FL416), IR66946-

3R-134-1-1(FL434), IR66946-3R-143-1-1

(FL443), IR66946-3R-178-1-1(FL478), IR66946-

3R-196-1-1(FL496), IR66946-3R-223-1-1

(FL523), IR66946-3R-230-1-1(FL530), IR66946-

3R-263-1-1(FL563) (the progenies of the cross

Pokkali/IR29 which were selected based on

different salinity tolerant abilities classified by

Gregorio (1997), Khao Mahk Khaek (KMK),

Daeng Dawk Gok (DDG), RD6, Khao Dawk

Mali105 (KDML105)(Thai landrace cultivars),

and Pokkali(tolerant cultivar). Since KDML105

and RD6 were moderately sensitive to salt (score

7) so these cultivars were used as susceptible

checks. The lines and cultivars used for vegetative

stage screening were selected from the lines or

cultivar found to vary on tolerance to salinity

during seedling stage. The 8 lines and cultivars

selected were Pokkali (tolerant check), IR29

(susceptible check), KMK, KDML105, FL478,

FL496, DDG, and RD6.

2. Screening at seedling stageSeeds of the 16 lines/cultivar were surface

sterilized with 10% Clorox (5.25%w/w sodium

hypochlorite) for 30 minutes, then rinsed with

distilled water. Sterilized seeds were incubated in

petridishes at room temperature (38-42∞C) for 5-7

days to germinate. Germinated seeds were placed

at 1 seed per a small hole on a styrofoam plate with

a nylon net supported at the bottom. The plates

were floated on a nutrient solution recommended

by Yoshida et al.(1976). After 14 days of growth,

the seedlings were subjected to salinization (EC 6

dS /m) by adding NaCl to the nutrient solution.



The nutrient solution was renewed once a week,

and its pH was maintained daily at 5.5 (adjusted by

adding either 1N NaOH or HCl). The seedlings

were grown in screenhouse at Ubon Ratchathani

Rice Research Center, Ubon Ratchathani, Thailand.

The experimental design was a 16 ¥ 2

factorial in RCB design with 3 replications. The

treatments consisted of 16 rice cultivars and 2

salinity levels of 0 and 6 dS /m. Each experimental

unit consisted of 18 plants. The salt tolerant scoring

(Table 1) was recorded 16 days after salinization

(Gregorio et al.,1997). At the same time, leaf and

shoot samples were taken for relative water content

and Na+ and K+ content determination,

respectively. For relative water content, the

youngest fully expanded leaf was used. One

centimeter long leaf sample cut at 1/3 of the leaf

from the leaf tip was taken. Two samples were

weighed to determine fresh weight (FW), then

soaked in distilled water at 25∞C for 4 hrs and

weighed again to record the turgid weight (TW)

and oven-dried at 80∞C for 24 hrs to determine the

dry weight (DW). The Relative Water Content

(RWC) was computed as follows

%RWC = FW DW

TW DW

--

¥100

Shoot samples for Na+ and K+ content

analysis were oven-dried for 3 days at 80∞C.

Dried samples were finely ground, and 0.3 g

powder from each sample was taken for

Na+ and K + analysis using atomic absorption

spectrophotometer.

3. Screening at vegetative stageThe purposes of this study were to confirm

the reliability of the screening at seedling stage

and to determine the differential responses to

salinity of the parents and progenies for

physiological characters including Na+/K+ ratio

in young leaves (1-4 leaves from the top), old

leaves(below the 4th leaves from the top) and

stems, RWC and salt tolerant scoring.

Preparation of pots : Black plastic bags

with 15 cm ¥ 15 cm surface area and 17 cm height

were used as experimental pots. Holes with 3-4

mm in diameter, were made at 2-cm spacing on the

side wall of these bags to allow movement of the

salinized water into the soil. A cotton bag was

placed inside each plastic bag and filled with

fertilized soil (45 mg N, 27 mg P2O5 and depending

on K treatment 18, 22, or 36 mg K2O /kg of soil

used)up to 2 cm below the rim of the bag. The

cotton bags were used once only. Then the bags

were placed in the plastic container filled up with

tap water to the same level as soil in plastic bags.

Six pregerminated seeds of each entry were

placed on soil surface of each bag. Two weeks

after seeding, seedlings were thinned to three per

bag. Water level was then raised to 1 cm above the

soil surface and maintained daily. Pesticides were

Table 1 Modified standard evaluation score (SES) of visual salt injury (Gregorio et al., 1997).

Score Observation Tolerance level

1 Normal growth Highly tolerant

3 Nearly normal growth;

leaf tips or few leaves whitish and rolled Tolerant

5 Growth severely retarded;

most leaves rolled; only a few are elongating Moderately tolerant

7 Complete cessation of growth ;

most leaves dry; some plants dying Susceptible

9 Almost all plant dead or dying Highly susceptible


applied to the plants as necessary.

Salinization : when the seedlings were 21

days old, water was siphoned from the contaniers;

3-hours elapsed before all water was drained out of

the bags. Salinized water solutions with

concentration of 4, 8 and 12 dS /m were prepared

by dissolving table salt (NaCl) in water while

stirring. The plastic container was filled up with

salinized water solution. The solution level was

maintained at 1 cm above soil surface by adding

tap water and salinity levels were monitored for

each treatment. Plants were exposed to different

salinity levels for 12 weeks when the experiment

was completed. The experiment was laid out in a

split plot design: Main plots were arranged as

4 ¥ 3 factorials ;

Factor 1: salinity levels = 4 levels ; 0, 4, 8,

and 12 dS /m

Factor 2: Potassium(K) application rates =

3 rates ; 31.25, 43.75, and 68.75 kg K2O /ha which

represented low, reccommended and high rate,

respectively. The purpose of different levels of K

application was to determine whether K level

affected Na absorption and hence salt tolerant

scoring of rice plants.

Sub plots were 8 lines/cultivars of rice

selected for difference in salinity tolerance from

screening at seedling stage. They were Pokkali,

IR29, KMK, KDML105, FL478, FL496, DDG

and RD.

The experiment was conducted in a

screenhouse at Ubon Ratchathani Rice Research

Center, Ubon Ratchathani, Thailand, during

June – September 2001. The average surrounding

temperature of the screenhouse was (day/night

temperature) 37/29∞C.

RESULTS

1. Seedling stage screeningThere was no significant difference in salt

tolerant scoring, RWC and Na+/K+ ratio when

seedlings were grown at the normal nutrient

solution(data not shown). Physiological traits

contributing to salt tolerance of rice grown in 6 dS/

m solution are shown in Table 2.

Visual symptom observation illustrated that

the lines which had low scores (salt tolerant)

compared to Pokkali(tolerant check) were FL416,

FL478, FL496, and FL530(Table 2). The lines

identified as moderately susceptible to salinity

(with intermediate scores) were FL358, FL367,

FL411, FL434, FL443, FL523, and FL563, while

the susceptible cultivars were RD6 and KDML105.

Genotypic ranking for RWC determined at

midday was similar to that determined at predawn.

The liness FL478, FL496 and FL530 had high

RWC compared to Pokkali (Table 2). The two

landrace cultivars, KMK and DDG, and FL416,

FL434, FL443, FL523, FL358, FL367, FL411,

FL563, and Pokkali were rated as having

moderately high RWC. Cultivars with low RWC

were RD6 and KDML105.

In this experiment, the lines FL496 and

FL530 had low Na+/K+ indicating that these lines

were tolerant to salinity compared to Pokkali with

respect to Na+/K+ ratio. The moderately tolerant

lines and cultivars were FL358, FL367, FL411,

FL416, FL434, FL443, FL478, FL523, FL563,

KMK and DDG, while the cultivars RD6 and

KDML105 were susceptible to salinity (with high

Na+/K+ ratio, Table 2).

2. Vegetative stage screening1 week after salinization (WAS), most rice

cultivars were scored 1 at salinity level of 4 dS/m

(Table 3). This indicated that at this salinity level

the salt content has mild effects on rice growth

except for IR29 (score 2). At 8 dS/m, IR29 was the

most susceptible (score 4) while KDML105 was

the second most susceptible (score 3). The tolerant

lines/cultivars identified were Pokkali, FL496 and

DDG which scored 1. At the salinity level of 12

dS/m, most lines scored 3 except for IR29 and

KDML105 which scored 5 and 4, respectively.

Similar trends were obtained 2 WAS at 4 dS/m. At


8 and 12 dS/m, salt affected these lines and cultivars

more severely than at 4 ds/m. After 12 weeks of

salinization, the rice plants were severely damaged

by the salt at these levels. Most susceptible

cultivatrs, such as IR29, RD6 and KDML105 were

dead or nearly dead, at 8 dS/m. While young leaves

of Pokkali, KMK and DDG were still green, old

leaves were dead. All plants were dead after

Table 3 Salinity tolerant scoring at 4 salinty levels recorded 1,2 and 12 weeks after salinization.

Variety 1 WAS 2 WAS 12 WAS0 dS/m 4 dS/m 8 dS/m 12 dS/m 0 dS/m 4 dS/m 8 dS/m 12 dS/m 0 dS/m 4 dS/m 8 dS/m 12 dS/m

Pokkali 1 a 1 b 1 e 3 cd 1 a 1 cd 2 d 3 d 1 a 1 bc 3 e 6 b

FL 496 1 a 1 b 1 e 3 cd 1 a 1 cd 3 c 4 c 1 a 2 b 7 c 8 a

DDG 1 a 1 b 1 e 3 cd 1 a 2 bc 3 c 4 c 1 a 1 bc 6 d 9 a

KMK 1 a 1 b 2 cd 3 cd 1 a 1 cd 3 c 4 c 1 a 1 bc 5 d 9 a

FL 478 1 a 1 b 2 cd 3 cd 1 a 2 bc 4 b 5 b 1 a 1 bc 7 c 9 a

KDML105 1 a 1 b 3 b 4 b 1 a 2 bc 4 b 5 b 1 a 1 bc 8 b 9 a

RD6 1 a 1 b 2 cd 3 cd 1 a 2 bc 4 b 5 b 1 a 1 bc 9 a 9 a

IR 29 1 a 2 a 4 a 5 a 1 a 3 a 5 a 7 a 1 a 7 a 9 a 9 a

Mean 1 1 2 3 1 2 4 5 1 2 7 9

CV (%) 37.1 24.1 23.0

Table 2 Physiological traits contributing to salinity tolerance in rice grown at salinity level of 6 dS/m.

Salt Plant Plant Relative water content (%)1/

tolerance weight (g) height (cm)

Lines/cultivars scoring* At 6 dS/m At 6 dS/m Midday Predawn Na+/K+ ratio*

FL358 5.7 b 0.638 bcd 71.85 ab 85.80 bc 84.28 de 0.246 bc

FL367 5.0 bcd 0.748 a-d 66.76 b-e 86.98 abc 87.35 b-e 0.266 bc

FL411 5.7 b 0.607 bcd 70.06 a-d 87.94 abc 91.61 a-d 0.241 bc

FL416 3.3 de 0.462 cd 51.96 fg 91.96 ab 94.11 ab 0.169 bc

FL434 5.3 bc 0.565 bcd 78.07 a 88.50 ab 86.73 b-e 0.182 bc

FL443 5.3 bc 0.578 bcd 64.41 cde 88.90 ab 85.04 cde 0.232 bc

FL478 3.0 e 0.665 a-d 55.72 efg 94.16 a 97.93 a 0.183 bc

FL496 3.0 e 0.523 bcd 60.68 c-f 94.28 a 96.68 a 0.158 c

FL523 4.7 b-e 0.633 bcd 78.73 a 91.06 ab 92.51 abc 0.200 bc

FL530 3.7 cde 0.506 cd 48.53 g 93.91 a 96.39 a 0.126 c

FL563 5.0 bcd 0.593 bcd 77.39 ab 86.04 bc 90.75 a-d 0.318 b

KMK 4.3 b-e 0.855 abc 63.42 cde 91.08 ab 90.34 a-d 0.209 bc

DDG 4.3 b-e 1.049 a 65.58 cde 92.00 ab 90.18 a-d 0.233 bc

RD6 7.7 a 0.338 d 59.83 def 81.35 c 82.06 e 0.729 a

KDML105 7.7 a 0.557 bcd 61.53 c-f 75.00 d 81.80 e 0.668 a

Pokkali 3.0 e 0.940 ab 80.61 a 93.27 a 94.12 ab 0.236 bc

CV(%) 32.3 29.0 8.5 4.2 4.5 71.8

1/ The data were collected 16 days after salinization


salinization at 12 dS/m for 12 weeks.

The results from this experiment indicated

that at salinity level of 4 dS/m, there was slight or

no effect on rice growth compared to that grown in

normal condition. Screening salt tolerant rice at

this condition(day/night temperature = 32-37/23-

29∞C) should not be at this level. Therefore the

screening was based on the data obtained from

salinity levels of 8 dS/m and 12 dS/m 2 weeks after

salinization. The line and cultivars which were

tolerant to these conditions were Pokkali, KMK,

FL496 and DDG. Moderately tolerant line and

cultivars were KDML105, FL478, and RD6. The

susceptible cultivar was IR29.

Total dry weight and total Na+ and K+ uptakePokkali had the highest total dry weight

(TDW) and high total Na+ and K+ uptake (Table 4)

rendering this cultivar to have balance Na+ and K+

in its cell. This implied that Pokkali had high salt

tolerant ability because of dilution effect. On the

other hand, some lines/cultivars had high Na+

uptake but low K+ uptake resulting in imbalance

Na+ and K+ in their cells. These lines/cultivars

included KDML105, RD6 and IR29 which were

rated as salt susceptible.

Effect of Potassium(K) application rateThere was no significant difference in salt

tolerant scoring among potassium application rates

of 31.25, 43.75, and 68.75 kg K2O ha-1(K1, K2,

and K3). This result indicated that the 3 potassium

application rates, which represented low,

recommended and high rate, had no significant

effect on salinity tolerance (Munns et al., 2002).

This suggested that some rice genotypes might

have some mechanisms, such as Na+ exclusion,

which allowed plants to handle high levels of salt

accumulation in their leaves to maintain low Na+/

K+ ratio(Gregorio and Senadhira,1993).

Tab

le 4

Tot

al d

ry w

eigh

t and

tota

l Na+

and

K+ u

ptak

e of

8 r

ice

lines

/cul

tivar

s at

4 s

alin

ity le

vels

.

Lin

es/c

ultiu

ars

Tot

al d

ry w

eigh

t (g/

plan

t)T

otal

Na+

upt

ake

(g/p

lant

)T

otal

K+ u

ptak

e (g

/pla

nt)

0 dS

/m4

dS /m

8 dS

/m12

dS

/m0

dS /m

4 dS

/m8

dS /m

12 d

S /m

0 dS

/m4

dS /m

8 dS

/m12

dS

/m

Pokk

ali

2.44

2 a

2.62

2 a

2.51

9 a

2.32

7 a

0.02

24 a

0.02

91 a

b0.

0393

a0.

0399

ab

0.03

24 a

0.03

26 a

0.02

91 a

0.02

39 a

IR 2

91.

375

a1.

108

c1.

696

b1.

713

ab0.

0111

a0.

0112

c0.

0301

ab

0.04

05 a

b0.

0148

c0.

0109

d0.

0156

b0.

0154

b

KM

K1.

661

a1.

584

bc2.

028

ab1.

588

ab0.

0139

a0.

0201

abc

0.03

34 a

b0.

0296

abc

0.02

01 b

c0.

0176

bcd

0.01

97 b

0.01

29 b

KD

ML

105

1.61

9 a

2.41

1 a

1.77

1 ab

1.53

0 b

0.01

35 a

0.03

35 a

0.03

39 a

b0.

0389

ab

0.01

99 b

c0.

0245

b0.

0152

b0.

0116

b

FL 4

781.

631

a1.

625

bc1.

768

ab1.

601

ab0.

0138

a0.

0185

bc

0.02

72 a

b0.

0346

abc

0.01

96 b

c0.

0172

bcd

0.01

76 b

0.01

47 b

FL 4

961.

319

a1.

981

ab1.

761

ab1.

467

b0.

0104

a0,

0204

abc

0.02

43 b

0.02

19 c

0.01

63 b

c0.

0199

bc

0.01

83 b

0.01

45 b

DD

G1.

177

a 1

.386

bc

1.65

9 b

1.51

3 b

0.00

82 a

0.01

55 b

c0.

0248

ab

0.02

70 b

c0.

0135

c0.

0152

cd

0.01

52 b

0.01

30 b

RD

61.

855

a2.

007

ab1.

781

ab1.

843

ab0.

0155

a0.

0257

ab

0.03

08 a

b0.

0439

a0.

0236

b0.

0216

bc

0.01

49 b

0.01

53 b

cv (

%)

23.8

30.1

25.7


Sodium concentrations in different parts ofrice plant

In young leaves, at 8 dS/ m and 12 dS/ m

levels of salinity, high Na+ concentration was

found in cultivars IR29, KMK, KDML105 and

RD6. Moderately high Na+ concentration was

found in FL478 and DDG, and low Na+

concentration were found in Pokkali and FL496.

In old leaves, similar trends were obtained; high

Na+ concentration was found in the cultivars IR29,

KDML105 and KMK. FL478 and DDG had

accumulated medium of Na+ concentration, while

Pokkali and FL496 had low Na+ concentration in

old leaves. In stems, a high concentration of Na+

was found at high salinity level (12 dS/ m) in IR29,

KMK, KDML105 FL478, and RD6. On the other

hand, DDG had moderately low Na+ concentration

and Pokkali and FL496 had low Na+ concentration

in stems (data not shown).

Potassium concentrations in different parts ofrice plant

There were high K+ concentrations in

young leaves of KMK, FL496 and RD6 at 0 dS /m

of salinity (Table 5). At 4 dS /m salinity level,

KMK had the highest K+ concentration, however

it was not significantly different from Pokkali,

FL496, DDG and RD6. At 8 and 12 dS /m, IR29

had the highest K+ concentration. In most salinity

levels, KDML105 had the lowest K+ concentration.

In the old leaves, for all salinity levels, Pokkali had

the highest K+ concentration except at 12 dS /m

where FL496 had the highest K+ concentration.

IR29 had the lowest K+ concentration in all cases

studied. In stems, the lines and cultivars which had

high K+ concentration at 0 dS /m were FL478,

Pokkali, KMK, FL496 and DDG, while RD6 had

moderately high K+ concentration IR29 and

KDML105 had low K+ concentration. Similar

trend was observed in stems of rice plant grown at

4, 8 and 12 dS /m where Pokkali had the highest K+

concentration, and KMK, FL478, FL496 and DDG

had moderately high K+ concentration. In all Tab

le 5

K+ c

once

ntra

tion

(%)

in 3

par

ts o

f r

ice

pla

nt a

t 1

wee

k a

fter

sal

iniz

atio

n.

Lin

es/c

ultiu

ars

You

ng l

eave

s (1

-4 le

aves

fro

m t

he t

op)

Old

lea

ves

(af

ter

the

4 th

lea

f f

rom

the

top

)St

em

0 dS

/m4

dS /m

8 dS

/m12

dS

/m0

dS /m

4 dS

/m8

dS /m

12 d

S /m

0 dS

/ m4

dS /m

8 dS

/m12

dS

/m

Pokk

ali

1.07

8 b

1.12

2 ab

c1.

062

ab1.

168

b1.

458

a1.

443

a1.

357

a1.

308

a1.

206

ab1.

056

a0.

926

a0.

888

a

IR 2

91.

125

b1.

038

bcd

1.12

9 a

1.34

4 a

1.21

5 b

0.94

5 c

0.93

5 d

1.00

3 b

0.97

0 d

0.69

5 e

0.55

7 d

0.48

2 e

KM

K1.

295

a1.

156

a1.

031

abc

0.98

4 de

1.30

4 ab

1.18

6 b

1.21

5 ab

c1.

024

b1.

182

ab0.

888

b0.

657

c0.

610

cd

KD

ML

105

1.13

3 b

0.95

1 d

0.85

8 d

0.90

4 e

1.33

0 ab

1.07

0 bc

0.97

3 d

1.09

3 b

1.08

5 c

0.75

6 de

0.62

2 cd

0.53

0 de

FL 4

781.

153

b1.

029

cd0.

988

bc1.

098

bc1.

331

ab1.

079

bc1.

173

bc1.

068

b1.

221

a0.

859

bc0.

751

b0.

640

c

FL 4

961.

265

a1.

070

abc

1.02

8 ab

c1.

129

bc1.

349

ab1.

218

b1.

327

ab1.

348

a1.

168

abc

0.85

1 bc

0.74

2 b

0.72

5 b

DD

G1.

142

b1.

061

abc

0.96

0 bc

1.07

2 bc

d1.

346

ab1.

205

b1.

079

cd0.

993

b1.

134

abc

0.83

4 bc

d0.

627

cd0.

630

c

RD

61.

263

a1.

143

ab0.

950

cd1.

043

cd1.

356

ab1.

210

b1.

087

cd1.

072

b1.

123

abc

0.78

0 cd

e0.

579

cd0.

591

cd

cv (

%)

9.5

1510

.8


salinity levels tested, IR29, KDML105 and RD6

had low K+ concentration (Table 5).

Sodium/Potassium ratios (Na+/K+ ratio) indifferent parts of rice plant

As for Na+/K+ ratios in young leaves, there

were no significant genotypic differences in Na+/

K+ ratio at 0 dS /m and 4 dS /m. The different

response appeared at salinity levels of 8 and 12 dS

/m. Low Na+/K+ ratios were found in the leaves of

Pokkali, FL478, FL496 and DDG, while high

Na+/K+ ratio was found in KDML105, RD6, KMK

and IR29 (Table 6). In the old leaves, at 0 dS /m

and 4 dS /m, Na+/K+ ratios of all lines and cultivars

did not show significant difference. At 8 and 12 dS

/m, there was tendency of low Na+/K+ in Pokkali,

FL478, FL496, KMK and DDG, which were

tolerant to salinity (low Na+/K+ ratio Table 6),

while high ratios were found in KDML105, RD6

and IR29. In stems, the Na+/K+ ratio appeared to

be higher than those in young and old leaves.

Pokkali expressed the lowest Na+/K+ ratio at 4, 8

and 12 dS /m while KDML105 had the highest

Na+/K+ ratio at 4 dS /m. At 8 and 12 dS/m, only

Pokkali and FL496 had low Na+/K+ ratio, while

FL478, DDG and KMK had intermediate Na+/K+

ratios, while IR29, KDML105 and RD6 had high

Na+/K+ ratios compared to Pokkali (Table 6).

DISCUSSION

The results of the study indicated that there

was slight or no effect of salinity on rice growth at

salinity level of 4 dS/m . This level might not be

suitable for screening salt tolerance in rice.

However, at 8 dS /m (day/night temperature = 37/

29∞C) most rice plants were dead 12 WAS.

Therefore, the screening should be conducted at

salinity level of 8 dS /m and 12 dS /m and data

should be collected 2 WAS. The shoot Na+/K+

ratio is considered to be a reliable parameter used

to evaluate salt tolerance ability of rice cultivars

(Gregorio et al., 1997; Chotechuen, 2001; Mishra Tab

le 6

Na+

/K+ r

atio

in

3 p

arts

of

ric

e p

lant

at

1 w

eek

aft

er s

alin

izat

ion.

Lin

es/c

ultiu

ars

You

ng l

eave

s (1

-4 le

aves

fro

m t

he t

op)

Old

lea

ves

(af

ter

the

4 th

lea

f f

rom

the

top

)St

em

0 dS

/m4

dS/ m

8 dS

/m12

dS

/m0

dS /m

4 dS

/m8

dS /m

12 d

S /m

0 dS

/m4

dS /m

8 dS

/m12

dS

/m

Pokk

ali

0.24

5 a

0.26

2 b

0.39

8 de

0.38

5 f

0.82

2 a

1.09

3 b

1.55

0 d

1.70

4 c

1.05

6 a

1.51

6 c

2.28

4 c

2.52

0 d

IR 2

90.

226

a0.

297

ab0.

571

bcd

0.80

8 cd

0.91

2 a

1.70

2 ab

3.06

9 a

3.05

8 ab

1.04

2 a

2.21

3 ab

c4.

077

a5.

671

a

KM

K0.

225

a0.

353

ab0.

619

bc1.

028

b0.

870

a1.

634

ab2.

230

bc2.

573

b1.

058

a2.

200

abc

3.50

4 ab

4.64

1 bc

KD

ML

105

0.19

5 a

0.47

2 a

0.99

3 a

1.26

9 a

0.86

8 a

1.92

6 a

3.19

6 a

3.58

3 a

1.31

9 a

2.63

6 a

3.98

2 a

5.60

5 a

FL 4

780.

201

a0.

336

ab0.

433

de0.

666

de0.

763

a1.

582

ab2.

085

cd2.

814

b0.

975

a2.

009

abc

2.89

4 bc

4.78

8 b

FL 4

960.

178

a0.

269

b0.

369

e0.

404

f0.

835

a1.

212

b1.

545

d1.

807

c0.

944

a1.

702

bc2.

682

c3.

138

d

DD

G0.

212

a0.

326

ab0.

491

cde

0.56

6 e

0.87

1 a

1.36

9 ab

2.33

5 bc

2.69

7 b

0.96

4 a

1.91

1 ab

c3.

428

ab4.

008

c

RD

60.

193

a0.

320

ab0.

739

b0.

920

bc0.

830

a1.

636

ab2.

742

ab3.

214

ab1.

044

a2.

356

ab4.

086

a4.

512

bc

cv (

%)

37.5

34.9

26.4


et al., 1998). Cultivars with low Na+/K+ ratio

usually have an ability to adjust the Na+ content in

parts of the plant to prevent toxicity of the ions. At

the same time, the rice plant also has K+ absorption

capacity to balance the Na+ in the cell (Gregorio et

al., 1993). Therefore, the cultivars which have the

ability of Na+/K+ balance (low Na+/K+ ratio) is

classified as salt tolerant cultivars. In this work, a

strong positive correlation (r = 0.84**) between

Na+/K+ ratio and salt tolerant scoring (Figure1a)

and negative correlation (r = - 0.71**) between

Na+/K+ ratio and relative water content (Figure1b)

in rice plant were found. Usually, when rice plants

are subjected to stress conditions caused by salinity,

the tolerant plants will markedly accumulate a

number of solute particles, i.e., proline, glycine

betain (Bray et al., 2000) and also K+ which is an

essential element in many enzyme activators or

cofactors and catalysts in plant mechanism(Evans

and Sorgor 1966). In general, many researchers

point out that in plants, high affinity K+ uptake

transporter correlates with low Na+ uptake. In

other words, the lower the Na+ uptake, the higher

the K+ uptake when rice plants are under

stresses(Munns et al., 2002; Amtmann and Sanders,

1999, Blumwald, 2000; Munns et al., 2003).

Therefore, the concentrations of proline and glycine

betain in plant cells are high. In this situation, the

percentages of water in the cells were also high due

to the osmotic adjustment. This was demonstrated

by high relaltive water content in plant under such

conditions. It could be concluded that under stress

conditions, plants with low Na+/K+ ratio or high

K+/Na+ ratio and high relative water content were

salinity tolerant.

At 8, and 12 dS /m, there were high external

Na+ concentrations which affected rice growth as

also reported by Greenway (1972) on its effect to

reduce availability of essential elements such as

K+ by nutrient deficiency. Sodium can partially

substitute K+ in a number of crops, while

substitution is minimally effective in others.

However, the critical level of K+ in plant tissue is

relatively high (about 200 ppm), and nearly all K+

are absorbed during vegetative growth (Gardner et

al., 1985). In this experiment, a higher level of K+

application (68.75 kg /ha as compared to 43.75 kg

/ha) was not able to compete with the high Na+

concentration (at 8 and 12 dS /m) and had no effect

on K+ concentration and salt tolerant ability. Salt

tolerant was also defined as genotypic differences

in biomass production in saline versus non-saline

condition over prolonged period of 3-4 weeks.

Short term experiment (1 week) measuring of

plant size revealed large decrease in growth rate

but little genotypic different (Munns and James,

Na+/K+ ratio

0.0 .1 .2 .3 .4 .5 .6 .7 .8

Salt

tole

ranc

e sc

orin

g

2

3

4

5

6

7

8

9

r = .84**

Na+/K+ ratio

0.0 .1 .2 .3 .4 .5 .6 .7 .8

%R

WC

7880828486889092949698

100

r = - 0.71**

Figure 1a Relationship between Na+/K+ ratio and

salt tolerant scoring of 16 rice cultivars/

lines grown in nutrient solution at 6

dS/m.

Figure 1b Relationship between Na+/K+ ratio and

%RWC of 16 rice cultivars/lines grown

in nutrient solution at 6 dS/m.


2003). Therefore, additional application of K

fertilizer in Northeast Thailand soil which is very

low in nutrient contents may not be enough to cope

with the high Na+ concentration in salinized soil

condition. The relation of salt tolerant lines to

plant size (both plant weight and plant height) in

this experiment indicated that there were 2 types of

salt tolerance ; the first type was associated with

the large plant size such as Pokkali (Table 2) with

plant weight of 0.940 g/plant, plant height of 80.61

cm, moderately low RWC, Na+/K+ ratio and salt

tolerant scoring of 3, 2 WAS. This salt tolerant

capability was a dilution effect of the large volume

of the vegetative shoot (Yeo and Flower, 1984)

and the ability of higher K+ uptake, resulting in

low Na+/K+ ratios in shoot and root (Neue, 1991).

The second type of salt tolerant mechanism was

observed in the lines FL478, FL496, and FL530

which had small plant sizes but high RWC both at

predawn and midday. Low Na+/K+ ratio and low

salt tolerant scoring indicated that these lines had

the mechanism to protect the water loss of stressed

plants. This was the type reported by Bolhar-Nor

Denkampf and Draxler (1993) in which they

clarified that during stomata opening, the starch in

chloroplast of guard cell, was degraded. This caused

K+ to move from subsidiary cells to enhance

osmotic value in vacuoles (osmotic adjustment)

and subsequently increased turgor pressure (leaf

water potential). The reduction of nutrient uptake

and Na+ accumulation in plants grown under high

saline medium were also found in long-term

response (Munns and Termat, 1986). The long-

term effect resulted from the accumulation of salt

within expanded leaves (Yeo et al., 1991). Yeo

and Flower (1986) also reported that the salinity

resistance was not conferred by a single factor, but

was indeed the sum of many contributory

physiological traits, which were not necessary

linked. Pokkali resistant ability, therefore, was

contributed not only by dilution effect but also by

second type of salt tolerant mechanism i.e. Na+

exclusion. Pokkali plant size itself rendered dilution

effect to contribute predominately, while in RILs

derived from Pokkali and IR29, Na+ exclusion

mechanism (Munns et al., 2002) was predominant.

The selection for salt tolerant parents can

be made using nutrient solution for screening at

seedling stage. The relationship between salinity

tolerant scoring and Na+/K+ ratio was also taken

into consideration because Na+/K+ ratio rather

than Na+ alone has been used as an index of

salinity tolerance for cultivars comparison in rice

(Asch et al., 2000; Zhu et al.,2001). The cultivars

selected in this manner were confirmed to be also

adapted well to salinized soil condition during

vegetative stage. The Na+/K+ ratio also indicated

that the tolerant cultivar/lines were Pokkali and

FL496. Moderately tolerant cultivar/lines were

FL478, DDG and KMK and susceptible cultivars

were IR29, KDML105 and RD6.

CONCLUSION

Screening for salt tolerant parent materials

was conducted based on physiological characters,

visual symptom(salt tolerant scoring), relative

water content, Na+ and K+ concentrations and

Na+/K+ ratios in shoot and different parts of rice

plant. The screening at seedling stage using nutrient

solution culture was the most appropriate and

reliable technique for a large number of plant

materials generated in each year. The screening at

vegetative stage confirmed that the lines were

salinity tolerant when grown in pots. In this study,

the lines identified as salt tolerant donors were

FL478, FL496 and FL530. KMK and DDG were

moderately tolerant donors (landrace cultivars)

and KDML105 and RD6 were salt susceptible

cultivars. The lines and cultivars which were salt

tolerant and moderately tolerant will be used as

salt tolerant parent in breeding program.

ACKNOWLEDGEMEMTS

We are grateful to Rockefeller Foundation,


for granting the scholarship and we would like to

acknowledge Dr. Shu Fukai (School of Land and

Food Science, the University of Queensland) for

editing this manuscript.

LITERATURE CITED

Akbar, M. and Yabuno, T. 1974. Breeding for

saline-resistant variety of rice. II. Comparative

performance of some rice varieties to salinity

during early development stage. Japan J.Breed. 24 : 176-181.

Amtmann, A. and Sanders, D. 1999. Mechanisms

of Na+ uptake by plant cells. Adv. Bot. Res.

29 : 76 –112.

Asch, F., Dingkuhn, M., Dorffing, K. and Meizan,

K. 2000. Leaf K/Na ratio predicts salinity

induced yield loss in irrigated rice. Euphytica113: 109-118.

Blumwald, E. 2000. Sodium transport and salt

tolerance in plant cells. Curr. Opin. Cell Biol.12 : 431 – 434.

Bolhar-Nor Denkampf, H. R. and Draxler, G.

1993. Functional leaf anatomy In D.O. Hall,

J.M. O. SCURLOCK, H.R. Bolhar-Nor

Denkampf, R.C. Leegood and S.P. Long (eds.)

Photosynthesis and Production in a Changing

Environment. A Field and LaboratoryManual. Chapman and Hall, London. pp.

99-112.

Bray, A.E., Bailey-Serres, J. and Weretilnyk, E.

2000. Responses to abiotic stresses. In B.

Buchanan, W. Gruissem,R. Jones, (eds.)

Biochemistry & Molecular Biology ofPlants. Rockville, Marryland, pp. 1158 -

1175.

Chotechuen,S. 2001. Responses of rice (Oryza

sativa L.) to salt stress. Paper presented to the

Rice Research Seminars held by Prachinburi

Rice Research Center and Ayutthaya Rice

Experiment Station on 5-6 March 2001 at

Aran Mermaid Hotel, Sra Kaew, Thailand.

pp.34-77.

Evan, H.J. and Sorger, G.J. 1966. Role of mineral

elements with emphasis on the univalent

cations. Annu. Rev. Plant Physiol.17: 47–76.

Gardner, F.P., Pearce, R.B. and Michell, R.L.

1985. Physiology of Crop Plants. The Iowa

State University Press, Ames, Iowa. pp.98-

131.

Greenway, H. 1972. Salinity plant growth and

metabolism. J. Aust. Insti. Agri. Sci. 39 : 24

– 34.

Gregorio, G.B. 1997. Tagging salinity tolerance

genes in rice using amplified fragment length

polymorphism (AFLP). UPLB, Ph.D Thesis,

118 p.

Gregorio, G.B., Dharmawansa, S. and Mendoza,

R.D. 1997. Screening rice for salinity

tolerance. IRRI Discussion Paper Series

NO.22 International Rice Research Institute,

Manila, Philippines. pp.2-23.

Gregorio, G.B. and Dharmawansa, S.1993. Genetic

analysis of salinity tolerance in rice (Oryza

sativa L.) Theor. Appl. Gen. 86 : 333-338.

Kaddah, M.T., Lehman, W.F. and Robinson, F.E.

1973. Tolerance of rice (Oryza sativa L.) to

salt during boot, flowering, and grain filling

stages, Agronomy J. 65: 845 – 847.

Mishra, B., Singh, R.K. and Jetly, V.1998.

Inheritance pattern of salinity tolerance in

rice. J. Genet. Breed. 52: 325-331.

Munns, R. and Termat, A. 1986. Whole plant

responses to salinity. Aust. J. Plant Physiol.13: 143 – 160.

Munns, R.1993. Physiological processes limiting

plant growth in saline soil : some dogmas and

hypothesis. Plant Cell Enviro.16 :15 – 24.

Munns, R., Hussain, S., Rivelli, A.R., James,

R.A., Condon, A.G., Linsay, M.P., Lagudah,

E.S., Schachtman, D.P. and Hare, R.A. 2002.

Avenues for increasing salt tolerance of crops,

and the role of physiological based selection

traits. Plant and Soil 247: 93-105.

Munns, R. and James, R.A. 2003. Screening


methods for salinity tolerance : a case study

with tetraploid wheat. Plant and Soil 253:

201-218.

Neue, H. U. 1991. Adverse soil tolerance of rice :

Mechanisms and screening techniques. pp.

243-250. In P. Deturck and F.N.

Ponnamperuma (eds.) Rice Production on

Acid Soil of the Tropics. Institute of

Fundamental Studies, Kandy, Sri Lanka.

Pearson, G.A. and Ayers, A. D. 1960. Rice as a

crop for salt-affected soil in process of

reclamation. U.S. Agri. Prod. Rep. 43 : 1-13.

Toenniessen, G.A. 1984. Review of the world

food situation and the role of salt-tolerant

plants. pp.399-413. In R.C. Staples, G.A.

Toenniessen (eds.) Salinity Tolerance in Plants

– Strategies for Crop Improvement. Wiley,

New York.

Yeo, A.R. and Flowers, T.J.1984. Mechanisms of

salinity resistance in rice and their role as

physiological criteria in plant breeding.

pp.151-170. In R.C. Staples, and G.A.

Toenniessen (eds.) Salinity Tolerance in

Plants–Strategies for Crop Improvement.

Wiley,New York.

Yeo, A. R. and Flowers, T. J. 1986. Salinity

resistance in rice (Oryza sativa L.) and a

pyramiding approach to breeding varieties

for saline soils. Aust. J. Plant Physiol. 13:

161 –173.

Yeo, A. R., Lee, K. S., Izard, P., Boursier, P. J. and

Flowers, T. J. 1991. Short and long – term

effects of salinity on leaf growth in rice (Oryza

sativa L.). J. Exp. Bot. 42: 881 – 889.

Yoshida, S., Forno, O. A., Cock, J. H. and Gomez,

K. A. 1976. Laboratory Manual for

Physiological Studies of Rice. International

Rice Research Institute. Manila. pp.61 –66.

Zhu, G.Y.,Kinet, J.M. and Lutts, S. 2001.

Characterisation of rice (Oryza sativa L.) F3

populations selected for salt resistance. I

Physiological behaviour during vegetative

growth. Euphytica 121: 25-263.


Weed Control Measures and Moisture Conservation PracticesEffects on Seedbank Composition and Vertical Distribution

in the Soil

Girma Woldetsadik1, Sombat Chinawong2, Rungsit Suwanketnikom3,Sunanta Juntakool3 and Visoot Verasan4

ABSTRACT

Change in the weed seedbank due to crop production practices is an important determination of

subsequent weed problems. Research was conducted to compare weed seed bank composition and

vertical distribution of weed seed in the soil among four weed control measures and four moisture

conservation practices at Dera Sub-Center and Melkassa Agricultural Research Center.

Seed numbers at 45 cm depth were lower in pre-emergence of primagram at the rate of 3 L/ha

treatment (137 seeds/m-2) and pre- and post-emergence of primagram at the rate of 3 L/ha plus 2, 4-D at

the rate of 1 L/ha (105 seeds/m-2) at Dera and Melkassa, respectively. More than 60% of the weed seed

bank was concentrated in the upper 15 cm of soil layer in post-emergence treatment at Dera and pre plus

post-emergence at Melkassa site. The seed bank of the moisture conservation treatments was more

uniformly distributed over depth and greater than the other systems. Chenopodium fasciulosum, Cyprus

rotundus, Eragrostis aspera, and Sorghum arundenanceum were the most commonly found in the seed

bank.

Key words: moisture conservation, flat bed, ridge, vertical distribution, and furrow

INTRODUCTION

The weed seed bank, which comprises of

viable seeds in the soil and on its surface, is the

principal source of annual weeds in the field crops.

Size and composition of the seed bank as well as

above ground weed flora reflect past and present

weed, crop, and soil management (Roberts et al.,

1981). Reducing the size of weed seed bank has

been a long-term goal of weed management

strategies, especially for field cropped continuously

(Schweizer et al., 1984). Additions and losses of

1 Bako Agricultural Research Center, P.O. Box-3, Bako, Ethiopia.2 Faculty of Agriculture at Kamphaeng Saen, Department of Agronomy, Kasetsart University, Nakom Pathom 73140, Thailand.3 Faculty of Agriculture, Department of Agronomy, Kasetsart University, Bangkok 10900, Thailand.


seed from the seed bank are affected by physical,

biological and management factors that interact

over time to result in shifts in weed flora (Cavers

and Benoit,1989).

Many weed infestations in cropping seasons

arise from the weed seed bank, so changes in the

seed bank due to agricultural management practices

ultimately result in changing in observed weed

flora. However, seed bank changes must be of

sufficient magnitude to produce detectable changes

in weed flora since only a small percentage of seed

residing in soil is expressed as flora during any

given growing season (Roberts, 1963; Harper,

1977).

Weed control by herbicides and mulch are

two primary practices that have an impact on weed

seed banks. Recognizing the importance of

herbicide and mulching in altering species

composition in the weed seed bank can lead to

improved strategies for weed management.

Weed seed depth in the soil influences

germination and seedling. Seed at or just below the

soil surface often germinate more than seed buried

deeper in the soil (Chepil, 1946; Herr et al., 1970).

Seed placed deep by plowing may remain dormant

until further tillage places them where germination

may occur. Weed species with long dormancy are

favored by plowing. Seed buried deep in the soil

also takes longer to emerge and develops seedling

characteristics than weed placed shallow (Mester

and Buhler, 1990). Weak seedlings are easier to

kill by chemical or mechanical methods than more

vigorous plants (Herr et al., 1970, Mester and

Buhler, 1990). Depth for optimum germination

and development varies among species. Thus, it is

very important to analyze the seed bank in the soil,

because seed density estimates is useful for

predicting weed infestation and enables a minimum

use of herbicides (Buhler et al., 1992).

Effective weed management is critical for

successful maize production. The use of herbicides

can also influence the species composition of the

seed bank, and may increase or decrease it,

depending on the chemicals used (Ball et al.,

1989), and they can also cause species shifting

(Roberts, 1968). In general, it can be said that

interactions among herbicides, land preparation

and cultural practice have altered the sizes and

natures of seed bank (Roberts, 1981). One key

factor that can influence herbicide effectiveness is

the propertied weed populations in a field. The

numbers of weed seed found in the soil seed bank

determine the propertied weed population in a

field. The seed bank is a continuous fluctuation

due to introductions of new weed seed and losses

of seed from germination or decay (Maxwell and

Ghersa, 1992). The soil seed bank population

declined by more than 90% after five years of

continuous maize with excellent control (Schweizer

et al., 1984). However, the seed bank rapidly

increased when the level of weed management

reduced. The level of input required to obtain

acceptable weed control is related to the size of the

seed bank.

Little work has been done quantifying depth

distribution of weed seed by herbicide and moisture

conservation practices under field conditions. No

previous work in herbicide and mulch effects on

seed germination was found in Ethiopia. This

experiment evaluated effects of herbicide and weed

management on numbers, depth distribution, and

viability of the soil weed seed bank.

The objective of this study was to compare

weed seed bank composition and vertical

distribution of weed seed in the soil among four

weed control measures and four moisture

conservation practices.


Study site and agronomic practicesA field experiment was conducted at Dera

sub-Center and Melkassa Agricultural Research

Center during the rainy season of 2003. The soil

types at Dera and Melkassa are diverse, most of

them are shallow and the organic matter content is

quite low between 0 and 2% in most areas, resulting

in poor water–holding capacity. The soils are

generally brown, grayish brown or light brown.

The textures of the soil are either clay loam, loam

or sandy loam. The experiment was laid out in a

split- split - plot design, comprising 3 levels of soil

depth [0-15 cm, 15-30 cm, and 30- 45 cm] in main

plots, 4 levels of fertilizer [N0, Control, N1, 10, N2,

20, and N3, 30 kg N/ha] in sub-plots and 4 weed

control treatments [W0, weedy check, W1,

primagram Gold 660 SC 3 L/ha (pre-emergence);

W2, primagram Gold 660 SC 3 L/ha and 2, 4-D 1



L/ha (pre and post-emergence); W3, 2, 4-D 1 L/ha

(post-emergence)] in sub- sub – plots and was

replicated thrice. Maize variety Melkassa-1 double

top-cross early duration was sown in rows of 25

cm apart on 8 July 2003 and 29 June 2003 at Dera

and Melkassa, respectively. Full dose of P2O5 and

half dose of N were applied basal and remaining N

was top dressed in 2 equal splits.

The cropping system was a maize -pulse

rotation, except for 2002 to 2003 where maize

followed maize. Agronomic practices and herbicide

treatments were consistent with those

recommended for maize in Melkassa agricultural

research center. Weed control treatments

representing those most commonly used in the

areas included primagram at the rate of 3 L/ha. The

herbicide was applied within 3 days after maize

planting and before weeds had begun to emerge.

Post-emergence of 2, 4-D was used at the rate of 1

L/ha one month after maize emergence.

Seed bank analysisStudies on seed bank in cultivated soil have

led to the development of many techniques for

estimating seed density from soil samples. To

determine the numbers of viable seed in the soil,

soil samples were taken on 15 May 2003, after

seedbed preparation but prior to seeding and

herbicide application. To measure the vertical

distribution of weed seed to a depth of 45 cm in the

soil, the soil was collected at surface of 0-15 cm,

16-30 cm and 31- 45 cm depths and the 3 samples

from each depth were pooled. The soil was air-

dried and sieved through a 2 mm screen to break up

large soil peds. The soil was spreaded in 22 cm

square trays and watered twice daily in a

greenhouse. Soil samples containing seeds were

watered as needed and maintained with water at a

depth of 1 cm. Weed seedlings that emerged were

identified and counted by species and removed

every 7 days from the beginning of weed emergence

until no weed emergence was observed at the seed-

leaf stage or 1-leaf stage to avoid the periodic

interaction of weeds, throughout the growing

period.

AnalysisAnalyses of variance (ANOVAs) were used

to test the effect of weed control and moisture

conservation practices on the seed banks and

vertical distribution of seeds in the soil. Means

were separated by LSD (0.05). Vertical distribution

of seed was expressed as a proportion of the total

seed bank.


Seed bank sizeSizes of the seed bank differed among

weed control measures and moisture conservation

practices (Table 1). The highest total seed number

was found in the Melkassa soil where 366 and 300

seeds/ m-2 were observed in post herbicide

application and ridge and furrow treatments,

respectively. Weedy check gave 222 total weed

seeds/ m-2 at Melkassa after post herbicide 2, 4-D.

Even though the total weed seeds at Dera for weed

control measurers were similar to the highest total

weed seeds/ 187 m-2 was from weedy check.

Primagram at the rate of 3 L/ha caused no difference

in seed numbers across weed management due to

effective weed control and low seed numbers. In

contrast, 2, 4-D at the rate 1 L/ha caused greatest

difference for weed management by depth.

Primagram plus 2, 4 –D had fewer seed in the soil

profile than 2, 4-D. Lower seed numbers in pre

plus post herbicide application than in control

might be due to better weed control in these weed

control treatments and to the stimulatory effect of

weed control in inducing weed seed germination

(Barralis and Chadoeuf, 1980). Seed numbers in

ridge and furrow were higher than those in flat bed;

flat bed plus straw mulching and ridge and furrow

plus straw mulching at the Melkassa site.

Differences among sites could be attributed in part

to differences in previous vegetative cover in


addition to differences in soil type. Pre-emergence

of primagram plus post-emergence of 2, 4-D at

both sites were best to suppress weeds from weed

control measures. The effect of mulch to control

weeds was found effective only at Melkassa.

Vertical distributionThe vertical distributions of seed in the soil

varied among weed control treatments (Table 2).

The top 0-15 cm of soil contained 43.5, 40.2, 40,

and 61 % of the seed bank in weedy check,

primagram, primagram plus 2, 4-D, and 2, 4-D at

Dera, respectively. At Melkasaa, the top 0-15 cm

of soil contained 59.1, 56.8, 62.7, and 58.3 % in

weedy check, primagram, primagram plus 2, 4-D

and 2, 4-D weed control measures, respectively

which were greater as compared to those at Dera

site. The highest concentrations of weed seed were

found in 0-15 cm layer. In 16-30 cm, distribution

of weed seeds in the soil declined and were found

in the ranges from 18.4 in post-emergence of 2, 4-

D to 46.5% in pre-emergence of primagram at

Dera and from 19.7 in post-emergence of 2, 4-D to

33.5% in pre-emergence of primagram at Melkassa,

which were almost similar in weed seeds in the soil

among the four weed management systems. Seeds

in 31-45 cm soil depth declined steadily and were

found in the range from 13.3 in pre-emergence of

primagram to 22.7 % in pre plus post-emergence

of primagram plus 2, 4-D at Dera in the order of

primagram at the rate of 3 L/ha, weedy check, 2, 4-

D at the rate of 1 L/ha and primagram @ 3 L/ha

plus 2, 4-D at the rate of 1 L/ha, respectively.

In moisture conservation practice, at Dera

flat bed plus straw mulching recorded the lowest

percent of weed seed at 0-15 cm soil depth and the

highest at 16-30 cm were recorded. Ridge and

furrow recorded the highest weed seed at 0-15 cm

and the lowest at 31-45 cm, while ridge and furrow

plus straw mulching recorded the lowest percent

of weed seed at 16-30 cm and the highest at 31-45

cm. At Melkassa flat bed plus straw mulching

showed the lowest percent of weed seed at 0-15 cm

and 31- 45 cm and the highest at 16-30 cm where

as flat bed without straw mulching treatment

recorded the highest percent of weed seed at 16-30

cm and the lowest at 16-30 cm. Ridge and furrow

with and without straw mulching expressed

medium percents of weed seed at the three depths

and which were not different from one another. At

both sites, flat bed plus straw mulching recorded

the lowest percentage of weed seed at 0-15 cm

layer.

The highest number of weed seed among

the four weed control measures was obtained from

weedy check followed by primagram @ 3 L/ha

treated plots (Table 3). Ridge and furrow also gave

the highest total weed seed followed by flat bed.

The control measure with mulch gave less number

of weed seed as compared to without mulch. Over

52.7% seeds were concentrated in the top 0-15 cm

layer in 2, 4-D @ 1 L/ha herbicide applied treatment

and flat bed planting method (Table 3). Weed

control measurers ranked 2, 4- D > weedy check >

primagram > primagram + 2, 4-D for total seed in

the top 45 cm soil depth. Depths by weed control

interaction were also significant at both sites.

Species compositionAt four weed control measures, twenty-

eight weed species were identified in the seed bank

or in the field. Almost all of the species observed

only in seed bank samples were relatively common

at the experimental site; hence it was surprising

that these species appeared in the soil samples.

Over 90 % of each seed bank among weed control

treatments and soils were composed of fewer

weed species with 18 species being common to

both sites (Table 4). The weeds in different

combinations were found at each seed bank. The

results showed that the weed seed bank was mainly

composed of Chenopodium fasciulosum,

Eragrostis aspera, Cyperus rotundus, Sorghum

arundenanceum, Flaveria trinerva, Argemone

mexicana, Hetilotropium cineraseioen,

Amaranthus hybridus, Anagalis arvensis, Nicandra


phylasoides, Datura stramonium, Galinsoga

parviflora, Euphorbia heterophylla, Launea

comuta, Portulaca oleracea, Foeum vulgare, etc.

at the two sites. Chenopodium fasciulosum was the

only species present in each weed control treated

plots constituted 21.7 % of the total seed bank

across two locations and consistently made up

greater percentages of the seed bank in pre-, pre -

plus post- and post- emergences than weedy check.

Cyprus rotundus, Eragrostis aspera and Sorghum

arundenanceum were the next most abundant

species at each site. Galinsoga parviflora was

Table 1 Numbers of seed per m2 at a depth of 45 cm in four-weed control measures and averaged

moisture conservation practices.

Total weed seeds Dera Melkassa

Weed control measures

W0 (weedy check) 187 222

W1 (Primagram) 184 144

W2 (Primagram + 2, 4-D) 137 105

W3 (2, 4-D) 135 366

LSD (0.05) 29.92 66.51

Moisture conservation practices

FB (Flat bed) 154 262

R & F (Ridge and furrow) 160 300

FB + SM (Flat bed + straw mulch) 167 168

R & F + SM (Ridge and furrow + straw mulch) 152 203

LSD (0.05) 35.22 37.08

Table 2 Influences of weed control measure and moisture conservation practices on vertical distributions

of total weed seed (%) to a 45 cm soil depth after maize cropping.

Dera Melkassa

Depth (cm)Weed control measures

W0 W1 W2 W3 W0 W1 W2 W3

0-15 43.5 40.2 40 61 59.1 56.8 62.7 58.3

15-30 39.8 46.5 37.3 18.4 28.7 33.5 22.9 19.7

30-45 16.8 13.3 22.7 20.6 12.2 9.7 14.4 22

Moisture conservation practices

Depth (cm) FB R & F FB + SM R & F + SM FB R & F FB + SM R and F + SM

0-15 48.8 51.8 36.5 48.6 61.3 59.2 58.8 55.9

15-30 31 33.8 46.9 30.3 23.2 28.7 30.4 26

30-45 20.2 10.6 10.8 21.1 15.5 12.1 10.8 18.1

W0 = weedy check, W1= primagram @ 3 L/ha, W2 = primagram @ 3 L/ha + 2, 4-D @ 1 L/ha, and W3 = 2, 4-D @ 1 L/ha. FB =

flat bed, R & F = ridge and furrow, FB + SM= flat bed + straw mulch, and R & F + SM = ridge and furrow + straw mulch. L1

= Dera and L2 = Melkassa


found in all weed control measures at both sites but

was not present in post herbicide treatment.

Commelina benghalensis, Agenatum spp,

Convolvulus spp were very small in number and

commonly found in Dera while Euphorbia

heterophylla, Leucas martini, Sidaalba spp, Tagets

minuta, Tribulus spp, and Portulaca oleracea

were found only in Melkassa.

CONCLUSION

In this study, the herbicides and residues

Table 3 Distribution of weed seed by soil depth as affected by weed control and moisture conservation

practices.

Weed seed as affected by depth (cm)

Weed control Moisture Dera Melkassa

conservation 0-15 16-30 31-45 0-15 16-30 31-45

No m-2

Weedy check (control)FB 35 18 3 42 16 14

R & F 24 20 5 31 11 6

FB + SM 28 14 5 21 10 16

R & F +SM 23 7 5 29 14 12

Primagram @ 3 L/ha (pre-emergence)FB 23 15 1 18 12 8

R & F 42 16 12 17 10 5

FB + SM 25 12 4 15 13 4

R & F + SM 12 20 2 27 7 8

Primagram @ 3 L/ha and 2, 4-D @ 1 L/ha (pre and post-emergence)FB 6 87 3 13 7 7

R & F 15 7 2 19 8 5

FB + SM 37 12 6 11 5 4

R & F + SM 16 19 6 15 5 6

2, 4-D @ 1 L/ha (post-emergence)

FB 33 4 5 85 22 19

R & F 13 8 6 82 25 25

FB + SM 14 8 2 13 8 7

R & F + SM 21 15 6 45 20 15

FB = flat bed, R & F = ridge and furrow, FB + SM= flat bed + straw mulch, and R & F + SM = Ridge and furrow + straw mulch.

L1 = Dera and L2 = Melkassa

could influence weed population levels, the rate of

population growth and species composition. This

might indicate that herbicides and residues were

effective enough to prevent excessive weed seed

production. This result was obtained from one-

year data, the long–term effects of herbicide and

residue on weed populations were not studied.

Further studies are needed to examine the

interaction effects of herbicide with residue

application, allelophaty and shading effects and

may result in reduced weed seedling emergence. It

also needs further study on farmer’s field where


Table 4 Seed bank compositions (45 cm deep) in two sites after one year of maize cropping with four

weed control measures.

Seed bank composition as affected by soil and weed control

Species Dera Melkassa

W0 W1 W2 W3 W0 W1 W2 W3

Chenopodium fasciulosum 43 89 56 55 10 30 9 11

Flaveria trinerva 12 3 1 1 24 13 7 12

Cyperus rotundus 26 19 7 2 12 6 1 27

Amaranthus hybridus 2 2 4 1 5 14 11 6

Eragrostis aspera 19 9 36 24 5 3 6 19

Argemone mexicana 15 9 7 3 16 7 5 3

Sorghum arundenanceum 26 13 3 22 9 2 1 19

Nicandra physaloides 1 3 5 3 3 4 12 6

Heliotropism cineraseioen 7 13 8 11 5 8 6 2

Anagalis arvensis 12 3 2 7 9 3 2 3

Datura stramonium 3 7 0 0 3 7 0 0

Euphorbia heterophylla * * * * 4 1 0 2

Galinsoga parviflora 1 3 1 0 1 3 1 0

Launea comuta 2 0 1 2 2 1 1 1

Portulaca oleracea * * * * 1 1 1 0

Foeum vulgare 1 1 1 2 0 0 0 2

Tagetes minuta * * * * 0 2 0 0

Erucastrum arabicum 2 3 3 4 0 1 1 0

Setaria pumila 1 1 1 0 0 1 1 0

Eleusine indica 0 1 0 0 0 1 0 0

Xanthium strumarium 2 0 2 0 1 0 0 0

Commelina benghalensis 1 0 0 0 * * * *

Ageratum conyzoides 1 3 0 1 * * * *

Convolvulus arvensis 3 0 0 1 * * * *

Leucas martini * * * * 0 1 0 1

Sidaalba spinosa * * * * 0 0 0 1

Tribulus terrestris * * * * 1 1 1 1

Values are means of 3 replications at Melkassa and Dera sites.* Species not found at that site;W0 = weedy check (control); W1 =

Primagram 660 SC @ 3 L/ha (pre emergence); W2 = Primagram 660 SC @ 3 L/ha and 2, 4-D @ 1 L/ha (pre and post emergence);

W3 = 2, 4-D @ 1 L/ha (post emergence). L1 = Dera and L2 = Melkassa

herbicides and residues were rarely or not applied.

The results and information obtained would

expedite the development of management strategies

to reduce populations of weed with seed bank

regeneration strategies and to manage the

successional dynamics of weed in weed control

measures.


LITERATURE CITED

Ball, D. A. and S. D. Miller. 1989. A composition

of techniques for estimation of arable soil

seed banks and relationship to weed flora.

Weed Res. 29: 365-373.

Barralis, G. and R. Chadoeuf. 1980. Etude de la

dynanique d’ une communaute adventice. 1.

Evolution de la flora adventice au course du

cycle vegetative d’une culture. Weed Res.20: 231-237.

Buhler, D. D., J. L. Gunsolus and D. F. Ralston.

1992. Integrated weed management

techniques to reduce herbicide inputs in

Soybean. Agron. J. 84: 973-978.

Burnside , O. C., R. S. Moomaw, F. W. Roeth, G.

A. Wicks and R. G. Wilson. 1986. Weed seed

demise in soil in weed free corn (Zea mays L.)

production across Nebraska. Weed Sci. 34:

248-251.

Cavers, P. B. and D. L. Benoit 1989. Seed banks

in arable land. pp 309-328 in M. A. Leck, V.

T. Parker and R. L. Simpson, eds. Ecology of

Soil Seed Banks. Academic Press New YorkChepil, W. S. 1946. Germination of weed seeds:

11. The influence of tillage treatments on

germination. Science. Agric 8: 347-357.

Harper, C. A. 1977.The seed bank, pp 83-110. In:

Population Biology of plants. AcademicPress, London.

Herr, D. E. and E. W. Stroube. 1970. Velvetleaf

control as influenced by herbicide placement

and seed depth. Weed Sci. 18: 459-461.

Maxwell, B.D. and C. Ghersa. 1992. The influence

of weed seed dispersion versus the effect of

competition on crop yield. Weed Techn. 6:

196-204.

Mester, J. C. and D. D.Buhler.1990. Effect of

seedling depth on Velvetleaf seedling

development and response to cynazine. WeedSci. 38: 34-38.

Roberts, H. A. 1963. Studies ion the weeds of

vegetable crops. 111. Effect of different t

primary cultivation on the w weed seeds in the

soil. J. of Ecology 51: 83-95.

Roberts, H. A. 1968. The changing population of

viable weeds seeds in an arable soil. WeedRes. 8: 253-256.

Roberts, H. A. and J. E. Neilson. 1981. Seed banks

in soils. Adv. Appl. Biology 6: 1-55.

Schweitzer, E. E. and R. L. Zimdahl. 1984. Weed

seed decline in irrigated soil after six years of

continuous corn (Zea mays L.) and herbicides.

Weed Sci. 32: 76-3.


Genetic Diversity of Elite and Exotic Oilseed MeadowfoamGermplasm using AFLP Markers

Sureeporn Katengam1 and Steven J. Knapp2

ABSTRACT

Amplified fragment length polymorphism (AFLP) is a PCR-based marker, which is suitable for

DNA fingerprinting. The AFLP fingerprinting has not been described in meadowfoam. This powerful

method was utilized to access genetic diversity of 41 meadowfoam accessions belonging to the genus

Limnanthes. The objectives were to estimate polymorphic information contents (PIC) for AFLP markers

and genetic distance among germplasm, and to assess the pattern of genetic diversity in meadowfoam

germplasm. One hundred and seventy six polymorphic AFLP markers were produced using 6 primer

combinations across 41 accessions. The PIC value ranged from 0.0 to 0.5 and 42 % of germplasm showed

high PIC scores in a range between 0.45 and 0.5. Genetic distance ranged from 0.14 to 0.55 with an

average of 0.44. The UPGMA clustering phenogram based on the distance matrix was consistent with

the known taxonomic classification. The first three principal coordinate analyses accounted for 37 % of

total variation of genetic distance estimated. Cluster analysis and principal component analysis clearly

separated L. floccosa from L. alba. Within L. alba, subspecies alba and versicolor were distinctly

separated into two groups. The results suggested genetic diversity among meadowfoam germplasm was

very high. This information is useful to layout framework for meadowfoam improvement thereby

enhancing productivity and performance of cultivated meadowfoam.

Key words: meadowfoam, Limnanthes sp., genetic diversity, DNA fingerprinting, AFLP

INTRODUCTION

Meadowfoam (Limnanthes sp.) is an annual

oil seed crop, native to Southern Oregon and

California (Mason, 1952). Seed oil of meadowfoam

contains unique unsaturated very long chain fatty

acids (C20 and C22) with outstanding oxidative

stability (Isbell, 1997). Cultivated meadowfoam

which belongs to section Inflexae, family

Limnanthaceae is based on Limnanthes alba. The

section Inflexae comprises of 4 species, namely L.

1 Department of Agronomy, Faculty of Agriculture, Ubon Ratchathani University, Ubon Ratchathani 34190, Thailand.

[email protected] Department of Crop and Soil Science, Oregon State University, Corvallis, OR, 97331 USA.


alba, L. floccosa, L. gracilis and L. montana. The

primary gene pool of L. alba is composed of L.

alba ssp. alba and L. alba ssp. versicolor, whereas

L. floccosa, L. gracilis, and L. montana are

identified as a secondary gene pool of L. alba.

Meadowfoam has been domesticated since

1973 (Jain, 1986). The L. alba was evaluated as

the most promising species in this genus for its

lower moisture requirements, adaptation to wide

ranges of environments, and high seed yield (Gentry

and Miller, 1965). The first non- shattering cultivar,

Foamore, developed for commercial production

was released in 1974 (Calhoun and Crane, 1975).

Breeding and cultivars development is underway

at the Oregon State University with the main goal

of increasing the productivity of meadowfoam by

developing superior cultivars, discovering and

developing novel phenotypes, and advancing our

understanding of the genetics of economically

important traits (Knapp and Crane, 1999).

Knowledge of the genetic diversity and

relationships among germplasm is essential to the

improvement of meadowfoam. Generally, the

genetic diversity of germplasm collections can be

obtained from pedigree records, morphological

traits, isozyme and DNA markers (Smith et al.,

1990; Mumm and Dudley, 1994). However, a

small number of polymorphic markers obtained

from isozyme markers and unfavorable phenotypic

expressions of some morphological traits due to

the environmental effects are known as limitations

of these markers (Smith et al., 1990). The advent

of DNA markers opens the ways to solve this

problem since DNA markers can reveal tremendous

number of genetic loci and they are phenotypic

neutral and not subjected to environmental effects.

A variety of DNA markers have been applied to

cultivar improvement and germplasm

management. Owing to its capacity to reveal a

large number of marker loci in a short period of

time (Vos et al., 1995), AFLP (Amplified fragment

length polymorphism) appears to be the leading

DNA-based marker systems for DNA fingerprints.

AFLP has shown to be a powerful tool for genetic

diversity study in many plant species such as

soybean (Maughan et al., 1996), barley (Ellis et

al., 1997), and rice (Zhu et al., 1998).

The high throughout AFLP markers were

employed to evaluate the genetic diversity among

recent meadowfoam germplasm. The objectives

were (1) to estimate polymorphic information

contents (PIC) for AFLP markers and estimate

genetic distance among inbreds, open-pollinated

cultivars, wild population and all genotypes, (2) to

assess the pattern of genetic diversity and

relationships of meadowfoam germplasm using

UPGMA cluster analysis and principal coordinate

analysis.

MATERIAL AND METHODS

Plant materialsA total of 41 meadowfoam accessions

representing nine inbred lines, eight open-

pollinated cultivars, and 24 wild populations were

included in this diversity study (Table1). The

meadowfoam seeds were germinated and grown

as described by Katengam et al. (2002). Leaves

from 50 to 55 day-old plants were harvested,

immediately frozen, and stored at -80∞C prior to

DNA extraction.

AFLP fingerprintsGenomic DNA was extracted from frozen

tissue using a protocol similar to Lodhi et al.

(1994) with minor modification. AFLP analysis

was carried out essentially as developed by

Keygene (Waeningen, NL) with the minor

modification that the selection of a subset of

fragments on steptavidin beads was omitted (Vos

et al., 1995). AFLP fingerprints were produced

using six MseI-EcoRI primer pairs with three

selective nucleotides (Table 2).

Data analysisGene diversity was used to describe the

relative value of AFLP marker with respect to the

degree of polymorphism exhibited for each

polymorphic locus. Thus,

Gene diversity = piik 2

1=Âwhere pi is the frequency of ith allele and k is the

number of alleles (Ott, 1991). Anderson et. al.

(1993) indicated that gene diversity was essentially

the same as the polymorphic information content

(PIC) as used by Botstein et al. (1980). This

parameter is sometimes called heterozygosity. Due



Table 1 Meadowfoam germplasm (41 accessions) for AFLP fingerprinting.

Accessions Description

1. OMF63 S5 Self-pollinated inbred line (Selected from OMF159)2. OMF64 S5 Self-pollinated inbred line (Selected from OMF160)3. OMF66 S5 Self-pollinated inbred line (Selected from OMF66)4. OMF109-1 Self-pollinated inbred line (Selected from Mermaid x OMF62/ OMF64)5. OMF109-2 Self-pollinated inbred line (Selected from Mermaid x OMF62/ OMF64)6. OMF109-3 Self-pollinated inbred line (Selected from Mermaid x OMF62/ OMF64)7. OMF40-11 (Mermaid S5) Insect-pollinated L. alba ssp. alba inbred line (Selected from PI

283703)8. LAG109 F4 Self-pollinated Mermaid x L. gracilis ssp. parishii inbred line9. LAG111 F4 Self-pollinated Mermaid x L. gracilis ssp. parishii inbred line

10. OMF66 (Redding) Wild L. alba ssp. versicolor population11. OMF 158 Wild L. alba ssp. versicolor population (Recollected PI 283705)12. OMF159 Wild L. alba ssp. versicolor population (Recollected PI 374791)13. OMF160 Wild L. alba ssp. versicolor population (Recollected PI 374801)14. OMF161 Wild L. alba ssp. versicolor population (Recollected PI 374802)15. OMF57 Wild L. alba ssp. versicolor population (UC328 or UC457)16. OMF52 Wild L. alba ssp. alba population (UC- Calaveras)17. OMF53 Wild L. alba ssp. alba population (UC-Sonoma)18. PI 374793 Wild L. alba ssp. alba population (Placer county)19. PI 374794 Wild L. alba ssp. alba population (Placer county)20. PI 374795 Wild L. alba ssp. alba population (Placer county)21. PI 374796 Wild L. alba ssp. alba population (Butte county)22. PI 374797 Wild L. alba ssp. alba population (Butte county)23. PI 374798 Wild L. alba ssp. alba population (Butte county)24. PI 367900 Wild L. alba ssp. alba population (Sacremento county)25. PI 374792 Wild L. alba ssp. alba population (Shasta county)26. Foamore Open-pollinated cultivar (Selected from PI 283704)27. Mermaid Open-pollinated cultivar (Selected from PI 283703)28. Floral Open-pollinated cultivar (Mermaid x L. floccosa ssp. grandiflora)29. Knowles (OMF69) Open-pollinated cultivar (Selected from bulk of L. alba ssp. alba)30. OMF86 Open-pollinated cultivar (Selected from Knowles)31. OMF78 Open-pollinated cultivar (Selected from intermating between L. alba

ssp.alba and ssp. versicolor)32. OMF87 High oil open-pollinated population (Selected from OMF62)33. PI 283724 L. gracilis ssp. parishii (Wild species)34. PI 420137 L. gracilis ssp. gracilis (Wild species)35. PI 283720 L. floccosa ssp. bellingeriana (Wild species)36. PI 420133 L. floccosa ssp. grandiflora (Wild species)37. PI 283719 L. floccosa ssp. floccosa (Wild species)38. OSU-LF-4 L. floccosa ssp. californica (Wild species)39. PI 283721 L. floccosa ssp. pumila (Wild species)40. PI 283725 L. montana (Wild species)41. OMF62-92 High oil open-pollinated population (L. alba ssp. alba)


to a bi-allelic feature, the PIC value for AFLP

markers therefore ranges from 0.0 (monomorphic)

to 0.5 (polymorphic).

Binary data representing the presence (1)

and absence (0) of specific AFLP marker was

generated. Only unambiguous polymorphic bands

were scored and entered into a binary matrix as

input for the genetic distance analysis. The genetic

distance of Roger as modified by Wright (1978)

was estimated among all genotypes using NTSYS-

pc, version 1.8 (Rohlf, 1993). A dendrogram was

subsequently generated by cluster analysis based

on the unweighted pair group method on the basis

of arithmetic averages (UPGMA) using a genetic

distance matrix. Goodness of fit of a cluster

analysis was tested using cophenetic correlation

(r) value from MXCOMP program in NTSYS,

which allowed direct comparison between the

original dissimilarity matrix that was clustered

and the cophenetic value matrix. Principal

coordinate analysis based on genetic distance

matrix was carried out using the PROC

PRINCOMP procedure of SAS (1996) (SAS

Institute, Inc., Cary, NC) to visualize the dispersion

of individuals in relation to the first three principal

axes of variation.

RESULTS

AFLP fingerprintingThe AFLP fingerprinting of 41 accessions

including nine inbred lines, eight open-pollinated

cultivars, and 24 wild populations of four species

(10 taxa) was performed using six MseI-EcoRI

primer combinations (Table 2). These primer

combinations were chosen based on previous

information of polymorphism level from screening

parents for AFLP meadowfoam mapping study

(Katengam et al., 2002). A total of 176 AFLP

markers were revealed from six primer

combinations, which were polymorphic between

two or more accessions across the 41 germplasm

(Table 3). The polymorphic markers from each

primer combination varied from 18 to 40 markers

Table 2 Oligonucleotide adapters and primers used for AFLP fingerprinting.

Adaptors:

EcoRI adaptors* 91M35 5’-CTCGTAGACTGCGTACC-3’

91M36 3’-CTGACGCATGGTTAA-5’

MseI adaptors* 92A18 5’-GACGATGAGTCCTGAG-3’

92A19 3’-TACTCAGGACTCAT-5’

AFLP primers

EcoRI +1** 92R11 5’-AGACTGCGTACCAATTC / A-3’

MseI +1** 92H20 5’-GACGATGAGTCCTGAGTAA / C-3’

EcoRI +3 92SO5 5’-GACTGCGTACCAATTC / ACA-3’

MseI +3 92G23 5’-GATGAGTCCTGAGTAA / CAG-3’

92G24 5’-GATGAGTCCTGAGTAA / CAT-3’

92G29 5’-GATGAGTCCTGAGTAA / CTG-3’

92G30 5’-GATGAGTCCTGAGTAA / CTC-3’

92F10 5’-GATGAGTCCTGAGTAA / CAC-3’

92F41 5’-GATGAGTCCTGAGTAA / CAA-3’

* = EcoRI and MseI adaptors were ligated onto the ends of genomic restriction fragments.

** = EcoRI+1 and MseI+1 primers were used in the preamplification of template DNA. The AFLP markers were generated using

pairs of EcoRI+3 and MseI +3 primers.


with an average of 29 markers per primer pair. The

sizes of these markers ranged from 50 to 250 bp.

Out of 176 AFLP markers, 142 and 138 AFLP

markers showed polymorphism in at least two

inbred lines and two cultivars, respectively,

whereas 175 AFLP markers revealed

polymorphism in at least two wild populations of

meadowfoam.

Gene diversity (Polymorphic informationcontent, PIC)

Estimation of gene diversity in

meadowfoam germplasm was represented by

polymorphic information content (PIC) value

which showed the probability of polymorphism

between two randomly lines. The PIC scores for

176 AFLP markers ranged from 0.0 to 0.5 (Figure1).

Mean PIC scores was 0.31 for inbred lines, 0.30

for open-pollinated cultivars, 0.40 for wild

populations, and 0.39 for all genotypes based on

142, 138, 175, and 176 polymorphic AFLP markers

respectively. The distribution of PIC score was

dramatically increased from 0.0 to 0.5. Nearly

half of markers (42.61%) showed maximum PIC

scores with a range of 0.45 - 0.50, indicating a high

genetic diversity in meadowfoam germplasm.

Distance analysisGenetic distance among 41 accessions

based on 176 AFLP markers was estimated using

Rogers genetic distance as modified by Wright

(1978), ranging from 0.14 to 0.55, with an average

of 0.44 (Table 4). The distances estimated among

nine inbred lines varied from 0.14 (between

OMF109-1 and OMF109-3) to 0.47 (between

LAG109F4 and OMF109-1, OMF109-2, and

OMF109-3) with an average of 0.39. OMF109-1,

OMF109-2 and OMF109-3 were related as they

were developed from the same cross (Mermaid x

OMF62/ OMF64), but were selected for different

fatty acid concentration profiles. OMF109-2 was

Table 3 Total numbers of informative AFLP marker detected with six primer combinations (one

EcoRI+3 primers (ACA) and six MseI+3 primers) used in diversity study.

Primers combinations (EcoRI+3 / MseI+3) Total polymorphic AFLP markers

ACACTC 28

ACACAG 38

ACACTG 40

ACACAC 35

ACACAA 18

ACACAT 20

Total 176

Average 29.33

Polymorphic information content

0.05

0.05-0.1

0.1-0.15

0.15-0.2

0.2-0.25

0.25-0.3

0.3-0.35

0.35-0.4

0.4-0.45

0.45-0.5

No.

of

AFL

P m

arke

rs

0

10

20

30

40

50

60

70

80

Figure 1 Distribution of polymorphic informa-

tion content (PIC ) scored for 176 AFLP

markers among 41 meadowfoam ac-

cessions.

Kasetsart J. (Nat. Sci.) 39 (2) 199T

able

4

Gen

etic

dis

tanc

e m

atri

x es

timat

ed b

y R

oger

-W f

rom

AFL

P fi

nger

prin

ts o

f 41

mea

dow

foam

acc

essi

ons.

12

34

56

78

910

1112

1314

1516

1718

1920

1. O

MF6

3S5

02.

OM

F64S

50.

310

3. O

MF6

6S5

0.33

0.44

04

OM

F109

-10.

370.

390.

460

5 O

MF1

09-2

0.37

0.38

0.46

0.25

06

OM

F109

-30.

360.

390.

450.

140.

270

7 O

MF4

0-11

0.35

0.43

0.42

0.42

0.43

0.43

08

LA

G10

9F4

0.41

0.44

0.45

0.47

0.47

0.47

0.38

09

LA

G11

1F4

0.35

0.42

0.40

0.44

0.44

0.44

0.42

0.30

010

OM

F66

0.31

0.41

0.32

0.42

0.43

0.43

0.41

0.45

0.42

011

OM

F158

0.30

0.38

0.35

0.42

0.42

0.42

0.41

0.43

0.38

0.32

012

OM

F159

0.33

0.41

0.39

0.42

0.42

0.42

0.40

0.47

0.41

0.37

0.35

013

OM

F160

0.32

0.38

0.38

0.41

0.42

0.41

0.41

0.46

0.42

0.35

0.37

0.34

014

OM

F161

0.32

0.38

0.38

0.37

0.37

0.36

0.44

0.47

0.41

0.34

0.34

0.38

0.38

015

OM

F57

0.43

0.49

0.47

0.47

0.47

0.49

0.43

0.44

0.44

0.47

0.45

0.46

0.47

0.46

016

OM

F52

0.43

0.49

0.45

0.49

0.48

0.50

0.42

0.47

0.44

0.47

0.47

0.48

0.48

0.48

0.42

017

OM

F53

0.38

0.44

0.43

0.46

0.46

0.46

0.40

0.47

0.44

0.42

0.43

0.44

0.42

0.43

0.47

0.44

018

PI

3747

930.

380.

460.

390.

450.

460.

450.

330.

410.

380.

410.

410.

430.

380.

410.

440.

440.

410

19 P

I 37

4794

0.39

0.49

0.39

0.45

0.44

0.46

0.42

0.47

0.43

0.41

0.43

0.45

0.43

0.43

0.47

0.44

0.44

0.36

020

PI

3747

950.

400.

470.

450.

500.

460.

490.

410.

450.

450.

440.

470.

460.

430.

440.

490.

470.

460.

420.

400

21 P

I 37

4796

0.38

0.46

0.41

0.47

0.45

0.48

0.41

0.46

0.42

0.40

0.41

0.43

0.41

0.43

0.45

0.48

0.46

0.34

0.38

0.39

22 P

I 37

4797

0.37

0.43

0.40

0.46

0.44

0.47

0.40

0.45

0.41

0.38

0.40

0.42

0.39

0.43

0.45

0.46

0.40

0.38

0.38

0.43

23 P

I 37

4798

0.37

0.40

0.42

0.46

0.45

0.46

0.42

0.47

0.44

0.40

0.40

0.40

0.39

0.43

0.49

0.46

0.43

0.40

0.42

0.42

24 P

I 36

7900

0.38

0.47

0.43

0.46

0.45

0.47

0.43

0.48

0.45

0.44

0.44

0.46

0.44

0.44

0.47

0.43

0.40

0.40

0.41

0.45

25 P

I 37

4792

0.38

0.42

0.40

0.42

0.43

0.42

0.41

0.46

0.42

0.40

0.38

0.36

0.37

0.39

0.46

0.43

0.43

0.41

0.41

0.47

26 F

oam

ore

0.37

0.45

0.40

0.46

0.46

0.45

0.43

0.49

0.43

0.44

0.41

0.41

0.38

0.45

0.49

0.44

0.44

0.38

0.40

0.42

27 M

erm

aid

0.40

0.46

0.43

0.49

0.48

0.49

0.37

0.48

0.42

0.46

0.43

0.44

0.42

0.46

0.46

0.45

0.43

0.40

0.41

0.44

28 F

lora

l0.

380.

430.

440.

450.

430.

460.

370.

410.

380.

420.

420.

400.

390.

450.

460.

420.

420.

400.

450.

4429

Kno

wle

s0.

400.

390.

440.

440.

430.

450.

400.

470.

430.

440.

430.

440.

400.

430.

460.

430.

430.

400.

440.

4230

OM

F86

0.37

0.39

0.42

0.41

0.43

0.42

0.38

0.46

0.45

0.41

0.41

0.42

0.37

0.42

0.47

0.47

0.40

0.40

0.46

0.44

31 O

MF7

80.

350.

400.

430.

390.

410.

400.

400.

490.

450.

410.

400.

400.

410.

400.

460.

460.

420.

390.

430.

4732

OM

F87

0.36

0.41

0.43

0.42

0.42

0.42

0.39

0.47

0.43

0.40

0.41

0.41

0.41

0.43

0.49

0.46

0.43

0.41

0.44

0.44

33 P

I 28

3724

0.47

0.51

0.49

0.49

0.48

0.50

0.47

0.46

0.44

0.52

0.48

0.49

0.50

0.49

0.45

0.48

0.45

0.47

0.47

0.51

34 P

I 42

0137

0.38

0.47

0.44

0.46

0.47

0.46

0.41

0.45

0.41

0.46

0.43

0.42

0.43

0.43

0.49

0.47

0.44

0.38

0.46

0.44

35 P

I 28

3720

0.47

0.52

0.49

0.49

0.50

0.50

0.48

0.48

0.46

0.53

0.51

0.54

0.52

0.52

0.47

0.46

0.47

0.49

0.45

0.49

36 P

I 42

0133

0.46

0.52

0.51

0.50

0.52

0.51

0.47

0.50

0.46

0.51

0.49

0.52

0.50

0.53

0.48

0.48

0.49

0.51

0.47

0.49

37 P

I 28

3719

0.46

0.52

0.51

0.50

0.51

0.49

0.49

0.46

0.45

0.49

0.49

0.55

0.51

0.52

0.48

0.48

0.50

0.49

0.47

0.49

38 O

SU-L

F 40.

470.

520.

500.

480.

490.

480.

480.

510.

480.

520.

520.

520.

510.

520.

480.

490.

510.

490.

460.

5139

PI

2837

210.

450.

490.

460.

460.

460.

460.

460.

490.

450.

470.

460.

490.

470.

470.

470.

450.

510.

480.

450.

4940

PI

2837

250.

380.

410.

430.

460.

450.

450.

430.

450.

400.

460.

390.

380.

440.

400.

460.

470.

410.

420.

460.

4441

OM

F622

90.

390.

420.

440.

430.

420.

440.

390.

450.

420.

460.

470.

450.

450.

460.

470.

440.

440.

450.

450.

47

200 Kasetsart J. (Nat. Sci.) 39 (2)T

able

4C

ontin

ued.

2122

2324

2526

2728

2930

3132

3334

3536

3738

3940

41

21 P

I 37

4796

0

22 P

I 37

4797

0.35

0

23 P

I 37

4798

0.37

0.39

0

24 P

I 36

7900

0.43

0.41

0.39

0

25 P

I 37

4792

0.42

0.41

0.40

0.40

0

26 F

oam

ore

0.40

0.40

0.35

0.40

0.40

0

27 M

erm

aid

0.43

0.42

0.43

0.44

0.42

0.41

0

28 F

lora

l0.

430.

420.

410.

420.

420.

410.

410

29 K

now

les

0.43

0.42

0.39

0.43

0.38

0.41

0.39

0.38

0

30 O

MF8

60.

440.

410.

380.

420.

400.

440.

380.

390.

340

31 O

MF7

80.

430.

430.

390.

430.

410.

400.

410.

360.

400.

340

32 O

MF8

70.

430.

400.

350.

420.

400.

420.

400.

410.

400.

380.

370

33 P

I 28

3724

0.47

0.49

0.48

0.45

0.46

0.46

0.46

0.47

0.48

0.47

0.49

0.49

0

34 P

I 42

0137

0.43

0.43

0.39

0.42

0.42

0.38

0.41

0.42

0.40

0.43

0.41

0.40

0.47

0

35 P

I 28

3720

0.51

0.49

0.49

0.46

0.49

0.52

0.49

0.51

0.50

0.48

0.52

0.50

0.44

0.49

0

36 P

I 42

0133

0.49

0.49

0.49

0.49

0.49

0.49

0.45

0.50

0.51

0.50

0.51

0.48

0.48

0.50

0.40

0

37 P

I 28

3719

0.50

0.48

0.51

0.51

0.50

0.51

0.48

0.52

0.51

0.50

0.51

0.49

0.48

0.51

0.32

0.41

0

38 O

SU-L

F 40.

520.

520.

500.

490.

480.

510.

480.

530.

490.

500.

520.

490.

490.

480.

390.

340.

410

39 P

I 28

3721

0.47

0.46

0.48

0.48

0.44

0.47

0.46

0.48

0.47

0.49

0.49

0.45

0.49

0.49

0.44

0.33

0.43

0.33

0

40 P

I 28

3725

0.44

0.42

0.41

0.44

0.39

0.41

0.40

0.43

0.40

0.40

0.42

0.38

0.46

0.39

0.52

0.51

0.52

0.49

0.48

0

41 O

MF6

229

0.49

0.46

0.45

0.46

0.44

0.46

0.44

0.41

0.42

0.43

0.42

0.40

0.47

0.47

0.49

0.49

0.49

0.47

0.40

0.43

0


selected for L. alba ssp. versicolor fatty acid

profile, which had high dienoic (22:2 D5D13) and

low erucic acid (22:1D13) content while OMF

109-3 was selected based on L. alba ssp. alba fatty

acid profile, which had high erucic acid but low

dienoic acid content. OMF109-1 was selected

based on heterozygote progeny, which had fatty

acid profiles between these two subspecies. In

contrast to OMF109 inbred lines, LAG109F4 was

derived from an interspecific cross between L.

alba (Mermaid) and L. gracilis ssp. parishii,

therefore it showed the greatest distance to those

derived from intersubspecific (L. alba) crosses.

Amongst the eight open-pollinated

cultivars, the distance estimated varied from 0.34

to 0.46 with an average of 0.40. The greatest

distance was found between Foamore and OMF62-

29, whereas the least distance or close relationship

was found between OMF86 and Knowles and

OMF86 and OMF78.

Among 24 wild meadowfoam populations,

the greatest distance (0.55) was found between

OMF159 (L. alba ssp. versicolor) and PI 283719

(L. floccosa ssp. floccosa). The least distance

(0.32) was found between two wild populations of

L. alba ssp. versicolor. The average genetic

distance among wild population was 0.45 indicating

high genetic diversity in these wild populations.

Cluster analysisCluster analysis using UPGMA

(unweighted pair group method based on arithmetic

mean) was performed to examine genetic

relationships among meadowfoam germplasms.

A phenogram was produced from the UPGMA

cluster analysis of genetic distance matrix for 41

accessions based on mean AFLP data from each

accession (Figure 2). There were three major

diverse clusters. The first cluster (I) was comprised

of L. alba ssp. versicolor. Wild populations of L.

alba ssp. versicolor and inbred lines derived from

them tended to group together in this cluster. The

second and largest cluster (II) was primarily

comprised of L. alba ssp. alba with two distinct

subclusters. The first subcluster included wild

populations of L. alba ssp. alba and the other

consisted of all elite germplasm (open-pollinated

cultivars) and wild populations of L. alba ssp. alba

(PI 374798), L. gracilis spp. gracilis (PI 420137),

and L. montana (PI 283725). The third cluster (III)

was composed of five taxa of L. floccosa including

subspecies bellingeriana, floccosa, grandiflora,

california and pumila (PI 283720, PI 283719, PI

420133, OSU-LF4 and PI 283721, respectively).

Three inbred lines (OMF62-29, LAG109-

F4 and LAG111-F4), separately formed a small

cluster far from the others (cluster IV). OMF62-

29 was high oil content enhanced germplasm

derived from L. alba ssp. alba, while LAG109-F4

and LAG111-F4 were inbred lines derived from

interspecific cross between L. alba ssp. alba

(Mermaid) and L. gracilis ssp. parishii. This

phenogram showed that these two inbred lines

were in between Mermaid and L. gracilis ssp.

parishii indicating that they were equally related

to their parents. The remaining two small clusters

(V and VI) consisted of two wild populations of L.

alba ssp. alba, OMF53 and PI 367900 and two

wild populations of L. alba alba (OMF52) and L.

alba versicolor (OMF57), respectively. The latter

was distantly related to their groups (Figure 2).

The goodness of fit of this UPGMA cluster

analysis was performed based on the cophenetic

correlation (r) value between the cophenetic value

matrix and the original distance matrix. The

cophenetic correlation was high (r = 0.85)

indicating a good fit of the UPGMA cluster analysis

performed.

Principal coordinate analysis (PCA)A two-dimensional presentation of genetic

distance produced by principal coordinate analysis

is shown in Figure 3. The first three principal

coordinates accounted for 37 % of the total variation

in AFLP-based genetic distance (the first, second,

and the third eigenvalues were 0.22, 0.09, and


0.5 0.4 0.3 0.2 0.1

OMF63 S5

OMF158OMF66OMF66 S5

OMF161OMF159OMF160PI 374792OMF64 S5OMF109-1OMF109-3OMF109-2OMF40-11PI 374793PI 374794

PI 374797PI 374795PI 374798OMF87FoamorePI 420137PI 283725MermaidFloralOMF78Knowles

PI 367900LAG109-F4LAG111-F4

OMF62-29OMF57OMF52PI 283724PI 283720PI 283719PI 420133OSU-LF4PI 283721

I

II

III

IV

V

VI

PI 374796

OMF86OMF53

Figure 2 A dendrogram produced by UPGMA clustering of Roger-W genetic distance based on AFLP

data among 41 meadowfoam accessions.

0.06, respectively). The first and the second as

well as the first and the third coordinate clearly

separated the wild populations of L. floccosa from

the other populations. Within wild populations of

L. alba, L. alba ssp. versicolor were clustered and

separated from the L. alba ssp. alba.

DISCUSSION

In this study, 41 accessions of meadowfoam

was fingerprinted including nine inbred lines, eight

open-pollinated cultivars and 24 wild populations.

One hundred seventy six polymorphic AFLP

makers were produced from six primer

combinations in 41 accessions whereas 141 and

138 AFLP markers were polymorphic among

inbred lines and open-pollinated cultivars,

respectively). Abundant of AFLP markers

providing an efficient mean to UPGMA cluster

analysis (Rohlf, 1993) with the cophenetic

correlation (r) of 0.84 suggested that a good fit of

cluster analysis was performed. The dendrogram

produced from UPGMA cluster analysis showed

concordance with the taxonomic classification

(Mason, 1952) and previous systematic and

phylogeny studies using morphological traits and

allozyme markers (McNeill and Jain, 1983).

The principal component analysis provided

three-dimensional presentation of estimated

genetic distance and supported the results of the


Limnanthes species contains a wide range

of mating systems from cleistogamy involving

full autogamy in L. floccosa through intermediate

stage in L. gracilis ssp. parishii, L. graclilis ssp.

gracilis, and L. montana, to L. alba with dominantly

protandous, showy, insect-pollinated flower and

with the lowest autofertility in the section Inflexae

(Mason, 1952; Arroyo, 1973). L. alba Benth. has

been domesticated since 1971, and several open-

pollinated cultivars have been developed for

commercial production. These results revealed

two distinct clusters including wild populations of

L. alba ssp. alba and L. alba spp versicolor (Figure

2) which was consistent with taxonomic

classification. Commercial open-pollinated

cultivars formed a subgroup within L.alba ssp.

alba cluster. Foamore was the first meadowfoam

cultivar developed (Calhoun and Crane, 1975),

followed by Mermaid and Floral (Calhoun and

Crane, 1984). All cultivars were developed by

mass selection. Knowles and OMF86 were closely

related since they were derived from OMF58 by

one and two cycles of recurrent half-sib family

selection, respectively. OMF78 was developed by

one cycle of recurrent half-sib family selection in

OMF59. The result from genetic distance showed

that all three cultivars were closely related.

L. alba was addressed as an outcrossing

species and primarily consisted of two subspecies,

alba and versicolor (Arroyo, 1973; Brown et al.,

1979). The mating systems of this species and the

other species in section Inflexae have been widely

investigated (Arroyo, 1975; McNeill and Jain,

1983). Several studies reported the presence of

self-pollinated progeny in wild populations of L.

alba (Arroyo, 1975). Knapp and Crane (1997)

screened 26 accessions of L. alba for self-pollinated

phenotypes and found that six populations of L.

alba ssp. versicolor produced seed in a high

percentage of flowers, which indicated that these

geographically isolated populations seemed to have

allelic diversity for self-pollination. L. alba ssp.

versicolor is distributed from ~37 to 41∞N and ~

Figure 3 Principal coordinate plots of 41

meadowfoam germplasm for the first,

second and third principal coordinates

estimated with 176 AFLP markers, us-

ing Roger-W distance matrix. ∑, L. alba

ssp. alba; , L. alba ssp. versicolor; D,

L. floccosa; —, L. glacillis and L.

montana; ¥¥¥¥¥, Interspecific hybrid.

UPGMA cluster analysis. L. floccosa subspecies

were distinctly separated from the others. Within

L. floccosa, two subgroups were clearly

distinguished in which L. floccosa subspecies

floccosa and bellingeriana were closely related

with genetic distance estimated at 0.32 (Table 4),

while the other members of this species,

grandiflora, californica, and pumila formed a

more distantly related groups. This result is not

only in agreement with previous studies of allozyme

markers and morphological traits (McNeill and

Jain, 1983) but agreed with morphological or

taxonomical classification of this species as

described by Mason (1952) and Arroyo (1973).

The subspecies floccosa and bellingeriana were

classified as fully autogamous, producing

cleistogamous flowers, while the remaining three

subspecies, grandiflora, californica, and pumila,

were assigned as semi-autogamous due to their

relatively more chasmogamous flowers and the

presence of a small degree of protandry.


120 to 123 ∞W in central and northern California

(Brown et al., 1979; McNeill and Jain, 1983).

Self- pollination seems to be concentrated in

populations originating near Redding California

(40.5 ∞N, 122.4 ∞W), and OMF66 (Redding) was

found to be a source of self-pollinated phenotypes

(Knapp and Crane, 1997). Self-pollinated inbred

lines have been developed from OMF66 and two

other wild populations of L. alba ssp. versicolor

(OMF159 and OMF160) (Table 1). The self-

pollinated inbred lines developed from these

species provide useful resources for developing

elite meadowfoam cultivars (Table 1).

CONCLUSION

AFLP fingerprinting was proven to be a

promising approach for evaluating genetic diversity

in addition to constructing a genetic linkage map

in meadowform. AFLPs revealed the great

diversity among meadowfoam germplasm, which

broadened the opportunity for meadowfoam

improvement. Wild species primarily contained

sources of many desirable genes underlying

important agronomic and quality traits.

Interspecific and intersubspecific hybridization

among meadowfoam germplasm provided an

opportunity to broaden the genetic base of

meadowfoam cultivars. Molecular breeding and

genome mapping underlying economically

important traits, particularly fatty acid

concentration (low erucic acid content) and self-

pollination traits, are underway in the laboratory,

and this information will be used in marker-assisted

selection for meadowfoam cultivar improvement.

ACKNOWLEDGEMENTS

This research was funded by the Paul C.

Berger Endowment and USDA (#58-5114-8-1021

and # 58-3620-8-107).

LITERATURE CITED

Anderson J.A., G.A. Churchill, J.E. Autrique,

S.D. Tankslay, and M.E. Sorrells. 1993.

Optimizing parental selection for genetic

linkage maps. Genome 36: 181-186.

Arroyo M.T.K. 1973. A taximmetric study of

intraspecific variation in autogamous

Limnanthes floccosa (Limnanthaceae).

Brittonia 25:177-191.

Arroyo M.T.K. 1975. Electrophoretic studies of

genetic variation in natural populations of

allogamous Limnanthes alba and autogamous

Limnanthes floccosa (Limnanthaceae).

Heredity 35(2):153-164.

Botstein D., R.L.White, M. Skolnick and R. Davis.

1980. Construction of genetic linkage map in

human using restriction fragment length

polymorphism. Am. J. Hum. Genet. 32:314-

331.

Brown C.R., H. Hauptli and S.K. Jain. 1979.

Variation in Limnanthes alba: A biosystematic

survey of germ plasm resources. Eco. Bot.33:267-274.

Calhoun W. and J.M. Crane. 1975. Registrationof ‘Foamore’ meadowfoam. Oregon Agr.

Exp. Sta., Corvallis.

Calhoun W. and J.M. Crane. 1984. Registrationof ‘Mermaid’ Meadowfoam. Oregon Agr.

Exp. Sta., Corvallis.

Ellis R.P., J.W. McNiclo, E. Baird, A. Booth, P.

Lawrence, B. Thomus and W. Powell. 1997.

The use of AFLP to examine genetic

relatedness in barley. Mol. breeding 3:359-

369.

Gentry H.S. and R.W Miller. 1965. The search for

new industrial crops. IV. Perspectives of

Limnanthes. Eco. Bot. 19:25-32.

Isbell T.A. 1997. Development of meadowfoam

as industrial crop through novel fatty acid

derivatives. Lipid Technol. 9:140-144.

Jain S. 1986. Domestication of Limnanthes

(Meadowfoam) as a new oil crop. In FAO/


IAEA (ed.) Plant domestication inducedmutation. Proceedings of an advisory group

meeting on the possible use of mutation

breeding for rapid domestication of new crop

plants. p 121-134, Vienna, Austria.

Katengam S., J.M. Crane and S.J. Knapp. 2002.

The development of a genetic map for

meadowfoam comprised of amplified

fragment length polymorphisms. Theor.Appl. Genet. 104:92-96.

Knapp S.J. and J.M. Crane. 1997. The development

of self-pollinated inbred lines of meadowfoam

by direct selection in open-pollinated. CropSci. 37:1770-1775.

Knapp S.J. and J.M. Crane. 1999. Breeding

advances and germplasm resources in

meadowfoam: A novel very long-chain

oilseed. In Janick J pp. 222-233. Perspectiveson new crops and uses. Ashs Press,

Alexandria, VA.

Lodhi M.A., G.-N. Ye, N.F. Weeden and B.I.

Reisch. 1994. A simple and efficient method

for DNA extraction from grapevine cultivars

and Vitis species. Plant Mol. Biol. Rep.12(1):6-13.

Mason C.T. 1952. A systematic study of the genus

Limnanthes. Bot. Publ. 25:455-512.

Maughan P.J., M.A. Saghai Maroof, G.R. Buss

and G.M. Huestis. 1996. Amplified fragment

length polymorphism (AFLP) in soybean:

species, diversity, inheritance, and near-

isogenic line analysis. Theor. Appl. Genet.93:392-401.

McNeill C.I. and S.K. Jain. 1983 Genetic

differentiation studies and phylogenetic

inference in the plant genus Limnanthes

(section Inflexae). Theor. Appl. Genet.66:257-269.

Mumm R.H. and J.W. Dudley. 1994. A

classification of 148 U.S. maize inbreds: I.

Cluster analysis based on RFLPs. Crop Sci.34:842-851.

Ott J. 1991. Analysis of human genetic linkage.John Hopkins University Press, Baltimore,

Maryland.

Rohlf F.J. 1993. NTSYS-pc numerical taxonomyand multivariate analysis system. Version

1.8. Exster Software, Setauket, New York.

SAS 6.12. 1996. SAS User’s guide. SAS institute

Inc. Cary, North Carolina.

Smith O.S., J.S.C. Smith, S.L. Bowen, R.A.

Tenborg and S.J. Wall, 1990. Similarities

among a group of elite inbreds as measured by

pedigree, F1 grain yield, grain yield, heterosis,

and RFLPs. Theor. Appl. Genet. 80:833-

840.

Vos P., R. Hoger, M. Bleeker, M. Reijans, T. van

de Lee, M. Hornes, A. Frijters, J. Pot, J.

Peleman, M. Kuiper, and M. Zabeau. 1995.

AFLP: a new technique for DNA

fingerprinting. Nucleic Acids. Res. 23(21):

4407-4414.

Wright S. 1978. Evolutions and the genetics ofpopulations. Vol.4 Variability within and

among natural populations. University of

Chicago Press, Chicago.

Zhu J., M.D.Gale, S. Quarrie, M.T. Jackson and

G.J. Bryan, 1998. AFLP markers for the study

of rice biodiversity. Theor. Appl. Genet.96:602-611.


Effects of Gramma Radiation on Azuki Bean Weevil,Callosobruchus chinensis (L.)

Jakarpong Supawan1, Praparat Hormchan1, Manon Sutantawong2

and Arunee Wongpiyasatid3

ABSTRACT

Effects of gamma radiation at various doses on eggs, larvae, pupae and adults of azuki bean weevil,

Callosobruchus chinensis (L.) were studied. It was found that there were no significant differences

between percent egg mortalities at 40 and 80 Gy and among 120, 160 and 180 Gy. As for the larvae,

percent mortalities at all doses were significantly different from that of the control. Significant differences

were also found among tested doses from that of the control in percent pupal mortality. After 4 and 7 days

of irradiation, percent adult mortalities were found to be significantly different from those of the control

at every dose. No significant differences were found among percent mortalities at all tested doses between

4 and 7 days. Percent sterilities at 100 and 120 Gy were not significantly different. When untreated male

mating with treated female, there were significant differences in fecundity from the control and from

among one another. In treated larvae at 100, 300 and 500 Gy, of all doses the degrees of melanization

decreased with the increasing dose. The percent phenoloxidase (PO) activities of the treated larvae

reduced compared with that of the control.

Key words: gamma radiation, azuki bean weevil, Callosobruchus chinensis, melanization, phenoloxidase

activity

INTRODUCTION

Irradiation of insect has received wide

attention in all aspects, from the fundamentals of

genetics, through the different approaches of

exploratory development to technology and

successful application to limited areas or to country-

wide insect-pest eradication programs. Radiation

studies on specific effects have been carried out on

male and female germ cells in Diptera,

Hymenoptera, Coleoptera and Hemiptera in order

to measure variation in response and sensitivity to

1 Department of Entomology, Faculty of Agriculture, Kasetsart University, Bangkok 10900, Thailand.2 Biological Science Division, Office of Atoms for Peace, Bangkok 10900, Thailand.3 Department of Applied Radiation and Isotopes, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand.


radiation at different stages of oogenesis and

spermatogenesis (IAEA, 1963).

Many species of stored-product pests are

cosmopolitan, but other serious pests such as the

khapra beetle (Trogoderma granarium), the larger

grain borer (Prostephanus truncatus), and various

species of legume weevil (Bruchidae) are not.

Irradiation has been proposed as a possible

quarantine treatment for various species of fruit

fly, mango seed weevil, and codling moth but not

for stored-product pests. However, irradiation has

good potential for control of these pests, especially

since the radiosensitivity of many of them is well

documented (Brower and Tilton, 1985).

The genera Bruchus and Callosobruchus

are particularly serious, and moderate to heavy

infestations have been reported in the region,

because the insects pass a major portion of their

life-cycles inside the seeds (Quraishi and Metin,

1963). The most serious of these species in Asia

are azuki bean weevil (Callosobruchus chinensis),

cowpea weevil (Callobruchus maculatus) and

graham bean weevil (Callosobruchus analis). The

three bruchid pests have different distribution

ranges. C. chinensis occurs in Asia, where it is a

pest on azuki bean, chickpea, cowpea, mungbean,

peanut, soybean and other grain legumes. The

aims of the study were to investigate the effects of

gamma radiation on various developmental stages

and reproduction of C. chinensis and to determine

if change in melanization of irradiated C. chinensis

larvae could be used to indicate irradiation exposure

including analysis of phenoloxidase activity.


Laboratory rearing of C. chinensisThe culture was started with eggs of

Callosobruchus chinensis (obtained from Stored

Pest Division, DOA) on Vigna angularis Sayi

seeds kept at room temperature of 27 ± 2∞C, 70 ±

10 %, relative humidity (RH) and 10 : 14 (light :

dark) photoperiod. This included the rearing of

larvae and pupae as well as the adults in azuki bean

seeds. All life stages were used in various

experiments.

Effects of gamma radiation on variousdevelopmental stages of C. chinensis

Eggs: four-day-old eggs from laboratory

culture were irradiated with 0, 40, 60, 120, 160 and

180 Gy in a Cobalt 60 gamma irradiator (Gamma

Cell 220) at the Office of Atoms for Peace. In each

replication, 100 eggs were taken and each dose

treatment was replicated three times. After

irradiation, the eggs were reared in petri dishes (1

cm tall ¥ 9 cm diameter). The viabilities of irradiated

and control egg were recorded at the end of ten

days. Wherever hatching occurred, the larvae were

provided with azuki bean seeds and allowed to

complete development.

Larvae: Seeds of 10-day-old egg laying

with larvae inside were irradiated with doses of 0,

100, 300, 500 and 800 Gy. For each dose, three

replications of 100 larvae each, were used. After

irradiation, the larvae were dissected from seeds,

reared in petri dishes (1 cm tall ¥ 9 cm diameter),

released over azuki bean seeds and kept at 27∞C.

Pupae: Seeds of 22-day-old egg laying

with pupae inside were irradiated with doses of 0,

100, 300, 500 and 800 Gy. Each group of irradiated

pupae consisted of 100 individuals dissected from

seeds. Replications were conducted the same way

as in the larvae.

After hatching, counts of dead and living

insects were recorded every day, together with

development and reproduction changes.

Adults: Approximately two-day-old adults

of C. chinensis were irradiated at 0, 300, 600 and

800 Gy. After irradiation, they were mated into

their groups. Data on insect emergence, insect

deformation and longevity were recorded.

Percent mortality was corrected using

Abbott’s formula. Duncan’s new multiple range

test (DNMRT) was used in statistical analysis.

Effects of gamma radiation on reproduction ofC. chinensis females

Mature females were irradiated with doses

of 0, 40, 60 and 80 Gy while males were irradiated

with 0, 50, 100 and 120 Gy respectively. Treated

and untreated adults were held in the same

conditions previously described. Fecundities and

percent sterilities of female were recorded from

the following crosses for each radiation dose.

Three replications were administered.

1. In each replication, 10 untreated males

were crossed with 10 untreated females (The



control)

UTM X UTF

2. 10 treated males of different doses were

crossed with 10 untreated females

TM X UTF

3. 10 treated females of different doses

were crossed with 10 untreated males

TF X UTM

Where UT stands for untreated, T stands

for treated, M stands for male and F stands for

female.

The fecundity was determined by counting

the eggs laid and the sterility by the percentages of

unhatching egg. The numbers of adult emerged in

each case were recorded as percentages of the

number emerged from the controls.

Effects of gamma radiation on melanization of10 days old C. chinensis

Mungbean seeds of 10 day-old egg laying

with larvae inside were irradiated with doses of 0,

100, 300 and 500 Gy. After irradiation, the larvae

were reared singly in petri dishes. The larvae were

cold-killed by placing in freezer (-20±1∞C) for 24

hours after irradiation of 1, 2, 3 weeks. They were

then removed from the freezer and kept at room

temperature. Melanization was evaluated visually.

The observations on melanization process were

made using stereomicroscope (30X). The color of

melanized body portion (black color) was shown

by photo.

Effects of gamma radiation on phenoloxidase(PO) activity in C. chinensis

Control and irradiated 10-day-old larvae

with 300 and 500 Gy were killed by placing them

in a freezer (-20±1∞C) for a few days. Larvae were

then removed from the freezer until used in the

experiments. PO activity was to be determined by

homogenizing each larval instar individually in

150 ml 0.1 M phosphate buffer solution (pH 6.5) in

glass apparatus. The tissue homogenate was added

with polyvinylpyrrolidone (oxidation protection)

and centrifuged at 12000 x g for 10 min. The

supernatant was used as the enzyme source.

Enzyme preparations were kept on ice until tested

to avoid possible auto oxidation. The substrate

was 3 mg/ml of L-dihydroxyphenylalanine (2-

methyl dopa, Sigma) in 0.1 M phosphate buffer

solution (pH 6.5). One hundred microlitre of the

supernatant was added to 150 ml substrate solution

and 750 ml 0.5 M phosphate buffer, mixed for few

seconds and incubated at 25∞C for 30 min. The

intensity of the red color produced was measured

by light absorption at 490 nm using a 6400

Spectrophotometer (Jenway). One unit of PO

activity was defined as the amount of enzyme, at

pH 6.5, producing a change of one absorbance unit

a 490 nm. Specific activity was measured as units

per milligram of protein (units/mg protein). Data

from spectrophotometric analyses were statistically

analyzed at 0.05 level of significance and means

separated by the Duncan’s new multiple range test

(DNMRT).


Effects of gamma radiation on variousdevelopmental stages of C. chinensis

EggsResults on egg irradiation at different doses

after 10 days are shown in Table 1. It was found

that the mortality of 4 days old egg increased with

the increasing dose. Percentages of mortality at 0,

40, 80, 120, 160 Gy were 0, 40.6, 75.2, 93.7, 98.3

respectively and no egg hatched after treated with

180 Gy. There were no significant differences

between percent mortalities of egg irradiated at 40

and 80 Gy and among those at 120, 160 and 180

Gy. Percent egg mortalities at every dose were

also found to be significantly different from that of

the control.

The experiment showed the eggs of C.

chinensis with 100 percent mortality at 180 Gy

which was in agreement with Sutantawong (1991)

who found that a dose of 180 Gy caused 100%


mortality in C. maculatus eggs.

LarvaeResults on larval irradiation at different

doses are shown in Table 2. The mortalities of 10

days old C. chinensis larvae were also found to

vary directly with the radiation doses. Larval

mortalities of full-grown larvae following

irradiation at 0, 100, 300, 500 and 800 Gy were 0,

27.37, 83.86, 96.14 and 100 % respectively. At

every dose, percent larval mortalities were noticed

to be significantly different from that of the control.

While there was no significant difference between

percent larval mortalities at 500 and 800 Gy, both

were significantly different from those at 100 and

300 Gy.

In the experiment, a dose of 800 Gy caused

100% larval mortality while OAEP (1968) and

Quraishi and Metin (1963) reported 16000 rad

(160 Gy) and 200 Gy respectively to cause 100%

mortality of 16 days old C. chinensis larvae after

hatching from eggs. Yet, the result was similar to

that reported by Sutantawong (1991) which found

500 Gy to cause 100% mortality of 7-10 days old

C. maculatus. It was possible that the different

result from that of Quraishi and Metin (1963) was

in the different ages of larvae used. The 8-day-old

larvae in their experiment, compared to the 10-day

old as in this study, were more sensitive to radiation

as suggested by Molin (2001). Hence, lower dose

(200 Gy) than the dose in the experiment was

required to cause 100% mortality.

PupaeThe results on the pupal irradiation are

shown in Table 3. The mortality of pupal irradiation

also increased with increasing dose. Percentages

of pupal mortality at 0, 100, 300, 500 and 800 Gy

were 0, 44.33, 69.41, 92.09 and 100 % respectively.

Some pupae were found dead inside the seeds,

hence they were not used in the treatments.

Table 1 Effects of gamma irradiation on percent

mortalities of 4-day-old egg of C.

chinensis 10 days after irradiation.

Dose (Gy) % mortality1/

0 0 a

40 40.6 b

80 75.2 b

120 93.7 c

160 98.3 c

180 100 c

1/ Means in column followed by the same letters are not

significantly different at 5% level as determined by

DNMRT.

Table 2 Effects of gamma irradiation on the

mortalities of 10-day-old C. chinensis

larvae 10 days after irradiation.

Dose (Gy) % larval mortality1/

0 0 a

100 27.37 b

300 83.86 c

500 96.14 d

800 100 d



DNMRT.



pupae 10 days after irradiation.

Dose (Gy) % pupal mortality1/

0 0 a

100 44.33 b

300 69.41 b

500 92.09 c

800 100 c



DNMRT.


Significant differences were found among all doses

from that of the control in percent pupal mortality.

It was also found that between 500 and 800 Gy, the

percent pupal mortality did not significantly

different from each other. There was no significant

difference of percent pupal mortality at 100 and

300 Gy as well.

The result was similar to those of

Sutantawong (1991) and Quraishi and Metin (1963)

which reported 100 percent pupal mortality of C.

maculatus and C. chinensis found at 500 and 470

Gy respectively. This study also agreed with Bhuiya

et al. (1985) who found 100 percent pupal mortality

of C. chinensis at 800 Gy.

AdultsTable 4 expresses the adult mortality to

increase when irradiated at higher doses. Four

days after irradiation, the 2-day old adults irradiated

at 0, 300, 600 and 800 Gy were noticed to have

percentages of mortality of 0, 16, 86.5 and 97

respectively, while after 7 days after irradiation

percent mortalities were 0, 81, 100 and 100

respectively. At every dose after 4 and 7 days of

irradiation, percent adult mortalities were found to

be significantly different from those of the control.

While percent mortalities at 600 and 800 Gy were

not significantly different from each other, they

were from that at 300 Gy 4 days after irradiation.

No significant differences among tested doses

after 7 days of irradiation were noticed.

Table 5 shows the percent mortalities of C.

chinensis adult between 4 and 7 days after

irradiation at various doses. It was revealed that

there were significant differences in percent

mortality between 4 and 7 days at 300 and 600 but

not at 800 Gy according to t-test analysis.

According to the results of Table 4 & 5, 800

Gy at 4 and 7 days or 600 Gy at 7 days after

irradiation should be recommended as the killing

doses and times for complete control of this insect.

However, from an economical point of view, 600

Gy at 7 days of irradiation is likely to cause less

expense since higher dose means greater cost

while longer checking days does not increase as

much expense.

Similar results were reported by

Sutantawong (1991) who revealed a dose of 1000

Gy 7 days after irradiated to give C. maculatus

100% mortality. The study also agreed with Kovacs

and Kiss (1985) who reported that with a treatment

of 0.8 kGy all imagoes of T. confusum were dead

7 days after irradiation.

As stated by Molin (2001), the effect of

radiation on insects are many and varied depending

primarily on the species, stage, age and physical

factor and the younger metamorphic stages of

insect are most radiosensitive than the older stages.Table 4 Effects of gamma irradiation on the


adults 4 and 7 days after irradiation.

Dose (Gy) % mortality after irradiation1/

4 days 7 days

0 0 a 0 a

300 16 b 81 b

600 86 c 100 b

800 97 c 100 b

1/ Means in each column followed by the same letters are not


DNMRT.



adult between 4 and 7 days after

irradiation.

Dose (Gy) % mortality after irradiation

4 days 7 days t value

300 16 81 -6.40*

600 86 100 -2.96*

800 97 100 -1.73

* Significant at .05 level.


Thus, eggs, larvae and young pupae are quite

easily killed within a reasonable period of time

after treatment, the old pupae and young adults are

relatively resistant to the lethal effects of irradiation.

Eggs are most sensitive to radiation. This was to

explain egg mortality occurring at lower doses as

in this study compared to the mortalities of other

stages at higher doses. Physical factors also have

an influence on the effects of radiation on insects.

The differences in temperature and humidity,

hence, caused different results to even 2-day old

difference in age as between larval mortality in the

study of Quraishi and Metin (1963) and this

experiment.

Effects of gamma radiation on fecundity andsterility of C. chinensis females

The effects of gamma irradiation with 2-

day-old C. chinensis males (TM) on the fecundities

and sterilities of untreated female (UTF) compared

with the control (UTF X UTM) are shown in Table

6. It was found that the numbers of egg/female at

each dose of untreated females were not

significantly different from the control or from

one another. While the percent sterilities at 100

and 120 Gy were not significantly different they

differed from that at 50 Gy.

When untreated male (UTM) mating with

treated female (TF) (Table 7), it was revealed that

at various doses of radiation the fecundity of

treated female decreased resulting in increased

percent sterilities from 0 to 80 Gy with dose of 80

Gy gave 100% sterility in female. When female

exposed to this dose was paired with a normal

adult of the opposite sex, eggs were laid on the

seeds but the eggs failed to hatch. It was also found

that there were significant differences in fecundity

from the control and among one another at all

doses. Percent sterilities at every tested dose were

significantly different from that of the control (0

Gy) but not among one another.

In sterility of female, the results agreed

with that of Brower and Tilton (1985) who showed

the doses required to sterilize stored-product insects

to vary widely from 70 Gy for the cowpea weevil,

C. chinensis and the female appeared more

susceptible than the male in all cases in the same

insects (OAEP, 1968). The result also agreed with

Sutantawong (1991) who suggested that female of

C. maculatus was susceptible than male at 40, 60

and 80 Gy.

According to the results in Table 6 & 7, the

lower dose in obtaining complete mortality in

females than males was required, hence, irradiated

Table 6 Effects of gamma irradiated 2-day-old

C. chinensis males mated with untreated

females (TM ¥ UTF) on the fecundities

and sterilities of untreated female.

Dose (Gy) Fecundity1/

(Eggs / female) % Sterility2/

0 32.93 a 76.15 a

50 33.23 a 79.67 b

100 28.63 a 90.40 c

120 28.53 a 91.47 c

1/2/ Means in each column followed by the same letters are not


DNMRT.

Table 7 Effects of gamma irradiated 2-day-old

C. chinensis females mated with

untreated males (TF ¥ UTM) on their

fecundities and sterilities.

Dose (Gy) Fecundity1/ % Sterility2/

(Eggs / female)

0 32.93 a 76.15 a

40 24.93 b 98.34 b

60 21.22 c 99.67 b

80 8.66 d 100 b

1/2/ Means in each column followed by the same letters are not


DNMRT.


females should be employed in SIT (Sterile Insect

Technique) program. However, since the females

in the nature are needed to reduce their fecundities,

it is more practical to release the irradiated males

in order to mate with these wild females even

though, in economical point of view, the expense

will be higher.

In most SIT program, however, both sexes

are released and the response of both males and

females to the sterilizing doses has to be assessed.

In some species the males are the more sensitive

sex, e.g., the screw-worm, Cochliomyia

hominivorax, in other species the females are, e.g.,

the medfly, Ceratitis capitata. In case where the

male is the more sensitive sex it would be very

advantage to have a system for the removal of

females so that a lower radiation dose could be

given to the male (Rechcigl and Rechcigl, 1998).

As suggested by Molin (2001), low

radiation doses cause insect sterilization or

genetically deformed gametes, which higher doses

required to induce insect death. In these

experiments, 600-800 Gy caused complete

mortality while only 100-120 Gy induced complete

sterility in C. chinensis adults. Higher radiation

doses cause multiple breaks in the chromosomes,

thus causing death or sterility.

Effects of gamma radiation on melanizationprocess of 10 days old C. chinensis larvae

Figure 1 presents the color of C. chinensis

at different doses. When larvae were treated with

various doses (0-500 Gy), the color variation of

the larvae ranged from black to creamy white. The

degree of melanization in treated larvae decreased

with the increasing doses.

Figure 1 Degrees of darkening in the larvae of azuki bean weevil, C. chinensis 3 days after treatment.

A. Control (30X)

B. Treated with 100 Gy (30X)

C. Treated with 300 Gy (30X)

D. Treated with 500 Gy (30X)


The results similarly agreed with Kongrat-

arpon (2002) who reported melanization occurred

in untreated young larvae of cigarette beetle larvae

after killing by freezing. Some parts of larvae body

became dark brown to black while the rest of the

body was yellow-white or grayed-yellow. In the

treated young larvae, non-melanization to slight

melanization occurred at 100, 300 and 500 Gy.

The degree of melanization in treated young larvae

decreased with the increasing dose. Ignatiwicz

and Banasik-Solgala (1997) reported that after the

irradiation treatment with doses ranging from 0.1

to 0.5 Gy, the melanization process was

significantly inhibited in young larvae cold-killed

after irradiation and old larvae of khapra beetle

melanized slowly after their death, so that some

visible in the body color were noted as late as after

24 hours. With similarity of this study to such

report, the change in melanization of the azuki

bean weevil, C. chinensis larvae could not be used

for indicating previous exposure of these insects to

irradiation, because of the great variability in

response of the melanization process to the

irradiation treatment.

Ignatowicz and Ibrahim (1996) found that

the melanization in irradiated young larvae of the

confused flour beetle, Tribolium confusum

DUAVL., was reduced in the first week after

treatment and completely inhibited in the second

week. Great variation of melanization in the

untreated old larvae partially obscured the effects

of gamma radiation on this process. However, the

melanization was considerably reduced in all

experiments involving old larvae. Banasik-solgala

and Stanislaw (1997) found that the degree of

melanization differed significantly between treated

and untreated the Indian meal moth (Plodia

interpunctella HBN), the Mediterranean flour moth

(Ephestia (Anagasta) kuehniella ZELL) and the

almond moth (Cadra cautella WLK).

Effects of gamma radiation on phenoloxidase(PO) activity in C. chinensis

Results of the experiment on the effects of

radiation on PO in larvae of azuki bean weevil, C.

chinensis, indicated variability of the response of

enzyme activity to gamma radiation. Three sets of

experiment were conducted to determine PO

activity spectrophotometrically. In this assay, the

frozen larvae were used to increase the activity of

the PO system (Mansour and Franz, 1996). The

results of PO activity measurement are shown in

Table 8.

The preliminary analysis of PO in the

irradiated larvae at 0, 100, 300 and 500 Gy showed

low activities of PO which was difficult to detect

because the larvae had small sizes that were not

easy in homogenizing. At 0, 100, 300 and 500 Gy,

PO activities of larva were found to be 2.500,

1.867, 0.851 and 0.409 units/mg protein

respectively.

The values obtained for PO activity showed

that with the increasing dose from 0 to 100, 300

and 500 Gy, PO activity decreased. The PO activity

was found to be significantly different between 0

and 500 Gy, while not differed from those at 100

and 300 Gy. No significant differences were found

among PO activities at 100, 300 and 500 Gy.

Percent reduction of PO activity also

decreased with the increasing dose where the

highest (83.67%) was at 500 Gy and lowest

(25.33%) at 100 Gy.

Irradiation was found to reduce or eliminate

melanization after death and reduced

phenoloxidase activity in the fourth instar of C.

chinensis larvae. The irradiation dose at which

melanization was essentially prevented was from

100 Gy up.

The result agreed with Lupa and Ignatowicz

(1999) who showed the highest doses to inhibit

enzyme activities in larvae of Mediterranean flour

moth and the confused flour beetle after irradiated.

There was a decreasing of PO activity and

significant difference when larvae were treated


with 0.3 (kGy) or higher dose of gamma radiation.

Lupa (2000), also reported that the

phenoloxidase activity in fourth instar of kaphra

beetle (Trogoderma granarium) larvae was

inhibited by 50 and 150 Gy after 1 week and larvae

irradiated with 100 and 300 Gy exhibited higher

enzyme activity compared to the control. Surisan

(2004) reported the changes in the effect of radiation

on PO activity on each instar of cotton bollworm.

At 75 and 150 Gy the irradiated insects were found

to have PO activities significantly less than that of

the control and percent reduction of PO activity

also increased from 4th to 1st instar. According to

Hara et al. (1991) the active units of cuticle-bound

phenoloxidase should be measured during

metamorphosis along with hemolymphatic

phenoloxidase activity in order to better understand

the physiological role of phenoloxidase in cuticle

tanning and sclerotization.

CONCLUSION

The egg, larval, pupal and adult mortalities

increased with the increasing dose with the egg

being the most sensitive. Sterilities of irradiated

male mating with untreated female were

significantly different at all doses while those of

irradiated female mating with untreated male

significantly different between 0 and other doses.

Lower dose was used for female than for male in

obtaining complete mortality. Melanization in

irradiated larvae decreased with increasing dose

and could not be used in detecting radiation

exposure due to great variability in response of the

process to irradiation. Percent PO activity decreased

with the increasing dose.

LITERATURE CITED

Banasik-Solgala, K. and I. Stanislaw. 1997.

Melanization process in irradiated larvae of

stored product moth. Polish J. Entomal. 66:

125-134.

Bhuiya, A.D., M. Ahmed, R. Rezaur, D.R. Seal, G.

Nahar, M.M. Islam and M.S. Islam. 1985.

Insect disinfestation of pulses by irradiation,

pp. 214-221. In J.H. Moy (ed.). RadiationDisinfestation of Food and AgriculturalProduct. University of Hawaii Press,

America.

Dobie, P., C.P. Harines, R.J. Hodges and P.F.

Prevett. 1991. Insects and Arachinids of

Table 8 Phenoloxidase (PO) activity and percent reduction of irradiated 10-day-old larvae of C.

chinensis at different radiation doses.

Dose (Gy) Mean of PO activity (units/mg protein) 1/2/

(% Reduction)

0 (control) 2.500 a

100 1.867 ab

(25.32)

300 0.851 ab

(65.96)

500 0.409 b

(83.67)

1/ Means followed by the same types of letter in the same columns are not significantly different from one another as determined

by DNMRT at 0.05 level.2/ Units of phenoloxidase at 490 nm.


Tropical Stored Products: Their Biologyand Identification (A Training Manual).Storage Department Tropical Development

and Research Institute. UK. 273 p.

Hara, T., T. Tsukamoto, K. Wantanabe, N.

Yamasaki and M. Funatsu. 1991. Properties

of cuticular phenoloxidase from pupae of the

housefly, Musca domestica L. Agric. Biol.Chem. 55(1): 13-17.

IAEA. 1963. Radiation and RadioisotopesApplied to Insects of AgriculturalImportance. International Atomic Energy

Agency, Vienna, Austria. 508 p.

Ignatowicz, S. and H.Z. Ibrahim. 1996. Reduced

melanization after death in larvae of the

confused flour beetle, Trogoderma confusum,

as a result of the irradiation treatment. PolishJ. Entomal. 65: 51-59.

Ignatowicz, S. and K. Banasik-Solgala. 1997.

Reduced melanization after death in larvae of

the khapra beetle, Trogoderma granarium, as

a result of the irradiation treatment.

Radiobiology 42: 801-802.

Kongrat-arpon, T. 2002. Effect of GammaRadiation on the Cigarette Beetle,Lasioderma serricorne (F.). M.S. Thesis,

Kasetsart University, Bangkok.

Kovacs, E. and I. Kiss. 1985. Disinfestation of

wheat germ, wheat, and dried mushrooms by

irradiation. pp. 189-198. In J.H. Moy (ed.).

Radiation Disinfestation of Food andAgricultural Product. University of Hawaii

Press, America.

Lupa, D.A. 2000. Reduced activity of

phenoloxidase after treatment with gamma

radiation in larvae of the kaphra beetle

(Trogoderma granarium Everts), as an easy

method to detect irradiated and nonirradiated

insects. Ann. Warsan Agri. Univ. Hort.

(Landscape Architecture) 21: 9-16.

Lupa, D.A. and S. Ignatowicz. 1999. Change of

phenoloxidase activity in stored product pest

treated with ionization radiation. Progress inPlant Protection 39(2): 458-462.

Mansour, M. and G. Franz. 1996. Effect of gamma

ray radiation on the biology of the Periplaneta

americana. Restaurator 21(1): 41-54.

Molin, R.A. 2001. Food Irradiation: Principlesand Applications. John Wiley and Sons,

Inc., New York. 469 p.

OAEP. 1968. Studies on the Use of GammaRadiation in the Control of Pea Weevil,Callosobruchus chinensis L. Bangkok,

Thailand. 18 p.

Quraishi, M.S and M. Metin. 1963. Radiosensitivity

of various stages of Callosobruchus chinensis

L., pp. 479-484. In International Atomic

Energy Agency. Radiation andRadioisotopes Applied to Insects ofAgricultural Importance. IAEA, Austria.

Rechcigl, J.E. and N.A. Rechcigl. 1998. Biologicaland Biotechnological Control of InsectPests. Lewis Publishers, America. 374 p.

Ress, D.P. 1960. Coleoptera, pp. 1 - 40. In B.

Subramanyam and D.W. Hagstrum (eds.).

Integrated Management of Insect in StoredProducts. Marcel Dekker, Inc., America.

Surisan, S. 2004. Gamma Radiation-InducedChanges in Haemocytes,Melanization and Phenoloxidase Activityof Cotton Bollworm, Heliothis armigeraHübner. M.S. Thesis, Kasetsart University,

Bangkok.

Sutantawong, M. 1991. Disinfestation of theCowpea Weevil, Callosobruchus maculatusF. in Stored Mungbean by GammaIrradiation. Office of Atomic Energy for

Peace, Bangkok 16 p.


Occurrence and Distribution of Major Seedborne Fungi Associatedwith Phaseolus Bean Seeds in Ethiopia

Mohammed Yesuf1 and Somsiri Sangchote2

ABSTRACT

A total of 245 seed samples of Phaseolus bean; 172 common beans, 51 climbing beans and 22

green beans were collected from various bean growing areas during 2003 crop season. The incidence and

severity of seed infection by the major fungal diseases of bean varied between localities, bean types and

cropping practices. Thirteen seed-borne fungal pathogens of different genus were identified from seed

samples collected from the major bean growing regions of Ethiopia. The incidence of different seedborne

fungi ranging between 0.2 to 14.5% was found to vary from location to location and growing conditions.

Among them, Colletotrichum lindemuthianum, Phaeoisariopsis griseola, and Ascochyta phaseolorum

were the most widespread and damaging seedborne fungal pathogens associated with Phaseolus bean

seeds in Ethiopia. From the total seed samples collected, 26.2%, 19.6% and 13.6% of common bean,

climbing bean and green bean respectively were infected by C. lindemuthianum, whereas infection by P.

griseola was 18.6% and 15.7% on common bean and climbing beans seeds respectively. Green bean seeds

were not infected by the latter two fungi. Seeds collected from south, southwest, and western part of

Ethiopia showed heavy seed infection by these major fungal pathogens, whereas seeds produced in dry

areas with minimum rainfall or under irrigation showed very low seed infection. Phytophthora rot of

beans caused by Phytophthora sp. was also detected from green bean pods and immature seeds produced

under irrigation in the central rift valley of the country. The geographic distribution of major seed-borne

fungi of Phaseolus beans was mapped.

Key words: Phaseolus beans, seedborne fungi, distribution, Ethiopia

INTRODUCTION

Different types of Phaseolus beans are

widely cultivated in various agro-ecological regions

of Ethiopia. The most important Phaseolus beans

grown in the country are dry common beans

(Phaseolus vulgaris L), climbing beans (Phaseolus

coccineus) and French/green beans (Phaseolus

vulgaris). The area covered by dry common bean

production alone in Ethiopia is estimated to be

1 Melkassa Agricultural Research Center, P. O. Box 436, Nazareth, Ethiopia.2 Department of Plant Pathology, Faculty of Agriculture, Kasetsart University, Bangkok 10900, Thailand.


more than 200,000 hectare annually (CSA, 2000).

Among the biotic stresses of Phaseolus

bean production in the country, diseases are

considered the major production threats. Under

optimum crop production management practices

of Ethiopia, there is a potential to produce common

beans of 2.5 - 3.0 tons/hectare (Amare, 1987).

However, the national average of bean yield in

Ethiopia is very low (0.6 - 0.7 ton/ha). The major

limiting factor for low yield is thought to be

diseases (Habtu et al., 1996). Seedborne fungi are

among the most important plant pathogens that

cause direct and indirect losses of the bean crop

throughout the world (Schwartz and Galvez, 1980;

Hall, 1994; Abdelmonem, 2000). Losses are

associated with high disease epidemic in the field

consequently causes reduced seed yield and quality

attributing to discoloration and blemish of seeds

and green pods (Hellene, 1988). Moreover,

seedborne pathogens play an important role for the

dispersal and dissemination of several

economically important diseases in the production

field within the country and between countries

(Agarwal and Sinclair, 1997). In a bean production

system like in Ethiopia where there is no seed

health and certification scheme, the risk of seed

infection by seedborne fungi could be high.

Different types of fungal diseases are

reported to cause damage on the bean plant in the

field. Among others, bean anthracnose

(Colletotrichum lindemuthianum), angular leaf spot

(Phaeoisariopsis griseola), aschochyta blight

(Ascochyta phaseolorum), ashy stem blight

(Macrophomina phaseolina) are reported to be the

major fungal pathogens causing foliar diseases of

Phaseolus beans under field condition in Ethiopia

(Habtu and Abiy, 1995, Habtu, et al., 1996).

However, the association of these major fungal

pathogens and other fungi with Phaseolus bean

seeds has not been adequately investigated.

Survey and identification of plant pathogens

is important to understand the association of

pathogens with a specific host plant and to describe

their geographic distribution (Agrios, 1997).

Application of GIS on plant disease distribution

map is helpful to store and refer spatially referenced

point data (Nelson, et al., 1999). Mathur (1995)

emphasised the need of surveys of seedborne

pathogen by having good amounts of sample unit

of different crop species and cultivars so that it

enables researchers to set research priorities. Detail

understanding of the type, occurrence, association

with seed, and geographic distribution of major

pathogens is a prerequisite to formulate rational

integrated and sustainable disease management

practices in different agro-ecologies. Therefore,

the occurrence of major seedborne fungi of

Phaseolus beans and their geographic distribution

are needed to be investigated.

The objective of this study was to investigate

the occurrence, and to map the geographic

distribution of major seedborne fungal pathogens

associated with Phaseolus bean seeds in various

bean growing agro-ecological regions of Ethiopia.


Survey areas and sample collectionDisease survey was carried out in the major

bean growing regions of Ethiopia representing the

lowlands (central rift valley), mid altitudes (north

and southern part) and highlands (north-west and

western part of the country) during the 2003 crop

season. A total of 245 Phaseolus bean seed samples

(172, 51, and 22) of common bean, climbing bean

and green bean respectively were collected from

the major bean-growing areas of the country. The

seed samples were collected from various sources

and cropping practices. These included farmers’

fields, experimental fields, large-scale commercial

farms and local markets. Geographic features like

latitude, longitude and altitude were recorded from

all surveyed areas using handheld Global

Positioning System (GPS), “Garmen” with 12

channels to trace back the specific locations and

types of seedborne fungi. All collected seed samples

were kept at +5 ∞C in a refrigerator. The seed

samples of different Phaseolus bean collected

from various agro-ecologies were tested for seed-

borne fungi by the standard blotter method

(International Seed Testing Association, 1993)

and agar plate method.

Detection of seedborne fungi using standardblotter method

Four hundred untreated and 1% sodium



hypochlorite pre-treated seeds were plated on four

layers of water soaked blotter, 10 seeds per petri

dish. The dishes were then incubated for seven

days in a growth chamber SGC097.C with

programmer, controller and sensor of combined

temperature 24 ± 1∞C and humidity probe under

95% humidity and 12 hours photoperiod of pre-

programmed light conditions (Maden et al., 1975).

The bottom of the compartment was filled with

clean water covered by a glass plate. Slight air

current was forced through a plastic pipe in the

center at the bottom against the underside of the

glass. Such high air humidity enabled the blotters

remoistened during incubation and facilitated

fungal sporulation. After seven days of incubation

samples were examined using stereomicroscope

with 20-30x magnification for the growth of fungi.

The habit characters of various fungi associated

with seeds was recorded. Further identification of

fungi was made using compound microscope.

Detection of seedborne fungi using agar platemethod

Seeds of Phaseolus bean treated with 1%

sodium hypochlorite were plated on potato dextrose

agar (PDA), and then incubated as it was done in

the standard blotter method. Microscopic

examination was made after seven days of

incubation. Comparison on the efficiency of

standard blotter and agar plate methods was also

made by using naturally infected seed lots of

common bean seed by the three major seed borne

fungi (C. lindemuthianum, A. phaseolorum, and P.

griseola).

Geographic distribution of major seedbornefungi of phaseolus beans

Important geographic features such as

latitude, longitude, and altitude were recorded

with the help of handheld Global Positioning

System (GPS) “Garmen” with 12 channels. Satellite

signals were received using the GPS and position

of the coordinates was recorded. The collected

coordinate information was downloaded into the

computer and all necessary descriptive information

were incorporated. Consequently, using the latitude

and longitude coordinates the survey route and

geographic distribution of major seedborne fungi

maps were prepared using Geographic Information

System (GIS) software (Arc view GIS) version

8.2.


Detection and identification of seedborne fungiof phaseolus beans

Symptoms of seed infection by seedborne

fungi varied depending on bean types and severity

of infection. Common bean seeds with severe

infection by C. lindemuthianum showed dark brown

spot associated with characteristic symptoms of

depressed sunken lesion both on pod and seed

surface (Figure 3 C & D). When seed infection was

not severe, it was difficult to differentiate symptoms

of different pathogen by the naked eyes.

This study showed that Phaseolus bean

seeds could be attacked by several economically

important seedborne fungal pathogens. A total of

thirteen seed-borne fungi of different genus was

identified. Seed infection levels of Phaseolus bean

with different seedborne fungi ranges between 0.2

- 14.5%. Bean seed infection levels varied between

localities, growing conditions, and bean types.

Seeds of common bean were more susceptible to

all major seedborne fungal pathogens than climbing

beans and green beans. High number of fungal

microflora was also associated with common bean

seeds (Table 1). The severity of seed infection by

the major pathogens was very high on seeds

collected from fields of small-scale farmer. This

was mainly because small-scale farmers did not

have the access to improve disease free seeds.

Instead, they used to grow their own saved seeds

from the previous harvests, without prior

knowledge on the primary inoculum present in the

seeds.


Bean anthracnose was found to be the most

widespread with high percent seed infection. C.

lindemuthianum was recovered from 26.2%, 19.6,

and 13.6% of the seed samples of common bean,

climbing bean and green bean respectively; whereas

P. griseola was recovered from 18.6% and 15.7%

of common bean and climbing bean respectively.

P. griseola and A. phaseolorum were not detected

from green bean seeds (Figure 1).

All these fungal pathogens are also

economically important seedborne pathogens of

beans in many parts of the world (Richardson,

1979). C. lindemuthianum is found to be the most

important seedborne pathogen and is found in high

frequency with severe crop damage followed by P.

griseola and A. phaseolorum in many bean growing

areas of Ethiopia. Chang et al., (2001) reported

that several species of fungal microflora were

isolated from naturally infected common bean

seeds in Canada and C. lindemuthianum was found

to be the most predominant fungus associated with

bean seeds.

In the western part of Ethiopia, small-scale

farmers grow climbing bean (Phaseolus coccineus)

in their gardens. All three major seed borne fungi

identified from common beans were also isolated

from climbing beans and all the isolates proved to

Phaseolus beans

Infe

cted

sam

ples

(%

)

0

5

10

15

20

25

30

C. lindemuthianum

P. griseola

A. phaseolorum

Com

mon

bea

ns

Clim

bing

bea

ns

Gre

en b

eans

26.2

18.6

9.9

19.6

15.7

9.8

13.6

0 0

Table 1 Numbers of infected seed sample and incidences (%) of seedborne fungi of common bean

(Phaseolus vulgaris) seed obtained from different agro-ecologies of Ethiopia during 2003 crop

season.

Fungi Numbers of infected sample Seed infection (%)

Range Average

Colletotrichum lindemuthianum 45 1.2-14.5 8.4

Phaeoisariopsis griseola 32 0.5-12.3 5.6

Ascochyta phaseolorum 17 0.6-7.4 3.7

Macrophomina phaseoli 10 0.3-4.5 2.8

Fusarium moniliforme 3 0.2-2.5 0.8

Fusarium oxysporum 5 0.3-2.3 0.9

Aspergillus sp. 6 1.2-4.5 2.4

Botrytis sp. 1 0.3-2.3 0.4

Phoma exigua 4 0.4-2.6 1.0

Sclerotinia sp. 1 0.6 0.6

Trichotecium sp. 3 0.2-1.6 0.5

Alternaria alternta 4 0.4-1.2 1.3

Figure 1 Infected samples (%) of Phaseolus bean

seeds by major seedborne fungi during

2003 crop season.


attack Phaseolus vulgaris. Since climbing bean is

a long duration crop and sometimes it can be

grown throughout the year as a garden crop, it

could serve as a primary source of inoculum of

these major seed borne pathogens from season to

season. Whereas, in the central rift valley of the

country where 60% of the country’s common bean

production is grown (Aleligne, 1990), green beans

are cultivated in state owned large commercial

farms and also private investors under irrigation in

continuous planting throughout the year. This type

of cropping system also creates favourable

condition for the survival and dissemination of

important seed borne pathogens such as C.

lindemuthianum. Noitalics rot of beans which is

caused by Phytophthora sp. was detected from

green bean pods and immature seeds collected

from commercial bean farms in the central rift

valley. This disease is common and causes damage

to the bean plant in the South and Central America

(Hall, 1994). In this study, Phytophthora rot of

beans caused by Phytophthora sp. was reported

for the first time in Ethiopia.

Detection of major seedborne pathogens

using the standard blotter method gave the higher

amount of seed infection (P < 0.01) as compared to

the agar plate method (Figure 4). This might be

due to the availability of free moisture on the moist

blotter papers than the agar plates. The bean seed

coat absorbed more water from the wet blotters

and enabled the fungi to easily grow and sporulate,

whereas on the agar plates, the bean seeds became

dry and sporulation of the fungus reduced as

compared to the standard blotter method. Blotter

method was the easiest, efficient and economical

detection technique for Colletotrichum

lindemuthianum, Ascochyta phaseolorum, and

Phaeoisariopsis griseola from naturally infected

Phaseolus bean seeds.

Geographic distribution of major seedbornefungi of Phaseolus beans

A total of thirteen seedborne fungi of

different genera were identified from samples

collected from various bean growing agro-

ecologies of Ethiopia (Table 1 & 2). Seed infection

of individual samples by different seedborne fungi

ranged from 0.2-14.5% and the most widespread

Table 2 Number of infected seed samples and infection percentage of climbing bean and green bean

seeds by different seedborne fungi obtained from different agro-ecologies of Ethiopia during

2003 crop season.

Seed infection (%)

Number of infected samples Range Average

Fungi Climbing Green Climbing Green Climbing Green

bean bean bean bean bean bean

Colletotrichum lindemuthianum 10 3 2.0 - 8.6 1.0 - 4.2 4.8 2.6

Phaeoisariopsis griseola 8 0 0.8 - 10.4 0 5.2 0

Ascochyta phaseolorum 5 0 1.1 - 6.4 0 3.2 0

Macrophomina phaseolina 4 0 0.4 - 2.4 0 1.2 0

Fusarium moniliforme 2 1 0.4 -1.8 0.9 1.0 0

Aspergillus sp. 4 2 1.6 - 5.8 1.1 - 2.6 3.2 1.4

Botrytis sp. 0 2 0 1.0 - 2.8 0 1.6

Tricotecium sp. 2 0 1.2 - 2.6 0 1.8 0

Phytophthora sp. 0 6 0 1.5 - 4.8 0 3.1

Alternaria alternata 6 2 1.8 - 6.5 1.0 - 2.3 3.4 1.7


seedborne fungi of Phaseolus beans were C.

lindemuthianum, P. griseola, and A. phaseolorum.

These pathogens attack major Phaseolus beans

and widely distributed in many of bean growing

areas of Ethiopia. The south, west, and north-

western part of the country is characterized with

high rainfall, whereas the central and eastern part

get low and erratic rainfall. However, during the

2003 crop season, there was high rainfall in all

parts of the country. This created favourable

condition for the development of some of the

seedborne pathogens even in the semi-arid areas.

In the western part of the country, where there is

high and frequent rainfall, all these pathogens

were found simultaneously in the same area or

bean field (Figure. 2). Bean anthracnose was the

most widespread and predominant fungal pathogen

in almost all bean growing areas of the country

including the lowland arid areas where there was

low and erratic rainfall during the crop season.

Even though there was field infection of beans by

bean anthracnose in the semi arid areas of the

central rift valley, seed infection by C.

lindemuthianum was very low. This might be due

to the late appearance of the disease in the field and

unfavourable environmental conditions during the

active growth stage of the bean plant. Angular leaf

spot and ascochyta blight were more abundant in

areas with high rainfall and humidity particularly

to the southern and western part of Ethiopia such

as Ambo, Bako, Pawe, Areka and Jimma area

(Figure 2). These fungi were not detected from

seeds collected from the central rift valley where

the bulk of dry common bean production is found

(Table 3).

Figure 2 Geographic distribution of major seedborne fungi of Phaseolus beans in Ethiopia during the

2003 main crop season (1- Colletotrichum lindemuthianum, 2-Phaeoisariopsis griseola, 3-

Ascochyta phaseolorum).


Table 3 Geographic distribution and severity of major seed borne fungi of common bean at different

bean growing areas during 2003 crop season.

Geographic Altitude Severity of major seedborne fungi

Location locations (M.a.s.l) C. lindemuthianum P. griseola A. phaseolorum

Melkassa 39 ∞.50' E 8 ∞.60' N 1650 ++ - -

Debrezeit 39 ∞.19' E 8 ∞.91' N 1940 ++ - -

Dejen 38 ∞.15' E 10 ∞.37' N 2476 + + -

Markos 37 ∞.91' E 10 ∞.58' N 2470 ++ + -

Finoteselam 37 ∞.43' E 10 ∞.84' N 1875 ++ ++ -

Chagni1 37 ∞.58' E 11 ∞.01' N 1665 ++ ++ -

Pawe1 36 ∞.60' E 11 ∞.50' N 1126 ++ +++ +

Bure1 37 ∞.21' E 10 ∞.77' N 2150 ++ +++ -

Guten 36 ∞.81' E 9 ∞.68’N 1404 + + -

Nekemt1 36 ∞.71' E 9 ∞.14' N 2130 +++ +++ ++

Bako1 37 ∞.23' E 9 ∞.11' N 1650 +++ +++ +++

Ambo1 37 ∞.86' E 8 ∞.64' N 2200 +++ +++ +++

Arjo1 36 ∞.71' E 8 ∞.75' N 2526 ++ + +

Bedele1 36 ∞.40' E 8 ∞.50' N 2027 ++ ++ +

Agaro1 36 ∞.64' E 7 ∞.89' N 1684 ++ +++ +

Jimma1 38 ∞.88' E 7 ∞.74' N 1750 ++ +++ ++

Wolkite1 37 ∞.96' E 8 ∞.42' N 1845 ++ ++ -

Areka1 37 ∞.78' E 7 ∞.08' N 1768 ++ +++ +

Sodo1 37 ∞.81' E 6 ∞.92' N 1850 ++ +++ -

Awassa1 38 ∞.52' E 7 ∞.30' N 1730 ++ +++ -

Shashemene 38 ∞.51' E 7 ∞.33’N 1940 ++ + -

Arssinegelle 38 ∞.79' E 7 ∞.41' N 1960 ++ + -

Ziway 38 ∞.94' E 8 ∞.06' N 1660 ++ + -

Meki 38 ∞.82' E 8 ∞.27' N 1675 + + -

Meisso 40 ∞.76' E 9 ∞.19' N 1350 + + -

Alemaya 42 ∞.27' E 9 ∞.43' N 2050 +++ - -

Shoarobit 40 ∞.00' E10 ∞.06' N 1260 ++ + -

Kemissie 39 ∞.97' E 10 ∞.14' N 1780 ++ + -

Kombolcha 39 ∞.85' E 11 ∞.11' N 1900 ++ ++ -

Dessie 39 ∞.70' E 11 ∞.08' N 1950 ++ + -

Kobbo 39 ∞.72' E 12 ∞.30' N 1450 ++ - -

1 - Areas with high rainfall

- = No disease; + = Slight; ++ = Moderate; +++ = Severe

M.a.s.l. - Meters above sea level


Figure 3 Symptoms of major seedborne fungal diseases under field condition on pods and seeds of

common beans, Paseolus vulgaris L. (A & B- Angular leaf spot, C & D-Anthracnose, E & F-

Ascochyta blight).

CONCLUSION

Basic information on the occurrences and

geographic distribution of major plant disease,

and their association with seeds is very important

for setting research priorities for further disease

management strategies in different agro-ecologies.

From this study, it could be concluded that several

seed borne fungi were highly associated with bean

seeds. The majority of fungi identified in this

study were known to be seed transmitted and

caused heavy crop loss in different bean growing

parts of the world. The presence of different seed

borne pathogens in bean seeds warrants for research

attention in the area of seed pathology. Seed

certification schemes must be strengthen during

production, and release of improved bean varieties

to the growers. Infection of bean seeds by large

number of seedborne fungi also implies the

importance of inspecting seed lots of Phaseolus

beans for potential seedborne pathogens in local

germplasm maintenance, introduction or exchange


Major seedborne fungal pathogens

Seed

infe

ctio

n (%

)

0

2

4

6

8

10

12

14

16

18

20Blotter methodAgar plate method

C. l

inde

mut

hian

um

P. g

rise

ola

A. p

hase

olor

um

a

b

a

b

a

b

Figure 4 Comparison of standard blotter and agar

plate methods for the detection of

seedborne fungi using naturally infected

common bean seeds.

between institutions and countries. Since there is

a change or shift in types and occurrence of plant

pathogens in different agro-ecologies, routine

disease surveys and identification work is important

for better understanding of fungal myco-flora

associated with Phaseolus bean seeds. Distribution

maps along with illustrations of the spread and

spatial patterns of diseases across regions should

be emphasised.

ACKNOWLEDGMENTS

The authors would like to acknowledge

Mr. Demeke Nigusse for his assistance on mapping

of survey routes and geographic distribution of

major seed borne fungi using GIS software.

This study was financed by the Ethiopian

Agricultural Research Organization (EARO)

through the Agricultural Research and Training

Project (ARTP).

LITERATURE CITED

Abdelmonem, A.M. 2000. Status of seed pathology

and seed health testing in Egypt. SeedSci. & Technol. 28: 533-547.

Agrios, G.N. 1997. Plant pathology. 4th ed.,

Academic Press, San Diego, 635p.

Agarwal, V.K. and J.B. Sinclair, 1997. Principlesof seed pathology, 2nd ed. CRC Lewis

publishers, New York. 539p.

Aleligne K. 1990. Farm survey and on-farm

research in haricot bean in the middle rift

valley of Ethiopia. In: Proceedings of anational workshop on research on haricotbean in Ethiopia, pp. 3-7, Addis Ababa,

Ethiopia, 1-3 October, 1990.

Amare, 1987. Haricot bean (Phaseolus vulgaris

L.) varieties performance and recommended

methods of production. pp. 229-251 In:

Proceedings of the 19th National CropImprovement Conference, Institute of

Agricultural Research, Addis Ababa, Ethiopia.

Central Statistical Authority (CSA). 2000. Area

under production of major crops. StatisticalBulletin No. 245, Addis Ababa, Ethiopia.

Chang, K.F., Y. yang, R. Conner, S.F. Hwang, and

R.J. Hward, 2001. Relationship of bean

seed infection with anthracnose to disease

development and seed microflora (Abstr.).

Can. J. Plant Pathol. 23: 196.

Habtu, A. and T. Abiy, 1995. Disease management

in lowland pulses: Progress and possibilities

for an integrated approach. In Habtu Assefa

(ed.). Proceedings of the 25th anniversaryof Nazareth Agricultural Research Center:

25 years of experience in lowland crops

research, 20-23 September 1995. Nazareth,

Ethiopia.

Habtu, A., S. Ivan, and J.C. Zadoks, 1996. Survey

of cropping practices and foliar disease of

common beans in Ethiopia. Crop Protection,

15 (2): 179-186.

Hall, R. 1994. Compendium of bean diseases.


2nd edition. APS Press, the American

Phytopathological Society, St. Paul,

Minnesota, 73p.

Hellene, R.D. 1988. Bean anthracnose fact sheet.Department of Plant pathology, New York

State Agricultural Experiment Station, Geneva

Cornell University, pp. 729-740.

International Seed Testing Association (ISTA)

1993. International rules for seed testing. SeedSci & Technol., 21, (Supplement), 288p.

Maden, S., D. Singh, S.B. Mathur, and P.

Neergaard, 1975. Detection and location of

seed-borne inoculum of Ascochyta rabiei and

its transmission in chickpea (Cicer arietinum).

Seed Sci. & Technol. 3: 667-681.

Maude, R.B. 1996. Seed-borne diseases and theircontrol: Principles and practices. CAB

International, University press, Cambridge,

280p.

Mathur, S.B. 1995. Some aspects of seed pathology

that deserve immediate attention. Indian J.Mycol. pl. Pathol. 25: 13-24.

Neergaard, P. 1977. Seed Pathology. Vol. I & II.

The Macmillan Press Ltd., London and

Asingstoke. 1187p.

Nelson, M.R., T.V. Orum, and R. Jaime-Garcia

1999. Application of Geographic Information

Systems and Geostatistics in Plant Disease

Epidemiology and management. Plant Dis.

83: 308-319.

Richardson, M.J. 1979. An annotated list ofseedborne diseases, Third edition,

Commonwealth Mycological Institute, Kew

Surrey, 320p.

Schwartz, H.F. and G.E. Galvez. 1980. Beanproduction problems: Diseases, insect, soil

and climate constraints of Phaseolus vulgaris,

Cali, Colombia, 424 p.


Short-Term Stressor Effects of Water Deprivation Priorto the Onset of Lay on Subsequent Reproductive Performance

of ISA Brown Pullets

Nirat Gongruttananun and Ratana Chotesangasa

ABSTRACT

The influence of water deprivation prior to the onset of production on sexual maturity and

subsequent reproductive performance was investigated in commercial pullets. Three hundred 16-week-

old ISA Brown pullets were used in this study. The birds were housed in cages (100 cages of 3 birds/cage)

situated in an open sided poultry shed and randomly divided into three treatment groups. The 3 groups

were 1) control (received feed with water at all times) 2) dehydrated and 3) dehydrated+NH4Cl.

Following an acclimatization period of 4 weeks, the birds in all groups were placed on a commercial layer

ration and the treatments began. In the dehydrated and dehydrated+NH4Cl groups, the drinking water was

removed completely from the pullets for 48 hours of dehydration but feed available at all times. The water

was then returned to the birds following the water withdrawal period, and thereafter until the end of the

trial. The pullets in the dehydrated+NH4Cl group were fed on the layer diet supplemented with 1%NH4Cl

throughout the experimental period. It was found that feed intake dropped rapidly, by approximately 50%,

when the birds were subjected to water deprivation. Neither age at first egg nor at 50-60 % of production

was influenced by the interruption of drinking water supply. Little difference in egg weight at first egg

was noted between the dehydrated birds and the normally hydrated hens. Pullets deprived of water were

slower coming into egg production especially during the first 2 weeks of production period, however,

thereafter egg production was similar among all treatment groups with numerical advantages for the

normally hydrated hens. There were no carryover effects of water deprivation on subsequent egg weight,

albumen weight, Haugh units, yolk weight or yolk color throughout the entire 12 weeks of the test. Body

weights were not different either before the treatment or at the end of the study (P>0.05).

The results might be interpreted as indicating that an interruption of drinking water supply for 2

days prior to the commencement of egg production induced transient effects of nutrient deficiency

resulted in retardation of reproductive development. The acidified layer ration failed to show any positive

effects on reproductive performance for the first 12 weeks of lay. The pullets, however, appeared to

overcome the detrimental effects as they approached sexual maturity.

Key words: water deprivation, sexual maturity, ammonium chloride, onset of lay, egg weight, egg

production

Department of Animal Science, Faculty of Agriculture, Kasetsart University, Bangkok 10900, Thailand.


INTRODUCTION

Water is recognized as a vitally important

nutrient for chickens. Lacking of drinking water

can have a disastrous effect on the performance of

the birds. Interruptions of the water supply could

happen as a result of failure of watering systems.

As cited by Marsden et al. (1965), there was an

occurrence of water deprivation at a turkey breeding

flock station due to the weekend attendant failing

to refill the water in the water supply. A mortality

rate of 13% was observed in older poults and 28%

in young birds deprived of drinking water for 2

days. Generally, the birds were weak, unsteady on

their feet, and some were dead. Symptoms were

ataxia and convulsions including violent flopping,

head retraction, gasping, falling over backwards,

and lying on their side or back, kicking and died.

The removal of drinking water in laying hens even

for a few days appears to have a long-term effect

on subsequent egg production. Sunde (1962)

reported that lack of drinking water for 36 hours

had a long term detrimental effect on body weight

and egg production. This author indicated that the

lowest egg production occurred on the fifth day

following water deprivation. Bierer et al. (1965)

found reduced feed intake and a sudden drop in

egg production in White Leghorn hens deprived of

water for 48 hours on Days 3 and 4 following water

deprivation. Adams (1973) deprived White

Leghorn laying hens of water for 48 to 72 hours

and found an 8-week period of declined egg

production. This finding was consistent with that

reported by Sunde (1962) who observed adverse

effects on egg production for 10 weeks in water-

deprived hens. Prolonged water deprivation has

been reported that severely affected the avian

kidneys resulting in renal failure and sudden death

(Siller, 1981). Glahn et al. (1988) demonstrated

that adding ammonium chloride in the diet could

reduce the incidence of kidney damages without

deleterious effects on egg production. However,

the many works reported have been conducted

with hens that had been in production for some

time and, thus, little or no information is available

on the influence of the dehydration mentioned on

sexual maturity.

The objective of the study was to determine

the effects of water deprivation for 48 hours, at the

point of lay, on sexual maturity and subsequent

reproduction performance in commercial pullets.

In addition, the use of a diet acidified with NH4Cl

in dehydrated birds was also examined, to

determine if dietary acidification could be effective

in preventing performance damages.


Three hundred ISA Brown pullets, 16 weeks

of age, were used in this study. The birds were

randomly divided and assigned into three treatment

groups as follows: 1) control, 2) dehydrated, and

3) dehydrated+NH4Cl. The pullets were allotted

to cages (100 cages of 3 birds/cage) placed in an

open-sided layer house, and received 16 hours of

light per day. An acclimatization period of 4

weeks was allowed. In the control group, the

pullets were provided food and water ad libitum

throughout the experimental period. In the

dehydrated and dehydrated+NH4Cl groups, the

drinking water was withdrawn completely for 48

hours of dehydration but feed was available at all

times. The water was then returned (rehydration)

to the dehydrated and dehydrated+NH4Cl groups

following the water withdrawal period. The pullets

in both control and dehydrated groups were fed a

layer diet, the birds in the dehydrated+NH4Cl

group received the layer diet supplemented with

1%NH4Cl throughout the study. The layer ration

used in this study was a mash commercial corn-

soybean ration, formulated to have a calculated

analysis of 3.25% Ca, 0.5% available phosphorous

(aP), and an ME value of 2,851 cal/kg. The acidified

layer diet was made by adding 0.45 kg of NH4Cl

to 44.9 kg of the layer diet (Glahn et al., 1988).

Observations of sexual maturity and subsequent



production performance of the experimental birds

were made and recorded. Daily egg production

was recorded and presented as the numbers of egg

per hen per 100 days (percentage of hen-day), and

feed intakes were determined during the 2 days of

water deprival period, and then biweekly thereafter

until 32 weeks of age. All eggs obtained from the

last day of each two weeks were assessed

individually for egg weight, albumen weight,

albumen height (Haugh units), yolk weight and

yolk color. Mean body weights of the experimental

bird were recorded at the beginning and the

termination of the trial.

The experiment was commenced in July,

2003 and terminated in December, 2003. The

average of minimum and maximum ambient

temperatures were 26.04 ± 1.98 ∞C and 33.29 ±0.66 ∞C, respectively. The statistical evaluation of

the data was performed by analysis of variance.

Repeated measures ANOVA was used where

appropriate. Mean values were compared using

Duncan’s multiple range test to determine

significance (Snedecor and Cochran, 1980).

Significance was assumed if P<0.05.

RESULTS

Generally, no deaths occurred during the

experimental period. The birds remained seemingly

in good health. During the water withdrawal period,

aggressive behavior was observed in the birds in

the water deprived groups, such as head movement

and nonnutritive pecking. On Day 2 of water

withdrawal, the dehydrated pullets appeared to

increase their alertness. The degree of aggression,

however, disappeared gradually as the drinking

water was reintroduced. Birds deprived of water

drank rapidly and avidly after the water was

returned.

Table 1 shows mean values of feed intake

of all treatment groups during and after water

deprivation. Feed intakes of the pullet in the two

groups of dehydration reduced sharply when the

drinking water was removed for 48 hours (P<0.05).

However, feed consumption of the dehydrated

hens caught up, and was similar to that of the

normally hydrated birds after 2 weeks of

replenishment of the drinking water and remained

so until the end of the experimental period (P>0.05).

Effects of water deprivation on ages at the onset of

lay and at 50% of production are illustrated in

Figure 1. Although the significant difference was

not noticed, the first egg of the control hens (144.7

days) was laid earlier than that of the dehydrated

birds (150 days) and dehydrated+NH4Cl hens

(148 days). However, the birds in all groups came

to 50% of egg production at the same age (P>0.05).

The size of the first egg was significantly affected

by the treatments as shown in Figure 2. It was

apparent that egg weight of the first egg of the

Table 1 Feed consumption during water deprivation and rehydration in the different treatment groups.

Group 48 hours-of Rehydration (weeks)

dehydration 2 4 6 8 10 12

Feed intake

(g/bird/2days) Feed intake (g/bird/d)

Control 155.0±12.5a 72.7±4.1 80.0±3.3 89.6±3.8 96.5±4.4 100.5±5.0 107.0±3.3

Dehydrated 77.0±8.3b 73.7±2.9 81.2±4.1 91.5±3.9 94.1±2.0 101.2±6.1 108.5±6.2

Dehydrated 62.5±7.9c 72.9±3.8 80.3±2.4 88.9±2.7 92.4±3.5 100.9±3.6 106.6±2.7

+NH4Cl

a-c Means with no common superscript differ significantly among groups (P<0.05).


control pullets (43.9 g) was similar to that of the

dehydrated birds (42.7 g), which was significantly

heavier than that of the dehydrated+NH4Cl hens

(39.3 g) (P<0.05). Influences of water deprivation

for 48 hours at the point of lay on subsequent egg

production are represented in Table 2. Obviously,

withdrawal of drinking water adversely affected

hen-day egg production during the initial state of

the production period. The dehydrated pullets laid

at a lower rate than did the normally hydrated

birds’ interval 20-22 weeks of age (P<0.05) whereas

egg production of the birds in the

dehydrated+NH4Cl group was not significantly

different from that of the birds in the control group.

However, thereafter egg production did not

significantly differ for all groups until the end of

the study. Details of egg quality measurement of

the different treatment groups for the first 12

weeks of production are summarized in Table 3.

Deprival of drinking water did not significantly

affect subsequent egg weight, or any aspect of egg

quality parameters throughout the experimental

period (P>0.05). Figure 4 presents mean values of

body weight of the different treatment hens at the

130

135

140

145

150

155

160

165

170

175

Age

(da

ys)

At first egg At 50% of lay

Group

control

dehydrated

dehydrated+ammonium chloride

Figure 1 Averages of age at the onset of lay and 50% of production of the different treatment hens.

36

37

38

39

40

41

42

43

44

45

46

Egg

wei

ght

(g)

Group

controldehydrateddehydrated+ammonium chloride

Figure 2 Averages of egg weight at first egg of the different treatment hens.


Table 3 Influences of interruption of drinking water supply for 48 hours prior to the onset of lay on

subsequent egg quality parameters in the different treatment groups.

Group Age (weeks)

22 24 26 28 30 32

Egg weight (g)

Control 48.7±2.6 50.8±1.0 54.9±1.3 56.4±2.6 55.6±1.9 56.1±3.3

Dehydrated 47.4±3.6 51.9±1.5 53.7±1.5 55.3±1.7 54.8±1.4 54.4±1.0

Dehydrated+NH4Cl 46.4±2.7 49.9±2.6 52.9±2.0 56.4±1.9 55.4±2.5 55.6±3.3

Haugh units

Control 93.6±4.5 96.8±5.2 91.8±2.4 88.7±4.2 90.0±2.2 87.1±6.0

Dehydrated 94.3±6.4 98.0±2.2 94.3±2.1 90.2±0.7 91.3±2.4 89.0±1.0

Dehydrated+NH4Cl 95.2±4.1 95.9±2.7 94.0±3.7 92.1±3.2 88.7±4.2 86.5±4.7

Yolk color

Control 10.1±0.8 11.8±0.3 11.9±0.3 12.5±0.2 11.8±0.3 12.1±0.1

Dehydrated 10.4±0.4 12.2±0.3 12.4±0.3 12.2±0.2 12.2±0.1 12.1±0.2

Dehydrated+NH4Cl 10.3±0.4 12.3±0.4 12.2±0.2 12.2±0.2 11.8±0.3 12.1±0.2

Yolk weight (%)

Control 22.5±1.2 22.5±0.8 22.7±0.4 22.9±0.4 23.2±1.2 22.6±0.3

Dehydrated 21.3±1.2 22.1±0.9 22.2±0.2 23.4±0.2 22.6±0.6 23.1±0.6

Dehydrated+NH4Cl 23.3±1.9 22.6±0.9 23.0±0.9 23.0±0.2 23.1±0.2 23.6±1.5

Albumen weight (%)

Control 67.8±1.4 68.1±0.6 67.8±0.7 67.7±0.8 67.5±0.8 67.3±0.7

Dehydrated 69.4±1.5 68.2±0.8 68.3±0.7 67.1±0.6 68.1±0.7 67.1±0.6

Dehydrated+NH4Cl 67.2±2.1 68.5±0.8 67.9±0.7 67.5±0.3 67.4±0.2 66.6±1.7

There were no significant differences between groups (P>0.05).

All data in the tables are given as mean ± standard deviation.

Table 2 Effects of water deprivation for 2 days during the pullet-laying transition period on subsequent

egg production in the different treatment groups.

Group Age interval (weeks)

20-22 22-24 24-26 26-28 28-30 30-32

Egg production (%)

Control 11.4±6.5a 34.8±8.5 58.8±7.8 74.9±4.2 74.6±6.3 82.8±3.5

Dehydrated 3.8±1.1b 32.3±10.6 54.2±8.9 71.9±7.3 73.0±6.3 82.2±3.8

Dehydrated+NH4Cl 6.8±4.8ab 24.4±14.0 49.5±13.5 74.6±3.8 73.7±7.2 80.2±6.0

a-b Means with no common superscript differ significantly among groups (P<0.05).


beginning and the end of the investigation. No

significant differences were observed in body

weight among groups before the treatment or at the

end of the experiment.

DISCUSSION

No adverse effects of water deprivation for

2 days at the onset of lay were observed on body

weights of the bird as illustrated in Figure 4.

However, the hens in the study were not weighed

during the water removal. The water deprival

period of 48 hours caused no apparent behavioral

ill effects on the hens, except for an aggressive

behavior. The pullets in both dehydration groups

showed an increase in aggressive pecks during the

water withdrawal period, especially on Day 2 of

water deprivation. The increased degree of

aggression might arise from frustration (Duncan

and Wood-Gush, 1971) resulting from hens

expecting the availability of drinking water but not

finding it. The finding of aggressive behavior in

the dehydrated hens in the study was similar to that

observed in White Leghorn laying hens deprived

of food (Simonsen, 1979). It is well documented

that deprivation of food or water enhances levels

of plasma corticosterone in avian species (Freeman

et al., 1980; Arnason et al., 1986). Corticosterone

hormone is secreted from the adrenal glands,

located anterior and medial to the cephalic lobe of

the avian kidneys (Ringer and Meyer, 1976).

Freeman et al. (1980) indicated that starvation was

a strong stimulus for the releasing of corticosterone

in domestic birds.

The findings of this study, that removal of

drinking water for 48 hours prior to the onset of lay

reduces feed intake, were consistent with those

reported by several workers (Sunde, 1962; Adams,

1973). Korr (1939) reported that the birds

responded to water deprivation by reducing

glomerular filtration rates (GFR) and urine flow

rates, and increasing the osmotic pressure of blood

and urine. Later, dehydration was reported to

induce an increase in plasma osmotic

concentrations (Koike et al., 1983). These findings

were consistent with those reported by Roberts

(1991).

Sexual maturity of the experimental pullets,

as estimated from age at first egg and at 50%

production, was not significantly affected by

interruption of drinking water which would be a

reflection of a similar body weight of the

experimental birds (Figure 4). Several reports

from previous studies (Summers et al., 1987;

Summers and Leeson, 1994) have shown that

there is a specific body weight threshold for the

onset of production of birds. The investigators

suggested that pullets must achieve the certain

body weight in order to trigger the onset of

production. It was well documented from previous

several studies (Brody et al., 1980; Leeson and

Caston, 1991) that regardless of feed or

management regimen, if the pullet had the chance

to consume diets at all times, the animal could

quickly catch up in body weight so that a uniform

weight was obtained at the onset of lay. The results

from the study indicated that, as the drinking water

was withheld, feed intake decreased markedly by

approximately 50-60 % comparing with that

consumed by the normally hydrated hens (Table

1). This might be comparable to the circumstance

of feed restriction for 2 days. However, it was of

interest to notice that feed intake in the two

dehydration groups increased rapidly and returned

to the normal value within 2 weeks after the birds

were allowed access to the drinking water again.

This demonstrated the ability of the pullets to

increase feed consumption in an attempt to achieve

a mature body weight as they approached sexual

maturity.

The results from this study revealed that

the interruption of drinking water for 48 hours at

the point of lay had minor effects on subsequent

reproductive performance due to feed restriction.

If one compared feed intake of the experimental

birds during water deprivation (Table 1) with the


average of egg weight at first egg (Figure 2), it was

likely that the amount of nutrient intake supported

egg size at the initial phase of lay. In addition, the

dehydrated pullets were coming into production

slowly than the conventionally reared birds as

evidenced by hen-day egg production during the

first 2 weeks of production period (Table 2), and

age at first egg. It did appear to be a trend in the

dehydrated birds for requirement of much more

times to be ready-to-lay pullets as compared with

the control pullets (Figure 1). Likewise, the trend

for slightly lower production was evident for the

hens in the two groups of water withdrawal as

depicted in Figure 3, confirming by the experiences

of others (Isaacks et al., 1960; Brake et al., 1985;

Summers et al., 1991) that feed restriction during

the prelaying phase had only a slightly adverse

effect on subsequent egg production. Summers et

al. (1991) deprived White Leghorn pullets of food

for 6 days at 17 weeks of age, they found that the

6-day feed withdrawal period delayed sexual

maturity and reduced egg weight only during the

first 2 weeks of production. Brake et al. (1985)

stated that nutrient intake during the prelay period

was a main factor influencing performance of

hens. These authors suggested that protein intake

0

10

20

30

40

50

60

70

80

90

20-22 22-24 24-26 26-28 28-30 30-32

Age interval (weeks)

Hen

-day

pro

duct

ion

(%)

control

dehydrated

dehydrated+ammonium chloride

1.1

1.2

1.3

1.4

1.5

1.6

Bod

y w

eigh

t (kg

)

At the beginning At the end

Age

controldehydrateddehydrated+ammonium chloride

Figure 4 Averages of body weight at the beginning and the end of the experiment of the different

treatment birds.

Figure 3 Hen-day egg production of hens on the different treatments from 20 to 32 weeks of age.


was the major factor allowing accumulation of

protein reserves in the body, which resulted in

increased egg production and egg weight.

Presumably, in this study, the dehydrated birds

consumed inadequate amount of nutrients

especially protein intakes, resulting from a

reduction of feed intake during 2 days of water

deprivation, in which the degrees of nutrient

deficiency would be so much severely that affected

reproductive development. This would indicate

that such conditions adversely affect carrying

reserves of body composition of the pullets at the

point of lay, thereby poor production and small

egg size at first egg were observed in the laying

house. Unfortunately, the data of gonads of the

birds in the current experiment were not

determined. Additional studies are needed to

conduct whether water deprivation influences the

development of reproductive organs of the domestic

fowl.

The results obtained from the study

indicated that withdrawal of drinking water supply

for 48 hours prior to the onset of egg production

had transiently adverse effects on subsequent

reproductive performance as evidenced by reducing

of egg weight at first egg and egg production

during the initial state of lay, suggesting that due

to reduced feed intake during the dehydration

period resulted in protein and/or nutrient

deficiencies, therefore growth and reproductive

development was retarded. However, the pullets

deprived of water tended to overcome the nutrient

restricted stress effect as they approached sexual

maturity. This would be a reflection of the ability

of the animals to compensate for such circumstance

by increasing feed intake and body composition.

Acidification diets with NH4Cl did not exhibit any

beneficial effects on sexual performance for the 12

weeks of observation period. These results were in

agreement with the results reported by Glahn et al.

(1988). The workers demonstrated that dietary

acidification with NH4Cl at a level of 1% did not

affect production performance. They also indicated

that a major advantage of the NH4Cl acidified diet

was that it reduced the incidence of kidney lesions.

CONCLUSION

Lacking of drinking water for 48 hours,

prior to the onset of lay, did not affect sexual

maturity, body weight, egg weight or subsequent

reproductive performance. However, the pullets

deprived of water commenced to lay slightly late

during the initial period of production. Obviously,

feed consumption of the birds dropped abruptly as

the birds were subjected to water deprivation. It

was possible that dehydration induced a short-

term stressor, leading to a reduction in feed intake.

Reduced feed consumption resulted in nutrient

deficiencies and retardation of reproductive

development. No beneficial effects on reproductive

performance of adding 1%NH4Cl to feed were

evident.

ACKNOWLEDGEMENTS

The authors wish to acknowledge the

financial support of the Kasetsart University

Research and Development Institute (KURDI) in

the carrying out of this work.

LITERATURE CITED

Adams, A.W. 1973. Consequences of depriving

laying hens of water a short time. Poultry Sci.52: 1221-1223.

Arnason, S.S., G.E. Rice, A. Chadwick and E.

Skadhauge. 1986. Plasma levels of arginine

vasotocin, prolactin, aldosterone and

corticosterone during prolonged dehydration

in the domestic fowl: effect of dietary NaCl.

J. Comp. Physiol B156:383-397.

Bierer, B.W., T.H. Eleazer and D.E. Roebuck.

1965. Effect of feed and water deprivation on

chickens of various ages. Poultry Sci. 44:

1351.


Brody, T., Y. Eitan, M. Soller, N. Nir and Z.

Nitsan. 1980. Compensatory growth and

sexual maturity in breeder females reared

under severe food restriction from day of

hatching. Br. Poultry Sci. 21: 437-446.

Brake, J., J.D. Garlich and E.D. Peebles. 1985.

Effect of protein and energy intake by broiler

breeders during the prebreeder transition

period on subsequent reproductive

performance. Poultry Sci. 64: 2335-2340.

Duncan, I.J.H. and D.G.M. Wood-Gush. 1971.

Frustration and aggression in the domestic

fowl. Anim. Behav. 19: 500-504.

Freeman, B.M., A.C.C. Manning and I.H. Flack.

1980. Short-term stressor effects of food

withdrawal on the immature fowl. Comp.Biochem. Physiol. 67A: 569-571.

Glahn, R.P., R.F. Wideman, Jr. and B.S. Cowen.

1988. Effect of dietary acidification and

alkalinization on urolith formation and renal

function in single comb white leghorn laying

hens. Poultry Sci. 67: 1694-1701.

Isaacks, R.E., B.L. Reid, R.E. Davies, J.H.

Quisenberry and J.R. Couch. 1960. Restricted

feeding of broiler type replacement stock.

Poultry Sci. 39: 339-346.

Koike, T.I., L.R. Pryor and H.L. Neldon. 1983.

Plasma volume and electrolytes during

progressive water deprivation in chickens

(Gallus domesticus). Comp. Biochem.Physiol. 74A: 83-87.

Korr, I.M. 1939. The osmotic function of the

chicken kidney. J. Cell. Comp. Physiol. 13:

175-179.

Leeson, S. and L.J. Caston. 1991. Growth and

development of Leghorn pullets subjected to

abrupt changes in environment temperature

and dietary energy level. Poultry Sci. 70:

1732-1738.

Marsden, S.J., G.S. McKee and M.L. Crandall.

1965. Water deprival and replenishment in

poults. Poultry Sci. 44: 793-797.

Ringer, R.K. and D.C. Meyer. 1976. Parathyroids,

Ultimobrachial Bodies, and the Pineal, pp.

359-371. In P.D. Sturkie (ed.). AvianPhysiology 3rd ed. Springer-Verlag, New

York.

Roberts, J.R. 1991. Effects of water deprivation on

renal function and plasma arginine vasotocin

in the feral chicken Gallus gallus

(Phasianidae). Aust. J. Zool. 39:439-446.

Siller, W.G. 1981. Renal pathology of the fowl –

A review. Avian Patho. 10: 187- 262.

Simonsen, H.B. 1979. Effect of feed withdrawal

on behavior and egg production in white

leghorns on litter and wire. Br. Vet. J. 135:

364-369.

Snedecor, G.W. and W.G. Cochran. 1980.

Statistical Methods 7th ed. The Iowa State

University Press, Ames, Iowa, USA.

Summers, J.D. and S. Leeson. 1994. Laying hen

performance as influenced by protein intake

to sixteen weeks of age and body weight at

point of lay. Poultry Sci. 73: 495-501.

Summers, J.D., S. Leeson and D. Spratt. 1987.

Rearing early maturing pullets. Poultry Sci.66:1750-1757.

Summers, J.D., D. Spratt and J.L. Atkinson. 1991.

Delaying sexual maturity of pullets by nutrient

restriction at the onset of production. Can. J.Anim. Sci. 71: 1215-1221.

Sunde, M.L. 1962. Amino acids, proteins, and

stuff. Poultry Sci. 41: 1688.


Pharmacokinetics and Withdrawal Times of Enrofloxacin in Ducks

Natthasit Tansakul1, Amnart Poapolathep1, Naruamol Klangkaew1,Napasorn Phaochoosak1 and Wanida Passudaruk2

ABSTRACT

The pharmacokinetic properties of enrofloxacin (EFX) were investigated in healthy ducks

following a single administration of EFX with a dose of 10 mg/kg of body weight by intravenous (i.v.),

intramuscular (i.m.), subcutaneous (s.c.) or oral (p.o.) route. The plasma concentration-time curve was

analyzed using a two compartment model. Mean peak plasma concentration of EFX was 11.49 ± 1.17,

5.65 ± 0.36, 4.99 ± 0.87 and 4.87 ± 0.69 mg/ml after i.v., i.m., s.c. and p.o. administration, respectively.

After a single i.v. administration, the pharmacokinetic parameters were found as follow; the elimination

half-life (t1/2b) = 6.47 ± 2.85 h, the elimination rate constant (Kel ) = 0.70 ± 0.06 h-1,the apparent volume

of distribution Vd(area) = 1.30 ± 0.22 L/kg and the total body clearance (ClB) = 0.89 ± 0.07 L/kg/h.

Difference enrofloxacin bioavailability following i.m., s.c. and p.o. administration were 98.77 ± 0.05 %,

85.11 ± 2.71 % and 80.35 ± 0.29%, respectively. The results of pharmacokinetic properties of EFX in

ducks should be provided with the dosage regimen, preslaughter withdrawal times and maximum residue

limits for ducks.

Key words: pharmacokinetic, withdrawal time, antibiotic, enrofloxacin, duck species

INTRODUCTION

Enrofloxacin (1-cyclopropyl-6-fluoro-1,4-

dihydro-4-oxo-7-[4-ethyl-1-piperazinyl] -3-

quinoline carboxylic acid) is an antimicrobial

substance which belongs to the fluoroquinolones

groups. This agent reportedly has excellent

activities against a wide range of aerobic gram-

negative bacteria. It is also active against gram-

positive bacteria and Mycoplasma spp. Therefore,

EFX is commercialized for animal use and potential

therapeutic application for many types of infection

(García-ovando et al.,1999). Similar to that of

1 Department of Pharmacology, Faculty of Veterinary Medicine, Kasetsart University, Bangkok 10900. Thailand.2 Department of Microbiology and Immunology, Faculty of Veterinary Medicine, Kasetsart University, Bangkok 10900.

Thailand.


other quinolones, these compounds act on inhibition

of DNA gyrase and exhibit a bactericidal and

mycoplasmacidal activity at low concentrations.

The efficacy of EFX reportedly inhibits in vivo

replication of certain organisms that are resistant

to antibacterial substances i.e., beta-lactam

antibiotics, aminoglycosides, tetracyclines, folic

acid antagonists and macrolides (Anadón et al.,

1995). Limited information is available on

disposition, metabolism and safety of EFX use in

commercial ducks. The objective of the present

study was to investigate the fundamental

pharmacokinetic value of EFX on ducks following


intravenous (i.v.), intramuscular (i.m.),

subcutaneous (s.c.) and oral (p.o.) administration.

Thereafter, the proper therapeutic regimen of EFX

should be concerned for ducks.


AnimalsHealthy ducks of an average 1.09 ± 0.24 kg

body weight, without previous treatment, were

used in the study. Ducks were separated into four

groups (30 ducks per group). The animals were fed

with a commercial standard diet that was free from

any chemotherapeutics three times per day. Water

supply was provided ad libitum. Throughout the

study they were housed in the animal cage at

Division of Experimental Animal, Faculty of

Veterinary Medicine, Kasetsart University.

Drug administration and sample collectionCommercial enrofloxacin containing 50

mg/ml (Baytril“ 5% sterile solution, Bayer AG,

Leverkusen,Germany) was prepared for i.v., i.m.,

s.c. and p.o. administrations at the same dose of 10

mg/kg body weight for each duck. Randomized

2.5 ml of heparinized blood were taken from the

brachial vein in the following preset times:0.0,

0.15, 0.30, 1, 2, 3, 4, 5, 6, 8, 10, 12, 20, 24, 30, 48,

54, 72, 78 and 92 h. Blood samples were collected

and centrifuged (3000 X g) for 15 min to collect

the plasma (García-ovando et al.,1999), placed in

a 1.5 ml Eppendorf vial (Laboratory Product, Inc.,

Rochester, NY.) and stored at -20∞C until analysis.

Method of analysisThe concentration of EFX was analyzed

using a microbiological diffusion method (Bennett

et al.,1966; Anhalt,1985; Limpoka,1992). The

method used Escherichia coli ATCC 25922

(Scientific and Technology Institute of Thailand)

as test organisms. Standard dose-response curves

were obtained using buffer EFX solution. The

motten agars were prepared by inoculated with the

organisms in broth. Then the medium was poured

32 ml into each 10 ¥ 15 cm glass plate. After

hardened, 10 mm diameter wells were punched 8

holes per plate. Then the plasma samples and

standard control (2 holes) were examined. The

samples were allowed to diffuse for 45 min at

room temperature prior to incubation for 24 h at

37∞C. Thereafter, the inhibition zone of the standard

preparations and samples were measured using a

caliper vernia. The concentrations were recorded

from plots of log concentration plus zone diameter

of plasma.

Calculation of pharmacokinetic parametersThe pharmacokinetic values of EFX on

plasma concentrations after a single i.v.

administration were evaluated by a semilogarithm

modified standard technique. A bi-exponential

equation was selected for all ducks having been

given the drug by the i.v. route and consequently

the data were described by a two-compartment

open model based on the criteria of improvement

in the sum square by plotting of residuals. The

following pharmacokinetic parameters were

obtained according to the conventional equations

previously described by Baggot (1977), Limpoka

(1992) and Craigmill et al. (1994).

The following equations were used to obtain

these pharmacokinetic parameters for a two-

compartment pharmacokinetic model.

t1/2a = ln 2/at1/2b = ln 2/bK21 = A(b) + B(a)/ A+B

Kel = (a)(b)/ K21

K12 = a+b- K21 - Kel

Vd(area) = Dose/Cpo

AUC = (A/a) + (B/b)

F = AUCother / AUCi.v

ClB = (Kel) (Vc)

The term of Cpo is the extrapolated plasma

concentration to determined the zero- time profile.

B was calculated from the elimination phase (B-

slope). A was calculated by the residual method


(O’Flaherty, 1981). The a and b are hybrid rate

constants describing the initial and terminal decline

in plasma concentration and are composed of the

microrate constants (K12,K21) of the model. The

t1/2a (distribution half-life), t1/2b (elimination half-

life), AUC (area under the curve), Vd(area) (apparent

volume of distribution during the post-distribution

phase), Bioavailability and ClB (total body

clearance) were calculated.

Statistic analysisThe pharmacokinetic parameters were

calculated by CA-Cricket Graph III, version 1.5J,

Computer Associates Inc., NY., U.S.A. Statistical

analysis of data was performed using Microsoft

Excel, Window XP.

RESULTS

After a single i.v. administration of 10 mg/

kg of body weight of EFX in ducks, the mean ± SD

pharmacokinetic parameters were calculated and

described by a two-compartment open model.

Distribution half-life (t1/2a ) was 0.60 ± 0.02 h,

whereas the elimination half-life(t1/2b) was 6.47 ±2.85 h. Table 1. presents the pharmacokinetic

parameters.

Comparison of the mean ± SD plasma

concentration-time profile of EFX at various routes

are shown in Table 2. and Figure 1. EFX was

absorbed rapidly. Concentrations of EFX peaked

within 30 min by i.m. administration while the

peak levels of s.c and p.o. administration were

found within 1 h. However, these levels were

higher than the therapeutic level (Anonymous;

1987).

DISCUSSION

Pharmacokinetic variables of EFX after

the i.v. administration were best described by a

two-compartment open model, with a rapid

distribution phase (t1/2a = 0.6 h) and a moderately

prolong elimination phase (t1/2b = 6.47 h).

Because of limited reports of

fluoroquinolones in ducks, the time to maximum

concentration (tmax) differed among enrofloxacin,

ciprofloxacin and norfloxacin in chicken were

applied as reference. The significant differences

(p<0.05) were found that the tmax of ciprofloxacin

(0.42 ± 0.08 h) (Atta and Sharif,1997) was reached

more rapidly than that of enrofloxacin (1.64 ± 0.04

h) (Anadón et al.,1995) and norfloxacin (1.99 ±0.17 h) (Laczay et al.,1998) after oral

administration. In addition, the peak plasma

concentration (Cmax) of ciprofloxacin was the

highest (4.67 ± 0.33 mg/ml),which was higher than

that of the enrofloxacin (2.44 ± 0.64 mg/ml) and

norfloxacin (1.46 ± 0.18 mg/ml).

Table 1 Pharmacokinetic data (mean ± SD) for

enrofloxacin determined following

intravenous administration at a single

dose of 10 mg/kg of body weight in

ducks.

Pharmacokinetic Enrofloxacin

parameters (units)

Cpo (mg/ml) 15.73 ± 2.60

A (mg/ml) 14.67 ± 3.38

a (h-1) 1.16 ± 0.03

B (mg/ml) 1.35 ± 0.90

b (h-1) 0.13 ± 0.07

t1/2a (h) 0.60 ± 0.02

t1/2b (h) 6.47 ± 2.85

K12 (h-1) 0.37 ± 0.003

K21 (h-1) 0.22 ± 0.13

Kel (h-1) 0.70 ± 0.06

Vd(area) (L/kg) 1.30 ± 0.22

ClB (L/kg/h) 0.89 ± 0.07

Bioavailabilityi.m. (%) 98.77 ± 0.05

Bioavailabilitys.c. (%) 85.11 ± 2.71

Bioavailabilityp.o. (%) 80.35 ± 0.29

Note: Pharmacokinetic parameters of EFX were determined by

a two-compartment pharmacokinetic model.


Table 2 Mean ± SD plasma concentrations of enrofloxacin in ducks following i.v., i.m., s.c. or p.o.

administration at a single dose of 10 mg/kg of body weight.

Hours after dosing Plasma concentrations (mg/ml)

i.v. i.m. s.c. p.o.

0.15 11.49 ± 1.17 5.65 ± 0.36 4.99 ± 0.87 4.87 ± 0.69

0.30 10.46 ± 1.85 8.97 ± 1.33 6.48 ± 1.07 5.55 ± 0.70

1.00 5.59 ± 0.28 7.99 ± 0.63 8.29 ± 0.62 7.61 ± 1.20

2.00 4.00 ± 0.25 7.60 ± 0.68 7.19 ± 1.12 5.99 ± 0.34

3.00 2.08 ± 0.36 6.76 ± 0.99 7.06 ± 0.58 5.74 ± 0.60

4.00 1.97 ± 0.15 5.72 ± 0.58 6.67 ± 0.73 4.62 ± 0.39

5.00 1.06 ± 0.16 3.90 ± 0.35 4.44 ± 0.68 4.43 ± 0.59

6.00 0.86 ± 0.19 3.57 ± 0.41 2.95 ± 1.13 3.99 ± 0.73

8.00 0.47 ± 0.07 1.51 ± 0.17 1.46 ± 0.10 2.28 ± 0.21

10.00 0.43 ± 0.04 1.43 ± 0.17 1.42 ± 0.25 2.06 ± 0.43

Figure 1 Comparative mean plasma concentration-time profile of enrofloxacin (EFX) following single

i.v.,i.m.,s.c.and p.o. administrations of 10 mg/kg b.w. in ducks.

Hour after dosing (hr.)

0 1 2 3 6 8 10

Plas

ma

conc

entr

atio

ns (u

g/m

l)

0

2

4

6

8

10

12

14

i.v.i.m.s.c.p.o.

4 5 7 9

A similar kinetic profile was also observed

in chickens after the i.v. administration, the biphasic

nature of the plasma concentration-time curve has

been reported for EFX (Anadón et al.,1995;García-

ovando et al.,1999). In the present study, the

elimination half-life (6.47 ± 2.85 h) was also

higher than that recorded in healthy dogs (3.4 h),

cattle (1.7h), sheep (3.7 h), horses (5.0 h) and pigs

(5.5 h) (Baggot,2001). However, this parameter

was lower than that previously reported in chickens

(6.99 ± 0.48 h) (García-ovando et al.,1999).

Fluoroquinolones are lipid-soluble

chemical agents, and their typical Vd values are 2-

4 L/kg (Brown,1996). Nevertheless, lower Vd

values i.e., 1.94 ± 0.14 L/kg have been reported for

EFX in chickens (García-ovando et al.,1999)


The mean ± SD bioavailability of EFX inducks was 98.77 ± 0.05% after the i.m.administration, therefore it is likely that the doseof EFX was almost completely absorbed. Thebioavailability value of the i.m. administrationwas also higher than those of the s.c. (85.11 ±2.71%) and p.o. (80.35 ± 0.29%). Moreover, thedrug was detected and remained in the plasma upto 20 h after the s.c. and i.m. administrations whileit was up to 24h after the i.v. and p.o.administrations.

In conclusion, The biphasic nature of plasmaconcentration-time curve suggested that a two-compartment pharmacokinetic model wouldprovide an accurate description of pharmacokineticbehaviors. The pattern of plasma concentration-time profiles between EFX and the otherfluoroquinolones were identical following i.m.,s.c. or p.o. administration . According to the resultsof this study a dose of 10 mg/kg body weight ofenrofloxacin in ducks may be appropriate for theroutes investigated. However, the tissue residuesshould be further determined by an HPLC assay toget insight into the tissue uptake and the properwithdrawal times of EFX in ducks.

ACKNOWLEDGEMENTS

This study was financially supported bythe Kasetsart University Research andDevelopment Institute (KURDI), THAILAND.The researchers wish to thank Prof. Dr. MalineeLimpoka for her advice on the method of analysis.Special thanks are extended to Kanokwan Bangnoiand Sasithorn Limsuwan for their advice on theuse of computer program.

LITERATURE CITED

Anadón, A., M.R. Martinez-Larranaga, M.J. Diaz,P. Bringas, M.A. Martinez, M.L. Fernandez-Cruz, M.C. Fernandez and R. Fernandez.1995.Pharmacokinetics and residues of enrofloxacinin chickens.Amer. J. of Vet. 56: 501-505.

Anhalt, J.P. 1985. Antimicrobial assays. In

Laboratory Procedures in ClinicalMicrobiology. 2nd ed., J.A. Sprinter Verlag,Washington, New York. 691 p.

Anonymous .1987. Baytril“ Broad-spectrumAntiinfective for the treatment of BacterialDiseases in Animals. Product information.Bayer Veterinary Department, Leverkusen,Germany. 27 p.

Atta, A.H. and L. Sharif. 1997. Pharmacokineticsof ciprofloxacin following intravenous andoral administration in broiler chickens. J. vet.Pharmacol. Therap. 20: 326-329.

Baggot, J.D.1977.Principles of Drug in DomesticAnimals.W.B. Saunders, Philadelphia. 238p.

Baggot, J.D.2001. The Physiological Basis ofVeterinary Clinical Pharmacology. IowaState University Press. 283p.

Bennett, J.V., J.L. Brodie, E.Benner and W.M.M.Kirby. 1966. Simplified, accurate method forantibiotic assay of clinical specimens. AppliedMicrobiology. 14: 170-175.

Brown, S.A. 1996. Fluoroquinolones in animalhealth. J.Vet.Pharmacol.Therap.19: 1-14.

Craigmill, A.L., S.F. Sundlof and J.E. Riviere.1994. Handbook of ComparativePharmacokinetics and Residues ofVeterinary Therapeutic Drugs, CRC Press,Inc., Boca Raton, Fla. 665 p.

García- ovando, H., N. Gorla, C. Luders, G. Poloni,C. Errecalde, G. Prieto and I. Puelles. 1999.Comparative pharmacokinetics ofenrofloxacin and ciprofloxacin in chickens.J. Vet. Pharmacol. Therap. 22: 209-212.

Laczay, P., G. Semjen and J. Lehel .1998.Comparative studies on the pharmacokineticsof norfloxacin in chickens,turkeys and geeseafter a single oral administration. J. Vet.Pharmacol. Therap. 21: 161-164.

Limpoka, M. 1992. Principle Pharmacokineticin Animals. Charulsanitwong, Bangkok. 195p.

O’ Flaherty, E. 1981.Toxicants and Drug:Kinetics and Dynamics.A Wiley IntersciencePublication, New York. 398 p.


Antimicrobial Resistance of Campylobacter jejuni Isolated fromChicken in Nakhon Pathom Province, Thailand

Jananya Sukhapesna1, Patamaporn Amavisit2, Worawidh Wajjwalku3,Arinthip Thamchaipenet4 and Thavajchai Sukpuaram5

ABSTRACT

Campylobacter jejuni isolated from retail market-chicken in Nakhon Pathom province were

determined for resistance to quinolone and other antimicrobial agents by broth microdilution method.

Sixty-eight C. jejuni strains were resistant to quinolone drugs including nalidixic acid (69.12%),

norfloxacin (69.12%), ciprofloxacin (58.82%) and marbofloxacin (25.00%). High proportions of the

isolates were resistant to tetracycline (77.94%), sulfamethoxazole (72.06%), kanamycin (51.47%),

ampicillin 47.06% and streptomycin (42.65%). Low proportions of the isolates were found resistance to

gentamycin (16.18%) and erythromycin (13.23%). Nearly 97% of the isolates were multiple resistances

to more than 4 antimicrobial agents tested.

Key words: Campylobacter jejuni, broth microdilution method, MICs

INTRODUCTION

Infection with Campylobacter species has

emerged worldwide as one of the leading causes of

diarrhea (Rautelin et al., 2003). C. jejuni is one of

the main species involved in human infection

(Saenz et al., 2000). The genotyping and serotyping

analysis revealed that poultry can be a source of

Campylobacter infection and the contamination

occurred by direct ingestion of undercooked food

or cross contamination of raw poultry to other

foods (Engberg et al.,2001).

Fluoroquinolones and macrolides have been

widely used for treatment of Campylobacter

1 Center for Agricultural Biotechnology, Kasetsart University, Kamphaengsean Campus, Nakhon Pathom 73140, Thailand.2 Department of Microbiology and Immunology, Faculty of Veterinary Medicine, Kasetsart University, Bangkok 10900,

Thailand.3 Department of Pathology, Faculty of Veterinary Medicine, Kasetsart University, Kamphaengsean Campus, Nakhon Pathom

73140, Thailand.4 Department of Genetics, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand.5 Department of Public Health, Faculty of Veterinary Medicine, Kasetsart University, Kamphaengsean Campus, Nakhon

Pathom 73140, Thailand.


infections (Aquino et al., 2002). The increasing

proportions of Campylobacter isolates have been

reported to be resistant to these drugs (Endtz et al.,

1991). The over uses of antimicrobial agents in

veterinary medicine or as feed additives might

result in the emergence and spread of resistance

among Campylobacter strains. This caused

potentially serious effects on food safety and

affected to both veterinary and human health

(Piddock et al., 2000).

In this study we determined the

antimicrobial resistance patterns of sixty-eight

C. jejuni isolated from chicken in Nakhon Pathom

province and evaluated the range of their minimum

inhibitory concentrations (MICs) using broth

microdilution method.


Bacterial strainsSixty-eight C. jejuni were isolated from

retail market-chicken in Nakhon Pathom province.

They were identified at Kamphaengsean Animal

Diagnostic Laboratory, Faculty of Veterinary

Medicine, Kasetsart University. These isolates

were confirmed to be C. jejuni by using PCR assay

following the method of Stucki et al. (1995). The

isolates were cultured on Columbia agar plate with

5% lysed horse blood and were incubated at 42∞Cfor 48 h in microaerophilic atmosphere. C. jejuni

colonies were transferred into 5 ml of Mueller-

Hinton broth and then incubated at 37∞C for 24 h

in microaerophilic atmosphere to produce a

suspension of 6 to 7 log CFU/ml.

Antimicrobial agentsSix groups of antimicrobial agents including

aminoglycoside (gentamycin, kanamycin and

streptomycin), macrolide (erythromycin),

penicillin (ampicillin), quinolone (ciprofloxacin,

marbofloxacin, nalidixic acid and norfloxacin),

sulfanamide (sulfamethoxazole) and tetracycline,

were used for this study.

Test procedureOne hundred microliters of two-fold

dilution of each antimicrobial agent ranging from

0.007 to 128 ml/ml was filled in each well of 96

wells sterile microtiter plates (Luber et al., 2003).

Each well for susceptibility testing was filled with

100 ml of 6 to 7 log CFU/ml of bacterial suspension,

and mixed gently. The plates were incubated at

37∞C under microaerophilic condition. The MICs

was evaluated at 24 hours later.


Minimum Inhibitory Concentrations

(MICs) were defined as the lowest concentration

that exhibits no growth of C. jejuni by visible

reading. The MICs of antimicrobial agents for C.

jejuni is presented in Table 1.

Each of isolates had different MICs in each

antimicrobial agents and different range of MICs.

The MICs range of gentamycin, kanamycin,

streptomycin, erythromycin, ampicillin,

ciprofloxacin, marbofloxacin, norfloxacin and

sulfamethoxazole were 0.03-2, 1-128, 0.12-16,

0.5-64, 0.25-64, 0.12-16, 0.007-8, 0.06-8 and

0.5-128 mg/ml, respectively. Nalidixic acid and

tetracycline had the same MICs range at 0.25-128

mg/ml. The MICs values of tetracycline, nalidixic

acid, kanamycin and sulfamethoxazole were higher

than other antimicrobial agents.

At present, there is no internationally

accepted criterion for susceptibility testing of

Campylobacter. Resistant breakpoint following

EUCAST (2000) method was used in this study. It

was classified as susceptibility or resistance

according to their individual MICs distribution of

each agent. When two or more sub-populations

were found, isolates with lower MICs were

classified as susceptibility, whereas those from

sub-populations with higher MICs were classified

as resistance.

The distribution of MICs of antimicrobial

agents against sixty-eight isolates of C. jejuni is

shown in Figure 1. The MICs of these antimicrobial

agents was a bimodal distribution. C. jejuni isolates

were categorized as resistance when MICs of

erythromycin, kanamycin, and sulfamethoxazole

exceeded 32 mg/ml; ampicillin and tetracycline

exceeded 16 mg/ml; nalidixic acid and streptomycin

exceeded 8 mg/ml; ciprofloxacin and

marbofloxacin exceeded 4 mg/ml, whereas C.

jejuni isolates were found as low resistance, when

MICs of gentamycin and norfloxacin exceeded 1

and 2 mg/ml, respectively.



Table 1 MICs of 11 antimicrobial agents against sixty-eight isolates of C. jejuni by broth microdilution

method.

MICs (mg/ml)

Agent1 128 64 32 16 8 4 2 1 0.5 0.25 0.12 0.06 0.03 0.015 0.007

GEN � � � � � � 2 9 � � 45 9 3

KAN 1 9 25 � � � 27 6�

STR � � � 8 21 � � � � 35 4

ERY � 3 6 � � 5 14 33 7

AMP � 4 8 20 � � � � 35 1

CIP � � � 8 11 21 � � � 23 5

MAR � � � � 3 14 � � � � � 7 33 10 1

NAL 1 9 7 22 8 4 9 4 2 2

NOR � � � � 1 9 37 � � � 13 8

SMX 12 35 2 � � � � 3 16

TET 8 7 32 6 � � � 1 10 4

1 GEN = gentamycin, KAN = kanamycin, STR = streptomycin, ERY = erythromycin, AMP = ampicillin, CIP = ciprofloxacin,

MAR = marbofloxacin, NAL = nalidixic acid, NOR = norfloxacin, SMX = sulfamethoxazole, TET = tetracycline2 MICs = Minimum Inhibitory Concentrations

Antimicrobial concentration ranges,

MIC50, MIC90, breakpoints of resistance and

percentage of resistance for C. jejuni are presented

in Table 2. MIC50 and MIC90 were MIC at which

50% and 90% of the isolates were inhibited,

respectively. The data indicated that

sulfamethoxazole had MIC50 and MIC90 higher

than other agents at 64 and 128 mg/ml, respectively,

Table 2 Antimicrobial agents, range of MICs, MIC50, MIC90, breakpoint of resistance and percentage

of resistance from sixty-eight C. jejuni isolates as determination by broth microdilution.

MIC (mg/ml) Break point of Percentage of

Agent Range MIC50 MIC90 resistance (mg/ml) resistance

Gentamycin 0.03-2 0.12 1 1 16.18%

Kanamycin 1-128 32 64 32 51.47%

Streptomycin 0.12-16 0.25 16 8 42.65%

Erythromycin 0.5-64 1 32 32 13.23%

Ampicillin 0.25-64 0.5 32 16 47.06%

Ciprofloxacin 0.12-16 4 8 4 58.82%

Marbofloxacin 0.007-8 0.03 4 4 25.00%

Nalidixic acid 0.25-128 16 64 8 69.12%

Norfloxacin 0.06-8 2 8 2 69.12%

Sulfamethoxazole 0.5-128 64 128 32 72.06%

Tetracycline 0.25-128 32 128 16 77.94%


while gentamycin had lowest values at 0.12 and 1

mg/ml, respectively.

Resistance to antimicrobial agents was

commonly detected among sixty eight C. jejuni

isolates. Resistance to four quinolones including

ciprofloxacin, marbofloxacin, nalidixic acid and

norfloxacin was found at 58.82, 25.00, 69.12 and

69.12%, respectively. The high prevalence of

quinolone resistant C. jejuni isolates was similar to

the finding of Engberg et al. (2001). However, the

prevalence of marbofloxacin resistant isolates of

C. jejuni in this study was relatively low (25.00%).

The reason was possibly due to marbofloxacin is

not commonly used as the other drugs in this

group. Endtz et al. (1991) reported that certain

isolates of C. jejuni were cross-resistant to various

Erythromycin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Kanamycin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

esSulfamethoxazole

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

esAmpicillin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Tetracycline

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Nalidixic acid

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Streptomycin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Ciprofloxacin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Marbofloxacin

0

20

40

60

80

100

0.01 0.06 0.5 4 32

MIC (ug/ml)

% is

olat

es

Gentamycin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Norfloxacin

0

20

40

60

80

100

0.03 0.25 2 16 128

MIC (ug/ml)

% is

olat

es

Figure 1 Distribution of MICs of 11 antimicrobial agents against sixty-eight isolates of C. jejuni from

chicken. Arrow indicated break point of resistance.


quinolones. Most nalidixic acid-resistant isolates

have been reported to be cross-resistant to one,

two or three drugs in this group (Reina et al.,

1995).

In Thailand, prevalence of quinolone

resistant Campylobacter species in broiler

farms was found to increase from 0% in 1987 to

84% in 1995 (Hoge et al., 1998). In addition,

Daniel et al. (2002) found that C. jejuni isolated

from Thai population was resistant to ciprofloxacin

in high proportion (77%). Quinolone-resistant

Campylobacter isolates in human and poultry

meat were also found in other countries, including

Canada (Gaunt and Piddock, 1996), Spain (Prats

et al., 2000), Senegal (Cardinale, 2003) and Taiwan

(Engberg et al., 2001). A fluoroquinolone-resistant

C. jejuni infection has also been associated with

foreign travel to many countries (Cardinale, 2003).

It was reported that most clinical isolates of C.

jejuni from U.S. troops in Thailand were resistant

to ciprofloxacin (Murphy et al., 1996).

Regarding the proportion of aminoglyco-

side resistance, the percentage of isolate resistance

in kanamycin was 51.47%, gentamycin was

16.18% and streptomycin was 42.65%. There was

a report previously indicated that aminoglycosides

resistance was less common in C. jejuni (Trieber

and Taylor, 2000). Campylobacter species isolated

from humans, pigs, cattle, and broilers were very

low in resistance to streptomycin (1%) (Fallon et

al., 2003). Moreover, it was demonstrated that

there was no Campylobacter from chicken isolates

that was resistant to aminoglycosides (Cabrita et

al., 1992). The contrary of aminoglycosides

resistance of C. jejuni from this work to the others

was possibly due to the administration of these

antimicrobial agents in veterinary medicine under

certain conditions in Thailand. Therefore,

resistances of these agents were found to increase

in certain isolates of C. jejuni.

In this study, the high proportions of C.

jejuni isolates were also resistant to other

antimicrobial agents including ampicillin

(47.06%), sulfamethozaxole (72.06%) and

tetracycline (77.94%). These results are in

accordance with the previous report that resistances

to ampicillin, chloramphenicol, sulfamethozaxole

and tetracycline of C. jejuni were commonly found

in Thailand (Daniel et al., 2002).

Low resistant level of erythromycin and

gentamycin was found in 9 and 11 isolates of C.

jejuni, respectively. However, MIC values of these

agents were different (Table 2). Since gentamycin

showed a very low MIC90, it seems reasonable to

consider this drug as an alternative for treatment of

C. jejuni infections under certain conditions in

Nakhon Pathom province, Thailand. Therefore, it

is necessary to select the antimicrobial wisely for

its effectiveness.

Resistance of C. jejuni isolates to a number

of antimicrobial agents is presented in Table 3.

Antimicrobial resistance (4 to 8 antimicrobial

agents) was detected in sixty-six isolates (97.06%).

Resistance to four or more of the drugs tested was

Table 3 Multi-antimicrobial resistant profiles of C. jejuni.

Number of antimicrobial agents Number of resistance isolates (%)

0 2 (2.94%)

4 10 (14.70%)

5 19 (27.94%)

6 26 (38.24%)

7 9 (13.24%)

8 2 (2.94%)


defined as multiple resistances. Only two isolates

of C. jejuni (2.94%) were susceptible to all

antimicrobial agents. All of multiple resistance of

C. jejuni was consisted of quinolone resistance.

One common resistance of this group was nalidixic

acid.

Most C. jejuni strains were resistant to

multi-antimicrobial agents tested. This was

possibly due to some development by genetic

change or physiological adaptation of the organisms

to increase antimicrobial tolerances in consequence

of previous exposure to antimicrobials. Therefore,

discontinuing the practice of routinely adding

growth promoters to animal feeds would reduce

the resistant strains of C. jejuni in animals in

Thailand. Long-term surveillance data are needed

to further evaluate the impact of any intervention

in antimicrobial usages. Moreover, antimicrobial

susceptibility testing methods for Campylobacter

species are needed to be harmonized and

standardized for predication that bacteria will

respond to treatment from an appropriate agent.

CONCLUSION

Quinolone–resistant Campylobactor is a

concerning issue in human medicine because of

failure in treatment of diarrhea cases with

quinolone. Scientists are monitoring the import

meat products to protect the consumers from this

organism. Most, C. jejuni of Thai–isolates have

shown high proportion of multiple resistance to

many kinds of antimicrobial agent used in the test

including quinolone. Erythromycin and

gentamycin are alternative drugs for treatment of

these cases. Good farming practice should be

implemented to reduce the usage of antimicrobials.

ACKNOWLEDGEMENTS

This work was supported by funding from

Center for Agricultural Biotechnology, Kasetsart

University, Kamphaengsean Campus. Gratitude

is expressed to Srisamai Viriyarampa for her

advices in microbiological techniques. We also

acknowledge the Faculty of Veterinary Medicine,

Kasetsart University for equipment supporting.

LITERATURE CITED

Aquino, M.H.C., A.L.L. Filgueiras, M.C.S. Ferreira

and S.S. Oliveinia. 2002. Antimicrobial

resistance and plasmid profiles of C. jejuni

and C. coli from human and animal sources.

Lett. Appl. Microbiol. 34: 149-153.

Cabrita, J.J., J. Rodrigues, F. Bragance, C. Morgado

and A. Goncalves. 1992. Prevalence biotypes,

plasmid profiles and antimicrobial resistance

of Campylobacter isolated from wild and

domestic animals from Northeast Portugal. J.Appl. Bacteriol. 73: 279-285.

Cardinale, E.J., A. Dromigny, F. Tall, M. Ndiaye,

M. Konte and J.D. Perrier. 2003.

Fluoroquinolone susceptibility of

Campylobacter strains, Senegal. Antimicrob.Agents Chemother. 58: 178-190.

Daniel, W.I., C.W. Hoge, P. Chittima, L. Bodhidatta

and K.W. Hickey. 2002. Competitive

antibiotic resistance of diarrheal pathogens

from Vietnam and Thailand, 1996-99. Emerg.Infect. Dis. 8: 175-201.

Endtz, H.P., G.J. Rujis, B. Klingeren, W. H. Jansen

and. R.P.Mouton. 1991. Quinolone resistance

in Campylobacter isolated from man and

poultry following the introduction of

fluoroquinolone in veterinary medicine. J.Antimicrob. Chemother. 27: 199-208.

Engberg, J., F.M. Aurestrup, D.E. Taylor and I.

Nachamkin. 2001. Quinolone and macrolide

resistance in C. jejuni and C. coli resistance

mechanisms and trend in human isolates.

Emerg. Infect. Dis. 7: 24-34.

EUCAST. 2000. Terminology relating to methods

for the determination of susceptibility of

bacteria to antimicrobial agents. Clin.Microbiol. Infect. 6: 503-508.


Fallon, R., N. Sullivan, M. Maher and C. Carroll.

2003. Antimicrobial resistance of C. jejuni

and C. coli isolates from broiler chickens

isolated at an Irish poultry processing plant.

Lett. Appl. Microbiol. 36: 277-281.

Gaunt, P.N. and L.J.V. Piddock. 1996.

Ciprofloxacin resistant Campylobacter spp.

In humans: an epidemiological and

laboratory study. J. Antimicrob. Chemother.37: 747-757.

Hoge, C.W., J.M. Gambel, C. Pitarangsri and P.

Echeverria. 1998. Trends in antibiotic

resistance diarrhea pathogens isolated in

Thailand over 15 years. Clin. Infect. Dis. 26:

341-345.

Luber, P., E. Bartelt, E. Genschow, J. Wagner and

H. Hahn. 2003. Comparison of broth

microdilution, E Test, and agar dilution

methods for antibiotic susceptibility testing

of Campylobacter jejuni and Campylobacter

coli. J. Clin. Microbiol. 41: 1062-8.

Murphy, G., P. Echeverria, L. Jackon, M. Arness

and C. Leboon. 1996. Ciprofloxacin and

azithromycin resistant Campylobacter causing

travelers’ s diarrhea in US troops deployed to

Thai in 1994. Clin. Infect. Dis. 22: 864-869.

Piddock, L.J.V., V. Ricci, K. Stanley and K.Jones.

2000. Activity of antibiotics used in human

medicine for C. jejuni isolated from farm

animals and their environment in Lancashire,

U.K. J. Antimicrob. Chemother. 46: 303-

306.

Prats, G., B. Mirelis, T. Lovet, C. Munoz, E. Miro

and F. Navarro. 2000. Antibiotic resistance

trends in enteropathogenic bacteria isolated

in 1985-1987 and 1995-1998 in Barcelona.

Antimicrob. Agents Chemother. 44: 1140-

1145.

Rautelin, H., A. Vierikko, M.L. Hanninen and M.

Vaara. 2003. Antimicrobial susceptibility of

Campylobacter strains isolated from Finnish

subjects infected domestically or from those

infected abroad. Antimicrob. AgentsChemother. 47: 102-105.

Reina, J., M.J. Ros and V. Fernandez-Baca. 1995.

Resistance to erythromycin in

fluoroquinolone-resistant Campylobacter

jejuni strains isolated from human feces. J.Antimicrob. Chemother. 35: 351-352.

Saenz, Y., M. Zarazaga, M. Lantero and F. Baquero.

2000. Antibiotic resistance in Campylobacter

strains isolated from animals, foods, and

human in Spain in 1997-1998. Antimicrob.Agents Chemother. 44: 267-271.

Stucki, U., J. Frey, J. Nicolet and A.P. Burnens.

1995. Identification of Campylobacter jejuni

on the basis of a species-specific gene that

encodes a membrane protein. J. Clin.Microbiol. 33: 855–859.

Trieber, C.A. and D.E.Taylor. 2000. Mechanisms

of antibiotic resistance in Campylobacter.

Am. Soc. Microbiol. 54: 441-454.


Hematology, Cytochemistry and Ultrastructure of Blood Cellsin Asiatic Black Bear (Ursus thibetanus)

Chaleow Salakij1, Jarernsak Salakij1, Nual-Anong Narkkong2,Ludda Trongwonsa3 and Rattapan Pattanarangsan4

ABSTRACT

Blood cells from adult Asiatic black bear (Ursus thibetanus) were examined and measured after

stainning with modified Wright stain and cytochemical stains, including Sudan Black B (SBB), Periodic

acid Schiff’s reaction (PAS), a-naphthyl acetate esterase (ANAE), acid phospatase (AcP) and b-

glucuronidase (b-glu). Red blood cells were uniform in shape, with 7.3 mm mean diameter in size and easy

to form rouleaux. Using scanning electron microscopic (SEM) examinations revealed normal and

abnormal red blood cell surfaces. Neutrophils contained several vacuoles as detected by light microscope

and revealed themselves as large granules under transmission electron microscope. Neutrophils stained

strongly positive with SBB, ANAE; weak positive with PAS and negative with AcP and b-glu. Using

SEM, neutrophil surfaces revealed several microvilli and some micropores. Eosinophils contained

numerous small round red refractive granules with some vacuoles. Eosinophils stained strongly positive

with SBB and ANAE but negative with PAS and b-glu. Basophils had variable numbers of intense

basophilic granules that obscured the very long lobulated nucleus. Basophils stained moderately positive

with SBB but strongly positive with ANAE. Lymphocytes were negative with SBB but have 3 patterns

of reactivity with ANAE, AcP and b-glu, including negative, focal dot and fine granular stainings.

Monocytes stained moderately with SBB and moderately to strongly with for ANAE and b-glu. The SEM

examinations could differentiate white blood cells by their surface contours. Transmission electron

microscopic examinations revealed organelles within all blood cells.

Key words: Asiatic black bear, blood cell, cytochemistry, morphology, ultrastucture

INTRODUCTION

Asiatic black bear (Ursus thibetanus) has

shaggy black fur with white crescent on the chest,

considerably larger than Malayan sun bear (Ursus

malayanus). Asiatic black bear has suffered from

habitat loss and is now rare in many areas (Francis,

2001). So this endangered species has been studied

1 Faculty of Veterinary Medicine, Kasetsart University, Kamphaengsaen, Nakorn Pathom 73140, Thailand.2 Central Instruments Center, Faculty of Science, Mahasarakarm University, Mahasarskarm 44150, Thailand.3 National Institute of Animal Health, Kasetklang, Jatujuk, Bangkok 10900, Thailand.4 Faculty of Veterinary Science, Mahidol University, Salaya, Nakorn Pathom 73170, Thailand.


intensively to determine the health status of the

individuals. Veterinary hematology serves as a

screening procedure to assess general health, the

body’s ability to fight infection in adjunct to patient

evaluation or diagnosis (Jain, 1993).

Differential white blood cell count is very

useful not only in numbering the white blood cells

but also provide evidence of anemic condition or


reveal the pathogenesis. Blood smear examinations

provide more information on morphology of red

blood cell, white blood cell and platelets (Mills,

1998). Cytochemical method is useful in diagnosis

of acute leukemia in human (Apibal, 1987;

Khemtonglang et al., 1997). The purpose of the

present study was to characterize the morphology,

cytochemical reaction and ultrastrucrure of blood

cells in Asiatic black bear.


From February to June 2003, five clinically

healthy Asiatic black bears in Khao Kheaw Open

Zoo were chemical restrained with xylaxine

(Rompun®) and ketamine. Two millilitres of blood

samples were collected from the jugular vein and

transferred to tubes containing ethylenediamine

tetraacetic acid (EDTA). Some of the blood without

anticoagulant was directly smeared on the slides.

Two Asiatic black bears were adult males and

three were adult females aging between 3-5 years

old. Hematology, plasma protein and fibrinogen

were determined by manual technique (Schalm et

al., 1975) within two hours after blood collection.

Two direct blood smears from each bear were

stained with a modified Wright and Wright’s stains.

A minimum of 200 leukocytes were counted for

differential leukocyte determinations. The

anticoagulated blood was used for reticulocyte

count by staining with new methylene blue stain

(Schalm et al., 1975). The percentage of

reticulocyte presented in 1,000 red blood cells

(RBC) was determined. For each hematologic

parameter, means, variances and standard errors

were calculated using SPSS® for Window™

(Norusis, 1993).

Cytochemical staining characteristics of

blood cells were evaluated using air-dried blood

smears from three Asiatic black bears. Cells were

stained with periodic acid Schiff’s reaction (PAS),

Sudan black B (SBB), a–naphthyl acetate

esterase (ANAE), acid phosphatase (AcP) and b–

glucuronidase (b–glu). Cytochemical procedures

used were the same as those previously described

(Salakij et al., 2002). Positive– and negative-

stained cells were differentiated by counting 500

cells on each of the cytochemically stained smears.

For scanning electron microscopy (SEM)

and transmission electron microscopy (TEM),

blood cells from three Asiatic black bears were

processed as described by Salakij et al. (2002).

Identification of blood cells by SEM and TEM was

based on the relative number, size, shape and

distribution of granules and on nuclear appearance.

RESULTS

There was no blood parasite detected in all

bears. Hematological data of Asiatic black bear

was tabulated (Table 1). White blood cell

differential counts were shown in Table 2. Blood

cell diameters were observed and calculated (Table

3). Cytochemical staining patterns of blood cells

were summarized (Table 4). The morphology under

light microscope, SEM, TEM and cytochemical

characteristics of individual blood cells were

evaluated, as described below.

ErythrocytesRed blood cells (RBCs) or erythrocytes

under light microscope showed uniform in shape

(Figure 1), slightly biconcave and slightly central

pallor as observed by SEM (Figure 4a, 4b) with 7.3

mm mean diameter in size (Table 3) and easy to

forming rouleaux. Some defected RBCs (Figure

4c), crenated RBCs (Figure 4d), echinocyte (Figure

5a, 5b), rubricyte (Figure 8b) and metarubricyte

(Figure 8c) were also observed. Mature red blood

cells were negative for all cytochemical stainings.

Ultrastructurally, mature RBC showed only

hemoglobin (Figure 8a) while the metarubricytes

showed some organelles (Figure 8c) and the

rubricytes showed more mitochondria (Figure 8b).


PlateletsBear platelets were approximately 1/5 to 1/

2 of RBC and had prominent reddish-purple

granules which were easily seen in modified Wright

stain (Figure 1d, f). Plateletes were not stained

with SBB but were moderately to strongly positive

with ANAE (Figure 3m). These platelets were

seldom seen on RBCs (Figure 5b), but gave rosette

formation on monocyte (Figure 5c) and aggregation

(Figure 5d). Ultrastructurally, platelets showed

dense granules, alpha-granules, glycogen granules

and microtubule (Figure 8d).

NeutrophilsNeutrophils were the most prevalent

leukocyte in Asiatic black bear (Table 2) with

neutrophil : lymphocyte ratio equal to 62 : 26.

With modified Wright stain, neutrophils showed

the same size as basophil (Table 3). Neutrophils

showed faintly stained cytoplasm which contained

indistinct pale granules and several vacuoles

Table 1 Hematology of Asiatic black bear.

Hematology Asiatic black bear (n = 5)

PCV (%) 41.4 ± 1.8

Hemoglobin (g/dL) 13.3 ± 0.6

RBC (1012/L) 6.08 ± 0.39

MCV (fL) 68.5 ± 2.1

MCHC (g/dL) 32.1 ± 0.6

WBC (1011/L) 6.99 ± 0.59

Band neutrophils (109/L) 0.11 ± 0.09

Segmented neutrophils (109/L) 4.15 ± 0.19

Lymphocytes (109/L) 1.84 ± 0.28

Monocytes (109/L) 0.13 ± 0.04

Eosinophils (109/L) 0.73 ± 0.18

Basophils (109/L) 0.03 ± 0.02

Band neutrophils (%) 1.7 ± 1.4

Segmented neutrophils (%) 60.3 ± 3.4

Lymphocytes (%) 25.8 ± 1.7

Monocytes (%) 1.8 ± 0.5

Eosinophils (%) 10.0 ± 1.5

Basophils (%) 0.4 ± 0.2

Plasma protein (g/dL) 8.1 ± 0.3

Fibrinogen (mg/dL) 220 ± 49

Reticulocyte (%) 0.0 ± 0.0

Table 2 White blood cell differential count in

Asiatic black bear.

Hematology Asiatic black bear

(n = 5)

Band neutrophils (%) 1.7 ± 1.4

Segmented neutrophils (%) 60.3 ± 3.4

Lymphocytes (%) 25.8 ± 1.7

Monocytes (%) 1.8 ± 0.5

Eosinophils (%) 10.0 ± 1.5

Basophils (%) 0.4 ± 0.2


(Figure 1a, 2a). Neutrophils have tight constricted

and multilobulated nuclei (Figure 1a, 2a). These

vacuoles shown in light microscope were revealed

as large granules in TEM (Figure 9a-d). Some

neutrophils (2-5%) of the female revealed sex

chromatin lobe.

Neutrophils stained strongly positive with

SBB (Figure 3a), moderately stained with ANAE

(Figure 3i) and negative with b-glu (Figure 3o).

Using SEM, neutrophil surfaces revealed several

microvilli and some micropores (Figure 6a, 6b,

6d). Ultrastructurally, neutrophils showed lobed

nuclei, small specific granules (Figure 11a, b),

large electron-dense granules (Figure 9a-c) and

glycogen granules (Figure 9d) were also detected.

EosinophilsEosinophils varied from 10 to 16 mm

(average 14 mm) in diameter (Table 3). Eosinophils

contained numerous small round red refractive

granules with some vacuoles (Figure 1b, 2b).

Eosinopil nuclei were less lobulated than those of

neutrophils, and tetralobed, trilobed or band-

shaped. Eosinophils stained moderately positive

with SBB (Figure 3b) and ANAE (Figure 3j) but

weak positive with b-glu (Table 4). Under SEM,

eosinophil surfaces revealed larger granule contour

(Figure 6c, 6d) than those of basophils (Figure 6e,

6f). Ultrastructurally, eosinophils showed lobed

nuclei, large pleomorphic granules with bar-shape

structures in some granules, Golgi apparatus, RER

and ribosomes (Figure 10a, 10b).

Table 4 Cytochemical staining patterns of blood cells from Asiatic black bear.

Cell type SBB PAS ANAE Acid phosphatase b-glucuronidase

Neutrophils + + + ++ - ±±

Eosinophils + - +++ - +

Basophils + + - ++ - -

Lymphocytes - - - / focal dot / fine granular - / focal dot / fine granular - / focal dot / fine granular

Monocytes + - + + - +

Platelets - - ++ - -

SBB indicates sudan black B; PAS, periodic acid-Schiff; and ANAE, a-naphthyl acetate esterase. Staining was scored as negative

(-), weak (±±, few positive cells), moderate (+), moderate to strong (++), or strong (+++).

Table 3 Mean ± SD of blood cell diameters (mm) in Asiatic black bear.

Cell type No. Asiatic black bear

Red blood cells 50 7.30 ± 0.84

Segmented Neutrophils 50 13.42 ± 1.10

Eosinophils 50 14.09 ± 1.66

Basophils 18 13.06 ± 1.95

Lymphocytes

small 50 8.28 ± 0.78

medium 50 11.02 ± 1.11

large 17 14.40 ± 0.48

Monocytes 50 15.28 ± 2.14


Figure 1 Light micrographs of blood cells in Asiatic black bear stained with modified Wright stain. a.

A segmented neutrophil with many cytoplasmic vacuoles. b. An eosinophil. c. A basophil (left)

and a segmented neutrophil. d. A lymphocyte (right) and a segmented neutrophil. e. A

monocyte. f. A basophil (right) and a segmented neutrophil.

Figure 2 Light micrographs of blood cells in Asiatic black bear stained with Wright’s stain (a-d). a. A

segmented neutrophil with some cytoplasmic vacuoles. b. An eosinophil. c. A basophil. d. A

lymphocyte (right) and a basophil. e. A leukocyte stained with new methylene blue whilst there

was no reticulocyte.


Figure 3 Cytochemical staining of blood cells in Asiatic black bear a. Sudan black-B (SBB) positive

segmented neutrophil. b. SBB positive in the periphery of the granules of the eosinophil. c.

SBB positive in basophil. d. Some small dots of SBB positive in a 14 mm monocyte. e. PAS

positive in segmented neutrophil. f. PAS positive in two segmented neutrophils and negative

in an eosinophil. g. PAS negative in a basophil comparing with PAS positive in two segmented

neutrophils. h. PAS negative in lymphocyte (lower right) and monocyte (upper left). i.-m.

ANAE reactivity in a segmented neutrophil (i), eosinophil (j), basophil (right) (k), monocyte

(upper right) (l) and in platelets (m). n. Negative AcP in a segmented neutrophil and a

monocyte. o. Negative b-glu in a segmented neutrophil and a basophil. p. Focal dot positive

of b-glu in a lymphocyte.


Figure 4 Scanning electron micrographs (SEM) of red blood cells (RBCs) in Asiatic black bear. a. A

cluster of RBCs showing bicocave disk and crenations. b. Higer magnification of crenated

RBCs. c. Defective RBCs. d. A cluster of four echinocytes.

Figure 5 SEM of RBCs and platelets in Asiatic black bear. a. An echinocyte. b. A platelet on an

echinocyte. c. Plateletes rosetting on a monocyte. d. Platelets aggregation next to neutrophil.


Figure 6 SEM of granulocytels in Asiatic black bear. a. A neutrophil showing short microvilli and some

micropores. b. A neutrophil with three micropores. c. An eosinophil. d. An eosinophil (right)

and a monocyte (left). e. A basophil showing small granule contour. f. A basophil.

Figure 7 SEM of agranulocytes in Asiatic black bear. a. A lymphocyte with several cytoplasmic blebs.

b. A monocyte with deep surface fissures.


Figure 8 Transmission electron micrographs (TEM) of blood cells in Asiatic black bear. a. Mature

erythrocytes. b. A rubricyte with some mitochondria (arrows). c. A metarubricyte. d. Platelets

containing mitochondria, vacuoles (v) and dense granules (*).


Figure 9 TEM of neutrophils in Asiatic black bear. a. A segmented neutrophil showing nucleus (N)

many fine granules and large granules (L). b. A segmented neutrophil showing two lobed

neucleus (N) with heterogeneous granule density of large granules (L). c. A segmented

neutrophil (N) with homogeneous granule density of large granules (L). d. A segmented

neutrophil (N) with many glycogen granules (arrows).


Figure 10 TEM of eosinophils and basophils in Asiatic black bear. a. An eosinophil showing neucleus

(N) and many granules. b. Higher magnification of eosinophil in (a) showing bar-shape

structures (arrows) in their granules. c., d. Basophils showing bilobed neuclei (N) and many

small heterogenous granules.


Figure 11 TEM of lymphocytes and monocytes in Asiatic black bear. a. A lymphocyte showing round

neucleus (N) and three mitochondria (arrows). b. A lymphocyte with round nucleus (N). c.

A monocyte with kidney-shaped nucleus (N). d. A monocyte with nucleus (N), many

mitochondria (MT), ribosomes, vacuoles (v) and pseudopodia (arrows).


BasophilsBasophils in were not frequently observed.

They varied from 11 to 15 mm (average 13 mm).

Basophils contained variable numbers of intensely

basophilic granules (Figure 1c, 1f, 2c, 2d) which

obscured the very long lobe nucleus. Basophil

granules stained with SBB more faintly than

granules of neutrophil (Figure 3c). Basophil

granules appeared red-brown when stained with

ANAE (Figure 3k) but weak positive or negative

with b-glu (Figure 3o). Under SEM, basophil

surfaces revealed smaller granule contour (Figure

6e, 6f) than those of eosinophils (Figure 6c, 6d).

Ultrastructurally, basophils showed lobed nuclei,

small heterogenous electron density granules, some

dense granules (Figure 10c) and some mitochondria

(Figure 10d).

LymphocytesLymphocytes in bears were variable in size

(7 to 16 mm diameter). Most lymphocytes were

small (Figure 1d, 2d) and medium size. Some

lymphocytes contained small azurophilic granules

in their cytoplasm. Lymphocytes were negative

for SBB but had 3 patterns of reactivity for ANAE

and b-glu, including negative, focal dot and fine

granular stainings. Under SEM, lymphocyte

surfaces revealed smooth bulging contour of round

nuclei (Figure 7a) with variable numbers of cell

membrane blebs. Ultrastructurally, lymphocytes

showed round nuclei with peripheral clumps of

heterochromatin and some mitochodria (Figure

11a, 11b).

MonocytesMonocytes in bears varied from 13 to 17

mm in diameter (Table 4). They are the largest

white blood cells with variable shape. The nuclei

were extremely variable but usually have lacy

chromatin (Figure 1e). The cytoplasm was blue-

gray and contained variable size of vacuoles (Figure

1e). Monocytes stained moderately positive with

SBB with faintly black, small granules scattered in

the cytoplasm (Figure 3d). They were moderately

to strongly positive with ANAE showing red-

brown fine granular pattern (Figure 3l), but they

were negative for PAS (Figure 3h) and b-glu.

Under SEM, monocyte surfaces revealed more

smooth membrane than those of neutrophil (Figure

7b) with deep fissure and micropores.

Ultrastructurally, monocytes showed variable

shape of nuclei with several mitochodria and

pseudopodia (Figure 11c, 11d).

DISCUSSION

In this report, we described the light

microscope, cytochemical features and

ultrastructure of blood cells in Asiatic black bear.

Although the erythrocytes in bears has special

features such as uniform in shape with central

pallor similar to those of dog, they are larger in

mean diameter (Jain, 1993) and easy to form

rouleaux. The RBC parameters of Asiatic black

bear in this study was similar to but had lower

number of leukocytes than those of Asiatic black

bear in the zoo of Czechoslovakia (Pospisil et al.,

1987). There was no blood parasite found in Asiatic

black bear of this study while there was a report of

filarid worm (Dirofilaria ursi) found in the

esophageal and tracheal connective tissue of the

male Asiatic black bear on Kyushu island

(Yokohata et al., 1990).

Ultrastructure of platelets in Asiatic black

bear were similar to those in bovine (Fern, 2000)

and African elephant (Du Plessis and Stevens,

2002). Platelets in Asiatic black bear were negative

for SBB which were the same as those in bovine,

cat, dog, horse and green sea turtle (Raskin and

Valenciano, 2000). Platelets in Asiatic black bear

positively stained with ANAE which were similar

to those of dog that were also positively stained

with non-specific esterase (Raskin and Valenciano,

2000). And they were similar to platelets in Asian

wild dog that were positive with ANAE and b-glu

(Salakij et al., 2000). So the ANAE stain would be


useful to differentiate megakaryocytic leukemia

in the bears when using with the other non-specific

esterase stains like those in human (Apibal, 1987).

The neutrophils in Asiatic black bear

contained large granules that were unstained similar

to vacuoles under light microscopy that were

characteristic of neutrophils in Asiatic black bear.

These vacuoles were large and variable in electron

density as shown by TEM. This characteristic was

not found in goat (Kramer, 2000) or in reindeer

(Henkel et al., 1999).

Characteristic features of basophils in

Asiatic black bear were similar to those in dog;

including having a long polymorphonuclear-

shaped nucleus that is longer than those of

neutrophil and thinner than those of most monocyte

(Willard et al., 1994). Basophils of Asiatic black

bear are the same size as neutrophils but smaller

than eosinophils (Table 2). These findings may be

useful in differentiation of degranulated basophils

from netrophils in Asiatic black bear.

The strongest reactivity of ANAE was found

in eosinophils which was different from eosinophils

in Asian wild dogs that were negative for ANAE

(Salakij et al., 2000). The bar-shape structures

found in some granules of the eosinophil in Asiatic

black bear by TEM were different from eosinophils

of dog, cat (Young, 2000), goat (Kramer, 2000)

and reindeer (Henkel et al., 1999).

In the Asiatic black bear, granulocytes and

monocytes stained with SBB. These findings are

useful in differentiating acute myelogenous

leukemia from acute lymphoblastic leukemia like

those in human (Apibal, 1987). ANAE, AcP and

b-glu staining characteristics of lymphocytes were

similar to those reported in human that can

differentiate T-lymphocytes (dot staining) from

non T-lymphocytes (negative or fine granular

staining) (Apibal, 1987; Khemtonglang et al.,

1997).

The basophils in Asiatic black bear were

strongly positive with SBB and ANAE which is

similar to reindeer basophils (Henkel et al., 1999).

The basophils in Asian wild dog were stained

strongly positive with SBB, ANAE and b-glu

(Salakij et al., 2000). So in the Asiatic black bear,

the SBB and ANAE are useful in the diagnosis

between basophilic leukemia and megakaryocytic

leukemia.

The ultrastructure of basophils in Asiatic

black bear showed smaller granules than those in

goat (Kramer, 2000) and in reindeer (Henkel et al.,

1999). The ultrastructure of lymphocytes and

monocytes in Asiatic black bear were not different

from those in pig (Steffens III, 2000) and in

reindeer (Henkel et al., 1999). Under SEM, the

eosinophils revealed large granule contour which

were easy to differentiate from basophils and

neutrophils. This is the first report on the surfaces

of blood cells in Asiatic black bear that could

demonstrate abnormal shape of red blood cells and

differentiation of white blood cells.

CONCLUSIONS

The neutrophils in Asiatic black bear

contained large granules that were unstained and

similar to vacuoles under light microscopy. This is

characteristic of neutrophils in Asiatic black bear.

The basophils in Asiatic black bear contained

many small granules that not stained metachromatic

with modified Wright stain so their nuclear outline

were clearly defied. The results of this study

provide more information on the morphology,

cytochemical staining and ultrastructural

characteristics of blood cells in Asiatic black bear.

This information adds to our understanding of

blood cells in healthy Asiatic black bears.

ACKNOWLEDGEMENTS

This research was supported in part by the

Kasetsart University Research and Development

Institute.


LlTERATURE CITED

Apibal, S. 1987. Laboratory diagnosis of acute

leukemia. J. Med. Tech. Assoc. Thailand15: 69-75.

du Plessis, L. and K. Stevens. 2002. Blood platelets

of the African elephant. J. Comp. Patho.127: 208-210.

Francis, C. M. 2001. A Photographic Guide toMammals of Thailand & South-East Asia.ASIABOOKS. Bangkok. 127 p.

Khemtonglang, N., C. Kitpetcharatana, B.

Petsuriya, K. Rukseree and A. Sriinsut. 1997.

Pattern of b-glucuronidase staining in acute

leukemias. J. Med. Tech. Assoc. Thailand25:66-78.

Kramer, J. W. 2000. Normal hematology of cattle,

sheep and goat, pp. 1075-1084. In B. F.

Feldman, J. G. Zinkl and N. C. Jain (eds.).

Schalm’s Veterinary Hematology 5thed.

Lippincott Williams and Wilkins,

Philadelphia.

Henkel, K. A., C. L. Swenson, B. Richardson and

R. Common. 1999. Morphology, cytochemical

staining and ultrastructure characteristics of

reindeer (Rangifer tarandus) leukocytes. Vet.Clin. Path. 28: 8-15.

Jain, N.C. 1993. Essentials of VeterinaryHematology. Lea and Febiger. Philadelphia,

417 p.

Mills, J. 1998. Interpreting blood smears (or What

blood smears are trying to tell you !). Aust.Vet. J. 76 : 596-600.

Norusis, M. J. 1993. SPSS“““““ for Window‘‘‘‘‘ BaseSystem User’s Guide Release 6.0. SPSS

Inc., Chicago, Illinois. 828 p.

Pospisil, J., F. Kaset and J. Vahara. 1987. Basic

haematological values in carnivores-I. The

Canidae, the Hyaenidae and the Ursidae.

Comp. Bioche. Physio. 86: 649-652.

Raskin, R. E. and A Valemciano. 2000.

Cytochemistry of normal leukocytes. pp. 327-

346. In B. F. Feldman, J. G. Zinkl and N. C.

Jain (eds.). Schalm’s VeterinaryHematology. 5thed. Lippincott Williams and

Wilkins, Philadelphia.

Salakij, C., J. Salakij, J. Rattanakunuprakarn, N.

Tengchaisri, W. Tunwattana and S. Apibal.

2000. Morphology and cytochemistry of blood

cells from Asian wild dog (Cuon alpinus).

Kasetsat J. (Nat. Sci.). 34:518-525.

Salakij, C., J. Salakij, S. Apibal, N. Narkkong, L.

Chanhome and N. Rochanapat. 2002.

Hematology, morphology, cytochemical

staining, and ultrastructural characteristics of

blood cells in King cobras (Ophiophagus

hannah). Vet. Clin. Path. 31 : 116-126.

Schalm, O. W., N. C. Jain and D. J. Carrol. 1975.

Schalm’s Veterinary Hematology. 3rded.

Lea and Febiger. Philadelphia. 807 p.

Steffens III WL. 2000. Ultrastructural features of

leukocytes, pp. 326-336. In B. F. Feldman, J.

G. Zinkl and N. C. Jain (eds.). Schalm’sVeterinary Hematology. 5thed. Lippincott

Williams and Wilkins, Philadelphia.

Willard, M. D., H. Tvedten and G. H. Turnwald.

1994. Small Animal Clinical Diagnosis byLaboratory Methods. 2nded. W. B. Saunders

Company. Philadelphia. 377 p.

Yokohata, Y., O., Fujita, M. Kamiya, T. Fujita, K.

Kaneko and M. Ohbayashi. 1990. Parasites

from the Asiatic black bear (Ursus thibetanus)

on Kyushu island, Japan. Journal of WildlifeDiseases. 26: 137-138.


Probiotic Properties of Bacillus pumilus, Bacillus sphaericusand Bacillus subtilis in Black Tiger Shrimp

(Penaeus monodon Fabricius) Culture

Watchariya Purivirojkul1, Monchan Maketon1 and Nontawith Areechon2

ABSTRACT

Three species of the genus Bacillus namely B. pumilus,B. sphaericus, and B. subtilis were isolated

from the intestine of Penaeus monodon and tested for possible potential as a probiotic in black tiger shrimp

rearing. The competition, colonization and inhibition activities of Bacillus spp. on Vibrio harveyi, a

known pathogen in black tiger shrimp aquaculture were conducted. Transmission electron microscope

observations showed the size of V. harveyi’s cell colonized by B. pumilus, B. sphaericus and B. subtilis

to be smaller compare with its normal cell. This morphological deviation was permanently changed in

every generation of V. harveyi. In addition, three species of the Bacillus could be grown in a wild range

of conditions, these include salinity between 0-8 % NaCl, pH from 4 to 11 and temperature ranging of

25-50 ∞C except B. sphaericus which was not amenable to grow at 50 ∞C. Therefore, B. pumilus, B.

sphaericus and B. subtilis showed promising potential to be used as a probiotic in black tiger shrimp.

As for probiotic properties in black tiger shrimp, the results shown that number of bacteria in

intestinal tracts of the shrimps were increased in all treatments from 216 to 803% of the control group,

and the amount of V. harveyi, were found reduced from 87.53 to 99.76% of the control, as also confirmed

by scanning electron microscope observations. B. subtilis, the mixture of B. sphaericus + B. subtilis, and

B. pumilus + B. sphaericus + B. subtilis in culture media showed immunostimulatory features measured

by total hemocytes, phenol oxidase, superoxide anion, clearance ability and bactericidal activity which

increased by 20.02-23.10, 26.02-39.43, 53.45-66.04, 44.68-59.57 and 50.00%, respectively.

Key words: Bacillus pumilus, Bacillus sphaericus, Bacillus subtilis, probiotic, Penaeus monodon

INTRODUCTION

Black tiger shrimp (Penaeus monodon

Fabricius) has been one of the important export

products of Thailand for more than a decade.

However, in 2002, Thai shrimp production fell

about 40 percent from 2001 to approximately

160,000 tons due to several diseases outbreak at

the beginning of the year. Farmers utilized a large

1 Department of Zoology, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand.2 Department of Aquaculture, Faculty of Fisheries, Kasetsart University, Bangkok 10900, Thailand.


quantities of antibiotics in trying to solve this

problem. Some antibiotic such as chloramphenicol

was found as a residue in shrimps exported from

China, Viet Nam, Indonesia and Thailand.

Recently, the European communities are focusing

in antibiotic residues in shrimp products by strictly

inspection for all shrimp products imported from

Asian countries.

The use of probiotic bacteria and some

immunostimulant substances such as glucan and

peptidoglycan have become popular methods

developed for fighting againt diseases since the

past decade (Fuller, 1992). Many genus of bacteria

were used as probiotic such as Vibrio (Gullian et

al., 2004) Bacillus spp. (Moriarty, 1998; Rengpipat

et al., 2000; Gullian et al., 2004), especially those

bacteria isolated from the intestine of Penaeus

monodon (Rengpipat et al., 2000). There are several

mechanisms of probiotic, these include production

of inhibitory compounds, competition for

chemicals or available energy, competition for

adhesion sites, enhancement of the immune

response and improvement of water quality

(Verschuere et al., 2000)

In our studies, we conducted both in vitro

and in vivo experiments by focusing on competitive

and inhibitive capabilities of bacteria in the genus

Bacillus in term of probiotic against Vibrio harveyi.

Moreover, we continued study these bacteria on

their effectiveness as probiotic properties in P.

monodon in terms of growth and immune response

indicated by total hemocytes, phenoloxidase

activity, superoxide anion, bactericidal activity

and clearance ability. Changing of bacterial

community in shrimp intestines when feed with

Bacillus spp. were also investigated by counting

the number of Bacillus spp. and Vibrio spp.,

scanning electron microscope (SEM) was used for

confirmation.


This research was performed in both in

vitro and in vivo systems.

Isolation and identification of Bacillus spp.Bacillus spp. were isolated from the

intestine of Penaeus monodon harvested from

shrimp farms in Chachoengsao province. Two

hundred samples of shrimp were investigated.

Intestine were dissolved in 5 ml of 1.5% NaCl per

animal and heat shock on water bath at 80 ∞C for

20 min followed by cold shock with normal tap

water. Then the solution was spreaded on plates

using spread plate technique on Nutrient agar

(NA) supplemented with 1.5% NaCl (w/v). These

plates were incubated at 35 ∞C for 24 h. Isolates

were purified by streaking on NA supplemented

with 1.5% NaCl (w/v). Catalase test were used for

identifying Bacillus species. Species identification

were done by VITEK 32 Bacillus (Biomérieux).

1. Colonization and inhibition activities of threebacteria on Vibrio harveyi in vitro

V. harveyi was isolated from black tiger

shrimp and streaked on TCBS (Thiosulfate Citrate

Bile Sucrose) agar. Bacillus spp. were cultured on

nutrient agar supplemented with 1.5% NaCl (w/

v), and both were incubated at 35 ∞C for 24 h.

Colonization activities tests were done on NA

supplemented with 1.5% NaCl (w/v) by cross

streak method. V. harveyi was streaked in the first

line and then Bacillus spp. was streaked

perpendicular to it. Each type of bacterium streaking

was done in triplicate and they were incubated at

room temperature for 24 h.

1.1 Morphological change of V. harveyiafter colonization

V. harveyi was isolated from the

colonization area, especially from the cross

streaking point as well as from the control. All

samples were cultured on TCBS agar and incubated

at 35 ∞C for 24 h. Growth rates of the experimental

V. harveyi were compared with the normal one.

Single colony was used to determine

morphological deviation by transmission electron

microscope (TEM)

1.2 Investigation for the possibility of V.harveyi to return to its normal shape

All V. harveyi samples isolated from the

cross streaking point were cultured on TCBS agar.

These plates were incubated at 35 ∞C for 24 h and

morphological change was determined. The

samples were then subcultured every 24 h for three

consecutive times. TEM was used to observe the



change of shapes during the subculture times for

investigation o the recovery.

1.3 Bacillus spp. growth in differentconditions

Temperature, salinity and pH were tested

for the effect on the growth of Bacillus spp. The

bacteria were prepared on NA supplemented with

1.5% NaCl (w/v) and incubated at 35 ∞C for 24

hours. For temperature study, a single colony was

selected from the plate and streaked on the NA

supplemented with 1.5% NaCl (w/v) and incubated

at 25 ∞C, 35 ∞C and 50 ∞C for 24 h. The single

colony of Bacillus spp. from the plate was

inoculated in NB supplemented with 1.5% NaCl

(w/v). 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10% NaCl were

added to the media for salinity study and the pH to

3, 4, 5, 6, 7, 8, 9, 10, 11, and 12 for pH study. Test

materials were incubated at 35 ∞C for 24 hours.

2. Probiotic and immunology properties of threebacteria against V. harveyi in vivo

Penaeus monodon was obtained from

shrimp farm in Chachoengsao Province, Thailand.

Shrimp with a mean fresh weight of approximately

8-12 g per animal were used. They were

acclimatised in an aerated aquarium system at 25

ppt and change water every week before testing.

2.1 Probiotic properties studyThree strains of Bacillus, they were B.

pumilus, B. sphaericus, B. subtilis. Each bacterium

was formulated at the concentration of 1011-1012

cfu/g powder using clay as a filter. The experiment

was designed as a CRD with 8 treatments and 3

replications each as shown in Table 1.

Each treatment was blended with the

shrimp’s feed, at the ratio of 5 g : 1 kg. feed and

then fed at 3% of the body weight, three times

daily. The growth rate were observed at 4 weeks

after feeding. Scanning electron microscope (SEM)

was used to confirm the present of Bacillus in

shrimp intestine.

2.2 Immunology studyPreparation of hemolymph samples0.5 ml of hemolymph from each sample

was withdrawn from base of the third walking leg

of the shrimp using a syrynge containing 1.5 ml

anticoagulant (K-199 + 5% L-cysteine) at 4 weeks

after feeding.

2.2.1 Total hemocytesAfter collected hemolymph, hemocytes

were counted using a hemocytometer and

calculated as number of blood cells (total hemocytes

per cubic millimeter)

2.2.2 Phenoloxidase activity assayThe method was modified from

Supamattaya et al. (2000).After the blood was

withdrawn, the hemocytes were washed three times

with shrimp saline (1000 rpm. 4 ∞C 10 min).

Hemocyte lysate (HLS) was prepared from

hemocytes in a cacodylate buffer pH 7.4 by using

Table 1 Eight treatments of probiotic properties study.

Treatment Species of Bacillus spp.

1 B. pumilus

2 B. sphaericus

3 B. subtilis

4 B. pumilus + B. sphaericus (1:1)

5 B. pumilus + B. subtilis (1:1)

6 B. sphaericus + B. subtilis (1:1)

7 B. pumilus + B. sphaericus + B. subtilis (1:1:1)

8 No Bacillus (control)


the sonicator at 30 amplitute for 5 second and the

suspension was then centrifuged at 10,000 rpm.,

4∞C for 20 min. The supernatant was collected as

HLS. Then 200 ml of trypsin 0.1% in cacodylate

buffer was mixed to the 200 ml HLS followed by

200 ml of L-dihydroxyphenylalanine (L-DOPA) 4

mg/ml as the substrate. Enzyme activity was

measured as the absorbance of dopachrome at 490

nm wavelength. Measurement of protein content

in HLS was made by using the method of Lowry

et al. (1951). The phenoloxidase activity was

calculated as the increasing of optimum density

(OD) per minute per mg of protein as :

1 unit of phenoloxidase = D OD490 / min/

mg protein

2.2.3 Superoxide anion (O2-)

The method was modified from

Supamattaya et al. (2000). The O2- was detected

by reduction of redox dye, nitroblue tetrazolium

(NBT). By this method, hemocytes were washed 3

times in K-199 solution. Living cells were separated

by using trypan blue solution and adjusted to

1¥107 cell/ml suspension. 200 ml of the cell

suspension of each sample was dropped into well

of a 96 microwell sterile plate. The plate was left

for 45 min at room temperature for incubation

period. Unattached cells were washed out by K-

199 solution. 100 ml of the reaction mixture (0.5

mg zymozan in 0.5 ml serum + 20 mg NBT in 1 ml

DMSO + K-199) was added to each well and the

reaction mixture incubated at 25 ∞C 60 min. NBT

was reduced by O2- during incubation period into

a water insoluable blue formozan. The reaction

was inhibited by putting 70% methanol into the

samples for 3 min and the samples were then

allowed to air dry. 120 ml of 2M NaOH and 140 ml

of dimethyl sulfoxide (DMSO) were added to each

well in order to dissolve the formozan. The

concentration of the prussian-blue-colored solution

was measured at 620 nm, KOH/DMSO was used

as a blank control. The amount of O2- was indicated

by the increasing in absorbance at 620 nm of 0.001

from control.

2.2.4 Clearance abilityThe method of study the clearance ability

was modified from Martin et al. (1993). The

bacterial pathogen V. harveyi was subcultured in

TSA (Tryptic Soy Agar) with 1.5% NaCl and

incubated at 35∞C for 24 hours. A single colony of

V. harveyi was solved in the 1.5% NaCl sterile

water. The suspension with the OD of 0.13

(2.45 ¥ 108 cfu/ml) measured by absorbance value

at 640 nm was used to count for the number of

bacteria after cultured on TCBS agar by spread

plate technique. Then 0.1 ml of the bacterial

suspension with the counted number was injected

to each tested shrimp while the control was injected

with saline water. Three hours after injections, 0.5

ml of the blood from each shrimp was withdrawn

for counting the number of bacteria by spread plate

technique and calculated for the difference.

2.2.5 Bactericidal activitySerum was separated from hemolymph of

each shrimp sample and diluted by 2.6% NaCl in

the proportion of 1:2, 1:4, 1:8, 1:16 and 1:32 serum

to NaCl. The 0.5 ml of each serum dilution and 0.5

ml of NaCl as the control were used in the study.

0.5 ml of bacterial suspension, V. harveyi, prepared

from the method as in number 2.2.4 was put into

each serum dilution including the control. The

treatments were incubated at room temperature

for 3 hours before counting the number of bacteria

made by a spread plate technique. Recording of

the results were made for the dilution that could

decrease 50% V. harveyi compared to the control.

2.3 Study for the bacterial concentrationin shrimps intestine

The average concentration of both probiotic

bacteria and V. harveyi were determined after 4

weeks of feeding. Scanning Electron Microscope

(SEM) was used for confirmation the results.

RESULTS

Isolation and identification of Bacillus spp.Out of twenty isolates from shrimp


intestines, there were only three species belong to

the genus Bacillus and were identified as B. pumilus,

B. sphaericus and B. subtilis. The percentage of

each species were shown in Table 2

1. Study on the colonization and inhibitionactivities of three bacteria on Vibrio harveyi invitro

The results showed that only 12 hours after

the tested, B. pumilus had inhibition effect against

V. harveyi as shown in Figure 1.1. On the test plate,

some clear zone area were existed and more

colonization areas were observed after 24 hours as

shown in Figure 1.2. On the other hand, the clear

zone did not appear in the colonized plates of both

B. sphaericus and B. subtilis.

1.1 Study on the morphological changedof V. harveyi after colonization

The cross streaking point between V.

harveyi and each colonizing Bacillus spp. on the

TCBS agar was isolated for V. harveyi. Results

showed that V. harveyi colonized by B. pumilus

had slower growth compare with the control but

the others two V. harveyi colonized by B. sphaericus

and B. subtilis had normal growth with slightly

change as shown by TEM (Figure 2)

In addition, the cell of V. harveyi colonized

by B. pumilus, was almost destroyed as shown in

Figure 2.2, this might occurred from some

metabolites produced from B. pumilus. The size of

normal V. harveyi and those colonized by B.

pumilus, B. sphaericus and B. subtilis were about

0.71 ¥ 1.54 mm, 0.50 ¥ 0.78 mm, 0.68 ¥ 0.96 mm

Table 2 The result of isolate Bacillus spp. from shrimp intestine by VITEK 32.

Species of Bacillus spp. Number of isolated Percent

Bacillus pumilus 1 5

Bacillus sphaericus 2 10

Bacillus subtilis 3 15

Non Bacillus species 14 70

Total 20 100

Figure 1 Colonization activities of Bacillus spp.

3 species on V. harveyi in vitro.

1.1 inhibition effect of B. pumilus

against Vibrio harveyi after 12 hours.

1.2 inhibition effect of B. pumilus


1.3 colonization activities of B.

sphaericus against Vibrio harveyi after 24 hours.

1.4 colonization activities of B. subtilis


and 0.68 ¥ 1.07 mm as shown in Figure 2.1, 2.2, 2.3

and 2.5 respectively.

1.2 Investigation for the possibility of V.harveyi to return to its normal shape

After three consecutive subcultures those

deviated V. harveyi colonized from each bacterium

on the TCBS agar at every 24 hours confirmed that

their morphologies were permanently changed


compare with the control. Thus, it might not

possible for the colonized V. harveyi to return to its

regular size and shape again as shown in Figure

2.3, 2.5 and 2.7

1.3 Bacillus spp. growth in differentconditions

Table 3 showed that B. pumilus and B.

subtilis could grow at 25-50 ∞C but B. sphaericus

could not grow at 50 ∞C. All three Bacillus species

could grow in the salinity ranging between 0-10%

and pH ranging from 4 to 11.

2. Study on the probiotic and immunologyproperties of three bacteria against V. harveyiin vivo

2.1 Probiotic properties study

Percent mean weight increasing of shrimps

after 4 weeks fed by the mixture of B. sphaericus

+ B. subtilis (55.72 ± 24.43) is significantly higher

(P<0.05) than the single formulation of B. pumilus,

B. sphaericus, B. subtilis and the mixture of B.

pumilus + B. sphaericus, the mixture of B. pumilus

+ B. subtilis, the mixture of B. pumilus + B.

sphaericus + B. subtilis and control which percent

mean weight increasing were 9.99 ± 5.71, 19.44 ±9.62, 24.17 ± 23.46, 19.94 ± 4.25, 13.63 ± 7.27,

18.77 ± 7.64 and 12.46 ± 3.22, respectively as

shown in Figure 3.

2.2 Immunology study2.2.1 Total hemocytesMean of total hemocytes from shrimp

hemolymph after cultured with 8 feeds for 4 weeks

Figure 2 Morphological structure of normal V. harveyi compared with those deviated from colonization

by TEM.

2.1 normal V. harveyi

2.2 V. harveyi colonized by B. pumilus

2.3 V. harveyi colonized by B. pumilus after three consecutive subcultures

2.4 V. harveyi colonized by B. sphaericus

2.5 V. harveyi colonized by B. sphaericus after three consecutive subcultures

2.6 V. harveyi colonized by B. subtilis

2.7 V. harveyi colonized by B. subtilis after three consecutive subcultures


showed that B. subtilis (11.03 ± 1.51 ¥ 106), the

mixture of B. sphaericus + B. subtilis (11.28 ± 1.88

¥106), and the mixture of B. pumilus + B. sphaericus

+ B. subtilis (11.31 ± 1.46 ¥ 106) were significantly

higher (P<0.05) than those fed with B. pumilus, B.

sphaericus, the mixture of B. pumilus + B.

sphaericus, the mixture of B. pumilus + B. subtilis

and control which total hemocytes were 9.78 ±0.82, 10.02 ± 1.35, 10.70 ± 1.74, 10.41 ± 1.65 ¥ 106

and 9.19 ± 0.81 ¥ 106 cell/ml., respectively, as

shown in Figure 4.

2.2.2 PhenoloxidaseMean of phenoloxidase from shrimp

hemocytes after fed with B. subtilis (297.04 ±20.69), the mixture of B. pumilus + B. subtilis

(307.56 ± 15.31) and the mixture of B. pumilus +

B. sphaericus + B. subtilis (310.58 ± 29.58) for 4

weeks were significantly higher (P<0.05) than fed

with B. pumilus, B. sphaericus, the mixture of B.

pumilus + B. sphaericus, the mixture of B.

sphaericus + B. subtilis and control which

phenoloxidase were 278.53 ± 56.47, 253.79 ±

Table 3 Growth of Bacillus spp. in NB at 25, 35 and 50 ∞C; NaCl 0-10% and pH ranging from 3-12.

B. pumilus B. sphaericus B. subtilis

Temperature

25 ∞C + + +

35 ∞C + + +

50 ∞C + - +

Salinity

0% NaCl + + +

1% NaCl + + +

2% NaCl + + +

3% NaCl + + +

4% NaCl + + +

5% NaCl + + +

6% NaCl + + +

7% NaCl + + +

8% NaCl + + +

9% NaCl + - +

10% NaCl + - +

pH

3 - - -

4 + + +

5 + + +

6 + + +

7 + + +

8 + + +

9 + + +

10 + + +

11 + + +

12 - - -


0

20

40

60

80

100

control 1 2 3 4 5 6 7

feeds

b bb

b

b bb

a 1 = B.pumilus 2 = B.sphaericus 3 = B.subtilis, 4 = B.pumilus + B. sphaericus 5 = B.pumilus + B. subtilis 6 = B.sphaericus + B. subtilis 7 = B.pumilus + B. sphaericus +

B.subtilis

Weight increasing (%)

x 106 cells/ml

0

5

10

15

control 1 2 3 4 5 6 7

feeds

bbb

b ba a aunit/min/mg.protein

0

100

200

300

400

control 1 2 3 4 5 6 7feeds

b

aa a

b

a aa

28.92, 234.52 ± 26.77, 280.70 ± 30.64 and 222.75

± 15.34, respectively, as shown in Figure 5.

2.2.3 Superoxide anionMean of superoxide anion from shrimp

hemocytes after fed with B. subtilis (8.13 ± 2.16),

the mixture of B. pumilus + B. sphaericus (8.33 ±1.66), the mixture of B. sphaericus + B. subtilis

(8.40 ± 1.41) and the mixture of B. pumilus + B.

sphaericus + B. subtilis (8.80 ± 1.31) for 4 weeks

were significantly higher (P<0.05) from B. pumilus,

B. sphaericus, the mixture of B. pumilus + B.

subtilis and control which superoxide anion were

5.63 ± 1.22, 4.60 ± 1.35, 7.63 ± 0.91 and 5.30 ±0.92, respectively, as shown in Figure 6.

2.2.4 Clearance abilityClearance ability of shrimp blood

circulation system after fed with 8 feeds for 4

weeks and then injected with V. harveyi found that

shrimp fed with B. subtilis (346.67 ± 70.95), the

mixture of B. sphaericus + B. subtilis (253.33 ±122.20) and the mixture of B. pumilus + B.

sphaericus + B. subtilis (320.00 ± 98.49) could

reduce V. harveyi in hemolymph significantly

different (P<0.05) from shrimp fed with B. pumilus,

B. sphaericus, the mixture of B. pumilus + B.

sphaericus, the mixture of B. pumilus + B. subtilis

and control which amount of V. harveyi cells were

456.67 ± 85.05, 660.00 ± 105.36, 483.33 ± 201.08,

420.00 ± 40.00 and 626.67 ± 192.96 cfu/ml,

respectively, as shown in Figure 7.

2.2.5 Bactericidal activityBactericidal activity from shrimp serum at

4 weeks after fed with B. sphaericus, the mixture

of B. pumilus + B. subtilis , the mixture of B.

sphaericus + B. subtilis and the mixture of B.

pumilus + B. sphaericus + B. subtilis (8.80 ± 1.31)

at the dilution ratio of 1:8 serum to brine could

killed 50 percent of V. harveyi. While, shrimp fed

Figure 3 Percent weight increase of the P. monodon after 4 weeks cultured with 8 feeds.

Figure 4 Average of the P. monodon hemocytes

after 4 weeks cultured with 8 feeds.

Figure 5 Average of the P. monodon

phenoloxidase after 4 weeks cultured

with 8 feeds.


with B. pumilus, B. subtilis, the mixture of B.

pumilus + B. sphaericus and control killed 50

percent of V. harveyi at the dilution ratio of 1:4, as

shown in Figure 8.

2.3 Study for the bacterial concentrationin shrimps intestine

2.3.1 Bacillus spp.The number of Bacillus spp. in shrimp

intestine after fed with B. pumilus (803.33 ±395.01), B. sphaericus (536.67 + 310.86), B.

subtilis (359.00 ± 270.63), the mixture of B. pumilus

+ B. sphaericus (373.33 ± 156.31), the mixture of

B. pumilus + B. subtilis (216.33 ± 94.87), the

mixture of B. sphaericus + B. subtilis (493.33 ±187.17) and the mixture of B. pumilus + B.

sphaericus + B. subtilis (476.00 ± 234.03) for 4

weeks were significantly different (P<0.05) from

control which number of Bacillus spp. average

was 1.00 ± 1.73 ¥ 104 cfu/g, as shown in Figure 9.

2.3.2 Vibrio spp.The number of Vibrio spp. in shrimp

intestine at 4 weeks after fed with normal feed

(control) (970.00 ± 285.13 cfu/g) was significantly

different (P<0.05) from shrimp fed with B. pumilus,

cfu/ml.

0

200

400

600

800

1000

control 1 2 3 4 5 6 7

feeds

b

a

b

a

a

aa a

unit

0

2

4

6

8

10

12

control 1 2 3 4 5 6 7

feeds

bcdd

ab aabc

a a

cd

feeds

% of V. harveyi decreased

0

50

100

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

1: 2

1: 4

1: 8

1: 1

6 1

: 32

control 1 2 3 4 5 6 7 * * * * * * * *

Figure 6 Average of the P. monodon superoxide

anion after 4 weeks cultured with 8

feeds.

Figure 7 Cells of V. harveyi from P. monodon

hemolymph after infected with V.

harveyi 3 hours after 4 weeks cultured

with 8 feeds.

Figure 8 Value of diluted serum to kill V. harveyi 50% (*) of P. monodon after 4 weeks of culture when

provided with 8 feeds (1 = B. pumilus, 2 = B. sphaericus, 3 = B. subtilis, 4 = B. pumilus + B.

sphaericus, 5 = B. pumilus + B. subtilis, 6 = B. sphaericus + B. subtilis, 7 = B. pumilus + B.

sphaericus + B. subtilis).


B. sphaericus, B. subtilis, the mixture of B. pumilus

+ B. sphaericus, the mixture of B. pumilus + B.

subtilis, the mixture of B. sphaericus + B. subtilis

and the mixture of B. pumilus + B. sphaericus + B.

subtilis which the number of Vibrio spp. in shrimp

intestine were 11.33 ± 1.16, 121.67 ± 11.50, 8.00

± 1.00, 26.00 ± 3.00, 32.00 ± 3.61, 10.00 ± 3.00,

2.33 ± 1.53 cfu/g, respectively, as shown in Figure

10.

2.3.3 Scanning electron microscopeSEM revealed that Bacillus spp. could

survived in shrimp intestine after treated with

either individual isolation or mixtures of two and

three of B. pumilus, B. sphaericus and B. subtilis

as shown in Figure 11.1. While, in control group,

only Vibrio spp. was observed in shrimp intestine

as shown in Figure 11.2.

DISCUSSION

Isolation of Bacillus spp. from shrimp

intestine found three species namely B. pumilus,

B. sphaericus and B. subtilis. Original of those

bacteria might came from water, soil, food or

normal flora in the intestine. Bonde (1981) reported

that in seawater dominated by B. licheniformis

followed by B. subtilis and B. pumilus. Other

species encountered in low numbers include B.

0

200

400

600

800

1000

1200

1400

Control 1 2 3 4 5 6 7

x 104 cfu/g

c

a

ab

abab

abab

b

feeds

Figure 9 The number of Bacillus spp. in P.

monodon intestine after 4 weeks of

culture when provided with 8 feeds.

Figure 11 Scanning Electron Microscope picture.

11.1 Scanning Electron Microscope picture of Bacillus spp. in shrimp intestine after fed

with mixture of B. pumilus, B. sphaericus and B. subtilis ( ¥ 10,000)

11.2 Scanning Electron Microscope picture of Vibrio spp. in shrimp intestine (control

group) (¥ 15,000)

0

500

1000

1500

Control 1 2 3 4 5 6 7

cfu/g

feedsaa

aa a a a

b

Figure 10 The number of Vibrio spp. in P.

monodon intestine after 4 weeks of

culture when provided with 8 feeds.


brevis, B. firmus and B. sphaericus, largely in

nonpolluted areas. In a numerical study of North

Sea sediments, Boeyé and Herts (1976) found that

B. subtilis, B. licheniformis and B. firmus strains

predominated. So, it was possible for B. pumilus,

B. sphaericus and B. subtilis to contaminate in

intestinal of shrimp by sea water.

In vitro production of inhibitory compounds

toward known pathogens for the considered species

has often been used in the selection of putative

probiotic strains (Verschuere et al., 2000). In this

study we demonstrated that the isolated Bacillus

spp. from intestinal of black tiger shrimp are

potential competitors for V. harveyi, the results

showed colonization effect of each Bacillus spp. to

V. harveyi in 24 hours. Special results we received

from this experiment, the effective of B. pumilus

could produce some substance and effective to

destroy V. harveyi confirm this result by clear

zone area in the test plates. The inhibition of B.

pumilus to V. harveyi observed clear zone of V.

harveyi colony in 12 hours and eradicate all of V.

harveyi colony in 48-72 hours. Although B.

sphaericus and B. subtilis did not show colonization

effect but they showed some inhibition effect to V.

harveyi, which confirmed by distorted shape of V.

harveyi by TEM the shape of V. harveyi had

smaller size and some area of cell wall was

destroyed. The abnormal shape of V. harveyi were

permanently changes confirmed with three

consecutive subcultures. Bacillus spp. might

produced some metabolites for instance antibiotic

(Williams and Vickers, 1986) or enzymes for

inhibition and/or digestion (Bruno and Montville,

1993)

Regarding to the growth of Bacillus spp.,

this experiment might confirmed that Bacillus

spp. could grow in a wide range of environment,

such as salinity up to 8% NaCl, pH 4-11 and

temperature up to 50 ∞C, althought B. sphaericus

could not grow at 50 ∞C but in the real culture,

water temperature never raise so high as 50 ∞C.

Using B. subtilis mixed with B. sphaericus

could increase weight of shrimp significantly

difference (P<0.05) from control and other

experiment groups, while others treatments had no

significant different from the control (P>0.05).

About immunoenhancement ability of

Bacillus spp., Bacillus had peptidoglycan in its

cell wall which could increase immune of shrimp

(Boonyaratpalin et al., 2000). In this research, B.

subtilis had highest efficiency to improve immune

parameters of black tiger shrimp consists of total

hemocyte, phenol oxidase, superoxide anion and

clearance ability. Others treatments containing B.

subtilis in the mixtures also increased immune

parameters significantly difference from the control

as well. These results were similar to Rengpipat et

al. (2000). But in our research we measured

superoxide anion instead of phagocytic activity

because superoxide anion produced in phagocytosis

process (Bell and Smith, 1993).

Furthermore, after fed shrimp with Bacillus

spp. for 4 weeks revealed that Bacillus spp. showed

superiority in competition and colonization to

Vibrio spp. in shrimp intestine.

CONCLUSIONS

In summary, it had been demonstrated that

B. pumilus, B. sphaericus and B. subtilis had good

properties to be used as probiotic for black tiger

shrimp. B. pumilus showed colonization activity

and produce inhibitory compounds while B.

sphaericus and B. subtilis showed only inhibitory

effects to V. harveyi. TEM studied found that cell

morphology of V. harveyi colonized by B. pumilus,

B. sphaericus and B. subtilis changed to smaller

sizes compared with normal cell. Moreover, three

species of Bacillus spp. could grow in a different

environment, including salinity 0-8% NaCl, pH 4-

11 and temperature 25-50 ∞C except B. sphaericus

which could not grow at 50 ∞C. When fed shrimp

with these Bacillus spp. showed

immunoenhancement capability and also

colonization and inhibition on V. harveyi in shrimp


intestine. Therefore, these Bacillus spp. might be

applied as good probiotic in shrimp aquaculture.

ACKNOWLEDGEMENTS

This research was supported by Kasetsart

University Research and Development Institute

(KURDI). We thank Professor Dr. Toshiaki Itami,

Miyazaki University for immune techniques.

Assoc. Prof. Kidchakarn Supamattaya, Prince of

Songkla University, Thailand for chemical formula.

Ms.Kanokphan Srimanopach, Department of

Fisheries for VITEK 32 instrument, Ms. Patcharee

Umrung, Central lab KU for TEM and Ms. Yuppade

Paowpan, Central lab KU for SEM.

LITERATURE CITED

Bell, K. L. and V. J. Smith. 1993. In vitro superoxide

production by hyaline cells of the shore crab

Carcinus maenas Dev. Comp. Immunol.17: 211-219.

Boeyé, A. and M. Herts. 1976. Numerical

taxonomy of Bacillus isolates from North Sea

sediments. Int. J. Syst. Bacteriol. 26:427-

441.

Bonde, G. J. 1981. Bacillus from marine habitats

: Allocation to phena established by

numerical Techniques, pp. 181-215. In R. C.

W. Berkeley and M. Goodfellow (eds.). TheAerobic Endospore-Forming Bacteria :Classification and Identification, Academic

Press, London.

Boonyaratpalin, M., K. Supamattaya, J.

Pongmaneerat, S. Boonyaratpalin and

Y. Toride. 2000. The stimulatory effect of

high and low molecule peptidoglycan

(PG) on immune responses in black

tiger shrimp, Penaeus monodon Fabricius.

Songklanakarin J. Sci. Technol. 22 (Suppl.)

: 689-696.

Bruno, M. E. C. and T. J. Montville. 1993. Common

mechanistic action of bacteriocins from lactic

acidbacteria. Appl. Environ. Microbiol.59:3003– 3010.

Fuller R. 1992. Probiotics. The Scientific Basis.

Chapman and Hall, London. 398 p.

Gullian, M., F. Thompson and J. Rodriguez. 2004.

Selection of probiotic bacteria and study on

their immunostimulatory effect in Penaeus

vannamei. Aquaculture 233: 1-14.

Lowry, O.H., N. J. Rosebrough, A. L. Farr and R.

J. Randall. 1951. Protein measurement with

the folin phenol reagent. J. Biol. Chem. 193:

265-275.

Martin, G. G., D. Poole, C. Poole, J. E. Hose, M.

Arias, L. Reynolds, N. Mckrell and A. Whang.

1993. Clearance of bacteria injected into the

hemolymph of the Penaeid shrimp, Sicyonia

ingentis. J. Inver. Pathol. 62: 308-315.

Moriarty, D.J.W. 1998. Control of luminous Vibrio

species in penaeid aquaculture ponds.

Aquaculture 164: 351-358.

Rengpipat, S., S. Rukpratanporn, S.

Piyatirativarakul and P. Menasveta. 2000.

Immunity enhancement in black tiger shrimp

(Penaeus monodon) by a probiont bacterium

(Bacillus S11). Aquaculture 191: 271-288.

Supamattaya, K., J. Pongmaneerat and T.

Klowklieng. 2000. The effect of b–glucan

(MacroGard®) on growth performance,

immune response and disease resistance in

black tiger shrimp, Penaeus monodon

Fabricius. Songklanakarin J. Sci. Technol.22 : 677-688.

Verschuere, L., G. Rombaut, P. Sorgeloos and W.

Verstraete. 2000. Probiotic Bacteria as

Biological Control Agents in Aquaculture.

Microbiol. Mol. Biol. R. 64: 655–671.

Williams, S. T., and J. C. Vickers. 1986. The

ecology of antibiotic production. Microb.Ecol. 12:43–52.


Extracts of Thai Indigenous Vegetables as Rancid Inhibitorin a Model System

Plernchai Tangkanakul, Gassinee Trakoontivakornand Chansuda Jariyavattanavijit

ABSTRACT

The antioxidant property of twenty five vegetables extracted with ethanolic and water was

determined by monitoring their capacities to scavenge the stable free- radicals DPPH. Oxidative rancidity

in oil-in-water emulsion model was evaluated by ferric thiocyanate (FTC) method. Total phenolic content

was also determined by Folin-Ciocalteu method. The ethanolic extracts were found to exhibit a higher

phenolic content as well as DPPH radical scavenging activities than the water extracts. However, the data

indicated that both ethanolic and water extracts had dramatically antioxidant activity determined by FTC

method. Sixteen and eighteen plants of ethanolic and water extracts, respectively, performed greater

rancid inhibition than synthetic antioxidant (BHA, 10 ppm). This study found no relationship between

antioxidant activities through the DPPH radicals scavenging or through lipid radicals scavenging.

Key words: vegetables, antioxidant capacity, free radical scavenging activity, total phenolics, rancidity

INTRODUCTION

Synthetic antioxidants have been applied

for decreasing lipid oxidation during storage of

processed food products. The use of chemical

additives has raised questions regarding food safety

and toxicity (Chang et al., 1977). Many research

works have been directed toward safe antioxidants

with high antioxidative activity from natural

sources. The antioxidant properties of herbs and

spices, cinnamon, turmeric, clove, black pepper,

nutmeg, dry ginger, rosemary, sage and paprika

were continuously reported (Chang et al., 1977;

Nakatani et al., 1986; Kikuzaki and Nakatani.,

1993; Tomaino et al., 2005). Many plant extracts

were revealed on antioxidation efficiency when

applied in oils, fats and fat containing foods, meat

Institute of Food Research and Product Development, Kasetsart University, Bangkok 10900, Thailand.


products (Karpin′ska et al., 2001). For example,

rosemary and sage prolong the induction period in

chicken fat and show antioxidant activity

comparable with butylated hydroxyanisole (BHA)

and butylated hydroxytoluene (BHT) (Bracco et

al., 1981).

For Thai local vegetables, study on

antioxidant activity was reported regarding to the

potency of over 100 varieties by using b-carotene

bleaching method ((Nakahara and

Trakoontivakorn, 1999; Trakoontivakorn and

Saksitpitak, 2000; Na Thalang et al., 2001). More

than fifty varieties were reported to contain

antioxidant more than 100 mg BHA equivalent in

100g fresh weight. With their high potential and

lack in information for rancid inhibition, the present

study was aimed to evaluate 25 Thai vegetables for

anti-rancidity in a model system. Extract solvents,

water and ethanol, were taken into a consideration.

Total phenolic content was examined to find a

relationship between phenolic content and

antioxidant activity of crude extracts from both

solvents.

Presenting antioxidant activity of plant

crude extract was expressed in many aspects.

These methods revealed differently on mechanisms

of antioxidant defense system, i.e., inhibition of

lipid peroxidation, reduction of lipid peroxyl

radicals or scavenging of oxygen and hydroxyl

radicals (Tsushida et al., 1994; Velioglu et al.,

1998; Kähkönen et al., 1999; Pulido et al., 2000).

Stable free radicals, frequently applied to study

natural antioxidant efficiency were 2,2’-azino-

b i s (3 - e thy lbenzo th i azo l i ne -6 - su l fon i c

acid (ABTS), DPPH or N,N-Dimethyl-p-

phenylenediamine dihydrochloride (DMPD).

Koleva et al. (2002) recommended using DPPH

due to being simple, rapid, convenient and

independent of sample polarity.

Antioxidant capacity was another

perspective to present antioxidant activity, applying

different methodologies such as TEAC (Trolox

equivalent antioxidant capacity, Cook et al., 1998),

b-carotene bleaching method expressed as BHA

content (Tsushida et al., 1994), stable radical

ABTS or DPPH expressed as vitamin C (Kim et al,

2002). These antioxidant capacity methods were

developed in an attempt to have a meaningful

interpretation relating to health. In this study,

rancid inhibition was concerned, therefore,

scavengers of DPPH were stated in equivalent to

BHA content.


Sample preparationTwenty five Thai indigenous vegetables

were either purchased from local markets or

collected from nature (Table 1). Edible portions of

vegetable were weighed in 10 g/plastic bags and

kept at -20 ∞C until extracting.

Vegetable extractionIndividual vegetable was extracted with 2

different solvents, 95% aqueous ethanol and

distilled water. The 10 g of frozen vegetables were

homogenized (Ultra Turrax) in 40 ml 95% aqueous

ethanol or distilled water at room temperature for

1 min. and centrifuged at 10000 g for 10 min. The

residue was re-extracted with either 95% aqueous

ethanol or distilled water and extracts were pooled

and made to 100 ml. The extracts were stored in

capped bottles and kept at -20∞C until further use

for antioxidant capacity and total phenolics

determinations and rancid inhibition test.

DPPH Radical scavenging activity andantioxidant capacity

DPPH scavenging activity was determined

using a modified method of Onichi et al. (1994).

The free radical scavenging activity of vegetable

extracts were tested, indicated as bleaching of the

stable 1,1 –diphenyl-2-picrylhydrazyl radical

(DPPH). A diluted extract of the right concentration

to posses not more than 60% scavenging activity

(%SA), 0.15 ml, was added to 0.9 ml of 0.1 mM

DPPH dissolved in 95% ethanolic solution. The

mixture was vertexed and allowed to stand at room

temperature. After 20 min., the absorbance was

recorded at 517 nm. 95% aqueous ethanol was

used as a control. Percentage of DPPH scavenging

activity (%SA) was calculated from this equation

(C-X)100/C, where C = absorbance of control and

X =absorbance of extract.

In order to express antioxidant activity of

plant extract to an easily understood manner,

antioxidant capacity as mg t-butylated

hydroxyanisole equivalent (BHAE) /g fresh

vegetable was introduced. A standard curve of t-

butylated hydroxyanisole (BHA) was obtained

from DPPH %SA (x) plotted against various BHA

concentrations (y). Prepared concentrations of

BHA solution were 0.1, 0.25, 0.5, 1.0, 2.5 and 5.0


276 Kasetsart J. (Nat. Sci.) 39 (2)T

able

1A

ntio

xida

nt c

apac

ity (

AC

), r

epor

ted

as B

HA

equ

ival

ent (

BH

AE

), a

nd to

tal p

heno

lic (

TP)

con

tent

, rep

orte

d as

gal

lic a

cid

equi

vale

nt (

GA

E),

of

vege

tabl

es.

Bot

anic

al n

ame

Loc

al n

ame

Tes

ted

Plac

e an

d tim

eD

ryE

than

olic

ext

ract

Wat

er e

xtra

ctR

atio

of

part

of

colle

ctio

n m

atte

rA

CT

PA

CT

PA

C,

(%)

(mg

BH

AE

(mg

GA

E/

(mg

BH

AE

(mg

GA

E/

EtO

H/w

ater

/g F

W)

g FW

)/g

FW

)g

FW)

Aga

nosm

a m

argi

nata

Sato

nL

eaf

Ubo

n R

atch

atha

ni,

31.1

613

1.72±1

4.38

42.1

6±5.

4917

.73±

3.15

18.4

6 ±0.

067.

4M

arch

Ana

card

ium

occ

iden

tale

Mam

uang

Lea

fSu

rat T

hani

, Mar

ch23

.06

176.

48±3

2.07

51.3

3±4.

4268

.25±

9.86

30.0

8±10

.67

2.6

him

apha

nA

rchi

dend

ron

jiri

nga

Nie

ngFr

uit

Sura

t Tha

ni, M

arch

28.8

615

.09±

7.64

6.71

±3.8

44.

78±3

.50

3.53

±1.3

73.

2B

arri

ngto

nea

acut

angu

laK

rado

n na

mL

eaf

Ubo

n R

atch

atha

ni,

19.6

131

.85 ±

0.81

11.8

6±0.

1119

.75±

1.76

7.12

±0.6

41.

6M

arch

Car

eya

spha

eric

aK

rado

n bo

gL

eaf

Sako

n N

akor

n,28

.15

174.

18±1

4.67

54.0

2±9.

4534

.63±

7.43

17.6

8 ±1.

145.

0Fe

brua

ryC

ente

lla

asia

tica

Bai

bua

bog

Lea

fSu

rat T

hani

, Mar

ch8.

721.

24±0

.15

1.18

±0.2

20.

14±0

.01

0.46

±0.0

58.

7C

olub

rina

asi

atic

aG

an to

ngL

eaf

Lam

pang

, Apr

il21

.94

4.06

±0.0

94.

94±0

.69

1.05±0

.23

3.86

±0.5

43.

9C

rato

xylu

m fo

rmos

umT

ewFl

ower

Sako

n N

akor

n,18

.74

53.9

1±0.

9733

.77±

3.54

31.0

4±6.

7116

.35 ±

1.95

1.7

Febr

uary

Cym

bopo

gon

citr

atus

.T

a ka

iSt

emB

angk

ok, M

ay17

.18

1.95

±0.0

31.

25±0

.04

0.51±0

.26

0.36

±0.0

23.

8D

rege

a vo

lubi

lis

Hua

nL

eaf

Lam

pang

, Apr

il14

.26

3.39

±0.0

73.

39±0

.08

1.56±0

.73

1.95

±0.0

52.

2E

ugen

ia g

rata

Sa m

egL

eaf

Sako

n N

akor

n,17

.60

88.7

2±0.

9525

.34±

4.96

46.7

7±9.

9111

.27±

0.91

1.9

Febr

uary

Fer

onie

lla

luci

daM

a su

ngFl

ower

Sako

n N

akor

n,24

.32

25.3

4±0.

3434

.51±

1.14

4.78±0

.59

12.8

7±2.

205.

3Fe

brua

ryG

linu

s op

posi

tifo

lius

Khe

e kh

omW

hole

Sako

n N

akor

n,8.

380.

03±0

.00

0.62

±0.0

30.

03±0

.03

0.38

±0.0

01.

0Fe

brua

ryG

loch

idio

n w

alli

chia

num

Mun

pu

Lea

fSu

rat T

hani

, Mar

ch25

.48

167.

22±1

9.06

49.0

6±7.

3233

.86±

4.15

11.4

3±0.

114.

9G

netu

m g

nem

onL

ieng

Lea

fSu

rat T

hani

, Mar

ch20

.14

1.20

±0.0

71.

76±0

.10

0.18±0

.06

1.74

±0.2

76.

7G

ymne

ma

inod

orum

Chi

engd

aL

eaf

Lam

pang

, Apr

il16

.76

1.77

±0.0

83.

61±0

.15

1.51±0

.88

3.60

±0.4

21.

2Ip

omoe

a aq

uati

caPh

ak b

ung

thai

Lea

f, s

tem

Ban

gkok

, May

8.22

9.56

±0.0

25.

74±0

.66

1.56±0

.05

1.91

±0.0

86.

1L

asia

spi

nosa

Phak

nha

mL

eaf,

ste

mSa

kon

Nak

orn,

8.16

1.65

±0.0

71.

11±0

.08

0.45±0

.02

0.73

±0.0

93.

7Fe

brua

ryL

euca

ena

leuc

ocep

hala

Kra

tinL

eaf,

ste

mB

angk

ok, M

ay16

.33

12.8

2±0.

157.

28±0

.37

4.60±1

.83

13.1

3±0.

302.

8L

imon

ophi

la a

rom

atic

aPh

ak k

a yh

eng

Lea

f, s

tem

Ubo

n R

atch

atha

ni,

9.76

10.4

8±0.

976.

51±0

.29

0.34±0

.15

0.97

±0.0

531

.2M

arch

Man

gife

ra in

dica

Ma

mua

ngL

eaf

Lam

pang

, Apr

il22

.47

114.

36±8

.39

40.2

2±4.

2737

.51±

8.06

12.5

1±1.

893.

0M

icro

mel

um m

inut

umM

hui

Lea

fSu

rat T

hani

, Mar

ch25

.18

24.7

6±2.

1312

.64±

0.49

6.13±0

.56

12.5

5±3.

264.

0P

arki

a sp

ecio

saSa

tor

Lea

fSu

rat T

hani

, Mar

ch26

.26

23.4

4±3.

7511

.73±

0.30

15.0

8±3.

115.

95±0

.34

1.6

Pas

sifl

ora

foet

ida

Ka

tok

rok

Lea

fU

bon

Rat

chat

hani

,19

.19

7.82±2

.80

5.02

±0.4

72.

55±0

.59

2.59

±0.5

03.

1M

arch

Spir

ogyr

a sp

.T

aoFi

lam

ents

Sako

n N

akor

n,3.

9014

.71±

2.17

4.08

±0.0

66.

74±0

.09

1.70

±0.1

12.

2Fe

brua

ry


mg/100 ml 95% ethanol. The regression line was

y = 0.0832x - 0.0469.

Determination of total phenolic contentsTotal phenolics were determined using the

Folin-Ciocalteau reagent, adopted from Singleton

and Rossi (1965). Two millilitres of suitable diluted

vegetable extracts was transferred and reacted

with 10 ml of Folin-Ciocalteau reagent (previously

diluted 10 fold with distilled water) in 25 ml

volumetric flask. After 30 sec. and before 8 min.,

8 ml of 7.5% of sodium carbonate was added and

mixed, and the contents of the flask made to

volume with distilled water. Solutions were heated

in a 40∞C water bath for 30 min. The color was

developed and absorbance measured at 765 nm.

The standard curve was prepared using 0, 0.5, 1.0

and 1.5 ml of gallate stock solution (8 mg/100ml)

in 25 ml volumetric flask. The regression line

between absorbance (y) and gallic acid content (x)

was y = 0.0046x + 0.0163. The results were

expressed as mg gallic acid equivalent / g of fresh

vegetable.

Antioxidative assay in model systemThe lipid oxidation was monitored by ferric

thiocyanate (FTC) method described by Kikuzaki

and Nakatani (1993). A 20 ml experimental mixture

made of 4 ml of an ethanolic extract (0.4 g fresh

vegetable), 4.1 ml of 2.51% linoleic acid in 99.5%

ethanol, 8 ml of 0.05 M phosphate buffer (pH 7.0)

and 3.9 ml of water was placed in a plastic bottle

(25 mm diameter, 60 mm height) with a screw cap.

For water extract, an assay mixture was prepared

the same as above except 3.9 ml of water was

replaced by 99.5% ethanol. The mixtures were

placed in an oven at 40∞C in the dark. FTC was

carried out by adding 0.1 ml of incubated mixture,

9.7 ml of 75% ethanol and 0.1 ml of 30%

ammonium thiocyanate into a test tube. Precisely

3 min after addition of 0.1 ml of 0.02M ferrous

chloride in 3.5% hydrochloric acid to the reaction

mixture, the absorbance of red color was measured

at 500 nm. Antioxidative assay was carried out

with 7 days interval until the absorbance of the

tested mixture reached maximum or 142 days.

BHA at concentration of 1.0 mg/100 ml and 5.0

mg/100 ml were used as a control.


Antioxidant capacityWithin twenty five vegetables, ethanolic

extract was found to obtain more antioxidant

capacity than water extract (Table 1). The result

agreed with Kaur and Kapoor (2000) who applied

b-carotene bleaching test. Ratio of antioxidant

capacity of components soluble in ethanol to water

of 25 tested vegetables varied in a range of 31.2 to

1.0, with mean 4.7 and median of 3.2. Limonophila

aromatica was the one that contained antioxidants

most susceptible to dissolve in ethanol. And it was

also revealed that antioxidant capacity of ethanolic

extracts was in different order from that of water

extracts, however, not dramatically.

A large variation in the antioxidant capacity

was observed in ethanolic extracts, ranging from

as high as 176.48 mg BHAE /g fresh weight of

Anacardium occidentale to as low as 0.03 mg

BHAE /g fresh weight of Glinus oppositifolius.

The ranking of five vegetables possessing high

antioxidant capacity through ethanol extraction

was Anacardium occidentale, Careya sphaerica,

Glochidion wallichianum, Aganosma marginata

and Mangifera indica. The result agreed to

Trakoontivakorn and Saksitpitak (2000) who

reported that methanolic extracts of Anacardium

occidentale, Careya sphaerica, Glochidion

wallichianum and Mangifera indica contained great

amount of antioxidants analyzed by b-carotene

bleaching method. The results of water extract

were found to have antioxidant capacities between

0.03 - 68.25 mg BHAE /g fresh vegetable. The five

greatest antioxidant capacities were Anacardium

occidentale, Eugenia grata, Mangifera indica,

Careya sphaerica and Glochidion wallichianum.


Total phenolic contentThe results of phenolic analysis of twenty

five vegetables are exhibited in Table 1. The

phenolic contents of the vegetable in ethanol

extraction varied from 0.62 mg GAE/g fresh weight

of Glinus oppositifolius to 54.02 mg GAE/g

fresh weight of Careya sphaerica. The ethanolic

extractable phenolic compounds greater than

25 mg GAE/g fresh vegetable were found

in Anacardium occidentale, Glochidion

wallichianum, Aganosma marginata, Mangifera

indica, Feroniella lucida, Cratoxylum formosum

and Eugenia grata. In water extraction, the total

phenolic contents was found highest in Anacardium

occidentale, 30.08 mg GAE/g fresh weight,

followed by Aganosma marginata, Careya

sphaerica, Cratoxylum formosum and Leucaena

leucocephala as 18.46, 17.68, 16.35 and 13.13 mg

GAE/g fresh weight, respectively. The present

study demonstrated that total phenolics in tested

Thai vegetables generally were greater than those

of Western herbs, 0.23 -17.51 mg GAE/g fresh

weight, reported by Zheng and Wang (2001).

Phenolic compounds were found to be

generally susceptible to dissolve more in ethanol

than in water. An exception was observed in

Leucaena leucocephala where extractable phenolic

compounds were almost double in water extract

than in ethanolic extract. Micromelum minutum,

Gymnema inodorum and Gnetum gnemon

possessed similar phenolic contents in both extract

media. A study reported on phenolic components

extracting ability that ethanol was less effective

than acetone when applied in Greek oregano and

summer savory. However, acetone extracts

exhibited incredible low in DPPH scavenging

activity (Exarchou et al., 2002). With this detection,

therefore, an extract medium should be taking into

account in order to obtain antioxidative compounds.

Phenolic contents existed in these twenty

five Thai indigenous vegetables were considered

as moderate to high. Converting vegetable weight

into dry weight basis, ethanolic extractable phenolic

compounds ranged from 7.28 mg GAE/g dry weight

in Cymbopogon citratus to 222.59 mg GAE/g dry

weight in Anacardium occidentale. Comparing to

the result of Kähkönen et al. (1999), the selected

cereals and vegetables contained greatly lower

amount of phenolics, 0.2 – 6.6 mg GAE/ g dry

weight and in moderate level in herb extracts, 9.1

– 23.1 mg GAE/g dry weight.

A relationship between phenolic content

and antioxidant activity was extensively

investigated, and both positive and negative

correlationships were demonstrated. Velioglu et

al. (1998), Rapisarda et al. (1999), Zheng and

Wang (2001), and many other research groups

stated that there was a positive correlation. In the

mean time, a few evidences of no significant

correlation were confronted (Heinonen et al., 1998;

Kähkönen et al., 1999). In this study, the regression

analysis was done separately on the extracting

medium, ethanol and water. The results revealed

that both ethanolic (R2 = 0.8893) and water extracts

(R2 = 0.6601) held a positive linear relationship

between phenolic content and antioxidant capacity

as displayed in Figure 1. The degree of correlation

coefficient pronounced greatly in ethanolic extracts

(R = 0.9430) and lesser in water extracts (R =

0.8125).

Antioxidative effect in model systemThe oxidation of oil-in-water emulsion was

monitored by the ferric thiocyanate (FTC) method.

The FTC method was used to measure the amount

of peroxide in initial stages of lipid oxidation.

Peroxide oxidizes ferrous iron to the ferric state

resulted in the formation of a red thiocyanate

complex. The determined values at 500 nm with

low absorbance indicated a high ability to delay

rancidity. On the contrary, if high absorbance

values and sharp increasing curve were noted, an

uncontrollable lipid oxidation stage was

confronted. In this study, extracts from 25

vegetables were analyzed and illustrated separately

by the extracting media (Figure 2).


R2 = 0.6601

R2 = 0.8893

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0.00 50.00 100.00 150.00 200.00

Antioxidant capacity (mg BHAE/g FW)

Tot

al p

heno

lics

(mg

GA

E/g

FW

)

Ethanolic extracts

Water extracts

Figure 1 Relationship between total phenolic content and antioxidant capacity of 25 vegetables,

extracted by ethanol and water.

Within 142 days of rancid inhibiting

evaluation, only 9 vegetable ethanol extracts were

lost or deteriorated in their abilities indicated by

reaching an absorbance of 0.4 (Figure 2a). They

were Cratoxylum formosum, Aganosma marginata,

Anacardium occidentale, Glinus oppositifolius,

Feroniella lucida, Micromelum minutum, Careya

sphaerica, Spirogyra sp. and Gnetum gnemon.

Absorbance of the control, no antioxidant in the

system, reached absorbance of 0.4 after day 7.

Seven out of 9 vegetables and 0.04 mg BHA

reached absorbance of 0.4 within 60 days. From

the whole experiment, only the extracts from Glinus

oppositifolius, Spirogyra sp., Gnetum gnemon,

0.04 mg BHA and 0.2 mg BHA displayed a sharp

raise curve similar to the control. A sharp changing

in absorbance is a common oxidation pattern

monitored by FTC method as shown in other

studies (Kikuzaki and Nakatani, 1993; Chen and

Ho, 1997). The indicator used to monitor the

generation of rancid odor from oil oxidation during

incubation was the absorbance of 0.4 by FTC as

reported by Chen and Ho (1997). Extract of

Anacardium occidentale and Careya sphaerica

could retain absorbance values around 0.4

indicating potential antioxidant plants, and there

were other 16 vegetables performed better than 0.2

mg BHA. The plant extracts that exhibited

substantial ability in rancid inhibition were

Gymnema inodorum, Dregea volubilis,

Cymbopogon citratus, Colubrina asiatica,

Passiflora foetida, Lasia spinosa and Centella

asiatica.

Water extracts illustrated a good result as

rancid inhibitors as well (Figure 2b). The plants

that possessed this property were Leucaena

leucocephala, Gymnema inodorum, Dregea

volubilis, Archidendron jiringa, Ipomoea aquatica,

Passiflora foetida and Spirogyra sp.. Water extracts

of plant that could not inhibit rancidity were

Glinus oppositifolius, Cratoxylum formosum,

Aganosma marginata, Gnetum gnemon, Lasia

spinosa, Cymbopogon citratus, Limonophila

aromatica and Centella asiatica.

It was interesting to point out that some

plants performed oppositely in controlling rancidity

in this model system. Water extract of Spirogyra

sp. contained effective antioxidants but not in

ethanolic extract. Whiles, ethanolic extracts of

Centella asiatica, Lasia spinosa and Cymbopogon

citratus were effective antioxidant, they became

ineffective when extracted with water. The plants


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 20 40 60 80 100 120 140

Time (days)

Abs

orba

nce

500

nm

AM

AO

AJ

BA

CS

CAU

CAB

CF

CC

DV

EG

FL

GO

GW

GG

GI

IA

LS

LL

LA

MI

MM

PS

PF

S

Control

BHA 0.04 mg

BHA 0.2 mg

a

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 20 40 60 80 100 120 140

Time (days)

Abs

orba

nce

500

nm

AM

AO

AJ

BA

CS

CAU

CAB

CF

CC

DV

EG

FL

GO

GW

GG

GI

IA

LS

LL

LA

MI

MM

PS

PF

S

Control

BHA 0.04 mg

BHA 0.2 mg

b

Figure 2 Oxidation of oil-in-water emulsion monitored by the ferric thiocyanate method (a) ethanol

extraction (b) water extraction. Aganosma marginata (AM), Anacardium occidentale (AO),

Archidendron jiringa (AJ), Barringtonea acutangula (BA), Careya sphaerica (CS), Centella

asiatica (CAU), Colubrina asiatica (CAB), Cratoxylum formosum (CF), Cymbopogon

citrates (CC), Dregea volubilis (DV), Eugenia grata (EG), Feroniella lucida (FL), Glinus

oppositifolius (GO), Glochidion wallichianum (GW), Gnetum gnemon (GG), Gymnema

inodorum (GI), Ipomoea aquatica (IA), Lasia spinosa (LS), Leucaena leucocephala (LL),

Limonophila aromatica (LA), Mangifera indica (MI), Micromelum minutum (MM), Parkia

speciosa (PS), Passiflora foetida (PF), Spirogyra sp.(S)


that displayed as a good antioxidant sources when

either extracted with water or ethanol were

Passiflora foetida, Dregea volubilis and Gymnema

inodorum.

Results from the model system experiment

revealed that plant extracts good in scavenging

DPPH radical was not always good in scavenging

lipid radicals. A relationship between BHAE

content and incubation time (reached absorbance

of 0.4) was investigated. BHAE content of tested

vegetables calculated from antioxidant capacity

which used DPPH scavenging method as a tool.

The regression values of ethanolic extract and

water extract were 0.093 and 0.0122, respectively

(Figure 3).

CONCLUSION

It was apparent from the study that plant

extracted by different solvents scavenged DPPH

radical and lipid radicals differently. Water extracts

of some vegetables were efficient comparable to

their ethanolic extracts for oxidative rancidity

inhibition. And the model of scavenging DPPH

radical could not be used to predict lipid oxidative

inhibition. This study also found that extractable

antioxidants from 0.4 g of some vegetables

demonstrated stronger rancid inhibition than the

synthetic antioxidant, 10 ppm BHA. It thus

constitutes an interesting source for use as natural

protecting agent to prevent oxidative deterioration

of food.

ACKNOWLEDGMENTS

Kasetsart University Research and

Development Institute (KURDI) financially

supported this research.

LITERATURE CITED

Bracco, U., J. Loliger and Viret J.L. 1981.

Production and use of natural antioxidans.

J. Am. Oil Chem. Soc. 58:686-690.Chang, S.S., B. Ostric-Matijasevic, O.A. Hsieh

and C.L. Chang. 1977. Natural antioxidants

from rosemary and sage. J. Food Sci. 42:1102-

1106.

Figure 3 Relationship between BHAE content obtained from antioxidant capacity and incubation time

to reach an absorbance of 0.4 by ferric thiocyanate method of 25 vegetables, extracted by

ethanol and water.

0

10

20

30

40

50

60

70

80

0 20 40 60 80 100 120 140 160 180 200

Incubation time (days)

BH

AE

con

tent

(m

g)

Ethanolic extracts

Water extracts


Chen, J.H. and C-T Ho. 1997. Antioxidant activities

of caffeic acid and its related hydroxycinnamic

acid compounds. J. Agric. Food Chem.

45:2374-2378.

Cook, J.A., D.J. Vanderjagt, A. Dasgupta, G.

Mounkaila, R.S. Glew, W. Blackwell and

R.H. Glew. 1998. Use of the Trolox assay to

estimate the antioxidant content of seventeen

edible wild plants of Niger. Life Sci. 63:105-

110.

Exarchou, V., N. Nenadis, M. Tsimidou, I.P.

Gerothanassis, A. Troganis and D. Boskou.

2002. Antioxidant activities and phenolic

composition of extracts from Greek oregano,

Greek sage and summer savory. J. Agric.Food Chem. 50:5294-5299.

Heinonen, M., P.J. Lehtonen and A. Hopia. 1998.

Antioxidative activity of berry and fruit wines

and liquor. J. Agric. Food Chem. 46:25-31.

Karpin′ska, M., J. Borowski and M. Danowska-

Oziewicz. 2001. The use of natural

antioxidants in ready-to-serve food. FoodChemistry 72:5-9.

Kaur, C. and H.C. Kapoor. 2002. Anti-oxidant

activity and total phenolic content of some

Asian vegetables. Inter. J. Food Sci. Tech.

37:153-161.

Kähkönen, M.P., A.L. Hopia, H.J. Vuorela, J-P.

Rauha, K. Pihlaja, T.S. Kujala and M.

Heinonen. 1999. Antioxidant activity of plant

extracts containing phenolic compounds.

J. Agric. Food Chem. 47:3954-3962.

Kikuzaki, H. and N. Nakatani. 1993. Antioxidant

effects of some ginger constituents. J. FoodSci. 58:1407-1410.

Kim, D-O., W.L. Ki, H.J. Lee and C.Y. Lee. 2002.

Vitamin Cequivalent antioxidant capacity

(VCEAC) of phenolic phytochemicals.

J. Agric. Food Chem. 50:3713-3717.

Koleva, I.I., T.A. van Beek, J.P.H. Linssen, A. de

Groot and L.N. Evstatieva. 2002. Screening

of plant extracts for antioxidant activity: a

comparative study on three testing methods.

Phytochemical Analysis 13:8-17.

Na Thalang, V., G. Trakoontivakorn and

K.Nakahara. 2001. Determination of

antioxidant activity of some commonly

consumed leafy vegetables in Thailand.

JIRCAS J. 9:39-46.

Nakahara, K. and G.Trakoontivakorn. 1999.

Antioxidative and autimutagenic properties

of some local agricultural products in Thailand.

In Highlight of Collaborative ResearchAcfivities between Thai ResearchOrganizations and JIRCAS. Provided for

JIRCAS Seminar in Bangkok 1999, Bangkok,

Thailand 3 March 1999. pp.141

Nakatani, N., R. Inatani, H. Ohta and A. Nishioka.

1986. Chemical constitutents of pepper and

application of food preservation, naturally

occurring antioxidative compounds.

Environmental Health Perspectives.

67:135-141.

Ohnishi, M., H. Morishita, H. Iwahashi, S. Toda,

Y. Shirataki, M. Kimura and R. Kido. 1994.

Inhibitory effects of chlorogenic acids on

linoleic acid peroxidation and haemolysis.

Phytochemistry 36 (3):579-583.

Pulido, R., L. Bravo and F. Saura-Calixto. 2000.

Antioxidant activity of dietary polyphenols

as determined by a modified ferric reducing/

antioxidant power assay. J. Agric. FoodChem. 48:3396-3402.

Rapisarda, P., A. Tomaino, R.L. Cascio, F. Bonina,

A. De Pasquale and A. Saija. 1999. Antioxidant

effectiveness as influenced by phenolic content

of fresh orange juices. J. Agric. Food Chem.

47:4718-4723.

Singleton, V.L. and J.A. Rossi. 1965. Colorimetry

of total phenolics with phosphomolybdic-

phosphotungstic acid reagents. Am. J. Enol.Vitic. 16:144-158.

Tomaino, A., F. Cimino, V. Zimbalatti, V. Venuti,

V. Sulfaro, A. De Pasquale and A. Saija.

2005. Influence of heating on antioxidant

activity and the chemical composition of some


spice essential oils. Food Chemistry 89:549-

554

Trakoontivakorn, G. and J. Saksitpitak. 2000.

Antioxidative potential of Thai indigenous

vegetable extracts. Food 30:164-176.

Tsushida, T., M. Suzuki and M. Kurogi. 1994.

Evaluation of antioxidant activity extracts

and determination of some active compounds.

Nippon Shokuhin Kogyo Gakhaishi.

41:611-618.

Velioglu, Y.S., G. Mazza, L. Gao and B.D. Oomah.

1998. Antioxidant activity and total phenolics

in selected fruits, vegetables and grain

products. J. Agric. Food Chem 46:4113-

4117.

Zheng, W. and S.Y. Wang. 2001. Antioxidant

activity and phenolic compounds in selected

herbs. J. Agric. Food Chem. 49:5165-5170.


Screening and Characterization of Lactic Acid Bacteria ProducingAntimicrobial Substance against Staphylococcus aureus

Chatinan Ratanapibulsawat1, Pumrussiri Kroujkaew2,Ohmomo Sadahiro2 and Sunee Nitisinprasert1

ABSTRACT

One isolate of lactic acid bacteria (LAB) producing antimicrobial substance (AMS) against

Staphylococcus aureus was selected and designed as LS2-30. Morphological, physical and biochemical

tests, and 16S rRNA analysis were performed and revealed that it was Pediococcus acidilactici. Optimal

conditions for growth and AMS production were determined. The optimum temperature and initial pH

for both growth and AMS production were 45 ∞C and pH 6, respectively. Cell-free culture fluid of LS2-

30 showed an inhibitory action against Staphylococcus aureus, Salmonella sp., Bacillus sp., and E. coli

but no activity against all LAB tested. When mode of action was studied, the cell-free culture fluid

exhibited a bactericidal mode of action but did not lyse its cells. After removing organic acids, partial

purified antimicrobial substance at pH 3 showed inhibitory activity as bactericidal action. Hydrolytic

enzyme treatment suggested that AMS structure was neither proteinaceous, starch nor lipid moiety in

nature. Therefore, the nature of inhibitory compound secreted by P. acidilactici LS2-30 was still unclear.

Key words: antimicrobial substance, lactic acid bacteria, Staphylococcus aureus

INTRODUCTION

Staphylococcus aureus is an important

opportunistic pathogen that causes a variety of

diseases in humans and animals. It is widely

recognized that antibiotic therapies formulated for

intramammary use are generally unsuccessful in

eliminating existing S. aureus infections or in

preventing the establishment of chronic udder

disease (Leitnera et al., 2003). Currently, more

than 95% of infectant with S. aureus infections

worldwide do not respond to first-line antibiotics

such as penicillin or ampicillin (Rubin et al.,

1999). Moreover, with increasing public concern

over food safety, there is regulatory pressure to

1 Department of Biotechnology, Faculty of Agro-Industry, Kasetsart University, Bangkok 10900, Thailand.2 Department of Animal Products, National Institute of Animal Industry (NIAI), Tsukuba, Ibaraki, 305-0901 Japan.


justify the use of therapeutic drugs in dairy cattle

and to reduce the incidence of residues in milk.

The emergence and spread of drug-resistant

staphylococci underscores the need to find new

modes of prevention and alternative non-antibiotic

treatment.

In a recent in vitro study, various substances

have been found and shown antimicrobial activity

against S. aureus, like, lactose chitosan derivative

against S. aureus CCRC 12657 (Chen and Chou,

2005), the extract of Panax ginseng root

(ginsenoside) (Hu et al., 2003) and lactoferrin with

penicillin G (Diarra et al., 2002). In addition, many

anti-S. aureus substances producing LAB have

been a great interest because most of the LAB are

considered as “Generally Recognized as Safe”

microorganism (Holzapfel et al., 1995). Several

LAB species were found to be a potent anti-S.

aureus substance producer such as Lactobacillus

plantarum BS, Lb. salivarius, Lb. plantarum 35d

(Messi et al., 2001), P. acidilactici F

(Osmanagaoglu et al., 1998) and Lb. acidophilus

LA1 (Bernet-camard et al., 1997). However, most

of them produced the substances that inhibited

themselves as well. Therefore, it is still important

to isolate a novel anti-S. aureus substance with

new properties and without significant inhibitory

activity against its host.

The objective of this study was to

characterize LAB producing antimicrobial

substance against S. aureus. The antibacterial

spectrum and their mode of action were described

as well.


Bacterial strains and growth conditionsAntimicrobial substance (AMS) producing

lactic acid bacteria isolated from silage and

fermented pork (Nham) were obtained from the

culture collection of Department of Biotechnology,

Kasetsart University (Thailand). They were

anaerobically grown in De Man Rogosa and Shape

(MRS, Merck) at 45 ∞C. All strains of S. aureus

isolated from mastitis cow milk from the National

Institute of Animal Health (Thailand) and

pathogenic bacteria used as indicator strains were

grown in Nutrient broth at 37 ∞C while LAB was

done in MRS at 37 ∞C. E. coli DH5a used as a host

for cloning PCR-amplified 16S rDNA was grown

in Luria broth at 37 ∞C. All strains were stored at

-80 ∞C in desirable media containing 20% (v/v)

glycerol, and propagated twice before experimental

use.

Screening of antimicrobial substance producingLAB

Bacterial cells were removed from the

overnight cultures by centrifugation at 10,956 g

for 15 min at 4∞C. Cell-free culture supernatant

(CFS) obtained was filter-sterilized with a 0.2 mm

pore size filter membrane and further assayed for

antimicrobial activity.

Determination of antimicrobial activityThe spot-on-lawn method was used to

determine antimicrobial activity. The desired

medium agar plates were overlaid with 5 ml of soft

medium agar containing 50 ml of an overnight

culture of the indicator strain. 10 ml of sample was

spotted onto the overlaid surface. The plates were

incubated for 6 and 12 h at 37∞C and determined

for antimicrobial substance (AMS) production.

The AMS titre was determined by the serial twofold

dilution method previously described by Mayr-

Harting et al. (1972). The activity was defined as

the reciprocal of the dilution after the last serial

dilution giving a zone of inhibition and expressed

as activity units (AU) per milliliter.

Identification of the LABSixteen-hour culture solution was

determined for morphology, gram stain, catalase

reaction and motility according to Axelsson (1998).

Growth of LAB in MRS broth at the temperatures

of 10, 45, 50 and 55 ∞C and NaCl concentration of

6% and 18% were determined by turbidity

measurement at 600 nm. CO2 production from

glucose was examined by the method of Gerhardt

et al. (1981). Carbohydrate fermentation patterns

were determined by API CHL50 tests (API,

BioMeriuex, France) as described by the

manufacturer. 16S rRNA analysis was done by the

modified method of Nitisinprasert et al. (2000).

Primer 1407B (5’-GACGGGCGGTGTGTAC-3’)

and 8UA (5’-AGAGTTTGATCCTGGCTC AG-

3’) were used as forward and reverse primers,

respectively. Amplification was carried out in a

thermo-cycler (Hybaid, UK) with annealing for 2

min at 55 ∞C. The purified DNA fragment was

ligated into pGEM-T® Easy vector (Promega,



USA) and then transformed into E. coli DH5acompetent cells. The presence of the PCR product

ligated to the plasmid was confirmed by whole cell

PCR reaction (Nitisinprasert et al., 2000) and

restriction analysis with EcoRI under standard

condition. Nucleotide sequences were determined

by Bioservice unit (BSU, Thailand). The sequences

of 16S rDNA clone were aligned with data from

GenBank by BLAST analysis.

Optimization of growth and AMS productionTo determine the effect of temperature and

pH on growth and AMS production, the culture

solution in MRS medium was incubated at the

temperatures of 30, 37, 45, 50 and 55∞C, and the

pH of 4, 5, 6, 7 and 8. Culture solution was sampled

every 2 h until 24 h and then determined the values

of turbidity at 600 nm, viable cell numbers, pH and

AMS activity against S. aureus TM67.

Preparation of partial purified AMS (ppAMS)Partially purified AMS (ppAMS) was

prepared by vacuum evaporator and cation

exchange chromatography. The CFS was

concentrated to 10-fold by vacuum evaporator.

One mililiter of 10-fold concentrated CFS was

loaded onto a Sephadex fast flow column

equilibrated with 20 mM sodium citrate buffer

(pH 3.0). After initial washing with 20 mM sodium

citrate buffer (pH 3.0), AMS was eluted with 1 M

NaCl in the same buffer at a flow rate of 1 ml/min.

The ppAMS obtained was free from lactic acid and

acetic acid tested by HPLC with an Aminex HPX-

87H cation exchanger. The acids were eluted with

8 mM sulphuric acid at the flow rate of 0.5 ml/min

and tartaric acid was used as internal standard.

Mode of actionTo determine the mode of action of CFS

and ppAMS, 1% (v/v) of the overnight culture

solution of S. aureus was inoculated into NB and

cultivated at 37∞C until the cells reached

logarithmic phase. Then, CFS at the concentrations

of 100, 200 and 300 AU/ml was added into the

culture broth and further cultivated at 37∞C. For

control, equal amount of sterile distilled water was

added instead of CFS. Viable cell numbers and

turbidity at 600 nm were monitored at 1-h intervals.

For ppAMS, 100 AU/ml of ppAMS was

directly added into 102 CFU/ml of S. aureus and

the number of viable cell was determined at 1-h

intervals. twenty mM sodium citrate buffer (pH

3.0) used instead of ppAMS was performed as a

control.

Structural determination by enzyme treatmentA ppAMS was treated with various enzymes

including pepsin, trypsin, proteinase K, protease,

a-chymotrypsin, papain, a-amylase and lipase.

Each enzyme was added to the ppAMS (100 AU/

ml) to obtain a final concentration of 1 mg/ml. The

mixture was incubated at the optimum temperature

of each enzyme for 18 h. At the end of incubation,

the residual activity of AMS was assayed.

RESULTS

Screening of AMS-producing lactic acidbacteria

Fifteen LAB strains isolated from silage

and fermented pork were screened for AMS

production as shown in Table 1. All CFS exhibited

higher inhibition number activity than CFS adjusted

pH to 5 and 6. It indicated a synergistic effect

between antimicrobial substance and organic acids.

The isolate designed as LS2-30 displayed the

highest inhibition number against 9 S. aureus

strains at different conditions of CFS, CFS adjusted

to pH 5 (CFS-5) and CFS adjusted to pH 6 (CFS-

6). Moreover, the AMS of LS2-30 was active

against food-borne and pathogenic bacteria not

only gram positive, S. aureus and Bacillus sp., but

also gram negative such as E. coli as shown in

Table 2. The inhibition activity against those

strains appeared at the range of 100 – 300 AU/ml

(data not shown). The highest activity obtained


was against S. aureus, Salmonella sp., Sal.

Typhimurium, B. cereus, and B. subtilis.

Interestingly, its AMS was not active against all

LAB, M. luteus, and L. innocua.

Identification of the strain LS2-30Morphological and biochemical tests were

used to identify this isolate. LS2-30 was gram-

positive, coccus-shaped, tetrad formed, without

catalase activity, non-motile and no spore forming.

It grew at high temperatures of 45, 50 and 55∞C,

and did not produce gas indicated as

homofermentative LAB. It also grew in MRS

broth with the pH adjusted to 4.5 and 9.6, and NaCl

concentration to 6% but not to 18%. Based on

Bergey’s Manual of systematic of bacteriology,

this isolate was tentatively assigned to the genus

Pediococcus. For species determination,

carbohydrate fermentation using API 50 CHL was

performed and analyzed by APILAB program.

The results showed that LS2-30 was similar to P.

pentosaceus at 92.8%. However, the difference in

maltose fermentation was found as shown in Table

3. Therefore, 1,500 bp of the 16S rRNA of LS2-30

was analyzed and the result showed 97% identity

to P. acidilactici B1104 (accession no. AJ

305322.1).

Optimization of growth and AMS productionThe effects of temperature and the initial

pH of medium on growth and AMS production

were monitored at various conditions. The

temperature effect was performed at 30, 37, 45, 50

and 55∞C as shown in Figure 1. Similar growth

profile was observed at 37 and 45∞C, however, the

maximum activity of 300 AU/ml obtained at 45 ∞C was greater and even maintained for longer time

of 24 h. At high temperature of 50 ∞C, both growth

and AMS activity reached the maximum after

18 h and then dropped dramatically while the one

of 55 ∞C was clearly inhibited after 6 h. It seems

that AMS produced after 6 h actively killed all

viable target cells at high temperature of 55 ∞C.

The initial pH of 4, 5, 6, 7 and 8 displayed

a significant effect on AMS production but not on

its growth as shown in Figure 2. At initial pH of 4,

5 and 6, AMS production started immediately

after growth began and reached a maximum of

200-300 AU/ml during stationary phase. The

maximum of AMS activity was observed after 8 h

of cultivation at initial pH of 6. At the higher pH of

7, only low activity of 100 AU/ml was detected

after 12 h while there was no AMS production

found at the initial pH of 8.

Mode of actionTo determine whether the AMS had

bactericidal or bacteriostatic effect on the sensitive

strains, the AMS treated condition of 100, 200 and

300 AU/ml and a control (without AMS) were

performed as shown in Figure 3. Comparing to the

control, the number of viable indicator cell from

Table 1 The antimicrobial activity of LAB 15

isolates against S. aureus 9 strains.

LAB %Inhibition number1/

CFS CFS-5 CFS-6

CS1-10 100 0 22.2

CS1-11 100 55.5 0

CS4-10 100 66.7 0

CS4-11 100 22.2 11.1

LG1-1 100 11.1 0

LG1-2 100 55.5 0

LG2-1 100 0 0

LS2-30 100 77.8 55.6

LS2-31 100 22.2 0

LS2-32 100 44.4 11.1

N1-9 100 44.4 0

SG1-1 100 22.2 0

SG1-2 100 11.1 0

SK1-3 100 55.6 11.1

SK1-4 100 11.1 0

1/ % Inhibition number = the number of inhibited S. aureus

/ total S. aureus tested


Table 2 The antibacterial spectrum of LS2-30.

Indicator strain Strain tested/inhibited

Staphylococcus aureus 9/9

Escherichia coli 13/13

Salmonella sp. 3/3

Salmonella Typhimurium TISTR 292 (ATCC 13311) 1/1

Bacillus cereus (isolated from silage) 1/1

Bacillus subtilis TISTR 025 1/1

Bacillus coagulans JCM2257 1/1

Bacillus subtilis JCM1465 1/1

Bacillus circulans JCM2504 1/0

Micrococcus luteus IFO12708 1/0

Listeria innocua (ATCC33090) 1/0

Pediococcus pentosaceus JCM5885 1/0

Enterococcus faecalis JCM5803 1/0

Lactococcus lactis ssp. lactis (ATCC19435) 1/0

Lactococcus lactis ssp. cremoris TUA1344L 1/0

Lactobacillus plantarum (ATCC14917) 1/0

Lactobacillus sakei subsp. sakei JCM1157 1/0

Leuconostoc mesenteroides ssp. mesenteroides JCM6124 1/0

the treated condition decreased immediately while

the optical density of cell suspension still remained.

This indicated that CFS of LS2-30 was bactericidal

without cell lysis.

For ppAMS (Figure 4), the viable cell of

2.5 log CFU/ml decreased to less than 10 cfu/ml

after 1 h incubation while the control declined

slowly. This confirmed bactericidal effect against

S. aureus TM67 while low pH of 3 displayed slight

inhibition of its growth.

Determination of compound structure byenzyme treatment

The structure of ppAMS produced from

LS2-30 was examined by various enzymes as

shown in Table 4. The antimicrobial activity of

ppAMS was not affected after treatment with

proteinase K, protease type XIII, papain, pepsin A,

trypsin type I, a-chymotrypsin type II, a-amylase

type X-A and lipase indicating that the ppAMS

was neither proteinaceous, starch nor lipid moiety

in nature.

DISCUSSION

Isolate LS2-30, obtained from silage and

identified as P. acidilactici, was selected to study

as an AMS producing strain. Recently, a number

of antagonistic substances from P. acidilactici

were reported, for example, lactic acid (Stiles,

1996), pediocin PA-1, pediocin AcH, pediocin

JD, pediocin SJ-1 (Schved et al., 1994), pediocin

5, pediocin L50, pediocin F (Osmanagaoglu et al.,

1998), pediocin AcM (Elegado et al., 1997). They

all, except lactic acid, had proteinaceous structure.

However, the AMS of LS2-30 seems to be different

from the results of various enzyme treatments, it

didn’t show the structure of either protein or

complex compounds of protein, starch or lipid

belonging to bacteriocin group IV. Therefore, it


Table 3 Carbohydrate fermentation pattern of

LS2-30 compared with P. pentosaceus.

Carbohydrate P. pentosaceus LS2-30

Glycerol ± -

Arabinose + +

Ribose + +

Xylose ± +

Galactose + +

Glucose + +

Fructose + +

Mannose + +

Rhamnose ± ?

Mannitol - -

Sorbitol - -

Amygdalin + +

Salicin + +

Cellubiose + +

Maltose + +

Lactose + ?

Melibiose ± -

Trehalose + +

Inulin ± -

Melezitose - -

Raffinose ± -

+ = positive; - = negative; ? = doubtful

closely related species (Klaenhammer, 1993),

therefore, it was possible that the AMS found was

not bacteriocin which correspond to above result

of structure determination. This was the first report

of non-bacteriocin substance excluding organic

acid produced by homofermentative lactic acid

bacteria. This property is an advantage to use this

compound as a biological preservative in the dairy

product and fermented food without any effect to

LAB used as a starter culture.

The AMS from this isolate were secreted

into culture medium at the beginning of the

logarithmic growth phase and reached the

maximum level during the early stationary phase

at 45∞C. Thus, the production of AMS from LS2-

30 was growth-associated. The same pattern

occurred with most of active compound produced

by other LAB, such as carnocin U149, leucocin S

and plancitaricin S. The initial optimum pH of

AMS production was pH 6 which was similar to

the previous reports (Mortvedt-Abildgaard et al.,

1995). In addition, this strain clearly showed that

the inhibitory activity remained stable although

the pH was down to pH 4 after prolonging

incubation, suggesting that the antimicrobial

substance was unaffected under acidic conditions.

This could be an important asset allowing its use

under acidic conditions such as fermented food.

Table 4 Factors affecting antimicrobial activity

of ppAMS.

Enzyme Activity (AU/ml)

Proteinase K 100

Protease type XIII 100

Papain 100

Pepsin A 100

trypsin type I 100

a-Chymotrypsin type II 100

a-Amylase type X-A 100

Lipase 100

Control 100

could be either tolerate to enzyme or new compound

structure.

Considering the inhibition ability, LS2-30

has shown the ability to secrete broad spectrum

AMS against different non-related genera,

including food-borne pathogens such as S. aureus,

Bacillus sp., Salmonella sp., and E. coli.

Interestingly, no inhibitory activity of LS2-30 was

observed against LAB tested. Most bacteriocins

produced by P. acidilactici displayed inhibitory

effect to both LAB and some pathogenic

microorganisms for example nisin, pediocin AcH

(Kalchayanand et al., 1992), salivacin 140 and

plantaricin 35d (Messi et al., 2001). Since the

definition of bacteriocin indicated that it will inhibit


Time (h)0 2 4 6 8 10 12 14 16 18 20 22 24

7

8

9

10

11

pH

0

2

4

6

Act

ivity

(AU

/ml)

0

100

200

300

400

500

600

Log

via

ble

cell

coun

t (C

FU/m

l)

Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

Act

ivity

(A

U/m

l)

0

100

200

300

400

500

600

Time (h)0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

Act

ivity

(A

U/m

l)

0

100

200

300

400

500

600

Time (h)0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

Act

ivity

(A

U/m

l)

0

100

200

300

400

500

600

Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

Act

ivity

(AU

/ml)

0

100

200

300

400

500

600

(a) (b)

(c) (d)

(e)

Figure 1 Influence of temperature on the cell growth (▲), pH (■ ) and antimicrobial substance

production (❏ ) by P. acidilactici LS2-30 at 30 ∞C (a), 37 ∞C (b), 45 ∞C (c), 50 ∞C (d) and

55 ∞C (e).

Both AMS production and cell growth were

observed at the same pH of 6.0 which was similar

pattern to pediocin PD-1 production from P.

damnosus NCFB 1832 (Nel et al., 2001). This

should be beneficial to gain the high cell number

and its products during fermentation.

The decline in number of living cells of S.

aureus recorded after the addition of antimicrobial

substance suggested that the mode of action was

bactericidal activity without cell lysis. Its action

was similar to plantaricin 35d properties as reported

by Messi et al. (2001). When high cell concentration

of target strain ( 108 cfu/ml ) were applied, the CFC

of 100, 200 and 300 AU/ml led to a decrease in

viability of only ca. 50% in 7, 4 and 4 h, respectively.

However, when low cell concentration of target

strain 102 cfu/ml was applied, only low dose of

ppAMS (100 AU/ml) could inhibit its growth.

This corresponded to the work done by Boucabeille

et al. (1997). It was found that linenscin OC2 of

10,520 AU/ml exhibited bactericidal against 108

cfu/ml of L. innocua while low dose of 850 AU/ml

did only bacteriostatic action. Therefore, the

inhibition efficiency would depend on the


Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

8

Act

ivity

(AU

/ml)

0

100

200

300

400

500

600

Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

8

Act

ivity

(A

U/m

l)

0

100

200

300

400

500

600

Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

8

Act

ivity

(AU

/ml)

0

100

200

300

400

500

600

Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

8

Act

ivity

(A

U/m

l)

0

100

200

300

400

500

600

Time (h)

0 2 4 6 8 10 12 14 16 18 20 22 24

Log

via

ble

cell

coun

t (C

FU/m

l)

7

8

9

10

11

pH

0

2

4

6

8

Act

ivity

(AU

/ml)

0

100

200

300

400

500

600

(a) (b)

(c) (d)

(e)

Figure 2 Influence of the initial pH of MRS broth on the cell growth (▲), pH (■ ) and antimicrobial

substance production (❏ ) by P. acidilactici LS2-30. pH of 4 (a), pH 5 (b), pH 6 (c), pH 7 (d)

and pH 8 (e).

concentration of antimicrobial substance and the

cell concentration of the indicator strain as also

supported by de Vuyst and Vandamme (1994).

The ambiguous structure of ppAMS still

remained. Further study on purification and

characterization will explain the inhibition

mechanism for application uses in food, feed and

pharmaceutical industries.

CONCLUSION

Fifteen LAB isolates have been screened

according to AMS production against 9

Staphylococcus sp. strain. One of them designed

as LS2-30 was classified as Pediococcus

acidilactici by morphological, physical and

biochemical tests, and 16S rRNA. The optimum

growth temperature and initial pH were 45 ∞C and

pH 6, respectively. CFC of LS2-30 showed an

inhibitory action against some pathogens and


Time (h)

0

Log

via

ble

coun

t

(cfu

/ml)

8

9

10

X X

X

X X X

0

0.5

1

1.5

2

0 2 4 6 8

Time (h)

OD

600

1 2 3 4 5 6 7 8 1 3 5 7

Time (h)

0

Log

via

ble

coun

t(c

fu/m

l)

1

2

3

1 2 3

Figure 3 Effect of LS2-30 CFS on the viability (a) and absorbance at 600 nm (b) of S. aureus. CFS was

added (arrow) at a final dose of 100 AU/ml (■ ), 200 AU/ml (▲), 300 AU/ml (X) and control

(◆ ).

Figure 4 Effect of ppAMS of LS2-30 on the

viability of S. aureus at 102 cfu/ml.

ppAMS was added at 100 AU/ml (▲)

and 20 mM sodium citrate buffer (pH

3.0) as control (◆ ).

foodborne bacteria such as S. aureus, Salmonella

sp., Bacillus sp., and E. coli . However, it didn’t

exhibit activity against all LAB tested. When

mode of action was studied, the CFS displayed a

bactericidal mode of action without cell lysis. The

ppAMS at pH 3.0 showed inhibitory activity as

bactericidal action at low concentration of 100

AU/ml against 102 CFU/ml of target strain. By

various hydrolytic enzymes treatment, it suggested

that the AMS was neither proteinaceous, starch

nor lipid moiety in nature. Therefore, the nature of

inhibitory compound secreted by P. acidilactici

LS2-30 is still unclear.

ACKNOWLEDGEMENTS

This work was supported in part by a

research grant from Graduated School, Kasetsart

University. The authors gratefully acknowledge

National Institutes of Animal Health, Ministry of

Agriculture and Cooperatives, for S. aureus used

as indicator strains.

LITERATURE CITED

Axelsson, L. 1998. Lactic acid bacteria:classification and physiology, pp. 1-72. In S.Salminen and A Von Wright (eds). LacticAcid Bacteria: Microbiology andFunctional Aspect. Marcel Dekker, Inc.,

New York.Bernet-Camard, M.F, V. Lievin, D. Brassart, J.R.

Neeser, A.L. Servin and S. Hudault. 1997.The human Lactobacillus acidophilus strainLA1 secretes a nonbacteriocin antimicrobialsubstance(s) active in vitro and in vivo. Appl.Environ. Microbiol. 63: 2747-2753.

Boucabeille, C., D. Mengin-Lecreulx, G. Henkes,J.M. Simonet and J. van Heijenoort. 1997.Antibacterial and hemolytic activities oflinenscin OC2, a hydrophobic substanceproduced by Brevibacterium linens OC2.FEMS Microbiol. Lett. 153: 295-301.

Chen, Y.L. and C.C. Chou. 2005. Factors affecting


the susceptibility of Staphylococcus aureusCCRC 12657 to water soluble lactose chitosanderivative. Food Microbiol. 22: 29–35.

Diarra, M.S., D. Petitclerc and P. Lacasse. 2002.Effect of lactoferrin in combination withpenicillin on the morphology and thephysiology of Staphylococcus aureus isolatedfrom bovine mastitis. J. Dairy Sci. 85:1141–1149.

Elegado, F.B., W.J. Kim and D.Y. Kwon. 1997.Rapid purification, partial characterization,and antimicrobial spectrum of the bacteriocin,Pediocin ACM, from Pediococcus acidilacticiM. Int. J. Food Microbiol. 37: 1-11.

Gerhardt, P., R.G.E. Murray, R.N. Costilow, E.W.Nester, W.A. Wood, N.R. Krieg, and G.B.Phillips. 1981. Manual of Methods forGeneral Bacteriology. Washington:American Society for Microbiology. P. 415.

Holzapfel, W.H., R. Geisen and U. Schillinger.1995. Biological preservation of foods withreference to protective cultures, bacteriocinsand food-grade enzymes. Int. J. FoodMicrobiol. 24: 343–62.

Hu, S., C. Concha, F. Lin and K.P. Waller. 2003.Adjuvant effect of ginseng extract on theimmune responses to immunization againstStaphylococcus aureus in dairy cattle. Vet.Immunol. 91: 29-37.

Kalchayanand, N., M.B. Hanilin and B. Ray. 1992.Sublethal injury makes Gram-negativebacteria sensitive to the bacteriocin, pediocinAcH and nisin. Lett. Appl. Microbiol. 15:239-243.

Klaenhammer, T.R. 1993. Genetics of bacteriocinsproduced by lactic acid bacteria. FEMSMicrobiol. Rev. 12: 39-86.

Leitnera, G., E. Lubashevskyb and Z. Traininb.2003. Staphylococcus aureus vaccine againstmastitis in dairy cows, composition andevaluation of its immunogenicity in a mousemodel. Vet. Immunol. Immunopathol. 93:159–167.

Mayr-Harting, A., A.J. Hedges and R.C.W.Berkeley. 1972. Methods for studyingbacteriocins, pp. 313-342. In J.R. Norris andD.W. Ribbon (eds.) Methods inMicrobiology (7a). Academic Press, NewYork.

Messi, P., M. Bondi, C. Sabia, R. Battini and G.Manicardi. 2001. Detection and preliminarycharacterization of a bacteriocin (plantaricin35d) produced by a Lactobacillus plantarumstrain. Int. J. Food Microbiol. 64: 193-198.

Mortvedt-Abildgaard, C.I., J. Nissen-Meyer, B.Jelle, B. Grenov, M. Skaugen and I.F Nes.1995. Production and pH-dependentbactericidal activity of lactocin S, a lantibioticfrom Lactobacillus sake L45. Appl. Environ.Microbiol. 61: 175-179.

Nitisinprasert, S., V. Nilphai, P. Bunyun, P. Sukyai,K. Doi and K. Sonomoto. 2000. Screeningand identification of effective thermotolerantlactic acid bacteria producing antimicrobialactivity against Escherichia coli andSalmonella sp. resistant to antibiotics.Kasetsart J. 34: 387-400.

Nel, H.A., R. Bauer, E.J. Vandamme and L.M.T.Dicks. 2001. Growth optimization ofPediococcus damnosus NCFB 1832 and theinfluence of pH and nutrients on the productionof pediocin PD-1. J. Appl. Microbiol. 91:1131-1138.

Osmanagaoglu, O., U. Gunduz, Y. Beyatli and C.Cokmus. 1998. Purification andcharacterization of pediocin F, a bacteriocinproduced by Pediococcus acidilactici F. Tr.J. Biol. 22: 217-228.

Rubin, R.J., C.A. Harrington, A. Poon, K. Dietrich,J.A. Greene and A. Moiduddin. 1999. Theeconomic impact of Staphylococcus aureusinfection in New York City hospitals. Emerg.Infect. Dis. 5: 9 –17.

Schved, F., Y. Henis and B.J. Juven. 1994.Response of spheroplasts and chelator-permeabilized cells of Gram-negative bacteriato the action of the bacteriocins pediocin SJ-1 and nisin. Inter. J. Food Microbiol. 21: 305-314.

Stiles, M. E. 1996. Biopreservation by lactic acidbacteria. Antonie Leeuwenhoek. 70: 331-345.

de Vuyst, L. and E.J. Vandamme. 1994.Bacteriocin of Lactic Acid Bacteria:Microbiology, Genetics and ApplicationsBlackie Academic and Professional, London,536 p.


Studies on Nham-Pla’s Processing by Using Rock Saltand Solar Salt

Mathana Sangjindavong, Pranisa Chuapoehuk and Daungdoen Vareevanich

ABSTRACT

Nham-Pla (fermented fish cakes) were made from striped snake-head fish (Channa striata), Nile

tilapia (Oreochromis niloticus), striped catfish (Pangasius hypophthalmus), and hybrid catfish (Clarias

macrocephalus x Clarias gariepinus). The results of the organoleptic test of uncooked Nham-Pla

prepared from rock salt showed differences (p<0.05) in appearance, texture, odor and average acceptability

scores with the exception for dark meat flesh from hybrid catfish which received the lowest acceptability

score of color. The organoleptic tests of cooked Nham-Pla prepared by using rock salt revealed that

Nham-Pla made from striped catfish received the highest acceptability score for the appearance

characteristic and there were no differences in odor, taste and average acceptability score for both samples

prepared from rock salt and solar salt. Nham-Pla made from Nile tilapia with solar salt received the lowest

acceptability score, but they received the highest acceptability score for texture. Nham-Pla made from

snake-head fish using solar salt received better acceptability scores than Nham-Pla prepared by using

rock salt. According to the results, rock salt and solar salt were important for Nham-Pla preparation based

on species of fresh water fish and some of their characteristics.

Key words: Nham-Pla, rock salt, solar salt

INTRODUCTION

Nham-Pla or sour fish cake is one of

fermented fishery products processed from mostly

fresh-water fish. Nham-Pla made from Notopterus

chitala are found in the market but the shapes of

the product and the ingredients were the same as

Som-Fug. Usually Som-Fug is made from minced

fresh water fish and the ingredients are salt, minced

garlic and cooked rice, but for Nham-Pla some

sliced pork skin are added the same as Nham-Moo

and carrot is used for decorating the products.

Because there were a lot of reservoirs in the country,

the people who live around that area have the great

chance to use some common fresh-water fish such

Department of Fishery Products, Faculty of Fisheries, Kasetsart University, Bangkok 10900, Thailand.


as Oreochromis niloticus, Pangasius

hypophthalmus, Channa striata, Clarias spp.,

Notopterus chitala and Notopterus notopterus for

Nham-Pla’s processing. Sangjindavong et al.

(2000) made a preliminary study on Nham-Pla

made from fresh water fish and marine fish. Sensory

evaluation tests showed that Nham-Pla made from

the fresh water fish had a higher acceptability

score than Nham-Pla made from marine fish.

Fermented fish products are divided into three

main categories (Amano, 1962): 1) Traditional

products with high salt content which involve

enzymes from fish muscle and intestinal organ.

Fish sauce and fish paste are categorized in this

group. 2) Traditional products with two steps

reaction: the first step involves enzymes from fish

muscle and the second step mainly adds

carbohydrate or microbial to start the final reaction.

Pla-Ra and Pickled fish are samples of this group.

3) Non-traditional products: acid such as

hydrochloric acid (HCI) is added in the processing

of fish sauce products.

Rock salt as a mineral occurs naturally in

the ground. It contains 98-99% sodium chloride

and it has a water insolubility level of about 0.5-

1.5%, being mainly calcium sulphate. Solar salt as

a natural product is obtained mainly through

evaporation of seawater. It contains 85% sodium

chloride and has a water insolubility level of less

than 0.03%. (http://www.lentech. Com/water-

softener-FAQ.htm)

The objective of this study was therefore to

assess the quality characteristics of fermented fish

cake prepared from four common varieties of

freshwater fish by comparing Nham-Pla which

using rock salt and solar salt.


Completely randomized design (CRD)

method with four treatments was adopted for this

research. Four kilograms of striped snake-head

fish (Channa striata), Nile tilapia (Oreochromis

niloticus), striped catfish (Pangasius

hypophthalmus), and hybrid catfish (Clarias

macrocephalus x C. gariepinus) were bought from

the market, cleaned, gutted and filleted, followed

by grinding for 10 minutes using Kitchen Aid.

Small amount of rock salt or solar salt, up to 30

grams, was added to 1,000 grams of fish minced

during the grinding process. All other ingredients

(Table 1) were added after grinding and then the

samples, weighing about 130-150 grams each,

were put into 6”¥9” plastic tubes, wrapped with

rubber bands on both ends, placed on trays and

kept at ambient temperature (28∞C-30∞C) for 4

days. Sensory evaluation, chemical analysis and

microbiological tests were conducted at

fermentation period of 4 days. Flow diagram of

Nham-Pla processing is shown in Figure 1.

Quality examinationSensory evaluationA sensory panel consisting of 10 persons

from the department of Fishery Products was

established. Hedonic scale ranged from 1 to 5 was

used to indicate the degree of acceptability of each

sample (1 = most unacceptable, 5 = most

acceptable). The evaluation criteria was manily

focused on appearance, color, odor, flavor and

texture (Watt et al., 1989). Statistical analyses

Table 1 Ingredients used in Nham-Pla

processing.

Ingredients Total weight (gms)

Fish minced 1,000

Pork skin, sliced 300

Rice, cooked 100

Carrot, sliced 150

Garlic, minced 150

Salt (Rock salt, Solar salt) 30

Chilli 10-20

Raw material

↓Filleting

↓Addition of salt while grinding for 10-15 minutes

↓Addition of other ingredients and mixed

↓Filling in plastic tubes (size 6”x9”) about 130-150 gms each

↓Fermentation

↓at ambient temperature (28 oC - 30 oC) for 4 days

↓Analysis

Figure 1 Flow diagram of Nham-Pla.



were carried out using Randomized completed

block design (RCBD) method. The mean from 3

replicates was compared by the methods of

Duncan’s new multiple range test (DMRT)

(Khuantham, 1996).

Chemical analysisThe value of pH was measured from 1:10

diluted samples (Metrohm 744 pH meter). Lactic

acid, expressed as total acid, was examined from

titrating diluted samples with 0.1 N standard NaOH,

using phenolphthaline as pH indicator and

calculating the lactic acid concentration using the

equation % lactic acid = (ml alkali x normality

alkali ¥ 9)/ weight of samples in grams (Frazier et

al., 1968). NaCl content was determined according

to Volhard (AOAC, 1990). Protein, fat and

moisture content were determined according to

AOAC (1984).

Microbiological analysisTotal viable aerobic counts (TVC) were

performed on Plate count agar (Merck KGaA),

and lactic acid bacteria counted on MRS agar

(Merck KGaA). Plates were incubated at 35-37∞C for 3 days (TVC count) and incubated at 25∞Cfor 3 days (lactic acid bacteria counted). Strains of

lactic acid bacteria were identified according to

Sharpe et al. (1966).

RESULTS

Chemical analysis

Proximate compositions of Nham-Pla are

shown in Table 2. and 3. Sensory evaluation are

shown in Table 7 and 8.

Microbiological analysisMicrobiological studied are shown in Table

4, 5 and 6

DISCUSSION

The effect of rock salt and solar salt on

Nham-Pla depended on the species of fish and salt

characteristics. Flavor of Nham-Pla related to

some lactic acid bacteria which might be found in

salt. Sangjindavong (1982) reported that 4 genera

of bacteria such as Staphylococcus sp., Bacillus

sp. Arthrobacter sp. and Corynebacterium sp.

were isolated from solar salt. Sangjindavong et al.

(2000) studied on Nham-Pla prepared from striped

catfish and Nile tilapia and lactic acid bacteria

were isolated namely :- Leuconostoc

mesenteroides, Lactobacillus plantarum,

Pediococcus damnosus, Lactobacillus leichmanii,

Lactobacillus delbruekii, Lactobacillus brevis and

Lactobacillus acidophilus. For this study, lactic

acid bacteria isolated from Nham-Pla were

Lactobacillus pentosus, Lactobacillus plantarum,

Lactobacillus brevis, Lactobacillus acidophilus

and Pediococcus pentosaceus. Some pathogenic

bacteria such as Staphylococcus aureus,

Clostridium perfringens, Salmonella spp., and

fungi were not detected from Nham-Pla.

Table 2 Chemical composition of Nham-Pla using rock salt.

Kinds of Nham-Pla Moisture pH Lactic acid Salt Protein Fat Ash

(%) (%) (%) (%) (%) (%)

Nile tilapia 75.11 3.81 2.37 1.63 15.58 1.06 2.14

Hybrid catfish 68.02 4.03 1.78 1.72 17.23 2.15 2.49

Striped snake-head fish 72.73 4.06 2.81 1.76 18.17 0.94 2.06

Striped catfish 71.26 4.05 2.17 1.73 16.97 3.25 2.24


Table 4 Total lactic acid bacteria in Nham-Pla.

Kinds of Nham-Pla Total lactic acid bacteria (CFU/g)

Rock salt Solar salt

Nile tilapia >300 >300

Hybrid catfish >300 >300

Striped snake-head fish >300 >300

Striped catfish >300 >300

Table 5 Total viable aerobic counts.

Kinds of Nham-Pla Total viable aerobic counts (CFU/g)

Rock salt Solar salt

Nile tilapia 4.65 ¥ 108 2.90 ¥ 109

Hybrid catfish 5.65 ¥ 108 9.25 ¥ 108

Striped snake-head fish 3.95 ¥ 108 6.10 ¥ 108

Striped catfish 1.64 ¥ 109 3.40 ¥ 108

Table 3 Chemical composition of Nham-Pla using solar salt.

Kinds of Nham-Pla Moisture pH Lactic acid Salt Protein Fat Ash

(%) (%) (%) (%) (%) (%)

Nile tilapia 75.58 4.06 2.48 1.59 17.12 1.07 2.33

Hybrid catfish 73.49 4.09 2.58 1.70 16.62 2.19 2.56

Striped snake-head fish 73.67 4.10 2.68 1.67 17.22 1.03 2.26

Striped catfish 73.16 4.09 2.32 1.8 17.52 3.01 2.35

Table 6 Coliforms, faecal coliforms and Escherichia coli in Nham-Pla.

Kinds of Nham-Pla Coliforms Faecal coliforms Escherichia coli

(MPN/gm) (MPN/gm) (MPN/gm)

Rock salt Solar salt Rock salt Solar salt Rock salt Solar salt

Nile tilapia >1100 >1100 <3 <3 <3 <3

Hybrid catfish >1100 >1100 <3 <3 <3 <3

Striped snake-head fish >1100 >1100 <3 <3 <3 <3

Striped catfish>1100 >1100 <3 <3 <3 <3


Table 7 Comparative sensory scores of uncooked Nham-Pla using rock salt and solar salt, fermented

at ambient temperature for 4 days.

Sensory properties

Kinds of Nham-Pla Appearance Texture Color Odor Average

acceptability

1. Striped snake-head fish 4.21a ± 0.11 3.00b ± 1.18 4.36a ± 0.11 3.93a ± 0.83 3.57abc ± 1.22

2. Striped catfish 4.04a ± 0.50 3.71a ± 0.83 3.82ab ± 0.54 3.68a ± 0.77 4.14a ± 0.53

3. Nile tilapia 3.43ab ± 0.65 3.57ab ± 0.65 3.82ab ± 0.88 2.96b ± 0.97 3.50abc ± 0.52

4. Hybrid catfish 3.54ab ± 1.04 4.07a ± 0.73 3.54b ± 1.05 3.43ab ± 1.16 3.71b ± 0.61

5. Striped snake-head fish 3.71ab ± 0.83 3.43ab ± 0.65 3.57b ± 0.94 3.00b ± 1.11 3.21bc ± 1.05

6. Striped catfish 3.15b ± 0.86 3.46ab ± 0.63 2.71c ± 1.00 2.86b ± 0.86 2.93c ± 0.62

7. Nile tilapia 3.79ab ± 1.05 3.43ab ± 0.76 3.85ab ± 0.95 3.36 ab ± 0.84 3.57abc ± 0.85

8. Hybrid catfish 3.86a ± 0.85 3.36 ab ± 0.84 3.79ab ± 0.89 3.21ab ± 0.89 3.64ab ± 0.93

Means followed by different letter are significantly different (P<0.05)

1-4 Nham-Pla; Using Rock salt

5-8 Nham-Pla; Using Solar salt

Table 8 Comparative sensory scores of cooked Nham-Pla using rock salt and solar salt, fermented at

ambient temperature for 4 days.

Sensory properties

Kinds of Nham-Pla Appearance Texture Color Odor Flavor Average

acceptability

1. Striped snake-head fish 3.93ab± 0.96 2.87c ± 1.25 4.00a± 0.85 3.30ab± 1.19 3.13ab± 1.25 3.30ab± 1.10

2. Striped catfish 4.20a ± 0.41 3.67ab± 0.90 4.00a± 0.53 3.60a ± 0.83 3.80a± 0.94 3.63a± 0.80

3. Nile tilapia 3.90ab ± 0.84 4.00a ± 0.85 3.90a± 0.93 3.40ab± 0.91 3.23ab± 1.24 3.63a± 1.01

4. Hybrid catfish 3.47b ± 0.83 3.67ab± 0.90 3.13b± 0.74 3.60a± 0.74 3.93a± 0.70 3.53a± 0.64

5. Striped snake-head fish 4.00ab ± 0.76 3.53b ± 0.92 3.87a± 0.83 3.32ab± 1.14 3.47 ab± 0.99 3.57a± 0.90

6. Striped catfish 3.73ab ± 0.70 3.20ac± 0.85 3.40ab± 0.83 3.50ab± 1.12 3.33 ab± 0.98 3.30ab± 0.88

7. Nile tilapia 2.66c ± 0.82 3.33bc ± 0.90 2.53c ± 0.74 2.73b ± 0.59 2.80b ± 0.86 2.67b ± 0.72

8. Hybrid catfish 4.00ab± 0.65 3.27bc ± 0.80 3.67ab ± 0.72 3.00ab± 1.13 3.27ab± 0.80 3.40a± 0.74

Means followed by different letter are significantly different (P<0.05)

1-4 Nham-Pla; Using Rock salt

5-8 Nham-Pla; Using Solar salt

CONCLUSIONS

Sensory properties of uncooked Nham-Pla

prepared from striped snake-head fish and striped

catfish using rock salt were highest based on the

criteria of appearance, color and odor at

fermentation period of 4 days while Nham-Pla

made from striped catfish using rock salt gave the

highest average acceptability. Sensory evaluation

of cooked Nham-Pla prepared from striped catfish

and Nile tilapia both using rock salt showed the

highest average acceptability. Nham-Pla made


from hybrid catfish using rock salt and Nham-Pla

made from Nile tilapia using solar salt gave the

lowest acceptability. Nham-Pla made from striped

snaked-head fish using rock salt had better texure

than using rock salt.

LITERATURE CITED

Amano, K. 1962. The influence of fermentation

on the nutritive value of fish with special

reference to fermented fish products of

Southeast Asia, pp. 180-200. In E. Heen and

R. Kruezer (eds.). Fish in Nutrition. Fishing

News (Books), London.

AOAC. 1984. Official Methods of Analysis. 14th

edition. Association of Official Analytical

Chemistry. Washington. D.C. 1141 p.


edition. Association of Official Analytical

Chemistry. Washington. D.C. 1296 p.

Frazier, W. C., E. H. Marth and R. H. Diebel. 1968.

Laboratory Manual for Food Microbiology.

4th edition. Burgess Publishing Companys

Minneapolis, MN. 122 p.

Horie, S. and N. Hinago. 1924. Bacterial flora in

imported solar salt. Bulletin of the JapaneseSociety of Scientific Fisheries 40(10) : 1059-

1062.

Khuantham, A. 1996. Principles of ExperimentalDesigns. Department of Statistics, Faculty of

Science, Kasetsart University, Bangkok. 227

p.

Saisithi, P. 1987. Traditional fermented fish

products with special reference to Thai

Products. Asean Food Journal 3(1) : 3-10.

Sangjindavong, M, P. Chuapoehuk and N.

Raksakulthai. 2000. Quality characteristics

of fermented sour fish cake (Nham-pla).

International Journal of Food Properties3(3) : 407-419.

Sangjindavong, M. 1982. Studies on bacteria

isolated from solar salt selling at Bangkok’s

market. Food. Jan-March : 56-63.

Watt, B. M., G. I. Ylimalei, L. E. Jefferg and L. G.

Elias. 1989. Basic Sensory Methods forFood Evaluation. The International

Development Research Center, Ottawa,

Canada. 160 p.

Water softener FAQ. http://www.lentech.com/

water-softener-FAQ.htm. (2/14/2005)


Product Development of Crocodile Jerky

Sinee Nongtaodum1, Nongnuch Raksakulthai2 and Mayuree Chaiyawat2

ABSTRACT

Crocodile jerky was developed from freshwater crocodile (Crocodylus siamensis) tail meat. The

Ratio Profile Test (RPT) was used to find the most acceptable product. Seasonings (soy sauce, sugar, and

pepper) and processing conditions (drying time and temperature, frying time and temperature) were

varied. The jerky sample with 8 % soy sauce, 6.5 % sugar, 2.5 % pepper and 3 % white sesame seed, dried

at 60∞C for 2 hours and fried at 160∞C for 1 minute received the highest acceptability score (P £ 0.05).

The shear force, L*, a*, b* values and aw of the prepared product were 25.8 N., 41.5, 5.3, 9.4 and 0.63,

respectively. The proximate composition of the fresh meat was 72.3 % moisture, 20.2 % protein, 5.5 %

fat, 1.0 % ash and 0.9 % carbohydrates, and the total viable bacterial count was 1.48 ¥ 106 CFU/g. The

proximate composition of the jerky was 13.9 % moisture, 48.2 % protein, 14.7 % fat, 5.3 % ash and 17.9

% carbohydrates. No microorganism was found in the jerky. The product shelf life was determined

according to the thiobarbituric acid number set at 2.5 mg malonaldehyde/Kg by accelerated temperature

test. The product was stored in aluminum foil laminated plastic (OPP/LDPE/Al/LDPE/OPP) bags at

ambient temperature (30±3∞C) under different conditions. The results showed that shelf life was 7 weeks

when packed under air, 9 weeks under air with moisture absorber, 14 weeks under air with oxygen

absorber and 13 weeks under modified atmosphere of 100 % nitrogen.

Key words: crocodile meat, crocodile jerky

INTRODUCTION

Freshwater, marine and hybrid crocodiles

are farm-raised successfully in Thailand. The most

commonly bred species is the freshwater crocodile

(Crocodylus siamensis). At present, the number of

crocodile farms is increasing. The main purpose of

farming is to produce and export crocodile leather

products. Crocodile meat, especially from the tail

part, is believed to be a nutraceutical and can be

cooked with herbs to cure asthma (Maneenopphol,

1998). It is popular among Chinese, Taiwanese,

and Korean tourists. Furthermore, canned stew,

1 Department of Food Technology, Faculty of Engineering and Industrial Technology, Silpakorn University, Sanamchandra

Palace Campus, Nakorn Pathom 73000, Thailand.2 Department of Fishery Products, Faculty of Fisheries, Kasetsart University, Bangkok 10900, Thailand.


soup and dried crocodile meat have been processed.

In Australia, there are many kinds of crocodile

products such as burgers and jerky (Rattanakorn,

1994). Jerky is a product rather similar to Thai

semi-dried beef or pork. Since it is light-weight

and shelf-stable, making crocodile jerky with a

Thai-style taste and flavor might expand the uses

of crocodile meat. Therefore, the objectives of this

study are as follows:

1. To determine the proximate

composition of fresh crocodile meat and prepared

crocodile jerky.

2. To develop a process for Thai-style

crocodile jerky.

3. To set up the thiobarbituric acid (TBA)

number as a quality index for the prepared crocodile

jerky.

4. To find the shelf life of the prepared

crocodile jerky under different storage conditions

at ambient temperature.


Process development of crocodile jerkyFrozen crocodile meat was sliced into thin

sheets of approximately 3 ¥ 6 ¥ 0.5 cm and

marinated with seasonings at 4-6∞C for 16 hours

before drying. The dried product was fried at 160∞C for 1 minute.

The Ratio Profile Test (RPT) of product

using soy sauce (6.0 %), sugar (4.5 %), ground

pepper (1.5 %) and sesame seeds (3.0 %) as in

semi-dried catfish sticks (Department of Fishery

Products, 2000) was conducted. The marinated

crocodile meat was dried at 60∞C for 3 hours then

fried at 160∞C for 1 minute. Sensory evaluation by

10-trained panelists for color, spice odor, hardness,

toughness, sweetness, saltiness and aftertaste was

carried out.

To develop the most acceptable product,

appropriate amounts of seasoning and spice were

studied by varying one factor at a time. The

variables were soy sauce (8, 8.5 and 9.0 % w/w of

crocodile meat); sugar (5.0, 6.5 and 8.0 % w/w)

and pepper (1.5, 2.0 and 2.5 % w/w). The prepared

samples were sensory evaluated by 30 panelists

for appearance, color, odor, taste, texture and

overall acceptability using a 9-point hedonic scale

(1 = dislike extremely and 9 = like extremely). The

experimental design used was a randomized

completely block design. The scores were

statistically analyzed for ANOVA, and Duncan’s

new multiple range tests were employed for

comparison among sample means.

Optimal drying temperature and time of

product with the highest acceptability score were

determined at 50 and 60∞C for 1.5 and 2 hours. The

experimental design was 2¥2 factorial. The samples

were sensory evaluated as above.

Quality of crocodile jerkyShear force of the prepared crocodile jerky

with the highest acceptability score was measured

with the TA-HD Texture Analyzer. L*, a* and b*

were measured by the Minolta CM-3500. aw was

measured with a Thermoconstanter Novasina TH

200.

Proximate compositions of fresh crocodile

meat and crocodile jerky were analyzed according

to AOAC (1984).

Total viable bacterial and Salmonella counts

were determined according to FDA (1984).

Deterioration at an accelerated temperatureThe prepared crocodile jerky was placed

on a glass plate with a cover and stored at 50∞C.

Samples were taken for the TBA number analysis

(Woods and Aurand, 1977; Shibata and Kinumaki,

1979) and the rancid odor was determined by 5

trained panelists every day. The TBA number at

the point that rancidity could be detected by the

panelists was considered as the end of storage

time.

ShelflifeSamples (20 g each) of prepared crocodile

jerky were packed in 12 ¥ 18 cm aluminum foil

laminated plastic bags (OPP/PE/Al/PE/OPP of

20/25/7/20/30 microns thickness) and stored at

ambient temperature (30±3∞C) under different

conditions namely: air, air with a moisture absorber

bag (silica gel), air with an oxygen absorber bag

(ferric) and modified atmosphere of 100 % nitrogen.

Samples were taken once a week for the

TBA number analysis until TBA number exceeded

the predetermined value. The experimental design

was a split plot with packaging conditions as a

main plot and storage time as a subplot.




Process development of crocodile jerkyRPT results are shown in Table 1. The

ratios of ideal and sample scores on toughness and

hardness were higher than 1.0, but those of spice

odor, sweetness and saltiness were lower than 1.0.

Therefore, drying time and temperature, amount

of spice and seasonings were adjusted in the next

step.

Sensory evaluation of samples prepared

with varying amounts of soy sauce, sugar and

pepper contents are shown in Tables 2-4. It may be

concluded that the appropriate amounts of soy

sauce, sugar and pepper were 8, 6.5 and 2.5% by

weight of the crocodile meat, respectively.

It was found that drying time and

temperature significantly affected the texture of

the product (P £ 0.05). Shear force of products

dried at 50 and 60∞C is shown in Table 5. Sensory

evaluation of samples is shown in Table 6. The

sample dried at 60∞C for 2 hours received the

highest sensory score for texture with the highest

shear force. The appearance, color and overall

acceptability scores were the highest as well.

It was recommended that jerky with the

size of 12.5 ¥ 5 ¥ 0.6 cm should be dried at a

temperature between 60 and 65∞C for at least 4 h

to keep it from spoiling (http://www.alljerky.com/

wwwboard/wwwboard.html).�The size of samples

in this study were smaller, thus the drying time

required at the same drying temperature was only

2 h.

Quality of crocodile jerky

Proximate compositions, total bacterial,

Salmonella counts of fresh crocodile meat, and

prepared crocodile jerky are shown in Table 7.

Mitchell et al. (1995) reported that the average

proximate compositions of fresh meat of

Crocodylus porosus and C. johnstoni were 75.9 %

moisture content, 21.1 % protein, 1.9 % fat and

0.95 % ash while Baek and Cadwallader (1997)

reported the proximate composition of the tail

meat of Alligator mississippiensis to be 29.1 %

protein and 2.9 % fat. The fat content of Crocodylus

siamensis from our study was as high as 5.5 %

which might be due to its feed. However, it could

be observed that the fat layer was separated from

the muscle.

It was found that the process of crocodile

jerky was effectively in reducing the number of

bacteria. The product had a shear force, L*, a* and

b* and aw of 25.8 N, 41.5, 5.3, 9.4 and 0.63,

respectively.

Deterioration at an accelerated temperatureThe TBA number was chosen as a quality

index in this study since it has been generally used

to test rancidity in meat (Green and Cumeze,

1982). In general, the acceptable TBA value in

food is less than 20 mg malonaldehyde/Kg sample

(Shamberger et al., 1977). However, different

types of food have different TBA values for a

threshold of rancid odor, e.g., in cooked ground

pork and beef, the values are 0.5-1.0 and 0.6-2.0

mg malonaldehyde/Kg, respectively (Tarladgis et

al. 1960). For frozen and canned fish, it was

reported as good quality when the TBA number

was less than 3.0, however they were still acceptable

when the number increased to 4-27 mg

malonaldehyde/Kg (Shamberger et al. 1977). For

fishmeal, the initial TBA number was reported at

21, and a strong rancid odor was found at around

300 mg malonaldehyde/Kg (Green and Cumeze,

1982). Rancid odor in semi-dried catfish sticks

could be detected when the TBA number exceeded

2.1 mg malonaldehyde/Kg (Pongchawee, 1994).

The initial TBA number of dried fish was 3.1

(Pigott and Tucker, 1990). The panelists accepted

fried pork sticks if the TBA number was lower

than 3.54 mg malonaldehyde/Kg (Niyomkiatkul,

1986).

It was found that panelists could detect a

rancid odor when the TBA number of the prepared


Table 2 Sensory evaluation scores of crocodile jerky with different soy sauce contents.

Sensory attribute Sensory evaluation score NS

Soy sauce content

8.0 % 8.5 % 9.0 %

Appearance 6.9 6.8 6.8

Color 7.0 6.9 6.8

Odor 6.9 6.8 6.8

Flavor 7.0 6.9 6.8

Texture 6.8 6.8 7.0

Overall acceptability 7.1 6.9 6.8

NS Not significantly different.

Table 1 Ratio Profile Test of crocodile jerky.

Attribute Ideal score (I) Sample score (S) S/I

Color 5.70 6.12 1.08

Spice odor 6.57 5.18 0.78

Toughness 4.40 5.00 1.14

Hardness 4.12 5.73 1.43

Sweetness 5.51 4.32 0.79

Saltiness 5.21 4.73 0.91

Aftertaste 0.50 0.52 1.04

jerky exceeded 2.5 mg malonaldehyde/Kg.

Therefore, this value was used as an index of

quality during the shelf life study.

ShelflifeTBA numbers of crocodile jerky stored

under different conditions are shown in Figure 1.

Packaging conditions and storage time significantly

affected the TBA number (P £ 0.05). The TBA

values of sample stored under air showed a trend

to be higher than those stored under air with

moisture absorber, under modified atmosphere

with nitrogen or under air with oxygen absorber. It

was also found that the TBA values of samples

stored under air with oxygen absorber during the

12, 13 and 14th week were not significantly different

(P > 0.05). According to the predetermined value

of TBA from the accelerated storage condition, the

shelf life of the products packed under air, air with

a moisture absorber, air with an oxygen absorber

and 100 % nitrogen were 7, 9, 14 and 13 weeks,

with the TBA numbers of 2.53, 2.63, 2.52 and 2.55

mg malonaldehyde/Kg samples, respectively. The

crocodile jerky product of Australia with 2 % fat

was reported to have a shelflife of at least 1 year

(www.alljerky.com/jerky_product.html). The

shorter shelflife of our products was resulted from

the higher fat content (14.7 %).

CONCLUSION

Thai-style crocodile jerky can be prepared

from the tail meat of cultured freshwater crocodile

with the acceptability scores of 7-8 (like moderately


Table 3 Sensory evaluation scores of crocodile jerky with different sugar contents.

Sensory attribute Sensory evaluation score1

Sugar content

5.0 % 6.5 % 8.0 %

Appearance 7.1 a 7.2 a 7.3 a

Color 7.1 a 7.2 a 7.2 a

Odor 7.0 a 7.2 a 7.2 a

Flavor 7.2 b 7.5 a 7.4 a

Texture 7.2 a 7.3 a 7.2 a

Overall acceptability 7.2 b 7.5 a 7.4 a

1Values in the same row followed by different letters are significantly different (P £ 0.05).

Table 5 Shear force of fried crocodile jerky dried at 50 or 60∞C for 1.5 or 2.0 hours.

Shear force (Newton)1

Drying temperature Drying time (hours)

50 ∞C 60 ∞C

1.5 12.4 d 14.1 c

2.0 16.0 b 20.4 a

1 Values followed by different letters are significantly different (P £ 0.05).

Table 4 Sensory evaluation scores of crocodile jerky with different pepper contents.

Sensory attribute Sensory evaluation score1

Pepper content

1.5 % 2.0 % 2.5 %

Appearance 7.4 a 7.4 a 7.4 a

Color 7.3 a 7.3 a 7.4 a

Odor 7.3 b 7.3 b 7.6 a

Flavor 7.1 c 7.4 b 7.8 a

Texture 7.1 a 7.1 a 7.2 a

Overall acceptability 7.1 c 7.4 b 7.7 a

1 Values in the same row followed by different letters are significantly different (P £ 0.05).

to like very much). TBA number can be used

effectively as a quality index. Product packed in

OPP/PE/Al/PE/OPP bag with oxygen absorber

could be kept for 14 weeks. However, since

deterioration of crocodile jerky is found to result

from rancidity, the use of vacuum packaging might

be able to extend shelf life of this product for

longer than 14 weeks.


Table 7 Proximate compositions and microbiological quality of crocodile meat and jerky.

Composition and microbiological quality Crocodile meat Crocodile jerky

Moisture (%) 72.3 13.9

Protein (%) 20.2 48.2

Fat (%) 5.5 14.7

Ash (%) 1.0 5.3

Carbohydrate (%) 0.9 17.9

Total bacterial count (CFU/g) 1.48 ¥ 106 N.D.

Salmonella (CFU/g) N.D. N.D.

N.D. not detected.

Table 6 Sensory evaluation scores of fried crocodile jerky dried at different times and temperatures.

Sensory attribute Sensory evaluation score

Drying temperature

50 ∞C 60 ∞C Means

Appearance

Drying time 1.5 h 6.7 6.9 6.8 b

2.0 h 7.0 7.5 7.3 a

Means 6.9 B 7.2 A

Color

Drying time 1.5 h 6.2 6.7 6.5 b

2.0 h 6.8 7.6 7.2 a

Means 6.5 B 7.1 A

Odor

Drying time 1.5 h 7.2 7.2 7.2

2.0 h 7.2 7.3 7.2

Means 7.2 7.2

Flavor

Drying time 1.5 h 7.3 7.4 7.4

2.0 h 7.3 7.5 7.4

Means 7.3 7.4

Texture

Drying time 1.5 h 6.5 6.8 6.6 b

2.0 h 6.9 7.4 7.2 a

Means 6.7 B 7.1 A

Overall acceptability

Drying time 1.5 h 6.9 7.1 7.0 b

2.0 h 7.0 7.3 7.2 a

Means 7.0 B 7.2 A

Means in the same row followed by different letters (A, B) are significantly different (P £ 0.05).

Means in the same column followed by different letters (a, b) are significantly different (P £ 0.05).


LITERATURE CITED


ed.; Association of Analytical Chemists,

Washington D.C. 1141 p.

Baek, H.H., and K.R. Cadwallader. 1997. Aroma

volatile in cooked alligator meat. J. Food Sci.62(2): 321-325

Department of Fishery Products. 2000. FreshwaterFish Processing. Faculty of Fisheries,

Kasetsart University 48 p.

FDA. 1984. Bacteriological Analytical Manualfor Food. 16th ed. Bureau of Food, Division

of Microbiology. Washington D.C. 224 p.

Green, B. E. and T. H. Cumeze. 1982. Relationship

between TBA numbers and inexperienced

panelists’ assessment of oxidized flavor in

cooked beef. J. Food Sci. 47: 52-54, 58.

http://www.alljerky.com/jerky_product.html. 11

Feb. 2005.

h t t p : / / w w w . a l l j e r k y . c o m / w w w b o a r d /

wwwboard.html. 11 Feb. 2005.

Maneenopphol, P. 1998. Basic knowledge in

crocodile farming. Modern Kaset. 8(3):1-

10.

Mitchell, G.E., A.W. Reed and D.B. Houlihan.

1995. Composition of crocodile meat

(Crocodylus porosus and C. johnstoni). FoodAustralia 47:321-325.

Niyomkiatkul, J. 1986. Effect of ButylatedHydroxy Anisole, Butylated HydroxyToluene and Sorbate on Quality of PorkStick. M.S. Thesis. Kasetsart University.

Bangkok.

Pigott, G. M. and B. W. Tucker. 1990. Seafood:Effects of Technology on Nutrition. Marcel

Dekker, Inc. New York. 362 p.

Pongchawee, K. 1994. Production of Semi-driedCatfish Stick and Shelf Life under ModifiedAtmosphere Packaging. M.S. Thesis.

Kasetsart University. Bangkok.

Rattanakorn, P. 1994. Utilization of crocodile.

Crocodile News 3(2): 5.

Shamberger, R.G., B. A. Shamberger and C. E.

Willis. 1977. Malonaldehyde content of food.

J. Nutr. 107: 1404-1409.

Figure 1 Change of TBA number of crocodile jerky stored under different conditions at ambient

temperature.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

TB

A N

umbe

r (m

g m

alon

alde

hyde

/kg)

1

2

3

4air

air + moisture absorber

100 % nitrogen

air + oxygen absorber

Storage time (week)


Shibata, N. and T. Kinumaki. 1979. An

improvement of TBA procedure as the

measure of the oxidative determination

occurring in fish oil: II- Intactsample

procedure. Bull. Jap. Soc. Sci. Fish. 45: 505-

509.

Tarladgis, B.G., B.M. Watt, M.T. Younathan and

L.R. Dugan, Jr. 1960. A distillation method

for quantitative determination of

malonaldehyde in rancid foods. J. Am. Chem.Soc. 37: 44-48.

Woods, A.E. and L.W. Aurand. 1977. LaboratoryManual in Food Chemistry. TheAVI

Publishing Co., Inc., Westport, Connecticut.

72 p.


Utilization of Fish Flour in Canned Concentrated Seasoning Stockfor Thai Foods Preparation

Plernchai Tangkanakul, Payom Auttaviboonkul, Patcharee Tungtrakul, Mantana RuamruxChidchom Hiraga, Kanjanarat Thaveesook and Montatip Yunchalad

ABSTRACT

Herring fish flour possessed high contents of protein, calcium and phosphorus at 64.70 g/100 g,

2,576 mg/100 g and 1,531 mg/100 g, respectively. Herring fish flour was used to replace fresh or

dehydrated fish meat in developing five canned concentrated seasoning stocks for Thai food; Nam Ya Pla

(fish curry sauce), Kaeng Som (spicy sour mixed vegetable), Kaeng Tai Pla (southern hot curry), Kaeng

Kua Fag (red curry with wax gourd), and Kaeng Tae Po (red curry with swamp morning glory). Sensory

test exhibited that a suitable amount of herring fish flour incorporated was 15% or 18% of curry paste,

that accounted for 2.6-5.1% in the recipe. It was determined that 100 g canned stocks provided protein,

calcium and phosphorus as 4.32 – 6.00 g, 126.5 – 136.4 mg and 83.9 – 108.6 mg, respectively. Calcium

content of fish flour incorporated dishes was high, supplying 10-25% of the adult Recommended Daily

Dietary Allowance in a 100 g portion.

Key words: fish powder, Thai curry, Thai dishes, processed foods, calcium, canned concentrated

seasoning stock

INTRODUCTION

Fish and fish products attracted food

technologists in development of varieties of food

due to their marvellous potential and health benefits

(Kinsella, 1986; Kinsella et al., 1990). Fish is an

excellent source of protein and rich in vitamins and

minerals (Guthrie, 1983). In Thailand, it was

reported that over 40 % of animal proteins

consumed were derived from fish, since a wide

spreading in rivers, lakes and rice fields (FAO,

1982). Besides its wide distribution, the price is

relatively lower than other animal proteins.

Surprisingly, in 58 developing countries, more

than 20% of animal protein supply came from fish

(FAO, 1982). It implied that there must be

Institute of Food Research and Product Development, Kasetsart University, Bangkok 10900, Thailand.


thousands of recipes prepared from fishes available

around the world.

Micronutrients in fish are renowned,

however, with different quantities depending on

the type of fishes. For example, cod, a white meat

fish, contains 60 mg sodium, 9 mg calcium and 0.1

mg iron, where as 123 mg, 60 mg and 1.2 mg,

respectively, were found in herring, which

determined from 100 g raw edible portion (Holland

et al., 1993). A gutted herring contained calcium

as high as 231-597 mg per 100 g fresh weight

(Tahvonen et al., 2000). Common fishes used in

Thai cooking are fresh water fishes, such as snake-

head fish, catfish and common silver barb. These

fishes contained calcium in a range of 18–32 mg,

0.6–1.0 mg iron and 189–287 mg phosphorus per

100 g raw edible portion (Institute of Nutrition,

1999). In Thailand, the sources of calcium could

come from dark green vegetables, dried shrimp

and small fishes. A national survey on dietary

intake of mineral in 1995 reported that calcium

consumption per day per person for Thai people

was 344 mg (Department of Health, 1995), while

the recommendation amount is 800 mg (Committee

on Recommended Daily Dietary Allowances,

1989). Thus, if there is herring in a convenient

form available in a market, it could supply as a

good source of calcium in fish-containing Thai

dishes. From health information, it is well-known

that calcium is related to health due to a property

on reducing the risk of developing osteoporosis.

At present, Thai people are concerned more

about food safety and health foods. And food

producers trend to produce easy-to-cook/

convenience products, serving a modern life style.

Hence, the objectives of this study were to evaluate

potential of utilizing herring fish flour into ten

different Thai dishes and to develop five most

accepted dishes into canned concentrated seasoning

stocks.


Herring fish flour was obtained from the

Norwegian Herring Oil and Meal Industry Research

Institute and kept at –18∞C until used. Ten popular

dishes of Thai foods were selected; Kaeng Luang

(yellow curry), Pad Thai (stir fried rice noodles),

Kaeng Pa (spicy curry without coconut milk), Pad

Khee Mao (spicy stir fried noodles), Kaeng Tai Pla

(southern hot curry), Kaeng Kua Fag (red curry

with wax gourd), Nam Ya Pla (fish curry sauce),

Kaeng Liang (spicy vegetable soup), Kaeng Som

(spicy sour mixed vegetable) and Kaeng Tae Po

(red curry with swamp morning glory). The nature

of these ten dishes was possible to add a fish

ingredient. Dish description and their major

ingredients were presented as followed:

Kaeng Luang Soup-type dish; dried chilli,

turmeric root, garlic, shallot,

shrimp paste, tamarind juice,

palm sugar, fish sauce, raw

papaya, fish.

Pad Thai Stirred-fried dish; rice noodle,

fish flour, hard tofu, fresh bean

sprouts, Chinese chive.

Kaeng Pa Soup-type dish; chilli, shallot,

garlic, shrimp paste, lemon

grass, galanga, coriander root,

kaffir lime peel, coriander seed,

cumin, pepper, fish, bamboo

shoot, long-yard bean, baby

corn, brinjal, fingerroot

(krachai), holy basil leaves.

Pad Khee Mao Stirred-fried dish; rice noodle,

pepper, garlic, fish flour, onion,

tomato, baby corn, cabbage, holy

basil leaves.

Kaeng Tai Pla Soup-type dish; chilli, shallot,

garlic, lemon grass, galanga,

kaffir lime peel, turmeric, shrimp

paste, fish, fermented fish

viscera, bamboo shoot, brinjal,

plate brush eggplant, long-yard

bean.

Kaeng Kua Fag Soup-type dish with coconut

milk; chilli, shallot, garlic,

galanga, lemon grass, shrimp

paste, coconut milk, dried fish,

wax gourd, kaffir lime leaves.

Nam Ya Pla Sauce with coconut milk; chilli,

shallot, fingerroot, garlic,

galanga, lemon grass, shrimp

paste, coconut milk, fish.

Kaeng Liang Soup-type dish; shallot, pepper,

shrimp paste, fingerroot, angled

gourd, straw mushroom,

pumpkin, baby corn, ivy gourd

leaves, hairy basil leaves, fish.

Kaeng Som Soup-type dish; dried red chilli,

shallot, fingerroot, shrimp paste,



fish, tamarind juice, Chinese

radish, long-yard bean, Chinese

cabbage.

Kaeng Tae Po Soup-type dish with coconut

milk; dried red chilli, shallot,

garlic, galanga, lemon grass,

pepper, kaffir lime peel,

coriander root, shrimp paste,

coconut milk, dried fish,

tamarind juice, kaffir lime juice,

swamp morning glory, kaffir

lime leaves.

Sensory evaluation of fish flour incorporatedThai dishes

Sensory evaluation was carried out using

25 panellists with a 9-hedonic scale method (9 =

like extremely, 8 = like very much, 7 = like

moderately, 6 = like slightly, 5 = neither like nor

dislike, 4 = dislike slightly, 3 = dislike moderately,

2 = dislike very much, 1 = dislike extremely). Two

sensory tests were carried out. The interested

attributes were colour, consistency, odour, taste,

texture and overall acceptance. The first test was

examined when ten dishes at any four different

percentages of fish flour; 6, 8, 10, 12, 15 or 18% of

curry paste were added. The result was used to

indicate five most acceptant dishes at the best

incorporated amount of fish flour. The second test

was run to evaluate taste preference of five selected

dishes prepared from canned concentrated

seasoning stock and their controls, the traditional

dishes. Five dishes which were prepared from

canned concentrated seasoning stock were mixed

with vegetables and water as shown in Table 1.

Processing concentrated seasoning stocksConcentrated seasoning stocks of five most

satisfied dishes, Nam Ya Pla, Kaeng Som, Kaeng

Tai Pla, Kaeng Kua Fag and Kaeng Tae Po, were

selected for canning. The pHs of the five dishes

were 5.97, 4.37, 5.73, 5.99 and 5.07, respectively.

The ingredients of five fish flour incorporated

concentrated seasoning stock are shown in Table

2. All concentrated stocks were packed in lacquered

tinplate can size No. 2 (307¥409) and processed at

Table 1 Preparation of five Thai dishes from canned concentrated seasoning stocks.

Product Seasoning stock,1 can (g) Water (g) Vegetables (g)

Nam Ya Pla 560 100 -

Kaeng Som 580 120 520

Kaeng Tai Pla 570 150 515

Kaeng Kua Fag 565 150 400

Kaeng Tae Po 580 100 275

Table 2 Ingredients of five canned concentrated seasoning stocks (g/100g total weight).

Product Fish flour Spices Condiment Coconut milk Chicken stock

Nam Ya Pla 4.5 10.0 4.8 80.7 -

Kaeng Som 4.6 10.0 20.0 - 65.4

Kaeng Tai Pla 5.1 11.6 17.4 - 65.9

Kaeng Kua Fag 4.0 8.2 8.5 79.3 -

Kaeng Tae Po 2.6 7.1 18.7 71.5 -


121∞C for 60 min, except Kaeng Som which was

processed for 30 min.

Chemical analysisHerring fish flour, canned concentrated

seasoning stock, foods prepared from concentrated

seasoning stock and its control recipe of five Thai

dishes were analyzed for nutritional compositions.

Proximate compositions; protein, fat, ash

and moisture content were determined by methods

of AOAC (1984). Total dietary fibre was carried

out by enzymatic–gravimetric method (Faulks and

Timms, 1985; Prosky et al., 1985).

The level of potassium, sodium, iron,

copper, zinc, cadmium, lead, calcium and

magnesium were analyzed through Atomic

Absorption Spectrophotometric method followed

AOAC (2000). The phosphorus content was

determined by means of colourimetric method

(AOAC, 1998).

Consumer testTwo hundred and eighty test subjects were

randomly selected for an acceptance test. Three

groups of people: 150 students, 80 teachers and 50

workers, were recruited. Each food prepared from

canned fish flour concentrated seasoning stock

(Table 1) was evaluated for its attributes of colour,

odour, taste and overall acceptability. Verbal scale

with 5 categories as follows: like very much, like,

neither like nor dislike, dislike, dislike very much,

was used for colour, odour and taste. For overall

acceptability, only 3 categories scale was used,

i.e., accepted, neither accepted nor unaccepted and

unaccepted.

Statistical analysisOne way analysis of variance and Duncan’s

multiple range test were applied to the results of

sensory data by using IRRISTAT software package

(IRRISTAT version 90-1). Significance of

differences of five dishes between control and its

herring fish flour incorporated dish was evaluated

using student’s t-test at levels of 0.01 and 0.05.


Nutrients of herring fish flourProximate analysis of herring fish flour is

shown in Table 3. Herring fish flour contained

64.70 g protein/100 g which exhibits a good protein

source. However, not only the protein content but

also the quality of protein is important to health.

Essential amino acid score is generally considered

the most logical method to assess protein quality.

The pattern of essential amino acids in herring fish

flour protein compared to the reference protein

according to FAO/WHO (1973) revealed that all

of essential amino acid contents are greater than

80% of FAO/WHO suggestion (data not shown).

It could be implied that incorporating fish flour

into food products would help improving nutritional

status to combat malnutrition problem. Fat content

was 10.52 g/100 g which was higher than fresh

water fish such as dried snake-head fish, 7.0 g/100

g (Ministry of Public Health, 1987).

Ash content was substantially high which

had a direct correlation with the quantities of the

minerals present (Table 4). Among them calcium

and phosphorus level were substantially high.

Thus, herring fish flour would be useful as a source

of many important minerals. Phithakpol (1984)

studied on calcium and phosphorus contents of

roller dried fish from sardine and threadfin bream.

The results showed that both minerals were

noticeably lower than those in the herring fish

flour (Table 4). The level of harmful heavy metals,

cadmium and lead, in tested herring fish flour was

low.

Sensory evaluation of ten Thai foods utilizingherring fish flour

Ten different Thai dishes which

incorporated with four various percentages of

herring fish flour were studied for sensory

evaluation on six attributes; colour, consistency,


Table 3 Proximate analysis of fish flours, processed foods and foods (per 100 g sample).

Raw material Moisture Protein Fat Ash Dietary fibre CHO* Energy

(g) (g) (g) (g) (g) (g) (Kcal)

Fish flour

Herring 9.31 64.70 10.52 10.81 4.13 0.53 355.60

Sardinea 5.55 75.20 5.26 5.19 0.72** 8.80 -

Threadfin breama 6.59 74.51 7.41 5.15 0.10** 6.34 -

Canned concentrated seasoning stocks

Nam Ya Pla 77.92 4.97 11.23 2.25 3.34 0.29 122.11

Kaeng Som 84.15 4.59 1.41 3.29 1.78 4.78 40.40

Kaeng Tai Pla 83.70 6.00 1.67 4.20 1.52 2.91 50.67

Kaeng Kua Fag 75.36 5.09 12.46 3.10 2.25 1.74 139.46

Kaeng Tae Po 65.65 4.32 13.85 2.70 2.47 11.01 185.97

Foods prepared from canned concentrated seasoning stocks

Nam Ya Pla 76.86 5.49 11.04 2.31 3.89 0.41 122.96

Kaeng Som 87.91 3.13 0.79 2.10 1.66 4.41 37.25

Kaeng Tai Pla 86.77 4.05 1.30 2.24 1.44 4.20 44.70

Kaeng Kua Fag 82.92 3.56 6.05 1.78 2.31 3.38 82.21

Kaeng Tae Po 76.20 3.74 7.47 2.42 2.22 7.95 113.99

Traditional foods (control)

Nam Ya Pla 80.14 3.74 11.23 1.86 2.76 0.27 117.11

Kaeng Som 89.11 2.58 0.54 1.83 1.99 3.95 30.98

Kaeng Tai Pla 87.71 3.68 1.09 2.01 1.39 4.12 41.01

Kaeng Kua Fag 85.34 2.88 5.38 1.62 2.14 2.64 70.50

Kaeng Tae Po 75.92 3.16 7.24 2.29 2.33 9.06 114.04

* carbohydrate, expressed as : 100 – (protein + fat + ash + dietary fibre + moisture)

** crude fibrea source: Phithakpol, 1984

odour, taste, texture and acceptability (Table 5).

Percentages of added herring fish flour in the

selected dishes were varied depending on the

nature of each dish. It was found that there were

many limitations for amount of herring fish flour

adding. Foods with small amount of herbs and

spices could incorporate less fish flour than foods

with high amount of spices, which could be

observed in Pad Thai and Pad Khee Mao. Colour

of most dishes was darkened when greater amount

of herring fish flour was added, lead to lower

acceptability, for example, Pad Thai, Pad Khee

Mao, Nam Ya Pla and Kaeng Som. From the

results, herring fish flour odour was detected in

two dishes of stirred fried noodle, Pad Thai and

Pad Khee Mao. Fish flour odour in soup-type

dishes were not substantially detected as indicated

by less score variation. Texture of both stirred

fried noodle dishes was influenced by high

percentage of fish flour. The effect was

mouthcoating of fish flour particles on the tongue.

For soup–type dishes, sandy mouth feel was

reported when high amount of fish flour was

applied. This effect was less pronounced in the


Table 4 Minerals contents in fish flours, processed foods and foods (mg per 100 g sample).

Raw material Na Ca P Fe Cu Pb K Mg Zn Cd

Fish flour

Herring 706.9 2,576 1,531 10.66 1.69 0.0064 904.4 140.25 11.36 0.03

Sardine a n/d 970 708 5.66 0.80 0.20 n/d n/d 4.52 0.05

Threadfin bream a n/d 514 692 1.59 0.40 0.21 n/d n/d 3.50 0.02

Canned concentrated seasoning stocks

Nam Ya Pla 506.6 126.5 102.4 1.21 0.11 n/d n/d n/d n/d n/d

Kaeng Som 861.2 135.1 83.9 0.86 0.08 n/d n/d n/d n/d n/d

Kaeng Tai Pla 1,000.6 136.4 92.9 1.82 0.10 n/d n/d n/d n/d n/d

Kaeng Kua Fag 734.6 129.8 108.6 1.04 0.17 n/d n/d n/d n/d n/d

Kaeng Tae Po 591.8 135.3 87.1 1.00 0.21 n/d n/d n/d n/d n/d

Foods prepared from canned concentrated seasoning stocks

Nam Ya Pla 1,055.0 204.8 175.8 1.83 0.20 n/d n/d n/d n/d n/d




Kaeng Tae Po 1,112.0 119.3 144.2 2.31 0.34 n/d n/d n/d n/d n/d

Traditional foods (control)

Nam Ya Pla 1,008.6 106.6 138.3 1.46 0.14 n/d n/d n/d n/d n/d




Kaeng Tae Po 996.6 47.6 68.7 1.44 0.31 n/d n/d n/d n/d n/d

a source: Phithakpol, 1984

n/d = not determined

dishes with coconut milk.

Overall acceptability was evaluated. And

dishes from top five at the highest incorporated

percentage (Table 5) were selected for further

canning process of concentrated seasoning stock.

The result of selected dishes with fish flour content

were Nam Ya Pla, 18% fish flour; Kaeng Tae Po,

15% fish flour; Kaeng Tai Pla, 15% fish flour;

Kaeng Kua Fag, 15% fish flour and Kaeng Som,

15% fish flour.

It was found that herring flour applied in

this study did not have any critical characteristic

on taste, odour or consistency of the dishes. A

similar research work was done in 1981 by

Phithakpol et al., demonstrated that the

incorporated roller-dried sardine and threadfin

bream had more effect on taste and odour than on

colour and consistency.

Chemical compositionsCanned concentrated seasoning stock

Nutrient compositions of the five canned

concentrated seasoning stock at zero month are

exhibited in Table 3. Protein content of all products

fell in the range of 4.32 – 6.00 g/100 g. Kaeng Tai

Pla contained the highest protein content. When


Table 5 Sensory evaluation of fish flour incorporated in 10 kinds of Thai dishes.

% Fish flour in curry paste Colour Consistency Odour Taste Texture Acceptability

Kaeng Luang8% 7.34a 6.86a 7.05a 7.02a 6.86a 6.95a

10% 7.14ab 6.50a 6.79ab 6.64a 6.76a 6.64a

12% 7.24ab 6.86a 6.98ab 6.79a 6.88a 6.81a

15% 6.95b 6.71a 6.68b 6.62a 6.52a 6.57a

Pad Thai6% 7.53a 6.78a 7.20a 7.32a 7.22a 7.32a

8% 7.40a 6.70a 7.05a 7.10a 6.85a 7.03a

10% 6.32b 5.88b 6.38b 6.65b 6.07b 6.30b

12% 5.60c 5.55b 6.00b 6.30b 5.65b 5.50c

Kaeng Pa8% 7.35a 7.10a 7.28a 7.35a 7.20a 7.34a

10% 7.40a 6.95a 7.03a 7.20ab 6.90a 7.05a

12% 7.38a 7.07a 7.03a 7.35a 7.00a 7.20a

15% 7.25a 6.75a 6.60a 6.93b 6.20b 6.43b

Pad Khee Mao6% 7.55a 7.34a 7.26a 7.24a 7.34a 7.24a

8% 7.24a 7.03a 6.97a 7.16a 6.71b 6.76a

10% 6.68b 6.03b 6.42b 6.63b 5.66c 5.74b

12% 6.13c 5.68b 5.95b 6.24b 5.18c 5.32b

Kaeng Tai Pla8% 7.58ab 7.53a 7.53a 7.74a 7.63ab 7.61a

10% 7.76a 7.71a 7.66a 7.82a 7.74a 7.76a

12% 7.45ab 7.68a 7.45a 7.63a 7.26bc 7.42ab

15% 7.37b 7.58a 7.37a 7.55a 6.95c 7.11b

Kaeng Kua Fag8% 7.35bc 7.53a 7.57a 7.63a 7.55a 7.60a

10% 7.72a 7.53a 7.55a 7.43ab 7.47a 7.55a

12% 7.50ab 7.47a 7.47a 7.53ab 7.45a 7.35a

15% 7.13c 7.32a 7.22a 7.22b 6.85b 6.88b

Nam Ya Pla10% 8.02a 8.00a 8.20a 8.24a 8.11a 8.39a

12% 7.70b 7.91a 7.87b 7.98b 7.65b 7.78b

15% 7.41b 7.96a 7.80bc 7.91b 7.35b 7.67b

18% 7.07c 7.89a 7.57c 7.72b 6.98c 7.35c

Kaeng Liang8% 7.40a 7.02a 7.43a 7.31a 7.43a 7.25a

10% 7.31a 7.14a 7.14ab 7.24a 7.05b 7.17b12% 7.45a 7.05a 7.02bc 6.76b 6.36c 6.74c

15% 7.43a 7.05a 6.79c 6.79b 6.38c 6.57c

Kaeng Som8% 7.97a 7.22a 7.56a 7.72a 7.61a 7.72a

10% 7.61b 7.06ab 7.25b 7.58ab 7.56a 7.56a

12% 7.03c 6.86b 7.25b 7.33bc 6.92b 6.94b

15% 6.89c 6.83b 6.86c 7.28c 6.47b 6.58b

Kaeng Tae Po8% 8.07a 7.52a 7.80a 7.75a 7.91a 8.02a

10% 7.73b 7.45a 7.55ab 7.45b 7.52b 7.50b

12% 7.66b 7.52a 7.52ab 7.50ab 7.50b 7.45b

15% 7.57b 7.48a 7.30b 7.41b 7.07c 7.25b

In a column, means for each attribute followed by a same letter are not significantly different at the 5 % level by DMRT


using coconut milk as an ingredient, fat content in

the products was markedly increased. Coconut

milk contained dishes, namely Kaeng Tae Po,

Kaeng Kua Fag and Nam Ya Pla, exhibited 11.23-

13.85 g fat in 100 g. Non coconut milk dishes,

Kaeng Som and Kaeng Tai Pla contained fat

dramatically lower of 1.41 and 1.67 g/100g,

respectively. Thus, the products without coconut

milk provided low calories which were 40.40 Kcal

for Kaeng Som and 50.67 Kcal for Kaeng Tai Pla.

As shown in Table 4, calcium and

phosphorus contents of all products fall in the

range of 126.5 – 136.4 mg/100 g and 83.9 – 108.6

mg/100 g, respectively. The ratio of calcium and

phosphorus in the products was 1.2-1.6:1. The

relationship between calcium and phosphorus in

the diet plays an important role in the absorption of

both minerals. A dietary ratio of 2 parts calcium to

1 to 2 parts phosphorus was known to promote the

highest level of calcium absorption (Guthrie, 1983).

Sodium content appeared to be high as

expected in canned concentrated seasoning stock.

The sodium level in Kaeng Tai Pla was found to be

the highest as 1000.6 mg in 100 g. The reason

might come from applying tai–pla (fermented fish

viscera), which contained 15,000 mg sodium in

100g (Institute of Nutrition, 1999), as an ingredient

in the dish. Iron contents ranged from 0.86-1.82

mg/100 g while copper level fell in the range of

0.08-0.21 mg/100 g.

Fish flour incorporated foods and controls

The contents of protein and ash were greater

in foods prepared from canned concentrated

seasoning stocks than their controls (Table 3). Fat,

dietary fibre and carbohydrate of five dishes

incorporating fish flour were comparable to their

controls recipes. The determined minerals in this

section were sodium, calcium, phosphorus, iron

and copper. In general, minerals level was higher

in fish flour incorporated foods than their control.

And there was a distinction in case of calcium

which the contents appeared to be 1.4 - 2.5 times

higher in all dishes using fish flour than those in

the controls (Table 4). The present calcium contents

were 90.9, 119.3, 122.3, 142.5 and 204.8 mg in

100g of Kaeng Kua Fag, Kaeng Tae Po, Kaeng

Som, Kaeng Tai Pla and Nam Ya Pla, respectively.

The Thai Recommended Daily Dietary Allowances

for adult on calcium is 800 mg (Committee on

Recommended Daily Dietary Allowances, 1989).

Therefore, consuming 100g portion of the above

dishes obtained calcium about 10-25% of RDA.

Sensory evaluation and nutrient of five selectedherring fish flour incorporated foods and theircontrols

Dishes prepared from seasoning stock

contained herring fish flour were tested along with

their controls (Table 6). There was no significant

differences found in all aspects on Kaeng Tai Pla

at P>0.01. While the control dishes of Nam Ya Pla

and Kaeng Som received significantly better scores

of all attributes (P<0.01). The results demonstrated

that colour was the renowned deteriorate factors.

Attributes on taste of the control did not alter much

from the fish flour incorporated seasoning stock in

Kaeng Tae Po and Kaeng Kua Fag. Even though

many attributes were not as good as the control, the

verbal explanation of foods contained herring fish

flour was positive expression as better than ‘like

slightly’. The overall acceptability scores of control

and fish flour incorporated food were relatively

good ranging from 7.43-7.91 and 6.61 – 7.64,

respectively.

Consumer testFoods prepared from five different

concentrate seasoning stocks were evaluated by

280 panelists (Figure 1). The ratings of colour for

the top 2 boxes (combined percentages of like very

much and like) for Kaeng Tae Po, Kaeng Tai Pla,

Kaeng Kua Fag, Keang Som and Nam Ya Pla were

75.7, 69.2, 64.6, 60.3 and 58.5%, respectively.

The result of the bottom 2 boxes scores

corresponded to dislike and dislike very much due


to off colour (Figure 1). For traditional recipe of

Nam Ya Pla, steamed snake-head fish meat which

is white in colour was used. Therefore, applying

fish flour in Nam Ya Pla was strongly affected the

colour. On the contrary, Kaeng Tai Pla had the

least effect since it contained fermented fish viscera,

which naturally dark in colour.

Odour of fish flour did not affect all five

recipes. The panelists responded to the odour of

products as like and like very much in a range of

62.3 – 76.8 %. Spices incorporated in curry pastes

probably masked the fish flour odour in addition to

a fish sensation that supposed to exist in the dishes.

The taste rating for the top 2 boxes was

great in Kaeng Tae Po, Nam Ya Pla, Kaeng Kua

Fag and Kaeng Som, while, Kaeng Tai Pla was

verbally indicated as most dislike among these

five dishes. However, the dislike reasons came

from chilli hotness or saltiness of product, not

from the taste of fish flour.

Considering the overall acceptability,

Kaeng Tae Po, Nam Ya Pla, Kaeng Som and

Kaeng Kua Fag obtained high percentage of 80.5,

78.3, 74.0 and 68.8, respectively, in the top 2

boxes. The sandy mouth feel was mentioned to

cause unacceptability in dishes without coconut

milk.

CONCLUSION

It was possible to use herring fish flour for

Thai food preparations. Applying amount was

varied due to the nature of foods. Some limitations

could be concluded as darken colour, fish flour

odour, cooking style, liquid available, coconut

milk present or sandy mouth feel.

Nutritious wise, herring fish flour possessed

good quality protein and high calcium content, in

addition to many minerals. The five selected Thai

dishes prepared by incorporating fish flour provided

90.9 – 204.8 mg of calcium per 100g, which were

1.4-2.5 times of the traditional dishes and

contributed 10-25% of RDA for calcium.

Table 6 Sensory evaluation of five herring fish flour incorporated dishes and their controls.

Food Mean±S.D.

Colour Consistency Odour Taste Texture Acceptability

Nam Ya Pla

Control 7.95±0.58 7.68±0.48 7.93±0.47 7.61±0.72 7.86±0.56 7.91±0.59

18% fish flour 6.23±0.83** 7.23±0.75** 6.64±0.68** 6.70±0.96** 6.86±0.85** 6.61±1.00**

Kaeng Som

Control 8.02±0.53 7.83±0.49 7.56±0.77 7.41±0.83 7.72±0.54 7.43±0.86

15% fish flour 6.33±0.83** 7.28±0.65** 6.65±1.00** 6.56±1.04** 6.74±0.80** 6.37±1.00**

Kaeng Tai Pla

Control 7.82±0.52 7.82±0.39 7.64±0.80 7.70±0.61 7.68±±0.65 7.80±0.65

15% fish flour 7.52±0.64* 7.75±0.43 7.54±0.58 7.70±0.70 7.34±0.50* 7.64±0.54

Kagng Kua Fag

Control 7.98±0.49 7.72±0.45 7.63±0.80 7.76±0.56 7.67±0.56 7.83±0.70

15% fish flour 7.04±0.66** 7.54±0.56* 7.11±0.81** 7.28±0.77* 6.87±1.02** 7.20±060**

Kaeng Tae Po

Control 7.68±0.73 7.68±0.48 7.48±0.63 7.30±0.84 7.64±0.49 7.54±0.69

15% fish flour 6.82±0.78** 7.41±0.57** 6.91±0.70** 6.98±0.97 7.04±0.72** 6.93±1.04*

** Significantly different at µ = 0.01 by student’s t-test

* Significantly different at µ = 0.05 by student’s t-test


Fig

ure

1Pe

rcen

tage

of

cons

umer

s on

pre

fere

nce

and

acce

ptan

ce to

war

d fi

ve T

hai d

ishe

s pr

epar

ed f

rom

can

ned

conc

entr

ated

sea

soni

ng s

tock

s.

020406080100

Nam

Ya

Pla

Kae

ng S

omK

aeng

Tai

Pla

Kae

ng K

ua F

agK

aeng

Tae

Po

% Consumers

colo

ur

020406080100

Nam

Ya

Pla

Kae

ng S

omK

aeng

Tai

Pla

Kae

ng K

ua F

agK

aeng

Tae

Po

% Consumers

tast

e

020406080100

Nam

Ya

Pla

Kae

ng S

omK

aeng

Tai

Pla

Kae

ng K

ua F

agK

aeng

Tae

Po

%Consumers

Lik

e ve

ry m

uch

+ li

ke

Nei

ther

like

nor

dis

like

Dis

like

+ d

islik

e ve

ry m

uch

odou

r

020406080100

Nam

Ya

Pla

Kae

ng S

omK

aeng

Tai

Pla

Kae

ng K

ua F

agK

aeng

Tae

Po

%Consumers

Acc

ept

Nei

ther

acc

ept n

or u

nacc

ept

Una

ccep

t

over

all a

ccep

tabi

lity


ACKNOWLEDGEMENTS

This study was carried out under theresearch project “Utilization of Fish Flour in SomeFood Products” funded by Norwegian Herring Oiland Meal Industry Research Institute.

LITERATURE CITED

Association of Official Analytical Chemists. 1984.Official method of analysis. (14th ed.)Arlington, Virginia.

Association of Official Analytical Chemists. 1998.Official method of analysis. (16th ed.) (P)method 995.11. Arlington, Virginia.

Association of Official Analytical Chemists. 2000.Official method of analysis. (17th ed.) (K)method 973.53, (Na) method 973.54, (Fe,Cu,Zn, Cd, Pb) method 999.11 (Ca, Mg) method.991.25. Gaithersburg, Maryland.

Committee on Recommended Daily DietaryAllowances, Department of Health, Ministryof Public Health. 1989. Recommended DailyDietary Allowances for Healthy Thais. TheWar Veterans Association of ThailandPublishing Co., Bangkok.

Department of Health, 1995. The 4th Nationalfood and nutrition survey. Ministry of PublicHealth, Thailand.

FAO 1982. Fishery products and the consumer indeveloping countries. Report of the FAO/NORAD round table discussion, FrazersHill,Malaysia 11-16 June 1982. FAO Fish.Rep. 271, 37 p.

FAO/WHO 1973. Energy and proteinrequirements. Report of a joint FAO/WHOAd Hoc Expert Committee. WHOTechnical report series no. 522; FAONutrition Meetings Report Series 52. WorldHealth Organization, Geneva.

Faulks, R.M. and S.B. Timms, 1985. A rapidmethod for determining the carbohydratecomponent of dietary fibre. Food Chemistry17:273-287.

Guthrie, H.A.1983. Introductory Nutrition, 5thed. St.Louis, Missouri, The C.V. MosbyCompany. pp.135, 614.

Holland B., J. Brown and D.H. Buss. 1993. Fishand Fish Products. Third supplement to 5thedition of McCance and Widdowson’s TheComposition of Foods. Royal Society ofChemistry, Cambridge.

Institute of Nutrition. 1999. Thai FoodComposition Tables. 1st ed. Bangkok,Thailand: Paluk Tai Co., Ltd.150 p.

Kinsella,J.A. 1986. Food components withpotential therapeutic benefits: The n-3polyunsaturated fatty acids of fish oils. FoodTechnology 4:89-97.

Kinsella, J.E., B. Lokesh and R.A. Stone. 1990.Dietary n – 3 polyunsaturated fatty acids andamelioration of cardiovascular disease:possible mechanisms. The American Journalof Clinical Nutrition 52:1 – 28.

Ministry of Public Health, Thailand. 1987. FoodComposition Table for Use in Thailand.Nutrition Division, Department of Health,Ministry of Public Health. 48 p.

Phithakpol, B., S. Sringam, N. Sarikaputi and T.Panayotou. 1981. Report on the acceptabilitytesting of fish protein concentrate type B androller – dried fish in Thailand. Institute ofFood Research and Product Development,Kasetsart University, Bangkok. Report forthe Food and Agricultural Organization ofthe United Nations Rome. 123 p.

Phithakpol, B. 1984. Marketability and feasibilityof pilot plant production of roller dried fish inThailand. Institute of Food Research andProduct Development, Kasetsart University,Bangkok. Report for the Food andAgricultural Organization of the UnitedNations Rome. 74 p.

Prosky, L., N. G. Asp, L. Furda, J.W. De Vries, T.F. Schweizer and B.F. Harland, 1985.Determination of total dietary fibre in foodsand food products: Collaborative study.Journal of the Association of OfficialAnalytical Chemists 68:677-679.

Tahvonen, R., T. Aro, J. Nurmi and H. Kallio.2000. Mineral content in Baltic herring andBaltic herring products. Journal of FoodComposition and Analysis 13:893-903.


Lightning Surge Response of Concrete Pole due to Effectof the Electrical Properties of Concrete based

on the Electromagnetic Field Method

Samroeng Hintamai and Jamnarn Hokierti

ABSTRACT

Lightning performance of overhead distribution line affects the cost of line construction. For

economical insulation coordination in distribution line design, it is necessary to accurately predict the

lightning surge overvoltage that occurs in the electric power system. In particularly, tower or pole surge

impedance is one of the most important parameters for lightning surge analysis of distribution lines. This

paper presents the lightning surge response of reinforced concrete pole due to the electrical properties of

concrete based on the electromagnetic field theory, which has never been considered in the previous

lightning surge analysis. The electrical properties of concrete were measured over the frequency of range

from 100 Hz to 40 MHz during 86 days after pouring. The concrete sample was mixed according to a

construction standard of the electrical distribution pole of Provincial Electricity Authority (PEA). The

cement/sand/aggregate ratio was about 1:1.5:3 and water/cement ratio was approximately 0.3. It was

found that the electrical properties of concrete varied significantly over the frequencies and time after

pouring. Therefore, lightning surge response of reinforced concrete pole depended on the electrical

properties of concrete. The results showed that surge impedance calculated by the proposed formula

agreed well with the other measured value obtained from reduced- scale test.

Key words: concrete pole, electromagnetic field method, electrical properties of concrete, lightning

surge response, surge impedance

INTRODUCTION

Thailand is in a tropical zone and has the

highest number of thunderstorm days in this zone,

about 50-120 days per year. For protection of

equipments in the electric power system, a lightning

overvoltage is a significant factor. Thus, lightning

surge analysis is essential for insulation design of

the electric power system. Particularly, tower surge

impedance is an important factor in the insulation

1 Bako Agricultural Research Center, Ethiopia.2 Department of Agronomy, Faculty of Agriculture at Kamphaeng Saen, Kasetsart University, Nakhon Pathom 73140, Thailand.3 Department of Agronomy, Faculty of Agriculture, Kasetsart University,Bangkok 10900, Thailand.


coordination design for transmission/distribution

lines. A number of experimental and theoretical

studies on tower surge impedance have been carried

out (Motoyama and Matsubara, 2000). However,

surge impedance of concrete pole has not been

studied enough for lightning surge analysis.

Representative methods to investigate the

tower surge characteristics include measurement

on real towers, measurement on reduced-scale

models, analytical study on simplified geometry


and numerical analysis based on the

electromagnetic theory.

The first theoretical formulation of tower

surge impedance was proposed by Jordan (1934).

In this formulation, based on Neumann’s

inductance formula, it was assumed that the current

distribution inside the tower was uniform from the

tower bottom.

Theoretical formulations of tower surge

impedance based on the electromagnetic field

theory were proposed by Lundholm et al.(1958),

Wagner and Hileman(1959), Sargent and

Darveniza (1969) and Okumura and Kijima(1985),

considering effects of the vector potential generated

by the injection current into the tower only.

Propagation velocity inside the tower was assumed

at the speed of light to the top of the tower.

However, the effect of return stroke current was

neglected. The tower was approximated as a vertical

cylinder having a height equal to the tower, and a

radius equal to the mean equivalent radius of the

tower. Propagation velocity inside the tower was

assumed at the speed of light.

Measurements of the surge impedance of

actual transmission towers were reported by Breuer

et al.(1958) and Caswell et al. (1958). In both

cases, a reflection method was used, and similar

impedance values of 165 ohms were obtained at

the tower top. Measured propagation velocity inside

the tower was almost the speed of light.

Another experimental value for actual

transmission towers was reported by Kawai (1964).

He used a direct method to measure tower surge

impedance, and obtained an impedance value of

100 ohms at the tower top. His experimental

results showed that the tower response to a vertical

current is different from the response to a horizontal

current. Measured propagation velocity inside the

tower was 70–80% of the speed of light.

Scale-model measurements were reported

by Chisholm (1983) and Wahab et al.(1987).

Chisholm used the time-domain reflectometry

(TDR) method to measure tower surge impedance.

These measurements were performed using both

horizontal and vertical current injection. Measured

propagation velocity inside the tower was the

speed of light. Wahab et al.(1987) used the direct

method to measure tower surge impedance for

various angles of current injection. Measured

propagation velocity inside the tower was 80–90%

of the speed of light. These results showed that the

tower surge impedance was strongly influenced

by the angle of current injection.

Field measurements of full-scale tower

impedance using the direct method were reported

by Ishii et al.(1991) and Yamada et al.(1995).

These measurements were performed using

inclined and horizontal current injection. Both

papers proposed a surge impedance of the tower

based on the Electromagnetic Transient

Program(EMTP). Propagation velocity inside the

tower was assumed at the speed of light.

Measurements of surge response of a

transmission tower to actual lightning were reported

by Matsumoto et al.(1995), Shinjo et al.(1997)

and Motoyama et al.(1997). All of them estimated

the surge impedance of the tower based on the

measurements, and proposed a model of the tower

based on the EMTP. The results showed that surge

response and surge impedance of the tower

depended on the lightning discharge path direction.

Theoretical work was reported by Ishii and

Baba(1997). They estimated the surge response of

a tower by numerical electromagnetic field analysis.

The calculated results were compared with field

test results (Yamada et al., 1995). The analysis

showed that surge response and surge impedance

of the tower depended on the arrangement of the

current lead.

Theoretical formulation of tower surge

impedance based on the electromagnetic field

theory was proposed by Motoyama and

Matsubara(2000). The analysis showed that the

tower surge impedance depended on the direction

and velocity of return stroke current.

Recently, theoretical formulation of pole


surge impedance of concrete pole based on the

electromagnetic field theory, including the effect

of direction and velocity of the return stroke current

and the electrical properties of concrete, was

proposed by Hintamai and Hokierti(2003). The

calculated result showed that surge impedance of

concrete pole depended on the electrical properties

of concrete.

In this paper, a new formula of surge

impedance of reinforced concrete pole based on

the electromagnetic field theory by taking the

effect of the electrical properties of concrete is

proposed. The electrical properties of concrete

were measured over the frequency range from 100

Hz to 40 MHz during 86 days after pouring.


1. Model of lightning return stroke currentIn this model, as the downward leader

nears the earth, an upward leader (or the return

stroke) is initiated progresses upwards with a

velocity vR neutralizing the charge lowered by the

preceding steeped leader (Chowdhuri et al., 2001).

The lightning channel then consists of a vertical

column; the lower part, containing current, is

rapidly expanding upwards, and the upper part,

containing the residual charge of the proceeding

steeped leader, is diminishing rapidly, as shown in

Figure 1.

The surge impedance of the return stroke is

a function of the height and the velocity of the

return stroke. However, the conservative

assumption that the stroke is a constant current

source is almost universally used, i.e., the surge

impedance of the stroke is infinite.

2. Electric field produced by a step currentThe geometry adopted for the calculation

of electromagnetic filed is shown in Figure 2. This

configuration is a crude approximation to the

lightning return stroke which travel up with a

propagation velocity vR in lossy dielectric medium

from the earth while removing negative charge

from the channel previously formed by a downward

moving, negative charged, cloud-to-ground leader.

From the electromagnetic field theory, the

general solution for electric fields in cylindrical

coordination at any point (r,f,z) is defined as

(Rubinstein and Uman, 1989),

EA

t= - - ∂

∂f , (1)

where

E : electric field intensity, V/m,f : scalar electric potential, V and

A : vector magnetic potential, Wb/m .From the Lorentz condition,

earth

cloud

residualcharge

returnstrokei

vR

Figure 1 Model of lightning return stroke.

h

z'

-z'

dz' i(z',t )

(r, φ, z)R

Rl

Perfectly Conducting ground

ˆxa

ˆya

ˆza

aφ

ˆra

ˆza

aφ

aθ

ˆsr

a

sr

rφ

θ

image

Figure 2 Geometry of a vertical conductor above

a perfectly conducting ground.


— + ∂∂

=.At

me f0 , (2)

where

m : permeability of medium, H/m and

e : complex permittivity of medium, F/m.

It can be shown that the inhomogeneous solutions

are

A r tI r t R c

Rdz

z

,, /( ) =

¢ -( ) ¢¢

Úmp4

0

, (3)

fpe

rr t

r t R c

Rdz

z

,, /( ) =

¢ -( ) ¢¢

Ú1

40

, (4)

where

z¢ : traveling distance of the current, m,

r¢ : observation coordinates,

r : source coordinates,

R r r= - ¢ : distance between and , m,

c : propagation velocity of the current in

medium, m/s,

r : line charge density, C/m and

I : current distribution, A.

The medium of the vertical structure being

considered here have conductivity s, a dielectric

constant eR and permeability m. Thus, the complex

permittivity is defined as (Plonus, 1988),

e e e s w= -0 R j / . (5)

The dipole technique uses infinitesimal

time-varying as the source of the electric and

magnetic fields. Since the vector potential A r can

be found from the current alone, expression f in

terms of A allows us to write (1) in terms of the

current distribution alone. By solving the Lorentz

condition, we obtain,

fme

fr t Adt tt

, . .( ) = - — + = -•( )Ú1

0

(6)

Substituting (6) into (1), we obtain

E r t A dtA

t

t

, . .( ) = — —( ) - ∂∂Ú

1

0me

(7)

where f(t=–•) since the current and charge

distributions are zeros for times less than a certain

time t0 .

In this analysis, a step current of magnitude

I0 is traveling up in the positive direction inside

the concrete pole at velocity vR. It is convenient to

use a mathematical expression describing both the

real current distribution and its image at the same

time as (Rubinstein and Uman, 1989),

i z t I u t z vR¢( ) = - ¢( ), / .0 (8)

The function u(x) is called the Heaviside

function and is defined as,

u xxx

( ) =≥

ÏÌÓ

◊0, < 0

1, 1

We allow for the presence of the conducting

plane by using the method of images. The vector

potential Ar in cylindrical coordination at any

point (r,f, z) can be integrated to yield,

AI h z h z r

z z r=

-( ) + -( ) +

- + +ÊË

ˆ¯

mp0

2 2

2 24ln . (9)

The heightcan be found by setting the

argument of the Heaviside function to zero. Solving

for and inserting into (9), the vector potential Ar

can be obtained as,

AI v t z v t z r

z z rr

R R=-( ) + -( ) +

- + +ÊË

ˆ¯

mp0

2 2

2 24ln . (10)

Meanwhile, we can calculate vector

potential at the same point from the image channel.

Since to change the actual channel to the image

channel all we can change z to –z, it is readily seen

that

AI v t z v t z r

z z ri

R R=+( ) + +( ) +

+ +ÊË

ˆ¯

mp0

2 2

2 24ln . (11)


The sign of the vector potential from image

is the same as from the source because the directions

of current are the same for both. The total vector

potential is then given by,

A z A z A zr i( ) = ( ) + -( ). (12)

Substituting (12) into (7), we obtain the

total electric field intensity in cylindrical

coordination at any (r,f, z) point as,

EI

j z rz

R= -

- +

Ê

ËÁ

ˆ

¯˜

60 10

0 2 2e s e w/. (13)

3. Surge impedance of concrete pole3.1 Single conductor modelIf the lightning stroke starts from the top of

the pole at, t=0 the pole top voltage Vtop is obtained

by,

V E dztop z

h

= -Ú 0

. (14)

For 0£t<(r/c), the wave front of the

electromagnetic wave does not pass the point.

Therefore,

Vtop = 0 . (15)

For (r/c) £ t <(h/vR + r/c), the wave front of

electromagnetic wave passes through the point

(r,f,z). Therefore,

VI

j

h h r

rtopR

=-

+ +Ê

ËÁÁ

ˆ

¯˜

60 0

0

2 2

e s e w/ln . (16)

For, t > h/vR + r/c the electromagnetic

wave reflected on the ground surface reaches the

pole top. Therefore,

V t V h v r ctop top R( ) = +( )/ / . (17)

Since the voltage Vtop is produced by the

vertical return stroke current I0, the pole surge

impedance Zpole is defined as,

ZV

I

jIn

h h r

r

t h r c

V h r c t h r c

poletop

R

R

top R R

=

=

£ ( )

-+ +Ê

ËÁÁ

ˆ

¯˜

( ) £ < +( )+( ) < +( )

Ï

Ì

ÔÔÔÔ

Ó

ÔÔÔÔ

0

0

2 2

0

60

,

/,

/ /

/ / , / /

0 t < r / c

r / c

e s e w

n

n n

(18)

Finally, the surge impedance of concrete

pole Zpole can be obtained as,

Zj

h h r

rpoleR

=-

+ +Ê

ËÁÁ

ˆ

¯˜

60

0

2 2

e s e w/ln , (19)

where h is the height of the pole.

3.2 Multiconductor modelAn actual concrete pole is composed of

multiconductor as shown in Figure 3. The five

conductors being short-circuited at its boundary

(top and/or bottom), the total impedance seen from

the top is given considering the mutual impedances

between the conductors by

Z Z Z Z Z

j

d

r

d

r

d

r r

j

h

d

pole

r

g GMR st

GMR st g

r

= + + +( )

=-( )

Ê

ËÁÁ

ˆ

¯˜

Ê

ËÁˆ

¯

Ê

ËÁ

ˆ

¯˜

+-( )

ÊËÁ

ˆ¯

11 12 13 15

05

4

0

2

60 2 4 2

16 2

60 2 2

e s e w

e s e w

/

ln ln

ln

/ln

,

,

(21)

where r r d d d dGMR st st st st st st, .= +( )1 2 1 22 24

If it is tedious to calculate each component

of the above equation, the total surge impedance is

easily evaluated as the impedance of an equivalent

circular single conductor with the following

geometrical mean radius


Zj

h h R

RpoleR

GMR

GMR=

-+ +Ê

ËÁÁ

ˆ

¯˜˜

60

0

2 2

e s e w/ln ,

(22)

where R r r dGMR g GMR st= ( ) ( )1 5 1 5 45 /,

/.


1. Electrical properties of concreteThe electrical properties of concrete were

measured over the frequency range from 100 Hz to

40 MHz during 86 days after pouring, using

Impedance/Gain-Phase Analyzer Hewlett Packard

4164A with an accuracy of 0.17 percent as shown

in Figure 3. The concrete sample was formed by

placing aluminum plate electrode about 1 cm in

thickness. The cement/sand/aggregate ratio in

preparing the concrete specimen was about 1:1.5:3

and the water/cement ratio was near 0.3.

Capacitance and dissipation factor were measured

by impedance analyzer. Dielectric constant and

electrical conductivity were calculated from the

measured values. Change of the dielectric constant

and the electrical conductivity with the curing

time and frequency are shown in Figures 5 and 6.

In the first 3 days, dielectric constant and

electrical conductivity decreased rapidly. After 10

days, their changes become very slow. These

changes show good consistency with the chemical

change and water content in the hardening period

of concrete. Therefore, frequency of lightning

current of the first strokes is 25 kHz (10/350 ms),

dielectric constant of concrete is about 16 and

electrical conductivity is about 0.01 mS/m. Another

frequency of lightning current of the subsequent

strokes is 1 MHz (0.25/100 ms), dielectric constant

of concrete is about 8 and electrical conductivity is

about 0.122 mS/m.

2. Propagation velocity of waveThe propagation velocity of return stroke

current inside a medium of complex permittivity eis varied inversely to the square root of the complex

permittivity as (Morshedy, 2000),

yx

z I

a) structure inside of concrete pole

b) cross section area

Figure 3 Multiconductor model of concrete pole.

Figure 4 Measurement of the electrical proper-

ties of concrete.


0

20

40

60

80

100

120

140

160

180

0.1 0.6 4.0 25.1 158.5 1000.0 6309.6 39810.7freq.(kHz)

diel

ectr

ic c

onst

ant

1st day

3rd day

6th day

10th day

15th day

20th day

28th day

55th day

86th day tf ( µs) 2500 416.6 62.5 10 1.57 0.25 0.04 0.006

Figure 5 Change of dielectric constant of concrete.

0.000

0.005

0.010

0.015

0.020

0.025

0.1 0.6 4.0 25.1 158.5 1000.0 6309.6 39810.7freq.(kHz)

cond

uctiv

ity(S

/m)

1st day

3rd day

6th day

10th day

15th day

20th day

28th day

55th day

86th day tf (µs) 2500 416 63 10 1.5 0.25 0.04 0.006

Figure 6 Change of electrical conductivity of concrete.

velocityjR

= ¥-

3 108

0e s e w/. (23)

The propagation velocity of return stroke

current inside the concrete pole is shown in Figure

7.

In Figure 7, the propagation velocity of

wave inside the concrete pole is about 70 to 103.6

m/ms as the frequency of lightning current is

between 25 kHz to 1 MHz.

3. Comparison between calculated result andmeasured result

To clarify the effectiveness of the proposed

formula, we show the comparison of surge

impedance of reinforced concrete pole calculated

by the proposed formula with the measured result

(Yamamoto et al., 1997). The surge impedance of

reinforced concrete pole was measured by scale

model technique. The height of reinforced concrete

pole is 14 m and a radius of 0.377 m. The hollow


steel reinforced concrete pole was composed of

two parts; iron cage and concrete part of about 2

mm in thickness. The surge impedance was

measured about 242 ohms, whereas calculated

result from this configuration by the proposed

formula is about 258 ohms. Therefore, it showed

that calculated value is different from measured

value about 6.8%.

4. Model of concrete poleThis study deals with 22 m in height of

concrete pole that imbedded a grounding lead wire

of 35 mm2 at the center of the pole from top to

bottom and supplemental steel reinforced at square

inside of solid taper pole, as shown in Figure 8.

5. Surge impedance of concrete poleFigure 9 shows the calculation of surge

impedance of concrete pole by equation (22) due

to the effect of the electrical properties of concrete.

The frequency of lightning current is between 25

kHz to 1 MHz, surge impedance of concrete pole

is about 80 to 119 ohms.

CONCLUSIONS

In this paper, the formulation of lightning

0

20

40

60

80

100

120

140

0.1 0.6 4.0 25.1 158.5 1000.0 6309.6 39810.7freq.(kHz)

prop

agat

ion

velo

city

(m/µ

s)

1st day

3rd day

6th day

10th day

15th day

20th day

28th day

55th day

86th day tf (µs) 2500 416.6 62.5 10 1.57 0.25 0.04 0.006

Figure 7 Propagation velocity of return stroke current inside the concrete pole.

Figure 8 Configuration of the reinforced con-

crete pole.

surge response of reinforced concrete pole based

on the electromagnetic field by taking the effect of

the electrical properties of concrete was obtained.

The results showed that surge impedance of

reinforce concrete pole depended on the geometry

of pole, the dielectric constant and the electrical

conductivity of concrete.

ACKNOWLEDGEMENTS

The authors wish to thank the Electrical

Properties Measurement Laboratory of the National

Metal and Material Technology Center for service


of the concrete electrical properties of

measurement.

LITERATURE CITED

Jordan, C.A. 1934. Lightning computations for

transmission lines with overhead ground wires

Part II. General Electric Review 34: 180-

185.

Lundholm, R., R.B. Finn, Jr. and W.S. Price. 1958.

Calculation of transmission line lightning

voltage by field concepts. AIEE Trans. PT.

77: 1271-1283.

Wagner, C.F. and A.R. Hileman. 1960. A new

approach to calculation of lightning

performance of transmission lines. AIEETrans. PT. 79: 589-603.

Sargent, A. and M. Darveniza. 1969. Tower surge

impedance. IEEE Trans. PAS. 88: 680-687

Okumura, K. and A. Kijima. 1985. A method for

computing surge impedance of transmission

line tower by electromagnetic field theory.

IEE of Japan Trans. B. 105: 733-740.

Breuer,G.D., A.J. Schultz, R.H. Schlomann and

W. S. Price. 1958. Field studies of the surge

response of a 345-kV transmission tower and

ground wire. AIEE Trans. PT. 77:

1392–1396.

Caswell, R.W., I.B. Johnson, E.F. Koncel, Jr. and

N. R. Schultz. 1958. Lightning performance

of 138-kV twin-circuit transmission lines of

Commonwealth Edison company—operating

experience and field studies. IEEE Trans.PT. 77: 1480–1491.

Kawai, M. 1964. Studies of the surge response on

a transmission line tower. IEEE Trans. PAS.

83: 30–34.

Chisholm,W.A., Y.L. Chow and K.D. Srivastava.

1983. Lightning surge response of

transmission towers. IEEE Trans. PAS. 102:

3232–3242.

Chisholm W.A. and Y.L. Chow. 1985. Travel time

of transmission towers. IEEE Trans. PAS.

104: 2922–2928.

Wahab, M.A.A., I. Matsubara and H. Kinoshita.

1987. An experimental evaluation of some

factors affecting tower surge impedance. IEEof Japan Trans. 107: 171–177.

Ishii M., T. Kawamura, T. Kouno, E. Ohsaki, K.

Murotani and T. Higuchi. 1991. Multistory

transmission tower model for lightning surge

analysis. IEEE Trans. PWRD. 6: 1327–1335.

Yamada T., A. Mochizuki, J. Sawada, E. Zaima,

T. Kawamura, A. Ametani, M. Ishii and S.

Kato. 1995. Experimental evaluation of a

UHV tower model for lightning surge analysis.

Figure 9 Surge impedance of concrete pole.

0

20

40

60

80

100

120

140

160

0.1 0.6 4.0 25.1 158.5 1000.0 6309.6 39810.7freq.(kHz)

surg

e im

peda

nce(

ohm

s)

1st day

3rd day

6th day

10th day

15th day

20th day

28th day

55th day

86th day tf ( µs) 2500 416.6 62.5 10 1.57 0.25 0.04 0.006


IEEE Trans. PWRD. 10: 393–402.

Matsumoto Y., O. Sakuma, K. Shinjo, M. Saiki, T.

Wakai, T. Sakai, N.Nagasaka, H. Motoyama

and M. Ishii. 1995. Measuring of lightning

surges on test transmission line equipped with

arresters struck by natural and triggered

lightning. IEEE Trans. PWRD. 11:

996–1002.

Shinjo K., Y. Matsumoto, O. Sakuma, T. Wakai,

T. Sakai, M. Ishii and H. Motoyama. 1997.

Characteristics of transient response of

okushishiku test transmission line struck by

natural and triggered lightning. IEE of JapanTrans. 117-B: 478–487.

Motoyama H., K. Shinjo, Y. Matsumoto and N.

Itamoto. 1998. Observation and analysis of

multiphase back flashover on the okushishiku

test transmission line caused by winter

lightning. IEEE Trans. PWRD. 13: 1319-

1398.

Ishii M. and Y. Baba. 1997. Numerical

electromagnetic field analysis of tower surge

response. IEEE Trans. PWRD. 12: 483–488.

Motoyama, H. and H.Matsubara. 2000. Analytical

and experimental study on surge response of

transmission tower. IEEE Trans. PowerDelivery 15: 812-819.

Hintamai, S. and J. Hokierti. 2003. Lightning

surge response analysis of concrete pole using

the electromagnetic field method, pp. 568-

573. In Transmission and DistributionConference and Exposition. Dallas Texas.

Chowdhuri, P. S. Li and P. Yan. 2001. Review of

research on lightning-induced voltages on an

overhead line. IEE Proc.-Gener. Transm.Distrib. 148: 91-95.

Rubinstein, M. and M.A.Uman. 1989. Methods

for calculating the electromagnetic fields from

a known source distribution: application to

lightning. IEEE Trans. Electromag.Compat. 31: 183-189.

Plonus, M.A. 1988. Applied Electromagnetics.International Editions. New York: McGraw-

Hill.

Morshedy, A.El. 2000. High-Voltage EngineerngTheory and Practice. New York: Marcel

Dekker, Inc.

Yamamoto, K., Z. Kawasaki, K. Matsuura, S.

Sekioka and S. Yokoyama. 1997. Study on

surge impedance of reinforced concrete pole

and grounding lead wire on distribution line

by experimental on reduced scale model, pp.

209-212. In Proc. 10th Int. Symp. HighVoltage Eng.

Documents

JH2005