comparison between two methods ofhaploid production in hexaploid triticale(x rriticosecale wittmaclà

J. El Haddour! o* , B . Mouatassim ", M . Mergoum b,N. Nsarellah"

a" Institut National de la Recherche Agronomique. BP 589, Setnt,Moroc co, 2 I 2 ( 0 3 ) 40 3 2 I 0. E lhaddoury24 @ yahoo.fr

b. N.D. S.U. Crops and weed Science, ND, 55108. ND.USA.

AL AWAMIA I 15 Vol.2 N"3, 2005

Abstract

In triticale breeding programs based on the doubled haploid technique, the rate of obtainedhaploid plants is important. Seeds of the Fl generation are costly and available only insmall quantities.In this study, two techniques have producing haploid plantations in triti-cale haggard been compared; thesis were: in vitro androgenesis through anther culture and,the intergeneric crossing with maiZe.

The resuh indicated that the in vitro androgenesis method was superior despite a high rateof obtained albinos. This method had a higher performance in producing haploid plants,with an average of 3.5 haptoid plants produced per cuhured spike. On the other hand, evenif producing haploid plants through crossing with Maize was less performing than antherculture overall, all plants regenerated were green. A possible explanation of this finding liesprobably in that the lower fertiliry rate is linked to incompatibitliry mechanisms, thus resul-ting in a low (0.38) average number of obtained plants per fertilised spike .The number of available FI seeds is important for the genetic variability and hence for thepotential recombination of traits. It is recommended that breeders use the androgenesis me-rhod.

Key words : Trit icale; in vitro androgenesis; intergeneric cross; doubled haploid.

:1.4, , .1t<. i , r l l A. oL.r iJ.sél 6rLloÇ CUi) ù,+ô LrLi . . . . - i t l l ia J)[ i a,

t 9

d -xul icl_r, i;a ùJi ,r,J.t JJè ,JÊJJIç+ t<-ri f* ;J1.- é 6. oL3;3 3rSl

.ÀJiJl ê. ç.9lrJ l3; , , i - l t ; . . t r ;

ciLi u313 dr. f.Jti d.i, L-lJJl oja o;,ellt1.rLl i i - 3,5 e Ui! ( .1,-^ l elJ js 4pf jJ l

é t :< l l l - s t l

6^*tJt t.l_,,j 6. t;Li. JEI a..À oLs;.r.sél .r,.rLl tjt j ôJjl-t .. CJI_/I c+. s+fLll J,^iJl uè#Yl alti.tl ; osJi

n:îl ry d'*ti: ,r! J-:r: cr!lr ..1r<,:.�;U ,il, Jl ù*,^:,1 aJuà,J i r.... tL.JlYl ùÈ+{Jl ip .r3 ,^.^ll .r.r,Jl lr!; 4.i; i l- ,: diJl orirtl ,.," ,ù 41U333*srSi

CE;l ,,r" J-ll .,3t{Sll l"*Jl ù^j é .l^^f f 6.>.-i-f l i.clr; 1gA JUII-L a,;l eu,J. çr+fi J*iJ, u+J:"yl dlul

. ÀJJl ê. g.rl!l ,erLJjJ.J.;SJl .r.rt-l erÇ ,Jq*jJl , Irtrdlf C,6Stl

20 AL AWAMIA I 15 Vol.2 N"3, 2005

RésuméIn comparaison de deux techniques de production de plantes haploïdcs, l'an'drogénèse in vitro

et Ie croisement intergénérique avec le mai's, a été conduite sur le triticale hexaploide -

Les résu1tats ont rnontré qu' en dépit du taux d'albinisme éIevé , l'androgénèse in vitro est plus

perform.ante comme rnéthode de production d'haploi'des, avec une moyenne de 3.5 plantes ha-

plordes par épi mis en culture.

ln production de plantes haploides par croisement avec le mais reste inférieure à celle par cul-

ture d'anthère, bien que la totalité des plantes régénérées soient chlorophyllienne'

Le faibte taux de fécondation dû probablement à des mécanismes d'incompatibilité fait qae le

nombre d'embryons sauvés et par conséquent celui de plantes haploi'des ne dépasse pas 0'38

plantes par épis castré.

Dans un progrcunrne d'amélioration génétique du triticale utilisant I'haplodiloïdisation comme

méthode de créationvariétale, le nombre de plantes haploïdes produites est très important car

la quantité de semences F I est faible et coîtteuse . II est donc recommandé en cas d'application

de t,haptométhode d'utiliser la culture d'anthère avec toutefois le transfert rapide des em-

bryons sur des milieux de régénération améIiorés.

Mots clés : Triticale ; androgenèse in vitro; croisement intergénérique ; haplodiplotdi-

sation.

AL AWAMIA ll5 Vol. 2 N'3.2005

IntroductionTriticale (X. Triticosecale Wittmack), is a man made species produced through crossingbetween durum wheat and rye (hexaploid triticale) or between bread wheat and rye (oc-taploid triticale). The first cross (wheat x rye), obtained were sterile (Wilson, 1976) orpartially fertile. With the discovery of colchicine, it became possible to routinely pro-duce fertile hexaploid triticale by doubling the number of chromosomes. This techniquepermitted crossing between different genotypes of wheat and rye and therefore, the èx-ploitation of a new genetic variability that allowed this new species to be an alternativecrop beside wheat and barley in some stressed environments. Indeed, triticale toleratesacid and poor soils, and resists better to many different biotic and abiotic stress than theother small grain cereals. In addition, triticale responds well favourably to fertiliser ap-plication and the use of low doses of nitrogen and phosphorus outputs an optimum pro-duction (Mergoum et al., 1992).Since with its self pollinated crop status triticale tolerates consanguinity, the haplodi-ploidisation or double haploid method is alternative to the classic techniques of self pol-linated plants improvement or selection (Pedigree, SSD or Bulk) and can be used in va-riety creation. This method presents the advantage to fix lines in a short time.In the same way and in the case of a limited introduction from the rye genome (addition,substitution or translocation lines), the application of the double haploid method permitsfixation to homozygoucity state in only one generation.The first studies on double haploid production of triticale used the in vitro androgenesisas method of haploid production. This technique permits good embryogenesis bù plantdevelopment is characterised by the production of a very large number of albino piants(ono et later,1976; Bernard, l97l-). This phenomenon could be explained by thè pre_sence of the rye genome (R), whose in vitro androgenesis is marked by the high percen-tage of albino plants (Wenzel et al.,1977; Marciniak et al., 199g). In triticale micro-spore culture, albinism remained a serious problem, which was reflected by occurrenceof more than 5ovo of the regenerates being albinos (Monostori et al., l99g).With other methods of haploid production in triticale, albinos plants do not generallyoccur. Crossing with Hordeum bulbosum is subject to incompatiUitity problems (Snapeet al 1979.) but fertilizing with maize pollen can be considered as an alternative methàdfor haploid production. Maize is insensitive to action of Krl and Kr2 alleles for incom-patibility with rye genome.

The objective of this study is to compare two techniques of haploid production in triti-cale, namely in vitro androgenesis and intergeneric crossing *ittt maize.This compari-son will permit to identify the most efficient and most economic technique for triticalegenetic improvement in the breeding programs of INRA, Morocco.

21

22 AL AWAMIA 115 Vol. 2 N'3.2005

Material and methodsThe plant material is constituted of twenty Fl hybrid of hexaploid triticale (X. Triticosecale

Wittmack) provided by the International Center for Wheat and Maize Improvement (CIM-

MYT inQ Mexico City D. F. Mexico (Table l) and a Moroccan maize variety (Doukkalia).

The triticale Fl hybrid plants were grown in the greenhouse. Planting was made at different

dates in order to coincide their flowering period with that of maize.

Table 1. Cross name and pedigree of 20 hybrids used in the comparison.

name1 53\MG/CMH7 7 .rr35 | CI0|lHT 7 Ar

1 53WG//ERIZO61 O/BULL- l - I2 'TCBI53WG/3/DAGRO/IBEXIICIVET-2

4 TCBI53WG//ERIZO-15iFAHAD-3) TCBI53WG/FAHAD-56 TCB153WG/FAHAD-6

RDV3A4ANATI-1I53WGLAMB-2

9 TCBI53WG^,IANATI-I110 TCBI53WG/SlPrKA-ll3lEDA-7llMzNZA75l4lGNrOIT_ TCBI53WGi3/PRESTO//2*TESMOI-/MUSX-60312 TCBI53WGRHINO-3/BULL-I.11 3

- TCB I53WG/3IRONDO/BANT-5//ANOAS-2

16 TCBI53WG/POLLMER-2.2.I1- TCBI53WG/3/TESMO-1/MUSX-603//FAHAD-4

I53WG//ANOAS-3ÆATU-419 TCBI53V/G//BULL-10/MANATI-I

For jn vitro androgenesis, spikes were cut when microspores were at the mid-uninucleate

and vacuolate stage of development and were pretreated for 7 days at 3oC in the dark (Picard

and De Buyser, 1975). The spikes were sterilise d in a 4Vo calcium hypochlorite solution for

3 minutes and rinsed with sterile water. The anthers were cultured under aseptic conditions

on C17 anther culture medium (Wang and Chen, 1986). After incubation at24oC in the dark,

the embryos that were developed out of the cultured anthers were transferred on R9 regene-

ration medium (Miller, 1963). The albino plants were counted, and eliminated.

For intergeneric crossing with maize, the method adopted in this study was described in

bread wheat (Laurie and Bennett, 1986). Spikes were emasculated two to four days before

anthesis and covered with glassine bags. The spikes were pollinated with freshly collected

AL AWAMIA 115 Vol. 2 N'3. 2005

maize pollen. At the same time, a solution with 100 mg l-lof 2,4-D (2,4-dichlorophenoxy-acetic acid) is injected in the upper cavity of the last inter node. One day after pollination asecond injection of the 2, 4-D is applied as well as the spraying of 75 mg l-' of gibberellicacid (GA) on the spike.

Two weeks after pollination, spikes were collected. Seeds were extracted from spikes andsurface sterilised ina(4Vo calcium hypochlorite solution for I minute, TOVo ethanolfor I mi-nute) and then rinsed with sterile distilled water. Embryos obtained were aseptically excisedand transferred on Murashige and Skoog medium (Murashige and Skoog , 1962) in plasticpetri dishes and incubated at24"C in the dark. After germination, plantlets were transferredto pots and placed in the greenhouse.

The green plants with developed roots obtained from both methods were transplanted to soiland grown in greenhouse. The chromosome numbers of the cultured plants were counted inroot tips using the standard feulgen technique. Haploid plants were treated withO.ZVo col-chicine, l7o DMSO for 4 hours, washed and replanted in order to obtain fertile double ha-ploid plants.

Statistical analysis was made in anther culture for the following traits: percentage of em-bryos expressed per 100 anthers cultured, green plants and albinos plants expressed per 100anthers cultured.

For the cross with maize, statistical analysis was made for the following traits: percentageof caryopses, embryos and green plants expressed per 100 pollinated florets.

The analysis of variance was made with the genotype as the source of variation in a rando-mized design. The separation of means was achieved by Duncan test with LSD at 5Vo levelof probability.

Results and discussionThe analysis of the variance showed a very highly significant effect of the genotype on thepercentage of embryos, green plants and albino plants in anther culture (Table 2).

Table 2. Analysis of variance: effect of genotype on the percentage of embryos, green plantsand number of albino plants.

Source of variation df 7o Embryos 7o green plants 7o albinos plants

z)

Genotype 19 4424,72 rrr(* 39254 * 'r* 815 ,5 * * *

Error* * *, significant at the I Vo" level.

Differences between genotypes were also important; the percentage of embryos varied fromÙvo to 2l55vo (genotype 8). Green plant percentages varied from \vo to 37,5Ea(genotype 3). The overall mean of embryos percentage is 66,27o, the overall mean greenplants percentage is 8,3Vo and the overall mean albino plants percentage is lSVo (Table 3).

20

24 AL AWAMIA 115 Vol. 2 N"3.2005

Table 3. Results of in vitro androgenesis obtained by culturing anthers from 20 triticale Fl

hybrids.

Spikes AnthersHvbridt ued cultured

Embryos obtained Green Plants Albinos Plants

Number 9o Number Vo Number lo

168 r4,3 j 2,4 g l6 9,5 i

84 20 23,8 i 2,4 e t 4 16,7 h

168 97 \ , t g 13 ,1 e 44 26.2 e

168 37,5 h 1 ,8 E 8,3 ij

84 0 l 0 h

r26 97 1 7 f 0,8 gh 9 7 ,1 j

84 9r r08,3 d 2,4 e 20 23,8 f

l 8 r 215,5 a 1 6 1 9 d 25 29,8 d

84 1 1 0 l 3 l c 26 3 l a 47,6 a

84r0 9,5 k 0 h

r l 84 93 l r0,7 d 1,2 gh

t2 168 r62 96,4 e 14,3 e 59 35,1 c

1 3 126 7,9 k 2,4 g

9,584t4

15 ef2 l84 148 176,2 b 6 f

r6 J J 39,3 h 0 l

r7 64'16,2 f l 6 1 9 d 4 l 48,8 a

r8 66 78,6 f 2 l 2 5 c 34 40,5 b

19 126 tzl 9 6 e 28,6 b z7 21,4 g

t2620 39,7 h 0 h 26 20,6 c

Total 2184 r446

ISD (percentage of embryos) = 2,6, LSD (green plant percentaTe) = 0,54, LSD (percentage of plantsalbino) = 0,57.Percentages followed by the same letter are not significantly dffirent at 0.05 probability by DuncanI,l.�D test

The embryos production is satisfactory, but the number of green plants is low. Only 182

green plants have been regenerated for more than 1446 embryos. In the same way, a high

percentage of albino plants is to be noted. The whole 182 green haploid plants were produ-

ced from 52 cultured spikes, which represents an average of 3,5 green haploid plants per

spike. The analysis of variance of seed set, number of embryos and green plants showed a

very highly significant effect of the genotype (Table 4).

395r82

AL AWAMIA 115 Vol. 2 N"3, 2005

Table 4. Analysis of variance: effect of triticale genotype on seed set, embryos and numberof green haploid plants.

Error

x**, significant at the I Vo" level.

The intergeneric cross with maize were marked by a low caryopses set percentage (63Va)(Table 5). Variations between genotypes were between\Vo and 307o (genotype 20), and theabsence of albinos plants is noted. Indeed, from 37 rescued embryos, 21 greenplants wereobtained. The number of embryos and green plants varies according to genotypes of Fl hyrbrids. All the2l green haploid plants are gotten from 54 emasculated spikes, which repre-sents an average 0,38 green haploid plants per spike. The repetition number must be increa-sed in order to obtain more caryopsis and embryos.

If the intergeneric crossing with maize is considered as an alternative method for haploidplant production in recalcitrant species to in vitro androgenesis, the case of triticale, asshown in this study, is to treat otherwise. Triticale, due to its hybrid structure (interspecificorigin), has two different responses for the two haploid plants production methods used.

25

20

source of variation df %o seed set zo Embryos zo Haploid plants

37,7 *** 46,19 *** 12,72 *,**

26 AL AWAMIA 115 Vol. 2 N'3. 2005

Table 5. Numbers and percentages of seed set, embryos and haploid plants obtained from

crossing 20 triticale Fl hybrids with maize.

Spikes Florcts Caryopses developed Embryoc obteined Green plrntsHybrld Fl

emrccuhted pollinated Number yo Number

0 d0 d0 g

8,8 def 3,8 e 1,25 cd

r7,5 b l 0 b 7,5 b

12,5 cd 2,5 cd, 0 d

0 d0 d0 g

0 d0 d0 g

0 d0 d0 g

12,5 cd l 0 b 7,5 b

l l 13,8 bc 2,5 cd 1,25 cd

0 d0 d0 gl0

0 d0 d0 gl l

0 d0 d0 g

f I 2,5 cd 1,25 cd

12,5 cd 5 c 2,5 cd

l 5 7,5 ef 5 c 0 d

5 b cl 0 b11,3 cde

0 d0 d0 g

0 d0 d0 g

0 d0 d0 g

3 0 at2 22,5 a t7,5a

2 l) t1080

LSD (percentage of seeds) = t 3 ,96., LSD (percentage of embryos) = 2 ,47 , LSD (percentage of plants)= 3.6.Percentages fotlowed by the same letter are not significantly dffirent at 0.05 probability by DuncanISD test.

ln in vitro androgenesis, the hybrid vigour especially appears in embryogenesis haploid(Figure 1 ) . Some genotype had 215 .5Vo in percentage of obtained embryos (genotype 8) .

AL AWAMIA 115 Vol.2 N"3.2005 21

Figurel. Embryoid formationobtained from trit icale antherscultured on Cl7 medium.

The crossing with the maize generates a low rate of seed set probably due to mechanisms ofincompatibil i ty that appear in species descending from interspecific hybridization. This mayexplain the low rate of seed set observed in this study: 6,3 seeds set for 100 flowers poll ina-ted. These results could be confirmed or improved by an ulterior study with enough repeti-t ion.

The anthers culture technique is simpler than crossing with maize which requires, in acldi-t ion, the adjustment of f lowerings periods of the two species and their genotypes. The perioclof t ime between the planting and germination of intended plants to the haploid productionand the production of doubled seeds is appreciably even for the two methocls: this period isbetween 9 and 11 months. Considering this results, crossing with maize does not offer an at-tractive alternative for the in vitro androgenesis in trit icale.

The bottleneck of the plants haploid ploduction thlough anther culture remains the greenplants development. A lot of progless is achieved by the improvement of environmentalconditions of culture, with new sources of carbon and nitrogen (Marciniak an<l al., 1998)that permitted the production of embryos in large numbers. Much effort must be also macleto improve surrounding conditions of regeneration stage in or<Jer to obtain more greenplants. Embryos obtained from anther culture are transformecl quickly to unorganizecl callusand necrosed, their fast transfer on l 'egeneration medium could increase the number of greenplants.

In bread wheat, the use of single regeneratecl medium which plantlets are regenerated cli-rectly from anthers and also imploved the frequencies of green plants regeneration (La-shermes, 1992). We observed the same phenomenon in trit icale, green haploid plants coulclbe effectively produced via anther culture with indol acetic acicl (AIA), instead of 2.4-D (Fi-gure 2). However, culture - regenerated medium should be improved.

28 AL AWAMIA I 15 Vol.2 N'3, 2005

Figure2. Direct regeneration from

triticale anthers.

ConclusionIn spite of the problem of albinism and the low rate of regeneration, the in vitro androgene-

sis remains the most advantageous method for producing haploid plants in triticale. With a

satisfactory embryogenesis, regeneration must be improved to increase the number of viable

haploicl plants necessary to recover a useful genetic variability prior to plant selection and

evaluation process.

AL AWAMIA 115 Vol.2 N'3. 2005

Références

Bernard S., 1977. Etude de quelques facteurs contribuant à la réussite d'androgénèse invitro chez le trit icale hexaploïde. Ann. Amélio. Plantes. 27(6),639-655.

Lashermes P., 1992.Improved anther culture method for obtaining regeneration in wheat(trit icum aestivum /.). J. Genet. & breed. 46:99-102.

Laurie D 4., bennett M D., 1986. Wheat x maize hybridization. Can. J. Genet. Cytol.28,3 r 3 - 3 1 6 .Marciniak K., Banaszak Z.,Wedzony M. 1998. Effect of genotype, medium and sugar on tri-ticale (X triticosecale wittmack.) Anther culture response. Cereal-research-communication.,2 6 ( 2 ) : 1 4 5 - 1 5 1 .

Mergoum M., Ryan J., Shroyer P., Abdel Monem M., 1992. Potential for adopting triticalein morocco. J . Nat . Recoure. L i fe Sci . Educ. 2 l (2) ,137-141.

Miller C O., 1963. Kinetin and Kinetin-l ike compounds, inerne methoden der Flanfenana-lyse, Linskens H. F. And Tracey M. V. (eds.). Springer, Berlin, Heidelberg, New York,194-202.

Monosrori T, Puolimatka M., Pauk T. 1998. Triticale (X. triticosecale wittmack) in vitro an-drogenesis isolated microspore culture. Novenytermeles.

41 (4) 37r-382.

Murashige T., Skoog F., 1962. A revised medium for rapid growth and bio assays with to-bacco tissue cultures. Physiol. Plant. 15, 413-497.

O'mara j g . , 19a8. Fer t i l i ty in hal lopolyploïds. Res . Genet . Soc. Am. l1 ,17-21.Ono h.,larter e n.,1916. Anther culture of trit icale. Crop. Sci. 16,120-122.

Picard E., De Buyser 1., 1975. Nouveaux résultats concernant la culture d'anthères de

Triticum aestivum /. Conditions de régénération des plantes haploïdes et production

de lignées entièrement homozygotes. C. R. Acad. Sci. Paris, t-281,989-992.

Snape W. J., Chapman J., Moss J., Blanchard C. E., Mil ler T.E. 19i9. The crossabil ity ofwheat varietes with hordeum bulbosum.Heridity 42,3, 291-295.

Wang P., Chen Y P., 1989. A study on the application of c17 medium for anther culture.

dara. Boinicia. 28. 46-49.

Wenzel G., Hoffman F., Thomas 8., 197'7 . Anther culture as a breeding tool in rape. Ploidylevel and phenotype of androgenetic planrs. Zplanzenziichtg.T8, | 49 - | 55.

wilson A. s. 1876. on wheat and rye hybrids. Trans. Proc. Bot. soc. Edinburgh, 12.,286-288.

29