REGULAR PAPER

Influence of germination date on Dioon edule (Zamiaceae) seedlingtolerance to water stress

Laura Yanez-Espinosa • Joel Flores •

Paulina S. Rodrıguez Millan • Gabriel Rubio Mendez

Received: 22 August 2013 / Accepted: 18 December 2013

� The Botanical Society of Japan and Springer Japan 2014

Abstract Dioon edule seedling mortality is mostly

attributed to dehydration by prolonged drought, even when

they present xeromorphic characteristics like the adult

plants. The effect of germination date (GD) and soil water

deficit on seedling tolerance to water stress was assessed.

The seedlings germinated and grown from mature seeds

every month from December to April GD were selected to

evaluate the leaf area, photosynthetic pigment content,

crassulacean acid metabolism (CAM) activity, stomatal

conductance (gs) and leaflet anatomy at soil water potential

(Ws) of 0.0 MPa (day 1), -0.1 MPa (day 40), -1.0 MPa

(day 90), -1.5 MPa (day 130), and a control (0.0 MPa at

day 130) to recognize differences due to leaf development.

The seedlings shifted from C3 to CAM cycling when

exposed to water stress at Ws of -1.0 MPa, like adult

plants. The March–April GD seedlings with undeveloped

sclerified hypodermis and stomata, presented reduced leaf

area, lower Chlorophyll a/b ratio, higher CAM activity and

midday partial stomatal closure when reached Ws of

-1.0 MPa. These have higher probability of dehydration

during severe drought (February–April) than those of the

December–February GD with similar Ws. Plants used for

restoration purposes must have full leaf development to

increase the survival.

Keywords CAM cycling � Cycads � Drought � Leaflet

anatomy � Mexico � Seedling mortality

Introduction

The endangered cycad Dioon edule Lindl. (Zamiaceae) is a

dioecious gymnosperm with a relatively limited production

of seeds and high mortality rate at the seed stage, up to

90 % under natural conditions (Vovides 1990). Seed

mortality has been attributed mainly to pre-germination

predation by both insects (Coleoptera) (Mora et al. 2013)

and the field mouse Peromyscus mexicanus (Rodentia:

Muridae) (Vovides et al. 2003).

Nevertheless, the high germination capacity of viable

seeds of D. edule, up to 75–98 % (Vovides 1990), com-

pensates for the aforementioned disadvantage. Although D.

edule seeds germinate in both shadow and light under

appropriate temperature and moisture in nursery condi-

tions, shadow increased 55.5 % germinability along

three months (Lopez-Ovando and Trevino-Garza 2008).

Therefore, there must be some ecological advantages for

seedlings provided by seed germination under shadow in

the field, as Dioon species most likely require low light

conditions during the initial growth stages, which are

L. Yanez-Espinosa (&)

Instituto de Investigacion de Zonas Deserticas, Universidad

Autonoma de San Luis Potosı, Altair 200, Col. del Llano,

San Luis Potosı, S.L.P 78377, Mexico

e-mail: lyaneze@uaslp.mx

L. Yanez-Espinosa � P. S. Rodrıguez Millan � G. Rubio Mendez

Programas Multidisciplinarios de Posgrado en Ciencias

Ambientales, Universidad Autonoma de San Luis Potosı, Av. Dr.

Manuel Nava 8, Zona Universitaria Poniente, San Luis Potosı,

S.L.P 78290, Mexico

P. S. Rodrıguez Millan

e-mail: paulina_rmillan@hotmail.com

G. Rubio Mendez

e-mail: rubio.gabriel864@gmail.com

J. Flores

Division de Ciencias Ambientales, Instituto Potosino de

Investigacion Cientıfica y Tecnologica, A.C. Camino a la Presa

San Jose 2055, Col. Lomas 4a Seccion, San Luis Potosı,

S.L.P 78216, Mexico

e-mail: joel@ipicyt.edu.mx

123

J Plant Res

DOI 10.1007/s10265-014-0624-4

provided by filtered sunlight in forest canopies or maternal

plants (Alvarez-Yepiz et al. 2011).

The seedling mortality rate observed in some popula-

tions reaches up to 40 % for D. edule (Pavon-Saure, pers.

comm.) attributed mainly to the effect of water deficit due

to high temperatures and prolonged dry season during the

first year (Vovides 1990; Vovides et al. 2002). However,

why does dehydration mortally affect seedlings during the

first year?

Considering that adult plants of D. edule have the typ-

ical cycad xeromorphic traits that could allow them to

survive prolonged drought, including leaves with a coria-

ceous texture, thick cuticle, heavy cutinization, sunken

stomata and the presence of mucilage (Medeghini Bonatti

and Baroni Fornasiero 1990), relatively low net CO2

assimilation and stomatal conductance to H2O, such as in

succulents and evergreen conifers (Marler and Willis

1997), and C3 metabolism when the available water supply

is sufficient, they can shift to a crassulacean acid metabo-

lism (CAM) photosynthetic pathway in response to

increased CO2 assimilation and water use efficiency under

water stress, triggered at a soil water potential below

-1.0 MPa (Silvera et al. 2010; Vovides et al. 2002). Has

also been observed that seedlings of D. edule begin to

absorb large volumes of water in a short period of time

when the soil water potential reaches -0.1 MPa (unpub-

lished data).

Here, we explored why not all D. edule seedlings sur-

vive the drought if they have the same xeromorphic traits

than adult plants that allow them to tolerate water stress,

addressing the following questions: (a) How the germina-

tion date (GD) and the water stress would affect the leaf

development? and (b) How soil water potential would

affect the seedling tolerance to water stress? To answer

these questions we examined leaf area, photosynthetic

pigments, CAM activity (DH?), stomatal conductance (gs)

and leaflet anatomical characters, at soil water potential

(Ws) of 0.0 MPa (day 1), -0.1 MPa (day 40), -1.0 MPa

(day 90), -1.5 MPa (day 130), and a control (0.0 MPa at

day 130) to recognize differences due to leaf development.

Materials and methods

Plant material

In Dioon spp. the embryos are completely developed

immediately after the seed is released from the cone and

can germinate (Calonje et al. 2011). The seeds were col-

lected from mature strobilus ready to release them in

October 2007, in a population located in the southern

region of the Sierra Madre Oriental in the state of San Luis

Potosi, Mexico, at 1,167 m above the sea level (a.s.l.)

along a 33� slope with N exposure, clay loam shallow

(40–45 cm) soil rich in Ca (103.8 meq 100 g-1). The mean

temperature varies from 28.4 �C in summer to 14.1 �C in

winter. The mean total precipitation is 684.5 mm, with a

prolonged dry season from November to May. The soil

water potential (Ws) during the driest period (February–

April) is -1.26 MPa (±0.318) and during the wettest

month (September) is -0.01 MPa (±0.01). The D. edule

population is distributed in the oak forest–submontane

scrub ecotone associated with Quercus polymorpha

Schltdl. and Cham., Q. laeta Liebm. (Fagaceae), Flouren-

sia laurifolia DC. (Asteraceae), Randia laetevirens Standl.

(Rubiaceae), Dodonaea viscosa Jacq. (Sapindaceae) and

Acacia angustissima (Mill.) Kuntze (Fabaceae).

To assess whether seeds were viable a float test in water

was applied (Calonje et al. 2011). Full seeds were placed in

plastic containers using a commercial germination mix

(peat, perlite and limestone, 3:1:1), with available soil

moisture maintained at -0.02 MPa (±0.0001), and

exposed to light under a 60 % black shade cloth at

144.1 lmol m-2 s-1 (±8.9) in a glasshouse. Any seed

with at least 2 mm of radicle growth was considered to

have germinated, and every ten days were recorded and

tagged until the end of the experiment.

Four weeks after the last seed germinated, the seedlings

tagged with monthly GD from December to April were

transplanted to black polyethylene bags containing 0.8 kg

of a mix of clay and coarse sand (3:1), housed under

glasshouse conditions and irrigated to field capacity.

Four weeks after the seedlings were transplanted, the

substrate was irrigated with 330 cm3 of water per plant,

and each bag was covered with plastic film to avoid

evaporation. Daily measurements of photosynthetic photon

flux density (PPDF), temperature, and relative humidity

were registered with a quantum light sensor and with a

temperature/relative humidity sensor data logger (Watch-

Dog 450, Spectrum Technologies, IL, USA). The mean

PPFD, air temperature and relative humidity ± SE mea-

sured at noon every day from June to September 2008 were

171.5 ± 2.2 lmol m-2 s-1, 33.5 ± 0.6 �C and 35.9 ±

1.7 %, respectively. The light intensity conditions simu-

lated those of a forest understory with a low PPFD

(100–200 lmol m-2 s-1).

The soil moisture was calculated gravimetrically using

the equation [(wet soil weight ? bag weight) - (soil dry

weight ? bag weight)/(soil dry weight)] 9 soil density.

The soil water potential (Ws) was calculated with a soil

water characteristic curve generated in the laboratory for

the substrate with a digital soil moisture tester (KS-D1,

Delmhorst Instrument Co., NJ, USA).

Seedlings tagged with monthly GD ranging from

December to April were assessed in the beginning of

the experiment (Ws = 0.0 MPa), 40 days later (Ws =

J Plant Res

123

-0.1 MPa), 90 days later (Ws = -1.0 MPa), 130 days

after (Ws = -1.5 MPa) and a control 130 days later

(Ws = 0.0 MPa) in order to compare hydrated versus

dehydrated seedlings at the end of the experiment and

exclude differences attributable to development of leaves.

The experimental design was completely randomized, with

three seedlings per GD and Ws.

Leaf area

Leaf area was measured by applying the WinFOLIA

(Regent Instruments Inc., Canada) image analysis software

program to scanned images of the leaves taken before

sampling.

Photosynthetic pigments

Chlorophyll a (Chl a), chlorophyll b (Chl b) and carote-

noids (Car) concentrations were calculated according to the

equation proposed by Lichtenthaler and Welburn (1983).

The chlorophyll a/chlorophyll b (Chl a/b) and chlorophyll

a ? b/carotenoids (Chl a ? b/Car) ratios were also cal-

culated. The average of three readings from leaf extracts

was analyzed with a visible range spectrophotometer

(Genesys 20 Model 4001/4, Thermo Spectronic, NY, USA)

at a wavelength of 470 nm for Chl a, 647 nm for Chl b and

663 nm for Car.

CAM activity

CAM activity was measured as DH? by calculating the

total titratable acids content according to the method pro-

posed by Vovides et al. (2002) for D. edule plants, which

involves collecting leaf samples at 6:00 a.m. (dark) and

7:00 p.m. (light) to calculate the titratable acidity of mac-

erated tissue with NaOH 0.01 N at pH 8. The mean of three

titrations per individual was thus determined.

Stomatal conductance (gs)

The gs measurements were performed with a leaf porometer

(SC-1, Decagon Devices, WA, USA) every 3 h over a 24 h

period. Leaflets were narrower than the diameter of the

cuvette; thus, two to three leaflets were inserted side by side

with no gap or overlap (Marler and Willis 1997). The stomatal

conductance of April GD plants with a Ws of 0.0 MPa was not

measured because the soft texture of the leaves would have

been damaged by the porometer sensor head.

Leaflet anatomy

One section of a leaflet was sampled in the middle leaf

region of December–April GD plants with Ws of 0.0 MPa,

fixed in 3 % glutaraldehyde solution and embedded in

glycol-methacrylate (Technovit� 7100, Heraeus Kulzer,

DE, USA). Transverse sections measuring 2 lm thick were

cut with a rotatory microtome (RM2125RT, Leica Micro-

systems, DE, USA), stained with 0.6 % brilliant cresyl blue

(Ruzin 1999) and observed under a light microscope (DM

2000, Leica Microsystems, DE, USA). Sections from the

same sample were slowly dried in a vacuum desiccator at

room temperature. The samples were mounted on double-

sided carbon tape on metal pins and sputtered with gold.

Stomata were observed under a scanning electron micro-

scope (Quanta 200, FEI, OR, USA).

Stomatal density was measured observing in five fields

of an epidermal impression of the leaflet adaxial surface

under a light microscope. The impression was made by

spreading a thin layer of nail varnish and removed after dry

by attaching clear sticky tape to it, peeling it from the

leaflet surface and sticking it to the slide.

Statistical analyses

Data analyses were performed using a two-way analysis of

variance (ANOVA) to examine the effects of GD, Ws and

their interaction on leaf area, DH? and the photosynthetic

pigment content (Chl a, Chl b, Car) and ratios (Chl a/b, Chl

a ? b/Car) for different Ws (0.0, -1.5 MPa). Leaflet sto-

matal density data analysis was performed with one-way

ANOVA to study the influence of GD. Differences among

the means were evaluated using Tukey’s test (P \ 0.05).

Linear regressions were performed with Ws as a predictor

of dark total H?, and stomatal density as a predictor of

stomatal conductance (gs) at 13:00, when they presented

the highest value throughout the day, and all variables were

previously transformed to their natural logarithm forms

except Ws.

The gs data analysis was performed with a two-way

repeated measures ANOVA, and because the assumption

of sphericity was violated, the degrees of freedom were

corrected with the Huynd–Feldt correction. Differences

among means were evaluated using the Bonferroni test

(P \ 0.05). All statistical analyses were performed with

XLSTAT software (2013.01.1 v, Addinsoft, NY, USA).

Results

Leaf area

No significance was indicated when analyzing the inter-

action between both of the factors for leaf area (n = 75,

d.f. = 16, F = 0.36, P [ 0.05). Although there was a

tendency for the December and January GD seedlings to

increase in leaf area, even though the soil water potential

J Plant Res

123

(Ws) diminished, and there was no difference from the

control. However, in February–April seedlings the leaf

area diminished when Ws diminished (-1.0 and

-1.5 MPa) relative to Ws control. Seedlings of the

December GD exhibited the largest leaf area and those

from April the smallest (n = 75, d.f. = 4, F = 9.03,

P \ 0.0001) (Fig. 1). Moreover, the leaf area of seedlings

growing with a Ws of 0.0 MPa were the smallest, and those

of the control at the end of the experiment were the largest

(n = 75, d.f. = 4, F = 7.73, P \ 0.0001) (Fig. 1).

Photosynthetic pigments

The effect of the interaction between the GD and Ws fac-

tors with respect to the photosynthetic pigment content in

leaves revealed that there was a statistically significant

difference for Chl a (n = 45, d.f. = 8, F = 43.58,

P \ 0.0001) and Car (n = 45, d.f. = 8, F = 317.28,

P \ 0.0001). The highest contents of Chl a and Car were

observed in the February–March GD seedlings with a Ws

of -1.5 MPa and in the January seedlings with the control;

the lowest contents were observed in the January–

December seedlings with Ws of 0.0 MPa and control for

Car and in the April seedlings with a Ws of -1.5 MPa for

Chl a (Table 1). Chl b also presented significant differ-

ences, and the highest content was observed in the April

seedlings with a Ws of -1.5 MPa (n = 45, d.f. = 8,

F = 78.78, P \ 0.0001) (Table 1).

Slight differences in the Chl a/b ratio over the range of

2.7–3.5 were observed both between the GD and Ws

(n = 45, d.f. = 8, F = 12.97, P \ 0.0001) (Fig. 2a), as

well as differences in the Chl a ? b/Car ratio over the

range of 2.7–3.8 (n = 45, d.f. = 8, F = 3.79, P \ 0.005)

(Fig. 2b). The highest value of Chl a/b was observed in the

March seedlings with a Ws of 0.0 MPa and the lowest in

Fig. 1 Leaf area by germination date (GD) and soil water potential

(Ws). Means are significantly different (P \ 0.05) if they do not share

the same letter (GD) and if they present asterisk (Ws)

Ta

ble

1L

eaf

chlo

rop

hy

lla

,ch

loro

ph

yll

ban

dca

rote

no

idco

nte

nt

inse

edli

ng

sb

yse

edG

Dan

dso

ilw

ater

po

ten

tial

See

dg

erm

inat

ion

gat

e(G

D)

So

ilw

ater

po

ten

tial

(Ws)

Ch

loro

ph

yll

a(l

gg

-1

fw)

Ch

loro

ph

yll

b(l

gg

-1

fw)

Car

ote

no

ids

(lg

g-

1fw

)

0.0

MP

a-

1.5

MP

aC

on

tro

l0

.0M

Pa

-1

.5M

Pa

Co

ntr

ol

0.0

MP

a-

1.5

MP

aC

on

tro

l

Dec

emb

er1

19

.7±

0.9

ab

13

6.8

±2

.6ab

10

1.9

±0

.8ab

42

.1±

0.5

bc

46

.3±

4.2

bc

33

.4±

0.8

c4

5.4

±0

.1e

51

.9±

0.6

d3

6.7

±0

.3f

Jan

uar

y9

1.5

±0

.4b

15

3.9

±0

.2ab

17

5.7

±1

.0ab

33

.3±

0.7

c5

3.9

±0

.5bc

57

.1±

0.6

bc

43

.1±

0.5

e6

1.1

±0

.1c

69

.2±

0.3

b

Feb

ruar

y1

36

.0±

1.5

ab

17

9.6

±0

.8a

14

4.2

±2

.4ab

61

.7±

0.5

b5

7.2

±0

.2bc

44

.6±

9.5

bc

63

.0±

0.7

c6

9.1

±0

.2b

52

.4±

0.2

d

Mar

ch1

37

.6±

1.0

ab

17

5.7

±1

.3ab

13

0.6

±0

.2ab

39

.2±

1.2

bc

59

.9±

0.6

bc

43

.8±

0.4

bc

54

.3±

0.6

d7

2.3

±0

.6a

51

.0±

1.2

d

Ap

ril

11

9.7

±0

.9ab

34

.8±

0.4

c1

01

.9±

0.8

ab

36

.8±

0.1

c1

44

.0±

0.9

a5

3.2

±0

.9bc

46

.4±

0.1

e4

8.1

±0

.5e

59

.6±

0.2

c

Mea

ns

are

sig

nifi

can

tly

dif

fere

nt

(P\

0.0

5)

ifth

eyd

on

ot

shar

eth

esa

me

lett

erin

the

sup

ersc

rip

ts

J Plant Res

123

the April seedlings with a Ws = -1.5 MPa; moreover, the

lowest value for seedlings containing a Ws of 0.0 MPa was

observed in the February seedlings, and no differences

were observed relative to the control. The highest values of

Chl a ? b/Car were observed in the April seedlings with a

Ws of -1.5 MPa and the December and April seedlings

with the control, while the lowest values were observed in

the January–February seedlings with a Ws of 0.0 MPa.

CAM activity

Positive values of DH? were observed in the seedlings

from December to April GD with a Ws of -1.0 and

-1.5 MPa; the January seedlings with a Ws of -1.5 MPa

exhibited negative DH? values (Fig. 3a), indicating an

increase in acidity during the night when the Ws reached

-1.0 MPa. However, the DH? difference between the

March and April seedlings containing a Ws of -1.5 MPa

was statistically significant (n = 60, d.f. = 12, F = 3.38,

P \ 0.005). Interestingly, the seedlings from the April GD

showed positive values when the Ws was -0.1 MPa

(Fig. 3a).

Approximately 58–80 % of the dark H? fluctuation is

explained by the Ws for the seedlings from February

(n = 12, d.f. = 1, F = 14.10, P \ 0.004, R2 = 0.585),

March (n = 12, d.f. = 1, F = 38.94, P \ 0.0001,

R2 = 0.796) and April (n = 12, d.f. = 1, F = 23.80,

P \ 0.0001, R2 = 0.704) (Fig. 3b). As the Ws diminished,

the dark H? increased twice its original value by the end of

the experiment in the April seedlings, by a factor of 1.5 in

the March seedlings and by 1.2 in the February seedlings,

increasing largely at a Ws of -1.0 MPa. In the December

(n = 12, d.f. = 1, F = 0.989, P [ 0.5, R2 = -0.1) and

January seedlings (n = 12, d.f. = 1, F = 0.676, P [ 0.5,

R2 = 0.430), the Ws had a slight effect on the dark H?

fluctuation (Fig. 3b).

Stomatal conductance (gs)

There was no statistical significance indicated when ana-

lysing the interaction among gs time measurements, GD

and Ws (n = 75, d.f. = 77.16, F = 1.30, P [ 0.05), but

there was a difference between gs time and GD (n = 75,

d.f. = 19.28, F = 2.23, P \ 0.05) for the February and

April seedlings, which presented the highest values of gs,

154 and 120 mmol m-2 s-1, at 10:00. When Ws was

-1.0 MPa the observed gs value was 80 mmol m-2 s-1,

Fig. 2 Chlorophyll a/b ratio a and chlorophyll a ? b/carotenoids

b by GD and soil water potential (Ws). Means are significantly

different (*P \ 0.05)

Fig. 3 CAM activity a and semi-log representation of dark H? b by

GD and soil water potential (Ws). Means are significantly different

(*P \ 0.05)

J Plant Res

123

the highest value recorded at 7:00, 10:00 and 13:00

(n = 75, d.f. = 19.28, F = 5.35, P \ 0.0001) (Fig. 4).

Association between stomatal conductance and gs

recorded at 13:00 was significant only when Ws was

-1.5 MPa, showing higher gs at lower stomatal density

(Table 2).

Leaflet anatomy

The first seedling leaf appeared 4–5 weeks after germina-

tion, completely expanded after 8 to 10 weeks and matured

2–3 weeks later. At the beginning of the experiment, the

December–February GD seedlings presented a fully

expanded leaf, that was coriaceous in texture and dark

green in color (Fig. 5a), whereas the leaves of the March

GD were smaller, not completely coriaceous and lighter

green. The April seedling leaves were the smallest; the

leaflets remained clustered in the apex and spread in the

base and were soft in texture and reddish light green

(Fig. 5b).

Along the transverse section, a leaflet epidermis was

composed of a single layer of thick-walled, oval or circular

cells in the December–March GD seedlings (Fig. 5c) and

thin-walled cells in the April seedlings (Fig. 5d), covered

by a thick cuticle and interrupted by stomata in the abaxial

epidermis (hypostomatic leaflet). The hypodermis occurred

only adaxially and was completely developed in leaflets of

the December–February seedlings, with one layer of thick-

walled, oval to circular sclerenchymatous cells (Fig. 5c). In

the March seedlings, the leaflet structure was not contin-

uous beneath the epidermis, and in the April seedlings, the

structure was not well developed (Fig. 5d). Mesophyll is

composed of single-layered palisade parenchyma beneath

the adaxial epidermis, and spongy parenchyma composed

of isodiametric to elongated cells, with large narrow

intercellular spaces (Fig. 5c). In leaflets of the April

seedlings, the palisade parenchyma was not completely

differentiated and spongy parenchyma cells were mostly

isodiametric with scarce small intercellular spaces

(Fig. 5d). Both the palisade and the mesophyll parenchyma

contained chloroplasts. No transfusion tissue was observed.

Several parallel vascular bundles were present, enclosed by

a sheath of slightly thickened parenchyma cells. Few

Table 2 Regression results for log gs (mmol m-2 s-1) at 13:00 h by

Ws

Ws Constant log stomatal

density (no. cm-2)

B r2

0.0 MPa 5.23 -0.32n.s. -0.31 0.10

-0.1 MPa 10.15 -0.92n.s. -0.50 0.26

-1.0 MPa 19.61 -1.92n.s. -0.67 0.45

-1.5 MPa 17.23 -1.62** -0.80 0.65

Control 5.60 -0.27n.s. -0.14 0.02

n.s. P [ 0.05, ** P \ 0.05

Fig. 4 Stomatal conductance

(gs) by soil water potential (Ws)

and GD. Shaded area = dark

period. Means are significantly

different (*P \ 0.05)

J Plant Res

123

sclerenchyma cells were scattered in the mesophyll or

vascular tissue region, but these were absent in the leaflets

of April seedlings. Conspicuous mucilage canals were

observed adaxially opposite to vascular bundles. Stomata

were sunken, with heavily lignified, thick-walled guard

cells, and overarched by subsidiary cells forming the epi-

stomatal chamber. A thick cuticle was observed up to the

central pore, and the substomatal chamber was lined with

Fig. 5 Seedling extended

leaflets when Ws was 0.0 MPa

(a), clustered leaflets in the apex

when Ws was -1.5 MPa (b).

Light microscope transverse

section (TS) of leaflet blade

mature (c) and developing (d).

TS of stoma mature (e), and

developing (f); scanning

electron microscope (SEM)

abaxial view of stoma with

epicuticular waxes mature with

a wide open (g), developing

with a narrow open (h).

c cuticle, e epidermis, ec

epistomatal chamber,

h hypodermis, m mucilage

canal, n encircling cell,

o occlusive cell, pp palisade

parenchyma, s stomata, sb

subsidiary cell, sc substomatal

chamber, sp spongy

parenchyma, v vascular bundle.

Scale bars = 100 lm for c, d,

30 lm for e, f and 5 lm for g, h

J Plant Res

123

spongy mesophyll cells (Fig. 5e). However, in March–

April seedlings, the stomata were still developing (Fig. 5d),

with thin-walled guard and subsidiary cells and a wide

lumen and a thin cuticle observed near the central pore

(Fig. 5f); the dorsal walls of guard cells were observed to

have lignified before the ventral walls. Subsidiary cells

were partially overlapped by two concentric rings of

encircling cells, which were raised to form a rim at the

same level as other epidermal cells, covered with a thick

cuticle and epicuticular wax tubules (Fig. 5g). The stoma

of the March–April seedlings showed only one ring of

raised encircling cells, but they were thin-walled and not

completely lignified, forming a narrow opening edge

(Fig. 5f, h).

Significant differences were observed among values of

leaflet stomatal density (n = 25, d.f. = 4, F = 15.41,

P \ 0.0001). The December–January GD seedlings pre-

sented the highest stomatal density, followed by the Feb-

ruary–March seedlings and the lowest value in the April

seedlings (Fig. 6).

Discussion

Dioon edule seedlings followed the developmental mor-

phology of normal cycad leaves from emergence to full

maturation (Norstog and Nicholls 1997). The GD affected

leaf area because there were differences in leaf structure

due to time elapsed for full development as control

showed. The effect of the Ws treatments was independent in

the December–January GD seedlings because the plants

continued growing during the experiment and the leaves

almost reached their mature size, only slightly reducing in

volume due to the thick-cuticle and the thick-walled epi-

dermal cells (De Micco and Aronne 2012). However, the

February–April GD seedlings showed smaller leaf area

when the Ws was lower, suggesting that dehydration could

be caused by the lack of developed xerophytic traits

because a sclerified leaf sustain lower water potentials

before the structure collapses and the xylem cavitate

(Hallik et al. 2009).

Differences in photosynthetic pigments may be associ-

ated with the mature functional array (Norstog and Nich-

olls 1997). Compared to Cycas revoluta (Cycadaceae),

whose adult sun leaves have been reported to possess a

total chlorophyll content of 3,050 lg g-1 and total yellow

Car content of 445 lg g-1 (Ida 1981), D. edule seedlings

are considerably inferior and share more in common with

Lepidozamia peroffskyana (Zamiaceae) in terms of typical

shade Chl a, Chl b and Chl a ? b/Car values.

In this study, the contents of Chl a, Chl b and Car

increased at Ws of -1.5 in most of the GD seedlings, which

may be caused by the fact that they were exposed to low

light and water stress, conditions favorable to increase the

chlorophyll content (Pirzad et al. 2011) as well as Car

required for the correct assembly of photosystems (Ca-

zzonelli 2011) enhancing their light harvesting at low light

levels.

The December–March GD and control seedlings pre-

sented Chl a/b ratios characteristic of plants growing in

environments with reduced PPFD (Lichtenthaler and

Burkart 1999) like understory trees in tropical rainforest

(Cao 2000). However, it is common for gymnosperms to

exhibit sun leaves with Chl a/b ratios ranging 2–3, par-

ticularly Cycas revoluta (2.9) (Ida 1981) and Lepidozamia

peroffskyana (2.1–3.3) (Medeghini Bonatti and Baroni

Fornasiero 1990). The low Chl a/b ratio of the April GD

plants when the soil water deficit was severe might be

attributed to water stress because Chl a is destroyed more

rapidly than Chl b by water stress when plants are vul-

nerable to drought (Cao 2000). These results suggest that

the April GD seedlings presented reduced photosynthetic

capacity and probably influenced the leaf area.

A Chl a ? b/Car weight ratio suggests that leaves might

be more yellowish green (Lichtenthaler 1987), but Lepi-

dozamia peroffskyana (Zamiaceae) showed a wide range of

values (Medeghini Bonatti and Baroni Fornasiero 1990)

according to leaf development stage. However, during the

slow leaf maturation process of D. edule, as in Ceratoz-

amia spp. (Zamiaceae), the photosynthetic apparatus pre-

sented ‘‘secondary’’ Car not involved in photosynthetic

processes until they were completely green and contained

the normal ‘‘primary’’ Car involved in photosynthetic

activity (Cardini and Bonzi 2005).

Dioon edule adult plants are CAM cycling (Reinert and

Blankenship 2010; Vovides et al. 2002), and this occurs in

plants that have a high stomatal conductance. They exhibit

normal C3 photosynthesis during the day but refix the dark

respiration of CO2 at night and have a diurnal fluctuation in

Fig. 6 Stomatal density by GD. Means are significantly different

(*P \ 0.05)

J Plant Res

123

the concentration of malic acid (Lambers et al. 2008). The

December–March seedlings exhibited C3 photosynthesis,

but shifted to CAM cycling when substrate reached a Ws of

-1.0 MPa as observed in adult plants, but with higher

concentrations than those reported in previous studies

(Vovides et al. 2002). However, the April seedlings

exhibited CAM cycling at higher Ws because underdevel-

oped leaf xeromorphic traits may have caused higher stress.

Although it has been suggested that at lower temperatures,

lower water stress and lower light intensity, D. edule adult

plants do not present CAM cycling (Vovides et al. 2002),

this experiment showed that seedlings under water stress

and low light intensity presented CAM cycling.

Seedlings with leaflets exhibiting fully developed

structural xeromorphic traits (December–January GD)

showed only slight increases in dark H?, whereas leaflets

with underdeveloped xeromorphic traits (February–April

GD) showed exponential increases in dark H? when Ws

decreased. Interestingly, some CAM plants have shown

increases in different organic acid content with increases in

osmotic pressure and, as a consequence, store a large

amount of water, making them less resistant to frost and

heat (Losch 1984).

The stomatal conductance of all the GD seedlings at

higher Ws showed a single mid-day peak characteristic of

C3 plants that are well supplied with water. Meanwhile, the

March–April GD seedlings at the lowest Ws showed

morning and late-afternoon peaks and mid-day partial

stomatal closure during the time of the day when the

evaporative demand was highest, characteristic of plants

growing in a water-limited environment (Lambers et al.

2008). When the Ws was -1.0 MPa, the gs increased near

the end of the night in all seedlings, as observed in other

CAM cycling species such as Talinum (Portulacaceae) and

Peperomia camptotricha (Piperaceae), which is explained

by the accelerated rate of malic acid formation toward the

end of the night (Harris and Martin 1991).

Some studies have reported that adult plants showing

mature dark foliage exhibit higher gas exchange values than

when fully expanded but more immature light-green foliage

(Marler and Willis 1997), possibly because stomata and

mesophyll are not full developed. However, in this experiment

there were no significant differences between the mature and

the immature leaves of the seedlings when the Ws was higher,

and low values were maintained, i.e., 32 % gs on average

reported for cultivated adult plants of D. edule growing in

Florida (155 mmol m-2 s-1) (Marler and Willis 1997).

The stomatal anatomy coincides with descriptions of

Cycas circinalis (Rao 1972) and Dioon spinulosum (Tang

et al. 2004). The sunken stomata of D. edule leaflets pro-

vide an additional advantage against drought by decreasing

gs, although the rate of CO2 influx is also reduced, and by

the creation of special microclimatic conditions such as

higher humidity directly above the stomatal pore (Roth-

Nebelsick 2007). The epistomatal chamber, with a depth of

20 lm, decreased the conductance by 30 % (Roth-Nebel-

sick 2007), and D. edule seedling stomata presented

chambers with an average depth of 20.34 lm; however,

seedlings from the March–April GDs showed less devel-

oped chambers, resulting in higher values of gs when water

stressed.

Stomatal density of seedlings was lower than reported

for adult D. edule but similar to other cycad species (Vo-

vides et al. 2002). The stomatal density presented a nega-

tive association with gs only when severe drought was

present (Ws of -1.5 MPa), although there is not a clear

relationship between stomatal density and gs (Camargo and

Marenco 2011).

Our experiment revealed the effect of GD and soil water

deficit (Ws) on seedling tolerance to water stress, helping to

understand the most relevant seedling establishment limi-

tation identified for the species D. edule. Remarkably D.

edule seedlings from all GDs are able to shift from C3 to

CAM photosynthetic pathway triggered at Ws of

-1.0 MPa, like adult plants. Particularly, the March–April

GD seedlings (61 %) with less developed xeromorphic

traits during the drought season (Ws of -1.0 to -1.5 MPa)

would have higher probability of death by dehydration.

Even less developed leaves have higher positive values of

DH?, they show midday partial stomatal closure, reduced

sclerification that may not sustain lower water potentials

and decreased photosynthetic capacity.

The seedling mortality has a relevant impact on population

structure and our results suggest that during prolonged and

severe drought, a low proportion of seedlings that have

developed completely the xeromorphic traits in the leaf

increase their probability of survival to secure the cohort. For

restoration purposes it is recommended to use plants with

xeromorphic traits of the leaf completely developed and

planting after the drought season to increase survival.

Acknowledgments This work was supported by The Cycad Soci-

ety, Inc. and PROMEP—SEP [UASLP-PTC-179]. Mexican govern-

ment authorized cone collection [SGPADGVS/00507/08]. We thank

NTBG College Biology Professors’ Fellowship for the basic knowl-

edge in cycads; Pablo Delgado provided assistance with SEM; Isaac

Jacob Chavez Acuna provided environmental data; Raymundo Mora

and Efraın Hernandez provided assistance in the field and laboratory.

References

Alvarez-Yepiz JC, Dovciak M, Burquez A (2011) Persistence of a

rare ancient cycad: effects of environment and demography. Biol

Conserv 144:122–130

Calonje M, Kay J, Griffith MP (2011) Propagation of cycad

collections from seed. Applied reproductive biology for conser-

vation. Sibbaldia 9:77–96

J Plant Res

123

Camargo MAB, Marenco RA (2011) Density, size and distribution of

stomata in 35 rainforest tree species in Central Amazonia. Acta

Amaz 41:205–212

Cao KF (2000) Leaf anatomy and chlorophyll content of 12 woody

species in contrasting light conditions in a Bornean heath forest.

Can J Bot 78:1245–1253

Cardini F, Bonzi LM (2005) Carotenoid composition and its

chemotaxonomic significance in leaves of ten species of the

genus Ceratozamia (Cycads). J Plant Physiol 162:517–528

Cazzonelli CI (2011) Carotenoids in nature: insights from plants and

beyond. Funct Plant Biol 38:833–847

De Micco V, Aronne G (2012) Morpho-anatomical traits for plant

adaptation to drought. In: Aroca R (ed) Plant responses to

drought stress, from morphological to molecular features.

Springer, Berlin, pp 37–61

Hallik L, Niinemets U, Wright IJ (2009) Are species shade and

drought tolerance reflected in leaf-level structural and functional

differentiation in Northern Hemisphere temperate woody flora?

New Phytol 184:257–274

Harris FS, Martin CE (1991) Correlation between CAM-cycling and

photosynthetic gas exchange in five species of Talinum (Port-

ulacaceae). Plant Physiol 96:1118–1124

Ida K (1981) Eco-physiological studies on the response of Taxodia-

ceous conifers to shading, with special reference to the behaviour

of leaf pigments distribution of carotenoids in green and

autumnal reddish brown leaves of gymnosperms. Bot Mag

(Tokyo) 94:41–54

Lambers H, Chapin FS, Pons TL (2008) Plant physiological ecology,

2nd edn. Springer, New York

Lichtenthaler HK (1987) Chlorophylls and carotenoids: pigments of

photosynthetic membranes. Methods Enzymol 148:350–382

Lichtenthaler HK, Burkart S (1999) Photosynthesis and high light

stress. Bulg J Plant Physiol 25:3–16

Lichtenthaler HK, Wellburn AR (1983) Determinations of total

carotenoids and chlorophylls a and b of leaf extracts in different

solvents. Biochem Soc T 11:591–592

Lopez-Ovando A, Trevino-Garza EJ (2008) Reproduction of chamal

(Dioon edule Lindley) by seed. Ra Ximhai 4:45–55

Losch R (1984) Species-specific responses to temperature in acid

metabolism and gas exchange performance of Macronesian

Sempervivoideae. In: Margaris NS, Arianoustou-Farragitaki M,

Oechel WC (eds) Being alive on land. Tasks for vegetation

science, vol 13. Dr. W. Junk, The Hague, pp 117–126

Marler TE, Willis LE (1997) Leaf gas-exchange characteristics of

sixteen cycad species. J Am Soc Hortic Sci 122:38–42

Medeghini Bonatti P, Baroni Fornasiero R (1990) Developmental

pattern and structural organization of leaf chloroplasts in

Lepidozamia peroffskyana. Aust J Bot 38:53–62

Mora R, Yanez-Espinosa L, Flores J, Nava-Zarate N (2013) Strobilus

and seed production of Dioon edule (Zamiaceae) in a population

with low seedlings density in San Luis Potosı, Mexico. Trop

Conserv Sci 6:268–282

Norstog KJ, Nicholls TJ (1997) The biology of the cycads. Cornell

University Press, Ithaca

Pirzad A, Reza Shakiba M, Zehtab-Salmasi S, Abolghasem Moham-

madi S, Darvishzadeh R, Samadi A (2011) Effect of water stress

on leaf relative water content, chlorophyll, proline and soluble

carbohydrates in Matricaria chamomilla L. J Med Plant Res

5:2483–2488

Rao LN (1972) Life history of Cycas circinalis. V. Seedling anatomy.

Proc Indian Acad Sci Sect B 76:47–57

Reinert F, Blankenship R (2010) Evolutionary aspects of crassulacean

acid metabolism. Oecol Aust 14:359–368

Roth-Nebelsick A (2007) Computer-based studies of diffusion

through stomata of different architecture. Ann Bot 100:23–32

Ruzin SE (1999) Plant microtechnique and microscopy. Oxford

University Press, Oxford

Silvera K, Neubig KM, Whitten WM, Williams NH, Winter K,

Cushman JC (2010) Evolution along the crassulacean acid

metabolism continuum. Funct Plant Biol 37:995–1010

Tang YJ, Liu N, Liao JP, Xie ZY, Wu QG, Chen JR (2004)

Systematics implications of pinna venation and pinna anatomy in

Zamiaceae. Acta Phytotax Sin 42:365–374

Vovides AP (1990) Spatial distribution, survival and fecundity of

Dioon edule (Zamiaceae) in a tropical deciduous forest in

Veracruz, Mexico, with notes on its habitat. Am J Bot

77:1532–1543

Vovides AP, Etherington JR, Dresser PQ, Groenhof A, Iglesias C,

Flores Ramırez J (2002) CAM-cycling in the cycad Dioon edule

Lindl. in its natural tropical deciduous forest habitat in central

Veracruz, Mexico. Bot J Linn Soc 138:155–162

Vovides AP, Perez-Farrera MA, Gonzalez-Astorga J, Gonzalez D,

Gregory T, Chemnic J, Iglesias C, Octavio-Aguilar P, Avendano

S, Barcenas C (2003) An outline of current knowledge of

Mexican cycads (Zamiaceae, Cycadales). Curr Top Plant Biol

4:159–174

J Plant Res

123