Alicia J. Manfre1,3, Gabrielle A. LaHatte1,4, Cynthia R. Climer1 and William R. Marcotte Jr.1,2,*
1Department
2Department
of Genetics & Biochemistry, 100 Jordan Hall, Clemson University, Clemson, SC 29634, USA
of Biological Sciences, 132 Long Hall, Clemson University, Clemson, SC 29634, USA
The end of orthodox seed development is typified by
a developmentally regulated period of dehydration
leading to the loss of bulk water from the entire structure.
When dehydration occurs, the cytoplasm condenses
and intracellular components become more crowded,
providing an environment amenable to numerous
undesirable interactions that can lead to protein
aggregation, denaturation and organelle–cell membrane
fusion. Acquisition of desiccation tolerance, or the
ability to withstand these very low water potentials and
consequent molecular crowding, has been correlated
with the accumulation of various protective compounds
including proteins and sugars. Among these are the
late embryogenesis abundant (LEA) proteins, a diverse
class of highly abundant, heat-stable proteins that
accumulate late in embryo maturation coincident with
the acquisition of desiccation tolerance. Previous work
led us to hypothesize that the protein ATEM6, one of
the two Arabidopsis thaliana group 1 LEA proteins, is
involved in regulating the rate at which water is lost
from the maturing embryo; homozygous atem6-1
mutants display premature dehydration of seeds at
the distal end of the silique. Here we demonstrate that
rehydrated, mature seeds from atem6-1 mutant plants
lose more water during subsequent air drying than wildtype seeds, consistent with a role for ATEM6 protein
in water binding/loss during embryo maturation.
In addition, and possibly as a result of premature
dehydration, mutant seeds along the entire length of
the silique acquire desiccation tolerance earlier than
their wild-type counterparts. We further demonstrate
Regular Paper
Seed Dehydration and the Establishment of Desiccation
Tolerance During Seed Maturation is Altered in the
Arabidopsis thaliana Mutant atem6-1
precocious, and perhaps elevated, expression of the
other A. thaliana group 1 LEA protein, ATEM1, that may
compensate for loss or ATEM6 expression. However, this
observation could also be consistent with acceleration of
the entire normal maturation program in atem6-1 mutant
embryos. Interestingly, ATEM6 protein does not appear
to be required in mature seeds for viability or efficient
germination.
Keywords: Arabidopsis • Dehydration • Desiccation tolerance
• Germination • LEA protein • Seed maturation.
Abbreviations: ANOVA, analysis of variance; daf, days after
flowering; GFP, green fluorescent protein; HSP, heat shock
protein; LEA, late embryogenesis abundant; LSD, least
significant difference; MS, Murashige and Skoog; PEG,
polyethylene glycol; PVDF, polyvinylidene fluoride.
Introduction
Plant embryo development can conveniently be divided
into two main phases, embryogenesis proper and maturation. During embryogenesis proper, important developmental cues lead to morphogenetic events that result in
specification of the basic body plan and establishment of
all the tissue types that will be found in the mature seed.
This is followed by a period of maturation during which
the body plan is elaborated, and storage compounds, that
will serve as an energy source for the germinating seedling
prior to photosynthetic competence, are accumulated. It
is also during the later stages of the maturation phase that
3Present
address: USDA-ARS-AFRS, 2217 Wiltshire Rd, Kearneysville WV 25430, USA.
address: Case Western Reserve University School of Law, 11075 East Boulevard, Cleveland, OH 44106, USA.
*Corresponding author: Email, [email protected]; Fax, +864-656-0119.
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185, available online at www.pcp.oxfordjournals.org
© The Author 2008. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.
All rights reserved. For permissions, please email: [email protected]
4Present
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
243
A. J. Manfre et al.
the embryo acquires desiccation tolerance and undergoes a
developmentally programmed dehydration event leading to
dormancy and a quiescent state.
During desiccation, mature orthodox seeds reach levels
of 5–10% water and can frequently be dried further to
1–5% water with little or no loss of viability. When dehydration occurs, the cytoplasm condenses and the intracellular components become more crowded. These conditions
provide an environment for numerous undesirable interactions that can lead to protein aggregation and denaturation
as well as organelle–cell membrane fusion (Hoekstra et al.
2001). Acquisition of desiccation tolerance, or the ability to
withstand these very low water potentials and subsequent
molecular crowding, has been correlated with the accumulation of various protective compounds including proteins
(Bartels et al. 1988, Vertucci and Farrant 1995, Black et al.
1999) and sugars (Sun et al. 1994, Black et al. 1999, Buitink
et al. 2000).
As the embryo prepares for desiccation, there is a shift
in carbohydrate accumulation from simple to more complex sugars. Sucrose, raffinose and stachyose are present at
low levels early in embryo development, but accumulate
to much higher levels just prior to desiccation (Ooms et al.
1992). The ability of sugars to form superviscous fluids, particularly the non-reducing sugars found in mature seeds, was
the original foundation for the prediction that intracellular
glasses contribute to desiccation tolerance in plants (Burke
1986, Williams and Leopold 1989). More recently, numerous
studies have evaluated the properties of model glasses as
well as the glassy matrix in anhydrobiotic plants and seeds
(Koster 1991, Leopold et al. 1994, Leprince et al. 1995, Wolkers et al. 1998). The overall conclusion of these studies is that
carbohydrates cannot be the only molecules involved in vitrification (Ooms et al. 1992). Because many proteins are also
abundant in the cytoplasm, it is likely that they also play a role
in glass formation (reviewed in Buitink and Leprince 2004).
Two classes of proteins most likely to make a major contribution to cellular stability in mature seeds include the
small heat shock proteins (HSPs) and the late embryogenesis
abundant (LEA) proteins. Small HSPs, that accumulate late
in seed development (Wehmeyer et al. 1996, Wehmeyer and
Vierling 2000), may help minimize the aggregation effects
of cytoplasm condensation (Feder and Hofmann 1999) by
acting as molecular chaperones and thereby contribute to
stabilization of a glassy state. LEA proteins are a diverse class
of highly abundant, heat-stable proteins that accumulate
late in embryo maturation and during the developmentally
regulated period of dehydration at the end of seed development. Accumulation of LEA proteins has been correlated with the acquisition of desiccation tolerance (Bartels
et al. 1988), and some members also accumulate in pollen,
another anhydrobiotic structure in plants. In addition,
as LEA proteins have also been shown to abrogate protein
244
aggregation in vitro (Goyal et al. 2005), they may function
in vivo in a manner similar to that of the small HSPs. It is likely,
therefore, that aqueous glass formation involves at least
oligosaccharides, HSPs and LEAs, together maintaining cellular
stasis and avoiding the damaging consequences of dehydration (reviewed in Hoekstra et al. 2001).
LEA expression is induced by the stress-related phytohormone ABA, and may be activated by either endogenous or
exogenously applied ABA. While LEA proteins are mainly
active during seed development, many LEA genes are
also inducible in response to water stress in vegetative and
reproductive structures (Berge et al. 1989, Morris et al. 1990,
Bostock and Quatrano 1992, Wilhelm and Thomashow
1993, Bies et al. 1998). Although individual LEA proteins have
been shown to represent up to 4% of total cellular protein
(Roberts et al. 1993), the transcripts and proteins rapidly
degrade upon imbibition (Roberts et al. 1993, Bies et al.
1998). It is therefore possible that some LEAs have a dual role
during the plant life cycle, and function as a storage protein
during germination as well as in desiccation tolerance during
seed development.
The D-19 family (Baker et al. 1988), or group 1 LEAs,
were first identified in wheat by Cuming and Lane (1979).
Since then, group 1 LEAs have been identified in many other
plants including Arabidopsis thaliana (Gaubier et al. 1993).
This group of proteins is typified by a highly conserved 20
amino acid signature motif that, although present only once
in many of the group 1 LEAs, may be tandemly repeated up
to four times (Espelund et al. 1992, Stacy and Aalen 1998,
Cuming 1999). Unlike some of the other, more broadly
expressed LEAs, group 1 proteins are expressed exclusively
in embryonic tissues, and are not inducible in adult tissues,
even in response to ABA (Bies et al. 1998).
Arabidopsis thaliana has two endogenous group 1 LEA
genes, ATEM1 and ATEM6 (Gaubier et al. 1993). ATEM1
expression precedes expression of ATEM6 by approximately
2 d, though mRNA and protein remain abundant for both
through maturation and into dry, mature seed. These two
genes encode very similar proteins that differ primarily
in the number of repeats in the conserved 20 amino acid
signature motif (four copies in ATEM1 vs. one copy in
ATEM6). The expression domain of ATEM1 is limited mainly
to the provascular tissues and the root tip, although there is
also some expression in the shoot meristem and vasculature
of the cotyledons (Vicient et al. 2000). ATEM6 has a much
larger expression domain, comprising essentially the entire
embryo with particularly strong expression in the shoot
meristem and provascular tissue.
Previous work in this lab identified an insertional mutation in the ATEM6 gene of Arabidopsis (Manfre et al. 2006).
The T-DNA insertion was present 2 bp 3′ of the TAA stop
codon, effectively separating the 3′-flanking region from
the remainder of the gene by >7 kb. This mutant, atem6-1,
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
contains no detectable ATEM6 protein in homozygous
mutant seed. The mutant also displays premature dehydration and maturation of seeds at the distal ends of siliques at
a time coincident with normal ATEM6 expression, suggesting that this protein may function to buffer the rate at which
water is lost from the seed. Introduction of a wild-type gene
into atem6-1 mutant plants (ATEM6-C) restores the wildtype phenotype.
To investigate further the in planta function of ATEM6 protein during seed development, we have evaluated the effect of
loss of ATEM6 expression on water loss rates in mature seeds.
In addition, we have characterized the germination capabilities of both mature and immature atem6-1 mutant seeds.
We present evidence that ATEM6 protein is not required in
mature seeds for viability and/or efficient germination. Further, we show that absence of ATEM6 results in accelerated
acquisition of desiccation tolerance in mutant seeds that may
be associated with changes in seed water loss.
A
4
mg H2O Lost mg−1 Dry Weight
Acquisition of desiccation tolerance
3
2
atem6-1
Wild-Type
1
0
0
10
20
30
40
50
60
70
80
90
Minutes of Slow Drying
Results
B
4
Seed water loss and water content assay
atem6-1
3.5
mg H2O mg−1 Dry Weight
Wild-Type
3
0.1
90 Min Dry
Hydrated
LiCl
0
Initial
Earlier work in this lab documented evidence of accelerated
seed dehydration and maturation (as evidenced by accumulation of seed coat pigmentation) of atem6-1 seeds at
the distal end of the silique (Manfre et al. 2006). Due to the
hydrophilic nature of ATEM6 protein and the fact that it is
expressed throughout maturing wild-type embryos, we postulated that ATEM6 protein might function to influence the
rate at which water is lost during seed development.
To test this hypothesis, water loss rates for fully hydrated,
mature seeds from atem6-1 and wild-type plants were determined. In addition, we assessed the water content of seeds at
different stages of treatment. Seeds were uniformly dried and
five replicates of 100 seeds each were re-hydrated overnight
at 4°C with sterile water. The seeds were then placed in an
aluminum pan and residual water removed by centrifugation (see Materials and Methods for details). Water loss from
the fully hydrated seeds was monitored by weighing on a
microbalance, with readings taken every minute for 90 min.
As can be seen in Fig. 1A, weight loss by drying is a biphasic process for both genotypes. Initial weight loss is relatively
rapid, and more than two-thirds of the total loss during the
90 min assay is achieved in the first 20 min. The rate of weight
loss is essentially identical for both genotypes during this time,
and we interpret this phase of drying to be due to the loss of
water that was absorbed by the outermost layers of the seed.
This is followed by a significantly slower rate of weight loss
that we attribute to water that is more tightly bound, most
probably water that has been taken up by the embryo during
the overnight imbibition period. Less than 0.20% of the total
weight loss occurs during the final 10 min of the assay, suggesting that any remaining water is very tightly bound.
Fig. 1 Water loss and water content assays. (A) Average weight loss
by slow drying for five replicate samples of atem6-1 mutants seeds
(open circles) and wild-type seeds (open squares). The data are
expressed as mg H2O lost per mg dry weight. (B) Average seed water
content of atem6-1 mutants seeds (gray bars) and wild-type seeds
(black bars) before and after overnight hydration and after slow
drying for 90 min. The data are expressed as mg H2O mg –1 dry weight.
Error bars in both panels represent the standard deviation.
Interestingly, and despite no difference in mature seed
water content before or after treatment in an LiCl hygrostat
(Fig, 1B, Initial and LiCl, respectively), atem6-1 seeds take
up more water than wild-type seeds (Fig, 1B, Hydrated).
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
245
A. J. Manfre et al.
On drying, both genotypes lose >97% of their water content
during the assay period, but mutant seeds lose more water
than wild-type seeds. Only part of this difference can be
accounted for by the increase in water uptake, and atem6-1
mutant seeds reach a significantly lower water content than
wild-type seeds (P < 0.001).
Polyethylene glycol germination assay
The visible phenotype of atem6-1 seeds, premature dehydration and maturation, is coincident with the time during
which wild-type embryos express ATEM6 protein. Despite
the lack of detectable ATEM6 protein, seeds from homozygous mutant plants are not only viable but germinate readily
under standard conditions. However, germination conditions
in the laboratory are usually near optimal, and any fitness
cost associated with the atem6-1 lesion may not be easily
observed. To determine if atem6-1 seeds are compromised
for germination under non-optimal conditions, germination
rates of wild-type and atem6-1 seeds were assessed on media
of decreasing water potential.
The high molecular weight solute polyethylene glycol
(PEG) 8000 was chosen to reduce the water potential of the
media based on experiments described by van der Weele
et al. (2000) and Penfield et al. (2001). Because autoclaving changes the size of the PEG polymer, all solutions were
filter sterilized before use. Five replicates of 100 seeds each
were plated on Murashige and Skoog (MS) media containing
different levels of PEG, stratified for 3 d at 4°C, and germination was scored after 5 d. The data in Fig. 2 represent the
average percentage germination. All seeds, regardless of
100
% Germination
80
60
40
20
0
0
5
10
% PEG
15
20
Fig. 2 PEG germination assay. Average percentage germination for five
replicate plates of atem6-1 mutants seeds (gray bars) and wild-type seeds
(black bars) on various concentrations of PEG. Error bars represent the
standard deviation.
246
genotype, displayed approximately 100% germination at 0,
5 and 10% PEG concentrations. The rates of germination
drop to approximately 55–75% at 15% PEG and to <25% at
the 20% PEG concentration. Although germination was significantly suppressed at 20% PEG, there was no statistically
significant difference between germination rates for wildtype and atem6-1 mutant seed (P = 0.08) at this or any other
level of PEG. These data demonstrate that loss of ATEM6 in
atem6-1 homozygous seed produces no difference in mature
seed viability or germination as compared with wild-type
seeds under the conditions tested.
Acquisition of desiccation tolerance (ADT) assay
The results presented above suggest that atem6-1 seeds
acquire essentially full desiccation tolerance even in the
absence of detectable ATEM6 protein. In fact, the visible phenotype of premature dehydration and maturation of seeds,
that is often limited to the first 4–5 seeds at the distal end
of the silique, suggests that the entire maturation program
may be altered in mutant seeds. If this is the case, mutant
seeds might be predicted to acquire desiccation tolerance
at a time different from the wild type. To examine this, and
to determine if there is a difference between plant-proximal
and plant-distal seeds during the maturation process, the
four proximal-most and four distal-most seeds from either
side of the replum of wild-type and atem6-1 plants were
dissected from siliques at 7, 10, 13 and 16 days after flowering (daf). These seeds were dried in a hygrostat for 2 d and
subsequently rehydrated. Germination was scored over 10 d
as no further germination was observed after 10 d postre-hydration. Some seeds were also plated directly (without
drying) on germination medium to assess fresh germination
rates.
Seeds placed directly on germination medium after
dissection (7, 10, 13 and 16 daf) displayed no difference in
germination rates between the two genotypes (data not
shown). There was, however, a marked difference in germination for the dried seeds. Fig. 3A and B show germination
rates for seeds dissected 10 daf, plant-proximal and plantdistal, respectively. As can be clearly seen, atem6-1 seeds
germinate much more readily than wild-type seeds in both
the proximal and distal positions, although location does
appear to have an effect. These differences are no longer
seen for seeds dissected at 13 daf when virtually all seeds
for both genotypes germinate over 10 d after rehydration (Fig. 3C, D). Essentially complete germination is also
obtained for seeds dissected from the plant-proximal end of
siliques 16 daf (Fig. 3E). Interestingly, seeds dissected from the
plant-distal end of siliques 16 daf show decreased germination rates for both genotypes (Fig. 3F). The reason for this is
not clear. No seeds dissected 7 daf germinated after drying,
regardless of genotype.
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
Acquisition of desiccation tolerance
A
B
100
100
atem6-1
Wild-Type
% Germination
% Germination
80
60
40
20
3 4 5 6 7 8 9
Days After Rehydration
10
D
100
80
% Germination
% Germination
2
60
40
20
1
2
3 4 5 6 7 8 9
Days After Rehydration
10
1
2
3 4 5 6 7 8 9
Days After Rehydration
10
1
2
3 4 5 6 7 8 9
Days After Rehydration
10
100
80
60
40
20
0
0
1
2
3 4 5 6 7 8 9
Days After Rehydration
10
F
100
80
% Germination
% Germination
40
0
1
E
60
20
0
C
80
60
40
20
100
80
60
40
20
0
0
1
2
3 4 5 6 7 8 9
Days After Rehydration
10
Fig. 3 Acquisition of desiccation tolerance (ADT) assay. Average percentage germination for dissected atem6-1 mutants seeds (gray bars) and
wild-type seeds (black bars) after drying and rehydration. (A, B) Ten daf, proximal and distal, respectively. (C, D) Thirteen daf, proximal and distal,
respectively. (E, F) Sixteen daf, proximal and distal, respectively. Error bars represent the standard deviation.
A fully fit three-way analysis of variance (ANOVA) was
performed to evaluate these differences in germination
more critically (Table 1). This analysis found that all three
factors of the model (genotype, seed position and daf)
explain significant proportions of the variation in percent
germination. A least significant difference (LSD) post hoc
test found that the atem6-1 genotype displayed significantly
higher germination (t > 0.315, P < 0.001) than either wildtype or ATEM6-C seeds. However, a significant genotype by
daf interaction (F4,162 = 13.702, P < 0.001) indicates that this
difference was only observed at 10 daf (LSD post hoc test
for 10 daf: t > 0.514, P < 0.001). Interestingly, seeds from the
proximal end of siliques were found to have higher germination rates than distal seeds (F1,162 = 12.737, P < 0.001), with
the greatest differences at 10 and 16 daf.
Developmental progression
The results presented above suggested that atem6-1 seeds
acquire essentially full desiccation tolerance even in the
absence of detectable ATEM6 protein and at a time earlier
than the wild-type. If this is the case, it should be possible
to demonstrate that other gene products associated with
desiccation tolerance also accumulate precociously. While
the details of this process are still poorly understood, one
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
247
A. J. Manfre et al.
Table 1 ANOVA, effect of genotype, seed position and daf on germination in the ADT assay
Source
DFa
Sum of squares
Mean square
F-ratio
P-value
Genotype
1
3.751
3.751
37.317
0
Position
1
0.665
0.665
6.611
0.011
Day
2
7.914
3.957
39.366
0
Genotype×position
1
0.323
0.323
3.218
0.076
Genotype×day
2
5.22
2.61
25.965
0
Position×day
2
0.333
0.166
1.656
0.196
Genotype×position×day
2
0.099
0.049
0.492
0.613
Error
162
10.856
0.101
a
DF, degrees of freedom.
gene product that could be immediately assessed was the
other group 1 LEA protein, ATEM1. The polyclonal antiwheat Em antibody already in hand (Manfre et al. 2006) recognizes both ATEM6 and ATEM1, and allowed examination
of ATEM1 expression in developing wild-type and atem6-1
seeds.
In order to compare the temporal expression pattern of
ATEM1 in wild-type and atem6-1 mutant seeds, the four
plant-proximal and plant-distal seeds from each side of the
central replum were harvested from siliques at some of the
same time points used for the ADT assay (10, 13 and 16
daf). Approximately 40 seeds were used for extraction and
35 µg of total soluble protein was electrophoresed on SDS–
polyacrylamide gels with mature wild-type seed protein
extracts (both proximal and distal) as controls. Proteins
were transferred to a polyvinylidene fluoride (PVDF) membrane for detection of ATEM proteins using the polyclonal
anti-wheat Em antibody.
In wild-type seeds, ATEM1 and ATEM6 proteins both
became detectable in plant-proximal and plant-distal seeds
at approximately 16 daf and remained detectable through
the mature seed stage (Fig. 4A). In atem6-1 seeds, however, ATEM1 protein became detectable at 13 daf (Fig. 4B),
consistent with a general acceleration of the maturation
process. Detection of ATEM1 protein in both plant-proximal
and plant-distal seeds is also consistent with the observance
of accelerated desiccation tolerance in seeds at both positions.
While these immunoblots do not represent a true quantitative
assessment of group 1 LEA expression, the level of ATEM1
protein in atem6-1 seeds appears to be significantly higher than
in wild-type seeds.
To assess further the expression of ATEM1 in the atem6-1
background, wild-type and atem6-1 mutant plants were
transformed with an ATEM1–green fluorescent protein (GFP)
fusion construct. This construct contains the entire ATEM1
gene including the promoter (with the 5′-untranslated
region), coding region (with intron) and 3′ control sequences,
into which a GFP coding region (Chiu et al. 1996) has been
248
A
kDa
17 -
1
2
3
4
5
6
7
8
9
ATEM1
11 -
B
kDa
17 11 -
ATEM6
1
2
3
4
5
6
7
8
9
ATEM1
ATEM6
Fig. 4 Immunoblot analysis of wild-type and atem6-1 seed proteins
during seed development. ATEM proteins were detected with a polyclonal
anti-wheat Em protein antibody that detects both ATEM1 and ATEM6.
(A) Wild-type; (B) atem6-1. Lanes 1, 10 daf proximal; 2, 10 daf distal; 3, 13
daf proximal; 4, 13 daf distal; 5, 16 daf proximal; 6, 16 daf distal; 7, mature
seed proximal; 8, mature seed distal; 9, wild-type mature seed control.
inserted 5′ of the ATEM1 stop codon. Therefore, it should be
possible to obtain information about both the temporal and
spatial pattern of ATEM1 expression. Developing seeds were
dissected from siliques at 12, 13, 14 and 15 daf, as described
for the ADT assay, and seed coats were removed to facilitate
GFP imaging.
GFP fluorescence is not detectable in either wild-type or
mutant embryos until 13 daf. However, while spatial expression at this stage does not appear to be significantly altered,
fluorescence intensity is noticeably higher in atem6-1 mutant
embryos than in the wild type (compare Fig. 5B and F).
The difference in expression levels between the wild type
and atem6-1 is even more dramatic in embryos 14 daf and
there appears to be noticeable expansion of the expression
domain in atem6-1 embryos. This observation is consistent
with elevated levels of ATEM1 protein seen in atem6-1 seeds
in Fig. 4. Interestingly, there is also a consistent difference in
expression level between distal and proximal embryos within
the two genotypes in which distal embryos have greater
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
Acquisition of desiccation tolerance
12 daf
13 daf
14 daf
15 daf
Wild-type
atem6-1
Fig. 5 Expression of the ATEM1-GFP fusion protein during seed development. Embryos were dissected from wild-type and atem6-1 seeds at 12,
13, 14 and 15 daf. In each panel, the lower and upper groups of embryos are from the proximal and distal ends, respectively, of a single silique.
Images were captured using a Nikon SMZ1500 dissecting scope with an X-CiteTM light source and Q-Capture software (version 2.68.6, Q-Imaging,
Burnaby BC, Canada).
fluorescence than proximal embryos. By 15 daf, distal atem6-1
embryos have lost most if not all of their chlorophyll (as evidenced by a reduction in autofluorescence) and begun to
dehydrate, whereas wild-type embryos still display some
chlorophyll autofluorescence and are fully hydrated.
Discussion
Acquisition of desiccation tolerance is a tightly regulated
process that occurs during the maturation phase of seed
development. Seeds activate a variety of genes and metabolic
pathways during this time, and the accumulation of sugars
as well as specific groups of proteins has been correlated
with maximal desiccation tolerance (Bartels et al. 1988).
We recently identified a mutant (atem6-1) that no longer
expresses ATEM6, one of the two group 1 LEA proteins in
Arabidopsis. The phenotype of this mutant, that includes
premature dehydration of plant-distal seeds, led us to suggest
that ATEM6 protein may be involved in regulating the rate
at which water is lost from the maturing seed (Manfre et al.
2006).
To assess the potential role of ATEM6 protein in seed
water loss in a more direct way, we measured the rate and
extent of weight loss for re-hydrated mature seeds after
removal of excess water. The results of this assay provided a
biphasic curve in which drying during the initial 20 min was
very rapid, and both the rate and amount of water lost from
wild-type and atem6-1 seeds was indistinguishable. During
development of the Arabidopsis seed, a hydrophilic polysaccharide (mucilage) is deposited in the outer cell layer of
the seed coat (Windsor et al. 2000). This epidermal layer of
mature seeds hydrates very rapidly on imbibition and forms
a hydrogel around the entire seed that, in isolation (i.e. a
non-soil environment), might be predicted to lose water
readily to the atmosphere. We attribute this rapid early
phase of drying to the loss of water that was absorbed by
the mucilage.
Weight loss after 20 min is much slower and we attribute
this phase of drying to water that is more tightly bound by
the mature seeds, potentially representing water absorbed
by the embryo. Notably, the amount of water lost by
atem6-1 seeds is greater than that by the wild type beginning at approximately 25 min into the assay. This difference
is statistically significant after 35 min of drying and remains
so for the duration of the assay. Seeds for both genotypes
lost <10 ng of weight (<0.20% of the total weight loss)
during the final 10 min, suggesting that water remaining
after 90 min is very tightly bound. An intriguing observation
of this experiment was that atem6-1 seeds appear to take
up significantly more water during imbibition than wild-type
seeds (Fig. 1B, Hydrated, P = 0.04). The additional water present in hydrated atem6-1 seeds only partially accounts for the
difference in water lost between wild-type and mutant seeds
in the assay, and atem6-1 seeds reach a statistically significantly lower water content than wild-type seeds (P < 0.001).
Certainly the mechanism(s) by which seeds lose water during
normal maturation on the plant will be somewhat different
from this assay. However, this result is consistent with the
phenotype of atem6-1 mutant plants and is also consistent
with a role for ATEM6 protein in embryonic water binding.
Greater water uptake by atem6-1 seeds also suggests a
role for ATEM6 during imbibition in which it may function
to limit and/or buffer water uptake. It is relatively well established that intracellular glasses form during late maturation
of angiosperm seeds and that these glasses are probably
composed of sugar–protein mixtures (Wolkers and Hoekstra
1997, Buitink and Leprince 2004). In fact, the glasses may
have a significant protein component (Wolkers et al. 1998,
Buitink et al. 2000). Due to their abundance in the cytoplasm
of embryonic cells during this time of seed development,
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
249
A. J. Manfre et al.
it is also likely that group 1 LEA proteins contribute directly
to glass stability. In the absence of ATEM6 protein, the stability of the glass present in the embryo may be compromised
such that more water can be absorbed. Much like it has been
proposed to prevent a rapid efflux of water during maturation, ATEM6 may help prevent a rapid influx of water during
imbibition that could be deleterious, particularly to membranous structures. Additional biophysical characterization
of atem6-1 mutant seeds may shed light on this possibility.
Considering that group 1 LEA protein structure and
expression pattern are so highly conserved across essentially
all angiosperms, monocot and dicot alike, and the Arabidopsis genome only encodes two group 1 LEAs, complete
absence of one might be anticipated to be associated with
a variety of effects. Despite this mutation, however, mature
dry seeds from atem6-1 plants appear fully viable and germinate readily under laboratory conditions. This result was
somewhat unexpected. Further, even when testing germination under suboptimal conditions, there was no observable
difference between wild-type and atem6-1 mutant seeds.
Clearly, ATEM6 protein is not essential for seed viability or germination proficiency, at least under the conditions tested.
While the complexity of desiccation tolerance in seeds
and the involvement of a number of different molecules may
make it difficult to assess the contribution of any individual
molecular species, the absence of ATEM6 protein in atem6-1
seeds clearly has an effect on seed maturation (Manfre et al.
2006). If ATEM6 makes an important and direct contribution to desiccation tolerance, then one could hypothesize
that the acquisition of tolerance in the mutant might be
delayed until later in seed maturation when other factors
have accumulated that can compensate for the absence of
ATEM6. In fact, our results demonstrate that desiccation
tolerance in atem6-1 mutant seeds is acquired earlier than
in the wild type, rather than later. Considering the potential
role of ATEM6 protein in water binding/loss during maturation, absence of this protein could lead to lower seed water
content at a time in development earlier than normal. This
could be perceived as a stress by the plant or, alternatively,
embryo water potential may serve as an endogenous cue
during maturation. Either of these possibilities could then
lead to a general acceleration of the normal maturation
program to help ensure seed survival. Interestingly, there
is also a difference in the acquisition of desiccation tolerance between plant-proximal and plant-distal seeds within
a silique wherein plant-proximal seeds develop tolerance
slightly earlier. However, this appears to be a general observation for both wild-type and atem6-1 mutant seeds and is,
therefore, not related to the presence or absence of ATEM6
protein.
If there is a general acceleration of the maturation
program, whether as the result of a perceived stress or of
an endogenous cue, it should be possible to demonstrate
250
premature accumulation of other molecular species involved
in desiccation tolerance. Unfortunately, our understanding
of the molecular mechanisms involved in this phenomenon
is still limited. As a first approximation, we have examined
expression of the other group 1 LEA protein, ATEM1, as
this molecule is equally likely to contribute to desiccation
tolerance and displays a similar temporal expression pattern
in the embryo. Both immunoblot and GFP fusion analyses
show that expression of ATEM1 is indeed detectable earlier in atem6-1 seeds than in wild-type seeds. In addition,
accelerated expression is observed in both plant-proximal
and plant-distal seeds, further supporting a mechanism
that affects maturation in all mutant seeds, not just those
in a plant-distal location. It is not yet known if expression
of ATEM1 in atem6-1 mutant seeds directly compensates
for the lack of ATEM6. Because the embryonic expression
patterns of these proteins only partially overlap (Bies et al.
1998), this might be unlikely in the absence of extensive
expansion of the ATEM1 expression domain. Our results
with the ATEM1-GFP protein fusion, however, may support some expansion of the ATEM1 expression domain, but
this possibility will need to be examined in a more critical
analysis.
This study provides evidence that the ATEM6 protein is
involved in a mechanism(s) associated with water retention/
loss during seed maturation. Whether ATEM6 is directly
involved in desiccation tolerance per se is not clear at this
time, but it is not required for viability or efficient germination under any conditions tested here. Absence of the
protein leads to premature dehydration of seeds, particularly those at a plant-distal location within the silique, and is
correlated with accelerated acquisition of desiccation tolerance in mutant seeds. In addition, expression of the other
A. thaliana group 1 LEA protein that may also be involved in
seed water relations, ATEM1, is similarly accelerated. While
not conclusive proof, these observations are consistent with
general acceleration of an otherwise normal maturation
process in atem6-1 mutant seeds. It will be of great interest
to examine the expression patterns of other gene products
that may also be involved in the acquisition of full desiccation
tolerance.
Materials and Methods
Plant materials
Wild-type A. thaliana plants were of the Columbia-0 ecotype. The atem6-1 T-DNA insertion line has been previously
described (Manfre et al. 2006).
Seed water loss and water content assay
Mature seeds from each genotype (100 seeds per replicate) were initially dried over a saturated LiCl solution
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
Acquisition of desiccation tolerance
(relative humidity of approximately 13%) in a hygrostat
for 2 d to ensure equal dehydration of all samples. Dried seeds
were re-hydrated overnight at 4°C in sterile water. The following day, seeds were transferred to aluminum pans (Part
No. 900786.901, TA Instruments, New Castle, DE, USA) and
several holes were punched in the bottom of the pan with a
26 gauge needle. The entire pan (seeds and residual water)
was transferred to a 2 ml conical bottom microcentrifuge
tube and spun at 470×g for 5 min to remove the water from
the pan. The pan with seeds was then placed in the weighing
chamber of a Mettler-Toledo UMT-2 balance with a Miracloth bag containing 4.0 g of powdered, freshly dried Drierite
(105°C overnight) and weight loss measured for 90 min. Data
were acquired using WinWedge® 32 Standard v. 3 software
(TAL Technologies, Philadelphia, PA, USA). After slow drying
for 90 min, the pans with seeds were placed in an oven at
105°C for 24 h and then reweighed. The weight of each pan
was determined after removing the dry seeds and subtracted
from all weights for that sample.
PEG germination assay
PEG 8000 at a concentration of 50% (w/w) was sterilized by
filtration through a 0.22 µm filter. MS medium [one packet
of MS minimal salts (#M6899, Sigma, St Louis, MO, USA)
supplemented with 1% sucrose, 0.8 mg l–1 thiamine, 0.5 mg
ml–1 pyridoxine, 0.5 mg ml–1 nicotinic acid and 50 mg l–1
myo-inositol] was mixed with 50% PEG (w/w) and water
to a final concentration of 0, 5, 10, 15 and 20% PEG and
0.25× MS. Mature seeds were uniformly dried by treatment
in a hygrostat containing a saturated LiCl solution (relative humidity approximately 13%) for 2 d and subsequently
surface sterilized using a vapor method (http://plantpath.
wisc.edu/∼afb/vapster.html). Five replicate plates with 100
seeds each on Whatman-1 filter paper saturated with 0.25×
MS medium ± PEG in plastic Petri plates (7 cm diameter)
were prepared. To minimize evaporation, each plate was
wrapped with Parafilm M. The seeds were stratified at 4°C
for 3 d and then grown at 24 ± 2°C under white fluorescent
light at 100 µmol m2 s–1 with a 16 h photoperiod. Seed germination was scored by radicle emergence after 5 d.
ADT germination assay
Developing seeds were dissected from siliques at four time
points during development as determined by tagging individual flowers on the day of anthesis. To collect immature
seeds, the siliques were opened from the base to the tip
with Dumont forceps (#4 and #5). All dissections were done
with sterile instruments in a laminar flow hood. Ten replicate plates for each data point were sown with 16 seeds
onto water-saturated 2.0 cm diameter Whatman No. 1 filter
paper in 2.5 cm diameter plastic Petri plates. The seeds were
dried over saturated LiCl in a hygrostat for 2 d to ensure equal
dehydration of all samples. Seeds were rehydrated by saturating the filter papers with 0.5× MS medium. To minimize
evaporation, each plate was wrapped with Parafilm M, then
placed in a pan containing water-saturated 3MM paper and
covered with a plastic dome lid to maintain humidity. Plates
were incubated under a 16 h photoperiod at 24 ± 2°C of
white fluorescent light at 250 µmol m2 s–1. Germination was
scored by radicle emergence once a day for 10 d.
Statistical analyses
Data from the water loss assay and PEG germination assay
were individually subjected to unpaired t-test analyses. Data
from the ADT germination assay were analyzed using a fully
fit three-way ANOVA and an LSD post hoc test using Systat
software (Version 10, Systat, Point Richmond, CA, USA). The
dependent variable for the three-way ANOVA, percentage
germination, was subjected to arc-sine square root transformation to meet the assumption of normality of residuals.
Immunoblot analysis
The four plant-proximal and plant-distal seeds from each
side of the central replum were harvested from wild-type
and atem6-1 siliques at 10, 13 and 16 daf. Total soluble
protein was isolated from approximately 40 seeds for each
sample by grinding in a solution of 10 mM Tris–HCl (pH
8.3), 10 mM NaCl in a microcentrifuge tube. The tubes were
vortexed well and centrifuged repeatedly for 2 min at maximum speed at 4°C until no solid material remained at the
bottom of the tube. Extracts were then frozen at −20°C and
put directly into the centrifuge to spin at maximum speed
at 4°C for 30 min. The supernatant was transferred to a new
tube and, heated to 80°C for 20 min followed by centrifugation at maximum speed at 4°C for 30 min. This supernatant
was transferred to a fresh tube and protein concentration
determined by bicinchoninic acid assay (Pierce, Rockford,
IL, USA). Extract volumes containing 35 µg of total soluble protein were brought to a final concentration of 25%
trichloroacetic acid, incubated on ice for 5 min and pelleted
at 4°C in a microcentrifuge at maximum speed for 5 min.
The supernatant was removed and the protein pellets were
washed with 100% acetone and spun for another 5 min at
4°C. The acetone was removed, and the pellets dried slightly
and resuspended in gel loading buffer. After resolution on
denaturing 15% Tris-glycine gels, proteins were electrotransferred to an Immobilon-P PVDF membrane (Millipore,
Billerica, MA, USA) at 100 V and 4°C for 1 h. Blots were processed using a primary polyclonal rabbit anti-wheat Em
antibody and a secondary alkaline phosphatase-conjugated
goat anti-rabbit antibody. Chemiluminescent detection was
performed using CSPD-ready-to-use (Roche Diagnostics,
Indianapolis, IN, USA) and images captured on a Fuji LAS1000plus (Fujifilm Life Science USA, Stamford, CT, USA).
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
251
A. J. Manfre et al.
Table 2 PCR primers
Primer name
Primer sequence (5′→3′)
ATEM1-L
CTACACCTTCAAAACCCTAAATCCAAATC
ATEM1-R
TGAGAGGTCTCTTAATCAGAAGTTTCTTG
ATEM1-StuI-L-C
GAAGCTCAAGAGCATCTTGCTGAAGGTATT
ATEM1-StuI-L-NC
CTAAGGCCTGTTGGTGAACTTTGACTCATC
ATEM1-StuI-R-C
AACAGGCCTTAGTAGTTACGAGCTACTAGTG
ATEM1-StuI-R-NC
CAAAAATGATACTTCCTCAAACTTTTTATAATC
GFP analyses
A 3,007 bp fragment of genomic DNA containing the
entire ATEM1 gene was PCR amplified using ATEM1-L and
ATEM1-R primers (Table 2) and blunt-end cloned into
pBluescript II KS+ at the SmaI site to create pBM487. A
unique StuI site was introduced at the ATEM1 stop codon
for insertion of a GFP-coding region. Two PCR products
were generated using the ATEM1-StuI-L-C/ATEM1-StuI-L-NC
(548 bp) and ATEM1-StuI-R-C/ATEM1-StuI-R-NC (498 bp)
primer pairs, digested with StuI and ligated together. The
product was then digested with BglII–NdeI and the 818 bp
fragment used to replace the corresponding fragment in
pBM487 to create plasmid pBM498, and the StuI site verified by sequence analysis. A jellyfish GFP-coding region was
isolated from the GFP(S65T) plasmid (Chiu et al. 1996) by
digestion with StuI–PstI and the PstI site rendered blunt by
treatment with the large Klenow fragment of Escherichia coli
DNA polymerase I. This fragment was ligated into the StuI
site of pBM498 to create pBM499 and verified by sequence
analysis. The entire ATEM1-GFP fragment was transferred to
pBI101 as an EcoRI–HindIII fragment to create pBM507, and
verified by restriction enzyme digestion.
Plasmid pBM507 was introduced into Agrobacterium
tumefaciens strain LBA4404 by heat shock (Chen et al. 1994)
and the resulting strain used to transform both wild-type
and atem6-1 A. thaliana plants using a floral dip method
(Clough and Bent 1998). Transformed seeds were selected
on kanamycin (50 µg ml–1) and verified by PCR. Developing
seeds were dissected from siliques at 12, 13, 14 and 15 d after
anthesis as described for the ADT assay. Seed coats were
removed and images captured using a Nikon SMZ1500 dissecting scope with an X-Cite™ light source and Q-Capture
software (version 2.68.6, Q-Imaging, Burnaby BC, Canada).
Magnification and exposure settings were held constant for
all images.
Funding
The National Research Initiative of the US Department
of Agriculture (USDA) Cooperative State Research, Education and Extension Service (CSREES; 2002-35318-12627
to W.R.M.); the CSREES/USDA (SC-1700200 to W.R.M.); the
252
Howard Hughes Medical Institute SC-LIFE Undergraduate
Research Program; the Clemson University Calhoun Honors
College.
Acknowledgments
We would like to thank Julie Lortz for excellent technical
assistance, and Michael Childress for invaluable assistance
with the statistical analyses. Plasmid GFP(S65T) was the kind
gift of Jen Sheen (through Shu-Hua Cheng). Technical Contribution Number 5298 of the Clemson University Experiment Station.
References
Baker, J., Steele, C. and Dure, L., III (1988) Sequence and characterization
of 6 Lea proteins and their genes from cotton. Plant Mol. Biol. 11:
277–291.
Bartels, D., Singh, M. and Salamini, F. (1988) Onset of desiccation
tolerance during development of the barley embryo. Planta 175:
485–492.
Berge, S.K., Bartholomew, D.M. and Quatrano, R.S. (1989) Control of
the expression of wheat embryo genes by abscisic acid. In Molecular
Basis of Plant Development. Edited by Goldberg, R.L. pp.193–201.
Alan R. Liss, New York.
Bies, N., Aspart, L., Carles, C., Gallois, P. and Delseny, M. (1998)
Accumulation and degradation of Em proteins in Arabidopsis
thaliana: evidence for post-transcriptional controls. J. Exp. Bot. 49:
1925–1933.
Black, M., Corbineau, F., Gee, H. and Come, D. (1999) Water content,
raffinose, and dehydrins in the induction of desiccation tolerance in
immature wheat embryos. Plant Physiol. 120: 463–471.
Bostock, R.M. and Quatrano, R.S. (1992) Regulation of Em gene
expression in rice. Interaction between osmotic stress and abscisic
acid. Plant Physiol. 98: 1356–1363.
Buitink, J., Hemminga, M.A. and Hoekstra, F.A. (2000) Is there a role for
oligosaccharides in seed longevity? An assessment of intracellular
glass stability. Plant Physiol. 122: 1217–1224.
Buitink, J. and Leprince, O. (2004) Glass formation in plant
anhydrobiotes: survival in the dry state. Cryobiology 48: 215–228.
Burke, M.J. (1986) The glassy state and survival of anhydrous biological
systems. In Membranes, Metabolism and Dry Organisms. Edited by
Leopold, A.C. pp. 358–363. Cornell University Press, Ithaca, NY.
Chen, H., Nelson, R.S. and Sherwood, J.L. (1994) Enhanced recovery of
transformants of Agrobacterium tumefaciens after freeze–thaw
transformation and drug selection. Biotechniques 16: 664–670.
Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H. and Sheen, J.
(1996) Engineered GFP as a vital reporter in plants. Curr. Biol. 6:
325–330.
Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplified method for
Agrobacterium-mediated transformation of Arabidopsis thaliana.
Plant J. 16: 735–743.
Cuming, A. (1999) LEA proteins. In Seed Proteins. Edited by Shewry, P.
and Casey, R. pp. 753–780. Kluwer Publishers, Dordrecht, The
Netherlands.
Cuming, A.C. and Lane, B.G. (1979) Protein synthesis in imbibing wheat
embryos. Eur. J. Biochem. 99: 217–224.
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
Acquisition of desiccation tolerance
Espelund, M., Saebøe-Larssen, S., Hughes, D.W., Galau, G.A., Larsen, F.
and Jakobsen, K.S. (1992) Late embryogenesis-abundant genes
encoding proteins with different numbers of hydrophilic repeats are
regulated differentially by abscisic acid and osmotic stress. Plant J. 2:
241–252.
Feder, M.E. and Hofmann, G.E. (1999) Heat-shock proteins, molecular
chaperones, and the stress response: evolutionary and ecological
physiology. Annu. Rev. Physiol. 61: 243–282.
Gaubier, P., Raynal, M., Hull, G., Huestis, G.M., Grellet, F., Arenas, C.,
et al. (1993) Two different Em-like genes are expressed in Arabidopsis
thaliana seeds during maturation. Mol. Gen. Genet. 238: 409–418.
Goyal, K., Walton, L.J. and Tunnacliffe, A. (2005) LEA proteins prevent
protein aggregation due to water stress. Biochem. J. 388: 151–157.
Hoekstra, F.A., Golovina, E.A. and Buitink, J. (2001) Mechanisms of
plant desiccation tolerance. Trends Plant Sci. 6: 431–438.
Koster, K.L. (1991) Glass formation and desiccation tolerance in seeds.
Plant Physiol. 96: 302–304.
Leopold, A.C., Sun, W.Q. and Bernal-Lugo, I. (1994) The glassy state in
seeds: analysis and function. Seed Sci. Res. 4: 267–274.
Leprince, O., Vertucci, C.W., Hendry, G.A.F. and Atherton, N.M. (1995)
The expression of desiccation-induced damage in orthodox seeds is
a function of oxygen and temperature. Physiol. Plant. 94: 233–240.
Manfre, A.J., Lanni, L.M. and Marcotte, W.R., Jr. (2006) The Arabidopsis
group 1 LATE EMBRYOGENESIS ABUNDANT protein ATEM6 is
required for normal seed development. Plant Physiol. 140: 140–149.
Morris, P.C., Kumar, A., Bowles, D.J. and Cuming, A.C. (1990) Osmotic
stress and abscisic acid induce expression of the wheat Em genes.
Eur. J. Biochem. 190: 625–630.
Ooms, J.J.J., deBruyn, S.M. and Karssen, C.M. (1992) Acquisition
of desiccation tolerance in seeds of Arabidopsis thaliana. In
Fourth International Workshop on Seeds. Edited by Come, D. and
Corbineau, F., Angers, France.
Penfield, S., Meissner, R.C., Shoue, D.A., Carpita, N.C. and Bevan, M.W.
(2001) MYB61 is required for mucilage deposition and extrusion in
the Arabidopsis seed coat. Plant Cell 13: 2777–2791.
Roberts, J.K., DeSimone, N.A., Lngle, W.L. and Dure, L. (1993) Cellular
concentrations and uniformity of cell-type accumulation of two lea
proteins in cotton embryos. Plant Cell 5: 769–780.
Stacy, R.A. and Aalen, R.B. (1998) Identification of sequence homology
between the internal hydrophilic repeated motifs of group 1 lateembryogenesis-abundant proteins in plants and hydrophilic repeats
of the general stress protein GsiB of Bacillus subtilis. Planta 206:
476–478.
Sun, W.Q., Irving, T.C. and Leopold, A.C. (1994) The role of sugar,
vitrification and membrane phase-transition in seed desiccation
tolerance. Physiol. Plant. 90: 621–628.
van der Weele, C.M., Spollen, W.G., Sharp, R.E. and Baskin, T.I. (2000)
Growth of Arabidopsis thaliana seedlings under water deficit studied
by control of water potential in nutrient-agar media. J. Exp. Bot. 51:
1555–1562.
Vertucci, C.W. and Farrant, J.M. (1995) Acquisition and loss of
desiccation tolerance. In Seed Development and Germination.
Edited by Kigel, J. and Galili, G. pp. 237–271. Marcel Dekker, Inc.,
New York.
Vicient, C., Hull, G., Guilleminot, J., Devic, M. and Delseny, M. (2000)
Differential expression of the Arabidopsis genes coding for Em-like
proteins. J. Exp. Bot. 51: 1211–1220.
Wehmeyer, N., Hernandez, L.D., Finkelstein, R.R. and Vierling, E. (1996)
Synthesis of small heat-shock proteins is part of the developmental
program of late seed maturation. Plant Physiol 112: 747–757.
Wehmeyer, N. and Vierling, E. (2000) The expression of small heat
shock proteins in seeds responds to discrete developmental signals
and suggests a general protective role in desiccation tolerance. Plant
Physiol. 122: 1099–1108.
Wilhelm, K.S. and Thomashow, M.F. (1993) Arabidopsis thaliana
Cor15b, an apparent homolog of Cor15a, is strongly responsive to
cold and ABA, but not drought. Plant Mol. Biol. 23: 1073–1077.
Williams, R.J. and Leopold, A.C. (1989) The glassy state in corn embryos.
Plant Physiol. 89: 977–981.
Windsor, J.B., Symonds, V.V., Mendenhall, J. and Lloyd, A.M. (2000)
Arabidopsis seed coat development: morphological differentiation
of the outer integument. Plant J. 22: 483–493.
Wolkers, W.F. and Hoekstra, F.A. (1997) Heat stability of proteins in
desiccation-tolerant cattail (Typha latifolia L) pollen—a Fourier
transform infrared spectroscopic study. Comp. Biochem. Phys. A 117:
349–355.
Wolkers, W.F., Oldenhof, H., Alberda, M. and Hoekstra, F.A. (1998) A
Fourier transform infrared microspectroscopy study of sugar glasses:
application to anhydrobiotic higher plant cells. Biochim. Biophys.
Acta 1379: 83–96.
(Received November 17, 2008; Accepted November 27, 2008)
Plant Cell Physiol. 50(2): 243–253 (2009) doi:10.1093/pcp/pcn185 © The Author 2008.
253