This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 140/4/1437    most recent pp.105.070508v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Wong, C. E. Articles by Moffatt, B. A. Search for Related Content PubMed PubMed Citation Articles by Wong, C. E. Articles by Moffatt, B. A. Agricola Articles by Wong, C. E. Articles by Moffatt, B. A.

ENVIRONMENTAL STRESS AND ADAPTATION TO STRESS

Transcriptional Profiling Implicates Novel Interactions between Abiotic Stress and Hormonal Responses in Thellungiella, a Close Relative of Arabidopsis^1,[W]

Department of Biology, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 (C.E.W., Y.L., C.D., M.G., B.A.M.); Département de mathématiques et de statistique, Pavillon Alexandre-Vachon, Université Laval, Quebec, Canada G1K 7P4 (A.L.); Department of Biology, McMaster University, Hamilton, Ontario, Canada L8S 4K1 (D.G., P.N., B.W., G.B.G., E.A.W.); and Department of Plant Sciences, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5A8 (G.R.G.)

	ABSTRACT

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Thellungiella, an Arabidopsis (Arabidopsis thaliana)-related halophyte, is an emerging model species for studies designed to elucidate molecular mechanisms of abiotic stress tolerance. Using a cDNA microarray containing 3,628 unique sequences derived from previously described libraries of stress-induced cDNAs of the Yukon ecotype of Thellungiella salsuginea, we obtained transcript profiles of its response to cold, salinity, simulated drought, and rewatering after simulated drought. A total of 154 transcripts were differentially regulated under the conditions studied. Only six of these genes responded to all three stresses of drought, cold, and salinity, indicating a divergence among the end responses triggered by each of these stresses. Unlike in Arabidopsis, there were relatively few transcript changes in response to high salinity in this halophyte. Furthermore, the gene products represented among drought-responsive transcripts in Thellungiella associate a down-regulation of defense-related transcripts with exposure to water deficits. This antagonistic interaction between drought and biotic stress response may demonstrate Thellungiella's ability to respond precisely to environmental stresses, thereby conserving energy and resources and maximizing its survival potential. Intriguingly, changes of transcript abundance in response to cold implicate the involvement of jasmonic acid. While transcripts associated with photosynthetic processes were repressed by cold, physiological responses in plants developed at low temperature suggest a novel mechanism for photosynthetic acclimation. Taken together, our results provide useful starting points for more in-depth analyses of Thellungiella's extreme stress tolerance.

The Shandong ecotype of Thellungiella grows in the high-salinity coastal areas in eastern China and has been proposed to be an appropriate relative of Arabidopsis (Arabidopsis thaliana) for studies of salinity tolerance mechanisms (Bressan et al., 2001; Inan et al., 2004). On the other hand, the Yukon ecotype was isolated in the Takhini Salt Flats near Whitehorse in the Yukon Territories, Canada, a subarctic and semiarid region (Warwick et al., 2004) characterized by multiple simultaneous abiotic stresses, including cold, drought, and high salinity. Under controlled conditions, both ecotypes can tolerate salinity as high as 500 mM NaCl (Inan et al., 2004; E.A. Weretinylk, personal communication). In addition, the Yukon ecotype of Thellungiella survives freezing to –18°C (M. Griffith, M. Timonin, and M. Saldanha, unpublished data) and can withstand water losses in excess of 40% of its fresh weight (E.A. Weretilnyk, personal communication), which are conditions far more extreme than those tolerated by Arabidopsis. Moreover, like Arabidopsis, Thellungiella displays many features characteristic of a genetic model system, including a short life cycle, self-fertility, transformability, and small genome size. Particularly important for this research is its high sequence similarity to Arabidopsis (Wang et al., 2004; Wong et al., 2005).

Transcript-profiling experiments using the Arabidopsis GeneChip array or full-length cDNA microarrays have shown that extensive changes occur in the transcriptome of Arabidopsis in response to drought, cold, or salinity stresses (Fowler and Thomashow, 2002; Kreps et al., 2002; Seki et al., 2002b). In fact, Kreps and coworkers reported that approximately 30% of the transcriptome on the Arabidopsis GeneChip 8 K oligoarray changed in association with stress treatments (Kreps et al., 2002).

Since expressed sequence tag analyses of Thellungiella clones revealed 90% to 95% identities between Thellungiella and Arabidopsis cDNA sequences (Wang et al., 2004; Wong et al., 2005), Arabidopsis microarrays have recently been used to perform systematic analyses of the gene expression profile in Thellungiella undergoing abiotic stress treatments (Taji et al., 2004; Volkov et al., 2004). Due to the specificity of the Arabidopsis GeneChip 22 K oligoarray, only one-fifth of the probes showed a significant signal after hybridization with Shandong ecotype mRNA. Of these, only 19 genes (0.5% of probe sets showing significant signals) were significantly up- or down-regulated after 5 d of 100 mM salt treatment. When an Arabidopsis full-length 8 K cDNA microarray was used in a comparative genomics study in salt tolerance between Arabidopsis and the Shandong ecotype of Thellungiella, only six genes were found to be differentially expressed by salt treatments in Thellungiella (Taji et al., 2004). This is attributed to a constitutive high level of expression of stress-responsive genes in Shandong ecotype in comparison to Arabidopsis and was proposed to be contributing to the extreme salt tolerance of Thellungiella (Taji et al., 2004). The constitutive high level of expression of stress-responsive genes in Shandong ecotype was also reported by a more recent study using a 25 K Arabidopsis 70-mer oligoarray that compared the transcript profiles of Arabidopsis and Thellungiella subjected to NaCl treatment (Gong et al., 2005). Despite these findings, the transcriptional responses of Thellungiella, in particular, the Yukon ecotype, to drought, cold, and rewatering after drought have yet to be explored.

Previously, we have reported the generation of 3,628 nonredundant cDNAs derived from stress-induced libraries of the Yukon ecotype of Thellungiella (Wong et al., 2005). These sequences were used to construct a cDNA microarray that has representative coverage of Thellungiella genome and is enriched for genes whose expression is associated with stress. Here we have used this microarray to perform an initial analysis of the molecular response of the Yukon ecotype of Thellungiella subjected to cold, exposure to salt, simulated drought, or simulated drought treatment followed by rewatering. The resulting expression profiles, though incomplete as they are limited to the transcripts represented by the array, provide a framework of molecular processes underlying abiotic stress tolerance in Thellungiella, which can be complemented at the level of proteins and metabolites.

Our experimental design primarily focuses on a slow and prolonged stress treatment applied to plants under controlled conditions (see "Materials and Methods") so as to approximate the type of exposure to individual stresses encountered by Thellungiella in the field. This approach contrasts with previous microarray studies in that they have focused on rapid, short-term abiotic stress treatments to identify genes involved in signaling pathways and transcriptional regulation of abiotic stress-induced genes in plants (Fowler and Thomashow, 2002; Kreps et al., 2002; Seki et al., 2002b; Taji et al., 2004). A consideration in this approach is that a rapidly imposed stress can result in a greater number of injury-related responses (for review, see Bray, 2002) and may not reveal adjustments required for long-term survival under stressful environmental conditions.

This study compares three stresses (cold, low water availability, and saline conditions) as well as recovery from water deficits in Thellungiella, using an expression-profiling strategy. We identify some intriguing features associated with Thellungiella's stress responses. The lower than expected degree of overlap among genes responsive to drought, cold, or salinity suggested that there are relatively few common end responses triggered by these stresses. In addition to activating the expression of some well-known stress-responsive genes, Thellungiella was found to down-regulate a large number of biotic stress-related genes under drought and salinity treatments. This infers the employment of a precise defense strategy under conditions of osmotic stress. A large number of photosynthetic-related transcripts were repressed in responses to cold stress. However, plants developed at low temperature demonstrated a high tolerance to photoinhibition of photosynthesis, which is difficult to reconcile based on known photoprotective processes. This suggests an important and perhaps novel photosynthetic response of Thellungiella to low temperature.

	RESULTS

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 

Physiological Characterization of Simulated Drought and Salinity Treatments

Water and solute potential measurements were carried out for well-watered control plants and plants that were subjected to simulated drought or saline conditions (see "Materials and Methods"). We consistently found, under the experimental conditions used, that Thellungiella was visibly wilted after 3 d of withholding water, when the leaf relative water content is about 65%. As shown in Table I , the leaf solute potentials became more negative as the plants were salinized or subjected to simulated drought, an indication that the solutes have become more concentrated under these conditions. In contrast to the salinity treatment, solute accumulation for drought-treated plants is accompanied by a loss of turgor, but both turgor and solute concentrations return to values comparable to those for control plants upon rewatering. Thus, our osmotic stress treatment strategies elicit stress-related responses as monitored by changes in plant water relations under the controlled experimental conditions we utilized.

To identify stress-associated genes using the Thellungiella cDNA microarrays, we have employed analysis methods that avoid the determination of transcriptional changes using cutoffs based upon fold-changes. Instead, an empirical Bayesian strategy was used to assess the significance of each treatment for each gene as outlined in "Materials and Methods" (Labbe, 2005; Labbe and Thompson, 2005).

Using this statistical test, we found the abundance of 101, 76, 22, and 46 transcripts to be significantly changed under drought, cold, high-salinity, and rewatering conditions, respectively. These transcripts were assigned a Munich Information Center for Protein Sequences (MIPS) protein code based on the best BLAST match. They range in fold-change relative to the corresponding control from 0.09 to 31 (Table II ). The former corresponds to a gene encoding a member of the protease/lipid transfer protein (LTP) family (At2g10940), while the latter is a sequence annotated as rab18 (At5g66400). The number of transcripts regulated by salinity was less than one-fifth and one-third of that of drought- and cold-regulated genes, respectively. This low level of transcriptional change has also been reported in two previous microarray studies of salinity effects using the Shandong ecotype of Thellungiella (Taji et al., 2004; Volkov et al., 2004).

View larger version (48K): [in this window] [in a new window]   Figure 1. Verification of microarray analysis by RT-PCR. RT-PCR analysis was carried out under linear amplification conditions for six orthologs of Arabidopsis as shown. An ortholog of Arabidopsis UBQ10 (At4g05320) was used as an internal control. Con, Control; Cld, cold; Dr, drought; SA, salinity.

View larger version (31K): [in this window] [in a new window]   Figure 2. Changes in transcript abundance in Thellungiella subjected to abiotic stress. A and B, Venn diagrams showing the number of transcripts differentially regulated during drought, cold, or high-salinity treatment (A), or drought or rewatering after drought (B). The total number of transcripts for each treatment is indicated in parentheses.

Clusters 1 and 6 contain a total of 61 transcripts that among the three stresses only respond to drought treatment. They represent 59% of all the 101 drought-regulated mRNAs. The greatest fold of induction was detected for a transcript encoding a homolog to a putative Arabidopsis alkaline -galactosidase that contains the PFAM profile of raffinose synthase (At3g57520), an enzyme involved in the biosynthesis of the raffinose family of oligosaccharides. Various genes previously reported to be abscisic acid (ABA) induced (Seki et al., 2002a) were also found to be up-regulated by drought. These ABA-responsive genes encode proteins that include galactinol synthase (At1g56600), ERD14 (At1g76180), LEA protein (At1g52690), protein phosphatase 2c (At1g72770), RD22 (At5g25610), and nodulin MtN3 family (At5g13170; Seki et al., 2002a), implicating the role of ABA in regulating drought responses in Thellungiella. This is consistent with the well-established role of ABA in drought stress (for review, see Zhu, 2002). Increases in transcript abundance were also found for genes involved in reactive oxygen species scavenging. It is well documented that abiotic stresses lead to overproduction of reactive oxygen species that cause damage to other molecules and cell structures (for review, see Apel and Hirt, 2004). For example, increased transcripts of metallothionein (At3g09390), peroxidase 42 (At4g21960), and glutathione S-transferase (At2g02390) encode proteins that could potentially function as scavengers for reactive oxygen species that arise as a consequence of drought.

Cluster 6 contains transcripts that are repressed under drought and, interestingly, these include defense-related transcripts that encode -1,3-glucanase (At3g57240) and osmotin-like protein (At4g11650). Similar transcripts have been reported to be induced by abiotic stimuli in pepper (Capsicum annuum) or Arabidopsis (Hong and Hwang, 2002; supplemental data in Seki et al., 2002a; supplemental data in Seki et al., 2002b).

We found 25 transcripts to be specifically expressed following a 3-week exposure to cold treatment (Cluster 4; Table II). Among these transcripts are 17 cold-responsive transcripts that encode a putative lipoxygenase (LOX; At1g17420), two members of the plant defensin protein (PDF1.2a and PDF 2.2), and chalcone synthase (At5g44420; At2g02100; At5g13930; Cluster 4; Table I). According to AraCyc, the Arabidopsis Biochemical Pathway tool (http://www.arabidopsis.org/tools/aracyc/), this putative LOX is predicted to be involved in the biosynthesis of the phytohormone jasmonic acid (JA; Bell and Mullet, 1993). The up-regulation of a LOX with a concomitant increase of PDF1.2a, a known JA-mediated defense marker (Penninckx et al., 1998), strongly implicates an increase of JA levels in response to cold. In addition, expression of chalcone synthase, a key enzyme in phenylpropanoid biosynthesis that is known to be transcriptionally regulated by JA (Richard et al., 2000) was also found in the same cluster. To date, only one report describes the changes in the Arabidopsis transcriptome associated with short- as well as long-term cold stresses (Fowler and Thomashow, 2002). Intriguingly, neither the transcript for the JA biosynthetic enzyme nor PDF1.2 is reported to be up-regulated by long-term cold treatment in Arabidopsis (supplemental data in Fowler and Thomashow, 2002). However, it is important to note that the long-term cold stress in that study was only 7 d in duration (Fowler and Thomashow, 2002).

Our analysis identified 26 cold-repressed transcripts (Cluster 9; Table II); one-half encode products that have putative functions associated with photosynthesis. These include genes that encode chlorophyll a/b-binding proteins, plastocyanin, and PSI subunit proteins. Similar genes have been reported to be repressed in Arabidopsis "cold" microarray studies (Fowler and Thomashow, 2002; Kreps et al., 2002; Seki et al., 2002b). Based on these findings, we further examined plants that were allowed to develop at low temperature for their tolerance to photoinhibition (Fig. 3A ) and their capacity for nonphotochemical quenching (NPQ), a major photoprotective mechanism (Fig. 3B; Müller et al., 2001). It is well established that development at low temperature results in an increased tolerance to photoinhibition (Huner et al., 1993; Gray et al., 1996, 2003), in part due to a reprogramming of photosynthetic carbon metabolism (Stitt and Hurry, 2002; Gray and Heath, 2005). Plants of Thellungiella grown under standard conditions demonstrated a 54% reduction in the maximum quantum yield of PSII photochemistry as measured by the F_v/F_m ratio in response to the 4-h photoinhibitory treatment (Fig. 3A). In contrast, plants developed at low temperature exhibited only a 10% reduction in F_v/F_m in response to the same photoinhibitory treatment (Fig. 3A). The capacity for NPQ formation was observed to be slightly higher in the plants developed at low temperature, although the difference was minimal (Fig. 3B).

View larger version (9K): [in this window] [in a new window]   Figure 3. Photosynthetic responses of Thellungiella. Plants were grown under standard growth conditions (Non-acclimated; ) or developed at low temperature (Acclimated; ) as described in "Materials and Methods." A, Photoinhibition at low temperature. Plants were exposed to 1,000 µmol photons m–2 s–1 at 4°C for 4 h, and the Fv/Fm ratio was determined pretreatment (black bars) and posttreatment (white bars). B, Induction of NPQ. Induction occurred at room temperature with a photon flux of 1,400 µmol photons m–2 s–1. All values represent means ± SD, n = 2 to 3.

Shared Responses

Clusters 2, 3, 5, 7, 8, 10, 12, and 13 consist of transcripts that are differentially regulated by more than one stress treatment relative to the control; the corresponding transcripts for each class are listed in Table II. As shown in Figure 2A, there was greater overlap between the drought- and salinity-induced changes as compared to the changes caused by drought and cold stresses. Out of 101 and 22 transcripts differentially regulated by drought treatment and salinity, respectively, there was an overlap of 15 sequences (68% of salinity-regulated transcripts), and all were repressed by both treatments. Based on best BLAST match, the corresponding gene products can be categorized under three major functional groups: signaling, photosynthesis, and defense related. A member of the basic helix-loop-helix transcription factor family (At1g61660) and a calmodulin-6 (At5g21274) represent the signaling group, whereas a putative chitinase (At2g43590), a pathogenesis-related protein (At1g75040), and a disease resistance protein (At5g41550) constitute the defense-related group. Meanwhile, 41% of the cold-regulated transcripts were also responsive to drought treatment. Among them were well-known cold- and drought-induced genes, such as ELIP (early light inducible), cor15b (cold-regulated protein), cor47, erd10 (early responsive to dehydration), cor6.6, Adh (alcohol dehydrogenase), and P5CS A (delta 1-pyrroline-5-carboxylate synthetase A; At2g39800), a gene encoding a product associated with Pro biosynthesis (Kreps et al., 2002; Seki et al., 2002b).

There is a higher percentage (68%) of genes that are also drought regulated in the salinity dataset when compared to that of the cold dataset (41%). This extensive overlap between the drought- and salinity-induced changes hints of commonalities between stress responses elicited by the drought and salinity treatments we used, suggesting cross talk between responses triggered by these stresses. This is consistent with previous observations on the greater overlap of drought- and high salinity-responsive gene expression in Arabidopsis (67%) in comparison to that of drought and cold (41%; Seki et al., 2002b).

There were four transcripts that are cold and salinity regulated (Clusters 10 and 13; Table II). Genes encoding a putative -glucosidase (At1g52400) and a member of lipid-associated protein family (At2g22170) were up-regulated under salinity treatment but repressed by cold (Cluster 13; Table II). No known function or molecular basis of their antagonistic responsiveness has been established for either corresponding protein.

The expression of 149 transcripts changed in association with at least one stress treatment; only six of these responded to all stress treatments (Fig. 2A). This suggests that the stress-responsive mechanisms in Thellungiella are divergent for each of these stresses, as reflected by the lack of sequence overlap in our previously reported expressed sequence tag collections of stress-induced libraries (Wong et al., 2005). Transcripts elevated by all three stresses encode the low temperature-induced 65-kD protein (RD29B; At5g52300), a lipid transfer protein 4 (LTP4; At5g59310), and the dehydrin Rab18 (At5g66400). The increase of rd29b and rab18 transcripts by abiotic stress is similar to that of Arabidopsis. For example, Seki and coworkers reported a greater than 2-fold increase for rd29b under all three stress treatments relative to the controls (Seki et al., 2002b). Though an increase of rab18 transcripts was not found in the same study (Seki et al., 2002b), a modest increase in the protein level has been reported in Arabidopsis after 7 d of cold treatment (Mantyla et al., 1995). Meanwhile, LTP4 transcripts were highly abundant in a salt-induced library of the Shandong ecotype of Thellungiella (Wang et al., 2004), and our microarray analysis revealed that LTP4 mRNAs, in addition to being induced by salt, are also induced by cold or drought treatment.

Drought and Rewatering Dataset

There were a total of 106 transcripts identified to be drought and/or rewatering responsive (Fig. 2B). Among these, 45 transcripts were differentially regulated by rewatering treatment when compared to the drought-treated control. The remaining 61 transcripts that were up- or down-regulated by drought were unchanged in rewatered plants relative to the drought sample.

We identified 19 rewatering-repressed and drought-inducible transcripts, and 21 rewatering-inducible and drought-repressed transcripts (Supplemental Table II). The former includes transcripts with homology to that encoding Arabidopsis dehydration-induced protein ERD10 (At1g20450), COR47 (At1g20440), and a LEA protein (At1g52690), while the latter consists of genes that may be repressed by drought and released from the repression by rewatering. This includes photosynthesis-related proteins and is consistent with photosynthesis being important in the process of recovery following rewatering.

	DISCUSSION

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Our microarray study primarily focused on Thellungiella exposed to various stresses over the course of days or weeks to identify genes that may be mechanistically involved in acclimation or stress resistance in the steady state. Among the 149 transcripts that changed in response to drought, cold, or salinity, only six were regulated by all three stresses (Fig. 2A). This is in spite of the fact that osmotic stress and the associated oxidative stress appear to be common consequences of exposure to drought, salinity, and cold (Apel and Hirt, 2004). The fact that the time period used to develop suitable drought and cold or salinity samples is not equivalent may have contributed to this finding. Nevertheless, the low degree of overlap among the transcripts associated with each stress suggests that the end responses triggered by these stresses are rather specific and suited to the particular stress conditions encountered.

Salinity-Induced Transcript Changes in Halophyte

The salinity dataset contains the fewest transcripts that were stress regulated when compared to drought and cold datasets. This corresponds to less than 1% of the sequences represented by our array. A similar low degree of transcriptional change was revealed by other "stress" microarray studies in Thellungiella (Taji et al., 2004; Volkov et al., 2004). Furthermore, many well-known drought- and salt-responsive genes were differentially regulated by drought treatment but not by high salinity in Thellungiella. For example, the accumulation of raffinose, an osmolyte in response to drought, can be inferred from the increased level of the transcripts galactinol synthase (At1g56600) and raffinose synthase (At3g57520), but neither transcript was found in the dataset of salinity-induced transcripts. Raffinose family oligosaccharides have been proposed to act as osmoprotectants in some plant species to allow osmotic adjustment of plant cells exposed to water deficit (for review, see Rathinasabapathi, 2000). Other proposed roles for these compounds include free radical scavenging, protection from photoinhibition, and metabolic detoxification (Orthen et al., 1994; Pharr et al., 1995), all of which may help a plant to survive various environmental stresses. However, the actual role of these osmolytes in cellular stress tolerance remains equivocal and has recently been challenged (Tunnacliffe and Lapinski, 2003; Zuther et al., 2004).

Nevertheless, the low degree of mRNA changes in salt-acclimated Thellungiella is intriguing. The degree of fold-change for salinity stress-responsive transcripts relative to the control under our experimental conditions may be far less pronounced and hence lie beyond the detection sensitivity of microarray analysis. While it is possible that more dramatic transcriptional changes occur in the roots instead of leaves or at an earlier time point following salt treatment, this seems unlikely given that microarray experiments using roots and 2-h salt-treated tissues have been performed by others with similar results (Taji et al., 2004; Volkov et al., 2004). An alternative explanation is offered by the report that salt tolerance in Thellungiella is associated with specific features of high selectivity for K⁺ accumulation and not Na⁺ after NaCl application, a response that is unlike its glycophytic relative Arabidopsis (Volkov et al., 2004). Thus, this predicts that the innate ability of Thellungiella to maintain ion homeostasis may allow its adaptation to a high-salinity environment with few transcriptional changes, in contrast to Arabidopsis. The remarkable halophytic nature may also be closely linked to the finding that many of these stress-inducible genes in Arabidopsis are constitutively overexpressed in Thellungiella and hence offered a greater innate tolerance to salt stress (Taji et al., 2004).

Abiotic Stress and Photosynthesis

The photosynthetic system in higher plants is highly susceptible to abiotic stresses (Huner et al., 1998). Under low temperatures, for instance, absorbed light energy cannot be used productively due to the inhibition of photosynthetic CO₂ assimilation and other metabolic processes (Huner et al., 1998). Such energy imbalance leads to an overexcitation of the photosynthetic apparatus that in turn increases the potential for photoinhibition and subsequently photooxidative damage (Huner et al., 1998). To avoid the energy imbalance resulting from abiotic stress, plants are able to down-regulate genes associated with photosynthetic light harvesting to reduce light energy absorbed. It is therefore not surprising that we found various photosynthesis-related genes repressed in our datasets, an observation that is consistent with previous similar microarray studies (Fowler and Thomashow, 2002; Kreps et al., 2002; Seki et al., 2002b). In fact, these genes constitute the largest functional group of transcripts (15%) identified to be down-regulated in our analysis. It is established that in Arabidopsis, transcripts encoding products associated with photosynthesis, such as chlorophyll a/b-binding proteins and Rubisco, decrease after transfer to low temperature (Strand et al., 1997). However, photosynthesis is able to recover in plants that develop at low temperature, in part due to a reprogramming of photosynthetic carbon metabolism (Strand et al., 1997; Stitt and Hurry, 2002; Gray and Heath, 2005). Stress responses often initiate transient physiological, biochemical, and molecular perturbations (Huner et al., 1993; Stitt and Hurry, 2002). These transient responses lead to stable, long-term adjustments that reflect complex developmental responses to the new conditions (Huner et al., 1993, 1998; Gray et al., 1996; Stitt and Hurry, 2002). Our observation that Thellungiella developed at low temperature exhibits an extremely high tolerance to photoinhibition would indicate that photosynthetic acclimation to low temperature had occurred (Fig. 3A). Whether the mechanism for this unusually high tolerance to photoinhibition is similar to that utilized by Arabidopsis (Gray et al., 2003) will require further experimentation.

Among the photosynthesis-related genes that were repressed are those that encode Rubisco subunits, oxygen-evolving complex, PSI and PSII subunits, and chlorophyll a/b-binding proteins. However, only two members of this group of genes were consistently down-regulated by all three stress treatments, and both of them belong to the chlorophyll a/b-binding protein family. The fact that we did not see all photosynthesis-related genes repressed in the same manner across all three stresses may reflect the different impact that these stresses have on photosynthesis in the short as well as the long term. For example, in Arabidopsis, down-regulation of photosynthetic gene expression after cold treatment is associated with the accumulation of hexose phosphate and soluble sugars (Strand et al., 1997, 2003). This is similar to the carbohydrate-mediated repression that is observed when sugars accumulate for other reasons (Krapp et al., 1993; Jang and Sheen, 1994; Sheen, 1994; Krapp and Stitt, 1995). In our analysis, we have observed an increased level of transcripts associated with Suc synthase under cold or drought treatment that may correlate with the carbohydrate-mediated repression of photosynthesis reported upon shifting plants to the cold (Strand et al., 1997, 2003).

Unlike light-harvesting proteins that are down-regulated by drought, cold, and high-salinity treatments, Thellungiella accumulates transcripts of ELIP to different levels in response to cold and drought treatment (At4g14690 and At3g22840; Cluster 2; Table II). ELIPs accumulate in thylakoid membranes during light stress, where they bind chlorophylls and carotenoids and provide protection against photooxidation (Montané and Kloppstech, 2000; Hutin et al., 2003). NPQ is a complex process considered to be a major photoprotective mechanism in all plant species by which excess light energy is released as heat (Müller et al., 2001). In barley (Hordeum vulgare), the accumulation of ELIPs is correlated with greater NPQ (Potter and Kloppstech, 1993; Krol et al., 1999). It is tempting to speculate that the ability of Thellungiella to accumulate ELIPs in response to cold or drought treatment serves a role in the dissipation of excess light energy early in the establishment of stress tolerance. However, when we examined the ability of plants developed at low temperature to generate NPQ in comparison to those grown under standard growth condition, minimal differences were observed (Fig. 3B). Thus, it is unlikely that this would account for the increased tolerance to photoinhibition that we also observed in these plants (Fig. 3A). This may suggest that alternative mechanisms exist in Thellungiella for dealing with energy imbalances at the level of photosynthesis in response to low temperature.

ABA and Drought/Salinity-Induced Responses

The phytohormone ABA is involved in regulating diverse plant processes, including gene expression during abiotic and biotic stresses. More specifically, ABA is known to mediate the expression of many stress-responsive genes as a result of drought or saline conditions, but may have a limited role in cold regulation (for review, see Zhu, 2002; Shinozaki et al., 2003). The apparent up-regulation of several genes known to be ABA responsive among products in our dataset associated with drought offers indirect evidence for the involvement of ABA in regulating the response of Thellungiella to drought.

There was a concomitant decrease of various defense-related transcripts noted in the drought and salinity datasets. Interestingly, ABA has been found to enhance disease susceptibility (Ward et al., 1989; McDonald and Cahill, 1999; Mohr and Cahill, 2003) and lead to a down-regulation in the expression of disease-resistance genes (Rezzonico et al., 1998; Audenaert et al., 2002; Anderson et al., 2004). Together, these correlative changes suggest that the down-regulation of defense-related transcripts observed in our study is likely associated with an increase in endogenous ABA levels in Thellungiella in response to osmotic stress, although this has yet to be directly verified.

Intriguingly, a number of reports have shown an induction of defense-related genes by similar stresses in Arabidopsis (supplemental data in Seki et al., 2002a; supplemental data in Seki et al., 2002b; Zimmermann et al., 2004). It is conceivable that the ability of this Thellungiella ecotype to tolerate semiarid conditions in the Yukon may be associated with its strategy to more precisely coordinate the antagonistic relationship between osmotic and biotic stress responses in comparison to plants that have a lower innate tolerance of stress, like Arabidopsis. A more recent study using a 25 K 70-mer Arabidopsis oligoarray that compared the transcript profiles of Arabidopsis and the Shandong ecotype of Thellungiella subjected to NaCl treatment have reached similar conclusions (Gong et al., 2005). It is likely that ABA is among the key players that allow the employment of a specific rather than general defense strategy. Although ABA has been known to mediate osmotic stress responses in Arabidopsis, there are reports that ABA-biosynthesis as well as ABA-responsive genes are more highly expressed in Thellungiella than in Arabidopsis (Taji et al., 2004; Gong et al., 2005).

The relatively lower percentage of ABA-responsive genes in our cold dataset in comparison to drought, in particular, is consistent with the emerging consensus that ABA may not play a critical role in the cold response (Seki et al., 2002a; Xiong and Zhu, 2002). For ABA-responsive genes induced by both cold and drought treatments, such as Adh, previous studies in Arabidopsis have shown induction by drought to be, in part, dependent on the action of ABA but the cold effect appears to be independent of ABA (Bruxelles et al., 1996).

Cell Wall-Related Transcripts

A number of transcripts encoding cell wall-related proteins are found in the drought-responsive transcript dataset. Transcripts encoding a xyloglucan:xylogycosyl transferase (At2g06850) and a putative endotransglycosylase (At5g65730) were down-regulated, whereas some of those encoding a potentially cell wall-associated LTP family of proteins were up-regulated (At4g33550, At5g59310) by drought treatment. Similar proteins have been implicated in cell wall loosening (Nishitani and Tominaga, 1992; Nieuwland et al., 2005).The plant cell wall plays a critical role in cell enlargement, which is an essential process in plant growth and development (Cosgrove, 2001). In addition, it has been documented that increased cell wall plasticity could potentially be a strategy employed in ameliorating the shear forces that arise with the massive loss of volume in mesophyll during drought in grasses (Balsamo et al., 2005). Perhaps this change in the cell wall can also provide a way to prevent extensive water loss during drought.

A possible explanation for the apparent antagonistic regulation of the two groups of cell wall-related transcripts observed in the drought dataset is that developmental and environmental cues may trigger the action of different groups of cell wall-loosening enzymes. In fact, there are two classes of enzymes involved in the cell wall-loosening process (for review, see Cosgrove, 2001). First, there are the primary enzymes that are directly involved in cell wall loosening as well as extension, which takes place during growth and development (Cosgrove, 2001). There are also secondary enzymes that act to weaken the wall without inducing wall extension (Cosgrove, 2001). There is minimum growth during drought and it is therefore conceivable that transcripts coding for the primary wall-loosening and extension proteins are repressed, while those that are responsible mainly for loosening the cell wall would be up-regulated, reflecting the need of Thellungiella for plastic cell walls in coping with drought stress. In line with this hypothesis is the down-regulation of two transcripts predicted to form the structural constituent of cell walls (At2g10940 and At1g21120) in the drought dataset.

JA Is Implicated in Mediating Cold Responses

Our analysis revealed transcriptional changes of some JA-related gene products (LOX, PDF1.2a, chalcone synthase) after cold treatment. The current knowledge of JA-related genes derives mostly from work on defense mechanisms during pathogenesis and wounding, and to some extent osmotic stress (for review, see Wasternack et al., 1998). A number of studies report that JA application in the form of methyl jasmonate could be used to reduce chilling injury in fruits (Gonzalez-Aguilar et al., 2004). It is likely that tissue damage associated with low temperature in chilling-sensitive plants induces responses similar to those occurring in pathogen-wounded plants. It is known that the primary site of injury (disruption) during a freeze-thaw cycle is the cellular membranes, particularly the plasma membrane (Steponkus, 1984). Furthermore, mechanical wounding, a process known to involve JA-mediated responses, has been reported to induce the cold regulatory pathway mediated by the CBF transcription factor in Arabidopsis (Cheong et al., 2002). These considerations and the correlative evidence from our analysis point to a potentially novel and exciting regulatory role of JA during cold responses in Thellungiella. Whether JA plays a critical role in the increase in freezing tolerance that occurs with cold acclimation of Thellungiella has to be examined further with metabolomics analysis and reverse-genetic approaches.

Transcript Changes Associated with Rewatering

For a plant to survive a drought stress, not only does it have to overcome drought-associated damages but also injuries that occur upon rewatering as a result of rapid water uptake (Stewart, 1989). Our experiments attempted to obtain insight into processes associated with the recovery process from a water deficit to gain a better understanding of the capacity of Thellungiella to recover from this stress. It is interesting to observe two patterns of expression for some well-known drought-responsive genes following rewatering after drought. Some transcripts, such as those for cor15b and ERD14, remained highly expressed 2 d after rewatering, whereas others, including KIN2 and COR47, were down-regulated upon rewatering. Whether these differences identify genes that play a direct role in recovery from a water deficit needs further investigation.

	CONCLUSION

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Our microarray analysis provides transcript profiles of drought, cold, high-salinity, or rewatering responses in Thellungiella. This analysis is a timely contribution to the emerging recognition of Thellungiella as a model species for the molecular elucidation of abiotic stress tolerance. Our results revealed interesting features and potentially valuable traits associated with the stress responses of Thellungiella. However, as with all microarray analyses, the interpretation of the transcript changes requires caution. Microarrays provide a sensitive but transient snapshot of gene expression. Furthermore, the changes in mRNA levels may not correlate with changes in protein or enzyme activity levels. Nevertheless, the results obtained from our microarray analysis provide useful starting points for more thorough investigation into the molecular mechanism behind the extreme stress tolerance of Thellungiella.

	MATERIALS AND METHODS

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 

Plant Materials and Stress Treatments

Plants of the Yukon ecotype of Thellungiella salsuginea (Pall.) O.E. Schulz (Al-Shehbaz et al., 1999; Cody, 2000) were grown in controlled environments with a day/night temperature regime of 22°C/10°C. An irradiance of 250 µmol photons m^–2 s^–1 over a 21-h daylength was provided. When the plants were 4 weeks old, they were subjected to stress treatments as described below. Plants were harvested 8 h after the lights came on in the growth chamber for the microarray experiment.

For the cold treatments, plants were shifted to a day/night temperature regime of 5°C/4°C with all other conditions constant and leaves were sampled after 3 weeks. For the photosynthetic experiments, the plants were allowed to develop under these conditions for an additional 8 weeks. Drought stress was simulated in pots by withholding water from 1-month-old plants until they wilted visibly (3 d). For the drought and rewatering treatment, plants were subjected to drought treatment and then rewatered and allowed to recover for 2 d. To have plants acclimated to high salinity, plants were watered with NaCl solutions at concentrations that increased by 50 mM increment every 3 d until the final concentration reached 300 mM; tissues were harvested 3 d later.

Physiological Characterization of Simulated Drought and Salinity Treatments

Water and solute potentials and relative water content measurements were carried out for well-watered control plants, plants that were subjected to simulated drought or saline conditions, or rewatering after water deficits as described by Weretilnyk et al. (2001).

Photosynthetic Measurements and Photoinhibition

F_v/F_m and experiments examining the rapid induction of NPQ were determined in planta at room temperature using a PAM-2000 portable chlorophyll fluorometer (Heinz Walz GmbH) as detailed by Baerr et al. (2005).

Photoinhibition of photosynthesis was induced at 4°C by exposure to a photon flux density of 1,000 µmol photons m^–2 s^–1 at the leaf surface for 4 h as described previously (Gray et al., 2003). Susceptibility to photoinhibition was quantified by monitoring changes in F_v/F_m.

RNA Extraction

Total RNA was extracted from only aboveground tissues as described previously (Wong et al., 2005) and were then purified using RNeasy columns (Qiagen). Three independent tissue collections and RNA extractions were performed for each of the treatments.

Amplification of cDNA Inserts and Microarray Preparation

A total of 3,628 unique clones derived from the abiotic stress libraries described previously (Wong et al., 2005) were used for the array printing. Based on sequence similarity, each of these probe sets was assigned an Arabidopsis (Arabidopsis thaliana) MIPS protein code as described by Wong et al. (2005). Target DNAs for spotting on the microarray were amplified by PCR using the SP6 and T7 promoter primers that flanked both sides of the cDNA inserts. Plasmid template (5 ng) was added to a 60-µL PCR mixture containing a total of 15 mM of each nucleotide, 60 µM of each primer, 120 mM MgCl₂, 600 mM Tris-HCl, pH 8.8, 3 M KCl, 0.48% Nonidet P-40, and 3 units of Taq polymerase (MBI Fermentas). Inserts were amplified using 40 cycles (94°C for 30 s, 56°C for 45 s, 72°C for 2 min) with an initial denaturation at 94°C for 2 min and a final extension at 72°C for 5 min. One microliter of each finished reaction was electrophoresed on a 1% agarose gel to confirm amplification quality and quantity. PCR products were then sent for purifications and printing on Corning GAPS II coated slides at the MicroArray Lab at the Biotechnology Research Institute (BRI), Montreal. Each target DNA was spotted in triplicate. A variety of control elements were also arrayed on the slide. These included multiple blank spots, buffer-only spots, and control DNA sequences derived from three transgenes (bacterial ACC deaminase gene, and fish chemokine and IL-1 genes), housekeeping genes (actin and tubulin), and known abiotic stress-induced genes (cor47, cor15a, rd20, and cbf1).

Fluorescent Probe Preparation

Forty micrograms of total RNA was reverse transcribed to synthesize aminoallyl-labeled cDNA, followed by coupling of the aminoallyl groups to either Cyanine 3 or 5 (Cy3/Cy5) fluorescent molecules using protocol from The Institute of Genome Research (http://www.tigr.org/tdb/microarray/protocolsTIGR.shtml). Anchored poly d(T) primers at a final concentration of 4 µM were used to prime the first-strand cDNA synthesis. Routinely, 2 ng of fish IL-1 mRNA was spiked into the cDNA synthesis. The spiking probe added to labeling permitted an independent cross-check of the results.

Hybridization and Washing

The cDNA microarrays were hybridized with Cy3 and Cy5 fluorescently labeled probe pairs of untreated/drought 3 d, untreated/cold acclimated, untreated/salt acclimated, and rewatering after drought/drought 3 d. Three biological replicates with dye swap as technical replicate were examined for all experimental conditions. The statistical analysis was based on a total of six replicates per experimental condition (three biological replicates plus three technical replicates).

Microarrays were prehybridized, hybridized, and washed according to the protocol provided by the MicroArray Lab at BRI (http://www.bri.nrc.gc.ca/microarraylab/labelling_en.pdf). An extra step of dipping the slides in isopropanol was performed prior to air drying the slides.

Scanning and Data Analysis

ScanArray confocal scanning system and QuantArray data acquisition software (Perkin-Elmer) were used according to the manufacturer's instructions to capture the data. Normalization between the Cy3 and Cy5 fluorescent dye emission intensities was achieved by adjusting the level of the photomultiplier gains. The signal intensities for all 14,688 spots were quantified and local background fluorescence values were automatically calculated by the QuantArray program. After local background subtraction, the data were imported into an algorithm that used a Bayesian approach to assess the significance of each treatment for each gene (Labbe, 2005; A. Labbe and M.E. Thompson, unpublished data). This methodology combines Bayesian analysis and frequentist theory in the context of multiple comparisons. Using the posterior probability of overexpression as a test statistic, we are taking advantage of the flexibility brought by Bayesian models. This probability is eventually regarded in the same spirit as a P value in the Benjamini-Hochberg procedure, to control for the multiplicity of the tests. By doing so, we are then taking advantage of both the control of false discoveries by using some well-known frequentist procedures and the flexibility brought by the Bayesian framework. The algorithm was implemented in an R environment for statistical analysis. Median values of triplicate spots for all probe sets were used in all calculations.

To correct for intensity-dependent bias and spatial bias simultaneously, a data transformation method called the Joint Lowess method (Cui et al., 2003) was used to normalize the data. The transformation was applied to each array individually. After normalization, differentially expressed genes between each control and treatment conditions were detected using a Bayesian model. The well-known gamma conjugate model, as described by Kendziorski et al. (2003), was fitted to the data, and parameters were obtained in an empirical Bayes manner. Note that different parameters were considered for each of the four treatment conditions and analyses were carried out separately. For each treatment, posterior probabilities of gene overexpression were computed for each gene. It has been shown that this type of probability shares the same property of uniformity with a P value under the null hypothesis of no differential expression (Dudley and Haughton, 2002). Such probabilities can thus be used as an input in the Benjamini-Hochberg procedure, which usually takes P values as an input (Benjamini and Hochberg, 1995), to control the false discovery rate. A false discovery rate of 1% was chosen for this experiment to address the problem of multiple testing. Considering these posterior probabilities in the same spirit as a P value enables us to combine the best of each approach (frequentist and Bayesian) in a unique procedure. The mathematical validity of this approach has been demonstrated (Labbe, 2005; Labbe and Thompson, 2005).

The list of genes identified to be significantly up- or down-regulated was then clustered using STATISTICA (version 6.1; Stats Soft). A Euclidean matrix was used to calculate the distances, and a complete linkage method was then applied for hierarchical clustering of these genes.

RT-PCR Analysis

First-strand cDNA was prepared from 5 µg of total RNA with the Superscript RT II kit (Invitrogen) and oligo(dT)₁₈ according to the manufacturer's instructions. A 0.5-µL aliquot of the total RT reaction volume (20 µL) was used as template in a 25-µL semiquantitative RT-mediated PCR amplification reaction. Number of cycles used for the transcripts investigated was routinely between 20 and 25, ensuring that the amount of amplified product remained in linear proportion to the initial template present in the reaction. Eight microliters of the PCR reaction was separated on 1% agarose gel containing 0.1 µg/µL ethidium bromide and visualized under UV light. The amount of amplified product was then estimated semiquantitatively using FluorChem Imaging System according to the manufacturer's instructions (Alpha Innotech).

Primers used were At1g56600F, CGAACCGGTCTTTGCAGCCACCG and At1g56600R, TGACGTACTGAAGCACACCGGCC; At5g59310F, AACCTTAGAAAACAAAAGTCAACTAAATCT and At5g59310R, CTAGACTTGGGTTGATGCTCTGT; At1g06430F, GGAGCTGATCTTGCAAACCTCTTG and At1g06430R, AGCCGACGAATCCATTAGCGAC; At5g52300F, ACAGGATCAGCCGTGATGACG and At5g52300R, TATTTTCCCCCTCCAATTCC; At5g66400F, GGTGGTTGCTCTTCCAACATGGCGTCTTACCAGAA and At5g66400R, CGGAATTCTTAACGACCACCACCACCAGGAAGTTTATC; and At4g05320F, AGTCGACCCTTCATTTGGTG and At4g05320R, TGATAAAGTAAGCAGGCGC.

	ACKNOWLEDGMENTS

 
We thank Wilson Sung for assistance in managing the microarray datasets, Jeffrey Pylatuik for providing tips on improving fluorescence probe labeling efficiency, Kazuhiro Fujiki for providing the clone for fish transgene, and Kevin Lanctot for writing an algorithm to convert QuantArray output files into a format accepted by the microarray statistical analysis.

Received August 28, 2005; returned for revision December 10, 2005; accepted January 19, 2006.

	FOOTNOTES

 
¹ This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada, Agriculture and Agri-Food Canada, the Canola Council of Canada, the Food System Biotechnology Centre at the University of Guelph, the Ontario Genomics Institute, and Performance Plants.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Barbara Moffatt (moffatt{at}uwaterloo.ca).

^[W] The online version of this article contains Web-only data.

Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.105.070508.

^* Corresponding author; e-mail moffatt{at}uwaterloo.ca; fax 1–519–7460614.

	LITERATURE CITED

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Al-Shehbaz IA, O'Kane SL Jr, Price RA (1999) Generic placement of species excluded from Arabidopsis (Brassicaceae). Novon 9: 296–307[CrossRef][ISI]

Anderson JP, Badruzsaufari E, Schenk PM, Manners JM, Olivia JD, Ehlert C, Maclean DJ, Ebert PR, Kazan K (2004) Antagonistic interaction between abscisic acid and jasmonate-ethylene signaling pathways modulates defense gene expression and disease resistance in Arabidopsis. Plant Cell 16: 3460–3479[Abstract/Free Full Text]

Apel K, Hirt H (2004) Reactive oxygen species: metabolism, oxidative stress, and signal transduction. Annu Rev Plant Biol 55: 373–399[CrossRef][Medline]

Audenaert K, De Meyer GB, Hofte MM (2002) Abscisic acid determines basal susceptibility of tomato to Botrytis cinerea and suppresses salicylic acid-dependent signaling mechanisms. Plant Physiol 128: 491–501[Abstract/Free Full Text]

Baerr JN, Thomas JD, Taylor BG, Rodermel SR, Gray GR (2005) Differential photosynthetic compensatory mechanisms exist in the immutans mutant of Arabidopsis thaliana. Physiol Plant 124: 390–402

Balsamo RA, Willigen CV, Boyko W, Farrant J (2005) Retention of mobile water during dehydration in the desiccation-tolerant grass Eragrostis nindensis. Physiol Plant 124: 336–342

Bell E, Mullet JE (1993) Characterization of an Arabidopsis-lipoxygenase gene responsive to methyl jasmonate and wounding. Plant Physiol 103: 1133–1137[Abstract]

Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate. A practical and powerful approach to multiple testing. J R Stat Soc Ser B 57: 289–300

Bray E (2002) Classification of genes differentially expressed during water-deficit stress in Arabidopsis thaliana: an analysis using microarray and differential expression data. Ann Bot (Lond) 89: 803–811[Abstract/Free Full Text]

Bressan RA, Zhang C, Zhang H, Hasegawa P, Bohnert H, Zhu JK (2001) Learning from the Arabidopsis experience. The next gene search paradigm. Plant Physiol 127: 1354–1360[Free Full Text]

Bruxelles C, Peacock WJ, Dennis ES, Dolferus R (1996) Abscisic acid induces the alcohol dehydrogenase gene in Arabidopsis. Plant Physiol 111: 381–391[Abstract]

Cheong YH, Chang H, Gupta R, Wang X, Zhu T, Luan S (2002) Transcriptional profiling reveals novel interactions between wounding, pathogen, abiotic stress, and hormonal responses in Arabidopsis. Plant Physiol 129: 661–677[Abstract/Free Full Text]

Cody WJ (2000) Flora of the Yukon Territory, Ed 2. NRC Research Press, Ottawa

Cosgrove DJ (2001) Wall structure and wall loosening. A look backwards and forwards. Plant Physiol 125: 131–134[Free Full Text]

Cui X, Kerr MK, Churchill GA (2003) Data transformation for cDNA microarray data. Stat Appl Genet Mol Biol 2: Article 4

Dudley RM, Haughton D (2002) Asymptotic normality with small relative errors of posterior probabilities of half-spaces. Ann Statist 30: 1311–1344[CrossRef]

Fowler S, Thomashow MF (2002) Arabidopsis transcriptome profiling indicates that multiple regulatory pathways are activated during cold acclimation in addition to the CBF cold response pathway. Plant Cell 14: 1675–1690[Abstract/Free Full Text]

Gong Q, Li P, Ma S, Rupassara I, Bohnert HJ (2005) Salinity stress adaptation competence in the extremophile Thellungiella halophila in comparison to its relative Arabidopsis thaliana. Plant J 44: 826–839[ISI][Medline]

Gonzalez-Aguilar GA, Tiznado-Hernandez ME, Zavaleta-Gatica R, Martinez-Tellez MA (2004) Methyl jasmonate treatments reduce chilling injury and activate the defense response of guava fruits. Biochem Biophys Res Commun 313: 694–701[CrossRef][ISI][Medline]

Gray GR, Heath D (2005) A global reorganization of the metabolome in Arabidopsis during cold acclimation is revealed by metabolic fingerprinting. Physiol Plant 124: 236–248

Gray GR, Hope BJ, Qin X, Taylor BT, Whitehead CL (2003) The characterization of photoinhibition and recovery during cold acclimation in Arabidopsis thaliana using chlorophyll fluorescence imaging. Physiol Plant 119: 365–375

Gray GR, Savitch LV, Ivanov AG, Huner NPA (1996) Photosystem II excitation pressure and development of resistance to photoinhibition. II. Adjustment of photosynthetic capacity in winter wheat and winter rye. Plant Physiol 110: 61–71[Abstract]

Hong JK, Hwang BK (2002) Induction by pathogen, salt and drought of a basic class II chitinase mRNA and its in situ localization in pepper (Capsicum annuum). Physiol Plant 114: 549–558[CrossRef][Medline]

Huner NPA, Öquist G, Hurry VM, Krol M, Falk S, Griffith M (1993) Photosynthesis, photoinhibition and low temperature acclimation in cold tolerant plants. Photosynth Res 37: 19–39

Huner NPA, Öquist G, Sarhan F (1998) Energy balance and acclimation to light and cold. Trends Plant Sci 3: 224–230[CrossRef][ISI]

Hutin C, Nussaume L, Moise N, Moya I, Kloppstech K, Havaux M (2003) Early light-induced proteins protect Arabidopsis from photooxidative stress. Proc Natl Acad Sci USA 100: 4921–4926[Abstract/Free Full Text]

Inan G, Zhang Q, Li PH, Wang ZL, Cao ZY, Zhang H, Zhang CQ, Quist TM, Goodwin SM, Zhu JH, et al (2004) Salt cress. A halophyte and cryophyte Arabidopsis relative model system and its applicability to molecular genetic analyses of growth and development of extremophiles. Plant Physiol 135: 1718–1737[Abstract/Free Full Text]

Jang JC, Sheen J (1994) Sugar sensing in higher plants. Plant Cell 6: 1665–1679[Abstract]

Kendziorski CM, Newton MA, Lan H, Gould MN (2003) On parametric empirical Bayes methods for comparing multiple groups using replicated gene expression. Stat Med 22: 3899–3914[CrossRef][ISI][Medline]

Krapp A, Hofmann B, Schafer C, Stitt M (1993) Regulation of the expression of rbcs and other photosynthetic genes by carbohydrates: a mechanism for the sink regulation of photosynthesis. Plant J 3: 817–828[CrossRef][ISI]

Krapp A, Stitt M (1995) An evaluation of direct and indirect mechanisms for the sink-regulation of photosynthesis in spinach: changes in gas-exchange, carbohydrates, metabolites, enzyme-activities and steady-state transcript levels after cold-girdling source leaves. Planta 195: 313–323[ISI]