Proteome analysis of Physcomitrella patens exposed to progressive dehydration and rehydration.

Physcomitrella patens is an extremely dehydration-tolerant moss. However, the molecular basis of its responses to loss of cellular water remains unclear. A comprehensive proteomic analysis of dehydration- and rehydration-responsive proteins has been conducted using quantitative two-dimensional difference in-gel electrophoresis (2D-DIGE), and traditional 2-D gel electrophoresis (2-DE) combined with MALDI TOF/TOF MS. Of the 216 differentially-expressed protein spots, 112 and 104 were dehydration- and rehydration-responsive proteins, respectively. The functional categories of the most differentially-expressed proteins were seed maturation, defence, protein synthesis and quality control, and energy production. Strikingly, most of the late embryogenesis abundant (LEA) proteins were expressed at a basal level under control conditions and their synthesis was strongly enhanced by dehydration, a pattern that was confirmed by RT-PCR. Actinoporins, phosphatidylethanolamine-binding protein, arabinogalactan protein, and phospholipase are the likely dominant players in the defence system. In addition, 24 proteins of unknown function were identified as novel dehydration- or rehydration-responsive proteins. Our data indicate that Physcomitrella adopts a rapid protein response mechanism to cope with dehydration in its leafy-shoot and basal expression levels of desiccation-tolerant proteins are rapidly upgraded at high levels under stress. This mechanism appears similar to that seen in angiosperm seeds.

Introduction

Bryophyte-like organisms were the first terrestrial plants that emerged more than 450 million years ago (Kenrick and Crane, 1997). Evolving from an aquatic, green algae-like ancestor, development of morphological and physiological features was principally aimed at adaptation to water deficiency for land colonization by these plants (Kenrick and Crane, 1997; Bateman ; Qiu ). As a representative bryophyte, Physcomitrella patens (hereafter Physcomitrella) has been recognized as a model system for the study of basic cytology and development in plant biology, as well as for understanding the molecular mechanisms during evolution that permitted the original bryophyte to cope with terrestrial habitats (Rensing ). Like other mosses, the life cycle of Physcomitrella is characterized by an alternation of two generations. Morphologically, dominant gametophytic generation is subdivided into two distinct developmental stages, i.e. protonema and gametophore. The former is composed of filamentous cells; the latter is made up of leafy-shoot tissues. It has been demonstrated that Physcomitrella can survive severe dehydration of their vegetative organs (e.g leafy-shoots), which is uncommon in higher vascular plants (Frank ; Oldenhof ; Cuming ). However, little is known about the molecular features responsible for the dehydration tolerance of leafy-shoots, the green haploid vegetative tissues, of Physcomitrella.

In general, tolerance to dehydration or complete desiccation is very common among mosses. Vegetative tissues of some species can withstand complete desiccation (drying to a water content of 10% or less), and then resume normal growth upon rehydration. A well-studied experimental model is moss, Tortula ruralis. It was reported that T. ruralis can tolerate a complete loss of free protoplasmic water (Wood and Oliver, 2004). Its cellular protection system was demonstrated to be essentially constitutive and contained key components of the system characterized within vascular plants, such as dehydrin protein and sugar (Oliver, 1991; Wood and Oliver, 2004). Moreover, the synthesis of proteins required for defence and repair occurred only under rehydration conditions (Oliver, 1991). Summarizing these results, Wood and Oliver (2004) proposed that vegetative tolerance in moss plants involved constitutive cellular protection coupled with a repair mechanism that is induced upon rehydration. However, the hypothesis seems to be controversial (Cuming ; Proctor ).

Compared with T. ruralis, vegetative tissues of Physcomitrella cannot tolerate complete desiccation, protonemal tissues in particular (Oldenhof ; Cuming ; Khandelwal ). Interestingly, its gametophore colonies survive in desiccated environments (Wang XQ et al., 2009). The tolerance discrepancies among tissues and species need to be clarified. Therefore, it is very necessary to explore the molecular details that contribute to the dehydration-tolerance response in various tissues of Physcomitrella. A survey based on microarray revealed that, in protonemal tissue of Physcomitrella, 130 genes are specifically induced by dehydration (Cuming ). The set of induced genes includes many members encoding late embryogenesis abundant (LEA) proteins and other homologues of angiosperm genes expressed during drought stress. In gametophore colonies of Physcomitrella, Frank identified 19 genes responsive to dehydration using cDNA macroarray. These genes include homologues that are known to be associated with abiotic stress in different species of vascular plants (Frank ). Proteomic analysis of the gametophore tissues revealed that, under desiccation conditions, the abundance of an additional 71 proteins is altered to cope with desiccation events (Wang XQ et al., 2009). Summarizing the previous results, the strategy followed by Physcomitrella for coping with dehydration/desiccation seems to be to mobilize the defence and repair system upon dehydration, which is in contrast to that seen in T. ruralis, but is similar to that seen in angiosperms. However, only a few dehydration-responsive genes/proteins belonging to a few functional categories have been identified in Physcomitrella (Frank ; Wang XQ et al., 2009). Furthermore, when microarray data of RNA expression in protonemal tissue (Cuming ) were compared with proteomics data from a gametophore colony (Wang XQ et al., 2009), huge discrepancies were observed between responses at the RNA and protein levels. Therefore, the mechanisms of dehydration/desiccation tolerance in vegetative tissues of Physcomitrella remain unclear.

Leafy-shoots of Physcomitrella are haploid organisms with stem, leaves, and filamentous rhizoids, which are morphologically analogous to the shoot of angiosperms. It was observed that these haploid plants can recover from severe dehydration. The molecular mechanisms underlying this phenotype remain poorly understood. Here, 2-D DIGE, plus traditional 2-DE coupled with mass spectrometry, was applied to identify dehydration- and rehydration-responsive proteins in Physcomitrella leafy-shoots. The earlier procedure for protein extraction (Cui , 2009) was modified to minimize losses and to optimize it for quantitative proteomics. The proteomic results were further confirmed using RT-PCR analysis of total RNA. Analysis of all the results taken together reveals a possible mechanism of cellular protection during dehydration in Physcomitrella leafy-shoots.

Materials and methods

Plant material cultivation

Physcomitrella patens ecotype ‘Gransden 2004’ was cultured on modified BCD medium (Nishiyama ) in a growth chamber at 23 °C with a 16 h photoperiod, and a light intensity of 150 μmol m−2 s−1. To obtain uniform leafy-shoots, 2-week-old gametophore colonies (mixed tissues of leafy-shoots and protonema) were ground gently with a homogenizer. The homogenate was resuspended in liquid BCD medium (1 mM, MgSO4, 10 mM KNO3, 45 μM FeSO4, 1.85 mM, KH2PO4, 1 mM, CaCl2, and trace elements, pH 6.5) and was transferred to a cellophane overlay on solid BCD medium containing 0.8% (w/v) agar, supplemented with 5 mM ammonium tartrate and 0.5% (w/v) glucose for protonema cultivation. After 8 d, the cellophane overlay bearing regenerated protonemal tissues was transferred onto ammonium tartrate-free BCD medium to allow the growth of leafy-shoots. After 20 d, leafy-shoots with stem, 7–9 fully extended leaves, and rhizoids were obtained and used in the study.

Physiological analysis of dehydration tolerance in leafy-shoots

To determine their tolerance for water deficiency, 20-d-old leafy-shoots were subjected to a progressive dehydration stress followed by rehydration. For dehydration treatment, leafy-shoots grown on cellophane were transferred to a flat beaker with nine layers of filter papers wetted with 1 ml BCD medium to avoid exposure to an abrupt change in humidity at the onset of the treatment. The beakers were then placed in a small growth chamber with 30% relative humidity, a 16 h photoperiod, and a light intensity of 150 μmol m−2 s−1. After dehydrating naturally and progressively for 3–4 d at 23 °C, the leafy-shoots were transferred to fresh solid BCD medium and allowed to revive for 6 d. During dehydration and rehydration, both water and chlorophyll content in leafy-shoots were determined at various time points according to a previously described method (Frank ). For every time point, six measurements were performed.

Extraction of total protein for 2-D DIGE

Total protein of leafy-shoots was extracted using a fractionation method (Cui , 2009) with modifications. Approximately 1 g of leafy-shoots was ground to a fine powder in liquid nitrogen. First, soluble proteins were extracted with buffer I (50 mM TRIS-HCl, pH 7.8, 10% glycerol, 2% β-mercaptoethanol, 1 mM PMSF, and 1 mM EDTA). After ultrasonication and centrifugation, the supernatant was collected in a test tube. In a second step, insoluble proteins associated with cellular structures and membrane were extracted with buffer II (100 mM phosphate buffer, pH 7.1, containing 0.2 M KCl, 10% glycerol, 2 mM MgSO4, 4% (w/v) CHAPS, 2% β-mercaptoethanol, 1 mM PMSF, and 1 mM EDTA), following ultrasonication and further extraction with 7 M urea, 2 M thiourea, and 30 mM DTT. After centrifugation, the supernatant was combined into the test tube. Subsequently, the combined supernatant was mixed with an equal volume of ice-cold phenol (Tris-buffered, pH 7.5) and centrifugation at 20 000 g for 30 min at 4 °C to separate the phenol and aqueous phases. The collected phenol phase was then mixed with 5 vols of cold, saturated ammonium acetate in methanol and left at –20 °C overnight to precipitate proteins. After centrifugation, the protein pellet was rinsed twice with cold acetone containing 13 mM DTT and then lyophilized. The resulting protein samples were used for quantitative DIGE analysis and traditional 2-DE. Protein content was estimated using a modified Bradford method as described previously by Cui . In addition, the total protein content of leafy-shoots was also determined using a modified Kjeldahl analysis (Gornall ; Wang YX et al., 2009).

Protein CyDye labelling

Proteins from both types of samples, untreated control leafy-shoots (20-d-old) or the dehydrated (3 d dehydration) and rehydrated leafy-shoots (6 d rehydration) were labelled with DIGE-specific Cy3 or Cy5 according to the manufacturer’s instructions (GE Healthcare) with the following modifications. Briefly, after resuspension in lysis buffer (7 M urea, 2 M thiourea, and 4% CHAPS) and adjusting the pH to 8.5, 50 μg of proteins were mixed with 300 pmol of CyDye and incubated on ice in the dark for 30 min. The labelling reaction was stopped by the addition of lysine. In the experiments, biological replicates were labelled reciprocally with either Cy3 or Cy5, and an internal standard was generated by pooling equal amounts of proteins from each sample labelled with DIGE-specific Cy2.

2-D DIGE and image analysis

Pairs of Cy3- and Cy5-labelled protein samples and a Cy2-labelled internal standard were mixed together in a 1:1:1 ratio and then subjected to IEF and SDS-PAGE as described previously by (Cui . Measurements were done in four replicates. The resulting gels were scanned using Typhoon 9400 Imager (GE Healthcare). The images were analysed using DeCyder v6.5 software according to the manufacturer’s user guide. Spot intensities were normalized based on the internal standard labelled with Cy2. Differentially-expressed spots were selected based on the P value of the t test (P <0.05); the presence in all replicates; and spot abundance ratio of >1.5.

Traditional 2-DE and image analysis

Traditional 2-DE was used to analyse the change of proteins extracted from the above plant samples again. For iso-electric focusing (IEF), 750 μg of proteins were re-suspended in IEF buffer and then loaded on to an 18 cm immobilized pH gradient (IPG) strip with a linear pH gradient ranging from 4 to 7. Following SDS-PAGE, gels were stained with Coomassie Brilliant Blue (CBB), and image and data analysis were carried out using our published protocol (Cui ). At least nine gels derived from three biological replicates of each sample were compared. A spot abundance ratio of >1.5 was set as a threshold to identify differentially-expressed proteins in this study.

In-gel tryptic digestion, mass spectrometry, and database search

The spots of interest were picked from corresponding preparative CBB stained gels and submitted to in-gel digestion with trypsin (Cui ). Extracted peptides were analysed by MALDI-TOF/TOF on a mass spectrometer, Ultraflex III from Bruker Daltonics (Bremen, Germany) as previously described by Chen . For the acquisition of mass spectra, 0.5 μl samples were spotted onto a MALDI plate, followed by 0.5 μl matrix solution (4 mg ml−1 α-cyano-4-hydroxycinnamic acid in 35% acetonitrile and 1% TFA). Mass data acquisitions were piloted by flexcontrol software v3.0 using batched-processing and automatic switching between MS and MS/MS modes. All MS survey scans were acquired over the mass range m/z 800–4500 in the reflection positive-ion mode and accumulated from 2000 laser shots with an acceleration of 23 kV. The MS spectra were externally calibrated using PeptideCalibStandard II (Bruker Daltonics) (1046.542, 1296.685, 1347.735, 1619.822, 2093.086, 2465.198, and 3147.471), and internally calibrated using trypsin autolytic products (m/z 842.509, 1045.564, 1940.935, and 2211.104) resulting in mass errors of <30 ppm. The MS peaks were detected with a minimum signal/noise (S/N) ratio ≥20 and cluster area S/N threshold ≥25 without smoothening and raw spectrum filtering. Peptide precursor ions corresponding to contaminants including keratin and the trypsin autolytic products were excluded in a mass tolerance of ±0.2 Da. The filtered precursor ions with a user-defined threshold (S/N ratio ≥50) were selected for the MS/MS scan. Fragmentation of precursor ions was performed using the LIFT positive mode. MS/MS spectra were accumulated from 4000 laser shots. The MS/MS peaks were detected on a minimum S/N ratio ≥3 and a cluster area S/N threshold ≥15 with smoothing. Mass spectra were evaluated using FlexAnalysis software. A database search for protein identification was performed using Mascot (http://www.matrixscience.com) and the current NCBI databases. Search parameters allowed for mass accuracy of ±50 ppm, one miscleavage of trypsin, oxidation of methionine, and carbamidomethylation of cysteine. Proteins were identified as the highest ranking results deduced by searching in NCBI-nr databases of green plants. For unambiguous identification of proteins, a matching of more than five peptides and a sequence coverage >15% was considered adequate.

RNA extraction and semi-quantitative RT-PCR

Total RNA was isolated from Physcomitrella leafy-shoots using the E.Z.N.A. Plant RNA Kit (Omega Bio-tek, Inc.) according to the manufacturer’s protocol II. Semi-quantitative RT-PCR was carried out as previously described (Chen ). Reverse transcription reactions were carried out using Oligo(dT) primers and the RevertAid First Strand cDNA Synthesis Kit (Fermentas). The exponential phase of RT-PCR was determined by measuring the aliquots of PCR products taken after different number of PCR cycles. The primer sequences used for each gene are provided in Supplementary Table S1 at JXB online. Physcomitrella actin3 cDNA was used as a positive control.

Results and discussion

Leafy-shoots can survive around 75% water loss

To assess the dehydration tolerance of Physcomitrella, leafy-shoots were subjected to a progressive dehydration stress followed by rehydration. Figure 1 shows changes in both water and chlorophyll contents during dehydration and rehydration. After dehydration for 3 d, only 22.6% of the water content was retained in the leafy-shoots. In other words, the leafy-shoots had lost around 75% of their water content by the third day of dehydration treatment. A further prolongation by one day of the dehydration treatment led to 94% water loss (Fig. 1A). In contrast to the rapid decrease in water content, only a slight decrease in chlorophyll content was observed in leafy-shoots after 3 or 4 d of dehydration (Fig. 1B). Subsequently, water recovery experiments were performed. As shown in Fig. 1, during the rehydration phase there was complete recovery of the water content in leafy-shoots that had been dehydrated for 3 or 4 d. By contrast, the chlorophyll content appeared to decrease during rehydration. Recovery of the chlorophyll content was only partial in plants that had been subjected to a 3-d dehydration phase, while in those subjected to 4-d dehydration, the chlorophyll content decreased rapidly and continuously during the rehydration phase (Fig. 1b). A rapid loss of chlorophyll during rehydration has previously been reported in Physcomitrella (Frank ).

Fig. 1.

Dynamic changes in (A) water and (B) chlorophyll contents in leafy-shoots of Physcomitrella during dehydration and recovery treatments. Leafy-shoots (20-d-old) under standard conditions of growth (diamonds) were subjected to progressive dehydration for 3 d (squares) or 4 d (triangles) followed by transfer to fresh medium (times indicated by arrows) and allowed to rehydrate for up to 6 d. Values plotted are mean ±SD of six independent measurements.

The observation of tolerance to approximately 75% of water loss in leafy-shoots of Physcomitrella is consistent with earlier observations in protonemal tissues of Physcomitrella (Cuming ), but is slightly lower than the estimated 90–95% in gametophore colonies of Physcomitrella (Frank ; Oldenhof ; Wang XQ et al., 2009). In contrast to desiccation-tolerant moss species such as T. ruralis (Oliver ), Atrichum androgynum (Mayaba , 2002), and Polytrichum formosum (Proctor ), the leafy-shoots used in the current investigation could not tolerate the completely air-dried state nor could they recover after remoistening (Fig. 1). Hence, Physcomitrella has been categorized as a moss species with an intermediate tolerance between desiccation-tolerance and desiccation-sensitive.

Protein content and extraction efficiency

Using a modified Kjeldahl analysis, the protein content in leafy-shoots was determined to be only 6.8±0.4 mg g−1 of fresh tissue. To minimize protein loss and to obtain highly resolved 2-DE gels, in this study, soluble and insoluble proteins were extracted separately from Physcomitrella leafy-shoots according to a previously described method (Cui ) with a few modifications. This fractionation procedure coupled with phenol extraction (see the Materials and Methods) consistently yielded 6.4±0.7 mg of total proteins g−1 fresh leafy-shoots based on Bradford’s assay. This indicates that 94.1% of proteins in leafy-shoots could be extracted by this procedure. This protocol, optimized for moss protein extraction, also yielded a high resolution of protein spots in 2-DE gels.

Proteome profiles of Physcomitrella leafy-shoots in response to dehydration and rehydration

Physiological assays for the content of water and chlorophyll demonstrated that Physcomitrella leafy-shoots could recover from around 75% of water deficiency, which is uncommon in shoots of most vascular plants. To understand the molecular details of the dehydration resistance response better, proteins were extracted from Physcomitrella leafy-shoots at different times during the dehydration (3 d) and subsequent rehydration (6 d) treatments and subjected to a comprehensive proteomics analysis. 2-DE maps were generated for untreated control leafy-shoots, dehydrated and rehydrated leafy-shoots using two complementary techniques, i.e. quantitative 2-D DIGE and traditional 2-DE stained with CBB. The parallel experiments permitted both the identification of dehydration- or rehydration-responsive proteins that could not be visualized earlier due to protein dye bias (Cui ), and an evaluation of the consistency of results generated by the different techniques.

Figure 2 represents DIGE images of total proteins from Physcomitrella leafy-shoot, displaying significant alteration of proteome profiles during dehydration and rehydration. To eliminate bias due to technical variations, each gel included an internal standard prepared by pooling aliquots of all the samples within the experiment. A total of 2308 spots were detected and analysed. Of these 100 were dehydration-responsive protein spots (Fig. 2A) and 82 rehydration-responsive protein spots (Fig. 2B) that were significantly up-regulated or down-regulated in leafy-shoots subjected to 3 d dehydration and rehydration for 6 d. The differentially-expressed protein spots were consistent with those from traditional 2-DE analysis. Apart from these, an additional 12 dehydration-responsive and 22 rehydration-responsive protein spots (spot ratio >1.5) were also visualized on triplicate 2-D gels stained with CBB (see Supplementary Fig. S1 at JXB online) as previously described by Cui . Thus a total of 216 differentially expressed protein spots, including 112 dehydration-responsive and 104 rehydration-responsive, were identified in Physcomitrella leafy-shoots (Fig. 2A, B). As expected, there is a considerable overlap between the two sets of proteomic data, with 74 protein spots showing differential expression in response to both dehydration and rehydration.

Fig. 2.

2-D DIGE analyses of proteins in response to dehydration and rehydration in Physcomitrella leafy-shoots. DIGE gels were generated using 18 cm pH 4–7 IPG strips and a 12.5% SDS-PAGE gel. Pooled samples with Cy2 label (blue) were used as the internal standard. Protein spots of interest are marked with spot numbers and their identities are given in Tables 1 and 2. (A) Proteins in leafy-shoot subjected to dehydration for 3 d (labelled with Cy5, red) compared with those in untreated control samples (labelled with Cy3, green). Proteins up-regulated during dehydration appear pink, while those down-regulated appear turquoise. (B) Proteins in leafy-shoots after rehydration for 6 d (Cy3, green) were compared with those in untreated control samples (Cy5, red). Proteins up-regulated during rehydration appear turquoise, while those down-regulated during rehydration appear pink. Numbers on the left indicate the molecular mass in kDa, while those on top indicate the pH value.

Table 1.

Functional categories of dehydration- and rehydration-responsive proteins in leafy-shoots of Physcomitrella. Average values for the fold change (up-regulation or down-regulation) are shown for each protein spot on 2DE.

No	Accession no	Gene product	Spot number	Up-regulation foldsa		Down-regulation foldsa
				Drought	Rehydration	Drought	Rehydration
Seed maturation
1	XP_001769448	Group 3 LEA protein	207	3.97	1.99
			206	3.43
			205	2.65
			204	3.00
			203	2.09
			217	7.72	1.92
			216	13.70
			215	4.32
			801	2.51
2	EDQ69797c	Group 3 LEA protein	243	5.87	2.91
			240	5.24	2.26
			630	2.70	1.61
			805	2.72
			807	1.64
3	EDQ50092c	Dehydrin	643	1.59	1.57
			256	1.94	2.13
			251	2.90
			252	3.01
			250	3.75	1.53
4	EDQ58357d	Dehydrin	231	2.88
5	EDQ83710c	Group 3 LEA protein	270	2.36	1.95
6	EDQ65434c	Group 3 LEA protein	282	5.99	6.01
7	EDQ55306c	Group 3 LEA protein	289	8.79	6.21
8	EDQ65435c	Group 3 LEA protein	291	3.44	3.56
9	XP_001778223	Group 3 LEA protein	684	4.21	2.82
10	EDQ83709c	Group 3 LEA protein	403	5.43	2.64
11	XP_001777806	Group 3 LEA protein	410	8.06	5.64
Disease/defence
12	EDQ62837d	Physcomitrin	439	2.08	7.60
13	AAV65396d	Physcomitrin	408	3.27	7.40
14	XP_001782104	Physcomitrin	685	1.53	2.78
15	XP_001769177c	Phosphatidylethanolamine	284	9.13	3.13
		binding protein
16	XP_001760409	Arabinogalactan protein	449	4.42	2.26
17	EDQ72147	Phospholipase D	209	2.38	2.19
			210	2.49	2.42
18	EDQ55079	Dehydroascorbate reductase	277	12.53	12.51
19	EDQ56799	GDP-D-mannose-3',5'-epimerase	904			–1.75
			905			–1.52
20	EDQ63468	Aldo-keto reductase	694	10.86b	4.30b
21	EDQ57313c	Aldo-keto reductase	691		2.39b
22	EDQ80964	Thioredoxin peroxidase	279	6.28	4.87
23	XP_001757965	2-Cys peroxiredoxin	735				–1.59
24	ABF66648	Lipoxygenase-2	615		1.58b
25	XP_001769187c	Aldehyde dehydrogenase	627		1.50b
26	XP_001785650	Aldehyde dehydrogenase	636		1.51b
27	XP_001758337	Benzoquinone reductase	679		1.61
28	XP_001762556	Benzoquinone reductase	680		1.54
29	XP_001784490	Quinone oxidoreductase-like protein	720			–1.90b	–2.26b
Protein synthesis, destination and storage
30	XP_001773059	Translation initiation factor 5A	686	2.76
31	XP_001771324	Translation initiation factor 4A	648	3.20b	3.24b
32	XP_001780334	Putative chaperonin 60 beta precursor	806	3.75	1.81
33	XP_001783048	Heat shock protein 70	621	1.56
34	XP_001772650	Heat shock protein 70-2	618	1.90b
35	XP_001781229	Heat shock protein 70-3	624		2.54b
36	XP_001779894	Heat shock protein	702			–1.65
37	XP_001770511	Heat shock protein 90/GRP94	607	1.53	1.56
38	XP_001775725	Heat shock protein Hsp100	610	1.53b	1.62b
39	EDQ74403	Putative chloroplast FtsH protease	234	1.99
40	XP_001769853	Putative FtsH-like protein Pftf precursor	631	2.00
41	EDQ59103	Putative serine carboxypeptidase II	402	2.10	1.93
42	XP_001769359	Gamma interferon inducible lysosomal thiol reductase	812	3.18	2.64
			910				–1.50
43	XP_001784758	Protein disulphide isomerase-like	639	5.51b	4.77b
			640		1.59
44	XP_001756944	Calnexin 1	647	1.55
Energy (photosynthesis and respiration)
45	EDQ55432c	Carbonic anhydrase	312			–2.67	–1.91
		Carbonic anhydrase	728			–1.91	–1.68
		Carbonic anhydrase	730			–1.90	–1.73
		Carbonic anhydrase	729			–1.77	–1.56
46	EDQ82463c	Rubisco activase	304			–2.17
47	XP_001779467	Rubisco activase	305				–1.66
48	XP_001776035	Rubisco activase	908				–1.68
49	NP_904194	Rubisco large subunit	709			–2.22
			740			–1.53	–1.74
50	XP_001760166	Sedoheptulose bisphosphatase	722				–1.78b
51	XP_001765395	23 kDa subunit of oxygen-evolving	745				–1.70
		system of photosystem II
52	XP_001752812	Chlorophyll a/b-binding protein	668		1.75b
53	XP_001775294	Photosystem II stability/assembly factor HCF136	690		1.90
54	EDQ57614c	Phosphoglycerate kinase	431	1.57	2.25
			430	1.50	2.08
55	XP_001776961	Transketolase	614	2.78	2.08
56	XP_001764998c	Transketolase	901			–1.69
57	XP_001783115c	Glyceraldehyde 3-phosphate dehydrogenase	661		1.88b
58	XP_001773841	Phosphoglucomutase	903			–1.65
59	XP_001756785	Putative ATP synthase	672	1.84b
Other and unknown proteins
60	EDQ82483	Nucleotide diphosphate kinase 2 protein	315			–1.68
61	XP_001778544	Glutamine synthetase	716				–1.58
62	XP_001755833	Cytosolic glutamine synthetase	713				–1.64b
63	XP_001764953	Arginase	909				- 2.29
64	XP_001760386	Dessication-related protein	655	2.31b	1.79b
65	EDQ50069	Predicted protein	223	14.95	8.46
			218	20.77	7.99
66	XP_001782231	Predicted protein	450	12.27	4.92
67	XP_001772850	Predicted protein	689	10.42	3.72
			266	5.44	3.95
68	XP_001784742	Predicted protein	442	10.29	2.74
			443	6.25	1.89
69	XP_001756904	Predicted protein	412	9.95	3.98
70	EDQ78135	Predicted protein	416	6.21	4.47
71	XP_001772188	Predicted protein	667	9.13	1.59
72	XP_001774835	Predicted protein	424	7.36	6.25
73	XP_001765013	Predicted protein	682	6.94	6.45
74	EDQ62252	Predicted protein	272	5.05	2.90
75	XP_001754730	Predicted protein	208	4.29	5.50
76	XP_001778432	Predicted protein	747	3.38b	2.44b
77	EDQ56935c	Predicted protein	405	3.14	2.72
78	XP_001762972c	Predicted protein	808	2.27
79	XP_001779748	Predicted protein	435	2.24	2.13
80	EDQ50069	Predicted protein	428	1.68	2.43
81	XP_001775296c	Predicted protein	608	1.70	1.62
82	XP_001770325	Predicted protein	638	1.96
83	XP_001756363	Predicted protein, ABA inducible	448	1.90
84	XP_001762972c	Predicted protein	269	1.75
85	EDQ73558	Predicted protein	817		2.59
86	XP_001766107	Predicted protein	695		1.51b
			625		2.36b
87	XP_001757198	Predicted protein	733			–1.58

Mean value from analysis of DIGE replicates unless otherwise stated.

Mean value from analysis of triplicate CBB stained gels.

Dehydration-responsive genes (n=19) identified previously by microarray analysis in protonemal tissues of Physcomitrella (Cuming ).

Dehydration-responsive gene (n=2) identified previously in gametophore colonies of Physcomitrella (Hoang Saavedra ).

Table 2.

Dehydration- and rehydration-responsive proteins of unknown function from leafy-shoots of Physcomitrella as identified by MALDI TOF MS/MS. Average values for the fold change (up-regulation or down-regulation) in the abundance of the proteins are shown.

No	Spot number	Experimental Mr/pI	Up-regulation foldsa		Down-regulation folda
			Drought	Rehydration	Drought	Rehydration
1	220	71.62/5.13	14.47	7.78
2	826	14.92/5.39	7.22	5.33
3	809	26.09/5.42	5.63	2.37
4	400	20.44/4.85	3.62	2.98
5	649	43.53/5.96	3.20
6	810	23.31/5.58	2.82	2.3
7	821	21.85/5.38	2.76	3.01
8	296	21.50/4.87	2.60	2.04
9	613	72.57/5.99	2.51	2.05
10	803	62.63/5.67	2.32
11	683	20.86/5.06	2.24	1.97
12	662	33.34/6.46	2.21b	3.06b
13	688	15.04/5.27	2.13	1.67
14	811	17.42/5.26	2.11	3.13
15	804	82.66/4.88	1.79
16	616	79.60/6.33		3.44b
17	820	23.72/5.59		3.63
18	818	64.52/6.14		2.87
19	816	64.72/5.94		2.24
20	906	17.06/5.58			–4.69
21	902	62.95/5.46			–1.71
22	708	50.00/5.34			–1.98b
23	750	40.05/6.32			–1.50b	–1.88b
24	739	15.31/4.74				–1.88
25	714	40.94/5.62				–1.81b
26	701	84.83/5.06				–1.63

Mean value from replicated DIGE analysis.

Mean value from analysis of triplicate CBB stained gels.

Identification of differentially-expressed moss proteins by MALDI-TOF/TOF MS

The differentially-expressed protein spots (n=142) were manually excised from CBB-stained preparative gels and subjected to MALDI TOF/TOF MS analysis. By combining information of peptide mass fingerprinting (PMF) and peptide sequence tags, 116 protein spots were identified successfully, corresponding to 87 genes. All proteins identified in this work are well matched with known sequences of Physcomitrella species (see Supplementary Table S2 at JXB online). This achievement could be largely due to the updated information of genome sequences of Physcomitrella (Rensing ) and the efficient algorithm based on MALDI TOF/TOF MS. However, most proteins were assigned to the group of predicted proteins in current release (v1.6). By sequence comparisons with recent entries in the NCBI-nr database, 63 of the differentially-expressed proteins could be annotated since they share significant sequence homology with known proteins. The remaining 24 proteins were assigned to the group of predicted proteins. These are proteins of unknown function, responsive to dehydration or rehydration stress in plants. These results are summarized in Tables 1 and 2. The MS/MS data set for all protein spot are provided in Supplementary data SM1 and Supplementary Table S2 at JXB online.

Functional classification of the dehydration- and rehydration-responsive proteins

Following the functional definitions of Bevan and the established features of metabolism (Buchanan ), the differentially- expressed proteins identified in this work could be classified into seven functional groups (Fig. 3). Except for 24 novel proteins with unknown functions, the majority of proteins were assigned to four categories, i.e. seed maturation, disease/defence, protein folding and stability, as well as energy. Interestingly, the proteins with the first three functional categories were mostly up-regulated. More strikingly, proteins in the functional category of seed maturation represented the largest group of up-regulated proteins, accounting for 29% and 18% of the identified dehydration- and rehydration-responsive proteins, respectively. This strongly suggests that leafy-shoots of Physcomitrella probably adopt a dehydration tolerance mechanism similar to that established for angiosperm seeds.

Fig. 3.

Distribution of functional categories of the differentially-expressed proteins in Physcomitrella leafy-shoots in response to dehydration (A) and rehydration (B). (This figure is available in colour at JXB online.)

Among the differentially-expressed proteins, a majority were up-regulated as shown in Table 1 and Fig. 3. During dehydration, the abundance of 78 protein spots increased and that of only 15 decreased; while during rehydration, the abundance of 69 protein spots increased and that of 15 decreased (Fig. 3). This implies that there is active protein synthesis in leafy-shoots of Physcomitrella during dehydration and rehydration. This contrasts with the previous view from the desiccation-tolerant moss, T. ruralis, wherein the synthesis of stress-responsive proteins occurred mainly during post-desiccation rehydration (Wood and Oliver, 2004), in agreement with observations in vascular plants (Moore ).

LEA proteins are the largest group of proteins up-regulated under dehydration stress

During dehydration and rehydration, a large number of LEA proteins (27 proteins spots) were strongly induced, presumably to cope with the huge water variation in leafy-shoots (Table 1). Quantitative DIGE analysis showed that, during dehydration, 27 protein spots presented a strong increase in abundance, with an average of 4.2-fold increase (range, 1.6–13.7-fold); while during the phase of rehydration 15 protein spots showed 1.5–6.2-fold increase in abundance (Fig. 4). Altogether 27 LEA protein spots were identified as products of 11 LEA genes in Physcomitrella. Of these, nine LEA proteins (21 spots) were annotated as group 3 LEA characterized by multiple repeated 11-mer sequences (Dure, 1993, 2001); the remaining two (6 spots) were identified as group 2 LEA (dehydrins) characterized by sequences corresponding to K-domain (EKKGIMDKIKEKLPG) and/or Y-domain (V/TDEYGNP) (Close et al., 1989, 1993; Close, 1996; Campbell and Close, 1997). However, there were no group 1 LEA proteins revealed in this study. Further details on the LEA proteins are provided in Supplementary data SM1 at JXB online.

Fig. 4.

Dynamic pattern of LEA protein spots in response to dehydration and rehydration in the leafy-shoots of Physcomitrella. Fold change in intensity of the different LEA protein spots (numbers on the x-axis) was calculated from their relative volumes with respect to the internal standard.

Interestingly, it was found that three LEA proteins in Physcomitrella (XP_001769448, EDQ69797, and EDQ50092) display multiple isoforms that appeared to respond dynamically to the process of dehydration and rehydration. These isoforms of LEA proteins occupied an acidic region with high molecular weight on gels (Fig. 2). One of these appears as five spots of 77 kDa and four spots of lower mass in a pI range of 5.35 and 5.48 (Fig. 2, main spots 207 and 217). Although protein isoforms have been frequently observed on 2-D gels, relatively little is known about isoforms of LEA protein (Tolleter ). The correlation between their multiple isoforms and the dehydration/rehydration conditions could provide an important clue for in-depth study of biological function of these LEA proteins.

About 90% of LEA proteins identified here have not been reported in previous proteome studies of Physcomitrella (Wang XQ et al., 2009; Sarnighausen ; Cho ). This could be due to several reasons including the plant material used, the efficiency of protein extraction, and the MS technique. However, the present proteomics data on leafy-shoots is in excellent agreement with the microarray results of Physcomitrella protonemal tissue (Cuming ). Sequence comparison confirmed that seven LEA genes that were up-regulated in dehydrated protonemal tissues encoded 15 LEA protein spots which were seen on the 2-D DIGE gels of dehydrated/rehydrated leafy-shoots (Table 1; Fig 2).

Basal level expression of LEA proteins under control conditions in Physcomitrella leafy-shoots

Most of LEA proteins had a detectable basal level of expression in untreated control leafy-shoots. There were 13 protein spots corresponding to the products of seven LEA genes which were expressed in untreated control plants as seen on 2-D gels (see Supplementary Fig. S1 at JXB online; Fig. 5). The expression profiles of these were studied using a semi-quantitative RT-PCR analysis (Fig. 6). Nine of the 11 LEA genes were significantly transcriptionally induced in response to dehydration and rehydration. Only two genes encoding protein spots 403 and 684 showed a smaller change at the transcript level. The majority of LEA genes had a detectable basal level of expression in control leafy-shoots, and their synthesis was induced when dehydration stress was applied (Fig. 6). A detectable basal level of expression of LEA proteins has not been reported in previous investigations on Physcomitrella (Frank ; Oldenhof ; Cuming ; Wang XQ et al., 2009). These unexpected findings partially support the controversial hypothesis based on a series of studies on desiccation-tolerance in moss T. ruralis, that the mechanism of desiccation tolerance in bryophytes probably involves a constitutive level of cellular protection (Wood and Oliver, 2004; Oliver ).

Fig. 5.

Basal level expression of LEA proteins in untreated control leafy-shoots of Physcomitrella as detected by resolution on 2-DE gels. Each spot of interest is represented by its relative volume, i.e. the intensity of each spot with respect to the total protein intensity for the same gel. The data presented are mean values ±SD from the analysis of triplicates of CBB-stained gels.

Fig. 6.

Transcription of Physcomitrella LEA genes. Semi-quantitative RT-PCR of 11 genes encoding LEA proteins in Physcomitrella was performed on total RNA isolated at indicated times (h) during the phases of dehydration and rehydration, starting from ‘Con’ (0 h). Numbers on the right are the spot numbers corresponding to the transcripts identified (left side) by sequence comparison. Expression of PpActin3 was used as an internal control.

The coherence between the responses at the protein and transcript levels indicates that LEA proteins are the largest group of proteins that accumulated significantly during dehydration in Physcomitrella. So far, hundreds of LEA proteins have been isolated from vascular and non-vascular plants (see reviews in Dure ; Close, 1997; Cuming, 1999; Shih ), as well as from microbes and animals (Volker ; Browne ; Gal ; Tunnacliffe ; Wang ). In most cases, LEA transcripts only appear under stress conditions or in embryos of plant seeds, and are therefore suggested to be associated with desiccation tolerance (see reviews in Bray, 1994; Ingram and Bartels, 1996; Cuming, 1999; Hundertmark and Hincha, 2008; Shih ). Numerous studies support the notion that the dehydration-induced LEA proteins function as cellular protectants to stabilize cellular components by preventing protein aggregation/denaturation and plasma membrane fusion caused by water loss (Chakrabortee ; Shih ). However, several other studies have revealed that LEA proteins may be constitutively expressed in plants such as pea (Robertson and Chandler, 1994), birch (Rinne ), Arabidopsis (Nylander ), citrus (Sanchez-Ballesta ), the moss T. ruralis (Wood and Oliver, 2004), similar to Physcomitrella shown in the present study. It is possible that a certain basal level of expression of LEA proteins is required for growth and development of leafy-shoots under unstressed conditions, or may represent a constitutively expressed defence system that is required at the onset of dehydration. The precise function of LEA proteins remains to be determined.

Leafy-shoots display an impressive set of defence mechanisms

Apart from LEA proteins, our proteomics data revealed 13 other proteins that are likely to be involved in the defence mechanisms of Physcomitrella. Of these, four proteins (seven spots) known to be associated with membrane components and membrane stabilization were highly expressed in dehydrated/rehydrated leafy-shoots. These are actinoporins (spots 439, 408, 685), phosphatidylethanolamine binding protein (spot 284), phospholipase D (spots 209, 210), and arabinogalactan protein (spot 449). Actinoporin is a homologue of pore-forming cytotoxin protein studied extensively in sea anemone (Hoang ). In leafy-shoots, three protein spots corresponding to actinoporins were found to increase in abundance during dehydration, with an average increase of 2-fold and to an average increase of about 6-fold during rehydration (Fig. 2; Table 1). RNA blot analysis revealed that actinoporins of Physcomitrella were induced by dehydration stress (Hoang ). These have been proposed to be adapted to bind phospholipid molecules via an atypical Trp cluster (Hoang ). On the basis of the observed highest abundance of actinoporins in rehydrated leafy-shoots, it is suggested that these proteins with a known hemolytic activity (Hoang ) play a key role in the recovery of the normal function of membranes.

Phosphatidylethanolamine-binding protein (PEBP, spot 284) was another phospholipid-associated protein identified in this study, which showed a remarkably high level of induced expression in dehydrated and rehydrated leafy-shoots (9-fold or 3-fold, respectively; Fig. 2; Table 1). Recently, four genes encoding PEBP have been annotated in the Physcomitrella genome (Hedman ). PEBP (spot 284) shows a low sequence similarity to all of them and, therefore, may be a new member of the PEBP family. PEBP family members are involved in many important biological processes, including membrane fluidity and membrane biogenesis (Moore ; Frayne ), anti-apoptotic responses (Zhang ), and the regulation of ABA signalling during seed germination (Xi ). It is possible that the PEPB proteins that appear to be highly expressed in Physcomitrella play a significant role in dehydration tolerance.

Arabinogalactan protein (AGP, spot 449) and phospholipase D (PLD, spots 209, 210) appeared to accumulate steadily throughout the dehydration–rehydration cycle (Table 1). It is well known that AGP is a typical glycosylphosphatidylinositol (GPI) anchored protein tightly bound to the outer side of plasma membranes in flowering plants (Schultz ; Eisenhaber ). Under osmotic stress due to high salt, a massive up-regulation of AGP is thought to act as a buffer zone to prevent direct interaction of the naked membrane with the wall matrix (Lamport ; Seitz ). The observed higher expression of AGP (>2-fold) in Physcomitrella leafy-shoots subjected to dehydration and rehydration suggests that this mechanism of membrane stabilization seen in flowering plants has been retained in the less evolved species. Interestingly, two isoforms of a key enzyme, phospholipase D (PLD, spots 209, 210) that controls AGP release under stress conditions (Munnik et al., 2000; Ruelland ) were also identified. To summarize, the above proteins associated with membrane components and membrane stabilization may be an indication that Physcomitrella has an ancient and efficient defence system to protect membranes again dehydration stress.

Among proteins in the defence category, the highest induction was found to be dehydroascorbate reductase (spot 277), which exhibits >12-fold higher expression during both dehydration and rehydration (Table 1). Dehydroascorbate reductase linked to ascorbate metabolism is one of the key enzymes responsible for the removal of active oxygen species (AOS). Other enzymes involved in AOS metabolism, for example, thioredoxin peroxidase (spot 279) and 2-Cys peroxiredoxin (spot 735), were also identified in the study (Table 1). In addition, it was found that expression of three proteins involved in quinone redox cycling was altered in the course of dehydration/rehydration, including two benzoquinone reductases (BR, spots 679, 680) and quinone oxidoreductase (QR)-like protein (spot 720). Several reports have supported the importance of these enzymes in protection against oxidative stress (Ernster ; Brock and Gold, 1996; Matvienko ). Therefore, in Physcomitrella, ascorbate metabolism and redox cycling of quinones could be dominant modules in its anti-oxidation defence system.

Importance of protein biosynthesis and quality control systems

Expression of 13 proteins that function in protein biosynthesis and quality control was enhanced in dehydrated/rehydrated leafy-shoots. These comprise eukaryotic initiation 4A (spot 648), eukaryotic initiation factor 5A-2 (spot 686), various HSPs (>60 kDa, seven spots), proteases (three spots), gamma interferon inducible lysosomal thiol reductase (GILT, spot 812), protein disulphide isomerase-like (PDI, spot 639), and calnexin 1 (spot 647) (Table 1). The first two proteins function in translation machinery and were up-regulated 2–3-fold in dehydrated leafy-shoots, suggesting that the protein biosynthesis in Physcomitrella was active during dehydration. This observation is in contrast to that in T. ruralis, wherein the synthesis of proteins during dehydration is highly unlikely due to the loss of polyribosomes (Oliver, 1991; Oliver and Bewley, 1997). It is suspected that the phase discrepancy in protein synthesis among bryophytes is related to their capability of dehydration tolerance.

An enhancement in the synthesis of proteins involved in processes for quality control (protein folding and stability, proteolysis) is a fundamental molecular response to various adverse environmental conditions (Hartl, 1996; Wickner ; Cui , 2009). These processes are aimed at maintaining proteins in their functional conformations and prevent protein aggregation. Several molecular chaperones (>60 kDa), such as Hsp 60, Hsp 70, Hsp 90, and Hsp100 (spots 806, 621, 618 607, 610), have been identified but no small molecular mass Hsps. In addition, several proteases such as chloroplast FtsH protease (spot 234), FtsH-like protein Pftf precursor (spot 631), and putative serine carboxypeptidase II (spot 402) were also dehydration-responsive (Table 1). Obviously, proteins belonging to the functional category, protein quality control, play an important role in the responses of Physcomitrella to dehydration stress. However, the fold up-regulation of these proteins was only an average of 1.99-fold, lower than that of proteins in categories such as seed maturation (average, 3.83-fold) and defence (average, 4.41-fold) (Table 1). By contrast, the functional category of protein quality control is relatively highly induced in vascular plants in the tolerance to diverse stresses (Cui , 2009). Thus, it is difficult to establish a cause-and-effect relationship between the lower expression of the protein quality control system with the higher capability of dehydration tolerance exhibited in Physcomitrella.

In this study, it was found that the abundance of gamma interferon inducible lysosomal thiol reductase (GILT, spot 812) and protein disulphide isomerase-like (PDI, spot 639) was increased more than 2-fold in dehydrated/rehydrated leafy-shoots (Table 1). GILT and PDI belong to the thiol reductase family, which is involved in the reduction, oxidation, and isomerization of protein disulphide bonds in cells (Lundstrom-Ljung and Holmgren, 1998; Bogunovic ). It has long been known that protein conformational fluctuations by thiol/disulphide transfer are involved in multiple biological processes including stress tolerance (Moriarty-Craige and Jones, 2004), as seen here for Physcomitrella exposed to dehydration/rehydration stress.

Changes in photosynthesis and energy metabolism

During dehydration and rehydration, the largest group of down-regulated proteins consisted of those involved in photosynthesis. The abundance of carbonic anhydrase (CA, four spots), an enzyme catalysing CO2 into bicarbonate (Khalifah, 1971; Moroney ), decreased throughout dehydration and rehydration treatments (Table 1). Similarly, expression of Rubisco activase (three spots) and Rubisco large subunit (two spots) also decreased. Since these enzymes function in the initial reactions of CO2 assimilation, the observed decreased levels of these enzymes strongly suggested that the uptake of CO2 might be affected under dehydration/rehydration conditions. However, key enzymes/proteins in the Calvin cycle and light reaction of photosynthesis did not exhibit significant changes in abundance on 2-D gels during dehydration. In this functional category, abundance of only four proteins (spots 668, 690, 722, 745, Table 1) was altered slightly during rehydration. The fact that the abundance of the major enzymes/proteins in the photosynthetic machinery was largely unperturbed during the dehydration/rehydration stress could be an important cellular and molecular basis for coping with drought stress in Physcomitrella.

On the other hand, there was a change in the abundance of four proteins involved in glycolysis and the pentose phosphate pathway (Table 1), namely, phosphoglycerate kinase (spots 431, 430), transketolase (spots 614, 901), glyceraldehyde 3-phosphate dehydrogenase (spot 661), and phosphoglucomutase (spot 903). Also, there was an increase in the abundance of a putative ATP synthase (spot 672). These observations may be an indication of an alteration in energetic status within cells of Physcomitrella during dehydration and rehydration.

Other metabolism and proteins of unknown function

During dehydration, expression of nucleoside diphosphate kinase (NDPK, spot 315) decreased. NDPKs are ubiquitous enzymes found in all organisms and cell types, and have distinct biological function in catalysing the synthesis of nucleoside triphosphates (Choi ). Abundance of three enzymes involved in both glutamine and arginine metabolism declined during rehydration (Table 1): two glutamine synthetases (spots 716 and 713) and an arginase (spot 909). The reduced levels of these enzymes reflect the repression of a fundamental nucleoside metabolism in leafy-shoots during dehydration/rehydration. Apart from the differentially-expressed proteins that are involved in distinct metabolic pathways, 27 spots corresponding to 24 different proteins (encoded by 24 unique genes) with no functional annotation also showed significant changes in abundance during dehydration and rehydration (Table 1). They are predicted or hypothetical proteins in the Physcomitrella genome database, 19 of which were reported for the first time as dehydration/rehydration-responsive proteins. A predicted protein with two isoforms (represented by spots 223 and 218) showed the highest induction of 14.95- to 20.77-fold during dehydration; another two proteins represented by spots 689 and 442, were also elevated by 10.42-fold and 10.29-fold under the same conditions. In addition, nine predicted proteins were up-regulated more than 5-fold. These proteins of unknown functions that are highly expressed under dehydration/rehydration conditions may be of interest for an in-depth study of dehydration tolerance in leafy-shoots of Physcomitrella.

Concluding remarks

Leafy-shoots of Physcomitrella appear to possess a complex molecular mechanism to cope with huge variations in water availability. By a comprehensive proteomic analysis supported by the confirmation of transcription of selected protein spots, major dynamic changes in protein synthesis were revealed in the haploid vegetable tissues of Physcomitrella during dehydration and rehydration. In contrast to other moss plants, Physcomitrella leafy-shoots appear to be well prepared to utilize their preformed defence system under dehydration and rehydration conditions, as indicated for example, by the basal level of expression of a substantial number of LEA proteins even in the presence of an adequate water supply. This unexpected finding, together with 87 dehydration-/rehydration-responsive proteins identified in the study, reveal some of the essential molecular details of dehydration responses in Physcomitrella. To our knowledge, most proteins identified here have not previously been described in Physcomitrella.

The experimental results presented here support the view that dehydration tolerance in moss plants has certain points of similarity with desiccation tolerance in angiosperm seed maturation. The proteomic analysis leads to the conclusion that leafy-shoots of Physcomitrella adopt a dual protective strategy against dehydration, which consists of maintaining a basal level of synthesis of stress-tolerance proteins and enhancing their synthesis further under stress. Such mechanisms may be poorly represented or absent in higher vascular plants that cannot survive in severe cellular dehydration. On the other hand, it was found that certain key molecular players that are known to contribute significantly to dehydration tolerance were induced in the leafy-shoots, indicating that these have been well conserved through evolution.

The data presented here provide a strong and reliable basis for future studies to elucidate the functional significance of specific proteins associated with plant adaptation to dehydration tolerance, a feature that is essential for land colonization.

Supplementary data

Supplementary data can be found at JXB online.

A list of all primers used in the study.

The MS/MS data set for all protein spot with peptide sequence tags.

Dynamic proteome profiles generated from traditional 2-DE.

Supplementary data SM1. Material on the identification of LEA proteins.