Differential gene expression in Pyropia columbina (Bangiales, Rhodophyta) under natural hydration and desiccation conditions

In rocky shores, desiccation is triggered by daily tide changes, and experimental evidence suggests that local distribution of algal species across the intertidal rocky zone is related to their capacity to tolerate desiccation. In this context, the permanence of Pyropia columbina in the high intertidal rocky zone is explained by its exceptional physiological tolerance to desiccation. This study explored the metabolic pathways involved in tolerance to desiccation in the Chilean P. columbina, by characterizing its transcriptome under contrasting conditions of hydration. We obtained 1,410 ESTs from two subtracted cDNA libraries in naturally hydrated and desiccated fronds. Results indicate that transcriptome from both libraries contain transcripts from diverse metabolic pathways related to tolerance. Among the transcripts differentially expressed, 15% appears involved in protein synthesis, processing and degradation, 14.4% are related to photosynthesis and chloroplast, 13.1% to respiration and mitochondrial function (NADH dehydrogenase and cytochrome c oxidase proteins), 10.6% to cell wall metabolism, and 7.5% are involved in antioxidant activity, chaperone and defense factors (catalase, thioredoxin, heat shock proteins, cytochrome P450). Both libraries highlight the presence of genes/proteins never described before in algae. This information provides the first molecular work regarding desiccation tolerance in P. columbina, and helps, to some extent, explaining the classical patterns of ecological distribution described for algae across the intertidal zone.

Water, and its intracellular balance, is a critical factor for all living organisms in both terrestrial and marine ecosystems.Mobile animals actively avoid desiccation, induced by water deficiency, while other organisms, such as resurrection plants -a small group of angiosperms that live in the most arid habitats of the world-are adapted to tolerate water losses of up to 90% (Gaff, 1987).This adaptation, in general terms, is based on the ability of an organism to equilibrate its internal water potential with the dry environment, and re-start normal functions when rehydrated (Alpert, 2000).Several studies using resurrection plants as model have been conducted to fully understand their impressive adaptation to desiccation (Scott, 2000).Recent advances in our understanding of the mechanisms of tolerance in these organisms have revealed changes at the morphological level, osmolites and protein synthesis, and a decline in ROS (reactive oxygen species) production and photosynthesis rate (Ingram & Bartels, 1996;Hoekstra et al., 2001;Bernacchia & Furini, 2004;Vicré et al., 2004;Dinakar et al., 2012).More specifically, it has been observed that the plant hormone ABA is accumulated in desiccated resurrection plants, which induces the expression of several proteins related to desiccation (Bartels et al., 1990;Dinakar et al., 2012).Also, leaves of resurrection plants tend to curl to reduce water loss and minimize oxidative damage due to desiccation (Vicré et al., 2004;Farrant et al., 2007;Toldi et al., 2009).Oxidative damage is attenuated or avoided by increasing antioxidant activity of some enzymes (e.g., ascorbate peroxidase, glutathione reductase, superoxide dismutase, among others) and levels of antioxidant compounds (e.g., anthocyanins) (Farrant et al., 2007;Toldi et al., 2009;Dinakar et al., 2012).Additionally, these plants reduce photosynthetic activity to minimize photo-oxidative damage that could lead to increased ROS levels.Thus, it seems clear that diverse metabolic pathways are involved in attenuating the oxidative stress condition caused by desiccation.
In the Chilean coastal ecosystems, P. columbina grows abundantly along the upper intertidal zone (Alveal, 1970;Santelices, 1989;Hoffmann & Santelices, 1997), where it is exposed to a wide range of environmentally stressful conditions, mainly desiccation-driven stress induced by low tides and air exposure (Contreras-Porcia et al., 2011a).P. columbina is well adapted to daily extremes, which range from exposure to water (full hydration) during high tides to long exposure to air (maximum desiccation) during low tides.Natural exposure to these extreme regimes have been described in other organisms exposed to desiccation.It is already known that Porphyra and Pyropia species have high tolerance to desiccation, and quickly recover photosynthetic activity once rehydrated after a period of desiccation (e.g., Smith & Berry, 1986;Kim et al., 2008;Contreras-Porcia et al., 2011a;Gao & Wang, 2012).Basic physiology of these organisms, including the mechanisms to tolerate environmentally stressful conditions, remains poorly studied.However, it is known that, in P. columbina, desiccation induces losses ca.96% of the water content and enhanced significantly the production of ROS (Contreras-Porcia et al., 2011a).The quick return of ROS to their basal levels during high tide is explained by an efficient activation of the antioxidant system.In comparison, species inhabiting the lower intertidal zone [e.g., Mazzaella laminarioides (Bory de Saint-Vincent) Fredericq, Ulva compressa Linnaeus and Lessonia spicata (Suhr) Santelices] are more sensitive to desiccation, and this seems related to the absence of efficient mechanisms to attenuate the over-production of ROS during rehydration (Contreras-Porcia et al., 2012;López-Cristoffanini et al., 2013).Flores-Molina et al. (unpublished data) recently provided some physiological and biochemical bases that help explain the role of desiccation on species distribution across the intertidal zone.They reported that sensitive species displayed i) inactivation of antioxidant enzymes, ii) over-oxidation of biomolecules, and iii) inactivation of the photosystem II.For example, Ulva compressa (Chlorophyta, Plantae) and Scytosiphon lomentaria (Lyngbye) Link (Ochrophyta, Chromista) inhabiting the mid-intertidal zone have lower tolerance to desiccation than P. columbina.However, they tolerate desiccation better than Lessonia spicata (Ochrophyta, Chromista) and Gelidium rex Santelices & I.A. Abbott (Rhodophyta, Plantae), both lower intertidal species.
Given the ecological and economic relevance of Pyropia and that some basic biochemical and physiological information on the mechanisms involved in tolerance to desiccation is known, we focused this work in determining the genes/proteins that are differentially expressed in P. columbina during the hydration-desiccation cycle, using Suppression Subtractive Hybridization (SSH) and ESTs determined by pyrosequencing (454 Life Sciences, Roche).The results of this study will help to elucidate the genetic basis underlying the high tolerance to desiccation displayed by this species, and will broaden our knowledge of the molecular biology/ecology of this organism and other macroalgal species.

Ethics statement
No specific permits were required for the described field studies.The study area is unrestricted to public access and use, and is not privately owned or designated as a protected area (reserves or parks).No protected or endangered species were involved in this study.Fronds of P. columbina were collected along 200-300 m of coastline during high (naturally hydrated plants, 100% relative water content (RWC)) and low tide (naturally desiccated plants, ca.4% RWC) in Maitencillo beach, Valparaiso (32º39.5´S,71º26.6´W).After collection, fronds were quickly rinsed (15-20 s) in 0.22 µm-filtered seawater, manually cleaned and frozen on site with liquid nitrogen.

RNA extraction
Total RNA was isolated from 20-30 g of fresh tissue of naturally hydrated and desiccated, pooled fronds of P. columbina.Tissue, frozen in liquid nitrogen, was homogenized in 25 mL of lysis buffer containing 4 M guanidinium thiocyanate, 25 mM EDTA, 200 mM sodium acetate, 2% polyvinylpyrrolidone (PVP-40) and 1% 2-mercaptoethanol.The homogenate was incubated for 10 min at 70°C with constant agitation in the presence of 20% sarcosin, and then centrifuged for 5 min at 16000 g.The RNA present in the supernatant solution was purified and re-extracted using an RNeasy mini Kit (Qiagen, Hilden, Germany), according to the manufacturer protocol.RNA quality and yield was assessed by spectrophotometry (NanoDrop TM 1000 Spectrophotometer, Thermo Scientific, DE, USA) and denaturing agarose gel electrophoresis.Finally, the mRNA was obtained from the total RNA extracted (ca.290-300 µg) using DynaBeads (Invitrogen, Oregon, USA).Prior to RNA extraction, all material was treated in 0.1% DEPC water.

Preparation of the cDNA libraries by SSH and next-generation sequencing
The synthesis of cDNAs, from both natural conditions (i.e., natural hydration and desiccation stress), were obtained using a SMART TM cDNA Library Construction Kit (Clontech, Mountain View, CA, USA), as described in Wellenreuther et al. (2004).Then, two subtracted cDNA libraries were constructed: one with genes expressed exclusively under hydration (UH) and the other by those expressed under desiccation stress (UD), as described in Diatchenko et al. (1996).cDNAs were then sequenced by 454-pyrosequencing (Margulies et al., 2005).cDNAs were ligated to 454 self-made adaptors with Multiplex Identifier Adaptors (MIDs) for the GS FLX Titanium Chemistry following Roche's technical bulletin TCB 09004 introducing SfiI-sites.The 454 libraries were immobilized on beads and clonally amplified using a "GS FLX Titanium LV emPCR Kit".The libraries were then sequenced using a "GS FLX Titanium Sequencing Kit XLR70" and a "GS FLX Titanium PicoTiterPlate Kit".All kits were purchased from Roche and used according to the manufacturer protocols.

Data assembly and bioinformatics analysis
Readings from both libraries were processed by selfwritten Perl scripts and assembled into coatings, i.e., representing putative transcripts using MIRA 3 assembly (Contreras-Porcia et al., 2011b).The expressed sequences tags (ESTs) determined from both libraries were subjected to separated bioinformatics analyses.ESTs were analyzed for sequence similarities using the BLASTX program (NCBI, MD, USA).Reading frames with the highest sequence similarity scores were used to analyze protein identity using the BLASTP program.Threshold values were set above 50 for high-scoring segment pairs, with a minimum significance at least of e ≤ 10 -4 and an identity higher than 30%.ESTs coding for known proteins were classified into functional categories by the KO (KEGG Orthology) database for ortholog grouping and hierarchical classification of genes, according their functionality (Kanehisa et al., 2004) by using the BLAST2GO software (Götz et al., 2008).Finally, the putative subcellular localization of the proteins was determined using the TARGETP (http://www.cbs.dtu.dk/services/TargetPEmanuelsson et al., 2000), WOLFPSORT (http://psort.ims.utokyo.ac.jp),ChloroP (http://www.cbs.dtu.dk/services/ChloroP Emanuelsson et al., 1999) and PSORTb (http://www.psort.org/psortbYu et al., 2010) servers.

EST sequencing and assembly
A total of 8,054 sequence reads were obtained from the library enriched with hydration-responsive exclusive transcripts (UH library, Table 1).Moreover, 8,432 sequence reads were obtained from the library enriched with desiccation-responsive exclusive transcripts (UD library).In the UH library, 49.2% of ESTs ranged from 200 to 500 bp, with an average size of 423 bp.In the UD library, 46% of the ESTs ranged from 200 to 500 bp, with an average size of 385 bp.The rest of the transcripts in both libraries ranged between 100 to 190 bp.The sequences of both libraries are available on the EMBL Nucleotide Sequence Database (http://www.ebi.ac.uk/ embl) with accession numbers HE858615 to HE859412 for UH and HE859413 to HE859937 for UD libraries.

DISCUSSION
The hydration-desiccation cycle in P. columbina generates an unbalance in the intracellular redox potential (Contreras-Porcia et al., 2011a), a situation that must be controlled by a coordinated cascade of responses that are induced differentially in each condition.Our results show that, during hydration, predominantly expressed genes were those involved in protein metabolism.Among them, we highlight several ribosomal proteins, translation initiation factors and elongation factors, proteases, the proteasome system and several ubiquitins (Table 1).Of particular relevance is the ubiquitin-proteasome system, present in all eukaryotes, which is activated in response to several abiotic stress factors and participates in tolerance mechanisms by removing unfolded proteins and proteins damaged during oxidative stress (Dreher & Callis, 2007;Pena et al., 2007;Kurepa et al., 2008).In this functional group three types of peptidylprolyl cistrans-isomerases (PPIases) were identified and found to be similar to those from Arthropoda and Mollusca.These proteins catalyze the cis-trans isomerisation of prolines (C 5 H 9 NO 2 ), and have been described in the processes of cellular signalling (with a calmodulin-binding domain), regulation of gene transcription, and acting as chaperones and folding catalysts.Specifically, in the gastropod Conus novaehollandiae these proteins facilitate the oxidative folding of several neurotoxic peptides (Safavi-Hemami et al., 2010).In plants, these proteins are involved in flowering (Wang et al., 2010) and in controlling cell proliferation, since PPIase expression increases in the presence of cytokinin (Vittorioso et al., 1998).PPIases are induced by wounding, heat and salt stress (Vucich & Gasser, 1996;Kurek et al., 1999), and have been directly involved in membrane protein folding (i.e., chloroplast and mitochondria) (Breiman et al., 1992).Thus, the quick re-establishment of the normal condition after desiccation in P. columbina could be in part explained by both: i) the re-folding of structural proteins with important functions such as those involved in transcriptional regulation, and ii) by the removal of oxidized proteins.In fact, levels of oxidised proteins measured during the hydration-desiccation cycle were consistently lower in P. columbina compared to those in sensitive species (Contreras-Porcia et al., 2011a, Flores-Molina et al., unpublished data).
Proteins that form part of the energy metabolism were more highly represented during hydration [i.e., Cytochrome C Oxidase subunits, NADH Dehydrogenase subunits, NADH-Ubiquinone Oxidoreductase subunits, F1F0-ATP Synthase subunits and ADP/ATP carrier proteins (Table 1)] than during desiccation stress (Table 2).Theoretically, maintaining the tolerance mechanisms required to buffer the effects of a stress imposes energy costs (e.g., Zagdańska, 1995).Acclimation to oxidative stress depends on a high availability of NADPH and ATP, since most of the intracellular metabolic reactions require energy.For example, the ubiquitin-mediated system for intracellular protein degradation is ATP-dependent in all organisms (Ciechanover et al., 1984).Additionally, the structure of the chromosomes plays a critical role in transcriptional regulation where the chromatin remodeling is also ATP-dependent (Luo & Dean, 1999).In the case of P. columbina it seems reasonable to hypothesize the occurrence of a higher energy production by the mitochondrial system during hydration, as many metabolic reactions are necessary to maintain a homeostatic redox state and, as a result, a healthy physiological condition during the hydration-desiccation cycle.Moreover, the decline in ATP production resulting from a general metabolic slow down during desiccation could be beneficial in preventing ROS production triggered by the electronic alterations induced by low water potential.In vascular plants under water stress, ROS production has been detected in apoplast, xylem vessels, chloroplasts and mitochondria (Mittler et al., 2004;Toldi et al., 2009).Therefore, it is possible that desiccation in P.
columbina induces ROS production in the organelles, a hypothesis that needs to be experimentally demonstrated.Additionally, the mitochondrial metabolism should be measured in order to demonstrate a decay-activation sequence of this system during the desiccation-hydration cycle.
In the context of ROS attenuation by the antioxidant system during the hydration-desiccation cycle, several antioxidant enzymes should be expressed.In fact, a peroxiredoxin (PRX) typical 2-Cys, an arachidonate 5-lipoxygenase, a glutathione S-transferase, and several cytochrome P450 and HSPs (heat shock proteins, 27, 40, 70, 80, 90 types) were detected during hydration.Moreover, during desiccation, enzymes such as thioredoxin (TRX), catalase and low variants of HSPs (70 and 90, gene sequences different from hydrated fronds) were expressed.PRXs are involved in detoxification of hydrogen peroxide, alkylhydroperoxides and peroxinitrites (Hall et al., 2009).These enzymes react at low peroxide concentrations and may become inactive at higher concentrations.In plants, PRX transcripts increase in response to different abiotic stresses such as salinity, drought, and heavy metals (Wood et al., 2003;Dietz, 2011).Their expression in algae has been poorly studied, although some studies indicate they are regulated by light, oxygen, copper, desiccation and redox state (Goyer et al., 2002;Contreras-Porcia et al., 2011a, 2001b;Lovazzano et al., 2013; see section 3 in Contreras-Porcia & López-Cristoffanini, 2012).In vascular plants, several PRXs have been described based on their catalytic mechanisms and subcellular localization (Baier & Dietz, 1997).In this study, PRX expresses mainly when fronds are hydrated, is localized exclusively in the chloroplast, and its reduction is TRX-dependent.The chloroplast TRX activity was also recorded during desiccation (Table 2).Therefore, it seems likely that P. columbina PRX play an important role in buffering oxidative stress and in post-desiccation detoxification of lipoperoxides in the chloroplast.However, other attenuation systems might also be operating in both environmental conditions in order to normalize the redox state imbalance.For example, a chlorophycean ferredoxin was identified during desiccation (Table 2).This is a small protein that plays a key role in electron distribution in the chloroplast (Schürmann & Buchanan, 2008) by regulating the chloroplast metabolic network through the TRX system, and contributes directly to ascorbate antioxidant protection (i.e., antioxidant compound and ascorbate peroxidase substrate) and PRX regeneration (Ceccoli et al., 2011;Dietz, 2011).
Another enzyme detected in hydrated fronds was a lipoxygenase (LOX, Table 1), a dioxygenase which peroxidates polyunsaturated fatty acids (Gigon et al., 2004).Lipid molecules produced by lipid degrading enzymes, such as oxylipins, can act as secondary messengers of stress-response signal transduction pathways (Blée, 2002;Vellosillo et al., 2007).However, a hyper-stimulation of lipoxygenase activity could induce an accumulation of lipope-roxides, which leads to cell damage and organelle dysfunction.For example, Contreras et al. (2009) demonstrated that hyper-activity of LOX led to an over-production of lipoperoxides, and at the end, to cell death in sensitive species under copper-induced oxidative stress.The particular expression of this enzyme during hydration could explain the exceptional control of the lipid peroxidation in P. columbina (Contreras-Porcia et al., 2011a) in comparison with several other algae (Flores-Molina et al., unpublished data).Moreover, the involvement of PRX during the hydration-desiccation cycle may additionally explain the effective ROS and lipoperoxide attenuation in this species.
Several cytochrome P450 variants were found in P. columbina under natural conditions of hydration and desiccation.These enzyme variants are present in all living species and catalyze the oxygenation of a high variety of substrates (Anzenbacher & Anzenbacherová, 2003).It has been demonstrated that P450s are induced by abiotic and biotic stress (Narusaka et al., 2004;Stolf-Moreira et al., 2011).Some P450s have been identified as ABA 8´hydroxylase that degrades ABA (abscidic acid) during the hydration-desiccation cycle (Kushiro et al., 2004;Shinozaki & Yamaguchi-Shinozaki, 2007).In vascular plants, ABA is overproduced during desiccation (as in P. columbina, Guajardo et al. pers. comm.), causes stomatal closure, and induces stress related genes.However, via the ABA 8´hydroxylase-P450, ABA concentration is reduced to basal levels during the transition from desiccation to hydration (Kushiro et al., 2004).ABA has been recently identified in several algal species (i.e., Tarakhovskaya et al., 2007;Yokoya et al., 2010), although its role in regulating the expression of genes associated with tolerance to abiotic/biotic stress has not being explored.In P. columbina the functional role of P450 could open new avenues to learn on tolerance pathways involved in managing environmental stressors, such as the ABA involvement in transcription of regulatory networks of desiccation stress signals and gene expression.Moreover, ABA in P. columbina could be involved in the up-regulation of several compounds like sugar, prolines and polyamines (e.g., putrescine, spermidine and spermine), which are known to increase their expression under water stress in vascular plants and algae (Guill & Tuteja, 2010;Alcázar et al., 2011;Kumar et al., 2011).Thus, it is also possible to hypothesize regarding the participation of the ABAindependent pathways in the regulatory response to dehydration stress.Indeed, a Clp-protease was identified during desiccation (Table 2), and the clp gene not only was induced by dehydration, but was also up-regulated during natural senescence (Nakashima et al., 1997).Analysis of the clp gene in transgenic plants indicates that the clp promoter contains cis-acting element(s) involved not only in ABA independent stress-responsive gene expression but also in senescence-activated gene expression (Simpson et al., 2003).Thus, several tolerance pathways, previously unknown in algae and other organisms, could be synergistically activated under particular environmental stress conditions.
Proteases are indispensable for the normal functioning of cells and tissues in all living organisms.However, their activities need to be correctly regulated (Habib & Fazili, 2007).During the hydration-desiccation cycle in P. columbina, several peptidases were identified (Tables 1 and 2).Even though the expression of these proteins is required for protein metabolism, an over-expression can be potentially harmful.During hydration, a metallocarboxypeptidase inhibitor (CPI) was identified, that forms a less active or fully inactive enzyme.In vascular plants, CPIs are activated by metals, mechanical wounds or insect injury (Villanueva et al., 1998;Habib & Fazili, 2007;Harada et al., 2010), and its expression is ABA-regulated (Villanueva et al., 1998).Thus, this type of protease in P. columbina could be part of the ABA genes that are regulated during the hydration-desiccation cycle needed to maintain cellular integrity under water deficiency.
ABC (ATP binding cassette) transporter sequences were identified in desiccated fronds.These transporters are involved in translocation of a wide variety of compounds across cell membranes, including carbohydrates, lipids, xenobiotics, drugs, and heavy metals (Ehrmann et al., 1998;Sipos & Kuchler, 2006;Contreras et al., 2010).These transporters have been reported during metal tolerance in vascular plants, carrying metal complexes from cytosol to the vacuole (Clemens, 2001).Also, in humans they appear associated with the protection of placental tissue by preventing cellular accumulation of cytotoxic compounds (Aye & Keelan, 2013).The participation of these transporters in desiccated P. columbina was unexpected, despite information which recently demonstrated that tolerance to desiccation in the free-living soil bacterium Rhizobium legumenosarum was associated with ABC-transporter activity (Vanderlinde et al., 2010).In the bacterium, a mutation in the ATP-binding component of a previously uncharacterized ATP transporter (Young et al., 2006) decreased the tolerance to desiccation due to low exopolysaccharides levels in the cell wall envelope of the mutant.That study demonstrated the crucial role of polysaccharides and their transport in organisms tolerant to desiccation.Therefore, our results suggest, for the first time in algae, the involvement of ABC-transporters in desiccation tolerance, possibly through cell wall stabilization during desiccation stress.
Several HSP were identified during hydration and desiccation stress.These proteins are recognized in prokaryotes and eukaryotes during responses to different physiological and environmental stress conditions (Feder & Hofmann, 1999).HSPs induction leads to a state of resistance for subsequent stress in the cell by preventing protein aggregation (Feder & Hofmann, 1999) and suppressing apoptosis (e.g., HSP90 see Beere et al., 2000 andRavagnan et al., 2001).HSP70 blocks apoptosis by binding apoptosis activating factor-1 (Apaf-1), thereby preventing the formation of the apoptosome complex (Ravagnan et al., 2001).In fish hepatocytes, HSPs induction is involved in stress tolerance by modulating the action of key proteins and kinases in the signal transduction pathways (Padmini & Usha-Rani, 2011).In the seaweeds Fucus serratus and F. vesiculosus, a small HSP has also been recorded in response to abiotic stress, but its specific role remains undetermined (Pearson et al., 2010).
During desiccation sequences of cell-wall hydrolases, glycosyl transferases, and chitin deacetylases (CDA) -all involved in cell wall metabolismwere identified.CDA, a type of carbohydrate esterase, hydrolyzes the acetamide group in the Nacetylglucosamine polymers derived from glucose (e.g., chitin), and promotes the formation of glucosamine units (e.g., chitosan).The substrates for this enzyme come from the carbon (fructose 6phosphate) and nitrogen (glutamine) metabolism (Ghormade et al., 2010).CDA was first discovered in extracts of the fungus Mucor rouxii (Araki & Ito, 1975).It was further reported associated with cell wall synthesis.CDA has also been reported in association with spore formation in yeast and attack-defense systems in plant-pathogen interactions.In P. columbina an over-activation of this enzyme is likely to occur, due to the induction, under desiccation stress, of a fructose-1,6-phosphatase (Table 2) that can overproduce fructose 6-phosphate, an important precursor of the CDA substrate.Thus, the potential induction of CDA in desiccated P. columbina may help in remodeling and maintaining cell wall integrity during growth, survival, and pathogenesis.

CONCLUSIONS
Despite their crucial ecological role as primary producers, molecular information on stress responses in intertidal macroalgae remains limited.Pyropia columbina, such as others Porphyra and Pyropia species, is a good model for unravelling some of the biological and molecular responses associated with desiccation and other environmental conditions that may cause oxidative stress.In this context, two subtracted EST libraries were constructed in order to understand the metabolic pathways active during the hydration-desiccation cycle in this species.Results showed that a significant portion of the transcripts had no known homologues in algae or other organisms as far as sequence data were available.These sequences are interesting since they could represent genes unique to this species.On the other hand, several genes/proteins not previously described in algae were differentially expressed in both environmental conditions.This information contributes to a better understanding the molecular mechanisms involved in tolerance to desiccation.However, a confirmation of the expression profiles of the reported genes by qPCR is needed to characterize, for example, temporality of the gene expression profile during the hydrationdesiccation cycle.Additionally, our study provides for the first time, a set of candidate genes for further examination of the physiological responses to other environmental stressors.This genetic background will broaden our understanding on physiological differences may contribute, or even determine, the ecological features of macroalgae inhabiting the intertidal rocky zone.

Table 1 .
Functional category, identity and potential cellular destination of proteins, and accession number of identified ESTs in P. columbina under natural hydration (UH library).E value: the best (lowest) Expect value (E value) of all alignments from that database sequence, D*: Putative destination.

Table 2 .
Functional category, identity and potential cellular destination of proteins and accession number of identified ESTs in Pyropia columbina under natural desiccation (UD library).Data from the hydrated library (UH) is presented in Table1.E value: the best (lowest) Expect value (E value) of all alignments from that database sequence, D*: Putative destination.
HE859496 Figure 1.Functional categorization of ESTs obtained from hydration and desiccation P. columbina libraries.Percentage values indicate the number of ESTs grouped in each functional group in relation to the total number of sequences obtained in each library.