The unicellular halotolerant green alga species Dunaliella are able to proliferate in extremely varied salinities by synthesizing intracellular glycerol and adjusting the cell shape and volume. However, some marine Dunaliella species such as Dunaliella tertiolecta are not able to regulate cell volume as an immediate response to counter external osmotic shock. Here we report that a rapid shock-response mechanism is present in Dunaliella tertiolecta, involving uptake of exogenous glycerol in response to hyperosmotic shock without changing cell volume, and this glycerol uptake activity is associated with the Dunaliella tertiolecta glycerol uptake protein 1 (DtGUP1) gene, which belongs to the membrane-bound O-acyltransferase. The mutant DtGUP1-E, in which the DtGUP1 gene is silenced, displayed an inability to take up glycerol from the medium and showed cell death under hyperosmotic shock. To our knowledge, this is the first time a gene product has been reported in Dunaliella tertiolecta that is involved in glycerol uptake activity under hyperosmotic stress.
Micro-organisms are constantly exposed to highly variable environmental conditions. Survival of the micro-organisms relies on their ability to respond rapidly to these changes and to adapt to them appropriately and effectively. For instance, microbial cells strategically counter decreased water activity by initially losing intracellular water, modifying ionic fluxes, and accumulating a specific solute intracellularly. These methods regulate the cytoplasm osmotically to that of the surrounding environment [1-4].
The genus Dunaliella (Chlorophyceae, Volvocales), a unicellular biflagellate green alga, is one of the most halophytic micro-organisms, capable of thriving under a wide range of salt concentrations ranging from as low as 0.05 m to salt saturation . The long-term osmoregulatory response of Dunaliella species to increased salinity largely involves synthesizing and accumulating intracellular glycerol . The glycerol synthesis pathway involves conversion of glucose from photosynthesis or starch breakdown to fructose-1,6-diphosphate and then to dihydroxyacetone phosphate, which is then converted to glycerol-3-phosphate by glycerol-3-phosphate dehydrogenase [7-9]. Finally, glycerol-3-phosphate is converted to glycerol by glycerol-3-phosphatase [10, 11]. In the case of decreasing salinity, excess glycerol is removed by dihydroxyacetone reductase-mediated oxidation to form dihydroxyacetone, followed by dihydroxyacetone kinase-mediated phosphorylation to form dihydroxyacetone phosphate [12-14].
As it lacks a rigid cell wall, an immediate osmoregulatory mechanism is vital for Dunaliella to cope with the sudden salinity changes. For instance, the halotolerant species Dunaliella salina responds to varied salinity by promptly changing its shape and volume to regulate intracellular osmolarity [15-17]. However, the initial osmotic response of marine Dunaliella species, for example Dunaliellatertiolecta is unclear. In the present study, the cell volume of D. tertiolecta was found to remain relatively unchanged under the osmotic shock conditions. This suggests that D. tertiolecta and other marine algae may possess alternative rapid mechanisms to counter sudden osmotic shock before initiation of glycerol synthesis.
Many yeast species can uptake and utilize glycerol, both as a sole carbon source and as an osmolyte . The glycerol uptake systems across the plasma membrane have been described and characterized in many osmotolerant yeasts . These include Na+/glycerol co-transport in Debaryomyces hansenii , H+/glycerol symport in Pichia sorbitophila , and an active transport mechanism in Zygosaccharomyces rouxii . Additionally, Saccharomyces cerevisiae has been shown to possess an active mechanism for putative glycerol uptake, driven by a proton symport system and associated with two genes, glycerol uptake protein 1 (GUP1) and glycerol uptake protein 2 (GUP2) . GUP1p is important for recovery from salt stress as mutants lacking GUP1p were unable to survive osmotic shock. The GUP1 gene encodes a multi-membrane spanning protein that belongs to the membrane-bound O-acyltransferase (MBOAT) family . Members of the MBOAT family are found in yeasts, mammals and algae, most of which have not been characterized .
To examine whether D. tertiolecta utilizes a similar osmoregulatory mechanism by taking in exogenous glycerol across the plasma membrane, the algal cells were subjected to hyperosmotic shock and the extracellular glycerol concentration was measured. A homolog of GUP1 in D. tertiolecta, DtGUP1, was isolated using rapid amplification of cDNA ends (RACE) and its role in the uptake of glycerol was investigated. In addition, transgenic D. tertiolecta in which the DtGUP1 gene was suppressed by RNA interference was generated and characterized. Our results show that DtGUP1 is induced by hyperosmotic shock and is involved in glycerol uptake activity. We propose that DtGUP1 encodes a protein that is involved in the uptake of glycerol into the algal cells under high salt-stress conditions.
Glycerol uptake in Dunaliella tertiolecta upon hyperosmotic shock
To investigate the ability of D. tertiolecta to take up exogenous glycerol, cells grown in 0.5 m NaCl to logarithmic phase were centrifuged, and the cell pellets were resuspended in 4 m NaCl for hyperosmotic shock treatment and 0.5 m NaCl for isosmotic treatment (control). Glycerol at a concentration of 5 mm was then added to both treatment groups at various time points and glycerol was assayed 5 min after addition. When D. tertiolecta cells were subjected to osmotic shock by an increase in NaCl from 0.5 to 4 m NaCl, the glycerol uptake rate increased to a maximum of 3.14 pg glycerol·cell−1·min−1 during the first 15 min, and slowly decreased to baseline after 30 min (Fig. 1). The response of the cells to the osmotic shock is too rapid for protein synthesis to occur, indicating the existence of a constitutive shock-response mechanism involving glycerol uptake. In the control culture (isosmotic treatment), the rate of glycerol uptake was briefly increased to 0.780 pg glycerol·cell−1·min−1 at 5 min, but returned to baseline after 15 min.
Isolation and sequence analysis of the DtGUP1 gene
Rapid amplification of cDNA ends (RACE) was performed, and a glycerol uptake protein 1 (GUP1) homolog in D. tertiolecta was identified and named DtGUP1. The coding region of DtGUP1 was 1167 bp long, encoding a protein containing 388 amino acids (GenBank accession number KC01328). A BLASTX (BLAST Scoring Parameters, Gertz E, 2005) search against the National Center for Biotechnology Information protein database (http://www.ncbi.nlm.nih.gov/protein) revealed that the DtGUP1 protein shared approximately 50% identity with an MBOAT family protein from Chlamydomonas reinhardtii (accession number XP_001689816.1; NCBI Reference Sequence), and 40% identity with the Saccharomyces cerevisiae Gup1p (accession number NP_011431.1; NCBI Reference Sequence). A conserved MBOAT family protein domain (Fig. S1) was predicted to be located between positions 22 and 319 (http://pfam.sanger.ac.uk/).
A hydropathy plot and secondary structure analysis (http://www.ch.embnet.org) for the deduced polypeptide indicated that there were seven transmembrane domains with a large extracellular hydrophilic region at the N-terminus of the protein (Fig. S2). In addition, the amino acid histidine, which is conserved in the active site of MBOAT family proteins , was present in the predicted protein in a position similar to those for C. reinhardtii and S. cerevisiae, i.e. position 256 in hydrophobic loop TM4 of the DtGUP1 protein (Figs S1 and S2). All these features strongly suggest that the DtGUP1 cDNA encodes a putative glycerol uptake protein of the MBOAT family in D. tertiolecta.
DtGUP1 is induced by hyperosmotic shock
To further examine the involvement of DtGUP1 in the glycerol uptake activity upon hyperosmotic shock, the relative expression of the DtGUP1 transcript of D. tertiolecta was measured using quantitative real-time PCR. As shown in Fig. 2, in response to hyperosmotic shock, there was a significant increase in the DtGUP1 transcript expression level within 0.25 h (15 min), in comparison to the control group. It is also worth mentioning that addition of exogenous glycerol to facilitate glycerol import under extreme salt-stress conditions was not involved in the induction of DtGUP1 expression (P > 0.05). Thus, DtGUP1 may be directly involved in glycerol uptake when exposed to hyperosmotic shock.
Generation of transgenic Dunaliella tertiolecta by RNA interference
RNA interference technology was used to verify the biological function of the targeted DtGUP1 gene in D. tertiolecta under adverse environmental conditions. A plasmid, DtGUP1-RNAi, was constructed to express hairpin RNA containing the endogenous DtGUP1 gene (Fig. S3A), and was transformed into D. tertiolecta by a glass bead-mediated gene transfer system. An antibiotic-resistant transformant was obtained, and genotyping PCR analysis revealed acquisition of the bleomycin gene (ble) encoding zeocin-resistance protein from the plasmid (Fig. S3B). The transgenic D. tertiolecta was named DtGUP1-E.
The DtGUP1 transcript level in DtGUP1-E was measured during hyperosmotic shock (Fig. 3). There was no significant increase in expression of the DtGUP1 gene in DtGUP1-E during 1 h of high-salt shock, compared with wild-type (P >0.05). This indicates that expression of the DtGUP1 gene was successfully suppressed in DtGUP1-E during hyperosmotic stress.
DtGUP1-E fails to take in exogenous glycerol
The DtGUP1 gene is presumed to encode a protein involved in regulation of glycerol uptake under high-salt stress in D. tertiolecta. Transgenic D. tertiolecta in which the DtGUP1 gene was suppressed, DtGUP1-E, was subjected to the same hyperosmotic shock of 4 m NaCl to evaluate a possible reduction in the glycerol uptake rate. The rate of glycerol uptake in the first 15 min was reduced to approximately one-third in the DtGUP1-E mutant compared with the wild-type (Fig. 4). From 30 min onwards, DtGUP1-E cells were incapable of taking up exogenous glycerol under hyperosmotic shock, and leakage of glycerol was observed. Thus DtGUP1 gene silencing in transgenic D. tertiolecta clearly reduced the rate of glycerol uptake into the cells when exposed to extreme salinity.
Mutant yeast in which genes encoding for glycerol uptake proteins had been deleted was shown to be less capable of recovering from salt stress than wild-type . Figure S4 illustrates the damaging effect of high salinity on transformed DtGUP1-E. Under extreme hyperosmotic shock, the cell volume of DtGUP1-E swelled and the cells subsequently broke up, whereas the cell volume in the wild-type remained relatively constant (Fig. 5 and Fig. S4). As Dunaliella species have no cell wall, dead Dunaliella cells disintegrate readily under hyperosmotic conditions. As shown in Fig. 6, the number of intact DtGUP1-E cells decreased rapidly in 4 m NaCl, but the wild-type cell concentration remained unchanged. This indicates that DtGUP1 expression in D. tertiolecta regulates exogenous glycerol uptake under hyperosmotic stress, and provides a survival strategy for the alga.
Glycerol is recognized as a major compatible solute that is accumulated intracellularly by yeasts and the alga Dunaliella to compensate for low water activity in high-salt environments. In osmotolerant yeasts, glycerol was found to be actively transported against a concentration gradient , and glycerol uptake activity has been shown to be indispensable for salt-stress survival . In this study, we observed that, under conditions of hyperosmotic stress, the marine alga Dunaliella tertiolecta displayed the ability to take up exogenous glycerol within a very short time frame (Fig. 1). This finding supports the hypothesis of a rapid shock-response mechanism in D. tertiolecta, involving the uptake of glycerol, similar to what was demonstrated in yeasts [21, 22]. We observed that all Dunaliella species obtained from culture collection centers, UTEX (University of Texas at Austin, TX) and ATCC (American Type Culture Collection at Manassas, Virginia) release approximately half of the glycerol synthesized into the culture medium (data not shown). This may represent wastage for the alga, but appears to act as an important survival strategy, as the cells may instantly take up the extracellular glycerol at times of osmotic shock.
Our results show that, in D. tertiolecta, transcription of DtGUP1 was induced when subjected to 4 m NaCl hyperosmotic shock, in agreement with yeast studies [26, 27]. The transcription induction was evident with and without addition of exogenous glycerol, over a very short time period from 0.25 to 1 h (Fig. 2). This clearly indicates that DtGUP1 expression in D. tertiolecta is highly associated with increased NaCl concentration, regardless of the presence of exogenous glycerol. Thus, it is postulated that DtGUP1 encodes a putative glycerol uptake protein that is associated with the glycerol uptake system and specifically dependent on Na+, and this shock response is transient.
The RNA interference experiment further confirmed the functionality of DtGUP1 in glycerol uptake under salt shock conditions. When the DtGUP1 gene was suppressed (in DtGUP1-E), transgenic D. tertiolecta was unable to take in exogenous glycerol when subjected to 4 m NaCl hyperosmotic shock (Fig. 4). This led to death of the transgenic cells (Figs 5 and 6) and leakage of glycerol (Fig. 4) in DtGUP1-E, due to cell rupturing under high-salt conditions. This is in agreement with the fact that D. tertiolecta has a less elastic membrane (Fig. 5) and glycerol metabolism only starts 6–8 h after a hyperosmotic shock [8, 28]. We reasoned that, as GUP1 is also implicated in a wide range of crucial processes for cell functioning, such as membrane composition/integrity , glycophosphatidylinositol (GPI-anchor) remodeling , the secretory/endocytic pathway , cytoskeleton polarization during mitosis/budding , and lipid metabolism/assembly , the suppression of DtGUP1 in D. tertiolecta diminished its osmotic response, making the alga incapable of surviving the high-salt stress. Taken together, DtGUP1 has been show to play an osmoregulatory role in D. tertiolecta related to glycerol uptake across the plasma membrane in response to increased salt concentration in the medium.
In conclusion, our findings provide evidence that osmotolerant algae such as D. tertiolecta are able to take up glycerol from the medium under hyperosmotic stress, and this glycerol uptake system is closely associated with the DtGUP1 gene of the MBOAT family. Such uptake represents a rapid shock-response mechanism to counter osmotic shock before synthesis of glycerol is initiated.
Microalga and growth conditions
Dunaliella tertiolecta strain LB-999 was obtained from the UTEX Culture Collection of Algae (University of Texas at Austin, TX). D. tertiolecta cells were grown in a batch system in sterile ATCC-1174 DA liquid medium (American Type Culture Collection at Manassas, Virginia) containing 0.5 m NaCl on a rotary shaker at 25 °C, under a 14 h light/10 h dark regime (50 μmol photons·m−2·s−1). The cell concentration was counted using a hemocytometer.
Determination of the glycerol uptake rate during hyperosmotic shock
D. tertiolecta cells were first grown in sterile ATCC-1174 DA liquid medium containing 0.5 m NaCl for 5 days before transferring to the dark for 1 day for adaptation. D. tertiolecta cells were harvested by centrifugation at 5000 g for 5 min at 4 °C. The microalgae pellets were resuspended in isovolumetric experimental medium containing either 4 m NaCl for hyperosmotic shock or 0.5 m NaCl for isosmotic treatment in the dark. A stock solution of 1 m glycerol was added to the suspension at various time points of 5, 15, 30, 60 and 120 min to obtain a concentration of 5 mm glycerol. The glycerol-treated D. tertiolecta cells were then incubated for 5 min before harvesting by centrifugation at 5000 g for 1 min at 4 °C, and the supernatants were retained for assay of exogenous glycerol uptake. The glycerol content was determined using a free glycerol determination kit (FG0100; Sigma, St Louis, MO, USA) according to the manufacturer's instructions. A negative control consisting of experimental medium only was also quantified. The rate of exogenous glycerol uptake was determined as ([glycerol]sample −[glycerol]medium) cell−1·min−1. The reason for determining the rate of glycerol uptake by measuring depletion of glycerol in the medium is because glycerol may be metabolized in the cells.
Determination of cell volume
Hyperosmotically shocked D. tertiolecta cells were photographed at 400× magnification under a light microscope at 0, 5, 15 and 30 min. Cell sizes were measured using imagej software (ImageJ, US National Institutes of Health, Bethesda, MD, USA), and cell volumes were estimated according to the geometric ellipsoid model .
Total RNA of D. tertiolecta was extracted using an RNeasy® plant mini kit (Qiagen, Valencia, CA, USA), according to the manufacturer's instructions. The extracted RNA was treated with DNase to eliminate genomic DNA contamination. Isolated total RNA was stored at −80 °C.
Total RNA was used to synthesize random-primed cDNA using the SuperScript™ III first-strand synthesis system (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The synthesized cDNA was stored at −20 °C.
Cloning the GUP1 gene from Dunaliella tertiolecta by rapid amplification of cDNA ends (RACE)
The 5′ and 3′ ends of the D. tertiolecta GUP1 (DtGUP1) gene were cloned using a SMART™ RACE cDNA amplification kit (Clontech, Mountain View, CA, USA). The design of the RACE primers was based on the mRNA sequence of the homologous Chlamydomonas reinhardtii GUP1 gene (CHLREDRAFT_146988, NCBI Gene Bank). Conserved regions were located by alignment with homologs from various species such as algae, yeasts and plants. The DtGUP1 5′ RACE product was generated using a nested universal primer from the kit as a forward primer and DtGUP1 gene-specific primers (5′-CGCCAGATGACCAAAAACTT-3′ and 5′-AAACCTGTGAGGGCAAAATG-3′) as reverse primers. The DtGUP1 3′ RACE product was generated using gene-specific forward primers 5′-CCGCACGCTGTACTTCAAT-3′ and 5′-TAGCCAAGTATGGCCTGCTG-3′ and a universal reverse primer from the kit. The RACE-PCR products were ligated into pGEM-T vector (Promega, Madison, WI, USA) and sequenced directly. RACE 5′ and 3′ overlapping sequences were assembled using the EMBOSS:merger program (http://emboss.bioinformatics.nl/cgi-bin/emboss/merger) to obtain the full-length sequence of DtGUP1. The coding sequence of DtGUP1 was confirmed by PCR amplification using high-fidelity Taq polymerase (Agilent Technologies, Santa Clara, CA, USA).
Analysis of DtGUP1 gene expression
In order to ascertain whether DtGUP1 responds to salt stress, D. tertiolecta cells that were grown to logarithmic phase under 0.5 m NaCl were resuspended in 4 m NaCl or 4 m NaCl supplemented with 5 mm glycerol for hyperosmotic treatments in the dark. Cells resuspended in 0.5 m NaCl or 0.5 m NaCl supplemented with 5 mm glycerol for isosmotic treatments served as controls. Treated D. tertiolecta cells were harvested at 0, 0.25, 0.5, 1, 2, 3, 4 and 8 h by centrifugation at 5000 g for 5 min at 4 °C for RNA extraction and cDNA synthesis. Growth of the cultures in 4 m NaCl supplemented with 5 mm glycerol was monitored from day 0 to day 3.
The relative gene expression of DtGUP1 in D. tertiolecta was analyzed by quantitative real-time PCR using an ABI PRISM® 7500 real-time PCR system (Applied Biosystems, Carlsbad, CA, USA) and 2× Maxima® SYBR Green/ROX qPCR Master Mix (Fermentas, Waltham, MA, USA) according to the manufacturer's instructions. The primers 5′-ACTCCTCCTACAACCGATGG-3′ and 5′-CAAGCTGCCACAAATAGGAA-3′ were used to amplify the DtGUP1 gene fragment (105 bp). The relative abundance of the D. tertiolecta β-tubulin (DtTUB) gene was also determined and used as an internal control. The primers used for amplifying the DtTUB gene fragment (123 bp) were 5′-CAGATGTGGGATGCCAAGAACAT-3′ and 5′-GTTCAGCATCTGCTCATCCACCT-3′. Quantitative real-time PCR conditions were as follows: an initial heating cycle of 95 °C for 3 min, and 40 cycles of 95 °C for 5 s and 60 °C for 15 s. The specificity of each pair of primers was checked by a melting curve. Each experiment was repeated three times.
A 363 bp fragment of the DtGUP1 coding sequence (bp 397–759) was used to construct DtGUP1-RNAi, which expressed hairpin RNA-containing hairpin RNA-containing sequences homologous to the target DtGUP1 gene driven by the RBCS2 promoter. The primers DtGUP1_SF_SpeI (5′-ATAACTAGTATGAGCCGCATGCTTTACTT-3′), introducing an SpeI site, and DtGUP1_SR_XbaI (5′-TAATCTAGATGCCACAAATAGGAACACCA-3′), introducing an XbaI site, were used to amplify the sense fragment. The primers DtGUP1_ASF_ClaI (5′-CCCATCGATTGCCACAAATAGGAACACCA-3′), introducing a ClaI site, and DtGUP1_ASR_EcoRI (5′-CCGGAATTCATGAGCCGCATGCTTTACTT-3′), introducing an EcoRI site, were used to amplify the antisense fragment. The plasmid also incorporated a 431 bp RBCS2 promoter and a 526 bp bleomycin resistance gene (ble) to confer zeocin-resistance selection, using the primers RBCS2_F_ApaI (5′-AAGGGCCCGGGATGGATTAAGGATCCAC-3′), introducing an ApaI site, RBCS2_R_XhoI (5′-CCCTCGAGCAACCAAGAC-CAAAAACACC-3′), introducing an XhoI site, ble-F2-XhoI (5′-AAACTCGAGATGGCCAAGCTGACCAGC-3′), introducing an XhoI site, and ble-R2-ClaI (5′-CACATCGATTTAGTCCTGCTCCTCGGC-3′), introducing an ClaI site. The PCR reaction, in a volume of 20 μL, was hot-started at 95 °C using 0.5 μL template, and continued for 38 cycles of denaturation at 95 °C for 30 s, annealing at 58 °C for 30 s, and extension at 72 °C for 1 min. An addition extension was performed at 72 °C for 5 min. All amplified fragments were then inserted into the pGreen-HY104 vector  before or after the GUS fragment (Fig. S3A).
Dunaliella tertiolecta transformation
D. tertiolecta cells were transformed using the glass bead method as described previously  with minor modifications. Cultures grown to 1–2 × 106 cells·mL−1 in sterile ATCC-1174 DA liquid medium containing 0.05 m NaCl were harvested by low-speed centrifugation 5000 g for 5 min at 4 °C and resuspended in 0.05 m NaCl ATCC-1174 DA liquid medium to a cell concentration of 1–2 × 108 cells·mL−1. A 300 μL aliquot of this cell suspension was added to a 1.5 mL sterile screw-cap tube containing 0.3 g acid-washed glass beads (425–600 μm diameter; Sigma), 100 μL 20% polyethylene glycol (PEG-8000; Sigma), 0.5–1 μg plasmid and 5 μL fish sperm DNA. After vortexing for 30 s at maximum speed, the cells were plated immediately onto 0.5 m NaCl ATCC-1174 DA medium agar plates containing 25 μg·mL−1 zeocin (Invitrogen) as the selection marker. Colonies that appeared within 1 week were picked and inoculated into liquid selective medium consisting of 0.5 m NaCl ATCC-1174 DA medium containing 25 μg·mL−1 zeocin, and genotyping PCR was performed to confirm the existence of the transgene using the primers ble-gen-F1 (5′-AAGCTGACCAGCGCCGTTC-3′) and ble-gen-R1 (5′-CTCGCCGATCTCGGTCAT-3′).
Each result shown is the mean of three replicated studies. Statistical analysis of the data was performed using the program SPSS-15 (SPSS Inc. Released 2006. SPSS for Windows, Version 15.0. Chicago, SPSS Inc.), and significance was determined at the 95% confidence limit.
This work was supported by the Agency for Science, Technology and Research (A*Star) Science and Engineering Research Council (SERC) Carbon Capture & Utilization Thematic Strategic Research Program (grant number 092-138-0023).