Hypoxia Influences Expression Profile of Pleckstrin Homology-Like Domain, Family A, Member 2 in Indian Catfish, Clarias Batrachu

Hypoxia influences expression profile of Pleckstrin homology-like domain, family A, member 2 in Indian catfish, Clarias batrachus (Linnaeus, 1758): A new candidate gene for hypoxia tolerance in fish

VINDHYA MOHINDRA*, RATNESH KTRIPATHI, PRABHAKER YADAV, RAJEEV KSINGH and KULDEEP KLAL National Bureau of Fish Genetic Resources (ICAR), Canal Ring Road, P. O. Dilkusha, Lucknow 226 002, India

*Corresponding author (Email, [email protected])

Several physiologically important genes were found to be regulated by hypoxia at the transcriptional level. The Pleckstrin homology-like domain, family A, member 2 (PHLDA2) gene was previously identified as an imprinted gene. The present study was aimed to determine the structure of complete cDNA and the deduced protein of PHLDA2 along with analysing the changes in its mRNA expression in Clarias batrachus tissues under hypoxic conditions. The complete cDNA of CbPHLDA2 gene consisted of 1009 nucleotides with an open reading frame of 417 nucleotides. The deduced CbPHLDA2 protein of 139 amino acids shared high homology with PHLD2A of other fishes as well as that of vertebrates. Importantly, a single amino acid (asparagine/lysine) insertion was identified in the PH domain of CbPHLDA2 and other fishes, which was absent in other vertebrates studied. Furthermore, under normoxic conditions, CbPHLDA2 was constitutively expressed with varying levels in analysed tissues. Short- and long-term hypoxia exposure resulted in significant changes in the expression of CbPHLDA2 in liver, spleen, head kidney, brain and muscle in a time-dependent manner. The results suggested that CbPHLDA2 might play an important role for adaptive significance under hypoxia.

[Mohindra V, Tripathi RK, Yadav P, Singh RK and Lal KK 2014 Hypoxia influences expression profile of Pleckstrin homology-like domain, family A, member 2 in Indian catfish, Clarias batrachus (Linnaeus, 1758): A new candidate gene for hypoxia tolerance in fish. J. Biosci. 39 433–442] DOI 10.1007/s12038-014-9426-z

1. Introduction of studies focused on these pathways in mammalian systems, limited studies are available on hypoxic response in aquatic Exposure of organisms to low-oxygen environment, or hyp- organisms, like fish, and Gracey et al. (2001) have shown oxia, induces an integrated response mediated by hypoxia- that there are many novel mRNAs which are differentially inducible transcription factor (HIF)-dependent and HIF- expressed under hypoxic conditions. Thus, identification of independent signalling pathways, which drives the expres- hypoxia-inducible mRNAs in fishes tolerant to low dis- sion of a number of genes for cellular adaptation (Scanlon solved oxygen conditions may define unique molecular and Glazer 2013). The expression of HIF-dependent genes events that would provide adaptation and survival to hypoxic regulates several biological processes, including cell prolif- environment (Geng et al. 2014). eration, angiogenesis, metabolism, apoptosis, immortaliza- The Indian catfish, Clarias batrachus (commonly known tion and migration (Harris 2002). Whereas HIF-independent as ‘magur’), is a freshwater air-breathing teleost species genes expression regulates the integrated stress response endemic to Indian subcontinent (Chonder 1999). The fish (ISR) and provides cellular adaptation to increased levels usually inhabits various habitats of low dissolved oxygen, of unfolded proteins generated endoplasmic reticulum (ER) i.e. wetlands, swamps, rivers, ponds and tanks, and burrows stress via unfolded protein response (UPR) pathway inside the mudflats during summer periods (Saha and Ratha (Rzymski and Harris 2007). Although, there are a number 2007; Mohindra et al. 2013a). Thus, C. batrachus is well

Keywords. CbPHLDA2; Clarias batrachus; hypoxia; differential expression http://www.ias.ac.in/jbiosci J. Biosci. 39(3), June 2014, 433–442, * Indian Academy of Sciences 433

Published online: 1 May 2014 434 Vindhya Mohindra et al. adapted to adverse ecological conditions and can be a useful respiration (referred as progressive hypoxia, PH) and model to study the mechanism of hypoxia tolerance as well thereafter, this DO was maintained by intermittent air as identification of candidate genes responsible for hypoxic circulation, for maximum of 12 h. Each experimental response (Mohindra et al. 2013b). group was treated with PH, 1 (H1), 6 (H6) or 12 During the course of our study for the identification of (H12) h, respectively (short-term). The corresponding hypoxia induced genes in C. batrachus,thePleckstrin control batches were kept under normoxia for same time homology-like domain, family A, member 2 (PHLDA2)gene periods as that of experimental once in similar manner. was found to be differentially expressed following hypoxia Before taking samples, fish were weighed and euthanized in exposures, from subtractive libraries data of C. batrachus tissues a 300 mg/L concentration of tricaine methanesulphonate (MS- (unpublished data). The PHLDA2 gene was originally identified 222, Sigma) to ameliorate suffering. Brain, muscle, liver, as an imprinted gene in placenta and liver (IPL) (Qian et al. spleen and head kidney samples were collected and snap 1997). PHLDA2 gene was found to be regulated by hypoxia in frozen in liquid N2 until further analysis. The tissue samples primary term human trophoblasts and its overexpression was were also collected from fish taken live in natural habitat associated with tumour suppression in osteosarcoma (Dai et al. (exposed to long-term hypoxia) during summer periods and 2012;Huanget al. 2012;Kimet al. 2007). The PHLDA2 control fish were corresponding lab-acclimatized ones. The protein consists of a single Pleckstrin homology domain with weight and length of fish collected from natural habitat was short N- and C-terminal extensions (Saxena et al. 2002). Among corresponding to those selected for lab experimental ones (30– the vertebrates, the PHLDA2 is highly conserved and primarily 80 g, 16–20 cm), and all the fish were sexually immature. involves in limiting placental growth (Saxena et al. 2002). The However, no information could be drawn about the fed status gene is maternally expressed and shows tissue-specific expres- of fish collected in natural habitat. The protocols for hypoxia sion with highest expression in placenta and low but detectable treatments followed were approved by Institutional Animal expression in fetal and adult liver and lung (Qian et al. 1997). Ethics Committee (IAEC). Further investigations showed that the PHLDA2 possess broad specificity and had moderate affinity for phosphatidylinositol phosphate (PIPs) (Saxena et al. 2002). However, the role of this 2.2 Identification of PHDLA2 transcript through SSH gene in response to hypoxia tolerance has not yet examined, and library construction apart from few studies from mammalian system, there were no reports available that describes the characterization, tissue dis- Total RNA was extracted from individual tissue samples tribution and/or expression pattern of PHLDA2 gene in fishes using the Nucleospin RNA II kit (Macherey-Nagel, Germa- under any ecological context. Thus, the present study was ny), as per the manufacturer’s instructions. Before cDNA undertaken to further gain insight about the molecular charac- synthesis, DNase I treatment was done to remove any geno- teristics of PHLDA2 cDNA and its expression pattern in differ- mic DNA contamination by adding 50 μL of 1U/μL RNase ent tissues of C. batrachus under hypoxia. free DNase I solution (EN0521; Thermos Scientific, Rock- ford, IL, USA) according to the manufacturer’s instruction. Further, 1 μg of total RNA was used to synthesize the first 2. Materials and methods strand cDNA using the Superscript III first-strand synthesis supermix for qRT-PCR kit (Invitrogen). The SSH library 2.1 Animals and hypoxic conditions (Super SMART cDNA Subtraction kit, Clontech) was constructed from samples collected from control and hypoxia For all the experiments, live fish (30–80 g, 16–20 cm) were challenged fish. The clones screened from library with in- collected from commercial catches and brought to laboratory serts were sequenced and EST sequences were analysed for for acclimation. During transportation, every care was taken the PHDLA2 transcripts. not to cause distress to the fish by putting them in sufficient area and volume of water. Fish were acclimatized at normoxia (5.00±0.1 mg/L, dissolved oxygen (DO)); at least 2.3 Characterization of full-length cDNA and structural for a month in tanks of 100 L capacity filled with 25 L of analysis of deduced protein of CbPHDLA2 water at 22±3°C. Processed feed of goat liver or flesh and soybean powder was provided to feed them once a day. Full-length cDNA of CbPHDLA2 was identified by the Feeding was stopped 48 h before the start of an experiment. similarity search and GeneScan tool (http://genes.mit.edu/ The hypoxic treatments to fish were as per Tripathi et al. GENSCAN.html). The protein-coding region was determined (2013). Briefly, fish were divided into 8 batches (4 control by selecting the longest amino acid sequence terminated and 4 experimental) of 3 fish each. For all the experimental before the polyadenylation signal and the deduced amino groups (without access to air), DO was brought from acid sequences were subjected to homology analysis using 5.00±0.1 mg/L to 0.98±0.1 mg/L by their own blastp program against protein database searches at NCBI

J. Biosci. 39(3), June 2014 A new candidate gene for hypoxia tolerance in fish 435

(http://www.ncbi.nlm.nih.gov/BLAST). Physiochemical and Nei 1987) based on the deduced amino acid sequence properties of CbPHDLA2 were identified through ProtParam of CbPHDLA2 and other sequences of PHDLA2 from fish- and ScanProsite tools on the ExPASy Server (http:// es, amphibians and mammalians representatives (table 1). www.expasy.org), while functional domains were identified The tree was drawn to scale, with bootstrap (2000 replicates) by SMART tool (smart.embl-heidelberg.de/). The identified using the Poisson correction model (Zuckerkandl and Pau- domains were further analysed by InterPro Scan (http:// ling 1965). All positions containing alignment gaps and www.ebi.ac.uk/Tools/pfa/iprscan/) and Superfamily 1.75 missing data were eliminated by pairwise deletion manner. tool (supfam.cs.bris.ac.uk). Subcellular localization was pre- Phylogenetic analysis was conducted in MEGA4 dicted through PSORT and iPSORT (http://psort.nibb.ac.jp/) (Tamura et al. 2007). (Nakai and Kanehisa 1992; Bannai et al. 2002). NetPhos 2.0 was used for consensus phosphorylation (serine, threonine or 2.5 Quantitative real-time PCR (qRT-PCR) and statistical tyrosine) site detection (www.cbs.dtu.dk/services/NetPhos/) analysis (Blom et al. 1999). The protocol for total RNA extraction and synthesis of first 2.4 Multiple sequence alignment and phylogenetic strand cDNA, from the individual tissue samples collected analysis from control and hypoxia challenged fish, was similar to as described above for library preparation. The expression of The deduced amino acid sequences of PHDLA2 from fishes, CbPHDLA2 was normalized against the expression of two amphibians and mammalian species (table 1) were obtained tissue specific reference genes, selected by pair wise fixed from NCBI database (http://www.ncbi.nlm.nih.gov/Entrez/) reallocation randomization test (Pfaffl et al. 2002). The and were aligned using the ClustalW2 programme (http:// reference genes were for Brain; 28S/α-tub,Liver;rpl30/ www.ebi.ac.uk/Tools/msa/clustalw2). The multiple sequence 28S, Muscle; rpl30/α-tub, Spleen; elf1α/28S, Head kidney; alignments were adjusted manually corresponding to differ- elf1α/α-tub in C. batrachus and their expression was not ent motifs and domains. Domain characterization of affected by hypoxia (Mohindra et al., unpublished). The CbPHDLA2 was according to Homo sapiens PHDLA2 pro- CbPHDLA2 and reference genes specific primers were de- tein (NP_003302.1), as present in the uniprot database signed using the PrimerQuestSM tool (Integrated DNA (http://www.uniprot.org/). An unrooted phylogenetic tree Technologies) (table 2). Further, 10-fold dilution series were was constructed by the neighbor-joining method (Saitou prepared to make standard curves from which primer effi- ciencies (E) was calculated according to the formula E = Table 1. PHLDA2 protein sequences accession numbers from 10−1/slope. The gel electrophoresis and dissociation curve Clarias batrachus and other species used for sequence alignment analysis was performed to confirm the specificity of the and phylogenetic analysis primer sets used. The total reaction volume for all qRT- μ μ μ Accession PCR reactions was 20 L (18 Lmastermixand2 L S. No. Protein Organism number undiluted cDNA made from 1 μg RNA/PCR product). The master mix contained 7.2 μLH2O, 0.8 μL of each primer 1 CbPHLDA2 Clarias batrachus KF577878 (0.4 μM final concentration) and 10.0 μLoftheSYBR 2 PHLDA2 Salmo salar NP_001134998.1 Green Mix (Roche Applied Science, Laval, PQ, Canada). 3 PHLDA2 Oryzias latipes XP_004084859.1 The following cycling conditions were used: (1) denatur- 4 PHLDA2 Oreochromis XP_003440449.1 ation, 5 min at 95°C; (2) amplification repeated 40 times, 10 s niloticus at 95°C, 10 s at 55°C, and 15 s at 72°C with ramp rate of 5 PHLDA2 Takifugu rubripes XP_003967405.1 4.4°C/s, 2.2°C/s, and 4.4°C/s, respectively; (3) melting curve 6 PHLDA2 Danio rerio NP_001018432.1 analysis, 5 sec at 95°C and 1 min at 65°C with ramp rate of 7 PHLDA2 Bos taurus NP_001069989.1 4.4 and 2.2°C/s, respectively, then up to 95°C at a rate of 8 PHLDA2 Xenopus laevis NP_001091206.1 0.1°C/s; (4) cooling, 10 s at 40°C with ramp rate of 2.2°C/s. 9 PHLDA2 Tupaia chinensis ELW71932.1 For each treatment and control, 3 individuals were analysed 10 PHLDA2 Homo sapiens NP_003302.1 in duplicate and reactions were performed in a Light Cycler 11 PHLDA2 Rattus norvegicus ABK76649.1 480 (Roche Applied Science) system. Second derivative 12 PHLDA2 Anas platyrhynchos XP_005019838.1 maximum method was employed to derive the Crossing point 13 PHLDA2 Ficedula albicollis NP_001186524.1 values (Ct) values, which were further compared and con- 14 PHLDA2 Equus caballus XP_001492569.2 verted to fold differences by the relative quantification meth- 15 PHLDA2 Mus musculus NP_033460.1 od using the relative expression software tool (REST) 384 v. 16 PHLDA2 Macropus eugenii BAJ09273.1 2(Pfafflet al. 2002). Absolute concentration values of CbPHLDA2 transcript were obtained by employing absolute

J. Biosci. 39(3), June 2014 436 Vindhya Mohindra et al.

Table 2. Primer sequences of PHLDA2 and reference genes in Clarias batrachus

Accession Amplicon Gene Number Primer (Forward) Primer (Reverse) E Size (bp)

PHLDA2 KF577878 GTGTCTGTTTGGCCTGTTGA CCCCACGTTTGAGAAGAAGA 1.96 98 rpl30 JK489341 TGCACGCTGGCCATCATTGAC AAATGGCTCTACTTCTCGCCCT 1.99 85 α-tub JK488355 GTCGAGCAAGGTAACGAACTTCAC GTGGATGGAGATACACTCACGCAT 1.92 111 elf1a GT157722 AATGGTTTAGATCTGCACCTGTTGCC ACCTTTATTAATTTCTCAGC 1.99 70 AGCTTTCTT 28S JK488212 TTTCAGGGCTAGTTGATTCGGCAG GGTTGATATAGACAGCAGGACGGT 1.92 92

Accession number, amplification efficiency (E) values and amplicon sizes, for the qRT-PCR analyses in the present study. PHLDA2 – pleckstrin homology-like domain, family A, member 2; rpl30 – ribosomal protein L30; α-tub – alpha tubulin, elf1a- elongation factor-1alpha; 28S – 28 S ribosomal RNA. quantification method corresponding to standard curve of 11–107). Superfamily 1.75 HMM library and genome assign- known concentration values with LightCycler® 480 Soft- ment server identified this protein as a member of Pleckstrin ware, Version 1.5. To examine the expression of CbPHLDA2 homology-like domain superfamily with e-value of 7.86e−20 transcript in C. batrachus tissues under normoxic conditions, (figure 2A). The post-translation modification (PTM) predic- absolute concentration values were analyzed with one-way tion by ScanProsite resulted in the identification of four po- ANOVA followed by the Tukey’s post-hoc test in SPSS v tential Casein kinase II, four Protein kinase C and one cAMP- 12.01 (SPSS, 2003). Significant differences were recorded at and cGMP-dependent protein kinase phosphorylation sites as P<0.05. well as one N-myristoylation site (table 3). Further, NetPhos 2.0 predicted the presence of seven serine, five threonine and one tyrosine phosphorylation sites in CbPHLDA2 (figure 2B). 3. Results SignalP 4.1 Server analysis revealed absence of signal peptide in CbPHLDA2 protein. Localization prediction by PSORT 3.1 Characterization of complete CDS and deduced and iPSORT tools showed CbPHLDA2 protein to be mainly protein of CbPHLDA2 localized in nucleus (43.5%), however, nuclear localization signal (NLS) was not found. The complete cDNA of CbPHLDA2 (accession no. ′ KF577878) consisted of 1009 nucleotides including a 5 - 3.2 Comparative and phylogenetic analysis untranslated region (UTR) of 175 nucleotides and 3′ UTR of 414 nucleotides (figure 1). The 417 nucleotides open reading Multiple sequence alignment of CbPHLDA2 protein with frame (ORF) encodes a protein of 139 amino acids with a fishes, amphibians and vertebrates showed that Pleckstrin predicted molecular weight of 16.05 kDa and the theoretical homology domain is highly conserved having isoelectric point of 8.80. Total number of positively and phosphoinositide binding site (figure 3). In C. batrachus negatively charged residues were 25 (Arg + Lys) and 21 and other fishes used in alignment, fish PHLDA2 protein (Asp + Glu) respectively. Based on concentration of Cyste- sequences contain a single amino acid insertion of Aspara- ine (Cys), Tryptophan (Trp) and Tyrosine (Tyr) residues, gine/Lysine in the PH domain, as compared to that of am- CbPHLDA2 was predicted to have extinction coefficient phibians and other vertebrates (figure 3). Phylogenetic values of 14230 M−1 cm−1 (assuming all pairs of Cys resi- -1 -1 analysis showed that CbPHLDA2 was clustered together dues form cystines) and 13980 M cm (assuming all Cys with the PHLDA2 proteins of fish group (figure 4). residues are reduced) at 280 nm measured in water. The instability and aliphatic index values were 42.77 and 71.51, respectively, while the grand average hydropathy value was 3.3 Expression of CbPHLDA2 gene under normoxic −0.704. Pairwise comparison showed that CbPHLDA2 condition shares high sequence identity with Salmo salar (86%), Oreochromis niloticus (84%), Oryzias latipes (83%), Under normoxic conditions, CbPHLDA2 was constitutively Takifugu rubripes (81%), Danio rerio (76%) as well as with expressed in all the examined tissues (Liver, Spleen, Head Homo sapiens (67%). kidney, Brain and Muscles). The transcript level was highest InterProScan analysis showed that CbPHLDA2 protein in head kidney (9.80×10−2±2.02x10−2) followed by spleen contains Pleckstrin homology domain (amino acid residues (5.97×10−3±1.19×10−3), muscle (2.87×10−3±7.11×10−4),

J. Biosci. 39(3), June 2014 A new candidate gene for hypoxia tolerance in fish 437

Figure 1. Complete nucleotide (1009 bp) and deduced protein (139 amino acid residues) sequence of PHLDA2 in Claris batrachus. The kozak motif (ATCATGA) and end codon (TGA) are highlighted with red colour. The identified Pleckstrin homology domain is underlined. The phosphoinositide-binding site is boxed.

Figure 2. (A) Pleckstrin homology domain prediction in the CbPHLDA2 protein of Clarias batrachus.(B) Prediction of serine, threonine and tyrosine phosphorylation sites in CbPHLDA2 protein of Clarias batrachus. X-axis represents the position of sequence, while Y-axis represents the phosphorylation potential (cut-off 0.5).

J. Biosci. 39(3), June 2014 438 Vindhya Mohindra et al.

Table 3. ScanProsite prediction of post-translation modification regulated in the liver (23.02-fold), while significantly sites found in the CbPHLDA2 protein of Clarias batrachus down-regulated in the spleen (1.98- fold) (figure 6) and head S. No. Post-translation modification Sites Sequence kidney (7.94- and 2.56-fold), respectively (figure 6). While, no significant difference was found in brain. After long-term 1 Casein kinase II 4–7 TpvE (natural) hypoxic conditions, CbPHLDA2 expression was phosphorylation 79–82 TtdD significantly up-regulated in liver (149.42-fold) and muscle 90–93 SgeD tissues (3.416 fold) (figure 6), while significantly down- 118–121 TrqD regulated in the brain (3.74-fold) (figure 6). No significant 2 Protein kinase C 8–10 SsK change was found in head kidney and spleen. phosphorylation 46–48 TqK 69–71 TgK 4. Discussion 132–134 SlR 3 cAMP- and cGMP- 29–32 KRkT 4.1 Characterization of CbPHLDA2: cDNA and protein dependent protein kinase phosphorylation 4 N-myristoylation 128–133 GQqeSL The present study had identified and characterized the fish PHLDA2 cDNA and deduced protein for the first time. The deduced amino acid sequence of CbPHLDA2 shared high − − brain (9.95×10 4±7.41×10 5) and was lowest in liver homology within the Pleckstrin homology domain among all − − (1.67×10 5±7.86×10 6). Its expression in these tissues was the analysed proteins, suggesting that the essential function significantly (p<0.05) different from each other (figure 5). of Pleckstrin homology domain in fishes may be similar to that in mammals (Saxena et al. 2002; Suzuki et al. 2011). The presence of potential PTM sites in the CbPHLDA2 3.4 Expression of CbPHLDA2 gene under hypoxic indicated that the protein might be phosphorylated and/or condition myristoylated, which is required to modulate protein function in maintaining cellular homeostasis (Karve and Cheema The short and long-term hypoxia resulted in significant 2011). Further, absence of signal peptide and NLS in change in the transcript levels of CbPHLDA2. Following CbPHLDA2 indicated the CbPHLDA2 protein to have dual short-term hypoxia, CbPHLDA2 transcript level was up- localization and could be located in both the nucleus and the

Figure 3. Multiple alignment and comparison of the deduced amino acid sequence of CbPHLDA2 of Clarias batrachus with fish, amphibians and vertebrates. Identical residues were marked by asterisk (*), Pleckstrin homology domain was underlined, Phosphoinositide (PI)-binding site is boxed. Symbol-▼ (red-coloured residues) denotes presence of one Asparagine/Lysine in fishes which are not present in amphibians and vertebrates. Potential post translational modification sites, casein kinase II (yellow-colour highlight), protein kinase C (green-colour highlight), cAMP- and cGMP-dependent protein kinase (pink-colour highlight) phosphorylation and N-myristoylation (residue in blue colour) sites are shown.

J. Biosci. 39(3), June 2014 A new candidate gene for hypoxia tolerance in fish 439

Figure 4. Phylogenetic analysis of PHLDA2 protein from Claris batrachus, and other fishes, amphibians and mammalian. Bootstrap values are given in percent and the scale bar indicates 0.2 substitution per site. extracellular compartment (Wang et al. 2004; Arnoysa and clading of CbPHLDA2 with fishes and other vertebrates was Wang 2007). Moreover, presence of phosphoinositide bind- consistence with the traditional taxonomy. Proteins with PH ing site suggested the interaction with PIP, PIP2 and PIP3 domains, like insulin receptor substrate (IRS protein), are proteins and plays an important role in different signaling reported to be activated by lipid second messenger phos- pathways (Saxena et al. 2002; Wang et al. 1994). Observa- phatidylinositol (3,4,5)-trisphosphate (PIP3) and facilitate tion of addition/insertion of asparagine/lysine at C-terminus the reduction in hepatic gluconeogenesis (Kalupahana et al. of PH domain in CbPHLDA2 and other fish species was 2011). As the CbPHLDA2 possesses the Pleckstrin homol- unique finding of this study. In contrast, PHLDA2 protein of ogy domain as well as PI binding site, it might interact with vertebrates contains poly-alanine at N-terminus, which is not PIP3 and modulate the metabolic responses in C. batrachus. present in fishes and amphibians. Moreover, taxon specific Thus, the identified modifications and substrate binding sites

Figure 5. Normoxic expression level of CbPHLDA2 mRNA in different tissues in Clarias batrachus as estimated by qRT-PCR. Expression values are given as nanogram (ng) of transcript estimated per microgram (μg) of total RNA. The letters a, b, c, d, e above the bars indicate homogenous subsets formed during Tukey’s post hoc test at significant differences (p<0.05) between tissues.

J. Biosci. 39(3), June 2014 440 Vindhya Mohindra et al.

Figure 6. Differential expression of CbPHLDA2 mRNA estimated by qRT-PCR, after short-term (PH, 1, 6 and 12 h) and long-term (NTR) hypoxia exposure in Clarias batrachus tissues. Y-axis represents the relative expression ratio (fold change) of CbPHLDA2 expression, mean±SE (N=3, in duplicate). X-axis represents hypoxic treatments as PH, progressive hypoxia upto 0.98±0.1 mg/L dissolved oxygen, H1, H6 and H12 (hypoxic time period, 1, 6 and 12 h at 0.98±0.1 mg/L, dissolved oxygen) and NTR, natural hypoxia exposure. Significant differences (p<0.05) in the expression levels of CbPHLDA2 in comparison to normoxic control group are indicated by asterisks (*) above bar.

in the CbPHLDA2 protein suggest that it might perform protein and have important role in their development (Qian specialized function in C. batrachus, which is needed to be et al. 1997). The PHLDA2 was found to be highly expressed established by further studies. in placenta and yolk sac and expressed at low level in liver (Frank et al. 1999). Moreover, it was reported that increased expression of hypoxia-inducible factor 2α (HIF-2α)was 4.2 Hypoxic regulation of CbPHLDA2 mRNA expression observed in growth restricted placentas suggesting that hypoxia may have a role to play in mal-development of the To our knowledge, this is the first report to describe the placental vasculature (Mccarthy et al. 2007). Also, it was effects of hypoxia on PHLDA2 gene expression in tissues found that down-regulation of PHLDA2 attenuates the im- other than placenta. PHLDA2 is a maternally expressed/ pact of hypoxia on placental growth (Kim et al. 2007). Thus, paternally silenced gene, imprinted in placenta and liver there seems a possibility that PHLDA2 might be a new target

J. Biosci. 39(3), June 2014 A new candidate gene for hypoxia tolerance in fish 441 gene for HIF-1α/2α, whose expression is regulated under Tolerance’. The financial support provided by NAIP-ICAR hypoxic conditions. is duly acknowledged. In the present study, highly up-regulated expression of CbPHLDA2, following different periods (short and long- term) of hypoxia, especially in liver tissue might be sugges- References tive of the altered hepatic metabolism, under hypoxic conditions. In mammals, PHLDA2 acts as a nutrient responsive gene and its higher expression results in increased adiposity Arnoysa EJ and Wang JL 2007 Dual localization: Proteins in (John and Tunster 2013) while, under hypoxic conditions, extracellular and intracellular compartments. Acta Histochemica – induction of HIF-1α expression in adipocytes has been di- 109 89 110 rectly linked to metabolic dysfunction (Sun et al 2013). Bannai H, Tamada Y, Maruyama O, Nakai K and Miyano S 2002 Extensive feature detection of N-terminal protein sorting signals. Moreover, HIF-1α has also been shown to play a role in Bioinformatics 18 298–305 liver metabolism by regulating hepatic glucose homeostasis Blom N, Gammeltoft S and Brunak S 1999 Sequence and structure- (Tajima et al 2009). In our earlier study also, the expres- based prediction of eukaryotic protein phosphorylation sites. sion of HIF-α subunits (HIF-1α,-2α and -3α) was found J. Mol. Biol. 294 1351–1362 to be differentially regulated in C. batrachus tissues Chi J-T, Wang Z, Nuyten DSA, Rodriguez EH, Schaner ME, Salim under hypoxia (Mohindra et al. 2013c). Thus, the results A, Wang Y and Kristensen GB et al. 2006 Gene expression suggest that PHLDA2 might be a new HIF target gene, programs in response to hypoxia: Cell type specificity and whose regulated expression can provide survival advan- prognostic significance in human cancers. PLoS Med. 3 e47 tage under hypoxia. However, more mechanistic studies Chonder SL 1999 Biology of finfish and shellfish (Howrah, India: – are needed to establish this hypothesis. In other tissues, SCSC Publisher) pp 303 313 there was no consistent pattern of its differential expres- Dai H, Huangy Y, Li Y, Meng G, Wang Y and Guo Q-N 2012 TSSC3 overexpression associates with growth inhibition, apo- sion under hypoxic conditions. It has been suggested ptosis induction and enhances chemotherapeutic effects in hu- that responses to hypoxia are evolutionarily conserved, man osteosarcoma. Carcinogenesis 33 30–40 but different cells respond to hypoxia in different fash- Denko NC, Fontana LA, Hudson KM, Sutphin PD, Raychaudhuri ion during adaptation, depending upon their microenvi- S, Altman R and Giaccia AJ 2003 Investigating hypoxic tumor ronment and baseline requirement of oxygen (Denko physiology through gene expression patterns. Oncogene 22 et al. 2003;Chiet al. 2006). 5907–5914 Frank D, Mendelsohn CL, Ciccone E, Svensson K, Ohlsson R and Tycko B 1999. A novel pleckstrin homology-related gene family 5. Conclusion defined by Ipl/Tssc3, TDAG51, and Tih1: tissue-specific expression, chromosomal location, and parental imprinting. – The present study has characterized the full-length cDNA of Mamm. Genome 10 1150 1159 Geng X, Feng J, Liu S, Wang Y, Arias C and Liu ZJ 2014 CbPHLDA2 and it was found that the PHLDA2 protein of Transcriptional regulation of hypoxia inducible factors alpha fishes has an insertion of single amino acid (Asn/Lys) in the (HIF-alpha) and their inhibiting factor (FIH-1) of channel catfish PH domain. Short and long-term hypoxia resulted in the (Ictalurus punctatus) under hypoxia. Comp. Biochem. Physiol. significant change in CbPHLDA2 expression, in a tissue- Part B, Biochem. Mol. Biol. 169 38–50. and time-dependent manner. Based on higher expression of Gracey AY, Troll JV and Somero GN 2001 Hypoxia-induced gene CbPHLDA2 observed in liver, under short as well as long- expression profiling in the euryoxic fish Gillichthys mirabilis. term hypoxic conditions, it was proposed that PHLDA2 may Proc. Natl. Acad. Sci. USA 98 1993–1998. act as a starvation signal for survival during times of energy Harris AL 2002 Hypoxia-a key regulatory factor in tumour growth. deficit, which possibly confers a survival advantage. How- Nat. Rev. Can. 2 38–47 ever, more studies are needed to further explore the mecha- Huang Y, Dai H and Guo Q-N 2012 TSSC3 overexpression reduces nisms of CbPHLDA2 expression in tissues and its possible stemness and induces apoptosis of osteosarcoma tumor- initiating cells. Apoptosis 17 749–761 association with HIF that may provide adaptive significance John R and Tunster S 2013 Epigenetic regulation of fetal growth under hypoxia. and healthy aging. Proceedings of Ageing - epigenetic mecha- nisms underlying lifelong health and wellbeing (eds) Ozanne and Reik (Birmingham, UK: 37th Congress of IUPS) SA61 Acknowledgements Kalupahana NS, Claycombe KJ and Moustaid-Moussa N 2011 (n-3) fatty acids alleviate adipose tissue inflammation and insulin resis- The authors are thankful to Director, NBFGR, for providing tance: mechanistic insights. Adv. Nutr. 2 304–316 facilities for this research work. This work was carried out Karve TM and Cheema AK 2011 Small changes huge impact: the under National Agricultural Innovation Project ‘Bio- role of protein posttranslational modifications in cellular homeo- prospecting of Genes and Allele Mining for Abiotic Stress stasis and disease. J. Amino Acids 2011 207691

J. Biosci. 39(3), June 2014 442 Vindhya Mohindra et al.

Kim H-S, Roha C-R, Chena B, Tyckoc B, Nelsona DM and Saxena A, Morozov P, Frank D, Musalo R, Lemmon MA, Skolnik Sadovsky Y 2007 Hypoxia Regulates the Expression of EY and Tycko B 2002 Phosphoinositide binding by the PHLDA2 in Primary Term Human Trophoblasts. Placenta 28 pleckstrin homology domains of Ipl and Tih1. J. Biol. Chem. 77–84 277 49935–49944 Mccarthy C, Cotter FE, Mcelwaine S, Twomey A, Mooney EE, Scanlon SE and Glazer PM 2013 Genetic Instability Induced by Ryan F and Vaughan J 2007 Altered gene expression pat- Hypoxic Stress. Stress-Induced Mutagenesis DOI 10.1007/978– terns in intrauterine growth restriction: potential role of 1-4614–6280–4_8 hypoxia. Amer. J. Obst. Gynecol. 196 70.e1–70.e6 SPSS 2003 Statistical Analysis Package SPSS version 12.01. Mohindra V, Tripathi RK, Singh A and Singh B 2013a Molecular (Chicago, Illinois: SPSS Inc.) characterization and expression analysis of a novel cystatin-like Sun K, Halberg N, Khan M, Magalang UJ and Scherer PE 2013 gene in a hypoxia-tolerant Indian catfish, Clarias batrachus Selective inhibition of hypoxia-inducible factor 1α ameliorates (Linnaeus, 1758). Fish Shellfish Immunol. 34 683–687 adipose tissue dysfunction. Mol. Cell Biol. 33 904–917. Mohindra V, Tripathi RK, Singh RK and Lal KK 2013b Molecular Suzuki S, Shaw G, Kaneko-Ishino T, Ishino F and Renfree characterization and expression analysis of PPP1R3C in MB 2011 Characterisation of marsupial PHLDA2 reveals hypoxia-tolerant Indian catfish, Clarias batrachus (Linnaeus, eutherian specific acquisition of imprinting. BMC Evol. 1758) under hypoxia. Gene 530 127–133 Biol. 11 244 Mohindra V, Tripathi RK, Singh RK and Lal KK 2013c Molecular Tajima T, Goda N, Fujiki N, Hishiki T, Nishiyama Y, Senoo-Matsuda characterization and expression analysis of three hypoxia- N, Shimazu M, Soga T, Yoshimura Y, Johnson RS and Suematsu M inducible factor alpha subunits, HIF-1α,-2α and -3α in 2009 HIF-1alpha is necessary to support gluconeogenesis hypoxia-tolerant Indian catfish, Clarias batrachus [Linnaeus, during liver regeneration. Biochem. Biophys. Res. Commun. 1758]. Mol. Biol. Rep. 40 5805–5815 387 789–794 Nakai K and Kanehisa M 1992 A knowledge base for predicting Tamura K, Dudley J, Nei M and Kumar S 2007 MEGA4: Molec- protein localization sites in eukaryotic cells. Genomics 14 897–911 ular Evolutionary Genetics Analysis (MEGA) software version Pfaffl MW, Horgan GW and Dempfle L 2002 Relative expression 4.0. Mol. Biol. Evol. 24 1596–1599 software tool (REST) for group-wise comparison and statistical Tripathi RK, Mohindra V, Singh A, Kumar R, Mishra RM and Jena analysis of relative expression results in real-time PCR. Nucl. JK 2013 Physiological responses to acute experimental hypoxia Acids Res. 30 e36 in the airbreathing Indian catfish, Clarias batrachus (Linnaeus, Qian N, Frank D, O'Keefe D, Dao D, Zhao L, Yuan L, Wang Q, 1758). J. Biosci. 38 373–383 Keating M, Walsh C and Tycko B 1997 The IPL gene on Wang DS, Shaw R, Winkelmann JC and Shaw G 1994 Binding of chromosome 11p15.5 is imprinted in humans and mice and is PH domains of beta-adrenergic receptor kinase and beta-spectrin similar to TDAG51, implicated in Fas expression and apoptosis. to WD40/beta-transducin repeat containing regions of the beta- Hum. Mol. Genet. 6 2021–2029 subunit of trimeric G-proteins. Biochem. Biophys. Res. Rzymski T and Harris AL 2007 The unfolded protein response and Commun. 203 29–35 integrated stress response to anoxia. Clin. Cancer Res. 13 2537–2540 Wang JL, Gray RM, Haudek KC and Patterson RL 2004 Saha N and Ratha BK 2007 Functional ureogenesis and adaptation Nucleocytoplasmic lectins. Biochim. Biophys. Acta 1673 to ammonia metabolism in Indian freshwater air-breathing 75–93 catfishes. Fish Physiol. Biochem. 33 283–295 Zuckerkandl E and Pauling L 1965 Evolutionary divergence Saitou N and Nei M 1987 The neighbor-joining method: A new and convergence in proteins; in Evolving genes and proteins method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4 (eds) V Bryson and HJ Vogel (New York: Academic Press) 406–425 pp 97–166

MS received 16 December 2013; accepted 10 March 2014

Corresponding editor: STUART ANEWMAN

J. Biosci. 39(3), June 2014