Introduction
Heat shock and many other stress conditions induce the rapid expression of several protein products, which are generally called heat shock proteins (HSPs), play important roles in cell survival and apoptosis (
Arrigo et al., 2005;
Fan et al., 2005;
Guo and Chen, 2008). HSPs with low molecular masses of 12-43 kDa are called small heat shock proteins (sHSPs), and they commonly share a homologous sequence of approximately 90 residues called the “α-crystallin domain” (
Drinkwater and Crowe, 1987;
Liang et al., 1997). The synthesis of small heat shock proteins is developmentally regulated in many organisms (
Tanguay et al., 1993;
Coca et al., 1996;
Haynes et al., 1996;
Linder et al., 1996;
Marin et al., 1996). In this context, the brine shrimp
Artemia is an interesting experimental model, for example, fully hydrated, post-gastrula cysts survive at least 4 years under anoxic conditions in a state of quiescence (
Hand and Gnaiger, 1988;
Clegg, 1994,
1995). However, the molecular basis for the remarkable stress resistance of
Artemia cysts is poorly understood. We previously cloned and characterized a partial Hsp26 cDNA sequence in
Artemia sinica, which is obviously highly expressed in early embryo development stage (
Jiang et al., 2007). The cysts, composed of approximately 4000 cells and enclosed in a shell which is impermeable to non-volatile molecules, are in a dormant state known as diapause (
Laksanalamai and Robb, 2004). Embryos in diapause state synthesize large amounts of a developmentally regulated but Hsp26 with different stress, which reaches peak in encysted embryos and remains at high levels until larvae emerge from cysts (
Jackson and Clegg, 1996;
Liang and MacRae, 1999;
MacRae, 2003;
Sun and MacRae, 2005). Hsp26 is thought to contribute to stress resisting in encysted
Artemia embryos by acting as a molecular chaperone. Hsp26 also functions in more than one major cell compartment is indicated by reversible cytoplasmic to nuclear translocation in
Artemia embryos during development, upon exposure to stress and by pH modulation in vitro (
Clegg, 1994;
Willsie and Clegg, 2001, 2002;
Sun and MacRae, 2005). Despite the potential importance of Hsp26 in
Artemia embryo development, the Hsp26 has been characterized only in
A. franciscana and it has not been examined systematically in any other
Artemia species to date. Therefore, to lead a theoretical direction for further research, the Hsp26 in
Artemia is systematically analyzed with bioinformatic tools and databases in this present paper.
Methods
Artemia culture and embryonic heat stress
Artemia cysts were collectted from 8 different geographic strains (Table 1). All collected Artemia cysts were rinsed three times in distilled water, then incubated in filtered natural sea water at 28°C for 12 h. Diapause stage embryos were heated at 39°C and 1000 lux light intensities for 1h in aerated hatch incubator and then were immediately collected for RNA extraction or frozen in liquid nitrogen for later use.
RNA extraction
RNA was isolated by Trizol Reagent (Invitrogen, Carlsbad, USA) according to the manufacturer’s instructions. To remove genomic DNA contamination, total RNA was digested with RNase-free DNase I (Promega, Madison, USA). The integrity of the RNA was examined by electrophoresis in 1% agarose gel containing formaldehyde. The concentration of RNA was determined by a spectrophotometer (DU-640; Beckman, Fullerton, USA) at 260 nm and 280 nm. Total RNA was stored at - 80°C for use.
Reverse transcription PCR (RT-PCR) and sequencing of Hsp26 gene
Reversely transcription (RT) were performed using the ExScript RT reagent kit (TaKaRa, Dalian, China) with oligo d(T)18 as primer. The final reaction volume contained 2.5 M oligo d(T)18, 50 mM Tris-HCl (pH 8.3), 75 Mm KCl, 10 mM dithiothreitol, 8 mM MgCl2, 0.5 mM dNTP, 1 U RNase inhibitor, 5 U reverse transcriptase, and 100 ng RNA. RT was carried out in PCR Thermal Cycler Dice (TaKaRa) at 42°C for 10 min, followed by an inactivation step at 95°C for 2 min. cDNA was used for PCR or stored at - 20°C for use. A pair of general primers 5′- TACGGAGGATTTGGTGGTATG -3′, forward and 5′- CTTGTTGATCTTGCTGGAGTTG -3′, reverse, were designed according to conserved regions of Hsp26 gene sequences of A. franciscana available from the GenBank database (accession no. AF031367). The PCR reactions were carried out in a final volume of 20 L containing buffer (2.0 L), 2.5 mM dNTP (1.6 L), 10 pM primer (1.0 L), 1 U of Taq polymerase, 1 L cDNA and water to 20 L. PCR was carried out by an initial denaturation at 95°C for 5 min, followed by 35 cycles of 94°C denaturation for 45 s, 55°C annealing for 1 min, and a 72°C extension for 1 min. Amplification cycles were followed by a final 7 min extension at 72°C. PCR products were separated electrophoretically in 2.0% agarose gels and stained with ethidium bromide. After separation, the PCR products were extracted with TaKaRa Taq Cycle Sequencing Core Kits, cloned into pMD 18-T vector, then sequenced using an ABI 1377 automated sequencer (Applied Biosystems, Foster City, USA).
Transcript element analysis and secondary structure of the 3′-untranslated region
cDNA sequence was aligned to its corresponding genome sequence (Accession no. DQ310575) by using the Spidey program (http://www.ncbi.nlm.nih.gov/spidey/) to infer the exon/intron boundaries and the 5′ and 3′-untranslated regions. Recently, the first noncoding exon sequence was acquired by 5′-RACE was also used in our studies (
Qiu et al., 2006). Searching Transcription Factor Binding Sites (Threshold: 90.0 point) was done using TFSEARCH (http://www.cbrc.jp/research/db/TFSEARCH.html). Secondary structure of the Hsp26 3′ -untranslated terminator region was predicted by RNAstructure 4.4 software.
The physico-chemical properties and homology of the deduced protein sequence of Hsp26
The physico-chemical properties of the protein were examined by “SeqFacts” on the website (http://bip.weizmann.ac.il/sqfbin/seqfacts), while subcellular localization was estimated by “PSORT II” (http://psort.hgc.jp/). The web server “NetPhos 2.0 Server” (http://www.cbs.dtu.dk/services/NetPhos/) was used to produces neural network predictions for serine, threonine and tyrosine phosphorylation sites. Protein secondary structure analysis using phyre software was online from http://www.sbg.bio.ic.ac.uk/phyre/html/index.html. The program Blastp was employed to investigate the NCBI protein database for nonredundant-proteins in order to understand whether similar proteins exist in other species. Sequence alignment was performed with the ClustalX (ver. 1.81) program and genetic distances were estimated using the ‘Molecular Evolutionary Genetic Analysis’ (MEGA 3) software package based on the Kimura-2 parameter model (transition and transversion). Neighbor-Joining (NJ) methods in MEGA3 were used to construct phylogentic trees. Statistical significance of groups within inferred trees was evaluated using the bootstrap method with 1000 replications.
Results
RNA extraction and cDNA sequencing
Total RNA electrophoresis result was shown in Fig. 1, and the absorbance ratio A260/A280 of 1.8-2.0 indicating the good integrity of the total RNA. The cDNA were acquired by RT-PCR and the products were sequenced by ABI 1377 automated sequencer. The 579 bp complete open reading frame (ORF) and the deduced protein sequence of Artemia sinica (YC population) of Hsp26 were shown in Fig. 2.
Transcript element analysis and secondary structure of the 3′ -untranslated region
All
Artemia Hsp26 cDNA has an overall length (ORF) range of 576-588 bp, and it encodes 192-195 aa proteins named Hsp26. It was observed that Hsp26 possessed 4 exons and 3 introns (Fig. 3) when the cDNA was aligned to genome sequence (DQ310575) in agreement with previous results (
Liang et al., 1997;
Qiu et al., 2006). The results indicated that all exons exactly matched the genome sequence, and all of 4 exons contained typical acceptor and donor splice sites. The 1000bp 5′-flanking region corresponding to Hsp26 promoter region contained lots of putative cis-acting transcription regulatory sites such as 12 sites of HSE (heat shock elements), 3 sites of Dfd, 2 sites of Croc, 1 sites of BR-C, and 1 sites of dl. In addition, cis-acting elements were also predicted within the first intron region, including 21 sites of HSE, 2 sites of AP-1, 1 sites of Dfd; 28 sites of HSE, 1 sites of dl, 3 sites of Dfd, 1 sites of CF2-II within intron 2; 14 sites of HSE, 4 sites of Dfd, 2 sites of Hb, 1 sites of dl, 1 sites of AP-1 within intron 3.
Secondary structure analysis in 3′-untranslated region
The 3′-untranslated region following the coding region of Hsp26 was rich in A+ T and 4 predicted stem structures (over the threshold -4.0 Kcal/mol) resided in the transcribed mRNA sequence corresponding to this region. The stems were tandem organized and this region formed a simply secondary structure with a free energy of -4.0 Kcal/mol (Fig. 4).
The physico-chemical properties of Hsp26 deduced protein
Hsp26 protein sequence was submitted to “SeqFacts” tool, and its theoretical molecular weight was estimated to be 20703.05 Da and the calculated protein extinction coefficient was 15220. Hsp26 protein was classified as an acidic protein because of its isoelectric point of 6.3450. The antigenic sites are located at amino acid residues 68-83, 133-148, respectively (Fig. 2). Submitting Hsp26 protein sequence to PSORT II, one seven-peptide nuclear localization signals (NLS) “PFRRRMM” was found, which suggested that the Hsp26 protein was hypothesized to be located inside the nucleus (Fig. 2). NetPhos 2.0 Server predicted the numbers of phosphorylation sites of serine, threonine and tyrosine are 9, 8 and 1, respectively (Fig. 2).
Homology between various sHSPs and phylogenetic tree construction
All the Artemia Hsp26 protein sequences, including the eight sequenced in this study and the two dwonloaded from GenBank database were aligned (Fig. 5). Hsp26 proteins share a conserved α-crystallin domain and a moderately conserved C terminus and have variable N-terminal regions. The α-crystallin domain mediates formation of dimers, fundamental units of oligomerization for many sHSPs. The homology of α-crystallin domains is higher than C-terminal or N-terminal sequences among ten Artemia Hsp26 proteins (Table 2 and Fig. 5). There exist four of special motifs in all Artemia species, a glycine-enriched motif (G content 45.5%, 10/22), two arginine-enriched motif (R content 60%, 6/10; 62.5%, 5/8) and a threonine and serine-enhanced motif (S and T content 69.2%, 9/13) (Fig. 5). The secondary structure elements predicted with the phyre program showed that Artemia Hsp26 consist of 4 helices and 10 sheets (Fig. 5). Moreover, we searched the nonredundant protein database with the Artemia sinica Hsp26 protein sequencs by using “Blastp” and the protein identities in the different species were as follows: sHSP in T. pseudospiralis 73/171 (42%); the hsp20.1 in B. mori 62/136 (45%); sHSP in T. castaneum 61/137 (44%); the 19.8 kDa small heat shock protein in C. fumiferana 49/98 (50%); heat shock protein 20.7 in L. migratoria 54/107 (50%); predicted similar to heat shock protein 1 in N. vitripennis 55/101 (54%); heat shock protein 19.7 in M. brassicae 54/98 (55%); small heat shock protein 19.7 in C. suppressalis 54/98 (55%); heat shock protein Hsp21.3 in L. sativae 54/91 (59%); 19.9 kDa small heat shock protein in C. fumiferana 66/173 (38%); crystallin, alpha B, isoform CRA_c in H. sapiens 54/103 (52%); crystallin, alpha B, isoform CRA_b in R. norvegicus 54/103 (52%); Alpha(B)-crystallin isoform 3 in C. familiaris 54/103 (52%); crystallin, alpha B in M. mulatta 51/98 (52%); crystallin, alpha B in B. taurus 51/98 (52%). A phylogenetic tree was constructed for 28 sHSP protein sequences, including 8 Artemia species sequences from this study, 6 Artemia species sequences from the NCBI databases and 14 sHSP sequences of other species (Fig. 6). The phylogenetic tree reveals a grouping of sHSPs into two main branches, one comprising invertebrate proteins, and the other one containing vertebrate proteins. Hsp26 first clustered with insect and arthropod, then with mammal sHSPs which indicated that Hsp26 of Artemia species is more closely related to the sHSPs from insect and arthropod than from mammalian. The branching pattern of the phylogenetic tree corresponded essentially with the evolutionary relationships among the species.
Discussion
Of the molecular chaperones, small heat shock proteins (sHSPs) are noted for their protective capability, representing the first line of defense against environmental and physiologic stresses (
Sun and MacRae, 2005;
Maimbo et al., 2007). Based on the fact that the Hsp26 proteins are of widespread occurrence when
Artemia species are under environmental and physiologic stresses and sHSPs have important roles in organism defense, the informatic identification and characterization of Hsp26 is an initial important step for better understanding of its precise function/role in
Artemia. In previous studies, yeast, many invertebrates and plant sHSP genes lack introns. However,
Artemia Hsp26 containing 3 introns, indicates that these introns are likely to play important roles in stress resistance. Our study showed that not only the 5′-flanking region and intron 1 contains many putative cis-acting sites (
Qiu et al., 2006) but also the intron 2 and intron 3 contain lots of cis-acting sites, including heat shock factors (HSFs), AP-1, Dfd, dl, CF2-II, Hb and AP-1binding sites.
In mammalian, hsp25 genes contain stress-responsive HSEs (
Kato et al., 2002;
Voellmy, 2004) and their expression is developmentally regulated in diverse embryonic cells and tissues (
Davidson et al., 2002). HSEs also occur in the 5′-flanking region of other sHSPs (
Candido, 2002;
Guan et al., 2004), implying control of gene expression by HSFs (heat shock factors) (
Pirkkala et al., 2001;
Qiu and MacRae, 2008). The other cis-acting elements, for emample, AP-1 (activator protein 1) cis-acting sites as potential modulators of expression (
Qiu et al., 2006), reflecting the diversity of regulatory elements in sHSP genes from different organisms (
Gopal-Srivastava et al., 1998;
Ilagan et al., 1999;
Qiu and MacRae, 2008). Furthermore, AP-1 as a member of the basic helix–loop–helix zipper transcription factor family can influence mammalian sHSP gene expression during embryo development and stress. All the putative cis-acting regulatory sequences still require further analysis to determine their regulatory significance. In 3′-untranslated region of
Artemia Hsp26 cDNA, the pairwised stems and the unpairwised loops constituted many stem-loops or “panhandles,” and they were followed by poly (A) sequences (Fig. 4). Therefore, this 3′-untranslated terminator contains the basic structure basis for transcriptional termination.
The predicted Mr 20.70305 kDa of Hsp26 is not consistent with the observed experiment Mr (~26kDa) of most Hsp26 proteins reported in GenBank. This smaller predicted Mr could be explained by the fact that post-translational modifications must occurred to Hsp26 protein in
Artemia. This is consistent with our predicted results that several serine, threonine and tyrosine phosphorylation sites were reside in Hsp26 protein sequences (Fig. 2). Protein phosphorylation at serine, threonine or tyrosine residues affects a multitude of cellular signaling processes. Some sHSPs, including Hsp26, have several putative phosphorylation sites that are characteristics of serine, threonine and tyrosine protein phosphorylation sites. The common presence of these sites in Hsp26 proteins suggests that these proteins are regulated and activated by similar phosphorylation and preceded further possible modification. These findings indicate that the phosphorylation in Hsp26 could be a general property of sHSPs and suggest potential functions of Hsp26 against environmental and physiologic stresses in
Artemia. It is a general rule of biology that living organisms or cells require a continuous and substantial flow of free energy to maintain their normal physiologic conditions. Organisms must have more energy need when outer stresses conditions occur. It is important to know whether phosphorylation site in Hsp26 is associated with energy production. However, it is noteworthy that the small heat shock protein family does not require nucleoside triphosphates (e.g. ATP) to perform chaperone functions in vitro (
Jakob et al., 1993;
Parsell and Lindquist, 1993;
Waters et al., 1996). Further study of the potential role of phosphorylation of Hsp26 in
Artemia embryo development progress is needed.
Small heat shock proteins enter nuclei during stress (
Arrigo and Welch, 1987;
Kato et al., 1993,
1994), but their mission within the nuclei remains elusive. For example, Hsp26 occurs in
Artemia nuclei during diapause and stress and may stabilize nuclear matrix proteins (
Willsie and Clegg, 2001,
2002), but whether Hsp26 as molecular chaperone moves into nuclei is unknown.
Artemia embryos Hsp26 nuclear migration may depend on nuclear localization signals (NLS) “PFRRRMM” which predicted in our study by PSORTII. Additionally, the predicted isoelectric point 6.3450 of Hsp26 also implied that this protein locating inside nuclei was possible.
Our objectives in this study were to analyze further the properties of Hsp26 and to explore more fully the relationship of Hsp26 protein to other small heat shock proteins through study and prediction of its molecular structure information. Additionally, we hoped to shed light on the role of Hsp26, thus contributing to our understanding of small heat shock proteins in other organisms.
Higher Education Press and Springer-Verlag Berlin Heidelberg