- Email: [email protected]
Structural evolution and tissue-specific expression of tetrapod-specific second isoform of secretory pathway Ca2+-ATPase
Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
Contents lists available at SciVerse ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Structural evolution and tissue-specific expression of tetrapod-specific second isoform of secretory pathway Ca2+-ATPase Nikolay B. Pestov a,⇑, Ruslan I. Dmitriev a, Maria B. Kostina a, Tatyana V. Korneenko a,b, Mikhail I. Shakhparonov a, Nikolai N. Modyanov b,⇑ a b
Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow 117871, Russia Department of Physiology and Pharmacology, University of Toledo College of Medicine, 3000 Arlington Ave., Toledo, OH 43614, USA
a r t i c l e
i n f o
Article history: Received 22 December 2011 Available online 3 January 2012 Keywords: Vertebrate evolution Gene duplication Enzyme isoforms Golgi complex Alternative splicing CPCA2 Ca-ATPase
a b s t r a c t Secretory pathway Ca-ATPases are less characterized mammalian calcium pumps than plasma membrane Ca-ATPases and sarco-endoplasmic reticulum Ca-ATPases. Here we report analysis of molecular evolution, alternative splicing, tissue-specific expression and subcellular localization of the second isoform of the secretory pathway Ca-ATPase (SPCA2), the product of the ATP2C2 gene. The primary structure of SPCA2 from rat duodenum deduced from full-length transcript contains 944 amino acid residues, and exhibits 65% sequence identity with known SPCA1. The rat SPCA2 sequence is also highly homologous to putative human protein KIAA0703, however, the latter seems to have an aberrant N-terminus originating from intron 2. The tissue-specificity of SPCA2 expression is different from ubiquitous SPCA1. Rat SPCA2 transcripts were detected predominantly in gastrointestinal tract, lung, trachea, lactating mammary gland, skin and preputial gland. In the newborn pig, the expression profile is very similar with one remarkable exception: porcine bulbourethral gland gave the strongest signal. Upon overexpression in cultured cells, SPCA2 shows an intracellular distribution with remarkable enrichment in Golgi. However, in vivo SPCA2 may be localized in compartments that differ among various tissues: it is intracellular in epidermis, but enriched in plasma membranes of the intestinal epithelium. Analysis of SPCA2 sequences from various vertebrate species argue that ATP2C2 gene radiated from ATP2C1 (encoding SPCA1) during adaptation of tetrapod ancestors to terrestrial habitats. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Mammalian Ca-ATPases, calcium pumps, are classified into three major groups: plasma membrane Ca-ATPases (PMCA), sarco-endoplasmic reticulum Ca-ATPases (SERCA) and secretory pathway Ca-ATPases (SPCA). They have major implications for normal and pathological physiology [1]. The first mammalian SPCA, SPCA1, was discovered in 1992 by Guntesky-Hamblin et al. [2] by molecular cloning from rat stomach and testis [3,4] and was shown to exhibit significant homology to the previously characterized yeast ATPase PMR1 [4]. Up to now, mammalian SPCA remained less studied than SERCA, PMCA, or PMR1, the yeast homolog of SPCA, which is required for normal Golgi function [5]. Since PMR1 mutants survive in a narrow interval of calcium and manganese concentrations [6] it was suggested that PMR1 is also involved in manganese transport. Superexpression of PMR1 allowed the direct demonstration of calcium transport, as well as sensitivities to vanadate, thapsigargin and ⇑ Corresponding authors. Fax: +1 419 383 2871. E-mail addresses: [email protected] (N.B. Pestov), nikolai.modyanov@utoledo. edu (N.N. Modyanov). 0006-291X/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.12.135
cyclopiazonic acid that are although different from those of other known Ca-ATPases [7]. This recombinant PMR1 was also purified and demonstrated directly to be able to translocate not only Ca, but also Mn. Substitution Q783A led to the loss of the ability to recognize Mn whereas Ca transport remained intact [8,9]. Thus it can be concluded that yeast PMR1 is a Golgi-resident (Ca2+, Mn2+)-ATPase, and its major physiological role is to provide glycoprotein processing enzymes with calcium and manganese. SPCAs attracted significant attention after it was found that mutations of Ca-ATPases SERCA2 and SPCA1 are associated with Darier and Hailey-Hailey [10] diseases, respectively. It is especially interesting that, although the latter genes are expressed ubiquitously, the only manifestations of their haploinsufficiencies are skin lesions. Human SPCA1 was expressed in yeast and was shown to have catalytic properties similar to those of PMR1, i.e. it is a Ca, MnATPase [11]. From this one can conclude that SPCA represents a conserved system of Ca handling that exists in all mammalian cells (SPCA1 transcripts were detected in all tissues tested [2,3]). However, in vertebrates the situation turned out to be more complex, because of the existence of a putative second isoform of SPCA that was first found in the course of a large-scale transcriptome study
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
[12]. This novel transcript has been cloned from human brain and named KIAA0703. Also, human genome projects allowed the sequencing of the complete human gene ATP2C2, and its localization on chromosome 16q23. Since then, the catalytic properties of overexpressed human SPCA2 highlighted small but important differences between SPCA pumps. For example, SPCA2 was found to have significantly higher Ca2+ affinity in comparison with SPCA1 [13–15]. Mammary gland contains significant amounts of SPCA2 and, more importantly its expression upregulates sharply during lactation [16], and drops just before mammary gland involution [17]. SPCA2 expression may be regulated by prolactin [18]. The recently reported ability of SPCA2 to interact and activate the Ca-channel Orai1 in mammary tumor cells, argues that SPCA2 may also have transport-independent functions [19]. The first SPCA2 coding sequence was cloned from human brain, and, indeed, it may have important functions in this organ. ATP2C2 is now considered as a candidate gene, mutations in which may be the cause of certain memory and speech disorders (reviewed in [20]). Here we report the sequence and intracellular localization of rat SPCA2, a detailed comparison of SPCA2 and SPCA1 expression in adult rat and newborn piglet tissues, and the phylogenetic relationships of SPCA2 from various vertebrate species. 2. Materials and methods An expanded Materials and Methods section can be found in the online supplement.
1299
resequenced. Xenopus laevis SPCA2 transcript sequence has been obtained by resequencing a partially sequenced cDNA clone. The nucleotide sequences reported here have been submitted to the GenBank™/EBI DATA Bank with accession number AF484685 for Rattus norvegicus SPCA2 and DQ420634.1 for X. laevis SPCA2. Tissue-specific expression of various transcripts was analyzed in rat, human and pig tissues by standard RT-PCR. 2.2. Localization of SPCA2 in cultured cells A full-length rat SPCA2 plasmid with C-terminally fused GFP was constructed and used for transfection followed by confocal fluorescent microscopy. 2.3. Immunochemical methods Recombinant proteins comprising the second cytoplasmic domains of SPCA1 and SPCA2 were expressed in Escherichia coli, purified and used for immunization of rabbits. The antibodies were used either as pan-SPCA probes or they were affinity purified to obtain a SPCA2 specific fraction. Membrane fractions were prepared from rat tissues employing several fractionation protocols and were used for western blotting with the antibodies followed by chemiluminescent detection. For immunohistochemistry, tissue sections were prepared by several different techniques, stained with the anti-SPCA2 antibodies followed by Alexa Fluor-conjugated secondary antibodies, and images collected using a fluorescent microscope. 3. Results
2.1. RT-PCR, cloning and sequencing 3.1. Sequence analysis of ATP2C2 gene and the encoded SPCA2 protein The full-length rat SPCA2 sequence was determined by cloning and sequencing of 30 - and 50 -RACE PCR products from rat duodenum. Full-length SPCA2 was cloned by long range RT-PCR and
We have successfully reconstructed the full-length rat SPCA2 transcript from duodenum using RACE PCR. The correctness of
Fig. 1. RT-PCR analysis SPCA1 and SPCA2 expression in rat and pig tissues. (A) Adult rat. (B) Newborn piglet. RT-PCR products from 0.05 lg total RNA were electrophoresed and stained with ethidium bromide. The numbers of cycles shown at the right were chosen for each isoform to compare expression levels in different tissues. The band intensities may not necessarily reflect relative contents of different isoforms in the same tissue.
1300
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
the assembled sequence has been confirmed by amplification of the full-length ORF. Homology search of the GenBank database against the assembled rat SPCA2 mRNA and protein indicates that human ATP2C2 gene encoding SPCA2 spans 95.4 kbp and is composed of 27 exons. Intron sizes vary from 82 bp (intron 21) to 29.8 kbp (intron 1). Exon 1 appears to include all of the 50 -UTR and exon 27 – all of the 30 -UTR. The mouse gene is shorter (56.5 kb) due to compacted introns (for example, intron 1 is 17.5 kbp). Positions of the exons with respect to the amino acid sequence are well conserved. Human and mouse genes are located on chromosomes 16q23 and 8, respectively. Exon–intron structures of ATP2C1 and ATP2C2 are similar, human ATP1C1 spans 107.5 kbp, also with a very long intron 1 (35.6 kbp). There are 27 exons altogether (28 exons for an alternatively spliced variant). Positions of the exons with respect to amino acid sequences are well conserved with the exception of the boundary between divergent exons 1 and 2. The amino acid sequence of SPCA2 deduced from the sequenced rat duodenum cDNA contains 944 residues. Predicted human SPCA2 protein contains 946 residues. Sequence alignment of rat and X. laevis SPCA2 with rat SPCA1 is shown in Supplementary Fig. 1. SPCA2 features a structure similar to other Ca pumps that fits to the conventional 10 transmembrane helices structure with large cytoplasmic domains. Rat SPCA2 has 85.0% residues identical to human SPCA2, 65.1% – to rat SPCA1, 43.9% - to yeast PMR1, 24.6% – to SERCA2. Especially divergent are N-terminal fragments encoded by exons 1 and 2 (Supplementary Fig. 2). Significantly, SPCA2 has a longer N-terminus than SPCA1 (mostly due to enlargement of the coding part in exon 1) and, as a result, has a somewhat larger molecular weight, 103 kDa, compared to 100 kDa of SPCA1.
3.2. Analysis of the 50 -end of human SPCA2 mRNA The previously reported structure of human SPCA2 (KIAA0703) [15] has an N-terminus without homology to any known protein. Similarity search indicates that its 50 -terminus originates from the 30 -terminus of intron 2. To clarify this issue, we have made an amplification with forward primers complementary to the 50 ends of KIAA0703 and to our theoretically predicted ‘‘canonic’’ 50 end (at translation initiation start) whereas the backward primer was complementary to exon 3 (Supplementary Fig. 3). All primers gave positive results with available samples of human cDNA (intestines and brain) although the ‘‘canonic’’ variant has an apparently higher level than KIAA0703. This indicates that both sequences exist in the transcriptome. However, considering the absence of homology between the N-terminus of KIAA0703 and any other Ca-ATPase, it is reasonable to suggest that KIAA0703 is a result of intron retention or transcription from an intron promoter. We obtained retentions of intron 5 and a small part of intron 1 in rat cDNA. Overrepresentation of SPCA2 variants with introns retained during molecular cloning is an interesting artifact. Alternatively, one can speculate that most of SPCA2 pre-mRNA is normally underspliced in vivo. This may be especially true for introns with complex splicing. Large introns may be removed stepwise through transient retention of internal exons [21], and this seems to be the case with exon 1 of rat ATP2C2. 3.3. Tissue distribution of SPCA2 transcripts in comparison with SPCA1 Analysis of tissue-specific expression of rat SPCAs is shown in Fig. 1A. In accordance with previous results of Guntesky-Hamblin et al. [2,3], SPCA1 is expressed ubiquitously. However, its level is
Fig. 2. Confocal imaging of subcellular distribution of SPCA2 tagged with green fluorescent protein in cultured cells. Mouse C2 myoblasts were transfected with a plasmid encoding the SPCA2-gfp chimera. (A and B) Fluorescence recorded at low (A) and high (B) detector sensitivities. (C) Phase contrast image of the cells. (D) A and C images merged.
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
variable, being apparently higher in kidney, testis, adrenal and mammary gland (without a strong effect of pregnancy and lactation), whereas some tissues, e.g. intestines, are relatively poor in SPCA1 transcripts. SPCA2 is very different – the expression level of rat SPCA2 varies between tissues to a great extent. It is the most abundant in lactating mammary gland, trachea, lung, intestines (especially in colon), skin and preputial gland. Trace levels of SPCA2 transcripts (at high number of cycles) can be detected in almost all tissues, although the expression in liver, adrenal, brain and skeletal muscle appears to be negligible. Lactating mammary gland has a much stronger signal than prelactating and quiescent glands from pregnant or virgin females. Another example of tissue-specific expression pattern has been obtained with the newborn pig (Fig. 1B). A larger size of pigs gave the possibility to study more tissues including several small organs. Pig SPCA1 is expressed almost ubiquitously, with the highest levels in choroid plexus, retina, adrenal, skin and salivary gland. Blood cells seem to have only traces of SPCA1. The tissue-specificity profile of pig SPCA2 is very similar to that in the rat. Strong signals are observed in lung and gastrointestinal tract (higher in large intestine) whereas tissues like liver and striated muscles have negligible levels of SPCA2. Interestingly, significant expression is observed in parathyroid gland. There are also some dissimilarities that can be attributed to either species- or age-specific differences, such as expression of SPCA2 in testis, ovary and adrenal (more pronounced in the piglet) and skin (higher in the rat). One difference is, however, very remarkable: pig SPCA2 shows the highest level in bulbourethral gland whereas this organ in the rat has a minute level of the transcript.
1301
Analysis of rat, mouse and human ESTs in GeneBank (not shown) corresponds very well with our RT-PCR analysis: rodent SPCA2 is the most abundant in lactating mammary gland and colon. Human data are not so similar: human prostate, unlike rodent prostate, may also contain SPCA2 indicating species-specific differences of SPCA2 usage in various exocrine glands. Note that no ESTs of SPCA2 can be found in brain or kidney, in line with our analyses but contrary to the results of KIAA0703 detection in the brain and kidney [12]. Also interesting is the presence of many ESTs in tumors of tissues where SPCA2 is expressed normally (colon, lung and mammary gland) [13,15]. 3.4. Cellular and subcellular localization of mammalian SPCA proteins For subcellular localization studies we constructed plasmid DNA which encodes full-length rat SPCA2 fused with green fluorescent protein (GFP) at the C-terminus. After transient transfection of mouse C2C12 myoblasts Golgi-like distribution of chimeras was observed: (Fig. 2). Similar results were obtained for all tested cell lines (human adenocarcinoma HT-29 cells, and CHO cells, results not shown). Using pan-SPCA polyclonal antibodies we detected SPCA1 and SPCA2 in membranes of rat tissues. SPCA1 was detected in all tissues tested at comparable levels except kidney, colon and heart, where its content was higher. Electrophoretic mobility of SPCA1 was about 100 kD in good accordance with its theoretical molecular weight (Fig. 3A). However, the SPCA2 signal, obtained with antibodies preabsorbed to remove cross-reactivity with SPCA1, was observed predominantly in membranes of distal colon (Fig. 3B). Also, SPCA2
Fig. 3. Immunoblotting detection of SPCA isoforms in rat tissues. (A) Detection of SPCAs with pan-SPCA-specific antibodies in lysates of various rat tissues. (B) Detection of SPCA2 with absorbed antibodies at a higher electrophoretic resolution in crudely fractionated rat brain and distal colon. 1,2 – brain; 3,4 – distal colon; 1,3 –fractions prepared to enrich plasma membranes; 2,4 – fractions of the remaining membranes. (C) Detection of SPCA2 with absorbed antibodies in rat distal colon fractionated to enrich certain cellular compartments. The strongest signal in the microsomal fraction is marked with box and asterisk. The lower panel shows detection of nongastric H,K-ATPase a-subunit (ang), a marker of apical plasma membranes.
1302
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
may be degraded (70 K band on blots). SPCA2 was detected as a weak doublet in brain and, in colon, upper band in the doublet with apparent molecular weight 103 kD exactly corresponded with the mass predicted from nucleotide sequence, whereas the lower, more intense band has the same electrophoretic mobility as SPCA2. This fact may reflect proteolytic processing or other post-translational modifications of SPCA2. We also performed membrane fractionation studies to observe subcellular localization of endogenously expressed SPCA2. Fractionation of rat colon membranes (Fig. 3C) demonstrates that the strongest SPCA2 signal is detected in plasma membrane enriched fraction. Immunohistochemical detection of SPCA2 using different fixatives and different embedding protocols (Fig. 4 and Supplementary Fig. 4) in rat skin and rat duodenum shows that, indeed, localization of SPCA2 may be different from that in cultured cells. In the skin, we observed perinuclear labeling consistent with localization in intracellular stores Fig. 4 and Supplementary Fig. 4. Quite the contrary, SPCA2 was detected mostly associated with plasma membranes in rat duodenum, thus confirming membrane fractionation experiment. 4. Discussion 4.1. Structure of SPCA2 SPCA2 has all of the typical features of Ca-ATPases. Also, conserved are most residues known to be important for sustaining Mn-transport in PMR1/yeast, such as Q747 (numeration according to rat SPCA2 sequence) [9], or Ca,Mn-dependent phosphorylation in human SPCA1. Alignment of SPCA1 and SPCA2 sequences (Supplementary Fig. 1) shows its close relatedness. However, SPCA2 has a significantly higher frequency of Glu leading to a lower pI (hSPCA2 – 5.5, hSPCA1 – 6.86, rSPCA2 – 5.9, rSPCA1 – 6.52). This makes SPCA2 more similar to PMR1 and SERCA polypeptides (pIs in the range 5.0–5.5). The major negatively charged cluster in SPCA2 is situated N-terminally, coordinates 25–46, being encoded by the boundary of the divergent exons 1 and 2. Additionally, a surface exposed loop in the nucleotide binding domain of SPCA2 (499–503) is also very acidic. Interestingly, an immediately following SPCA2specific residue (Y505) is strongly predicted to be a site of tyrosine phosphorylation. Search for other possible protein modification sites indicate that both SPCAs may be O-glycosylated in a Thr-rich fragment that lies in a lumen-exposed loop between transmembrane helices 7 and 8 (for example, rat SPCA2s has a STPRTTT stretch). Another prominent feature of SPCA2 is a significant bias in composition of hydrophobic amino acids toward Leu. Frequency of Leu is about 20% higher in SPCA2 than in SPCA1. For example, in the first transmembrane region (M1) LIMLLL in SPCA1 corresponds to LILLLL in all SPCA2s. The Leu-rich transmembrane domain of cadherin was implicated in self-association [22] thus it is possible to speculate that this region of SPCAs is involved in interactions with other proteins. Another speculation may be that in SPCAs this region does not adopt the a-helical conformation. Indeed, it is known that in SERCA the N-terminal part of M1 undergoes large lateral movements [23]. 4.2. Evolution of SPCA pumps Fig. 4. Immunohistochemical labeling of rat tissues with antibodies against SPCA2. Tissues were fixed in Carnoy solution, embedded in PEG, and stained with rabbit polyclonal antibodies against SPCA2. (A and C) Black and white images were obtained by inversion of those from red channels for easier viewing. (B and D) Merged images of fluorescent antibodies detecting anti-SPCA2 antibodies (red fluorescence) and nuclei stained in blue with DAPI (B) or in green with SYBR Green (D). Bars: 25 lm (A), 10 lm (D). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
SPCA is conserved through fungi to animals and there is no evidence for SPCA in plants. Genome and ESTs analysis indicates that yeast and most animals possess only one isoform. Invertebrates (Drosophila melanogaster, Anopheles gambiae, Caenorhabditis elegans) have one SPCA gene and their encoded proteins are slightly more homologous to SPCA1 than to SPCA2. Fish genomes (Fugu rubripes
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
and others) appear to have only one SPCA gene that encodes a protein also more related to SPCA1. On the other hand, several available ESTs of SPCAs in an amphibian (X. laevis) and a bird (Gallus gallus) are easy to separate into two groups with more homology either to SPCA1 or to SPCA2. An apparent phylogenetic tree of SPCA2 proteins is shown in Supplementary Fig. 5. This is an evidence that SPCA2 is present in tetrapods including modern amphibians, reptiles, birds and mammals. Because SPCA2 is expressed in lung/trachea it is reasonable to hypothesize that SPCA2 originated from SPCA1 by a gene duplication at the time acquisition of lungs or the transition from aquatic to terrestrial environments. The absence of SPCA2 in fugu genome, however, does not provide any evidence to discriminate between these two possibilities because aquatic teleosteans are known to loose primitive fish lungs [24]. 4.3. Cellular and subcellular distribution of SPCA2 Here, SPCA2 subcellular localization was found to be dependent on tissue and cell type: predominantly plasma membrane in case of rat intestines, but intracellular in epidermal keratinocytes. Predominantly Golgi-like localization of both SPCA1 and SPCA2 pumps in various cultured cell lines was reported previously [13–15], and confirmed in this report. However, some differences may exist as well: association with lipid rafts is more pronounced in the case of SPCA1 than in SPCA2 [25]. It is not surprising that in terminally differentiated cells of epithelial tissues the situation may be different and the dynamic equilibrium may be shifted to the plasma membrane. Indeed, the detection of SPCA in rat liver membranes fractionated by density centrifugation gives a much stronger signal in plasma membrane than in the Golgi fraction [26]. Importantly, SPCA1 has been demonstrated to be present in milk fat globule [27]. Also, in pancreatic acini, SPCA2 is not concentrated in Golgi, instead, it colocalizes with SERCA [28]. These data suggest that in vivo SPCAs are recycling pumps present in Golgi, secretory vesicles and plasma membranes, as proposed for neuroendocrine cells [29]. This behavior is very common for proteins that were long considered trans-Golgi markers (for example, the Mn-requiring enzyme b-galactosyltransferase [30]). Acknowledgments This work was supported by the Russian Foundation for Basic Research (Grants 10-04-01206 and 11-04-12112), MCB program of the Russian Academy of Sciences and funds from University of Toledo College of Medicine. We thank Drs. A. Kitayama, C. Terasaka, M. Mochii, N. Ueno, T. Shin-I. and Y. Kohara for a Xenopus laevis cDNA clone and Dr. Ronald Mellgren for valuable comments on the manuscript . Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.12.135. References [1] M. Brini, E. Carafoli, Calcium pumps in health and disease, Physiol. Rev. 89 (2009) 1341–1378. [2] D.M. Clarke, G.E. Shull, Molecular cloning and tissue distribution of alternatively spliced mRNAs encoding possible mammalian homologues of the yeast secretory pathway calcium pump, Biochemistry 31 (1992) 7600–7608. [3] G.E. Shull, D.M. Clarke, A.M. Gunteski-Hamblin, CDNA cloning of possible mammalian homologs of the yeast secretory pathway Ca2+-transporting ATPase, Ann. N. Y. Acad. Sci. 671 (1992) 70–80. [4] H.K. Rudolph, A. Antebi, G.R. Fink, C.M. Buckley, T.E. Dorman, J. LeVitre, L.S. Davidow, J.I. Mao, D.T. Moir, The yeast secretory pathway is perturbed by mutations in PMR1, a member of a Ca2+-ATPase family, Cell 58 (1989) 133–145.
1303
[5] A. Antebi, G.R. Fink, The yeast Ca2+-ATPase homologue, PMR1, is required for normal Golgi function and localizes in a novel Golgi-like distribution, Mol. Biol. Cell 3 (1992) 633–654. [6] G. Dürr, J. Strayle, R. Plemper, S. Elbs, S.K. Klee, P. Catty, D.H. Wolf, H.K. Rudolph, The medial-Golgi ion pump PMR1 supplies the yeast secretory pathway with Ca2+ and Mn2+ required for glycosylation, sorting, and endoplasmic reticulum-associated protein degradation, Mol. Biol. Cell 9 (1998) 1149–1162. [7] A. Sorin, G. Rosas, R. Rao, PMR1, a Ca2+-ATPase in yeast Golgi, has properties distinct from sarco/endoplasmic reticulum and plasma membrane calcium pumps, J. Biol. Chem. 272 (1997) 9895–9901. [8] Y. Wei, J. Chen, G. Rosas, D.A. Tompkins, P.A. Holt, R. Rao, Phenotypic screening of mutations in PMR1, the yeast secretory pathway Ca2+/Mn2+-ATPase, reveals residues critical for ion selectivity and transport, J. Biol. Chem. 275 (2000) 23927–23932. [9] D. Mandal, T.B. Woolf, R. Rao, Manganese selectivity of PMR1, the yeast secretory pathway ion pump, is defined by residue Gln783 in transmembrane segment 6, J. Biol. Chem. 275 (2000) 23933–23938. [10] Z. Hu, J.M. Bonifas, J. Beech, G. Bench, T. Shigihara, H. Ogawa, S. Ikeda, T. Mauro, E.H. Epstein Jr., Mutations in ATP2C1, encoding a calcium pump, cause HaileyHailey disease, Nat. Genet. 24 (2000) 61–65. [11] V.-K. Ton, D. Mandal, C. Vahadji, R. Rao, Functional expression in yeast of the human secretory pathway Ca2+, Mn2+-ATPase defective in Hailey-Hailey disease, J. Biol. Chem. 277 (2002) 6422–6427. [12] K. Ishikawa, T. Nagase, M. Suyama, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, O. Ohara, Prediction of the coding sequences of unidentified human genes. X, DNA Res. 5 (1998) 169–176. [13] J. Vanoevelen, L. Dode, K. van Baelen, R.J. Fairclough, L. Missiaen, L. Raeymaekers, F. Wuytack, The secretory pathway Ca2+/Mn2+-ATPase 2 is a Golgi-localized pump with high affinity for Ca2+ ions, J. Biol. Chem. 280 (2005) 22800–22808. [14] L. Dode, J.P. Andersen, J. Vanoevelen, L. Raeymaekers, L. Missiaen, B. Vilsen, F. Wuytack, Dissection of the functional differences between human secretory pathway Ca2+/Mn2+-ATPase (SPCA) 1 and 2 isoenzymes by steady-state and transient kinetic analyses, J. Biol. Chem. 281 (2006) 3182–3189. [15] M. Xiang, D. Mohamalawari, R. Rao, A novel isoform of the secretory pathway Ca2+, Mn2+-ATPase, hSPCA2, has unusual properties and is expressed in the brain, J. Biol. Chem. 280 (2005) 11608–11614. [16] H.M. Faddy, C.E. Smart, R. Xu, G.Y. Lee, P.A. Kenny, M. Feng, R. Rao, M.A. Brown, M.J. Bissell, S.J. Roberts-Thomson, G.R. Monteith, Localization of plasma membrane and secretory calcium pumps in the mammary gland, Biochem. Biophys. Res. Commun. 369 (2008) 977–981. [17] T.A. Reinhardt, J.D. Lippolis, Mammary gland involution is associated with rapid down regulation of major mammary Ca2+-ATPases, Biochem. Biophys. Res. Commun. 378 (2009) 99–102. [18] U. Anantamongkol, H. Takemura, T. Suthiphongchai, N. Krishnamra, Y. Horio, Regulation of Ca2+ mobilization by prolactin in mammary gland cells: possible role of secretory pathway Ca2+-ATPase type 2, Biochem. Biophys. Res. Commun. 352 (2007) 537–542. [19] M. Feng, D.M. Grice, H.M. Faddy, N. Nguyen, S. Leitch, Y. Wang, S. Muend, P.A. Kenny, S. Sukumar, S.J. Roberts-Thomson, G.R. Monteith, R. Rao, Storeindependent activation of Orai1 by SPCA2 in mammary tumors, Cell 143 (2010) 84–98. [20] D.F. Newbury, A.P. Monaco, Genetic advances in the study of speech and language disorders, Neuron 68 (2010) 309–320. [21] A.R. Hatton, V. Subramaniam, A.J. Lopez, Generation of alternative Ultrabithorax isoforms and stepwise removal of a large intron by resplicing at exon-exon junctions, Mol. Cell 2 (1998) 787–796. [22] O. Huber, R. Kemler, D. Langosch, Mutations affecting transmembrane segment interactions impair adhesiveness of E-cadherin, J. Cell Sci. 112 (1999) 4415–4423. [23] C. Toyoshima, H. Nomura, Structural changes in the calcium pump accompanying the dissociation of calcium, Nature 418 (2002) 605–611. [24] S.F. Perry, R.J.A. Wilson, C. Straus, M.B. Harris, J.E. Remmers, Which came first, the lung or the breath? Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 129 (2001) 37–47. [25] S. Baron, P. Vangheluwe, M.R. Sepúlveda, F. Wuytack, L. Raeymaekers, J. Vanoevelen, The secretory pathway Ca2+-ATPase 1 is associated with cholesterol-rich microdomains of human colon adenocarcinoma cells, Biochim. Biophys. Acta 1798 (2010) 1512–1521. [26] R.I. Dmitriev, N.B. Pestov, T.V. Korneenko, E.S. Sorokina, A.A. Kazmin, M.B. Kostina, M.I. Shakhparonov, Liver SPCA, a secretory pathway Ca2+-ATP-ase, is predominantly localized in plasma membrane, Biol. Membr. 20 (2003) 480– 485. [27] T. Reinhardt, A. Filoteo, J. Penniston, R. Horst, Ca2+-ATPase protein expression in mammary tissue, Am. J. Physiol. Cell Physiol. 279 (2000) 1595–1602. [28] V.C. Garside, A.S. Kowalik, C.L. Johnson, D. DiRenzo, S.F. Konieczny, C.L. Pin, MIST1 regulates the pancreatic acinar cell expression of Atp2c2, the gene encoding secretory pathway calcium ATPase 2, Exp. Cell Res. 316 (2010) 2859– 2870. [29] K. Mitchell, P. Pinton, A. Varadi, C. Tacchetti, E.K. Ainscow, T. Pozzan, R. Rizzuto, G.A. Rutter, Dense core secretory vesicles revealed as a dynamic Ca2+ store in neuroendocrine cells with a vesicle-associated membrane protein aequorin chimaera, J. Cell Biol. 155 (2001) 41–51. [30] D.H. Dubois, B.D. Shur, Cell surface b-1,4-galactosyltransferase. A signal transducing receptor? Adv. Exp. Med. Biol. 376 (1995) 105–114.
Contents lists available at SciVerse ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Structural evolution and tissue-specific expression of tetrapod-specific second isoform of secretory pathway Ca2+-ATPase Nikolay B. Pestov a,⇑, Ruslan I. Dmitriev a, Maria B. Kostina a, Tatyana V. Korneenko a,b, Mikhail I. Shakhparonov a, Nikolai N. Modyanov b,⇑ a b
Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow 117871, Russia Department of Physiology and Pharmacology, University of Toledo College of Medicine, 3000 Arlington Ave., Toledo, OH 43614, USA
a r t i c l e
i n f o
Article history: Received 22 December 2011 Available online 3 January 2012 Keywords: Vertebrate evolution Gene duplication Enzyme isoforms Golgi complex Alternative splicing CPCA2 Ca-ATPase
a b s t r a c t Secretory pathway Ca-ATPases are less characterized mammalian calcium pumps than plasma membrane Ca-ATPases and sarco-endoplasmic reticulum Ca-ATPases. Here we report analysis of molecular evolution, alternative splicing, tissue-specific expression and subcellular localization of the second isoform of the secretory pathway Ca-ATPase (SPCA2), the product of the ATP2C2 gene. The primary structure of SPCA2 from rat duodenum deduced from full-length transcript contains 944 amino acid residues, and exhibits 65% sequence identity with known SPCA1. The rat SPCA2 sequence is also highly homologous to putative human protein KIAA0703, however, the latter seems to have an aberrant N-terminus originating from intron 2. The tissue-specificity of SPCA2 expression is different from ubiquitous SPCA1. Rat SPCA2 transcripts were detected predominantly in gastrointestinal tract, lung, trachea, lactating mammary gland, skin and preputial gland. In the newborn pig, the expression profile is very similar with one remarkable exception: porcine bulbourethral gland gave the strongest signal. Upon overexpression in cultured cells, SPCA2 shows an intracellular distribution with remarkable enrichment in Golgi. However, in vivo SPCA2 may be localized in compartments that differ among various tissues: it is intracellular in epidermis, but enriched in plasma membranes of the intestinal epithelium. Analysis of SPCA2 sequences from various vertebrate species argue that ATP2C2 gene radiated from ATP2C1 (encoding SPCA1) during adaptation of tetrapod ancestors to terrestrial habitats. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Mammalian Ca-ATPases, calcium pumps, are classified into three major groups: plasma membrane Ca-ATPases (PMCA), sarco-endoplasmic reticulum Ca-ATPases (SERCA) and secretory pathway Ca-ATPases (SPCA). They have major implications for normal and pathological physiology [1]. The first mammalian SPCA, SPCA1, was discovered in 1992 by Guntesky-Hamblin et al. [2] by molecular cloning from rat stomach and testis [3,4] and was shown to exhibit significant homology to the previously characterized yeast ATPase PMR1 [4]. Up to now, mammalian SPCA remained less studied than SERCA, PMCA, or PMR1, the yeast homolog of SPCA, which is required for normal Golgi function [5]. Since PMR1 mutants survive in a narrow interval of calcium and manganese concentrations [6] it was suggested that PMR1 is also involved in manganese transport. Superexpression of PMR1 allowed the direct demonstration of calcium transport, as well as sensitivities to vanadate, thapsigargin and ⇑ Corresponding authors. Fax: +1 419 383 2871. E-mail addresses: [email protected] (N.B. Pestov), nikolai.modyanov@utoledo. edu (N.N. Modyanov). 0006-291X/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.12.135
cyclopiazonic acid that are although different from those of other known Ca-ATPases [7]. This recombinant PMR1 was also purified and demonstrated directly to be able to translocate not only Ca, but also Mn. Substitution Q783A led to the loss of the ability to recognize Mn whereas Ca transport remained intact [8,9]. Thus it can be concluded that yeast PMR1 is a Golgi-resident (Ca2+, Mn2+)-ATPase, and its major physiological role is to provide glycoprotein processing enzymes with calcium and manganese. SPCAs attracted significant attention after it was found that mutations of Ca-ATPases SERCA2 and SPCA1 are associated with Darier and Hailey-Hailey [10] diseases, respectively. It is especially interesting that, although the latter genes are expressed ubiquitously, the only manifestations of their haploinsufficiencies are skin lesions. Human SPCA1 was expressed in yeast and was shown to have catalytic properties similar to those of PMR1, i.e. it is a Ca, MnATPase [11]. From this one can conclude that SPCA represents a conserved system of Ca handling that exists in all mammalian cells (SPCA1 transcripts were detected in all tissues tested [2,3]). However, in vertebrates the situation turned out to be more complex, because of the existence of a putative second isoform of SPCA that was first found in the course of a large-scale transcriptome study
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
[12]. This novel transcript has been cloned from human brain and named KIAA0703. Also, human genome projects allowed the sequencing of the complete human gene ATP2C2, and its localization on chromosome 16q23. Since then, the catalytic properties of overexpressed human SPCA2 highlighted small but important differences between SPCA pumps. For example, SPCA2 was found to have significantly higher Ca2+ affinity in comparison with SPCA1 [13–15]. Mammary gland contains significant amounts of SPCA2 and, more importantly its expression upregulates sharply during lactation [16], and drops just before mammary gland involution [17]. SPCA2 expression may be regulated by prolactin [18]. The recently reported ability of SPCA2 to interact and activate the Ca-channel Orai1 in mammary tumor cells, argues that SPCA2 may also have transport-independent functions [19]. The first SPCA2 coding sequence was cloned from human brain, and, indeed, it may have important functions in this organ. ATP2C2 is now considered as a candidate gene, mutations in which may be the cause of certain memory and speech disorders (reviewed in [20]). Here we report the sequence and intracellular localization of rat SPCA2, a detailed comparison of SPCA2 and SPCA1 expression in adult rat and newborn piglet tissues, and the phylogenetic relationships of SPCA2 from various vertebrate species. 2. Materials and methods An expanded Materials and Methods section can be found in the online supplement.
1299
resequenced. Xenopus laevis SPCA2 transcript sequence has been obtained by resequencing a partially sequenced cDNA clone. The nucleotide sequences reported here have been submitted to the GenBank™/EBI DATA Bank with accession number AF484685 for Rattus norvegicus SPCA2 and DQ420634.1 for X. laevis SPCA2. Tissue-specific expression of various transcripts was analyzed in rat, human and pig tissues by standard RT-PCR. 2.2. Localization of SPCA2 in cultured cells A full-length rat SPCA2 plasmid with C-terminally fused GFP was constructed and used for transfection followed by confocal fluorescent microscopy. 2.3. Immunochemical methods Recombinant proteins comprising the second cytoplasmic domains of SPCA1 and SPCA2 were expressed in Escherichia coli, purified and used for immunization of rabbits. The antibodies were used either as pan-SPCA probes or they were affinity purified to obtain a SPCA2 specific fraction. Membrane fractions were prepared from rat tissues employing several fractionation protocols and were used for western blotting with the antibodies followed by chemiluminescent detection. For immunohistochemistry, tissue sections were prepared by several different techniques, stained with the anti-SPCA2 antibodies followed by Alexa Fluor-conjugated secondary antibodies, and images collected using a fluorescent microscope. 3. Results
2.1. RT-PCR, cloning and sequencing 3.1. Sequence analysis of ATP2C2 gene and the encoded SPCA2 protein The full-length rat SPCA2 sequence was determined by cloning and sequencing of 30 - and 50 -RACE PCR products from rat duodenum. Full-length SPCA2 was cloned by long range RT-PCR and
We have successfully reconstructed the full-length rat SPCA2 transcript from duodenum using RACE PCR. The correctness of
Fig. 1. RT-PCR analysis SPCA1 and SPCA2 expression in rat and pig tissues. (A) Adult rat. (B) Newborn piglet. RT-PCR products from 0.05 lg total RNA were electrophoresed and stained with ethidium bromide. The numbers of cycles shown at the right were chosen for each isoform to compare expression levels in different tissues. The band intensities may not necessarily reflect relative contents of different isoforms in the same tissue.
1300
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
the assembled sequence has been confirmed by amplification of the full-length ORF. Homology search of the GenBank database against the assembled rat SPCA2 mRNA and protein indicates that human ATP2C2 gene encoding SPCA2 spans 95.4 kbp and is composed of 27 exons. Intron sizes vary from 82 bp (intron 21) to 29.8 kbp (intron 1). Exon 1 appears to include all of the 50 -UTR and exon 27 – all of the 30 -UTR. The mouse gene is shorter (56.5 kb) due to compacted introns (for example, intron 1 is 17.5 kbp). Positions of the exons with respect to the amino acid sequence are well conserved. Human and mouse genes are located on chromosomes 16q23 and 8, respectively. Exon–intron structures of ATP2C1 and ATP2C2 are similar, human ATP1C1 spans 107.5 kbp, also with a very long intron 1 (35.6 kbp). There are 27 exons altogether (28 exons for an alternatively spliced variant). Positions of the exons with respect to amino acid sequences are well conserved with the exception of the boundary between divergent exons 1 and 2. The amino acid sequence of SPCA2 deduced from the sequenced rat duodenum cDNA contains 944 residues. Predicted human SPCA2 protein contains 946 residues. Sequence alignment of rat and X. laevis SPCA2 with rat SPCA1 is shown in Supplementary Fig. 1. SPCA2 features a structure similar to other Ca pumps that fits to the conventional 10 transmembrane helices structure with large cytoplasmic domains. Rat SPCA2 has 85.0% residues identical to human SPCA2, 65.1% – to rat SPCA1, 43.9% - to yeast PMR1, 24.6% – to SERCA2. Especially divergent are N-terminal fragments encoded by exons 1 and 2 (Supplementary Fig. 2). Significantly, SPCA2 has a longer N-terminus than SPCA1 (mostly due to enlargement of the coding part in exon 1) and, as a result, has a somewhat larger molecular weight, 103 kDa, compared to 100 kDa of SPCA1.
3.2. Analysis of the 50 -end of human SPCA2 mRNA The previously reported structure of human SPCA2 (KIAA0703) [15] has an N-terminus without homology to any known protein. Similarity search indicates that its 50 -terminus originates from the 30 -terminus of intron 2. To clarify this issue, we have made an amplification with forward primers complementary to the 50 ends of KIAA0703 and to our theoretically predicted ‘‘canonic’’ 50 end (at translation initiation start) whereas the backward primer was complementary to exon 3 (Supplementary Fig. 3). All primers gave positive results with available samples of human cDNA (intestines and brain) although the ‘‘canonic’’ variant has an apparently higher level than KIAA0703. This indicates that both sequences exist in the transcriptome. However, considering the absence of homology between the N-terminus of KIAA0703 and any other Ca-ATPase, it is reasonable to suggest that KIAA0703 is a result of intron retention or transcription from an intron promoter. We obtained retentions of intron 5 and a small part of intron 1 in rat cDNA. Overrepresentation of SPCA2 variants with introns retained during molecular cloning is an interesting artifact. Alternatively, one can speculate that most of SPCA2 pre-mRNA is normally underspliced in vivo. This may be especially true for introns with complex splicing. Large introns may be removed stepwise through transient retention of internal exons [21], and this seems to be the case with exon 1 of rat ATP2C2. 3.3. Tissue distribution of SPCA2 transcripts in comparison with SPCA1 Analysis of tissue-specific expression of rat SPCAs is shown in Fig. 1A. In accordance with previous results of Guntesky-Hamblin et al. [2,3], SPCA1 is expressed ubiquitously. However, its level is
Fig. 2. Confocal imaging of subcellular distribution of SPCA2 tagged with green fluorescent protein in cultured cells. Mouse C2 myoblasts were transfected with a plasmid encoding the SPCA2-gfp chimera. (A and B) Fluorescence recorded at low (A) and high (B) detector sensitivities. (C) Phase contrast image of the cells. (D) A and C images merged.
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
variable, being apparently higher in kidney, testis, adrenal and mammary gland (without a strong effect of pregnancy and lactation), whereas some tissues, e.g. intestines, are relatively poor in SPCA1 transcripts. SPCA2 is very different – the expression level of rat SPCA2 varies between tissues to a great extent. It is the most abundant in lactating mammary gland, trachea, lung, intestines (especially in colon), skin and preputial gland. Trace levels of SPCA2 transcripts (at high number of cycles) can be detected in almost all tissues, although the expression in liver, adrenal, brain and skeletal muscle appears to be negligible. Lactating mammary gland has a much stronger signal than prelactating and quiescent glands from pregnant or virgin females. Another example of tissue-specific expression pattern has been obtained with the newborn pig (Fig. 1B). A larger size of pigs gave the possibility to study more tissues including several small organs. Pig SPCA1 is expressed almost ubiquitously, with the highest levels in choroid plexus, retina, adrenal, skin and salivary gland. Blood cells seem to have only traces of SPCA1. The tissue-specificity profile of pig SPCA2 is very similar to that in the rat. Strong signals are observed in lung and gastrointestinal tract (higher in large intestine) whereas tissues like liver and striated muscles have negligible levels of SPCA2. Interestingly, significant expression is observed in parathyroid gland. There are also some dissimilarities that can be attributed to either species- or age-specific differences, such as expression of SPCA2 in testis, ovary and adrenal (more pronounced in the piglet) and skin (higher in the rat). One difference is, however, very remarkable: pig SPCA2 shows the highest level in bulbourethral gland whereas this organ in the rat has a minute level of the transcript.
1301
Analysis of rat, mouse and human ESTs in GeneBank (not shown) corresponds very well with our RT-PCR analysis: rodent SPCA2 is the most abundant in lactating mammary gland and colon. Human data are not so similar: human prostate, unlike rodent prostate, may also contain SPCA2 indicating species-specific differences of SPCA2 usage in various exocrine glands. Note that no ESTs of SPCA2 can be found in brain or kidney, in line with our analyses but contrary to the results of KIAA0703 detection in the brain and kidney [12]. Also interesting is the presence of many ESTs in tumors of tissues where SPCA2 is expressed normally (colon, lung and mammary gland) [13,15]. 3.4. Cellular and subcellular localization of mammalian SPCA proteins For subcellular localization studies we constructed plasmid DNA which encodes full-length rat SPCA2 fused with green fluorescent protein (GFP) at the C-terminus. After transient transfection of mouse C2C12 myoblasts Golgi-like distribution of chimeras was observed: (Fig. 2). Similar results were obtained for all tested cell lines (human adenocarcinoma HT-29 cells, and CHO cells, results not shown). Using pan-SPCA polyclonal antibodies we detected SPCA1 and SPCA2 in membranes of rat tissues. SPCA1 was detected in all tissues tested at comparable levels except kidney, colon and heart, where its content was higher. Electrophoretic mobility of SPCA1 was about 100 kD in good accordance with its theoretical molecular weight (Fig. 3A). However, the SPCA2 signal, obtained with antibodies preabsorbed to remove cross-reactivity with SPCA1, was observed predominantly in membranes of distal colon (Fig. 3B). Also, SPCA2
Fig. 3. Immunoblotting detection of SPCA isoforms in rat tissues. (A) Detection of SPCAs with pan-SPCA-specific antibodies in lysates of various rat tissues. (B) Detection of SPCA2 with absorbed antibodies at a higher electrophoretic resolution in crudely fractionated rat brain and distal colon. 1,2 – brain; 3,4 – distal colon; 1,3 –fractions prepared to enrich plasma membranes; 2,4 – fractions of the remaining membranes. (C) Detection of SPCA2 with absorbed antibodies in rat distal colon fractionated to enrich certain cellular compartments. The strongest signal in the microsomal fraction is marked with box and asterisk. The lower panel shows detection of nongastric H,K-ATPase a-subunit (ang), a marker of apical plasma membranes.
1302
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
may be degraded (70 K band on blots). SPCA2 was detected as a weak doublet in brain and, in colon, upper band in the doublet with apparent molecular weight 103 kD exactly corresponded with the mass predicted from nucleotide sequence, whereas the lower, more intense band has the same electrophoretic mobility as SPCA2. This fact may reflect proteolytic processing or other post-translational modifications of SPCA2. We also performed membrane fractionation studies to observe subcellular localization of endogenously expressed SPCA2. Fractionation of rat colon membranes (Fig. 3C) demonstrates that the strongest SPCA2 signal is detected in plasma membrane enriched fraction. Immunohistochemical detection of SPCA2 using different fixatives and different embedding protocols (Fig. 4 and Supplementary Fig. 4) in rat skin and rat duodenum shows that, indeed, localization of SPCA2 may be different from that in cultured cells. In the skin, we observed perinuclear labeling consistent with localization in intracellular stores Fig. 4 and Supplementary Fig. 4. Quite the contrary, SPCA2 was detected mostly associated with plasma membranes in rat duodenum, thus confirming membrane fractionation experiment. 4. Discussion 4.1. Structure of SPCA2 SPCA2 has all of the typical features of Ca-ATPases. Also, conserved are most residues known to be important for sustaining Mn-transport in PMR1/yeast, such as Q747 (numeration according to rat SPCA2 sequence) [9], or Ca,Mn-dependent phosphorylation in human SPCA1. Alignment of SPCA1 and SPCA2 sequences (Supplementary Fig. 1) shows its close relatedness. However, SPCA2 has a significantly higher frequency of Glu leading to a lower pI (hSPCA2 – 5.5, hSPCA1 – 6.86, rSPCA2 – 5.9, rSPCA1 – 6.52). This makes SPCA2 more similar to PMR1 and SERCA polypeptides (pIs in the range 5.0–5.5). The major negatively charged cluster in SPCA2 is situated N-terminally, coordinates 25–46, being encoded by the boundary of the divergent exons 1 and 2. Additionally, a surface exposed loop in the nucleotide binding domain of SPCA2 (499–503) is also very acidic. Interestingly, an immediately following SPCA2specific residue (Y505) is strongly predicted to be a site of tyrosine phosphorylation. Search for other possible protein modification sites indicate that both SPCAs may be O-glycosylated in a Thr-rich fragment that lies in a lumen-exposed loop between transmembrane helices 7 and 8 (for example, rat SPCA2s has a STPRTTT stretch). Another prominent feature of SPCA2 is a significant bias in composition of hydrophobic amino acids toward Leu. Frequency of Leu is about 20% higher in SPCA2 than in SPCA1. For example, in the first transmembrane region (M1) LIMLLL in SPCA1 corresponds to LILLLL in all SPCA2s. The Leu-rich transmembrane domain of cadherin was implicated in self-association [22] thus it is possible to speculate that this region of SPCAs is involved in interactions with other proteins. Another speculation may be that in SPCAs this region does not adopt the a-helical conformation. Indeed, it is known that in SERCA the N-terminal part of M1 undergoes large lateral movements [23]. 4.2. Evolution of SPCA pumps Fig. 4. Immunohistochemical labeling of rat tissues with antibodies against SPCA2. Tissues were fixed in Carnoy solution, embedded in PEG, and stained with rabbit polyclonal antibodies against SPCA2. (A and C) Black and white images were obtained by inversion of those from red channels for easier viewing. (B and D) Merged images of fluorescent antibodies detecting anti-SPCA2 antibodies (red fluorescence) and nuclei stained in blue with DAPI (B) or in green with SYBR Green (D). Bars: 25 lm (A), 10 lm (D). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
SPCA is conserved through fungi to animals and there is no evidence for SPCA in plants. Genome and ESTs analysis indicates that yeast and most animals possess only one isoform. Invertebrates (Drosophila melanogaster, Anopheles gambiae, Caenorhabditis elegans) have one SPCA gene and their encoded proteins are slightly more homologous to SPCA1 than to SPCA2. Fish genomes (Fugu rubripes
N.B. Pestov et al. / Biochemical and Biophysical Research Communications 417 (2012) 1298–1303
and others) appear to have only one SPCA gene that encodes a protein also more related to SPCA1. On the other hand, several available ESTs of SPCAs in an amphibian (X. laevis) and a bird (Gallus gallus) are easy to separate into two groups with more homology either to SPCA1 or to SPCA2. An apparent phylogenetic tree of SPCA2 proteins is shown in Supplementary Fig. 5. This is an evidence that SPCA2 is present in tetrapods including modern amphibians, reptiles, birds and mammals. Because SPCA2 is expressed in lung/trachea it is reasonable to hypothesize that SPCA2 originated from SPCA1 by a gene duplication at the time acquisition of lungs or the transition from aquatic to terrestrial environments. The absence of SPCA2 in fugu genome, however, does not provide any evidence to discriminate between these two possibilities because aquatic teleosteans are known to loose primitive fish lungs [24]. 4.3. Cellular and subcellular distribution of SPCA2 Here, SPCA2 subcellular localization was found to be dependent on tissue and cell type: predominantly plasma membrane in case of rat intestines, but intracellular in epidermal keratinocytes. Predominantly Golgi-like localization of both SPCA1 and SPCA2 pumps in various cultured cell lines was reported previously [13–15], and confirmed in this report. However, some differences may exist as well: association with lipid rafts is more pronounced in the case of SPCA1 than in SPCA2 [25]. It is not surprising that in terminally differentiated cells of epithelial tissues the situation may be different and the dynamic equilibrium may be shifted to the plasma membrane. Indeed, the detection of SPCA in rat liver membranes fractionated by density centrifugation gives a much stronger signal in plasma membrane than in the Golgi fraction [26]. Importantly, SPCA1 has been demonstrated to be present in milk fat globule [27]. Also, in pancreatic acini, SPCA2 is not concentrated in Golgi, instead, it colocalizes with SERCA [28]. These data suggest that in vivo SPCAs are recycling pumps present in Golgi, secretory vesicles and plasma membranes, as proposed for neuroendocrine cells [29]. This behavior is very common for proteins that were long considered trans-Golgi markers (for example, the Mn-requiring enzyme b-galactosyltransferase [30]). Acknowledgments This work was supported by the Russian Foundation for Basic Research (Grants 10-04-01206 and 11-04-12112), MCB program of the Russian Academy of Sciences and funds from University of Toledo College of Medicine. We thank Drs. A. Kitayama, C. Terasaka, M. Mochii, N. Ueno, T. Shin-I. and Y. Kohara for a Xenopus laevis cDNA clone and Dr. Ronald Mellgren for valuable comments on the manuscript . Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.12.135. References [1] M. Brini, E. Carafoli, Calcium pumps in health and disease, Physiol. Rev. 89 (2009) 1341–1378. [2] D.M. Clarke, G.E. Shull, Molecular cloning and tissue distribution of alternatively spliced mRNAs encoding possible mammalian homologues of the yeast secretory pathway calcium pump, Biochemistry 31 (1992) 7600–7608. [3] G.E. Shull, D.M. Clarke, A.M. Gunteski-Hamblin, CDNA cloning of possible mammalian homologs of the yeast secretory pathway Ca2+-transporting ATPase, Ann. N. Y. Acad. Sci. 671 (1992) 70–80. [4] H.K. Rudolph, A. Antebi, G.R. Fink, C.M. Buckley, T.E. Dorman, J. LeVitre, L.S. Davidow, J.I. Mao, D.T. Moir, The yeast secretory pathway is perturbed by mutations in PMR1, a member of a Ca2+-ATPase family, Cell 58 (1989) 133–145.
1303
[5] A. Antebi, G.R. Fink, The yeast Ca2+-ATPase homologue, PMR1, is required for normal Golgi function and localizes in a novel Golgi-like distribution, Mol. Biol. Cell 3 (1992) 633–654. [6] G. Dürr, J. Strayle, R. Plemper, S. Elbs, S.K. Klee, P. Catty, D.H. Wolf, H.K. Rudolph, The medial-Golgi ion pump PMR1 supplies the yeast secretory pathway with Ca2+ and Mn2+ required for glycosylation, sorting, and endoplasmic reticulum-associated protein degradation, Mol. Biol. Cell 9 (1998) 1149–1162. [7] A. Sorin, G. Rosas, R. Rao, PMR1, a Ca2+-ATPase in yeast Golgi, has properties distinct from sarco/endoplasmic reticulum and plasma membrane calcium pumps, J. Biol. Chem. 272 (1997) 9895–9901. [8] Y. Wei, J. Chen, G. Rosas, D.A. Tompkins, P.A. Holt, R. Rao, Phenotypic screening of mutations in PMR1, the yeast secretory pathway Ca2+/Mn2+-ATPase, reveals residues critical for ion selectivity and transport, J. Biol. Chem. 275 (2000) 23927–23932. [9] D. Mandal, T.B. Woolf, R. Rao, Manganese selectivity of PMR1, the yeast secretory pathway ion pump, is defined by residue Gln783 in transmembrane segment 6, J. Biol. Chem. 275 (2000) 23933–23938. [10] Z. Hu, J.M. Bonifas, J. Beech, G. Bench, T. Shigihara, H. Ogawa, S. Ikeda, T. Mauro, E.H. Epstein Jr., Mutations in ATP2C1, encoding a calcium pump, cause HaileyHailey disease, Nat. Genet. 24 (2000) 61–65. [11] V.-K. Ton, D. Mandal, C. Vahadji, R. Rao, Functional expression in yeast of the human secretory pathway Ca2+, Mn2+-ATPase defective in Hailey-Hailey disease, J. Biol. Chem. 277 (2002) 6422–6427. [12] K. Ishikawa, T. Nagase, M. Suyama, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, O. Ohara, Prediction of the coding sequences of unidentified human genes. X, DNA Res. 5 (1998) 169–176. [13] J. Vanoevelen, L. Dode, K. van Baelen, R.J. Fairclough, L. Missiaen, L. Raeymaekers, F. Wuytack, The secretory pathway Ca2+/Mn2+-ATPase 2 is a Golgi-localized pump with high affinity for Ca2+ ions, J. Biol. Chem. 280 (2005) 22800–22808. [14] L. Dode, J.P. Andersen, J. Vanoevelen, L. Raeymaekers, L. Missiaen, B. Vilsen, F. Wuytack, Dissection of the functional differences between human secretory pathway Ca2+/Mn2+-ATPase (SPCA) 1 and 2 isoenzymes by steady-state and transient kinetic analyses, J. Biol. Chem. 281 (2006) 3182–3189. [15] M. Xiang, D. Mohamalawari, R. Rao, A novel isoform of the secretory pathway Ca2+, Mn2+-ATPase, hSPCA2, has unusual properties and is expressed in the brain, J. Biol. Chem. 280 (2005) 11608–11614. [16] H.M. Faddy, C.E. Smart, R. Xu, G.Y. Lee, P.A. Kenny, M. Feng, R. Rao, M.A. Brown, M.J. Bissell, S.J. Roberts-Thomson, G.R. Monteith, Localization of plasma membrane and secretory calcium pumps in the mammary gland, Biochem. Biophys. Res. Commun. 369 (2008) 977–981. [17] T.A. Reinhardt, J.D. Lippolis, Mammary gland involution is associated with rapid down regulation of major mammary Ca2+-ATPases, Biochem. Biophys. Res. Commun. 378 (2009) 99–102. [18] U. Anantamongkol, H. Takemura, T. Suthiphongchai, N. Krishnamra, Y. Horio, Regulation of Ca2+ mobilization by prolactin in mammary gland cells: possible role of secretory pathway Ca2+-ATPase type 2, Biochem. Biophys. Res. Commun. 352 (2007) 537–542. [19] M. Feng, D.M. Grice, H.M. Faddy, N. Nguyen, S. Leitch, Y. Wang, S. Muend, P.A. Kenny, S. Sukumar, S.J. Roberts-Thomson, G.R. Monteith, R. Rao, Storeindependent activation of Orai1 by SPCA2 in mammary tumors, Cell 143 (2010) 84–98. [20] D.F. Newbury, A.P. Monaco, Genetic advances in the study of speech and language disorders, Neuron 68 (2010) 309–320. [21] A.R. Hatton, V. Subramaniam, A.J. Lopez, Generation of alternative Ultrabithorax isoforms and stepwise removal of a large intron by resplicing at exon-exon junctions, Mol. Cell 2 (1998) 787–796. [22] O. Huber, R. Kemler, D. Langosch, Mutations affecting transmembrane segment interactions impair adhesiveness of E-cadherin, J. Cell Sci. 112 (1999) 4415–4423. [23] C. Toyoshima, H. Nomura, Structural changes in the calcium pump accompanying the dissociation of calcium, Nature 418 (2002) 605–611. [24] S.F. Perry, R.J.A. Wilson, C. Straus, M.B. Harris, J.E. Remmers, Which came first, the lung or the breath? Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 129 (2001) 37–47. [25] S. Baron, P. Vangheluwe, M.R. Sepúlveda, F. Wuytack, L. Raeymaekers, J. Vanoevelen, The secretory pathway Ca2+-ATPase 1 is associated with cholesterol-rich microdomains of human colon adenocarcinoma cells, Biochim. Biophys. Acta 1798 (2010) 1512–1521. [26] R.I. Dmitriev, N.B. Pestov, T.V. Korneenko, E.S. Sorokina, A.A. Kazmin, M.B. Kostina, M.I. Shakhparonov, Liver SPCA, a secretory pathway Ca2+-ATP-ase, is predominantly localized in plasma membrane, Biol. Membr. 20 (2003) 480– 485. [27] T. Reinhardt, A. Filoteo, J. Penniston, R. Horst, Ca2+-ATPase protein expression in mammary tissue, Am. J. Physiol. Cell Physiol. 279 (2000) 1595–1602. [28] V.C. Garside, A.S. Kowalik, C.L. Johnson, D. DiRenzo, S.F. Konieczny, C.L. Pin, MIST1 regulates the pancreatic acinar cell expression of Atp2c2, the gene encoding secretory pathway calcium ATPase 2, Exp. Cell Res. 316 (2010) 2859– 2870. [29] K. Mitchell, P. Pinton, A. Varadi, C. Tacchetti, E.K. Ainscow, T. Pozzan, R. Rizzuto, G.A. Rutter, Dense core secretory vesicles revealed as a dynamic Ca2+ store in neuroendocrine cells with a vesicle-associated membrane protein aequorin chimaera, J. Cell Biol. 155 (2001) 41–51. [30] D.H. Dubois, B.D. Shur, Cell surface b-1,4-galactosyltransferase. A signal transducing receptor? Adv. Exp. Med. Biol. 376 (1995) 105–114.