View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Genomics 88 (2006) 791–800 www.elsevier.com/locate/ygeno

Identification and characterization of five-transmembrane isoforms of human vasoactive intestinal and pituitary adenylate cyclase-activating polypeptide receptors

Payman Baradar Bokaei a,b,1, Xue-Zhong Ma b,c,1, Bartosz Byczynski a,b, Jeremy Keller a,b, ⁎ Darinka Sakac b, Soad Fahim b, Donald R. Branch a,b,c,d,e,

a Institute of Medical Science, University of Toronto, Toronto, ON, Canada M5S 1A8 b Research and Development, Canadian Blood Services, Toronto, ON, Canada M5G 2M1 c Department of Medicine, University of Toronto, Toronto, ON, Canada M5S 1A8 d Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada M5S 1A8 e Division of Cell and Molecular Biology, Toronto General Research Institute, 67 College Street, Toronto, ON, Canada M5G 2M1 Received 7 April 2006; accepted 18 July 2006 Available online 24 August 2006

Abstract

The seven-transmembrane (7TM) G--coupled neuroendocrine receptors VPAC1 (HGNC approved symbol VIPR1) and VPAC2 (HGNC approved gene symbol VIPR2) are expressed in different tissues and involved in the regulation of important biological functions. We now report the identification and characterization of novel five-transmembrane(5TM) forms of both human VPAC1 and human VPAC2. These alternatively spliced variant mRNAs result from the skipping of exons 10/11, spanning the third intracellular loop, the fourth extracellular loop, and the transmembrane regions 6 and 7, producing in-frame 5TM receptors predicted to lack a G-protein-binding motif. RT-PCR showed that these 5TM receptors are differentially expressed in transformed and normal cells. Translation of the 5TM protein was demonstrated by transfection and expression in CHO cells. Following stimulation, differential signaling of the 7TM versus 5TM forms was shown both for the activation of adenylate cyclase and for tyrosine phosphorylation. The identification of these splice variants in various cells and their expression and differential signal transduction compared to the 7TM form suggest that these novel receptors have biological relevance. © 2006 Elsevier Inc. All rights reserved.

Keywords: Vasoactive intestinal peptide receptors; Splice variants; Five-transmembrane; VIPR1; VPAC1; VIPR2; VPAC2; GPCR

The human vasoactive intestinal peptide and pituitary the HT29 human colonic adenocarcinoma cell line library [4]. adenylate cyclase-activating polypeptide family The VPAC1 receptor comprises 457 amino acids and its gene consists of three receptors, PAC1 (HGNC approved gene spans 22 kb, which is composed of 13 exons ranging in size symbol ADCYAP1R1), VPAC1, and VPAC2 [1,2]. These from 42 to 1400 bp [4]. VPAC1 is evolutionarily conserved receptors belong to the class II subfamily of the seven- and structurally characterized by its 7TM domains with an transmembrane (7TM) G-protein-coupled receptor (GPCR) extracellular N-terminus and a cytoplasmic C-terminus [5]. superfamily [1,2]. VPAC1 was first cloned from a rat VPAC1 is found in certain regions of the brain, notably the cDNA library by cross-hybridization with a receptor piriform cortex, the septal nucleus, and the hippocampal [3]. The human VPAC1 cDNA was first characterized from formation [6,7]. VPAC1 is also a predominant receptor expressed on other cells and tissues, particularly on resting T lymphocytes and peripheral blood monocytes [8–10].In ⁎ Corresponding author. Division of Cell and Molecular Biology, Toronto addition, studies have shown that tumor-derived breast and General Research Institute, 67 College Street, Toronto, 67 College St., ON, Canada M5G 2M1. Fax: +1 416 974 9575. intestinal cell lines, as well as some malignant T E-mail address: [email protected] (D.R. Branch). lymphoblastic cell lines, exhibit VPAC1 mRNA in large 1 These authors contributed equally to this work. amounts [10–12]. Furthermore, our laboratory has recently

0888-7543/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygeno.2006.07.008 792 P.B. Bokaei et al. / Genomics 88 (2006) 791–800 shown that VPAC1 binds to human immunodeficiency virus VPAC1 stimulation, which causes an increase in HIV-1 type 1 (HIV-1) to transduce an intracellular signal that productive infection [13]. results in facilitation of HIV-1 productive infection [13]. In this study, we describe the identification, isolation, and The VPAC2 receptor was initially cloned from a rat pituitary characterization of two novel human 5TM VPAC receptor cDNA library [14] and subsequently from a human placenta isoforms, one of VPAC1 and one of VPAC2. This is the first cDNA library [15]. VPAC2 is present in human skeletal muscle, description of these isoforms in human cells. These splice- intestine, and brain, especially in the thalamus and cerebral variant VPAC isoforms are differentially expressed in normal cortex [16,17]. VPAC2 is encoded by 13 exons and the human human primary peripheral blood mononuclear and cord blood gene spans 117 kb [15]. VPAC2 is expressed more selectively cells and in malignant hematopoietic- and epithelial-derived than is VPAC1. VPAC2 is found at very low levels on resting cancer cell lines. Using transfection, we confirmed that the 5TM monocytes and T cells but its expression is increased after isoform, predicted by sequence analysis to lack a G-protein- activation of these cells [10]. binding site, can be translated and has significantly decreased Significant differences exist between VPAC1 and VPAC2, ability to increase intracellular cAMP in response to vasoactive both in structure and in function [18,19]. Indeed, there are a intestinal peptide (VIP) and secretin compared to cells number of reports indicating opposing functions for these two transfected to express the 7TM receptor. However, despite a receptors [20,21]. Recently, we have shown that VPAC2 lack of G protein binding and adenylate cyclase activation, stimulation by specific dramatically inhibits HIV-1 stimulation of the 5TM isoform was, nonetheless, able to acti- productive infection [22]. The opposite effect is observed after vate protein tyrosine kinase activity that showed a differential

Fig. 1. Sequence alignment of seven- and five-transmembrane VPAC1 and VPAC2 receptors. P.B. Bokaei et al. / Genomics 88 (2006) 791–800 793 pattern of resulting phosphotyrosine-containing com- cells. A strategy using specific primers designed to amplify pared to the 7TM receptor. These results suggest that both both 5TM and 7TM VPAC1 isoforms (Fig. 3A) was isoforms of VPAC1 and VPAC2 would have biological confirmed using Hut78 cells (Fig. 3B). Using this RT-PCR relevance. demonstrated differential patterns of mRNA expression in different cells. In normal, nonactivated PBMCs, and malig- Results nant cell lines Hut78, Raji, and Colo205, expression of both isoforms was evident. However, in normal cord blood Sequence alignment of VPAC1 and VPAC2 isoforms PBMCs and malignant CaCo2 and transformed 293 cells, only the 7TM form of VPAC1 was found. Furthermore, the T In our effort to clone full-length human VPAC1 from cell leukemia-derived Kit225 and ACHN -derived cells normal peripheral blood mononuclear cells (PBMCs) and do not express either VPAC1 isoform. VPAC2 from the human Sup-T1 cell line, we unexpectedly A strategy using nested RT-PCR (Fig. 4A) was optimized identified splice variant receptor transcripts. These variant and used to examine the expression of 5TM and 7TM VPAC2 receptors of VPAC1 and VPAC2, bearing 264- and 222-bp receptor isoforms. The ability of the RT-PCR to amplify both deletions, respectively, showed an open reading frame that transcripts was confirmed using Sup-T1 cells (Fig. 4B). The would encode receptors lacking, respectively, 88 and 74 results reveal limited expression of the 5TM isoform but also amino acid residues (Fig. 1). Interestingly, analysis of the differential expression of the VPAC2 mRNA (Fig. 4C). Normal deduced amino acid sequences revealed that the truncated nonactivated PBMCs express low levels of 7TM VPAC2 VPAC1 and VPAC2 isoforms lack the regions in the third mRNA as previously published [9,10]; however, these cells do intracellular loop, the fourth extracellular loop, and TM6–7 not express the 5TM isoform (Fig. 4C). Following activation (Figs. 1 and 2A). These events result in with PHA/IL-2 or with PHA alone, the 7TM isoform of VPAC2 transcripts that encode a putative five-transmembrane domain shows increased expression and the 5TM isoform is now also form (Fig. 2B). expressed; indeed, the only other cells tested that expressed 5TM VPAC2 transcripts were Sup-T1 cells, the original cells VPAC1 and VPAC2 expression in normal and malignant cells from which the two isoforms were cloned. Neither Daudi nor THP-1 cells express either transcript. The other cell lines It has been reported that 7TM VPAC1 and VPAC2 are, or examined expressed only the 7TM VPAC2. These results were can be induced to be, expressed in most human tissues confirmed by Southern blot, which showed hybridization of a [23,24]. As the identification of human 5TM isoforms has full-length VPAC2 radioactive probe to only those bands never been reported, we examined the expression of 7TM corresponding to the 5TM and 7TM positive (+ve) controls compared to 5TM VPAC1 and VPAC2 isoforms in different (Fig. 4D).

Fig. 2. (A) Deletions of 7TM VPAC1 and VPAC2 nucleotides that result in 5TM truncated isoforms. The deleted residues in the mutants are indicated by dashes. (B) Schematic representation of the 7TM and 5TM VPAC1 and VPAC2 receptors. The deletion in the third intracellular loop, the fourth extracellular loop, and TM6–7 results in the 5TM truncated isoforms. Light shading indicates deletions. The arrows indicate the connecting point of TM5 to the C-terminus. 794 P.B. Bokaei et al. / Genomics 88 (2006) 791–800

Fig. 3. Differential expression of 7TM and 5TM VPAC1 transcripts in normal and malignant cells. (A) Schematic representation of the structure of the full-length VPAC1 mRNA showing the locations of specific primer pairs for amplification of the VPAC1 cDNA (see Materials and methods). The shaded area represents the deletion that creates the 5TM isoform. (B) RT-PCR result using the primer pair indicated in A to amplify both 7TM and 5TM VPAC1 isoforms from Hut78 cells. (C) Detection of VPAC1 transcripts in different cells. Results are from normal human PBMCs, normal cord blood mononuclear cells (Cord Blood), epithelial cancer- derived cell lines (Colo205, CaCo2, ACHN), a transformed epithelial cell line (293), and hematopoietic cancer-derived cell lines (Raji, Hut78, and Kit225). Positive controls consisted of Hut78 and cDNA for the cloned either 5TM or 7TM VPAC1. Negative control was water.

Protein expression of 5TM isoform lead to the generation of a signaling pathway(s) that ultimately leads to a specific cell function. To examine whether 5TM Because it has been reported that certain mutations in the VPAC isoforms retain functionality, particularly as they are GPCR may result in poor protein expression and/or impair the predicted by sequence analyses to lack a G-protein-binding site, stability of the receptor on the membrane [25,26],itisimportantto we compared the abilities of the 7TM and 5TM isoforms to show that the novel alternative splicing we have documented induce increased production of cAMP and also examined their results in protein production. Due to a lack of potent antibodies to ability to activate tyrosine kinases. The response to VIP caused VPAC receptors, to address this issue, we used transfection of an approximately fourfold increase in cAMP production in CHO cells with VPAC1 5TM and 7TM cDNA in pcDNA3.1, transfected CHO cells with 7TM VPAC1 compared to empty- which contains an Xpress tag so as to be able to monitor the vector control. Response to VIP by 5TM-transfected cells did protein expression by immunoprecipitation of metabolically not result in an increase in cAMP (Fig. 6A). Receptor-mediated labeled protein. Fig. 5 shows our results that indicate the trans- cellular responses to secretin were also tested and revealed lation of these isoforms to the predicted molecular sized proteins results similar to those obtained for VIP (Fig. 6A). and support previous reports of production of other cell surface A series of experiments was next performed to evaluate proteins having mutations encoding a truncated receptor [27,28]. tyrosine kinase activation following VIP or secretin stimulation of 7TM and 5TM VPAC1 isoforms. In this regard, transfected Functionality of VPAC isoforms CHO cells were serum starved overnight and then stimulated with 50 μM VIP or secretin for 30 min at 37°C. Cells were GPCR responses are critically dependent on the functional washed with PBS and lysed in radioimmunoprecipitation assay interaction between the receptor and its ligand and the (RIPA) lysis buffer. Although no differences in phosphorylation subsequent G-protein-dependent activation. These interactions between the 5TM- and the 7TM-transfected cells was apparent P.B. Bokaei et al. / Genomics 88 (2006) 791–800 795

Fig. 4. Differential expression of 7TM and 5TM VPAC2 transcripts in normal and malignant cells. (A) Schematic representation of the structure of the full-length VPAC2 mRNA showing the locations of specific primer pairs for first- and second-round amplification of the VPAC2 cDNA. The shaded area represents the deletion that creates the 5TM isoform (see Materials and methods). (B) RT-PCR result using the primer pairs indicated in (A) to amplify both 7TM and 5TM VPAC2 isoforms from Sup-T1 cells. (C) Detection of VPAC2 transcripts in different cells. Results are from normal human PBMCs, a transformed epithelial cell line (293), and hematopoietic cancer-derived cell lines (Sup-T1, Jurkat, Daudi, Raji, and THP-1). Positive controls consisted of Sup-T1 and cDNA for the cloned either 5TM or 7TM VPAC2. Negative control was water. RT- PCR for protein phosphatase-1 (PP-1) was used as a housekeeping gene control for mRNA quality. (D) Southern blot analysis of RT-PCR results from (C). after stimulation with VIP for 30 min, both showed increased these phosphorylated proteins remains to be determined, it can phosphorylation after stimulation compared to empty vector be speculated that these differences in tyrosine phosphorylation control. The lack of differential tyrosine phosphorylation may may lead to generation of a different signaling pathway(s) and be due to some cross-reactivity of the VIP with endogenous that this differential tyrosine phosphorylation may lead to VPAC2 on these cells. In contrast to stimulation with VIP, unique cell activity or function. Further investigations are Western blot analysis of phosphotyrosine-containing proteins necessary to elucidate the specific signaling pathways following after 30 min stimulation with secretin showed differential activation of the different VPAC receptor isoforms. tyrosine kinase activation (Fig. 6B). Proteins of approximately 50 and 60 kDa showed no increased phosphorylation after Discussion secretin stimulation in the 5TM- compared to the 7TM- transfected cells, while proteins of approximate molecular Different isoforms of GPCR have been previously identified sizes of 30 and 38 kDa showed increased phosphorylation for a number of receptors. In one report, identification and compared to 7TM-transfected cells. Although the identity of characterization of a 5TM domain variant isoform of the NTS2 796 P.B. Bokaei et al. / Genomics 88 (2006) 791–800

VPAC2 receptor has been previously described in mouse [34]. This VPAC2 receptor isoform was identified in immune cells and shown to have a deletion in the carboxyl-terminal end of the seventh transmembrane domain. These investigators also showed that wild-type and the deletion variant VPAC2 bound the same amount of 125I-VIP with similar affinity. However, unlike wild-type VPAC2, the deletion variant VPAC2 did not transduce VIP-elicited increases in intracellular concentrations of cAMP. It was concluded that a natural deletion of a part of the last transmembrane domain of mouse VPAC2 abrogates signaling functions. Recently, however, this group has shown a differential signaling pathway for wild-type compared to the deletion variant of VPAC2 [35]. Using production of IL-4, they reported that exogenous VIP increases IL-4 production from wild-type VPAC2 transfectants, but decreased IL-4 production when using the deletion variant VPAC2 transfectants [35]. Data also supported their conclusion that the ratio of expression of these two forms of VPAC2 contributes to the overall agonist Fig. 5. Expression of VPAC1 7TM and 5TM protein isoforms in CHO cells. effect and that these two VPAC2 receptor isoforms have CHO cells transfected with pcDNA3.1 containing cDNA to express either 7TM 35 opposing functions and can compete with each other in the or 5TM VPAC1 were identified by autofluorography after S metabolic production of IL-4. labeling, immunoprecipitation with anti-Xpress antibody, and SDS–PAGE. Lane 1 represents CHO cells transfected with empty pcDNA3.1 vector. Lane 2 In our study, we have detected two naturally occurring shows CHO cells transfected with full-length cDNA VPAC1 5TM isoform in human 5TM isoforms of VPAC receptors that are differentially pcDNA 3.1 (39 kDa). Lane 3 shows CHO cells transfected with full-length 7TM expressed in different human cell types. Several distinct modes VPAC1 cDNA in pcDNA3.1 (53 kDa). of GPCR ligand binding have been observed in which any truncation of the GPCR protein could influence ligand binding receptor in the rat central nervous system was and thus affect signaling pathways that could result in various described [29]. These investigators demonstrated that alter- and specific functional activities [36,37]. Therefore, we native splicing of the rat NTS2 receptor gene generated a 5TM investigated and compared the expression and response to domain variant (vNTS2) that is coexpressed with the full-length ligand of the predicted 5TM VPAC isoforms to those of their NTS2 receptor throughout the brain and spinal cord. These 7TM counterpart. Showing that both 5TM and 7TM transcripts findings suggested that the rat vNTS2 is a functional receptor are found differentially expressed in both transformed and that may play a role in NT signaling in the mammalian central normal cell types suggests that expression of these proteins may nervous system. In another study, 5TM mutants of the have biological relevance. chemokine receptors CCR5 and CXCR4 were reported [30]. We predicted from sequence analyses that the 5TM isoforms It was found that 5TM isoforms were produced by the N- would not activate G proteins but that they may, nevertheless, terminal segment connecting directly to TM3 as a result of a transduce a signal when stimulated with ligand. As the 5TM deletion of TM1–2 and the first intracellular and extracellular isoforms for both VPAC1 and VPAC2 have similar structures, loops. In this study it was also shown that the 5TM CCR5 and we used only VPAC1 to examine the functionality of these 5TM CXCR4 isoforms functioned as normal chemokine receptors in isoforms. In this regard, we investigated the ability of VIP and mediating different cell function [30]. It was concluded that in secretin to activate adenylate cyclase and induce cAMP the case of CCR5 and CXCR4, 5TM isoforms appear to meet production and also examined tyrosine phosphorylation follow- the minimum structural requirements for a functional GPCR. ing receptor stimulation. These studies revealed that the 5TM Importantly, investigators have previously reported splice isoform has a severe inability to activate adenylate cyclase and variants of the VPAC family member receptor PAC1 [31–33]. produce cAMP. This was predicted due to deleted sequences In one report [31], splice variants of PAC1 that exhibited altered necessary for heterotrimeric G protein interaction. Of interest adenylate cyclase stimulation and differential abundance were was that when stimulated with secretin, the 5TM isoform identified in the frog. In another study [32], Chaudhary and actually resulted in a significant reduction in the endogenous Baumann found PAC1 splice variants, but not VPAC2 variants, cAMP levels (Fig. 6A). This could occur as a result of ligand- in the rat brain. Stimulation of the PAC1 variants with PACAP induced activation of a phosphodiesterase [38]. Activation of indicated similar potency of these variant receptors. However, phosphodiesterase resulting in decreased cAMP production has in a recent study [33], variants of PAC1 were found in human been previously described [39]. Furthermore, this effect on neuroblastoma cells that demonstrated differential signaling and cAMP by secretin suggests that if phosphodiesterase is the authors concluded that the variant PAC1 receptors have an activated, it must be through a G-protein-coupled-independent important role in the developing nervous system. mechanism. There are a number of reports of G-protein- Although there have not been any reports until our work of independent signaling mechanisms [40–42]. Indeed, despite splice variants of human VPAC1 or VPAC2, an isoform of the lack of G protein binding, the 5TM isoform was able to activate P.B. Bokaei et al. / Genomics 88 (2006) 791–800 797

Fig. 6. Differential signal transduction of 7TM and 5TM VPAC1 receptors. (A) 105 CHO Flp-In cells/ml stably expressing 7TM or alternatively spliced 5TM VPAC1 receptor were untreated (control) or treated with 50 μM VIP or secretin for 30 s. cAMP production was measured by EIA. (B) 3×106 CHO Flp-In cells expressing either 7TM or 5TM VPAC1 were serum starved overnight and then stimulated with 50 μM VIP or secretin for 30 min at 37°C. Cells were washed with PBS and lysed in RIPA lysis buffer. Western blot analysis of total cell lysates after SDS–PAGE using anti-phosphotyrosine antibody is presented. Molecular size markers are indicated on the right. Arrows indicate positions of phosphotyrosine proteins of interest. Western blot for β-actin served as loading control. tyrosine kinase activity, supporting a signaling pathway that is isoform on cancer cells, as it binds ligand but does not activate independent of heterotrimeric G protein activation. Indeed, adenylate cyclase, may compete with the 7TM isoform, as has previous investigators have reported tyrosine kinase-dependent, previously been suggested for IL-4 production with mouse G-protein-coupled-independent signaling pathways [43]. Thus, VPAC2 variants [35]. Thus, further study of the expression of our results suggest that a different signaling pathway(s) that the 5TM isoform in various cancers is warranted. It is also very involves G-protein-independent mechanisms may be associated curious that in cord blood there was no evidence of a 5TM with the 5TM isoform. Our results contrast with a previous isoform of either VPAC1 or VPAC2; however, in adult human report showing that a 5TM isoform of CCR5 had no effect on PBMCs both isoforms were clearly found, with VPAC2 receptor activation compared to 7TM [30]; however, this may isoforms requiring cell activation for optimal expression. be due to the fact that the 5TM isoform in this study continued These data have identified and characterized the region of to retain the ability to bind G proteins. specific alternative splicing of the first human examples of a Our observation that these isoforms of VPAC1 and VPAC2 5TM isoform of both VPAC1 and VPAC2. We have shown the are differentially expressed in various cell types is of particular differential expression of these splice variants in both primary interest. It is intriguing to speculate that since these receptors human peripheral blood cells and various transformed cell lines, have differential signaling, their differential expression, parti- including malignant cancer-derived cell lines. Finally, we have cularly in malignant cells, could have biological significance. demonstrated differential signaling of the 5TM isoform Indeed, it is known that VPAC receptors can be overexpressed in compared to the 7TM form. Taken together, our findings certain cancers and that stimulation of VPAC receptors on cancer provide valuable knowledge of 5TM isoforms of VPAC cells can increase their growth rate [44,45]. Our findings that receptors and suggest these have biological relevance. Further 5TM isoforms of VPAC1 or VPAC2 may be found in malignant studies are currently under way to elucidate better their potential cells raises the intriguing possibility that expression of the 5TM role in biological processes. 798 P.B. Bokaei et al. / Genomics 88 (2006) 791–800

Materials and methods (data not shown). CHO cells were plated in 100-mm dishes (4×106 cells/dish) and transfected with 10 μg of the recombinant plasmid containing the wild-type Cells and variant VPAC1 receptors or pcDNA3.1 empty vector by using the CaPO4 precipitation method. Transfected cells were grown in culture medium (DMEM) supplemented with 10% (v/v) heat-inactivated FBS, 10 μg/ml gentamicin, and Cell lines used in this study were obtained from the American Type Culture Collection (Manassas, VA, USA). CHO Flp-In cells were purchased from Gibco 1% L-glutamine in a fully humidified atmosphere of 95% air and 5% CO2 at (Burlington, ON, Canada). Whole blood was obtained through the Canadian 37°C. Resistant cells were selected by addition of Geneticin (G418) to a final concentration of 0.5 mg/ml. After 10–15 days of selection, individual colonies Blood Services from normal volunteers after they consented to donate blood. PBMCs were isolated from whole blood using a Ficoll–Hypaque density were isolated under limiting dilution in 96-well microtiter plates and grown until gradient. Nonadherent cells were cultured in RPMI 1640 medium supplemented confluence, trypsinized, and further expanded in six-well microtiter plates. The with 10% heat-inactivated fetal bovine serum (FBS; Sigma, Canada) and 0.1% cell clones expressing the different recombinants were selected and used for subsequent studies. Alternatively, the DNA fragments encoding both 7TM and gentamicin. The cultures were maintained at 37°C in a humidified atmosphere of 5% CO and 95% air, at a density of 106 cells/ml by addition of culture 5TM VPAC1 were double digested and subcloned into the pcDNA5/FRT vector 2 – medium every other day. Adherent cell lines were maintained in Dulbecco’s (Invitrogen). Briefly, the HindIII XhoI fragment from the TA vectors, which modified Eagle’s medium (DMEM) (Gibco BRL), supplemented with 10% FBS contain DNA segments that encode 7TM/5TM VPAC1, were subcloned individually into pcDNA5/FRT (Invitrogen). The resulting plasmids were and 100 units/ml each penicillin and streptomycin and grown in 5% CO2 and 95% air at 37°C. CHO Flp-In cells were grown in HAM F-12 nutrient mixture cotransfected along with pOG44 into CHO Flp-In cells (Invitrogen) to induce a (Gibco) containing 10% FBS and 0.125% hygromycin. The cultures were site-specific integration event. Stable integrants were then screened. For verification of the expression of the 7TM and 5TM human VPAC1 receptor maintained at 37°C in a fully humidified atmosphere of 5% CO2 and 95% air, at 6 in the transfected cells, 1 μg of purified mRNA from transfected cells was used a density of 10 cells/ml by addition of culture medium every other day. The viability of all cells was confirmed by trypan blue exclusion (viability >96%) for cDNA synthesis. RT-PCR was carried out with specific VPAC1 receptor prior to experiments. primers and analyzed as described above.

RT-PCR for detection of VPAC transcripts cDNA cloning and sequencing For identification of splice-variant isoforms of both VPAC1 and VPAC2 For cloning of full-length VPAC1 and VPAC2, mRNAs were isolated from receptors in different human cells, total RNA was isolated with the TRIzol normal human primary cells and the Sup-T1 cell line, respectively, using the reagent (Invitrogen) according to the manufacturer’s recommendations. Two QuikPrep Micro mRNA purification (Amersham Pharmacia Biotech, Inc. micrograms of RNA was subjected to reverse transcription with the First-Strand Canada). Reverse transcription was accomplished using the First-Strand cDNA cDNA synthesis kit (Amersham Pharmacia Biotech) according to the synthesis kit (Amersham Pharmacia Biotech) according to the manufacturer’s manufacturer’s directions, with the exception that a mixture of three oligo(dT) directions with the exception that a mixture of three oligo(dT) primers was used primers was used as the first-strand primer containing a C, A, or G at the 3′ as the first-strand primer containing a C, A, or G at the 3′ position [46]. PCR was position [46]. The cDNA was amplified by either PCR or nested-PCR with the performed in a Perkin–Elmer GeneAmp PCR 9700 thermocycler and reactions same reagents and the following primers: VPAC1, 5′-TGCTTGCC- contained 1× Olerup SSP PCR Master Mix with 1 unit of Taq polymerase GTCTCCTTCTTCTCT-3′ (sense) and 5′-GATCCTGGTGGTCAGACCAGG- (Genovison, Inc., West Chester, PA, USA), 25 μM forward and reverse primers, GAGA-3′(antisense), corresponding to nt 822–844 and 1479–1455 of the and 2 μl of template cDNA described above, plus a no-template control, in 25-μl VPAC1 receptor (Accession No. L13288); VPAC2, first-round PCR, 5′- volumes. After initial denaturation at 95°C for 2 min, 35 cycles of amplification GTACTGCATCATGGCCAAC-3′ (sense) and 5′-GTGGGCCTC- were carried out at 94°C for 1 min, 60°C for 1 min, 72°C for 1 min, with a final CCGCCGCGTCC-3′ (antisense) corresponding to nt 792–810 and 1516–1497 extension step of 7 min at 72°C. The nucleotide sequences of oligonucleotide of the VPAC2 receptor (Accession No. L36566); and nested-PCR, 5′- primers used for the PCR were designed based on the published sequences CCCTAGAAGGTGCTTCCTGG-3′ (sense) and 5′-GATGTCTCTAGAT- (GenBank database Accession No. L13288, VPAC1, Sense F, 5′-CTCAGGG- GACCGAGGTCTCCGTTTG-3′ (antisense), corresponding to nt 870–889 and CAGACCATGCGCCCGCCA-3′, and Antisense R, 5′-CAGACCAGGGA- 1479–1456. The products spanned 658 (VPAC1, 7TM), 394 (VPAC1, 5TM), 610 GACTTCGGCTTG-3′, corresponding to, respectively, nt 82–106 and 1467– (VPAC2, 7TM), and 388 bp (VPAC2, 5TM). 1445 of the VPAC1 receptor; and Accession No. L36566, VPAC2, Sense F, 5′- CACGCTGAGCTCGGGATGCGGA-3′, and Antisense R, 5′-AGATGACC- GAGGTCTCCGTTTG-3′, corresponding to nt 147–168 and 1477–1456, Southern blot respectively). The sense and antisense primer pairs used were specific for the amplification of the full-length VPAC receptors as well as the TM5–7 domains For confirmation of PCR products corresponding to the expected 7TM and and C-terminus regions. Following PCR amplification, 20 μl of PCR product 5TM of VPAC2 receptor transcripts, DNA fragments were separated on 1.5% was subjected to electrophoresis on a 1.5% agarose gel and stained with agarose gel and transferred onto Nybond-N+ nylon membrane (Amersham ethidium bromide. The DNA fragments obtained were purified by using the Biosciences). Southern blotting of DNA was performed according to the QIAquick gel extraction kit (Qiagen, Inc., Mississauga, ON, Canada) and then protocol (PT1190-1) from Clontech Laboratories, Inc. As hybridization probes, cloned directly into the pCR2.1-TOPO vector (Invitrogen Canada, Inc., we used the 1.46-kb full-length VPAC2 derived from Sup-T1 cells and the DNA Burlington, ON, Canada). All of the positive clones with a cDNA insert that sequence based on the published sequence (GenBank database Accession No. potentially contained multitranscripts of VPAC1 and VPAC2 were sequenced by L136566). The DNA probes were labeled using [32P]dCTP (Amersham) by the dideoxy method using reagents (Amersham Pharmacia Biotech) with the T7 random priming using the Multiprime DNA labeling system (Amersham and M13 primers. The sequences for VPAC1 and VPAC2 were analyzed using Pharmacia Biotech) according to the manufacturer’s directions. The membrane the software DNASIS Max 2.0 and BLAST in the GenBank database. was washed with hybridization washing buffer and then exposed to X-ray film (Hyperfilm MP). Construction of stable VPAC-expressing cell lines Metabolic labeling The cDNAs encoding both full-length wild-type human VPAC1 receptor and the 5TM variant truncated isoform were excised from the above TA cloning To show the 7TM and 5TM VPAC1 cDNA result in protein expression, we vectors by digestion with BglII and XbaI and inserted into the BamHI and XbaI used metabolic labeling of newly synthesized proteins by 35S-labeled amino sites of the mammalian expression vector pcDNA3.1 (Invitrogen), which acids as previously described [45,46]. Briefly, 100-mm plates of confluent CHO contains an Xpress-tag insert. CHO cells were selected for transfection as these cells transfected using the pcDNA3.1 vector were starved of cysteine (Cys) and cells do not express VPAC1 mRNA, although VPAC2 mRNA was detected (Met) for 1 h and then incubated at 37°C in Cys- and Met-free P.B. Bokaei et al. / Genomics 88 (2006) 791–800 799 medium (Sigma–Aldrich Chemical, Oakville, ON, Canada) containing [3] T. Ishihara, R. Shigemoto, K. Mori, K. Takahashi, S. Nagata, Functional 0.25 mCi each of 35S-labeled Met and Cys (1175 and 1075 Ci/mmol sp act, expression and tissue distribution of a novel receptor for vasoactive respectively; ICN, Costa Mesa, CA, USA) for 4 h. Following incubation, cells intestinal polypeptide, Neuron 8 (1992) 811–819. were washed with cold PBS and lysed in ice-cold RIPA lysis buffer (1% Nonidet [4] S.P. Sreedharan, J.X. Huang, M.C. Cheung, E.J. Goetzl, Structure,

P-40, 0.1% SDS, 0.1% Na3 deoxycholate, 50 mmol/L Hepes, pH 7.3, 150 mmol/ expression, and chromosomal localization of the type I human vasoactive L NaCl, 2 mmol/L sodium orthovanadate, 50 μmol/L ZnCl2, 2 mmol/L EDTA, intestinal peptide receptor gene, Proc. Natl. Acad. Sci. USA 92 (1995) 50 μM/L NaF, 50 μM orthophosphate, 2 mmol/L phenylmethylsulfonyl 2939–2943. fluoride) for 10 min on ice. The lysate was precleared twice by adding it to [5] C.D. Ulrich, M. Holtmann, L.J. Miller, Secretin and vasoactive intestinal unconjugated protein A–Sepharose CL4 B beads (Amersham Bioscience) with peptide receptors: members of a unique family of G protein-coupled rotation for 1 h. The precleared lysate was then incubated for 2 h with anti- receptors, Gastroenterology 114 (1998) 382–397. Xpress antibody that had been prebound to protein A–Sepharose beads. The [6] K. Suda, D.M. Smith, M.A. Ghatei, J.K. Morphy, S.R. Bloom, beads were washed and boiled in reducing sample buffer, and the 35S-labeled Investigation and characterization of receptors for pituitary adenylate proteins were separated on SDS–PAGE. The gel was fixed in 7% acetic acid for cyclase-activating polypeptide in human brain by radio ligand binding 15 min followed by soaking for 30 min in Amplify (Amersham Bioscience). The and chemical cross-linking, J. Clin. Endocrinol. Metab. 72 (1991) gel was dried on a Bio-Rad Model 583 gel dryer at 80°C for 2 h and allowed to 958–964. expose film (Kodak BioMax MR) by autofluorography. [7] X. Hou, A. Vandermeers, P. Gourlet, MC. Vandermeers-Piret, P. Robber- echt, Structural requirements for the occupancy of rat brain PACAP Cyclic AMP assay receptors and adenylate cyclase activation, Neuropharmacology 33 (1994) 1189–1195. The biological activity of 7TM and 5TM human VPAC1 receptors was [8] D. Ganea, Regulatory effects of vasoactive intestinal peptide on cytokine assessed by their ability to elicit intracellular cAMP generation in response to production in central and peripheral lymphoid organs, Adv. Neuroimmu- – VIP and secretin stimulation. In summary, the transfected Flp-In CHO cells (105 nol. 6 (1996) 61 74. cells/ml in 100-μl volumes) were grown in 96-well microtiter plates overnight. [9] M. Delgado, C. Martinez, M.C. Johnson, R.P. Gomariz, D. Ganea, To measure the cAMP accumulation, cells were then stimulated with 50 μM VIP Differential expression of vasoactive intestinal peptide receptors 1 and 2 or 50 μM secretin for 30 s at 37°C. The medium was aspirated, cells were lysed, (VIP-R1 and VIP-R2) mRNA in murine lymphocytes, J. Neuroimmunol. – and cAMP was quantified by using a cAMP enzyme immunoassay (EIA) system 68 (1996) 27 38. (Amersham Pharmacia Biotech), following the manufacturer’s instructions. [10] M. Lara-Marquez, M. O'Dorisio, T. O'Dorisio, M. Shah, B.K. Karacay, Selective and activation-dependent regulation of vasoactive intestinal peptide receptor type 1 and type 2 in human T Western blot analysis cells, J. Immunol. 166 (2001) 2522–2530. [11] J.A. Waschek, M.L. Richards, D.T. Bravo, Differential expression of VIP/ To determine the effect of ligand stimulation on tyrosine phosphorylation, PACAP receptor in breast, intestinal and pancreatic cell lines, Cancer Flp-In CHO cells (3×106) transfected with either the 7TM or the 5TM form Lett. 92 (1995) 143–149. μ of VPAC1 were serum starved overnight and stimulated with 50 M VIP or [12] J.A. Waschek, D.T. Bravo, ML. Richards, High levels of vasoactive secretin for 30 min at 37°C. Following the treatment, cells were washed twice intestinal peptide/pituitary adenylate cyclase-activating peptide receptor with ice-cold PBS and then lysed in RIPA lysis buffer for 10 min on ice. mRNA expression in primary and tumor lymphoid cells, Regul. Pept. 60 After centrifugation, cell lysates were mixed with SDS sample buffer and (1995) 149–157. – boiled. Proteins were separated by 10% SDS PAGE and blotted onto [13] D.R. Branch, L.J.E. Valenta, S. Yousefi, D. Sakac, R. Singla, M. Bali, B.M. nitrocellulose membranes. The membranes were blocked for 1 h at room Sahai, X.Z. Ma, VPAC1 is a cellular neuroendocrine receptor expressed on temperature using Tris-buffered saline (TBS) containing 0.8% BSA (Bioshop, T cells that actively facilitates productive HIV-1 infection, AIDS 16 (2002) Burlington, ON, Canada) plus 0.045% fish gelatin (Sigma–Aldrich). The 309–319. – membranes were washed in TBS Tween 20 and then incubated for 2 h with [14] E.M. Lutz, W.J. Sheward, K.M. West, J.A. Morrow, G. Fink, A.J. Harmar, monoclonal anti-phosphotyrosine antibody (4G10; Upstate Biotechnology, The VIP2 receptor: molecular characterisation of a cDNA encoding a Inc., Lake Placid, NY, USA), washed, and then incubated with goat anti- novel receptor for vasoactive intestinal peptide, FEBS Lett. 334 (1993) mouse IgG conjugated to horseradish peroxidase (Bio-Rad, Mississauga, ON, 3–8. Canada) followed by enhanced chemiluminescence (Bioscience Corp., Baie [15] J.E. Adamou, N. Aiyar, S. Van Horn, N.A. Elshourbagy, Cloning and ’ D Urfé, QC, Canada) used to visualize the proteins. Membranes were functional characterization of the human vasoactive intestinal peptide stripped (30 min at 50°C using 100 mM 2-mercaptoethanol and 2% SDS in (VIP)-2 receptor, Biochem. Biophys. Res. Commun. 209 (1995) – β – 62.5 mM Tris HCl, pH 6.9) and then reprobed using anti- -actin (Sigma 385–392. Aldrich) to control for loading. [16] P. Vertongen, B. Velkeniers, E. Hooghe-Peters, P. Robberecht, Differential alternative splicing of PACAP receptor in pituitary cell subpopulations, Mol. Cell. Endocrinol. 113 (1995) 131–135. Acknowledgment [17] T.B. Usdin, T.I. Bonner, E. Mezey, Two receptors for vasoactive intestinal polypeptide with similar specificity and complementary distributions, This work was supported by a Graduate Fellowship Award Endocrinology 135 (1994) 2662–2680. from the Canadian Blood Services to P.B.B. [18] M. Delgado, E.J. Munoz-Elias, R.P. Gomariz, D. Ganea, Vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide prevent inducible nitric oxide synthase transcription in macrophages by References inhibiting NF-kappa B and IFN regulatory factor 1 activation, J. Immunol. 162 (1999) 4685–4696. [1] B. Martin, R.L. Maturana, R. Brenneman, T. Walent, M.P. Mattson, S. [19] P. Nicole, K. Du, A. Couvineau, M. Laburthe, Site-directed mutagenesis of Maudsley, Class II G protein-coupled receptors and their ligands in human vasoactive intestinal peptide receptor subtypes VIP1 and VIP2: neuronal function and protection, Neuromol. Med. 7 (2005) 3–36. evidence for difference in the structure–function relationship, J. Pharma- [2] M. Laburthe, A. Couvineau, P. Gaudin, J.J. Maoret, C. Rouyer-Fessard, P. col. Exp. Ther. 284 (1998) 744–750. Nicole, Receptors for VIP, PACAP, secretin, GRF, , GLP-1, and [20] M. Xia, G.O. Gaufo, Q. Wang, S.P. Sreedharan, E.J. Goetzl, Transduction other members of their new family of G protein-linked receptors: of specific inhibition of HuT 78 human T cell chemotaxis by type I structure–function relationship with special reference to the human VIP- vasoactive intestinal peptide receptors, J. Immunol. 157 (1996) 1 receptor, Ann. N. Y. Acad. Sci. 805 (1996) 94–109. 1132–1138. 800 P.B. Bokaei et al. / Genomics 88 (2006) 791–800

[21] M. Tsutsumi, T.H. Claus, Y. Liang, Y. Li, L. Yang, J. Zhu, F. Dela Cruz, X. human neuroblastoma cells: consequences for signaling by VIP and Peng, H. Chen, S.L. Yung, S. Hamren, J.N. Livingston, C.Q. Pan, A potent PACAP, Mol. Cell. Neurosci. 31 (2006) 193–209. and highly selective VPAC2 agonist enhances glucose-induced [34] C. Grinninger, W. Wang, K.B. Oskoui, J.K. Voice, E.J. Goetzl, A natural release and glucose disposal: a potential therapy for type 2 diabetes, variant type II G protein-coupled receptor for vasoactive intestinal peptide Diabetes 251 (2002) 1453–1460. with altered function, J. Biol. Chem. 279 (2004) 40259–40262. [22] P. Baradar Bokaei, X.Z. Ma, D. Sakac, D.R. Branch, HIV-1 proviral [35] M.-C. Huang, A.L. Miller, W. Wang, Y. Kong, S. Paul, E.J. Goetzl, integration is inhibited by specific agonist stimulation of the VPAC2 Differential signalling of T cell generation of IL-4 by wild-type and short- neuroendocrine receptor (submitted for publication). deletion variant of type 2 G protein-coupled receptor for vasoactive

[23] M. Laburthe, A. Couvineau, J.C. Marie, VPAC receptors for VIP and intestinal peptide (VPAC2), J. Immunol. 176 (2006) 6640–6646. PACAP, Recept. Channels 8 (2002) 137–153. [36] C. Pantaloni, P. Brabet, B. Bilanges, A. Dumuis, S. Houssami, D. Spengler, [24] D. Vaudry, B.J. Gonzales, M. Basille, L. Yon, A. Fournier, H. Vaudry, J. Bockaert, L. Journot, Alternative splicing in the N-terminal extracellular Pituitary adenylate cyclase activating polypeptide and its receptors: from domain of the pituitary adenylate cyclase-activating polypeptide (PACAP) structure to function, Pharmacol. Rev. 52 (2000) 269–324. receptor modulates receptor selectivity and relative potencies of PACAP- [25] Q.I. Qy, F. Wang, H.T. Zhang, J.C. Wang, H.P. Xiao, M.H. Wang, Y.F. 27 and PACAP-38 in phospholipase C activation, J. Biol. Chem. 271 Han, R.M. Zhang, S.H. Tao, Z.W. Luo, Induced recovery of defective (1996) 22146–22151. membrane expression of a CC 5 mutant by [37] K.M. Braas, V. May, Pituitary adenylate cyclase-activating polypeptides phytohemagglutinin, Cell Mol. Life Sci. 60 (2003) 2492–2500. directly stimulate sympathetic neuron Y release through PAC [26] S. Venkatesan, A. Petrovic, D.I. Van Ryk, M. Locati, D. Weissman, P.M. (1) receptor isoform activation of specific intracellular signaling pathways, Murphy, Reduced cell surface expression of CCR5 in CCR5Delta 32 J. Biol. Chem. 274 (1999) 27702–27710. heterozygotes is mediated by gene dosage, rather than by receptor [38] R.K. Sharma, S.B. Das, A. Lakshmikuttyamma, P. Selvakumar, A. sequestration, J. Biol. Chem. 277 (2002) 2287–2301. Shrivastav, Regulation of calmodulin-stimulated cyclic nucleotide phos- [27] P. Gaudin, A. Couvineau, C. Rouyer-Fessard, J.J. Maoret, M. Laburthe, phodiesterase (PDE1), Int. J. Mol. Med. 18 (2006) 95–105. The human vasoactive intestinal peptide/pituitary adenylate cyclase [39] L. Kang, L.E. Nagy, Chronic ethanol feeding suppresses {beta}-adrenergic activating peptide receptor 1 (VPAC1): constitutive activation by receptor-stimulated lipolysis in adipocytes isolated from epididymal fat, mutations at threonine 343, Biochem. Biophys. Res. Commun. 254 Endocrinology (2006) (Epub ahead of print). (1999) 15–20. [40] M. Wong, S. Uddin, B. Majchrzak, T. Huynh, A.E. Proudfoot, L.C. [28] P. Gaudin, J.J. Maoret, A. Couvineau, C. Rouyer-Fessard, M. Laburthe, Platanias, E.N. Fish, Rantes activates Jak2 and Jak3 to regulate Constitutive activation of the human vasoactive intestinal peptide 1 engagement of multiple signaling pathways in T cells, J. Biol. Chem. receptor, a member of the new class II family of G protein-coupled 276 (2001) 11427–11431. receptors, J. Biol. Chem. 273 (1998) 4990–4996. [41] J.G. Williams, Serpentine receptors and STAT activation: more than one [29] A. Perron, P. Sarret, L. Gendron, T. Stroh, A. Beaudet, Identification and way to twin a STAT, Trends Biochem. Sci. 24 (1999) 333–334. functional characterization of a 5-transmembrane domain variant isoform [42] P. Schnabel, H. Gas, T. Nohr, M. Bohm, G protein-independent stimulation of the NTS2 in rat central nervous system, J. Biol. of human myocardial phospholipase C by mastoparan, Br. J. Pharmacol. Chem. 280 (2005) 10219–11027. 122 (1997) 31–36. [30] K. Ling, P. Wang, J. Zhao, Y.L. Wu, Z.J. Cheng, G.X. Wu, W. Hu, L. Ma, [43] A.T. Ting, K.J. Einspahr, R.T. Abraham, P.J. Leibson, Fc gamma G. Pei, Five-transmembrane domains appear sufficient for a G protein- receptor signal transduction in natural killer cells: coupling to phos- coupled receptor: functional five-transmembrane domain chemokine pholipase C via a G protein-independent, but tyrosine kinase-dependent receptors, Proc. Natl. Acad. Sci. USA 96 (1999) 7922–7927. pathway, J. Immunol. 147 (1991) 3122–3127. [31] D. Alexandre, H. Vaudry, L. Grumolato, V. Turquier, A. Fournier, S. Jegou, [44] T.W. Moody, J.M. Hill, R.T. Jensen, VIP as a trophic factor in the CNS and Y. Anouar, Novel splice variants of type I pituitary adenylate cyclase- cancer cells, 24 (2003) 163–177. activating polypeptide receptor in frog exhibit altered adenylate cyclase [45] S. Schulz, C. Rocken, C. Mawrin, W. Weise, V. Hollt, S. Schulz, stimulation and differential relative abundance, Endocrinology 143 (2002) Immunocytochemical identification of VPAC1, VPAC2, and PAC1 2680–2692. receptors in normal and neoplastic human tissues with subtype-specific [32] P. Chaudhary, T.K. Baumann, Expression of VPAC2 receptor and PAC1 antibodies, Clin. Cancer Res. 10 (2004) 8235–8242. receptor splice variants in the trigeminal ganglion of the adult rat, Brain [46] X.Z. Ma, T. Jin, D. Sakac, S. Fahim, X. Zhang, Y. Katsman, M. Bali, D.R. Res. Mol. Brain Res. 104 (2002) 137–142. Branch, Abnormal splicing of SHP-1 protein tyrosine phosphatase in [33] E.M. Lutz, E. Ronaldson, P. Shaw, M.S. Johnson, P.J. Holland, R. human T cells: implications for lymphomagenesis, Exp. Hematol. 31

Mitchell, Characterization of novel splice variants of the PAC1 receptor in (2003) 131–142.