A Natural Variant Type II G Protein-coupled Receptor for Vasoactive Intestinal Peptide with Altered Function

Abstract

The vasoactive intestinal peptide (VIP) and its G protein-coupled receptors VPAC1 and VPAC2 prominently mediate diverse physiological functions in the neural, endocrine, and immune systems. A deletion variant of mouse VPAC2 has been identified in immune cells that lacks amino acids 367–380 at the carboxyl-terminal end of the seventh transmembrane domain. When expressed at equivalent levels in a human Jurkat T cell line, which has very low endogenous expression of human VPAC1 and VPAC2, wild-type and deletion-variant VPAC2 bound the same amount of 125I-VIP with similar affinity. Unlike wild-type VPAC2, however, deletion-variant VPAC2 did not transduce VIP-elicited increases in intracellular concentration of cyclic AMP, chemotaxis, or suppression of generation of interleukin-2. Natural deletion of part of the last transmembrane domain of VPAC2 thus abrogates signaling functions without apparent alterations of expression or ligand binding. The vasoactive intestinal peptide (VIP) and its G protein-coupled receptors VPAC1 and VPAC2 prominently mediate diverse physiological functions in the neural, endocrine, and immune systems. A deletion variant of mouse VPAC2 has been identified in immune cells that lacks amino acids 367–380 at the carboxyl-terminal end of the seventh transmembrane domain. When expressed at equivalent levels in a human Jurkat T cell line, which has very low endogenous expression of human VPAC1 and VPAC2, wild-type and deletion-variant VPAC2 bound the same amount of 125I-VIP with similar affinity. Unlike wild-type VPAC2, however, deletion-variant VPAC2 did not transduce VIP-elicited increases in intracellular concentration of cyclic AMP, chemotaxis, or suppression of generation of interleukin-2. Natural deletion of part of the last transmembrane domain of VPAC2 thus abrogates signaling functions without apparent alterations of expression or ligand binding. Vasoactive intestinal peptide (VIP) 1The abbreviations used are: VIP, vasoactive intestinal peptide; PACAP, pituitary adenylate cyclase-activating peptide; GPCR, G protein-coupled receptor; VPAC1, type I VIP GPCR; VPAC2, type II VIP GPCR; IL-2, interleukin 2; [cAMP]i, intracellular concentration of cyclic AMP; hHPRT, human hypoxanthine phosphoribosyltransferase; TF, transfectant; FAM, 6-carboxyfluorescein; TAMRA, 6-carboxytetramethylrhodamine. and pituitary adenylate cyclase-activating peptide (PACAP) are the two members of a cluster of neuroendocrine and immune peptide mediators that are most closely related structurally and by shared use of two G protein-coupled receptors (GPCRs) termed VPAC1 and VPAC2 (1Sreedharan S.P. Huang J.X. Cheung M.C. Goetzl E.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 2939-2943Crossref PubMed Scopus (160) Google Scholar, 2Harmar A.J. Arimura A. Gozes I. Journot L. Laburthe M. Pisegna J.R. Rawlings S.R. Robberecht P. Said S.I. Sreedharan S.P. Wank S.A. Waschek J.A. Pharmacol. Rev. 1998; 50: 265-270PubMed Google Scholar, 3Lutz E.M. Shen S. Mackay M. West K. Harmar A.J. FEBS Lett. 1999; 458: 197-203Crossref PubMed Scopus (22) Google Scholar). VPAC1 and VPAC2 each couple to at least two different G proteins and consequently transduce increases in intracellular concentrations of calcium and cyclic AMP, as well as an array of cellular functional responses to VIP and PACAP. VPAC2 is inducible on lymphocytes and regulates the balance between two major types of immune responses (4Voice J.K. Dorsam G. Lee H. Kong Y. Goetzl E.J. FASEB J. 2001; 15: 2489-2496Crossref PubMed Scopus (80) Google Scholar, 5Goetzl E.J. Voice J.K. Shen S. Dorsam G. Kong Y. West K.M. Morrison C.F. Harmar A.J. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 13854-13859Crossref PubMed Scopus (123) Google Scholar). The structures of the genes encoding both VPAC1 and VPAC2, and a GPCR termed PAC1 (6Pisegna J.R. Wank S.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6345-6349Crossref PubMed Scopus (373) Google Scholar) with exclusive specificity for PACAP, are far more complex than those for most other GPCRs. This genomic complexity is considered to explain in part the many different alternatively spliced and other structurally distinct forms of PAC1 (7Zhou C.J. Kikuyama S. Arimura A. Shioda S. Ann. N. Y. Acad. Sci. 2000; 921: 373-374Crossref PubMed Scopus (6) Google Scholar, 8Ajpru S. McArthur A.J. Piggins H.D. Sugden D. Brain Res. Mol. Brain Res. 2002; 105: 29-37Crossref PubMed Scopus (26) Google Scholar). However, no such natural mutant forms of VPAC1 and VPAC2 have been identified to date. A search for mutant forms of VPAC2 was carried out by PCR amplification of RNA sequences encoding VPAC2 that were isolated from lymphoid organs of the immune system. The discovery of a naturally occurring variant VPAC2, which lacks a 14-amino acid sequence at the cytosolic end of the seventh transmembrane domain, prompted initial studies of its expression, VIP-binding properties, and characteristics of transduction of cellular biochemical and functional responses to VIP. This deletion-variant VPAC2 appears to bind VIP normally in lymphocyte transfectants but unlike wild-type VPAC2 does not transduce VIP-elicited increases in intracellular concentration of cyclic AMP ([cAMP]i), decreases in generation of interleukin 2 (IL-2), or chemotactic responses. Identification, Cloning, and Sequencing of cDNAs Encoding Mouse Wild-type and Deletion-mutant VPAC2 GPCRs—Total RNA was isolated from thymus and spleen tissues of normal 6–8-week-old C57BL/6 female mice with TRIzol (Invitrogen) for synthesis of first-strand cDNAs using the Superscript II system (Invitrogen). VPAC2 DNAs in the first-strand cDNA array were amplified by PCR with platinum TaqDNA polymerase (Invitrogen) and primers for sequences at the ends of the coding region: 5′-gac tag tcc acc atg agg gcg tcg gtg gtg ctg a-3′ (forward (SpeI)) and 5′-gga att cct aaa tga ctg aag tct ctg act gca gga a-3′ (reverse (EcoRI)). Individual VPAC2 DNAs isolated by 1% agarose gel electrophoresis were introduced into pBluescript II SK(+) phagemids (Stratagene, La Jolla, CA) at the SpeI/EcoRI sites for transformation into Library-Efficiency DH5α Escherichia coli (Invitrogen), selection of colonies with PCR-detectable VPAC2 inserts in plasmid mini-preparations (Qiagen, Valencia, CA), and determination of the sequences of these VPAC2 DNAs with primers for vector sequences: 5′-ctcgaaattaaccctcactaaa-3′ (forward) and 5′-gaattgtaatacgactcactatag-3′ (reverse) (ELIM Biopharmaceuticals, Inc., Hayward, CA). Multiple DNAs encoding wild-type and deletion-variant VPAC2 receptors were subcloned into the pIRESneo3 bicistronic expression vector (BD Biosciences) at the NheI/EcoRI sites and re-sequenced with the primers: 5′-ctcactatagggagacccaagctt-3′ (forward) and 5′-cgcttttgagagggagtactcac-3′ (reverse). Generation of Jurkat Human T Cell Transfectants and Characterization of Their VPAC2 Expression—Replicate suspensions of 107 Jurkat T cells in Opti-MEM I (University of California San Francisco Cell Culture Facility) were incubated as recommended with a mixture of Gene-PORTER transfection reagent and 40 μg of purified pIRES plasmids containing wild-type or deletion-mutant VPAC2 DNA, followed 4 h later by addition of Booster Express reagent number 1 (Gene Therapy Systems, Inc., San Diego, CA). After 24 h, the Jurkat cells were washed and cultured for 2 days in complete RPMI 1640 medium containing 10% fetal bovine plasma, 100 units/ml penicillin G, and 50 μg/ml streptomycin. Geneticin (Invitrogen) then was added to the cultures in steps of 200, 400, and 600 μg/ml every 4–6 days and cultures continued further for 7–10 days prior to quantification of their expression of VPAC GPCRs by real-time PCR. Real-time PCR was carried out as described in an ABI Prism 7700 sequence detection system (9Graeler M. Goetzl E.J. FASEB J. 2002; 16: 1874-1878Crossref PubMed Scopus (192) Google Scholar, 10Wang W. Graeler M.H. Goetzl E.J. FASEB J. 2004; 18: 1043-1045Crossref PubMed Scopus (39) Google Scholar) using the following primers and probes: hVPAC1, 5′-acaaggcagcgagtttggat-3′ (forward), 5′-gtgcagtggagcttcctgaac-3′ (reverse), and 5′-FAM-ccacccttctggtcgccacagctat-TAMRA-3′ (probe); hVPAC2, 5′-ctgccaatgtgggagagacc-3′ (forward), 5′-tctgggaacgtctctgaccat-3′ (reverse), 5′-FAM-acggtgccctgcccaaaagtcttc-TAMRA-3′ (probe); mVPAC2, 5′-agctgggagccgggacta-3′ (forward), 5′-cacggtgtatctgtagggcactt-3′ (reverse), and 5′-FAM-ctgcacagctggtccatgtcccg-TAMRA-3′ (probe); hHPRT, 5′-ggcagtataatccaaagatggtcaa-3′ (forward), 5′-gtctggcttatatccaacacttcgt-3′ (reverse), and 5′-TET-caagcttgctggtgaaaaggacccc-TAMRA-3′ (probe). Western blots of proteins extracted from each transfectant (TF) and untransfected Jurkat T cells, as described (11Xia M. Sreedharan S.P. Goetzl E.J. J. Clin. Immunol. 1996; 16: 21-30Crossref PubMed Scopus (27) Google Scholar), were developed with a rabbit antibody to the amino-terminal sequence 54–70 (12Kaltreider H.B. Ichikawa S. Byrd P.K. Ingram D.A. Kishiyama J.L. Sreedharan S.P. Warnock M.L. Beck J.M. Goetzl E.J. Am. J. Respir. Cell Mol. Biol. 1997; 16: 133-144Crossref PubMed Scopus (150) Google Scholar). Mouse anti-human CD45 monoclonal antibody (clone 69, IgG1, PharMingen) was used to standardize the quantity of cellular proteins added to each well. Intact cellular binding of 3-[125I]iodotyrosyl-VIP (specific activity = 2,000 Ci/mmol, PerkinElmer Life Sciences) was quantified as described (13Sreedharan S.P. Patel D.R. Xia M. Ichikawa S. Goetzl E.J. Biochem. Biophys. Res. Commun. 1994; 203: 141-148Crossref PubMed Scopus (60) Google Scholar). Assessment of Biochemical and Functional Responses of Jurkat T Cell TFs to VIP—To measure VIP-evoked increases in the intracellular concentration of cyclic AMP, [cAMP]i, replicate 0.5-ml aliquots of 106 TFs/ml of RPMI 1640 with 1 mg/ml crystallized bovine serum albumin and 5 μm Ro20-1724 (Biomol, Plymouth Meeting, PA) were added to 24-well plates. After 15 min of preincubation at 37 °C in humidified 5% CO2 in air, 10–9 to 10–6m VIP, 10–6m VPAC1-selective agonist [Lys15,Arg16,Leu27]VIP-(1–7)/GRF-(8–27) or 10–6m VPAC2-selective agonist Ac-[Glu8,OCH3-Tyr10,Lys12,Nle17,Ala19,Asp25,Leu26,Lys27,28]VIP-cyclo-(21–25) (14Gourlet P. Vandermeers A. Vertongen P. Rathe J. De Neef P. Cnudde J. Waelbroeck M. Robberecht P. Peptides. 1997; 18: 1539-1545Crossref PubMed Scopus (167) Google Scholar, 15Xia M. Sreedharan S.P. Bolin D.R. Gaufo G.O. Goetzl E.J. J. Pharmacol. Exp. Ther. 1997; 281: 629-633PubMed Google Scholar) were added, incubations continued for 20 min at 37 °C, and T cells were pelleted by centrifugation at 4 °C for 10 min at 400 × g. Results of preliminary studies at 2, 20, and 60 min showed maximal levels of cAMP at 20 min. After removal of supernates, the T cells were disrupted by incubation for 10 min at 37 °C in 0.25 ml per well of 0.1 m HCl with 1% (v/v) Triton X-100 and re-centrifuged at 4 °C for 10 min at 1,200 × g. The [cAMP]i in the lysate supernates was quantified directly after acetylation with the Correlate-EIA kit (Assay Designs, Inc., Ann Arbor, MI); sensitivity = 0.035 nm and mean recovery = 85%. To determine the extent of VIP suppression of generation of human IL-2 by stimulated Jurkat T cells, replicate 0.6-ml aliquots of 2 × 106 Jurkat cells/ml of complete RPMI 1640 were added to 24-well plates that had been precoated with 0.75 μg/well of mouse anti-human CD3 monoclonal antibody (PharMingen). After addition of 10–9 to 10–6m VIP or 10–6m VPAC1-selective agonist or VPAC2-selective agonist, wells received 20 ng/ml phorbol myristate acetate, and incubations were continued for 48 h at 37 °Cin5%CO2 in humidified air. At 24 and 48 h, 200-μl portions of supernatant were harvested for immunoassays of IL-2 (R&D Systems, Inc., Minneapolis, MN). Chemotaxis of Jurkat T cells was assessed as described (16Zheng Y. Kong Y. Goetzl E.J. J. Immunol. 2001; 166: 2317-2322Crossref PubMed Scopus (87) Google Scholar) after 4 h of incubation in Transwell chambers over 5-μm pore filters, which had been pretreated in 1 μg/ml of human type IV collagen and dried. Stimuli included 10–9 to 10–6m VIP and 100 nm human recombinant SDF-1α chemokine (CXCL-12, Peprotech, Inc., Rocky Hills, NJ). The results of this and other assays are presented as mean ± S.D., and t tests for statistical significance were conducted with the Prism 4 system (GraphPad Software, Inc., San Diego, CA). A variant form of mouse VPAC2 was identified in multiple clones from both spleen and thymus. The sequence showed a deletion of 42 bases and re-alignment, which led to loss of 14 amino acids (367–380) from the carboxyl-terminal end of the seventh transmembrane domain (Table IA). This region may be vital for both cellular behavior and signal transduction of VPAC2, as it is in some other GPCRs with short intracellular loops. Therefore, studies were designed to examine these characteristics of mouse VPAC2 GPCR stable TFs in Jurkat human T cells that have low levels of endogenous human VPAC2 receptors and barely detectable levels of human VPAC1 receptors (Table IB). After several weeks of Geneticin-selection of Jurkat T cells transfected with pIRESneo3-VPAC2 plasmids, the mRNAs encoding wild-type and deletion-variant (deletion-mutant) mouse VPAC2 were substantially more abundant than that encoding endogenous human wild-type VPAC2 (Table IB).Table IStructures of mVPAC2 GPCRs and levels of mRNA expression in Jurkat T cell transfectantsA. StructuresDNA sequencesWild-type1141ttccttccagggcctggtggtagcagttctatactgcttcctgaacagtgaggtacagtg1200Deletion mutantttccttccag–––––––––––––––––––––gtacagtgAmino acid sequencesWild type358FELCVGSFQGLVVAVLYCFLNSEVQCELKRRW389Deletion mutantFELCVGSFQ–––––––VQCELKRRWB. Real-time PCR quantification of mRNA expressionVPAC GPCRJurkat T cell mVPAC2 TFsam or h, mouse or human mRNA; the intensity of the real-time PCR signal for a human HPRT standard in each sample was set at 100.UntransfectedWild-type mVPAC2 TFDeletion mutant mVPAC2 TFhVPAC10.18 ± 0.01bmRNA level relative to that of hHPRT, mean ± S.D.0.24 ± 0.060.42 ± 0.22hVPAC24.96 ± 0.182.21 ± 0.122.06 ± 0.08mVPAC20.00 ± 0.0011.7 ± 1.2512.1 ± 0.20a m or h, mouse or human mRNA; the intensity of the real-time PCR signal for a human HPRT standard in each sample was set at 100.b mRNA level relative to that of hHPRT, mean ± S.D. Open table in a new tab Similarly, Western blots of membrane proteins extracted from Jurkat cells showed that total VPAC2 protein levels were substantially higher in wild-type and deletion-mutant mouse VPAC2 TFs than untransfected Jurkat cells (Fig. 1). Although this polyclonal anti-VPAC2 antibody specific for the amino terminus does not distinguish between mouse and human VPAC2 GPCRs, the mRNA measurements indicate that transfection with either mouse VPAC2 decreases the level of message for endogenous human VPAC2 (Table IB). Thus it is assumed that the higher total levels of VPAC2 receptor protein in TFs are attributable nearly totally to the respective mouse VPAC2 GPCR (Fig. 1). The specific binding of 125I-VIP by 106 wild-type and deletion-mutant mouse VPAC2 TFs, respectively, was a mean (±S.D.) of 18,635 ± 3,078 dpm and 19,277 ± 3,552 dpm, as contrasted with 2,490 ± 838 for untransfected Jurkat T cells (n = 2, in duplicate). The respective mean Kd values were 17, 19 and 15 nm. Thus any differences in signaling and transduction of functions between the two forms of mouse VPAC2 are not attributable to altered binding of VIP. One of the principal signals transduced by VIP binding to VPAC2 is a Gαs-mediated increase in [cAMP]i. Several hundredfold increases in the [cAMP]i in wild-type VPAC2 TFs were elicited by VIP in a concentration-dependent relationship up to a maximal level similar to that attained by forskolin (Fig. 2). In contrast, deletion-mutant VPAC2 TFs showed no rise in [cAMP]i at any concentration of VIP. Untransfected Jurkat T cells that express only low levels of human native VPAC2 showed only modest increases in [cAMP]i, which were significant at the highest two concentrations of VIP but never exceeded 10-fold. The striking increments in [cAMP]i evoked by forskolin in all Jurkat cell lines were indistinguishable. In two of the studies where mean increases in [cAMP]i in wild-type VPAC2 TFs were 238-fold and 366-fold with 10–8m and 10–7m VIP, respectively, mean increases were 184-fold for 10–6m VPAC2-selective agonist and 3.2-fold for 10–6m VPAC1-selective agonist. Thus increases in [cAMP]i were mediated solely by signals from VPAC2 GPCRs. The known consistent suppression of T cell generation of IL-2 by VIP was confirmed through VIP concentration-dependent inhibition of IL-2 generation by wild-type VPAC2 TFs stimulated for 24 and 48 h (Fig. 3). After 24 h, inhibition of IL-2 generation in wild-type VPAC2 TFs was significant at 10–8 to 10–6m VIP and achieved a maximal level of over 60% at 10–6m VIP. After 48 h, inhibition was significant at 10–7 and 10–6m VIP with maximal suppression of 52% at 10–7m VIP. IL-2 generation by stimulated untransfected Jurkat T cells was just marginally suppressed only by 10–6m VIP at 24 h and not at all at 48 h. There was no inhibitory effect of any concentration of VIP at either time on IL-2 generation by stimulated deletion-mutant TFs (Fig. 3). In two of the studies where mean suppression of IL-2 generation in wild-type VPAC2 TFs were 30 and 48% with 10–8m and 10–7m VIP, respectively, mean inhibition was 44% for 10–6m VPAC2-selective agonist and only 5% for 10–6m VPAC1-selective agonist. As for [cAMP]i responses, suppression of IL-2 generation was mediated solely by signals from VPAC2 GPCRs. Wild-type VPAC2 TFs showed significant chemotactic responses to 10–9 to 10–6m VIP, concentration-dependent increments in responses, and maximal chemotaxis at 10–7m VIP, as had been observed for human blood and mouse spleen T cells (Fig. 4) (17Xia M. Leppert D. Hauser S.L. Sreedharan S.P. Nelson P.J. Krensky A.M. Goetzl E.J. J. Immunol. 1996; 156: 160-167PubMed Google Scholar). In contrast, untransfected Jurkat T cells showed marginal chemotactic responses only to 10–8 and 10–7m VIP, and deletion-mutant VPAC2 TFs did not respond at all to VIP (Fig. 4). It is not yet known whether this deletion-variant arises from a mutation, alternative splicing, or another mechanism. It also is difficult to predict the effects on VPAC2 structure of deleting amino acids 367–380, which constitute the carboxyl-terminal end of the seventh transmembrane domain (Table IA). Two predictions seem certain, both of which are consistent with the observed properties of the deletion-variant mouse VPAC2 in Jurkat T cells. First, most sites of post-translational modifications involved in signaling and for docking of signaling proteins are in the carboxyl-terminal tail. The likely retraction and distortion of this intracellular tail induced by the amino acid sequence deletion would be expected to alter signaling and consequently cellular functional effects. Thus it is not surprising that TFs expressing deletion-mutant VPAC2 GPCRs at a level equal to that of the wild-type VPAC2 GPCRs in parallel TFs (Table IB and Fig. 1) fail to manifest VIP-evoked increases in [cAMP]i, suppression of IL-2 generation, or chemotaxis (Figs. 2, 3, 4). Second, results of studies of structural determinants of VIP binding strongly support requirements for the extracellular amino terminus, portions of the first and second extracellular loops, and part of the fifth transmembrane domain but not the seventh transmembrane domain or carboxyl terminus (18Nicole P. Du K. Couvineau A. Laburthe M. J. Pharmacol. Exp. Ther. 1998; 284: 744-750PubMed Google Scholar, 19Juarranz M.G. Van Rampelbergh J. Gourlet P. De Neef P. Cnudde J. Robberecht P. Waelbroeck M. Mol. Pharmacol. 1999; 56: 1280-1287Crossref PubMed Scopus (31) Google Scholar, 20Park C.G. Ganguli S.C. Pinon D.I. Hadac E.M. Miller L.J. J. Pharmacol. Exp. Ther. 2000; 295: 682-688PubMed Google Scholar). Thus normal expression and VIP-binding properties of the deletion-mutant VPAC2 GPCR are not unexpected. What remains to be elucidated about the deletion-variant VPAC2 GPCR is the mechanism of its generation, its cell and tissue distribution, determinants of the regulation of its expression and after cellular and its natural functional in and other physiological and by the GPCR