- Email: [email protected]
Expression of the Rat Adrenomedullin Receptor or a Putative Human Adrenomedullin Receptor Does Not Correlate with Adrenomedullin Binding or Functional Response
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
244, 832–837 (1998)
RC988349
Expression of the Rat Adrenomedullin Receptor or a Putative Human Adrenomedullin Receptor Does Not Correlate with Adrenomedullin Binding or Functional Response Scott P. Kennedy,*,1 Dexue Sun,* Joseph J. Oleynek,† Christopher F. Hoth,* Jimmy Kong,† and Roger J. Hill† *Department of Molecular Sciences and †Department of Cardiovascular and Metabolic Diseases, Pfizer Inc., Groton, Connecticut 06340
Received February 17, 1998
There has been considerable difficulty in defining distinct adrenomedullin (AM) binding sites and function in vivo. However, a rat adrenomedullin receptor (rAMR) and a putative human adrenomedullin receptor (hAMR) have recently been reported. We attempted to confirm and extend the pharmacological characterization of these cloned receptors. COS-7 cells transfected with rAMR or epitope tagged rAMR display abundant rAMR mRNA expression and cell-surface receptor localization. Specific 125I-AM binding is detected in transfected cells; however, similar levels of binding are also detected in cells transfected with vector DNA alone. This AM binding site fails to mediate any changes in cAMP in response to AM. In contrast, Swiss 3T3 cells, expressing specific endogenous AM receptors, display AM binding and functional cAMP responses. Transfection studies performed with the putative hAMR yield similar results. These data suggest that the proposed rAMR and hAMR do not represent authentic adrenomedullin receptors. q 1998 Academic Press
Adrenomedullin (AM) is a 52 amino acid vasoactive peptide member of the calcitonin gene related peptide (CGRP) super family (7,8). It is produced from a variety of vascularized tissues as preproadrenomedullin (185 amino acids) and processed into AM (95-146) as well as other biologically active peptides, including proadrenomedullin (22-41), prodepin (45-92) and adrenotensin (153-185) (7-9). AM displays a wide variety of in vivo activities in the vasculature, adrenal gland, kidney, pituitary gland and brain (7,8). Circulating plasma AM 1 Address correspondence to Dr. Scott P. Kennedy, Pfizer Inc., Department of Molecular Sciences, Eastern Point Rd, Groton, CT 06340. Fax: (860) 441-3783. E-mail: [email protected].
concentrations have been found to be elevated and/or correlated with severity of disease state in essential hypertension, primary aldosteronism, chronic renal failure, congestive heart failure, acute myocardial infarction, and hyperthyroidism (7). It has not been possible to determine if some or all of the actions of AM are mediated through specific AM receptors, since CGRP shares many of the same biological actions and binding sites with AM, and highly selective ligands are not yet available. However, specific AM binding sites have been demonstrated in rat vascular smooth muscle cells (10), bovine endothelial cells (11), rat lung and vascular tissues (12), mouse astrocytes (13), NG108-15 cells (14) and Swiss 3T3 cells (15). The most direct evidence for a specific AM receptor comes from studies reporting the cloning and expression of a putative rat adrenomedullin receptor (rAMR) (2). These studies were performed in transfected COS-7 cells, where this receptor bound 125I-rat adrenomedullin (rAM) with high affinity (KDÅ 8.2 nM) and mediated a cAMP response to rAM (EC50Å 7 nM), but not to CGRP. No further characterization of this receptor has been reported. Very recently, a putative human adrenomedullin receptor (hAMR) cDNA was isolated from lung which displayed 73% homology with the rAMR although no expression or pharmacology studies were reported (5). There clearly exists a need for defining the specific role of AM and its related peptides in normal and pathological states, and a potential therapeutic value in developing pharmacological agonists or antagonists of AM action. Therefore, the goal of the present study was to confirm and extend the pharmacological characterization of the cloned rat and human adrenomedullin receptors. MATERIALS AND METHODS Materials. Restriction endonucleases and other molecular biological reagents were purchased from New England Biolabs unless oth-
832
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
erwise noted. All molecular biological methods were performed according to manufacturer kit protocols or according to Sambrook et al. (1). COS-7 cells (ATCC, Rockville, MD) and Swiss 3T3 fibroblasts (Dr. Shama Kajiji, Pfizer Inc., Groton, CT), were maintained in DMEM with 10% fetal calf serum (Gibco-BRL, Grand Island, NY or 10% calf serum, respectively. 125I-labeled and unlabeled rat (1-50) and human (1-52) adrenomedullin were from Peninsula laboratories (Belmont, CA). The human A2a adenosine receptor cDNA was obtained from the Garvan Institute (Sydney, Australia). Receptor cloning. Rat lung mRNA (a kind gift of Maynard D. Carty, Pfizer Inc., Groton CT) was primed with random hexamers and reverse-transcribed with Superscript RT (Gibco BRL). The rat adrenomedullin receptor (rAMR) was then cloned as two overlapping PCR fragments (#1Å 196 bp 5*-untranslated region (UTR) and 700 bp coding region and #2Å 139 bp of 3*- UTR and 730 bp coding region). Primer sets, based on published rAMR sequence (2), included:#1F-AGCCAGTCCTTCCCACCCTCTCCCACAGC (forward), #1R-TAAACACTGCGATGAGAGGAAAAG (reverse), #2F-AGCGCCACCAGCACCGAATACG (forward) #2R-TGCGTCCCGAACCTTCTCCCTCTCCCTACA (reverse). PCR was performed according to the Gene-Amp PCR reagent protocol (Perkin Elmer) for 35 cycles at 947C, 30s; 637C, 30s; 727C, 2 min. PCR products were cloned into pCR II TA cloning vector (Invitrogen) and sequence verified on an Applied Biosystems 373-A sequencer. A single nucleotide difference from the published rAMR sequence at amino acid valine-104 ( GTT instead of GTG) was noted and also present in the published mouse sequence (3). The full-length rAMR cDNA was assembled by digesting PCR#1 fragment with EcoRI/Bsu36I, PCR-#2 fragment with Bsu36I/NotI, and ligating these fragments into pcDNA3 digested with EcoRI and NotI. An amino terminal FLAG-epitope tagged rAMR (FLAGrAMR) was generated by PCR as described (4) and encoded MKTIIALSYIFCLVFA (leader sequence)-DYKDDDDA (FLAG epitope)-Met (initiation methionine). A putative human adrenomedullin receptor (hAMR) cDNA clone was identified in a human prostate cDNA library constructed in pBluescript (Incyte Pharmaceuticals, Inc. Palo Alto, CA) based on homology with the rAMR sequence. This clone (pBShAMR) contained 193 bp of 5*-UTR sequence and 784 bp of coding region. The remainder of the hAMR cDNA was cloned by library/nested PCR using a human heart plasmid cDNA library (Gibco BRL). PCR was performed using the Expand Long Template PCR System (Boehringer Mannheim) with specific primer CTCCCCCTCCTGGCAGCGTTACC and T7 primer, TAATACGACTCACTATAGGGAGAGAGCTAT (30 cycles at 947C, 10s; 657C, 30s; 687C, 2 min), followed by a nested PCR with a second specific primer CGGCAGCCAGGACAACCCAAGAG and a poly-A primer, GGCCACGCGTCGACTAGTACTTTTTTTTTTTTTTTTT (30 cycles at 947C, 30s; 607C,30s; 727C, 1 min). The resultant PCR fragment was subcloned into pCR II TA cloning vector to generate pCRIIhAMR. The full-length hAMR cDNA was assembled by three-way ligation of a XhoI(CT filled)/BglI fragment from pBShAMR, a BglI/EcoRI fragment from pCRIIhAMR, and BamHI(GA filled) /EcoRI digested pcDNA3. The sequence was identical to the published hAMR sequence (5). In vitro transcription/translation. In vitro transcription/translation of pcDNA3hAMR was performed with the TNT Coupled Reticulocyte Lysate System followed by S35-labeling with canine pancreatic microsomal membranes (Promega). Samples were analyzed by 8-16% gradient SDS-PAGE gel electrophoresis followed by autoradiography (6). Transfection. COS-7 cells were transfected with 10 mg of DNA in T75 flasks using Superfect (Qiagen) or DOTAP (Boehringer Mannheim). Three hours after transfection, cells were trypsinized and plated into 12-well tissue culture plates. Cells were analyzed for ligand binding, cAMP responses, mRNA expression and immunofluorescence microscopy 48 h after transfection. Northern analysis. Total RNA was prepared from transfected COS-7 cells with a commercial RNA Isolation Kit (Stratagene). Elec-
trophoresed RNA was transferred onto Hybond-N Nylon (Amersham) and membranes were hybridized with probes (EcoRI fragment of rAMR or Pst-HindIII fragment of hAMR) labeled with [32P]dCTP in Prime-a-Gene Labeling System (Promega), in Rapid-Hyb buffer (Amersham) at 687C for 2 h. Blots were washed twice with 21SSC at rt. followed by 0.11SSC for 20 minutes at 687C and exposed 6-18 h to Kodak X-omat film. Ligand binding. Cells were rinsed once with 1 ml of binding buffer (20 mM HEPES, 5 mM MgCl2, 100 mM NaCl, 5 mM KCl, containing 10 mg/ml leupeptin, 0.2 mg/ml bacitracin, and 100 mM PMSF). Binding reactions were carried out for 90 min. at rt. with constant slow rotation in a reaction volume of 0.5 ml binding buffer containing 200 to 350 pM (approximately 200 nCi) 125I-rat or human adrenomedullin. Cells were then washed three times with 1 ml of ice-cold binding buffer, after which 0.5 ml NaOH was added for 10 min. Cell lysate was removed and counted in a gamma counter. Nonspecific binding was determined in the presence of 500 nM unlabeled rat or human adrenomedullin. cAMP assay. Media was removed and 0.5 ml media containing 10 mM HEPES, 20 mM RO-20-1724 and 1 U/ml adenosine deaminanse was added. Following preincubation for 10 min. at 377C, 5 mM forskolin or specific receptor agonist was added for 10 min. Reactions were terminated by the addition of 1.0 N HCl followed by centrifugation at 2000 1 g for 10 minutes. Sample supernatants were removed and cAMP levels determined by commercial radioimmunoassay (New England Nuclear, Inc.). Immunofluorescence microscopy. Flag-tagged rAMR transfected COS-7 cells were cultured on glass coverslips in 6-well plates, rinsed with PBS, fixed in 3.7% formaldehyde for 30 min. and washed three times with PBS-Glycine (10 mM) at rt. Non-specific sites were then blocked for 20 min. at rt. in PBS-Glycine containing 0.5% BSA followed by incubation with anti-FLAG M2 antibody (10 mg/ml, Eastman Kodak Co.) at 377C for 10 min. Coverslips were washed in PBSGlycine and incubated with a 1:50 dilution of FITC-conjugated goat anti-mouse IgG (Pierce ). Samples were visualized and recorded with a Zeiss Axiovert 135 fluorescence microscope/Sony color video/printer- UP5600MD.
RESULTS The rAMR cDNA was isolated from rat lung RNA by reverse transcriptase polymerase chain reaction, sequence verified and subcloned into the mammalian expression vector pcDNA-3 (pcDNArAMR). Preliminary experiments indicated that COS-7 cells treated with transfection reagent Superfect (or DOTAP), plasmid DNA, or media alone did not express specific 125I-rAMbinding sites (data not shown). To verify published rAMR binding studies, COS-7 cells were transiently transfected with pcDNArAMR and Superfect. As shown in Fig. 1A, rAMR plasmid transfected cells (lane b) abundantly expressed specific rAMR mRNA (Ç1.8 Kb) not detected in vector transfected cells (lane a). In agreement with published data, specific 125I-rAMbinding was detected in rAMR transfected cells. However, similar binding was also observed in control vector transfected cells (Fig 2). To further characterize this apparent rAM-binding site, additional members of the CGRP super family were examined. Unlike many reported AM binding sites, 125I-rAM binding to transfected COS-7 cells was not inhibited by unlabeled CGRP (1mM), and direct binding of 125I-CGRP, 125I-AM
833
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Expression of the rat adrenomedullin receptor in transfected COS-7 cells. (A) Northern analysis of rAMR expression in COS-7 cells transfected with (a) pcDNA3, (b) pcDNA3rAMR or (c) pcDNA3FLAGrAMR (1 mg total RNA/lane). (B) Cell surface FLAG expression detected in non-permeabilized COS-7 cells transfected with pcDNA3 (top) or pcDNA3FLAGrAMR (bottom) by immunofluorescence microscopy. Magnification, 250X.
(13-52), 125I-AM (22-41) or 125I-salmon calcitonin (sCT) was not detected (data not shown). To ensure that our transfection protocol resulted in efficient receptor expression, a human adenosine A2a receptor (hA2aR) cDNA was transfected into COS-7 cells. As shown in Fig. 2, abundant hA2aR, which is functionally coupled to a Gs-protein similar to the reported rAMR, was detected by [3H]CGS-21680 ligand binding. In addition, an epitope (FLAG) tagged rAMR was abundantly expressed on the plasma membrane surface of non-permeabilized transfected cells, as detected by immunofluorescence microscopy, in the absence of specific 125I-rAM binding above vector control (Fig. 1B and 2; mRNA expression: Fig. 1A, lane c). These data indicate that the absence of AM binding was not due to deficient cell surface rAMR expression. Finally, specific 125I-rAM-binding was verified in Swiss 3T3 cells which have been shown to express specific AMR in the absence of CGRP receptors (Fig. 2). As an alternative method for detecting rAMR expression, rAMR transfected cells were examined for rAMinduced cAMP responses. This receptor has been re-
ported to be positively coupled to adenylate cyclase. As shown in Fig. 3, specific agonist stimulated cAMP responses were detected in hA2aR transfected COS-7 cells and Swiss 3T3 cells containing endogenous AMR. However, no responses to AM were measurable in rAMR transfected COS-7 cells. In addition, the lack of receptor function was not specific to the COS-7 cell background as no cAMP response or specific AM binding was detected in HEK 293T cells transfected with rAMR (data not shown). A putative hAMR with 73% homology to the rAMR was recently published, although no expression or functional studies were reported (5). To pharmacologically characterize and verify this putative hAMR, a fulllength cDNA was assembled from a human prostate cDNA library clone and reverse transcribed heart RNA (see Materials and Methods). In addition to sequencing, the hAMR open reading frame was confirmed by in vitro transcription/translation, resulting in a single labeled protein migrating at the expected (Ç45 kDa) molecular weight (Fig. 4A, lane a). This translated hAMR was sensitive to temperature-induced aggregation, a
834
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. 125I-labeled Rat adrenomedullin binding to COS-7 cells transfected with rAMR and Swiss 3T3 cells. Rat 125I-AM (rAM) binding to COS-7 cells transfected with control pcDNA3 (vector) or pcDNA3rAMR (rAMR ) and to Swiss 3T3 cells (3T3) was measured as described in Materials and Methods. Shown in right graph inset is [3H]CGS-21680 binding to COS-7 cells transfected with hA2aR. Rat AMR transfection data are Mean { S.E.M. of four experiments performed in triplicate . Rat AMR experiments include data with FLAG-tagged receptors, and vector data include one experiment using hA2aR.
characteristic of some membrane proteins (Fig. 4A, lane b). The hAMR cDNA was subcloned into pcDNA3 and transfected into COS-7 cells which demonstrated abundant Ç1.8 kb hAMR mRNA (Fig. 4B). However, no specific human 125I-AM (hAM) binding was evident above that detected with vector DNA alone (Fig 5). Similarly, no direct binding of 125I-AM (13-52), 125I-AM (22-41) or 125IsCT was observed (data not shown). In addition, no functional cAMP response to hAM was observed in COS-7 cells transfected with hAMR (Fig. 3). Finally, stable or transient expression of hAMR in 293 HEK cells showed no binding or functional responses to AM while cAMP induction by sCT (16) and isoproterenol through endogenous Gs-coupled receptors were detected (data not shown).
ogy to the rAMR also failed to yield binding and functional characteristics expected of an authentic AMR. It is currently unknown why we were not able to confirm the published rAMR binding studies. There have been no reports confirming the identity of the rAMR since the original publication. There are however, several reports in the literature which have initially mis-identified receptor types including the interleukin-8 receptor (17), a putative vasoactive intestinal peptide receptor (18), and a putative NPY receptor (19). The possibility for transfected genes to upregulate functional expression of other endogenous receptors has been suggested as contributing to these erroneous results (18). As noted in the present study, transfection of COS-7 cells with plasmid and transfection agent induced a specific AM binding site. In addition, Han et al. have reported high levels of AM binding in COS-7 cells presumably due to endogenous AM receptors. It is not obvious from the reported rAMR studies whether control vector DNA transfections were performed. Other studies have highlighted the importance of appropriate cell background in the identification of orphan receptors. Aiyar et al. (20) demonstrated binding and functional responses of a putative CGRP-1 receptor only when it was expressed in stable 293 HEK cells, and not in COS-7 cells. It is possible that our particular COS-7 cell line does not allow for effective high-affinity G-protein coupling/function for the AMR although we demonstrated coupling/function for the hA2aR. We also obtained identical results in the 293 HEK cell. In addition, we were not able to demonstrate direct binding of CGRP or a variety of preproAM peptides including the putative AM antagonist (22-52), which should detect
DISCUSSION The data presented in this report suggest that the gene previously described as the rat adrenomedullin receptor is not an authentic adrenomedullin receptor. While we were able to confirm specific 125I-rAM binding to COS-7 cells transfected with the rAMR cDNA, equivalent binding was detected on cells transfected with expression vector alone. Efficient cell surface recombinant receptor expression was verified in parallel immunofluorescence experiments utilizing a FLAG-tagged rAMR. Binding assays were validated utilizing endogenous AMR expressed in Swiss 3T3 cells and specific ligand binding to COS-7 cells transfected with hA2aR. In addition, no rAMR functional coupling to adenylate cyclase was observed in rAMR transfected cells, while cAMP responses were detected in agonist stimulated hA2aR transfected cells and in rAM treated Swiss 3T3 cells. Finally, a putative hAMR which has high homol-
FIG. 3. Adrenomedullin stimulation of cAMP production in COS7 cells transfected with hAMR or rAMR, and Swiss 3T3 cells. Specific cAMP production was measured in COS-7 cells transfected with pcDNArAMR (rAMR), pcDNAhAMR (hAMR), hA2aR, or in 3T3 cells, in response to the appropriate agonist; rat AM (rAM, 100 nM), human AM (hAM, 100 nM) or CGS-21680 (50 nM). Forskolin (5 mM) treatment of COS-7 and 3T3 cells produced Ç40-75 pmol cAMP/ ml/100,000 cells. Data are presented as mean { S.E.M.of triplicate determinations of a representative experiment.
835
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Expression of the putative human adrenomedullin receptor. (A) In vitro transcription/translation of pcDNA3hAMR without (a) and with (b) heating (907C) prior to SDS-PAGE and autoradiographic analysis. (B) Northern analysis (5 mg total RNA/lane) of hAMR expression in COS-7 cells transfected with pcDNA3 (a) or pcDNA3hAMR (b).
low affinity, uncoupled receptors (14). These studies would suggest that the true ligand for this receptor is not likely to be in the CGRP super-family, even though the rAMR and hAMR show distant homology with the RDC1 receptor which has been reported to bind both CGRP and AM (2,5).
The potential mis-identification of the rAMR could have significant impact on the interpretations of studies which have utilized this DNA sequence to study its localization in normal and pathological tissues in situ (21) as well as during embryogenesis (22). It also could misdirect the development of therapeutically relevant
FIG. 5. Displacement of 125I-labeled human adrenomedullin binding on COS-7 cells transfected with hAMR and Swiss 3T3 cells. Total human 125I-AM binding was measured on COS-7 transfected with pcDNA3 control (vector) or pcDNA3hAMR (hAMR) at increasing concentrations (0-500 nM) of unlabeled hAM and to Swiss 3T3 cells (3T3, right panel). Data are presented as mean { S.E.M. of triplicate determinations of a representative experiment. 836
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
agents which act specifically at the AMR and restores the question of whether molecularly distinct AMR’s exist separate from CGRP receptors. In summary, the current study indicates that the published rAMR and hAMR should be reclassified as orphan receptors. ACKNOWLEDGMENTS The authors recognize Ms. Ocean Pellett for tissue culture assistance, Ms. Yevette Clancy for DNA sequencing, Dr. Tom Turi for assistance with immunofluorescence experiments and Dr. John F. Thompson for critical review of the manuscript.
REFERENCES 1. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 2. Kapas, S., Catt, K. J., and Clark, A. J. L. (1995) J. Biol. Chem. 270, 25344–25347. 3. Saeki, Y., Ueno, S., Mizuno, R., Nishimura, T., Fujimura, H., Nagai, Y., and Yanagihara, T. (1993) FEBS Lett. 336, 317–322. 4. Goodman, O. B., Krupnick, J. G., Santini, F., Gurevich, V. V., Penn, R. B., Gagnon, A. W., Keen, J. H., and Benovic, J. L. (1996) Nature 383, 447–450. 5. Hanze, J., Dittrich, K., Dotsch, J., and Rascher, W. (1997) Biochem. Biophys. Res. Commun. 240, 183–188. 6. Laemmli, U. K. (1970) Nature 227, 680–685. 7. Kitamura, K., and Eto, T. (1997) Curr. Op. Nephr. Hyper. 6, 80– 87. 8. Richards, A. M., Nicholls, M. G., Lewis, L., and Lainchbury, J. G. (1996) Clin. Sci. 91, 3–16.
9. Gumusel, B., Chang, J-K, Hao, Q., Hyman, A., and Lippton, H. (1996) Peptides 17, 461–465. 10. Ishizaka, Y., Tanaka, M., Kitamura, K., Kangawa, K., Minamino, N., Matsuo, H., and Eto, T. (1994) Biochem. Biophys. Res. Commun. 200, 642–646. 11. Shimekake, Y., Nagata, K., Ohta, S., Kambayashi, Y., Teraoka, H., Kitamura, K., Eto, T., Kangawa, K., and Matsuo, H. (1995) J. Biol. Chem. 270, 4412–4417. 12. Owji, A. A., Smith, D. M., Coppock, H. A., Morgan, D. G. A., Bhogal, R., Ghatei, M. A., and Bloom, S. R. (1995) Endocrinology 136, 2127–2134. 13. Yeung, V. T. F., Ho, S. K. S., Nicholls, M. G., and Cockram, C. S. (1996) J. Neuro. Res. 46, 330–335. 14. Zimmerman, U., Fischer, J. A., Frei, K., Fischer, A. H., Reinscheid, R. K., and Muff, R. (1996) Brain Res. 724, 238–245. 15. Withers, D. J., Coppock, H. A., Seufferlein, T. S., Smith, D. M., Bloom, S. R., and Rozengurt, E. (1996) FEBS lett. 378, 83–87. 16. Han, Z-Q., Coppock, H. A., Smith, D. M., Noorden, S. V., Makgoba, M. W., Nicholl, C. G., and Legon, S. (1997) J. Mol. Endo. 18, 267–272. 17. Thomas, K. M., Taylor, L., and Navarro, J. (1991) J. Biol. Chem. 266, 14839–14841. 18. Cook, J. S., Wolsing, D. H., Lameh, J., Olson, C. A., Correa, P. E., Sadee, W., Blumenthal, E. M., and Rosenbaum, J. S. (1992) FEBS Lett. 300, 149–152. 19. Herzog, H., Hort, Y. J., Shine, J., and Selbie, L. A. (1993) DNA Cell Biol. 12, 465–471. 20. Aiyar, N., Rand, K., Elshourbagy, N. A., Zeng, Z., Adamou, J. E., Bergsma, D. J., and Li, Y. (1996) J. Biol. Chem. 271, 11325– 11329. 21. Miller, M. J., Martinez, A., Unsworth, E. J., Thiele, C. J., Moody, T. W., Elsasser, T., and Cuttitta, F. (1996) J. Biol. Chem. 271, 23345–23351. 22. Montuenga, L. M., Martinez, A., Milller, M. J., Unsworth, E. J., and Cuttitta, F. (1997) Endocrinology 138, 440–451.
837
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
244, 832–837 (1998)
RC988349
Expression of the Rat Adrenomedullin Receptor or a Putative Human Adrenomedullin Receptor Does Not Correlate with Adrenomedullin Binding or Functional Response Scott P. Kennedy,*,1 Dexue Sun,* Joseph J. Oleynek,† Christopher F. Hoth,* Jimmy Kong,† and Roger J. Hill† *Department of Molecular Sciences and †Department of Cardiovascular and Metabolic Diseases, Pfizer Inc., Groton, Connecticut 06340
Received February 17, 1998
There has been considerable difficulty in defining distinct adrenomedullin (AM) binding sites and function in vivo. However, a rat adrenomedullin receptor (rAMR) and a putative human adrenomedullin receptor (hAMR) have recently been reported. We attempted to confirm and extend the pharmacological characterization of these cloned receptors. COS-7 cells transfected with rAMR or epitope tagged rAMR display abundant rAMR mRNA expression and cell-surface receptor localization. Specific 125I-AM binding is detected in transfected cells; however, similar levels of binding are also detected in cells transfected with vector DNA alone. This AM binding site fails to mediate any changes in cAMP in response to AM. In contrast, Swiss 3T3 cells, expressing specific endogenous AM receptors, display AM binding and functional cAMP responses. Transfection studies performed with the putative hAMR yield similar results. These data suggest that the proposed rAMR and hAMR do not represent authentic adrenomedullin receptors. q 1998 Academic Press
Adrenomedullin (AM) is a 52 amino acid vasoactive peptide member of the calcitonin gene related peptide (CGRP) super family (7,8). It is produced from a variety of vascularized tissues as preproadrenomedullin (185 amino acids) and processed into AM (95-146) as well as other biologically active peptides, including proadrenomedullin (22-41), prodepin (45-92) and adrenotensin (153-185) (7-9). AM displays a wide variety of in vivo activities in the vasculature, adrenal gland, kidney, pituitary gland and brain (7,8). Circulating plasma AM 1 Address correspondence to Dr. Scott P. Kennedy, Pfizer Inc., Department of Molecular Sciences, Eastern Point Rd, Groton, CT 06340. Fax: (860) 441-3783. E-mail: [email protected].
concentrations have been found to be elevated and/or correlated with severity of disease state in essential hypertension, primary aldosteronism, chronic renal failure, congestive heart failure, acute myocardial infarction, and hyperthyroidism (7). It has not been possible to determine if some or all of the actions of AM are mediated through specific AM receptors, since CGRP shares many of the same biological actions and binding sites with AM, and highly selective ligands are not yet available. However, specific AM binding sites have been demonstrated in rat vascular smooth muscle cells (10), bovine endothelial cells (11), rat lung and vascular tissues (12), mouse astrocytes (13), NG108-15 cells (14) and Swiss 3T3 cells (15). The most direct evidence for a specific AM receptor comes from studies reporting the cloning and expression of a putative rat adrenomedullin receptor (rAMR) (2). These studies were performed in transfected COS-7 cells, where this receptor bound 125I-rat adrenomedullin (rAM) with high affinity (KDÅ 8.2 nM) and mediated a cAMP response to rAM (EC50Å 7 nM), but not to CGRP. No further characterization of this receptor has been reported. Very recently, a putative human adrenomedullin receptor (hAMR) cDNA was isolated from lung which displayed 73% homology with the rAMR although no expression or pharmacology studies were reported (5). There clearly exists a need for defining the specific role of AM and its related peptides in normal and pathological states, and a potential therapeutic value in developing pharmacological agonists or antagonists of AM action. Therefore, the goal of the present study was to confirm and extend the pharmacological characterization of the cloned rat and human adrenomedullin receptors. MATERIALS AND METHODS Materials. Restriction endonucleases and other molecular biological reagents were purchased from New England Biolabs unless oth-
832
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
erwise noted. All molecular biological methods were performed according to manufacturer kit protocols or according to Sambrook et al. (1). COS-7 cells (ATCC, Rockville, MD) and Swiss 3T3 fibroblasts (Dr. Shama Kajiji, Pfizer Inc., Groton, CT), were maintained in DMEM with 10% fetal calf serum (Gibco-BRL, Grand Island, NY or 10% calf serum, respectively. 125I-labeled and unlabeled rat (1-50) and human (1-52) adrenomedullin were from Peninsula laboratories (Belmont, CA). The human A2a adenosine receptor cDNA was obtained from the Garvan Institute (Sydney, Australia). Receptor cloning. Rat lung mRNA (a kind gift of Maynard D. Carty, Pfizer Inc., Groton CT) was primed with random hexamers and reverse-transcribed with Superscript RT (Gibco BRL). The rat adrenomedullin receptor (rAMR) was then cloned as two overlapping PCR fragments (#1Å 196 bp 5*-untranslated region (UTR) and 700 bp coding region and #2Å 139 bp of 3*- UTR and 730 bp coding region). Primer sets, based on published rAMR sequence (2), included:#1F-AGCCAGTCCTTCCCACCCTCTCCCACAGC (forward), #1R-TAAACACTGCGATGAGAGGAAAAG (reverse), #2F-AGCGCCACCAGCACCGAATACG (forward) #2R-TGCGTCCCGAACCTTCTCCCTCTCCCTACA (reverse). PCR was performed according to the Gene-Amp PCR reagent protocol (Perkin Elmer) for 35 cycles at 947C, 30s; 637C, 30s; 727C, 2 min. PCR products were cloned into pCR II TA cloning vector (Invitrogen) and sequence verified on an Applied Biosystems 373-A sequencer. A single nucleotide difference from the published rAMR sequence at amino acid valine-104 ( GTT instead of GTG) was noted and also present in the published mouse sequence (3). The full-length rAMR cDNA was assembled by digesting PCR#1 fragment with EcoRI/Bsu36I, PCR-#2 fragment with Bsu36I/NotI, and ligating these fragments into pcDNA3 digested with EcoRI and NotI. An amino terminal FLAG-epitope tagged rAMR (FLAGrAMR) was generated by PCR as described (4) and encoded MKTIIALSYIFCLVFA (leader sequence)-DYKDDDDA (FLAG epitope)-Met (initiation methionine). A putative human adrenomedullin receptor (hAMR) cDNA clone was identified in a human prostate cDNA library constructed in pBluescript (Incyte Pharmaceuticals, Inc. Palo Alto, CA) based on homology with the rAMR sequence. This clone (pBShAMR) contained 193 bp of 5*-UTR sequence and 784 bp of coding region. The remainder of the hAMR cDNA was cloned by library/nested PCR using a human heart plasmid cDNA library (Gibco BRL). PCR was performed using the Expand Long Template PCR System (Boehringer Mannheim) with specific primer CTCCCCCTCCTGGCAGCGTTACC and T7 primer, TAATACGACTCACTATAGGGAGAGAGCTAT (30 cycles at 947C, 10s; 657C, 30s; 687C, 2 min), followed by a nested PCR with a second specific primer CGGCAGCCAGGACAACCCAAGAG and a poly-A primer, GGCCACGCGTCGACTAGTACTTTTTTTTTTTTTTTTT (30 cycles at 947C, 30s; 607C,30s; 727C, 1 min). The resultant PCR fragment was subcloned into pCR II TA cloning vector to generate pCRIIhAMR. The full-length hAMR cDNA was assembled by three-way ligation of a XhoI(CT filled)/BglI fragment from pBShAMR, a BglI/EcoRI fragment from pCRIIhAMR, and BamHI(GA filled) /EcoRI digested pcDNA3. The sequence was identical to the published hAMR sequence (5). In vitro transcription/translation. In vitro transcription/translation of pcDNA3hAMR was performed with the TNT Coupled Reticulocyte Lysate System followed by S35-labeling with canine pancreatic microsomal membranes (Promega). Samples were analyzed by 8-16% gradient SDS-PAGE gel electrophoresis followed by autoradiography (6). Transfection. COS-7 cells were transfected with 10 mg of DNA in T75 flasks using Superfect (Qiagen) or DOTAP (Boehringer Mannheim). Three hours after transfection, cells were trypsinized and plated into 12-well tissue culture plates. Cells were analyzed for ligand binding, cAMP responses, mRNA expression and immunofluorescence microscopy 48 h after transfection. Northern analysis. Total RNA was prepared from transfected COS-7 cells with a commercial RNA Isolation Kit (Stratagene). Elec-
trophoresed RNA was transferred onto Hybond-N Nylon (Amersham) and membranes were hybridized with probes (EcoRI fragment of rAMR or Pst-HindIII fragment of hAMR) labeled with [32P]dCTP in Prime-a-Gene Labeling System (Promega), in Rapid-Hyb buffer (Amersham) at 687C for 2 h. Blots were washed twice with 21SSC at rt. followed by 0.11SSC for 20 minutes at 687C and exposed 6-18 h to Kodak X-omat film. Ligand binding. Cells were rinsed once with 1 ml of binding buffer (20 mM HEPES, 5 mM MgCl2, 100 mM NaCl, 5 mM KCl, containing 10 mg/ml leupeptin, 0.2 mg/ml bacitracin, and 100 mM PMSF). Binding reactions were carried out for 90 min. at rt. with constant slow rotation in a reaction volume of 0.5 ml binding buffer containing 200 to 350 pM (approximately 200 nCi) 125I-rat or human adrenomedullin. Cells were then washed three times with 1 ml of ice-cold binding buffer, after which 0.5 ml NaOH was added for 10 min. Cell lysate was removed and counted in a gamma counter. Nonspecific binding was determined in the presence of 500 nM unlabeled rat or human adrenomedullin. cAMP assay. Media was removed and 0.5 ml media containing 10 mM HEPES, 20 mM RO-20-1724 and 1 U/ml adenosine deaminanse was added. Following preincubation for 10 min. at 377C, 5 mM forskolin or specific receptor agonist was added for 10 min. Reactions were terminated by the addition of 1.0 N HCl followed by centrifugation at 2000 1 g for 10 minutes. Sample supernatants were removed and cAMP levels determined by commercial radioimmunoassay (New England Nuclear, Inc.). Immunofluorescence microscopy. Flag-tagged rAMR transfected COS-7 cells were cultured on glass coverslips in 6-well plates, rinsed with PBS, fixed in 3.7% formaldehyde for 30 min. and washed three times with PBS-Glycine (10 mM) at rt. Non-specific sites were then blocked for 20 min. at rt. in PBS-Glycine containing 0.5% BSA followed by incubation with anti-FLAG M2 antibody (10 mg/ml, Eastman Kodak Co.) at 377C for 10 min. Coverslips were washed in PBSGlycine and incubated with a 1:50 dilution of FITC-conjugated goat anti-mouse IgG (Pierce ). Samples were visualized and recorded with a Zeiss Axiovert 135 fluorescence microscope/Sony color video/printer- UP5600MD.
RESULTS The rAMR cDNA was isolated from rat lung RNA by reverse transcriptase polymerase chain reaction, sequence verified and subcloned into the mammalian expression vector pcDNA-3 (pcDNArAMR). Preliminary experiments indicated that COS-7 cells treated with transfection reagent Superfect (or DOTAP), plasmid DNA, or media alone did not express specific 125I-rAMbinding sites (data not shown). To verify published rAMR binding studies, COS-7 cells were transiently transfected with pcDNArAMR and Superfect. As shown in Fig. 1A, rAMR plasmid transfected cells (lane b) abundantly expressed specific rAMR mRNA (Ç1.8 Kb) not detected in vector transfected cells (lane a). In agreement with published data, specific 125I-rAMbinding was detected in rAMR transfected cells. However, similar binding was also observed in control vector transfected cells (Fig 2). To further characterize this apparent rAM-binding site, additional members of the CGRP super family were examined. Unlike many reported AM binding sites, 125I-rAM binding to transfected COS-7 cells was not inhibited by unlabeled CGRP (1mM), and direct binding of 125I-CGRP, 125I-AM
833
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Expression of the rat adrenomedullin receptor in transfected COS-7 cells. (A) Northern analysis of rAMR expression in COS-7 cells transfected with (a) pcDNA3, (b) pcDNA3rAMR or (c) pcDNA3FLAGrAMR (1 mg total RNA/lane). (B) Cell surface FLAG expression detected in non-permeabilized COS-7 cells transfected with pcDNA3 (top) or pcDNA3FLAGrAMR (bottom) by immunofluorescence microscopy. Magnification, 250X.
(13-52), 125I-AM (22-41) or 125I-salmon calcitonin (sCT) was not detected (data not shown). To ensure that our transfection protocol resulted in efficient receptor expression, a human adenosine A2a receptor (hA2aR) cDNA was transfected into COS-7 cells. As shown in Fig. 2, abundant hA2aR, which is functionally coupled to a Gs-protein similar to the reported rAMR, was detected by [3H]CGS-21680 ligand binding. In addition, an epitope (FLAG) tagged rAMR was abundantly expressed on the plasma membrane surface of non-permeabilized transfected cells, as detected by immunofluorescence microscopy, in the absence of specific 125I-rAM binding above vector control (Fig. 1B and 2; mRNA expression: Fig. 1A, lane c). These data indicate that the absence of AM binding was not due to deficient cell surface rAMR expression. Finally, specific 125I-rAM-binding was verified in Swiss 3T3 cells which have been shown to express specific AMR in the absence of CGRP receptors (Fig. 2). As an alternative method for detecting rAMR expression, rAMR transfected cells were examined for rAMinduced cAMP responses. This receptor has been re-
ported to be positively coupled to adenylate cyclase. As shown in Fig. 3, specific agonist stimulated cAMP responses were detected in hA2aR transfected COS-7 cells and Swiss 3T3 cells containing endogenous AMR. However, no responses to AM were measurable in rAMR transfected COS-7 cells. In addition, the lack of receptor function was not specific to the COS-7 cell background as no cAMP response or specific AM binding was detected in HEK 293T cells transfected with rAMR (data not shown). A putative hAMR with 73% homology to the rAMR was recently published, although no expression or functional studies were reported (5). To pharmacologically characterize and verify this putative hAMR, a fulllength cDNA was assembled from a human prostate cDNA library clone and reverse transcribed heart RNA (see Materials and Methods). In addition to sequencing, the hAMR open reading frame was confirmed by in vitro transcription/translation, resulting in a single labeled protein migrating at the expected (Ç45 kDa) molecular weight (Fig. 4A, lane a). This translated hAMR was sensitive to temperature-induced aggregation, a
834
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. 125I-labeled Rat adrenomedullin binding to COS-7 cells transfected with rAMR and Swiss 3T3 cells. Rat 125I-AM (rAM) binding to COS-7 cells transfected with control pcDNA3 (vector) or pcDNA3rAMR (rAMR ) and to Swiss 3T3 cells (3T3) was measured as described in Materials and Methods. Shown in right graph inset is [3H]CGS-21680 binding to COS-7 cells transfected with hA2aR. Rat AMR transfection data are Mean { S.E.M. of four experiments performed in triplicate . Rat AMR experiments include data with FLAG-tagged receptors, and vector data include one experiment using hA2aR.
characteristic of some membrane proteins (Fig. 4A, lane b). The hAMR cDNA was subcloned into pcDNA3 and transfected into COS-7 cells which demonstrated abundant Ç1.8 kb hAMR mRNA (Fig. 4B). However, no specific human 125I-AM (hAM) binding was evident above that detected with vector DNA alone (Fig 5). Similarly, no direct binding of 125I-AM (13-52), 125I-AM (22-41) or 125IsCT was observed (data not shown). In addition, no functional cAMP response to hAM was observed in COS-7 cells transfected with hAMR (Fig. 3). Finally, stable or transient expression of hAMR in 293 HEK cells showed no binding or functional responses to AM while cAMP induction by sCT (16) and isoproterenol through endogenous Gs-coupled receptors were detected (data not shown).
ogy to the rAMR also failed to yield binding and functional characteristics expected of an authentic AMR. It is currently unknown why we were not able to confirm the published rAMR binding studies. There have been no reports confirming the identity of the rAMR since the original publication. There are however, several reports in the literature which have initially mis-identified receptor types including the interleukin-8 receptor (17), a putative vasoactive intestinal peptide receptor (18), and a putative NPY receptor (19). The possibility for transfected genes to upregulate functional expression of other endogenous receptors has been suggested as contributing to these erroneous results (18). As noted in the present study, transfection of COS-7 cells with plasmid and transfection agent induced a specific AM binding site. In addition, Han et al. have reported high levels of AM binding in COS-7 cells presumably due to endogenous AM receptors. It is not obvious from the reported rAMR studies whether control vector DNA transfections were performed. Other studies have highlighted the importance of appropriate cell background in the identification of orphan receptors. Aiyar et al. (20) demonstrated binding and functional responses of a putative CGRP-1 receptor only when it was expressed in stable 293 HEK cells, and not in COS-7 cells. It is possible that our particular COS-7 cell line does not allow for effective high-affinity G-protein coupling/function for the AMR although we demonstrated coupling/function for the hA2aR. We also obtained identical results in the 293 HEK cell. In addition, we were not able to demonstrate direct binding of CGRP or a variety of preproAM peptides including the putative AM antagonist (22-52), which should detect
DISCUSSION The data presented in this report suggest that the gene previously described as the rat adrenomedullin receptor is not an authentic adrenomedullin receptor. While we were able to confirm specific 125I-rAM binding to COS-7 cells transfected with the rAMR cDNA, equivalent binding was detected on cells transfected with expression vector alone. Efficient cell surface recombinant receptor expression was verified in parallel immunofluorescence experiments utilizing a FLAG-tagged rAMR. Binding assays were validated utilizing endogenous AMR expressed in Swiss 3T3 cells and specific ligand binding to COS-7 cells transfected with hA2aR. In addition, no rAMR functional coupling to adenylate cyclase was observed in rAMR transfected cells, while cAMP responses were detected in agonist stimulated hA2aR transfected cells and in rAM treated Swiss 3T3 cells. Finally, a putative hAMR which has high homol-
FIG. 3. Adrenomedullin stimulation of cAMP production in COS7 cells transfected with hAMR or rAMR, and Swiss 3T3 cells. Specific cAMP production was measured in COS-7 cells transfected with pcDNArAMR (rAMR), pcDNAhAMR (hAMR), hA2aR, or in 3T3 cells, in response to the appropriate agonist; rat AM (rAM, 100 nM), human AM (hAM, 100 nM) or CGS-21680 (50 nM). Forskolin (5 mM) treatment of COS-7 and 3T3 cells produced Ç40-75 pmol cAMP/ ml/100,000 cells. Data are presented as mean { S.E.M.of triplicate determinations of a representative experiment.
835
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Expression of the putative human adrenomedullin receptor. (A) In vitro transcription/translation of pcDNA3hAMR without (a) and with (b) heating (907C) prior to SDS-PAGE and autoradiographic analysis. (B) Northern analysis (5 mg total RNA/lane) of hAMR expression in COS-7 cells transfected with pcDNA3 (a) or pcDNA3hAMR (b).
low affinity, uncoupled receptors (14). These studies would suggest that the true ligand for this receptor is not likely to be in the CGRP super-family, even though the rAMR and hAMR show distant homology with the RDC1 receptor which has been reported to bind both CGRP and AM (2,5).
The potential mis-identification of the rAMR could have significant impact on the interpretations of studies which have utilized this DNA sequence to study its localization in normal and pathological tissues in situ (21) as well as during embryogenesis (22). It also could misdirect the development of therapeutically relevant
FIG. 5. Displacement of 125I-labeled human adrenomedullin binding on COS-7 cells transfected with hAMR and Swiss 3T3 cells. Total human 125I-AM binding was measured on COS-7 transfected with pcDNA3 control (vector) or pcDNA3hAMR (hAMR) at increasing concentrations (0-500 nM) of unlabeled hAM and to Swiss 3T3 cells (3T3, right panel). Data are presented as mean { S.E.M. of triplicate determinations of a representative experiment. 836
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC
Vol. 244, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
agents which act specifically at the AMR and restores the question of whether molecularly distinct AMR’s exist separate from CGRP receptors. In summary, the current study indicates that the published rAMR and hAMR should be reclassified as orphan receptors. ACKNOWLEDGMENTS The authors recognize Ms. Ocean Pellett for tissue culture assistance, Ms. Yevette Clancy for DNA sequencing, Dr. Tom Turi for assistance with immunofluorescence experiments and Dr. John F. Thompson for critical review of the manuscript.
REFERENCES 1. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 2. Kapas, S., Catt, K. J., and Clark, A. J. L. (1995) J. Biol. Chem. 270, 25344–25347. 3. Saeki, Y., Ueno, S., Mizuno, R., Nishimura, T., Fujimura, H., Nagai, Y., and Yanagihara, T. (1993) FEBS Lett. 336, 317–322. 4. Goodman, O. B., Krupnick, J. G., Santini, F., Gurevich, V. V., Penn, R. B., Gagnon, A. W., Keen, J. H., and Benovic, J. L. (1996) Nature 383, 447–450. 5. Hanze, J., Dittrich, K., Dotsch, J., and Rascher, W. (1997) Biochem. Biophys. Res. Commun. 240, 183–188. 6. Laemmli, U. K. (1970) Nature 227, 680–685. 7. Kitamura, K., and Eto, T. (1997) Curr. Op. Nephr. Hyper. 6, 80– 87. 8. Richards, A. M., Nicholls, M. G., Lewis, L., and Lainchbury, J. G. (1996) Clin. Sci. 91, 3–16.
9. Gumusel, B., Chang, J-K, Hao, Q., Hyman, A., and Lippton, H. (1996) Peptides 17, 461–465. 10. Ishizaka, Y., Tanaka, M., Kitamura, K., Kangawa, K., Minamino, N., Matsuo, H., and Eto, T. (1994) Biochem. Biophys. Res. Commun. 200, 642–646. 11. Shimekake, Y., Nagata, K., Ohta, S., Kambayashi, Y., Teraoka, H., Kitamura, K., Eto, T., Kangawa, K., and Matsuo, H. (1995) J. Biol. Chem. 270, 4412–4417. 12. Owji, A. A., Smith, D. M., Coppock, H. A., Morgan, D. G. A., Bhogal, R., Ghatei, M. A., and Bloom, S. R. (1995) Endocrinology 136, 2127–2134. 13. Yeung, V. T. F., Ho, S. K. S., Nicholls, M. G., and Cockram, C. S. (1996) J. Neuro. Res. 46, 330–335. 14. Zimmerman, U., Fischer, J. A., Frei, K., Fischer, A. H., Reinscheid, R. K., and Muff, R. (1996) Brain Res. 724, 238–245. 15. Withers, D. J., Coppock, H. A., Seufferlein, T. S., Smith, D. M., Bloom, S. R., and Rozengurt, E. (1996) FEBS lett. 378, 83–87. 16. Han, Z-Q., Coppock, H. A., Smith, D. M., Noorden, S. V., Makgoba, M. W., Nicholl, C. G., and Legon, S. (1997) J. Mol. Endo. 18, 267–272. 17. Thomas, K. M., Taylor, L., and Navarro, J. (1991) J. Biol. Chem. 266, 14839–14841. 18. Cook, J. S., Wolsing, D. H., Lameh, J., Olson, C. A., Correa, P. E., Sadee, W., Blumenthal, E. M., and Rosenbaum, J. S. (1992) FEBS Lett. 300, 149–152. 19. Herzog, H., Hort, Y. J., Shine, J., and Selbie, L. A. (1993) DNA Cell Biol. 12, 465–471. 20. Aiyar, N., Rand, K., Elshourbagy, N. A., Zeng, Z., Adamou, J. E., Bergsma, D. J., and Li, Y. (1996) J. Biol. Chem. 271, 11325– 11329. 21. Miller, M. J., Martinez, A., Unsworth, E. J., Thiele, C. J., Moody, T. W., Elsasser, T., and Cuttitta, F. (1996) J. Biol. Chem. 271, 23345–23351. 22. Montuenga, L. M., Martinez, A., Milller, M. J., Unsworth, E. J., and Cuttitta, F. (1997) Endocrinology 138, 440–451.
837
AID
BBRC 8349
/
694d$$$741
03-13-98 13:51:31
bbrcg
AP: BBRC