(CANCER RESEARCH 58. 1866-1871. May I. 1998]

Expression of All Known Vasopressin Receptor Subtypes by Small Cell TumorsImplies a Multifaceted Role for this Neuropeptide1

William G. North,2 Michael J. Fay,3 Kenneth A. Longo, and Jiniin Du

Department of Physiology. Diminuitili Medicai School, Lebanon, New Hampshire 03756

ABSTRACT

Vasopressin is one of several small neuropeptides that are reported tobe autocrine growth factors for small cell carcinoma of the lung (SCCL).It has been assumed that this peptide exercises its ntitogenic influencesthrough the Vasopressin V,., receptor, and we have previously demonstrated that this receptor is expressed by classical and variant SCCL.Activation of the vasopressin Vla receptor produces changes in phospho-lipases C, D, and \2, in protein kinase C, and in Qr* mobilization. This

study demonstrates that SCCL cells express not only vasopressin V,.,receptors but also iiiRNAs and proteins representing normal Vlb receptors and V2 receptors. They were also shown to express mRNA for ahuman form of the putative receptor rabbit vasopressin-activated calcium-mobilizing receptor (VACM-1). Additionally, SCCL tumor cells were

found to express mRNA and protein representing a possible nonfunctional, shortened, "diabetic" form of the vasopressin V., receptor that is

the product of incomplete posttranscriptional splicing. At least four ofthese five vasopressin receptors were produced by cell lines exemplifyingclassical and variant forms of SCCL. No differences in the sequences forthe V, receptors between classical and variant SCCL were found. However, although the nature and expression of both vasopressin V, receptorsand human VACM are apparently unaffected by dedifferentiation inSCCL, only the abnormal (and probably nonfunctional) form of the V,receptor could be demonstrated in variant cell line NCI H82. Functionalengagement of vasopressin V, receptors is reported to produce rises incAMP and activation of protein kinase A, whereas stimulation of V,breceptors is believed to produce similar changes to those produced by VIareceptors, i.e., activation of phospholipases and of protein kinase C.Stimulation of VACM receptors raises intracellular free C'ir ' through

currently unknown but phosphoinositide-independent mechanisms. The

presence of all known vasopressin receptors that are, together, potentiallycapable of inducing several different transduction cascades in small celltumor cells suggests that this peptide serves a multifaceted role in tumorphysiology.

INTRODUCTION

Vasopressin is a nonapeptide amide that is produced by magnocel-

lular neurons of the hypothalamus and serves as the antidiuretichormone in man (1). This physiological role is exercised by thepeptide through V2 subtype receptors, whereas vasopressin aids in theregulation of blood pressure by acting through V,., subtype receptorsand functions in the regulation of the adrenal-stress axis by acting

through V,b subtype receptors. All three of these vasopressin receptors are part of a neuropeptide receptor superfamily comprised ofstructurally related proteins with seven transmembrane domains (2,3). Although these three receptor subtypes are thought to represent allof the receptors for vasopressin, an additional and nonhomologous

Received 11/25/97; accepted 3/4/98.The costs of publication of this article were defrayed in pan by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported in part by United States Public Health Service Award CA

19613. M. J. F. was supported as a postdoctoral fellow on NIH Training Grant T32 DK07508. and K. A. L. was supported by fellowships from the Dolores Zhorhab LiebmannFoundation and the Stephen J. Ryan Foundation.

2 To whom requests for reprints should be addressed, at Department of Physiology.

Dartmouth Medical School, Borwell 752E. 1 Medical Center Drive, Lebanon, NH 03756.3 Present address: Department of Pharmacology, Midwestern University, Downers

Grove, IL 60515.

protein, named VACM-1,4 has recently been identified as a putative

receptor of rabbit medullary kidney cells (4). Unlike the other threereceptors, VACM-1 appears to possess only one transmembrane do

main. We have provided a substantial body of evidence showing thevasopressin gene is expressed by all small cell tumors (5), andvasopressin is reported to be a mitogenic factor for SCCL (6). Recently, we demonstrated functional vasopressin V,a receptors arepresent, not only on classical forms of SCCL but also on variant formsthat have undergone some degree of dedifferentiation (7). We havealso shown classical SCCL can express a human counterpart of theputative VACM-1 receptor (8). In this study, we examined classical

and variant forms of SCCL to determine whether they can expressmore than one of these and other vasopressin receptor subtypes.5

MATERIALS AND METHODS

Cell Culture and Human Tissues. The classical SCCL cell line NCI H345and the variant cell line NCI H82 were purchased from the American TypeTissue Culture Collection (Rockville, MD). Both cell lines have been shown toexpress vasopressin gene-related products by immunocytochemistry (9) and byWestern analysis.*1 Cells were maintained in RPMI 1640 (Sigma Chemical Co.,

St. Louis. MO) supplemented with 10% fetal bovine serum (Hyclone, Logan.UT) at cell densities of 105-106/ml in a humidified atmosphere of 5% CO, at37°C.Human liver and lung tissue specimens, frozen in liquid nitrogen within

10 min after surgery, were obtained from the Cooperative Human TissueNetwork (Philadelphia. PA).

RNA Isolation. Poly(A)+ RNA was isolated from cells and tissues byoligo(dT) cellulose chromatography using the method of Badley (10), asdescribed previously (11). Total RNA was isolated from normal liver andkidney tissue using TRIzol reagent (Life Technologies. Inc., Gaithersburg,MD) prior to poly(A)+ extraction.

RT-PCR. Poly(A)+ RNA (1-5 ¿ig)from tumor cells, liver, or lung was

added into a Superscript preamplification system (Life Technologies. Inc.) forthe synthesis of the first cDNA strand by using an oligo(dT) primer and reversetranscriptase (400 units), and the product used directly for PCR. PCR with theGeneAmp PCR reagent kit (Perkin-Elmer. Foster City, CA) was performed in

a thermocycler (ERICOMP. San Diego. CA). The reaction mixtures wereoverlaid with 50 ¿¿Iof mineral oil and subjected to an initial denaturation at97°Cfor 8 min. followed by 30 cycles comprising a denaturation step at 95°Cfor 30 s, an annealing step (primers to the template) at 58°Cfor 1 min 30 s, andan extension step at 72°Cfor 1 min 30 s. At the completion of the cyclingreaction, an additional extension step at 72°Cfor 10 min was performed. The

primers selected for PCR and sequencing of V,b and V, receptors are summarized in Fig. 1, whereas those used for V,., receptors and human VACM arepresented elsewhere (7, 8).7 These primers were synthetic 18, 19, 20. and 22

oligomers designed to yield overlapping PCR products of 454, 669, and 858 bpfor Vu receptors; synthetic 18 and 20 oligomers yielding overlapping PCRproducts of 903, 768, 535, and 589 bp for Vlb receptors; synthetic 20 oli-

gomers providing overlapping PCR products of 512 and 862 bp for V2receptors; and synthetic 20 oligomers yielding overlapping PCR products of674 and 193 bp for human VACM (8). The PCR products for V, and V,receptors collectively spanned the entire reading frame of the receptor mRNAs.PCR products were extracted with an equal volume of chloroform and exam-

4The abbreviations used are: VACM. vasopressin-activated calcium-mobilizing pro

tein; SCCL, small cell carcinoma of the lung; RT-PCR, reverse transcriptase-PCR.5 The nucleotide sequences reported in this paper have been submitted to GenBank

(accession nos. AF030625, AF030512, AF030626, AF017061, and AF032388).6 M. J. Fay, J. Du, and W. G. North, unpublished data.7 K. A. Longo, M. J. Fay, J. Du, and W. G. North, unpublished data.

1866

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

VASOPRESSIN RECEPTORS IN SMALL CELL CARCINOMA

¡nedon 2% agarose gels. For some PCRs, products were labeled using 10¿iCi/yn![a-'2P]dCTP (3000 Ci/mmol; Dupont/NEN, Boston, MA), and these

products were isolated by 4% PAGE at 100 V for 4 h in Tris-borate-EDTAbuffer. DNase-free RNase (Boehringer Mannheim, Indianapolis, IN) andRNase-free DNase (Life Technologies, Inc.) were used in some reactions to

validate that the products of PCR did not result from genomic DNA contamination.

Cloning and Sequencing. PCR products (1 fj.1,4-12 ng) of vasopressin

receptors were ligated into a pCR vector, and 2 /nl of the ligation mixture weretransformed in One Shot Competent Cells using a TA Cloning Kit (Invitrogen.San Diego, CA). Plasmid clones were prepared with a Wizard Minipreps DNApurification system (Promega, Madison, WI) and screened by £coRIdigestionand agarose gel electrophoresis. At least two positive clones of each PCRproduct were chosen for double-strand cDNA sequencing with a DyeDeoxyTerminator Cycle Sequencing Kit (Applied Biosystems/Perkin-Elmer). The

primers chosen for PCR amplifications, as described in Fig. 1, and vectoruniversal primers (Ml3 Forward, M13 Reverse, and T7) together served assequencing primers. The protocol for sequencing cloned cDNA ( 1 p.g) wasmodified as follows: 97°Cfor 2 min; 30 cycles of 95°Cfor 30 s, 58°Cfor 1min 30 s, and 72°Cfor 1 min 30 s; and 72°Cextension for 10 min. The

products were separated by phenol-chloroform fractionation and precipitated

with 100% ethanol. Automated DNA sequencing was performed using a model373 DNA Sequencer from Applied Biosystems.

Northern Blot Analysis. Northern blot analysis was performed with 5 /igof poly(A)+ RNA from NCI H82 cells that was first denatured and thenfractionated on a 1.2% formaldehyde-agarose gel. Separated products were

transferred to nitrocellulose (Nitropure, MSI, Westboro, MA) with a 10x SSCtransfer buffer, and the membrane was vacuum-dried at 80°Cfor 2 h. Prehy-bridization was performed at 37°Cin 6X SSC. 50% formamide, 5X Den-hardt's solution, 100 mg/ml sheared salmon sperm DNA, and 0.1% SDS (Life

Technologies. Inc.) for 4 h. Two 48-bp synthesized oligodeoxynucleotide

probes (Keystone. Menlo Park, CA). each in the amount of 25 ng. were usedto identify mRNA for human V,a receptor. Single and specific probes of 768bp and 862 bp obtained from PCR were used in Northern analyses for V,breceptor mRNA and V-, receptor mRNA. respectively. Probes were [12P]dCTP-

labeled (3000 Ci/mmol: Dupont/NEN) using the DECA prime II DNA labelingkit for the three large probes and the KinaseMax DNA kit for the 48-bp probes(Ambion, Austin, TX) and hybridized with the membrane at 42°Cfor 18 h, as

reported elsewhere (7). The membrane was washed twice at ambient temperature, first with 2 X SSC solution containing 0.1 % SDS and then in a 0.1 X SSCsolution containing 0.1% SDS. It was finally washed twice at 56°Cin 0.1X

SSC containing 0.1% SDS and subjected to autoradiographic analysis.Generation of Polyclonal Antibodies. Rabbit polyclonal antibodies to the

human vasopressin V,a, Vlb, and V, receptors and an abnormal form of V,receptors were generated using procedures similar to those described previously (12, 13). Decapeptide amides, nonapeptide amides, or decapeptidesrepresenting 8-10-amino acid unique sequences in each of the four receptorforms with an added NH2-terminal tyrosine (to permit iodide labeling), when

this amino acid was not already present in the sequence chosen, were used asantigens. Specifically, the antigen for vasopressin Vhl receptor was the de-capeptide. YVNNVTKARD-NH,, which comprised the 9-amino acid unique

sequence from 194 to 202; the antigen for vasopressin Vlb receptor wasYPLWDANPTP-NH,. which represented the unique sequence from 5 to 13;the antigen for vasopressin V2 receptor was YVEGGSGVTD-NH2, represent

ing the sequence from 183 to 191: and the antigen for the putative abnormalreceptor was YLEGGCSRG-OH, representing its COOH-terminal (302-309)

octapeptide sequence. These peptides were synthesized using a SYNERGYmodel 432A Applied Biosystems Peptide Synthesizer and the Merrifieldsolid-phase procedure. The prepared peptides were first complexed in a molar

ratio of 100:1 to bovine thyroglobulin with glutaraldehyde. and the complexwas dialyzed (3 X 1000 distilled water), lyophilized, and restored in physiological saline at a peptide equivalent concentration of 1 mg/ml. These solutionsof the complex were emulsified in a ratio of 1:1 with Fruend's adjuvant

(complete for initial immunization, incomplete for boosts), and immunizationswere performed at multiple intradermal sites along the backs of animals. Thetiters of harvested antisera were evaluated by examining their capacity to bindto individual iodinated peptides (12, 13). Antibodies from antisera (Vivian forV,a receptor, Bivily for V|b receptor. Rocket for V2 receptor, and Abner for theabnormal V, receptor) obtained in this manner were isolated as IgG2b fractions

by protein A-Sepharose (Sigma) chromatography for use in Western blot

analysis at concentrations of 10 jug/ml.Western Analysis. Samples for Western analysis were prepared by soni

cating cells (0.4-1.2 X IO7) or tissue (100 trig) in 1.Oml of lysis buffer 110 mM

Tris-HCl, 5 mM EDTA. 50 mM NaCl. 30 mM NaPO4. 50 mM NaF, 0.1 mM

Na,VO4, 1 mM phenylmethylsultbnyl fluoride, 0.2 units/ml aprotinin. and0.1% Triton X-100 (pH 7.6)]. Homogenates were centrifuged for 2 min at

12,000 X g and ambient temperature. Soluble extracts were mixed 1:1 with 2XSDS/Tris-HCl sample buffer (pH 8.7) containing 50 mM DTT, and proteinswere reduced by heating at 100°Cfor 5 min. These proteins were then

subjected to SDS-PAGE on 12% gels at pH 9.3 using the method of Laemmli(14). Separated proteins were electrophoretically transferred in 20 mM Tris-

glycine to Immobilon polyvinylidene difluoride membranes. These membranetransfers were dried and incubated for 16 h at 4°C,with a 10 /ig/ml of a protein

A-purified. rabbit polyclonal antibody preparation raised against peptide rep

resenting Vla, Vlh. or V-, receptors or V2 abnormal receptors (see above) inTris-HCl buffer [10 mM Tris-HCl. 100 mM NaCl. 5% BSA, and 0.1% Tween

20 (pH 7.4)]. Following three washings in buffer (pH 7.4), membrane transferswere incubated with a goat antirabbit IgG-horseradish peroxidase conjugate for1 h. Each was removed by washing (three) in Tris-HCl buffer (pH 7.4).

Transfers were sometimes placed in a 1:50 dilution of 33% H2O2 in methanolfor 30 min before blocking (to destroy endogenous peroxidase activity in thesample). Immunoreactive proteins in peroxidase-treated transfers were visual

ized using an ECL Western Blotting Detection System (Amersham LifeSciences, Arlington Heights, IL) with exposure of X-ray film from 30 to 120 s.Prestained SDS-PAGE standard proteins (Sigma) were used as molecular size

markers.

RESULTS

RT-PCR, Cloning, and DNA Sequencing. Using forward andreverse primers selected to provide overlapping sequences coveringthe entire reading frames of vasopressin Vla and Vlh mRNAs, RT-PCR of poly(A)+ RNA preparations from cells of human small cell

tumor lines NCI H345 and NCI H82 and from human liver yielded, ineach case, a single product of the expected size predicted from thecDNA for the human forms of these receptors derived from liver (15)and blood vessels (16) or pituitary (17, 18). For each receptor, theseproducts presumably spanned the intronic segments of the genes anddemonstrated, by the absence of these introns, that all of the productswere generated from RNA and not DNA template. Data for V,.,receptors has been presented elsewhere (7). Four of the productsobtained for the V,b receptor (903, 768, 535, and 589 bp) are shownin Fig. 1. Using forward and reverse primers selected to provideoverlapping sequences covering the entire sequence of vasopressin V-,mRNA (Fig. 1), RT-PCR of poly(A)+ RNA preparations from NCI

H345, NCI H82, and lungs yielded, in some cases, single products ofa size predicted from published data on cDNA forms of this receptorfrom kidneys (19) and lung (20). However, for both cell lines, whenprimers spanning intron 2 were used, a product larger by the size ofintron 2 (—100 bp) than the size of 862 bp, predicted from thestructure of V2 receptor mRNA, was obtained. Although normal-sized

products and the larger product were obtained from classical cell lineNCI H345, only the larger product could be demonstrated for NCIH82 variant cells (Fig. 2b). This larger form was not present in kidneytissues but seemed to be present as a minor component in normal lung(20). Primers selected from rabbit VACM-1 to amplify two regions of

a human VACM generated, from NCI H82 cells, single products of674 and 193 bp (Fig. 2r), sizes predicted from both the rabbitsequence and the sequence from classical SCCL (8). In all cases,RT-PCR production of cDNA products was unaffected by prior

DNase treatment of poly(A)+ RNA preparations, whereas no products were generated when initial treatment with RNase was performedor when reverse transcriptase was omitted from the reaction mixture.The RT-PCR products representing Vlh and V2 receptors of NCI H82

1867

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

VASOPRESSIN RECEPTORS IN SMALL CELL CARCINOMA

era B11

Br2 BrO

fOr1,903bp

MM,768bp

(2r2, 535bp

(2rO, 589bp

PrimerForwardFOFlF2ReverseRlR2RONuclcolide-123-

-10412-29736-753763-7801252-12711308-1325Length201818182018Sequence5'-tccctgtcattctcaacgct-3'5'-cctctgtgggatgccaac-3'5'-acttgggacaggccctca-3'5

'-ccgagt2gtggcagctaa-3'5'-aagatgatggtctcagcggt-3'5'-gagaacctccactagtcc-3'Strand+++..•

B

Exonl25bp

Exon2 Exon 3203bp

-AAA3'

512bp

862bp

DirectionForwardReversePrimerAI

ACBDLenglh(bp)2020202020Nucleolidc40-5972-91379-398564-583

1221-1240Sequence(5'

to3')caggccclcagaacacctgc

atgctcatggcgtccaccacaccgcttccgtgggccagatctgagaaggagcgagaaggc

ggaaggcagclgagcttclc

Fig. 1. Synthetic primers selected for RT-PCR and sequencing from lung and the

SCCL cell lines NCI H345 and NCI H82 of the vasopressin Vlh receptor (Ai and thevasopressin V, receptor (ß).These oligomers generated overlapping PCR products thatcollectively spanned beyond the entire reading frame of the receptor mRNAs and the twointrons of the genes. Sequences obtained from these products were identical to sequencesof mRNAs published earlier for normal human tissues by us and others (2. 17-20), and inaddition, a partially spliced V2 mRNA that retains intron 2 of the gene and can give riseto a COOH-terminally truncated receptor. This last form was the only one apparent in thevariant NCI H82 cell line.

are represented in Fig. 2. Cloning and sequencing of all V,a- andVlb-related products and V2-related products of predicted size showed

them to collectively provide a complete characterization of human VlamRNA for the two cell lines from —23at the 5' end (23 bases priorto the reading frame) through 1224 at the 3' end (18 bases beyond thereading frame), for V]h mRNA from 123 bases beyond the 5' end to52 bases beyond the 3' end, and for V2 mRNA from 32 bases beyondthe 5' end to 53 bases beyond the 3' end. The sequences for the

vasopressin V, receptor mRNAs of the two cell lines (NCI H345 andNCI H82) were identical and had exact sequence homology with thesequence of human Vla mRNA published by Thibonnier et al. (15) forhuman liver and the sequence of human Vlh mRNA published bySugimoto et al. (17). One sequence of human V-, receptor mRNA

from NCI H345 was identical to that published by us (20) for humanlung and the genomic structure for the receptor earlier published byBirnbaumer et al. (19). Enlarged products obtained with V2 receptorprimers from both classical and variant cells were found throughsequencing to contain the entire 106 bases of intron 2 in addition tosequence for V2 mRNA. If these larger forms (as it seems) representan incompletely spliced form of V2 receptor mRNA, such an mRNAshould give rise to a COOH-terminally truncated receptor lacking

transmembrane domain seven and having a unique COOH-terminal

structure.Northern Analysis. Blot hybridization of poly(A)+ RNA from

NCI H82 using the two nonredundant 48-bp cDNA probes for V)a

receptor revealed a single band, representing the mRNA, at approximately 1.9 kb, a size in excellent agreement with that of the mRNAfor rat vasopressin V,a receptor from liver (15). A single band of —5.2

kb was obtained in Northern blots using the probe specific for V,breceptor mRNA, and a single band at —1.9kb obtained using a probe

specific for V2 receptor mRNA. Although the band with the V2receptor probe could represent an abnormally spliced V2 receptormRNA that retains intron 2 (see above), such an abnormal productshould contain only an additional 109 bases, making it difficult todistinguish them on Northern blots from a normal form. Results fromNorthern analyses of Vlb and V, mRNAs are represented in Fig. 3.The sizes of the mRNAs for the three vasopressin receptors (Vla, V,b,and V-)) are all in excellent agreement with the sizes of corresponding

mRNAs found for respective rat and human normal tissues by us andothers (7, 17, 20).

Western Analysis. The use of the protein A-isolated IgG2b prep

aration of polyclonal antibodies against the V,a receptor (Vivian 3)and both the alkaline phosphatase and the ECL procedures in eachcase revealed the presence of major protein bands at Mr —70,000 and-43,000 in Western blots from SDS-PAGE separations from NCIH82, and fresh-frozen human liver. These data have been detailed

elsewhere (7). Western analysis with polyclonal antibodies to Vlbreceptor (Bivily 3) gave a prominent protein band at Mr —39,000.

Western analysis with polyclonal antibodies to the vasopressin V2receptor (Rocket 3) was performed only on extracts from NCI H82cells, and this demonstrated prominent protein bands at M, —12,000,—17,000, and —33,000. These probably represented forms of theabnormal V, receptor because, from RT-PCR, this is the only V2

structure indicated to be present in NCI H82 cells, and these antibodies recognize an epitope in the NH2-terminal region of the V2 receptor

shared by both normal and abnormal proteins. When antibodiesagainst the abnormal vasopressin V2 structure (Abner 3) were used onthese same extracts, prominent protein bands at Mr 36,000 and 50,000and minor bands at Mr 43,000 and 61,000 were displayed. Because theA/r 36,000/33,000 protein seems to react with both antibody preparations used here, it very likely represents intact abnormal V2 receptor.Failure to observe prominent bands at Mr 17,000 and 12,000 with theCOOH-terminally directed antibodies suggests these proteins couldrepresent NH2-terminal degradation fragments of an abnormal V2

receptor. Data from Western analyses performed using the ECLprocedure and antibodies against Vlh and V2 receptor proteins fromNCI H82 are illustrated in Fig. 4.

DISCUSSION

These data are the first to demonstrate that mRNAs for all threerecognized vasopressin receptors (Vlu, Vlb, and V2) are expressed bysmall cell carcinoma cells. This expression is found in variant SCCL,exemplified here by cell line NCI H82. Hence, it would seem thatgene expression of all three vasopressin receptors is a common featureof SCCL and, unlike the expression of many other substances, is notturned off by the process of cellular dedifferentiation (21, 22). Additionally, the mRNAs for each V, receptor subtype were found to havesequences identical to those expressed by normal tissues (2, 15-18). A

sequentially normal form of the human V2 receptor mRNA (19, 20)was also present in NCI H345 cells, but this normal form wasapparently absent from NCI H82 cells (see below).

Northern analysis showed all of the receptor mRNAs from SCCLwere of the expected size. Hence, messages for at least vasopressin

1868

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

VASOPRESSIN RECEPTORS IN SMALL CELL CARCINOMA

1 2

Fig. 2. RT-PCR products obtained with a poly(A) +RNA preparation from NCI H82 small-cell carcinoma

cells and separated on 2% agarose (Vib receptor products; A), 4% PAGE (V2 receptor products; ß),and 2%agarose (human VACM product: Ci. Selected primersfor V|b and V2 receptors (see Fig. I) yielded overlapping cDNAs that provided the sequence for the entirereading frame of human vasopressin Vlh and V, receptor mRNAs. A, VJb receptor. Lane I, 100-bp DNAladder; Lane 2, 589-bp product of expected size obtained with primers f2 and rO;Lane 3. 535-bp product ofexpected size obtained with primers f2 and r2. B, V2receptor. Lane I, 512-bp product of expected size obtained using primers A and B; Lane 2, 968-bp product

of enlarged size obtained using primers C and D; Lane3, 862-bp product of normal lung obtained using primers C and D. C human VACM. Lane I, 100-bp DNA

ladder; Lanes 2 and 3, products of 674 and 193 bp of thepredicted size, respectively, obtained using primers selected from the rabbit structure (4, 8).

700,600'500-400'300'

B

968bp>862bp>

512bp>

1 2 DNAladderbp

—1000— 900— 800

—700

— 600

—500

—400

Vla and V,h receptors, which can give rise to normal and presumablyfunctional receptor proteins, are present in SCCL. In fact, thesemRNAs seem, in each case, to give rise to the expected receptorproteins because proteins reactive with antibodies against individualreceptors were found in SCCL cell extracts. The molecular weights(apparent) found here for the V, receptor proteins, ranging in sizefrom 39,000 to 70,000, are in good agreement with those published byothers for human and animal receptors (7, 19, 23).

In addition to normal receptor mRNAs, RT-PCR and sequencing

revealed that an abnormal form of V2 receptor mRNA is also expressed by cells representing both classical and variant SCCL. Thisabnormal mRNA was, notably, the only form of the V-, receptormRNA that could be demonstrated in variant NCI H82 cells. Itapparently arises from incomplete posttranscriptional splicing andretains the entire intron 2 of the V2 receptor gene. Inclusion of thisintronic segment introduces a tag stop codon into the reading framecommencing at base position 1359 of the gene. This base is located ashort distance into the intron, and the mRNA, therefore, predictablytranslates a COOH-terminally truncated receptor similar (but not

identical) to that found in an inherited form of diabetes insipidusknown as the "Utah" type. Both receptors lack the seventh transmem

brane domain and COOH terminus of the normal V2 receptor. Itssimilarity to the Utah receptor suggests such an abnormal receptor ofSCCL would be nonfunctional and might serve as a "null" receptor if

it is able to bind the peptide and be expressed at the surface of cells(24). That such an mRNA is present in SCCL and gives rise to aprotein product is evidenced by the ability of antibodies against theunique peptide sequence at the COOH terminus to demonstrate proteins in Western profiles of SCCL. One of these proteins, of Mr—36,000, was found to also react with antibodies directed against the

NH2-terminal region of the receptor, and this protein is of a size

predicted for the nonglycosylated form of abnormal V-, receptor.

Another protein of Mr 50,000 could be a glycosylated form because aglycosylation site is present in the NH2-terminal region of the protein

(25). With respect to this possibility, a carbohydrate side group wasshown to account for a Mr 66,000 form of the Mr 45,000 neuropeptideY2 receptor (26). However, it is not yet clear why the M, 50,000protein (and minor proteins of Mr 43,000 and 61,000) appeared toreact only with antibodies recognizing the abnormal COOH terminus.

We have recently shown for the first time that a human counterpartto rabbit VACM-1 (4) is expressed by classical SCCL (8), and our

current studies demonstrate that this human VACM is also expressedby variant SCCL. Identical products appear to be amplified throughRT-PCR from classical and variant forms of SCCL. The sequence of

this putative receptor in SCCL shows human and rabbit proteins havea high degree of homology (8).7 Despite rabbit VACM-1 being shown

to produce calcium mobilization in response to activation by vaso-

1869

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

VASOPRESSIN RECEPTORS IN SMALL CELL CARCINOMA

VlbKbjH7.5— "*

4.4 —4 5.2 Kb

BV2

1.9 Kb

RNAladder

Kb—4.4

—2.4

—1.4Fig. 3. Northern analysis performed on poly(A) + RNA from SCCL cell line NCI H82

with specific cDNA probes for V]h mRNA (A) and V2 mRNA (B), following separationon a 1.2% formaldehyde-agarose gel. A. 768-bp probe used for Vlh mRNA hybridizeswith an mRNA of approximately 5.2 kb. B. 862-bp probe used for V, mRNA hybridizeswith an mRNA of 1.9 kb. Hybridization was performed with 32P probes for 18 h at 42°C,and the membrane was visualized by exposing X-ray film for 48 h at —80°C.The

reference RNA ladder was from Life Technologies, Inc.

pressin, its role as a receptor for this peptide still remains to be clearlyestablished.

Therefore, SCCL seemingly can express as many as five possiblevasopressin receptors, comprising: normal and presumably functional

forms of the three established serpentine receptors known as the V,a,V,b, and V2 subtypes; an abnormal (and presumably nonfunctional)form of the V2 receptor; and a human counterpart of VACM-1 fromrabbit kidneys. Expression of both V, subtype vasopressin receptorsand the putative hVACM receptor does not appear to undergo changeduring tumor dedifferentiation. It is, therefore, clear that the refrac-tiveness of variant SCCL to vasopressin with respect to changes inintracellular Ca2+ (7) is not due to a loss in the expression of these

vasopressin subtype receptors.The presence of as many as five types of vasopressin receptors in

SCCL, four of which can potentially activate a variety of intracellularcascades, implies that the peptide is exercising a multifaceted ratherthan a simple mitogenic role in the growth and survival of SCCL.Vasopressin Vla and V,b receptors are reported to activate phospho-lipases A2, C, and D and protein kinase C, raise intracellular freeCa2+, and increase phosphorylation of mitogen-activated protein ki

nase and of focal adhesion kinase (2,27-31). VACM is reported toraise intracellular free Ca2+ through currently unknown mechanisms

that do not include phospholipase C and inositol phosphates (4).Alternatively, vasopressin V2 receptors are known to activate

adenylate cyclase and protein kinase A and raise cellular levels ofcAMP (2, 32). This very different manner in which vasopressin canaffect cells through its V, receptors might mean that the peptideexercises a very different role on tumor cells when acting throughthese receptor proteins. Taylor and coworkers (33) provided evidence that vasopressin cannot only stimulate growth of MCF-7breast cancer cells in culture but also inhibit their growth. Although such a growth inhibition by vasopressin has not yet beendemonstrated in SCCL, it is possible a role of tumor growthinhibition for the peptide is exercised through vasopressin V,subtype receptors (opposing growth-promoting actions throughother vasopressin subtype receptors). In support of this propositionwas the earlier finding that growth inhibition in tumors can beassociated with changes in cAMP (34). If correct, the possibleabsence of normal V2 subtype from the variant cells of NCI H82could be reminiscent of acquired resistance by tumor cells to thegrowth-inhibiting effects of transforming growth factor ßthroughgeneration of an abnormal type 2 receptor for this peptide (35). Theavailability of peptide agonists/antagonists for vasopressin V2 receptors makes such a proposition now readily testable in SCCLcells.

t

V2~

33Kd> ^- _

— 8Fig. 4. Western blot analysis from 12% SDS-PAGE with an ECL detection system (Amersham) showing dithioerythritol-reduced proteins from SCCL cell line NCI H82 that were

immunoreactive with rabbit polyclonal antibodies. A, proteins of A/r —39,000, immunoreactive with antibodies against human Vlh receptor (Bivily 3). B. proteins of A/r —12,000.—17,000. and —33,000. immunoreactive with antibodies against human V2 receptor (Rockie 3). C. proteins of M, —36.000, —43,000. —50.000, and —61.000. immunoreactive with

antibodies against an abnormal human V2 receptor (Abner 3). The transfers were first reacted with peroxide in ethanol to block endogenous peroxidase activity, and the polyvinylidenedifluoride membrane was then blocked with a 5% BSA solution in Tris-HCl-Triton X-100. Incubation with the protein A-isolated primary antibodies was performed for 16 h at 4°C.

and incubation with the goat-antirabbit IgG-horseradish peroxidase conjugate (Life Technologies, Inc.) in Tris-HCl-Triton X-100 was performed for l h at ambient temperature.Immunoreactive proteins were visualized on X-ray film exposed to the transfer for 2 min, 30 s, and l min (A, B. and C, respectively).

1870

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

VASOPRESSIN RECEPTORS IN SMALL CELL CARCINOMA

ACKNOWLEDGMENTS

We are thankful for the excellent technical assistant provided by SusanGagnon, Timothy Moser, and Alison North.

REFERENCES

1. North, W. G., Biosynthesis of vasopressin and neurophysins. In: D. M. Gash and G. J.Boer (eds.), Vasopressin: Principles and Properties, pp. 175-209. New York: Plenum

Press, 1987.2. Carmichael. M. C., and Kumer. R. Molecular biology of vasopressin receptors.

Semin. Nephrol. 14: 341-348, 1994.3. Fahrenholz, F.. Akhunddova. A., Buchner, H., and Gorbulev, V. V, lysine vasopres

sin receptor and oxytocin receptor from pig LLC-PK1 cells: two new members of thevasopressin-oxytocin receptor family. In: P. Gross, D. Richter, and G. I. Robertson(eds.), Vasopressin, pp. 45-57. Paris: John Libbey Eurotext. 1993.

4. Bumatowska-Hledin. M. A., Spielman, W. S.. Smith, W. L., Shi, P.. Meyer. J. M.. andDewitt, D. L. Expression cloning of an AVP-activated, calcium-mobilizing receptorfrom rabbit kidney medulla. Am. J. Physiol.. 268: F1198-F1210, 1995.

5. Friedmann, A. S., Malott, K. A., Memoli. V. A., Pai, S. I., Yu, X.. and North W. G.Products of vasopressin gene expression in small-cell carcinoma of the lung. Br. J.Cancer, 69: 260-263, 1994.

6. Woll, P. J., and Rozengurt, E. Neuropeptides as growth regulators. Br. Med. Bull., 45;492-505, 1989.

7. North, W. G., Fay, M. J., Longo, K., and Du. J. Functional V, vasopressin receptorsare present in both variant and classical forms of small-cell tumor cells. Peptides. 18:985-993, 1997.

8. Longo. K. A., Fay, M. J.. and North. W. G. Evidence for the expression of a novelvasopressin-activated calcium mobilizing receptor (VACM-1 ) in human breast cancer

and lung cancer. Mol. Biol. Cell. 7: 3761. 1996.9. Fay. M. J.. Friedmann. A. S., Yu, X.. and North. W. G. Vasopressin and vasopressin

receptor immunoreactivity in variant and classical small-cell lung carcinoma celllines: disruption in the activation cascade of V,„-receptorsin variant SCCL. CancerLett., 82: 167-174, 1994.

10. Badley. J. E. A simple, rapid method for the purification of Poly A+ RNA.BioTechniques. 6: 114-116, 1988.

11. North, W. G., and Yu, X. Vasopressin mRNA and neurophysin-related cell-surfaceantigen (NRSA) in small-cell carcinoma. Peptides, 14: 303-307, 1993.

12. North, W. G., LaRochelle, F. T., Hajdar, J., Sawyer. W. H., and Valtin. H. Characterization of an antiserum used in a radioimmunoassay for arginine-vasopressin:implications for reference standards. Endocrinology, 103: 1976-1984, 1978.

13. North, W. G. Neuropeptide production by small cell carcinoma: vasopressin andoxytocin as plasma markers of disease. J. Clin. Endocrino!. Metab.. 73: 1316-1320,

1991.14. I ,11-ninili.U. K. Cleavage of structural proteins during the assembly of the head of

bacteriophage T4. Nature (Lond.), 227.- 680-685. 1970.15. Thibonnier M., Auzan. C.. Madhun. Z.. Wilkins, P.. Berti-Mattera. L., and Clauser. E.

Molecular cloning, sequencing, and functional expression of a cDNA encoding thehuman V,., vasopressin receptor. J. Biol. Chem.. 269.- 3304-3310, 1994.

16. Hirasawa A., Shibata. K.. Kotosai, K., and Tsujimoto, G. Cloning, functional expression and tissue distribution of human cDNA for the vascular-type vasopressinreceptor. Biochem. Biophys. Res. Commun.. 203: 72-79, 1994.

17. Sugimoto, T., Saito, M., Mochizuki. S., Watanabe. Y.. Hashimoto. S.. andKawashima. H. Molecular cloning and functional expression of a cDN A encoding thehuman Vlh vasopressin receptor. J. Biol. Chem., 269: 27088-27092. 1994.

18. de Keyzer. Y.. Auzan. C.. Lenne. F., Beldjord, C.. Thibonnier. M., Bertagna, X., andClauser. E. Cloning and characterization of the human V3 pituitary vasopressinreceptor. FEES Lett., 356: 215-220, 1994.

19. Birnbaumer, M.. Siebold, A., Rosenthal, W., and Biche!, D. G. The human ADHreceptor: the gene, the cDNA, the protein. In: P. Gross, D. Richter, and G. I.Robertson (eds.), Vasopressin, pp. 19-31. Paris: John Libbey Eurotext. 1993.

20. Fay, M. J., Du. J.. Yu, X.. and North. W. G. Evidence for expression of vasopressinV2 receptor mRNA in human lung. Peptides. 17: 477-481. 1996.

21. Brambilla. E., Moro, D., Gazzeri, S.. Brichon, P. Y., Nagy-Mignotte, H., Morel, F.,

Jacrot, M., and Brambilla. C. Cytotoxic chemotherapy induces cell differentiation insmall-cell lung carcinoma. J. Clin. Oncol.. 9: 50-61, 1991.

22. Carney. D. N., Gazdar. A. F.. Bepler. G.. Guccion. J. G., Marangos. P. J.. Moody,T. W., Zweig, M. H., and Minna. J. D. Establishment and identification of small celllung cancer cells lines having classic and variant features. Cancer Res.. 45: 2913-

2923. 1985.23. Kojro. E.. Eich. P., and Fahrenholz. F. Purification and molecular structure of the

photoaffmily labeled V2-vasopressin receptor from bovine kidney. In: P. Gross, D.

Richter, and G. L. Robertson (eds.). Vasopressin. p. 582. Paris: John Libbey Eurotext,1993.

24. Sadeghi, H. M., Innamorati. G., and Birnbaumer. M. An X-linked NDI mutationreveals a requirement for cell surface V2R expression. Mol. Endocrino!.. //: 706-

713, 1997.25. Brownstein. M. J.. and Lolait. S. J. Hereditary nephrogenic diabetes insipidus. In: P.

Gross. D. Richter, and G. I. Robertson (eds.), Vasopressin, pp. 33-43. Paris: John

Libbey Eurotext, 1993.26. Wimalawansa. S. J. Purification and biochemical characterization of neuropeptide Y2

receptor. Am. Soc. Biochem. Mol. Biol.. 270: 18523-18530, 1995.

27. Thibonnier M.. Bayer. A.I.. and Leng. Z. Cytoplasmic and nuclear signaling pathwaysof the Vrvascular vasopressin receptors. Regul. Pept.. 45: 79-84. 1993.

28. Bilezikjian. L. M.. Blound. A.I., and Vale. W. W. The cellular actions of vasopressinon corticotrophs of the anterior pituitary: resistance to glucocorticoid action. Mo!.Endocrino!.. /: 451-458, 1987.

29. Zachary, I., Sinnett-Smith. J., and Rozengurt, E. Stimulation of tyrosine kinase

activity after addition of bombesin. vasopressin. and endothelin to intact cells. J. Biol.Chem.. 266: 24126-24133. 1991.

30. Zachary. I., Sinnett-Smith. J.. and Rozengurt. E. Bombesin. vasopressin, and endo

thelin stimulation of tyrosine phosphorylation in Swiss 3T3 cells. J. Biol. Chem., 267:19031-19034, 1992.

31. Zachary. I.. Sinnett-Smith, J.. Turner, C. E., and Rozengurt, E. Bombesin. vasopressin. and endothlin rapidly stimulates tyrosine phosphorylation of focal adhesion-associated protein paxillin in Swiss 3T3 cells. J. Biol. Chem.. 268: 22060-22065,

1993.32. Kimura. K. Expression of vasopressin V2 receptor in Xenopits laevis oocytes by

porcine kidney cell line (LLC-PK) messenger RNA. Biochem. Biophys. Res. Commun., 174: 149-155, 1990.

33. Taylor. A. H., Ang. V. T. Y.. Jenkins, J. S., Silverlight. J. J.. Coombes, R. C., andLauqmani. Y. A. Interaction of vasopressin and oxytocin with human breast cancercells. Cancer Res.. 50: 7882-7886, 1990.

34. Kaur, G.. Viallet, J.. Laborda. J.. Blair. O.. Gazdar. A. F.. Minna. J. D.. and Sausville.E. A. Growth inhibition by cholera toxin of human lung carcinoma cell lines:correlation with GM1 ganglioside expression. Cancer Res., 52: 3340-3346, 1992.

35. Jonge, R. R., Garrigue-Antar. L.. Vellucci. V. F.. and Reiss. M. Frequent inactivationof the transforming growth factor beta type II receptor in small-cell lung carcinomacells. Oncol. Res.. 9: 89-98. 1997.

1871

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

1998;58:1866-1871. Cancer Res   William G. North, Michael J. Fay, Kenneth A. Longo, et al.   NeuropeptideSmall Cell Tumors Implies a Multifaceted Role for this Expression of All Known Vasopressin Receptor Subtypes by

  Updated version

  http://cancerres.aacrjournals.org/content/58/9/1866

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/9/1866To request permission to re-use all or part of this article, use this link

Research. on November 12, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from