sst2 Somatostatin Receptor Expression Reverses ...cancerres.aacrjournals.org/content/canres/57/5/956.full.pdf · lnstitut de Recherche des Maladies de l'Appareil Digestif. 2 To whom

ICANCER RESEARCH57. 956-962. March I. 9971

ABSTRACT

Among the five cloned somatostatin receptor subtypes (sat! to sst5), sst2mediates the antiproliferative effect of somatostatin analogues in vitro.Somatostatin analogues have been shown to inhibit cell growth in vitro andin vivo in pancreatic cancer models that expressed sst2. We recentlydemonstrated the loss of sst2 gene expression in human pancreatic adenocarcinomas and most of the derived pancreatic cancer cell lines. In the

present study, we corrected the sst2 defect in human pancreatic cancerBxPC-3 and Capan-1 cells by stable transfection with human sst2 eDNA.

In the absence of exogenous ligand, both BxPC-3 and Capan-1 cellsexpressing sst2 showed a significant reduction in cell growth. This inhibitory effect was blocked by treatment with antiserum to somatostatin.sst2-expressing cells produced somatostatin-like immunoreactivity thatmainly corresponded to somatostatin 14, indicating the induction of a

negative autocrine loop. In other respects, sst2 expression in Capan-1 cellsinduced a significant reduction of clonogenicity in soft agar. Moreover, asignificantly reduced (Capan-1 cells) or suppressed (BxPC-3 cells) tumor

growth in athymic nude mice was observed. The reversal of tumorigenicity induced by the restoration of sst2 expression suggests that the loss ofsst2 contributes to the malignancy of human pancreatic cancers.

INTRODUCTION

Somatostatin is a widely distributed peptide that negatively regulates a number of cellular processes including the growth of multipleepithelial cell types (1). Studies have shown that somatostatin and itsstable analogues suppress the growth of various normal and cancercells expressing somatostatin receptors (2, 3). Evidence exists for adirect antiproliferative effect mediated by specific cell surface receptors. Five subtypes of somatostatin receptors have been cloned fromhuman, mouse, and rat (4, 5). It was found that sst23 mediates theantiproliferative effect of stable long-acting somatostatin analogues invitro through the activation of a phosphotyrosine phosphatase (6, 7).

The role of sst2 in the negative control of cell proliferation is strengthened by the presence of this subtype in rat and human breast cancercells, small cell lung cancer cells, and pancreatic AR4â€”2Jand MIAPaCa-2 cancer cells, which respond in vitro to the growth-inhibitoryeffect of somatostatin analogues (2, 6, 8â€”10).We recently observedthat sst2 is expressed in normal human exocrine pancreas but not inexocrine pancreatic carcinomas and their metastasis as well as in mostof human pancreatic cancer cell lines (1 1). These results could explainthe observations that in exocrine pancreatic cancer patients receivingsomatostatin analogue treatment, tumor growth is not influenced to a

Received 9/4/96; accepted 1/2/97.The costs of publication of this article were defrayed in part 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.

I Supported in part by Grant 6755 from the Association pour la Recherche contre le

Cancer, Grant 9407556 from the Conseil Regional Midi PyrÃ©nÃ©es,and a grant from thelnstitut de Recherche des Maladies de l'Appareil Digestif.

2 To whom requests for reprints should be addressed, at Institut National de Ia Sante

Ct de Ia Recherche MCdicale U151, lnstitut Louis Bugnard, I avenue J. PoulhÃ¨s, Bat L3,Centre Hospitalier Universitaire Rangueil 31054 Toulouse, CÃ©dex,France. Phone: 33-5-61-32-24-07; Fax: 33-5-62-26-40-12; E-mail: [email protected].

@ The abbreviations used are: sst2. cloned somatostatin receptor subtype 2: RI, reversetranscription.

measurable extent in the majority of cases (12â€”14).We postulated thatin these carcinomas with poor prognosis, the loss of sst2 may lead toa deficiency in cell growth negative control by somatostatin and thuscontribute to tumor growth and progression.

To support this hypothesis, we corrected the endogenous defect ofsst2 in two human pancreatic cancer cell lines, BxPC-3 and Capan- 1,by stably expressing human sst2 in these cells. We examined theeffect of sst2 expression on cell growth, transformation in vitro, andtumorigenicity in vivo. Restoration of sst2 expression in human pancreatic cancer cells generated an inhibitory autocrine loop by stimulating sst2 ligand production. The consequent constitutive activationof sst2 reversed both the in vitro and in vivo malignant properties. Wepropose that the malignancy of human pancreatic cancer is in partassociated with a loss of sst2 expression.

MATERIALS AND METHODS

Materials. Geneticin (G4l 8), Moloney murine leukemia virus reverse transcriptase, and Lipofectin were from Life Technologies, Inc. (Cergy Pontoise,France). RNAzoI and oligonucleotides were from Bioprobe (Montreuil-sousBois, France). RNAsin was from Promega (CharbonniÃ¨res, France). Taqpolymerase was from Beckman (Gagny, France). Somatostatin antibodies werefrom Neosystem (Strasbourg, France). Human sst2 cDNA was provided by Dr.0. I. Bell (HowardHughesMedicalInstitute,Chicago,IL).OctreotideandTyr@-octreotidewere from Sandoz, Basel, Switzerland.

Transfections and Cell Culture. Human pancreatic tumor BxPC-3 andCapan-I cells were transfected with pRS2 dicistronic mammalian expressionvector containing the 1.35-kbp fragment encompassing the open reading frameof human sst2 cDNA (6), using Lipofectin reagent. Stable BxPC-3 and Capan-l transfectants were selected and cultured in DMEM and RPMI 1640,respectively, supplemented with 10% FCS and 0.3 mg/ml Geneticin. Geneti

cm-resistant clones were examined for their ability to bind [â€˜25ITyr@]octreotide. Cells were concomitantly transfected with a mock vector

devoid of sst2 cDNA and used as control clones.Binding Studies. Tyr@-octreotidewas radioiodinated and purified by high

performance liquid chromatography as described (15). Transfected BxPC-3and Capan-1 cells were cultured in 35-mm-diameter dishes until confluence(5 X l0@ cells/dish), washed with cold Krebs/HEPES buffer (pH 7.4), and

incubated at 25Â°C for 120 mm in a final volume of 1.5 ml of KrebsIHEPES

buffer containing 15 mg/ml BSA, 0.3 mg/mI soybean trypsmn inhibitor, 0.5

mg/ml bacitracin, and 20 @Mto I nM of [â€2̃5I-Tyr@octreotide (specific activity,

900 Ci/mmol). Cells were then washed twice with KrebsIHEPEScontaining 15mg/ml BSA and collected after a 10-mm incubation in 0.lN NaOH fordetermination of bound radioactivity (6). Nonspecific binding was determined

in the presence of 1 @zMoctreotide.Cell Growth Assay. BxPC-3 and Capan-1 transfected cells cultured re

spectively in DMEM or RPMI 1640 containing 10% FCS were plated in

35-mm-diameter dishes at 6 X l0@ cells/ml (2 ml/dish). After an 18-h attach

ment phase, cells were cultured in serum-free medium overnight (time 0). The

culture medium was then replaced by fresh medium with or without 10%FCS,and cells were cultured for 5 days. To evaluate the effect of somatostatinantiserum on cell proliferation, cells were grown in 24-well plates (6 X l0@cells/well) in medium supplemented with 10% FCS. After an 18-h attachment

phase, cells were cultured in serum-free medium overnight (time 0). Then,culture medium was replaced by fresh medium containing 10% FCS with orwithout somatostatin antibodies added every 48 h (dilution, 1:500). Cell

956

sst2 Somatostatin Receptor Expression Reverses Tumorigenicity of Human

Pancreatic Cancer Cells1

Nathalie Delesque, Louis Buscail,2 Jean-Pierre EstÃ¨ve, Nathalie Saint-Laurent, Catherine Muller,Gisbert Weckbecker, Christian Bruns, Nicole Vaysse, and Christiane Susini

Institut National de Ia Sante el de Ia Recherche MÃ©dicaleVlSi, institut Louis Bugnard, Centre Hospitalier Universitaire Rangueil. 31054 Toulouse CÃ©dex.France (N. D.. L B..i-P. E., N. S-L. N. V.. C. S.J; Laborazoire de Pharmacologie et Toxicologie Fondamentales. Centre National de Ia Recherche Scientifique. 31077 Toulouse, France (C. MI: andSandoz Pharina Ltd 4002, Basle, Switzerland 1G. W., C'. B.J

on March 31, 2019. © 1997 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

sst2 REVERSES TUMORIGENICITYOF HUMAN PANCREATIC CANCER

maintained in pathogen-free conditions. These mice were used as transplantrecipients at the age of 8 weeks. Transfected BxPC-3 and Capan- 1 cells wereinjected s.c. in 0.4 ml ofDMEM or RPMI 1640 (7â€”10mice/clone). s.c. tumorswere measured periodically in two dimensions using vernier calipers. Tumorvolume was determined by the equation

L

growth was measured by cell counting using a Coulter counter model 2M

(Coulter Electronics, Hialeah, FL) as described (6).RIA for Somatostatin. Transfected BxPC-3 and Capan-l cells were grown

in 175-cm2 flasks (2.2 X 106 cells/flask) in DMEM and RPMI 1640, respec

tively, supplemented with 10% FCS. After an 18-h attachment phase, cells

were cultured in serum-free medium overnight (time 0). Then, culture mediumwas replaced by fresh medium containing 10% FCS, and cells were culturedfor 72 h. Pooled culture media (40 mI/assay from three 175-cm2 flaskscorresponding to a total of 26.4 X 106 BxPC-3 control cells, 19.8 X 106BxPC-3/sst2 cells, 35.2 X 106Capan-l control cells, and 27.3 X 106Capanl/sst2 cells) were collected, acidified with trifluoroacetic acid, and subsequently concentrated using SepPak C18cartridges (Waters, Les Ulis, France).The adsorbed peptides were eluted with 80% acetonitrile and 0.1% trifluoroacetic acid. The eluates were evaporated under vacuum and lyophilized. Thedried samples were analyzed for immunoreactivity. The cells were washedtwice, counted using Coulter counter model 2M, extracted in H20, and boiledfor 15 mm. Somatostatin-like immunoreactivity was measured in cell extractsand media by RIA using specific antiserum directed against the central sequence of somatostatin 14 used at a 1:80,000 dilution, the radioligand [V25-Tyr' â€˜]somatostatin,and the standard somatostatin as described (16). This assaydetects somatostatin 14and molecular forms of somatostatin 14extended at theamino terminus, including somatostatin 28 and prosomatostatin. The minimaldetectable dose was 1 fmolltube.

Gel Filtration Chromatography. Pooled cell culture media (800 ml) fromBxPC-3 and Capan-1 cells expressing sst2 cultured in medium supplementedwith 10% FCS for 72 h were acidified and concentrated using SepPak C18cartridges. The adsorbed peptides were eluted with 80% acetonitrile and 0.1%trifluoroacetic acid. The eluate was evaporated, the dried sample was extracted

with 10% acetic acid, and soluble material was applied to a Sephadex G-50column (1.5 X 90 cm; Pharmacia) equilibrated in 10% acetic acid. Fractions

(5.4 ml) were collected at a flow rate of 8.1 mI/h and lyophilized, andsomatostatin-like peptides were evaluated using RIA. The column was cali

brated with 1 ng of somatostatin 14 and 1 ng of somatostatin 28 applied to thecolumn and measured by RIA on the column effluent.

RT-PCR.Culturedcellsweregrownin5-cm-diameterdishesfor48 h,andtotal RNA was extracted as described (17, 18).RT was carried out as describedpreviously using 1 p@gof total RNA (11). PCR was then performed asdescribed (6, 11), using specific sense and antisense primers (2.2 @tMforhuman preprosomatostatin and 0.3 @tMfor @-actin).PCR was carried out on aDNA thermal cycler (Trio Thermobloc-Biometra, Gottingen, Germany) for 25(j3-actin) or 35 (preprosomatostatin) cycles consisting of denaturation for 1mm at 94Â°C,annealing for 1 mm (54Â°Cfor (3-actinand 59Â°Cfor preprosomatostatin), and extension at 72Â°Cfor 90 s. The amplification was terminatedby a final extension step at 72Â°Cfor 10 mm. The following pairs of specificprimers for @3-actin(sense, 5'TCACGCCATCCTGCGTCTGGACT3'; antisense, 5'CCGGACTCATCGTACTCCT3'; Ref. 19) and human preprosomatostatin (sense, (5'TCCAGCTCGGCTTTCGCGGC3'); antisense,(5'TCAG1TFC'FAATGCAAGGGTCTCGC3'; Ref. 20) produced DNA frag

ments of 517 and 487 bp, respectively. PCR amplification of both targetsample and j3-actin were run simultaneously. To confirm that PCR products

resulted from cDNA templates rather than genomic DNA, parallel RT-PCRreactions were carried out for each sample in the absence of reverse tran

scriptase during the RT procedure. These procedures and PCR reactions onwater were used as a negative control of reaction. Amplified fragments wereseparated by 7% PAGE, stained with ethidium bromide, and exposed to UV

light.Soft Agar Assay. Dishes (35 mm in diameter) were precoated with 2 ml of

DMEM or RPMI 1640 containing 20% FCS and 0.8% agar (Bacto-agar;DIFCO). This agar underlayer was allowed to solidify before use. Two ml ofDMEM or RPM! 1640 containing 20% FCS, 0.4% agar, and different quantities of BxPC-3 or Capan-l transfected cells (500â€”100,000 cells) were overlaid onto the precoated dishes. All dishes were incubated at 37Â°Cin 5% C02,95% air. After 26 days, cell colonies were visualized by staining with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (5 mg/mI in phosphatebuffer; Sigma, Saint Quentin-Fallavier, France), and quantification was determined by image analysis (Biocom apparatus).

Tumor Implantation and Growth Studies in Nude Mice. Exponentiallygrowing cells (10 X 106)were inoculated s.c. into athymic female nude mice(Swiss nude/nude-; IFFA-CREDO, l'Arbresle, France) that were bred and

V = W2 X -2

in which W = width and L = length of tumor. Mice were sacrificed at 4 and

9 weeks after the inoculation of Capan-l and BxPC-3 cells, respectively. Allanimal procedures were in accordance with the guidelines of the institutionalanimal care committee.

Statistical Analysis. Statistical comparisons between control cells andcells expressing sst2 were performed using Student's t test (P < 0.05 wasconsidered significant).

RESULTS

Constitutive Inhibition of BxPC-3 and Capan-1 Cell Growth bysst2 Expression. The human somatostatin receptor sst2 was stablyexpressed in BxPC-3 and Capan-l cells. Several Geneticin-resistantclones were obtained, and sst2 receptors were characterized by binding studies using [125I-Tyr@]octreotide as a tracer. Among them, twoclones were selected, BxPC3/sst2 and Capanl/sst2, that expressedsst2 with an equilibrium constant (Kd) of 0.8 Â±0.16 and 0.3 Â±0.06nM, respectively, and a maximal binding capacity of 20 Â± 3.6 and

6 Â±1.1 fmoL/mg, respectively (mean Â±SE of two experiments intriplicate). Cells were concomitantly transfected with a mock vectordevoid of sst2 cDNA and used as control clones. These clones did notbind [125I-Tyr@]octreotide (data not shown).

Growth curves for sst2-expressing cells and control cells weregenerated to determine whether the introduction of exogenous sst2cDNA influences growth parameters. As shown in Fig. 1, BxPC-3 andCapan- 1 cells expressing sst2 showed a significant reduction in cellgrowth in the absence of exogenous ligand when compared with that

Cc,

0

x

w

:@z-j-jw0

Fig. I . Effect of sst2 expression on human pancreatic cancer cell growth. Clonal celllines were derived from BxPC-3 (A and C) and Capan-l (B and D) cells transfected withthe mock vector (U) or the vector encoding human sst2 eDNA (s), respectively. Cells(12 X l0@/35-mm plate) were grown in medium supplemented with 10% FCS. After an18-hattachment phase, cells were cultured in serum-free medium ovemight (time 0). Thencells were cultured for 5 days in medium containing (A and B) or not containing (C andD) 10% FCS, and cell growth was measured at the indicated times by cell counting(Coulter counter model 2M; Coulter Electronics). Results are expressed as the cellnumber/plate (mean Â±SE of three separate experiments performed in triplicate).

0 1 2 345 012345

DAYS

957



sst2 REVERSES TUMORIGENICITY OF HUMAN PANCREATIC CANCER

S286,5K S1412.4K

0.016

0.014

vo

C0

â€˜130.012CU

0E0.

0.01>@:@tCU

@ 0.008C

EEas

.@

C

CU

U)0CUE0U)

1 2 3 4 5 6 7 8 9 1011 1213141516171819202122232425262728293031 3233343536373839404142434445

fraction number

Fig. 2. Somatostatin-like immunoreactivity profile after Sephadex G-50 filtration of media from BxPC-3 cells expressing (@) or not expressing (EIJ)sst2. Cells were cultured for72 h in DMEM with 10% FCS, and cell media (800 ml) were collected, concentrated, treated with 10% acetic acid, and applied to a Sephadex G-50 filtration column. Each fraction(5.4 ml) was lyophilized, and the immunoreactivity of each fraction was assayed by RIA. Arrows, exclusion volume (VO) and the elution position of cytochrome C ( I2.4 K), aprotinin(6.5 K), somatostatin 28 (528; 3.15 K). and somatostatin 14 (S14; I.64 K).

observed from control cells. Both in serum-containing medium and inserum-free medium, the decrease in proliferation of BxPC-3 cellsexpressing sst2 was significant after 24 h of culture when comparedwith that of control cells, and the decrease continued throughout theexperiment (Fig. 1, A and C). After 5 days of culture, the proliferationofBxPC-3/sst2 was reduced by 41 Â±2 and 100% (mean Â±SE; n 3;P < 0.001) when cells were cultured in serum-containing medium(Fig. 1A) and in serum-free medium (Fig. 10, respectively. Similarresults were obtained with Capan-l cells (Fig. 1, B and D). Significantdecrease of sst2-expressing cell proliferation was found after 3 days ofculture in serum-containing medium (Fig. IB) and after 24 h of culturein serum-free medium (Fig. 1D) when compared with that of controlcells. After 5 days of culture of Capan-l/sst2, proliferation wasreduced by 25 Â±2 and 68 Â±3% in medium with or without serum,respectively (mean Â±SE; n = 3; P < 0.01). These results wereobserved with two different clones of each cell line stably expressingsst2 (data not shown). Treatment of BxPC-3 and Capan-l cells withthe stable somatostatin analogue octreotide at a concentration of 1 nMdid not provide additional inhibition of cell growth irrespective of thetime of treatment (data not shown). The doubling time of sst2-expressing cells versus control cells was 47 Â±1 versus 36 Â±2.7 h forBxPC-3 cells (P < 0.02) and 34 Â±0.8 versus 3 1 Â±0.7 h for Capan- 1cells (P = 0.05) when cells were cultured in serum-containing medium. When Capan- 1 cells were cultured in serum-free medium, thedoubling time of ssi2-expressing cells versus control cells was84 Â±4.2 versus 52 Â±3.5 h (P < 0.001). These results demonstratedthat the stable expression of sst2 results in an inhibition of BxPC-3

and Capan- 1 cell growth in both the presence and absence of serum,suggesting a constitutive activation of sst2.

Stable Expression of sst2 in Pancreatic Cancer Cells CausedActivation of sst2 by a Somatostatin-dependent Autocrine Pathway. To answer the question of whether sst2 expressed in pancreaticcancer cells was constitutively active, we analyzed the stimulation ofsst2 by the production of endogenous somatostatin-like immunoreactivity. To determine whether BxPC-3 and Capan-l cells expressingsst2 might synthesize and secrete somatostatin, we first examined theendogenous production of somatostatin-like immunoreactivity in cellextracts and media obtained from cells expressing or not expressingsst2. Somatostatin-like immunoreactivity was undetectable in cellextracts from cells expressing or not expressing sst2, irrespective ofthe cell line. However, after 3 days of culture, the level of somatostatin-like immunoreactivity in cell medium obtained from BxPC-3control cells was 0.5 Â± 0.05 fmol/l06 cells. This level reached0.8 Â±0.08 fmol/106 cells (mean Â±SE; n = 3; P < 0.05) in the culturemedium obtained from BxPC-3/sst2. Somatostatin-like immunoreactivity was undetectable in media from Capan-l control cells after 3days of culture but was measured in media of Capan-l/sst2 cells at alevel of 0.1 Â±0.02 fmol/106 cells (mean Â±SE; n = 3). These resultsindicated that somatostatin-like immunoreactivity was significantlyincreased in sst2-expressing cells and suggested that stable sst2 expression led to the production of somatostatin-like peptides.

We next investigated which forms of somatostatin were present incell culture medium. In view of the extremely low levels of immunoreactivity in the culture medium of Capan-l cells, this study was

958




performed on BxPC-3 cells. Cells expressing or not expressing sst2were cultured in serum-containing medium, and pooled cell mediawere treated as described in â€œMaterialsand Methodsâ€•for the analysisof somatostatin-like immunoreactivity. As illustrated in Fig. 2, whencell medium from control cells was loaded onto a Sephadex G-50column, two peaks of somatostatin-like immunoreactivity were observed. A major peak (fractions 7â€”13;elution volume, 48 ml) thataccounted for 75% of total immunoreactivity represented high apparent molecular mass immunoreactive materials and could correspondto somatostatin proforms. A minor peak (25% of total immunoreactivity) with an elution volume of 173 ml (fractions 29â€”33)coelutedwith somatostatin 14 standard. In contrast, when cell medium fromsst2-transfected BxPC-3 cells was loaded onto the column, 83% of thetotal immunoreactivity coeluted within three peaks corresponding tothe mature forms of somatostatin, somatostatin 28 (fractions 24â€”27;elution volume, 140 ml), and somatostatin 14 (fractions 31â€”33;elution volume, 173 ml), and the shorter form somatostatin 12 (fractions35â€”37;elution volume, 194 ml). Seventeen percent of the total immunoreactivity coeluted in a minor peak of high apparent molecularmass immunoreactive materials (fractions 7â€”10;elution volume, 48ml) that could correspond to somatostatin proforms.

To further support the hypothesis of sst2-induced production ofsomatostatin-like immunoreactivity in pancreatic cancer cells, we nextexamined the expression of preprosomatostatin mRNA by RT-PCRanalysis. As observed in Fig. 3, when cells were cultured for 48 h, thelevel of preprosomatostatin transcripts was up-regulated in sst2-expressing BxPC-3 and Capan-l cells when compared with that ofcontrol cells.

To investigate whether the secreted immunoreactive somatostatinaccounted for the inhibition of cell growth observed in cells expressing sst2, we then evaluated the effect of the addition of somatostatinantiserum to the cell culture medium. A 96-h exposure to antisomatostatin antiserum did not modify the cell growth of BxPC-3 andCapan-1 control cells (data not shown). Conversely, the inhibition ofcell growth observed in BxPC3/sst2 and Capan-l/sst2 cells was significantly blocked by 44 Â±5.2 and by 62 Â±21%, respectively (meanof three experiments in triplicate) after a 96-h exposure to antisomatostatin antiserum. These results indicate that somatostatin antiserum neutralized the effect of secreted somatostatin and nullified theinhibitory effect of sst2 expression on cell growth.

Stable sst2 Expression Reduced Tumorigenicity of BxPC-3 andCapan-1 Cells: Anchorage-independent Growth. The ability toform colonies in soft agar, which is reflective of malignant transfor

A

B

Fig. 4. Anchorage-independent colony formation in soft agarose of Capan-l cells.Exponentially growing control cells transfected with mock vector (A) and sst2-expressingcells (B) were overlaid (1.5 X l03/dish in 2 ml of RPMI 1640 containing 20% FCS and0.4% agar) on 2 ml of underlayer of 0.8% agarose in the same medium in 6-well plates(35 mm in diameter). Cell colonies were visualized by staining with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide after 26 days of incubation.

mation, was determined in BxPC-3 and Capan-l cells. Independent ofthe number of cells plated onto soft agar, we observed an anchorageindependent growth for Capan-l cells but not for BxPC-3 cells. Wethen compared the clonogenic potential of control and Capan-l/sst2cells in semisolid agarose medium plated with 1000 cells/dish. Asshown in Fig. 4, Capan-l/sst2 cells had a lower cloning efficiency insemisolid medium than did the control cells. The number of coloniesformed by Capan-l/sst2 cells was reduced by 89% when comparedwith that formed by control cells [35 Â±4 colonies/dish (sst2-expressing cells) versus 304 Â±10 colonies/dish (control cells), mean Â±SEof two experiments in quadruplicate].

Tumorigenicity in Vivo. Inoculation of exponentially growingcells into athymic mice was performed at 10 X 106 cells/site, and theprogression of xenografts was followed. After a lag time of 40 daysafter inoculation, xenografts from BxPC-3 control cells grew rapidly,with a doubling time of 7.9 days. By contrast, no significant growthof xenografts from BxPC-3/sst2 cells was observed up to day 66, andtumor volume remained unchanged (Fig. 5A). Similar results wereobtained after another run of experiments using different clonesexpressing or not expressing sst2 (data not shown). Xenografts fromCapan-l control cells grew rapidly up to day 29, but those fromsst2-expressing Capan- 1 cells grew at a significantly slower rate (thexenograft size was less than 53 Â± 13% of that of control cells;P < 0.05; Fig. 5B). The doubling time of xenografts from control andssl2-expressing Capan-l cells was 9 and 12 days, respectively.

959

Ml 234

Fig. 3. Effect of sst2 expression on preprosomatostatin mRNA levels in pancreaticcancer cells. After an 18-h attachment phase, cells were cultured in serum-free mediumovemight and then cultured for 48 h in medium containing 10% FCS. RI-PCR analysiswas performed on total RNAs extracted from clonal cell lines containing mock vector(Capan-l, Lane 1; BxPC-3, Lane 3) or expressing sst2 (Capan-l, Lane 2; BxPC-3, Lane4). The resulting PCR products using specific primers for preprosomatostatin (A; 487 bp)and @3-actin(B; 517 bp) were analyzed by PAGE after ethidium bromide staining. M, DNAsize markers (pOEMÂ®;Promega).




Furthermore, sst2 is present in normal human exocrine pancreas,colon, and prostate tissues but is not expressed in pancreatic andadvanced colorectal carcinomas and their metastasis as well as inprostate cancer ( 11, 22). The loss of sst2 gene expression and thus ofthe consequent mediation of negative regulation of cell proliferationin neoplastic tissues may provide a growth advantage for these tumors

and explain in part the lack of anti-growth effect of somatostatinanalogue therapy on several advanced-stage carcinomas ( 12). In anattempt to support the molecular basis of this hypothesis, we correctedthe sst2 receptor defect in the human pancreatic cancer BxPC-3 andCapan- I cells by stably expressing sst2.

We demonstrated that the sst2 transfectants showed a significantlyreduced cell growth in vitro when compared with BxPC-3 and Capan-I control cells. We established that Capan-l cells expressing sst2showed a decreased clonogenicity in agar. Furthermore, our resultsclearly demonstrated that xenografts from BxPC-3 and Capan-l cellsexpressing sst2 showed a suppressed or a reduced growth. All of theseeffects were observed in the absence of added exogenous ligand.

The inhibition of cell proliferation observed in sst2-expressing cellsoccurred independently of the presence of serum in the culture medium and was associated with an increase in the level of preprosomatostatin mRNA as well as an increase in production of endogenoussomatostatin-like peptides. Moreover, the somatostatin antibodies thatcan neutralize the somatostatin-like peptides secreted by sst2-expressing cells counteracted the sst2-induced inhibition of cell proliferation,indicating that suppression of the interaction between secreted ligandand sst2 receptors interrupted the sst2-induced autocrine loop. Allthese findings support the concept that the expression of sst2 induceda constitutive activation of the receptor resulting from the induction ofsst2 ligand and led to the generation of a negative autocrine growthregulatory loop.

As previously observed in breast cancer cells and colonic tumorcells (23, 24), human pancreatic cancer Capan- 1 and BxPC-3 cells arecapable of synthesizing endogenous somatostatin-like peptides. InBxPC-3 cells, both control and sst2-expressing cells produced somatostatin-like peptides. This production was significantly increased insst2-expressing cells, as reflected by the increase of steady-statepreprosomatostatin mRNA and peptide levels. In Capan-l controlcells, no immunoreactive materials could be detected, but preprosomatostatin mRNA was detected after RT-PCR analysis. The synthesisof somatostatin-like peptides was clearly increased in sst2-expressingCapan- I cells, as shown by the detection of somatostatin immunoreactivity in the culture medium from these cells and the up-regulationof the level of preprosomatostatin mRNA. However, due to the lowlevel of immunoreactive materials in Capan-l/sst2 cells, the secretedforms of somatostatin-like peptides could not be investigated.

As many peptide hormones, somatostatin is synthesized as aninactive precursor, prosomatostatin, that undergoes posttranslationalprocessing by specific endoproteolytic enzymes belonging to thesubtilisin-like serine protease family (25). Processing occurs principally at the COOH-terminal part of the molecule to generate twobiologically active peptides, somatostatin 14 and somatostatin 28, anamino-terminal-extended form of somatostatin 14. In BxPC-3 controlcells, most of the somatostatin immunoreactivity corresponded tosomatostatin proforms. Conversely, in sst2-expressing BxPC-3 cells,the secreted forms corresponded primarily to somatostatin 14, suggesting that sst2 expression up-regulated the endoproteolytic processing of prosomatostatin.

Both binding capacity and somatostatin-like immunoreactivitywere 3- and 8-fold higher, respectively, in BxPC-3/sst2 than in Capanl/sst2 cells. These results may be correlated with the higher inhibitorygrowth response both in vitro and in vivo assays observed in BxPC3/sst2 cells when compared to Capan-l/sst2 cells. Finally, the auto

960

A

0 10 20 30 40 50 60 70

Days

0 10 20 30

(I)

EE

asEI 600

C,,

EE0E0>

0EI-.

Days

Fig. 5. Xenograft formation by BxPC-3 (A) and Capan-l (B) cells. Exponentiallygrowing cells (control cells expressing mock vector, 0; sst2-expressing cells, @)wereinoculated s.c. (10 X lO@cells/site) into athymic female nude mice (7â€”10animals/groupfor Capan- I and BxPC-3 cells, respectively). Tumors were measured on the indicated daysin two dimensions using vemier calipers. Values are means Â±SE from 7-10 xenografts.

DISCUSSION

Pancreatic adenocarcinoma is actually the fourth leading cause ofdeath from malignant diseases in Western countries. The poor prognosis is explained by: (a) late diagnosis because of the absence ofspecific early symptoms; and (b) a tendency of the tumor to spreadrapidly with the frequent occurrence of metastasis even from smallprimary tumors (21). In the present study, we demonstrated theintrinsic role of sst2 in the negative growth regulation of humanpancreatic cancer cells. The restoration of sst2 expression after stabletransfection in the two cell lines, Capan-l and BxPC-3, had a directimpact on the malignant properties of these cells due to the acquisitionof an autocrine-negative activity of sst2.

It is known that somatostatin exerts an antiproliferative effect onvarious normal and cancerous cells by either indirectly inhibitinghormone/growth factor release and angiogenesis or directly acting ontarget cells (2, 10). We recently demonstrated that sst2 mediates theantiproliferative effect of the somatostatin analogues octreotide andvapreotide through the stimulation of tyrosine phosphatase activity(6, 7).




Classification and nomenclature of somatostatin receptors. Trends Pharmacol. Sci.,16: 86â€”88,1995.

6. Buscail, L.. Delesque. N.. EstCve, i-P., Saint-Laurent, N., Prats, H., Clerc, P.,Robberecht, P., Bell, G. I., Liebow, C., Schally. A. V., Vaysse, N., and Susini, C.Stimulation of tyrosine phosphatase and inhibition of cell proliferation by somatostatin analogues: mediation by human receptor subtypes SSTRI and SSTR2. Proc. NatI.Acad.Sci.USA, 91:2315â€”2319.1994.

7. Buscail, L., EstÃ¨ve, i-P., Saint-Laurent, N., Bertrand, V., Reisine, I., O'Carroll,A-M., Bell, G. I., Schally, A. V., Vaysse, N., and Susini, C. Inhibition of cellproliferation by the somatostatin analogue RC-160 is mediated by somatostatinreceptor subtypes SSTR2 and SSTR5 through different mechanisms. Proc. NatI.Acad.Sci.USA, 92: 1580â€”1584,1995.

8. Liebow, C., Reilly, C.. Serrano, M., and Schally. A. V. Somatostatin analogues inhibitgrowth of pancreatic cancer by stimulating tyrosine phosphatase. Proc. NatI. Acad.Sci. USA, 86: 2003â€”2007, 1989.

9. Viguerie, N., Tahiri-iouti, N., Ayral, A-M., Cambillau, C., Scemama, J-L., BastiÃ©.M-i., Knuhtsen. S., EstÃ¨ve, i-P., Pradayrol, L., Vaysse, N., and Susini, C. Directinhibitory effects of a somatostatin analog, SMS 201-995, on AR4â€”2Jcell proliferation via pertussis toxin-sensitive guanosine trtphosphate-binding protein-independent mechanism. Endocrinology. 124: 1017â€”1025.1989.

10. Lamberts, S. W. J., Krenning. E. P.. and Reubi, J. C. The role of somatostatin andits analogs in the diagnosis and treatment of tumors. Endocr. Rev., 12: 450â€”482,I991.

I I . Buscail, L., Saint-Laurent, N., Chastre, E., Vaillant, I. C., Gespach, C., CapellÃ , G..Kalthoff, H., Lluis, F., Vaysse, N., and Susini, C. Loss of sst2 somatostatin receptorgene expression in human pancreatic and colorectal cancer. Cancer Res., 56: 1823â€”1827,1996.

12. Klijn, J. G. M., Hoff, A. M., Planting, A. S. I., Verweij, I., Kok, T., Lamberts,S. W. J., Potengen. H., and Foekens, J. A. Treatment of patients with metastaticpancreatic and gastrointestinal tumours with the somatostalin analogue sandostatin: a Phase II study including endocrine effects. Br. i. Cancer. 62: 627â€”630,1990.

13. Huguier, M., Samana. G., Testart, i., Mauban, S., Fingerhut, A., Nassar, J., Houry. S..Jaeck, D., De Mestier, P., Favre, i. P., Michot, F., Vidrequin, A., Mantion. G.,VeyriÃ¨res, M., Fourtanier, G., Lointier, P., and Gignoux, M. Treatment of adenocarcinoma of the pancreas with somatostatin and gonadoliberin (luteinizing hormonereleasing hormone). Am. J. Surg., 164: 348â€”353, 1992.

14. Friess, H., Buchler, M., Beglinger, C., Weber, A., Kunz, i., Fritsch, K., Dennler, H. I.,and Beger. H. G. Low-dose octreotide treatment is not effective in patients withadvanced pancreatic cancer. Pancreas, 8: 540â€”545, 1993.

15. Knuhtsen, S., EstÃ¨ve,J. P., Bemadet, B., Vaysse, N., and Susini, C. Molecularcharacterization of the solubilized receptor of somatostatin from rat pancreatic acinarmembranes. Biochem. J., 254: 641â€”647, 1988.

16. Chayvialle, I. A., Miyata, M., Rayford, P. L., and Thompson, i. C. Effect oftest meal,intragastric nutrients. and intraduodenal bile on plasma concentrations of immunoreactive somatostatin and vasoactive intestinal peptide. Gastroenterology. 79: 844â€”852, 1980.

17. Chomczynski. P., and Sacchi, N. Single-step method of RNA isolation by acidguanidium thiocyanate-phenol-chloroform extraction. Anal. Biochem.. 162: 156â€”159, 1987.

18. Dukas. K.. Sarfati, P., Vaysse, N., and Pradayrol, L. Quantification of changes in theexpression of multiple genes by simultaneous polymerase chain reactions. Anal.Biochem., 215: 66â€”72,1993.

19. Alonso, S., Minty, A., Bourlet, Y., and Buckingham, M. Comparison of threeactin-coding sequences in the mouse; evolutionary relationships between the actingenes of warm-blooded vertebrates. i. Mol. Evol., 23: 11â€”22,1986.

20. Shen, L. P., and Rutter, W. i. Sequence of the human somatostatin I gene. Science(Washington DC), 234: 168â€”170,1984.

2 1. Warshaw, A. L., and Fernandez-Del Castillo, C. Pancreatic carcinoma. N. Eng.J. Med., 326: 455â€”465,1992.

22. Reubi, J. C., Waser, B., Schaer, J. C., and Markwalder, R. Somatostatin receptors inhuman prostate and prostate cancer. I. Clin. Endocrinol. & Metab., 80: 2806â€”2814,I995.

23. Nelson, I., Cremin, J. M., and Murphy. R. F. Synthesis of somatostatin by breastcancer cells and their inhibition by exogenous somatostatin and sandostatin. Br. i.Cancer, 59: 739â€”742,1989.

24. De BruIne, A. P., de Vries, J. E., Dinjens. W. N. M., Moerkerk, P. T., Van der Linden.E. P. M.. Pijls, M. M. I., Ten Kate, J., and Bosman, F. I. Human Caco-2 cellstransfected with C-Ha-Ras as a model for endocrine differentiation in the largeintestine. Differentiation, 53: 51â€”60,1993.

25. Patel, Y., and Galanopoulou, A. Processing and intracellular targeting of prosomatostatin-derived peptides: the role of mammalian endoproteases. Somatostatin and itsreceptors. Ciba Found. Symp., 190: 26â€”50,1995.

26. Rauly, I., Saint-Laurent, N., Delesque, N., Buscail, L., EstÃ¨ve,I. P., Vaysse. N.. andSusini, C. Induction of a negative autocrine loop by expression of SSTR2 somatostatin receptor in NIH-3T3 cells. I. Clin. Invest., 97: 1874â€”1883, 1996.

27. Ohmura, E., Lkada, M., and Onoda, N. Insulin-like growth factor I and transforminggrowth factor a as autocrine growth factors in human pancreatic cancer cell growth.Cancer Res., 50: 19â€”27,1990.

28. Korc, M., Chandrasekar, B., Yamanaka, Y.. Friess, Y., BUehler, M. W., and Beger,H. G. Overexpression of the epidermal growth factor receptor in human pancreaticcancer is associated with concomitant increases in the levels of epidermal growthfactor and transforming growth factor a. i. Clin. invest., 90: 1352â€”1360,1992.

29. Smith, I. P., Fantaskey. A. P., Liu, G., and Zagon, I. S. Identification of gastrin as agrowth peptide in human pancreatic cancer. Am. I. Physiol.. 268: Rl35â€”Rl4l, 1995.

30. NÃ¨gre.F., Fagot-Revurat, P., Bouisson, M., Rehfeld, J. F., Vaysse, N., and Pradayrol,

crime-negative activity of the receptor in transfectants depends uponthe levels of sst2 expression and of peptide production. This autocrineloop could control the tumorigenic properties of pancreatic cancercells.

We recently reported that expression of sst2 in NIH 3T3 cells alsoinduces a negative autocrine loop that results from the up-regulationof preprosomatostatin gene expression (26). Our results demonstratingthat the expression of ssi2 in two human pancreatic tumor cellsgenerates a negative autocrine loop by stimulating the production ofsst2 ligand support the hypothesis that the induction of somatostatinby sst2 may be a general mechanism by which the sst2 ligand receptorsystem negatively regulates cell proliferation.

It has been demonstrated that many autocrine factors such asinsulin-like growth factor I (27), epidermal growth factor/transforming growth factor a (28), gastrin (29), and glycine-extended gastrin(30) stimulate the growth of pancreatic tumor cells. However, thepresent study is, to our knowledge, the first demonstration that inhuman pancreatic cancer cells, the expression of a receptor coupled toG protein leads to the reversal of malignancy in vitro and in vivo.Nevertheless, mechanisms implicated in the antineoplastic effect mediated by sst2 remain to be investigated. Preliminary results obtainedin BxPC-3 cells argue in favor of a cell cycle blockade in the G1â€”Stransition phase in sst2 transfectants.

The molecular pathology of human pancreatic cancer has beenpoorly understood until now. The role of K-ras oncogene has beenestablished, and its mutation occurs in more than 90% of cases ofpancreatic carcinomas (3 1). A mutation of the tumor suppressor genep53 occurringin 50%of cases has been observed,as well as a loss ofheterozygosity at chromosome Yip involving the site of the p53 gene(32). Recently, new putative tumor suppressor genes have been described such as P16/MTS-1 or DPC4 with regard to the deletion ormutations observed at chromosomes 9p2l and l8q21, respectively(33, 34). Human sst2 somatostatin receptor gene has been found to be

localized on chromosome l7q24 (35). To our knowledge, mutationsor deletion at chromosome l7q have not been observed in pancreaticcancer but a neighbor region at locus l7q2l bears the BRCA I antioncogene (breast cancer I gene), a mutation or deletion of whichpredisposes one to breast and ovarian carcinomas (36). Additionalstudies are needed to elucidate the mechanisms by which sst2 geneexpression was lost in human pancreatic cancer. This could resultfrom gene alterations, as recently observed in human small cell lungcancer cells (37), and/or transcriptional or posttranscriptional defects.

In conclusion, we observed that the tumorigenicity of Capan- 1 andBxPC-3 cells is reversible upon restoration of sst2 expression. Theabsence of sst2 expression in human pancreatic cancer and the subsequent loss of the sst2-induced autocrine loop could play a role inderegulation of cell growth and tumor development.

ACKNOWLEDGMENTS

We thank Dr. 0. I. Bell for providing human sst2 cDNA, P. Dumont fortechnical assistance, and Dr. A. Estival for providing pancreatic cancer cellsand for helpful advice.

REFERENCES

I. Lewin, M. J. M. The somatostatin receptors in the GI tract. Annu. Rev. Physiol., 54:455â€”469,1992.

2. Schally, A. V. Oncological applications of somatostatin analogues. Cancer Res., 48:6977â€”6985,1988.

3. Weckbecker, G., Raulf, F., Stolz, B., and Bruns, C. Somatostatin analogs for diagnosis and treatment of cancer. Pharmacol. & Ther., 60: 245â€”260,1993.

4. Bell, G. I., and Reisine, I. Molecular biology of somatostatin receptors. TrendsNeurosci., 16: 34â€”38,1993.

5. Hoyer, D., Bell, G. I., Berelowitz, M., Epelbaum, J.. Feniuk, W.. Humphrey, P. P. A.,O'Carroll, A-M., Patel, Y. C., Schonbrunn, A., Taylor, J. E., and Reisine, I.

961




L. Autocrine stimulation of AR4â€”2irat pancreatic tumor cell growth by glycineextended gastrin. tnt. J. Cancer, 66: 653-658, 1996.

31. BerthClCmy,P., Buisson, M., Escourrou, I., Vaysse, N., Rumeau, J. L., and Pradayrol,L. Identification of gastrin as a growth peptide in human pancreatic cancer. Ann.intem.Med.,123:188â€”191,1995.

32. Pellegata, N. S., Sessa, F., Renault, B., Bonato, M., Leoan, B. E., Solcia, E., andRanzani, G. N. K-ms and p53 gene mutations in pancreatic cancer: ductal andnonductal tumors progress through different genetic lesions. Cancer Res., 54: 1556â€”1460. 1994.

33. Huang, L., Goodrow, I. L., Zhang, S. Y., Klein-Szanto, J. P., Chang, H., and Ruggeri,B. A. Deletion and/or mutation analyses of the P16/Mi'S-I tumor suppressor gene inhuman ductal pancreatic cancer reveals a higher frequency of abnormalities intumor-derived cell lines than in primary ductal adenocarcinomas. Cancer Res., 56:1137â€”1141,1996.

34. Hahn, S. A., Schutte, M., Hoque, A. I., Moskaluk, C. A., da Costa, L. I., Rozenblum,E., Weinstein, C. L., Fischer, A., Yeo, C. J., Hruban, R. H., and Kem, S. E. DPC4,a candidate tumor suppressor gene at human chromosome l8q2l . I . Science (Washinglon DC), 271: 350â€”353, 1996.

35. Yamada, Y., Stoffel, M., Espinosa, R., Xiang, K. S., Seino, M., Seino, S., Le Beau,M. M., and Bell, G. I. Human somatostatin receptor genes: localization to humanchromosomes 14, 17, and 22 and identification of simple tandem repeat polymorphisms. Genomics, 15: 449â€”452, 1993.

36. Knudson, A. G. Antioncogenes and human cancer. Proc. Natl. Acad. Sci. USA, 90:10914â€”10921,1993.

37. Zhang, C. Y., Yokogoshi, Y., Yoshimoto, K., Fujinaka, Y., Matsumoto, K., andSaito, S. Point mutation of the somatostatin receptor 2 gene in the human smallcell lung cancer cell line COR-Ll03. Biochem. Biophys. Res. Commun., 210:805â€”815,1995.

962



1997;57:956-962. Cancer Res Nathalie Delesque, Louis Buscail, Jean-Pierre Estève, et al. Tumorigenicity of Human Pancreatic Cancer Cellssst2 Somatostatin Receptor Expression Reverses

Updated version

http://cancerres.aacrjournals.org/content/57/5/956

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

[email protected] 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/57/5/956To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

sst2 Somatostatin Receptor Expression Reverses ...cancerres.aacrjournals.org/content/canres/57/5/956.full.pdf · lnstitut de Recherche des Maladies de l'Appareil Digestif. 2 To whom