Langenbecks Arch Chir (1995) 380:90-95 © Springer-Verlag 1995 S. Jonas • M. John • J. Boese-Landgraf - R. Hfiring G. Prevost- E Thomas - S. Rosewicz - E.-O. Riecken B. Wiedenmann. R Neuhaus Somatostatin receptor subtypes in neuroendocrine tumor cell lines and tumor tissues Received: 12 September 1994 Subtypen des Somatostatinrezeptors in neuro- endokrinen Tumorzellinien und Tumorgeweben Abstract Somatostatin receptor scintigraphy (SRS) is positive in approximately 80% of all patients who have been found to have neuroendocrine (NE) gastroenteropan- creatic (GEP) tumors. The reasons for negative results are unclear. The aim of the present study was identification of the specific somatostatin receptor (SSTR) subtypes that are responsible for the in vivo binding of the widely used som- atostatin (SST) analogues octreotide and lanreotide in hu- man neuroendocrine gastroenteropancreatic tumors. Ten patients were subjected to SRS with radiolabeled octreo- tide. Following surgical resection, tumor tissues were an- alyzed for SSTR subtype mRNA expression by the reverse transcription-polymerase chain reaction (RT-PCR). In ad- dition, SSTR subtype transcripts were investigated by Northern blot analysis and RT-PCR in neuroendocrine tu- mor cell lines. Expression of SSTR at the protein level was studied by chemical cross-linking experiments. Three pa- tients were negative by SRS. However, RT-PCR revealed most prominently SSTR 2 expression in all tumor speci- mens. In addition, all tumor tissues analyzed by chemical crosslinking exhibited SST-14 binding sites, indicating that at least some NE tumors were false-negative on SRS. S. Jonas • R Neuhaus ([]) Freie Universit~it Berlin, Universit~itsklinikum Rudolf Virchow, Department of Surgery, Augustenburger Platz 1, D-13353 Berlin, Germany J. Boese-Landgraf • R. H~ring Freie Universit~it Berlin, Universit~itsklinikum Benjamin Franklin, Department of Surgery, Berlin, Germany M. John • S. Rosewicz. E.-O. Riecken • B. Wiedenmann Freie Universit~t Berlin, Universit~itsklinikum Benjamin Franklin, Department of Gastroenterology, Berlin, Germany G. Prevost Institut d'Oncologie, Bobigny, France F. Thomas Ipsen Biotech, Paris, France The heterogeneous SSTR subtype pattern of NE tumor cell lines with either SSTR 1 (BON, RIN 38) or SSTR 2 (QGP 1, AR 42 J) as the predominant subtype suggests an additional role for other SSTR subtypes than SSTR 2. Zusammenfassung In 80% der F~ille von neuroendo- krinen (NE-) Tumoren des gastroenteropankreatischen (GEP-) Systems ergibt die Somatostatinrezeptorszintigra- phie (SRS) positive Ergebnisse, wobei Ursachen negativer Befunde nicht bekannt sin& Ziel dieser Studie war die Identifizierung spezifischer Subtypen des Somatostatinre- zeptors (SSTR), an die in NE-Tumoren des GEP-Systems die verbreiteten Somatostatin(SST-)analoge Oktreotid und Lanreotid in vivo binden. Bei 10 Patienten mit NE-Tu- moren wurde eine SRS mit markiertem Oktreotid durchgefiihrt und anschlieBend eine chirurgische Resek- tion vorgenommen. Der Nachweis der mRNA-Expression von SSTR-Subtypen im Tumorgewebe erfolgte mittels re- verser Transkription und Polymerasekettenreaktion (RT-PCR). Weiterbin wurden Transkripte von SSTR-Sub- typen in NE-Tumorzellinien dutch Northern-blot-Analyse und RT-PCR untersucht. Die Expression yon SSTR-Sub- typen auf der Proteinebene wurde anhand chemischer Quervernetzungsexperimente nachgewiesen. Ergebnisse der SRS waren in 3 yon 10 Patienten mit einem NE-Tumor des GEP-Systems negativ. In allen Tumorgeweben wurde mittels RT-PCR haupts~chlich die Expression des SSTR 2-Subtyps nachgewiesen. Weiterhin konnten in allen un- tersuchten Tumorgeweben bei der chemischen Querver- netzung SST 14-Bindungsstellen dargestellt werden, so dag yon falsch-negativen Ergebnissen der SRS ausgegan- gen werden muB. Das heterogene Muster yon SSTR-Sub- typen in NE-Tumorzellinien mit vomehmlicher SSTR 1- oder SSTR 2-Expression in BON- und RIN 38- bzw. in QGP 1- und AR 42 J-Zellinien k6nnte ftir eine zus~itzliche Rolle anderer Subtypen als SSTR 2 sprechen. Schliisselwi~rter Somatostatinrezeptor-Subtypen Neuroendokrine Tumorzellinien Gastroenteropankreatische Tumore
`
`Roxane Labs., Inc.
`Exhibit 1038
`Page 001
`
`

`
`91 Introduction Neuroendocrine (NE) tumor cells of the gastroenteropan- creatic (GEP) system contain high-affinity binding sites for somatostatin (SST) and its analogues octreotide and lanreotide [23-26]. Stable somatostatin analogues (e.g., octreotide and lanreotide) are clinically used both to lo- calize NE GEP tumors by somatostatin receptor scintig- raphy (SRS) and, therapeutically, to control specific hy- persecretion syndromes [9, 12, 28, 33]. Although most of these tumors are positive on SRS, approximately 20-25% of all NE GEP tumors are not detected by SRS when oc- treotide is used as ligand [28, 33]. Interestingly, some of the scintigraphically negative tumors exhibit in vitro bind- ing sites for the "pan-ligand" somatostatin 14 (SST 14), but not for octreotide. This suggests that different soma- tostatin receptor subtypes must exist in NE tumor cells [11, 13]. Recently, five distinct somatostatin receptor (SSTR) subtypes have been identified and characterized by molec- ular cloning and expression studies [1, 2, 4, 5, 15-17, 22, 27, 29-32]. All SSTRs have been shown to be G-protein- coupled and to exhibit seven putative membrane-spanning domains [1, 17]. SSTR subtypes differ in their affinities for specific ligands, such as SST- 14, SST-28, or stable SST analogues, such as octreotide [2, 22]. In vitro experiments showed that lanreotide and octreotide mainly bind S STR 2 with high affinity. Thus, different SSTR subtypes ex- pressed in NE GEP tumors cells may account for the pos- itive or negative results obtained with SRS. In the present study, we investigated the expression of SSTR 1-3 mRNA in both pancreatic NE cell lines and NE GEP tumor tissues. In addition, expression of SSTR at the protein level was studied by chemical crosslinking experi- ments using SST-14 as ligand. Materials and methods Cell culture The NE pancreatic tumor cell lines AR 42 J (rat) [7], RIN 38 (rat) [ 18], BON (human) [6] and QGP 1 (human) [8] were cultured in plas- tic flasks or in 60-mm-plastic petri dishes as previously described. Cells were harvested at >90% confluence. Tumor tissues Tumor tissues of 10 patients with NE GEP tumors were obtained by surgery at the Universitfitsklinikum Rudolf Virchow (UKRV) and Universit~tsklinikum Benjamin Franklin (UKBJ), Berlin, Germany. Informed consent was obtained from all patients, and the study was performed in accordance with the standards set by the ethical com- mittee of the UKRV and UKBJ. Tissue samples were quick-frozen in liquid nitrogen and stored at -80 °C. The NE tumor histology was verified by conventional and immunohistological methods prior to either RNA or protein preparation. Somatostatin receptor scintigraphy (SRS) was performed preop- eratively as previously described [28, 33]. RNA isolation RNA extraction from reference tissue (rat brain) and cells (AR 42 J, RIN 38, BON, QGP 1) was performed by the guanidinium isothio- cyanat/cesiumchloride method as previously described [14]. Subse- quent mRNA preparation was clone by the use of a PolyAtract kit (Promega, Serva, Heidelberg, FRG) according to the manufacturer's instructions. Total RNA from minor amounts of tissue was purified following the method of Chomczynski and Sacchi [3] with minor modifica- tions. Briefly, 10-20 mg tumor tissue were homogenized in guani- dinium thiocyanate homogenization buffer. RNA purification was performed in an acidified solution by phenol/chloroform/isoamyl al- cohol extraction. Aqueous phase was subsequently extracted twice by chloroform. RNA samples were precipitated using isopropanol, the pelleted RNA was washed twice with 75% ethanol, stored in 75% ethanol at -80 °C and redissolved in RNase-free water prior to use for North- ern blotting or reverse transcription. Northern blotting mRNA analysis was done by Northern blotting using 20 ~tg of dena- tured mRNA separated on a formaldehyde-agarose-gel and trans- ferred to a nylon membrane (Hybond N, Amersham, Braunschweig, Germany) by standard capillary blotting methods [14]. pPUC 18 containing mouse SSTR 1 cDNA and pGEM 3 Z con- taining SSTR 2 cDNA were kind gifts from S. Seino, Chiba, Japan. pBluescript SK containing rat SSTR 3 cDNA was a kind gift of W. Meyerhof, Hamburg, FRG. A 0.95-kb Pstl/Sall fragment of the mouse SSTR 1, a 1.2-kb Xbal fragment of the mouse SSTR 2 and a 3.6-kb EcoRI fragment of the rat SSTR 3 were used as SSTR-sub- type-specific cDNA probes. The respective cDNAs were labeled 32 with P-dCTP (NEN) using a random prime labeling kit (Amer- sham, Braunschweig, FRG). Hybridization using SSTR-specific ra- diolabeled cDNA probes was carried out overnight in buffer contain- ing formamide at 42 °C, followed by high-stringency washes. North- ern blots were exposed to X-ray films at -80 °C 24-72 h. Reverse transcription-polymerase chain reaction (RT-PCR) First-strand cDNA for PCR was generated from about 1 gg of total RNA or mRNA. Reverse transcription was carried out according to standard protocols of the manufacturer using M-MLV reverse tran- scriptase (Gibco-BRL, Berlin, Germany). Control experiments were carried out without M-MLV reverse transcriptase or with water in- stead of mRNA solution (control RT). Unless otherwise indicated, the RT mixture was used directly as template for PCR in a 1/500 dilution. Control amplifications were performed with control RT as template to exclude contamination of the reaction mixtures. No am- plified cDNA was found in any of the control experiments. The iden- tity of the amplified cDNA fragments was confirmed by Southern blotting and subsequent hybridization with SSTR-subtype-specific cDNA probes as described below. For semiquantitative determination of SSTR subtype mRNA of NE tumor cell lines, RT mixtures were amplified at consecutive PCR cycles using subtype-specific primer pairs: human SSTR 1: 5'-TCCCAGAACGGGACCTTGAGC and 5'-AACCTGGGCGTGT GGGTGCTA, human SSTR2: 5'-GGCTCTGTGGTGTCAAC- CAAC and 5'-TGCACCATCAACTGGCCAGGT, human SSTR 3: 5'-TCGGTGTCCACGACCTCAGAA and 5'-CGCTACCTGGCC GTGGTACAT, rat SSTR 1: 5'-GGTCAGGGTAGCGCCATTCTC and 5'-TGCGGTGCGTGAGAAGACCAC, rat SSTR 2: 5'-CTG- GCCTCCGGAGCAACCAGT and 5'-CGTGGTCTCATTCAG- CCGGGA, rat SSTR 3: 5'-CTGGACACGTCCCTGGGGAAT and 5'-ATGAAGGCTGTTCGCCAGGCA. The reaction was carried out in 10 mM Tris-HC1 buffer pH 9.0 containing 50 mM KC1, 0.1% Tri- ton X- 100, 1.0 mM MgC12, 100 gM of each dNTR 12.5 pmol of each primer and 2.5 U Taq DNA Polymerase (Promega-Serva, Heidel- berg, FRG) in a final volume of 50 ~tl. Reactions were performed in
`
`Roxane Labs., Inc.
`Exhibit 1038
`Page 002
`
`

`
`92 a Trio Thermocycler (Biometra, G6ttingen, FRG). Following an in- itial denaturing step for 5 min at 95 °C, Taq DNA polymerase was added. The amplification program consists of 28-38 cycles with 30 s denaturation at 94 °C, 1 min annealing at 63 °C and 1.5 min extend- ing at 72 °C. Human and rat glyceraldehyde-3-phosphate dehydrogenase-spe- cific primers (Clontech, ITC, Heidelberg, FRG) were used to ampli- fy the respective cDNA fragment (about 1000 bp in size) as internal standard according to the manufacturer's instructions. Amplificates from consecutive cycles were analyzed by agarose gel electrophoresis (2% agarose), DNA was visualized by ethidium bromide, and photographs of the respective gels were scanned and subsequently examined by densitometry. The PCR resulted in single bands of the expected size (human SSTR 1,417 bp; human SSTR 2, 531 bP; human SSTR 3, 420 bp; rat SSTR 1, 428 bp, rat SSTR 2, 521 bp; rat SSTR 3,551 bp). RNA samples from tumor tissued were analyzed in RT-PCR using a pair of degenerated primers chosen from highly conserved regions of all known SSTR subtypes from rat, mice and hu- mans (srcom 1: 5'-AYCGITAYSTGGCYGTRGTICAYC, srcom 2: 5'-VGG -GTTKGCRCAGCTRTTRGCRTA). The amplification was carried out in 10 mM Tris HC1 buffer at pH 9.0, containing 50 mM KC1, 0.1% Triton X-100, 1.6 mM MgCI 2, 200 gM of each dNTP, 50 pmol of each primer and 2.5 U Taq DNA Polymerase (Promega- Serva, Heidelberg, FRG) in a final volume of 50 gl. Following an initial denaturing step for 3 min at 95 °C, Taq DNA polymerase was added. The amplification program consists of 35 cycles with 40 s denaturation at 95 °C, 1.5 min annealing at 55 °C and 1.5 min ex- tending at 72 °C. Aliquots of the PCR mixtures were analyzed by agarose gel elec- trophoresis using a 2% agarose gel. The PCR resulted in cDNA frag- ments of the expected size of about 500 kb. After electrophoresis, gels were denatured in buffer A (0.5 M NaOH, 1.5 M NaC1) about 30 min and subsequently blotted on positively charged nylon mem- branes (Amersham Hybond N+, Braunschweig, Germany) for a min- imum of 2 h using a standard capillary blotting protocol with buf- fer A as blot buffer. After blotting, membranes were washed briefly in neutralization buffer and baked for 2 h at 80 °C. Southern blot analysis Southern blots of PCR amplificates were probed with fluorescin-la- beled probes of the respective S S TR cDNA fragments. Labeling, hy- bridization and ECL detection were carried out according to the man- ufacturers protocol (Amersham, Braunschweig, Germany) with three high-stringency washes at 65 °C. " Chemical crosslinking of somatostatin receptors with SST-14 Chemical crosslinking of SSTR with radiolabeled SST-14 was per- formed as previously described [19, 20]. Briefly, frozen tumor tis- sue samples or frozen cell pellets from about 106 cells of each cell line (AR 42 J, RIN 38, BON) were homogenized by sonication in cold buffer containing 20 mM Tris-HC1 pH 7.4, 1 mM PMSF, 20% glycerol, diluted at 1 g protein/ml and stored at -20 °C. The protein concentration was determined by a colorimetric method (Sigma, Par- is, France). The solution was diluted at a protein concentration of 1 mg/ml in cold binding buffer containing 20 mM Tris-HC1 pH 7.4, 5 mM MgCI 2, 1 mM PMSF. Then 100 ~tl of the respective protein solution was mixed with 50 gl of 125I-labeled somatostatin-14 (105 dpm) solubilized in 100 mM Tris-HCl pH 7.4, 5 mM MgC12, 1 mM PMSF, 0.1% bovine serum albumin (BSA). The mixture was incubated at 4 °C for 30 min. Chemical crosslinking was achieved by the use of 0.1 mM ethylene-glycol-bis(succinimidyl-succinate) (Sigma, Paris, France), and crosslinking was done at 4 °C for 15 min. Crosslinking was stopped by the addition of 50 gl gel loading buf- fer (125 mM Tris-HC1 pH 7.4, 10% SDS, 50% glycerol, 20% mer- captoethanol, 0.025% bromophenol blue). To confirm specific bind- ing of SST-14, control binding experiments were performed with un- labeled SST-14 in excess. Samples were heated at 95 °C for 2 min and proteins were separated in 12.5% SDS polyacrylamide gel. Gels were dried and autoradiographed at -80 °C. Results SSTR subtype expression in NE tumor cell lines The distribution of SSTR 1, SSTR 2, SSTR 3 was exam- ined by Northern blotting in rat brain as reference tissue and in cell lines from rat (AR 42 J, RIN) and humans (QGP 1, BON). By Northern blotting, readily detectable levels of SSTR 1 mRNA (3.6 kb) were found only in rat brain and RIN insulinoma cells (Fig. t, Table 1). After pro- longed times of exposure, a weak signal for SSTR 1, mRNA was found in the case of AR 42 J cells. SSTR 2 mRNA was detected at high levels in rat brain and in AR 42 J cells, but only at a low signal strength in RIN cells (Fig. 1, Table 1). Transcripts of 2.2 kb hybridized with the SSTR 2 probe. The SSTR 3 mRNA transcripts of approx- imately 4.4 kb were only detected in rat brain and not in any of the cell lines studied (Fig. 1, Table 1). SSTR subtype gene expression was also analyzed by RT-PCR technique using SSTR subtype specific primer pairs in the subsequent PCR step. In all PCR experiments, the amplification products were of the expected sizes Fig. 1 SSTR subtype mRNA expression in NE pancreatic tumor cell lines as revealed by Northern blot analysis. Poly(A) RNA samples from cell lines (RIN 38, QGP 1, BON, AR 42 J) and rat brain (control) were denatured, electrophor- esed, transferred to nylon mem- branes and hybridized with radiolabeled SSTR subtype- specific cDNA probes SSTR 1 3.6 kb- SSTR 2 SSTR 3 2.2 kb- 4.4 kb-
`
`Roxane Labs., Inc.
`Exhibit 1038
`Page 003
`
`

`
`Table 1 Specific somatostatin receptor (SSTR) subtype mRNA ex- pression in neuroendocrine (NE) tumor lines as revealed by North- ern blot analysis (+ hybridization signal; (+) low signal after pro- longed exposure time; - negative) SSTR 1 SSTR 2 SSTR 3 Rat brain + + + QGP 1 - - - BON - - - RIN 38 + (+) - AR 42 J (+) + - 'x" _ok" _¢b" 93 Table 2 SSTR subtype mRNA levels and SST-14 binding in NE pancreatic tumor cell lines. Relative abundance of SSTR subtype transcripts was estimated from the signal strength of the electrophor- esed amplifications yielded by consecutive PCR cycles in relation to glyceraldehyde-3-phosphate dehydrogenase mRNA expression (n.d. not detected; n.t. not tested; RT-PCR: (+), +, ++, +++ relative abun- dance; chemical crosslinking: + signal for specific SST-14 binding, i.e., specific binding of radiolabeled SST-14 to polypeptide of 42, 57, 70 or 90 kDa in SDS-PAGE) RT-PCR Crosslinking SSTR 1 SSTR 2 SSTR 3 SST-14 QGP 1 ++ + n.d. n.t. BON +++ + (+) + AR 42 J ++ +++ (+) + RIN 38 +++ ++ (+) + Table 3 SSTR subtype expression in human NE GEP tumor tissues. RT-PCR analysis was performed using degenerated primer pairs. PCR-generated DNA fragments were analyzed by Southern blot analysis using SSTR subtype-specific probes ]-, (+), +, ++, intensi- ty of the hybridization signal] (SRS somatostatin receptor scinti- graphy (performed preoperatively in all patients: + positivity as described elsewhere [7, 8]); chemical crosslinking: + positive signal for specific SST-14 binding as described in Table 2, p primary tu- mor, m metastasis) Tumor tissue RT-PCR SRS Cross- linking SSTR1 SSTR2 SSTR3 Insulinoma (liver m.) + ++ - - + Atypical carcinoid (p.) - ++ - - + Carcinoid 1 (ovary, m.) (+) ++ - - + Carcinoid 2 (liver m.) + ++ - + + Carcinoid 3 (p.) + ++ - + n.t. Carcinoid 4 (p.) - ++ - + + Carcinoid 5 (liver m.) - ++ - + + Gastrinoma 1 (p.) - ++ - + + Gastrinoma 2 (liver m.) + ++ - + + Gastrinoma 3 (liver m.) (+) ++ - + n.t. (Fig. 2). In addition, the specificity of the primer pairs was confirmed by Southern blotting and subsequent hybridiza- tion of the respective amplified DNA fragments with SSTR subtype specific cDNA probes (data not shown). Figure 1 shows the results of a representative RT-PCR experiment using mRNA of the indicated NE tumor cells lines with subtype specific primer pairs. The abundance of QGP1 BON AR42J RIN38 Fig. 2 SSTR subtype mRNA expression in NE pancreatic tumor cell lines as revealed by RT-PCR analysis. Poly(A) RNA samples from cell lines were reversely transcribed, subsequent PCR was performed using SSTR subtype specific primer pairs (h, human; r, rat). Reac- tion mixtures were analyzed by electrophoresis in ethidium bromide stained agarose gels. The PCR resulted in single bands of the expect- ed size (hSSTR 1 - 417 bp, hSSTR 2 - 531 bp, hSSTR 3 - 420 bp, rSSTR 1 - 428 bp, rSSTR 2 - 521 bp, rSSTR3 - 551 bp) the subtype transcripts (Table 2) was estimated from the signal strength of the electrophoresed amplificates yielded by consecutive cycles in relation to glyceraldehyde-3- phosphate dehydrogenase mRNA expression. SSTR 1 transcripts were detected most prominently in RIN 38 cells and in BON cells, and at lower levels in AR 42 J and QGP 1 cells. The SSTR 2 transcript was found to be the predom- inant SSTR mRNA in AR42 J and QGP 1 cells and to oc- cur at lower levels in BON and RIN cells. The expression levels of SSTR 3 mRNA were very low in both BON and AR 42 J cells, and not detectable in QGP 1 and RIN cells. The expression of SSTR at the protein level was con- firmed by chemical crosslinking experiments using SST- 14 as ligand. As shown in Table 2, all cell lines tested ex- hibited positive signals for SST-14 binding. Cell lines pos- itive by chemical crosslinking showed specific binding of radiolabeled SST-14 to polypeptides of 42, 57, 70 or 90 kDa in SDS-PAGE, owing to the molecular heteroge- neity of the S STR found in this analytical approach [21]. SSTR subtypes in human NE GEP tumors Tumor tissue samples of ten patients preoperatively sub- jected to somatostatin receptor scintigraphy (SRS) were studied by RT-PCR and chemical crosslinking (Table 3). The NE GEP tumors of seven patients (4 with carcinoids, 3 with gastrinomas) were positive on SRS. In three patients with, variously, an insulinoma, an atypical carcinoid and an ileal carcinoid, the primary tumor or metastases failed to be detected by SRS. Interestingly, all tumor tissues in- vestigated were positive by chemical crosslinking with SST-14 (Table 3). As mentioned above, tumor tissues pos- itive by chemical crosslinking showed specific binding of radiolabeled SST-14 polypeptides of 42, 57 or 90 kDa in
`
`Roxane Labs., Inc.
`Exhibit 1038
`Page 004
`
`

`
`94 SDS-PAGE. Thus, all tumor tissues tested exhibited som- atostatin-binding sites in vitro. RT-PCR analysis of the ex- pression of the SSTR subtype transcripts using degener- ated primer pairs revealed a heterogeneous pattern. SSTR 2 mRNA was found at high levels in all tumor tissues, whereas SSTR 1 subtype mRNA was detectable in only six of the ten tissue samples. SSTR 3 transcripts were not de- tected in any of the tumor tissues tested under the PCR con- ditions used. Discussion Binding sites for SST-14, SST-28 and the clinically rele- vant SST analogues octreotide or lanreotide have been found in a large number of NE tumors by in vitro radioli- gand binding techniques [23-26]. Using SRS, in vivo bind- ing of the SST analogue octreotide to NE tumors was initially observed by Lamberts and co-workers [9, 12] and was later confirmed by others [28, 33]. To date, the expres- sion of the known SSTR subtypes (for nomenclature see [2]) has only been examined in a small number of patients with NE tumors of the pancreas and intestine by a non- quantitative RT-PCR approach [ 10]. The authors suggested that only SSTR 2 was functionally relevant, but the study lacks data on the expression of SSTRs at the protein level. In addition, the data obtained were not correlated to in vivo binding conditions, i.e. SRS. In this study, we have assessed the composition of S STR subtypes in human NE tumor tissue by RT-PCR and by in vitro crosslinking using SST-14 as the SSTR ligand as well as by in vivo imaging using SRS. In addition, we studied SSTR subtypes in NE tumor cell lines by Northern blot- ting, RT-PCR and in vitro crosslinking using SST-14 as SSTR ligand. SRS was performed in all ten patients and failed to de- tect tumors in three patients. As demonstrated both by RT-PCR and by crosslinking experiments, all tumor tissues exhibited at least one SSTR subtype transcript and all tu- mor tissues investigated contained SST-14-binding sites. At present, we cannot exclude the possibility that the ex- pression of SSTRs was due to the presence of some nor- mal tissue surrounding the tumors and expressing SSTR. However, our data indicate that at least some NE tumor le- sions were false-negative by SRS. The observed SSTR subtype pattern determined by RT-PCR of NE GEP tumor tissues and cell lines implies a central role for SSTR 2 as target for octreotide or lanreo- tide in tumor diagnostics (SRS) and therapy (control ofhy- persecretion syndromes). However, as SRS-negative tu- mor tissues showed somatostatin-binding sites in vitro, the development of additional SSTR-subtype-specific ligands may be proposed for these tumors to avoid false-negative results in SRS. In addition, the heterogeneous SSTR subtype pattern of NE tumor cell lines with either SSTR 1 (BON, RIN) or SSTR 2 (QGP 1, AR 42 J) as the predominant subtype sug- gests an additional role for other SSTR subtypes. Thus, NE GEP tumor tissues and tumor cell lines should be also stud- ied by semiquantitative PT-PCR to determine the complete SSTR subtype pattern, including the recently described subtypes 4 and 5. Further studies, using SSTR subtype-specific ligands may help to clarify the role of SSTR subtypes other than SSTR 2. Such investigations may lead to improved diag- nosis and treatment of NE GEP tumor disease. Acknowledgements The authors thank I. Eichhorn for expert tech- nical assistance. This study was supported in part by grants from the Deutsche Krebshilfe/Dr.-M.-Scheel-Stiftung (W31/91/Wil), the Verum-Stiftung, and the Deutsche Forschungsgemeinschaft (SFB 366/A5) to B.W. References 1. Bell GI, Reisine T (1993) Molecular biology of somatostatin re- ceptors. Trends Neurosci 16:34-38 2. Bruns C, Weckbecker G, Raulf F, Kaupmann K, Schoeffter P, Hoyer D, Ltibbert H (1994) Molecular pharmacology of soma- tostatin receptor subtypes. Ann NY Acad Sci (in press) 3. Chomczynski R Sacchi N (1987) Single-step method of RNA isolation by acid guanidinium thiocyanata-phenol-chloroform extraction. Anal Biochem 162:156-159 4. Corness JD, Demchyshyn LL, Seeman P, Tol HHM van, Srik- ant CB, Kent G, Patel YC, Niznik HB (1993) A human soma- tostatin receptor (SSTR 3), located on chromosome 22, displays preferential affinity for somatostatin- 14 like peptides. FEBS Lett 321:279-284 5. Demchyshyn LL, Srikant CB, Sunahara RK, Kent G, Seeman P, Tol HHM van, Panetta R, Patel YC, Niznik HB (1993) Cloning and expression of a human somatostatin-14-selective receptor variant (somatostatin receptor 4) located on chromosome 20. Mol Pharmacol 43:894-901 6. Evers BM, Townsend CM, Upp J, Allen E, Hurlbut S, Kim S, Rajaraman S, Singh R Reubi JC, Thompson J (1991) Establish- ment and characterization of human carcinoid in nude mice and effect of various agents on tumor growth. Gastroenterology 101:303-311 7. Jessop NW, Hay RJ (1980) Characteristics of two rat pancreat- ic exocrine cell lines derived from transplantable tumors. In Vi- tro Cell Dev Biol 16:212 8. Kaku M, Nishiyama T, Yagawa K, Abe M (1980) Establishment of a carcinoembryonic antigen-producing cell line from human pancreatic carcinoma. Gann 71:596-601 9. Krenning ER Breeman WAR Kooij PPM, Lameris JS, Bakker WH, Koper JW, Ausema I, Reubi JC, Lamberts SWJ (1989) Lo- calisation of endocrine-related tumours with radioiodinated an- alogue of somatostatin. Lancet I:242-244 10. Kubota A, Yamada Y, Kagimoto S, Shimatsu A, Imamura M, Tsuda K, Imura H, Seino S, Seino Y (1994) Identification of somatostatin receptor subtypes and an implication for the effi- cacy of somatostatin analogue SMS 201-995 in treatment of hu- man endocrine tumors. J Clin Invest 93:132 i-1325 11. Kvols LK, Brown ML, O'Connor MK, Hung JC, Hayostek RJ, Reubi JC, Lamberts SWJ (1993) Evaluation of a radiolabeled somatostatin analog (I-123 octreotide) in the detection and lo- calisation of carcinoids and islet cell tumors. Radiology 187:129-133 12. Lamberts SWJ, Bakker WH, Reubi JC, Krenning EP (1990) Somatostatin-receptor imaging in the localization of endocrine tumors. N Engl J Med 323:1246-1249 13. Lamberts SWJ, Hofland LJ, van Koetsveld PM, Reubi JC, Bruin- ing HA, Bakker WH, Krenning EP (1990) Parallel in vivo and in vitro detection of functional somatostatin receptors in human endocrine pancreatic tumors: consequences with regard to diag-
`
`Roxane Labs., Inc.
`Exhibit 1038
`Page 005
`
`

`
`nosis, localization, and therapy. J Clin Endocrinol Metab 71:566-574 14. Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 15. Panetta R, Greenwood MT, Warszynska A, Demchyshyn LL, Day R, Niznik HB, Srikant CB, Patel YC (1994) Molecular clon- ing, functional characterization, and chromosomal localization of a human somatostatin receptor (somatostatin receptor type 5) with preferential affinity for somatostatin-28. Mol Pharmacol 45:417-427 16. Patel YC, Greenwood M, Kent G, Panetta R, Srikant CB (1993) Multiple gene transcripts of the somatostatin receptor SSTR 2: tissue selective distribution and cAMP regulation. Biochem Biophys Res Commun 192:288-294 17. Patel YC, Greenwood MT, Warszynska A, Panetta R, Srikant CB (1994) All five cloned human somatostatin receptors (hSSTR 1-5) are functionally coupled to adenytyl cyclase. Bio- chem Biophys Res Commun 198:605-612 18. Philippe J, Chick WL, Habener J (1987) Multipotential pheno- typic expression of genes encoding peptide hormones in rat in- sulinoma cell lines. J Clin Invest 79:351-358 19. Prevost G, Lanson M, Thomas F, Veber N, Gonzalez W, Beaupain R, Starzec A, Bogden A (1992) Molecular heteroge- neity of somatostatin analog BIM23014C receptors in human breast carcinoma using cross-linking assay. Cancer Res 52:843- 85O 20. Prevost G, Provost R Salle V, Lanson M, Thomas F (1993) A crosslinking-assay allows the detection of receptors for the som- atostatin analogue, lanreotide in human breast tumors. Eur J Can- cer 11:1589-1592 21. Prevost G, Hosford D, Thomas F (1994) Receptors for somatos- tatin and somatostatin analogs in human breast tumors. Ann NY Acad Sci (in press) 22. Rens-Domiano S, Reisine T (1992) Biochemical and function- al properties of somatostatin receptors. J Neurochem 58: 1987-1996 23. Reubi JC, Haecki WH, Lamberts SWJ (1987) Hormone-produc- ing gastrointestinal tumors contain a high density of somatosta- tin receptors. J Clin Endocrinol Metab 65:1127-1134 95 24. Reubi JC, Maurer R, von Werder K, Torhorst J, Klijn JGM, Lam- berts SWJ (t987) Somatostatin receptors in human endocrine tumors. Cancer Res 47:551-558 25. Reubi JC, Kvols LK, Waser B, Nagorney DM, Heitz PU, Char- boneau JW, Reading CC, Moertel C (1990) Detection of soma- tostatin receptors in surgical and percutaneous needle biopsy samples of carcinoids and islet cell carcinomas. Cancer Res 50:5969-5977 26. Reubi JC, Laissue J, Waser B, Horisberger U, Schaer JC (1994) Expression of somatostatin receptors in normal, inflamed and neoplastic human gastrointestinal tissues. Ann NY Acad Sci, in press 27. Rohrer L, Raulf F, Bruns C, Hofst~idter F, B~ittner R, Schfile R (1993) Cloning and characterization of a fourth human soma- tostatin receptor. Proc Natl Acad Sci USA 90:4196-4200 28. Scherfibl H, B~der M, Fett U, Hamm B, Schmidt-Gayk H, Kop- penhagen K, Dop F-J, Riecken E-O, Wiedenmann B (1993) Som- atostatin-receptor imaging of neuroendocrine gastroenteropan- creatic tumors. Gastroenterology 105:1705-1709 29. Xu Y, Song J, Bruno JR Berelowitz M (1993) Molecular clon- ing and sequencing of a human somatostatin receptor, hSSTR 4. Biochem Biophys Res Commun 193:648-652 30. Yamada Y, Post SR, Wang K, Tager HS, Bell GI, Seino S (1992) Cloning and functional characterization of a family of human and mouse somatostatin receptors expressed in brain, gastroin- testinal tract, and kidney. Proc Natl Acad Sci 89:251-255 31. Yamada Y, Reisine T, Law SF, Ihara Y, Kubota A, Kagimoto S, Seino M, Seino Y, Bell GI, Seino S (1992) Somatostatin recep- tors, an expanding gene family: cloning and functional charac- terization of human SSTR 3, a protein coupled to adenylyl cy- clase. Mol Endocrinol 6:2136-2142 32. Yamada Y, Kagimoto S, Kubota A, Koichiro Y, Masuda K, Yoshimichi S, Ihara 5(, Li Q, Imura H, Seino S, Seino Y (1993) Cloning, functional expression and pharmacological character- ization of a fourth (hSSTR 4) and fifth (hSSTR 5) human som- atostatin receptor subtype. Biochem Biophys Res Commun 195:844-852 33. Zimmer T, Ziegler K, B~ider M, Fett U, Harem B, Riecken E-O, Wiedenmann B (1994) Localisation of neuroendocrine tumors of the upper gastrointestinal tract. Gut 35:471-475
`
`Roxane Labs., Inc.
`Exhibit 1038
`Page 006