- Email: [email protected]
Somatostatin receptor subtype gene expression in human and rodent tumors
Life Soences, Vol 53, pp 85-90 Printed m the USA
Pergamon Press
SOMATOSTATIN RECEPTOR SUBTYPE GENE EXPRESSION IN HUMAN AND RODENT TUMORS Peter A. Eden and John E Taylor *Blomeasure Inc, 27 Maple St, Milford, MA 01757-3650 (Received m final form Aprd 19, 1993) Summary_ Somatostatin (SRIF) analogues display anti-tumor properties beheved to be medmted by specific cell surface somatostatin receptors (SSTR). SSTR subtypes have unique pharmacologacal properties, including speofic GTP-bmdmg protein coupling, aon channel regulation, and cAMP inhibition, therefore, identification of Isotypes expressed in tumor cells facilitates current efforts to design potent antitumor SRIF analogues Human and rodent sohd, transplantable tumors and tumor cell lanes were examined for gene expression of SSTR1, SSTR2 and SSTR3 by reverse transcnptaon of tumor mRNA and subsequent amphfication of cDNA by the polymerase chain reaction, using SSTR subtype-specific ohgonucleotide primers. SSTR2 mRNA transcripts were observed In all of the tumor cell lines examined. SSTR1 gene expression was seen m several human and rat tumor types, and SSTR3 gene expression observed in two rodent tumor types. SSTR mRNAposatave tumors are expected to possess membrane-bound receptors whach could potentially interact with anti-tumor SRW analogues In addition to its classical neuroendocnne, neural, and gastrointestinal properties (1-4), the tetradecapeptlde SRIF may function as a paracnne or autocrane inhibitor of cell proliferation and tissue growth. This activity as most evident with respect to tumor growth, where potent octapeptide SRIF analogues (e.g., BIM-23014, 3-(2-Naphthyl)-D-Ala-cyclo[Cys-Tyr-D-Trp-Lys-Val-Cys]-ThrNH2; SMS-201-995, D-Phe-cyclo[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-ol)have been shown to Inhabit tumor cell prohferation in v u r o and tumor growth t n v t v o (5-10). The use of SRIF analogues to inhibit tumor growth relies upon the presence of specific receptors on the cell surface. Indeed, tumor cells have been shown to express SSTRs, strongly suggesting that SRIF may play a role in tumor cell function (4). The diverse biological effects of SRIF are mediated by a family of receptors with chsunct but overlapping tissue distributions, umque pharmacological properties and potentially different functions (11) such as adenylyl cyclase, aon channel and protein phosphatase regulaUon (12, 13, 14). Studtes of tissue distnbunon have shown a wade range of SSTR expression including the central nervous system and peripheral tissues such as stomach, antestine and pancreas (15). The recent cloning of several SSTR genes has increased understanding of SSTR structure and function. To date subtypes SSTR1, SSTR2, SSTR3, SSTR4 and SSTR5 have been cloned and partially characterized (15, 16, 17, 18). The S STRs belong to a superfamily of GTP-blndmg protein (G-protem)-coupled receptors that have seven membrane-spanning dommns (19), extracellular sites for N-linked glycosylation, and mtracellular serme/threonine residues which could be phosphorylated by senne/threonine-speclfic klnases. Alternative splicing of SSTR2 mRNA (20) and tumor-specific post-translataonal processing of SSTR2 (21) provades additional structural Correspondence: Peter A. Eden, Ph.D. Baomeasure Inc, 27 Maple St, Milford, MA 01757-3650 0024-3205/93 $6 00 + 00 Copyright © 1993 Pergamon Press Ltd All rights reserved
86
Somatostatin Receptor Expression m Tumors
Vol. 53, No. 1, 1993
(and hkely functional) heterogeneity. SSTR1 apparently does not couple to G-proteins but binds SRIF with high affinity; SSTR2 is associated wxth pertussis toxm-sensiuve G-proteins but has not been shown to mediate SRIF mlubltlOn of adenylyl cyclase; SSTR3 binds SRIF with high affinity and couples to G-proteins and adenylyl cyclase (11). Although the SSTR subtypes share many biochemical and pharmacological charactenstlcs, their sequences are relatively divergent (15). SSTR sequence dissinularity and alternative processing of isotypes likely promote the diverse biological actions of SRIF due to potent:ally disparate receptor-coupled signal transduclaon mechanisms. Reverse transcription-polymerase chain reaction (RT-PCR) is a highly sensmve method of gene expression analysis which specifically detects the presence of messenger RNA. RT-PCR Involves several steps: isolauon of RNA from cells or tissue, generation of a DNA copy of the mRNA transcript (cDNA) using a reverse transcnptase enzyme, and amplification of cDNA by the PCR (22) using oligonucleotide primers specific for the cDNA of interest. Due to the sensltave nature of the PCR, even low levels of mRNA expression which might prove undetectable using Northern blotting techniques can be observed. This technique was applied to human and rodent sohd, transplantable tumors or tumor cell lines in order to detect gene expression of SSTR1, SSTR2 and SSTR3 subtypes. Methods RNA Isolation. Tumor tassue samples (listed m Table I) originally obtained from ATCC were selected from the Biomeasure, Inc. tumor bank and passaged either as m v t t r o cell cultures, or sohd, transplantable tumors m mice as described previously (10). Total cellular RNA was isolated from tumor ussue using the Tri-reagent RNA isolation k~t available from Molecular Research Center, lnc., Cmcinnata, OH. After preclpltauon with 2 volumes of ethanol and .3 M sodmm acetate, RNA was suspended m sterile, filtered water. Concentrations of RNA were determined by A260 and 1% agarose/6% formaldehyde gel electrophoresls, and material stored at -80°C until use. Ohgonucleotlde Primers and SSTR clones. Ohgonucleotlde PCR primers (typically 17-21 nucleoudes in length) were complementary to sequences that either flank the protein cochng regions or exast within 5' translated regions of the SSTR genes Each primer was designed with sufficient nucleotIde sequence divergence so as to preclude cross-reactivity with known SSTR famdy members. SSTR cDNA clones constructed m pGEM-3Z cloning vector (Promega) were provided by Dr. G 1. Bell, U. Chicago. RT-PCR: Reverse transcription was performed m 500 Ixl Eppendorf tubes according to instructions included with the Gene Amp RNA PCR kit (obtained from Perkm Elmer Cetus), with the following modificat:ons: prior to addition of reverse transcnptase enzyme, the RT mixture (containing 1 ktg RNA, 5 mM MgCI2, 1X PCR II buffer [.5M KC1, .1M Tns-HCI, pH 8.3], 1 mM each deoxynucleosldetnphosphate, 1 U RNAse inhibitor, 2.5 IIM random hexamer primers) recewed 1 U (0.4 l.tg) RNAse-free DNAse 1 (Promega) and incubated at 37°C for 30 nun to destroy any genonuc DNA. The DNAse enzyme was then macuvated by mcubauon at 95°C for 5 nun. After coohng to room temperature, 50 U reverse transcnptase enzyme and a mineral oil overlay were added, and reverse transcription carried out as follows: the RT rmxtures were Incubated at 42°C for 15 mm., the temperature increased to 99°C for 5 mm., then soaked for 5 mln. at 5°C. The thermal cycling was carded out m a Minicycler thermal cycler (MJ Research, lnc., Watertown, MA). Resultant cDNA was treated with chloroform to remove residual nuneral oil and allquoted for PCR. RT mixtures devoid of reverse transcnptase enzyme served as controls for presence of residual genonuc DNA PCR DNA amphfication of SSTR cDNA transcripts was performed using a GeneAmp kit (Perkin Elmer) m 500 I.tl Eppendorf tubes, as follows" RT mixtures received 75 pmol PCR pnmer/reacuon, 0.5 I~1 (2 5 U) "Taq" DNA polymerase/reaction, 2 mM MgC12, 1X PCR II reaction buffer and a mineral oll overlay. Final sample volume was 100 I.tl and cychng condmons used m the Mlnicycler thermal cycler were as follows: lmtial denaturauon at 95°C for 2 mm, 94"C for 1 min (denaturation), 68°C for 1 mm (primer annealing and extension). The 35 cycle thermal profile included an 8 nun, 68°C extension step at the end of the program Ahquots (8 Ixl) of each reactaon
Vol. 53, No 1, 1993
Somatostatln Receptor Expression in Tumors
87
were examined by 1.5% agarose gel electmphoresis in T A E buffer (.04 M Tris-acetate, .001 M EDTA) at 70 V for 1.5 h using a mini-sub gel electrophoresis apparatus (BioRad). Upon compleuon o f electrophoresis the D N A was stained with ethidmm bronude, visuahzed using a Spectrohne transllluminator and photographed with a Polaroid instant camera using Polaroid type 667 film
Results Tumor samples examined m this study showed evidence o f SSTR1, SSTR2 and SSTR3 gene expression. SSTR2 m R N A was detected m every tumor cell type exarmned; th~s supports b i n d i n g d a t a i n d i c a t i n g a w i d e - s p r e a d d ~ s t r i b u u o n o f the r e c e p t o r d e t e r m i n e d by membrane/ra&ohgand binding (18/23, 19/24). The mouse B 16 melanoma cells showed quahtatively weak expression o f SSTR2 and SSTR3. Non-tumor prostate tissue was used with the SSTR2specific primers and resulted in no product, providing a negative control for this primer set (not shown). The results o f the RT-PCR analysis are presented in Table I.
TABLE I Somatostain R~¢eptor Subtype Gene Expression
Humim turlaors H S C L C NCI-H69 (cl) H S C L C NCI-H345 (cl) H S C L C LX-1 (s) melanoma H187 (s) H N S C L C A549 (cl) H N S C L C H165 (cl) colon CX-5 (s) breast M D A MB231 (cl) breast ZR-75-1 (cl) breast MCF7 (cl)
SSTR1 + + -
SSTR2
SSTR3
+ + + + + + + + + +
Rodent tumors acmar pancreas AR42J (cl) breast MT/W9a-R (s) melanoma B16 (cl) hepatoma M5123 (s) prostate R-3327 (s)
+ + +/-
+ + +/+ +
+/+ -
HSCLC: human small cell lung carcinoma; HNSCLC: human non-small cell lung carcinoma. The "(cl)" and "(s)" denote tumor cell hne or sohd tumor, respectively. The "+" denotes presence of mRNA expression; the "-" denotes no detectable gene expression, and "+/-" m&cates apparent low gene expression.
In order to ensure PCR products were amplified from the appropriate SSTR genes, i.e. no cross-reactivity occurred due to presence m tumor cells of other SSTR transcripts, SSTR1, SSTR2 and SSTR3 c D N A clones were used as template with the SSTR1, SSTR2 and SSTR3-speclfiC primer pairs Under the stringent condmons employed for RT-PCR in this study (primer annealing temperatures not lower than 68°C), no non-specific PCR products were seen, indicating that cross-
88
Somatostatm Receptor Expression m Tumors
Vol 53, No 1, 1993
reactivaty did not occur (not shown). Furthermore, the mRNA transcript-derived amphfication products were of the expected sizes: SSTR1, 409 base pairs (bp); SSTR2, 342 bp; SSTR3, 361 bp, and corresponded in size to respective SSTR clone-derived PCR products. Minute amounts of residual genormc DNA can produce false posluve results when using a sensitive gene expression analysis technique such as RT-PCR. Therefore, m all reactions DNA was destroyed with DNAse prior to reverse transcription. Reactions devoid of reverse transcnptase enzyme served as controls for the possibility of genomic DNA presence as these should not contain cDNA template material. Presence of a PCR product m the reactions containing reverse transcriptase therefore indicated DNA amphficauon products were generated only from mRNA template expressed in tumor cells. RT-PCR reaclaons containing no added RNA template did not result in amphfied product with any of the SSTR-speclfic pnmer paws employed in this study. The addiuon of DNAse did not appear to affect the random hexamer DNA primers used in the reverse transcnpUon procedure. Figure 1 shows representative RT-PCR reactions using SSTR1, SSTR2 and SSTR3specific primers with tumor RNA. Slrmlar results were seen w~th the other tumor samples used in this study.
+
2
3
+
-
+
4
5
6
+
7
8
bp 1078
-
603
-
310
-
Fig 1 SSTR subtype gene expression in tumors: agarose gel containing amplified cDNA from representative SSTR mRNA transcripts. The size in base pmrs (bp) of marker fragments is indicated. Presence or absence of reverse transcriptase enzyme in each reaction is denoted by a "+" or "-" above appropriate lanes. Lane one contains fragments of Hae III digest of Phage X174 used as a molecular weight marker; lanes 2 and 3 contain SSTR1 DNA amplified from HSCLC NCI-H69 tumor RNA with and without reverse transcriptase enzyme (rt enzyme), respectwely; lanes 4 and 5 contain SSTR2 DNA amplified from HSCLC NCI-H69 tumor RNA with and without rt enzyme, respecuvely; lanes 6 and 7 contain SSTR3 DNA amplified from rat M5123 tumor RNA with and without rt enzyme, respectively; lane 8 contains a representative control RT-PCR reaction using SSTR2-specxfic primers with no added template RNA. DNA fragments corresponding to SSTR1 (409 bp, lane 2), SSTR2 (342 bp, lane 4) and SSTR3 (361 bp, lane 6) are clearly seen only when reverse transcnptase enzyme was included in the RT-PCR reacuon
Discussion In this study we report gene expression of SSTR subtypes in human and rodent tumors. We
Vol 53, No 1, 1993
Somatostatln Receptor Expression m Tumors
89
have taken measures to ensure the specificity of the amphfied cDNA products. (a) PCR products generated under stringent DNA amphfication conditions were of the predxcted fragment sizes based on primer anneahng sites chosen from known SSTR cDNA sequences. (b) The PCR primers specific for each SSTR subtype were designed either from untranslated regions of the genes (which have been shown in many gene families to be umque to each particular gene) or from non-conserved 5' coding regions. The primers used in this study did not amplify other SSTR genes as evidenced by results of PCR experiments using all primer sets w~th cloned SSTR1, SSTR2 and SSTR3 genes. (c) Primers specific to SSTR1, SSTR2 or SSTR3 amplified from RNA only one DNA band corresponding in size to amphfied DNA from each respective cloned gene. (d) Treatment of the samples with RNAse prior to RT-PCR resulted xn no amplified product under any condition, inchcatang that amplification was due solely to mRNA presence (data not shown). (e) Presence of DNAse as well as the presence or absence of reverse transcnptase enzyme m each RT-PCR reactaon ensured mRNA-denved DNA amphficaUon. All of the above reasons indicate that the amphfied DNA fragments seen in thts study correspond to SSTR1, SSTR2 or SSTR3 mRNA. Previous radioligand binding studies (23, 24) have demonstrated that the same panel of tumors or tumor cell hnes examined in this study were variably enriched m SRIF receptors which chsplayed the pharmacological specificity of the SSTR2 subtype. Quantitatively, the htghest SSTR2 concentrauons were detected in the AR42J pancreas and the HSCLC NCI-H69 and NCI-H345 tumor cell lines. Much lower levels of SSTR2 were seen in colon, hepatoma, prostate and breast, B 16 melanoma cells showed very low levels of SSTR2, winch correlates with the quahtatlvely weak gene expresslon of SSTR2 observed in these cells. These observations support the present study by demonstrating that SSTR2 gene expression results in detectable surface membrane receptors. Comparable receptor binding assays for SSTR1 and SSTR3 membrane receptor expression have not been reported. In conclusion, RT-PCR was used to detect gene expression of specific somatostatm receptor subtypes m human and rodent tumors and tumor cell hnes. The use of an mRNA template for DNA amphficatlon by the PCR provided a sensitive and selective method of gene expresston analysis Tumor cell types exhibiting SSTR gene expresolon are expected to possess the corresponding SSTR(s) on their cell surfaces, and such reformation facihtates the development of anti-tumor SRIF analogues Acknowledgments We thank Dr. Graeme I. Bell and Dr. Kazukl Yasuda for SSTR clones and many helpful discussions. R¢ferences 1. 2. 3. 4 5. 6. 7 8. 9. 10. 11. 12.
P. BRAZEAU, W. VALE, R. BURGUS, N. LING, M BUTCHER, J. RIVIER and R GUILLEMIN, Science 179 77-79 (1973). S. REICHLIN, N. Engl. J. Med. 309 1495-1501 (1983) S. REICHLIN, J. Lab. Chn. Med. 109 320-326 (1987). J. EPELBAUM, Prog. Neuroblol. 27 63-100 (1986). A.V. SCHALLY, Cancer Res. 48 6977-6895 (1988). J E. TAYLOR, A.E. BOGDEN, J.-P. MOREAU and D.H COY, Bxochem. Blophys. Res. Comm. 153 81-86 (1988). H. PARMER, A.E. BOGDEN, M. MOLLARD, B. de ROUGE, R H. PHILLIPS, and S L. LIGHTMAN, Cancer Treat. Rev. 16 95-115 (1989). S.W.J LAMBERTS, E.P. KRENNING, and J.-C. REUBI, Endocnne Rev 12 450482 (1991). A.E. BOGDEN, J.E. TAYLOR, J.-P MOREAU, and D.H. COY, Cancer Res. 50 2646-2650 (1990a). A.E. BOGDEN, J.E. TAYLOR, J -P. MOREAU, D H. COY, and D.J. LePAGE, Cancer Res. 50 4360-4365 (1990b). Y. YAMADA, T. REISINE, S.F. LAW, Y. IHARA, A. KUBOTA, S KAGIMOTO, M. SEINO, G.I. BELL, and S. SEINO, Molec. Endo. 6 2136-2142 (1992). K.H. JACOBS, K. AKTORIES, and G SCHULTZ, Nature 303 177-178 (1983)
90
13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.
Somatostatin Receptor Expressionin Tumors
Vol 53, No 1, 1993
H.-L. WANG, T. REISINE, and M. DICHTER, NeuroscL 8 335-342 (1990). C. LIEBOW, C. RE1LLY, M. SERRANO and A.V. SCHALLY, Proc. Natl. Acad Scl USA 86 2003-2007 (1989). Y. YAMADA, S.R POST, K. WANG, H.S. TAGER, G.I. BELL and S. SEINO, Proc Natl. Acad. Sci. 89 251-255 (1992). K. YASUDA, S. RENS-DOMIANO, C. D. BREDER, S. F. LAW, C. B. SAPER, T. REISINE and G. I. BELL, J. Biol. Chem., 267 20422-20428 (1992). A.-M. O'CARROLL, S. J. LOLAIT, M. KONIG and L. C. MAHAN, Mol. Pharmacol., 42 939-946 (1992). J.F. BRUNO, Y. XU, J. SONG and M. BERELOWlTZ, Proc. Natl. Acad. ScL USA, 89 11151-11155 (1992). H.G. DOHLMAN, J. THORNER, M.G. CARON and R.J LEFKOWlTZ, Ann. Rev. Blochem. 60 653-688 (1991). M. VANETFI, M. KOUBA, X. WANG, G. VOGT and V. HOLLT, FEBS 311 290-294 (1992). J.E. TAYLOR, P.A. EDEN and T. REISINE, In Preparataon. R.K. SAIKI, Science 239 487-491 (1988). J.E. TAYLOR, Proc. Winter Neuropeplade Conf., Brockenndge, CO (1993). J.E. TAYLOR, S.R. KEYES, and A.E. BOGDEN, Growth Factors, Peptldes and Receptors '92 XIIth George Washington Unlvers~ty Spring Symposium Abst. # 141, Washington, D.C (1992)
Pergamon Press
SOMATOSTATIN RECEPTOR SUBTYPE GENE EXPRESSION IN HUMAN AND RODENT TUMORS Peter A. Eden and John E Taylor *Blomeasure Inc, 27 Maple St, Milford, MA 01757-3650 (Received m final form Aprd 19, 1993) Summary_ Somatostatin (SRIF) analogues display anti-tumor properties beheved to be medmted by specific cell surface somatostatin receptors (SSTR). SSTR subtypes have unique pharmacologacal properties, including speofic GTP-bmdmg protein coupling, aon channel regulation, and cAMP inhibition, therefore, identification of Isotypes expressed in tumor cells facilitates current efforts to design potent antitumor SRIF analogues Human and rodent sohd, transplantable tumors and tumor cell lanes were examined for gene expression of SSTR1, SSTR2 and SSTR3 by reverse transcnptaon of tumor mRNA and subsequent amphfication of cDNA by the polymerase chain reaction, using SSTR subtype-specific ohgonucleotide primers. SSTR2 mRNA transcripts were observed In all of the tumor cell lines examined. SSTR1 gene expression was seen m several human and rat tumor types, and SSTR3 gene expression observed in two rodent tumor types. SSTR mRNAposatave tumors are expected to possess membrane-bound receptors whach could potentially interact with anti-tumor SRW analogues In addition to its classical neuroendocnne, neural, and gastrointestinal properties (1-4), the tetradecapeptlde SRIF may function as a paracnne or autocrane inhibitor of cell proliferation and tissue growth. This activity as most evident with respect to tumor growth, where potent octapeptide SRIF analogues (e.g., BIM-23014, 3-(2-Naphthyl)-D-Ala-cyclo[Cys-Tyr-D-Trp-Lys-Val-Cys]-ThrNH2; SMS-201-995, D-Phe-cyclo[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-ol)have been shown to Inhabit tumor cell prohferation in v u r o and tumor growth t n v t v o (5-10). The use of SRIF analogues to inhibit tumor growth relies upon the presence of specific receptors on the cell surface. Indeed, tumor cells have been shown to express SSTRs, strongly suggesting that SRIF may play a role in tumor cell function (4). The diverse biological effects of SRIF are mediated by a family of receptors with chsunct but overlapping tissue distributions, umque pharmacological properties and potentially different functions (11) such as adenylyl cyclase, aon channel and protein phosphatase regulaUon (12, 13, 14). Studtes of tissue distnbunon have shown a wade range of SSTR expression including the central nervous system and peripheral tissues such as stomach, antestine and pancreas (15). The recent cloning of several SSTR genes has increased understanding of SSTR structure and function. To date subtypes SSTR1, SSTR2, SSTR3, SSTR4 and SSTR5 have been cloned and partially characterized (15, 16, 17, 18). The S STRs belong to a superfamily of GTP-blndmg protein (G-protem)-coupled receptors that have seven membrane-spanning dommns (19), extracellular sites for N-linked glycosylation, and mtracellular serme/threonine residues which could be phosphorylated by senne/threonine-speclfic klnases. Alternative splicing of SSTR2 mRNA (20) and tumor-specific post-translataonal processing of SSTR2 (21) provades additional structural Correspondence: Peter A. Eden, Ph.D. Baomeasure Inc, 27 Maple St, Milford, MA 01757-3650 0024-3205/93 $6 00 + 00 Copyright © 1993 Pergamon Press Ltd All rights reserved
86
Somatostatin Receptor Expression m Tumors
Vol. 53, No. 1, 1993
(and hkely functional) heterogeneity. SSTR1 apparently does not couple to G-proteins but binds SRIF with high affinity; SSTR2 is associated wxth pertussis toxm-sensiuve G-proteins but has not been shown to mediate SRIF mlubltlOn of adenylyl cyclase; SSTR3 binds SRIF with high affinity and couples to G-proteins and adenylyl cyclase (11). Although the SSTR subtypes share many biochemical and pharmacological charactenstlcs, their sequences are relatively divergent (15). SSTR sequence dissinularity and alternative processing of isotypes likely promote the diverse biological actions of SRIF due to potent:ally disparate receptor-coupled signal transduclaon mechanisms. Reverse transcription-polymerase chain reaction (RT-PCR) is a highly sensmve method of gene expression analysis which specifically detects the presence of messenger RNA. RT-PCR Involves several steps: isolauon of RNA from cells or tissue, generation of a DNA copy of the mRNA transcript (cDNA) using a reverse transcnptase enzyme, and amplification of cDNA by the PCR (22) using oligonucleotide primers specific for the cDNA of interest. Due to the sensltave nature of the PCR, even low levels of mRNA expression which might prove undetectable using Northern blotting techniques can be observed. This technique was applied to human and rodent sohd, transplantable tumors or tumor cell lines in order to detect gene expression of SSTR1, SSTR2 and SSTR3 subtypes. Methods RNA Isolation. Tumor tassue samples (listed m Table I) originally obtained from ATCC were selected from the Biomeasure, Inc. tumor bank and passaged either as m v t t r o cell cultures, or sohd, transplantable tumors m mice as described previously (10). Total cellular RNA was isolated from tumor ussue using the Tri-reagent RNA isolation k~t available from Molecular Research Center, lnc., Cmcinnata, OH. After preclpltauon with 2 volumes of ethanol and .3 M sodmm acetate, RNA was suspended m sterile, filtered water. Concentrations of RNA were determined by A260 and 1% agarose/6% formaldehyde gel electrophoresls, and material stored at -80°C until use. Ohgonucleotlde Primers and SSTR clones. Ohgonucleotlde PCR primers (typically 17-21 nucleoudes in length) were complementary to sequences that either flank the protein cochng regions or exast within 5' translated regions of the SSTR genes Each primer was designed with sufficient nucleotIde sequence divergence so as to preclude cross-reactivity with known SSTR famdy members. SSTR cDNA clones constructed m pGEM-3Z cloning vector (Promega) were provided by Dr. G 1. Bell, U. Chicago. RT-PCR: Reverse transcription was performed m 500 Ixl Eppendorf tubes according to instructions included with the Gene Amp RNA PCR kit (obtained from Perkm Elmer Cetus), with the following modificat:ons: prior to addition of reverse transcnptase enzyme, the RT mixture (containing 1 ktg RNA, 5 mM MgCI2, 1X PCR II buffer [.5M KC1, .1M Tns-HCI, pH 8.3], 1 mM each deoxynucleosldetnphosphate, 1 U RNAse inhibitor, 2.5 IIM random hexamer primers) recewed 1 U (0.4 l.tg) RNAse-free DNAse 1 (Promega) and incubated at 37°C for 30 nun to destroy any genonuc DNA. The DNAse enzyme was then macuvated by mcubauon at 95°C for 5 nun. After coohng to room temperature, 50 U reverse transcnptase enzyme and a mineral oil overlay were added, and reverse transcription carried out as follows: the RT rmxtures were Incubated at 42°C for 15 mm., the temperature increased to 99°C for 5 mm., then soaked for 5 mln. at 5°C. The thermal cycling was carded out m a Minicycler thermal cycler (MJ Research, lnc., Watertown, MA). Resultant cDNA was treated with chloroform to remove residual nuneral oil and allquoted for PCR. RT mixtures devoid of reverse transcnptase enzyme served as controls for presence of residual genonuc DNA PCR DNA amphfication of SSTR cDNA transcripts was performed using a GeneAmp kit (Perkin Elmer) m 500 I.tl Eppendorf tubes, as follows" RT mixtures received 75 pmol PCR pnmer/reacuon, 0.5 I~1 (2 5 U) "Taq" DNA polymerase/reaction, 2 mM MgC12, 1X PCR II reaction buffer and a mineral oll overlay. Final sample volume was 100 I.tl and cychng condmons used m the Mlnicycler thermal cycler were as follows: lmtial denaturauon at 95°C for 2 mm, 94"C for 1 min (denaturation), 68°C for 1 mm (primer annealing and extension). The 35 cycle thermal profile included an 8 nun, 68°C extension step at the end of the program Ahquots (8 Ixl) of each reactaon
Vol. 53, No 1, 1993
Somatostatln Receptor Expression in Tumors
87
were examined by 1.5% agarose gel electmphoresis in T A E buffer (.04 M Tris-acetate, .001 M EDTA) at 70 V for 1.5 h using a mini-sub gel electrophoresis apparatus (BioRad). Upon compleuon o f electrophoresis the D N A was stained with ethidmm bronude, visuahzed using a Spectrohne transllluminator and photographed with a Polaroid instant camera using Polaroid type 667 film
Results Tumor samples examined m this study showed evidence o f SSTR1, SSTR2 and SSTR3 gene expression. SSTR2 m R N A was detected m every tumor cell type exarmned; th~s supports b i n d i n g d a t a i n d i c a t i n g a w i d e - s p r e a d d ~ s t r i b u u o n o f the r e c e p t o r d e t e r m i n e d by membrane/ra&ohgand binding (18/23, 19/24). The mouse B 16 melanoma cells showed quahtatively weak expression o f SSTR2 and SSTR3. Non-tumor prostate tissue was used with the SSTR2specific primers and resulted in no product, providing a negative control for this primer set (not shown). The results o f the RT-PCR analysis are presented in Table I.
TABLE I Somatostain R~¢eptor Subtype Gene Expression
Humim turlaors H S C L C NCI-H69 (cl) H S C L C NCI-H345 (cl) H S C L C LX-1 (s) melanoma H187 (s) H N S C L C A549 (cl) H N S C L C H165 (cl) colon CX-5 (s) breast M D A MB231 (cl) breast ZR-75-1 (cl) breast MCF7 (cl)
SSTR1 + + -
SSTR2
SSTR3
+ + + + + + + + + +
Rodent tumors acmar pancreas AR42J (cl) breast MT/W9a-R (s) melanoma B16 (cl) hepatoma M5123 (s) prostate R-3327 (s)
+ + +/-
+ + +/+ +
+/+ -
HSCLC: human small cell lung carcinoma; HNSCLC: human non-small cell lung carcinoma. The "(cl)" and "(s)" denote tumor cell hne or sohd tumor, respectively. The "+" denotes presence of mRNA expression; the "-" denotes no detectable gene expression, and "+/-" m&cates apparent low gene expression.
In order to ensure PCR products were amplified from the appropriate SSTR genes, i.e. no cross-reactivity occurred due to presence m tumor cells of other SSTR transcripts, SSTR1, SSTR2 and SSTR3 c D N A clones were used as template with the SSTR1, SSTR2 and SSTR3-speclfiC primer pairs Under the stringent condmons employed for RT-PCR in this study (primer annealing temperatures not lower than 68°C), no non-specific PCR products were seen, indicating that cross-
88
Somatostatm Receptor Expression m Tumors
Vol 53, No 1, 1993
reactivaty did not occur (not shown). Furthermore, the mRNA transcript-derived amphfication products were of the expected sizes: SSTR1, 409 base pairs (bp); SSTR2, 342 bp; SSTR3, 361 bp, and corresponded in size to respective SSTR clone-derived PCR products. Minute amounts of residual genormc DNA can produce false posluve results when using a sensitive gene expression analysis technique such as RT-PCR. Therefore, m all reactions DNA was destroyed with DNAse prior to reverse transcription. Reactions devoid of reverse transcnptase enzyme served as controls for the possibility of genomic DNA presence as these should not contain cDNA template material. Presence of a PCR product m the reactions containing reverse transcriptase therefore indicated DNA amphficauon products were generated only from mRNA template expressed in tumor cells. RT-PCR reaclaons containing no added RNA template did not result in amphfied product with any of the SSTR-speclfic pnmer paws employed in this study. The addiuon of DNAse did not appear to affect the random hexamer DNA primers used in the reverse transcnpUon procedure. Figure 1 shows representative RT-PCR reactions using SSTR1, SSTR2 and SSTR3specific primers with tumor RNA. Slrmlar results were seen w~th the other tumor samples used in this study.
+
2
3
+
-
+
4
5
6
+
7
8
bp 1078
-
603
-
310
-
Fig 1 SSTR subtype gene expression in tumors: agarose gel containing amplified cDNA from representative SSTR mRNA transcripts. The size in base pmrs (bp) of marker fragments is indicated. Presence or absence of reverse transcriptase enzyme in each reaction is denoted by a "+" or "-" above appropriate lanes. Lane one contains fragments of Hae III digest of Phage X174 used as a molecular weight marker; lanes 2 and 3 contain SSTR1 DNA amplified from HSCLC NCI-H69 tumor RNA with and without reverse transcriptase enzyme (rt enzyme), respectwely; lanes 4 and 5 contain SSTR2 DNA amplified from HSCLC NCI-H69 tumor RNA with and without rt enzyme, respecuvely; lanes 6 and 7 contain SSTR3 DNA amplified from rat M5123 tumor RNA with and without rt enzyme, respectively; lane 8 contains a representative control RT-PCR reaction using SSTR2-specxfic primers with no added template RNA. DNA fragments corresponding to SSTR1 (409 bp, lane 2), SSTR2 (342 bp, lane 4) and SSTR3 (361 bp, lane 6) are clearly seen only when reverse transcnptase enzyme was included in the RT-PCR reacuon
Discussion In this study we report gene expression of SSTR subtypes in human and rodent tumors. We
Vol 53, No 1, 1993
Somatostatln Receptor Expression m Tumors
89
have taken measures to ensure the specificity of the amphfied cDNA products. (a) PCR products generated under stringent DNA amphfication conditions were of the predxcted fragment sizes based on primer anneahng sites chosen from known SSTR cDNA sequences. (b) The PCR primers specific for each SSTR subtype were designed either from untranslated regions of the genes (which have been shown in many gene families to be umque to each particular gene) or from non-conserved 5' coding regions. The primers used in this study did not amplify other SSTR genes as evidenced by results of PCR experiments using all primer sets w~th cloned SSTR1, SSTR2 and SSTR3 genes. (c) Primers specific to SSTR1, SSTR2 or SSTR3 amplified from RNA only one DNA band corresponding in size to amphfied DNA from each respective cloned gene. (d) Treatment of the samples with RNAse prior to RT-PCR resulted xn no amplified product under any condition, inchcatang that amplification was due solely to mRNA presence (data not shown). (e) Presence of DNAse as well as the presence or absence of reverse transcnptase enzyme m each RT-PCR reactaon ensured mRNA-denved DNA amphficaUon. All of the above reasons indicate that the amphfied DNA fragments seen in thts study correspond to SSTR1, SSTR2 or SSTR3 mRNA. Previous radioligand binding studies (23, 24) have demonstrated that the same panel of tumors or tumor cell hnes examined in this study were variably enriched m SRIF receptors which chsplayed the pharmacological specificity of the SSTR2 subtype. Quantitatively, the htghest SSTR2 concentrauons were detected in the AR42J pancreas and the HSCLC NCI-H69 and NCI-H345 tumor cell lines. Much lower levels of SSTR2 were seen in colon, hepatoma, prostate and breast, B 16 melanoma cells showed very low levels of SSTR2, winch correlates with the quahtatlvely weak gene expresslon of SSTR2 observed in these cells. These observations support the present study by demonstrating that SSTR2 gene expression results in detectable surface membrane receptors. Comparable receptor binding assays for SSTR1 and SSTR3 membrane receptor expression have not been reported. In conclusion, RT-PCR was used to detect gene expression of specific somatostatm receptor subtypes m human and rodent tumors and tumor cell hnes. The use of an mRNA template for DNA amphficatlon by the PCR provided a sensitive and selective method of gene expresston analysis Tumor cell types exhibiting SSTR gene expresolon are expected to possess the corresponding SSTR(s) on their cell surfaces, and such reformation facihtates the development of anti-tumor SRIF analogues Acknowledgments We thank Dr. Graeme I. Bell and Dr. Kazukl Yasuda for SSTR clones and many helpful discussions. R¢ferences 1. 2. 3. 4 5. 6. 7 8. 9. 10. 11. 12.
P. BRAZEAU, W. VALE, R. BURGUS, N. LING, M BUTCHER, J. RIVIER and R GUILLEMIN, Science 179 77-79 (1973). S. REICHLIN, N. Engl. J. Med. 309 1495-1501 (1983) S. REICHLIN, J. Lab. Chn. Med. 109 320-326 (1987). J. EPELBAUM, Prog. Neuroblol. 27 63-100 (1986). A.V. SCHALLY, Cancer Res. 48 6977-6895 (1988). J E. TAYLOR, A.E. BOGDEN, J.-P. MOREAU and D.H COY, Bxochem. Blophys. Res. Comm. 153 81-86 (1988). H. PARMER, A.E. BOGDEN, M. MOLLARD, B. de ROUGE, R H. PHILLIPS, and S L. LIGHTMAN, Cancer Treat. Rev. 16 95-115 (1989). S.W.J LAMBERTS, E.P. KRENNING, and J.-C. REUBI, Endocnne Rev 12 450482 (1991). A.E. BOGDEN, J.E. TAYLOR, J.-P MOREAU, and D.H. COY, Cancer Res. 50 2646-2650 (1990a). A.E. BOGDEN, J.E. TAYLOR, J -P. MOREAU, D H. COY, and D.J. LePAGE, Cancer Res. 50 4360-4365 (1990b). Y. YAMADA, T. REISINE, S.F. LAW, Y. IHARA, A. KUBOTA, S KAGIMOTO, M. SEINO, G.I. BELL, and S. SEINO, Molec. Endo. 6 2136-2142 (1992). K.H. JACOBS, K. AKTORIES, and G SCHULTZ, Nature 303 177-178 (1983)
90
13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.
Somatostatin Receptor Expressionin Tumors
Vol 53, No 1, 1993
H.-L. WANG, T. REISINE, and M. DICHTER, NeuroscL 8 335-342 (1990). C. LIEBOW, C. RE1LLY, M. SERRANO and A.V. SCHALLY, Proc. Natl. Acad Scl USA 86 2003-2007 (1989). Y. YAMADA, S.R POST, K. WANG, H.S. TAGER, G.I. BELL and S. SEINO, Proc Natl. Acad. Sci. 89 251-255 (1992). K. YASUDA, S. RENS-DOMIANO, C. D. BREDER, S. F. LAW, C. B. SAPER, T. REISINE and G. I. BELL, J. Biol. Chem., 267 20422-20428 (1992). A.-M. O'CARROLL, S. J. LOLAIT, M. KONIG and L. C. MAHAN, Mol. Pharmacol., 42 939-946 (1992). J.F. BRUNO, Y. XU, J. SONG and M. BERELOWlTZ, Proc. Natl. Acad. ScL USA, 89 11151-11155 (1992). H.G. DOHLMAN, J. THORNER, M.G. CARON and R.J LEFKOWlTZ, Ann. Rev. Blochem. 60 653-688 (1991). M. VANETFI, M. KOUBA, X. WANG, G. VOGT and V. HOLLT, FEBS 311 290-294 (1992). J.E. TAYLOR, P.A. EDEN and T. REISINE, In Preparataon. R.K. SAIKI, Science 239 487-491 (1988). J.E. TAYLOR, Proc. Winter Neuropeplade Conf., Brockenndge, CO (1993). J.E. TAYLOR, S.R. KEYES, and A.E. BOGDEN, Growth Factors, Peptldes and Receptors '92 XIIth George Washington Unlvers~ty Spring Symposium Abst. # 141, Washington, D.C (1992)