(CANCER RESEARCH 49, 2755-2760, May 15, 1989] Corticotropin-releasing Factor and Vasopressin Production in the Rat Pituitary Tumor 7315a: Biochemical and Immunohistochemical Studies1

MaríaInésMoranoand Fernando E. Estivariz2

Centro de Estudios Endocrinos, Facultad de Ciencias Medicas, Universidad Nacional de La Plata, Casilla de Correo 455, 1900 La Plata, Argentina

ABSTRACT (16). We have found it interesting to study the production and secretion of hypothalamic factors by this tumor, principally In order to investigate the production and secretion of hypothalamic those involved in the control of POMC peptides secretion. In factors by the prolactin and proopiomelanocortin (POMC)-derived, pep- tide-producing, transplanta blu rat pituitary tumor 7315a, we determined the present report, we determined the tumor content and in the concentrations of corticotropin-releasing factor (CRF)- and vasopres- vitro secretion of CRF and AVP, compared with the production sin (AVP)-like immunoreactivities (IR) in the tumor extracts [14.0 ±1.6 of rat prolactin and POMC peptides in the 7315a tumor. In (SE) and 4.2 ±0.9 pmol/g, respectively] and incubation media (0.26 ± addition, we have investigated the Chromatographie character 0.01 and 0.07 ±0.01pmol/107 cells/h, respectively). Total peptide content istics of the CRF- and AVP-like immunoreactivities, and fi correlated well with tumor weight. Moreover, there is a very good nally, we have explored the possibility of distinct populations correlation between the CRF and AVP IR, but not as good between CRF of tumor cells producing the different hormones. or AVP IR and POMC peptide IR tumor contents. Tumor extracts were chromatographed on Sephadex G-75 and compared with chromatograms of stalk median eminence (SME) extracts from normal Buffalo rats. CRF MATERIALS AND METHODS IR in tumor chromatograms gave an unusual pattern of peaks. About 31% of the total CRF IR was eluted in the high molecular weight region. Materials. All reagents used were of analytical grade. Collagenase The major portion of CRF IR was located in a wide region of lower Type IV, deoxyribonuclease I type DM-100, BSA (RIA grade), molec molecular weight. The AVP radioimmunoassay revealed a similar pattern ular weight markers for gel filtration chromatography, PMSF, 3,3'- of peaks in tumor and SME chromatograms. A propressophysin-like diaminobenzidine, Polypep, and dextran (M, 70,000) were provided by peak and a smaller peak coeluting with synthetic AVP were detected. Sigma, St. Louis, MO. Sephadex G-75 Superfine and chromatography Immunohistochemical staining of consecutive sections of the tumor indi columns were from Pharmacia, Uppsala, Sweden. PAP was from Dako cated that AVP and CRF are often found in the same cell, but the CRF Corporation, Santa Barbara, CA. Norit-A-activated charcoal and and AVP-producing cells are clearly distinct from the POMC peptide- polyethylene glycol (M, 6,000) were purchased from Pablo Zubizarreta producing cells. Ward, Buenos Aires, Argentina. Vycor glass was from Corning Glass Works, New York, NY. Penicillin and streptomycin were supplied by Lepetit Laboratories, Buenos Aires, Argentina. Eagle I lei.a medium INTRODUCTION was from Difco, Detroit, MI. Radioiodine (125I isotope) was from The presence of POMC3-derived peptides has been demon Amersham International Pic., Bucks. Radioiodinated LH-RH was from New England Nuclear, Boston, MA. Synthetic human ACTH (1-39) strated in human pituitary and ectopie ACTH-secreting tumors which may give rise to Cushing's syndrome (1-5). Moreover, [hACTH (1-39)], ACTH (1-24), ACTH (17-39), and a-MSH were supplied by Drs. P. A. Desaulles and W. Rittel, Ciba-Geigy, Basle, since the first demonstration of CRF-like activity in two types Switzerland. Synthetic JV-terminal POMC (1-48) [N-POMC (1-48)] of ectopie ACTH-producing tumors (pancreatic tumor and lung was a gift from Dr. J. Ramachandran, Genentech Inc., San Francisco, oat cell carcinoma) by Upton and Amatruda (6), considerable CA. Human N-POMC (1-76) [hN-POMC (1-76)] and h/3-LPH were evidence has been accumulating on ectopie production of both purified at St. Bartholomew's Hospital, London and kindly donated by ACTH and CRF in various human tumors (7-11). In the rat, Dr. P. J. Lowry, Department of Physiology and Biochemistry, Univer there is some information about CRF-like activity in the ACTH, sity of Reading, England. 73-MSH was a gift from Dr. P. Owens, growth hormone and prolactin-producing pituitary tumor MtT/ University of Newcastle, New South Wales, Australia, and MT was donated by Dr. M. Manning, Medical College of Ohio, OH. Human ß- F4 (12). On the other hand, it has been suggested that tumors endorphin, (Phe2,Nle4) ACTH (1-24), rat and ovine CRF (rCRF and can produce factors which resemble other hypothalamic hor oCRF), Tyr°-rCRF(rCRF with a tyrosine in position 0), sauvagine, mones and affect pituitary hormone secretion (13-15). (Arg')-AVP, (Arg')-AVT, LH-RH, and human angiotensins II and III The 7315a tumor is a Irimethyl-aniline induced trunspkmta (hA II and hA III) were purchased from Peninsula Laboratories, Bel- ble rat pituitary tumor that secretes prolactin and POMC- mont, CA. OT (Syntocinon) was from Sandoz Laboratories, Buenos derived peptides. In this respect, our previous studies have Aires, Argentina. Desacetyl a-MSH and C-terminal a-MSH antiserum analyzed the entire spectrum of POMC peptides produced by were kindly provided by Dr. A. Eberle, Institute of Molecular Biology the 7315a tumor, as a first step towards explaining the effects and Biophysics, Zurich, Switzerland. Rabbit anti-0-endorphin E-286, exerted by the tumor on the hypothalamo-pituitary-adrenal axis anti-rCRF 11 and anti-A II E-44 sera were donated by Dr. E. Spinedi, CRRIE, La Plata, Argentina. The anti-AVP/AVT sera AVPR and AVP4 Received 8/12/88; revised 1/18/89; accepted 1/23/89. were a gift from Dr. S. Blähser,Justus Liebig Universität, Giesen, The costs of publication of this article were defrayed in part by the payment Germany, and rabbit anti-oCRF serum was donated by Dr. G. Gillies, of page charges. This article must therefore be hereby marked advertisement in St. Bartholomew's Hospital, London, England. The anti-N-POMC (1- accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' Work supported by Grant PID-3-901102 from the National Research Council 76) sera C4 and D4, anti-ACTH serum AC 1-7 and goat anti-rabbit 7- of Argentina (CONICET). globulin were produced in our laboratory. Rat prolactin RIA materials 2 Present address: Neuroendocrinology Unit, Department of Neurobiology, were supplied by the NIADDK Rat Pituitary Hormone Distribution 1MBICE, Casilla de Correo 403, 1900 La Plata, Argentina. To whom requests for reprints should be addressed. Program. 3The abbreviations used are: POMC, proopiomelanocortin; A, angiotensin; Biological Specimens. The 7315a pituitary tumor was kindly provided ACTH, adrenocorticotropin; AVP, vasopressin; AVT, vasotocin; BSA, bovine by Dr. R. M. MacLeod from the Department of Medicine, University serum albumin; CRF, corticotropin-releasing factor; IR, immunoreactivity; LH- of Virginia, School of Medicine, Charlottesville, VA, in 1981 and RH, luteinizing hormone-releasing hormone; /5-LPH, /3-lipotropin; MSH. mela- notropin; MT, mesotocin; OT, oxitocin; PAP, peroxidase-antiperoxidase com maintained in our laboratory since then by successive transfer in Buffalo plex; PMSF, phenylmethylsulphonyl fluoride; N-POMC, N-terminal POMC; rats originally obtained from Simonsen Labs, Gilroy, CA, and subse RIA, radioimmunoassay; SME, stalk median eminence. quently bred in our laboratory. Transplants were performed as described 2755

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research. CRF AND AVP IN 7315a TUMOR previously (16) by s.c. inoculation of tumor pieces or dispersed tumor Ojeda, and McCann (22), using an antiserum provided by Dr. A. cells (approximately IO7) suspended in Eagle's HeLa medium, in the Arimura (Tulane University). The detection limit of the assay was 4 suprascapular region of inbred adult Buffalo rats. Animals were kept pg/tube. The rat prolactin RIA was performed using the NIADDK RIA at 21 ±2°Cin a 12-h light/12-h darkness cycle with water and food kit. The intra- and interassay coefficients were less than 8 and 12%, available ad libitum. Since the discovery of CRF and AVP in our tumor respectively, in all the RIA systems employed. line, we have tested and found these peptides in another sample of the Immunohistochemical Methods. One portion of each tumor was 7315a tumor from Dr. MacLeod, recently provided by him. routinely fixed in Elftman's fluid (23), embedded in paraffin, and Peptide Extraction and Chromatography. Tumor tissue was removed sectioned at 3-^m thickness. The presence of POMC peptides, CRF from donor rats between Days 15 and 50 after inoculation. Segments and AVP were immunohistochemically demonstrated as published of nonnecrotic tumor tissue were rapidly homogenized in either 200 elsewhere (24). Adjacent sections were immunostained by the peroxi- mM HC1 or acetic acid (1 M) with PMSF (l IHM)(l g tumor/2.5 ml dase-antiperoxidase (PAP) unlabeled antibody enzyme technique, using acid mixture) with or without brief boiling to inactivate proteolytic as the oxygen acceptor 3,3'-diaminobenzidine. For comparison, pitui- enzymes. Tumor homogenates were immediately centrifuged at 12,000 taries and SME from normal adult Buffalo rats were fixed and treated x g at 0°Cfor60 min. SME (consisting of the hypothalamic-hypophy- in the same way. As negative controls, the first antisera were replaced sial stalk, with a small portion of basal hypothalamic tissue attached to by rabbit normal serum or absorbed against their corresponding anti it) was dissected from normal adult Buffalo rats according to Linton gens. The antibodies employed were identical to those described in "Radioimmunoassays." and Lowry (17), and processed in conditions similar to those for the tumor tissue. Supernatants from tumor and SME extracts were frozen at -20°Cuntil assay for hormone content or immediately applied (5- 10 ml) to a Sephadex G-75 column as described below. RESULTS Gel filtration was performed on a Sephadex G-75 Superfine column All results described in this paper were obtained with tumors (3 x 80 cm) using 265 mM formic acid with or without 70 mM 2- from 40-56 generation transfers from the original tumor in our mercaptoethanol, and an upward flow of 6 ml/h according to Morano and Estivariz (16). Fractions of 2 ml were collected and frozen at —¿20°Claboratory. Implantation of the 7315a tumor resulted in pal pable tumor tissue after 15 days. Thereafter, a linear increase unless immediately processed for RIA. The column was calibrated with molecular weight markers and the following peptides: BSA, rat prolac- in tumor size was observed until 50 days. tin, rCRF, and AVP. Exclusion and bed volumes were determined by The peptide extraction procedures were selected based on the UV absorption monitoring to the column eluted for each run. Chro maximum recovery of ACTH and rCRF when these peptides matography was performed three times with the same tumor extract, were added during the tumor homogenization. Two different variations in retention times being less than 3% from run to run. variants for the preparation of dispersed tumor cells were Preparation and Incubation of Dispersed Tumor Cells. Approximately employed (mechanical and collagenase dispersion). In both 5 g tumor tissue were rapidly excised from surrounding normal tissue, methods, approximately the same number of viable tumor cells and minced finely in 50 ml of chilled Krebs-Ringer-bicarbonate-glucose and similar basal secretion rate were obtained. RIA dilution buffer containing 5 g BSA per liter and antibiotics supplemented with 10% Eagle's HeLa medium. The suspension was transferred to a plastic curves obtained with tumor extracts and incubation media were parallel to the respective standards with all RIA systems em tube and this procedure was repeated once more with remaining frag ployed (result not shown) thus making it possible to quantify ments of tissue. The resulting suspension was then filtered through a 100-Mmmesh gauze and centrifuged at 100 x g for 20 min at 4°C.The the amounts of the different peptides produced by the 7315a cell pellet was washed twice and suspended gently in 10 ml incubation tumor. medium. An aliquot of cell suspension was mixed with an equal volume Table 1 summarizes the tumor content and in vitro secretion of Trypan Blue solution in saline (4 g per liter) and loaded in a by tumor cell of all peptides investigated. In addition to rat hemocytometer. Usually, between 4 and 8 x IO8cells/5 g tumor were prolactin and POMC-derived peptides (16, 25), the tumor obtained and the cell viability was about 95%. The cell suspension was produces the neuropeptides CRF and AVP but not AII or LH- adjusted to 10 cells/ml with incubation medium and preincubated for RH. The concentrations of CRF and AVP IR in the tumor 60 min in a shaking incubator at 37°Cunder 95% air-5% CO2 atmos extracts and incubation media are smaller than those of rat phere. After preincubation, the cells were centrifuged, the medium prolactin and POMC peptides (except o-MSH). For compari discarded and the same volume of medium was added to the cell pellet son, the CRF and AVP contents in the SME were 2.1 ±0.3 which was then gently resuspended. The cell suspension was rapidly and 19.8 ±3.5 pmol/SME (mean ±SE, N = 7), respectively. distributed in plastic tubes (1 ml per tube), and incubated for 2 h under identical conditions as for preincubation. After this, the tubes were Moreover, we found CRF and AVP IR in a new sample of the centrifuged at 100 x g for 20 min. The supernatants were extracted, 7315a tumor obtained from Dr. R. M. MacLeod. However, the frozen immediately, and stored at -20°Cfor hormone measurements. concentrations of CRF and AVP IR in the original tumor line In other experiments, collagenase-dispersed tumor cells were ob were smaller by a factor of at least 10 (result not shown). tained and incubated as described previously (18). The CRF and AVP tumor contents as measured by RIA from Radioimmunoassays. RIAs for ACTH, N-POMC, /3-LPH/endorphin, and a-MSH were performed as described previously (16, 19). A CRF Table 1 Tumor content and in vitro secretion of rat prolactin, RIA was employed according to Spinedi and Rodriguez (20); it had a proopiomelanocortin (POMC)-derived peptides, and neuropeptides detection limit of 2 pg/tube. [125I]Tyr°-rCRFwasused as tracer. The in the 7315a tumor Tumor extracts were obtained with 200 mM HC1 (l g tumor/2.5 ml); tumor anti-CRF serum 11 had 0.1% cross-reactivity with oCRF and no cells were mechanically dispersed from the same tumors and incubated for l h as crossreactivity with sauvagine, AVP, and angiotensins; the anti-CRF described in "Materials and Methods." Values are means ±SE; N = 3 tumors. F8B3 cross-reacted fully with oCRF. The AVP RIA had a detection content secretion limit of 0.4 pg/tube. Synthetic AVP was used for both tracer and (pmol/gtumor)106.0 (pmol/107cells/h)8.43 standards. Two antisera (AVPR and AVP4) were used; they both reacted RatprolactinACTH/V-terminal ±9.897.8 ±0.431.06 fully with AVT and showed less than 0.01% cross-reactivity with MT, ±6.195.2 ±0.041.26 OT, rCRF, and POMC peptides. The A II RIA had a detection limit POMC/3-LPH/endorphina-MSHrCRFAVPLH-RHAHTumor±5.890.5 ±0.030.85 of 2 pg/tube. The A II was labeled using an identical procedure to the ±9.82.8 ±0.090.06 ±0.214.0 ±0.010.26 one previously described (21). The anti-A II serum E-44 showed ap ±1.64.2 ±0.010.07 proximately 40% cross-reactivity with A III and 1% with A I. The LH- ±0.9<0.02<0.01Basal ±0.01 RH RIA was generously performed by Dr. S. Chiocchio (Instituto de Neurobiologia, Buenos Aires, Argentina) according to Negro-Vilar, 2756

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research. CRF AND AVP IN 7315a TUMOR rats through 10 transfer generations of the 7315a correlated BSA rPRL rCRF AVP 125I well with tumor growth between Days 15 to 50 after inoculation (r > 0.90; P < 0.01). Moreover, there is a very good correlation between the CRF and AVP IR contents of the tumors, but not II as good between CRF or AVP IR and POMC peptides IR (e.g., ACTH IR) tumor contents (/>< 0.01; Fig. 1). Fig. 2a shows the CRF RIA elution profiles of Sephadex G- . 3 75 chromatograms of SME extracts from normal Buffalo rats, obtained with two different antisera. With antiserum 11, the Chromatographie profile of CRF IR presented similar charac .2 teristics to those previously obtained using Wistar or Sprague- Dawley rats (17, 26); namely, a major peak eluting at the .1 position expected for synthetic rCRF (94% of the total CRF IR) and a minor peak of larger molecular weight, which prob ably represents the precursor form of rCRF (27). Antiserum 0.0 0.2 F8B3 detected a single peak coinciding with synthetic rCRF. A 0.4 0.6 0.8 1.0 K„ pool of tumors and five individual tumors from different trans fer generations were analyzed chromatographically. All gave similar Chromatographie profiles. Such profiles obtained with extracts of two individual tumors (Nos. 59 and 169) are shown in Fig. 2, b and c. Tumor chromatograms from either acetic .2 oc acid (1 M) extracts containing PMSF (1 mM) with brief boiling u (Fig. 2Ä)or HC1 (200 mM) extracts eluted with 2-mercaptoeth- o anol containing solvent (Fig. 2c), gave essentially the same e pattern of peaks with respect to CRF IR. The CRF RIA profiles o. obtained with both antisera displayed marked heterogeneity at an elution position corresponding to distribution coefficients 0.0 0.2 0.4 0.6 1.0 K, (KD) between 0.50 and 0.75 (Fig. 2, b and c). Antiserum 11 c) revealed an additional peak eluting at the position expected for the precursor which accounted for 31 ±2% of the total CRF IR of the tumor Chromatographie profiles. Finally, the addition

0.99 1.2

0.9

0.6 1.0 Kr

Fig. 2. CRF immunoreactive profiles of Sephadex G-75 chromatograms of 0.3 extracts from: a, 12 Buffalo rat stalk median eminence; b, 4 g of tumor No. 59; and c. 4 g of tumor No. 169. Solid lines, profiles obtained with no. 11 antiserum; dashed lines, those obtained using F8B3 antiserum. The extracts a and b were 0.08 0.24 0.40 homogenized in acetic acid (1 M) containing PMSF (1 mM) with brief boiling, and the extract c was obtained with HCI (200 mM). The column was eluted with nmol AVP/ TUMOR 265 mM formic acid (a and b) or 265 mM formic acid with 2-mercaptoethanol (70 mM) (c). Arrows, eluting positions of standard peptides and proteins. Points, individual(A1,,).CRF fractions. immunoreactivity Elution is positions expressed are as indicated pumi synthetic as distribution rCRF per coefficientfraction (2 0.77 ml). rPRL, rat prolactin.

24 of [125I]Tyr°-rCRFto tumor tissue at the time of homogen i/a- 18 tion did not yield any labeled product with different elution characteristics of rCRF (not shown). 12 Fig. 3 shows the Chromatographie profiles of AVP IR of rat SME extracts (Fig. 3a) and the tumor extract described in Fig. 2b (Fig. 3¿>).InSME extracts, the chromatography confirmed previous reports on other rat strains (28). A peak with a molecular weight of about 20,000 could be seen which probably 0.6 1.2 1.8 2.4 represents propressophysin (29), but the bulk of the AVP IR nmol rCRF/ TUMOR (94% of total) coeluted with synthetic AVP. In agreement with Fig. 1. Correlation between CRF immunoreactivity (IR) and AVP IR (upper), and ACTH IR and CRF IR (lower) contents of 7315a tumors. Points, individual rat SME chromatograms, tumor chromatograms displayed a tumors extirpated at different times after inoculation (see "Materials and Meth similar pattern of peaks with all the extraction procedures ods"). Tumor tissue was extracted with 200 mM HCI (l g tumor/2.5 ml), r, employed. With antiserum AVP4, the AVP RIA indicated the correlation coefficient. Tumor contents are expressed in nmol synthetic rCRF, synthetic AVP, and synthetic hACTH (l-39)/whole tumor, respectively. In both presence of only one peak which coeluted with synthetic AVP. cases correlation coefficients were significantly different from zero (P < 0.01 ). Moreover, antiserum AVPR showed higher levels of AVP IR in 2757

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research. CRF AND AVP IN 7315a TUMOR

BSA rPRL rCRF AVP 125I review, see réf.31). In addition, we have demonstrated these peptides in the original 7315a tumor line from Dr. MacLeod. However, we have not been able to detect two other neuropep- IH tides in the tumor tissue (i.e., LH-RH and A II). a) Upton and Amatruda (6) first demonstrated CRF-like biolog 15 ical activity in human tumor extracts. Later, other cases of human CRF-producing tumors were established (32, 33). More over, the presence of ACTH and CRF IR has been described in human lung cancer and pheochromocytoma (9), ovarian tumors (34), bronchial carcinoids (10-11), and paraganglioma (35). On 10 the other hand, AVP IR has been extensively reported from patients with oat cell carcinoma of the lung (15). With respect to rat transplantable tumors, the study of Suda et al. (12) only examined the existence of CRF-like bioactivity in extracts of the pituitary tumor MtT/F4. The present communication is, z then, the first demonstration of CRF and AVP IR together in o a rat hormone-producing tumor. Absolute amounts of CRF and AVP detected in tumor ho- 0.0 0.2 0.4 0.6 0.8 1.0 K.. mogenates were, respectively, about seven and 23 times less than those of rat prolactin and POMC peptides. In addition, dispersed cell from the 7315a tumor secreted the same peptides b) found in the tumor extracts. On the other hand, there was very good correlation between the CRF IR and AVP IR contents in whole tumors obtained from 10 transfer generations. However, the correlation between CRF IR and ACTH IR contents from the same tumors was not as good. The apparent lack of corre lation of CRF and AVP versus POMC peptides contents in the tumor can be explained by the observation of immunohisto- chemical heterogeneity in the 7315a tumor. The immunohis- tochemical staining for N-POMC, CRF, and AVP revealed at least two distinct cell populations, one positive for POMC peptides and the other positive for CRF and/or AVP. We have 0.0 0.2 0.4 0.6 0.8 1.0 Kr previously described that the staining patterns of serial sections with anti-rat prolactin and anti-POMC peptides were clearly Fig. 3. AVP ¡mmunoreactiveprofiles of Sephadex G-75 Superfine chromato- grams of extracts from: a, 12 Buffalo rat stalk median eminence; A,4 g of tumor distinct (36). Although the production of more than one hor No. 59. Solid lines, profiles obtained with AVPR antiserum; dashed lines, those mone or neuropeptide is a common finding in pituitary and obtained using AVP., antiserum. The column was eluted with 265 mM formic nonpituitary tumors (7-12, 34, 35, 37, 39), the mechanism by acid. Arrows, eluting positions of standard peptides and proteins; points, individual fractions. Elution positions are indicated as distribution coefficients (A'«).AVP which the cells producing either peptide differentiate during immunoreactivity is expressed as pmol synthetic AVP per fraction (2 ml). rPRL, tumor evolution remains unresolved. Moreover, the copackage rat prolactin. of CRF and AVP within the same neurosecretory vesicles in terminals of a population of hypothalamic neurons of normal rat SME than did AVP4. Recoveries of CRF and AVP IR after rats (38) indicates that a similar coexpression of both genes chromatography fluctuated between 85 and 95%. occurs in normal tissue. On the other hand, the presence of Immunohistochemical staining (PAP method) revealed the ACTH and CRF in two distinct populations of tumor cells was presence of immunoreactive N-POMC, rCRF, and AVP in the previously identified in a human bronchial carcinoid (11). cytoplasm of many tumor cells (Fig. 4). In addition ACTH, ß- In order to investigate the biochemical characteristics of CRF LPH/endorphin, and a-MSH were also located in tumor cells and AVP IR, we have found it interesting to observe whether by immunohistochemistry (not shown). The study of consecu any differential processing of the precursor forms of CRF and tive sections of the tumors indicated that AVP and CRF are AVP occurred in the 7315a tumor compared with normal rat often found in the same cell. Serial section showed, in addition, SME or other tumors (15, 17, 26, 28, 34, 35). Gel filtration the CRF and AVP-secreting cells are clearly distinct from the chromatography of SME homogenates of Buffalo rats did not POMC peptide-secreting cell population (Fig. 4). show any differences with regard to CRF or AVP IR from those of other rat strains previously investigated (17, 26, 28). The Chromatographie profiles of tumor CRF IR gave an unusual DISCUSSION pattern of peaks. About 31% of the total CRF IR was eluted in The hormone-producing tumor 7315a was isolated from the the high molecular weight region, at the position expected for pituitary gland of a female Buffalo rat (25). Since then it has the CRF precursor (27). Although this peak may be caused by been reported as a transplantable rat pituitary tumor that se aggregated or protein-bound CRF (26), the dissociating condi cretes prolactin and a small amount of ACTH (30). In a recent tions employed to elute the column (265 mM formic acid) make study (16) we have assessed the production of several peptides such an association unlikely. Moreover, this IR peak was not derived from different region of POMC in the 7315a tumor. detected using the anti-oCRF serum F8B3, indicating that this Our present investigation indicates that our 7315a tumor line antibody does not recognize the pro-CRF molecule. In contrast also produces the neurohormones CRF and AVP, both involved with the homogeneous peak of rCRF found in SME extracts, in the regulation of pituitary POMC peptide secretion (for in tumor chromatograms the major portion of CRF IR was 2758

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research. CRF AND AVP IN 7315a TUMOR •¿;*•

•¿

Fig. 4. Immunoperoxidase staining of se rial histological sections of the 731 Sa tumor with anti-AVP (a), and TV-terminal proopi- omelanocortin (¿>),andanti-CRF (c) sera. Ar rows, same cell (x 190).

100 b eluted in a large region between /TD0.50 and 0.75 with all the and all those mentioned for their gifts of antisera, peptides, and drugs. extraction and elution procedures employed. In the case of Thanks are extended to Jorge Bevilacquea, Monica Carino, Frida AVP IR, two peaks were detected with the antiserum AVPR at Winschu-Bustos, and RubénParedes for helpful technical assistance. elution positions compatible with propressophysin (29) and true AVP, in both SME and 7315a tumor extracts. This is REFERENCES consistent with the data from most human oat cell carcinomas 1. Ueda, M., Takechi, T., Abe, K., Miyakawa, S., Oli 11:11111,S.,and Yanaihara, of the lung with ectopie AVP production which contain pro N. 0-Melanocyte-stimulating hormone immunoreactivity in human pituitar- pressophysin (15). The observation that antiserum AVP4 has ies and ectopie adrenocorticotropin-producing tumors. J. Clin. Endocrinol. no cross-reaction with propressophysin and it cross-reacts with Metab., 50: 550-556, 1980. AVP to a lesser extent than antiserum AVPR indicates the two 2. Shibasaki, T., Masui, H., Sato, G., Ling, N., and Guillemin, R. Secretion pattern of pro-opiomelanocortin-derived peptides by a pituitary adenoma antisera recognize different portions and/or structural confor from a patient with Cushing's disease. J. Clin. Endocrinol. Metab., 52: 350- mation of the AVP sequence. 353, 1981. 3. Suda, T., Tozawa, F., Yamaguchi, H., Shibasaki, T., Demura, H., and The production of the 7315a tumor of CRF and AVP, which Sin/mm'. K. Multiple forms of immunoreactive ß-endorphinare present in are secreted in vitro and probably circulate at physiologically an ectopie adrenocorticotropin-producing tumor but not in normal pituitary significant levels in tumor-bearing rats, raises the interesting or pituitary adenomas. J. Clin. Endocrinol. Metab., 54: 167-171, 1982. question of whether these peptides may act on the pituitary and 4. De Keyzer, I., Bertagna, X., Lenne, F., Girard, F., Luton, J. P., and Kahn, A. Altered proopiomelanocortin gene expression in adrenocorticotropin- cause increased POMC peptides release. Recent investigations producing nonpituitary tumors. Comparative studies with corticotropic ade conducted in our laboratory support this assertion (39). In this nomas and normal pituitaries. J. Clin. Invest., 76: 1892-1898, 1985. respect, Belsky et al. (33) previously demonstrated excessive 5. Hale, A. C., Besser, G. M., and Rees, L. H. Characterization of proopiome- lanocortin-derived peptides in pituitary and ectopie adrenocorticotrophin- ACTH production from the anterior pituitary due to CRF secreting tumours. J. Endocrinol., 108: 49-56, 1986. secretion by metastatic medullary carcinoma of the thyroid. In 6. Upton, G. V., and Amatruda, T. T. Evidence for the presence of tumor addition, it is possible that the CRF and AVP produced by the peptides with corticotropin-releasing-factor-like activity in the ectopie ACTH syndrome. N. Engl. J. Med., 285:419-424, 1971. tumor may stimulate POMC peptides secretion of other tumor 7. Birkenhäger,J. C., Upton, G. V., Seldenrath, H. J., Krieger, D. F., and cells in a paracrine manner. Alternatively, since in normal Tashian, A. H. Medullary thyroid carcinoma: ectopie production of peptides development pituitary corticotrophs are derived from undiffer- with ACTH-like, corticotrophin releasing factor-like and prolactin produc entiated embryonic cells, probably induced by the same hy- tion-stimulating activities. Acta Endocrinol., 83: 280-292, 1976. 8. Hashimoto, K., Takahara, J., Ogawa, N., Yunoki, S., Ofuji, T., Arata, A., pothalamic principles that regulate their function in postnatal Kanda. S., and Torada, K. Adrenocorticotropin, 0-lipotropin, ß-endorphin, life, it is tempting to speculate that tumor CRF and AVP may and corticotropin-releasing factor-like activity in an adrenocorticotropin- contribute to differentiate the tumor POMC peptide-secreting producing nephroblastoma. J. Clin. Endocrinol. Metab., 50:461-465, 1980. 9. Suda, T., Tomori, N., Tozawa, F., Demura, H., Sin/urne, K., Mouri, T., cells or at least stimulate their proliferation. This possibility is Miura, Y., and Sanao. N. Immunoreactive corticotropin and corticotropin- supported by the observation of a GH3 line which produced releasing factor in human hypothalamus, adrenal, lung cancer, and pheo- chromocytoma. J. Clin. Endocrinol. Metab., 58: 919-924, 1984. only prolactin; when Pitressin (posterior pituitary extract with 10. Zarate, A., Kovacs, K., Flores, M., Moran, C., and Félix,I.ACTH and CRF- AVP) was added to this GH3 cell culture, the percentage of producing bronchial carcinoid associated with Cushing's syndrome. Clin. prolactin-positive cells remained constant, but the percentage Endocrinol., 24: 523-529, 1986. 11. Schteingart, D. E., Lloyd, R. V., Akil, H., Chandler, W. F., Ibarra-Perez, G., of ACTH-positive cells increased starting 10 h after stimulation Rosen, S. G., and Ogletree, R. Cushing's syndrome secondary to ectopie (40). corticotropin-releasing hormone-adrenocorticotropin secretion. J. Clin. En In conclusion, this is the first demonstration of CRF and docrinol. Metab., 63: 770-775, 1986. 12. Suda, T., Demura, H., Demura, R., Wakabayashi, I., Nomura, K., Odagiri, AVP IR, immunologically and chromatographically similar to E., and Shizume, K. Corticotropin-releasing factor-like activity in ACTH those of SME, in a rat pituitary tumor. Further studies are in producing tumors. J. Clin. Endocrinol. Metab., 44:440-445, 1977. progress to investigate the effects exerted by these tumor neu- 13. Shalet, S. M., Beardwell, C. G., Macfarlane, I. A., Elison, M. L., Norman, rohormones on the hypothalamopituitary-adrenal axis of the C. M., Rees, L. H., and Hughes, M. Acromegaly due to production of a growth hormone releasing factor by a bronchial carcinoid tumour. Clin. 7315a tumor-bearing rats. Endocrinol., 10:61-67, 1979. 14. Guillemin, R., Brazeau, P., Bohlen, P., Esch, F., Ling, N., and Wehrenberg, W. B. Growth hormone releasing factor from a human pancreatic tumor that ACKNOWLEDGMENTS causes acromegaly. Science (Wash. DC), 218: 585-587, 1984. 15. Yamaji, T., Ishibashi, M., and Hori, T. Propressophysin in human blood: a The authors wish to thank Dr. F. C. Iturriza for the photographic possible marker of ectopie vasopressin production. J. Clin. Endocrinol. work, Dr. R. M. MacLeod who generously provided the 7315a tumor. Metab., 59:505-512, 1984. 2759

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research. CRF AND AVP IN 7315a TUMOR

16. Morano, M. I., and Estivariz, F. E. Chromatographie characteristics of pro- chromatography of hypothalamic corticotrophin-releasing factors. Neuroen- opiomelanocortin-derived peptides from the rat transplantable tumour docrinology, 38: 17-24, 1984. 7315a. J. Endocrinol., 112:417-425, 1987. 29. Schmale, H., Heinsohn, S., and Richter, D. Structural organization of the 17. Union, E. A. and Lowry, P. J. Radioimmunoassay and Chromatographie rat gene for the arginine vasopressin-neurophysin precursor. EMBO J., 2: characterisation of CRF-41-like immunoreactivity in hyputbaiami from sev 763-767, 1983. eral species. Neuropeptides, 3: 45-52, 1982. 30. Judd, A. M., Koike, K., Schettini, G., Login, I. S., Hewlett, E. L., Yasumoto, 18. Castro, M. G., Estivariz, F. E., and Iturriza, F. C. The regulation of the T., and MacLeod, R. M. Dopamine decreases 7315a tumor cell prolactin cortico-melanotrophic cell activity in aves. I. Evaluation and selection of in release induced by calcium mobilization. Endocrinology, 117: 1215-1221, vitro systems for testing A( Til-releasing substances in the duck pituitary. 1985. 31. Antoni, F. A. Hypothalamic control of adrenocorticotropin secretion: ad Comp. Biochem. Physiol., 79A: 169-173, 1984. vances since the discovery of 41-residue corticotropin-releasing factor. En 19. Estivariz, F. E., Morano, M. I., Carino, M., Jackson, S., and Lowry, P. J. docrine Rev., 7:351-378, 1986. Adrenal regeneration in the rat is mediated by mitogenic .V-terminal pro- 32. Kirkland, S. C., and Ellison, M. L. Secretion of corticotropin releasing factor- opiomelanocortin peptides generated by changes in precursor processing in like activity by a human bronchial carcinoid cell line. J. Endocrinol., 103: the anterior pituitary. J. Endocrino!., ¡16:206-217, 1988. 85-90, 1984. 20. Spinedi, E., and Rodriguez, G. Angiotensin II and adrenocorticotropin 33. Belsky, J. L., Cuello, B., Swanson, L. W., Simmons, D. M., Jarrett, R. M., release: mediation by endogenous corticotropin-releasing factor. Endocrinol and Braza, F. Cushing's syndrome due to ectopie production of corticotropin- ogy, 119:1397-1402,1986. releasing factor. J. Clin. Endocrinol. Metab., 60:496-500, 1985. 21. Mukherjee, A., Kulkami, P., McCann, S. M., and Negro-Vilar, A. Evidence 34. Suda, T., Tomori, N., Yajima, F., Odagiri, E., Demura, H., and Shizume, K. Characterization of immunoreactive corticotropin and corticotropin-releas for the presence of angiotensin II receptors in rat anterior pituitary mem branes. Endocrinology, 110: 665-667,1982. ing factor in human adrenal and ovarian tumours. Acta Endocrino!., Ill: 546-552,1986. 22. Negro-Vilar, A., Ojeda, S. R., and McCann, S. M. Catecholaminergic mod ulation of luteinizing hormone-releasing hormone release by median emi 35. Hashimoto, K., Suemaru, S., 1Indori, T., Sugawara, M., Ota, Z., Takata, S., Hamaya, K., Doi, K., and Chrétien,M. Multiple endocrine neoplasia with nence in vitro. Endocrinology, ¡04:1749-1757, 1979. Cushing's syndrome due to paraganglioma producing corticotropin-releasing 23. Elftman, H. A chrome-alum fixative for the pituitary. Stain Technol., 32: factor and adrenocorticotropin. Acta Endocrinol., 113:189-195, 1986. 25-29, 1957. 36. Herreros, G. R., Morano, M. I., and Iturriza, F. C. Hallazgo de metástasis 24. Iturriza, F. C., Estivariz, F. E., and Levitin, H. P. Coexistence of a-melano- del tumor corticotropo 7315a. Estudios histológicos e inmunohistoquímicos cyte-stimulating hormone and adrenocorticotrophin in all cells containing de las lesiones primarias y secundarias. In: A. Arguelles, I. Molocznik, and either of the two hormones in the duck pituitary. Gen. Comp. Endocrino!., A. Guitelman (eds.), Prolactina, Aspectos Fisiológicos, Bioquímicosy Clín «.•110-115,1980. icos, pp. 83-95. Buenos Aires: Academia Nacional de Ciencias, 1985. 25. Bates, R. W., Garrison, M. M., and Morris, H. P. Comparison of two 37. McComb, D. J., Kovacs, K., Beri, J., and Zak, F. Pituitary adenomas in old different transplantable mammotropic pituitary tumors. Hormone content Sprague-Dawley rats; a histologie, ultrastructural, and immunocytochemical and effect on host. Proc. Soc. Exp. Biol. Med., 123:67-70, 1966. study. J. Nati. Cancer Inst., 73:1143-1166, 1984. 38. Whitnall, M. H., and Gainer, H. Ultrastructural studies of peptide coexist 26. Yasuda, N., and Yasuda. Y. Studies on the site of origin and molecular form ence in corticotropin-releasing factor- and arginine-vasopressin-containing of the extrahypothalamic corticotrophin-releasing factor (CRF) in rats. Acta neurons. In: T. W. Moody (ed.), Neural and Endocrine Peptides and Recep Endocrinol., 709:145-152,1985. tors, pp. 159-175. Plenum Publishing Corporation, 1986. 27. Shibahara, S., Morimoto, Y., Furutani, Y., Notake, M., Takahashi, H., 39. M. I. Morano, and F. E. Estivariz. Influences of the pituitary tumor 7315a Shimi/u. S., Horikawa, S., and ¡Suma.S.Isolation and sequence-analysis of on the hypothalamic-pituitary-adrenal axis in the rat. 8th International the human corticotropin-releasing factor precursor gene. EMBO J., 2: 775- Congress of Endocrinology. Kyoto, Japan, 1988. 779, 1989. 40. Nakane, P. K., Sétáló,G.,and Mazurkiewicz, J. E. The origin of ACTH cells 28. Gillies, G. E., Puri, A., Linton, E. A., and Lowry, P. J. Comparative in rat anterior pituitary. Ann. NY Acad. Sci., 297: 201-204, 1977.

2760

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research. Corticotropin-releasing Factor and Vasopressin Production in the Rat Pituitary Tumor 7315a: Biochemical and Immunohistochemical Studies

María Inés Morano and Fernando E. Estivariz

Cancer Res 1989;49:2755-2760.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/49/10/2755

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/49/10/2755. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1989 American Association for Cancer Research.