8/8/2019 Identification and Cellular Localization of Growth Hormone

1/3

0021-972x/94/7902-0670$03.w/0Journal of Clinical Endocrinology and MetabolismCopyright 0 1994 by The Endocrine Society Vol. 79, No. 2Prrnted in U.S.A.

IDENTIFICATION AND CELLULAR LOCALIZATION OF GROWTH HORMONE RECEPTOR GENE EXPRESSIONIN THE HUMAN OVARYFADY I. SHARARA AND LYNNETTE K. NIEMANDevelopmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutesof Health, Bethesda, MDAddress for correspondence: Lynnette K. Nieman, M.D.Building 10, Room lON262NIH9000 Rockville PikeBethesda, MD 20892ABSTRACT While in vi vo and in vitro studies in rodents, pigs and women suggest that growth hormone (GH) canstimulate ovarian steroidogenesis, it is not known if this effe ct is mediated by a direct action on the ovary. Theabsence of GH receptor (GHR) messenger RNA would mitigate against a direct ovarian effect . We used the reversetranscriptase-polymerase chain reaction and in situ hybridization to examine whether the GHR mRNA was present inhomogenates of seven human ovaries or in tissue sections of ten ovaries. GHR gene expression was detected inPCR products after Southern blot hybridization using an oligoprobe directed to the intracellular domain sharing nohomology to the prolactin receptor. In situ hybridization using the same digoxigenin-labeled oligoprobe localized theGHR mRNA in the granulosa cells of dominant and antral follicles, corpus luteum, corpora albicans and theendothelium of blood vessels. GHR mRNA was not detected in preantral follicles, theta interna, theta externa,oocytes, or stroma. The presence of GHR mRNA in human granulosa cells and corpus luteum, taken together withprevious studies showing GH-induced stimulation of estradiol and progesterone secretion, suggest that GH may playa direct role in the development of the human follicle.

IntroductionAnimal and human studies suggest that GH mightpotentiate the eff ect of FSH on oocyte recruitment, andestradiol and progesterone production (1-6). While thisovarian effec t could be mediated through systemic orintraovarian insulin-like growth factor-l (IGF-I) production(l), a direct eff ect of GH cannot be excluded (25).Recently , Lobie demonstrated immunocytochemicallocalization of the GHR/binding protein in the rat oocyte,granulosa cell, theta interna, theta externa and corpusluteum (7). To our knowledge, studies of GHR geneexpression and cellular localization have not beenperformed in the human ovary. We used reversetranscriptase-polymerase chain reaction (RT-PCR) andin situ hybridization (ISH) to examine the expression andlocation of GHR mRNA in the human ovary.Materials and MethodsSubjects and tissue collection: Ovaries were obtainedduring surgery for non-ovarian disease. Seven werecollected prospectively from 4 women (age 35 - 43) withregular menses. Immediately after removal, a portion ofovary was frozen on dry ice and stored at -70 C untilRNA extraction. 10 ovaries from 9 other women (age 35- 52) were identified by a retrospective review ofpathology specimens and used for ISH.RNA extraction: RNA extraction was performed usingthe guanidium thiocyanate method modified by Chirgwin(8). Electrophoresis of total RNA on a 1% agaroseI2.2 Mformaldehyde gel showed intact 28s and 18s bands.Primers and p&es: The primers and probes for GHRand glyceraldehyde 3-phosphate dehydrogenase(GAPDH) were synthesized using a nucleic acidsynthesizer. Probes were 3-tail labeled withdigoxigenin-UTP according to the manufacturer s

instructions (Boehringer Mannheim Co, Indianapolis,IN).GHR primers-upstream: 5-GCC AAT GAC ATA CATGAG GGT ACC-3 (nucleotides 1201-l 225);downstream: 5-TGG GAC AGG CAT CTC AGA ACCTGG-3 (nucleotides 1755-1779); the 578 bp codingregion of GHR chosen for amplification corresponds toan intracellular portion of the GHR that has no homologywith the prolactin receptor. The GHR oligoprobe 5-GGACAT CCC TGC CTT ATT CTT TTG GCC CGG GGAAAG GAC CAC ACT A-3 represented a 46 bpsequence from the intracellular region flanked by theprimers (bp 1489-l 535). The corresponding sensesequence was used as a negative control for ISH. TheGAPDH primers flanked a 195 bp region. GAPDHprimers: upstream- 5CCA TGG AGA AGG CTG GGG-3 (nucleotides 388-405); downstream- 5-CAA AGTTGT CAT GGA TGA CC-3 (nucleotides 563-582).GAPDH probe: 5-CTA AGC AGT TGG TGG TGC A-3(nucleotides 531-549).RT-PCR: 1 ug of total RNA was reverse transcribed intocorresponding cDNA using a RNA RT-PCR kit (Perkin-Elmer-Cetus, Norwalk, -CT), in a reaction cycleconsistina of 15 min at 42 C. 5 min at 99 C and 5 min at5 C, and-amplified by 30 cycles of PCR (94 C for 1.5min, 55 C for 1 min, 72 C for 1.5 minlcycle). Controlsamples were run in an identical fashion without theaddition of RNA.Southern blot hybridization: After electrophoresis on a2% agarose gel, the transcript sizes were determined bycomparison with a DNA molecular weight marker(Molecular Weight Marker XI, Boehringer Mannheim Co,Indianapolis, IN). The PCR products were thentransferred onto a positively charged nylon membraneand the DNA was cross-linked onto the membrane using

670

at UECE on March 31, 2009jcem.endojournals.orgDownloaded from

http://jcem.endojournals.org/http://jcem.endojournals.org/http://jcem.endojournals.org/

8/8/2019 Identification and Cellular Localization of Growth Hormone

2/3

RAPID COMMUNICATIOKS 671a stratalinker (Stratagene, La Jolla , CA) The membranewas incubated for one hour in prehybrid ization mixture(5X SSC [1X SSC is 0.15 M sodium chloride and 0.015M sodium citrate, pH 7 41, 1% Blocking reagent[Boehringer Mannheim Co, Indianapolis, IN], 0.1% N-lauroylsarcosine, 0.02% sodrum dodecyl sulfate (SDS)),and hybridized at 52 C for 5-6 hours with a solutioncontaining the prehybridizatlon mixture and thedigoxigenin 3-tail labeled oligoprobe (5 nmol/ml). Lumi-Phos (Boehringer Mannheim Co, Indranap-olrs, IN) wasused to detect the digoxigenln-labeled hybridizedproducts on Kodak XAR films (Rochester, NY)In s/U hybridmtion: Sections of 10 urn were cut seriallyfrom paraffin-embedded blocks, placed on sialinatedslrdes, and stored at room temperature for ISH. Prior toISH, the sections were deparaffinized by washing inxylene and serial ethanol. ISH was carned out accordingto the method of Boehringer Mannheim, using a 16 hourIncubation at 52 C with a drgoxigenin- labeled GHRprobe (100 pmoV500 ul) The slides were then seriallywashed in 4X, 2X, 1X and 0 5X SSC, incubated for onehour in a blocklng solution (2% normal sheep serum,0 1% Tnton X-100) and Incubated with an ant!-digoxigenin antibody conjugated to alkalinephosphatase (Boehnnger Mannhelm Co, Indianapolis,IN) for 5-6 hours, followed by a wash to remove

unbound antibody. After hybridization was detectedusing a color development mixture (NBT and X-phosphate, Boehringer Mannheim Co, Indianapolis, IN),the slides were air dryed, mounted and photographed.ResultsThe appropriate size PCR-amplified products for GHR

(578 bp) and GAPDH (195 bp) were detected bySouthern blot hybridization in all 7 ovaries (Fig 1). Thedoublet band in some lanes may represent a splicingvariant of the GHR (9). In situ hybridization localizedGHR mRNA to the granulosa cel l layer of dominant(n=2), developing (n=2) and atretic (n=lO) follicles, largeand small cells of corpora lutea (n=4), corpora albicans(>2/section) and the endothel ium of blood vessels. GHRmRNA was not detected in primordial follicles andoocytes (5 ovaries), theta, or stromal cells. Controlsections lacked specific staining (Fig 2).

578bp. m - -*m-IIFigure 1. Southern blot hybridization of GHR PCR geneproducts from 7 human ovaries. The double band insome samples may represent a splicing variant.

C 0

Figure 2. In situ hybndtzation for GHR mRNA. A Dominant fol lic le (A, x40) and atrehc foll icle (8, x10) with messagein the granulosa cells and corpora alblcans (B). Theta cells are negative. C. Sense probe applied to corpus luteum(x10). D Oocytes are negative (x40) E Posltrve signal In corpus albicans and blood vessels (x10). F. Positive signalIn corpus luteum (x10) A _ F, stroma IS negative

at UECE on March 31, 2009jcem.endojournals.orgDownloaded from

http://jcem.endojournals.org/http://jcem.endojournals.org/http://jcem.endojournals.org/

8/8/2019 Identification and Cellular Localization of Growth Hormone

3/3

672 RAPID COMMUNICATIONS JCEt.cM.1994Vol79. No

DiscussionRecently, the role of GH in ovarian physiology hasbeen the subject of increasing interest (10). In femalerats, GH deficiency delays puberty, and decreases theestradiol response to hCG (11); GH administrationreverses these effects (2,ll). Similarly, in humans,exogenous GH treatment induces puberty in GHdeficient children with pubertal delay (12). While somereports show that GH reduced the dose of humanmenopausal gonadotropins needed to develop folliclesfor in vitro fertilization, others failed to confirm this effect(13,14). Moreover, successful pregnancy was achievedin a patient with Laron-type dwarfism, characterized bycongenital IGF-I deficiency secondary to a GH receptorabnormality (15). Taken together, these data suggestthat GH has a permissive and modulatory rather than anobligatory role in follicular development (15).GH might affect ovarian function by stimulation ofhepatic IGF-I, or by induction of intra-ovarian IGF-1 orother factor(s) that modulate FSH action. In vitro studieshave suggested a direct effect: GH augmented estrogenand progesterone secretion by rat, porcine and humangranulosa cells above that induced by maximally-stimulating doses of PMSG (3), and stimulated folliculargrowth, oocyte maturation and estradiol production inperfused rabbit ovaries (4) and human granulosa cells(5) in the absence of LH and FSH.GH may bind to receptors for GH, prolactin or placentallactogen (17). By targeting an intracellular portion of theGHR bearing no similarities to the prolactin or placentallactogen receptors, we have localized a pure ovarianGHR. Similarly, choosing a probe from the intracellularand not extracellular domain of the GHR excludedidentification of the GH-binding protein.This study demonstrates for the first time the presenceand cellular localization of GHR gene expression in theintact human ovary. These data agree with previouspreliminary findings showing GHR in the highlyluteinized granulosa cells, but not in theta-interstitialcells, obtained after gonadotropin stimulation during IVFcycles (10). The localization of GHR to the granulosaand corpus luteum, taken with the previous reports ofGH-augmented folliculogenesis and steroidogenesis,support a direct action of GH, via its receptor, tomodulate ovarian steroidogenesis. As an earlier studyfailed to detect IGF-I mRNA in the granulosa cells ofdominant or developing follicles (16), we speculate thatGH may act through a mechanism independent of IGF-1induction.Acknowledgement We thank Drs J. Palmer and C.MacLeod, and the other gynecologists and pathologistsat Holy Cross Hospital for their role in tissue collection.

References1. Adashi EY, Resnick CE, Hurwitz A, et al. 1991Insulin-like growth factors: the ovarian connection. HumReprod. 6:1213-1219.2. Jia XC, Kalmin J, Hsueh AJW. 1986 Growth hormoneenhances follicle-stimulating hormone-induceddifferentiation of cultured rat granulosa cells.Endocrinology. 118:1401-09.

3. Hutchinson LA, Findlay JK, Herington AC. 1988Growth hormone and insulin-like growth factor-laccelerate PMSG-induced differentiation of granulosacells. Mol Cell Endocrinol. 55:61-69.4. Yoshimura Y, Nakamura Y, Koyama N, et al. 1993Effects of growth hormone on follicle growth, oocytematuration, and ovarian steroidogenesis. Fertil Steril.59:917-23.5. Mason HD, Martikainen H, Beard RW, et al. 1990Direct gonadotrophic effect of growth hormone onoestradiol production by human granulosa cells in vitro.J Endocrinol. 126:Rl-4.6. Tapanainen J, Martikainen H, Voutilainen R, et al.1992 Effect of growth hormone administration on humanovarian function and steroidogenic gene expression ingranulosa-luteal cells. Fertil Steril. 58:726-32.7. Lobie PE, Breipohl W, Garcia Aragon J, Waters MJ.1990 Cellular localization of the growth hormonereceptor/binding protein in the male and femalereproductive systems. Endocrinology. 126:2214-21.8. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ.1979 Isolation of biologically active ribonucleic acid fromsources enriched in ribonuclease. Biochemistry.18 :5294-g.9- Mercado M, DaVila N, McLeod JF, Baumann G. 1994Distribution of growth hormone receptor messengerribonucleic acid containing and lacking exon 3 in humantissues. J Clin Endocrinol Metab. 78:731-5.10. Katz E, Ricciarelli E, Adashi EY. 1993 The potentialrelevance of growth hormone to female reproductivephysiology and pathophysiology. Fertil Steril. 59:8-34.11. Ramaley JA, Phares CK. 1980 Delay of pubertyonset in females due to suppression of growth hormone.Endocrinology. 106:1989-93.12. Sheikholislam BM, Stempfel RS. 1972 Hereditaryisolated somatotropin deficiency: Effects of humangrowth hormone administration. Pediatrics. 49:362-374.13 Homburg R, West C, Torresani T, Jacobs HS. 1990Cotreatment with human growth hormone andgonadotropins for induction of ovulation: a controlledclinical trial. Fertil Steril. 53:254-260.14. Younis JS, Simon A, Koren R, et al. 1992 The effectof growth hormone supplementation on in vitrofertilization outcome: a prospective randomized placebo-controlled double-blind study. Fertil Steril. 58:575-80.15. Dor J, Ben-Shlomo I, Lunenfeld B, et al. 1992Insulin-like growth factor-l (IGF-I) may not be essentialfor ovarian follicular development: Evidence from IGF-Ideficiency. J Clin Endocrinol Metab. 74:539-542.16. El-Roeiy A, Chen X, Roberts VJ, et al. 1993Expression of insulin-like growth factor (IGF-I) and IGF-II and the IGF-I, IGF-II, and insulin receptor genes andlocalization of the gene products in the human ovary. JClin Endocinol Metab. 77:141 l-8.17. Kelly PA, Djiane J, Postel-Vinay MC, Edery M. 1991The prolactin/growth hormone receptor family. EndocrRev. 12:235-51.

at UECE on March 31, 2009jcem.endojournals.orgDownloaded from

http://jcem.endojournals.org/http://jcem.endojournals.org/http://jcem.endojournals.org/