- Email: [email protected]
PTHrP Receptor mRNA
Cultured Human Fibroblasts and Not Cultured Human Keratinocytes Express a PTH/PTHrP Receptor mRNA Nancy M. Hanafin,* Tai C. Chen,* Gerhardt Heinrich,")' Gino V. Segre,$ and Michael F. Holick* vitamin 1). Skin, and Bone Research Laboratory. Sections of 'Endocrinology and tMolecular Medicine. nepartniLMits of Medicine and Physiology. Boston Univorsit)' Medical CLMiter; and ^[^Endocrinc Unit. Department ot Medicine. Massachusetts General Mospiral. Harvard Medical School. Boston. MassuclnisL-tts, U.S.A.
There is increasing evidence that parathyroid hormone (PTH) and PTH-related peptides (PTHrP) are involved in normal skin cell grovi^th; therefore, we investigated whether the PTH/PTHrP receptor was expressed in cultured human keratinocytes and dermal fihroblasts. Northern analyses of poly (A)^ RNA isolated from cultured fihrohlasts revealed two PTH/ PTHrP receptor transcripts with one major hand at 2.5 kh and one minor hand at 2.3 kh. These transcripts were consistent with those found in human osteosarcoma cells, which are known to express PTH/ PTHrP-R mRNAs. In contrast, after repeated Northem analyses no PTH/PTHrP receptor transcripts were found in poly (A)""" RNA isolated from cultured keratinocytes. Reverse-transcriptase/nested polymerase chain reaction analyses of total RNA isolated
from cultured keratinocytes and fihrohlasts confirmed the Northern analyses data that the PTH/ PTHrP receptor was expressed in cultured fihrohlasts hut not in cultured keratinocytes. When cultured fihrohlasts and keratinocytes were exposed to 10 M PTH (1-34) there was a twofold increase in cAMP levels in the fihrohlasts and no demonstrahle increase was noted in keratinocytes. These results suggest that skin fihrohlasts possess the classical PTH/PTHrP receptor and are target cells for PTH and PTHrP w^hereas keratinocytes do not have the receptor and are unresponsive to its N-terminal agonist in the stimulation of cAMP formation. Key wovds: cAMPI nested PCR/3T3 fibrohlasts. J Invest Dennatol 105:133137, 1995
P
A variety of normal tissues synthesize and secrete PTHrP |3]. In fact, the presence of PTHrP bioactivity in uonmalignant cells was first demonstrated in conditioned medium harvested trom contitieut human keratinocyte culttires 111)]. However, the physiologic role and the fnnctional pathway ofthis peptide in normal skin is still unclear, although some have suggested that PTI irP may function as an atitocrine or paracrine factor |3,1 1]. In vitro and /'/( vivo studies in our laboratory have demonstrated that PTH (1-34) and PTHrP (1-34) are potent inhibitors of epidermal cell proliferation and PTH (7-34), an antagonist ofthe PTH/PTHiP receptor, can block the antiproliferative etfect of PTH (1-34) and stimulate epidermal proliferation in cultured keratinocytes 112,13]. Additionally. Kaiser cl al 114], after expressing an antisense PTHrP sequence in an established luiman keratinocyte cell line, demonstrated enhanced keratinocyte cell growth. In considering these responses of keratinocytes to PTH and PTHiP, we wondered if the PTH/PTHiP receptor, presumably at least one mediator for the biologic expression of both the PTH and PTHrP hormones, was present in cultured human keratiuocytes and fihroblasts. Because PTH stimulates the fonnation of cyclic adenosine monophosphate (cAMP) in target tissues known to express the receptor, we measured intracelhilar cAMP levels in keratinocytes and fihroblasts after treatment with PTH (1-34). To determine mRNA expression, we also performed Northern analyses on poly (A) ' from cultured keratinocytes and fibroblasts using a cDNA that codes for a common lniman PTH/PTHrP receptor 11 5]. Additionally, we performed reverse transcriptase (RT) and nested
arathyroid honnonc—related peptide's (PTHrP) bioactivity was initially identified in tnmors associated with humoral hypercalcemia of lnalignancy |1,2]. The 141 amino acid PTHrP shares 70% homology with PTH in its first 13 amino acids bnt diverges completely in its primary structme thereafter. Studies using synthetic PTHrP N terminal fragmetits have demonstrated that fi'agments that bound to the PTH receptor have caused biologic responses similar to PTH in cultured bone and kidney cell.s | 3 | . Several groups have postulated that a single receptor species, at least in bone and kidney cells, may mediate many pliysiologic fiinctions ot hoth PTH and PTHrP [ 4 - 6 | . A human PTH/pTHrP receptor cDNA encoding a 593 amino acid protein has been cloned (7]. Using a cDNA probe for this PTH/PTHrP receptor, Urena ct al \H] demonstrated PTH/ PTHrP receptor transcripts in many nonclassical PTH target organs and tissues, including the skin. Two equally abundant PTH/PTHrP receptor traii.scripts were present in poly (A) RNA i.solated trom whole rat skin; a 2.3—2..S-kb transcript, identical to that found in bone and kidney and a smaller 2.()-2.2-kb transcript. The data support early experimentation that suggested PI H receptors were present in skin fibroblasts [9].
Manuscript received Occemhcr 17, 1994; final revision received February 17, 1995; accepted for puhlication Mardl 3. f995. Reprint requests to: M. P. Holick, Vitamin D. Skin, and Uonc Research Laboratory, Boston University Medical C'eiiter, SO East C^oncord Street. M-1013, Boston, MA fl211K.
()fl22-2()2X/')5/S09.50 • SSD1()022-202X(95)()OI60-M • Copyright « - 1995 liy The Society fbr Invosdgalix e Dermatology. Inc. 133
134
HANAFIN tif .-l
polymerase chain reaction (PCR) analyses on total lVNA isolated from human keratinocytes and both human and mouse fibroblasts, EXPERIMENTAL P R O C E D U l ^ S Cell Culture Keratinocytes were grown in culttire using a modification of Rheinwald and Green [16J, Keratinocytes were obtained From neonatal foreskins and plated on a lethally irradiated 3 r3 feeder layer in a serum-free MCDB 1 53 basal medium containing 0,15 niM calcium and supplemented with growth factors, including bovine pituitary extract, 3 /xg/ml; epidermal growth factor, 25 ng/ml; insulin, 5 ;u,g/ml; hydrocortisone, 200 ng/ml; cholera toxin, 0,1 /xg/ml, and prostaglandin E,, 50 ng/ml [17], Cells were fed and maintained in the same medium without cholera toxin and hydrocortisone; three times per week. Second-passage keratinocytes grown on 100-mm or 35-mm dishes wore used for Northern analyses or cAMP determinations, respectively. To avoid the potential contamination of 3T3, some primary cultured keratinocytes wcM-e grown without a 3T3 feeder layer for rcversetranscriptase/polymerase chain reaction (RT/PCR) analyses. Normal fibroblasts were grown from human foreskins in Dulbecco's modified Eagle's medium supplemented with 5% fetal bovine scrum as previously described |18], For R T / P C R experiments, cells were harvested from single 100-nim dishes. Total RNA from human osteoblast-like osteosarcoma cells (SaOS-2) for both Northern and RT/PCR analyses was extracted and purified as described cAMP Determination For cAMP experiments, koratinocytes and fibroblasts were plated on 35-nim dishes. Growth factors were removed from the keratinocyte media and fetal bovine serum was removed from the fibroblast media 2 d before the experiment. At the experiment's initiation, cells were preincubated for 30 min at 37°C in 0,5 ml of fresh MCDB-153 or DMEM medium (Sigma, St, Louis, MO) for keratinocytes and fibroblasts, respectively, followed by treating the cells with human parathyroid hormone (hPTH 1-34) (Bachcm, Torrance, CA) at a final concentration of 10 ^ M or vehicle (phospliate-buftcrc'd saline) only. Following incubation for 1(1 mill at 37°C, the reaction was terminated by removing the medium and adding 0,5 ml of 0,1 N NaOH, The resulting cell suspension was neutralized with 1 N HCl and aliquots were cjuantitatcd using a commercial cAMP assay kit (Amersham, Arlington Heights, IL), Standard error of the mean was calculated from triplicate determinations. Northern Analysis Total cellular RNA was prepared from koratinocytes and fibroblasts using a modification of the guanidinium thiocyanatc-cesium chloride technique [19], Total lUMA from mouse kidney was prepared in a similar manner. Poly (A)^ RNA was further isolated from keratinocyte and fibroblast by utilizing a PolyATtract mRNA Isolation System (Promega, Madison, Wl), Total RNA anci poly (A) ' samples were denatured and fractionated on a 1%/37% agarose/formaldehydc gel containing ethidium bromide. An RNA ladder (Gibco/BRL, Grand Island, NY) was also run to determine message size at the hybridization bands. The RNA was transferred via capillary action to a nylon inembrane, Hybond N (Amersham) and cross-linked to the membrane via an ultraviolet transilluminator (IBI), Membranes were prehybridizcd at 42°C for 2 h in a solution containing 50% formamide, 1 00 ^g/ml denatured nonhomologous DNA, 5 X SSC, 0,5% sodium dodecyl sulfate, 10 X Denhardt's, fifty millimoles per liter of" Na2PO^, and 10% dextran sulfate (Sigma), A cDNA (Gcnbank accession number, L043()8) encoding a human PTH/ PTHrP receptor [15] was labeled with [ a " P ] dCTP (Amersham) via the Prime-It II Random Primer Labeling Kit (Stratagene, La Jolla, CA), A Sophadex G-5() column (Boehringer Mannheim Corp, Indianapolis, IN) was used to remove unincorporated free nucleotide. All blots were stripped and rchybridized with a human glyceraldehydc-3-phosphatc dehydrogcnasc (GAPDH) oligonucleotide (Oncogene Science, Uniondalc, NY), whose sequence was based on the cDNA sequence cloned by Tso and colleagues [20],
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
The oligonucleotide was 5' labeled with [7'"P] ATP (Amersham), A Sephadex G-25 column (Boehringer Mannheim Corp) was used to remove unincorporated free nucleotidc. Hybridizations took place overnight at 65°C in a solution containing 1 M NaCl, 50 niM Tris-HCl (pH 7,5), 10% dextran sulfate, 1% sodium dodecyl sulfate, and 100 /j,g/ml denatured nonhomologous DNA (Sigma), Autoradiography of blots was carried out at —7(1°C using Hyperfilm (Amersham) and an intensifying screen (Dupoiit/NEN, Boston, MA), RT/Nested PCR Analysis Total RNA aliquots (1 /xg) isolated from cultured human keratinocytes, grown either with or without a 3T3 feeder layer, human fibroblasts, human osteoblast-like osteosarcoma cells (SaOS-2), and immortalized mouse fibroblasts (3T3s) were added to separate reverse transcriptase (RT) reactions [21], The buffer consisted of 7,0 mM MgCU, 50 mM KCl, 10 mM Tris-HCl (pH 8,3), deoxynucleotide triphosphates at 1 mM each, 1 unit RNase inhibitor, 2,5 /xM random hexamers, and 2,5 units of cloned Moloney Murine Leukemia Virus (M-MLV) Reverse Transcriptase (Perkin-Elmer Cetus/Applied Biosystems, Foster City, CA), The final volume for each reaction was 20 fjd. The reaction mixtures were incubated at room temperature for 10 min, 42°C for 45 mill, 99°C for 5 min, and 5°C for 5 min, PCR amplifications were carried out in the Gene Amp PCR system 9600 (Perkin-Elmer Cetus), Amplifications were done in a final buffer volume of 100 fxX consisting of 2 mM MgCU, 50 mM KCl, 10 mM Tris-HCL (pH 8,3), and 2,5 units AmpIiTaq DNA Polymerase (Perkin-Elmer Cetus), PCR for glyceraldehyde-3-phosphate dehydrogeiiase (GAPDH) was performed using ] 0 [xM of the forward primer 5'-TCCCATCACCATCTTCCA-3' and 10 /xM o f t h e reverse primer 5'-GTCCACCACCCTGTTGCT-3', PCR conditions were 95°C for 2 min, followed by 35 cycles of 95°C for 30 seconds, 60°C for 30 seconds, and 72°C for 30 seconds with a final extension at 72°C for 5 min. Nested PCR [22] was used for the PTH/PTHiP-R. The outer forward primer 5'-GGGCACCAGGTGAAGTGGT-3' and the outer reverse primer 5'-GGTTGCTCTGACACCGACCC-3' were used for the first amplification, PCR conditions were 95°C for 2 min, followed by 30 cycles at 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds with a 3-second extension per cycle. Final extension was at 72°C for 5 min. For the second amplification, 5 /xl ofthe lirst-round amplification mixture was added to the same mixture as in the previous amplification but this time the inner primer concentrations were doubled to 20 /xM rather than 10 /xM, The inner forward primer was 5'-CGGGAGGTATTTGACCGCCTAG-3' and the inner reverse primer was 5'-CAGAATCCAGTAGTAGTTGG-3'. PCR conditions were the same as for the first amplification except cycle duration was 25, Samples of 20 /ixl of PCR reaction products were analyzed by electrophoresis on a 2% agarose gel. Bands were visualized by ethidium bromide staining. At some amplifications [a'^P] dCTP (Amersham) was added to reaction mixtures and gels were dried and exposed to Hyperfilm (Amersham), The PCR amplification products for PTH/PTHrP-R were further analyzed for authenticit)' by restriction enzyme digestion with Fok 1, Bsg I, and Drd I (New England Biolabs, Beverly, MA),
IUESULTS Stimulation of cAMP Production in Fibroblasts Treated with hPTH (1-34) The ability of a PTH N-terminal fragment to stimulate cAMP production was studied in cultured normal human keratinocytes and dermal fibroblasts. Treatment of fibroblasts with 1 0 " ' M of hPTH (1-34) caused a twofold increase in cAMP levels over the basal level (4,2 ± 0,1 pmol/dish versus 2,0 ± 0,05 pmol/dish). The data were significantly different and p was <0,005, In contrast, when keratinocytes were treated with 10 ' M of liPTH (1-34) no stimulation of cAMP formation was observed. Treated keratinocytes were 4,0 ± 0,2 pmol/dish and untreated keratinocytes were 3,9 ± 0,1 pmol/dish. There was no significant difference between these data points. Because hPTH (1-34) stimulated cAMP
VOL. 105. NO. 1 JULY
PTH/PTHrP RECEPTOR mRNA IN FIIJROHLASTS
135
A PTH/PTHrP-R 2.5 Kb
1 2
3
4
5
1 2
3 4 5
12
3 4 5 6 7
GAPDH
e
Figure 1. PTH/PTHrP-R expression in cultured human fibroblasts as sbown by Northern analysis and RT/nested PCR analysis. A) Northern blot analysis of total RNA prepared from SaOS-2 {lattes I attd J) and monse kidney cells {latte 4) and poly (A) ' RNA from keratinocytes {latte 2) and fibroblasts {latte 5). Tbe blots were h\'bridized witb tbe fuU-lengtb cDNA encoding tbe buman PTH/PTHrP receptor. B) Lane I was RT and GAPDH PCR amplification of total lOvIA from human keratinocytes cultured witbout a ^'V3 feeder layer. Lane 2 was an aliquot ot tbese same cells after RT and nested PCR amplification vvith PTH/ PTHrP-R primers. Lanes J through 6 are RT/uested PCR samples from buman keratinoeytes cultured witb 3T3 cells, human fibroblasts. luiman osteosarcoin.i cells, and .^T."^ cells, respectively. Lane 7 is tbe nested PCR product from tbe human PTH/PTHrP receptor cDNA.
GAPDH PTH/PTHrP-R
formation in cultured fibroblasts but not in cultured keratinocytes, we examined whether cultured human tibroblasts and keratinocytes possessed the same PTH/PTHrP receptors as those found in kidney and bone cells. Northern Blot Analysis Utilizing the recently cloned, fuUlengtb human PTH/1'THrP receptor cONA [15], Northern analysis was performed on poly (A) ' RNA from cultured human keratinocytes and fibroblii.sts. Jii all hybridization e.xperiment.s, approximately 1 0 /xg total RNA from mouse kidney cells and/or buman osteoblast-like osteosarcoma cells (SaOS-2) vvere used as positive controls. Lane 1 of Fig IA contained the positive control SaOS-2 cells. Lane 2 contained 5 iig of poly (A) ^' RNA from keratinocytes. No signal was visible in this sample. Lanes 3 and 4 contained SaOS-2 and mouse kidney total RNA, respectively. As can bo seen, a bigbly e.xpressed 2.5 kb PTH/PTHrP receptor niRNA was found in both SaOS-2 and mouse kidney cells. Also visible in botb samples was a less prominent transcript at 2.3 kb. Interestingly, three smaller transcripts were expressed in the SaOS-2 bone cells. Lane 5 contained 5 jug of poly (A) ' RNA isolated from fibroblasts. The major 2.5-kb transcript was clearly expressed as well as another at 2.3 kb. No smaller transcripts were observed even after prolonged exposure of tbe autoradiograpb. Etbidium bromide staining of the gel and rebybridization of tbe membrane witb an oligonuclcotide specific for buman GAPDH confirmed sample integrity and relative loading concentrations. In Fig IA, tbe GAPDH signals \vere visible in all ofthe lanes e.vcept for lane 4, the mouse kidney cells. Because tbe GAPDH oligonucleotide probe was specific for human cells, uo detectable signal was e.xpected. Apparent and expected were the intensities of tbe GAPDH signals in the poly (A) ' RNA from koratinocytes and fibroblasts as compared to tbe GAPDH signals from SaOS-2 total RNA. In a comparison hybridization experiment, five times tbe amount of poly (A) ' RNA from keratinocytes as from fibrobla.st.s .still did not .show PTH/PTHrP-R expression, even with prolonged exposure of tbe autoradiograpb (data not shown). In a total of five separate analyses, the PTH/ PTHrP receptor messages were consistently found in poly (A) ' IOvFA samples prepared from cultured dermal fibroblasts but not from cultured keratinocytes. RT/NESTED PCR Analysis Reverse transcriptase and nested PCR was used to analyze total RNA isolated from buman keratinocytes and fibroblasts as well as from monse immortalized fibroblasts (3T3s) and from human osteosarcoma cells (SaOS-2). In these experiments, tbe success of tbe cl3NA .synthesis from tlic R T
reaction was verified by amplifying an aliquot of tbe RT product witb primers for the GAPDH constitutive gene and confirming by gel electrophoresis the presence ofthe predicted PCR product. For PTH/PTHrP analyses, parallel first-amplification reactions were performed on cDNA aliquots and PCR produces were analyzed by gel electropboresis. Tbe first amplification of a bnman PTH/ PTHrP receptor cDNA produced tbe predicted 675-bp prodnct and tbe second amplification produced tbe predicted 36()-bp product. Nested PCR analyses of R T products from buman keratinocytes tbat were initially cultured on a 3T3 feeder layer, as well as from human fibroblasts, monse 3T3 fibroblasts, and human osteosarcoma cells demonstrated identical results to those from tbe bnman PTH/PTHrP receptor cDNA. In contrast, nested PCR analyses of RT products from buman keratinocytes tbat were cultured entirely without a 3T3 feeder layer (Fig IB) demonstrated neither a first nor second PCR amplification product. Lane I contained tbe RT/GAPDH-amplified PCR product from keratinocytes grown without a 3T3 feeder layer. Tbe predicted 762-bp product is clearly visible. Lane 2 contained an identical aliquot from tbis R T reaction tbat was amplified via tbe nested PCR protocol previously described. No PCR products were visible. Lane 3 contained an RT/nested PCR sample from keratinocytes tbat were cultured witb a .3T3 feeder layer. Lanes 4 through 6 were RT/nested PCR samples from luiman fibroblasts, luiman osteosarcoma cells, and mouse 3T3 fibroblasts, respectively. Lane 7 contained tbe nested PCR product from tbe buman PTH/PTHrP receptor cDNA. As can be seen, lanes 3 through 7 displayed identical second-amplification PCR products ofthe predicted 360 bp size. First-ainplificatioii PCR products of 675 bp were also identical for tbese samples and parallel amplifications with GAPOH primers produced the predicted product (data not shown). Restriction enzyme digests of the amplification products witb Fok I, Bsg 1, and l^rd 1 furtber confirmed tbeir autbenticity. DISCUSSION Although we |12,13J have demonstrated that specific PTH and PTHrJ^ N-tci-niinal fragments are antiproliferative agents in cnltnred luiman keratinocytes and in monse epidermis in vii'o. Nortliern analyses indicated the PTH/PTHrP receptor niRNA was not expressed in cultured keratinocytes. In contrast, cultured dermal fibroblasts expressed two PTH/PTHrP receptor transcripts with one major band at 2.5 kb and one minor band at 2.3 kb. The transcripts observed in fibroblasts were consistent witb tbose seen in hybridizations of RNA isolated from SaOS-2 cells and monse
136
HANAFIN i:r A
kidney cells, which are known to be PTH/PTHrP receptor positive [8]. The presence of two transcripts may be due to alternative transcription start sites, splicing, or the presence of closely related mRNAs. To increase detection sensitivity for the I'TH/PTHrP transcript, we utilized RT/ncsted PCR. The results confirmed the Northern analyses data and clearly showed that the PTH/PTHrP receptor message was present in human fibroblasts, huinan osteosarcoma cells and mouse 3T3 cells. Interestingly, tho enhanced sensitivity of RT/ncsted PCR revealed the PTH/PTHrP receptor in keratinocytes ctiltured t^n a 3T3 feeder layer. Because we found the PTH/PTHrP receptor in 3T3 cells, it was thought that R T / PCR amplification of RNA isolated from keratinocytes grown on a 3T3 feeder layer could generate receptor-positive amplification products. RT/nested PCR analyses of keratinocytes that were cultured without a 3T3 feeder layer demonstrated no PTH/PTHrP receptor, suggesting that the product observed was derived from the 3T3 cells of the feeder layer and not from the keratinocytes. The data are consistent with the additional ohservation that intracellular cAMP levels increased when fibroblasts were treated with PTH (1-34) (10 ^ M), but in identically treated keratinocytes there was no detectable cAMP response. Because there is uo detectable PTH/PTHrP receptor in human cultured keratinocytes, the data raise the possihility that PTHs and PTHrPs antiproliferative effects on keratinocytes may he mediated by PTH/PTHrP receptors located in adjacent fihrohlasts. Such a pathway is believed to be one way PTH increases osteoclastic bone resorption. Because mature osteoclasts do not have PTH receptors, PTH may first act on osteoblasts and/or stromal fibrohlasts to increase bone resorption [23|. Moreover, paracrine- and/or autocrine-mediated effects on skin cells hy PTHrP are also possihle. It is also possible that there are different receptors for PTH/ PTHrP in keratinocytes and fihroblasts. This idea is strengthened from the variability in the biologic potency of different PTH/ PTHrP analogs in numerous tissues [3]. Studies of the hiologic effects of several PTHrP fragments in different cell types have suggested that there may ho a PTHrP receptor distinct from the classic PTH/PTHrP receptor, which is coupled to adenylate cyclase through the G-protein [7]. For example, recombinant PTHrP (1-141) and PTHrP (1-108) were two to four times more potent than PTHrP (1-34) in the stinuilatioii of plasminogen activator activity in UMR cells |24[. Similarly, hPTHrP (1-34) and hPTHrP (1-108) were not active in inhibiting bone resorption, whereas the carboxy-terminal fragment hPTHrP (107-139) was a potent inhihitor of bone resorption [25|. Such data suggest that a distinct PTHrP receptor(s) may recognize specific fragments. Similarly, other studies suggest that particular PTHrP fragments may induce difFerent signal-transduction mechanisms. Specifically, Orloff rt al 1261 using squamous carcinoma cells demonstrated that hPTHrP (67-86)NH, or hPTHrP (1-141) caused a rapid and transient rise in intracellular calcium in a dose-dependent manner between 10 " M and 10 " M but there was no stimulation of cAMP formation detectable even with 10 '' M hPTHrP (67-86). In a separate study, Orloffcl al |27| stiggest human carcinoma cells and normal keratinocytes express mRNAs related to hut distinct from the human PTH receptor mRNA. If there are distinct receptors and presumably specific PTH and PTHrP fragment activation sites on these receptors, then this could explain no cAMP formation by PTH (1-34) and no PTH/PTHrP receptor expression in keratinocytes. Through cAMP determinations. Northern analyses, and R T / nested PCR analyses, this study localizes the human PTH/PTHrP receptor in fihroblasts but not in keratinocytes. Numerous questions remain, however, as to whether the hiologic activities of PTH and PTHrP in skin cells are mediated through paracrine and/or autocrine pathways or whether identical or distinct receptors exist. More research is necessary to understand the hiologic activation of endogenously occurring IHII and PTHrP iu skin cells and the potential therapeutic use for these hormones in regulating skin cell growth and differentiation.
THE JOURNAL OF INVESTIGATIVE DERMATOl.OGY
This tt'oik tvas sttftpotted itt pnti hy NIHgtants DI\4.i690 and MOIRR()0.'t3.i. IVe thattk Dai'id M. Jackson for his e.\eelleiii gtaphics assistance tt'iilt the ntaitiisctipi.
REFERENCES 1.
2.
3. 4.
5.
6.
7.
8.
9. 10. I I. 12.
13. 14. 15.
16. 17.
18.
I 9. 2(1.
21. 22.
23. 24.
Stewart AF, Horst R, Deftos LJ, Cadman EC, Lang R, Broad.is AE: Biochemical evaluation of patients with cancer-associated hyjiercah emia: e\idence for humoral and nonhumoral groups. ;V luiglj Med 3(J3:1 377-1383. 19811 Strewler GJ, Williams RD, Nissenson RA: I lunian renal carcinoma cells produce hyperialceniia in the nude mouse and a novel protein recognized by panuhyroid hormone receptors. / C/IH Ini'esi 7l:76')-774. 1983 OrloffJJ, Wu 1 L, Stewart Al": Parathyroid hormone-like proteins: biochemical responses and receptor iiKeiactions. Endocr Rev 1(1:476-495, 198'; Juppner II, Abou-Samra AB, Uueuo S, Giu WX, Potts JV Jr. Segre C;V: fhe parathytoid hormoue like peptide associated with humoral hypercillcemia of malignancy and parathyroid hormone bind to the same rece|)tor OTI the plasnia membrane of ROS 17/2.8 cells. / Biot Chetn 263:8557-8560, 1988 Nissenson RA. Karpf I), Hambino T. WiiierJ, Canga M, Nyiredy K, Arnaud CD: c;ovalent labeling of a higli-alFmity guanyl micleotide-sensitive parathyroid hormone receptor in canine renal cortex. Biochetnisiiy 26:1 874-1878, 1987 Fukayama S, Bosnia TJ, Goad DL, Voelkel EF, Tashjian AH Jr: Human parathyroid lK>riiioiie (PTI I)-reiated protein and luiman P'l'M: comparative biological activities ou huinan bone cells aud bone resorption. Ettdocritioloi^y 123:2841-2848, 1988 Juppner FL Abou-Samra AB, Freeman M, Kong XF, Schipani F., Richards J, Kolakowski LF, Flock J, Potts JT Jr, Kronenberg HM, Segre C:V: A G protein-linked receptor for paratbyroid bornione and parathyroid hormonerelated peptide. Science 254:1024-1026, 1991 Urena P, Kong XF, Abou-Samra AB, Juppner 11, Krouenberg HM, Potts JTJr, Segre GV: Paratbyroid bornione (P FI l)/PTH-related peptide receptor message ribomicleic acids are widely dislribuled in rat tissues. F.itdocrittolot;y 133:617-623, 1993 Goldring SR. Mabaffey JE, Rosenblatt M, Dayer JM, Potts |T Jr, Krane SM: Paratbyroid liormone iiiliibitors: comparison of biological acti\ity in bone and skin derived tissue. / ('litt Endocrinol Meiah 48:655-659, 1979 Mereiidini> JJ Jr, Insogna KL, Milstoue LM. Broadus AE, Stewart AF: CAiltured buman keratinocytes produce a parathyroid hormoue-like protein. Science 231:388-390, 1986 OrlolfJJ, Wu TL. Stewart AF: Parathyroid hormone-like proteins: biochemical responses and receptor interactions. lUtdocr Rev IO:476-4')5, 1989 Flolick MF, Nussbaum S, i'ersous KS: PI I l-like humoral hypercalcemia factor (I IMF) of malignancy may be an epidetmal differentiation factor: synthetic hllHF (I-34)NII, inliibits prolileiation and induces terminal differentiation of ctiltured btiman keratinocytes. / Bone Mineral Res 3:S2I4, 19S8 Holick MF, Ray S, C;hen VC. lian X, Persons K: A paradiyroid hormone antagonist stimtilates epidermal prolifeiatit>n and hair growth in mice. Proc .\'
VOL. 105, NO. I
PTH/PTHrP H.ECEPTOR mRNA IN FIBROBLASTS
JULY 1995
Fenton AJ, Kemp BE, Kent GN, MoseleyJM, Zheng MH. Rowe DJ, BrittoJM, Martin TJ, Nicholson GC: A carboxyl-teriuinal peptiiie from the parathyroid hormone-related protein inhibits bone resorption by osteoclasts. J:;iAimiw;i)V)' 1 29:1 762-1 768, 1991 Orloff JJ, Kats Y, Milnick M, Gasalla-1 lertaiz J, Isales CM: Evidence for a receptor on squamous carcinoma cell Hues wbich recognizes a mid-region
27.
137
fragment of parathyroid hornioue-related proteiu, PFIlrl' (67-K(i)NH,. / ;ii)/ii' Mittetal Res 8:S133, 1993 Orlolf JJ. Urena P, Scbipani E, Ribaudo AE, Milstoue LM. Pllilbrick WM, Abou-Samra AB, Segre CIV. Juppner 11:1 kiniaii squamous carciiu>iiia cells aud normal keratinocytes express mRNA, related to btit dislinct from litunan P i l l receptor mRNA. / i!..m-A/IMIT,.; K« 7:.i23(). 1992
There is increasing evidence that parathyroid hormone (PTH) and PTH-related peptides (PTHrP) are involved in normal skin cell grovi^th; therefore, we investigated whether the PTH/PTHrP receptor was expressed in cultured human keratinocytes and dermal fihroblasts. Northern analyses of poly (A)^ RNA isolated from cultured fihrohlasts revealed two PTH/ PTHrP receptor transcripts with one major hand at 2.5 kh and one minor hand at 2.3 kh. These transcripts were consistent with those found in human osteosarcoma cells, which are known to express PTH/ PTHrP-R mRNAs. In contrast, after repeated Northem analyses no PTH/PTHrP receptor transcripts were found in poly (A)""" RNA isolated from cultured keratinocytes. Reverse-transcriptase/nested polymerase chain reaction analyses of total RNA isolated
from cultured keratinocytes and fihrohlasts confirmed the Northern analyses data that the PTH/ PTHrP receptor was expressed in cultured fihrohlasts hut not in cultured keratinocytes. When cultured fihrohlasts and keratinocytes were exposed to 10 M PTH (1-34) there was a twofold increase in cAMP levels in the fihrohlasts and no demonstrahle increase was noted in keratinocytes. These results suggest that skin fihrohlasts possess the classical PTH/PTHrP receptor and are target cells for PTH and PTHrP w^hereas keratinocytes do not have the receptor and are unresponsive to its N-terminal agonist in the stimulation of cAMP formation. Key wovds: cAMPI nested PCR/3T3 fibrohlasts. J Invest Dennatol 105:133137, 1995
P
A variety of normal tissues synthesize and secrete PTHrP |3]. In fact, the presence of PTHrP bioactivity in uonmalignant cells was first demonstrated in conditioned medium harvested trom contitieut human keratinocyte culttires 111)]. However, the physiologic role and the fnnctional pathway ofthis peptide in normal skin is still unclear, although some have suggested that PTI irP may function as an atitocrine or paracrine factor |3,1 1]. In vitro and /'/( vivo studies in our laboratory have demonstrated that PTH (1-34) and PTHrP (1-34) are potent inhibitors of epidermal cell proliferation and PTH (7-34), an antagonist ofthe PTH/PTHiP receptor, can block the antiproliferative etfect of PTH (1-34) and stimulate epidermal proliferation in cultured keratinocytes 112,13]. Additionally. Kaiser cl al 114], after expressing an antisense PTHrP sequence in an established luiman keratinocyte cell line, demonstrated enhanced keratinocyte cell growth. In considering these responses of keratinocytes to PTH and PTHiP, we wondered if the PTH/PTHiP receptor, presumably at least one mediator for the biologic expression of both the PTH and PTHrP hormones, was present in cultured human keratiuocytes and fihroblasts. Because PTH stimulates the fonnation of cyclic adenosine monophosphate (cAMP) in target tissues known to express the receptor, we measured intracelhilar cAMP levels in keratinocytes and fihroblasts after treatment with PTH (1-34). To determine mRNA expression, we also performed Northern analyses on poly (A) ' from cultured keratinocytes and fibroblasts using a cDNA that codes for a common lniman PTH/PTHrP receptor 11 5]. Additionally, we performed reverse transcriptase (RT) and nested
arathyroid honnonc—related peptide's (PTHrP) bioactivity was initially identified in tnmors associated with humoral hypercalcemia of lnalignancy |1,2]. The 141 amino acid PTHrP shares 70% homology with PTH in its first 13 amino acids bnt diverges completely in its primary structme thereafter. Studies using synthetic PTHrP N terminal fragmetits have demonstrated that fi'agments that bound to the PTH receptor have caused biologic responses similar to PTH in cultured bone and kidney cell.s | 3 | . Several groups have postulated that a single receptor species, at least in bone and kidney cells, may mediate many pliysiologic fiinctions ot hoth PTH and PTHrP [ 4 - 6 | . A human PTH/pTHrP receptor cDNA encoding a 593 amino acid protein has been cloned (7]. Using a cDNA probe for this PTH/PTHrP receptor, Urena ct al \H] demonstrated PTH/ PTHrP receptor transcripts in many nonclassical PTH target organs and tissues, including the skin. Two equally abundant PTH/PTHrP receptor traii.scripts were present in poly (A) RNA i.solated trom whole rat skin; a 2.3—2..S-kb transcript, identical to that found in bone and kidney and a smaller 2.()-2.2-kb transcript. The data support early experimentation that suggested PI H receptors were present in skin fibroblasts [9].
Manuscript received Occemhcr 17, 1994; final revision received February 17, 1995; accepted for puhlication Mardl 3. f995. Reprint requests to: M. P. Holick, Vitamin D. Skin, and Uonc Research Laboratory, Boston University Medical C'eiiter, SO East C^oncord Street. M-1013, Boston, MA fl211K.
()fl22-2()2X/')5/S09.50 • SSD1()022-202X(95)()OI60-M • Copyright « - 1995 liy The Society fbr Invosdgalix e Dermatology. Inc. 133
134
HANAFIN tif .-l
polymerase chain reaction (PCR) analyses on total lVNA isolated from human keratinocytes and both human and mouse fibroblasts, EXPERIMENTAL P R O C E D U l ^ S Cell Culture Keratinocytes were grown in culttire using a modification of Rheinwald and Green [16J, Keratinocytes were obtained From neonatal foreskins and plated on a lethally irradiated 3 r3 feeder layer in a serum-free MCDB 1 53 basal medium containing 0,15 niM calcium and supplemented with growth factors, including bovine pituitary extract, 3 /xg/ml; epidermal growth factor, 25 ng/ml; insulin, 5 ;u,g/ml; hydrocortisone, 200 ng/ml; cholera toxin, 0,1 /xg/ml, and prostaglandin E,, 50 ng/ml [17], Cells were fed and maintained in the same medium without cholera toxin and hydrocortisone; three times per week. Second-passage keratinocytes grown on 100-mm or 35-mm dishes wore used for Northern analyses or cAMP determinations, respectively. To avoid the potential contamination of 3T3, some primary cultured keratinocytes wcM-e grown without a 3T3 feeder layer for rcversetranscriptase/polymerase chain reaction (RT/PCR) analyses. Normal fibroblasts were grown from human foreskins in Dulbecco's modified Eagle's medium supplemented with 5% fetal bovine scrum as previously described |18], For R T / P C R experiments, cells were harvested from single 100-nim dishes. Total RNA from human osteoblast-like osteosarcoma cells (SaOS-2) for both Northern and RT/PCR analyses was extracted and purified as described cAMP Determination For cAMP experiments, koratinocytes and fibroblasts were plated on 35-nim dishes. Growth factors were removed from the keratinocyte media and fetal bovine serum was removed from the fibroblast media 2 d before the experiment. At the experiment's initiation, cells were preincubated for 30 min at 37°C in 0,5 ml of fresh MCDB-153 or DMEM medium (Sigma, St, Louis, MO) for keratinocytes and fibroblasts, respectively, followed by treating the cells with human parathyroid hormone (hPTH 1-34) (Bachcm, Torrance, CA) at a final concentration of 10 ^ M or vehicle (phospliate-buftcrc'd saline) only. Following incubation for 1(1 mill at 37°C, the reaction was terminated by removing the medium and adding 0,5 ml of 0,1 N NaOH, The resulting cell suspension was neutralized with 1 N HCl and aliquots were cjuantitatcd using a commercial cAMP assay kit (Amersham, Arlington Heights, IL), Standard error of the mean was calculated from triplicate determinations. Northern Analysis Total cellular RNA was prepared from koratinocytes and fibroblasts using a modification of the guanidinium thiocyanatc-cesium chloride technique [19], Total lUMA from mouse kidney was prepared in a similar manner. Poly (A)^ RNA was further isolated from keratinocyte and fibroblast by utilizing a PolyATtract mRNA Isolation System (Promega, Madison, Wl), Total RNA anci poly (A) ' samples were denatured and fractionated on a 1%/37% agarose/formaldehydc gel containing ethidium bromide. An RNA ladder (Gibco/BRL, Grand Island, NY) was also run to determine message size at the hybridization bands. The RNA was transferred via capillary action to a nylon inembrane, Hybond N (Amersham) and cross-linked to the membrane via an ultraviolet transilluminator (IBI), Membranes were prehybridizcd at 42°C for 2 h in a solution containing 50% formamide, 1 00 ^g/ml denatured nonhomologous DNA, 5 X SSC, 0,5% sodium dodecyl sulfate, 10 X Denhardt's, fifty millimoles per liter of" Na2PO^, and 10% dextran sulfate (Sigma), A cDNA (Gcnbank accession number, L043()8) encoding a human PTH/ PTHrP receptor [15] was labeled with [ a " P ] dCTP (Amersham) via the Prime-It II Random Primer Labeling Kit (Stratagene, La Jolla, CA), A Sophadex G-5() column (Boehringer Mannheim Corp, Indianapolis, IN) was used to remove unincorporated free nucleotide. All blots were stripped and rchybridized with a human glyceraldehydc-3-phosphatc dehydrogcnasc (GAPDH) oligonucleotide (Oncogene Science, Uniondalc, NY), whose sequence was based on the cDNA sequence cloned by Tso and colleagues [20],
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
The oligonucleotide was 5' labeled with [7'"P] ATP (Amersham), A Sephadex G-25 column (Boehringer Mannheim Corp) was used to remove unincorporated free nucleotidc. Hybridizations took place overnight at 65°C in a solution containing 1 M NaCl, 50 niM Tris-HCl (pH 7,5), 10% dextran sulfate, 1% sodium dodecyl sulfate, and 100 /j,g/ml denatured nonhomologous DNA (Sigma), Autoradiography of blots was carried out at —7(1°C using Hyperfilm (Amersham) and an intensifying screen (Dupoiit/NEN, Boston, MA), RT/Nested PCR Analysis Total RNA aliquots (1 /xg) isolated from cultured human keratinocytes, grown either with or without a 3T3 feeder layer, human fibroblasts, human osteoblast-like osteosarcoma cells (SaOS-2), and immortalized mouse fibroblasts (3T3s) were added to separate reverse transcriptase (RT) reactions [21], The buffer consisted of 7,0 mM MgCU, 50 mM KCl, 10 mM Tris-HCl (pH 8,3), deoxynucleotide triphosphates at 1 mM each, 1 unit RNase inhibitor, 2,5 /xM random hexamers, and 2,5 units of cloned Moloney Murine Leukemia Virus (M-MLV) Reverse Transcriptase (Perkin-Elmer Cetus/Applied Biosystems, Foster City, CA), The final volume for each reaction was 20 fjd. The reaction mixtures were incubated at room temperature for 10 min, 42°C for 45 mill, 99°C for 5 min, and 5°C for 5 min, PCR amplifications were carried out in the Gene Amp PCR system 9600 (Perkin-Elmer Cetus), Amplifications were done in a final buffer volume of 100 fxX consisting of 2 mM MgCU, 50 mM KCl, 10 mM Tris-HCL (pH 8,3), and 2,5 units AmpIiTaq DNA Polymerase (Perkin-Elmer Cetus), PCR for glyceraldehyde-3-phosphate dehydrogeiiase (GAPDH) was performed using ] 0 [xM of the forward primer 5'-TCCCATCACCATCTTCCA-3' and 10 /xM o f t h e reverse primer 5'-GTCCACCACCCTGTTGCT-3', PCR conditions were 95°C for 2 min, followed by 35 cycles of 95°C for 30 seconds, 60°C for 30 seconds, and 72°C for 30 seconds with a final extension at 72°C for 5 min. Nested PCR [22] was used for the PTH/PTHiP-R. The outer forward primer 5'-GGGCACCAGGTGAAGTGGT-3' and the outer reverse primer 5'-GGTTGCTCTGACACCGACCC-3' were used for the first amplification, PCR conditions were 95°C for 2 min, followed by 30 cycles at 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds with a 3-second extension per cycle. Final extension was at 72°C for 5 min. For the second amplification, 5 /xl ofthe lirst-round amplification mixture was added to the same mixture as in the previous amplification but this time the inner primer concentrations were doubled to 20 /xM rather than 10 /xM, The inner forward primer was 5'-CGGGAGGTATTTGACCGCCTAG-3' and the inner reverse primer was 5'-CAGAATCCAGTAGTAGTTGG-3'. PCR conditions were the same as for the first amplification except cycle duration was 25, Samples of 20 /ixl of PCR reaction products were analyzed by electrophoresis on a 2% agarose gel. Bands were visualized by ethidium bromide staining. At some amplifications [a'^P] dCTP (Amersham) was added to reaction mixtures and gels were dried and exposed to Hyperfilm (Amersham), The PCR amplification products for PTH/PTHrP-R were further analyzed for authenticit)' by restriction enzyme digestion with Fok 1, Bsg I, and Drd I (New England Biolabs, Beverly, MA),
IUESULTS Stimulation of cAMP Production in Fibroblasts Treated with hPTH (1-34) The ability of a PTH N-terminal fragment to stimulate cAMP production was studied in cultured normal human keratinocytes and dermal fibroblasts. Treatment of fibroblasts with 1 0 " ' M of hPTH (1-34) caused a twofold increase in cAMP levels over the basal level (4,2 ± 0,1 pmol/dish versus 2,0 ± 0,05 pmol/dish). The data were significantly different and p was <0,005, In contrast, when keratinocytes were treated with 10 ' M of liPTH (1-34) no stimulation of cAMP formation was observed. Treated keratinocytes were 4,0 ± 0,2 pmol/dish and untreated keratinocytes were 3,9 ± 0,1 pmol/dish. There was no significant difference between these data points. Because hPTH (1-34) stimulated cAMP
VOL. 105. NO. 1 JULY
PTH/PTHrP RECEPTOR mRNA IN FIIJROHLASTS
135
A PTH/PTHrP-R 2.5 Kb
1 2
3
4
5
1 2
3 4 5
12
3 4 5 6 7
GAPDH
e
Figure 1. PTH/PTHrP-R expression in cultured human fibroblasts as sbown by Northern analysis and RT/nested PCR analysis. A) Northern blot analysis of total RNA prepared from SaOS-2 {lattes I attd J) and monse kidney cells {latte 4) and poly (A) ' RNA from keratinocytes {latte 2) and fibroblasts {latte 5). Tbe blots were h\'bridized witb tbe fuU-lengtb cDNA encoding tbe buman PTH/PTHrP receptor. B) Lane I was RT and GAPDH PCR amplification of total lOvIA from human keratinocytes cultured witbout a ^'V3 feeder layer. Lane 2 was an aliquot ot tbese same cells after RT and nested PCR amplification vvith PTH/ PTHrP-R primers. Lanes J through 6 are RT/uested PCR samples from buman keratinoeytes cultured witb 3T3 cells, human fibroblasts. luiman osteosarcoin.i cells, and .^T."^ cells, respectively. Lane 7 is tbe nested PCR product from tbe human PTH/PTHrP receptor cDNA.
GAPDH PTH/PTHrP-R
formation in cultured fibroblasts but not in cultured keratinocytes, we examined whether cultured human tibroblasts and keratinocytes possessed the same PTH/PTHrP receptors as those found in kidney and bone cells. Northern Blot Analysis Utilizing the recently cloned, fuUlengtb human PTH/1'THrP receptor cONA [15], Northern analysis was performed on poly (A) ' RNA from cultured human keratinocytes and fibroblii.sts. Jii all hybridization e.xperiment.s, approximately 1 0 /xg total RNA from mouse kidney cells and/or buman osteoblast-like osteosarcoma cells (SaOS-2) vvere used as positive controls. Lane 1 of Fig IA contained the positive control SaOS-2 cells. Lane 2 contained 5 iig of poly (A) ^' RNA from keratinocytes. No signal was visible in this sample. Lanes 3 and 4 contained SaOS-2 and mouse kidney total RNA, respectively. As can bo seen, a bigbly e.xpressed 2.5 kb PTH/PTHrP receptor niRNA was found in both SaOS-2 and mouse kidney cells. Also visible in botb samples was a less prominent transcript at 2.3 kb. Interestingly, three smaller transcripts were expressed in the SaOS-2 bone cells. Lane 5 contained 5 jug of poly (A) ' RNA isolated from fibroblasts. The major 2.5-kb transcript was clearly expressed as well as another at 2.3 kb. No smaller transcripts were observed even after prolonged exposure of tbe autoradiograpb. Etbidium bromide staining of the gel and rebybridization of tbe membrane witb an oligonuclcotide specific for buman GAPDH confirmed sample integrity and relative loading concentrations. In Fig IA, tbe GAPDH signals \vere visible in all ofthe lanes e.vcept for lane 4, the mouse kidney cells. Because tbe GAPDH oligonucleotide probe was specific for human cells, uo detectable signal was e.xpected. Apparent and expected were the intensities of tbe GAPDH signals in the poly (A) ' RNA from koratinocytes and fibroblasts as compared to tbe GAPDH signals from SaOS-2 total RNA. In a comparison hybridization experiment, five times tbe amount of poly (A) ' RNA from keratinocytes as from fibrobla.st.s .still did not .show PTH/PTHrP-R expression, even with prolonged exposure of tbe autoradiograpb (data not shown). In a total of five separate analyses, the PTH/ PTHrP receptor messages were consistently found in poly (A) ' IOvFA samples prepared from cultured dermal fibroblasts but not from cultured keratinocytes. RT/NESTED PCR Analysis Reverse transcriptase and nested PCR was used to analyze total RNA isolated from buman keratinocytes and fibroblasts as well as from monse immortalized fibroblasts (3T3s) and from human osteosarcoma cells (SaOS-2). In these experiments, tbe success of tbe cl3NA .synthesis from tlic R T
reaction was verified by amplifying an aliquot of tbe RT product witb primers for the GAPDH constitutive gene and confirming by gel electrophoresis the presence ofthe predicted PCR product. For PTH/PTHrP analyses, parallel first-amplification reactions were performed on cDNA aliquots and PCR produces were analyzed by gel electropboresis. Tbe first amplification of a bnman PTH/ PTHrP receptor cDNA produced tbe predicted 675-bp prodnct and tbe second amplification produced tbe predicted 36()-bp product. Nested PCR analyses of R T products from buman keratinocytes tbat were initially cultured on a 3T3 feeder layer, as well as from human fibroblasts, monse 3T3 fibroblasts, and human osteosarcoma cells demonstrated identical results to those from tbe bnman PTH/PTHrP receptor cDNA. In contrast, nested PCR analyses of RT products from buman keratinocytes tbat were cultured entirely without a 3T3 feeder layer (Fig IB) demonstrated neither a first nor second PCR amplification product. Lane I contained tbe RT/GAPDH-amplified PCR product from keratinocytes grown without a 3T3 feeder layer. Tbe predicted 762-bp product is clearly visible. Lane 2 contained an identical aliquot from tbis R T reaction tbat was amplified via tbe nested PCR protocol previously described. No PCR products were visible. Lane 3 contained an RT/nested PCR sample from keratinocytes tbat were cultured witb a .3T3 feeder layer. Lanes 4 through 6 were RT/nested PCR samples from luiman fibroblasts, luiman osteosarcoma cells, and mouse 3T3 fibroblasts, respectively. Lane 7 contained tbe nested PCR product from tbe buman PTH/PTHrP receptor cDNA. As can be seen, lanes 3 through 7 displayed identical second-amplification PCR products ofthe predicted 360 bp size. First-ainplificatioii PCR products of 675 bp were also identical for tbese samples and parallel amplifications with GAPOH primers produced the predicted product (data not shown). Restriction enzyme digests of the amplification products witb Fok I, Bsg 1, and l^rd 1 furtber confirmed tbeir autbenticity. DISCUSSION Although we |12,13J have demonstrated that specific PTH and PTHrJ^ N-tci-niinal fragments are antiproliferative agents in cnltnred luiman keratinocytes and in monse epidermis in vii'o. Nortliern analyses indicated the PTH/PTHrP receptor niRNA was not expressed in cultured keratinocytes. In contrast, cultured dermal fibroblasts expressed two PTH/PTHrP receptor transcripts with one major band at 2.5 kb and one minor band at 2.3 kb. The transcripts observed in fibroblasts were consistent witb tbose seen in hybridizations of RNA isolated from SaOS-2 cells and monse
136
HANAFIN i:r A
kidney cells, which are known to be PTH/PTHrP receptor positive [8]. The presence of two transcripts may be due to alternative transcription start sites, splicing, or the presence of closely related mRNAs. To increase detection sensitivity for the I'TH/PTHrP transcript, we utilized RT/ncsted PCR. The results confirmed the Northern analyses data and clearly showed that the PTH/PTHrP receptor message was present in human fibroblasts, huinan osteosarcoma cells and mouse 3T3 cells. Interestingly, tho enhanced sensitivity of RT/ncsted PCR revealed the PTH/PTHrP receptor in keratinocytes ctiltured t^n a 3T3 feeder layer. Because we found the PTH/PTHrP receptor in 3T3 cells, it was thought that R T / PCR amplification of RNA isolated from keratinocytes grown on a 3T3 feeder layer could generate receptor-positive amplification products. RT/nested PCR analyses of keratinocytes that were cultured without a 3T3 feeder layer demonstrated no PTH/PTHrP receptor, suggesting that the product observed was derived from the 3T3 cells of the feeder layer and not from the keratinocytes. The data are consistent with the additional ohservation that intracellular cAMP levels increased when fibroblasts were treated with PTH (1-34) (10 ^ M), but in identically treated keratinocytes there was no detectable cAMP response. Because there is uo detectable PTH/PTHrP receptor in human cultured keratinocytes, the data raise the possihility that PTHs and PTHrPs antiproliferative effects on keratinocytes may he mediated by PTH/PTHrP receptors located in adjacent fihrohlasts. Such a pathway is believed to be one way PTH increases osteoclastic bone resorption. Because mature osteoclasts do not have PTH receptors, PTH may first act on osteoblasts and/or stromal fibrohlasts to increase bone resorption [23|. Moreover, paracrine- and/or autocrine-mediated effects on skin cells hy PTHrP are also possihle. It is also possible that there are different receptors for PTH/ PTHrP in keratinocytes and fihroblasts. This idea is strengthened from the variability in the biologic potency of different PTH/ PTHrP analogs in numerous tissues [3]. Studies of the hiologic effects of several PTHrP fragments in different cell types have suggested that there may ho a PTHrP receptor distinct from the classic PTH/PTHrP receptor, which is coupled to adenylate cyclase through the G-protein [7]. For example, recombinant PTHrP (1-141) and PTHrP (1-108) were two to four times more potent than PTHrP (1-34) in the stinuilatioii of plasminogen activator activity in UMR cells |24[. Similarly, hPTHrP (1-34) and hPTHrP (1-108) were not active in inhibiting bone resorption, whereas the carboxy-terminal fragment hPTHrP (107-139) was a potent inhihitor of bone resorption [25|. Such data suggest that a distinct PTHrP receptor(s) may recognize specific fragments. Similarly, other studies suggest that particular PTHrP fragments may induce difFerent signal-transduction mechanisms. Specifically, Orloff rt al 1261 using squamous carcinoma cells demonstrated that hPTHrP (67-86)NH, or hPTHrP (1-141) caused a rapid and transient rise in intracellular calcium in a dose-dependent manner between 10 " M and 10 " M but there was no stimulation of cAMP formation detectable even with 10 '' M hPTHrP (67-86). In a separate study, Orloffcl al |27| stiggest human carcinoma cells and normal keratinocytes express mRNAs related to hut distinct from the human PTH receptor mRNA. If there are distinct receptors and presumably specific PTH and PTHrP fragment activation sites on these receptors, then this could explain no cAMP formation by PTH (1-34) and no PTH/PTHrP receptor expression in keratinocytes. Through cAMP determinations. Northern analyses, and R T / nested PCR analyses, this study localizes the human PTH/PTHrP receptor in fihroblasts but not in keratinocytes. Numerous questions remain, however, as to whether the hiologic activities of PTH and PTHrP in skin cells are mediated through paracrine and/or autocrine pathways or whether identical or distinct receptors exist. More research is necessary to understand the hiologic activation of endogenously occurring IHII and PTHrP iu skin cells and the potential therapeutic use for these hormones in regulating skin cell growth and differentiation.
THE JOURNAL OF INVESTIGATIVE DERMATOl.OGY
This tt'oik tvas sttftpotted itt pnti hy NIHgtants DI\4.i690 and MOIRR()0.'t3.i. IVe thattk Dai'id M. Jackson for his e.\eelleiii gtaphics assistance tt'iilt the ntaitiisctipi.
REFERENCES 1.
2.
3. 4.
5.
6.
7.
8.
9. 10. I I. 12.
13. 14. 15.
16. 17.
18.
I 9. 2(1.
21. 22.
23. 24.
Stewart AF, Horst R, Deftos LJ, Cadman EC, Lang R, Broad.is AE: Biochemical evaluation of patients with cancer-associated hyjiercah emia: e\idence for humoral and nonhumoral groups. ;V luiglj Med 3(J3:1 377-1383. 19811 Strewler GJ, Williams RD, Nissenson RA: I lunian renal carcinoma cells produce hyperialceniia in the nude mouse and a novel protein recognized by panuhyroid hormone receptors. / C/IH Ini'esi 7l:76')-774. 1983 OrloffJJ, Wu 1 L, Stewart Al": Parathyroid hormone-like proteins: biochemical responses and receptor iiKeiactions. Endocr Rev 1(1:476-495, 198'; Juppner II, Abou-Samra AB, Uueuo S, Giu WX, Potts JV Jr. Segre C;V: fhe parathytoid hormoue like peptide associated with humoral hypercillcemia of malignancy and parathyroid hormone bind to the same rece|)tor OTI the plasnia membrane of ROS 17/2.8 cells. / Biot Chetn 263:8557-8560, 1988 Nissenson RA. Karpf I), Hambino T. WiiierJ, Canga M, Nyiredy K, Arnaud CD: c;ovalent labeling of a higli-alFmity guanyl micleotide-sensitive parathyroid hormone receptor in canine renal cortex. Biochetnisiiy 26:1 874-1878, 1987 Fukayama S, Bosnia TJ, Goad DL, Voelkel EF, Tashjian AH Jr: Human parathyroid lK>riiioiie (PTI I)-reiated protein and luiman P'l'M: comparative biological activities ou huinan bone cells aud bone resorption. Ettdocritioloi^y 123:2841-2848, 1988 Juppner FL Abou-Samra AB, Freeman M, Kong XF, Schipani F., Richards J, Kolakowski LF, Flock J, Potts JT Jr, Kronenberg HM, Segre C:V: A G protein-linked receptor for paratbyroid bornione and parathyroid hormonerelated peptide. Science 254:1024-1026, 1991 Urena P, Kong XF, Abou-Samra AB, Juppner 11, Krouenberg HM, Potts JTJr, Segre GV: Paratbyroid bornione (P FI l)/PTH-related peptide receptor message ribomicleic acids are widely dislribuled in rat tissues. F.itdocrittolot;y 133:617-623, 1993 Goldring SR. Mabaffey JE, Rosenblatt M, Dayer JM, Potts |T Jr, Krane SM: Paratbyroid liormone iiiliibitors: comparison of biological acti\ity in bone and skin derived tissue. / ('litt Endocrinol Meiah 48:655-659, 1979 Mereiidini> JJ Jr, Insogna KL, Milstoue LM. Broadus AE, Stewart AF: CAiltured buman keratinocytes produce a parathyroid hormoue-like protein. Science 231:388-390, 1986 OrlolfJJ, Wu TL. Stewart AF: Parathyroid hormone-like proteins: biochemical responses and receptor interactions. lUtdocr Rev IO:476-4')5, 1989 Flolick MF, Nussbaum S, i'ersous KS: PI I l-like humoral hypercalcemia factor (I IMF) of malignancy may be an epidetmal differentiation factor: synthetic hllHF (I-34)NII, inliibits prolileiation and induces terminal differentiation of ctiltured btiman keratinocytes. / Bone Mineral Res 3:S2I4, 19S8 Holick MF, Ray S, C;hen VC. lian X, Persons K: A paradiyroid hormone antagonist stimtilates epidermal prolifeiatit>n and hair growth in mice. Proc .\'
VOL. 105, NO. I
PTH/PTHrP H.ECEPTOR mRNA IN FIBROBLASTS
JULY 1995
Fenton AJ, Kemp BE, Kent GN, MoseleyJM, Zheng MH. Rowe DJ, BrittoJM, Martin TJ, Nicholson GC: A carboxyl-teriuinal peptiiie from the parathyroid hormone-related protein inhibits bone resorption by osteoclasts. J:;iAimiw;i)V)' 1 29:1 762-1 768, 1991 Orloff JJ, Kats Y, Milnick M, Gasalla-1 lertaiz J, Isales CM: Evidence for a receptor on squamous carcinoma cell Hues wbich recognizes a mid-region
27.
137
fragment of parathyroid hornioue-related proteiu, PFIlrl' (67-K(i)NH,. / ;ii)/ii' Mittetal Res 8:S133, 1993 Orlolf JJ. Urena P, Scbipani E, Ribaudo AE, Milstoue LM. Pllilbrick WM, Abou-Samra AB, Segre CIV. Juppner 11:1 kiniaii squamous carciiu>iiia cells aud normal keratinocytes express mRNA, related to btit dislinct from litunan P i l l receptor mRNA. / i!..m-A/IMIT,.; K« 7:.i23(). 1992