BIOCHEMICAL
Vol. 182, No. 2, 1992 January 31, 1992
EVIDENCE
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS Pages 481-486
FOR THE EXPRESSION OF GROWTH RECEPTORS IN HUMAN PLACENTA
HORMONE
Francis FRANKENNE*, Eliane ALSAT**, Marie-Louise SCIPPO*, Ahmed IGOUT*, Georges HENNEN* and Daniele EVAIN-BRION** *Human Biochemistry, University of Liege, B23,4OOO Liege, Belgium **Laboratoire de Physiopathologie du Developpement, CNRS URA 1337. Ecole Normale Superieure, 75230 Paris CEDEX 05, France Received
November
14,
1991
ABSTRACT : Since human placenta produces a growth hormone variant, it seemed important to search for evidence of GH receptors in that organ. Evidence for the expression of the GH receptor (GHR) gene was obtained by northern blot analysis. In addition, GHR poly A+ RNA was detected in RNA from cultured trophoblastic cells, but not from placenta fibroblasts. There was a low but significant specific binding of pituitary GH-N and placental GH-V to placenta plasma membranes. Both variants apparently bound to the same receptor, which is present in the first trimester as well as in the term placenta. These results suggest that placental GH may have paracrine or autocrine 0 1992 Academic Press, Inc. functions in the placenta
Human placenta has been shown to produce its own growth hormone variant (1,2). That variant is specified by the hGH-V gene (3,4) in the syncytiotrophoblast (5). The GH-V protein is likely to share some activities of pituitary GH-N since it displays a higher potency for binding to rabbit liver GH receptors (1). Placental GH-V replaces pituitary GH-N in maternal serum after mid-pregnancy, but has not been detected in the fetal circulation (1). The hGH-N gene is intensively expressed in the pituitary, but not in normal placenta (2). Nevertheless, trophoblastic cells in culture are able to produce the GH-N protein in some circumstances (6). Some uneventful pregnancy cases associated with a deletion of the hGH-V gene in the feto-placental genome have been reported (7). Recently, we have obtained evidence that the hGH-N gene is expressed in such a placenta (to be published). Beside a potential endocrine role that placental GI-Is could exert in the maternal organism, 0006-291X/92
481
$1.50
Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
182,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
local activity is also possible, which would imply the existence of GH receptors in human placenta (1). Therefore, the primary objective of this work was to investigate the potential expression of GH receptor in the human placenta in order to obtain evidence for a potential autocrine or paracrine activity of GH in that organ.
MATERIALS
AND METHODS
Placental cell membranes were isolated according to the method of Marshall et al (8). Ret GH-V (9)(20 pg in 25 ~1 0.2 M sodium borate pH 9.5), ret GH-N (gift from Kabi Pharmacia, Stockholm, Sweden) and hPL (RIA, NIDDK) were labeled using 1 mCi Na [I2511 (Medgenix, Fleurus, Belgium) and 5 p.g Iodo-Gen (Pierce, Rockford, IL, USA) for 5 min, then gel filtered on a 0.9 X 60 cm Sephacryl S-200 column (Kabi Pharmacia, Stockholm, Sweden) in 0.05 M PBS pH 7.4 containing 0.1% BSA. Protein estimation was carried out by the method of Lowry et al (10). For binding analysis, plasma membranes (70-150 pg protein per tube) were incubated at 21-25°C for 16 h with [125I] ret GH-N or [1251’Jret GH-V (20 - 30.000 cpm per tube) in the presence and absence of excess unlabeled GH-N or GH-V (100 ng per tube) in 50 mM Tris-HCl pH 7, 0.1% BSA in a total volume of 0.5 ml. All incubations were performed in duplicate. The radioactivity of the bound hormone was counted after addition of 2 ml of cold buffer per tube, centrifugation and aspiration of the supematant. Human trophoblastic cells from first trimester and term placenta were isolated and cultured as described previously (11). Total RNA was extracted (12) and Poly A+ RNA isolation and slot-blotting were performed as previously reported (2). For northern analysis, 40 ng of term placental poly A+ RNA were loaded on to a 1% agarose glyoxal gel (13), then submitted to electrophoresis and transferred to nitrocellulose membrane (Millipore, Bedford, MA, USA). A 500 bp fragment of the human GH receptor cDNA encompassing a translated 70 bp portion corresponding to the 18 amino acids of the signal peptide and to the 6 fust amino acids of the mature protein, and a 430 bp non translated 5’ flanking region (14), was [32P] labeled by primed synthesis as described elsewhere (13). Hybridizations were performed as previously reported (2) and the blots were exposed (Hyperfilm p max, Amersham, Brussels, Belgium) at -7O“C for various periods of time. 482
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
RESULTS
Northern blot analysis of placental polyA+ RNA, using a [32P] GHR cDNA probe gave the same pattern as liver polyA+ RNA, with an intense band migrating at 4.4 Kbp (fig.1). By dot blot analysis, polyA+ RNA extracted from cultured trophoblasts gave positive, dose-dependent signals after incubation with [32P]GHR cDNA. (fig .2) The signals seen in the lane corresponding to the polyA+ RNA from fibroblast did not decrease in intensity with the doses, and then corresponded to a non-specific hybridization. The specific binding of [lzr]hPL,
[l25I]GH-N (pituitary GH) and [t25I]GH-V (placental
GH) to plasma membranes from first trimester and term human placenta have been assayed. Specific binding was recorded with [ 1251]GH-N as well as [ 1251JGH-V (Table I), but not with [ 12511hPL.
A
L
P
+ -4.4
2 Kb
01 Fieure.
02
C
0.5 0.25
22
0.125
P
0.062
if
0-03'
Autoradiograph of Northern blot analysis (1% agarose gel) of poly A+ RNA hybridized with a [32P] labeled human GH receptor cDNA probe. RNA samples were from human liver (40 pg, lane L) and term placenta (40 pg, lane PI. Size (kilobases) of the specifically RNA is shown by the arrow.
Fieure.
B
hybridized
the GH receptor ply(A)+
Slot blot of poly A+ RNA from human placental cells hybridized with hGH receptor cDNA probe. Decreasing amounts of poly A+ RNA from first trimester (A) and term (B) trophoblast cells cultured for five days and from first trimester cultured placental fibroblasts (C) were slot blotted onto a nitrocellulose sheet and hybridized with 32P-labeled GHR cDNA as described in Materials and Methods.
483
Vol.
182,
No.
2, 1992
m.
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Specific binding of [t25I]hGH-1 and [ 125I]hGH-2 to membrane fractions from human placenta
Placenta
Specific binding (%)
Weeks of pwzwv [*251]hGH-1
@ituitay)
[‘=I]hGH-2
@laced)
A
8
1
0.9
B
9
0.7
1.2
C
8.5
2.4
1.5
D
8
0.9
2.7
E
39
2.9
2.3
F
38
1.1
1.04
Parallel experiments performed using GH-N mass amounts with [ 125I]GH-V or GH-V mass amounts with [1251j GH-N for non specific binding assessment, gave essentially the same values, which were below 11% of the added radioactivity for [1251]GH-V and below 5.5% for [1251]GH-N. No difference was observed in the GH-V or GH-N specific binding with the length of pregnancy.
DISCUSSION The data demonstrate the presence of hGH R mRNA in human placenta. Furthermore, as shown by specific placental cell cultures, the GHR gene is expressed in the trophoblastic cells, but not in mesenchymal cells, like fibroblasts. In addition, we were able to detect the presence of membrane binding sites for both pituitary GH-N and placental GH-V in human placenta, while essentially no specific binding was found for hPL. The recorded binding values, ranging from 0.7 to 2.9 %, are close to the ones reported for human liver membranes, using pituitary GH-N (15). Although the recorded specific binding ranged between 10% and 20% above the nonspecific binding figures, their low absolute values prevented any meaningful application of Scatchard analysis. GH-N and GH-V are very likely to bind to a unique type of receptor, since they are able to completely displace each other. Membrane fractions prepared from first trimester or term placenta yielded bindings which were not significantly different. Nevertheless, further studies are needed to assess 484
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
potential differences in affinity or capacity of placental GH receptors at different stage of development. That GH-V is expressed together with GH receptors in the trophoblast strongly suggests the involvment of placental GH in an autocrine or paracrine mechanism of action. It is therefore likely to exert a regulatory activity on the production of IGF-I in the placenta (16). ACKNOWLEDGMENTS
We thank M. Horward for excellent technical assistance and A. Dukers for the preparation of the manuscript. We are particularly grateful to Dr. Michel Gossens for the generous gift of the GH receptor cDNA probe. The purified hPL preparation was provided by NIDDK. This work was supported by grants from “La Fondation de France”, from INSERM and Belgian “Communaute fraqaise”
(accord INSERM-CFB), from the Belgian “Region
Wallonne” (grant n”1532) and from the EEC (PL 91004). REFERENCES
1.
Frankenne F., Closset J., Gomez F., Scippo M.L., Smal J. and G. Hennen (1988) J Clin Endocrinol Metab, 66 : 1171-l 180, .
2.
Frankenne F., Rentier-Dentier Fr., Scippo M.L., Martial J. and G. Hennen (1987) J Clin Endocrinol Metab, 64 : 635-637.
3.
Frankenne F., Scippo M.L., Van Beeumen J., Igout A. and G. Hennen (1990) J Clin Endocrinol Metab, 71 : 15-18.
4.
Seeburg P.H. (1982) DNA 1: 239-249.
5.
Jara C.S., Salud A.T., Bryant-Greenwood G.D., Pirens G., Hennen G. and F. Frankenne (1989) J Clin Endocrinol Metab, 69 : 1069-1072.
6.
Evain-Brion D., Alsat E., Mirlesse V., Dodeur M., Scippo M.L., Hennen G. and F. Frankenne (1990) Hot-m. Res., 33 : 256-259.
7.
Wurzel J., Parks J.S., Herd J.E., P.V. Nielsen (1982) DNA 1 : 251-257.
8.
Marshall R.N., Underwood L.E. and S.J. Joina (1974) J Clin Endocrinol Metab, 39 : 283-292.
9.
Igout A., Scippo M.L., Frankenne F. and G. Hennen (1989) Nucleic Acids Res, 17, nolO, 3998.
10.
Lowry O.H., Rosebrough J., Farr A.C. and R.J. Randal (1951) J Biol Chem, 193 : 265-271.
11. Alsat E. , Mirlesse V., Fondacci C., Dodeur M. and D. Evain-Brian (1991) J Clin Endocrinol Metab, 73 : 288-295 485
Vol.
1992
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
No.
12.
Chomczynski P. (1990) Anal B&hem,
13.
Maniatis T., Frish E.F. and J. Sambrook (1982). “Molecular cloning A Laboratory manual, Cold Spring Harbor, New York.
14.
Amselem S., Duquesnoy P., Attree O., Novelli G., Bousnina S., Postel-Vinay, M.C. and M. Goossens (1989) N Engl J Med, 321 : 989-995.
15.
Hocquette J.F., Postel-Vinay MC., Kayser C., De Hemptinne B.and A. AmarCostesec (1989) Endocrinology 125 : 2167-2174.
16.
2,
BIOCHEMICAL
182,
162 : 156-159.
Mills N.C., D’Ercole A.J., Underwood L.E. and J. Ilan (1986) Mol Biol Reprod. 11 : 231-236.
486
Vol. 182, No. 2, 1992 January 31, 1992
EVIDENCE
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS Pages 481-486
FOR THE EXPRESSION OF GROWTH RECEPTORS IN HUMAN PLACENTA
HORMONE
Francis FRANKENNE*, Eliane ALSAT**, Marie-Louise SCIPPO*, Ahmed IGOUT*, Georges HENNEN* and Daniele EVAIN-BRION** *Human Biochemistry, University of Liege, B23,4OOO Liege, Belgium **Laboratoire de Physiopathologie du Developpement, CNRS URA 1337. Ecole Normale Superieure, 75230 Paris CEDEX 05, France Received
November
14,
1991
ABSTRACT : Since human placenta produces a growth hormone variant, it seemed important to search for evidence of GH receptors in that organ. Evidence for the expression of the GH receptor (GHR) gene was obtained by northern blot analysis. In addition, GHR poly A+ RNA was detected in RNA from cultured trophoblastic cells, but not from placenta fibroblasts. There was a low but significant specific binding of pituitary GH-N and placental GH-V to placenta plasma membranes. Both variants apparently bound to the same receptor, which is present in the first trimester as well as in the term placenta. These results suggest that placental GH may have paracrine or autocrine 0 1992 Academic Press, Inc. functions in the placenta
Human placenta has been shown to produce its own growth hormone variant (1,2). That variant is specified by the hGH-V gene (3,4) in the syncytiotrophoblast (5). The GH-V protein is likely to share some activities of pituitary GH-N since it displays a higher potency for binding to rabbit liver GH receptors (1). Placental GH-V replaces pituitary GH-N in maternal serum after mid-pregnancy, but has not been detected in the fetal circulation (1). The hGH-N gene is intensively expressed in the pituitary, but not in normal placenta (2). Nevertheless, trophoblastic cells in culture are able to produce the GH-N protein in some circumstances (6). Some uneventful pregnancy cases associated with a deletion of the hGH-V gene in the feto-placental genome have been reported (7). Recently, we have obtained evidence that the hGH-N gene is expressed in such a placenta (to be published). Beside a potential endocrine role that placental GI-Is could exert in the maternal organism, 0006-291X/92
481
$1.50
Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
182,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
local activity is also possible, which would imply the existence of GH receptors in human placenta (1). Therefore, the primary objective of this work was to investigate the potential expression of GH receptor in the human placenta in order to obtain evidence for a potential autocrine or paracrine activity of GH in that organ.
MATERIALS
AND METHODS
Placental cell membranes were isolated according to the method of Marshall et al (8). Ret GH-V (9)(20 pg in 25 ~1 0.2 M sodium borate pH 9.5), ret GH-N (gift from Kabi Pharmacia, Stockholm, Sweden) and hPL (RIA, NIDDK) were labeled using 1 mCi Na [I2511 (Medgenix, Fleurus, Belgium) and 5 p.g Iodo-Gen (Pierce, Rockford, IL, USA) for 5 min, then gel filtered on a 0.9 X 60 cm Sephacryl S-200 column (Kabi Pharmacia, Stockholm, Sweden) in 0.05 M PBS pH 7.4 containing 0.1% BSA. Protein estimation was carried out by the method of Lowry et al (10). For binding analysis, plasma membranes (70-150 pg protein per tube) were incubated at 21-25°C for 16 h with [125I] ret GH-N or [1251’Jret GH-V (20 - 30.000 cpm per tube) in the presence and absence of excess unlabeled GH-N or GH-V (100 ng per tube) in 50 mM Tris-HCl pH 7, 0.1% BSA in a total volume of 0.5 ml. All incubations were performed in duplicate. The radioactivity of the bound hormone was counted after addition of 2 ml of cold buffer per tube, centrifugation and aspiration of the supematant. Human trophoblastic cells from first trimester and term placenta were isolated and cultured as described previously (11). Total RNA was extracted (12) and Poly A+ RNA isolation and slot-blotting were performed as previously reported (2). For northern analysis, 40 ng of term placental poly A+ RNA were loaded on to a 1% agarose glyoxal gel (13), then submitted to electrophoresis and transferred to nitrocellulose membrane (Millipore, Bedford, MA, USA). A 500 bp fragment of the human GH receptor cDNA encompassing a translated 70 bp portion corresponding to the 18 amino acids of the signal peptide and to the 6 fust amino acids of the mature protein, and a 430 bp non translated 5’ flanking region (14), was [32P] labeled by primed synthesis as described elsewhere (13). Hybridizations were performed as previously reported (2) and the blots were exposed (Hyperfilm p max, Amersham, Brussels, Belgium) at -7O“C for various periods of time. 482
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
RESULTS
Northern blot analysis of placental polyA+ RNA, using a [32P] GHR cDNA probe gave the same pattern as liver polyA+ RNA, with an intense band migrating at 4.4 Kbp (fig.1). By dot blot analysis, polyA+ RNA extracted from cultured trophoblasts gave positive, dose-dependent signals after incubation with [32P]GHR cDNA. (fig .2) The signals seen in the lane corresponding to the polyA+ RNA from fibroblast did not decrease in intensity with the doses, and then corresponded to a non-specific hybridization. The specific binding of [lzr]hPL,
[l25I]GH-N (pituitary GH) and [t25I]GH-V (placental
GH) to plasma membranes from first trimester and term human placenta have been assayed. Specific binding was recorded with [ 1251]GH-N as well as [ 1251JGH-V (Table I), but not with [ 12511hPL.
A
L
P
+ -4.4
2 Kb
01 Fieure.
02
C
0.5 0.25
22
0.125
P
0.062
if
0-03'
Autoradiograph of Northern blot analysis (1% agarose gel) of poly A+ RNA hybridized with a [32P] labeled human GH receptor cDNA probe. RNA samples were from human liver (40 pg, lane L) and term placenta (40 pg, lane PI. Size (kilobases) of the specifically RNA is shown by the arrow.
Fieure.
B
hybridized
the GH receptor ply(A)+
Slot blot of poly A+ RNA from human placental cells hybridized with hGH receptor cDNA probe. Decreasing amounts of poly A+ RNA from first trimester (A) and term (B) trophoblast cells cultured for five days and from first trimester cultured placental fibroblasts (C) were slot blotted onto a nitrocellulose sheet and hybridized with 32P-labeled GHR cDNA as described in Materials and Methods.
483
Vol.
182,
No.
2, 1992
m.
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Specific binding of [t25I]hGH-1 and [ 125I]hGH-2 to membrane fractions from human placenta
Placenta
Specific binding (%)
Weeks of pwzwv [*251]hGH-1
@ituitay)
[‘=I]hGH-2
@laced)
A
8
1
0.9
B
9
0.7
1.2
C
8.5
2.4
1.5
D
8
0.9
2.7
E
39
2.9
2.3
F
38
1.1
1.04
Parallel experiments performed using GH-N mass amounts with [ 125I]GH-V or GH-V mass amounts with [1251j GH-N for non specific binding assessment, gave essentially the same values, which were below 11% of the added radioactivity for [1251]GH-V and below 5.5% for [1251]GH-N. No difference was observed in the GH-V or GH-N specific binding with the length of pregnancy.
DISCUSSION The data demonstrate the presence of hGH R mRNA in human placenta. Furthermore, as shown by specific placental cell cultures, the GHR gene is expressed in the trophoblastic cells, but not in mesenchymal cells, like fibroblasts. In addition, we were able to detect the presence of membrane binding sites for both pituitary GH-N and placental GH-V in human placenta, while essentially no specific binding was found for hPL. The recorded binding values, ranging from 0.7 to 2.9 %, are close to the ones reported for human liver membranes, using pituitary GH-N (15). Although the recorded specific binding ranged between 10% and 20% above the nonspecific binding figures, their low absolute values prevented any meaningful application of Scatchard analysis. GH-N and GH-V are very likely to bind to a unique type of receptor, since they are able to completely displace each other. Membrane fractions prepared from first trimester or term placenta yielded bindings which were not significantly different. Nevertheless, further studies are needed to assess 484
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
potential differences in affinity or capacity of placental GH receptors at different stage of development. That GH-V is expressed together with GH receptors in the trophoblast strongly suggests the involvment of placental GH in an autocrine or paracrine mechanism of action. It is therefore likely to exert a regulatory activity on the production of IGF-I in the placenta (16). ACKNOWLEDGMENTS
We thank M. Horward for excellent technical assistance and A. Dukers for the preparation of the manuscript. We are particularly grateful to Dr. Michel Gossens for the generous gift of the GH receptor cDNA probe. The purified hPL preparation was provided by NIDDK. This work was supported by grants from “La Fondation de France”, from INSERM and Belgian “Communaute fraqaise”
(accord INSERM-CFB), from the Belgian “Region
Wallonne” (grant n”1532) and from the EEC (PL 91004). REFERENCES
1.
Frankenne F., Closset J., Gomez F., Scippo M.L., Smal J. and G. Hennen (1988) J Clin Endocrinol Metab, 66 : 1171-l 180, .
2.
Frankenne F., Rentier-Dentier Fr., Scippo M.L., Martial J. and G. Hennen (1987) J Clin Endocrinol Metab, 64 : 635-637.
3.
Frankenne F., Scippo M.L., Van Beeumen J., Igout A. and G. Hennen (1990) J Clin Endocrinol Metab, 71 : 15-18.
4.
Seeburg P.H. (1982) DNA 1: 239-249.
5.
Jara C.S., Salud A.T., Bryant-Greenwood G.D., Pirens G., Hennen G. and F. Frankenne (1989) J Clin Endocrinol Metab, 69 : 1069-1072.
6.
Evain-Brion D., Alsat E., Mirlesse V., Dodeur M., Scippo M.L., Hennen G. and F. Frankenne (1990) Hot-m. Res., 33 : 256-259.
7.
Wurzel J., Parks J.S., Herd J.E., P.V. Nielsen (1982) DNA 1 : 251-257.
8.
Marshall R.N., Underwood L.E. and S.J. Joina (1974) J Clin Endocrinol Metab, 39 : 283-292.
9.
Igout A., Scippo M.L., Frankenne F. and G. Hennen (1989) Nucleic Acids Res, 17, nolO, 3998.
10.
Lowry O.H., Rosebrough J., Farr A.C. and R.J. Randal (1951) J Biol Chem, 193 : 265-271.
11. Alsat E. , Mirlesse V., Fondacci C., Dodeur M. and D. Evain-Brian (1991) J Clin Endocrinol Metab, 73 : 288-295 485
Vol.
1992
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
No.
12.
Chomczynski P. (1990) Anal B&hem,
13.
Maniatis T., Frish E.F. and J. Sambrook (1982). “Molecular cloning A Laboratory manual, Cold Spring Harbor, New York.
14.
Amselem S., Duquesnoy P., Attree O., Novelli G., Bousnina S., Postel-Vinay, M.C. and M. Goossens (1989) N Engl J Med, 321 : 989-995.
15.
Hocquette J.F., Postel-Vinay MC., Kayser C., De Hemptinne B.and A. AmarCostesec (1989) Endocrinology 125 : 2167-2174.
16.
2,
BIOCHEMICAL
182,
162 : 156-159.
Mills N.C., D’Ercole A.J., Underwood L.E. and J. Ilan (1986) Mol Biol Reprod. 11 : 231-236.
486
