0021-972x/92/7401-0211$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society
Vol. 74, No. 1 Printed in U.S.A
Trophoblast-Derived Transforming Growth Factor-& Suppresses Cytokine-Induced, but not GonadotropinReleasing Hormone-Induced, Release of Human Chorionic Gonadotropin by Normal Human Trophoblasts* NOBORU MATSUZAKI, YING LI, KAZUO MASUHIRO, TOUSHUN JO, KOUICHIROU SHIMOYA, TAKESHI TANIGUCHI, FUMITAKA SAJI, AND OSAMU TANIZAWA Department
of
Obstetrics and Gynecology,
Osaka University
Medical
ABSTRACT. Using a transforming growth factor-p (TGFp)sensitive cell line, MvlLu (or CCL 64), we demonstrated that trophoblasts predominantly produced a latent form of TGF& After converting latent TGFfi to active TGFp in oitro by acid (pH 2.5), alkali (pH lO.O), or heat (90 C; 10 min) treatment, addition of rabbit anti-TGFB, antiserum resulted in the elimination of TGFfl activity, thus suggesting that trophoblasts produced at least a certain amount of latent TGFP,. To investigate the role of TGF& in placental hormonogenesis, we first studied the effect of recombinant (r) TGFp, on the production of interleukin-6 (IL-6) and hCG by trophoblasts. rTGF/3, exerted no inhibitory activity on IL-6 and hCG production. The effect of rTGF& on cytokine-induced IL-6 and hCG release was then
School, Osaka 553, Japan
examined. While rTGF& failed to inhibit basal hCG secretion, it did inhibit recombinant tumor necrosis factor-a (rTNFa)induced IL-6 release as well as rTNFu- and rIL-g-induced hCG release in a dose-dependent manner. However, rIL-la-induced IL-6 and hCG release was remarkably sensitive to rTGF&mediated suppression. In contrast, GnRH-induced hCG release, the response of which is independent of the IL-6 and IL-6 receptor system in trophoblasts, was completely resistant to rTGF&. Thus, trophoblast-derived TGF& is an important regulatory molecule of cytokine-dependent, but not cytokine-independent, hCG release, possibly by converting latent TGFO to active TGF@ at the local site of trophoblasts. (J Clin Endocrinol Metab
T
74:
211-216,
1992)
using anti-TGF@, antiserum, we examined whether trophoblasts produce TGFP, molecules. Trophoblast-derived interleukin-1 (IL-l) and tumor necrosis factor-a (TNFa) synergistically induce IL-6 release, and the released IL-6 activates the IL-6 receptor (IL-6-R)-mediated signal transduction pathway to release hCG by normal trophoblasts; this is a cytokine-mediated regulatory cascade, resulting in hCG release (5, 6, 7). GnRHmediated hCG release (8), on the other hand, is independent of the cytokine-mediated hCG release mechanism (5). To investigate the TGF@-mediated regulatory mechanism in placental hormone function, we studied the effect of TGF/3 on cytokine- and GnRH-mediated hCG release as well as its effect on the production of trophoblast-derived IL-6 and hCG.
RANSFORMING growth factor-p (TGFB) is a multifunctional polypeptide with a mol wt of 25 kDa, composed of two disulfide-linked 12.SkDa subunits (1). TGF@ was initially defined as being related to the induction of malignant cells in viva (2). It contributes to the regulation of cell growth and differentiation because of its unique regulatory activities (2). There is increasing evidence showing unique TGFP-mediated effects on endocrinological systems (3). Little information, however, is available regarding TGFP-mediated modification of normal human placental functions. Using a TGFP-sensitive cell line, MvlLu (4), we first addressed the question of whether trophoblasts produce an active or latent form of TGFP-like molecule. Then, Received March 11, 1991. Address all correspondence and requests for reprints to: N. Matsuzaki, M.D., Ph. D., Department of Obstetrics and Gynecology, Osaka Universitv Medical School. l-1-50, Fukushima, Fukushima-ku, Osaka 553, Japah. * This work was supported by Grants-in-Aid for Scientific Research (no. 02454382.63480367. and 01570930) from the Ministry of Educaiion, Science, and Culture, Tokyo, Japan.
Materials
and Methods
Reagents Purified Escherichia coli-derived recombinant kindly provided by Dainippon Pharmaceutical
(r) TNFu! was Co. (Osaka,
X11
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
212
MATSUZAKI
ET AL.
JCE&M.1392 Vol74. No I
Japan). rIL-lcr and rIL-lb were gifts from Otsuka Pharmaceutical Co. Ltd. (Tokushima, Japan). rIL-6 was a kind gift from Dr. T. Kishimoto (Osaka University, Osaka, Japan). rTGF& and rabbit anti-TGF& immunoglobulin G (polyclonal antiTGF& neutralizing antiserum) were purchased from CosmoBio Co. (Tokyo, Japan). A GnRH agonist (GnRHa), [DLeua,Des-Gly’o-NH2]GnRH-N-ethylamide acetate (leuprolide acetate), was kindly provided by Takeda Pharmaceutical Co. (Osaka, Japan).
IL-6 and hCG release, rTGF& was added to the trophoblast culture, which was simultaneously stimulated with 10.0 Kg/L rIL-6, 20 pg/L rIL-ln, 200 pg/L rTNFq or various concentrations of GnRHa (leuprolide acetate). One hundred and fifty minutes after the stimulation, the culture supernatants were collected and assayed for hCG and IL-6. Each experiment was repeated at least three times, with identical results.
Preparation of purified trophoblasts
For measurement of the IL-6 titer, 1 X 10’ cells of an IL-6dependent murine hybridoma, MH60.BSF-2, were cultured in 96-well flat bottom microplates (Corning) with either test samples or standard rIL-6 for 48 h at 37 C in an incubator. The cultured cells were pulsed with 1 &i [3H]thymidine (15.7 Ci/ mM; New England Nuclear) for the last 6 h. The results were expressed as the mean t- SEM micrograms of IL-6 per L produced by the test cells, as described previously (8).
Fresh placentas were obtained from legal abortions carried out on social grounds at 7-9 weeks gestation. The placenta was thoroughly washed, and the villous tissue was separated from the connective tissue and minced on ice. To prepare single cell suspensions of trophoblasts, villous tissue (2-3 g) was first incubated with type III collagenase (Funakoshi, Tokyo, Japan) and DNase (Sigma Chemical Co., St. Louis, MO) at 37 C for 30 min in a shaking water bath. The enzyme-treated cells were filtered through a mesh and centrifuged at 800 x g for 6 min. To obtain purified trophoblasts, enzyme-treated placental cells were centrifuged twice on a 60% Percoll (Pharmacia Fine Chemicals, Piscataway, NJ) gradient for 15 min at 800 x g. The sedimented cells were then washed twice with culture medium to remove the collagenase. Cell viability, estimated by the trypan blue exclusion test, was more than 90%. Dispersed trophoblasts were adjusted to 2 x lo8 cells/L in culture medium and plated onto collagen-IA-coated dishes (Nitta Gelatin, Osaka, Japan). Cells (2 x 1OAcells/L) were cultured for 2 days at 37 C in 24-well flat bottom microplates (Corning, Iwaki Glass, Tokyo, Japan) (5-7). Bioassay for TGF/3 TGF/3 in the culture supernatants of purified trophoblasts was quantitated by comparing the level of inhibition of MvlLu mink lung cell (CCL 64) proliferation in supernatant dilution curves with that of rTGF@, (4). MvlLu (1 x lo7 cells/L) was seeded and cultured for 22 h. For the last 4 h, 1 &X/well [“HI thymidine (New England Nuclear, Boston, MA) was added. To detect the active form of TGFp, the fetal calf serum-free culture medium of trophoblasts was collected after 24-h culture. To activate the latent form of TGFfl to its active form, the medium was dialyzed against either acidic (pH 2.5) or alkaline solution (pH 10.0) overnight and then redialyzed with phosphate-buffered saline, pH 7.4, for 24 h or heated at 90 C for 10 min and cooled on ice. For TGF@ neutralization studies, rabbit antiTGFj3, antiserum was used. The neutralization rate (percentage) was calculated by comparing the count of [“Hlthymidine uptake between (antiserum-added group - background) and (antiserum-free group - background). rTGF& addition to trophoblast culture To observe the effect of TGFP, on IL-6 and hCG production, rTGF& was added at the initiation of (day 0) or 24 h after (day 1) culture of dispersed trophoblasts. Forty-eight hours after the start of culture, the culture supernatants were removed and assayed for IL-6 and hCG. To examine the effect of TGFp, on
Bioassay for IL-6
Hormonal assaysand statistics hCG released into the medium was measured by an enzyme assay (Mochida Co., Tokyo, Japan), using a monoclonal antibody specific for the P-subunit of hCG (5,9). Statistical analysis of the results for significance was performed by Duncan’s test after a one-way analysis of variance.
Results Production
of TGF/3 by normal trophoblasts
Figure 1A shows that trophoblasts produced predominantly a latent form, rather than an active form, of a TGFB-like molecule (P < 0.0005), because TGFP-like activity was detected predominantly in culture supernatants that had been treated with acid, alkali, or heat. Almost negligible TGFp activity was detected in untreated culture supernatants of trophoblasts, regardless of the presence or absence of fetal calf serum. As shown in Fig. lB, TGFB-mediated activity in the culture supernatants that had been treated with acid was neutralized in a dose-dependent manner by the addition of the rabbit anti-TGFP, antiserum. Effect of TGF@, on hCG and IL-6 production The results in Table 1 show no significantly inhibitory or augmenting effect of rTGF& on the titer of trophoblast-derived IL-6 or hCG. Measurement of IL-6 by a bioassay was not affected by the presence of rTGFPl (data not shown). Suppression
of rIL-6-induced
hCG release by rTGFDl
Table 2 shows that simple addition of rTGF& to cultured trophoblasts alone failed to augment or suppress hCG release, while addition of rIL-6 induced hCG release. As shown in Fig. 2, rTGFPl addition resulted in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
TGF&-MEDIATED
REGULATION
OF hCG
RELEASE
(B)
(A) 500. FIG. 1. Production of TGF@ by purified trophoblasts (A) and neutralization of trophoblast-derived TGFp activity by rabbit anti-TGF& antiserum (B). A, TGFP activity was assayed using a TGFP-sensitive cell line, MvlLu. The values represent the mean f SEM of triplicate determinations. ****, P < 0.0001. B, After conversion to active TGF@ by acid treatment, the culture supernatant (0.1 rg/L TGFP activity) was mixed with the antiserum. The neutralization rate (percentage) was calculated according to the formula described in Materials and
213
r-
-*t**
400,
~ 3oo 1 $ ‘, G ’ 2oo
I
100
Methods.
0 TABLE 1. Effect of rTGF& tion of IL-6 and hCG Pretreatment rTGI% (10 wdJ rTGFh (10 rg/L) Control medium
addition to trophoblasts Day of addition
IL-6 (&L) 5.0 + 0.8 4.5 + 0.7 5.6 f 1.2
0
1 0
on their produc-
hCG (IU/L) 1270 + 457 1007 +- 384 1227 * 293
1 : *-I--
rTGFfil (10 rg/L) was added either at initiation of (day 0) or after 24 h (day 1) of the trophoblast culture. After 48 h (day 2), the culture supernatants were collected for titration of IL-6 and hCG. The values represent the mean + SEM of triplicates. TABLE 2. Effect of rTGF& on hCG release from trophoblasts Stimulants Control medium rIL-6 rTGF&
Cont. (fig/L) 10 1.0 10
hCG release (W/L) 117 f 14 188 + 25b 120 k 10 129 + 14’
Cultured trophoblasts were incubated with various stimulants for 150 min. The culture supernatants were removed to determine hCG titer. The values represent the mean f SEM of triplicate determination. b P < 0.01 vs. control. ‘P = NS vs. control.
the suppression of rIL-6-induced hCG release in a dosedependent fashion at doses between 0.01-10 pg/L. This suppression was not due to impairment of cell viability, because rTGF& addition did not alter the cell viability, as determined by the trypan blue dye exclusion test (data not shown). Suppression of rIL-laIL-6 release
and rTNFwinduced
hCG and
As shown in Fig. 3A, the rTNFa-mediated hCG response was suppressed by rTGF& in a dose-dependent
I L-6 (IO!%/LI
(-)
(+)
AGF-/3 (U?./L)
i-!
i-)
(+)
(+I
(+)
(+)
(+I
0.001
0.01
0 1
1.0
10
FIN:. 2. rTGFP, suppresses rIL-6-mediated hCG release. Purified trophoblasts stimulated with rIL-6 (10 kg/L) were cultured in the presence or absence of rTGF/% at various concentrations. The values represent the mean -C SEM of triplicate determinations. *, P < 0.05; **, I’ < 0.01.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
214
MATSUZAKI
ET AZ,
JCE ti M. 1992 Voli4.Nol
(A
FIG. 3. rTGFp, suppresses rTNFcu-mediated and rIL-la-mediated hCG and IL-6 release. Purified trophoblasts stimulated with either rTNFa (200 pg/L; A and B) or rIL-la (20 rg/L; C and D) were cultured in the presence or absence of rTGFj3 at various concentrations. The values represent the mean f SEM of hCG (A and C) and IL-6 (B and D) in triplicate determinations. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
II
__,-. )EFp 1:: l”,iLi
c+, c-i
,+b 001
(Cl
T
1 manner. Figure 3B demonstrates that rTGF& also inhibited rTNFa-induced IL-6 release. The inhibition ratio of rTNFa-induced IL-6 release by rTGF& was similar to that of either rTNFa- or rIL-6-induced hCG release by rTGF& (Figs. 3, A and B, and 2). Figure 3C, in contrast, shows that rTGF/3, completely ‘blocked rIL-la-induced hCG release even at a concentration of 0.001 pg/L (P < 0.01). Moreover, as demonstrated in Fig. 3D, the effect of rTGF/3 on rIL-la-induced IL-6 release was extremely sensitive compared with its effect on rTNFa-induced IL6 release even at a concentration of 0.001 pg/L rTGF& (P < 0.001; Fig. 3B). Similar rTGFP1-mediated suppression was observed with trophoblasts stimulated with rIL18 (data not shown). Similar TGF&-mediated suppression was observed after rTGFPl addition even after rILla- or rTNFa-mediated stimulation of trophoblasts (data not shown).
--A I
Effect of rTGF&
on GnRH-induced
*em-
hCG release
We tested whether TGFP, was capable of suppressing GnRH-mediated hCG release. As illustrated in Fig. 4, rTGF& failed to significantly suppress GnRH-mediated hCG release (P = NS, GnRHa-stimulated us. GnRHaplus rTGF/&-stimulated groups), while it completely suppressed rIL-g-mediated hCG release (P < 0.01). Discussion Production of TGFP by the human placenta has been reported. Assoian et al. (10) isolated TGFp from the platelets in the placenta, while Madisen et al. (11) purified TGFP from human placental tissue. Our present study revealed that trophoblasts in the placenta predominantly produced a latent form of a TGFP-like molecule, the activity of which was significantly neutralized by
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
TGF&-MEDIATED Stimulants
rTGF-fl,
hCG
Release
I Control
NIL-6
Medwm
REGULATION
(IU/L! 1
-
(10 /x/L) I- +
GI+?H&oo~~cUL)
I+
I- +
GnRHa
(10 nmo~/~)
GnRHa
( 1 ” ml/L)
0
I 50
100
150
200
250
FIG. 4. rTGF& fails to suppress GnRHa-mediated hCG release. Trophohlasts were stimulated with either rIL-6 (10 pg/L) or various concentrations of GnRHa in the presence or absence of rTGF fil (10 pg/L). The values represent the mean f SEM of triplicate determinations. *, PC 0.05; **, P < 0.01.
rabbit anti-TGF& antiserum. Trophoblasts, therefore, produce at least a certain amount of TGFP,, although it is not clear whether trophoblasts produce other types of TGFp such as TGF& (12), -iij3 (13), and -p4 (14). A latent form of TGFP has been reported to be incapable of exerting various biological activities associated with TGF@ because of its failure to bind to native TGFfiR (15). Latent TGFp can be converted to its active form in uitro by various methods, including changes in pH (3, 6), thermal activation (3), treatment with proteases (16) as well as glycosidases (17), and cell to cell contact (18). Such conversion of TGFp has been reported to occur in activated immunocompetent cells that generate local pH shift, proteases, and sialidases to produce active TGFP (19). It was described in a recent study that the addition of anti-TGFP, monoclonal antibody resulted in augmentation of lymphocyte proliferation in culture supernatants in which only endogenously derived latent TGF/3 activity could be detected (20). This suggests that even endogenously produced TGFP, i.e. latent TGFP, might exert regulatory activities after being converted to active TGF@ at a local site. Oberbauer et al. (21) reported that TGFp significantly suppressed epidermal growth factor-stimulated hCG production in Jar choriocarcinoma cells. However, few reports are available that show TGFP-mediated regulatory activities on normal human placental hormone release and production. Although simple addition of TGF@ showed no inhibition or augmentation of the basal hCG secretion, our present study revealed the unique regulatory activities of TGFP on the cytokine-mediated regulatory network; both trophoblast-derived IL-l and TNFa synergistically stimulate hCG release by inducing tro-
OF hCG
RELEASE
215
phoblast-derived IL-6 release and activating IL-6-R-mediated signal transduction pathways (7). rTGF& markedly suppressed rIL-la-, rTNFLu-, or IL-6-induced hCG release. rTGF0, also inhibited rIL-lor- or rTNFa-induced IL-6 release by trophoblasts, which response shows differential sensit,ivity to the concentration of rTGF& added; compared with the magnitude of its effect on rTNFa-induced IL-6 and hCG release, rTGF& showed marked suppression of rIL-la-induced release. In contrast, GnRH, which triggers hCG release independently of the IL-6 and IL-6-R system in trophoblasts (5), shows complete resistance to TGFPI-mediated inhibitory activity. TGFB,, therefore, showed preferential suppressive action on cytokine-mediated, but not GnRH-mediated, hCG release by regulating the amount of IL-6 released and the IL-6-induced hCG release. Rather, GnRH-induced hCG secretion has been reported to be up-regulated by activin and down-regulated by inhibin, two other cytokine members of the TGFp family (22). hCG release, thus, might be down-regulated by cytokines in the TGFp family, TGF/& and inhibin. The cellular and molecular mechanisms of TGFpmediated suppression have been extensively investigated in other cellular systems, such as hepatocytes (23). Dubois et al. (24) defined TGF@ as a potent inhibitor of IL1-R expression, which results in TGFP-mediated suppression of IL-l-induced responses. However, such mechanisms might not be operative in our system, because rTGFpl was capable of suppressing rIL-la-induced IL-6 and hCG release even after IL-l and IL-l-R interaction, since TGF& was capable of suppressing rIL-lainduced IL-6 and hCG release even when it was added after rIL-lu stimulation of trophoblasts. The finding that TGFP, inhibited cytokine-mediated IL-6 and hCG release even after addition of the cytokines also excluded the possibility that TGF/?, might competitively bind to IL-l-R (6), TNFn-R (25), or IL-6-R (5) on trophoblasts, thereby serving as a cytokine inhibitor itself. In contrast to the TGFB-mediated regulatory activity on IL-6 and hCG release, TGFp exerted no inhibitory or augmenting effect on IL-6 or hCG production. These findings together with TGFP’s effect on GnRH-induced hCG release exclude the possibility that rTGF& is toxic to trophoblasts and thereby reduces IL-6 and hCG release, indicating that TGF& produced by trophoblasts, platelets, and monocytes in the placenta might act as a physiological regulator in the cytokine-dependent hCG release mechanism in an autocrine or paracrine fashion. It seems possible to speculate that such a TGFP-mediated regulatory function might be supported through TGF@ receptors and postreceptor-mediated signal transduction pathways. In hematopoietic cells, three distinct receptor molecules for TGFP, i.e. with molecular sizes of 65, 85, and 260 kDa, have been demonstrated (26).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
216
MATSUZAKI
Trophoblasts seem to possess a TGFP- and TGF&Rspecific signal transduction pathway common to immunocompetent cells, hematopoietic cells, hepatocytes, and synovial cells. Further experiments will be required to demonstrate the presence of TGFP-R on trophoblasts and characterize the nature and mechanism of postTGF/3-R intracellular signal transduction pathways that participate in the regulation of hCG release. References 1. Massague J. The TGF-j3 family of growth, differentiation. Cell. 1987;49:431-8. 2. Sporn MB, Roberts AB, Wakefield LM, Assoian RK. Transforming-8: biological function and chemical structure. Science. 1986;233:532-4. 3. Kim I-U, Shomberg DW. The production of transforming growth factor+3 activity by rat granulosa cell cultures. Endocrinology. 1989;124:1345-51. 4. Holly RW, Armour R, Baldwin JH, Greenfield S. Activity of a kidney epithelial cell growth on lung and mammary cells. Cell Biol Int Rep. 1983;7:141-7. 5. Nishino E, Matsuzaki N, Masuhiro K, et al. Trophoblast-derived IL-6 regulates human chorionic gonadotropin release through IL6 receptor on human trophoblasts. J Clin Endocrinol Metab. 1990;71:436-41. 6. Masuhiro K, Matsuzaki N, Nishino E, et al. Trophoblast-derived Interleukin-1 (IL-I) stimulates the release of human chorionic gonadotropin by activating IL-6 and IL-g-receptor system in the first trimester human trophoblasts. J Clin Endocrinol Metab. 1991;72:594-601. 7. Ying L, Matauzaki N, Masuhiro K, et al. Trophoblast-derived tumor necrosis factor-a induces release of human chorionic gonadotropin using IL-6- and IL-6-receptor dependent system in the normal human trophoblasts. J Clin Endocrinol Metab. In Press. 8. Kameda T, Matsuzaki N, Sawai K, et al. Production of interleukin6 by normal trophoblast. Placenta. 1990;11:205-13. 9. Iwashita M, Watanabe M, Adachi T, et al. Effect of gonadal steroids on gonadotropin-releasing hormone stimulated human chorionic gonadotropin release by trophoblast cells. Placenta. 1989;10:103-12. 10. Assoian RK, Komoriya A, Meyers CA, Miller DM, Sporn MB. Transforming growth factora in human platelets. J Biol Chem. 1983,258:7155-60. 11. Mad&n L, Webb NR, Rose TM, Marquard H, Ikeda T, Twarzik D, Term placenta contains transforming growth factors. Biochm Biophy Res Commun. 1982;106:354-61.
ET AL.
JCE& M .19!32 Voli4.
No I
12. Madisen L, Webb NR, Rose TM, et al. Transforming growth factor02: cDNA cloning and sequence analysis. DNA. 1988,7:1-8. 13. Derynck R, Lindquist PB, Lee A, et al. A new type of transforming growth factor-& TGF-(33. EMBO J. 198&7:3737-43. 14. Jakowlew SB, Dillard PJ, Sporn MB, Roberts AB. Complementary deoxyribonucleic acid cloning of a messenger ribonucleic acid encoding transforming growth factor 84 from chicken embryo chondrocytes. Mol Endocrinol. 1988,2:1186-95. 15. Mivazono K. Hellman U. Wernstedt C. Heldin C-H. Latent hieh molecular weight complex of transforming growth factor-fil-puhfication from human platelets and structural characterization, J Biol Chem. 1988,263:6407-15. 16. Lyons RM, Kreski-Oja J, Moses HL. Proteolytic activation of latent transforming growth factor-8 from tibroblast-conditioned medium. J Cell Biol. 198&106;1659-65. 17. Mivazono K. Heldin CH. Role of carbohvdrate structures in TGF81 latency. Nature. 1989;338:158-60. ” 18. Antonelli-Orlidge A, Saunders KB, Smith SR, D’Amore PA. An activated form of transforming growth factor-8 is produced by coculture of endothelial cells and pericytes. Proc Nat1 Acad Sci USA. 1989;86:4544-8. 19. Pilatte Y, Bignon J, Lambre CR. Lysosomal and cytosolic sialidases in rabbit alveolar macrophages: demonstration of increased lysosomal activity after in uiuo activation with bacillus CalmetteGuerin. Biochim Biophys Acta. 1987;923:150-5. 20. Lucas C, Bald LN, Fendly BM, et al. The autocrine production of transforming growth factor-& during lymphocyte activation. A study with a monoclonal antibody-based ELIZA. J Immunol. 1990;145:1415-22. 21. Oberbauer AM, Linkhart TA, Mohan S, Longo LD. Fibroblast growth factor enhances human chorionic gonadotropin synthesis independent of mitogenic stimulation in Jar choriocarcinoma cells. Endocrinology. 198&123:2696-700. 22. Petraalia J. Vale W. Inhibin and activin modulate the - F. Vauahan release of gonadotropin-releasing hormone, human chorionic gonadotropin and progesterone from cultured human placental cells. Proc Nat1 Acad Sci USA. 1989;86:5114-7. 23. Taylor AW, Ku N-O, Mortensen RF. Regulation of cytokineinduced human C-reactive protein production by transforming growth factor. J Immunol. 1990;145:2507-13. 24. Dubois CM, Ruscetti FW, Palaszynski EW, Falk LA, Oppenheim JJ, Keller JR. Transforming growth factor (3 is a potent inhibitor of interleukin-I (IL-l) receptor expression: proposed mechanism of inhibition of IL-1 action.J Exp Med. 1990;172:737-44. 25. Eades DK. Cornelius P. Pekela PH. Characterization of tumor necrosis factor receptor in human placenta. Placenta. 1988;9: 247-51. 26. Cheifetz S, Weatherbee JA, Tsang ML-S, et al. The transforming growth factor system, a complex pattern of cross-reactive ligands and receptors. Cell. 1987;48:409-15.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
Vol. 74, No. 1 Printed in U.S.A
Trophoblast-Derived Transforming Growth Factor-& Suppresses Cytokine-Induced, but not GonadotropinReleasing Hormone-Induced, Release of Human Chorionic Gonadotropin by Normal Human Trophoblasts* NOBORU MATSUZAKI, YING LI, KAZUO MASUHIRO, TOUSHUN JO, KOUICHIROU SHIMOYA, TAKESHI TANIGUCHI, FUMITAKA SAJI, AND OSAMU TANIZAWA Department
of
Obstetrics and Gynecology,
Osaka University
Medical
ABSTRACT. Using a transforming growth factor-p (TGFp)sensitive cell line, MvlLu (or CCL 64), we demonstrated that trophoblasts predominantly produced a latent form of TGF& After converting latent TGFfi to active TGFp in oitro by acid (pH 2.5), alkali (pH lO.O), or heat (90 C; 10 min) treatment, addition of rabbit anti-TGFB, antiserum resulted in the elimination of TGFfl activity, thus suggesting that trophoblasts produced at least a certain amount of latent TGFP,. To investigate the role of TGF& in placental hormonogenesis, we first studied the effect of recombinant (r) TGFp, on the production of interleukin-6 (IL-6) and hCG by trophoblasts. rTGF/3, exerted no inhibitory activity on IL-6 and hCG production. The effect of rTGF& on cytokine-induced IL-6 and hCG release was then
School, Osaka 553, Japan
examined. While rTGF& failed to inhibit basal hCG secretion, it did inhibit recombinant tumor necrosis factor-a (rTNFa)induced IL-6 release as well as rTNFu- and rIL-g-induced hCG release in a dose-dependent manner. However, rIL-la-induced IL-6 and hCG release was remarkably sensitive to rTGF&mediated suppression. In contrast, GnRH-induced hCG release, the response of which is independent of the IL-6 and IL-6 receptor system in trophoblasts, was completely resistant to rTGF&. Thus, trophoblast-derived TGF& is an important regulatory molecule of cytokine-dependent, but not cytokine-independent, hCG release, possibly by converting latent TGFO to active TGF@ at the local site of trophoblasts. (J Clin Endocrinol Metab
T
74:
211-216,
1992)
using anti-TGF@, antiserum, we examined whether trophoblasts produce TGFP, molecules. Trophoblast-derived interleukin-1 (IL-l) and tumor necrosis factor-a (TNFa) synergistically induce IL-6 release, and the released IL-6 activates the IL-6 receptor (IL-6-R)-mediated signal transduction pathway to release hCG by normal trophoblasts; this is a cytokine-mediated regulatory cascade, resulting in hCG release (5, 6, 7). GnRHmediated hCG release (8), on the other hand, is independent of the cytokine-mediated hCG release mechanism (5). To investigate the TGF@-mediated regulatory mechanism in placental hormone function, we studied the effect of TGF/3 on cytokine- and GnRH-mediated hCG release as well as its effect on the production of trophoblast-derived IL-6 and hCG.
RANSFORMING growth factor-p (TGFB) is a multifunctional polypeptide with a mol wt of 25 kDa, composed of two disulfide-linked 12.SkDa subunits (1). TGF@ was initially defined as being related to the induction of malignant cells in viva (2). It contributes to the regulation of cell growth and differentiation because of its unique regulatory activities (2). There is increasing evidence showing unique TGFP-mediated effects on endocrinological systems (3). Little information, however, is available regarding TGFP-mediated modification of normal human placental functions. Using a TGFP-sensitive cell line, MvlLu (4), we first addressed the question of whether trophoblasts produce an active or latent form of TGFP-like molecule. Then, Received March 11, 1991. Address all correspondence and requests for reprints to: N. Matsuzaki, M.D., Ph. D., Department of Obstetrics and Gynecology, Osaka Universitv Medical School. l-1-50, Fukushima, Fukushima-ku, Osaka 553, Japah. * This work was supported by Grants-in-Aid for Scientific Research (no. 02454382.63480367. and 01570930) from the Ministry of Educaiion, Science, and Culture, Tokyo, Japan.
Materials
and Methods
Reagents Purified Escherichia coli-derived recombinant kindly provided by Dainippon Pharmaceutical
(r) TNFu! was Co. (Osaka,
X11
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
212
MATSUZAKI
ET AL.
JCE&M.1392 Vol74. No I
Japan). rIL-lcr and rIL-lb were gifts from Otsuka Pharmaceutical Co. Ltd. (Tokushima, Japan). rIL-6 was a kind gift from Dr. T. Kishimoto (Osaka University, Osaka, Japan). rTGF& and rabbit anti-TGF& immunoglobulin G (polyclonal antiTGF& neutralizing antiserum) were purchased from CosmoBio Co. (Tokyo, Japan). A GnRH agonist (GnRHa), [DLeua,Des-Gly’o-NH2]GnRH-N-ethylamide acetate (leuprolide acetate), was kindly provided by Takeda Pharmaceutical Co. (Osaka, Japan).
IL-6 and hCG release, rTGF& was added to the trophoblast culture, which was simultaneously stimulated with 10.0 Kg/L rIL-6, 20 pg/L rIL-ln, 200 pg/L rTNFq or various concentrations of GnRHa (leuprolide acetate). One hundred and fifty minutes after the stimulation, the culture supernatants were collected and assayed for hCG and IL-6. Each experiment was repeated at least three times, with identical results.
Preparation of purified trophoblasts
For measurement of the IL-6 titer, 1 X 10’ cells of an IL-6dependent murine hybridoma, MH60.BSF-2, were cultured in 96-well flat bottom microplates (Corning) with either test samples or standard rIL-6 for 48 h at 37 C in an incubator. The cultured cells were pulsed with 1 &i [3H]thymidine (15.7 Ci/ mM; New England Nuclear) for the last 6 h. The results were expressed as the mean t- SEM micrograms of IL-6 per L produced by the test cells, as described previously (8).
Fresh placentas were obtained from legal abortions carried out on social grounds at 7-9 weeks gestation. The placenta was thoroughly washed, and the villous tissue was separated from the connective tissue and minced on ice. To prepare single cell suspensions of trophoblasts, villous tissue (2-3 g) was first incubated with type III collagenase (Funakoshi, Tokyo, Japan) and DNase (Sigma Chemical Co., St. Louis, MO) at 37 C for 30 min in a shaking water bath. The enzyme-treated cells were filtered through a mesh and centrifuged at 800 x g for 6 min. To obtain purified trophoblasts, enzyme-treated placental cells were centrifuged twice on a 60% Percoll (Pharmacia Fine Chemicals, Piscataway, NJ) gradient for 15 min at 800 x g. The sedimented cells were then washed twice with culture medium to remove the collagenase. Cell viability, estimated by the trypan blue exclusion test, was more than 90%. Dispersed trophoblasts were adjusted to 2 x lo8 cells/L in culture medium and plated onto collagen-IA-coated dishes (Nitta Gelatin, Osaka, Japan). Cells (2 x 1OAcells/L) were cultured for 2 days at 37 C in 24-well flat bottom microplates (Corning, Iwaki Glass, Tokyo, Japan) (5-7). Bioassay for TGF/3 TGF/3 in the culture supernatants of purified trophoblasts was quantitated by comparing the level of inhibition of MvlLu mink lung cell (CCL 64) proliferation in supernatant dilution curves with that of rTGF@, (4). MvlLu (1 x lo7 cells/L) was seeded and cultured for 22 h. For the last 4 h, 1 &X/well [“HI thymidine (New England Nuclear, Boston, MA) was added. To detect the active form of TGFp, the fetal calf serum-free culture medium of trophoblasts was collected after 24-h culture. To activate the latent form of TGFfl to its active form, the medium was dialyzed against either acidic (pH 2.5) or alkaline solution (pH 10.0) overnight and then redialyzed with phosphate-buffered saline, pH 7.4, for 24 h or heated at 90 C for 10 min and cooled on ice. For TGF@ neutralization studies, rabbit antiTGFj3, antiserum was used. The neutralization rate (percentage) was calculated by comparing the count of [“Hlthymidine uptake between (antiserum-added group - background) and (antiserum-free group - background). rTGF& addition to trophoblast culture To observe the effect of TGFP, on IL-6 and hCG production, rTGF& was added at the initiation of (day 0) or 24 h after (day 1) culture of dispersed trophoblasts. Forty-eight hours after the start of culture, the culture supernatants were removed and assayed for IL-6 and hCG. To examine the effect of TGFp, on
Bioassay for IL-6
Hormonal assaysand statistics hCG released into the medium was measured by an enzyme assay (Mochida Co., Tokyo, Japan), using a monoclonal antibody specific for the P-subunit of hCG (5,9). Statistical analysis of the results for significance was performed by Duncan’s test after a one-way analysis of variance.
Results Production
of TGF/3 by normal trophoblasts
Figure 1A shows that trophoblasts produced predominantly a latent form, rather than an active form, of a TGFB-like molecule (P < 0.0005), because TGFP-like activity was detected predominantly in culture supernatants that had been treated with acid, alkali, or heat. Almost negligible TGFp activity was detected in untreated culture supernatants of trophoblasts, regardless of the presence or absence of fetal calf serum. As shown in Fig. lB, TGFB-mediated activity in the culture supernatants that had been treated with acid was neutralized in a dose-dependent manner by the addition of the rabbit anti-TGFP, antiserum. Effect of TGF@, on hCG and IL-6 production The results in Table 1 show no significantly inhibitory or augmenting effect of rTGF& on the titer of trophoblast-derived IL-6 or hCG. Measurement of IL-6 by a bioassay was not affected by the presence of rTGFPl (data not shown). Suppression
of rIL-6-induced
hCG release by rTGFDl
Table 2 shows that simple addition of rTGF& to cultured trophoblasts alone failed to augment or suppress hCG release, while addition of rIL-6 induced hCG release. As shown in Fig. 2, rTGFPl addition resulted in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
TGF&-MEDIATED
REGULATION
OF hCG
RELEASE
(B)
(A) 500. FIG. 1. Production of TGF@ by purified trophoblasts (A) and neutralization of trophoblast-derived TGFp activity by rabbit anti-TGF& antiserum (B). A, TGFP activity was assayed using a TGFP-sensitive cell line, MvlLu. The values represent the mean f SEM of triplicate determinations. ****, P < 0.0001. B, After conversion to active TGF@ by acid treatment, the culture supernatant (0.1 rg/L TGFP activity) was mixed with the antiserum. The neutralization rate (percentage) was calculated according to the formula described in Materials and
213
r-
-*t**
400,
~ 3oo 1 $ ‘, G ’ 2oo
I
100
Methods.
0 TABLE 1. Effect of rTGF& tion of IL-6 and hCG Pretreatment rTGI% (10 wdJ rTGFh (10 rg/L) Control medium
addition to trophoblasts Day of addition
IL-6 (&L) 5.0 + 0.8 4.5 + 0.7 5.6 f 1.2
0
1 0
on their produc-
hCG (IU/L) 1270 + 457 1007 +- 384 1227 * 293
1 : *-I--
rTGFfil (10 rg/L) was added either at initiation of (day 0) or after 24 h (day 1) of the trophoblast culture. After 48 h (day 2), the culture supernatants were collected for titration of IL-6 and hCG. The values represent the mean + SEM of triplicates. TABLE 2. Effect of rTGF& on hCG release from trophoblasts Stimulants Control medium rIL-6 rTGF&
Cont. (fig/L) 10 1.0 10
hCG release (W/L) 117 f 14 188 + 25b 120 k 10 129 + 14’
Cultured trophoblasts were incubated with various stimulants for 150 min. The culture supernatants were removed to determine hCG titer. The values represent the mean f SEM of triplicate determination. b P < 0.01 vs. control. ‘P = NS vs. control.
the suppression of rIL-6-induced hCG release in a dosedependent fashion at doses between 0.01-10 pg/L. This suppression was not due to impairment of cell viability, because rTGF& addition did not alter the cell viability, as determined by the trypan blue dye exclusion test (data not shown). Suppression of rIL-laIL-6 release
and rTNFwinduced
hCG and
As shown in Fig. 3A, the rTNFa-mediated hCG response was suppressed by rTGF& in a dose-dependent
I L-6 (IO!%/LI
(-)
(+)
AGF-/3 (U?./L)
i-!
i-)
(+)
(+I
(+)
(+)
(+I
0.001
0.01
0 1
1.0
10
FIN:. 2. rTGFP, suppresses rIL-6-mediated hCG release. Purified trophoblasts stimulated with rIL-6 (10 kg/L) were cultured in the presence or absence of rTGF/% at various concentrations. The values represent the mean -C SEM of triplicate determinations. *, P < 0.05; **, I’ < 0.01.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
214
MATSUZAKI
ET AZ,
JCE ti M. 1992 Voli4.Nol
(A
FIG. 3. rTGFp, suppresses rTNFcu-mediated and rIL-la-mediated hCG and IL-6 release. Purified trophoblasts stimulated with either rTNFa (200 pg/L; A and B) or rIL-la (20 rg/L; C and D) were cultured in the presence or absence of rTGFj3 at various concentrations. The values represent the mean f SEM of hCG (A and C) and IL-6 (B and D) in triplicate determinations. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
II
__,-. )EFp 1:: l”,iLi
c+, c-i
,+b 001
(Cl
T
1 manner. Figure 3B demonstrates that rTGF& also inhibited rTNFa-induced IL-6 release. The inhibition ratio of rTNFa-induced IL-6 release by rTGF& was similar to that of either rTNFa- or rIL-6-induced hCG release by rTGF& (Figs. 3, A and B, and 2). Figure 3C, in contrast, shows that rTGF/3, completely ‘blocked rIL-la-induced hCG release even at a concentration of 0.001 pg/L (P < 0.01). Moreover, as demonstrated in Fig. 3D, the effect of rTGF/3 on rIL-la-induced IL-6 release was extremely sensitive compared with its effect on rTNFa-induced IL6 release even at a concentration of 0.001 pg/L rTGF& (P < 0.001; Fig. 3B). Similar rTGFP1-mediated suppression was observed with trophoblasts stimulated with rIL18 (data not shown). Similar TGF&-mediated suppression was observed after rTGFPl addition even after rILla- or rTNFa-mediated stimulation of trophoblasts (data not shown).
--A I
Effect of rTGF&
on GnRH-induced
*em-
hCG release
We tested whether TGFP, was capable of suppressing GnRH-mediated hCG release. As illustrated in Fig. 4, rTGF& failed to significantly suppress GnRH-mediated hCG release (P = NS, GnRHa-stimulated us. GnRHaplus rTGF/&-stimulated groups), while it completely suppressed rIL-g-mediated hCG release (P < 0.01). Discussion Production of TGFP by the human placenta has been reported. Assoian et al. (10) isolated TGFp from the platelets in the placenta, while Madisen et al. (11) purified TGFP from human placental tissue. Our present study revealed that trophoblasts in the placenta predominantly produced a latent form of a TGFP-like molecule, the activity of which was significantly neutralized by
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
TGF&-MEDIATED Stimulants
rTGF-fl,
hCG
Release
I Control
NIL-6
Medwm
REGULATION
(IU/L! 1
-
(10 /x/L) I- +
GI+?H&oo~~cUL)
I+
I- +
GnRHa
(10 nmo~/~)
GnRHa
( 1 ” ml/L)
0
I 50
100
150
200
250
FIG. 4. rTGF& fails to suppress GnRHa-mediated hCG release. Trophohlasts were stimulated with either rIL-6 (10 pg/L) or various concentrations of GnRHa in the presence or absence of rTGF fil (10 pg/L). The values represent the mean f SEM of triplicate determinations. *, PC 0.05; **, P < 0.01.
rabbit anti-TGF& antiserum. Trophoblasts, therefore, produce at least a certain amount of TGFP,, although it is not clear whether trophoblasts produce other types of TGFp such as TGF& (12), -iij3 (13), and -p4 (14). A latent form of TGFP has been reported to be incapable of exerting various biological activities associated with TGF@ because of its failure to bind to native TGFfiR (15). Latent TGFp can be converted to its active form in uitro by various methods, including changes in pH (3, 6), thermal activation (3), treatment with proteases (16) as well as glycosidases (17), and cell to cell contact (18). Such conversion of TGFp has been reported to occur in activated immunocompetent cells that generate local pH shift, proteases, and sialidases to produce active TGFP (19). It was described in a recent study that the addition of anti-TGFP, monoclonal antibody resulted in augmentation of lymphocyte proliferation in culture supernatants in which only endogenously derived latent TGF/3 activity could be detected (20). This suggests that even endogenously produced TGFP, i.e. latent TGFP, might exert regulatory activities after being converted to active TGF@ at a local site. Oberbauer et al. (21) reported that TGFp significantly suppressed epidermal growth factor-stimulated hCG production in Jar choriocarcinoma cells. However, few reports are available that show TGFP-mediated regulatory activities on normal human placental hormone release and production. Although simple addition of TGF@ showed no inhibition or augmentation of the basal hCG secretion, our present study revealed the unique regulatory activities of TGFP on the cytokine-mediated regulatory network; both trophoblast-derived IL-l and TNFa synergistically stimulate hCG release by inducing tro-
OF hCG
RELEASE
215
phoblast-derived IL-6 release and activating IL-6-R-mediated signal transduction pathways (7). rTGF& markedly suppressed rIL-la-, rTNFLu-, or IL-6-induced hCG release. rTGF0, also inhibited rIL-lor- or rTNFa-induced IL-6 release by trophoblasts, which response shows differential sensit,ivity to the concentration of rTGF& added; compared with the magnitude of its effect on rTNFa-induced IL-6 and hCG release, rTGF& showed marked suppression of rIL-la-induced release. In contrast, GnRH, which triggers hCG release independently of the IL-6 and IL-6-R system in trophoblasts (5), shows complete resistance to TGFPI-mediated inhibitory activity. TGFB,, therefore, showed preferential suppressive action on cytokine-mediated, but not GnRH-mediated, hCG release by regulating the amount of IL-6 released and the IL-6-induced hCG release. Rather, GnRH-induced hCG secretion has been reported to be up-regulated by activin and down-regulated by inhibin, two other cytokine members of the TGFp family (22). hCG release, thus, might be down-regulated by cytokines in the TGFp family, TGF/& and inhibin. The cellular and molecular mechanisms of TGFpmediated suppression have been extensively investigated in other cellular systems, such as hepatocytes (23). Dubois et al. (24) defined TGF@ as a potent inhibitor of IL1-R expression, which results in TGFP-mediated suppression of IL-l-induced responses. However, such mechanisms might not be operative in our system, because rTGFpl was capable of suppressing rIL-la-induced IL-6 and hCG release even after IL-l and IL-l-R interaction, since TGF& was capable of suppressing rIL-lainduced IL-6 and hCG release even when it was added after rIL-lu stimulation of trophoblasts. The finding that TGFP, inhibited cytokine-mediated IL-6 and hCG release even after addition of the cytokines also excluded the possibility that TGF/?, might competitively bind to IL-l-R (6), TNFn-R (25), or IL-6-R (5) on trophoblasts, thereby serving as a cytokine inhibitor itself. In contrast to the TGFB-mediated regulatory activity on IL-6 and hCG release, TGFp exerted no inhibitory or augmenting effect on IL-6 or hCG production. These findings together with TGFP’s effect on GnRH-induced hCG release exclude the possibility that rTGF& is toxic to trophoblasts and thereby reduces IL-6 and hCG release, indicating that TGF& produced by trophoblasts, platelets, and monocytes in the placenta might act as a physiological regulator in the cytokine-dependent hCG release mechanism in an autocrine or paracrine fashion. It seems possible to speculate that such a TGFP-mediated regulatory function might be supported through TGF@ receptors and postreceptor-mediated signal transduction pathways. In hematopoietic cells, three distinct receptor molecules for TGFP, i.e. with molecular sizes of 65, 85, and 260 kDa, have been demonstrated (26).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
216
MATSUZAKI
Trophoblasts seem to possess a TGFP- and TGF&Rspecific signal transduction pathway common to immunocompetent cells, hematopoietic cells, hepatocytes, and synovial cells. Further experiments will be required to demonstrate the presence of TGFP-R on trophoblasts and characterize the nature and mechanism of postTGF/3-R intracellular signal transduction pathways that participate in the regulation of hCG release. References 1. Massague J. The TGF-j3 family of growth, differentiation. Cell. 1987;49:431-8. 2. Sporn MB, Roberts AB, Wakefield LM, Assoian RK. Transforming-8: biological function and chemical structure. Science. 1986;233:532-4. 3. Kim I-U, Shomberg DW. The production of transforming growth factor+3 activity by rat granulosa cell cultures. Endocrinology. 1989;124:1345-51. 4. Holly RW, Armour R, Baldwin JH, Greenfield S. Activity of a kidney epithelial cell growth on lung and mammary cells. Cell Biol Int Rep. 1983;7:141-7. 5. Nishino E, Matsuzaki N, Masuhiro K, et al. Trophoblast-derived IL-6 regulates human chorionic gonadotropin release through IL6 receptor on human trophoblasts. J Clin Endocrinol Metab. 1990;71:436-41. 6. Masuhiro K, Matsuzaki N, Nishino E, et al. Trophoblast-derived Interleukin-1 (IL-I) stimulates the release of human chorionic gonadotropin by activating IL-6 and IL-g-receptor system in the first trimester human trophoblasts. J Clin Endocrinol Metab. 1991;72:594-601. 7. Ying L, Matauzaki N, Masuhiro K, et al. Trophoblast-derived tumor necrosis factor-a induces release of human chorionic gonadotropin using IL-6- and IL-6-receptor dependent system in the normal human trophoblasts. J Clin Endocrinol Metab. In Press. 8. Kameda T, Matsuzaki N, Sawai K, et al. Production of interleukin6 by normal trophoblast. Placenta. 1990;11:205-13. 9. Iwashita M, Watanabe M, Adachi T, et al. Effect of gonadal steroids on gonadotropin-releasing hormone stimulated human chorionic gonadotropin release by trophoblast cells. Placenta. 1989;10:103-12. 10. Assoian RK, Komoriya A, Meyers CA, Miller DM, Sporn MB. Transforming growth factora in human platelets. J Biol Chem. 1983,258:7155-60. 11. Mad&n L, Webb NR, Rose TM, Marquard H, Ikeda T, Twarzik D, Term placenta contains transforming growth factors. Biochm Biophy Res Commun. 1982;106:354-61.
ET AL.
JCE& M .19!32 Voli4.
No I
12. Madisen L, Webb NR, Rose TM, et al. Transforming growth factor02: cDNA cloning and sequence analysis. DNA. 1988,7:1-8. 13. Derynck R, Lindquist PB, Lee A, et al. A new type of transforming growth factor-& TGF-(33. EMBO J. 198&7:3737-43. 14. Jakowlew SB, Dillard PJ, Sporn MB, Roberts AB. Complementary deoxyribonucleic acid cloning of a messenger ribonucleic acid encoding transforming growth factor 84 from chicken embryo chondrocytes. Mol Endocrinol. 1988,2:1186-95. 15. Mivazono K. Hellman U. Wernstedt C. Heldin C-H. Latent hieh molecular weight complex of transforming growth factor-fil-puhfication from human platelets and structural characterization, J Biol Chem. 1988,263:6407-15. 16. Lyons RM, Kreski-Oja J, Moses HL. Proteolytic activation of latent transforming growth factor-8 from tibroblast-conditioned medium. J Cell Biol. 198&106;1659-65. 17. Mivazono K. Heldin CH. Role of carbohvdrate structures in TGF81 latency. Nature. 1989;338:158-60. ” 18. Antonelli-Orlidge A, Saunders KB, Smith SR, D’Amore PA. An activated form of transforming growth factor-8 is produced by coculture of endothelial cells and pericytes. Proc Nat1 Acad Sci USA. 1989;86:4544-8. 19. Pilatte Y, Bignon J, Lambre CR. Lysosomal and cytosolic sialidases in rabbit alveolar macrophages: demonstration of increased lysosomal activity after in uiuo activation with bacillus CalmetteGuerin. Biochim Biophys Acta. 1987;923:150-5. 20. Lucas C, Bald LN, Fendly BM, et al. The autocrine production of transforming growth factor-& during lymphocyte activation. A study with a monoclonal antibody-based ELIZA. J Immunol. 1990;145:1415-22. 21. Oberbauer AM, Linkhart TA, Mohan S, Longo LD. Fibroblast growth factor enhances human chorionic gonadotropin synthesis independent of mitogenic stimulation in Jar choriocarcinoma cells. Endocrinology. 198&123:2696-700. 22. Petraalia J. Vale W. Inhibin and activin modulate the - F. Vauahan release of gonadotropin-releasing hormone, human chorionic gonadotropin and progesterone from cultured human placental cells. Proc Nat1 Acad Sci USA. 1989;86:5114-7. 23. Taylor AW, Ku N-O, Mortensen RF. Regulation of cytokineinduced human C-reactive protein production by transforming growth factor. J Immunol. 1990;145:2507-13. 24. Dubois CM, Ruscetti FW, Palaszynski EW, Falk LA, Oppenheim JJ, Keller JR. Transforming growth factor (3 is a potent inhibitor of interleukin-I (IL-l) receptor expression: proposed mechanism of inhibition of IL-1 action.J Exp Med. 1990;172:737-44. 25. Eades DK. Cornelius P. Pekela PH. Characterization of tumor necrosis factor receptor in human placenta. Placenta. 1988;9: 247-51. 26. Cheifetz S, Weatherbee JA, Tsang ML-S, et al. The transforming growth factor system, a complex pattern of cross-reactive ligands and receptors. Cell. 1987;48:409-15.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 13:44 For personal use only. No other uses without permission. . All rights reserved.
