Transforming growth factor-/fe (TGF-βs) are potent regulators of cell growth and differentiation. Expression of the closely related TGF-β subtypes in vivo is differentially regulated both temporally and spatially. Members of the steroid hormone superfamily may play an important role in this gene- and tissue-specific regulation. We have shown that anti-estrogens induce the production of TGF-β1 in mammary carcinoma cells and fetal fibroblasts, whereas retinoic acid specifically induces TGF-β2 in primary epidermal kératinocytes. The induction of TGF-β2 by retinoids is accompanied by an increase in TGF-β2 mRNAs, but little change in transcription rates, suggesting an effect of retinoids on message stability or processing. In contrast, TGF-β1 mRNA levels are unchanged by anti-estrogen treatment, suggesting these compounds may regulate the translatability of the TGF-β1 message or some post-translational processing event. We have identified a stable stem-loop structure in the 5′ untranslated region (UTR) of the TGF-β mRNA that inhibits translation of a heterologous reporter gene, and we are investigating the possibility that anti-estrogens may regulate the activity of this element, and hence the translatability of the TGF-β1 message. A significant fraction (25–90%) of the TGF-β induced by retinoids and anti-estrogens is in the biologically active rather than the latent form. We have shown that active TGF-β has a much shorter in vivo half-life than latent TGF-β, suggesting that the TGF-β induced by retinoids and steroids may act locally at the site of production. Since many tumor cells retain sensitivity to the growth inhibitory effects of active TGF-β, the use of members of the steroid hormone superfamily for inducing this potent growth inhibitor locally at the tumor site may have therapeutic potential.

Members of the transforming growth factor-β superfamily have emerged as key regulators of many aspects of cell growth, differentiation and function (for review, see Roberts and Sporn, 1990). Five structurally highly related TGF-β subtypes have been identified to date, and multiple subtypes appear to be expressed in all species examined. Thus TGF-βs 1, 2 and 3 are expressed in human; 1, 2,3 and 4 in chick; and 2 and 5 in frog. In situ and immunohistochemical studies suggest that these isoforms are differentially regulated, both temporally and spatially (Heine et al. 1987; Pelton et al. 1989; Miller et al. 1989). While the different TGF-β subtypes are equipotent in the majority of biological assays in vitro (reviewed by Roberts and Sporn, 1990), there is some indication that in more complex systems, involving interactions between different cell types, the TGF-β subtypes differ in their activities. For example, only TGF-β2 and TGF-β3 are active in inducing mesoderm formation in Xenopus embryos (Roberts et al. 1990). The significance of multiple TGF-β subtypes may therefore be twofold: (1) different regulatory elements controlling expression of the various TGF-β subtypes may allow differential regulation of the same biological activity in an organ- and time-specific manner, and (2) subtype switching in a given tissue may result in expression of a different biological activity in the context of that tissue. The recent cloning and analysis of the promoters for the TGF-β1, 32 and genes indicate that these are very different in structure, suggesting one potential mechanism for the differential regulation of TGF-β subtypes (Kim et al. 1989; Noma et al.; Lafyatis et al. unpublished).

We have examined the possible role of members of the steroid hormone superfamily in the regulation of TGF-β subtypes. The goal here has been (1) to identify endogenous regulatory molecules responsible for the selective expression of different TGF-β subtypes observed in vivo, and (2) to identify analogs or antagonists of these agents which may allow us to manipulate the endogenous levels of various TGF-β subtypes in vivo, in a target-specific manner.

Several members of the steroid hormone superfamily have been shown to regulate the production of TGF-β subtypes

The data in Table 1 summarize work from our laboratories and others showing an induction of TGF-β family members by steroids and related compounds in a variety of different target cell types. In addition to these examples of positive regulation, it should be noted that negative regulatory effects have been observed in other systems. For example, estrogen appears to depress TGF-β2 and TGF-β3 mRNAs in human breast carcinoma cells (Arrick, Korc and Derynck, 1990), and to cause a decrease in TGF-β protein (subtype not identified) in the same cells (Knabbe et al. 1987), and progestins decrease the level of TGF-β1 mRNA in the T47D human breast carcinoma line (Murphy and Dotzlaw, 1989). It is apparent that members of the steroid hormone superfamily can regulate the production of distinct TGF-β subtypes in vitro, in a target-specific manner. The inhibitory effects of the anti-estrogens and gestodene on breast carcinoma cells, and of retinoids on kératinocytes, are partially reversed by neutralizing antibodies to TGF-β (Knabbe et al. 1987; Colletta et al. 1990a,b; Glick et al. 1989). This suggests that the induction of TGF-β subtypes has functional significance in the mechanism of action of steroids and retinoids.

Table 1.

Induction of TGF-β subtypes by members of the steroid hormone superfamily

Induction of TGF-β subtypes by members of the steroid hormone superfamily
Induction of TGF-β subtypes by members of the steroid hormone superfamily

The regulation of TGF-β family members by steroids/retinoids is not simply an in vitro curiosity, since the induction of TGF-β2 in kératinocytes by retinoids has now been confirmed in vivo (Glick et al. 1989). Retinoic acid was applied to the shaved skin of adult Balb/c mice, and treated and control skins were processed for immunohistochemistry 48 h later. Using subtype-specific anti-TGF-β antibodies, a dramatic increase in immunoreactive TGF-β2 was demonstrated in the follicular and interfollicular epithelium of retinoid-treated skins. Little staining for TGF-β1 was observed in either treated or untreated skins. Thus retinoic acid can also function as subtype-specific inducer of TGF-β in the more complex in vivo situation. Retinoids have been implicated in the control of proliferation and normal differentiation in a variety of epithelia (Sporn and Roberts, 1984). Since TGFβc are strongly growth inhibitory for many epithelial cells (reviewed by Roberts and Sporn, 1990), inducing terminal differentiation in some instances, this indicates that the observed actions of retinoids in vivo may be due in part to their ability to modulate local levels of TGF-β2 in target epithelia.

Regulation of TGF-f> subtypes by retinoids and steroids is frequently at the post-transcriptional level

Retinoic acid treatment of mouse kératinocytes results in a greater than 20-fold increase in the steady-state levels of the four major transcripts coding for TGF-β2 (Glick et al. 1989). However, nuclear run-on experiments indicate that there is no increase in the rate of transcription of the TGF-β2 mRNAs on retinoic acid treatment, so the observed increase in transcript levels must be due to post- transcriptional effects (Glick et al. 1989). Possible mechanisms include effects of retinoids on message processing, transport or stability. Post-transcriptional effects of retinoids on other mRNAs have been observed, indicating this may be a major pathway by which retinoids control gene expression (Smits et al. 1987).

In contrast, the induction of TGF-β1 in breast carcinoma cells in response to anti-estrogens or gestodene, and in fibroblasts in response to anti-estrogens, is accompanied by little or no change in TGF-β1 mRNA levels (Colletta et al. 1990a; Colletta et al. 19906). For example, treatment with 500 nM gestodene causes a greater than 90-fold induction of TGF-β1 protein, with less than a 3-fold increase in corresponding mRNA levels (Colletta et al. 1990a). This suggests that the anti-estrogens and gestodene may be affecting the efficiency of translation of the TGF-β mRNA, or enhancing some subsequent post-translational step, such as post- translational processing or secretion. Comparably large effects on TGF-β1 protein levels without effects on the mRNA are observed in other systems. For example, activation of lymphocytes with phytohaemagglutinin is accompanied by a large increase in TGF-β mRNA within 6 h, but an increase in secreted TGF-β protein is not observed for a further 2 days (Kehrl et al. 1986). Similarly, activation of monocytes with lipopolysaccharide results in increased TGF-β secretion with no change in mRNA levels (Assoian et al. 1987). Important translational or post- translational regulatory effects on the TGF-β1 message are therefore observed in a variety of systems. Experimentally, the presence of TGF-β1 mRNA should not be assumed necessarily to indicate the presence of the cognate protein.

In common with a number of cytokines and oncogenes, the TGF-β1 mRNA has a long (>800bp) 5’ untranslated region (UTR) (Kim et al. 1989). Computer-aided analysis of possible secondary structure formation in the TGF-β1 mRNA indicates the potential presence of two highly stable (AG≈-16.84 kJ) (1 kJ=0.239kcal) stem-loop structures in the 5′ UTR, shown diagrammatically in Fig. 1 (S.-J. Kim, unpublished). The ferritin mRNA has a very similar stem-loop structure in the 5′ UTR, and this stem-loop has been shown to be the site of binding of a regulatory protein that acts as a repressor of ferritin mRNA translation under conditions of low availability of iron (Klausner and Harford, 1989). By analogy, we propose that the putative stem-loops in the 5′ UTR of the TGF-/J1 mRNA may be involved in regulating the translatability of the TGF-β1 message in a similar fashion. Preliminary experiments in which cDNA fragments of the 5’ UTR were inserted upstream from the coding sequence of a human growth hormone reporter gene, indicate that the downstream stem-loop structure has an inhibitory effect on the translation of the reporter gene, in certain cell backgrounds (S.-J. Kim, unpublished). Further work is underway to define more closely the structures and specific sequence elements that are responsible for translational regulation of this message.

Fig. 1.

Diagrammatic representation of the TGF-β1 mRNA, showing location of potential stem-loop structures in the 5′ untranslated region.

Fig. 1.

Diagrammatic representation of the TGF-β1 mRNA, showing location of potential stem-loop structures in the 5′ untranslated region.

A significant fraction of the TGF-fi induced by steroids or retinoids is in the biologically active form

The TGF-βs differ from the majority of growth regulatory factors in that they are generally synthesized and secreted in a biologically latent form, which must be activated before TGF-β can exert its biological effects on target cells (Lawrence et al. 1984; Wakefield et al. 1987). For TGF-β, the latent form is a non-covalent complex in which the homodimeric active TGF-β1 is non-covalently associated with a dimer of the remainder of its precursor ‘pro’ region, and this in turn is disulfide-bonded to a third, structurally unrelated, protein of 135Mr×10−3 (Wakefield et al. 1988; Miyazono et al. 1988). The nature of the activation mechanism in vivo is unclear, but may involve proteases, and in some instances be dependent on cell-cell interactions (Antonelli-Orlidge et al. 1989; Sato and Rifkin, 1989).

For most cells in culture, the TGF-β secreted is more than 95% latent (Wakefield et al. 1987). However, for cells induced to secrete TGF-β in response to retinoids and steroids, a significant fraction of the secreted TGF-β is in the biologically active form (see Table 2). The active fraction ranges from 25% to nearly 100% of the total TGF-β secreted. Using latent TGF-β1 generated by combining iodinated active TGF-/51 with the TGF-β1 precursor pro region from recombinant sources, we have demonstrated that the latent form of TGF-β1 has a greatly extended plasma half-life in vivo when compared with active TGF-β1 (see Table 3) (Wakefield et al. 1990). This is consistent with data from our laboratory and others suggesting that active TGF-β1 may be rapidly cleared from the extracellular fluid and plasma as a complex with alpha-2-macroglobulin (O’Connor-McCourt and Wakefield, 1987; Coffey et al. 1987). The latent form of TGF-β does not bind to alpha-2-macroglobulin and cannot be cleared by this route (Wakefield et al. 1988). We have therefore proposed that whereas latent TGF-β may be able to exert a long-range, endocrine type of action, active TGF-/5 probably acts very locally to its site of production, in an autocrine/paracrine fashion (see Fig. 2).

Table 2.

The fraction of TGF-β that is induced in the active, as opposed to biologically latent, form in response to members of the steroid hormone superfamily

The fraction of TGF-β that is induced in the active, as opposed to biologically latent, form in response to members of the steroid hormone superfamily
The fraction of TGF-β that is induced in the active, as opposed to biologically latent, form in response to members of the steroid hormone superfamily
Table 3.

Comparison of the plasma half-lives of active and latent forms of TGF--β

Comparison of the plasma half-lives of active and latent forms of TGF--β
Comparison of the plasma half-lives of active and latent forms of TGF--β
Fig. 2.

Active and latent TGF-β1 may have a different range of action. Active TGF-β1 has a very short plasma half-life. Active TGF-/J1 secreted by cells is likely to bind rapidly to the ubiquitous cell surface binding proteins for TGF-β1, or to be cleared from the vicinity, possibly as a complex with alpha-2-macroglobulin (a-2M). This is likely to restrict the site of action of active TGF-β to target cells close to the site of production. By contrast, latent TGF-β does not bind to alpha-2-macroglobulin, and has a much longer plasma half-life than active TGF-β. It may therefore be carried by the circulation to more distant targets, and exert a more long-range endocrine type of action.

Fig. 2.

Active and latent TGF-β1 may have a different range of action. Active TGF-β1 has a very short plasma half-life. Active TGF-/J1 secreted by cells is likely to bind rapidly to the ubiquitous cell surface binding proteins for TGF-β1, or to be cleared from the vicinity, possibly as a complex with alpha-2-macroglobulin (a-2M). This is likely to restrict the site of action of active TGF-β to target cells close to the site of production. By contrast, latent TGF-β does not bind to alpha-2-macroglobulin, and has a much longer plasma half-life than active TGF-β. It may therefore be carried by the circulation to more distant targets, and exert a more long-range endocrine type of action.

The observation that steroids and retinoids frequently induce TGF-β in its active form is important for two reasons. First, as indicated above, it means that the TGF-β induced by these agents is likely to have a very local action. Second, any cell in the vicinity that possesses TGF-β receptors is a potential target for the induced TGF-β. This contrasts with the situations where latent TGF-β is induced, since cells that are capable of activating latent TGF-β have yet to be identified.

Implications for chemoprevention or therapy of epithelial malignancies

TGF-/Ss are potent inhibitors of the growth of normal epithelial cells (for review, see Roberts and Sporn, 1990). One might expect therefore that loss of response to this endogenous growth inhibitor could contribute to the genesis of epithelial tumors. Indeed, there is experimental evidence in the rat tracheal epithelial system suggesting that neoplastic progression following carcinogen treatment is accompanied by increased resistance to growth inhibition by TGF-β (Hubbs, Hahn and Thomassen, 1989). Similarly, less aggressive, well-differentiated human colon carcinoma cells are growth-inhibited by TGF-β, whereas more agressive, poorly differentiated colon tumor cells are not (Hoosein et al. 1989). However, a recent survey of the literature indicated that of the 37 human carcinoma cell lines analysed for a response to TGF-β in vitro, more than half (20) of these retained the ability to be inhibited by TGF-β (Wakefield and Sporn, 1990). This suggests that there may be a window of opportunity during neoplastic progression, when raising local TGF-β concentrations in the vicinity of the developing tumor might restore a measure of growth control and slow down the progression of the affected cells to full-blown malignancy. Since resistance to TGF-β appears to increase with malignant progression, and because tumors become more heterogeneous as they progress, it would obviously be most effective to intervene as early as possible, preferably during the preneoplastic phase, to prevent tumor development.

The pleiotropic effects of TGF-βe on multiple cell types suggests that systemic elevation of TGF-β levels in a clinical situation should probably be avoided. Ideally, therefore, one would like to develop pharmacological agents that could cause a highly localized induction of TGF-β in specific target tissues. Analogs and antagonists of members of the steroid hormone superfamily seem particularly promising in this regard; only cells expressing the appropriate ligand receptor will be targets for the steroid action. Furthermore, as noted above, the TGF-β secreted in response to these agents is largely in the biologically active form, which means that its action will be local, and that even tumor cells that lack or have lost the ability to activate the latent form will be growth-inhibited.

From the point of view of achieving very specific tissue targeting, one molecule of potential interest is the synthetic progestin, gestodene, which is a component of an oral contraceptive preparation widely used in Europe. Although gestodene binds to the classical progestin receptor, an additional novel binding site for this agent has been demonstrated in malignant breast tissue and cells lines (Iqbal et al. 1986; Colletta et al. 1989). This binding site does not appear to be present in normal cells or cells of other malignancies (Iqbal et al. 1986). We have shown that, acting through the novel gestodene binding site, gestodene will induce TGF-β1 in breast cancer cell lines and inhibit their growth (Colletta et al. 1990a). Other progestins that only bind to the classical progesterone receptor have no effect.

Here, therefore, we may have a prototype of a novel form of chemopreventive agent. In the normal breast, and other tissues, gestodene would bind to the progestin receptor and exert its normal progestogenic activity. However, in malignant breast tissue that expresses the unique gestodene binding site, gestodene may bind to this site and induce TGF-β production, thereby slowing growth of the malignant cells (Fig. 3). It is not yet known at what stage in malignant progression the gestodene binding site is first expressed, but obviously the earlier this occurs, the greater the chance of effective prevention. While it remains to be shown that gestodene will actually prevent development of breast cancers in vivo, the work nevertheless indicates that agents may be found that can cause induction of TGF-β subtypes in very restricted target tissues, and suggests that the use of pharmacological compounds to induce local production of endogenous growth inhibitors in malignant or premalignant cells represents an important new approach to the problem of prevention and treatment of epithelial malignancies.

Fig. 3.

Possible roles of the synthetic progestin, gestodene, in normal and malignant tissues. In normal tissues, gestodene binds to the classical progesterone receptor and exerts a progestagenic effect. In malignant breast cells, which express a novel binding site for gestodene, gestodene induces the production of TGF-β1, which may then inhibit the growth of the malignant cell in an autocrine fashion. For more details, see text.

Fig. 3.

Possible roles of the synthetic progestin, gestodene, in normal and malignant tissues. In normal tissues, gestodene binds to the classical progesterone receptor and exerts a progestagenic effect. In malignant breast cells, which express a novel binding site for gestodene, gestodene induces the production of TGF-β1, which may then inhibit the growth of the malignant cell in an autocrine fashion. For more details, see text.

Members of the steroid hormone superfamily have been shown to regulate the production of TGF-β subtypes in a variety of target cell types, and experiments with neutralizing antibodies indicate that the TGF-βs may be local mediators of some of the biological activities of steroids/retinoids. Regulation of TGF-β subtypes by these agents appears to involve predominantly post-transcriptional mechanisms, and much of the TGF-β secreted is in the biologically active, not the more common latent, form. Since active TGF-β is much more rapidly cleared from the circulation than the latent form, this suggests that the TGF-β induced by steroids/retinoids may have a very local action. Since TGF-βs are highly potent inhibitors of epithelial cell growth, the targetted local induction of this peptide family could be exploited in the development of novel chemopreventive or chemotherapeutic strategies for the management of epithelial malignancies.

Antonelli-Orlidge
,
A.
,
Saunders
,
K. B.
,
Smith
,
S. R.
and
D’Amore
,
P. A.
(
1989
).
An activated form of transforming growth factor /? is produced by cocultures of endothelial cells and pericytes
.
Proc. natn. Acad. Sci. U.S.A.
86
,
4544
4548
.
Arrick
,
B. A.
,
Korc
,
M.
and
Derynck
,
R.
(
1990
).
Differential regulation of expression of three transforming growth factor-/? species in human breast cancer cell lines by estradiol
.
Cancer Res.
50
,
299
303
.
Assoian
,
R. K.
,
Fleurdelys
,
B. E.
,
Stevenson
,
H. C.
,
Miller
,
P. J.
,
Madtes
,
D. K.
,
Raines
,
E. W.
,
Ross
,
R.
and
Sporn
,
M. B.
(
1987
).
Expression and secretion of type fi transforming growth factor by activated human macrophages
.
Proc. natn. Acad. Sci. U.S.A.
84
,
6020
6024
.
Coffey
,
R. J. Jr
,
Kost
,
L. J.
,
Lyons
,
R. M.
,
Moses
,
H. L.
and
Larusso
,
N. F.
(
1987
).
Hepatic processing of transforming growth factor /? in the rat
.
J. clin. Invest.
80
,
750
757
.
Colletta
,
A. A.
,
Howell
,
F. V.
and
Baum
,
M.
(
1989
).
A novel binding site for a synthetic progestagen in breast cancer cells
.
J. Steroid Biochem.
33
,
1055
1061
.
Colletta
,
A. A.
,
Wakefield
,
L. M.
,
Howell
,
F. V.
,
Danielpour
,
D.
,
Baum
,
M.
and
Sporn
,
M. B.
(
1990a
).
The growth inhibition of human breast cancer cells by a novel synthetic progestin is partly mediated by the induction of transforming growth factors /?
.
J. Clin. Invest., in press
.
Colletta
,
A. A.
,
Wakefield
,
L. M.
,
Howell
,
F. V.
,
Van Roozendaal
,
K. E. P.
,
Danielpour
,
D.
,
Ebbs
,
S. R.
,
Sporn
,
M. B.
and
Baum
,
M.
(
1990b
).
Antioestrogens induce the secretion of active transforming growth factor f> from human fibroblasts
.
Br. J. Cancer, in press
.
Glick
,
A. B.
,
Flanders
,
K. C.
,
Danielpour
,
D.
,
Yuspa
,
S. H.
and
Sporn
,
M. B.
(
1989
).
Retinoic acid induces transforming growth factor-/?2 in cultured kératinocytes and mouse epidermis
.
Cell Reg.
1
,
87
97
.
Heine
,
U. I.
,
Munoz
,
E. F.
,
Flanders
,
K. C.
,
Ellingsworth
,
L. R.
,
Lam
,
H.-Y. P.
,
Thompson
,
N. L.
,
Roberts
,
A. B.
and
Sporn
,
M. B.
(
1987
).
Role of transforming growth factor-/? in the development of the mouse embryo
.
J. Cell Biol.
105
,
2861
2876
.
Hoosein
,
N. M.
,
McKnight
,
M. K.
,
Levine
,
A. E.
,
Mulder
,
K. M.
,
Childress
,
K. E.
,
Brattain
,
D. E.
and
Brattain
,
M. G.
(
1989
).
Differential sensitivity of subclasses of human colon carcinoma cell lines to the growth inhibitory effects of transforming growth factor-/?!
.
Expl Cell. Res.
181
,
442
453
.
Hubbs
,
A. F.
,
Hahn
,
F. F.
and
Thomassen
,
D. G.
(
1989
).
Increased resistance to transforming growth factor beta accompanies neoplastic progression of rat tracheal epithelial cells
.
Carcinogenesis
10
,
1599
1605
.
Iqbal
,
M. J.
,
Colletta
,
A. A.
,
Houmayoun-Valyani
,
S. D.
and
Baum
,
M.
(
1986
).
Differences in oestrogen receptors in malignant and normal breast tissue as identified by the binding of a new synthetic progestogen
.
Br. J. Cancer
54
,
447
-
452
.
Kehrl
,
J. H.
,
Wakefield
,
L. M.
,
Roberts
,
A. B.
,
Jakowlew
,
A.
,
Alvarez-Mon
,
M.
,
Derynck
,
R.
,
Sporn
,
M. B.
and
Fauci
,
A. S.
(
1986
).
Production of transforming growth factor-/? by human T lymphocytes and its potential role in the regulation of T cell growth
.
J. exp. Med.
163
,
1037
1050
.
Kim
,
S.-J.
,
Glick
,
A.
,
Sporn
,
M. B.
and
Roberts
,
A. B.
(
1989
).
Characterization of the promoter region of the human transforming growth factor-/?l gene
.
J. biol. Chem.
264
,
402
408
.
Klausner
,
R. D.
and
Harford
,
J. B.
(
1989
).
Cis-trans models for post-transcriptional gene regulation
.
Science
246
,
870
872
.
Knabbe
,
C.
,
Lippman
,
M. E.
,
Wakefield
,
L. M.
,
Flanders
,
K. C.
,
Kasid
,
A.
,
Derynck
,
R.
and
Dickson
,
R. B.
(
1987
).
Evidence that transforming growth factor-/? is a hormonally regulated negative growth factor in human breast cancer cells
.
Cell
48
,
417
428
.
Komm
,
B. S.
,
Terpening
,
C. M.
,
Benz
,
D. J.
,
Graeme
,
K. A.
,
Gallegos
,
A.
,
Korc
,
M.
,
Greene
,
G. L.
,
O’Malley
,
B. W.
and
Haussler
,
M. R.
(
1988
).
Estrogen binding, receptor mRNA, and biologie response in osteoblast-like osteosarcoma cells
.
Science
241
,
81
84
.
Kyprianou
,
N.
and
Isaacs
,
J. T.
(
1989
).
Expression of transforming growth factor-/! in the rat ventral prostate during castration-induced programmed cell death
.
Motee. Endocr.
3
,
1515
1522
.
Lawrence
,
D. A.
,
Bircher
,
R.
,
Kryceve-Martinerie
,
C.
and
Jullien
,
P.
(
1984
).
Normal embryo fibroblasts release transforming growth factors in a latent form
.
J. cell. Physiol.
121
,
184
188
.
Miller
,
D. A.
,
Lee
,
A.
,
Matsui
,
Y.
,
Chen
,
E. Y.
,
Moses
,
H. L.
and
Derynck
,
R.
(
1989
).
Complementary DNA cloning of the murine transforming growth factor-/53 (TGF-β3) precursor and the comparative expression of TGF-β3 and TGF-β1 messenger RNA in murine embryos and adult tissues
.
Molec. Endocr.
3
,
1926
1934
.
Miyazono
,
K.
,
Hellman
,
U.
,
Wernstedt
,
C.
and
Heldin
,
C-H.
(
1988
).
Latent high molecular weight complex of transforming growth factor j81: purification from human platelets and structural characterization
.
J. biol. Chem.
263
,
6407
6415
.
Murphy
,
L. C.
and
Dotzlaw
,
H.
(
1989
).
Regulation of transforming growth factor-alpha and transforming growth factor-/! messenger ribonucleic acid abundance in T-47D, human breast cancer cells
.
Molec. Endocr.
3
,
611
617
.
O’Connor-McCourt
,
M. D.
and
Wakefield
,
L.
(
1987
).
Latent transforming growth factor-/! in serum: a specific complex with alpha-2-macroglobulin
.
J. biol. Chem.
262
,
14 090
14 099
.
Pelton
,
R. W.
,
Nomura
,
S.
,
Moses
,
H. L.
and
Hogan
,
B. L. M.
(
1989
).
Expression of transforming growth factor fS RNA during murine embryogenesis
.
Development
106
,
759
767
.
Pfeilschifter
,
J.
and
Mundy
,
G. R.
(
1987
).
Modulation of type fi transforming growth factor activity in bone cultures by osteotropic hormones
.
Proc. natn. Acad. Sci. U.S.A.
84
,
2024
2028
.
Roberts
,
A. B.
,
Kondaiah
,
P.
,
Rosa
,
F.
,
Watanabe
,
S.
,
Good
,
P.
,
Roche
,
N. S.
,
Rebbert
,
M. L.
,
Dawid
,
I. B.
and
Sporn
,
M. B.
(
1990
).
Mesoderm induction in Xenopus laevis distinguishes between the various TGF-β isoforms
.
Growth Factors, in press
.
Roberts
,
A. B.
and
Sporn
,
M. B.
(
1990
).
The transforming growth factor-/Is
. In
Handbook of Experimental Pharmacology
(ed.
M. B.
Sporn
and
A. B.
Roberts
), vol.
95
, pp.
419
-
472
. Springer-Verlag, Heidelberg.
Sato
,
Y.
and
Rifkin
,
D. B.
(
1989
).
Inhibition of endothelial cell movement by pericytes and smooth muscle cells: activation of a latent transforming growth factor-/31-like molecule by plasmin during co-culture
.
J. Cell Biol.
109
,
309
315
.
Smits
,
H. L.
,
Floyd
,
E. E.
and
Jetten
,
A. M.
(
1987
).
Molecular cloning of sequences regulated during squamous differentiation of tracheal epithelial cells and controlled by retinoic acid
.
Molec. cell. Biol.
7
,
4017
4023
.
Sporn
,
M. B.
and
Roberts
,
A. B.
(
1984
).
In The Retinoids
(ed.
M. B.
Sporn
,
A. B.
Roberts
and
D. S.
Goodman
), vol.
1. New York: Academic Press
.
Wakefield
,
L. M.
,
Smith
,
D. M.
,
Flanders
,
K. C.
and
Sporn
,
M. B.
(
1988
).
Latent transforming growth factor-/! from human platelets: a high molecular weight complex containing precursor sequences
.
J. biol. Chem.
263
,
7646
7654
.
Wakefield
,
L. M.
,
Smith
,
D. M.
,
Masui
,
T.
,
Harris
,
C. C.
and
Sporn
,
M. B.
(
1987
).
Distribution and modulation of the cellular receptor for transforming growth factor-β J. Cell Biol.
105
,
965
975
.
Wakefield
,
L. M.
and
Sporn
,
M. B.
(
1990
).
Suppression of carcinogenesis: a role for TGF-/! and related molecules in prevention of cancer
. In
Tumor Suppressor Genes
(ed.
G.
Klein
), pp.
217
-
243
. Marcel Dekker, Inc., New York.
Wakefield
,
L. M.
,
Winokur
,
T. S.
,
Hollands
,
R. S.
,
Christopherson
,
K.
,
Levinson
,
A. D.
and
Sporn
,
M. B.
(
1990
).
Recombinant latent transforming growth factor-β1 has a longer plasma half-life in rats than active transforming growth factor-β1, and a different tissue distribution
.
J. Clin. Invest., in press
.