An Active Site of Transforming Growth Factor-β1 for Growth Inhibition and Stimulation
1999; Elsevier BV; Volume: 274; Issue: 39 Linguagem: Inglês
10.1074/jbc.274.39.27754
ISSN1083-351X
AutoresShuan Shian Huang, Mi Zhou, Frank E. Johnson, Huey‐Sheng Shieh, Jung San Huang,
Tópico(s)Growth Hormone and Insulin-like Growth Factors
ResumoTransforming growth factor-β (TGF-β) is a bifunctional growth regulator. It inhibits growth of many cell types, including epithelial cells, but stimulates growth of others (e.g. fibroblasts). The active site on the TGF-β molecule, which mediates its growth regulatory activity, has not been defined. Here, we show that antibody to a TGF-β1 peptide containing the motif WSLD (52nd to 55th amino acid residues) completely blocked both 125I-TGF-β1 binding to TGF-β receptors and TGF-β1-induced growth inhibition in mink lung epithelial cells. Site-directed mutagenesis analysis revealed that the replacement of Trp52 and Asp55 by alanine residues diminished the growth inhibitory activity of TGF-β1 by ∼90%. Finally, while wild-type TGF-β1 was able to stimulate growth of transfected NIH 3T3 cells, the double mutant TGF-β1 W52A/D55A was much less active. These results support the hypothesis that the WSLD motif is an active site of TGF-β1, which is important for growth inhibition of epithelial cells and growth stimulation of fibroblasts. Transforming growth factor-β (TGF-β) is a bifunctional growth regulator. It inhibits growth of many cell types, including epithelial cells, but stimulates growth of others (e.g. fibroblasts). The active site on the TGF-β molecule, which mediates its growth regulatory activity, has not been defined. Here, we show that antibody to a TGF-β1 peptide containing the motif WSLD (52nd to 55th amino acid residues) completely blocked both 125I-TGF-β1 binding to TGF-β receptors and TGF-β1-induced growth inhibition in mink lung epithelial cells. Site-directed mutagenesis analysis revealed that the replacement of Trp52 and Asp55 by alanine residues diminished the growth inhibitory activity of TGF-β1 by ∼90%. Finally, while wild-type TGF-β1 was able to stimulate growth of transfected NIH 3T3 cells, the double mutant TGF-β1 W52A/D55A was much less active. These results support the hypothesis that the WSLD motif is an active site of TGF-β1, which is important for growth inhibition of epithelial cells and growth stimulation of fibroblasts. transforming growth factor-β disuccinimidyl suberate Dulbecco's modified Eagle's medium polymerase chain reaction kilobase pair(s) base pair Chinese hamster ovary Transforming growth factor-β (TGF-β)1 is a family of 25-kDa structurally homologous dimeric proteins containing one interchain and four intrachain disulfide bonds (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). It is a bifunctional growth regulator, inhibiting cell growth of most cell types (including epithelial cells, endothelial cells, smooth muscle cells, and lymphocytes) but stimulating proliferation of others (such as fibroblasts) (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). TGF-β has many other biological activities, such as stimulation of extracellular matrix biosynthesis, angiogenesis, and differentiation of several cell lineages (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). It has been implicated in the processes of wound repair and morphogenesis (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). Isoforms in mammalian species, including TGF-β1, TGF-β2, and TGF-β3, exhibit ∼70% sequence homology and have similar biological activities (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). However, activities of the isoforms differ in certain cell types or systems (4Jennings J.C. Mohan S. Linkhart T.A. Widstrom R. Baylink D.J. J. Cell. Physiol. 1988; 137: 167-172Crossref PubMed Scopus (156) Google Scholar, 5Merwin J.R. Newman W. Becall L.D. Tucker A. Madri J. Am. J. Pathol. 1991; 138: 37-51PubMed Google Scholar, 6Qian S.W. Burmester J.K. Sun P.D. Huang A. Ohlsen D.J. Suardet L. Flanders K.C. Davies D. Roberts A.B. Sporn M.B. Biochemistry. 1994; 33: 12298-12304Crossref PubMed Scopus (16) Google Scholar, 7Shah M. Foreman D.M. Ferguson M.W.J. J. Cell Sci. 1995; 108: 985-1002Crossref PubMed Google Scholar). For example, TGF-β1 is more potent than TGF-β2 in inhibiting growth of endothelial cells (4Jennings J.C. Mohan S. Linkhart T.A. Widstrom R. Baylink D.J. J. Cell. Physiol. 1988; 137: 167-172Crossref PubMed Scopus (156) Google Scholar, 5Merwin J.R. Newman W. Becall L.D. Tucker A. Madri J. Am. J. Pathol. 1991; 138: 37-51PubMed Google Scholar, 6Qian S.W. Burmester J.K. Sun P.D. Huang A. Ohlsen D.J. Suardet L. Flanders K.C. Davies D. Roberts A.B. Sporn M.B. Biochemistry. 1994; 33: 12298-12304Crossref PubMed Scopus (16) Google Scholar), while TGF-β3 antagonizes the activities of TGF-β1 and TGF-β2 in an animal model system of wound healing (7Shah M. Foreman D.M. Ferguson M.W.J. J. Cell Sci. 1995; 108: 985-1002Crossref PubMed Google Scholar). TGF-β2 appears to bind to α2-macroglobulin more strongly than TGF-β1(8Danielpour D. Sporn M.B. J. Biol. Chem. 1990; 265: 6973-6977Abstract Full Text PDF PubMed Google Scholar). The molecular basis of the different activities of TGF-β isoforms is not well understood (4Jennings J.C. Mohan S. Linkhart T.A. Widstrom R. Baylink D.J. J. Cell. Physiol. 1988; 137: 167-172Crossref PubMed Scopus (156) Google Scholar, 5Merwin J.R. Newman W. Becall L.D. Tucker A. Madri J. Am. J. Pathol. 1991; 138: 37-51PubMed Google Scholar, 6Qian S.W. Burmester J.K. Sun P.D. Huang A. Ohlsen D.J. Suardet L. Flanders K.C. Davies D. Roberts A.B. Sporn M.B. Biochemistry. 1994; 33: 12298-12304Crossref PubMed Scopus (16) Google Scholar, 7Shah M. Foreman D.M. Ferguson M.W.J. J. Cell Sci. 1995; 108: 985-1002Crossref PubMed Google Scholar, 8Danielpour D. Sporn M.B. J. Biol. Chem. 1990; 265: 6973-6977Abstract Full Text PDF PubMed Google Scholar). X-ray crystallographic and nuclear magnetic resonance spectroscopic analyses of TGF-β1 and TGF-β2 have revealed generally similar three-dimensional structures, with differences in certain regions of the molecule (9Schlunegger M.P. Grütter M.G. Nature. 1992; 353: 430-434Crossref Scopus (285) Google Scholar, 10Schlunegger M.P. Grütter M.G. J. Mol. Biol. 1993; 231: 445-458Crossref PubMed Scopus (63) Google Scholar, 11Hinck A.P. Archer S.J. Qian S.W. Roberts A.B. Sporn M.B. Weatherbee J.A. Tsang M.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1996; 35: 8517-8534Crossref PubMed Scopus (154) Google Scholar, 12Daopin S. Piez K.A. Ogawa Y. Davies D.R. Science. 1992; 257: 369-373Crossref PubMed Scopus (376) Google Scholar, 13Archer S.J. Box A. Roberts A.B. Sporn M.B. Ogawa Y. Piez K.A. Weatherbee J.A. Tsang H.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1993; 32: 1164-1174Crossref PubMed Scopus (53) Google Scholar, 14Archer S.J. Box A. Roberts A.B. Sporn M.B. Ogawa Y. Piez K.A. Weatherbee J.A. Tsang H.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1993; 32: 1152-1163Crossref PubMed Scopus (40) Google Scholar). Using a domain swap approach, Qian and her co-workers (15Qian S.W. Burmester J.K. Merwin J.R. Madri J.A. Sporn M.B. Roberts A.B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6290-6294Crossref PubMed Scopus (44) Google Scholar) demonstrated that residues 40–82 play important roles in the activity of a particular TGF-β isoform. They subsequently showed that residues 45 and 47 determine the binding affinities of TGF-β isoforms toward α2-macroglobulin (16Burmester J.K. Qian S.W. Roberts A.B. Huang A. Amatayakul-Chantler S. Suardet L. Odartchenko N. Madri J.A. Sporn M.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8628-8632Crossref PubMed Scopus (37) Google Scholar) and that residues 91–96 are important in the interaction of TGF-β1 with soluble type II TGF-β receptor (17Qian S.W. Burmester J.K. Tsang M.L.-S. Weatherbee J.A. Hinck A.P. Ohlsen D.J. Sporn M.B. Roberts A.B. J. Biol. Chem. 1996; 271: 30656-30662Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar,18Burmester J.K. Qian S.W. Ohlsen D. Phan S. Sporn M.B. Roberts A.B. Growth Factors. 1998; 15: 231-242Crossref PubMed Scopus (22) Google Scholar). Recently, we determined the TGF-β antagonist activities of synthetic pentacosapeptides with overlapping amino acid sequences covering most of the TGF-β1 molecule (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Of these seven pentacosapeptides, only the one containing residues 41–65 of TGF-β1 (termed β125-(41–65)) exhibited potent TGF-β antagonist activity (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). The replacement of both residues Trp52 and Asp55 by alanine completely abrogated the TGF-β antagonist activity of this pentacosapeptide (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Multiple conjugation of β125-(41–65) to the carrier proteins bovine serum albumin and carbonic anhydrase conferred TGF-β agonist activity as measured by growth inhibition (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). These results led us to identify structurally unrelated TGF-β partial agonists, which also possess two or more WXXD motifs per dimer or molecule (20Leal S.M. Liu Q. Huang S.S. Huang J.S. J. Biol. Chem. 1997; 272: 20572-20576Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 21Huang S.S. Cerullo M.A. Huang F.W. Huang J.S. J. Biol. Chem. 1998; 273: 26030-26041Google Scholar). Based on these studies, we hypothesized that the WSLD motif (52nd to 55th amino acid residues of TGF-β1) is an active site that interacts with TGF-β receptors; this interaction leads to growth inhibition. To test this hypothesis, we prepared an antibody that specifically reacts with β125-(41–65) and determined its effect on the biological activities of TGF-β1. In addition, we generated TGF-β1 mutant proteins in which residues Trp52 and/or Asp55 are replaced by alanine and compared the biological activity of wild type TGF-β1 with those of the mutants. In this communication, we show that this antibody blocks 125I-TGF-β binding to TGF-β receptors and abolishes TGF-β1-induced growth inhibition in mink lung epithelial cells (Mv1Lu cells). We also demonstrate that the TGF-β1 mutant in which both Trp52 and Asp55 are replaced with alanine residues has diminished growth inhibitory activity in Mv1Lu cells and has diminished ability to induce autocrine growth of transfected NIH 3T3 cells. Na125I (17 Ci/mg) and [methyl-3H]thymidine (67 Ci/mmol) were purchased from ICN Radiochemicals (Irvine, CA). High molecular mass protein standards (myosin, 205 kDa; β-galactosidase, 116 kDa; phosphorylase, 97 kDa; bovine serum albumin, 66 kDa) were purchased from Sigma. Disuccinimidyl suberate (DSS) was obtained from Pierce. TGF-β1 was purchased from Austral Biologicals (San Ramon, CA). Pan-specific TGF-β neutralizing antibody and porcine TGF-β1 were purchased from R & D Systems, Inc. (Minneapolis, MN). TGF-β1 immunoassay kit was obtained from Promega (Madison, WI). The pentacosapeptide β125-(41–65), whose amino acid sequence corresponds to the 41st to 65th amino acid residues of TGF-β1, was prepared as described previously (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Porcine TGF-β1 cDNAs, pPTGFbeta1 and pSQneo-TGF-β1, were obtained from American Type Culture Collection (Manassas, VA) and Drs. Su Wen Qian and Anita B. Roberts (NCI), respectively. Diff-Quik solutions were obtained from American Scientific Products (McGraw Park, IL). β125-(41–65) was conjugated to bovine thyroglobulin using glutaraldehyde according to the procedure described previously (22Huang S.S. Huang J.S. J. Biol. Chem. 1988; 263: 12608-12618Abstract Full Text PDF PubMed Google Scholar). β125-(41–65)-thyroglobulin conjugate was injected subcutaneously with adjuvants into rabbits for generation of antiserum (22Huang S.S. Huang J.S. J. Biol. Chem. 1988; 263: 12608-12618Abstract Full Text PDF PubMed Google Scholar). The antibody (IgG) to β125-(41–65) was purified with protein A-Sepharose using standard procedures. Its specificity was verified by Western blot analysis and enzyme-linked immunosorbent assay. 125I-TGF-β1 was prepared by iodination of TGF-β1 with Na125I and chloramine T as described previously (20Leal S.M. Liu Q. Huang S.S. Huang J.S. J. Biol. Chem. 1997; 272: 20572-20576Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 23Liu Q. Huang S.S. Huang J.S. . 1997; 272: 18891-18895Google Scholar). The specific radioactivity of 125I-TGF-β1 was 1–3 × 105 cpm/ng. Mv1Lu cells grown on 24-well clustered dishes were incubated with 0.1 nm125I-TGF-β1 and various concentrations of antibody to β125-(41–65) both with and without 10 μm β125-(41–65), a specific TGF-β peptide antagonist, in binding buffer (20Leal S.M. Liu Q. Huang S.S. Huang J.S. J. Biol. Chem. 1997; 272: 20572-20576Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 23Liu Q. Huang S.S. Huang J.S. . 1997; 272: 18891-18895Google Scholar).125I-TGF-β1 was preincubated with antibody to β125-(41–65) at room temperature for 1 h prior to incubation with cells. After 2.5 h at 0 °C, the cells were washed with binding buffer, and the cell-associated radioactivity was determined. The specific binding of125I-TGF-β1 to TGF-β receptors in Mv1Lu cells was estimated by subtracting nonspecific binding (obtained in the presence of 10 μmβ125-(41–65)) from total binding. All experiments were carried out in quadruplicate. Mv1Lu cells grown on 35-mm Petri dishes were incubated with 0.1 nm125I-TGF-β1 in the presence of various concentrations of control IgG or antibody to β125-(41–65) or 10 μmβ125-(41–65) (for measurement of nonspecific binding) in binding buffer (20Leal S.M. Liu Q. Huang S.S. Huang J.S. J. Biol. Chem. 1997; 272: 20572-20576Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 23Liu Q. Huang S.S. Huang J.S. . 1997; 272: 18891-18895Google Scholar). After 2.5 h at 0 °C,125I-TGF-β1-affinity labeling was carried out in the presence of DSS (20Leal S.M. Liu Q. Huang S.S. Huang J.S. J. Biol. Chem. 1997; 272: 20572-20576Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 23Liu Q. Huang S.S. Huang J.S. . 1997; 272: 18891-18895Google Scholar). The125I-TGF-β1-affinity-labeled TGF-β receptors were analyzed by 5% SDS-polyacrylamide gel electrophoresis under reducing conditions and autoradiography. Various concentrations of TGF-β1 and 50 μg/ml antibody to β125-(41–65) or control IgG were preincubated for 1 h at room temperature and then added to Mv1Lu cells grown on 24-well clustered dishes with DMEM containing 0.1% fetal calf serum. After 16 h at 37 °C, the cells were pulsed with 1 μCi/ml [methyl-3H]thymidine for 4 h. The cells were then washed twice with 10% trichloroacetic acid and once with ethanol:ether (2:1, v/v). The [methyl-3H]thymidine incorporation into cellular DNA was then determined by liquid scintillation counting. A two-step PCR procedure was used to generate point mutations. Primers included: A, 5′ GAATTCAGATCTGAGATGGCGCCTTCGGGGCTGC (TGF-β1 906–918); B, 5′ GAATTCAGATCTTCAGCTGCACTTGCAGGAACGC (TGF-β12057–2078); C, 5′ CCCTACATCGCCAGCCTAGACACT; D, 5′ CTGAGTGTCTAGGCTGGCGATGTA; E, 5′ GGAGCCTAGCCACTCAGTACAGCAAGG; F, 5′ GCTGTACTGAGTGGCTAGGCTCCAGATG; G, 5′ CCCTACATCGCGAGCCTAGCCACTCAGTAC; H, 5′ GTACTGAGTGGCTAGGCTCGCGATGTAGGG; I, 5′ CCTACATCTGGGCGCTAGACACTCAG; J, 5′CTGAGTGTCTAGCGCCCAGATGTAGG. The underlined nucleotides indicate the mutations. To introduce the tryptophan 52 to alanine (W52A) mutation, porcine TGF-β1 cDNA (pPTGFbeta1 from American Type Culture Collection) and primers A/D and C/B were used in the first PCR reaction to generate ∼1-kb and 200-bp fragments, respectively. The reaction mixture was treated with 5 units of Klenou fragment and gel-purified. The ∼1-kb and 200-bp fragments were then used as templates in the second PCR reaction using Primers A and B to generate the full-length cDNA. The PCR product was cloned into the pT7 Blue T-Vector, and the plasmid was purified. The purified plasmid was cut with BsgI to produce a 340-base pair fragment containing the mutated sequence. This 340-base pair fragment was then ligated to pSV·SPORT 1-TGF-β1, which was digested with BsgI to delete the corresponding fragment. This ligated product was purified and identified. For D55A, S53A, and W52A/D55A mutants, the above procedure was repeated, but using A/F and E/B for D55A mutation, A/J and I/B for S53A mutation, and A/H and G/B for W52A/D55A mutation. The plasmids containing wild-type TGF-β1 and TGF-β1 mutant cDNAs in pSV·SPORT 1 were then cut with KpnI and XbaI. The ∼1.6-kb cDNA insert was then ligated to the expression vector pMSXND. NIH 3T3 and Chinese hamster ovary (CHO) cells were transfected with pMSXND vector only or pMSXND containing the insert of wild-type or mutant TGF-β1 cDNA using the calcium phosphate-transfection method. NIH 3T3 and CHO cells stably expressing vector only, wild-type TGF-β1 cDNA, or mutant TGF-β1 cDNAs were selected with 800 μg/ml G418. Four or more clones for the expression of each construct were isolated. The production of wild-type and mutant TGF-β1was carried out according to the procedure of Qian et al.(15Qian S.W. Burmester J.K. Merwin J.R. Madri J.A. Sporn M.B. Roberts A.B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6290-6294Crossref PubMed Scopus (44) Google Scholar). After acid activation and lyophilization, the conditioned media were subjected to reverse-phase high performance liquid chromatography (C8 column) using a linear gradient of acetonitrile (0 to 70%) in 10% trifluoroacetic acid. Wild-type and mutant TGF-β1 were eluted at ∼30% acetonitrile. Wild-type and mutant TGF-β1 were quantitated by enzyme-linked immunoassay using the protocol provided by the manufacturer (Promega, Madison, WI). During the assay, an internal standard of porcine TGF-β1was included in the samples. The putative active-site motif (WSLD) is located in a loop (residues 46–56) of the TGF-β molecule, which is accessible to solvent (9Schlunegger M.P. Grütter M.G. Nature. 1992; 353: 430-434Crossref Scopus (285) Google Scholar, 10Schlunegger M.P. Grütter M.G. J. Mol. Biol. 1993; 231: 445-458Crossref PubMed Scopus (63) Google Scholar, 11Hinck A.P. Archer S.J. Qian S.W. Roberts A.B. Sporn M.B. Weatherbee J.A. Tsang M.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1996; 35: 8517-8534Crossref PubMed Scopus (154) Google Scholar, 12Daopin S. Piez K.A. Ogawa Y. Davies D.R. Science. 1992; 257: 369-373Crossref PubMed Scopus (376) Google Scholar, 13Archer S.J. Box A. Roberts A.B. Sporn M.B. Ogawa Y. Piez K.A. Weatherbee J.A. Tsang H.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1993; 32: 1164-1174Crossref PubMed Scopus (53) Google Scholar, 14Archer S.J. Box A. Roberts A.B. Sporn M.B. Ogawa Y. Piez K.A. Weatherbee J.A. Tsang H.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1993; 32: 1152-1163Crossref PubMed Scopus (40) Google Scholar). We predicted that antibody raised to β125-(41–65) containing this motif would block both TGF-β1 binding to TGF-β receptors and the biological activities of TGF-β1. To test this, we prepared rabbit antiserum to β125-(41–65) using the thyroglobulin conjugate of β125-(41–65) as antigen. The effect of antibody to β125-(41–65) on125I-TGF-β1 binding to TGF-β receptors in Mv1Lu cells was then determined. As shown in TableI, the antiserum to β125-(41–65) specifically immunoprecipitated125I-TGF-β1, while nonimmune serum did not significantly immunoprecipitate 125I-TGF-β1. The purified antibody to β125-(41–65) quantitatively inhibited 125I-TGF-β1 binding to Mv1Lu cells (Fig. 1 A). At ∼75 μg/ml, the antibody to β125-(41–65) completely abrogated the specific binding of 0.1 nm125I-TGF-β1 to TGF-β receptors in Mv1Lu cells. By contrast, control IgG did not show any effect on125I-TGF-β1 binding to cells at the same concentration. The 125I-TGF-β1-affinity labeling analysis revealed that, like β125-(41–65), a TGF-β antagonist, the antibody to β125-(41–65) effectively blocked125I-TGF-β1 binding to all TGF-β receptor types, including type I, II, III, and V TGF-β receptors (TβR-I, TβR-II, TβR-III, TβR-V) (Fig. 1 B, lanes 3and 4). To define the functional significance of the inhibition of TGF-β1 binding to TGF-β receptors, we determined the effect of antibody to β125-(41–65) on TGF-β1-induced growth inhibition of Mv1Lu cells, as measured by [methyl-3H]thymidine incorporation into cellular DNA. As shown in Fig. 1 C, the antibody blocked TGF-β1-induced inhibition of DNA synthesis in Mv1Lu cells. At 50 μg/ml, the antibody to β125-(41–65) completely blocked the inhibition of DNA synthesis induced by 0.25–8 pmTGF-β1 in these cells. The control IgG did not affect the TGF-β1-induced inhibition of DNA synthesis. These results are consistent with the reports that the loop (residues 46–56, which contain the WSLD motif) included in amino acid residues 41–65 is exposed (9Schlunegger M.P. Grütter M.G. Nature. 1992; 353: 430-434Crossref Scopus (285) Google Scholar, 10Schlunegger M.P. Grütter M.G. J. Mol. Biol. 1993; 231: 445-458Crossref PubMed Scopus (63) Google Scholar, 11Hinck A.P. Archer S.J. Qian S.W. Roberts A.B. Sporn M.B. Weatherbee J.A. Tsang M.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1996; 35: 8517-8534Crossref PubMed Scopus (154) Google Scholar, 12Daopin S. Piez K.A. Ogawa Y. Davies D.R. Science. 1992; 257: 369-373Crossref PubMed Scopus (376) Google Scholar, 13Archer S.J. Box A. Roberts A.B. Sporn M.B. Ogawa Y. Piez K.A. Weatherbee J.A. Tsang H.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1993; 32: 1164-1174Crossref PubMed Scopus (53) Google Scholar, 14Archer S.J. Box A. Roberts A.B. Sporn M.B. Ogawa Y. Piez K.A. Weatherbee J.A. Tsang H.L.-S. Lucas R. Zhang B.-L. Wenker J. Torchia D.A. Biochemistry. 1993; 32: 1152-1163Crossref PubMed Scopus (40) Google Scholar). The results also suggest that this loop may be involved in the interaction of TGF-β1 with TGF-β receptors.Table IImmunoprecipitation of 125I-TGF-β1 by antiserum to β125-(41–65)125I-TGF-β1immunoprecipitatedcpmControl2,741 ± 101+Non-immune serum3,575 ± 821+Antiserum to β125-(41–65)13,092 ± 1,714125I-TGF-β1 (0.1 ng) was incubated with 5 μl of antiserum to β125-(41–65) or non-immune serum in 100 μl of 25 mm Hepes buffer, pH 7.4, or 0.15 mNaCl containing 0.2% bovine serum albumin. After 8 h at 4 °C, the immunocomplexes were precipitated with protein A-Sepharose (1.5 h, 4 °C), washed, and counted. The experiments were carried out in triplicate. Open table in a new tab 125I-TGF-β1 (0.1 ng) was incubated with 5 μl of antiserum to β125-(41–65) or non-immune serum in 100 μl of 25 mm Hepes buffer, pH 7.4, or 0.15 mNaCl containing 0.2% bovine serum albumin. After 8 h at 4 °C, the immunocomplexes were precipitated with protein A-Sepharose (1.5 h, 4 °C), washed, and counted. The experiments were carried out in triplicate. To confirm that the WSLD motif (52nd to 55th residues) is important in the interaction of TGF-β1 with TGF-β receptors, we generated porcine wild-type TGF-β1 and TGF-β1 mutants in which Trp52, Ser53, and/or Asp55 were replaced by alanine residues by stably expressing their cDNA constructs in NIH 3T3 and CHO cells. The residue Leu54 was not replaced, since the corresponding residue is not conserved in TGF-β2, which has the motif WSSD. The growth inhibitory activities of wild-type and mutant TGF-β1 purified from the culture media of transfected NIH 3T3 and CHO cells were determined by measuring their inhibitory activity on DNA synthesis in Mv1Lu cells. As shown in Fig. 2, both wild-type TGF-β1 and TGF-β1 S53A mutant (in which the 53rd residue, serine, is replaced by alanine) inhibited [methyl-3H]thymidine incorporation into cellular DNA with identical IC50 of 0.3 ± 0.1 pm and 0.3 ± 0.2 pm (mean ± S.E.,n = 7 experiments), respectively. In contrast, the TGF-β1 mutants TGF-β1 W52A, TGF-β1 D55A, and TGF-β1 W52A/D55A had diminished activities in inhibition of DNA synthesis of Mv1Lu cells with IC50 of 0.6 ± 0.2, 1.3 ± 0.4, and 3.0 ± 0.8 pm (mean ± S.E., n = 7 experiments), respectively. The double mutant TGF-β1W52A/D55A appeared to possess only ∼10% of the activity of wild-type TGF-β1. These results suggest that theWSLD motif plays an important role in the activity of TGF-β1. TGF-β is a bifunctional growth regulator (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). It inhibits cell growth of most cell types but stimulates growth of other cell types such as NIH 3T3 cells (1Massagué J. Annu. Rev. Cell Biol. 1990; 6: 597-641Crossref PubMed Scopus (3004) Google Scholar, 2Roberts A.B. Sporn M.B. Sporn M.B. Roberts A.B. in Handbook of Experimental Pharmacology: Peptide Growth Factors and Their Receptors. Springer-Verlag, New York1990: 419-472Google Scholar, 3Wright J.A. Turley E.A. Greenberg A.H. Crit. Rev. Oncog. 1993; 4: 473-492PubMed Google Scholar). During culture of NIH 3T3 cells stably transfected with wild-type TGF-β1 cDNA, we noticed that these transfected cells proliferated faster than NIH 3T3 cells stably transfected either with vector only or with TGF-β W52A/D55A cDNA. To see if the faster growth of the transfected cells is due to autocrine stimulation by the transfected wild-type TGF-β1, we determined the effect of neutralizing antibody to TGF-β1 on growth of NIH 3T3 cells stably transfected with wild-type TGF-β1 cDNA. As shown in Fig.3, A, C, andE, NIH 3T3 cells expressing wild-type TGF-β1proliferated faster than cells expressing TGF-β1W52A/D55A and control cells (transfected with vector only) as determined by an autocrine growth assay in which autocrine-stimulated cells grow to form larger size colonies compared with cells without this stimulation. The order of the growth rates is: cells expressing wild-type TGF-β1 > cells expressing TGF-β1W52A/D55A > control cells (Fig. 3, A, C, and E). Both cells expressing wild-type TGF-β1and TGF-β1 W52A/D55A exhibited similar levels of wild-type and mutant TGF-β1 transcripts (data not shown). On the other hand, culture of NIH 3T3 cells expressing wild-type TGF-β1 and TGF-β1 W52A/D55A in the presence of a neutralizing antibody to TGF-β1 had diminished growth rates, which were comparable with that of NIH 3T3 cells transfected with vector only (Fig. 3, B, D, andF). These results indicate that the faster proliferation of NIH 3T3 cells expressing wild-type TGF-β1 is due to autocrine stimulation by expressed wild-type TGF-β1. These results also suggest that the TGF-β1 W52A/D55A mutant has significantly diminished ability to induce autocrine growth of transfected NIH 3T3 cells when compared with wild-type TGF-β1. We have shown that specific antibody to β125-(41–65) at ∼75 μg/ml completely blocks 125I-TGF-β1 binding to TGF-β receptors and also blocks TGF-β1-induced inhibition of DNA synthesis in Mv1Lu cells. The antibody reported here appears to be more potent than those reported by Flanders et al. (24Flanders K.C. Roberts A.B. Ling N. Fleurdelys B.E. Sporn M.B. Biochemistry. 1988; 27: 739-746Crossref PubMed Scopus (76) Google Scholar). Of their five antisera to TGF-β1 peptide fragments, those directed against residues 50–75 and 78–109 blocked only 40 and 80% of 125I-TGF-β1 binding to cells, respectively, at 450 μg/ml. So far, only antibodies raised to the intact TGF-β dimer have been shown to completely block TGF-β activities (24Flanders K.C. Roberts A.B. Ling N. Fleurdelys B.E. Sporn M.B. Biochemistry. 1988; 27: 739-746Crossref PubMed Scopus (76) Google Scholar, 25Wang J. Sun L. Myeroff L. Wang X. Gentry L.E. Yang J. Liang J. Zborowska E. Markowitz S. Wilson J.K.V. Brattain M.G. J. Biol. Chem. 1995; 270: 22044-22049Abstract Full Text Full Text PDF PubMed Scopus (328) Google Scholar). The potent TGF-β antagonist activity of our antibody to β125-(41–65) strongly suggests that the region of the amino acid residues 41–65 of TGF-β1 is exposed and may be involved in receptor binding. The putative active site WSLD of TGF-β1, which is included in the amino acid sequence of β125-(41–65), was proposed based on the following observations. 1) Of seven synthetic pentacosapeptides with overlapping amino acid sequences covering most of the TGF-β1 molecule, only the peptide of residues 41–65 (β125-(41–65)) exhibits potent TGF-β antagonist activity (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). 2) The β125-(41–65) structural variant β125-(41–65) W52A/D55A, in which both Trp52 and Asp55 are replaced by alanine residues, does not show a significant TGF-β antagonist activity (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). 3) Multiple conjugation of β125-(41–65) to carrier proteins confers TGF-β agonist activity as measured by growth inhibition but not transcriptional activation (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Here we show that the double mutations of Trp52 and Asp55 and single mutation of either Trp52 or Asp55 of TGF-β1 diminish the growth inhibitory activity of TGF-β1 by 90 and 50–75%, respectively. Not only does the double mutant TGF-β1 W52A/D55A possess only 10% activity of wild-type TGF-β1, but it also appears to have significantly diminished ability to stimulate growth of transfected NIH 3T3 cells by an autocrine mechanism. These results support the hypothesis that the WSLD is an active site of TGF-β1(Fig. 4). However, the inability of the double mutations of Trp52 and Asp55 to completely abolish the TGF-β activities implies that there are other binding sites in addition to the WSLD site. One candidate appears to be the region of residues 91–96, since this region is important for TGF-β1 binding to TβR-II in solution and TGF-β receptors in cells (17Qian S.W. Burmester J.K. Tsang M.L.-S. Weatherbee J.A. Hinck A.P. Ohlsen D.J. Sporn M.B. Roberts A.B. J. Biol. Chem. 1996; 271: 30656-30662Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar) and since antibodies to residues 78–109 block TGF-β1 binding to TGF-β receptors in cells (23Liu Q. Huang S.S. Huang J.S. . 1997; 272: 18891-18895Google Scholar). The three-dimensional configuration of residues 91–96 seems to be required to constitute this binding site, because a synthetic pentacosapeptide β125-(81–105) with amino acid residues 81–105 of TGF-β1 does not show any TGF-β antagonist activity (18Burmester J.K. Qian S.W. Ohlsen D. Phan S. Sporn M.B. Roberts A.B. Growth Factors. 1998; 15: 231-242Crossref PubMed Scopus (22) Google Scholar). It is noteworthy that the proposed receptor binding sites (Fig. 4), which are contributed by the two monomers in TGF-β1, are similar to those of vascular endothelial cell growth factor; the vascular endothelial cell growth factor receptor is in contact with both subunits of the ligand, a disulfide-linked homodimer protein (26Wiesmann C. Fuh G. Christinger H.W. Eigenbrot C. Wells J.A. de Vos A.M. Cell. 1997; 91: 696-704Abstract Full Text Full Text PDF Scopus (415) Google Scholar). The hypothesis of the two major binding sites (the WSLD motif and a C-terminal site, residues 91–96), which are contributed by the two TGF-β1 monomers, is supported by the following observations: 1) the formation of disulfide-linked dimers of TGF-β1 is important for its activities (27Amatayakul-Chantler S. Qian S.W. Gakenheimer K. Böttinger E.P. Roberts A.B. Sporn M.B. J. Biol. Chem. 1994; 269: 27687-27691Abstract Full Text PDF PubMed Google Scholar); 2) antibodies to these two binding sites block TGF-β1binding to TGF-β receptors and TGF-β1-stimulated activities (Ref. 24Flanders K.C. Roberts A.B. Ling N. Fleurdelys B.E. Sporn M.B. Biochemistry. 1988; 27: 739-746Crossref PubMed Scopus (76) Google Scholar and this study); 3) like TGF-β2, TGF-β1/TGF-β2 (92–98) hybrids fail to bind to the soluble type II TGF-β receptor (18Burmester J.K. Qian S.W. Ohlsen D. Phan S. Sporn M.B. Roberts A.B. Growth Factors. 1998; 15: 231-242Crossref PubMed Scopus (22) Google Scholar); and 4) the TGF-β1 W52A/D55A mutant exhibits diminished growth regulatory activities as compared with those of wild type TGF-β1 (this study). The importance of theWLSD motif in mediating growth regulatory activity of TGF-β1 is also supported by the findings that several structurally unrelated proteins that contain two or moreW XX D motifs per dimer or molecule show TGF-β agonist activity in growth inhibition (20Leal S.M. Liu Q. Huang S.S. Huang J.S. J. Biol. Chem. 1997; 272: 20572-20576Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar, 21Huang S.S. Cerullo M.A. Huang F.W. Huang J.S. J. Biol. Chem. 1998; 273: 26030-26041Google Scholar, 28Leal S.M. Huang S.S. Huang J.S. J. Biol. Chem. 1999; 274: 6711-6717Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar) and that β125-(41–65)-protein conjugates containing multiple WLSD motifs exhibit growth inhibitory activity (19Huang S.S. Liu Q. Johnson F.E. Konish Y. Huang J.S. J. Biol. Chem. 1997; 272: 27155-27159Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). We thank Drs. Su Wen Qian and Anita B. Roberts, NCI, for providing porcine TGF-β1 cDNA (in pSQneo vector). We also thank John McAlpin for typing the manuscript.
Referência(s)