Human transforming growth factor-beta 3: recombinant expression, purification, and biological activities in comparison with transforming growth factors-beta 1 and -beta 2

Recent cDNA characterization has predicted the existence of a new member of the transforming growth factor family, transforming growth factor-beta 3 (TGF beta 3). However, nothing is known about the biological activities of the TGF beta 3 protein, since it has not been purified from any natural sources. We report here the recombinant expression in mammalian cells and the purification to apparent homogeneity of human TGF beta 3. The TGF beta 3 was evaluated in comparison with purified TGF beta 1 and TGF beta 2 in several assays for its effects on stimulation or inhibition of proliferation of mammalian cells. These analyses revealed that TGF beta 3 exerts activities similar to the two other TGF beta species, but that there are distinct differences in potencies between the different TGF beta forms depending on the cell type and assay used.     

Type beta transforming growth factor from feline sarcoma virus-transformed rat cells. Isolation and biological properties

We have isolated a strongly mitogenic, type beta transforming growth factor (beta TGF) released by Snyder-Theilen feline sarcoma virus-transformed rat embryo (FeSV-Fre) cells that induces phenotypic transformation of normal NRK cells when they are concomitantly stimulated by analogues of epidermal growth factor (EGF). Molecule filtration chromatography separates beta TGF from an EGF-like TGF (eTGF) which is also present in acid extracts from medium conditioned by FeSV-Fre cells (J. Massagué, (1983) J. Biol. Chem. 258, 13606-13613). Final purification of beta TGF is achieved by reverse phase high pressure liquid chromatography (HPLC) on octadecyl support, molecular filtration HPLC, and nonreducing dodecyl sulfate-polyacrylamide gel electrophoresis steps, yielding a 300,000-fold purified polypeptide with a final recovery of 21%. The purified rat beta TGF consists of two Mr = 11,000-12,000 polypeptide chains disulfide-linked as a Mr = 23,000 dimer. Induction of anchorage-independent proliferation of NRK cells by rat beta TGF depends on the simultaneous presence of eTGF or EGF. In the presence of a saturating (300 pM) concentration of either rat eTGF or mouse EGF, half-maximal anchorage-independent proliferation of NRK cells is obtained with 4-6 pM rat beta TGF. In the presence of a saturating (20 pM) concentration of rat beta TGF, half-maximal anchorage-independent proliferation of NRK cells is obtained with either rat eTGF or mouse EGF at a 50-70 pM concentration. Rat beta TGF is also able to induce DNA synthesis and cell proliferation on growth-arrested NRK, human lung, and Swiss mouse 3T3 fibroblast monolayers, this effect being half-maximal at 2-3 pM beta TGF for NRK cells. These results identify eTGF and beta TGF as the two synergistically acting factors responsible for the transforming action of culture fluids from FeSV-Fre cells.     

Partial purification of an immunosuppressive protein from a human tumor cell line and analysis of its relationship to transforming growth factor beta

The A673 human rhabdomyosarcoma cell line constitutively produces an acid-soluble, potent immunosuppressive factor (ISF), which inhibits T-cell proliferation. We have partially purified this factor from the culture supernatant of A673 cells by a sequence of acid extraction, gel filtration, cation exchange chromatography, and reverse-phase HPLC. Characterization studies indicate that ISF is similar or identical to transforming growth factor beta (TGF beta). ISF exhibits a molecular weight of 25 kDa in sodium dodecyl sulfate-polyacrylamide gels. ISF, like TGF beta, is a very basic protein (pI = 9.5) that is sensitive to reduction. Anti-TGF beta 1 antibodies completely block ISF activity in the thymocyte assay. Furthermore, ISF, like TGF beta, stimulated the anchorage-independent growth of normal rat kidney fibroblasts in soft agar.     

Comparison of the biological actions of TGF beta-1 and TGF beta-2: differential activity in endothelial cells

Beta transforming growth factor (TGF beta) has multiple in vitro biological effects including stimulation or inhibition of proliferation of specific cell types. A second major form of TGF beta, TGF beta-2, has recently been isolated from porcine platelets, from bovine bone matrix, and from several other sources. The two forms of TGF beta are biologically equipotent with the exception that TGF beta-2 was much less active than TGF beta-1 for inhibition of proliferation of a rat pleuripotent hematopoietic stem cell line. During the purification of beta TGF from bone, we obtained two fraction pools that differed in their ability to inhibit 3H-thymidine incorporation into aortic endothelial cells (AEC). We therefore compared highly purified TGF beta-1 and TGF beta-2 isolated from porcine platelets for inhibition of DNA synthesis in mink lung epithelial cells (MvILu), and in AEC, and for stimulation of 3H-thymidine incorporation in calvarial bone cells (CBC) in 3 experiments. TGF beta-1 and TGF beta-2 inhibited cell proliferation in MvILu with no significant differences in the ED50 (31 +/- 8 pg/ml vs 23 +/- 7). TGF beta-2 was much less potent than TGF beta-1 in inhibiting DNA synthesis in AEC (6310 +/- 985 pg/ml vs 101 +/- 34). The reduced specific activity of TGF beta-2 was also observed in adrenal capillary endothelial cells. Both beta-1 and beta-2 stimulated proliferation of CBC (ED50 26 +/- 2 pg/ml vs 10 +/- 4). We also examined the specificity of the MvILu and AEC inhibition assays. Epidermal growth factor (EGF), platelet derived growth factor (PDGF), acidic and basic fibroblast growth factors (FGF), skeletal growth factor (SGF)/insulin-like growth factor-II (IGF-II), and insulin-like growth factor-I (IGF-I) did not inhibit DNA synthesis in either assay system. However, when the growth factors were added to maximal inhibiting concentrations of TGF beta-1, both acidic and basic FGF significantly reduced TGF beta-1 inhibition in AEC. We conclude that (1) inhibition of DNA synthesis in endothelial cells is relatively specific for TGF beta-1, (2) inhibition of DNA synthesis in MvILu is a sensitive and specific assay for generic TGF beta activity but does not distinguish beta-1 from beta-2, (3) the relative inhibition of DNA synthesis in MvILu and AEC may provide a means to quantitatively estimate TGF beta-1 and TGF beta-2, and (4) both TGF beta-1 and TGF beta-2 are potent mitogens for chicken embryonic calvarial bone cells.     

Purification of a growth factor related to platelet-derived growth factor and a type beta transforming growth factor secreted by mouse neuroblastoma cells. A general strategy for the purification of basic polypeptide growth factors.

A general strategy was developed for the purification of basic polypeptide growth factors. This method is a combination of gel filtration, weak-cation-exchange h.p.l.c. and reverse-phase h.p.l.c., separating the proteins according to size, charge and hydrophobicity respectively. All steps are carried out at low pH with exclusively volatile acidic buffer solutions. The sterile conditions and easy removal of salt by freeze-drying facilitate the detection of the growth factors by biological assays. By using this method, homogeneous preparations of two basic growth factors were purified in high yield from mouse-neuroblastoma-Neuro-2A-cell-conditioned medium. It is shown that these purified factors are biochemically and immunologically related to platelet-derived growth factor and type beta transforming growth factor from human platelets.     

A beta-type transforming growth factor, present in conditioned cell culture medium independent of cell transformation, may derive from serum

An alpha-type transforming growth factor (TGF alpha) is produced at high levels by rat embryo cells transformed by the Snyder-Theilen strain of feline sarcoma virus (FeSV). Addition of 2 ng mouse epidermal growth factor (mEGF) during purification identified the presence of a second, EGF-dependent growth factor of the TGF beta type (TGF beta) in this conditioned medium. This factor had an approximate Mr of 12,000 and eluted at 37% acetonitrile during high performance liquid chromatography. This extracellular type of TGF beta activity also was present in conditioned medium of rat cells after infection with a transformation defective strain of Abelson leukemia virus, and hence expression of this growth factor activity was independent of cell transformation. Moreover, the presence of an EGF-dependent, 12,000 Mr clonogenic activity in extracts of bovine serum alone suggests serum as an origin for the B-type transforming growth factor initially observed in conditioned medium of Snyder-Theilen FeSV transformed cells. This does not, however, preclude the possibility that TGF beta is also secreted by the transformed rat embryo cells themselves.     

Detection of mRNA for the transforming growth factor beta family in human articular chondrocytes by the polymerase chain reaction

The Transforming Growth Factor-beta (TGF beta) family of polypeptides elicits diverse biological actions on a wide range of cell types. There are known to be several isoforms of TGF beta coded for by different genes, with possibly differential expression and potencies. We have demonstrated that there is constitutive expression of three forms of transforming growth factor beta in adult human articular chondrocytes. The presence of 10% fetal calf serum in the culture medium may influence expression. The addition of transforming growth factor beta or interleukin 1 beta to the culture medium does not appear to consistently influence the expression of TGF beta by the cells.     

Transforming growth factor type beta in normal human urine

TGF beta has been identified in normal human urine specimens from five individuals studied for five consecutive days. The peptide was extracted from urine using Sepralyte C1 beads. Detectable levels of [125I]TGF beta competing activity as measured by radioreceptor assay was found in about half of the specimens studied. The protein isolated from urine using C1 Sepralyte beads was further purified using Biogel P-60 column chromatography. Fractions were tested for TGF beta and EGF competing activity using radioreceptor assays. TGF beta and EGF extracted from urine are clearly separated by column chromatography. Two distinct EGF peaks and a single TGF beta peak were observed. Fractions having [125I]TGF beta competing activity were pooled and further purified using reverse-phase HPLC. HPLC fractions having [125I]TGF beta competing activity were tested for bioactivity using a soft agar assay. The fractions were capable of stimulating soft agar growth of AKR-2B (clone 84A) cells and cross reacted with a TGF beta antibody in a radioimmunoassay. The presence of TGF beta in normal human urine was also demonstrated by immunoblotting. These results also suggest that C1 bead extraction of urine specimens can be used as a rapid first step in purification of TGF beta.     

Divergent action of transforming growth factor beta on DNA synthesis in human foetal liver cells

Purified human transforming growth factor beta (TGF beta) was found to be a potent inhibitor of DNA synthesis in human foetal hepatocytes. Half-maximal inhibition occurred at 0.5-1 pM and was reversible, with an increase in DNA synthesis within 24 h following removal of TGF beta. By contrast, in the same cultures, 'fibroblast-like' non-hepatocytes retained the ability to synthesize DNA in the presence of up to 200 fold higher doses of TGF beta. This differential response to TGF beta suggests that it may act as an important cell growth regulator in the human foetal liver.     

Altered response to growth factors or retinoic acid in phenotypic transformation of normal rat kidney cells expressing human c-fos gene.

Anchorage-independent growth of normal rat kidney (NRK) fibroblast in soft agar depends on both transforming growth factor beta (TGF beta) and epidermal growth factor (EGF). To examine whether c-fos protein is involved in phenotypic transformation of NRK cells, we have transfected and isolated several NRK cell lines that carry the human c-fos gene fused to the metallothionein IIA promoter. A transfectant, Nf-1, had constitutive levels of the human c-fos expression. Anchorage-independent growth of Nf-1 was already stimulated by EGF alone, and the colony sizes of Nf-1 were comparable to those of the parental NRK in the presence of both EGF and TGF beta. Anchorage-independent growth of NRK could be observed in the presence of TGF beta or retinoic acid or platelet derived growth factor (PDGF) and EGF. No growth of NRK in soft agar appeared when basic fibroblast growth factor (bFGF) and EGF were present. By contrast, anchorage-independent growth of Nf-1 was surprisingly enhanced by EGF and TGF beta or retinoic acid or PDGF or bFGF. Expression of the human c-fos gene may compensate the signal to phenotypic transformation induced by TGF beta as well as retinoic acid or PDGF or bFGF.     

Purification of human epidermal growth factor (urogastrone) from urine

Human epidermal growth factor has been isolated from a concentrated chromatographic eluate of human urine. The purification method utilizes six chromatographic steps including adsorption to aminoethylcellulose (AE-11), gel filtration on Sephadex G-50, carboxymethylcellulose (CM-52) chromatography, ion-exchange HPLC and reverse-phase HPLC. The final product appears homogeneous and identical to pure gamma-urogastrone when analyzed by polyacrylamide gel electrophoresis and reverse-phase HPLC using two eluent systems. The yield of the method described above allowed the development of a sensitive radioimmunoassay system for this growth factor.     

Transforming growth factor-beta induction of type-1 plasminogen activator inhibitor. Pericellular deposition and sensitivity to exogenous urokinase

The human tumor cell line HT-1080 was used as a model system to study the effects of transforming growth factor-beta (TGF beta) on polypeptide synthesis and proteolytic activity of malignant cells. Confluent cultures were exposed to TGF beta under serum-free conditions, and alterations in the production of proteins were examined by metabolic labeling and polypeptide analysis. TGF beta induced the synthesis and secretion of the Mr 47,000 endothelial type plasminogen activator inhibitor (PAI-1) as shown by reverse zymography, immunblotting, and immunoprecipitation analyses. TGF beta-induced PAI-1 was rapidly deposited in the growth substratum of the cells as shown by metabolic labeling and extraction of the cultures with sodium deoxycholate. Using pulse-chase experiments, we found a relatively fast turnover of substratum-associated PAI-1. Exogenously added urokinase released PAI-1 from the substratum even in the presence of the plasmin inhibitor aprotinin, suggesting a direct effect of urokinase. Immunoreactive complexes of higher molecular weight were subsequently detected in the medium. Epidermal growth factor, transforming growth factor-alpha, platelet-derived growth factor, and insulin did not elicit similar effects on the amount of PAI-1. TGF beta also inhibited the anchorage-independent growth of HT-1080 cells at the same concentrations at which it induced PAI-1. These results indicate that TGF beta can modulate the extracellular proteolytic activity of cultured cells by enhancing the secretion and deposition of PAI-1 into their microenvironment. It remains to be established whether TGF beta inhibition of anchorage-independent growth of these cells is associated with the induction of PAI-1.     

Effects of Transforming Growth Factor β1 on Astroglial Cells in Culture

The effects of transforming growth factor beta 1 (TGF beta 1) on DNA synthesis and functional differentiation of astroglial cells cultured in serum-free medium were investigated. TGF beta 1 diminished and delayed the peak of DNA synthesis induced by serum. TGF beta 1-treated cells were larger than control cells. This factor delayed the appearance of process-bearing cells induced by acidic fibroblast growth factor treatment and also affected the astrocyte-specific enzyme glutamine synthetase (GS), whose accumulation is under hydrocortisone (HC) control. TGF beta 1 inhibited the induction of GS activity by HC in a dose- and time-dependent manner. Moreover, pretreatment with TGF beta 1 for 4 h maintained the inhibition of GS activity for approximately 16 h after removal of this factor from culture medium. These results suggest that TGF beta 1 may be an important regulator of astrocyte growth and differentiation.     

HIV-1 transactivator protein Tat induces proliferation and TGF beta expression in human articular chondrocytes

The human immunodeficiency virus-1 (HIV-1) protein Tat binds to cell surface antigens and can regulate cellular responses. Tat has similar immunosuppressive effects as transforming growth factor-beta (TGF beta) and both inhibit lymphocyte proliferation. TGF beta is expressed by primary human articular chondrocytes and is their most potent growth factor. The present study analyzed the interactions of TGF beta and HIV Tat in the regulation of human articular chondrocytes. Synthetic or recombinant full-length Tat (1-86) induced chondrocyte proliferation and this was of similar magnitude as the response to TGF beta. Tat peptides that did not contain the RGD motif had similar chondrocyte stimulatory activity as full-length Tat. Among a series of Tat peptides, peptide 38-62 which contains the basic domain was the only one active, suggesting that this region is responsible for the effects on chondrocyte proliferation. Full-length Tat and peptide 38-62 synergized with TGF beta and induced proliferative responses that were greater than those obtained with any combination of the known chondrocyte growth factors. Further characterization of the interactions between Tat and TGF beta showed that Tat increased synthesis and TGF beta activity and TGF beta 1 mRNA levels. The stimulatory effects of Tat and peptide 38-62 on chondrocyte proliferation were reduced by neutralizing antibodies to TGF beta and by TGF beta antisense oligonucleotides. These results identify a virally encoded protein and a synthetic peptide derived from it as novel and potent chondrocyte growth stimuli which act at least in part through the induction of TGF beta.     

Modulation of transforming growth factor alpha-dependent expression of epidermal growth factor receptor gene by transforming growth factor beta, triiodothyronine, and retinoic acid

We have investigated the actions of transforming growth factor (TGF) type alpha on epidermal growth factor (EGF) receptor mRNA expression in MDA-468 human mammary carcinoma cells in serum-free media. We found that exposure of MDA-468 cells to TGF alpha results in elevated levels of EGF receptor mRNA. This increase in mRNA accumulation showed time and dose dependence. Addition of TGF beta 1 enhanced the accumulation of EGF receptor mRNA induced by TGF alpha in a time- and dose-dependent manner. We also found that triiodothyronine at physiological concentrations exerts synergistic control on the action of TGF alpha alone, or in association with TGF beta 1, on EGF receptor mRNA expression. Similarly, retinoic acid treatment also enhanced in a time- and dose-dependent manner the TGF alpha-dependent response of EGF receptor mRNA and acted synergistically with TGF beta 1. The results described here suggest that optimum regulation of EGF receptor gene expression by TGF alpha is a complex process involving synergistic interactions with heterologous growth factors and hormones.     

Cellular receptors for type beta transforming growth factor. Ligand binding and affinity labeling in human and rodent cell lines

Type beta transforming growth factor (beta TGF) purified from human platelets to homogeneity as judged by NH2-terminal amino acid sequence analysis has been labeled with 125I to characterize its interaction with cellular receptors. Binding of 125I-beta TGF to target cells is temperature- and time-dependent, specific, saturable, and reversible. About 1.6-1.9 X 10(4) binding sites/cell with high affinity for beta TGF (Kd = 5.6-7.8 X 10(-11) M and 9.1-14 X 10(-11) M, respectively) are found in NRK-49F and BALB/c 3T3 cells. beta TGF receptors do not appear to undergo acute down-regulation by the ligand. Specific binding of 125I-beta TGF has been observed in several human, rat, and mouse fibroblast lines and in some, but not all, tumor-derived cell lines examined. 125I-beta TGF has been cross-linked to intact cells and isolated membrane preparations using disuccinimidyl suberate. Cells and isolated membranes from human, rat, and mouse origin affinity labeled with 125I-beta TGF exhibit a major labeled species of approximately 280 kilodaltons that has the properties of high affinity and specificity expected from a physiologically relevant beta TGF receptor. Minor labeled species of 70-90 kilodaltons are also labeled by 125I-beta TGF, but they correspond to molecular species with low apparent affinity (Kd approximately 10(-8) M) for 125I-beta TGF.     

Melanoma growth stimulatory activity: isolation from human melanoma tumors and characterization of tissue distribution

Melanoma growth stimulatory activity (MGSA) is an acid and heat stable, auto-stimulatory growth factor which was first isolated from culture medium conditioned by the Hs294T human melanoma cell line. In this report, we describe the purification of MGSA from acid ethanol extracts of Hs294T tumors grown in nude mice using a series of Bio-Gel P30, reverse phase-high performance liquid chromatography and heparin-sepharose steps. This modified procedure provides a 10-fold improved yield of MGSA over previously published procedures. Purified MGSA-stimulated melanoma cell growth in both 3H-thymidine and cell number assays over a concentration range of 0.06 to 6 ng/ml. The MGSA bioactivity was primarily associated with fractions which exhibited molecular weights of 16 and 13-14 Kd based upon sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Purified platelet-derived growth factor (PDGF), insulin-like growth factor (IGF-1), transforming growth factor-beta (TGF beta), and epidermal growth factor (EGF) in combination with TGF beta did not stimulate 3H-thymidine incorporation in Hs294T cells under the conditions used for MGSA bioassay. Monoclonal antibody to MGSA was used to screen melanoma and benign nevus cultures as well as fixed sectioned tissue for MGSA. The majority of the melanoma cultures were MGSA positive, while most nevus cultures were MGSA negative. However, when fixed sectioned tissue was screened for MGSA immunoreactivity, melanoma tissue was MGSA positive and three-fourths of the benign nevi were MGSA positive. In addition, epidermal keratinocytes and several tissues exhibiting proliferative disorders contained immunoreactive MGSA. These data suggest that MGSA may be a normal regulator of growth and that the microenvironment of the cell may regulate both production of MGSA and response to MGSA.     

Suppression of the EGF-dependent induction of c-myc proto-oncogene expression by transforming growth factor beta in a human breast carcinoma cell line

Alterations in c-myc proto-oncogene expression after treatment of human mammary carcinoma MDA-468 cells with epidermal growth factor (EGF) and/or transforming growth factor beta (TGF beta) have been investigated. A stimulation of c-myc messenger RNA was detected within 60 min after treatment with EGF. This induction persisted for at least 24 hr, albeit to a lower extent. The early and late increase in c-myc mRNA levels induced by EGF were inhibited by the presence of TGF beta. TGF beta alone induced little change in c-myc mRNA levels. The effect of TGF beta represents a novel action of this hormone at the level of gene expression.     

Transforming growth factor beta induces the production of interleukin 6 by human peripheral blood mononuclear cells

Previous studies have indicated that the cytokine transforming growth factor beta 1 (TGF beta 1) has immunosuppressive properties and can inhibit the production of tumor necrosis factor (TNF) and Interleukin 1 (IL 1) by human peripheral blood mononuclear cells. In this study, we have examined the effects of TGF beta 1 on the production of Interleukin 6 (IL 6) by human peripheral blood mononuclear cells. Treatment with only TGF beta 1 leads to the induction of IL 6, and this was both dose- and time-dependent. The effect of TGF beta 1 was evident at the level of IL 6 mRNA, suggesting TGF beta 1-induced de novo synthesis of IL 6. Induction of IL 6 by TGF beta 1 was specific, as other cytokines made by mononuclear cells (TNF and IL 1) were not induced by TGF beta 1. Furthermore, when a panel of stimuli were compared for their ability to induce IL 1, TNF and IL 6 in the presence or absence of TGF beta 1, IL 6 levels were augmented in the presence of TGF beta 1, while the induction of IL 1 and TNF was inhibited significantly. These results indicate that TGF beta 1 has complex effects on the production of cytokines by peripheral blood mononuclear cells and that TGF beta 1 is not inhibitory for all cytokine production. The ability of TGF beta 1 to induce IL 6 suggests that IL 6 may mediate some of the effects of TGF beta 1.     

Altered regulation of protein disulfide isomerase in cells resistant to the growth-inhibitory effects of transforming growth factor beta 1

A murine keratinocyte cell line that is resistant to the growth-inhibitory effects of transforming growth factor beta 1 (TGF beta 1) was examined for differential gene expression patterns that may be related to the mechanism of the loss of TGF beta 1 responsiveness. Cells that were resistant to the growth-inhibitory effects of TGF beta 1 (KCR cells) were derived from K-ras-transformed BALB/MK keratinocytes (KC cells). Using a subtractive hybridization procedure with KC and KCR mRNAs, we isolated a complementary DNA clone for murine protein disulfide isomerase (PDI). The mRNA for PDI is inhibited by TGF beta 1 treatment in the parental KC cells, but not in the TGF beta 1-resistant KCR cells. Similar PDI down-regulation also occurs in other TGF beta-sensitive cells, but not in a human pancreatic carcinoma cell line which is insensitive to the growth-inhibitory effects of TGF beta 1. The results suggest that misregulation of PDI, an important component of co- and posttranslational modification systems, may be involved in the mechanism by which some cells escape from the growth-inhibitory effects of TGF beta 1.