Inhibition of transforming growth factor beta signaling in MCF-7 cells results in resistance to tumor necrosis factor alpha: a role for Bcl-2.

Transforming growth factor beta (TGF-beta) is a multifunctional cytokine capable of regulating diverse cellular processes. In this study we investigated the effect of autocrine TGF-beta signaling on tumor necrosis factor (TNF) alpha-induced cell death. We abrogated the TGF-beta autocrine loop by overexpression of a truncated TGF-beta type II receptor in MCF-7 breast carcinoma cells and found that this generated resistance to TNF-alpha-induced cytotoxicity. To elucidate the molecular basis of the influence of TGF-beta on TNF-alpha-induced cytotoxicity, we evaluated the expression levels or activities of proteins involved in TNF-alpha signal transduction or the regulation of apoptosis in general in TGF-beta-responsive and TGF-beta-nonresponsive MCF-7 cells. We observed no significant difference in the expression of TNF-alpha receptors or the TNF receptor-associated death domain protein. In addition, downstream activation of nuclear factor kappaB by TNF-alpha was not altered in cells that had lost TGF-beta responsiveness. Analysis of members of the Bcl-2 family of apoptosis-regulatory proteins revealed that Bcl-X(L) and Bax expression levels were not changed by disruption of TGF-beta signaling. In contrast, the TGF-beta-nonresponsive cells expressed much higher levels of Bcl-2 protein and mRNA than did cells with an intact TGF-beta autocrine loop. Furthermore, restoration of a TGF-beta signal to MCF-7 cells that had spontaneously acquired resistance to TGF-beta caused a reduction in Bcl-2 protein expression. Taken together, our data indicate that loss of autocrine TGF-beta signaling results in enhanced resistance to TNF-alpha-mediated cell death and that this is likely to be mediated by derepression of Bcl-2 expression.

Transactivation of the p21 promoter by BRCA1 splice variants in mammary epithelial cells: evidence for both common and distinct activities of wildtype and mutant forms.

We have evaluated the transcriptional activation of a human p21 promoter reporter construct by transfection of BRCA1 expression constructs into tumorigenic and nontumorigenic human breast cell lines. Two cell lines with wildtype p53 (MCF-7 and MCF10A) demonstrated transcriptional activation of the p21 promoter by full-length BRCA1 (BRCA1L) as well as by two splice variants that lack most of exon 11 (BRCA1S and BRCA1S-9,10). In contrast, two cell lines with mutant p53 (MDA-231 and HCC1937) were inactive. Co-transfection of BRCA1L with BRCA1S or BRCA1S-9,10 exhibited synergistic p21 promoter activation, due to augmented expression of the cytomegalovirus promoter-based BRCA1 expression constructs. We examined the transcriptional activity of two known sequence alterations in BRCA1, one that results in a carboxy-terminal truncation of BRCA1 and is clearly pathogenic, and the other a missense mutation that is suspected of predisposing to cancer. Although both mutations have been shown to be defective in some assays of transactivation, we observed both mutations to be fully active in activation of the p21 promoter when incorporated in the full-length BRCA1L. In contrast, these mutations rendered BRCA1S inactive. These observations indicate that such transcriptional assays cannot serve as the basis for a functional appraisal of BRCA1 sequence alterations encountered in the course of genetic testing.

Methylation of CpGs as a determinant of transcriptional activation at alternative promoters for transforming growth factor-beta3.

The human transforming growth factor-beta3 (TGF-beta3) gene has a typical CpG island, the core of which is centered just upstream of its principle promoter. Activation of an alternative downstream promoter, leading to the production of a truncated mRNA lacking the portion of the 5' noncoding region responsible for translational inhibition of TGF-beta3 mRNA, is only evident in breast cancer cells. We compared the methylation status of genomic DNA isolated from a panel of breast (SKBR-3 and T47-D) and non-breast cancer (HT-1080, A673, and A375) cell lines by sequencing sodium metabisulfite-treated DNA. In all cell lines, the core of the TGF-beta3 CpG island was predominantly unmethylated, irrespective of promoter usage associated with that cell line. In contrast, we observed a marked difference in methylation at 19 CpG sites immediately flanking and downstream of the alternative promoter's transcription initiation site. Specifically, the non-breast cancer cell lines exhibited nearly complete methylation of these CpG sites, whereas in the breast cancer cell lines, these CpGs were predominantly unmethylated. Our data support the hypothesis that methylation of a limited number of CpGs at the periphery of an otherwise unmethylated CpG island underlies the transcriptional repression of the downstream promoter in non-breast cancer cells, thereby serving to regulate the use of alternative promoters for TGF-beta3.

Characterization of functional messenger RNA splice variants of BRCA1 expressed in nonmalignant and tumor-derived breast cells.

BRCA1 has been identified as a tumor suppressor gene that is mutated in many cases of inherited breast and ovarian cancer. Recent data suggest that multiple splice forms of BRCA1 exist, but the structure and function of these alternative transcripts have not been elucidated. By sequence analysis of reverse transcription-PCR products, we have determined that a major splice form of BRCA1 expressed in malignant and nonmalignant breast epithelial cells contains an in-frame deletion of 3309 nucleotides from exon 11. A second alternative splice event results in the in-frame deletion of the 123 nucleotides that make up exons 9 and 10. These splice variants are found on polysomes and are therefore predicted to encode 80-85-kDa BRCA1-derived proteins lacking approximately 60% of the internal amino acids that constitute full-length BRCA1.

Therapeutic implications of the TGF-beta system.

This review considers the various roles of the TGF-beta system in mammary carcinogenesis, tumor progression, and cellular responses to therapeutic measures. The paradigm that has evolved from the work of many investigators suggests that loss of tumor cell responsiveness to the effects of TGF-beta can result in a crucial shift in the net effect of TGF-beta within the context of the tumor-host interaction. Principal elements of host-tumor interactions in which this shift may play out, including immune suppression, angiogenesis, and modification of the surrounding extracellular matrix by tumor cells, are potentially amenable to manipulation. Additional effects of TGF-beta, such as suggested by reports of its ability to alter the drug resistance of tumor cells and the drug sensitivity of normal tissues, suggest that appropriate molecular intervention designed to affect the TGF-beta system might constitute an effective therapeutic strategy.

Enhanced translational efficiency of a novel transforming growth factor beta 3 mRNA in human breast cancer cells.

The mRNA for transforming growth factor beta 3 (TGF-beta 3) includes a long (1.1-kb) 5' noncoding region which exerts a potent inhibitory effect on translational efficiency. We now report that many human breast cancer cell lines (T47-D, SK-BR-3, ZR-75-1, and BT-474) express two mRNA species for TGF-beta 3: the 3.5-kb transcript previously described as the only TGF-beta 3 mRNA species in cells and a novel 2.6-kb transcript which lacks approximately 870 nucleotides from the 5' noncoding region. The 5' end of the shorter transcript was sequenced, establishing it to be a 5' truncation of the full-length TGF-beta 3 transcript. Estradiol decreased mRNA levels of both TGF-beta 3 mRNA transcripts to an equivalent degree in estrogen receptor-positive cells. In contrast, the synthetic progestin gestodene altered the relative abundance of the two transcripts, preferentially diminishing the expression of the 2.6-kb transcript. The potential for enhanced mRNA translation attributable to the shorter 5' noncoding region was evaluated by transfection of cells with chimeric plasmid constructs in which the transcription unit consisted of coding sequence for chloramphenicol acetyltransferase downstream of the 5' noncoding sequence from TGF-beta 3. The translational efficiency of chloramphenicol acetyltransferase-encoding mRNA containing the shorter 5' noncoding region of the 2.6-kb TGF-beta 3 transcript was approximately seven times greater than with the full-length 5' noncoding region of TGF-beta 3. Polysome analysis of TGF-beta 3 mRNA in SK-BR-3 cells supported the hypothesis that the 2.6-kb transcript was more actively engaged in translation.

Association of heterogeneous nuclear ribonucleoprotein A1 and C proteins with reiterated AUUUA sequences.

Post-transcriptional regulatory mechanisms have been shown to play a major role in gene expression in eukaryotic cells. The presence of a reiterated pentamer (AUUUA) in the 3'-untranslated region (UTR) of mRNAs encoding lymphokines, cytokines, transcription factors, and proto-oncogenes has been shown to be associated with rapid turnover and translation attenuation. Cytoplasmic proteins (70, 50, 43, 36, and 25 kDa) capable of specifically binding to RNAs containing these AU-rich sequences were identified in human peripheral blood T lymphocytes. Levels of the 36-kDa protein were markedly increased following transcriptional, but not translational inhibition, a feature recently reported for hnRNP A1, a protein of comparable mass. Antibodies directed against heterogeneous nuclear ribonucleoproteins (hnRNPs) A1 and C immunoprecipitated 36- and 43-kDa proteins that had bound the AUUUA-rich region contained in the 3'-UTR of granulocyte-macrophage colony-stimulating factor mRNA. Recombinant hnRNP A1 was shown to preferentially bind to RNAs containing AUUUA sequences in a specific manner, and displayed comparable patterns to the 36-kDa AU-specific binding proteins following partial proteolysis. These data identify for the first time hnRNP A1 and C as cytoplasmic proteins in human lymphocytes that are capable of specifically associating with reiterated AUUUA sequences present in the 3'-UTR of labile mRNAs. As such, they may play a role as trans-acting factors in the modulation of cytoplasmic mRNA turnover and translation, in addition to their previously characterized roles as pre-mRNA binding proteins involved in nuclear mRNA processing.

Immunohistochemical staining for transforming growth factor beta 1 associates with disease progression in human breast cancer.

The transforming growth factor beta s (TGF-beta) comprise a family of M(r) 25,000 pluripotent growth factors which have been implicated in the development and progression of human breast cancer. Conflicting data suggest that TGF-beta has the potential to either inhibit or promote the progression of mammary neoplasia. We therefore examined a pathological library of malignant breast biopsy specimens to determine the prevalence and distribution of immunoreactivity with antibodies specific for the three mammalian isoforms of TGF-beta (beta 1, beta 2, and beta 3). We found that intense staining for TGF-beta 1 was positively associated with rate of disease progression, and that this was independent of age, stage, nodal status, or estrogen receptor status (P = 0.009).

Altered metabolic and adhesive properties and increased tumorigenesis associated with increased expression of transforming growth factor beta 1.

Transforming growth factor-beta (TGF-beta) is a potent mediator of cell proliferation and extracellular matrix formation, depending on the cell type and the physiological conditions. TGF-beta is usually secreted in a "latent" complex that needs activation before it can exert its effects. Several observations correlate increased expression of TGF-beta 1 with tumorigenesis. To evaluate the physiological relevance of increased TGF-beta 1 synthesis in tumor cells we established cell clones overexpressing TGF-beta 1 and observed the resulting physiological changes in TGF-beta overproducing cells in vitro and in vivo. As a model system we used the human E1A-transformed 293 tumor cells, which are insensitive to the direct growth modulatory effects of TGF-beta. The selection of this cell line allows an assessment of physiological alterations independent of TGF-beta induced proliferative changes. The use of two TGF-beta 1 expression vectors containing either the natural or a modified TGF-beta 1 precursor cDNA permitted the establishment of separate 293 cell lines overexpressing latent or active TGF-beta. Comparison of the resulting changes in glycolytic rate, adhesiveness and integrin and plasminogen activator expression established that, in vitro, both types of clones behaved similarly, indicating that expression of latent TGF-beta induces autocrine changes in the tumor cells and thus suggesting that some level of cell-associated activation occurs. TGF-beta overexpression resulted in an increased metabolic rate due to enhanced glycolysis, a property long associated with tumor cells. This increased glycolysis was not associated with altered proliferation. Cells overexpressing TGF-beta also displayed enhanced fibronectin mRNA and plasminogen activator synthesis and increased adhesiveness in vitro. They showed enhanced survival when plated sparsely on plastic in the absence of serum, and attached more readily to laminin. In addition, synthesis of several beta 1 integrins, in particular the alpha 1/beta 1, alpha 2/beta 1, and alpha 3/beta 1, all of which recognize laminin, were enhanced. Finally, cells overexpressing active TGF-beta, but not latent TGF-beta, also showed increased tumorigenicity in nude mice. Thus, an increase in endogenous TGF-beta synthesis confers several proliferation-independent phenotypic changes which may be of significance for the survival of the tumor cell inoculum or its subsequent growth, and for tumor formation and development. In the case of cells expressing active TGF-beta, the release of active TGF-beta into the vicinity of the tumor cells may also result in a more hospitable environment for tumor growth.

Regulation of growth of normal ovarian epithelial cells and ovarian cancer cell lines by transforming growth factor-beta.

OBJECTIVE: The purpose of this study was to study the role of transforming growth factor-beta in regulation of proliferation of normal and malignant ovarian epithelial cells. STUDY DESIGN: We examined production of and responsiveness to transforming growth factor-beta in primary monolayer cultures of epithelial cells from five normal human ovaries and in five ovarian cancer cell lines. RESULTS: In normal ovarian epithelial cells, proliferation always was inhibited by transforming growth factor-beta (greater than 40%) (p less than 0.01). Among the cancer cell lines, proliferation of one was markedly inhibited (greater than 95%) (p less than 0.01), two were only modestly inhibited (15% to 20%) (p less than 0.05), and two were unaffected. In addition, we found that all of the normal ovarian epithelial cells and four of five ovarian cancer cell lines produce transforming growth factor-beta ribonucleic acid and protein. CONCLUSIONS: These data suggest that transforming growth factor-beta may act as an autocrine growth inhibitory factor for normal ovarian epithelium in vivo. Because most of the ovarian cancer cell lines are relatively resistant to the growth inhibitory effect of transforming growth factor-beta and because one cell line does not produce transforming growth factor-beta, it is possible that loss of the transforming growth factor-beta pathway may play a role in the development of some ovarian cancers.

Inhibition of translation of transforming growth factor-beta 3 mRNA by its 5' untranslated region.

We have cloned and sequenced the 5' untranslated region of the transforming growth factor-beta 3 (TGF-beta 3) mRNA as well as the adjacent genomic sequence. S1 nuclease analysis identified a single transcription start site. We have thus determined that the 5' untranslated region is about 1.1 kb long and contains 11 open reading frames. In vitro translation of the TGF-beta 3 precursor coding sequence was markedly inhibited by the presence of the 5' untranslated region. Similarly, when the 5' untranslated region of TGF-beta 3 was introduced upstream of the coding sequence of chloramphenicol acetyltransferase, in vitro translation was inhibited. Furthermore, upon transfection into 293 cells, chloramphenicol acetyltransferase expression was inhibited by the 5' untranslated region of TGF-beta 3. The degree of translational inhibition was inversely proportional to the amount of transfected DNA. Mutation analysis implicated multiple segments of the 5' untranslated region as contributing to the inhibitory effect. Deletion of much of the 5'-most 640 nucleotides, including 8 of the 11 upstream ATGs, relieved much but not all of the inhibitory influence of the 5' untranslated region of TGF-beta 3 mRNA. The two upstream open reading frames closest to the initiator codon for the TGF-beta 3 coding sequence also decreased translational efficiency, since mutation of either ATG resulted in increased translation. Transfection results with T47-D cells, a cell line which expresses TGF-beta 3 mRNA, were similar to those obtained with the 293 cell line. Thus, TGF-beta 3 mRNA is a recent example of an expanding group of growth-related mRNAs in which the 5' untranslated region contains upstream open reading frames and other sequences which inhibit translation.

Differential regulation of expression of three transforming growth factor beta species in human breast cancer cell lines by estradiol.

Transforming growth factor (TGF)-beta is a potent regulator of many cell functions and a growth inhibitor for mammary epithelial cells. We now know of three highly homologous members of the human TGF-beta gene family. We have studied the expression of TGF-beta 1, -beta 2, and -beta 3 mRNA in four human breast cancer cell lines. Using the RNase protection assay, we have detected mRNA expression of TGF-beta 1, -beta 2, and -beta 3 by T-47D cells, TGF-beta 1 and -beta 3 by ZR-75-1 cells, and TGF-beta 1 by MCF-7 cells. Treatment of these estrogen receptor-positive cells with 10 nM estradiol for 48 h resulted in decreased mRNA levels of TGF-beta 2 and -beta 3 but did not affect mRNA levels of TGF-beta 1. Expression of TGF-beta 1 and -beta 2 mRNA by an estrogen receptor-negative cell line, MDA-MB-231, was not changed by estradiol treatment. Treatment of cells with the antiestrogen tamoxifen (1 microM) did not significantly alter mRNA levels for any of the three TGF-beta species. We have further determined that estradiol treatment of T-47D was associated with diminished secretion of TGF-beta into the medium. Both TGF-beta 1 and -beta 2 inhibited the proliferation of MCF-7 cells, and neither protein affected the growth of T-47D cells. TGF-beta 1 was at least 10-fold more potent than TGF-beta 2 at inhibiting the growth of MCF-7 cells.

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.

Hydrogen peroxide from cellular metabolism of cystine. A requirement for lysis of murine tumor cells by vernolepin, a glutathione-depleting antineoplastic.

The sesquiterpene lactone antineoplastic vernolepin acutely depletes murine tumor cell glutathione (GSH), and lyses the cells by an unknown mechanism that is enhanced synergistically by inhibition of GSH synthesis with buthionine sulfoximine (BSO) (Arrick et al. 1983. J. Clin. Invest. 71:258). We found here that lysis of P815 mastocytoma cells by vernolepin, with or without BSO, required cystine in the culture medium. Addition of catalase markedly suppressed vernolepin-mediated cytolysis in cystine-containing media, suggesting the involvement of hydrogen peroxide in the cytolytic action of vernolepin. Consistent with this, inhibition of tumor cell glutathione disulfide reductase with 1,3-bis(2-chloroethyl)-1-nitrosourea or inhibition of endogenous catalase with aminotriazole synergistically augmented cytolysis by vernolepin. Moreover, H2O2 was released by suspensions of P815 cells in cystine-containing buffer (63 pmol/10(6) cells X h). Omission of cystine reduced the rate of H2O2 accumulation 10-fold. No H2O2 was detected without cells. Cytolysis by vernolepin could be restored in cystine-deficient medium by several other disulfides, themselves noncytolytic, such as disulfiram and oxidized Captopril, as well as by cysteine. In contrast, withholding two other essential amino acids (leucine or tryptophan) or adding cycloheximide did not interfere with cytolysis by vernolepin. These results suggest that cellular uptake of disulfides of physiologic and pharmacologic interest may be followed by their intracellular reduction and autooxidation with generation of H2O2. This previously unrecognized source of intracellular oxidant stress may be an important component of injury to GSH-depleted cells.

Glutathione metabolism as a determinant of therapeutic efficacy: a review.

Glutathione, as the chief nonprotein intracellular sulfhydryl, affects the efficacy and interactions of a variety of antineoplastic interventions, mainly through nucleophilic thioether formation or oxidation-reduction reactions. Thus, glutathione plays a role in the detoxification and repair of cellular injury by such diverse agents as mechlorethamine, melphalan, cyclophosphamide, nitrosoureas, 6-thiopurine, 4'-(9-acridinylamino)methanesulfon-m-anisidide, the quinone antibiotics (including Adriamycin, daunorubicin, and mitomycin C), the sesquiterpene lactones (such as vernolepin), and other sulfhydryl-reactive diterpenes (like jatrophone). Glutathione may play a similar role in host and tumor cell responses to radiation, hyperthermia, and the reactive reduction products of oxygen secreted by inflammatory cells. Further, glutathione participates in the formation of toxic metabolites of such chemotherapeutics as azathioprine and bleomycin and may affect the cellular uptake of other agents, such as methotrexate. It seems likely that alterations in glutathione metabolism of tumor or host as a result of one therapeutic intervention may affect the outcome of concurrent treatments. Knowledge of these interactions may be useful in designing combination therapy for neoplastic disease.

Inhibition of glutathione synthesis augments lysis of murine tumor cells by sulfhydryl-reactive antineoplastics.

GSH plays an important role in cellular defense against a wide variety of toxic electrophiles via the formation of thioether conjugates. We studied the role of GSH in murine tumor cell defense against a novel class of sulfhydryl-reactive antineoplastics, the sesquiterpene lactones (SL). Incubation of P815 mastocytoma cells with any of the four SL tested (vernolepin, helenalin, elephantopin, and eriofertopin) for 1 h resulted in 70-97% depletion of GSH. The importance of GSH resynthesis upon exposure of tumor cells to SL was evaluated with the use of buthionine sulfoximine (BSO), a selective, nontoxic inhibitor of gamma-glutamylcysteine synthetase. Inhibition of GSH synthesis with 0.2 mM BSO markedly enhanced SL-mediated cytolysis of four murine tumor cell lines. A 6- to 34-fold reduction in the amount of SL causing 50% lysis was obtained with BSO. Addition of BSO to P815cells either during or immediately after a 1-h pulse with 10 micrograms/ml of vernolepin increased cytolysis from less than 3% to 78-82%. However, a 1.5-h delay in the addition of BSO to such cells, which allowed for substantial resynthesis of GSH, reduced cytolysis to 30%. Recovery of GSH synthetic capacity after BSO treatment correlated with loss of the synergistic effect of BSO on lysis by vernolepin. BSO did not augment cytolysis by six other antineoplastics (doxorubicin, mitomycin C, vinblastine, cytosine arabinoside, maytansine, and 1,3-bis-[2-chloroethyl]-1-nitrosourea [BCNU]). Of these, only BCNU depleted cellular GSH. Lysis by jatrophone, another GSH-depleting antitumor agent, was increased 21-fold by BSO. Since prolonged incubation with BSO alone results in near-complete GSH depletion without loss of cell viability, SL-mediated cytolysis is probably not a result of GSH depletion. We have demonstrated, however, a critical role for GSH synthetic capacity as a determinant of tumor cell susceptibility to cytolysis by SL. GSH also plays an important role in cellular defense against oxidative injury. Vernolepin, acting as a GSH-depleting agent, markedly sensitized tumor cells to lysis by H2O2 (greater than 6.5-fold increase with 20 micrograms/ml of vernolepin). These findings suggest the possibility that the coordinated deployment of sulfhydryl-reactive antitumor agents, BSO, and oxidative injury might constitute an effective chemotherapeutic strategy.

The cutaneous infiltrates of leprosy: cellular characteristics and the predominant T-cell phenotypes.

We report on the characteristics of cells in the cutaneous lesions and blood of 21 patients with lepromatous, tuberculoid, and intermediate forms of leprosy. A large proportion of the infiltrates in lepromatous lesions consist of macrophages heavily parasitized with Mycobacterium leprae. The T cells in the lesions are devoid of OKT4/Leu 3a-positive ("helper") cells and consist almost exclusively of OKT8/Leu 2a-positive ("suppressor") populations. In contrast, the tuberculoid infiltrates contain well-organized epithelioid and giant-cell granulomas and only remnants of bacilli, and the predominant T cell is from the OKT4/Leu 3a-positive subset. In both tuberculoid and lepromatous infiltrates, T cells and macrophages expressed HLA-DR antigen. No marked alteration in the distribution of blood T-cell phenotypes was noted. We conclude that there is a marked difference between T-cell subsets in lepromatous and tuberculoid infiltrates, which may influence the microbicidal activity of macrophages in the lesions.