A novel cell culture model for studying differentiation and apoptosis in the mouse mammary gland.

BACKGROUND: This paper describes the derivation and characterization of a novel, conditionally immortal mammary epithelial cell line named KIM-2. These cells were derived from mid-pregnant mammary glands of a mouse harbouring one to two copies of a transgene comprised of the ovine beta-lactoglobulin milk protein gene promoter, driving expression of a temperature-sensitive variant of simian virus-40 (SV40) large T antigen (T-Ag). RESULTS: KIM-2 cells have a characteristic luminal epithelial cell morphology and a stable, nontransformed phenotype at the semipermissive temperature of 37 degrees C. In contrast, at the permissive temperature of 33 degrees C the cells have an elongated spindle-like morphology and become transformed after prolonged culture. Differentiation of KIM-2 cells at 37 degrees C, in response to lactogenic hormones, results in the formation of polarized dome-like structures with tight junctions. This is accompanied by expression of the milk protein genes that encode beta-casein and whey acidic protein (WAP), and activation of the prolactin signalling molecule, signal transducer and activator of transcription (STAT)5. Fully differentiated KIM-2 cultures at 37 degrees C become dependent on lactogenic hormones for survival and undergo extensive apoptosis upon hormone withdrawal, as indicated by nuclear morphology and flow cytometric analysis. KIM-2 cells can be genetically modified by stable transfection and clonal lines isolated that retain the characteristics of untransfected cells. CONCLUSION: KIM-2 cells are a valuable addition, therefore, to currently available lines of mammary epithelial cells. Their capacity for extensive differentiation in the absence of exogenously added basement membrane, and ability to undergo apoptosis in response to physiological signals will provide an invaluable model system for the study of signal transduction pathways and transcriptional regulatory mechanisms that control differentiation and involution in the mammary gland.

NF-kappaB inhibits apoptosis in murine mammary epithelia.

The transcription factor NF-kappaB is a key modulator of apoptosis in a variety of cell types, but to date this specific function of NF-kappaB has not been demonstrated in epithelia. Here, we describe the activation of NF-kappaB during post-lactational involution of the mouse mammary gland, a period of extensive apoptosis of luminal epithelial cells. Significantly, active NF-kappaB localized exclusively to nonapoptotic epithelial cells both in vivo and in the mammary epithelial cell line, KIM-2, transduced with an NF-kappaB-dependent green fluorescent protein reporter. Activation of NF-kappaB in vitro coincided with a decrease in the cytosolic repressor, IkappaBalpha. Furthermore, induction of NF-kappaB either by extracellular ligands or, more specifically, by inhibition of the IkappaB repressor with adenoviral constructs expressing antisense mRNA, resulted in enhanced survival of KIM-2 cells. Therefore, although coincident with induction of apoptosis both in vivo and in vitro, NF-kappaB appeared to exert a selective survival function in epithelial cells. This study highlights for the first time a role for NF-kappaB in modulating apoptosis in epithelium.

NF-kappaB and apoptosis in mammary epithelial cells.

A number of transcription factors have been identified as regulators of mammary development, including Stat5 and C/EBPbeta (1-3). In this review we summarize evidence which suggests that the NF-kappaB family of transcription factors also has a role in mammary gland development. NF-kappaB was originally described as a mediator of inflammatory reactions and cellular responses to viral pathogens. More recently it has been shown to possess an anti-apoptotic effect in a variety of cell types by regulating apoptosis-related genes. In the light of this function in other tissues, and the observation that aberrant activation of NF-kappaB can be associated with mammary tumors, we discuss the potential role of this transcription factor in modulating mammary epithelial apoptosis and involution of the mammary gland.

Ternary complex factors Elk-1 and Sap-1a mediate growth hormone-induced transcription of egr-1 (early growth response factor-1) in 3T3-F442A preadipocytes.

In our search for transcription factors induced by GH, we have analyzed immediate early gene activation in a model of GH-dependent differentiation. Here we describe the activation of early growth response factor-1 (egr-1) in GH-stimulated 3T3-F442A preadipocytes and the transcription factors responsible for its transactivation. Binding activity of egr-1 in electrophoretic mobility shift assay (EMSA) increased transiently 1 h after GH stimulation, accompanied by a concomitant increase in egr-1 mRNA. egr-1 induction appeared not to be related to proliferation since it was amplified in quiescent preadipocytes at a time when cells were refractive to GH-stimulated DNA synthesis. Truncations of the proximal 1 kb of the egr-1 promoter revealed that a 374-bp region (-624 to -250) contributes about 80% of GH inducibility in 3T3-F442A cells and approximately 90% inducibility in CHO-K1 cells. This region contains three juxtaposed SRE (serum response element)/Ets site pairs known to be important for egr-1 activity in response to exogenous stimuli. Site-specific mutations of individual SRE and Ets sites within this region each reduced GH inducibility of the promoter. Use of these site-specific mutations in EMSA showed that disruption of either Ets or SRE sites abrogated ternary complex formation at the composite sites. DNA binding of ternary complexes, but not binary complexes, in EMSA was rapidly and transiently increased by GH. EMSA supershifts indicated these ternary complexes contained serum response factor (SRF) and the Ets factors Elk-1 and Sap-1a. Coexpression of Sap-1a and Elk-1 resulted in a marked increase in GH induction of egr-1 promoter activity, although transfection with expression vectors for either Ets factor alone did not significantly enhance the GH response. We conclude that GH stimulates transcription of egr-1 primarily through activation of these Ets factors at multiple sites on the promoter and that stabilization of ternary complexes with SRF at these sites maximizes this response.

Growth hormone and insulin-like growth factor I induce bone morphogenetic proteins 2 and 4: a mediator role in bone and tooth formation?

GH is known to increase the formation of bone and hard tissues of the tooth (dentine, cementum, and enamel), as do bone morphogenetic proteins. GH receptors are expressed in these tissues and could mediate local growth responses. Here we report that both GH and insulin-like growth factor I (IGF-I) are able to increase expression of bone morphogenetic protein-2 and -4 messenger RNAs 4- to 5-fold in human dental pulp fibroblasts in vitro. Induction was seen at physiological concentrations of hormone (25-100 ng/ml GH; 50-200 ng/ml IGF-I) and reached a maximum at 4-8 h. Immunoblot analysis demonstrated that the increase in messenger RNAs resulted in an increase in expressed protein. Anti-IGF-I inhibition experiments indicate that GH is able to induce the response without a requirement for local IGF-I production. These results raise the possibility that bone morphogenetic proteins mediate the local osteogenic actions of GH and IGF-I, and lend support to the view that GH can act through the mediation of factors other than IGF-I. These factors may combine with IGF-I in different tissues to enhance GH action and specificity.

Activation of chimeric and full-length growth hormone receptors by growth hormone receptor monoclonal antibodies. A specific conformational change may be required for full-length receptor signaling.

Signal transduction by the growth hormone receptor (GHR) occurs through growth hormone (GH)-induced dimerization of two GHRs to form a trimeric complex. It is thought that dimerization alone is sufficient for signaling, since monoclonal antibodies (mAbs) against the extracellular domain of the GHR elicit proliferation of FDC-P1 cells transfected with a chimeric receptor comprising the extracellular domain of the GHR and the fibronectin and cytoplasmic domains of the murine granulocyte colony-stimulating factor receptor. We have screened 14 GHR mAbs for proliferative activity against characterized FDC-P1 and BaF-B03 cell lines stably expressing the full-length human, rabbit, or rat GHR, or the chimeric human GHR/granulocyte colony-stimulating factor receptor, and for transactivation of the c-fos promoter and STAT activation. With the chimeric receptor, eight mAbs were able to elicit proliferation, although there was no correlation between inhibition of hormone binding and agonist activity. In contrast, no mAbs were able to act as agonists with the full-length GHR FDC-P1 cell lines, although nine competed with GH for binding. A weak proliferative response was observed in the BaF-B03 cell lines with two of the mAbs (263 and 1C9), and the addition of anti-mouse F(ab)2 resulted in increased signaling in the hGHR BaF-B03 cell line to a plateau of 28 +/- 4% of the GH maximum for mAb 263. These data could indicate considerable stringency in the ability of mAbs to correctly dimerize the full-length GHR. However, the ability of mAb 263 to stimulate a mutant hGHR altered in the F'-G' loop of domain 2 was nearly abolished, concurrent with an increased affinity of this mAb for the receptor. Since the F'-G' loop undergoes a conformational change on GH binding and is necessary for full proliferative signaling, we propose that in addition to promoting receptor dimerization, mAb 263 may induce specific changes in receptor conformation similar to GH, which are required for the biological response.

Growth factor modulation of mitogenic responses and proteoglycan synthesis by human periodontal fibroblasts.

In order to understand the relationship between specific growth factors and matrix synthesis by periodontal cells, we have investigated the effects of platelet-derived growth factor BB (PDGF-BB), insulin-like growth factor-I (IGF-1), and growth hormone on DNA and proteoglycan synthesis by cultured human gingival and periodontal ligament fibroblasts in vitro. PDGF-BB and IGF-1, but not growth hormone, were mitogenic for both periodontal ligament fibroblasts and gingival fibroblasts, although the periodontal ligament cells responded more strongly. The mitogenic response was accompanied by alterations in expression of matrix proteoglycan mRNA. For both the gingival and periodontal ligament cells, there was a decrease in mRNA for decorin and an increase in mRNA for versican following exposure to IGF-1 and PDGF-BB. Although no change was seen in response to PDGF, biglycan mRNA level was increased by IGF-1 in periodontal ligament fibroblasts. With the gingival fibroblasts, biglycan mRNA levels were unaffected by IGF-1, PDGF-BB, or growth hormone. These findings suggest variable responses of fibroblasts to growth factors depending upon anatomical site within the periodontium. Moreover, there appears to be a correlation between cell proliferation and the types of proteoglycan synthesised with decorin expression being suppressed, and versican being increased during fibroblast proliferation.

Growth hormone and colony-stimulating factor 1 share multiple response elements in the c-fos promoter.

The c-fos protooncogene is induced by GH rapidly, but transiently. Induction requires C kinase activation and the serum response element, and recent binding studies have also implicated the sis-inducible element (SIE). However, no systematic study of the promoter elements responsible for transactivation by GH has been undertaken. Here we used Chinese hamster ovary K1 cells transiently cotransfected with rabbit GH receptor and c-fos promoter-luciferase constructs to demonstrate that the major region responsible for GH induction is located between 284-396 base pairs upstream of the transcription start site. Full induction by GH requires all of the known elements located in this region to be intact, including the SIE or signal transducer and activator of transcription binding element. We also report novel negative elements located around -216 upstream of the start site that reduce induction by GH and provide gel shift evidence for factors binding in this region. Cotransfection of Chinese hamster ovary K1 cells with c-fms and c-fos promoter constructs followed by the addition of CSF-1 revealed that these same c-fos elements contribute to transactivation by c-fms. Serum also uses the same elements to induce c-fos expression, except for the SIE. These results indicate that GH receptor and c-fms tyrosine kinase operate through multiple common response elements to regulate c-fos gene expression despite their structural differences.

Early responses of trans-activating factors to growth hormone in preadipocytes: differential regulation of CCAAT enhancer-binding protein-beta (C/EBP beta) and C/EBP delta.

Using the 3T3-F442A preadipocyte line as a model of GH-dependent differentiation, early changes in the DNA-binding affinity of transcription factors in response to GH addition were investigated. Addition of 50 ng/ml human GH to cells in chemically defined medium led to a rapid increase in binding activity of activator protein 1 (AP-1) and CCAAT enhancer-binding protein (C/EBP), which was significant at 30 min and reached maximal induction by 2 h (3-fold for AP-1, 2.5-fold for C/EBP). Induction in AP-1 DNA binding correlates with a concomitant GH trans-activation of c-jun and c-fos genes described previously. Using specific antibodies in electrophoretic mobility shift assays and Western blots, it was shown that the increase in activity of C/EBP is the result of an increase in synthesis of two alternatively translated forms of C/EBP beta: 40-C/EBP beta and 23-C/EBP beta. This increase in protein was not accompanied by alteration in mRNA level and could be blocked by a Janus kinase 2 tyrosine kinase inhibitor and a C kinase inhibitor at concentrations shown to inhibit GH-dependent activation of microtubule-associated protein (MAP) kinases. Concomitant with the translationally activated increase in C/EBP beta, a GH-dependent increase was observed in C/EBP delta transcription. This was accompanied by an increase in mRNA for C/EBP delta, which was superinduced by cycloheximide and, unlike the increase in C/EBP beta protein, was not observed with insulin. Thus GH exerts its effects on C/EBP isoforms at two levels: transcriptional activation of C/EBP delta and translational activation of C/EBP beta. It is proposed that GH-dependent phosphorylation results in the efficient translation of 40-C/EBP beta and 23-C/EBP beta (the mouse homolog of the inhibitor liver-enriched inhibitory protein), and that together with the induction of C/EBP delta, these may be involved in initiating the adipocyte differentiation program.

Signal transduction by the growth hormone receptor.

It has been proposed that dimerization of identical receptor subunits by growth hormone (GH) is the mechanism of signal transduction across the cell membrane. We present here data with analogs of porcine GH (pGH), with GH receptors (GHR) mutated in the dimerization domain and with monoclonal antibodies to the GHR which indicate that dimerization is necessary but not sufficient for transduction. We also report nuclear uptake of GH both in vivo and in vitro, along with nuclear localization of the receptor and GH-binding protein (GHBP). This suggests that GH acts directly at the nucleus, and one possible target for this action is a rapid increase in transcription of C/EBP delta seen in 3T3-F442A cells in response to GH. This tyrosine kinase-dependent event may be an archetype for induction of other immediate early gene transcription factors which then interact to determine the programming of the subsequent transcriptional response to GH.

Major histocompatibility complex variants and articular disease severity in rheumatoid arthritis.

In the light of associations previously described between DR4, DQw7, the C4B null allele and Felty's syndrome, and between Dw14, the C4A null allele and rheumatoid vasculitis, we have looked for associations between these and other DRB1, DQB, DQA and C4 encoded variants, and articular disease severity assessed radiologically in 119 subjects with RA but without major extra-articular features. The association of DR4 with more severe disease and DR1 and DR2 with mild disease was confirmed but no associations were found between the presence of DQw7, Dw14, C4A or C4B null alleles and articular disease severity. This suggests that the associations noted previously between these markers and extra-articular disease features of Felty's syndrome or rheumatoid vasculitis are unlikely to reflect an overall effect on disease severity but more likely to represent associations with these particular extra-articular disease subsets.

HLA associations in subjects with rheumatoid arthritis and bronchiectasis but not with other pulmonary complications of rheumatoid disease.

We have examined HLA-DR, DQA and DQB variants in 72 controls, 153 subjects with RA without extra-articular features and in subjects with the rheumatoid pulmonary complications of interstitial fibrosis (23) peripheral airways disease (13) and in 41 subjects with RA and bronchiectasis. Subjects with RA alone showed the expected association with HLA-DR4 (79%) but those with RA and co-existent pulmonary fibrosis were less likely to be DR4 positive (61%). No other HLA-DR variants were significantly increased in the different disease groups. HLA-DQB1*0501 which types serologically as DQw1 was increased in subjects with RA and peripheral airways disease as compared to rheumatoid subjects with normal lung function, but these differences were not statistically significant. DQB1*0601 was increased in subjects with bronchiectasis with or without RA (but only significantly so in RA-BR subjects) DQB1*0301, DQB1*0201 and DQA1*0501 frequencies were also increased in subjects with RA and bronchiectasis as compared to those with RA alone.

Complement C4 null alleles as a marker of gold or D-penicillamine toxicity in the treatment of rheumatoid arthritis.

C4 null alleles and HLA-DR antigens were defined in 48 rheumatoid arthritis (RA) subjects who had developed renal or heamatological side effects to gold or penicillamine, as compared to 33 RA subjects who had received the drugs for similar time periods without developing side effects. A C4A null allele was found in 56% of subjects with and 31% of those without side effects (P = 0.027, relative risk 2.8). A similar but statistically non-significant trend was observed with the C4B null allele (P = 0.64) resulting in a higher risk of drug toxicity in rheumatoid patients bearing either a C4A or C4B null allele (relative risk 5.7). Frequencies of DR3 and DR4 were similar in the two groups.