Fibroblast growth factor-1-inducible gene FR-17 encodes a nonmuscle α-actinin isoform

Polypeptide growth factor binding to cell surface receptors activates a cytoplasmic signaling cascade that ultimately promotes the expression of specific nuclear genes. As an approach to investigate the molecular mechanism of fibroblast growth factor (FGF)-1 mitogenic signaling, we have begun to identify and characterize FGF-1-inducible genes in murine NIH 3T3 cells. Here we report that one of these genes, termed FGF-regulated (FR)-17, is predicted to encode a nonmuscle isoform of α-actinin, an actin cross-linking protein found along microfilaments and in focal adhesion plaques. FGF-1 induction of α-actinin mRNA expression is first detectable at 2 h after mitogen addition and is dependent on de novo RNA and protein synthesis. Maximal α-actinin mRNA expression, corresponding to an approximately nineteenfold level of induction, is present after 12 h of FGF-1 stimulation. Western blot analysis indicated that FGF-1 stimulated cells also produce an increased amount of α-actinin protein. The FGF-1-related mitogen FGF-2, calf serum, several of the polypeptide growth factors present in serum, and the tumor promoter phorbol myristate acetate can also induce α-actinin mRNA expression. Finally, nonmuscle α-actinin mRNA is expressed in vivo in a tissue-specific manner, with relatively high levels detected in adult mouse intestine and kidney. These results indicate that nonmuscle α-actinin is a serum-, polypeptide growth factor-, and tumor promoter-inducible gene in mouse fibroblasts. © 1996 Wiley-Liss, Inc.     

Pleiotrophin gene expression is highly restricted and is regulated by platelet-derived growth factor.

Pleiotrophin (PTN) is a growth and neurite extension promoting polypeptide which is highly expressed in brain and in tissues derived from mesenchyme. The PTN gene is developmentally regulated and is closely related to the MK and RI-HB genes, both of which are developmentally regulated and induced by retinoic acid. We now have screened 17 cell lines and report that expression of the PTN gene in these cells is restricted to embryo fibroblasts and intestinal smooth muscle cells. However, NIH 3T3 cells stimulated by the platelet-derived growth factor (PDGF) express a marked increase in levels of PTN mRNA whereas retinoic acid failed to increase levels of PTN mRNA in NIH 3T3 cells or in F9 embryonal carcinoma cells within 72 hours of exposure. The results suggest that expression of the PTN gene is highly restricted and that the PTN gene is a new member of the PDGF-induced cytokine family.     

The levels of vascular smooth as well as skeletal muscle actin mRNAs differ substantially among both myoblast and fibroblast lines with different skeletal myogenic potentials.

Little is known about the factors which regulate vascular smooth muscle (vsm) actin gene expression during skeletal myogenesis in culture. We have therefore looked for differences in the levels of accumulation of vsm actin mRNA among six mouse cell lines differing in apparent myogenic potential or in the complement of myogenesis determination genes which they express: NIH 3T3 and 10T1/2 non-myogenic fibroblasts and four myogenic lines--3T3-MyoD1 and 10EMc11s, MyoD/myogenin expressing sublines of the fibroblast lines, derived by transfer into the parent lines of a MyoD cDNA expression construct; C2C12, which expresses all four known myogenesis determination genes; and BC3H1, which expresses myf-5, myogenin, little herculin, and no MyoD. In differentiated cells of all four myogenic lines, vsm actin mRNA was expressed at levels dramatically higher than in growth-arrested NIH 3T3 cells, consistent with expression of vsm actin mRNA as an intrinsic part of the skeletal myogenic program somehow directed by myogenesis determination gene products. Interestingly, however, the level of vsm actin mRNA in growth arrested C3H10T1/2 fibroblasts was also dramatically higher than that in NIH 3T3. In view of these findings, and of the relative ease with which 10T1/2 as opposed to NIH 3T3 cells can be converted to myogenic lines, we hypothesize that factors which can act to regulate vsm actin gene expression in the absence of myogenesis determination gene expression may also influence the skeletal myogenic potential of the cells in which they are found. Among the myogenic lines, the ratio of vsm to skm actin mRNA was highest in BC3H1 cells, raising the possibility that were these cells forced to express MyoD and/or more herculin, as do the other myogenic lines, the ratio would decrease. Thus both fibroblast and myogenic lines will be useful for investigating the mechanisms controlling skeletal myogenesis and vsm and skm actin gene expression during myogenesis.     

The levels of vascular smooth as well as skeletal muscle actin mRNAS differ substantially among both myoblast and fibroblast lines with different skeletal myogenic potentials.

Little is known about the factors which regulate vascular smooth muscle (vsm) actin gene expression during skeletal myogenesis in culture. We have therefore looked for differences in the levels of accumulation of vsm actin mRNA among six mouse cell lines differing in apparent myogenic potential or in the complement of myogenesis determination genes which they express: NIH 3T3 and 10T1/2 non-myogenic fibroblasts and four myogenic lines--3T3-MyoD1 and 10EMc11s, MyoD/myogenin expressing sublines of the fibroblast lines, derived by transfer into the parent lines of a MyoD cDNA expression construct; C2C12, which expresses all four known myogenesis determination genes; and BC3H1, which expresses myf-5, myogenin, little herculin, and no MyoD. In differentiated cells of all four myogenic lines, vsm actin mRNA was expressed at levels dramatically higher than in growth-arrested NIH 3T3 cells, consistent with expression of vsm actin mRNA as an intrinsic part of the skeletal myogenic program somehow directed by myogenesis determination gene products. Interestingly, however, the level of vsm actin mRNA in growth arrested C3H10T1/2 fibroblasts was also dramatically higher than that in NIH 3T3. In view of these findings, and of the relative ease with which 10T1/2 as opposed to NIH 3T3 cells can be converted to myogenic lines, we hypothesize that factors which can act to regulate vsm actin gene expression in the absence of myogenesis determination gene expression may also influence the skeletal myogenic potential of the cells in which they are found. Among the myogenic lines, the ratio of vsm to skm actin mRNA was highest in BC3H1 cells, raising the possibility that were these cells forced to express MyoD and/or more herculin, as do the other myogenic lines, the ratio would decrease. Thus both fibroblast and myogenic lines will be useful for investigating the mechanisms controlling skeletal myogenesis and vsm and skm actin gene expression during myogenesis.     

Bystander effect of normal fibroblasts for macrophages co-cultured with susceptible transformed target cells

Macrophages attack and kill pathologically changed, transformed and tumor cells. However, in some cases they may also support tumor growth, modulate the action of anticancer drugs, and even facilitate the development of drug resistance in tumor cells. Here we present data that bystander fibroblasts NIH3T3 were not only resistant to murine macrophages J774.2 but also blocked their killing action towards murine transformed fibroblasts L929. Macrophages were isolated from mixed cultures by means of CD11b specific immunomagnetic beads, and changes induced by their former co-culturing were studied using DNA microarray technology and other tests. An expression of candidate genes coding for cytokines and for signal transduction pathway proteins was estimated in macrophages in different variants of their co-culture with target cells. Changes in expression of mRNA for interleukin 1beta, NFkappaB, IkappaBalpha, gadd45, and CD5 were detected as the most prominent in the macrophages co-cultured with the transformed cells. Bystander NIH3T3 fibroblasts abolished these changes in the macrophages J774.2, and the level of expression of the above mentioned genes was close to the level seen in the macrophages which did not exert cytotoxicity towards the target fibroblasts. Potential implications and research perspectives of using the macrophage-target cell co-cultures with different bystander cellular partners are discussed.     

Mitogenic growth factors regulate differentially early gene mRNA expression: a study on two clones of 3T3 fibroblasts.

The relationship between cell proliferation and mRNA levels of the immediate early genes c-fos, c-jun, and jun B has been investigated in two clones of 3T3 fibroblasts (D1-3T3 and N2-3T3) upon treatment with basic fibroblast growth factor (bFGF), thrombin, phorbol 12-myristate 13-acetate (PMA) and dibutyryl cyclic AMP (Bt2cAMP). The 3T3-derived clone D1-3T3 almost stops dividing upon serum deprivation, while the N2-3T3 clone does not. The proliferation of the two clones was stimulated by thrombin and PMA and inhibited by Bt2cAMP. Basic FGF stimulated the growth of D1-3T3 but partly inhibited that of N2-3T3 cells. In spite of variable mitogenic response, immediate early genes, c-fos, c-jun, jun B, and c-myc, were induced by the growth factors and by PMA in both cell clones. In our experimental conditions the early gene mRNAs were expressed independently; i.e., the expression of one protooncogene had no bearing on the expression of the other. The cell growth was not directly related to the expression of a particular protooncogene mRNA. Data are presented showing that early gene mRNA expression induced by bFGF or thrombin was not mediated by protein kinase C activation while thrombin-induced mitosis was. Basic FGF induced a part of c-jun mRNA expression, but not mitosis, through a pertussis toxin-sensitive mechanism.     

Bovine milk lactoferrin induces synthesis of the angiogenic factors VEGF and FGF2 in osteoblasts via the p44/p42 MAP kinase pathway

Lactoferrin (LF) belongs to the transferrin family and is present in several physiological fluids, including milk and colostrum. LF has recently been identified as an anabolic factor for bone. Here we investigated whether bovine LF (bLF) induces synthesis of angiogenic factors by osteoblasts. If so, we examined the underlying mechanism. We found that bLF purified from milk increased the mRNA expression of vascular endothelial growth factor (VEGF) and fibroblast growth factor-2 (FGF2) in murine osteoblast-like MC3T3-E1 cells and primary murine osteoblasts in a time- and dose-dependent manner. Furthermore, bLF increased VEGF and FGF2 protein levels in MC3T3-E1 cells. In addition, treatment of MC3T3-E1 cells with bLF rapidly induced phosphorylation of p44/p42 mitogen-activated protein (MAP) kinase. The bLF-mediated increases in VEGF and FGF2 mRNA and protein were inhibited by U0126, a specific inhibitor of the upstream kinase that activates p44/p42 MAP kinase (MEK). Taken together, our results strongly suggest that bLF induces VEGF and FGF2 synthesis in a p44/p42 MAP kinase-dependent manner in MC3T3-E1 cells.     

Overexpression of the type II transforming growth factor-beta receptor inhibits fibroblasts proliferation and activates extracellular signal regulated kinase and c-Jun N-terminal kinase

Transforming growth factor-beta (TGF-beta) is a bimodal regulator of cellular growth. The cellular effects of TGF-beta depend on the intensity of signals emanating from TGF-beta receptors. Low levels of receptor activity are sufficient to stimulate cell proliferation, while higher degrees of receptor activation are associated with growth inhibition. To study the mechanisms of these effects, a tetracycline-inducible expression system was used to overexpress type II TGF-beta receptors in NIH 3T3 fibroblasts. Overexpressed type II TGF-beta receptors suppressed fibroblast proliferation elicited by TGF-beta1, fibroblast growth factor (FGF) or platelet-derived growth factor (PDGF). Accompanying these anti-proliferative effects, increases in extracellular-signal regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) activity were detected. Furthermore, PDGF alpha-, but not PDGF beta-receptor protein levels were reduced by type II TGF-beta receptor overexpression. In conclusion, our system is an excellent tool to study the molecular mechanisms of growth inhibition by TGF-beta in fibroblasts. Activation of JNK and ERK, or modulation of PDGF receptor expression may be involved in this process.     

Growth-Regulatory Activity of the Growth Arrest-Specific Gene, GAS1, in NIH3T3 Fibroblasts

The growth arrest-specific gene, Gas-1, is preferentially expressed in quiescent NIH3T3 cells and inhibits DNA synthesis, suggesting that Gas-1 may be a tumor suppressor gene. When GAS1 cDNA, under the control of the strong constitutive CMV promoter, was transfected into NIH3T3 cells, no stable transfectant cell lines were produced, confirming that high levels of expression of GAS1 mRNA inhibit proliferation. GAS1, under the control of a dexamethasone-inducible promoter, was also transfected into NIH3T3 cells, resulting in normal numbers of transfectant clones. When expression of GAS1 mRNA was induced with dexamethasone, the growth rate was greatly inhibited. Morphological changes characteristic of growth arrest were also observed. To determine if antisense inhibition of expression of Gas-1 will transform normal fibroblasts, GAS1 cDNA, cloned in the antisense orientation, was transfected into NIH3T3 cells and expression of endogenous Gas-1 mRNA was inhibited. The GAS1-antisense cells had altered morphology and grew to a much higher saturation density than control cell lines with a loss of contact inhibition. However, there was no change in requirements for serum or any development of anchorage-independence. Antisense inhibition of expression of GAS1 is therefore insufficient to transform the cells, suggesting that additional genetic events are required for a fully malignant phenotype.     

FGF21 attenuates lipolysis in human adipocytes - a possible link to improved insulin sensitivity

Fibroblast growth factor 21 (FGF21) is active in murine adipocytes and has beneficial metabolic effects in animal models of type 2 diabetes mellitus. We assessed whether FGF21 influences lipolysis in human adipocytes and 3T3-L1 cells. FGF21 had no short-time effect (h) while a 3-day incubation with FGF21 attenuated hormone-stimulated lipolysis. FGF21 did not influence the mRNA expression of genes involved in regulating lipolysis, but significantly reduced the expression of the lipid droplet-associated phosphoprotein perilipin without affecting differentiation. Via reduced release of fatty acids into the circulation, the anti-lipolytic effect could be a mechanism through which FGF21 promotes insulin sensitivity in man.     

The Mycoplasma hyorhinis p37 Protein Rapidly Induces Genes in Fibroblasts Associated with Inflammation and Cancer

The p37 protein at the surface of Mycoplasma hyorhinis cells forms part of a high-affinity transport system and has been found associated with animal and human cancers. Here we show in NIH3T3 fibroblasts, p37 rapidly induces the expression of genes implicated in inflammation and cancer progression. This gene activation was principally via the Tlr4 receptor. Activity was lost from p37 when the C-terminal 20 amino acids were removed or the four amino acids specific for the hydrogen bonding of thiamine pyrophosphate had been replaced by valine. Blocking the IL6 receptor or inhibiting STAT3 signalling resulted in increased p37-induced gene expression. Since cancer associated fibroblasts support growth, invasion and metastasis via their ability to regulate tumour-related inflammation, the rapid induction in fibroblasts of pro-inflammatory genes by p37 might be expected to influence cancer development.     

E2A-HLF-mediated cell transformation requires both the trans-activation domains of E2A and the leucine zipper dimerization domain of HLF.

The E2A-HLF fusion gene, formed by the t(17;19)(q22;p13) translocation in childhood acute pro-B-cell leukemia, encodes a hybrid protein that contains the paired trans-activation domains of E2A (E12/E47) linked to the basic region/leucine zipper DNA-binding and dimerization domain of hepatic leukemia factor (HLF). To assess the transforming potential of this novel gene, we introduced it into NIH 3T3 murine fibroblasts by using an expression vector that also contained the neomycin resistance gene. Cells selected for resistance to the neomycin analog G418 formed aberrant colonies in monolayer cultures, marked by increased cell density and altered morphology. Transfected cells also grew readily in soft agar, producing colonies whose sizes correlated with E2A-HLF expression levels. Subclones expanded from colonies with high levels of the protein reproducibly formed tumors in nude mice and grew to higher plateau-phase cell densities in reduced-serum conditions than did parental NIH 3T3 cells. By contrast, NIH 3T3 cells expressing mutant E2A-HLF proteins that lacked either of the bipartite E2A trans-activation domains or the HLF leucine zipper domain failed to show oncogenic properties, including anchorage-independent cell growth. Thus, both of the E2A trans-activation motifs and the HLF leucine zipper dimerization domain are essential for the transforming potential of the chimeric E2A-HLF protein, suggesting a model in which aberrant regulation of the expression pattern of downstream target genes contributes to leukemogenesis.