Farnesyl transferase inhibitors induce apoptosis of Ras-transformed cells denied substratum attachment

Farnesyl transferase inhibitors (FTIs) are a novel class of antitumor drugs that block the oncogenic activity of Ras. Because FTIs lack significant cell toxicity in vitro and in vivo, a significant question is how they cause tumor regression. We now report that FTIs are in fact potent activators of apoptosis in Ras-transformed cells if attachment to substratum is prevented. When cultured at high density or on polyHEMA, a nonadherent substrate, Ras-transformed cells exhibited massive DNA degradation and cell death within 24 h of treatment with the FTI L-739,749. Death was p53-independent and was inhibited by the apoptosis suppressor BCL-XL. Furthermore, apoptosis was significantly attenuated by ectopic expression of a farnesyl-independent form of RhoB, a Rho protein previously implicated as a critical target for inhibition by FTIs. The findings suggest a link between FTIs and Rho-dependent adhesion signaling. Furthermore, our work indicates that FTIs revert cells to a state in which cell-substratum attachment is necessary for viability and suggests that apoptosis forms the basis for drug-induced tumor regression.

BIN1 is a novel MYC-interacting protein with features of a tumour suppressor

BIN1 is a novel protein that interacts with the functionally critical Myc box regions at the N terminus of the MYC oncoprotein. BIN1 is structurally related to amphiphysin, a breast cancer-associated autoimmune antigen, and RVS167, a negative regulator of the yeast cell cycle, suggesting roles in malignancy and cell cycle control. Consistent with this likelihood, BIN1 inhibited malignant cell transformation by MYC. Although BIN1 is expressed in many normal cells, its levels were greatly reduced or undetectable in 14/27 carcinoma cell lines and 3/6 primary breast tumours. Deficits were functionally significant because ectopic expression of BIN1 inhibited the growth of tumour cells lacking endogenous message. We conclude that BIN1 is an MYC-interacting protein with features of a tumour suppressor.

Resistance of a variant ras-transformed cell line to phenotypic reversion by farnesyl transferase inhibitors

Pharmacological inhibitors of the housekeeping enzyme farnesyl transferase (FT) inhibit the growth of ras-transformed cells in vitro and in vivo without antiproliferative effects on normal cells. In one direction to analyze the basis for this selectivity and to study modes of drug resistance that arise in animals, we characterized a variant ras-transformed cell line, 749r-1, which was resistant to phenotypic reversion with FT inhibitors. The transformed phenotype, growth potential, and actin cytoskeleton of 749r-1 cells were unaffected by treatment with the FT inhibitor 1-739,749 at concentrations up to 30-fold higher than those sufficient to revert ras-transformed cells. Resistance correlated with a reduced ability of L-739,749 to inhibit the farnesylation of Ras and lamin B and with a reduction in the susceptibility of endogenous FT to drug inhibition. These effects were not due to mutation of the FT subunits, changes in intracellular drug accumulation, or amplification of the multiple drug resistance gene (MDR). However, a similar reduction in the ability of L-739,749 to inhibit Ras farnesylation was also seen in ras-transformed cells rendered resistant by ectopic expression of farnesyl-independent RhoB, suggesting some mechanistic overlap. We concluded that 749r-1 cells sustained a stable alteration that conferred drug resistance by a novel mechanism.

c-Myc induces apoptosis in epithelial cells by both p53-dependent and p53-independent mechanisms

We tested the hypothesis that wild-type p53 activity is required for c-Myc-dependent apoptosis in epithelial cells. Primary baby rat kidney epithelial cell lines were generated by immortalization through the concerted action of c-Myc and a temperature-sensitive (ts) dominant inhibitory mutant allele of p53 (BRK myc/p53ts cells). When shifted to the permissive temperature for wild-type p53 activity, the BRK myc/p53ts cells underwent growth arrest and apoptosis. However, apoptosis also could be induced by serum deprivation at the nonpermissive temperature, when p53 was in the mutant state. Bcl-2 suppressed both modes of cell death. Apoptosis induced by wild-type p53 but not by serum deprivation was accompanied by G1 cell cycle arrest and increased expression of the Bcl-2 antagonist Bax. We concluded that c-Myc could induce apoptosis in epithelial cells by at least two mechanisms that could be distinguished by their p53 requirement. Our results support the possibility that c-Myc-dependent cell death might be exploited for therapeutic ends during carcinoma development, without regard to p53 status of the target cell.

Evidence that farnesyltransferase inhibitors suppress Ras transformation by interfering with Rho activity

Small-molecule inhibitors of the housekeeping enzyme farnesyltransferase (FT) suppress the malignant growth of Ras-transformed cells. Previous work suggested that the activity of these compounds reflected effects on actin stress fiber regulation rather than Ras inhibition. Rho proteins regulate stress fiber formation, and one member of this family, RhoB, is farnesylated in vivo. Therefore, we tested the hypothesis that interference with RhoB was the principal basis by which the peptidomimetic FT inhibitor L-739,749 suppressed Ras transformation. The half-life of RhoB was found to be approximately 2 h, supporting the possibility that it could be functionally depleted within the 18-h period required by L-739,749 to induce reversion. Cell treatment with L-739,749 disrupted the vesicular localization of RhoB but did not effect the localization of the closely related RhoA protein. Ras-transformed Rat1 cells ectopically expressing N-myristylated forms of RhoB (Myr-rhoB), whose vesicular localization was unaffected by L-739,749, were resistant to drug treatment. The protective effect of Myr-rhoB required the integrity of the RhoB effector domain and was not due to a gain-of-function effect of myristylation on cell growth. In contrast, Rat1 cells transformed by a myristylated Ras construct remained susceptible to growth inhibition by L-739,749. We concluded that Rho is necessary for Ras transformation and that FT inhibitors suppress the transformed phenotype at least in part by direct or indirect interference with Rho, possibly with RhoB itself.

Activated H-ras rescues E1A-induced apoptosis and cooperates with E1A to overcome p53-dependent growth arrest

The adenovirus E1A oncogene products stimulate DNA synthesis and cell proliferation but fail to transform primary baby rat kidney (BRK) cells because of the induction of p53-mediated programmed cell death (apoptosis). Overexpression of dominant mutant p53 (to abrogate wild-type p53 function) or introduction of apoptosis inhibitors, such as adenovirus E1B 19K or Bcl-2 oncoproteins, prevents E1A-induced apoptosis and permits transformation of BRK cells. The ability of activated Harvey-ras (H-ras) to cooperate with E1A to transform BRK cells suggests that H-ras is capable of overcoming the E1A-induced, p53-dependent apoptosis. We demonstrate here that activated H-ras was capable of suppressing apoptosis induced by E1A and wild-type p53. However, unlike Bcl-2 and the E1B 19K proteins, which completely block apoptosis but not p53-dependent growth arrest, H-ras expression permitted DNA synthesis and cell proliferation in the presence of high levels of wild-type p53. The mechanism by which H-ras regulates apoptosis and cell cycle progression is thereby strikingly different from that of the E1B 19K and Bcl-2 proteins. BRK cells transformed with H-ras and the temperature sensitive murine mutant p53(val 135), which lack E1A, underwent growth arrest at the permissive temperature for wild-type p53. p53-dependent growth arrest, however, could be relieved by E1A expression. Thus, H-ras alone was insufficient and cooperation of H-ras and E1A was required to override growth suppression by p53. Our data further suggest that two complementary growth signals from E1A plus H-ras can rescue cell death and thus permit transformation.

Critical role of Rho in cell transformation by oncogenic Ras

We demonstrate that Rho, a regulator of cytoskeletal actin, is necessary for Ras transformation. A dominant inhibitory Rho gene (RhoBN19) specifically suppressed Rat1 cell focus formation induced by oncogenic Ras but not by Raf. An activated Rho gene (RhoBV14) lacked focus formation activity but augmented the focus formation activity of both oncogenes. NIH3T3 cell lines expressing RhoBV14 grew to higher saturation density and displayed reduced serum and anchorage requirements for growth. We concluded that Rho played a role in cell growth regulation and was required for transformation by oncogenic Ras but not Raf. A model for Ras signal transduction proposing separate Rho-dependent and Raf-dependent pathways is discussed.

c-Myc cooperates with activated Ras to induce the cdc2 promoter

Expression of c-myc with constitutively active mutants of the ras gene results in the cooperative transformation of primary fibroblasts, although the precise mechanism by which these genes cooperate is unknown. Since c-Myc has been shown to function as a transcriptional activator, we have examined the ability of c-Myc and activated Ras (H-RasV-12) to cooperatively induce the promoter activity of cdc2, a gene which is critical for cell cycle progression. Microinjection of expression constructs encoding H-RasV-12 and c-Myc along with a cdc2 promoter-luciferase reporter plasmid into quiescent cells led to an increase in cdc2 promoter activity approximately 30 h after injection, a period which coincides with the S-to-G2/M transition in these cells. Expression of H-RasV-12 alone weakly activated the cdc2 promoter, while expression of c-Myc alone had no effect. Mutants of c-Myc lacking either the leucine zipper dimerization domain or the phosphoacceptor site Ser-62 could not cooperate with H-RasV-12 to induce the cdc2 promoter. These mutants also lacked the ability to cooperate with H-RasV-12 to stimulate DNA synthesis. Deletion analysis identified a distinct region of the cdc2 promoter which was required for c-Myc responsiveness. Taken together, these observations suggest a mechanistic link between the molecular activities of c-Myc and Ras and induction of the cell cycle regulator Cdc2.

Farnesyltransferase inhibition causes morphological reversion of ras-transformed cells by a complex mechanism that involves regulation of the actin cytoskeleton

A potent and specific small molecule inhibitor of farnesyl-protein transferase, L-739,749, caused rapid morphological reversion and growth inhibition of ras-transformed fibroblasts (Rat1/ras cells). Morphological reversion occurred within 18 h of L-739,749 addition. The reverted phenotype was stable for several days in the absence of inhibitor before the transformed phenotype reappeared. Cell enlargement and actin stress fiber formation accompanied treatment of both Rat1/ras and normal Rat1 cells. Significantly, inhibition of Ras processing did not correlate with the initiation or maintenance of the reverted phenotype. While a single treatment with L-739,749 was sufficient to morphologically revert Rat1/ras cells, repetitive inhibitor treatment was required to significantly reduce cell growth rate. Thus, the effects of L-739,749 on transformed cell morphology and cytoskeletal actin organization could be separated from effects on cell growth, depending on whether exposure to a farnesyl-protein transferase inhibitor was transient or repetitive. In contrast, L-739,749 had no effect on the growth, morphology, or actin organization of v-raf-transformed cells. Taken together, the results suggest that the mechanism of morphological reversion is complex and may involve farnesylated proteins that control the organization of cytoskeletal actin.

Ras regulatory interactions: novel targets for anti-cancer intervention?

Advances in the understanding of Ras oncoprotein function suggest novel points for anti-tumor intervention. First, upstream-acting guanine nucleotide exchange factors and SH2/SH3 domain-containing adaptor proteins that link Ras with growth factor receptor tyrosine kinases have recently been characterized. Second, work on downstream-acting Ras effector functions including the Ras GTPase-activating protein (p120GAP) and the Raf kinase has revealed direct biochemical interactions that are functionally required for oncogenic Ras signalling. We summarize progress in these areas and discuss the potential for novel applications to anti-cancer chemotherapy.

Negative growth selection against rodent fibroblasts targeted for genetic inhibition of farnesyl transferase

The Ras oncoprotein must be modified by farnesyl transferase (FTase) for biological activity. Therefore, inhibition of FTase may offer a means to block ras induced cell transformation. To address this hypothesis, we have introduced antisense and dominant inhibitory FTase expression plasmids into a panel of normal, mutant ras-, and mos- transformed rodent fibroblasts in an effort to genetically suppress FTase activity. Antisense FTase constructs reduced colony formation efficiency approximately 29% in normal and approximately 41% in ras-transformed cells relative to control plasmids. In contrast, antisense FTase plasmids did not exhibit a statistically significant effect on colony formation efficiency in mos-transformed transfectants. FTase alpha N199K is a mutant form of the alpha subunit of FTase that exhibits dominant inhibitory activity versus native FTase. Only mos-transformed transfectants exhibited expression of alpha N199K RNA in 15 of 16 fibroblast lines that were randomly selected and characterized. Our data suggest that genetic inhibition of FTase may result in a selection against animal cell growth.

Characterization of recombinant human farnesyl-protein transferase: cloning, expression, farnesyl diphosphate binding, and functional homology with yeast prenyl-protein transferases

We have isolated cDNAs encoding the alpha and beta subunits of human farnesyl-protein transferase (FPTase). The proteins encoded by these two cDNAs are 93-95% identical to the corresponding subunits of bovine and rat FPTase and show regions of homology with proteins encoded by Saccharomyces cerevisiae prenyl-protein transferase genes. Human FPTase expressed in Escherichia coli from a translationally coupled operon had kinetic properties similar to those of FPTase isolated from bovine brain. Examination of farnesyl diphosphate binding indicated that while neither individual subunit was capable of isoprenoid binding, a radiolabeled farnesyl diphosphate analog could be specifically photo-cross-linked to the beta subunit of FPTase holoenzyme. To further analyze subunit structure-function and to detect functional similarities with yeast prenyl-protein transferases (FPTase and two geranylgeranyl-protein transferases), amino acid changes homologous to those found in mutant yeast prenyl-protein transferase subunits were made in the subunits of human FPTase. Substitutions in either the alpha or beta subunits that decrease the activity of yeast prenyl-protein transferases were also observed to impair human FPTase. Kinetic analyses showed that these mutant human FPTases have Km and kcat values that are altered with respect to wild-type human FPTase.

Recognition by Max of its cognate DNA through a dimeric b/HLH/Z domain

The three-dimensional structure of the basic/helix-loop-helix/leucine zipper domain of the transcription factor Max complexed with DNA has been determined by X-ray crystallography at 2.9 A resolution. Max binds as a dimer to its recognition sequence CACGTG by direct contacts between the alpha-helical basic region and the major groove. This symmetric homodimer, a new protein fold, is a parallel, left-handed, four-helix bundle, with each monomer containing two alpha-helical segments separated by a loop. The two alpha-helical segments are composed of the basic region plus helix 1 and helix 2 plus the leucine repeat, respectively. As in GCN4, the leucine repeat forms a parallel coiled coil.

Biphasic effect of Max on Myc cotransformation activity and dependence on amino- and carboxy-terminal Max functions

In Ras cotransformation assays, Max exhibited a biphasic effect on Myc transformation activity. Cotransfection of low levels of Max expression plasmid stimulated Myc transformation activity, but cotransfection of high levels suppressed it. Mutations in the functionally undefined Max amino- and carboxy-terminal regions outside of the B/HLH/LZ motif partly separated these activities, suggesting various modes of Max regulation. We demonstrate that the Max protein is a nuclear protein in vivo and identify a carboxy-terminal region similar to nuclear localization signals whose integrity is necessary for efficient localization. Two mutants that delete amino- or carboxy-terminal consensus signals for casein kinase II (CKII) exhibited altered gel mobility and DNA-binding potential in vitro and showed modified transforming potential in the Ras cotransformation assay, suggesting that CKII or a CKII-related enzyme may regulate Max function in vivo. Our data suggest that both the ratio of Myc/Max hetero-oligomers to Max homo-oligomers and Max-specific regulation can contribute to determining the biological activity of Myc in vivo.

A new bind for Myc

Recent studies centered on the c-Myc basic/helix-loop-helix/leucine zipper (B/HLH/LZ) motifs have led to the identification of a DNA recognition sequence for c-Myc and the isolation of a novel protein that forms a DNA-binding complex with c-Myc in vitro. These advances may make it possible to address directly the long-standing question of c-Myc function in vivo.

Association of Myn, the murine homolog of max, with c-Myc stimulates methylation-sensitive DNA binding and ras cotransformation

Myn, a novel murine approximately 18 kd basic/helix-loop-helix/"leucine zipper" (B/HLH/LZ) protein, forms a specific DNA-binding complex with the c-Myc oncoprotein through the HLH/LZ motif in both proteins. c-Myc/Myn recognizes a c-Myc-binding site (GACCACGTGGTC) with higher affinity than either protein by itself. CpG methylation of the recognition site greatly inhibits DNA binding, suggesting that DNA methylation may regulate the c-Myc/Myn complex in vivo. In 3T3 fibroblasts, Myn mRNA levels are induced several-fold by serum with delayed early kinetics, suggesting regulation by immediate-early gene products. Coexpression of Myn in a myc/ras rat embryo fibroblast focus formation assay specifically augmented c-myc transforming activity. We suggest that interaction of Myn with c-Myc stabilizes sequence-specific DNA binding in vivo.

Mbh 1: a novel gelsolin/severin-related protein which binds actin in vitro and exhibits nuclear localization in vivo

We describe the characterization of a novel cDNA, mbh1 (myc basic motif homolog-1), which was found during a search for candidate factors which might interact with the c-Myc oncoprotein. Embedded within the amino acid sequence encoded by mbh1 is a region distantly related to the basic/helix-loop-helix (B/HLH) DNA-binding motif and a potential nuclear localization signal. Mbh1 encodes a polypeptide structurally similar to the actin-severing proteins gelsolin and severin. Translation of mbh1 RNA in rabbit reticulocyte extracts produces an approximately 45 kd protein capable of binding actin-coupled agarose beads in vitro in a Ca2(+)-dependent manner. Antiserum raised to a trpE/mbh1 bacterial fusion protein recognizes an approximately 45 kb protein in murine 3T3 fibroblasts, suggesting that the cDNA encodes the complete Mbh1 protein. Examination of Mbh1 localization in 3T3 fibroblasts by indirect immunofluorescence reveals a larger cell population showing diffuse staining, and a smaller population exhibiting a distinct nuclear stain. Western analysis corroborates this intracellular localization and indicates that total cellular levels and localization of Mbh1 are not affected by the cell growth state. The data suggest that Mbh1 may play a role in regulating cytoplasmic and/or nuclear architecture through potential interactions with actin.

Methylation-sensitive sequence-specific DNA binding by the c-Myc basic region

The function of the c-Myc oncoprotein and its role in cell growth control is unclear. A basic region of c-Myc is structurally related to the basic motifs of helix-loop-helix (HLH) and leucine zipper proteins, which provide sequence-specific DNA binding function. The c-Myc basic region was tested for its ability to bind DNA by attaching it to the HLH dimerization interface of the E12 enhancer binding factor. Dimers of the chimeric protein, termed E6, specifically bound an E box element (GGCCACGTGACC) recognized by other HLH proteins in a manner dependent on the integrity of the c-Myc basic motif. Methylation of the core CpG in the E box recognition site specifically inhibited binding by E6, but not by two other HLH proteins. Expression of E6 (but not an E6 DNA binding mutant) suppressed the ability of c-myc to cooperate with H-ras in a rat embryo fibroblast transformation assay, suggesting that the DNA recognition specificity of E6 is related to that of c-Myc in vivo.

Posttranscriptional regulation of cellular gene expression by the c-myc oncogene

The c-myc oncogene has been implicated in the development of many different cancers, yet the mechanism by which the c-myc protein alters cellular growth control has proven elusive. We used a cDNA hybridization difference assay to isolate two genes, mr1 and mr2, that were constitutively expressed (i.e., deregulated) in rodent fibroblast cell lines immortalized by transfection of a viral promoter-linked c-myc gene. Both cDNAs were serum inducible in quiescent G0 fibroblasts, suggesting that they are functionally related to cellular proliferative processes. Although there were significant differences in cytoplasmic mRNA levels between myc-immortalized and control cells, the rates of transcription and mRNA turnover of both genes were similar, suggesting that c-myc regulates mr1 and mr2 expression by some nuclear posttranscriptional mechanism. mr1 was also rapidly (within 2 h) and specifically induced by dexamethasone in BALB/c cell lines expressing a mouse mammary tumor virus long terminal repeat-driven myc gene, under conditions where other growth factor-inducible genes were unaffected. A frameshift mutation in the mouse mammary tumor virus myc gene destroyed the dexamethasone stimulation of mr1, indicating that c-myc protein is required for the effect. As in the myc-immortalized cells, the induction of mr1 by c-myc occurred without detectable changes in mr1 transcription or cytoplasmic mRNA stability, implicating regulation, either direct or indirect, through a nuclear posttranscriptional mechanism. These results provide evidence that c-myc can rapidly modulate cellular gene expression and suggest that c-myc may function in gene regulation at the level of RNA export, splicing, or nuclear RNA turnover.