OncoPPi-informed discovery of mitogen-activated protein kinase kinase 3 as a novel binding partner of c-Myc.

Mitogen-activated protein kinase kinase 3 (MKK3) is a dual threonine/tyrosine protein kinase that regulates inflammation, proliferation and apoptosis through specific phosphorylation and activation of the p38 mitogen-activated protein kinase. However, the role of MKK3 beyond p38-signaling remains elusive. Recently, we reported a protein-protein interaction (PPI) network of cancer-associated genes, termed OncoPPi, as a resource for the scientific community to generate new biological models. Analysis of the OncoPPi connectivity identified MKK3 as one of the major hub proteins in the network. Here, we show that MKK3 interacts with a large number of proteins critical for cell growth and metabolism, including the major oncogenic driver MYC. Multiple complementary approaches were used to demonstrate the direct interaction of MKK3 with MYC in vitro and in vivo. Computational modeling and experimental studies mapped the interaction interface to the MYC helix-loop-helix domain and a novel 15-residue MYC-binding motif in MKK3 (MBM). The MBM in MKK3 is distinct from the known binding sites for p38 or upstream kinases. Functionally, MKK3 stabilized MYC protein, enhanced its transcriptional activity and increased expression of MYC-regulated genes. The defined MBM peptide mimicked the MKK3 effect in promoting MYC activity. Together, the exploration of OncoPPi led to a new biological model in which MKK3 operates by two distinct mechanisms in cellular regulation through its phosphorylation of p38 and its activation of MYC through PPI.

Alterations in c-Myc phenotypes resulting from dynamin-related protein 1 (Drp1)-mediated mitochondrial fission.

The c-Myc (Myc) oncoprotein regulates numerous phenotypes pertaining to cell mass, survival and metabolism. Glycolysis, oxidative phosphorylation (OXPHOS) and mitochondrial biogenesis are positively controlled by Myc, with myc-/- rat fibroblasts displaying atrophic mitochondria, structural and functional defects in electron transport chain (ETC) components, compromised OXPHOS and ATP depletion. However, while Myc influences mitochondrial structure and function, it is not clear to what extent the reverse is true. To test this, we induced a state of mitochondrial hyper-fission in rat fibroblasts by de-regulating Drp1, a dynamin-like GTPase that participates in the terminal fission process. The mitochondria from these cells showed reduced mass and interconnectivity, a paucity of cristae, a marked reduction in OXPHOS and structural and functional defects in ETC Complexes I and V. High rates of abortive mitochondrial fusion were observed, likely reflecting ongoing, but ultimately futile, attempts to normalize mitochondrial mass. Cellular consequences included reduction of cell volume, ATP depletion and activation of AMP-dependent protein kinase. In response to Myc deregulation, apoptosis was significantly impaired both in the absence and presence of serum, although this could be reversed by increasing ATP levels by pharmacologic means. The current work demonstrates that enforced mitochondrial fission closely recapitulates a state of Myc deficiency and that mitochondrial integrity and function can affect Myc-regulated cellular behaviors. The low intracellular ATP levels that are frequently seen in some tumors as a result of inadequate vascular perfusion could favor tumor survival by countering the pro-apoptotic tendencies of Myc overexpression.

C-MYC overexpression is required for continuous suppression of oncogene-induced senescence in melanoma cells.

Malignant melanomas often harbor activating mutations in BRAF (V600E) or, less frequently, in NRAS (Q61R). Intriguingly, the same mutations have been detected at higher incidences in benign nevi, which are largely composed of senescent melanocytes. Overexpression of BRAF(V600E) or NRAS(Q61R) in human melanocytes in vitro has been shown to induce senescence, although via different mechanisms. How oncogene-induced senescence is overcome during melanoma progression remains unclear. Here, we report that in the majority of analysed BRAF(V600E)- or NRAS(Q61R)-expressing melanoma cells, C-MYC depletion induced different yet overlapping sets of senescence phenotypes that are characteristic of normal melanocytes undergoing senescence due to overexpression of BRAF(V600E) or NRAS(Q61R), respectively. These senescence phenotypes were p16(INK4A)- or p53-independent, however, several of them were suppressed by genetic or pharmacological inhibition of BRAF(V600E) or phosphoinositide 3-kinase pathways, including rapamycin-mediated inhibition of mTOR-raptor in NRAS(Q61R)-expressing melanoma cells. Reciprocally, overexpression of C-MYC in normal melanocytes suppressed BRAF(V600E)-induced senescence more efficiently than NRAS(Q61R)-induced senescence, which agrees with the generally higher rates of activating mutations in BRAF than NRAS gene in human cutaneous melanomas. Our data suggest that one of the major functions of C-MYC overexpression in melanoma progression is to continuous suppress BRAF(V600E)- or NRAS(Q61R)-dependent senescence programs.

c-Myc depletion inhibits proliferation of human tumor cells at various stages of the cell cycle.

A major role for c-Myc in the proliferation of normal cells is attributed to its ability to promote progression through G(1) and into S phase of the cell cycle. The absolute requirement of c-Myc for cell cycle progression in human tumor cells has not been comprehensively addressed. In the present work, we used a lentiviral-based short hairpin RNA (shRNA) expression vector to stably reduce c-Myc expression in a large number of human tumor cell lines and in three different types of normal human cells. In all cases, cell proliferation was severely inhibited, with normal cells ultimately undergoing G(0)/G(1) growth arrest. In contrast, tumor cells demonstrated a much more variable cell cycle response with cells from several lines accumulating in S or G(2)/M phases. Moreover, in some tumor lines, the phase of cell cycle arrest caused by inhibition of c-Myc could be altered by depleting tumor suppressor protein p53 or its transcriptional target p21(CIP/WAF). Our data suggest that, as in the case of normal cells, c-Myc is essential for sustaining proliferation of human tumor cells. However its rate-limiting role in cell cycle control is variable and is reliant upon the status of other cell cycle regulators.

Transformation, genomic instability and senescence mediated by platelet/megakaryocyte glycoprotein Ibalpha.

GpIbalpha, a subunit of the von Willebrand factor receptor, functions during blood clotting to promote platelet adhesion and activation. GpIbalpha is widely expressed, is positively regulated by c-Myc and is essential for the promotion of c-Myc-mediated chromosomal instability. We now show that GpIbalpha is also a classical oncoprotein in which its deregulated expression leads to transformation, reduced growth factor requirements, increased resistance to apoptosis, and, in primary cells, p53-dependent senescence. Finally, GpIbalpha also promotes double-stranded DNA breaks, and induces profound nuclear dysmorphology, indicating that, in addition to its direct transforming function, it displays genotoxicity at several distinct levels. These findings identify novel functions for GpIbalpha and pathways through which c-Myc mediates transformation and global genomic destabilization.

Functional and physical communication between oncoproteins and tumor suppressors.

The discovery of oncogenes (c-onc's) and tumor suppressors (TS's) has led to the concept that cancer arises from defects in each of these classes of genes or their products. More recently, it has been appreciated that c-onc and TS proteins often affect one another's functions. Within this context, I review the two classical TS's, p53 and the retinoblastoma protein, and the consequences of their inactivation. The various forms of genomic instability (GI) that underly the high mutation rates of transformed cells are then discussed. Particular emphasis is placed upon the concept that GI is not only an integral part of the transformed state but is a prerequisite. Increased oxidative DNA damage, and/or an inabiliy to repair it, can lead to GI. The review then discusses recent observations showing that loss of the TS protein peroxiredoxin 1 (prdx 1) and increased expression of the c-onc protein c-Myc, each leads to increased oxidative DNA damage. The critical nature of the c-onc-TS interaction is underscored by that occurring between prdx1 and c-Myc, with the former protein regulating the production of DNA-damaging reactive oxygen species by the latter. The intimate association between these proteins and others serves as a paradigm for the exquisite balancing act that c-onc's and TS's must maintain in order to properly control normal DNA replication and cellular proliferation while simultaneously minimizing the acquisition of potentially neoplastic mutations.

Dynamic in vivo interactions among Myc network members.

Members of the Myc oncoprotein network (c-Myc, Max, and Mad) play important roles in proliferation, differentiation, and apoptosis. We expressed chimeric green fluorescent protein (GFP) fusions of c-Myc, Max, and three Mad proteins in fibroblasts. Individually, c-Myc and Mad proteins localized in subnuclear speckles, whereas Max assumed a homogeneous nuclear pattern. These distributions were co-dominant and dynamic, however, as each protein assumed the pattern of its heterodimeric partner when the latter was co-expressed at a higher level. Deletion mapping of two Mad members, Mad1 and Mxi1, demonstrated that the domains responsible for nuclear localization and speckling are separable. A non-speckling Mxi1 mutant was also less effective as a transcriptional repressor than wild-type Mxi1. c-Myc nuclear speckles were distinct from SC-35 domains involved in mRNA processing. However, in the presence of co-expressed Max, c-Myc, but not Mad, co-localized to a subset of SC-35 loci. These results show that Myc network proteins comprise dynamic subnuclear structures and behave co-dominantly when co-expressed with their normal heterodimerization partners. In addition, c-Myc-Max heterodimers, but not Max-Mad heterodimers, localize to foci actively engaged in pre-mRNA transcription/processing. These findings suggest novel means by which Myc network members promote transcriptional activation or repression.

Inverse regulation of cyclin B1 by c-Myc and p53 and induction of tetraploidy by cyclin B1 overexpression.

We have shown previously that mitotic spindle inhibitors allow the c-Myconcoprotein to uncouple mitosis from DNA synthesis, resulting in the acquisition of tetraploidy. This can also occur in the absence of spindle inhibition if c-Myc deregulation is combined with inactivation of the p53 tumor suppressor. Under these conditions, cyclin B1 protein is induced but retains its normal cell cycle regulation. We now show that the cyclin B1 promoter is directly but oppositely regulated by c-Myc and p53. Enforced expression of cyclin B1 also induces tetraploidy, either after mitotic spindle inhibition or in the absence of such inhibition if cyclin B1 is coexpressed with c-Myc. Cyclin B1 represents a new class of c-Myc target genes that is also regulated by p53. It is also the first identified downstream effector of c-Myc able to produce the chromosomal instability that characterizes virtually all tumor cells.

Mmip-2/Rnf-17 enhances c-Myc function and regulates some target genes in common with glucocorticoid hormones.

Members of the Mad family of basic-helix-loop-helix-leucine zipper proteins inhibit the transcriptional activity of the c-Myc oncoprotein. Mmip-2/Rnf-17 is a RING-finger protein that interacts with all four known Mad proteins, redistributes them to the cytoplasm, and thus enhances c-Myc function. We generated cell lines in which Mmip-2/Rnf-17 was rendered glucocorticoid (GC)-inducible. Stable expression of Mmip-/Rnf-17 resulted in the expected transport of the most abundant endogenous mad protein, Mxi1, to the cytoplasm. Compensatory increases in Mxi1 and Mad3 transcripts, similar to those previously described in Mad1 null hematopoietic cells, were also seen. Mmip-2/Rnf-17 also sensitized cells to several different pro-apoptotic stimuli and regulated a subset of c-Myc target genes. Unexpectedly, some of these genes were also found to be modulated solely by GCs. Thus, the inhibition of Mad proteins by Mmip-2/Rnf-17 modulates c-Myc function by enhancing its ability to regulate a subset of its potential target genes. Our results also identify a previously unrecognized overlap between genes regulated by c-Myc- and GCs and provide a potential molecular basis for their regulation of common cellular functions.

Promotion of growth and apoptosis in c-myc nullizygous fibroblasts by other members of the myc oncoprotein family.

c-myc nullizygous fibroblasts (KO cells) were used to compare the abilities of c-myc, N-myc and L-myc oncoproteins to accelerate growth, promote apoptosis, revert morphology, and regulate the expression of previously described c-myc target genes. All three myc oncoproteins were expressed following retroviral transduction of KO cells. The proteins all enhanced the growth rate of KO cells and significantly shortened the cell cycle transition time. They also accelerated apoptosis following serum deprivation, reverted the abnormal KO cell morphology, and modulated the expression of previously described c-myc target genes. In most cases, L-myc was equivalent to c-myc and N-myc in restoring all of the c-myc-dependent activities. These findings contrast with the previously reported weak transforming and transactivating properties of L-myc. Myc oncoproteins may thus impart both highly similar as well as dissimilar signals to the cells in which they are expressed.

Genetic dissection of c-myc apoptotic pathways.

All biological functions mediated by the c-myc oncoprotein require an intact transactivation domain (TAD). We compared TAD mutants for their ability to promote apoptosis of 32D myeloid cells in response to interleukin-3 (IL-3) deprivation and exposure to chemotherapeutic drugs, and to activate ornithine decarboxylase, an endogenous c-myc target. Different sub-regions of the TAD were required to mediate each function. cDNA microarrays were then used to identify multiple c-myc-regulated transcripts, some of which were also modulated by IL-3 or cytotoxic drugs, as well as by specific sub-regions of the TAD. Several of the c-myc-regulated transcripts had also been previously identified as targets for IFN-gamma. The functional consequences of their deregulation were manifested by a marked sensitivity of c-myc-overexpressing cells to IFN-gamma-mediated apoptosis. Our results establish that several well-characterized functions of c-myc are separable and correlate with the expression of a novel group of target genes, some of which also mediate the apoptotic action of IFN-gamma.

Mmip-2, a novel RING finger protein that interacts with mad members of the Myc oncoprotein network.

Mad proteins are basic-helix-loop-helix-leucine zipper (bHLH-ZIP)-containing members of the myc oncoprotein network. They interact with the bHLH-ZIP protein max, compete for the same DNA binding sites as myc-max heterodimers and down-regulate myc-responsive genes. Using the bHLH-ZIP domain of mad1 as a yeast two-hybrid 'bait', we identified Mmip-2, a novel RING finger protein that interacts with all mad members, but weakly or not at all with c-myc, max or unrelated bHLH or bZIP proteins. The mad1-Mmip-2 interaction is mediated by the ZIP domain in the former protein and by at least two regions in the latter which do not include the RING finger. Mmip-2 can disrupt max-mad DNA binding and can reverse the suppressive effects of mad proteins on c-myc-responsive target genes and on c-myc + ras-mediated focus formation in fibroblasts. Tagging with spectral variants of green fluorescent protein showed that Mmip-2 and mad proteins reside in separate cytoplasmic and nuclear compartments, respectively. When co-expressed, however, the proteins interact and translocate to the cellular compartment occupied by the more abundant protein. These observations suggest a novel way by which Mmip-2 can modulate the transcriptional activity of myc oncoproteins.

MYC oncogenes and human neoplastic disease.

c-myc, N-myc and L-myc are the three members of the myc oncoprotein family whose role in the pathogenesis of many human neoplastic diseases has received wide empirical support. In this review, we first summarize data, derived mainly from non-clinical studies, indicating that these oncoproteins actually serve quite different roles in vivo. This concept necessarily lies at the heart of the basis for the observation that the deregulated expression of each MYC gene is reproducibly associated with only certain naturally occurring malignancies in humans and that these genes are not interchangeable with respect to their aberrant functional consequences. We also review evidence implicating each of the above MYC genes in specific neoplastic diseases and have attempted to identify unresolved questions which deserve further basic or clinical investigation. We have made every attempt to review those diseases for which significant and confirmatory evidence, based on studies with primary tumor material, exists to implicate MYC members in their causation and/or progression.

C-myc overexpression and p53 loss cooperate to promote genomic instability.

p53 monitors genomic integrity at the G1 and G2/M cell cycle checkpoints. Cells lacking p53 may show gene amplification as well as the polyploidy or aneuploidy typical of many tumors. The pathways through which this develops, however, are not well defined. We demonstrate here that the combination of p53 inactivation and c-myc overexpression in diploid cells markedly accelerates the spontaneous development of tetraploidy. This is not seen with either N-myc or L-myc. Tetraploidy is accompanied by significantly higher levels of cyclin B and its associated cdc2 kinase activity. Mitotic spindle poisons accelerate the appearance of tetraploidy in cells either lacking functional p53 or overexpressing c-myc whereas the combination is additive. Restoration of p53 function in cells overexpressing c-myc causing rapid apoptosis, indicating that cells yet to become tetraploid have nonetheless suffered irreversible genomic and/or mitotic spindle damage. In the face of normal p53 function, such damage would either be repaired or trigger apoptotis. We propose that loss of p53 and overexpression of c-myc permits the emergence and survival of cells with increasingly severe damage and the eventual development of tetraploidy.

Differential apoptotic behaviors of c-myc, N-myc, and L-myc oncoproteins.

c-, N-, and L-myc are related nuclear oncoproteins that bind similar DNA sites and cooperate with activated ras oncogenes to transform primary fibroblasts. Although c-myc can also promote apoptosis in some cells after growth factor withdrawal or exposure to cytotoxic agents, roles for N- and L-myc in apoptosis remain undetermined. To address this, c-, N-, or L-myc were stably expressed in the interleukin 3 (IL-3)-dependent 32D hematopoietic cell line. The apoptotic response of each cell line was assessed after IL-3 withdrawal or treatment with four structurally unrelated cytotoxic agents. All three oncoproteins accelerated apoptosis after IL-3 withdrawal. In contrast, whereas c-myc overexpression generally sensitized cells to cytotoxic drugs, N-myc and L-myc overexpression produced resistance. myc expression tended to be associated with a more robust G2-M arrest after drug exposure, but this did not correlate with drug sensitivity or resistance. Bcl-2 and Bcl-X(L) protected control cells against apoptosis after either IL-3 withdrawal or drug exposure, although in some cases this effect could be overridden by myc oncoproteins, particularly N-myc and L-myc. Our results suggest that the apoptotic pathways activated upon IL-3 withdrawal and cytotoxic drug treatment are distinct and differentially affected by members of the myc and Bcl-2 families.

Distinct apoptotic responses imparted by c-myc and max.

The c-myc oncoprotein accelerates programmed cell death (apoptosis) after growth factor deprivation or pharmacological insult in many cell lines. We have shown that max, the obligate c-myc heterodimeric partner protein, also promotes apoptosis after serum withdrawal in NIH3T3 fibroblasts or cytokine deprivation in interleukin-3 (IL-3)-dependent 32D murine myeloid cells. We now show that c-myc- and max-overexpressing 32D cells differ in the nature of their apoptotic responses after IL-3 removal or treatment with chemotherapeutic compounds. In the presence of IL-3, c-myc overexpression enhances the sensitivity of 32D cells to Etoposide (Sigma, St Louis, MO), Adriamycin (Pharmacia, Columbus, OH), and Camptothecin (Sigma), whereas max overexpression increases sensitivity only to Camptothecin. Drug treatment of c-myc-overexpressing cells in the absence of IL-3 did not alter the spectrum of drug sensitivity other than to additively accelerate cell death. In contrast, enhanced sensitivity to Adriamycin, Etoposide, and Taxol (Bristol-Meyers Squibb, Princeton, NJ) was revealed in max-overexpressing cells concurrently deprived of IL-3. Differential rates of apoptosis were not strictly correlated with the ability of the drugs to promote G1 or G2/M arrest. Ectopic expression of Bcl-2 or Bcl-XL blocked drug-induced apoptosis in both cell lines. In contrast, whereas Bcl-2 blocked apoptosis in both cell lines in response to IL-3 withdrawal, Bcl-XL blocked apoptosis in max-overexpressing cells but not in c-myc-overexpressing cells. These results provide mechanistic underpinnings for the idea that c-myc and max modulate distinct apoptotic pathways.

Commonly occurring loss and mutation of the MXI1 gene in prostate cancer.

One of the most common chromosomal abnormalities in prostate cancer involves loss of 10q22-qter. Rarely, a smaller deletion, involving 10q24-q25, has been observed, suggesting the presence of a tumor suppressor gene at this site. We previously demonstrated that the MXI1 gene maps to 10q24-q25 and is mutated in some tumors with cytogenetically detectable deletions of this locus. MXI1 encodes a basic-helix-loop-helix protein that suppresses the transcriptional activity of the MYC oncoprotein by competing for the common dimerization partner, MAX, and binding to identical DNA sites. Because more than 90% of prostate tumors contain no cytogenetic abnormality of 10q, the relevance of MXI1 loss and/or mutation to the vast majority of cases remains unclear. We prospectively evaluated prostate tumors for loss of MXI1 by fluorescence in situ hybridization (FISH) and cytogenetic techniques. Twenty-one of 40 tumors (53%) demonstrated loss of a single MXI1 allele as determined by FISH. Ten cases with cytogenetically normal 10qs, but with FISH-documented deletion of MXI1, were examined at the molecular level, and eight mutations were identified, albeit at low frequency. Five of the mutant proteins were unable to bind DNA in association with MAX. We conclude that MXI1 gene loss in prostate cancer is common and most frequently involves a cytogenetically undetectable deletion.

Lack of transcriptional repression by max homodimers.

Max, a basic-helix-loop-helix-leucine zipper (bHLH-ZIP) protein, plays a central role in the transcriptional regulation of myc oncoprotein-responsive genes. Myc-max heterodimers bind to consensus E-box motifs near or within the promoters of these genes and activate gene expression, whereas heterodimers between max and members of the mad family of bHLH-ZIP proteins promote transcriptional repression. In contrast to all other members of the myc network, max readily homodimerizes and binds to identical E-box sites in vitro. However, the role for max homodimers in transcriptional repression in vivo is unclear. Upstream stimulatory factor (USF) is a bHLH-ZIP protein which does not interact with members of the myc-max-mad family. By replacing the HLH-ZIP domain of max with that from USF, we created a chimeric protein, max(USF), which was indistinguishable from max with respect to its ability to homodimerize and bind DNA. As expected, however, max(USF) was unable to heterodimerize with any of the tested max partner proteins and was incapable of suppressing c-myc target genes. Thus, transcriptional repression is an exclusive property of max-mad heterodimers and cannot be achieved by max homodimers alone.

Mmip1: a novel leucine zipper protein that reverses the suppressive effects of Mad family members on c-myc.

C-myc, a member of the basic helix-loop-helix-leucine zipper (bHLH-ZIP) protein family activates target genes in heterodimeric association with another bHLH-ZIP protein, Max. Max readily homodimerizes, competes with C-myc-Max heterodimers, and represses transcription. Four additional bHLH-ZIP proteins, Mad1, Mxi1, Mad3 and Mad4, heterodimerize with Max and also repress transcription of c-myc-responsive genes. We employed a yeast two-hybid approach to identify proteins which interact with Mxi. We identified a novel ZIP-containing protein, Mmip1 (Mad member-interacting protein 1) that strongly dimerizes with all four Mad members, but not with c-myc, Max, or with unrelated HLH proteins. The Mmip1-Mxi association is mediated by the ZIP domain of each polypeptide and is as strong or stronger than the associations between c-myc and Max or Max and Mxi1. In vitro, Mmip1 can inhibit DNA binding by Max-Mad heterodimers and, in vivo, can reverse the suppressive effects of Mad proteins on c-myc functions. Mmipl is found in a variety of cells types, is induced by serum stimulation, and can be co-immunoprecipitated from fibroblasts in association with Mxi1. By interfering with the dimerization between Max and Mad family member proteins, Mmip1 can indirectly up-regulate the transcriptional activity of c-myc and suppress the antiproliferative actions of Mad proteins.

Establishment of an apoptosis-resistant and growth-controllable cell line by transfecting with inducible antisense c-Jun gene.

F-MEL cells were transfected with the c-jun antisense gene located downstream of a glucocorticoid-inducible MMTV promoter, and the obtained cells were named c-jun AS cells. When the c-jun AS cells were treated with dexamethasone (DEX) in DMEM supplemented with 10% serum, the growth of the cells was completely suppressed for a duration of 16 days with a high cell viability exceeding 86%. The c-jun expression in the c-jun AS cells was suppressed moderately in the absence of DEX and strongly in the presence of DEX. The c-jun AS cells grew well and reached a density of 10(6) cells/mL without supplementation of any serum components. Viability was greater than 80% after the cells had been cultured for 8 days in the absence of DEX. The c-jun AS cells stayed at a constant cell density and high viability above 80% for 8 days when they were cultured in the presence of DEX under serum deprivation. In contrast, the wild type F-MEL cells were unable to grow and died by apoptosis in 3 days under serum deprivation. Internucleosomal cleavage of DNA, a landmark of apoptosis, was clearly detectable. Thus the c-jun AS cell line that is resistant to apoptosis induced by serum deprivation and can reversibly and viably be growth-arrested was established. A dual-signal model was proposed to explain the experimental result, the interlinked regulation of apoptosis, and growth by c-jun.