![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Molecular and Cellular Biology, June 2006, p. 4226-4239, Vol. 26, No. 11
0270-7306/06/$08.00+0 doi:10.1128/MCB.01959-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Departments of Pharmacology,1 Genetics, Norris Cotton Cancer Center, Dartmouth Medical School, One Medical Center Drive, Lebanon, New Hampshire 03756,2 Department of Molecular Biology, Princeton University, Princeton, New Jersey 085403
Received 6 October 2005/ Returned for modification 12 December 2005/ Accepted 12 March 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MBIV is not a simple loss-of-function mutant because it is hyperactive for G2 arrest in primary cells. Microarray analysis of genes regulated by N-MycMBIV reveals that it is weakened for transactivation and repression but not nearly as defective as N-MycMBII. Although the mutated region is not part of the previously defined DNA binding domain, we find that N-MycMBIV has a significantly lower affinity for DNA than the wild-type protein in vitro. Furthermore, chromatin immunoprecipitation shows reduced binding of N-MycMBIV to some target genes in vivo, which correlates with the defect in transactivation. Thus, this conserved domain has an unexpected role in Myc DNA binding activity. These data also provide a novel separation of Myc functions linked to the modulation of DNA binding activity. |
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Myc is a basic helix-loop-helix (bHLH)/leucine zipper transcription factor that operates as a heterodimer with Max. Myc transactivates in part by binding to the TRRAP complex, recruiting histone acetylase activity to cellular promoters (7, 22, 39, 40). There is also evidence that Myc activates genes using CBP and Skp2 as cofactors (30, 55, 56), and Myc may also promote transcriptional elongation through recruitment of the positive transcription elongation factor b (16). Whole-genome array analysis has revealed that Myc binds to a large fraction of cellular promoters, activating around 5 to 15% of genes weakly depending on the cellular context and experimental system (6, 12, 20, 24, 27, 33, 37, 46, 50, 57). Defining which genes are responsible for driving the different Myc phenotypes continues to be a challenge.
The Myc family members, c-Myc, N-Myc, and L-Myc, all function as oncogenes in different tumors and have a high degree of sequence conservation (11). c-Myc and N-Myc are particularly well conserved and have equivalent oncogenic activities (43). Furthermore, their coding regions can substitute for each other in mouse development (36). Myc proteins are also well conserved across species, which is reflected in the observation that the Drosophila myc gene, dmyc, can functionally substitute for mammalian c-Myc (49, 54). Characterization of the Myc protein has revealed several domains that are critical for different activities. The DNA binding domain was established by comparison to other bHLH proteins and found to be essential for gene activation (4, 5). In the N terminus, the prominent evolutionarily conserved domains are called Myc homology boxes (MBs). Of particular note is MBII, which is necessary for full transcriptional activation and repression. Mutation of this domain inhibits the binding of most Myc coactivators and also inhibits most Myc phenotypes (13, 18, 23, 43, 47, 53). In fact, the MBII deletion phenotype is so severe that MBII may be interpreted to be necessary for the integrity of the N terminus as a whole rather than as a distinct domain. The MBI domain has a more complex phenotype which is dependent on the extent of the mutation. MBI is a site of phosphorylation and is the only consistent site of Myc protein mutation found in some tumors (3, 35). MBI mutations can affect Myc protein turnover (26, 51), and larger deletions indicate that this domain is necessary for full activity in transformation assays (30, 53). Recent work has defined a third region of the Myc N terminus, MBIII, as a mediator of apoptosis, transformation, and tumorigenesis (28).
Here we report the first characterization of a conserved domain within the N-Myc protein, Myc317-337, designated MBIV. Deletions within this domain cause unique changes in the induction of apoptosis, transformation, and G2 arrest that are reflective of an unexpected defect in Myc DNA binding activity. These data suggest novel conformational and/or cofactor requirements that mediate Myc biological activity.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MBIV mutant was created using standard techniques and verified by sequencing. The mouse N-MycMBII mutant was previously described (43). Rat c-myc null fibroblasts (HO15.19 cells), human embryonic kidney 293 cells, human primary fibroblast BJ cells, RK3E cells, Rat-1a fibroblasts, rat embryo fibroblasts (REFs), mouse embryo fibroblasts (MEFs), and retroviral producer PhoeNX cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS). PhoeNX cells were transfected by using Fugene (Roche) according to the manufacturer's instructions, and infections were performed using standard protocols. Cells were maintained in 150 µg/ml hygromycin (Calbiochem) to select for cells containing LXSH-based constructs and 0.5 mg/ml G418 (Sigma) to select for pcDNA3.1-based constructs. Cell counts. Cells (2 x 104) were plated in six-well plates. Cells were trypsinized each day and counted using a hemocytometer, with at least 100 cells being counted. Cell counts were normalized to cells counted on the first day after seeding for each cell line. The experiment was performed twice.
CFSE staining. Cells (106) were trypsinized, resuspended in 1 ml of 0.1% FCS-phosphate-buffered saline containing 5 µM, 5 (and 6)-carboxyfluorescein diacetate (CFDA) succinimidyl ester (SE) (CFSE; Molecular Probes catalog no. of C1157), and incubated for 10 min at 37°C. Cells were washed three times in phosphate-buffered saline and plated onto three 10-cm-diameter plates. Cells were trypsinized on days 1, 2, and 3 and analyzed by fluorescence-activated cell sorter (FACS) using a FACScan. Mean fluorescence intensity values were measured, and differences between measurements on subsequent days were calculated.
Apoptosis. Cells maintained in 0.1% serum for 24 h were ethanol fixed and incubated overnight in 8 µg/ml propidium iodide (PI; Sigma) and 40 µg/ml RNase H (Sigma). PI content was analyzed using FACScan.
Transformation and soft agar assays.
REFs were extracted from day-15 rat embryos. Two days after extraction, 70% confluent 10-cm-diameter plates were transfected with 1 µg human H-RasG12V and 1 µg CßF (vector), CßF N-MycWT, N-MycMBII, N-MycMBIV, and 0.1 µg pcDNA3.1 (G418 resistance). For each transfection, three plates were maintained for 10 days in DMEM-5% FCS, and one plate was selected with G418. After 10 days, foci were counted and normalized to G418-resistant colonies. Foci formed relative to N-MycWT were calculated. For soft agar, Rat-1a fibroblasts were infected with LXSH N-MycWT, N-MycMBII, N-MycMBIV, or vector control, and pools were drug selected with hygromycin. Cells were plated in soft agar in triplicate in six-well plates. The bottom layer consisted of 0.6% noble agar-DMEM-10% FCS, and the upper layer consisted of 1 x 104 cells suspended in 0.3% noble agar-DMEM-10% FCS. After 12 days, the number of foci larger than 50 µm was counted in five fields from each well. Foci formed relative to N-MycWT were calculated. For focus formation, 104 cells of the same Rat-1a lines above were mixed with 2 x 106 Rat-1a cells and plated. Cells were maintained for 10 days, fixed in methanol, and stained with 0.1% methylene blue. Foci were counted for cultures in triplicate. The RK3E transformation assay was performed as described previously (21). Briefly, viral supernatants from a 10-cm plate of PhoeNX cells were diluted and used to infect a 10-cm plate of RK3E cells with LXSH (vector) or the same vector with N-MycWT or N-MycMBIV. Cells were placed in selection to measure transfection efficiency or maintained for 3 weeks to allow focus formation. After 3 weeks, cells were fixed in methanol and stained with 0.1% methylene blue. Hygromycin-resistant colonies and foci were counted to determine the relative number of foci formed for two independent experiments.
Subcellular fractionation.
Myc null cells expressing N-MycWT, N-MycMBII, N-MycMBIV, and vector control were lysed by Dounce homogenization in hypotonic buffer. Nuclei and cytosol were separated by centrifugation at 1,600 x g. Nuclei were subsequently washed twice, and cytosol was clarified by further centrifugation at 6,000 x g. Nuclei were solubilized in F buffer containing 0.5% Triton. Protein content of the fractions was measured using the Bradford assay, and samples were normalized for protein content. Nuclear and cytosolic material from an equivalent amount of cells was immunoblotted for N-Myc, RNA polymerase II, and tubulin.
Coimmunoprecipitation. Plates (with 10- or 15-cm diameters) of subconfluent cells were lysed in 1 or 2 ml of F buffer (52), and extracts were incubated overnight at 4°C with 20 µl anti-FLAG antibody-conjugated beads (Sigma) or 50 µl protein A/G beads with 1 µg of the relevant antibody. Beads were washed five times and resuspended in loading buffer. Immunoprecipitated protein was resolved on a sodium dodecyl sulfate-polyacrylamide gel, transferred to a polyvinylidene difluoride membrane, and blotted. Polyclonal anti-N-Myc antibodies and polyclonal anti-Max antibody were purchased from Santa Cruz Biotech.
G2 arrest.
Primary human fibroblast cells were transduced with retrovirus generated by transfecting packaging cells with LPCX, LPC-N-MycERTAM, LPC-N-MycMBII-ERTAM, or LPC-N-MycMBIV-ERTAM. Transduced cells were selected with 0.8 µg/ml puromycin. Resulting lines were not treated or treated for 48 h with 0.8 µM 4-hydroxytamoxifen (OHT). MEFs were transduced with retrovirus generated by transfecting packaging cells with LXSH, LXSH N-Myc, and LXSH N-MycMBIV. Cells were cultured in 0.5% FCS for 12 h. Cells were harvested for FACS analysis by trypsinization, fixed with ethanol, and then stained with propidium iodide using standard staining protocols. Samples were analyzed using the single laser FACScan. Resulting files were analyzed using FlowJo software.
EMSA. Plates (10-cm diameter) of 293 cells were transfected with 1 µg Cß S-Max and 1 µg Cß S FLAG-N-Myc constructs for 2 days and lysed with 1 ml F buffer. Cell extract (2 µl) was mixed with 19 µl electrophoretic mobility shift assay (EMSA) buffer (20 mM Tris, pH 8.4, 50 mM KCl, 3 mM MgCl2, 1 mM dithiothreitol, 1 mM ß-mercaptoethanol, 8% glycerol) and 1 µg salmon sperm DNA for 30 min at room temperature. Double-stranded, end-labeled probe (0.2 ng) (GATCCGTAAGACCACGTGGTCGTCAGGATC; bold type indicates the Myc/Max consensus binding site) was added for 30 min at room temperature. Complexes were identified by incubating 0.2 µg of the relevant polyclonal antibody with the reaction mixtures and assaying for a supershift or loss of band. Complexes were separated on 5% acrylamide 0.5x Tris-borate-EDTA-polyacrylamide gels run at 250 V at 4°C and visualized by phosphorimager. Bands were quantitated using ImageQuant software.
ChIP. Chromatin immunoprecipitations (ChIPs) were carried out using an Upstate chromatin immunoprecipitation assay kit according to the manufacturer's instructions. PCR was carried out on resultant DNA samples using the following primer pairs: NME1, TGCTGAGAGGGAAAGGAGACAGG and TCAGCAAGCACCGCCGAAGTACC; NPM1, GGCCTACCTCACACTGTTGGAGG and TGAAGGTTGGCAAGAGTCGTAGAGC; CAD, CACTGGAACCAAGCAAACACC and AAGGTTGCTGCTGTGGAACTACC; HSP60, ACGAGACACTCACCATGACTTACG and TAGAAAGTGCTCCCCTTTTCTTC; RUVBL1, ACCCTTGCTGGCTAATGTGATCCT and AGATTGAGGAGGTGAAGAGCACCA; and p27, CGAAGCTTCAGTGATCAAGTGTACTGG and CCTTAAGAGCGCGCTCGCCAGC. IPs were carried out using polyclonal anti-N-Myc antibody (Santa Cruz) and monoclonal anti-CBP antibody (Lab Vision).
Microarrays.
Human BJ fibroblasts were stably transduced with hygromycin retrovirus constructs encoding LXSH N-MycWT, N-MycMBII, N-MycMBIV, or empty vector. Polyclonal cultures were expanded for the minimum time to reach two 15-cm dishes of subconfluent, log-phase cells. RNA was harvested by using an RNAeasy kit, and the RNA integrity and concentration were verified by gel electrophoresis. An independent culture of log-phase BJ cells was used to prepare a standard control RNA. The experimental and control RNAs were directly converted to cDNA with either Cy5- or Cy3-modified dCTP using an Agilent Technologies fluorescent direct label kit. Each Cy5-labeled experimental cDNA was mixed with a constant amount of control Cy3-labeled cDNA prepared from asynchronously growing, untransfected BJ fibroblasts and hybridized overnight to Agilent Human 1A version 2 oligonucleotide microarrays. Microarrays were washed and processed according to the supplier. Slides were scanned using a Genepix4000B (Axon Instruments). Only genes with a signal >20% above background and with at least 70% good data across the arrays were considered for further analysis. The data were centered to the average of the two vector-only arrays and filtered for average changes in signal for either N-MycWT or mutants that were above the threshold indicated. The microarray data for this study are fully accessible at https://genome.unc.edu.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MBIV is also compared to the well-established loss-of-function deletion mutant N-MycMBII.
|
MBIV is normal for induction of proliferation but defective for apoptosis.
Accelerated rates of cell proliferation and apoptosis are well-characterized responses to Myc expression. The myc null fibroblast cell line HO15.19 was used to investigate the ability of N-MycMBIV to induce proliferation and apoptosis without the complication of endogenous c-Myc (38). N-MycWT and mutants were introduced into myc null fibroblasts by retroviral infection, and pools were selected. Equivalent amounts of N-MycMBIV protein and N-MycWT protein were expressed, demonstrating that this mutation does not affect protein synthesis or turnover rate (Fig. 2A). Like N-MycWT, N-MycMBIV was found to be nuclear by subcellular fractionation, demonstrating that the deletion does not modulate the normal localization of N-Myc (Fig. 2A).
|
MBIV-expressing cells also had the same 23-h doubling time as N-MycWT. N-MycMBII-expressing cells were partially defective for proliferation, having a doubling time of about 30 h, consistent with previous results with c-Myc (10). The proliferation rate was independently measured by CFSE cell staining. At the start of the assay, cells were stained with this inert, fluorescent dye, and FACS analysis was used to measure cell fluorescence on three subsequent days. Proliferation is proportional to the rate of fluorescence loss, because the CFSE is diluted as the cells divide (44, 45). The mean CFSE loss per day was 1.9-fold for the vector control cells (Fig. 2C and D). This value is higher than the dilution expected for a 45-h doubling time, because some dye is also lost by leakage and bleaching. N-MycWT- and N-MycMBII-expressing cells had daily CFSE losses of 2.8-fold and 2.3-fold, respectively, correlating with their doubling times of 23 h and 30 h, respectively (43). N-MycMBIV-expressing cells had a daily CFSE loss of 2.8-fold (Fig. 2C and D), and therefore this mutant induces a proliferation rate indistinguishable from N-MycWT-expressing cells in two independent assays. Loss of CSFE staining was uniform across the cell population, indicating that all cells proliferate at the same rate and that there are no growth-arrested cells in the population.
The ability of N-MycMBIV to induce apoptosis following serum deprivation was also measured. Apoptosis was quantitated by measuring the proportion of cells with sub-G1 DNA content by PI staining and FACS analysis (Fig. 3). After 24 h of serum deprivation, 22% of MycWT cells had a sub-G1 DNA content, whereas vector control and N-MycMBII cells were largely resistant to apoptosis, with values of 3.6% and 6.1%, respectively. Interestingly, the N-MycMBIV cells were also resistant to apoptosis induced by serum deprivation, with only 7.2% cells having sub-G1 DNA content.
|
MBIV is defective for some transformation assays.
Exogenous Myc expression in primed cell systems can induce cell transformation. Since each transformation assay models subtly different aspects of tumorigenesis, we used four different well-established Myc transformation assays to determine the ability of N-MycMBIV to induce cell transformation: the REF-based Ras/Myc focus formation assay, the Rat-1a soft agar assay, the Rat-1a focus formation assay, and the RK3E focus formation assay.
In the REF-based Ras/Myc focus formation assay, rat embryo fibroblasts are transfected with H-rasG12V and myc. Cooperation between these oncogenes results in foci on the cell monolayer (31). We transfected REFs with H-rasG12V and either vector control, N-mycWT, N-mycMBII, or N-mycMBIV. After 10 days, foci were counted and normalized for transfection efficiency (Fig. 4A). N-mycWT and N-mycMBIV induced equivalent numbers of foci, whereas N-mycMBII was defective, as previously reported (43).
|
MBII, N-mycMBIV, or vector control and drug selected stable cell lines. Again, we found that expression levels of these proteins were comparable (not shown). When plated in soft agar, N-mycWT and N-mycMBIV induced equivalent numbers of colonies larger than 50 µm, whereas N-mycMBII was defective (Fig. 4B). There were also no qualitative or kinetic differences between the appearances of the N-mycWT- and N-mycMBIV-induced colonies (Fig. 4C).
Rat-1a cells can also be used for a focus formation assay. At confluence, Rat-1a cells form a tight monolayer, and expression of exogenous Myc allows cells to pile up into foci. We used the same cells as for the soft agar assay and assayed for focus formation when cells expressing N-Myc were plated at low density in a field of nontransformed cells (see Materials and Methods). Surprisingly, we found that N-MycWT-expressing cells formed foci, whereas N-MycMBIV-expressing cells did not (Fig. 4E and F). This was unexpected, since N-mycMBIV was fully active in focus formation in cooperation with H-rasG12V and in soft agar colonies.
In the rat RK3E cell transformation assay, exogenous Myc expression induces E1A-immortalized rat kidney cells to grow in well-defined foci on a monolayer (21). RK3E cells were infected with retroviral vectors expressing either nothing, N-MycWT, or N-MycMBIV. As with the myc null fibroblasts, the expression levels of N-MycWT and N-MycMBIV were comparable (not shown). Parallel cultures of RK3E cells were either left to grow to confluence and maintained for 3 weeks to allow focus formation or selected for the hygromycin resistance gene on the expression vector to assay for infection efficiency. Cells expressing N-MycWT but not the vector control became transformed, piling up into foci. Expression of N-MycMBIV also resulted in focus formation but with only 40% of the efficiency of N-MycWT (Fig. 4D). These data, while not as dramatic as the defect for Rat-1a focus formation, are consistent with a defect for N-MycMBIV in Myc-dependent focus formation in monolayer culture in the absence of an H-rasG12V oncogene.
In summary, N-MycMBIV has activity equivalent to N-MycWT in the REF focus and Rat-1a soft agar assays but is defective in the Rat-1a and RK3E focus assays. The differential activities of N-MycMBIV in the latter assays suggest that this mutant can transform cells but is defective for some activities demanded by more stringent assays.
MycMBIV is hyperactive for G2 arrest.
Myc is well established as driving proliferation in many cell systems; however, in primary cells, expression of Myc can induce a G2 growth arrest (19). To assess the impact of our Myc mutants on G2 arrest, we created estrogen receptor (ER) fusion proteins for each and stably introduced them into human BJ fibroblasts. In estrogen receptor fusions, MycER is inactive until the addition of the ligand OHT (17, 34). After 48 h of tamoxifen addition, the proportion of cells in each phase of the cell cycle was measured by PI staining and FACS analysis (Fig. 5A, two representative experiments). As previously reported, MycWT induces a readily measurable G2 arrest that is dependent on activation of the MycWT-ER protein by tamoxifen in this system. Also consistent with previous reports, the MycMBII mutant fails to induce a G2 arrest. Surprisingly, the MycMBIV mutant was considerably more active at inducing a G2 arrest than MycWT. Comparable stimulation of G2 arrest was observed in three independent experiments.
|
MBIV, and control vectors. Log-phase cells grown in 10% FCS had equivalent FACS profiles (not shown). After 12 h of serum starvation, N-MycWT-expressing cells had a small but reproducible increase in the proportion of cells in G2. Consistent with the observation in human fibroblasts, N-MycMBIV-expressing cells had reproducibly more cells in G2 than MycWT (Fig. 5B).
N-MycMBIV is partially defective for transactivation and repression.
N-MycMBIV is normal for induction of cell proliferation, defective for induction of apoptosis and in some transformation assays, and hyperactive for G2 arrest. We wanted to understand the modulation of these cell phenotypes in terms of Myc target gene activation and repression. Myc is a transcription factor, and previous studies have shown that the expression of Myc activates and represses a large number of genes by both direct and indirect mechanisms (14, 58).
To compare the differential effects of N-MycWT, N-MycMBIV, and N-MycMBII on gene regulation, we extracted RNA from human BJ fibroblasts stably expressing these proteins and analyzed them by hybridization to Agilent Technologies 21K Oligo microarray (representing 15,463 unique unigene clusters). Equivalent Myc protein expression in these cell lines was established by Western blotting (Fig. 6A). Duplicates of each sample were hybridized, and the number of genes with a minimum change (n-fold) above several arbitrarily defined thresholds was calculated for each cell line (Fig. 6B and C). Genes were considered to be Myc responsive if they were activated or repressed by 1.5-fold or more compared to the vector control in the average of two arrays.
|
1,500 genes >1.5-fold and 390 genes >2-fold. Among the activated genes were many previously described direct Myc targets (www.myc-cancer-gene.org), including ornithine decarboxylase, prothymosin, and nucleolin (see Table S1 in the supplemental material). We also found that many ribosomal protein genes (47 different genes) and genes encoding mitochondrial ribosomal proteins (20 different genes) were upregulated (6). N-MycMBIV upregulated significantly fewer genes than MycWT (Fig. 6B), i.e., 1,000 genes >1.5-fold and 160 genes >2-fold.
Repression by Myc proteins paralleled activation, with the cells expressing N-MycWT showing decreased expression of a larger number of genes (1,050 genes >1.5-fold and 260 genes >2-fold) than the cells expressing N-MycMBIV (500 genes >1.5-fold and 100 genes >2-fold). Thus, N-MycMBIV appears to be equally defective for gene activation and repression. N-MycMBIV did not appear to regulate a different set of genes than N-MycWT, because on average 70 to 80% of N-MycMBIV-regulated genes were also regulated by N-MycWT. This degree of overlap within this experimental system strongly indicates that N-MycMBIV regulates mainly a subset of the N-MycWT-regulated genes.
We were interested to know if we could account for the phenotypes of the N-MycMBIV and N-MycMBII mutants by examining the gene families they regulate. Sorting of the activated and repressed genes for N-MycWT and N-MycMBIV into functional categories by GO::TermFinder analysis did not reveal any significant differences between the functions of the regulated genes on a global scale (9). N-MycWT did not regulate additional gene families compared to N-MycMBIV; it simply regulated a larger number of genes in each family (data not shown).
Clustering of the regulated genes for the arrays revealed the profile of gene activation and repression for each N-Myc protein (Fig. 6C). Consistent with the overall number of genes that are activated and repressed, the general profiles are very similar between N-MycWT and N-MycMBIV. However, blocks of genes that were activated or repressed only by N-MycWT and not by N-MycMBIV are evident (see Tables S1 and S2 in the supplemental material). We infer that the genes that differ between N-MycWT and N-MycMBIV are likely candidates to play a role in apoptosis, transformation, and G2 arrest. A list of genes that are differentially regulated by N-MycWT and N-MycMBIV and that also have a potential role in cell growth and differentiation is shown in Table 1. Noteworthy genes include the established Myc target genes that mediate transformation: CDCA7 (JPO1) (48), which is activated by N-MycWT but more weakly by N-MycMBIV, and MTA1 (59), which is activated more strongly by N-MycMBIV. SHMT2, which mediates part of the proliferative activity of Myc (42), is activated by both N-MycWT and N-MycMBIV. Analysis to date has not revealed any one particular gene from this list to be responsible for the phenotypes of N-MycMBIV, and it may be that the coordinate regulation of a combination of these genes is necessary for each phenotype.
|
MBII was very different from that of the other N-Myc proteins, indicating a general defect in both activation and repression of cellular genes. N-MycMBII activated only 167 genes >1.5-fold and 16 genes >2-fold. Repression was similarly defective, with only 212 genes repressed >1.5-fold and 21 genes repressed >2-fold. These values range from 5 to 20% of the number of N-MycWT-regulated genes. The vast majority of genes activated or repressed by N-MycWT and N-MycMBIV were only weakly activated or repressed by N-MycMBII. However, from a careful inspection of the profiles it is important to note that most of the genes that were regulated by N-MycWT were also regulated to a limited degree by N-MycMBII compared to the vector control, except that the extent of activation or repression was severely blunted and no longer passed the threshold of significance criteria. Even genes that continue to pass the threshold cutoff were almost always activated or repressed to a lesser extent than by MycWT (see Table S3 in the supplemental material).
N-MycMBIV is defective for DNA binding in vitro.
The microarray gene activation profile indicated that a differential effect on cellular gene regulation was the basis for the distinct phenotype of the N-MycMBIV mutant. In particular, N-MycMBIV was deficient in both activating and repressing approximately 40% of the genes that were responsive to N-MycWT. These data suggested that N-MycMBIV might have an altered interaction with chromosomal binding sites and/or nuclear cofactors. To explore these possibilities, we assessed the ability of this mutant to bind to DNA and to recruit nuclear factors.
The most direct mechanism by which Myc is known to transactivate gene expression is by binding to the TRRAP complex, thereby recruiting histone acetyltransferase activity to promoters (39, 40). We first investigated whether N-MycMBIV could bind to TRRAP by coimmunoprecipitation. As previously published, TRRAP can be coimmunoprecipitated with N-MycWT but not N-MycMBII using an anti-Myc antibody (Fig. 7A). There was no detectable difference in TRRAP binding between N-MycMBIV and N-MycWT, consistent with previous mapping of the TRRAP binding domain to the MBI and MBII regions, which are identical in these proteins.
|
MBIV is not part of the minimal functional DNA binding domain, it maps closest to this domain along the linear sequence. We therefore assessed whether Max binding and/or DNA binding was compromised in the N-MycMBIV mutant. We assayed for Myc-Max binding by coimmunoprecipitation. N-MycWT, N-MycMBII, and N-MycMBIV were expressed at comparable levels (Fig. 7B, upper panel), and immunoprecipitation of N-Myc led to coprecipitation of comparable levels of Max protein (Fig. 7B, lower panel). We next analyzed the DNA binding affinity of N-Myc/Max complexes by EMSA using the Myc binding site GACCACGTGGAC (bold indicates the Myc/Max consensus binding site) as a probe (Fig. 7C). N-MycWT and N-MycMBII formed Myc/Max complexes on GACCACGTGGAC with similar intensities, whereas the N-MycMBIV/Max band was reproducibly much weaker. Incubation of the EMSA reactions with anti-FLAG antibody supershifted the Myc/Max complexes, confirming their identity. Quantitation of three independent experiments revealed that N-MycMBIV DNA binding was reduced to only 20% of the affinity of N-MycWT (Fig. 7D). This reduced DNA binding activity is most likely the cause of the defect in gene activation and repression by the N-MycMBIV mutant.
N-MycMBIV displays defects in DNA binding to some Myc targets.
To establish that N-MycMBIV is defective for DNA binding in vivo as well as in vitro, we measured N-MycWT and N-MycMBIV binding to the promoters of a group of Myc-regulated genes by using ChIP (8). We measured Myc binding to two genes equivalently upregulated by N-MycWT and N-MycMBIV (NME1 and NPM1) and three genes with reduced activation by N-MycMBIV (CAD, HSP60, and RUVBL1) (Fig. 8A). We also measured Myc binding to Myc-repressed gene p27.
|
MBIV, or vector control. We confirmed that this antibody was able to immunoprecipitate both N-Myc proteins equivalently (data not shown). The resultant coprecipitated DNA or DNA from 1% of the IP input material was used as a template for PCR using primers directed to the promoter E-boxes of the Myc-upregulated genes (Fig. 8B). The promoter of a Myc-repressed gene (p27) was also included, because although repressed genes do not have Myc consensus binding sites, some have been found to weakly bind to Myc (37). For the genes expressed equivalently in response to both N-MycWT and N-MycMBIV, NME1 and NPM1, Myc binding was found to be equivalent. For the genes with reduced expression in response to MycMBIV (CAD, HSP60, and RUVBL1), N-MycMBIV bound more weakly than N-MycWT, despite both Myc proteins being expressed at similar levels (Fig. 2A). N-MycMBIV also bound more weakly than N-MycWT to the p27 promoter. Quantitation of the PCRs, followed by normalization against input material, showed that the CAD promoter had the most differential Myc binding, with N-MycWT binding >5-fold over background, compared to N-MycMBIV, which bound only 2-fold over background (Fig. 8B, lower panel). We infer that subtle differences in the individual promoters govern the relative binding of N-MycWT and N-MycMBIV.
CBP is a c-Myc cofactor which binds to the c-Myc C terminus and is recruited to c-Myc target genes (55). Since N-MycMBIV is defective for DNA binding and the deletion maps near where CBP was reported to bind to c-Myc, we wanted to determine whether CBP recruitment to target genes is reduced by this mutant. ChIP was performed using the same cell lines as above, and complexes were immunoprecipitated using an anti-CBP antibody. When PCR was performed on the CBP-coprecipitated promoters, N-MycWT and N-MycMBIV were found to stimulate recruitment of CBP to Myc-regulated gene promoters by equivalent amounts (see Fig. S1 in the supplemental material). Therefore, N-MycMBIV is not defective for gene activation because it is defective for CBP recruitment.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The MBIV domain is necessary for maximal DNA binding.
Since the structure has not been solved for any of the full-length Myc proteins, little can be said about the secondary structure or the position of the MBIV domain in the tertiary structure. Following the observation that c-Myc was a DNA binding protein, the minimal DNA binding domain was deduced from comparison with other bHLH proteins (4, 5). The crystal structure of the Myc/Max DNA binding domains in association with a consensus DNA sequence is known, but the MBIV region was not included (41). Here we demonstrate that although MBIV lies outside this minimal DNA binding domain, it is necessary for full Myc DNA binding activity within the intact protein. The only known protein with a direct role in Myc DNA binding activity is Max, but we find that Max dimerization is unaffected by the MBIV deletion. Since MBIV lies just upstream of the bHLH domain in the primary amino acid sequence, we suggest that this segment may lie close to the DNA binding domain in the tertiary structure. Deletion of MBIV may alter the structure of the DNA binding domain, or it could change the affinity of Myc for a nuclear cofactor that coordinates with Myc to regulate DNA binding. This putative altered structure or cofactor must exist in the nuclear lysates used for EMSA experiments, since these show reduced DNA binding activity for the mutant (Fig. 7). A nonspecific DNA binding sequence was described for c-Myc, but this sequence is not well conserved in N-Myc and does not map to the conserved MBIV domain (15). Furthermore, the N-MycMBIV mutant is localized to the nucleus with the same efficiency as N-MycWT (Fig. 2A). Consistent with a weakened affinity for DNA, N-MycMBIV was found to activate and repress significantly fewer genes than N-MycWT. Curiously, despite this markedly weakened DNA binding ability, N-MycMBIV retains wild-type activity for cooperation with H-RasG12V and rescue of the proliferation defect in myc null cells. Therefore, a high affinity for DNA is not well correlated with Myc biological activity. Further investigation will be required to determine if there is any distinguishing feature of the promoters of the genes that are not activated or repressed by N-MycMBIV.
MBIV is required for apoptosis and some transformation assays.
Microarray analysis showed that N-MycMBIV upregulated and repressed about 60% as many genes as N-MycWT. Despite the large aberration in gene expression, cell proliferation was found to be unaffected by this mutation. Either Myc may induce proliferation by redundant pathways or the major proliferative genes may fall mainly within the gene set that is regulated equivalently between N-MycWT and N-MycMBIV.
Some transformation assays and apoptosis are defective in cells expressing N-MycMBIV and therefore appear to require a more robust Myc target gene response. Either the full complement of Myc target genes required to trigger these pathways is lacking or the genes governing these pathways are not being activated or repressed efficiently enough by N-MycMBIV. The N-Myc MBIV domain contrasts with a recently characterized nearby domain, MBIII (28). A Myc MBIII deletion increases apoptosis but decreases transformation. We are currently testing individual genes which are upregulated or downregulated by N-MycWT but not N-MycMBIV for the ability to contribute to apoptosis or transformation. However, the coordinated expression of a number of genes may be required to drive these phenotypes.
MycMBIV is not a general Myc loss-of-function mutant.
Unlike MycMBII, which has been demonstrated here and elsewhere to have a loss of function for almost all Myc phenotypes (18, 19, 23, 29, 53), MycMBIV exhibits a gain of function for inducing G2 arrest in primary fibroblasts. No genes appear to be activated or repressed by MycMBIV above the levels seen for a MycWT, and therefore the superinduction of a G2 arrest does not appear to be due to simple hyperactivation or repression of a Myc target gene(s). This suggests that the most likely explanation for the exaggerated G2 arrest phenotype is that Myc regulates genes which both drive and oppose G2 arrest. In the case of MycMBIV, the genes which oppose G2 arrest may be muted, leaving the genes that drive G2 arrest to dominate.
Myc can induce a complex set of phenotypes in many different cell backgrounds. It remains unclear if the distinct phenotypes are a response to different target genes or a cell context-dependent response to a common target gene set. In the present study, we show that mutation of an evolutionarily conserved Myc domain results in the loss of a subset of target genes which correlates with enhancement and inhibition of different Myc phenotypes. This is in contrast with previously described Myc mutations, which either have no effect on or inhibit Myc phenotypes. The fact that this mutant can reduce Myc-induced apoptosis and transformation but enhance Myc-induced G2 arrest argues that the distinct target genes are involved in different phenotypes. Further analysis may allow the identification of the specific gene or genes involved in transformation or apoptosis.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by a grant from the National Cancer Institute to M.D.C. M.L.W. was supported by Howard Hughes Medical Institute Biomedical Research Support Award 76200-560801 to Dartmouth College, American Cancer Society Institutional Award IRG-82-003-17 to the Norris Cotton Cancer Center, and a grant from the V Foundation for Cancer Research. M.L.W. is a V Scholar of the V Foundation for Cancer Research.
|
FOOTNOTES |
|---|
Supplemental material for this article may be found at http://mcb.asm.org/.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Askew, D. S., R. A. Ashmun, B. C. Simmons, and J. L. Cleveland. 1991. Constitutive c-myc expression in an IL-3-dependent myeloid cell line suppresses cell cycle arrest and accelerates apoptosis. Oncogene 6:1915-1922.[Medline]
3. Bhatia, K., K. Huppi, G. Spangler, D. Siwarski, R. Iyer, and I. Magrath. 1993. Point mutations in the c-Myc transactivation domain are common in Burkitt's lymphoma and mouse plasmacytomas. Nat. Genet. 5:56-61.[Medline]
4. Blackwell, T. K., L. Kretzner, E. M. Blackwood, R. N. Eisenman, and H. Weintraub. 1990. Sequence-specific DNA binding by the c-Myc protein. Science 250:1149-1151.
5. Blackwood, E. M., and R. N. Eisenman. 1991. Max: a helix-loop-helix zipper protein that forms a sequence-specific DNA-binding complex with Myc. Science 251:1211-1217.
6. Boon, K., H. N. Caron, R. van Asperen, L. Valentijn, M. C. Hermus, P. van Sluis, I. Roobeek, I. Weis, P. A. Voute, M. Schwab, and R. Versteeg. 2001. N-myc enhances the expression of a large set of genes functioning in ribosome biogenesis and protein synthesis. EMBO J. 20:1383-1393.[CrossRef][Medline]
7. Bouchard, C., O. Dittrich, A. Kiemaier, K. Dohmann, A. Menkel, M. Eilers, and B. Luscher. 2001. Regulation of cyclin D2 gene expression by the Myc/Max/Mad network: Myc-dependent TRRAP recruitment and histone acetylation at the cyclin D2 promoter. Genes Dev. 15:2042-2047.
8. Boyd, K. E., and P. J. Farnham. 1997. Myc versus USF: discrimination at the cad gene is determined by core promoter elements. Mol. Cell. Biol. 17:2529-2537.[Abstract]
9. Boyle, E. I., S. Weng, J. Gollub, H. Jin, D. Botstein, J. M. Cherry, and G. Sherlock. 2004. GO::TermFinder—open source software for accessing gene ontology information and finding significantly enriched gene ontology terms associated with a list of genes. Bioinformatics 20:3710-3715.
10. Bush, A., M. Mateyak, K. Dugan, A. Obaya, S. Adachi, J. Sedivy, and M. Cole. 1998. c-myc null cells misregulate cad and gadd45 but not other proposed c-Myc targets. Genes Dev. 12:3797-3802.
11. Cole, M. D. 1986. The myc oncogene: its role in transformation and differentiation. Annu. Rev. Genet. 20:361-385.[CrossRef][Medline]
12. Coller, H. A., C. Grandori, P. Tamayo, T. Colbert, E. S. Lander, R. N. Eisenman, and T. R. Golub. 2000. Expression analysis with oligonucleotide microarrays reveals that MYC regulates genes involved in growth, cell cycle, signaling, and adhesion. Proc. Natl. Acad. Sci. USA 97:3260-3265.
13. Conzen, S. D., K. Gottlob, E. S. Kandel, P. Khanduri, A. J. Wagner, M. O'Leary, and N. Hay. 2000. Induction of cell cycle progression and acceleration of apoptosis are two separable functions of c-Myc: transrepression correlates with acceleration of apoptosis. Mol. Cell. Biol. 20:6008-6018.
14. Dang, C. V. 1999. c-Myc target genes involved in cell growth, apoptosis, and metabolism. Mol. Cell. Biol. 19:1-11.[Medline]
15. Dang, C. V., H. van Dam, M. Buckmire, and W. M. F. Lee. 1989. DNA-binding domain of human c-Myc produced in Escherichia coli. Mol. Cell. Biol. 9:2477-2486.
16. Eberhardy, S. R., and P. J. Farnham. 2002. Myc recruits P-TEFb to mediate the final step in the transcriptional activation of the cad promoter. J. Biol. Chem. 277:40156-40162.
17. Eilers, M., D. Picard, K. R. Yamamoto, and J. M. Bishop. 1989. Chimaeras of myc oncoprotein and steroid receptors cause hormone-dependent transformation of cells. Nature 340:66-68.[CrossRef][Medline]
18. Evan, G. I., A. H. Wyllie, C. S. Gilbert, T. D. Littlewood, H. Land, M. Brooks, C. M. Waters, L. Z. Penn, and D. C. Hancock. 1992. Induction of apoptosis in fibroblasts by c-myc protein. Cell 69:119-128.[CrossRef][Medline]
19. Felsher, D. W., A. Zetterberg, J. Zhu, T. Tlsty, and J. M. Bishop. 2000. Overexpression of MYC causes p53-dependent G2 arrest of normal fibroblasts. Proc. Natl. Acad. Sci. USA 97:10544-10548.
20. Fernandez, P. C., S. R. Frank, L. Wang, M. Schroeder, S. Liu, J. Greene, A. Cocito, and B. Amati. 2003. Genomic targets of the human c-Myc protein. Genes Dev. 17:1115-1129.
21. Foster, K. W., S. Ren, I. D. Louro, S. M. Lobo-Ruppert, P. McKie-Bell, W. Grizzle, M. R. Hayes, T. R. Broker, L. T. Chow, and J. M. Ruppert. 1999. Oncogene expression cloning by retroviral transduction of adenovirus E1A-immortalized rat kidney RK3E cells: transformation of a host with epithelial features by c-MYC and the zinc finger protein GKLF. Cell Growth Differ. 10:423-434.
22. Frank, S. R., M. Schroeder, P. Fernandez, S. Taubert, and B. Amati. 2001. Binding of c-Myc to chromatin mediates mitogen-induced acetylation of histone H4 and gene activation. Genes Dev. 15:2069-2082.
23. Freytag, S. O., C. V. Dang, and W. M. F. Lee. 1990. Definition of the activities and properties of c-myc required to inhibit cell differentiation. Cell Growth Differ. 1:339-343.[Abstract]
24. Frye, M., C. Gardner, E. R. Li, I. Arnold, and F. M. Watt. 2003. Evidence that Myc activation depletes the epidermal stem cell compartment by modulating adhesive interactions with the local microenvironment. Development 130:2793-2808.
25. Gallant, P., Y. Shiio, P. F. Cheng, S. M. Parkhurst, and R. N. Eisenman. 1996. Myc and Max homologs in Drosophila. Science 274:1523-1527.
26. Gregory, M. A., and S. R. Hann. 2000. c-Myc proteolysis by the ubiquitin-proteasome pathway: stabilization of c-Myc in Burkitt's lymphoma cells. Mol. Cell. Biol. 20:2423-2435.
27. Guo, Q. M., R. L. Malek, S. Kim, C. Chiao, M. He, M. Ruffy, K. Sanka, N. H. Lee, C. V. Dang, and E. T. Liu. 2000. Identification of c-myc responsive genes using rat cDNA microarray. Cancer Res. 60:5922-5928.
28. Herbst, A., M. T. Hemann, K. A. Tworkowski, S. E. Salghetti, S. W. Lowe, and W. P. Tansey. 2005. A conserved element in Myc that negatively regulates its proapoptotic activity. EMBO Rep. 6:177-183.[CrossRef][Medline]
29. Kenney, A. M., M. D. Cole, and D. H. Rowitch. 2003. Nmyc upregulation by sonic hedgehog signaling promotes proliferation in developing cerebellar granule neuron precursors. Development 130:15-28.
30. Kim, S. Y., A. Herbst, K. A. Tworkowski, S. E. Salghetti, and W. P. Tansey. 2003. Skp2 regulates Myc protein stability and activity. Mol. Cell 11:1177-1188.[CrossRef][Medline]
31. Land, H., L. F. Parada, and R. A. Weinberg. 1983. Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature 304:596-602.[CrossRef][Medline]
32. Landay, M., S. K. Oster, F. Khosravi, L. E. Grove, X. Yin, J. Sedivy, L. Z. Penn, and E. V. Prochownik. 2000. Promotion of growth and apoptosis in c-myc nullizygous fibroblasts by other members of the myc oncoprotein family. Cell Death Differ. 7:697-705.[CrossRef][Medline]
33. Li, Z., S. Van Calcar, C. Qu, W. K. Cavenee, M. Q. Zhang, and B. Ren. 2003. A global transcriptional regulatory role for c-Myc in Burkitt's lymphoma cells. Proc. Natl. Acad. Sci. USA 100:8164-8169.
34. Littlewood, T. D., D. C. Hancock, P. S. Danielian, M. G. Parker, and G. I. Evan. 1995. A modified oestrogen receptor ligand-binding domain as an improved switch for the regulation of heterologous proteins. Nucleic Acids Res. 23:1686-1690.
35. Lutterbach, B., and S. R. Hann. 1994. Hierarchical phosphorylation at N-terminal transformation-sensitive sites in c-Myc protein is regulated by mitogens and in mitosis. Mol. Cell. Biol. 14:5510-5522.
36. Malynn, B. A., I. M. de Alboran, R. C. O'Hagan, R. Bronson, L. Davidson, R. A. DePinho, and F. W. Alt. 2000. N-myc can functionally replace c-myc in murine development, cellular growth, and differentiation. Genes Dev. 14:1390-1399.
37. Mao, D. Y., J. D. Watson, P. S. Yan, D. Barsyte-Lovejoy, F. Khosravi, W. W. Wong, P. J. Farnham, T. H. Huang, and L. Z. Penn. 2003. Analysis of Myc bound loci identified by CpG island arrays shows that Max is essential for Myc-dependent repression. Curr. Biol. 13:882-886.[CrossRef][Medline]
38. Mateyak, M. K., A. J. Obaya, S. Adachi, and J. M. Sedivy. 1997. Phenotypes of c-Myc-deficient rat fibroblasts isolated by targeted homologous recombination. Cell Growth Differ. 8:1039-1048.[Abstract]
39. McMahon, S. B., H. A. Van Buskirk, K. A. Dugan, T. D. Copeland, and M. D. Cole. 1998. The novel ATM-related protein TRRAP is an essential cofactor for the c-Myc and E2F oncoproteins. Cell 94:363-374.[CrossRef][Medline]
40. McMahon, S. B., M. A. Wood, and M. D. Cole. 2000. The essential cofactor TRRAP recruits the histone acetyltransferase hGCN5 to c-Myc. Mol. Cell. Biol. 20:556-562.
41. Nair, S. K., and S. K. Burley. 2003. X-ray structures of Myc-Max and Mad-Max recognizing DNA. Molecular bases of regulation by proto-oncogenic transcription factors. Cell 112:193-205.[CrossRef][Medline]
42. Nikiforov, M. A., S. Chandriani, B. O'Connell, O. Petrenko, I. Kotenko, A. Beavis, J. M. Sedivy, and M. D. Cole. 2002. A functional screen for Myc-responsive genes reveals serine hydroxymethyltransferase, a major source of the one-carbon unit for cell metabolism. Mol. Cell. Biol. 22:5793-5800.
43. Nikiforov, M. A., S. Chandriani, J. Park, I. Kotenko, D. Matheos, A. Johnsson, S. B. McMahon, and M. D. Cole. 2002. TRRAP-dependent and TRRAP-independent transcriptional activation by Myc family oncoproteins. Mol. Cell. Biol. 22:5054-5063.
44. Nikiforov, M. A., I. Kotenko, O. Petrenko, A. Beavis, L. Valenick, I. Lemischka, and M. D. Cole. 2000. Complementation of Myc-dependent cell proliferation by cDNA expression library screening. Oncogene 19:4828-4831.[CrossRef][Medline]
45. Obaya, A. J., I. Kotenko, M. D. Cole, and J. M. Sedivy. 2002. The proto-oncogene c-myc acts through the cyclin-dependent kinase (Cdk) inhibitor p27(Kip1) to facilitate the activation of Cdk4/6 and early G(1) phase progression. J. Biol. Chem. 277:31263-31269.
46. O'Connell, B. C., A. F. Cheung, C. P. Simkevich, W. Tam, X. Ren, M. K. Mateyak, and J. M. Sedivy. 2003. A large scale genetic analysis of c-Myc-regulated gene expression patterns. J. Biol. Chem. 278:12563-12573.
47. Oster, S. K., D. Y. Mao, J. Kennedy, and L. Z. Penn. 2003. Functional analysis of the N-terminal domain of the Myc oncoprotein. Oncogene 22:1998-2010.[CrossRef][Medline]
48. Prescott, J. E., R. C. Osthus, L. A. Lee, B. C. Lewis, H. Shim, J. F. Barrett, Q. Guo, A. L. Hawkins, C. A. Griffin, and C. V. Dang. 2001. A novel c-Myc-responsive gene, JPO1, participates in neoplastic transformation. J. Biol. Chem. 276:48276-48284.
49. Schreiber-Agus, N., D. Stein, K. Chen, J. S. Goltz, L. Stevens, and R. A. DePinho. 1997. Drosophila Myc is oncogenic in mammalian cells and plays a role in the diminutive phenotype. Proc. Natl. Acad. Sci. USA 94:1235-1240.
50. Schuhmacher, M., F. Kohlhuber, M. Holzel, C. Kaiser, H. Burtscher, M. Jarsch, G. W. Bornkamm, G. Laux, A. Polack, U. H. Weidle, and D. Eick. 2001. The transcriptional program of a human B cell line in response to Myc. Nucleic Acids Res. 29:397-406.
51. Sears, R., F. Nuckolls, E. Haura, Y. Taya, K. Tamai, and J. R. Nevins. 2000. Multiple Ras-dependent phosphorylation pathways regulate Myc protein stability. Genes Dev. 14:2501-2514.
52. Sommer, A., K. Bousset, E. Kremmer, M. Austen, and B. Luscher. 1998. Identification and characterization of specific DNA-binding complexes containing members of the Myc/Max/Mad network of transcriptional regulators. J. Biol. Chem. 273:6632-6642.
53. Stone, J., T. de Lange, G. Ramsay, E. Jakobovits, J. M. Bishop, H. Varmus, and W. Lee. 1987. Definition of regions in human c-myc that are involved in transformation and nuclear localization. Mol. Cell. Biol. 7:1697-1709.
54. Trumpp, A., Y. Refaeli, T. Oskarsson, S. Gasser, M. Murphy, G. R. Martin, and J. M. Bishop. 2001. c-Myc regulates mammalian body size by controlling cell number but not cell size. Nature 414:768-773.[CrossRef]
