Search

We have developed a statistical regression modeling approach to discover genes that are differentially expressed between two predefined sample groups in DNA microarray experiments. Our model is based on well-defined assumptions, uses rigorous and well-characterized statistical measures, and accounts for the heterogeneity and genomic complexity of the data. In contrast to cluster analysis, which attempts to define groups of genes and/or samples that share common overall expression profiles, our modeling approach uses known sample group membership to focus on expression profiles of individual genes in a sensitive and robust manner. Further, this approach can be used to test statistical hypotheses about gene expression. To demonstrate this methodology, we compared the expression profiles of 11 acute myeloid leukemia (AML) and 27 acute lymphoblastic leukemia (ALL) samples from a previous study (Golub et al. 1999) and found 141 genes differentially expressed between AML and ALL with a 1% significance at the genomic level. Using this modeling approach to compare different sample groups within the AML samples, we identified a group of genes whose expression profiles correlated with that of thrombopoietin and found that genes whose expression associated with AML treatment outcome lie in recurrent chromosomal locations. Our results are compared with those obtained using t-tests or Wilcoxon rank sum statistics.

Basic helix-loop-helix (bHLH) transcription factors are known to function during mammalian neurogenesis. Here we show that transient transfection of vectors expressing neuroD2, MASH1, ngn1 or related neural bHLH proteins, with their putative dimerization partner E12, can convert mouse P19 embryonal carcinoma cells into differentiated neurons. Transfected cells express numerous neuron-specific proteins, adopt a neuronal morphology and are electrically excitable. Thus, the expression of neural bHLH proteins is sufficient to confer a neuronal fate on uncommitted mammalian cells. Neuronal differentiation of transfected cells is preceded by elevated expression of the cyclin-dependent kinase inhibitor p27(Kip1) and cell cycle withdrawal. This demonstrates that the bHLH proteins can link neuronal differentiation to withdrawal from the cell cycle, possibly by activating the expression of p27(Kip1). The ability to generate mammalian neurons by transient expression of neural bHLH proteins should create new opportunities for studying neurogenesis and devising neural repair strategies.

The myogenic progenitors of epaxial (paraspinal and intercostal) and hypaxial (limb and abdominal wall) musculature are believed to originate in dorsal-medial and ventral-lateral domains, respectively, of the developing somite. To investigate the hypothesis that Myf-5 and MyoD have different roles in the development of epaxial and hypaxial musculature, we further characterized myogenesis in Myf-5- and MyoD-deficient embryos by several approaches. We examined expression of a MyoD-lacZ transgene in Myf-5 and MyoD mutant embryos to characterize the temporal-spatial patterns of myogenesis in mutant embryos. In addition, we performed immunohistochemistry on sectioned Myf-5 and MyoD mutant embryos with antibodies reactive with desmin, nestin, myosin heavy chain, sarcomeric actin, Myf-5, MyoD and myogenin. While MyoD(-/-) embryos displayed normal development of paraspinal and intercostal muscles in the body proper, muscle development in limb buds and brachial arches was delayed by about 2.5 days. By contrast, Myf-5(-/-) embryos displayed normal muscle development in limb buds and brachial arches, and markedly delayed development of paraspinal and intercostal muscles. Although MyoD mutant embryos exhibited delayed development of limb musculature, normal migration of Pax-3-expressing cells into the limb buds and normal subsequent induction of Myf-5 in myogenic precursors was observed. These results suggest that Myf-5 expression in the limb is insufficient for the normal progression of myogenic development. Taken together, these observations strongly support the hypothesis that Myf-5 and MyoD play unique roles in the development of epaxial and hypaxial muscle, respectively.

The differentiation of muscle cells is controlled by the MyoD family of transcription factors. This family is regulated by extracellular growth factors that transmit largely unknown signals into the cells. Here we show that the activity of extracellular signal-regulated protein kinase (ERK), a kinase that is part of the mitogen-activated protein kinase (MAPK) cascade, is low in myoblasts and is induced with the onset of terminal differentiation of C2 cells. ERK activity is also induced in fibroblasts that were modified to express MyoD, but not in the parental fibroblast cells. Thus, ERK induction is an intrinsic property of muscle cells. A specific MAPK kinase inhibitor (PD098059) that was added to C2 cells partially inhibited the fusion of myoblasts to multinucleated myotubes without affecting the expression of muscle-specific markers. This inhibitor blocked the induction of MyoD expression that normally takes place during terminal differentiation. Two lines of evidence suggest that the MAPK cascade induces the activity of MyoD: 1) the expression of constitutively activated forms of MEK1 or Raf1 enhanced the transcriptional activity of MyoD in 10T1/2 fibroblasts; and 2) the addition of PD098059 to fibroblast cells expressing a conditional MyoD-estrogen fusion protein significantly inhibited the expression of MyoD-responsive genes. Our results indicate that the MAPK pathway is activated in differentiating muscle cells and that it positively regulates the expression and activity of MyoD protein.

Each of the myogenic helix-loop-helix transcription factors (MyoD, Myogenin, Myf-5, and MRF4) is capable of activating muscle-specific gene expression, yet distinct functions have not been ascribed to the individual proteins. We report here that MyoD and Myogenin mRNAs selectively accumulate in hindlimb muscles of the adult rat that differ in contractile properties: MyoD is prevalent in fast twitch and Myogenin in slow twitch muscles. The distribution of MyoD and Myogenin transcripts also differ within a single muscle and correlate with the proportions of fast glycolytic and slow oxidative muscle fibres, respectively. Furthermore, the expression of a transgene consisting of a muscle-specific cis-regulatory region from the myoD gene controlling lacZ was primarily associated with the fast glycolytic fibres. Alteration of the fast/slow fibre type distribution by thyroid hormone treatment or by cross-reinnervation resulted in a corresponding alteration in the MyoD/Myogenin mRNA expression pattern. These findings show that the expression of specific myogenic helix-loop-helix regulators is under the control of innervation and humoral factors and may mediate differential control of contractile protein gene expression in adult muscle.

MyoD is a master regulatory gene for myogenesis. Under the control of a retroviral long terminal repeat, MyoD was expressed in a variety of differentiated cell types by using either a DNA transfection vector or a retrovirus. Expression of muscle-specific proteins was observed in chicken, human, and rat primary fibroblasts and in differentiated melanoma, neuroblastoma, liver, and adipocyte lines. The ability of MyoD to activate muscle genes in a variety of differentiated cell lines suggests that no additional tissue-specific factors other than MyoD are needed to activate the downstream program for terminal muscle differentiation or that, if such factors exist, they are themselves activated by MyoD expression.