P0, a major structural protein of peripheral myelin, is a homophilic adhesion molecule and maps to chromosome 1q22-q23, in the region of the locus for Charcot-Marie-Tooth neuropathy type 1B (CMT1B). We have investigated P0 as a candidate gene in two pedigrees with CMT1B and found point mutations which are completely linked with the disease (Z = 5.5, theta = 0). The mutations, glutamate substitution for lysine 96 or aspartate 90, are located in the extracellular domain, which plays a significant role in myelin membrane adhesion. Individuals with CMT1B are heterozygous for the normal allele and the mutant allele. Our results indicate that P0 is a gene responsible for CMT1B.

Tumor cells arising from a particular tissue may exhibit the same gene expression patterns as their precursor cells. To test this proposition, we have analyzed the expression of a neural RNA-binding protein, Musashi1, in primary human central nervous system (CNS) tumors. In rodents, Musashi1 is expressed predominantly in proliferating multipotent neural precursor cells, but not in newly generated postmitotic neurons. The expression of Musashi1 is downregulated with the successive progression of neurogenesis. In normal adult human tissues, we detected low levels of Musashi1 expression in brain and testis by RT-PCR analysis. In an RNA panel of 32 cancer tissues and cell lines, elevated expression of Musashi1 was seen in all five malignant gliomas studied, in contrast to the slight expression seen in other tumor cells, including those in several melanomas and a prostate cancer. Western blot analysis showed strong Musashi1 expression in malignant gliomas compared with nonneoplastic brain tissue. Glioblastomas, the most malignant form of glioma, showed higher Musashi1 expression than less malignant gliomas by immunohistochemical analysis. Tumors with strong Musashi1 expression tended to have high proliferative activity. Thus, the expression of Musashi1 correlated with the grade of the malignancy and proliferative activity in gliomas. These results suggest that primary CNS tumors may share gene expression patterns with primitive, undifferentiated CNS cells and that Musashi1 may be a useful marker for the diagnosis of CNS tumors.

Islet-2 is a LIM/homeodomain-type transcription factor of the Islet-1 family expressed in embryonic zebrafish. Two Islet-2 molecules bind to the LIM domain binding protein (Ldb) dimers. Overexpression of the LIM domains of Islet-2 or the LIM-interacting domain of Ldb proteins prevented binding of Islet-2 to Ldb proteins in vitro and caused similar in vivo defects in positioning, peripheral axonal outgrowth, and neurotransmitter expression by the Islet-2-positive primary sensory and motor neurons as the defects induced by injection of Islet-2-specific antisense morpholino oligonucleotide. These and other experiments, i.e., mosaic analysis, coexpression of full-length Islet-2, and overexpression of the chimeric LIM domains derived from two different Islet-1 family members, demonstrated that Islet-2 regulates neuronal differentiation by forming a complex with Ldb dimers and possibly with some other Islet-2-specific cofactors.

The telencephalon shows the greatest degree of size variation in the vertebrate brain. Understanding the genetic cascade that regulates telencephalon growth is crucial to our understanding of how evolution of the normal human brain has supported such a variation in size. Here, we present a simple and quick approach to analyze this cascade that combines caged-mRNA technology and the use of antisense morpholino oligonucleotides in zebrafish embryos. Lhx2, a LIM-homeodomain protein, and Six3s (Six3b and Six3a), another homeodomain proteins, show very similar expression patterns early in forebrain development, and these are known to be involved in the growth of this part of the brain. The telencephalon of six3b and six3a double morphant (six3 morphant) embryos is markedly reduced in size due to impaired cellular proliferation. Head-specific overexpression of Lhx2 by photoactivation of a caged-lhx2 mRNA completely rescued this size reduction, whereas similar head-specific activation of Six3b could not rescue the knockdown effect of lhx2. In the forebrain of medaka embryos, Six3 facilitates cellular proliferation by sequestration of Geminin from Cdt1, a key component in the assembly of the prereplication complex. Our results suggest that Lhx2 may mediate an alternative or parallel pathway for control of cellular proliferation in the developing forebrain via Six3.

No abstract