Spinal muscular atrophy (SMA) is a common, often fatal, autosomal recessive disease leading to progressive muscle wasting and paralysis as a result of degeneration of anterior horn cells of the spinal cord. A gene termed survival of motor neurons (SMN), at 5q13, has been identified as the determining gene of SMA (Lefebvre et al., 1995). The SMN gene is deleted in > 98% of SMA patients, but the function of the SMN protein is unknown. In searching for hnRNP‐interacting proteins we found that SMN interacts with the RGG box region of hnRNP U, with itself, with fibrillarin and with several novel proteins. We have produced monoclonal antibodies to the SMN protein, and we report here on its striking cellular localization pattern. Immunolocalization studies using SMN monoclonal antibodies show several intense dots in HeLa cell nuclei. These structures are similar in number (2–6) and size (0.1–1.0 micron) to coiled bodies, and frequently are found near or associated with coiled bodies. We term these prominent nuclear structures gems, for Gemini of coiled bodies.

We have investigated the use of single-stranded oligodeoxynucleotides (ssODN) to produce specific single-base alterations in episomal and chromosomal DNA in mouse embryonic stem (ES) cells. Two different reporter genes, EGFP and LacZ, each with a single point mutation that inactivates reporter activity, were used. ssODN homologous to the target sequence, except for a single mismatch at the mutant base, were used to correct the mutant reporter genes. When tested in CHO-K1 cells, the ssODN showed correction rates of 0.5-1.0%, consistent with prior reports. ssODN in the antisense orientation provided higher rates of gene conversion than those in the sense orientation for both reporter genes. Nuclear extracts from mouse ES cells exhibited nearly the same correction activity as extracts from CHO-K1 cells. ssODN corrected the mutant bases of both episomal and chromosomal mutant reporter genes in mouse ES cells. Although the efficiency of gene correction observed in ES cells is low, approximately 10 À4, these results demonstrate that ssODN can produce single-base alterations in the genomic DNA of mouse ES cells. As conversion efficiency is improved by the continued development of oligonucleotide structure and DNA delivery methods, ssODN could be used to produce ES cells with specific mutations in any gene in a single step. The targeted ES cells could in turn be used to create accurate mouse models of inherited diseases.

The centrosome, an organelle that functions as the major microtubule-organizing center, plays an essential role in the formation of the mitotic spindle and guiding accurate chromosome segregation. Centrosome aberrations are frequently associated with various forms of human cancers and it is thought that defects in this organelle contribute to genomic instability and malignant transformation. We recently identified and characterized a centrosome-localized protein complex that is comprised of Su48 and Nde1. Disruption of the normal function of these proteins leads to abnormal cell division. To extend our understanding of how this protein complex operates, we sought to identify Nde1-interacting molecules by the yeast two-hybrid screening method. Here, we demonstrate that both Nde1 and Su48 can associate with p78/MCRS1, a protein implicated in cancer development. We found that, whereas the majority of p78 localizes to the nucleus as reported in earlier studies, a fraction of the p78 protein can be detected in the centrosome. Moreover, we determined that a region containing the forkhead-associated domain of p78 is involved in association with Nde1 and Su48, as well as in centrosomal localization. We also provide evidence that the association between p78 and Nde1 is regulated by phosphorylation on Nde1. Furthermore, abrogation of the endogenous p78 function by small interfering RNA knockdown causes cell death and a modest delay in mitosis. These results indicate that a subset of the p78 proteins comprises a component of the centrosome and that p78 is essential for cell viability.