Human Artificial Chromosomes for Cancer Research and Functional Genomics

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

The organization and the degree of divergence of the human rDNA units within an individual NOR are only partially known. To address this lacuna, we applied the TAR cloning technology to isolate individual rDNA units from chromosomes 21 and 22. This approach revealed an unexpectedly high level of heterogeneity in mature 18S/28S rRNA sequences, raising the possibility of corresponding variations in ribosome dynamics. A functional analysis of these variants in vivo, i.e., in the presence of many copies of rRNA genes in acrocentric chromosomes, is impossible. To overcome this obstacle, we are using a yeast rDNA assay developed by Dr. Susan Liebman's lab. They constructed the yeast strain, in which the chromosomal rDNA repeats were completely deleted, and cell growth was supported by the presence of a single kind of rDNA repeat units on a 2-micron DNA plasmid. Using plasmid shuffling, this system can be used to examine the effects of any mutational changes of rDNA on growth of yeast cells, on translational accuracy, and on rRNA structures involved in the ribosome assembly and/or function. Using this system, we are testing dozen sequence variants identified in 18S and 28S regions that are conservative between human and yeast. Specifically, each variant in the conservative segments is introduced to the corresponding position of yeast rDNA and a modified complete yeast rDNA unit 9 kb in size was used for plasmid shuffling. We have now applied the TAR cloning strategy to characterize the structure and variations of rDNA in the mouse genome. Mice were chosen because genetically engineered mouse models (GEMMs) of human disorders, including cancer and aging, have been created. Although rDNA heterogeneity and instability in mice are well documented, until now only one mouse rDNA unit has been assembled based on the 45S rRNA sequence and one available BAC carrying the IGS region. Such a non-representative reference sequence precludes employing computational and bioinformatic methods to identify rDNA variants. To isolate mouse rDNA units, the TAR vector carrying YAC and BAC cassettes was used. Unexpectedly, the size of the isolated clones did not correspond to the published 45 kb mouse rDNA reference sequence BK000946. 24 independent TAR-isolated rDNA-containing clones are represented by three-size classes (35-40 kb, 40-45 kb and 45-50 kb). DNA sequencing confirmed the size of inserts in each BAC clone. Their sequence comparison revealed that size difference is determined by different size of the IGS region. Note that such size variations in the IGS region have never been observed for human and primate rDNA units. Because the observed size heterogeneity of rDNA units makes it difficult to assemble a reference sequence, we assembled three consensus sequences, i.e., 1 (38,954 bp), 2 (42,007 bp) and 3 (46,251 bp) corresponding to three types of rDNA units. Comparison of these consensus sequences revealed both insertions and deletions of SINE and other repetitive elements within the IGS region. In contrast, the region encoding 45S rRNA is relatively stable. Analysis of this region revealed 206 variations between TAR-isolated rDNA units and the published "standard" 45S rRNA sequence. We plan to continue the characterization of mouse rDNA. rDNA units will be TAR-cloned from three different mouse strains to clarify if these IGS size variants are common in mice. Finally, we will employ radial TAR cloning developed in our lab to identify sequences flanking the rDNA clusters in the mouse chromosomes [distal (DJ) and proximal (PJ) regions]. Information about these sequences is critical to analyze the dynamics of mouse NORs during aging and carcinogenesis. Mammalian centromeres direct faithful genetic inheritance and are typically characterized by regions of highly repetitive and rapidly evolving DNA. In the recent collaborative work, we described centromere innovations within a mouse species. We focused on a mouse species, Mus pahari, that we found has one chromosome with a massive expansion of a newly evolved repeat array that houses 20,000 functional CENP-B boxes: 100-fold more than on the other M. pahari centromeres. The balance of pro- and anti-microtubule-binding by the new centromere permits it to segregate during cell division with high fidelity alongside the older ones whose sequence creates a markedly different molecular composition. We previously constructed a synthetic HAC vector (tetO-HAC) that has a great potential for the study of assembly and maintenance of human kinetochore as well as for gene therapy and synthetic biology. To extend the utility of the tetO-HAC, we have developed two assays for measuring chromosome instability (CIN). These assays allow: i) ranking different anticancer drugs and cytotoxic plant extracts with respect to their effects on chromosome transmission fidelity and ii) identifying new CIN genes that may be targets for cancer therapy. Using HAC-based assays we were able to rank different anticancer drugs with respect to their effects on CIN. Drugs with various mechanisms of action were included in the analysis. New and potentially less toxic compounds that selectively elevate CIN in cancer cells identified by the HAC-based screening tool could lay the foundation for new treatment strategies for cancer. Recently, the HAC-based assay was used to evaluate contribution of the most common glioma missense mutations in IDH1 and TP53 genes on chromosome transmission. Based on this analysis, IDH1R132H and TP53R248Q missense mutations result in a high level of CIN in cancer cells. We also investigated the sensitivity of malignant glioma cells carrying these mutations to CIN-inducing drugs - paclitaxel, vincristine, and etoposide - compared to conventional temozolomide. New and potentially less toxic agents that selectively elevate CIN to promote cancer cell death identified in this study could lay the foundation for new treatment strategies of gliomas. An abnormal chromosome number is a feature of most solid tumors and is often accompanied by an elevated rate of chromosome instability (CIN). Gain or loss of entire chromosomes leads to large-scale changes in gene copy number and expression levels. Mutations in CIN genes are thought to be an early event in tumor development. At present, approximately 400 human genes that control proper chromosome transmission have been annotated with gene ontology terms, while systematic CIN gene screens in yeast have revealed more than 900 genes. Therefore, it may be supposed that many human CIN genes remain unidentified. In our previous work, we developed a high-throughput assay for identification of new human CIN genes using the tetO-HAC expressing a degron-destabilized EGFP. In the current study, we have been used an available at NCATS Ambion collection of 19,000 siRNAs covering the whole human genome for identification of new CIN genes. As a result, 250 new CIN candidate genes have been identified. The experiments on reconfirmation of 250 CIN candidate genes have been performed using newly developed siRNAs and CRISPR knock-out experiments and live-cell microscopy. As a result, 44 new CIN genes were identified. The experiments are in progress to clarify a function of these genes in proper chromosome transmission. Identification of new CIN genes should create opportunities for the development of new therapeutic strategies to target the CIN phenotype of cancer cells.