Positional analyses of BRCA1-dependent expression in Saccharomyces cerevisiae
Mutations in BRCA1 account for a significant proportion of hereditary breast and ovarian cancers, but analysis of BRCA1 function is complicated by pleiotropic effects and binding partners (Pol II holoenzyme and transcription factors, chromatin remodelers, recombination complexes and E3 ligases). In vertebrate cells, efforts to elucidate BRCA1 transcriptional effects have focused on specific genes or restricted portions of the genome – limiting analyses of BRCA1 effects on adjoining DNA sequences and along chromosome lengths. Here, we use microarray analyses on the genetically tractable yeast cell system to elucidate BRCA1-dependent genome-wide positional effects on both gene induction and repression. Yeast responses may be of clinical relevance based on findings that BRCA1 severely diminishes yeast growth kinetics but that BRCA1 mutated at sites identified from breast tumors is no longer able to retard yeast. Our analysis reveals that BRCA1 acts through both transcription factors to up-regulate specific loci and chromatin remodeling complexes to effect global changes in gene expression. BRCA1 also exhibits gene repression activities. Cluster-functional analysis reveals that these repressed factors are required for mitotic stability and provide a novel molecular explanation for the conditional lethality observed between BRCA1 and chromosome segregation genes.
RVS support from the Susan G. Komen for the Cure Foundation (Award BCTR0707708)
Gene Expression Profiles of HCT116-p53-/- cells Treated with Paclitaxel or Stathmin-siRNA
Among the most difficult cancers to treat are those having mutated or non-functional p53, with concurrent overexpression of stathmin (stmn1), amicrotubule (MT) destabilizer (e.g. Yuan et al. 2006. J. Pathol. 209: 549-8). Alli et al. (2007. Oncogene. 26:1003-12) demonstrated that this type of cancer cell undergoes apoptosis in response to stmn1 knock down by siRNA. The mechanism by which stmn1 siRNA promotes apoptosis in cancerous cell lines is not understood, while efficient delivery of targeted siRNA therapies remains an elusive challenge. In this work, we aim to understand the apoptotic processes activated by stmn1 siRNA in a p53-null human colon cancer cell line, HCT116-p53-null, and specifically what is unique about stmn1 siRNA induced apoptosis compared to that induced by paclitaxel treatment, a MT stabilizing drug. We find that both stmn1 depletion and paclitaxel addition slowed cell growth and caused ~25% cell death beginning ~48 h after initial treatment. Microarray analyses of gene expression profiles were then examined to provide an unbiased screen for genes up or down regulated by either treatment. Time courses (~55-72 h) were analyzed for annotated genes showing expression fold changes of greater than or equal to 2. In stmn1 siRNA treated cells, we found 215 down-regulated and 157 up-regulated genes. Paclitaxel treatment resulted in 162 down-regulated and 642 up-regulated genes. Of the differentially regulated genes from each treatment, the number of shared genes included 27 of 350 (8%) down-regulated genes and 89 of 710 (13%) up-regulated genes. These data indicate that most expression changes were unique to each treatment. Comparing those genes whose expression is oppositely changed between the two treatments also demonstrated a low level of overlap (95 of 1081 genes, or ~9%) between treatments. To date our results indicate that stmn1 siRNA and paclitaxel induce apoptosis through unique upstream signals, although each treatment acts to stabilize MTs. Continued analysis and microarray verification are ongoing to confirm our conclusions. Understanding how stmn1 siRNA induces apoptosis could lead to identification of novel chemotherapy targets.
Partially funded by an HHMI grant to Lehigh and NIH (LC).
Loss of Endogenous Oncoprotein18/Stathmin in Mouse Embryo Fibroblasts Induces Changes in Tubulin Isoform Expression with Minimal Changes to Microtubule Dynamics
Op18/Stathmin is a ubiquitous microtubule (MT) destabilizing protein linked to cancer and cell health: Op18 is over-expressed in leukemias and its expression level correlates with breast cancer stage progression. We are using MEFs (mouse embryonic fibroblasts) WT (+/+), heterozygous (+/-), or knockout (-/-) for the Op18 gene to further characterize Op18s roles in MT polymerization and dynamics. MT polymer level and nucleation rate increased with loss of Op18. In contrast, loss of one or both copies of the Op18 gene results in surprisingly modest changes to MT dynamics. For example MT dynamicity, a measure of total tubulin addition and loss from MT ends, and catastrophe frequency were similar in all three lines. Since MT dynamics depend on tubulin (Tb) isotype composition (Panda et al., 1994), we used quantitative reverse transcription-PCR to measure differences in mRNA levels for each Tb isotype. The alpha-Tb I mRNA level did not change across genotypes, but protein level nearly doubled in the (-/-) line compared to WT. Cells (-/-) for Op18 also under-express mRNA for Tbs II (20% less) and IV (36% less) and over-express Tb III (78% more) compared to WT. This change in Tb isotype expression is consistent with that observed in taxol-resistant breast cancer cells (Shalli et al., 2005). Other differentially regulated mRNAs include increases in alpha-Tb VII and VIII, and Tb VII. We conclude that cells respond to loss of Op18 by changing the ratio of tubulin isoforms, allowing cells to maintain dynamic MT turnover. We hypothesize that it is changes to MT polymer content and specific Tb isoform expression, rather than changes to MT dynamics, that are responsible for Op18s role in cancer and cell survival. Thanks to Jutta Marzillier, Lehigh Genomics Facility, and G. Shyumyatsky, Rutgers University (mice).
Funded by NIH.