Project description:Recurrent somatic hotspot mutations of DICER1 appear to be clustered around each of four critical metal binding residues in the RNase IIIB domain of DICER1. This domain is responsible for cleavage of the 3’ end of the 5p-miRNA strand of a pre-mRNA hairpin. To investigate the effects of these cancer-associated “hotspot” mutations we engineered mouse Dicer1-deficient ES cells to express wild-type and an allelic series of the mutant human DICER1 variants. Global miRNA and mRNA profiles from cells carrying the metal binding site mutations were compared to each other and wild-type human DICER1. The miRNA and mRNA profiles generated through the expression of the hotspot mutations were virtually identical, and the DICER1 hotspot mutation carrying cells were distinct from both wild-type and Dicer1-deficient cells. Further, miRNA profiles showed mutant DICER1 results in a dramatic loss in processing of mature 5p-miRNA strands but were still able to create 3p-strand miRNAs. Messenger-RNA profile changes were consistent with the loss of 5p-strand miRNAs and showed enriched expression for predicted targets of the lost 5p derived miRNAs. We therefore conclude that cancer-associated somatic hotspot mutations of DICER1, affecting any one of four metal binding residues in the RNase IIIB domain, are functionally equivalent with respect to miRNA-processing and are hypomorphic alleles, yielding a global loss in processing of mature 5p-strand miRNA. We further propose that this resulting 3p-strand bias in mature miRNA expression likely underpins the oncogenic potential of these hotspot mutations. A total of 28 Affymetrix Mouse Gene ST arrays were done for mRNA expression profiling of various DICER1 mutants (n=14), wildtype controls (n=6), vector only (n=3) and parental cell lines (n=5).
Project description:DICER1 plays a critical role in microRNA (miRNA) biogenesis. Recurrent somatic “hotspot” mutations at four mental binding sites within the RNase IIIb domain of DICER1, were identified in ovarian sex cord-stromal tumors and have since been described in other pediatric tumors. In this study, we identified and characterized DICER1 hotspot mutations in endometrial cancers derived from The Cancer Genome Atlas (TCGA) and our local tumor bank. DICER1 hotspot mutations are found in ~2% of endometrial tumors. Using Illumina and Sanger targeted resequencing we observed biallelic DICER1 mutations in more than 50% of cases with hotspot mutations and identified an additional recurrent mutation G1809R in 2 cases. Through small RNA deep sequencing and real-time PCR, we demonstrated mutations that add a positively charged side chain to residue 1809 have similar detrimental effects on 5p miRNA production as mutations at metal binding sites. In one case G1809R was compound heterozygous with a germline S839F mutation, which contributes to loss of DICER1 expression by promoting protein degradation. As expected, 5p miRNAs are globally reduced in tumors and cell lines with hotspot mutations. Pathway analysis of gene expression profiles indicated that genes derepressed due to loss of 5p miRNAs are strongly associated with cell cycle related pathways. Using a Dicer null cell line model, we demonstrated that DICER1 hotspot mutants abolished the inhibitory effects of wildtype DICER1 on cell proliferation upon re-expression. Furthermore, targets of let-7 family miRNAs are enriched among the upregulated genes, suggesting loss of let-7 may be impacting downstream pathways.
Project description:Recurrent somatic hotspot mutations of DICER1 appear to be clustered around each of four critical metal binding residues in the RNase IIIB domain of DICER1. This domain is responsible for cleavage of the 3’ end of the 5p-miRNA strand of a pre-mRNA hairpin. To investigate the effects of these cancer-associated “hotspot” mutations we engineered mouse Dicer1-deficient ES cells to express wild-type and an allelic series of the mutant human DICER1 variants. Global miRNA and mRNA profiles from cells carrying the metal binding site mutations were compared to each other and wild-type human DICER1. The miRNA and mRNA profiles generated through the expression of the hotspot mutations were virtually identical, and the DICER1 hotspot mutation carrying cells were distinct from both wild-type and Dicer1-deficient cells. Further, miRNA profiles showed mutant DICER1 results in a dramatic loss in processing of mature 5p-miRNA strands but were still able to create 3p-strand miRNAs. Messenger-RNA profile changes were consistent with the loss of 5p-strand miRNAs and showed enriched expression for predicted targets of the lost 5p derived miRNAs. We therefore conclude that cancer-associated somatic hotspot mutations of DICER1, affecting any one of four metal binding residues in the RNase IIIB domain, are functionally equivalent with respect to miRNA-processing and are hypomorphic alleles, yielding a global loss in processing of mature 5p-strand miRNA. We further propose that this resulting 3p-strand bias in mature miRNA expression likely underpins the oncogenic potential of these hotspot mutations.
Project description:Although early developmental processes involve cell fate decisions that define the body axes and establish progenitor cell pools, development does not cease once cells are specified. Instead, most cells undergo specific maturation events where changes in the cell transcriptome ensure that the proper gene products are expressed to carry out unique physiological functions. Pancreatic acinar cells mature post-natally to handle an extensive protein synthetic load, establsih organized apical-basal polarity for zymogen granule trafficking, and assemble gap-junctions to perimt efficient cell-cell communication. Despite significant progress in defining transcriptional networks that control initial acinar cell specification and differentiation decisions, little is know regarding the role of transcription factors in the specification and maintenance of maturation events. One candidate maturation effector is MIST1, a secretory cell-restricted transcription factor that has been implicated in controlling regulated exocytosis events in a number of cell types. Embryonic knock-out of MIST1 generates acinar cells that fail to establish an apical-basal organization, fail to properly localize zymogen granule and fail to communicate intra-cellularly, making the exocrine organ highly suceptible to pancreatic diseases. In an effort to identify the gene expression differences responsible for MIST1 regulating mature acinar properties. We generated a tamoxifen-inducible mouse model where MIST1 expression could be activated in vivoand performed gene expression arrays on wildtype, MIST1-null, and induced MIST1 pancreatic RNA. RNA was isolated from pancreata of 8 week old mice using the Qiagen RNeasy Midi kit. Pancreta of wildtype, MIST1-null, and MIST1-null with a tamoxifen inducible MIST1-expressing transgene were harvested 36 hours post-tamoxifen administration. Therefore, this experiment provides information on steady-state gene expression differences between wildtype and MIST1-null mice as well as immediate gene expression changes induced by MIST1 expression.