Search

Breast cancer is a heterogeneous genetic disease driven by the accumulation of individual mutations per tumor. Whole-genome sequencing approaches have identified numerous genes with recurrent mutations in primary tumors. Although mutations in well characterized tumor suppressors and oncogenes are overrepresented in these sets, the majority of the genetically altered genes have so far unknown roles in breast cancer progression. To improve the basic understanding of the complex disease breast cancer and to potentially identify novel drug targets or regulators of known cancer-driving pathways, we analyzed 86 wild-type genes and 94 mutated variants for their effect on cell growth using a serially constructed panel of MCF7 cell lines. We demonstrate in subsequent experiments that the metal cation transporter CNNM4 regulates growth by induction of apoptosis and identified a tumor suppressive role of complement factor properdin (CFP) in vitro and in vivo. CFP appears to induce the intracellular upregulation of the pro-apoptotic transcription factor DDIT3 which is associated with endoplasmic reticulum-stress response.

Nascent strand sequencing (NS-seq) is used to discover DNA replication origins genome-wide, allowing identification of features for their specification. NS-seq depends on the ability of lambda exonuclease (λ-exo) to efficiently digest parental DNA while leaving RNA-primer protected nascent strands intact. We used genomics and biochemical approaches to determine if λ-exo digests all parental DNA sequences equally. We report that λ-exo does not efficiently digest G-quadruplex (G4) structures in a plasmid. Moreover, λ-exo digestion of nonreplicating genomic DNA (LexoG0) enriches GC-rich DNA and G4 motifs genome-wide. We used LexoG0 data to control for nascent strand–independent λ-exo biases in NS-seq and validated this approach at the rDNA locus. The λ-exo–controlled NS-seq peaks are not GC-rich, and only 35.5% overlap with 6.8% of all G4s, suggesting that G4s are not general determinants for origin specification but may play a role for a subset. Interestingly, we observed a periodic spacing of G4 motifs and nucleosomes around the peak summits, suggesting that G4s may position nucleosomes at this subset of origins. Finally, we demonstrate that use of Na+ instead of K+ in the λ-exo digestion buffer reduced the effect of G4s on λ-exo digestion and discuss ways to increase both the sensitivity and specificity of NS-seq.

The CDKN2A locus encodes for two distinct tumor suppressor proteins, p16 INK4A and p14 ARF , involved in cell cycle regulation. CDKN2A germline mutations have been associated with familial predisposition to melanoma and other tumor types. Besides bona-fide pathogenic mutations, many sequence variants have been identified, but their effect is not well known. We detected the p.Gly23Asp missense mutation in one of the two tested melanoma patients of a family with three melanoma cases. Even though the mutated amino acid is located in a conserved domain that specifically binds to and blocks the function of CDK4/6, its lack of segregation with disease suggested a series of functional assays to discriminate between a pathogenic variant and a neutral polymorphism. The effect of this mutation has been investigated exploiting four p16 INK4A properties: its ability (i) to bind CDK4, (ii) to inhibit pRb phosphorylation, (iii) to evenly localize in the cell, and (iv) to cause cell cycle arrest. The mutant protein properties were evaluated transfecting three different cell lines (U2-OS and NM-39, both p16-null, and SaOS 2, p53 and pRb-null) with plasmids expressing either p16 wt , p16 23Asp , or the p16 32Pro pathogenic variant. We found that p16 23Asp was less efficient than p16 wt in CDK4 binding, in inhibiting pRb phosphorylation, in inducing G1 cell cycle arrest; moreover, its pattern of distribution throughout the cell was suggestive of protein aggregation, thus assessing a pathogenic role for p16 23Asp in familial melanoma.

Origins of DNA replication (ORIs) occur at defined regions in the genome. Although DNA sequence defines the position of ORIs in budding yeast, the factors for ORI specification remain elusive in metazoa. Several methods have been used recently to map ORIs in metazoan genomes with the hope that features for ORI specification might emerge. These methods are reviewed here with analysis of their advantages and shortcomings. The various factors that may influence ORI selection for initiation of DNA replication are discussed.

[Background]: Germline mutations in the BRCA1 and BRCA2 genes are associated with increased risks of breast and ovarian cancers. Although several common variants have been associated with breast cancer susceptibility in mutation carriers, none have been associated with ovarian cancer susceptibility. A genome-wide association study recently identified an association between the rare allele of the single-nucleotide polymorphism (SNP) rs3814113 (ie, the C allele) at 9p22.2 and decreased risk of ovarian cancer for women in the general population. We evaluated the association of this SNP with ovarian cancer risk among BRCA1 or BRCA2 mutation carriers by use of data from the Consortium of Investigators of Modifiers of BRCA1/2. [Methods]: We genotyped rs3814113 in 10 029 BRCA1 mutation carriers and 5837 BRCA2 mutation carriers. Associations with ovarian and breast cancer were assessed with a retrospective likelihood approach. All statistical tests were two-sided. [Results]: The minor allele of rs3814113 was associated with a reduced risk of ovarian cancer among BRCA1 mutation carriers (per-allele hazard ratio of ovarian cancer = 0.78, 95% confidence interval = 0.72 to 0.85; P = 4.8 × 10-9) and BRCA2 mutation carriers (hazard ratio of ovarian cancer = 0.78, 95% confidence interval = 0.67 to 0.90; P = 5.5 × 10-4). This SNP was not associated with breast cancer risk among either BRCA1 or BRCA2 mutation carriers. BRCA1 mutation carriers with the TT genotype at SNP rs3814113 were predicted to have an ovarian cancer risk to age 80 years of 48%, and those with the CC genotype were predicted to have a risk of 33%. [Conclusion]: Common genetic variation at the 9p22.2 locus was associated with decreased risk of ovarian cancer for carriers of a BRCA1 or BRCA2 mutation., Spanish National Cancer Center (CNIO) and the Spanish Consortium: Partially supported by Fundación Mutua Madrileña, Asociación Española Contra el Cáncer, and the Spanish Ministry of Science and Innovation (FIS PI08 1120).

Purpose A growing number of sequence changes of unknown clinical significance are being identified in the BRCA1 gene. However, these variants cannot be used for identification and surveillance of at-risk individuals unless their pathogenic role can be demonstrated. The frequency of these variants makes research on this subject a relevant topic in the field of predisposition to breast and ovarian cancers. Herein, we investigate the pathogenicity of the BRCA1 p.Val1688del (c.5181_5183delGTT) variant, which recurs in our population. Patients and Methods Recent studies have drawn attention to different strategies that, if considered singly, do not usually provide sufficient power to firmly state for or against causality, thus forcing to a re-evaluation of the literature on each specific variant. To increase the power of our study, we used a recently described strategy that integrates data from multiple independent evidences. By this approach, we analyzed data from the comprehensive study of 12 breast/ovarian cancer families carrying p.Val1688del. Results We succeeded in integrating five independent evidences of disease causality including segregation, tumor pathology, and evolutionary and epidemiologic data. Under this model, we obtained a final score of 349,000:1 in favor of disease causality. This result largely matches established cutoffs, and thus is readily translatable into a clear clinical message. Conclusion We show that p.Val1688del is a pathogenic mutation deriving from a common founder. Notably, this study alone increases by 15% the number of BRCA1-positive families in our patients’ cohort, thus substantially contributing to explain many of the families wherein prediction of a BRCA1 mutation contrasted with the absence of a molecular recognizable defect.