Exploring the link between MORF4L1 and risk of breast cancer

Abstract

Introduction: Proteins encoded by Fanconi anemia (FA) and/or breast cancer (BrCa) susceptibility genes cooperate in a common DNA damage repair signaling pathway. To gain deeper insight into this pathway and its influence on cancer risk, we searched for novel components through protein physical interaction screens.

Methods: Protein physical interactions were screened using the yeast two-hybrid system. Co-affinity purifications and endogenous co-immunoprecipitation assays were performed to corroborate interactions. Biochemical and functional assays in human, mouse and Caenorhabditis elegans models were carried out to characterize pathway components. Thirteen FANCD2-monoubiquitinylation-positive FA cell lines excluded for genetic defects in the downstream pathway components and 300 familial BrCa patients negative for BRCA1/2 mutations were analyzed for genetic mutations. Common genetic variants were genotyped in 9,573 BRCA1/2 mutation carriers for associations with BrCa risk.

Results: A previously identified co-purifying protein with PALB2 was identified, MRG15 (MORF4L1 gene). Results in human, mouse and C. elegans models delineate molecular and functional relationships with BRCA2, PALB2, RAD51 and RPA1 that suggest a role for MRG15 in the repair of DNA double-strand breaks. Mrg15-deficient murine embryonic fibroblasts showed moderate sensitivity to γ-irradiation relative to controls and reduced formation of Rad51 nuclear foci. Examination of mutants of MRG15 and BRCA2 C. elegans orthologs revealed phenocopy by accumulation of RPA-1 (human RPA1) nuclear foci and aberrant chromosomal compactions in meiotic cells. However, no alterations or mutations were identified for MRG15/MORF4L1 in unclassified FA patients and BrCa familial cases. Finally, no significant associations between common MORF4L1 variants and BrCa risk for BRCA1 or BRCA2 mutation carriers were identified: rs7164529, Ptrend = 0.45 and 0.05, P2df = 0.51 and 0.14, respectively; and rs10519219, Ptrend = 0.92 and 0.72, P2df = 0.76 and 0.07, respectively.

Conclusions: While the present study expands on the role of MRG15 in the control of genomic stability, weak associations cannot be ruled out for potential low-penetrance variants at MORF4L1 and BrCa risk among BRCA2 mutation carriers.

Figure 1

Mrg15 deficiency confers sensitivity to γ-radiation. Mrg15-null murine embryonic fibroblasts (MEFs) show intermediate sensitivity to γ-radiation relative to controls (WT, wild-type; Morf4l1^-/-, Mrg15-deficient; and Atm^-/-, Atm-deficient). *Significant differences between WT and Morf4l1^-/- MEFs (one-tailed t test, P < 0.01).

Figure 2

Mrg15 deficiency impairs Rad51 foci formation and reduces Brca1 and Brca2 levels. (a) Left panel: number of cells with Rad51 nuclear foci (>4 foci per nuclei) in wild-type and Morf4l1^-/- murine embryonic fibroblast (MEF) clones after (16 hours) treatment with 10 Gy. **Significant difference (two-tailed t test, P < 0.001). Right panel: representative images of Rad51 and pS139-H2ax immunodetection in cultures counted above for foci. DAPI, 4,6-diamidino-2-phenylindole; IR, γ-irradiated. (b) Levels of Brca1, Brca2 and Rad51, and control Actb, in whole cell extracts of Morf4l1^-/- MEFs and wild-type counterparts (three cell clones of each genotype are shown).

Figure 3

Phenotypic study of Caenorhabditis elegans brc-2 and mrg-1 mutants. (a) Representative images of meiotic cells at the distal part, near the gonad bend. RAD-51 foci are bright and nuclear in wild-type (WT) animals whereas RAD-51 foci appear less intense and weakly diffuse in the cytoplasm, reduced but often dispersed and intense in the nuclei, or absent in brc-2, mrg-1 and rad-51 mutants, respectively. There are more RPA-1 nuclear foci in each of the three mutants than in WT animals. 4,6-Diamidino-2-phenylindole (DAPI) panels are merged with the red channel (for WT and brc-2 mutant) and with the green channel (for rad-51 mutant). *Abnormal chromosomal compaction. (b) Quantitation of RAD-51 and RPA-1 foci per nuclei in several germ cell lines of WT animals and brc-2 and mrg-1 mutant animals. Number of cells scored (n) and standard deviation of the mean indicated. **Significant differences relative to WT (Mann-Whitney U test, P < 0.001). (c) SYTO-12 staining in synchronized adult worms. Left top panel: an animal heterozygous for the brc-2 mutation (according to green fluorescent protein expression at the pharynx) shows WT SYTO-12 staining (that is, one to two labeled cells at the gonad bend). Right top and left bottom panels: an increase in SYTO-12-positive cells in the germline of brc-2 and mrg-1 mutants, respectively. Right bottom panel: magnification of the highlighted area in the left panel.

Figure 4

Variation at the MORF4L1 locus and breast cancer risk. (a) SNPs with previous suggestive evidence of association with breast cancer (BrCa) risk in the general population (P_2df < 0.01) [67], genes and the linkage disequilibrium structure around MORF4L1 in HapMap Caucasians (data release 27). (b) Hazard ratio (HR) estimates of association of rs7164529 (top panels) and rs10519219 (bottom panels) with BrCa risk among BRCA1 (left panels) and BRCA2 (right panels) mutation carriers. Graphs show HRs and 95% confidence intervals of heterozygotes and minor allele homozygotes for all participating centers except for rs10519219 and relatively small groups (less than five individuals with the minor genotype). Size of the rectangle is proportional to the corresponding study precision.