Rat Mcs1b is concordant to the genome-wide association-identified breast cancer risk locus at human 5q11.2 and MIER3 is a candidate cancer susceptibility gene

Abstract

Low-penetrance alleles associated with breast cancer risk have been identified in population-based studies. Most risk loci contain either no or multiple potential candidate genes. Rat mammary carcinoma susceptibility 1b (Mcs1b) is a quantitative trait locus on RN02 that confers decreased susceptibility when Copenhagen (COP)-resistant alleles are introgressed into a Wistar Furth (WF)-susceptible genome. Five WF.COP congenic lines containing COP RN02 segments were compared. One line developed an average of 3.4 ± 2.0 and 5.5 ± 3.6 mammary carcinomas per rat ± SD when females were Mcs1b-resistant homozygous and Mcs1b heterozygous, respectively. These phenotypes were significantly different from susceptible genotype littermates (7.8 ± 3.1 mean mammary carcinomas per rat ± SD, P = 0.0001 and P = 0.0413, respectively). All other congenic lines tested were susceptible. Thus, Mcs1b was narrowed to 1.8 Mb of RN02 between genetic markers ENSRNOSNP2740854 and g2UL2-27. Mammary gland-graft carcinoma susceptibility assays were used to determine that donor (P = 0.0019), but not recipient Mcs1b genotype (P = 0.9381), was associated with ectopic mammary carcinoma outcome. Rat Mcs1b contains sequence orthologous to human 5q11.2, a breast cancer susceptibility locus identified in multiple genome-wide association studies. Human/rat MAP3K1/Map3k1 and mesoderm induction early response (MIER; MIER3)/MIER3 are within these orthologous segments. We identified MIER3 as a candidate Mcs1b gene based on 4.5-fold higher mammary gland levels of MIER3 transcripts in susceptible compared with Mcs1b-resistant females. These data suggest that the human 5q11.2 breast cancer risk allele marked by rs889312 is mammary gland autonomous, and MIER3 is a candidate breast cancer susceptibility gene.

Figure 1. Rat Chr 2 map of WF.COP lines delimiting Mcs1b to 1.8 Mb

Markers used to genotype WF.COP congenics are listed in relative positions on the y-axis. Lines are labeled with letter-number combinations and designated with filled dark-gray bars to indicate Mcs1b resistant alleles. Lines that are drawn with unfilled bars represent COP intervals incapable of conferring decreased susceptibility or resistance to mammary carcinoma development. The filled light-gray bars at ends of each congenic segment are intervals of unknown genotype. Lines T and K were published previously [33].

Figure 2. Rat Mcs1b is mammary gland autonomous

Percentage of mammary gland graft positive recipients that developed ectopic mammary gland carcinomas are shown for each susceptible (S) and Mcs1b resistant (R) donor:recipient group. Groups with a S donor are shown as filled bars, and groups with a R donor are shown as unfilled bars. The total number of mammary gland graft positive recipients that were evaluated for tumor outcome in each group were, respectively, 27, 22, 23, and 18 for S:S, R:R, S:R, and R:S.

Figure 3. Rat Mcs1b contains the ortholog of GWAS identified human 5q11.2 breast cancer risk allele marked by SNP rs889312 and multiple human/rat conserved transcripts

Genetic markers used to map Mcs1b to this region are marked on the x-axis. Ends of relevant congenic lines are included for orientation. Potential candidate breast cancer susceptibility gene transcripts mapping to rat Mcs1b are shown as filled black bars that represent exonic and intronic DNA. Gene names are flanking 5′-UTRs of respective transcripts. Elements shown as gray bars and designated as A, B, C, or D correspond to predicted Actbl2, ENSRNOG00000013098, C5orf35, and ENSRNOG00000034909, respectively.

Figure 4. Rat Mcs1b resistant genotype is associated with higher body weight and rat/human Mier3/MIER3 levels are higher in rat mammary and human breast carcinomas

Panel A: Lower body weight at 12 weeks of age was observed in mammary carcinoma susceptible (diamonds) compared to Mcs1b resistant females (squares) with DMBA and without (P < 0.0001 and P = 0.0007, respectively). Body weight was significantly higher in susceptible females that received DMBA compared to without (P < 0.0001). Panel B: Rat Mier3 transcript levels were significantly higher in DMBA-induced mammary carcinomas (triangles) compared to non-diseased mammary gland tissue (circles, P = 0.0120). Mean ± SD are graphed for each variable. Panel C: Human MIER3 was significantly higher in breast carcinomas compared to pathologically normal breast tissues. Oncomine (www.oncomine.org, reference of main text) was used to query The Cancer Genome Atlas (cancergenome.nih.gov) gene expression database. Shown are box plots of log₂ median centered MIER3 transcript levels for invasive ductal breast carcinomas (IDBC, n=392) and invasive lobular breast carcinomas (ILBC, n=36) compared to pathologically normal breast tissues (Breast, n=61). *P < 0.05. ^†P = 0.0569.

Figure 5. Ectopically-expressed eGFP-MIER3 localizes to nuclei in breast cancer cell lines

Plasmid expression vectors containing eGFP_MIER3 ORFs were transiently transfected into MDA-MB-231 or T47D breast cancer cell lines. At 24 h cells were fixed in formalin and stained with DAPI to visualize nuclei. Confocal microscopy was used to visualize DAPI staining and green fluorescence. Green fluorescence from eGFP-MIER3 and DAPI nuclear staining co-localized in both cell lines (Merged).