Experiment
To characterize the genetic architecture of hybrid male sterility on a genome wide scale an F2 intercross between two wild derived lines of inbred mice were used.
M. m. domesticus mice, WSB/EiJ and M. m. musculus, PWD/PhJ mice were crossed in reciprocal directions to generate F1 hybrids that were then bred with siblings. Pups were weaned into same sex groups at 21 days and males were separated into individual cages at ~56 days. Males were then phenotype at 70 days.
Five diagnostic measures of subfertility and sterility were quantified: testis weight (absolute weight of right testis), sperm density (estimated using a Makler chamber), sperm head morphology ( measured in samples collected from the epididymides), proportion of abnormal sperm (4 primary abnormalities: proximal bent tail, distal bent tail, missing head or missing tail, and severely amorphous head) and cross sectional area of seminiferous tubules (measured only in stage VII tubules).
Genomic DNA was extracted from liver tissue. SNP markers were designed from the Perlegen phase 4 release of the mouse resequencing project. A total of 331 SNP markers were genotyped using the Sequenom iPLEX MassARRAY system. After conservative filtering steps for quality control, 198 SNPs were genotyped across the autosomes, X Chromosome, Y Chromosome and mitochondrion of 310 F2 male mice.
All phenotypes were analyzed using standard interval mapping except for abnormal sperm types, which were analyzed using the extended Haley-Knott method. Genome wide significance thresholds were calculated from 1000 permutations of the autosomes and separate permutations for the X chromosome. For phenotypes that did not follow normal distribution results from several alternative mapping procedures were compared, including nonparametric interval mapping, interval mapping for binary traits, and a two part model for phenotypes that exhibited a spike near zero in the distribution.
Standard, single QTL interval mapping detected significant QTL for each hybrid sterility phenotype in the F2 male mice [Table 3, Fig 4, Fig S2].
Each X-linked QTL associated with the M. m. musculus genotype (PWD/PhJ) was associated with a large reduction in fertility. In contrast, infertility was associated with both
M. m. domesticus (WSB/EiJ) and M. m. musculus (PWD/PhJ) alleles at autosomal QTL. For 6 of the 11 QTL the allele causing hybrid male sterility was recessive. One case of underdominance, where the least fertile genotype was associated with heterzygous males, was observed at locus Spdq2 on Chr 17. [Table 3.]
Two-QTL model analysis was performed and several pairs of QTL were identified affecting each other in an additive manner; however, little evidence of epistatic interactions were observed among the QTL.
Multiple models that jointly fit any number of QTL were also considered using an automated forward/backward step wise search algorithm. In addition to statistical improvements over the one-QTL model several additional QTL were found [Table 4].
In single QTL interval mapping QTL Rttwq1 (right testis weight QTL 1) mapped to Chromosome 2 with a LOD score of 5.42 at 30.0 cM (96.0 Mb), the 1.5 LOD interval spanned from 52.2 to 123.9 Mb. In the multiple model analysis the QTL mapped to 33 cM on Chr 2, LOD=6.06, CI 52.2-128.5 Mb, accounting for 5.32 % of trait variance with the WSB/EiJ allele, inherited recessively, associated with sterility.
In single QTL interval mapping QTL Rttwq2 (right testis weight QTL 2) mapped to Chromosome 4 with a LOD score of 8.66 at 45.9 cM (108.9 Mb), the 1.5 LOD interval spanned from 98.8 to 115.3 Mb. In the multple model analysis the QTL also mapped to 45.9 cM on Chr 4, LOD=14.40, CI 98.8-115.3, accounting for 13.49% of trait varaince with the PWD/PhJ allele contributing to sterility in an additive inheritance pattern.
In single QTL interval mapping QTL Rttwq3 (right testis weight QTL 3) mapped to Chromosome 10 with a LOD score of 6.36 at 20.0 cM (69.3 Mb), the 1.5 LOD interval spanned from 55.1 to 87.4 Mb. In the multiple model analysis the QTL mapped to 21 cM on Chr 10, LOD=9.33, CI 51.3-82.7, accounting for 9.33 % of trait variance, with the WSB/EiJ allele contributing to sterility in an additive inheritance pattern.
Additional right testical weight QTL were mapped in the multiple model analysis:
QTL Rttwq4 (right testis weight QTL 4) mapped to 54.0 cM on Chr 3, LOD=8.06, CI 108.5-143.7 Mb, accounting for 7.19% of trait varaince with PWD/PhJ alleles contributing to sterility when inherited in a dominant pattern.
QTL Rttwq5 (right testis weight QTL 5) mapped to 60.0 cM on Chr 4, LOD=5.40, CI 121.8-148.0 Mb, accounting for 5.40% of trait variation with WSB/EiJ alleles contributing to sterility in a additive pattern.
QTL Rttwq6 (right testis weight QTL 6) mapped to 3.0 cM on Chr 13, LOD=10.54, CI 12.2-24.0 Mb, accounting for 9.58% of trait variance with PWD/PhJ alleles contributing to sterility in an additive pattern.
QTL Rttwq7 (right testis weight QTL 7) mapped to 24.0 cM on Chr 17, LOD=3.74, CI 3.1-73.2 Mb, accounting for 3.23% of trait variance with underdominant inheritance contributing to sterility.
For relative right testis weight, an interaction between QTL Rttwq4 on Chr 3 and QTL Rttwq6 on Chr 13 included a significant epistatic effect resulting in significantly lower weight when the genotypes at both QTL were homozygous for M. m. musculusPWD/PhJ alleles.
The LOD score was 6.1 and accounted from 5.4% of trait variance.
Focusing on interactions involving the X chromosome, conditioned on PWD/PhJ (M. m. musculus) alleles, 5 autosomal QTL were detected interacting with the X Chromosome. For each abnormal sperm QTL in the interaction there were higher proportions of abnormal sperm when the X-linked PWD/PhJ allele was combined with autosomal QTL homozygous for the M.m domesticusWSB/EiJ alleles. [Table 5.]
Analysis Tools