Genetic mapping and developmental timing of transmission ratio distortion in a mouse interspecific backcross
© Eversley et al. 2010
Received: 23 May 2010
Accepted: 3 November 2010
Published: 3 November 2010
Skip to main content
© Eversley et al. 2010
Received: 23 May 2010
Accepted: 3 November 2010
Published: 3 November 2010
Transmission ratio distortion (TRD), defined as statistically significant deviation from expected 1:1 Mendelian ratios of allele inheritance, results in a reduction of the expected progeny of a given genotype. Since TRD is a common occurrence within interspecific crosses, a mouse interspecific backcross was used to genetically map regions showing TRD, and a developmental analysis was performed to identify the timing of allele loss.
Three independent events of statistically significant deviation from the expected 50:50 Mendelian inheritance ratios were observed in an interspecific backcross between the Mus musculus A/J and the Mus spretus SPRET/EiJ inbred strains. At weaning M. musculus alleles are preferentially inherited on Chromosome (Chr) 7, while M. spretus alleles are preferentially inherited on Chrs 10 and 11. Furthermore, alleles on Chr 3 modify the TRD on Chr 11. All TRD loci detected at weaning were present in Mendelian ratios at mid-gestation and at birth.
Given that Mendelian ratios of inheritance are observed for Chr 7, 10 and 11 during development and at birth, the underlying causes for the interspecific TRD events are the differential post-natal survival of pups with specific genotypes. These results are consistent with the TRD mechanism being deviation from Mendelian inheritance rather than meiotic drive or segregation distortion.
Commonly used inbred mouse strains, which trace their genetic ancestry primarily to the Mus musculus domesticus subspecies , have extensive interspecific polymorphic differences when compared to Mus spretus. Because of the large number of polymorphisms that are distributed across the genome, interspecific crosses are frequently used to map genes responsible for variation in a variety of phenotypic traits . In crosses between M. musculus and M. spretus only interspecific backcrosses using hybrid females are possible since hybrid males are sterile. However, interspecific backcrosses often result in skewed distributions in the inheritance of polymorphic alleles from the hybrid females, a phenomenon called transmission ratio distortion (TRD) [3–8]. Transmission ratio distortion is defined as statistically significant deviation from the expected 1:1 Mendelian ratios of allele inheritance, resulting in a reduction of the expected progeny of a given genotype.
Transmission ratio distortion involving M. spretus crosses was first identified during linkage testing on Chromosomes (Chrs) 2, 4 and 10 [8–11]. Subsequent efforts attempted to map the causative loci influencing TRD in four backcrosses involving M. spretus . Transmission ratio distortion has also been observed in wild M. musculus populations involving Chr 1 and in commonly derived inbred strains on Chr 11 [12–15].
Among the causes of TRD are meiotic drive, segregation distortion (SD), and deviation from Mendelian inheritance (DMI) . The defining characteristic of meiotic drive is that TRD occurs during female meiosis . Consequently, the resulting gametes are not lost and fertility is unaffected, but the inheritance of adjacent neutral polymorphisms is affected [17, 18]. Meitoic drive is one of the more common examples in which a "selfish gene" drives the preferential selection and fertilization of an oocyte [6, 17]. An example of meiotic drive at the second meiotic division can be seen in the DDK syndrome at the Om locus on mouse Chr 11 [19, 20].
Segregation distortion is due to a chromosomal transmission imbalance that typically occurs after meiosis but prior to fertilization. This mechanism is responsible for the SD system in Droshophila melanogaster and the mouse t-haplotype [21–25]. Finally, DMI occurs as a result of post-fertilization lethality of embryos or neonates with a particular genotype. Therefore, DMI can be used to map loci at which specific alleles have detrimental effect on survival. This is particularly interesting in crosses between closely related species because DMI may provide an important tool to study the genetics of speciation.
In this study we report three independent occurrences of TRD caused by post-meiotic lethality in a single interspecific backcross population between A/J (M. musculus) and SPRET/EiJ (M. spretus) mouse inbred strains. Preferential transmission of M. musculus alleles is observed on Chr 7 and of M. spretus alleles on Chrs 10 and 11. In addition, the Chr 11 TRD is modified by a locus on Chr 3. All three loci showing TRD are consistent with a DMI cause since allele-specific losses are not observed until after birth.
SNP markers displaying TRD
Among 19 SNPs on Chr 10, a single peak was evident at rs4228380 (Figure 1B). The model posits that a single distorted locus is located at 48.5 cM with an expected distortion of 71%. This is in good agreement with the predictions of the GF model (GF = 12.08, 19 d.f., not significant p = 0.88). Incremental adjustments of the location and TRD had no affect on the minimum GF.
Chromosome 11 shows broad distortion spanning 16 SNPs incrementally spaced with a maximum peak of 67% TRD (Figure 1C). After adjusting the location of GF and TRD, the best GF was found to be near rs13481119 at 45 cM distal to the centromere (GF = 3.631, 16 d.f., not significant p = 0.99).
SNP markers showing an interaction influencing TRD
Chr 3 rs3694133
Timing of the TRD effect
Three independent loci causing strong transmission ratio distortion favoring A alleles on Chr 7 and S alleles on Chrs 10 and 11 were identified. This is the third report of TRD involving Chr 11, but the first to document an interaction with Chr 3 [12, 14]. The distorting losses all occur after birth but prior to weaning and are likely the result of allelic incompatibilities influencing pup survival.
The region of maximum distortion on Chr 7 has been reported to harbor many imprinted genes influencing fetal and placental growth, postnatal growth, lethality and viability . Most notably, this region contains orthologs of the imprinted gene cluster involved in the neurological disorders Prader-Willi Syndrome and Angelman Syndrome; Prader-Willi Syndrome is caused by the lack of a functional paternal copy (maternally imprinted), while Angelman Syndrome is caused by a lack of maternal copy . Investigation into the genes underlying Prader-Willi Syndrome reveals that Necdin mutants display respiratory instability and die within the first week after birth . Forty-five percent of mice with maternally transmitted Angelman Syndrome die seven days after birth or display a reduction in post-natal growth and viability [29, 30].
It is more plausible that the distortion evident in this study was due to maternally inherited chromosomes with a paternal imprint. Angelman Syndrome, rather than Prader-Willi Syndrome, is caused by a loss of maternal genetic contribution. Although, Prader-Willi Syndrome causes postnatal lethality, the phenotype of Angelman Syndrome is not 100% penetrant and is consistent with the abundance of homozygous animals in the interspecific backcross. The TRD on Chr 7 could be due to variants within the orthologous Angelman Syndrome imprinting center or strain-specific methylation patterns causing silencing of maternally transmitted S allele(s). Meiotic transmission of the A allele by the heterozygous mother results in a viable homozygous mouse and is unlikely to be influenced by methylated silencing. Imprinting defects have been reported in interspecific crosses [31–33], and disregulation of imprinting is thought to be common cause of hybrid disgenesis .
The TRD on Chr 10 has been previously reported using C57BL/6J and SPRET/EiJ backcrosses [7, 10]; the genetic similarity between A/J and C57BL/6J suggests that the distorted locus is conserved in these strains . In previous reports the Chr 10 TRD was reported to extend from Myb to Ifg1 with an apex at Col6a1/Col6a, near D10Mit242 (41.2 cM) . A similar TRD spanning more than 15 cM and with maximum peak at D10Bir9 (30 cM) was also reported for an interspecific backcross . Reciprocal crosses confirmed a preferential selection of heterozygous alleles over homozygous C57BL/6J alleles similar to the data presented here.
Unlike Chr 7, the TRD interval on Chr 10 is not known to contain imprinted genes. It is possible that the TRD on Chr 10 may be influenced by global strain-specific methylation differences resulting in an incompatibility that is not completely penetrant.
Although epistatic interactions influencing TRD have been reported [6, 36, 37], the current data reports an independent modifier that is not located in a region of statistically significant distortion yet has a striking effect on the Chr 11 TRD. Transmission ratio distortion on Chr 11 is increased among backcross progeny homozygous for A/J genotypes on Chr 3 suggesting a maladaptive incompatibility between Chr 3 AA and Chr 11 AA . There is a possibility for a third undetected modifier influencing the TRD of Chr 11 since the least represented progeny are Chr 3 AA , Chr 11 AA .
Although there are now two reports of TRD occurring in a similar region on Chr 11, the data suggests different mechanism are responsible. Unlike the previous report, which reported a sex skewed TRD without a Chr 3 interaction in a cross between two M. musculus-derived strains , the TRD of Chr 11 reported here does not differentially affect the sexes. Additionally, it is unlikely to be due to the early embryonic lethal DDK syndrome since the timing of TRD is between birth and weaning. Further phenotypic characterization of developing pups will be required to identify the physiological cause of DMI on Chrs 7, 10 and 11 in mouse interspecific backcrosses.
In summary, Mendelian ratios of inheritance occur for Chr 7, 10 and 11 during development and at birth, but not at weaning. Additionally, Chr 3 genotypes influence TRD of Chr 11. The data indicate a differential post-natal survival of pups with specific genotypes. These results are consistent with the TRD mechanism being deviation from Mendelian inheritance rather than meiotic drive or segregation distortion.
All mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Female A/J (A) mice (M. musculus) were crossed to male SPRET/EiJ (S) mice (M. spretus) to generate interspecific hybrids. In all crosses dams are listed first and sire second. Female ASF1 hybrids were backcrossed to A/J males to generate a segregating population of 235 (ASF1)A backcross mice in the initial experiment. Mice were euthanized at eight months of age and DNA extracted from tail and liver tissue using a Purgene DNA Extraction kit (Promega, Madison, WI).
Additional cohorts of (ASF1)A offspring were generated and euthanized within two days of birth (n = 88), or pre-natally between 12.5 - 19.5 days post-coitus (n = 60). For pre-natal samples, placenta and embryo weights were recorded and DNA was extracted from tail tissue by phenol-chloroform extraction.
All animal studies were approved by the Institutional Animal Care and Use Committee of University of North Carolina at Chapel Hill.
DNA samples from (ASF1)A mice were commercially genotyped using 254 informative single nucleotide polymorphisms (SNP) markers (Illumina, San Diego, CA). Additional genotyping was done using a custom Sequenom MassArray SNP Genotyping platform containing 182 SNP markers (Geneseek, Lincoln, NE). Sequenom SNP markers were selected from NCBI Build 37 placed at approximately 10-15 cM intervals (Additional file 1, Table S1).
The microsatellite markers D7Mit309, D10Mit145, D10Mit108, D11Mit338, D11Mit152 and D11Mit225 (Operon Biotechnologies, Huntsville, AL) were used to test for TRD in embryos and neonates. Standard PCR methods were used [38, 39]. PCR products were fractionated using 2%-4% agarose gels and stained with ethidium bromide for visualization.
Evaluation of loci for TRD was performed on each cross using Chi square analysis with one degree of freedom. A corrected version of Montagutelli's GF test for single or multiple loci showing TRD was used as previously described [6, 19]. A correlation analysis was performed to detect associations between genotype and placenta or birth weight. Pair-wise analyses were also performed genome-wide with the primary TRD loci to detect modifier loci. Significance thresholds were corrected for multiple testing .
The authors thank members of our lab for assistance and helpful discussions. CDE was supported by pre-doctoral fellowship CA123636 from the National Institutes of Health. DWT acknowledges support by National Institutes of Health grants CA079869 and CA105417.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.