Fernando Pardo-Manuel de Villena

Recombination is proportional to the number of chromosome arms in mammals

In sexually reproducing organisms recombination contributes significantly to the processes of DNA repair, fidelity of chromosome segregation, and generation of genetic diversity. Abnormal recombination is associated with decreased fitness of offspring, aneuploidy, and several human diseases (Chaganti et al. 1974; Charlesworth and Barton 1996; Hassold et al. 2000). Despite its importance, little is known of how the rate and pattern of recombination is regulated in higher eukaryotes, although sex, DNA sequence, chromatin structure, chromosomal location, and chromosome size have been shown to influence the recombination rate (Robinson 1996). Mammals show great interspecific variability of recombination rate. For example, the level of recombination in human is more than twice that observed in the mouse. Neither the total DNA content nor the diploid number of chromosomes account for these differences. The DNA content of the mammalian genome is fairly constant and, on average, the haploid genome consists of 3.0 × 10 bp (Sober 1968). Studies in Drosophila and other organisms indicate that proper chromosome segregation generally requires at least one recombination event per chromosome (Mather 1936). Because crossing-over is associated with proper distributive segregation of each bivalent at the first meiotic division, the diploid number of chromosomes establishes the minimum number of recombination events that occurs per meiosis. However, this parameter correlates poorly with the observed recombination frequency in mammals (Burt and Bell 1987). In an under-appreciated report, Dutrillaux (1986) demonstrated empirically that there is a strong correlation between number of chiasma (NC) and the haploid number of chromosome arms. Although he did not provide the correlation statistics for his regression analysis, the variance explained is very high (R 4 0.90; F < 1.32 × 10; b in Fig. 1a). Dutrillaux concluded that NC is proportional to the haploid number of chromosome arms (NF/2) of the karyotype, where NF is the “fundamental number” (Matthey 1945). This result was in agreement with a cytological study that indicated the occurrence of an obligatory chiasmata in each chromosomal arm in human males (Hulten 1974). In the intervening years, genome-wide recombination studies have permitted the establishment of linkage maps of several mammalian species (see below) and provided important information about the process of meiotic non-disjunction, the major cause of aneuploidy (Hassold et al. 2000). In addition, recombination still remains at the center of one of the most vexing problems in evolutionary biology, sex. Burt and Bell (1987) used mammalian chiasma frequencies as a test of two competing hypotheses to explain the selective force resulting in sex and recombination; the Red Queen and the Tangled Bank (Maynard Smith 1978; Bell 1982; Charlesworth 1985). The Red Queen hypothesis proposes that recombination is selected to counter the antagonistic advances made by the biotic environment, particularly parasites, during the previous generations. This hypothesis predicts that recombination is positively correlated with age to maturity, because the longer the time between generations, the more contrary will be the environment and the more intense the selection (Bell 1982). Burt and Bell (1987) demonstrated that the number of chiasma in excess of one per bivalent (“excess of chiasma” or EC) was strongly correlated with age to maturity. However, the number of chromosome arms was not among the parameters tested (haploid number, age to maturity, litter size, and adult body weight) for correlation with EC in this study. We have performed a regression analysis, using 21 of the 24 nondomesticated mammalian species reported by Burt and Bell (1987) for which the karyotypes are available to us. We find that, in agreement with the expectations of Dutrillaux, there is a very strong correlation (R 4 0.79; F < 7.16 × 10; a in Fig. 1a) between NC and the haploid number of chromosome arms (FN/2). The small differences between the two regression lines (NC 4 0.84(N/2) + 5.1 and NC 4 0.85(NF/2) + 7.3; a and b respectively in Fig. 1a) can be ascribed to intra-specific variability, methodological differences, and/or sampling errors. Another estimate of the genome-wide recombination rate in a species is the total length of the linkage map. Linkage maps have been reported for 11 mammalian species. The species for which linkage maps are available may be divided into two groups based on the number of independent maps, map coverage and density, and the number of informative meioses analyzed. In the first group, we include species with more than one reported map, extensive genome coverage, and moderate to high marker density. This group includes human, mouse, pig, dog, rat, and cattle (Serikawa et al. 1992; Barendse et al. 1994; Ellegren et al. 1994; Archibald et al. 1995; Dib et al. 1996; Dietrich et al. 1996, Ma el al. 1996; Rohrer et al. 1996; Bihoreau et al. 1997; Kappes et al. 1997; Andoh et al. 1998; Broman et al. 1998; Rhodes et al. 1998; Szpirer et al. 1998; Neff et al. 1999; Paszek et al. 1999; Steen et al. 1999, Werner et al. 1999; Kitada et al. 2000; McCarthy et al. 2000; Mellersh et al. 2000; Mouse Genome Informatics, http:// www.informatics.jax.org September 2000). As observed for NC, the size of the linkage map is strongly correlated with the haploid number of chromosome arms (FN/2) (R 4 0.81; F < 0.014; Fig. 1b). In Figure 1b, we have also included five species (cat, goat, sheep, horse and baboon, plotted as open circles) for which linkage maps are deficient in one or more of the criteria mentioned above (Crawford et al. 1995; Vaiman et al. 1996; Lindgren et al. 1998; Schibler et al. 1998; Menotti-Raymond et al. 1999; Kiguwa et al. 2000; Rogers et al. 2000; Swinburne et al. 2000). Although these Correspondence to: C. Sapienza; E-mail: sapienza@unix.temple.edu

Natural selection and the function of genome imprinting: beyond the silenced minority

Most hypotheses of the evolutionary origin of genome imprinting assume that the biochemical character on which natural selection has operated is the expression of the allele from only one parent at an affected locus. We propose an alternative - that natural selection has operated on differences in the chromatin structure of maternal and paternal chromosomes to facilitate pairing during meiosis and to maintain the distinction between homologues during DNA repair and recombination in both meiotic and mitotic cells. Maintenance of differences in chromatin structure in somatic cells can sometimes result in the transcription of only one allele at a locus. This pattern of transcription might be selected, in some instances, for reasons that are unrelated to the original establishment of the imprint. Differences in the chromatin structure of homologous chromosomes might facilitate pairing and recombination during meiosis, but some such differences could also result in non-random segregation of chromosomes, leading to parental-origin-dependent transmission ratio distortion. This hypothesis unites two broad classes of parental origin effects under a single selective force and identifies a single substrate through which Mendels first and second laws might be violated.

Confirmation of maternal transmission ratio distortion at Om and direct evidence that the maternal and paternal “DDK syndrome” genes are linked

The polar, preimplantation-embryo lethal phenotype known as the“DDK syndrome” in the mouse is the result of the complex interaction of genetic factors and a parental-origin effect. We previously observed a modest degree of transmission-ratio distortion in favor of the inheritance of DDK alleles in the Ovum mutant (Om) region of Chromosome (Chr) 11, among offspring of reciprocal F1-hybrid females and C57BL/6 males. In this study, we confirm that a significant excess of offspring inherit DDK alleles from F1 mothers and demonstrate that the preference for the inheritance of DDK alleles is not a specific bias against the C57BL/6 allele or a simple preference for offspring that are heterozygous at Om. Because none of the previous genetic models for the inheritance of the“DDK syndrome” predicted transmission-ratio distortion through F1 females, we reconsidered the possibility that the genes encoding the maternal and paternal components of this phenotype were not linked. We have examined the fertility phenotype of N2 females and demonstrate that the inter-strain fertility of these females is correlated with their genotype in the Om region. This result establishes, directly, that the genes encoding the maternal and paternal components of the DDK syndrome are genetically linked.

A high-resolution map of the regulator of the complement activation gene cluster on 1q32 that integrates new genes and markers.

Abstract Sixteen microsatellite markers, including two described here, were used to construct a high-resolution map of the 1q32 region encompassing the regulator of the complement activation (RCA) gene cluster. The RCA genes are a group of related genes coding for plasma and membrane associated proteins that collectively control activation of the complement component C3. We provide here the location of two new genes within the RCA gene cluster. These genes are PFKFB2 that maps 15 kilobases (kb) upstream of the C4BPB gene, and a gene located 4 kb downstream of C4BPA, which seems to code for the 72 000 Mr component of the signal recognition particle (SRP72). Neither of these two genes is related structurally or functionally to the RCA genes. In addition, our map shows the centromere-telomere orientation of the C4BPB/MCP linkage group, which is: centromere-PFKFB2-C4BPB-C4BPA-SRP72-C4BPAL1-C4BPAL2-telomere, and outlines an interval with a significant female-male recombination difference which suggests the presence of a female-specific hotspot(s) of recombination.

The maternal DDK syndrome phenotype is determined by modifier genes that are not linked to Om

Abstract. The DDK syndrome is a polar, early embryonic lethal phenotype caused by incompatibility between a maternal factor of DDK origin and a paternal gene of non-DDK origin. Both maternal factor and paternal gene have been mapped to the Om locus on mouse Chromosome (Chr) 11. The paternal contribution to the syndrome has been shown to segregate as a single locus. Although the inheritance of the maternal contribution has not been characterized in depth, it as been assumed to segregate as a single locus. We have now characterized the segregation of the DDK fertility phenotype in over 240 females. Our results demonstrate that females require at least one DDK allele at Om to manifest the syndrome. However, the DDK syndrome inter-strain cross-fertility phenotype of heterozygous females is highly variable and spans the gamut from completely infertile to completely fertile. Our results indicate that this phenotypic variability has a genetic basis and that the modifiers of the DDK syndrome segregate independently of Om.

H2-haplotype-dependent unequal transmission of the 1716 translocation chromosome from Ts65Dn females

Ts65Dn is a segmentally trisomic Chromosome (Chr) 16 mouse model for Down’s syndrome (DS; Davisson et al. 1993; Reeves et al. 1995). Ts65Dn mice (2n 4 41) are trisomic for a segment of mouse Chr 16 that spans just proximal to Appto Mx1,homologous to the DS “critical region” in human Chr 21 (Davisson et al. 1993; Reeves et al. 1995). This trisomy is compatible with survival to adulthood. Ts65Dn results from a balanced translocation of the distal end of Chr 16 to the centromeric end of Chr 17 to form a 17 small extra chromosome (Davisson et al. 1993; Fig. 1). Ts65Dn males are sterile, whereas Ts65Dn females are fertile, but the transmission of the 17 16 Chr is non-Mendelian (Davisson et al. 1993). The cause for this transmission ratio distortion (TRD) is unknown (Davisson et al. 1993). However, it has been noted that significant but modest deviations from Mendelian expectations should be replicated in independent experiments to exclude random events (Montagutelli et al. 1996; Pardo-Manuel de Villena et al. 2000). We performed two crosses between Ts65Dn females and euploid (B6EiC3H-a/A)F1 males that provide the opportunity to test whether the non-Mendelian transmission of the 17 16 Chr is reproducible. As is shown in Tabl e 1 a significant excess of euploid offspring is found in both crosses (H 0: Mendelian inheritance of the 17 chromosome,x 4 5.57, 1 d. f.,P < 0.025; andx 4 9.83, 1 d. f.,P < 0.002, in crosses I and II, respectively). These results confirm that the unequal recovery of trisomic progeny, at birth, is a constant feature in these crosses and not the result of chance events. All these crosses show similar and modest levels of TRD [61 ± 9, 67 ± 14, and 62 ± 7, Davisson and coworkers (1993), cross I and cross II, respectively; TRD is expressed as percent ± SD). Incomplete selection is normally interpreted as the result of epistatic effects of alleles at other locus or loci (Montagutelli et al. 1996). Here we present evidence for a locus (or loci) linked to the H2 region on Chr 17 that modifies the levels of TRD. This locus was identified on the basis of non-independent segregation of H2 haplotypes and transmission of the 17 16 chromosome. TheH2 haplotype was determined in parents and offspring because the original goal of our experiments was the characterization of Ts65Dn mice as a model for immunological alterations in DS (Paz-Miguel et al. 1999; F. Leyva-Cobia ́n and J.E. Paz-Miguel, unpublished observations). In both crosses, the haplotype of the sires isH2; in cross I the haplotype of the dams is H2 and in cross II the haplotype of the dams is H2. This mating protocol allows us to test whether recovery of trisomic progeny is independent of theH2 haplotype. In cross II independence between the transmission of the 17 16 translocation chromosome and H2 haplotype can be rejected (H 0: independent assortment, x 2 4 9.94, 2 d. f., P < 0.02, corrected for performing two tests, one for the presence of TRD and another for independent assortment). Table 1 shows that failure of the null hypothesis results from the strong selection against trisomic progeny in offspring inheriting the H2 haplotype (82% and 18% euploid and trisomic offspring, respectively) and Mendelian assortment in offspring inheriting the H2 haplotype (51% and 49% euploid and trisomic offspring, respectively). Lastly,H2 offspring show an intermediate level of TRD that is consistent with a selection against the 17 16 Chr in association with theH2 haplotype. The allele-specific effect of the H2 haplotype in the transmission of 17 16 chromosome to the offspring is supported by the results of cross I, in which a similar trend for selection against trisomic progeny in association with the H2 haplotype is observed (Table 1). Davisson and coworkers (1993) proposed that the deficiency in maternal transmission of the 17 16 Chr “was due to either in utero or preweaning loss of trisomic progeny or segregation distortion.” Determining the origin of maternal TRD is of great importance because it offers the possibility of identifying genes involved in either unequal segregation during female meiosis or embryoniclethal mutations. A reduction in both litter size and the number of litters in Ts65Dn females has been reported previously (Davisson et al. 1993). The reduction in fecundity in Ts65Dn females suggests that TRD may result from differential survival of trisomic and euploid embryos owing to gene imbalance for regions on two different mouse chromosomes (Fig. 1). The Chr 16 region present in the 17 Chr is unlikely to be involved in the origin of TRD because no TRD has been reported for Ts1Cj (Sago et al. 1998). Ts1Cj is a partial Chr 16 trisomy model for DS in which the distal part of Chr 16 has been translocated to the end of mouse Chr 12 (Sago et al. 1998). This indicates that gene imbalance in the region present in the rearranged chromosomes of both Ts65Dn and Ts1Cje (Sod1-Mx1) can be excluded as the cause for TRD (Fig. 1). Carriers of the 17 16 Chr are also partially trisomic for the proximal region of Chr 17 (Fig. 1). This region is thought to be small, on the basis of the absence of markers D17Mit58 and D17Mit34in somatic cell hybrids containing the 17 16 Chr, whereas they are present in the reciprocal 16 17 translocation product (Reeves et al. 1995). Although this region is not expected to contribute to the DS phenotype (Reeves et al. 1995), two features make it especially relevant for the origin of TRD; this region includes the centromere of the 17 16 Chr and is linked to the modifier locus identified in this report. If TRD results from unequal segregation of the 17 16 Chr during female meiosis, then the centromere of the 17 16 Chr must play a role in this phenomenon. Different alleles at the three Chr 17 centromeres may result in pairing differences and unequal segregation. Unequal segregation during female meiosis has been demonstrated in the mouse in several instances (Gropp and Winking 1981; Agulnik et al. 1991; Pachierotti et al. 1995; Pardo-Manuel de Villena et al. 2000). Robertsonian translocations (2n 4 39) are subject to unequal segregation during female meiosis, independent * Both authors contributed equally to this work.