Chromosomal evolution in the South American marsh rats, Holochilus chacarius and Holochilus brasiliensis.

Abstract

Chromosomal rearrangements in natural populations are often presumed to be underdominant and therefore important in speciation or else heterotic and consequently adaptive, but are rarely viewed as neutral. This study tests these alternative possibilities for chromosomal evolution in the South American marsh rat, genus Holochilus. Chromosomal variation was described in two samples (collected in 1979 and 1986) from one locality in Paraguay and from 8 additional localities in Paraguay and Argentina. Wild-caught animals were used to establish a breeding colony, and the effects of Robertsonian B-chromosome variability on pairing, recombination, and segregation in meiosis were studied. Results indicate that Holochilus chacarius has an unusually high level of chromosomal polymorphism, characterized by four distinct classes of rearrangements. These polymorphisms are temporally stable in nature and do not appear to affect fertility in laboratory crosses. The variation is widespread, but shows no geographic pattern. Chromosomal data indicate that H. brasiliensis vulpinus is a distinct species from H. chacarius, and that different patterns of chromosomal variability characterize these two groups. Studies of meiosis reveal that Rb heterozygotes regularly pair in prophase I. Rb heterozygotes have significantly more chiasmata than homozygotes, and this difference is due to an increase in the number of terminally located chiasmata. Rates of non-disjunction were uniformly low ($<$10%) in homozygous individuals and individuals heterozygous for 1, 2, 3, and 4 Rb rearrangements, including translocations with monobrachial homology. No significant differences were found in rates of non-disjunction among any of the animals studied, and there was no trend for higher mal-segregation with increasing structural heterozygosity. Meiotic drive in males appears to favor the transmission of acrocentrics over metacentrics in Rb heterozygotes. B-chromosomes did not usually pair in meiosis, although some achiasmatic associations were observed. B-chromosomes did not affect the number or distribution of chisamata among the A complement. The segregation of B-chromosomes appears to be random, with no evidence for either Mendelian inheritance or meiotic drive. These data fail to support the underlying assumptions in two different chromosomal speciation models, fail to support a "parasitic" model of B-chromosome evolution, and fail to support a "heterotic" model of B-chromosome evolution.