An unusual supernumerary chromosome has been reported for two related avian species, the zebra and Bengalese finches. This large, germline-restricted chromosome (GRC) is eliminated from somatic cells and spermatids and transmitted via oocytes only. Its origin, distribution among avian lineages, and function were mostly unknown so far. Using immunolocalization of key meiotic proteins, we found that GRCs of varying size and genetic content are present in all 16 songbird species investigated and absent from germline genomes of all eight examined bird species from other avian orders. Results of fluorescent in situ hybridization of microdissected GRC probes and their sequencing indicate that GRCs show little homology between songbird species and contain a variety of repetitive elements and unique sequences with paralogs in the somatic genome. Our data suggest that the GRC evolved in the common ancestor of all songbirds and underwent significant changes in the extant descendant lineages.

Guppy X and Y chromosomes represent an early stage in sex chromosome divergence. Synapsis and recombination between X and Y chromosomes attract special attention because recombination suppression promotes their differentiation, but previous studies have given contradictory results. Linkage analysis indicated that recombination between X and Y was extremely rare (<10%) and occurred in the medial part of the Y chromosome, while cytological analysis demonstrated regular association between the distal ends of the X and Y at diakinesis. In this study, we examine pairing and recombination between X and Y chromosomes using immunolocalization of MLH1 to mark recombination nodules, and genomic in situ hybridization with a male DNA probe to identify the Y-specific heterochromatic region. Pairing between X and Y is initiated distally. Single crossovers were detected in 87% of XY synaptonemal complexes, most often in the distal region and less frequently in a median position indicating that end-to-end associations between X and Y are chiasmatic. Thus, we suggest that the very low frequency of recombination detected by linkage analysis in a previous study resulted from a lack of informative markers in distal regions.

GmbH, Germany). The location of each captured immunostained spread was recorded so that it could be relocated on the slide after FISH. Electron microscopy was carried out using a JEM-1400 electron microscope (JEOL, Tokyo, Japan) at 80 kV. All microscopy studies were carried out at the Center for Microscopic Analysis of Biological Objects of SD RAS (Novosibirsk, Russia). Corel PaintShop Pro X6 (Corel) was used for a correction of image brightness and contrast. Chromosome measurements and generation of recombination maps of GRCs. Centromeres were identified by ACA foci. MLH1 signals were only scored if they were localized on SCs. The length of the SC was measured in micrometers and the positions of MLH1 foci in relation to the centromere were recorded using MicroMeasure 3.3 30. SCs of GRC and macrochromosomes were identified by their relative lengths and centromeric indexes. SC1 is the largest submetacentric. SC2 and SC3 are large subacrocentrics of similar sizes but different centromeric indexes. SC4 is middle size metacentric, SC5 and SC6 are subacrocentrics of the same size, which differ from each other in the centromeric indexes. On bone marrow metaphase chromosome spreads, Z and W are identified as a pair of non-matching macrochromosomes: metacentric and submetacentric (correspondingly). At SC spreads, ZW is identified as macrobivalent with misaligned centromeres and/or asynapsed ends of the axial elements. GRC is identified as the only acrocentric macrobivalent or univalent. To generate recombination maps, we divided the length of the SC into equal intervals approximately equal to 1 µm and plotted the proportion of MLH1 foci located in each interval. STATISTICA 6.0 software package (StatSoft, Tulsa, OK, USA) was used for descriptive statistics. All results were expressed as mean ± SD; p < 0.05 was considered as statistically significant.