We examined seven adult males and seven female nestlings collected in two mixed forest parks in Novosibirsk (54,50N; 83,05E and 55.09N, 82.95E). The males were captured with bird nets at the beginning of the breeding season. Nestling females on days 3–6 after hatching were collected from the nests.

Mitotic chromosome spreads were prepared from short-term bone marrow cell cultures incubated in Dulbecco’s modified Eagle’s medium (ThermoFisher Scientific, cat# 41965062) with 10 μg/ml colchicine (Sigma, cat# C9754) for 2 h at 37°C. The cells were swollen in 0.56% KCl, fixed in fresh Carnoy’s solution (methanol:glacial acetic acid, 3:1). The cell suspension was dropped on clean, cold, wet microscope slides and spread by air-drying. Chromosomes were stained with DAPI dissolved in Vectashield antifade solution (Vector Laboratories, cat# H-1200-10, United States).

Meiotic chromosome spreads were prepared from a suspension of testicular cells of adult males treated with hypotonic solution (0.88% KCl) for 3 h at 37°C and then with Carnoy’s solution. The cell suspension was dropped onto clean, cold, wet coverslips (60 × 24 mm, 0.17 mm thick), and dried. Chromosomes were stained with 0.1% Giemsa solution (Biovitrum, cat# 20-043\L).

Chromosome spreads for SC analysis were prepared from testes and ovaries by the drying down method (Peters et al., 1997). Testes and ovaries were dissected and placed for 30–60 min in an extraction buffer containing 30 mM Tris (Helicon, cat# 77-86-1, Russia), 50 mM sucrose (Sigma, cat# S7903-1KG), 17 mM trisodium citrate dehydrate (Chimmed, cat# A1227436-500.0, Russia), 5 mM EDTA (Panreac&AppliChem, cat# A5097), and pH 8.2. Then, small pieces of testis or the whole ovary were macerated in 40 µl of 100 mM of sucrose, pH 8.2, on a glass slide. A fine suspension was made, and 20 µl of the suspension was gently dropped at the slide moistened by 1% paraformaldehyde (Sigma-Aldrich, cat# 158127) solution, pH 9.2, containing 0.15% Triton X-100 (Sigma, cat# T8787). The slides were dried for 2 h, washed in 0.4% Kodak Photo-Flo 200 (Kodak, cat# 742057), and dried at room temperature.

Immunostaining was performed according to Anderson et al. (1999) using rabbit polyclonal anti-SYCP3 (1:500; Abcam, cat# ab15093), mouse monoclonal anti-MLH1 (1:50; Abcam, cat# ab14206), human anticentromere (ACA) (1:100; Antibodies Inc., cat# 15–234), and rabbit polyclonal anti-H3K9me3 (1:50; Abcam, cat# ab8898) primary antibodies. The secondary antibodies used were Cy3-conjugated goat antirabbit (1:500; Jackson ImmunoResearch, cat# 111-165-144), FITC-conjugated goat antimouse (1:50; Jackson ImmunoResearch, cat# 115-095-003), and AMCA-conjugated donkey antihuman (1:100; Jackson ImmunoResearch, cat# 709-155-149). Antibodies were diluted in PBT [3% bovine serum albumin and 0.05% Tween 20 in phosphate-buffered saline (PBS)]. Slides were incubated in a solution of 10% PBT for 40 min to reduce the unspecific binding of the antibodies. Primary antibody incubations were performed overnight in a humid chamber at 37°C and secondary antibody incubations for 1 h at 37°C. After antibody incubations, slides were washed three times in PBST (PBS with 0.1% Tween 20) for 10 min. Slides were mounted in Vectashield antifade mounting medium (Vector Laboratories, cat# H-1000-10, United States).

The DNA probe for the great tit GRC was prepared by microdissection of 15 round Giemsa-positive bodies located near spermatocytes and spermatids at the conventionally prepared meiotic chromosome spreads. According to Pigozzi and Solari (1998) they contain GRC ejected from the germ cells. DNA isolated from the microdissected material was amplified with the GenomePlex Single Cell Whole Genome Amplification Kit (Sigma-Aldrich, cat# WGA4) according to the manufacturer’s protocol (Zadesenets et al., 2017). The obtained PCR products were labeled with Flu-dUTP (Genetyx, cat# N-801000, Russia) in additional PCR cycles. DNA probes for zebra finch and pale martin GRCs were prepared as described earlier (Torgasheva et al., 2019). C₀t-1 DNA isolation from pale martin was performed as described earlier (Trifonov et al., 2017).

FISH with the DNA probes on the SC spreads were performed according to a standard protocol with salmon sperm DNA (Ambion, cat# AM9680, United States) as a DNA carrier (Trifonov et al., 2017). Chromosomes were counterstained with DAPI dissolved in Vectashield antifade solution.

Images of DAPI-stained metaphase chromosomes and SC spreads after immunostaining and FISH were captured using a CCD camera installed on an Axioplan 2 compound microscope (Carl Zeiss, Germany) equipped with filter cubes #49, #10, and #15 (ZEISS, Germany) using ISIS4 (METASystems GmbH, Germany).

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 (Reeves, 2001). SCs of GRC and macrochromosomes were identified by their relative lengths and centromeric indices. STATISTICA 6.0 software package (StatSoft, Tulsa, OK, United States) was used for descriptive statistics. All results were expressed as mean ± SD.

A diploid chromosome number (2n) in the somatic cells of the great tit was 80 and corresponded to that previously described (Nanda et al., 2011). Macrochromosome 1 was metacentric; 2 was subacrocentric; 3, 4, 5, and 6 were submetacentric; and all other chromosomes but five were telocentric, forming a row gradually decreasing in length. Z chromosome was a submetacentric macrochromosome of intermediate size. The W chromosome was a large microchromosome. Pachytene karyotype contained six macrobivalents, 33 microbivalents, the sex bivalent (ZZ in males, ZW in females), and an additional bivalent or univalent of a large acrocentric chromosome, which was not present on the bone marrow spreads (Figure 1). We identify this chromosome as a GRC.

We detected mosaicism for the number of GRC copies in pachytene oocytes (Supplementary Table 1). Three out of seven females contained two copies of GRC in all examined cells at the pachytene stage (Figure 1B). Four females were mosaic for GRC copy number. Most of their oocytes contained two GRCs. The proportion of cells with one GRC varied from 2 to 26% (Supplementary Table 1). The GRC bivalents appeared as normal autosomal bivalents with at least one MLH1 focus (Figure 1B) and were distinguished as the only acrocentrics. The GRC univalents were distinguished from the bivalents by a lack of MLH1 signals and less intense labeling with antibodies to SYCP3 (Figure 1C). They were significantly longer than the bivalents (19.4 ± 3.0 and 14.1 ± 5.4 µm, respectively; Mann–Whitney test, p = 0.02).

GRC bivalents differed substantially in average number and distribution of MLH1 foci from the macrobivalents of comparable size (SC1: 17.2 ± 3.7 µm and SC2: 14.7 ± 3.1 µm). Most of the GRC bivalents contained one MLH1 signal. It was always located near the centromere (Supplementary Image 1). The bivalents with two or three MLH1 signals were rare (4.0 and 0.4%, respectively). The average number of MLH1 signals per GRC bivalent was 1.05 ± 0.2. The macrobivalents 1 and 2 contained a significantly higher number of MLH1 foci (3.5 ± 0.9 and 2.7 ± 0.8, respectively, Mann–Whitney test, p = 0.00), which showed a rather even distribution with peaks near the bivalent ends (Supplementary Figure S1).

In total, we analyzed 612 spermatocytes from seven males of the great tit. In all analyzed cells, GRC occurred as a condensed body extensively labeled by antibodies to the centromere proteins. SYCP3 signal was only observed near the proximal end of GRC or its both ends as single or double dots or short lines (Figure 1D).

To estimate the copy number variation of GRC at different stages of spermatogenesis, we used antibodies against histone H3 trimethylated at lysine 9 (H3K9me3). It is a marker of heterochromatic transcriptionally repressed regions (Nicetto and Zaret, 2019). The zebra finch GRC is enriched in H3K9me3 compared with the basic chromosome set during prophase I and after the elimination (del Priore et al., 2014).

Most male germline cells of the great tit contain one strong H3K9me3 signal marking the GRC (Figure 2A). In 15 cells out of the 411 examined (3.6%) we detected two H3K9me3 signals (Figure 2B). Spermatids and spermatozoa usually do not show the H3K9me3 signals. Near some of these cells, we detected condensed chromatin bodies with strong H3K9me3 signal (Figure 2C). Apparently, they were GRCs ejected from the cells. Surprisingly, we found the H3K9me3 signal in a few spermatozoa (3 out of 880 cells examined) (Figure 2D).

To estimate a homology between GRC and the chromosomes of the basic set of the great tit, we performed reverse FISH with the GRC DNA probe. It produced a strong specific signal on the condensed GRC in pachytene spermatocytes (Figure 3A). It also faintly labeled different regions of chromosomes of the basic set. In pachytene oocytes, FISH with suppression of repeated sequences using C₀t-1 DNA produced a strong specific signal on the whole GRC except short pericentromeric region (Figure 3B), where most MLH1 foci were located (Supplementary Image 1). It also labeled pericentromeric regions of most microchromosomes (Figure 3B). We did not detect GRC probe signals in post-meiotic cells.

To estimate a homology between GRC of different species, we carried out cross-species FISH of the zebra finch and pale martin GRC DNA probes with the great tit oocytes and the great tit GRC DNA probe with the zebra finch and pale martin spermatocytes (Figures 3C–F). The zebra finch and pale martin GRC probes produced a clear hybridization signal at the GRC of the great tit. They also faintly labeled its W chromosome (Figures 3C,D). Additionally, the zebra finch GRC DNA probe produced a distinct signal at the short arm of the SC2 (Figure 3C) and the pale martin GRC DNA probe in the middle of the long arm of the SC5 (Figure 3D). The great tit GRC probe gave a strong signal at the middle of one zebra finch microbivalent and the distal half of one pale martin microbivalent (Figures 3E,F). We also found the signals of the great tit GRC probe at telomeres of some macro- and microbivalents of pale martin. We did not detect signals of great tit GRC probe at zebra finch and pale martin GRCs.