Genome assembly of the Neotropical marsh rat Holochilus nanus (Cricetidae: Sigmodontinae) brings insights on B and sex chromosome evolution

TL;DR

This study sequenced the genome of a marsh rat to better understand the evolution of B and sex chromosomes in rodents.

Contribution

The paper provides new insights into B and sex chromosome evolution through genome assembly and analysis in Holochilus nanus.

Findings

Ancient genome duplications were identified that may originate from B chromosomes.

Shared sequence blocks were found between B and Y chromosomes, suggesting genomic mosaicism.

Expressed sequences in B chromosomes suggest a functional role beyond repetitive DNA.

Abstract

The Neotropical region comprises approximately 27% of mammal diversity, and rodents of the tribe Oryzomyini represent a significant portion of that. This diversity is reflected in the karyotype variability of the tribe, with a huge number of chromosomal rearrangements involving autosomal, sex, and B chromosomes. Supernumerary B chromosomes were described for more than 10 species, four of them belonging to the genus Holochilus. Therefore, we sequenced the genome of two H. nanus specimens with different karyotypes: a female with (HNA-XXB) and a male without (HNA-XY) a B chromosome. We also sequenced previously flow-sorted chromosomes from this species: two B (HNA-B1, HNA-B2) and the Y chromosome (HNA-Y). Genome assemblies of HNA-XY and HNA-XXB were compared, enabling the identification of ancient genome duplications that could result from fragments of the B chromosome. In addition, more than fifty scaffolds containing sequence blocks shared between the libraries of HNA-B1, HNA-B2, and HNA-Y were found. The sequence blocks mapped in metaphases of H. nanus presented hybridization signals on the centromeric region of the chromosomes, highlighting that the centromeric composition of H. nanus is highly variable. In addition, qPCR and RT-qPCR analysis evidenced that these sequences are expressed, indicating a role in the genome structure. Briefly, supernumeraries of H. nanus appear to be a mosaic of the genome and may contain genes and sequences crucial for its maintenance, in addition to its repetitive DNA fraction.

The online version contains supplementary material available at 10.1007/s00335-026-10205-8.

Figures

Genome assembly statistics of (a) HNA-XY and (b) HNA-XXB. c Comparative alignment between the assemblies HNA-XY and HNA-XXB, insert of genomics rearrangements highlighted in (d) and (e). d Details of the main divergent point between the genomes (arrow heads), and e green/blue dots corresponding to ancient genome duplications (arrow heads). The mean percent identity of the diagonal line is represented by the color scale shown in the bar at the bottom left of the figure

Plots representing the per base coverage ratio between HNA-XY (blue) and HNA-XXB (red) genomes, for the scaffolds: a S_3326, b S_3354, c S_8120, d S_23778, and e S_32154. FISH with the S_30215 probe on metaphase of HNA-XXB, the metaphase is represented as: f DAPI staining, g FITC signals, h DAPI and FITC merged, and i G-banding pattern. qPCR and RT-qPCR analysis of sequence blocks on HNA-XY (blue) and HNA-XXB (red) samples: j gene dosage ratio of gDNA; and k relative expression of cDNA

Full text

Introduction

The Neotropical region comprises about 1800 species of mammals, corresponding to 27% of the class diversity (Túnez et al. 2021; Burgin et al. 2025). Rodent species of the tribe Oryzomyini represent a significant amount of this diversity, reaching around 180 extant species (Percequillo et al. 2021; Burgin et al. 2025). These species are distributed from the southeastern United States to the southernmost portion of South America, occupying all biomes through these territories (Prado and Percequillo 2013, 2018).

Concerning cytogenetic features, Oryzomyini are also highly diverse (Di-Nizo et al. 2017; Moreira et al. 2020). Diploid number (2n) ranges from 16 to 88, and autosomal and sex chromosome polymorphisms were reported across all the group phylogeny (Reig et al. 1990; Barros et al. 1992; Moreira et al. 2020). A comparative cytogenetic analysis by Zoo-FISH using the entire chromosome set of Holochilus nanus (2n = 56 + 2Bs, XY) as probes was performed in 15 species of Oryzomyini (Moreira et al. 2022). The results showed many chromosomal rearrangements, including autosomal, sex, and B chromosomes, involved in the karyotype evolution of the group (Moreira et al. 2022).

Supernumerary B chromosomes are extra genomic elements, mainly composed of repetitive DNA (Trifonov et al. 2010; Vujošević et al. 2018), and present in more than 1600 eukaryotic species (D’Ambrosio et al. 2017; Jones 2017). B chromosomes are most frequently found in rodents within the mammalian class, and approximately 66 species harbor these chromosomes (Vujošević et al. 2018). Only in the tribe Oryzomyini, more than 10 species were already described, containing B chromosomes on their karyotype complement (Moreira et al. 2020), four of them belong to the genus Holochilus (Yonenaga-Yassuda et al. 1987; Sangines and Aguilera 1991; Nachman 1992; Moreira et al. 2020). B chromosome probes of H. nanus presented hybridization signals in the autosomal, sex, and/or B chromosomes of 13 Oryzomyini species (Ventura et al. 2015; Moreira et al. 2022). Repetitive DNA analysis of these isolated B chromosomes highlighted that they are enriched by Short Interspersed Nuclear Element (SINE), Long Terminal Repeats (LTR), and simple repeats (Moreira et al. 2023).

Repetitive DNA sequences play an important role in chromosomal evolution, being responsible for maintaining chromosome integrity or inducing the occurrence of rearrangements (Morrish et al. 2002; Erickson et al. 2011). In the Neotropical subfamily Sigmodontinae, an expansion of the Endogenous Retrovirus (ERV) mysTR and an extinction of Long Interspersed Nuclear Element (LINE) were suggested to be responsible for the high karyotype variability during Sigmodontinae radiation (Cantrell et al. 2005; Rinehart et al. 2005; Erickson et al. 2011). Almost half of the repetitive DNA content of the H. nanus genome is composed of LTR elements. In addition, the landscape of repetitive DNA in this species shows a first insertion wave of LINE, followed by an expansion of LTR/ERV (Moreira et al. 2023).

Herein, we performed an entire genomic analysis of two H. nanus specimens, a male (HNA-XY) and a female (HNA-XXB), plus two B (HNA-B1 and HNA-B2) and the Y chromosome (HNA-Y) of H. nanus, which were previously isolated by flow sorting. We also performed a comparative analysis between the genome assembly of HNA-XY and HNA-XXB to identify and characterize B chromosome sequences and understand the contribution of its composition in the H. nanus karyotype and genomic variability.

Materials and methods

Cell cultures and sampling

Our sample consists of fibroblast cell lines of two specimens of H. nanus: (i) a male (BIO 634) without a B chromosome, 2n = 56 and FN = 56 (HNA-XY); and (ii) a female (BIO 327) harboring one B chromosome, 2n = 56 + 1B and FN = 56 (HNA-XXB). These fibroblast lines were previously established at the cell collections of the Laboratório de Citogenética de Vertebrados, Instituto de Biociências, Universidade de São Paulo, Brazil. Both specimens were collected at São Bento (02°43′S; 44°50′W), Maranhão State, Brazil, and deposited at the Museu de Zoologia of Universidade de São Paulo. Cell lines were cultured in Dulbecco’s modified Eagle’s medium, supplemented with 20% of fetal bovine serum to obtain whole genomic DNA (gDNA), RNA, and chromosome suspensions (Freshney 1986). gDNA were purified with the PureLink^®^ Genomic DNA Kit (Invitrogen™). RNA was purified with the PureLink™ RNA Mini Kit (Invitrogen™), treated with DNase (Thermo Fisher Scientific, USA), and converted to cDNA libraries with the High-Capacity RNA-to-cDNA™ Kit (Applied Biosystems).

Genomic sequencing

Illumina paired-end sequencing was performed from HNA-XY and HNA-XXB gDNA. In addition, the chromosome probes corresponding to two B chromosomes (HNA-B1 and HNA-B2) plus the Y chromosome (HNA-Y) of H. nanus were also sequenced. These chromosome probes were previously isolated by chromosome flow sorting and amplified by a degenerate oligonucleotide-primed PCR with the 6 MW primer (Ventura et al. 2015). Sequencing was performed using the Macrogen Inc. (Korea) service. Libraries were constructed with TruSeq Nano DNA Kit. For each sample, fragments of 151 bp were generated.

Genome assembly

Raw read quality was analyzed using FastQC (Andrews 2010). Low-quality reads were discarded, and Illumina adapters were trimmed using Trimmomatic-0.39 (Bolger et al. 2014) and BBMap-38.49 (Bushnell et al. 2017). For the gDNA data set, the first 15 bases of the read were cut off, and reads shorter than 95 bp were dropped. For the chromosome probes data set, the filtering processes were performed in three steps: (i) trimmed Illumina adapters; (ii) trimmed 6 MW primer sequence (5’ CCGACTCGAGNNNNNNATGTGG ‘) and quality filter; and (iii) cut off the first 30 bases of the reads (https://github.com/MoreiraCN/Filtering_Illumina_sequences). Filtered reads were used to assemble the genomes of HNA-XY and HNA-XXB using Meraculous-v2.2.6 (Chapman et al. 2011). The assembly of each sample, along with the filtered reads, was used to reassemble these genomes using SPAdes-3.14.0 (Nurk et al. 2013; https://github.com/MoreiraCN/Assembling_Illumina_sequences). Metrics values of final assemblies were obtained with QUAST-5.0.2 (Gurevich et al. 2013) and Assembly-Stats (https://github.com/rjchallis/assembly-stats). Mitochondrial DNA (mtDNA) of HNA-XY and HNA-XXB was assembled from the raw data using NOVOPlasty-4.3.1 (Dierckxsens et al. 2017). Illumina adapters were trimmed with Trimmomatic-0.39 (Bolger et al. 2014), and the mtDNA of Mus musculus (GRCm38 - NCBI RefSeq assembly GCF_000001635.20) was used as seed.

Genome annotation

Prediction of protein-coding genes on the genome assembly of HNA-XY and HNA-XXB was performed with Braker-v28.2 (Brůna et al. 2021; https://github.com/MoreiraCN/Genome_annotation_Braker). Both genomes were soft-masked by RepeatMasker-4.1.1 (Smit et al. 2013), using a custom repeat library created with RepeatModeler (Smit and Hubley 2015). Genes were annotated using the similarity with protein sequences of M. musculus (GRCm39 - https://ftp.ensembl.org/pub/release-110/fasta/mus_musculus/pep/) as a reference with Blast (Altschu et al. 1990). Blast results were filtered to keep primarily high protein alignment coverage (> 90%) and high sequence identity (> 90%) for each annotation. Next, unused IDs and unidentified annotations were re-analyzed following the same criteria, but now with a relaxed filter: first, keep annotation IDs with > 70% identity and > 70% of alignment coverage. The number of annotated genes for each assembly was also accessed using the BUSCO tool, from the dataset glires_odb10 (Simão et al. 2015). mtDNA of HNA-XY and HNA-XXB were annotated using Mito Fish (Bernt et al. 2013 (https://mitofish.aori.u-tokyo.ac.jp/annotation/input/).

Genomic rearrangements

The occurrence of genomic rearrangements between HNA-XY and HNA-XXB was identified using the approach of whole genome pairwise sequence alignments. The genomes were aligned using minimap2-2.24 (Li 2018) and the output file was used to generate a dot plot with R script DotPlotly (https://github.com/tpoorten/dotPlotly). Different parameters were tested in the command line, in addition to self-alignments, to better interpret the results. Finally, the following parameters were used: -s (color alignments by similarity) -t (calculate the identity for on-target alignments) -m 500 (filter out alignments shorter than 500 bp) -q 1000 (filter out queries with alignment length below 1000 bp) -k 56 (diploid chromosome number).

Coverage ratio analysis

Coverage ratio analysis was conducted using the pipeline CovDetect (Valente et al. 2014; https://github.com/ivanrwolf/CovDetect/tree/master). Briefly, filtered libraries (HNA-XXB, HNA-B1, HNA-B2, HNA-Y) were aligned to the HNA-XY assembly using bowtie-v2.4.2 (Langmead and Salzberg 2012). Per base coverage was calculated using bedtools depth for each library (https://bedtools.readthedocs.io/en/latest/). The coverage was used as input to CovDetect to retrieve sequence blocks. Regions with less than 15x coverage were discarded, and coverage regions with more than 10.000 bp detected between HNA-XY and HNA-XXB were kept. For the remaining libraries (HNA-XY/HNA-B1, HNA-XY/HNA-B2, HNA-XY/HNA-Y), regions with more than 200 bp were kept (https://github.com/MoreiraCN/Identification_of_sequence_blocks/tree/main). To select some of these regions for fluorescent in situ hybridization (FISH), quantitative PCR (qPCR), and reverse transcriptase-qPCR (RT-qPCR) analysis, each region was plotted using the package Sushi of R (https://bioconductor.riken.jp/packages/3.4/bioc/html/Sushi.html).

Primers construction

A total of 21 regions were selected for FISH, qPCR, and RT-qPCR analysis, and sets of primers were designed for these regions. In addition, the set of primers described by Moreira et al. (2023) for the gene YWHAZ (tyrosine 3-monooxygenase / tryptophan 5-monooxygenase activation protein, zeta polypeptide) was used as a control in the qPCR and RT-qPCR analysis (De Spiegelaere et al. 2015). A list of all primers used, and their sequences are available in Supplementary Table S1.

Fluorescent in situ hybridization

Sequences corresponding to selected regions were amplified by a PCR reaction and used as probes on metaphases of HNA-XY and HNA-XXB. The amplification reaction was performed with Platinum™ Taq DNA Polymerase (Invitrogen™) using gDNA of HNA-XY and HNA-XXB, with the thermocycling conditions of: 94 °C–5 min; 45 cycles: 94 °C–1 min, 60 °C–1 min, and 72 °C–45 s; and 72 °C–5 min. Amplicons were labeled by Nick translation with biotin-16-dUTP (Nick Translation mix, Roche Applied Science). Chromosomes on slides were denatured in 70% formamide/2xSSC (saline-sodium citrate buffer) at 75 °C for 90 s. The hybridization mix consisted of 100 ng of the labeled probe in 50% formamide/2xSSC and was denatured for 10 min at 98 °C and added to the slides. Slides were incubated at 37 °C for 24 h. After the hybridization, the slides were washed in three baths of 2xSSC at 42 °C for 5 min. Immunodetection was performed with avidin+FITC conjugates (Roche Applied Science), and slides were mounted with DAPI 1:500 in Slowfade (Life Technologies). Chromosomes were identified by the G-banding pattern produced after DAPI staining. Analysis was performed in the BX61 Olympus microscope (Olympus, Tokyo, Japan) with an Olympus DP71 digital camera. Metaphase plate images were analyzed using the software Gimp.

Quantitative PCR and reverse transcriptase-qPCR

qPCR and RT-qPCR analysis were performed on a Bio-Rad CFX96TM Real-Time System, using the RealQ Plus Master Mix Green with high Rox (Ampliqon, Odense, Denmark). Values of Gene Dose Ratio (GDR) were obtained by the amplification of gDNA, and the relative expression was calculated by the amplification of cDNA. Cycling conditions were: 95 °C–10 min; 40 cycles: 95 °C–15 s and 60 °C–1 min; 60 °C–1 min; and 95 °C–50 s. GDR was obtained using the method 2 − ΔCt (Bel et al. 2011), and relative expression was determined by the method ΔΔCt (De Santis et al. 2011). The single-copy gene YWHAZ was used as a reference in both methods, as previously described by Moreira et al. (2023). Comparative analysis between HNA-XY and HNA-XXB was performed using the Mann–Whitney test, p values < 0.05 were accepted as significant.

Data Availability

Raw data are available in the database of the National Center for Biotechnology (NCBI), accession numbers are: (i) SRR22427722 for HNA-XY; (ii) SRR22681972 for HNA-XXB; (iii) SRR22443438 for HNA-B1; (iv) SRR22443437 for HNA-B2; and (v) SRR22443436 for HNA-Y (Supp. Table S2). Assemblies are available at NCBI, accession numbers are: (i) SRR22750343 for HNA-XY; and (ii) SRR22750344 for HNA-XXB (Supp. Table S2).

Results

Genome assembly and gene annotation

Millions of Illumina paired-end reads were generated, for entire genomes, the average of sequenced reads was 1,017,179,804, and for isolated chromosomes was 44,263,058 (Supp. Table S2). After trimming, the average reads were 949,091,492 and 39,015,772, for the whole genome and the isolated chromosomes, respectively (Supp. Table S2). It was possible to recover an average of 40,3x genome depth for HNA-XY and HNA-XXB, even after filtering. The assemblies presented very similar metrics, HNA-XY with: 247,806 scaffolds; 2,4 GB total length; 40,58% GC content; 61,893 bp scaffold N50; and 645,799 bp longest scaffold (Fig. 1a, Supp. Table S2), and HNA-XXB with: 117,748 scaffolds; 2,2 GB total length; 40,50% GC content; 42,754 bp scaffold N50; and of 534,501 bp longest scaffold (Fig. 1b, Supp. Table S2). The assembly of mtDNA recovered only one contig with 16,363 bp for both genomes (Supplementary file). Concerning the annotation of genes, 39,229 genes and 41,151 transcripts were recovered for HNA-XY and 33,446 genes and 35,042 transcripts for HNA-XXB assemblies (Supp. Table S2). After the Blast similarity analysis, the final number of annotated genes was 14,350 in HNA-XY and 14,612 in HNA-XXB (Supp. Table S2). For the mtDNA assembly, 37 genes were annotated in HNA-XY and HNA-XXB: 13 genes; 22 tRNAs; and 2 rRNAs (Supp. Fig. S1–2).

Fig. 1. Genome assembly statistics of (a) HNA-XY and (b) HNA-XXB. c Comparative alignment between the assemblies HNA-XY and HNA-XXB, insert of genomics rearrangements highlighted in (d) and (e). d Details of the main divergent point between the genomes (arrow heads), and e green/blue dots corresponding to ancient genome duplications (arrow heads). The mean percent identity of the diagonal line is represented by the color scale shown in the bar at the bottom left of the figure

Fig. 2. Plots representing the per base coverage ratio between HNA-XY (blue) and HNA-XXB (red) genomes, for the scaffolds: a S_3326, b S_3354, c S_8120, d S_23778, and e S_32154. FISH with the S_30215 probe on metaphase of HNA-XXB, the metaphase is represented as: f DAPI staining, g FITC signals, h DAPI and FITC merged, and i G-banding pattern. qPCR and RT-qPCR analysis of sequence blocks on HNA-XY (blue) and HNA-XXB (red) samples: j gene dosage ratio of gDNA; and k relative expression of cDNA

Genome rearrangements

The assemblies of HNA-XY and HNA-XXB were alignment within each other to identify genomic rearrangements between them. The diagonal line resulting from this alignment evidenced a high proportion of homologous sequences, expected for two genomes from the same species (Fig. 1c). However, it was possible to identify gaps in the synteny, indicating the occurrence of genomic rearrangements (Fig. 1d), in addition to duplications observed in the HNA-XXB genome (Fig. 1e). This rearrangement detected could result from the difference in the karyotype composition between HNA-XY and HNA-XXB.

Sequence blocks from B and sex chromosomes

Coverage ratio analysis allows the identification of scaffolds containing a high amount of B and sex chromosome sequences. A total of 34,411 scaffolds containing sequence blocks larger than 200 bp were identified between HNA-XY and HNA-XXB, 109 between HNA-XY and HNA-B1, 128 between HNA-XY and HNA-B2, and 176 between HNA-XY and HNA-Y. To recover larger scaffolds containing sequence blocks between HNA-XY and HNA-XXB, we performed a new filter to identify only scaffolds containing sequence blocks larger than 10.000 bp. Therefore, 713 scaffolds containing sequence blocks were identified between HNA-XY and HNA-XXB. The selected scaffolds between HNA-XY and the chromosome libraries (HNA-B1, HNA-B2, and HNA-Y) can be classified as: (i) 56 scaffolds in common between HNA-B1, HNA-B2 and HNA-Y; (ii) 32 scaffolds in common between HNA-B1 and HNA-B2; (iii) 10 scaffolds in common between HNA-B1 and HNA-Y; (iv) 4 scaffolds in common between HNA-B2 and HNA-Y; (v) 11 scaffolds exclusives of HNA-B1; (vi) 36 scaffolds exclusives of HNA-B2; and (vii) 106 scaffolds exclusives of HNA-Y.

We generated plots for each selected scaffold, and based on these plots, we selected scaffolds to perform FISH, qPCR, and RT-qPCR analysis. The scaffolds selected were: (i) S_3326, S_3354, S_8120, S_23778 and S_32154 between HNA-XY and HNA-XXB (Fig. 2a–e); (ii) S_72479 and S_82143 between HNA-XY and HNA-B1 (Supp. Fig. S3a–b); (iii) S_45658 and S_80568 between HNA-XY and HNA-B2 (Supp. Fig. S3c–d); (iv) S_48272, S_106655 and S_119473 between HNA-XY and HNA-Y (Supp. Fig. S3e–g); (v) S_25900 and S_30215 between HNA-XY, HNA-B1 and HNA-B2 (Supp. Fig. S4a–d); (vi) S_69887 and S_73484 between HNA-XY, HNA-B1 and HNA-Y (Supp. Fig. S5a–d); (vii) S_50734 and S_105625 between HNA-XY, HNA-B2 and HNA-Y (Supp. Fig. S6a–d); and (viii) S_30750, S_53901 and S_66582 between HNA-XY, HNA-B1, HNA-B2 and HNA-Y (Supp. Fig. S7a–i).

Experimental validation

FISH with probes from selected sequence blocks presented the same hybridization signal on metaphases of HNA-XY and HNA-XXB, highlighting the centromeric region of all chromosomal pairs (Supp. Fig. S8–27). However, in some cases hybridization signals were stronger, for instance for the sequence S_30215 (Fig. 2f–i), or weaker, for the sequence S_119473 (Supp. Fig. S19). Moreover, the sequences S_72479 (Supp. Fig. S13), S_45658 (Supp. Fig. S15), and S_66582 (Supp. Fig. S27) also presented a faint hybridization signal dispersed throughout all the chromosomes. No difference was detected in the hybridization pattern between probes labeled with gDNA of HNA-XY and HNA-XXB. qPCR analysis was used to compare the quantity of copies between HNA-XY and HNA-XXB genomes. GDR was higher in the genome HNA-XXB for almost all analyzed sequences (Fig. 2j). The exceptions were S_3326, S_32154, S_25900, S_69887, S_48272, and S_106655, where the number of copies was similar or higher in the HNA-XY genome. Expression of the sequence blocks was verified by amplifying the cDNA of HNA-XY and HNA-XXB with the same primers designed for GDR analysis (Supp. Table S1). Comparisons between HNA-XY and HNAXXB expression showed no significant difference (Fig. 2k), and the majority of sequences presented a low expression level, except S_80568 and S_105625.

Discussion

A review performed by Ruban et al. (2017) highlighted the two main ways to identify B chromosome sequences: (i) sequencing the B chromosome isolated by flow sorting or microdissection; and (ii) sequencing two genomes of the same species, with one of them harboring a B chromosome. In this report, we used both approaches to identify B chromosome sequences of H. nanus. Thus, we sequence the entire genome of a male without a B chromosome, HNA-XY, and a female with one B chromosome, HNA-XXB. In addition to two B chromosomes, HNA-B1 and HNA-B2, plus the Y chromosome, HNA-Y, of H. nanus, previously isolated by flow sorting and fragmented by a DOP-PCR reaction (Ventura et al. 2015). We assembled the genomes of HNA-XY and HNA-XXB (Fig. 1a, b), although it was not possible to assemble the genomes corresponding to the B and Y chromosomes of H. nanus. Unfortunately, the high amount of repetitive DNA sequences typical of B and sex chromosomes becomes a challenge when assembling libraries obtained with short-read sequence technology (Trifonov et al. 2010; Treangen and Salzberg 2012; Vujošević et al. 2018).

Genome assemblies were used to annotate protein-coding genes, with M. musculus as a reference. More than ten thousand genes were annotated in each assembly, 14,350 in HNA-XY and 14,612 in HNA-XXB. We suggested that the 262 additional genes recovered for the HNA-XXB genome are part of the genomic content of the B chromosome of HNA-XXB. The number of annotated genes, tRNAs, and rRNAs in the mtDNA of H. nanus was the same as reported for M. musculus (Quiros et al. 2017).

Genome assemblies of HNA-XY and HNA-XXB were compared to identify the presence of genomic rearrangements. The comparative analysis evidenced an almost continuous diagonal red line (Fig. 1c), indicating that these two genomes are highly similar to each other, despite the difference in karyotype composition (2n = 56 + XY x 2n = 56 + XXB). These results reinforce the conservative nature of mammalian genomes, despite the presence of all chromosomal rearrangements within the group (O’Brien et al. 1999; Graphodatsky et al. 2011). Additionally, it was possible to identify gaps in the diagonal line indicating the occurrence of genomic rearrangements (Fig. 1d) and some green/blue dots dispersed on the left side of the diagonal line (Fig. 1e), corresponding to HNA-XXB genome. These dots represent ancient genome duplications and could be fragments of the B chromosome of HNA-XXB that diverged from the remaining genome. Thus, we suggest that the B chromosome of *H. nanus * is composed of fragmented sequences dispersed throughout the genomic complement. A similar analysis performed between two specimens of the cichlid fish Astatotilapia latifasciata, one of them harboring a B chromosome, highlighted similar results, with the presence of insertions, deletions, and duplications in these genomes (Jehangir et al. 2019).

The identification of regions with different coverage ratios in each library, sequence blocks, was performed using the protocol described by Valente et al. (2014). More than fifty scaffolds containing sequence blocks shared between the sequence libraries of HNA-B1, HNA-B2, and HNA-Y were found. Previous studies using FISH analysis mapped sex and B chromosome sequences from H. nanus in the X, Y, and B chromosomes of several Oryzomyini species, suggesting a common origin of these chromosomes in the tribe (Ventura et al. 2015; Moreira et al. 2022). In addition, the characterization of the repetitive DNA content of HNA-B1, HNA-B2, and HNA-Y evidenced that these elements are enriched with transposable elements, such as SINE and LTR (Moreira et al. 2023). Now, we highlighted that sequences shared between these chromosomes are also present on the centromeric region of all chromosomes of the complement (Fig. 2f–i). Together, those results reinforce the role of the repetitive DNA portion (transposable elements and satellite DNA) that comprises the sex and B chromosomes of H. nanus in the karyotypic reshuffling of species of Oryzomyini. Similarities between sex and B chromosomes are usually found, these elements share many features, for instance, the univalency during meiosis and accumulation of repetitive DNA (Camacho et al. 2000). B chromosomes can also originate from sex chromosomes or vice versa (Camacho et al. 2011). In the rodent species Apodemus flavicollis, for example, the B chromosome originated from the pericentromeric region of sex chromosomes (Rajičić et al. 2017). Furthermore, the use of bioinformatic approaches suggested that the B chromosome of A. latifasciata is composed of degenerated genes and transposable elements from the autosomal complement (Valente et al. 2014; Coan and Martins 2018).

The sequence blocks selected were mapped in the chromosomes of HNA-XY and HNA-XXB by FISH. All of them presented a hybridization signal on the centromeric region of the chromosomes, similar to the one found with the hybridization of the probe SatDNA-HNA-1 on metaphases of HNA-XY and HNA-XXB (Moreira et al. 2023). These results suggested that sequences that compose the centromeric region of H. nanus chromosomes are highly variable with numerous copies. In addition, RT-qPCR analysis showed that these sequences are expressed, indicating a possible role in the genome. Centromeric regions are usually composed of satellite DNA, which represents the majority of repetitive DNA sequences of eukaryotic genomes (Plohl et al. 2012; Biscotti et al. 2015). Transcripts of satellite DNA are usually associated with centromeric functions, kinetochore formation, or chromosome segregation (Plohl et al. 2012; Biscotti et al. 2015). Our findings pointed out that the role of repetitive DNA in the Neotropical rodent species is far from being understood, playing a role in the sex and B chromosomes of these species.

Finally, there are still many gaps about the widespread occurrence of B chromosomes in rodent species, in the present report, we shed light over this question. The B chromosome of H. nanus species seems to be a mosaic of the genomes, probably containing genes and sequences important for its maintenance, in addition to its repetitive DNA fraction. All these features reinforce the role of H. nanus as a model species to understand the huge chromosomal radiation of Neotropical rodents.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1