From germline genome to highly fragmented somatic genome: genome-wide DNA rearrangement during the sexual process in ciliated protists

Abstract

Genomes are incredibly dynamic within diverse eukaryotes and programmed genome rearrangements (PGR) play important roles in generating genomic diversity. However, genomes and chromosomes in metazoans are usually large in size which prevents our understanding of the origin and evolution of PGR. To expand our knowledge of genomic diversity and the evolutionary origin of complex genome rearrangements, we focus on ciliated protists (ciliates). Ciliates are single-celled eukaryotes with highly fragmented somatic chromosomes and massively scrambled germline genomes. PGR in ciliates occurs extensively by removing massive amounts of repetitive and selfish DNA elements found in the silent germline genome during development of the somatic genome. We report the partial germline genomes of two spirotrich ciliate species, namely Strombidium cf. sulcatum and Halteria grandinella, along with the most compact and highly fragmented somatic genome for S. cf. sulcatum. We provide the first insights into the genome rearrangements of these two species and compare these features with those of other ciliates. Our analyses reveal: (1) DNA sequence loss through evolution and during PGR in S. cf. sulcatum has combined to produce the most compact and efficient nanochromosomes observed to date; (2) the compact, transcriptome-like somatic genome in both species results from extensive removal of a relatively large number of shorter germline-specific DNA sequences; (3) long chromosome breakage site motifs are duplicated and retained in the somatic genome, revealing a complex model of chromosome fragmentation in spirotrichs; (4) gene scrambling and alternative processing are found throughout the core spirotrichs, offering unique opportunities to increase genetic diversity and regulation in this group.

Supplementary information: The online version contains supplementary material available at 10.1007/s42995-023-00213-x.

Keywords: Alternative processing; Ciliates; Gene scrambling; Genome rearrangement; Germline genome; Somatic genome.

Fig. 1

Macronucleus (MAC) genome assembly and features of chromosomes and introns of Strombidium cf. sulcatum. A The schema illustrates the canonical structure of nanochromosomes in the MAC of S. cf. sulcatum. B The size distribution of contigs with 0, 1 or 2 telomeres in S. cf. sulcatum. C Correlation between the GC-content (%) and mean base depth of each contig. X-axis shows the GC-content (%) and Y-axis shows the mean base depth. Each dot represents one contig. Scaffolds with either 0/1/2 telomeres share a similar GC-distribution. D The size distribution of 6240 identified introns. E The base composition of 10 nt at both ends of each intron. F Nanochromosomes with complete genes were assessed for gene content; only 3.7% of nanochromosomes contain two genes and 0.7% contain three or more genes. G The proportion of cis-nanochromosomes (two genes located on the same DNA strand) and trans-nanochromosomes (two genes located on opposite strands)

Fig. 2

Homologous genes of Halteria grandinella, Strombidium cf. sulcatum, Oxytricha trifallax and Euplotes vannus. A Homologous clusters of the four species. The digits indicate the number of homologous clusters. B Number of homologous genes and clusters in each combination between the four species. Blocks on the left represent the quantity of genes, the darker the blocks the higher the number of genes. Bar graph refers to the proportion of genes from species indicated by the right blocks. C Pairwise comparison of genes among the four species. Each block indicates the number of homologous genes. The digits indicate the percentage of homologous genes in species with an asterisk (*) labeled. D Morphology of the four species involved in this study. Arrows indicate micronuclei

Fig. 3

Chromosome breakage site (CBS) models of Halteria grandinella, Strombidium cf. sulcatum, Oxytricha trifallax and Euplotes vannus. A The cartoon shows models for CBS retention (top) or elimination (bottom). “m” and “n” denote the ends of two adjacent MAC chromosomes, corresponding to the breakage points “m” and “n” in the MIC genome. The CBSs are retained in somatic chromosomes if n–m < 0, while they are deleted in the case of n–m > 0. B–E The putative CBS size distribution and base composition of CBSs in the most abundant size and flanked sequences in the four species. Dashed boxes in (B–E) denote CBSs (B–E) and nearby consensus elements (B–D)

Fig. 4

The nucleotide bias among somatic nanochromosomes and subtelomeric regions. A–D The top panels of each species show the nucleotide bias of nanochromosomes. All contigs capped with 2-telomeres were split into 250 bins after the removal of telomere sequences. The nucleotide composition of each bin was calculated and is shown as a line chart. AT-rich regions were found at both ends of the chromosomes. The terminal 100 bp of the subtelomeric regions were extracted to count the AT-content using a sliding window of 10 bp and step size of 1 bp, shown as heatmaps in the middle panels. Each row of the heatmap reflects one chromosome. The AT-content of each window is represented by the color range from warm (red) to cold (blue). Seqlogos in the bottom panels show the conserved bases of the subtelomeric region. A distinct boundary (blue arrows in A, B) in the heatmaps indicates the highly-conserved sites at the 8th to 10th nucleotides (A) and 18th to 22nd nucleotides (B) from both ends of the vast majority of nanochromosomes

Fig. 5

Comparison of genome rearrangements among Halteria grandinella, Strombidium cf. sulcatum, Oxytricha trifallax and Euplotes vannus. A, C, D The size distribution of scrambled and unscrambled MDSs, IESs and pointers of the four species. Data in (A, C) are displayed in two scales. The seqlogos in (D) refer to sequence motifs of the most abundant pointers in each species. B The distribution of MDS density of each MAC contig in the four species. MPKA: MDS count per kilo bases of a MAC contig

Fig. 6

Scrambled genome rearrangement patterns in Halteria grandinella, Strombidium cf. sulcatum, Oxytricha trifallax and Euplotes vannus. A The most common pattern of unscrambled gene rearrangement. Colorful blocks indicate the MDSs, grey boxes represent the IESs. B–D Scrambled gene rearrangement models. The MDSs in (B) are in different orientations and the ones in (C) are in reverse order. The MIC-1 and MIC-2 in (D) represent different MIC chromosomes, or may be interrupted as a result of insufficient sequencing and is regarded as a case of (C) in the downstream analysis. E The number of scrambled chromosomes as shown in (B–D). “Concurrent” refers to chromosomes that follow either the (B) or C/D pattern. F The patterns of alternative processing events. A-type: the same MDSs in the micronuclear DNA are processed into multiple distinct somatic chromosomes. B-type: the MDS from one chromosome is nested by, or overlaps with, the IES from another chromosome. G The proportion of alternatively processed genes involving A-type and B-type. H An example case of alternative fragmentation in Euplotes vannus. The MDSs on one MIC contig are shared by five MAC contigs (MAC 2–5 contained both telomeres; MAC-1 had only one telomere. MDS2 is nested by MDS3). I The rearrangement patterns of chromosomes encoding homologs of the single-copy gene “Calcineurin-like phosphoesterase” in the four species. Other areas of the MIC are represented by short lines that are not to scale