The DNA sequence of chromosome I of an African trypanosome: gene content, chromosome organisation, recombination and polymorphism

Abstract

The African trypanosome, Trypanosoma brucei, causes sleeping sickness in humans in sub-Saharan Africa. Here we report the sequence and analysis of the 1.1 Mb chromosome I, which encodes approximately 400 predicted genes organised into directional clusters, of which more than 100 are located in the largest cluster of 250 kb. A 160-kb region consists primarily of three gene families of unknown function, one of which contains a hotspot for retroelement insertion. We also identify five novel gene families. Indeed, almost 20% of predicted genes are members of families. In some cases, tandemly arrayed genes are 99-100% identical, suggesting an active process of amplification and gene conversion. One end of the chromosome consists of a putative bloodstream-form variant surface glycoprotein (VSG) gene expression site that appears truncated and degenerate. The other chromosome end carries VSG and expression site-associated genes and pseudogenes over 50 kb of subtelomeric sequence where, unusually, the telomere-proximal VSG gene is oriented away from the telomere. Our analysis includes the cataloguing of minor genetic variations between the chromosome I homologues and an estimate of crossing-over frequency during genetic exchange. Genetic polymorphisms are exceptionally rare in sequences located within and around the strand-switches between several gene clusters.

Figure 1

Annotated features of the T.brucei chromosome I DNA sequence. The map shows the position and orientation of genes, putative CDSs, pseudogenes, DNA repeats and defined promoter sequences. All short CDSs (<150 codons) for which there is additional evidence of transcription (such as conservation in other species) are shown (see text). Coloured boxes above the central line represent genes oriented 5′→3′ left to right, and below in the reverse direction. The CDSs are coloured with respect to GO process categories: hypothetical proteins are predicted from DNA sequence alone, while proteins labelled ‘process unknown’ have homology to other proteins in T.brucei and/or other species but their biological process has not been defined.

Figure 2

Chromosome I structure with polymorphism and crossover frequency. The compressed chrI map in (C) shows the position and orientation of the gene clusters illustrated in greater detail in Figure 1. (A) Histogram showing the distribution of microsatellites containing >10 repeats in Tb927 chrI. (B) Histogram showing the distribution of SSPs. In (A) and (B), the width of each bar corresponds to 10 kb. (C) The distribution of directional gene clusters on chrI. Clusters are boxed to reveal those transcribed from left to right (above the line) and from right to left (below the line), and the strand-switch regions between them. Individual CDSs observed on the opposite strand to large clusters are either pseudogenes or annotated as hypothetical proteins, and may not be CDS (see Fig. 1). Clusters of RHS and LRRP CDSs are boxed with a dotted line and exclude two intact CDSs (LRRP1 and 1.380). (D) The position of polymorphic markers used to generate the genetic map. (E) A genetic crossover map of Tb927 chrI, created by analysing hybrid progeny from a cross between Tb927 and Tb247, showing the physical and genetic distances between markers (kb and cM respectively) and the physical size of the recombination unit (kb/cM).

Figure 3

Repeated DNA length polymorphism within and between cloned stocks of T.brucei. Columns 1, 4, 7, 10 = Tb927; columns 2, 5, 8, 11 = Tb247; columns 3, 6, 9, 12 = Tb427. L = 100 bp ladder. (A) Restriction fragments containing tandemly arrayed DNA sequences. Top left, tubulin gene array (unit size 3641 bp): lanes 1–3, HpaI/NcoI; lanes 4–6, SspI/XmnI, fragment lengths in Tb927chrIa sequence file 16 235 and 16 247 bp, respectively (but not represented in full, see Fig. 1), length polymorphism between stocks ∼35 kb. Top right, histone H3 gene array (unit size 1584 bp): lanes 1–3, Sau3AI; lanes 4–6, PstI/EcoRI, fragment lengths in sequence file 11 469 and 11 664 bp, respectively, length polymorphism ∼30 kb. Bottom left, Family 15 tandem array (unit size 3100 bp): lanes 1–3, SspI; lanes 4–6, BglII, fragment lengths in sequence file 18 171 and 26 134 bp, length polymorphism ∼15 kb. Bottom right, 123-bp repeat array in Tb927.1.1740: lanes 1–3, HindIII/SspI; lanes 4–6, XmnI, fragment lengths in sequence file 20 838 and 23 880 bp, length polymorphism ∼15 kb. (B) Length polymorphism of PCR amplification products containing microsatellite. Left to right: msat D2 in Tb.1.1780 and msats E5, 18 and 15 in intergenic sequence.