Chapter 19 : DNA Recombination Strategies During Antigenic Variation in the African Trypanosome

Architecture and singular transcription of variant surface glycoprotein (VSG) gene expression sites in Trypanosoma brucei. The four line diagrams show cartoon representations of telomeric VSG expression sites. The top diagram shows a generic bloodstream expression site (BES), while the two diagrams below display examples of variant BES ( 48 ) in which VSG pseudogenes (ψ, peach box) are found (BES 14) or where there has been loss of several expression site associated genes (ESAGs; dark blue box) or pseudogenes (light blue box) (BES 10). The final line diagram shows a VSG expression site (MES) used in metacyclic form T. brucei, which are found in the tsetse; here, the RNA polymerase I (Pol I) promoter (flag) does not drive expression of ESAGs, as it does in the BES, but only the VSG (red box), which in all cases is found adjacent to the telomere (telo; vertical line). Upstream of the MES promoter, several ESAG pseudogenes have been described, suggesting that these sites were derived from the BES. Arrays of 70-bp DNA repeats in the BES and MES are shown (hatched box), which always appear to be upstream of VSG genes or pseudogenes. Only one BES or MES is actively transcribed at a time in a single cell. A bloodstream form T. brucei cell is shown, in which the nucleus is diagrammed. The single active BES (red, extended arrow denotes transcription) is shown associated with the expression site body (ESB, small green circle), which is spatially distinct from the nucleolus (large green circle), though both subnuclear structures are sites of RNA Pol I transcription. Silent BES (three are shown in black; truncated arrow denotes limited transcription) do not associate with the ESB or nucleolus.

The variant surface glycoprotein (VSG) gene archive in Trypanosoma brucei. Whole T. brucei chromosomes are shown separated by pulsed field gel electrophoresis and stained with ethidium bromide. To the left of the gel, the positions of the megabase chromosomes, intermediate chromosomes and minichromosomes that comprise the nuclear genome are indicated, including the size and number of the different chromosome classes. To the right of the gel, the different loci in which VSGs are found are indicated (bloodstream expression site (BES), mini, array), including the number of VSGs in each locus type and whether they are functional (intact, red box) or are pseudogenic (ψ, peach box). BES denotes VSGs in expression sites that are used in the mammalian bloodstream and are found in the megabase and intermediate chromosomes. Mini denotes VSGs found in the minichromosomes, and array denotes VSGs found in the subtelomeres of the megabase chromosomes. In each case the presence or absence of a number of sequence features in addition to the VSG is shown: the telomere (vertical line), 70-bp repeats (widely hatched box), expression site-associated genes (black box), the RNA Pol I promoter (arrow) and 177-bp repeats (narrow hatched box).

Hierarchy of variant surface glycoprotein (VSG) gene switching by recombination during infections by Trypanosoma brucei. The graph depicts the log of the number of T. brucei cells in a cow up to 70 days after infection (day 0). The schematic below details the timing of activation, by switching, of the different VSGs found in the genome (VSG type): silent telomeric VSGs (telomere) are activated more frequently than intact, subtelomeric array VSGs (array), which in turn are activated more frequently than VSG pseudogenes (pseudo). Gene conversion is the most frequent route for the above activation events, and the features associated with gene conversion of each VSG type are diagrammed. The VSG expressed before a switch (blue box) is transcribed (dotted arrow) from a bloodstream expression site (BES), in which the VSG is adjacent to the telomere (vertical line) and flanked upstream by 70-bp repeats (hatched box) and expression site associated genes (ESAGs; black boxes). The amount of sequence copied during VSG gene conversion is shown. For telomeric VSGs the sequence copied normally encompasses the VSG open reading frame (red box) and extends upstream to the 70-bp repeats, but also can extend further upstream into the ESAGs if the silent VSG is in an inactive BES; the downstream conversion limit may be the end of the VSG, but can also extend to the telomere from either a minichromosome VSG or inactive BES. Gene conversion of an intact subtelomeric array VSG is more limited in the range of sequence copied. In segmental VSG gene conversion parts of multiple, normally nonfunctional VSG pseudogenes (orange, red or brown boxes) are combined to generate a novel mosaic VSG; though this is shown to occur in the BES, it is not known if this is the location of gene assembly. Note also, the VSG pseudogene donors are shown for convenience as a contiguous array; in fact, segmental gene conversions using adjacent genes have never been observed.

Complexity of variant surface glycoprotein (VSG) mosaics formed by segmental gene conversion in Trypanosoma brucei. (A) Where the 3′ boundary of segmental gene conversion occurs within the coding sequence of the VSG (3′ donation), part or all of the previously expressed C-terminal-domain-(CTD) -encoding region of VSG is retained, allowing the expression of a large contingent of silent VSGs (red box) that contain frameshifts or stop codons towards their 3′ ends (frameshift or premature stop codon indicated by an asterisk); as in Fig. 3 , the recipient VSG (blue) is shown in the bloodstream expression site (BES) and the extent of conversion is indicated (NTD denotes N-terminal domain). Donors of VSGs formed in this way were found to share little sequence similarity over their whole sequence. (B) Mosaic VSGs can allow (partial) expression of pseudogene VSGs. Donors of VSGs (pink box) formed in this way share relatively high levels of sequence similarity (73% identity at the nucleotide level). (C) Segmental gene conversion yields diverse products: the diagram shows nine different VSGs detected during chronic infections ( 124 ); different donors are indicated in different colours, with 3′ donors indicated by hatching.

Models for variant surface glycoprotein (VSG) recombination during antigenic variation in Trypanosoma brucei. (A) Recombination is shown initiated by a DNA double-strand break (DSB) in the 70 bp repeats (hatched box) upstream of the VSG (black arrow) in the bloodstream expression site (BES) (ESAGs and promoter are not shown). Only factors that have been examined for a role in VSG switching are indicated; those shown in color have been found to act, while those for whom no evidence of a role in VSG switching has been found are shown in gray. DSB processing to reveal 3′ single-stranded ends is, in part, catalyzed by MRE11-RAD50-XRS2/NBS1 (MRX), generating a substrate on which RAD51 forms a nucleoprotein filament; note, however, that a further exonuclease (not shown) normally acts with MRX of both ends of the DSB are processed. RAD51 function is mediated by a number of factors: BRCA2 influences RAD51 filament dynamics, while the detailed roles of RAD51 paralogs (RAD51-3, RAD51-4, RAD51-5 and RAD51-6 in T. brucei) are unclear. RAD51 catalyzes repair by homology-dependent invasion of the single-stranded end into intact DNA (gray lines), containing a silent VSG (gray arrow). Mismatch repair constrains homologous recombination to act only on sufficiently homologous sequences. Three pathways for DSB repair have been described and may contribute to VSG switching. (B) DSB repair; here, newly synthesized DNA is copied from the intact DNA duplex and remains base-paired, generating Holliday junction structures whose enzymatic resolution can lead to gene conversion with (not shown) or without (shown) crossover of flanking sequence. In T. brucei, RMI1-TOP3 has been shown to suppress crossover, by perhaps acting on the Holliday junctions. (C) Synthesis-dependent strand annealing; here, newly synthesized DNA is displaced from the intact duplex and reanneals with homologous sequence at the DSB, allowing synthesis of the other strand. Break-induced replication is shown in (D); in this mechanism, an origin-independent replication fork forms on the strand invasion intermediate allowing replication to the chromosome end.