Search

A main determinant of prolonged Trypanosoma brucei infection and transmission and success of the parasite is the interplay between host acquired immunity and antigenic variation of the parasite variant surface glycoprotein (VSG) coat. About 0.1% of trypanosome divisions produce a switch to a different VSG through differential expression of an archive of hundreds of silent VSG genes and pseudogenes, but the patterns and extent of the trypanosome diversity phenotype, particularly in chronic infection, are unclear. We applied longitudinal VSG cDNA sequencing to estimate variant richness and test whether pseudogenes contribute to antigenic variation. We show that individual growth peaks can contain at least 15 distinct variants, are estimated computationally to comprise many more, and that antigenically distinct 'mosaic' VSGs arise from segmental gene conversion between donor VSG genes or pseudogenes. The potential for trypanosome antigenic variation is probably much greater than VSG archive size; mosaic VSGs are core to antigenic variation and chronic infection.

Patterns of genetic diversity in parasite antigen gene families hold important information about their potential to generate antigenic variation within and between hosts. The evolution of such gene families is typically driven by gene duplication, followed by point mutation and gene conversion. There is great interest in estimating the rates of these processes from molecular sequences for understanding the evolution of the pathogen and its significance for infection processes. In this study, a series of models are constructed to investigate hypotheses about the nucleotide diversity patterns between closely related gene sequences from the antigen gene archive of the African trypanosome, the protozoan parasite causative of human sleeping sickness in Equatorial Africa. We use a hidden Markov model approach to identify two scales of diversification: clustering of sequence mismatches, a putative indicator of gene conversion events with other lower-identity donor genes in the archive, and at a sparser scale, isolated mismatches, likely arising from independent point mutations. In addition to quantifying the respective probabilities of occurrence of these two processes, our approach yields estimates for the gene conversion tract length distribution and the average diversity contributed locally by conversion events. Model fitting is conducted using a Bayesian framework. We find that diversifying gene conversion events with lower-identity partners occur at least five times less frequently than point mutations on variant surface glycoprotein (VSG) pairs, and the average imported conversion tract is between 14 and 25 nucleotides long. However, because of the high diversity introduced by gene conversion, the two processes have almost equal impact on the per-nucleotide rate of sequence diversification between VSG subfamily members. We are able to disentangle the most likely locations of point mutations and conversions on each aligned gene pair.

Pathogens often persist during infection because of antigenic variation in which they evade immunity by switching between distinct surface antigen variants. A central question is how ordered appearance of variants, an important determinant of chronicity, is achieved. Theories suggest that it results directly from a complex pattern of transition connectivity between variants or indirectly from effects such as immune cross-reactivity or differential variant growth rates. Using a mathematical model based only on known infection variables, we show that order in trypanosome infections can be explained more parsimoniously by a simpler combination of two key parasite-intrinsic factors: differential activation rates of parasite variant surface glycoprotein (VSG) genes and density-dependent parasite differentiation. The model outcomes concur with empirical evidence that several variants are expressed simultaneously and that parasitaemia peaks correlate with VSG genes within distinct activation probability groups. Our findings provide a possible explanation for the enormity of the recently sequenced VSG silent archive and have important implications for field transmission.

African trypanosomes evade humoral immunity through antigenic variation whereby, they switch expression of the variant surface glycoprotein (VSG) gene encoding their glycoprotein surface coat. Switching proceeds by duplication from an archive of silent VSG genes into a transcriptionally active locus, and precedent suggests silent genes can contribute, combinatorially to formation of expressed, functional genes through segmental gene conversion. The genome project has revealed that most of the silent archive consists of hundreds of VSG genes in subtelomeric tandem arrays, and that most of these are not functional genes. The aim of this review is to explore links between the uncovered trypanosome genotype and the phenotype of antigenic variation, stretching from the broad phenotype-transmission in the field and the overcoming of herd immunity-to events within single infections. Highlighting in particular the possible impact of phenotype selection on the evolution of the VSG archive and the mechanisms for its expression leads to a theoretical framework to further our understanding of this complex immune evasion strategy.

African trypanosomes evade humoral immunity through antigenic variation, whereby they switch expression of the gene encoding their VSG (variant surface glycoprotein) coat. Switching proceeds by duplication of silent VSG genes into a transcriptionally active locus. The genome project has revealed that most of the silent archive consists of hundreds of subtelomeric VSG tandem arrays, and that most of these are not functional genes. Precedent suggests that they can contribute combinatorially to the formation of expressed, functional genes through segmental gene conversion. These findings from the genome project have major implications for evolution of the VSG archive and for transmission of the parasite in the field.

Antigenic variation in African trypanosomes displays a degree of order that is usually described as 'semi-predictable' but which has not been analysed in statistical detail. It has been proposed that, during switching, the variable antigen type (VAT) being inactivated can influence which VAT is subsequently activated. Antigenic variation proceeds by the differential activation of members of the large archive of distinct variable surface glycoprotein (VSG) genes. The most popular model for ordered expression of VATs invokes differential activation probabilities for individual VSG genes, dictated in part by which of the four types of genetic locus they occupy. We have shown, in pilot experiments in cattle, correlation between the timing of appearance of VSG-specific mRNA and of lytic antibodies corresponding to seven VSGs encoded by single-copy genes. We have then determined the times of appearance of VAT-specific antibodies, as a measure of appearance of the VATs, in a statistically significant number of mouse infections (n=22). There is a surprisingly high degree of order in temporal appearance of the VATs, indicating that antigenic variation proceeds through order in the probability of activation of each VAT. In addition, for the few examples of each available, the locus type inhabited by the silent 'donor' VSG plays a significant role in determination of order. We have analysed in detail previously published data on VATs appearing in first relapse peaks, and find that the variant being switched off does not influence which one is being switched on. This differs from what has been reported for Plasmodium falciparum var antigenic variation. All these features of trypanosome antigenic variation can be explained by a one-step model in which, following an initial deactivation event, the switch process and the imposition of order early in infection arise from the inherent activation probabilities of the specific VSG being switched on.

Antigenic variation in African trypanosomes proceeds by switching between different variant surface glycoprotein (VSG) molecules, whose extensive epitope differences enable evasion of antibody responses. Each trypanosome has approximately 1000 basic copy VSG genes inside chromosomes and a subset located at telomeres. Switching usually involves different individual basic copy genes being duplicated, as an expression linked copy, into a transcriptionally active site. In a few cases expression linked copies with a number of point mutations have been observed, leading to the suggestion that point mutagenesis provides another mechanism of antigenic variation. The most extensive example is a VSG gene that is normally activated in the metacyclic population in the tsetse fly, but the point mutations were detected in expression linked copies generated during bloodstream infection, after prolonged growth and selection. It was suggested that particularly telomeric or metacyclic VSG genes might undergo point mutagenesis during expression linked copy formation. To test this we have cloned 3 trypanosomes very soon after they had generated, during mouse infection, expression linked copies of the metacyclic VSG gene ILTat 1.22 and have detected only a single point mutation which is present in one expression linked copy, but not the corresponding basic copy, gene. This mutation does not prevent binding of a neutralizing antibody. Extensive VSG gene point mutagenesis may be a consequence merely of prolonged growth and extensive selection. There is not a single reported case of a point mutated VSG presenting a completely new set of exposed epitopes, suggesting point mutagenesis is unlikely to be an authentic mechanism for antigenic variation.

Procyclic and epimastigote forms of Trypanosoma congolense express an immunodominant glutamic acid/alanine-rich protein (GARP) that covers the parasite surface. Although GARP shows no sequence similarity to procyclins from T. brucei, the general characteristics of the two sets of surface glycoproteins suggest that they have analogous functions, in much the same way that variant surface glycoproteins with unrelated primary sequences fulfil the same function in bloodstream form trypanosomes. Since T. brucei and T. congolense do not follow the same pathway through the tsetse fly, one possible function of procyclins might be to direct parasites to the correct compartments. As a first step towards testing this hypothesis, we have produced stably transformed procyclic forms of T. brucei in which the GARP coding region has been integrated into a procyclin expression site. GARP can be detected on the surface of these transgenic trypanosomes, uniformly distributed within the endogenous procyclin coat, but there are differences in post-translational modification when it is expressed in T. brucei rather than in T. congolense. The fact that GARP is readily accessible to antibodies which were raised against a bacterial fusion protein led us to examine its potential as a selectable surface marker for transfection. We have established a rapid and simple procedure for isolating stable transformants that provides an alternative to conventional methods of selection for antibiotic resistance.

Five monoclonal antibodies (mAb) were raised that bound to the surface of procyclic stage Trypanosoma congolense with high intensity in immunofluorescence. Immunoblot analysis of trypanosome lysates using 3 of these mAb revealed a diffuse SDS-PAGE band of 36-40 kDa. The purified antigen did not react with Coomassie Blue or silver stains, but did stain blue with Stains-all, indicating acidity. For the one mAb tested, the epitope was periodate-sensitive and therefore probably glycan. Although this antigen shares properties with procyclin/PARP, which forms a surface coat on procyclic Trypanosoma brucei, a search in T. congolense for homologues of a procyclin/PARP gene revealed only non-coding sequence of partial similarity. Using a differential screen, a procyclic stage T. congolense cDNA clone was isolated that encoded a putative 256-amino acid protein containing 2 peptides chemically sequenced independently by Beecroft et al. [36]. The protein, termed glutamate and alanine-rich protein (GARP), has potential hydrophobic leader and tail sequences (the latter with potential for replacement by a glycosyl phosphoinositol anchor) and no potential N-linked glycosylation sites. It has no significant sequence homology with known proteins. Antibodies against a translational fusion of GARP bound specifically in Western blots to a band very similar to that detected by the mAb and also to the purified antigen. Immunogold electron microscopy revealed a dense packing of the antigen on the cell surface. It appears that procyclic T. brucei and T. congolense have major surface proteins with structural analogy, but with no sequence homology.

The Trypanosoma brucei metacyclic population in the salivary glands of the tsetse fly displays a characteristic set of variable antigen types (VATs) which represents only a restricted part of the parasite's total VAT repertoire. After introduction into the mammalian host by fly bite, the metacyclics transform into bloodstream forms which retain expression of the metacyclic VATs. Specific antibodies, both polyvalent and monoclonal, have been used to neutralize separately 4 individual VATs from metacyclic populations. Control experiments and visual observation confirmed lysis of each VAT. On injection of the surviving trypanosomes, after washing, into mice each neutralized VAT was nevertheless expressed within a few days. Simultaneous neutralization of 2 metacyclic VATs which usually switch to one another in bloodstream infections did not prevent expression of either on subsequent injection into mice. Expression of neutralized VATs was not influenced by the antigenic composition of the population originally ingested by the tsetse fly. Metacyclic forms and their immediate successors thus appear to switch rapidly to expression of other metacyclic VATs in bloodstream populations.

A group of 27 variable antigen type (VAT)-specific monoclonal antibodies (McAbs) have been made against metacyclic forms of a cloned stock of Trypanosoma brucei rhodesiense. In combination, these labelled in immunofluorescence 99.3% of trypanosomes in salivary probes from tsetse flies. The 0.7% of unlabelled trypanosomes were believed to be uncoated forms. The ability of a mixture of antibodies to kill metacyclics in vitro by complement-mediated lysis, thus neutralizing their infectivity for mice, was tested. The antibody mixture consisted of 24 McAbs plus 3 VAT-specific rabbit antisera. In 12 replicate experiments this mixture of antibodies prevented infection of mice. Parallel controls showed that neutralization was probably antibody-mediated and VAT specific. However, we have not been able to repeat these results on a long-term basis; this may be due to a loss of neutralizing activity by one of the McAbs. The successful neutralization experiments indicate that the number of VATs in the metacyclic repertoire of one stock of T. b. rhodesiense is limited to at most 27.

The sequence of appearance of specific lytic activity against more than 20 variable antigen types (VATs) of Trypanosoma vivax in the serum of 27 animals belonging to 5 species has been examined. For each host species there was a characteristic course of infection, with differences in height and duration of parasitaemia and in pathogenicity. The sequence of antigenic variation was similar in all host species, with some VATs consistently eliciting response more rapidly than others. The predominant group, comprising VATs which apparently developed within the first 3 weeks, varied in size according to the total number of trypanosomes in the bloodstream within that period, suggesting there is a spectrum, rather than discrete groupings, in the hierarchy of VAT expression. There was very little evidence for differences in appearance of VATs between host species; the only clear example was one VAT which apparently did not develop in one host species. The sequence of antigenic variation in T. vivax seems to be determined by the parasite rather than the host species.

The variable antigen repertoire expressed by metacyclic Trypanosoma brucei rhodesiense is not influenced by the anamnestic expression whereby the variable antigen type (VAT) ingested by a tsetse fly is present at high levels in early bloodstream populations of fly-infected mice. This has been demonstrated by feeding to Glossina morsitans a trypanosome line expressing a VAT which is normally a component of the metacyclic repertoire. The VAT did not constitute a significantly increased proportion of the resultant metacyclic population which would have occurred had anamnestic expression and metacyclic expression been linked. Five other metacyclic VATs were also present at control levels. We conclude that the mechanisms of expression of VATs in the fly and in the mammal are independently controlled.

Rates at which Trypanosoma brucei rhodesiense trypanosomes switch from expression of one variable antigen type (VAT) to that of another have been determined in cloned populations that have been recently tsetse-fly transmitted. Switching rates have been determined between several, specific pairs of VATs in each population. High rates of switching were observed in 2 cloned trypanosome lines, each derived from a separate cyclical transmission of the same parental stock and each expressing a different major VAT. Five estimates of the switching rate between one particular pair of VATs were consistently high (approximately 1 x 10(-3) switches/cell/generation). These high switching rates were similar both in bloodstream populations of mice and in populations confined to subcutaneously implanted growth chambers in mice, thus indicating that the interaction of the bloodstream population with other trypanosome populations in the lymphatics or extravascular sites in systemic infections did not influence the estimates of the rate of switching. Fourteen estimates were made of VAT-specific switching rates in bloodstream infections involving 8 combinations from among 6 VATs. Switching rate estimates were VAT-specific and showed considerable variation between different combinations of VATs--from 1.9 x 10(-6) to 6.9 x 10(-3) switches/cell/generation. The rates of switching to different metacyclic-VATs were, however, very similar. Summation of between 3 and 5 VAT-specific switching rate values in each of 4 experiments conducted in bloodstream infections has provided minimum estimates of the overall rate of antigenic variation: 2.0-9.3 x 10(-3) switches/cell/generation. These values are between 20 and 66,000-fold higher than previously published estimates.(ABSTRACT TRUNCATED AT 250 WORDS)

A pleomorphic line of Trypanosoma brucei rhodesiense expressing a single variable antigen was used to quantify the rate of loss of the surface coat from bloodstream forms transforming to procyclics in the tsetse fly, Glossina morsitans, and in in vitro culture. Loss of variable antigen occurred at similar rates in the crop and anterior portion of the midgut of tsetse flies and in in vitro culture, but in the posterior portion of the fly midgut it occurred 2-3 times faster. The posterior portion of the midgut is the most important site for transformation of bloodstream-form trypanosomes to procyclics, and the dynamics of at least one component of this process are therefore not accurately paralleled in vitro.

Trypanosoma brucei contains a repertoire of more than 100 different genes for Variant Surface Glycoproteins (VSGs). A small and strain-specific fraction of these genes is expressed in the salivary glands of the tsetse fly (M-genes), giving rise to metacyclic Variable Antigen Types (M-VATs). Antibodies produced in a chronic trypanosome infection initiated by syringe inoculation of bloodstream forms into mammals (i.e. against B-VATs), will react with most of the M-VATs suggesting that these B-VATs express VSG genes that are similar or identical to M-genes. We have cloned DNA complementary to the VSG mRNA of four of such B-VATs and used this to characterize the corresponding VSG genes. In three of the four VATs we find a single VSG gene hybridizing with the cDNA probe and we provide supporting evidence that this gene is expressed as an M-gene. In the bloodstream repertoire these genes appear to be activated by duplicative translocation to another telomere. In all four variants the putative M-genes are telomeric and in the three cases where the location of the genes on chromosome-sized DNA molecules could be determined, the genes were located in large DNA, whereas the majority of the telomeric VSG genes are in chromosomes less than 1000 kb. Our results are best explained by models for M-gene activation involving telomeric expression sites for these genes which are separate from those used by bloodstream forms. The implications of these results for vaccination are discussed.

Trypanosoma brucei rhodesiense subpopulations in different sites within the body of infected mice were isolated and enumerated on day 6 of cyclically transmitted infections. Most trypanosomes were in the blood vasculature and spleen but approximately 6% occurred in lymph nodes and about 9% were extravascular. Most of the extravascular trypanosomes were in the peritoneal and pleural cavities; significant numbers also occurred in the brain and kidneys. Six major variable antigen types (VATs) were detected by immunofluorescence using specific antisera and monoclonal antibodies. The prevalence of each VAT was essentially the same in subpopulations in the blood, mesenteric and inguinal lymph nodes, brain, kidneys and peritoneal and pleural cavities. This similarity of VAT composition in different subpopulations is probably caused by high rates of dynamic interchange of trypanosomes between sites. Extravascular trypanosomes, therefore, form a significant proportion of the total population in acute infections of mice but they do not appear to play any special role in the population biology of antigenic variation at this stage of infection.