Skip to main content
NCBI home page
Molecular and Cellular Biology logoLink to Molecular and Cellular Biology
. 1998 Jan;18(1):198–205. doi: 10.1128/mcb.18.1.198

Frequent Loss of the Active Site during Variant Surface Glycoprotein Expression Site Switching In Vitro in Trypanosoma brucei

Mike Cross 1, Martin C Taylor 1,†, Piet Borst 1,*
PMCID: PMC121476  PMID: 9418867

Abstract

African trypanosomes undergo antigenic variation of their variant surface glycoprotein (VSG) coat to avoid being killed by their mammalian hosts. The active VSG gene is located in one of many telomeric expression sites. Replacement of the VSG gene in the active site or switching between expression sites can give rise to a new VSG coat. To study Trypanosoma brucei VSG expression site inactivation rather than VSG gene switching, it is useful to have an in vitro negative-selection system independent of the VSG. We have achieved this aim by using a viral thymidine kinase (TK) gene. Following integration of the TK gene downstream of the 221a VSG expression site promoter, transformant cell lines became sensitive to the nucleoside analog 1-(2-deoxy-2-fluoro-8-d-arabinofuranosyl)-5-iodouracil. These TK trypanosomes were able to revert to resistance at a rate approaching 10−5 per cell per generation. The majority of revertants expressed a new VSG gene even though there had been no selection against the VSG itself. Analysis of these switched variants showed that some had shut down TK expression via an in situ expression site switch. However, most variants had the complete 221 expression site deleted and another VSG expression site activated. We speculate that a new VSG expression site cannot switch on without inactivation of the old site.

The protozoan parasite Trypanosoma brucei lives within the bloodstream of its mammalian host. In order to avoid destruction by the host immune response, T. brucei periodically changes its variant surface glycoprotein (VSG) coat, a process termed antigenic variation (see references 6 and 10 for recent reviews). The expressed VSG gene is invariably found close to the telomere in a long polycistronic transcription unit called the expression site. There are around 20 VSG expression sites (13, 26) which are highly homologous and include a number of expression site-associated genes (ESAGs) besides the VSG gene (30). Some of the ESAGs encode proteins required for the uptake of host macromolecules, for example, ESAGs 6 and 7, which comprise the heterodimeric transferrin receptor (reviewed in references 6 and 27). Normally, only one expression site is transcribed at a time, giving rise to a single set of ESAGs and a VSG coat composed of a single protein species (for a review, see reference 6). In addition to the expression sites, there are about 1,000 different VSG genes located in large arrays in the interior of the larger chromosomes and at the telomeres of approximately 100 minichromosomes (39).

VSG switching can occur through DNA rearrangements such as gene conversion and reciprocal recombination (see Fig. 4; see reference 2 for a recent review). In the former, a copy of a silent VSG gene replaces the previously active VSG gene. If the donor VSG is part of another expression site, then the gene conversion can include a number of ESAGs as well. A similar outcome can arise from a reciprocal recombination reaction, although in this case there is no duplication or loss of the genes involved. In addition to these mechanisms of switching, the trypanosome can activate a new expression site and silence the old one, a process termed an in situ switch. This is different from the other mechanisms in that it can occur in the absence of any detectable DNA rearrangement (19, 41). How the switch in transcription between VSG expression sites occurs and how all expression sites but one are silenced are not known, although it has been suggested that some form of epigenetic mechanism might be involved (16, 18, 31).

FIG. 4.

FIG. 4

Open in a new tab

Mechanisms of VSG expression site switching applicable to the TK-expressing trypanosomes. In each case, the 221 VSG expression site (thick line) and another VSG expression site, X (thinner line), are shown. Promoters (flags) and the telomeric VSG gene (221, shaded box; X, open box) are indicated. The HYG and TK cassettes located near the promoter in the 221 VSG expression site and active transcription (broken arrow) are also shown. Recombination reactions which switch expression site sequences containing the TK gene can occur either by nonreciprocal gene conversion (thin line [events 1 and 2]) or by reciprocal exchange (cross [events 3 and 4]). Note that the boundaries of these exchanges can theoretically take place upstream of the promoter in the 50-bp repeat array and encompass a variable amount of expression site sequence. The prediction of the type of VSG coat expressed and the genotype of the HYG, TK, and VSG 221 markers in each switch event are also given.

Previous studies of VSG switching have relied heavily upon the use of animals to select the rare VSG switch variants from a population of trypanosomes. With this approach, it is difficult to manipulate the conditions of selection. Quantitative study of factors which may influence VSG switching and switch variant selection is also far from easy. To overcome these limitations, in vitro selection by complement-mediated lysis using antisera raised against the parental VSG type (trypanolysis) has been used (24). However, this procedure has proved unreliable, as trypanosomes trapped in clumps survive the antibody treatment. In addition, methods which select purely against the VSG antigen type are not suitable for the study of VSG expression site switching. This is because a large proportion of switch variants can arise through replacement of the VSG gene alone, leaving the rest of the active expression site unchanged (2).

Following the development of stable-transfection techniques, it has recently become possible to tag an individual VSG expression site with drug resistance genes (5, 18, 31). This allows positive selection for expression of that site in liquid culture, providing the opportunity to examine specific expression site activation events and to monitor consecutive on-off-on switches (19, 26, 32). However, this type of experiment does not simulate the situation in vivo, where immune attack against the parent population selects for a switch in expression to any one of the 20 VSG expression sites the trypanosome has to offer. To this end, we set about testing a negative-selection system for T. brucei, making use of the thymidine kinase (TK)-thymidylate kinase gene from herpes simplex virus type 1 (HSV-1). This enzyme can phosphorylate a wide variety of nucleoside analogs which subsequently act as competitive inhibitors of DNA polymerase or as DNA chain terminators, leading to cell death. We and others have previously demonstrated that TK can be used as a negatively selectable marker in the procyclic insect stage of T. brucei (22, 36). Here, we report the use of TK in the bloodstream stage of the parasite to successfully select in vitro trypanosomes which had switched expression of the VSG expression site.

MATERIALS AND METHODS

Trypanosomes, plasmid constructs, and transformation.

T. brucei 221a bloodstream-form trypanosomes (MiTat 1.2a) of strain 427 (11) were used and were grown in vitro at 37°C in HMI-9 culture medium (17) with the following modifications: no Serum Plus or extra thymidine was added, and 20% instead of 10% fetal bovine serum was used. Growth of 221a trypanosomes under these conditions appeared no different from growth in standard HMI-9 medium. For growth assays, trypanosomes were seeded at 5 × 104 cells/ml in the presence or absence of 20 μg of 1-(2-deoxy-2-fluoro-8-d-arabinofuranosyl)-5-iodouracil (FIAU).

Plasmid construct p221-TK was derived from plasmid enES-2 (31) and carries the hygromycin phosphotransferase gene (HYG) and the HSV-1 TK gene (see Fig. 1). It contains the following sequences inserted in the polylinker of the pBluescript KS(+) vector (5′→3′): the 190-bp SalI-SphI fragment from the 221 VSG expression site, the 240-bp βα intergenic region of the tubulin array as a splice acceptor site, the HYG gene, the 402-bp intergenic region between actin genes 1 and 2 as a polyadenylation-splice acceptor site, the HSV-1 TK gene, the 330-bp αβ intergenic region of the tubulin array as a polyadenylation site, and the 685-bp SphI-StuI fragment from the 221 VSG expression site. A detailed description of how this plasmid was constructed can be obtained from the authors upon request.

FIG. 1.

FIG. 1

Open in a new tab

Integration of the HSV-1 TK gene into the 221 VSG expression site. A schematic drawing of a telomeric VSG expression site, modelled after that of Revelard et al. (30), is shown, indicating the site of integration of the construct p221-TK (enlarged section). The ESAGs (numbered open boxes), the VSG gene (shaded box), repeat arrays (striped boxes), the VSG expression site promoter (flag), and the telomere repeats (triangles) are indicated. Construct p221-TK contains flanking sequences (solid black bars) derived from the 221a expression site which act as homology regions to allow recombination into the expression site. The hygromycin phosphotransferase (HYG) and HSV-1 TK cassettes are indicated (see Materials and Methods for details). Restriction enzyme sites are as follows: S, SalI; Sp, SphI; and St, StuI.

Trypanosomes were transformed with 5 μg of NotI-XhoI-digested plasmid p221-TK DNA as described previously (23). Transformants were selected for resistance to 2.5 μg of hygromycin B per ml. Correct integration of the p221-TK construct into the 221 VSG expression site was determined by standard molecular biological techniques (34). Two cell lines shown to contain the correct integration, called HTK3 and HTK16, were used for further analysis.

Negative selection of HTK trypanosomes for VSG switch variants.

HTK trypanosomes were routinely cultivated in the presence of 20 μg of hygromycin per ml. For switching experiments, the cells were washed in medium containing no hygromycin and used to inoculate a fresh culture, again containing no hygromycin, at a density of 2.5 × 103 to 5 × 103 cells/ml. Cells were harvested when the culture had grown to 1 × 106 to 2 × 106 cells/ml and were distributed over 96-well plates at 104/well in 100 μl of medium containing 20 μg of FIAU per ml. After 6 or 7 days, clonal outgrowth of wells was scored and FIAUr trypanosome cell lines were tested for sensitivity to 20 μg of hygromycin per ml. Cell lines displaying a FIAUr Hygs phenotype were then immediately expanded in vitro for preparation of crude cell lysates and DNA (see below).

Luria-Delbrück fluctuation test.

The rate of reversion to FIAU resistance was measured with the Luria-Delbrück fluctuation test (21). HTK3 cells were distributed over two 96-well plates at 10/well in 200 μl of culture medium. After 5 days of growth, cells in the 48 central wells from each plate were counted and taken for further analysis. Replica cultures in which the contents of the wells were transferred to 2 ml of fresh medium supplemented with 20 μg of FIAU per ml with or without hygromycin (20 μg/ml) in 24-well plates were set up. Outgrowth of wells was scored after 7 days. The frequency of reversion (a) was then assessed by substituting experimental results into the equation a = (−lnP0 · ln2)/N, where P0 is the proportion of wells without cell growth and N is the number of trypanosomes per culture upon addition of FIAU.

Analysis of VSG switch variants.

Crude cell lysates of HTK cell lines selected for the FIAUr Hygs phenotype were prepared from 1 × 106 to 2 × 106 cells which had been washed twice with PSG (59.4 mM Na2HPO4, 3.1 mM NaH2PO4, 43.2 mM NaCl, 55.5 mM glucose [pH 8]), resuspended in 10 μl of sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer, and boiled for 5 min. Proteins were separated on an 8% polyacrylamide–SDS gel and visualized by staining with Coomassie brilliant blue. Small-scale genomic DNA preparations and dot blot hybridization were done as described elsewhere (23). The probes used were the coding regions of the HYG and TK genes and a 590-bp PstI fragment of pTcV221.5 specific for the 221 VSG gene (3). To control for DNA loading, a fragment containing the calmodulin intergenic repeat region was used (23).

Pulsed-field gel electrophoresis.

Chromosome separations were performed by contour-clamped homogeneous electric field (CHEF) electrophoresis with a Bio-Rad CHEF-DR II system. The gel (0.8% FMC Fastlane agarose) was run for 72 h at 12°C at 2 V/cm with a switching time of 900 s. TAFE buffer (1×) (10 mM Tris, 0.5 mM EDTA-free acid [Titriplex II], 4.4 mM glacial acetic acid) was used and was changed once halfway through the run. Each gel lane contained 1.25 × 107 trypanosomes embedded in low-melting-point agarose (Gibco BRL) as described previously (38). Hansenula wingei chromosomes (Boehringer) were used as DNA size markers. Following staining with ethidium bromide, the gel was blotted onto a Zeta-Probe GT nylon membrane (Bio-Rad) by the alkaline transfer method and hybridized as described by the manufacturer. The probes used were VSG V02 (33), VSG 1.8 (24), and 50-bp repeats (41).

RESULTS

Integration of a TK gene in the 221 VSG expression site confers sensitivity to nucleoside analogs.

We and others have previously reported that procyclic trypanosomes engineered to express the HSV-1 TK gene are sensitive to nucleoside analogs added to the culture medium (22, 36). As this system could be used to select cells which had ceased expression of active TK protein, it appeared suitable for the study of VSG expression site switching in vitro. By placing TK in a VSG expression site, it should be possible through the lethal combination of TK and nucleoside analog to mimic the negative selection imposed by the host immune response against a VSG antigenic type. We designed a construct with the HSV-1 TK gene downstream of the hygromycin phosphotransferase gene such that both markers would become integrated 272 bp downstream of the promoter of the 221 VSG expression site (Fig. 1). This location was chosen since an inserted marker at that position did not affect the ability of the expression site to switch off (31) and because it would allow us to focus on events whereby the entire expression site had become inactivated. Electroporation of bloodstream-form 221a trypanosomes with this construct yielded six clones from three experiments. Two clones, HTK3 and HTK16, were analyzed in detail. Correct targeting of the construct to the 221 expression site in each cell line was confirmed by Southern hybridization (data not shown).

HTK trypanosomes were found to grow normally in liquid culture (Fig. 2). They were inhibited and killed by the nucleoside analog FIAU, whereas 221a wild-type cells were unaffected. After an interval, however, HTK trypanosomes began to grow out at a rate which matched that of the wild type, suggesting reversion to a TK− phenotype. This phenomenon was also observed with other nucleoside analogs (bromovinyldeoxyuridine and ethyldeoxyuridine [data not shown]). It has previously been reported that reversion of TK-expressing procyclic trypanosomes is due to inactivating point mutations in the TK gene (36). This could also explain reversion of the bloodstream HTK cells. However, in this case, VSG expression site switching should also lead to FIAU resistance.

FIG. 2.

FIG. 2

Open in a new tab

Effect of FIAU on the growth of TK trypanosomes. TK-expressing trypanosomes (diamonds) and wild-type parent trypanosomes (stars) were grown in medium with (broken line) or without (solid line) the nucleoside analog FIAU. Trypanosomes were counted from samples of each culture at given time points.

The majority of FIAU-resistant cells appear to express a VSG different from 221.

On the basis of sensitivity to hygromycin, it is possible to discriminate between HTK revertants which arise due to mutation of the TK gene and those which have inactivated the 221 expression site. This is because in trypanosomes in which the TK gene has been mutated, the HYG gene in the active 221 expression site is still transcribed, resulting in hygromycin resistance. In contrast, if the 221 expression site has been inactivated, then no hygromycin phosphotransferase is produced and the cells are sensitive to hygromycin. In order to allow VSG expression site switch variants to arise and survive, HTK trypanosomes were grown in the absence of hygromycin for 3 days prior to negative selection. FIAU was then added to the culture, which was distributed over 96-well microtiter dishes under conditions in which clonal outgrowth of revertants occurred (see Materials and Methods). FIAU-resistant clones were picked and screened for hygromycin resistance. We found from over 10 experiments that the frequency of FIAUr Hygr clones (most likely TK− mutants) was on the order of 10−6, although on rare occasions this rose to 10−4. We have no simple explanation for this variation. TK− mutants do not appear to have any growth advantage over TK+ cells in culture, nor do they preferentially survive stresses such as cryopreservation (data not shown). The frequency of FIAUr Hygs clones ranged from 1 × 10−5 to 3 × 10−5. Thus, in most experiments the majority of revertants showed the hygromycin-sensitive phenotype. An estimation of the rates of VSG expression site switching and mutation is presented below.

If the hygromycin-sensitive revertants had indeed undergone a VSG expression site switch as expected, then they should express a VSG protein different from 221. As VSG is the most abundant protein of the bloodstream trypanosome, it can be directly visualized in crude cell lysates following SDS-PAGE and Coomassie staining. Different VSG proteins often separate from each other under denaturing conditions, so gel migration can be used as a guide for VSG type. Figure 3 shows the protein analysis of hygromycin-sensitive revertants from a typical FIAU selection experiment. The 221 protein expressed by the parent HTK cell line is the most intensely staining band. This was confirmed by Western blotting with anti-221 antibodies (data not shown). Most of the FIAUr Hygs clones appear to have lost this protein and to have acquired another intensely staining band, indicating expression of a VSG different from 221. Some clones still appear to express the 221 VSG (clones 6, 8, and 10). Clones of this type were investigated further by Western analysis with anti-221 antibodies (data not shown). Some were indeed positive for 221, as would be expected if the TK and HYG genes had been lost following gene conversion by the homologous region of another VSG expression site (see below). Others, however, did not react with the polyclonal 221 antiserum, suggesting the presence of a different VSG that happens to comigrate with 221 during SDS-PAGE. On the basis of gel migration, we have observed six different VSG variants in the switching experiments done to date. This is likely to be an underestimate of the actual number of switch variant types for the reason outlined above. The switch variant which occurred most frequently (20 to 40%) is characterized by a VSG which migrates differently from VSG 221 at 67 kDa during SDS-PAGE (for example, clones 1, 5, 7, and 13 in Fig. 3). Variants that express a VSG of this size have been observed previously in switching experiments in which mice were infected with strain 427 221a trypanosomes (31). This VSG, designated V02, is the most common type which our 221-expressing trypanosome strain switches to in vivo. VSG V02 shows cross-reactivity with a polyclonal serum raised against VSG 221 (31), and this was also found for all 11 clones tested expressing the 67-kDa VSG (data not shown). Hence, V02 also appears to be the favored switch variant following in vitro selection of HTK trypanosomes.

FIG. 3.

FIG. 3

Open in a new tab

Protein analysis of putative TK trypanosome switch variants. Crude protein lysates from the TK-expressing parent cell line and from cells selected for FIAU resistance and hygromycin sensitivity in one experiment were analyzed by SDS-PAGE and Coomassie staining. An Eagle Eye (Stratagene) scan of the stained gel is shown. The VSG protein is the most prominent band, as indicated to the right of each lane (stars). The position of the 221 VSG is shown.

Measurement of VSG expression site switching and TK inactivation rates.

It is difficult to derive a reliable estimate for the VSG switching rate in vivo due to the inherent variation exhibited by such biological phenomena. This is not the case with an in vitro system, however, in which it is possible to determine the rate of a particular event by the Luria-Delbrück fluctuation test (21). The rate of reversion to a TK− phenotype in the presence of hygromycin gives an estimate for the mutation rate of the gene TK. In the absence of hygromycin, the rate of reversion reflects the rate of mutation plus the rate of expression site switching. Thus, it is possible to calculate the switching rate from these two values. The results of the fluctuation tests are shown in Table 1. This analysis indicated that the mutation rate of TK in the 221 expression site is 3.8 × 10−7 cell−1 generation−1. This rate is similar to that observed for reversion to FIAU resistance in procyclic trypanosomes which express HSV-1 TK from the ribosomal DNA array (1.2 × 10−7 cell−1 generation−1 [36]). By subtracting the TK inactivation rate from the rate of reversion in the absence of hygromycin, we estimate the rate of VSG expression site switching to be 7.2 × 10−6, almost 20-fold higher than the rate of mutation.

TABLE 1.

Experimental data for fluctuation analysis of reversion of the TK trypanosomesa

Parameter and formula Result with the indicated selective agent(s)
FIAU FIAU + hygromycin
No. of trypanosomes
 Initial (Ni) 10 10
 Final (Nf)  2.9 × 105  2.9 × 105
Total no. of cell generations per culture: (Nf − Ni)/ln 2 4.18 × 105 4.18 × 105
Experimental cultures
 Total 96 96
 Without growth 4 82
 Proportion with no revertants (P0) 0.042 0.854
Avg no. of revertants per culture: m = −ln P0 3.17 0.16
Rate of reversion (per cell per generation): m/total no. of generations 7.58 × 10−6 3.8 × 10−7
Open in a new tab
a

The rate of reversion to FIAU resistance in the presence or absence of hygromycin for cell line HTK3 was calculated by using the Luria-Delbrück fluctuation test equation described in Materials and Methods. 

Characterization of the events leading to VSG expression site switching in the HTK trypanosomes.

There are a number of different mechanisms by which the TK-expressing trypanosomes could become resistant to FIAU and sensitive to hygromycin, as outlined in Fig. 4. It is possible to discriminate between these events by examining the VSG coat expressed, the presence or absence of the HYG, TK, and VSG 221 marker genes, and the chromosomal locations of these marker genes in the different switch variants. A total of 32 FIAUr Hygs clones derived from four independent experiments were analyzed for their VSG coat by SDS-PAGE and Coomassie staining and by Western blotting and for their marker genotype by DNA dot blot hybridization. The results are summarized in Table 2. Although the relative frequencies of different switch events varied somewhat from experiment to experiment, as can be expected from natural fluctuation within distinct populations, a broad trend was observed. The majority of clones (87%) were found to express a VSG different from 221. Of the cells still expressing 221, all had lost both the HYG and the TK genes. The most likely explanation for this result is that the region of the 221 expression site which contains the integrated construct had been replaced by sequences from another VSG expression site via gene conversion (event 2 in Fig. 4). Although this type of event is not regarded as an expression site switch in the strict sense (i.e., there is no antigenic switch), the gene conversion could encompass a large portion of the expression site and therefore switch the expression of a number of different ESAGs.

TABLE 2.

Switching profiles of the TK trypanosomes following selection with FIAUa

Switch eventb Genotype No. of clones
Total (%)
Expt 1, HTK3 Expt 2, HTK3 Expt 3, HTK3 Expt 4, HTK16
Marker loss (1) HYG− TK− 221− 3 5 5 10 71
Marker loss (2) HTG− TK− 221+ 0 0 1 3 13
Recip recom (3, 4) HYG+ TK+ 221+ 0 0 0 0 0
In situ (5) HYG+ TK+ 221+ 4 0 1 0 16
Open in a new tab
a

FIAU-resistant hygromycin-sensitive clones from four separate experiments were analyzed with respect to the presence and chromosomal location of the HYG, TK, and VSG 221 marker genes, the results of which are summarized. 

b

The type of switch event as designated in Fig. 4 (Recip recom, reciprocal recombination). 

The clones expressing a VSG other than 221 could be divided into two classes, depending on the presence or absence of the marker genes. A total of 13% of these switch variants had retained the HYG, TK, and VSG 221 genes and could be discriminated on the basis of the chromosomal locations of these genes. By pulsed-field gel electrophoresis analysis (data not shown), all were found to retain the three marker genes on the same chromosome band as the parent HTK strain. This indicates an in situ switch (event 5 in Fig. 4) rather than a reciprocal recombination (event 3). The lack of telomere exchange events was expected, given that the TK gene is located just downstream of the expression site promoter in the HTK trypanosomes. Surprisingly, over 70% of switch variants were found to have lost the HYG, TK, and VSG 221 genes.

Most variants have lost the 221 expression site and activated another VSG expression site.

The experiments described above suggested that the majority of switch variants had inactivated the TK gene by deleting the entire 221 expression site. To investigate this further, we performed a chromosomal analysis of clones which showed this type of switch event. We focused upon those variants which had switched to expression of the VSG V02 gene, since this was the most common switch variant type and since we had probes with which to detect V02 sequences. Chromosomal DNAs from 11 V02 switch variants were size fractionated by pulsed-field gel electrophoresis. Conditions were chosen to optimize separation of the ca. 3.2-Mbp chromosome that contains the telomeric VSG 221 gene, which in this strain is present as a single copy. Following electrophoresis, the gel was stained with ethidium bromide and then blotted onto a nylon membrane and hybridized with a panel of DNA probes (Fig. 5).

FIG. 5.

FIG. 5

Open in a new tab

Chromosomal analysis of V02 switch variants in which the markers have been lost. Chromosomal DNA of the parent HTK cell line and of V02 switch variants which had lost HYG, TK, and VSG 221 were separated by CHEF, and the gel was stained with ethidium bromide (panel EtBr). Lane mkr, H. wingei chromosomes. The position of the chromosome which bears the 221 VSG expression site is indicated (221). The gel was then transferred to a nylon membrane and hybridized with the probes indicated below each panel.

Hybridization with V02 showed that eight switch variants appeared to be the same as the HTK parent. From analyzing the ethidium bromide stain, we found that many of these clones showed large rearrangements of the 3.2-Mbp chromosome. Of the remaining three clones, two (F2 and F8) showed an additional V02 signal at the position of the 3.2-Mbp chromosome, indicating gene conversion of the 221 expression site by the V02 site (note that variant F8 is not unambiguously interpretable [see below]). The other clone (2B3) also showed hybridization of the 3.2-Mbp chromosome with V02, but this was not the result of a gene conversion, since the V02 copy on a small chromosome had been lost. Since this small chromosome contains the V02 expression site (31) and since clone 2B3 expresses V02 ESAG 6 (data not shown), a likely explanation for this event is that the chromosome had recombined with the one containing the 221 expression site but with loss of the 221 expression site sequences. A similar type of complex switch has been reported previously (37).

We then probed for the VSG 1.8 gene, since a copy of this comigrates with the 221 gene and most likely resides on the same chromosome. The pattern of hybridization in the 3-Mbp size range with the VSG 1.8 probe confirmed the results of ethidium bromide staining. In some clones, such as 2C5, F12, and F22, the 3.2-Mbp chromosome was reduced by up to 200 kbp. Other clones showed an increase in size of the 3.2-Mbp chromosome (F9) or even a duplication (F1 and F17; note that for F17, the larger of the duplicated chromosomes is barely resolved from the compression zone in this gel). Variant F8 showed a smear of hybridization, the upper part of which was also detected by the V02 probe. As the remaining chromosomes in this lane had resolved well, our interpretation of the switch event for clone F8 is a V02 gene conversion, but one which results in a chromosome that is unstable and collapses to a form which has lost the V02 gene. A procyclic acidic repetitive protein A (PARP A) locus has previously been mapped to the 221-containing chromosome (15). Hybridization with PARP A gave a pattern identical to that of VSG 1.8 (data not shown), again confirming the rearrangement of the 3.2-Mbp chromosome in these clones.

Bloodstream-form VSG expression sites invariably contain a long stretch of 50-bp repeats located upstream of the expression site promoter (41). The 221 expression site has approximately 40 kbp of 50-bp repeats in our 221a trypanosomes (32). We questioned whether these repeats were still present in the rearranged 3.2-Mbp chromosome in the V02 switch variants. The 3.2-Mbp chromosome of clones 2B3, F2, and F8, which contains a copy of VSG V02, hybridized with the 50-bp repeat probe as expected. Variant RD4, which has a 3.2-Mbp chromosome similar in size to that of the parent HTK cell line, also retained the 50-bp repeats on this chromosome. However, all the other clones showed either no or only faint hybridization of the rearranged 3.2-Mbp chromosome to the 50-bp repeat probe, regardless of whether this chromosome had increased or decreased in size. The same result was obtained when the hybridized membrane was washed under conditions of low stringency (data not shown). The results suggest that in these switch variants the entire 221 VSG expression site, including the 50-bp repeats, had been lost and that the V02 expression site had been activated in situ without rearrangement.

DISCUSSION

In this report, we demonstrate that a viral TK gene can be used as a negatively selectable marker in bloodstream-form T. brucei suitable for studying VSG expression site switching in vitro. We think that this TK system will prove a useful addition to the tools available for investigation of the mechanisms of antigenic variation. Indeed, TK-expressing trypanosomes have already been used to see how changing the level of DNA modification J can affect VSG expression site switching (40). We have estimated the rates of switching and mutation of the VSG expression site by the Luria-Delbrück fluctuation test. Although we have not directly demonstrated that inactivation of TK in bloodstream trypanosomes can arise from point mutation of the TK gene, this seems likely, since this mechanism had been observed previously in procyclic trypanosomes (36). On the basis of the inactivation of TK in the absence of an expression site switch, we calculate the mutation rate in bloodstream trypanosomes to be 3.8 × 10−7 per cell per generation. This is similar to what we observed for TK in the ribosomal DNA of procyclic trypanosomes.

We found that the rate of inactivation of the 221 VSG expression site approaches 10−5 per cell per generation in the TK-transformed cells. Almost 90% of inactivation events resulted in the expression of a new VSG gene. The rate of switching is greater than what we observe with other 427 221a trypanosome lines, for which switch variants were selected in vivo in immunized mice. Rudenko et al. (31) reported rates of switching in vivo of between 10−6 and 10−7 for both the wild type and transformants containing a HYG gene integrated near the 221 expression site promoter. McCulloch et al. (23) observed switching rates of between 1.2 × 10−6 and 1 × 10−7 (with an average of 6 × 10−7) for three different 427 221a transformant lines in nine experiments, with wild-type switching occurring at a rate of 3 × 10−7. Since VSG switch variants were generated in vitro for both types of experiment, the difference in switching frequency is probably a consequence of the method of selection. For example, the majority of switch variants arise late in the growth of the initial culture, when cell numbers are highest. Due to the slow turnover of VSG, these late switchers can have a mixed coat long enough to be a target for immune destruction. In contrast, the combination of relatively rapid dilution of TK activity to a sublethal level and slow action of the selection system would greatly favor survival of the late switchers in vitro. The elevated rate of switching is not simply due to the presence of the TK gene, since 427 221a trypanosomes containing only a HYG gene in the 221 expression site also switch in vitro at rates approaching 10−5 (32).

We observed at least six different types of VSG switch variant. We think that these variants represent different expression sites switched on following inactivation of the TK-marked 221 expression site. The criterion we used to distinguish between different expression sites is migration of VSG protein during SDS-PAGE. This assay is rather crude, in that different VSGs can migrate similarly, so it is possible that the actual number of sites which can be activated could be nearer the 20 or so in the trypanosome’s repertoire. However, some expression sites might be activated at a low frequency or might be nonfunctional. Another possibility is that some expression sites are not suitable for growth under particular culture conditions. We have found that variability in ESAGs 6 and 7, which encode the trypanosome transferrin receptor, may be required primarily to allow T. brucei to cope with the diversity of transferrins in a range of mammalian hosts (6). Support for this idea comes from switching experiments using different sera in the culture medium, which we find can greatly influence the number and type of variants that survive the selection (12).

When we analyzed the type of events which led to VSG switching, we found that 18% were in situ switches, while the remainder involved loss of the 221 expression site. For the latter, we expected that the active expression site had been replaced via gene conversion by another VSG expression site, as described previously (26, 29, 35, 37). However, in the sample of 11 V02 VSG switch variants we analyzed, only 2 contained a new copy of the V02 gene in the 3.2-Mbp chromosome. The majority of events therefore appear to be complex switches involving chromosome rearrangement and V02 expression site activation. One possibility is a gene conversion of the 221 expression site by another expression site, which subsequently becomes silenced upon activation of V02. Such events have been suggested before (25). We found, however, that the 50-bp repeats were absent from the 3.2-Mbp chromosome in all but one of the remaining switch variants. As these repeats are invariably found upstream of the expression site promoter (41), we conclude that 8 of 11 clones have undergone an unknown rearrangement that did not involve another VSG expression site.

Although complex VSG switches following immune selection in animals have been noted before (25, 35), they have been considered to be rare events. The experiments reported here indicate that they can be rather frequent. We suspected that chromosomal rearrangement may be a consequence of using TK and nucleoside analogs which can interfere with DNA replication. However, two lines of evidence argue against this. First, the majority of switch variants arise during expansion of trypanosomes in the absence of FIAU. Second, and more compelling, is the occurrence of similar chromosomal rearrangements during switching from the 221 or the V02 expression site in 427 transformant cell lines which do not contain the TK gene (32). The very low level of transcription from a silenced 221 promoter (26, 31) did not produce sufficient TK activity to be lethal to the cell, since we did obtain clones which had switched off the 221 site. Besides, we find that expression of TK integrated in the tubulin array, where the level of transcription is much higher than in a silent VSG expression site, does not make the trypanosome sensitive to FIAU under the conditions used in the selection procedure (4, 12).

It is difficult to make a valid comparison of the frequencies of switch event types in vivo with the in vitro switching experiments reported here, due to the nature of the selection system. To survive FIAU selection in vitro, the trypanosome has to prevent by some means the expression of the TK gene located near the active expression site promoter. Replacement of the VSG gene at the end of the expression site, an event often observed in vivo, simply won’t do. However, it is interesting that in a study of switching events in animals, Myler et al. (25) reported that in 7 of 19 switches the previously active site had been lost and another VSG expression site on a different chromosome had been activated. This type of switch has all the hallmarks of the expression site loss event we describe here. As 5 of the 19 switches reported by Myler were the result of a simple VSG gene conversion, an event excluded by the TK selection system, the frequency of complex expression site loss switches in vivo is 50%. This is not significantly different from what we observe in vitro.

Although more work is required to establish the mechanism of deletion, several alternatives can be envisaged. Deletion could arise through aberrant interchromosomal recombination promoted by a repeated element, such as a retrotransposon (e.g., TRS/INGI [28]). This would account for the large size difference in the 3.2-Mbp chromosome of the switch variants. Chromosome breakage and healing, phenomena seen often in Plasmodium spp. (20), would also delete the telomeric VSG expression site. Another possibility is that destruction of the expression site may be the result of a normal switching event which has gone wrong. For instance, a cut in the expression site postulated to initiate a gene conversion reaction is followed by gap widening and a search for homology (reviewed in references 2 and 8). If this search fails, then the site may be destroyed by exonuclease activity followed by recombination events to heal the chromosome.

We consider it highly unlikely that VSG expression site deletion contributes to survival of T. brucei in the wild. We think that the gross DNA rearrangements observed by us represent background genetic noise, made audible by a powerful selection system amplifying rare events. Trypanosome lines recently passaged through tsetse flies switch coat at rates up to 10−2 per cell division, i.e., 3 to 4 orders of magnitude higher than the rates of the rodent-adapted strains studied by molecular biologists (reviewed in reference 2). If this rapid switching includes in situ switching (mechanism 5 in Fig. 4), a point that remains to be verified, then the rare expression site deletions observed here would have no physiological significance.

Nevertheless, we think that these rare events are of special interest because they provide insight into the mechanism of expression site switching. The frequent deletion of the previously active site during the switches analyzed here and in other switches using positive selection (32) strongly suggests that activation of a new site cannot readily occur without inactivation of the old one. It now appears that loss of the old site helps to activate the new one, rather than that different sites are activated and inactivated independently, as previously thought (1, 9; see reference 7 for discussion). There are other lines of evidence arguing against independent activation-inactivation of sites. Davies et al. (14) have recently reported very high switch rates following removal of a putative stabilizing element, and we have found that two expression sites cannot be maximally active at the same time, showing that in situ switches must involve some kind of interaction between the sites involved (7). The TK system described here should prove useful in dissecting this interaction.

ACKNOWLEDGMENTS

We thank Magali Berberof, Pat Blundell, Ines Chaves, Anita Dirks-Mulder, Herlinde Gerrits, Rudo Kieft, Ronald Plasterk, Gloria Rudenko, and Fred van Leeuwen for helpful discussions and critical reading of the manuscript.

This work was supported by grants from the European Commission to M.C., from the Wellcome Trust to M.C.T., and from the Netherlands Organization for Scientific Research (NWO/SON) to P.B.

REFERENCES

  • 1.Baltz T, Giroud C, Baltz D, Roth C, Raibaud A, Eisen H. Stable expression of two variable surface glycoproteins by cloned Trypanosoma equiperdum. Nature. 1986;319:602–604. doi: 10.1038/319602a0. [DOI] [PubMed] [Google Scholar]
  • 2.Barry J D. The relative significance of mechanisms of antigenic variation in African trypanosomes. Parasitol Today. 1997;13:212–217. doi: 10.1016/s0169-4758(97)01039-9. [DOI] [PubMed] [Google Scholar]
  • 3.Bernards A, De Lange T, Michels P A, Liu A Y, Huisman M J, Borst P. Two modes of activation of a single surface antigen gene of Trypanosoma brucei. Cell. 1984;36:163–170. doi: 10.1016/0092-8674(84)90085-0. [DOI] [PubMed] [Google Scholar]
  • 4.Blundell, P. A., and P. Borst. 1997. Unpublished data.
  • 5.Blundell P A, Rudenko G, Borst P. Targeting of exogenous DNA into Trypanosoma brucei requires a high degree of homology between donor and target DNA. Mol Biochem Parasitol. 1996;76:215–229. doi: 10.1016/0166-6851(95)02560-x. [DOI] [PubMed] [Google Scholar]
  • 6.Borst P, Bitter W, Blundell P A, Cross M, McCullouch R, Rudenko G, Taylor M C, Van Leeuwen F. The expression sites for variant surface glycoproteins of Trypanosoma brucei. In: Hide G, Mottram J C, Coombs G H, Holmes P H, editors. Trypanosomiasis and leishmaniasis: biology and control. Oxford, United Kingdom: British Society for Parasitology/CAB International; 1997. pp. 109–131. [Google Scholar]
  • 7.Borst, P., W. Bitter, P. A. Blundell, R. McCullouch, I. M. F. Chaves, M. Cross, H. Gerrits, F. Van Leeuwen, M. C. Taylor, and G. Rudenko. Control of VSG gene expression sites in Trypanosoma brucei. Mol. Biochem. Parasitol., in press. [DOI] [PubMed]
  • 8.Borst P, Rudenko G, Taylor M C, Blundell P A, Van Leeuwen F, Bitter W, Cross M, McCullouch R. Antigenic variation in trypanosomes. Arch Med Res. 1996;27:379–388. . (Review.) [PubMed] [Google Scholar]
  • 9.Cornelissen A W, Johnson P J, Kooter J M, Van der Ploeg L H, Borst P. Two simultaneously active VSG gene transcription units in a single Trypanosoma brucei variant. Cell. 1985;41:825–832. doi: 10.1016/s0092-8674(85)80063-5. [DOI] [PubMed] [Google Scholar]
  • 10.Cross G A. Antigenic variation in trypanosomes: secrets surface slowly. Bioessays. 1996;18:283–291. doi: 10.1002/bies.950180406. . (Review.) [DOI] [PubMed] [Google Scholar]
  • 11.Cross G A, Manning J C. Cultivation of Trypanosoma brucei sspp. in semi-defined and defined media. Parasitology. 1973;67:315–331. doi: 10.1017/s0031182000046540. [DOI] [PubMed] [Google Scholar]
  • 12.Cross, M., and P. Borst. 1997. Unpublished data.
  • 13.Cully D F, Ip H S, Cross G A. Coordinate transcription of variant surface glycoprotein genes and an expression site associated gene family in Trypanosoma brucei. Cell. 1985;42:173–182. doi: 10.1016/s0092-8674(85)80113-6. [DOI] [PubMed] [Google Scholar]
  • 14.Davies K P, Carruthers V B, Cross G A M. Manipulation of the vsg cotransposed region increases expression-site switching in Trypanosoma brucei. Mol Biochem Parasitol. 1997;86:163–177. doi: 10.1016/s0166-6851(97)02853-3. [DOI] [PubMed] [Google Scholar]
  • 15.Gottesdiener K, Garcia-Anoveros J, Lee M G, Van der Ploeg L H. Chromosome organization of the protozoan Trypanosoma brucei. Mol Cell Biol. 1990;10:6079–6083. doi: 10.1128/mcb.10.11.6079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gottschling D E, Aparicio O M, Billington B L, Zakian V A. Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell. 1990;63:751–762. doi: 10.1016/0092-8674(90)90141-z. [DOI] [PubMed] [Google Scholar]
  • 17.Hirumi H, Hirumi K. Continuous cultivation of Trypanosoma brucei blood stream forms in a medium containing a low concentration of serum protein without feeder cell layers. J Parasitol. 1989;75:985–989. [PubMed] [Google Scholar]
  • 18.Horn D, Cross G A. A developmentally regulated position effect at a telomeric locus in Trypanosoma brucei. Cell. 1995;83:555–561. doi: 10.1016/0092-8674(95)90095-0. [DOI] [PubMed] [Google Scholar]
  • 19.Horn D, Cross G A M. Analysis of Trypanosoma brucei vsg expression site switching in vitro. Mol Biochem Parasitol. 1997;84:189–201. doi: 10.1016/s0166-6851(96)02794-6. [DOI] [PubMed] [Google Scholar]
  • 20.Lanzer M, Fischer K, Le Blancq S M. Parasitism and chromosome dynamics in protozoan parasites: is there a connection? Mol Biochem Parasitol. 1995;70:1–8. doi: 10.1016/0166-6851(95)00021-r. . (Review). [DOI] [PubMed] [Google Scholar]
  • 21.Luria S E, Delbruck M. Mutations of bacteria from virus sensitivity to virus resistance. Genetics. 1943;28:491–511. doi: 10.1093/genetics/28.6.491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Lye L, Wang C C. Thymidine kinase as a selectable marker for studying the biogenesis of glycosomes in Trypanosoma brucei. Exp Parasitol. 1996;82:211–217. doi: 10.1006/expr.1996.0026. [DOI] [PubMed] [Google Scholar]
  • 23.McCulloch R, Rudenko G, Borst P. Gene conversions mediating antigenic variation in Trypanosoma brucei can occur in variant surface glycoprotein expression sites lacking 70-base-pair repeat sequences. Mol Cell Biol. 1997;17:833–843. doi: 10.1128/mcb.17.2.833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Michels P A, Van der Ploeg L H, Liu A Y, Borst P. The inactivation and reactivation of an expression-linked gene copy for a variant surface glycoprotein in Trypanosoma brucei. EMBO J. 1984;3:1345–1351. doi: 10.1002/j.1460-2075.1984.tb01975.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Myler P J, Aline R F, Jr, Scholler J K, Stuart K D. Multiple events associated with antigenic switching in Trypanosoma brucei. Mol Biochem Parasitol. 1988;29:227–241. doi: 10.1016/0166-6851(88)90078-3. [DOI] [PubMed] [Google Scholar]
  • 26.Navarro M, Cross G A. DNA rearrangements associated with multiple consecutive directed antigenic switches in Trypanosoma brucei. Mol Cell Biol. 1996;16:3615–3625. doi: 10.1128/mcb.16.7.3615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Overath P, Stierhof Y D, Wiese M. Endocytosis and secretion in trypanosomatid parasites—tumultuous traffic in a pocket. Trends Cell Biol. 1997;7:27–33. doi: 10.1016/S0962-8924(97)10046-0. [DOI] [PubMed] [Google Scholar]
  • 28.Pays E. Genome organization and control of gene expression in trypanosomatids. In: Broda P M A, Oliver S G, Sims P F G, editors. The eukaryotic genome—organisation and regulation. Cambridge, United Kingdom: Society for General Microbiology, Cambridge University Press; 1992. pp. 127–160. [Google Scholar]
  • 29.Pays E, Van Assel S, Laurent M, Dero B, Michiels F, Kronenberger P, Matthyssens G, Van Meirvenne N, Le Ray D, Steinert M. At least two transposed sequences are associated in the expression site of a surface antigen gene in different trypanosome clones. Cell. 1983;34:359–369. doi: 10.1016/0092-8674(83)90370-7. [DOI] [PubMed] [Google Scholar]
  • 30.Revelard P, Lips S, Pays E. A gene from the VSG expression site of Trypanosoma brucei encodes a protein with both leucine-rich repeats and a putative zinc finger. Nucleic Acids Res. 1990;18:7299–7303. doi: 10.1093/nar/18.24.7299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Rudenko G, Blundell P A, Dirks-Mulder A, Kieft R, Borst P. A ribosomal DNA promoter replacing the promoter of a telomeric VSG gene expression site can be efficiently switched on and off in T. brucei. Cell. 1995;83:547–553. doi: 10.1016/0092-8674(95)90094-2. [DOI] [PubMed] [Google Scholar]
  • 32.Rudenko, G., I. M. F. Chaves, A. Dirks-Mulder, and P. Borst. 1997. Unpublished data.
  • 33.Rudenko G, McCulloch R, Dirks-Mulder A, Borst P. Telomere exchange can be an important mechanism of variant surface glycoprotein gene switching in Trypanosoma brucei. Mol Biochem Parasitol. 1996;80:65–75. doi: 10.1016/0166-6851(96)02669-2. [DOI] [PubMed] [Google Scholar]
  • 34.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
  • 35.Shea C, Glass D J, Parangi S, Van der Ploeg L H. Variant surface glycoprotein gene expression site switches in Trypanosoma brucei. J Biol Chem. 1986;261:6056–6063. [PubMed] [Google Scholar]
  • 36.Valdes J, Taylor M C, Cross M A, Ligtenberg M J, Rudenko G, Borst P. The viral thymidine kinase gene as a tool for the study of mutagenesis in Trypanosoma brucei. Nucleic Acids Res. 1996;24:1809–1815. doi: 10.1093/nar/24.10.1809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Van der Ploeg L H, Cornelissen A W, Michels P A, Borst P. Chromosome rearrangements in Trypanosoma brucei. Cell. 1984;39:213–221. doi: 10.1016/0092-8674(84)90207-1. [DOI] [PubMed] [Google Scholar]
  • 38.Van der Ploeg L H, Schwartz D C, Cantor C R, Borst P. Antigenic variation in Trypanosoma brucei analyzed by electrophoretic separation of chromosome-sized DNA molecules. Cell. 1984;37:77–84. doi: 10.1016/0092-8674(84)90302-7. [DOI] [PubMed] [Google Scholar]
  • 39.Van der Ploeg L H, Valerio D, De Lange T, Bernards A, Borst P, Grosveld F G. An analysis of cosmid clones of nuclear DNA from Trypanosoma brucei shows that the genes for variant surface glycoproteins are clustered in the genome. Nucleic Acids Res. 1982;10:5905–5923. doi: 10.1093/nar/10.19.5905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Van Leeuwen, F., R. Kieft, M. Cross, and P. Borst. 1997. Unpublished data.
  • 41.Zomerdijk J C, Ouellette M, ten Asbroek A L, Kieft R, Bommer A M, Clayton C E, Borst P. The promoter for a variant surface glycoprotein gene expression site in Trypanosoma brucei. EMBO J. 1990;9:2791–2801. doi: 10.1002/j.1460-2075.1990.tb07467.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Molecular and Cellular Biology are provided here courtesy of Taylor & Francis

RESOURCES