Cell-cycle-regulated control of VSG expression site silencing by histones and histone chaperones ASF1A and CAF-1b in Trypanosoma brucei

Abstract

Antigenic variation in African trypanosomes involves monoallelic expression and reversible silencing of variant surface glycoprotein (VSG) genes found adjacent to telomeres in polycistronic expression sites (ESs). We assessed the impact on ES silencing of five candidate essential chromatin-associated factors that emerged from a genome-wide RNA interference viability screen. Using this approach, we demonstrate roles in VSG ES silencing for two histone chaperones. Defects in S-phase progression in cells depleted for histone H3, or either chaperone, highlight in particular the link between chromatin assembly and DNA replication control. S-phase checkpoint arrest was incomplete, however, allowing G2/M-specific VSG ES derepression following knockdown of histone H3. In striking contrast, knockdown of anti-silencing factor 1A (ASF1A) allowed for derepression at all cell cycle stages, whereas knockdown of chromatin assembly factor 1b (CAF-1b) revealed derepression predominantly in S-phase and G2/M. Our results support a central role for chromatin in maintaining VSG ES silencing. ASF1A and CAF-1b appear to play constitutive and DNA replication-dependent roles, respectively, in the recycling and assembly of chromatin. Defects in these functions typically lead to arrest in S-phase but defective cells can also progress through the cell cycle leading to nucleosome depletion and derepression of telomeric VSG ESs.

Figure 1.

Candidate VSG ES regulators. (A) The plot shows the BF^T1 outputs from a genome-scale RNAi screen for viability; RNAi induced in bloodstream-form cells for 3 days. Previously identified VSG ES regulators, new candidate regulators and core histones are highlighted. A z-score >3.3 indicates statistical significance. (B) Heat maps showing previously identified and new candidate VSG ES regulators. BF^T1, as above; BF^T2, RNAi induced in bloodstream-form cells for 6 days; PF, RNAi induced in procyclic cells; DIF, RNAi induced throughout differentiation from bloodstream to procyclic form. Red indicates a loss-of-fitness. Data derived from previous study (15).

Figure 2.

Candidate VSG ES regulators localize to the nucleus. Immunofluorescence detection of epitope-tagged proteins; the regions outlined in the phase panels indicate the regions shown in the (immuno)fluorescence panels. The nucleus (N) and kinetoplast (K, mitochondrial genome) were stained with the DNA intercalating dye, DAPI. Scale bar, 5 µm.

Figure 3.

Candidate VSG ES regulators are essential for bloodstream stage cell survival. Growth curves for (A) bloodstream and (B) insect stage cells following induction of RNAi (+Tet). In each case, data were derived from four and two independent strains, respectively. Error bars represent 1 SD (standard deviation). RNAi was induced in the presence of 1 μg/ml tetracycline.

Figure 4.

Histone chaperones maintain VSG ES silencing. (A) Schematic map indicating the location of the GFP:NPT reporter downstream of the silent VSG221 ES promoter; these cells express VSGX from a different ES. (B) CAF-1b or ASF1A knockdown coincides with derepression of a silent VSG ES in bloodstream-form cells; data from two independent strains shown for CAF-1b and ASF1A. Coomassie (coo)-stained gels serve as loading controls. RNAi was induced in the presence of 1 μg/ml tetracycline for 24 h.

Figure 5.

Histone H3 depletion leads to VSG ES derepression. (A) Western blot showing depletion of histone H3. (B) Histone H3 RNAi knockdown leads to a severe growth defect and (C) derepression of a previously silent VSG221 ES promoter in bloodstream-form cells after 7-h induction, as determined by western blotting. Coomassie-stained gels serve as loading controls. Data derived from three independent strains. Error bars represent 1 SD. RNAi was induced in the presence of 1 μg/ml tetracycline.

Figure 6.

Histone H3 depletion triggers an S-phase defect and G₂/M-specific VSG ES derepression. (A) Analysis of DAPI-stained cells after 7-h induction of histone H3 RNAi. K, kinetoplast; N, nucleus. Data derived from three independent strains. Error bars represent 1 SD. (B) Flow cytometry analysis showing cell cycle population distributions (left hand panels) and GFP fluorescence levels (right hand panels) after 4 and 7 h of induction. (C) Flow cytometry analysis showing GFP fluorescence from wild-type cells (upper panel) and cells with the GFP:NPT cassette at the active VSG221 ES promoter (lower panel). (D) Flow cytometry analysis of the GFP positive population from B; histone H3 RNAi for 7 h. Horizontal lines indicate the cell cycle stages, and inset table details the distribution of cells in each stage. RNAi was induced in the presence of 1 μg/ml tetracycline.

Figure 7.

Depletion of ASF1A leads to VSG ES derepression at all cell cycle stages. (A) Analysis of DAPI-stained cells after 24-h induction of ASF1A RNAi. Data were derived from two independent strains. Error bars represent 1 SD. (B) Flow cytometry analysis showing cell cycle population distributions (left hand panels) and GFP fluorescence levels (right hand panels) after 24-h induction. (C) Western blot showing histone H3 levels following ASF1A depletion. The coomassie (coo)-stained gel serves as a loading control. (D) Flow cytometry analysis of the GFP-positive population. Other details as in Figure 6D.

Figure 8.

Depletion of CAF-1b leads to VSG ES derepression during S-phase and G₂/M. (A) Analysis of DAPI-stained cells after 24-h induction of CAF-1b RNAi. Data were derived from two independent strains. Error bars represent 1 SD. (B) Flow cytometry analysis showing cell cycle population distributions (left hand panels) and GFP fluorescence levels (right hand panels) after 24-h induction. (C) Western blot showing histone H3 levels following CAF-1b depletion. The coomassie (coo)-stained gel serves as a loading control. (D) Flow cytometry analysis of the GFP positive population from B. Other details as in Figure 6D. (E) Nucleosome ladders generated by micrococcal nuclease digestion; equal numbers of cells were analysed in each sample and digestion was carried out between 10 and 60 min.