A var gene promoter implicated in severe malaria nucleates silencing and is regulated by 3' untranslated region and intronic cis-elements

Abstract

Questions surround the mechanism of mutually exclusive expression by which Plasmodium falciparum mediates activation and silencing of var genes. These encode PfEMP1 proteins, which function as cytoadherent and immunomodulatory molecules at the surface of parasitised erythrocytes. Current evidence suggests that promoter silencing by var introns might play a key role in var gene regulation. To evaluate the impact of cis-acting regulatory regions on var silencing, we generated P. falciparum lines in which luciferase was placed under the control of an UpsA var promoter. By utilising the Bxb1 integrase system, these reporter cassettes were targeted to a genomic region that was not in apposition to var subtelomeric domains. This eliminated possible effects from surrounding telomeric elements and removed the variability inherent in episomal systems. Studies with highly synchronised parasites revealed that the UpsA element possessed minimal activity in comparison with a heterologous (hrp3) promoter. This may result from the integrated UpsA promoter being largely silenced by the neighbouring cg6 promoter. Our analyses also revealed that the DownsA 3' untranslated region further decreased the luciferase activity from both cassettes, whereas the var A intron repressed the UpsA promoter specifically. By applying multivariate analysis over the entire cell cycle, we confirmed the significance of these cis-elements and found the parasite stage to be the major factor regulating UpsA-promoter activity. Additionally, we observed that the UpsA promoter was capable of nucleating reversible silencing that spread to a downstream promoter. We believe these studies are the first to analyse promoter activity of Group A var genes, which have been implicated in severe malaria, and support the model that var introns can further suppress var expression. These data also suggest an important suppressive role for the DownsA terminator. Our findings imply the existence of multiple levels of var gene regulation in addition to intrinsic promoter-dependent silencing.

Fig. 1

Generation and molecular characterization of luciferase reporter cassettes integrated into the Plasmodium falciparum Dd2^attB parasite line. (A) Schematic of the genomic layout of the Group A var gene PFD1235w (previously named AL010226) in the P. falciparum 3D7 strain. Flags represent putative transcription start sites. Bars represent DNA regions chosen for analysis. The diagram is not drawn to scale. (B) Schematic of the integration of a generic luciferase reporter cassette-bearing pCBD-P plasmid (containing the attP element) into the attB locus of the Dd2^attB parasite line. The attB docking site was engineered into the cg6 gene by homologous recombination (Nkrumah et al., 2006). Co-transfection of pCBD-P plasmids with the integrase-encoding plasmid pInt resulted in site-specific integration of the pCBD-P plasmid into the genome, flanked by attL and attR sequences. Upon recombination, the luciferase cassette resided downstream of the cg6 amino-terminal region, followed by the pBluescript plasmid backbone and the bsd selectable marker cassette placed in the opposite orientation to the reporter. Restriction digest fragment sizes refer to the integration of the pCBD-P-ALH plasmid. The positions of the probes used for Southern blot hybridization are indicated as bars above the Dd2^attB/Luc locus (black bar: cg6 probe; gray bar: luciferase probe). EV, EcoRV; K, KpnI. (C) Experimental design of luciferase reporter assays, depicting the cassettes subcloned into pCBD-P and integrated into Dd2^attB. The grey arrow indicates the luciferase open reading frame. UpsA, the 2.15 kb 5’ untranslated region (UTR) of PFD1235w; DownsA, the 1.4 kb 3’ UTR of PFD1235w; Group A intron, the 0.74 kb PFD1235w intron. (D) PCR reactions documenting the integration of luciferase reporters into Dd2^attB. attB: Amplification of the Dd2^attB parent strain and designated luciferase reporter lines with primers 3’F1 (cg6-specific) and T3 (pBS plasmid backbone). varA and hrp3: Results of PCR over the attL recombination site using the primer 3’F1 and either the UpsA-specific primer 1089 (varA) or the hrp3-specific primer 1090 (hrp3). attR: PCR amplification over the attR junction using primers 594 (specific to the cam promoter driving the bsd selectable marker in all pCBD-P plasmids) and T3. The parasite line ALA2 yielded similar results (data not shown). (E) Southern blot hybridization documented the insertion of single-copy luciferase cassettes into the attB site of the Dd2^attB parasite. Genomic DNA was isolated from parasite lines and digested with EcoRV or KpnI. Hybridization with a cg6 probe revealed a 13.5 kb fragment in the EcoRV-digested Dd2^attB genomic DNA that was absent in all second generation attP recombinants. This band was replaced in all lines by a 9.6 kb band corresponding to the attR junction and the human dhfr cassette, and a band of variable size (due to the different promoters tested) that corresponded to the attL junction and the 5’ portion of the luciferase cassette. The smaller band sizes are as follows: ALH, ALHi and ALA cassettes: 7.5 kb; HLH, HLHi and HLA cassettes: 7.1 kb; ∂LH cassette: 5.3 kb. Hybridization with a luciferase probe showed bands consisting of the entire inserted pCBD-P plasmid, the attR junction, and the plasmid backbone of the integrated attB locus. There was no evidence of plasmid retention or multi-copy insertion, which would have been apparent as bands ~3 kb smaller in size. Sizes were as follows: ALH: 12.2 kb; ALHi: 12.9 kb; ALA: 13.1 kb; ∂LH: 10.0 kb; HLH: 11.8 kb; HLHi: 12.6 kb; and HLA: 12.7 kb. The autoradiograph of the KpnI digested genomic DNA from the hrp3 containing lines HLH, HLHi and HLA was cropped to remove intervening lanes.

Fig. 2

The DownsA terminator (corresponding to the downstream non-coding region from the Plasmodium falciparum Group A var gene PFD1235w) decreases luciferase expression of multiple promoters, while the intron from this same Group A gene represses the var promoter specifically. (A) Experimental design and order of experiments performed on reporter containing parasites. Preclonal parasite lines are denoted with the prefix “Dd2^attB/”. (B) The stage-specific luciferase activity of integrated luciferase cassettes was assessed by harvesting parasites during early ring (eR), late ring (lR), trophozoite (T) and schizont (S) stages. Firefly luciferase activities of integrated reporter cassettes HLH (plotted on the left Y-axis), as well as ALH and ∂LH (plotted on the right Y-axis) are expressed as relative luciferase units (RLU) per million parasites (RPM). These data were obtained from a representative time course experiment performed in triplicate. (C) Ratios of expression from the luciferase cassettes, with significance derived from repeated measures ANOVA with Dunnett’s post-test (or a paired t-test in the case of ∂LH). ALH is defined as the control for UpsA promoter-containing lines and the promoter-less ∂LH line, HLH is defined as the control for the hrp3 promoter-containing lines. ALH and HLH are given ratios of one for comparative purposes. **, P value ≤ 0.01; ***, P value ≤ 0.001.

Fig. 3

Regulatory elements from the Plasmodium falciparum Group A var gene PFD1235w repress cumulative luciferase activity from the UpsA promoter and affect the timing of maximal transcription. (A) Histogram of the percentages of each developmental stage obtained at the harvested time points. Representative Giemsa-stained images are presented, with the border color corresponding to the bar color on the histogram. Parasite lines were subjected to repeat rounds of double sorbitol synchronization, with the initial round performed just after blood cell invasion and a second one performed 12-15 h later. Replicate flasks were prepared to minimize parasite culture handling. (B) relative luciferase units (RLU) per million parasites (RPM) obtained from the time course experiment. (C) From this experiment, the cumulative area under the curve values were derived using a trapezoidal formula.

Fig. 4

Regulatory elements from the Plasmodium falciparum Group A var gene PFD1235w repress cumulative UpsA–produced transcript levels. UpsA is the upstream non-coding region from the Group A var gene PFD1235w. (A-D) Luciferase (A) and bsd (C) transcript values assessed by quantitative real-time PCR with cDNA preparations. These values were normalized to the copy number of actin transcripts, a widely used reference gene. The cumulative area under the curve data are presented for luciferase in (B) and bsd in (D).

Fig. 5

The Plasmodium falciparum UpsA promoter (from the PFD1235w var gene) is associated with epigenetic changes in the surrounding locus and affects the expression of surrounding elements. (A) Growth curves of lines cultured in the presence of the selection agents WR99210 and BSD (open symbols) or WR99210 alone (closed symbols). Parasite cultures were seeded into replicate wells at 1% parasitemia, all with 2.5 nM WR99210 but half with 2 μg/mL BSD. HLH transgenes containing the hrp3 promoter were not affected by the application of BSD pressure, while the growth of UpsA promoter-containing lines was delayed. Prior to application of drug, normal parasite growth was verified by attainment of parasitemia above 5% in complete medium. (B, C) In vitro SYBR Green I incorporation assays were performed against BSD (B) or WR99210 (C) using the ALH parasite lines selected in the presence of WR99210 and BSD (ALH-WB) or WR99210 alone (ALH-W). As a control, we included the HLH parasite line cultured in medium containing both WR99210 and BSD. IC₅₀ values (shown as mean ± S.E.M., calculated from six independent experiments) were derived by regression analysis. (D-E) Luciferase activities of BSD pressured versus non-pressured cell populations, measured in relative luciferase units per million parasites. (D) Time-course of the luciferase activities of BSD sensitive and resistant populations of ALH parasites. As a control, we included the ALA1 parasite line that was cultured in the presence of WR99210 and BSD. (E) Luciferase activity of HLH cultures grown in the presence (HLH1-WB) or absence (ALA1-W) of BSD, harvested during late ring stages.