Epigenetic control of the variable expression of a Plasmodium falciparum receptor protein for erythrocyte invasion

Abstract

Plasmodium falciparum can invade erythrocytes by redundant receptors, some of which have variable expression. A P. falciparum clone Dd2 requiring erythrocyte sialic acid for invasion can be switched to a sialic acid-independent progeny clone Dd2NM by growing the Dd2 clone with neuraminidase-treated erythrocytes. The RH4 gene is transcriptionally up-regulated in Dd2NM compared to Dd2, despite the absence of DNA changes in and around the gene. We determined the epigenetic modifications around the transcription start site (TSS) at the time of expression of RH4 in Dd2NM (44 h) and at an earlier time when RH4 is not expressed (24 h). At 44 h, the occupancy of the +1 nucleosome site downstream of the TSS of the active RH4 gene in Dd2NM was markedly reduced compared to Dd2; no difference was observed at 24 h. At 44 h, histone modifications associated with up-regulation were positively correlated to the active RH4 gene of Dd2NM compared to Dd2; no differences were observed at 24 h. Histone H3K9 trimethylation (a marker for silencing) was higher in Dd2 than Dd2NM along the 5'-UTRs of the RH4 gene at both 44 and 24 h. Our data indicate that the failure of Dd2 to express the sialic acid-independent invasion receptor gene RH4 is associated with the epigenetic silencing mark H3K9 trimethylation present throughout the cycle.

Fig. 1.

Transcription of the RH4 and PEBL genes at 24 h and 44 h of the 48-h asexual blood stage cycle. (A) Schematic structures of the RH4 and PEBL genes and identification of the transcription start sites (TSSs) of each gene. Exon-intron structures of the RH4 and PEBL genes are drawn based on gene sequences in GenBank (release no. AF420309 for RH4 and release no. AY032735 for PEBL) and from reverse transcription-PCR results. Black boxes represent the different exons in the two genes. A star indicates the frame-shift mutation in the 5′ end of the ORF of PEBL. Two arrows in opposite orientations indicate the transcription start sites and transcription orientations of the RH4 and PEBL genes. Three dashed lines indicate multiple TSS positions 3′ to the longest TSS in PEBL (−917 bp). A poly(dA)poly(dT) tract in the region between the transcription start sites of the two genes is indicated by an arrowhead. Total RNA samples from Dd2 and Dd2NM were collected at 24 or 44 h postinvasion for reverse transcription and real-time PCR analysis of transcription of the RH4 (B) and PEBL (C) genes. AMA1 (PlasmoDB ID: PF11_0344) (D), a gene expressed only at 44 h, and an asexual stage, silent var gene Svar1 (PlasmoDB ID: PF07_0051) (E) were studied at the same times. For each real-time PCR, the signal from each gene was normalized by a housekeeping gene 60S ribosomal protein L18 (60S L18, MAL13P1.209; Fig. S2) to generate the relative expression value for each gene. Experiments were independently performed at least three times. The SE is shown as an error bar.

Fig. 2.

Nucleosome mapping in the intergenic region between the start codons for RH4 and PEBL for 44 h parasites. MNase-digested DNA samples from Dd2 and Dd2NM at 44 h were purified on 3.5% agarose gel and prepared for Solexa sequencing (A) and real-time PCR (B). The nucleosomes are shown by the gray ovals with marked numbers. The first nucleosome 3′ to the TSS of RH4 or PEBL was defined as +1. The nucleosome resides in the region between the two TSSs of RH4 and PEBL was defined as −1. The +1 nucleosome of RH4 is boxed and showed a lower number of sequence tags for Dd2NM than for Dd2. (B) Tiling-arranged PCR primer sets were used in real-time PCR to obtain the enrichment signals of nucleosomal DNA. Nucleosomal DNA enrichment values are presented as the percentage of a well-positioned nucleosome with the SE based on three biologically independent samples. Each PCR of Dd2 and Dd2NM were compared by the unpaired t test with Welch’s correction, **P < 0.01, *P < 0.05. (C) Enzymatic digestion of Dd2 and Dd2NM chromatin. The MNase digested (10 U/mL) or undigested (0 U/mL) DNA samples were analyzed on 2% agarose gel electrophoresis. DNA molecular weights were marked in bp on the left. Mononucleosome (1N), dinucleosome (2N), and trinucleosome (3N) released by MNase treatment were indicated on the right.

Fig. 3.

Histone modifications in the RH4 and PEBL genes for 44 h old parasites. Distribution of histone modifications determined by ChIP generated by real-time PCR (Fig. 2B) for the RH4 and PEBL genes from Dd2 and Dd2NM measured in the positions identified in A. After MNase digested chromatin samples were evenly divided, each aliquot was immunoprecipitated by various antibodies against acetylation at histone H3K9/K14 (B), acetylation at histone H4K5/K8/K12/K16 (C), histone H3K4 trimethylation (D), and histone H3K9 trimethylation (E). An active gene AMA1 (A) and a silent var gene Svar1 (Sv) in 44 h parasites were analyzed as controls. For each PCR product position, signals from the ChIP assay are presented as the percentage of input DNA. Experiments were independently performed at least three times to generate the average value of each histone modification signal at each position. The SE is shown as an error bar.

Fig. 4.

Histone H3K9 trimethylation is an epigenetic marker for silencing observed at 24 h for RH4 and PEBL. The chromatin samples from Dd2 and Dd2NM were prepared for nuclesome mapping (A) and ChIP assays for acetylation at histone H3K9/K14 (B), acetylation at histone H4K5/K8/K12/K16 (C), histone H3K4 trimethylation (D), and histone H3K9 trimethylation (E). Tiling-arranged PCR primer sets were used only for scanning occupancy of the +1 TSS nucleosome of the RH4. As described in the legend of Fig. 2, nucleosomal DNA enrichment values are presented as the percentage of a well-positioned nucleosome with the standard error based on three biologically independent samples. Enrichment of histone modifications at −1, +1, and +2 nucleosome positions surrounding the RH4 TSS was determined as described in the legend of Fig. 3. AMA1 and Svar1 genes were tested as control as described in the legend of Fig. 3.