Targetome Analysis of Malaria Sporozoite Transcription Factor AP2-Sp Reveals Its Role as a Master Regulator

Abstract

Malaria transmission to humans begins with sporozoite infection of the liver. The elucidation of gene regulation during the sporozoite stage will promote the investigation of mechanisms of liver infection by this parasite and contribute to the development of strategies for preventing malaria transmission. AP2-Sp is a transcription factor (TF) essential for the formation of sporozoites or sporogony, which takes place in oocysts in the midguts of infected mosquitoes. To understand the role of this TF in the transcriptional regulatory system of this stage, we performed chromatin immunoprecipitation sequencing (ChIP-seq) analyses using whole mosquito midguts containing late oocysts as starting material and explored its genome-wide target genes. We identified 697 target genes, comprising those involved in distinct processes parasites experience during this stage, from sporogony to development into the liver stage and representing the majority of genes highly expressed in the sporozoite stage. These results suggest that AP2-Sp determines basal patterns of gene expression by targeting a broad range of genes directly. The ChIP-seq analyses also showed that AP2-Sp maintains its own expression by a transcriptional autoactivation mechanism (positive-feedback loop) and induces all TFs reported to be transcribed at this stage, including AP2-Sp2, AP2-Sp3, and SLARP. The results showed that AP2-Sp exists at the top of the transcriptional cascade of this stage and triggers the formation of this stage as a master regulator. IMPORTANCE The sporozoite stage plays a central role in malaria transmission from a mosquito to vertebrate host and is an important target for antimalarial strategies. AP2-Sp is a candidate master transcription factor for the sporozoite stage. However, study of its role in gene regulation has been hampered because of difficulties in performing genome-wide studies of gene regulation in this stage. Here, we conquered this problem and revealed that AP2-Sp has the following prominent features as a master transcription factor. First, it determines the repertory of gene expression during this stage. Second, it maintains its own expression through a transcriptional positive-feedback loop and induces all other transcription factors specifically expressed in this stage. This study represents a major breakthrough in fully understanding gene regulation in this important malarial stage.

Keywords: malaria; sporozoite; transcription factor; transcriptional regulation.

FIG 1

ChIP-seq of AP2-Sp. (A) Motif associated with summits of AP2-Sp peaks in experiment 1. Binding sequences nearest to the summit were visualized with WebLogo software. (B) Distance from the summit to the nearest motif sequence was calculated in each AP2-Sp peak, and the distribution is shown with a histogram. The horizontal axis indicates the distance from the summit to the nearest motif sequence (the bin size is 100 bp). (C) ChIP-seq peaks of AP2-Sp upstream of target genes. Seven genes functionally different in the sporozoite stage were selected (see also Fig. 2D and E). Binding motifs of AP2-Sp under the peaks are indicated by bars. Peaks were visualized by using Integrative Genomics Viewer (IGV) software. CSP, circumsporozoite protein; IMC1a, inner membrane complex protein 1a; SERA5, serine repeat antigen 5; MAEBL, membrane-associated erythrocyte binding-like protein; SPECT2, sporozoite micronemal protein essential for cell traversal 2; RON2, rhoptry neck protein 2; UIS3, upregulated in sporozoite 3; IK2, eukaryotic translation initiation factor 2-alpha kinase 2. (D) Comparison of the mapped views of ChIP-seq peaks obtained by two independent experiments. Reads were mapped on chromosome 14. (E) Numbers of common peaks between experiment 1 (1,270 peaks) and experiment 2 (1,156 peaks) within the selected distance were plotted. The plot indicates that approximately 80% (895 peaks) of the peaks in experiment 2 had counterparts in experiment 1 within 150 bp. (F) Venn diagram showing common peaks of two ChIP-seq experiments. Peaks whose summits were within 150 bp were regarded as common. (G) Coverage maps of two ChIP-seq experiments were created by setting positions of peak summits identified in experiment 1 in the center.

FIG 2

Targetome of AP2-Sp. (A) Common target genes predicted by two ChIP-seq experiments. (B) Common target genes were classified according to functional categories. The number of target genes in each group is shown on a pie graph. (C) GO enrichment analysis. Terms enriched in target genes (P < 0.01) are listed. BP, biological process; CC, cellular component. (D) Products of target genes are grouped by localization to subcellular structures characteristic to motile stages: microneme, rhoptry, and pellicle structure. Proteins belonging to the pellicle structure are further classified into cytoskeletal proteins and proteins involved in gliding motility. These assignments were performed according to previous literatures. GAMA* was additionally predicted as a target manually using ChIP-seq peaks and RNA-seq data. IMC, inner membrane complex; ISP, IMC subcompartment protein; ISC, IMC suture component; PHIL1, photosensitized INA-labeled protein 1; SPM, subpellicular microtubules; GAP, glideosome-associated protein; GAPM, glideosome-associated protein with multiple-membrane spans; GAC, glideosome-associated connector; ADF, actin-depolymerizing factor; MTIP, myosin A-tail interacting protein; ELC, myosin essential light chain; RON, rhoptry neck protein; ARNP, apical rhoptry neck protein; ASP, apical sushi protein; AARP, apical asparagine-rich protein; RhopH3, high-molecular-weight rhoptry protein 3; RAMA, rhoptry-associated membrane antigen; RALP1, rhoptry-associated leucine zipper-like protein 1; ROP14, rhoptry protein 14; RAP, rhoptry-associated protein; ICP, inhibitor of cysteine proteases; MAEBL, merozoite apical erythrocyte-binding ligand; TRAP, thrombospondin-related anonymous protein; TREP, TRAP-related protein; TRP1, thrombospondin-related protein 1; GAMA, GPI-anchored micronemal antigen; CSP, circumsporozoite protein; SPECT, sporozoite microneme protein essential for cell traversal; PLP1, perforin-like protein1; AMA1, apical membrane antigen 1; CelTOS, cell transversal for ookinetes and sporozoites; TLP, TRAP-like protein; TRAMP, thrombospondin-related apical membrane protein; STAPR, secreted protein with altered thrombospondin repeat domain; TRSP, thrombospondin-related sporozoite protein; ROM, rhomboid protease; APH, acylated pleckstrin-homology domain-containing protein. (E) Parasite progression through the sporozoite stage, starting with sporogony and ending with development into the liver stage, is illustrated. Target genes are categorized according to the steps they are involved in. The five genes marked by an asterisk were additionally predicted as targets manually using ChIP-seq peaks and RNA-seq data. GEST, gamete egress and sporozoite traversal protein; SSP3, sporozoite surface protein 3; SERA, serine repeat antigen 5; ORP2, oocyst rupture protein 2; SIAP1, sporozoite invasion-associated protein 1; CRMP, cysteine repeat modular protein; PL, phospholipase; Puf2, Pumilio-FBF family protein 2; IK2, eukaryotic translation initiation factor 2-alpha kinase 2; UIS, upregulated in infectious sporozoite; LISP1, liver specific protein 1; ZIPCO, ZIP domain-containing protein; AQP, aquaglyceroporin; SPELD, sporozoite surface protein essential for liver-stage development.

FIG 3

Targetome of AP2-Sp in sporozoite transcriptome. (A) Plasmodium berghei genes excluding putative subtelomeric genes (4,654 genes) were ordered according to the RPKM values obtained by RNA-seq analyses in the oocyst/oocyst sporozoite (14 days after infective blood meal) and in the salivary gland sporozoite (21 days after infective blood meal). The proportion of target genes in the top 100 highly expressed genes is shown as a pie graph for each transcriptome (blue). Nontarget genes are divided into two groups: genes related to protein synthesis (orange) and other genes (green). The number of target genes in each group is shown on a pie graph. (B) P. berghei genes were grouped according to RPKM values (in log₂ scale), and the percentage of target genes in each group is plotted in a bar graph.

FIG 4

Positive transcriptional feedback is essential for maintaining AP2-Sp expression. (A) Clusters of AP2-Sp peaks and corresponding putative binding motifs in the upstream region of the AP2-Sp gene. Binding motifs of AP2-Sp under the peaks are indicated by bars. Transcripts of AP2-Sp obtained by RNA-seq were mapped onto the genome parallel to ChIP-seq data. Blue and red rectangles indicate reads mapped to forward and reverse strands of the P. berghei genome, respectively. The image was created using IGV software. (B) Schematic diagram of AP2-Sp_pro_mut parasite preparation. Mutant parasites were prepared from pbcas9 using the double CRISPR method. Positions targeted by gRNA are indicated by scissor characters. Vertical bars indicate mutated binding motifs of AP2-Sp (23 motifs including overlapping motifs). They were mutated by adding 16 mutations (basically, [T/C]GCA[T/C][G/A] was changed to TGAA[T/C][G/A]), which were confirmed by sequencing (lower panels). (C) RT-qPCR assays of AP2-Sp transcripts were performed for pbcas9 and AP2-Sp_pro_mut parasites. Total RNA was extracted from midguts at 14 days after infective blood meal. The 60S ribosomal protein L 21 mRNA was used as an internal control. The results are shown as expression of AP2-Sp relative to that in pbcas9. The data are averages of three biologically independent experiments (± the standard errors [SE]). Student t tests were used for statistical analyses. P values of <0.05 were considered significant (*). (D) The expression of AP2-Sp target genes was compared between pbcas9 and AP2-Sp_pro_mut parasites using RNA-seq. The RPKM values of each gene were normalized with average RPKM values of nontarget genes, and the log₂ (fold change) for these values was calculated for pbcas9 and AP2-Sp_pro_mut parasites. Analyses were performed in two independent clonal parasite lines described earlier, and the results of the target (upper panel) and other (lower panel) genes are shown as histograms. (E) The expression of eight target genes was compared using RT-qPCR assays in pbcas9 and AP2-Sp_pro_mut parasites. Assays were performed as described for panel C. The 60S ribosomal protein L 21 mRNA was used as an internal control. The results are shown as gene expression relative to that in pbcas9. Data represent the averages of three biologically independent experiments (± the SE). *, P < 0.05; **, P < 0.01. RON2, rhoptry neck protein 2; CSP, circumsporozoite protein; TRAP, thrombospondin-related anonymous protein; MAEBL, merozoite apical erythrocyte-binding ligand; SPECT2, sporozoite microneme protein essential for cell traversal 2; UIS4, upregulated in infectious sporozoite 4.

FIG 5

TFs specific to the sporozoite stage are targets of AP2-Sp. (A) Clusters of AP2-Sp peaks and putative binding motifs of AP2-Sp under the peaks in the upstream region of the AP2-Sp2 gene. Binding motifs of AP2-Sp under the peaks are indicated by bars. Transcripts of AP2-Sp obtained by RNA-seq were mapped onto the genome parallel to ChIP-seq data. Blue and red rectangles indicate reads mapped to forward and reverse strands of the P. berghei genome, respectively. The image was created using IGV software. (B) Clusters of AP2-Sp peaks and putative binding motifs of AP2-Sp in the upstream region of the AP2-Sp3 gene. (C) Clusters of AP2-Sp peaks and putative binding motifs of AP2-Sp in the upstream region of the SLARP gene. (D) Clusters of AP2-Sp peaks and putative binding motifs of AP2-Sp in the upstream region of the AP2-L gene E. Schematic diagram of the preparation of AP2-Sp2_pro_mut parasites. Mutant parasites were prepared from pbcas9 using the double CRISPR method. Positions targeted by gRNA are indicated by scissor characters. Vertical bars indicate mutated binding motifs of AP2-Sp. Basically, [T/C]GCA[T/C][G/A] was changed to TGAA[T/C][G/A]. Mutations were confirmed by sequencing (lower panels). Two clones were prepared independently, in which seven (clone 1) and nine (clone 2) motifs (not including overlapped ones) were found to be mutated. (F) RT-qPCR assays of AP2-Sp2 transcripts. Experiments were performed between pbcas9 and AP2-Sp2_pro_mut parasites at 14 days after infective blood meal using mosquito midguts as starting material. The 60S ribosomal protein L 21 mRNA was used as an internal control. The results are shown as expression of AP2-Sp2 relative to that in pbcas9. The data are averages of three biologically independent experiments (± the SE). The statistical analyses were performed using Student t tests (*, P < 0.05).