Schematic depiction of apicoplast ACP and the FASII pathway in P. falciparum.

CoA = coenzyme A, ACPS = holo-ACP synthase, FabD = malonyl-CoA:ACP S-malonyltransferase, 4-PP = 4-phosphopantetheine. The ACP structural models are based on the X-ray structures of P. falciparum holo-aACP (PDB 3GZM)36 and E. coli acyl-ACP (PDB 5USR)45.

Knockout or knockdown of ACP expression blocks parasite growth and apicoplast biogenesis.

(A) Synchronous growth assay of ΔACP PfMev parasites cultured ±50 µM mevalonate (Mev). Parasitemia values are the average ± SD from biological triplicate samples. (B) Genomic PCR analysis and live parasite imaging show selective loss of the apicoplast genome and disrupted apicoplast morphology based on fluorescence of the ACPL-GFP marker protein in PfMev parasites. Nuc. = nuclear gene (LDH, Pf3D7_1324900), Api. = apicoplast gene (SufB, Pf3D7_API04700), Mito. = mitochondrial gene (CoxI, Pf3D7_MIT02100). (C) Synchronous growth assay of apicoplast ACP-aptamer/TetR-DOZI Dd2 parasites cultured ±1 µM aTc and ±200 µM IPP. Parasitemia values are the average ± SD from biological triplicates. Inset: western blot analysis of parasites harvested after 1, 3, or 5 days at the indicated growth conditions and probed with anti-EF1α (cytosolic loading control) or anti-HA (ACP) antibodies. (D) Quantitative PCR analysis of the apicoplast:nuclear genome ratio for parasites in panel C cultured 84 hours in the indicated conditions, based on amplification of apicoplast (SufB, ClpM: Pf3D7_API03600, TufA: Pf3D7_API02900) relative to nuclear (STL: Pf3D7_0717700, I5P: Pf3D7_0802500, ADSL: Pf3D7_0206700) genes. Indicated qPCR ratios were normalized to +aTc and are the average ± SD of biological triplicates. Significance was analyzed by unpaired Student’s t-test to determine the indicated p values. (E) Immunofluorescence microscopy of parasites in panel C cultured for 5 days (120 hours) in the indicated conditions and stained with anti-apicoplast ACP or DAPI (nucleus). Below: population analysis of apicoplast morphology scored for disrupted (dispersed), punctate, or elongated GFP signal in 50 total parasites from biological triplicates.

Holo-ACP synthase but not FabD is essential for blood-stage parasites and apicoplast biogenesis.

(A) Synchronous growth assay of ΔACPS PfMev parasites cultured ±50 µM mevalonate. (B) Genomic PCR analysis and (C) live parasite imaging show selective loss of the apicoplast genome and disrupted apicoplast morphology based on fluorescence of the ACPL-GFP marker protein in PfMev parasites. (D) Synchronous growth assay of ΔFabD PfMev parasites cultured ±50 µM mevalonate. (E) Genomic PCR analysis and (F) live parasite imaging indicate retention of the apicoplast genome and normal apicoplast morphology based on ACPL-GFP fluorescence in ΔFabD parasites. Parasitemia values are the average ± SD from biological triplicates. Nuc. = nuclear gene (LDH), Api. = apicoplast gene (SufB), Mito. = mitochondrial gene (CoxI), DIC = differential interference contrast.

Apicoplast ACP associates with apicoplast pyruvate kinase II.

(A) Proximity biotinylation study of Dd2 parasites episomally expressing apicoplast ACP tagged with C-miniTurbo, showing log2 enrichment ratio of known apicoplast-targeted proteins compared to parental Dd2 parasites, based on spectral intensity of proteins detected by tandem mass spectrometry. (B) Immunoprecipitation-mass spectrometry (IP-MS) study of apicoplast ACP interacting proteins, based on anti-HA-tag IP of apicoplast versus mitochondrial ACP-HA2 in Dd2 parasites lysed in Triton X-100 or RIPA buffer. Axes display the log2 enrichment ratio for detection of each protein in IP samples of aACP versus mACP. Dashed diagonal line has a slope of one. (C) IP-MS analysis based on anti-HA-tag IP of lysates from Dd2 parasites episomally expressing wildtype (WT) or the S95A mutant of apicoplast ACP-HA2 and lysed in digitonin or RIPA buffer. Axes display the log2 enrichment ratio for detection of each protein in IP samples of WT versus S95A ACP. Dashed diagonal line has a slope of one. The identity of numbered proteins in panels A-C is shown in Supplementary File 1. A list of all proteins detected by mass spectrometry is shown in Figure 4-source data 2. (D) Affinity pull-down and western blot analysis of lysates from E. coli bacteria heterologously expressing parasite His6-tagged PKII or apicoplast ACP-HA2 (WT or S95A mutant). PKII was affinity isolated from bacterial lysates using nickel-nitrilotriacetic acid (Ni-NTA) resin, eluted with free imidazole, and analyzed by western blot for co-purification with ACP. Membranes were probed with anti-His6 and anti-HA-tag antibodies. The signal intensity for WT or S95A ACP relative to PKII in pulldown samples was quantified by densitometry in biological triplicate samples and plotted as the average ± SD, with significance analyzed by Student’s t-test to determine the indicated p value.

ACP knockdown destabilizes PKII and impairs apicoplast DNA and RNA levels.

(A) Western blot analysis of apicoplast ACP-aptamer/TetR-DOZI Dd2 parasites episomally expressing pyruvate kinase II (PKII) C-Myc3 and cultured for 3 days ±1 µM aTc with 200 µM IPP. The blot was probed with α-Myc (PKII), α-HA (ACP), and α-εF1β (cytosolic loading control) antibodies and stained with Ponceau S for total protein. (B) Densitometry quantification of PKII and ACP protein levels normalized to εF1β based on western blot analysis of three biological replicates. Graph depicts the average ± SD, normalized to +aTc conditions. (C) Quantitative PCR analysis of DNA isolated from biological triplicate samples of apicoplast ACP-aptamer/TetR-DOZI parasites cultured for 36 or 84 hours ±1 µM aTc with 200 µM IPP, with normalization of Ct values averaged from three apicoplast genes (SufB, TufA, ClpM) to Ct values averaged from three nuclear genes (STL, I5P, ADSL). Graphs depict the average ± SD of each timepoint, normalized to +aTc conditions. (D) Quantitative RT-PCR analysis of RNA isolated from biological triplicate samples obtained from parasites cultured under identical conditions as for panel C to determine normalized levels of apicoplast (SufB, TufA, ClpM) relative to nuclear (STL, I5P, ADSL) transcripts. Indicated p values were determined by Student t-test analysis.

Scheme of ACP interaction with PKII and dysfunctions upon ACP knockdown.

Loss of ACPS is expected to phenocopy loss of ACP. PEP = phosphoenolpyruvate, (d)NTPs = nucleoside triphosphates and deoxynucleoside triphosphates. The aACP structural model is based on PDB 3GZM and 5USR. The structural model of aACP bound to PKII was generated using AlphaFold 3.

Integration PCRs for deletion or modification of the apicoplast ACP gene.

(A) Genotyping PCR for successful deletion of the apicoplast ACP gene (Pf3D7_0208500) in polyclonal PfMev parasites showing selective amplification of sequence at the 5’ and 3’ ends of the drug-resistance marker (Δ5’ and Δ3’) in knockout parasites but not in parental PfMev parasites and exclusive amplification of sequence at the 5’ and 3’ ends of the ACP CDS in WT but not ΔACP parasites. Genotype PCR analysis confirming on-target integration of the 3’ aptamer/TetR-DOZI cassette at the apicoplast ACP locus in polyclonal PfMev (B) or clonal Dd2 parasites (C) and the absence of the unmodified wildtype (WT) locus, with parental parasites serving as negative control for integration.

Additional microscopy images ΔACP PfMev NF54 parasites.

Phase = phase contrast images, api-SFG = ACPL-superfolder GFP, DAPI = nuclear DNA stain.

Synchronous growth assays (A), fluorescence microscopy (B), and genomic qPCR (C) of ACP knockdown in PfMev NF54 parasites.

For the growth assay in panel A, cultures were split on day 6. Samples for genomic qPCR were cultured as indicated for ≥8 days. Gene copy numbers were based on amplification of apicoplast (SufB: Pf3D7_API04700, ClpM: Pf3D7_API03600, TufA: Pf3D7_API02900) or nuclear (ADSL: Pf3D7_0206700) relative to nuclear (STL: Pf3D7_0717700, I5P: Pf3D7_0802500) genes. Indicated qPCR ratios were normalized to +aTc and are the average ± SD of biological triplicates.

Additional IFA images for apicoplast morphology upon ACP knockdown in Dd2 parasites in +aTc (A) or -aTc/+IPP (B) conditions.

BF = bright field.

Validation of custom anti-apicoplast ACP antibody.

Indicated amount of purified aACP antigen was loaded and fractionated on SDS-PAGE gel, transferred to membrane, blocked, and probed with 1:1000 dilution of rabbit pre-bleed serum or final serum, washed, and probed with donkey anti-rabbit IRDye680 secondary antibody.

Integration PCRs for deletion of ACPS (A) or FabD (B) in PfMev parasites.

Additional microscopy images of apicoplast morphology in ΔACPS (A) and ΔFabD (B) PfMev parasites.

Phase = phase contrast, api-SFG = apicoplast-targeted superfolder GFP.

Growth of ΔACP (A) and ΔFabD (B) PfMev parasites in low-lipid conditions with daily media changes.

Western blot detection of biotinylated proteins in lysates from parental, aACP-miniTurbo, or aACP-BioID2 Dd2 parasites.

Enriched protein interactors of apicoplast ACP detected by proximity biotinylation or immunoprecipitation.

(A) Apicoplast ACP interactors identified by miniTurbo versus BioID2. (B) Interacting proteins of ACP enriched in IP samples of apicoplast compared to mitochondrial ACP in Triton X-100 or RIPA buffer. (C) Interacting proteins of ACP enriched in IP samples of WT compared to S95A ACP in RIPA or digitonin buffer. All axes display the log2 enrichment ratio.

Replicate western blot images for Ni-NTA pulldown of PKII association with WT (A and B) or Ser95Ala (C and D) ACP in E. coli.

Replicate western images for ACP knockdown and PKII levels

AlphaFold3 model of apicoplast ACP bound to pyruvate kinase II, rendered in PyMOL.

The unstructured N-terminus of parasite PKII, which is expected to function as the apicoplast-targeting sequence and be cleaved upon organelle import, was not included in modeling. The presence of the 4-PP group was modeled by superposition with ACP in the PDB file 5USR.