A mutagenesis screen for essential plastid biogenesis genes in human malaria parasites

Abstract

Endosymbiosis has driven major molecular and cellular innovations. Plasmodium spp. parasites that cause malaria contain an essential, non-photosynthetic plastid-the apicoplast-which originated from a secondary (eukaryote-eukaryote) endosymbiosis. To discover organellar pathways with evolutionary and biomedical significance, we performed a mutagenesis screen for essential genes required for apicoplast biogenesis in Plasmodium falciparum. Apicoplast(-) mutants were isolated using a chemical rescue that permits conditional disruption of the apicoplast and a new fluorescent reporter for organelle loss. Five candidate genes were validated (out of 12 identified), including a triosephosphate isomerase (TIM)-barrel protein that likely derived from a core metabolic enzyme but evolved a new activity. Our results demonstrate, to our knowledge, the first forward genetic screen to assign essential cellular functions to unannotated P. falciparum genes. A putative TIM-barrel enzyme and other newly identified apicoplast biogenesis proteins open opportunities to discover new mechanisms of organelle biogenesis, molecular evolution underlying eukaryotic diversity, and drug targets against multiple parasitic diseases.

Fig 1. A fluorescent reporter for apicoplast loss in P. falciparum is specifically degraded in the apicoplast.

(A) Strategy for a conditional GFP reporter that fluoresces upon apicoplast loss. (B) Reporter construct for expression of apicoplast-targeted GFP (ACP_L-GFP) and cytoplasmic mCherry used for integration into P. falciparum Dd2^attB parasites. (C) Ratio of GFP:mCherry fluorescence in untreated versus actinonin/IPP-treated parasites expressing ACP_L-GFP-degron. Data are shown as mean ± SEM (n = 3). ****P < 0.0001, unpaired two-tailed t test. See also S1 Fig. Tabulated data are shown in S1 Data. (D) GFP protein levels in untreated versus actinonin/IPP-treated parasites expressing GFP-EcssrA. Higher molecular weight species in GFP blot from unprocessed ACP_L is indicated with black arrowhead. (E) Flow cytometry plots showing mCherry and GFP fluorescence in untreated versus actinonin/IPP-treated parasites expressing GFP-EcssrA. Uninfected RBCs are gated away (lower left quadrants), and the percentage mCherry⁺, GFP⁺ parasites in the gated populations relative to the total number of cells quantified are indicated. (F) Representative live-cell fluorescent images of untreated and actinonin/IPP-treated parasites expressing mCherry and GFP-EcssrA. Hoechst stains for parasite nuclei. Scale bar 5 μm. ACP, acyl-carrier protein; ACP_L, ACP leader peptide; a.f.u., arbitrary fluorescence units; EcssrA, E. coli ssrA peptide; GFP, green fluorescent protein; IPP, isopentenyl diphosphate; RBC, red blood cell.

Fig 2. A screen for apicoplast(−) mutant clones identifies candidate biogenesis genes.

(A) Schematic for selection and characterization of apicoplast(−) mutant clones. (B) Representative flow cytometry histograms of ENU_1 population (shown in D) GFP fluorescence over the course of successive fluorescence-activated cell sorts. Distribution of GFP fluorescence for pure populations of apicoplast(+) and apicoplast(−) parasites are shown for comparison in grey and blue, respectively. (C) Apicoplast genome coverage in parental strains and mutant clones from whole-genome sequencing. Tabulated data are shown in S2 Data. (D) Genomic loci of SNVs identified in coding regions of mutant clones. The mutagenized populations from which each clone was derived are indicated. For specific mutations, see also Table 1 and S2 Table. a.f.u., arbitrary fluorescence units; EMS, ethyl methanesulfonate; ENU, N-ethyl-N-nitrosourea; FACS, fluorescence-activated cell sorting; GFP, green fluorescent protein; IPP, isopentenyl diphosphate; NGS, next generation sequencing; SNP, single nucleotide polymorphism; SNV, single nucleotide variant.

Fig 3. The I437S variant disrupts PfFtsH1 protease and ATPase activity in vitro.

(A) Domain organization of PfFtsH1 showing location of I437S. The known ATPase-inactive (E249Q) and peptidase-inactive (D493A) variants are also shown. (B) Peptidase activity of recombinant PfFtsH1 WT and mutants. Data are shown as mean ± SEM (n = 3). ****P < 0.0001, unpaired two-tailed t test. Tabulated data are shown in S3 Data. (C) ATPase activity of recombinant PfFtsH1 WT and mutants. Data are shown as mean ± SEM (n = 3). ***P = 0.0009, **P = 0.0016 adjusted P values compared to WT, ordinary one-way ANOVA with Tukey’s multiple comparisons test. Tabulated data are shown in S3 Data. PfFtsH1, P. falciparum ATP-dependent zinc metalloprotease 1; TM transmembrane; V, velocity; WT, wild-type.

Fig 4. PfAtg7 is a cytoplasmic protein required for PfAtg8 membrane conjugation and apicoplast biogenesis.

(A) Domain organization of PfAtg7 showing location of the identified stop codon. (B) Growth time course of PfAtg7-TetR/DOZI knockdown parasites in the absence of ATc with and without IPP. Data are shown as mean ± SD (n = 3). ****P < 0.0001 adjusted P value (−ATc compared to −ATc/+IPP), two-way ANOVA with Sidak’s multiple comparisons test. Tabulated data are shown in S4 Data. (C) Representative immunofluorescence images of PfAtg7-TetR/DOZI parasites showing apicoplast morphology (anti-ACP) in +ATc and −ATc/+IPP parasites at reinvasion cycle 7. Scale bar 5 μm. (D) Western blot of apicoplast-targeted ClpP in PfAtg7-TetR/DOZI parasites in the presence and absence of ATc at reinvasion cycle 7. Processed ClpP after removal of its transit peptide is approximately 25 kDa, while full-length protein is about 50 kDa. Result is representative of n = 2 blots. (E) Live-cell fluorescent images of GFP-PfAtg8 upon PfAtg7 knockdown 24 hours post ATc removal, prior to apicoplast loss. Scale bar 5 μm. ACP, acyl-carrier protein; ATc, anhydrotetracycline; Clp, caseinolytic protease; ClpP, Clp protease; GFP, green fluorescent protein; IPP, isopentenyl diphosphate; PfAtg7, P. falciparum autophagy-related protein 7; TetR/DOZI, tetracycline repressor/development of zygote inhibited fusion protein.

Fig 5. Three “conserved Plasmodium proteins of unknown function” are required for apicoplast biogenesis.

(A) Domain organization of Pf3D7_0518100, Pf3D7_1305100, and Pf3D7_1363700 showing location of identified stop codons. (B, E, and H) Growth time course of Pf3D7_0518100, Pf3D7_1305100, and Pf3D7_1363700-TetR/DOZI knockdown parasites in the absence of ATc with and without IPP. Data are shown as mean ± SD (n = 3). ****P < 0.0001 adjusted P value (−ATc compared to −ATc/+IPP), two-way ANOVA with Sidak’s multiple comparisons test. Tabulated data are shown in S4 Data. (C, F, and I) Representative immunofluorescence images of indicated TetR/DOZI parasites showing the apicoplast (anti-ACP) in +ATc and −ATc/+IPP parasites at reinvasion cycle 7. Scale bars 5 μm. (D, G, and J) Western blot of apicoplast-targeted ClpP in indicated TetR/DOZI parasites in the presence and absence of ATc at reinvasion cycle 7. Processed ClpP after removal of its transit peptide is approximately 25 kDa, while full-length protein is about 50 kDa. Results are representative of n = 2 western blots. ACP, acyl-carrier protein; ATc, anhydrotetracycline; Clp, caseinolytic protease; ClpP, Clp protease; IPP, isopentenyl diphosphate; TetR/DOZI, tetracycline repressor/development of zygote inhibited fusion protein.

Fig 6. Evolutionary divergence from IGPS proteins suggests a novel apicoplast biogenesis function for PfAMR1.

(A) Conservation of active-site residues between IGPS and IGPS-like proteins. Sequences of IGPS homologs identified in V. brassicaformis and P. falciparum and known IGPS enzymes from E. coli and Saccharomyces cerevisiae were aligned. Active-site residues identified from structure-function studies in IGPS enzymes are listed. ^dCatalytic residues. See also S6 Fig. (B) Schematic of IGPS and IGPS-like protein loss in P. falciparum and Cryptosporidium parvum, respectively. (C) Complementation of an E. coli trpC mutant (trpC9800) with IGPS and IGPS-like proteins from V. brassicaformis and P. falciparum grown on minimal media (M9) + antibiotic resistance (carbenicillin) + and trp. Empty vector condition contains only antibiotic resistance. BL21 Star (DE3) strain was used for the WT condition. (D) Same complementation as in panel C but grown without trp. AMR, apicoplast-minus IPP-rescued; AS, anthranilate synthase; IGPS, indole-3-glycerol phosphate synthase; IPP, isopentenyl diphosphate; PRT, phosphoribosyltransferase; trp, tryptophan; TS, tryptophan synthase; WT, wild-type.