The Metabolite Repair Enzyme Phosphoglycolate Phosphatase Regulates Central Carbon Metabolism and Fosmidomycin Sensitivity in Plasmodium falciparum

Abstract

Members of the haloacid dehalogenase (HAD) family of metabolite phosphatases play an important role in regulating multiple pathways in Plasmodium falciparum central carbon metabolism. We show that the P. falciparum HAD protein, phosphoglycolate phosphatase (PGP), regulates glycolysis and pentose pathway flux in asexual blood stages via detoxifying the damaged metabolite 4-phosphoerythronate (4-PE). Disruption of the P. falciparumpgp gene caused accumulation of two previously uncharacterized metabolites, 2-phospholactate and 4-PE. 4-PE is a putative side product of the glycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase, and its accumulation inhibits the pentose phosphate pathway enzyme, 6-phosphogluconate dehydrogenase (6-PGD). Inhibition of 6-PGD by 4-PE leads to an unexpected feedback response that includes increased flux into the pentose phosphate pathway as a result of partial inhibition of upper glycolysis, with concomitant increased sensitivity to antimalarials that target pathways downstream of glycolysis. These results highlight the role of metabolite detoxification in regulating central carbon metabolism and drug sensitivity of the malaria parasite.IMPORTANCE The malaria parasite has a voracious appetite, requiring large amounts of glucose and nutrients for its rapid growth and proliferation inside human red blood cells. The host cell is resource rich, but this is a double-edged sword; nutrient excess can lead to undesirable metabolic reactions and harmful by-products. Here, we demonstrate that the parasite possesses a metabolite repair enzyme (PGP) that suppresses harmful metabolic by-products (via substrate dephosphorylation) and allows the parasite to maintain central carbon metabolism. Loss of PGP leads to the accumulation of two damaged metabolites and causes a domino effect of metabolic dysregulation. Accumulation of one damaged metabolite inhibits an essential enzyme in the pentose phosphate pathway, leading to substrate accumulation and secondary inhibition of glycolysis. This work highlights how the parasite coordinates metabolic flux by eliminating harmful metabolic by-products to ensure rapid proliferation in its resource-rich niche.

Keywords: CRISPR; Plasmodium falciparum; antimalarial; fosmidomycin; glycolysis; isoprenoid; metabolic regulation; metabolism; metabolomics; parasitology; pentose.

FIG 1

PGP is required for normal growth of P. falciparum asexual stages, and Δpgp mutant parasites selectively accumulate metabolites in the PPP and two nonstandard metabolites. (A) Fluorescence microscopy of live PGP-GFP-infected RBCs. Infected RBCs were labeled with DAPI and visualized by differential interference contrast (DIC) and fluorescence microscopy. GFP fluorescence was present throughout the cytoplasm. (B) The asexual growth of Δpgp strain- versus wild-type (WT)-infected RBCs was monitored daily over a 13-day period by flow cytometry following SYTO61 labeling. Data are presented as the mean ± standard errors of the mean (SEM) cumulative parasitemia normalized to the WT day 13 data point (100%) from three independent experiments. Statistical significance was determined using a paired Student's t test at day 13 (*, P < 0.05; ***, P < 0.001). (C) Intracellular metabolite profiling of WT- and Δpgp mutant-infected RBCs by LC-MS. Individual metabolites are plotted as fold change (log₂) versus –log₁₀(P) for the Δpgp mutant parasites compared to WT parasites. Annotated metabolites were verified using standards, with the exception of gluconate-X-P (where the position of the P is unknown). Ribose-5-P and ribulose-5-P coelute under the chromatography conditions used. Two peaks of unknown identity, m/z 168.9909 and m/z 214.9968, also increased in the mutant line. (D) WT- and Δpgp mutant-infected RBCs were incubated with different concentrations of l- or d-lactate, and intracellular levels of 2-phospholactate were measured by GC-MS. Data are presented as fold change versus the WT (no lactate) condition. Data are presented as the means ± SEM from three independent experiments performed on different days. Statistical significance was determined using an unpaired Student's t test for the condition of WT with no lactate versus Δpgp mutant (**, P < 0.01), and one-way analysis of variance (ANOVA) was used for all lactate stimulation conditions (*, P < 0.05; **, P < 0.01; all other comparisons were nonsignificant). (E) Structure of 2-phospho-d-lactate (C₃H₇O₆P).

FIG 2

2-Phospho-d-lactate is generated via the glyoxalase pathway. (A) Schematic of the methylglyoxal detoxification system in P. falciparum. GloI and cGloII catalyze the conversion of methylglyoxal to d-lactate. P. falciparum also expresses a GloI-like protein (GILP) and tGloII, which are both targeted to the apicoplast. (B) In vitro production of d-lactate in ΔgloI parasites versus WT parasites. Data are presented as the percentage of d-lactate production normalized to the WT 60-min time point (100%) after experimental background subtraction. Data are presented as the means ± SEM from three independent experiments performed on different days, and statistical significance was determined using an unpaired Student's t test at the 60-min time point (***, P < 0.001). (C) WT- and ΔgloI-infected RBCs were incubated with 0 or 1 mM methylglyoxal (MG), and the intracellular abundance of 2-phospholactate was measured by GC-MS. Data are presented as fold change relative to the condition of WT with 0 mM methylglyoxal. Data are presented as the means ± SEM from three independent experiments performed on different days. (D) Extracellular d-lactate excretion was measured in uninfected RBCs (uRBC) and WT-infected RBCs (iRBC) after sample deproteination using a d-lactate plate assay (Cayman Chemicals). Cultures were set at 2.5% hematocrit (and 1.8% to 4% parasitemia for infected RBC). Medium was collected after 26 to 29 h of culture (ring to trophozoite stages for infected RBC). Data are presented as the means ± SEM from four independent repeats collected on different days, and statistical significance was determined using an unpaired Student's t test (*, P < 0.05).

FIG 3

Loss of PGP leads to inhibition of 6-PGD and enhanced flux through the oxidative PPP. (A) Chemical structure of 4-PE (C₄H₉O₈P) and GC-MS confirmation of 4-PE in P. falciparum. (B) Synthetic 4-PE was added to lysates of saponin-purified trophozoites, together with 6-phosphogluconate (6-PG), and the activity of 6-PGD was assessed by measurement of production of ribulose-5-P by GC-MS. Data are presented as percentages of the no 4-PE condition, with means ± SEM from four to five independent experiments performed on different days, and statistical significance was determined using one-way ANOVA in comparison to the condition of 0 mM 4-PE (**, P < 0.01; ****, P < 0.0001). (C) Schematic of the pentose phosphate pathway (PPP) in P. falciparum. Where relevant, carbon backbones are presented as gray (unlabeled) or red (¹³C-labeled) circles. The relative activity of the oxidative and nonoxidative PPP was monitored with 1,2-¹³C₂-glucose incorporation into ribose-5-P, with one and two labeled carbons corresponding to oxidative and nonoxidative PPP activity, respectively. (D and E) The ribose-5-P pool sizes (D) and percent label into ribose-5-P following 1,2-¹³C₂-glucose labeling (E). M + 1 represents the fraction of ribose-5-P derived from the oxidative arm, M + 2 represents the contribution of the nonoxidative arm, and M + 3 comprises both oxidative and nonoxidative labeling. Data are presented as the means ± SEM from three independent experiments performed on different days. Statistical significance was determined using unpaired t testing (ns, nonsignificant; **, P < 0.01; ***, P < 0.001). (F) The percent label into 6-phosphogluconate (6-P-gluconate) following 30 min of ¹³C₆-glucose incorporation (1:1 ¹²C:¹³C). Data are presented as the means ± SEM from three independent experiments performed on different days.

FIG 4

6-PGD disruption leads to glycolytic inhibition and increased oxidative PPP flux. (A) Inducible disruption of 6-PGD via DiCre-dependent gene excision and glmS ribozyme-mediated transcript decay. (B) Confirmation of 6-PGD depletion by Western blotting following rapamycin (Rap; 100 nM) and glucosamine (GlcN; 2.5 mM) addition. HA-tagged 6-PGD was detected at the expected size (anti-HA), and anti-P. falciparum BiP was used as a loading control. Indicated sizes are in kilodaltons. (C) Cell viability of DiCre-6-PGD and the parental DiCre lines in the presence or absence of rapamycin (100 nM) and glucosamine (2.5 mM), over four replication cycles. Flow cytometry was performed using SYTO61 labeling, and the data are presented as the treated/untreated ratio, with means ± SEM from three independent experiments. (D) Metabolic phenotype of 6-PGD depletion in trophozoite-stage infected RBCs. Metabolite pools are presented as mean arbitrary ion counts (±SEM) from three independent experiments.

FIG 5

Loss of PGP leads to increased sensitivity to fosmidomycin. (A) DHAP and phosphoenolpyruvate, derived from glycolysis, are imported into the apicoplast for synthesis of isoprenoids via the 1-deoxy-d-xylulose 5-phosphate/2-C-methyl-d-erythritol 4-phosphate pathway. Fosmidomycin is a competitive inhibitor of the first committed enzyme in this pathway, and its efficacy is decreased when glycolytic flux is increased (26). (B) Synchronized WT- and Δpgp-infected RBCs were treated with increasing concentrations of fosmidomycin for 72 h, and the growth percentage was determined by flow cytometry following SYTO61 labeling. Data are presented as the means ± SEM from three independent experiments performed on different days.

FIG 6

PGP-mediated metabolite repair maintains flux through glycolysis and the pentose phosphate pathway. PGP dephosphorylates 2-phospholactate (produced by an unspecified kinase[s] or a putative by-product of pyruvate kinase) and 4-phosphoerythronate (4-PE; a putative by-product of GAPDH). Loss of PGP leads to accumulation of 4-PE and partial inhibition of 6-PGD, with concomitant accumulation of 6-phosphogluconate and inhibition of enzymes in upper glycolysis. The PGP metabolic phenotype can be recapitulated by partial knockdown of 6-PGD. Abbreviations: glucose-6-phosphate isomerase (PGI), pyruvate kinase (PK), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 6-phosphogluconate dehydrogenase (6-PGD), glyoxylase I (GloI), phosphoglycolate phosphatase (PGP).