Glutathione reductase-null malaria parasites have normal blood stage growth but arrest during development in the mosquito

Abstract

Malaria parasites contain a complete glutathione (GSH) redox system, and several enzymes of this system are considered potential targets for antimalarial drugs. Through generation of a gamma-glutamylcysteine synthetase (gamma-GCS)-null mutant of the rodent parasite Plasmodium berghei, we previously showed that de novo GSH synthesis is not critical for blood stage multiplication but is essential for oocyst development. In this study, phenotype analyses of mutant parasites lacking expression of glutathione reductase (GR) confirmed that GSH metabolism is critical for the mosquito oocyst stage. Similar to what was found for gamma-GCS, GR is not essential for blood stage growth. GR-null parasites showed the same sensitivity to methylene blue and eosin B as wild type parasites, demonstrating that these compounds target molecules other than GR in Plasmodium. Attempts to generate parasites lacking both GR and gamma-GCS by simultaneous disruption of gr and gamma-gcs were unsuccessful. This demonstrates that the maintenance of total GSH levels required for blood stage survival is dependent on either de novo GSH synthesis or glutathione disulfide (GSSG) reduction by Plasmodium GR. Our studies provide new insights into the role of the GSH system in malaria parasites with implications for the development of drugs targeting GSH metabolism.

FIGURE 1.

Generation of parasites lacking expression of GR (Δgr parasites). A, schematic representation of the disruption vector pL1282, the WT gr locus (numbers indicate the exons) and the locus after disruption. Lines within exon 3 represent the conserved domains of GR. Restriction sites (K, KpnI; X, XbaI) and the size of restriction fragments used for diagnostic Southern blots (see C) and the tgdhfr-selectable marker are indicated. This construct was used for generation of Δgr1 and -2 and deletes part of exon 3. B, schematic representation of the disruption vector pL1538, the WT gr locus, and the locus after disruption. Restriction sites (H, HindIII; Xb, XbaI) and size of restriction fragments used for diagnostic Southern blots (see D) and the yfcu-hdhfr-selectable marker are indicated. This construct was used for generation of Δgr4 and deletes the complete gr ORF. C, Southern analysis of separated chromosomes (left) and digested genomic DNA (right), confirming correct disruption of gr in Δgr1 and -2 parasites. Hybridization of Δgr chromosomes with the 3′dhfr probe recognizes the construct integrated into gr on chromosome 10, the endogenous dhft-ts gene on chromosome 7, and the integrated GFP construct on chromosome 3. The 3′gr probe hybridizes to the gr gene on chromosome 10 (right lane). Hybridization of digested DNA with the tgdhfr-ts and the gr ORF probes recognized the expected DNA fragments indicated in A. Genomic DNA from WT and Δgr parasites were digested with ClaI (C)/XhoI (Xh). D, Southern analysis of separated chromosomes (left) and HindIII/XbaI-digested genomic DNA (right), confirming correct disruption of gr in Δgr4 parasites. Hybridization of Δgr chromosomes with the 3′dhfr probe recognizes the construct integrated into gr on chromosome 10, the endogenous dhft-ts gene on chromosome 7, and the integrated GFP construct on chromosome 3. Hybridization of digested DNA with the 3′gr probe recognizes the expected DNA fragments indicated in B. E, Northern analysis of gr transcription. Blood stages RNA was hybridized with the 3′gr probe and as a loading control with the L644R probe recognizing the large subunit ribosomal RNA. In WT parasites, a transcript of ∼2.8 kb was detected. In the Δgr1 and Δgr2 parasites, a smaller transcript of ∼2.0 kb was detected, corresponding to exons 1 and 2 and part of exon 3, which remain in the genome after integration of pL1282 (see A). In Δgr4, no transcripts were detected. F, Western analysis of GR expression. Protein extracts of blood stages were reacted with a rabbit polyclonal antiserum against P. falciparum GR. In WT, the GR protein of 56 kDa is detected, which is absent in the Δgr parasites (see also supplemental Fig. S2). Antibody PbB5 is used as a loading control. Uninfected mouse red blood cells (MRBC) were run in each experiment.

FIGURE 2.

Phenotype of Δgr parasites; GR activity, GSH production, in vivo blood stage growth, and sensitivity to methylene blue and eosin B. A, GR activity in blood stage parasites as determined by monitoring the disappearance of NADPH in real time by a decrease in the absorbance at 340 nm in the absence (−) or presence of GSSG (+). Measurements were significantly different between WT, Δgr1, and Δgr2 parasites (p < 0.0001, t test with Bonferroni's correction for multiple comparisons). Bars, S.D. B, GSH levels in blood stages of WT, Δgr1, and Δgr2 parasites. One-way analysis of variance showed significant differences between WT and Δgr parasites (p < 0.0001; unpaired t test with the Welch correction for differences in variance). The median and 10th and 90th percentiles are shown. C, mice infected with Δgr and WT parasites show a comparable course of infection (parasitemia) in Swiss-CD1 mice infected with 1–2 × 10² parasites. The average parasitemia of groups of five mice is shown. Bars, S.D. D, the course of parasitemia in groups of five mice infected with Δgr or WT parasites and treated with MB for a period of 6 days is shown. Parasitemia in untreated mice infected with WT is shown as a control. No differences were found between the growth pattern of MB-treated WT and Δgr mutants. E, inhibition of the in vitro growth of blood stages MB (left) and eosin B (right). Inhibition of growth was determined by measuring the rate of DNA synthesis using flow cytometry (FC assay). EC₅₀ values are shown in the graphs. Error bars, S.D.

FIGURE 3.

Development of oocysts of Δgr parasites. A, light microscope pictures of oocysts of Δgr parasites in the midgut of A. stephensi mosquitoes at days 8 and 12 p.i., showing an aberrant morphology and reduced size. Scale bar, 25 μm. B, GFP-expressing oocyst of WT and Δgr parasites, showing reduced size, reduced GFP expression, and lower numbers of oocysts (see Table 1). Scale bar, 50 μm. C, day 2 and day 12 oocysts stained with antibodies against different oocyst proteins: Pbs21, a surface protein of ookinetes and young oocysts; PbCap380, a protein of the oocyst capsule; and PbCSP, a surface protein of sporozoites. Antibody-stained oocysts were detected by indirect fluorescence microscopy using Rhodamine Red^TM-X goat anti-mouse IgG (H + L) or Texas Red®-X goat anti-rabbit IgG (H + L) secondary antibodies (Invitrogen). Scale bar, 25 μm.

FIGURE 4.

Morphology of oocysts of Δgr parasites. Transmission electron microscope analysis of WT (a–c) and Δgr 2 (d–f) oocysts at day 6 postinfection of A. stephensi mosquitoes. WT oocysts show abundant ER, mitochondria (m), and large nuclei (n). Oocysts of Δgr parasites have a smaller size and show limited expansion of the ER, small nuclei with electron-lucent nucleoplasm, damaged mitochondria (m in a and inset), and large vacuoles (v). Scale bars, 2 μm.

FIGURE 5.

Importance of GSH enzymes in the Plasmodium parasite life cycle. A, analysis of mutants lacking expression of either GR or γ-GCS shows that neither the reduction of GSSG nor the de novo synthesis of GSH is essential for parasite blood stage multiplication (lower panels). It has been proposed that the intracellular blood stage parasites (gray circles) are able to take up and use GSH directly from the host erythrocyte (pink circles). Blood stage parasites lacking both γ-GCS and GR are not viable, indicating that blood stages cannot completely rely on host enzymes or alternative pathways of GSH reduction for their GSH metabolism (upper right panel). B, development of the extracellular oocyst stage in the mosquito is completely dependent on reduction and synthesis of GSH by its own enzymes. The black arrows represent the possible transport of GSH and GSSG in and/or out of the parasite.