An apicoplast-resident folate transporter is essential for sporogony of malaria parasites

Abstract

Malaria parasites are fast replicating unicellular organisms and require substantial amounts of folate for DNA synthesis. Despite the central role of this critical co-factor for parasite survival, only little is known about intraparasitic folate trafficking in Plasmodium. Here, we report on the expression, subcellular localisation and function of the parasite's folate transporter 2 (FT2) during life cycle progression in the murine malaria parasite Plasmodium berghei. Using live fluorescence microscopy of genetically engineered parasites, we demonstrate that FT2 localises to the apicoplast. In invasive P. berghei stages, a fraction of FT2 is also observed at the apical end. Upon genetic disruption of FT2, blood and liver infection, gametocyte production and mosquito colonisation remain unaltered. But in the Anopheles vector, FT2-deficient parasites develop inflated oocysts with unusual pulp formation consisting of numerous single-membrane vesicles, which ultimately fuse to form large cavities. Ultrastructural analysis suggests that this defect reflects aberrant sporoblast formation caused by abnormal vesicular traffic. Complete sporogony in FT2-deficient oocysts is very rare, and mutant sporozoites fail to establish hepatocyte infection, resulting in a complete block of parasite transmission. Our findings reveal a previously unrecognised organellar folate transporter that exerts critical roles for pathogen maturation in the arthropod vector.

Keywords: Plasmodium; apicoplast; folate; malaria; membrane transport protein; sporogony.

Figure 1. BT1-family transport proteins in Plasmodium.

(a) Phylogenetic analysis reveals duplication of an ancestral folate transporter in apicomplexan parasites. Shown is a maximum likelihood tree of 38 biopterin transporter 1 (BT1) family proteins from apicomplexan parasites, related Chromalveolata and from additional representative eukaryotic species, as well as one related protein of the major facilitator superfamily (MFS) of bacterial origin. Open circles, nodes with bootstrap values of more than 90%. Closed circles, nodes with values of 65–90%. The Plasmodium folate transporters FT1 (blue) and FT2 (green) are highlighted. (b) Structure homology modelling of Plasmodium berghei FT2 indicates the signature MFS architecture, characterised by two separate domains (BT1 domains) surrounding a translocation centre, each containing six transmembrane helices. Shown are top (top) and side views (bottom). The phospholipid bilayer is indicated by grey dashed lines. The model was obtained with I-TASSER (Yang & Zhang, 2015) using the Escherichia coli YajR transporter structure (PDB ID: 3wdoA; Jiang et al., 2013) as the homology template. Coverage, 0.875; Z-score, 3.70; C-score, −1.12; TM-score, 0.832

Figure 2. FT2 is an apicoplast protein and broadly expressed during Plasmodium life cycle progression.

(a, b) Expression of FT2 throughout the Plasmodium berghei life cycle. ft2-tag_apiGFP parasites were imaged live at different stages. Shown are representative images of blood stages (a) and mosquito and liver stages (b), including the fluorescent signals of tagged FT2 (red, first row), the apicoplast marker api-GFP (green, second row) a merge of both signals (third row) as well as a merge of differential interference contrast images (DIC) with Hoechst 33342 nuclear stain (DNA, blue, fourth row). Note that a fraction of FT2 is observed in non-apicoplast structures in merozoites, sporulating oocysts and sporozoites (white arrowheads). Yellow arrowheads, apical prominence in merozoites. Extra-parasitic red fluorescence during the liver stage is due to autofluorescence of host cell vesicular compartments and was also observed in non-fluorescent WT parasites and uninfected host cells. (c) FT2 shows dual localisation to the apicoplast and to the apical end in sporozoites. Shown is a sporozoite recorded during motility, including the signal of tagged FT2 (top) and a merge of DIC with Hoechst 33342 nuclear stain (DNA, blue, bottom). Green arrowhead, apicoplast-localised FT2. Yellow arrowhead, FT2 at sporozoite apex. The dual localisation pattern is representative of all analysed sporozoites. n = 90 sporozoites from two independent feeding experiments. Bars, 5 μm for all except oocysts and liver stages (10 μm)

Figure 3. Efficient blood propagation and sexual differentiation in the absence of FT2.

(a) Replacement strategy for the deletion of Plasmodium berghei FT2. The locus was targeted with a replacement plasmid containing the 5′ and 3′ flanking regions, an expression cassette for high-level cytoplasmic GFP fluorescence (green), and the drug-selectable hDHFR-yFcu cassette (blue). Shown are the WT locus (top), the transfection vector (middle) and the recombined locus (bottom). WT-specific and integration-specific (INT) primer combinations are indicated by arrows and expected fragments by dotted lines. (b) Diagnostic PCRs of the WT locus (top) and of the drug-selected and isolated ft2⁻ parasites (bottom), using the primer combinations depicted in (a). Note the generation of a second independent FT2 loss-of-function mutant expressing fluorescent markers of the apicoplast and mitochondrion (see Figure S2a,b). (c, d) Efficient parasite proliferation in the murine bloodstream. (c) Intravital competition assay. Equal numbers of mCherry-fluorescent Berred WT and GFP-fluorescent ft2⁻ blood stage parasites were co-injected intravenously into mice and peripheral blood was analysed daily by flow cytometry (Matz, Matuschewski, & Kooij, 2013). Mean values (±SD) are shown. n.s., non-significant; two-way ANOVA; n = 3. (d) Serial passage of infected blood reveals no significant fitness loss of asexual ft2⁻ blood stage parasites. Co-infected blood, as described in (c), was serially transferred into naïve mice after 1 week of infection and the ratio of WT and ft2⁻ blood stages was determined by flow cytometry. Mean values (±SD) are shown. n.s., non-significant; one-way ANOVA; n = 3. (e–g) Normal sexual differentiation in the absence of FT2. 10⁷ WT or ft2⁻ blood stage parasites were injected intravenously into mice and peripheral blood was analysed 3 days later for gametocyte conversion (e), exflagellation (f) and ookinete conversion (g). Shown are the percentages of mature gametocytes among all blood stages, as identified in Giemsa-stained thin blood films, exflagellation centres per microscopic field and the percentage of fully matured ookinetes among in vitro cultivated P28-positive parasites. Shown are individual data points as well as mean values (bars). n.s., non-significant; Student’s t test; n ≥ 6

Figure 4. Prominent swelling and cytoplasmic malformations in FT2-deficient oocysts.

(a) Normal mosquito midgut colonisation in the absence of FT2. Shown are oocyst numbers of individual infected midguts and the mean infection intensity (bars) on Day 10 after the blood meal. n.s., non-significant; Student’s t test; n ≥ 46 midguts from three independent blood meals. (b, c) Oocyst sporulation is severely impaired upon loss of FT2. Sporulating oocysts were quantified (b) and imaged by differential interference contrast (DIC) (c) on Day 17 after the blood meal. ***, p < .01; Student’s t test; n = 4 independent blood meals. Percentages indicate the mean sporulation frequency and are derived from analysis of >600 oocysts from >30 infected midguts each. (d, e) Marked swelling of FT2-deficient mature oocysts. Infected midguts were imaged live 10 (d) and 17 days after the blood meal (e). Shown is the area occupied by individual oocysts in live fluorescence micrograph. n.s., non-significant; *, p < .05; Student’s t test; n ≥ 254 oocysts from four independent blood meals. (f, g) FT2-deficient oocysts form large intracellular malformations. (f) Shown is the percentage of oocysts containing intraparasitic accumulations 10 and 17 days after the blood meal. n.s., non-significant; ***, p < .01; Student’s t test; n = 4 independent blood meals. Percentages are derived from analysis of >600 oocysts from >30 infected midguts each. (g) Depicted are differential interference contrast images of 17 days old oocysts, showing various stages of either sporulation (WT) or cytoplasmic clumping (ft2⁻). White arrowheads, sporoblast; yellow arrowheads, intraparasitic accumulations. (h, i) Intracellular malformations displace cytoplasmic GFP and several parasite organelles in FT2-deficient oocysts. (h) Shown are images of a sporulating WT oocysts at Day 14 (top) and ft2⁻ oocysts on Days 17 and 22 after the blood meal (bottom), including the fluorescent signal of cytoplasmic GFP (green, first column), Hoechst 33342 nuclear stain (DNA, blue, second column), DIC images (third column), and two distinct merges of DIC with either GFP or DNA (last two columns). Note the presence of cytoplasmic GFP within the sporoblast (white arrowhead) and the absence thereof from the intraparasitic accumulations (yellow arrowhead). Also note the complete vesiculation of the ft2⁻ oocysts at Day 22. (i) Shown are 17 days old ft2⁻_{mito-mCh/api-GFP} oocysts at different stages of cytoplasmic clumping, including the individual signals of api-GFP (green, first column), mito-mCherry (red, second column), Hoechst 33342 nuclear dye (DNA, blue, third column), DIC images (fourth column) and a merge of all channels (fifth column). Note the displacement of apicoplast, mitochondrion and nuclei by the intraparasitic accumulations (yellow arrowheads). Bars, 10 μm

Figure 5. FT2-deficient oocysts accumulate single-membrane vesicles. Shown are transmission electron micrographs of 17 days old oocysts.

(a) ft2⁻ oocysts showing four stages of vesiculation. Initially, vesicles emerge from the oocyst periphery (1) and then form large congregations, which frequently associate with ducts of similar electron density, that appear to emerge from the oocyst periphery (2, 3). Some vesicles appear to fuse to form large cavities covered in single-membrane vesicles (4, 5). Upon progressing vesiculation, vesicles are predominantly found in the oocyst periphery, leaving only a central portion of cytoplasm unaffected (6). Red arrowhead, vesicle accumulation; yellow arrowheads, duct. (b) Occasional budding of malformed ft2⁻ sporozoites. Shown are sporulating WT (left) and ft2⁻ oocysts (right). Note that vesiculation also occurs in sporulating ft2⁻ oocysts. Budding ft2⁻ sporozoites showed signs of membrane deformation not commonly observed in WT. Yellow arrowhead, budding sporozoite apex; red arrowhead, membrane deformation. The light pink pseudo coloration highlights vesicle formations. Bars, 2 μm. (c) Quantification of salivary gland-associated sporozoites 21 days after the blood meal. Depicted are averaged sporozoite burdens from ~50 mosquitoes per feeding as well as mean values (bars). ***, p < .001; Student’s t test; n = 4 independent blood meals. (d) Proposed stages of oocyst development in WT parasites (left) or in the absence of FT2 (right). Upon commencing sporogony, vesicles are trafficked to the oocyst periphery to supply membrane for sporoblast formation, thereby promoting sporozoite production in WT parasites. In ft2⁻ oocysts, vesicles fail to undergo coordinated fusion, thus leading to the formation of vesicular patches, which associate with the periphery via membranous ducts and ultimately fuse. Continued trafficking of vesicles to the ft2⁻ oocyst periphery leads to the concentration of cytoplasm and parasite organelles in the oocyst centre

Figure 6. FT2 is required for formation of infectious sporozoites and dispensable for liver stage maturation.

(a) Sporozoites originating from FT2-deficient oocysts do not elicit malaria. Kaplan–Meier analysis of time to development of patent blood infection. Naïve mice were intravenously injected with 10,000 WT or ft2⁻ sporozoites, or were subjected to bites by mosquitoes that had received blood meals from mice infected with WT or ft2⁻ parasites. Peripheral blood was monitored daily by microscopic analysis of Giemsa-stained thin blood films. ***, p < .001; Mantel–Cox test; n = 3 mice from two independent feeding experiments. (b) Sporozoites originating from FT2-deficient oocysts are non-invasive. Huh-7 cells were inoculated with an equal mix of mCherry-fluorescent Berred WT and GFP-fluorescent ft2⁻ sporozoites obtained from independent preparations. Shown are percentages of WT and ft2⁻ liver stage parasites as quantified by live fluorescence microscopy 48 hr after inoculation. The dashed line represents the 50:50 ratios expected in the case of normal ft2⁻ invasion. n = 509 parasites from two independent infection experiments. Note that only once a single GFP-fluorescent liver stage was observed. (c–g) Phenotypic rescue during mosquito infection restores ft2⁻ sporozoite infectivity and reveals redundant functions of FT2 during liver stage development. Mosquitoes were fed on mice co-infected with mCherry-fluorescent Berred WT and GFP-fluorescent ft2⁻ parasites, leading to cross-fertilisation and phenotypic rescue during mosquito infection. (c) Kaplan–Meier analysis of time to development of patent blood infection. Naïve mice were intravenously injected with 30,000 sporozoites isolated from salivary glands of WT x ft2⁻-infected mosquitoes and peripheral blood was monitored daily by flow cytometry. n.s., non-significant; Mantel–Cox test; n = 9 mice from three independent feeding experiments. (d) Representative dot plot obtained by flow cytometry on Day 3 after transmission. Double fluorescent parasites are due to inter-chromosomal recombination events during mosquito infection and denote FT2-deficient parasites. (e) Quantification of FT2-deficient liver stages in vitro and emerging blood stages in vivo after inoculation with the mixed sporozoite population. Expected values reflect the fractions in the case of unaltered parasite fitness and are based on the percentage of observed double-fluorescent sporozoites. Shown are mean values (±SD). n.s., non-significant; one-way ANOVA and Tukey’s multiple comparison test; n = 3 independent feeding experiments, including data from 1.269 sporozoites, 2.643 liver stages and 751 first-generation blood stage parasites. (f) Representative images of in vitro cultivated WT and ft2⁻ liver stages at different times after infection with sporozoites isolated from WT x ft2⁻-infected mosquitoes. Shown is a merge of cytoplasmic fluorescence (red, WT; green, ft2⁻), Hoechst 33342 nuclear stain (blue) and differential interference contrast images. Values represent the mean area occupied by liver stages in fluorescence micrographs. n.s., non-significant; Student’s t test; n ≥ 150 liver stages from three independent experiments. Bar, 10 μm. (g) Diagnostic PCR of genomic parasite DNA isolated from the blood of a co-infected donor mouse prior to mosquito feeding (top) and from a recipient mouse after transmission (bottom). Primer combinations are as indicated in Figure 3a