Conditional Degradation of Plasmodium Calcineurin Reveals Functions in Parasite Colonization of both Host and Vector

Abstract

Functional analysis of essential genes in the malarial parasite, Plasmodium, is hindered by lack of efficient strategies for conditional protein regulation. We report the development of a rapid, specific, and inducible chemical-genetic tool in the rodent malaria parasite, P. berghei, in which endogenous proteins engineered to contain the auxin-inducible degron (AID) are selectively degraded upon adding auxin. Application of AID to the calcium-regulated protein phosphatase, calcineurin, revealed functions in host and vector stages of parasite development. Whereas depletion of calcineurin in late-stage schizonts demonstrated its critical role in erythrocyte attachment and invasion in vivo, stage-specific depletion uncovered roles in gamete development, fertilization, and ookinete-to-oocyst and sporozoite-to-liver stage transitions. Furthermore, AID technology facilitated concurrent generation and phenotyping of transgenic lines, allowing multiple lines to be assessed simultaneously with significant reductions in animal use. This study highlights the broad applicability of AID for functional analysis of proteins across the Plasmodium life cycle.

Graphical abstract

Figure 1

Generation of a Functional AID System in Plasmodium berghei to Examine Calcineurin Function (A) Auxin promotes interaction of TIR1 (an F box protein, in green) with the AID degron tagged protein (blue). The AID-tagged protein (red) is recruited to the Skp, Cullin, F-box-containing complex (SCF), a multi-protein E3-ligase complex, resulting in ubiquitination and degradation of the target protein. Schematic adapted from Nishimura et al. (2009). (B) Expression and localization of PbCnA-AID-HA at the indicated stages of Plasmodium life cycle. Fixed and permeablised parasites were probed with indicated primary antibodies. Scale bar, 5 μm. (C) Robust and efficient depletion of PbCnA-AID-HA, upon addition of auxin in both schizonts and gametocytes, as measured by western blotting. Enolase serves as a loading control. (D) Conditional depletion of PbCnA-AID-HA is reliant on auxin, TIR1, and the proteasome. PbCnA-AID protein levels in a non-TIR1 background is resistant to auxin-mediated depletion (left panel). Pre-incubation with proteasome inhibitor 1 μM MG132 for 1 hr blocks PbCnA-AID depletion by auxin (right panel), as shown by western blotting. Enolase serves as a loading control. See also Figure S1.

Figure 2

Calcineurin Regulates Erythrocyte Invasion by the Merozoite and Male Gametogenesis (A) Sustained (24.5 hr) auxin treatment from the ring stage has no effect on the number of merozoites produced in PbCnA-AID parasites (25 segmented schizonts counted per condition, n = 4 experiments). (B) Pre-incubation of mature PbCnA-AID schizonts with auxin for 30 min results in 86% reduction of in vivo erythrocytic invasion by merozoites as measured by flow cytometry (n = 3 experiments). (C) AMA1 processing in free merozoites (± auxin) determined by exposing merozoites to extracellular conditions ([ec]; high Na⁺ and low K⁺) versus intracellular conditions ([ic]; low Na⁺ and high K⁺) and measured by western blotting. Both supernatant (S) and pellet (P) fractions were probed with mAb28G2, which recognizes the C terminus of AMA-1. (D) Merozoite attachment to erythrocytes is regulated by PbCnA-AID. Isolated merozoites (± auxin; ± CytD) were incubated with erythrocytes under shaking conditions. After 10 min, cells were fixed and merozoite attachment/invasion was assessed by microscopy (n = 3 experiments with at least ten fields each containing ∼350 erythrocytes per experimental condition). See also Figures S2C–S2E. (E) Gametocyte production is unaffected by PbCnA-AID depletion. Synchronized ring-stage parasites were treated with auxin (15 min post-invasion) and grown ex-vivo, and gametocytemia was determined 32 hr later (n = 4 experiments). (F) Gametocyte emergence from erythrocytes is unaffected by PbCnA-AID depletion. Gametocytes were pre-treated with auxin for 45 min and subsequently activated for gametogenesis by addition of RPMI + 100 μM xanthurenic acid and a drop in temperature from 37°C to 21°C (n = 2 experiments with 100 gametocytes counted per experimental condition). (G) Reduction in the number of male gametocytes replicating their DNA with PbCnA-AID depletion. Proportion of male gametocytes undergoing DNA replication was determined at 0 and 8 min post-activation and is expressed as a percentage of the non-auxin-treated control line at 8 min (n = 3 experiments). Control lines (Con) are OsTIR1 expressing with an unmodified pbcna locus (pG230). For all panels, auxin = 500 μM IAA, mean ± SEM; two-tailed t test for paired observations: [^∗], p < 0.05; [^∗∗], p < 0.01; and [^∗∗∗],p < 0.001. See also Figure S2.

Figure 3

Calcineurin Regulates Gamete Fertilization and Host Cell Colonization of both Ookinetes and Sporozoites (A) PbCnA-AID depletion in gametocytes resulted in 90% reduction in ookinetes formed. Conversion rate is reported as the percentage of female gametes forming ookinetes (n = 4 experiments). (B) Following PbCnA-AID depletion prior to gametocyte activation, ookinete micronemal and motor proteins, known to be expressed post-fertilization, become nearly undetectable by western blot. Enolase serves as the loading control. (C) Ookinete motility (measured by distance covered over time) is unaffected by PbCnA-AID depletion (n = 21 instances per condition; bottom and top of box denote first and third quartiles, respectively; whiskers denote minimum and maximum; p = 0.9 for PbCnA-AID line ± auxin, 0.6 for Con line ± auxin). (D) Secretion of the micronemal proteins CTRP and chitinase into the supernatant is unaffected by PbCnA-AID depletion in mature ookinetes. (E) Oocyst numbers in mosquito midgut upon PbCnA-AID depletion in mature ookinetes. Auxin was added to parasite cultures 6 hr post-induction of gametogenesis. Mature ookinetes were fed to mosquitoes, and 7 days later, midguts were dissected and GFP-positive oocysts were counted (n = minimum of 50 mosquitoes for each condition; mean ± SD). Also see independent repeat in Figure S3B. (F) Sporozoite invasion was examined by incubating mosquito salivary gland sporozoites with auxin (90 min), followed by addition to HepG2 hepatocytes. Sporozoite invasion was assessed 2 hr later and calculated as the proportion of intracellular sporozoites to total sporozoites (intracellular + extracellular). For both control and experimental line, 100% is the proportion of intracellular sporozoites without auxin treatment (n = 2 experiments). (G) Incubation with auxin (90 min) of mosquito salivary gland sporozoites from a GFP-expressing version of PbCnA-AID parasites reduces EEFs in HepG2 hepatocytes by 46%. For both control and experimental line, 100% is the number of GFP-positive EEFs without auxin treatment (n = 3 experiments). For all panels, auxin denotes 500 μM IAA; control line (Con) is the TIR1-expressing parent (pG230 for [A] and [B); pG402 for [C]–[G]); mean ± SEM; two-tailed t test for paired observations: [^∗∗], p < 0.01; [^∗∗∗∗], p < 0.0001. See also Figure S3.

Figure 4

Conditional Protein Depletion Demonstrates CDPK1 and PPKL Are Dispensable for Asexual Growth but Crucial for Ookinete Development (A) Robust depletion of both CDPK1 and PPKL protein with C-terminal AID-HA epitope tags, upon addition of auxin for 45 min in schizonts, as measured by western blotting. Enolase serves as the loading control. (B) Continuous treatment with auxin (24.5 hr) from ring stage has no effect on schizogony in both PbCDPK1-AID and PbPPKL-AID parasite lines. Scale bar, 5 μm. (C) Pre-incubation of mature PbCDPK1-AID or PbPPKL-AID schizonts with auxin for 45 min had no effect on erythrocytic invasion by merozoites as measured by flow cytometry (n = 3 experiments) (D) Robust depletion of both CDPK1 and PPKL protein with C-terminal AID-HA epitope tags, upon addition of auxin for 45 min in gametocytes, as measured by western blotting. Enolase serves as the loading control. (E) PbCDPK1-AID or PbPPKL-AID depletion in gametocytes resulted in abnormal ookinete formation. Zygote-to-ookinete development was visualized by probing against the surface protein marker, P25 (in red, DAPI: blue). (F) Proportion of spherical/retort and normal ookinete forms observed upon PbPPKL or PbCDPK1 depletion. Gametocytes were pre-incubated with auxin for 45 min followed by induction of gametogenesis. Zygote-to-ookinete conversion was determined by examining and counting cells stained with α-P25 antibody. Scale bar, 5 μm. (G) CDPK1 regulates protein levels of translationally controlled genes during zygote-to-ookinete development. Gametocytes were incubated with auxin for 45 min prior to induction of gametogenesis. 24 hr later, parasite pellets were examined by western blotting. While expression of P-28 was unaffected in PbCDPK1-AID-depleted parasites, WARP, CTRP, and Chitinase levels were significantly reduced. Enolase serves as loading control. See also Figure S4.

Figure 5

Multiplex Generation and Phenotyping of Transgenic Lines (A) Schematic of vectors generated for simultaneous transfection and generation of multiple transgenic lines. In addition to C-terminally tagging the gene of interest with AID-2xHA, the plasmid expresses the drug selectable marker (hdhfr) and a fluorescence marker (gfp, cfp, or mcherry) driven by the bidirectional pbeef1α promoter. (B) Fluorescent parasites indicate successful generation of three transgenic lines 8 days post-transfection. (C) Graph of FACS where gates indicate the three types of collected fluorescence-positive cells. Fifty IRBCs of each fluorescence type were intravenously administered to three naive mice to generate GFP-, CFP-, and mCherry-expressing isogenic parasite lines, respectively. (D) Schematic of duplexed in vivo invasion assay where CFP-expressing PbCDPK1-AID schizonts are mixed with GFP-expressing PbCnA-AID line prior to intravenous injection. 15 min later, four to five blood drops are collected by tail prick and cultured for 12 hr followed by flow cytometry analysis. The gated (G) DiCycle ruby positive (infected IRBC) can be further distinguished into either singly or dually fluorescent populations to determine parasitemia. (E) Duplexed in vivo invasion assay where CFP-expressing PbCDPK1-AID and GFP-expressing PbCnA-AID (± auxin) are mixed and intravenously injected (mean ± SEM; two tailed t test for paired observations: p < 0.05; [^∗∗]). (F) Schematic of duplexed ookinete motility assay where mCherry-expressing PbDozi-AID ookinetes are mixed with GFP expressing PbCnA-AID line in Matrigel. Reconstructed tracks of motile ookinetes with frames collected every 10 s for 15 min (right panel). (G) Ookinete motility (measured by distance covered over time) was measured for mixed red PbDozi-AID and green PbCnA-AID ookinetes (± auxin) (n = 15 instances per condition; bottom and top of box denote first and third quartiles, respectively; whiskers denote minimum and maximum; p = 0.14 for PbCnA-AID line ± auxin and 0.35 for PbDozi-AID line ± auxin). Scale bar denotes 5 μm unless otherwise indicated. See also Figure S5.