Metabolism of host lysophosphatidylcholine in Plasmodium falciparum-infected erythrocytes

Abstract

The human malaria parasite Plasmodium falciparum requires exogenous fatty acids to support its growth during the pathogenic, asexual erythrocytic stage. Host serum lysophosphatidylcholine (LPC) is a significant fatty acid source, yet the metabolic processes responsible for the liberation of free fatty acids from exogenous LPC are unknown. Using an assay for LPC hydrolysis in P. falciparum-infected erythrocytes, we have identified small-molecule inhibitors of key in situ lysophospholipase activities. Competitive activity-based profiling and generation of a panel of single-to-quadruple knockout parasite lines revealed that two enzymes of the serine hydrolase superfamily, termed exported lipase (XL) 2 and exported lipase homolog (XLH) 4, constitute the dominant lysophospholipase activities in parasite-infected erythrocytes. The parasite ensures efficient exogenous LPC hydrolysis by directing these two enzymes to distinct locations: XL2 is exported to the erythrocyte, while XLH4 is retained within the parasite. While XL2 and XLH4 were individually dispensable with little effect on LPC hydrolysis in situ, loss of both enzymes resulted in a strong reduction in fatty acid scavenging from LPC, hyperproduction of phosphatidylcholine, and an enhanced sensitivity to LPC toxicity. Notably, growth of XL/XLH-deficient parasites was severely impaired when cultured in media containing LPC as the sole exogenous fatty acid source. Furthermore, when XL2 and XLH4 activities were ablated by genetic or pharmacologic means, parasites were unable to proliferate in human serum, a physiologically relevant fatty acid source, revealing the essentiality of LPC hydrolysis in the host environment and its potential as a target for anti-malarial therapy.

Keywords: fatty acid; lysophospholipase; lysophospholipid; malaria; serine hydrolase.

Fig. 1.

Serine hydrolase inhibitors block LPC hydrolysis in situ. (A) Fatty acids derived from LPC hydrolysis reduce C4,C9-FA incorporation into parasite neutral lipids. LPC or the non-hydrolysable LPC analog lyso-PAF were co-incubated with C4,C9-FA for 30 min and incorporation into DAG and TAG was quantified and normalized to a no-LPC control (set to 100). Means and SD are from three independent experiments. Significance of the differences in the 30 µM LPC and lyso-PAF means was assessed using an unpaired, two-tailed t-test with Welch’s correction for unequal variance (hereafter referred to as “Welch’s t-test”); P-values for DAG and TAG are indicated. (B) C4,C9-FA/LPC competition assay for identifying inhibitors of in situ LPC hydrolysis. BTC, BODIPY-TR-ceramide internal standard; TLC, thin layer chromatography. (C) Definition of an inhibition profile for in situ LPC hydrolysis. Inhibitors were evaluated in a C4,C9-FA/LPC competition assay at 10 µM. C4,C9-FA fluorescence intensities were normalized to a no-LPC, no inhibitor control (set to 100). Means and SD are from three independent experiments. Statistical significance (inhibitor vs. DMSO control) was assessed using Welch’s t-test; P-values below 0.05 are displayed above the bars. (D) Structure of AKU-010. (E) Inhibition of serine hydrolases in lysates of MACS-enriched, P. falciparum–infected erythrocytes as assessed by TAMRA-FP profiling. AA74-1 was included to suppress the strong signal from erythrocyte acylpeptide hydrolase (APEH), which was present as 80- and 55-kDa species (far left lane, black asterisks). Red asterisks indicate species with the inhibition profile AKU-010 > AKU-005 >> AKU-006, JW642. (F) TAMRA-FP profiling of uninfected erythrocytes. Inhibition of erythrocyte serine hydrolases was only observed with IDFP and AA74-1. Black asterisk, APEH. Sizes of markers are indicated in kDa.

Fig. 2.

Two exported serine hydrolases are targets of AKU-010. (A) TAMRA-FP profiling of MACS-enriched ΔXL1, ΔXL2, and 3D7 iRBCs. XL1 is indicated with a blue arrow. (B) TAMRA-FP profiling of MACS-enriched ΔXL2 and 3D7 iRBCs. Three AKU-010-inhibited species corresponding to XL2 are indicated with red asterisks. (C) XL1 and XL2 are exported to the host cell. Upper: Schematic for fractionation of MACS-enriched 3D7 parasites into saponin supernatant and pellet fractions. Lower: TAMRA-FP labeling reveals that XL1 (blue arrow) and XL2 (red asterisks) appear predominantly in the saponin supernatant containing soluble erythrocyte proteins, whereas the intracellular serine hydrolase PfPARE (black asterisk) appears exclusively in the parasite pellet. In A–C, sizes of protein standards are given in kDa. In C, XL1 migrates between the 100- and 150-kDa markers, similar to A. (D) Representative thin-layer chromatograms depicting the effects of LPC on incorporation of OA into parasite lipids in 3D7 and ΔXL1/2 parasites. The two DAG species are 1,2- and 1,3-isomers. BTC, BODIPY-TR-ceramide internal standard. Note the differential effect of LPC on OA labeling of PC in the two lines. (E) Oleate alkyne/LPC competition profiling of single and double XL1/2 knockout lines. Ratios below 1 (dotted line) indicate suppression of OA incorporation in the presence of LPC. Means and SD are from at least three independent experiments. Significance relative to 3D7 was assessed within lipid groups with Welch’s t-test. *, P < 0.05.

Fig. 3.

Exported XL2 and intraparasitic XLH4 govern the metabolism of exogenous LPC in asexual parasites. (A) TAMRA-FP profiling of MACS-enriched parental 3D7 and XLH4-YFP-expressing parasites. YFP tagging of XLH4 results in a mass increase of ~30 kDa. A ~270-kDa erythrocyte serine hydrolase, which is not inhibited by AKU-010 (Fig. 1F), is shown as a loading control. (B) A live parasitized erythrocyte exhibiting intraparasitic XLH4-YFP fluorescence. (C) Immunofluorescence localization of XLH4-HA. A 3D7 parasite imaged with identical parameters is shown as a negative control. In B and C: Hoechst 33342 fluorescence (DNA) is pseudocolored magenta. BF, bright-field. (Scale bar, 3 µm.) (D) TAMRA-FP profiling demonstrates the loss of XLH4 (green asterisk) in MACS-enriched double (ΔXL2/XLH4) and quadruple (QKO) knockout lines. Full gel images are shown in SI Appendix, Fig. S3G. (E) Oleate alkyne/LPC competition profiling of XLH4 single and XL/XLH multiple knockout lines. A ratio of 1 (dotted line) indicates no competition from LPC hydrolysis. Means and SD are from at least three independent experiments. Significance relative to 3D7 was assessed within lipid groups with Welch’s t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. (F) Enhanced OA labeling of PC in QKO parasites is not due to erythrocyte acyltransferase activity. Equivalent numbers of MACS-enriched QKO iRBC or uninfected erythrocytes (uRBC) were labeled with OA with or without LPC as indicated, in duplicate. Phospholipids were resolved by TLC. (G) TAMRA-FP profiling of MACS-enriched parasites reveals comparable XL2 expression in the complemented QKO parasite (QKO/XL2c) and parental 3D7 lines. The loading control is that described in A. (H) In vitro lysophospholipase activity in lysates of equivalent numbers of MACS-enriched iRBC or uRBC. Percent of TopFluor LPC hydrolysis is shown for serial threefold dilutions of lysates. The dotted line indicates 10% substrate hydrolysis. Means and SD are from three independent experiments.

Fig. 4.

Loss of XL2 and XLH4 activities impairs fatty acid scavenging from LPC, exacerbates LPC toxicity, and abrogates growth in serum. (A) QKO parasites have diminished ability to use LPC as a sole source of fatty acids. Ring-stage parasites were cultured for one generation in a minimal-lipid medium containing the indicated range of concentrations of each of LPC 16:0 and 18:1 (“2LPC”) or in standard culture medium containing 0.5% Albumax. Relative parasitemias were calculated by dividing the parasitemia at each 2LPC concentration by that in 0.5% Albumax. Means and SD are shown for four independent experiments. Significance relative to 3D7 was assessed for 15 to 25 μM concentrations with Welch’s t-test. *, P < 0.05; **, P < 0.01. (B) QKO parasites are hypersensitive to LPC toxicity. 3D7 and QKO parasites were cultured for 48 h in complete RPMI supplemented with LPC 18:1 (1.95 to 240 µM) and parasite growth was quantified using SYBR Green. Points were fitted to a four-parameter dose–response curve. Data are from two independent experiments. (C and D) Parasites lacking XL2 and XLH4 do not proliferate in medium containing human serum. Parasite lines were grown in RPMI medium supplemented with either 0.5% Albumax I (C) or 10% pooled human serum (D). Parasitemia was determined from >1,000 RBCs on Giemsa-stained smears. Cumulative parasitemia is the parasitemia on day X multiplied by fold-subculture up to that point. Where parasite growth was observed, data were natural-log transformed and fitted by linear regression. One of two independent experiments with similar results is shown. (E) AKU-010 inhibits the growth of 3D7 and ΔXL2 parasites in serum but not Albumax I. Ring-stage parasites were cultured for 96 h in media containing 10% human serum or 0.5% Albumax I and AKU-010 (2.2 nM to 5 µM). Parasite growth was quantified using SYBR Green. Means and SD are from three independent experiments. Serum data points are fitted to a four-parameter dose–response curve and Albumax (Alb) data points are shown with a connecting line. (F) AKU-010 sensitizes 3D7 parasites to LPC in Albumax-containing medium. Ring-stage parasites were cultured for 48 h with LPC 18:1 (7.5 to 240 µM), with or without 5 µM AKU-010. Parasite growth was quantified using SYBR Green. Means and SD are from three independent experiments. Points were fitted to a four-parameter dose–response curve.

Fig. 5.

LPC metabolism in P. falciparum and consequences of the loss of XL2 and XLH4 activities. (A) In wild-type parasites, LPC is hydrolyzed to fatty acids and glycerophosphocholine (GPC) either in the host erythrocyte by XL2 or within the parasite by XLH4. The liberated fatty acids (FA) serve as precursors for lipid synthesis while GPC is hydrolyzed to glycerol-3-phosphate (G3P) and choline (Cho) by glycerophosphocholine phosphodiesterase (GDPD). Exogenous fatty acids can also be taken up and incorporated into parasite lipids. Arrows indicate metabolic reactions with the corresponding enzymes in magenta and gray lines indicate transport processes. RBC, red blood cell; PV, parasitophorous vacuole. (B) Loss of XL2 and XLH4 greatly diminishes the rate of hydrolysis of LPC (a low level of hydrolysis is likely occurring but is not depicted). LPC accumulating in the infected erythrocyte may be directly acylated to PC through the activity of an LPC acyltransferase (LPCAT) with acyl-CoA serving as co-substrate. Exogenous free fatty acids serve as the primary FA source for parasite lipid synthesis.