The tyrosine transporter of Toxoplasma gondii is a member of the newly defined apicomplexan amino acid transporter (ApiAT) family

Abstract

Apicomplexan parasites are auxotrophic for a range of amino acids which must be salvaged from their host cells, either through direct uptake or degradation of host proteins. Here, we describe a family of plasma membrane-localized amino acid transporters, termed the Apicomplexan Amino acid Transporters (ApiATs), that are ubiquitous in apicomplexan parasites. Functional characterization of the ApiATs of Toxoplasma gondii indicate that several of these transporters are important for intracellular growth of the tachyzoite stage of the parasite, which is responsible for acute infections. We demonstrate that the ApiAT protein TgApiAT5-3 is an exchanger for aromatic and large neutral amino acids, with particular importance for L-tyrosine scavenging and amino acid homeostasis, and that TgApiAT5-3 is critical for parasite virulence. Our data indicate that T. gondii expresses additional proteins involved in the uptake of aromatic amino acids, and we present a model for the uptake and homeostasis of these amino acids. Our findings identify a family of amino acid transporters in apicomplexans, and highlight the importance of amino acid scavenging for the biology of this important phylum of intracellular parasites.

Fig 1. Phylogenetic analysis of ApiAT family proteins.

Consensus maximum likelihood tree of ApiAT family proteins. The tree was generated from a multiple sequence alignment of 66 putative ApiAT proteins from a range of apicomplexans, with 464 residues used in the analysis. Bootstrap values are depicted by black circles (>90% support), blue circles (>75–90% support), or white circles (55–75% support). The tree is unrooted. Abbreviations: Bb, Babesia bovis; Cp, Cryptosporidium parvum; Et, Eimeria tenella; Nc, Neospora caninum; Pb. Plasmodium berghei; Pf, Plasmodium falciparum; Tg, Toxoplasma gondii; Ta, Theileria annulate; ug, ungrouped.

Fig 2. Expression and localization analysis of T. gondii ApiAT family proteins.

(A-E) Western blots with anti-HA antibodies to measure the expression and molecular mass of tagged TgApiAT proteins in tachyzoites stages of the parasite. Western blots with antibodies against GRA8 and Tom40 were used to test for the presence of protein in samples where the HA-tagged TgApiAT protein was not detected. (F-I) Immunofluorescence assays with anti-HA antibodies to determine the localisation of HA-tagged TgApiAT proteins (green in merge). Samples were co-labelled with antibodies against the plasma membrane marker P30 (red in merge). TgApiAT3-3-HA-expressing parasites were co-transfected with the trans-Golgi network (TGN) marker Stx6-GFP [63], and labelled with anti-HA (red in merge), anti-P30 (plasma membrane, PM; blue in merge) and anti-GFP (green in merge) antibodies. All scale bars are 2 μm.

Fig 3. Genetic disruption of T. gondii ApiAT family proteins reveals the importance of TgApiAT2 and TgApiAT5-3 for parasite growth in vitro.

(A-F) Plaque assays depicting growth of disrupted TgApiAT strains and their corresponding parental WT strain. 150 parasites were added to wells of a 6-well plate and cultured for 9 days in DMEM. (A) WT (RHΔhxpgrt) and apiAT1^Δ54–534 parasites grown in DMEM (left) or RPMI (right). (B) WT (TATi/Tomato), apiAT2^Δ138–588 and apiAT2^Δ138–588 parasites complemented with a constitutively expressed TgApiAT2 (apiAT2^Δ138-588/cTgApiAT2). (C) WT (TATi) and apiAT3 sub-family mutants. (D) WT (TATi/Tomato) and apiAT5 sub-family mutants. (E) WT (TATi/Tomato) and apiAT6 sub-family mutants. (F) apiAT7 sub-family mutants. Note that the TATi/Tomato strain served as WT strain for the apiAT2, apiAT5, apiAT6, and apiAT7 sub-family mutants, and the identical image of the TATi/Tomato plaque assay is shown in B, D and E to facilitate interpretation of the data. All images are from the same experiment, and are representative of three independent experiments.

Fig 4. Analysis of [¹³C] amino acid uptake into WT and apiAT5-3^Δ188–504 parasites reveals a role for TgApiAT5-3 in amino acid homeostasis.

(A-B) Extracellular WT (TATi/Tomato) or apiAT5-3^Δ188–504 tachyzoites were incubated in medium containing [¹³C]-L-amino acids for 15 min. Polar metabolites were extracted and amino acid abundance (A) and levels of [¹³C]-amino acid enrichment (B) in WT (black) and apiAT5-3^Δ188–504 (red) tachyzoites determined by GC-MS. Only L-amino acids that could be detected in all experiments are shown. The data are averaged from three independent experiments and error bars represent ± s.e.m. (*, P < 0.05; **, P < 0.01; ***, P < 0.001, ****, P < 0.0001; Student’s t test. Where significance values are not shown, the differences were not significant; P > 0.05).

Fig 5. TgApiAT5-3 is an L-tyrosine transporter that is stimulated by the presence of L-tyrosine on the trans side of the membrane.

(A) Timecourse for the uptake of L-Tyr into X. laevis oocytes expressing TgApiAT5-3 (squares) or into uninjected oocytes (circles). Uptake was measured in the presence of 1 mM L-Tyr containing 0.5 μCi/ml [¹⁴C]Tyr. Each data point represents the mean uptake in 10 oocytes from a single experiment ± standard deviation, and the data are representative of 3 independent experiments. A first order rate equation was fitted to each timecourse (R² = 0.97 for TgApiAT5-3-expressing oocytes and R² = 0.77 for uninjected controls). Both the rate constant for L-Tyr uptake and the maximal L-Tyr uptake measured in TgApiAT5-3-expressing oocytes were significantly higher than those measured in uninjected oocytes (P < 0.01, Student’s t tests). (B) TgApiAT5-3-expressing oocytes (squares) and uninjected oocytes (circles) were preloaded (P.L.) with L-Tyr by incubation in 2.5 mM unlabelled L-Tyr (filled symbols) for 32 or 72 hr, respectively, or not preloaded (open symbols). Following the preincubation period, uptake of L-Tyr was measured in a solution of 1 mM L-Tyr containing 0.5 μCi/ml [¹⁴C]Tyr. Data show the mean uptake in 10 oocytes from a single experiment ± standard deviation, and are representative of 3 independent experiments. First order rate equations were fitted to the uptake timecourses for the preloaded and non-preloaded TgApiAT5-3-injected oocytes (R² = 0.98 for preloaded, and R² = 0.95 for non-preloaded oocytes). Both the first order rate constants for L-Tyr uptake and the maximal L-Tyr uptake were significantly higher in preloaded compared to non-preloaded TgApiAT5-3-expressing oocytes (P < 0.01, Student’s t tests). (C) TgApiAT5-3-expressing oocytes were preloaded by incubation in 1 mM L-Tyr containing 0.5 μCi/ml [¹⁴C]Tyr for 32 hr. Subsequent efflux (filled symbols) and retention (open symbols) of the preloaded substrate was measured over the timecourse indicated, in the presence of an extracellular medium containing 2.5 mM L-Tyr (squares) or extracellular medium lacking of L-Tyr (circles). Data show the mean efflux and retention ± standard deviation in 3 replicates (measuring efflux/retention from 5 oocytes each) from a single experiment, and are representative of 3 independent experiments. (D) Trans-stimulated initial rate kinetic analysis of L-Tyr transport by TgApiAT5-3. The rate of L-Tyr uptake was measured at a range of [L-Tyr] concentrations in the external medium (i.e. [L-Tyr]_cis) in TgApiAT5-3-expressing oocytes preloaded with 0 mM to 2.5 mM L-Tyr (i.e. [L-Tyr]_trans). The TgApiAT5-3-mediated uptake (calculated by subtracting the uptake in uninjected oocytes from the uptake in TgApiAT5-3-expressing oocytes) at each [L-Tyr]_trans condition tested conformed to a Michaelis-Menten kinetic model (R² > 0.90 for all non-linear regressions). The data were fitted to a Scatchard linear regression (0.89 ≤ R² ≤ 0.98 for all linear regressions). Data show the mean uptake rate ± standard deviation in 10 oocytes from a single experiment, and are representative of 2 independent experiments.

Fig 6. TgApiAT5-3 is an exchanger for aromatic and large neutral amino acids.

(A) TgApiAT5-3-expressing oocytes were pre-injected with a range of L-amino acids at a calculated oocyte cytosolic concentration of 5 mM (with the exception of L-Tyr (§) which was preloaded into oocytes via incubation in 2.5 mM L-Tyr for 32 hr), or were not pre-injected (ND96 condition). Subsequent uptake of 1 mM L-amino acids containing 0.5 or 1 μCi/ml [¹⁴C]-labelled amino acid (cis-Substrate) was measured over 10 minutes and normalised to uptake per minute. Each box in the heat map shows the mean uptake in 10 oocytes from a single experiment, representative of 3 independent experiments. The statistical analyses compare pre-injected/pre-loaded oocytes to ND96 controls for each substrate tested (*, P < 0.05, one-way ANOVA, Dunnet’s post-hoc test. Where significance values are not shown, the differences are not significant, P > 0.05). (B) TgApiAT5-3-expressing oocytes were preloaded with a range of L-amino acids containing 0.5 or 1 μCi/ml [¹⁴C] radiolabelled amino acids (calculated final concentrations shown beneath each substrate), and efflux of these substrates was measured over 5 min in the absence of external amino acids (ND96) or in the presence of 5 mM external amino acids (with the exception of L-Tyr (§), which was present at a concentration of 2.5 mM), and normalised to efflux per minute. Each box in the heat map shows the mean rate of efflux from 3 replicates (each comprised of 5 oocytes) from a single experiment, representative of 3 independent experiments. Statistical analyses compare trans substrates to ND96 controls for each efflux substrate tested (*, P < 0.05, one-way ANOVA, Dunnet’s post-hoc test. Where significance values are not shown, the differences are not significant, P > 0.05).

Fig 7. TgApiAT5-3 mediates the uptake of L-tyrosine and L-phenylalanine into T. gondii.

Initial rate of uptake of (A) [¹⁴C]Tyr, (B) [¹⁴C]Phe, and (C) [¹⁴C]Arg, in WT (RHΔhxgprt/Tomato), apiAT5-3^Δ188–504, and apiAT5-3^Δ188-504/cTgApiAT5-3 strain parasites. Uptake was measured in PBS-glucose containing either 60 μM unlabelled L-Tyr and 0.1 μCi/ml [¹⁴C]Tyr (A), 15 μM unlabelled L-Phe and 0.1 μCi/ml [¹⁴C]Phe (B), or 100 μM unlabelled L-Arg and 0.1 μCi/ml [¹⁴C]Arg (C). The initial rates of transport for each substrate were computed from the initial slopes of the fitted single-order exponential curves (S8 Fig), and represent the mean ± SEM from three independent experiments (* P<0.05; ** P<0.01; n.s. = not significant; Student’s t test).

Fig 8. In vitro growth of parasites lacking TgApiAT5-3 is modulated by the concentration of aromatic amino acids in the growth medium, and TgApiAT5-3 is important for parasite virulence.

(A-C) Fluorescence growth assay for WT (RHΔhxgprt/Tomato, black), apiAT5-3^Δ188–504 (red) and apiAT5-3^Δ188-504/cTgApiAT5-3 strain parasites cultured for 5 days in DMEM containing a range of L-Tyr (A), L-Phe (B), or L-Trp (C) concentrations. Growth is expressed as a percentage of maximum growth measured on day 5 for each parasite strain. Sigmoidal curves were fitted to the data in (A). All data shown are averaged from three biological replicates (mean ± SEM). (D) Balb/c mice were infected intraperitoneally with 1,000 WT (black), apiAT5-3^Δ188–504 (red), or apiAT5-3^Δ188-504/cTgApiAT5-3 (blue) strain parasites and monitored for symptoms of toxoplasmosis. Data for WT and apiAT5-3^Δ188–504 parasites are derived from 2 biological replicates consisting of 5 mice each, whereas data for apiAT5-3^Δ188-504/cTgApiAT5-3 parasites is derived from a single experiment consisting of 5 mice.

Fig 9. Model for the roles of ApiAT proteins in T. gondii tachyzoites.

Depiction of a T. gondii parasite (blue) inside a host cell. Aromatic amino acids, including L-Tyr, L-Phe and L-Trp, and large neutral amino acids are thought to be translocated across the parasitophorous vacuole membrane surrounding the parasite (dashed line) through non-selective channels [64]. TgApiAT5-3 (red cylinder) functions as the major L-Tyr uptake pathway in T. gondii. Additionally, TgApiAT5-3 functions as an exchanger, exporting aromatic and large neutral amino acids from the parasite, and thereby contributing to the homeostasis of these amino acids. The uptake of L-Phe and L-Trp is primarily mediated by alternate, and as yet undefined, uptake pathways (green cylinder). These alternate pathways can mediate sufficient L-Tyr uptake for parasite growth in the absence of TgApiAT5-3 at high L-Tyr concentrations (when L-Phe and L-Trp concentrations are not correspondingly high). We have previously demonstrated that TgApiAT1 (blue cylinder) facilitates the uptake of L-Arg into parasites [9], and propose that other ApiAT-family proteins expressed in tachyzoites (gray and purple cylinders) facilitate the transport of other amino acids. Of these, TgApiAT2 and TgApiAT6-1 (purple cylinders) are important for parasite growth.