TgApiAT6-1 is important for parasite proliferation

In a previous study of the ApiAT family in T. gondii, we demonstrated that TgApiAT6-1 localized to the plasma membrane of the parasite [20]. We were unable to disrupt the TgApiAT6-1 locus using CRISPR/Cas9 genome editing, suggesting that TgApiAT6-1 (www.toxodb.org gene identifier TGME49_240810) may be essential for proliferation of the tachyzoite stage [20]. To test this hypothesis, and to facilitate subsequent functional characterisation of TgApiAT6-1, we generated a regulatable parasite strain, rTgApiAT6-1, in which TgApiAT6-1 expression could be knocked down through the addition of anhydrotetracycline (ATc) (S1 Fig). A 3’ hemagglutinin (HA) epitope tag was introduced into the rTgApiAT6-1 locus and knockdown of the TgApiAT6-1 protein was observed after 2 days cultivation in the presence of ATc (Fig 1A). The rTgApiAT6-1 strain expressed a tdTomato transgene to allow measurement of growth in the absence or presence of ATc for 7 days using fluorescence proliferation assays, as previously described [27]. Compared to parasites cultured in the absence of ATc, we observed a major defect in proliferation in rTgApiAT6-1 parasites cultured in the presence of ATc (conditions under which TgApiAT6-1 is knocked down; Fig 1B). ATc had no effect on the proliferation of wild type (WT) parasites under the same conditions (Fig 1C). These data indicate that the knockdown of rTgApiAT6-1 is associated with a severe impairment of parasite proliferation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. TgApiAT6-1 is important for parasite proliferation.

A. Western blot analysis of proteins extracted from rTgApiAT6-1-HA parasites cultured in the presence of ATc for 0–2 days, and detected with antibodies against HA or TgTom40 (as a loading control). B-D. Fluorescence growth assays measuring the proliferation of rTgApiAT6-1 parasites (B), WT parasites (C), or rTgApiAT6-1 parasites complemented with a constitutively-expressed copy of TgApiAT6-1 (rTgApiAT6-1/cTgApiAT6-1; D). Parasites were cultured for 6 or 7 days in the absence (black) or presence (red) of ATc. Parasite proliferation is expressed as a percentage of parasite proliferation in the -ATc condition on the final day of the experiment for each strain. Data are averaged from 3 technical replicates (± S.D.) and are representative of three independent experiments.

https://doi.org/10.1371/journal.ppat.1009835.g001

To determine whether the proliferation phenotype we observed was due solely to depletion of TgApiAT6-1, we introduced a constitutively-expressed copy of TgApiAT6-1 into rTgApiAT6-1 parasites, generating the strain rTgApiAT6-1/cTgApiAT6-1. The presence of the constitutively-expressed TgApiAT6-1 fully restored parasite proliferation in the presence of ATc (Fig 1D). Together, these data indicate that TgApiAT6-1 is important for proliferation of the tachyzoite stage of T. gondii parasites.

TgApiAT6-1 is a high affinity lysine transporter

In order to investigate whether rTgApiAT6-1 is an amino acid transporter, we incubated rTgApiAT6-1 parasites grown for two days in the absence or presence of ATc in amino acid-free RPMI 1640 medium supplemented with 2 mg/ml of a [¹³C]-labelled amino acid mix. Parasites were incubated in the [¹³C]-labelled amino acid mix for 15 mins, then polar metabolites were extracted and analysed by GC-MS. We compared the fractional abundance of ¹³C-labelled amino acids to the total abundance of each amino acid following the 15 min uptake period (Fig 2A). Of the 17 amino acids detected by GC-MS, only the uptake of ¹³C-Lys was significantly reduced when TgApiAT6-1 expression was knocked down. These findings were replicated in a recent study that adopted a similar approach in measuring the uptake of ¹³C-labelled amino acids in parasites depleted in TgApiAT6-1 [34]. These data suggested that TgApiAT6-1 may be a Lys transporter, although it could also mediate the uptake of other amino acids not detected under the transport conditions of the experiments, or not detected by GC-MS, such as Arg. Arg is readily converted into L-ornithine during in-line heating and separation for GC-MS such that very little remains chemically unmodified [35,36].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. TgApiAT6-1 is a cationic and neutral amino acid transporter with high affinity for Lysine.

A. Analysis of [¹³C] amino acid uptake in parasites expressing or lacking TgApiAT6-1. rTgApiAT6-1 parasites were cultured for 2 days in the absence (black) or presence (red) of ATc until natural egress, then incubated in medium containing [¹³C]-L-amino acids for 15 min. Metabolites were extracted and the fractions of [¹³C]-L-amino acids determined by GC-MS. Amino acids are represented by single letter codes; OxoP, 5-oxoproline. The data represent the mean ± S.D of 3 replicate experiments (*, P < 0.001, Student’s t test. Where significance values are not shown, differences were not significant; P > 0.05). B. Uptake of a range of amino acids into oocytes expressing TgApiAT6-1. Uptake was measured in the presence of 100 μM unlabelled substrate and 1.0 μCi/ml [³H] or [¹⁴C] substrate. Amino acid substrates are represented by single letter codes, while for other metabolites: Pu, putrescine; Sp, spermidine; and GA, γ-amino butyric acid (GABA). Each bar represents the mean ± S.D. uptake of 10 oocytes for a single experiment, and each is representative of three independent experiments. The uptake into uninjected oocytes (shown in S3A Fig) was subtracted for all substrates tested. Statistical analysis compares injected oocyte uptake to uninjected oocyte uptake for the same substrate (*P < 0.05, one-way ANOVA, Dunnett’s post-hoc test). C-D. Inhibition of Arg uptake into TgApiAT6-1-expressing oocytes by a range of amino acids. Uptake of 100 μM unlabelled Arg and 1.0 μCi/ml [¹⁴C]Arg was measured in the presence of 1 mM (C) or 10 mM (D) of the competing amino acid. Amino acid substrates are represented by single letter codes. Each bar represents the mean ± S.D. uptake of 10 oocytes for a single experiment, and are representative of three independent experiments. The first bar in each graph represents the Arg-only uptake control. The uptake in uninjected oocytes (shown in S3B and S3C Fig for the 1 mM and 10 mM competition experiments, respectively) has been subtracted for all conditions. Statistical analysis compares all bars to the Arg uptake control (*, P < 0.05, one-way ANOVA, Dunnett’s post-hoc test). E-F. Steady-state kinetic analysis of Lys (E) and Arg (F) uptake into TgApiAT6-1-expressing oocytes. Uptake was measured at a range of concentrations of unlabelled Lys (E) or Arg (F) as indicated on the x-axis and 1.0 μCi/ml [¹⁴C]Arg or [¹⁴C]Lys. Each data point represents the mean ± S.D. uptake of 10 oocytes for a single experiment, and are representative of three independent experiments. The uptake into uninjected oocytes has been subtracted for all substrate concentrations tested.

https://doi.org/10.1371/journal.ppat.1009835.g002

To characterise further the substrate specificity of TgApiAT6-1, and to investigate the transport mechanism, we expressed TgApiAT6-1 in Xenopus laevis oocytes. We demonstrated that TgApiAT6-1 was expressed on the plasma membrane of oocytes, where it is detectable at two distinct molecular masses: one at ~48 kDa, consistent with a monomer of TgApiAT6-1, and another at ~95 kDa, which may represent a dimeric form of the protein (S2A Fig). After optimising its expression in oocytes (S2B and S2C Fig), we investigated the substrate specificity of TgApiAT6-1. We measured the uptake of a range of radiolabelled amino acids and amino acid derivatives in TgApiAT6-1-expressing oocytes, a selection of which are shown in Fig 2B. Consistent with the metabolomics data, TgApiAT6-1 mediated Lys uptake (Fig 2B). Notably, TgApiAT6-1 also mediated uptake of Arg and some neutral amino acids including Met and Leu (Fig 2B).

In the experiment measuring the uptake of [¹³C]-labelled amino acids into parasites, neither Met nor Leu showed discernible changes in uptake following knock down of TgApiAT6-1 (Fig 2A). This may be because TgApiAT6-1 has a higher affinity for Lys than for the neutral amino acids, such that under the conditions of the ¹³C-labelled amino acid uptake experiment, the Lys in the medium excluded the other amino acids from the active site of the transporter. To test whether this was the case, we measured TgApiAT6-1-mediated uptake of Arg in oocytes in the presence of a 10-fold (Fig 2C) or 100-fold (Fig 2D) higher concentration of other, unlabelled amino acids. At a 10-fold higher concentration of the unlabelled amino acid, only Lys inhibited Arg uptake (Fig 2C); however, at 100-fold higher concentrations, numerous neutral amino acids including Met, Leu, Phe and His partially inhibited Arg uptake (Fig 2D). This is consistent with the transporter having a higher affinity for Lys than for the other unlabelled amino acids tested.

To test the affinity of TgApiAT6-1 for Lys and Arg, we measured the uptake kinetics of these amino acids. The rate of substrate uptake for both Lys and Arg into oocytes expressing TgApiAT6-1 remained constant throughout the first 10 min of uptake reactions (S2D Fig) and subsequent experiments were performed within this timeframe. We found that TgApiAT6-1 has a much higher affinity for Lys than for Arg (K_0.5 for Lys was 22.8 μM ± 2.9 μM; K_0.5 for Arg was 748 μM ± 260 μM; Fig 2E and 2F and Table 1), although the maximal rate (V_max) of Arg uptake (169 ± 8 pmol/10 min/oocyte; Fig 2E and 2F and Table 1) greatly exceeded that for Lys (25.4 ± 1.5 pmol/10 min/oocyte; Fig 2E and 2F and Table 1). Further evidence for the preference of TgApiAT6-1 for Lys over Arg was obtained by calculating the apparent specificity constant (k_cat/ K_0.5) for both substrates, which was ~five-fold greater for Lys uptake than Arg (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. TgApiAT6-1 steady-state kinetic parameters for Lys and Arg.

https://doi.org/10.1371/journal.ppat.1009835.t001

Our previous study of the selective Arg transporter TgApiAT1 indicated that this transporter is electrogenic (i.e. transport of Arg across the membrane is coupled to a net movement of an elemental charge), and that the electrogenicity of TgApiAT1-mediated Arg transport is due to the positive charge of Arg at neutral pH [27]. We investigated whether TgApiAT6-1 is also electrogenic. When TgApiAT6-1-expressing oocytes were voltage clamped to −50 mV and perfused with 1 mM Arg, an inward current was observed (Fig 3A), reminiscent of the Arg-induced currents observed previously for TgApiAT1 [27] and other eukaryotic AA⁺ transporters [37,38]. The first phase of these currents is represented by a steep inward current peaking at approximately –50 nA, before the second phase shows a gradual relaxation towards an equilibrated inward current of approximately –20 nA (Fig 3A). On removal (washout) of Arg from the medium, the current showed an overshoot, increasing to beyond the pre-substrate perfusion baseline current (Fig 3A), with the magnitude of this overshoot increasing with the duration of the 1 mM Arg perfusion (Fig 3B). The biphasic current pattern disappears when TgApiAT6-1 expressing, voltage-clamped oocytes were pre-injected with 1 mM Arg (Fig 3C). Together, these data can be explained by TgApiAT6-1 facilitating the bi-directional transport of Arg (i.e. into and out of the oocyte). In this scenario, the biphasic current and overshoot observed in oocytes reflect the movement of charge out of the oocyte as the intracellular concentration of Arg increases following uptake, something that is not observed in Arg-injected oocytes, in which the intracellular Arg concentration is high from the beginning of the experiment, and from which Arg efflux is occurring throughout.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Arg and Lys compete for the same binding site in TgApiAT6-1.

Electrophysiology measurements in TgApiAT6-1 expressing oocytes. All currents were recorded in two-voltage clamp configuration to record membrane current. All oocytes were voltage clamped at −50 mV and the application of substrate (Arg or Lys) are indicated by the dashed horizontal lines above the tracings. Representative current tracings were normalised to 0 nA to remove background (non-substrate induced) current. A. Representative current tracing of TgApiAT6-1 expressing oocytes upon the addition and subsequent washout of 1 mM extracellular Arg ([Arg]_o) with no pre-injection of substrate. The perfusion buffer used was ND96 (pH 7.3). B. Representative current tracing of a TgApiAT6-1 expressing oocyte repeatedly pulsed with 1 mM Arg for 1 min, 2 min, and 10 min with 5 min gaps in between pulses. The perfusion buffer used was ND96 (pH 7.3). C. Representative current tracing of a TgApiAT6-1 expressing oocyte pre-injected with 1 mM Arg ([Arg]_i = 1 mM) upon the addition and subsequent washout of 1 mM extracellular Arg ([Arg]_o). The perfusion buffer used was ND96 (pH 7.3). D. Representative current tracings of TgApiAT6-1 expressing oocytes upon the addition and subsequent washout of either 1 mM extracellular Arg ([Arg]_o) or Lys ([Lys]_o) with no pre-injection of substrate. The perfusion buffer used was ND96 (pH 7.3). E-F. Inhibition kinetics of increasing [Lys] on Arg-induced inward currents in TgApiAT6-1 expressing oocytes. All oocytes were voltage-clamped at −50 mV and exposed to a range of Arg concentrations at one of 3 concentrations of Lys (0, 50, 500 μM) until a new steady-state current baseline was achieved. Each data point represents the mean ± S.D. steady-state inward current (n = 14 oocytes). All 3 inhibitory [Lys] plots were fitted to either the Michaelis-Menten equation (E) with R² values of 0.97 (0 mM Lys), 0.84 (50 μM), 0.63 (500 μM), or a Lineweaver-Burke linear regression (F) with R² values of 0.94 (0 mM Lys), 0.93 (50 μM), and 0.67 (500 μM).

https://doi.org/10.1371/journal.ppat.1009835.g003

To determine whether the observed currents are a direct consequence of the movement of Arg, or whether inorganic ions contribute to the current, we examined Arg-induced currents in voltage-clamped oocytes (−50 mV) in buffers with different salt compositions. Arg-stimulated currents gave similar values independent of salt composition (S4A and S4B Fig), consistent with Arg being the current-generating ion. Furthermore, the unidirectional TgApiAT6-1-mediated uptake of Arg into oocytes was unaffected by the absence of Na⁺, K⁺, Cl^-, Mg²⁺ or Ca²⁺ from the medium (S4C Fig), consistent with these ions not being required for co-transport with the cationic amino acid.

We also measured Lys-induced currents in voltage-clamped TgApiAT6-1 expressing oocytes, and observed small currents of ~1–2 nA above background (Fig 3D), consistent with the low maximum velocity of TgApiAT6-1 in transporting Lys (Fig 2E and Table 1). The small relative magnitude of the Lys-mediated currents in our set-up precluded the use of electrophysiology to characterise Lys transport. Together, our electrophysiology studies demonstrate that the currents observed for TgApiAT6-1 are carried by AA⁺ substrates of the transporter and not by any other biologically relevant ion.

Our earlier data indicated that Lys can inhibit Arg uptake into oocytes (Fig 2C and 2D). We therefore investigated whether Arg and Lys compete for the same binding site of the TgApiAT6-1 transporter. To do this, we exploited the observation that Arg, but not Lys, induces appreciable currents in voltage-clamped oocytes expressing TgApiAT6-1 (Fig 3D). We measured the steady-state kinetics of Arg-induced currents in the presence of increasing concentrations of Lys. Lys acted as a high affinity competitive inhibitor of Arg, with K_0.5 values for Arg increasing from 0.19 mM ± 0.05 mM at 0 mM Lys, to 2.3 mM ± 0.2 mM at 50 μM Lys, to 28 mM ± 8 mM at 500 μM Lys (Fig 3E). By contrast, the V_max of Arg-induced currents remained constant at approximately −40 nA over the range of Lys concentrations tested. The changes in K_0.5 are readily observed as a change in the slope of the Lineweaver-Burke linear regressions, while the intersection of the three regression lines at the ordinate (1/V_Arg) indicates similar V_max values (Fig 3F). These data are consistent with Arg and Lys binding to the same binding site of TgApiAT6-1, and with these substrates competing for transport by this protein. The higher affinity of TgApiAT6-1 for Lys compared to Arg means that, in a physiological setting, the contribution of TgApiAT6-1 to Arg uptake will vary with the concentration of Lys, increasing as the [Arg]:[Lys] ratio increases. We demonstrated in a previous study that Lys uptake by the parasite can be inhibited by Arg, and that Arg uptake in parasites lacking the selective Arg transporter TgApiAT1 is inhibited by Lys [27]. This is consistent with the competition between these substrates for uptake by TgApiAT6-1 that we observed in the oocyte experiments (Fig 3E and 3F).

To test whether TgApiAT6-1 contributes to Lys uptake in parasites, we measured the uptake of Lys in TgApiAT6-1 parasites cultured in the absence or presence of ATc for 2 days. We observed a significant ~16-fold decrease in the initial rate of Lys uptake upon TgApiAT6-1 knockdown (P < 0.01; ANOVA; Figs 4A and S5A). We next investigated the contribution of TgApiAT6-1 to Arg uptake. We measured the uptake of [¹⁴C]Arg in TgApiAT6-1 parasites cultured in the absence or presence of ATc for 2 days. We observed a significant ~five-fold decrease in the initial rate of Arg uptake upon TgApiAT6-1 knockdown (P < 0.01; ANOVA; Figs 4B and S5B). These data are consistent with TgApiAT6-1 mediating the uptake of both Lys and Arg into the parasite. Neither Lys nor Arg uptake was impaired in WT parasites cultured in the presence of ATc (Figs 4A and 4B and S5C and S5D). Likewise, uptake of 2-deoxy-glucose, a glucose analogue, was unaffected upon TgApiAT6-1 knockdown (S5E Fig). These data indicate that the observed defects in Lys and Arg uptake in the rTgApiAT6-1 strain were not the result of ATc addition, or of a general impairment of solute uptake or parasite viability.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. TgApiAT6-1 mediates uptake of Lys and Arg in T. gondii parasites.

A-B. Initial rate of Lys (A) and Arg (B) uptake in WT and rTgApiAT6-1 parasites cultured in DME in the absence (black) or presence (red) of ATc for 2 days. Uptake was measured in 50 μM unlabelled Lys and 0.1 μCi/ml [¹⁴C]Lys (A) or 80 μM unlabelled Arg and 0.1 μCi/ml [¹⁴C]Arg (B). Initial rates were calculated from fitted curves obtained in time-course uptake experiments (S5 Fig). Data represent the mean initial uptake rate ± S.D. from three independent experiments.(**, P = 0.01; n.s. not significant; ANOVA with Sidak’s multiple comparisons test).

https://doi.org/10.1371/journal.ppat.1009835.g004

We have previously shown that parasites lacking the selective Arg transporter TgApiAT1 could proliferate in medium containing high concentrations of Arg relative to Lys [27]. This suggested the presence of a second Arg transporter, which our data now indicate is TgApiAT6-1. To test whether the proliferation of parasites lacking TgApiAT6-1 was similarly dependent on the concentrations of Arg or Lys in the culture medium, we cultured rTgApiAT6-1 parasites in the absence or presence of ATc and at a range of [Lys] or [Arg]. In conditions in which TgApiAT6-1 is expressed (-ATc), we observed defects in parasite proliferation at [Lys] or [Arg] below ~50 μM (S6 Fig), consistent with previous data indicating that the parasite is auxotrophic for these amino acids [27]. In the absence of TgApiAT6-1 (+ATc), we observed minimal parasite proliferation regardless of the concentration of either Lys or Arg (S6 Fig). This indicates that, unlike TgApiAT1, defects in the proliferation of parasites lacking TgApiAT6-1 cannot be rescued by modifying the concentration of its substrates in the culture medium.

The genome of T. gondii parasites encodes homologues of at least some enzymes that function in the so-called diaminopimelate pathway for Lys biosynthesis [24,39]. It is conceivable, therefore, that parasites can compensate for the loss of TgApiAT6-1 by synthesising Lys via this pathway. The final putative enzyme in this pathway is termed TgLysA (diaminopimelate decarboxylase; TGME49_278740). To characterise the importance of the Lys biosynthesis pathway in T. gondii tachyzoites, we disrupted the TgLysA genomic locus in RHΔhxgprt/Tomato strain parasites using a CRISPR/Cas9-based genome editing approach. The resultant 2 bp deletion in the TgLysA locus causes a frameshift mutation in the TgLysA open reading frame, leading to the predicted production of a truncated TgLysA protein lacking residues 281 to 686, and loss of key active site residues that are predicted to render the enzyme non-functional (S7A Fig; [40]). We termed this strain lysA^Δ281–686. To test the importance of TgLysA for parasite proliferation, we performed fluorescence proliferation assays on WT or lysA^Δ281–68 parasites cultured in medium containing 0–800 μM [Lys]. As demonstrated previously ([27], S6A Fig), proliferation of WT parasites is inhibited at [Lys] below ~50 μM (S7B Fig). lysA^Δ281–68 parasites exhibit a similar inhibition of proliferation at [Lys] below 50 μM, and a virtually identical growth response across the entire range of tested [Lys] (S7B Fig). The observed dispensability of TgLysA for tachyzoite proliferation is consistent with data from an earlier CRISPR-based screen that examined the importance of all genes encoded in the parasite genome [33].

Together, the data in S6 and S7 Figs indicate that: a) T. gondii tachyzoites are auxotrophic for Lys; b) proliferation of parasites lacking TgApiAT6-1 cannot be rescued by increasing the concentrations of Lys or Arg; and c) the putative Lys biosynthesis pathway in these parasites is either not functional or is unable to supply the Lys requirements needed for parasite viability. Given that the Arg transporter TgApiAT1 is still expressed when TgApiAT6-1 is depleted [31], it is unlikely that the growth defects observed upon TgApiAT6-1 depletion result solely from defects in Arg uptake. These data therefore support the hypothesis that TgApiAT6-1 is essential because it is required for the uptake of Lys. In summary, our data are consistent with the hypothesis that TgApiAT6-1 is the primary and essential Lys uptake pathway into tachyzoite-stage parasites, as well as having a role in the uptake of Arg.

Mechanisms of cationic amino acid acquisition by TgApiAT6-1 and TgApiAT1

In order to understand the integrated contributions of both TgApiAT6-1 and TgApiAT1 to parasite AA⁺ acquisition, we investigated the transport mechanisms of the two proteins in more detail. Our observation of TgApiAT6-1-mediated Arg outward currents (Fig 3B and 3C), and our previous work on the related aromatic and neutral amino acid exchanger TgApiAT5-3 [20], suggested that TgApAT6-1 and TgApiAT1 might operate as (bidirectional) amino acid exchangers. To explore this possibility, we first asked whether TgApiAT6-1 and TgApiAT1 could efflux substrates. Oocytes expressing TgApiAT6-1 were preloaded with [¹⁴C]Lys and the efflux of radiolabel was measured in the presence or absence of 1 mM unlabelled Lys in the extracellular medium (Fig 5A). Similarly, oocytes expressing TgApiAT1 were preloaded with [¹⁴C]Arg and the efflux of radiolabel was measured in the presence or absence of 1 mM unlabelled extracellular Arg (Fig 5B). For both TgApiAT6-1 and TgApiAT1, the presence of substrate in the extracellular medium gave rise to an increase in effluxed [¹⁴C]substrate and a decrease in oocyte-retained [¹⁴C]substrate over time (Fig 5A and 5B); i.e. extracellular Lys induced a ‘trans-stimulation’ of the efflux of [¹⁴C]Lys from oocytes expressing TgApiAT6-1, and extracellular Arg induced a trans-stimulation of the efflux of [¹⁴C]Arg from oocytes expressing TgApiAT1. Neither efflux of preloaded substrate nor trans-stimulation of either substrate was observed in H₂O-injected control oocytes (S8A and S8B Fig). These results indicate that, in addition to the net substrate uptake that we have demonstrated previously (Fig 2; [27]), both transporters are capable of substrate exchange and can be trans-stimulated; i.e. the unidirectional flux of radiolabel is stimulated by the presence of substrate at the opposite (trans) face of the membrane.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Trans-simulation of AA⁺ transport in TgApiAT6-1- and TgApiAT1-expressing oocytes.

A-B. TgApiAT6-1-injected (A) and TgApiAT1-injected (B) oocytes were pre-loaded with either 1 mM unlabelled Lys and 1.0 μCi/ml of [¹⁴C]Lys (TgApiAT6-1) or 1 mM unlabelled Arg and 1.0 μCi/ml of [¹⁴C]Arg (TgApiAT1) for 3–6 hr. The retention of substrates in TgApiAT6-1- or TgApiAT1-expressing oocytes were measured in the presence of 1 mM external substrate (closed symbols) or in the absence of an external substrate (open symbols). Data points represent the mean ± S.D. from 3 batches of 5 oocytes from one experiment, and are representative of 3 independent experiments. C-D. Arg efflux and retention in TgApiAT6-1 expressing (C) and TgApiAT1 expressing (D) oocytes in the presence of candidate trans-stimulating substrates. Oocytes were pre-loaded (PL) with Arg by first microinjecting Arg to a final concentration of ~5 mM, then incubated oocytes in a solution containing 1 mM unlabelled Arg and 1.0 μCi/ml of [¹⁴C]Arg for 1 hr (TgApiAT6-1) or 2 hr (TgApiAT1). This pre-loading was followed by addition of 1 mM unlabelled amino acids or amino acid derivatives to the outside of the oocyte. Efflux of pre-loaded Arg (Arg_o; black bars) and retention of pre-loaded Arg (Arg_i; white bars) were measured after 1 hr (TgApiAT6-1) or 2 hr (TgApiAT1) in the presence of 1 mM of the metabolites or with ND96 buffer in the extracellular medium (−). The horizontal dashed line across both figures indicates the amount of Arg pre-loaded (PL) into oocytes (left-most bar). Amino acid substrates are represented by single letter codes, while for other metabolites: Cr, creatine; Ag, agmatine; Sp, spermidine; Pu, putrescine; Ci, citrulline; Ur, urea; and Or, ornithine. Each bar represents the mean ± S.D. efflux or retention in from 3 batches of 5 oocytes from one experiment, and are representative of three independent experiments. Statistical analysis compares all bars to oocyte efflux in the presence of ND96 (−) (pH 7.3) (* P < 0.05, one-way ANOVA, Dunnett’s post-hoc test). E-F. Lys (E) or Arg (F) uptake into TgApiAT6-1 expressing oocytes pre-loaded with a range of candidate trans-stimulating substrates. TgApiAT6-1-injected oocytes were pre-loaded by microinjecting the indicated substrates to a final concentration of ~5 mM (or with the same volume of H₂O as a control), and the uptake of 15 μM Lys and 1.0 μCi/ml of [¹⁴C]Lys (E) or 1 mM Arg and 1.0 μCi/ml of [¹⁴C]Arg (F) was determined. Uptake of Arg and Lys into control oocytes not expressing TgApiAT6-1 using the same trans-stimulation conditions (shown in S3D and S3E Fig for Arg and Lys uptake, respectively) were subtracted for all conditions. Amino acid substrates are represented by single letter codes, while for other metabolites: Cr, creatine; Ag, agmatine; Sp, spermidine; Pu, putrescine; Ci, citrulline; and Or, ornithine. Each bar represents the mean ± S.D. uptake of 10 oocytes for a single experiment, and are representative of three independent experiments. Statistical analysis compares all bars to Lys or Arg uptake in control H₂O ‘pre-loaded’ oocytes (* P < 0.05, one-way ANOVA, Dunnett’s post-hoc test).

https://doi.org/10.1371/journal.ppat.1009835.g005

Our earlier results (Fig 2) indicated that TgApiAT6-1 could also mediate the flux of some neutral amino acids. To examine whether other substrates trans-stimulate AA⁺ flux via TgApiAT6-1 and TgApiAT1, we measured the efflux of [¹⁴C]Arg via TgApiAT6-1 and TgApiAT1 using a range of amino acids and amino acid derivatives as counter-substrates. The strongest trans-simulation of TgApiAT6-1-mediated [¹⁴C]Arg efflux was observed in the presence of several cationic amino acids (e.g. Arg, His, Orn), as well as by the large neutral amino acids Leu and Met (Fig 5C). Notably, when Lys was used as a counter-substrate for TgApiAT6-1, Arg efflux was significantly lower than when measured in the absence of a counter-substrate. As the rate of transport for any substrate is determined by the slowest step in the transport mechanism (i.e. by the rate-limiting step of the transport cycle), we hypothesise that the counter-transport of extracellular Lys via TgApiAT6-1 into the oocyte represents a rate-limiting step in which the rate at which [¹⁴C]Arg efflux can occur via TgApiAT6-1 is limited by the rate at which (unlabelled) Lys is transported into the oocyte. This notion is supported by the very low maximal Lys transport rate relative to maximal Arg transport rate (see Fig 2E and 2F and Table 1). We observed significant trans-stimulation of Lys efflux by large neutral amino acids and by cationic amino acids, although, unlike the effects of Lys on Arg efflux, none of the tested counter substrates inhibited Lys efflux (S3F Fig).

In contrast to the range of cationic and neutral amino acids that were able to trans-stimulate Arg efflux by TgApiAT6-1, we were unable to detect trans-stimulation of [¹⁴C]Arg efflux through TgApiAT1 using any counter-substrate other than Arg itself (Fig 5D). This suggests that TgApiAT1 does not transport substrates other than Arg, and is consistent with our previous study that indicated TgApiAT1 is a highly-selective Arg transporter [27].

To determine whether the specificity of trans-stimulation holds true for transport in both directions, we reversed the direction of substrate flux in TgApiAT6-1 expressing oocytes, and measured the trans-stimulation of Lys uptake by a range of substrates. Oocytes expressing TgApiAT6-1 were microinjected with a range of amino acids and arginine-derived metabolites to a final concentration of approximately 5 mM and the uptake of [¹⁴C]Lys was measured. Cationic amino acids and a number of neutral and hydrophilic amino acids trans-stimulated Lys uptake via TgApiAT6-1 (Fig 5E). None of the trans-stimulating amino acids increased the rate of Lys uptake beyond that observed under conditions of trans-stimulation by intracellular Lys. Next, we measured trans-stimulation of the uptake of [¹⁴C]Arg via TgApiAT6-1. As observed with the efflux experiments, several cationic (Arg, Orn) and large neutral amino acids (Val, Leu, Met, Phe) trans-stimulated Arg uptake into TgApiAT6-1-expressing oocytes (Fig 5F). By contrast, uptake of Arg with Lys present on the other side of the membrane was lower than for any other trans-stimulating substrate, and lower even than non-trans-stimulated uptake. This mirrors our observation of reduced Arg efflux when external Lys is present (Fig 5C), and further supports the hypothesis that the slow counter-transport of Lys acts as a rate-limiting step in the transport cycle of TgApiAT6-1 under the conditions of these transport assays.

In summary, these experiments indicate that transport of AA⁺ by TgApiAT6-1 can be trans-stimulated by a range of cationic and neutral amino acids at both the intra- and extra-cellular face of the membrane. Together these results are consistent with Lys being a high-affinity but low V_max substrate of TgApiAT6-1 in comparison to Arg, which has a lower affinity for the transporter but a much higher maximal rate of transport. The data in Fig 5C and 5F are also consistent with the low maximal rate of Lys transport by TgApiAT6-1 setting an upper limit (rate-limitation) to the speed at which Arg can be taken up or effluxed by TgApiAT6-1 under conditions in which Lys is present.

Our data indicate that TgApiAT1, a highly selective Arg transporter, is trans-stimulated strongly by Arg (Fig 5D). This could limit the net accumulation of Arg within parasites, with one molecule of Arg effluxed for every molecule that is transported in. Similarly, TgApiAT6-1, which exhibits little unidirectional efflux in the absence of trans-substrate and has a higher affinity for Lys than other amino acids, may be limited in its capacity to accumulate Lys and other substrates. We therefore utilised the oocyte expression system to investigate whether TgApiAT6-1 and TgApiAT1 are capable of net substrate accumulation, testing whether the intracellular concentration of amino acid substrates reached a level higher than the extracellular concentration.

To measure whether AA⁺ accumulation occurs in TgApiAT6-1 and TgApiAT1-expressing oocytes, we used a targeted LC-MS/MS approach to measure intra-oocyte substrate concentrations while accounting for the endogenous metabolism of these substrates [36]. TgApiAT6-1 or TgApiAT1-expressing oocytes were incubated in a solution containing 1 mM Lys or Arg for 50–54 hr, a timeframe that enabled us to observe the accumulation of specific amino acids over time, which we then quantified using external calibration curves (S9A and S9B Fig). TgApiAT6-1 expressing oocytes accumulated Lys to an intracellular concentration more than two-fold higher than the extracellular concentration, with full electrochemical equilibrium not yet reached at the final time point (Fig 6A, closed squares). Similarly, Arg accumulated to an estimated intracellular concentration some four- to five-fold higher than the extracellular concentration in oocytes expressing TgApiAT6-1, reaching equilibrium after ~40 hr (Fig 6B, closed squares). By contrast, in H₂O-injected control oocytes the intracellular concentrations of Lys and Arg increased to a value that was approximately the same as the extracellular concentration after ~24 hr, remaining at this level for the duration of the experiment (Fig 6A and 6B, closed circles). Of the metabolites reliably identified by LC-MS/MS across all conditions, only Lys and metabolic products of Lys (S1 Table) and Arg (S2 Table) displayed large intracellular accumulation in AA⁺-incubated oocytes expressing TgApiAT6-1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. TgApiAT6-1 and TgApiAT1 are net accumulators of cationic amino acids.

A-C. Time-course measuring the Lys (A) or Arg (B-C) concentration in TgApiAT6-1 expressing oocytes (A-B; squares), TgApiAT1 expressing oocytes (C; squares), or H₂O-injected oocytes (A-C; circles) incubated in 1 mM Lys (A) or 1 mM Arg (B-C) as quantified by LC-MS/MS. Following 32–34 hr of incubation measuring accumulation inside oocytes, samples were split into two groups, one continuing with substrate incubation (closed symbols), the other with substrate replaced by substrate-free incubation media (open symbols). Each data point represents the mean intracellular amino acid concentration ± S.D. of 12 individual oocytes (substrate-incubated) or 3 batches containing 5 oocytes each (substrate-free), and are representative of 3 independent experiments.

https://doi.org/10.1371/journal.ppat.1009835.g006

Removing the extracellular substrate from TgApiAT6-1 expressing oocytes after 32–34 hr resulted in a decrease in intracellular concentrations of both Lys and Arg (Fig 6A and 6B, open squares); this was not observed in H₂O-injected oocytes (open circles) and therefore is unlikely to be a result of these amino acids being incorporated into proteins and/or by conversion into other metabolites. Instead, these data are consistent with TgApiAT6-1 mediating the net efflux of amino acids from oocytes when external substrate is absent. Together with other results, these data indicate that TgApiAT6-1 is able to mediate the accumulation of cationic amino acids.

TgApiAT1 also mediated a substantial accumulation of Arg, with the intracellular concentration of Arg reaching a level some three-fold higher than the extracellular concentration after 32 hr (Fig 6C, closed squares), then decreasing following the removal of Arg from the medium (Fig 6C, open squares). Oocytes expressing TgApiAT1 displayed a slower accumulation of Arg than did oocytes expressing TgApiAT6-1. As was observed for oocytes expressing TgApiAT6-1, Arg was the only compound shown to undergo substantial intracellular accumulation in oocytes expressing TgApiAT1 and incubated in the presence of extracellular Arg (S2 Table).

Together, these data indicate that TgApiAT6-1 and TgApiAT1 function as uniporters, with the capacity to mediate amino acid exchange. Both transporters have the capacity to accumulate cationic substrate to concentrations higher than that in the extracellular medium. However, TgApiAT6-1 has a broad specificity for many AA⁺, large neutral amino acids and arginine metabolites and a very high selectivity for Lys, such that when Lys is present it reduces the transport rate for all other substrates. By contrast, Arg transport by TgApiAT1 is trans-stimulated solely by Arg (and no other amino acid), consistent with previous data indicating the high selectivity of this transporter for Arg [27].

Ethics statement

Xenopus laevis frog maintenance and oocytes preparation was approved by the Australian National University Animal Experimentation Ethics Committee (Protocols A2014/20 and A2020/48). Before surgeries to extract oocytes, frogs were anesthetized by submersion in a 0.2% (w/v) tricaine methanesulfonate (MS-222) solution made in tap water and neutralized with Na₂HCO₃ for 15–40 minutes until frogs exhibited no reaction upon being turned upside down.

Parasite culture and proliferation assays

Parasites were maintained in human foreskin fibroblasts in a humidified 37°C incubator with 5% CO₂. Parasites were cultured in Roswell Park Memorial Institute 1640 (RPMI) medium supplemented with 1% (v/v) fetal calf serum, 2 mM L-glutamine, 50 U/ml penicillin, 50 μg/ml streptomycin, 10 μg/ml gentamicin, and 0.25 μg/ml amphotericin b. Where applicable, anhydrotetracycline (ATc) was added to a final concentration of 0.5 μg/ml with 0.025% (v/v) ethanol added to relevant vehicle controls. In experiments wherein we varied the concentration of Lys in the growth medium, we generated homemade RPMI medium containing a range of [Lys], as described previously [31].

Fluorescence proliferation assays were performed as described previously [27]. Briefly, 2,000 tdTomato-expressing parasites were added to wells of an optical bottom 96-well plate containing a monolayer of host cells. Fluorescence was read regularly using a FLUOstar Optima plate reader (BMG).

Generation of genetically modified T. gondii parasites

To generate a T. gondii strain in which we could knock down expression of TgApiAT6-1, we replaced the native TgApiAT6-1 promoter region with an ATc-regulatable promoter using a double homologous recombination approach. First, we amplified the 5’ flank of TgApiAT6-1 with the primers ApiAT6-1 5’ flank fwd (5’-GACTGGGCCCCTTCATTTCTTCGCAACGTGACAAGC) and ApiAT6-1 5’ flank rvs (5’-GACTCATATGCCGACTTGCTTGAAGAACCTGCG). We digested the resulting product with PspOMI and NdeI and ligated into the equivalent sites of the vector pPR2-HA₃ [54], generating a vector we termed pPR2-HA₃(ApiAT6-1 5’ flank). Next, we amplified the 3’ flank with the primers ApiAT6-1 3’ flank fwd (5’- GATCAGATCTAAAATGGCGTCCTCGGACTCGAAC) and ApiAT6-1 3’ flank rvs (5’-CTAGGCGGCCGCGAGTTCGGAGGACGATCCAGAGG). We digested the resulting product with BglII and NotI and ligated into the equivalent sites of the vector pPR2-HA₃(ApiAT6-1 5’ flank) vector. We linearized the resulting plasmid with NotI and transfected this into TATi/Δku80 strain parasites [55] expressing tandem dimeric Tomato RFP [27]. We selected parasites on pyrimethamine and cloned parasites by limiting dilution. We screened clones for successful integration of the ATc regulatable promoter using several combinations of primers. To test for 3’ integration, we used the ApiAT6-1 3’ screen fwd (5’-CCGCAGTGGACGGACACC) and ApiAT6-1 3’ screen rvs (5’-CAGTTCCGCTCGGTTGCTTG) primers to detect the presence of the native locus, and the t7s4 screen fwd (5’- ACGCAGTTCTCGGAAGACG) and ApiAT6-1 3’ screen rvs (above) primers to detect the presence of the modified locus. To screen for 5’ integration, we used the ApiAT6-1 5’ screen fwd (5’-CTGGAGAAGTGTGTGAGGAGC) and ApiAT6-1 screen rvs (5’-GAGTGGAGACGCTGCGACG) primers to test for the presence of the native locus, and ApiAT6-1 5’ screen fwd (above) and DHFR screen rvs (5’-GGTGTCGTGGATTTACCAGTCAT) primers to detect the presence of the modified locus. We termed the resulting strain regulatable (r)TgApiAT6-1. To enable us to measure knockdown of the ATc-regulatable TgApiAT6-1 protein, we integrated a HA tag into the rTgApiAT6-1 locus by transfecting a TgApiAT6-1-HA 3’ replacement vector, described previously [20], into this strain.

To complement the rTgApiAT6-1 strain with a constitutively-expressed copy of TgApiAT6-1, we amplified the TgApiAT6-1 open reading frame with the primers TgApiAT6-1 orf fwd (5’-GATCAGATCTAAAATGGCGTCCTCGGACTCGAAC) and TgApiAT6-1 orf rvs (5’- GATCCCTAGGAGCGGAGTCTTGCGGTGGC) using genomic DNA as template. We digested the resulting product with BglII and AvrII and ligated into the equivalent sites of the pUgCTH₃ vector [27]. We linearised the resulting vector with MfeI, transfected into rTgApiAT6-1 parasites, and selected on chloramphenicol. We cloned drug resistant parasites before subsequent characterisation.

To produce a parasite strain containing a frameshift mutation in TgLysA, we first generated a single guide RNA (sgRNA)-expressing vector that targeted the TgLysA locus. To do this, we introduced a sgRNA-encoding sequence targeting TgLysA into the pSAG1::Cas9-U6::sgUPRT vector (Addgene plasmid #54467) by Q5 mutagenesis, as described previously [56], using the primers TgLysA CRISPR fwd (5’-AGGCGCCTCCAAATTTCTCAGTTTTAGAGCTAGAAATAGCAAG) and universal rvs (5’-AACTTGACATCCCCATTTAC). We transfected the resulting vector into RHΔhxgprt/Tomato parasites [57], then cloned Cas9-GFP-expressing parasites by fluorescence activated cell sorting 3 days after transfection. We amplified and sequenced the region around the TgLysA mutation site in clonal parasites with the primers TgLysA seq fwd (5’-GCTTGGCTGAGCTTTTGCT) and TgLysA seq rvs (5’-GCACTGCAGGTTGACTTGG). We selected a clone containing a 2 bp deletion in the locus for subsequent characterisation.

Radiolabel uptake assays in T. gondii parasites

Radiolabel uptake assays with extracellular T. gondii parasites were performed as described previously [58]. Briefly, WT and rTgApiAT6-1 parasites were cultured in Dulbecco’s Modified Eagle’s medium (containing 400 μM Arg and 800 μM Lys) in the absence or presence of ATc for 2 days. Parasites were isolated from host cells and resuspended in Dulbecco’s PBS (pH 7.4) supplemented with 10 mM glucose (PBS-glucose) and a mix of radiolabelled and unlabelled amino acids. Specifically, Lys uptake was measured by incubation in 0.1 μCi/ml [¹⁴C]Lys and 50 μM Lys, Arg uptake was measured by incubation in 0.1 μCi/ml [¹⁴C]Arg and 80 μM unlabelled Arg, and 2-deoxyglucose (2-DOG) was measured by incubation in 0.2 μCl/ml [1-¹⁴C]2-DOG and 25 μM unlabelled 2-DOG. Samples were incubated at 37°C, with aliquots removed at regular intervals and centrifuged through an oil mixture containing 84% (v/v) PM-125 silicone fluid and 16% (v/v) light mineral oil to separate parasites from unincorporated radiolabel. Samples were lysed and incorporated radiolabel was measured using a scintillation counter.

Parasite protein detection

Parasite protein preparations and associated western blotting was performed as described previously [59]. Membranes were probed with rat anti-HA (Sigma, clone 3F10; 1:500 dilution) and rabbit anti-TgTom40 ([60]; 1:2,000 dilution) primary antibodies, and horseradish peroxidase-conjugated goat anti-rat (Abcam catalogue number ab97057; 1:5,000 dilution) and goat anti-rabbit (Abcam catalogue number ab97051; 1:10,000 dilution) secondary antibodies. Blots were imaged on a ChemiDoc MP imaging system (Biorad).

[¹³C]-Amino acid labelling in parasites

Labelling experiments with a mixture of [¹³C]-amino acids were performed as described previously [20]. Briefly, rTgApiAT6-1 parasites were cultured for 2 days in the absence or presence of ATc. Egressed parasites were incubated in amino acid-free RPMI supplemented with 2 mg/ml algal [¹³C]amino acid mix (Cambridge Isotope Laboratories, catalogue number CLM-1548) for 15 min at 37°C, and parasite metabolites extracted in chloroform:methanol:water (1:3:1 v/v/v) containing 1 nmol scyllo-inositol as an internal standard. Polar metabolites in the aqueous phase were dried, methoxymated (20 mg/ml methoxyamine in pyridine overnight), then trimethylsilylated (N,O-bis(trimethylsilyl)trifluoroacetamide with 1% trimethylsilyl) for 1 hr at room temperature. Samples were analysed using GC-MS as described [61], and the fractional labelling of detected amino acids was determined using DExSI software [62].

Xenopus laevis oocytes preparation and transporter expression

Xenopus laevis oocytes were removed by abdominal surgical incision and extraction as previously reported [63]. To generate a vector from which we could express complementary RNA (cRNA) encoding the HA-tagged TgApiAT6-1 protein, we amplified the TgApiAT6-1 open reading frame with the primers TgApiAT6-1 oocyte fwd (5’-GATCCCCGGGCCACCATGGCGTCCTCGGACTCGAAC) and TgApiAT6-1 orf rvs (above). We digested the resulting product with XmaI and AvrII and ligated into the equivalent sites of the vector pGHJ-HA [27]. We prepared cRNA for injection and in vitro protein expression in oocytes as previously described [63–65]. The pGHJ-TgApiAT6-1 and pGHJ-TgApiAT1 [27] plasmids were linearized by digestion with NotI (NEB) for 2 hr. Following in vitro synthesis and purification, cRNA was quantified using a Tecan Infinite M1000 Pro (Tecan Group, Mӓnnedorf, Switzerland) spectrophotometer (OD₂₆₀/OD₂₈₀). For all transporter assays in oocytes, cRNA was micro-injected into stage 5 or 6 oocytes using a Micro4 micro-syringe pump controller and A203XVY nanoliter injector (World Precision Instruments, Sarasota, FLA, U.S.A.). Oocytes were maintained in oocyte Ringer’s buffer with added gentamycin (50 μg/ml) and Ca²⁺ (1.8 mM) as previously described [66]. We determined the optimal expression conditions for TgApiAT6-1 such that all oocyte experiments were conducted on days 4 to 6 post-cRNA injection and by micro-injection of 15ng cRNA/oocyte (S2B and S2C Fig).

Xenopus laevis oocyte uptake and efflux assays

Methods optimised for the study of ApiAT family transporters in X. laevis oocytes have been detailed previously [20]. For simple radiolabel uptake experiments, oocytes were washed four times in ND96 buffer (96 mM NaCl, 2 mM KCl, 1 mM MgCl₂, 1.8 mM CaCl₂, 5 mM HEPES, pH 7.3) at ambient temperature. In initial time-course experiments (S2D Fig) it was determined that the uptake of radiolabelled forms of both Lys and Arg were linear with time for over 10 min. In subsequent experiments estimates of initial uptake rates were therefore made using an incubation period of 10 mins except where indicated in the axis title. Uptake of radiolabel was quenched by washing oocyte batches four times in ice-cold ND96. For ion replacement uptake experiments in which alternative salt buffers were used, uptake was quenched by washing oocytes in the alternative buffer. The salt composition of these alternative buffers is indicated in the figure legends and detailed in S3 Table.

For uptake experiments in which trans-stimulation was investigated, 100 mM solutions of potential substrates (in ND96) were pre-injected at 25 nl/oocyte using a Micro4 micro-syringe and A203XVY nanoliter injector (World Precision Instruments). Oocytes were then incubated on ice for 30 mins prior to uptake experiments to allow for membrane recovery. The injection of 25 nl/oocyte of 100 mM solutions of the candidate substrates gave an estimated cytosolic concentration of 5 mM, based on an assumed free aqueous volume of 500 nl/oocyte [63,67]. Calculations of cytosolic concentrations from pre-injection should be treated as approximations only, as stage 5 or 6 oocyte diameters vary from 1−1.3 mm and aqueous oocyte volumes range from 368 to > 500 nl [67,68].

In conducting radiolabel efflux experiments, oocytes were pre-loaded with radiolabelled substrate using one of two different methods.

For the first method (Fig 5C and 5D), batches of 5 oocytes were pre-injected with unlabelled Arg calculated to an approximate cytosolic concentration of 5 mM as described in the preceding paragraph. This pre-injection was used to stimulate the subsequent loading (over a 1 or 2 hr incubation period) of 1 mM [¹⁴C]Arg. After the loading period the extracellular radiolabel was washed away and efflux measurements were conducted.

For the second method (Figs 5A and 5B and S8A and S8B), radiolabelled substrate was simply added to the external oocyte-bathing solution and the samples were then incubated at 16–18°C for 3–28 hr to allow the uptake of radiolabel. The loading-time varied, depending on whether oocytes were expressing TgApiAT6-1 or TgApiAT1 (in which case the radiolabel was taken up relatively quickly through these transporters) or whether the oocytes were the H₂O-injected controls (in which case radiolabel was taken up more slowly). The loading time was chosen so as to ensure that in each case the amount of radiolabel taken up by the oocytes was approximately the same. In the case of the H₂O-injected oocytes (i.e. oocytes not expressing TgApiAT6-1 or TgApiAT1) the preloading incubation time was typically 21 to 28 hr.

Pre-loading of oocytes with radiolabel was followed by quenching of the loading process by washing the oocytes in ice-cold ND96 solution, then initiation of the efflux by replacing the ice-cold solution with ambient-temperature solutions containing potential trans-stimulating substrates, as described in the figure legends. Groups of 5 oocytes at a time were incubated in 500 μl of extracellular solution for times indicated in individual figures, following which 100 μl was removed (for estimation of the extracellular concentration of radiolabel) and efflux immediately quenched by washing the oocytes with 4 × ice-cold ND96. The washed oocytes were transferred immediately to 96-well plates for estimation of the amount of radiolabel retained within the oocytes at the time of sampling.

To determine the efflux that was attributable to each of the two transporters of interest, the amounts of radioactivity measured in the extracellular medium and retained within the oocytes in the experiments with control H₂O-injected oocytes, were subtracted from those measured in TgApiAT6-1-expressing and TgApiAT1-expressing oocytes.

All oocyte uptake and efflux experiments were performed using either [¹⁴C]- or [³H]-labelled forms of the amino acid being studied, with the total substrate concentrations and concentration of [¹⁴C]- or [³H]-labelled amino acids specified in the relevant figure legends. The specific activities of [¹⁴C]-labelled compounds were as follows: Lys, 301 μCi/mmol; Arg, 312 μCi/mmol; Met, 68 μCi/mmol; Leu, 328 μCi/mmol; Ile, 328 μCi/mmol; Gln, 253 μCi/mmol; Glu, 281 μCi/mmol; Asp, 200 μCi/mmol; Ala, 151 μCi/mmol; Pro, 171 μCi/mmol; and Gly, 107 μCi/mmol (Gly). For [³H]-labelled compounds, specific activities were: putrescine, 28 Ci/mmol; spermidine, 32.35 Ci/mmol; and γ-[2-3-³H(N)]GABA, 89 Ci/mmol. Following the relevant incubation periods, oocytes or aliquots of solutions were distributed into OptiPlate96-well plates (Perkin-Elmer) and oocytes were lysed overnight in 10% (w/v) SDS. Microscint-40 scintillation fluid (Perkin-Elmer) was added to the samples, and plates covered and shaken for 5 min before radioactivity was counted on a Perkin-Elmer MicroBeta² 2450 microplate scintillation counter.

Oocyte protein detection assays

Oocyte surface biotinylation and whole membrane preparations were performed as described previously [63,66], using equivalent numbers of oocytes per experiment. During surface biotinylation, 15 oocytes were selected 4–6 days post cRNA injection, washed 3 × in ice-cold PBS (pH 8.0), and incubated for 25 mins at ambient temperature in 0.5 mg/ml of EZ-Link Sulfo-NHS-LC-Biotin (Thermo Fisher Scientific). To quench biotinylation, oocytes were washed 3 × in ice-cold PBS and then solubilised in lysis buffer (20 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% v/v Triton X-100) on ice for 1.5 hr. To remove unsolubilised material, samples were centrifuged at 16,000 × g, with the supernatant retained. To purify biotinylated proteins, the supernatant mixed was mixed with streptavidin-coated agarose beads (Thermo Fisher Scientific). The mixture was incubated with slow rotation for >2 hr before the beads were washed four times with lysis buffer then dissolved in SDS-PAGE sample buffer. For whole cell membranes, 25 oocytes were triturated in homogenisation buffer (50 mM Tris-HCl pH 7.4, 100 mM NaCl, 1 mM EDTA, protease inhibitors) and the resulting homogenate centrifuged at 2,000 × g at 4°C for 10 min to remove nuclei and cell debris, then at 300,000 × g for 1 hr at 4°C to pellet membranes. Samples were washed 3 × with homogenisation buffer and then solubilised with the same buffer containing 4% (w/v) SDS, and then prepared for SDS-PAGE.

Oocyte metabolite extraction and LC-MS/MS data acquisition

Xenopus laevis oocytes injected with either TgApiAT1 cRNA, TgApiAT6-1 cRNA or H₂O were incubated with substrate at concentrations, pH and temperatures indicated in figure legends. All incubations occurred in substrates solved in 1 × ND96 buffer as described previously [36]. For each condition, 12 oocytes were washed × 3 with 1 × ND96 solution not containing substrate under ambient conditions, before addition of 1 × ND96 containing substrate. Oocytes requiring the replacement of one incubation solution with another (e.g. to examine efflux following pre-loading) were washed × 4 in the ice-cold replacement 1 × ND96 solution (minus substrate) before the addition of 1 × ND96 containing substrate at ambient temperature. Oocyte incubations were quenched by placing oocyte-containing tubes on ice and washing × 4 with 1 ml of ice-cold MilliQ H₂O. Polar metabolites were extracted using a two-stage liquid-liquid phase extraction. The first extraction was in chloroform:water:methanol (1:1:3) to isolate aqueous metabolites. Oocytes were lysed and titurated in this mixture and then centrifuged at 13,000 × g for 5 min to remove cell debris and unsolved material. The second extraction involved adding 1:5 H₂O:mixture, which precipitated hydrophobic solutes. The supernatant was centrifuged again at 13,000 × g for 5 min, which left 3 clear phases: a floating aqueous phase of water/methanol, a dense organic phase and an interphase containing a clear white precipitate. The upper aqueous phase was removed and the organic phase and interphase discarded. Samples were desiccated on a vacuum centrifuge, then re-solved with acetonitrile/H₂O (80%/20% v/v).

Chromatographic separation was performed on an Ultimate 3000 RSLC nano Ultra high performance liquid chromatography (UHPLC) system (Dionex) by using hydrophilic interaction ion chromatography with a ZIC cHILIC column (3.0 μm polymeric, 2.1x150mm; Sequant, Merck) as described previously [36]. The gradient started with 80% mobile phase B (Acetonitrile; 0.1% v/v Formic acid) and 20% mobile phase A (10mM ammonium formate; 0.1% v/v formic acid) at a flow rate of 300 μl/min, followed by a linear gradient to 20% mobile phase B over 18 min. A re-equilibration phase of 12 mins using 80% mobile phase B was done with the same flow rate, making a total run time of 30 mins. The column was maintained at 40°C and the injection sample volume was 4μl. The mass detection was carried out by Q-Exactive Plus Orbitrap mass spectrometer (Thermo Scientific, Waltham, MA, USA) in positive electrospray mode. The following settings were used for the full scan MS: resolution 70,00, m/z range 60–900, AGC target 3 × 10⁶ counts, sheath gas 40 l min^-1, auxiliary gas 10 l min^-1, sweep gas 2 l min^-1, and capillary temperature 250°C, spray voltage +3.5kV. The MS/MS data was collected through data dependent top 5 scan mode using high-energy c-trap dissociation (HCD) with resolution 17,500, AGC target 1 × 10⁵ counts and normalized collision energy (NCE) 30. The rest of the specifications for the mass spectrometer remained unchanged from the vendor recommended settings. A pooled sample of all extracts was used as a quality control (QC) sample to monitor signal reproducibility and stability of analytes. Blank samples and QC samples were run before and after the batch and QC samples were run within the batch to ensure reproducibility of the data. Arg and Lys were calibrated as an external standard by first making serial concentration and then aliquoting the same volume from each (8 μl) to give the following amounts (pmol): 80, 200, 400, 800, 2000, 4800. Calibration curves for both metabolites were freshly determined with each new batch of LC-MS/MS run. The acquired raw metabolite data were converted into mzXML format and processed through an open source software MS-DIAL [69]. Identification of metabolites was performed by first using publicly available MS/MS libraries matching exact mass (MS tolerance 0.01 Da) and mass fragmentation pattern (MS tolerance 0.05 Da), and then further confirmed by standards through retention time where possible. Raw peak height was used for the quantification of metabolites.

Electrophysiological recordings in X. laevis oocytes

Single oocytes were recorded in either unclamped mode to record membrane potential (E_m) or in two-voltage clamp configuration at a set membrane potential to record membrane currents. Perfusion of different buffers and substrate solutions was controlled by valve release and stop, and perfusion rate either gravity-fed or controlled by a peristaltic pump (Gilson, Middleton, WI, U.S.A.). The E_m of oocytes was recorded in unclamped mode using conventional boronsilicate glass microelectrodes capillaries (World Precision Instruments) filled with 3 M KCl to a resistance of 5–10 MΩ. In two-voltage clamp configuration, the same experimental setup was followed with the exception that borosilicate glass microelectrodes were filled with 3M KCl with a tip resistance of: 1.5 ≥R_e≥ 0.5 MΩ. Oocytes were impaled and allowed to recover for 10 mins under constant perfusion to a steady-state E_m before recordings began. All E_m recordings were conducted in ND96 (pH 7.3) buffer. The amplifier was placed in set-up (current clamp) mode and the oocytes impaled with both the voltage sensing and current passing microelectrode. Before voltage clamping, the amplifier output current was set to zero to normalise currents recorded in voltage clamp mode. A test membrane potential pulse was also routinely administered and current output adjusted using amplifier gain and oscillation control (clamp stability), until the response time was sufficiently rapid (i.e. replicating the square wave-form with less than 5 μs maximal response time). Oocytes were clamped at −50 mV unless indicated otherwise.

All E_m and membrane current recordings were made with voltage commands generated using a Axon GeneClamp 500B amplifier (Axon Instruments, Union City, CA, U.S.A.) connected to 1×LU and 10×MGU head stages. Amplifier gain was set at × 10. All output signals were low-pass filtered at 1 kHz. The analogue signal was converted into digital by a Digidata 1322A (Axon Instruments), and data were sampled at 3–10 Hz using Clampex v 10.0 software (Axon Instruments). Various buffers of different salt composition were utilised during free voltage and two-voltage clamp recordings, the composition of which are provided in S3 Table.

Data analysis and statistics

Data analyses for the radiolabelled uptake experiments in parasites were performed using GraphPad Prism (Version 8). All oocyte data were analyzed using OriginPro (2015). All data displayed in figures represent the mean ± S.D. except where otherwise indicated. Unless uptake data from uninjected oocytes is included in figures, uptake in uninjected oocytes was subtracted from uptake in cRNA-injected oocytes to give the ‘transporter-dependent uptake’. All data sets assumed Gaussian normalcy which was tested by running a Shapiro-Wilk test prior to analysis. Normalcy was determined as at the P < 0.05 threshold. Multi-variant experiments with three or more experimental conditions were subjected to a one-way ANOVA with Dunnett’s post-hoc test and significance tested at the p < 0.05 level.

The setting of initial rate membrane transport conditions were established by fitting time-course uptake data in oocytes to: (1) which is a Box Lucas 1 model with zero offset [70], where S_t, and S₀, and are the amount of substrate (S) at variable time (t), or when t = 0, S_max is the vertical asymptote of substrate amount, and k is the 1^st order rate constant.

The theoretical reversal membrane potential (E_rev) for monovalent cationic amino acids (AA⁺) in oocytes (S9D Fig) was determined using the Nernst equation: (2)

Where, R, T and F have their usual values and meanings, z = +1 for the AA⁺ charge.

Calibration curves for LC-MS/MS quantification of Lys and Arg concentration in oocytes were fitted to a linear equation. Likewise, Lineweaver-Burke linear regressions of Michaelis-Menten steady-state kinetic data were also fitted to linear equations. All curve fittings were evaluated using adjusted goodness of fit R² values as quoted in figure legends. All non-linear fitting was conducted using the Levenburg-Marquardt algorithm, with iteration numbers varying from 4 to 11 before convergence was attained.