Trypanosoma brucei aquaglyceroporin 2 is a high-affinity transporter for pentamidine and melaminophenyl arsenic drugs and the main genetic determinant of resistance to these drugs

Abstract

Objectives: Trypanosoma brucei drug transporters include the TbAT1/P2 aminopurine transporter and the high-affinity pentamidine transporter (HAPT1), but the genetic identity of HAPT1 is unknown. We recently reported that loss of T. brucei aquaglyceroporin 2 (TbAQP2) caused melarsoprol/pentamidine cross-resistance (MPXR) in these parasites and the current study aims to delineate the mechanism by which this occurs.

Methods: The TbAQP2 loci of isogenic pairs of drug-susceptible and MPXR strains of T. brucei subspecies were sequenced. Drug susceptibility profiles of trypanosome strains were correlated with expression of mutated TbAQP2 alleles. Pentamidine transport was studied in T. brucei subspecies expressing TbAQP2 variants.

Results: All MPXR strains examined contained TbAQP2 deletions or rearrangements, regardless of whether the strains were originally adapted in vitro or in vivo to arsenicals or to pentamidine. The MPXR strains and AQP2 knockout strains had lost HAPT1 activity. Reintroduction of TbAQP2 in MPXR trypanosomes restored susceptibility to the drugs and reinstated HAPT1 activity, but did not change the activity of TbAT1/P2. Expression of TbAQP2 sensitized Leishmania mexicana promastigotes 40-fold to pentamidine and >1000-fold to melaminophenyl arsenicals and induced a high-affinity pentamidine transport activity indistinguishable from HAPT1 by Km and inhibitor profile. Grafting the TbAQP2 selectivity filter amino acid residues onto a chimeric allele of AQP2 and AQP3 partly restored susceptibility to pentamidine and an arsenical.

Conclusions: TbAQP2 mediates high-affinity uptake of pentamidine and melaminophenyl arsenicals in trypanosomes and TbAQP2 encodes the previously reported HAPT1 activity. This finding establishes TbAQP2 as an important drug transporter.

Keywords: aquaporin; drug transport; parasite; protozoan; resistance mutation.

Figure 1.

Expression of TbAQP2 in multidrug-resistant trypanosomes sensitizes to (a) pentamidine and (b) Cymelarsan but not (c) diminazene or (d) phenylarsine oxide. EC₅₀ values were obtained using the Alamar blue assay and bars represent the mean and SEM of 3 to >10 independent determinations; data were analysed for significant differences from the wild-type control using one-way ANOVA/Tukey's test (GraphPad Prism 5.0). NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; unless otherwise indicated, relative to wild-type controls. EV, empty-vector control.

Figure 2.

Expression of TbAQP2 and chAQP2/3 in aqp2 null and aqp2/aqp3 null trypanosomes. EC₅₀ values were determined for (a) pentamidine, (b) diminazene, (c) Cymelarsan and (d) phenylarsine oxide. All data are the mean of ≥10 independent determinations. See legend of Figure 1 for details. Significance was tested relative to the wild-type control unless otherwise indicated.

Figure 3.

Localization of GFP-chAQP2/3 expressed in bloodstream-form aqp2/aqp3 null trypanosomes. (a) Western blot using anti-GFP antiserum as described previously. Blotting was performed after induction with 1 μg/mL tetracycline (+) or without induction as control (−). The Coomassie-stained panel shows relative loading. (b) GFP-chAQP2/3 was localized to the plasma membrane. Blue colour is DAPI staining of nucleus and kinetoplast. The scale bar represents 10 μm.

Figure 4.

Uptake of 30 nM [³H]pentamidine by aqp2 null and wild-type control cells. Cells of T. b. brucei strain 2T1 (closed symbols) or the derived aqp2 null strain (open symbols) were incubated with [³H]pentamidine for 90 s in the presence or absence of various concentrations of unlabelled pentamidine (circles) or propamidine (squares). The incubation was stopped by the addition of 1 mL ice-cold 1 mM pentamidine and immediate centrifugation through oil. The experiment shown is representative of four identical but independent experiments, each performed in triplicate and showing virtually identical outcomes. Data points are mean ± SEM of triplicates.

Figure 5.

[³H]pentamidine uptake by various s427-derived trypanosome lines. (a) [³H]pentamidine concentration was 50 nM, reflecting predominantly HAPT1-mediated uptake. (b) [³H]pentamidine concentration was 1 μM, reflecting predominantly LAPT1 uptake. Rates were determined from the slopes of time courses over 10 min with timepoints at 0, 1, 3, 5, 7.5 and 10 min. Lines were linear, with none of the lines showing significant deviation from linearity in a runs test (GraphPad Prism 5.0). Slopes of control time courses in the presence of 1 mM unlabelled pentamidine were all non-significantly different from zero (F-test, GraphPad Prism 5.0), whereas uptake with unlabelled inhibitor was highly significantly different from zero (typically P < 0.0001). Results are the mean of three to four independent experiments, each performed in triplicate. (c, d) Uptake of [³H]pentamidine in bloodstream forms of B48/AQP2 was determined as described in the legend to Figure 4. The experiments shown are representative of three independently performed replicates. (c) Uptake of 30 nM [³H]pentamidine. (d) Uptake of 1 μM [³H]pentamidine. Solid squares, coincubation with various concentrations of unlabelled pentamidine; open circles, coincubation with propamidine. **P < 0.01; ***P < 0.001.

Figure 6.

Effect of expression of T. brucei aquaporins on drug susceptibility of Leishmania mexicana. TbAQP2 and TbAQP3 were expressed in promastigotes using the pNUS vector. Two independent clones of each resulting cell line and the promastigotes transfected with the ‘empty’ pNUS vector were tested for sensitivity to (a) pentamidine, (b) Cymelarsan, (c) diminazene and (d) amphotericin B using the Alamar blue fluorimetric assay. Bars are means of three to eight independent determinations; error bars are SEM. ***P < 0.001 by one-way ANOVA with Tukey's correction (GraphPad Prism 5.0). EV, empty-vector control.

Figure 7.

Expression of T. b. brucei aquaporins in promastigotes of Leishmania mexicana. (a) Specific uptake of 100 nM [³H]pentamidine over 5 min in L. mexicana promastigotes transfected with empty pNUS vector (control), or promastigotes transfected with TbAQP2 or with TbAQP3. In each case, mediated uptake of 100 nM radiolabel was compared with total association of [³H]pentamidine with the cell pellet in the presence of a saturating concentration (1 mM) of unlabelled pentamidine. The data shown are the mean of triplicates ± SEM. ***P < 0.001 by one-way ANOVA, compared with all other datasets. (b) Time course of 50 nM [³H]pentamidine uptake, over 15 s, using L. mexicana promastigotes transformed with TbAQP2 in the presence and absence of 1 mM unlabelled pentamidine. Uptake at 50 nM pentamidine was linear (r² = 0.98) and rapid (0.032 ± 0.002 pmol/10⁷ cells/s, compared with 0.00026 ± 1.8 × 10⁻⁶ pmol/10⁷ cells/s in the presence of 1 mM pentamidine). (c) Characterization of 20 nM [³H]pentamidine uptake in L. mexicana promastigotes expressing TbAQP2, in the presence of varying concentrations of unlabelled inhibitor. (d) Michaelis–Menten plot of 20 nM [³H]pentamidine uptake; conversion of pentamidine inhibition plot in (c).

Figure 8.

Expression of synthetic aquaporin constructs in bloodstream forms of the aqp2/aqp3 double-null line. EC₅₀ values were determined for (a) pentamidine, (b) Cymelarsan, (c) diminazene and (d) phenylarsine oxide. Expression was induced by the addition of 1 μg/mL tetracycline 24 h before setting up the plates for the assay, followed by 48 h of incubation of the cells in the presence of a doubling dilution of the test compound followed by a further 24 h of incubation with the dye before fluorescence was determined. All data are the mean of at least three independent determinations; error bars are SEM. Statistical significance with aqp2/aqp3 null was determined using a one-way ANOVA with Tukey's correction (GraphPad Prism 5.0); **P < 0.02; ***P < 0.01. In the Cymelarsan panel (b), it is also indicated that Groups 4 and 5 are significantly different from Group 3.