Courses.bio.unc.edu

Chloroquine Transport via the Malaria Parasite’s
Chloroquine Resistance Transporter
The following resources related to this article are available online at
www.sciencemag.org (this information is current as of December 25, 2009 ):
including high-resolution figures, can be found in the online Updated information and services,
version of this article at:
Supporting Online Material
cites 21 articles
subject collections
reprints
of this article or about obtaining permission to reproduce
this article in whole or in part can be found at:
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the ScienceAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright2009 by the American Association for the Advancement of Science; all rights reserved. The title uptake of [3H]CQ was measured in an acidicmedium (pH = 6.0), in which the majority of CQwas protonated. Oocytes expressing PfCRTCQR showed a marked (typically 5-fold, and up to 10-fold) increase in [3H]CQ uptake relative to non- injected controls and to oocytes expressingPfCRTCQS (Fig. 2A). This is consistent withPfCRTCQR, but not PfCRTCQS, mediating the Rowena E. Martin,1* Rosa V. Marchetti,1 Anna I. Cowan,2 Susan M. Howitt,1 transport of [3H]CQ. (The membrane potential and cytosolic pH in PfCRTCQS-expressing oocyteswere the same as those in PfCRTCQR-expressing The emergence and spread of chloroquine-resistant Plasmodium falciparum malaria parasites oocytes; table S2.) In contrast, oocytes injected has been a disaster for world health. Resistance is conferred by mutations in the Chloroquine Resistance Transporter (PfCRT), an integral membrane protein localized to the parasite’s native (that is, motif-replete, nonharmonized) internal digestive vacuole. These mutations result in a marked reduction in the accumulation of PfCRTCQR did not show increased [3H]CQ up- chloroquine (CQ) by the parasite. However, the mechanism by which this occurs is unclear. We take, nor was the protein present at significant expressed both wild-type and resistant forms of PfCRT at the surface of Xenopus laevis oocytes. The levels in the plasma membrane (fig. S3, A and B).
resistant form of PfCRT transported CQ, whereas the wild-type protein did not. CQ transport via The sequence modifications made here there- the mutant PfCRT was inhibited by CQ analogs and by the resistance-reverser verapamil. Thus, fore facilitated the functional expression of CQ resistance is due to direct transport of the drug via mutant PfCRT.
T76K and S163R (15) mutations in PfCRTCQR Malaria,aninfectiousdiseasethatisprev- Thedigestivevacuoleisalysosomalorganelle, restoreCQsensitivitytoCQRparasites(9,16).
and the targeting of PfCRT to this compartment The introduction of either one of these changes to is likely to be mediated by discrete endosomal- PfCRTCQR, each of which entailed the addition a drug that was cheap, safe, and effective. CQ lysosomal trafficking motifs. Upon expression of of a positive charge to the putative substrate- PfCRT in Xenopus oocytes, such motifs may cause binding site of the protein (10, 11), resulted in the falciparum Chloroquine Resistance Transporter the protein to be targeted to analogous organelles, loss of CQ transport activity (Fig. 2B). In con- (PfCRT) (1, 2) and is associated with a marked rendering direct measurements of PfCRT function trast, the introduction of K76T to PfCRTCQS did reduction in CQ accumulation by the parasite impractical. We therefore identified and removed not result in a significant increase in [3H]CQ up- (3, 4). CQ is a diprotic weak base (pKa of 8.1 and multiple putative trafficking motifs from both take (Fig. 2B). (PfCRTCQS K76T did localize to 10.2, where Ka is the acid dissociation constant), termini of the PfCRT protein sequence (fig. S1).
the oocyte plasma membrane; fig. S4.) The K76T with the relative proportions of the neutral, In addition, the PfCRT coding sequence for mutation is therefore necessary but not sufficient mono-protonated (CQH+), and di-protonated for the transport of CQ via PfCRT. This is con- ) species varying with pH (table S1).
harmonized to facilitate correct folding of the sistent with the other PfCRT mutations acting in The neutral species enters the parasite and its protein (12, 13). Hemagglutinin (HA)–tagged synergy with K76T to confer CQ resistance.
internal compartments via simple diffusion.
forms of this modified version of the PfCRT CQ transport showed a strong dependence on When the base enters the acidic environment of sequence were expressed in Xenopus oocytes, and the pH of the medium (Fig. 2C). Under alkaline the parasite’s digestive vacuole [pH ~ 5 (5–8)], conditions, [3H]CQ was taken up to similarly high the equilibrium is shifted toward the CQH 2+ PfCRT (PfCRTCQS and PfCRTCQR, respectively) levels in noninjected oocytes and oocytes ex- species, which is unable to diffuse across the to the oocyte plasma membrane was confirmed pressing PfCRTCQR or PfCRTCQS. This is likely membrane and becomes trapped, thereby accu- to represent simple diffusion of uncharged CQ mulating to high concentrations within this com- The successful expression of (motif-free, (uptake was nonsaturable at pH = 7.4 and 8.4; partment. CQ is thought to exert its antimalarial codon-harmonized) PfCRT at the oocyte surface fig. S5). In contrast, at pH = 5.0 to 6.9, CQ effect here by interfering with the detoxification enabled us to investigate the transport activity of transport in oocytes expressing PfCRTCQR was of heme, which is released as a byproduct of the protein. Except where specified otherwise, much higher than that in noninjected oocytes or The key resistance-conferring mutation in Fig. 1. Immunolocalization of PfCRT in the Brightfield
PfCRT is the replacement of a lysine (K) with a threonine (T) at position 76 (9). This K76T mutation occurs in a region of the protein that as the pigment layer. This, in turn, sur- is predicted to be involved in substrate recog- rounds a cytoplasm crowded with yolk sacs nition (10). It is never found in isolation, but is always accompanied by a number of what are organelles (23). Expression of C-terminally thought to be compensatory mutations in the pro- HA-tagged PfCRTCQR or PfCRTCQS results, in tein (11). We compared the function of mutant each case, in the appearance of a fluores- PfCRT from the CQ-resistant (CQR) P. falciparum indicating that both proteins are expressed strain Dd2 with that of wild-type PfCRT from the CQ-sensitive (CQS) strain D10 (fig. S1).
is not present in noninjected oocytes. Simi- lar results were obtained with N-terminally Research School of Biology, The Australian National Uni- versity, Canberra, Australian Capital Territory 0200, Aus- tralia. 2The John Curtin School of Medical Research, The Australian National University, Canberra, Australian Capi- *To whom correspondence should be addressed. E-mail:[email protected] Non-injected
25 SEPTEMBER 2009 VOL 325 SCIENCE www.sciencemag.org oocytes expressing PfCRTCQS [P < 0.05, analysis tential undergoing a depolarization over the same Vmax of 67 T 13 pmol hour−1 per oocyte (Fig. 2E, of variance (ANOVA)], with the maximum dif- inset) for the transport of CQ via PfCRTCQR.
ference (that is, the pH optimum for PfCRTCQR- (The presence of a high extracellular concentra- mediated CQ transport) observed at pH = 6.0 creased with increasing concentrations of un- tion of CQ did not affect the membrane poten- (fig. S6). This pH dependence is consistent with labeled CQ (Fig. 2E), which is consistent with a tial or cytosolic pH of noninjected PfCRTCQS- or CQ being transported in its mono- or di-protonated saturable transport mechanism. In contrast, raising forms. Indeed, depolarization of the membrane the concentration of unlabeled CQ had little ef- Verapamil increases the accumulation of CQ potential by the replacement of extracellular Na+ fect on [3H]CQ transport in PfCRTCQS-expressing by resistant parasites in vitro and thereby increases with K+ (at pH = 6.0; table S3) resulted in a 25 T and noninjected oocytes. This is consistent with their sensitivity to CQ (4). Verapamil inhibited 2% reduction in PfCRTCQR-mediated CQ transport the entry of the drug into these oocytes being via the transport of CQ via PfCRTCQR (Fig. 2F and (Fig. 2D; P < 0.005, paired t test). The decrease simple diffusion of the neutral species. A least- Table 1; half-maximum inhibitory concentration in PfCRTCQR-mediated CQ transport as pH was squares fit of the data to the Michaelis-Menten IC50 = 30 T 3 mM), as did a range of quinolines reduced from 6.9 to 5.5 (Fig. 2C) may have been equation yielded an apparent Michaelis constant including quinine and amodiaquine (table S5). In due, at least in part, to the oocyte membrane po- KM(CQ) of 245 T 3 mM and a maximum velocity contrast, piperaquine and artemisinin (both clin- Time (min)
[CQ] (µM)
High [K+]
[CQ] (µM)
[Verapamil] (µM)
Fig. 2. Transport properties of PfCRTCQR in Xenopus oocytes. (A) Oocytes bars). (D) Effect of depolarization of the oocyte plasma membrane (by expressing PfCRTCQR (solid circles) showed a marked increase in CQ trans- replacement of Na+ with K+ in the extracellular medium) on [3H]CQ up- port relative to noninjected oocytes (solid triangles) and oocytes expressing take into noninjected oocytes (white bars), oocytes expressing PfCRTCQS PfCRTCQS (open circles). Rates of CQ uptake (pmol hour−1 per oocyte; n = 3 T (gray bars), and oocytes expressing PfCRTCQR (black bars). PfCRTCQR- SEM, estimated from uptake at 60 min) were as follows: noninjected, 1.14 T expressing oocytes, but not PfCRTCQS-expressing or noninjected oocytes, 0.16; PfCRTCQS-, 1.14 T 0.19; and PfCRTCQR, 5.54 T 0.44. The PfCRTCQR- showed a significant decrease in CQ uptake when depolarized [P < 0.005 mediated uptake of [3H]CQ (obtained by subtracting uptake in oocytes and P > 0.05 (ANOVA), respectively]. (E) Effect of unlabeled CQ on the expressing PfCRTCQR from that in PfCRTCQS-expressing oocytes) was ap- uptake of [3H]CQ by noninjected oocytes (solid triangles) and oocytes proximately linear with time for at least 4 hours (inset). (B) Introduction expressing either PfCRTCQR (solid circles) or PfCRTCQS (open circles). The of K76T to PfCRTCQS did not increase CQ transport to above that mea- inset shows the [CQ]-dependence of PfCRTCQR-mediated uptake, which sured in oocytes expressing PfCRTCQS or in noninjected (ni) oocytes (P > was calculated by subtracting the uptake measured in oocytes express- 0.05, ANOVA). The introduction of T76K or S163R to PfCRTCQR resulted ing PfCRTCQS from that in oocytes expressing PfCRTCQR at each CQ in the loss of PfCRTCQR-associated CQ transport ([3H]CQ uptake in these concentration. (F) Inhibition by verapamil of the uptake of [3H]CQ by oocytes did not differ significantly from that in noninjected oocytes or oocytes expressing PfCRTCQR (solid circles) or PfCRTCQS (open circles). In from oocytes expressing PfCRTCQS; P > 0.05, ANOVA). (C) pH depen- all panels, uptake is shown as mean T SEM from three to five separate dence of [3H]CQ uptake into noninjected oocytes (white bars), oocytes experiments, within which measurements were made from 10 oocytes per expressing PfCRTCQS (gray bars) and oocytes expressing PfCRTCQR (black www.sciencemag.org SCIENCE VOL 325 25 SEPTEMBER 2009 8. T. N. Bennett et al., Mol. Biochem. Parasitol. 133, 99 Table 1. IC50 values for the inhibition of PfCRTCQR-mediated CQ transport by a number of drugs and peptides. PfCRTCQR-mediated CQ transport was calculated by subtracting the uptake measured in oocytes 9. V. Lakshmanan et al., EMBO J. 24, 2294 (2005).
expressing PfCRTCQS from that in oocytes expressing PfCRTCQR. The data are shown in fig. S7 and Fig. 2F.
10. R. E. Martin, K. Kirk, Mol. Biol. Evol. 21, 1938 IC50 values were derived by least-squares fit of the equation Y = Ymin + [(Ymax – Ymin)/(1 + ([inhibitor]/IC50)C], where Y is PfCRTCQR-mediated CQ transport, Ymin and Ymax are the minimum and maximum values of Y, and 11. P. G. Bray et al., Mol. Microbiol. 56, 323 (2005).
12. P. Cortazzo et al., Biochem. Biophys. Res. Commun. 293, C is a constant. All values are mean T SEM from n = 3 or 4 separate experiments, within which mea- surements were made from 10 oocytes per treatment.
13. A. A. Komar, T. Lesnik, C. Reiss, FEBS Lett. 462, 387 14. Materials and methods are available as supporting 15. Single-letter abbreviations for the amino acid residues are as follows: F, Phe; G, Gly; H, His; K, Lys; L, Leu; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.
16. D. J. Johnson et al., Mol. Cell 15, 867 (2004).
17. D. A. van Schalkwyk, T. J. Egan, Drug Resist. Updat. 9, 18. P. G. Bray et al., Mol. Microbiol. 62, 238 (2006).
19. C. P. Sanchez et al., Biochemistry 44, 9862 (2005).
20. C. P. Sanchez et al., Mol. Microbiol. 64, 407 (2007).
21. B. Naude, J. A. Brzostowski, A. R. Kimmel, T. E. Wellems, ically effective against both CQS and CQR par- sensitive) CQ-mediated efflux of H+ from the 22. A. M. Lehane, K. Kirk, Antimicrob. Agents Chemother. 52, asites) had no effect. Amantadine exhibits some digestive vacuole of CQR parasites (22). The antimalarial activity in vitro, particularly against achievement of a robust expression system for 23. C. A. Wagner, B. Friedrich, I. Setiawan, F. Lang, S. Bröer, CQR parasites (16), and also inhibited transport PfCRT has the potential to facilitate the rational Cell. Physiol. Biochem. 10, 1 (2000).
design of novel CQ-like drugs that bypass the 24. We thank J. Abbey, R. Summers, E. Baker, and R. Slatyer for technical assistance. This work was supported by the Several peptides were found to cause a pro- resistance mechanism and/or the design of clin- Australian National Health and Medical Research Council nounced inhibition of CQ transport via PfCRTCQR ically effective resistance-reversing agents.
(NHMRC) (grant 471472) and the Australian Research (table S5). Most of the peptides that are active Council (grant DP0559433). R.E.M. was supported by an against PfCRTCQR have key elements of the CQ- NHMRC Australian Biomedical Fellowship (fellowship resistance reverser pharmacophore [hydrogen bond 1. D. A. Fidock et al., Mol. Cell 6, 861 (2000).
2. A. B. Sidhu, D. Verdier-Pinard, D. A. Fidock, Science 298, acceptor and two hydrophobic aromatic rings (17)] (table S6). This pharmacophore can be viewed as www.sciencemag.org/cgi/content/full/325/5948/1680/DC1 3. C. D. Fitch, Proc. Natl. Acad. Sci. U.S.A. 64, 1181 (1969).
defining the basic elements involved in interactions 4. D. J. Krogstad et al., Science 238, 1283 (1987).
between PfCRTCQR and substrates or inhibitors.
5. R. Hayward, K. J. Saliba, K. Kirk, J. Cell Sci. 119, 1016 The concentration dependence of inhibition 6. N. Klonis et al., Biochem. J. 407, 343 (2007).
of CQ transport was determined for a number of 7. Y. Kuhn, P. Rohrbach, M. Lanzer, Cell. Microbiol. 9, 1004 compounds (Table 1 and fig. S7). YPWF-NH2 (endomorphin-1; an opioid receptor agonist) wasthe most effective peptide inhibitor of PfCRTCQR- mediated CQ uptake, with an IC50 comparableto that of quinine and verapamil. Measurements of [3H]YPWF-NH2 uptake in oocytes express-ing different PfCRT constructs revealed that PfCRTCQR, but not PfCRTCQR-T76K, PfCRTCQR-S163R, or PfCRTCQS, mediates the transport ofthis peptide (fig. S8).
via mutant PfCRT, which provides an explana- Liam J. Holt,1* Brian B. Tuch,2* Judit Villén,3* Alexander D. Johnson,2 tion for the phenomenon of CQ resistance, as well as for the reversal of CQ resistance by reversingagents such as verapamil. The presence of a posi- To explore the mechanisms and evolution of cell-cycle control, we analyzed the position tive charge (K76 or R163) in the PfCRT substrate- and conservation of large numbers of phosphorylation sites for the cyclin-dependent kinase Cdk1 in the budding yeast Saccharomyces cerevisiae. We combined specific chemical inhibition interacting with the transporter. The K76T muta- of Cdk1 with quantitative mass spectrometry to identify the positions of 547 phosphorylation tion removes the positive charge, altering the sub- sites on 308 Cdk1 substrates in vivo. Comparisons of these substrates with orthologs throughout strate specificity of PfCRT to allow the transport the ascomycete lineage revealed that the position of most phosphorylation sites is not of the protonated drug. In the parasite, the pres- conserved in evolution; instead, clusters of sites shift position in rapidly evolving disordered ence of mutant PfCRT on the digestive vacuole regions. We propose that the regulation of protein function by phosphorylation often will allow the protonated drug to be transported depends on simple nonspecific mechanisms that disrupt or enhance protein-protein interactions.
down its electrochemical gradient, out of the vac- The gain or loss of phosphorylation sites in rapidly evolving regions could facilitate the uole, and thus away from its site of action (fig. S9).
evolution of kinase-signaling circuits.
This mechanism is consistent with recent studiesimplicating PfCRTCQR in the transport of [3H]CQ and provide insights into the mechanisms and in CQR parasites (18–20) and in Dictyostelium evolution of regulation by phosphorylation. We discoideum transformants expressing PfCRT at therefore developed methods for comprehensive endosomal membranes (21). It is also consistent tification and analysis of Cdk substrates would identification of the sites of Cdk1 phosphoryl- with the recent demonstration of a (verapamil- enhance our understanding of cell-cycle control ation on large numbers of substrates in vivo. We 25 SEPTEMBER 2009 VOL 325 SCIENCE www.sciencemag.org

Source: http://courses.bio.unc.edu/2010Spring/Biol402/Papers%20for%20Discussion/chloroquinetransport.pdf