Chronic exposure to arsenic in drinking water can ... - Semantic Scholar

Download PDF
32 downloads 0 Views 731KB Size Report
Comment
Leishmania donovani was serially passaged in mice exposed to ar- senic in drinking water at ... reported in human liver biopsies. After five monthly passages in ...
Chronic exposure to arsenic in drinking water can lead to resistance to antimonial drugs in a mouse model of visceral leishmaniasis Meghan R. Perrya, Susan Wylliea, Andrea Raabb, Joerg Feldmannb, and Alan H. Fairlamba,1 a

Division of Biological Chemistry and Drug Discovery, College of Life Sciences, University of Dundee, Dundee DD1 5EH, Scotland, United Kingdom; and bCollege of Physical Sciences–Chemistry, Trace Element Speciation Laboratory, University of Aberdeen, Aberdeen AB24 3UE, Scotland, United Kingdom

Edited by Thomas E. Wellems, National Institutes of Health, Bethesda, MD, and approved October 1, 2013 (received for review June 18, 2013)

The Indian subcontinent is the only region where arsenic contamination of drinking water coexists with widespread resistance to antimonial drugs that are used to treat the parasitic disease visceral leishmaniasis. We have previously proposed that selection for parasite resistance within visceral leishmaniasis patients who have been exposed to trivalent arsenic results in cross-resistance to the related metalloid antimony, present in the pentavalent state as a complex in drugs such as sodium stibogluconate (Pentostam) and meglumine antimonate (Glucantime). To test this hypothesis, Leishmania donovani was serially passaged in mice exposed to arsenic in drinking water at environmentally relevant levels (10 or 100 ppm). Arsenic accumulation in organs and other tissues was proportional to the level of exposure and similar to that previously reported in human liver biopsies. After five monthly passages in mice exposed to arsenic, isolated parasites were found to be completely refractory to 500 μg·mL−1 Pentostam compared with the control passage group (38.5 μg·mL−1) cultured in vitro in mouse peritoneal macrophages. Reassessment of resistant parasites following further passage for 4 mo in mice without arsenic exposure showed that resistance was stable. Treatment of infected mice with Pentostam confirmed that resistance observed in vitro also occurred in vivo. We conclude that arsenic contamination may have played a significant role in the development of Leishmania antimonial resistance in Bihar because inadequate treatment with antimonial drugs is not exclusive to India, whereas widespread antimonial resistance is.

|

drug resistance sodium antimony gluconate environmental pollution

| treatment failure |

V

isceral leishmaniasis (VL) is a systemic illness caused by the obligate intracellular protozoan parasites Leishmania donovani and Leishmania infantum. The parasite, in amastigote form, multiplies within macrophages of the spleen, liver, and bone marrow causing fever, anorexia, weight loss, and hepatosplenomegaly (1). Effective chemotherapy for this condition is essential because untreated VL is fatal and accounts for 41,000 deaths per year (2). Currently, there are four main antileishmanial drugs available: antimonial preparations, amphotericin B, miltefosine, and paromomycin (3). Antimonial preparations have been used for almost a century and remain an essential part of the treatment of VL in South America and sub-Saharan Africa (2). However, by the end of the twentieth century, their efficacy in Bihar, which houses 90% of India’s large VL burden (4), had decreased to cure rates of less than 50% (5). Consequently, use of antimonial drugs is no longer recommended in the Indian subcontinent (6). The underlying reasons for this epidemic of resistance in Bihar are not fully understood. In our previous publication we proposed that the presence of arsenic in drinking water in Bihar has contributed to the gradual decline in efficacy of antimonial preparations for VL in this region (7). Subsequent to the 1970s, when there was a large-scale insertion of shallow tube wells to provide clean drinking water in India, it was found that the Bihari population was at risk from arsenic exposure due to contamination from naturally occurring trivalent arsenic in the 19932–19937 | PNAS | December 3, 2013 | vol. 110 | no. 49

groundwater accessed by these wells (8). The elements antimony and arsenic have a long therapeutic history, are closely related in the periodic table, and share many similar chemical properties (9). In the 1980s, when Leishmania antimonial resistance was first suspected, parasitologists used stepwise exposure to trivalent arsenic in the laboratory to induce antimonial cross-resistance and study mechanisms of resistance (10). Our hypothesis is that a similar selection process could occur in VL patients who have been chronically exposed to environmental arsenic, such that selection for parasite resistance to arsenic would result in crossresistance to antimonial drugs. The incidence of arsenic contamination, VL, and antimonial resistance from all of the available data from surveys, journals, and Web sites was previously collated by our group (7). These data showed that in 10 out of 38 districts in Bihar, arsenic exposure, endemic VL, and reported antimonial resistance coexist. Thus, in these areas, arsenic contamination of the groundwater has the potential to contribute to the development of Leishmania parasite antimonial resistance, giving epidemiological plausibility to the hypothesis. In this study we test our hypothesis using a mouse model of arsenic exposure and chronic VL. Our findings demonstrate that Leishmania parasites resistant to pentavalent antimonials can be created through oral exposure to arsenic in drinking water of mice infected with L. donovani. Results Quantitation of Arsenic Exposure in BALB/c Mice by Inductively Coupled Plasma Mass Spectrometry (ICP-MS). The hypothesis that Leishmania

parasites could become resistant to antimonial preparations Significance Bihar, India, is the only region where arsenic contamination of drinking water coexists alongside endemic visceral leishmaniasis and where widespread resistance to pentavalent antimonial drugs (e.g., Pentostam) occurs. We have proposed that selection for parasite resistance in infected individuals exposed to environmental arsenic results in cross-resistance to the related metalloid antimony. To test this hypothesis, we have serially passaged Leishmania donovani in mice receiving arsenic in drinking water at environmentally relevant levels. In support of this hypothesis, we show that after five monthly passages, Leishmania parasites become stably resistant to Pentostam in vitro and in vivo. Author contributions: M.R.P., S.W., J.F., and A.H.F. designed research; M.R.P., S.W., and A.R. performed research; M.R.P., S.W., A.R., J.F., and A.H.F. analyzed data; and M.R.P., S.W., A.R., J.F., and A.H.F. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. Freely available online through the PNAS open access option. See Commentary on page 19666. 1

To whom correspondence should be addressed. E-mail: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1311535110

SEE COMMENTARY

Fig. 1. Arsenic levels in organs of BALB/c mice. At day 28 (A) and day 56 (B), three mice from each group (clear bars, 0 ppm; gray bars, 10 ppm; and black bars, 100 ppm) were culled, and liver, spleen, bone, and kidneys were harvested, homogenized, and digested before analysis of their total arsenic content by ICPMS (39). In the case of spleen, kidney, and bone of the D28 control samples, these are the weighted mean of two independent analyses. Values are mean ± SE in mg per kg wet weight. *P < 0.05 compared with 0-ppm group.

through exposure to arsenic within their host relies on the presence of arsenic in the organs in which they reside. This was assessed through exposing uninfected BALB/c mice to arsenic at 0, 10, and 100 ppm for a total of 56 d and measuring arsenic accumulation in organs and other tissues by ICP-MS. The mice in the 10-ppm group, which is equivalent to tube well arsenic concentrations found in Bihar, gained weight in a similar fashion to the control group at 0 ppm and had an average fluid intake of 2 mL·d−1 (equivalent to 20 μg arsenic per mouse per day). However, mice in the 100-ppm group gained weight at a slower rate, and their fluid intakes progressively decreased from an average of 1 mL·d−1 to 0.5 mL·d−1 (equivalent to 50–100 μg arsenic per mouse per day), indicating cumulative toxicity at the higher arsenic concentration. The arsenic content of standard laboratory chow was determined to be 0.048 ± 0.005 μg·g−1. Food intake was not measured, but based on a daily intake of 16 g per 100 g body weight for BALB/cByJ mice (11), arsenic intake from food is negligible (0.15 μg per mouse per day). The tissue levels of arsenic correlated with the concentration of arsenic in the animals’ drinking water (Fig. 1). In mice exposed to 10 ppm arsenic, levels were maintained or increased between day 28 (Fig. 1A) and day 56 (Fig. 1B), times corresponding to the start and end of the experimental infection. However, in the 100ppm group, the mean arsenic content of all tissues except the spleen decreased, most likely as a result of the decreased fluid intake in the 100-ppm group. The mean arsenic contents of infected liver and spleens isolated on day 56 were comparable to those from uninfected ani-

mals (Table 1), suggesting that the presence of the Leishmania parasite did not affect the metabolism of arsenic. Likewise, the level of residual total antimony in livers of infected mice 72 h after the end of treatment with Pentostam was not significantly altered between non–arsenic-exposed and arsenic-exposed groups (Table 1), suggesting that exposure to arsenic does not interfere with the accumulation of antimony in the infected tissues. The amounts of infected spleen from the 0-ppm group available for arsenic analysis (Table 1) were limited as a result of the need to isolate parasites for further passage into mice and for drug sensitivity assays. The high SE for the spleen samples from the control mice is a result of the low arsenic content and the low sample amounts resulting in highly variable count rates during ICP-MS analysis indicating a large relative variability of arsenic in the control group. An independent measurement of splenic arsenic content from three infected mice at 28 d was performed using larger splenic samples, and the results are shown in parentheses in Table 1. The elevated levels of arsenic seen in the liver and kidney would be expected because these organs are the main sites of arsenic metabolism and excretion, respectively (12). Liver biopsy of chronically exposed patients in Bangladesh demonstrated high levels of arsenic, between 0.5 and 6 mg per kg dry weight (13). Levels between 0.4 and 1.5 mg per kg wet weight were seen in the livers of BALB/c mice exposed to arsenic at 100 ppm throughout the arsenic exposure time course (days 7, 14, 28, and 56). These experimental results are comparable to those found with similar exposures in C3H mice in the literature (14, 15).

Table 1. Total arsenic and antimony in organs of BALB/c mice exposed to arsenic in drinking water

Organ Liver Infected Uninfected Spleen Infected Uninfected

Total antimony, mg·kg−1

0 ppm

10 ppm

100 ppm

0 ppm

10 ppm

100 ppm

0.028 ± 0.011 0.0020 ± 0.0006

0.12 ± 0.05 0.12 ± 0.02**

0.41 ± 0.09*** 0.53 ± 0.14**

2.3 ± 0.1 ND

2.2 ± 0.4 ND

2.0 ± 0.1 ND

0.12 ± 0.08 (0.01 ± 0.0004)† 0.0026 ± 0.0007

0.23 ± 0.18 0.12 ± 0.02**

0.35 ± 0.28 0.40 ± 0.12*

2.4 ± 0.2 ND

2.6 ± 0.1 ND

1.1 ± 0.2* ND

Arsenic levels were measured by ICP-MS in the livers and spleens of mice that have been exposed to 0, 10, and 100 ppm for 56–63 d. The results represent the mean and SE of measurements from three infected mice at passages 2, 3, 4, and 5 for liver and at passages 3 and 5 for spleen and from uninfected mice at 56 d. Total antimony results are mean and SE of measurements from three mice from each arsenic exposure level at fifth passage, 72 h post-s.c. injection of 50 mg·kg−1 of Pentostam. ND, not determined. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with control. † Results of an independent experiment.

Perry et al.

PNAS | December 3, 2013 | vol. 110 | no. 49 | 19933

MICROBIOLOGY

Total arsenic, mg·kg−1

Generation of a Pentostam-Resistant Leishmania Line Through in Vivo Exposure to Arsenic. LV9 parasites which were exposed to arsenic

in murine hosts in serial passages (Fig. 2) developed resistance to antimonial preparations when assessed in vitro in macrophages (Fig. 3). The level of ex vivo amastigote sensitivity to Pentostam in the in-macrophage assay gradually decreased at each passage. After the third passage the EC50 for parasites exposed at 10 ppm was fivefold greater than the 0-ppm control group (0 ppm, 43 ± 6 μg·mL−1; 10 ppm, 228 ± 75 μg·mL−1), and no response to Pentostam was seen at 500 μg·mL−1 for 100 ppm-exposed parasites. After the fifth passage, the 0-ppm mice had an EC50 of 38.5 ± 3.9 μg·mL−1, and no response to Pentostam at 500 μg·mL−1 was observed in either of the arsenic-exposed parasite groups (Fig. 3). Following passage in mice for a further 4 mo without arsenic exposure, complete resistance to Pentostam at 500 μg·mL−1 was retained, whereas the control group had an EC50 of 60.6 ± 13 μg·mL−1. In Vivo Assessment of Leishmania Resistance. On the fifth passage, BALB/c mice exposed to arsenic at 0, 10, and 100 ppm were treated in vivo with drug vehicle, Pentostam, or miltefosine. There was a significant difference in response (P < 0.05) to Pentostam treatment with less than 20% suppression of parasite load in the arsenic-exposed groups compared with 66.4% suppression in the 0-ppm control group (Fig. 4). In Fig. 4A, miltefosine at 12 mg·kg−1 had a similar effect on the parasite burden in all three groups. When the data are presented as a percentage of the control in Fig. 4B, there is a significant difference between the response of 0-ppm and 10-ppm groups, which may be due to variability in parasite burden in the control groups. The lack of efficacy of Pentostam in vivo in the arsenic-exposed groups confirms that the cross-resistance developed by the LV9 parasites is relevant in vivo as well as in vitro.

Discussion The isolated development of decreased antimonial efficacy in Bihar, India, when the drug remains efficacious worldwide in other VL-endemic communities has previously been attributed to erratic prescribing practices and India’s unregulated private healthcare system (16, 17). This study demonstrates, through an in vivo model, that it is biochemically and physiologically plausible that the transmission of L. donovani through arsenic-exposed populations has contributed to the development of antimonial-resistant parasites. The Indian subcontinent is the only region where antimonial

Fig. 2. Procedure for generation of an antimonial-resistant line in vivo through oral exposure to arsenic.

19934 | www.pnas.org/cgi/doi/10.1073/pnas.1311535110

Fig. 3. Susceptibility of in-macrophage ex vivo amastigotes to Pentostam. Ex vivo amastigotes recovered from BALB/c mice exposed to 10 ppm (■) and 100 ppm (▲) arsenic were assessed for their sensitivity to Pentostam for 72 h within mouse peritoneal macrophages (38). Amastigotes from the 0-ppm (●) BALB/c mice had an EC50 value for Pentostam of 38.5 ± 3.9 μg·mL−1, whereas parasites recovered from the 100-ppm group of mice remained insensitive to the drug at concentrations up to 500 μg·mL−1.

efficacy is less than 50% (5), and it is the only region where arsenic contamination of the groundwater and VL coexist (18). Arsenic has long been known to induce cross-resistance to antimony in vitro (10), but this phenomenon has not been demonstrated in vivo. This study shows that the speed of resistance development is related to the levels of arsenic exposure: at high levels of hepatic arsenic of between 0.4 and 1.5 mg·kg−1 seen in the 100-ppm group, resistance in L. donovani parasites developed relatively rapidly over a few months, whereas resistance took longer to develop at the lower level of exposure of 10 ppm. This latter level, being equivalent to a 60-kg person drinking 3 L of water contaminated with arsenic at 1,200 μg·L−1 (19), is in the upper bracket of arsenic concentrations found in Bihar (8). The limitations of this work concern how far the models we have used can relate to the development of antimonial resistance in Leishmania in patients in Bihar. Our model of arsenic exposure used high concentrations that recreated in months a physiological situation that would have developed in arsenic-exposed persons over years, as confirmed by the ICP-MS measurements. Thus, even though 10 ppm is equivalent to a high level of arsenic contamination, it is possible that through accumulation, these hepatic arsenic levels can be reached over time at lower exposure levels in mice. Furthermore, the incubation period of VL in humans ranges from weeks to many months (1); therefore, it is possible that at these and lower levels of hepatic arsenic exposure, resistance in the Leishmania parasite may have a chance to develop. There was no significant difference seen in antimony levels between arsenic-exposed and non–arsenic-exposed postPentostam treatment livers, indicating that competing metabolism of the two metalloids is unlikely to be a contributory factor in the decreased efficacy of Pentostam in vivo. The stability of drug resistance following passage of resistant lines through non–arsenic-exposed mice implies that the mechanisms of resistance have been acquired at a genetic rather than at an epigenetic level within the parasite. This stability, together with an increased fitness, has been demonstrated in field isolates (20). In relation to this hypothesis, this allows for dissemination of resistant strains through non–arsenic-exposed persons and into areas where there is no arsenic contamination through human migration. Perry et al.

SEE COMMENTARY

Fig. 4. Effect of drug treatment on the parasite burden of arsenic-exposed BALB/c mice. Parasite burden is expressed as total parasite numbers per liver (A) and response to treatment as a percentage of the respective untreated control group (B). Following 28 d of infection, at each arsenic exposure level group (clear bars, 0 ppm; gray bars, 10 ppm; and black bars, 100 ppm), BALB/c mice were treated daily for 5 d with either s.c. injections of drug vehicle alone (n = 4, 5, and 5 at each arsenic exposure level), 50 mg·kg−1 Pentostam s.c. (n = 3, 5, and 5), or miltefosine 12 mg·kg−1 (n = 3, 4, and 4). Values are mean ± SE. *P < 0.05 compared with 0-ppm group.

Perry et al.

to antimony was not determined, but a related study found no correlation between parasite resistance and clinical outcome to antimonial therapy (35). Detailed epidemiological work is underway to explore the relationship between arsenic exposure and antimonial treatment failure in VL in Bihar, including assessment of established clinical risk factors. This work highlights the need to consider the environmental context in which a parasite is propagating. There are parallels with development of chloroquine resistance in the malarial parasite following the public health program of mass administration of chloroquine medicated salt in the 1950s and 1960s in South America, Southeast Asia, and Africa (36). The presence of poor prescribing practice in Bihar and the coexistence of arsenic contamination of the groundwater in this region provide multiple ways for the parasite to be exposed to subtherapeutic doses of the closely related metalloids leading to resistance. We conclude that arsenic contamination may have played a significant role in Bihar because issues of inadequate treatment courses are not exclusive to India, but low antimonial efficacy is. Materials and Methods Ethics Statement. All animal experiments were approved by the Ethical Review Committee at the University of Dundee and performed under the Animals (Scientific Procedures) Act 1986 (UK Home Office Project License PPL 60/4039) in accordance with the European Communities Council Directive (86/609/EEC). Animals. BALB/c mice (commercially acquired from Harlan) were used for all in vivo experiments. Cell Lines. L. donovani (LV9 strain; World Health Organization designation MHOM/ET/67/HU3) ex vivo amastigotes derived from hamster spleens, as previously described (37), were used to infect BALB/c mice by i.p. injection of 1 × 107 amastigotes (200 μL). Amastigotes were then propagated through BALB/c mice, with infected spleens harvested at 1–2 monthly intervals and the recovered parasites used to infect further groups of mice. These parasites were propagated in this manner for 1 y before use in the creation of an in vivo resistant line. Chemicals. All chemicals used are from Sigma Aldrich unless otherwise stated: AsIII (as sodium meta arsenite) and SbV (as Pentostam, sodium stibogluconate free of m-chlorocresol; a gift from GlaxoSmithKline). In Vitro in-Macrophage Drug Sensitivity Assay. Sensitivity of the BALB/c ex vivo amastigotes in macrophage was assessed, as described previously (38). EC50 values were calculated using a two-parameter equation in GraFit (version 5.0.13; Erithacus Software):

PNAS | December 3, 2013 | vol. 110 | no. 49 | 19935

MICROBIOLOGY

Resistance in L. donovani has been studied for over 30 y using both laboratory-generated resistant strains and clinical isolates. These differing methods have led to varying reports of resistance mechanisms (16, 21). The main mechanisms identified relate to the parasites increasing their defense against oxidative stress by manipulating their thiol metabolism and changes in transport of both drug and thiols (16, 22–25). Although ex vivo Pentostamsensitive and -resistant parasites are able to infect mouse peritoneal macrophages, we have been unable to transform these parasites into insect-stage promastiogtes for biochemical studies. The in-macrophage assay system is well established as an assay for assessing parasite susceptibility but has been criticized for nonstandardization across research groups (26) and not being reflective of the clinical prognosis (27). In the case of antimonial drugs specifically, this lack of clinical correlation may be due to the dual mechanism of action of Pentostam, which exerts its direct toxic effects on the Leishmania parasite not only after reduction to SbIII but also via indirect effects on the host immune system leading to increased oxidative and nitrosative stress (28). Therefore, the results of the in vivo treatment with Pentostam provide important confirmation that the resistance developed by the arsenic-exposed Leishmania parasites is relevant within a physiological system. There is some evidence that chronic arsenic exposure leads to immunosuppression (29, 30). This could have contributed to the decreased efficacy of in vivo antimonials in our model because part of their mode of action is driven by the immune system (28). However, the current literature is sporadic and can be contradictory (31, 32), so no concrete conclusions can be drawn without further work in this area. Moreover, it cannot explain the parasite resistance observed in vitro in macrophages. Outside of Bihar, elevated rates of treatment failure for pentavalent antimonials have been reported from Nepal and Peru (33, 34). In Nepal the overall treatment failure rate with sodium stibogluconate was ∼1 in 10. This is considerably lower than failure rates in Bihar which fall below 50% (5), and one of the significant risk factors for treatment failure identified in this Nepalese study was living in a district that borders Bihar. Other important clinical risk factors were prolonged course of infection and differences in treatment administration. In Peru, a study on the use of pentavalent antimonials in cutaneous leishmaniasis reported a 24.4% failure rate. Patients who experienced treatment failure in Peru were more likely to have had a short disease duration, have more than one CL lesion, have either Leishmania peruviana or Leishmania braziliensis isolated, and have lived in an area of endemicity for less than 72 mo (34). In vitro susceptibility

y=

100% 1+

x IC50

!s :

Oral Exposure of BALB/c Mice to AsIII in Drinking Water. In designing the arsenic exposure model, the concentrations of arsenic to be administered to mice were chosen to recreate two situations reported from the Indian subcontinent. Using an established dose-equivalent equation based on body surface area (19), BALB/c mice drinking water containing AsIII at 10,000 μg·L−1 (hereafter 10 ppm, equivalent to 133 μM) is equivalent to a 60-kg man drinking 3 L·d−1 (18) of water contaminated with arsenic at 1,200 μg·L−1. This concentration has been reported in arsenic-contaminated tube wells in Bihar (8). The higher level of 100,000 μg·L−1 (hereafter 100 ppm, equivalent to 1.33 mM) is equivalent to concentrations greater than those reported from Bihar but was chosen to recreate arsenic levels previously detected on liver biopsy in patients chronically exposed to arsenic (13). BALB/c mice, aged between 6 and 10 wk, were divided into three groups that received either normal tap water (hereafter 0 ppm), arsenic at 10 ppm (10 mg·L−1), or arsenic at 100 ppm (100 mg·L−1). The water of both groups was changed one to two times per week, water intake was measured, and the animals were weighed weekly. All mice were fed standard autoclavable murine chow pellets RM1 A (Special Diets Services). At days 7, 14, 28, and 56, three mice from each exposure level were euthanized. Harvested liver and kidney were placed in CK14 tubes containing ceramic beads, and harvested spleen was placed in MK28R tubes containing metal beads (PeqlabBiotechnologie GmbH) with 500 μL double-distilled water and homogenized at 2 × 5,000 rpm for 10 s with a 20-s gap, using a PreCelleys 24 homogenizer (Bertin Technologies). Harvested bone was crushed with a pestle and mortar before mechanical homogenization, in MK28R tubes, at 2 × 5,900 rpm for 30 s with a 15-s gap. Blood cells and sera were separated by centrifugation at 15,000 × g for 90 s. Samples of murine chow were also collected for digestion and analysis. Samples (100-mg wet weight) were digested in closed microwave vessels using 500 μL nitric acid and 100 μL hydrogen peroxide followed by microwave digestion, further dilution with double-distilled water, and analysis by ICP-MS (39). Each mouse tissue sample was measured in triplicate and expressed as mg per kg wet weight. The average limit of detection was 0.2 μg As per kg wet weight. Arsenic values measured in certified reference 1. Herwaldt BL (1999) Leishmaniasis. Lancet 354(9185):1191–1199. 2. Guerin PJ, et al. (2002) Visceral leishmaniasis: Current status of control, diagnosis, and treatment, and a proposed research and development agenda. Lancet Infect Dis 2(8): 494–501. 3. Chappuis F, et al. (2007) Visceral leishmaniasis: What are the needs for diagnosis, treatment and control? Nat Rev Microbiol 5(11):873–882. 4. Singh RK, Pandey HP, Sundar S (2006) Visceral leishmaniasis (kala-azar): Challenges ahead. Indian J Med Res 123(3):331–344. 5. Olliaro PL, et al. (2005) Treatment options for visceral leishmaniasis: A systematic review of clinical studies done in India, 1980-2004. Lancet Infect Dis 5(12):763–774. 6. Matlashewski G, et al. (2011) Visceral leishmaniasis: Elimination with existing interventions. Lancet Infect Dis 11(4):322–325. 7. Perry MR, et al. (2011) Visceral leishmaniasis and arsenic: An ancient poison contributing to antimonial treatment failure in the Indian subcontinent? PLoS Negl Trop Dis 5(9):e1227. 8. Chakraborti D, et al. (2003) Arsenic groundwater contamination in Middle Ganga Plain, Bihar, India: A future danger? Environ Health Perspect 111(9):1194–1201. 9. Yan S, Jin L, Sun H (2005) 51Sb Antimony in Medicine. Metallotherapeutic Drugs and Metal-Based Diagnostic Agents—The Use of Metals in Medicine (John Wiley, Chichester, UK), pp 441–461. 10. Dey S, et al. (1994) High level arsenite resistance in Leishmania tarentolae is mediated by an active extrusion system. Mol Biochem Parasitol 67(1):49–57. 11. Bachmanov AA, Reed DR, Beauchamp GK, Tordoff MG (2002) Food intake, water intake, and drinking spout side preference of 28 mouse strains. Behav Genet 32(6):435–443. 12. Vahter M, Concha G (2001) Role of metabolism in arsenic toxicity. Pharmacol Toxicol 89(1):1–5. 13. Mazumder DNG (2005) Effect of chronic intake of arsenic-contaminated water on liver. Toxicol Appl Pharmacol 206(2):169–175. 14. Kitchin KT, Conolly R (2010) Arsenic-induced carcinogenesis—Oxidative stress as a possible mode of action and future research needs for more biologically based risk assessment. Chem Res Toxicol 23(2):327–335. 15. Ahlborn GJ, et al. (2009) Impact of life stage and duration of exposure on arsenicinduced proliferative lesions and neoplasia in C3H mice. Toxicology 262(2):106–113. 16. Croft SL, Sundar S, Fairlamb AH (2006) Drug resistance in leishmaniasis. Clin Microbiol Rev 19(1):111–126. 17. Hasker E, et al. (2010) Management of visceral leishmaniasis in rural primary health care services in Bihar, India. Trop Med Int Health 15(Suppl 2):55–62. 18. Ravenscroft P, Brammer H, Richards K (2009) Arsenic Pollution: A Global Synthesis (Wiley-Blackwell, Chichester, UK).

19936 | www.pnas.org/cgi/doi/10.1073/pnas.1311535110

material (Seronorm Trace Elements Urine; Seronorm) were 197 ± 5 μg As per L (n = 3, certified value 184 ± 17 μg As per L), and those measured in TORT-2 (Lobster Hepatopancreas; National Research Council Canada; http://archive. nrc-cnrc.gc.ca/obj/inms-ienm/doc/crm-mrc/eng/TORT-2_e.pdf) were 20 ± 0.4 mg As per kg dry weight (certified 21.6 ± 1.8 mg As per kg dry weight). In Vivo Selection of Pentostam Resistance. BALB/c mice, aged between 6 and 10 wk, were divided into three groups receiving AsIII at 0, 10, or 100 ppm in their drinking water. The mice were preexposed to arsenic for 1 mo before infection by i.p. injection of 0.2 mL of 1 × 107 L. donovani ex vivo splenic amastigotes from BALB/c mice, as described above. At 28-d postinfection, the mice were euthanized, and ex vivo amastigotes from the spleens of each group were prepared separately. These amastigotes were either used to reinfect further arsenic preexposed groups of mice or, using the in-macrophage assay described above, assessed for their sensitivity to Pentostam. See Fig. 1 for an experimental flow diagram. Samples of Leishmania-infected liver and spleen were also prepared and analyzed for arsenic content by ICP-MS, as above. Following the fifth passage, ex vivo amastigotes were used to infect groups of mice that had not been exposed to arsenic to look for stability of resistance. These parasites were passaged in BALB/c mice for 4 mo before reassessment of their sensitivity to Pentostam using the in-macrophage assay. In Vivo Assessment of Parasite Resistance. In vivo sensitivity to Pentostam and miltefosine was assessed at the end of the fifth passage. At 28 d postinfection, groups of mice from each exposure level were treated with drug vehicle only (sterile water, s.c. injection), Pentostam (50 mg·kg−1, s.c.), or miltefosine (12 mg·kg−1, orally) once daily for 5 d. Fresh drug dosing solutions were prepared daily. On day 35 postinfection, 72 h after the final treatment, all animals were euthanized, and parasite burdens were determined by blinded counting. The total number of parasites per liver was calculated by multiplying the parasites per organ cell nucleus by 2 × 105 by the weight of the liver in mg (40). The differences in response to Pentostam and miltefosine between arsenic exposure groups were compared using the unpaired Student t test. Liver and spleens from treated animals were prepared as described above and analyzed for both arsenic and antimony by ICP-MS. ACKNOWLEDGMENTS. This work was supported by the Wellcome Trust Grants 090665 (to M.R.P.) and 079838 and 083481 (to A.H.F.).

19. Reagan-Shaw S, Nihal M, Ahmad N (2008) Dose translation from animal to human studies revisited. FASEB J 22(3):659–661. 20. Vanaerschot M, et al. (2011) Antimonial resistance in Leishmania donovani is associated with increased in vivo parasite burden. PLoS ONE 6(8):e23120. 21. Singh N, Singh RT, Sundar S (2003) Novel mechanism of drug resistance in kala azar field isolates. J Infect Dis 188(4):600–607. 22. Borst P, Ouellette M (1995) New mechanisms of drug resistance in parasitic protozoa. Annu Rev Microbiol 49:427–460. 23. Ouellette M, et al. (1995) Transport of antifolates and antimonials in drug-resistant Leishmania. Drug Transport in Antimicrobial and Anticancer Chemotherapy, Infectious Disease and Therapy, eds Scully BE, Neu HC (Marcel Dekker, New York), Vol 17, pp 377–402. 24. Ashutosh SS, Sundar S, Goyal N (2007) Molecular mechanisms of antimony resistance in Leishmania. J Med Microbiol 56(Pt 2):143–153. 25. Aït-Oudhia K, Gazanion E, Vergnes B, Oury B, Sereno D (2011) Leishmania antimony resistance: What we know what we can learn from the field. Parasitol Res 109(5):1225–1232. 26. Seifert K, Escobar P, Croft SL (2010) In vitro activity of anti-leishmanial drugs against Leishmania donovani is host cell dependent. J Antimicrob Chemother 65(3):508–511. 27. Rijal S, et al. (2007) Antimonial treatment of visceral leishmaniasis: Are current in vitro susceptibility assays adequate for prognosis of in vivo therapy outcome? Microbes Infect 9(4):529–535. 28. Rais S, et al. (2000) Sodium stibogluconate (Pentostam) potentiates oxidant production in murine visceral leishmaniasis and in human blood. Antimicrob Agents Chemother 44(9):2406–2410. 29. Biswas R, et al. (2008) Analysis of T-cell proliferation and cytokine secretion in the individuals exposed to arsenic. Hum Exp Toxicol 27(5):381–386. 30. Bishayi B, Sengupta M (2003) Intracellular survival of Staphylococcus aureus due to alteration of cellular activity in arsenic and lead intoxicated mature Swiss albino mice. Toxicology 184(1):31–39. 31. Soto-Peña GA, et al. (2006) Assessment of lymphocyte subpopulations and cytokine secretion in children exposed to arsenic. FASEB J 20(6):779–781. 32. Wu MM, Chiou HY, Ho IC, Chen CJ, Lee TC (2003) Gene expression of inflammatory molecules in circulating lymphocytes from arsenic-exposed human subjects. Environ Health Perspect 111(11):1429–1438. 33. Rijal S, et al. (2010) Clinical risk factors for therapeutic failure in kala-azar patients treated with pentavalent antimonials in Nepal. Trans R Soc Trop Med Hyg 104(3):225–229. 34. Llanos-Cuentas A, et al. (2008) Clinical and parasite species risk factors for pentavalent antimonial treatment failure in cutaneous leishmaniasis in Peru. Clin Infect Dis 46(2):223–231.

Perry et al.

SEE COMMENTARY

38. Wyllie S et al. (2012) The anti-trypanosome drug fexinidazole shows potential for treating visceral leishmaniasis. Sci Transl Med 4(119):119re1. 39. Williams PN, et al. (2007) Greatly enhanced arsenic shoot assimilation in rice leads to elevated grain levels compared to wheat and barley. Environ Sci Technol 41(19):6854–6859. 40. Stauber LA (1958) Host resistance to the Khartoum strain of Leishmania donovani. Rice Inst Pam 45(1):80–96.

MICROBIOLOGY

35. Yardley V, et al. (2006) American tegumentary leishmaniasis: Is antimonial treatment outcome related to parasite drug susceptibility? J Infect Dis 194(8):1168–1175. 36. Payne D (1988) Did medicated salt hasten the spread of chloroquine resistance in Plasmodium falciparum? Parasitol Today 4(4):112–115. 37. Wyllie S, Fairlamb AH (2006) Refinement of techniques for the propagation of Leishmania donovani in hamsters. Acta Trop 97(3):364–369.

Perry et al.

PNAS | December 3, 2013 | vol. 110 | no. 49 | 19937