- Email: [email protected]
Evaluation of the anticryptococcal activity of the antibiotic polymyxin B in vitro and in vivo
International Journal of Antimicrobial Agents 41 (2013) 250–254
Contents lists available at SciVerse ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Evaluation of the anticryptococcal activity of the antibiotic polymyxin B in vitro and in vivo Bing Zhai, Xiaorong Lin ∗ Department of Biology, 3258 Texas A&M University, College Station, TX 77843-3258, USA
a r t i c l e
i n f o
Article history: Received 13 August 2012 Accepted 2 November 2012 Keywords: Cryptococcosis Cryptococcus neoformans Polymyxin B Fluconazole Capsule
a b s t r a c t Polymyxin B (PMB), a cationic lipid oligopeptide used to treat Gram-negative bacterial infections, was previously identified to possess broad-spectrum antifungal activity and to work synergistically with azole antifungals in vitro. Here we evaluated the efficacy of PMB against Cryptococcus neoformans in vitro and in vivo and explored the mechanism of the hypersensitivity of this fungus to this compound. Using comparative time-course assays, PMB was found to kill both proliferative and quiescent cryptococcal cells in vitro. Presence of the polysaccharide capsule, a characteristic feature of Cryptococcus, significantly enhances the susceptibility of this fungus to the fungicidal activity of PMB. Furthermore, PMB is able to reduce the tissue fungal burden both in intravenous and inhalation models of murine cryptococcosis at a level comparable with the commonly used antifungal fluconazole. These findings suggest that PMB could provide an additional option for treatment against systemic cryptococcosis. © 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Invasive fungal infections have emerged worldwide as a serious threat to public health owing to the increasing population of immunocompromised patients such as individuals infected by human immunodeficiency virus (HIV) and recipients of cancer chemotherapy or tissue transplantation [1,2]. For example, systemic cryptococcosis, predominantly caused by the fungal pathogen Cryptococcus neoformans, is responsible for ca. 25–30% of deaths of acquired immune deficiency syndrome (AIDS) patients (over 0.6 million) each year [3]. Cryptococcosis in heart and small bowel transplant recipients has a mortality rate of 50% [4]. Such poor outcome is partly due to the extremely limited number of clinically available antifungal agents. Currently, the clinical treatment for cryptococcosis relies on amphotericin B, which was discovered over 60 years ago [5], and azole antifungals [commonly fluconazole (FLC)] [6]. The fungistatic nature of azoles renders it necessary to apply azoles as long-term maintenance treatment, which could lead to the emergence of resistant fungal strains [7]. Therefore, it is important to investigate new compounds to enrich the repertoire of antifungal agents. Through a screen of the Johns Hopkins Clinical Compound Library (Johns Hopkins School of Public Health, Baltimore, MD), we found that the antibiotic polymyxin B (PMB) possesses antifungal activity. It can act either alone or in combination with azoles
∗ Corresponding author. Tel.: +1 979 845 7274; fax: +1 979 845 2891. E-mail address: [email protected] (X. Lin).
against various species of fungal pathogens [8]. PMB is a cationic lipid oligopeptide that is mostly used to treat infections caused by multidrug-resistant Gram-negative bacteria [9,10]. This positively charged molecule has a neutralising effect on the negatively charged bacterial outer membrane largely due to the presence of lipopolysaccharide. This electrostatic interaction is the critical initial step [11]. Therefore, PMB perturbs the integrity of the bacterial membrane, eventually resulting in cell lysis [12]. Several other cationic peptide antibiotics, such as polymyxin E and omiganan, also display antifungal activity [13,14]. It has been proposed that these compounds also disrupt the integrity of plasma and/or vacuolar membranes in eukaryotic fungi and cause cell lysis. Compared with Candida or Aspergillus spp., the fungus Cryptococcus is more susceptible to PMB [8]. Among fungi, the presence of a polysaccharide capsule is a unique feature in Cryptococcus. The capsule is an important virulence factor as it protects Cryptococcus cells from many environmental and host-relevant stresses, such as phagocytosis [15]. Production of the capsule is greatly induced under host-relevant conditions (high temperature, high CO2 concentration, and neutral or alkaline pH). Because this polysaccharide layer primarily contributes to the high negative charge of cryptococcal cells [16], we suspect that the presence of the capsule may contribute to the hypersusceptibility of Cryptococcus to PMB. Clinically achievable serum concentrations of PMB are ca. 6.25–50 mg/L [17], which are below the in vitro inhibitory concentrations against most fungal pathogens but fall within the range of those of Cryptococcus isolates [8,14]. Therefore, we decided to evaluate the in vivo efficacy of PMB in two murine models of cryptococcosis. Naturally, hosts inhale cryptococcal cells from the
0924-8579/$ – see front matter © 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. http://dx.doi.org/10.1016/j.ijantimicag.2012.11.006
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
environment. The fungal cells are either cleared in the lungs or establish latent infections [18]. When the immune system of the host is compromised (e.g. HIV infection), cryptococcal cells can activate, disseminate from the lungs and eventually cause fatal systemic cryptococcosis [18]. In the inhalation model, animals are inoculated intranasally with Cryptococcus cells, which is analogous to the natural route of infection. These animals primarily develop pulmonary infections. In the intravenous (i.v.) model, animals are infected intravenously with fungal cells to represent disseminated cryptococcosis. Although the poor penetration of PMB to the central nervous system may limit its effect on brain infections, our data on the tissue fungal burden of the lungs and kidney suggest a modest efficacy of this compound against systemic cryptococcal infections. 2. Materials and methods 2.1. Strains and media The Cryptococcus strains used in this study include the serotype A reference strain H99, the capsule mutant strain cap59 derived in the H99 background [19] and the serotype D strain XL280 [20]. Strains were maintained on yeast–peptone–dextrose (YPD) medium (Difco, Houston, TX). Time-course assays were performed using phosphate-buffered saline (PBS) or RPMI-1640 medium buffered with 4-morpholinepropanesulfonic acid (MOPS) (BioWhittakerTM ; Lonza Inc., Allendale, NJ). 2.2. Compounds and animals For in vitro studies, polymyxin B sulfate was dissolved in water at a stock concentration of 20 mg/mL and FLC was dissolved in water at a stock concentration of 2 mg/mL; both of the drugs were further diluted with PBS buffer or RPMI medium to the indicated working concentrations. For animal studies, PMB was dissolved in 0.9% saline at a stock concentration of 20 mg/mL and was diluted with 0.9% saline for injection. FLC was dissolved in 0.9% saline at 4 mg/mL. Female A/J mice (6–8 weeks old) were purchased from The Jackson Laboratory (Bar Harbor, ME). 2.3. In vitro time-course assays of effect of the capsule on drug efficacy Time-course assays were employed to evaluate the effect of the polysaccharide capsule on drug efficacy. To induce capsule production, wild-type H99 cells were cultured on RPMI agar medium at 37 ◦ C in 5% CO2 . To suppress capsule production, H99 cells were cultured on YPD agar with addition of 1 M NaCl at 30 ◦ C in ambient air. After 3 days of incubation, the cells grown under each condition were collected and washed twice in PBS buffer. The cells were then suspended in PBS buffer to obtain a cell density of 1500–2000 cells/mL. The cell suspensions obtained from these two different growth conditions were then treated with PMB (8 mg/L). A no-drug treatment was used as the positive control. At different time points (0, 1, 2, 4, 6 and 8 h post inoculation), aliquots of cell suspensions were spread onto drug-free solid YPD medium to determine the number of viable cells by colony counting. 2.4. In vitro study of the effect of the capsule on drug efficacy using an acapsular mutant To quantify the MIC100 of PMB against the wild-type H99 strain and the cap59 mutant, fungal cells of the two strains were cultured in liquid YPD medium overnight (30 ◦ C, 250 rpm). Cells were then collected by centrifugation and washed twice in PBS buffer. Cells were inoculated into RPMI medium to reach a density of 1500–2000 cells/mL. The stock PMB solution was directly diluted in
251
RPMI medium to reach drug concentrations of 0, 2, 4, 6, 8, 10, 12, 14, 16 and 20 g/mL. Wells that contained no Cryptococcus cells were included as negative controls. The MIC100 of PMB was defined as the lowest drug concentration that resulted in a 100% decrease in absorbance at an optical density of 600 nm compared with that of the control in drug-free medium. 2.5. In vivo murine models of cryptococcosis Animal models of systemic cryptococcosis were induced by two infectious routes, namely i.v. and intranasal infection. In each infection model, mice were assigned to four treatment groups: control (0.9% saline); FLC; PMB; and the drug combination. Five animals per group were used for the fungal burden assay and ten animals per group were used for the survival study. All of the drugs were administered intraperitoneally. For the i.v. infection, each mouse was challenged with 1.0 × 104 H99 cells at Day 0. Drug treatment with FLC (10 mg/kg/day), PMB (2.5 mg/kg/day) or the drug combination started on Day 1 and lasted for 5 consecutive days. Mice were given drugs once every 24 h. Mice were then sacrificed on Day 6 and their kidneys were harvested for determination of tissue fungal burden. For the intranasal infection, animals were first sedated with ketamine and xylazine and then 1.0 × 105 H99 cells suspended in 50 L of saline were slowly inoculated into the left nostril of sedated animals. In the survival study, drug treatment with FLC (16 mg/kg/day), PMB (2.5 mg/kg/day) or the drug combination was initiated at Day 2 and lasted for 5 consecutive days. In the fungal burden study, drug treatment with FLC (4 mg/kg/day), PMB (2.5 mg/kg/day) or the drug combination was initiated at Day 10 and lasted for 3 consecutive days. Mice were sacrificed 24 h after the last dose of treatment and their lungs were harvested. The kidneys or lungs were homogenised in 2 mL of cold PBS buffer using an IKA® Ultra-Turrax® T-18 Homogenizer (IKA Works, Wilmington, NC) with the same setting for each type of organ. Tissue suspensions were serially diluted (10×), plated onto YPD agar and incubated at 30 ◦ C for 2 days such that the colonies became visible in order to calculate CFUs. Animals were weighed daily and were monitored twice a day for disease progression and potential severe side effects, including weight loss, gait changes, laboured breaths or fur ruffling. Animal experiments were performed according to the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at Texas A&M University (College Station, TX) (animal protocol permit no. 2011-22). 2.6. Statistical analysis One-way analysis of variance (ANOVA) tests in the fungal burden studies were performed using GraphPad Prism 5 (GraphPad Software Inc., La Jolla, CA). A P-value of <0.05 was considered significant. 3. Results 3.1. Polymyxin B is fungicidal against both proliferative and non-proliferative cryptococcal cells in vitro Our previous study indicated that PMB possesses antifungal activity and can dramatically enhance the potency of FLC against many fungal pathogens in vitro [8]. These assays were performed with fungal cells undergoing active proliferation. To examine whether PMB is fungicidal both against proliferative and nonproliferative cryptococcal cells, a comparative time-course assay was performed of cell viability of H99 and XL280 strains in vitro treated with PMB, FLC or the drug combination. Fungal cells in RPMI
252
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
Fig. 1. Polymyxin B (PMB) is fungicidal against proliferative and quiescent Cryptococcus cells of the serotype A strain H99 and the serotype D strain XL280. (A and C) H99 cells were inoculated into RPMI medium (A) or phosphate-buffered saline (PBS) (C) and were cultured without any drug (control) or in the presence of fluconazole (FLC) (8 mg/L), polymyxin B (PMB) (16 mg/L) or a combination of these two drugs. (B and D) XL280 cells were inoculated into RPMI medium (B) or PBS (D) and were cultured without any drug (control) or in the presence of FLC (6 mg/L), PMB (10 mg/L) or a combination of these two drugs. At the indicated time points, aliquots of cell suspensions were transferred and plated onto drug-free agar medium. CFUs were measured after 2 more days of incubation.
medium or suspended in PBS buffer were cultured to represent rapidly proliferative or quiescent cells, respectively. As shown in Fig. 1, the survival curves of H99 and XL280 to the drug treatments displayed similar patterns. In RPMI medium, fungal cells with all of the drug treatments were cleared within the experimental period (Fig. 1A and B). The number of viable H99 cells started to decrease within 6 h in the presence of PMB compared with the no-drug control. Similarly, viable XL280 cells started to decrease within 4 h after being treated with PMB. The drug combination showed the strongest effect on H99 cells, indicating a synergistic interaction of the two drugs, whereas PMB displayed a more dominant role in the drug combination in clearing XL280 cells. Similarly, the fungal cells of both strains in PBS buffer were completely cleared by PMB (Fig. 1C and D). These results indicate that PMB is fungicidal and is potent against Cryptococcus cells irrespective of the cell growth status. In contrast, FLC was effective in clearing rapidly proliferative cells incubated in RPMI medium (Fig. 1A and B) but it did not show any effect on clearance of non-proliferative fungal cells in PBS buffer by itself (Fig. 1C and D). FLC also failed to enhance the potency of PMB against cryptococcal cells in PBS buffer. This lack of efficacy of FLC on quiescent fungal cells has been observed previously and is consistent with its fungistatic nature [21]. These results indicate that PMB is fungicidal and is more potent than FLC against quiescent fungal cells. 3.2. The polysaccharide capsule of Cryptococcus facilitates the fungicidal activity of polymyxin B Given that isolates of Cryptococcus are much more susceptible to PMB compared with isolates of Candida and Aspergillus spp. [8], we hypothesise that differences in cell surface composition may contribute to the difference in their susceptibility to PMB. The polysaccharide capsule is a prominent and unique structure present in Cryptococcus that is negatively charged [16]. Therefore, we suspect that presence of the capsule may contribute to the hypersensitivity of this fungus to PMB. To test this hypothesis,
we first compared the inhibitory concentration of PMB against the acapsular cap59 mutant [19] with the wild-type H99. The cap59 mutant is more resistant to PMB compared with H99, with a 25% higher MIC100 (MIC100 of cap59 = 10 mg/L; MIC100 of the wildtype H99 = 8 mg/L). This suggests that the lack of capsule confers the cryptococcal resistance against the stress caused by PMB. As the mutation in the cap59 strain could result in other physiological changes in the fungal cells that may account for the drug resistance [22], we decided to test the effect of the capsule on Cryptococcus sensitivity to PMB by examining the drug sensitivity of wild-type H99 cells with different size of capsules. Wild-type H99 cells were cultured under a capsule-inducing condition that is relevant to the host physiological condition (mammalian cell culture medium with neutral pH, 37 ◦ C, 5% CO2 ), and under a capsule-suppressing condition (YPD media with addition of a high concentration of salt). The difference in capsule size of the H99 cells cultured under these two conditions was apparent after 3 days of incubation by India ink staining (Fig. 2A and B). Cells grown in these conditions were then collected individually, washed and suspended in PBS buffer. The susceptibility of these cells to PMB (8 mg/L) was measured by a time-course cell viability assay. We chose to maintain cells in the non-proliferative state in PBS buffer during the assay to avoid potential complication of changes in capsule size during cell proliferation if RPMI medium was used. The survival rates of these fungal cells treated with PMB were examined at different time points. As shown in Fig. 2C, >50% of the fungal cells with a large capsule were cleared by PMB within 2 h, whilst it took >8 h to reach a similar clearance level of cells with no visible capsule. Taken together, these results indicate that the capsule of Cryptococcus renders this fungus more susceptible to PMB. 3.3. Polymyxin B modestly reduces the kidney fungal burden in the intravenous infection model of systemic cryptococcosis In the in vivo study, a murine model of systemic cryptococcosis where animals were infected intravenously was adopted. Pilot
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
253
Fig. 2. The capsule enhances the susceptibility of Cryptococcus to polymyxin B (PMB). (A and B) Cells were examined by India ink staining; the capsule excludes Indian ink and appears as a halo surrounding the yeast cell, with the size of the halo reflecting the size of the capsule. (A) Capsule induced; and (B) capsule suppressed. (C) Fungal cells previously grown at each condition were suspended in phosphate-buffered saline with addition of PMB at a concentration of 8 mg/L. At the indicated time points, aliquots of cell suspensions were transferred and plated onto drug-free agar medium to determine CFUs.
experiments showed that the kidney fungal burden displayed relatively low variations among individual animals in this model. Previous studies on PMB had defined a safe dose of ≤2.5 mg/kg/day [23]; therefore, we set all the PMB treatments at this dose in the animal experiments. As shown in Fig. 3, PMB modestly reduced the tissue fungal burden in the kidney, although the reduction is not statistically significant (P = 0.058). Surprisingly, the drug combination did not show any synergistic effect. An animal survival study was also performed using this model, but no significant difference was observed among the groups (data not shown). The rapid progression of the disease in this model might prevent us from observing any apparent effect of the drug treatments on animal survival as the animals succumbed to the cryptococcal infection within 7 days after inoculation even with the FLC therapy [21].
drug combination did not show any apparent advantage in prolonging animal survival compared with treatment with a single drug. We reasoned that the high dose of FLC, although by itself more effective, may mask the potential synergistic effect between FLC and PMB in vivo. The modest differences in animal survival among these groups may also prevent us from seeing the potential synergy between the drugs. Therefore, the dose of FLC was reduced to 4 mg/kg/day in the intranasal infection model and the lung fungal burden was examined instead. Lungs are the primary infectious site in this model. As shown in Fig. 4B, the 3-day treatment with PMB alone significantly reduced the lung fungal burden (P < 0.05).
3.4. Polymyxin B slightly prolongs the survival period of animals, reduces the fungal burden in the lungs, and works synergistically with fluconazole in an intranasal infection model We further tested the effect of these treatments in a murine inhalation model of cryptococcosis where the infection route reflects the natural route of cryptococcal infection. The dose of FLC was increased from 10 mg/kg/day to 16 mg/kg/day to achieve a better protective result. As shown in Fig. 4A, treatment with PMB or FLC modestly prolonged the survival of infected mice. However, the
Fig. 3. Polymyxin B (PMB) modestly reduces the fungal burden of the kidney in a murine model of disseminated cryptococcosis. Mice were challenged intravenously with H99 cells and were treated with saline, PMB (2.5 mg/kg/day), fluconazole (FLC) (10 mg/kg/day) or the drug combination for 5 consecutive days and were then sacrificed. The fungal burden of the kidneys was determined by calculating CFUs. PMB alone was able to modestly reduce the fungal burden (P = 0.058). However, there was no significant difference among the three drug treatment groups.
Fig. 4. Polymyxin B (PMB) modestly prolongs animal survival (A) and is effective in reducing lung fungal burden (B) in a murine model of pulmonary cryptococcosis. Mice were challenged intranasally with H99 cells. (A) Treatment started at Day 2 and lasted for 5 consecutive days. Survival of animals in each group was examined. (B) The lungs of infected mice were dissected and homogenised after 3 days of drug treatment. PMB alone was able to reduce the fungal burden in the lungs (*P < 0.05) and the drug combination displayed the best protection (**P < 0.001). FLC, fluconazole.
254
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
The FLC group did not show any protective effect, as expected with such a low dose used. However, the drug combination displayed a strong inhibitory effect on fungal proliferation (P < 0.001), indicating synergy between FLC and PMB. Taken together, these data suggest that PMB is effective in vivo against cryptococcal infections. 4. Discussion In this study, the efficacy of PMB against Cryptococcus was examined both in vitro and in vivo. The fungicidal nature of PMB, its relative high potency against Cryptococcus, and its synergy with FLC suggest its potential in augmenting the effects of current FLC therapies. The benefits of research on existing clinical compounds that exert antifungal activity are attracting more attention in recent years [21]. Existing information regarding the pharmacokinetics and safety of these drugs could mean a much shorter period of investigation before their clinical use as antifungals. The antifungal property of the polymyxins and other cationic peptide antibiotics has recently been proposed by several groups [8,13,14]. These studies suggested that the mechanisms of the antifungal property of these drugs are analogous to their antibacterial effect, that is disruption of the integrity of membranes (cytoplasmic and/or vacuolar). Owing to the huge differences between the composition and structure of eukaryotic and bacterial membranes (e.g. the presence or absence of sterols in the membrane, the prevalence of negatively charged lipids, etc.), fungi are more resistant to PMB and other cationic antibiotics compared with Gram-negative bacteria. However, Cryptococcus is more susceptible to PMB than other tested fungal pathogens [8]. The evidence presented here suggests that the polysaccharide capsule of Cryptococcus may contribute to this hypersensitivity. This unique negatively charged layer may enrich the effective concentration of cationic molecules in the microenvironment of the cell surface and render the Cryptococcus cells more susceptible to PMB. Compared with the in vitro potency of PMB, the in vivo efficacy of this drug against Cryptococcus is modest in our animal models. One possible reason is the difference between the in vitro and in vivo conditions. Non-specific binding of PMB to proteins in serum or host tissues may decrease the effective concentration of the drug against Cryptococcus in vivo. This is supported by previous studies showing that polymyxins lost ca. 50% of their activity in the presence of serum [23,24]. In addition, the animal models used in this study may diminish the observable effect of PMB in vivo. The A/J mouse is highly susceptible to cryptococcal infections, and the H99 strain is one of the most virulent clinical isolates [25]. The rapid progression of cryptococcosis in the models used here may underestimate the drug effect. A better effect of PMB might be expected on a more robust animal model, such as rat or rabbit, or when the animals are infected by a less-virulent cryptococcal strain. However, the fact that PMB displayed a modest efficacy in vivo against Cryptococcus, especially in the clearance of tissue fungal burden, in the models used in this study suggests the potential of PMB as an option for the treatment of cryptococcosis. Furthermore, PMB could be used in topical and eye treatments to combat local fungal infections, where much higher doses of PMB are often used to treat bacterial infections. Funding: This work was supported by the American Heart Association (grant 0BGIA3740040 to XL) and the Norman Hackerman Advanced Research Program (grant 01957 to XL). Competing interests: None declared. Ethical approval: This study was performed according to the guidelines of the National Institutes of Health (NIH) and the Institutional Animal Care and Use Committee (IACUC) at Texas A&M
University (College Station, TX). The animal models and procedures used have been approved by the IACUC at Texas A&M University (protocol no. 2011-22). References [1] Mitchell TG, Perfect JR. Cryptococcosis in the era of AIDS–100 years after the discovery of Cryptococcus neoformans. Clin Microbiol Rev 1995;8: 515–48. [2] Anaissie EJ. Diagnosis and therapy of fungal infection in patients with leukemia–new drugs and immunotherapy. Best Pract Res Clin Haematol 2008;21:683–90. [3] Park BJ, Wannemuehler KA, Marston BJ, Govender N, Pappas PG, Chiller TM. Estimation of the current global burden of cryptococcal meningitis among persons living with HIV/AIDS. AIDS 2009;23:525–30. [4] Wu G, Vilchez RA, Eidelman B, Fung J, Kormos R, Kusne S. Cryptococcal meningitis: an analysis among 5,521 consecutive organ transplant recipients. Transpl Infect Dis 2002;4:183–8. [5] Laniado-Laborín R, Cabrales-Vargas MN. Amphotericin B: side effects and toxicity. Rev Iberoam Micol 2009;26:223–7. [6] Perfect JR, Dismukes WE, Dromer F, Goldman DL, Graybill JR, Hamill RJ, et al. Clinical practice guidelines for the management of cryptococcal disease: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis 2010;50:291–322. [7] Bicanic T, Harrison T, Niepieklo A, Dyakopu N, Meintjes G. Symptomatic relapse of HIV-associated cryptococcal meningitis after initial fluconazole monotherapy: the role of fluconazole resistance and immune reconstitution. Clin Infect Dis 2006;43:1069–73. [8] Zhai B, Zhou H, Yang L, Zhang J, Jung K, Giam CZ, et al. Polymyxin B, in combination with fluconazole, exerts a potent fungicidal effect. J Antimicrob Chemother 2010;65:931–8. [9] Zavascki AP, Goldani LZ, Li J, Nation RL. Polymyxin B for the treatment of multidrug-resistant pathogens: a critical review. J Antimicrob Chemother 2007;60:1206–15. [10] Evans ME, Feola DJ, Rapp RP. Polymyxin B sulfate and colistin: old antibiotics for emerging multiresistant Gram-negative bacteria. Ann Pharmacother 1999;33:960–7. [11] Clausell A, Garcia-Subirats M, Pujol M, Busquets MA, Rabanal F, Cajal Y. Gram-negative outer and inner membrane models: insertion of cyclic cationic lipopeptides. J Phys Chem B 2007;111:551–63. [12] Hancock RE. Peptide antibiotics. Lancet 1997;349:418–22. [13] Fritsche TR, Rhomberg PR, Sader HS, Jones RN. Antimicrobial activity of omiganan pentahydrochloride against contemporary fungal pathogens responsible for catheter-associated infections. Antimicrob Agents Chemother 2008;52:1187–9. [14] Ben-Ami R, Lewis RE, Tarrand J, Leventakos K, Kontoyiannis DP. Antifungal activity of colistin against Mucorales species in vitro and in a murine model of Rhizopus oryzae pulmonary infection. Antimicrob Agents Chemother 2010;54:484–90. [15] Kozel TR, Gotschlich EC. The capsule of Cryptococcus neoformans passively inhibits phagocytosis of the yeast by macrophages. J Immunol 1982;129:1675–80. [16] Nosanchuk JD, Casadevall A. Cellular charge of Cryptococcus neoformans: contributions from the capsular polysaccharide, melanin, and monoclonal antibody binding. Infect Immun 1997;65:1836–41. [17] Sarria JC, Angulo-Pernett F, Kimbrough RC, McVay CS, Vidal AM. Use of intravenous polymyxin B during continuous venovenous hemodialysis. Eur J Clin Microbiol Infect Dis 2004;23:340–1. [18] Lin X, Heitman J. The biology of the Cryptococcus neoformans species complex. Annu Rev Microbiol 2006;60:69–105. [19] Chang YC, Kwon-Chung KJ. Complementation of a capsule-deficient mutation of Cryptococcus neoformans restores its virulence. Mol Cell Biol 1994;14:4912–9. [20] Lin X, Hull CM, Heitman J. Sexual reproduction between partners of the same mating type in Cryptococcus neoformans. Nature 2005;434:1017–21. [21] Zhai B, Wu C, Wang L, Sachs MS, Lin X. The antidepressant sertraline provides a promising therapeutic option for neurotropic cryptococcal infections. Antimicrob Agents Chemother 2012;56:3758–66. [22] García-Rivera J, Chang YC, Kwon-Chung KJ, Casadevall A. Cryptococcus neoformans CAP59 (or Cap59p) is involved in the extracellular trafficking of capsular glucuronoxylomannan. Eukaryot Cell 2004;3:385–92. [23] Hermsen ED, Sullivan CJ, Rotschafer JC. Polymyxins: pharmacology, pharmacokinetics, pharmacodynamics, and clinical applications. Infect Dis Clin North Am 2003;17:545–62. [24] Li J, Milne RW, Nation RL, Turnidge JD, Smeaton TC, Coulthard K. Use of high-performance liquid chromatography to study the pharmacokinetics of colistin sulfate in rats following intravenous administration. Antimicrob Agents Chemother 2003;47:1766–70. [25] Zaragoza O, Alvarez M, Telzak A, Rivera J, Casadevall A. The relative susceptibility of mouse strains to pulmonary Cryptococcus neoformans infection is associated with pleiotropic differences in the immune response. Infect Immun 2007;75:2729–39.
Contents lists available at SciVerse ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Evaluation of the anticryptococcal activity of the antibiotic polymyxin B in vitro and in vivo Bing Zhai, Xiaorong Lin ∗ Department of Biology, 3258 Texas A&M University, College Station, TX 77843-3258, USA
a r t i c l e
i n f o
Article history: Received 13 August 2012 Accepted 2 November 2012 Keywords: Cryptococcosis Cryptococcus neoformans Polymyxin B Fluconazole Capsule
a b s t r a c t Polymyxin B (PMB), a cationic lipid oligopeptide used to treat Gram-negative bacterial infections, was previously identified to possess broad-spectrum antifungal activity and to work synergistically with azole antifungals in vitro. Here we evaluated the efficacy of PMB against Cryptococcus neoformans in vitro and in vivo and explored the mechanism of the hypersensitivity of this fungus to this compound. Using comparative time-course assays, PMB was found to kill both proliferative and quiescent cryptococcal cells in vitro. Presence of the polysaccharide capsule, a characteristic feature of Cryptococcus, significantly enhances the susceptibility of this fungus to the fungicidal activity of PMB. Furthermore, PMB is able to reduce the tissue fungal burden both in intravenous and inhalation models of murine cryptococcosis at a level comparable with the commonly used antifungal fluconazole. These findings suggest that PMB could provide an additional option for treatment against systemic cryptococcosis. © 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Invasive fungal infections have emerged worldwide as a serious threat to public health owing to the increasing population of immunocompromised patients such as individuals infected by human immunodeficiency virus (HIV) and recipients of cancer chemotherapy or tissue transplantation [1,2]. For example, systemic cryptococcosis, predominantly caused by the fungal pathogen Cryptococcus neoformans, is responsible for ca. 25–30% of deaths of acquired immune deficiency syndrome (AIDS) patients (over 0.6 million) each year [3]. Cryptococcosis in heart and small bowel transplant recipients has a mortality rate of 50% [4]. Such poor outcome is partly due to the extremely limited number of clinically available antifungal agents. Currently, the clinical treatment for cryptococcosis relies on amphotericin B, which was discovered over 60 years ago [5], and azole antifungals [commonly fluconazole (FLC)] [6]. The fungistatic nature of azoles renders it necessary to apply azoles as long-term maintenance treatment, which could lead to the emergence of resistant fungal strains [7]. Therefore, it is important to investigate new compounds to enrich the repertoire of antifungal agents. Through a screen of the Johns Hopkins Clinical Compound Library (Johns Hopkins School of Public Health, Baltimore, MD), we found that the antibiotic polymyxin B (PMB) possesses antifungal activity. It can act either alone or in combination with azoles
∗ Corresponding author. Tel.: +1 979 845 7274; fax: +1 979 845 2891. E-mail address: [email protected] (X. Lin).
against various species of fungal pathogens [8]. PMB is a cationic lipid oligopeptide that is mostly used to treat infections caused by multidrug-resistant Gram-negative bacteria [9,10]. This positively charged molecule has a neutralising effect on the negatively charged bacterial outer membrane largely due to the presence of lipopolysaccharide. This electrostatic interaction is the critical initial step [11]. Therefore, PMB perturbs the integrity of the bacterial membrane, eventually resulting in cell lysis [12]. Several other cationic peptide antibiotics, such as polymyxin E and omiganan, also display antifungal activity [13,14]. It has been proposed that these compounds also disrupt the integrity of plasma and/or vacuolar membranes in eukaryotic fungi and cause cell lysis. Compared with Candida or Aspergillus spp., the fungus Cryptococcus is more susceptible to PMB [8]. Among fungi, the presence of a polysaccharide capsule is a unique feature in Cryptococcus. The capsule is an important virulence factor as it protects Cryptococcus cells from many environmental and host-relevant stresses, such as phagocytosis [15]. Production of the capsule is greatly induced under host-relevant conditions (high temperature, high CO2 concentration, and neutral or alkaline pH). Because this polysaccharide layer primarily contributes to the high negative charge of cryptococcal cells [16], we suspect that the presence of the capsule may contribute to the hypersusceptibility of Cryptococcus to PMB. Clinically achievable serum concentrations of PMB are ca. 6.25–50 mg/L [17], which are below the in vitro inhibitory concentrations against most fungal pathogens but fall within the range of those of Cryptococcus isolates [8,14]. Therefore, we decided to evaluate the in vivo efficacy of PMB in two murine models of cryptococcosis. Naturally, hosts inhale cryptococcal cells from the
0924-8579/$ – see front matter © 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. http://dx.doi.org/10.1016/j.ijantimicag.2012.11.006
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
environment. The fungal cells are either cleared in the lungs or establish latent infections [18]. When the immune system of the host is compromised (e.g. HIV infection), cryptococcal cells can activate, disseminate from the lungs and eventually cause fatal systemic cryptococcosis [18]. In the inhalation model, animals are inoculated intranasally with Cryptococcus cells, which is analogous to the natural route of infection. These animals primarily develop pulmonary infections. In the intravenous (i.v.) model, animals are infected intravenously with fungal cells to represent disseminated cryptococcosis. Although the poor penetration of PMB to the central nervous system may limit its effect on brain infections, our data on the tissue fungal burden of the lungs and kidney suggest a modest efficacy of this compound against systemic cryptococcal infections. 2. Materials and methods 2.1. Strains and media The Cryptococcus strains used in this study include the serotype A reference strain H99, the capsule mutant strain cap59 derived in the H99 background [19] and the serotype D strain XL280 [20]. Strains were maintained on yeast–peptone–dextrose (YPD) medium (Difco, Houston, TX). Time-course assays were performed using phosphate-buffered saline (PBS) or RPMI-1640 medium buffered with 4-morpholinepropanesulfonic acid (MOPS) (BioWhittakerTM ; Lonza Inc., Allendale, NJ). 2.2. Compounds and animals For in vitro studies, polymyxin B sulfate was dissolved in water at a stock concentration of 20 mg/mL and FLC was dissolved in water at a stock concentration of 2 mg/mL; both of the drugs were further diluted with PBS buffer or RPMI medium to the indicated working concentrations. For animal studies, PMB was dissolved in 0.9% saline at a stock concentration of 20 mg/mL and was diluted with 0.9% saline for injection. FLC was dissolved in 0.9% saline at 4 mg/mL. Female A/J mice (6–8 weeks old) were purchased from The Jackson Laboratory (Bar Harbor, ME). 2.3. In vitro time-course assays of effect of the capsule on drug efficacy Time-course assays were employed to evaluate the effect of the polysaccharide capsule on drug efficacy. To induce capsule production, wild-type H99 cells were cultured on RPMI agar medium at 37 ◦ C in 5% CO2 . To suppress capsule production, H99 cells were cultured on YPD agar with addition of 1 M NaCl at 30 ◦ C in ambient air. After 3 days of incubation, the cells grown under each condition were collected and washed twice in PBS buffer. The cells were then suspended in PBS buffer to obtain a cell density of 1500–2000 cells/mL. The cell suspensions obtained from these two different growth conditions were then treated with PMB (8 mg/L). A no-drug treatment was used as the positive control. At different time points (0, 1, 2, 4, 6 and 8 h post inoculation), aliquots of cell suspensions were spread onto drug-free solid YPD medium to determine the number of viable cells by colony counting. 2.4. In vitro study of the effect of the capsule on drug efficacy using an acapsular mutant To quantify the MIC100 of PMB against the wild-type H99 strain and the cap59 mutant, fungal cells of the two strains were cultured in liquid YPD medium overnight (30 ◦ C, 250 rpm). Cells were then collected by centrifugation and washed twice in PBS buffer. Cells were inoculated into RPMI medium to reach a density of 1500–2000 cells/mL. The stock PMB solution was directly diluted in
251
RPMI medium to reach drug concentrations of 0, 2, 4, 6, 8, 10, 12, 14, 16 and 20 g/mL. Wells that contained no Cryptococcus cells were included as negative controls. The MIC100 of PMB was defined as the lowest drug concentration that resulted in a 100% decrease in absorbance at an optical density of 600 nm compared with that of the control in drug-free medium. 2.5. In vivo murine models of cryptococcosis Animal models of systemic cryptococcosis were induced by two infectious routes, namely i.v. and intranasal infection. In each infection model, mice were assigned to four treatment groups: control (0.9% saline); FLC; PMB; and the drug combination. Five animals per group were used for the fungal burden assay and ten animals per group were used for the survival study. All of the drugs were administered intraperitoneally. For the i.v. infection, each mouse was challenged with 1.0 × 104 H99 cells at Day 0. Drug treatment with FLC (10 mg/kg/day), PMB (2.5 mg/kg/day) or the drug combination started on Day 1 and lasted for 5 consecutive days. Mice were given drugs once every 24 h. Mice were then sacrificed on Day 6 and their kidneys were harvested for determination of tissue fungal burden. For the intranasal infection, animals were first sedated with ketamine and xylazine and then 1.0 × 105 H99 cells suspended in 50 L of saline were slowly inoculated into the left nostril of sedated animals. In the survival study, drug treatment with FLC (16 mg/kg/day), PMB (2.5 mg/kg/day) or the drug combination was initiated at Day 2 and lasted for 5 consecutive days. In the fungal burden study, drug treatment with FLC (4 mg/kg/day), PMB (2.5 mg/kg/day) or the drug combination was initiated at Day 10 and lasted for 3 consecutive days. Mice were sacrificed 24 h after the last dose of treatment and their lungs were harvested. The kidneys or lungs were homogenised in 2 mL of cold PBS buffer using an IKA® Ultra-Turrax® T-18 Homogenizer (IKA Works, Wilmington, NC) with the same setting for each type of organ. Tissue suspensions were serially diluted (10×), plated onto YPD agar and incubated at 30 ◦ C for 2 days such that the colonies became visible in order to calculate CFUs. Animals were weighed daily and were monitored twice a day for disease progression and potential severe side effects, including weight loss, gait changes, laboured breaths or fur ruffling. Animal experiments were performed according to the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at Texas A&M University (College Station, TX) (animal protocol permit no. 2011-22). 2.6. Statistical analysis One-way analysis of variance (ANOVA) tests in the fungal burden studies were performed using GraphPad Prism 5 (GraphPad Software Inc., La Jolla, CA). A P-value of <0.05 was considered significant. 3. Results 3.1. Polymyxin B is fungicidal against both proliferative and non-proliferative cryptococcal cells in vitro Our previous study indicated that PMB possesses antifungal activity and can dramatically enhance the potency of FLC against many fungal pathogens in vitro [8]. These assays were performed with fungal cells undergoing active proliferation. To examine whether PMB is fungicidal both against proliferative and nonproliferative cryptococcal cells, a comparative time-course assay was performed of cell viability of H99 and XL280 strains in vitro treated with PMB, FLC or the drug combination. Fungal cells in RPMI
252
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
Fig. 1. Polymyxin B (PMB) is fungicidal against proliferative and quiescent Cryptococcus cells of the serotype A strain H99 and the serotype D strain XL280. (A and C) H99 cells were inoculated into RPMI medium (A) or phosphate-buffered saline (PBS) (C) and were cultured without any drug (control) or in the presence of fluconazole (FLC) (8 mg/L), polymyxin B (PMB) (16 mg/L) or a combination of these two drugs. (B and D) XL280 cells were inoculated into RPMI medium (B) or PBS (D) and were cultured without any drug (control) or in the presence of FLC (6 mg/L), PMB (10 mg/L) or a combination of these two drugs. At the indicated time points, aliquots of cell suspensions were transferred and plated onto drug-free agar medium. CFUs were measured after 2 more days of incubation.
medium or suspended in PBS buffer were cultured to represent rapidly proliferative or quiescent cells, respectively. As shown in Fig. 1, the survival curves of H99 and XL280 to the drug treatments displayed similar patterns. In RPMI medium, fungal cells with all of the drug treatments were cleared within the experimental period (Fig. 1A and B). The number of viable H99 cells started to decrease within 6 h in the presence of PMB compared with the no-drug control. Similarly, viable XL280 cells started to decrease within 4 h after being treated with PMB. The drug combination showed the strongest effect on H99 cells, indicating a synergistic interaction of the two drugs, whereas PMB displayed a more dominant role in the drug combination in clearing XL280 cells. Similarly, the fungal cells of both strains in PBS buffer were completely cleared by PMB (Fig. 1C and D). These results indicate that PMB is fungicidal and is potent against Cryptococcus cells irrespective of the cell growth status. In contrast, FLC was effective in clearing rapidly proliferative cells incubated in RPMI medium (Fig. 1A and B) but it did not show any effect on clearance of non-proliferative fungal cells in PBS buffer by itself (Fig. 1C and D). FLC also failed to enhance the potency of PMB against cryptococcal cells in PBS buffer. This lack of efficacy of FLC on quiescent fungal cells has been observed previously and is consistent with its fungistatic nature [21]. These results indicate that PMB is fungicidal and is more potent than FLC against quiescent fungal cells. 3.2. The polysaccharide capsule of Cryptococcus facilitates the fungicidal activity of polymyxin B Given that isolates of Cryptococcus are much more susceptible to PMB compared with isolates of Candida and Aspergillus spp. [8], we hypothesise that differences in cell surface composition may contribute to the difference in their susceptibility to PMB. The polysaccharide capsule is a prominent and unique structure present in Cryptococcus that is negatively charged [16]. Therefore, we suspect that presence of the capsule may contribute to the hypersensitivity of this fungus to PMB. To test this hypothesis,
we first compared the inhibitory concentration of PMB against the acapsular cap59 mutant [19] with the wild-type H99. The cap59 mutant is more resistant to PMB compared with H99, with a 25% higher MIC100 (MIC100 of cap59 = 10 mg/L; MIC100 of the wildtype H99 = 8 mg/L). This suggests that the lack of capsule confers the cryptococcal resistance against the stress caused by PMB. As the mutation in the cap59 strain could result in other physiological changes in the fungal cells that may account for the drug resistance [22], we decided to test the effect of the capsule on Cryptococcus sensitivity to PMB by examining the drug sensitivity of wild-type H99 cells with different size of capsules. Wild-type H99 cells were cultured under a capsule-inducing condition that is relevant to the host physiological condition (mammalian cell culture medium with neutral pH, 37 ◦ C, 5% CO2 ), and under a capsule-suppressing condition (YPD media with addition of a high concentration of salt). The difference in capsule size of the H99 cells cultured under these two conditions was apparent after 3 days of incubation by India ink staining (Fig. 2A and B). Cells grown in these conditions were then collected individually, washed and suspended in PBS buffer. The susceptibility of these cells to PMB (8 mg/L) was measured by a time-course cell viability assay. We chose to maintain cells in the non-proliferative state in PBS buffer during the assay to avoid potential complication of changes in capsule size during cell proliferation if RPMI medium was used. The survival rates of these fungal cells treated with PMB were examined at different time points. As shown in Fig. 2C, >50% of the fungal cells with a large capsule were cleared by PMB within 2 h, whilst it took >8 h to reach a similar clearance level of cells with no visible capsule. Taken together, these results indicate that the capsule of Cryptococcus renders this fungus more susceptible to PMB. 3.3. Polymyxin B modestly reduces the kidney fungal burden in the intravenous infection model of systemic cryptococcosis In the in vivo study, a murine model of systemic cryptococcosis where animals were infected intravenously was adopted. Pilot
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
253
Fig. 2. The capsule enhances the susceptibility of Cryptococcus to polymyxin B (PMB). (A and B) Cells were examined by India ink staining; the capsule excludes Indian ink and appears as a halo surrounding the yeast cell, with the size of the halo reflecting the size of the capsule. (A) Capsule induced; and (B) capsule suppressed. (C) Fungal cells previously grown at each condition were suspended in phosphate-buffered saline with addition of PMB at a concentration of 8 mg/L. At the indicated time points, aliquots of cell suspensions were transferred and plated onto drug-free agar medium to determine CFUs.
experiments showed that the kidney fungal burden displayed relatively low variations among individual animals in this model. Previous studies on PMB had defined a safe dose of ≤2.5 mg/kg/day [23]; therefore, we set all the PMB treatments at this dose in the animal experiments. As shown in Fig. 3, PMB modestly reduced the tissue fungal burden in the kidney, although the reduction is not statistically significant (P = 0.058). Surprisingly, the drug combination did not show any synergistic effect. An animal survival study was also performed using this model, but no significant difference was observed among the groups (data not shown). The rapid progression of the disease in this model might prevent us from observing any apparent effect of the drug treatments on animal survival as the animals succumbed to the cryptococcal infection within 7 days after inoculation even with the FLC therapy [21].
drug combination did not show any apparent advantage in prolonging animal survival compared with treatment with a single drug. We reasoned that the high dose of FLC, although by itself more effective, may mask the potential synergistic effect between FLC and PMB in vivo. The modest differences in animal survival among these groups may also prevent us from seeing the potential synergy between the drugs. Therefore, the dose of FLC was reduced to 4 mg/kg/day in the intranasal infection model and the lung fungal burden was examined instead. Lungs are the primary infectious site in this model. As shown in Fig. 4B, the 3-day treatment with PMB alone significantly reduced the lung fungal burden (P < 0.05).
3.4. Polymyxin B slightly prolongs the survival period of animals, reduces the fungal burden in the lungs, and works synergistically with fluconazole in an intranasal infection model We further tested the effect of these treatments in a murine inhalation model of cryptococcosis where the infection route reflects the natural route of cryptococcal infection. The dose of FLC was increased from 10 mg/kg/day to 16 mg/kg/day to achieve a better protective result. As shown in Fig. 4A, treatment with PMB or FLC modestly prolonged the survival of infected mice. However, the
Fig. 3. Polymyxin B (PMB) modestly reduces the fungal burden of the kidney in a murine model of disseminated cryptococcosis. Mice were challenged intravenously with H99 cells and were treated with saline, PMB (2.5 mg/kg/day), fluconazole (FLC) (10 mg/kg/day) or the drug combination for 5 consecutive days and were then sacrificed. The fungal burden of the kidneys was determined by calculating CFUs. PMB alone was able to modestly reduce the fungal burden (P = 0.058). However, there was no significant difference among the three drug treatment groups.
Fig. 4. Polymyxin B (PMB) modestly prolongs animal survival (A) and is effective in reducing lung fungal burden (B) in a murine model of pulmonary cryptococcosis. Mice were challenged intranasally with H99 cells. (A) Treatment started at Day 2 and lasted for 5 consecutive days. Survival of animals in each group was examined. (B) The lungs of infected mice were dissected and homogenised after 3 days of drug treatment. PMB alone was able to reduce the fungal burden in the lungs (*P < 0.05) and the drug combination displayed the best protection (**P < 0.001). FLC, fluconazole.
254
B. Zhai, X. Lin / International Journal of Antimicrobial Agents 41 (2013) 250–254
The FLC group did not show any protective effect, as expected with such a low dose used. However, the drug combination displayed a strong inhibitory effect on fungal proliferation (P < 0.001), indicating synergy between FLC and PMB. Taken together, these data suggest that PMB is effective in vivo against cryptococcal infections. 4. Discussion In this study, the efficacy of PMB against Cryptococcus was examined both in vitro and in vivo. The fungicidal nature of PMB, its relative high potency against Cryptococcus, and its synergy with FLC suggest its potential in augmenting the effects of current FLC therapies. The benefits of research on existing clinical compounds that exert antifungal activity are attracting more attention in recent years [21]. Existing information regarding the pharmacokinetics and safety of these drugs could mean a much shorter period of investigation before their clinical use as antifungals. The antifungal property of the polymyxins and other cationic peptide antibiotics has recently been proposed by several groups [8,13,14]. These studies suggested that the mechanisms of the antifungal property of these drugs are analogous to their antibacterial effect, that is disruption of the integrity of membranes (cytoplasmic and/or vacuolar). Owing to the huge differences between the composition and structure of eukaryotic and bacterial membranes (e.g. the presence or absence of sterols in the membrane, the prevalence of negatively charged lipids, etc.), fungi are more resistant to PMB and other cationic antibiotics compared with Gram-negative bacteria. However, Cryptococcus is more susceptible to PMB than other tested fungal pathogens [8]. The evidence presented here suggests that the polysaccharide capsule of Cryptococcus may contribute to this hypersensitivity. This unique negatively charged layer may enrich the effective concentration of cationic molecules in the microenvironment of the cell surface and render the Cryptococcus cells more susceptible to PMB. Compared with the in vitro potency of PMB, the in vivo efficacy of this drug against Cryptococcus is modest in our animal models. One possible reason is the difference between the in vitro and in vivo conditions. Non-specific binding of PMB to proteins in serum or host tissues may decrease the effective concentration of the drug against Cryptococcus in vivo. This is supported by previous studies showing that polymyxins lost ca. 50% of their activity in the presence of serum [23,24]. In addition, the animal models used in this study may diminish the observable effect of PMB in vivo. The A/J mouse is highly susceptible to cryptococcal infections, and the H99 strain is one of the most virulent clinical isolates [25]. The rapid progression of cryptococcosis in the models used here may underestimate the drug effect. A better effect of PMB might be expected on a more robust animal model, such as rat or rabbit, or when the animals are infected by a less-virulent cryptococcal strain. However, the fact that PMB displayed a modest efficacy in vivo against Cryptococcus, especially in the clearance of tissue fungal burden, in the models used in this study suggests the potential of PMB as an option for the treatment of cryptococcosis. Furthermore, PMB could be used in topical and eye treatments to combat local fungal infections, where much higher doses of PMB are often used to treat bacterial infections. Funding: This work was supported by the American Heart Association (grant 0BGIA3740040 to XL) and the Norman Hackerman Advanced Research Program (grant 01957 to XL). Competing interests: None declared. Ethical approval: This study was performed according to the guidelines of the National Institutes of Health (NIH) and the Institutional Animal Care and Use Committee (IACUC) at Texas A&M
University (College Station, TX). The animal models and procedures used have been approved by the IACUC at Texas A&M University (protocol no. 2011-22). References [1] Mitchell TG, Perfect JR. Cryptococcosis in the era of AIDS–100 years after the discovery of Cryptococcus neoformans. Clin Microbiol Rev 1995;8: 515–48. [2] Anaissie EJ. Diagnosis and therapy of fungal infection in patients with leukemia–new drugs and immunotherapy. Best Pract Res Clin Haematol 2008;21:683–90. [3] Park BJ, Wannemuehler KA, Marston BJ, Govender N, Pappas PG, Chiller TM. Estimation of the current global burden of cryptococcal meningitis among persons living with HIV/AIDS. AIDS 2009;23:525–30. [4] Wu G, Vilchez RA, Eidelman B, Fung J, Kormos R, Kusne S. Cryptococcal meningitis: an analysis among 5,521 consecutive organ transplant recipients. Transpl Infect Dis 2002;4:183–8. [5] Laniado-Laborín R, Cabrales-Vargas MN. Amphotericin B: side effects and toxicity. Rev Iberoam Micol 2009;26:223–7. [6] Perfect JR, Dismukes WE, Dromer F, Goldman DL, Graybill JR, Hamill RJ, et al. Clinical practice guidelines for the management of cryptococcal disease: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis 2010;50:291–322. [7] Bicanic T, Harrison T, Niepieklo A, Dyakopu N, Meintjes G. Symptomatic relapse of HIV-associated cryptococcal meningitis after initial fluconazole monotherapy: the role of fluconazole resistance and immune reconstitution. Clin Infect Dis 2006;43:1069–73. [8] Zhai B, Zhou H, Yang L, Zhang J, Jung K, Giam CZ, et al. Polymyxin B, in combination with fluconazole, exerts a potent fungicidal effect. J Antimicrob Chemother 2010;65:931–8. [9] Zavascki AP, Goldani LZ, Li J, Nation RL. Polymyxin B for the treatment of multidrug-resistant pathogens: a critical review. J Antimicrob Chemother 2007;60:1206–15. [10] Evans ME, Feola DJ, Rapp RP. Polymyxin B sulfate and colistin: old antibiotics for emerging multiresistant Gram-negative bacteria. Ann Pharmacother 1999;33:960–7. [11] Clausell A, Garcia-Subirats M, Pujol M, Busquets MA, Rabanal F, Cajal Y. Gram-negative outer and inner membrane models: insertion of cyclic cationic lipopeptides. J Phys Chem B 2007;111:551–63. [12] Hancock RE. Peptide antibiotics. Lancet 1997;349:418–22. [13] Fritsche TR, Rhomberg PR, Sader HS, Jones RN. Antimicrobial activity of omiganan pentahydrochloride against contemporary fungal pathogens responsible for catheter-associated infections. Antimicrob Agents Chemother 2008;52:1187–9. [14] Ben-Ami R, Lewis RE, Tarrand J, Leventakos K, Kontoyiannis DP. Antifungal activity of colistin against Mucorales species in vitro and in a murine model of Rhizopus oryzae pulmonary infection. Antimicrob Agents Chemother 2010;54:484–90. [15] Kozel TR, Gotschlich EC. The capsule of Cryptococcus neoformans passively inhibits phagocytosis of the yeast by macrophages. J Immunol 1982;129:1675–80. [16] Nosanchuk JD, Casadevall A. Cellular charge of Cryptococcus neoformans: contributions from the capsular polysaccharide, melanin, and monoclonal antibody binding. Infect Immun 1997;65:1836–41. [17] Sarria JC, Angulo-Pernett F, Kimbrough RC, McVay CS, Vidal AM. Use of intravenous polymyxin B during continuous venovenous hemodialysis. Eur J Clin Microbiol Infect Dis 2004;23:340–1. [18] Lin X, Heitman J. The biology of the Cryptococcus neoformans species complex. Annu Rev Microbiol 2006;60:69–105. [19] Chang YC, Kwon-Chung KJ. Complementation of a capsule-deficient mutation of Cryptococcus neoformans restores its virulence. Mol Cell Biol 1994;14:4912–9. [20] Lin X, Hull CM, Heitman J. Sexual reproduction between partners of the same mating type in Cryptococcus neoformans. Nature 2005;434:1017–21. [21] Zhai B, Wu C, Wang L, Sachs MS, Lin X. The antidepressant sertraline provides a promising therapeutic option for neurotropic cryptococcal infections. Antimicrob Agents Chemother 2012;56:3758–66. [22] García-Rivera J, Chang YC, Kwon-Chung KJ, Casadevall A. Cryptococcus neoformans CAP59 (or Cap59p) is involved in the extracellular trafficking of capsular glucuronoxylomannan. Eukaryot Cell 2004;3:385–92. [23] Hermsen ED, Sullivan CJ, Rotschafer JC. Polymyxins: pharmacology, pharmacokinetics, pharmacodynamics, and clinical applications. Infect Dis Clin North Am 2003;17:545–62. [24] Li J, Milne RW, Nation RL, Turnidge JD, Smeaton TC, Coulthard K. Use of high-performance liquid chromatography to study the pharmacokinetics of colistin sulfate in rats following intravenous administration. Antimicrob Agents Chemother 2003;47:1766–70. [25] Zaragoza O, Alvarez M, Telzak A, Rivera J, Casadevall A. The relative susceptibility of mouse strains to pulmonary Cryptococcus neoformans infection is associated with pleiotropic differences in the immune response. Infect Immun 2007;75:2729–39.