- Email: [email protected]
Characterising the post-antifungal effects of micafungin against Candida albicans, Candida glabrata, Candida parapsilosis and Candida krusei isolates
International Journal of Antimicrobial Agents 35 (2010) 80–84
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Short communication
Characterising the post-antifungal effects of micafungin against Candida albicans, Candida glabrata, Candida parapsilosis and Candida krusei isolates Katherine T. Nguyen a , Philip Ta a , Bich Thu Hoang a , Shaoji Cheng a,b , Binghua Hao a,b , M. Hong Nguyen a,b,c,d , Cornelius J. Clancy a,b,d,∗ a
Department of Medicine, University of Florida College of Medicine, Gainesville, FL, USA University of Pittsburgh, Pittsburgh, PA, USA Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Gainesville, FL, USA d North Florida/South Georgia Veterans Health System, Gainesville, FL, USA b c
a r t i c l e
i n f o
Article history: Received 28 January 2009 Accepted 2 September 2009 Keywords: Micafungin Time–kill Post-antifungal effect Candida
a b s t r a c t In this study, we measured time–kills and post-antifungal effects (PAFEs) for micafungin against Candida albicans (n = 4), Candida glabrata (n = 3), Candida parapsilosis (n = 3) and Candida krusei (n = 2) isolates and further characterised the PAFEs. Minimum inhibitory concentrations (MICs) were 0.5–1.0 mg/L against C. parapsilosis and 0.008–0.125 mg/L against the other species. Micafungin caused kills >1 log at 1 × MIC, 4 × MIC (range 1.19–3.10 log) and 16 × MIC (2.27–3.68 log), achieving fungicidal levels (≥3 log) against nine isolates. One-hour drug exposure during PAFE experiments resulted in kills of 0.73–2.88 log and 1.72–3.55 log at 4× and 16× MIC, respectively, achieving fungicidal levels against five isolates. Isolates of each species collected 8 h after a 1-h exposure to micafungin (4× and 16× MIC) were hypersusceptible to sodium dodecyl sulphate (SDS) and Calcofluor White. Cells tested during the PAFE period demonstrated cell wall disturbances as evident on electron micrographs as well as significant reductions in adherence to epithelial cells. Phagocytosis by J774 macrophages was significantly enhanced for three PAFE isolates tested. Micafungin is fungicidal and exerts PAFEs that kill diverse Candida spp., disturb cell walls of viable organisms, reduce adherence and enhance susceptibility to phagocytosis. © 2009 Published by Elsevier B.V. and the International Society of Chemotherapy.
1. Introduction Micafungin is an echinocandin antifungal that inhibits the synthesis of cell wall -1,3-d-glucan. MIC90 values (minimum inhibitory concentrations for 90% of organisms) of micafungin are low against Candida albicans (0.03 mg/L), Candida glabrata (0.015 mg/L), Candida tropicalis (0.06 mg/L) and Candida krusei (0.12 mg/L) but are higher against Candida parapsilosis and Candida guilliermondii (2 mg/L and 1 mg/L, respectively) [1]. Micafungin was fungicidal against C. albicans, C. glabrata, C. tropicalis and C. krusei isolates during time–kill experiments at 4× and 16× MIC [2] and was fungistatic against C. guilliermondii [3]. In post-antifungal effect (PAFE) experiments, 1-h exposure to micafungin at 4× MIC resulted in prolonged growth inhibition of C. albicans, C. glabrata, C. tropicalis and C. krusei isolates [2]. To our knowledge, time–kill and PAFE data have not been published against C. parapsilosis. In this study, we tested the hypotheses that the PAFEs of micafungin
inhibit growth of diverse Candida spp. and cause cell wall changes that decrease cell integrity and adherence to host cells and increase susceptibility to killing by phagocytes. 2. Materials and methods 2.1. Micafungin, Candida isolates and growth rates Micafungin was provided by Astellas Pharma, Inc. (Deerfield, IL). Ten Candida isolates (3 C. albicans, 3 C. glabrata, 3 C. parapsilosis and 1 C. krusei) were recovered from the bloodstream of patients. Candida albicans ATCC 90028 and C. krusei ATCC 6258 were included as controls. In vitro growth rates in the absence of drug were determined in yeast–peptone–dextrose (YPD) and RPMI 1640 liquid media at 30 ◦ C and 37 ◦ C in microtitre plates as described previously [4]. 2.2. Antifungal susceptibility testing
∗ Corresponding author. Present address: University of Pittsburgh, 867 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA. Tel.: +1 412 624 0309; fax: +1 412 648 8455. E-mail address: [email protected] (C.J. Clancy).
Micafungin MICs were determined using a broth microdilution technique in RPMI 1640 medium buffered to pH 7.0 with MOPS, as recommended by the Clinical and Laboratory Standards Institute [5]. Significant antifungal carry-over [defined as a
0924-8579/$ – see front matter © 2009 Published by Elsevier B.V. and the International Society of Chemotherapy. doi:10.1016/j.ijantimicag.2009.09.003
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
reduction in colony-forming units (CFU) >25% of the control] was excluded using standard methods [6]. The starting inoculum was 0.5–2.5 × 103 cells/mL. MICs were read as the lowest concentration that caused 50% diminution of growth below control levels [5]. The range of micafungin concentrations was 0.008 mg/L to 32 mg/L. Each isolate was tested at least three times. 2.3. Time–kill and post-antifungal effects Time–kills and PAFEs were measured simultaneously for each isolate using previously described methods [7]. The starting inoculum was 1–5 × 105 CFU/mL. For each isolate, micafungin concentrations tested were 0.25×, 1×, 4× and 16× MIC. Time points were 0 (to corroborate starting inocula), 2, 4, 8, 12, 24 and 48 h. In PAFE experiments, cells were washed after 1 h incubation with micafungin and were re-suspended in warm RPMI. Time–kill and PAFE experiments were conducted twice for each isolate and data are presented as mean values. Fungicidal activity was defined as ≥3 log kill compared with the starting inoculum over the 48-h study period. 2.4. Sensitivity to cell wall agents [8] Candida cells were exposed to micafungin at 0×, 1×, 4× and 16× MIC. After 1 h of exposure, cells were washed and re-suspended in YPD liquid medium for 8 h at 35 ◦ C, similar to the methods for PAFE experiments. Cells recovered at 8 h were then subcultured in YPD liquid medium with 1% glucose until exponential phase, at which point they were diluted to an optical density at 599 nm (OD599 ) of 0.1. Undiluted and serial 10-fold dilutions of each culture (4 L) were spotted onto YPD plates containing Calcofluor White (40 mg/L) or sodium dodecyl sulphate (SDS) (0.01–0.04%). Plates were incubated at 30 ◦ C for 72 h. Experiments were performed in triplicate. The 8-h time point was chosen to assure adequate time for PAFEs to be manifest.
occasions. Differences between isolates exposed to micafungin and corresponding controls were determined using Student’s t-test. 2.7. Macrophage phagocytosis Methods previously published by our laboratory were used to investigate macrophage phagocytosis [8]. Candida albicans S, C. parapsilosis A and C. glabrata 1 cells were recovered as described for the adherence assays, opsonised with 50% human serum for 30 min at 37 ◦ C, washed with PBS and transferred to RPMI. Cells of cell line J774A.1 (American Type Culture Collection, Manassas, VA) were challenged with Candida cells at a ratio of 1:10 and were incubated for 2 h at 37 ◦ C, after which macrophages were completely lysed with cold sterile distilled water. Candida colonies were enumerated following incubation at 30 ◦ C for 48 h. The percentage of surviving Candida was calculated in comparison with the CFUs of Candida grown in the same conditions without macrophages. The assays were performed in triplicate and were repeated at least twice.
Table 1 Micafungin minimum inhibitory concentrations (MICs) and reductions in starting inocula of Candida isolates during time–kill and post-antifungal effect (PAFE) experiments. Isolate/MIC (g/mL)
Micafungin concentration (× MIC)
Time–kill killing (log)a
PAFE killing (log)a
C. albicans S/0.125
16 4 1
3.38 3.10 2.62
2.55 2.52 2.12
C. albicans ATCC/0.03
16 4 1
2.27 2.18 1.27
1.72 1.48 N/A
C. albicans 1/0.008
16 4 1
2.94 1.19 1.12
3.00 1.39 0.48
C. albicans 2/0.008
16 4 1
2.51 2.35 1.66
1.88 0.73 N/A
C. parapsilosis A/1.0
16 4 1
3.11 2.73 1.52
2.08 1.54 N/A
C. parapsilosis 1/0.5
16 4 1
3.30 2.66 1.33
2.23 1.53 N/A
C. parapsilosis S/0.5
16 4 1
3.15 2.60 1.93
3.20 1.63 N/A
C. glabrata 1/0.125
16 4 1
3.28 2.43 1.82
3.43 2.88 1.14
C. glabrata 2/0.03
16 4 1
3.55 2.64 1.33
3.55 2.82 N/A
C. glabrata 3/0.015
16 4 1
3.68 2.86 1.43
2.60 2.39 N/A
C. krusei ATCC/0.06
16 4 1
3.42 3.00 2.33
3.12 2.79 1.90
C. krusei 1/0.03
16 4 1
3.08 2.89 1.90
2.91 2.64 0.43
2.5. Electron microscopy Transmission electron microscopy was performed by the North Florida/South Georgia Veterans Health System Electron Microscopy Laboratory (Gainesville, FL). Candida cells recovered 8 h after a 1-h exposure to micafungin at 0×, 1× and 4× MIC were cultured on Sabouraud dextrose agar (SDA) plates for 24 h at 35 ◦ C. Cells collected from the plates were fixed at 4 ◦ C in 0.1 M sodium cacodylate buffer (pH 7.2) containing 2% glutaraldehyde and 2% paraformaldehyde. Samples were then dehydrated through a graded series of ethanol and embedded in Lowicryl K4M (Electron Microscopy Sciences, Hatfield, PA). Thin sections were imaged with a Zeiss EM902 electron microscope. 2.6. Adherence to human buccal epithelial cells (BECs) Methods previously published by our laboratory were used to determine adherence to human BECs [9]. BECs were collected from three investigators by gently scraping the cheek mucosa with a cotton swab. Candida cells recovered 8 h after a 1-h exposure to micafungin at 4× MIC were cultured on SDA for 24 h at 35 ◦ C. For the assay, 0.5 mL of washed BECs (1 × 105 /mL) were incubated in a glass tube with 0.5 mL of washed Candida cells (1 × 106 /mL) in phosphate-buffered saline (PBS) at 37 ◦ C for 1 h with shaking; for the control, 0.5 mL of BECs were mixed with 0.5 mL of PBS. Following incubation, the cells were vacuum-filtered and pressed onto glass slides for Gram staining. The number of Candida cells attached to 100 BECs was counted under light microscopy. Each experiment was performed in duplicate on at least two separate
81
N/A, not applicable (no reductions in colony counts compared with the starting inoculum). a Time–kill and PAFE killing data are presented as the largest log reduction in colony counts at each concentration compared with the starting inoculum over the 48-h study period.
82
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
Differences between isolates exposed to micafungin and controls were determined using Student’s t-test. 3. Results 3.1. Micafungin exerts significant time–kills and prolonged post-antifungal effects Micafungin MICs and the results of time–kill and PAFE experiments are summarised in Table 1. Representative time–kill and PAFE curves are shown in Fig. 1 for C. albicans strain S and C. parapsilosis strain A. Micafungin at 0.25× MIC did not result in significant killing of strains and data are not included. The ranges of MICs were consistent with large surveillance studies of micafungin activity in vitro [1]. In time–kill experiments, micafungin caused kills >1 log at 1×, 4× and 16× MIC against all 12 isolates. At 16× MIC, the range of log time–kills was 2.27 to 3.68. Micafungin was fungicidal against nine isolates (defined as ≥3 log kill), including all C. parapsilosis, C. glabrata and C. krusei but only one of the C. albicans. At 4× MIC the range of log time–kills was 1.19 to 3.10, and micafungin was fungicidal against one C. albicans and one C. krusei isolate. One-hour exposure to micafungin during PAFE experiments also resulted in significant time–kills, ranging from 1.72 to 3.55 log at 16× MIC and 0.73 to 2.88 log at 4× MIC. Indeed, 1-h exposure to 16× MIC was fungicidal against five isolates. Moreover, re-growth of each isolate was inhibited for ≥12 h after micafungin was washed out, consistent with sustained PAFEs. Candida isolates that were not killed by micafungin during time–kill and PAFE experiments did not exhibit elevated MICs upon re-testing, nor was growth in liquid media reduced. 3.2. Micafungin post-antifungal effects cause changes to Candida cell walls To characterise PAFEs on cell walls, cells of seven isolates (C. albicans S, C. albicans ATCC 90028, C. parapsilosis A, C. parapsilosis 1, C. glabrata 1, C. glabrata 3 and C. krusei 1) were recovered 8 h after a
1-h exposure to micafungin at 0, 1×, 4× and 16× MIC. Viable cells of each isolate recovered after exposure to micafungin at 4× and 16× MIC were significantly more susceptible to Calcofluor White (40 mg/L) and increasing concentrations of SDS (0.01–0.04%) than control cells (no drug) (data not shown). In addition, C. albicans S and C. krusei 6258 cells exposed to 1× MIC for 1 h were significantly more susceptible to these agents than controls. The cell walls of four strains (C. albicans S, C. parapsilosis A, C. glabrata 1 and C. krusei 1) were examined by electron microscopy. Cells recovered 8 h after a 1-h exposure to micafungin at 0, 1× and 4× MIC were demonstrated to be viable by culturing on SDA for 24 h at 35 ◦ C. Colonies were selected from plates for electron microscopy. All control cells exhibited intact cell walls with normal inner and outer layers. At 4× MIC, cell walls were markedly disturbed with loss of distinct inner and outer layers (Fig. 2). At 1× MIC, the cell wall and plasma membrane of C. albicans S more closely resembled cells recovered from 4× MIC than controls. For the other isolates, 1× MIC cells more closely resembled controls. 3.3. Micafungin post-antifungal effects diminish adherence of Candida isolates to buccal epithelial cells We tested the hypothesis that cells whose cell walls were altered by micafungin PAFEs would be impaired in adherence by co-incubating human BECs with C. albicans S, C. parapsilosis A and C. glabrata 1 cells recovered 8 h after a 1-h exposure to micafungin at 4× MIC. The relative adherence of C. albicans S, C. parapsilosis A and C. glabrata 1 cells was 58 ± 18%, 69 ± 28% and 62 ± 21%, respectively (vs. 100% for corresponding controls) (P ≤ 0.01). 3.4. Candida cells tested during the post-antifungal effect period are more susceptible to phagocytosis by macrophages Finally, we tested the hypothesis that the changes induced by PAFEs would make cells more susceptible to phagocytosis by coincubating J774 macrophages with C. albicans S, C. parapsilosis A and C. glabrata 1 cells that had been recovered after exposure to
Fig. 1. Representative time–kill and post-antifungal effect (PAFE) curves for Candida isolates at micafungin concentrations of 1×, 4× and 16× the minimum inhibitory concentration (MIC). CFU, colony-forming units.
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
83
Fig. 2. Electron micrographs of (A) viable Candida albicans S cells and (B) viable Candida glabrata 1 cells recovered 8 h after a 1-h incubation with micafungin at 0× the minimum inhibitory concentration (MIC) (control), 1× MIC and 4× MIC. Note: in (B) the cells were mislabelled as C. albicans in the second line of the legend beneath each image.
micafungin at 4× MIC as in the previous experiments. For C. albicans S, the percentage survival of cells tested during the PAFE period and of control cells in phagocytosis assays was 61 ± 8% vs. 82 ± 9%, respectively (P = 0.008). Corresponding values were 52 ± 6% vs. 78 ± 11% for C. parapsilosis A (P = 0.004) and 55 ± 12% vs. 74 ± 4% for C. glabrata 1 (P = 0.01). 4. Discussion In this study, we demonstrated that micafungin killed 12 Candida isolates during time–kill experiments at concentrations equal to or greater than the MIC, achieving fungicidal levels against nine isolates. The findings with C. albicans, C. glabrata and C. krusei corroborated a smaller time–kill study [2]. Fungicidal activity against C. glabrata and C. krusei, in particular, is noteworthy given the acquired and intrinsic resistance of these species to fluconazole. More importantly, this is the first report that micafungin was fungicidal against C. parapsilosis. Our findings suggest that micafungin killed C. parapsilosis at least as well as the other species, providing sufficient drug concentrations above the MIC were used. These results are potentially significant because elevated echinocandin
MICs exhibited by many C. parapsilosis isolates have raised concerns about using these agents to treat infections by this species [10]. Along these lines, it is notable that the highest MIC in this study (exhibited by a C. parapsilosis isolate) was 1 mg/L. Therefore, for each isolate tested micafungin concentrations that resulted in kills were lower than the plasma peak concentration (ca. 8–12 mg/L) typically attained with conventional dosing [11]. As with the time–kills, the PAFE results both confirmed and extended prior observations. Similar to a previous report [5], micafungin exerted prolonged PAFEs against C. albicans, C. glabrata and C. krusei. We documented similar PAFEs against C. parapsilosis isolates, the first time that such results have been reported. One-hour exposure to micafungin was fungicidal against five isolates and inhibited re-growth of all isolates for ≥12 h. It is notable that Candida isolates that were not killed by micafungin did not exhibit elevated MICs upon re-testing, indicating that growth was not due to acquired resistance or pre-existing resistant subpopulations. Rather, it is possible that these isolates were dormant or replicated slowly in the presence of micafungin [7,12]. If so, they were not irreversibly impaired, as growth in liquid media after recovery from time–kill and PAFE experiments was not reduced.
84
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
Although viable, Candida cells that were recovered from PAFE experiments following exposure to 4× MIC demonstrated cell wall disturbances by electron microscopy, including profound thinning of the inner and outer cell wall layers. Indeed, cell wall abnormalities were present 32 h following drug exposure (8-h recovery in YPD liquid medium and then 24 h growth on agar). The isolates that were exposed to micafungin were also more susceptible to cell wall-active drugs than unexposed controls. These findings are consistent with sustained effects of micafungin on the -1,3-dglucan synthase complex. Taken with the level of kills seen after 1 h exposure to micafungin, our findings suggest that the drug rapidly associates with its target and then maintains activity against it. Alternatively, micafungin might intercalate with the phospholipid bilayer of the cell membrane and subsequently access its target over time. Finally, we demonstrated that cell wall changes resulting from the PAFEs of micafungin are associated with decreased adherence to BECs and increased susceptibility to killing by macrophages. These findings are consistent with a report that exposure to micafungin attenuated the adherence of a C. albicans isolate to epithelial cells [13]. Along these lines, we hypothesise that micafungin alters the distribution of adhesins and phagocyte ligands on the Candida cell surface. Recent studies have demonstrated that subinhibitory concentrations of caspofungin unmask cell wall -glucans, resulting in greater -glucan receptor-mediated expression of inflammatory cytokines by macrophages [14]. Under normal growth conditions, -glucans are not recognised by phagocytic cells because they are buried beneath a mannoprotein coat [14]. Although micafungin inhibits -1,3-glucan synthesis, the global disruption of cell wall architecture that results from drug exposure might paradoxically increase -glucan exposure. Taken together, the adhesion and phagocytosis data suggest that Candida cells not killed by micafungin might be less able to establish invasive disease and persist within infected hosts. Although the relevance of PAFEs to the treatment of infections in vivo remains unproven [15], both PAFEs and the direct fungicidal activity of micafungin and other echinocandins might contribute to their clinical efficacy. Acknowledgments Experiments were performed in Dr Clancy’s laboratory at North Florida/South Georgia Veterans Health System, Gainesville, FL.
Funding: This project was funded by a research grant to CJC from Astellas Pharma, Inc. Competing interests: CJC has received research support from Pfizer, Inc. and Merck, Inc., and was supported by the Medical Research Service of the Department of Veterans Affairs. Ethical approval: Not required. References [1] Pfaller MA, Boyken L, Hollis RJ, Kroeger J, Messer SA, Tendolkar S, et al. In vitro susceptibility of invasive isolates of Candida spp. to anidulafungin, caspofungin, and micafungin: six years of global surveillance. J Clin Microbiol 2008;46:150–6. [2] Ernst EJ, Roling EE, Perzold CR, Keele DJ, Klepser ME. In vitro activity of micafungin (FK-463) against Candida spp.: microdilution, time–kill, and postantifungaleffect studies. Antimicrob Agents Chemother 2002;46:3849–53. [3] Cantón E, Pemán J, Sastre M, Romero M, Espinel-Ingroff A. Killing kinetics of caspofungin, micafungin, and amphotericin B against Candida guilliermondii. Antimicrob Agents Chemother 2006;50:2829–32. [4] Cheng S, Clancy CJ, Checkley MA, Zhang Z, Wozniak KL, Seshan KR, et al. The role of Candida albicans NOT5 in virulence depends upon diverse host factors in vivo. Infect Immun 2005;73:7190–7. [5] Clinical and Laboratory Standards Institute. Reference method for broth dilution antifungal susceptibility testing of yeasts; approved standard. 3rd ed. Document M27-A3. Wayne, PA: CLSI; 2007. [6] Klepser ME, Ernst EJ, Lewis RE, Ernst ME, Pfaller MA. Influence of test conditions on antifungal time–kill curve results: proposal for standardized methods. Antimicrob Agents Chemother 1998;5:1207–12. [7] Clancy CJ, Huang H, Cheng S, Derendorf H, Nguyen MH. Characterizing the effects of caspofungin on Candida albicans, Candida parapsilosis, and Candida glabrata isolates by simultaneous time–kill and postantifungal-effect experiments. Antimicrob Agents Chemother 2006;50:2569–72. [8] Cheng S, Clancy CJ, Zhang Z, Hao B, Wang W, Iczkowski KA, et al. Uncoupling of oxidative phosphorylation enables Candida albicans to resist killing by phagocytes and persist in tissue. Cell Microbiol 2007;9:492–501. [9] Cheng S, Clancy CJ, Checkley MA, Handfield M, Hillman JD, Progulske-Fox A, et al. Identification of Candida albicans genes induced during thrush offers insight into pathogenesis. Mol Microbiol 2003;48:1275–88. [10] Walsh TJ. Echinocandins—an advance in the primary treatment of invasive candidiasis. N Engl J Med 2002;347:2070–2. [11] Theuretzbacher U. Pharmacokinetics/pharmacodynamics of echinocandins. Eur J Clin Microbiol Infect Dis 2004;23:805–12. [12] Bryan LE. Two forms of antimicrobial resistance: bacterial persistence and positive function resistance. J Antimicrob Chemother 1989;23:817–23. [13] Borg-von Zepelin M, Zaschke K, Gross U, Monod M, Müller FM. Effect of micafungin (FK463) on Candida albicans adherence to epithelial cells. Chemotherapy 2002;48:148–53. [14] Wheeler RT, Fink GR. A drug-sensitive genetic network masks fungi from the immune system. PLOS Pathog 2006;2:e35. [15] Andes D. In vivo pharmacodynamics of antifungal drugs in treatment of candidiasis. Antimicrob Agents Chemother 2003;47:1179–86.
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
Short communication
Characterising the post-antifungal effects of micafungin against Candida albicans, Candida glabrata, Candida parapsilosis and Candida krusei isolates Katherine T. Nguyen a , Philip Ta a , Bich Thu Hoang a , Shaoji Cheng a,b , Binghua Hao a,b , M. Hong Nguyen a,b,c,d , Cornelius J. Clancy a,b,d,∗ a
Department of Medicine, University of Florida College of Medicine, Gainesville, FL, USA University of Pittsburgh, Pittsburgh, PA, USA Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Gainesville, FL, USA d North Florida/South Georgia Veterans Health System, Gainesville, FL, USA b c
a r t i c l e
i n f o
Article history: Received 28 January 2009 Accepted 2 September 2009 Keywords: Micafungin Time–kill Post-antifungal effect Candida
a b s t r a c t In this study, we measured time–kills and post-antifungal effects (PAFEs) for micafungin against Candida albicans (n = 4), Candida glabrata (n = 3), Candida parapsilosis (n = 3) and Candida krusei (n = 2) isolates and further characterised the PAFEs. Minimum inhibitory concentrations (MICs) were 0.5–1.0 mg/L against C. parapsilosis and 0.008–0.125 mg/L against the other species. Micafungin caused kills >1 log at 1 × MIC, 4 × MIC (range 1.19–3.10 log) and 16 × MIC (2.27–3.68 log), achieving fungicidal levels (≥3 log) against nine isolates. One-hour drug exposure during PAFE experiments resulted in kills of 0.73–2.88 log and 1.72–3.55 log at 4× and 16× MIC, respectively, achieving fungicidal levels against five isolates. Isolates of each species collected 8 h after a 1-h exposure to micafungin (4× and 16× MIC) were hypersusceptible to sodium dodecyl sulphate (SDS) and Calcofluor White. Cells tested during the PAFE period demonstrated cell wall disturbances as evident on electron micrographs as well as significant reductions in adherence to epithelial cells. Phagocytosis by J774 macrophages was significantly enhanced for three PAFE isolates tested. Micafungin is fungicidal and exerts PAFEs that kill diverse Candida spp., disturb cell walls of viable organisms, reduce adherence and enhance susceptibility to phagocytosis. © 2009 Published by Elsevier B.V. and the International Society of Chemotherapy.
1. Introduction Micafungin is an echinocandin antifungal that inhibits the synthesis of cell wall -1,3-d-glucan. MIC90 values (minimum inhibitory concentrations for 90% of organisms) of micafungin are low against Candida albicans (0.03 mg/L), Candida glabrata (0.015 mg/L), Candida tropicalis (0.06 mg/L) and Candida krusei (0.12 mg/L) but are higher against Candida parapsilosis and Candida guilliermondii (2 mg/L and 1 mg/L, respectively) [1]. Micafungin was fungicidal against C. albicans, C. glabrata, C. tropicalis and C. krusei isolates during time–kill experiments at 4× and 16× MIC [2] and was fungistatic against C. guilliermondii [3]. In post-antifungal effect (PAFE) experiments, 1-h exposure to micafungin at 4× MIC resulted in prolonged growth inhibition of C. albicans, C. glabrata, C. tropicalis and C. krusei isolates [2]. To our knowledge, time–kill and PAFE data have not been published against C. parapsilosis. In this study, we tested the hypotheses that the PAFEs of micafungin
inhibit growth of diverse Candida spp. and cause cell wall changes that decrease cell integrity and adherence to host cells and increase susceptibility to killing by phagocytes. 2. Materials and methods 2.1. Micafungin, Candida isolates and growth rates Micafungin was provided by Astellas Pharma, Inc. (Deerfield, IL). Ten Candida isolates (3 C. albicans, 3 C. glabrata, 3 C. parapsilosis and 1 C. krusei) were recovered from the bloodstream of patients. Candida albicans ATCC 90028 and C. krusei ATCC 6258 were included as controls. In vitro growth rates in the absence of drug were determined in yeast–peptone–dextrose (YPD) and RPMI 1640 liquid media at 30 ◦ C and 37 ◦ C in microtitre plates as described previously [4]. 2.2. Antifungal susceptibility testing
∗ Corresponding author. Present address: University of Pittsburgh, 867 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA. Tel.: +1 412 624 0309; fax: +1 412 648 8455. E-mail address: [email protected] (C.J. Clancy).
Micafungin MICs were determined using a broth microdilution technique in RPMI 1640 medium buffered to pH 7.0 with MOPS, as recommended by the Clinical and Laboratory Standards Institute [5]. Significant antifungal carry-over [defined as a
0924-8579/$ – see front matter © 2009 Published by Elsevier B.V. and the International Society of Chemotherapy. doi:10.1016/j.ijantimicag.2009.09.003
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
reduction in colony-forming units (CFU) >25% of the control] was excluded using standard methods [6]. The starting inoculum was 0.5–2.5 × 103 cells/mL. MICs were read as the lowest concentration that caused 50% diminution of growth below control levels [5]. The range of micafungin concentrations was 0.008 mg/L to 32 mg/L. Each isolate was tested at least three times. 2.3. Time–kill and post-antifungal effects Time–kills and PAFEs were measured simultaneously for each isolate using previously described methods [7]. The starting inoculum was 1–5 × 105 CFU/mL. For each isolate, micafungin concentrations tested were 0.25×, 1×, 4× and 16× MIC. Time points were 0 (to corroborate starting inocula), 2, 4, 8, 12, 24 and 48 h. In PAFE experiments, cells were washed after 1 h incubation with micafungin and were re-suspended in warm RPMI. Time–kill and PAFE experiments were conducted twice for each isolate and data are presented as mean values. Fungicidal activity was defined as ≥3 log kill compared with the starting inoculum over the 48-h study period. 2.4. Sensitivity to cell wall agents [8] Candida cells were exposed to micafungin at 0×, 1×, 4× and 16× MIC. After 1 h of exposure, cells were washed and re-suspended in YPD liquid medium for 8 h at 35 ◦ C, similar to the methods for PAFE experiments. Cells recovered at 8 h were then subcultured in YPD liquid medium with 1% glucose until exponential phase, at which point they were diluted to an optical density at 599 nm (OD599 ) of 0.1. Undiluted and serial 10-fold dilutions of each culture (4 L) were spotted onto YPD plates containing Calcofluor White (40 mg/L) or sodium dodecyl sulphate (SDS) (0.01–0.04%). Plates were incubated at 30 ◦ C for 72 h. Experiments were performed in triplicate. The 8-h time point was chosen to assure adequate time for PAFEs to be manifest.
occasions. Differences between isolates exposed to micafungin and corresponding controls were determined using Student’s t-test. 2.7. Macrophage phagocytosis Methods previously published by our laboratory were used to investigate macrophage phagocytosis [8]. Candida albicans S, C. parapsilosis A and C. glabrata 1 cells were recovered as described for the adherence assays, opsonised with 50% human serum for 30 min at 37 ◦ C, washed with PBS and transferred to RPMI. Cells of cell line J774A.1 (American Type Culture Collection, Manassas, VA) were challenged with Candida cells at a ratio of 1:10 and were incubated for 2 h at 37 ◦ C, after which macrophages were completely lysed with cold sterile distilled water. Candida colonies were enumerated following incubation at 30 ◦ C for 48 h. The percentage of surviving Candida was calculated in comparison with the CFUs of Candida grown in the same conditions without macrophages. The assays were performed in triplicate and were repeated at least twice.
Table 1 Micafungin minimum inhibitory concentrations (MICs) and reductions in starting inocula of Candida isolates during time–kill and post-antifungal effect (PAFE) experiments. Isolate/MIC (g/mL)
Micafungin concentration (× MIC)
Time–kill killing (log)a
PAFE killing (log)a
C. albicans S/0.125
16 4 1
3.38 3.10 2.62
2.55 2.52 2.12
C. albicans ATCC/0.03
16 4 1
2.27 2.18 1.27
1.72 1.48 N/A
C. albicans 1/0.008
16 4 1
2.94 1.19 1.12
3.00 1.39 0.48
C. albicans 2/0.008
16 4 1
2.51 2.35 1.66
1.88 0.73 N/A
C. parapsilosis A/1.0
16 4 1
3.11 2.73 1.52
2.08 1.54 N/A
C. parapsilosis 1/0.5
16 4 1
3.30 2.66 1.33
2.23 1.53 N/A
C. parapsilosis S/0.5
16 4 1
3.15 2.60 1.93
3.20 1.63 N/A
C. glabrata 1/0.125
16 4 1
3.28 2.43 1.82
3.43 2.88 1.14
C. glabrata 2/0.03
16 4 1
3.55 2.64 1.33
3.55 2.82 N/A
C. glabrata 3/0.015
16 4 1
3.68 2.86 1.43
2.60 2.39 N/A
C. krusei ATCC/0.06
16 4 1
3.42 3.00 2.33
3.12 2.79 1.90
C. krusei 1/0.03
16 4 1
3.08 2.89 1.90
2.91 2.64 0.43
2.5. Electron microscopy Transmission electron microscopy was performed by the North Florida/South Georgia Veterans Health System Electron Microscopy Laboratory (Gainesville, FL). Candida cells recovered 8 h after a 1-h exposure to micafungin at 0×, 1× and 4× MIC were cultured on Sabouraud dextrose agar (SDA) plates for 24 h at 35 ◦ C. Cells collected from the plates were fixed at 4 ◦ C in 0.1 M sodium cacodylate buffer (pH 7.2) containing 2% glutaraldehyde and 2% paraformaldehyde. Samples were then dehydrated through a graded series of ethanol and embedded in Lowicryl K4M (Electron Microscopy Sciences, Hatfield, PA). Thin sections were imaged with a Zeiss EM902 electron microscope. 2.6. Adherence to human buccal epithelial cells (BECs) Methods previously published by our laboratory were used to determine adherence to human BECs [9]. BECs were collected from three investigators by gently scraping the cheek mucosa with a cotton swab. Candida cells recovered 8 h after a 1-h exposure to micafungin at 4× MIC were cultured on SDA for 24 h at 35 ◦ C. For the assay, 0.5 mL of washed BECs (1 × 105 /mL) were incubated in a glass tube with 0.5 mL of washed Candida cells (1 × 106 /mL) in phosphate-buffered saline (PBS) at 37 ◦ C for 1 h with shaking; for the control, 0.5 mL of BECs were mixed with 0.5 mL of PBS. Following incubation, the cells were vacuum-filtered and pressed onto glass slides for Gram staining. The number of Candida cells attached to 100 BECs was counted under light microscopy. Each experiment was performed in duplicate on at least two separate
81
N/A, not applicable (no reductions in colony counts compared with the starting inoculum). a Time–kill and PAFE killing data are presented as the largest log reduction in colony counts at each concentration compared with the starting inoculum over the 48-h study period.
82
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
Differences between isolates exposed to micafungin and controls were determined using Student’s t-test. 3. Results 3.1. Micafungin exerts significant time–kills and prolonged post-antifungal effects Micafungin MICs and the results of time–kill and PAFE experiments are summarised in Table 1. Representative time–kill and PAFE curves are shown in Fig. 1 for C. albicans strain S and C. parapsilosis strain A. Micafungin at 0.25× MIC did not result in significant killing of strains and data are not included. The ranges of MICs were consistent with large surveillance studies of micafungin activity in vitro [1]. In time–kill experiments, micafungin caused kills >1 log at 1×, 4× and 16× MIC against all 12 isolates. At 16× MIC, the range of log time–kills was 2.27 to 3.68. Micafungin was fungicidal against nine isolates (defined as ≥3 log kill), including all C. parapsilosis, C. glabrata and C. krusei but only one of the C. albicans. At 4× MIC the range of log time–kills was 1.19 to 3.10, and micafungin was fungicidal against one C. albicans and one C. krusei isolate. One-hour exposure to micafungin during PAFE experiments also resulted in significant time–kills, ranging from 1.72 to 3.55 log at 16× MIC and 0.73 to 2.88 log at 4× MIC. Indeed, 1-h exposure to 16× MIC was fungicidal against five isolates. Moreover, re-growth of each isolate was inhibited for ≥12 h after micafungin was washed out, consistent with sustained PAFEs. Candida isolates that were not killed by micafungin during time–kill and PAFE experiments did not exhibit elevated MICs upon re-testing, nor was growth in liquid media reduced. 3.2. Micafungin post-antifungal effects cause changes to Candida cell walls To characterise PAFEs on cell walls, cells of seven isolates (C. albicans S, C. albicans ATCC 90028, C. parapsilosis A, C. parapsilosis 1, C. glabrata 1, C. glabrata 3 and C. krusei 1) were recovered 8 h after a
1-h exposure to micafungin at 0, 1×, 4× and 16× MIC. Viable cells of each isolate recovered after exposure to micafungin at 4× and 16× MIC were significantly more susceptible to Calcofluor White (40 mg/L) and increasing concentrations of SDS (0.01–0.04%) than control cells (no drug) (data not shown). In addition, C. albicans S and C. krusei 6258 cells exposed to 1× MIC for 1 h were significantly more susceptible to these agents than controls. The cell walls of four strains (C. albicans S, C. parapsilosis A, C. glabrata 1 and C. krusei 1) were examined by electron microscopy. Cells recovered 8 h after a 1-h exposure to micafungin at 0, 1× and 4× MIC were demonstrated to be viable by culturing on SDA for 24 h at 35 ◦ C. Colonies were selected from plates for electron microscopy. All control cells exhibited intact cell walls with normal inner and outer layers. At 4× MIC, cell walls were markedly disturbed with loss of distinct inner and outer layers (Fig. 2). At 1× MIC, the cell wall and plasma membrane of C. albicans S more closely resembled cells recovered from 4× MIC than controls. For the other isolates, 1× MIC cells more closely resembled controls. 3.3. Micafungin post-antifungal effects diminish adherence of Candida isolates to buccal epithelial cells We tested the hypothesis that cells whose cell walls were altered by micafungin PAFEs would be impaired in adherence by co-incubating human BECs with C. albicans S, C. parapsilosis A and C. glabrata 1 cells recovered 8 h after a 1-h exposure to micafungin at 4× MIC. The relative adherence of C. albicans S, C. parapsilosis A and C. glabrata 1 cells was 58 ± 18%, 69 ± 28% and 62 ± 21%, respectively (vs. 100% for corresponding controls) (P ≤ 0.01). 3.4. Candida cells tested during the post-antifungal effect period are more susceptible to phagocytosis by macrophages Finally, we tested the hypothesis that the changes induced by PAFEs would make cells more susceptible to phagocytosis by coincubating J774 macrophages with C. albicans S, C. parapsilosis A and C. glabrata 1 cells that had been recovered after exposure to
Fig. 1. Representative time–kill and post-antifungal effect (PAFE) curves for Candida isolates at micafungin concentrations of 1×, 4× and 16× the minimum inhibitory concentration (MIC). CFU, colony-forming units.
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
83
Fig. 2. Electron micrographs of (A) viable Candida albicans S cells and (B) viable Candida glabrata 1 cells recovered 8 h after a 1-h incubation with micafungin at 0× the minimum inhibitory concentration (MIC) (control), 1× MIC and 4× MIC. Note: in (B) the cells were mislabelled as C. albicans in the second line of the legend beneath each image.
micafungin at 4× MIC as in the previous experiments. For C. albicans S, the percentage survival of cells tested during the PAFE period and of control cells in phagocytosis assays was 61 ± 8% vs. 82 ± 9%, respectively (P = 0.008). Corresponding values were 52 ± 6% vs. 78 ± 11% for C. parapsilosis A (P = 0.004) and 55 ± 12% vs. 74 ± 4% for C. glabrata 1 (P = 0.01). 4. Discussion In this study, we demonstrated that micafungin killed 12 Candida isolates during time–kill experiments at concentrations equal to or greater than the MIC, achieving fungicidal levels against nine isolates. The findings with C. albicans, C. glabrata and C. krusei corroborated a smaller time–kill study [2]. Fungicidal activity against C. glabrata and C. krusei, in particular, is noteworthy given the acquired and intrinsic resistance of these species to fluconazole. More importantly, this is the first report that micafungin was fungicidal against C. parapsilosis. Our findings suggest that micafungin killed C. parapsilosis at least as well as the other species, providing sufficient drug concentrations above the MIC were used. These results are potentially significant because elevated echinocandin
MICs exhibited by many C. parapsilosis isolates have raised concerns about using these agents to treat infections by this species [10]. Along these lines, it is notable that the highest MIC in this study (exhibited by a C. parapsilosis isolate) was 1 mg/L. Therefore, for each isolate tested micafungin concentrations that resulted in kills were lower than the plasma peak concentration (ca. 8–12 mg/L) typically attained with conventional dosing [11]. As with the time–kills, the PAFE results both confirmed and extended prior observations. Similar to a previous report [5], micafungin exerted prolonged PAFEs against C. albicans, C. glabrata and C. krusei. We documented similar PAFEs against C. parapsilosis isolates, the first time that such results have been reported. One-hour exposure to micafungin was fungicidal against five isolates and inhibited re-growth of all isolates for ≥12 h. It is notable that Candida isolates that were not killed by micafungin did not exhibit elevated MICs upon re-testing, indicating that growth was not due to acquired resistance or pre-existing resistant subpopulations. Rather, it is possible that these isolates were dormant or replicated slowly in the presence of micafungin [7,12]. If so, they were not irreversibly impaired, as growth in liquid media after recovery from time–kill and PAFE experiments was not reduced.
84
K.T. Nguyen et al. / International Journal of Antimicrobial Agents 35 (2010) 80–84
Although viable, Candida cells that were recovered from PAFE experiments following exposure to 4× MIC demonstrated cell wall disturbances by electron microscopy, including profound thinning of the inner and outer cell wall layers. Indeed, cell wall abnormalities were present 32 h following drug exposure (8-h recovery in YPD liquid medium and then 24 h growth on agar). The isolates that were exposed to micafungin were also more susceptible to cell wall-active drugs than unexposed controls. These findings are consistent with sustained effects of micafungin on the -1,3-dglucan synthase complex. Taken with the level of kills seen after 1 h exposure to micafungin, our findings suggest that the drug rapidly associates with its target and then maintains activity against it. Alternatively, micafungin might intercalate with the phospholipid bilayer of the cell membrane and subsequently access its target over time. Finally, we demonstrated that cell wall changes resulting from the PAFEs of micafungin are associated with decreased adherence to BECs and increased susceptibility to killing by macrophages. These findings are consistent with a report that exposure to micafungin attenuated the adherence of a C. albicans isolate to epithelial cells [13]. Along these lines, we hypothesise that micafungin alters the distribution of adhesins and phagocyte ligands on the Candida cell surface. Recent studies have demonstrated that subinhibitory concentrations of caspofungin unmask cell wall -glucans, resulting in greater -glucan receptor-mediated expression of inflammatory cytokines by macrophages [14]. Under normal growth conditions, -glucans are not recognised by phagocytic cells because they are buried beneath a mannoprotein coat [14]. Although micafungin inhibits -1,3-glucan synthesis, the global disruption of cell wall architecture that results from drug exposure might paradoxically increase -glucan exposure. Taken together, the adhesion and phagocytosis data suggest that Candida cells not killed by micafungin might be less able to establish invasive disease and persist within infected hosts. Although the relevance of PAFEs to the treatment of infections in vivo remains unproven [15], both PAFEs and the direct fungicidal activity of micafungin and other echinocandins might contribute to their clinical efficacy. Acknowledgments Experiments were performed in Dr Clancy’s laboratory at North Florida/South Georgia Veterans Health System, Gainesville, FL.
Funding: This project was funded by a research grant to CJC from Astellas Pharma, Inc. Competing interests: CJC has received research support from Pfizer, Inc. and Merck, Inc., and was supported by the Medical Research Service of the Department of Veterans Affairs. Ethical approval: Not required. References [1] Pfaller MA, Boyken L, Hollis RJ, Kroeger J, Messer SA, Tendolkar S, et al. In vitro susceptibility of invasive isolates of Candida spp. to anidulafungin, caspofungin, and micafungin: six years of global surveillance. J Clin Microbiol 2008;46:150–6. [2] Ernst EJ, Roling EE, Perzold CR, Keele DJ, Klepser ME. In vitro activity of micafungin (FK-463) against Candida spp.: microdilution, time–kill, and postantifungaleffect studies. Antimicrob Agents Chemother 2002;46:3849–53. [3] Cantón E, Pemán J, Sastre M, Romero M, Espinel-Ingroff A. Killing kinetics of caspofungin, micafungin, and amphotericin B against Candida guilliermondii. Antimicrob Agents Chemother 2006;50:2829–32. [4] Cheng S, Clancy CJ, Checkley MA, Zhang Z, Wozniak KL, Seshan KR, et al. The role of Candida albicans NOT5 in virulence depends upon diverse host factors in vivo. Infect Immun 2005;73:7190–7. [5] Clinical and Laboratory Standards Institute. Reference method for broth dilution antifungal susceptibility testing of yeasts; approved standard. 3rd ed. Document M27-A3. Wayne, PA: CLSI; 2007. [6] Klepser ME, Ernst EJ, Lewis RE, Ernst ME, Pfaller MA. Influence of test conditions on antifungal time–kill curve results: proposal for standardized methods. Antimicrob Agents Chemother 1998;5:1207–12. [7] Clancy CJ, Huang H, Cheng S, Derendorf H, Nguyen MH. Characterizing the effects of caspofungin on Candida albicans, Candida parapsilosis, and Candida glabrata isolates by simultaneous time–kill and postantifungal-effect experiments. Antimicrob Agents Chemother 2006;50:2569–72. [8] Cheng S, Clancy CJ, Zhang Z, Hao B, Wang W, Iczkowski KA, et al. Uncoupling of oxidative phosphorylation enables Candida albicans to resist killing by phagocytes and persist in tissue. Cell Microbiol 2007;9:492–501. [9] Cheng S, Clancy CJ, Checkley MA, Handfield M, Hillman JD, Progulske-Fox A, et al. Identification of Candida albicans genes induced during thrush offers insight into pathogenesis. Mol Microbiol 2003;48:1275–88. [10] Walsh TJ. Echinocandins—an advance in the primary treatment of invasive candidiasis. N Engl J Med 2002;347:2070–2. [11] Theuretzbacher U. Pharmacokinetics/pharmacodynamics of echinocandins. Eur J Clin Microbiol Infect Dis 2004;23:805–12. [12] Bryan LE. Two forms of antimicrobial resistance: bacterial persistence and positive function resistance. J Antimicrob Chemother 1989;23:817–23. [13] Borg-von Zepelin M, Zaschke K, Gross U, Monod M, Müller FM. Effect of micafungin (FK463) on Candida albicans adherence to epithelial cells. Chemotherapy 2002;48:148–53. [14] Wheeler RT, Fink GR. A drug-sensitive genetic network masks fungi from the immune system. PLOS Pathog 2006;2:e35. [15] Andes D. In vivo pharmacodynamics of antifungal drugs in treatment of candidiasis. Antimicrob Agents Chemother 2003;47:1179–86.