Medical Mycology 2002, 40, 21±26 Accepted 25 April 2001

Transcription

1 ã Medical Mycology 2002, 40, 21±26 Accepted 25 April 2001 Effect of granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor on polymorphonuclear neutrophils, monocytes or monocyte-derived macrophages combined with voriconazole against Cryptococcus neoformans T. CHILLER*y, K. FARROKHSHAD*, E. BRUMMER*y & D. A. STEVENS*y *Division of Infectious Diseases, Department of Medicine, Santa Clara Valley Medical Center, and California Institute for Medical Research, San Jose, CA; ystanford University School of Medicine, Stanford, CA, USA The antifungal activity of voriconazole () was tested against Cryptococcus neoformans (Cn) with and without the addition of polymorphonuclear neutrophils (PMN), monocytes or monocyte-derived macrophages (MDM) in vitro. Human effector cells with and without the addition of were incubated with Cn for 24 h. PMN, mono and MDM alone resulted in 61%, 34% and 23% inhibition of Cn, respectively (n ˆ 3, P<0 01). at 0 01 and 0 05 m g ml 1 alone resulted in 48% inhibition and 19% killing (n ˆ 6). The addition of at 0 01 and 0 05 m g ml 1 to human effector cells enhanced killing of Cn by 51% and 71% for the PMN, 41% and 58% for the mono, and 14% and 34% for the MDM, respectively. The addition of either granulocyte colony-stimulating factor (G-CSF) or granulocyte-macrophage colony stimulating factor (GM-CSF) signi cantly enhanced the ability of human effector cells to kill Cn. G-CSF and GM-CSF plus PMN resulted in 47% and 46% killing, respectively; GM-CSF plus monocytes or MDM resulted in 31% or 22% killing, respectively. G-CSF and GM-CSF further enhanced the collaborative killing effect of human effector cells and. At 0 01 and 0 05 m g ml 1 of, G-CSF or GM-CSF enhanced PMN killing to 92% and 93% or 87% and 94%, respectively. GM-CSF enhanced both mono and MDM with at 0 01 and 0 05 m g ml 1 in killing Cn to 62% and 86%, and 61% and 84%, respectively. These results suggest that would have good ef cacy in the treatment of Cn infection in humans. Furthermore, would have enhanced ef cacy in clinical settings where either G- CSF or GM-CSF was being used. Keywords colony-stimulating factors, Cryptococcus neoformans, human effector cells, voriconazole Introduction Cryptococcus neoformans is an important cause of disease in immunocompromised individuals. Most often, Correspondence: Dr. D. A. Stevens, Division of Infectious Diseases, Department of Medicine, Santa Clara Valley Med Center, 751 S. Bascom Ave, San Jose, CA 95128, USA. Tel.: ; Fax: ; stevens@leland.stanford.edu. disseminated infection localizes to the central nervous system (CNS) causing meningitis, which develops in as many as 10 15% of patients with AIDS. Other conditions associated with impaired cellular immunity, such as corticosteroid use and hematological malignancies, also put patients at risk for C. neoformans infection. As the fungus normally enters the body via the respiratory tract, the cellular response in the lung is important in host defense against infection. Experimenã 2002 ISHAM, Medical Mycology, 40, 21±26

2 22 Chiller et al. tal evidence in animal models and clinical studies has demonstrated that the alveolar macrophages play an important role in the rst line of defense in the lung [1,2]. Studies have further shown that peripheral polymorphonuclear neutrophils (PMN) and monocytes can kill C. neoformans in vitro and appear to be important in host defense in vivo [3,4]. It has been previously shown that macrophage colony stimulating factor (M-CSF) can enhance the ability of host effector cells to inhibit C. neoformans in vitro [5]. Cytokines have also been shown to enhance anti-cryptococcal activity of effector cells from patients. G-CSF and GM-CSF given to patients had increased effector cell anti-cryptococcal activity in vitro [6]. Voriconazole () is a new, oral triazole, with a broad spectrum of activity against a variety of fungi [7]. In vitro studies have demonstrated up to 10-fold greater activity against C. neoformans than currently available azoles and in vivo studies demonstrate good central nervous system penetration [8,9]. Given these studies, should have potential as a new agent in the therapy of cryptococcosis. It has been previously shown that can collaborate with human effector cells stimulated with either G- CSF or GM-CSF and enhance both inhibition and killing of Aspergillus fumigatus and Candida albicans [10,11]. In this study we have tested whether can collaborate with human effector cells and G-CSF or GM-CSF to increase killing of C. neoformans. Materials and methods C. neoformans An isolate of C. neoformans var. neoformans serotype A (46545) was used in all experiments. This isolate was originally obtained from the Research Center for Pathogenic Fungi and Microbial Toxicosis at Chiba University in Japan. It was stored at the California Institute for Medical Research either at 4 o C or room temperature. Subcultures were grown at 35 o C for 24 h on Sabouraud s glucose agar (SGA) plates, harvested, washed once, counted, and resuspended at a concentration of cells ml 1 in RPMI-1640 plus 10% autologous human serum (referred to as complete tissue culture medium, CTCM). Inocula were dispensed as a volume of 0 1 ml ( cells) per well in sets of quadruplicate wells in a microtest 96-well at-bottom plate. was supplied by P zer, Groton, Conn. powder was rst dissolved into solution in dimethyl sulfoxide (DMSO) and then diluted with distilled water to a nal concentration of 2 mg ml 1 and stored at 4 o C. The desired concentrations for the studies were made from the stock solution and prepared in CTCM. Preliminary experiments showed no effect of DMSO alone, at the concentration that would be present in drug dilutions, on effector cell activity. G-CSF and GM-CSF Recombinant human G-CSF was supplied by Amgen, Thousand Oaks, CA, USA. G-CSF was supplied at a concentration of 0 3 mg ml 1 and was diluted to the required concentration with RPMI GM-CSF was supplied by Immunex Corp., Seattle, WA, USA, and was stored in aliquots of IU ml 1 at 80 o C. Appropriate dilutions were made in RPMI The concentrations of both CSFs studied were shown optimal in prior studies. PMN assay PMN were isolated from heparinized human blood by 6% dextran-70 sedimentation followed by densitygradient centrifugation of the buffy coat diluted 1:1 with RPMI 1640 on Histopaque 1077 (Sigma Chemical Co., St. Louis, MO, USA). The pelleted cells (PMN and some red blood cells, RBC) were collected in 0 85% NH 4 Cl to lyse RBC. PMN were washed, counted, and suspended in CTCM. The nal concentration used for all experiments was cells ml 1, giving an effector:target (E:T) ratio of 10:1 The assay was performed using quadruplicate wells for each condition tested. PMN in a volume of 0 1 ml were added to microtest plate wells containing washed C. neoformans yeast cells with and without. PMN were not added to some sets of quadruplicate wells containing. G-CSF at a concentration of 500 ng ml 1 was added to wells containing both PMN alone and PMN with. After cultures were incubated for 24 h at 37 o C in 5% CO 2 95% air, the contents of the well were removed and added to sterile water, allowing for lysis of PMN and release of yeast cells. Complete removal of well contents was con rmed microscopically. Appropriate dilutions of harvested cultures were made and incubated on blood agar plates (BAP) for 48 h at 35 o C. Colony counts were made and number of colony-forming units (CFU) were determined per culture. Monocyte assay Peripheral blood mononuclear cells (PBMC) were isolated from heparinized human blood by 6% dextran- 70 sedimentation followed by density-gradient centrifu-

3 G-CSF and GM-CSF against Cryptococcus neoformans 23 gation of the buffy coat diluted 1:1 with RPMI 1640 on Histopaque 1077 (Sigma). The PBMC layers were harvested, washed, counted, and suspended in CTCM; 0 1 ml of a suspension of 10 6 cells ml 1 was added to wells of 96-well microtiter plates and incubated for 1 h at 37 o C in 5% CO 2 95% air. The supernatants from these wells were discarded and non-adherent cells (monocytes) were washed away with two washings of RPMI There were approximately 10 20% adherent cells remaining in the wells. Yeast cells in a volume of 0 1 ml were added to wells at a concentration of cells ml 1. The estimated E:T ratio for these experiments was 10 20:1. at appropriate concentrations was added to wells with and without monocytes. GM-CSF at a concentration of 500 U ml 1 was added to wells containing monocytes with and without. Cultures were incubated for 24 h at 37 o C in 5% CO 2 95% air and yeast cells were harvested and counted as detailed in the PMN assay section. Monocyte-derived macrophage assay PBMC were isolated from human donors as described above. 0 1 ml of PBMC at cells ml 1 were put into individual wells of 96-well microtiter plates and incubated in RPMI-1640 with 10% human serum for a total of 5 days. The media was changed after 48 h. The supernatants were then removed and the non-adherent cells were washed away twice with RPMI The remaining adherent cells (approximately well 1 ) were used as monocyte-derived macrophages (MDM) in coculture with C. neoformans yeast cells as described above for the monocyte assay. E:T ratios were approximately 20:1. at appropriate dilutions in 0 1 ml was added to wells containing no MDMs, MDMs with, and MDMs with both and GM-CSF. GM-CSF at a concentration of 500 U ml 1 were added to wells with and without MDM as was done for both PMN and monocyte assays detailed above. Cultures were in incubated for 24 h at 37 o C in 5% CO 2 95% air and yeast cells were harvested and counted as detailed in the PMN assay section. Quantitative analysis Table 1 Inhibition and killing of C. neoformans by voriconazole () in CTCM* (m g ml 1 ) Inhibition of growth was calculated as a percentage of inhibition as compared to the growth of the control. The formula [1 (Experimental CFU/Control CFU)] 100 was used to determine the percentage of inhibition or fungistasis. Killing ability in the assays was calculated by rst determining if the resultant CFU were less than the inoculum and then the formula [1 (Experimental CFU/ Inoculum CFU)] 100 was employed to determine percentage killing. ANOVA with the Bonferroni correction was used for statistical analysis of data with signi cance assumed to be P<0 05. Results % Inhibition, mean SEM Activity of against C. neoformans % Killing, mean SEM Py Ref < < < < < < < 0 01 *Complete tissue culture medium (CTCM) consists of RPMI 1640 with 10% human sera. ycompared with CTCM alone. Table 1 demonstrates the activity of against C. neoformans. As can be seen, concentrations of below 0 05 m g ml 1 exhibited dose-dependent inhibition. Concentrations of 0 05 m g ml 1 of and above demonstrated dose-dependent killing of C. neoformans. The results represent C. neoformans incubated in CTCM, which contains 10% human sera. As such, profound inhibition and killing of C. neoformans by, as seen in Table 1, is in part due to the effects of sera on the inhibition of C. neoformans. Similar experiments performed in RPMI 1640 without the presence of sera (data not shown) also exhibited dose-dependent inhibition, but not killing, of C. neoformans. Effect of G-CSF and GM-CSF on PMN activity with against C. neoformans PMNs alone in 24-h culture with C. neoformans resulted in 61% inhibition of growth as compared with controls, whereas the addition of G-CSF increased the activity to result in 47% killing of C. neoformans (Table 2). As can be seen in Table 2, PMN and collaborated at concentrations of 0 01 and 0 05 m g ml 1 resulting in 51% and 71% killing, respectively. G-CSF also exhibited the ability to enhance the activity of PMNs in combination with (P< 0 01). The addition of G-CSF increased killing in the two concentrations to 92% and 93%, respectively. When GM-CSF was used to enhance the activity of PMNs combined with, similar results were

4 24 Chiller et al. Table 2 Effect of G-CSF on PMN activity against C. neoformans and collaboration with Condition ( m g ml 1 ) % Inhibition, % Killing, mean SD mean SD P* versus reference condition PMN alone PMN (0 01 m g ml 1 ) PMN (0 05 m g ml 1 ) CTCM PMN < < 0 01 PMN G-CSFy < < 0 01 < < 0 01 < 0 05 PMN GM- CSFz < < 0 01 < < 0 01 < 0 05 *P value represents comparison with speci ed condition. yg-csf concentration used was 500 ng ml 1. zgm-csf concentration used was 500 U ml 1. obtained (Table 2). GM-CSF signi cantly increased killing of C. neoformans by PMNs and combined with at concentrations of 0 01 and 0 05 m g ml 1 to 87% and 94%, respectively (Table 2). Effect of GM-CSF on monocyte activity combined with against C. neoformans Monocyte monolayers had signi cant inhibitory activity against C. neoformans, but alone did not kill (Table 3). Monocytes in 24-h culture inhibited cryptococcal growth by 34% (P< 0 01) as compared with controls. at concentrations of 0 01 and 0 05 m g ml 1 demonstrated synergy with monocytes and resulted in killing of C. neoformans at 41% and 58%, respectively. Moreover, Table 3 Condition GM-CSF signi cantly enhanced the activity of monocytes and resulted in 31% killing. The combination of GM-CSF with monocytes and further enhanced the killing of C. neoformans (Table 3). At concentrations of 0 01 and 0 05 m g ml 1 of, GM-CSF enhanced monocytes to kill 62% and 86% of C. neoformans, respectively. Effect of GM-CSF on MDMs combined with against C. neoformans Effect of GM-CSF on monocyte activity against C. neoformans and collaboration with (m g ml 1 ) % Inhibition, % Killing, mean SD mean SD MDMs were also tested for their ability to inhibit the growth of C. neoformans. Alone, as shown in Table 4, MDMs inhibited the growth of C. neoformans by 23%. The addition of GM-CSF to MDMs increased the P* versus reference condition Mono alone Mono (0 01 m g ml 1 ) CTCM Mono < < 0 01 Mono GM-CSFy < < 0 01 < < 0 01 < 0 05 Mono (0 05 m g ml 1 ) *P value represents comparison with speci ed condition. ygm-csf concentration used was 500 U ml 1.

5 G-CSF and GM-CSF against Cryptococcus neoformans 25 Table 4 Effect of GM-CSF on MDM activity against C. neoformans and collaboration with Condition (m g ml 1 ) % Inhibition, % Killing, mean SD mean SD P* vs reference condition MDM alone MDM (0 01 m g ml 1 ) MDM (0 05 m g ml 1 ) CTCM MDM < < 0 01 MDM GM-CSF y < < 0 01 < < 0 01 < 0 01 *P value represents comparison with speci ed condition. ygm-csf concentration used was 500 U ml 1. activity of the cells resulting in the ability to kill C. neoformans. GM-CSF enhanced MDMs to produce 22% killing (Table 4). The addition of to MDM cocultures resulted in killing of C. neoformans (Table 4). at concentrations of 0 01 and 0 05 m g ml 1 collaborated with MDMs and killing at 14% and 34%, respectively. Finally, GM-CSF was able to further enhance the amount of killing done by MDMs combined with. Concentrations of 0 01 and 0 05 m g ml 1 of combined with MDMs showed increased killing with GM-CSF at levels of 61% and 84%, respectively. Discussion In this study, we report the ability of G-CSF and GM- CSF to enhance effector cell activity against C. neoformans and to further collaborate with for enhanced killing. The effect of G-CSF on PMN activity against C. neoformans has been reported with PMNs isolated from HIV patients [6], which demonstrated less cryptococcal activity than PMNs from normal hosts. The authors demonstrated that PMNs from these patients exhibit increased activity against C. neoformans in vitro with the addition of G-CSF. In our study with PMNs from normal hosts, we were able to increase the anticryptococcal activity from inhibition to killing of C. neoformans with the addition of either G-CSF or GM-CSF. This is the rst report on the ability of GM- CSF to enhance the activity of PMNs against C. neoformans in vitro. was able to collaborate with PMNs alone and with PMNs stimulated with either G-CSF or GM-CSF to enhance the ability to kill C. neoformans. Prior studies have demonstrated the ability of monocytes and PMNs to kill C. neoformans in vitro [2]. Furthermore, it has been shown that GM-CSF can increase the activity of monocytes and macrophages against C. neoformans [12,13]. In this study, we used E:T ratios of monocytes and MDMs which only allowed for inhibition, not killing, of C. neoformans, and we were able to enhance cell activity to kill C. neoformans with the addition of GM-CSF. Similarly, was able to interact with the effector cells and kill C. neoformans. also demonstrated the ability to enhance the killing activity of GM-CSF treated cells. The results of this study support a potential advantage in using G-CSF and GM-CSF to treat patients with cryptococcosis. They also support that could have increased ef cacy against C. neoformans in clinical settings where these cytokines are used. Acknowledgements This work was supported in part by a grant from P zer and Co., New York, NY, USA. References 1 Weinberg PB, Becker S, Granger DL, Koren HS. Growth inhibition of Cryptococcus neoformans by human alveolar macrophages. Am Rev Respir Dis 1987; 136: Miller MF, Mitchell TG. Killing of Cryptococcus neoformans strains by human neutrophils and monocytes. Infect Immun 1991; 59: Karaoui, RM, Hall, NK, Larsh, HW. Role of macrophages in immunity and pathogenesis of experimental cryptococcosis induced by the airborne route. II. Phagocytosis and intracellular fate of Cryptococcus neoformans. Mykosen 1977; 20: Levitz SM, Farrell TP, Maziarz RT. Killing of Cryptococcus neoformans by human peripheral blood mononuclear cells stimulated in culture. J Infect Dis 1991; 163: Nassar F, Brummer E, Stevens DA. Effect of in vivo macrophage colony-stimulating factor on fungistasis of bronchoalveolar and peritoneal macrophages against Cryptococcus neoformans. Antimicrob Agents Chemother 1994; 38:

6 26 Chiller et al. 6 Coffey MJ, Phare SM, George S, Peters-Golden M, Kazanjian PH. Granulocyte colony-stimulating factor administration to HIV-infected subjects augments reduced leukotriene synthesis and anticryptococcal activity in neutrophils. J Clin Invest 1998; 102: Marco F, Pfaller MA, Messer SA, Jones RN. Antifungal activity of a new triazole, voriconazole (UK-109,496), compared with three other antifungal agents tested against clinical isolates of lamentous fungi. Med Mycol 1998; 36: Pfaller MA, Zhang J, Messer SA, et al. In vitro activities of voriconazole, uconazole, and itraconazole against 566 clinical isolates of Cryptococcus neoformans from the United States and Africa. Antimicrob Agents Chemother 1999; 43: Schwartz S, Milatovic D, Thiel E. Successful treatment of cerebral aspergillosis with a novel triazole (voriconazole) in a patient with acute leukaemia. Br J Haematol 1997; 97: Vora S, Purimetla N, Brummer E, Stevens DA. Activity of voriconazole, a new triazole, combined with neutrophils or monocytes against Candida albicans: effect of granulocyte colony-stimulating factor and granulocyte-macrophage colonystimulating factor. Antimicrob Agents Chemother 1998; 42: Vora S, Purimetla N, Brummer E, Stevens DA. Activity of voriconazole combined with neutrophils or monocytes against Aspergillus fumigatus: effects of granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor. Antimicrob Agents Chemother 1998; 42: Levitz SM. Activation of human peripheral blood mononuclear cells by interleukin-2 and granulocyte-macrophage colonystimulating factor to inhibit Cryptococcus neoformans. Infect Immun 1991; 59: Chen GH, Curtis JL, Mody CH, et al. Effect of granulocytemacrophage colony-stimulating factor on rat alveolar macrophage anticryptococcal activity in vitro. J Immunol 1994; 152: