Effect of Pichia Anomala Killer Toxin on Candida Albicans

Medical Mycology 1998, 36, 199–204 Accepted 16 January 1998 Effect of Pichia anomala killer toxin on Candida albicans

H. L. MATHEWS,∗ S. CONTI,† L. WITEK-JANUSEK‡ & L. POLONELLI† ∗Departments of Microbiology and Immunology and ‡Maternal Child Health, Loyola University of Chicago, Maywood, IL, USA; and †Istituto di Microbiologia, Universita` degli Studi di Parma, V.le A. Gramsci 14, 43100 Parma, Italy

The effect of a Pichia anomala killer toxin upon a Candida albicans-sensitive strain was studied. Yeast and hyphae, after treatment with the toxin, were less capable of uptaking either [3H]-uridine or [35S]-methionine. In addition, the hyphal form of the fungus appeared to be less capable of DNA synthesis after toxin treatment. No effect of the killer toxin was shown upon a natural resistant mutant of the source strain. These data suggest that, similar to other killer yeast toxins, the toxin of P. anomala can produce a number of quantifiable effects upon sensitive C. albicans cells. Keywords Candida albicans, hyphae, Pichia anomala killer toxin, yeast

Introduction as the incorporation of radioisotopes into the fungus. Although the assays were developed for the measure- Killer yeast toxins are either glycosylated or non-gly- ment of lymphocyte-mediated antifungal activity, other cosylated acidic proteins that are encoded by double- antifungal mediators can also be assayed for their effects stranded DNA or RNA plasmids, or by chromosomal upon C. albicans. In the study reported herein, the genes [1–6]. These killer toxins are produced by a number effect of P. anomala killer toxin on C. albicans was of different fungi and exert their effect via a variety of evaluated by use of these and other previously de- different mechanisms. In the case of the toxins produced veloped procedures [18,19]. by Saccharomyces cerevisiae, the toxins bind to target cell wall receptors and induce target cell membrane permeability leading to cell death [3].The KT28 toxin of Materials and methods S. cerevisiae has been speculated to inhibit DNA syn- Fungal cultures thesis [7]. The toxin of Pichia kluyveri causes target Pichia anomala (ATCC 96603) was used for the pro- cell leakage of potassium ions and ATP, a decrease in duction of the yeast killer toxin. C. albicans strain UP intracellular pH and inhibition of the uptake of amino 10S (sensitive to the activity of the P. anomala killer acids [8,9]. The cell wall receptors for S. cerevisiae toxin toxin), C. albicans strain UP 10R (a toxin resistant have been shown to be b-1-6--glucan as well as a mutant) [20], and C. albicans (ATCC 58716) were used mannoprotein, while the P. anomala toxin receptor is b- throughout this investigation. Cultures were stored at 1-6--glucan [10–12]. The K1 toxin and the P. anomala 25 °C on Sabouraud glucose agar (SGA) (Becton toxin are lethal for sphaeroplasts of Candida albicans Dickinson, Cockeysville, MD, USA). Cells used for [12,13]. Little is known about the nature and mode of experimentation were cultured overnight at 37 °Con action of the P. anomala killer toxin, although the toxin SGA, collected as isolated colonies, and washed once is encoded by a nuclear gene and possesses a large range in Hanks’ balanced salts solution (HBSS). C. albicans of intergeneric activity [14,15]. hyphal forms were obtained by incubation at 37 °Cin Previous work from one of our laboratories has RPMI 1640. Inoculation of 2×105 yeast cells/ml yielded resulted in the development of rapid and reproducible approximately 100% hyphal segments when incubated assays to measure growth inhibition of C. albicans [16, for 2 h at 37 °C. 17]. Fungal growth inhibition was assessed optimally Preparation of P. anomala killer toxin Correspondence: Dr Herbert Mathews, Department of Microbiology and Immunology, Loyola University of Chicago, Maywood, IL, USA. The P. anomala toxin was prepared as described pre- Tel. +708 216 4586; Fax. +708 216 9574; [email protected]. viously and diluted in HBSS at pH 4·0, pH 5·0, pH 6·0

 1998 ISHAM 200 Mathews et al.

Fig. 1 Growth inhibition of C. albicans strain UP 10S by P. anomala killer toxin at varying pH. (a) The effect of the toxin on hyphae at pH 4·0, Β; pH 5·0, Φ; pH 6·0, Α; pH 7·0, Κ. (b) The effect of the toxin on yeast at pH 4·0, Χ; pH 5·0, Ε; pH 6·0, Μ; pH 7·0, Ν. C. albicans growth inhibition was assessed by the incorporation of [3H]-uridine. Data are presented as a percentage of growth control. Mean incorporated DPM for C. albicans hyphae at pH 4·0 was 13 361, at pH 5·0 was 15 909, at pH 6·0 was 17 084 and at pH 7·0 was 43 053 for a representative and individual experiment. Mean incorporated DPM for C. albicans yeast at pH 4·0 was 1546, at pH 5·0 was 2311, at pH 6·0 was 3976 and at pH 7·0 was 5683 for a representative and individual experiment. A pH of 5·0 was used for further experiments. Data are expressed as the mean±SD of three independent experiments.

or pH 7·0 [15]. Serial dilution of a single lot of toxin were obtained by incubation at 37 °Cin5%CO2 for was used for all experiments. 2 h at the indicated pH. Culture supernatants were removed from hyphae at the end of the 2-h incubation period and the P. anomala killer toxin was added at C. albicans growth inhibition the indicated dilutions to each fungal culture for an

The antifungal activity (denoted percentage of growth additional 3-h incubation at 37 °Cin5%CO2 at the control) of the P. anomala killer toxin for C. albicans indicated pH; 0·05 ml of RPMI 1640 containing 1 lCi was determined as described below. Briefly, fungal cells of [3H]-uridine (ICN Radiochemicals, Irvine, CA, USA) used for experimentation were collected from isolated, or 1 lCi of [35S]-methionine (New England Nuclear, overnight SGA colonies, and washed once in HBSS. Boston, MA, USA) was added to individual wells at a Yeast form cells were resuspended to 2×105/ml in pH of 7·0. RPMI 1640 at pH 4·0, pH 5·0, pH 6·0 or pH 7·0. 1×104 C. albicans yeast cells were placed in 96-well, flat- cells were then added to individual wells of 96-well, bottom plates (Corning) in HBSS 15 min prior to the flat-bottom plates (Corning). C. albicans hyphal forms addition of toxin to facilitate the association of the

 1998 ISHAM, Medical Mycology, 36, 199–204 Effect of P. anomala killer toxin 201 yeast with the plastic surface of the assay wells [19]. Statistics The yeasts (1×104) were diluted in HBSS at a pH of 4·0. Comparisons were performed by Student’s t-test ana- After 15 min the culture supernatants were removed. To lysis as described previously [16]. the yeast was added the toxin at the indicated pH for a 3-h incubation at 37 °Cin5%CO2. The toxin was removed and radioisotopes added to individual assay Results wells as described above. Both hyphae and yeasts were Effect of P. anomala killer toxin on C. albicans strain incubated for 1 h at 37 °C, 5% CO2. To each well 25 U lyticase (Sigma Chemical, St Louis, MO, USA) in 50 ll UP 10S at varying pH HBSS was added for 0·5 h at 25 °C. Cells were then In Fig. 1 the effect of the P. anomala toxin on the harvested with a PHD harvester and associated radio- hyphal and yeast stage of C. albicans was determined activity determined with a Beckman LS5801 b liquid at various pH. The toxin was active against both forms scintillation counter. Percentage growth inhibition of C. of the fungus. However, the yeast form of the fungus albicans={[(DPM C. albicans control)−(DPM (toxin was more susceptible to the toxin than the hyphal form, and C. albicans)]/[DPM C. albicans control]}×100 was P<0·05 at a reciprocal dilution of the toxin of 32. calculated as described previously [18]. Data are pre- No other statistical difference was determined between sented as percentage of growth control and calculated experimental groups. Data are presented as percentage as (100−% growth inhibition). All cultures were pre- of growth control as judged by uptake of [3H]-uridine pared at least in triplicate and the mean inhibition of into the two forms of the fungus. those values determined. Effect of P. anomala toxin on C. albicans strain UP Assessment of antifungal activity by DAPI staining 10S uptake of [3H]-uridine and [35S]-methionine A correlation was sought between the treatment of C. The toxin was equally capable of inhibiting the in- albicans hyphae with P. anomala killer toxin and its corporation of [3H]-uridine and [35S]-methionine into effect on hyphae as judged by 4′,6-diamidino-2-pheny- hyphae and yeast (Fig. 2). The antifungal activity was lindole (DAPI) staining [19]. This fluorescent dye binds independent of the radioactive precursor employed. to the AT-rich regions in the minor groove of double- The yeast form of the fungus was more susceptible to stranded DNA; 5×103 C. albicans yeast in 150 llof the toxin than the hyphal form of the fungus, P<0·05 RPMI-1640 were prepared in Lab-Tek 8-well Chamber at a reciprocal dilution of the toxin of 8. No other

Slides (Nunc) for 2 h at 37 °Cin5%CO2 at pH 7·0. statistical difference was determined between ex- The supernatants were removed with a pasteur pipette perimental groups. Similar data were obtained for C. and 50 ll aliquots of P. anomala killer toxin at pH 5·0 albicans strain 58716, but in all cases the effect on was added to individual wells of the chamber slides growth of the fungus was reduced when compared to in a final volume of 150 ll of RPMI-1640. The chamber strain UP 10S. In all cases, the observed antifungal slides were incubated for 3 h at 37 °Cat5%CO2.At effect was not detected above pH 7·0 (data not shown). the end of the incubation, the fluid was removed with No effect of the toxin was observed for C. albicans a pasteur pipette and the individual chambers washed strain UP 10R. 3× with 0·05  phosphate buffered saline (PBS), pH 7·0. The chamber slides were incubated in PBS at Effect of P. anomala toxin on C. albicans strain UP 22 °C for varying periods of time, as indicated in 10S nuclei Fig. 3. PBS was removed and hyphae fixed with 2% paraformaldehyde at 4 °C for 2 h. The hyphae were The above experimental results show that P. anomala washed with PBS 3×; 150 ll of 2·5 lg/ml DAPI dye toxin inhibits the growth of C. albicans as judged by in PBS, pH 7·4, was added to each chamber and the reduced uptake of [3H]-uridine and [35S]-methionine. incubated at room temperature for 45 min, protected In order to assess the effect of P. anomala toxin on from light. The fluid was removed and each chamber the nuclei of C. albicans, the DNA-binding dye 4,6- washed once with PBS and then a coverslip fixed to diamidino-2-phenylindole (DAPI) was used. C. albicans the slide and the hyphae viewed microscopically. The hyphae stained with DAPI have demonstrable nuclei nuclei were scored and the percentage of nucleated within each hyphal segment (untreated) (Fig. 3). When hyphal segments determined per at least 100 hyphal hyphae were treated for 3 h with P. anomala toxin and segments. stained with DAPI, all hyphal segments contained a

 1998 ISHAM, Medical Mycology, 36, 199–204 202 Mathews et al.

Fig. 2 Growth inhibition of C. albicans strain UP 10S by P. anomala toxin at pH 5·0. C. albicans growth inhibition was assessed by the incorporation of [3H]-uridine (Φ)or[35S]-methionine (Κ) into (a) hyphae, or incorporation of [3H]-uridine (Ε)or[35S]-methionine (Ν) into (b) yeast. Data are presented as a percentage of growth control. Data are expressed as the mean±SD of at least two independent experiments. nucleus. However, when hyphae were treated with P. this investigation, P. anomala toxin has been shown to anomala toxin for 3 h, the P. anomala toxin removed, inhibit the growth of C. albicans hyphae and yeast as and the hyphae incubated for additional periods of measured by the inhibited uptake of [3H]-uridine and time, a sequential loss of hyphal nuclei was observed [35S]-methionine [18]. Such measurement is accurate, (P<0·05 at 9 h, and P<0·01 at 21 h) when compared to efficient and a convenient means by which to assess untreated hyphae. No effect of the toxin was observed antifungal activity. against yeast nuclei. The results suggest that the toxin can exert its effect either directly or indirectly upon precursor in- corporation of radioisotopes into both RNA and pro- Discussion tein of C. albicans. A similar method has been described P. anomala killer toxin has been assayed routinely previously for other yeast killer toxins [8,9]. The P. by either survival determination on solid agar or by anomala toxin has been reported to be inactive at a pH measurement of zones of growth inhibition [3,14]. Al- greater than 7·0 [15,20]. In this study an antifungal though suitable for measurement of the effect of killer effect was seen at varying pH, but the most dramatic toxin, these assays can be laborious and involve effect was seen against the yeast at 6·0 and 7·0. The measurement of small zones of growth inhibition. In yeasts in this study were diluted in HBSS at pH 4·0

 1998 ISHAM, Medical Mycology, 36, 199–204 Effect of P. anomala killer toxin 203

in this study reflect a more generalized growth inhibitory effect upon the fungus. These antifungal effects on C. albicans hyphae may not be a single event but rather a combination of antifungal effects that may be po- tentially cytostatic or cytocidal. The effect of P. anomala killer toxin on C. albicans may not only be growth inhibitory as judged by decreased uptake of radio- isotopic precursors for fungal RNA and protein but also may directly affect the DNA of the fungus.

Acknowledgements This work was supported in part by USPHS grant AI- Fig. 3 Effect of P. anomala toxin at pH 5·0 on C. albicans strain UP 10S hyphal nuclei as judged by DAPI staining. Hyphae were 31227 and by the Cancer Federation, H.L.M. L.W.-J. either untreated or treated with P. anomala toxin for 3 h, the toxin was supported by USPHS NR-00085. L.P. was sup- removed and the hyphae incubated for additional periods of time ported by the Italian National Research Council (CNR) at 22 °C for a total of 3, 6, 9 or 21 h. Hyphae were stained with DAPI and scored microscopically with UV illumination for the Target Project ‘Prevention and Control of Disease Fac- presence of nuclei within hyphal segments. Data are expressed as tors’ subproject ‘Etiology of Infectious Diseases’ (con- the mean±SD of three independent experiments. tract no. 95.00793.PF41-115-15895) and by Ministero delle Sanita`, Istituto Superiore di Sanita` VIII Progetto AIDS, Rome, Italy (contract no. 9305/32). and then shifted to a higher pH. Without this dilution in HBSS at pH 4·0, antifungal activity was not observed. References These data are consistent with P. anomala toxin exerting 1 Young TW. Killer yeasts. In: Rose AH, Harrison JS, eds. The an antifungal effect like that observed when C. albicans Yeasts. New York: Academic Press, 1987: 131–164. has been treated with IL-2 activated lymphocytes. 2 Radler F, Herzberger S, Schonig I, Schwarz P. Investigation of a We have shown previously that activated lympho- killer strain of Zygosaccharomyces bailii. J Gen Microbiol 1993; cytes inhibit the capacity of C. albicans hyphae to 139: 495–500. incorporate [3H]-uridine into RNA and [35S]-methionine 3 Tipper DJ, Bostian KA. Double-stranded ribonucleic acid killer systems in yeasts. Microbiol Rev 1984; 48: 125–156. into protein [18]. An additional measure of the anti- 4 Zorg J, Kilian S, Radler F. Killer toxin producing strains of the fungal effect of IL-2-activated lymphocytes was the yeasts Hanseniaspora uvarum and Pichia kluyveri. Arch Microbiol analysis of the capacity of treated hyphal segments to 1988; 149: 261–267. return to the yeast phase of growth [18,20]. The ability 5 Starmer WT, Ganter PF, Aberdeen V, Lachance M-A, Phaff HJ. of hyphal segments to return to the yeast phase of The ecological role of killer yeasts in natural communities of yeasts. Can J Microbiol 1987; 33: 783–796. growth has been reported as an index of the cytocidal 6 Kimura T, Kitamoto N, Matsuoka K, Nakmura K, Iimura Y, capacity of effector cell populations for C. albicans Kito Y. Isolation and nucleotide sequences of the genes encoding hyphae [21]. Herein, P. anomala toxin was incubated killer toxins from Hansenula mrakii and Hansenula saturnus. Gene with preformed hyphae for 3 h and then hyphal seg- 1993; 137: 265–270. ments analysed microscopically for nuclear content by 7 Schmitt MJ, Brendel M, Schwarz R, Radler F. Inhibition of DNA synthesis in Saccharomyces cerevisiae by yeast killer toxin staining with DAPI. A reduced number of nuclei were KT28. J Gen Microbiol 1989; 135: 1529–1535. detected. It appeared that the effect of P. anomala toxin 8 Middlebeck EJ, Crutzen AQH, Vogels GD. Effects of potassium on C. albicans hyphae resulted in an absence of fungal and sodium ions on the killing action of a Pichia kluyveri toxin nuclei. This nuclear effect was not immediate. Rather, in cells of Saccharomyces cerevisiae. Antimicrob Agents Chemother the lack of DAPI staining nuclei required a period of 1980; 14: 519–524. 9 Middlebeck EJ, Stumm C, Vogels GD. Effects of Pichia kluyveri time during which P. anomala killer toxin was not toxin on sensitive cells. Antonie van Leeuwenhoek J Microbiol required. This apparent effect on nuclei was similar to Serol 1980; 46: 205–220. the effect produced in hyphae by antifungal lympho- 10 Nakajima T, Aoyama K, Ichishima E, Matsuda K. Structural cytes [19,22] and by an inhibitor of DNA synthesis, analysis of b-glucans from a killer toxin sensitive yeast, Sac- hydroxyurea [23,24]. charomyces cerevisiae, and a killer resistant mutant. Ag Biol Chem 1989; 47: 2953–2955. It is unlikely that lymphocytes and the P. anomala 11 Schmitt M, Radler F. Mannoprotein of the yeast cell wall as the killer toxin exert their antifungal effect in precisely the primary receptor for the killer toxin of Saccharomyces cerevisiae same manner. Rather, it is likely that the methods used strain 28. J Gen Microbiol 1987; 133: 3347–3354.

 1998 ISHAM, Medical Mycology, 36, 199–204 204 Mathews et al.

12 Sawant AD, Ahearn DG. Involvement of a cell wall receptor in 19 Mathews HL, Kasprowicz DJ, Sto¨ver AG. Direct effect of inter- the mode of action of an anti-Candida toxin of Pichia anomala. leukin-2 activated lymphocytes upon Candida albicans. In: Vanden Antimicrob Agents Chemother 1990; 34: 1331–1335. Bossche H, Stevens DA, Odds FC, eds. Proc Host–Fungus In- 13 Zhu H, Bussey H. The K1 toxin of Saccharomyces cerevisiae kills terplay. Bethesda: National Foundation for Infectious Diseases, spheroplasts of many yeast species. Appl Environ Microbiol 1989; 1997: 149–157. 55: 2105–2107. 20 Conti SC, Cantelli M, Gerloni P, Fisicaro W, Maglaian D, 14 Polonelli L, Morace G. Reevaluation of the yeast killer phe- Bertolotti P, Mozzoni D, Sullivan D, Coleman D, Polonelli L. nomenon. J Clin Microbiol 1986; 24: 866–869. Killer factor interference in mixed opportunistic yeast cultures. Mycopathologia 15 Polonelli L, Morace G. Production and characterization of yeast 1996; 135: 1–8. 21 Hashimoto T. A micro method for the quantitative determination killer toxin monoclonal antibodies. J Clin Microbiol 1987; 25: of the viability of Candida albicans hyphae. J Microbiol Method 460–462. 1983; 1: 89–98. 16 Beno DWA, Mathews HL. Growth inhibition of Candida albicans 22 Beno DWA, Mathews HL. Growth inhibition of Candida albicans by IL-2 induced splenocytes. Infect Immun 1992; 60: 853– by interleukin-2 induced lymph node cells. Cell Immunol 1990; 863. 128: 89–100. 17 Beno DWA, Sto¨ver AG, Mathews HL. Growth inhibition of 23 Soll DR, Stasi M, Bedell G. The regulation of nuclear migration + Candida albicans hyphae by CD8 lymphocytes. J Immunol 1995; and division during pseudo-mycelium outgrowth in the dimorphic 154: 5273–5281. yeast Candida labicans. Exp Cell Res 1978; 116: 207–215. 18 Beno DWA, Mathews HL. Quantitative measurement of lympho- 24 Soll DR, Bedell G, Thiel J, Brummel M. The dependency of cyte mediated growth inhibition of Candida albicans. J Immunol nuclear division on volume in the dimorphic yeast Candida al- Methol 1993; 164: 155–164. bicans. Exp Cell Res 1981; 133: 55–62.

 1998 ISHAM, Medical Mycology, 36, 199–204