WO1989002744A1

WO1989002744A1 - Synergistic antifungal protein and compositions containing same

Info

Publication number: WO1989002744A1
Application number: PCT/US1988/003420
Authority: WO
Inventors: Walden K. Roberts; Claude P. Selitrennikoff; Bridget E. Laue
Original assignee: The University Of Colorado Foundation, Inc.
Priority date: 1987-10-02
Filing date: 1988-10-03
Publication date: 1989-04-06
Also published as: AU2557288A

Abstract

Novel plant proteins (SAFPs) which synergize the activity of antifungal antibiotics are identified. SAFPs are demonstrated to synergize antifungal antibiotics, such as nikkomycins, polyoxins and amphotericins. SAFPs also display antifungal activity against several species of fungi, including strains of Candida, Trichoderma, Neurospora and strains of the plant pathogens Fusarium, Rhizoctonia and Chaetomium. Synergistic antifungal compositions containing SAFP and antifungal antibiotics are provided. In particular, synergistic compositions of corn-SAFP and nikkomycin are found to be effective as antifungal compositions, especially against the opportunistic human pathogen Candida albicans. Method for employing SAFPs and synergistic compositions containing them for the inhibition of fungi are provided. In addition, a method for purifying SAFP from grain meal is provided.

Description

SYNERGISTIC ANTIFUNGAL PROTEIN AND COMPOSITIONS CONTAINING SAME

This invention was made with partial government support under contract number DCB 8500233 awarded by the National Science Foundation. The government has certain rights in this invention.

Field of the Invention

This invention relates to novel antifungal plant proteins which synergize and enhance the activity of antifungal antibiotics which_. are designated synergistic antifungal proteins (SAFPs) , especially corn-SAFP. This invention also relates to antifungal compositions containing SAFP alone or in combination with an antifungal antibiotic, particularly those which affect fungal cell wall synthesis. Most particularly the invention relates to antifungal compositions containing corn-SAFP in combination with nikkomycin. SAFPs are useful for the inhibition of fungal growth. The synergistic antifungal compositions of the present invention are generally useful in. vitro and in vivo for inhibition of fungal growth and for combatting fungal infections. Corn-SAFP/nikkomycin compositions are particularly useful in inhibiting the growth of the opportunistic human pathogen Candida albicans and for combatting candidal infections.

Background of the Invention

There is a significant need for effective antimycotic drugs especially for the treatment of systemic fungal infections which are life-threatening, common complications in immune-compromised patients, see for example Hart et al. (1969) J. Infect. Dis. 12j):169-191. Among the most virulent organisms are strains of the yeast Candida, most particularly strains of Candida albicans. While there are several effective topical agents for treatment of candidiasis, treatment of systemic infection is much more difficult. The drug of choice for systemic infection is amphotericin B, however this drug is highly toxic to the host (see, for example, Medoff and Kobayashi (1980) New Eng. J. Med. 302:145-155) . Antimycotic agents that are more effective and/or less toxic than existing drugs are highly desirable.

Several classes of nucleoside antibiotics, including polyoxins (Hori et al. (1971) Agr. Biol. Chem. 15:1280; Hori et al. (1974) Agr. Biol. Chem. 3_8:699; Sasaki et al. (1968) Ann. Phytopathol. Soc. Japan 34,:272) and nikkomycins (Dahn et al. US patent 4,046,881 and 4,158,608; Zahner et al. US patent 4,287,186; Hagenmaier et al. US patent 4,315,922) have been reported. Polyoxins and nikkomycins are reported to be useful in agriculture against phytopathogenic fungi and insect pests. Early reports indicated that polyoxins were not effective against zoopathogenic fungi, such as Candida albicans (see, for example, Gooday (1977) J. Gen. Microbiol. 99:1; Shenbaga urthi et al.. (1983) J. Med. Chem. 26:1518-1522) . It was believed that the polyoxins were not taken up by target cells. More recently, polyoxins have been reported to inhibit the growth in vitro of certain zoopathogenic fungi including Candida albicans and Cryptococcus neoformans when provided at illimolar concentrations (Becker et al. Antimicro. Agents Chemother. (1983) 23:926- 929 and Mehta et ayL. (1984) Antimicro.^" Agents Chemother. 2_5:373-374) . Nikkomycins X and Z have now also been reported to inhibit growth of Candida albicans in vitro (Yadan et al. (1984) J. Bacteriol. 160:884-888? McCarthy et al. (1985) J. Gen. Microbiol. 13L:775-780) . Polyoxins and nikkomycins are similar in structure and apparently both act as competitive inhibitors of chitin synthetase (Endo et al. (1970) J. Bacteriol. 104.:189-196; Muller et al. (1981) Arch. Microbiol. 130:195-197) . Chitin is an essential component of the cell wall of most fungi. Nikkomycins appear, however, to be more effective (about 100 fold) against certain fungi, for example C. albicans, than polyoxins which is in part due to a higher affinity of nikkomycin for chitin synthetase and more rapid uptake of nikkomycin by C. albicans cells (McCarthy et al. (1985) supra) . The activity of polyoxins and nikkomycins is reported to be inhibited by peptides, such as those present in rich media (Becker et al. , 1983, supra; McCarthy et al. , 1985, supra; Mehta et al. , 1984, supra) . Peptides are believed to inhibit uptake of the antibiotic by target cells. The usefulness of nikkomycins and polyoxins for clinical applications such as in the treatment of systemic fungal infection, where peptide inhibition is likely, is expected to be limited as the concentrations of antibiotic required for effective fungal inhibition are not likely to be achieved in vivo.

Mixtures of antimicrobial agents, particularly mixtures in which the components have different modes of action have been used in antimicrobial compositions to broaden activity spectrum or to minimize the occurrence of resistant strains. Further, certain of these mixtures can display an enhanced antimicrobial activity, greater than the additive activity of the individual components, due to synergy. For example, Gisi et l. (1985) Trans. Br. Mycol. Soc. 85.:299-306 reported that a number of fungicide mixtures displayed synergistic activity against phytopathogenic fungi in field tests. The maximum synergy ratio reported was 7, that is a 7-fold enhancement of activity over the calculated additive effect. Fungicide mixtures can also show antagonism with reduced activity of the combination compared to the individual components. It has recently been reported (Hector and Braun (1986) Antimicro. Agent Chemother. J29:389-394) that mixtures of either nikkomycin Z or nikkomycin X with papulacandin B, an inhibitor of /3-glucan synthesis, display synergistic antifungal activity against Candida albicans. Activity enhancements up to about 10 were reported.

Certain enzymes have also been reported to synergize the effect of antifungal agents. Lysozy e has been reported to synergize the activity of amphotericin B against Candida albicans and Coccidioides immitis (Collins and Pappagianis (1974) Sabouraudia 12:329-340). Natural mixtures of mycolytic enzymes of fungal origin, designated mycolases, were reported to have a synergistic effect on the activity of the antifungal drugs amphotericin B and nystatin (Davies and Pope (1978) Nature 273:235-6; Pope and Davies (1979) Postgraduate Med. J. 5.5:674-676). The in vitro MICs (minimum inhibitory concentrations) of these antifungal drugs were lowered about 5 to 10-fold in combinations with mycolase. In related n vivo experiments in a mouse model, fungal mycolase was reported to enhance the effectiveness of amphotericin B and nystatin against systemic infection of Candida albicans. It was suggested that mycolase, which is believed to be a mixture of carbohydrases, enhances penetration of the antibiotic into fungal cells. Fungal mycolases, alone, were described as very effective at releasing protoplasts from A. fumi atus and C. albicans in vitro and were also reported to have some effect, alone, against systemic fungal infection in the mouse model system. In contrast a mixture of the carbohydrases, chitinase (β-1, 4 N-acetyl-D-glucosaminidase) and laminarinase (jS-1,3(4)-glucanase) while reported to effect protoplast release from Asperσillus fumicratus and Candida albicans did not enhance the effectiveness of amphotericin B and nystatin in vivo. - Recently, in similar in vitro and in vivo experiments with fungal mycolase/ amphotericin B mixtures, only slight enhancement of antifungal activity by a fungal mycolase was reported (Chalkley et al. (1985) Sabouraudia 23.:147-164) . This report suggests that the difference in results may be associated with the lower chitinase or lower 0-1,6-D- glucanase activities in their preparation of mycolase compared to that employed in the previous experiments. The specific protein or proteins in mycolase that effect antibiotic enhancement have not been identified. Some bacterial mycolases have also been reported to affect enhancements (about 2-fold) of the activity of amphotericin B (Oranusi and Trinci (1985) Microbios 4.:17-30). Again, no specific enzyme activity was associated with synergy.

Plants appear to have a variety of mechanisms for protecting themselves against infection by viruses, bacteria, fungi and insects.^' These mechanisms are believed to include the presence of inhibitory substances in plant tissue or plant excretions. Such inhibitory substances may be present constitutively in the plant or induced by infection and may be low molecular weight compounds such as inhibitins or phytoalexins or certain proteins, for example, peroxidases, proteinase inhibitors, chitinases or /3-1,3-glucanases. In most cases, the inhibitory function of these substances have not been demonstrated.

In addition to the specific need for more effective, clinically useful antifungal agents, there is a general need for effective, natural, biodegradable antifungal agents, particularly for use in agriculture against plant pathogenic fungi. Such natural antifungal agents may be considered to be ecologically preferable to chemical fungicides. To be economically useful, especially in agricultural applications, such natural antifungal agents should be available in large amounts from inexpensive sources.

It has been recently reported, Roberts and Selitrennikoff (1986) Biochem. Biophys. Acta 880:161-170, that a class of plant proteins called ribosome-inactivating proteins are effective against fungi, for example, Trichoderma reesei. These authors have also reported the presence in barley, corn, wheat and rye of another class of antifungal proteins, designated AFPs which inhibit growth of some fungi, including T. reesei, in vitro. The present work is an extension of this work with plant antifungal protein.

Summary of the Invention

The present invention is based on the surprising finding that a newly identified class of plant proteins, designated synergizing antifungal proteins or SAFPs, significantly enhance the activity of antifungal agents which are inhibitory to the synthesis of fungal cell walls, including polyoxins, nikkomycins and amphotericins. SAFPs can be isolated from grains or germinating grains, including wheat, rye, barley and corn. The level of SAFP is enhanced in germinating grains, particularly wheat and rye. Corn is preferred as a source of SAFP because active corn- SAFP is readily isolated in partially purified or substantially pure form and in relatively large amounts. Furthermore, corn protein extracts containing corn-SAFP were found to be more stable than similar preparations of rye, wheat or barley, which facilitated isolation and purification of SAFP from corn.

SAFP is itself inhibitory to fungi (including strains of Neurospora and Trichoderma) , yeasts (including strains of Candida) , and plant pathogenic fungi (including strains of Rhizoctonia. Chaetomium and Fusariu ) , and particularly Neurospora crassa and Trichoderma reesei. SAFP is in particular inhibitory.to the growth of N. crassa, T. reesei and Candida albicans, as well as to Fusarium moniliforme, F. proliferatum and F. sacchari.

Antifungal activity of SAFP against Neurospora, Trichoderma _r Rhizoctonia, Fusarium and Chaetomium has been assessed in n vitro agar plate assays. In similar plate assays, SAFP showed no growth inhibition of yeasts, including strains of -Candida . Rhodotorula and Saccharomvces. However, inhibition assays performed in liquid medium demonstrated that SAFP itself also inhibited the growth of yeasts, particularly Candida albicans. The synergizing activity of SAFP has been assessed in ih vitro assays, for example, as the enhancement of the antifungal activity of nikkomycin Z or nikkomycin X against the opportunistic human pathogen Candida albicans. Corn-SAFP is an approximately 19 kd protein, as assessed in SDS-PAGE electrophoresis, which displays in particular synergistic anti-Candida activity and anti-Neurospora activity. Corn- SAFP, in substantially pure form or in partially pure form, also displays antifungal activity against strains of Trichoderma. Candida, Fusarium, Rhizoctonia and Chaetomium. In particular, corn steepwater and protein concentrates thereof, which display a distinct protein band at 19 kd and which display synergistic anti-Candida activity, are also found to be inhibitory to the growth of fungi, including Trichoderma. Candida, Neurospora and the plant pathogens Fusarium. Rhizoctonia and Chaetomium. Corn-SAFP has been isolated in substantially pure form by methods described herein, as demonstrated by the absence of contaminating protein bands in conventional protein gel electrophoresis, as shown in Figure 4. Substantially pure corn-SAFP displays no detectible chitinase activity, mannanase or N-_/3-acetylhexosaminidase activity. Substantially pure corn-SAFP preparations include those in which the 19 kd corn SAFP protein represents about 90% or more of the total protein present in the preparation. In in vitro synergy plate assays, corn-SAFP was found to greatly enhance the anti-candidal activity of nikkomycin X or Z up to about 100 fold, while in liquid culture assays, enhancements of up to 1000 fold were observed. Corn-SAFP also displayed significant enhancement (about 10-fold) of the activity of polyoxin against Candida albicans and also enhanced (about 3-fold) the activity of amphotericin B against this yeast.

The present invention discloses a novel class of plant proteins, SAFPs, found in grains such as corn, wheat, barley and rye, which, in addition to having antifungal activity, enhance the antifungal activity of antimycotics, particularly those which inhibit the synthesis of fungal cell walls. In a specific embodiment, the present invention provides corn-SAFP in substantially pure form, having synergistic antifungal activity and antifungal activity. The invention also provides partially purified corn-SAFP preparations having both synergistic antifungal activity and antifungal activity.

SAFP can be employed as an antifungal agent against strains of fungi including, among others, Neurospora, Trichoderma. Candida, Fusarium, Rhizoctonia and Chaetomium. Fungal growth inhibition can be accomplished by applying SAFP, in substantially pure form, in partially pure form, or in crude extracts, to a fungal habitat. The amount of SAFP that is applied is such that its concentration in the fungal habitat is effective for growth inhibition of that fungus. The amount of SAFP required for fungal growth inhibition depends on the desired application. The amount of SAFP required for use against a particular fungus in a particular habitat can be readily determined employing appropriate in vitro or in vivo assays that are well known in the art, such as those described herein. For example, it was found that the minimum amount of substantially pure corn-SAFP required for inhibition of Neurospora crassa in in vitro hyphal extension inhibition assays was about 0.3 μg protein/disc. In a similar assay it was found that the minimum amount of substantially pure corn-SAFP required to inhibit Trichoderma crassa was about 3 μg protein/disc. In liquid medium, it was found that between about 10 to 30 -μg/ml of substantially pure corn-SAFP was required to inhibit Candida albicans. SAFP can in general be employed in any fungal habitat in which it returns antifungal activity.

The present invention further provides antifungal compositions which contain SAFP in combination with an antifungal antibiotic. SAFP being present in such compositions at a revel sufficient to synergize or enhance the antifungal effect of the antibiotic. The antibiotic being present at a sufficient level that the composition has antifungal activity, i.e. inhibits fungal growth. Synergistic compositions are those in which the MIC of the antibiotic in the composition is lower than the MIC of the antibiotic in the absence of SAFP. The antifungal compositions of the present invention preferably contain corn-SAFP. While, in principle, any antifungal agent particularly those that inhibit fungal wall synthesis such as amphotericin B, polyoxin and nikkomycin are useful in the compositions of the present invention, compositions containing nikkomycins are preferred. Compositions containing nikkomycin X. or nikkomycin Z are more preferred and compositions containing nikkomycin X or nikkomycin Z in combination with corn-SAFP are most preferred.

The amounts of SAFP and a particular antifungal antibiotic that in combination produce a synergistic antifungal composition will vary dependent upon the desired application of the composition. The amounts of SAFP and antibiotic required in a particular application against a particular fungus can be readily determined employing appropriate in vitro or in vivo assays that are well known to the art such as those described herein. For example, it was found that compositions containing about 50 μg/ml partially purified corn-SAFP (fraction CMS) and about .06 μg/ml nikkomycin displayed synergistic antifungal activity, particularly against Candida albicans in plate diffusion disc assays (Table 4) . It was found that concentrations of partially purified corn-SAFP (fraction CMS) of at least about 10 μg/ml in combination with about 0.8 μg/ml nikkomycin retained antifungal activity against Candida albicans as measured in plate diffusion disc assays. It was further demonstrated that concentrations of substantially pure corn-SAFP fractions of about 0.3 μg protein/ml or greater, in combination with concentrations of nikkomycin of about 0.17 μg/ml or greater, inhibited growth of C. albicans in liquid medium.

The synergistic compositions of the present invention are useful in general, as antifungal agents effective against a variety of fungi including both phytopathogenic and zoopathogenic fungi. These synergistic compositions are particularly useful against strains of Candida and Rhodotorula and are most particularly useful against the opportunistic human pathogen Candida albicans. The compositions can in general be employed in any fungal habitat in which SAFP and the antibiotic retain activity.

Brief Description of the Figures

Figure 1 is an elution profile from the initial CM- Sephadex column purification step of SAFP from corn protein extracts. Protein in each 6 ml fraction was quantified by measurement of absorbance at 280 nm. Bound protein was eluted with a linear salt gradient (0.01-0.2 M NaCl) . One minor and two major protein peaks were eluted. The results of anti-Candida synergy assays of nikkomycin Z by individual fractions are given beneath the protein fraction profile. Synergy is quantified as strong inhibition (++) , weak inhibition (+) and no inhibition (-) of Candida albicans in synergy plate assays. Nikkomycin Z synergy was assayed on Candida albicans suspension plates (carrot juice agar medium) by adding 30 μl of a 1:10 dilution of each column fraction with 25 ng of antibiotic to assay discs. Only the third peak contained synergistic activity. Fractions 48-55 contained the majority of the desired activity and were combined for further purification.

Figure 2 shows elution profiles from CM-Sephadex column purification of SAFP from corn protein extracts. A flow rate of 1 ml/ in was employed in these separations. Protein in each 6 ml fraction was quantified by measurement of absorbance at 280 nm. Bound protein was eluted with a linear salt gradient (0.01 - 0.2 M NaCl) . Four peaks were eluted. Figure 2A displays the quantitative results of antifungal assays, while Figure 2B displays the quantitative results of enzyme assays across the four peaks. Absorbance at 280 nm is represented in both A and B by closed circles, solid lines. The results of hyphal extension inhibition of T. reesei (open circles, solid line) , hyphal extension inhibition of N. crassa (closed circles, dashed lines) and synergistic anti-Candida activity (closed circles, dotted line) are presented on panel A. The results of chitinase (closed circles, dotted line) , glucanase (open circles, solid line) and 3-N-acetyl- hexosaminidase (closed circles, dashed line) assays are presented in panel B.

Figure 3 is an elution profile of phenyl Sepharose column chromatograph purification of corn-SAFP. Equivalent samples of protein from peak 3 (Figure 2) were washed through the column in 1 M ammonium sulfate (open circles) and 0.1 M sodium chloride (closed circles) and bound protein was subsequently eluted with 50% ethylene glycol. Protein was quantified by measurement of absorbance at 280 nm.

Figure 4 is a photograph of an SDS-polyacrylamide gel electrophoresis of protein fractions from the CM-Sephadex separation (Figure 2) and phenyl Sepharose separation (Figure 3) . Panel A, lanes 1 to 5, contain approximately 5 μg samples of CM-Sephadex column fractions 28, 35, 43, 48 and 52, respectively. Panel A, lanes 6-8, contain approximately 5 μg samples of combined fractions 42-48 from the CM-Sephadex column, and phenyl Sepharose peaks 1 and 2, respectively. Five-fold higher concentrations of these same samples are contained in Panel A., lanes 9-11. Panel B contains two separate phenyl Sepharose column isolates of peak 1 (lanes 1 and 2) , two isolated of peak 2 (lanes 3 and 4) and a single isolate of peak 3 (lane 5) . Panels A and B also contain molecular weight standards as indicated.

Detailed Description of the Invention

The term synergy as used herein applies to the enhancement of antifungal activity of certain antibiotics by certain plant proteins, SAFPs. Synergy can be quantitatively measured as the lowering of the minimum inhibitory concentration (MIC) of an antibiotic effected by combining it with an SAFP and is specifically measured herein as enhancement of the activity of antimycotics against Candida albicans in in vitro assays. When used herein, the term significant enhancement of antifungal activity refers to enhancements of 10-fold or greater. The MIC is generally defined as the highest dilution (i.e. lowest concentration) of an agent that inhibits growth of a microorganism. In liquid medium, the MIC is usually defined as the lowest concentration of an agent which prevents visible growth of a standard inoculum which is measure by culture turbidity. Inhibition plate assays in which discs impregnated with an antimicrobial agent are placed on microbial lawns can also be used to assess MICs (diffusion disc assays). As described in Example 2, the MIC in disc diffusion assays is defined as the lowest concentration of an agent applied to a disc which gives a measurable zone of growth inhibition of the microbial lawn. MICs are determined empirically and often display strain and media dependence.

The effectiveness of an antibiotic agent in vivo is generally assessed in animal model systems, such as those described in Pope and Davies (1979) supra; and Chalkley et al. (1985) supra. In such experiments, effectiveness is assessed as survival or cure rate. Comparisons of the effectiveness of different antibiotic agents is assessed as increases in survival or cure rates.

The present work is an extension of experiments with antifungal proteins (AFPs) which were isolated from barley, corn and wheat (Roberts and Selitrennikoff (1988) J. Gen. Microbiol. 134:169-176) . These proteins inhibited growth of Trichoderma, Phvcomyces and Alternaria and have been shown to have endochitinase activity. Wheat and barley AFP chitinases did not inhibit growth of Neurospora, in contrast to corn AFP preparations. Growth of the important human pathogen Candida albicans was found to be resistant to inhibition by the AFPs in agar plate assays. AFPs were then assessed to determine if they synergized with antifungal antibiotics to lower the MICs of the antibiotics. Selected results of such experiments are summarized in Table 1. Nikkomycin, a mixture of nikkomycin Z and X, synergized with all AFP preparations, but synergy was particularly dramatic with corn-AFP preparations. Polyoxin synergized significantly with corn and wheat AFP preparations, while modest synergy was observed with combinations of amphotericin and AFP preparations from barley and corn. In contrast, no synergy was observed with papulocandin and AFP preparations. Wheat and barley AFPs (Table 1) were purified to homogeneity. The corn-AFP preparation (Table 1) when flashed through a Sephadex column was shown to contain multiple protein peaks (Figure 1).

Corn-AFP preparations were found to contain a mixture of several proteins. Using synergy with nikkomycin to inhibit the growth of C. albicans as an activity assay, the synergizing activity in corn-AFP preparations was found to reside in a single protein fraction from CM-Sephadex column chromatography, see Figure 1. Further purification of this fraction using conventional hydrophobic column chromatography with phenyl-Sepharose resulted in the isolation of an approximately 19 kd protein. The 19 kd protein which effected strong enhancement of nikkomycin activity was designated a corn-SAFP.

Table 1: Effect of AFP preparations on Antibiotic MIC against Candida albicans

MIC against Candida albicans (μg/disc)

AFP^a Nikkomycin Papulocandin Polyoxin B Amphotericin B

None 0.17 0.5 50.0 5.0

Barley 0.05 0.5 (3x)^b

Blue corn <0.0017 0.5 (>100x)

Yellow corn <0.0017 0.5 (>100x)

Wheat 0.05 0.5 (3x)

^a AFP was supplied at 15 μg protein/disc. AFP fractions were prepared as described in Roberts and Selitrennikoff (1987) J". Gen. Microbiol., supr . b The number in parentheses refers to the fold reduction in MIC Since a significant loss in specific synergizing activity was observed in the conventional phenyl-Sepharose chromatography step, efforts were made to improve the purification of corn-SAFP. Improved purification of corn- SAFP was obtained by carrying out the CM-Sephadex chromatography at a slower flow rate than had been employed in previous separations and more importantly, by employing a novel phenyl-Sepharose chromatographic procedure. Slower elution in the CM-Sephadex step resulted in four distinct protein peaks (Figure 2) rather than the three peaks observed previously (Figure 1) . Synergistic anti-Candida activity was found only in peak 3. Anti-Neurospora activity was also confined to peak 3, while • anti- Trichoderma activity was observed in all peak fractions. All four peaks were also assayed for chitinase, glucanase and β-N-acetylhexσsaminidase activity. None of these enzyme activities coincided with the anti-Neurospora or synergistic anti-Candida activity of peak 3.

Corn-SAFP was then further purified employing a novel method of hydrophobic column chromatography. Fractions from the CM-Sephadex column that contained synergistic anti-Candida activity were combined and subjected to phenyl-Sepharose column chromatography. This separation was carried out by loading the column at a lower salt concentration than is typically employed in order to reduce the hydrophobic interactions between the proteins and the column. Bound protein was then eluted with 50% ethylene glycol. This procedure resulted in the profile of Figure 3 containing three bands, one which passed directly through the column (1) , a second which was somewhat retarded (2) , and a third smaller band which was eluted with 50% ethylene glycol (3). SDS-PAGE electrophoresis (Figure 4) demonstrated that peak 2 from the low salt phenyl-Sepharose separation contained apparently homogeneous 19 kd protein (corn-SAFP) . This peak 2 was also demonstrated (see Table 2) to contain all synergistic antifungal activity as well as all anti-Neurospora activity. Peak 2 also contained anti-Trichoderma activity.

Table 2: Antifungal and Enzymatic Activities of Proteins Purified by Phenyl Sepharose Chromatography

Fraction Chitinase Glucanase Anti-Trichoderma Anti-Neurospora Anti-Candi Assayed Activity³ Activity*⁵ Activity⁰ Activity⁰ Activity

0.9

>5 >10

0.3 0.2

NA^e

^a Reducing sugar released after incubation at 37° for 4 h (μ moles glucose/mg protein). ^b Reducing sugar released after incubation at 37° for 20 min. (μ moles glucos protein) .

° Minimum amount of protein required to inhibit fungal growth (μg protein/disc) in the presence of sub-inhibitory levels of nikkomycin.

Protein peaks from chromatography in 0.1 M NaCl (Fig. 2) NA = No activity detected.

The association of synergistic anti-Candida activity with the 19 kd corn protein was confirmed by an experiment in which substantially pure 19 kd corn-SAFP of peak 2 (Figure 3) was subjected to electrophoresis at pH 6.0 in a non-denaturing acrylamide gel. The protein from this gel was allowed to diffuse into an agar plate containing freshly seeded C.. albicans and sub-inhibitory concentrations of nikkomycin. A strong zone of growth inhibitor was observed only at a position in the agar which coincided with the 19 kd corn-SAFP band.

Earlier experiments demonstrated that the 19 kd SAFP could be extracted form cornmeal using 0.05 M acetic acid, and all was stable at moderate temperatures. Thus, it seemed possible that corn-SAFP might survive the low pH and elevated temperatures of the corn steeping process and be present in an active form in steepwater. Accordingly, steepwater samples were obtained from the Adolph Coors corn refining plant, Johnstown, Colorado, and analyzed for antifungal activity. The 35 hour light steepwater sample was effective at inhibiting growth of Trichoderma reesei and Candida albicans in the presence of sub-inhibitory levels of nikkomycin. This activity was lost following heating at 90°C for 15 minutes. Analysis of the proteins in steepwater by SDS-PAGE showed a broad protein smear accompanied by a distinct protein band at 19 kd. The synergistic anti-Candida activity of the extract indicated that the 19 kd SAFP was intact and active in steepwater.

Additional experiments showed that the synergistic antifungal and antifungal activity in steepwater could be precipitated using ammonium sulfate, ethanol, or acetone. Moreover, testing these concentrated preparations on a number of fungal plant pathogens showed that they inhibited growth on agar of pathogenic strains of Trichoderm , Rhizoctonia, and Fusarium.

Efforts were then made to identify SAFPs in sources other than corn. Seeds are known to synthesize large amounts of new enzymes (e.g., glucanases) on germination. Accordingly, wheat and rye were allowed to germinate for three days, after which protein extracts were prepared as in AFP preparations. These protein extracts contained high SAFP activity and were found to lower the MIC of nikkomycin against C. albicans by about 100-fold. The wheat and rye SAFPs could be partially purified by the same procedure used for corn-SAFP. However, the wheat and rye SAFP preparations, in contrast to preparations from corn, lost activity after several days storage at 4°C and have not as yet been further characterized..

Chitinase and glucanase preparations from several other sources were also tested in the synergy assay. No synergy was found with chitinases from Serratia marcescens, Pseudomonas stuzeri. or Streptomyces griseus or in glucanase preparations from Penicilliu or mollusk. Significant synergy was observed, however, with a partially purified glucanase preparation from the fungus Rhizopus and in commercial bacterial (Arthrobacter luteus) enzyme mixture containing both chitinase and glucanase called Zymolase (available from Sigma Chemical Co., St. Louis, MO) . The nature of the synergizing enzymes in these preparations has not been identified, and it is not known whether they act by a mechanism that is similar to plant SAFPs. The synergizing activity in these preparations may be due to minor components in the mixtures.

The anti-Candida synergy that is observed with the SAFP/antibiotic compositions of the present invention is surprising, since it is not predictable that a particular combination of two antimicrobial agents, even those which have different modes of action, will be synergistic.

The very strong synergy observed in the corn- SAFP/nikkomycin composition of the present invention against Candida strains was also surprising. Typically enhancements due to synergy are observed to be in the range of 10 fold or less. For the corn-SAFP/nikkomycin compositions, the MIC of nikkomycin was lower by up to 100 fold in plate assays. In similar inhibition assays of Candida albicans done in liquid media, the MIC of nikkomycin was also lowered by up to about 100 fold (Table 3).

Table 3: Growth Inhibition of C. albicans in Liquid Culture

Corn SAFP in Wells (μg/ml)^A

Nikkomycin in Wells (μg/ml) 30 100 300

0

0.017

0.050

0.17

0.5

1.7

5

17

A Each well contained 150 μl of 2% carrot juice inoculated with C. albicans suspension to give an absorbance at 630 nm of 0.005. Fungal growth was scored as +++, ++, + or no growth (-) after visual inspection for turbidity.

Corn-SAFP/nikkomycin compositions were found to be effective against several strains of C. albicans which varied in their sensitivity toward nikkomycin (Table 4) . In each case, corn-SAFP significantly synergized the effect of nikkomycin and lowered the MIC of nikkomycin in compositions by about 33 to 100 fold. Corn-SAFP/ nikkomycin compositions were also found to be effective against Candida albicans on poor or rich medium. Corn-SAFP synergized nikkomycin activity under conditions (i.e., rich media) when the activity of nikkomycin is attenuated by peptide inhibition.

The mechanism by which SAFP synergizes the action of polyoxins, nikkomycins and amphotericins is not known. One possible mechanism is that SAFP acts to increase penetration of the antibiotics into the target fungi. This could occur as the result of degradation or per eabilization of the fungal cell wall by SAFP. Fungal cell walls are composed of chitin, glucans with 0-1,3 or β- 1,6- linkages and mannans with α-1,6, α-1,2 or α-1,3- linkages. It has been demonstrated, however, that corn- SAFP does not have chitinase, glucanase or mannanase activity. Characterization of the specific enzyme activity of SAFP is not necessary for the practice of the present invention. Table 4 : Comparison of Growth Inhibition of Various strains of Candida albicans on Different Growth Media

MIC Nikkomycin Z¹

Candida Carrot Juice Agar Nutrient Agar strain Nikk Nikk/ Nikk Nikk/

Corn-SAFP² Corn-SAFP²

I³ 170 1.7 II 170 1.7 III 50 0.5 IV 17 0.5

Minimum Inhibitory Concentration (MIC) of Nikkomycin Z in units of ng/disc. All plates were incubated for 16 h at 37°C.

In assays of Nikk/Corn-SAFP, the concentration of Corn- SAFP was 15 μg/disc. Corn-SAFP was partially purified through the CM-Sephadex step (fraction CMS) .

I is laboratory strain B-311 and strains II-IV are separate clinical isolates.

The following examples are intended for illustrative purposes only and are not intended to limit the scope of the invention.

Example 1; Isolation and Purification of Synergistic Antifungal Protein (SAFP) from Corn

Antifungal protein (AFP) extracts, containing SAFP and other antifungal proteins, was prepared from corn using methods similar to those described in Roberts and

Selitrennikoff, 1986 (supra) . The initial protein extraction was carried out at 4°C. Purified fractions could be stored at -70°C essentially without loss of activity.

Corn meal was obtained either from the refrigerator section of health food stores or by grinding dried corn finely in a coffee grinder. AFP could be extracted from corn meal using either acidic or neutral pH buffers (pH ranging from about

4.0 to 7.0) In a typical extraction, corn meal (2 lbs) was stirred for X h at 4°C in 2 1 of 50 mM NaCl/2.5 mM sodium

^" phosphate (pH 7.0)/5.0 mM EDTA. The resulting suspension was centrifuged (10,000 x g, 20 min, 4°C) and the supernatant protein extract was saved.

Ammonium sulfate was slowly added with stirring to the corn protein extract, to obtain a 30% saturated solution which was left overnight at 4°C. The solution was then centrifuged (10,000 x g, 20 min) and the supernatant .was made 55% saturated with ammonium sulfate. The solution was left 20 min at 4°C and centrifuged (10,000 x g, 20 min) . Precipitate pellets from both .ammonium sulfate fractionations were saved and resuspended in 80 ml of 10 mM NaCl/5.0 mM sodium phosphate (pH 7.0)/1.0 mM EDTA resuspension buffer. The resuspended 30%-55% protein fraction was then clarified by centrifugation and dialyzed (16 h, 4°C) against 2 x 1 1 of resuspension buffer.

The dialyzed solution was again clarified by centrifugation prior to further purification by chromatography on carboxymethyl-Sephadex (CM-Sephadex) . Sephadex column chromatography was carried out at 23°C. The dialyzed protein fraction, was passed through a CM-Sephadex column (C-50-120, Sigma Chemical Co., St. Louis, MO) equilibrated with resuspension buffer. The column was prepared by soaking 2 g -of CM-Sephadex in resuspension buffer and adding the slurry to a column to form approximately 60 ml of packed gel volume. The dialyzed protein fraction was washed through the column with buffer until the absorbance of the effluent at 280 nm had fallen to approximately 0.2; typically this required washing with about 400 ml of buffer. In a typical preparation, about 1700 g of protein was added to the column, of which 2/3 washed through and 1/3 remained bound. All of the AFP activity remained bound to the column.

Antifungal protein was then eluted from the Sephadex column using a linear salt gradient prepared conventionally by running 220 ml of 200 mM NaCl/ 5 mM sodium phosphate (pH 7.0)/I mM EDTA into 220 ml of resuspension buffer. Fractions (6 ml) were collected at 3-minute intervals. This gradient eluted three protein peaks (as assayed by absorbance at 280 nm) as shown in Figure 1; one minor component eluting early and two major components eluting later. The latest eluting peak contained all of the SAFP activity. Fractions containing the highest synergistic antifungal activity, fractions 48-55 (Figure 1) , were collected, and concentrated 3-fold by ultra-filtration (Amicon YM-10 filter) . This fraction, designated fraction CMS, retained antifungal and synergistic antifungal properties and was employed in the in vitro anti-Candida plate assays presented in Table 4 (infra) .

Further purification of fraction CMS was attempted employing phenyl-Sepharose column chromatography employing 1 M (NH₄)2S0 at 4⁰C. Two protein peaks (assayed by absorbance at 280 nm) eluted from the column, the first after about 8 ml of buffer, the second after about 25 ml of buffer. Synergizing activity was found only in the second protein peak. A significant loss (80%) in specific synergizing activity was observed on this step of purification. The reason for this loss was not established.

Fraction CMS retained antifungal activity against T. reesei, but did not inhibit C. albicans in agar plate assays. This fraction was found to contain all synergistic antifungal activity against C. albicans. This fraction as assayed by SDS-PAGE was found to contain several protein peaks, including a peak at about 19 kd.

The SAFP fraction obtained by conventionally run phenyl-Sepharose chromatography retained some activity against T. reesei as well as synergistic anti-Candida activity. This fraction appeared by SDS-PAGE to contain a single protein band at a molecular weight of approximately 19 kd. When electrophoresis of this fraction was run using high protein loading (30-50 μg protein/lane) , it was found that the fraction also contained several higher molecular weight bands. These other bands were estimated to represent about 10% of the protein of the fraction. In subsequent enzyme assays this fraction was found also to contain chitinase activity.

A second procedure was found to result in improved purification of 19 kd corn-SAFP. Ammonium sulfate fractionation of corn protein extract was performed as described above. The dialyzed 30%-55% fraction was subjected to CM-Sephadex chromatography, essentially as described above. However, the chromatography was carried out at a slower flow rate (1 ml/min) , which resulted in the elution of four distinct peaks (Fig. 2) . Synergistic anti- Candida activity was confined to peak 3. This peak was also found to contain growth inhibitory activity against Neurospora crassa. Anti-Neurospora activity was found only in corn AFP preparations, not in AFP preparations of wheat and barley. Anti-Trichoderma activity was found in all four peaks. Chitinase, glucanase (01,3- and 01,6-) and 0- N-acetylhexosaminidase activities were also assayed across the four peaks. Chitinase was found in all four peaks. A single peak of glucanase activity at fraction 47 and a single peak of 0-N-acetyl hexosaminidase at fraction 40 were detected. Anti-Neurospora and synergistic anti- Candida activity peaked at fraction 44. These antifungal activities did not coincide with any of the enzyme activities tested.

The fractions associated with synergistic anti-Candida activity (42-48) were combined, concentrated and subjected to further purification. Hydrophobic column chromatography employing phenyl-Sepharose (CL-4B, Sigma Chemical Co., St. Louis, Missouri) was used. The sample was loaded onto the column at a low salt concentration employing 0.1 M NaCl to reduce the hydrophobic interaction between proteins and the column materials. The elution profile is shown in Figure 3 (closed circles) , compared to the elution profile when higher salt concentrations (1 M ( H4)₂S0₄) were employed. A large protein fraction passed directly through the column (peak 1) , followed by another large fraction whose passage was retarded (peak 2). A third smaller fraction (peak 3) was then eluted with 50% ethylene glycol. The three peak fractions from phenyl-Sepharose chromatography were assayed for enzyme activities and antifungal activities as shown in Table 2. Peak 1 contained most of the chitinase activity, a small amount of glucanase activity, and most of the anti-Trichoder a activity. No anti-Neurospora or synergistic anti-Candida activity was found in peak 1. Peak 3 contained most of the glucanase activity and no other detectible enzymatic or antifungal activity. Peak 2 contained all of the anti- Neurospora and synergistic anti-Candida activity and a smaller activity against Trichoderma. Peak 2 contained no detectible chitinase activity and a very small amount of glucanase.

The purity of the various fractions from the CM- Sephadex and phenyl-Sepharose column separations were assessed by SDS-PAGE (Figure 4) . Electrophoresis of the CM-Sephadex fractions 28, 35, 43, 48 and 52 (Figure 4A, lanes 1-5 respectively) showed different protein species in the fractions as expected. Combined fractions 42-48, as well as the phenyl-Sepharose fractions peak 1 and peak 2, were analyzed at low loading (about 5 μg protein/lane, Figure 2A, lanes 6-8) and at 5-fold higher loading (lanes 9-11) . In addition, two separate phenyl-Sepharose chromatography isolates of peak 1 (Figure 2B, lanes 1 and 2) and peak 2 (Figure 2B, lanes 3 and 4) and a peak 3 fraction (lane 5) were analyzed. It is apparent that peak 1 contains multiple protein species present in fractions 42-48, peak 2 contains an apparently homogenous 19 kd protein, corn-SAFP and peak 3 contains two 30 kd protein species and a small amount of the 19 kd protein. It is estimated that 19 kd corn-SAFP isolated in peak 2 of the phenyl-Sepharose elution is about 95-98% pure.

Example 2: In vitro Assays of Antifungal Activity and Synergy

Antifungal activity of protein extracts and fractions was assayed by inhibition of hyphal extension of Trichoderma reesei on agar plates. These assays were performed as described in Roberts and Selitrennikoff, 1986 (supra) in 100 x 15 mm petri plates containing 10 ml of carrot juice agar medium. Five 0.25 inch diameter sterile blank paper discs were placed firmly on the agar, one in the center of a plate and four others at a distance of 1.2 cm around the central disc. Samples to be tested for antifungal activity were diluted in buffered saline, 130 mM NaCl/10 mM sodium phosphate (pH 7.4) , and 25 μl portions were added to each of the 4 peripheral discs. Control plates were prepared by addition of 25 μl of buffered saline to each of the peripheral discs. The concentration of AFP or SAFP fractions added to discs was measured in μg total protein of the fraction. Conidia from T. reesei, for example ATCC culture 13631, grown on a suitable agar medium were suspended in 1 ml of buffered saline, agitated vigorously to form a slightly turbid suspension and 25 μl of the suspension was added to the central disc of assay and control plates. Plates were incubated at 23°C for approximately 72 h until mycelial growth from the central disc had enveloped peripheral discs of the control plates. The formation of crescents of inhibition around sample discs indicated that the sample contained effective concentrations of antifungal agent. Inhibition assays using Neurospora crassa (for example, 74-OR8-la, Fungal Genetic Stock Center, Hu boldt State College, Arcata, California) , Phvcomyces blakesleeanus (for example ATCC 8743a) and Alternaria alternaria (for example ATCC 16086) were performed in a similar manner except that growth conditions of the fungi were altered appropriately and Phvcomyces was assayed on potato-dextrose agar medium. All of the grain AFPs acting alone exhibited antifungal activity toward Trichoderma. Phyco vces. and Alternaria. Wheat and barley AFPs did not inhibit growth of Neurospora.

Inhibition assays of Candida albicans and other yeasts, including Saccharomyces and Rhodotorula, were performed using a modified disc assay in which the organism was suspended in carrot juice agar medium and potential inhibitors were added to blank paper discs which were placed on the agar surface. Candida albicans cultures, for example, strain B-311 (ATCC 32354) , were prepared by inoculating 5 ml of sterile liquid medium (1% glucose, 0.5% yeast extract) with a loopful of Candida and incubating the culture overnight at 37°C without shaking. Candida suspension plates were prepared by adding 1 ml of the overnight culture of Candida to 100 ml of liquid carrot juice agar at 45°C, mixing and dispensing 10 ml of the agar into petri plates. After the Candida agar suspensions had solidified, five blank paper discs were placed as described above on each plate. Sample and control solutions (30 μl) were added to discs. Plates were incubated at 37°C overnight and examined for zones of growth inhibition.

In plate assays, fungal inhibition was quantified by measuring the highest dilution (i.e., lowest total protein concentration) of a sample that caused a detectible zone of inhibition around an assay disc. None of the grain SAFPs alone inhibited growth of Candida, Saccharomyces or Rhodotorula. Inhibition of Candida and Rhodotorula in plate assays was observed only when SAFPs were assayed in the presence of sub-inhibitory concentrations of antifungal antibiotics, especially nikkomycin.

Antibiotic synergy plate assays were used to assay for corn-SAFP activity. Inhibition by known antifungal antibiotics alone was assayed on plates and quantified, as described above, by adding 30 μl of a buffered saline solution of known concentration of antibiotic to each assay disc. In inhibition plate assays, the approximate MIC (minimum inhibitory concentration) of an antibiotic was defined as the lowest concentration that caused a detectible zone of growth inhibition. Synergy of inhibition of known antifungal antibiotics by protein fractions, extracts or solutions was assayed by adding 30 μl of a 1:1 admixture of antibiotic solution and protein solution to assay discs. Synergy was quantified by comparison of the MIC of the antibiotic alone to the MIC of the antibiotic/protein composition. Since none of the AFPs or SAFPs alone inhibits growth of C. albicans on plates, any decrease in antibiotic MIC was scored as synergy. A synergy ratio (antibiotic MIC/antibiotic/protein combination MIC) can be employed for comparisons.

Liquid culture assays were also employed to assess antifungal and synergistic antifungal activity of SAFP, particularly against Candida albicans. Liquid culture assays were performed by inoculating the organism into wells of a 96-well tissue culture plate containing 150 μl of medium (2% carrot juice medium was employed for assays with C. albicans) in each well. Varying concentrations of nikkomycin, SAFP and mixtures of the two agents were introduced into the wells. Plates were incubated in static culture at 37°C for 24 hours, and wells were then examined visually for culture turbidity. Results of an experiment assessing corn-SAFP/nikkomycin activity against C. albicans are shown in Table 3, where growth was scored as +++, ++, + or no growth (-) . In this experiment, the initial C. albicans inoculum resulted in an absorbance reading (650 nm) of about 0.005.

Nikkomycin alone inhibited growth of C. albicans at concentrations greater than 5 μg/ml. Corn-SAFP alone was also found to inhibit growth of C. albicans. This result was surprising in light of disc diffusion assays which showed no C. albicans inhibition with SAFP alone. The reason for these differing results is not known. In any event, corn-SAFP displays synergizing activity with nikkomycin in liquid assays. The MIC of nikkomycin was reduced from about 17 to about 0.17 μg/ml by the addition of 0.3 μg/ml of corn-SAFP. Corn-SAFP employed in these experiments was that purified by phenyl-Sepharose chromatography (peak 2, Figure 3).

Protein extracts, fractions and solutions were quantified for total protein μg/ml using the Bradford dye- binding method.

Carrot juice medium was prepared by first autoclaving 20 g of carrot slices in 180 ml water. The resulting carrot juice was then diluted 1:9 (v/v) with water, agar (2% w/v) was added and the mixture was again autoclaved. Assays were also performed using a richer nutrient broth agar medium Partially purified nikkomycin, which is a combination of nikkomycin X and Z, approximately 70% pure, was obtained as a gift from Bayer A.-G. A formulation that includes nikkomycin X and Z is commercially available in Europe as an agricultural fungicide. Nikkomycin Z and nikkomycin X can be purified from this mixture by known methods (Zahner et al. (1981) U.S. Patent 4,287,186). Nikkomycin Z is also commercially available (Calbiochem, San Diego, CA, cat. no. 481995) . Amphotericin B was obtained from commercial sources (Sigma, St. Louis, MO) and papulacandin B, approximately 80% pure, was a gift from Ciba-Geigy Corp. (Basel, Switzerland) . Polyoxin B was purified from polyoxin AL wettable. powder as described in Selitrennikoff (1982) Neurospora Newsletter, no. 29, p.27.

Example 3: Chemical and Biological Properties of Corn SAFP The grain SAFPs are all highly basic proteins as evidenced by their strong binding to CM-Sephadex.

Corn-SAFP partially purified by CM-Sephadex (fraction CMS) displayed both chitinase and 0-1,3 glucanase activity in addition to antifungal activity against T. reesei and N. crassa, and synergistic activity in combination with antifungal antibiotics, especially nikkomycin, against Candida albicans. Corn-SAFP purified by phenyl Sepharose retained both antifungal activity against T. reesei and N. crassa. and antifungal antibiotic synergy against Candida albicans. Phenyl-Sepharose-purified corn-SAFP displayed no chitinase, mannanase or 0-N-acetylhexosaminidase activity, and little or no glucanase activity.

Chitinase, 0-1,3 glucanase, 0-1,6 glucanase, mannanase and ø-N-acetylhexosaminidase activities were assayed by measuring the increase of reducing sugar, analogous to the procedure of Dubois et al. (1956) Anal. Chem. 28:350-356.

Example 4: Comparison of corn-SAFP/nikkomvcin anti- Candida synergy on different growth media

Synergy assays were performed as described above with

Candida. except that assays were also performed on a rich nutrient broth agar medium. Nutrient agar assay plates were prepared as above, substituting a commercial nutrient agar medium for carrot juice agar. Incubation times were modified appropriately.

In the growth medium comparison, relative inhibition by nikkomycin Z and corn-SAFP/nikkomycin Z mixtures was assayed. In corn-SAFP/nikkomycin Z mixtures, an excess of SAFP (15 μg protein) as fraction CMS was added to each assay disc and the concentration of nikkomycin Z was varied. . The MIC of nikkomycin Z in the presence and absence of SAFP was determined as the lowest concentration of the antibiotic that affects measurable growth inhibition of Candida albicans. As shown in Table 4, the MIC of nikkomycin Z against Candida -albicans grown on nutrient agar was found to be about 9 fold higher than the MIC against C. albicans grown on the less-rich carrot juice medium. It is believed that the higher MIC on rich medium is due to the presence of inhibitory levels of peptides in the medium. Corn-SAFP was found to lower the MIC of nikkomycin Z on both rich and poor media. Interestingly, in most cases the nikkomycin Z MIC was lowered about 100 fold in the presence of SAFP on both media. The minimum amount of partially purified corn-SAFP (fraction CMS) required to synergize with nikkomycin Z (25 ng/disc or about 0.8 μg/ml) was approximately 0.3 μg protein per disc (about 10 μg/ml) for assays carried out in both rich and poor media.

Example 5: Relative sensitivity of Candida albicans strains to Nikkomvcin/corn-SAFP compositions

Several recent clinical isolates of Candida albicans were obtained from Dr. B. Reller, Department of Medicine,

University Hospital, Denver, Colorado. The sensitivity of the clinical isolates to the synergistic nikkomycin/corn-

SAFP composition was assayed and compared to that of the laboratory isolate used in the initial assays. Assays were performed as described above, employing Nikkomycin Z

(Calbiochem) and purified corn-SAFP. Assays were done on carrot juice agar as well as on nutrient broth agar plates.

The results are presented in Table 4. Nikkomycin Z MICs were determined alone and in the presence of an excess of SAFP (15μg protein/disc) provided as fraction CMS. There was wide variation in strain sensitivity to both nikkomycin Z alone and SAFP/nikkomycin Z mixtures. In all cases, the MIC of nikkomycin was lowered in the presence of SAFP. SAFP synergy was about as effective on poor medium as on rich medium.

Claims

We claim:

1. Corn-synergistic antifungal protein in substantially pure form which has a molecular weight of approximately 19 kd and which significantly enhances the antifungal activity of nikkomycin against strains of Candida albicans in in vitro assays.

2. The antifungal protein of claim 1 which also displays growth inhibition of strains of the fungus Neurospora in in vitro assays.

3. A method of inhibiting the growth of a fungus which comprises applying to said fungus or a habitat of said fungus the synergistic antifungal protein of claim 1.

4. The method of claim 3 wherein said fungus is a strain selected from the group consisting of strains of Candida, Trichoderma Neurospora , Rhizoctonia, Fusarium and Chaetomium.

5. The method of claim 3 wherein said fungus is. a plant pathogen.

6. The method of claim 3 wherein said synergistic antifungal protein is comprised in corn steepwater or a protein concentrate thereof.

7. A synergistic antifungal composition which comprises an antifungal antibiotic and a synergistic antifungal protein said synergistic antifungal protein being present in sufficient amount to synergize the antifungal effect of said antifungal antibiotic, said synergistic antifungal protein being a protein isolated from corn, said antifungal antibiotic being selected from the group of antifungal antibiotics which comprises nikkomycins, polyoxins and amphotericins.

8. The synergistic antifungal composition of claim 7 that inhibits growth of strains of the species Candida.

9. The synergistic antifungal composition of claim 8 that inhibits growth of strains of Candida albicans.

10. The synergistic antifungal composition of claim 7 that inhibits growth of strains of the species Rhodotorula.

11. The synergistic antifungal composition of claim 7 wherein said antifungal antibiotic is a nikkomycin.

12. The synergistic antifungal composition of claim 11 wherein said nikkomycin is selected from the group consisting of nikkomycin X and nikkomycin Z.

13. The synergistic antifungal composition of claim 12 wherein said nikkomycin is present at a concentration greater than or equal to about 0.03 times the MIC of said nikkomycin alone.

14. The synergistic antifungal composition of claim 12 wherein said nikkomycin is present at a concentration greater than or equal to about 0.01 times the MIC of said nikkomycin alone.

15. The synergistic antifungal composition of claim 7 wherein said corn-SAFP migrates as *a single protein band on SDS-PAGE and has a molecule weight of about 19 kD.

16. The synergistic antifungal composition of claim 7 in which said synergistic antifungal protein is comprised in a partially purified CM-Sephadex chro atograph fraction.

17. The synergistic antifungal composition of claim 16 wherein said synergistic antifungal protein is comprised in protein fraction CMS.

18. The synergistic antifungal composition of claim 17 wherein corn protein fraction CMS is present at a concentration of about 50 μg protein/ml.

19. The synergistic antifungal composition of claim 13 wherein corn protein fraction CMS is present at a concentration of about 10 μg protein/ml.

20. The synergistic antifungal composition of claim 7 wherein said synergistic antifungal protein is comprises in corn steepwater or a protein concentrate thereof.

21. A method of inhibiting the growth of a fungus which comprises applying to said fungus or a habitat of said fungus the composition of claim 7.

22. The method of claim 21 in which said fungus is selected fro - the group of fungi consisting of Candida and Rhodotorula.

23. The method of claim 22 in which said fungus is a strain of Candida«,

24. The method of claim 23 in which said fungus is a strain of Candida albicans.

25. The method of claim 21 in which said antifungal antibiotic is a nikkomycin.

26. The method of claim 21 wherein said nikkomycin is present at a concentration greater than or equal to about 0.03 times the MIC of said nikkomycin alone.

27. The method of claim 21 wherein said nikkomycin is present at a concentration greater than or equal to about 0.01 times the MIC of said nikkomycin alone.

28. A method of inhibiting the growth of a strain of Candida albicans which comprises applying to Candida albicans or a habitat of Candida albicans a synergistic antifungal composition which comprises corn-SAFP and a nikkomycin that has anti-Candida activity, corn-SAFP and said nikkomycin being present in concentrations sufficient to cause growth inhibition of said strain of Candida albicans.

29. The method of claim 28 wherein said nikkomycin is nikkomycin X.

30. The method of claim 28 wherein said nikkomycin is nikkomycin Z.

31. The method of claim 28 wherein said nikkomycin is present at a concentration greater than or equal to about 0.03 times the MIC of said nikkomycin alone.

32. The method of claim 28 wherein said nikkomycin is present at a concentration greater than or equal to about 0.01 times the MIC of said nikkomycin alone.

33. The method of claim 28 wherein the corn-SAFP is present in amounts sufficient to lower the MIC of nikkomycin by about 33-100 fold relative to the MIC of nikkomycin alone for said strain of Candida albicans.

34. A method of obtaining substantially pure corn-SAFP from corn ^~~ meal which comprises the steps of:

1) preparing an AFP protein extract by extracting corn meal using either acidic or neutral pH buffers and centrifuging the resulting suspension;

2) subjecting said AFP protein extract to ammonium sulfate fractionation to obtain a 30-55% protein fraction;

3) subjecting said 30-55% protein fraction to chromatographic separation on carboxymethyl-Sephadex eluting protein fractions with a linear salt gradient (0.01-0.2 M NaCl) ;

4) selecting and collecting from said carboxymethyl- Sephadex eluted protein fractions that fraction which displays synergistic anti-Candida activity;

5) subjecting said fraction which displays synergistic anti-Candida activity to phenyl-Sepharose column chromatography employing a low salt concentration (0.1 M NaCl) buffer and selecting from the protein fractions eluted from said phenyl-Sepharose column the protein fraction containing synergistic anti-Candida activity and collecting said phenyl-Sepharose SAFP containing fraction to obtain substantially pure corn-SAFP.