Imernatumal 3tmnlal of F.tul Ahcrohu~k~gv. 17 (1'492) 145- t 58 ,-, 1992 El~cv,cr Se,cncc Publi~hcr~ B.V. All r,ghts re~el~cd OIt~8-ledlS/92/$llS.01)

145

FOOD 1H)555

Media for detecting and enumerating yeasts and moulds Larry R. Bcuchat I"(md Safi'lv oral (~uaht~" l:nha~cement i.ahora~.rv. IX,par/merit of b'~s~t S( ten( ~' and Technoh~n', Un, cr~ay ,) th'or~a. ( ; n l f i . , G.4. U.SA

Dilution plating techniques are designed to determine populatum~ ol, viable fimpal prupagules per unit, weight, or vu|um¢ of fi~(I. Direct, pl;.Lting tcChnlqm:,,, on the ~lher hand. are designed to as~,ess the inl,ernal rnyct)flilra iff individual pieces of hxlds, e.g., ,,cuds nr dried fruW,, and re,,u|ts arc expressed a', a percentage of infecl,ed pieces. L~dh techniques are used by industry .'rod regulatory agencies l,o moral,or leveN of fung:ll conlaminal,ion ;it variou', stages of fl'4)d handling, sl,oring, proee.,,.sing and marketing. Peptone (().lr;) water is eummonly u~cd as a diluent fi~r ',ample,, to be homugenh,ed or blended. ')aflured diluenl,s containing up In 30% gluco,,e t~r ~)c; sucrose are recommended fi)r enuincr:ding xerophiles. Nt~ nne medium i-,,,,~tisfactory fl)r detecl.itm or ,.'numeration of yea,,ts :rod mtmld.., in all f,x~ds. Dich[nran rose bengal chloramphenicol agar, o,(ytctr:icycline glucose yeast extract agar and ruse bengal chhiramphenicol agar are superit)r to .',etdified Ix1tah~ dc~tro~e ,ear for enurncralion of yeasl,s and moulds. D,chhnan 18c~ g|ycerul agar pe~G~,,,~ ~'~:11for enumerating rm)der:dc!y xelx)philic yea,~l,s and moulds. Fasl,idinus xerophiles require medm containing high content,rations of sugars a n d / o r .,~)dium chloride. Media have been formulated It) detect, i~)l,entially aflats~xigenie aspurgilli and mycotoxigenie strains or pemcdlia and fusaria, but increased selectivity :rod specificity of media for del,ecl,ing mycol,oxigenic moulds are needed, lieal,-resistant m~mld a,,cC~Slx~res~flen require heat treat,taunt prior to plating in order 1o activate the germination pr4~cess. The spread-plate technique is sl,rongly preferred over the wmr-platu technique for enumerating yeast,, and moulds. The recommended incubation l,emperature is 25°C, hal, incubation time between plating and eunnl,mg colonies ranges from 5 days [or delerminafion ~)f general p~)pulatitms of myc~flora to 4 week,, o r more fiw fastidious xerophiles. There is a need ¢or new and improved media fi~r selectively i~lating various groups, genera, species a n d / o r strains of fungi capable of growing only under spcofie env,ronmental conditions, e.g.. low a,~ or. in the cas~ of suhlethally injured cells, under condil,ions wh,ch fi~cflitate resuscil,al,ion, Improved media are needed which accurately del,ccl, moulds producing specific mye~m~xins in a wide range of fiK~l types. Key words: Yet, st: Mould', Fungi: Xerophlhc; Ileat resiqanl,: Mycol,oxigeni~.; Diluent: Select,ire media; Enumeration

Introduction Y e a s t s a n d m o u l d s a r e d i s t r i b u t e d w i d e l y in soil, w a t e r a n d air. C o n s e q u e n t l y , unprocessed materials of both plant and animal origin are contaminated with fungi

(3~m.slumdem'e address" Larry R. Beuehal,. F~x~l Sal'ety and Quality Enhancement Lalmratory. Depart-

ment of Food Science and Technology. Universil,y of Georg,a. Griffin. GA 30223°1797, USA.

146

at the time they reach the food manufacturer. Processing schemes can either render the finished food product free of fungi or merely reduce populations. Given enough time, survivors may grow and eventually spoil the product. The detection and enumeration of viable yeasts and moulds in unprocessed and processed foods is an integral part of total quality management programs, and can be used to monitor the effectiveness of sanitation practices at each step during post-harvest and -slaughter handling, processing and distribution of foods. Dilution plating techniques are designed to determine populations of viable fungal propagules per unit weight or volume of food. Direct plating techniques, on the other hand, are designed to assess the internal mycoflora of individual pieces of foods, e.g., seeds or dried fruits, and results are expressed as a percentage of infected pieces. The direct plating technique is qualitative, then, rather than quantitative. Both techniques arc used by industry as well as regulatory agencies to monitor relative levels of fungal contamination at various stages of food handling, storing, processing and marketing. Procedures for analysis of any given product may differ from laboratory to laboratory. Certainly, procedures differ for analysing various types of foods. There are excellent reviews and descriptions of dilution and direct plating procedures in the literature (Booth, 1971a,b; Corry, 1982a; Jarvis et al., 1983; Koburger and Marth, 1984; Mislivec and Stack, 1984; King et al., 1986; Jarvis and Williams, 1987; Beuehat and Hocking, 1990) and the reader is referred to these publications. For detailed descriptions of common foodborne fungi, see Pitt and Hocking (1985a) and Samson and van Reenen-Hoekstra (1988). The following text provides essentially only an introduction to procedures for dilution and direct plating for the purpose of determining mycological quality of foods.

Sampling procedure Procedures for selecting samples of foods for mycological analyses differ depending upon the consistency or physical state of the food. Replicate samples representative of the lot to be examined must be obtained. The size of the sample for dilution plating may range from 5 to 250 g (ml). Samples selected for direct plating should consist of at least 100 pieces, s;x to ten pieces per 9-cm plate. All foods, whether solid or liquid, should be thoroughly mixed before samples are taken. Dry foods such as nuts and cereal grains are usually ground to a powder if they are to be analysed using the dilution plating technique. Sampling procedures are outlined by the International Commission on Microbiological Specifications of Foods (ICMSF, 1986).

Preparation of samples Preparation of samples for dilution plating may consist of manually shaking with a known volume of diluent, direct blending with a diluent in a blender or

147 pummelling with a diluent in a Stomacher. in some instances, rinsing of the entire sample with a diluent may be necessary. Jarvis et al. (1983) reported that higher populations are usually obtained by stomaching or blending than by shaking techniques. Hastings ¢t aL (1986) observed that there were no significant differences between mould populations detected in samples prepared by blending or stomaching. Methods by which samples are homogenized prior to secondary dilution are described by Jarvis and Williams (1987). Soaking foods in diluent prior to mixing and dilutiht can have at least three effects (Seller, 1986). First, soaking may allow resuscitation of sublethally damag~,d cells from dried, intermediate moisture or acidic foods. Soaking may also facilitate better release of cells which are present within tissues. Thirdly, with hard or sharp materials such as cereals and nuts, the softening effect of soaking will help prevent damage to stomacher bags used for preparing primary dilutions. After homogenizing, samples should not be allowed to stand tot more than a few minutes before removing a portion for subsequent dilution and plating, since fungal propagules may settle to the bottom of the container, resulting in under- or overestimation of populations, depending upon the location in the mixture from which the portion is withdrawn. To assess the internal mycoflora of cereal grains, nuts and dried fruits and vegetables by direct plating, various chemicals are used for surface disinfection. These include 3% hydrogen peroxide, 2% potassium permanganate, 75% ethanol, 0.001% mercuric chloride and 0.35% sodium or calcium hypochlorite (Booth, 1971a). Contact time ranges from 1 to 5 min, followed by rinsing with sterile distilled water. Andrews (1986) stated that most food particles should be effectively surface disinfected by immersion in a 0.35% (w/v) sodium hypochlorite solution for 2 min. Rinsing with sterile water is not essential, provided samples are adequately drained after disinfection.

Diluent

Peptone (0.1%) water and 0.05-0.1 M potassium phosphate buffer (pH 7.0) are among the most commoniy used diluents for yeast and mould enumeration. However, several other diluents may be employed, depending upon the type of food under investigation. For example, it is important to use a diluent containing a sufficient amount of solute to minimize osmotic shock to xerophilic mould and osmophilic yeast cells when serial dilutions are made prior to plating. Buffered diluents containing up to 30% glucose or 60% sucrose have been used ior this purpose. Death of halophilic fungi is minimized by incorporating 6 to 10% sodium chloride into diluents. Surface active agents may be added to diluents to reduce clumping of mould spores and conidia. For a discussion of diluents suitable for various foods, see King et al. (1986) and Jarvis and Williams (1987).

148 Culture media General purpose media Because foods differ widely in composition and the environments to which they have been exposed, the types and populations of mycoflora also differ. Pitt (1986) characterized the ideal enumeration medium as having the following attributes. It should suppress bacterial growth completely, without affecting growth of food fungi. It should be nutritionally adequate and support the growth of relatively fastidious fungi. The radial growth of colonies should be constrained but spore germination should not be inhibited. Unfortunately. no one medium is satisfactory for detection or enumeration of yeasts and moulds in all foods. Traditionally, acidified potato dextrose agar (PDA) has been used for enumeration, but this medium is not an exceptionally good nutrient source and it may inhibit resuscitation of injured cells due its low pH (3.5). Antibiotic supplemented media such as oxytetracycline glucose yeast extract (OGY) agar (Mossel et al., 1962, 1970), rose bengal chlortetracycline (RBC) agar (Jarvis, 1973) and dichloran rose bengal ehioramphenicol (DRBC) agar (King et al., 1979) are superior to acidified media for both dilution and direct plating techniques. Monographs have been published for all these media in the ICFMH Pharmacopoeia (Baird et al., 1987). They are less inhibitory to injured yeasts and moulds, more effective in inhibiting bacterial growth and less likely to cause precipitation of food particles because of their higher pH (5-6). Basal media to which ehloramphenicol, ehlortetracycline, oxytetracycline, gentamicin, kanamycin or streptomycin is added at concentrations up to 100 t.tg/nll include OGY, DRBC, plate count agar, mycophil agar, malt agar and tryptone glucose yeast extract agar. The effectiveness of most antibiotics is diminished at alkaline pH, so if highIy alkaline foods are being analysed, adjustment of the medium pH to less than 8.0 may be necessary to minimize bacterial growth, Rose bengal (Jarvis, 1973; King el at., 1979) and dichloran (2,6-dichloro-4nitroaniline) have been successfully used to control growth of spre:,.ding moulds (King et al., 1979). Caution should be taken to avoid exposure of rose bengal supplemented media to light, since cytotoxic breakdown products (Banks et al., 1985) may result in underestimation of mycoflora in samples. Two techniques have been developed for general enumeration of yeasts and moulds which are a clear departure from traditional methods. The Hydrophobic Grid Membrane Filter (HGMF) technique, using an automated ISO-GRID TM sample processor counting system was developed by Brodsky et al. (1982). The HGMF is essentially a membrane base upon which is applied a hydrophobic grid. The food sample is passed through the filter before transferring it to the surface of PDA (pH 5.6) supplemented with 100 ppm chloramphenicol and chlortetracycline. Plates are incubated for 48 + 4 h at 25°C before colonies are counted. A culture film (PetrifilmTM YM) method has been developed for enumerating yeasts and moulds in foods. A dry selective indicator medium on a special support film is rehydrated upon application of 1 ml of diluted sample. The sample is then covered with a second film to prevent drying during incubation at 25"C for 4 or 5

14q days. This system has been demonstrated to perform well for enumerating yeasts and moulds in a wide range of types of foods (Beuchat et al., 1990, 1991).

Xerophilic fimgi While many yeasts and moulds capable of growing at low a,,. can also grow at high a~, some species require low a~ in recovery media to enable growth and colony development. Reduction of a~ is achieved by supplementing basal media with sodium chloride or polyhydric alcohols such as glycerol, glucose, fructose or sucrose. Dichloran 18% glycerol (DG 18) agar CHocking and Pitt, 1980; Baird et al., 1987), glucose yeast extract sucrose agar, yeast extract sucrose agar, malt salt agar (King et al., 1980), 25% glycerol nitrate agar (Pitt, 1979) and malt yeast 5% salt 12% glucose agar (Pitt and Hocking, 1985b) are but a few of the mycological media formulated Io select for fungi capable of growing at low a,,. Others are described by Beuchat and Hocking (199[)). D G I 8 agar (a,~ 0.95) was developed for enumeration of moderately' ":erophilie moulds and osmophilic yeasts in products such as grains, flours, nuts and spices. The medium supports growth of the Aspergiilus reslrictus series, Wallemia sebio Zygosaccharomyces roltrii, Debaryomyces hansenii and many peniefllia and other aspergiili (Beuehat and Hocking, 19911)~ Unlortunately, some Eurotium species grow too rapidly on DG 18 agar, obscuring the growth of other xerophiles. The use of surfactants and fungicides to control spreading of Eurotium colonies on D G I 8 agar has been recommended (Beuchat and Tapia de Daza, 1992; Tapia de Daza and Beuchat, 1992). Differences exist in tolerance of xerophilic fungi to various solutes and hydrogen ion concentration. In an investigation of germination and growth of six xerophilic moulds, Pitt and Hocking (1977) concluded thai a universal isolation medium for enumerating xerophilic fungi could be based on glycerol or glucose/fructose, but hot o , .~odium chloride as the a~-Iimiting solute. Apart from species of the Aspergilhts glaucus and A, restrictus groups, Xeromyces bisponJs, Wallemia sebi, Cho,sosporium species and some Eurotium and Scopulariopsis species, the vast majority of moulds arc not able to grow in the a,~ range of 0.61-0.70. Likewise, only a few species of yeasts will grow in this aw range. Those that can grow, have been described as osmophilic and can cause spoilage of high-sugar products. Particularly troublesome is Zygosaccharomyces rowtii. Raw sugar cane, maple, chocolate and fruit syrups, honey, confectionery products, jams and jellies, fruit concentrates and dried fruits are most vulnerable to degradation by Z. rola'ii. Spoilage of these products may also be due to growth of Z. bailii, but this yeast is more likely to b e implicated in acidic foods such as salad dressings, mayonnaise and pickles with somewhat higher a~. As with xerophilic moulds, plating media with reduced a w are required for recovering osmophillc yeasts from foods. Potato dextrose agar supplemented *o contain 60% sucrose (Restaino el al., 1985), a yeast extract medium containing 50% glucose (Jermini et al., 1987) and a wide range of other sugar-supplemented media (Scarr, 1959; King et al.. 1986; Beuchat and Hocking, 1990) are suitable for

150

isolating and enumerating osmophilic yeasts. Often, xerophilic moulds will also develop on these media.

Heat-resistant moulds Spoilage of thermally processed fruits and fruit products by heat-resistant strains of Byssochlamys fuh'a, B. nicea, Neosartorya ).schcri, Talaromyces flavus, Talaromyces bacillisponrs and Eupenicillium brifeldiauum have been recognized for some time (Beuchat and Rice, 1979; Beuehat and Pitt, 1992~ Ascospores of these moulds show high heat resistance, in some instances comparable to bacterial spores. Because of their low incidence in fruit products (less than 10 per 100 g or ml), relatively large samples must be analysed. Centrifugation may be used to concentrate ascospore~ i~ h'q:,,.'d fruit products, the force and time necessary being influenced by volume, viscosity and specific gravity of the sample. Heat-resistant ascospores may require heat activation before germination and growth will occur (Splittstoesser et al., 1972; Katan, 1985; Beuchat, 1987). Samples are heated at 75-80°C for 1 h before 50-ml quantities are combined with 100 mi of 1.5 × strength potato dextrose agar in 150.ram Petri plates. Plates are incubated at 30°C for up to 30 days before colonies are counted. Most viable aseospores will germinate and form visible colonies within 10 days; however, heat-injured and other debilitated ascospores may require additional time to form colonies. An alternative method which avoids error due to aerial contamination involves heating samples in flat-sided bottles followed by incubation on their sides. Colonies develop on the surface of the fruit product. This system is suitable for pulps and homogenates.

Mycotoxigenic moulds Several media have been formulated for use in recovering specific genera or groups of yeasts and moulds based on metabolic activities. Among these are media suitable for enumeration of moulds capable of producing mycotoxins. These media are of value to any quality assurance program concerned with mycological quality of foods from a public health standpoint. The presence of mycotoxin producers in a food does not necessarily mean that mycotoxins are present. On the other hand, the absence of mycotoxin producers is not evidence that mycotoxins are not present in the food, since growth followed by death of mycotoxigenic moulds may have occurred at some point prior to analysis. Nevertheless, media formulated to select for myeotoxin producers are of interest to the food mycologist and considerable effort is being expended in several laboratories to develop these media for routine use. Detection of Aspergillus flavus and Aspergillus parasiticus in foods indicates the possibility of aflatoxin contamination. In connection with their studies on aflatoxin contamination of grains, Bothast and Fennell (1974) developed a diagnostic medium which could be used to rapidly differentiate between members of the A. flavus group and other common storage fungi. The medium contains ferric citrate which promotes the formation of a persistent bright yellow-orange reverse pigmentation of colonies. The medium (Aspergiilus differential medium, ADM) also

15i

effectively distinguishes between members of the A. fiat'us group and other Aspergillus species of interest in medical mycology (Salkin and Gordon, 1975). Noting that supplementation of rose bengal streptomycin agar with botran (2,6-dichloro-4-nitroaniline) facilitated the isolation and enumeration of A. flatus from peanuts (Bell and Crawford, 1967), ADM was modified by Hamsa and Ayres (1977) to yield a medium with greater selectivity. Hara et al. (1974) developed a medium to simplify the screening of large numbers of A. flat'us isolates for aflatoxin production. Detection of aflatoxin-posirive strains utilized ultraviolet-induced fluorescence of aflatoxin produced in a Czapek's solution agar containing corn steep liquor, HgCI z and (NH.~)H.,PO.~ instead of NaNO 3. Pitt et al. (1983) developed a selective medium for enumerating A. flatus and A. parasiticus based on the medium of Bothast and Fenncll (1974). Results can be obtained after 42 h incubation at 30°C, making the medium suitable for use in industrial quality control laboratories. Results were reproducible and comparable with those obtained using standard fungal enumeration media incubated for much longer periods. A very low percentage of t'alse positives or negatives was reported. This medium is included in the ICFMH Pharmacopoeia {Curtis and Baird, 1993). A simple method for screening aflatoxin-producing .4. flart~s and A. parasiticus using ultraviolet photography has been developed (Yabe et al., 1987). In ultraviolet photographs of colonies, aflatoxin producers appear as grey or black colonies, whereas non-producers appear as white colonies. The technique may be useful as a simple, safe and rapid method of screening aflatoxin-producing moulds. A selective medium was developed by Frisvad (1983, 1986) to screen for Penicilliwn riridicanm~ and Penicillium ~'errucldosum as an aid in the examination of stored cereal products for toxic metabolites. The medium, pentachloronitrobenzene rose bengal yeast extract sucrose (PRYES) agar, contains 15% sucrose, chloramphenicol and chlortetracycline (50 mg/I), rose bengal (25 mg/I) and pentachloronitrobenzene (100 mg/I) as selective agents. Members of the Mucorales are completely inhibited, allowing important storage moulds to grow. Ochratoxin A and citrinin producers in the P. viridicatum group I! appear as violet brown reverse on PRYES agar, whereas producers of xanthomegnin and viomellein (P. viridicamm group I and P. auraatiogriseum) are indicated by their yellow reverse and obverse colors. The medium appears to be quite suitable |'~r .,,creening for nephrotoxin-producing fungi in cereals. Dichloran rose bengal yeast extract sucrose (DRYES) agar was originally formulated as a screening medium to detect Pemcillium ~'errucosum, but may be used for a more general determination of toxigenic Penicillium and Aspergillus species (Frisvad et al., 1990a). They recommend the use of both DG18 agar and DR'YES agar if populations and identity of fungi and their myeotoxin production is important in tht: examination of dried products. Three media have been developed for selective isolation and direct identification of various groups of moulds associated with cereal products, meat and cheese (Frisvad et al., 1990b). Acetic acid dichloran yeast extract sucrose (AD~t ES) agar is useful for selective and indicative isolation of moulds '~laoiling acid-preserved

152 bread. Roqucfortine C and PR-toxin produced by Penicillium roqueforti var. roqueforti on ADYES agar can be detected by an agar plug method. Dichloran creatine (DC) agar is an effective screening medium for penicillin associated with meat, cheese, nuts and other lipid and/or protein foods. The third medium, dichloran yeast extract sucrose (DYES) agar, was developed for qualitative determination of toxigenic penicillin and aspergilli. Creatine sucrose agar, containing creatine as a sole nitrogen ~uree and bromocresol purple as a pH indicator, is reported to be a good medium to differentiate subgenera of penicillin and to subdivide taxa and subgroups based on mycotoxin production into approximately two equal groups (Frisvad, 1985). Andrews and Pitt (1986) developed dichloran chloramphenicol peptone agar (DCPA), a selective medium for isolating Fusarium species and some dematiaceous hyphomycetes (e.g., Alternaria spp.) from cereals, The medium was shown to select against species of Aspergilhts, Penicillium, Cladosporium and mucoraeeous fungi. Fusarium species produced well-formed colonies with good conidial production, permitting rapid identification. The medium can also be used for the identification of Fusarium species (Hocking and Andrews, 1987). Conner (1990) reported that DCPA and DCPA supplemented with 0.5 !~g/ml crystal violet were effective in suppressing Aspergillus and Penicitlium while allowing selective growth of Fusarium species. The possibility of using the growth response of Fusarium species on tannin sucrose agar as an aid to identification was investigated by Thrane (1986). Of the eleven Fusarium species investigated, eight were able to grow on tannin sucrose agar. The difference in growth response proved to be an additional character usable for identification of species. A selective medium for Fusarium species was developed using Czapek-Dox agar containing iprodione [3-(3,5-dichlorophenyl)-N(I-methyl-ethyi)-2,4-dioxo-I-imidazolidine.carboxamide] (3 me/l) and dichloran (2 rag/I) (Abiidgren et al., 1987). This medium (CZID near) is selective against numerous species of Altenmria, Epicoccum, Penicillium and mucoraceous fungi. Fusaria produce large, easily recognizable colonies on CZID agar, facilitating isolation and subculture. Andrews (1990) evaluated media for differentiation of potentially toxigenic Alternaria alternata from the Alternaria state of Pleospora infectoria. The two species can be differentiated on diehioran chloramphenicol peptone agar (DCPA) but are more readily differentiated on dichloran chloramphenicol malt agar (DCMA).

Selectice media for yeasts Relatively few media are available specifically for enumerating or at lca:.t facilitating the growth of yeasts at the expense of moulds and bacteria. However, since yeasts may be the predominant microflora in some foods, e,g., confectionery products and fruit juice concentrates, special media for enumeration of yeasts are desirable. Oxytetracycline glucose yeast extract (OGYE) agar (Mossel et al., 1970; Baird et ai., 1987) has been widely used for many years, especially in European countries, as

153 a general medium for enumerating yeasts. The medium loses its bacteriostatie effect if incubated at temperatures creater than 25°C. Moulds will also develop colonies on OGYE agar. For selective recovery of psychrotrophic yeasts from chilled proteinaceous foods, oxytetracycline and gentamicin seem to be the bacteriostatic combination of choice (Dijkmann et al., 1979). Tryptone glucose yeast extract (TGY) agar supplemented with antibiotics can also be used successfully for products containing high populations of yeasts. Davenport ~1980)oullined a guide to media and methods for studying yeasts and yeast-like organisms in fotKis. Media have also been developed for enumerating yeasts in specific food products. For example, Kish et al. (1983) formulated a sclective medium for wine yeasts. The medium, containing 1511mg/I bisulphite and 12% (by volume) ethanol, was suitable for the enumeration of wine yeasts when present at low populations in the natural microflora during early stages of grape juice fermentation, in the brewing industry, lysine agar (Waiters and Thiselton+ 1953; Lin, 1975), Schwarz differential medium, and other selective media (Lin, 1973) are used to detect wild yeasts. Heard and Fleet (198{~)evaluated selective media for enumeration of yeasts during wine fermentation. Colonies of Sacdmromyces ceret,isiae dominated on malt extracl agar and sometimes masked the presence of other genera. Lysine agar suppressed the growth of S. ceret+isiae and enabled the enumeration of non.Saccharomyces species such as Kloeckera apiculata, Candida stellata and Saccharomycodes hedwigii. Growth of non-Saccharomyces yeasts on ethanol sulphite agar was variable, Molybdate agar fortified with 0.1Z~i% calcium propionate was successfully used by Rale and Vakil (1984) to selectively isolate and identify yeasts from tropical fruits. This medium requires a longer incubation period for full expression of the natural pigmentation of yeasts.

Plating technique The spread-plate technique is strongly preferred to the pour-plate technique for enumeration of yeasts and moulds in foods using dilution plating. Spread plating avoids any risk of the.~mal inactivation of fungal propagules which may be associated with the pour-plate technique and facilitates maximum exposure of cells to atmospheric oxygen, thus allowing sporulation to proceed unencumbered in those instances when identification of mycoflora is desired, Media should be prepared and i~oured {ca. 2(I ml per plate) 16-40 h in advance of use to facilitate 'drying' of the surface. Generally, 0.1 ml of appropriately diluted sample is deposited in duplicate or triplicate on the surface of media and then spread evenly over the surface using a sterile bent glass rod. Rods should not exceed 2 mm in diameter in order to minimize the amount of sample adhering to them at the end of the spreading procedure. Larger samples, e.g, up to 0.3 ml of a 10-n dilution of sample, can be spread on each plate to facilitate enumeration of low populations of yeasts and moulds.

154

Incubation Media used for general enumeration of yeasts and moulds should be incubated at 25°C for 4 or, preferably, 5 days before colonies are counted. Higher temperatures, e.g., 40-45°C, must be used for thermophiles whereas incubation at 5-15°C is needed for psychrotrophic fungi. The incubation time necessary before counting colonies of specific genera, species or groups of fungi varies. For example, accurate enumeration of moderate xerophiles on DOi8 agar may require 6-8 days incubation and mycelial development by fastidious xerophiles may require 4 weeks or longer. Petri plates should be incubated in an upright position and should not be disturbed until colonies are ready to count. Movement of plates can result in release of mould conidia or spores and subsequent development of satellite colonies which would give an overestimation of population in the test sample. If spreading moulds are a problem, ope may be forced to count colonies after 3 and again after 5 days of incubation. No attempt should be made to select plates containing 30-300 colonies for counting. Rather, select dilutions giving 10-100 colonies on 9-cm plates or 20-200 colonies on 14-cm plates. Several colonies should be picked from countable plates periodically and examined microscopically by wet mount to ensure that bacteria are not growing. Development of bacterial colonies is more likely to occur on acidified media than on antibiotic-supplemented media. Non-linearity of counts from dilution plating often occurs, i.e., 10-fold dilutions of samples often do not resutt in 10-fold reductions in numbers of colonies recovered on plating media. This has been attributed to fragmentation of mycelia and breaking of spore clumps during dilution in addition to competitive inhibition when large numbers of colonies are present on plates (Jarvis et al., 1983).

Interpretation of data Unprocessed fruits, vegetables, grains and nuts may harbour large 0opulations of yeasts and/or moulds without apparent spoilage. Counts exceeding 10~ C F U / g are not uncommon. The same populations in processed ingredients or finished products may indicate spoilage. The significance of high populations in either unprocessed or processed foods and food ingredients can only be assessed by reference to the history of the product and its intended treatment before it reaches the consumer. Results from direct plating reveal the extent of internal infection of samples and therefore may give an indication of the environmental conditions under v, hich the product was stored prior to analysis. The major genera of moulds detected using the direct plating techniques may also give some indication concerning time at which infection occurred, i.e., in the field or during storage. The presence of potentially mycotoxigenic meulds in the ir~ternal areas of grains or nuts, however, is not evidence that mycotoxins are present. Conversely, the absence or low

155

incidence of mycotoxigenic moulds is not evidence that the product does not contain mycotoxins.

Precautions

Mycological media should be monitored for performance by comparing recovew of yeasts and moulds as well as unwanted bacterial colonies. Seller (1985) concluded that the Miles-Misra method (Corry, 1982b) can be used for assessing reel, very of certain yeasts and for ensuring that bacteria are adequately inhibited. The method is not suitable for use with the more rapidly spreading yeasts and moulds which require a stab inoculation method and measurement of rate of colony diameter increase with incubation time. Strains of microorganisms selected for monitoring performance of media should exhibit stable characteristics which do not change after repeated subculturing. A range of microorganisms representing rapid- and slow-growing yeasts, moulds and bacteria should be included. Microorganisms likely to be present in the food under study should be selected for the evaluation. A proportion of mould and yeast cells in any given food being analysed may be metabolically or structurally injured as a result of physical or chemical stress (Beuchat, 1984). Such cells require optimum nutrient, pH, osmotic and temperature conditions for recovery if subsequent repair and colony formation are to occur on enumeration media. Infections, iutox~cations and respiratory problems as a result of inhaling fungal spores can occur in laboratory personnel working with fungi. It is important to not remove lids from Petri plates containing colonies of moulds unless absolutely necessary for purposes of identification. All plates on which colonies have developed should be properly enclosed and heat sterilized before discarding. Anyone who has mishandled moulds in a laboratory will be remembered by those who work in that laboratory for many years to come.

References Ahildgren, M.P., Lund, F., Thrane, U. and Elmholt, S. (198"/) Czapek-Dox agar containing iprodione and dichloran as a selectiv~ medium for the ~solatnon of Fuscmum species. Len. Appl. Micriobiol. 5, 83-86. Andrews, S. (1986) Optimizalion of conditions for the surface disinf,~ction of sorghum and sultanas using sodium hypochlorile solutions. In: A.D. King, J.I. Pill, L.R. Bcuchat and 3.E.L Corry (Eds.), Methods for Mycological F.,xaminatnon of Foods. NATO ASI Series A: Life Sciences, % [ 122. Plenum Press, New York. pp. 28-32. Andrews. S. (1990} Differentiation of AIternarla species isolated from cereals on a dichloran malt extract agar. International Workshop on Standardization of Methods for the Mycological Examinalion of Foods, Progr. and Abstracts. Baarn. The Netherlands. p. 25. Andrews, S. and Pitt, J.I. (1986) Selective mcdnum for isolation of Fe~anum species and dematiac¢ous hyphomycetes from cereals. Appl. Environ. Microbiol. 51. [235-1238.

156 Baird, R,M., Corry, J.E.L. and Curtis. G.D.W. (Eds.)(1987) Pharmacop.~eia of Cultiv¢ Media for Food Microbiology. Int. J. Food Microbiol. 5. 187-300, Banks, J.G.. Board. R.G., Carger, J. and Dodge, A.D. (1985) The eylotoxic and photodynamic inactivation of microorgamsms by rose bengal. J. Appl. Bacteriol. 58., 392-400. Bell, D.K, and Crawford, J.L. (1907) A hotran-amended mud;urn for i olating ,4spergillus tim'us from peal,,uts and soil. Phytopathology 57, 939-941. Beuchat, L.R. (1984) injury and repair of yeasts and moulds. In: Rev~-valof Injured MIcrobe~ M.H.E. Andrews and A.D, Russell (Eds.). Academic Press, London. pp.-293-308. Beuchat, L,R. (1987) Influence of organic acids on the heat resistance characteristics of Talaromyces fiat'us aseospores. Int. J. Food MicrobioL 6, 97-105. Beuchat, L.R. and Hocking. A.D. (1990) Some considerations when analysing foods for the presence of xerophilie fungi. J. Food Prot. 53, 984-989. Beuchat, L.R., Nail, B.V.. Brackelt, R.E. and Fox, T.L. (1990) Evaluation of a culture film (PetrifilmTM YM) methods for enumerating yeasts and molds in dairy and high-acid foods. J. Food Prof. 53, 864, 869-874. Beuchat, L.R., Nail, B.V., Braekelt, R.E, and Fox, T.L, (1991)Comparison of Ihe PctrifilmTM yeast and mold culture film method to conventional methods for enumerating yeasts and molds in foods. J. Food Prot. 54, 443-447. Beuchat, L.R. and Pitt, J.l. (1992) Detectton and enumeration of heat resistant molds. In: C. Vanderzant and D. Splettstoesser (Eds.), Compendium of Methods for the Microbiological Examination of Foods. Am. Public Health Assoc., W~shington. DC. pp. 239-249. Beuchat, L.R. and Rice, S.L. (1979) Byssocllamys spp. and their importance in processed fruits. Advan. Food Res. 25. 237-288. Beuchat, L.R. and Tapia de Daza, M.S, (1992) Evaluation of chemicals for restricting colony spreading by a xerophilie mold, Eurotmm amstetodami, on dichloran-18% glycerol agar. Appl. Environ. MicrobioL 58, 2093-2095. Booth, C, (1971a) Fungal culture media, in: C. Booth, (Ed.) Methods in Microbiology, Vol. 4, Academic Press, Lonoon. pp. 50-93. Booth, C. (1971b) introduction to general methods. In: C. Booth (Ed.), Methods in Microbiology, Vol. 4. Academic Press, London. Bothast, R.J. and Fennell, D,I. (1974) A medium for rapid identification and enumeration of Aspergiltus flat'us and related organisms. Mycologia 66, 365-369. Brodsky, M.N., Enlis, P., Entis, M.P., Sharpe, A.N. and Jarvis. G.A. (1982) Determination of aerobic plate and yeast mold counts in foods using an automated hydrophobic grid-membrane filler technique. J. Food. Prot. 45, 301-304. Conner, D.E. (1991)) Evaltlali0n of method~ for selective enumeration ~f Ft,~,rimn spp. in foodstulls. International Workshop on Standardization of Methods for the Mycological Examination of Foods, Progr. and Abstracts. Baarn, The Netherlands. p. 26. Corry, LE.L. (I982a) Assessment of the selectivity and productivity of media used in analytical mycology. Arch, Lebensm. Hyg, 33, 160-164. Corry, J.E.L. (1982b) Quality assessment of culture media by the Miles-Misra method. In: J.E.L. Corry (Ed.), Quality Assurance and Quality Control in Microbiological Culture Media. G.I.T,-Verlag, Darmstadt, pp. ,21-37. Curtis, G.D.W. and Baird, R,M. (Eds.) (1993) Pharmacopoeia of caltive media for food microbiology: additional monographs (2). Int. J. Food Mierobiol. 17, in press. Davenport, R.R. 0980) An outline to media and methods for studying yeasts and yeast-like organisms. In: F,A. Skinner, S.M. Passmore and R.R. Davenport (Eds.), BiologY and Activities of Yeasts. Soc, Appl, Bacteriol. Series No. 9, Academic Press, London, pp. 261-278. Dijkmann, K.E., Koopmans, J. and Mossel, D.A.A. (1979) The recovery and identification of pschrotrophic yeasts from chilled and frozen comminuted fresh meats. J. Appl. Bacteriol. 47, ix. Frisvad, J,C, (1983) A selective and indicative medium for groups of Penicillium z'iridicatum producing different mycotoxins in cereals. J. Appl. Bacteriol. 54, 409-416. Frlwad, J.C. (1985) Creatine sucrose agar, a differential medium for mycotoxin producing terverocdlatc Penieillium species. Lett. Appl Micriobiol. 1, 109-113.

157 Frisvad. J.D. (1986) Selective medium for Po~icilhum e'indicamm in cereals. In: A.D. King, J.l. Pitt, L.R,, Bcuchat and J.E.L, Corry (Eds.), Methods for the Mycological Examination of Food. NATO ASI Series A: Life Sciences, Vol. 122, Plenum Publ. Co., New York, pp. 132-135. Frnsvad, J.C.. Fillenbo~g, O. and Throne, U. (Iq~0a) Colhboldhv¢ .')Ladyon media for detecting and enumerating toxigenic fungi. 2nd International Workshop on the Standardization of Methods for the Mycolog,cal Examination of Foods. Progr. and Abstracts. Baarn. The Netherlands, p. 2"/. Friend, J.C., Lund, F., Filtenborg, O. and Thrane. U. (199061 Selective media for the detection oflb© associated toxigenic mycoflora of cereal products, meat and cheese. 2rid International Workshop on the Standardization of Methods for the Mycological Examination of Foods, Progr. and Abstracts. Baarn, The Netherlands. pp. 27-28. Hamsa, T.A.P. and Ayres, J.C. (1977) A differential medmm for the isolation of Aspergdlttsflm'us from cotton~ed. J. Food Sci. 42. 449-453. Hara, S., Fennell, D.I. and Hesseltine, C.W. (1974) Aflatoxin-producing strains of Aspergillt~s]lat'us detected by fluorescence of agar medium under ultraviolet light. Appl. Microbiol, 27, 1118-1123, Hastings, J.W, Tsai, W,Y.J. and Bullerman, L.B. 0986) Comparison of stomaching versus blending in sample preparation for mold enumeration, in: A.D. King, J.I. Pitt, L.R. Bcochat and .LE,L Cony (Eds.), Methods for Mycological Examination of Foods, NATO AS1 Series A: Life Sciences, Vol, 122, Plenum Press, New York, pp. 7-9, Heard, G,M. and Fleet, G.H. (lq86) Evaluation of selective medta for enumeration of yeasts during wine fermentation. J. Appl. Bacleriol. 60, 477-481. Hocking, A D. and Andrews, S. {1987) Dichloran chloramphenicol peptone agar as an identification medmm for FItsaritm) species and some dcmatiaceous hypbomycetes. Trans. Br. MycoL Soc. 89, 239-244. Hocking, A.D. and Pill, J.l. (1980) Dichloran glycerol medium for enumeration of xerophilic fungi from low-moisture foods Appl. Environ. Microbiol. 39, 4S8-492. International Commission on Microbiological Specification for Foods. (1986}. Microorganisms in Foods. 2. Sampling for Microbiological Analysis: Principles and Specific Applications. Unk'ersity of Toronto Press, "i'oronto. Jarvis, B. (1973) Comparison of an improved rose-bengal-chlortetracycline agar with other media for the selective isolation and enumeration of moulds and yeasts in food. J. Appl. BaclerioL 36, 723 -727. Jarvis, B., Seller, D.A.L., Ould, A.J.L. and Williams, A.P. (1983) Observations on the enumeration of moulds in food and feeding stuffs. J. Appl Bacteriol. 55, 325-336. Jarvis, B. and Williams, A.P. (1987) Methods [3r detecting fungi in foods and beverages. In: L.R. Beuchat (Ed.), Food and Beverage Mycology, Van Noslrand Reinhold, New York, pp. 599-636. Jermini, M.F.G., Geiges, O, and Schmidt-Lorenz, W. (1987) Detection isolation and identification of osmoiolerant yeasts from high-sugar products. J. Food Plot. 50, 468-472. Katan, T. (1985) Heat activation of dormant ascospores of Talaromyces 17at'us. Trans. Br. Mycol. So(=. 84, 748-750. King, A.D., Hocking, A.D. and Pitt, J,l. (1979) Dichloran rose bengal medium for enumeration and isolation of molds from foods. Appl. Environ. Mierobiol. 3~, 959-964, King, A.D., Pitt, J.l., Beuchat, L.R. and Corry, J.E.L. {l~ds.) (1986) Methods for the Mycological Examination of Food. NATO AS! Series A, Life Sciences, Vol. 122. Plenum Press, New York. Kish, S,, Share R. and Margalith, P. (1983) A note on a seleclive medium for the wine yeasts. J. Appl. Bacteriol. 55, 177-179. Kohurger, ,]'.A. and Marth, E.H. (1984) Yeasts and molds, In: M. Speck (Ed.). Compendium of Methods for the Microbiological Examination of Foods, Am. Public Health Assoc. Washington, DC, pp. 197-202. Lin, Y. (1973) Detection of wild yeasts in the brewery: a new criterion. Brewers Digest 48, 60-69. Lin, Y, (1975} Detection of wild yeasts in the brewery: efficiency of differential media. J, Inst. Brew. 81, 410-417. Mislivec, P.B. and Stack, M.E. (1984) Enumeration of yeasts and molds and production of toxins. In: Bacteriological Analytical Manual, 6th Edn. US Food and Drug Administration, Washington, DC. Mosse], D.A.A,, Visser, M. and Mengerink, W.Hj. (1962) A comparison of media for the enumeration of moulds and yeasts for foods and vegetables. Lab, Prac. 11, 109-112.

158 Morsel. D.A.A., Kieynen-Scmmeling, A.M.C., Vineentie, H.M., Beerens, H. and Catsavas, M. (1070) Oxytelracycline glucose yeast extract afar for selective enumeration of moulds and yeasts in foods and clinical material. J. Appl. Bacleriol. 33, 454-457. Pitt, J.l. (1079) The Genus PemciUium and its Teleomorphic Slates Eupcnicilllum and Talaromyces. Academic Press, London. Pitt, J,I. (1986) Properties or the ideal enumeration medium. In: A.D. King, J.I. Pill, L.R. Beuchat and J.E.L Corry (Eds.I, Methods for the Mycological Examination of Food, NATO ASI Series A: Life Sciences, Vol. 122. Plenum Publ. Co., New York, pp. I32-132. Pitt, J.I. and Hocking, A.D. (1977) influence of .volute and hydrogen ion concentration on the water relations of some xerophilic fungi. ,L Gen. Microbiol. 101, 35-40, Pitt, J.L and Hocking, A.D. (1985a) Fungi and Food Spoilage. Academic Press, Sydney. Pill, J.l. and Hocking. A.D. (t985b) New species of fungi from Indonesian dried fish. Mycolaxon 22, 197-208. Pitt, J,l,, Hocking, A.D. and Glenn. D.R, (1983) An improved medium for the detection of Aspergillt~ flal'us and Asl~,rgillusparasiticus. J. Appl. Bacteriol. 54. 109-114. Rale, V.B. and Vakil, J.R. (1984) A note on an improved molybdatc agar for the selecliv¢ isolation of yeasts from tropical fruits. J. Appl. Bactcriol. 56, 409-4)3. Restaino, L, Bills, S. and Lenovich. L.M. (1985)Growth response of an osmotolerant, sorhate-reslstant Saccharomvces rowfii strain: evaluation of plating media. J. Food Prot. 48, 207-209. Salkin, I.F, and Gordon, M.A. (1975) Evaluation of Aspergillus differential medium. J. Clin. MicrobioL 2, 74-75. Samson, R.A. at~d van Rcenen-Hoekstra, E.S. (198g) Introduction to Food-Borne Fungi. 3rd Edn. Centraalbureau voor Schimmelculturces. Baarn, The Netherlands. Scarr, M.P. (1959)Selective media used in the microbiological examination of sugar products. J. Sci. Food Agric. 10, 678-681. Seller, D.A.L. (1985) Monitoring mycological media. Int..L Food MicrobioL 2, 123-131. Seller, D.A.L. (1986) Effecl of presoaking on recovery of fungi from cereals and cereal products. In: A.D. King, LI. Pitt, L.R, Beuchat and J.E,L. Corry (Eds.L Methods for the Mycological Examinalion of Food. NATO ASI Series A., Life Sciences, Vol. 122. Plenum Press. New York, pp. 26-28. Splittstoesser, D.F, Wilkison, M. and Harrison, N. (1972) Heat activation of Byssochlamys fuh'a ascospores. J. Milk Food Technol. 35, 399-401. Tapia de Daza, M.S. and Beuchat, L.R. (1992) Suitability of modified dichloran glycerol (DGIS) agar for enumerating unstressed and temperature-stressed xerophili¢ molds. Food Mierobiol. 10, in press. Thrane, U. (1986) The ability of common Fusarium species to grow on tannin-sucrose agar. [¢ll. AppL Microbiol. 2, 33-35. Waiters, L.S. and Thisellon, MA. (1953) Utilization of lysine by yeasts. J. Inst. Brew. 59, 401-404. Yab¢, K,, Ando. Y,, llo, M. and Terakado, N. (1987) Simple method for .screening aflatoxin-producing molds by UV photography. Appl. Environ. Microbiol. 53, 230-234.