ISOLATION AND IDENTIFICATION OF FUNGI FROM MUSHROOM COMPOSTS AND EVALUATION OF THEIR BIOLOGICAL ACTIVITY

K c y words: A n t i b a c t e r ~ a l act ivi ty , a n t i f u n g a l act ivi ty , A s p e r g ~ l l ~ i s f 'u~,~igutils, rnushrooln compost, nichodernia harzirrnu~ii.


INTRODUCTION
Straw mushroom (Volvariella volvacea) and oyster mushroom (Plez~rotus spp.) are cultivated a t the research station of Export Development Board (EDB), Ratmalana, Sri Lanka, to develop growth conditions and improve yields of mushrooms for local consumption and export. Oyster compost samples a r e packed in polybags and subjected to steam sterilization before mushroom spawn is introduced. Components of straw mushroom compost once mixed are only subjected to a self heating process where temperature of compost piles sl~ould theoretically increasc to about 70" C due to natural activity of rnicrobcs in composting substrates.' High temperature cnvironments within compost substrates, during the 2 different composting procedures should remove plant pathogenic, saprophytic, parasitic fungi a s well as insects and other unwanted microorganisms. Howevcr, complete elimination of unwanted mycoflora does not take place during the composting processes. Many thermophilic, heat tolerant and mesophilic fungal contaminants remain. The above fungi can contaminate the spawn added to various mushroom compost substrates, leading to unusual discoloration of the substrates. The surviving fungal flora can compete with mushroom spawn for nutrients, moisture, and could also act as antagonistic organisms to mushroom mycelia, leading to the reduction in the yield of mus11i-oom fruit bodies, affecting t11c quality of thc final product . One objective of' our research project was t o identify and isolate mycoflora which survive self heating or high temperature treatment in mushroom composts. The second objective was to screen t h e identified mycoflora f'or their capacity to produce biologically active compounds such as antifungal and antibacterial substances on selected media.

METHODS AND MATERIALS
Isolatttzg nmzd idemttifyi~zg fzsmzgi fi-oln compost: Straw compost ( straw, horse manure, kapok waste, water) and Oyster compost (saw dust, rice bran, calcium ca~bonate, magnesium sulphate, water) samples were collected fi-om EDB Research Station 111 I-atmalana, Srl Lanka. Oyster mushroom conlpost samples were at dii'ferent stagcs as fbllows: a) after packing in polybags, no visual contamination, b) after pacliing, contaminated with mold flora, c) samples in polybags to which nlush~.oorn spawn had been added, no visual contamination, d) samples after addition of spawn whlch are contaminated with mold flora. Total o f 8 polybags of oyster compost a t a-d stages were randomly selected from a series of'polybags arranged on shelves inside a mushroom cultivation shed. From each polybag 3xlg samples of compost were withdrawn randomly for isolating fungi (total # of samples from 4 stages = 24). Straw mushroom samples were collected, 1) during the self heating process where the temperature of piles would have been around 70 O C and 2) after spawn has been added. Twenty 1 g (ten from each stage) compost samples were collected randomly from different areas of 4 compost piles (five sampleslpi lc). Oyster and straw compost samples were sprinkled on yeast extract starch agar (soluble starch 15.0g, yeast extract 4.0g, K,HPO, 1.0g, MgS0,.7H20 0.5g distilled water 11) and plates were placed inside a moist chamber (glass tank lined with several layers of moist filter paper) , incubated in an incubator (Sailyo,35,40,45, 50U C f'or 7 days.' (10 replicate plateslten~perature). Isolated fungi were transferred to yeast extract starch agar slants for identification and storage. Pure cultures of isolated fungi were prepared and identification was carried out with the aid of slide cultures, temporary mounts and publisl~ed keyshooks .
Prepar.nt~oiz of'liquzd cr~ltz~res: Isolated Aspergillz~s fismnigatrcs was grown on yeast extract starch agar plates and incubated for 10 days at 30" C. Yeast extract starc11 liquid medium was used to promote the growth of the above fungus.' Fifteen Erlenmeyer flasks each containing 100 ml of medium were sterilized a t 121" C and 1.03 kg ~m -~ pressure for 20 mia. Each cooled flask was inoculated aseptically with 2 discs of 0.5 cm diameter pure culture of Aspergillrts fulnigatrss and flasks were ii~cubated for 12 days a t 30°C.' Isolated Trichodermna lzarzialtr~ln was grown on a liquid medium (maltose 20.0gJ glucose 10.0g, peptone 2.0g, yeast extract l.Og, MgSO,O.fig,0.5g,CaCI,0.05g,FeCl,,0.01g, ZnSO, 0.01g, distilled w a t c r lI).4r' Fifteen Erlenrnyer flasks containing (100 ml each) medium were inoculated aseptically with 2 discs of 0.5 cm diameter of a pure culture of T lzarzsanr~~n grown on yeast extract starch agar, and t h e flasks were incubated a t 30 "C for 21 day~.~."fter incubation period liquid media were separated from Aspergsllz~s or Trichoder~na fungal mats by filtering through Whatman No. 1 filter paper. Filtrates from 10 flasks (for each fungus) were pooled, pH noted, freeze dried and saved for bioas~ays.~."

Biocissc~ys fbr cintifi~ngal activity:
(1) Disc dif'f~~sion bioassay-Crude residues (2 mg each) of A. fiim~gatzrs and I: were dissolved in 0.5 nll of water. Pure cultures of' three plant pathogenic fungl, namely Fr~sarsz~7-rz oxysporz~nz, R h i z o c t o~~r a solans, and Colletotrzclz~~m l~ndenzs~tlzianz~~n were selected as test organisms. Cultures of the 3 test fungi in slants were transferred to petri dishes containing potato dextrose agar (PDA). After one day, sterile filter paper discs (Whatman # 40, diameter 1.3 cm) Impregnated with various levels ofA. fzrlnigatus crude extracts (to give concentrations of' 100-1000 pg) were placed on plates about 2 cm away from the margin of each plate. Three discslconcentratlodtest organism and one water control was used per plate The total 11~1rnbcr of' plates used per selected level of extract per test orgallism was fif'tcen. All plates were incubated fbr 1-4 days at 30') C. Inhibition effect of' elude extracts of' A. fir~n~gatz~s on each test organism was recorded by tneasu1,lng rllametcrs of' illhibitioll zones of treatments and comparing them with the controls.""'The above bioassay was repeated using higher levels (1000 -10,000 yg) of' crude extracts of A. fi~lnigatz~s. The same bioassay was adapted to test the antifungal properties of aqueous crude extracts of T. karzialzrrm; the experimental conditions were identical to the ones used for A. fumigatus., (2) Claclosporiz~m TLC bioassay -Cladosporiu~n cladosporoides spore suspensions were pl.cpared in the following manner: Cladosporiz~~n was grown on PDA medium and l~lcctbated for 10 days at 30° C. A pure culture of 7-8 day old Cladosporsz~ln grown on PDA medium in 100 ml Erlenmeyer flasks was selected for the bioassay.
A 25 rnl portion of Czapex dox nutrient liquid medium (sucrose 30.0g, KNO, 2.0g, lI,HPO, 1.0g MgS0,.7H20 0.5g, KC1 0.5g, FeSO, 0.01g, distilled watcr 11) was pourcd into culture flasks and shaken for 20 mill on a shaker (IUCHI SEIEDO, model SRR-2) to facilitate release of c~n i d i a .~.~ The suspension was filtered through a muslin cloth and the conidia in the filtrate were adjusted to 5x106 conidia per ml.617 Crude residue (2 mg) of Aspergillus furnigatus dissolved in 0.5 ml of methano1:water (1:l) mixture was spotted on five silica gel 60 F-254 plates (5cmx20cm) and the plates were placed in a glass tank and developed in ethyl acetate:hexane (1:l) mobile phase. The developed plates were air-dried and sprayed with a suspension of conidia of Cladosporiz~nz in Czapex dox nutrient solution and incubated in a moist chamber for 48 h at 30° C. Presence of iid~ibition zones (white) was noted against a green backgroui~d.~.~ The above experiment was repeated using crudc extracts of T Izarzinnz~in.

Streptococczis viridans, Staplzylococcus aureus, Klebsiella aerogerzes and
Escherichia coli were selected as test organisms. Each bacterial species was grown on 50 ml nutrient broth cultures in Erlenmeyer flasks and incubated for 1 day a t 30" C. A dilution series of each culture was prepared aseptically to study the growth (density) of' each bacterium. Dilution 10." which co~~tained about 2 x 10.' cells/ml was selected as the most suitable dilution. A set 01' 15 nutrient agar plates were inoculated with 0.5 ml of selected dilution of each bacterium and inoculum was spread evenly with the aid of'a sterile glass rod. Plates were incubated for 30 min. Three sterile filter paper discs (Whatman # 40, diameter 1.3 cm) inlpreg~~atcd with various volumes of aqueous crucle extracts of A. frtmigatus (to give concentrations of 10-40-pg) were placed on incubated agar plates about 2 cm from the edge of each plate. One disc moistened with sterile distilled water (control) was also placed in each plate. The size ofinl~ibition zones around the discs (indicating antibacterial activity) were measured after 1-2 day incubation period a t 30° C.4.b The above bioassay was repeated using higher levels (50-2000 pg) of the same crude extract. The sanie bioassay was adapted to test the antibacterial activity of T. hurziancrm; the experirne~i~al conditions were identical to the ones used for A. fitmigat~s.~." (2) Hamburger & Cordell bioautographic TLC assay -Forty eight Iioul-old bacterial cultures of Stapltylococcus aureus and Streptococcus oil-iclcrr~s i n 11 u trient b r o t h cultures were centrifuged a t x2000g i n a rcfrigerated high speed microcentrifuge (KOKUSAN ENSHINKI, model H-1500 SR). Liquid portion was discarded and the pellets were mixed with 20 ml of fresh nutrient agar broth to prepare each bacterial suspension. Crude residue (2 mg) ofA. fumigatzts dissolved in 0.5 ml of methano1:water (1:l) was spotted on ten silica gel 60 F-254 plates (5cmx20cm) and the plates were developed in a glass tank containing ethyl acetate:hcxane (1:l). Five air dricd plates were sprayed with Stapllvlococcr~s and another five with Streptococcus bacterial broth suspension. All TLC platcs were incubated in a humid chamber (glass tank) a t 30° C overnight. Plates were sprayed with a solution of tetrazolium chloride (20mglml) and incubated in a moist chamber a t 37O C for 4 h and platcs treated with 70% alcol~ol were clicckcd for. white inhibition zones against a pink background after one day8 The same bioassay was repeated using crude residue (2 mg) of 5 ? harzianum.

Extraction of active compolinds from tlze fungal residues: Aspergilbts firmigatus
and Diclzoderma harzianum crude residues from culture filtrates were subjected to a solvent extraction procedure in order to facilitate the separation and purification of active compounds. Hexane, ethyl acetate, methylene chloride, ether and metllano1 (5 ml each) were used as solvents. Weight of residues in each fraction was recorded after vapourizing the solvents on a rotavapour (BUCHI, model 912439).!' Known weights of residues of A. fiimigatus and T. harziunum were dissolved in water and the disc diffusion bioassay was carried out using different volumes of extracts (to give 10-100 pg concentrations). The test organisms used were

Streptococcr~s viridans, Staphylpcoccus anreus, K. aerogenes and E.coli .4.6
Purification atzd separation of active fractions rrsilzg column and thin layer chromatography: (a) Residues of' methanol fraction of A. firmigatus (obtained from the previous experiment) were purified on a silica gel-60 column using 50 ml each of methylene chloridc:methanol(7:3), and methylene chloride:methanol(6:4) and pure methanol. Three fractions collected ti-om the column were spotted on silica gel-60 TLC plates (20x20 cm) and the plates were developed in a moist chamber using methylene chloride: methanol (7:3) as the mobile phase." Three fractions -AM-1, AM-2 and AM-3 were identified under long wave uv ( 356 nm). Methanol was removed using a rotavapour and the three fractions were dissolved in 0.5 ml each water. Various volumes of aqueous fractions (to give concentrations of 10-40 pg) were tested for antibacterial activity against Streptococcus viridans and Staphylococczrs aurezrs using the disc diffusion method and Hamburger and Cordell assay.

RESULTS
Four species of fungi namely Aspergillz~s fiunigatzcs, Clzaetomiz~m thermoplzile, Mrrcor pzrsi11~~s and I).iclzodermu Izarzia~zt~rn were isolated and identified from straw and oyster mushroom compost samples incdbated at a range of temperatures. Mucor pirsillirs was isolated a t a frequency of 50% and 60% respectively from y e a s t extract s t a r c h a g a r plates incubated a t 5 different temperatures after sprinkling oyster and straw compost samples taken from different stages of processing. Chaetomium tlzermophile could be isolated a t a lower frequency of 30-35% from plates sprinkled with oyster and straw composts. T h a r z i a n z~m a n d A. f'r~migutzrs occurred in all plates (100% frequency) incubated with both types of compost samples a t t e m p e r a t u r e s of 30-50" C, ii-respective of the stage of processing of' composts.
When antifungal activity of crude extracts ofA. fi~~nzgutrrs and T I~urztu17~11n was checked using the disc diffusion bioassay, inhibitory effect on the coloily growth of' the 3 test fungi-Rhizoctorzzu solatzi, Fusarium oxysporirrin and Colletotric/ruii~ Li~zdemutlzianr~~n was observed a t fairly high levels of 5000 .and 10,000 yg. Thc effect of A. fr~172.igatz~s and Tharzianum crude extracts on F. o x y s p o~-r~m was more prominent a t 5000 a n d 10,000 yg levels. Results indicated a 16.7% and 30% m e a n reduction of colony r a d i u s (cm) of F. oxysporum exposed t o 5000 a n d 10,000 pg levels of crude extracts of A. fumigatus respectively. The mean reduction of the colony radius of the same fungus exposed to the crude extracts of T.
Izarziu~zr~m was 13.4% a n d 50% a t 5000 a n d 10,000 pg levels respectively. However, t h e same test fungi were insensitive to lower levels (100-1000 pg) of' crude extracts from both A. fi~migatzrs and T Izarziaizum.

L L O I L Z I I H was
Whcil antifungal activity of' crude extracts of A. fninigatz~s and ' I : hcir-. t e s t e d usings Cladosporiurn TLC bioassay, clear inhibition zones could not be detected.
Staphylococcus azirerrs, Streptococcus viridans a n d K. aerogenes were inhibited a t relatively high (50-2000 yg) and low (10-40 yg) levels of crude extracts of A. fi~migatus during disc diffusion antibacterial bioassay (Table 1). E.coli showed sensitivity to A. funzigatus extracts a t 30-40 pg levels and no sensitivity a t lowcr levels.  (Table 2). At low levels, a gradual increase in the amount ofthe crude extract (10-40 pg) demonstrated a slight increase i n inhibitory action against t h e above 3 test bacteria. E.coli indicated sensitivity to crude extracts a t 30-40 pg levels and no sensitivity to lower concentrations of the same extracts. Clear inhibition zones of antibacterial activity a g a i n s t St?-eptococcns andStaplzylo-cocci~s could not be dctected in TLC plates spotted with crude extracts of A. fimzgatus and T h u l -z n r~~~t n . However, a reduction of the growth of' both types of bacteria was seen near thc base line in all TLC plates.

E.coli
L'no effect no effect 1.Sf 0. Purification and separation of active fractions ofA. fi~migatus using column and thin layer chromatography resulted in identifying 3 active fractions namely, AM-1, AM-2 and AM-3 in methanol which were later dissolved in water and testcd fbr anti bacterial activity. When the above fractions were subjected to disc diffusion bioassay, all three aclueous fractions sl~owed activity against Sfr~plrylococc~~s n~lreus and St)-eptococci~s u~~.l,dn~zs. F1.action AM-1 sliowed thc highest act~vi t,y a g > l~l l s t Strcptococcz~s a n d Stuplzylococci~s, whereas AM-2 and AM-3 dcmonstratecl similar inhibitory action against the same bacteria compared to the control. (Table  3). When Hamburger & Cordell TLC bioassay was carried out, only one fraction (AM-1) showed antibacterial activity against Streptococcus viridans and Stuplylococcus aureus whereas AM-2 and m1-3 were not active against the same bacteria. The active spot (Rf=0.91) appeared as a yellow circle of 1 cm diameter on a pink background on TLC plates.; this antibacterial fraction was labelled as AM1-H. Aqueous fractions, TE-1, TE-2 and TE-3, which were prepared from I: harziarzz~m ethyl acetate fraction when tested using the Hamburger & Cordell assay, did not display any antibacterial activity. Four aqueous fractions of T l~ta~ziarrz~m -TM-1, TM-2 TM-3 and TM-4 (which were initially extracted to methanol ), wl~eil tested exhibited coilsiderable antibacterial activity against both Streptococcus viridans and Staplzylococc~~s aureus ( Table 4). Control "no effect . .

no effect
,'mean of 45 replicates?standard deviation; 'mean of 15 replicates of control.

DISCUSSION
The present survey indicates t h e occurrence of thermophilic fungi such a s Aspergilltss fi~migatzrs, CI~aetomir~m ther~nophile and Mz~corpusillus in straw and oystcr compost substrates a t different levels of processing. Trichoder~na harzia11~11n was also present in all the compost samples tested. Sincc most of the isolated fungi could theoretically grow a t a wide temperature range of 20-5SU C; and iT).iclzoderrr~a around 10-40" C, conipost samples were incubated in an incubator under humid conditions a t 30-50° C to isolate fungi. Eventhough two separ a t e composting procedures a r e adapted a t EDB, Ratmalana for straw and oyster compost substrates, types of fungi present on both substrates were t h e same a n d their frequencies of occurrence were also relatively similar. M.
prrsillus, C. tkermophile, A. firmigatus have been reported as thermophilous fungi living in various mushroom compost^.'^^ A. ftrlnigat~rs which i s considered a s a human and animal pathogen, has been recovered previously from soil, compost and plant substrates."1° A. firlr~igatirs which is the causal agent of aspergillosis, farmer's lung disease and other related disorders, has also been isolated from body fluids of acquired immunodeficiency syndrome (AIDS) patients.'" The above fungus produces many secondary toxic metabolites such a s fumitremorgin, gliotoxin, fumitoxin which attack the central nervous system leading to tremorgenic effects in affected individual^.'^. A. fumigatus which can survive a t a range of water activities (a,") from 0.82-0.97 a t 40" C has been found to grow on composts at high temperature upto 55°C and low oxygen tensions.1° Since the above fungus could cause deleterious effects in man, necessary steps should be taken to decontaminate compost before introducing mushroom spawn.
Many different weed moulds have been recognized in compost, competing with mycelia of mushrooms for nutrients, water and space. Weed moulds such as Clzaetonzirrtn globosr~m can cause severe reductions in yield if the composting procedure is inefficient or if substrates infested with spores of molds are used in the preparation of compost.'l However, Clzaetomium thermopltile has been identified as a useful fungus which helps to increase temperature in compost piles along with actinomycetes and bacteria, during composting.ll Species of Dichoderma namely I: viridae, T harzianum, T hnmatum and T. ko~zingii which are considered as weed moulds appear as dense green mats on mushroom composts. Results of previous surveys indicate that 10-80% of bags in mushroom houses in the United Kingdom, being infected with Tkickoderma resulting in considerable loss of income for the grower.ll T Itamatwn and ?: 1ta.rzianrrnz can also affect the growth of oyster mushroom ( Pleurotz~s spp.) through the production of a diffusible, non-volatile toxin which has been shown to kill mycelia of t h e above m u~h r o o m .~~ Wet and improperly pasteurized soil or compost components could assist the development of Trichoderma on composts.ll Mushroom cultivated on heat treated straw have previously been colonized by the cellulose competing antagonist -II h a m a t u~n . '~ Relatively high levels (5000-10,000 pg) of crude aqueous extract of A.fi~migatus when subjccted to disc diffusion assay, demonstratccl antifungal activity agalnst the three test fungi. The inability to identify inhibition zones during the Cladosporiz~m TLC bioassay could be due to the low polarity of active con~pounds in crude extracts of A. ficmigatzss. Siilcc lower levels of crude extracts (<I000 yg) were not active against test fungi during disc diffusion assay, further studies on antifungal properties were not conducted. Few strains of thcrmotolerant A. fz~migatzss from soil have been shown to produce several antifungal compounds when optimum conditions are provided. A family of antifungal compounds named as "sphingofungins" identified by growingA. fzsmigatus on solid millet-based medium, has been characterized as potent inhibitors of first enzyme (serine palmitoyl transferase) i n the sphingolipid nietabolic pathway in mammals.Vne of the aims of the present survey was to isolate new antifiungal/antibacterial compounds from A. f~u~zigatzss grow11 on yeast extract starch liquid medium.
Dnring our survey we were successful in demonstrating thc ability of A . ficm~gatus in producing antibacterial compound(s) on yeast extract starch medium. Since AM-1 to AM-3 fractions of A. firnzigatus were.cqually sensitive to Streptococczc.~ viridans and Staplzylococczrs aurezrs, we believe that our future research in identifying and characterizing purified fractions could make a valuable colitribution towards developing new therapeutic agents for the treatment of plant and 1 or human bacterial diseases.
Present survey demonstrated the antifungal nature of crude cxtl-acts of' T. ':zal-zialz~~~?z at fairly high levels (5000-10,000 pg ) against three phytopatl~ogenic f'i~ngi. Since tlie lower levels (100-1000 pg) when tested using disc diffusion bioassays were not active against all 3 fungi we focussed our research only on antibacterial assays. However, in a recent study of chitinolytic enzynles extracted from T Izarzianvm has demonstrated antifungal activity against Fzrsal-iunz spp., Pytlzium and Botrytis, specially in combination with enzymes from Enterobacter cloacae.'%nother survey indicated the ability of T lzarzianzem in produciilg "trichorzins" and "harzianins" (linear peptides belonging to the peptaibol dass) on a liquid medium containing glucose and seven mineral salts; trichorzins and harziailins have exhibited antifungal activity against pliytopathogenic Sclel-otium ce11zvortrm .lsl Another biologically active compound known as "trichodermiii" has been isolated from Trichoderma species such as T. Itarzianunz, T viridae, T honingii, and T. l i g t~o r~~i ?~ grown on barley seeds (solid-phase fermentation).14~'Commercial preparations of'trichodermin are being used in Russia to manage a wide range of phytopathogenic fungi causing diseases on many ecoilomical crops such as tomato, capsi-cum, cucumber and cereal^.'^,^^ Ability of t h e fiichoderma i n producing antifungal compounds could vary depending on m a n y conditions such a s Piclzoderlna species selected, solidlliquid medium used for fermentation a n d incu bation conditions. During t h e present survey, only aqueous TM fractions of T. harzianum exhlbitcd antibacterial activity against Streptococcus uiridans and Staplzylococczts CLILI.C'ZLS. This proves that the active compound is fairly nonpolar, hence in future a differtent solvent system will be used as the mobile phase during Hamburger and Cordell bioautographic TLC assay for better separation of active components. Large scale fermentation of ' I : ltarzianr~m and A. filnigatrss on the 2 selected media to obtain fairly high levels of crude extracts h a s already been initiated. We propose to idcntify TM and AM fractions from Tlzarzianum and A. fumigatus and purify both fractions using chromatography techniques before elucidating the structures with the aid of NMR and GC-Mass spectroscopy.