Isolation of Aureobasidium pul / uJans from Zimbabwean sources and glucosidase activities of selected isolates

The yeast-like fungus Aureobasidium pullulans (DeBary) Arnaud has been isolated and identified for the first time in Zimbabwe from diverse sources including fruits, leaves and commercial manure. Selective enrichment of an appropriate substrate using a minimal salts broth with shaking for two days at 25°C followed by plating on a corresponding solid medium, proved to be a reliable method for isolating the fungus. Extracellular a. and l3-glucosi-


Introduction
Aureobasidium pullulans (DeBary) Amaud is a yeast-like fungus which is usually associated with plant tissue and plays a role in carbon mineralisation processes (Hudson 1972, Burn s 1982. Its cu ltural and microscopic features include variab le levels of dark pigmentation on agar media and the presence of septate hyphae, ch lamydospores and budding cells, Since the early seventies, there has been a growing research interest in A. pullulans due to its products and properties that have proven or potential economic importance. These include an extracellular polysaccharide pullu lan , many hydrolytic enzymes, single cell proteins, to lerance to certain pollution levels and roles in biodeterioration of materials. An extensive literature review is available on the various economic aspects of the organism (Deshpande el ai, 1992).
The above mentioned literature review covered the global distrib ution of A. pullulans and recognised its widespread occurrence. However, it was stated that th e organism seemed to be most common in temperate zones with Britain and the United States being the sources of most reported isolations. A list of countri es (covering tem perate, mediterranean, arid and tropical zones) which reported the presence of A pullulans was presented, Egypt and South Africa were the only African countries mentioned. We are aware of some enzymatic studies carried out in South Africa on A. pullulans (for example, Myburgh et ai, 1991a and b) but the strain was obtained from an American Culture Co llection, Although the presence of A. puJlu/ans is at a very low frequency, only about 3% of a total of 27 ran dom samples of Nigerian cassava flour from one locality and none of a total 88 samples from seven other localities yielded the organism (Abba Kareem dases produced in broth cultures by selected isolates were generally low; yields ranged from 0.7-1.4 units/ml and 1,7-2,5 units/ml respectively. Of six (mainly complex) nitrogen sources, NaN03 supported the highest level of extracellular a-glucosidase in A. pullulans Mn· 2 despite a very low final pH (2.42). However, addition of Tween 80 to a 24 hour old culture led to a three-fold increase in yield of extracellular a-glucosidase (approx, 6.0U/ml). and Okagbue 1991), isolation of the organism from other sources has not been reported in the African hinterland.
Although strains of A. puflufans could diHer in certain physiological properties, they have shown consistency in ability to grow on some carbohydrates (Dennis and Buhagiar 1973) and in production of certain extracellular hydrolytic enzymes (Fedenci 1982), Their production of amylases, xylanases, (land B-g lucosidases have attracted much attention because the enzymes are potentially usefu l for converting plant polymeric biomass (especially starches and lignocellulosic materials) into single cell protem and biofuels. Therefore, relevant studies on the organism have been intensified in developed countries such as Italy and the United Siaies (see, for example, Federici and D'elia 1983, Saha el ai, 1993, Saha el ai, 1994. However, as most th ird world countries have agrobased economies which produce large quantities of plant biomass and as increased research on biomass conversion strategies have been recommended tor them (Anon 1979), it seemed appropriate that A. pullulans should be of research interest in th ird world countries of Sub-Saharan Africa. The aim of this study was to isolate A pullu/ans from Zimbabwean habitats and to explore the physiological diversity of the isolates including their ability to produce a-and 8-glucosidases.

Enrichment isolation of A.pullulans
Vegetable materials, especially leaves, were collected from certain plants/trees (Table 1) at the main campus of the Bulawayo Polytechn ic and subjected 10 an enrichment procedure described by Pollock ef al. (1992) . It involved incubation of the samples in minimal salts broth of the following composition (per litre of disti lled water) : 1 9 of (NH.),HPO . • ; 0. 5g of NaGI ; 0.05g of MgSO,.7H·O; 2g 01 K:H PO.; O.Olg each of FeSO., MnSO., and ZnSO. HGI to pH 7.0; and 109 sucrose in 250m I conical flasks and shaken in an orbital shaker at 25"G fo r two days. The enrichment promoted development of yeast-like cells wh ich remained mainly in suspension when the shaking was stopped . Subsequently, samples of the suspension we re inoculated onto agar plates containing the same minimal salts medium and incubated at the same tempe rature. Colonies wh ich developed after 3-4 days were visually and microscopically assessed for typical A. pulll.llans cha racteristics such as presence of septate mycel ia, bl astospores and chlamydospores; development or otherwise, of black/brown pigmentation as the co lon ies became ol der, were also noted (De Hoog 1998).

Representative colonies with 'desired' characteristics were
isolated by conventional methods and kept on potato dextrose agar (PDA) slants in the refri gerator.

Assimilation of carbon compounds by selected isolates
Twenty of ou r isolates were tested for the ability to assimilate carbon co mpounds by using the auxanographic method normally applied to yeasts (Barnetl et al. 1983) . The basal medium used was yeast nitrogen base (Difco) and incubations were at 25"C for 7 days. In our experience assimilation patterns of A. pullulans on carbon sou rces can be reliably assessed with auxonograms in a manner similar to that of well known yeasts which can be concluded in 7 days (Barnet et al. 1983).

Cultivation of selected A. pullulans isolates for production of extracellular glucosidases
Seven iso lates, arbit ra rily chosen, were cultivated in shake flasks. For a-g lucosidase production, 50ml of medi um (in duplicate 250m I Erl enmeyer flasks) 01 Ihe following composition (gi l): maltose, 10; NaNO" 2.0; FeSO •. 7H,0, 0.01 ; MgSO •. 7H,O, 0. 5; NaGI, 0.5; yeast extract (Oxoid), OA , was inoculated with a loopful of each organism taken from the  After four days, the yeast cells were removed by centrifugation at 1 800g for twenty minutes. Biomass was washed with distilled water, dried at 105"G for 4h and Ihen weighed in dry pre-weighed aluminium pans . The supernatan t was used as the crude enzyme preparation and was preserved with 0.2g/1 sodium azide if it was not immedia tely us ed and sto red at 4"C. The amount of biomass per ml of cu lture was determined and that fo rmed the basis for computing the amount of enzyme produced per gram biomass produced in the cu lture. For B~glucosidase production , the procedure was similar to Ihat of "-glucosidase production except that the medium contained in addition , KH,PO., 1 gil and bactocellobiose, 1 gil was substituted for maltose.

Enzyme assays
For (':(-and B-glu cosidase activities, respectively, the reaction mix tures contained 1ml of 1% maltose or 0 .1% bactocellobiose, lml 0150mM acetate buffer pH 4.5 or pH 5.0 and 1 ml of crude enzyme prepa ration . Incubations were at SouC for 30min. For both assays, the amount of glucose liberated during the incubation was ox idised using dinitrosalicyl ic acid (ON SA) reagent. One millilitre of DNSA reagent and 0.3ml of Rochelle salt (sodium-potassium tartarate) were added and the mixture was boiled for Sm in. After cooling to room tempe rature, the amount of redUCing sugar was measured by spectrophotometric absorbance readings at 57Snm and a glucose sta ndard curve was used. O ne unit of enzyme was defined as the amount of enzyme that liberated 0.01 mg of reducing sugar as glucose per minute under these conditions. Enzyme unit per gram biomass was computed based on the biomass per ml which was determined as has been explained ea rl ier.

Effect of nitrogen sources on production of extracellular a -glucosidase by A. pullulans Mn'
An inoculum cu ltUre was prepared with A. pullulans Mn ' as described above for B-glucosidase produ ction except that the medium contained Fe S04.7H20 at a concent ration of 0.001 gi l. Additionally, KH,PO. at 1.0g/1 was included and pH was adjusted to 5.0 using 1 M HGI. Two millilitres 01 the inoculum culture was used to inoculate each of the duplicate flasks containing 100ml of a basal fermentation medium (of the same composition as the inocu lum culture medium) su pplemented with one of the following nitrogen sources: NaNO" 2.0g/l; (NH.),SO., 2 .0g /l and peptone , tryptone, yeast extract and meat extract (Lab Lemco), each 5.0g/l. The fermentation flasks were incubated on the rotary shaker at 200rpm at 25'G for four days a~er which a-glucosidase activities and linal pH of the broths were determined.

Effect of TWeen 80 on production of extracellular a -glucosidases by A. pullulans Mn'
The procedure was the same as that in which the effecl of Soulh African Journal of Botany 2001, 67 157-160 nitrogen sources was exammed, except that Tween 80, a surfacta nt, was added (at 0.2% and 0.4% (vlv) levels) to 24 hourold duplicate fermentation cultures contalnmg NaNO) as the sole nitrogen source. A control, in duplicate, was also prepared to measure enzyme production in the absence of surfactants.

Results and Discussion
Our preliminary observations of A. pullulans in this laboratory were purely by chance. Initially the organism was found among yeasts growing on potato dextrose agar (PDA) plates inoculated with samples of jUice from locally grown grapes. Subsequently, the organism occurred in pla tings (on PDA) of ripe fruits of marula plant (Sclerocarya caffra) . Again, we obtained some A. pullulans isolates when samples of commercial manure as well as rotten avocado fruits (Persea americana) were plated out to recover probable deteriorative fungi. These chance isolations and absence of information on the occurrence of A. pullulans In this part of Africa (based on Deshpande el al. 1992) prompted us to undertake the screening exercise reported in this work to obtain some reasonably reliable information on its occurrence and on the possibility of its isolation under defined conditions. Interestingly, virtuaJly all the sources screened by enrichment yielded A. pullulans. The colonies resulting from the enrichment (including those obtained by chance as pointed out above) developed black to brown pigmentation in aged cultures and microscopy revealed septate hyphae, blastospores and chlamydospores; these attributes are consistent with the description of A. pullulans given by De Haag (1998). Further, to confirm the identity of the organisms we took cognisance of the characters which have been found useful for distinguishing A. pullulans from A. prunorum and Trichosporon pullulans (Dennis and Buhagiar 1973). Thus our isolates utilised glycerol (unlike T pullulans) and galactose (unlike A. prunorum).
Other substrates utilised for growth by our A. pullulans isolates were cellobiose, xylose , maltose , sucrose, pectin , xylan , dextrin, arbutin , glucuronic aCid and sodium acetate. While the responses of our isolates on most of these carbohydrates are in line with published observations (Dennis and Buhagiar 1973), utilisation of the last three substrates by A. pullulans does not seem to have been published. Unfortunately our isolates fa iled to assimilate salicin , inulin and lactose. and exhibited variable responses on arabitol and rib itol (Table 2). They also fa iled to utilise casein, gelatin , vanillin and tributyrin; these results are at variance with those reported for the strains exam ined by Federici el a/. (1982). It is possible that the use of auxanograms and seven day incubation period in this study had an adverse effect on the abilities of strains to assimilate the complex substrates (casein. gelatin. etc) which were readily utilised by aerobic shaken cultures.
As mentioned earlier, research on production of hydrolytic enzymes including ex-and l1-glucosidase, by A. pul/ulans, is being in tensified : the later enzymes mediate terminal hydrolytic activities involved in saccharification of starch and cellulose, respectively, for various commercial applications. Table 3 shows that all our selected isolates produced appreciable levels of the two enzymes. However, each organism 159

ThIS study
Denn is and Buhag lar appeared to produce a significantly higher level of 11-than of ex-glucosidase under the conditions employed in this study. Unfortunately, the levels of both enzymes produced by all the strains are generally low, the approximate maximum level of l1-glucosidase being only 2.5 U/ml of broth; this result is in agreement with the observations made on one strain of A. pullulans by Saha ef al. (1993Saha ef al. ( , 1994. It should be noted that the cited observations were made in what appeared to be pioneer studies on glucosidase activities of A. pullulans and that the workers also examined the effect of carbon sources on the production of the enzymes by the organism. To supplement thei r observations we have examined the effects of nitrogen sources and of a surfacta nt, Tween 80, on yields of a-glucosidase. Table 4 shows that although all the tested nitrogen sources supported appreciable levels of extracellular a-glucosidase production , the highest yield of the enzyme occurred in presence of NaNO] despite a significantly low final pH. This finding contradicts a study (Takii ef a/. 1995) in which two complex nitrogen sources (meat extract and pep-  tone) were used to optimise a-glucos idase production by a the rmophilic Bacillus sp. Further work would be needed to test the effect of combinations of complex nitrogen sources on peJiormance of A. pulfulans strains and to determine whether th e enzyme produced in presence of NaNO, would func1ion at the very low final pH (2.42). It was observed further in th is study ( Table 5) that addition of up to 0.4% Tween 80 to the cu lture medium containing NaNO, as the nitrogen source led to up to a three-fold increase in lhe extracelullar a -gl ucosidase achieved in the fe rmentation. Th is is consistent with the findings of Reese (1972) that surtactants caused several fungi to produce extracellularly, up to 20 fold increases of certain hydrolytic enzymes including cellulase, fJ-glucosidase and amylase. It was suggested that surtactan ts enhanced release of enzymes into the cultu re medium by increasing cell membrane permeability.
Overall , th is study has shown that A. pullulans occurs to an appreaciable extent on Zimbabwean vegetation (Tab le 1) and tha t it can be reliably isolated by enrichment; the observations extend existing knowledge (reviewed by Deshpande ef al. 1992) of the global distribution and micro-ecological habitats of the fungus. Also the study has confirmed that A. pullulans strains generally produce appreciable extracellular levels of 0:-and fJ-glucosidases and has shown that use of NaND:, as th e nitrogen source in culture medium an d additi on of Tween 80 (up to 0.4%) to 24 hour-old cu lture of A. pullulans may boost o:-glucosidase content of the broth.