Substrate-induced lipase gene expression and aflatoxin production in Aspergillus parasiticus and Aspergillus flavus

Authors


Jiujiang Yu, USDA/ARS, Southern Regional Research Center, 1100 Robert E. Lee Blvd., New Orleans, LA 70179, USA (e-mail: jiuyu@srrc.ars.usda.gov).

Abstract

Aims: To establish a relationship between lipase gene expression and aflatoxin production by cloning the lipA gene and studying its expression pattern in several aflatoxigenic and nontoxigenic isolates of Aspergillus flavus and A. parasiticus.

Methods and Results: We have cloned a gene, lipA, that encodes a lipase involved in the breakdown of lipids from aflatoxin-producing A. flavus, A. parasiticus and two nonaflatoxigenic A. flavus isolates, wool-1 and wool-2. The lipA gene was transcribed under diverse media conditions, however, no mature mRNA was detected unless the growth medium was supplemented with 0·5% soya bean or peanut oil or the fungus was grown in lipid-rich medium such as coconut medium. The expression of the lipase gene (mature mRNA) under substrate-induced conditions correlated well with aflatoxin production in aflatoxigenic species A. flavus (SRRC 1007) and A. parasiticus (SRRC 143).

Conclusions: Substrate-induced lipase gene expression might be indirectly related to aflatoxin formation by providing the basic building block ‘acetate’ for aflatoxin synthesis. No direct relationship between lipid metabolism and aflatoxin production can be ascertained, however, lipase gene expression correlates well with aflatoxin formation.

Significance and Impact of the Study: Lipid substrate induces and promotes aflatoxin formation. It gives insight into genetic and biochemical aspects of aflatoxin formation.

Introduction

Aflatoxins, produced predominantly by fungi such as Aspergillus flavus and A. parasiticus, are among the most potent natural carcinogens known. Contamination of agricultural commodities with aflatoxins are of serious health and economic concern worldwide (Jelinek et al. 1989). Tremendous progress has been recently made in aflatoxin research. In turn the discovery of the aflatoxin pathway gene cluster (Trail et al. 1995; Yu et al. 1995; Brown et al. 1996) led to a rapid identification of the majority of genes involved in aflatoxin biosynthesis (Yabe et al. 1989, 1998; Bhatnagar et al. 1992, 2002; Cleveland and Bhatnagar 1992; Chang et al. 1993, 1995; Payne et al. 1993; Yu et al. 1993, 1995, 1998, 2000a, b; Trail et al. 1995; Cleveland et al. 1997; Woloshuk et al. 1997; Payne and Brown 1998; Woloshuk and Prieto 1998).

Aflatoxin production is known to be influenced by many biotic and abiotic environmental factors such as temperature, water stress, pH, nutritional conditions (Adye and Mateles 1964; Dutton 1988; Payne and Brown 1998; Chang et al. 2000), and interaction between host and invading fungi. It is known that carbon and nitrogen sources play a vital role in the regulation of aflatoxin production (Adye and Mateles 1964; Dutton 1988; Payne and Brown 1998; Chang et al. 2000; Yu et al. 2000c; Aziz et al. 2002). Simple sugars such as glucose, sucrose, maltose and galactose induce aflatoxin production, while complex carbohydrates such as starch and peptone do not (Adye and Mateles 1964; Payne and Brown 1998). Glucose or sucrose has the same effect on inducing aflatoxin formation.

Aflatoxin biosynthesis is a complex process. The aflatoxin pathway gene cluster within a 80-kb DNA region in A. flavus and A. parasiticus (Yu et al. 1995; Cleveland et al. 1997) are under the tight control of a positive regulatory gene aflR (Chang et al. 1993, 1995; Payne et al. 1993; Ehrlich et al. 1999) and aflJ (Meyers et al. 1998; Chang and Yu 2002). The genetic regulation of aflatoxin production in response to environmental and nutritional factors however remains poorly understood as is the regulation and interaction of the aflR gene with other pathway regulators or signals (Chang et al. 1995; Payne and Brown 1998).

Although the relationship between lipids and aflatoxin production (Fanelli et al. 1983; Fanelli and Fabbri 1989; De Luca et al. 1995; Fanelli et al. 1995; Aziz et al. 2002) and lipase activity in nonaflatoxigenic fungi (Ohnishi 1982a, b; Yamaguchi et al. 1991; Tsuchiya et al. 1996) have been reported, no expression of lipase genes in aflatoxigenic fungi was reported. The effects of lipids on fungal growth and aflatoxin production have been studied in A. flavus and A. parasiticus by many researchers (Fanelli et al. 1983, 1995; Fanelli and Fabbri 1989; De Luca et al. 1995) who demonstrated that lipophilic epoxy fatty acids stimulated aflatoxin production in toxigenic fungi (Fanelli et al. 1983). Ergosterol oxidation was also found to induce both fungal growth and aflatoxin production (De Luca et al. 1995). Saturated free fatty acids support fungal growth and aflatoxin production, while the unsaturated free fatty acids inhibit fungal growth (Fanelli and Fabbri 1989). Lipoperoxidation of unsaturated free fatty acids produces lipoperoxides, which are presumed to inhibit fungal growth and reduce aflatoxin biosynthesis (Fanelli and Fabbri 1989).

Lipases are lipolytic enzymes produced by nonaflatoxigenic fungi closely related to A. flavus such as A. oryzae (Ohnishi 1982a, b; Tsuchiya et al. 1996) and other fungi not related to A. flavus such as Penicillium camembertii (Yamaguchi et al. 1991) which by hydrolysis of triglycerides modify fungal lipid composition. In order to establish the linkage between lipid metabolism and aflatoxin production at a molecular level, we have cloned a gene, lipA, encoding a lipase from aflatoxigenic strains A. parasiticus SRRC 143 and A. flavus SRRC 1007. In addition, lipA from two nonaflatoxigenic A. flavus isolates, wool-1 and wool-2, demonstrating the ability to degrade wool possibly by proteases and lipases, was also cloned in this study. The expression patterns of lipA in the two aflatoxigenic species, the two nonafltoxigenic A. flavus isolates, and an A. parasiticus nonaflatoxigenic aflR-disrupted mutant CS10N2-disA(Takahashi et al. 2002) were studied under aflatoxin-inducive and noninducive medium conditions.

Materials and methods

Fungal strains and culture conditions

Fungal strains used in this study were the following: (i) A. parasiticus SRRC 143 (ATCC 56775) which produces aflatoxins B1, B2, G1 and G2; (ii) A. flavus SRRC 1007 which produces aflatoxins B1 and B2; (iii) two nonaflatoxigenic A. flavus isolates, wool-1 and wool-2, respectively, which do not produce any aflatoxin or aflatoxin intermediates and are capable of degrading wool; and (iv) an aflR-disrupted mutant strain of A. parasiticus SRRC 143, CS10N2-disA, in which the aflatoxin pathway positive regulatory gene, aflR, was disrupted. As a result, no aflatoxin pathway genes are expressed and no aflatoxin is produced (in this aflR-disabled mutant). Fungal strains were maintained on potato dextrose agar plates (Difco Laboratories, Detroit, MI, USA). For conidial production, spores were inoculated on Petri dishes (100 × 15 mm style; Falcon, Becton Dickinson and Company, Franklin Lakes, NJ, USA) containing 20 ml of V8 agar solid medium [50 ml l−1 V8 juice, a commercial beverage (Campbell Soup Company, Camden, NJ, USA, http://www.v8juice.com] consisting of eight vegetable juices, 20 g l−1 agar, pH 5·2). The cultures were incubated at 30°C temperature. For detection of gene expression and aflatoxin production, fungal mycelia were grown as submerged cultures in aflatoxin-inducive GMS medium [containing (g l−1): sucrose, 50; ammonium sulphate, 3; potassium phosphate, 10; magnesium sulphate, 2; micronutrient mixture; 1 ml, pH 4·5] (Adye and Mateles 1964) and nonaflatoxin-inducive peptone mineral salt (PMS) medium in which sucrose was replaced with an equal amount of peptone (Adye and Mateles 1964). For detection of lipA gene expression under lipid-induced conditions, the sucrose mineral salt (GMS) and PMS media were supplemented with 0·5% soya bean or peanut oil. The coconut medium (CAM) was reported to support aflatoxin production (Davis et al. 1987; Dyer and McCammon 1994) like sucrose or glucose. Coconut medium is a lipid-rich medium and was also used to induce lipA gene expression. The CAM was made by blending 200 g of commercially sold solid coconut with 400 ml hot distilled water. The soluble coconut nutrients were retained after filtering out undissolved material by passage through Miracloth, the resulting liquid filtrate was adjusted to pH 7·0 using 3 m NaOH, and after sterilization was used to support submerged fungal growth. Growth media for the aflR-disrupted mutant (CS10N2-disA) was supplemented with uracil at a final concentration of 0·1 mg ml−1 as the pyrG gene was also disrupted to create marker selection system for fungal transformation in this mutant. All liquid cultures were grown for 72 h at 30°C with constant shaking at 150 rev min−1.

Cloning and DNA sequencing

The lipase gene, lipA, was cloned from A. parasiticus SRRC 143 and A. flavus SRRC 1007 by polymerase chain reaction (PCR) using primers designed based on a homologue of lipA in A. oryzae, mdlB (accession no. D85895). The primers used for lipA gene cloning are, forward primer (lipF2): 5′-CCAAGCTTTGCAACCAAGCCTGTCG-3′ and reverse primer (lipR2): 5′-CTGCAGGTGTAGTGTGCTTGGCCGA-3′. The corresponding cDNA sequences were also cloned by reverse transcriptase PCR (RT-PCR) using primers designed at the putative translation start and stop sites. The primers are, forward primer (lipF1): 5′-GATGCGCTTCCTCTCCGGCTTCG-3′ and reverse primer (lipF1): 5′-TTAGCGGAAGGGCAATCCATGAC-3′. The genomic and cDNA sequences of lipA gene were determined by direct sequencing of the PCR products. The two pairs of primers were also used for genomic DNA (PCR products) and cDNA (RT-PCR products) sequencing as sequencing primers. All of the primers used in this study were made with a DNA synthesizer (model 380A; Applied Biosystems, Foster City, CA, USA).

RNA purification and RT-PCR experiments

All of the total RNAs were isolated from 72 h mycelia of the fungal strains in respective liquid media (GMS and PMS) using RNeasy Total RNA Kit (Qiagen Inc., Valencia, CA, USA) from the mycelia. The mRNA was purified subsequently using PolyATract mRNA isolation System (Promega Corp., Madison, WI, USA). First strand cDNA was synthesized using Advantage RT-for-PCR Kit (Clontech, Palo Alto, CA, USA). The first strand cDNA was used as the template in RT-PCR experiment to detect lipA mRNAs under different medium conditions. RT-PCR was performed with denaturing at 94°C for 15 s, annealing at 56°C for 30 s, extension at 72°C for 2 min, 30 cycles with a final extension at 72°C for 5 min. PCR products were separated in 1% agarose gel under 80 V for 90 min.

Aflatoxin extraction and thin-layer chromatography

Aflatoxins were extracted from the fungal mycelia grown in different liquid media after 72 h of growth at 30°C with an acetone/chloroform solvent system (1 : 1 ratio, v/v; Dutton 1988). The extracted aflatoxin metabolites were assayed by loading 3 μl of a 300-μl sample onto thin-layer chromatography (TLC) plates (catalog no. 7001-04, silica gel; J. T. Baker, Inc., Phillipsburg, NJ, USA) and subsequent separation achieved with an ether : methanol : water (96 : 3 : 1, v/v/v) solvent system. The solvent separated aflatoxins were visualized under 365 nm long wave ultraviolet and the amount of aflatoxins was quantified visually by comparing the brightness of the spots with a series of standard samples of known concentration by densitometry. The aflatoxin standards were purchased from Sigma Co. (St Louis, MO, USA) and replicate analyses were performed.

GenBank accession numbers

The lipA gene sequence data from A. parasiticus SRRC 143 and from A. flavus SRRC 1007 have been submitted to GenBank databases under accession numbers AF404488 and AF404489, respectively (http://www.ncbi.nlm.nih.gov).

Results

The lipA gene encodes a polypeptide consisting of 306 amino acid residues

The lipA gene from A. parasiticus SRRC 143 and A. flavus SRRC 1007 was cloned by PCR using primers based on a homologue of lipA in A. oryzae, mdlB (Tsuchiya et al. 1996). The genomic DNA sequences from A. parasiticus and A. flavus were found to be 1654 and 1665 bp, respectively. The corresponding cDNA sequences were also cloned by RT-PCR using primers designed at the putative translation start and stop sites. Two intervening sequences of 52 bp each were found near the N-terminal region. The coding region consists of 921 bp, capable of encoding 306 amino acids (AA) (Fig. 1) with a calculated molecular mass of 33·5 kDa. There were 27 nucleotide substitutions and deletions within the coding region of the lipase from A. flavus compared with that of A. parasiticus. However, only five of them resulted in changes of AA residues (Fig. 1). The genomic DNA sequences of A. flavus lipA and A. oryzae mdlB are identical (Tsuchiya et al. 1996) (Fig. 1) probably because these fungal species are highly related. The homology of the AA sequences encoded by lipA between A. parasiticus and A. flavus (A. oryzae) is 98·7%. However, a relatively significant degree of homology (62·7%) at the AA level was also found between A. parasiticus and a nonrelated fungal species P. camembertii (Yamaguchi et al. 1991; Fig. 1).

Figure 1.

Comparisons of the lipase amino acid sequences from Aspergillus parasiticus, A. flavus, A. oryzae and Penicillium camembertii. A.p., A.f., A.o. and P.c. are abbreviations for sequences from A. parasiticus SRRC 143, A. flavus SRRC 1007, A. oryzae, and P. camembertii, respectively. The numbers to the left indicate the amino acid (AA) sequence starting from the translation start methionine (M). Identical AA are represented by a dash (–) and a gap is represented by a dot (.). The translation stop codon is represented by a star (*). The first AA residue of putative mature enzyme (Tsuchiya et al. 1996) at position 29 (D) is printed in boldface letter and underlined. The conserved AA sequence motif for lipolytic enzymes (GXSXG) is underlined and the conserved AA residues are printed in boldface letters. The serine (S), aspartic acid (D) and histidine (H), that form a catalytic triad, are printed in boldface letters and double underlined. The putative N-glycosylation site (Tsuchiya et al. 1996) at position 252 is also underlined

Expression of lipA gene under lipid-induced condition

The functional expressions of the lipA gene of these Aspergillus species under several nutritional conditions were studied by RT-PCR experiments. When A. parasiticus and A. flavus mycelia were grown in aflatoxin-inducive (GMS) and nonaflatoxin-inducive (PMS) media, no mature lipA mRNA (921 bp) was detected from mycelium grown in either medium (Table 1). When the GMS or PMS growth media were supplemented with 0·5% soya bean oil or peanut oil, the mature lipA mRNAs (921 bp) were readily detected (Table 1; Fig. 2a,b ). The genomic and cDNA sequences of lipA from the two A. flavus isolates, wool-1 and wool-2, were shown to be identical to that of A. flavus SRRC 1007 (data not shown). The expression pattern of lipA from these two isolates was also the same as that of A. parasiticus and A. flavus (Fig. 2a, b) in that in the presence of oil the mature mRNAs were readily detected. The CAM is oil rich and often used for aflatoxin production assays (Arseculeratne et al. 1969). When the fungal mycelia were grown in CAM, the lipA mature mRNAs from A. parasiticus SRRC 143 and A. flavus SRRC 1007 were detected (Fig. 2c). When the two nonaflatoxingenic A. flavus isolates, wool-1 and wool-2, were grown in the CAM, the lipA mature mRNAs were detected, but the level was significantly reduced based on the band intensity of RT-PCR amplification. The mature mRNAs detected from 72 h mycelia of the two A. flavus isolates, wool-1 and wool-2, were (estimated by visualization and by densitometry measurement) ca 20 and 60%, respectively, of the wild type strain A. flavus SRRC 1007 (Fig. 2c) grown under similar conditions.

Table 1.  Expression of lipA (mature mRNA) and aflatoxin production in Aspergillus under different nutritional media conditions
StrainsMedia*lipA Mature mRNA detected (%)Aflatoxins producedRelative amount of aflatoxins (%)
  1. *Fungal mycelia were grown in these media with or without 0·5% oil for 72 h except noted otherwise.

  2. †Oil was added in the concentration of 0·5% to the media at the time of inoculation except noted below.

  3. ‡PMS medium for 48 h, then adding 0·5% oil and growing for additional 24 h.

A. parasiticus SRRC 143GMSNot detectedB1,B2,G1,G2100
A. parasiticus SRRC 143GMS + oil†100B1,B2,G1,G2100
A. parasiticus SRRC 143PMSNot detectedNo0
A. parasiticus SRRC 143PMS + oil100B1,B2,G1,G260
A. parasiticus SRRC 143PMS (48 h)  + oil (24 h)‡50B1,B2,G1,G220
A. parasiticus SRRC 143Coconut100B1,B2,G1,G2100
A. flavus SRRC 1007GMSNot detectedB1,B2100
A. flavus SRRC 1007GMS + oil100B1,B2100
A. flavus SRRC 1007PMSNot detectedNo0
A. flavus SRRC 1007PMS + oil100B1,B210
A. flavus SRRC 1007Coconut100B1,B2100
A. flavus isolate wool-1GMSNot detectedNo0
A. flavus isolate wool-1GMS + oil100No0
A. flavus isolate wool-1PMSNot detectedNo0
A. flavus isolate wool-1PMS + oil100No0
A. flavus isolate wool-1Coconut20No0
A. flavus isolate wool-2GMSNot detectedNo0
A. flavus isolate wool-2GMS + oil100No0
A. flavus isolate wool-2PMSNot detectedNo0
A. flavus isolate wool-2PMS + oil100No0
A. flavus isolate wool-2Coconut60No0
A. parasiticus CS10N2-disAGMSNot detectedNo0
A. parasiticus CS10N2-disAGMS + oil100No0
A. parasiticus CS10N2-disAPMSNot detectedNo0
A. parasiticus CS10N2-disAPMS + oil100No0
A. parasiticus CS10N2-disACoconut100No0
Figure 2.

Detection of the lipA transcript by RT-PCR. RT-PCR was performed to detect lipA mature mRNAs under different nutritional conditions from A. parasiticus SRRC 143, A. flavus SRRC 1007, and the two nonaflatoxigenic A. flavus isolates, wool-1 and wool-2, respectively. (a) Fungal mycelia grown in GMS medium supplemented with 0·5% soya bean oil. (b) Fungal mycelia grown in PMS medium supplemented with 0·5% soya bean oil. The lipase gene (lipA) was transcribed and mature mRNAs were detected (lanes c), but with less intensity compared with GMS medium supplemented with oil; and (c) fungal mycelia grown in coconut medium. The lipase gene (lipA) was transcribed and mature mRNAs were detected (lanes c). A.p. 143 represents PCR products amplified from A. parasiticus SRRC 143; A.f.1007 represents PCR products amplified from A. flavus SRRC 1007; A.f. wool-1 represents PCR products amplified from A. flavus isolate wool-1; A.f. wool-2 represents PCR products amplified from A. flavus isolate wool-2. Lane ‘g’ represents genomic PCR products amplified from genomic DNA purified from the respective fungal mycelia. Lane ‘c’ represents RT-PCR products amplified from first strand cDNA of respective fungal strain

Independent expressions of the lipA gene and aflatoxin pathway genes

Equivalent amounts of aflatoxins were produced in both GMS medium and GMS medium containing 0·5% soya bean oil (Fig. 3, lanes 1 and 2, and lanes 6 and 7). Furthermore aflatoxin was detected in nonaflatoxin-inducive medium (PMS) when supplemented with 0·5% soya bean oil (Fig. 3, lanes 4 and 9). However, unlike aflatoxin pathway genes, the lipA gene is expressed and processed in the presence of peptone in the PMS medium as long as it was supplemented with 0·5% soya bean oil. The mature lipA mRNA was detected in both aflatoxin-inducive (GMS) and nonaflatoxin-inducive (PMS) media when supplemented with 0·5% soya bean oil. In the nonaflatoxigenic A. flavus isolates, wool-1 and wool-2, no aflatoxin was produced irrespective of medium utilized. When the two isolates were grown in either GMS or PMS medium containing 0·5% oil, the lipA gene was transcribed and mature mRNA was detected (Table 1).

Figure 3.

Thin-layer chromatographic (TLC) assay of aflatoxin production in different media conditions. Aflatoxins were extracted from a mixture of 72 h mycelia and the growth medium. The extracted aflatoxins were loaded onto TLC plates. TLC plates were developed in an ether : methanol : water (96 : 3 : 1, v/v/v) solvent system. A total of 20 μg of known compounds of aflatoxin (B1, B2, G1 and G2) standards (AF Std) were loaded onto the TLC plates for quantitating aflatoxins produced in the various media conditions. Lanes 1–5 are A. parasiticus SRRC 143 and lanes 6–9 are A. flavus SRRC 1007. Lanes 1 and 6 are GMS, lanes 2 and 7 are GMS supplemented with 0·5% soya bean oil, lanes 3 and 8 are PMS, and lanes 4 and 9 are PMS supplemented with 0·5% soya bean oil. Lane 5 is A. parasiticus SRRC 143 grown in PMS for 48 h and then adding 0·5% soya bean oil and continuing growth for additional 24 h

The aflR, a positive regulatory gene, is responsible for the activation of the aflatoxin biosynthetic pathway genes (Chang et al. 1995). In order to investigate whether aflR exerts any influence on the expression of the lipA gene, production of mature mRNA by the aflR-disrupted mutant strain of A. parasiticus CS10N2-disA (Takahashi et al. 2002) was assessed. In this mutant, no aflatoxin pathway genes were transcribed and no aflatoxins were produced. RT-PCR of lipA gene, using primers which function effectively with A. parasiticus CS10N2-disA, demonstrated that the lipA gene was transcribed under all medium conditions used and that mature mRNA was detected only when 0·5% oil was supplemented in the growth media (Table 1). In other words, when the aflatoxin pathway regulatory gene, aflR, was disabled, lipA gene expression was not affected. No aflR binding motif (TCGxxxxxCGA; Ehrlich et al. 1999) was identified in the promoter region of the lipA gene sequence. Therefore, it is unlikely that the transcription of the lipA gene is under the regulation of aflR, as is the case with the majority, if not all, of the aflatoxin pathway genes (Chang et al. 1995; Ehrlich et al. 1999). Expression of lipA is induced by lipid, while aflatoxin-inducive or nonaflatoxin-inducive media alone did not support the expression of lipA. However, the aflatoxin pathway gene expression and eventually aflatoxin production is not totally dependent on lipid induction (Table 1; Fig. 3) as hexoses such as glucose and sucrose are known to be very good supporters of aflatoxin formation. The results demonstrated that lipid was required for the expression of lipA, but aflatoxin production did not necessarily require lipid. Lipases are substrate-induced enzymes. The metabolite ‘acetate’ may be produced during hydrolysis of lipids by lipases. Acetate is believed to be the earliest aflatoxin precursor. The aflatoxin production in response to lipase gene expression might be indirectly related to toxin synthesis through providing the building block ‘acetate’ for aflatoxin biosynthesis. It seems, therefore, that the expression of lipA gene has no direct relationship with the expression of aflatoxin pathway genes (aflR) and so no direct relationship exists between lipid metabolism and aflatoxin formation.

Lipid induces, supports or promotes aflatoxin production

Aflatoxigenic fungal strains grown in the different media (GMS, PMS and each supplemented with 0·5% oil) were assessed for the level of aflatoxin production. The amount of aflatoxin produced in GMS medium was treated as 100% (Table 1; Fig. 3, lanes 1 and 6). When the fungus was grown in the GMS medium supplemented with 0·5% soya bean oil, no noticeable increase in aflatoxin production was detected (Table 1; Fig. 3, lanes 2 and 7). When the fungus was grown in the PMS medium, no aflatoxin was detected (Table 1; Fig. 3, lanes 3 and 8). When this PMS medium was supplemented with 0·5% soya bean oil, marked aflatoxin production was experienced (Fig. 3, lane 4, Fig. 3, lane 9) albeit not as great as supported by GMS medium. The results demonstrated that the fungus may use oil as carbon source and that lipid may also promote the expression of aflatoxin pathway genes and support aflatoxin formation albeit at low efficiency compared with glucose or sucrose rich media.

Fungal mycelia from PMS medium (nonaflatoxin-inducive condition) supplemented with 0·5% oil, produced aflatoxins within 24 h (Fig. 3, lane 5). The whole process of signal transduction involves many genetic and biochemical steps. At a minimum consisting of lipid-induced lipA gene transcription, transcript processing, lipid degradation, induction of aflatoxin pathway gene expression and finally aflatoxin biosynthesis. Approximately 5–10% of aflatoxins (Fig. 3, lane 5) were accumulated within a 24-h culture compared with a 72-h culture grown in the aflatoxin-inducive media (Fig. 3, lane 1 or 2). When the fungal mycelia is grown in the PMS medium supplemented with 0·5% oil for 72 h (Fig. 3, lane 4), the amount of aflatoxins produced is about 60% of that produced by growing in the GMS medium for the same duration (Fig. 3, lane 1 or 2). In contrast fungal mycelia grown in the GMS medium, and GMS supplemented with 0·5% oil, revealed no significant difference in the amount of aflatoxin production (Fig. 3, lanes 1 and 2, lanes 6 and 7) between species and supplementation or lack of supplementation.

Discussion

The first AA residue of mature lipase enzyme in the closely related species A. oryzae and also in the nonrelated species P. camembertii was reported to be an aspartic acid (Tsuchiya et al. 1996, accession no. D85895; Yamaguchi et al. 1991, accession no. D90315). This aspartic acid reside at position 29 (Fig. 1 printed in boldface letter and underlined) was identified as N-terminal residue of putative mature lipase in A. parasiticus and A. flavus. Excluding the 28 AA of leader sequence, the putative mature enzyme should consist of 278 AA with a calculated molecular mass of 30·4 kDa. It was reported (Toida et al. 1995) that the purified lipase protein (L2) from A. oryzae was between 39 and 41 kDa. It is possible that the molecular weight discrepancy occuring between purified native mature enzyme and the deduced molecular mass was due to glycosylation of the enzyme. A consensus N-glycosylation site (NTTV) was also identified in this lipase from A. parasiticus and A. flavus to be at position 252 (Fig. 1; Yamaguchi et al. 1991; Tsuchiya et al. 1996). The conserved AA sequence motif (GXSXG) at position 171–175 and a catalytic triad consisting of serine (S-173), aspartic acid (D-226) and histidine (H-288) for lipolytic enzymes were identified as well (Fig. 1; Yamaguchi et al. 1991; Tsuchiya et al. 1996).

BLAST search against the NCBI (http://www.ncbi.nlm.nih.gov) GenBank database and A. nidulans genome database (http://www-genome.wi.mit.edu/annotation/fungi/aspergillus/) for the cloned lipA gene sequence demonstrated that no significant sequence homology has been identified in the database at nucleotide and AA levels. However, we identified an A. nidulans lipA gene (accession no. AF424740) encoding for triacylglycerol lipase in the NCBI database with no sequence homology between the lipA in A. parasiticus, A. flavus and A. nidulans. The lipA gene in A. nidulans is probably doing the same function as our cloned example.

It was demonstrated that supplementation of lipid (0·5% oil) in the nonaflatoxin-conducive medium (PMS) promoted aflatoxin biosynthesis (Fig. 3, lanes 5 and 9). Aflatoxin biosynthesis involves over 20 aflatoxin pathway genes and their encoded enzymes. We postulate that the function of the lipids could be either as a carbon source or as an inducer for aflatoxin production. However, unlike sucrose or glucose (GMS medium), the lipids are not the preferred carbon source for aflatoxin production. These data do suggest that lipase or specific lipase actions involved in lipid metabolism may be important in promoting aflatoxin formation, possibly during fungal invasion of lipid-rich environments (such as seeds of corn, cotton and peanuts). The cloning and expression of lipase gene may further help our understanding of fungal invasion, the complex regulation of aflatoxin biosynthesis, and the potential existence of signal transduction between primary and secondary metabolism.

Acknowledgements

The A. flavus isolates wool-1 and wool-2 were kindly provided by Dr Jeanette M. Cardamone (USDA/ARS, Eastern Regional Research Center, Wyndmoor, PA, USA). We are grateful to Drs Kenneth C. Ehrlich and Perng-Kuang Chang for critical reviews of this manuscript. We thank Michelle Faust for technical assistance and Janell Becker for secretarial help.

Ancillary