Aflatoxin biosynthesis cluster gene cypA is required for G aflatoxin formation

Abstract

Aspergillus flavus isolates produce only aflatoxins B1 and B2, while Aspergillus parasiticus and Aspergillus nomius produce aflatoxins B1, B2, G1, and G2. Sequence comparison of the aflatoxin biosynthesis pathway gene cluster upstream from the polyketide synthase gene, pksA, revealed that A. flavus isolates are missing portions of genes (cypA and norB) predicted to encode, respectively, a cytochrome P450 monooxygenase and an aryl alcohol dehydrogenase. Insertional disruption of cypA in A. parasiticus yielded transformants that lack the ability to produce G aflatoxins but not B aflatoxins. The enzyme encoded by cypA has highest amino acid identity to Gibberella zeae Tri4 (38%), a P450 monooxygenase previously shown to be involved in trichodiene epoxidation. The substrate for CypA may be an intermediate formed by oxidative cleavage of the A ring of O-methylsterigmatocystin by OrdA, the P450 monooxygenase required for formation of aflatoxins B1 and B2.

FIG. 1.

Structures of aflatoxins B1, B2, G1, and G2 and the aflatoxin precursors, OMST and 11-hydroxyOMST (11-OH-OMST).

FIG. 2.

Characteristics of the norB-cypA region in different Aspergillus species. (A) Schematic diagram of the norB-cypA sequences of different aflatoxin biosynthesis gene cluster homologs. Thick arrows indicate coding regions and direction of transcription of norB and cypA. Gaps represent deletions of 32 and 854 bp in A. flavus NRRL3357 and AF13, and 1516 bp in A. flavus AF70 and A. oryzae ATCC12892, ATCC46244, and FRR2874 when the sequences are compared to the sequence of A. parasiticus in this region. Additional smaller deletions or insertions are marked by asterisks (11 bp in A. parasiticus NRRL2999 at bp 1166, 13 bp in BN008R, and 4 bp in A. nomius NRRL13137). The positions of oligonucleotide primers AP1729 and AP3551 are indicated by small arrows. (B) Agarose gel (1.0%) electrophoresis of PCR fragments obtained by amplification of different Aspergillus DNAs with primers AP1729 and AP3551. Abbreviations: AF, A. flavus; AP, A. parasiticus; AO, A. oryzae; AN, A. nomius.

FIG. 3.

Preparation and characterization of cypA disruptants. (A) Schematic of the cypA disruption plasmid, pCypDV, and the expected product obtained by homologous recombination. Directions of transcription of norB, cypA, and aflT are shown as thick arrows. The insert niaD cassette DNA (shaded box) was obtained as a XbaI fragment from pSL82 (7). P1 to P6 are approximate annealing sites for the oligonucleotide primers used to test for insertion of the XhoI/SphI-digested plasmid fragment in transformants. K, KpnI; X, XhoI; S, SphI; Xb, XbaI. (B) A representative silica gel TLC profile of aflatoxins produced by pCypDV transformants. Locations of authentic aflatoxin standards are indicated on the right. Aflatoxins were visualized under 366-nm light and the negative image is shown. T2 and T12 are transformants that do not produce G aflatoxins; T8 produces both B and G aflatoxins. (C) Bands on a 1.0% agarose gel after electrophoresis of PCR products obtained with oligonucleotide primer pairs P1-P2, P3-P4, and P5-P6 used to check cypA disruption transformants. The positions of the primers are shown in panel A.

FIG. 4.

Northern hybridization analysis of A. parasiticus and A. flavus RNAs with probes for the genes norB and cypA. A. parasiticus BN009E transformant T8, the cypA-knockout transformant, T2, A. flavus AF13, and A. flavus AF70 were grown on peptone minimal salts (P) and glucose minimal salts (G) medium for 3 days and 20 μg of total RNA was separated on a 0.4 M formaldehyde-1.2% agarose gel and transferred to a Nytran⁺ membrane for probing with radiolabeled norA and cypA probes. The approximate sizes of the bands are shown on the right side of the blots.

FIG. 5.

Alignment of CypA protein sequences from isolates of three Aspergillus species that produce both B and G aflatoxins and Tri4 from G. zeae, GenBank accession number AAM48924. Locations of presumed consensus P450 catalytic and other domains are shown as bars under the sequence. Shaded regions represent locations of amino acid identity in the sequences. Basic amino acids adjacent to the P-X-P motif and the heme-binding loop are indicated by arrows. The different helix domains are indicated by letters and brackets above the sequence.

FIG. 6.

Scheme showing the possible conversion of OMST to aflatoxins B1 and G1 (the steps would be similar for the conversion of dihydro-OMST to aflatoxin B2 and G2). The conversion of OMST to aflatoxin B1 is predicted to involve formation of an A-ring-opened intermediate, structure 1, as suggested by Udwary et al. (29). Oxidation of this intermediate by CypA is predicted to give the epoxide which is subsequently converted to the G aflatoxins.