Melanin biosynthesis in the maize pathogen Cochliobolus heterostrophus depends on two mitogen-activated protein kinases, Chk1 and Mps1, and the transcription factor Cmr1

Abstract

The maize pathogen Cochliobolus heterostrophus requires two mitogen-activated protein kinases (MAPKs), Chk1 and Mps1, to produce normal pigmentation. Young colonies of mps1 and chk1 deletion mutants have a white and autolytic appearance, which was partially rescued by a hyperosmotic environment. We isolated the transcription factor Cmr1, an ortholog of Colletotrichum lagenarium Cmr1 and Magnaporthe grisea Pig1, which regulates melanin biosynthesis in C. heterostrophus. Deletion of CMR1 in C. heterostrophus resulted in mutants that lacked dark pigmentation and acquired an orange-pink color. In cmr1 deletion strains the expression of putative scytalone dehydratase (SCD1) and hydroxynaphthalene reductase (BRN1 and BRN2) genes involved in melanin biosynthesis was undetectable, whereas expression of PKS18, encoding a polyketide synthase, was only moderately reduced. In chk1 and mps1 mutants expression of PKS18, SCD1, BRN1, BRN2, and the transcription factor CMR1 itself was very low in young colonies, slightly up-regulated in aging colonies, and significantly induced in hyperosmotic conditions, compared to invariably high expression in the wild type. These findings indicate that two MAPKs, Chk1 and Mps1, affect Cmr1 at the transcriptional level and this influence is partially overridden in stress conditions including aging culture and hyperosmotic environment. Surprisingly, we found that the CMR1 gene was transcribed in both sense and antisense directions, apparently producing mRNA as well as a long noncoding RNA transcript. Expression of the antisense CMR1 was also Chk1 and Mps1 dependent. Analysis of chromosomal location of the melanin biosynthesis genes in C. heterostrophus resulted in identification of a small gene cluster comprising BRN1, CMR1, and PKS18. Since expression of all three genes depends on Chk1 and Mps1 MAPKs, we suggest their possible epigenetic regulation.

FIG. 1.

Schematic representation of the fungal DHN-melanin biosynthesis pathway. PKS18 (polyketide synthase), BRN2 (T4HN reductase), BRN1 (T3HN reductase), SCD1 (scytalone dehydratase), and p-diphenol oxidase are enzymes presumably involved in the indicated biosynthetic steps.

FIG. 2.

CMR1 gene features and domain architecture of the protein product.

FIG. 3.

Characterization of the cmr1 mutant phenotype. (A) Colonies of wild type and Δcmr1 mutant; (B) Δcmr1 mutants form light brown pseudothecia when crossed with one another and with the wild-type strain; (C) growth inhibition of mutant and wild-type strains as a result of UV irradiation. Wild-type and mutant conidia were allowed to attach to the slides (45 min); mycelial debris and conidia that remained unattached were washed off. Following UV irradiation germ tubes of the mutant stopped further growth. The energy of irradiation is indicated. wt, wild type.

FIG. 4.

Pigmentation and autolytic phenotypes. Five-day-old stationary liquid cultures of Δcmr1, Δchk1, and Δmps1 mutants and the wild-type strain (WT) were grown in complete medium with xylose (CMX) with or without addition of 1.5 M sorbitol.

FIG. 5.

RT-PCR analysis of expression of the melanin biosynthesis genes BRN1, BRN2, and SCD1 (A) and PKS18 (B) in Δcmr1, Δchk1, and Δmps1 mutants and the wild-type strain (wt). The stationary liquid cultures were grown for 3 and 5 days in CMX (3d and 5d, respectively) or 5 days in CMX supplemented with 1.5 M sorbitol (srb). Expression of the actin-encoding gene (ACT) indicates relative RNA quantities in each sample. ACT₁ and ACT₂ indicate different primer pairs used for PCR (Table 1).

FIG. 6.

Expression of sense and antisense CMR1. (A) Semiquantitative RT-PCR analysis of CMR1 gene expression in wild type and Δchk1 and Δmpk1 mutants grown as described in the Fig. 5 legend. Polyadenylated mRNA was primed with standard oligo(dT) primer. Two primer pairs were used for the subsequent PCR: ctrl-s and V-a, and s and a (Table 1). Expression of the actin-encoding gene (ACT₂) indicates relative RNA quantities in each sample. (B) Schematic representation of the primer and intron positions as related to the CMR1 open reading frame (ORF). Small arrows indicate primers used for RT and subsequent PCR. Open squares denote introns in the sense CMR1 transcript, whereas the filled square indicates an intron in the antisense CMR1. (C) Analysis of sense and antisense CMR1 transcript levels by directional cDNA synthesis. RNA used for RT-PCR was prepared from wild type and Δchk1 and Δmps1 mutants grown for 3 days in CMX, as described in the Fig. 5 legend. To examine antisense CMR1 expression, cDNA was synthesized with CMR1 primer s and actin primer act2-a to serve as an internal control. For sense CMR1 expression, RT was performed with CMR1 primer a and actin primer act2-a. Primers s and a (and act2-s and act2-a for actin) were used for the subsequent PCRs. Note that for detection of the antisense CMR1 34 PCR cycles were required, compared to 28 cycles for the sense transcript. wt, wild type.

FIG. 7.

Construction of CMR1p-GFP reporter strain and analysis of CMR1 promoter activity in the absence of Cmr1 protein. (A) CMR1p-GFP transformation vector. Abbreviations: Ptrp, constitutive promoter; NAT, nourseothricin acetyltransferase-encoding gene; p-XhoI-s, p-BamHI-a, 3′-BglI-s, 3′-SacI-a, GFP-BamHI-s, and GFP-SpeI-a, primers used for vector construction. (B) RT-PCR analysis of GFP reporter expression in parental CMR1p-GFP strain and in mps1-deleted progeny. Expression of the actin-encoding gene indicates relative RNA quantities in each sample. Numbers 1, 11, and 5 represent different transformants.

FIG. 8.

Organization of the melanin biosynthesis gene clusters in Aspergillus fumigatus, Alternaria brassicicola, Magnaporthe grisea, and Cochliobolus heterostrophus. The accession numbers of the indicated genes are listed in Table 2.