Identifying Genes Devoted to the Cell Death Process in the Gene Regulatory Network of Ustilago maydis

Abstract

Cell death is a process that can be divided into three morphological patterns: apoptosis, autophagy and necrosis. In fungi, cell death is induced in response to intracellular and extracellular perturbations, such as plant defense molecules, toxins and fungicides, among others. Ustilago maydis is a dimorphic fungus used as a model for pathogenic fungi of animals, including humans, and plants. Here, we reconstructed the transcriptional regulatory network of U. maydis, through homology inferences by using as templates the well-known gene regulatory networks (GRNs) of Saccharomyces cerevisiae, Aspergillus nidulans and Neurospora crassa. Based on this GRN, we identified transcription factors (TFs) as hubs and functional modules and calculated diverse topological metrics. In addition, we analyzed exhaustively the module related to cell death, with 60 TFs and 108 genes, where diverse cell proliferation, mating-type switching and meiosis, among other functions, were identified. To determine the role of some of these genes, we selected a set of 11 genes for expression analysis by qRT-PCR (sin3, rlm1, aif1, tdh3 [isoform A], tdh3 [isoform B], ald4, mca1, nuc1, tor1, ras1, and atg8) whose homologues in other fungi have been described as central in cell death. These genes were identified as downregulated at 72 h, in agreement with the beginning of the cell death process. Our results can serve as the basis for the study of transcriptional regulation, not only of the cell death process but also of all the cellular processes of U. maydis.

Keywords: U. maydis; apoptosis; autophagy; cell death; necrosis; regulatory networks; transcription factors.

FIGURE 1

Orthologous proteins shared between U. maydis, A. nidulans, N. crassa, and S. cerevisiae. (A) Venn diagram comparing orthologous clusters of whole proteomes. (B) The bar plot graph shows the number of orthologous clusters by organism. (C) The plot indicates the number of clusters that are organism specific or shared by 2, 3, or 4 organisms. (D) For the 2326 clusters shared by 4 organisms, the protein abundance levels (in percentages and absolute numbers) are shown for each organism.

FIGURE 2

GRN of U. maydis. TFs (yellow nodes) and TGs (blue nodes). The most connected TFs (nodes) are UMAG_02449 (BHLH domain-containing protein), UMAG_03536 (hypothetical protein), UMAG_06256 (Zn(2)-C6 fungal-type domain-containing protein), UMAG_02835 (hypothetical protein), UMAG_10417 (hypothetical protein), UMAG_03280 (hypothetical protein), UMAG_10426 (pH-response TF pacC), UMAG_10368 (hypothetical protein), UMAG_01597 (Uncharacterized protein), and UMAG_15042 (Cell pattern formation-associated protein ust1).

FIGURE 3

Communities in the network. The richest biological processes for each community were identified and hierarchically clustered based on Euclidean distance measures and Ward’s method for linkage analysis. Each row represents the GO term for biological processes, and each column represents the community ID.

FIGURE 4

Cell death module in U. maydis. Nodes in yellow represent TFs, and nodes in blue are TGs; the gray edges represent regulatory interactions. The size of a node is proportional to its degree of output.

FIGURE 5

Gene expression related to cell death in U. maydis. 1 × 10⁶ cells⋅ml^–1 were incubated in MC medium for 24 h (exponential phase), 48 h (stationary phase begins), and 72 h (aged yeast cells) and the expression of genes related to cell death was measured by qRT-PCR. (A) Log₂ fold change in the expression of genes of U. maydis at 48 h of growth. (B) Log₂ fold change in the expression of genes of U. maydis at 72 h of growth. The actin gene was used as an endogenous control, and the relative expression ratio of target genes was calculated by the 2^{− ΔΔCT} method. All results are from three independent experiments. The values presented are the means ± SD for each group, ns is P-value > 0.05 and ***is P-value < 0.0001 vs. Control.