Overlapping Podospora anserina Transcriptional Responses to Bacterial and Fungal Non Self Indicate a Multilayered Innate Immune Response.

Recognition and response to non self is essential to development and survival of all organisms. It can occur between individuals of the same species or between different organisms. Fungi are established models for conspecific non self recognition in the form of vegetative incompatibility (VI), a genetically controlled process initiating a programmed cell death (PCD) leading to the rejection of a fusion cell between genetically different isolates of the same species. In Podospora anserina VI is controlled by members of the hnwd gene family encoding for proteins analogous to NOD Like Receptors (NLR) immune receptors in eukaryotes. It was hypothesized that the hnwd controlled VI reaction was derived from the fungal innate immune response. Here we analyze the P. anserina transcriptional responses to two bacterial species, Serratia fonticola to which P. anserina survives and S. marcescens to which P. anserina succumbs, and compare these to the transcriptional response induced under VI conditions. Transcriptional responses to both bacteria largely overlap, however the number of genes regulated and magnitude of regulation is more important when P. anserina survives. Transcriptional responses to bacteria also overlap with the VI reaction for both up or down regulated gene sets. Genes up regulated tend to be clustered in the genome, and display limited phylogenetic distribution. In all three responses we observed genes related to autophagy to be up-regulated. Autophagy contributes to the fungal survival in all three conditions. Genes encoding for secondary metabolites and histidine kinase signaling are also up regulated in all three conditions. Transcriptional responses also display differences. Genes involved in response to oxidative stress, or encoding small secreted proteins are essentially expressed in response to bacteria, while genes encoding NLR proteins are expressed during VI. Most functions encoded in response to bacteria favor survival of the fungus while most functions up regulated during VI would lead to cell death. These differences are discussed in the frame of a multilayered response to non self in fungi.

Functional significance of four successive glycine residues in the pyrophosphate binding loop of fungal 6-oxopurine phosphoribosyltransferases.

Hypoxanthine-guanine phosphoribosyltransferase (HGPRT) is a key enzyme of the purine recycling pathway that catalyzes the conversion of 5-phospho-ribosyl-α-1-pyrophosphate and guanine or hypoxanthine to guanosine monophosphate (GMP) or inosine monophosphate (IMP), respectively, and pyrophosphate (PPi). We report the first crystal structure of a fungal 6-oxopurine phosphoribosyltransferase, the Saccharomyces cerevisiae HGPRT (Sc-HGPRT) in complex with GMP. The crystal structures of full length protein with (WT1) or without (WT2) sulfate that mimics the phosphate group in the PPi binding site were solved by molecular replacement using the structure of a truncated version (Δ7) solved beforehand by multiwavelength anomalous diffusion. Sc-HGPRT is a dimer and adopts the overall structure of class I phosphoribosyltransferases (PRTs) with a smaller hood domain and a short two-stranded parallel ß-sheet linking the N- to the C-terminal end. The catalytic loops in WT1 and WT2 are in an open form while in Δ7, due to an inter-subunit disulfide bridge, the catalytic loop is in either an open or closed form. The closure is concomitant with a peptide plane flipping in the PPi binding loop. Moreover, owing the flexibility of a GGGG motif conserved in fungi, all the peptide bonds of the phosphate binding loop are in trans conformation whereas in nonfungal 6-oxopurine PRTs, one cis-peptide bond is required for phosphate binding. Mutations affecting the enzyme activity or the previously characterized feedback inhibition by GMP are located at the nucleotide binding site and the dimer interface.

Two structurally similar fungal prions efficiently cross-seed in vivo but form distinct polymers when coexpressed.

HET-s is a prion protein of the filamentous fungus Podospora anserina. An orthologue of this protein, called FgHET-s has been identified in Fusarium graminearum. The region of the FgHET-s protein corresponding to the prion forming domain of HET-s, forms amyloid fibrils in vitro. These fibrils seed HET-s(218-289) fibril formation in vitro and vice versa. The amyloid fold of HET-s(218-289) and FgHET-s(218-289) are remarkably similar although they share only 38% identity. The present work corresponds to the functional characterization of the FgHET-s(218-289) region as a prion forming domain in vivo. We show that FgHET-s(218-289) is capable of prion propagation in P. anserina and is able to substitute for the HET-s PFD in the full-length HET-s protein. In accordance with the in vitro cross-seeding experiments, we detect no species barrier between P. anserina and F. graminearum PFDs. We use the yeast Saccharomyces cerevisiae as a host to compare the prion performances of the two orthologous PFDs. We find that FgHET-s(218-289) leads to higher spontaneous prion formation rates and mitotic prion stability than HET-s(218-289). Then we analysed the outcome of HET-s(218-289)/FgHET-s(218-289) coexpression. In spite of the cross-seeding ability of HET-s(218-289) and FgHET-s(218-289), in vivo, homotypic polymerization is favoured over mixed fibril formation.

Lethal accumulation of guanylic nucleotides in Saccharomyces cerevisiae HPT1-deregulated mutants.

Guanylic nucleotide biosynthesis is a conserved and highly regulated process. Drugs reducing GMP synthesis affect the immunological response and mutations enabling guanylic-derivative recycling lead to severe mental retardation. While the effects of decreased GMP synthesis have been well documented, the consequences of GMP overproduction in eukaryotes are poorly understood. In this work, we selected and characterized several mutations making yeast hypoxanthine-guanine phosphoribosyltransferase insensitive to feedback inhibition by GMP. In these mutants, accumulation of guanylic nucleotides can be triggered by addition of extracellular guanine. We show that such an accumulation is highly toxic for yeast cells and results in arrest of proliferation and massive cell death. This growth defect could be partially suppressed by overexpression of Rfx1p, a transcriptional repressor of the DNA damage response pathway. Importantly, neither guanylic nucleotide toxicity nor its suppression by Rfx1p was associated with an alteration of forward mutation frequency.

Temporal gene expression of 3-ketoacyl-CoA reductase is different in high and in low erucic acid Brassica napus cultivars during seed development.

The membrane-bound acyl-CoA elongase complex is a key enzyme responsible for erucoyl-CoA synthesis. Among the four putative genes encoding the four moieties of this complex in Brassica napus seeds, only one has been characterized, the Bn-fae1 gene, which encodes the 3-ketoacyl-CoA synthase. The genes encoding the other enzymes (3-ketoacyl-CoA reductase, 3-hydroxyacyl-CoA dehydratase and trans-2,3-enoyl-CoA reductase) have not been identified. We cloned two 3-ketoacyl-CoA reductase cDNA isoforms, Bn-kcr1 and Bn-kcr2, from B. napus seeds. Their function was identified by heterologous complementation in yeast by restoring elongase activities. The comparison of Bn-kcr mRNA expression in different B. napus tissues showed that the genes were preferentially expressed in seeds and roots. We also investigated the regulation of gene expression in High Erucic Acid Rapeseed (HEAR) and in Low Erucic Acid Rapeseed (LEAR) cultivars during seed development. The co-expression of Bn-fae1 and Bn-kcr observed in HEAR cultivar during seed development was different in LEAR cultivar, suggesting that expression of both genes was directly or indirectly linked.

Rvs161p and sphingolipids are required for actin repolarization following salt stress.

In Saccharomyces cerevisiae, the actin cytoskeleton is depolarized by NaCl stress. In this study, the response was maximal after 30 min, and then actin patches repolarized. Rvs161p was required for actin repolarization because the rvs161delta mutant did not repolarize actin patches after growth in a salt medium. Mutations suppressing the rvs161delta-related salt sensitivity all occurred in genes required for sphingolipid biosynthesis: FEN1, SUR4, SUR2, SUR1, and IPT1. These suppressors also suppressed act1-1-related salt sensitivity and the defect in actin repolarization of the rvs161delta mutant, providing a link between sphingolipids and actin polarization. Indeed, deletion of the suppressor genes suppressed the rvs161delta defect in actin repolarization in two ways: either actin was not depolarized at the wild-type level in a set of suppressor mutants, or actin was repolarized in the absence of Rvs161p in the other suppressor mutants. Rvs161p was localized as cortical patches that concentrated at polarization sites, i.e., bud emergence and septa, and was found to be associated with lipid rafts. An important link between sphingolipids and actin polarization is that Rvs161p was required for actin repolarization and was found to be located in lipid rafts.