Live cell imaging of septin dynamics in Ustilago maydis.

Septins are highly conserved cytoskeletal proteins involved in a variety of biological processes such as cell polarization and cytokinesis. In humans, functional defects in these proteins have been linked to cancer and neuronal diseases. In recent years, substantial progress has been made in studying the structure of septin subunits and the formation of defined heteromeric building blocks. These are assembled into higher-order structures at distinct subcellular sites. An important microscopic approach in studying septin assembly and dynamics is the use of septins tagged with fluorescent proteins. This revealed, eg, that septins form rings during cytokinesis and that septins build extended filaments partially colocalizing with actin cables and microtubules. Here, we describe extensive live cell imaging of septins in the model microorganism Ustilago maydis. We present techniques to study dynamic localization of protein and septin mRNA on shuttling endosomes as well as colocalization of proteins at these highly motile units. Moreover, FLIM-FRET experiments for analyzing local protein interactions are presented. Importantly, these imaging approaches transfer well to other fungal and animal model systems for in vivo analysis of septin dynamics.

A reverse genetic approach for generating gene replacement mutants in Ustilago maydis.

We describe a versatile strategy for generating gene replacement mutants in the phytopathogenic fungus Ustilago maydis. The system includes the choice of 32 different insertion cassettes for genetic engineering purposes, such as gene disruption and more sophisticated insertions of reporter genes, heterologous promoters or combinations of the two. PCR-amplified flanking sequences needed for homologous recombination are ligated to the respective insertion cassettes via SfiI sites. As proof of principle we generated two replacement mutants in which the endogenous promoter of the pheromone gene mfa1 drives expression of the Green Fluorescent Protein gene (gfp). Simultaneously, expression of the mfa1 ORF is controlled either by the carbon source-regulated crg1 promoter or the nitrogen source-regulated nar1 promoter. In both cases gfp expression was pheromone-inducible and pheromone expression was only detected when the heterologous promoters were active.

Structure-function analysis of lipopeptide pheromones from the plant pathogen Ustilago maydis.

Mating of two haploid cells is a prerequisite for the successful infection of corn by the pathogenic fungus Ustilago maydis. Cell-cell recognition is mediated by small lipopeptide pheromones. Genes encoding pheromone precursors as well as pheromone receptors are located in the a mating type locus. Two pheromones are known, the tridecapeptide a1 and the nonapeptide a2, both of which contain an S-prenylated cysteine methyl ester at the C-terminus. It has previously been shown that synthetic pheromones are active in a biological test system. Here, we used the same assay to perform a detailed analysis of synthetic a1 and a2 pheromones. Testing of truncated derivatives of a1 and a2 revealed that in both cases the pheromone function is less sensitive to N-terminal than to C-terminal truncations. Replacement of each amino acid in the a1 pheromone by either alanine or the corresponding D-amino acids revealed that four positions are important for function: the two central glycines (positions 5 and 9), proline at position 7 and tyrosine at position 10. By introducing different naturally occurring as well as synthetic amino acids at position 10, we demonstrate that the presence of an aromatic side chain at this position is necessary for function. We propose a model in which a cis peptide bond at proline 7 favours the formation of a type II' beta turn of the a1 pheromone backbone with glycine 9 in position i+1 (where i refers to the first position of the beta turn). As a result, tyrosine 10, at position i+2 of the turn, would be highly exposed and could be inserted into a structurally well-defined binding pocket of the receptor. The latter may represent an important facet of receptor specificity.

Comparative analysis of the plant mRNA-destabilizing element, DST, in mammalian and tobacco cells.

The labile SAUR transcripts from higher plants contain a conserved DST sequence in their 3'-untranslated regions. Two copies of a DST sequence from soybean are sufficient to destabilize reporter transcripts in cultured tobacco cells whereas variants bearing mutations in the conserved ATAGAT or GTA regions are inactive. To investigate the potential for conserved recognition components in mammalian and plant cells, we examined the function of this instability determinant in mouse NIH3T3 fibroblasts and tobacco BY2 cells. In fibroblasts, a tetrameric DST element from soybean accelerated deadenylation and decay of a reporter transcript. However, a version mutated in the ATAGAT region was equally effective in this regard, and a tetrameric DST element from Arabidopsis was inactive. In contrast, the soybean DST element was more active as an mRNA instability element than the mutant version and the Arabidopsis element, when tested as tetramers in tobacco cells. Hence, the plant DST element is not recognized in animal cells with the same sequence requirements as in plant cells. Therefore, its mode of recognition appears to be plant-specific.

A homologue of the transcriptional repressor Ssn6p antagonizes cAMP signalling in Ustilago maydis.

In Ustilago maydis, cAMP signalling is crucial for successful infection of maize plants. Strains are non-pathogenic if mutated in any of the currently identified components of this signalling pathway, such as the alpha-subunit of a heterotrimeric G protein Gpa3, the adenylate cyclase Uac1 and the regulatory and catalytic subunit of protein kinase A Ubc1 and Adr1 respectively. Deletion of gpa3, uac1 or adr1 triggers filamentous growth, and certain point mutations in gpa3 and ubc1 that mimic a high cAMP level display a glossy colony phenotype. Screening an autonomously replicating U. maydis library in such a background (gpa3Q206L), we identified sql1 as a suppressor of the glossy colony phenotype. Interestingly, only alleles carrying C-terminal truncations of Sql1 were able to complement the mutant phenotype, suggesting a gain-of-function by these variants. Sql1 is a functional homologue of the yeast transcriptional repressor Ssn6p and contains 10 tetratricopeptide repeats (TPRs), of which the first six are important for suppressor function. Truncated sql1 alleles that suppress the glossy colony phenotype of gpa3Q206L strains induce filamentous growth when introduced in wild type. Filamentation of these strains is reversed in the presence of cAMP. We present a model in which Sql1 is part of an evolutionary conserved Sql1-Tup1 transcriptional repressor complex that antagonizes cAMP signalling by repressing cAMP-regulated genes.

Activation of the cAMP pathway in Ustilago maydis reduces fungal proliferation and teliospore formation in plant tumors.

In the corn smut fungus Ustilago maydis, mating of two haploid sporidia is a prerequisite for subsequent colonization of the host. Cyclic AMP (cAMP) and pheromone signals have been implicated in this developmental program. The cAMP pathway is also needed for subsequent fungal development in planta, as null mutants in any component of the pathway fail to form tumors. Here we show that moderate activation of the pathway conferred either by mutation in the Galpha subunit or by mutation in the regulatory subunit of the protein kinase A influences tumor morphology. In the resulting tumors, the amount of fungal material is drastically reduced and fungal development is arrested at the stage of sporogenic hyphae. We conclude that tight regulation of the cAMP pathway is crucial for fungal development within the plant but does not interfere with the tumor induction process.

Fungal-plant signalling in the Ustilago maydis-maize pathosystem.

The smut fungus Ustilago maydis needs the host plant maize for completion of its sexual life cycle. Recent experiments highlight the importance of cAMP and mitogen-activated protein (MAP) kinase signalling for cell fusion as well as for subsequent stages of plant colonisation and induction of disease symptoms. During these distinct developmental stages the same signalling cascades must be differentially regulated and accommodate multiple inputs and outputs.

The MAP kinase kpp2 regulates mating and pathogenic development in Ustilago maydis.

In the phytopathogenic fungus Ustilago maydis, fusion of compatible haploid cells is a prerequisite for infection. This process is genetically controlled by the biallelic a locus, encoding pheromone precursors and receptors. These are presumed to be coupled to a heterotrimeric G protein and a MAP kinase cascade, leading to activation of the HMG domain transcription factor Prf1. Here, we have demonstrated that putative MAP kinase sites in Prf1 are required for its activity during mating. In addition, we have identified a gene, kpp2, which encodes a putative MAP kinase related to Pmk1 of Magnaporthe grisea and Fus3p of Saccharomyces cerevisiae. kpp2 deletion mutants are attenuated in several steps of development: cell fusion, induction of pheromone-responsive genes and pathogenicity. Epistasis analysis shows that kpp2 does not affect pheromone gene expression through the cAMP signalling cascade. Pathogenicity of kpp2 mutants can be partially restored by overexpressing the b genes, indicating a regulation of Prf1 by Kpp2. These data support the hypothesis that the MAP kinase Kpp2 transmits the pheromone signal.

PcMYB1, a novel plant protein containing a DNA-binding domain with one MYB repeat, interacts in vivo with a light-regulatory promoter unit.

Light regulatory unit 1 (LRU1) is necessary for and sufficient to mediate light-dependent activation of the chalcone synthase (CHS) minimal promoter in Petroselinum crispum. This composite promoter unit consists of at least two distinct cis-acting elements, designated ACECHS and MRECHS, both of which are required for light induction. The ACGT-containing element ACECHS interacts with common plant regulatory factors (CPRFs) which belong to the basic region/leucine zipper (bZIP) class of transcription factors. Here, we demonstrate that MRECHS, originally identified as an in vivo DNA footprint, is a MYB recognition element. This element possesses a functional core that is essential for light responsiveness and is specifically recognized by two distantly related MYB-like proteins: MYB305 and the novel factor MYB1 from P. crispum. PcMYB1 was identified by both its specific binding to MRECHS in vitro and recognition of MRECHS in vivo. The deduced amino acid sequence revealed that PcMYB1 contains only one MYB-like repeat. This portion of the protein constitutes the DNA-binding domain. Mutational analysis of PcMYB1 in combination with sequence comparison suggests the presence of a helix-turn-helix structure containing a recognition helix that is sufficient for sequence-specific binding. The structure of this distinct MYB-like DNA-binding domain appears to be conserved in proteins from all three eukaryotic phyla.

The transcriptional regulator CPRF1: expression analysis and gene structure.

Many eukaryotic DNA-binding proteins share a conserved amino acid sequence known as the basic region leucine zipper (bZIP) domain. bZIP proteins recognise DNA, upon dimerization, in a sequence-specific manner. The Common Plant Regulatory Factor 1 (CPRF1) is a bZIP transcription factor from parsley (Petroselinum crispum), which recognises defined elements containing ACGT cores. CPRF1 genomic DNA was cloned and the gene was sequenced. Analysis of the sequence data revealed the existence of 12 exons and 11 introns within a stretch of about 9 kb. A second RNA species hybridising to CPRF1 probes was identified as an alternatively spliced, additional CPRF1 transcript containing intron 8. This polyadenylated RNA species showed accumulation characteristics very similar to those of the CPRF1 mRNA. CPRF1 specifically binds an ACGT-containing element which is located within the composite regulatory unit that is necessary and sufficient for light activation of the parsley chalcone synthase (CHS) minimal promoter. Expression studies at the mRNA level demonstrated that CPRF1 mRNA is present in all organs of light-grown plants in which CHS mRNA expression is detectable, and light-dependent CHS mRNA accumulation was shown to be blocked by cycloheximide. Therefore, translation of a protein factor, possibly CPRF1, may be a prerequisite for CHS promoter activation.

Functional analysis of a light-responsive plant bZIP transcriptional regulator.

Common plant regulatory factor 1 (CPRF1) is a parsley basic region/leucine zipper (bZIP) transcription factor that recognizes specific nucleotide sequences containing ACGT cores. Such a sequence is contained within LRU1, the composite light regulatory unit that is necessary and sufficient for light-dependent activity of the parsley chalcone synthase (CHS) promoter. After light treatment of both etiolated and green seedlings, CPRF1 mRNA levels increased prior to CHS mRNA accumulation. The change in CPRF1 mRNA leads to a light-responsive increase in CPRF1 protein. Transient expression analysis in parsley protoplasts using the CPRF1 promoter fused to the beta-glucuronidase (GUS) open reading frame indicated that light-dependent CPRF1 mRNA accumulation was under transcriptional control. The 5' untranslated region of the CPRF1 gene includes a cis-acting nucleotide sequence that contains two ACGT elements at a distance of 12 bp between their palindromic centers. This feature is reminiscent of as-1 and octopine synthase (ocs) elements identified in promoters from plant pathogens. This double ACGT Element element, designated dACECPRF1, stimulated transcription when placed 5' to a heterologous core promoter. CPRF1 bound to dACECPRF1 DNA as well as to the ACGT element from the CHS promoter in vitro. Cotransfection experiments demonstrated that CPRF1 interacts with these elements in vivo and that overexpression of CPRF1 actually reduced light-dependent transcription from the CHS promoter. CPRF1 thus appears to contribute to the regulation of the CPRF1 gene and to interfere with the activities of light-regulated promoters.

Polyubiquitin gene expression and structural properties of the ubi4-2 gene in Petroselinum crispum.

Ubiquitin is an omnipresent protein found in all eukaryotes so far analysed. It is involved in several important processes, including protein turnover, chromosome structure and stress response. Parsley (Petroselinum crispum) contains at least two active polyubiquitin (ubi4) genes encoding hexameric precursor proteins. The deduced amino acid sequences of the ubiquitin monomers are identical to one another and to ubiquitin sequences from several other plant species. Analysis of the promoter region of one ubi4 gene revealed putative regulatory elements. In parsley plants, the ubi4 mRNAs were the predominant ubiquitin mRNAs and were present at comparable levels in all plant organs tested. In cultured parsley cells, high levels of ubiquitin gene expression remained unaffected by heat shock, elicitor or light treatment.