Invited Talks

Session I, Genomic Analysis

The MIPS Neurospora crassa Database - MNCDB.
Gertrud Mannhaupt and Werner Mewes. GSF-Forschungszentrum f. Umwelt und Gesundheit, Institut f. Bioinformatik, 85764 Neuherberg, Germany

 The German project of sequencing Neurospora crassa linkage groups II and V started in 1998 and is close to completion. The two chromosomes are estimated to be 4.6 Mb and 9.2 Mb in length, respectively. The total genome has a length of about 43 Mb which is more than 3 times larger than the first completely sequenced eukaryotic genome of Saccharomyces cerevisiae. The MNCDB database at MIPS contains about 17 Mb of genomic sequences. 8 Mb could be definitely assigned to LG II and V and about 9 Mb could be partly assigned to different chromosomes. The entire dataset was automatically analysed using the PEDANT system. PEDANT is a fully automated system using a wide spectrum of sequence analysis and structure prediction tools. 4700 genes are predicted by the program FGENESH, recently trained for Neurospora (V. Solovyev). The results of the capacious analysis provided by PEDANT can be viewed on the project page.
Moreover detailed manual supervised gene modelling and annotation of the genemodels have been carried out for at least all sequences belonging to LG II and V. More than 2500 proteins were processed manually. Beside annotating different features they are classified and assigned to functional categories. The deduced protein sequences are concurrently analysed by the PEDANT system. The genome of Neurospora crassa will complement the information supplied by the ones of S. cerevisiae, C. elegans and D. melanogaster for several aspects. Comparisons of these genomes to the one of N. crassa - the filamentous fungal model organism - will aid the definition of entities required for multi-cellularity and tissue organization.
Summaries of data and related information, such as EST hits, BLAST hits, literature links, DNA and Protein sequences, or functional classification is available on each gene/protein entry and can be viewed on our project homepage.
http://mips.gsf.de/proj/neurospora

The pdx-1 (snz-1/sno-1) region of the Neurospora crassa genome
Donald O. Natvig, Laura E. Bean, William H. Dvorachek, Jr., Edward L. Braun, Allison Errett, Gregory S. Saenz, Mara D. Giles, Margaret Werner-Washburne, and Mary Anne Nelson. Department of Biology, University of New Mexico, Albuquerque, NM

The pdx-1 region of the N. crassa genome contains homologs of two closely-linked stationary phase genes, SNZ1 and SNO1, from Saccharomyces cerevisiae. Homologs of SNZ1 encode extremely highly conserved proteins that have been implicated in pyridoxine (vitamin B6) metabolism in the filamentous fungi Cercospora nicotianae and Aspergillus nidulans. In N. crassa, SNZ and SNO homologs map to the region occupied by pdx-1 (pyridoxine requiring), a gene known in N. crassa since the earliest studies with biochemical mutants, but which was not sequenced previously. Combining genomic sequence data, existing high-resolution mapping data from the pdx-1 region, and sequence data obtained from known mutants previously classified as pdx-1, we demonstrated that the pdx-1 mutant phenotype can derive from mutations in either the SNZ or SNO homolog of N. crassa. This provided the first firm experimental link between SNZ and SNO functions. One N. crassa mutant with a disrupted SNO homolog was at one time designated pdx-2. It now appears appropriate to reserve the pdx-1 designation for the N. crassa SNZ homolog and pdx-2 for the SNO homolog. Analysis of 36 kbp in the pdx region revealed at least 12 protein coding genes, supporting a previous conclusion of high gene densities (12,000-13,000 total genes) for N. crassa. With the exception of pdx-1 and pdx-2 there is no evidence of shared function among the genes in this region.

Gene expression analysis using oligo based microarrays.
Nisha Sahay. Product Manager, QIAGEN Inc., Valencia, CA 91355

Gene expression analysis is a powerful step in the diagnosis of various diseases, appraisal of disease progression, evaluation of drug therapy effectiveness and the estimation of the severity of trauma, as well as many other potential uses. To date the focus of this type of analysis is in screening large numbers of genes (sometimes whole genomes) by utilizing DNA microarrays or in the quantitative analysis of smaller numbers of genes using various quantitative PCR methods. Early generation DNA microarrays based on cDNAs or ORFs were not very specific and also not very quantitative. The lack of specificity of the ORF-based microarrays is mostly due to cross-hybridization of related or overlapping genes found in complex genomes. At QIAGEN Operon we have addressed many of these limitations of the ORF-based microarrays by designing oligos that are concentration normalized, hybridization temperature normalized as well as sequence optimized. In order to achieve good sensitivity while maintaining specificity, the sequences we design are relatively long oligonucleotides.

Microarrays for Neurospora crassa.
Craig R. Tomlinson and Mario Medvedovic. Hyacinth Genomics, LLC, Cincinnati, OH.

The primary goal of Hyacinth Genomics, LLC, is to provide oligonucleotides representing the expressed genome of model systems that have made seminal contributions to biology. Many research groups lack sufficient libraries for genomics studies and prominent among them are researchers who work with Neurospora crassa. To fulfill this need, we propose to generate an oligonucleotide library representing the genes of N. crassa to be available as arrayed DNAs in microtiter plates and on microscope slides. Using the recently completed N. crassa genome sequence database, the latest advances in bioinformatics will be applied to the sequence design of more than 12,000 oligonucleotides using an approach that will minimize cross-hybridization and maximize specificity and sensitivity. Studies will be carried out to demonstrate the efficacy of the N. crassa oligonucleotide sequences. Our purpose here is to assess the interest and support of our plan to supply affordable microarrays to the N. crassa community. We intend to keep costs to a minimum by including the production of the N. crassa oligonucleotide library as part of the budget of an NIH-sponsored Small Business Initiative Research grant. The providing of arrayed DNAs representing the N. crassa genome would greatly promote genomics studies in this significant model system.

Session II, Signaling and Development

The clock in Neurospora.
Jennifer Loros and Jay Dunlap. Departments of Genetics and Biochemistry, Dartmouth Medical School

Neurospora is a uniquely tractable model system for the analysis of the molecular basis of eukaryotic circadian oscillatory systems. Molecular bases for the period length and sustainability of the rhythm, light and temperature resetting of the circadian system, and for gating of light input and light effects are becoming understood, and Neurospora promises to be an amenable system for examining the role of coupled feedback loops in the clock. Many of these insights have been shown to have, or to foreshadow, direct parallels in mammalian systems, including specifically the mechanism of light entrainment, the involvement of PAS:PAS heterodimers as transcriptional activators in essential clock-associated feedback loops, and the dual role of FRQ in the loop as both an activator and a repressor; similarities extend to the primary sequence level in at least one case, that of WC-1 and BMAL1. Work on circadian output in Neurospora, using several approaches, has identified numerous regulated genes and has been at the forefront of studies aimed at understanding clock control of gene expression.
This work was supported by grants from the National Science Foundation MCB-0084509 to J.J.L., the National Institutes of Health MH44651 to J.C.D. and J.J.L., MH01186 and GM34985 to J.C.D. and the Norris Cotton Cancer Center core grant at Dartmouth Medical School.

Roles for WHITE COLLAR-1 in circadian and general photoperception in Neurospora crassa. (updated abstract follows)
Kwangwon Lee, Jennifer J. Loros, and Jay C. Dunlap. Department of Biochemistry and Department of Genetics, Dartmouth Medical School. Hanover NH. 03755.

The transcription factors WHITE COLLAR-1 (WC-1) and WHITE COLLAR-2 (WC-2) interact to form a heterodimeric complex (WWC) that is essential for all light- mediated processes in Neurospora. Based on the blind phenotypes of mutants and sequence comparisons with known photoreceptor proteins, WC-1 has been proposed to be the Neurospora blue light photoreceptor. WCC also plays a distinct non-light related role as the transcriptional activator in the FREQUENCY(FRQ)/WCC feedback loop that is central to the Neurospora circadian system. To further investigate the roles of WC-1, we analyzed the phenotypes of five different wc-1 mutant alleles for 1) FRQ expression in constant darkness 2) light induction of FRQ, and WC-1, 3) post-transcriptional modification of FRQ, 4) light induction of developmentally and light inducible genes. Our data show that WC-1 is involved in FRQ expression at at least two different levels, and only a small N-terminal fraction of the full length of WC-1 is necessary for light induction of frq and other light-inducible genes. Functional WC-1 is necessary for the WCC formation. We propose that WC-1 is not a light- receptor for the circadian clock.
This work was supported by grants from the National Science Foundation MCB-0084509 to J.J.L., the National Institutes of Health MH01186 and GM34985 to J.C.D., GM20553 to K. L. and the Norris Cotton Cancer Center core grant at Dartmouth Medical School.

UPDATED ABSTRACT: ROLES FOR WHITE COLLAR-1 IN CIRCADIAN AND GENERAL PHOTOPERCEPTION IN Neurospora crassa.
Kwangwon Lee, Jennifer J. Loros, and Jay C. Dunlap. Department of Biochemistry and Department of Genetics, Dartmouth Medical School. Hanover NH. 03755.

The transcription factors WHITE COLLAR-1 (WC-1) and WHITE COLLAR-2 (WC-2) interact to form a heterodimeric complex (WWC) that is essential for all light-mediated processes in Neurospora. Based on the blind phenotypes of mutants and sequence comparisons with known photoreceptor proteins, WC-1 has been proposed to be the Neurospora blue light photoreceptor. WCC also plays a distinct non-light related role as the transcriptional activator in the FREQUENCY(FRQ)/WCC feedback loop that is central to the Neurospora circadian system. To further investigate the roles of WC-1, we analyzed the phenotypes of five different wc-1 mutant alleles for 1) FRQ expression in constant darkness 2) light induction of FRQ, and WC-1, 3) post-transcriptional modification of FRQ, 4) light induction of developmentally and light inducible genes. Our data show that WC-1 is involved in FRQ expression at at least two different levels, and WC-1 is necessary for light induction of frq and other light-inducible genes. Microarray analysis has revealed novel roles for WC-1 in light transduction pathways. Supported by NIH (GM20553 to KL, GM34985 to JCD) and NSF (0084509 to JJL).

Identification of novel components involved in controlling circadian rhythmicity using genetic and genomic approaches.
Deborah Bell-Pedersen. Department of Biology, Texas A&M University, College Station, TX 77843

To identify signaling components of the circadian clock system in Neurospora crassa, we have carried out a genetic selection to isolate mutations that alter the expression of clock-controlled genes (ccgs). This selection is based on the differential expression of the ccgs in response to the presence or absence of the clock gene product FRQ. For example, ccg-1 expression is repressed and ccg-2 expression is activated when FRQ is present in the cell. The promoter region of the ccg-1 gene was ligated to mtr. The mtr gene encodes a neutral amino acid permease that allows both positive and negative selection. The fusion constructs were transformed into both bd; mtr; trp-2 and bd;frq10;mtr;trp-2 strains. Reconstruction experiments using the ccg-1:mtr transformants showed FRQ-dependent production of the amino acid permease. Both strains were subjected to UV light mutagenesis and assayed for growth under the opposite conditions. Out of 100,000 colonies screened for each transformant, we identified 15 mutant strains that yielded the desired growth on the selective media. The phenotypes of these mutations will be discussed. In a related study, we are using transcriptional profiling to help us understand the mechanisms by which the clock regulates development and other critical cellular events. N. crassa DNA microarrays were probed with cDNA produced from mRNA obtained from cultures harvested at different times of the day, during developmental induction and from a strain that overexpresses WC-1. These analyses are allowing us to identify genes under circadian control at the level of transcript accumulation, and to identify novel candidate clock components.

The clock kinases in the Neurospora circadian clock.
Yuhong Yang, Ping Cheng, and Yi Liu. UT Southwestern Medicla Center, Dallas, TX.

Phosphorylation of clock proteins represents an important mechanism regulating circadian clocks. In Neurospora, clock protein FREQUENCY is progressively phosphorylated over time, and its level decreases when it is extensively phosphorylated. To identify the kinase phosphorylating FRQ and to understand the function of FRQ phosphorylation, a FRQ phosphorylating kinase was purified and identified as casein kinase II. Disruption of the catalytic subunit gene of CKII in Neurospora resulted in hypophosphorylation and increased levels of FRQ protein. In addition, the circadian rhythms of frq RNA, FRQ protein, and clock-controlled genes are abolished in the CKII mutant. Our data suggest that the phosphorylation of FRQ by casein kinase II may have at least three functions: it decreases the stability of FRQ, reduces the protein complex formation between FRQ and the WHITE COLLAR proteins, and is important for the closing of the Neurospora circadian negative feedback loop. Taken together, our results suggest that CKII is an essential component of the Neurospora circadian clock.

Life without heterotrimeric G alpha proteins.
Katherine A. Borkovich. Department of Plant Pathology, University of California, Riverside.

Our laboratory investigates heterotrimeric G protein signal transduction pathways in the filamentous fungus Neurospora crassa. We have cloned, mutated and characterized three Galpha, one Gbeta and one Ggamma subunits from this species. Our results have demonstrated crucial roles for G proteins in regulation of cell proliferation, asexual and sexual differentiation, and stress tolerance, through both cAMP-dependent and independent pathways. Now we report the relative contributions of the three Galpha proteins, GNA-1, GNA-2 and GNA-3, to these processes, through the construction of strains containing multiple Galpha mutations. GNA-1 and GNA-3 are required for normal adenylyl cyclase activity, and these subunits contribute to hyphal elongation, conidiation and sexual fertility. GNA-2 appears to play a supporting role, in that functions for the gna-2 gene are only revealed in a gna-1 or gna-3 mutant background. Epistatic relationships between mcb (a mutation in the regulatory subunit of cAMP-dependent protein kinase, or PKA) and the gna-1 and gna-3 genes suggest that GNA-1 and GNA-3 lie upstream of PKA during the regulation of several processes in N. crassa.

Session III, Gene Regulation/Gene Silencing

DNA unpaired in meiotic prophase triggers specific silencing of itself and its homologs.
Robert L. Metzenberg, Patrick K. T. Shiu, Namboori B. Raju, and Denise Zickler*. Stanford University, Stanford, CA. *Universite Paris-Sud, Orsay, France.

 In the sexual phase of Neurospora, fusion of nuclei of opposite mating type results in a diploid zygote which quickly enters prophase of the first meiotic division. Homologs become intimately paired during pachytene of prophase1. Any DNA that is not so paired generates a signal in which a double-stranded RNA is apparently made as an intermediate. The result is destruction of mRNA encoded by the unpaired DNA or by any gene of the same sequence, paired or unpaired. Thus any gene present in an odd number of copies in the two parents taken together will be unexpressed, as will any gene present in an even number, but located differently in the two parents so that it cannot be paired. This system of meiotic silencing by unpaired DNA (MSUD) offers a unifying explanation of several seemingly unrelated details of Neurospora biology. These include the ascus- dominance of some classical mutants, of some deletion mutants, and of gene insertion mutants; the barrenness of strains with segmental duplications; and (in part),the barrenness of crosses of N. crassa to related species like N. tetrasperma and N. sitophila. A mutant (Sad-1) in which the gene for RNA-directed RNA polymerase is deleted allows even unpaired genes to be expressed. Thus the mutation suppresses partially or completely the MSUD underlying these examples.

RID, a DNA methyltransferase-like protein that is essential for repeat-induced point mutation in Neurospora crassa.
Michael Freitag, Gregory O. Kothe, Rebecca Williams and E. U. Selker. Institute for Molecular Biology, University of Oregon, Eugene, OR.

All organisms have mechanisms to safeguard the integrity of their genomes. During sexual development Neurospora crassa inactivates duplicated DNA segments by "repeat-induced point mutation" (RIP). RIP operates on paired DNA repeats, introduces G:C to A:T transition mutations and creates targets for subsequent DNA methylation in vegetative tissue. The mechanism of RIP and its relationship to DNA methylation are poorly understood. We show that a cytosine methyltransferase-like protein is essential for RIP, the first implication of a eukaryotic DNA methyltransferases in mutagenesis under natural conditions. Disruption of the rid (RIP-defective) gene did not affect normal development or vegetative DNA methylation. We isolated conserved rid genes from N. intermedia and N. tetrasperma, and portions of the gene from the homothallic N. africana, N. terricola, N. pannonica and N. galapagosensis. We propose that RID-like proteins constitute a new class of fungal DNA methyltransferases that are generally important during sexual development and for gene silencing processes.

Control of DNA methylation by a histone H3-K9 methyltransferase in Neurospora crassa.
Hisashi Tamaru & Eric U. Selker. Institute of Molecular Biology, University of Oregon, Eugene, OR97403, USA.

 Cytosine methylation is the only known epigenetic mark on DNA in eukaryotes. In contrast histones, which wrap DNA around them, are marked by a variety of combinations of different modifications such as acetylation, phosphorylation and methylation at a variety of positions. DNA methylation is essential for the normal development of mammals and plants and is involved in epigenetic processes such as X-chromosome inactivation, genomic imprinting and silencing of transposons. However, little is know about how DNA methylation is controlled. In an attempt to tag a previously identified DNA methylation gene dim-1 (defective in methylation) by insertional mutagenesis we unexpectedly generated a mutation in a novel gene, dim-5,which like the dim-2(DNA methyltransferase) mutation, eliminates all or nearly all DNA methylation. We mapped dim-5between trp-4and leu-2on LG IV. A contig containing both trp-4and leu-2was found in a N. crassagenomic sequence data base, based on their expected homology to S. cerevisiae TRP4and LEU1,respectively. We scrutinized the interval between the putative trp-4and leu-2for DIM-5 candidates using BLASTx. One striking candidate was found to encode a protein related to histone H3-K9 methyltransferases involved in heterochromatin silencing in fission yeast and fly. We confirmed that the candidate gene is dim-5 by complementation and quelling tests using DNA fragments containing the gene. Sequencing of the dim-5 gene in dim-5 mutant identified a nonsense mutation in a SET domain that is known to be required for several previously identified histone H3 methyltransferases. Biochemical studies on recombinant DIM-5 demonstrated that this protein specifically methylates K9 of histone H3. Substitutions of K9 in histone H3 caused dominant loss of DNA methylation in vivo. We conclude that histone H3-K9 methylation signals DNA methylation in Neurospora crassa.

New insights into recombination in Neurospora.
David EA Catcheside, P Jane Yeadon, J Paul Rasmussen, Frederick J Bowring and Lin Koh. School of Biological Sciences, Flinders University, PO Box 2601, Adelaide, SA 5001, Australia.

Meiotic recombination in the his-3 locus is predominantly initiated at or close to cog^L, a recombination hotspot within the promoter region of the nearby gene lpl. We have found that interposition of blocks of sequence heterology between cog^L and his-3 scarcely affect the frequency of recombination within his-3, showing that propagation of recombination events from this hotspot does not require contiguous tracts of sequence homology in the parental chromosomes. His+ progeny are readily obtained from crosses heterozygous for a his-3 point mutation and a chromosomal translocation that disrupts his-3 between the point mutation and cog^L. A substantial proportion of the recombinant progeny show evidence of sequence exchange both sides of the translocation breakpoint, suggesting that DNA repair synthesis can make multiple excursions between the parental chromosomes. This would resurrect a copy choice mechanism, long out of favour, to explain some recombination events.

Session IV, Organelle Biogenesis/Metabolic Regulation

Function and assembly of the preprotein translocase of the outer mitochondrial membrane (the TOM complex).
Frank E. Nargang and Rebecca D. Taylor. Department of Biological Sciences, University of Alberta, Edmonton, Alberta.

The TOM complex recognizes mitochondrial preproteins synthesized in the cytosol and translocates them across, or into, the outer mitochondrial membrane. The complex contains receptor proteins with domains that extend into the cytosol as well as integral membrane proteins that form the core TOM complex. The major component of the core complex is Tom40, which forms the translocation pore. The tom40 gene has been shown to be essential in both yeast an N. crassa. Cells with reduced levels of Tom40 grow very slowly and contain small mitochondria that lack cristae. Mitochondria isolated from such strains are deficient in their ability to import mitochondrial preproteins and are deficient in the core TOM complex components Tom22 and Tom6 suggesting that the import and/or stability of these proteins is dependent on the presence of Tom40. We have analyzed mutant variants of Tom40 for in vivo function and for their ability to be assembled into the TOM complex in vitro. Assembly of Tom40 into the complex occurs through a series of intermediates. Our analysis of Tom40 variants suggests that a high molecular weight intermediate on the assembly pathway contains only newly imported subunits and that conserved regions in the N-terminus of the protein play an important role in the assembly process.

Protein translocation into mitochondria: Purification and characterization of the inner membrane protein translocases of Neurospora crassa.
Christian Kozany, Dejana Mokranjac, Andreja Vasiljev, Walter Neupert, and Holger Prokisch. Institut für Physiologische Chemie, Munich, Germany.

Mitochondria are essential organelles of eukaryotic cells. The large majority of the proteins required for mitochondrial structure and function is encoded by nuclear genes which have to be imported into mitochondria. Import of mitochondrial preproteins is mediated by a general translocase in the outer membrane, the TOM complex, and by two distinct translocases in the mitochondrial inner membrane, the TIM23 complex and the TIM22 complex. The TOM complex must specifically recognise mitochondrial precursor proteins synthesized in the cytosol and translocate them across the outer membrane. For further import into or across the inner membrane the TOM complex cooperates with both TIM complexes. The TIM23 complex mediates import of preproteins with N-terminal presequences, whereas hydrophobic proteins with internal targeting signals are inserted via the TIM22 complex. The translocation pathway has been studied in detail in the yeast system, however, not much is known about the precise composition and molecular structure of the TIM complexes. For a detailed biochemical characterisation of the TIM complexes we cloned the genes from N. crassa and generated via different methods strains expressing histedinyl-tagged Tim components. The possibility to obtain large amounts of Neurospora organelles allowed the purification of the low abundant TIM complexes in substantial amounts. The characterisation of the TIM complexes will be presented.

Transcripts and transcript-binding proteins in mitochondria of Neurospora crassa.
Robert Brambl, Nora Plesofsky, and Jill Kleidon. Department of Plant Biology, University of Minnesota, Saint Paul.

We compared expression elements of three disparate groups of mitochondrial genes in Neurospora crassa, COB, COX1, and the clustered ATP8-ATP6-mtATP9-COX2. To identify promoter sequences we employed the published N. crassa consensus sequence for COB, and we found closely related sequences within the 5'-UTRs of both COX1 and the ATP8-COX2 transcriptional units. The COX1 RNA includes two flanking URF sequences, but the more distant 3'-flanking ND1 and 5'-flanking tRNA^cysare not included in the mature COX1 transcript. The ATP8, ATP6, mtATP9, and COX2 are expressed as a single transcript which does not include adjacent 5'-URF sequence. Primer extension analysis of the 5'-UTR of the COX1 transcript showed that there was only one 5'-end for the COX1 gene transcript, which is 73 nt downstream of the consensus promoter sequence and is the first nt 3' of the sequence for the intervening tRNA^cys. Primer extension analysis of the 5'-UTR ATP8-COX2 sequence showed that the 5'-end for this transcript was the first nt 3' of the consensus promoter sequence. We per49 formed gel-shift experiments to detect proteins in mitochondria that bind to transcripts as possible regulatory proteins. Results with the 5'-UTR RNAs of COB, COX1, and ATP8-COX2 suggest that they may be binding both unique proteins and an overlapping group of two to four proteins of ~155 to 45 Mr. We successively deleted regions of the RNA 5'-UTRs to identify sequences that bound these proteins. We identified similar predicted stem-loop secondary structures in the protein-binding regions of all three UTRs.

Complex I, sex and alternative NADH dehydrogenases in Neurospora mitochondria.
Arnaldo Videira. Instituto de Biologia Molecular e Celular and Instituto de Ciências Biomédicas de Abel Salazar, University of Porto, Portugal.

Respiratory chain complex I couples electron transfer with proton translocation through the inner membrane of mitochondria, thus participating in oxidative phosphorylation. The transfer of electrons from NADH to ubiquinone is mediated by protein-bound prosthetic groups. Complex I contains about 40 proteins of dual genetic origin distributed in two major domains, the peripheral and membrane arms. While 14 of these subunits are conserved in prokaryotes and constitute a “minimal structure” for activity, the others are present exclusively in eukaryotes. Most of the Neurospora polypeptides have been cloned and/or identified. A thoroughly genetic analysis has provided insights into the biogenesis, structure and function of complex I. An overview of the effects of disruption and site-directed mutagenesis of specific complex I genes, including the generation of models for human mitochondrial disease, will be presented. Complex I seems to be dispensable for Neurospora grown under vegetative conditions but it is required for sexual development. The sexual phase of the fungal life cycle is blocked in homozygous crosses between complex I mutants. In addition to complex I, organisms also contain non-proton-pumping alternative NADH dehydrogenases with varying number and specificity. Like plants, N. crassa contains 3-4 alternative NAD(P)H dehydrogenases in the mitochondrial inner membrane. One of them is located in the outer face of the membrane (external enzyme), working as a calcium-dependent oxidase of cytosolic NADPH. At least another external enzyme and a NADH dehydrogenase working with matrix substrates (internal enzyme) are present in the fungus.

Translational control by upstream open reading frames in mRNA.
Matthew Sachs. Department of Biochemistry and Molecular Biology, OGI School of Science and Engineering, Oregon Health & Science University, Beaverton, OR.

Translational control by upstream open reading frames (uORFs) in the 5'-leaders of prokaryotic mRNAs and eukaryotic mRNAs is an increasingly well-documented form of genetic control. Arg-specific translational regulation by the arginine attenuator peptide (AAP), encoded by a uORF in the Neurospora crassa arg-2 specifying the small subunit of carbamoyl phosphate synthetase, represents a clear example of such a control phenomenon. Using cell-free translation systems and a primer extension inhibition (toeprint) assay, we showed that synthesis of the 24-residue N. crassa AAP in high Arg causes ribosomes to stall at the uORF termination codon, blocking ribosomes from scanning to the downstream initiation codon. The AAP also causes Arg-regulated stalling of ribosomes involved in elongation when it is fused directly to the N-terminus of a luciferase reporter, suggesting that it blocks a step common to elongation and termination. The peptide sequence is important, and mutation of an evolutionarily conserved Asp residue to Asn eliminates regulation in vivo and in vitro in both N. crassa and Saccharomyces cerevisiae. Regulation appears independent of the charging status of tRNA. We have further developed the cell-free N. crassa system so that we can radiolabel nascent AAP-containing peptides and directly assess the regulatory effects of Arg on polypeptide synthesis. The data demonstrate that, when Arg causes ribosomes to stall during elongation following synthesis of the AAP, the nascent peptide remains associated with ribosomes, and stalled ribosomes can subsequently resume polypeptide synthesis. To characterize new mutations potentially affecting Arg-specific translational control, we have adapted a microtiter plate assay to monitor the affects of mutations affecting the growth of an N. crassa arg-12^s pyr-3 arg-2-hph through their effects on the expression of the arg-2 and arg-2-hph genes.

Sugar sensing in Neurospora.
Daniel Ebbole and Xin Xie. Department of Plant Pathology and Microbiology, Texas A & M University, College Station Tx.

The ability to sense the presence of a preferred carbon source is a fundamental property of growing cells. In yeast, several mechanisms are involved in glucose sensing and controlling gene expression. One mechanism is to employ glucose transporter homologs that have evolved as glucose receptor/sensors but that are not themselves functional glucose transporters. We have analysed the N. crassa rco-3 gene and conclude that it functions as a sugar sensor rather than a sugar transporter. To gain further evidence for the role of rco-3 as a regulator we have isolates several suppressors of rco-3 and are examining the effect of the rco-3 mutant on gene expression. The characterization of the suppressed rco-3 mutants and transcriptional profiling of the rco-3 mutant strain is helping us define the genetic pathway responsible for sugar sensing.

Regulation of sulfur metabolism in Neurospora crassa.
John V. Paietta. Department of Biochemistry and Molecular Biology, Wright State University, Dayton, OH.

The sulfur regulatory system of Neurospora crassa is composed of a group of structural genes (e.g., ars-1, arylsulfatase) that are under coordinate control of the CYS3 positive regulator and SCON negative regulators. We are constructing a comprehensive model of sulfur metabolism and have begun by further defining the role of the CYS3 regulator. CYS3-dependent sulfur regulated expression has been found in a number of genes involved in cysteine and methionine metabolism (e.g, methionine synthase, cystathionine gamma-lyase, homoserine transacetylase); thereby expanding the role of CYS3. Further, in vitro binding site studies have been used to more accurately determine the CYS3 recognition sequence. The combination of gene expression profile data, based on the availability of the genomic sequence, along with binding site data will allow for greater selectivity in a genome-wide analysis for genes controlled by this system.

Session V, Cell Biology and Morphogenesis

Analysis of dynactin-membrane interaction in Neurospora.
I. H. Lee, S. Kumar and M. Plamann. School of Biological Sciences, University of Missouri-Kansas City, Kansas City, MO 64110-2499.

Dynactin has been proposed to link the microtubule-associated motor cytoplasmic dynein with membranous cargo; however, the mechanism by which dynactin-membrane interaction is regulated is unknown. Actin-related protein 1 (Arp1) is the most abundant subunit of dynactin, and it forms a short filament to which additional subunits associate. A pointed-end binding subcomplex binds to one end of the Arp1 filament, while a shoulder/sidearm complex containing p150Glued binds to other end. p150Glued of dynactin has been shown to mediate dynein-dynactin interaction; however, it has not been found to play a direct role in membrane- binding. We analyzed 43 p150Glued mutants, and we found that C-terminal deletions that remove the terminal domains result in constitutive dynactin-membrane binding. We propose that the C-terminal domains of p150Glued regulate dynactin- membrane binding through a steric mechanism that controls accessibility of the Arp1 filament of dynactin to membranous cargo. We also examined mutants defective in the Arp1 pointed-end subcomplex. We found that these mutants show stronger dynein/dynactin-membrane interaction relative to wild-type suggesting that the Arp1 pointed-end complex may also play a role in regulating interaction of dynactin with membranous cargo.

Crystal structure of Hex-1 reveals its mechanism of self-assembly and evolutionary origin.
Greg Jedd¹, Ping Yuan², D. Kumaran³, S. Swaminathan³, Nam-Hai Chua¹ and K. Swaminathan². ¹Laboratory of Plant Molecular Biology, The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399, USA,²Macromolecular X-ray Crystallography Laboratory, Institute of Molecular Agrobiology, National University of Singapore, 1 Research Link, Singapore 117604, ³Biology Department, Brookhaven National Laboratory, Upton, NY 11973-5000, USA.

The Woronin body is a specialized peroxisomal vesicle that is restricted to the filamentous Ascomycotina and its major protein component, Hex-1, self-assembles to form the vesicle's crystalline core. We present the crystal structure of Hex-1 at 1.78 Å and provide the structural basis for its self-assembly. The protein consists of two mutually perpendicular B-barrels. The N-terminal barrel contains six B-strands and the C-terminal domain contains a five-stranded barrel and a flanking Alpha-helix. The crystal reveals three primary intermolecular contacts that promote self-assembly through the formation of crosslinked Hex-1 filaments. Self-assembly is abolished in vitro and in vivo by mutations in intermolecular contact residues, indicating that the crystal structure is a valid representation of the Woronin body-core. In spite of sharing a weak sequence homology, the tertiary structures of Hex-1 and the eukaryotic translation initiation factor 5a (eIF-5A) are very similar, suggesting an ancestral link between them. Since eIF-5A does not self-assemble, comparison of these two protein structures suggests how a crystalline protein complex evolved from a soluble precursor.

Morphological miscellany.
Tony Griffiths. Botany, UBC, 6270 University Blvd., Vancouver, BC. Canada V6T 1Z4.

Recent work has focused on trying to understand growth and branching of Neurospora through morphological mutants. One project (A. Virag) attempts to isolate mutations of the actin gene by obtaining cytochalasin resistant mutants. Preliminary indications suggest that at least one resistant mutant of this type is at the actin locus. Another approach (O. Gavric) is to piece together the available morphological mutants into groups that might represent developmental hierarchies, by studying interactions in double mutants, and by structural and physiological similarities between mutants. Several novel types of gene interaction have been observed. Some overall patterns are emerging. Finally we are extending earlier studies on morphological mutants by obtaining chemical profiles of the mutants (E. Jovel).

Antibiotic drugs and the vacuolar ATPase.
Barry Bowman and Emma Jean Bowman. MCD Biology, University of California, Santa Cruz. CA.

The vacuolar ATPase generates an electrochemical gradient across membranes of the Golgi, vacuole and other organelles in the endomembrane system. The enzyme is composed of 13 types of subunits and functions as a rotary motor. Several different kinds of antibiotics have evolved that are highly potent, specific inhibitors. To identify the binding site of the antibiotics we have selected mutant strains of Neurospora crassa that are resistant to the toxic effects of bafilomycin. One class of mutant strains has single point mutations in the vma-3 gene, which encodes a 16 kDa proteolipid subunit of the vacuolar ATPase. Three different residues were altered among the bafilomycin-resistant strains. The mutated sites appear to be on the outer face of the "rotor" sector of the enzyme. The positions of two mutated residues corresponds precisely to the positions of mutated residues in the homologous subunit of the mitochondrial ATPase. In the mitochondrial ATPase the mutations confer resistance to oligomycin, an antibiotic that has no effect on the vacuolar ATPase. These results suggest that vacuolar and mitochondrial ATPases have an ancient, conserved antibiotic binding site. As the sequence of the polypeptides diverged, new antibiotics have arisen that target the same vulnerable site in this family of enzymes.