The Neurospora crassa Genome: Cosmid Libraries Sorted by Chromosome

Corresponding author: Jonathan Arnold, Genetics Department, University of Georgia, Athens, GA 30602., arnold{at}uga.edu (E-mail)

Communicating editor: Z-B. ZENG

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A Neurospora crassa cosmid library of 12,000 clones (at least nine genome equivalents) has been created using an improved cosmid vector pLorist6Xh, which contains a bacteriophage origin of replication for low-copy-number replication in bacteria and the hygromycin phosphotransferase marker for direct selection in fungi. The electrophoretic karyotype of the seven chromosomes comprising the 42.9-Mb N. crassa genome was resolved using two translocation strains. Using gel-purified chromosomal DNAs as probes against the new cosmid library and the commonly used medium-copy-number pMOcosX N. crassa cosmid library in two independent screenings, the cosmids were assigned to chromosomes. Assignments of cosmids to linkage groups on the basis of the genetic map vs. the electrophoretic karyotype are 93 ± 3% concordant. The size of each chromosome-specific subcollection of cosmids was found to be linearly proportional to the size of the particular chromosome. Sequencing of an entire cosmid containing the qa gene cluster indicated a gene density of 1 gene per 4 kbp; by extrapolation, 11,000 genes would be expected to be present in the N. crassa genome. By hybridizing 79 nonoverlapping cosmids with an average insert size of 34 kbp against cDNA arrays, the density of previously characterized expressed sequence tags (ESTs) was found to be slightly <1 per cosmid (i.e., 1 per 40 kbp), and most cosmids, on average, contained an identified N. crassa gene sequence as a starting point for gene identification.

FOR 60 years Neurospora crassa has served as a model system for exploring what genes do (DAVIS 2000 ). Work by BEADLE and TATUM 1941 with N. crassa resulted in the development of the one gene-one enzyme hypothesis, the first close connection between genes and biochemical function. Of the estimated 10,000–13,000 genes in the N. crassa genome, over 1200 genes have been identified by phenotype and/or their map location (NELSON et al. 1997 ; RADFORD and PARISH 1997 ). More than 2000 unique cDNAs have been sequenced (NELSON et al. 1997 ; ZHU et al. 2001 ; M. A. NELSON personal communication). Significantly, 60% of these unique cDNAs have no known homologs in the Saccharomyces cerevisiae genome (GOFFEAU et al. 1996 ) or in genes identified in other organisms.

As a study in contrasts this project is linked by strategy and resources to the Pneumocystis (the major killer of AIDs patients) genome project, thereby representing the highs and lows of the fungal kingdom (CUSHION and ARNOLD 1997 ). One is free living, and one has an obligate lifestyle. One is nested phylogenetically within a large diverse group of ascomycetes, while the other is alone on a branch arising basally in the ascomycete lineage (BERBEE and TAYLOR 1993 ). By comparing these organisms' entire genomes we can begin to understand what it is to be a fungus and what genes define a fungus by reference to other sequenced fungal genomes, such as those of S. cerevisiae and Schizosaccharomyces pombe. Initial results on the Pneumocystis genome project are presented in this issue (SMULIAN et al. 2001 ).

The N. crassa genome is estimated to be 42.9 Mb in size (ORBACH 1992 ). Mapping and sequencing of this genome are underway. Our strategy (Fig 1) begins with tiling each of the seven chromosomes in parallel with nonoverlapping cosmids and linking adjacent "tiles" by hybridization to a common cosmid. This strategy relies upon the availability of a library sorted by chromosome to give a sevenfold speedup to the rate of physical mapping and to detect repeats in clones sorting to multiple chromosomes. The result will be a nonredundant chromosome walk across each chromosome with markers at 29-kb intervals on average and with a portrait of repeats (AIGN et al. 2001 ; BHANDARKAR et al. 2001 ; HALL et al. 2001 ). This procedure was used previously to create a physical map of the Aspergillus nidulans genome (PRADE et al. 1997 ) and is based on the procedure used for physical mapping of the S. pombe genome (MAIER et al. 1992 ; HOHEISEL et al. 1993 ; MIZUKAMI et al. 1993 ). The physical map can then be linked to the genetic map by a variety of strategies, including complementing mutations on the genetic map with cosmids on the physical map and hybridizing cosmids to pools of known cDNAs (BALDING and TORNEY 1997 ). Cosmid end-sequencing of this library is providing sequenced-tagged connectors to the underlying genomic sequence, genetic maps, and expressed sequence tag (EST) collections, as well as a shotgun sequencing of the N. crassa genome (MAHAIRAS et al. 1999 ). As with the Human Genome Project, the cosmid ends are being used to "map by sequencing" to 3.3 kb resolution and then will be used to carry out "sequence extension" to yield a final draft of the N. crassa genomic sequence.

View larger version (41K): In this window In a new window Download PPT slide   Figure 1. The organization of this genome project. The physical map of the N. crassa genome is being constructed at two resolutions by tiling chromosomes with bacterial artificial chromosomes (BACs; Germany; AIGN et al. 2001 ) and cosmids [University of Georgia (UGA); BHANDARKAR et al. 2001; HALL et al. 2001]. To assist in gene identification an EST project is underway at University of New Mexico (UNM; NELSON et al. 1997 ), University of Oklahoma (OU), and Dartmouth Medical School (DMS; ZHU et al. 2001 ). Cosmid end-sequencing of the library described here is being carried out at UGA. The genomic sequencing is being carried out in a shared fashion by partnering institutions, using shotgun subcloning of cosmids and BACs tiled in the physical map (BEAN et al. 2001 ). Cosmids or BACs are physically sheared, end-repaired, phosphorylated, size-selected to 1–4 kb, and ligated into SmaI-digested pUC18. Gaps in sequence assemblies are closed by primer walking. The data generated are available on the world wide web in both raw and annotated form at http://gene.genetics.uga.edu and http://www.mips.biochem.mpg.de/proj/neurospora/.

In support of this project, our approach to obtain the N. crassa physical map has been to isolate N. crassa chromosomes by pulsed-field gel electrophoresis (PFGE), taking advantage of translocation strains to resolve similarly sized chromosomes (ORBACH et al. 1988 ), and to use them as probes against a large new cosmid library in a modified cosmid vector. This approach allows the sorting of cosmids on the basis of their hybridization to chromosome-specific probes (BRODY et al. 1991 ). In addition, Orbach and Sachs generated a cosmid library expected to contain 3.8 N. crassa genome equivalents in the vector pMOcosX (ORBACH 1994 ); this library has also been sorted with chromosome-specific probes in these studies.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains:
Most strains are available from the Fungal Genetics Stock Center (FGSC), Kansas City, Kansas. N. crassa 74-OR23-1A (FGSC no. 987) was the source of genomic DNA, and the sources of all strains are given in the tables and figures. Chromosomal DNA was derived from translocation strains, T(VIIL IVR)T54M50 (FGSC no. 2466), which was in turn derived from 74-OR23-1A by UV irradiation (PERKINS 1997 ), and T(IVR VIR)ALS159 (FGSC no. 2100), which was derived by UV irradiation from rg-1, cr-1 (PERKINS 1997 ).

DNA preparations:
Intact N. crassa chromosomal DNA was prepared by the agarose spheroplast method (ORBACH et al. 1988 ). N. crassa 74-OR23-1A genomic DNA was prepared by the method of YELTON et al. 1984 . Cosmid DNAs were isolated from overnight cultures grown in Luria broth (LB) media using the High Pure DNA isolation kit (Roche).

Construction of cosmid vector pLorist6Xh:
Lorist6 (GIBSON et al. 1987 ) was digested with BamHI and ligated to a BamHI-XhoI linker oligonucleotide (5'-GATCCCTCGAGG-3') to generate an intermediate plasmid pLorist6X, which, using the half-site fill-in method (ZABAROVSKY and ALLIKMETS 1986 ), would allow cloning of non-size-fractionated inserts without generation of chimeric clones. To allow insertion of a selectable marker for fungal transformation, pLorist6X was then digested with SalI. The ends were filled in using the Klenow fragment and treated with thermosensitive alkaline phosphatase (GIBCO-BRL, Gaithersburg, MD) to dephosphorylate the blunt ends. The 1.4-kb HpaI fragment from pCB1004 (CARROLL et al. 1994 ) encoding bacterial hygromycin B phosphotransferase (Hy^R) gene under the control of the A. nidulans trpC promoter was then ligated with SalI-digested pLorist6X to generate the new Lorist6 variant, pLorist6Xh (Fig 2). This vector can be used directly for fungal transformation with the hygromycin phosphotransferase as the selectable marker.

View larger version (15K): In this window In a new window Download PPT slide   Figure 2. The new vector pLorist6Xh has a XhoI site for automatic insert size selection and a hygromycin cassette for fungal transformation. The vector pLorist6Xh is derived from Lorist6 (GIBSON et al. 1987 ). Both vectors, pLorist6Xh and pMO-cosX, are similar in possessing an XhoI cloning site and hygromycin-resistance cassette (HyR). Genes shown include hygromycin B phosphotransferase (HyR), neomycin phosphotransferase (neoR), and bacteriophage genes (P, O, cII, and cro). Terminators are denoted with a T.

Genomic DNA library construction:
The vector pLorist6Xh was digested with XhoI followed by a partial fill-in reaction with Klenow using dTTP and dCTP. N. crassa genomic DNA (10 µg) was digested with MboI (at a concentration between 0.01 and 0.016 units/µg of DNA) for 1 hr at 37°. The MboI-digested genomic DNA was partially filled with dGTP and dATP using Klenow. Ligation reactions were set up using varying amounts of the vector (1–2 µg) and insert DNA (0.5–2.0 µg). After overnight incubation at 4°, the DNA was precipitated using sodium acetate and ethanol. The DNA was finally resuspended in 4 µl of 10 mM Tris-HCl, 0.1 mM EDTA (pH 8.0) buffer. A 0.4-µl aliquot was used for a test packaging to judge the efficiency of the ligation, followed by a full-scale packaging reaction using the Gigapack II XL packaging extracts (Stratagene, La Jolla, CA). The packaged phage was plated using Escherichia coli DH5 (mcr^-) cells. Kanamycin-resistant colonies (12,000) were picked to 125 microtiter plates (designated H001–H125) containing LB with kanamycin (50 µg/ml). A copy of the new pLorist6Xh library has been deposited with the FGSC for distribution to the research community (Fungal Genetics Stock Center, Department of Microbiology, University of Kansas Medical Center, Kansas City, KS 66160-7420; MCCLUSKEY and KINSEY 2000 ).

The pcosAX library was kindly provided by R. Aramayo (MCCLUSKEY and KINSEY 2000 ). A preexisting pMOcosX library (plates originally designated X1–X25 are now X107–X131, and plates originally designated as G1–G25 are now X132–X156) was obtained from the FGSC (MCCLUSKEY and KINSEY 2000 ).

Pulsed-field gel electrophoresis:
PFGE was performed using CHEF-DRII (Bio-Rad Laboratories, Richmond, CA) units. Gels (150 ml) of 0.8% type IV agarose (Amresco) were run in 1x Tris Acetate EDTA (TAE) buffer at 14°. Three sets of separation conditions were used to separate the chromosomes representing the seven linkage groups of N. crassa. In the first, a two-step procedure resolved the two largest chromosomes corresponding to linkage groups I and V: step 1, 32 V with 11,000-sec pulse times for 192 hr; and step 2, 54 V with 3600-sec pulse times for 84 hr. The second separation conditions resolved the chromosomes of intermediate sizes corresponding to linkage groups II, III, and IV. This separation used three steps, each at 60 V: step 1, 3000 sec pulse times for 73 hr; step 2, 2700-sec pulse times for 30 hr; and step 3, 2200-sec pulse times for 88 hr. The third set of conditions resolved the smallest chromosomes corresponding to linkage groups VI and VII. This separation also used three steps, all at 50 V: step 1, 3000-sec pulse times for 73 hr; step 2, 2700-sec pulse times for 30 hr; and step 3, 2200-sec pulse times for 99 hr.

Chromosomal DNA isolation and labeling:
Chromosomal DNA was isolated from agarose slices obtained from the PFGE gels with the Prep-A-Gene DNA isolation kit (Bio-Rad). The DNA was radiolabeled using the High Prime DNA labeling kit (Roche Molecular Biochemicals) according to the manufacturer's directions, using between 20 and 50 µCi of -labeled ³²P-dCTP (Amersham, Arlington Heights, IL) per reaction with a specific activity >1 x 10⁸ dpm/µg (BRODY et al. 1991 ).

DNA hybridization:
Cosmid libraries were stamped onto nylon membranes (S&S Nytran Maximum Strength Plus (Schleicher & Schuell, Keene, NH), PALL Biodyne B (Gelman, Ann Arbor, MI), or Hybond-XL (Amersham) with a 96-pin replicator in a 3 x 3 array of 9 microtiter plates per membrane manually or in a 6 x 6 array of up to 36 microtiter plates per membrane with a BioGrid high density stamping robot (BioRobotics). Inocula were allowed to grow overnight at 37° on LB agar plates containing kanamycin (50 µg/ml) for the pLorist6Xh library or containing ampicillin (50 µg/ml) for the pMOcosX library. Membranes were treated to lyse the colonies with the following: (1) 10% SDS for 5 min; (2) 1.5 M NaCl, 0.5 M NaOH for 5 min; (3) 0.5 M Tris-HCl, 1.5 M NaCl, pH 7.2, for 5 min; (4) 0.3 M NaCl, 0.3 M Na citrate, pH 7.0 (2x SSC), for 5 min. The treated membranes were air-dried for 30 min, and the DNA was cross-linked to the nylon membrane by UV radiation (3–12 min) with a UV cross-linker (Stratagene) or alternatively by baking at 80° for 2 hr.

A set of membranes representing the complete library was prehybridized at 65° for 2 hr with 10–12 ml of modified hybridization buffer containing casein hydrolysate instead of bovine serum albumin. Gentle agitation was provided in a hybridization oven (Hybaid). After prehybridization, 5–20 µl of radioactive probe (>10⁸ cpm) was added. The membranes were hybridized with probe for 12–18 hr at 65°. The membranes were then washed twice in 2x SSC, 1% SDS at 65° for 30 min with agitation. Two subsequent washes were done in 0.5x SSC at 65° for 30 min. Membranes were then removed, blotted dry, and electronically autoradiographed on a Packard Instant Imager (Packard, Meriden, CT). They were also exposed to X-ray film (BIOMAX MR) at -80° with intensifying screens.

Fungal transformation:
N. crassa transformation was performed as described by CASE et al. 1979 and modified to use Novozyme 234 (VOLLMER and YANOFSKY 1986 ).

DNA sequencing:
Cosmid H123E02, isolated from the pLorist6Xh cosmid library, complemented a quinic acid (qa)-2 mutant of N. crassa by transformation (CASE et al. 1979 ) and hybridized to cloned qa DNA. Since by these criteria it was likely to contain at least part of the qa gene cluster (GEEVER et al. 1989 ), it was sequenced using the shotgun subcloning method (ROE et al. 1996 ). Cells containing the cosmid were grown at 37° overnight (250 rpm) in a 500-ml Ehrlenmeyer flask containing 200 ml LB media. At least 75 µg of cosmid DNA was obtained by a double acetate cleared lysate method (ROE et al. 1996 ), modified as per http://www.genome.ou.edu/Db/AcetateProcV3.html). The cosmid DNA was nebulized (physically sheared) using an IPI Medical Products nebulizer 4207 at 10 psi for 2.5 min. The nebulized DNA was end-repaired, T4 polynucleotide kinase-treated, and size-selected (1.5–3.0 kb) using agarose gel electrophoresis for subcloning into SmaI-digested pUC18 (ROE et al. 1996 ). Electrocompetent E. coli XL1-Blue (Stratagene) cells were transformed with the ligated DNA and plated on LB-ampicillin agar with isopropyl thiogalactoside and X-GAL (1 mg each per plate). White colonies (576) were inoculated into six deep 96-well blocks (Marsh Biomedical, Rochester, NY) containing 1.2 ml of LB media. The blocks were incubated overnight at 37° on a rotary shaker at 250 rpm.

DNA sequencing templates were generated by a Biomek-2000-based double-stranded DNA sequencing template isolation procedure (ROE et al. 1996 ). Taq-polymerized cycle sequencing was performed using BigDye terminator chemistry according to manufacturer's specification (PE Applied Biosystems, Foster City, CA). Unincorporated dye-terminator was removed with ethanol precipitation. Gel electrophoresis was performed on a 96-lane ABI 377 sequenator. The same protocol for sequencing pUC18 templates was used to sequence 4406 cosmid ends from plates 1–40 of the Lorist6Xh library to provide gene assignments to cosmids with one difference: gel electrophoresis was performed on a 96-capillary ABI 3700 sequenator.

Sequencing data were processed and assembled by software packages Phred/Crossmatch/Phrap running on a Unix server (EWING et al. 1998 ; EWING and GREEN 1998 ). BLASTx searches were done in 600-bp sliding windows (with an overlap of 60 bp) along the cosmids against GenPept 114 (ALTSCHUL et al. 1990 ). Positions of coding regions in H123E02 were refined by FastX comparisons of H123E02 sequence against the identified BLASTx hits in GenPept 114 (Wisconsin package, version 10, Genetics Computer Group, Madison, WI).

cDNA library:
The cDNA library was kindly provided by M. A. Nelson and D. Natvig and is described in NELSON et al. 1997 .

GenBank accessions:
The GenBank accessions for cosmid H123EO2 and Lorist6Xh sequences are AC006498 and AF237704, respectively.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Cosmid libraries:
The cosmid vector pLorist6Xh, based on pLorist6 (GIBSON et al. 1987 ), was engineered as described in MATERIALS AND METHODS to contain a unique XhoI cloning site and a hygromycin B phosphotransferase (hph) cassette (CARROLL et al. 1994 ; Fig 2). By half-site fill-in of the XhoI site in the vector and half-site fill-in of genomic DNA digested with MboI, concatamers of genomic DNA cannot be cloned into the vector (ZABAROVSKY and ALLIKMETS 1986 ). The hph cassette allows direct selection for transformation of fungi with cosmid vector. One key difference between pLorist6Xh and pMOcosX (ORBACH 1994 ) is that pLorist6Xh has a origin of replication and not a colE1 origin. Thus, pLorist6Xh is maintained at lower copy numbers in bacteria, reducing the effects of the expression of cloned sequences toxic to bacterial growth. The hph cassette in pLorist6Xh is driven by the A. nidulans trpC promoter, whereas the cassette in pMOcosX is driven by the N. crassa cpc-1 promoter; the absence of N. crassa sequence in the vector backbone reduces potential complications for analyses of N. crassa sequences cloned into the vector. The hph cassette used in pLorist6Xh was previously modified to eliminate restriction enzyme sites, increasing the versatility of the vector (CARROLL et al. 1994 ). Finally, pLorist6Xh selection in bacteria is accomplished using kanamycin rather than ampicillin, offering cleaner selection for extended growth (GIBSON et al. 1987 ).

A cosmid library was generated for N. crassa 74-OR23-1A consisting of 12,000 clones in pLorist6Xh from genomic DNA as described in MATERIALS AND METHODS. In the studies described below, this new library was augmented with the existing Orbach-Sachs pMOcosX N. crassa genomic DNA cosmid library of 4800 clones available from FGSC (MCCLUSKEY and KINSEY 2000 ). Cosmid insert sizes were measured for 81 pLorist6Xh and 75 pMOcosX cosmids chosen at random by digesting each with BamHI and sizing the resulting fragments from each digestion-reaction on agarose gels. On the basis of these results (data not shown), the average insert size in each cosmid library was estimated to be 34 ± 0.7 kb. Given an average insert size of 34 kb and estimated genome size of 42.9 Mb (ORBACH 1992 ), the combined cosmid libraries consisting of 16,800 clones were estimated to contain at least 13 N. crassa genome equivalents. The number of genome equivalents may be larger because the average insert size of 34 kb may be downwardly biased due to comigrating bands that may have been counted as one fragment.

N. crassa karyotype:
We obtained separation of all seven chromosomes of N. crassa (Fig 3) using three different conditions for PFGE and two translocation strains. Fig 3A shows the separation of linkage groups I and V of the wild-type strain OR74A. Using a second set of conditions (Fig 3B), linkage groups II, III, and IV were well resolved. Using this second set of conditions, linkage groups I and V were not well resolved, nor were linkage groups VI and VII.

View larger version (25K): In this window In a new window Download PPT slide   Figure 3. All seven N. crassa chromosomes are separated by pulsed-field gel electrophoresis with the use of translocation strains ALS154 and T54M50. (A) Chromosomes corresponding to linkage groups I and V (linkage group designations) are separated by PFGE as described in MATERIALS AND METHODS for large chromosomes. A. nidulans and S. pombe chromosomes ranging in size from 2.9 to 5.0 Mb and 3.5 to 5.7 Mb, respectively, were used as size standards. (B) Chromosomes corresponding to linkage groups II, III, and IV (linkage group designations) are separated by PFGE as described in MATERIALS AND METHODS. (C) Chromosomes corresponding to linkage groups VI and VII are separated by PFGE as described in MATERIALS AND METHODS for small chromosomes. Lane 1, N. crassa chromosomes of the wild-type (OR74A) strain; lane 2, chromosomal DNA from translocation strain T(IVR VIR)ALS159 with the truncated translocation chromosome isolated; lane 3, chromosomal DNA from translocation strain T(VIIL IVR)T54M50 with a truncated translocation chromosome running at the same size as the 2.9-Mb chromosome corresponding to linkage group IV of A. nidulans.

The problem of resolving the chromosomes corresponding to linkage groups VI and VII, each 4.0 Mb in size, was solved by using a third set of conditions and two different translocation strains. In wild type, linkage groups VI and VII can be partially resolved under this third set of conditions as a doublet-band (Fig 3C lane 1). Translocation strain T(IVR VIR)ALS159 contains a quasiterminal translocation of the right arm of linkage group IV to the right arm of linkage group VI (PERKINS 1997 ). We used this strain to increase the size of the rest of linkage group VI, which now migrates at a position different from that of linkage group VII (Fig 3C, lane 2). The smaller chromosome visible in this lane (indicated by an arrow) presumably represents the shortened linkage group IV. Strain T(VII IVR) T54M50 contains a quasiterminal translocation of the left arm of linkage group VII to the right arm of linkage group IV (PERKINS 1997 ). We used this strain to isolate linkage group VI, which migrates as a unique band due to the altered size of the linkage group VII chromosome (Fig 3C, lane 3). ORBACH et al. 1988 and BALLARIO et al. 1989 also used this approach to remove the linkage group VII component out of the linkage group VI and VII doublet. The smaller chromosome visible in Fig 3C, lane 3 (indicated by an arrowhead) represents the shortened linkage group VII chromosome (BALLARIO et al. 1989 ). Assignments of chromosomes to linkage groups VI and VII were confirmed by hybridization of nine genetic markers on linkage group VI and four genetic markers on linkage group VII hybridized to the cosmid library sorted by chromosome (data not shown).

Assigning cosmids to chromosomes:
The procedure for assigning cosmids to chromosomes was to radiolabel chromosomal DNA from each of the seven chromosomes resolved as in Fig 3 and to hybridize these radiolabeled chromosomal DNA probes to cosmid libraries arrayed on nylon membranes. A total of 16,800 clones from both libraries (12,000 Lorist6Xh clones and 4800 pMOcosX clones available from FGSC) were hybridized twice to each chromosome probe. In the first hybridization analysis, which used manually stamped membranes, fewer cosmids could be assigned to any chromosome than in the second hybridization experiment, which used robotically stamped membranes (Table 1). This was because the robot more reproducibly delivered similar quantities of cells to replicate membranes (data not shown). By combining the results of these two hybridization analyses, 13,882 clones were assigned to one or more chromosomes. Assignments are available at http://gene.genetics.uga.edu.

How repeatable were these hybridization results? Of the 97,174 hybridization signals that represent 13,882 clones hybridized to each of seven chromosome-specific probes, 90% of the signals behaved similarly in both experiments. This percentage is termed the repeatability of cosmid assignments to linkage groups. On the basis of these data, cosmids were classified into three groups (Table 1). S-clones (74.6%) hybridized to only one of the chromosome-specific probes. R-clones (24.6%) hybridized to two or more chromosome probes, but not all probes. A-clones (0.7%) hybridized to every chromosome probe (Table 2).

The number of S-clones identified by each chromosome-specific probe was highly correlated to the physical size of the chromosome used as the probe in both hybridization experiments (Table 1). That the number of clones identified by hybridization with chromosome-specific probes was linearly proportional to the sizes of the chromosome used as probes indicates that the libraries are representative of the genome. In the first experiment using manually stamped membranes, the number of S-clones obtained with the linkage group IV probe was less than the number that would be expected on the basis of the strong correlation observed in the other experiments. This discrepancy arose because we were unable to score some of the membranes that contained cosmids from the pMOcosX library in this instance.

The R- and A-clones hybridized to two or more chromosomes. Of the 3419 R-clones, which hybridized to multiple chromosomes but not to all chromosomes, 2830 clones hybridized to two chromosomes, 381 to three, 126 to four, 52 to five, and 30 clones to six chromosomes. Hybridization of the R-clones and the A-clones to probes representing multiple chromosomes is likely due to repeated sequences present in more than one probe preparation or incomplete resolution of the two largest linkage groups.

There is a set of 107 A-clones that hybridize to all seven chromosomes. Out of these, 52 clones contain the rDNA sequences based on hybridization experiments with the N. crassa rDNA repeat (Table 3).

Experimental determination of gene representation in cosmid libraries and analysis of the validity of assigning cosmids to chromosome by chromosome-specific probing:
The theoretical expectation discussed above is that the pLorist6Xh library contained at least nine genome equivalents and the commonly used portion of the pMOcosX library contained at least four genome equivalents. Therefore, a given single-copy gene should be represented, on average, by 13 independent clones in the libraries. To test this, five different plasmids, each containing a cloned gene from a different linkage group, were used to probe the libraries (Table 3). Each gene was represented by at least 9 or more independent clones identified by two independent hybridizations with gene-specific probes. Variation in the number of clones hybridizing to a particular gene plasmid is within the range of variation observed in the number of probes hybridizing to clones in the library (http://gene.genetics.uga.edu; see physical maps). Complementing clones originally identified by probes with single-copy genes in Table 3 from known linkage groups were all assigned to the same linkage group as found by chromosome-specific probings.

One additional experiment was performed to assess the accuracy of cosmid assignments to linkage groups: the sequencing of 4406 cosmid ends from the first 40 plates of the Lorist6Xh library (http://gene.genetics.uga.edu; see cosmid ends). Each cosmid end was used as a query sequence in a BLASTn search to assign sequenced N. crassa genes in public databases and on the genetic map to cosmids. A total of 96 genes on the genetic map were assigned to cosmids by hybridization (Table 3), by complementation (Table 4), or by cosmid end-sequencing. A total of 89/96 = 0.93 ± 0.3 or 93% ± 3% of the genes assigned on the basis of chromosome-specific probes were concordant with their independent assignment based on genetic mapping (PERKINS et al. 2000 ). This percentage is termed the concordancy in cosmid assignments to linkage groups.

The ribosomal rRNA genes are clustered on linkage group V (PERKINS et al. 1982 ). Their representation in the pLorist6Xh and pMOcosX libraries, and a third genomic DNA library prepared from a mat-a wild-type strain (pcosAX), was examined by probing with an rDNA probe. Given that there are 185 copies of the rDNA repeat (KRUMLAUF and MARZLUF 1980 ), clones for rDNA were underrepresented in each of the cosmid libraries, but the underrepresentation was less severe in the pLorist6Xh collection (113 clones out of 12,000) than in the other two libraries tested (pMOcosX, 12 clones out of 4800; and pcosAX, 20 clones out of 6300).

To search for telomere sequences in the library, preliminary hybridization tests were conducted with an end-labeled telomeric oligonucleotide repeat (TTAGGG)₄ probe (SCHECHTMAN 1987 ). The negative results (data not shown) did not indicate the presence of cosmids containing telomere sequences in the library (though positive controls containing cloned telomeric DNA were detected in parallel), suggesting that it is unlikely that the libraries contain telomere sequences.

Fungal transformation:
Cosmid clones identified by hybridization with gene probes for am, his-3, pan-2, qa-2, trp-1, and trp-3 were tested to determine whether they would functionally complement the mutant phenotype following transformation. In all cases, transformation with the cosmid DNAs successfully complemented the mutant phenotype (Table 4).

EST density:
To estimate the feasibility of anchoring EST data obtained from cDNAs (NELSON et al. 1997 ) to N. crassa genomic DNA cloned in these cosmid libraries, a pool of four to seven different cosmid S-clones, one from each linkage group, were hybridized to 2976 unique cDNAs arrayed in duplicate on Nylon membranes. In total, 12 such pools of four to seven different S-clones were used as probes (Table 5). Each S-clone used was different; e.g., cosmids in each pool were nonoverlapping with cosmids in other pools. Each pool of S-clones hybridized to a distinct collection of ESTs. A total of 67 unique ESTs in the 2976-clone sample hybridized to the 79 cosmids sampled. Given that the 79 cosmids represent 2686 kb of genomic DNA, then, on average, this EST library contains one EST per 40 kb of genomic sequence, or slightly <1 EST per cosmid. A hard problem with eukaryotes is identifying genes in anonymous DNA sequence, and the presence of an EST in each cosmid on average provides a starting point for gene identification.

Sequencing of qa gene cluster cosmid:
A cosmid (H123E02) found to complement the qa-2 mutation of N. crassa in this study was completely sequenced to obtain an estimate of gene density in N. crassa as shown in Fig 4. The cosmid contained most of the qa cluster described by GEEVER et al. 1989 . A total of 12 genes were found by open reading frame (ORF) identification and BLASTx searches (ALTSCHUL et al. 1990 ; Table 6). Two of the genes (elf-1B and rehydrin homolog) were found by sequencing insert ends in a cDNA library (NELSON et al. 1997 ) and confirmed by a tBLASTx search (ALTSCHUL et al. 1990 ). An additional leucine tRNA gene was found by sequencing (HUIET et al. 1984 ) and confirmed by a BLASTn search (ALTSCHUL et al. 1990 ). With this cosmid insert being 44 kb, this translated into 1 gene per 3.4 kb. In the N. crassa sequenced cosmid, similarity to a fatty acid transporter gene (fat1) in the plant pathogen Cochliobolus heterostrophus was identified, which is consistent with N. crassa serving as a good model for related plant pathogens.

View larger version (19K): In this window In a new window Download PPT slide   Figure 4. Cosmid H123-E02 contains 13 genes in 44 kbp of insert. Most of the qa cluster for quinic acid metabolism is found on this cosmid. The Lorist6Xh is 6.6 kbp. +, Watson strand; -, the Crick strand.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Resources described here provide a foundation for functional genomics in nearly 60 years of Neurospora genetics, enabling researchers a unique opportunity to understand the emergent properties of more complex eukaryotes like development, biological clocks (LEE et al. 2000 ), and transvection (ARAMAYO and METZENBERG 1996 ) in a simple microbial system. This article describes an important step toward the goals of the N. crassa genome project, the generation of a new comprehensive cosmid library for N. crassa to be used for creating a physical map and ultimately the complete sequence of the N. crassa genome. The library was made in a modified cosmid vector pLorist6Xh, which contains a hygromycin phosphotransferase hph gene cassette, allowing the direct selection of clones from this library for fungal transformation. The pLorist6Xh ( origin of replication) has an advantage over pMOcosX and other vectors that utilize an E. coli origin of replication (col E1 ori). The vector pLorist6Xh is low-copy number, and cosmid/cosmid hybridizations are observed to have a reduced background. The pLorist6Xh vector has been recently used to construct cosmid libraries in Pneumocystis carinii (M. T. CUSHION and A. G. SMULIAN, unpublished results), A. flavus, and Candida albicans (H. S. KELKAR and J. ARNOLD, unpublished results).

Clones from the new pLorist6Xh library (12,000) were pooled with 4800 clones from an existing pMOcosX cosmid library made by M. Orbach and M. Sachs (MCCLUSKEY and KINSEY 2000 ) to obtain an approximate total of 16,800 cosmid clones. This theoretically yields a representation of at least 13 genome equivalents. Past use of cosmid libraries on the A. nidulans physical mapping project and current examination of this library indicate that the use of two vectors is beneficial in reducing cloning bias (Table 3).

The use of the restriction enzyme MboI (a four-base cutter) that cleaves regardless of cytosine methylation is considered a good way to obtain partial digests of N. crassa DNA for cloning. However, it is unlikely to cut in telomeric regions that are composed of TTAGGG repeats (SCHECHTMAN 1987 ), nor are these repeats likely to have natural ends compatible with the half-filled XhoI in the vector used for cloning. It is, however, likely that the cosmid library contains mitochondrial sequences since no specific steps were taken to eliminate mitochondrial DNA as with the Aspergillus mapping project (BRODY et al. 1991 ). These sequences should be detectable by hybridization with mitochondrial DNA probes.

We are capitalizing on the electrophoretic karyotype of N. crassa to map all seven chromosomes in parallel (AIGN et al. 2001 ; BHANDARKAR et al. 2001 ; HALL et al. 2001 ). The PFGE gel-isolated chromosomal DNA allowed us to subdivide our cosmid libraries into chromosome-specific subcollections, in which 74.6% (= 100 x 10,356/13,882) of the cosmids were uniquely assignable to individual chromosomes (Table 2). The assignment of cosmids to chromosomes by chromosome-specific probes can be compared with published chromosome assignments of cosmids in the widely used pMOcosX portion of our library methods (ROSA et al. 1997 ; SCHMIDHAUSER et al. 1997 ; WAN et al. 1997 ). These independent assignments are 91% identical. This 91% concordancy in assignments is not statistically different from the 93 ± 3% concordancy reported here.

If the concordancy and repeatability of cosmid assignments to linkage groups were not different, the conclusion would be that the repeatability of hybridizations of chromosome-specific probes to cosmids is sufficient to explain differences in cosmid assignments to linkage groups on the basis of the genetic map vs. the PFGE karyotype. Estimated concordancy (93 ± 3%) and repeatability (90%) of cosmid assignments to linkage groups were not significantly different. The conclusion is that any discordancy observed is explainable by the extent to which chromosome-specific probings are not repeatable.

Subdividing the library by chromosomes produces a resource that is valuable to the community. The subcollection of cosmids with unique chromosome assignments (the 10,356 S-clones in Table 1) by itself constitutes an eight-genome equivalent cosmid library. Each chromosome-specific collection from this study is available from FGSC and will allow accelerated cloning of genes by complementation when their chromosome assignment is known, with varied pooling schemes described before (METZENBERG and KANG 1987 ; BALDING and TORNEY 1997 ).

The collections of R- and A-clones may prove useful in the identification of DNA repeat families (ENKERLI et al. 1997 ). For example, a subset of clones containing rDNA sequences (52 clones) was found among the A-clones (107 clones). The collection of A-clones in the A. nidulans project mapped to the centromeres (PRADE et al. 1997 ). Some of these N. crassa A-clones may provide entry points to other centromeres (CENTOLA and CARBON 1994 ) and may help in the identification of additional families of repeats beyond the rDNA multigene family. Subcollections of R- and A-clones are available from the FGSC as well.

By virtue of constructing a physical map of the A. nidulans genome (PRADE et al. 1997 ) with markers every 29 kb, a surprising observation was made. In A. nidulans there is an alternating banding pattern in the distribution of repeated sequences along each chromosome. One striking contrast between the genomes of N. crassa and A. nidulans is evidence for a process of premeiotic inactivation of duplicated sequences known as repeat-induced point (RIP) mutation in N. crassa but not in A. nidulans (SELKER 1990 ). When a nucleus containing a duplicated sequence participates in meiosis, both copies of any duplicated sequences are selectively mutated. The N. crassa genome thus has a method for detecting and altering duplicated sequences. The prediction is that the overall genome organization of repeats in N. crassa will be strikingly different from that of A. nidulans (PRADE et al. 1997 ). Currently, in N. crassa, only 25% of the cosmids appear to have repeats that are located on multiple chromosomes as opposed to 34% in A. nidulans. The presence of RIPing in N. crassa may not be the only explanation for this observation because A. flavus appears to possess an even higher proportion of uniquely assignable clones (see Table 2; H. S. KELKAR and J. ARNOLD, unpublished observations).

The majority of the R-clones involve clones hybridizing to chromosomes corresponding to linkage groups I or V and may represent insufficient separation of chromosomes corresponding to linkage groups I and V by PFGE. This problem would arise even if the cosmid libraries were made from gel-isolated chromosomes (ZOLAN et al. 1992 ). If we set aside the chromosome corresponding to linkage group V, then the fraction of detectable repeats located on more than one chromosome is 6%. If we add to this the rDNA sequences with a repeat unit of 8600–9700 bp iterated tandemly 185 times on linkage group V (KRUMLAUF and MARZLUF 1980 ), we obtain 1.59–1.79 Mb of tandemly repeated rDNA or 4% repetitive rDNA to be added to the 6% of repetitive DNA dispersed on more than one chromosome. Our estimate of 10% repetitive DNA in the genome is then identical to that in KRUMLAUF and MARZLUF 1980 . The only difference is that our estimate of 4% rDNA in the N. crassa genome is smaller than the 7% rDNA reported in KRUMLAUF and MARZLUF 1980 .

Initial sequencing of a cosmid (H123E02) from the pLorist6Xh library also provides some information on the number of genes in the N. crassa genome. The 13 genes found is the same number found in BEAN et al. 2001 . Using the relationship of gene density to genome size, KUPFER et al. 1997 obtained a prediction of at least 8000–9000 genes in the N. crassa genome. From sequencing random cDNAs, NELSON et al. 1997 estimated 10,000–13,000 genes in the N. crassa genome. From sequencing one cosmid clone of N. crassa and from accompanying sequence and transcription analyses of a total of 260 kb (Table 7), it appears that estimating 11,000 genes in N. crassa is not unreasonable with 1 gene every 4 kb. N. crassa is thus as gene rich as Drosophila melanogaster with 13,600 genes (ADAMS et al. 2000 ). In the N. crassa cosmid that we sequenced, the average length of identified genes was 1.5 kb (Table 6). BEAN et al. 2001 report an average gene length of 0.9 kb in their sequenced cosmid (Table 7). NELSON et al. 1997 report an average insert size of 1.52 kb on 1409 cDNAs, and this insert size helps to establish a lower limit on the average size of a gene. If the gene is 1.52–2.00 kb, then at least 53–70% of the N. crassa genome is expressed. This gene density in N. crassa of 1 gene every 4 kb is lower than that in S. cerevisiae (1 gene every 2 kb) or P. carinii (1 gene every 2 kb; SMULIAN et al. 2001 ) and approaches that of Caeneorhabditis elegans (1 gene every 5 kb C. ELEGANS CONSORTIUM 1998); and that of D. melanogaster (1 gene every 13 kb for a total of 13,600 genes; ADAMS et al. 2000 ). Initial N. crassa cDNA sequencing indicates that 56% of these genes have no homology to genes in GenBank, including those in the S. cerevisiae genome (NELSON et al. 1997 ), a level typical of fungal genomes (PRADE et al. 2001 ). In C. elegans, 58% of the genes have protein products with no BLASTx matches outside the Nematoda. Thus, >50% of the time we can expect a gene with a novel function in N. crassa.

	ACKNOWLEDGMENTS

The support and advice of Dr. Bruce A. Roe and his laboratory (Oklahoma University, Norman, Oklahoma), as well as the generous exchange of materials (the cDNA library) and ideas with Drs. Mary Anne Nelson and Don Natvig (University of New Mexico, Albuquerque, NM) were very valuable. We appreciate comments on the manuscript by David Perkins, Bob Metzenberg, Ulrich Schulte, Jorg Hoheisel, and anonymous reviewers. J.A., H.K., M.O., and M.S. performed cosmid library construction. J.G., H.K., and S.C. played a role in PFGE. J.G. performed DNA/DNA hybridizations. M.C., S.O., and H.K. carried out complementation experiments. D.H., H.K., J.R.W., J.K.W., and M.J.W. performed sequencing and sequence analysis. C.K. enabled the design, acquisition, and use of robotics instrumentation on the project. We gratefully acknowledge the support from the National Science Foundation in the form of grants MCB-9630910 (M.C., S.C., M.S., J.A.) and BIR-9512887 (J.A., S.C., C.K.), and from the Georgia Research Alliance (J.A.).

Manuscript received September 1, 2000; Accepted for publication December 13, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, M. D., S. E. CELNIKER, R. A. HOLT, C. A. EVANS, and J. D. GOCAYNE et al., 2000  The genome sequence of Drosophila melanogaster.. Science 287:2185-2195[Abstract/Free Full Text].

AIGN, V., U. SCHULTE, and J. D. HOHEISEL, 2001  Hybridization-based mapping of Neurospora crassa linkage groups II and V. Genetics 157:1015-1020[Abstract/Free Full Text].

ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS, and D. J. LIPMAN, 1990  Basic local alignment search tool. J. Mol. Biol. 215:403-410[Medline].

ARAMAYO, R. and R. L. METZENBERG, 1996  Meiotic transvection in fungi. Cell 86:103-113[Medline].

BALDING, D. J. and D. C. TORNEY, 1997  The design of pooling experiments for screening a clone map. Fungal Genet. Biol. 21:302-307[Medline].

BALLARIO, P., G. MORELLI, E. SPORENO, and G. MACINO, 1989  Cosmids from the Vollmer-Yanofsky library identified with a chromosome VII probe. Fungal Genet. Newsl. 36:38-39.

BEADLE, G. W. and E. L. TATUM, 1941  Genetic control of biochemical reactions in Neurospora.. Proc. Natl. Acad. Sci. USA 27:499-506[Free Full Text].

BEAN, L. E., W. H. DVORACHEK, E. L. BRAUN, A. ERRETT, and G. S. SAENZ et al., 2001  Analysis of the pdx-1 (snz-1/sno-1) region of the Neurospora crassa genome: correlation of pyridoxine-requiring phenotypes with mutations in two structural genes. Genetics 157:1067-1075[Abstract/Free Full Text].

BERBEE, M. L. and J. W. TAYLOR, 1993  Dating the evolutionary radiations of the true fungi. Can. J. Bot. 71:1114-1127.

BHANDARKAR, S. M., S. A. MACHAKA, S. S. SHETE, and R. N. KOTA, 2001  Parallel computation of a maximum-likelihood model for chromosome reconstruction. Genetics 157:1021-1043[Abstract/Free Full Text].

BRODY, H., J. GRIFFITH, A. J. CUTICCHIA, J. ARNOLD, and W. E. TIMBERLAKE, 1991  Chromosome-specific recombinant libraries from the fungus Aspergillus nidulans.. Nucleic Acids Res. 19:3105-3109[Abstract/Free Full Text].

CARROLL, A. M., J. A. SWEIGARD, and B. VALENT, 1994  Improved vectors for selecting resistance to Hygromycin. Fungal Genet. Newsl. 41:22.

CASE, M. E., M. SCHWEIZER, S. R. KUSHNER, and N. H. GILES, 1979  Efficient transformation of Neurospora crassa by utilizing hybrid plasmid DNA. Proc. Natl. Acad. Sci. USA 76:5259-5263[Abstract/Free Full Text].

Genome sequence of the nematode C. elegans: a platform for investigating biology. (1998) Science 282:2012-2018[Abstract/Free Full Text].

CENTOLA, M. and J. CARBON, 1994  Cloning and characterization of centromeric DNA from Neurospora crassa.. Mol. Cell. Biol. 14:1510-1519[Abstract/Free Full Text].

CUSHION, M. T. and J. ARNOLD, 1997  Proposal for a Pneumocystis Genome Project. J. Euk. Microbiol 44:7S[Medline].

DAVIS, R. H., 2000 Neurospora Contributions of a Model Organism. Oxford University Press, New York.

ENKERLI, J., G. BHATT, and S. F. COVERT, 1997  Nht1, a transposable element cloned from a dispensable chromosome in Nectria haematococca MPVI. Mol. Plant Microbe Int. 10:742-749.

EWING, B. and P. GREEN, 1998  Base-calling of automated sequencer traces using Phred II. Error probabilities. Genome Res. 8:186-194[Abstract/Free Full Text].

EWING, B., L. HILLIER, M. C. WENDL, and P. GREEN, 1998  Base-calling of automated sequencer traces using Phred I. Accuracy assessment. Genome Res. 8:175-185[Abstract/Free Full Text].

GEEVER, R. F., L. HUIET, J. A. BAUM, B. M. TYLER, and V. B. PATEL et al., 1989  DNA sequence organization and regulation of the qa (quinic acid) gene cluster of Neurospora crassa.. J. Mol. Biol. 207:15-37[Medline].

GIBSON, T. J., A. ROSENTHAL, and R. H. WATERSTON, 1987  Lorist6, a cosmid vector with BamHI, NotI, ScaI, and HindIII cloning sites and altered neomycin phosphotranserase gene expression. Gene 53:283-286[Medline].

GOFFEAU, A., B. G. BARRELL, H. BUSSEY, R. W. DAVIS, and B. DUJON et al., 1996  Life with 6000 genes. Science 274:546-567[Abstract/Free Full Text].

HALL, D., J. WANG, and S. M. BHANDARKAR, 2001  ODS2: a multiplatform software application for creating integrated physical and genetic maps. Genetics 157:1045-1056[Abstract/Free Full Text].

HOHEISEL, J. D., E. MAIER, R. MOTT, L. MCCARTHY, and A. V. GRIGORIEV et al., 1993  High resolution cosmid and P1 maps spanning the 14 Mb genome of the fission yeast S. pombe.. Cell 73:109-120[Medline].

HUIET, L., B. M. TYLER, and N. H. GILES, 1984  A leucine tRNA gene adjacent to the QA gene cluster of Neurospora crassa.. Nucleic Acids Res. 12:5757-5765[Abstract/Free Full Text].

KRUMLAUF, R. and G. A. MARZLUF, 1979  Characterization of the sequence complexity and organization of the Neurospora crassa genome. Biochemistry 18:3705-3713[Medline].

KRUMLAUF, R. and G. A. MARZLUF, 1980  Genome organization and characterization of the repetitive and inverted repeat DNA sequences in Neurospora crassa.. J. Biol. Chem. 255:1138-1145[Free Full Text].

KUPFER, D. M., C. A. REECE, S. W. CLIFTON, B. A. ROE, and R. A. PRADE, 1997  Multicellular ascomycetous fungal genomes contain more than 8000 genes. Fungal Genet. Biol. 21:364-372[Medline].

LEE, K., J. J. LOROS, and J. C. DUNLAP, 2000  Interconnected feedback loops in the Neurospora circadian system. Science 289:107-110[Abstract/Free Full Text].

MAHAIRAS, G. G., J. C. WALLACE, K. SMITH, S. SWARTZELL, and T. HOLZMAN et al., 1999  Sequence-tagged connectors: a sequence approach to mapping and scanning the human genome. Proc. Natl. Acad. Sci. USA 17:9739-9744.

MAIER, E., J. D. HOHEISEL, L. MCCARTHY, R. MOTT, and A. V. GRIGORIEV et al., 1992  Complete coverage of the Schizosaccharomyces pombe genome in yeast artificial chromosomes. Nat. Genet. 1:273-277[Medline].

MCCLUSKEY, K. and J. A. KINSEY, 2000  Fungal Genetics Stock Center catalogue of strains. Fungal Genet. Newsl. 47(Suppl.):6-9. Ed. 8.

METZENBERG, R. L. and S. KANG, 1987  A simple strategy for identifying a single complementing clone in a 96-well microtiter dish. Fungal Genet. Newsl. 34:44-45.

MIZUKAMI, T., W. I. CHANG, I. GARKAVTSEVE, N. KAPLAN, and D. LOMARDI et al., 1993  A 13 kb resolution cosmid map of the 14 Mb fission yeast genome by nonrandom sequence-tagged site mapping. Cell 73:121-132[Medline].

NELSON, M. A., S. KANG, E. L. BRAUN, M. E. CRAWFORD, and P. L. DOLAN et al., 1997  Expressed sequences from conidial, mycelial and sexual stages of Neurospora crassa.. Fungal Genet. Biol. 21:348-363[Medline].

ORBACH, M. J., 1992  One-liners. Fungal Genet. Newsl. 39:92.

ORBACH, M. J., 1994  A cosmid with a HyR marker for fungal library construction and screening. Gene 150:159-162[Medline].

ORBACH, M. J., D. VOLLRATH, R. W. DAVIS, and L. YANOFSKY, 1988  An electrophoretic karyotype of Neurospora crassa.. Mol. Cell. Biol. 8:1469-1473[Abstract/Free Full Text].

PATEL, V. B., M. SCHWEIZER, C. C. DYKSTRA, S. R. KUSHNER, and N. H. GILES, 1982  Genetic organization and transcriptional regulation in the qa gene cluster of Neurospora crassa.. Proc. Natl. Acad. Sci. USA 78:5783-5787.

PERKINS, D. D., 1997  Chromosome rearrangements in Neurospora and other filamentous fungi. Adv. Genet. 36:239-398[Medline].

PERKINS, D. D., A. RADFORD, D. NEWMEYER, and M. BJORKMAN, 1982  Chromosomal loci of Neurospora crassa.. Microbiol. Rev. 46:426-570[Free Full Text].

PERKINS, D. D., M. S. SACHS, and A. RADFORD, 2000  Chromosomal Loci of Neurospora crassa.. Academic Press, San Diego in press.

PRADE, R. A., J. GRIFFITH, K. KOCHUT, J. ARNOLD, and W. E. TIMBERLAKE, 1997  In vitro reconstruction of the Aspergillus (=Emericella) nidulans genome. Proc. Natl. Acad. Sci. USA 94:14564-14569[Abstract/Free Full Text].

PRADE, R. A., P. AYOUBI, S. KRISHNAN, S. MACWANA, and H. RUSSELL, 2001  Accumulation of stress and cell wall degrading enzyme associated transcripts during asexual development in Aspergillus nidulans.. Genetics 157:957-967[Abstract/Free Full Text].

RADFORD, A. and J. H. PARISH, 1997  The genome and genes of Neurospora crassa.. Fungal Genet. Biol. 21:258-266[Medline].

RANDALL, T. A. and R. L. METZENBERG, 1998  The mating type locus of Neurospora crassa: identification of an adjacent gene and characterization of transcripts surrounding the idiomorphs. Mol. Gen. Genet. 259:615-621[Medline].

ROE, B. A., J. S. CRABTREE and A. S. KHAN, 1996 DNA Isolation and Sequencing. Wiley, New York.

ROSA, A. L., S. D. HEDO, E. D. TEMPORINI, G. A. BORIOLI, and M. R. MAUTINO, 1997  Mapping chromosome landmarks in the Centromere I region of Neurospora crassa.. Fungal Genet. Biol. 21:315-322[Medline].

RUSSELL, P. J., S. WAGNER, K. D. RODLAND, R. L. FEINBAUM, and J. P. RUSSELL et al., 1984  Organization of the ribosomal ribonucleic acid genes in various wild-type and wild-collected strains of Neurospora.. Mol. Gen. Genet. 196:275-282[Medline].

SCHECHTMAN, M. G., 1987  Isolation of telomere DNA from Neurospora crassa.. Mol. Cell. Biol. 7:3168-3177[Abstract/Free Full Text].

SCHMIDHAUSER, T. J., Y.-Z. LIU, H. LIU, and S. ZHOU, 1997  Genome analysis in Neurospora crassa; cloning of four loci arginine-1 (arg-1), methionine-6 (met-6), unknown-7 (un-7), and ribosome production-1 (rip-1) and associated chromosome walking. Fungal Genet. Biol. 21:323-328[Medline].

SELKER, E. U., 1990  Premeiotic instability of repeated sequences in Neurospora crassa.. Annu. Rev. Genet. 24:579-613[Medline].

SMULIAN, A. G., T. SESTERHENN, R. TANAKA, and M. T. CUSHION, 2001  The ste3 pheromone receptor gene of Pneumocystis carinii is surrounded by a cluster of signal transduction genes. Genetics 157:991-1002[Abstract/Free Full Text].

VOLLMER, S. J. and C. YANOFSKY, 1986  Efficient clones of genes of Neurospora crassa.. Proc. Natl. Acad. Sci. USA 83:4869-4873[Abstract/Free Full Text].