Long, Interrupted Conversion Tracts Initiated by cog in Neurospora crassa

Corresponding author: David E. A. Catcheside, School of Biological Sciences, Flinders University, Bedford Park, S.A. 5042, Australia, david.catcheside{at}flinders.edu.au (E-mail).

Communicating editor: R. H. DAVIS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Multiple polymorphisms distinguish Emerson and Lindegren strains of Neurospora crassa within the histidine-3 gene and in its distal flank. Restriction site and sequence length polymorphism in a set of 14 PCR products covering this 6.9-kb region were used to identify the parental origin of DNA sequence information in prototrophic progeny of crosses heterozygous for auxotrophic mutations in his-3 and the silent sequence differences. Forty-one percent of conversion tracts are interrupted. Where the absence of rec-2⁺ permits activity of the recombination hotspot cog, conversion appears to originate at cog and conversion tracts are up to 5.9 kb long. The chromosome bearing cog ^L, the dominant allele that confers a high frequency of recombination, is almost invariably the recipient of information. In progeny from crosses heterozygous rec-2/rec-2⁺, conversion tracts are much shorter, most are not initiated at cog and either chromosome seems equally likely to be converted. Although 32% of his-3 prototrophs have a crossover that may be associated with conversion, it is suggested that the apparent association between conversion and crossing over at this locus may be due to confounding of coincidental events rather than to a mechanistic relationship.

THE recombination hotspot cog, located 2.3–3.2 kb distal of his-3 (YEADON and CATCHESIDE 1995A ), exists as two functional alleles. The dominant allele of Lindegren origin, cog ^L (formerly called cog⁺), promotes recombination at a similar level to that seen at Saccharomyces cerevisiae (yeast) hotspots, significantly higher than the allele from Emerson, cog ^E (ANGEL et al. 1970 ; CATCHESIDE 1979 ; YEADON and CATCHESIDE 1995A ). In the absence of the unlinked regulatory gene rec-2⁺, cog ^L increases his-3 prototroph frequencies by a factor of eight when compared to similar crosses homozygous cog ^E (ANGEL et al. 1970 ), to a maximum of 0.5% (P. J. YEADON, unpublished results). Crossing over between his-3 and the distal flanking gene, adenine-3, is increased fourfold by cog ^L (ANGEL et al. 1970 ) to ~9%, from 2% in crosses homozygous cog ^E (CATCHESIDE and ANGEL 1974 ; CATCHESIDE 1979 ; YEADON and CATCHESIDE 1995A ). In the presence of rec-2⁺, recombination between his-3 alleles is reduced 30-fold in crosses containing cog ^L and fourfold in crosses homozygous cog ^E, in each case to the same low level (ANGEL et al. 1970 ). Crossing over between his-3 and ad-3 is not significantly reduced by rec-2⁺ in crosses homozygous cog ^E, but is reduced to 2% in crosses homozygous cog ^L (ANGEL et al. 1970 ; CATCHESIDE 1979 ). In cog ^L/cog ^E heterozygotes, recombination is preferentially initiated on the chromosome carrying cog ^L, which is also more likely to be the recipient of information (CATCHESIDE 1977 ).

Allelic recombination that results in prototrophs such as those assayed in this study has long been assumed to result from gene conversion rather than crossing over within the gene. Conversion was first recognized by aberrant segregation of alleles in Neurospora tetrads, reflecting nonreciprocal transfer of information (MITCHELL 1955 ). Tetrads have not yet been used to investigate cog or rec genes. However, previous studies have determined that allelic recombination at other loci in Neurospora is primarily nonreciprocal (CASE and GILES 1958 ; STADLER 1959 ; SUYAMA et al. 1959 ). If tetrads are not analyzed, it has been concluded that conversion is responsible for allelic recombination when flanking markers are not recombined (MURRAY 1960 ; STADLER and TOWE 1963 ). More recently, chromatids that have experienced conversion have been identified in yeast by the possession of tracts of DNA sequence information from one parent flanked by both sequence information and more distant genetic markers from the other ( JUDD and PETES 1988 ; SYMINGTON and PETES 1988 ). Chromatids that have experienced a crossover can be detected by their new combination of the distant flanking markers.

The effect of cog ^L on both allelic recombination and crossing over suggests a mechanistic association between conversion at his-3 and crossing over distal of it. However, rec-2⁺ has a differential effect on conversion and crossing over in crosses homozygous cog ^E, consistent with suggestions that conversion and crossing over may proceed by separate mechanisms (POWERS and SMITHIES 1986 ; CARPENTER 1987 ; BOWRING and CATCHESIDE 1996 ). If this is the case, the association could reflect the higher frequency of conversion due to cog ^L, increasing the likelihood of crossing over occurring later, for example by a resultant close local association of chromatids.

Lindegren and Emerson strains of Neurospora crassa are polymorphic both in the intergenic region containing cog distal of his-3 (YEADON and CATCHESIDE 1995A ) and within the his-3 structural gene (P. J. YEADON and D. E. A. CATCHESIDE, unpublished results). Sequence divergence (calculated as described in MIYAMOTO et al. 1988 ) is 3.1% between his-3 and cog, 1.4% within cog and 0.5% in the his-3 coding sequence (P. J. YEADON, unpublished results). Such sequence divergence provides silent markers enabling determination of the parental origin of sequence information over an extended region in recombinant progeny of meioses at a resolution higher than that reported previously.

Although much lower levels of sequence heterology introduced in the MAT region of S. cerevisiae affect recombination (BORTS and HABER 1987 , BORTS and HABER 1989 ; BORTS et al. 1990 ) by reducing crossing over and increasing conversion and ectopic events (BORTS and HABER 1987 ), similar levels of heterology at other loci in yeast (SYMINGTON and PETES 1988 ; MALONE et al. 1994 ) and Schizosaccharomyces pombe (fission yeast; GRIMM et al. 1994 ) had no effect. The higher levels of polymorphism used here in Neurospora appear to have little effect compared to the magnitude of the 30-fold reduction resulting from the presence of rec-2⁺ (ANGEL et al. 1970 ). In crosses heterozygous cog ^L/cog ^E and homozygous rec-2, truncation of the heterology proximal of cog leads to no more than a further doubling of prototroph frequency (P. J. YEADON and D. E. A. CATCHESIDE, unpublished results).

We have used 14 PCR products ranging from 330 to 540 bp in length each having easily detectable restriction site polymorphism (RSP) or sequence length polymorphism (SLP) to investigate the molecular outcome of conversion within his-3 and 3.8 kb distal of this gene. A HpaI RSP allowed extension of the analysis to 6 kb proximal of his-3. The RSPs and SLPs were used to determine the parental origin of each segment in 38 progeny prototrophic for histidine from crosses heteroallelic for cog and his-3. Progeny from diploids (Figure 1) homozygous rec-2 and heterozygous rec-2 /rec-2⁺ were examined to investigate differences in outcome due to the inactivation of the cog function by rec-2⁺, and from diploids in which the his-3 mutation closer to cog was cis (K1201/K26 and K504/K26) or trans (K26/K874) to cog ^L to detect differences resulting from the bias for conversion to be initiated on the cog ^L chromosome. The large number of heterologies scored (16 in 6.9 kb, including the sequence variations responsible for the his-3 mutations) permits measurement of the length of conversion tracts, and detection of discontinuity and location of crossovers between flanking markers arginine-1 (proximal of his-3) and ad-3 (distal), if they occur within the region surveyed.

View larger version (20K): [in this window] [in a new window]   Figure 1. —Crosses from which histidine prototrophs were selected. P and D denote the proximal and distal flanking markers on the chromosome carrying the most proximal his-3 allele, and p and d denote the flanking marker alleles on the chromosome carrying the most distal his-3 allele. The centromere is between his-3 and arg-1. All crosses are heterozygous for his-3 alleles (K1201/K26, K504/K26 or K26/K874), for cog alleles and for flanking markers arg-1 and ad-3. From his-3 to arg-1 is ~25 cM; from his-3 to ad-3 9 cM in the absence of rec-2+, but 2 cM when rec-2+ is present.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Origin of Neurospora parental strains:
K26 (Table 1) was isolated in a strain of Lindegren 25a origin (CATCHESIDE and ANGEL 1974 ); K1201, K504, K874, arg-1 K166 and ad-3 K118 in Emerson a (CATCHESIDE and ANGEL 1974 ). T10988 and T10990 were generated by STEVE FITTER and T10997 by FRED BOWRING in this laboratory. F strains are from the collection of D. G. CATCHESIDE.

Culture methods:
Methods were those described by BOWRING and CATCHESIDE 1996 , except that crosses were supplemented with 200 /ml histidine, 500 µg/ml alanine, 500 µg/ml arginine and 200 µg/ml adenine. Vegetative cultures were supplemented with 500 µg/ml alanine, 500 µg/ml arginine, 200 µg/ml histidine, and 400 µg/ml adenosine as required.

Isolation of recombinant progeny:
Ascospores were treated as described in CATCHESIDE 1981 except that plates contained 0.05% of glucose and fructose in place of sucrose, were supplemented with arginine, adenosine and alanine and incubated for 3 days at 34°. his⁺ colonies were picked to slopes and grown at 25°. Cultures were streaked for single colonies and reisolated before further analysis. Flanking markers were determined by the ability of a prototroph to grow either without adenosine or without arginine.

T11245-T11320 (Table 1) are histidine prototrophs isolated for the purpose of conversion tract mapping. T11245-T11252 were derived from a cross between T10988 and T4393, T11253-T11260 from a cross between T10990 and F7446, and T11261-T11268 from a cross between T10990 and F7448. T11269-T11274, T11302-T11305 and T11320 were derived from three separate crosses between T10987 and T4398. T11275-T11277 and T11306-T11308 were from two independent crosses between T10990 and T10997.

Preparation of PCR template:
Quick template DNA was made from each progeny strain as described in YEADON and CATCHESIDE 1996 . For the parental strains, genomic DNA was prepared as described by YEADON and CATCHESIDE (1994).

PCR amplification:
PCR was performed for 40 cycles (SAIKI et al. 1988 ) using a PTC-100 Thermal Sequencer (MJ Research Inc., supplied by Bresatec) and Taq DNA polymerase (BTQ-1; Bresatec) in a total reaction volume of 50 µl. Fifty nanograms of genomic DNA or 2 µl of quick template DNA was used as template in each reaction. Annealing was at 50° and MgCl₂ 2.5 mM.

PCR primers:
his-3 primers were designed using the program PCRPRIM on ANGIS from the sequence of histidine-3 published by LEGERTON and YANOFSKY 1985 and corrected as necessary where sequence information (P. J. YEADON and D. E. A. CATCHESIDE, unpublished results) conflicted with the published sequence. Primers distal of his-3 (Figure 2) were designed using sequence previously obtained for the intergenic regions (YEADON and CATCHESIDE 1995A ). The P1 pair of primers (Figure 2) was designed after a sequence walk proximal of his-3.

View larger version (21K): [in this window] [in a new window]   Figure 2. —Scale map of the his-3 to cog region showing the segments amplified by PCR and restriction site polymorphisms and the Guest element used to identify the parental origin of each segment. RSPs, his-3 mutant sites and insertions that result in SLPs shown above the line are in Emerson, those below the line are in Lindegren. Measurements are in bp from the start codon of his-3. The designations of PCR segments are arbitrary, reflecting the names given to primer pairs as they were designed.

Restriction digests and electrophoresis:
PCR products were digested with 3 units of the appropriate restriction enzyme (New England Biolabs) for 90 min as described by the manufacturer and the products resolved by electrophoresis on 3% NuSieve 3:1 agarose (FMC Bioproducts) in TAE, 3 V/cm for 3 hr.

Detection of RSP and SLP in the Lindegren and Emerson parents:
Genomic DNA from T10987 (Lindegren descent) and F7448 (Emerson descent) was used as template to amplify DNA segments (Figure 2) predicted from the sequence to have RSPs or SLPs that differentiate each parent. PCR products were digested where necessary with appropriate restriction enzymes and fragments resolved by electrophoresis (data not shown). Each pair of PCR products yielded the expected distinguishable patterns, length polymorphisms in GAP(X) and C8, and a restriction site present in only one parent in the remainder. The polymorphism in C8 reflects the presence of the inverted repeat transposable element Guest in Emerson (YEADON and CATCHESIDE 1995B ) yielding a product 102 bp longer than that from Lindegren.

The HpaI RSP 6 kb proximal of his-3, discovered during mapping of genomic DNA (YEADON and CATCHESIDE 1995A , Figure 2), assisted location of crossovers proximal of his-3 and was detected by Southern analysis as described in YEADON and CATCHESIDE 1995A . The probe used was JY25 (YEADON and CATCHESIDE 1995A ).

Determination of the parental origin of sequence segments from recombinant progeny:
Fourteen of the 15 segments (Figure 2) were amplified by PCR from the 38 histidine prototrophic progeny, and the parental origin of each identified from their restriction pattern or, in the case of C8, from the length polymorphism due to Guest. The segment labeled GAP(X) (Figure 2) was not used since reliable sequence was difficult to obtain, impeding identification of RSPs and, since a single site could not be surveyed, the parental origin of segments different in length from that in either parent would be uncertain.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

his⁺ recombinant progeny:
Histidine prototrophs were selected from crosses homozygous rec-2 or heterozygous rec-2/rec-2⁺ (Figure 1) having the his-3 allele closer to cog either on the chromosome bearing cog ^L (K1201/K26 and K504/K26) or on the chromosome bearing cog ^E (K26/K874). Since all crosses were heterozygous for different his-3 mutant alleles, his-3⁺ progeny are either the result of a crossover between the sites responsible for the his-3 mutations or conversion of one marker, with or without a crossover elsewhere, or more rarely result from a reversion event.

Potential revertants:
In two of the his-3⁺ progeny (T11251 and T11273) the only alteration detected was that of the marker responsible for the his-3 mutation (K26 and K504, respectively; data not shown). Thus these prototrophs could have resulted from reversion and have not been considered further.

Potential sibling pairs:
Five pairs of prototrophs isolated from crosses heterozygous rec-2 /rec-2⁺ were each identical for all markers assayed including mating type and, where relevant, am allele. Because of the small number of prototrophic progeny available in the crosses in which they arose, they are in each case >95% likely to be from the same pair of sibling spores. The pairs are T11270 and T11271, T11276 and T11277, T11302 and T11303, T11304 and T11305, T11306 and T11308. Siblings are identified in Table 1 by (v), (w), (x), (y) and (z) and in Figure 3 by an asterisk after the strain number of one of a pair. Henceforth in this study, each pair is considered as a single strain, the asterisk after the number identifying it as a sibling pair.

View larger version (86K): [in this window] [in a new window]   Figure 3. —The parental origin of DNA from histidine prototrophic progeny of crosses detailed in Figure 1. E indicates DNA of Emerson origin, L of Lindegren origin. PCR-amplified segments (P1–C9), the proximal HpaI RSP (H) and the positions of the mutant sites are shown above the appropriate column. P is the proximal (arg-1 or arg+), D the distal (ad-3 or ad+) flanking marker. Distances from the start codon of his-3 in bp are shown below each marker. X indicates where a crossover has occurred. Where two locations are equally likely, each position is marked X?. To assist interpretation, the last two columns (X and Int) indicate, respectively, the number of crossovers (-, 1 or 2) between flanking markers and whether the conversion tract is interrupted (+) or not (-). Strains marked with an asterisk indicate pairs that are x95% likely to be from sibling spores.

Bias in the chromosome receiving information:
Among the 23 progeny of crosses homozygous rec-2, four strains (T11247, T11265, T11266 and T11267) have exchanges with no detectable conversion (henceforth termed "simple crossovers") between the his-3 mutations. In two others, T11262 and T11264, there are alternate positions in which the crossover could have occurred, indicated by the symbol "X?" in Figure 3. In 16 of these 23 progeny the Lindegren chromosome must be the recipient of information, and in T11261 alone is Emerson most probably the recipient of information. It is clear that the Lindegren chromosome, which carries the high frequency cog^L allele, is converted more often than the Emerson chromosome. In contrast, in the 11 progeny from crosses in which rec-2⁺ is present, the two chromosomes seem equally likely to be converted: one has a simple crossover (T11269), there are four in which Emerson is converted (T11270*, T11274, T11307 and T11320), three where Lindegren is converted (T11275, T11276*, and T11306*) and three (T11272, T11302* and T11304*) where the recipient of information cannot be determined.

Positions of crossovers:
Nine of the 23 progeny from rec-2 /rec-2 crosses (39%) have at least one crossover between arg-1 and ad-3 (Figure 3). Three of these are between arg-1 and the HpaI RSP (H) 6 kb proximal of his-3 (in T11264, T11266 and T11268), one is between H and his-3 (in T11252), seven are within his-3 (in T11245, T11247, T11262, T11264, T11265, T11266 and T11268) and one (T11261) is between cog and ad-3. Seven of the 11 progeny from crosses in which rec-2⁺ was present (64%) have crossovers between the flanking markers (Figure 3). Two of these are between arg-1 and H (T11276* and T11307), four are within the his-3 gene (T11269, T11272, T11302* and T11304*) and one (T11275) between the C3 and C9 heterologies distal of cog.

Length of conversion tracts:
Conversion tracts in prototrophs from crosses homozygous rec-2 vary in length (Figure 3). All prototrophs, except those with simple crossovers and the two potential revertants, show conversion of more than one marker. The longest continuous tracts are those in T11254 and T11259 that cover the region between the R1 and C9 heterologies, 5.6 kb apart. RSP typing distal of C9 showed that the tract in T11254 terminates <185 bp distal of the heterology in C9, and that in T11259 ends >300 bp and <1178 bp distal of the C9 heterology (data not shown). Thus the longest tract extends >5.9 kb. However, without tetrad analysis the possibility that these long tracts represent double crossovers cannot be discounted, although interference makes this improbable since the his-3–ad-3 interval is at maximum 9 cM (CATCHESIDE 1979 ) and is probably <45 kb (J. P. RASMUSSEN, personal communication).

In contrast, if T11264, T11266 and T11268 are considered to contain conversion tracts, these could terminate anywhere between the HpaI RSP (H) and arg-1. The arg-1–his-3 interval, which includes the centromere, is up to 25 cM (P. J. YEADON, unpublished results) and has at least 830 kb (ROSA et al. 1997 ) and probably >1.2 Mb (A. L. ROSA, personal communication) of DNA sequence. Thus conversion tracts in T11264, T11266 and T11268 would be >7.5 kb and conceivably up to 1.2 Mb long. Since interference in Neurospora has little effect over distances of >1 Mb (FOSS et al. 1993 ), it is not unlikely that two crossovers could occur between his-3 and arg-1. Although identification of RSPs proximal of his-3 may alter the conclusion, from current genotype information T11264, T11266 and T11268 seem more likely to result from double crossovers than from long conversion tracts.

The presence of rec-2⁺ results in shorter conversion tracts that do not extend distal of his-3. The longest continuous conversion tracts in these progeny are in T11270* and T11274. The tracts cover R1, K504 and P1 and are thus at least 940 bp long.

Discontinuity in conversion tracts:
Among progeny from crosses homozygous rec-2, there are 17 that show evidence of conversion (Figure 3). Of these, conversion tracts are discontinuous in eight (47%). Ten of the 11 progeny from crosses including rec-2⁺ (Figure 3) show evidence of conversion and in three of these the tracts are discontinuous (30%). The difference between discontinuity of tracts in crosses in which rec-2⁺ was present or absent is not significant (² = 0.25 with Yates' correction; P > 0.5). In total, 11 of 27 (41%) conversion tracts are discontinuous.

Gradient of conversion frequency:
For the progeny of crosses homozygous rec-2, the distribution of conversion frequencies is bimodal (Figure 4A). There is a peak of conversion within the his-3 gene, as is expected since the progeny were selected as histidine prototrophs. The second peak of conversion close to the C3 heterology within cog is not expected as it was not a direct consequence of selection. The probability that the three highest values for conversion outside his-3 fall within the region that includes the difference between cog ^L and cog ^E is 0.008 (3/10 x 2/9 x 3/8), making the conjunction of proximity to cog and the peak of unselected conversion significant. Moreover, in the progeny of crosses heterozygous rec-2 /rec-2⁺, where cog is inactive, the peak at cog is absent (Figure 4B).

View larger version (16K): [in this window] [in a new window]   Figure 4. —Conversion frequency for unselected markers in the his-3 to cog region. Data from crosses homozygous rec-2 (A) are shown above the map and from crosses heterozygous rec-2 /rec-2+ (B) below the map. The vertical axis indicates the number of times a site is scored as converted. Where the uncertainty of the position of crossovers affects the decision of whether a site is converted or not, it is scored as 0.5. Distances are to scale and measurements are in bp from the start codon of the his-3 gene.

Association of crossovers with conversion:
At least one crossover between the flanking markers arg-1 and ad-3 occurred in 47% of the prototrophs (16 of 34; Figure 3). Of the prototrophs from crosses homozygous rec-2, three of the crossovers (in T11264, T11266 and T11268) are between H and arg-1 and are likely to be too distant to be associated with conversion. Those crossovers in T11252 and T11261 are also sufficiently distant that their association with conversion is doubtful. This leaves seven crossovers that may be associated with conversion initiated by cog ^L, although four of these are simple crossovers, with no evidence of conversion. Thus, ignoring simple crossovers, 16% of prototrophs (three of 19) from crosses homozygous rec-2 have crossovers apparently associated with conversion. If simple crossovers are included as potentially associated with conversion, then 30% of these prototrophs (seven of 23) have an associated crossover.

Five of the seven crossovers in progeny from crosses heterozygous rec-2/rec-2⁺ are within the region surveyed; one is a simple crossover (T11269), one (T11275) is sufficiently distant that an association with conversion is doubtful, but three (in T11272, T11302* and T11304*) are at ends of conversion tracts and thus may be associated with conversion. The two remaining crossovers (in T11276* and T11307) are >6 kb from the proximal end of his-3, and are unlikely to be associated with conversion at this locus. Of the 10 prototrophs from crosses heterozygous rec-2/rec-2⁺ (cog inactive) that have evidence of conversion, three (30%) have crossovers that may be associated with conversion. If simple crossovers are included, 36% of these prototrophs (four of 11) have crossovers that may be associated with conversion.

In total, 11 of 34 prototrophs (32%) have crossovers that may be associated with conversion. The presence or absence of rec-2⁺ has no significant effect on the association between conversion and crossing over in prototrophic progeny (² = 0.31 with Yates' correction; P > 0.5).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Although it is not known whether conversion is initiated by a double-strand break (DSB) in Neurospora, the evidence is very good that this is the case in yeast. DSBs have been detected at several yeast hotspots (SUN et al. 1989 ; CAO et al. 1990 ; FAN et al. 1995 ). Conversion intermediates with paired Holliday junctions have been isolated from yeast (SCHWACHA and KLECKNER 1994 , SCHWACHA and KLECKNER 1995 ). Such structures are consistent with the double-strand-break repair (DSBR) model of recombination (SZOSTAK et al. 1983 ) and more recent refinements by SUN et al. 1991 in which the DSB results in little or no gap but is flanked by 3'-overhanging single-stranded tails. In Neurospora, the regulation of recombination by naturally polymorphic rec genes (reviewed in CATCHESIDE 1986 ), not yet detected in S. cerevisiae or S. pombe, provides an additional analytical tool. Regulation of recombination in the his-3 region of Neurospora by the unlinked gene rec-2, the dominant allele of which (rec-2⁺) prevents the initiation of recombination at cog (CATCHESIDE and ANGEL 1974 ), allows separation of events at his-3 into those that are cog-related and those that are not. Apart from the additional aspect of local regulation of recombination in Neurospora, there are many similarities between cog and yeast hotspots that suggest a common mode of activity.

Convertants at his-3 manifest a bias in the direction of information transfer. Of the 23 prototrophs from crosses lacking rec-2⁺, in only one is Emerson almost certainly the recipient of information (Figure 3), confirming the earlier suggestion that the chromosome carrying cog^L is preferentially converted (CATCHESIDE 1977 ). Likewise in both budding and fission yeasts, where a hotspot has alleles that differ in activity such as the ade6 M26 hotspot of S. pombe (GUTZ 1971 ) or the promoter deletion that removes hotspot activity at ARG4 of S. cerevisiae (NICOLAS et al. 1989 ), the chromosome on which recombination is initiated is the recipient of information. Indeed, in the absence of crossing over, the DSBR model for yeast recombination predicts information transfer only to the broken chromosome (SZOSTAK et al. 1983 ; ORR-WEAVER and SZOSTAK 1985 ). However, in the current model (SUN et al. 1991 ), since conversion is due to mismatch repair (MMR) of heteroduplex rather than gap-filling, failure of MMR results in post-meiotic segregation (PMS) and the potential of information transfer to the unbroken chromatid.

Conversion tracts at different hotspots vary in length, dependent both on the locus and the mode of selection of progeny. The length of conversion tracts at his-3 in Neurospora is within the range of those measured in yeast at several loci, including the HIS4-LEU2 interval (WHITE and PETES 1994 ) where conversion tracts are often >5 kb long. However, lengths of 1.5 kb are more common at most loci ( JUDD and PETES 1988 ; BORTS and HABER 1989 ). Minimum lengths of conversion tracts at ade6 in S. pombe range from 670 to 1290 bp, depending on the hotspot allele, with M26 generating shorter tracts (GRIMM et al. 1994 ). Conversion tracts at his-3 can be over 5.9 kb long (Figure 3), but the distance between the recombinator and the gene may select for a subset enriched for the longest tracts in prototrophic progeny. Conversion tracts selected to give prototrophs at the am locus of Neurospora appear generally shorter, as 30% are <741 bp long (F. J. BOWRING and D. E. A. CATCHESIDE, unpublished results). Since the recombinator at am is close to the 5' end of the gene and extension of heteroduplex beyond the more 3' mutant site is unlikely to yield prototrophs, this system may select shorter tracts.

The degree of discontinuity of conversion tracts appears to vary between loci as well as between species. Forty-one percent of conversion tracts in this study are discontinuous (Figure 3). At ade6 in S. pombe, only seven of 210 conversion tracts were discontinuous, with the reservation that only seven points in 3.7 kb were monitored (GRIMM et al. 1994 ). In S. cerevisiae, conversion tracts have been scored as discontinuous from <10% (JUDD and PETES 1988 ; SYMINGTON and PETES 1988 ; BORTS and HABER 1989 ) to 25% (BORTS and HABER 1987 ) of the time. The variation in apparent degree of discontinuity may be an artefact of the measurement: the fewer heterologies that can be scored, the more likely a tract will be seen to be continuous. It is also possible that a difference in MMR between yeast and Neurospora (discussed below) could account for the relatively high frequency of discontinuous tracts at his-3.

Conversion gradients have been detected at most recombination hotspots in both budding and fission yeasts (NICOLAS et al. 1989 ; SCHULTES and SZOSTAK 1990 ; SYMINGTON et al. 1991 ; DETLOFF et al. 1992 ; GRIMM et al. 1994 ; MALONE et al. 1994 ). Similarly, the peak of unselected conversion within the cog region in progeny of rec-2 diploids (Figure 4) suggests that when cog is active in the absence of rec-2⁺, most and possibly all conversion tracts in this region are initiated at cog. Conversion in the much rarer prototrophs from crosses heterozygous rec-2 /rec-2⁺ peaks in the 5' end of his-3 (Figure 4) and does not extend distal of the gene (Figure 3), suggesting that when cog is inactivated residual conversion is initiated at another site proximal of his-3 or within its 5' end. This rec-2⁺-insensitive recombinator may be functionally monomorphic as both chromosomes are equally likely to be converted in prototrophs from crosses that include rec-2⁺ (Figure 3). These results are consistent with deductions from data on recombination at his-3 in the translocation mutant TM429, generated in a cog⁺ (now designated cog ^L) strain (CATCHESIDE and ANGEL 1974 ). Since conversion could only cross the translocation breakpoint if cog ^L was in the intact chromosome, CATCHESIDE and ANGEL 1974 concluded that, in the absence of rec-2⁺, most conversion is initiated at cog ^L. Conversion was also found to be initiated from the proximal end of his-3, although at a much reduced frequency and insensitive to rec-2⁺ (CATCHESIDE and ANGEL 1974 ).

In both the double-strand-break repair (SUN et al. 1991 ) and the single-strand gap repair (RADDING 1982 ) models, the establishment of recombination intermediates results in heteroduplex DNA. Since PMS is much rarer than conversion in yeast (reviewed in PETES et al. 1991 ), most mismatches must be repaired by the MMR system (HOLLIDAY 1964 ). Little is known about MMR in Neurospora, although homologues of Escherichia coli MMR genes mutL and mutS have recently been cloned (H. INOUE, personal communication). The Neurospora mus-18 gene, responsible for repair of UV-induced DNA damage, has no sequence similarity to other repair enzymes and appears to operate by a novel mechanism (YAJIMA et al. 1995 ), which may imply that MMR in Neurospora differs in mechanism from that in yeast. It is therefore possible that PMS is more common in Neurospora than in yeast, as appears to be the case in the filamentous fungus Ascobolus in which aberrant 4:4 segregation, rare or absent in yeast, is detected (NICOLAS and ROSSIGNOL 1989 ). A lower efficiency of MMR in Neurospora than in yeast could account for the relatively high frequency of discontinuous conversion tracts seen at his-3 (Figure 3).

Since the pattern of recombination products associated with cog^L is within the spectrum generated by yeast hotspots, it is likely that the recombination mechanisms are similar. Thus it is probable that conversion at his-3 in Neurospora is initiated, as in yeast, by a DSB. However, provided there is no bias in strand scission, the DSBR model for yeast recombination (ORR-WEAVER and SZOSTAK 1985 ) predicts that conversion events have a 50% chance of an associated crossover. Flanking marker recombination in convertants varies ~50% at many loci of Neurospora (FINCHAM 1967 ; FOSS et al. 1993 ) and yeast (FOGEL et al. 1979 ; GILBERTSON and STAHL 1996 ). However, when a correction for incidental exchanges is made (STADLER 1973 ), crossovers are associated with about one-third of conversion events (FOGEL et al. 1979 ; FOSS et al. 1993 ). In the his-3–cog system, 30% of prototrophs from crosses in which cog is active have crossovers that may be associated with conversion (Figure 3), consistent with figures obtained by correction for incidental exchanges (FOSS et al. 1993 ). Since the deviation from the predicted 50% association between conversion and crossing over is similar at most loci studied, the data presented here may reflect an insufficiency in the model rather than a difference in mechanism between Neurospora and yeast.

If the recombination intermediate were frequently resolved in a way that could not result in crossovers, as suggested for mating type switching in yeast (NASMYTH 1982 ), possibly by a topoisomerase (THALER et al. 1987 ; MCGILL et al. 1989 ; GILBERTSON and STAHL 1996 ) or by scission of one junction and migration of the second to the resulting nicks (GILBERTSON and STAHL 1996 ), this would retain the bias in information transfer but resolve the low association with crossing over. Crossovers could result from resolution of the conversion intermediate only by cutting at both Holliday junctions, giving a theoretically equal chance of conversion with crossing over or conversion without crossing over. Thus the degree of association between conversion and crossing over would depend on the relative frequencies of the two modes of resolution.

It is possible however that conversion and crossing over proceed by different mechanisms as suggested previously (POWERS and SMITHIES 1986 ; CARPENTER 1987 ; BOWRING and CATCHESIDE 1996 ). Molecular investigation of tetrads showing conversion at the ARG4 locus of yeast did not support the crossover pathway of the DSBR model (GILBERTSON and STAHL 1996 ). A study employing molecular markers closely flanking the am locus, which is in a region where exchange events are rare, concluded that conversion at this locus appears unassociated with crossing over (BOWRING and CATCHESIDE 1996 ), despite the fact that many loci of Neurospora (including am) show a significant positive correlation between conversion and crossing over (summarized in FOSS et al. 1993 ). Since it seems unlikely that the mechanism of conversion will vary between loci of the same organism, the apparent association between conversion and crossing over at his-3 may be due to the relative frequency of the two types of event in this region rather than a mechanistic relationship. Although the size of the data set prevents definite conclusions of this nature, coupling these results with those obtained at am (BOWRING and CATCHESIDE 1996 ) suggests a potential flaw in the conclusion of a mechanistic relationship between conversion and crossing over at hotspots of yeast, where recombination occurs as frequently as at cog. Alternatively, regulation of the frequency of alternate modes of resolution of a conversion intermediate may differ between loci.

In conclusion, although our data are consistent with a DSBR pathway for conversion in Neurospora, the double-strand-break repair model (ORR-WEAVER and SZOSTAK 1985 ) insufficiently explains the low number of crossovers associated with conversion initiated at cog ^L. Clearly, the same problems of interpretation encountered by workers with yeast (FOGEL et al. 1979 ; GILBERTSON and STAHL 1996 ) exist in Neurospora, despite the differences between this organism and yeast. This work has identified the tools needed for more detailed analyses, including tetrad data, required to differentiate between recombination models currently under debate.

	ACKNOWLEDGMENTS

We are indebted to FRANK STAHL for his comments on versions of this manuscript. Our thanks to DUNCAN MACKAY for help with a statistical problem and to FRED BOWRING for critical discussion. This work was supported by a grant from the Flinders University Research Budget.

Manuscript received February 19, 1997; Accepted for publication September 4, 1997.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

FNGEL, T., B. AUSTIN, and D. G. CATCHESIDE, 1970  Regulation of recombination at the his-3 locus in Neurospora crassa. Aust. J. Biol. Sci. 23:1229-1240[Medline].

BORTS, R. H. and J. E. HABER, 1987  Meiotic recombination in yeast: alteration by multiple heterozygosities. Science 237:1459-1463[Abstract/Free Full Text].

BORTS, R. H. and J. E. HABER, 1989  Length and distribution of meiotic gene conversion tracts and crossovers in Saccharomyces cerevisiae. Genetics 123:69-80[Abstract].

BORTS, R. H., W.-Y. LEUNG, W. KRAMER, B. KRAMER, and M. WILLIAMSON et al., 1990  Mismatch repair-induced meiotic recombination requires the PMS1 gene product. Genetics 124:573-584[Abstract].

BOWRING, F. J. and D. E. A. CATCHESIDE, 1996  Gene conversion alone accounts for more than 90% of recombination events at the am locus of Neurospora crassa. Genetics 143:129-136[Abstract].

CAO, L., E. ALANI, and N. KLECKNER, 1990  A pathway for generation and processing of double-strand breaks during meiotic recombination in S. cerevisiae. Cell 61:1089-1101[Medline].

CARPENTER, A. T. C., 1987  Gene conversion, recombination nodules and the initiation of meiotic synapsis. BioEssays 6:232-236[Medline].

CASE, M. E. and N. H. GILES, 1958  Recombination mechanisms at the pan-2 locus in Neurospora crassa. Cold Spring Harbor Symp. Quant. Biol. 23:119-135[Medline].

CATCHESIDE, D. G., 1977 The Genetics of Recombination. Edward Arnold, London.

CATCHESIDE, D. E. A., 1979  Effect of rec-2⁺ on the formation of double mutant recombinants in Neurospora crassa. Aust. J. Biol. Sci. 32:261-265.

CATCHESIDE, D. E. A., 1981  Genes in Neurospora that suppress recombination when they are heterozygous. Genetics 98:55-76[Abstract].

CATCHESIDE, D. E. A., 1986  A restriction and modification model for the initiation and control of recombination in Neurospora. Genet. Res. Camb. 47:157-165[Medline].

CATCHESIDE, D. G. and T. ANGEL, 1974  A histidine-3 mutant, in Neurospora crassa, due to an interchange. Aust. J. Biol. Sci. 27:219-229[Medline].

DETLOFF, P., M. A. WHITE, and T. D. PETES, 1992  Analysis of a gene conversion gradient at the HIS4 locus in Saccharomyces cerevisiae. Genetics 132:113-123[Abstract].

FAN, Q., F. XU, and T. D. PETES, 1995  Meiosis-specific double-strand breaks at the HIS4 recombination hotspot in the yeast Saccharomyces cerevisiae: control in cis and trans. Mol. Cell. Biol. 15:1679-1688[Abstract].

FINCHAM, J. R. S., 1967  Recombination within the am gene of Neurospora crassa. Genet. Res. Camb. 9:49-62.

FOGEL, S., R. K. MORTIMER, K. LUSNAK, and F. TAVARES, 1979  Meiotic gene conversion: a signal of the basic recombination event in yeast. Cold Spring Harbor Symp. Quant. Biol. 43:1325-1341.

FOSS, E., R. LANDE, F. W. STAHL, and C. M. STEINBERG, 1993  Chiasma interference as a function of genetic distance. Genetics 133:681-691[Abstract].

GILBERTSON, L. A. and F. W. STAHL, 1996  A test of the double-strand break repair model for meiotic recombination in Saccharomyces cerevisiae. Genetics 144:27-41[Abstract].

GRIMM, C., J. BAHLER, and J. KOHLI, 1994  M26 recombinational hotspot and physical conversion tract analysis in the ade6 gene of Schizosaccharomyces pombe. Genetics 135:41-51.

GUTZ, H., 1971  Site specific induction of gene conversion in Schizosaccharomyces pombe. Genetics 69:317-337[Free Full Text].

HOLLIDAY, R., 1964  A mechanism for gene conversion in fungi. Genet. Res. Camb. 5:282-304.

LEGERTON, T. L. and C. YANOFSKY, 1985  Cloning and characterisation of the multifunctional his-3 gene of Neurospora crassa. Gene 39:129-140[Medline].

JUDD, S. R. and T. D. PETES, 1988  Physical lengths of meiotic and mitotic conversion tracts in Saccharomyces cerevisiae.. Genetics 118:401-410[Abstract].

MALONE, R. E., S. KIM, S. A. BULLARD, S. LUNQUIST, and L. HUTCHINGS-CROW et al., 1994  Analysis of a recombination hotspot for gene conversion occurring at the HIS2 gene of Saccharomyces cerevisiae. Genetics 137:5-18[Abstract].

MCGILL, C., B. SHAFER, and J. STRATHERN, 1989  Coconversion of flanking sequences with homothallic switching. Cell 57:459-467[Medline].

MITCHELL, M. B., 1955  Aberrant recombination of pyridoxine mutants of Neurospora. Proc. Natl. Acad. Sci. USA 41:216-220.

MIYAMOTO, M. M., B. F. KOOP, J. L. SLIGHTOM, M GOODMAN, and M. R. TENMANT, 1988  Molecular systematics of higher primates: genealogical relations and classification. Proc. Natl. Acad. Sci. USA 85:7627-7631[Abstract/Free Full Text].

MURRAY, N. E., 1960  Complementation and recombination between methionone-2 alleles in Neurospora crassa. Heredity 15:207-217.

NASMYTH, K. A., 1982  Molecular genetics of yeast mating type. Annu. Rev. Genet. 16:439-500[Medline].

NICOLAS, A. and J.-L. ROSSIGNOL, 1989  Intermediates in homologous recombination revealed by marker effects in Ascobolus. Genome 31:528-535.

NICOLAS, A., D. TRECO, N. P. SCHULTES, and J. W. SZOSTAK, 1989  Iden- tification of an initiation site for meiotic gene conversion in the yeast Saccharomyces cerevisiae.. Nature 338:35-39[Medline].

ORR-WEAVER, T. L. and J. W. SZOSTAK, 1985  Fungal recombination. Microbiol. Rev. 49:33-58[Free Full Text].

PETES, T. D., R. E. MALONE and L. S. SYMINGTON, 1991 Recombination in yeast, pp. 407–521 in The Molecular and Cellular Biology of the Yeast Saccharomyces (Genome Dynamics, Protein Synthesis and Energetics). Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

POWERS, A. and O. SMITHIES, 1986  Short gene conversions in the human fetal globin region: a by-product of chromosome pairing during meiosis? Genetics 112:343-358[Abstract].

RADDING, C. M., 1982  Homologous pairing and strand exchange in genetic recombination. Annu. Rev. Genet. 16:405-437[Medline].

ROSA, A. L., S. D. HAEDO, 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].

SAIKI, R. K., D. H. GELFLAND, S. STOFFEL, S. J. SCHARF, and R. HIGUCHI et al., 1988  Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487-491[Abstract/Free Full Text].

SCHULTES, N. P. and J. W. SZOSTAK, 1990  Decreasing gradients of gene conversion on both sides of the initiation site for meiotic recombination at the ARG4 locus in yeast. Genetics 126:813-822[Abstract].

SCHWACHA, A. and N. KLECKNER, 1994  Identification of joint molecules that form frequently between homologs but rarely between sister chromatids during yeast meiosis. Cell 76:51-63[Medline].

SCHWACHA, A. and N. KLECKNER, 1995  Identification of double Holliday junctions as intermediates in meiotic recombination. Cell 83:1-20[Medline].

STADLER, D. R., 1959  Gene conversion of cysteine mutants in Neurospora. Genetics 44:647-655[Free Full Text].

STADLER, D. R., 1973  The mechanism of intragenic recombination. Annu. Rev. Genet. 7:113-127[Medline].

STADLER, D. R. and A. M. TOWE, 1963  Recombination of allelic cysteine mutants in Neurospora. Genetics 48:1323-1344[Free Full Text].

SUN, H., D. TRECO, and J. W. SZOSTAK, 1989  Double strand breaks at an initiation site for meiotic gene conversion. Nature 338:87-90[Medline].

SUN, H., D. TRECO, and J. W. SZOSTAK, 1991  Extensive 3'-overhanging, single-stranded DNA associated with the meiosis-specific double-strand breaks at the ARG4 recombination initiation site. Cell 64:1155-1161[Medline].

SUYAMA, J., K. D. MUNKRES, and V. W. WOODWARD, 1959  Genetic analysis of the pyr-3 locus of Neurospora crassa: the bearing of recombination and gene conversion upon intraallelic linearity. Genetica 30:293-311[Medline].

SYMINGTON, L. S. and T. D. PETES, 1988  Expansions and contractions of the genetic map relative to the physical map of yeast chromosome III. Mol. Cell. Biol. 8:595-604[Abstract/Free Full Text].

SYMINGTON, L. S., A. BROWN, S. G. OLIVER, P. GREENWELL, and T. D. PETES, 1991  Genetic analysis of a meiotic recombination hot-spot on chromosome III of Saccharomyces cerevisiae. Genetics 128:717-727[Abstract].

SZOSTAK, J. W., T. L. ORR-WEAVER, R. J. ROTHSTEIN, and F. W. STAHL, 1983  The double-strand-break repair model for recombination. Cell 33:25-35[Medline].

THALER, D. S., M. M. STAHL, and F. W. STAHL, 1987  Tests of the double-strand-break repair model for Red-mediated recombination of phage and plasmid dv. Genetics 110:501-511.

WHITE, M. A. and T. D. PETES, 1994  Analysis of meiotic recombination events near a recombination hotspot in the yeast Saccharomyces cerevisiae.. Curr. Genet. 26:21-30[Medline].

YAJIMA, H., M. TAKAO, S. TASUHIRA, J. H. ZHAO, and C. ISHII et al., 1995  A eukaryotic gene encoding an endonuclease that specifically repairs DNA damaged by ultraviolet light. EMBO J. 14:2393-2399[Medline].

YEADON, P. J. and D. E. A. CATCHESIDE, 1995a  The chromosomal region which includes the recombinator cog in Neurospora crassa is highly polymorphic. Curr. Genet. 28:155-163[Medline].

YEADON, P. J. and D. E. A. CATCHESIDE, 1995b  Guest: a 98 bp inverted repeat transposable element in Neurospora crassa. Mol. Gen. Genet. 247:105-109[Medline].

YEADON, P. J. and D. E. A. CATCHESIDE, 1996  Quick method for producing template for PCR from Neurospora cultures. Fungal Genet. Newsl. 43:71.

This article has been cited by other articles:

				M. D. Yandeau-Nelson, Q. Zhou, H. Yao, X. Xu, B. J. Nikolau, and P. S. Schnable MuDR Transposase Increases the Frequency of Meiotic Crossovers in the Vicinity of a Mu Insertion in the Maize a1 Gene Genetics, February 1, 2005; 169(2): 917 - 929. [Abstract] [Full Text] [PDF]

				P. J. Yeadon, F. J. Bowring, and D. E. A. Catcheside Alleles of the Hotspot cog Are Codominant in Effect on Recombination in the his-3 Region of Neurospora Genetics, July 1, 2004; 167(3): 1143 - 1153. [Abstract] [Full Text] [PDF]

				P. J. Yeadon, L. Y. Koh, F. J. Bowring, J. P. Rasmussen, and D. E. A. Catcheside Recombination at his-3 in Neurospora Declines Exponentially With Distance from the Initiator, cog Genetics, October 1, 2002; 162(2): 747 - 753. [Abstract] [Full Text] [PDF]

				H. Yao, Q. Zhou, J. Li, H. Smith, M. Yandeau, B. J. Nikolau, and P. S. Schnable From the Cover: Molecular characterization of meiotic recombination across the 140-kb multigenic a1-sh2 interval of maize PNAS, April 30, 2002; 99(9): 6157 - 6162. [Abstract] [Full Text] [PDF]

				P. J. Yeadon, J. P. Rasmussen, and D. E. A. Catcheside Recombination Events in Neurospora crassa May Cross a Translocation Breakpoint by a Template-Switching Mechanism Genetics, October 1, 2001; 159(2): 571 - 579. [Abstract] [Full Text] [PDF]

				H. Yao, Q. Zhou, J. Li, H. Smith, M. Yandeau, B. J. Nikolau, and P. S. Schnable From the Cover: Molecular characterization of meiotic recombination across the 140-kb multigenic a1-sh2 interval of maize PNAS, April 30, 2002; 99(9): 6157 - 6162. [Abstract] [Full Text] [PDF]