Crosses involving the Adiopodoumé strain of Neurospora crassa are defective for repeat-induced point mutation (RIP), a genome defense mechanism of fungi. We show here that the Adiopodoumé strain possesses an incompletely penetrant and variably expressive dominant suppressor of RIP (Srp) that maps to an 34-kbp genome segment that is 26 kbp proximal to mat on linkage group IL. Gene disruption experiments revealed that Srp is the upr-1 allele of Adiopodoumé (upr-1^Ad) that is contained within this segment. The upr-1 gene codes for the catalytic subunit of the translesion DNA polymerase- (Pol-) and it is unusually polymorphic in Neurospora. That the upr-1 gene contains upstream ORFs that overlap with the main ORF is potentially relevant to the incomplete penetrance and variable expressivity of the suppressor. Crosses between heterokaryons that contain upr-1^Ad and strains that prevent mating events involving nuclei that contain upr-1^Ad yielded no progeny in which RIP had occurred, consistent with the idea that the suppressor encoded by upr-1^Ad is diffusible. The potential involvement of the Pol- subunit in two functions, translesion DNA synthesis and RIP regulation, might account for the rapid evolution of its gene in Neurospora.

---

AN unusual strain of Neurospora crassa was isolated in the 1950s from Adiopodoumé (a village and research station near Abidjan, Cote D'Ivoire, West Africa), in that it was the only wild-isolated Neurospora strain found harboring active copies of a LINE-like transposable element named Tad (KINSEY and HELBER 1989). All the other wild isolates of Neurospora examined (including those of species other than N. crassa) contained only mutation-inactivated relics of Tad (KINSEY 1989; KINSEY et al. 1994). No other active transposable element has since been found in Neurospora. Subsequently, the Adiopodoumé strain was identified as one of 7 wild-isolated N. crassa strains (of 450 screened) possessing the ability to dominantly suppress repeat-induced point mutation (RIP) (NOUBISSI et al. 2000, 2001; BHAT et al. 2003). RIP is a genome defense process of fungi that occurs during the premeiosis of a sexual cross and hypermutates and methylates duplicated DNA sequences, thereby protecting the genome against the proliferation of transposable elements and other parasitic DNA (CAMBARERI et al. 1989; SELKER 1990; GALAGAN et al. 2003; GALAGAN and SELKER 2004). Presumably the dominant RIP suppressor phenotype facilitated Tad's survival in the Adiopodoumé strain. Screening for the dominant RIP suppressors was done by crossing the wild isolates with strains of the opposite mating type bearing the transgene Dp(erg-3), which duplicates a segment of the erg-3 gene. Crosses of the Dp(erg-3) strains with most wild isolates yielded RIP-induced erg-3 mutant progeny at frequencies in the 2–25% range, but in the crosses with the dominant RIP suppressors (and with a derivative of the Adiopodoumé A strain called T-430-Hyg^r a) this frequency was consistently <0.5%. The RIP suppressor phenotype also extended to a duplication of the downy gene (VYAS et al. 2006). Presumably it applies generally. Initial mapping studies suggested that the Adiopodoumé strain contains a dominant RIP suppressor (hereafter referred to as Srp for "Suppressor of RIP") linked to the mating-type locus (mat) on linkage group (LG) IL (BHAT et al. 2003). Here we show that Srp is the Adiopodoumé allele of the ultraviolet photoreactivation-1 (upr-1) gene. This gene encodes the catalytic subunit of the translesion DNA polymerase- (Pol-) (SAKAI et al. 2002). This finding revealed that at least a subset of upr-1 alleles can act as potential regulators of RIP. A role for upr-1 in RIP regulation could have driven its rapid evolution, as evidenced by the unusually high polymorphism exhibited by this gene in N. crassa (TAMULI et al. 2006).

Gene and phenotype symbols:

Gene symbols are italicized, while phenotype symbols are not. For example, the nonitalicized symbol Srp⁺ signifies nonsuppressor phenotype, nonitalicized Srp^+/– signifies intermediate phenotype, and nonitalicized Srp^– signifies suppressor phenotype.

General genetic manipulation of N. crassa and the assay for RIP:

Neurospora genetic analysis was essentially as described by DAVIS and DE SERRES (1970). The frequency of RIP-induced erg-3 mutant progeny among ascospores harvested at 31 days was used as a measure of RIP efficiency and determined as the proportion of colonies with the erg-3 mutant morphology, as seen under a dissection microscope.

General strains:

Unless otherwise stated all N. crassa strains were obtained from the Fungal Genetics Stock Center (FGSC, University of Missouri, Kansas City). They included the standard Oak Ridge (OR) strains 74-OR23-1 A (FGSC 987) and OR8-1 a (FGSC 988) and the mutants erg-3 a (FGSC 2725) and erg-3 A (FGSC 3439). The erg-3 mutation confers a female-sterile phenotype. The Round spore (R) strains R A (FGSC 4022) and R a (FGSC 4023) contain an ascus-dominant 20- to 30-kbp deletion that when crossed with the wild type causes unpairing of the R⁺ allele and its consequent meiotic silencing; consequently, all eight ascospores become round rather than ellipsoid. R also confers a female-sterile phenotype. The auxotrophic strains are arg-1 A (FGSC 325), arg-3 A (FGSC 1068), leu-3 A (FGSC 4002), leu-3 a (FGSC 4003), his-3 A (FGSC 6103), his-3 a (FGSC 7088), his-1 A (FGSC 401), his-1 a (FGSC 402), trp-1 A (FGSC 2038), and trp-1 a (FGSC 2039) and the wild-isolated strain Adiopodoumé A (FGSC 430) and its derivativeT-430-Hyg^r a (FGSC 8609) that was derived from Adiopodoumé A by replacement of the mat A idiomorph by transformation with DNA of the mat a idiomorph (ANDERSON et al. 2001). Both Adiopodoumé A and T-430-Hyg^r a display the dominant RIP suppressor phenotype linked to mat on LG IL (BHAT et al. 2003). The translocation strains are T(VILIR)IBj5 cpc-1 A (FGSC 4433) and T(VILIR)IBj5 cpc-1 a (FGSC 4434), in which a segment of VIL, extending from cpc-1 through ylo-1, is translocated to IR. One-third of the viable progeny from a cross of the translocation with a normal sequence contain a duplication of the translocated segment, designated Dp(VILIR)IBj5. The duplication is stably barren in crosses. The duplication is 400 kbp in size and serves as an effective dominant suppressor of RIP (VYAS et al. 2006).

The Dp(erg-3) A and a strains, described by BHAT et al. (2003) and PRAKASH et al. (1999), contain the transgene Dp(erg-3) that duplicates a fragment of the LG IIIR gene ergosterol-3 (erg-3) tagged with the bacterial hph gene for resistance to hygromycin. The ectopically duplicated fragment serves to target RIP to erg-3. The resulting RIP-induced erg-3 mutant progeny can be scored under a dissection microscope by their distinct colony morphology on Vogel's-sorbose agar medium (NOUBISSI et al. 2000). The Dp(dow) A and a strains were previously referred to as Dp1.5dow^ec A and a (VYAS et al. 2006). They contain the hph-tagged transgene Dp(dow) that duplicates a fragment of the LG IIIR gene downy (dow) and targets RIP to dow. The mus-26 (NcRev7) mutant strains were constructed in the laboratory and have been described by TAMULI et al. (2006).

Strains Sad-2 (RIP32) A (RLM 30-12) and Sad-2 (RIP32) a (RLM33-12) (SHIU et al. 2006) were kindly provided by Robert L. Metzenberg (Department of Biology, California State University, Northridge, CA). They were used to construct the heterokaryons [(leu-3 Srp; Sad-2 a) + (erg-3; his-1 a)] and [(Dp(IBj5); trp-1; Sad-2 a) + (erg-3; his-1 a)]. These heterokaryons were crossed with the strain R; Sad-2; Dp(dow) A.

PCR amplification, molecular markers, and transformation:

Oligonucleotide primers for PCRs were custom purchased from Bioserve (Hyderabad, India). The reaction conditions, other molecular methods, and transformation protocols were as described in BHAT et al. (2004). PCR-based RFLPs were identified to distinguish between the Adiopodoumé and OR genomes in the mat-proximal region of LG IL (Figure 1B). Supplemental Table 1 lists the primers used to amplify 11 genomic segments (P1–P11) from the mat-proximal sequence of super-contig 7.3, as well as the restriction enzymes that enable one to distinguish between the Adiopodoumé and OR alleles of the amplified fragment.

View larger version (11K): In this window In a new window Download PPT slide   FIGURE 1.— (A) Prototrophic progeny produced by crossover if Srp is proximal to mat. For class 1 prototrophs: (I) Srp a, (II) Srp A, and (III) Srp+ A. For class 2 prototrophs: (I) Srp+ A, (II) Srp+ a, and (III) Srp a. If, instead, Srp is distal to mat, then crossovers in II yield Srp+ a in class 1 and Srp A in class 2 but the products of crossover in I and III remain unaffected. (B) Molecular markers in the mat-proximal region. The distances (kilobase pairs) between consecutive pairs of markers from the left (mat) are 12.1, 12.8, 4.1, 2.7, 4.9, 7.0, 7.0, 5.9, 5.6, 8.5, 66.7, and >1042. (C) Localization of Srp to an 34-kbp region proximal to mat. Locations of the crossovers that produced the class 1 Srp A prototroph no. 117 and the class 2 Srp a prototroph no. 51 are shown. P2.1 and P2.2 are single-nucleotide polymorphisms (SNPs) in the interval P2–P3 and P8.1 and P8.2 are SNPs in the interval P8–P9.

 

An OR A-specific DNA insert in molecular marker P7:

The molecular marker P7 is a 1658-bp PCR amplicon when OR A DNA was used as the template, but a smaller fragment was amplified when the template DNA was from the Adiopodoumé strain, OR a, or any of several other wild-isolated N. crassa mat A and mat a strains, or from the N. tetrasperma strain 85 A. A 369-bp sequence (contig 7.3, bases 345,297–345,665) is present in the fragment amplified from OR A, but is missing from all the other strains examined. The OR A-specific "insert" is a composite of three smaller sequences: an 133-bp "left" segment that is repeated at 21 other genomic locations in the sequenced genome, an 45-bp "middle" segment repeated at one other location, and an 171-bp "right" segment that is repeated at 24 other locations. These three segments (or any two) are not found together at any other location in the N. crassa genome sequence. The 369-bp sequence showed limited homology (supplemental Table 2) with the In(het-6) brk1 RS element described by MICALI and SMITH (2006).

Strains used to make Srp-homozygous crosses:

Four leu-3 Srp a segregants with the Srp^– phenotype (398 a, 475 a, 514 a, and 603 a) were obtained from the cross leu-3 A x T-430-Hyg^r a. Segregant 475 a was crossed with arg-1 A and a prototrophic Srp^– A progeny was designated 117 A. From 117 A x Dp(erg-3) a we obtained three mat A segregants, 117-11 A, 117-44 A, and 117-66 A, that were Srp^–, and one mat A segregant, 117-25 A, that was a nonsuppressor (Table 1). Four Srp arg-3 a segregants were obtained from either the cross arg-3 A x T-430-Hyg^r a or arg-3 A x 475 a. Of 393 F₁ segregants from the first cross, 317 were arg-3⁺ (215 a and 102 A) and 76 were arg-3^– (3 a and 73 A). Of 49 leu-3⁺ segregants examined from the second cross, 34 were arg-3⁺ A, 10 were arg-3⁺ a, 4 were arg-3^– A, and 1 was arg-3^– a. The four Srp arg-3 a segregants from these two crosses were confirmed to possess the Srp^– phenotype and designated 33 a, 82 a, 112 a, and 114 a. Two prototrophic Srp^– a segregants from 82 a x leu-3 A were designated 82-46 a and 82-51 a (Table 1). A mat a segregant was obtained from 117-11 A x 82-51 a, which showed an intermediate phenotype (Srp^+/–) and was designated 82-51-6 a. In sum, the strains 117-11 A, 117-44 A, 117-66 A, 82-46 a, and 82-51 a had the Srp^– phenotype; strain 117-25 A was Srp⁺ and 82-51-6 a was Srp^+/– (Table 1). In all seven strains the genomic region around mat was derived from the Adiopodoumé strain (data not shown), which suggested that the Srp⁺ and Srp^+/– phenotypes of 117-25 A and 82-51-6 a were not due to crossing over in this region. Crosses among all these strains were homozygous for Srp (Table 1).

View this table: In this window In a new window   TABLE 1 Strains used to perform Srp-homozygous crosses

 

upr-1 gene disruptions:

Oligonucleotide primers 1F 5' TGG TTG GAC GAG AGT TCT GC and 2R 5' TTT CTT GGC TTG ACA CCG GC 3', and 10F 5' ACT GCT GAA TAA CCG GCA GC and P2R 5' TCC TGA GGA AAG CCA TGA CC, were used to PCR amplify 1410- and 2186-bp segments, respectively, from the 5' and 3' of upr-1^Ad. These segments were cloned in the same orientation into the KpnI and EcoRV sites, respectively, of the vector pCSN44, to generate the plasmid pRT1. The KpnI and EcoRV sites of pCSN44 flank the bacterial hph gene for resistance to hygromycin. Using pRT1 as template and the primers 1F (above) and 11R 5' CCT GCA TGT CGA CCA TAA GC, we amplified an 6-kbp DNA fragment that contained the hph gene flanked by 1410 and 1313 bp, respectively, of the upr-1^Ad 5' and 3' sequences. The amplified DNA was gel purified and transformed by electroporation into the Adiopodoumé strain (FGSC 430). Of 12 hygromycin-resistant transformants examined, the genomic DNA of 3 (nos. 9, 21, and 24) could template the PCR amplification of a 2312-bp fragment with the primers AD62R 5' TGT GCG GCA ATA CCA TGT GC (which anneals to a sequence 150 bp upstream of the 5' end of primer 2R and is therefore absent from the transforming DNA) and AD63F 5' GTC GAC GGT ATC GAT AAG CTT G (which anneals to the pCSN44 multiple-cloning site present only in the transforming DNA). A PCR product is obtained only if the template DNA includes a upr-1 allele disrupted by homologous integration of the transforming DNA. Consequently these primers are "disruption specific." In contrast, the primers 3F 5' CGC TTT GTG CAA CCA GTT CC and 8R 5' GCT ATC CTG AAC ATG ATG GAC amplify a 2519-bp fragment only from an intact upr-1^Ad allele but not from the disrupted allele and hence are "intact allele specific." Transformants 9, 21, and 24 gave PCR products with both the disruption- and intact allele-specific primers, thereby showing that they were heterokaryons that also contained both transformed and untransformed nuclei. These heterokaryons were crossed with OR a. Hygromycin-resistant mat A progeny from these crosses are homokaryons that inherit the disrupted allele, whereas the hygromycin-sensitive mat A progeny inherit the intact upr-1^Ad allele. We could also verify the gene disruption by Southern hybridization because a KpnI restriction site present in the intact allele is absent from the disrupted allele (Figure 2). Disruption of upr-1^Ad conferred an increased sensitivity to UV, a characteristic of upr-1 null mutants (R. TAMULI and D. P. KASBEKAR, unpublished results), thus providing evidence that upr-1^Ad indeed encodes an active Pol-.

View larger version (61K): In this window In a new window Download PPT slide   FIGURE 2.— Disruption of upr-1Ad. (A) The top line is a schematic of the target upr-1Ad allele, and the middle line represents the transforming DNA in which the hph gene (hatched segment) is flanked by 5' and 3' upr-1Ad sequences (respectively, on the right and left of hph). The bottom line shows the disrupted allele resulting from homologous crossovers in the flanking upr-1Ad sequences. The bent arrows represent the primers used to construct the transforming DNA (see text for details) as well as the "disruption-specific" primers AD62R and AD63F and the "intact-specific" primers 8R and 3F. The inverted triangle represents the intron. K, KpnI restriction sites. Note the loss of the middle KpnI site from the disruption allele. (B) Confirmation of upr-1Ad disruption by Southern analysis. Genomic DNA of the Adiopodoumé strain (430), of the three primary transformants (9, 21, and 24), and of four of their hygromycin-resistant mat A homokaryotic progeny (9-20, 21-58, 24-94, and 24-110) was digested with KpnI and probed with the upr-1Ad allele obtained by PCR amplification with the primers 1F and 12R. Note that the 2812- and >3460-bp fragments resulting, respectively, from digestion at the right and middle and the middle and left KpnI sites of the intact upr-1Ad allele are replaced in the disruption progeny strains by a single fragment of 6 kbp due to loss of the middle site from the disrupted allele. A 143-bp fragment expected from the N-terminal end of upr-1Ad is not visible in this blot. Note also that the primary transformants contain both the intact and the disrupted allele as expected for heterokaryons. Although the disruption band is not seen as prominently in the DNA from no. 24, it is seen in its hygromycin-resistant mat A progeny 24-94 and 24-110.

 

Srp shows incomplete penetrance and variable expressivity:

Segregation of the Srp phenotype was examined in 120 F₁ progeny from the cross T-430-Hyg^r a x 74-OR23-1 A by crossing them to Dp(erg-3) strains of the opposite mating type and determining the frequency of erg-3 mutant progeny in the F₂. All 65 mat A progeny were Srp⁺ (erg-3 mutation frequency >1.0%), whereas 51 of the 55 mat a progeny were Srp^– (erg-3 mutation frequency <0.5%) and 4 had an intermediate (Srp^+/–) phenotype (erg-3 mutation frequency 0.5–1.0%). These results confirmed the linkage of Srp to mat. That 4/55 mat a but 0/65 mat A progeny showed an intermediate phenotype alerted us to the possibility that occasionally (here, 7%) Srp might show variable expressivity and confer an Srp^+/– rather than an Srp^– phenotype.

Of 733 F₁ progeny examined from the cross T-430-Hyg^r a x leu-3 A, 673 were parental type with respect to leu-3 and mat (452 leu⁺ a and 221 leu-3 A) and 60 (8.2%) were crossover types (32 leu⁺ A and 28 leu-3 a). All the crossover progeny and 42 leu-3 A and 54 leu⁺ a progeny were tested for their Srp phenotype. The 74 mat A progeny were all Srp⁺, whereas 64 of the 82 mat a progeny were Srp^–, 8 were Srp^+/–, and 10 were Srp⁺. Although the 10 Srp⁺ a progeny could, in principle, have arisen by crossover between Srp and mat, the absence of the complementary Srp^– A progeny led us to postulate that they might not be products of crossover but might reflect an extreme version of the Srp^+/– phenotype. That is, Srp might also be incompletely penetrant.

To confirm the variable expressivity and incomplete penetrance of Srp, we tested Srp-homozygous crosses for the production of non-Srp^– progeny. The cross T-430-Hyg^r a x Adiopodoumé A is infertile (ANDERSON et al. 2001; BHAT et al. 2003), but crosses of the Srp a strains T-430-Hyg^r a, 82-46 a, and 82-51 a with the Srp A strains 117-11 A, 117-44 A, and 117-66 A were fertile (see MATERIALS AND METHODS for strain descriptions). Five of 30 and 10 of 18 progeny examined, respectively, from crosses involving T-430-Hyg^r a, or crosses of 117-11 A with 82-46 a or 82-51 a, showed the Srp^+/– or Srp⁺ phenotypes (Table 2). Confirmation of Srp's incomplete penetrance signaled that the Srp⁺ segregants from a Srp x Srp⁺ cross need not necessarily be of the Srp⁺ genotype and cautioned us to use only the Srp^– segregants (whose genotype is necessarily Srp) for mapping Srp.

View this table: In this window In a new window   TABLE 2 Phenotypes of progeny from Srp-homozygous crosses

 
The non-Srp^– phenotypes of Srp strains were reversible through crosses. Strains 117-25 A and 82-51-6 a are, respectively, Srp⁺ and Srp^+/–, but derive their LG IL region around mat from the Adiopodoumé background (data not shown; see MATERIALS AND METHODS for strain descriptions). Two of the 20 progeny tested from 117-25 A x 82-51-6 a showed reversion to Srp^– (Table 2). Seven other progeny that inherited the mat locus from the Srp^+/– parent showed a further conversion to Srp⁺.

Genetic localization of Srp:

We performed crosses between the strains arg-1 A or arg-3 A and the four leu-3 Srp a strains 398 a, 475 a, 514 a, and 603 a (see MATERIALS AND METHODS) and between leu-3 A and the four Srp arg-3 a strains 33 a, 82 a, 112 a, and 114 a (see MATERIALS AND METHODS). Most progeny from these crosses are auxotrophic for either leucine or arginine, but prototrophs (designated, respectively, as classes 1 and 2 from the two sets of crosses) are produced by crossover between leu-3 and arg-1 (or arg-3), which are, respectively, distal and proximal to mat. Since Srp and mat were closely linked, the auxotrophic markers also bracketed Srp, and the prototrophs were enriched for crossovers between Srp and mat. Depending on whether Srp is proximal or distal to mat, crossover between them yields class 1 and class 2 prototrophs that are, respectively, Srp^– A and Srp⁺ a (Figure 1A) or Srp⁺ a and Srp^– A. Of 81 class 2 prototrophs tested, 80 were parental types (42 Srp⁺ A + 38 Srp^– a) and one, no. 87, was the crossover type, Srp⁺ a (data not shown). This was consistent with Srp being proximal to mat, but also with the alternative model that the prototrophy of no. 87 resulted from a crossover outside mat-Srp and that its Srp⁺ phenotype was due to the incomplete penetrance of Srp.

Of 177 class 1 prototrophs (83 mat a and 94 mat A), 2 (nos. 117 and 141) were Srp^– A (data not shown), as would be expected only if Srp was proximal to mat. The remaining 92 mat A class 1 prototrophs were Srp⁺ and presumably resulted from crossovers proximal to Srp. Of the 83 mat a class 1 prototrophs, 40 were Srp⁺ (data not shown) and were attributed to incomplete penetrance; thus penetrance can occasionally be as low as 52%.

The genetic interval to which Srp maps is bounded distally by the more proximal of the two crossovers that produced the class 1 Srp^– A prototrophs 117 and 141 and proximally by the most distal of the 38 crossovers that produced the class 2 Srp^– a prototrophs (i.e., crossovers in III). The crossovers were mapped with respect to molecular polymorphisms between Adiopodoumé and OR in the mat-proximal region of LG IL (Figure 1B and supplemental Table 1). Figure 1C summarizes this analysis for the crossovers that generated no. 117 and the class 2 Srp^– a prototroph no. 51. The remaining 37 class 2 crossovers were proximal to that in no. 51 and the other class 1 crossover was distal to that in no. 117 (data not shown). Srp was localized to the 34-kbp interval between the SNPs P2.1 and P8.2. Consistent with this localization, prototroph no. 87 was found to have resulted from a crossover between mat and P1 (data not shown).

Srp is identical with upr-1^Ad:

The 34-kbp Srp interval contains 10 genes (or hypothetical genes) designated (proximal to distal) NCU01942.1–NCU01951.1 (http://www.broad.mit.edu/cgi-bin/annotation/fungi/neurospora_crassa_7). The sequence of the corresponding interval from the Adiopodoumé strain (accession no. DQ 387872) revealed at least one nonsynonymous difference between the Adiopodoumé and the OR alleles of five genes (NCU01942.1, NCU01946.1, NCU01947.1, NCU01950.1, and NCU01951.1; supplemental Table 3). The largest candidate gene, NCU01951.1, also known as upr-1, had 118 nucleotide differences between its Adiopodoumé and OR alleles (upr-1^Ad and upr-1^ORA), including 66 nonsynonymous changes. Crosses homozygous for RIP-induced mutants of upr-1 are RIP competent (TAMULI et al. 2006), but this did not disqualify upr-1^Ad from being Srp.

If Srp was identical with upr-1^Ad, disruption of upr-1^Ad should abolish the Srp^– phenotype. We used PCR to amplify an 6-kbp DNA fragment that contained the bacterial hph gene for resistance to hygromycin flanked by 1410 and 1313 bp of sequences from, respectively, the 5' and 3' of upr-1^Ad (Figure 2, also see MATERIALS AND METHODS). The amplified DNA was transformed into the Adiopodoumé strain by electroporation and transformants were selected on hygromycin medium. In N. crassa, transforming DNA can integrate either by nonhomologous end joining into an ectopic site or, less frequently, by homologous integration. We identified three transformants (nos. 9, 21, and 24) in which integration had occurred by homologous recombination, thereby disrupting the upr-1^Ad allele in the transformed nucleus (Figure 2). The transformants were heterokaryons; that is, they also contained additional untransformed nuclei. The heterokaryons were crossed with OR a, and hygromycin-resistant mat A progeny from these crosses represented homokaryons that contained the disrupted allele, whereas the hygromycin-sensitive mat A progeny were homokaryons with the intact upr-1^Ad allele. All segregants that inherited the disrupted upr-1^Ad allele were Srp⁺, whereas the Srp^– or Srp^+/– phenotypes were seen only among segregants that inherited the intact upr-1^Ad allele (Figure 3). Since disruption of upr-1^Ad abolished the Srp^– and Srp^+/– phenotypes, we could conclude that Srp is identical with upr-1^Ad.

View larger version (16K): In this window In a new window Download PPT slide   FIGURE 3.— Frequency of RIP-induced erg-3 mutant progeny from crosses of Dp(erg-3) a with the primary transformants 9, 21, and 24 and with the hygromycin-sensitive (shaded circles) and hygromycin-resistant (open circles) mat A homokaryotic progeny. In each section the primary transformant is indicated by the star (nos. 9, 21, and 24, respectively, in the bottom, middle, and top sections) and the circles represent its progeny homokaryons from crosses with OR a. Note that all segregants homokaryotic for the disrupted upr-1Ad allele (open symbols) were Srp+ (erg-3 frequency >1%) whereas the Srp– or Srp+/– phenotypes (erg-3 frequencies <0.5% and 0.5–1.0%, respectively) were seen only among the heterokaryotic primary transformants and the homokaryotic segregants that inherited the intact upr-1Ad allele (shaded symbols). Note also that several segregants with the intact upr-1Ad allele show the Srp+ phenotype, thereby demonstrating the incomplete penetrance of Srp.

 
The mus-26 gene encodes a regulatory subunit of Pol- (SAKAI et al. 2003). Homozygous mus-26 mutant crosses are not ordinarily RIP defective (TAMULI et al. 2006). We found that RIP was suppressed in the cross Srp; mus-26 (84) a x mus-26 (27); Dp(erg-3) A (frequency of erg-3 progeny <0.5). Therefore suppression of RIP by upr-1^Ad is independent of the regulatory subunit.

A difference in dominant suppression of RIP by Srp and that by large chromosome segment duplications:

The heterokaryon [(leu-3 Srp; Sad-2 a) + (erg-3; his-1 a)] was isolated on the basis of the complementation between its auxotrophic components and used to make crosses with the strain R; Sad-2; Dp(dow) A. Mating of R; Sad-2; Dp(dow) A with the leu-3 Srp; Sad-2 a component is nonproductive due to homozygosity for Sad-2 and crosses of R; Sad-2; Dp(dow) A with adventitiously formed erg-3; his-1 a homokaryons also are nonproductive because both R and erg-3 confer female sterility. However, the heterokaryon is female fertile and therefore progeny can be obtained from the mating between its erg-3; his-1 a component and R; Sad-2; Dp(dow) A. We examined 120 erg⁺ progeny from this cross but none were RIP-induced dow mutants. In contrast, the control Sad-2; Dp(dow) A x erg-3 a crosses yielded RIP-induced dow mutants among the erg⁺ progeny at frequencies typically >8%. These results suggested that the Srp^– phenotype of Srp is able to spread to the Srp⁺ x Srp⁺ asci of the mosaic perithecia that form in [Srp + Srp⁺] x Srp⁺ crosses. (Strictly speaking, since RIP actually occurs in the premeiotic dikaryon that precedes karyogamy, it is more accurate to say the Srp^– phenotype can spread into the premeiotic dikaryons that go on to generate productive karyogamies. The term "asci" is used here merely as a matter of convenience.) This result is consistent with the idea that RIP suppression is due to the synthesis of a diffusible, upr-1^Ad -encoded, suppressor Pol- subunit.

The 400-kbp duplication Dp(IBj5) is an effective dominant suppressor of RIP (VYAS et al. 2006). We constructed the heterokaryon [(Dp(IBj5); trp-1; Sad-2 a) + (erg-3; his-1 a)] and crossed it with the R; Sad-2; Dp(dow) A strain. Again, progeny are generated only from the mating of R; Sad-2; Dp(dow) A with the erg-3; his-1 a component of the heterokaryon. Seven of the 70 erg⁺ progeny examined were RIP-induced dow mutants. Thus, although the RIP suppressor phenotype of Dp(IBj5) is dominant within an ascus, in a mosaic perithecium it did not affect asci that do not contain the Dp. That is, the presumed titration of the RIP machinery requires the presence of the Dp in the ascus.

Conclusions:

We have shown that Srp, the dominant RIP suppressor of the wild-isolated Adiopodoumé strain, is identical with its upr-1 allele (upr-1^Ad). Although Pol- and the other translesion polymerases (Pol-, Pol-, Pol-, and Rev1) are dispensable for RIP (TAMULI et al. 2006), a RIP-suppressive Pol- catalytic subunit encoded by upr-1^Ad might be well placed to interfere with the assembly of other factors essential for RIP in Neurospora. RIP suppression by upr-1^Ad did not require the Pol- regulatory subunit encoded by mus-26. Recently, ACHARYA et al. (2006) established that the Saccharomyces cerevisiae Rev1 protein binds separately to the catalytic and regulatory subunits of Pol- (i.e., Rev3 and Rev7, respectively), suggesting the possibility of separate Rev1–Rev3 and Rev1–Rev7 complexes in vivo. Since Rev1 has been suggested to play a role in assembling Pol-, Pol-, Pol-, and Pol- at the replication fork in mouse and human cells (PRAKASH et al. 2005), it is possible that Rev1 could play a role in assembling the Pol- catalytic subunit at replication forks separately from the regulatory subunit encoded by mus-26. Presumably, the association of a complex without the mus-26-encoded regulatory subunit would not be capable of polymerization but may be capable of inhibiting the association of other factors essential for RIP. The Neurospora Rev1 homolog, encoded by the mus-42 gene, is located 358 kbp distal to upr-1 on LG IL, which is why we have not yet obtained the rare mus-42 upr-1^Ad crossover type that would enable us to put this idea to a genetic test.

Both upr-1^ORA and upr-1^Ad possess out-of-frame ATG codons that initiate short open reading frames upstream of the main upr-1 ORF (TAMULI et al. 2006). Similar upstream out-of-frame ORFs (uORFs) are conserved in the yeast and human homologs of upr-1 (GIBBS et al. 1998). Since uORFs reduce the translational efficiency of the main ORF (MORRIS and GEBALLE 2000), it has been suggested that they serve to keep cellular levels of Pol- very low (LAWRENCE 2002). It is tempting to speculate that the uORFs might contribute to the variable expressivity and incomplete penetrance of Srp. That is, Srp strains that are otherwise genetically identical might differ in uORF expression and this might generate differences in the amount of the suppressor Pol- translated from the main ORF, thus giving rise to the variable expressivity and incomplete penetrance. That is, the variable expressivity and incomplete penetrance might reflect consequences of differences in uORF activity. If the effective cellular levels of the suppressor subunit decrease to below the threshold required to suppress RIP, then a upr-1^Ad strain would mimic the upr-1 null mutant. Incomplete penetrance of the Srp phenotype allows us to reconcile our observations of RIP suppression with the observations of RIP in other Srp-homozygous crosses (ANDERSON et al. 2001 and this work). Temporary loss of suppressor function and its partial or complete recovery in subsequent generations might also involve an epigenetic change reminiscent of paramutation in maize or metastable epialleles in other systems (HOLLICK et al. 1997; POIRIER et al. 2002; RAKYAN et al. 2002; CHANDLER and STAM 2004; DELLA VEDOVA and CONE 2004) or it can be due to stochastic events during development (LALUCQUE and SILAR 2004). Although we have no experimental evidence as yet, it is conceivable that Srp strains are effectively heterokaryotic and express upr-1^Ad in only a subset of nuclei and epigenetically switch it off in other nuclei. Among the progeny from crosses parented by such a heterokaryon, those inheriting the epigenetically silenced upr-1^Ad gene would be phenotypically Srp⁺. The range in Srp's penetrance seen in this work (52–100%) might reflect the range in the proportion of nuclei in which upr-1^Ad is silenced. In sum, the rapidly evolving upr-1 gene (TAMULI et al. 2006) might represent the key to solve the riddle of Tad's survival in the Adiopodoumé strain, wrapped in the mystery of incompletely penetrant dominant RIP suppressors, inside the enigma of RIP regulation.

David Perkins and two anonymous reviewers made several useful comments on earlier versions of this article. We thank Parmit Singh, Srividhya Iyer, and Mukund Ramakrishnan for reconfirming the results of some experiments. G. Giridharan helped us to prepare the figures. Charges for strains from the Fungal Genetics Stock Center (FGSC) were generously waived. The FGSC is supported by National Science Foundation grant BIR-9222772. R.T. was supported by Junior and Senior Research Fellowships from the Council of Scientific and Industrial Research–University Grants Commission, New Delhi.

We dedicate this article to the memory of David D. Perkins (1919–2007).

ACHARYA, N., R. E. JOHNSON, S. PRAKASH and L. PRAKASH, 2006 Complex formation with Rev1 enhances the proficiency of yeast DNA polymerase for mismatch extension and for extension opposite from DNA lesions. Mol. Cell. Biol. 26: 9555–9563.[Abstract/Free Full Text]

ANDERSON, C., T. TANG and J. A. KINSEY, 2001 Elimination of active Tad elements during the sexual phase of the Neurospora crassa life cycle. Fungal Genet. Biol. 33: 49–57.[CrossRef][Medline]

BHAT, A., F. K. NOUBISSI, M. VYAS and D. P. KASBEKAR, 2003 Genetic analysis of wild-isolated Neurospora crassa strains identified as dominant suppressors of repeat-induced point mutation. Genetics 164: 947–961.[Abstract/Free Full Text]

BHAT, A., R. TAMULI and D. P. KASBEKAR, 2004 Genetic transformation of Neurospora tetrasperma, demonstration of repeat-induced point mutation (RIP) in self-crosses and a screen for recessive RIP-defective mutants. Genetics 167: 1155–1164.[Abstract/Free Full Text]

CAMBARERI, E. B., B. C. HENSEN, E. SCHABTACH and E. U. SELKER, 1989 Repeat-induced G-C to A-T mutations in Neurospora. Science 244: 1571–1575.[Abstract/Free Full Text]

CHANDLER, V. L., and M. STAM, 2004 Chromatin conversations: mechanisms and implications of paramutation. Nat. Rev. Genet. 5: 532–544.[CrossRef][Medline]

DAVIS, R. H., and F. J. DE SERRES, 1970 Genetic and microbiological research techniques for Neurospora crassa. Methods Enzymol. 17: 79–143.

DELLA VEDOVA, C. B., and K. C. CONE, 2004 Paramutation: the chromatin connection. Plant Cell 16: 1358–1364.[Free Full Text]

GALAGAN, J. E., and E. U. SELKER, 2004 RIP: the evolutionary cost of genome defense. Trends Genet. 20: 417–423.[CrossRef][Medline]

GALAGAN, J. E., S. E. CALVO, K. A. BORKOVICH, E. U. SELKER, N. D. READ et al., 2003 The genome sequence of the filamentous fungus Neurospora crassa. Nature 422: 859–868.[CrossRef][Medline]

GIBBS, P. E. M., W. G. MCGREGOR, V. M. MAHER, P. NISSON and C. W. LAWRENCE, 1998 A human homolog of the Saccharomyces cerevisiae REV3 gene, which encodes the catalytic subunit of DNA polymerase . Proc. Natl. Acad. Sci. USA 95: 6876–6880.[Abstract/Free Full Text]

HOLLICK, J. B., J. E. DORWEILER and V. L. CHANDLER, 1997 Paramutation and related allelic interactions. Trends Genet. 13: 302–308.[CrossRef][Medline]

KINSEY, J. A., 1989 Restricted distribution of the Tad transposon in strains of Neurospora. Curr. Genet. 15: 271–275.[CrossRef][Medline]

KINSEY, J. A., and J. H. HELBER, 1989 Isolation of a transposable element from Neurospora crassa. Proc. Natl. Acad. Sci. USA 86: 1929–1933.[Abstract/Free Full Text]

KINSEY, J. A., P. W. GARRETT-ENGELE, E. B. CAMBARERI and E. U. SELKER, 1994 The Neurospora transposon Tad is sensitive to repeat-induced point mutation (RIP). Genetics 138: 657–664.[Abstract]

LALUCQUE, H., and P. SILAR, 2004 Incomplete penetrance and variable expressivity of a growth defect as a consequence of knocking out two K⁺ transporters in the euascomycete fungus Podospora anserina. Genetics 166: 125–133.[Abstract/Free Full Text]

LAWRENCE, C. W., 2002 Cellular roles of DNA polymerase and Rev 1 protein. DNA Repair 1: 425–435.[CrossRef][Medline]

MORRIS, D. R., and A. P. GEBALLE, 2000 Upstream open reading frames as regulators of mRNA translation. Mol. Cell. Biol. 20: 8635–8642.[Free Full Text]

MICALI, C. O., and M. L. SMITH, 2006 A nonself recognition gene complex in Neurospora crassa. Genetics 173: 1991–2004.[Abstract/Free Full Text]

NOUBISSI, F. K., K. MCCLUSKEY and D. P. KASBEKAR, 2000 Repeat-induced point mutation (RIP) in crosses with wild-isolated strains of Neurospora crassa: evidence for dominant reduction of RIP. Fungal Genet. Biol. 31: 91–97.[CrossRef][Medline]

NOUBISSI, F. K., K. APARNA, K. MCCLUSKEY and D. P. KASBEKAR, 2001 Evidence for dominant suppression of repeat-induced point mutation (RIP) in crosses with the wild-isolated Neurospora crassa strains Sugartown and Adiopodoumé-7. J. Genet. 80: 55–61.[Medline]

POIRIER, C., A. YOSHIKI, K. FUJIWARA, J. L. GUÉNET and M. KUSAKABE, 2002 Hague (Hag): a new mouse hair mutation with an unstable semidominant allele. Genetics 162: 831–840.[Abstract/Free Full Text]

PRAKASH, A., S. SENGUPTA, K. APARNA and D. P. KASBEKAR, 1999 The erg-3 (sterol 14, 15 reductase) gene of Neurospora crassa: generation of null mutants by repeat-induced point mutation and complementation by proteins chimeric for human lamin B receptor sequences. Microbiology 145: 1443–1451.[Abstract]

PRAKASH, S., R. E. JOHNSON and L. PRAKASH, 2005 Eukaryotic translesion synthesis DNA polymerases: specificity of structure and function. Annu. Rev. Biochem. 74: 317–353.[CrossRef][Medline]

RAKYAN, V. K., M. E. BLEWITT, R. DRUKER, J. I. PREIS and E. WHITELAW, 2002 Metastable epialleles in mammals. Trends Genet. 18: 348–351.[CrossRef][Medline]

SAKAI, W., C. ISHII and H. INOUE, 2002 The upr-1 gene encodes a catalytic subunit of the DNA polymerase which is involved in damage-induced mutagenesis in Neurospora crassa. Mol. Genet. Genomics 267: 401–408.[CrossRef][Medline]

SAKAI, W., Y. WADA, Y. NAOI, C. ISHII and H. INOUE, 2003 Isolation and genetic characterization of the Neurospora crassa REV1 and REV7 homologs: evidence for involvement in damage-induced mutagenesis. DNA Repair 2: 337–346.[CrossRef][Medline]

SHIU, P. K., D. ZICKLER, N. B. RAJU, G. RUPRICH-ROBERT and R. L. METZENBERG, 2006 SAD-2 is required for meiotic silencing by unpaired DNA and perinuclear localization of SAD-1 RNA-directed RNA polymerase. Proc. Natl. Acad. Sci. USA 103: 2243–2248.[Abstract/Free Full Text]

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

TAMULI, R., C. RAVINDRAN and D. P. KASBEKAR, 2006 Translesion DNA polymerases Pol , Pol , Pol , Pol and Rev1 are not essential for repeat-induced point mutation in Neurospora crassa. J. Biosci. 31: 557–564.[CrossRef][Medline]

VYAS, M., C. RAVINDRAN and D. P. KASBEKAR, 2006 Chromosome segment duplications in Neurospora crassa and their effects on repeat-induced point mutation (RIP) and meiotic silencing by unpaired DNA. Genetics 172: 1511–1519.[Abstract/Free Full Text]

Communicating editor: J. J. LOROS

International Access Link

Online ISS 1091-6490
Copyright 2008 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu