A Family of cAMP-Response-Element-Related DNA Sequences With Meiotic Recombination Hotspot Activity in Schizosaccharomyces pombe

Corresponding author: Gerald R. Smith, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. North, A1-162, Seattle, WA 98109-1024., gsmith{at}fhcrc.org (E-mail)

Communicating editor: M. LICHTEN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The heptamer sequence ATGACGT is essential for activity of the M26 meiotic recombination hotspot in the ade6 gene of Schizosaccharomyces pombe. Hotspot activity is associated with binding of the heterodimeric transcription factor Atf1·Pcr1 to M26. We have found that the sequences (C/T/G) TGACGT also bound Atf1·Pcr1 and acted as meiotic hotspots, but unlike M26 they must be followed by A or C for Atf1·Pcr1 binding and hotspot activity. The basis of the hotspot activity of CTGACGTA (ade6-3013) appears to be identical to that of M26: hotspot activity of both sequences was abolished in cells mutant for atf1, pcr1, spc1, or wis1 and was undetectable in mitotic recombination and in meiotic recombination when located on a plasmid. Both hotspot sequences were sites of micrococcal nuclease hypersensitivity in meiotic chromatin, suggesting that they create an open chromatin structure during meiosis at the site of the hotspots. The newly identified hotspot sequences (C/T/G)TGACGT(A/C) and M26 are closely related to the cAMP response element (CRE) consensus sequence for binding of cAMP-responsive transcription factors such as Atf1·Pcr1, suggesting a link between transcription and meiotic recombination. These results significantly expand the list of identified sequences with meiotic recombination hotspot activity in S. pombe from a single sequence to a family of CRE-related sequences.

RECOMBINATION between homologous DNA duplexes plays important roles during mitosis in the repair of DNA damage and during meiosis in the generation of genetic diversity and in the proper segregation of chromosomes. Meiotic recombination does not occur evenly throughout the genome but is elevated in certain regions, called hotspots. Meiotic recombination hotspots have been identified in many organisms from bacteria to mammals (SMITH 1994 ; LICHTEN and GOLDMAN 1995 ).

The M26 hotspot of the fission yeast Schizosaccharomyces pombe has been particularly well characterized (reviewed by FOX and SMITH 1998 ). This hotspot results from a G T transversion in the coding region of the ade6 gene and elevates intragenic recombination up to 15-fold relative to the nearby M375 mutation, an identical G T transversion in the preceding codon (GUTZ 1971 ; SZANKASI et al. 1988 ). M26 also undergoes gene conversion 10 times more frequently than M375 and demonstrates disparity of conversion, with M26 being preferentially converted to wild type.

The hotspot activity of M26 depends on a specific nucleotide sequence, the heptamer ATGACGT (the site of the M26 mutation is underlined; SCHUCHERT et al. 1991 ). Prior to the work described here, the M26 heptamer was the only sequence known to have hotspot activity in S. pombe. Extensive mutational analysis demonstrated that hotspot activity required all seven of these nucleotides; however, mutations outside the heptamer had no significant effect on recombinant frequency. The apparently unique nucleotide sequence of M26 is unusual for meiotic recombination hotspots and greatly facilitates study of the mechanism by which recombination is elevated at this hotspot.

The M26 heptamer sequence is specifically bound by a heterodimeric complex of the proteins Atf1 and Pcr1 (KON et al. 1997 ), transcription factors that bind the cAMP response element (TAKEDA et al. 1995 ; KANOH et al. 1996 ; SHIOZAKI and RUSSELL 1996 ; WATANABE and YAMAMOTO 1996 ). Hotspot activity correlates with the ability of mutant sequences to bind the Atf1·Pcr1 heterodimer (WAHLS and SMITH 1994 ; KON et al. 1997 ). Transcriptional activity of Atf1·Pcr1 requires the product of the wis1 gene, a protein kinase that phosphorylates Spc1, which in turn phosphorylates Atf1 following activation of a stress-induced protein kinase cascade (SHIOZAKI and RUSSELL 1995 , SHIOZAKI and RUSSELL 1996 ; DEGOLS et al. 1996 ; STETTLER et al. 1996 ; WILKINSON et al. 1996 ; SHIEH et al. 1997 ). The product of the spc1 gene is also required for M26 hotspot function (KON et al. 1998 ); however, the relationship between transcription and recombination is not clear (KON et al. 1997 ; see DISCUSSION).

Several lines of evidence support a role for chromatin structure in hotspot activity. The M26 heptamer is inactive on a plasmid and in several transplacements in which 3- to 6-kb regions of DNA containing the ade6-M26 gene were transplaced to other genomic locations (PONTICELLI and SMITH 1992 ; VIRGIN et al. 1995 ). In contrast, the heptamer is active in either orientation at several other sites, when made by site-directed mutagenesis of 1–3 bp without accompanying disruption of gross chromosomal structure (FOX et al. 1997 ). Certain large chromosomal integrations may disrupt a chromatin structure required for hotspot function. The site of the M26 mutation is also associated with chromatin remodeling, as revealed by changes in patterns of micrococcal nuclease-sensitive sites in chromatin. In cells bearing the ade6-M375 mutation, MNase-sensitive sites appear at 150-bp intervals, reflecting nucleosome phasing, along the entire ade6 coding region. In contrast, the nuclease-sensitivity pattern in cells bearing the M26 heptamer sequence is similar to that in naked DNA, suggesting that nucleosome positioning is rearranged. In addition, a new hypersensitive site appears near the M26 heptamer sequence. This hypersensitive site, together with a hypersensitive site in the 5' noncoding region of the ade6-M26 locus, is present in chromatin isolated from premeiotic cells and becomes pronounced as cells enter meiosis or become stressed (MIZUNO et al. 1997 ). Thus, the M26 heptamer sequence may promote hotspot activity through the local opening of chromatin, allowing recombination-specific proteins access to the DNA.

Work on the mammalian transcription factor ATF1 has defined a binding consensus sequence, cAMP response element (CRE), which is present in genes that are transcriptionally regulated by ATF1 in response to intracellular cAMP levels (MONTMINY et al. 1986 ). This consensus, TGACGTC, bears striking similarity to the M26 heptamer sequence ATGACGT: both have in common the hexamer TGACGT. S. pombe protein extracts contain a factor that specifically binds to a DNA probe containing the sequence CTGACGTAAC with this hexamer (JONES and JONES 1989 ). S. pombe atf1 was also cloned as a gene called gad7, required for the starvation-induced arrest of cells in G₁ (KANOH et al. 1996 ). The Gad7 (Atf1) protein binds to a DNA probe containing the CRE sequence CTGACGTCAG. These studies suggested that S. pombe Atf1·Pcr1 can bind to sequences other than the M26 heptamer. From the work cited above, one such sequence appeared to be CTGACGT, and it seemed plausible that other M26-like sequences may also be binding sites for Atf1·Pcr1. Since binding to M26 correlates with hotspot activity, these sequences may also be novel meiotic recombination hotspots. To test this hypothesis, we constructed S. pombe strains in which the ade6 gene was altered at the site of the M26 heptamer to make the following sequences: CTGACGTN, TTGACGTN, and GTGACGTN. Atf1·Pcr1 bound to a subset of these sequences, but binding depended on the nucleotide following the heptamer. Sequences that bound Atf1·Pcr1 in vitro had meiotic recombination hotspot activity in vivo. These studies have thus identified a new family of sequences that can act as meiotic recombination hotspots in S. pombe.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

S. pombe strains and crosses:
The wild-type strains used for mutagenesis and preparation of protein extracts were GP18 (h^- leu1-32) and GP20 (h⁺ leu1-32). Strains containing the mutant ade6 alleles studied here are listed in Table 1. Strains GP14 (h⁺ ade6-52) and GP5 (h⁺ ade6-M216) were used for the crosses in Fig 2. The homothallic strain used in Table 4 was GP341 (h⁹⁰ ade6-469 ura4-294). Strains mutant for atf1, pcr1, spc1, or wis1, provided by K. Shiozaki and P. Russell (SHIOZAKI and RUSSELL 1996 ), have the following genotypes: JM544 (GP2204, h⁺ wis1::ura4⁺ leu1-32 ura4-D18); KS1517 (GP2206, h⁺ spc1::ura4⁺ leu1-32 ura4-D18); KS1512 (GP2208, h⁺ atf1::ura4⁺ leu1-32 ura4-D18); KS1648 (GP2435, h^- pcr1::his7⁺ his7-366 ura4-D18 leu1-32). The unstable diploids used for preparation of chromatin were GP2611 (h⁺/h^- ade6-3013/ade6-3013 leu1-32/leu1-32 ura1-61/+ pro1-1/+) and GP2612 (h⁺/h^- ade6-3015/ade6-3015 leu1-32/leu1-32 ura1-61/+ lys3-37/+). Genealogies and sources of strains are available on request.

View larger version (30K): In this window In a new window Download PPT slide   Figure 1. In vitro binding to DNA probes containing CRE-like sequences. – extract, control reactions containing M26 probe and lacking protein extract; F, free probe; B, bound probe. (A) Binding reactions containing wild-type extract and probes as indicated. The CRE-like sequences in these probes were ade6-M26 (ATGACGTGAG), ade6-M375 (AGGACGTGAG), ade6-3012 (CTCACGTAAC), ade6-3014 (CTCACGTAAT), ade6-3013 (CTGACGTAAC), and ade6-3015 (CTGACGCAAT). (B) Binding of extracts from wild-type and mutant strains to probe containing ade6-3013 (CTGACGTAAC). (C) Binding of extracts from wild-type and mutant strains to probe containing ade6-M26 (ATGACGTGAG).

View larger version (56K): In this window In a new window Download PPT slide   Figure 2. Hotspot activity of CRE-like sequences. Strains carrying the sequences indicated were crossed with strains carrying either ade6-52 (A) or ade6-M216 (B). Recombinant frequencies are arithmetic means of 4–8 independent crosses, with standard error of the mean indicated by bars. Subsequent crosses with the ade6-3017 strains, recovered from frozen stocks, produced recombinant frequencies only slightly greater than those with ade6-M26 strains (data not shown); the basis of this difference is unclear.

All strains were grown on rich yeast extract agar (YEA) solid or yeast extract liquid (YEL) liquid media (GUTZ et al. 1974 ) or on modified Edinburgh minimal medium 2 (EMM2; NURSE 1975 ) supplemented as appropriate. Sporulation was at 25° on sporulation agar (SPA) solid medium supplemented with adenine and leucine (GUTZ et al. 1974 ). The frequency of recombination between ade6 alleles in heterothallic crosses was determined by random spore analysis (PONTICELLI et al. 1988 ). Spores were plated on supplemented EMM2 with or without adenine to determine total spores or Ade⁺ recombinants, respectively. Each cross was repeated from independent cultures at least four times, and more than 50 colonies were counted for each determination. The frequency of Ade⁺ revertants, as determined by meiotic selfings, was <0.1/10⁶ viable spores (data not shown; DEVEAUX et al. 1992 ).

ade6 alleles:
ade6 alleles used in recombination studies were ade6-M26 (G1010T), ade6-M375 (G1007T), ade6-469 (C2342T), ade6-M216 (G921A) (SZANKASI et al. 1988 ), and ade6-52 (G1670A) (M. E. FOX, unpublished data). Nucleotides are numbered according to SZANKASI et al. 1988 ; the ade6 coding region extends from 875 to 2530. It should be noted that the mutation corresponding to the ade6-52 allele differs from that previously reported (FOX et al. 1997 ). The G1670A mutation was identified by sequencing of genomic DNA isolated from three different ade6-52 strains from our laboratory and a different ade6-52 strain provided by J. Kohli. ade6 alleles created in this study are described in Table 1.

Mutagenesis of the chromosomal ade6 locus:
Mutations were created in the chromosomal ade6 locus by site-directed mutagenesis of the ade6⁺ gene on plasmid pMF1 (1.45-kb BamHI-XhoI fragment of ade6⁺ subcloned into pKS⁺; Stratagene, La Jolla, CA), followed by transformation of strain GP18 to adenine auxotrophy with a linear DNA fragment bearing the desired mutation, as follows. Site-directed mutagenesis of plasmid pMF1 was by polymerase chain reaction (PCR)-based mutagenesis (FOX et al. 1997 ) or by the Morph mutagenesis system (5Prime 3Prime, Boulder, CO) and was confirmed by sequencing. Mutations were transferred to the chromosomal ade6 locus by transformation (ITO et al. 1983 ). Strain GP18 (h^- leu1-32) was cotransformed with at least 1 µg of a 1.45-kb BamHI-XhoI fragment of ade6 bearing the desired mutation and 1 µg of circular plasmid pDB248' (BEACH et al. 1982 ) carrying the Saccharomyces cerevisiae LEU2 marker, which complements the S. pombe leu1-32 mutation. Leu⁺ transformants were selected on EMM2 medium with limiting adenine (10 µg/ml) on which Ade^- cells produce red colonies (GUTZ et al. 1974 ). Ade^- strains were purified on media containing leucine, and Leu^- segregants were identified. Chromosomal mutagenesis was confirmed by PCR amplification and sequencing and by Southern blot analysis (data not shown).

Binding assays:
The DNA probes used were 215-bp dsDNA fragments containing the mutated ade6 sequences, prepared by StyI digestion of plasmid DNA and gel purification. Probes were end-labeled using [-³²P]dCTP and the Klenow fragment of Escherichia coli DNA polymerase I (New England Biolabs, Beverly, MA).

Crude whole-cell protein extracts were prepared from exponentially growing cultures of strains GP20, GP2204, GP2206, GP2208, and GP2435 as follows. Cell pellets (2 g wet weight) were lysed in an equal volume of 50 mM HEPES-NaOH (pH 7.9), 500 mM NaCl, 20% glycerol, 1 mM EDTA, 1 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, 10 µg/ml bestatin, 1 µg/ml leupeptin, 1 µg/ml pepstatin by the addition of acid-washed glass beads and 10 rounds of vortexing for 30 sec each with cooling on ice between each round. Following centrifugation at 15,000 x g for 30 min at 4°, the supernatant was removed and protein concentration determined using a modified Bradford assay (Bio-Rad, Richmond, CA). Extracts were diluted with the buffer above to 10 mg/ml, stored in aliquots at -70°, and thawed on ice immediately before use.

Binding reactions were performed in a final volume of 10 µl containing 0.2 ng labeled probe DNA, 10 µg protein extract, 1 µg poly(dI-dC) (Sigma, St. Louis), 12% glycerol, 12 mM HEPES-NaOH (pH 7.9), 5 mM MgCl₂, 4 mM Tris-HCl (pH 7.9), 0.6 mM EDTA, and 0.6 mM DTT. Reactions were incubated at room temperature for 30 min, and the products were separated on 5% nondenaturing polyacrylamide gels at 16 V/cm in 1x TGE buffer [50 mM Tris-HCl (pH 7.9), 380 mM glycine, 2 mM EDTA]. Gels were dried onto Whatman 3MM paper and autoradiographed or analyzed using a Molecular Dynamics (Sunnyvale, CA) PhosphorImager and ImageQuant software.

In vitro binding site selection:
DNA sequences bound by Atf1·Pcr1 were selected using a modified Selex procedure as described (TUERK and GOLD 1990 ; CUI et al. 1995 ). The probe used was a 65-bp dsDNA molecule containing a central core of 11 randomized nucleotides centered over the site of the M26 heptamer and flanked on the left by GG and 24 bp of ade6 DNA and on the right by 20 bp of the ade6 DNA and GGATCCGG to provide restriction sites for convenient cloning (5' GGAATTCTCCTGCCAAACAAATTGATNNNNNNNNNNNGCACATTGATGCATCATTTAGGATCCGG). This probe was end-labeled using T4 polynucleotide kinase (New England Biolabs) and incubated with protein extract from strain GP18. The products were analyzed as above using an equivalent probe containing the M26 heptamer sequence as a control marker for migration of bound probe. The region of the gel corresponding to the position of bound probe was excised and crushed in 100 µl of H₂O. After elution from the gel, nucleic acids were precipitated with ethanol and used as template in a PCR amplification with primers hybridizing to the ends of the 65-bp probe (5' GGAATTCTCCTGCCAAAC and 5' CCGGATCCTAAATGATGC). The PCR product was end-labeled and used as a probe in a subsequent round of binding and fractionation. After three rounds of binding and PCR amplification, bound probe was readily detectable as a shifted band migrating at the position of bound probe containing the M26 heptamer. PCR products from rounds three and four were subcloned into pCRII (Invitrogen, San Diego), and the inserts were sequenced. Since binding to the desired DNA targets should depend on Atf1 protein, specificity of binding to candidate sequences was determined in binding reactions containing BamHI-digested plasmid DNA and protein extract from atf1⁺ and atf1 cells.

Micrococcal nuclease sensitivity of chromatin:
Meiotic chromatin structure of ade6 was analyzed as described (MIZUNO et al. 1997 ). Chromatin was isolated from diploid strains GP2611 and GP2612, homozygous for ade6-3013 and ade6-3015, respectively, at 0 and 3 hr after induction of sporulation. Chromatin was digested with 0, 20, and 30 units/ml of micrococcal nuclease (Amersham Pharmacia). MNase-digested DNA was cleaved with XhoI and the resulting restriction fragments were analyzed by Southern hybridization using a probe for the 3' terminus of the ade6 coding region (a 238-bp XhoI-EcoRI fragment). Digestion of naked genomic DNA was as described (MIZUNO et al. 1997 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In vitro binding of Atf1·Pcr1 to a probe containing CTGACGTAAC:
To test the hypothesis that multiple sequences can bind Atf1·Pcr1, we constructed plasmids carrying mutated sequences at the site of the M26 heptamer in the ade6 gene. Initial experiments involved the candidate sequence CTGACGTAAC (JONES and JONES 1989 ; KANOH et al. 1996 ). Gel-shift assays using wild-type S. pombe extract showed binding to 215-bp dsDNA fragments containing CTGACGTAAC (ade6-3013), but not to probes containing the closely related sequences CTCACGTAAC (ade6-3012), CTCACGCAAT (ade6-3014), or CTGACGCAAT (ade6-3015) (Fig 1A; these sequences differ from ade6-3013 at the underlined positions). Binding to the probe containing CTGACGTAAC (ade6-3013) was abolished in the absence of Atf1 or Pcr1 and was reduced in the absence of Spc1 or Wis1 (Fig 1B). Similar requirements for binding were observed with a probe containing the M26 heptamer (Fig 1C). These probes were therefore bound by a factor whose synthesis or activity is under the control of atf1, pcr1, spc1, and wis1. The similar electrophoretic mobilities of the numerous complexes reported here (Fig 1 and additional data not shown) suggest that this factor is Atf1·Pcr1 itself, as shown for M26 with purified Atf1·Pcr1 protein (WAHLS and SMITH 1994 ; KON et al. 1997 ). These binding results suggested that CTGACGTAAC (ade6-3013) might be a recombination hotspot similar to M26.

CTGACGTAAC (ade6-3013) is a meiotic recombination hotspot:
To test this hypothesis, the sequence CTGACGTAAC (ade6-3013) and the control sequence CTGACGCAAT (ade6-3015) were introduced into the ade6 gene on the chromosome and assayed for recombination hotspot activity. The ade6-3015 control was selected because it involves the same number and type of base-pair changes as ade6-3013 and thus reduces potential complications due to differential mismatch correction. Meiotic crosses were performed between strains carrying these alleles and strains carrying either the ade6-52 or the ade6-M216 allele, which map to either side of M26. For comparison, crosses were also performed with strains carrying the ade6-M26 hotspot mutation or its nonhotspot control ade6-M375. In crosses with ade6-52 the presence of the CTGACGTAAC sequence (ade6-3013) increased the Ade⁺ recombinant frequency about 13-fold relative to the control CTGACGCAAT (ade6-3015); the hotspot activity of 3013 was indistinguishable from the 13-fold enhancement of M26 relative to M375 (Fig 2A; Table 2). In crosses with ade6-M216, the 3013 mutation increased the recombinant frequency 6-fold relative to its control (Fig 2B); this also is comparable to the 5-fold hotspot activity of M26 relative to M375 in crosses with ade6-M216 (Fig 2B). 3013 is thus active as a hotspot in crosses with markers to either side of itself, as previously shown for M26 (GUTZ 1971 ). As further controls, strains with the sequences CTCACGTAAC (ade6-3012) and CTCACGCAAT (ade6-3014), which did not bind Atf1·Pcr1 (Fig 1A), were constructed and tested for hotspot activity in crosses with ade6-52 and ade6-M216. For both sequences recombinant frequencies were comparable to those obtained with M375 (Fig 2). Thus, in vitro binding of protein correlated with hotspot activity in vivo. These results established that ade6-3013 creates a meiotic recombination hotspot with properties similar to those of M26 (see additional results below).

Other CRE-related sequences that bind Atf1·Pcr1 and act as recombination hotspots:
The sequence previously shown to bind Gad7 (Atf1), CTGACGTCAG (ade6-3017) (KANOH et al. 1996 ), was similarly tested and found to bind Atf1·Pcr1 in vitro and to have recombination hotspot activity in vivo (Table 3; Fig 2). This result suggested that multiple related sequences might bind Atf1·Pcr1 and have hotspot activity.

In parallel studies, a search for DNA sequences that are bound by Atf1·Pcr1 was performed using a modification of the Selex approach (TUERK and GOLD 1990 ; CUI et al. 1995 ). This search was not carried to exhaustion but did yield several sequences that demonstrated Atf1-dependent binding (data not shown). These included the M26 heptamer and the sequence GTGACGTAAG. The similarity of this sequence to those shown above to have hotspot activity suggested that other sequences related to the CRE consensus also act as hotspots. To test this suggestion by more rigorous genetic means, a series of mutant sequences, (C/T/G)TGACGTN, was generated and introduced into the ade6 gene; the four sequences ATGACGTN had previously been tested during identification of the M26 heptamer and found to be active as meiotic recombination hotspots (SCHUCHERT et al. 1991 ). This series of sequences and their corresponding ade6 allele designations are shown in Table 1.

In vitro binding assays were performed using probes carrying these mutated sequences with extract from wild-type cells or from atf1 or pcr1 mutant cells (Table 3). The following sequences bound significantly more protein than the others: ATGACGTGAG, CTGACGTAAG, CTGACGTCAG, GTGACGTAAG, GTGACGTCAG, TTGACGTAAG, and TTGACGTCAG (only the underlined nucleotides differ among these sequences). In all cases significant binding was dependent on the presence of Atf1 and Pcr1 (Table 3 and additional data not shown).

To test whether these Atf1·Pcr1-binding sequences have recombination hotspot activity, strains bearing the sequences in Table 1 on their chromosomes were crossed with strains bearing either the ade6-52 allele or the ade6-M216 allele, and recombinant frequencies were determined (Fig 2). All sequences that displayed in vitro binding above background were also meiotic recombination hotspots, while those that had no significant binding displayed recombinant frequencies comparable to those of M375. Hotspot activity thus correlated with in vitro binding, although the variability in the binding assay precludes a quantitative comparison.

These results indicate that the (C/G/T)TGACGT heptamers are recombination hotspots, but only if followed by either A or C. This contrasts with M26 for which changing the position following ATGACGT from G to A, C, or T has no significant effect on hotspot activity (SCHUCHERT et al. 1991 ). Limited studies suggest that hotspot activity is independent of the next two nucleotides as well: CTGACGTAAC (ade6-3013), CTGACGTAAT (ade6-3019), and CTGACGTAGG (ade6-3018) all demonstrated Atf1-dependent protein binding and had comparable hotspot activity in crosses with both ade6-52 and ade6-M216 (Fig 1 and Fig 2 and data not shown).

The sequence GTGACGTGAG (ade6-3025) had an intermediate recombinant frequency in crosses with both ade6-52 and ade6-M216; recombinant frequencies were increased approximately threefold relative to equivalent crosses involving ade6-M375 but decreased relative to M26 (Fig 2). The original mutational analysis of the M26 heptamer also found recombinant frequencies for this sequence reduced approximately threefold relative to M26 but above that with ade6-M375 (SCHUCHERT et al. 1991 ); this was a significant reduction, although not as great as that observed for other single base changes at the first position of the M26 heptamer (ATGACGT). The ade6-3025 sequence did not bind Atf1·Pcr1 significantly above background levels (Table 3), but a low level might not have been detected in this assay.

In summary, M26 is not the only sequence that displays meiotic hotspot activity. Rather, there is a family of related sequences, ATGACGT and (C/G/T)TGACGT(C/A), that bind Atf1·Pcr1 and function as meiotic recombination hotspots.

The CTGACGTAAC sequence (ade6-3013) forms a meiosis-specific hotspot with properties similar to those of the M26 hotspot:
The sequences shown above to have hotspot activity are closely related to the M26 heptamer sequence and, like M26, bind Atf1·Pcr1, suggesting that the mechanism of the new hotspots may be similar to that of M26. The following experiments were performed to determine whether one of the novel hotspots, CTGACGTAAC (ade6-3013), behaves in a manner similar to M26. The control used was CTGACGCAAT (ade6-3015), a sequence with the same number and type of nucleotide changes but with a recombinant frequency similar to that of M375 (Fig 2; Table 2).

In vitro binding and hotspot activity require Atf1 and Pcr1: To test the genetic requirements for binding and hotspot activity, null alleles of atf1, pcr1, wis1, or spc1 were introduced into strains carrying either ade6-M26, ade6-M375, ade6-3013, or ade6-3015. As shown previously, hotspot activity of M26 requires Atf1 and Pcr1 (Table 2; KON et al. 1997 ). Furthermore, hotspot activity of M26 appears to require activation of the protein kinase cascade involving Wis1 and Spc1, since M26 was inactive in mutants lacking Wis1 (Table 2) or Spc1 (Table 2; KON et al. 1998 ). The 3013 hotspot behaved like the M26 hotspot. As noted previously, the 3013 sequence increased recombinant frequency 13-fold relative to its control (3015), a hotspot activity comparable to that of M26 relative to M375, and the 3013 hotspot activity was also dependent on Atf1, Pcr1, Wis1, and Spc1 (Table 2). In vitro binding to probe containing 3013 or M26 was abolished in the absence of Atf1 and Pcr1 and reduced, but not abolished, in the absence of Wis1 or Spc1 (Fig 1B and Fig C). Thus, the genetic and biochemical properties of the 3013 hotspot resemble those of M26.

The 3013 sequence is inactive in mitosis and when on a plasmid: Activity of the M26 hotspot is meiosis specific: no enhancement of recombinant frequency relative to that with M375 is seen during mitotic growth (PONTICELLI et al. 1988 ; SCHUCHERT and KOHLI 1988 ). M26 is also inactive in plasmid-by-chromosome crosses when on the plasmid but active when on the chromosome (PONTICELLI and SMITH 1992 ). The 3013 hotspot was tested to determine whether its activity is meiosis- and context-dependent in a similar manner. The control used was ade6-3015, which was inactive for both in vitro binding and in vivo hotspot activity (Fig 1A and Table 2). Strains bearing this hotspot or the control sequence were transformed with a plasmid carrying the ade6-469 allele [a C T mutation located 1.33 kb from 3013 (SZANKASI et al. 1988 )] and grown mitotically to allow plasmid-by-chromosome recombination to generate Ade⁺ products. The presence of the 3013 hotspot had no significant effect on mitotic recombinant frequency (Table 4A). The 3013 hotspot was active, however, when meiosis was induced in these transformed cultures (data not shown). Hotspot activity of the 3013 sequence, like that of M26, is thus meiosis specific.

To test for meiotic activity of 3013 when on a plasmid, we constructed homothallic strains bearing ade6-469 on the chromosome and the hotspot alleles or their controls on a 3-kb fragment of ade6 in the plasmid pade6 (SZANKASI et al. 1988 ). The frequency of meiotic Ade⁺ recombinants was nearly the same for all four alleles (Table 4B), demonstrating that neither the M26 nor the 3013 hotspot was active when on a plasmid. These results suggest that hotspot activity of the 3013 hotspot, like that of M26, is dependent on its genomic location and may require a specific chromatin structure.

Chromatin structure is altered at the 3013 sequence: To determine whether chromatin structure of the ade6 gene is altered in strains carrying the ade6-3013 hotspot, chromatin isolated from diploids homozygous for ade6-3013 or the control ade6-3015 was digested with MNase and analyzed by Southern blotting (Fig 3). MNase sensitivity of the ade6 region in the 3015 control had a pattern similar to that observed previously for both wild-type and ade6-M375 (MIZUNO et al. 1997 ). In contrast, the pattern of MNase-sensitive sites around the 3013 sequence (ca. +50 to +800) was different and was more similar to that observed in the naked DNA digestion (Fig 3; additional data not shown). In addition, a new hypersensitive site was observed in 3013 at the site of the new sequence, which was absent in 3015 chromatin. This hypersensitive site was detectable at low levels in premeiotic cultures and increased in intensity following induction of meiosis. Such modified patterns in the MNase sensitivity are very similar to those at the M26 site (MIZUNO et al. 1997 ). Thus, both the M26 and 3013 hotspots are associated with specific remodeling of chromatin in the region of ade6 as cells enter meiosis.

View larger version (71K): In this window In a new window Download PPT slide   Figure 3. Chromatin structural change at the ade6-3013 hotspot. Before (0 hr) and after (3 hr) meiotic induction, chromatin was isolated from strain GP2611 (ade6-3013) and from the negative control strain GP2612 (ade6-3015). Samples were analyzed by partial digestion of chromatin with MNase (0, 20, and 30 units/ml) followed by XhoI digestion and indirect end-labeling of Southern blots as described (MIZUNO et al. 1997 ). One microgram of genomic DNA was analyzed in each lane. An arrow indicates the position of the 3013 mutation. The nucleotide positions of restriction sites for the size markers are indicated relative to the first A of the ade6 coding sequence (SZANKASI et al. 1988 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We show here that M26, previously considered a unique sequence (SCHUCHERT et al. 1991 ; LICHTEN and GOLDMAN 1995 ; FOX and SMITH 1998 ), is not the only meiotic recombination hotspot in S. pombe and describe a family of sequences related to the M26 heptamer that act as hotspots in S. pombe. Since an exhaustive search for sequences that bind the hotspot-activating transcription factor Atf1·Pcr1 was not performed, there may well be additional sequences that function as hotspots. In the currently sequenced 11.3 Mb of S. pombe DNA there are 286 copies of the M26 hotspot sequence and 587 copies of the new hotspot sequences (Cre hotspots, Table 5). If these naturally occurring sequences act similarly to enhance recombination, our findings suggest that recombination hotspots contribute more to meiotic recombination than previously calculated (KON et al. 1997 ).

The recombination hotspot sequences described here demonstrate an unusual dependence upon surrounding nucleotides. The M26 heptamer, ATGACGT, is active irrespective of the nucleotide following the heptamer. In contrast, the novel hotspot sequences described here display a strong dependence on the nucleotide following the heptamer: CTGACGT, TTGACGT, and GTGACGT are all active if followed by a C or A, but inactive if followed by a G or T (Fig 2; Table 5). Nucleotides within a consensus are typically considered to be independent: the effects of changing one base to another are generally independent of changes at another position, and interdependence of nucleotides, such as that between nucleotides at positions 1 and 8 in Table 5, is unusual. The mammalian CRE binding protein CREB1 shows a similar interdependence of nucleotides within the CRE consensus sequence for binding and transcriptional activation (BENBROOK and JONES 1994 ). In this case, sequences in which the G at position 6 (Table 5) is mutated to either C or A are active only if the nucleotide at position 3 is a G and that at position 8 is a C. These observations underscore the necessity of testing multiply mutated sequences to determine the nucleotide sequence(s) necessary for activity.

Activity of the M26 hotspot is associated with binding of the transcription factor Atf1·Pcr1 (KON et al. 1997 ). It appears that binding of Atf1·Pcr1 is also required for activity of the novel hotspots identified here. All sequences associated with hotspot function bound a factor whose synthesis or activity is controlled by atf1 and pcr1 (Fig 1; Table 3); we infer that this factor is Atf1·Pcr1 itself (see RESULTS). Binding of this factor was not significantly above background with sequences that did not enhance recombination. The correlation of binding and hotspot activity suggests that Atf1·Pcr1 directly activates the M26 hotspot (WAHLS and SMITH 1994 ) as well as the family of hotspots studied here.

The Atf1 transcription factor is activated through a stress-induced protein kinase cascade in which Wis1 phosphorylates Spc1, which in turn phosphorylates Atf1 (SHIOZAKI and RUSSELL 1995 , SHIOZAKI and RUSSELL 1996 ; DEGOLS et al. 1996 ; STETTLER et al. 1996 ; WILKINSON et al. 1996 ; SHIEH et al. 1997 ). Spc1 and Wis1 are required for transcriptional activity of Atf1, although it is not clear whether phosphorylation itself is required. Recombination hotspot activity of M26 also requires Spc1 and Wis1 (KON et al. 1998 ; Table 2), indicating that this protein kinase cascade is also required for the recombination functions of Atf1·Pcr1 bound to the M26 heptamer. For the one other sequence tested further, 3013, hotspot activity similarly required Atf1, Pcr1, Spc1, and Wis1 (Table 2). Interestingly, for both M26 and 3013 in vitro binding was undetectable in the absence of Atf1 or Pcr1 but was detectable at reduced levels in the absence of Wis1 or Spc1 (Fig 1). The reduced levels of binding in the absence of these protein kinases might reflect reduced transcription of atf1 in these mutants (SHIOZAKI and RUSSELL 1996 ; KON et al. 1998 ), although Atf1·Pcr1 protein levels are reported to be unaltered in these mutants (DEGOLS et al. 1996 ; KON et al. 1998 ). Alternatively, the reduced levels of binding might reflect a weaker affinity of unphosphorylated Atf1 protein relative to phosphorylated, although treatment of purified Atf1·Pcr1 protein with phosphatase is reported not to affect binding (KON et al. 1998 ). Further studies are required to elucidate the role of the Spc1 and Wis1 protein kinases in meiotic recombination hotspot activity.

The hotspot sequences described here are very similar to the CRE consensus sequence TGACGT(C/A) that is required for transcriptional activation of cAMP-responsive genes (MONTMINY et al. 1986 ). However, the relationship between recombination and transcription is complex. It has been suggested that hotspot activity of M26 requires transcription of the ade6 gene. Deletion of 500 bp containing the putative ade6 promoter abolishes M26 activity (ZAHN-ZABAL et al. 1995 ), and replacement of the weak ade6 promoter with the strong adh1 promoter increases recombination at ade6 (GRIMM et al. 1991 ). However, both of these results could reflect the gross structural changes made to the ade6 region, rather than effects on transcription of ade6. Evidence against a requirement for transcription of ade6 is the observation that ade6 transcript levels are equivalent in M26 and M375 strains and are not affected by mutations in atf1 or pcr1 (KON et al. 1997 ). In addition, while the Atf1 transcription factor and the Wis1 and Spc1 kinases are required for both transcriptional activation and for meiotic hotspot activity, the stress response and hotspot activity are separable: pcr1 is required for M26 hotspot activity but not for the stress response (KON et al. 1997 , KON et al. 1998 ). It is possible that Atf1 has other, as yet unidentified, binding partners that are required for transcriptional activation in response to stress.

In addition to the genetic requirements for atf1, pcr1, wis1, and spc1 discussed above, the 3013 hotspot shares other properties with M26: both hotspots are inactive during mitosis and when located on a plasmid. It has been proposed that hotspot activity of M26 requires a particular chromatin structure that is absent on a plasmid and during mitosis (PONTICELLI and SMITH 1992 ; VIRGIN et al. 1995 ; FOX et al. 1997 ). Indeed, the entry of cells into meiosis is associated with an increase in the intensity of a micrococcal nuclease hypersensitive site at M26 and in the 5' noncoding region of the ade6-M26 locus, indicating that chromatin structure becomes more open at a time when the hotspot is active (MIZUNO et al. 1997 ). A similar micrococcal nuclease hypersensitive site was observed at the site of the 3013 hotspot; as with M26, the chromatin structure at this sequence became more "open" as cells entered meiosis (Fig 3). It is possible that local opening of the chromatin at hotspot sequences increases accessibility for binding of proteins required for recombination, including Atf1·Pcr1. Alternatively, Atf1·Pcr1 constitutively bound to hotspot sequences may promote local chromatin remodeling in response to meiotic induction. Meiosis-specific recombination factors may be recruited to such open chromatin regions. Although it is not clear how sequences such as M26 and 3013 act to increase recombinant frequency, it is likely that they act in a similar manner, since both require the Atf1·Pcr1 transcription factor and the Wis1 and Spc1 kinases, and they induce similar specific changes to local chromatin structure.

The elucidation of a new family of meiotic recombination hotspots in S. pombe will facilitate further studies on the mechanism of meiotic recombination hotspots and the role they play in determining the frequency and distribution of recombination during meiosis.

	ACKNOWLEDGMENTS

We are grateful to Paul Russell, Kazuhiro Shiozaki, and Jürg Kohli for providing strains and for sharing unpublished data and to Joe Calvo for helpful discussion. We thank Marcella Cervantes, Liz Greene, Randy Schreckhise, and Jennifer Young for technical assistance and Sue Amundsen, Jirair Bedoyan, Luther Davis, Joe Farah, Randy Schreckhise, Walt Steiner, Andrew Taylor, and Jennifer Young for helpful discussions and comments on the manuscript. This work was supported by grant GM31693 from the National Institutes of Health to G.R.S.; by grants from the Human Frontier Science Program and the Ministry of Education, Science, Culture, and Sports (Japan) to K.O.; and by grants from the Biodesign Research Program of RIKEN and CREST of Japan Science and Technology to Takehiko Shibata.

Manuscript received March 28, 2000; Accepted for publication May 15, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BEACH, D., M. PIPER, and P. NURSE, 1982  Construction of a Schizosaccharomyces pombe gene bank in a yeast bacterial shuttle vector and its use to isolate genes by complementation. Mol. Gen. Genet. 187:326-329[Medline].

BENBROOK, D. M. and N. C. JONES, 1994  Different binding specificities and transactivation of variant CRE's by CREB complexes. Nucleic Acids Res. 22:1463-1469[Abstract/Free Full Text].

CUI, Y., Q. WANG, G. D. STORMO, and J. M. CALVO, 1995  A consensus sequence for binding of Lrp to DNA. J. Bacteriol. 177:4872-4880[Abstract/Free Full Text].

DEGOLS, G., K. SHIOZAKI, and P. RUSSELL, 1996  Activation and regulation of the Spc1 stress-activated protein kinase in Schizosaccharomyces pombe.. Mol. Cell. Biol. 16:2870-2877[Abstract].

DEVEAUX, L. C., N. A. HOAGLAND, and G. R. SMITH, 1992  Seventeen complementation groups of mutations decreasing meiotic recombination in Schizosaccharomyces pombe.. Genetics 130:251-262[Abstract].

FOX, M. E., and G. R. SMITH, 1998 Control of meiotic recombination in Schizosaccharomyces pombe, pp. 345–378 in Progress in Nucleic Acid Research and Molecular Biology, edited by K. MOLDAVE. Academic Press, New York.

FOX, M. E., J. B. VIRGIN, J. METZGER, and G. R. SMITH, 1997  Position- and orientation-independent activity of the Schizosaccharomyces pombe meiotic recombination hot spot M26.. Proc. Natl. Acad. Sci. USA 94:7446-7451[Abstract/Free Full Text].

GRIMM, C., P. SCHAER, P. MUNZ, and J. KOHLI, 1991  The strong ADH1 promoter stimulates mitotic and meiotic recombination at the ADE6 gene of Schizosaccharomyces pombe.. Mol. Cell. Biol. 11:289-298[Abstract/Free Full Text].

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

GUTZ, H., H. HESLOT, U. LEUPOLD and N. LOPRIENO, 1974 Schizosaccharomyces pombe, pp. 395–446 in Handbook of Genetics, edited by R. C. KING. Plenum Press, New York.

ITO, H., Y. FUKUDA, K. MURATA, and A. KIMURA, 1983  Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153:163-168[Abstract/Free Full Text].

JONES, R. H. and N. C. JONES, 1989  Mammalian cAMP-responsive element can activate transcription in yeast and binds a yeast factor(s) that resembles the mammalian transcription factor ATF. Proc. Natl. Acad. Sci. USA 86:2176-2180[Abstract/Free Full Text].

KANOH, J., Y. WATANABE, M. OHSUGI, Y. IINO, and M. YAMAMOTO, 1996  Schizosaccharomyces pombe gad7⁺ encodes a phosphoprotein with a bZIP domain, which is required for proper G1 arrest and gene expression under nitrogen starvation. Genes Cells 1:391-408[Abstract].

KON, N., M. D. KRAWCHUK, B. G. WARREN, G. R. SMITH, and W. P. WAHLS, 1997  Transcription factor Mts1/Mts2 (Atf1/Pcr1, Gad7/Pcr1) activates the M26 meiotic recombination hotspot in Schizosaccharomyces pombe.. Proc. Natl. Acad. Sci. USA 94:13756-13770.

KON, N., S. C. SCHROEDER, M. D. KRAWCHUK, and W. P. WAHLS, 1998  Regulation of the Mts1-Mts2-dependent ade6-M26 meiotic recombination hot spot and developmental decisions by the Spc1 mitogen-activated protein kinase of fission yeast. Mol. Cell. Biol. 18:7575-7583[Abstract/Free Full Text].

LICHTEN, M. and A. S. H. GOLDMAN, 1995  Meiotic recombination hotspots. Annu. Rev. Genet. 29:423-444[Medline].

MIZUNO, K.-I., Y. EMURA, M. BAUR, J. KOHLI, and K. OHTA et al., 1997  The meiotic recombination hot spot created by the single-base substitution ade6-M26 results in remodeling of chromatin structure in fission yeast. Genes Dev. 11:876-886[Abstract/Free Full Text].

MONTMINY, M. R., K. A. SEVARINO, J. A. WAGNER, G. MANDEL, and R. H. GOODMAN, 1986  Identification of a cyclic-AMP-responsive element within the rat somatostatin gene. Proc. Natl. Acad. Sci. USA 83:6682-6686[Abstract/Free Full Text].

NURSE, P., 1975  Genetic control of cell size at cell division in yeast. Nature 256:547-551[Medline].

PONTICELLI, A. S. and G. R. SMITH, 1992  Context dependence of a eukaryotic recombination hotspot. Proc. Natl. Acad. Sci. USA 89:227-231[Abstract/Free Full Text].

PONTICELLI, A. S., E. P. SENA, and G. R. SMITH, 1988  Genetic and physical analysis of the M26 recombination hotspot of Schizosaccharomyces pombe.. Genetics 119:491-497[Abstract/Free Full Text].

SCHUCHERT, P. and J. KOHLI, 1988  The ade6-M26 mutation of Schizosaccharomyces pombe increases the frequency of crossing over. Genetics 119:507-515[Abstract/Free Full Text].

SCHUCHERT, P., M. LANGSFORD, E. KÄSLIN, and J. KOHLI, 1991  A specific DNA sequence is required for high frequency of recombination in the ade6 gene of fission yeast. EMBO J. 10:2157-2163[Medline].

SHIEH, J.-C., M. G. WILKINSON, V. BUCK, B. A. MORGAN, and K. MAKINO et al., 1997  The Mcs4 response regulator coordinately controls the stress-activated Wak1-Wis1-Sty1 MAP kinase pathway and fission yeast cell cycle. Genes Dev. 11:1008-1022[Abstract/Free Full Text].

SHIOZAKI, K. and P. RUSSELL, 1995  Cell-cycle control linked to extracellular environment by MAP kinase pathway in fission yeast. Nature 378:739-743[Medline].

SHIOZAKI, K. and P. RUSSELL, 1996  Conjugation, meiosis, and the osmotic stress response are regulated by Spc1 kinase through Atf1 transcription factor in fission yeast. Genes Dev. 10:2276-2288[Abstract/Free Full Text].

SMITH, G. R., 1994  Hotspots of homologous recombination. Experientia 50:234-241[Medline].

STETTLER, S., E. WARBRICK, S. PROCHNIK, S. MACKIE, and P. FANTES, 1996  The Wis1 signal transduction pathway is required for expression of cAMP-repressed genes in fission yeast. J. Cell Sci. 109:1927-1935[Abstract].

SZANKASI, P., W. D. HEYER, P. SCHUCHERT, and J. KOHLI, 1988  DNA sequence analysis of the ade6 gene of Schizosaccharomyces pombe: wild-type and mutant alleles including the recombination hotspot allele ade6-M26.. J. Mol. Biol. 204:917-925[Medline].

TAKEDA, T., T. TODA, K. KOMINAMI, A. KOHNOSU, and M. YANAGIDA et al., 1995  Schizosaccharomyces pombe atf1⁺ encodes a transcription factor required for sexual development and entry into stationary phase. EMBO J. 14:6193-6208[Medline].

TUERK, C. and L. GOLD, 1990  Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249:505-510[Abstract/Free Full Text].

VIRGIN, J. B., J. METZGER, and G. R. SMITH, 1995  Active and inactive transplacement of the M26 recombination hotspot in Schizosaccharomyces pombe.. Genetics 141:33-48[Abstract].

WAHLS, W. P. and G. R. SMITH, 1994  A heteromeric protein that binds to a meiotic homologous recombination hotspot: correlation of binding and hotspot activity. Genes Dev. 8:1693-1702[Abstract/Free Full Text].

WATANABE, Y. and M. YAMAMOTO, 1996  Schizosaccharomyces pombe pcr1⁺ encodes a CREB/ATF protein involved in regulation of gene expression for sexual development. Mol. Cell. Biol. 16:704-711[Abstract].

WILKINSON, M. G., M. SAMUELS, T. TAKEDA, W. M. TOONE, and J.-C. SHIEH et al., 1996  The Atf1 transcription factor is a target for the Sty1 stress-activated MAP kinase pathway in fission yeast. Genes Dev. 10:2289-2301[Abstract/Free Full Text].

ZAHN-ZABAL, M., E. LEHMANN, and J. KOHLI, 1995  Hot spots of recombination in fission yeast: inactivation of the M26 hot spot by deletion of the ade6 promoter and the novel hotspot ura4-aim.. Genetics 140:469-478[Abstract].

This article has been cited by other articles:

				W. Reiter, S. Watt, K. Dawson, C. L. Lawrence, J. Bahler, N. Jones, and C. R. M. Wilkinson Fission Yeast MAP Kinase Sty1 Is Recruited to Stress-induced Genes J. Biol. Chem., April 11, 2008; 283(15): 9945 - 9956. [Abstract] [Full Text] [PDF]

				P. Calabrese A population genetics model with recombination hotspots that are heterogeneous across the population PNAS, March 13, 2007; 104(11): 4748 - 4752. [Abstract] [Full Text] [PDF]

				W. W. Steiner and G. R. Smith Natural Meiotic Recombination Hot Spots in the Schizosaccharomyces pombe Genome Successfully Predicted from the Simple Sequence Motif M26 Mol. Cell. Biol., October 15, 2005; 25(20): 9054 - 9062. [Abstract] [Full Text] [PDF]

				D. W. Pryce, A. Lorenz, J. B. Smirnova, J. Loidl, and R. J. McFarlane Differential Activation of M26-Containing Meiotic Recombination Hot Spots in Schizosaccharomyces pombe Genetics, May 1, 2005; 170(1): 95 - 106. [Abstract] [Full Text] [PDF]

				M. Pineda-Krch and R. J. Redfield Persistence and Loss of Meiotic Recombination Hotspots Genetics, April 1, 2005; 169(4): 2319 - 2333. [Abstract] [Full Text] [PDF]

				W. W. Steiner and G. R. Smith Optimizing the Nucleotide Sequence of a Meiotic Recombination Hotspot in Schizosaccharomyces pombe Genetics, April 1, 2005; 169(4): 1973 - 1983. [Abstract] [Full Text] [PDF]

				J. A. Farah, G. Cromie, W. W. Steiner, and G. R. Smith A Novel Recombination Pathway Initiated by the Mre11/Rad50/Nbs1 Complex Eliminates Palindromes During Meiosis in Schizosaccharomyces pombe Genetics, March 1, 2005; 169(3): 1261 - 1274. [Abstract] [Full Text] [PDF]

				N. A. Rana, N. D. Ebenezer, A. R. Webster, A. R. Linares, D. B. Whitehouse, S. Povey, and A. J. Hardcastle Recombination hotspots and block structure of linkage disequilibrium in the human genome exemplified by detailed analysis of PGM1 on 1p31 Hum. Mol. Genet., December 15, 2004; 13(24): 3089 - 3102. [Abstract] [Full Text] [PDF]

				M. K. Davidson, H. K. Shandilya, K. Hirota, K. Ohta, and W. P. Wahls Atf1-Pcr1-M26 Complex Links Stress-activated MAPK and cAMP-dependent Protein Kinase Pathways via Chromatin Remodeling of cgs2+ J. Biol. Chem., December 3, 2004; 279(49): 50857 - 50863. [Abstract] [Full Text] [PDF]