- Email: [email protected]
Schizosaccharomyces pombe apn1 encodes a homologue of the Escherichia coli endonuclease IV family of DNA repair proteins
Biochimica et Biophysica Acta 1396 Ž1998. 15–20
Short sequence-paper
Schizosaccharomyces pombe apn1 encodes a homologue of the Escherichia coli endonuclease IV family of DNA repair proteins Dindial Ramotar ) , Julie Vadnais, Jean-Yves Masson, Stephane Tremblay MaisonneuÕe-Rosemont Hospital, Research Center, 5415 Boul. de L’ Assomption, Montreal, Que., Canada H1T 2M4 Received 11 July 1997; revised 18 August 1997; accepted 22 August 1997
Abstract The Apn1 protein of the budding yeast Saccharomyces cereÕisiae is a DNA repair enzyme that hydrolyzes apurinicrapyrimidinic ŽAP. sites and removes 3X-blocking groups present at single strand breaks of damaged DNA. Yeast cells lacking Apn1 are hypersensitive to DNA damaging agents that produce AP sites and DNA strand breaks with blocked 3X-termini. In this study, we showed that the fission yeast Schizosaccharomyces pombe bears a homologue, Spapn1, that is 45% identical to S. cereÕisiae Apn1. However, the Spapn1 gene is apparently not expressed. Active expression of S. cereÕisiae Apn1 in S. pombe conferred no additional resistance to DNA damaging agents. These data suggest that the pathway by which S. pombe repairs AP sites is independent of a functional Apn1-like AP endonuclease. q 1998 Elsevier Science B.V. Keywords: AP endonuclease; DNA repair; Yeast
Reactive oxygen species such as superoxide, hydrogen peroxide, and hydroxyl radical are produced by a multitude of sources that include ionizing radiation, chemical oxidants, and aerobic metabolism w1,2x. These active oxygen species, particularly the hydroxyl radical w2x, damage the cellular DNA by producing a plethora of oxidative DNA lesions, for example, apurinicrapyrimidinic ŽAP. sites and DNA strand breaks bearing blocked 3X-termini, such as 3X-phosphate and 3X-phosphoglycolate w1,3,4x. Among these lesions, AP sites are known targets for the production of mutations w5–7x. AP sites are removed from damaged DNA by AP endonucleases, which
)
Corresponding author: Fax: q1 Ž514. 252-3430; E-mail: [email protected]
belong to two distinct families, Exo III and Endo IV w8–12x. The Exo III family members are Mg 2q-dependent enzymes and these include Escherichia coli exonuclease III, Drosophila Rrp1, mouse Apex, human AperHap1rRef1, and plant Arp1 w11–13x. There is no Exo III homolog in S. cereÕisiae, but this organism contains an enzyme, Pde1, with similar enzymatic properties as exonuclease III w13,14x. Cells that lack the Exo III member are unable to repair damage to their chromosomal DNA and this may account for the early embryonic lethality observed in the apex -r- mouse null mutant w4,15x. There are fewer members in the Endo IV family and these include E. coli endonuclease IV, S. cereÕisiae Apn1, S. pombe Spapn1, and Caenorhabditis elegans CeApn1 w13x. Only endonulease IV and Apn1 are functionally well characterized w13x. The AP endonucleases of both families make an incision on the 5X side of abasic
0167-4781r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 7 - 4 7 8 1 Ž 9 7 . 0 0 1 6 0 - 7
16
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
sites to yield free 3X hydroxyl groups for DNA repair synthesis by DNA polymerases w8–11x. S. cereÕisiae apn1 deletion mutants are hypersensitive to agents that produce AP sites, such as methyl methane sulfonate Ž MMS. w9x. In addition, apn1 deletion mutants accumulate mutations spontaneously at a high rate w9x, which are primarily due to unrepaired AP sites w6,16x. Thus, Apn1 play a critical role by repairing endogenously produced DNA lesions in order to avoid genetic variation. A homologue of Apn1, endonuclease IV Žendo IV. , also exists in E. coli w17x. Both Apn1 and endo IV are structurally and functionally related. This is supported by the finding that when Apn1 is expressed in E. coli, it can functionally substitute for endo IV in the repair of damaged DNA in vivo w18x. Likewise, endonuclease IV can functionally replace Apn1 in the repair of damaged DNA in vivo w19x. These findings prompt us to examine if Apn1rendo IV homologue is present and functionally conserved in higher eukaryotic cells. To initiate this study, we chose the fission yeast S. pombe for many reasons, particularly, since it has biological processes that are conserved between both S. cereÕisiae and human cells w20x. We first tested if S. pombe contains an AP endonuclease activity that could act on the synthetic substrate used for monitoring Apn1 and endo IV w21x. Rather surprisingly, no detectable AP endonuclease activity was found in extracts prepared from three different S. pombe wild-type strains SP720, FW9, and Q353 ŽTable 1.. In parallel experiments, Apn1 and endo IV were readily detected in their respective extracts Ž Table 1. . Addition of various cofactors such as metal ions and ATP, either singly or in combinations, to the S. pombe crude extracts failed to reveal any AP endonuclease activity. Other conditions such as changing pH, temperature, NaCl concentration and buffer conditions also proved unsuccessful. Extracts prepared from S. pombe at various stages in exponential growth or starved cells also contained no AP endonuclease activity. Moreover, extracts derived from cells treated with either H 2 O 2 Ž1 mM. or MMS Ž 0.1%. for 30 min also showed no AP endonuclease activity. Our observation was confirmed by two independent laboratories ŽDrs. M. Sander, NIEHS and Y. Kow, Emory, GA, USA. that used oligonucleotide substrates that bear a single AP site w22x. Addition of fix amounts Ž10 Units. of either purified Apn1 or endo IV to
Table 1 Level of AP endonuclease activity in crude extracts derived from S. cereÕisiae, E. coli, and S. pombe Strains S. cereÕisiae DBY747 Žwild-type. DRY370 Ž apn1D . DRY370rpAPN1 DRY370rpExoIII E. coli AB1157 Žwild-type. BW528 Ž xth -; nfo - . S. pombe SP720 Žwild-type. DRP721 Ž Spapn1D . SP972 Q353 FW9 SP720qpurified Apn1 SP720qpurified endo IV SP720rpAPN1 SP720rpExoIII
AP endonuclease ŽUnitsrmg protein. 60 -1 660 430 12 -1 -1 -1 -1 -1 -1 10 10 560 470
Crude extracts were assayed for AP endonuclease activity using a synthetic substrate w17x. The values represent the average of three independent crude extract determinations.
crude S. pombe extracts resulted in 100% detection of the exogenously added enzymes Ž Table 1. . This result suggests that the crude extract did not interfere with the detection of AP endonuclease activity. In addition, no degradation of either purified S. cereÕisiae Apn1 or E. coli endo IV was observed in the crude S. pombe extract. However, these data do not exclude the possibility that S. pombe AP endonuclease is more labile to extraction. We decided to examine more thoroughly whether an Apn1rendo IV-homologue exist in S. pombe. Two degenerated pimers were made against two of the highly conserved regions ŽENTrMAG and GIPLVLET. shared between E. coli endo IV and S. cereÕisiae Apn1 w17x. The primers were used with template DNA, prepared directly from whole cells of S. pombe, in a polymerase chain reaction w23x. The primers amplified a 430 bp fragment from S. pombe genomic DNA. This fragment was used as a 32 Plabeled probe to screen a S. pombe genomic library constructed on a cosmid Ž generously provided by Dr. Hoheisel, ICRF, UK. w24x. Two overlapping clones were isolated that were derived from chromosome III of S. pombe. Both clones were located ; 200 kb
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
from the ade6 gene. Further analysis revealed that the 430 bp fragment hybridized specifically to a 800 bp BamHIrHindIII DNA fragment contained in the cosmid clones. This 800 bp fragment was used as a probe to isolate the corresponding cDNA from a S. pombe cDNA library constructed in lZAPII phage Žgenerously provided by Dr. David Beach, Cold Spring Harbor, NY.. The cDNA was rescued as a Amp plasmid using the Sol bacterial strain Ž Stratagene, USA.. This plasmid carried an insert 2.8 kb, which contained three open reading frames Ž ORF., ORF1, 2, and 3, determined by DNA sequence analysis w25x. While ORF1 and ORF2 were completed, ORF3 was incomplete. We showed that ORF1 encodes a protein that is highly conserved in nature and belongs to the family of QM transcription factors w26x. ORF2 is predicted to encode a protein with 324 amino acid residues ŽFig. 1., which shared 40.5%, 45%, and 42% identity, respectively, to E. coli endonuclease IV, S. cereÕisiae Apn1, and a recently isolated homologue, CeApn1, from the nematode Caenorhabditis elegans ŽFig. 2. w27x. We referred to ORF2 as Spapn1. These four proteins constitute the Endo IV family of AP endonucleases and they share five highly conserved regions ŽI, II, III, IV, V., except for the Spapn1 protein which has an extra 24 amino acid inserted in region I Ž Fig. 2. . Since no measurable AP endonuclease activity was detected in S. pombe crude extracts, we examine whether the Spapn1 gene is actually expressed. Using
17
the coding region of the Spapn1 gene as a probe, we were unable to detect its message by Northern blot analysis Ždata not shown. . In control experiments, the ura4 gene mRNA was easily detected Ždata not shown.. Using another approach involving reverse transcriptase also did not detect the S. pombe apn1 mRNA. In E. coli, but not S. cereÕisiae, the endonuclease IV gene is strongly induced by superoxide generating agents, such as paraquat and menadione w28x. However, such agents also did not reveal any expression of the Spapn1 gene. Altogether, these data suggest that either the Spapn1 mRNA is weakly expressed and therefore, cannot be detected by the above approaches or the gene is not expressed. Among the known Endo IV family members, S. pombe is the only member that has a 24 amino acid insertion in one of the highly conserved regions Ž Fig. 2.. Careful analysis revealed that the 24 amino acid lies within a highly conserved intron sequence Ž Fig. 3.. Out of 73 S. pombe introns examined w29x, Spapn1 DNA sequence contained the second most abundant 5X-splice site sequence Ž GTATGT. , the most abundant branch ŽCTAAC. , and 3X-splice site sequences ŽTAG. w29x. The percentage of pyrimidine is 80% from the branch to the 3X-splice sequence, consistent with that present in other S. pombe introns w29x. Moreover, the Spapn1 intron is 72 nucleotides in length which is within the range of 40–80 nucleotides present in 77% of known S. pombe genes bearing introns w29x. Collectively, these similarities
Fig. 1. Nucleotide and deduced amino acid sequence of the Spapn1 gene. The Genebank accession number for Spapn1 is U33635.
18
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
Fig. 2. Comparison of the predicted amino acid sequence of four members of the Endo IV family. The five highly conserved regions ŽI, II, III, IV, and V. are boxed.
suggest that Spapn1 bears a highly conserved intron, and if removed, will produce an identical conserved Region I as that of S. cereÕisiae Apn1 or E. coli endonuclease IV. It appears then that the Spapn1 cDNA we isolated is an artefact of reverse transcriptase copying contaminating genomic DNA during the cDNA library preparation. This is supported by the fact that when the Spapn1 gene was isolated by PCR
from the chromosomal DNA of the wild-type strain SP720 or Q353, it also contained the same intron fragment. Nonetheless, studies are in progress to delete the intron and to test if the resulting Spapn1 gene expresses AP endonuclease activity when place next to a heterologous promoter, for example, the ADH promoter. We have not yet explored the precise molecular defect that prevents expression of Spapn1,
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
Fig. 3. Illustration of the portion of the Spapn1 gene containing a conserved intron sequence. If Spapn1 gene is expressed, then accurate splicing of its message is expected to yield the conserved region I.
but the presumptive promoter region from y625 to q1 nucleotide will be tested for its ability to direct expression of the reporter gene lacZ encoding bgalactosidase. Because the Spapn1 gene is apparently not expressed, we anticipated that cross-specie expression of an AP endonuclease could confer additional drug resistance to S. pombe. We therefore construct the S. cereÕisiae APN1 gene such that it was under the control of the ADH promoter in the vector pART1 Žgenerously provided by Dr. Howard Lieberman, Columbia University, NY, USA. , which is capable of directing protein expression in either S. cereÕisiae or S. pombe. The resulting plasmid, pAPN1 was introduced into both S. cereÕisiae and S. pombe w30,31x. Crude extract derived from S. pombe bearing the pART1 vector contained no measurable AP endonuclease activity ŽTable 1.. In contrast, the plasmid pAPN1 directed the expression of a substantial amount of AP endonuclease in S. pombe ŽTable 1.. Similar level of activity was also found in crude extracts derived from S. cereÕisiae apn1 deletion mutant harboring pAPN1. Western blot analyses confirmed that pAPN1 directed the expression of the native 40 kDa Apn1 polypeptide, which was detected by anti-Apn1 antibodies w32x. No such polypeptide was expressed by the vector pART alone. Despite the expression of Apn1 in S. pombe, it provided no additional MMS resistance to S. pombe ŽFig. 4, bar 6. as measured by gradient plate assay w32x. In this assay, cells that are sensitive to the drug grow only a short distance along the gradient, whereas resistant cells will grow further along the gradient. As expected pAPN1 conferred full wild-type resistance to MMS to the S. cereÕisiae apn1 deletion mutant ŽFig. 4, bar 3. w32x. A remote possibility, however, is that S. cereÕisiae Apn1 does not enter the S. pombe
19
nucleus perhaps because of its larger size or its unique nuclear localization signal that is not recognized by S. pombe nuclear localization machinery w19,32x. To circumvent this problem, a smaller AP endonuclease Ž30 kDa. of bacterial origin, exonuclease III, was expressed in both organisms. While, pExoIII conferred MMS resistance to S. cereÕisiae ŽFig. 4, bar 4., it conferred no MMS resistance to S. pombe ŽFig. 4, bar 7. . The data suggests that AP endonuclease may not be the critical enzyme that repairs abasic sites caused by MMS in S. pombe chromosomal DNA. Consistent with this interpretation is that deletion of the Spapn1 gene resulted in the S. pombe mutant strain DRP720 that showed no sensitivity to MMS Ž Fig. 4, bar 8. . From the above data, it would appear that S. pombe uses an alternative pathway to repair AP sites. Such pathway could use an AP lyase instead to nick the AP sites. This is a likely pathway since an AP lyase, Spnth, has been characterized from S. pombe
Fig. 4. MMS-resistance of S. cereÕisiae and S. pombe strains. Bars 1–4 indicate the following S. cereÕisiae strains, respectively, DBY747rpART1; DRY370 deleted for the APN1 gene and bearing pART1; DRY370rpAPN1; and DRY370rpEXOIII. Bars 5–8 indicate the following S. pombe strains, respectively, SP720 Žwild-type.rpART1; SP720rpAPN1; SP720rpEXOIII; and DRP721 was derived from SP720, except deleted for the Spapn1 gene. Gradient plate assay was done according to Ramotar et al. w32x. Growth all along the gradient is taken to be 100%. The bottom layer of the gradient contained 0.03% MMS.
20
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
w33x. Spnth is structurally and enzymatically related to E. coli endonuclease III, which has both DNA glycosylase and AP lyase activities w33x. Moreover, we recently demonstrated that expression of endonuclease III in S. cereÕisiae can functionally replace the biological role of Apn1 in the repair of AP sites w16x. The fact that S. pombe contains an endonuclease III homologue strongly suggests that this organism could use this enzyme to also repair AP sites. If so, one would expect that S. pombe nth deletion mutants would be hypersensitive to agents that produce AP sites, but that remains to be determined. This work was supported by a grant to D.R. from the Natural Science and Engineering Research Council of Canada. J.-Y.M. received a graduate student fellowship from the Fonds pour la formation de Chercheurs et d’Aide a´ la Recherche du Quebec, and D.R. is a Career Scientist of the National Cancer Institute of Canada.
References w1x C. von Sonntag, The Chemical Basis of Radiation Biology, Taylor and Francis, London, 1987. w2x J.A. Imlay, S. Linn, Science 240 Ž1988. 1302–1309. w3x F. Hutchison, Prog. Nucleic Acids Res. Mol. Biol. 32 Ž1985. 115–154. w4x B. Demple, A.W. Johnson, D. Fung, Proc. Natl. Acad. Sci. U.S.A. 83 Ž1986. 7731–7735. w5x L.A. Loeb, B.D. Preston, Annu. Rev. Genet. 20 Ž1986. 201–230. w6x B.A. Kunz, E.S. Henson, H. Roche, D. Ramotar, T. Nunoshiba, B. Demple, Proc. Natl. Acad. Sci. U.S.A. 91 Ž1994. 8165–8169. w7x P.E. Gibbs, C.W. Lawrence, J. Mol. Biol. 251 Ž1995. 229–236. w8x P. Doetsch, R.P. Cunningham, Mutat. Res. 236 Ž1990. 173–201.
w9x D. Ramotar, S.C. Popoff, E.B. Gralla, B. Demple, Mol. Cell. Biol. 11 Ž1991. 4537–4544. w10x T. Lindahl, Nature 362 Ž1993. 709–715. w11x B. Demple, L. Harrison, Annu. Rev. Biochem. 63 Ž1994. 915–948. w12x G. Barzilay, I.D. Hickson, BioEssays 17 Ž1995. 713–719. w13x D. Ramotar, Biochem. Cell Biol. Ž1997. in press. w14x M. Sander, D. Ramotar, Biochemistry 36 Ž1997. 6100–6106. w15x S. Xanthoudakis, R.J. Smeyne, J. D Wallace, T. Curran, Proc. Natl. Acad. Sci. U.S.A. 93 Ž1996. 8919–8923. w16x J.-Y. Masson, D. Ramotar, Mol. Microbiol. 24 Ž1997. 711– 721. w17x S.C. Popoff, A.S. Spira, A.W. Johnson, B. Demple, Proc. Natl. Acad. Sci. U.S.A. 87 Ž1990. 4193–4197. w18x D. Ramotar, S.C. Popoff, B. Demple, Mol. Microbiol. 5 Ž1991. 149–155. w19x D. Ramotar, B. Demple, J. Biol. Chem. 271 Ž1996. 7368– 7374. w20x J. Hayles, P. Nurse, Genetics of the fission yeast Schizosaccharomyces pombe, Annu. Rev. Genet. 26 Ž1992. 373–402. w21x J.D. Levin, B. Demple, Nucleic Acids. Res. 18 Ž1990. 5069–5075. w22x R.K. Singhal, R. Prasad, S.H. Wilson, J. Biol. Chem. 270 Ž1995. 949–957. w23x D. Ramotar, Mol. Cell. Biochem. 145 Ž1995. 185–187. w24x J.D. Hoheisel, E. Maier, R. Mott, L. McCarthy, A.V. Grigoriev, L.C. Schalkwyk, D. Nizetic, F. Francis, H. Lehrach, Cell 73 Ž1993. 109–120. w25x F. Sanger, S. Nicklen, A.R. Coulson, Proc. Natl. Acad. Sci. U.S.A. 74 Ž1977. 5463–5467. w26x J-Y. Masson, J. Vadnais, D. Ramotar, Gene 170 Ž1996. 153–154. w27x J.-Y., Masson, S. Tremblay, D. Ramotar, Gene Ž1996. in press. w28x E. Chan, B. Weiss, Proc. Natl. Acad. Sci. U.S.A. 84 Ž1987. 3189–3193. w29x G. Prabhala, G.H. Rosenberg, N.F. Kaufer, Yeast 8 Ž1992. 171–182. w30x R.D. Gietz, R.H. Schiestl, Yeast 7 Ž1991. 253–263. w31x D. Warshawsky, L. Miller, Biotechniques 16 Ž1994. 798. w32x D. Ramotar, R. Lillis, C. Kim, B. Demple, J. Biol. Chem. 268 Ž1993. 20533–20539. w33x T. Roldan-Arjona, C. Anselmino, T. Lindahl, Nucleic Acids Res. 24 Ž1996. 3307–3312.
Short sequence-paper
Schizosaccharomyces pombe apn1 encodes a homologue of the Escherichia coli endonuclease IV family of DNA repair proteins Dindial Ramotar ) , Julie Vadnais, Jean-Yves Masson, Stephane Tremblay MaisonneuÕe-Rosemont Hospital, Research Center, 5415 Boul. de L’ Assomption, Montreal, Que., Canada H1T 2M4 Received 11 July 1997; revised 18 August 1997; accepted 22 August 1997
Abstract The Apn1 protein of the budding yeast Saccharomyces cereÕisiae is a DNA repair enzyme that hydrolyzes apurinicrapyrimidinic ŽAP. sites and removes 3X-blocking groups present at single strand breaks of damaged DNA. Yeast cells lacking Apn1 are hypersensitive to DNA damaging agents that produce AP sites and DNA strand breaks with blocked 3X-termini. In this study, we showed that the fission yeast Schizosaccharomyces pombe bears a homologue, Spapn1, that is 45% identical to S. cereÕisiae Apn1. However, the Spapn1 gene is apparently not expressed. Active expression of S. cereÕisiae Apn1 in S. pombe conferred no additional resistance to DNA damaging agents. These data suggest that the pathway by which S. pombe repairs AP sites is independent of a functional Apn1-like AP endonuclease. q 1998 Elsevier Science B.V. Keywords: AP endonuclease; DNA repair; Yeast
Reactive oxygen species such as superoxide, hydrogen peroxide, and hydroxyl radical are produced by a multitude of sources that include ionizing radiation, chemical oxidants, and aerobic metabolism w1,2x. These active oxygen species, particularly the hydroxyl radical w2x, damage the cellular DNA by producing a plethora of oxidative DNA lesions, for example, apurinicrapyrimidinic ŽAP. sites and DNA strand breaks bearing blocked 3X-termini, such as 3X-phosphate and 3X-phosphoglycolate w1,3,4x. Among these lesions, AP sites are known targets for the production of mutations w5–7x. AP sites are removed from damaged DNA by AP endonucleases, which
)
Corresponding author: Fax: q1 Ž514. 252-3430; E-mail: [email protected]
belong to two distinct families, Exo III and Endo IV w8–12x. The Exo III family members are Mg 2q-dependent enzymes and these include Escherichia coli exonuclease III, Drosophila Rrp1, mouse Apex, human AperHap1rRef1, and plant Arp1 w11–13x. There is no Exo III homolog in S. cereÕisiae, but this organism contains an enzyme, Pde1, with similar enzymatic properties as exonuclease III w13,14x. Cells that lack the Exo III member are unable to repair damage to their chromosomal DNA and this may account for the early embryonic lethality observed in the apex -r- mouse null mutant w4,15x. There are fewer members in the Endo IV family and these include E. coli endonuclease IV, S. cereÕisiae Apn1, S. pombe Spapn1, and Caenorhabditis elegans CeApn1 w13x. Only endonulease IV and Apn1 are functionally well characterized w13x. The AP endonucleases of both families make an incision on the 5X side of abasic
0167-4781r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 7 - 4 7 8 1 Ž 9 7 . 0 0 1 6 0 - 7
16
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
sites to yield free 3X hydroxyl groups for DNA repair synthesis by DNA polymerases w8–11x. S. cereÕisiae apn1 deletion mutants are hypersensitive to agents that produce AP sites, such as methyl methane sulfonate Ž MMS. w9x. In addition, apn1 deletion mutants accumulate mutations spontaneously at a high rate w9x, which are primarily due to unrepaired AP sites w6,16x. Thus, Apn1 play a critical role by repairing endogenously produced DNA lesions in order to avoid genetic variation. A homologue of Apn1, endonuclease IV Žendo IV. , also exists in E. coli w17x. Both Apn1 and endo IV are structurally and functionally related. This is supported by the finding that when Apn1 is expressed in E. coli, it can functionally substitute for endo IV in the repair of damaged DNA in vivo w18x. Likewise, endonuclease IV can functionally replace Apn1 in the repair of damaged DNA in vivo w19x. These findings prompt us to examine if Apn1rendo IV homologue is present and functionally conserved in higher eukaryotic cells. To initiate this study, we chose the fission yeast S. pombe for many reasons, particularly, since it has biological processes that are conserved between both S. cereÕisiae and human cells w20x. We first tested if S. pombe contains an AP endonuclease activity that could act on the synthetic substrate used for monitoring Apn1 and endo IV w21x. Rather surprisingly, no detectable AP endonuclease activity was found in extracts prepared from three different S. pombe wild-type strains SP720, FW9, and Q353 ŽTable 1.. In parallel experiments, Apn1 and endo IV were readily detected in their respective extracts Ž Table 1. . Addition of various cofactors such as metal ions and ATP, either singly or in combinations, to the S. pombe crude extracts failed to reveal any AP endonuclease activity. Other conditions such as changing pH, temperature, NaCl concentration and buffer conditions also proved unsuccessful. Extracts prepared from S. pombe at various stages in exponential growth or starved cells also contained no AP endonuclease activity. Moreover, extracts derived from cells treated with either H 2 O 2 Ž1 mM. or MMS Ž 0.1%. for 30 min also showed no AP endonuclease activity. Our observation was confirmed by two independent laboratories ŽDrs. M. Sander, NIEHS and Y. Kow, Emory, GA, USA. that used oligonucleotide substrates that bear a single AP site w22x. Addition of fix amounts Ž10 Units. of either purified Apn1 or endo IV to
Table 1 Level of AP endonuclease activity in crude extracts derived from S. cereÕisiae, E. coli, and S. pombe Strains S. cereÕisiae DBY747 Žwild-type. DRY370 Ž apn1D . DRY370rpAPN1 DRY370rpExoIII E. coli AB1157 Žwild-type. BW528 Ž xth -; nfo - . S. pombe SP720 Žwild-type. DRP721 Ž Spapn1D . SP972 Q353 FW9 SP720qpurified Apn1 SP720qpurified endo IV SP720rpAPN1 SP720rpExoIII
AP endonuclease ŽUnitsrmg protein. 60 -1 660 430 12 -1 -1 -1 -1 -1 -1 10 10 560 470
Crude extracts were assayed for AP endonuclease activity using a synthetic substrate w17x. The values represent the average of three independent crude extract determinations.
crude S. pombe extracts resulted in 100% detection of the exogenously added enzymes Ž Table 1. . This result suggests that the crude extract did not interfere with the detection of AP endonuclease activity. In addition, no degradation of either purified S. cereÕisiae Apn1 or E. coli endo IV was observed in the crude S. pombe extract. However, these data do not exclude the possibility that S. pombe AP endonuclease is more labile to extraction. We decided to examine more thoroughly whether an Apn1rendo IV-homologue exist in S. pombe. Two degenerated pimers were made against two of the highly conserved regions ŽENTrMAG and GIPLVLET. shared between E. coli endo IV and S. cereÕisiae Apn1 w17x. The primers were used with template DNA, prepared directly from whole cells of S. pombe, in a polymerase chain reaction w23x. The primers amplified a 430 bp fragment from S. pombe genomic DNA. This fragment was used as a 32 Plabeled probe to screen a S. pombe genomic library constructed on a cosmid Ž generously provided by Dr. Hoheisel, ICRF, UK. w24x. Two overlapping clones were isolated that were derived from chromosome III of S. pombe. Both clones were located ; 200 kb
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
from the ade6 gene. Further analysis revealed that the 430 bp fragment hybridized specifically to a 800 bp BamHIrHindIII DNA fragment contained in the cosmid clones. This 800 bp fragment was used as a probe to isolate the corresponding cDNA from a S. pombe cDNA library constructed in lZAPII phage Žgenerously provided by Dr. David Beach, Cold Spring Harbor, NY.. The cDNA was rescued as a Amp plasmid using the Sol bacterial strain Ž Stratagene, USA.. This plasmid carried an insert 2.8 kb, which contained three open reading frames Ž ORF., ORF1, 2, and 3, determined by DNA sequence analysis w25x. While ORF1 and ORF2 were completed, ORF3 was incomplete. We showed that ORF1 encodes a protein that is highly conserved in nature and belongs to the family of QM transcription factors w26x. ORF2 is predicted to encode a protein with 324 amino acid residues ŽFig. 1., which shared 40.5%, 45%, and 42% identity, respectively, to E. coli endonuclease IV, S. cereÕisiae Apn1, and a recently isolated homologue, CeApn1, from the nematode Caenorhabditis elegans ŽFig. 2. w27x. We referred to ORF2 as Spapn1. These four proteins constitute the Endo IV family of AP endonucleases and they share five highly conserved regions ŽI, II, III, IV, V., except for the Spapn1 protein which has an extra 24 amino acid inserted in region I Ž Fig. 2. . Since no measurable AP endonuclease activity was detected in S. pombe crude extracts, we examine whether the Spapn1 gene is actually expressed. Using
17
the coding region of the Spapn1 gene as a probe, we were unable to detect its message by Northern blot analysis Ždata not shown. . In control experiments, the ura4 gene mRNA was easily detected Ždata not shown.. Using another approach involving reverse transcriptase also did not detect the S. pombe apn1 mRNA. In E. coli, but not S. cereÕisiae, the endonuclease IV gene is strongly induced by superoxide generating agents, such as paraquat and menadione w28x. However, such agents also did not reveal any expression of the Spapn1 gene. Altogether, these data suggest that either the Spapn1 mRNA is weakly expressed and therefore, cannot be detected by the above approaches or the gene is not expressed. Among the known Endo IV family members, S. pombe is the only member that has a 24 amino acid insertion in one of the highly conserved regions Ž Fig. 2.. Careful analysis revealed that the 24 amino acid lies within a highly conserved intron sequence Ž Fig. 3.. Out of 73 S. pombe introns examined w29x, Spapn1 DNA sequence contained the second most abundant 5X-splice site sequence Ž GTATGT. , the most abundant branch ŽCTAAC. , and 3X-splice site sequences ŽTAG. w29x. The percentage of pyrimidine is 80% from the branch to the 3X-splice sequence, consistent with that present in other S. pombe introns w29x. Moreover, the Spapn1 intron is 72 nucleotides in length which is within the range of 40–80 nucleotides present in 77% of known S. pombe genes bearing introns w29x. Collectively, these similarities
Fig. 1. Nucleotide and deduced amino acid sequence of the Spapn1 gene. The Genebank accession number for Spapn1 is U33635.
18
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
Fig. 2. Comparison of the predicted amino acid sequence of four members of the Endo IV family. The five highly conserved regions ŽI, II, III, IV, and V. are boxed.
suggest that Spapn1 bears a highly conserved intron, and if removed, will produce an identical conserved Region I as that of S. cereÕisiae Apn1 or E. coli endonuclease IV. It appears then that the Spapn1 cDNA we isolated is an artefact of reverse transcriptase copying contaminating genomic DNA during the cDNA library preparation. This is supported by the fact that when the Spapn1 gene was isolated by PCR
from the chromosomal DNA of the wild-type strain SP720 or Q353, it also contained the same intron fragment. Nonetheless, studies are in progress to delete the intron and to test if the resulting Spapn1 gene expresses AP endonuclease activity when place next to a heterologous promoter, for example, the ADH promoter. We have not yet explored the precise molecular defect that prevents expression of Spapn1,
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
Fig. 3. Illustration of the portion of the Spapn1 gene containing a conserved intron sequence. If Spapn1 gene is expressed, then accurate splicing of its message is expected to yield the conserved region I.
but the presumptive promoter region from y625 to q1 nucleotide will be tested for its ability to direct expression of the reporter gene lacZ encoding bgalactosidase. Because the Spapn1 gene is apparently not expressed, we anticipated that cross-specie expression of an AP endonuclease could confer additional drug resistance to S. pombe. We therefore construct the S. cereÕisiae APN1 gene such that it was under the control of the ADH promoter in the vector pART1 Žgenerously provided by Dr. Howard Lieberman, Columbia University, NY, USA. , which is capable of directing protein expression in either S. cereÕisiae or S. pombe. The resulting plasmid, pAPN1 was introduced into both S. cereÕisiae and S. pombe w30,31x. Crude extract derived from S. pombe bearing the pART1 vector contained no measurable AP endonuclease activity ŽTable 1.. In contrast, the plasmid pAPN1 directed the expression of a substantial amount of AP endonuclease in S. pombe ŽTable 1.. Similar level of activity was also found in crude extracts derived from S. cereÕisiae apn1 deletion mutant harboring pAPN1. Western blot analyses confirmed that pAPN1 directed the expression of the native 40 kDa Apn1 polypeptide, which was detected by anti-Apn1 antibodies w32x. No such polypeptide was expressed by the vector pART alone. Despite the expression of Apn1 in S. pombe, it provided no additional MMS resistance to S. pombe ŽFig. 4, bar 6. as measured by gradient plate assay w32x. In this assay, cells that are sensitive to the drug grow only a short distance along the gradient, whereas resistant cells will grow further along the gradient. As expected pAPN1 conferred full wild-type resistance to MMS to the S. cereÕisiae apn1 deletion mutant ŽFig. 4, bar 3. w32x. A remote possibility, however, is that S. cereÕisiae Apn1 does not enter the S. pombe
19
nucleus perhaps because of its larger size or its unique nuclear localization signal that is not recognized by S. pombe nuclear localization machinery w19,32x. To circumvent this problem, a smaller AP endonuclease Ž30 kDa. of bacterial origin, exonuclease III, was expressed in both organisms. While, pExoIII conferred MMS resistance to S. cereÕisiae ŽFig. 4, bar 4., it conferred no MMS resistance to S. pombe ŽFig. 4, bar 7. . The data suggests that AP endonuclease may not be the critical enzyme that repairs abasic sites caused by MMS in S. pombe chromosomal DNA. Consistent with this interpretation is that deletion of the Spapn1 gene resulted in the S. pombe mutant strain DRP720 that showed no sensitivity to MMS Ž Fig. 4, bar 8. . From the above data, it would appear that S. pombe uses an alternative pathway to repair AP sites. Such pathway could use an AP lyase instead to nick the AP sites. This is a likely pathway since an AP lyase, Spnth, has been characterized from S. pombe
Fig. 4. MMS-resistance of S. cereÕisiae and S. pombe strains. Bars 1–4 indicate the following S. cereÕisiae strains, respectively, DBY747rpART1; DRY370 deleted for the APN1 gene and bearing pART1; DRY370rpAPN1; and DRY370rpEXOIII. Bars 5–8 indicate the following S. pombe strains, respectively, SP720 Žwild-type.rpART1; SP720rpAPN1; SP720rpEXOIII; and DRP721 was derived from SP720, except deleted for the Spapn1 gene. Gradient plate assay was done according to Ramotar et al. w32x. Growth all along the gradient is taken to be 100%. The bottom layer of the gradient contained 0.03% MMS.
20
D. Ramotar et al.r Biochimica et Biophysica Acta 1396 (1998) 15–20
w33x. Spnth is structurally and enzymatically related to E. coli endonuclease III, which has both DNA glycosylase and AP lyase activities w33x. Moreover, we recently demonstrated that expression of endonuclease III in S. cereÕisiae can functionally replace the biological role of Apn1 in the repair of AP sites w16x. The fact that S. pombe contains an endonuclease III homologue strongly suggests that this organism could use this enzyme to also repair AP sites. If so, one would expect that S. pombe nth deletion mutants would be hypersensitive to agents that produce AP sites, but that remains to be determined. This work was supported by a grant to D.R. from the Natural Science and Engineering Research Council of Canada. J.-Y.M. received a graduate student fellowship from the Fonds pour la formation de Chercheurs et d’Aide a´ la Recherche du Quebec, and D.R. is a Career Scientist of the National Cancer Institute of Canada.
References w1x C. von Sonntag, The Chemical Basis of Radiation Biology, Taylor and Francis, London, 1987. w2x J.A. Imlay, S. Linn, Science 240 Ž1988. 1302–1309. w3x F. Hutchison, Prog. Nucleic Acids Res. Mol. Biol. 32 Ž1985. 115–154. w4x B. Demple, A.W. Johnson, D. Fung, Proc. Natl. Acad. Sci. U.S.A. 83 Ž1986. 7731–7735. w5x L.A. Loeb, B.D. Preston, Annu. Rev. Genet. 20 Ž1986. 201–230. w6x B.A. Kunz, E.S. Henson, H. Roche, D. Ramotar, T. Nunoshiba, B. Demple, Proc. Natl. Acad. Sci. U.S.A. 91 Ž1994. 8165–8169. w7x P.E. Gibbs, C.W. Lawrence, J. Mol. Biol. 251 Ž1995. 229–236. w8x P. Doetsch, R.P. Cunningham, Mutat. Res. 236 Ž1990. 173–201.
w9x D. Ramotar, S.C. Popoff, E.B. Gralla, B. Demple, Mol. Cell. Biol. 11 Ž1991. 4537–4544. w10x T. Lindahl, Nature 362 Ž1993. 709–715. w11x B. Demple, L. Harrison, Annu. Rev. Biochem. 63 Ž1994. 915–948. w12x G. Barzilay, I.D. Hickson, BioEssays 17 Ž1995. 713–719. w13x D. Ramotar, Biochem. Cell Biol. Ž1997. in press. w14x M. Sander, D. Ramotar, Biochemistry 36 Ž1997. 6100–6106. w15x S. Xanthoudakis, R.J. Smeyne, J. D Wallace, T. Curran, Proc. Natl. Acad. Sci. U.S.A. 93 Ž1996. 8919–8923. w16x J.-Y. Masson, D. Ramotar, Mol. Microbiol. 24 Ž1997. 711– 721. w17x S.C. Popoff, A.S. Spira, A.W. Johnson, B. Demple, Proc. Natl. Acad. Sci. U.S.A. 87 Ž1990. 4193–4197. w18x D. Ramotar, S.C. Popoff, B. Demple, Mol. Microbiol. 5 Ž1991. 149–155. w19x D. Ramotar, B. Demple, J. Biol. Chem. 271 Ž1996. 7368– 7374. w20x J. Hayles, P. Nurse, Genetics of the fission yeast Schizosaccharomyces pombe, Annu. Rev. Genet. 26 Ž1992. 373–402. w21x J.D. Levin, B. Demple, Nucleic Acids. Res. 18 Ž1990. 5069–5075. w22x R.K. Singhal, R. Prasad, S.H. Wilson, J. Biol. Chem. 270 Ž1995. 949–957. w23x D. Ramotar, Mol. Cell. Biochem. 145 Ž1995. 185–187. w24x J.D. Hoheisel, E. Maier, R. Mott, L. McCarthy, A.V. Grigoriev, L.C. Schalkwyk, D. Nizetic, F. Francis, H. Lehrach, Cell 73 Ž1993. 109–120. w25x F. Sanger, S. Nicklen, A.R. Coulson, Proc. Natl. Acad. Sci. U.S.A. 74 Ž1977. 5463–5467. w26x J-Y. Masson, J. Vadnais, D. Ramotar, Gene 170 Ž1996. 153–154. w27x J.-Y., Masson, S. Tremblay, D. Ramotar, Gene Ž1996. in press. w28x E. Chan, B. Weiss, Proc. Natl. Acad. Sci. U.S.A. 84 Ž1987. 3189–3193. w29x G. Prabhala, G.H. Rosenberg, N.F. Kaufer, Yeast 8 Ž1992. 171–182. w30x R.D. Gietz, R.H. Schiestl, Yeast 7 Ž1991. 253–263. w31x D. Warshawsky, L. Miller, Biotechniques 16 Ž1994. 798. w32x D. Ramotar, R. Lillis, C. Kim, B. Demple, J. Biol. Chem. 268 Ž1993. 20533–20539. w33x T. Roldan-Arjona, C. Anselmino, T. Lindahl, Nucleic Acids Res. 24 Ž1996. 3307–3312.