Expression of UMP1 is inducible by DNA damage and required for resistance of S. cerevisiae cells to UV light

It has recently been shown that the UMP1 gene of Saccharomyces cerevisiae encodes a small. short-lived protein engaged in 20S proteasome formation. The results presented in this paper demonstrate that ULMP1 expression is induced by the DNA damaging agents methyl methanesulfonate (MMS) and UV light as well as by hydroxyurea (HU), an inhibitor of DNA replication. MMS induction of UMP1 expression occurs at the transcriptional level and is independent of the activity of the regulatory checkpoint kinases encoded by MEC1. RAD53 or DUN1. It is also shown that the disruption of UMP1 causes increased sensitivity of yeast cells to killing by UV radiation, but only slight sensitivity to HU treatment, and does not cause any increase in the killing effect of MMS.

The product of the DNA damage-inducible gene of Saccharomyces cerevisiae, DIN7, specifically functions in mitochondria

We reported previously that the product of the DNA damage-inducible gene of Saccharomyces cerevisiae, DIN7, belongs to a family of proteins that are involved in DNA repair and replication. The family includes S. cerevisiae proteins Rad2p and its human homolog XPGC, Rad27p and its mammalian homolog FEN-1, and Exonuclease I (Exo I). Here, we report that Din7p specifically affects metabolism of mitochondrial DNA (mtDNA). We have found that dun1 strains, defective in the transcriptional activation of the DNA damage-inducible genes RNR1, RNR2, and RNR3, exhibit an increased frequency in the formation of the mitochondrial petite (rho(-)) mutants. This high frequency of petites arising in the dun1 strains is significantly reduced by the din7::URA3 allele. On the other hand, overproduction of Din7p from the DIN7 gene placed under control of the GAL1 promoter dramatically increases the frequency of petite formation and the frequency of mitochondrial mutations conferring resistance to erythromycin (E(r)). The frequencies of chromosomal mutations conferring resistance to canavanine (Can(r)) or adenine prototrophy (Ade(+)) are not affected by enhanced synthesis of Din7p. Experiments using Din7p fused to the green fluorescent protein (GFP) and cell fractionation experiments indicate that the protein is located in mitochondria. A possible mechanism that may be responsible for the decreased stability of the mitochondrial genome in S. cerevisiae cells with elevated levels of Din7p is discussed.

Different repair of O6-methylguanine occurring in DNA modified by MMS in vivo or in vitro

Lack of the adaptive response effect on the level of GC-->AT transitions induced by methyl methanesulfonate (MMS) in E. coli [Sledziewska-Gójska, E. (1993) The level of GC-->AT transitions induced by MMS is not affected by adaptive response of Escherichia coli K12. Mutation Res., 294, 1-8.] can be explained by MMS inactivation of the ada encoded O6-methylguanine-DNA methyltransferase [Takahashi, K.Y., Kawazoe, K., Sakumi, Nakabeppu Y. and M. Sekiguchi (1988) Activation of Ada protein as a transcriptional regulator by direct alkylation with methylating agents, J. Biol. Chem., 263, 13490-13492; Sledziewska-Gójska, E. (1995) Inactivation of O6-methylguanine-DNA methyltransferase in vivo by SN2 alkylating agents, Mutation Res., 336, 61-67]. To evaluate this explanation and clarify the origin of MMS-induced GC-->AT transitions, we compared the repair of DNA treated by MMS in vivo or in vitro. Replication forms of lacZ mutants of E. coli phage M13mp18 were used to analyse the effect of the adaptive response on the frequency of GC-->AT transitions occurring in control and mismatch repair deficient strains. It was shown that DNA lesions, leading to GC-->AT transitions, induced by MMS in vivo are not repaired in adapted E. coli cells. In contrast, induction of the adaptive response causes efficient repair of these DNA lesions induced by MMS in vitro. This repair is consistent with the assumption that GC-->AT transitions induced by MMS are originated by O6-methylguanine and that MMS treatment of the cells during in vivo mutagenesis interfere with the adaptation mediated repair of the lesion. In agreement with this we have shown that treatment of the adapted cultures with 5 mM MMS completely blocks repair of in vitro modified DNA. Increased level of GC-->AT transitions induced by MMS occurs in mutS- strains. These mutations are avoided in adapted mutS- cells, when induced by MMS in vitro. This confirms that mismatch repair system of E. coli recognises mismatches formed in DNA by O6-methylguanine.

The antimutagenic effect of a truncated epsilon subunit of DNA polymerase III in Escherichia coli cells irradiated with UV light

It has previously been suggested that inhibition of the proofreading 3'-5' exonuclease activity of DNA polymerase may play an important role in generation of UV-induced mutations in Escherichia coli. Our previous work showing that overproduction of epsilon, the proofreading subunit of DNA polymerase III, counteracts the SOS mutagenic response of E. coli seemed to be consistent with this hypothesis. To explore further the nature of the antimutagenic effect of epsilon we constructed plasmid pMK17, which encodes only two of the three highly conserved segments of epsilon--ExoI and ExoII; the third segment, ExoIII, which is essential for 3'-5' exonuclease activity, is deleted. We show that at 40 degrees C, overproduction of the truncated epsilon subunit significantly delays production of M13 phage, suggesting that the protein retains its capacity to bind to DNA. On the other hand, the presence of pMK17 in a trpE65 strain growing at 40 degrees C causes a 10-fold decrease in the frequency of UV-induced Trp+ mutations. This antimutagenic effect of the truncated epsilon is effectively relieved by excess UmuD,C proteins. We also show that the presence of plasmid pIP21, which contains the dnaQ49 allele encoding an epsilon subunit that is defective in proofreading activity, almost completely prevents generation of UV-induced mutations in the trpE65 strain. We propose that the DNA binding ability of free epsilon, rather than its 3'-5' exonuclease activity, affects processing of premutagenic UV-induced lesions, possibly by interfering with the interaction between the UmuC-UmuD'-RecA complex and Pol III holoenzyme. This interaction is probably a necessary condition for translesion synthesis.

Inactivation of O6-methylguanine-DNA methyltransferase in vivo by SN2 alkylating agents

The cellular level of O6-methylguanine-DNA methyltransferase (MGMT) is important in mutagenic, carcinogenic and therapeutic effects of alkylating agents. I have investigated how SN2 alkylating agents affect the activity of MGMT in vivo. As a model, adapted cultures of E. coli K12 strain AB2497 containing 2400 +/- 430 molecules of MGMT per cell were used. MGMT activity was assayed in the cell extracts of adapted cultures challenged with various doses of MMS, DMS and for comparison the SN1 alkylating agents, MNNG and MNU. In control non-adapted cultures, with low constitutive levels of MGMT, the mutagenic potential of various doses of different alkylating agents was estimated to correlate with the O6-methylguanine content produced in DNA by various treatments. Inactivation of MGMT by MNNG or MNU occurs only in doses able to produce a highly mutagenic level of O6-methylguanine in DNA, which is consistent with the consumption of MGMT activity in the DNA repair process. In contrast, non-mutagenic doses of MMS or DMS are sufficient to inactivate MGMT in adapted E. coli cells. It may be concluded that SN2 alkylating agents can block the main pathway of O6-methylguanine-DNA repair in vivo.

Different UmuC requirements for generation of different kinds of UV-induced mutations in Escherichia coli

An Escherichia coli strain bearing the dnaQ49 mutation, which results in a defective epsilon subunit of DNA polymerase III, and carrying the lexA71 mutation, which causes derepression of the SOS regulon, is totally unable to maintain high-copy-number plasmids containing the umuDC operon. The strain is also unable to maintain the pAN4 plasmid containing a partial deletion of the umuD gene but retaining the wild-type umuC gene. These results suggest that a high cellular level of UmuC is exceptionally harmful to the defective DNA polymerase III of the dnaQ49 mutant. We have used this finding as a basis for selection of new plasmid umuC mutants. The properties of two such mutants, bearing the umuC61 or umuC95 mutation, are described in detail. In the umuC122::Tn5 strain harbouring the mutant plasmids, UV-induced mutagenesis is severely decreased compared to that observed with the parental umuDC+ plasmid. Interestingly, while the frequency of UV-induced GC-->AT transitions is greatly reduced, the frequency of AT-->TA transversions is not affected. Both mutant plasmids bear frameshift mutations within the same run of seven A residues present in umuC+; in umuC61 the run is shortened to six A whereas in umuC95 is lengthened to eight A. We have found in both umuC61 and umuC95 that translation is partially restored to the proper reading frame. We propose that under conditions of limiting amounts of UmuC, the protein preferentially facilitates processing of only some kinds of UV-induced lesions.

The level of GC-->AT transitions induced by MMS is not affected by the adaptive response in Escherichia coli K12

I have examined the effect of the adaptive response on the frequency of MMS-induced umuC-dependent AT-->TA transversions and umuC-independent, GC-->AT transitions. It was found that the induction of the adaptive response causes a moderate decrease (50-60%) in the frequency of AT-->TA transversions and surprisingly has no effect on the level of GC-->AT transitions. In contrast, a dramatic decrease in MNNG-induced mutations has been observed in adapted cultures. However, this effect was completely abolished by MMS pretreatment of adapted cells before MNNG challenge. A mechanism for the MMS interference with the repair of MNNG-induced mutations in adapted cells is proposed.

Mutagenesis of Escherichia coli: a method for determining mutagenic specificity by analysis of tRNA suppressors

A method for estimating mutagenic specificity in Escherichia coli (argE3, hisG4, thr-1, supE44), based upon the isolation of Arg+ or His+ revertants and identification of tRNA suppressors, is described. The method gives an insight not only into mutagenic pathways but also into the functioning of tRNA. With N-methyl-N'-nitro-N-nitrosoguanidine, 98% of mutations are GC----AT transitions. With N4-hydroxycytidine, 100% are AT----GC transitions. With hydroxylamine, apart from GC----AT transitions, approximately 30% of Arg+ revertants are formed by GC (or AT)----TA transversions. When the chemistry of the mutagenic attack is known, the method allows us to discriminate whether mutations occur on the transcribed or non-transcribed strands of DNA. It has been found that reversion of argE3 to Arg+ is a better monitor of mutagenic pathways than reversion of hisG4 to His+.

Alternative pathways of methyl methanesulfonate-induced mutagenesis in Escherichia coli

Methyl methanesulfonate (MMS) induced mutagenesis is known to be largely dependent on functional umuCD and recA genes. By phenotypic analysis of Arg+ (argE3, ochre) revertants according to their reversion of the mutations his-4 (ochre) and thr-1 (amber), we attempted to deduce the specificity and/or sites of MMS-induced mutations. It is shown that: (1) MMS-induced, umuC-dependent Arg+ revertants (which prevail in bacteria proficient in mismatch repair) result from a different mutational pathway from umuC-independent ones. UmuC-dependent Arg+ revertants belong to class 2 (Arg+His+Thr-), and umuC-independent ones to class 1 (Arg+His-Thr-). (2) The mismatch repair system very efficiently prevents mutations induced by MMS. We found that in the mutS strain, deficient in mismatch repair, class 1 Arg+ revertants are the most numerous, whereas class 2 Arg+ revertants occur at similar levels in MMS-treated mutS and mutS+ strains. Therefore the mismatch repair system very efficiently prevents formation of umuC-independent Arg+ revertants, but exerts negligible or no effect on umuC-dependent Arg+ revertants. (iii) Both mutS umuC and mutS recA strains, are highly mutable by MMS.

Effect of proofreading and dam-instructed mismatch repair systems on N4-hydroxycytidine-induced mutagenesis

The role of the proofreading (3' leads to 5' exonuclease) function of T4 DNA polymerase and the mismatch repair system of E. coli on N4-hydroxycytidine (oh4Cyd) induced mutagenesis was investigated. oh4Cyd-induced mutation is strongly suppressed when the proofreading activity increases as a result of the presence of tsCB87--antimutator polymerase or elevated temperature (43 degrees C vs 30 degrees C). Mutagenic activity of oh4Cyd, however, is little, if at all, affected by the presence of the tsLB56 mutator allele of T4 DNA polymerase with suppressed proofreading activity. This leads to the conclusion that oh4Cyd nucleotides are not frequently removed by proofreading activity of wild-type T4 DNA polymerase. The number of mutations induced by oh4Cyd increases 3- to 5-fold due to damage of the genes mutS, mutL, uvrE, but not mutR. Dam- cells are more sensitive to, and hypermutable by, oh4Cyd in comparison with dam+ cells. This is compatible with the notion that oh4C residues are recognised and excised by mismatch repair enzymes. The results indicate that neither the proofreading function of T4 DNA polymerase, nor the mismatch repair enzymes, are responsible for the high specificity of oh4Cyd which causes AT leads to GC transition.