Mol Gen Genet (1987) 209:494-498 MG'(3 © Springer-Verlag 1987 Effects of modulation of RNase H production on the recovery of DNA synthesis following UV-irradiation in Escherichia coli Serge Casaregola, Mohammed Khidhir, and I. Barry Holland Department of Genetics, University of Leicester, University Road, Leicester, LEI 7RH, UK Summary. The requirements for the recovery of DNA syn- thesis in UV-irradiated Escherichia coli were analysed in strains having varied levels of RNase H and RecA protein. We have previously shown (Khidhir et al. 1985) that the recovery of DNA synthesis in E. coli following UV treat- ment is an inducible SOS function requiring protein synthe- sis. We proposed that this reflected the need for the synthe- sis of specific induced replisome reactivation factor(s) for recovery. In this study we now show that recovery of DNA synthesis can in fact take place in the absence of protein synthesis in a mutant laking RNase H and having high (constitutive) levels of RecA protein. We also show that expression of rnh is inhibited during the SOS response in recA + but not in a recA- strain. The results are discussed in relation to the mechanism of recovery of DNA synthesis following UV irradiation in E. coli. Key words: UV irradiation - DNA replication - RNase H Introduction UV-irradiation transiently inhibits DNA synthesis in Escherichia coli in a dose dependent manner (Doudney 1972; Burton and Holland 1983). Thus, for example, the rate of DNA synthesis, as measured by pulse labelling with thymidine, returns to the rate of the untreated control after 30-45 rain following a dose of 10 J/m 2 (Burton and Holland 1983; Khidhir et al. 1985). The observed inhibition of DNA synthesis is not, however, due to the synthesis of an induc- ible SOS factor and we have concluded (Khidhir et al. 1985) that replication forks are blocked in vivo at UV-lesions such as dimers. Since newly synthesized DNA following UV-irradiation in E. coli contains gaps apparently opposite dimers (Rupp and Howard-Flanders 1968), replisomes must at some stage by-pass such lesions and we have suggested previously that such a by-pass mechanism is crucial to the recovery of DNA synthesis and hence of survival of the bacteria (Khidhir et al. 1985). Analysis of the nature of this process has re- vealed that recovery is dependent upon an inducible SOS function under recA-lexA control. We have also shown that RecA protein itself is directly involved in recovery, in addi- tion to its role in inactivating LexA under these conditions. However, the presence of high levels of RecA protein was Offprint requests to: I.B. Holland not sufficient to allow recovery of DNA synthesis in the absence of protein synthesis, leading us to conclude that at least one additional, inducible factor was required. We proposed that this SOS inducible function, or IRR factor(s) (for induced replisome reactivation), is either necessary for active movement of stalled replication complexes past dimers (without base insertion) or for reinitiation of replica- tion down stream of a dimer following by-pass by the repli- some, again uncoupled from polymerisation (Khidhir et al. 1985). So-called stable DNA replication (Kogoma and Lark 1975) in E. coli defines a form of replication which con- tinues for several rounds despite the inhibition of protein synthesis. In contrast, protein synthesis is required for initi- ation of normal DNA replication from oriC facilitated by DnaA protein (reviewed by Kornberg 1980). Stable DNA replication (Sdr), which appears to involve alternative meth- ods of RNA priming of initiation (Kogoma 1986) may, however, arise in different ways. One form of stable replica- tion is reported to be an inducible, SOS function (hence iSdr) in E. coli (Kogoma et al. 1979). In addition, Kogoma (1978) has isolated sdrA mutants in which stable replication is constitutive. The sdrA mutants have extremely low levels of RNase H and the sdrA locus has in fact been identified as the structural gene (rnh) for RNase H (Ogawa et al. 1984). Although both types of Sdr require a functional RecA protein for sustained replication, it has been reported that iSdr does not involve detectable changes in the level of RNase H in the cells (Bialy and Kogoma 1986; our un- published data). Several studies have now indicated that RNase H is an important discriminatory factor, ensuring the fidelity of the normal initiation of DNA replication in E. coli by eliminat- ing potential RNA primers at sites other than oriC (Ko- goma and von Meyenburg 1983 ; Lindhal and Lindha11984; Ogawa and Okazaki 1984; Horiuchi et al. 1984). Consistent with this view, rnh- mutants have been shown to initiate DNA replication independently of dnaA from several sites distinct from oriC (De Massy et al. 1984; T. Horiuchi, per- sonal communication). In view of these findings, the induction of a capacity for novel RNA priming might play a crucial role in the reinitiation of DNA replication from abnormal sites, such as those associated with UV-lesions. We therefore investi- gated the recovery of DNA synthesis in strains in which the level of RNase H has been manipulated in various ways. In addition, we have also analysed the expression of the