A smaller form of the sliding clamp subunit of DNA polymerase III is induced by UV irradiation in Escherichia coli.

The β subunit of DNA polymerase III holoenzyme of Escherichia coli is a 40.6-kDa protein that functions as a sliding DNA clamp (Stukenberg, P. T., Studwell-Vaughan, P. S., and O'Donnell, M.(1991) J. Biol. Chem. 266, 11328-11334). It is responsible for tethering the polymerase to DNA and endowing it with the high processivity required for DNA replication. Here and in a companion study (Paz-Elizur, T., Skaliter, R., Blumenstein, S., and Livneh, Z.(1996) J. Biol. Chem. 271, 2482-2490) we report that the dnaN gene, encoding the β subunit, contains an internal in-frame gene, termed dnaN*, that encodes a smaller form of the β subunit. The novel 26-kDa protein, termed β*, is UV-inducible, and when overexpressed from a plasmid under an inducible promoter, it increases up to 6-fold the UV resistance of E. coli cells. These findings suggest that the β* protein functions in a reaction associated with DNA repair or recovery of DNA replication in UV-irradiated cells.

UV irradiation of Escherichia coli cells produces in DNA primarily cyclobutyl pyrimidine dimers and pyrimidinepyrimidone (6 -4) 1 adducts that are responsible for most of the mutagenic and inactivating effects of UV irradiation (1). The immediate cellular response to UV irradiation is an arrest of chromosome replication in order to allow a period of elimination of DNA damage by DNA repair mechanisms (2). Many of the genes known to be involved in these processes, such as uvrA, uvrB, recA, umuD, and umuC are regulated by the SOS regulatory network, whose primary function is to help the cell in coping with DNA damage (3,4). However, UV irradiation induces also heat shock genes (5) and other genes (6) which affect the post-UV physiology of the cell.
We have previously examined in detail the replication of UV-irradiated DNA with purified proteins (7)(8)(9)(10). These studies revealed that the ␤ subunit of DNA polymerase III holoenzyme, the major replicase of the E. coli chromosome (11), limits the ability of the purified polymerase to bypass UV lesions during in vitro replication of UV-irradiated single-stranded DNA (10). Consistent with this result, overproduction of the ␤ subunit from a plasmid caused a reduction in UV resistance and in UV mutagenesis of E. coli cells (12). This involvement of the ␤ subunit in UV irradiation effects prompted us to examine whether it may be present in a different form in UV-irradiated cells. Here and in two companion studies (27,28), we report that UV irradiation induces a shorter form of the ␤ subunit that functions in vitro as an alternative DNA polymerase clamp. We suggest that this protein functions in DNA synthesis associated with post-UV recovery in E. coli.
Plasmids-Plasmid pUN234 carries the dnaN gene cloned under the lac promoter in plasmid pUC18 (12). Plasmid pUN234FS2 is a derivative of pUN234 containing a ϩ4 insertion/frameshift mutation in the dnaN upstream to dnaN* (12). Plasmid pBSOW1 contains dnaN* expressed under the lac promoter in plasmid pBluescript SK ϩ , and plasmid pBSW7 is similar to pBSOW1 except that dnaN* was cloned in the opposite orientation. These plasmids were constructed by cloning, in two orientations, the EcoRV-(1871)-BanI-(2832) dnaN* DNA fragment from plasmid pUN234 into the EcoRV site in plasmid pBluescript SK ϩ .
Proteins and Chemicals-The ␤ subunit of DNA polymerase III was purified as previously described (13). It was then fractionated by 10% SDS-PAGE, and a gel slice containing 250 -500 g of the protein was ground, mixed with complete Freund's adjuvant, and injected into young female rabbits. The anti-␤ antibodies obtained were affinitypurified on purified ␤* that was fixed onto a nitrocellulose membrane (Schleicher & Schuell, 0.2 m) as described elsewhere (14). Purified ␤* was prepared from an overproducing cell (28). Restriction nucleases were purchased from Pharmacia and from New England Biolabs. T4 DNA ligase was the product of Stratagene, and DNA polymerase I and calf alkaline phosphatase were from U. S. Biochemical Corp.
UV Irradiation-Cultures were UV-irradiated at 254 nm using a low pressure mercury germicidal lamp. The dose rate was 0.1-0.5 J m Ϫ2 s Ϫ1 as determined by a UV products radiometer equipped with a UVX-25 sensor.
Analysis of ␤* Expression in Cells Harboring dnaN and dnaN* Plasmids-E. coli MC4100XL cells harboring plasmids with the dnaN or the dnaN* genes cloned under the lac promoter were treated with 0.5 mM IPTG to induce expression from the lac promoter. Total cellular proteins were fractionated by SDS-PAGE, transferred to a nitrocellulose membrane, and detected with the affinity-purified anti-␤ antibodies using enhanced chemiluminescence for detection (ECL, Amersham Corp.). The following plasmids were tested: pUN234, pUN234FS2, pUC18, pBSOW1, and pBSW7.
Identification of the Cellular ␤*-MC4100 cells were grown in a M9 medium supplemented with 0.2% glucose, vitamin B 1 , and the amino acids histidine, arginine, proline, threonine, and isoleucine (20 mg/liter each). The protein extracts from cells growing at the late logarithmic phase (OD 595 ϭ 1.5) were fractionated by two-dimensional gel electrophoresis according to O'Farrell (15), after which they were transferred onto a nitrocellulose membrane and probed with anti-␤ antibodies using enhanced chemiluminescence for detection (ECL, Amersham Corp.).
Induction of ␤* by UV Irradiation-E. coli MC4100 were grown to OD 595 ϭ 0.3 in a minimal A medium supplemented with vitamin B 1 , 0.2% glucose, and histidine, arginine, proline, threonine, and isoleucine * This research was supported by grants from the Dorot Science Fellowship Foundation and the Scheuer Research Foundation of the Israel Academy of Sciences, and from The Forchheimer Center for Molecular Genetics. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
(20 mg/liter each). The cells were then UV-irradiated at a dose of 50 J m Ϫ2 and returned for further growth. At the indicated time points, cell samples were concentrated to OD 595 ϭ 20, and the amount of total protein in the cell suspension was determined according to Bradford (16) after solubilization of the cells with 0.1 M NaOH. Equal amounts of protein were fractionated by 10% SDS-PAGE, and detection of ␤* was performed by immunoblotting with anti-␤ antibodies.
Effect of Overproduction of ␤* on UV Resistance-AB1157XL or MC4100XL cells harboring the dnaN* plasmids were grown overnight at 30°C in LB containing 0.2% glucose and 100 g/ml ampicillin. They were then centrifuged, resuspended in an equal volume of LB, divided into two, and diluted 1:100 with LB containing 0.2% glucose or with LB containing 0.5 mM IPTG, in the presence of 100 g/ml ampicillin. Cultures were grown to OD 595 ϭ 0.5 at 30°C (glucose) or 37°C (IPTG). The cells were then washed, resuspended in 10 mM Tris⅐HCl, pH 7.5, 0.15 M NaCl, and 5-ml portions were UV-irradiated with agitation at doses of up to 125 J m Ϫ2 . Following irradiation the cultures were diluted and spread on LB plates containing ampicillin and either 0.2% glucose or 0.5 mM IPTG, according to their growth in culture. Cells without plasmids were treated by the same procedure, except that the antibiotics were omitted. Percent survival was calculated by dividing the number of colonies arising from an irradiated culture by the number of colonies arising from the unirradiated culture.

DNA Sequence Elements Typical of a Gene Are Present
Inside the Coding Region of the dnaN Gene-A computer analysis of the dnaN gene which codes for the ␤ subunit (17,18), revealed that its coding sequence contains elements typical of the control region of a gene. This putative gene was termed dnaN*. The DNA sequence elements include a promoter, a Shine-Dalgarno sequence, an in-frame ATG initiation codon, and the sequence 5Ј-CGCTGTCTACCCTGCCAGCG-3Ј (positions 1960 -1979), closely homologous to the consensus sequence of the binding site of the LexA repressor (SOS box), 5Ј-NNCTGT-NTatNcaNNCAGNN-3Ј (3) (Fig. 1, upper panel). Such a configuration of a gene nested inside another gene is not uncommon among viruses, however it is quite rare among bacterial chromosomal genes (19). The putative dnaN* gene product represents the C-terminal portion of the ␤ subunit, with a calculated molecular mass of 26 kDa (Fig. 1, lower panel). These DNA sequence homologies raised the possibility of a UV-inducible, internal in-frame gene that expresses a shorter form the ␤ subunit sliding DNA clamp.
A Plasmid Carrying the Intact dnaN Gene Does Not Express ␤* Unless the Expression of the ␤ Subunit Is Inactivated-We examined whether plasmids carrying the entire dnaN gene direct the synthesis of ␤*. First we used plasmid pUN234 (12), which carries the intact dnaN gene cloned under the lac promoter. Since dnaN* is included within dnaN, we expected that ␤* would be expressed from this plasmid in addition to the ␤ subunit. Cells harboring pUN234 were grown in parallel in medium containing either glucose, a catabolic repressor of the lac operon, or IPTG, a synthetic inducer of the lac operon. Total cellular protein was then fractionated by SDS-PAGE and immunoblotted with polyclonal antibodies against the ␤ subunit. Since ␤* is identical to the C-terminal two-thirds of the ␤ subunit, we expected a good cross-reactivity with ␤*. Purified ␤* that was prepared from an overproducing cell (28) served both as an electrophoretic marker and for the affinity purification of the polyclonal antibodies.
As can be seen in Fig. 2 (lane 4), the ␤ subunit was overproduced; however, no protein of 26 kDa was detected. Puzzled by the inability of this plasmid to express ␤*, we examined its expression from plasmid pUN234FS2, which contains a ϩ4 insertion/frameshift mutation inside the dnaN gene upstream to the beginning of the dnaN* gene. As can be seen (Fig. 2, lane  6), the mutation essentially eliminated the synthesis of the ␤ subunit from the plasmid, and at the same time a band of ␤* could be detected. This represents most likely ␤* synthesized from the mutated plasmid. A possible explanation is that the ␤ subunit itself is a negative regulator of ␤* expression, and when present in large amounts, it inhibits the expression of ␤*. This experiment also suggests that the 26-kDa protein was not formed by proteolysis of the ␤ subunit either in vivo or during extract preparation and handling, otherwise it would have been expected to appear in lane 4. Notice that a 26-kDa protein was not observed also in lanes 2 or 3 in Fig. 2 suggesting that   FIG. 1. The control region of the putative dnaN* gene residing inside the coding sequence of dnaN. The sequence that shows homology to a LexA binding site (SOS boxlike sequence), the Ϫ35 and Ϫ10 regions of the putative promoter, the Shine-Dalgarno (SD) sequence and the putative ATG initiation codon are underlined. Also shown are the two promoters of the adjacent downstream recF gene. Those were previously mapped inside the coding sequence of dnaN (26). Sequence coordinates are according to Ohmori et al. (18). The lower panel shows schematically the relationship between dnaN and dnaN*, and their protein products. cellular ␤* is present in much smaller amount than the ␤ subunit.
The ␤* Protein Is Expressed from a Plasmid Carrying the dnaN* Gene-We next cloned the dnaN* itself under the inducible lac promoter, to yield plasmid pBSOW1, and examined its ability to direct the synthesis of ␤*. As can be seen in Fig. 3, extracts prepared from cells grown with IPTG (but not in its absence) contained a protein of 26 kDa that reacted with anti-␤ antibodies. In contrast, extract from cells with plasmid pBSW7, in which dnaN* was cloned in the opposite orientation to the lac promoter, did not contain ␤*. Thus, the dnaN* gene directs the synthesis of ␤* when present on a plasmid.
Identification of the Cellular ␤*-In order to identify the cellular ␤* protein, we probed by Western blot analysis a protein extract prepared from late logarithmic/early stationary phase E. coli MC4100 using affinity-purified antibodies. As can be seen in Fig. 4 (lane E), a protein of 26 kDa reacted with the anti-␤ antibodies, and exhibited an electrophoretic mobility identical to the overproduced and purified ␤*. Extracts obtained from early logarithmic cells gave a weaker ␤* band (see Fig. 5), indicating that it is induced in the stationary phase. In order to verify that this is the cellular ␤*, we analyzed it by two-dimensional gel electrophoresis. The single 26-kDa protein exhibited a pI of 5.4 (Fig. 4), in good agreement with the theoretical pI value of 5.46, calculated based on the amino acid composition of ␤*, and identical to the pI of the overproduced ␤* (28). Thus, based on its molecular mass, isoelectric point, and reactivity with purified anti-␤ antibodies, this band was assigned as ␤*, the cellular product of dnaN*.
UV Irradiation Induces ␤*-The induction of ␤* was examined directly by immunoblot analysis using purified anti-␤ antibodies. As can be seen in Fig. 5, UV irradiation caused an increase in the level of ␤*, peaking at 1-1.5 h after irradiation. Based on densitometric tracing of the blot, the extent of induction was 7-fold, similar to the induction observed with dnaN*-lacZ translational gene fusions (27).
Overproduction of ␤* Increases UV Resistance-What is the physiological role of ␤*? The ␤* protein is not a general substitute for the ␤ subunit which is needed constantly during the FIG. 2. Expression of proteins from dnaN plasmids. E. coli MC4100XL cells harboring plasmids with the dnaN gene cloned under the lac promoter were treated with IPTG to induce expression from the lac promoter. Equal amounts of total cellular proteins were fractionated by SDS-PAGE, transferred to a nitrocellulose membrane and detected with affinity purified anti-␤ antibodies using enhanced chemiluminescence as described under "Materials and Methods." The plasmids tested were plasmid pUN234, carrying the dnaN gene cloned under the lac promoter in plasmid pUC18 (lanes 3 and 4), plasmid pUN234FS2, a derivative of pUN234, containing a ϩ4 insertion/frameshift mutation in the dnaN upstream to dnaN* (lanes 5 and 6), and the control vector pUC18 (lane 2). Lane 1 contains purified ␤ subunit and ␤* markers.
FIG. 3. Expression of proteins from dnaN* plasmids. E. coli MC4100XL cells harboring plasmids with the dnaN* gene cloned under the lac promoter were treated with IPTG to induce expression from the lac promoter. Total cellular proteins were fractionated by SDS-PAGE, transferred to a nitrocellulose membrane, and detected with affinitypurified anti-␤ antibodies using enhanced chemiluminescence as described under "Materials and Methods." In addition to the ␤ subunit and ␤* the antibodies cross-reacted with two other proteins whose signals on the immunoblots exhibited large variations. Plasmid pB-SOW1 contains dnaN* expressed under the lac promoter in plasmid pBluescript SK ϩ , and plasmid pBSW7 is similar to pBSOW1 except that dnaN* was cloned in the opposite orientation. replication of the chromosome. This is indicated by our inability to complement a dnaN59 temperature-sensitive mutation with a ␤*-producing plasmid. The UV inducibility of ␤* and its relation to the ␤ subunit of DNA polymerase III raised the possibility that it is involved in a recovery function such as DNA repair. To examine this possibility we have determined the effect of overexpression of ␤* on the UV resistance of cells using plasmid pBSOW1 (Fig. 3) that expresses ␤* from the inducible lac promoter.
As can be seen in Table I, induction of the synthesis of ␤* from plasmid pBSOW1 by IPTG caused an increase of up to 6-fold in the UV resistance of two different strains, compared to the same cells grown with glucose and in the absence of IPTG, conditions under which the lac promoter was repressed. Such increase in UV survival was not observed with the control plasmid pBSW7, in which dnaN* was cloned opposite to the lac promoter, or with the cells without any plasmid (Table I). It is noteworthy that when the intact ␤ subunit was overproduced from the same plasmid, it caused a decrease in UV resistance of the cells, opposite to the effect of ␤* (12). Overproduction of the ␣ subunit of DNA polymerase III had no effect on UV resistance (12). These results are consistent with a model in which ␤* participates in a recovery process in the cell, which is limited by the amount of ␤*. Overproduction of ␤* would be expected to facilitate this recovery reaction, and thus increase UV resistance. Possible pathways to be affected are DNA repair or the reactivation of DNA replication (2).
It was recently shown by x-ray crystallography that the ␤ subunit is composed of three structurally similar domains, and it dimerizes to form a hexagon-like ring (20). ␤* contains precisely two of the three domains of the ␤ subunit, raising the possibility that it forms a trimeric alternative clamp that functions in the UV-irradiated cell with one of the DNA polymerases. Such a clamp activity is demonstrated in our companion study (28). The function of the ␤ subunit is carried out in eukaryotes by the proliferating cell nuclear antigen (PCNA), which serves as the processivity clamp of DNA polymerase ␦ (21). PCNA and the ␤ subunit are structurally very similar, forming nearly identical hexagonal rings (22,23). However, in contrast to the dimeric structure of the 40.6 kDa ␤ subunit, PCNA is 29 kDa and forms a trimer (23). Thus ␤* bears resemblance to PCNA with a possible evolutionary link. In this context it is interesting to note that PCNA is the target for several regulatory mechanisms that coordinate the response of mammalian cells to UV irradiation (24,25).

TABLE I
Overproduction of ␤* increases UV resistance Overnight cultures of AB1157XL or MC4100XL cells harboring the various plasmids were washed and grown in parallel to OD 595 -0.5 in fresh medium under conditions in which the lac promoter was repressed (30°C in the presence of glucose) or fully induced (37°C in the presence of IPTG). The cells were then washed, resuspended in 10 mM Tris⅐HCl, pH 7.5, 0.15 M NaCl, and UV-irradiated with agitation at the indicated UV doses. Following irradiation the cultures were diluted and spread on LB plates containing ampicillin and either 0.2% glucose or 0.5 mM IPTG, according to their growth in culture. Cells without plasmids were treated by the same procedure, except that ampicillin was omitted.