The Bacteriophage P1 HumD Protein Is a Functional Homolog of the Prokaryotic UmuD′-Like Proteins and Facilitates SOS Mutagenesis in Escherichia coli

Restoration of mutagenesis functions to a normally nonmutable lexA(Def) recA430 strain by P1 HumD. The ability of E. coli UmuD′ and P1 HumD to restore mutagenesis functions to JS431, a recA430 lexA(Def) strain, was monitored in a qualitative plate assay by following reversion of the trpE65(oc) allele in the absence of exogenous DNA damage (−) or after exposing the cells to MMS (+). recA430 lexA(Def) strains are nonmutable because they are unable to posttranslationally mediate the cleavage of UmuD to UmuD′ and therefore give rise to essentially the same number of Trp⁺ revertants as the isogenic ΔumuDC strain, RW262. Mutagenesis can, however, be restored to JS431 by providing a recombinant UmuD′ intrans. As can be seen, the extent of restoration is dependent on the copy number of the plasmid; pRW66 is a low-copy-number UmuD′ plasmid, and pOS32 is a high-copy-number UmuD′ plasmid. Similarly, HumD also restores mutagenesis to JS431, with the extent of mutagenesis related to the copy number/expression of the relevant HumD plasmid: pOS31, low-copy-number HumD; pOS30, high-copy-number HumD translated in the same direction as LacZ′ (expressed from the vector polylinker); pOS31, high-copy-number HumD translated in the opposite direction to LacZ′. The data presented are the means from three independent isolates and three plates per isolate. The error bars represent the standard error of the mean for each experiment.

Specificity of HumD’s ability to restore mutagenesis functions in a ΔumuDC recA718 lexA(Def) strain. The specificity of HumD’s ability to restore mutagenesis functions to RW264 was assayed by introducing compatible plasmids expressing various combinations of E. coli UmuD′ and/or UmuC, pKM101 MucA′ and/or MucB, and R391 RumA′ and/or RumB together with high-copy-number HumD. Mutagenesis was monitored as described in the legend to Fig. 2 by following spontaneous reversion of the trpE65(oc) allele. Strain EW181 demonstrates the level of spontaneous mutagenesis promoted by the chromosomally encoded UmuD(D′) proteins, and RW264 is the isogenic ΔumuDC strain. As can be seen, background levels of spontaneous mutagenesis were observed with RW264 harboring pRW274 (low-copy-number E. coli UmuC), pOS30 (high-copy-number HumD), pOS32 (high-copy-number E. coli UmuD′), pOS33 (high-copy-number R391 RumA′), pOS34 (high-copy-number pKM101 MucA′), pOS35 (low-copy-number R391 RumB), pOS37 (low-copy-number pKM101 MucB) alone or with pRW274/pOS33 (low-copy-number E. coli UmuC and high-copy-number R391 RumA′), pRW274/pOS34 (low-copy-number E. coli UmuC and high-copy-number pKM101 MucA′), and pOS37/pOS30 (low-copy-number pKM101 and high-copy-number HumD). A twofold increase over background was observed with pOS35/pOS30 (low-copy-number R391 RumB and high-copy-number HumD). In dramatic contrast, significant levels of spontaneous mutagenesis, comparable to that seen in theumu⁺ strain, were observed with pRW274/pOS32 (low-copy-number E. coli UmuC and high-copy number E. coli UmuD′) and pRW274/pOS30 (low-copy-number E. coliUmuC and high-copy-number HumD). The data presented are the means from three independent isolates and three plates per isolate. The error bars represent the standard error of the mean for each experiment.

Restoration of UV resistance to a normally UV-sensitive ΔumuDC lexA(Def) recA718 strain by HumD-UmuC. The ability of various plasmids to restore UV resistance to RW264 [ΔumuDC lexA(Def) recA718 uvrA155] was assayed by exposing plasmid containing cultures to UV light and plating serial dilutions on LB agar plates. Surviving colonies were scored after 24 h of incubation at 37°C. The data presented are the means from three independent isolates and three plates per UV dose. Strains analyzed were RW264 alone (⧫), RW264/pOS30 (high-copy-number HumD) (□), RW264/pRW274 (low-copy-number E. coli UmuC) (▴), RW264/pRW274/pOS30 (low-copy-number E. coli UmuC with high-copy-number HumD) (X), and RW264/pRW274/pOS32 (low-copy-numberE. coli UmuC with high-copy-number E. coli UmuD′) (●).

Overproduction and purification of HumD. Bacteriophage P1 HumD was purified from BL21(λDE3) cells carrying the HumD-overproducing plasmid pOS36. Proteins were separated in SDS–15% polyacrylamide gels, and proteins were visualized after staining with Coomassie blue. Lanes U and I are whole-cell extracts from uninduced and IPTG-induced cells, respectively; lanes A, D, H, and G are fractions obtained after selective ammonium sulfate precipitation and DEAE-Sephacel, hydroxyapatite, and AcA54 gel filtration chromatography, respectively. The position of monomeric HumD is marked on the right, and the molecular masses of marker proteins are indicated (in kilodaltons) on the left.

HumD is a dimer in solution. To determine if HumD is capable of forming a dimer with itself, HumD was chemically cross-linked with glutaraldehyde as described in Materials and Methods. Untreated (−) and cross-linked (+) samples were separated in an SDS–15% polyacrylamide gel and transferred to an Immobilon P membrane. The monomeric/multimeric state of HumD was determined by probing the membrane with polyclonal antibodies to HumD and subsequent visualization with the CSPD-Star chemiluminescent substrate. The positions of HumD monomers, dimers, and higher-order structures are indicated on the right.

Damage-inducible expression of HumD. (A) Nucleotide sequence of the humD promoter/operator region. The initiator codon (formylmethionyl [fmet]) of HumD is at the far right and boxed; positions of the −35 and −10 promoter elements and of the ribosome binding site (RBS) are overlined. The proposed LexA-binding site is located between the −10 promoter element and the RBS and is also boxed. It should be noted that this LexA-binding site deviates from the consensus at only three positions. (B) Steady-state levels of HumD expressed from pOS30 in an undamaged or MMC-treatedlexA⁺ strain (RW406) or in an undamagedlexA51(Def) strain (RW260). The position of HumD is indicated on the right. As can be seen, the HumD antiserum recognizes another cellular protein in addition to HumD. The identity of this protein is unknown, but it serves as a useful internal control, ensuring that equal amounts of protein extract have been applied to the gel.

In vivo stability of HumD. Plasmid pOS30 was introduced into the Δ(umuDC)596::ermGT recA⁺ lexA51(Def) strain EC10, and the relative stability of HumD was measured after protein synthesis was inhibited by the addition of chloramphenicol (100 μg/ml) at time zero. Additional aliquots were removed at 10-min intervals. Whole-cell extracts were separated in an SDS–15% polyacrylamide gel, and proteins were visualized using HumD antibodies and the CSPD-Star chemiluminescence assay. Track S is ∼20 ng of highly purified HumD protein; track Δ is an extract of the strain lacking pOS30. As shown in Fig. 7, the HumD antiserum recognizes another cellular protein in addition to HumD that serves as a useful internal control, ensuring that equal amounts of protein extract have been applied to the gel. The position of HumD is indicated on the right.

Stabilization of E. coli UmuC by HumD. A plasmid expressing UmuC alone (pRW274) or coexpressing HumD (pOS30) was introduced into the Δ(umuDC)596::ermGT recA⁺ lexA51(Def) strain, EC10. The relative stability of UmuC was assayed after protein synthesis was inhibited by the addition of chloramphenicol (100 μg/ml) at time zero. Whole-cell extracts were separated in an SDS–15% polyacrylamide gel, and proteins were visualized by using UmuC antibodies and the CSPD-Star chemiluminescence assay. The position of UmuC is indicated by an arrow at the left.