Search

Two enzymes of base excision repair (BER), uracil DNA glycosylase (UDG) and DNA polymerase beta (beta pol), from HeLa cells co-eluted from Superose 12 FPLC columns. The UDG was completely displaced from 150-180-kDa fractions to 30- 70-kDa fractions by brief treatment with 0.5 N NaCl, pH 3.0, as expected when protein-protein associations are disrupted, but beta pol was not displaced by this treatment. UDG was not essential to the presence of beta pol in the 150-180-kDa enzyme complex. beta pol and UDG apparently reside in separate but co-eluting structures. Immunoaffinity chromatography showed that the association of UDG and beta pol was accounted for by attachment in common to DNA and that the association was abolished by eliminating DNA. Evidence for base excision repairosomes containing UDG and beta pol in protein-protein assemblies was not found. However, UDG and human AP endonuclease (HAP1) were associated with HSP70 and HSP27, which are present in 150-180-kDa and 30-70-kDa proteins of cell sonicates. The association of HSPs with BER enzymes was confirmed by hydroxyl radical protein-protein footprinting and immunoaffinity tests. The association of HSPs and BER enzymes is a novel finding. HSP binding may account for the presence of BER enzymes in the two large size class fractions and HSPs may have functional roles in BER.

It was demonstrated previously that a deoxyribophosphodiesterase (dRpase) activity is associated with the DNA repair enzyme exonuclease I, and that this activity is stimulated by the addition of the E. coli single-stranded DNA-binding protein (Ssb). This activity catalyzes the release of deoxyribose-phosphate groups at apurinic/apyrimidinic (AP) sites in the DNA that have been cleared by the action of an AP endonuclease. We have now used the yeast two-hybrid system to demonstrate that a protein-protein interaction occurs between exonuclease I and Ssb. When the E. coli ssb gene was fused in frame to the DNA-activating domain of the GAL4 transcriptional activator and the exonuclease I gene was fused in frame to the DNA-binding domain, a functional GAL4 transcriptional activator was produced as determined by growth of yeast on selective medium and the measurement of beta-galactosidase activity. We have also demonstrated that Ssb can stimulate the dRpase activity of exonuclease I using double-stranded bacteriophage M13 DNA containing several strand interruptions at incised AP sites. These results suggest that Ssb may be required for efficient base-excision repair in bacteria.

Expression of the prokaryotic gene for chloramphenicol acetyltransferase (EC 2.3.1.28) (CAT) in primate cells transfected with X-irradiated plasmid pSV2CAT was determined in transient expression assays. CAT expression did not depend upon the presence of supercoiled plasmids, but relaxed circular forms were essential. X-ray conversion of relaxed circles to linear forms paralleled the loss of CAT expression, with identical D0's in the first part of dose-response curves. X-ray-induced loss of supercoiled forms was complete at much lower doses. The D0 for inactivation of CAT expression by X irradiation of the plasmids in 1 mM Tris buffer was 270 Gy; it was 13 Gy for plasmids irradiated in water. The D0's for conversion of pSV2CAT to relaxed circle forms were only one-seventh as large as the D0's for CAT inactivation after X-ray in water or in 1 mM Tris buffer. Expression of the CAT gene in some representative repair-deficient human fibroblasts transfected with X-irradiated pSV2CAT was less than in monkey CV-1 cells or cell lines from normal human subjects. These results demonstrate a novel means to study low levels of X-ray damage in DNA correlating specific X-ray damage in the DNA with expression of the gene in unirradiated primate cells.

Radiation-induced single-strand breaks were found throughout the 172 bp repeat units of African green monkey component alpha DNA. Two kinds of 3'-ends of 5'-32P-labeled restriction fragments were found, as previously described by others. After irradiation in vitro, the yield of single-strand breaks was 4 X 10(-5) breaks/nucleotide/Gy, as determined by analyses in DNA sequencing type gels. Protection from X-ray damage was found when the DNA received 150 Gy in the presence of 2-mercaptoethanol. The results demonstrate a very sensitive quantitative means to study the role of indirect effects of ionizing radiation on strand-break induction and protection at the base sequence level. Component alpha DNA was isolated from irradiated CV-1 cells and was analyzed for single-strand breaks. Under these conditions the frequency of breaks was less than the frequency obtained when purified DNA was irradiated. The methodology is presented because of its relevance to the study of DNA strand breakage in living cells.