Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

The rejoining of double-strand breaks in DNA by human cell extracts

The rejoining of double-strand breaks in DNA by human cell extracts Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 © 7990 Oxford University Press Nucleic Acids Research, Vol. 18, No. 21 6205 The rejoining of double-strand breaks in DNA by human cell extracts Phillip North, Anil Ganesh and John Thacker* MRC Radiobiology Unit, Chilton, Didcot, Oxon 0X11 ORD, UK Received September 4, 1990; Revised and Accepted October 9, 1990 ABSTRACT A double-strand DNA break was introduced at a specific In the present study we have attempted to refine the site within the lacZ gene of plasmid pUC18 using one measurement and analysis of DNA-break rejoining with the use of nuclear extracts from human cells applied to defined molecules of several restriction enzymes, and the plasmid carrying specific enzymatically-induced dsb. As prototypes we exposed to nuclear extracts from human cell lines. have compared the activities of extracts from an A-T cell line Physical rejoining of DNA was monitored by Southern analysis after gel separation, and the fidelity of (AT5BIVA) with those from a line showing normal radiation- rejoining by expression of the lacZ gene after bacterial sensitivity (MRC5V1). The simple recombinant plasmid pUC18 was used as a substrate because it has a number of different transformation with the treated plasmid. Breaks at the enzyme break sites closely spaced at the same location (the Sa/I and EcoRI sites were rejoined by extracts to form multicloning site) on the molecule. This site is within the lacZ circular monomers, but the efficiency of rejoining was much higher at the Sa/1 site. Measurement of rejoining gene of pUC18, allowing the fidelity of rejoining to be assessed at several adjacent sites having different types of by expression of normal gene activity after extract treatment. termini, consistently showed a range of efficiencies with 5' 4-base > 3' 4-base overhangs and 4-base > MATERIALS AND METHODS 2-base > no overhang. Similar efficiencies were found for nuclear extracts from transformed cell lines, both Cells and extracts from a 'normal' individual and an ataxia-telangiectasia The transformed human cell lines MRC5V1 (17) and AT5BIVA (A-T) patient, and from a non-transformed normal cell (18) and the non-transformed HF12 cells (19) were grown in culture. In contrast at some sites, especially those with Eagle's Minimal Essential Medium (MEM), supplemented with a low rejoin efficiency, the fidelity of rejoining was very 1 mM glutamine, 10% foetal calf serum and antibiotics, at 37°C much lower for the A-T extracts than for normal cell in 5% CO :95% air. The sensitivity of these cell lines to extracts. Mis-rejoining was, however, unrelated to ionising radiation was checked by colony-forming assay; rejoin efficiency at other sites, suggesting that factors AT5BIVA was approximately 4-fold more sensitive to X-rays such as the exact sequence at the break site on the than MRC5V1 or HF12 (data not shown), in agreement with molecule may also influence the fidelity of rejoining. previous data (12,13). Cells were held as frozen stocks until required for extract preparation; they were grown up to final numbers in excess of INTRODUCTION 108 cells using large numbers of flasks or petri dishes. Extract procedures were based on those of Lopez et al. (20); ice-cold The introduction of defined and localized sites of damage into conditions were used throughout. Cells were scraped into purified DNA molecules, followed by exposure of the DNA to phosphate-bufferred saline (PBS) using a 'rubber policeman' and a cellular environment, is an important technique in the molecular washed 3 times in PBS with centrifugation. The cell pellet was analysis of DNA repair and mutagenesis in mammalian cells resuspended in 10 ml buffer A (Tris-HCl, pH 7.5, 20 mM; (e.g., 1 —5). Localization of a damage site allows it to be followed MgCl , 0.5 mM; dithiothreitol, 0 s mM; KC1, 0.5 mM; CaCl , thereafter for mechanistic analysis of subsequent processing. 2 2 An important type of DNA damage, induced by agents such 2 mM) containing 250 mM sucrose, and held for 15 min. After centrifugation, the cells were resuspended in buffer A without as ionising radiation, is the double-strand break (dsb). A number sucrose but with PMSF (phenylmethylsulphonyl fluoride, 1 mM) of studies implicate the dsb as a lethal lesion in cells (6 — 8), and and homogenized (40 strokes) in a Dounce homogeniser to release some radiation-sensitive mutants of mammalian cells appear to have a reduced ability to rejoin this type of damage (9 — 11). intact nuclei. Nuclei were pelletted (2000 g, 1 min) and washed Individuals with the disorder ataxia-telangiectasia (A-T), who 4 times in buffer A with triton X-100 (0.5%) and 3 times without triton but with PMSF. Nuclei were resuspended in 2 ml buffer suffer both radiosensitivity and cancer-proneness (12-14), may B (NaCl, 500 mM; EDTA, 10 mM; PMSF, 0.1 mM) and also be unable to repair DNA breaks adequately in comparison sonicated for 4 X15 sec bursts. Debris was pelletted in microtubes to normal human cells (15,16). • To whom correspondence should be addressed Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 6206 Nucleic Acids Research, Vol. 18, No. 21 (12000 rpm for 30 min). Protein was precipitated from the 12 — 14 hr growth the bacteria were harvested and plasmid supernatant by adding ammonium sulphate (0.34 g/ml) and isolated using the alkaline-lysis method (21). After exposure to stirring for 2 hrs. The precipitate was pelletted at 15000 g for various restriction enzymes (see text) the DNA was run on either 20 min; then it was dissolved and dialysed overnight in buffer 1% agarose or 5% polyacrylamide gels. C (Tris-HCl, pH7.5, 50 mM; EDTA, 0.1 mM; 2-mercaptoethanol, 10 mM; PMSF, 0.1 mM; glycerol, 10%) RESULTS before use or storage at -70° C for up to 6 months. Protein Conditions for break rejoining by human nuclear extracts concentration in the extracts was measured using the Biorad Extracts prepared from nuclei of the radiation-sensitive cell line protein assay. AT5BIVA and its normal counterpart MRC5V1 were found to Plasmid preparation be able to rejoin broken plasmid molecules to form both circular monomers and linear catenates. A typical experiment showing HB101 bacteria containing the plasmid pUC18 were grown in the untreated, broken and extract-treated plasmid DNA is given bulk culture, and plasmid DNA was isolated and purified using in Fig. 1 to illustrate the formation of rejoined molecules and the standard methods (21). Double-strand breaks were introduced controls. The enzyme T4 ligase was used as a positive control at single sites in the multicloning region of pUC18, to interrupt in experiments, while nuclear extracts were boiled as negative the lacZ gene, using restriction endonucleases (Bethesda Research controls. It is seen that unbroken DNA has predominantly closed Labs) as shown below. circular monomers (CC) although some open circular (OC) and dimeric forms are present; when broken with endonuclease a GCCAAGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCCCGGGTACCGAGCTCGAATTC Hin&lll Pstl BmnM EcdRl single linear (LIN) band is found. Rejoining with extract or T4 Sail/ SstI ligase yields a set of CC-forms (topoisomers, seen when gels Accl/Hincll are run without ethidium bromide present) and linear multimers, while boiled extract does not. The bacterial transformation The linearized plasmid was phenol/chloroform purified; breakage frequencies found with these DNA samples are given in the was checked by agarose gel electrophoresis and Southern legend to Fig.l; it was found that measured transformation analysis. frequencies and crude visual estimates of the amount of rejoined circular monomers on gel blots were in general quantitative Reaction conditions agreement. As expected, breaking the DNA reduced the Plasmid DNA and protein extracts were mixed in 50 /xl reactions containing Tris.HCl (pH7.5, 65.5 mM), MgSO (10 mM), ATP (1 mM),EDTA (91 nM), 2-mercaptoethanol (9.1mM), PMSF (91 nM), glycerol (9.1 %), and DNA (40 /xg/ml). Nuclear-extract protein (0.2-3. 8 mg/ml) or T4 ligase (BRL; 0.01 U) was added to start the reaction. Incubation was at 14°C for 20—26 hrs. DNA was subsequently purified by phenol/chloroform extraction and ethanol precipitation. Some reactions were flash frozen in liquid nitrogen and stored at —70°C before purification. Gel electrophoresis and Southern analysis After a brief heating (65° for 10 min) a quarter of the sample was run on a 1 % agarose gel to estimate the DNA concentration. Subsequently 1 ng of DNA from each sample was run on a 1 % (LiNr agarose gel (without ethidium bromide, unless otherwise noted) <oc> in TBE buffer (21), blotted on to nitrocellulose (Schleicher and (UN) Schull), and probed with pUC18 labelled by the random oligonucleotide method (22) to a specific activity of 0.4-2.0X10 9 cpm//tg DNA. — (CO — oc Bacterial transformation and rejoin fidelity Bacterial transformations were carried out using the E. coli strain CC DH5alpha (23) and the standard transformation protocol of Hanahan (24). The transformants were selected on LM plates containing ampicillin (100 jig/ml) and X-gal (5-Bromo- 4-chloro-3-indolyl-/3-D-galactopyranoside:40 jtg/ml). Bacterial viability was assessed on plates containing X-gal but no ampicillin. White colonies (not expressing /S-galactosidase — EcoRI-cut — uncut activity) were streaked onto LM plates with ampicillin and X- gal for confirmation. Transformation frequencies were expressed as transformants per viable cell to allow comparison between Fig. 1. Southern blot hybridization to unbroken or £coR]-cut (treated and untreated) pUCI8 DNA after electrophorctic separation. The extract protein concentration different experimental samples. was 0.6 mg/ml. DNA forms: CC, closed circular, OC, open circular; LJN, linear (also multimeric forms shown by powers). These samples yielded bacterial Analysis of mis-rejoined plasmid DNA transformation frequencies per 106 viable cells as follows: uncut 392; White colonies, previously streaked onto LM plates, were used cut/untreated 1.1; cut/extract treated 4.3; cut/boiled-extract treated 1.1; cut/T4-ligase treated 277. to inoculate cultures in nutrient broth with ampicillin. After Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 Nucleic Acids Research, Vol. 18, No. 21 6207 transformation frequency by about 400-fold (25,26), while extract conditions, rejoining by T4 ligase was more efficient at the EcoRI (and T4 ligase) treatment of broken DNA increased this frequency site than at the Sail site (data not shown). significantly. Fidelity of rejoining of endonuclease-broken plasmid The influence of reaction conditions was explored as follows: (a) protein concentration: the production of CC monomers and The fraction of pUC18 plasmids yielding white colonies, by mis- other forms was concentration dependent, but saturated at rejoining of the broken lacZ gene, was measured in control different concentrations depending on the site of the break. This samples consisting of unbroken DNA, untreated broken DNA, is illustrated in Fig.2 for pUC18 broken at the Ecc&l or Sail or broken DNA exposed to boiled extracts or T4 DNA ligase. sites (see also below). It is seen (Table 2) that this fraction is extremely low in unbroken (b) plasmid DNA concentration: preliminary experiments made DNA, while untreated broken DNA gives a higher white colony fraction presumably due to the action of the bacterial host cells at a j: i ratio (27) favouring the formation of circular monomers yielded few rejoined molecules, while the relatively high on the DNA ends (see Discussion). DNA treated with boiled concentration of 40 /ig/ml gave consistent formation of circular nuclear extracts is 'mis-rejoined' to a similar extent to untreated monomers and linear dimers. broken DNA; a small increase may occur with £coRI-broken DNA, but the numbers counted are relatively few. Treatment (c) cofactors: the formation of circular monomers required ATP with T4 ligase reduces the mis-rejoined fraction in broken DNA and magnesium ions, but not dNTPs (data not shown). A long by approximately 10-fold at either break-site. incubation time was allowed for maximum formation of rejoined molecules, although we have found that this can be shortened The results of large-scale transformation experiments to give substantially without loss of circular monomer formation. measurable frequencies of white colonies from extract-treated DNA are shown in Tables 3 and 4, for £coRI-broken and Sa/I- Rejoining of broken plasmid by normal and A-T nuclear broken DNA respectively. The fraction of mis-rejoined plasmids extracts found for £coRI-broken DNA exposed to extracts from MRC5V1 cells is relatively similar to that for untreated broken DNA Extensive data were obtained for pUC18 DNA broken at the EcoRl or Sail sites. Fig. 3 gives an example of Southern blot data (compare Tables 2 and 3), but this fraction is very high after for rejoining at these sites by extracts from nuclei of A-T and exposure to AT5BIVA nuclear extracts. This result was normal cells. It is seen that extracts from the different cell types gave approximately the same amount of rejoined material and very similar transformation frequencies (Fig.3 legend). It should be noted, however, that £c<?RI-broken DNA was rejoined with lower efficiency than &z/I-broken DNA by the nuclear extracts (longer exposure of the gel blot in Fig.3 showed the formation of circular monomers with £coRI-broken DNA). Combined data from several experiments gave the average transformation (LIN) frequencies shown in Table 1; overall these showed a difference of about 5-fold between EcoRI-broken and Sa/I-broken DNA frequencies after rejoining with extracts from either cell type. This difference between the two break-sites was also reflected LIN in the protein-concentration dependence of monomer and multimer formation (Fig.2). In contrast, under these reaction CC-forms EcoRI-cut Sail-cut T4 ligase treated MRC5 AT5 MRC5 AT5 Fig. 3 . Southern blot hybridization to broken pUC18 DNA after treatment with oc different cell extracts at the same protein concentration (0.6 mg/ml) or T4 ligase. LIN iii DNA forms are indicated as in Fig.l . Bacterial transformation frequencies per 106 viable cells were as follows: uncut 108; £coRI-cut: /untreated 1.7, /MRC5-extract treated 4.2, /AT5-extract treated 4.2; Safl-cut: /MRC5-extract treated 25.8, /AT5-extract treated 33.2. Table 1. Transformation frequencies for DNA treated extracts from human cell nuclei - PROTEIN CONCENTRATION Ipg/mll 28 260 560 8*0 1120 28 280 560 M0 112<> „ Cell type Frequency per 103 viable bacteria* 140 420 700 980 1300 U0 «20 '00 980 1300 ( £coRI-break Sofl-break Ratio MRC5V1 1.5(7) 9.5 (3) 0.16 AT5BIVA 1.9(16) 8.0 (4) 0.23 Fig. 2. Effect of increasing AT5BIVA nuclear extract protein concentrations on the rejoining of pUC18 broken at the £ooRl or Sail sites. Gel run in the presence "Numbers of expts. shown in parentheses; these extracts had a range of protein of ethidium bromide, to consolidate the CC forms in one position; note that this concentrations, but similar data were found for comparisons at the same protein position is slightly different from that for the original (bacterial) uncut material. concentration. Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 6208 Nucleic Acids Research, Vol. 18, No. 21 reproducible within a factor of about 2 for independent extracts concentrations (giving a transformation frequency reduction of and experiments (Table 3). In addition, several samples of DNA 10 or more; i.e., similar to the frequencies found for EcoRI- broken with EcoRl were used, and these were put through more broken DNA). However, the fraction of white colonies was not stringent purification procedures prior to extract treatment, to increased above that found at the higher protein concentrations ensure that these results did not arise from some spurious effect tested (data not shown). peculiar to the DNA sample (note also that the rejoining of the Rejoin efficiency and fidelity of AT5BIVA extracts at other EcoRI-broken DNA by T4 ligase was not compromised, break sites indicating that poor fidelity did not arise from repair of DNA damaged in the initial production of breaks by the endonuclease). Rejoin efficiency may be affected by degree of 'cohesiveness' Thus it appears that the A-T nuclear extract lacks fidelity in of the DNA ends, including both the type and number of bases rejoining these simple breaks. involved, although it is not clear whether this also affects the rejoin fidelity. Therefore, we examined the efficiency and fidelity For the rejoining of Sa/I-broken DNA (Table 4) a different of rejoining by AT5BIVA nuclear extracts at several other sites picture was found; both MRC5V1 and AT5BIVA extract- in die pUC18 multicloning region (see Methods). Some of these treatments gave similarly low fractions of white colonies. Since, for a given protein concentration, rejoining at the Sail site is much sites, like EcoRI and Sail, generate 5' 4-base overhangs (BamHl, Hindin), while others give 3' 4-base overhangs (Pstl, SsiT). In more efficient than for the EcoRl site (Fig.2) we also measured addition, we tested Hindi which yields non-cohesive ends, and the mis-rejoined fraction for Sa/I-broken DNA at very low protein Accl which breaks the pUC18 DNA at the same site as Sail but has a 5' 2-base overhang. Table 2. Mis-rejoining in control DNA samples As yet, with //mcll-broken DNA we have seen no increase Treatment White:blue ratio1 Percent whites in transformation frequency over control (broken untreated DNA) £coR]-break Safl-break of broken DNA EcoRI-break Safl-break levels after treatment with any extract and very little formation of circular monomers (although some linear dimers; data not Untreated 6: 2873 (5) 7:1546 (7) 0.21 0.45 shown). Table 5 shows representative transformation frequencies Boiled extract 41:5796(11)3:579(2) 0.70 0.52 T4 ligase 6:15089(8) 1:1581(2) 0.04 0.06 for DNA broken at each of the other sites, and the proportion [ Unbroken2 2:35243 (18) 0.005 ] of mis-rejoined molecules found. It is seen that /Icd-broken DNA is rejoined with relatively low efficiency, even when compared 'number of expts. shown in parentheses to £coRI-broken DNA, and has a high fraction of mis-rejoined ^broke n DNA treated with nuclear extracts also gave a very low fraction of plasmids with AT5BIVA extract treatment. The 5' 4-base white colonies Table 3. £coRI-break mis-rejoining by human nuclear extracts AT5BIVA Extract Expt MRC5V1 (protein Percent White:blue (protein Percent no. no. White:blue mg/ml) ratio mg/ml) whites ratio whites 1 (0.2) 0.45 557:4948 (0.6) 10.1 1 4: 874 620:4928 (0.9) 11.2 368:3284 (1.7) 10.1 70:1014 (2.5) 6.5 2 3: 877 (0.2) 0.34 662:5868 (3.8) 10.1 3 678:4221 (0.6) 13.8 4 544:3501 (3.8) 11.2 2 5 3:1187 (1.7) 0.25 130:2271 (0.6) 5.4 15: 254 (1.3) 5.6 3 6 9:2251 (0.6) 0.40 115:1653 (0.6) 6.5 98:1505 (0.9) 6.1 7 7:3901 (0.6) 0.18 (0.6) 0.27 8 6:2199 9 0: 169' (0.6) <0.59 10 1: 531 (0.6) 0.25 'DNA concentration lower than in other treatments (j:i >2) Table 4. Safl-brcak mis-rejoining by human nuclear extracts Extract Expt MRC5V1 AT5BIVA no. no. White:blue (protein Percent White:blue (protein Percent ratio mg/ml) whites ratio mg/ml) whites 1 1 0:1508 (0.2) <0.07 1:3557 (0.6) 0.03 (3.7) 5:2822 0.18 2 0:1787 (3.7) <0.06 3 0: 214 (1.7) <0.4 6 1:1271 (0.6) 0.08 (1.3) 2:1289 0.15 3 4 0:2097 (0.4) <0.05 0:1792 (0.6) <0.0 6 0.22 0: 744 (0.9) <0.1 3 3:1333 (0.6) 5 1:1805 (0.6) 0.06 Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 Nucleic Acids Research, Vol. 18, No. 21 6209 overhang sites (BamHI, HindUT) have relatively high rejoin Rejoin efficiency of non-transformed cell extracts efficiencies and a much smaller mis-rejoined fraction than £coRI- The nuclear extracts used for the above assay were isolated from broken DNA (Table 5), although this fraction is considerably transformed cell lines, which may be atypical of normal (diploid) higher than in controls (Table 2) or Sa/I-broken DNA (Tables cells isolated freshly from tissue. However, early passage human 4 and 5). The 3 ' 4-base overhangs have a somewhat lower rejoin fibroblasts were found to show similar relative rejoin efficiencies efficiency than the 5' 4-base overhangs, but these two sites show as those established for the transformed lines over a range of a striking difference in the fidelity of rejoin (percent whites, Table break-sites (Fig.4). Mis-rejoined fractions were relatively low 5). Thus, /tol-broken DNA gave a relatively high mis-rejoined at all break-sites (data not shown). fraction while &/I-broken DNA showed a mis-rejoined fraction only slightly higher than that found for Sa/I-broken DNA. Analysis of mis-rejoined plasmid molecules When pUC18 is broken with PvwII a fragment of 322 base pairs, including the whole multicloning site, is released. Plasmid isolated Table 5. Transformation frequency and mis-rejoined fractions of pUC18 treated from individual white bacterial colonies gave a range of PvuU with AT5BIVA extracts after breakage at different sites fragment sizes, with the majority showing losses of < 100 base Enzyme Overhang Transformation White:blue Percent pairs, for both MRC5V1 and AT5BIVA treatments. Plasmids sequence frequency' ratio2 whites which cut once or did not cut with PvuII were analysed on agarose gels using larger fragments generated by BgH or Seal digests. Ecofd 5' AATT 1.3 358:5683 5.9 Fig.5 shows the overall distribution of deletion sizes for 153 Sail 5' TCGA 20 3:5096 0.06 5' GATC 9.2 plasmids isolated (13 MRC5V1-, 140 AT5BIVA-extract treated). Bamm 47:3688 1.3 5 ' AGCT Hindm 13 58:5012 1.1 Pstl 3 ' TGCA 3.2 98:5102 1.9 Sal 3 ' AGCT 5.6 DISCUSSION 9:4139 0.22 Accl 5' CG 0.21 29:560 4.9 Rejoin efficiency 'Values given per 105 viable bacteria, all from the same expt. to ensure We have been able to measure successfully the rejoining of comparability. 'model' DNA double-strand breaks by human nuclear extracts, 2Using extracts 2 and 3, with protein concentrations in the range 0.6— 1.3 under simple reaction conditions. The rejoining appears to be mg/ml. taken to completion by the extracts; it should be noted that the DNA is repurified and heated after extract treatment, so that proteins will be removed and weak DNA bonding associations will be broken before analysis. However, we have found that I Break - si te s — the efficiency of the rejoining process varies with the break site. (extract treated) The 5-fold difference established for the rejoin efficiency of Sail breaks and £coRI breaks was supported by data for other break- sites. In part these data suggest that the cohesiveness of ends is £ £ * & ^ * £ .£ important for rejoining efficiency, as seen for example by the £ r? tr -9 -!P n° Co i~ low rejoining efficiency of HincU- (non-cohesive) or of Accl- breaks (2-base overhang). It is possible that the EccRUSaR difference is explained similarly, since EcdSl termini (5' AATT overhang; 8 hydrogen bonds) will be more weakly associated than Sail termini (5' TCGA overhang; 10 bonds). However, the data for other break sites indicate that other factors, such as the polarity of the termini or possibly specific base-sequence and 4 5 400 S00 200 300 Bau pair* Fig. 4. Efficiency of rejoining of different break-sites in pUC18 by extracts from untransformed HF12 cells (at 0.72 mg/ml protein), compared to unbroken or T4 ligase-rejoined material (gel run with ethidium bromide). Transformation Fig. 5. Distribution of deletion sizes in mis-rejoined plasmids after extracMreatment frequencies per 10s viable bacteria were: uncut (untreated) 294, EcoRl 7.4, Sail and selection for lack of expression of the lacZ gene (white colonies). Closed 62, Accl 0.8, Bamm 12.5, Pstl 1.5, Sstl 2.6, Hindm 29. bars, AT5BIVA extract-treated; hatched bars, MRC5V1 extract-treated. Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 6210 Nucleic Acids Research, Vol. 18, No. 21 topological location of these sites, may influence their response. deficiency (34). However, we cannot at present comment on the Thus Sail, BamHl- and //mdTII-breaks with 5' 4-base overhangs relevance of our data to the genetic defect in the A-T disorder; were rejoined somewhat more efficiently than the 3' -base extracts from only one A-T cell line were used and it will be overhangs given by Pstl- and Ssfl-breaks. These data for the necessary to test extracts from the cells of other patients before mammalian nuclear extracts contrast with the profile found for generalizations can be made. T4 ligase where, under the same reaction conditions, the rejoin efficiencies were: HitvttR 2: Sstl = Pstl > EcoW > BamHl ACKNOWLEDGEMENTS > Sail > Accl (data not shown). We have found, in agreement with our previous cellular studies This study was supported in part by the Commission of the (5), that extracts from transformed A-T and normal cells do not European Communities contract B16-E-144-UK. differ in rejoin efficiency. These results are in keeping with measurements showing similar DNA ligase activities in cells of REFERENCES these genotypes (28). We have now shown, in addition, that extracts from early-passage (diploid) human cells give similar 1. Brown.T.C. and JiricnyJ. (1987) Ceil, 50, 945-950. 2. Dolan.M.E., Oplinger.M. and Pegg.A.E. (1988) Mutation Res., 193, relative efficiencies of rejoin at the different sites tested (Fig.4). 131-137. 3. Moriya.M., Takeshita.M., Johnson.F., Peden.K., Will.S. and Grollman.A.P. Rejoin fidelity (1988) Proc. Natl. Acad. Sd. USA, 85, 1586-1589. 4. Ellison.K.S., Dogliotti.E. and EssigmanJ.M. (1989) Mutation Res., 220, While nuclear extracts from the different cell types have similar 93-100 . rejoin efficiencies, we have found differences in the fidelity with 5. ThackerJ. (1989) Mutation Res. 220, 187-204. which these extracts rejoin broken DNA at certain sites. Although 6. Frankenberg.D., Frankenberg-Schwager.M., Bloecher.D. and Harbich.R. it is known that bacteria will mis-rejoin a proportion of broken (1981) Radial. Res., 88, 524-532. plasmid molecules (29; see also Table 2), high mis-rejoin 7. Sargentini.N.J. and Smhh.K.C. (1986) Radial. Res., 107, 58-72 . 8. Bryant.P.E. (1985) Int. J. Radial. Biol., 48, 55-60 . frequencies were only found after human nuclear extract 9. Kemp,L., Sedgwkk.S.G. and Jeggo.P.A. (1984) Mutation Res., 132, treatment. Extracts from the nuclei of the AT5BFVA cell line, 189-196. in particular, give a highly significant increase in the loss of lacZ 10. Wlodek.D. and Hittelman.W.N. (1987) Radiat. Res., 112, 146-155. gene function when rejoining certain break-sites. It is notable that 11. Stamato.T.D., DipatriA and GiacctaA (1988) Radiat. Res., 115,325-333. those break-sites (especially EcoRl and Accl) which yield a low 12. Taykn\A.M.R., Hamden,D.G., Ariett,C.F., Harcourt,S.A., Lehmann.A.R., Stevens.S. and Bridges.B.A. (1975) Nature (London) 258, 427-429. fidelity of rejoining with the A-T extract also have relatively low 13. Cox,R., and Masson.W.K. (1980) Int. J. Radiat. Biol. 38, 575-576. rejoin efficiencies. This suggests that where breaks are more 14. Boder,E. (1985) In Gatti.R.A. and Swift.M. (pis.) Ataxia-Telangiectasia: difficult to rejoin, some extracts will act on the broken DNA Genetics, Neuropathology and Immunology of a Degenerative Disease of to remove sequence. Childhood. Liss, New York, pp. 1-63 . 15. McKinnon.P.J. (1987) Human Genet. 75, 197-208. However, the two 3' 4-base overhang sites show similar 16. ThackerJ. (1989) Bioessays, 11, 58-62 . 'intermediate' rejoin efficiencies but give quite different mis- 17. Huschtscha.L.I., and HoUiday.R. (1983)7. CellSd., 63, 77-99 . rejoined fractions (Table 5), in one case (Sstf) similar to that for 18. Day.R.S., ZioUcowskJ.C.HJ., Scudiero.D.A., Meyer.S.A., Lubiniecki,A.S., the site having the highest rejoin efficiency (Sail). It is notable Giradi.A.J., Galloway.S.M. and Bynum.G.D. (1980) Nature (London), 288, that these two endonucleases generate the same sequence at the 724-727 . 19. Cox.R. and Masson,W.K. (1974) Int. J. Radiat. Biol., 26, 193-196. overhang (5' TCGA or 3' AGCT), while the reverse of this 20. Lopez,B., Rousset.S. and Coppey.J. (1987) Nucleic Acids Res., 15, sequence (HindHl break site, 5' AGCT) does not show high 5643-5655. fidelity rejoining. This result suggests that the nuclear extract 21 . Sambrook,J., Fritsch.E.F., and Maniatis.T. (1989) Molecular Cloning-A may contain some very specific means of ensuring that certain Laboratory Manual, 2nd. Edn. Cold Spring Harbor Lab. Press, New York. sequences are rejoined faithfully. 22. Feinberg.A.P. and Vogelstein.B. (1983) Anal. Biochem. 132, 6-13 . 23. Bethesda Research Laboratories (1986) Focus, 8, 9. It is known that broken DNA ends act as sites of action or 24. Hanahan.D. (1985) In Glover.D.M. (ed.) DNA Cloning, IRL Press, Oxford., recognition for enzymes which degrade or delete sequence. Vol.1, pp. 109-135. Simple mechanisms for lack of fidelity, such as an increased 25. Cohen.S.N., Chang.A.C.Y. and Hsu.L. (1972) Proc. Natl. Acad. Sd. USA 69, 2110-2114. nuclease activity in the A-T extract, were not revealed by an 26. Conley.E.C. and SaundersJ.R. (1984) MoL Gen. Genet., 194,211-218. increase in the extent of degradation of DNA on gels after 27. Dugaiczyk.A., Boycr.H.W. and Goodman,H.M. (1975)7. Mol. Biol., 96, treatment with A-T extracts, and no differences between cell lines 171-184. were found in simple nuclease assays (data not shown). 28. Willis.A.E. and Lindahl.T. (1987) Nature (London), 325, 355-357. Alternatively, there may be lack of 'protection' of DNA ends 29. Thompson.R. and Achtman.M. (1979) Mol. Gen. Genet., 169, 49-57 . 30. AkroydJ., and Symonds.N. (1986) Gene, 49, 273-282. in the A-T extract (5), allowing degradative processes access to 31. de Vries.E., van Driel,W., Bergsma.W.G., Amberg.A.C. and van der the DNA. Such protection would arise minimally from the weak Vliet.P. (1989) / . Mol. Biol., 208, 65-78 . hydrogen-bonded association of the broken ends, more strongly 32. Conley.E.C, Saunders,V.A., Jackson.V., and SaundersJ.R. (1986) Nudeic from proteins which bind to the ends, and maximally from rapid Acids Res., 14, 8919-8932. ligation. DNA binding proteins protecting broken ends have been 33. Pfeiffer,P. and Vielmetter.W. (1988) Nucleic Acids Res., 16, 907-924. 34. Malyim,B.A., Blackwell.T.K., Fulop.G.M., Rathbun,G.A., Furley.A.J.W., identified, for example, in infectivity studies of bacteriophage Ferrier.P., Heinke,L.B., Phillips.R.A., Yancopoulos.G.D. and Alt.F.W. Mu (30) and in HeLa cells infected with adenovirus (31). (1988) Cell, 54, 453-460. Additionally, in studies of break rejoining in other organisms, such as Escherichia coli (32) and extracts from Xenopus laevis eggs (33), it has been suggested that 'accessory' (DNA binding) proteins are involved in the rejoining process. A defect in a complex that holds free ends together after endonucleolytic breakage has also been proposed in studies of aberrant immune gene rearrangements in mice suffering severe combined immune http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nucleic Acids Research Oxford University Press

The rejoining of double-strand breaks in DNA by human cell extracts

Download PDF
6 pages

Loading...

Page 2

Loading...

Page 3

Loading...

Page 4

Loading...

Page 5

Loading...

Page 6

 
/lp/oxford-university-press/the-rejoining-of-double-strand-breaks-in-dna-by-human-cell-extracts-le0cAE7a4b

References (1)

Publisher
Oxford University Press
Copyright
© 1990 Oxford University Press
ISSN
0305-1048
eISSN
1362-4962
DOI
10.1093/nar/18.21.6205
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 © 7990 Oxford University Press Nucleic Acids Research, Vol. 18, No. 21 6205 The rejoining of double-strand breaks in DNA by human cell extracts Phillip North, Anil Ganesh and John Thacker* MRC Radiobiology Unit, Chilton, Didcot, Oxon 0X11 ORD, UK Received September 4, 1990; Revised and Accepted October 9, 1990 ABSTRACT A double-strand DNA break was introduced at a specific In the present study we have attempted to refine the site within the lacZ gene of plasmid pUC18 using one measurement and analysis of DNA-break rejoining with the use of nuclear extracts from human cells applied to defined molecules of several restriction enzymes, and the plasmid carrying specific enzymatically-induced dsb. As prototypes we exposed to nuclear extracts from human cell lines. have compared the activities of extracts from an A-T cell line Physical rejoining of DNA was monitored by Southern analysis after gel separation, and the fidelity of (AT5BIVA) with those from a line showing normal radiation- rejoining by expression of the lacZ gene after bacterial sensitivity (MRC5V1). The simple recombinant plasmid pUC18 was used as a substrate because it has a number of different transformation with the treated plasmid. Breaks at the enzyme break sites closely spaced at the same location (the Sa/I and EcoRI sites were rejoined by extracts to form multicloning site) on the molecule. This site is within the lacZ circular monomers, but the efficiency of rejoining was much higher at the Sa/1 site. Measurement of rejoining gene of pUC18, allowing the fidelity of rejoining to be assessed at several adjacent sites having different types of by expression of normal gene activity after extract treatment. termini, consistently showed a range of efficiencies with 5' 4-base > 3' 4-base overhangs and 4-base > MATERIALS AND METHODS 2-base > no overhang. Similar efficiencies were found for nuclear extracts from transformed cell lines, both Cells and extracts from a 'normal' individual and an ataxia-telangiectasia The transformed human cell lines MRC5V1 (17) and AT5BIVA (A-T) patient, and from a non-transformed normal cell (18) and the non-transformed HF12 cells (19) were grown in culture. In contrast at some sites, especially those with Eagle's Minimal Essential Medium (MEM), supplemented with a low rejoin efficiency, the fidelity of rejoining was very 1 mM glutamine, 10% foetal calf serum and antibiotics, at 37°C much lower for the A-T extracts than for normal cell in 5% CO :95% air. The sensitivity of these cell lines to extracts. Mis-rejoining was, however, unrelated to ionising radiation was checked by colony-forming assay; rejoin efficiency at other sites, suggesting that factors AT5BIVA was approximately 4-fold more sensitive to X-rays such as the exact sequence at the break site on the than MRC5V1 or HF12 (data not shown), in agreement with molecule may also influence the fidelity of rejoining. previous data (12,13). Cells were held as frozen stocks until required for extract preparation; they were grown up to final numbers in excess of INTRODUCTION 108 cells using large numbers of flasks or petri dishes. Extract procedures were based on those of Lopez et al. (20); ice-cold The introduction of defined and localized sites of damage into conditions were used throughout. Cells were scraped into purified DNA molecules, followed by exposure of the DNA to phosphate-bufferred saline (PBS) using a 'rubber policeman' and a cellular environment, is an important technique in the molecular washed 3 times in PBS with centrifugation. The cell pellet was analysis of DNA repair and mutagenesis in mammalian cells resuspended in 10 ml buffer A (Tris-HCl, pH 7.5, 20 mM; (e.g., 1 —5). Localization of a damage site allows it to be followed MgCl , 0.5 mM; dithiothreitol, 0 s mM; KC1, 0.5 mM; CaCl , thereafter for mechanistic analysis of subsequent processing. 2 2 An important type of DNA damage, induced by agents such 2 mM) containing 250 mM sucrose, and held for 15 min. After centrifugation, the cells were resuspended in buffer A without as ionising radiation, is the double-strand break (dsb). A number sucrose but with PMSF (phenylmethylsulphonyl fluoride, 1 mM) of studies implicate the dsb as a lethal lesion in cells (6 — 8), and and homogenized (40 strokes) in a Dounce homogeniser to release some radiation-sensitive mutants of mammalian cells appear to have a reduced ability to rejoin this type of damage (9 — 11). intact nuclei. Nuclei were pelletted (2000 g, 1 min) and washed Individuals with the disorder ataxia-telangiectasia (A-T), who 4 times in buffer A with triton X-100 (0.5%) and 3 times without triton but with PMSF. Nuclei were resuspended in 2 ml buffer suffer both radiosensitivity and cancer-proneness (12-14), may B (NaCl, 500 mM; EDTA, 10 mM; PMSF, 0.1 mM) and also be unable to repair DNA breaks adequately in comparison sonicated for 4 X15 sec bursts. Debris was pelletted in microtubes to normal human cells (15,16). • To whom correspondence should be addressed Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 6206 Nucleic Acids Research, Vol. 18, No. 21 (12000 rpm for 30 min). Protein was precipitated from the 12 — 14 hr growth the bacteria were harvested and plasmid supernatant by adding ammonium sulphate (0.34 g/ml) and isolated using the alkaline-lysis method (21). After exposure to stirring for 2 hrs. The precipitate was pelletted at 15000 g for various restriction enzymes (see text) the DNA was run on either 20 min; then it was dissolved and dialysed overnight in buffer 1% agarose or 5% polyacrylamide gels. C (Tris-HCl, pH7.5, 50 mM; EDTA, 0.1 mM; 2-mercaptoethanol, 10 mM; PMSF, 0.1 mM; glycerol, 10%) RESULTS before use or storage at -70° C for up to 6 months. Protein Conditions for break rejoining by human nuclear extracts concentration in the extracts was measured using the Biorad Extracts prepared from nuclei of the radiation-sensitive cell line protein assay. AT5BIVA and its normal counterpart MRC5V1 were found to Plasmid preparation be able to rejoin broken plasmid molecules to form both circular monomers and linear catenates. A typical experiment showing HB101 bacteria containing the plasmid pUC18 were grown in the untreated, broken and extract-treated plasmid DNA is given bulk culture, and plasmid DNA was isolated and purified using in Fig. 1 to illustrate the formation of rejoined molecules and the standard methods (21). Double-strand breaks were introduced controls. The enzyme T4 ligase was used as a positive control at single sites in the multicloning region of pUC18, to interrupt in experiments, while nuclear extracts were boiled as negative the lacZ gene, using restriction endonucleases (Bethesda Research controls. It is seen that unbroken DNA has predominantly closed Labs) as shown below. circular monomers (CC) although some open circular (OC) and dimeric forms are present; when broken with endonuclease a GCCAAGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCCCGGGTACCGAGCTCGAATTC Hin&lll Pstl BmnM EcdRl single linear (LIN) band is found. Rejoining with extract or T4 Sail/ SstI ligase yields a set of CC-forms (topoisomers, seen when gels Accl/Hincll are run without ethidium bromide present) and linear multimers, while boiled extract does not. The bacterial transformation The linearized plasmid was phenol/chloroform purified; breakage frequencies found with these DNA samples are given in the was checked by agarose gel electrophoresis and Southern legend to Fig.l; it was found that measured transformation analysis. frequencies and crude visual estimates of the amount of rejoined circular monomers on gel blots were in general quantitative Reaction conditions agreement. As expected, breaking the DNA reduced the Plasmid DNA and protein extracts were mixed in 50 /xl reactions containing Tris.HCl (pH7.5, 65.5 mM), MgSO (10 mM), ATP (1 mM),EDTA (91 nM), 2-mercaptoethanol (9.1mM), PMSF (91 nM), glycerol (9.1 %), and DNA (40 /xg/ml). Nuclear-extract protein (0.2-3. 8 mg/ml) or T4 ligase (BRL; 0.01 U) was added to start the reaction. Incubation was at 14°C for 20—26 hrs. DNA was subsequently purified by phenol/chloroform extraction and ethanol precipitation. Some reactions were flash frozen in liquid nitrogen and stored at —70°C before purification. Gel electrophoresis and Southern analysis After a brief heating (65° for 10 min) a quarter of the sample was run on a 1 % agarose gel to estimate the DNA concentration. Subsequently 1 ng of DNA from each sample was run on a 1 % (LiNr agarose gel (without ethidium bromide, unless otherwise noted) <oc> in TBE buffer (21), blotted on to nitrocellulose (Schleicher and (UN) Schull), and probed with pUC18 labelled by the random oligonucleotide method (22) to a specific activity of 0.4-2.0X10 9 cpm//tg DNA. — (CO — oc Bacterial transformation and rejoin fidelity Bacterial transformations were carried out using the E. coli strain CC DH5alpha (23) and the standard transformation protocol of Hanahan (24). The transformants were selected on LM plates containing ampicillin (100 jig/ml) and X-gal (5-Bromo- 4-chloro-3-indolyl-/3-D-galactopyranoside:40 jtg/ml). Bacterial viability was assessed on plates containing X-gal but no ampicillin. White colonies (not expressing /S-galactosidase — EcoRI-cut — uncut activity) were streaked onto LM plates with ampicillin and X- gal for confirmation. Transformation frequencies were expressed as transformants per viable cell to allow comparison between Fig. 1. Southern blot hybridization to unbroken or £coR]-cut (treated and untreated) pUCI8 DNA after electrophorctic separation. The extract protein concentration different experimental samples. was 0.6 mg/ml. DNA forms: CC, closed circular, OC, open circular; LJN, linear (also multimeric forms shown by powers). These samples yielded bacterial Analysis of mis-rejoined plasmid DNA transformation frequencies per 106 viable cells as follows: uncut 392; White colonies, previously streaked onto LM plates, were used cut/untreated 1.1; cut/extract treated 4.3; cut/boiled-extract treated 1.1; cut/T4-ligase treated 277. to inoculate cultures in nutrient broth with ampicillin. After Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 Nucleic Acids Research, Vol. 18, No. 21 6207 transformation frequency by about 400-fold (25,26), while extract conditions, rejoining by T4 ligase was more efficient at the EcoRI (and T4 ligase) treatment of broken DNA increased this frequency site than at the Sail site (data not shown). significantly. Fidelity of rejoining of endonuclease-broken plasmid The influence of reaction conditions was explored as follows: (a) protein concentration: the production of CC monomers and The fraction of pUC18 plasmids yielding white colonies, by mis- other forms was concentration dependent, but saturated at rejoining of the broken lacZ gene, was measured in control different concentrations depending on the site of the break. This samples consisting of unbroken DNA, untreated broken DNA, is illustrated in Fig.2 for pUC18 broken at the Ecc&l or Sail or broken DNA exposed to boiled extracts or T4 DNA ligase. sites (see also below). It is seen (Table 2) that this fraction is extremely low in unbroken (b) plasmid DNA concentration: preliminary experiments made DNA, while untreated broken DNA gives a higher white colony fraction presumably due to the action of the bacterial host cells at a j: i ratio (27) favouring the formation of circular monomers yielded few rejoined molecules, while the relatively high on the DNA ends (see Discussion). DNA treated with boiled concentration of 40 /ig/ml gave consistent formation of circular nuclear extracts is 'mis-rejoined' to a similar extent to untreated monomers and linear dimers. broken DNA; a small increase may occur with £coRI-broken DNA, but the numbers counted are relatively few. Treatment (c) cofactors: the formation of circular monomers required ATP with T4 ligase reduces the mis-rejoined fraction in broken DNA and magnesium ions, but not dNTPs (data not shown). A long by approximately 10-fold at either break-site. incubation time was allowed for maximum formation of rejoined molecules, although we have found that this can be shortened The results of large-scale transformation experiments to give substantially without loss of circular monomer formation. measurable frequencies of white colonies from extract-treated DNA are shown in Tables 3 and 4, for £coRI-broken and Sa/I- Rejoining of broken plasmid by normal and A-T nuclear broken DNA respectively. The fraction of mis-rejoined plasmids extracts found for £coRI-broken DNA exposed to extracts from MRC5V1 cells is relatively similar to that for untreated broken DNA Extensive data were obtained for pUC18 DNA broken at the EcoRl or Sail sites. Fig. 3 gives an example of Southern blot data (compare Tables 2 and 3), but this fraction is very high after for rejoining at these sites by extracts from nuclei of A-T and exposure to AT5BIVA nuclear extracts. This result was normal cells. It is seen that extracts from the different cell types gave approximately the same amount of rejoined material and very similar transformation frequencies (Fig.3 legend). It should be noted, however, that £c<?RI-broken DNA was rejoined with lower efficiency than &z/I-broken DNA by the nuclear extracts (longer exposure of the gel blot in Fig.3 showed the formation of circular monomers with £coRI-broken DNA). Combined data from several experiments gave the average transformation (LIN) frequencies shown in Table 1; overall these showed a difference of about 5-fold between EcoRI-broken and Sa/I-broken DNA frequencies after rejoining with extracts from either cell type. This difference between the two break-sites was also reflected LIN in the protein-concentration dependence of monomer and multimer formation (Fig.2). In contrast, under these reaction CC-forms EcoRI-cut Sail-cut T4 ligase treated MRC5 AT5 MRC5 AT5 Fig. 3 . Southern blot hybridization to broken pUC18 DNA after treatment with oc different cell extracts at the same protein concentration (0.6 mg/ml) or T4 ligase. LIN iii DNA forms are indicated as in Fig.l . Bacterial transformation frequencies per 106 viable cells were as follows: uncut 108; £coRI-cut: /untreated 1.7, /MRC5-extract treated 4.2, /AT5-extract treated 4.2; Safl-cut: /MRC5-extract treated 25.8, /AT5-extract treated 33.2. Table 1. Transformation frequencies for DNA treated extracts from human cell nuclei - PROTEIN CONCENTRATION Ipg/mll 28 260 560 8*0 1120 28 280 560 M0 112<> „ Cell type Frequency per 103 viable bacteria* 140 420 700 980 1300 U0 «20 '00 980 1300 ( £coRI-break Sofl-break Ratio MRC5V1 1.5(7) 9.5 (3) 0.16 AT5BIVA 1.9(16) 8.0 (4) 0.23 Fig. 2. Effect of increasing AT5BIVA nuclear extract protein concentrations on the rejoining of pUC18 broken at the £ooRl or Sail sites. Gel run in the presence "Numbers of expts. shown in parentheses; these extracts had a range of protein of ethidium bromide, to consolidate the CC forms in one position; note that this concentrations, but similar data were found for comparisons at the same protein position is slightly different from that for the original (bacterial) uncut material. concentration. Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 6208 Nucleic Acids Research, Vol. 18, No. 21 reproducible within a factor of about 2 for independent extracts concentrations (giving a transformation frequency reduction of and experiments (Table 3). In addition, several samples of DNA 10 or more; i.e., similar to the frequencies found for EcoRI- broken with EcoRl were used, and these were put through more broken DNA). However, the fraction of white colonies was not stringent purification procedures prior to extract treatment, to increased above that found at the higher protein concentrations ensure that these results did not arise from some spurious effect tested (data not shown). peculiar to the DNA sample (note also that the rejoining of the Rejoin efficiency and fidelity of AT5BIVA extracts at other EcoRI-broken DNA by T4 ligase was not compromised, break sites indicating that poor fidelity did not arise from repair of DNA damaged in the initial production of breaks by the endonuclease). Rejoin efficiency may be affected by degree of 'cohesiveness' Thus it appears that the A-T nuclear extract lacks fidelity in of the DNA ends, including both the type and number of bases rejoining these simple breaks. involved, although it is not clear whether this also affects the rejoin fidelity. Therefore, we examined the efficiency and fidelity For the rejoining of Sa/I-broken DNA (Table 4) a different of rejoining by AT5BIVA nuclear extracts at several other sites picture was found; both MRC5V1 and AT5BIVA extract- in die pUC18 multicloning region (see Methods). Some of these treatments gave similarly low fractions of white colonies. Since, for a given protein concentration, rejoining at the Sail site is much sites, like EcoRI and Sail, generate 5' 4-base overhangs (BamHl, Hindin), while others give 3' 4-base overhangs (Pstl, SsiT). In more efficient than for the EcoRl site (Fig.2) we also measured addition, we tested Hindi which yields non-cohesive ends, and the mis-rejoined fraction for Sa/I-broken DNA at very low protein Accl which breaks the pUC18 DNA at the same site as Sail but has a 5' 2-base overhang. Table 2. Mis-rejoining in control DNA samples As yet, with //mcll-broken DNA we have seen no increase Treatment White:blue ratio1 Percent whites in transformation frequency over control (broken untreated DNA) £coR]-break Safl-break of broken DNA EcoRI-break Safl-break levels after treatment with any extract and very little formation of circular monomers (although some linear dimers; data not Untreated 6: 2873 (5) 7:1546 (7) 0.21 0.45 shown). Table 5 shows representative transformation frequencies Boiled extract 41:5796(11)3:579(2) 0.70 0.52 T4 ligase 6:15089(8) 1:1581(2) 0.04 0.06 for DNA broken at each of the other sites, and the proportion [ Unbroken2 2:35243 (18) 0.005 ] of mis-rejoined molecules found. It is seen that /Icd-broken DNA is rejoined with relatively low efficiency, even when compared 'number of expts. shown in parentheses to £coRI-broken DNA, and has a high fraction of mis-rejoined ^broke n DNA treated with nuclear extracts also gave a very low fraction of plasmids with AT5BIVA extract treatment. The 5' 4-base white colonies Table 3. £coRI-break mis-rejoining by human nuclear extracts AT5BIVA Extract Expt MRC5V1 (protein Percent White:blue (protein Percent no. no. White:blue mg/ml) ratio mg/ml) whites ratio whites 1 (0.2) 0.45 557:4948 (0.6) 10.1 1 4: 874 620:4928 (0.9) 11.2 368:3284 (1.7) 10.1 70:1014 (2.5) 6.5 2 3: 877 (0.2) 0.34 662:5868 (3.8) 10.1 3 678:4221 (0.6) 13.8 4 544:3501 (3.8) 11.2 2 5 3:1187 (1.7) 0.25 130:2271 (0.6) 5.4 15: 254 (1.3) 5.6 3 6 9:2251 (0.6) 0.40 115:1653 (0.6) 6.5 98:1505 (0.9) 6.1 7 7:3901 (0.6) 0.18 (0.6) 0.27 8 6:2199 9 0: 169' (0.6) <0.59 10 1: 531 (0.6) 0.25 'DNA concentration lower than in other treatments (j:i >2) Table 4. Safl-brcak mis-rejoining by human nuclear extracts Extract Expt MRC5V1 AT5BIVA no. no. White:blue (protein Percent White:blue (protein Percent ratio mg/ml) whites ratio mg/ml) whites 1 1 0:1508 (0.2) <0.07 1:3557 (0.6) 0.03 (3.7) 5:2822 0.18 2 0:1787 (3.7) <0.06 3 0: 214 (1.7) <0.4 6 1:1271 (0.6) 0.08 (1.3) 2:1289 0.15 3 4 0:2097 (0.4) <0.05 0:1792 (0.6) <0.0 6 0.22 0: 744 (0.9) <0.1 3 3:1333 (0.6) 5 1:1805 (0.6) 0.06 Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 Nucleic Acids Research, Vol. 18, No. 21 6209 overhang sites (BamHI, HindUT) have relatively high rejoin Rejoin efficiency of non-transformed cell extracts efficiencies and a much smaller mis-rejoined fraction than £coRI- The nuclear extracts used for the above assay were isolated from broken DNA (Table 5), although this fraction is considerably transformed cell lines, which may be atypical of normal (diploid) higher than in controls (Table 2) or Sa/I-broken DNA (Tables cells isolated freshly from tissue. However, early passage human 4 and 5). The 3 ' 4-base overhangs have a somewhat lower rejoin fibroblasts were found to show similar relative rejoin efficiencies efficiency than the 5' 4-base overhangs, but these two sites show as those established for the transformed lines over a range of a striking difference in the fidelity of rejoin (percent whites, Table break-sites (Fig.4). Mis-rejoined fractions were relatively low 5). Thus, /tol-broken DNA gave a relatively high mis-rejoined at all break-sites (data not shown). fraction while &/I-broken DNA showed a mis-rejoined fraction only slightly higher than that found for Sa/I-broken DNA. Analysis of mis-rejoined plasmid molecules When pUC18 is broken with PvwII a fragment of 322 base pairs, including the whole multicloning site, is released. Plasmid isolated Table 5. Transformation frequency and mis-rejoined fractions of pUC18 treated from individual white bacterial colonies gave a range of PvuU with AT5BIVA extracts after breakage at different sites fragment sizes, with the majority showing losses of < 100 base Enzyme Overhang Transformation White:blue Percent pairs, for both MRC5V1 and AT5BIVA treatments. Plasmids sequence frequency' ratio2 whites which cut once or did not cut with PvuII were analysed on agarose gels using larger fragments generated by BgH or Seal digests. Ecofd 5' AATT 1.3 358:5683 5.9 Fig.5 shows the overall distribution of deletion sizes for 153 Sail 5' TCGA 20 3:5096 0.06 5' GATC 9.2 plasmids isolated (13 MRC5V1-, 140 AT5BIVA-extract treated). Bamm 47:3688 1.3 5 ' AGCT Hindm 13 58:5012 1.1 Pstl 3 ' TGCA 3.2 98:5102 1.9 Sal 3 ' AGCT 5.6 DISCUSSION 9:4139 0.22 Accl 5' CG 0.21 29:560 4.9 Rejoin efficiency 'Values given per 105 viable bacteria, all from the same expt. to ensure We have been able to measure successfully the rejoining of comparability. 'model' DNA double-strand breaks by human nuclear extracts, 2Using extracts 2 and 3, with protein concentrations in the range 0.6— 1.3 under simple reaction conditions. The rejoining appears to be mg/ml. taken to completion by the extracts; it should be noted that the DNA is repurified and heated after extract treatment, so that proteins will be removed and weak DNA bonding associations will be broken before analysis. However, we have found that I Break - si te s — the efficiency of the rejoining process varies with the break site. (extract treated) The 5-fold difference established for the rejoin efficiency of Sail breaks and £coRI breaks was supported by data for other break- sites. In part these data suggest that the cohesiveness of ends is £ £ * & ^ * £ .£ important for rejoining efficiency, as seen for example by the £ r? tr -9 -!P n° Co i~ low rejoining efficiency of HincU- (non-cohesive) or of Accl- breaks (2-base overhang). It is possible that the EccRUSaR difference is explained similarly, since EcdSl termini (5' AATT overhang; 8 hydrogen bonds) will be more weakly associated than Sail termini (5' TCGA overhang; 10 bonds). However, the data for other break sites indicate that other factors, such as the polarity of the termini or possibly specific base-sequence and 4 5 400 S00 200 300 Bau pair* Fig. 4. Efficiency of rejoining of different break-sites in pUC18 by extracts from untransformed HF12 cells (at 0.72 mg/ml protein), compared to unbroken or T4 ligase-rejoined material (gel run with ethidium bromide). Transformation Fig. 5. Distribution of deletion sizes in mis-rejoined plasmids after extracMreatment frequencies per 10s viable bacteria were: uncut (untreated) 294, EcoRl 7.4, Sail and selection for lack of expression of the lacZ gene (white colonies). Closed 62, Accl 0.8, Bamm 12.5, Pstl 1.5, Sstl 2.6, Hindm 29. bars, AT5BIVA extract-treated; hatched bars, MRC5V1 extract-treated. Downloaded from https://academic.oup.com/nar/article/18/21/6205/2388438 by DeepDyve user on 14 August 2020 6210 Nucleic Acids Research, Vol. 18, No. 21 topological location of these sites, may influence their response. deficiency (34). However, we cannot at present comment on the Thus Sail, BamHl- and //mdTII-breaks with 5' 4-base overhangs relevance of our data to the genetic defect in the A-T disorder; were rejoined somewhat more efficiently than the 3' -base extracts from only one A-T cell line were used and it will be overhangs given by Pstl- and Ssfl-breaks. These data for the necessary to test extracts from the cells of other patients before mammalian nuclear extracts contrast with the profile found for generalizations can be made. T4 ligase where, under the same reaction conditions, the rejoin efficiencies were: HitvttR 2: Sstl = Pstl > EcoW > BamHl ACKNOWLEDGEMENTS > Sail > Accl (data not shown). We have found, in agreement with our previous cellular studies This study was supported in part by the Commission of the (5), that extracts from transformed A-T and normal cells do not European Communities contract B16-E-144-UK. differ in rejoin efficiency. These results are in keeping with measurements showing similar DNA ligase activities in cells of REFERENCES these genotypes (28). We have now shown, in addition, that extracts from early-passage (diploid) human cells give similar 1. Brown.T.C. and JiricnyJ. (1987) Ceil, 50, 945-950. 2. Dolan.M.E., Oplinger.M. and Pegg.A.E. (1988) Mutation Res., 193, relative efficiencies of rejoin at the different sites tested (Fig.4). 131-137. 3. Moriya.M., Takeshita.M., Johnson.F., Peden.K., Will.S. and Grollman.A.P. Rejoin fidelity (1988) Proc. Natl. Acad. Sd. USA, 85, 1586-1589. 4. Ellison.K.S., Dogliotti.E. and EssigmanJ.M. (1989) Mutation Res., 220, While nuclear extracts from the different cell types have similar 93-100 . rejoin efficiencies, we have found differences in the fidelity with 5. ThackerJ. (1989) Mutation Res. 220, 187-204. which these extracts rejoin broken DNA at certain sites. Although 6. Frankenberg.D., Frankenberg-Schwager.M., Bloecher.D. and Harbich.R. it is known that bacteria will mis-rejoin a proportion of broken (1981) Radial. Res., 88, 524-532. plasmid molecules (29; see also Table 2), high mis-rejoin 7. Sargentini.N.J. and Smhh.K.C. (1986) Radial. Res., 107, 58-72 . 8. Bryant.P.E. (1985) Int. J. Radial. Biol., 48, 55-60 . frequencies were only found after human nuclear extract 9. Kemp,L., Sedgwkk.S.G. and Jeggo.P.A. (1984) Mutation Res., 132, treatment. Extracts from the nuclei of the AT5BFVA cell line, 189-196. in particular, give a highly significant increase in the loss of lacZ 10. Wlodek.D. and Hittelman.W.N. (1987) Radiat. Res., 112, 146-155. gene function when rejoining certain break-sites. It is notable that 11. Stamato.T.D., DipatriA and GiacctaA (1988) Radiat. Res., 115,325-333. those break-sites (especially EcoRl and Accl) which yield a low 12. Taykn\A.M.R., Hamden,D.G., Ariett,C.F., Harcourt,S.A., Lehmann.A.R., Stevens.S. and Bridges.B.A. (1975) Nature (London) 258, 427-429. fidelity of rejoining with the A-T extract also have relatively low 13. Cox,R., and Masson.W.K. (1980) Int. J. Radiat. Biol. 38, 575-576. rejoin efficiencies. This suggests that where breaks are more 14. Boder,E. (1985) In Gatti.R.A. and Swift.M. (pis.) Ataxia-Telangiectasia: difficult to rejoin, some extracts will act on the broken DNA Genetics, Neuropathology and Immunology of a Degenerative Disease of to remove sequence. Childhood. Liss, New York, pp. 1-63 . 15. McKinnon.P.J. (1987) Human Genet. 75, 197-208. However, the two 3' 4-base overhang sites show similar 16. ThackerJ. (1989) Bioessays, 11, 58-62 . 'intermediate' rejoin efficiencies but give quite different mis- 17. Huschtscha.L.I., and HoUiday.R. (1983)7. CellSd., 63, 77-99 . rejoined fractions (Table 5), in one case (Sstf) similar to that for 18. Day.R.S., ZioUcowskJ.C.HJ., Scudiero.D.A., Meyer.S.A., Lubiniecki,A.S., the site having the highest rejoin efficiency (Sail). It is notable Giradi.A.J., Galloway.S.M. and Bynum.G.D. (1980) Nature (London), 288, that these two endonucleases generate the same sequence at the 724-727 . 19. Cox.R. and Masson,W.K. (1974) Int. J. Radiat. Biol., 26, 193-196. overhang (5' TCGA or 3' AGCT), while the reverse of this 20. Lopez,B., Rousset.S. and Coppey.J. (1987) Nucleic Acids Res., 15, sequence (HindHl break site, 5' AGCT) does not show high 5643-5655. fidelity rejoining. This result suggests that the nuclear extract 21 . Sambrook,J., Fritsch.E.F., and Maniatis.T. (1989) Molecular Cloning-A may contain some very specific means of ensuring that certain Laboratory Manual, 2nd. Edn. Cold Spring Harbor Lab. Press, New York. sequences are rejoined faithfully. 22. Feinberg.A.P. and Vogelstein.B. (1983) Anal. Biochem. 132, 6-13 . 23. Bethesda Research Laboratories (1986) Focus, 8, 9. It is known that broken DNA ends act as sites of action or 24. Hanahan.D. (1985) In Glover.D.M. (ed.) DNA Cloning, IRL Press, Oxford., recognition for enzymes which degrade or delete sequence. Vol.1, pp. 109-135. Simple mechanisms for lack of fidelity, such as an increased 25. Cohen.S.N., Chang.A.C.Y. and Hsu.L. (1972) Proc. Natl. Acad. Sd. USA 69, 2110-2114. nuclease activity in the A-T extract, were not revealed by an 26. Conley.E.C. and SaundersJ.R. (1984) MoL Gen. Genet., 194,211-218. increase in the extent of degradation of DNA on gels after 27. Dugaiczyk.A., Boycr.H.W. and Goodman,H.M. (1975)7. Mol. Biol., 96, treatment with A-T extracts, and no differences between cell lines 171-184. were found in simple nuclease assays (data not shown). 28. Willis.A.E. and Lindahl.T. (1987) Nature (London), 325, 355-357. Alternatively, there may be lack of 'protection' of DNA ends 29. Thompson.R. and Achtman.M. (1979) Mol. Gen. Genet., 169, 49-57 . 30. AkroydJ., and Symonds.N. (1986) Gene, 49, 273-282. in the A-T extract (5), allowing degradative processes access to 31. de Vries.E., van Driel,W., Bergsma.W.G., Amberg.A.C. and van der the DNA. Such protection would arise minimally from the weak Vliet.P. (1989) / . Mol. Biol., 208, 65-78 . hydrogen-bonded association of the broken ends, more strongly 32. Conley.E.C, Saunders,V.A., Jackson.V., and SaundersJ.R. (1986) Nudeic from proteins which bind to the ends, and maximally from rapid Acids Res., 14, 8919-8932. ligation. DNA binding proteins protecting broken ends have been 33. Pfeiffer,P. and Vielmetter.W. (1988) Nucleic Acids Res., 16, 907-924. 34. Malyim,B.A., Blackwell.T.K., Fulop.G.M., Rathbun,G.A., Furley.A.J.W., identified, for example, in infectivity studies of bacteriophage Ferrier.P., Heinke,L.B., Phillips.R.A., Yancopoulos.G.D. and Alt.F.W. Mu (30) and in HeLa cells infected with adenovirus (31). (1988) Cell, 54, 453-460. Additionally, in studies of break rejoining in other organisms, such as Escherichia coli (32) and extracts from Xenopus laevis eggs (33), it has been suggested that 'accessory' (DNA binding) proteins are involved in the rejoining process. A defect in a complex that holds free ends together after endonucleolytic breakage has also been proposed in studies of aberrant immune gene rearrangements in mice suffering severe combined immune

Journal

Nucleic Acids Research – Oxford University Press

Published: Nov 21, 1990

There are no references for this article.