Alkaline Sucrose Gradient Sedimentation Analysis Research Articles

Metabolic fate of [3H]1-beta-D-arabinofuranosyl-5-[(E)-2-bromovinyl]uracil in herpes simplex virus type 1-infected cells.

The metabolic fate of 1-beta-D-arabinofuranosyl-5-[(E)-2-bromovinyl]uracil (BV-araU) in herpes simplex virus type 1-infected cells was studied using tritium-labeled BV-araU. [3H]BV-araU was selectively taken-up by infected cells. Approximately 10% of the total uptake of [3H]BV-araU was recovered from the acid-insoluble fraction at any time post-infection. Both cellular uptake of [3H]BV-araU and its incorporation into the acid-insoluble fraction increased with increasing incubation time through 8 hr post-infection. Uptake of [3H]BV-araU and its incorporation into the acid-insoluble fraction also increased proportionally to the duration of exposure to [3H]BV-araU. An alkaline sucrose gradient sedimentation analysis revealed that the radioactive DNA obtained from cells pulse-labeled with [3H]BV-araU were small DNA fragments which remained at the top following a chasing period in isotope-free medium, whereas that pulse-labeled with [3H]thymidine was chased to a fraction of high molecular weight DNA. Nuclease P1 digestion reduced 99% of the [3H]BV-araU-labeled DNA extracted from infected cells to a low molecular weight. Following digestion of [3H]BV-araU-labeled DNA with micrococcal nuclease and spleen exonuclease, all of the radioactivity was recovered as [3H]BV-araU 3'-monophosphate. Thus, BV-araU strongly inhibits the elongation of viral DNA strands as demonstrated by the alkaline sucrose gradient sedimentation analysis, whereas at least a portion of the [3H]BV-araU is incorporated inside viral DNA strands in infected cells.

The mechanism of recA polA lethality: suppression by RecA-independent recombination repair activated by the lexA(Def) mutation in Escherichia coli.

The mechanism of recA polA lethality in Escherichia coli has been studied. Complementation tests have indicated that both the 5'-->3' exonuclease and the polymerization activities of DNA polymerase I are essential for viability in the absence of RecA protein, whereas the viability and DNA replication of DNA polymerase I-defective cells depend on the recombinase activity of RecA. An alkaline sucrose gradient sedimentation analysis has indicated that RecA has only a minor role in Okazaki fragment processing. Double-strand break repair is proposed for the major role of RecA in the absence of DNA polymerase I. The lexA(Def)::Tn5 mutation has previously been shown to suppress the temperature-sensitive growth of recA200(Ts) polA25::spc mutants. The lexA(Def) mutation can alleviate impaired DNA synthesis in the recA200(Ts) polA25::spc mutant cells at the restrictive temperature. recF+ is essential for this suppression pathway. recJ and recQ mutations have minor but significant adverse effects on the suppression. The recA200(Ts) allele in the recA200(Ts) polA25::spc lexA(Def) mutant can be replaced by delta recA, indicating that the lexA(Def)-induced suppression is RecA independent. lexA(Def) reduces the sensitivity of delta recA polA25::spc cells to UV damage by approximately 10(4)-fold. lexA(Def) also restores P1 transduction proficiency to the delta recA polA25::spc mutant to a level that is 7.3% of the recA+ wild type. These results suggest that lexA(Def) activates a RecA-independent, RecF-dependent recombination repair pathway that suppresses the defect in DNA replication in recA polA double mutants.

Enhancing effect of heterocyclic amines and ß-carbolines on UV or chemically induced mutagenesis in E. coli

Most heterocyclic amines and ß-carbolines — harman, norharman, harmine, harmaline — enhanced UVC (254 nm) induced mutagenesis without microsomal activation in E. coli B/r WP2. 3-Amino-1,4-dimethyl-5 H-pyrido[4,3- b]indole (Trp-P-1) was most effective and increased UVAB (295–400 nm) induced mutations as well as UVC induced ones. Trp-P-1 enhanced the frequencies of mutations induced by not only UV but also 4-nitroquinoline-1-oxide (4NQO) or 2-(2-furyl)-3-(5-nitro-2-furyl)acrylamide (AF2), while it showed little effect on N-methyl- N′-nitro- N-nitrosoguanidine (MNNG) or γ-ray induced mutagenesis. Trp-P-1 decreased the survival of UVC irradiated cells of CM571 recA. However, these effects of Trp-P-1 on UVC induced mutagenesis and lethality were not observed in WP2s uvrA which is excisioon repair deficient. The alkaline sucrose gradient sedimentation analysis demonstrated that Trp-P-1 blocked the incision step in DNA excision repair. Further, pretreatment with Trp-P-1 before UVC irradiation showed no effect on UVC induced mutagenesis. Similar effects were also seen in the case of harman or norharman. These results suggest that heterocyclic amines and ß-carbolines inhibit DNA excision repair directly or indirectly, thus enhancing UV or chemically induced mutagenesis.

Differences in DNA damage induced by mutagenic and nonmutagenic 4-aminoazobenzene derivatives in Escherichia coli

DNA lesions produced in Escherichia coli AB2500 (uvrA) exposed to the carcinogen N-hydroxy-3-methoxy-4-aminoazobenzene (N-OH-3-MeO-AAB) or the noncarcinogen N-hydroxy-2-methoxy-4-aminoazobenzene (N-OH-2-MeO-AAB) were investigated by alkaline sucrose gradient sedimentation and 32P-postlabeling analysis. Alkali-labile sites appeared to be formed equally in cells treated with both aminoazobenzene derivatives. 32P-Postlabeling analysis revealed that the 3-MeO-AAB-DNA adduct level was 25-fold higher than that for 2-MeO-AAB-DNA adducts. In addition to major adducts, 4 minor spots were detected in N-OH-3-MeO-AAB-treated cells, while only one major adduct was found in N-OH-2-MeO-AAB-treated cells. The mutagenecities and cytotoxicities were also determined with E. coli with different repair capacities; we found that repair of 3-MeO-AAB damages is strongly dependent on the UVR repair system. Moreover, N-OH-3-MeO-AAB, but not N-OH-2-MeO-AAB, could induce recA and umuC gene expression, which was higher in uvrA strains than in the wild type.

The role of excision repair in the removal of transient benzo[a]pyrene-induced DNA lesions in Chinese hamster ovary cells

In Chinese hamster ovary (CHO) cells, benzo[a]pyrene induces both persistent and transient lesions that are detected by alkaline sucrose gradient sedimentation analysis (ASG sites). The transient lesions disappear within 15 min while the persistent lesions can be detected for several hours following treatment. Although the persistent ASG sites are believed to be repaired by excision repair, the process responsible for the disappearance of the transient ASG sites is unknown. To determine the contribution of excision repair to the removal of these transient lesions, CHO cells were treated with benzo[a]pyrene (B(a)P) in the presence of the inhibitors of excision repair, araC and novobiocin. The results indicate that: (1) araC inhibits the removal of persistent, but not the transient B(a)P-induced ASG sites; (2) novobiocin, a putative inhibitor of the incision step of DNA excision repair, reduced the number of lesions detected immediately following treatment, indicating that many of these lesions may represent single-strand discontinuities generated during repair; and (3) the lesions detected in the presence of novobiocin disappear rapidly following treatment. Based on these results, we concluded that B(a)P-induced transient ASG sites are repaired by a process other than excision repair.

Mechanism of Killing Chinese Hamster Ovary Cells Heated in G 1 : Effects on DNA Synthesis and Blocking in G 2

To determine where in the cell cycle Chinese hamster ovary cells die following heating in G1, a mild hyperthermia treatment, i.e., 10 or 11.5 min at 45.5 degrees C, resulting in 40-50% cell kill was used. After a 7-14-h delay in G1, the cells heated in G1 eventually entered S phase and replicated all their DNA. Both an autoradiographic analysis with tritiated thymidine and a bromodeoxyuridine-propidium iodide bivariate analysis by flow cytometry revealed that both clonogenic and nonclonogenic cells were delayed in progression through S phase for at least 4 h. Then they completed replication of all their DNA and entered G2. Alkaline sucrose gradient sedimentation analysis revealed that these heated cells could complete replicon elongation into cluster-sized molecules of 120-160 S which persisted for 2-12 h after heating. However, further replicon elongation into multicluster-sized molecules greater than 160 S required an additional 12 h in heated cells compared to the 4 h needed in unheated control cells. Our results when compared with the literature suggest that when G1 cells are heated to a survival level of about 50%, the nonclonogenic cells recover from a long delay in G1, traverse S at a reduced rate, and then die either in G2 or as multinucleated cells after an aberrant division.

Apurinic site induction in the DNA of cells heated at hyperthermic temperatures.

The induction of DNA damage in cells heated at hyperthermic (43-48 degrees C) temperatures was determined by alkaline filter elution and alkaline sucrose gradient-sedimentation analysis of cell DNA denatured at pH 13.0. A class of DNA lesion which converted to strand breaks during denaturation of DNA at pH 13.0 was produced randomly throughout the cell DNA at temperatures as low as 43 degrees C. Induction of this lesion occurred with a T0 of 90 and 10 min at 45 and 48 degrees C, respectively. We estimate that these pH 13.0-detectable DNA lesions are produced in the cell DNA with a frequency of approximately 75 and 660 per min of heating at 45 and 48 degrees C, respectively. Since the lesions were quantitatively converted to DNA strand breaks at pH 13.0 with a half-time of 30 min, or less, we suggest that these pH 13.0-detectable DNA lesions are heat-induced, abasic DNA sites. The induction of these lesions does not appear to be directly involved in the initial heat-induced inhibition of DNA synthesis. The presence of these lesions cannot be excluded as an explanation for the long-term inhibition of replicon initiated in heated cells.

Involvement of DNA polymerases in the repair of DNA damage by benzo[ a]pyrene in cultured Chinese hamster cells

Mechanisms for induction of single-strand scissions in DNA by S9-activated benzo[ a]pyrene (B[a]P) and their repair in cultured Chinese hamster V79 cells were investigated with inhibitors of DNA-repair synthesis using alkaline sucrose gradient sedimentation analysis. The marked induction of single-strand scissions in DNA was observed following 3 h treatment of V79 cells with 5 μg/ml of B[a]P. These DNA lesions were repaired to the control level within 4 h after removal of B[a]P. The simultaneous addition of inhibitors of DNA-repair synthesis, 1-β- D-arabinofuranosylcytosine (araC) plus hydroxyurea with B[a]P did not increase the formation of DNA single-strand scissions. When these inhibitors were added after removal of B[a]P, however, they significantly blocked the rejoining of DNA-strand scissions. On the other hand, when aphidicolin, a specific inhibitor of DNA polymerase α, was used instead of araC, a partial inhibition of the rejoining was observed, and further addition of 2′,3′-dideoxythymidine, an inhibitor of DNA polymerase β, augmented the inhibitory effect. These results indicate that B[a]P-induced single-strand scissions of DNA in V79 cells could be repaired mostly by excision repair which involved DNA polymerase α and a non-α polymerase, presumably polymerase β.

Inhibition by ganciclovir of cell growth and DNA synthesis of cells biochemically transformed with herpesvirus genetic information.

The ability of LM cells, thymidine kinase-deficient LM cells (LMTK-), and LMTK- cells transformed to the LMTK+ phenotype by herpes simplex virus type 1 genetic information (LH7 cells) to anabolize the acyclovir congener ganciclovir was examined. About 50-fold more ganciclovir triphosphate was produced by LH7 cells than by either LM or LMTK- cells. Growth inhibition studies indicated that 180 and 120 microM ganciclovir were required to achieve 50% growth inhibition of LM and LMTK- cells, respectively; only 0.07 microM ganciclovir was necessary to achieve 50% inhibition of LH7 cells. DNA synthesis in the transformed cells was significantly reduced by ganciclovir treatment, whereas ganciclovir had little effect on DNA synthesis in the nontransformed cells. Alkaline sucrose gradient sedimentation analysis of transformed cellular DNA indicated that LH7 DNA synthesized in the presence of ganciclovir chased into mature DNA. Both LM and LH7 DNA synthesized in the presence of ganciclovir exhibited a concentration-dependent reduction in the rate of elongation into mature DNA. Finally, [14C]ganciclovir was incorporated internally into the growing chains of LH7 cells.

Characterization of DNA replication at a restrictive temperature in a mouse DNA temperature-sensitive mutant, tsFT20 strain, containing heat-labile DNA polymerase alpha activity.

tsFT20 cells derived from a mouse mammary carcinoma cell line FM3A have temperature-sensitive DNA polymerase alpha activity (Murakami, Y., Yasuda, H., Miyazawa, H., Hanaoka, F., and Yamada, M. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 1761-1765). DNA replication in tsFT20 cells at the restrictive temperature (39 degrees C) has been characterized in detail. DNA-synthesizing ability of these cells was measured by [3H] thymidine incorporation and autoradiography. The incorporation of [3H]thymidine decreased rapidly after temperature shift-up, and the incorporation was less than 20% of the initial level after 4 h at 39 degrees C. The rapid decrease correlated well with the decrease in the grain number in the individual nucleus but not with the number of cells with labeled nuclei. Alkaline sucrose gradient sedimentation analysis and DNA fiber autoradiography revealed that DNA chain elongation proceeded normally within a replicon in the temperature-sensitive cells incubated at the restrictive temperature and the DNA elongation rate did not change during the incubation at the restrictive temperature up to at least 6 h. On the other hand, the maturation of replicon-sized DNA to higher molecular weight DNA was retarded or inhibited in the temperature-sensitive cells at the restrictive temperature. The analysis of the center to center distance between replicons by DNA fiber autoradiography revealed that the frequency of replicon initiation decreased in tsFT20 cells at 39 degrees C.

Repair of DNA single-strand breaks in radiation-sensitive mutants of mouse cells

The radiation-sensitive mutant M10 of mouse lymphoma L5178Y cells was examined for its ability to rejoin DNA single-strand breaks induced by γ-rays. The alkaline sucrose gradient sedimentation analysis revealed that M10 cells repaired single-strand breaks but simultaneously produced increasing amounts of small DNA fragments with time of postirradiation incubation, something which was not observed in L5178Y cells. Since small fragments did not appear in M10 cells irradiated at room temperature, DNA fragmentation may result from cold treatment during irradiation followed by incubation at 37°C. This indicates that the cold susceptibility is characteristic of M10 cells and is not related to radiation sensitivity of this mutant. This conclusion is supported by the finding that no DNA degradation takes place after cold treatment with a subsequent incubation in the other radiosensitive mutant LX830 that belongs to the same complementation group as M10.

DNA lesions, chromosomal aberrations and G 2 delay in CHO cells cultured in medium containing bromo- or chloro-deoxyuridine

Experiments have been performed to investigate whether BrdUrd- and CldUrd-substituted DNA contains lesions causing a delay in cell-cycle progression and induction of chromosomal aberrations. The presence of lesions has been determined directly by alkaline sucrose gradient and nucleoid sedimentation analysis and indirectly by screening for induced chromosomal aberrations. The influence of inhibitors of DNA repair (caffeine and 3-aminobenzamide) or DNA synthesis (hydroxyurea) on the frequencies of such aberrations has been estimated. It is found that BrdUrd and CldUrd are cytotoxic when present in DNA. No randomly located DNA breaks could be detected under neutral conditions, but BrdUrd-substituted DNA was found to contain numerous alkali-labile sites. CldUrd at high concentrations causes G 2 delay, similar to the action of known DNA-damaging agents. The extent of delay depends on the pattern of incorporation of the analogue, i.e., incorporation for two cell cycles causes the longest delay, growth for 12 h in CldUrd followed by 12 h in dThd-containing medium causes a lesser delay and the delay is not significant when the cells are incubated in the analogue for only 12 h prior to fixation. Numerous chromatid type aberrations are present in cells incubated at the highest CldUrd concentration, and their induction follows the pattern of induction of G 2 delay, indicating the sharing of a common lesion. Caffeine, 3-aminobenzamide and hydroxyurea increase the number of chromosomal aberrations when added 2 h before fixation. The significance of these results is discussed.

Transient DNA lesions induced by benzo[ a]pyrene in Chinese hamster ovary cells

Alkaline sucrose gradient sedimentation analysis was used to detect DNA lesions induced by benzo[ a]pyrene (B(a)P) in Chinese hamster ovary cells. The number of lesions detected immediately following treatment with 10 −4 M B(a)P was related directly to the duration of treatment. When treated cells were incubated in a B(a)P-free medium, the majority of lesions disappeared rapidly and could no longer be detected 15 min following treatment. These data indicate that a population of B(a)P-induced DNA lesions may be removed by a rapid DNA-repair process. The transient nature of such lesions should be considered when assays for DNA damage or repair are designed and interpreted.

Purification and characterization of a factor stimulating DNA polymerase alpha activity from mouse FM3A cells.

A protein factor which stimulates DNA polymerase alpha activity on heat-denatured DNA has been purified from mouse FM3A cells. The final preparation had a specific activity of 43,000 units/mg protein and lacked detectable DNA polymerase, RNA polymerase, DNA-dependent- and independent ATPase, exo- and endodeoxyribonuclease and phosphatase activities. The stimulating factor sedimented at 2.9S in a glycerol gradient. Sodium dodecyl sulfate polyacrylamide gel electrophoresis of the glycerol gradient fraction revealed the presence of a major band of 36,000 daltons, the amount of which corresponded well with the level of stimulating activity. The stimulation by the factor was specific for heat-denatured DNA, and a little or no stimulation was observed with native DNA, ribo- and deoxyribohomopolymers and single stranded circular DNA. Alkaline sucrose gradient sedimentation analysis of the reaction products revealed that newly synthesized DNA was covalently linked to the termini of heat-denatured DNA. The average chain length of the elongated span determined by the digestion with micrococcal nuclease and phosphodiesterase II, did not differ between in the presence and absence of the stimulating factor, suggesting that the stimulation by the factor was due to the increase in the initiation frequency of DNA synthesis from the 3'-hydroxyl terminus of heat-denatured DNA.

Chinese hamster ovary mutant UV-1 is hypomutable and defective in a postreplication recovery process.

CHO-UV-1 is a mutant of the Chinese hamster cell CHO-K1 hypersensitive to killing by ultraviolet light but with normal resistance to X-ray. It is also hypersensitive to killing by ethyl methane sulfonate. Hybrid clones formed bu fusing UV-1 and Chinese hamster lung cells display the normal ultraviolet resistance of the latter. The sensitive phenotype behaves, therefore, in a genetically recessive manner. Ultraviolet sensitivity of UV-1 is not associated with a deficiency in excision repair. Alkaline sucrose gradient sedimentation analysis of nascent DNA from ultraviolet-irradiated cells reveals that UV-1 is, however, markedly deficient in postreplication recovery. Furthermore, UV-1 has a lower rate of induced mutation to 6-thioguanine resistance than does the parental cell when treated with ultraviolet light or ethyl methane sulfonate. These results suggest that the phenotype of UV-1 is due to a mutation in a form of postreplication recovery which in normal cells is error prone.

Repair replication of viral and cellular DNA in herpes simplex virus type 2-infected human embryonic and xeroderma pigmentosum cells

When human embryonic fibroblasts were infected with herpes simplex virus type 2 (HSV-2) at a multiplicity of 5 PFU/cell, repair replication of cellular DNA was detectable by 3–5 hr postinfection (p.i.). Between 12 and 14 hr p.i., predominantly repair replication of cellular DNA was noted but some repair and semiconservative replication of viral DNA occurred. Almost all DNA synthesis between 22 and 24 hr p.i. was derived from repair synthesis of cellular and viral DNA. Alkaline sucrose gradient sedimentation analysis revealed that cellular DNA of HSV-2-infected cells suffered extensive strand breaks by 14 hr p.i. Significant repair replication of cellular DNA was induced by HSV-2 even in the presence of 100 μg/ml of phosphonoacetic acid which inhibits viral DNA synthesis to undetectable levels. Repair replication of cellular and viral DNA was also detected in infected xeroderma pigmentosum (XP12BE) cells. The results suggest that HSV-2 induced repairable damage in cellular DNA without viral DNA replication and that native HSV-2 DNA which accumulated in the infected cells had a repairable structure (probably strand breaks) that did not require incision for repair.

Enhanced capacity of DNA repair in human cytomegalovirus-infected cells.

Plaque formation in Vero cells by UV-irradiated herpes simplex virus was enhanced by infection with human cytomegalovirus (HCMV), UV irradiation, or treatment with methylmethanesulfonate. Preinfection of Vero cells with HCMV enhanced reactivation of UV-irradiated herpes simplex virus more significantly than did treatment with UV or methylmethanesulfonate alone. A similar enhancement by HCMV was observed in human embryonic fibroblasts, but not in xeroderma pigmentosum (XP12BE) cells. It was also found that HCMV infection enhanced hydroxyurea-resistant DNA synthesis induced by UV light or methylmethanesulfonate. Alkaline sucrose gradient sedimentation analysis revealed an enhanced rate of synthesis of all size classes of DNA in UV-irradiated HCMV-infected Vero cells. However, HCMV infection did not induce repairable lesions in cellular DNA and did not significantly inhibit host cell DNA synthesis, unlike UV or methylmethanesulfonate. These results indicate that HCMV enhanced DNA repair capacity in the host cells without producing detectable lesions in cellular DNA and without inhibiting DNA synthesis. This repair appeared to be error proof for UV-damaged herpes simplex virus DNA when tested with herpes simplex virus thymidine kinase-negative mutants.

Inhibition by acyclovir of cell growth and DNA synthesis of cells biochemically transformed with Herpesvirus genetic information

Thymidine kinase-deficient LM cells (LMTK −-) biochemically transformed to the TK + phenotype with herpes simplex virus genetic information showed an increased uptake of and ability to phosphorylate the acyclic nucleoside analog 9-(2-hydroxyethoxymethyl)guanine (acyclovir, acycloguanosine, acyclo-Guo). In growth inhibition studies the TK + transformants were much more sensitive to inhibition with acyclovir than the untransformed cells (13- to 90-fold more sensitive). The synthesis of DNA in the transformed cells was significantly reduced by acyclovir treatment, whereas acyclovir had little effect on the DNA synthesis of the untransformed cells. Alkaline sucrose gradient sedimentation analysis of cellular DNA synthesized in the presence of acyclovir showed that, in contrast to untreated untransformed cells, the DNA newly synthesized by transformed cells was considerably smaller in size. In pulse-chase experiments the small fragments of DNA synthesized in the presence of acyclo-Guo were not chased to high molecular weight DNA. Finally, acyclo-Guo was shown to be incorporated terminally at 3′-ends of growing DNA chains in replicating cells.

Effect of hyperthermia on DNA single-strand breaks induced by bleomycin in Hela cells.

The effect of hyperthermia (43 approximately 45 degrees) on DNA single-strand breaks induced by bleomycin was examined in HeLa cells by the alkaline sucrose gradient sedimentation analysis. When HeLa cells were exposed to a temperature of 43 degrees approximately 45 degrees for 1 hr, only a small number of DNA single-strand breaks than bleomycin treatment at 37 degrees. The DNA strand breaks induced by bleomycin were almost completely repaired by post-treatment incubation of the cells in the control medium at 37 degrees. On the contrary, the repairing capacity of the cells was markedly inhibited by post-treatment incubation at 43 degrees, and this inhibition of cell repair seemed irreversible because no repair was found after an additional incubation of the cells at 37 degrees for 1 hr. These results support the possibility that the mechanism of synergism between the effect of bleomycin and hyperthermia may be related to the inhibition of DNA repair.

Replication of vaccinia DNA in mouse L cells III. Intracellular forms of viral DNA

Parental and replicating vaccinia DNA molecules, labeled with [ 3H]thymidine or [ 14C]thymidine, present in infected L cells were analyzed by sedimentation in neutral and alkaline sucrose gradients. Alkaline sucrose gradient sedimentation analysis of labeled parental genomes present in infected cells showed that: (a) Such genomes were not degraded to acid-soluble products during the infection cycle; (b) cell-associated, cross-linked parental molecules (102 and 90–92 S) were “nicked”; parental DNA molecules sedimenting at 70–72 S were detected, but further nicking or degradation did not occur; and (c) molecules sedimenting at 90–92 S appeared to accumulate in the cytoplasm of infected cells and could serve as the templates for semiconservative DNA replication. When analyzed in neutral sucrose gradients, about 90% of viral DNA molecules labeled from 1 to 2 hr postinfection were associated with large aggregates or complexes which pelleted under the conditions of analysis used in these studies. With time (2–3 hr) after infection, viral DNA molecules could be dissociated from such aggregates and resolved by sedimentation in neutral sucrose gradients. The dissociation of labeled viral DNA molecules from complexes required continuous protein synthesis. Viral DNA could be released from complexes by treatment with alkali or digestion with S-1 nuclease, but not with RNase or Pronase. The results suggest that single-stranded (ss) DNA regions per se or proteins having affinity for ssDNA may bind the replicating vaccinia DNA molecules together in complexes.