AGE-RELATED ALTERATIONS IN HUMAN CHROMOSOME COMPOSITION AND DNA CONTENT IN VITRO DURING SENESCENCE

Different theories of ageing involving somatic mutations, error catastrophe, compensation and repair, and programmed ageing were subjected to analysis for their feasibility from experimental data. Due to the relative difficulty of carrying out longitudinal studies in human subjects in vivo and the long periods involved, most of these experiments dealt with in vitro systems. I. Fibroblast cells in culture were found to be ideal materials for demonstrating senescence for a particular species and at the terminal end the cultures showed certain changes associated with age. II. It appears that the ideas regarding the alterations in DNA content at the terminal stages of the culture or otherwise are related to the tissues concerned and the duration of the cultured condition. III. The importance of DNA content specially related to the concept of stability of the DNA strand supports indirectly the error catastrophe theories. The rate of net DNA synthesis, the size of the replicon and duration of S phase is reported not to change during in vitro ageing. The regulation system of DNA replication may however change, but this alteration does not affect the DNA replication machinery. Following cell fusion studies it has been hypothesized that senescent human diploid fibroblasts contain a diffusible inhibitor which blocks cells at G₁ phase. Some immortal cell lines like HeLa and SV40-transformed cells would contain a dominant inducer that could override or inactivate the putative inhibitor, the nature of which is not yet clear. IV. DNA repair competence of fibroblast cultures is reported to decline near the end of in vitro life-span. However, it has been noted that the human skin fibroblasts from both young and old donors are equally proficient in repairing damage by UV light. V. The replication patterns of chromosomes from both senescent embryonic fibroblasts and early-passage adult skin fibroblasts were essentially identical. There were very few differences between the early-passage embryonic and adult skin cells. It was concluded that the terminal replication pattern of fibroblasts changes very little with cellular ageing. VI. A statistically significant increase in sister chromatid exchange frequency has been reported during the terminal part of fibroblast cultures. VII. Electron-microscopic studies have confirmed the increased organization of microfillaments into bundles in senescent cells. The presence of a rigid cytoskeletal structure may contribute in part to the inability of such cells to replicate. VIII. Age-related increase in nuclear proteins was attributed to accumulation of residual acid proteins. Densitometric analysis showed that histone HI was low in late-passage cells and H₄ fraction increased relatively at the terminal phase. IX. In contrast to age-matched controls, fibroblasts cultured from progeria and Werner's syndrome undergo significantly low population doubling. Metaphase plates from these patients demonstrated a much higher frequency of chromosomal abnor malities than normal fibroblasts. Frequency of sister chromatid exchange in cells from Fanconi's anaemia did not show any significant change as compared with control sets. X. Significantly lower Feulgen DNA values have been recorded from lymphocytes of the elderly as compared with younger ones, indicated by hypodiploidy as well as by the individual amounts of euchromatin and heterochromatin. However, later data from flow-cytometric measurements indicated that DNA content was the same for all age groups. XI. The UV-induced unscheduled DNA synthesis in lymphocytes of 80 to 90 year-old individuals was reduced as compared to younger persons. However, the rate of repair of DNA strand breaks is apparently constant in all the age groups. XII. Increase in aneuploidy from lymphocyte cultures of aged individuals has been recorded by many workers. XIII. Significantly lower titres of serine, threonine, histidine, ornithine and lysine have been observed in aged persons, and only the first three decreased equally in both sexes. Some of the amino acids were influenced by sex hormones. XIV. Study of the frequency of spontaneous sister chromatid exchanges showed that neither intra-individual variation between replicate cultures established from the same blood sample nor variation among samples from the same individual initiated at different times was significant. However, sensitivity to induced sister chromatid exchange is reported to increase with age.     

Delayed repair of DNA single-strand breaks does not increase cytogenetic damage

DNA damage and cytogenetic effects of ionizing radiation were investigated in Chinese hamster ovary (CHO) cells and unstimulated human peripheral blood lymphocytes. DNA damage and repair were analysed by alkaline elution under conditions that predominantly measured DNA single-strand breaks (ssb). X-radiation (2.5 Gy) induced ssb in both CHO cells and unstimulated lymphocytes, and the breaks were repaired within 30 and 90 min, respectively. This rapid repair was delayed by the poly(ADP-ribose) polymerase inhibitor, 3-aminobenzamide (3AB). The cytogenetic effects of the 3AB-induced delay in DNA repair were examined by analysing sister chromatid exchange (SCE) frequency in CHO cells and fragmentation of prematurely condensed chromosomes (PCC) in unstimulated human lymphocytes after 2.5 Gy of X-rays. Although 3AB delayed the rejoining of DNA ssb, this delay did not result in increased cytogenetic damage manifested as either SCE or fragmentation of PCC. These results indicate that the rapidly rejoining DNA ssb are not important in the production of chromosome damage.     

Chromosome instability associated with human alphoid DNA transfected into the Chinese hamster genome.

Repetitive DNA sequences have been implicated in the mediation of DNA rearrangement in mammalian cells. We have tested this hypothesis by using a dihydrofolate reductase (DHFR) expression vector into which candidate sequences were inserted. DHFR- Chinese hamster ovary (CHO) cells were transfected with this vector, the amplification of which was then selected for by methotrexate (MTX) exposure. Cells transfected with the vector alone (and resistant to 0.02 or 1.0 microM MTX) or with a poly(dG-dT) insert (and resistant to 0.05 or 1.0 microM MTX) showed little change in chromosome aberrations or sister chromatid exchange frequencies. In contrast, transfection of DHFR- CHO cells with a vector containing either of two distinct 0.34-kilobase human alphoid DNA segments (and selection to 0.05 to 10.0 microM MTX) showed an approximately 50% increase in chromosome number and marked changes in chromosome structure, including one or two dicentric or ring forms per cell. The sister chromatid exchange frequency also increased, to more than double the frequency of that in cells transfected without insert or those containing poly(dG-dT). In situ hybridization of one 0.34-kilobase insert in some cells suggested clustering of homologous sequences in structurally abnormal recipient CHO cell chromosomes. The approach described provides an introduction to a unique means for a coordinate molecular and cytological study of dynamic changes in chromosome structure.     

THE EFFECT OF ETHIDIUM BROMIDE ON MITOCHONDRIAL DNA SYNTHESIS AND MITOCHONDRIAL DNA STRUCTURE IN HELA CELLS

The synthesis of mitochondrial DNA (mDNA) in HeLa cells is selectively inhibited by relatively low concentrations of ethidium bromide. After exposure of cells to strongly inhibitory concentrations of the drug, the apparent superhelix density of mDNA is rapidly increased, as judged by its buoyant density in CsCl in the presence of ethidium bromide. Mitochondrial DNA synthesized in the presence of partially inhibitory concentrations of ethidium bromide is also altered in its buoyant density in the presence of the dye, but is more heterogeneous in this respect. However, the change in buoyant density of newly synthesized mDNA may be explained by changes in structure other than a change in superhelix density, as indicated by its increased resistance to digestion by pancreatic DNase.     

DNA fork displacement rates in synchronous aneuploid and diploid mammalian cells.

DNA fork displacement rates were measured in Chinese hamster ovary cells (CHO), human HeLa cells and human diploid fibroblasts. For CHO cells two independent techniques were used: one based on CsCl equilibrium density gradients and the other on 313 nm photolysis of incorporated bromodeoxyuridine (BrdUrd). Both methods indicated that there was no significant variation in fork displacement rates in CHO cells as they progressed through S phase. Asynchronous CHO cultures displayed the same average value (1.0 micron/min) and range of values as found in synchronous cells. In contrast, the rate of DNA fork displacement in HeLa cells, measured by the BrdUrd-313 nm method, increased continuously from 0.8 micron/min in early S to 2.5 micron/min in late S. For human diploid fibroblasts, in early S, the rate was approximately 0.7 micron/min and decreased to a minimum of 0.5 micron/min in mid S. The replication fork displacement rate then increased to a maximum of 0.9 micron/min in late S and declined again before the end of S phase. This pattern of DNA fork displacement rates roughly paralleled the overall thymidine incorporation rate and appears quite different from the patterns found for HeLa and CHO cells.     

DNA-mediated transfer of a human DNA repair gene that controls sister chromatid exchange.

The Chinese hamster cell line mutant EM9, which has a reduced ability to repair DNA strand breaks, is noted for its highly elevated frequency of sister chromatid exchange, a property shared with cells from individuals with Bloom's syndrome. The defect in EM9 cells was corrected by fusion hybridization with normal human fibroblasts and by transfection with DNA from hybrid cells. The transformants showed normalization of sister chromatid exchange frequency but incomplete correction of the repair defect in terms of chromosomal aberrations produced by 5-bromo-2'-deoxyuridine.     

Evidence for the formation of hybrid DNA during mitotic recombination in Chinese hamster cells

Direct evidence is provided for the formation of hybrid DNA during mitotic recombination in CHO cells. The cells were labeled for one round of replication in medium containing BUdR, so that the density of the DNA was heavy light (HL) and then returned to light medium. Further DNA synthesis, during either repair or chromosome replication, can only result in HL or fully light (LL) DNA; however, the formation of hybrid DNA as part of the process of recombinational repair will produce some fully heavy (HH) DNA.A small fraction of DNA containing regions of HH DNA has been detected on neutral CsCl gradients, and the amount of this DNA is increased by treatment of the cells with mitomycin C. Increasing doses of mitomycin C produce similar increases in both the amount of HH DNA and the frequency of sister chromatid exchanges measured cytologically. This correlation provides evidence that the HH DNA is hybrid DNA, formed as an intermediate in recombinational repair.     

Search for DNA interchange corresponding to sister chromatid exchanges in Chinese hamster ovary cells.

Chinese hamster ovary cells (CHO) grown for one cycle in bromodeoxyuridine (BrdU) contain a small amount (0.5%) of unusually dense double stranded DNA. This dense DNA has been previously interpreted as being bifilarly substituted with BrdU and hence evidence that sister chromatid exchange (SCE) formation proceeds via the Holliday model of recombination. However, the amount of this dense DNA is 100 times greater than that expected based on the SCE frequency in similarly cultured CHO cells, and it is not increased by treating the cells with mitomycin C. Moreover, contrary to expectations for bifilary substituted DNA, the amount of this dense DNA is not reduced by growing BrdU-labeled cells for a second cycle in TdR. Finally, DNA isolated from CHO cells contains a minor band (0.5%) with a density 0.025 gm/cc greater than that of the main band, whether or not BrdU has been incorporated. These results call into question the identification of this unusually dense DNA as bifilarly substituted and hence its previously postulated relationship to SCE formation.     

Chromatid damage induced by fluorescent light during G2 phase in normal and Gardner syndrome fibroblasts. Interpretation in terms of deficient DNA repair

Skin fibroblasts from Gardner syndrome (GS) compared with those from normal donors showed a significantly higher incidence of chromatid gaps and breaks following exposure to low-intensity, cool-white fluorescent light during G2 phase of the cell cycle. Considerable evidence supports the concept that chromatid gaps and breaks seen directly after exposure to DNA-damaging agents represent unrepaired DNA single- and double-strand breaks respectively. The changes in incidence of chromatid aberrations with time after light exposure are consistent with the sequence of events known to follow DNA damage and repair. Initially, the incidence of light-induced chromatid gaps was equivalent in GS and normal fibroblasts. In the normal cells, the chromatid gaps disappeared by 1 h post-exposure, presumably as a result of efficient repair of DNA single-strand breaks. In contrast, the incidence of gaps increased in GS cells by 0.5 h followed by a decrease at 1 h and concomitant increase in chromatid breaks. It appears from these findings that the increased incidence of chromatid damage in GS fibroblasts results from deficient repair of DNA single-strand breaks which arise from incomplete nucleotide excision of DNA damage during G2 phase.     

Repair of benzo[a]pyrene-initiated DNA damage in human cells requires activation of DNA polymerase alpha

Normal human fibroblasts treated with r-7,t-8-dihydroxy-t-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE) yielded DNA polymerase alpha with elevated levels of activity, incorporated [3H]thymidine as a function of unscheduled DNA synthesis, and exhibited restoration of normal DNA-strand length as a function of unscheduled DNA synthesis. Lipoprotein-deficient fibroblasts treated with BPDE did not show elevated levels of DNA polymerase alpha activity, exhibited minimal [3H]thymidine incorporation, and had fragmented DNA after 24 h of repair in the absence of lipoprotein or phosphatidylinositol supplementation. When DNA polymerase beta activity was inhibited, cells with normal lipoprotein uptake exhibited [3H]thymidine incorporation into BPDE-damaged DNA but did not show an increase in DNA-strand length. DNA polymerase alpha activity and [3H]thymidine incorporation in lipoprotein-deficient fibroblasts increased to normal levels when the cells were permeabilized and low-density lipoproteins or phosphatidylinositol were introduced into the cells. DNA polymerase alpha isolated from normal human fibroblasts, but not from lipoprotein-deficient fibroblasts, showed increased specific activity after the cells were treated with BPDE. When BPDE-treated lipoprotein-deficient fibroblasts were permeabilized and 32P-ATP was introduced into the cells along with lipoproteins, 32P-labeled DNA polymerase alpha with significantly increased specific activity was isolated from the cells. These data suggest that treatment of human fibroblasts with BPDE initiates unscheduled DNA synthesis, as a function of DNA excision repair, which is correlated with increased activity of DNA polymerase alpha, and that increased DNA polymerase alpha activity may be correlated with phosphorylation of the enzyme in a reaction that is stimulated by low-density lipoprotein or by the lipoprotein component, phosphatidylinositol.     

Microfluorometric analysis of cellular DNA following incorporation of bromodeoxyuridine.

Bromodeoxyuridine (BrdU) incorporation into cellular DNA has a differential effect on the cell-associated fluorescence of several DNA-specific dyes. After cells were treated with BrdU, flow microfluorometry was used to study the relative increase or decrease influorescence of stained cells. Bromodeoxyuridine incorporation into CHO cells increased the fluorescence of mithramycin-, olivomycin-, or chromomycin-stained cells, decreased that of propidium iodide-stained cells, and had little, if any, effect on the fluorescence of acriflavine Feulgen-stained cells. Changes in relative fluorescence of cell associated dyes are due to changes in the amounts of dye bound to cells with BrdU-substituted DNA. Colorimetric and absorbance measurement of DNA content showed that BrdU does not alter the diploid DNA content of CHO cells; however, BrdU induces perturbations in the distribution of cells about the cell cycle which cause an increase in average DNA content.     

Identification of RecQL1 as a Holliday junction processing enzyme in human cell lines

Homologous recombination provides an effective way to repair DNA double-strand breaks (DSBs) and is required for genetic recombination. During the process of homologous recombination, a heteroduplex DNA structure, or a ‘Holliday junction’ (HJ), is formed. The movement, or branch migration, of this junction is necessary for recombination to proceed correctly. In prokaryotes, the RecQ protein or the RuvA/RuvB protein complex can promote ATP-dependent branch migration of Holliday junctions. Much less is known about the processing of Holliday junctions in eukaryotes. Here, we identify RecQL1 as a predominant ATP-dependent, HJ branch migrator present in human nuclear extracts. A reduction in the level of RecQL1 induced by RNA interference in HeLa cells leads to an increase in sister chromatid exchange. We propose that RecQL1 is involved in the processing of Holliday junctions in human cells.     

Effects of ultraviolet irradiation on the rate and sequence of DNA replication in synchronized Chinese hamster cells.

The effects of ultraviolet light (UV) irradiation on the rate of DNA replication in synchronized Chinese hamster ovary (CHO) cells were investigated. A technique for measuring semiconservative DNA replication was employed that involved growing the cells in medium containing 5-bromodeoxyuridine and subsequently determining the amount of DNA that acquired hybrid buoyant density in CsCl density gradients. One of the advantages of this technique was that it allowed a characterization of the extent of DNA replication as well as rate after irradiation. It was found that while there was a dose-dependent reduction in the rate of DNA replication following UV-irradiation, doses of up to 10 J/m2 (which produce many dimers per replication) did not prevent the ultimate replication of the entire genome. Hence, we conclude that dimers cannot be absolute blocks to DNA replication. In order to account for the total genome replication observed, a mechanism must exist that allows genome replication between dimers. The degree of reduction in the rate of replication by UV was the same whether the cells were irradiated at the G1-S boundary or 1 h into S-phase. Previous work had shown that cells in early S-phase are considerably more sensitive to UV than cells at the G1-S boundary. Experiments specifically designed to test for reiterative replication showed that UV does not induce a second round of DNA replication within the same S-phase.     

Recombination and ligation of transfected DNA in CHO mutant EM9, which has high levels of sister chromatid exchange.

Transformation frequencies were measured in CHO mutant EM9 after transfection with intact or modified plasmid pSV2-gpt. The mutant and wild-type strain behaved similarly under all conditions except when homologous recombination was required to produce an intact plasmid. Therefore, the defect of the mutant which renders it slow in DNA strand break rejoining and high in sister chromatid exchange induction reduces its ability to recombine foreign DNA molecules.     

Measurement of DNA damage in mammalian cells using flow cytometry

A technique for the detection of DNA damage induced by radiation insult has been developed. Cells were lysed with a buffer containing 2 M sodium chloride to release the DNA in a supercoiled form, the nucleoid. These were stained with the DNA intercalating dye, ethidium bromide, and exposed to laser light within a flow cytometer. Scattered and fluorescent light was analyzed from the laser/nucleoid interaction following irradiation of viable cells with gamma rays. The addition of ethidium bromide to prepared nucleoids caused a reduction in scattered light due to condensation of the nucleoid. Irradiation of cells prior to nucleoid production and ethidium bromide treatment restricted this condensation and produced a dose-dependent increase in laser scatter. Nucleoids derived from human lymphocytes showed enhanced light scatter from 5 Gy, compared to Chinese hamster ovary (CHO) fibroblasts where doses above 10 Gy were required. Up to 30 Gy CHO nucleoids showed a dose-dependent reduction in the ethidium bromide fluorescence. This technique allows detection of altered light scattering and fluorescent behavior of nucleoids after cellular irradiation; these may be related to structural changes within the nucleus induced by the radiation. The use of flow cytometry compared to other methods allows a rapid analysis of nuclear damage within individual cells.     

Induction of sister chromatid exchanges by chemical mutagens and its possible relevance to DNA repair

Several chemical mutagens were found to induce sister chromatid exchanges in Chinese hamster chromosomes. Among them, effects of 4NQO and MMC were very similar to those of UV light in that the exchange frequency increased with increasing dose of chemicals and that it was markedly lowered in the presence of 1 mM caffeine during a post-treatment period. The frequency of proflavin-induced sister chromatid exchanges was also found to be dose dependent, but it was insensitive to the caffeine post-treatment. On the other hand, no appreciable increase was detected in the incidence of sister chromatid exchanges in MNNG-treated cells over a 100-fold range of variation in chemical dose. Caffeine by itself raised the exchange frequency only slightly over a control level. It was found that 4NQO and MMC exerted remarkable delayed effects on the exchange induction, whereas proflavin did not. This seems to suggest that the lesions caused by the former mutagens would be long-lived and repeatedly provoke sister chromatid exchanges. These data imply that there are several possible ways in which the initial DNA lesions ultimately lead to the formation of sister chromatid exchanges, and that at least UV-, 4NQO- and MMC-induced sister chromatid exchanges would have evolved through a caffeine sensitive repair process, probably related to a post-replication repair of DNA damage.     

Reinitiation of host DNA synthesis in senescent human diploid cells by infection with simian virus 40

Human diploid fibroblasts, TIG-1, cease to proliferate at about 60-62 population doubling level. In their senescent state used in this study, the percentage of nuclei labeled by [3H]thymidine for 48 h was around 1-2% in fresh medium containing 5-40% fetal bovine serum. The percentage of labelled nuclei increased up to 10-fold after infection with SV40. This increase reflects stimulation of cell DNA synthesis because: 1. The increase also occurred when ts A900 was used for infection at the non-permissive temperature, under these conditions viral DNA synthesis is inhibited; 2, the increase paralleled the stimulation of [3H]thymidine incorporation into DNA in a Hirt-precipitate fraction from SV40-infected cells. UV-irradiated SV40 had reduced ability to induce DNA synthesis. A viable deletion mutant of SV40, d1940, had almost the same activity to induce cell DNA synthesis as did wild-type SV40. Equilibrium density gradient centrifugation analysis of DNA labelled with 5-bromodeoxyuridine (BrdU) supported semiconservative replication rather than repair synthesis. We conclude that a considerable fraction of human diploid cells in a senescent population initiate host DNA replication by infection with SV40, although these cells cannot be stimulated with fetal bovine serum.     

Sister chromatid exchange induced by anti-herpes drugs.

The rate of sister chromatid exchange induced by several anti-herpes agents was measured to assess their potential mutagenicity. The agents--5-iodo-deoxyuridine (IDU), 5-trifluoromethyl-deoxyuridine (TFT), and [E]-5-(2-bromovinyl)-deoxyuridine (BVDU)--were incubated at various concentrations with human lymphocytes and fibroblasts, and that rate of sister chromatid exchanges was measured. In lymphocytes and fibroblasts BVDU and IDU did not induce exchange except at concentrations of 50 mg/l, while TFT increased the rate of exchange at a concentration of 0.5 mg/l. The rate of sister chromatid exchange is a sensitive index of chromosomal damage, and these findings provide information on the safety of some of the antiherpes agents tested. TFT increased the rate of exchange at a concentration that coincides with its minimal antiviral concentration, but BVDU did not induce exchange at therapeutic concentrations.