The specificity of different classes of ethylating agents toward various sites in RNA.

The alkyl products of neutral in vitro ethylation of TMV-RNA by [14C]diethyl sulfate, [14C]ethyl methanesulfonate, and [14C]ethylnitrosourea have been determined and found to differ significantly depending on the ethylating agent. Diethyl sulfate and ethyl methanesulfonate ethylate the bases of TMV-RNA in the following order: 7-ethylguanine greater than 1-ethyladenine, 3-ethylcytidine greater than 7-ethyladenine, 3-ethyladenine, O6-ethylguanosine, 3-ethylguanine. Ethyl methanesulfonate was more specific for the 7 position of guanine, and other derivatives were found in lesser amounts than with diethyl sulfate. Neither reagent caused the formation of detectable amounts (smaller than 0.26 percent) of 1-ethylguanine, 1,7-diethylguanine, N2-ethylguanine, N6-ethyladenine, N4-ethylcytidine, or 3-ethyluridine. Identified ethyl bases account for over 85% of the total radioactivity of [14C]ethyl methanesulfonate and [14C]diethyl sulfate treated TMV-RNA. Phosphate alkylation accounts for about 13 and 1%, respectively, In contrast, [14C]ethylnitrosourea-treated TMV-RNA, while reacting to a similar extent (15-70 ethyl groups/6400 nucleotides), is found to cause considerably more phosphate alkylation. Upon either U4A RNase or acid hydrolysis up to 60% of the radioactivity is found as volatile ethyl groupw in the form of [14C]ethanol, and a further 15% appears to be primarily ethyl phosphate and nucleosides with ethylated phosphate. Of the remaining radioactivity, half is found as O6-ethylguanosine, the major identified ethyl nucleoside. Other ethyl bases found in ethylnitrosourea-treated TMV-RNA are 7-ethylguanine greater than 1-ethyladenine, 3-ethyladenine, 7-ethyladenine, 3-ethylcytidine, and 3-ethylguanine. It appears that ethylnitrosourea preferentially alkylates oxygens, and that formation of phosphotriesters is by far the predominant chemical event. Since the number of ethyl groups introduced into TMV-RNA by ethylnitrosourea is similar to the number of lethal events, one may conclude that phosphate alkylation leads to loss of infectivity. None of the three ethylating agents studied are strongly mutagenic on TMV-RNA or TMV. The role of phosphate alkylation in regard to in vivo mutagenesis and oncogenesis remains to be established. At present it appears possible that the extent of this reaction may correlate better with the oncogenic effectiveness of different ethylating agents, than the extent of any base reaction. Unfractionated HeLa cell RNA is ethylated primarily in acid labile manner even by diethyl sulfate and ethyl methanesulfonate, a fact that is attributed to its high content of low molecular weight trna rich in terminal phosphates which alkylate readily.     

The specificity of different classes of ethylating agents toward various sites of HeLa cell DNA in vitro and in vivo.

The sites and extent of ethyl products of neutral ethylation of HeLa cell DNA by [14-C]diethyl sulfate, [14-C]ethyl methanesulfonate, and [14-C]ethylnitrosourea have been determined in vitro and in vivo, and found to differ significantly depending on the ethylating agents. Diethyl sulfate and ethyl methanesulfonate ethylate the bases of HeLa cell DNA in the following order: 7-ethylguanine greater than 3-ethyladenine greater than 1-ethyladenine, 7-ethyladenine greater than 3-ethylguanine, 3-ethylcytosine, O-6-ethylguanine. Ethyl bases accounted for 84-87% of the total ethyl groups associated with HeLa cell DNA. Ethylnitrosourea, in contrast, has particular affinity for the O-6 position of guanine. It ethylates the bases of HeLa cell DNA in the following order: O-6-ethylguanine, 7-ethylguanine greater than 3-ethyladenine greater than 3-ethylguanine, 3-ethylthymine greater than 1-ethyladenine, 7-ethyladenine, 3-ethylcytosine. Ethylation of the bases only accounts for 30% of the total ethylation in the case of ethylnitrosourea. The remaining 70% of the [14-C]ethyl groups, introduced in vivo and in vitro, are in the form of phosphotriesters which after perchloric acid hydrolysis are found as [14-CA1ethanol and [14-C]ethyl phosphate. In contrast, phosphotriesters amounted to only 8-20% of total ethylation in in vivo or in vitro diethyl sulfate and ethyl methanesulfonate treated HeLa cell DNA, and 25% of the total methylation in in vitro methylnitrosourea treated HeLa cell DNA. Alkylation at the N-7 and N-3 positions of purines in DNA destabilizes the glycosidic linkages. Part of 7-ethylguanine and 3-ethyladenine are found to be spontaneously released during the ethylation reaction. Incorporation of the 14-C of the alkylating agents into normal DNA bases of HeLa cells can be eliminated by performing the alkylations, in the presence of cytosine arabinoside, for 1 hr.     

Tumour induction by low molecular weight alkylating agents

Low molecular weight alkylating carcinogens, such as nitroso compounds, alkylate guanine of DNA to 7-alkylguanine, but the amount of this product correlates poorly with tumour induction. Loveless postulated that a minor product of alkylation, O-(6)-alkylguanine, may be responsible for mutagenesis and carcinogenesis. He showed that methyl methanesulphonate (MMS) does not produce O-(6)-methylguanine from deoxyguanosine, and in the present study it failed to induce thymic lymphomas or pulmonary adenomas in inbred Swiss mice. Loveless gave evidence that ethyl methanesulphonate (EMS), methylnitrosourea (MNU) and ethylnitrosourea (ENU) did produce O-(6)-alkylguanine, and all three induced pulmonary adenomas in the present study. It has also been shown that both of the alkylnitrosoureas induced thymic lymphomas but ethyl methanesulphonate did not.     

Mutagenicity of dichlorvos and methyl methanesulphonate for Escherichia coli WP2 and some derivatives deficient in DNA repair

The mutagenic and lethal action of methyl methanesulphonate (MMS) and dichlorvos (DDVP) has been studied on Escherichia coli WP2 and some derivatives deficient in DNA repair genes. The exrA⁺ and recA⁺ alleles were necessary for significant mutagenesis by either compound, and the uvrA gene affected neither the lethal nor mutagenic responses. Increased sensitivity to both compounds was shown by the exrA and uvrAexrA strains and in a more pronounced way by the uvrApolA, recA, and uvrAexrApolA strains.Bacteria deficient at the polA locus were 2 and 3 times more mutable by DDVP and MMS respectively, consistent with the hypothesis that the absence of the polA system for the repair of single-strand gaps results in a greater proportion of the total repair being channelled through the error-prone exrA⁺/recA⁺-dependent system. Single-strand breaks were detectable by alkaline sucrose gradient centrifugation after both MMS and DDVP treatment of polA bacteria. Thus in all the tests carried out, both compounds showed similar patterns of activity, and the results are consistent with their known ability to alkylate DNA. The chief differences were quantitative; sensitivity increases were far more pronounced with MMS which was also a far more potent mutagen than DDVP.     

A comparison of mutation induction by 14 MeV fast neutrons and 137 Cs -rays in meiotic spermatocytes in the silkworm

Barley seeds were treated with methyl methanesulphonate (MMS) and ethyl methanesulphonate (EMS), stored at 15% water content and washed for 16–24 h. These treatments resulted in an increase of toxic and genetic effects. In teh DNA of embryos of such stored MMS- and EMS-treated seeds, a strong enhancement of the amount of single-strand breaks and/or alkali-labile sites took place. In contrast, the amount of alkylated sites, particularly of 7-methylguanine, was somewhat lower. It can be that the depurination and/or backbone breakage, which proceeds during the storage period, is responsible for the enhancement of toxic and genetic effects, whereas the influence of the alkylation of DNA during the storage period by the unreacted residual mutagen is negligible.     

Some chemical aspects of dose-response relationships in alkylation mutagenesis

Alkylation of DNA can lead to induction of potentially miscoding groups (promutagenic) or potentially template-inactivating groups (lethal). The proportions of these are found to vary with the chemical nature of the alkylating agent. Agents of low Swain and Scott s factor (or those tending to Ingold's S_Ni type) react relatively more extensively at O-atom sites in DNA, and yield relatively more of the miscoding O₆-alkylguanine residues. Phosphotriester formation is also relatively more extensive with S_Ni agents.Inactivation of DNA can result from depurinations, strand breakage, and cross-linkage.Both promutagenic and lethal lesions are subject to repair; 3 principal enzymatic systems appear to exist; one for excision and repair of cross-links or aralkyl groups resembles the uvr system; others for repair of single-strand breaks parallel repair of X-ray-induced breaks (exr, rec systems); another, less well defined at present, recognizes certain methylated bases, and depurinated sites (probably Goldthwait's endonuclease II).These factors can be shown to influence dose-response in alkylation mutagenesis. This, broadly, can be classified as linear with the promutagenic group-inducing or directly miscoding agents, and is independent of cytotoxicity; whereas with other agents non-linear response parallels the occurrence of “shouldered” survival curves, and reflects mutation induction by “repairs errors”.Additionally, alkylation of cellular constituents other than DNA, e.g. repair enzymes, may influence dose response, and will again depend on chemical reactivity of the agent.     

Mutagenicity of hycanthone in Drosophila melanogaster

The schistosomicidal agent hycanthone was tested for mutagenicity in Drosophila melanogaster. The compound was administered either by injection into adult males or by larval feeding. The following types of genetic damage were measured:(1) complete and mosaic sex-linked recessive lethal mutations; (2) II–III translocations; and (3) dominant lethals.In postmeiotic germ cells, especially in late spermatids, a pronounced increase was found in the frequency of sex-linked recessive lethals, both completes and mosaics. By contrast, translocations and dominant lethals were not induced.     

The reaction of alkylating agents with bacteriophage R17. Biological effects of phosphotriester formation

The extent of biological inactivation and of the degradation of the RNA after reaction of bacteriophage R17 with ethyl methanesulphonate, isopropyl methanesulphonate and N-ethyl-N-nitrosourea was studied. Formation of breaks in the RNA chain probably results from hydrolysis of phosphotriesters formed in the alkylation reactions. Near neutral pH the ethyl and isopropyl phosphotriesters are sufficiently stable for the kinetics of the hydrolysis reaction to be followed. Results indicate that the rate of hydrolysis increases rapidly as the pH is raised. The evidence shows that a phosphotriester group does not itself constitute a lethal lesion. The extent of phosphotriester formation by the different agents is discussed in terms of reaction mechanism.     

Inactivation and mutation of coliphage T4 by aliphatic nitrosamides and methanesulphonates: in vitro recovery of infectivity of T4 inactivated by isopropyl methanesulphonate.

The inactivation and mutation (to r phenotype) of extracellular coliphage T4 wild-type by the monofunctional alkylating agents N-methyl- and N-ethyl-N-nitrosourea and isopropyl methanesulphonate were investigated. The rate and extent of change in phage infectivity observed during the post-treatment period were found to correlate with what is known of the mechanisms by which these agents react in vitro. Loss of phage infectivity was found to occur during the period following treatment with these agents, but that resulting from treatment with isopropyl methanesulphonate was preceded, in the first 24 to 48 h, by a recovery of infectivity. This suggested that changes in phage infectivity occurring after treatment with monofunctional alkylating agents are resultant of various processes which diversely promote loss and recovery of infectivity. The mutagenicity of N-methyl-N-nitrosourea was similar to that of its ethyl homologue at a level of phage survival of 4 x 10-3, but less than that of isopropyl methanesulphonate. At a level of survival of 3 x 10-2 ethyl methanesulphonate was a mutagenic as its isopropyl homologue, but methyl methanesulphonate was only slightly if at all mutagenic. These results could not be correlated with the compounds' reaction mechanisms. The efficiency of isopropyl methanesulphonate (compared with its toxicity to phage) was found to decrease as the severity of the dose was increased.     

Mutagenicity in Salmonella typhimurium TA98 and TA100 of nitroso and respective hydroxylamine compounds

Five aromatic nitroso compounds were prepared and their mutagenicity in Salmonella typhimurium strains TA98 and TA100 compared with that of the corresponding hydroxylamines and the previously studied nitroarenes. A remarkable correspondence of the dose-response curves was observed between the nitroso and the respective hydroxylamine compounds. This effect could be observed in TA98 and TA100. It was only marginally dependent on the metabolical activation by rat liver S9-mix. Even the presence of a bulky alkyl substituent either near to the functional group, or far away from it, previously shown to considerably influence the mutagenic properties of nitroarenes, does not remarkably affect the properties of the nitroso and hydroxylamine species. The similarity between the latter two is likely to be due to a fast reduction of the nitrosoarenes to the hydroxylamine species under the test conditions. It seems that enzymes are not responsible for that reduction step, because sterical crowding near the functional group does not influence that behaviour.The test results of the aromatic hydroxylamines bearing a bulky substituent show that there are at least two ways to influence the mutagenicity of an aromatic nitro compound by such a group. A substituent near the functional group (ortho-position) disturbs the enzymatic reduction of the nitro group, because 3-tert-butyl-4-hydroxylaminobiphenyl and its corresponding nitroso compound are highly mutagenic, whereas 3-tert-butyl-4-nitrobiphenyl was previously shown to be inactive even after addition of S9-mix. In contrast, 4'-tert-butyl-4-hydroxylaminobiphenyl with the tert-butyl group "far away" from the hydroxylamino functionality clearly shows decreased mutagenic activity suggesting a different influence of a substituent in that position. In addition, the substance shows only little cell toxicity even at higher concentrations. Both effects could be due to a reduced effective dose of the hydroxylamine in the cells compared to the non-alkylated compound, caused by a faster degradation of the hydroxylamine or a hindered interaction between that substance and the cells.     

Effect of N-nitroso-N-methylurea on viability and mutagenic response of repair-deficient strains of Escherichia coli

Various E. coli mutants, deficient in DNA repair, differed in their response to increasing concentrations of N-nitroso-N-methylurea (NMU).Loss of viability due to exposure to NMU was greatest in those strains with a reduced capacity for repair of single-strand breaks. Viability of wild-type and uvrA^? strains was not affected by NMU concentrations up to 3.0 mM. Some loss of viability occurred, at the higher NMU concentrations, in both strains carrying exrA^? while strains carrying uvrA^?polA^? or recA^? were the most sensitive. The results support the hypothesis that the lethal effect of NMU on repair-deficient E. coli was due to its ability to induce single-strand breaks.Induction of mutations by NMU was observed in all the strains used and the results suggested that NMU damage per se was the major mutational event. The dose response curve for induction of revertants by NMU was, however, influenced by the repair system(s) present. The number of revertants scored at the higher NMU concentrations was greater in those strains lacking the recA and polA dependent repair functions than in the wild-type strain. However, at NMU concentrations below 2.0 mM the numbers of revertants induced in exrA^? carrying strains, prossessing accurate rec-dependent repair, were lower than the comparable wild-type values. The evidence suggests that the uvrA gene product also acts on some, possibly non-mutagenic, types of NMU damage and that error-prone repair of these lesions increases the number of potential revertants.     

The nature of reverse mutations in Drosophila melanogaster

A selection system for the screening of reversions has been constructed and used to test reversions of lethals located in the proximal region of the X chromosome of Drosophila and of Kpn mutations.Spontaneous and induced reversions have been screened, X-rays and ethyl methanesulphonate (EMS) being the mutagens used in the induction experiments.No genuine back-mutation was found in 6·10⁵ gametes scored. Sterile reversions of all four lethals tested were obtained. Their frequency suggested that at least in three of the lethals the sterile reversions represented “escapers” of the lethal effect rather than true revertants.Three fertile reversions of l^x4 were found and analyzed. All three were autosomal suppressors, located on the second chromosome, allelic to each other, dominant in males and recessive in females.One fertile reversion of l^3DES was found to be an X-linked suppressor. It is suggested that this suppressor is a Y-suppressed lethal, showing a V-type position effect, resulting from an aberration included in the proximal heterochromatin of the X chromosome.Reversions of Kpn were obtained at a similar rate to that found in previous reports²².The absence of true back-mutants in our experiments, in contrast to findings in previous reports, is discussed. From the existing literature on spontaneous and induced back-mutations in Drosophila melanogaster it appears that for several mutations the rates of forward and back-mutation are of the same order of magnitude. It is suggested that reported cases of back-mutations represent mainly inter- and intrachromosomal recombination in duplicated regions rather than mutational events and that the frequency of true back-mutation in Drosophila is usually of an order of magnitude, similar to that known for microorganisms and fungi.     

Properties of 3-methyladenine-DNA glycosylase from Escherichia coli.

An Escherichia coli enzyme that releases 3-methyladenine and 3-ethyladenine in free form from alkylated DNA has been purified 2800-fold in 7% yield. The enzyme does not liberate several other alkylation products from DNA, including 7-methylguanine,O6-methylguanine, 7-methyladenine, N6-methyladenine, 7-ethylguanine, O6-ethylguanine, and the arylalkylated purine derivatives obtained by treatment of DNA with 7-bromomethyl-12-methylbenz[a]anthracene. The reaction of the enzyme with alkylated DNA leads to the introduction of apurinic sites but no chain breaks (less than one incision per ten apurinic sites), and there is no detectable nuclease activity with native DNA, depurinated DNA, ultraviolet-irradiated DNA, or X-irradiated DNA as potential substrates. The enzyme is termed 3-methyladenine-DNA glycosylase. It is a small protein, Mr = 19 000, that does not require divalent metal ions, phosphate, or other cofactors in order to cleave base-sugar bonds in alkylated DNA.     

The isolation and properties of δ-tocotrienol from Hevea latex

1. delta-Tocotrienol (8-methyltocotrienol) was isolated from the latex of Hevea brasiliensis. This new member of the tocopherol family is a pale-yellow oil at room temperature. 2. The properties of delta-tocotrienol are very similar to those of delta-tocopherol and the small differences can be explained by the change in side chain. 3. The ultraviolet and infrared spectra of delta-tocotrienol were determined and a conversion factor for use with the Emmerie-Engel reaction was worked out. Details are given for the chromatography of delta-tocotrienol on thin layers (adsorption and partition) and reversed-phase paper, and the nitroso derivatives were formed. 4. An ethyl carbonate ester of delta-tocotrienol was prepared and compared with a similar ester of delta-tocopherol. 5. Hydroxymethylation of delta-tocotrienol followed by reduction gave beta-tocotrienol as a major product.     

Function and organization of the G region of bacteriophage Mu; Host specificity determined by gene splicing

Chinese hamster ovary (CHO) cells were exposed to [³H]ethyl nitrosourea (ENU) or [³H]ethyl methanesulfonate (EMS) and the following DNA ethylation products were quantitated: 3- and 7-ethyladenine, O²-ethylcytosine, 3-, 7- and O⁶-ethylguanine, O²- and O⁴-ethyldeoxythymidine and the representative ethylated phosphodiester, deoxythymidylyl (3′–5′)ethyl-deoxythymidine. When mutations at the hypoxanthine-guanine phosphoribosyl transferase (HGPRT) locus induced by these same treaments were compared with the observed ethylation products, mutations were found to correlate best with 3- and O⁶-ethylguanine. EMS induced approximately twice as many sister-chromatid exchanges (SCEs) as ENU at doses yielding equal mutation frequencies. When SCEs were indirectly compared with DNA ethylation products, 3-ethyladenine and ethylated phosphodiesters related best to SCE formation. Because mutation and SCE induction appear, at least in part, to be related to different DNA adducts, SCE induction by simple ethylating agents may not be a quantitative indicator of potentially mutagenic DNA damage.     

Genetics of somatic mammalian cells XII: Mutagenesis by carcinogenic nitroso compounds

Four known carcinogenic compounds, all of which contain the nitroso group, have been found to be effective in producing cell killing, chromatid breaks, mutagenesis and chromatid rearrangements in Chinese hamster ovary cells. Single cell survival curves revealed these agents to exhibit a 50,000-fold difference in their mean lethal dose (D_o) values. However, all agents yielded constant values for their efficiencies of production of chromatid breaks and single gene mutagenesis, but not chromatid rearrangements, when these were expressed in terms of the D_o value of each agent. Three of the carcinogens produced single-hit and one produced a multiple-hit survival curve. The system offers distinct advantages in study of the interrelationships between mutagenic and carcinogenic actions in mammalian cells.     

Comparative study of the clastogenic efficiency of ethyl methanesulfonate and diethyl sulfate in Drosophila melanogaster mature sperm

Chromosome loss and translocation tests were carried out in Drosophila melanogaster sperm, stored in untreated females for up to 24 days, to compare the clastogenicity of ethyl methanesulfonate (EMS) and diethyl sulfate (DES). The sex-linked recessive lethal test was used as a "biological dosimeter" and the following results were obtained: The yield of 2-3 translocations induced by both mutagens increased steadily with storage, being significantly higher after EMS than after DES treatment. The frequencies of partial losses induced by EMS and DES were similar and increased with storage. With up to 11 days' storage, the frequency of complete loss induced by DES was higher than that induced by EMS and remained unchanged when storage was extended to 24 days. Complete loss induced by EMS increased significantly with further storage (12-24 days). With DES, complete (but not partial) loss was detected with a dose at which EMS failed to modify the control values. These data suggest that the lower recovery of II-III translocations after treatment with DES does not result from a low breaking capacity but from a diminished or delayed rejoining of the induced breaks. This could be due to a physiological impairment of the treated cells by the high toxicity of DES or to an actual lower rejoinability of the broken ends. The differential recovery of complete and partial losses after DES treatment further suggests that the mechanisms leading to the fixation of both types of damage are somehow different, and that processes intervening in the recovery of partial losses are less affected, or not at all, by the proposed reduced rejoining of chromosome breaks.     

The molecular basis for biological inactivation of nucleic acids. The action of methylating agents on the ribonucleic acid-containing bacteriophage R17

1. The inactivation of an RNA-containing bacteriophage after reaction with four methylating agents was studied. Measurements of the extent of methylation of the RNA and of the nature and amounts of the various reaction products were made. In experiments with dimethyl sulphate and methyl methanesulphonate inactivation can be quantitatively accounted for by methylation at two of the positions involved in hydrogen bonding: N-1 of adenine and N-3 of cytosine. In experiments with N-methyl-N-nitrosourea and N-methyl-N'-nitro-N-nitrosoguanidine methylation at N-1 of adenine and N-3 of cytosine accounts for only about one-half of the observed inactivation. Scission of the RNA chain during reaction accounts for a further 20% of the inactivation. To account for the remainder it seems necessary to postulate that formation of O(6)-methylguanine constitutes a lethal lesion. 2. Breaks in the RNA chain formed on reaction with the nitroso derivatives presumably result from methylation of the phosphate diester group followed by hydrolysis of the unstable triester thus formed.     

DNA photobinding of 7-methylpyrido[3,4-c]psoralen and 8-methoxypsoralen. Effects on macromolecular synthesis, repair and survival in cultured human cells

The photobinding to DNA of tritiated 7-methylpyrido[3,4-c]psoralen (MPP), a recently synthesized monofunctional compound of therapeutical interest, and of 8-methoxypsoralen (8-MOP) was determined in cultured normal human fibroblasts. Employing compounds at 10(-6) M, MPP photobinds approximately 11 times more efficiently than 8-MOP: one molecule is fixed respectively per 7.5 X 10(4) or 8.1 X 10(5) base pairs/kJ . m-2 of 365-nm radiation (UVA). Removal of bound material from DNA is slow and limited in 48-72 h of post-treatment incubation to 30-40% of initial adducts formed by MPP and to 50-60% of those of 8-MOP. For equivalent photobinding MPP and 8-MOP induce similar inhibitions of DNA synthesis. However, the recovery of DNA synthesis during post-treatment incubation is lower after photoaddition of MPP than after that of 8-MOP. MPP also exerts a much higher lethal effect than 8-MOP: one lethal hit corresponds to about 4400 and to 19,900 adducts per cell respectively. Alkaline elution experiments confirmed the monofunctional nature of MPP and indicated that in MPP-damaged cells DNA breaks accumulate with time of post-treatment incubation. In contrast, after photoaddition of 3-carbethoxypsoralen (3-CPs), another monofunctional furocoumarin, or irradiation with 254-nm UV, DNA breaks are induced only transiently. In 8-MOP-treated cells, DNA cross-links appear to be partially repaired. In conclusion, MPP monoadducts turn out to constitute more cytotoxic lesions than 8-MOP mono- and bi-adducts.     

Effects of caffeine on chromosome aberrations and sister-chromatid exchanges induced by mitomycin C in BrdU-labeled human chromosomes.

The BrdU-Hoechst staining technique has been used in analyzing the effect of caffeine (CAF) on chromosome aberrations and sister-chromatid exchanges (SCEs) induced by mitomycin C (MC). CAF increased the frequency of SCE in MC-treated chromosomes in all specimens. The combination of MC and CAF caused a remarkable increase in all types of chromosome aberrations, but the most startling effect was the appearance of many cells with multiple aberrations (shattered chromosomes). The BrdU-Hoechst technique showed that the shattered chromosomes did not appear in cells that had replicated only once, but did occur in cells which replicated twice in the presence of MC and CAF. The large majority of chromatid breaks observed did not involve areas common to SCE; and the SCE frequency significantly increased in spite of the existence of multiple breaks. This indicates that very few of the breaks are incomplete exchanges and that the mechanism for formation of SCE might be different from that of chromosome breaks. In another experiment, monofunctional-MC (M-MC) had a small effect on SCE rates, though it induced shattered chromosomes with CAF post-treatment. Possible differences in the mechanisms leading to SCE and chromosome breaks are discussed.