Development of an 125I-postlabeling assay as a simple, rapid, and sensitive index of DNA-protein cross-links.

A rapid, simple, and sensitive 125I-postlabeling technique has been developed to allow detection of DNA-protein cross-links induced by environmental contaminants and carcinogens. This method is based on specific incorporation of 125I into tyrosine residues associated with DNA. Cultured Chinese hamster ovary cells were exposed to various crosslinking agents, e.g., UV light, K2CrO4, or NiCl2. DNA was isolated by proteinase K/phenol/chloroform. The residual peptides cross-linked to DNA were radioiodinated with Na125I and chloramine T. After repeated precipitation with ethanol, the radioactivity was determined. The 125I method was compared with a 3[H]-tyrosine prelabeling method and found to be of similar sensitivity.

Introduction DNA-protein cross-links (DPCs) are thought to be important genotoxic lesions induced by environmental contaminants and carcinogens (1,2). They are not easily repaired, and may impede cellular replication, transcription, or repair processes (1,2). Various methods have been developed for purification and detection of DPCs. Some of them, however, are plagued by problems related to high background, due to DNA isolation without proteinase digestion (1,2). Recently, we designed an effective and sensitive assay for residual amino acid-DNA cross-links following proteinase K digestion in Chinese hamster ovary (CHO) cells that were prelabeled with a radioactive amino acid (3). The assay has been used to measure DPC formation in tissue culture induced by nickel and chromium compounds. Unfortunately, this assay can only be used in systems where cellular proteins can be radiolabeled.
Iodine isotopes, especially 125I, offer several advantages in postlabeling methods. Iodine can be incorporated into the aromatic side-chain of tyrosine residues with This paper was presented at the Second International Meeting on Molecular Mechanisms of Metal Toxicity and Carcinogenicity held [10][11][12][13][14][15][16][17]  relative ease, yielding a stable radiolabeled compound. Iodine may also substitute into other amino acids (e.g., histidine) but the reaction rate is much less than that for tyrosine. We and others have shown that tyrosine is the predominant amino acid cross-linked to DNA by cross-linking agents including y-radiation and some transition metals (3)(4)(5)(6). The short half-life (60 days) is an advantage in isotope disposal. The counting efficiency of l I is high. The isotopic abundance is higher for current preparations of i25j (up to 100%), giving an effective labeling rate. The radiation emitted by 125I is less penetrating than that of i3'I thus presenting a reduced radiation hazard.
In this article, we describe a simple, rapid, and effective assay fori2 I-postlabel- DNA was isolated from CHO cells as previously described (3). Briefly, cells were scraped with a rubber policeman, collected and washed by repeated centrifugation and suspension. The cell pellet was lysed in 10 mM Tris, pH 8.0, containing 100 mM NaCl, and 0.5% SDS. RNase (10 mg/ml) was added, and a high concentration of proteinase K (500 mg/ml) was applied to hydrolyze the protein. The DNA pellet containing residual polypeptides and amino acids, was repeatedly extracted with phenol/chloroform, and the final aqueous layer was precipitated with ethanol. 125 _RadoaCtiVe Posdabeling i25I-postlabeling of residual polypeptides and amino acids associated with the DNA was performed by modification of the iodination technique described by Neuer et al. (7). Fifty micrograms DNA precipitated with ethanol was suspended in 100 ml of 2% SDS, 30% urea and 0.5M Tris-HCI, pH 7.6, and mixed with 10 mCi of Nai25I (17.4 Ci/mg) and 5 ml of chloramine T solution (6 mg/ml). For rapid and efficient iodination, reagents in the reaction mixture were highly concentrated. After 2 min at room temperature, the iodine was reduced by addition of 10 ml of 50% l-mercaptoethanol. DNA was collected together with the associated radiolabeled tyrosine by Environmental Health Perspectives The effect of temperature ( residual peptides associated with DI trol and UV-exposed CHO cells. lodin at the indicated temperatures, pH 7. for 5 min, after which samples we ethanol in five cycles. described (3), and the results were compared with that of the 125I-postlabeling. (A scheme for the 1251I-labeling technique is shown in Figure 1).

Results and Discussion
Initially, experiments were performed to establish the optimal conditions for radioiodination of tyrosine residues associated with DNA in control and UV-treated CHO cells. Variables such as pH, temperature, reaction time, ethanol precipitation procedures, and 125I/DNA ratios were optimized. As shown in Figure 2, chloramine T-mediated iodination was strongly temperature-dependent, and a maximum yield of iodine incorporation was achieved at 250 to 350C. Further increases of reaction temperature resulted in a decrease in iodinacol. tion yield, which was probably due to increased side reactions at higher temperaipitation with tures. Some investigators prefer to carry out olved in 10 mM iodination protocols on ice to reduce side rporated 125j in reactions and oxidative damage induced by I off. The DNA chloramine T; however, we found that iodadioactivity in a ination at temperatures less than 25°C also nples were used reduced the reaction rate. We suggest that absorbance at iodination at room temperature is both was expressed as suitable and convenient. d assay described Figure 3 shows that the optimal pH -ding to a series value for iodination of tyrosine residues sults). was 7.5. Above pH 8 there was a tendency ition efficiency, toward decreased iodination yield, and Labeled with 3Habove pH 9, the reaction became highly the cross-linking inefficient. Since the isotope Na'25I was -d as previously usually supplied as a solution in 0.1 N NaOH, the composition of the other reactants must be modified to buffer the iso-*UV(4K/m') tope to pH 7.5. Figure 4 shows that chloramine Tmediated iodination of tyrosine residues was instantaneous, such that the minimum reaction time required to obtain a maximal yield of iodination at room temperature was 1 min. After this, the yield remained constant for 20 min. Incubation for much longer times caused oxidative damage to DNA and peptides, and consequently decreased the product yield. To ensure an adequate mixture of reactants, we chose 2 min as the standard incubation time.
The effects of ethanol precipitation on 0°X>__0 recovery of 125I-labeled tyrosine residues is shown in Figure 5. Following iodination, peptides associated with DNA isolated from control and UV-exposed CHO cells. lodination was conducted at the indicated pH, at room temperature, 4 mCi/10mg DNA for 5 min, after which samples were precipitated five times with ethanol.
radioactivity represented tyrosine residues crosslinked to DNA.
To establish the ratio of 1251/ DNA for optimal efficiency of radioiodination, the ratio of iodine to DNA in reaction mixtures was varied and the specific activity of the products was determined. The curve shown in Figure 6 demonstrates that the optimal yield of radioiodination is achieved at iodine to DNA ratios of 1. of residual peptides associated with DNA isolated from control and UV-exposed CHO cells. lodination was conducted at room temperatures, pH 7.5, 4 mCi/10 mg DNA for the indicated intervals, and then samples were precipitated five times with ethanol.  Comiro lished above, we examined DPC forma-UV(4CJ/o2) tion in cultured CHO cells induced by UV, HCHO, chromium(VI), nickel(II), and cis-Pt agents which have been demonstrated by other studies to yield DPCs by quite different mechanisms (2). UV light produces radicals in the ring structure of DNA bases, which then react with adjacent proteins to form covalent bonds (8). HCHO forms covalent bonds between the S amino groups of lysine residues and adjacent DNA bases (9). Chromium(III), the reduced form of chromium(VI), and the dechlorinated form of cis-Pt have an affinity for histidyl, methionyl, and cysteinyl residues of proteins and the nitrogens or phosphates of DNA bases (10). b* 7 8 9 Nickel(II) was thought to form stable proa (No.) tein-nickel(II)-DNA complexes, and a itation on~l 1poststrong interaction between nickel(II) and ted with DNA i-s amino terminal residues and imidazole D cells.Iodination group of histidine residue was demon-pH 7.5, 4 mCi/10 strated (11). On the other hand, increasing e precipitated with evidence suggests that some transition metted.
als may generate reactive oxygen species (ROS), which may indirectly mediate DNA damage, protein oxidation, and DPC d UV-treated formation (12,13). Regardless of the mechrther increases anisms by which DPCs are formed, Lny increase in increasing levels of DPCs were likely to the resulting affect the structure and function of DNA. ns that under It is noteworthy that nucleoprotein in norall the iodine mal cells, as an important component for sine residues maintenance of DNA conformation, replien saturated. cation, and transcription, is also closely Lditions estab-associated with DNA by phosphotriester bond between internucleotide phosphate or other unknown bonds. The presence of Rato of 12-1 to DNA ( Ci/10ig ) Figure 6. The effect of temperature on 'rl-postlabeling of residual peptides associated with DNA isolated from control and UV-exposed CHO cells. lodination was conducted at indicated ratios of 251/DNA at room temperature, pH 7.5, for 5 min, and then the samples were precipitated with ethanol. these nucleoproteins results in higher background and lower sensitivity of most DPC assays. To solve this background problem, we previously developed a new method for analysis of residual amino acid-DNA cross-links as an index of DPC formation after prelabeling CHO cells with radioactive amino acid and isolation of DNA by standard proteinase K/phenol extraction (3). This method has been demonstrated to be highly sensitive for isolation of DPCs from chromium(VI)and nickel(II)-treated CHO cells (3). The assay presented here has obvious advantages over the previous method, in which the prelabeling with a radioactive amino acid is substituted by 125I-postlabeling of tyrosine residue associated with DNA. Figure  7 illustrates that all five test agents induced increased 1251I-labeling efficiency in a dosedependent manner. Particularly significant increases were noted for UV light and chromium(VI 3H-tyrosine prelabelling ( cpm/lOgLg DNA ) Figure 7. The use of the "l-postlabeling assay for determination of DPCs formed in CHO cells by some cross-linking agents. CHO cells were exposed to various concentrations of the indicated agents. Following these treatments, residual peptide-DNA complexes were isolated by proteinase K/ phenol/ chloroform extraction. n1'postlabelling was conducted as described in Materials and Methods. Each point presents the mean ± SEM from three separate samples.
shown. In a comparison between the postlabeling and prelabeling protocols with CHO cells exposed to UV, chromium(VI) 4 in = I 0.
T Figure 8. The correlation between '251-postlabeling and 3H-tyrosine prelabeling assays. CHO cells were prelabeled with 3H-tyrosine for 24 hr prior to exposure to indicated agents. DNA was isolated by proteinase/ phenol/chloroform extraction. Residual tyrosine complexed with DNA was determined as previously described (3). The results were compared with that of the l251-postlabeling method. and nickel(II), a significant correlation between the two assays is found (in Figure  8, r=0.864, p<0.01). In fact, it appears that the postlabeling method is more sensitive than prelabeling assay. The 1251 -postlabeling assay allows the sensitive detection of DPCs in human samples from individuals exposed to environmentally adverse agents. 125 In summary, I-postlabeling technique can be used to detect DPC formation. It can be applied to experimental studies and can be used in risk assessment and epidemiologic investigations of human population exposed to environmental carcinogens or mutagens.