Protein-DNA interactions in the human beta-globin locus control region hypersensitive site-2 as revealed by four nitrogen mustards

The influence of chromatin structure on DNA damage induced by nitrogen mustard and cisplatin analogues

The interaction of anti-tumour drugs with reconstituted chromatin has been investigated using defined nucleosomal complexes. This allowed the effect of nucleosome cores on drug-induced DNA damage to be assessed for four nitrogen mustard analogues, dimethylsulphate and three cisplatin analogues. A defined nucleosomal complex was employed that contained two precisely positioned nucleosome cores. The construct was then subjected to drug treatment, and the resulting DNA damage was quantitatively analysed using a Taq DNA polymerase stop assay. At the sites of damage, densitometric comparisons between purified and reconstituted DNA were used to evaluate the influence of nucleosomal core proteins on specific drug-DNA interactions. Results were combined with previous data obtained for other DNA-damaging drugs investigated using the same nucleosomal construct. For most of the DNA-damaging agents studied, this method revealed protection at the positioned nucleosome cores and indicated that the preferred site of DNA binding for these compounds was in the linker region of the construct. Statistical analyses confirmed the significant level of damage protection conferred by the nucleosome cores and revealed differences between the examined compounds. Larger compounds generally displayed a greater tendency to target the linker region of the nucleosomal DNA and were impeded from damaging nucleosomal core DNA. In contrast, smaller molecules had greater access to nucleosomal core DNA.

Footprinting the 'essential regulatory region' of the retinoblastoma gene promoter in intact human cells

The retinoblastoma tumour suppressor protein is a key cell cycle regulator. Protein-DNA interactions at the retinoblastoma (RB1) promoter, including the 'essential regulatory region', were investigated using novel DNA-targeted nitrogen mustards in intact human cells. The footprinting experiments were carried out in two different environments: in intact HeLa and K562 cells where the access of DNA-targeted probes to chromatin is affected by cellular protein-DNA interactions associated with gene regulation; and in purified DNA where their access is unencumbered by protein-DNA interactions. Using the ligation-mediated PCR (LMPCR) technique, the sites of damage were determined at base pair resolution on DNA sequencing gels. Our results demonstrate that, in intact cells, footprints were observed at the E2F, ATF and RBF1/Sp1 DNA binding motifs in the RB1 promoter. In addition, a novel footprint was observed at a previously unidentified cycle homology region (CHR) and at four uncharacterised protein-DNA binding sites. In further experiments, nitrogen mustard-treated cells were FACS sorted into G1, S and G2/M phases of the cell cycle prior to LMPCR analysis. Expression of the RB1 gene is cell cycle-regulated and footprinting studies of the promoter in FACS-sorted cells indicated that transcription factor binding at the GC box, CHR binding motif and the 'essential regulatory region' are cell cycle dependent.

Genomic and phylogenetic footprinting at the epsilon-globin silencer region in intact human cells

Chromatin structure at the silencer region of the epsilon-globin promoter was investigated using novel nitrogen mustards as probes of protein--DNA interactions. Sites of protection and enhancement that corresponded to known transcription factor binding sites were detected in both K562 and HeLa cells at this gene region. Protection was observed at several sites including the GATA-1/YY1 motifs. Of particular interest was a large 155 bp footprint that was observed at the epsilon-globin gene silencer region of the promoter. This large footprint was consistent with the presence of a positioned nucleosome core in intact human cells at this silencer region. Additionally, the DNA sequence at the epsilon-globin silencer and promoter was compared for 11 mammalian species. Significant areas of conservation were found that correlated with known transcription factor binding motifs. This phylogenetic footprinting analysis was compared to the genomic footprinting data at the epsilon-globin silencer region.

The effect of DNA-alkylating agents on gene expression from two integrated reporter genes in a mouse mammary tumor line

A model system was developed to investigate the effects of DNA alkylating agents on cellular gene expression. The cytomegalovirus immediate-early promoter (CMV) and the mouse mammary tumor virus promoter (MMTV) were coupled separately to the luciferase reporter gene and stably expressed in cultured cells. The change in luciferase activity was used as a measure of gene expression inhibition. Seven well-characterized DNA alkylating agents of varied DNA adduct-forming ability were evaluated in this system. The major groove binders/intercalators (that form guanine adducts) increased CMV-luciferase activity above background, while minor groove binders (that form adenine adducts) all decreased it. The MMTV-luciferase activity was remarkably different to the CMV-luciferase activity and was inhibited to the greatest extent by the minor groove alkylators. One of these, a polybenzamide with spatially separated alkylating groups, inhibited gene expression to a greater extent than inhibition of general DNA or RNA synthesis.

Alkylating agent and chromatin structure determine sequence context-dependent formation of alkylpurines

We determined the adduct maps of S(N)1 and S(N)2 alkylating agents in cultured human cells (in vivo) and in vitro to probe DNA-protein interactions along sequences of the promoter and exon 1 of the Fragile-X mental retardation 1 (FMR1) gene. Using ligation-mediated polymerase chain reaction (LMPCR), we compared the piperidine-sensitive alkylpurines sites generated by treating cultured cells (in vivo) and naked DNA (in vitro) with S(N)1 (N-methyl-N-nitrosourea, N-nitroso(acetoxymethyl)methylamine and 1-methyl-3-nitro-1-nitrosoguanidine) and S(N)2 alkylating agents (dimethyl sulfate (DMS), methane sulfonic acid methyl ester, iodo methane, diethyl sulfate, methane sulfonic acid ethyl ester and iodo ethane). The FMR1 promoter has four sites where DNA-protein interactions are observed. In these regions, the S(N)1 methylating agent reactions produced only hypo-reactive sites. In contrast, iodoalkane S(N)2 alkylating agents (MeI and EtI) reactions generated only hyper-reactive sites. Although there are hyper-reactive sites for the other S(N)2 reagents, the hyper-reactive site at +14 on the FMR1 map is more pronounced for the sulfate and sulfonate-derived alkylating agents than for the iodoalkanes. However, DMS modification in the presence of methyl sulfone, a compound that does not alkylate DNA, eliminates the hyper-reactive site observed at +14. This suggests that the electron-rich oxygen atoms of the sulfate and sulfonate-derived S(N)2 alkylating agent structure position the alkylating moiety to the neighboring N-7-guanine position to favor alkyl transfer to the guanine. Using KMnO(4) to probe for single-strand DNA, an unpaired cytosine base was detected at the 5'-side of the hyper- reactive guanine base at position +14, consistent with the formation of a local DNA single-strand bulge. In conclusion, we show that the sequence context-dependent formation of alkylpurines is determined by the chemical nature of the alkylating agent, the DNA sequence context, chromatin structure, and the presence of other non-reactive molecules that can inhibit alkylation.

The effect of chromatin structure on cisplatin damage in intact human cells

The influence of chromatin structure on cisplatin DNA damage was investigated in intact human cells. The epsilon-globin gene promoter was utilised as the target DNA sequence and the terminal transferase-dependent PCR technique was employed to examine adduct formation at base pair resolution. It was found that cisplatin preferentially damaged at runs of consecutive guanine bases in intact cells. By comparing the relative intensity of adduct formation in intact cells and in purified genomic DNA, it was possible to assess the influence of chromatin proteins on the extent of cisplatin DNA damage. Enhanced damage in intact cells was found at the CACC site where a member of the Sp1 family of proteins is thought to bind. It is postulated that protein binding at this site bends the DNA double-helix so that enhanced cisplatin binding occurs. The altered DNA binding of cisplatin in the presence of chromatin proteins could be important in the properties of cisplatin as an anti-tumour drug.

A large "footprint" at the boundary of the human beta-globin locus control region hypersensitive site-2

The 5'-boundary region of the human beta-globin locus control region hypersensitive site-2 (HS-2) was examined for protein-DNA interactions. The HS-2 is an erythroid specific DNase I hypersensitive site that extends for approximately 600 bp. Erythroid K562 cells and non-erythroid HeLa cells were damaged by bleomycin and hedamycin--these agents are able to "footprint" nucleosome cores and proteins bound to DNA. The fragments generated by DNA damage were amplified by the ligation-mediated polymerase chain reaction with primers specific for the 5'-boundary region of HS-2 and examined at base pair resolution on DNA sequencing gels. The intensity of damage in intact cells was compared with that in purified DNA. The comparison between intact cells and purified DNA revealed a protected region of 226 bp with bleomycin and 182 bp with hedamycin in K562 cells. The length of the protected region was consistent with the presence of a nucleosome core. We postulate that an erythroid-specific protein binds next to the positioned nucleosome at the boundary of HS-2 to prevent sliding of the nucleosome into the hypersensitive site--this would also account for the large size of the protected region. HeLa cells (lacking a hypersensitive site in the beta-globin cluster) did not have an area of protection in this region.

A survey of the sequence-specific interaction of damaging agents with DNA: emphasis on antitumor agents

This article reviews the literature concerning the sequence specificity of DNA-damaging agents. DNA-damaging agents are widely used in cancer chemotherapy. It is important to understand fully the determinants of DNA sequence specificity so that more effective DNA-damaging agents can be developed as antitumor drugs. There are five main methods of DNA sequence specificity analysis: cleavage of end-labeled fragments, linear amplification with Taq DNA polymerase, ligation-mediated polymerase chain reaction (PCR), single-strand ligation PCR, and footprinting. The DNA sequence specificity in purified DNA and in intact mammalian cells is reviewed for several classes of DNA-damaging agent. These include agents that form covalent adducts with DNA, free radical generators, topoisomerase inhibitors, intercalators and minor groove binders, enzymes, and electromagnetic radiation. The main sites of adduct formation are at the N-7 of guanine in the major groove of DNA and the N-3 of adenine in the minor groove, whereas free radical generators abstract hydrogen from the deoxyribose sugar and topoisomerase inhibitors cause enzyme-DNA cross-links to form. Several issues involved in the determination of the DNA sequence specificity are discussed. The future directions of the field, with respect to cancer chemotherapy, are also examined.

Protein-DNA footprinting of the human epsilon-globin promoter in human intact cells using nitrogen mustard analogues and other DNA-damaging agents

Nitrogen mustard analogues, bleomycin and dimethyl sulphate (DMS) have been used as probes of protein-DNA interactions in intact human cells. The sites of damage have been determined at base pair resolution in the single copy epsilon-globin gene promoter in erythroid K562 cells, non-erythroid HeLa cells and purified DNA. Exponential amplification of gene-specific damage fragments was achieved using the ligation-mediated polymerase chain reaction (LMPCR) technique and analysed on DNA sequencing gels. A comparison of the relative damage band intensities between purified DNA and intact cells revealed several significant differences - both protection (footprint) and enhancement. These differences occurred at putative transcription factor binding sites and hence are thought to be due to protein-DNA interactions. A major feature of the band intensity ratio plots was the footprint observed at the CCAAT box binding motif as revealed by nitrogen mustard analogues. Enhanced band intensity (hypersensitivity) was displayed at the 5'- and 3'-ends of the CCAAT box in K562 cells - this feature was absent in HeLa cells and in vitro reconstitutions. A footprint was found at the GATA-1 motif in K562 cells that was also absent in non-expressing HeLa cells. Footprints were also evident at the TATA box, CACC box and the epsilonF1 DNA binding motif in K562 cells.

The interaction of DNA-targeted platinum phenanthridinium complexes with DNA

Cisplatin analogues were synthesised that consisted of platinum(II) diamine complexes tethered via a polymethylene chain ( n = 3, 5, 8 and 10) to a phenanthridinium cation. Both chloro and iodo leaving groups were examined. DNA adduct formation was quantitatively analysed using a linear amplification system with the plasmid pGEM-3Zf(+). This system utilised Taq DNA polymerase to extend from an oligonucleotide primer to the damage site. This damage site inhibited the extension of the DNA polymerase. The products were electrophoresed on a DNA sequencing gel enabling adduct formation to be determined at base pair resolution. The damage intensity at each site was determined by densitometry. The platinum phenanthridinium complexes were shown to damage DNA at shorter incubation times than cisplatin. To produce similar levels of damage, an 18 h incubation was required for cisplatin compared to 30 min for the n = 3 platinum phenanthridinium complexes; this indicates that the intercalating chromophore causes a large increase in the rate of platination. A reaction mechanism involving direct displacement of the chloride by the N-7 of guanine may account for the rate increase. These results indicate that further development of these compounds could lead to more effective cancer chemotherapeutic agents.