Synergistic modulation of cyclobutane pyrimidine dimer photoproduct formation and deamination at a TmCG site over a full helical DNA turn in a nucleosome core particle

Abstract

Sunlight-induced C to T mutation hotspots in skin cancers occur primarily at methylated CpG sites that coincide with sites of UV-induced cyclobutane pyrimidine dimer (CPD) formation. The C or 5-methyl-C in CPDs are not stable and deaminate to U and T, respectively, which leads to the insertion of A by DNA polymerase η and defines a probable mechanism for the origin of UV-induced C to T mutations. We have now determined the photoproduct formation and deamination rates for 10 consecutive T=(m)CG CPDs over a full helical turn at the dyad axis of a nucleosome and find that whereas photoproduct formation and deamination is greatly inhibited for the CPDs closest to the histone surface, it is greatly enhanced for the outermost CPDs. Replacing the G in a T=(m)CG CPD with A greatly decreased the deamination rate. These results show that rotational position and flanking sequence in a nucleosome can significantly and synergistically modulate CPD formation and deamination that contribute to C to T mutations associated with skin cancer induction and may have influenced the evolution of the human genome.

Figure 1.

Deamination-bypass mechanism for UV-induced C to T mutations. (A) Formation of a CPD at a T^mC site and its deamination. (B) Deamination bypass pathway for the formation of UV-induced C to T mutation at T^mCG sites in which polymerase η inserts an A opposite the T resulting from deamination of the 5-methylC in the CPD formed by UV light.

Figure 2.

System for studying effect of nucleosome structure on deamination of T^mCG CPDs. (A) Ten 147 base pair duplexes ds1–10 were constructed that would position a methyl-C (within AT^mCG) through 10 consecutive rotational positions located at the nucleosome dyad. (B) Rotational positions of the ^mC of the T^mCG sites in the nucleosome. (C) Hydroxyl radical footprinting of nucleosomes reconstituted with equal molar amounts of the ten 147-mer duplexes. Cleavage intensities are shown on the right. Arrows indicate the site of ^mC for the numbered substrate.

Figure 3.

Deamination time course with reconstituted nucleosomes. (A) Scheme for determining the extent of deamination that makes use of site-specifically ³²P-labeled ^mdC substrates ds1–10. (B) Chicken blood nucleosomes that had been exchanged with ds3 containing the internally labeled AT^mCG were irradiated to form AT=^mCG CPDs that were allowed to deaminate over time, converting ³²P-^mC to ³²P-dT within the CPD. At specific time intervals aliquots of the reaction were removed and photoreverted with photolyase and degraded by P1 nuclease. The samples were first electrophoresed on a 10% denaturing PAGE gel first to separate the digestion products by size. The band containing the ³²P-^mdC and ³²P-dT was then separated by electrophoresis in a pH 3.5 citrate gel as shown in panel (B) with the bands migrating from the bottom to the top. Panel (C) shows the plot of the extent of deamination with time that has been fit to a first order process using an aliquot of each sample that was allowed to deaminate completely to establish the total amount of CPD that was produced.

Figure 4.

Photoproduct yield and deamination half-lives as a function of rotational position. (A) Bar graph showing the relative percent of hydroxyl radical cleavage at each rotation position of the ^mC in the nucleosome core particle. (B) Bar graph of T^mC CPD yield at each rotational position compared to that in free DNA (dashed line). (C) Bar graph of the deamination rate constants for the T^mC CPDs at each rotational position compared to that in free DNA (dashed line). (D) Relative differences between the transition state free energies for hydroxyl radical cleavage, photoproduct formation and deamination at each rotational position.

Figure 5.

Orientations of the G•^mC base pairs relative to the histone surface. Orientations were approximated from the crystal structure of a nucleosome core particle 1KX5.pdb and arranged from top to bottom in order of slowest to fastest photoproduct formation and deamination. The rates fit best as a function of the angle between a vector bisecting the C6-N1 and N3-C4 bonds and a perpendicular vector to the histone core surface.

Figure 6.

Temperature factors for a high resolution nucleosome core particle structure. (A) Plot of average temperature factors for all atoms of the nucleotide in the 1KX5.pbd structure that corresponds to the indicated position of the ^mC. (B) Structure of the −3 to +6 section of the 1KX5.pdb structure shown in spacefill corresponding to the ^mC sites numbered 1–10 in this study, colored according to their relative temperature factors with red as the highest and blue as the lowest. The complementary strand, −6 to +3, is shown in wireframe and colored in the same way as for the ^mC containing strand.

Figure 7.

Mutagenic potential of T^mCG sites as a function of rotational position. The product of the relative rate of CPD formation and CPD deamination is plotted versus rotational position.