Cytosine deamination and the precipitous decline of spontaneous mutation during Earth's history

Abstract

The hydrolytic deamination of cytosine and 5-methylcytosine residues in DNA appears to contribute significantly to the appearance of spontaneous mutations in microorganisms and in human disease. In the present work, we examined the mechanism of cytosine deamination and the response of the uncatalyzed reaction to changing temperature. The positively charged 1,3-dimethylcytosinium ion was hydrolyzed at a rate similar to the rate of acid-catalyzed hydrolysis of 1-methylcytosine, for which it furnishes a satisfactory kinetic model and a probable mechanism. In agreement with earlier reports, uncatalyzed deamination was found to proceed at very similar rates for cytosine, 1-methylcytosine, cytidine, and cytidine 5'-phosphate, and also for cytosine residues in single-stranded DNA generated from a phagemid, in which we sequenced an insert representing the gene of the HIV-1 protease. Arrhenius plots for the uncatalyzed deamination of cytosine were linear over the temperature range from 90 °C to 200 °C and indicated a heat of activation (ΔH(‡)) of 23.4 ± 0.5 kcal/mol at pH 7. Recent evidence indicates that the surface of the earth has been cool enough to support life for more than 4 billion years and that life has been present for almost as long. If the temperature at Earth's surface is assumed to have followed Newton's law of cooling, declining exponentially from 100 °C to 25 °C during that period, then half of the cytosine-deaminating events per unit biomass would have taken place during the first 0.2 billion years, and <99.4% would have occurred during the first 2 billion years.

Keywords: HIV-1 protease; cytosine deamination; heat mutagenesis; spontaneous mutation.

Fig. 1.

1,3-Dimethylcytosine as a potential model for the acid-catalyzed hydrolysis of 1-methylcytosine.

Fig. 2.

Influence of pH on the deamination of 1-methylcytosine (blue) and 1,3-dimethylcytosine (red) at 130 °C. Heats of ionization of acetate (pH 3–5), phosphate (pH 5.5–8.5), and bicarbonate (pH 8.7–10) buffers were used to correct pH values of the samples—actually measured at 25 °C—to the corresponding values at 130 °C. Broken lines with unit slopes have been added as a visual reference.

Fig. 3.

Temperature dependence of rate constants for the deamination of 1-methylcytosine at pH 2.45 (red) and cytosine at pH 7.0 in phosphate buffer (blue), in the range between 90° and 200 °C. For 1-methylcytosine, k₂₅ = 2.8 × 10⁻¹⁰ s⁻¹, ΔH^‡ = 23.0 kcal/mol. For cytosine, k₂₅ = 1.6 × 10⁻¹⁰ s⁻¹, ΔH^‡ = 23.4 kcal/mol.

Fig. 4.

C-to-T mutations (per 1,000 cytosine sequenced) as a function of incubation time at 90 °C. The regression line fits the equation (C-to-T mutations per thousand positions) = 0.22 + 0.35 (hours of incubation), with R² = 1.00.

Fig. 5.

Hypothetical decline in temperature (°C) as a function of time elapsed (by) since the earth's surface reached 100 °C (Eq. 1), assuming that T_∞ = 0 °C.

Fig. 6.

Decline in the rate of mutation in single-stranded DNA with time, per unit of biomass, based on an assumed first-order decline in temperature, from 100 °C at 4 billion years ago to 25 °C at present, with T_∞ = 0 °C, and ΔH^‡ = 24 kcal/mol. The broken line, added solely as a visual reference, shows the behavior that would be expected for a simple first order decline in the rate constant as a function of time.

Fig. 7.

First-order rate constants (s⁻¹) in neutral solution at 25° (blue) and 100 °C (red) for the deamination (a), depurination (u), and depyrimidination (y) of deoxyguanosine, deoxythymidine, deoxycytidine, and deoxyadenosine, from ref. . Thus, Ca, Aa, and Ga represent the deamination of deoxycytidine, deoxyadenosine, and deoxyguanosine; Cy and Ty represent cleavage of the cleavage of the linkage between a pyrimidine base and 2-deoxyribose; and Au and Gu represent cleavage of the linkage between a purine base and 2-deoxyribose.