Phenotypic and genetic consequences of protein damage

Abstract

Although the genome contains all the information necessary for maintenance and perpetuation of life, it is the proteome that repairs, duplicates and expresses the genome and actually performs most cellular functions. Here we reveal strong phenotypes of physiological oxidative proteome damage at the functional and genomic levels. Genome-wide mutations rates and biosynthetic capacity were monitored in real time, in single Escherichia coli cells with identical levels of reactive oxygen species and oxidative DNA damage, but with different levels of irreversible oxidative proteome damage (carbonylation). Increased protein carbonylation correlates with a mutator phenotype, whereas reducing it below wild type level produces an anti-mutator phenotype identifying proteome damage as the leading cause of spontaneous mutations. Proteome oxidation elevates also UV-light induced mutagenesis and impairs cellular biosynthesis. In conclusion, protein damage reduces the efficacy and precision of vital cellular processes resulting in high mutation rates and functional degeneracy akin to cellular aging.

Figure 1. Proteome carbonylation correlates with cellular biosynthetic capacity.

Exponentially growing E.coli strains with different levels of chaperone activity and translation errors show different levels of (a) protein carbonylation and (b) bacteriophage λ single burst size. (c) (χ) There is a negative correlation between total proteome carbonylation and biosynthetic capacity measured as single burst size of bacteriophage λ. Strain identity corresponding to the numbers is listed in (d). Results are means of 3 measurements, each in triplicate. The error bars represent the standard deviation. R² value of the linear fit is indicated in panel (c). “Oe” stands for over-expression.

Figure 2. Mutation rate correlates with total proteome carbonylation.

(a) The fraction of cells with a MutL-CFP focus (mutation rate) decreases with increasing chaperone activity and translation accuracy in wild type (white bars) and MutH deficient E.coli (black bars). (b) 8-oxo-guanine and ROS (DHR123 fluorescence) levels in each strain, in the absence and presence of 1 mM trolox, (c) Correlation between the fraction of cells with MutL-CFP foci and total proteome carbonylation. Strain identity corresponding to the numbers is listed in Figure 1D. Results are given as mean of 3 measurements, each in triplicate. Error bars represent the standard deviation. “Oe” stands for over-expression.

Figure 3. Trolox reduces the amount of protein carbonylation and the mutation rates.

Parallel decrease in mutation rate (white and black circle) (fraction of cells with a MutL-CFP focus) and in constitutive protein carbonylation (white and black square) in the presence of 1 mM trolox (black symbols) relative to no trolox controls (white symbols). The results are given as mean of 3 measurements, each in triplicate. Error bars represent the standard deviation.

Figure 4. UV-induced mutation frequencies correlate with total proteome carbonylation.

(a) Dose response of UVC induced mutation frequency in E.coli strains with increasing chaperone activity and translation accuracy. (b) Protein carbonylation at the end of the recovery period correlates with the UVC radiation dose. (c) Correlation between the UVC induced mutation frequency and protein carbonylation measured immediately after irradiation. Spontaneous mutation frequencies for each strain are labelled with an X. The results of mutation frequencies are given as median of 3 measurements, each in triplicate. Protein carbonylation measurements are presented as mean of 3 measurement, each in triplicate. Error bars represent the standard deviation. “Oe” stands for over-expression.

Figure 5. Protein damage reduces the efficiency of DNA repair.

The level of 8-oxo-guanine in DNA (a) immediately after irradiation and (b) at the end of the recovery period when it increases differently with the UVC radiation depending on the fidelity of protein biosynthesis and chaperone activity. (c) 8-oxo-guanine level correlates with the protein carbonylation level immediately after irradiation. (d) 8-oxo-guanine level after post-irradiation incubation correlates with protein carbonylation. 8-oxo-guanine and protein carbonylation levels prior to irradiation are labeled with an X. The results are given as mean of 2 measurements, each in triplicate. Error bars represent the standard deviation. R² value of the linear fit is indicated in panel (a). “Oe” stands for over-expression.