Frailty, sex, and poverty are associated with DNA damage and repair in frail, middle-aged urban adults

Abstract

Frailty is an age-related syndrome characterized by reduced recovery from stressors and increased risks of morbidity and mortality. Although frailty is usually studied in those over 65 years, our previous work showed that frailty is both present and a risk factor for premature mortality in midlife. We identified altered gene expression patterns and biological pathways associated with inflammation in frailty. Evidence suggests DNA oxidation damage related to inflammation accumulates with age, and that DNA repair capacity (DRC) declines with age and age-related conditions. We hypothesized that inter-individual differences in DNA oxidation damage and DRC are associated with frailty status and poverty level. Using the CometChip assay, we assessed baseline single-strand breaks and hydrogen peroxide (H₂O₂)-induced DNA oxidation damage and DRC in non-frail and frail middle-aged African American and White individuals with household incomes above and below poverty. Analysis of baseline single-strand breaks showed no associations with frailty, poverty, race, or sex. However, we identified an interaction between frailty and poverty in H₂O₂-induced DNA oxidation damage. We also identified interactions between sex and frailty as well as sex and poverty status with DRC. The social determinant of health, poverty, associates with DRC in men. Baseline DNA damage, H₂O₂-induced DNA damage as well as DRC were associated with serum cytokine levels. IL-10 levels were inversely associated with baseline DNA damage as well as H₂O₂-induced DNA damage, DRC was altered by IL-4 levels and sex, and by TNF-α levels in the context of sex and poverty status. This is the first evidence that DRC may be influenced by poverty status at midlife. Our data show that social determinants of health should be considered in examining biological pathways through which disparate age-related health outcomes become manifest.

Keywords: CometChip assay; Cytokines; DNA oxidation damage; DNA repair; Frailty; Race; Single-strand breaks; Social determinants of health.

Figure 1:. DNA damage assessed by the CometChip.

A) Schematic of CometChip experimental workflow. Participant PBMCs and control TK6 cells were added to the CometChip and allowed to settle into the microwells. A section of the CometChip was treated with H₂O₂. Part of the CometChip was cut off and placed in a 37°C incubator for two hours before lysing, while the other section was placed immediately into lysis solution. B) Representative images of individual comets in the CometChip assay of baseline DNA damage, H₂O₂-induced damage, and damage remaining after cells were allowed to repair for 2 hours. Scale bar = 100 μm

Figure 2:. CometChip measurement of DNA damage and repair in cryopreserved human PBMCs.

PBMCs were assessed for baseline damage, H₂O₂-induced damage, and H₂O₂- induced then repaired DNA damage. All participants were sorted into tertiles based on baseline DNA damage. Boxplots represent the summary data for baseline damage, H₂O₂-induced damage, and H₂O₂- induced then repaired DNA damage for each baseline damage tertile. NT = baseline DNA damage

Figure 3:. Baseline DNA damage by cohort demographics.

For A) frailty status, B) poverty status, C) race, and D) sex, baseline DNA damage datapoints are represented by a scatterplot with a boxplot overlaid; the line in the middle of the box indicates median value, and the box defines values from 25th percentile to 75th percentile. The whiskers extend to the farthest point within 1.5 * the inter-quartile range.

Figure 4:. Associations between DNA damage and repair measurements with frailty, sex, and poverty.

Linear regression models were built to assess potential relationships between H₂O₂-induced DNA damage or DNA repair capacity and cohort demographics. The plots show DNA damage and repair measurements as scatter points with overlaid boxes indicating estimated values from the linear regression model and their standard errors. A) There is a significant interaction between frailty status and poverty status for H₂O₂-induced DNA damage (P = 0.023). B) There is a significant interaction between frailty status and sex for DNA repair capacity (P = 0.015). C) There is a significant interaction between poverty status and sex for DNA repair capacity (P = 0.019).

Figure 5:. Relationships between serum cytokine levels and DNA damage or repair measurements.

Linear regression models were built to assess potential relationships between DNA damage and repair measurements, serum cytokine levels, and cohort demographics. Plots show data scatter points with overlaid ribbons indicating estimated values from the linear regression model and their standard errors. A) There is a significant relationship between IL-10 and baseline DNA damage (P = 0.026). B) There is a significant relationship between IL-10 and H₂O₂-induced DNA damage (P = 0.013). C) There is a significant interaction between sex and IL-4 for DRC (P = 0.022). D) There is a significant interaction among sex, poverty status, and TNF-α for DRC (P = 0.048)

Figure 6:. Schematic representation of factors that contribute to interindividual differences in H₂O₂-induced DNA damage and DNA repair capacity.

This figure was created with BioRender.com.