Changes in oxidative damage, inflammation and [NAD(H)] with age in cerebrospinal fluid

Abstract

An extensive body of evidence indicates that oxidative stress and inflammation play a central role in the degenerative changes of systemic tissues in aging. However a comparatively limited amount of data is available to verify whether these processes also contribute to normal aging within the brain. High levels of oxidative damage results in key cellular changes including a reduction in available nicotinamide adenine dinucleotide (NAD(+)), an essential molecule required for a number of vital cellular processes including DNA repair, immune signaling and epigenetic processing. In this study we quantified changes in [NAD(H)] and markers of inflammation and oxidative damage (F2-isoprostanes, 8-OHdG, total antioxidant capacity) in the cerebrospinal fluid (CSF) of healthy humans across a wide age range (24-91 years). CSF was collected from consenting patients who required a spinal tap for the administration of anesthetic. CSF of participants aged >45 years was found to contain increased levels of lipid peroxidation (F2-isoprostanes) (p = 0.04) and inflammation (IL-6) (p = 0.00) and decreased levels of both total antioxidant capacity (p = 0.00) and NAD(H) (p = 0.05), compared to their younger counterparts. A positive association was also observed between plasma [NAD(H)] and CSF NAD(H) levels (p = 0.03). Further analysis of the data identified a relationship between alcohol intake and CSF [NAD(H)] and markers of inflammation. The CSF of participants who consumed >1 standard drink of alcohol per day contained lower levels of NAD(H) compared to those who consumed no alcohol (p<0.05). An increase in CSF IL-6 was observed in participants who reported drinking >0-1 (p<0.05) and >1 (p<0.05) standard alcoholic drinks per day compared to those who did not drink alcohol. Taken together these data suggest a progressive age associated increase in oxidative damage, inflammation and reduced [NAD(H)] in the brain which may be exacerbated by alcohol intake.

Figure 1. Inter-correlation between CSF NAD(H), oxidative damage and inflammatory markers. (A) Positive association between CSF 8 OHdG and CSF total antioxidant capacity.

A significant positive association was observed between CSF 8(TAC) (p = 0.05, n = 37). Comparisons were made using multiple linear regression controlling for age. (B) Positive association between CSF 8 OHdG and CSF F2 Isoprostane levels. A significant positive association was observed between CSF 8 OHdG and F2 Isoprostane levels (p = 0.01, n = 34). Comparisons were made using multiple linear regression controlling for age. (C) Inverse association between CSF [NAD(H)] and CSF F2 Isoprostane levels. A significant inverse association was observed between CSF [NAD(H)] and F2 Isoprostane levels (p = 0.02, n = 48). Comparisons were made using the Pearson correlation coefficient and multiple linear regression controlling for age.

Figure 2. Positive association between plasma and CSF NAD(H) levels.

A significant positive relationship was observed between plasma and CSF NAD(H) concentrations (p = 0.03, n = 38). An increase of one µg/mL in plasma [NAD(H)] was associated with a 0.11 µg/mL increase in CSF [NAD(H)]. Comparisons were made using multiple linear regression controlling for age.

Figure 3. Association between alcohol consumption and CSF (A) [NAD(H)] (B) IL-6 levels. (A) Alcohol consumption is associated with decreased levels of CSF NAD(H).

Participants who consumed zero (n = 32), >0–1 (n = 14) and >1 (n = 8) standard alcoholic drinks per day were found to contain significantly different levels of CSF NAD(H) (p = 0.02). Specifically the CSF of participants who consumed >1 standard drink of alcohol per day contained significantly lower levels of NAD(H) compared to those who reported consuming zero drinks per day; 62.39±19.93 vs. 86.93±25.32 µg/mL respectively (p<0.05). Comparisons were made using the Kruskal-Wallis with Dunn’s post-hoc test. Error bars represent 95% confidence intervals. (B) Alcohol consumption is associated with increased levels of CSF IL-6. A significant increase in CSF IL-6 was observed in participants who drank >1 (p<0.05) and >0–1 (p<0.05) standard alcoholic drinks per day compared to those who drank zero; 2.12±1.28, 2.25±1.66 vs. 1.16±1.67 pg/mL respectively. Comparisons were made using the Kruskal-Wallis with Dunn’s post-hoc test. Error bars represent 95% confidence intervals.