Markers of oxidative damage to lipids, nucleic acids and proteins and antioxidant enzymes activities in Alzheimer's disease brain: A meta-analysis in human pathological specimens

Abstract

Oxidative stress and decreased cellular responsiveness to oxidative stress are thought to influence brain aging and Alzheimer's disease, but the specific patterns of oxidative damage and the underlying mechanism leading to this damage are not definitively known. The objective of this study was to define the pattern of changes in oxidative-stress related markers by brain region in human Alzheimer's disease and mild cognitive impairment brain tissue. Observational case-control studies were identified from systematic queries of PubMed, ISI Web of Science and Scopus databases and studies were evaluated with appropriate quality measures. The data was used to construct a region-by-region meta-analysis of malondialdehyde, 4-hydroxynonenal, protein carbonylation, 8-hydroxyguanine levels and superoxide dismutase, glutathione peroxidase, glutathione reductase and catalase activities. We also evaluated ascorbic acid, tocopherol, uric acid and glutathione levels. The analysis was complicated in several cases by publication bias and/or outlier data. We found that malondialdehyde levels were slightly increased in the temporal and occipital lobes and hippocampus, but this analysis was significantly impacted by publication bias. 4-hydroxynonenal levels were unchanged in every brain region. There was no change in 8-hydroxyguanine level in any brain region and protein carbonylation levels were unchanged except for a slight increase in the occipital lobe. Superoxide dismutase, glutathione peroxidase and reductase and catalase activities were not decreased in any brain region. There was limited data reporting non-enzymatic antioxidant levels in Alzheimer's disease brain, although glutathione and tocopherol levels appear to be unchanged. Minimal quantitative data is available from brain tissue from patients with mild cognitive impairment. While there is modest evidence supporting minor regional changes in markers of oxidative damage, this analysis fails to identify a consistent pattern of pro-oxidative changes and accumulation of oxidative damage in bulk tissue analysis in the setting of Alzheimer's disease, as has been widely reported.

Keywords: Ascorbic acid; Carbonylation; Catalase; Glutathione; Hydroxyguanine; Hydroxynonenal; Malondialdehyde; Mild cognitive impairment; Superoxide dismutase; Tocopherol.

Figure 1:. Malondialdehyde levels in Alzheimer’s disease brain

In the raw analysis, malondialdehyde levels appear to be significantly increased in the frontal, temporal and occipital lobes and in the hippocampus. There is significant evidence of publication bias in the analyses for the frontal, temporal and parietal lobes – correction for this bias reduces the observed increase in the frontal and temporal lobes, although both remain significantly increased. The analysis of the occipital lobe is impacted by an outlier study; were the analysis repeated without Balazs 1994, it would no longer suggest a significant increase. There is no statistical heterogeneity in the analyses, except in the temporal lobe which is highly heterogeneous (p<0.0001). The increase in malondialdehyde level in the hippocampus is not impacted by publication bias, heterogeneity or outlier effects.

Figure 2:. Hydroxynonenal and protein carbonylation levels in Alzheimer’s disease brain

There is no significant increase in 4-hydroxynonenal in any studied brain region, although there is significant heterogeneity in the analysis. Protein carbonylation levels are similarly unchanged in most studied brain regions with the exception of the occipital lobe which has a 12% increase in protein carbonylation, p=0.04.

Figure 3:. Nuclear 8-hydroxyguanine levels in Alzheimer’s disease brain

8-hydroxyguanine levels are unchanged in AD. Three studies (Gabbita 1998, Wang 2005 and Bradley-Whitman 2013) were contributed by the University of Kentucky (UK); these were aggregated prior to inclusion in this analysis except in the cerebellum where the results of the three studies were highly discordant.

Figure 4:. Superoxide dismutase activity in Alzheimer’s disease brain

Superoxide dismutase activity is significantly increased only in the parietal lobe. One study has been excluded because the assay used measured total protein abundance rather than enzymatic activity.[89] Additionally, Richardson 1993 (abstract only) and Ramassamy 1999 (data not reported, but described as unchanged) were excluded

Figure 5:. Glutathione peroxidase and reductase activities in Alzheimer’s disease brain

There is no evidence of significant change in glutathione peroxidase activity in any brain region. There was no evidence of heterogeneity of publication bias in this analysis. For glutathione reductase, adequate data to perform meta-analysis was only available for the brain regions shown. There was no evidence of a significant change in glutathione reductase activity in any brain region in Alzheimer’s disease.

Figure 6:. Catalase activity in Alzheimer’s disease brain

There is no evidence of significant change in catalase activity in any brain region in Alzheimer’s disease. There is significant heterogeneity in the data in the cerebellum and temporal and parietal lobes. It is difficult to evaluate to publication bias given the small number of studies, but for the frontal lobe which has the largest number of studies, there is no evidence of publication bias.

Figure 7:. Non-enzymatic antioxidant levels in Alzheimer’s disease brain

The major finding of this portion of the analysis is that there is not much data available describing the levels of non-enzymatic antioxidants in the brain. There is enough data for glutathione to make an analysis in three regions (the frontal lobe, occipital lobe and hippocampus) and there is no change in glutathione level in any of these regions. Three studies have measured alpha-tocopherol levels in the brain. There is not enough data to perform a region-by-region analysis, but we have lumped all the regions together to produce a general estimate of alpha-tocopherol levels in Alzheimer’s disease brain – there does not appear to be a significant change. Additionally, alpha-tocopherol level in CSF was unchanged. Finally, we were unable to find a single study reporting ascorbic acid levels in the brain in Alzheimer’s disease, but five studies have evaluated CSF and the ascorbic acid level in CSF in Alzheimer’s disease appears significantly depleted.