The Roles of Biomarkers of Oxidative Stress and Antioxidant in Alzheimer's Disease: A Systematic Review

Purpose. Oxidative stress plays an important role in the pathogenesis of Alzheimer's disease (AD). This paper aims to examine whether biomarkers of oxidative stress and antioxidants could be useful biomarkers in AD, which might form the bases of future clinical studies. Methods. PubMed, SCOPUS, and Web of Science were systematically queried to obtain studies with available data regarding markers of oxidative stress and antioxidants from subjects with AD. Results and Conclusion. Although most studies show elevated serum markers of lipid peroxidation in AD, there is no sufficient evidence to justify the routine use of biomarkers as predictors of severity or outcome in AD.


Introduction
Alzheimer's disease is the most common form of dementia in the elderly and is characterized by a progressive loss of cognitive capacity and severe neurodegeneration. The pathophysiologic process is posited to be initiated by extracellular fibrillary -amyloid deposition, with subsequent intraneuronal hyperphosphorylated tau protein aggregation [1]. Mutations in the amyloid precursor protein (APP), presenilin-1 (PS1), or PS2 that alter APP metabolism favor the production of a fibrillary form, A . Such findings form the basis of the amyloid cascade hypothesis of AD pathophysiology [2].
Although this amyloid cascade hypothesis may be the underlying pathogenesis for the familial form of AD, increasing evidence suggests that oxidative stress has a key role in late-onset sporadic forms, which are the majority of AD cases. Abnormal levels of oxidative stress have been reported in Alzheimer's disease in both the brain and blood stream [3,4]. Changes in Alzheimer's disease that produce a prooxidative imbalance have been attributed to decrease in antioxidant defenses, toxicity related to amyloid-, and/or altered metal metabolism in the brain and peripheral tissues [3,4] ( Figure 1).
The possible benefits of biomarkers in clinical practice include outcome prediction in AD patients that may further influence therapeutic regimens. The aim of this review is to determine whether biomarkers of oxidative stress can play an important prognostic role in the outcome of AD. The successful translation of these approaches to the clinics offers the promise of not only improving outcome prediction but also a more scientific basis for therapeutic options.

Methods
Studies were identified from a systematic search of PubMed, Scopus databases, Google Scholar, and the reference lists of all included studies and major relevant review papers. To find all of the relevant articles, PubMed was searched using the key words: "TBARS, " "oxidized LDL, " "protein carbonyls, " "8-HOG, " "antioxidant, " and "Alzheimer's disease" in various combinations. Case-control studies with human subjects were considered for inclusion. The articles selected were published in English between January 1985 and September 2013.

Biomarkers of Lipid Peroxidation.
Lipid peroxidation is one of the major consequences of oxidative imbalancemediated injury to the brain. It causes changes in the fluidity and permeability of cell membranes and impairs the activity of membrane-bound enzymes. Lipid peroxidation also leads to the production of conjugated diene hydroperoxides and unstable substances that disintegrate into various aldehydes like malondialdehyde, 4-hydroxynonenal, and TBARS.
Several studies demonstrate that serum or plasma TBARS level in AD subjects is significantly higher than in controls [7,13,15,[20][21][22][23][24][25][26][27][28], while others observe no significant difference between AD subjects and controls [29][30][31][32][33][34][35] (Table 1). Results regarding erythrocyte TBARS level in AD are also controversial. Some studies observe higher erythrocyte TBARS levels in AD [5,7,31], while others observe no difference between AD and controls [6,15,29,36]. A meta-analysis regarding blood TBARS level in AD and mild cognitive impairment reveals that TBARS levels are significantly elevated in Alzheimer's disease plasma/serum [37]. However, findings may vary by different patients selection criteria: some researchers observe AD subjects with MMSE 7-24 points do not have significantly higher serum TBARS than controls [34], while others observe AD subjects with 7-20 points as well as ADAS-cog 10-35 points have significantly higher serum TBARS than controls [20]. These suggest that MMSE alone is not enough to discriminate those with higher TBARS from those with lower TBARS. The lack of a link between MMSE and TBARS is also reported by another study, which suggests that MMSE is not correlated with TBARS [7], while plasma TBARS level may actually increase with the severity of cognitive dysfunction in AD [7].  [15] 20 AD pts (mean age = 80.1). Follow-up at memory center for more than 6 months, no evidence of nutritional behavior, mean duration of AD was 3.9 years, 19 pts on cholinergic therapy. 23 controls, mean age 76.0

Plasma/ erythrocytes
Significantly higher in AD in plasma level ( = 0.036). No significant difference between AD and control in erythrocytes level. Ox-LDL has been suggested to be produced by oxidized phospholipids released from brain tissue into circulation [38]. Ox-LDL is a promising marker of oxidative injury of the whole body. It may also be a peripheral marker that is linked to the severity of oxidative damage in the presence of dementia [23,39,40]. Serum ox-LDL level is universally higher in AD than in controls in the three studies.

Biomarkers of Antioxidants.
Antioxidants are suggested as potential indirect markers of oxidative stress processing in the brain of patients with AD (Table 3). Oxidative stress has been speculated to cause antioxidant consumption and thus, a decline in antioxidant levels [51]. Nonenzymatic compounds with antioxidant properties include vitamin A/carotenoids, vitamins C and E, and uric acid. On the other hand, antioxidant enzymes that vary on condition of oxidative stress in AD remain unsettled, since antioxidant enzymes like glutathione peroxidase (GPx) and superoxide dismutase (SOD) may be induced by oxidative stress (to increase their level or activity) or consumed (to decrease their level or activity) [37].

Uric Acid.
Three studies elaborate that plasma or serum uric acid level is significantly lower in AD [52][53][54], while three other studies do not observe this difference [29,55,56]. It is possible that excluding patients with metabolic syndrome plays an important role in measurements of significantly lower uric acid level in AD [52,54]. This suggests that metabolic syndrome may interfere with the level of uric acid as an indirect marker of oxidative stress in AD.

Vitamin E.
Most of the studies in the literature report that plasma or serum vitamin E level is significantly lower in AD [7,9,10,15,16,25,54,[57][58][59][60]. However, vitamin E supplementation does not seem to improve prognosis in AD [61]. So far, erythrocyte and platelet vitamin E levels are not different between AD and controls [15,29].

Vitamin C.
Vitamin C concentration in plasma or serum is also found to be significantly lower in AD in most literature [10,16,25,54,60,62,63]. Furthermore, vitamin C is especially significantly lower in moderate and severe AD [62].

Vitamin A.
All of the studies on serum or plasma vitamin A levels establish a significant difference between AD and control [10, 16, 25, 54, 63, 64].

Superoxide Dismutase (SOD).
As an antioxidant enzyme, SOD may be induced or consumed by oxidative stress [65]. It is one of the most studied antioxidant enzymes in AD. Several studies find no difference in SOD between AD and controls [15,29,31,36,66]. Some find significantly lower erythrocyte level [5,11,54,67] and plasma/serum level [11,26] in AD, while others find significantly higher erythrocyte [6,7,21,22,[68][69][70][71] and plasma [28] levels in AD. However, only three among these studies have a case number more than 80 [7,22,68] and all of them either include only mild-to-moderate AD [22,68] or have clear severity-classification of AD [7]. All three demonstrate significantly higher erythrocyte SOD level in AD [7,21,22]. One study further establishes that leukocyte SOD level is higher in moderate AD than in mild AD, higher in mild AD than in controls, and higher in mild AD than in severe AD [7]. This suggests that SOD level is induced by oxidative stress in the early stages of AD and is consumed in the later stage. Studies with different findings may be due to limitations of small sample size [26,66] or loose inclusion criteria [15,21,29,36,54].

Glutathione Peroxidase (GPx)/Reduced Glutathione (GSH).
Glutathione peroxidase, another antioxidant enzyme, may be also induced or consumed under conditions of oxidative stress [65]. Some studies demonstrate no difference in GPx level between AD and controls [15,22,29,31]. Some demonstrate lower GPx in AD [26-28, 36, 54], while others observe higher GPx [5,7,24,27]. As for the balance between reduced and oxidized GSH (GSSG), some studies show no significant difference between AD and controls [6,20,29], while others demonstrate a balance towards GSSG in AD with statistical significance [7,32,41,72,73]. The only study with a case number more than 100 demonstrates significantly higher plasma, erythrocyte, and leukocyte GPx and GSSG levels in severe AD than in moderate AD, in moderate AD than in mild AD, and in mild AD than in controls [7]. Other studies may be limited by their small sample size [6,15,26] or different patient characteristics [20,31,54].

Biomarkers of DNA Oxidation .
8-OHdG is the most commonly studied biomarker for oxidative DNA. Most studies in the literature reveal significantly higher 8-OHdG in AD [4,16,74]. Among the three, two studies demonstrated significantly higher peripheral 8-OHdG in lymphocytes and leukocytes [4,16].

Biomarkers of Central Nervous System.
As for CSF antioxidant, the difference of vitamin E level between AD and controls remains controversial [58,64]. Although one study shows no difference in CSF vitamin C level between AD and controls [75], two other studies reveal that CSF vitamin C level is lower [60,64] and one study even has negative findings, which may be due to the small sample size [75]. Only one study about CSF vitamin A level shows no difference between AD and control [64].
In terms of markers of oxidative stress measured closer to the brain, one of the studies that measured brain 8-OHdG directly found elevated 8-OHdG in AD [74]. CSF protein carbonyl level is universally significantly higher in AD than in controls [18,45,46]. Furthermore, several studies measure protein carbonyls directly in the brain postmortem [14,43,47]. Although most studies on the hippocampus have found exclusively significantly higher protein carbonyl levels in AD, one study on other brain regions such as the neocortex, amygdala, brainstem, and cerebellum has also

Brain
Significantly higher in mild AD and MCI than preclinical and control.
found significantly increased protein carbonyl levels in AD [47]. Regarding nitrated protein that represents protein peroxidation, there is no consistency among different research groups on assessing nitrated protein level in AD. Although direct measurement from brain tissue postmortem reveals consistently significantly higher nitrated protein in AD [19,43,49], only two studies demonstrate significantly higher CSF nitrate protein level in AD [46,48]. Two other studies report no difference between AD and controls [18,50].

Conclusions
Most studies show that serum markers of lipid peroxidation are elevated in Alzheimer's disease. However, there is insufficient evidence to justify the routine use of biomarkers as predictors of severity or outcome in AD.