ABSTRACT Oxidative stress results from an imbalance of oxidant and antioxidant homeostasis in the body. Toxic build-up of reactive oxygen species, reactive nitrogen species and free radicals can disrupt redox signalling and interfere with normal physiology. The brain is particularly vulnerable to oxidative damage, and there is strong evidence that oxidative stress plays an important role in the pathology of Alzheimer’s disease (AD), leading to the degradation of neurones and disruption of synaptic transmission.
Biomarkers of oxidative stress in patients with AD include products of lipid peroxidation, protein oxidation and DNA/RNA oxidation, samples of which are taken from the patient’s plasma, urine or cerebrospinal fluid. Evidence was limited to support the use of oxidative stress biomarkers as a diagnostic tool for AD, however, when coupled with psychological examinations assessing cognitive function, there were positive correlations between increasing systemic protein carbonyl levels and a decline in cognitive function. There is still ambiguity surrounding the clinical relevance of peripheral oxidative stress biomarkers as results are conflicting, which likely to arise from differing quantification methods, sample preparations and study populations assessed.
Biomarkers with positive results need further confirmation with larger sample sizes and studies applying them in a clinical setting. INTRODUCTION This review will appraise the current literature surrounding the presence of oxidative stress in Alzheimer’s disease patients and whether the use of biomarkers for diagnosis and disease progression are clinically relevant. Oxidative stress, defined as the result of an imbalance in oxidant and antioxidant homeostasis in the body, leads to a toxic build-up of reactive oxygen species (ROS) 1, reactive nitrogen species (RNS) and free radicals, which can disrupt redox signalling and interfere with normal physiology 2. Free radicals describe an atom or group of atoms that have at least one unpaired electron on their outer-orbital shell. The most prominent free radicals in human physiology are those derived from oxygen, known collectively as reactive oxygen species (ROS).
ROS are produced under physiological conditions as by-products of metabolism and produce intermediates required for mitochondrial oxidative phosphorylation 3. Free radicals can also be produced as a side product of arachidonic acid metabolism and nitric oxide synthesis – biochemical pathways that play important roles in the inflammatory response 4. The imbalance between the excessive production of (ROS) and the body’s antioxidant enzymes such as Superoxide Dismutase, Glutathione Peroxide and NADPH oxidase, can lead to disruption of normal physiology as radicals are highly reactive. Radicals may oxidise vital enzymes, lipid membranes and DNA, that can lead to irreversible chemical modifications and induce cellular death 5 7. The modifications of lipids and proteins during oxidative stress generate oxidation-specific epitopes.
These act as damage-associated molecular patterns (DAMPs), specifically HMGB1, that trigger an inflammatory response through pattern recognition receptor binding 7. Biomarkers are defined as an objective and quantifiable parameter that can be measured as an indicator of a biological or pathogenic process 13 14. The ability to accurately measure the level of oxidative stress through a reliable biomarker shows great potential for clinical environments. However, for a biomarker to be clinically relevant it must fulfil the following requirements: be specific for a particular disease, have a prognostic value and correlate with disease activity 5. As highlighted by Selleck et al.
, although the rate of biomarker discovery has increased significantly in recent times, the pace at which these are translated into clinical use is often slow, which leads to a gap between discovery and clinical use. This review focusses specifically on biomarkers of oxidative stress related to Alzheimer’s disease (AD), the most common neurodegenerative disease. Oxidative stress and inflammation both represent potential targets for novel therapeutics to treat AD, as the brain is particularly vulnerable to oxidative stress damage due to its high rate of oxygen consumption, low concentration of antioxidant enzymes relative to other tissues and abundance of lipids 6 8 12. Biomarkers have potential to be useful to increase the accuracy of diagnosis and act as an indirect measure of disease progression in AD patients 15. There is strong evidence that oxidative damage is important in the pathology of AD, which leads to the degradation of neurones and disruption of synaptic transmission 10 12.
Biomarkers of oxidative stress in patients with AD include products of lipid peroxidation, protein oxidation and DNA/RNA oxidation 9, samples of which are commonly taken from the patient’s plasma, urine or cerebrospinal fluid (CSF). DISCUSSION One crucial question raised when discussing the clinical relevance of oxidative stress biomarkers is where is the best place to take samples from. Samples can be taken directly from the blood plasma or serum, urine or the cerebrospinal fluid (CSF) 12.
It is argued that the CSF is the closest representation to the brain’s biochemical components as it exchanges extracellular fluid and contains molecules produced by neurons and astrocytes 12. Blood or urine, although more accessible and less invasive in regard to sample extraction, largely represent systemic oxidative stress and inflammation. Biomarkers produced by the brain will be diluted, requiring extremely sensitive methods of analysis, which are often expensive and require specialist training 12. Biomarkers of lipid peroxidation – Isoprostanes One way to quantify oxidative stress injury is to measure lipid peroxidation 16. Lipid peroxidation is the mechanism where lipids are attacked by free radical species that extract a hydrogen atom from side-chain carbon atoms. The greater the number of carbon-carbon bonds in the chain, the greater the chances of hydrogen extraction. This explains why unsaturated lipids are particularly susceptible to peroxidation. One leading theory is that free-radical oxidation of lipids contributes to the pathology of oxidative stress in the brain in neurodegenerative diseases 17 23.
Isoprostanes are novel biomarkers of oxidative stress, generated by the free-radical peroxidation of arachidonic acid (AA) and are important in the pathogenesis of AD 18 22. The measurement of F2-isoprostanes by mass spectrometry has been considered gold-standard for a marker of lipid peroxidation via oxidative stress in vivo, due to their stability 23. F2-isoprostanes are also unaffected by lipid content in the diet, providing a reliable indicator of oxidative stress. Study results have consistently shown increased levels of F2-Isoprostane levels in the CSF of AD patients. Montine et al 19 performed an in vivo investigation of quantitative biomarkers of free radical injury in latent AD.
CSF was obtained by lumbar puncture at all institutions in the morning to limit potential circadian fluctuation. CSF F2-isoprostane concentrations from 412 volunteers who ranged in age from 21 to 89 (mean age + SD = 59 ± 17 years) with an average F2-isoprostane concentration of 29.2 ± 8.4 pg/ml. By using CSF A?42 and tau as biomarkers, results obtained by Montine et al indicated the concentrations of F2-isoprostanes increased in individuals who display signs of latent AD (high CSF tau and low A?42). The validity of these results obtained could be increased by comparing the concentrations of F2-isoprostane and A?42/Tau in two separate groups: those who are clinically normal and those diagnosed with progressed AD, as some of the ‘clinically normal’ patients who showed signs of latent AD, may not actually go on to develop the disease. Additionally, the method of sample extraction via lumbar puncture may be considered invasive and not clinically viable in out-of-hospital clinics such as GP surgeries. The presence of elevated isoprostane levels in the CSF was also outlined by Grossman et al 24, who found that the increase in isoprostane was relatively specific for AD patients compared to patients with frontotemporal dementia.
The increase in isoprostane level in AD patients 61.4 (±29.7) pg/ml was significant (p<0.01) compared to healthy controls 27.8 (±8.3) pg/ml. CSF samples were taken from autopsies of AD patients, so the severity of disease progression was not known, which may have provided further insight into whether the biomarker can detect oxidative stress in latent AD patients, before serious symptoms present.Kim et al 18 measured urinary concentrations of F2-isoprostanes in AD patients against a control with normal neurologic function.
Concentrations of F2-isoprostanes were measured using gas chromatography mass spectrometry, an attractive technique for urinary sample analysis due to its specificity and sensitivity 20. Results showed a significant increase in urinary concentrations of PGF2? in AD patients versus controls. Urinary concentration of 8-isoPGF2? was elevated compared to controls, however, the difference was not significant and therefore its utility as a biomarker is reduced. In contrast, a study by Montine et al 21 found conflicting evidence that F2-isoprostanes are not reproducibly increased in the plasma and urine in patients with AD, and highlighted the results do not reflect the CNS levels of these biomarkers.
Study results regarding F2-isoprostanes as biomarkers of oxidative stress taken from plasma and urine are currently inconsistent, and although more convenient and less invasive for sample extraction, the measurement of free F2-isoprostanes in the plasma or urine do not reveal whether they originated from oxidative stress in the brain 23.Lowered CSF F2-isoprostanes suggest a reduction in oxidative stress in the brain 25. Measuring these levels of CSF F2-isoprostanes may provide utility as a biomarker to measure the effects of AD treatments in clinical trials and as a measure of cognitive decline and a preclinical marker of neurodegenerative disease 25. Isoprostanes provide clinicians with a useful biomarker for oxidative stress, particularly in AD patients when measured in the CSF. However, the utility of F2-isoprostanes as a biomarker is limited by the cost of quantification with chromatography coupled with mass spectrometry 36.
Therefore F2-isoprostanes are unlikely to be clinically relevant until the assay cost reduces. Biomarkers of DNA oxidation – 8-Hydroxy-2′-deoxyguanosine (8-OHdG)During oxidative stress, ROS/RNS can attack DNA leading to the hydroxylation of bases. 8-Hydroxy-2′-deoxyguanosine (8-OhdG) is one of the most common biomarkers of DNA oxidation and it has been observed that the deposition of A? plaques, which play an important role in the pathology of AD, follows increasing levels of 8-OhdG 27. DNA oxidation may also be caused by attacks from products of ROS-induced modifications of lipids and proteins 36. Oxidative DNA damage was measured in a study by Mecocci et al 28, who assessed the levels of 8-OhdG in peripheral lymphocytes by high-performance liquid chromatography.
DNA isolated from lymphocytes of patients with AD contained higher levels of 8-OHdG compared with the controls (P<0.001), additionally an inverse relationship between the levels of 8-OHdG in DNA and plasma levels of antioxidant proteins was identified. Methods for analysing 8-OhdG include a single-cell gel electrophoresis (SCGE) or 'comet assay', which is extremely rapid and sensitive at measuring DNA damage and is considered an economically viable technique 29. However, it is noted in the protocol that the overall procedure takes at least two days, which may reduce its relevance in a clinical environment where rapid diagnostic results are needed. Migliore et al 27 performed a study measuring the level oxidative DNA damage in peripheral leukocytes of AD and mild cognitive impairment (MCI) patients compared to healthy controls. Results showed significantly increased primary DNA damage in leukocytes for AD and also MCI patients compared with controls. This study highlighted that increased levels of oxidised DNA can be detected in MCI patients, who are considered to be in an early neurodegenerative disease state that can result in AD. This may provide clinicians with an effective pre-symptomatic diagnosis of AD risk, which could allow patients to be prescribed AD medication earlier.
However, it is noted that oxidative DNA damage decreases with the progression of AD, meaning the elevated biomarkers will only be detectable and beneficial in early-stage AD patients, meaning its use as a biomarker for AD progression would be limited. Biomarkers of protein oxidation – Protein carbonyls Neuronal damage in the brains of AD patients is associated with increased protein oxidation and nitration, and has been strongly linked to proteolysis and neuronal cell death in the disease progression of AD 30. Protein carbonylation is a chemically stable and irreversible modification and presents one of the most widely used measures of protein oxidation 31. Protein carbonyls can either be formed by the oxidation of amino acid side chains, or through oxidative cleavage of proteins 2 31. A significant advantage of using protein carbonyls as a biomarker of oxidative stress is that most assays do not require specific or expensive equipment and can be performed in a standard biochemistry laboratory 2 32. The detection of protein carbonyls is highly sensitive and involves derivatisation with 2,4-dinitrophenylhydrazine (DNPH), leading to the formation of 2,4-dinitrophenyl (DNP). The DNP group absorbs UV light and can be quantified using a spectrophotometric assay 32. Alternatively, an enzyme-linked immunosorbent assay (ELISA) can be performed for levels of anti-DNP antibodies to quantify total protein carbonyl groups.
ELISA’s are considered more sensitive than DNPH spectrophotometric assays, as free DNPH and non-proteins can be easily removed through washing to reduce interference 32. Relative ease of use and detection provided by this assay shows strong signs of clinical relevance, as protein carbonyls could be quantified from patient samples in a clinical setting. Ahmed et al 30 quantified the levels of protein oxidation and nitration products in the CSF of AD patients against controls using liquid chromatography/mass spectrometry.
The sample size for this study was small, which may have been due to the invasive nature of lumbar punctures reducing the number of willing study participants. Results showed that concentrations of 3-Nitrotyrosine (+60%), N?-carboxymethyl-lysine (+71%), 3DG-H (+57%) and N-formylkynurenine (+15%) in the CSF of AD patients were all significantly increased compared with normal controls (P <0.05). From performing a mini-mental state examination (MMSE) to assess cognitive function, Ahmed et al found a significant negative correlation between MMSE score and concentration of fructosyl-lysine, which has previously been identified to increase with rising severity of AD 30. This study provides evidence that the quantification of protein carbonyls in the CSF can be used as biomarkers of oxidative stress and relate to cognitive function in AD patients. There is potential for this to be integrated into a clinical setting to improve the diagnosis of cognitive decline in early AD, thus allowing patients to receive therapeutic intervention before severe symptoms arise. Cristalli et al 34 measured levels of protein carbonyls in the peripheral blood of 110 AD patients.
The data concluded that the level of protein carbonyls was significantly higher in AD patients compared with controls in plasma, erythrocyte and leukocyte samples. In addition, subjects were categorised based on the severity of their disease (mild, intermediate, severe), which was determined by neurological examinations alongside an MMSE and classified according to the Global Deterioration Scale. This proved extremely useful in presenting how the concentration of oxidative stress biomarkers varied not just in AD patients alone, but also depending on their disease progression. Cristalli et al showed the concentration of protein carbonyls in plasma, erythrocyte and leukocyte samples correlated with an increased in disease severity (decrease in MMSE). This inverse correlation may provide clinicians with useful data that has the potential to support a patient’s diagnosis, but also their disease progression.
3-Nitrotyrosine Another biomarker of protein oxidation/nitration is 3-nitrotyrosine (3-NT). Nitration of tyrosine is a stable marker of oxidative/nitrative stress and can occur within polypeptide sequences or free tyrosine amino acids 2. Peroxynitrite (ONOO-), a common reactive nitrogen species (RNS), is formed in the reaction between nitric oxide NO• and superoxide O2 •-. Peroxynitrite can go onto react with tyrosine residues, which can be measured through analysing the levels of 3-NT product with HPLC/MS 33 35. ELISA assays are available for 3-NT quantification, but their utility is limited by low sensitivity and differing affinities of antibodies for different nitrated proteins 35.Literature surrounding 3-NT as a biomarker of oxidative stress in patients with AD is conflicting and often includes studies with small sample sizes. For example, Korolainen and Pirttilä 36 and Ryberg et al 33 conducted studies with just 22 and 17 AD patients respectively. Neither studies found a significant difference in the concentration of nitrated protein in the CSF or plasma between AD and control, which may have been due to unrepresentative sample sizes.
On the other hand, the study by Ahmed et al 30 had a slightly higher sample size (n = 32) and found significantly higher nitrated protein in AD patients (p < 0.05). Ryberg et al outlined that human CSF contains nitrites, meaning in vitro formation of 3-NT is possible, which may increase sample 3-NT. Therefore, preparation techniques will likely be more reliable if 3-NT is reduced after sample extraction to prevent further nitration, thus its clinical relevance as a biomarker of oxidative stress is limited.
CONCLUSION In conclusion, oxidative stress biomarkers are found to increase in patients with Alzheimer’s disease, however their utility and clinical relevance to study disease progression in AD patients is debatable. Despite the aforementioned biomarkers being largely non-specific for oxidative stress resulting from AD – thus limiting their utility as a direct diagnostic indicator – the quantification of protein carbonyls in the CSF appears to be the most clinically relevant from the literature cited. Its utility alongside a mini-mental state examination, can provide information to improve the diagnosis of cognitive decline in early AD, which would allow patients to receive therapeutic intervention before severe symptoms arise. Additionally, its measurement could be used in a clinical environment to assess disease progression in response to therapeutics. It is argued that oxidative stress biomarkers obtained through CSF samples provide the closest representation of oxidative stress in the brain resulting from AD. However, on the whole, there is still ambiguity surrounding the use of peripheral oxidative stress biomarkers as some studies discourage their use whilst others consider them to be clinically relevant, especially when diagnosing patients with mild cognitive decline and latent AD. Differences in results are likely to arise from differing quantification methods, sample preparations and study populations assessed. Further research is needed to examine the utility of oxidative stress biomarkers as indirect markers of disease progression in AD patients, before they can be applied to a clinical setting or clinical trials as the data is still conflicting.
Biomarkers with positive findings such as protein carbonyls still need further confirmation with larger sample sizes and studies applying them in a clinical setting.