Age-dependent accumulation of dicarbonyls and advanced glycation endproducts (AGEs) associates with mitochondrial stress

Abstract

Aging is a strong risk factor for brain dementia and cognitive decline. Age-related accumulation of metabolites such as advanced glycation end products (AGEs) could serve as danger signals to initiate and accelerate disease process and neurodegeneration. The underlying causes and consequences of cerebral AGEs accumulation remain largely unknown. Here, we comprehensively investigate age-related accumulation of AGEs and dicarbonyls, including methylglyoxal (MG), glyoxal (GO), and 3-deoxyglucosone (3-DG), and the effects of mitochondrial reactive oxygen species (ROS) on cerebral AGEs accumulation, mitochondrial function, and oxidative stress in the aging human and mouse brain. We demonstrate that AGEs, including arginine and lysine derived N(6)-carboxymethyl lysine (CML), Nε-(1-Carboxyethyl)-l-lysine (CEL), and methylglyoxal-derived hydro-imidazolone-1 (MG-H1), were significantly elevated in the cerebral cortex and hippocampus with advanced age in mice. Accordingly, aging mouse and human brains revealed decrease in activities of mitochondrial respiratory chain complexes I & IV and ATP levels, and increased ROS. Notably, administration of mitoTEMPO (2-(2,2,6,6-Tetramethylpiperidin-1-oxyl-4-ylamino)-2-oxoethyl)triphenylphosphonium chloride (mTEMPO), a scavenger of mitochondrial ROS, not only suppressed ROS production but also reduced aged-induced accumulation of AGEs and dicarbonyls. mTEMPO treatment improved mitochondrial respiratory function and restored ATP levels. Our findings provide evidence linking age-related accumulation of toxic metabolites (AGEs) to mitochondrial oxidative stress. This highlights a novel mechanism by which AGEs-dependent signaling promotes carbonyl stress and sustained mitochondrial dysfunction. Eliminating formation and accumulation of AGEs may represent a new therapeutic avenue for combating cognitive decline and mitochondrial degeneration relevant to aging and neurodegenerative diseases including Alzheimer’s disease.

1. Introduction

Aging is associated with diverse deleterious changes that lead to the accumulation of toxic metabolites, resulting in gradual decline in the organism’s ability to adapt to metabolic stress, progressive loss of biological and physiological functions, increased susceptibility to disease and eventual death [1–3]. Advancing age is the greatest risk factor of developing neurodegenerative dementias, including Alzheimer’s disease (AD). The rapid growth of the aging population and the socioeconomic and personal cost of health care for people with AD make it increasingly important to understand the aging process and the mechanisms of aging linked to age-related neurodegenerative disease.

Glycation is spontaneous process that occurs endogenously as the result of inevitable production of metabolic by-products, which contributes to the post-translational modification of proteins [4]. Advanced glycation endproducts (AGEs) are the irreversible products of non-enzymatic glycation of proteins by reducing sugars or dicarbonyls [4–7]. Dicarbonyls such as methylglyoxal (MG), glyoxal (GO), and 3-deoxyglucosone (3-DG) are formed endogenously and are known as reactive AGE precursors [4,8,9]. Such dicarbonyl species are more reactive than sugars making them the most important glycating agents in the rapid modification of cellular and extracellular proteins to generate AGEs [4]. Aging disrupts the balance between the formation and natural clearance of AGE’s and their precursors (dicarbonyls), resulting in the accumulation of AGEs and dicarbonyls. AGEs and dicarbonyls play roles in the development and progression/aggravation of many degenerative diseases, including diabetes, atherosclerosis, cardiovascular disease, and AD [8,10–14]. AGEs, such as Nε-carboxymethyl-lysine (CML) and Nε-(1-carboxyethyl) lysine (CEL), have been detected in intracellular protein deposits in neurofibrillary tangles [15,16] and the cerebrospinal fluid [17]. AGEs derived from arginine residues modified by methylglyoxal are known as methylglyoxal hydroimidazolone (MG-H1) and are accumulated in both protein and free adduct forms to a quantitatively greater extent than other AGEs [18,19]. In AD, AGEs modify crosslinking of long-lived proteins such as β-amyloid peptide (Aβ) and hyperphosphorylated tau protein, which are hallmarks of AD pathology [12–14,20]. In addition to accumulation of damaged macromolecules in cells and tissues, one common denominator of aging is mitochondrial dysfunction. Progressive glycation leads to impairment in mitochondrial energy production and increased oxidative stress. Excessive AGEs and dicarbonyl compounds have detrimental effects on mitochondrial respiratory function and the permeability of mitochondrial membranes [5,21]. Thus, age-related accumulation of harmful metabolites such as AGEs could be endogenous danger signals that initiate and/or promote mitochondrial perturbation and dementia. Several lines of evidence suggest that there are specific mitochondrial targets of glycation. Mitochondrial dysfunction itself has been implicated in disease and aging, however, the molecular mechanisms underlying glycation of biomolecules induced-mitochondrial dysfunction in the aging brain remain elusive.

In the present study, we comprehensively investigated the accumulation of cerebral AGEs and reactive dicarbonyls in aging mouse and human brains and the effect of mitochondrial ROS on the accumulation of diacarbonyls and AGEs and mitochondrial function. These studies addressed the following key questions: 1) Are the levels of AGE adducts and dicarbonyls increased with aging? 2) Do reactive dicarbonyls and AGE adducts contribute to age-related oxidative stress and mitochondrial dysfunction? 3) Could mitochondrial targeted antioxidant mitigate accumulation of AGE adducts and ameliorate mitochondrial function? Our studies provide substantial evidence that increased accumulation of dicarbonyls and AGE adducts results in oxidative stress and mitochondrial dysfunctions with advancing age and, in contrast, that the administration of mitochondria-specific antioxidants reverses these detrimental effects.

2. Material and methods

2.1. Animal study

Both male and female wild type (WT) mice (Charles River C57BL/6 strain) from 3 to 30 months of age were used in this study. Mice were sacrificed by cervical dislocation and brain tissues were collected and snap frozen in liquid nitrogen prior to storage at ‒80 °C for later analysis in the described experiments. All studies on mice were performed in accordance with the National Institutes of Health guidelines for animal care with the approval of the Institutional Animal Care and Use Committee of the University of Kansas-Lawrence and Columbia University.

2.2. Pharmacological treatment

In vivo experiment, MitoTEMPO [(2-(2,2,6,6-Tetramethylpiperidin-1-oxyl-4-ylamino)-2-oxoethyl)triphenylphosphonium chloride (mTEMPO), Catalog #SML0737; Sigma–Aldrich] was dissolved in sterile 0.9% saline, and then passed through a 0.22 μm filter to sterilize the mTEMPO solution prior to use. Wild-type C57/B6 mice at 21 months of age were subjected to intraperitoneal injection of mTEMPO at a dose of 0.25 mg/kg/day or sterile saline (Vehicle) for 3 months. Brain tissues were collected and stored as above.

In vitro brain slice experiment, mTEMPO was prepared at 10 mM in ddH₂O as a stock solution and passed through a 0.22 μm filter. The sterile mTMEPO solution was made small aliquots and kept at ‒20 °C. Brain slices including cortex and hippocampus (coronal section, 400 μm in thickness) prepared from mice were kept in an interface chamber at 29 °C and perfused with oxygenated artificial cerebrospinal fluid (ACSF) continuously bubbled with 95% O₂ and 5% CO₂. The composition of the ACSF was: 124 mM NaCl, 4.4 mM KCl, 1 mM Na₂HPO₄, 25 mM NaHCO₃, 2 mM CaCl₂, 2 mM MgCl₂ and 10 mM glucose. Then, brain slices were pre-treated with Mito-TEMPO (2 μM) or vehicle for 10 min followed by treatment with MG-derived AGE (100 μg/ml) prepared as described [5] or MG (1 μM) for 1 h during perfusion. Afterwards, the treated brain slices were snap frozen in liquid nitrogen and stored at ‒80 °C for later analysis.

2.3. Human tissues

Human brain tissues of the temporal cortex (temporal pole, including Brodmann area 38, which is the apparent rostral origin of the superior, middle and inferior temporal gyri) from aged individuals and young controls were obtained from the New York Brain Bank and ADRC at Columbia University (Detailed information for each of the cases studied is shown in Table 1). Informed consent was obtained from all subjects.

Table 1.

The information on the human brain tissues used in the experiments.

Case	Gender	Age(yr)	PMI(hr)	Break stage	Neuritic plaques- Temporal pole	Neuritic plaques-Hippocampus
NO1	M	89	4.8	0	Very rare	Absent
NO2	M	62	4.3	0	Absent	Absent
NO3	M	78	5.3	0	Absent	Absent
NO4	F	76	3.6	0/I	Absent	Absent
NO5	M	89	3.0	I/I	Absent	Absent
NO6	M	89	4.9	II/II	Very rare	Very rare → none in subiculum
NO7	F	84	5.6	0	Diffuse → none	Rare→none in subiculum
NO8	F	89	4.3	II/0	Absent	Absent
Mean ± SE	3F/5M	82 ± 3.41	4.475 ± 0.305
YC1	F	33	3.7	0	Absent	Absent
YC2	F	33	8	0	Absent	Absent
YC3	M	41	1.7	0	Absent	Absent
YC4	F	28	2.1	0	Absent	Absent
Mean ± SE	3F/1M	33.75 ± 2.69	3.875 ± 1,44

Note: NO, normal old; YC, young control; F, female; M, male; PMI, postmortem interval; SE: standard error.

2.4. Immunodetection of AGE-adducts

We use immunodot blot to represent the distribution of groups of AGE modified protein as points on a simple scale, allowing us to confirm the total amount of glycation adducts in each experimental group. Use of immunodot blot to determine AGE levels were described in our previous study [5]. Briefly, equal amounts of total protein extracts (2 μg) from cortex or hippocampus of mouse brains at 3, 6, 12, 20 and 30 months of age were spotted onto nitrocellulose membranes (Bio-Rad). β-actin (Cat #A5441, Sigma-Aldrich) probe of corresponding dot blots was used as a total protein loading control. Primary antibodies against total AGEs (generated in our lab) [5], MG-AGE (Cat #STA 011, Cell Biolabs), MG-H1 (Cat #HM5017, Hycult Biotech), CML (Cat #KAL-KH024, Cosmo Bio), or CEL (Cat #HPA008023, Sigma-Aldrich) at 1:3000, the corresponding secondary antibody HRP-conjugated anti-guinea pig, anti-rabbit or anti-mouse IgG at 1:10000, and Chemiluminescent Substrate (Catalog #34580, Thermo Fisher Scientific) were used to detect AGE-adducts of proteins. Protein signals were visualized using a FluorChem HD2 imaging system. The intensity of signal was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). For ELISA, total AGEs, MG-AGEs and CML in mouse brain cortex or human brain homogenates were determined using mouse AGEs, MG and CML ELISA kit (Cat #MBS704846 and #MBS756134 and #MBS265592, MyBioSource) or human AGEs, MG and CML ELISA kit (Cat #LS-F4568–1, Lifespan Biosciences, Inc, #MBS2602652, MyBioSource and #E12798h-96, Lifeome Biolabs), respectively, according to the manufacturer’s protocol.

2.5. RP-HPLC analysis of α-dicarbonyl compounds in brain tissues

A small piece of brain tissue was suspended in 350 μl Milli-Q water, homogenized with a Pellet Pestle motor (Kontes), and sonicated (3 × 30s) on ice. Supernatant from tissue lysates was collected by centrifugation (12,000×g for 10 min). Protein concentration in the tissue supernatant was measured by Bradford assay. All the samples were aliquoted and stored at ‒80 °C for later use. Quantification and analysis of α-dicarbonyl compounds in these samples were performed using o-phenylenediamine (OPD) dihydrochloride derivatizaton as previously described with some modifications [22–24]. Three α-dicarbonyl compounds, including glyoxal (GO), methylglyoxal (MG), and 3-deoxyglucosone (3-DG), were analyzed by measuring their corresponding quinoxaline derivatives under a reverse phase high-performance liquid chromatography (RP-HPLC) procedure coupled to UV detection [22,25]. A Shimadzu LC system (Kyoto, Japan) equipped with an LC-20AD pump, an LC-20AD/AT low-pressure gradient former, an SIL-10ADvp auto--sampler, a CTO-10AS VP oven, and a DAD detector (SPD-M20A) was used. For quinoxatline formation, 200 μl of tissue supernatant at equal protein concentration was incubated with 25 μl (20 mM) OPD and 1.5 μl (2.5 mM) 5-MQ at 37 °C for 24 h. The reaction was stopped by adding 25 μl (5 M) PCA (Perchloric acid), followed by centrifugation at 12,000×g for 10 min to separate the precipitated protein pellets. 50 μl clear supernatant was injected into an ACE-C18 column (250 mm × 4.6 mm, 5 μm) and the separation was carried out at a flow rate of 0.7ml/min. Elution was performed isocratically with a mixture of 0.1% (v/v) acetic acid in water (HPLC grade) and acetonitrile (ACN) (40:60, v/v). Quantitative analysis relies on comparing the retention times of the peaks in an analyzed sample with those of known standards. Positive identification can be made if the retention time of a peak in the analyzed sample is the same as the standard. Peak areas were used for quantitative chromatographic calculations and same method was used for area calculation for the standard and sample peak. As 5-MQ was used as the internal standard, the peak area ratio between the analyzed sample and the internal standard was used to evaluate the levels of MG, GO and 3-DG in mouse brain tissues at various ages.

2.6. Measurement of enzyme activities associated with respiratory chain complexes

Activity of key enzymes associated with the respiratory chain was measured using homogenates of the cerebral cortex including hippocampus as previously described [26]. Briefly, cerebral cortices were homogenized in Cell Lysis buffer (Cell Signaling Technology). Protein extracts (50 μg) were used for complex I (COX I, NADH:ubiquinone oxidoreductase) and IV (COX IV, cytochrome c oxidase) assays. Complex I activity was determined by monitoring the change in absorbance at 340 nm every 20 s for 6 min. Complex IV activity was determined by cytochrome c reduction rate. Briefly, the protein samples were added to a cuvette containing 0.475 ml of assay buffer (10 mM Tris-HCL, pH 7.0, and 120 mM KCl), and the reaction volume was brought to 0.525 ml with enzyme dilution buffer (10 mM Tris-HCL, pH 7.0 and 250 mM sucrose). The reaction was initiated by the addition of ferrocytochrome c substrate solution (0.22 mM) into the cuvette. The rate of change in absorbance at 550 nm was recorded using a kinetic program with 5 s delay and 10 s intervals for 6 readings using an Amersham Biosciences Ultrospect 3100 pro spectrophotometer.

2.7. Measurement of ATP level

ATP levels were determined using an ATP Bioluminescence Assay Kit (Roche) following the manufacturer’s instructions [26]. Briefly, brain tissues were homogenized in the lysis buffer provided, incubated on ice for 30 min, and centrifuged at 12,000×g for 10 min. ATP levels were then measured in the subsequent supernatants using a Luminescence plate reader (Molecular Devices). A 1.6s delay time after substrate injection and 10s integration time were used.

2.8. Hydrogen peroxide (H₂O₂) assay

Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (Cat#A22188; Invitrogen) was used for measuring the amount of hydrogen peroxide in brain cortices according to the manufacturer’s instructions as described [5]. Briefly, 50 μg of protein from homogenized brain samples was used to determine the H₂O₂ level. The values were reported after subtracting the background, which was determined using a reaction without tissue lysate.

2.9. Statistical analysis

Statistical analysis was performed using StatView statistics software. Differences among means were assessed by one-way analysis of variance (ANOVA) followed by Fisher’s protected least significant difference for post hoc comparisons. P < 0.05 was considered significant. All data were expressed as the mean ± SEM.

3. Results

3.1. AGEs accumulate with aging in mouse brains

Given that both aging and the presence of AGEs are considered to be risk factors in the development of age-related chronic complications, such as diabetes, renal failure, cardiovascular disorder, and neurodegenerative disease, we sought to determine whether aging altered levels of glycated adducts in the cortex and hippocampus of mouse brains. Protein extracted from cortex and hippocampus of mice at 3, 6, 12, 20 and 30 months of age were subjected to immunodot blot for determination of AGEs, including total AGEs, MG-AGE, MG-H1, CML, and CEL, with their respective antibodies, and normalized to β-actin. Significantly elevated levels of total AGEs, MG-AGE, MG-H1, CML, and CEL were observed in mice at 12, 20, and 30 months of age compared with young mice at 3 months of age (Fig. 1A–E). Consistent with these observations, increased levels of AGEs, including total AGEs (Fig. 1F), MG-AGE (Figure S1A) & CML (Figure S1B) were seen in the brains of aging mice by quantitative ELISA. These data demonstrate that increased accumulation of AGEs in mouse brains occurs in an age-dependent manner.

Fig. 1. Determination of AGE adducts in aging mouse brain.

(A–E) Quantification of immunodot intensity of total AGEs (A), MG-AGEs (B), MG-H1 (C), CML (D), and CEL (E) normalized to β-actin were performed in the cortical homogenates from indicated mice. β-actin was used as a protein loading control. (F) Total AGE levels were measured by ELISA in cortical homogenates of indicated mice. Data are presented as mean ± SEM (n = 3). *P < 0.05, ns > 0.05 versus 3-month old mice.

3.2. Aging induces elevation of dicarbonyl compounds in mouse brains

Given that reactive dicarbonyls such as MG, GO and 3-DG are considered the most important glycation agents, leading to the formation of AGEs and causing cellar and mitochondrial perturbation, we determined the tissue levels of cerebral reactive α-dicarbonyl compounds in mouse brain (hippocampus and cerebral cortex) by our modified RP-HPLC assay employing quinoxaline derivatives as external standards and 5-MQ as an internal standard. Two specific peaks were recorded in HPLC chromatograms of standards: one of which is an external standard corresponding to quinoxaline derivatized MG (Fig. 2A), GO (Fig. S2A) or 3-DG (Fig. S3A), respectively, and the other is an internal standard (5-MQ). Representative HPLC chromatograms indicated the presence of MG (Fig. 2B–E), GO (Fig. S2B–E), and 3DG (Fig. S3B–E) in the cerebral cortex and hippocampus of the indicated age groups of mice. All peaks corresponding to MG, GO and 3-DG showed similarity with the specific peaks of external standards (quinoxaline derivatives) and were eluted at the same retention times. Well-defined peaks of quinoxaline derivatives were well resolved from other peaks in the chromatographic analysis. However, extra peaks other than internal standard are characteristic of dicarbonyls and hence these peaks were integrated for quantification of MG, GO and 3-DG. We observed increased levels of quinoxaline derivatives of dicarbonyls in mouse brains with increasing age with the highest levels in cortical tissues of 30-month-old mice. In the cortex, levels of MG (Fig. 2F) were higher than GO (Fig. 2G) and 3-DG (Fig. 2H). Additionally, levels of cortical MG (Fig. 2F) and 3-DG (Fig. 2H) were increased as early as 6 months of age in comparison with GO levels in which increased starting at 12 months of age (Fig. 2G). The degree of all peaks corresponding to MG, GO and 3-DG in 3-month-old mice was negligible (not shown). Thus, accumulation of these deleterious dicarbonyls starts at a certain age and exerts chronic effects in aging by inducing toxicity and oxidative stress, contributing to age-related dementia and cognitive decline.

Fig. 2. Determination of reactive dicarbonyls in aging mouse brain.

Representative HPLC chromatogram showed external and internal standard of MG (A), and cortical and hippocampal MG (B–E) in the indicated mice. Quantification of levels of MG (F), GO (G), and 3-DG (H) were performed in cortex and hippocampus of indicated mice. Data are presented as mean ± SEM (n = 3), *P < 0.01 versus young mice samples as a control group.

3.3. Accumulation of AGEs and dicarbonyls affects ROS generation and mitochondrial function in the aging brain

To determine whether aging-induced accumulation of diccarbonyls and AGEs disturbs mitochondrial function, we assessed mitochondrial respiratory function and energy metabolism by measuring activities of mitochondrial respiratory chain complexes I (COX I) and IV (COX IV) and ATP levels in aging mice. Compared to 3-month-old young mice, activity of COX I and COX IV was gradually reduced in cortex and hippocampus by 30–50% from 12 to 30 month-old mice (Fig. 3A–B). In parallel, cortical mitochondria revealed significant reduction in ATP levels with advancing age compared to young mice (Fig. 3C), suggesting age-related impairment in mitochondrial respiratory function and energy metabolism. Mitochondria are a main source for production of ROS, which accelerate the formation of AGE-modified proteins, including glycated tau and Aβ, and exaggerate AGEs-induced oxidative stress and toxicity [6,14]. We therefore determined whether increased age elevates ROS and oxidative stress. As shown in Fig. 3D, hydrogen peroxide (H₂O₂) was significantly elevated in the cortex of mice with increasing age. These results indicate that age-related excessive accumulation of AGE and dicarbonyls generation correlates to the overproduction and accumulation of ROS and mitochondrial dysfunction.

Fig. 3. Mitochondrial function in aging mouse brain.

(A–C) Activity of complex I (A), complex IV (B) and ATP levels (C) were determined in cortex and hippocampus of indicated mice. (D) Levels of H₂O₂ were measured in cortex of indicated mice. *p < 0.05 compared to other groups of mice. Data are expressed as fold-increase relative to young control mice. N = 3–5 mice per group.

To extend these results to human samples, we evaluated levels of AGE-adducts and dicarbonyls in aged human brains and their effects on mitochondrial function and oxidative stress. Similarly, AGEs, including total AGE, MG-AGE, MG-H1, CML and CEL, were significantly elevated in the temporal cortex of normal aged human brains compared to young brains by quantification of immunodot blotting (Fig. 4A–E). Further confirmation of significantly elevated levels of these AGEs was achieved by quantitative ELISA in aged human brains, including total AGEs (Fig. 4F), MG-AGE (Fig. S4A) and CML (Fig. S4B). Representative HPLC chromatograms showed sharp elution peaks of MG (Fig. 5A & B), GO (Fig. S5A & B) or 3-DG (Fig. S5C & D) with internal standard in young and aged human brains. Levels of MG (Fig. 5C), GO (Fig. 5D) and 3-DG (Fig. 5E) quantified by HPLC were also robustly enhanced in aged human brains. These results demonstrate that accumulation of dicarbonyls derived AGE-adducts occurs in the aged brain. Accordingly, the activity of mitochondrial key enzymes (complex I and IV) and ATP production were significantly reduced (Fig. 6A–C), while H₂O₂ levels were greatly elevated (Fig. 6D) in aged human brains compared to young brains. These results demonstrate a link between progressively increased dicarbonyls-mediated formation/accumulation of AGEs and sustained mitochondrial oxidative stress in the aging brain.

Fig. 4. Determination of AGE adducts in young and aged human brains.

(A–E) Quantification of immunodot intensity of total AGEs (A), MG-AGEs (B), MG-H1 (C), CML (D) and CEL (E) normalized to β-actin were performed in young (age: 33.75 ± 2.69) and old brain (age: 81.4 ± 3.06). β-actin was used as a protein loading control. (F) Total AGE levels were measured by ELISA in young and old human brains Data are presented as mean ± SEM (n = 3–5). *P < 0.01 vs young brain.

Fig. 5. Determination of reactive dicarbonyls in young and aged human brains.

Representative HPLC chromatogram showed peaks for MG in young (A) and aged cortex (B). Quantification of MG (C), GO (D), and 3-DG (E) were performed in young and old cortical homogenates. Data are presented as mean ± SEM (n = 3–8). *P < 0.05 versus young brains.

Fig. 6. Assessment of mitochondrial function and ROS in young and aged human brains.

(A–D) Activity of complex I (A), complex IV (B), levels of ATP (C) and H₂O₂ (D) were measured in young and aged brain. Data presented as mean ± SEM (n = 4–8). *P < 0.05 or ^#P < 0.01 versus young brain as a control group.

3.4. Effect of mitochondrial ROS on cerebral AGEs and mitochondrial function in vivo in aged mice

To determine whether mitochondrial ROS are involved in accumulation of AGEs, we examined the effect of scavenging mitochondria-derived ROS by administration of mitoTEMPO (mTEMPO), a scavenger of mitochondrial ROS, to aged mice on AGEs accumulation and mitochondrial function. It is clear that mice treated with mTEMPO exhibit not only blunted ROS production (Fig. 7A), but also largely reduced cellular AGE-adducts (Fig. 7B–F & Fig. S6A–C). Moreover, we demonstrated that mTEMPO treatment increased ATP levels (Fig. 7G) and activity of mitochondrial respiratory chain complex (I & IV) (Fig. 7H and I) in the cerebral cortex. These results suggest that scavenging mitochondrial ROS significantly eliminates the formation and accumulation of AGEs and improves mitochondrial function.

Fig. 7. Effect of mitochondrial ROS on cerebral AGEs and mitochondrial function in mouse brain in vivo.

Mice at 21–22 months of age were daily intraperitoneal injection of mTEMPO or vehicle for three months, and then cortical homogenates were prepared and subjected to the indicated analyses. H₂O₂ (A) levels were assayed in cerebral cortex of vehicle- or mTEMPO-treated mice. Quantification of immunodot intensity of total AGEs (B), MG-AGEs (C), MG-H1 (D), CEL (E) and CML (F) normalized to β-actin were performed in vehicle- or mTEMPO-treated mice. β-actin was used as a protein loading control. Cortical ATP (G), complex I (H), complex IV levels (I) were determined in vehicle- or mTEMPO-treated mice. Data presented as mean ± SEM (n = 4–5). *P < 0.05 versus vehicle group.

3.5. Blocking mitochondrial ROS attenuates AGEs- or MG-induced mitochondrial dysfunction and oxidative stress in vitro in brain slices

To further determine a direct link of mitochondrial ROS to AGEs- or MG-induced mitochondrial stress, we conducted in vitro brain slice experiments. Due to the absence of a blood supply to the tissue in this experiment, vehicle, mTEMPO, AGEs or MG was delivered to the brain tissue by passive diffusion via perfusion solution [artificial cerebrospinal fluid (ACSF)] continuously bubbled with 95% O₂ and 5% CO₂. The slices treated with either AGEs or MG exhibited significant decreases in activity of mitochondrial respiratory chain complex I and IV (Fig. 8A and B, Fig. 8E and F) and ATP levels (Fig. 8C and G), and increase in H₂O₂ levels (Fig. 8D and H). Treatment with mTEMPO reversed these detrimental effects (Fig. 8A–H) as shown by increased complex I and IV activity and ATP levels and reduced H₂O₂ levels. Thus, suppression of mitochondrial ROS protects against AGEs- or MG-mediated aberrant mitochondrial function and oxidative stress.

Fig. 8. Effect of mTEMPO on AGEs- or MG-induced ROS generation and mitochondrial dysfunction in vitro WT mouse brain slices.

Complex I (A, E), complex IV (B, F), ATP (C, G) & H₂O₂ (D, H) levels were determined in the perfused WT brain slices with treatment of AGEs (100 μg/ml) or MG (1 μM) in the presence or absence of mTEMPO (2 μM). Vehicle treated WT mouse brain slices were used as a control group. Data presented as mean ± SEM (n = 3–4), *P < 0.05, versus vehicle or mTEMPO treated group.

4. Discussion

Mounting evidence links dicarbonyl stress and subsequent glycation to the aging process in human brains [27–29]. AGEs are formed endogenously as a part of normal metabolism and their accumulation is part of the normal aging process [27], however, excessive accumulation of AGEs accelerates the aging process and perturbs mitochondrial and synaptic function [7]. This is illustrated by the present study in both mouse and human brains showing the detrimental effects of AGEs on mitochondrial respiratory function, energy metabolism, and increased mitochondrial oxidative stress in AGEs-insulted brain slices and the AGEs-rich aged brain. AGEs also induce loss of synapses and impair learning and memory [5,7]. Furthermore, AGEs are highly accumulated in brains both with advanced age and/or with disease progression, including AD. High levels of AGEs have been shown in cortical neurons of older adults, in cerebrospinal fluid [17], neurofibrillary tangles, and senile plaques from patients with AD [20,30], which are positively correlated with severity of cognitive impairment [28,31]. In line with these observations, the present study demonstrates that levels of AGE adducts, including MG-AGE, MG-H1, CML, and CEL, were significantly increased in mouse brains in an age-dependent manner and reached the highest levels at 30 months of age. These AGE-modified products are also highly prevalent in aged human brains. Consistent with our results, other group’s report shows increased AGE-modified proteins using an anti-CML antibody in the brains of 22-month old mice compared with 2-month-old mice [32]. Since AGE formation and accumulation are enhanced in aging-induced diseases, including AD, age-related accumulation of AGEs could serve as danger signals to initiate/promote mitochondrial and synaptic stress leading to neurodegeneration, cognitive impairment, memory loss, or AD pathology [13,33].

The reactive dicarbonyls, including MG, GO and 3-DG, are major precursors of various AGEs. Among these dicarbonyl compounds, MG is one of the best-known glycation agents linked to age-related cell injury and cognitive impairment via its direct cytotoxicity and/or initiating the formation of MG-derived AGEs [5,31,34–36]. The MG-H1 adduct is a marker of MG-induced glycation of arginine residues while CML and CEL are markers of lysine amino acid protein modification [37]. There is increasing evidence from pathological and epidemiological studies that dicarbonyls-induced oxidative stress and formation of AGEs are related to the risk of adverse aging-related outcomes and with poor survival [38]. In the present study, we demonstrated increased levels of cerebral dicarbonyls, including MG, GO and 3-DG, in mouse brains in an age-dependent manner. The highest levels of these dicarbonyl compounds were evidenced at 30 months of age. Similarly, MG, GO and 3-DG are highly prevalent in aged human brains. Of these dicarbonyls, MG levels were higher than GO and 3-DG in aging brains. Given that high MG levels are associated with a faster rate of cognitive decline in human [39], MG-derived AGE products might be responsible, at least in large part, for potentiating brain dysfunction and aging-related complications.

It is known that dicarbonlys and AGEs exaggerate oxidative stress-induced neuronal perturbation via interaction with the AGE receptor or binding protein [5,6,13,14]. Indeed, age-related progressive accumulation of AGEs associates with elevation of reactive oxygen species including H₂O₂ levels. The concentration of H₂O₂ affects mitochondrial biogenesis and enzyme activity [40]. Modest induction of H₂O₂ has been shown to promote mitochondrial biogenesis [41], while high levels of H₂O₂ induces apoptosis [42], decreases mitochondrial enzyme activity, and leads to mitochondrial dysfunction. Moreover, mitochondrial respiratory chain complexes are proton pumps that generate the transmembrane proton gradient necessary to drive ATP generation by ATP synthase. Any changes in the electron transport chain (ETC) system would impact mitochondrial ATP generation and any ensuing mitochondrial processes. The significant decrease in enzyme activities of complexes I and IV and ATP levels in mouse brains with increasing age suggest a deterioration in the ETC system of AGE accumulated tissues. Here we provided the evidence of sustained mitochondrial stress with significant impairment in mitochondrial respiratory function and reduced ATP levels in both muse and human aging brains.

Excess ROS production has been also linked to activation of the stress kinases. It has been shown that an increase in AGEs accumulation activates stress kinase signaling [43]. We believe that the enhanced ROS production by high levels of AGE accumulation in aged mouse and human brains contributes to disruption of mitochondrial respiratory chain activity and impairment in energy metabolism. Given that mitochondria-derived ATP is important for maintaining normal synaptic function [44], AGE-mediated ROS overproduction and the decline in synaptic mitochondrial ATP could contribute to mitochondrial dysfunction or synaptic loss. In parallel, RAGE is a receptor for AGEs [45]. RAGE expression is elevated by AGEs and Aβ relevant to aging, diabetes, AD-affected brain and AD transgenic mice overexpressing Aβ [32,46–49]. Overexpression of RAGE in neurons or microglia accelerates Aβ accumulation, neuroinflammation, and deficits in learning and memory [50,51]. Interaction of AGEs with RAGE activates signal transduction, such as p38, ERK1/2 mitogen-activated protein kinases, and nuclear transcription factor κB (NFκB) [6,52,53], resulting in metabolic changes and cellular perturbation during aging and age-related neurodegenerative disease [6,7,13,14]. Loss RAGE or defective RAGE signaling protects against toxic metabolites (AGEs or Aβ)-mediated oxidative stress, synaptic dysfunction, cerebral Aβ accumulation, and improves cognitive function [7,50,51,54]. Therefore, RAGE may be important for AGEs-induced mitochondrial stress, impairment in respiratory function and energy metabolism. Blockade of RAGE may have protective effect on AGEs-induced mitochondrial perturbation and cognitive dysfunction resulting from progressive AGE accumulation on brain aging.

Although cells are equipped with an impressive antioxidant enzyme defense system against toxic metabolites, which are generated as byproducts of processes that are essential to cellular survival, this defense system is insufficient to prevent oxidative damage to tissues and cells under certain pathophysiological conditions. Mitochondrial targeted antioxidants are known as a useful tool for examining the role of mitochondrial ROS emission on cellular abnormalities of several chronic disease [55–58]. We hypothesized that elevated levels of dicarbonyls and AGEs-induced oxidative stress might consume much of the supply of constitutive antioxidant enzymes, in turn, the reduced antioxidant defenses directly alter synaptic activity and neurotransmission in neurons, leading to cognitive dysfunction. Therefore, we exogenously administrated mTEMPO, a mitochondrial-targeted antioxidant to aged mice to investigate whether scavenging mitochondrial ROS influences AGE accumulation and mitochondrial function. Intriguingly, mTEMPO treatment not only suppressed oxidative stress and restored mitochondrial function but also significantly reduced age-related accumulation of cerebral AGEs. Accordingly, suppressing mitochondrial ROS by mTEMPO significantly attenuated mitochondrial dysfunction insulted by MG or AGEs in vitro brain slices. Thus, mitochondrial ROS contributes importantly to the formation and accumulation of AGEs and AGEs-induced alterations in mitochondrial function. Application of mitochondrial-targeted antioxidant may have benefit for clearance of accumulated dicarbonyls and AGEs and restoration of mitochondrial function.

In conclusion, we demonstrate that elevated formation and accumulation of dicarbonyls and AGEs trigger overproduction of ROS in aging brains, which links to impaired mitochondrial respiration and energy metabolism. Eliminating mitochondrial ROS greatly prevents the progressive accumulation of cerebral AGEs and dicarbonyls. Thus, AGE-mediated signal transduction via ROS/dicarbonyls could enhance neuronal perturbation and AD pathology through mitochondrial dysfunction, synaptic injury, and cognitive decline. Furthermore, activation of antioxidant enzymes or application of mitochondrial-targeted antioxidant may present a new therapeutic approach for halting AD progression at the early stage through mitochondrial quality control combined with eliminating age-related accumulation of toxic metabolites such as AGEs and dicarbonyls.