Alzheimer’s disease (AD) is a life-threatening neurodegenerative disorder. The primary clinical manifestations of AD are cognitive impairment, intellectual decline, and personality changes [1]. With the aging of human society, the number of patients with AD continues to increase. AD is caused by genetic, environmental, and lifestyle factors that affect neuronal cell degradation over time and severely affect quality of life in older people. Intensive research on the pathogenesis of AD is ongoing [2].
A major cause of AD is oxidative stress, which leads to neuronal injury and death. Oxidative stress is caused by an imbalance in the production and elimination of oxygen free radicals, thus resulting in the accumulation of oxygen active substances in the body and a subsequent stress response [3]. High levels of reactive oxygen species (ROS) can cause oxidation of DNA, proteins, lipids, and other biological macromolecules, thus affecting the function and structure of biological molecules. ROS can also alter signal transduction and result in abnormal cell function and apoptosis. Therefore, inhibiting oxidative stress may be a feasible approach in the treatment of AD [4].
The history of ginseng (Panax ginseng C. A. Mey.) is described in "Shen Nong’s Materia Medica." Ginseng exerts effects in traditional Chinese medicine such as increasing vitality, supporting the spleen and lungs, delaying aging, and rejuvenating the nerves; its individual components and metabolic components also show biological activity [5]. Ginsenoside compound K [20-O-β-D-glucopyranosyl-20(S)-protopanaxadiaol (CK)], one of the primary active metabolites of protopanaxadiol-type ginsenosides, is produced by the intestinal flora from ginsengdiol saponins; moreover, CK has biological properties, such as antiaging, neuroprotective, antioxidative, and anti-inflammatory effects [6]. In our previous study [7], we confirmed that CK has neuroprotective properties; however, no analyses of the antioxidative and neuroprotective effects of CK have used 1H nuclear magnetic resonance (1H-NMR)-based metabolomics.
Recently, metabolomic techniques, an important part of systems biology approaches, have been widely used to discover novel biomarkers [8, 9]. 1H-NMR-based metabolomics has been applied to examine overall metabolic changes in organisms, thus enabling evaluation of the efficacy of traditional Chinese medicine formulations, including their pharmacological and toxicological mechanisms [10]. Cell metabolomics research has rapidly and intuitively described the metabolism of specific cell types with little interference, thus revealing the overall performance of physiological functions at the cellular level [11].
Neuronal damage caused by oxidative stress can lead to neurodegenerative diseases such as AD. Excessive production of ROS triggers intracellular oxidative stress. Hydrogen peroxide (H2O2) is a reactive oxygen species and an important inducer of oxidative stress, which is used in the study of neurodegenerative diseases, such as AD, that are caused by oxidative stress [12]. We treated HT22 cells with H2O2 to establish an intracellular model of AD. In MTT assays, 8 μM CK significantly increased the survival rate of HT22 cells ( Figure 1 ). Thus, 8 μM CK was used for metabolomics analysis.
Effects of CK treatment on cell growth.
Viability of cells treated with different concentrations of CK. Model: 400 μM H2O2, CK1: 4 μM CK + 400 μM H2O2, CK2: 6 μM CK + 400 μM H2O2, and CK3: 8 μM CK + 400 μM H2O2. b P < 0.01 vs. control group; c P < 0.05 and d P < 0.01 vs. model group. Values represent mean ± SD (n = 8).
We then investigated the potential protective effects of CK against H2O2-induced changes in endogenous metabolites of HT22 cells with 1H-NMR-based metabolomics. Differentially present metabolites in each treatment group were screened and identified, and the primary metabolic pathways were analyzed.
Analysis of the 1H-NMR data combined with a comparison of previous studies enabled assignment of signals of the main metabolite components in the three groups of cells ( Figure 2, Table S1 ). The peak values of the 1H-NMR spectra were normalized in MestReNova software, and the normalized peak data were imported into SIMCA-P software for partial least squares-discriminant analysis (PLS-DA) analyses.
Typical 1H NMR spectra of HT22 cells.
(a) Control group and model group. (b) Model group and CK group.
The PLS-DA method removes interfering factors that are not associated with sample classification and maximizes the differences between groups. The results revealed differences in cell metabolite composition between the control and model groups ( Figure 3a ). The values of parameters R2Y and Q2Y were 0.815 and 0.986, respectively, thus indicating that the established PLS-DA model had good fit and predictability ( Figure 3b ). The CK group cluster was distinct from the model group cluster ( Figure 3c ), thereby indicating that the CK group differed from the model group.
(a) PLS-DA score plot of the control and model groups. (b) Permutation test of the control and model groups. (c) PLS-DA score plot of the CK and model groups. (d) Loading plot. (e) PLS-DA score plot of the control, model, and CK groups.
The loading plot ( Figure 3d ) showed that the metabolites farther away from the origin could be considered candidate biomarkers. The metabolites with variable importance (VIP) values (VIP1 and VIP2) > 1 and P < 0.05 were highlighted as potential biomarkers. A total of 20 metabolites were considered as potential biomarkers that inhibit oxidative stress injury. Detailed information on these biomarkers and trends in changes in these biomarkers between groups are shown in Table 1 .
Potential biomarkers of oxidative stress.
| No. | ppm | MW | Metabolite | KEGG ID | Model/control | CK/model |
|---|---|---|---|---|---|---|
| 1 | 3.238 | 146.20 | Acetylcholine | C01996 | ↑b | ↓d |
| 2 | 3.278 | 125.15 | Taurine | C00245 | ↑b | ↓c |
| 3 | 1.358 | 90.08 | Lactate | C00186 | ↑b | ↓c |
| 4 | 3.598 | 75.07 | Glycine | C00037 | ↓ | ↑d |
| 5 | 3.838 | 89.09 | Alanine | C00041 | ↑b | ↓c |
| 6 | 2.438 | 118.088 | Succinic acid | C00042 | ↑b | ↑ |
| 7 | 3.718 | 193.199 | Phenylacetylglycine | C05598 | ↑b | ↓c |
| 8 | 1.958 | 60.05 | Acetic acid | C00033 | ↓b | ↑d |
| 9 | 4.358 | 244.2 | Uridine | C00299 | ↑a | ↓ |
| 10 | 3.758 | 147.13 | Glutamate | C00025 | ↓a | ↑ |
| 11 | 0.958 | 131.18 | Leucine | C00123 | ↑a | ↓ |
| 12 | 2.678 | 149.21 | Methionine | C00073 | ↑a | ↓ |
| 13 | 3.958 | 131.133 | Creatine | C00300 | ↑a | ↓c |
| 14 | 3.958 | 219.23 | Pantothenic acid | C00864 | ↑a | ↓c |
| 15 | 3.198 | 162.20 | O-Acetylcholine | C01996 | ↓a | ↑d |
| 16 | 8.478 | 45.01 | Formate | C00058 | ↑a | ↓ |
| 17 | 0.998 | 117.15 | Valine | C00183 | ↑a | ↓ |
| 18 | 2.038 | 175.14 | N-Acetylaspartate | C01042 | ↓a | ↑c |
| 19 | 3.078 | 113.1 | Creatinine | C00791 | ↓a | ↓ |
| 20 | 3.558 | 180.16 | Inositol | C00137 | ↓a | ↑d |
Notes: a P < 0.05, b P < 0.01, model group vs. control group; c P < 0.05, d P < 0.01, CK group vs. model group.
These biomarkers were imported into the MetaboAnalyst 3.0 online analysis tool to perform enrichment and pathway analyses (metabolic pathways with an influence value > 0.1). The results suggested that CK regulates taurine and hypotaurine metabolism, pyruvate metabolism, alanine aspartate and glutamate metabolism, and glycine serine and threonine metabolism ( Figure 4a and b ), thus decreasing oxidative stress damage.
Altered metabolic pathways in H2O2-damaged cells, visualized as (a) bubble plots and (b) an enrichment overview.
a: Alanine, aspartate, and glutamate metabolism; b: taurine and hypotaurine metabolism; c: glutathione metabolism; d: pyruvate metabolism; e: glycine, serine, and threonine metabolism; f: pantothenate and CoA biosynthesis; g: methane metabolism; h: glyoxylate and dicarboxylate metabolism; i: inositol phosphate metabolism; and j: D-glutamine and D-glutamate metabolism.
Taurine, a well-known antioxidant, has potential antiapoptotic properties and exerts important regulatory effects on brain function; it has neuroprotective effects and improves cognitive function [13]. Taurine represents a key protein transformation node involved in amino acid and energy metabolism: it limits peroxidation, regulates glutamate metabolism, inhibits nerve cell apoptosis, and protects nerve cell function. Increasing evidence suggests that taurine-induced neuroprotection is mediated by the antagonism of glutamate-induced excitotoxicity [14]. Taurine content was much higher in the model groups than the control group, possibly as a compensatory response to antioxidant damage [15, 16]. Taurine content decreased after treatment with CK, thus indicating that CK decreased oxidative stress and eliminated the compensatory response.
Glycine is an amino acid with a simple structure that counteracts the production of ROS and attenuates H2O2-induced oxidative damage in neurons [17]. CK is also involved in glutathione metabolism, and glycine is the main precursor of glutathione synthesis [18]. Glutathione is an endogenous antioxidant that plays a key role in the defense against oxidative stress in the brain and counteracts neurotoxicity due to peroxides such as H2O2 [19]. Impaired glutathione metabolism may lead to neurodegenerative diseases such as AD. Therefore, maintaining glutathione metabolism homeostasis plays a crucial role in neuronal protection [20]. Our results showed that CK is involved in glycine metabolism and increased glycine content, and thus plays a role in regulating glutathione metabolism and H2O2-induced oxidative decreasing damage in HT22 cells.
Glutamate, another essential precursor for glutathione synthesis, participates in glutathione metabolism [21] and is also an important neurotransmitter in the brain. Intraneuronal homeostasis of glutamate is essential for neuronal energy metabolism and amino acid metabolism; glutamate affects synaptic transmission and brain function. Several studies have shown diminished glutamate levels in the brains of patients with AD [22], in agreement with the results of the present study suggesting that CK may enhance ATP ( Figure 5 ) levels in HT22 cells by regulating glutamate metabolism.
Effects of CK on intracellular ATP levels of HT22 cells.
Model: 400 μM H2O2, CK1: 4 μM CK + 400 μM H2O2, CK2: 6 μM CK + 400 μM H2O2, and CK3: 8 μM CK + 400 μM H2O2. b P < 0.01 vs. control group; d P < 0.01 vs. model group. Values represent mean ± SD (n = 8).
Some studies have indicated that phosphoinositide 3-kinase (PI3K)/protein kinase B (PKA, also known as AKT) signaling is an important therapeutic target for the treatment of AD. Moreover, PI3K/AKT signaling is involved in the regulation of neuronal oxidative stress and energy metabolism in AD [23]. Therefore, we next detected the expression levels of components of the PI3K/AKT signaling pathway through western blotting. PI3K is a lipid kinase that generates phosphatidylinositol-3,4,5-trisphosphate, which in turn promotes the translocation of AKT to the plasma membrane. PI3K/AKT signaling influences neuronal plasticity, cell survival, proliferation, and apoptosis inhibition [24]. AKT, an important upstream regulator of GSK-3β, increases GSK-3β phosphorylation, thus inactivating GSK-3β. The antagonistic effects of GSK-3β activity influence central-nervous-system axon regeneration. Of note, inhibitors of GSK-3β have been postulated to exert neuroprotective effects [25]. The present results indicated that CK increased the expression of components of the PI3K/AKT signaling pathway and improved regulation of the expression of proteins involved in energy-metabolism pathways ( Figure 6 ).
Western blotting analysis of the expression of proteins associated with energy metabolism.
Model: 400 μM H2O2, CK1: 4 μM CK + 400 μM H2O2, CK2: 6 μM CK + 400 μM H2O2, and CK3: 8 μM CK + 400 μM H2O2 (a) and comparison of expression levels (b). b P < 0.01 vs. control group; c P < 0.05, d P < 0.01 vs. model group. Values represent mean ± SD (n = 3).
We explored the effects of CK on endogenous meta bolites in oxidatively damaged neurons in vitro. CK also has neuroprotective effects in vivo [26]. Whether CK can cross the blood–brain barrier is unknown. Karpagam et al. have shown that CK can cross the blood–brain barrier, according to ADMET assays [27], but further investigation is needed to determine whether this crossing also occurs in vivo.
