INTRODUCTION
Chronic fatigue syndrome (CFS) is a complicated disorder characterized by extreme fatigue that cannot be explained by any underlying medical condition. Most of the symptoms, such as loss of memory or concentration, unexplained muscle or joint pain, headaches, unrefreshing sleep, and extreme exhaustion, are neural and psychiatric,1) suggesting an involvement of disorders in neuronal–endocrine–immune interactions.2)
Although the term CFS first appeared over 30 years ago, the diagnosis of this illness is still based on symptoms, and the etiology is still unclear. An agreement on the etiology and pathophysiology of CFS has not been reached yet.3,4) A multifactorial etiology is accepted by most researchers.5–7) Disturbance in immunity and the involvement of oxidative and nitrosative stress (O&NS) are reported by various researchers to be associated with the onset of CFS.8,9) When the peripheral immune system is activated it produces a series of pro-inflammatory cytokines, such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α.10,11) This pro-inflammatory signal triggers the production of the same cytokines by glial cells in the brain.12,13) After being stimulated by the presence of cytokines, microglia also produces and releases other pro-inflammatory molecules and nitric oxide (NO).14) Elevated NO leads to increased superoxide and peroxynitrite. This in turn leads to oxidative stress and further induce the production of proinflammatory cytokines.15,16) All these factors may finally contribute to neurodegeneration and reduced neurogenesis, which are common pathological changes in many diseases.17–19)
Kampo medicines Hochuekkito (TJ-41) and Hachimijiogan (TJ-7) are believed to relieve the symptoms in individuals with poor physical strength who tend to get tired easily.20,21) TJ-41 was reported to be effective in improving human immunity22,23) and reducing the expression of inflammatory cytokines24,25) and NO.26) TJ-7 could be beneficial in relieving fatigue,27) decreasing oxidative stress,28,29) and improving cognitive function.30)
During the past 10 years, we have developed a CFS animal model using six injections of heat-killed Brucella abortus (BA) antigen, which could extend the duration of fatigue that mimics the real chronic course of human CFS.31) Running wheel activity is suppressed for at least 6 weeks after the last injection, which helps us to test treatments. The disadvantage is that a large number of mice (about 1/3) die during the six injections.31)
The aim of this research is to: 1) improve the animal model to reduce the death rate; 2) confirm neuroinflammation, oxidative stress, and neurogenesis in the brain of CFS animals; and 3) study the effects of Kampo medicine.
MATERIALS AND METHODS
Animals and Living ConditionsFemale BALB/c mice (20–24 g, 8 weeks old) were procured from CLEA Japan (Tokyo, Japan). The animals were housed 5 per cage on 12 h light/dark cycle with food and water available ad libitum. Temperature was maintained 22–23°C. Running wheel cages included a running wheel (23 cm in diameter), counters, cribs, and water taps (a detailed description of this cage is provided in our prior research).31) The mice were free to stay in the rest area or enter the running wheel. Our research protocol was approved by the Animal Experimental Committee of Kanazawa Medical University.
Induction of the CFS ModelHeat-killed BA ring test antigen was obtained from the National Veterinary Services Laboratories in the United States. In order to obtain an appropriate concentration of BA, the stock solution was centrifuged and resuspended in saline. The CFS animal model was induced by six repeated injections of BA antigen solution (0.2 mL per mouse) intraperitoneally every 2 weeks. The first dose for the CFS group was decreased to 0.6 × 1010 particles to avoid a large number of mice dying, and the following five injections were 2.5 × 1010. The control group was injected with 0.2 mL saline six times. Mice were placed on the running wheel cages the day before and one day after each injection for 24 h (in and out at 13:00). Then they were put back in cages where they lived together.
5-Bromodeoxyuridine (BrdU) was given to mice orally; 0.8 mg/mL solution was made, and 4% sucrose was added in order to neutralize the unpleasant taste of BrdU. Drinking water was changed to BrdU solution at the 15th week for 7 d. Mice were killed 6 weeks after the last injection (at the end of the 16th week).
Kampo Medicine InterventionTJ-41 and TJ-7 were obtained from Tsumura & Co. (Tokyo, Japan). TJ-41 contained a mixture of spray-dried hot water extracts of 10 medicinal plants: Astragali radix (16.7%), Atractyloclis lanceae rhizoma (16.7%), Ginseng radix (16.7%), Angelicase radix (12.5%), Bupleuri radix (8.3%), Zizyphi fructus (8.3%), Aurantii nobilis pericarpium (8.3%), Glycyrrhizae radix (6.3%), Cimicifugae rhizoma (4.2%) and Zingiberis rhizome (2.0%). TJ-7 is an extract from a mixture of Rehmanniae radix, Corni fructus, Dioscoreae rhizome, Alismatis rhizome, Hoelen, Moutan cortex, Cinnamomi cortex, and Aconiti tuber (27.3, 13.6, 13.6, 13.6, 13.6, 11.4, 4.5, and 2.3%). Detailed chemical profiling of TJ-41 and TJ-7 could be found in references.32,33) A 0.5% concentration of TJ-41 and 0.5% TJ-7 were added into rodent diet CE-2 (CLEA Japan). The daily dosage of each drug was about 800 mg/kg body weight (daily consumption of diet was assumed as 4 g and body weight as 25 g). For a human, the common dosage of TJ-41 is 100 mg/kg, and that of TJ-7 is 80 mg/kg body weight (with human body weight assumed as 50 kg). Intervention with TJ-41 and TJ-7 started from the 12th week and lasted for 4 weeks until the end of the research.
Brain Fixation and HistologyMice were anesthetized with an overdose of chloral hydrate (Nacalai, Japan). A cannula was inserted into the left ventricle. Mice were perfused with saline (about 20 mL) and about 50 mL of 4% paraformaldehyde/phosphate-buffered saline (PBS) solution. Both the saline and the paraformaldehyde solution were ice-cold. Then the brain was taken out, post-fixed in 4% paraformaldehyde/PBS overnight, and immersed in 30% sucrose/PBS overnight. Along the sagittal suture, the brain was separated and put in a deep freezer until use. Coronal sections were cut through the entire dentate gyrus from interaural 2.46 to 0.00 mm by a cryostat microtome. Thickness was set at 35 µm, and about 70 sections could be obtained from each sample. Sections were placed in a bath of antifreeze (2 : 3 : 5 ratio of glycerol : ethylene glycol : PBS) in a 96-well plate and stored at −20°C.
ImmunohistochemistryFor Ki67 and 4-hydroxynonenal (4-HNE) staining, sections were mounted on the fine frost slides (Matsunami, Japan) and dried out. Antigen retrieval and 3% H2O2 processing were done before incubation in rabbit anti-Ki67 antibody (1 : 4,000, Vector Laboratories, VP-K451) and mouse anti-4-HNE antibody (1 : 5, JaICA, MHN-100P). Then secondary antibody (1 : 200, Vector Laboratories, BA-1000; 1 : 50 mouse immunoglobulin G kappa binding protein conjugated to horseradish peroxidase (m-IgGκ BP-HRP), Santa Cruz Biotechnology, sc-516102) incubation and a standard strept avidin-biotin complex (SABC, Nacalai, Japan, 30462, only for Ki-67) and diaminobenzidine (DAB, Nacalai, 25985) procedure were done. For doublecortin (DCX) staining, free-floating sections were incubated with goat anti-DCX antibody (1 : 400, Santa Cruz Biotechnology, SC-8066), followed by the secondary antibody (Vector Laboratories, BA-9500) incubation and the same SABC/DAB kit procedure.
For BrdU/neuron-specific nuclear protein（NeuN） double fluorescent staining, the sections were incubated in 2 M HCl for 20 min at 37°C and neutralized in boric acid (pH 8.5) before incubation with BrdU (1 : 400, Accurate Chemical and Scientific, OBT0030G) and NeuN antibodies (1: 400, Millipore, MAB377). Then sections were incubated with the fluorescent secondary antibodies Alexa Fluor 488 and Alexa Fluor 594 (1 : 200, Thermos-Fisher, A-11001 and A-11007). Images of sections were captured on a Zeiss Axiovert 200M microscope.
Eight sections (every eighth section from among the 70 sections, representing the whole hippocampal area) were taken out from a 96-well plate to represent the entire rostrocaudal extent of the dentate gyrus. All the labeled cells (Ki-67, DCX and NeuN/BrdU) in the granule cell layer were counted on the microscope. Then cell number was multiplied by 2 (left and right sides of brain) and 8 (every eighth section selected) to get the total quantity for each mouse.
Western BlotAfter mice were anesthetized with an overdose of diethyl ether (Nacalai), they were decapitated, and their brains were harvested. Then hippocampus and cortex were separated and stored at −80°C until use. Tissues were homogenized in a lysis buffer (50 mM Tris–HCL, 150 mM NaCl, 1% Triton X-100) with protease inhibitor cocktail, 1 µg/mL (Nacalai, 08714-04), and agitated for 40 min at 4°C. Tissue homogenates were then centrifuged at 12000 rpm for 20 min, and the supernatant was collected. Total protein concentration was determined by Bradford assay. A 6× sample buffer with reducing agent (Nacalai, 09499-14) was added into the tissue homogenates and heated to 95°C for 7 min.
Samples (total protein concentration of 4–20 µg) were loaded into each well, and then sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) proteins were transblotted to a Polyvinylidene fluoride (PVDF) membrane. The membrane was incubated with anti-β-actin (Cloud-Clone Corp., CAB340Hu22, 1 : 10000), Anti-IL-1β (Abcam, ab9722, 1 : 3000), Anti-IL-6 (Abcam, ab7737, 1 : 1000), Anti-Interferon gamma (IFN-γ, Abcam, ab133566, 1 : 3000), Anti-Nitrotyrosine (3-NT, Santa Cruz, sc-32757, 1 : 50), and Anti-4-HNE (R&D, MAB3249, 1 : 100) at 4°C overnight. This was followed by secondary antibody incubation (m-IgGκ BP-HRP, Santa Cruz Biotechnology, sc-516102, 1 : 4000; Goat anti-Rabbit IgG HRP, Abcam, ab6721, 1 : 3000) at room temperature for 1 h. Full film scans of the Western blot data were obtained with Fusion FX7 (Vilber Lourmat, France). The expressions of proteins were quantified by measuring band intensities using ImageJ software (NIH, Bethesda, MD, U.S.A.). The band intensity values of the protein of interest were adjusted to that of actin and standardized to the control group (but not for 3-NT and 4-HNE, since no bands were detected in the control group).
Statistical AnalysisAll data were expressed as the mean ± standard error of the mean (S.E.M). and were analyzed by Student’s independent t-test. All statistical analyses were performed with SPSS 22 software. Significance was reached at values of p < 0.05.
RESULTS
Running-Wheel ActivitiesDuring the injections, only two mice in the research group died. Both of them died after the 6th injection. At the baseline, the running wheel activities did not differ between the CFS and control groups (18217.9 ± 3184.9 vs. 17963.6 ± 2238.0). After the first two injections, the running wheel activity of CFS dropped markedly until the end of the 16th week, remaining at only 51.8% of the baseline. There were no significant differences between the Kampo group and the CFS vacant group with regard to running wheel activities (Fig. 1).
Immunochemistry StainingThere were no significant differences among the CFS, Kampo, and control groups in Ki-67, DCX, and NeuN/BrdU staining (Figs. 2a, b, c, e). 4-HNE positive cells could be seen all over the brain in CFS vacant and Kampo group, but not for the Control group (Fig. 2d).
Results of Western BlotAfter repeated injections of BA, the expressions of IL-1β, IL-6, 3-NT, and 4-HNE were significantly elevated in both the cortex and the hippocampus. Expression for 3-NT and 4-HNE was absent both in the cortex and the hippocampus in the control group. Treatment with TJ-41 and TJ-7 resulted in decreased expression of IL-1β, IL-6, and IFN-γ in the hippocampus but not in the cortex compared to the CFS vacant group, while no reductions of 3-NT and 4-HNE could be confirmed either in the cortex or in the hippocampus. Among the decreased expressions of IL-1β, IL-6, and IFN-γ in hippocampus, only IL-6 was decreased remarkably (Fig. 3).
DISCUSSION
In the current study, we decreased the dosage of BA to 25% in the first injection which resulted in significant reduction of mice death rate. Only 2 animals died (10%) out of 20 mice in the CFS group. This could be due to the decreased immune response or hypersensitivity reaction from the reduced dosage. Decreased dosage of BA would have helped the animals to adapt with the altered immune system and to withstand the additonal dosage.
A large number of CFS patients endure recurrent, persistent, or subacute bacterial and viral infections. To mimic the infections, we performed six repeated intraperitoneal injections in mice. Then we noticed increased expressions of cytokines (IL-1β, IL-6) and oxidative stress markers (3-NT, 4-HNE) in both the cortex and the hippocampus of CFS mice. This could be due to the intrusive bacteria that caused activation of Toll-like receptors (TLRs). The activated TLR-4 complex leads to the upregulated transcription of nuclear factor-kappaB (NF-κB) and the production of pro-inflammatory cytokines.34) Second, the cytokine message transmission from the peripheral to the central nervous system (CNS). The mechanisms governing the entry and exit of immune cells/cytokines in and out of the CNS used to be poorly understood.35,36) However Louveau et al. recently found functional lymphatic vessels in the CNS that are connected to the deep cervical lymph nodes.37) This finding suggests that peripheral inflammatory signals can overcome the blood-brain barrier and communicate with the CNS.
The evidence supporting the existence of neuroinflammatory processes in individuals with CFS also includes the findings of Nakatomi38) and Barnden.39) Activated microglia and dysfunctional astrocytes were identified by positron emission tomography (PET) and magnetic resonance imaging (MRI). The activated microglia releases a wide range of neurotoxins, including pro-inflammatory cytokines40) and NO.41) Although at physiological concentrations NO is a major neuromodulator, elevated levels of NO promotes neuropathogenesis.42) Increased NO levels and superoxide elevate the generation of peroxynitrite anions, which are very toxic.43) Peroxynitrite can trigger many kinds of pathologic abnormalities, including depletion of glutathione, lipid peroxidation in membranes, nitration of protein tyrosine residues, and damage to DNA.8) The neuroinflammation and O&NS interact and reinforce each other, and could contribute to the onset and retain of CFS.
Although it is generally believed that increased cytokines and NO negatively regulate hippocampal neurogenesis, no decreased neurogenesis was identified (by Ki-67, DCX, and BrdU staining) in the hippocampus of CFS mice in our study, which suggests that decreased neurogenesis might not be a pathologic pathway of CFS. This result supports the consensus that loss of newborn neurons is common in depression44) but not important to CFS. Decreased neurogenesis is mostly due to decreased levels of brain-derived neurotrophic factor (BDNF), which is detected in clinical depression and animal models of depression45,46) but not in CFS. The unchanged neurogenesis might be a unique feature of CFS that helps distinguish it from depression.
The results of the present investigation support the theory that TJ-41 and TJ-7 could help to decrease the levels of inflammatory cytokines. However, TJ-41 and TJ-7 could not improve the symptoms of CFS that might be related to their inability to decrease oxidative stress and to reduce the increased levels of cortical cytokines. In the current study, we could not examine the expression of additional cytokines and oxidative stress markers. Future studies could focus to elucidate the mechanisms that debilitate neural functions and screen more potent drugs that could decrease neuroinflammation and oxidative stress in brain.
