Background
Fatigue makes a significant contribution to the global disease burden. A spectrum of fatigue-related syndromes, known as chronic fatigue syndrome (CFS), includes cognitive problems, sleep disturbances, malaise and gastrointestinal symptoms [
1]. CFS is characterized by persistent and severe fatigue [
2]. In addition to fatigue, CFS patients also complain of somatic symptoms, such as irritable bowel syndrome (IBS) and migraines, and these symptoms can be worsened by stress and over exercise [
3‐
5]. According to reports published by the US Centres for Disease Control and Prevention (CDC), the incidence of CFS may be as high as 1 % in the US population, and the female to male ratio is 4:1 [
6].
Although many studies regarding fatigue have been published in recent decades, the exact pathophysiological mechanisms of fatigue are not well delineated. Several lines of evidence indicate that non-viral pathogen entry into the body induces the expression of cytokines in the central nervous system (CNS) [
7]. Recent investigations have revealed that the interaction between inflammatory pathways and the neuroendocrine system is associated with the manifestation of symptoms such as pain, fatigue, impaired memory and depression, which largely characterize at least some patients suffering from CFS [
8]. Induction of interleukin (IL)-1β expression and elevated inflammatory mediators in the brain are prerequisites for decreased locomotor activity and other fatigue-related behaviours [
9,
10]. Moreover, many CFS patients demonstrate abnormal hypothalamic-pituitary-adrenal (HPA) axis activity [
11,
12]. In patients with fatigue-predominant CFS, there was significant enhancement in nocturnal ACTH-adrenal signalling and marginally increased inhibitory feedback compared with that of healthy controls [
13]. In addition, fatigue improvement (defined as a reduction in fatigue score) has been reported in response to low-dose hydrocortisone therapy [
14]. The involvement of HPA axis dysfunction suggested that the diencephalon was the brain region responsible for fatigue sensation, although the molecular/cellular mechanisms of this process were still unclear.
In recent years, increased attention has been paid to the importance of the cytosolic signalling pathways of inflammation [
15]. Among these, the NOD-like receptor pyrin domain containing 3 (NLRP3) inflammasome is of particular interest. The NLRP3 inflammasome can be activated by a number of different stimuli known as danger-associated molecular patterns (DAMPs) [
16]. Among these stimuli, the production of reactive oxygen species (ROS) and mitochondrial dysfunction are also major contributors to the development of fatigue-related sensations [
17]. Once activated, the NLRP3 inflammasome can trigger increased induction of caspase-1 cleavage and IL-1β production [
18]. Cordero et al.’s study reveals that coenzyme Q10 (CoQ10) deficiency can induce NLRP3 inflammasome activation in the peripheral blood mononuclear cells (PBMCs) of CFS patients [
19].
However, whether the NLRP3 inflammasome in the CNS is involved in the development of fatigue sensation remains elusive. Therefore, the present study was designed to assess NLRP3 inflammasome activation status and reveal its potential involvement in the pathogenesis of peripheral immune system-induced fatigue. Behavioural and biochemical alterations were also measured to evaluate fatigue in a mouse model of CFS.
Methods
Chemicals and reagents
Lipopolysaccharide (LPS) from Escherichia coli 0111:B4 (catalogue: L2630-10MG) was purchased from Sigma-Aldrich (St. Louis, MO, USA); anti-mouse NLRP3 Ab (AG-20B-0014) was from AdipoGen Corp. (San Diego, CA); anti-mouse caspase-1 p10 Ab and anti-mouse actin Ab were from Santa Cruz Biotechnology, Inc. (Dallas, Texas); and anti-mouse IL-1β Ab was from Cell Signaling Technology (Beverly, MA). For immunofluorescence staining, anti-mouse caspase-1 p10 Ab (sc-22166) was from Santa Cruz Biotechnology, Inc. (Dallas, Texas); anti-mouse IL-1β Ab (ab9722) was from Abcom (Massachusetts, USA); Alexa Fluor 488 Goat anti-Rabbit, Alexa Fluor 555 Goat anti-Mouse and Alexa Fluor 647 Goat anti-Rabbit were from Life Technology (Shanghai, People’s Republic of China). The reagents listed above were prepared and used according to the manufacturer’s instructions.
Animals
C57BL/6 female mice (wild-type (WT) control) weighing 20–23 g were purchased from Shanghai Super-B&K Laboratory Animal Corp. Ltd. C57BL/6 NLRP3 KO female mice were provided by the Model Animal Research Centre of Nanjing University (AAALAC accredited). The animals were housed in a pathogen-free animal facility with 12-h light and 12-h dark cycles (8:00–20:00 light and 20:00–8:00 dark) with free access to food and water.
Treatment conditions
To induce the CFS model, the mice were injected intraperitoneally with 3 mg/kg of LPS or normal saline [
20]. Twenty-four hours after injection, the mice were subjected to a swim stress test for 20 min. The mice were forced to swim individually in a transparent plastic square jar (25 cm × 31 cm) containing 15-cm-deep water at 23 ± 2 °C. A light lead sinker (5 % of the body weight) was attached to the tail root of each mouse. The mice were then removed from the pool and dried with a clean towel. The pool water was replaced after each session. The following four groups of mice were created to test the effect of LPS treatment during the swim test: (a) control (saline)/no swim; (b) LPS/no swim; (c) control/swim; and (d) LPS/swim. To confirm our findings and test the importance of NLRP3 inflammasome activation in the CFS model, the weight-matched NLRP3 KO female mice were also subjected to LPS challenge and swim stress testing. Normal saline was administered to the control mice. Each group contained 8–10 animals. All of the animal studies were approved by the Ethics Committee of the Second Military Medical University, and all procedures were performed in compliance with the Guideline for Care and Use of Laboratory Animals published by the National Institutes of Health, USA.
Sample collection
The mice were anesthetized with pentobarbital sodium (100 mg/kg) and then sacrificed. The whole brain was rapidly extracted from the animals and placed on ice. The diencephalon tissue (mainly the hypothalamus) was quickly dissected and then frozen in liquid nitrogen. A representative image of the dissected region is shown in Additional file
1: Figure S1. Blood was collected by cardiac puncture and centrifuged at 3000
g for 15 min at 4 °C (Eppendorf 5801R centrifuge, Germany), and the serum was collected to measure the levels of certain biochemical mediators in the serum. The right quadriceps femoris was immediately dissected and weighed. Muscle samples were manually homogenized with a glass homogenizer using ice-cold physiological saline. Homogenates were centrifuged for 15 min at 3000
g (Eppendorf 5801R centrifuge, Germany), and the supernatants were collected to determine malondialdehyde (MDA) concentrations.
Assessment of behavioural parameters
The behavioural parameters consisted of locomotor activity assessment and a rota-rod test. After the experimental procedure, the animals were dried with a clean towel and allowed to rest in their original cages for 30 min. The locomotor activity was monitored for a total of 1 h using the Neuroscience Behaviour Mouse Cage Rack System (ShangHai Biowill Co., Ltd, Shanghai, China). This system contained 12 individual automated cages, which each continuously recorded the animal’s movement using a micro video camera. The video was then analysed using the native Motor Monitor software (ShangHai Biowill Co., Ltd). The data were collected and analysed in 10-min time intervals. For the rota-rod test, the animals were first conditioned at a constant speed of 10 rpm for a period of 5 min. The animals that failed the first conditioning were given two additional conditioning periods. For the assessment, the mice were individually placed on the rota-rod, which was adjusted to a speed of 30 rpm. The fall-off time was recorded for each mouse, and the longest period any animal could be maintained on the rod was 600 s.
Assessment of biochemical mediators
Mouse serum levels of lactic acid and MDA were determined using commercial assay kits (Xi Tang Biotechnologies Co., Ltd, Shanghai, China). Muscle levels of MDA were measured as well. IL-1β and IL-6 mouse serum levels were measured with ELISA kits (Dakewe Biotech Company Ltd., Shenzhen, China) according to the manufacturer’s instruction.
Real-time PCR analysis for NLRP3 and pro-IL-1β mRNA
The total RNA was extracted from frozen diencephalon tissues using Trizol reagent (Life Technologies, USA). The homogenate was mixed with 200 μl chloroform and then centrifuged at 12,000
g for 15 min at 4 °C. The aqueous phase (approximately 0.5 ml of the upper layer) was precipitated with an equal volume of isopropanol and centrifuged at 12,000
g for 10 min at 4 °C. The final total RNA pellet was resuspended in 20 μl DEPC water. Reverse transcription was performed with 1 μg total RNA using the Transcriptor First Strand cDNA Synthesis Kit (Roche Ltd, Swiss). A total of 2 μl first-strand cDNA solution was used for real-time RT-PCR in combination with a Fast Start Universal Probe Master Mix (ROX). All experiments were run in triplicate. The real-time PCR was run on an Applied Biosystems 7500 Real-Time PCR System (Life Technologies Corporation, USA). The primers used for RT-PCR are listed in Table
1. The threshold cycle (CT) of the target product was normalized to that of the internal standard GADPH.
Table 1
Primer used for real-time PCR in this study
NLRP3 | ACCAGCCAGAGTGGAATGAC | ATGGAGATGCGGGAGAGATA |
Pro-IL-1β | CTCACAAGCAGAGCACAAGC | TCCAGCCCATACTTTAGGAAGA |
GADPH | GTGTTTCCTCGTCCCGTAGA | AATCTCCACTTTGCCACTGC |
Western blot analysis
The brain samples were lysed with RIPA Lysis Buffer (Aidlab Biotechnologies Co., Ltd, Beijing, China) supplemented with a protease inhibitor “cocktail”, and protein concentrations in the extracts were measured using the BCA Protein Assay Kit (Aidlab Biotechnologies Co., Ltd, Beijing, China). An equal amount of each extract was separated by SDS-PAGE and then transferred onto nitrocellulose membranes. The membranes were blocked in a 5 % non-fat milk solution for 2 h at 24 °C. The blots were then incubated with primary antibodies for 24 h at 4 °C and then incubated with secondary antibodies for 2 h at 4 °C. Full film scans of the Western blot data were obtained with an Amersham Imager 600 (GE Healthcare Bio-Sciences AB, Sweden). The protein expression levels were quantified by measuring band intensities using ImageJ software (NIH, USA). The band intensity values of the proteins of interest were normalized to that of actin.
Immunofluorescence staining
After anesthetisation, the mice were transcardially perfused with normal saline (0.9 %), and brain tissues were fixed in a fresh 4 % paraformaldehyde solution (pH 7.4) at 4 °C. Coronal sections (30 μm) containing the diencephalon were prepared for immunofluorescence staining. Primary antibodies against NLRP3, caspase-1 and IL-1β were also used to delineate respective inflammasome components. DAPI (4′,6-diamidino-2-phenylindole) was used for nuclear staining. Alexa Fluor 488, Alexa Fluor 555 and Alexa Fluor 647 were used as secondary antibodies. Images of the stained specimens (five mice per group) were captured using an Olympus Research Inverted System Microscope IX71 (Olympus, Japan).
Statistical analysis
The results were measured and expressed as the mean ± SEM. Locomotor activity data were first analysed by a repeated measures analysis of variance (ANOVA) to determine the effects of the treatment. Other variations among groups were analysed by a univariate ANOVA followed by Dunnett’s test to compare any two groups. The comparisons made between two groups were evaluated using Student’s independent t test. All statistical analyses were performed using SPSS 21 software. Statistical significance was defined as p < 0.05.
Discussion
Experimental fatigue is classified into four categories: (1) physical fatigue, such as forced exercise and swimming; (2) mental fatigue; (3) environmental fatigue, such as heat exposure; and (4) immunologically induced fatigue [
23]. Among these models, immunologically induced fatigue is usually considered to be associated with the neuroendocrine-immune interactions [
24]. Lipopolysaccharide (LPS) treatment is commonly used to mimic bacterial infection and is also known to induce a behavioural sickness, which can be used as a model of fatigue [
25,
26]. In this study, we established an LPS-induced fatigue model with the addition of swim stress, which is a combination of immunological and physical fatigue.
In the behavioural test, our study demonstrated that wild-type mice showed significantly decreased locomotor activity only following LPS/swim stress treatment. Mice treated with either LPS alone or subjected to only swim stress also exhibited decreased activity relative to controls, but this difference was not statistically significant (Fig.
1). These results conflict with those from previously published reports [
27], and this discrepancy can possibly be explained by the low dose of LPS used here (3 mg/kg) and the relatively mild swim stress condition we applied. In our preliminary experiment, an intraperitoneal LPS injection at a dose of 10 mg/kg significantly affected locomotor activity without swim stress (data not shown). In addition, other published studies were mainly based on poly(I:C) intraperitoneal injection [
27,
28]. Strain differences may also explain the different results, but apart from the knockout mouse strain, we did not use any other mouse strains besides C57BL/6 mice. These differences may also be attributed to various experimental designs. Our results indicated that although latent or mild infection itself was insufficient to induce fatigue, it could increase vulnerability to stress and exhaustion. Biochemical parameters including lactic acid, MDA, IL-1β and IL-6 are also involved in fatigue-associated illness [
29]. So we measured their levels in our established mouse fatigue model. Lactic acid is a product of glycolysis under anaerobic conditions, and it accumulates during high-intensity physical exercise. As lactic acid accumulates, muscle tissue and blood pH decreases, which harms muscle tissue and causes fatigue [
30,
31]. We tested serum lactic acid levels to reflect physical fatigue and found no significant difference between groups. Intense exercise may cause an imbalance between the body’s oxidation and anti-oxidation systems, thereby producing more ROS. ROS attack polyunsaturated fatty acids (PUFA), which can lead to lipid peroxidation [
32]. MDA is one of the degradation products in the lipid peroxidation process [
33]. Thus, MDA is an ideal parameter to measure and to understand oxidation status and the production of ROS. In light of CFS pathophysiology, we tested MDA levels in both serum and muscle tissue and found no significant difference in either tissue among the four groups. When compared with other groups, the LPS/swim stress mice produced more serum IL-1β and IL-6, which are two important inflammatory biomarkers (Fig.
2). Taken collectively, our results suggest that the fatigue symptom observed in the LPS/swim stress mice was mainly attributed to inflammation-induced fatigue rather than muscular weakness, muscular pain or lipid peroxidation.
The NLRP3 inflammasome has recently emerged as an unexpected sensor of metabolic danger and stress [
34]. ROS production and mitochondrial dysfunction have been reported to trigger NLRP3 inflammasome activation [
35]. Indeed, NLRP3 activation has been implicated in the development of many major diseases such as gout, type 2 diabetes, obesity-induced insulin resistance and depression [
36,
37]. NLRP3 protein expression can be highly induced by stimulation with LPS through Toll-like receptor 4 (TLR-4), and IL-1β is the main cytokine produced by NLRP3 [
38]. It has been reported that IL-1β levels in the brain or other tissues are highly correlated with CFS symptoms [
39]. In this study, we investigated the role of NLRP3 inflammasome activation during LPS-induced fatigue. The NLRP3 inflammasome was strongly activated in LPS/swim stress mice, and exposure to swim stress significantly enhanced LPS-induced NLRP3 expression and activation (Fig.
4). Moreover, we found that NLRP3 activation was essential to fatigue pathogenesis, as shown by the attenuated behavioural performance observed in NLRP3 KO mice (Fig.
5). In Western blot and immunofluorescence analysis, the NLRP3 KO mice showed significantly decreased caspase-1 activation as well as reduced IL-1β production in the mouse diencephalon (Fig.
7). Interestingly, in the biochemical analysis, NLRP3 KO mice had significantly decreased the IL-1β serum levels, whereas the IL-6 levels in these mice were not different from those of the wild-type mice, which suggested that such fatigue model was mainly IL-1β-dependent. These results, together with other reports underlining the importance of IL-1β in CFS [
10,
40], indicated that LPS-induced fatigue is an IL-1β-dependent disorder and that NLRP3/caspase-1 inhibition therapy may be a promising option for anti-fatigue therapy. To the best of our knowledge, this is the first study to report NLRP3 inflammasome activation in the CNS in a model of fatigue.
NLRP3 inflammasome activation is often described in terms of a two-step process requiring two signals [
41]. For example, many initial signals, such as LPS, are used to prime the cell by inducing NLRP3 expression, and then a secondary signal forms the inflammasome complex and leads to the cleavage of caspase-1 and the maturation of IL-1β [
16,
42]. In vitro, this signal distinction is often less clear [
43,
44]. This is because the priming stimulus might itself lead to the release of activators such as ATP. In our results, although neither LPS treatment nor exposure to swim stress alone was able to trigger significant behavioural and biochemical changes, the combination of both could significantly promote NLRP3 activation in the mouse diencephalon. Our data are consistent with the two-step activation process. Our results showed slight caspase-1 p10 induction and IL-1β secretion in the LPS/no swim stress mice, and swim stress enhanced this effect similar to administering ATP in vitro. Overall, our study demonstrated that LPS/swim stress promotes robust NLRP3 inflammasome activation in the CNS.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
L-JZ and G-CL designed the experiments and supervised the project. Z-TZ, X-MD, X-JM, Y-GL and Y-CC conducted the experimental research. YZ, J-KC and C-LY analysed the data and wrote and edited the manuscript. G-CL provided the NLRP3 KO mice. All authors read and approved the final manuscript.