RESEARCH ARTICLE

A Chronic Fatigue Syndrome Model Demonstrates Mechanical Allodynia and Muscular Hyperalgesia via Spinal Microglial Activation Masaya Yasui,1,2 Takashi Yoshimura,1 So Takeuchi,1 Kyohei Tokizane,1 Makoto Tsuda,2,3 Kazuhide Inoue,2,3 and Hiroshi Kiyama1,2 Patients with chronic fatigue syndrome (CFS) and fibromyalgia syndrome (FMS) display multiple symptoms, such as chronic widespread pain, fatigue, sleep disturbance, and cognitive dysfunction. Abnormal pain sensation may be the most serious of these symptoms; however, its pathophysiology remains unknown. To provide insights into the molecular basis underlying abnormal pain in CFS and FMS, we used a multiple continuous stress (CS) model in rats, which were housed in a cage with a low level of water (1.5 cm in depth). The von Frey and Randall–Seritto tests were used to evaluate pain levels. Results showed that mechanical allodynia at plantar skin and mechanical hyperalgesia at the anterior tibialis (i.e., muscle pain) were induced by CS loading. Moreover, no signs of inflammation and injury incidents were observed in both the plantar skin and leg muscles. However, microglial accumulation and activation were observed in L4–L6 dorsal horn of CS rats. Quantification analysis revealed a higher accumulation of microglia in the medial part of Layers I–IV of the dorsal horn. To evaluate an implication of microglia in pain, minocycline was intrathecally administrated (via an osmotic pump). Minocycline significantly attenuated CS-induced mechanical hyperalgesia and allodynia. These results indicated that activated microglia were involved in the development of abnormal pain in CS animals, suggesting that the pain observed in CFS and FMS patients may be partly caused by a mechanism in which microglial activation is involved. GLIA 2014;00:000–000

Key words: microglia, chronic stress, pain, chronic fatigue syndrome, fibromyalgia

Introduction

P

atients with cryptogenic syndromes, such as chronic fatigue syndrome (CFS) and fibromyalgia syndrome (FMS), display chronic widespread pain (diffuse hyperalgesia and/or allodynia) and multiple symptoms, including fatigue, sleep disturbance, malaise, and cognitive dysfunction (Clauw, 2009; Clauw and Chrousos, 1997; Fukuda et al., 1994). Although these diseases show a substantial overlap of symptoms, their etiologies remain largely unclear. Currently, the pathogenesis for these diseases is proposed to result from the disintegration of the nervous, immune and endocrine systems from prolonged stresses including mental and physiological

burdens (Clauw, 2010). An animal model is necessary to provide pathophysiological insights into these syndromes at the molecular level. However, there are no currently established animal models to address the molecular basis underlying these syndromes. We have recently developed a chronic or continuous stress (CS) rat model that partially mimics the symptoms associated with CFS (2011; Konishi et al., 2010; Ogawa et al., 2005, 2009, 2012; Tanaka et al., 2003). In this model, a rat is placed in a cage with a very low level of water (1.5 cm in depth) for 1–5 days to supply a continuous multiple stresses (i.e., CS). By using this model, we have investigated alterations in the expression of genes in several organs, and

View this article online at wileyonlinelibrary.com. DOI: 10.1002/glia.22687 Published online Month 00, 2014 in Wiley Online Library (wileyonlinelibrary.com). Received Dec 26, 2013, Accepted for publication Apr 17, 2014. Address correspondence to Dr. Hiroshi Kiyama, Department of Functional Anatomy and Neuroscience, Nagoya University Graduate School of Medicine, 65 Tsurumaicho, Showa-ku, Nagoya, Aichi 466–8550, Japan. E-mail: [email protected] From the 1Department of Functional Anatomy and Neuroscience, Graduate School of Medicine, Nagoya University, Nagoya, Aichi, Japan; 2Core Research for Evolutional Science and Technology (CREST) of the Japan Science and Technology Agency, Saitama, Japan; 3Department of Molecular and System Pharmacology, Graduate School of Pharmaceutical Sciences, Kyushu University Fukuoka, Japan.

C 2014 Wiley Periodicals, Inc. 1 V

also reported dramatic changes in the pituitary gland at the molecular and cellular levels (Konishi et al., 2010, 2011; Ogawa et al., 2005, 2009, 2012; Tokizane et al., 2013). For instance, the expression levels of metallopeptidases are markedly altered; the damage-induced neuronal endopeptidase (Kiryu-Seo et al., 2000; Nagata et al., 2010) is downregulated in the anterior lobe (AL; Ogawa et al., 2005), whereas the Neprilysin (another member of the metallopeptidase family) is upregulated in the intermediate lobe (IL; Ogawa et al., 2005). Moreover, somatotrophs in the AL of pituitary exhibit atrophy, and growth hormone secretion is suppressed (Ogawa et al., 2012). A DNA microarray analysis in the pituitary gland further demonstrated a list of genes whose expression levels are altered (Konishi et al., 2010, 2011; Tokizane et al., 2013). Furthermore, rats exposed to CS show significant activation of melanotrophs in the IL, evidenced by the active secretion of alpha-melanocyte stimulating hormone (a-MSH), followed by the degeneration of melanotrophs (Ogawa et al., 2009). Since the higher level of blood a-MSH in patients suffering from CFS for less than 5 years is evidenced, this CS model may be concluded as a useful model for studying CFS (Shishioh-Ikejima et al., 2010). One of characteristic symptoms seen in patients with FMS and CFS is abnormal muscle pain, such as hyperalgesia (Clauw, 2009; Clauw and Chrousos, 1997; Fukuda et al., 1994). Although we have previously revealed several characteristic alterations in the CS animal model, it remains unclear whether this model displays abnormal pain. We here examined whether CS rats also show abnormal (muscle) pain, such as mechanical allodynia and hyperalgesia. In addition, we investigated the pathophysiology of the pain associated with this animal model, and microglia in spinal cord (SC) has merged as a pathogenesis of the abnormal pain.

Materials and Methods Experimental Animals Male Sprague-Dawley rats (SLC, Hamamatsu, Japan; 250 g) were used in this study. They were singly housed under a light-dark cycle (12:12 h) and kept in a temperature-controlled room (23 6 1) C with ad libitum access to food and water. 70 animals were used in this study. All rats were acclimatized for at least 1 week prior to the experiment and maintained in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996). The study was conducted with the approval of the local animal ethics committee in accordance with the regulations for animal experiments at Nagoya University (permission No. 24294), the Animal Protection and Management Law of Japan (No. 105), and the Ethical Issues of the International Association for the Study of Pain (Zimmermann, 1983).

CS Animal Model Rats were randomly assigned to CS or no CS (NCS) control. For the CS model, 8-week-old rats were transferred to cages filled with

2

water (23 6 1) C to a height of 1.5 cm for 1–5 days. Rats were replaced in fresh water cages every 24 h (Tanaka et al., 2003). NCS rats were maintained (for 1–5 days) in their home cages and transferred to new cages every 24 h. Body weight increased each day for only NCS rats. Previous studies show no significant change in blood glucose concentration and rectal temperature between the two groups (Ogawa et al., 2012). Behavior and sleep–wake states have also been previously described (Ogawa et al., 2012). To examine microglial proliferation in the SC, the thymidine analog bromodeoxyuridine (BrdU; 200 mg/kg, 10 mg/mL in saline) was injected intraperitoneally during the stress loading (for 5 days, n 5 3).

Behavioral Testing Randall–Selitto Test. All behavioral tests were performed blindly to the investigator. A Randall–Selitto analgesimeter (Ugo Basile, Italy) was used to measure the pressure pain threshold of the deep tissues. To accurately evaluate pain of the deep tissue, we used surface anesthetic (EMLA cream: AstraZeneca, UK). EMLA cream was applied to the shaved skin over the lower hind leg extensors for 30 min and then removed with ethanol (Nasu et al., 2010). The animals were restrained with a towel around the trunk for calming purposes, and treated gently during the experiments. A cone-shaped pusher with a rounded tip (diameter: 6 mm, and made in our laboratory) was applied to the belly of the lower hind leg extensors, including the tibialis anterior (TA) muscle, through shaved skin. A probe with a tip diameter 2.6 mm allows for the measurement of the muscular mechanical nociceptive threshold (Nasu et al., 2010; Takahashi et al., 2005). The speed of applied force was set at 16 g/s and the cutoff point was set at 250 g to avoid damage to the tissue. The intensity of the pressure that caused an escape reaction was defined as the withdrawal threshold. Training sessions were carried out for at least 4 consecutive days. Measurements were performed seven times at 30-s intervals, and the mean value (excluding minimum and maximum values) was taken as the nociceptive threshold. Von Frey Hair Test. The mechanical withdrawal threshold of the skin over the hindpaw plantar surface was measured with von Frey hairs (VFHs; North Coast Medical, San Jose, CA). Rats were placed on the metal mesh floor in cubical Plexiglas chambers. The animals were allowed to acclimatize to their surroundings for at least 15 min before testing. The filament was applied ten times in an ascending order of force (0.16, 0.4, 0.6, 1, 1.4, 2, 4, 6, 8, and 10 g) at intervals of 5 s. The threshold was determined by the method of limits (i.e., increasing and decreasing the forces of VFHs). When an animal showed at least 1 withdrawal response among trials and did not show any withdrawal by one grade lower filament, we then take the former (higher) filament force as a positive response (Yasui et al., 2012). This test was carried out in a blinded manner.

Reverse Transcription Polymerase Chain Reaction The plantar skin, TA muscle, and SC dorsal horn from CS and NCS rats were removed and quickly frozen in liquid nitrogen. At least five rats were used in each group. Total RNA was purified using the acid guanidine isothiocyanate/phenol/chloroform method and converted to complementary DNA with SuperScript III (Invitrogen, Carlsbad, CA).

Volume 00, No. 00

Yasui et al.: Microglia-Mediated Pain Under Chronic Stress

Amplification was performed as follows: glyceraldehyde-3-phosphate dehydrogenase (sense 50 -CTACATGGTCTACATGTTCCAGTATG30 , antisense 50 -AGTTGTCATGGATGACCTTGG-30 ; 26 cycles), interleukin 6 (IL-6; sense 50 -GGATACCACCCACAACAGAC-30 , antisense 50 -CATTGGAAGTTGGGGTAGGA-30 ; 34 cycles), interleukin 1b (IL-1b; sense 50 -GTGTCTGAAGCAGCTATGGC-30 , antisense 50 -TCCTTGTACAAAGCTCATGG-30 ; 32 cycles for TA and 34 cycles for Plantar skin), tumor necrosis factor a (TNF-a; sense 50 -CCACGCTCTTCTGTCTACTG-30 , antisense 50 -CCTCGTCCC TTGAAGAGAAC-30 ; 32 cycles for TA and 34 cycles for Plantar skin), at 94 C for 30 s, 60 C for 30 s and 72 C for 30 s. Chemokine (C-C motif) ligand 2 (CCL2; sense 50 -CTGTCTCAGCCAGAT GCAGT-30 , antisense 50 -GTGCTTGAGGTGGTTGTGGA-30 ; 33 cycles), at 94 C for 30 s, 61 C for 30 s and 72 C for 30 s. The amplified products were electrophoresed on 1.0% agarose gels and stained with Ethidium Bromide. For the positive control of inflammation, the complete Freund’s adjuvant (CFA; 100 lL) was injected into the crural muscles and plantar skin.

MD). The mean values of the number and area of Iba1-positive cells were obtained from two sections for each SC from each group.

Pharmacology Effects of intrathecal administration of the inhibitor of microglial activation, minocycline (Sigma-Aldrich), were examined. Under 2% isoflurane anesthesia, rats were implanted with a 32-gauge intrathecal catheter (ReCathCo, Allison Park, PA) in the lumbar enlargement (close to L4–5 segments) for intrathecal drug administration. The catheter placement was verified by hindlimb paralysis induced by intrathecal administration of lidocaine (2%, 5 lL). Rats that showed paralysis were excluded from the experiments. Minocycline (10 lg/ lL) or saline (as vehicle control) was intrathecally administered in naive rats. Alzet 2002 mini-osmotic pumps (Durect Corp., Cupertino, CA) were used for the continuous delivery of minocycline or saline at a rate of 0.5 lL/h. The infusion pump was connected to the catheter and the pump was implanted under the back skin. After a 5-day recovery from the operation, the animals were then moved to the water cage.

Immunohistochemistry

Statistical Analyses

After behavioral testing, animals were anesthetized with an intraperitoneal injection of sodium pentobarbital (45 mg/kg) and transcardially perfused with saline followed by Zamboni’s fixative [0.1 M phosphate buffer saline (PBS) containing 2% paraformaldehyde and 0.2% picric acid]. Bilateral L1–L6 SC segment, the plantar skin, and the TA were immediately dissected post-fixed in the same fixative for 12 h at 4 C, followed by immersion in 30% sucrose in 0.1 M PBS at 4 C for 2 days. The SCs were then frozen in liquid nitrogen and cooled isopentane, and processed for sectioning using the cryostat. The SCs were serially cut on a cryostat (20-lm thickness), and sections were thaw-mounted on Superfrost Plus microscope slides (Matsunami, Tokyo, Japan) and dried at room temperature overnight. To visualize the pathological changes, sections were exposed to the hematoxylin and eosin (HE) stain. Sections were blocked with 1% BSA and 0.3% Triton-X100 (Sigma-Aldrich, St Louis, MO), and incubated with the primary antibody, rabbit anti-Iba1 (1:1,000; Wako, Osaka, Japan), antiglial fibrillary acidic protein (anti-GFAP, 1:2,000; Sigma-Aldrich), or rat anti-BrdU (1:200; AbD Serotec, Oxford, UK) overnight at 4 C. Sections were then incubated with the secondary antibody, Alexa 488 donkey anti-rabbit immunoglobulin G (IgG; 1:1,000) or Alexa 594 donkey anti-rat IgG (1:1,000; Molecular Probes, Invitrogen, Carlsbad, CA), 90 min at room temperature.

All data are expressed as the mean 6 standard error of the mean, and were analyzed by a 2-way analysis of variance (ANOVA) with repeated measures followed by the Holm–Sidak multiple comparison test, except for the von Frey test. Data from this test were analyzed using a nonparametric two-way repeated measures ANOVA on ranks, followed by the Holm–Sidak comparison test. Significance was reached at values of P < 0.01.

Quantification of Activated Microglia Two tissue sections from each SC were randomly selected, and the number and area of Iba1-positive cells were measured under fluorescence microscopy. Only microglia containing a distinct nucleus was measured. Three observation fields (lateral, central, and medial field) of 10,000 lm2 in each of the dorsal horn regions (Laminae I and II, Laminae III and IV) were captured using a digital camera attached to the fluorescence microscope. The total area and number of Iba1positive microglial cells within the six fields was quantified using image processing and analyzing software, Image J (NIH, Bethesda,

Month 2014

Results CS Rats Exhibited Mechanical Allodynia and Hyperalgesia CS rats demonstrated tactile allodynia and mechanical hyperalgesia in bilateral hind limbs after 5 days CS. The von Frey test showed a significant decrease in PWT in rats exposed for 2 days after CS, and this response was restored to the control level on the third day after CS (Fig. 1A). In the Randall–Sellito test, the pressure threshold was significantly decreased immediately after CS induction (i.e., on day 1 after CS loading), and was maintained at this level for the following 4 days (Fig. 1B). CS Induced the Activation of Microglia in the Dorsal Horn Because pain behavior tests were examined in the hind limb, we examined microglial response in the SC at L1–L6 levels. A substantial increase in Iba-1-positive cells (i.e., microglia) was observed in the dorsal horn of L5 and L6, and a marginal to moderate increase was consistently seen in L2–L4 (Fig. 2A,D–I). Such marked and consistent increase in microglia was not observed in other regions of SC under 5 days CS (Fig. 2A), although a few, if any, microglial cells in the ventral horn had a slightly activated morphology at 5 day CS. As for astrocyte, we examined the morphology and number 3

prominently increased in the medial part. In the medial region of the dorsal horn, microglia were increased in both the superficial layers (Layers I and II) and deeper layers (Layers III and IV) of the lamina. The average number of microglia per unit area was relatively higher in superficial layers than in deeper layers (even for control rats); however, the increase in microglial number was approximately twofold for both layers (Fig. 3B,D). A similar result was observed in the area occupied by microglia (Fig. 3C,E). We next examined the alteration of microglial localization in L5 during stress loading (Fig. 4). In control rats, microglial cells were localized evenly throughout the dorsal horn (Fig. 4A). However, microglial numbers were significantly increased, particularly in the medial region, with 4 days of CS (Fig. 4E). A substantial increase was found on the fifth day of CS (Fig. 4F). These results indicated that the increase and activation of microglia occurred at the later stage of CS (4–5 days). To address whether the increase in microglia is due to proliferation or/and migration, we injected BrdU during the stress loading (5 days). In the dorsal horn of CS animals, several BrdU positive microglia were observed (Fig. 4G–I). In addition there also existed significant number of BrdU negative microglia in the same region, suggesting that the increase in microglial cells in dorsal horn under CS would be due to both proliferation and migration. FIGURE 1: Time course of mechanical withdrawal threshold following continuous stress (CS) loading. The mechanical threshold and the pressure pain threshold were measured by the von Frey hair test (A) and Randall–Selitto test (B), respectively. Tactile allodynia at plantar surface was induced for 2 days after CS loading (A), and the mechanical hyperalgesia at tibialis anterior muscle was induced for 4 days. NCS; no-continuous stress, right; right leg or paw, left; left leg or paw. n 5 6 (each group), ***P < 0.001, **P < 0.01, and *P < 0.05.

of astrocytes in CS animals using GFAP immunostaining, however no significant changes in number and morphology were observed throughout the lumber dorsal horn (Fig. 2B,C). Alterations of microglial number and morphology were observed to be negligible in L1 (Fig. 2D). Microglia in the dorsal horn of L5 and L6 (Fig. 2H,I, respectively) showed morphology of a typical activated shape—short processes and a thick cell body (see the enlarged box in Fig. 2H). Accumulation of microglia was mainly observed in the medial part of the dorsal horn, with prominent presence in deeper layers. The increase in microglial number and occupied area were more prominent in the medial part rather than in the lateral part. Statistical significance against control (NCS) in number was found at the medial and central, but not at the lateral, level (Fig. 3). These results confirmed the immunohistochemical observations, and thus suggested that microglia were 4

Apparent Inflammatory and Injury Incidents Were Absent in Peripheral Tissue Although this model did not induce any physical injuries to peripheral tissue, we examined the status of both injury and inflammation in the plantar skin and the TA muscle where the von Frey test and Randall–Selitto test was applied, respectively. HE staining did not reveal any inductions of inflammatory cell (i.e., lymphocytes and macrophage) migration and accumulation, as well as tissue degeneration (Fig. 5A,B). To rule out the possibilities of inflammation, muscular fatigue and injury, we examined peripheral serum levels of C reactive protein (CRP), lactate (LA), and creatine kinase (CK), respectively. No significant changes in these levels were found (Fig. 5C-E). These data suggested that peripheral tissue inflammation and muscular injury were not evident. In addition we further examined the expression of messenger RNAs (mRNAs) for inflammatory cytokines (IL-1b, IL-6, TNF-a) and chemokine (CCL2) using polymerase chain reaction (PCR). The expressions of these mRNAs in planter skin and leg muscles did not alter between CS and control animals except CCL2 mRNA in plantar skin (Fig. 5F). Minocycline Suppressed Pain Behavior and Microglial Activation We used minocycline to determine whether inhibition of microglial activation suppresses pain behavior under CS Volume 00, No. 00

Yasui et al.: Microglia-Mediated Pain Under Chronic Stress

FIGURE 2: Microglial accumulation and activation in dorsal horn of lumber (L) 1–6 after 5 days of CS loading. (A) Accumulation of Iba1positive cells (microglia) was seen in the dorsal horn of spinal cord, and such accumulation was not seen in other region of the spinal cord. (B) and (C) demonstrated GFAP immunostaining in the dorsal horn (L5) of control (B) and CS (C) animals. (D–I) Iba1-positive cells in dorsal horn of L1–L6 (D–I, respectively), (Box in H) a representative image of microglia at high power magnification. Prominent microglial accumulation and activation are observed in the dorsal horn of L5 (H) and L6 (I). Scale bar 5 200 lm (A) and 100 lm (B–I). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

conditions. According to the von Frey test, administration of minocycline significantly (P 5 0.012) attenuated CS-induced tactile allodynia, although the results of the PWT indicated a trend for minocycline to not fully suppress CS-mediated pain behavior (Fig. 6A). In the Randal Selitto test, CS-induced mechanical hyperalgesia was almost completely suppressed by minocycline (P 5 0.025; Fig. 6B). Similar results were obtained in both hind limbs. Because minocycline successfully suppressed pain behavior, we then evaluated the morphology of microglia in the dorMonth 2014

sal horn from minocycline treated CS rats. Vehicle-treated CS rats showed prominent microglial accumulation in the medial region of L5 dorsal horn (Fig. 7A), whereas markedly less microglia was observed in minocycline-treated CS rats despite some microglia with activated phenotypes remaining in deeper medial region (Fig. 7B). Quantification analyses revealed that microglial number and their occupied area in minocyclinetreated rats were significantly attenuated (P 5 0.032 in Fig. 7C, and P 5 0.003 in Fig. 7D) in Layers I and II (Fig. 7C,D). Similar tendency was also observed in the deeper layer (Fig. 7E,F); 5

FIGURE 3: Quantification of microglial number and morphology in L5 spinal cord. (A) A unit area (104 lm2) is designated at six regions within the dorsal horn (superficial medial, superficial central, superficial lateral, deeper medial, deeper central, and deeper lateral). (B and C) Microglial cell number (B) and occupied area (C) in superficial layers (I/II). (D and E) Microglial cell number (D) and occupied area (E) in deeper layers (III/IV). n 5 4 (each group), ***P < 0.001, **P < 0.01, and *P < 0.05. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

however, the reduction was less compared with the superficial layers (P 5 0.071 in Fig. 7E, and P 5 0.023 in Fig. 7F).

Discussion The present study demonstrated that CS elicited mechanical allodynia and muscular hyperalgesia (myalgia) accompanied by microglial activation in the dorsal horn of the SC. Strikingly, suppression of microglial activation by minocycline attenuated the allodynia and hyperalgesia, suggesting that microglial activation may be a cause of the pain, which occurs under chronic stress. 6

Allodynia and hyperalgesia are types of abnormal pain that occurs as a consequence of peripheral inflammation and nerve injury. Several studies have addressed the molecular mechanisms underlying inflammatory and neuropathic pain, with evidence showing that spinal microglia play a major role in eliciting abnormal pain (Ellis and Bennett, 2013; Inoue and Tsuda, 2009; Ji et al., 2013; Koizumi et al., 2013; Tsuda et al., 2003, 2013). The model used in this study did not induce direct nerve damages although the microglial activation in the dorsal horn was evident. In addition, morphological inflammatory incidents were not found in the plantar skin (used for the von Frey test) and TA muscle (used for the Volume 00, No. 00

Yasui et al.: Microglia-Mediated Pain Under Chronic Stress

FIGURE 4: Profile of microglial activation and accumulation during CS in dorsal horn of L5 spinal cord. Iba-1 positive cells (microglia) in dorsal horn of L5 spinal cord in NCS (A), and CS for 1 day (B), 2 days (C), 3 days (D), 4 days (E), and 5 days (F). Scale bar 5 100 lm. Most of the BrdU positive nuclei [red: indicated by arrowhead in (G)] locate in Iba1 positive (H) microglial cell (G–I). (I) shows merged images of (G) and (H). Scale bar 5100 mm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Randall–Seritto test). Furthermore, levels of CRP, CK, and LA in serum from the CS animals were comparable to those from control animals, suggesting that severe muscular damage and inflammation were not evident. We further examined the expression levels of mRNA for the inflammatory cytokines such as IL-6, IL-1b, and TNF-a, as well as the chemokine CCL2, using tissues from foot paw and TA muscle tissue by PCR, however no increases were detected except CCL2 mRNA in plantar skin (Fig. 5). The reason why the mRNA level of CCL2 was increased in skin but not in muscle is obscure, but significant accumulation of macrophage and other immune cells were not observed histologically. Altogether, these data would suggest that the increased number and activation of microglial cells in the dorsal horn of CS animals were not due to nerve injury or peripheral inflammation. The area that was occupied by the accumulation of activated microglia in CS animals was not completely overlap with the area where the accumulation of microglial cells was observed in animals with nerve injury and tissue inflammation. In CS animals, the area was predominantly restricted to the medial half of the dorsal horn, specifically from the superficial to deep layers. However, in the sciatic nerve injury Month 2014

model (or nerve injury-induced neuropathic pain model), activated microglia are observed in a much wider area of the dorsal horn (from medial to lateral), specifically with more localization in the superficial layers (I–III; Maeda et al., 2010). This difference in the localization of microglial cells between CS and nerve injury models may also support the idea that the microglial accumulation in CS animals is not due to inflammation nor nerve injury. There remains a possibility that microglia of the current study may have been stimulated by a subset of dorsal root ganglion (DRG) neurons that were distinct from those activated by inflammation and nerve injury. Previous anatomical data have reported that DRG neurons projecting into the tibial nerve terminate in the medial half of L5 dorsal horn (Molander and Grant, 1986). Further evidence was shown with a retrograde tracer dye injected into the plantar skin, which was transported to this region (Molander and Grant, 1985). Thus, the area in which microglial accumulation occurred in the present study corresponded to the area in which sensory information from the plantar and calf skin, and calf muscles, innervate. Our results do not provide evidence of injury and inflammation in those regions. Although no damage and inflammation was 7

FIGURE 5: Morphology of the plantar skin and the tibialis anterior muscle. Representative HE stainings of the plantar skin (A) and the tibialis anterior (B) after 5 days CS loading, No histological signs for inflammation and injury were evident in both tissues. Levels of creatine kinase (CK; C), C-reactive protein (CRP; D), and lactate (LA; E) in serum from NCS and CS animals. There were no significant differences in the levels between NCS (n 5 4) and CS (n 5 13). Scale bar 5 50 lm (A, B). (F) demonstrates mRNA expressions for inflammatory cytokines and CCL2 in the plantar skin and TA from CS and NCS. No significant differences between CS and NCS were observed in IL1b, IL-6 and TNF-a, mRNAs, whereas CCL2 mRNA was slightly increased in the plantar skin. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

detected, it would be plausible to assume that a continuous subliminal stimulation from the regions where the tibial nerve innervates may lead to the local activation of microglial cells in the dorsal horn. This subliminal stimulation may elicit activation of microglial cells, thus causing abnormal pain. Another characteristic of microglial localization in this study was the restricted increase in the number of microglia specifically in L4–L6, but not in L1–L3, thoracic and cervical 8

SC. The area, L4–L6, is associated with the hind limb, particularly leg and foot. CS rats demonstrate substantially increased locomotive activity throughout the day and night time, and also occupy a standing position with lower limbs for much longer periods compared with rats in the control cage (Ogawa et al., 2012). Therefore, the tonus of muscle, particularly the antigravity muscle, such as the soleus, is likely to be higher. Therefore, in the present study, this hypertonic Volume 00, No. 00

FIGURE 6: Minocycline suppresses abnormal pain behaviors. An intrathecal administration of minocycline partially suppresses the paw withdrawal threshold in the von Frey test (A) and reduces the pressure pain threshold of the anterior tibialis muscle in the Randall– Selitto test (B). Control (NCS): n 5 10, Vehicle: n 5 4, Minocycline (Mino): n 5 5, **P < 0.01, and *P < 0.05.

FIGURE 7: Microglial accumulation and activation under CS are suppressed by minocycline treatment. (A) Accumulation of Iba1-positive cells (microglia) in dorsal horn of 5-day CS-loaded rats. (B) The treatment of minocycline substantially reduced number of the Iba1positive microglia (B). (C and D) Minocycline treatment (Mino) suppressed the numbers of Iba1-positive cells (C, P 5 0.032) and their occupied area within 104 mm2 (D, P 5 0.003) in superficial Layers I and II. (E and F) Minocycline treatment also suppressed the number (E, P 5 0.071) and the occupied area (F, P 5 0.023) in deeper Layers III and IV, however the reduced number is less significant in deeper layers (E) than in superficial layers (C). *P < 0.05 and **P < 0.01. Scale bar 5 100 lm (A and B). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

muscular sensory information, possibly via the tibial nerve, may be transmitted to the medial region of L4–L6 dorsal horn, thus eliciting microglial activation in this region. This may suggest that a sort of activity dependent microglial activation may exist, although further evidences are needed. Recently, Arima et al. (2012) have demonstrated in a mouse model of experimental autoimmune encephalomyelitis in which autoreactive T cells enter the central nervous system (CNS) via the L5 SC, and this specific pathway is due to gravity-induced activation of sensory neurons by the soleus muscle in the leg (Arima et al., 2012). CS rats of the present study likely received a relatively stronger gravity burden continuously, thereby activating the L5 spinal nerve pathway. Thus, unidentified factors may have activated along this pathway affecting microglial cells in this region. Nevertheless, a continuous activation of some sensory stimuli from the hind limb may be a reason why microglial activation was restricted to L4–L6 of the SC. Abnormal pain may be caused by alterations in microglia located in the upper brain (e.g., cerebral cortex and limbic system) rather than in the SC. Our preliminary studies showed no prominent changes in microglia in the upper brain regions (data not shown). However, some studies reported that chronic stress could alter morphology of microglial cells in cerebral cortex. In a chronic stressinduced depressive model of mice, modestly increased numbers of Iba-1-positive cells, as well as an alteration of microglial morphology, in the prefrontal area occurs in susceptible animals compared with chronic stress-resilient animals (Couch et al., 2013; Tynan et al., 2010). Interestingly, minocycline treatment was shown to improve stress-induced cognitive impairment via the modification of microglial activity (Hinwood et al., 2012). A chronic stress model has recently shown increased ramification of microglia together with increased expression of b1-integrin in the prefrontal cortex (Hinwood et al., 2013). Furthermore, chronic restraint stress significantly increases cyclooxygenase 2 (Cox2) labeling in microglia/macrophages of the hippocampus and cortex of stressed mice, with no change in their cell numbers (Gerecke et al., 2013). Therefore, the involvement of activated microglia in the cerebral cortex and limbic area under chronic stress is attractive to consider when relating their activation responses to the mechanisms underlying abnormal pain caused by chronic stress. Although in this model we could not find any prominent activation of microglial cells in cerebral cortex and limbic area, a higher correlation between pain behavior and the responses of microglia to minocycline treatment in dorsal horn under CS should be emphasized. In addition, we could not rule out a possibility that a disorder of other central analgesic systems is implicated under CS, because minocycline treatment did 10

not completely suppress allodynia in the present study. Further more detail investigation of microglial changes in the brain under CS treatment is worth exploring. Establishing animal models with the full list of symptoms of CFS and FMS is actually impossible. However, the model used in the present study demonstrated some symptoms, such as sleep and endocrine system disorders (Ogawa et al., 2009, 2012), as well as abnormal pain (in this study). Interestingly, all these disorders have been shown to originate from CNS dysfunctions, such as hypothalamic dopaminergic attenuation (Konishi et al., 2010; Ogawa et al., 2012), and activation of spinal microglia (results from this study). Therefore, the present study together with the previous studies may be providing evidences that the etiology for CFS and FMS is originated from some disorders of CNS function under chronic stress.

Acknowledgment Grant sponsors: CREST, JST, Japan, and Ministry of Education, Culture, Sports, Science and Technology of the Japanese Government. We are grateful to Ms. Y. Tabata and N. Tawarayama, for their technical assistance, and Ms. A. Asano for secretarial assistance.

References Arima Y, Harada M, Kamimura D, Park JH, Kawano F, Yull FE, Kawamoto T, Iwakura Y, Betz UA, Marquez G, Blackwell TS, Ohira Y, Hirano T, Murakami M. 2012. Regional neural activation defines a gateway for autoreactive T cells to cross the blood-brain barrier. Cell 148:447–457. Clauw DJ. 2009. Fibromyalgia: An overview. Am J Med 122:S3–S13. Clauw DJ. 2010. Perspectives on fatigue from the study of chronic fatigue syndrome and related conditions. PM R 2:414–430. Clauw DJ, Chrousos GP. 1997. Chronic pain and fatigue syndromes: Overlapping clinical and neuroendocrine features and potential pathogenic mechanisms. Neuroimmunomodulation 4:134–153. Couch Y, Anthony DC, Dolgov O, Revischin A, Festoff B, Santos AI, Steinbusch HW, Strekalova T. 2013. Microglial activation, increased TNF and SERT expression in the prefrontal cortex define stress-altered behaviour in mice susceptible to anhedonia. Brain Behav Immun 29:136–146. Ellis A, Bennett DL. 2013. Neuroinflammation and the generation of neuropathic pain. Br J Anaesth 111:26–37. Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A. 1994. The chronic fatigue syndrome: A comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med 121:953–959. Gerecke KM, Kolobova A, Allen S, Fawer JL. 2013. Exercise protects against chronic restraint stress-induced oxidative stress in the cortex and hippocampus. Brain Res 1509:66–78. Hinwood M, Morandini J, Day TA, Walker FR. 2012. Evidence that microglia mediate the neurobiological effects of chronic psychological stress on the medial prefrontal cortex. Cereb Cortex 22:1442–1454. Hinwood M, Tynan RJ, Charnley JL, Beynon SB, Day TA, Walker FR. 2013. Chronic stress induced remodeling of the prefrontal cortex: Structural re-

Volume 00, No. 00

Yasui et al.: Microglia-Mediated Pain Under Chronic Stress organization of microglia and the inhibitory effect of minocycline. Cereb Cortex 23:1784–1797. Inoue K, Tsuda M. 2009. Microglia and neuropathic pain. Glia 57:1469–1479. Ji RR, Berta T, Nedergaard M. 2013. Glia and pain: Is chronic pain a gliopathy? Pain 154:S10–S28. Kiryu-Seo S, Sasaki M, Yokohama H, Nakagomi S, Hirayama T, Aoki S, Wada K, Kiyama H. 2000. Damage-induced neuronal endopeptidase (DINE) is a unique metallopeptidase expressed in response to neuronal damage and activates superoxide scavengers. Proc Natl Acad Sci USA 97:4345–4350. Koizumi S, Ohsawa K, Inoue K, Kohsaka S. 2013. Purinergic receptors in microglia: Functional modal shifts of microglia mediated by P2 and P1 receptors. Glia 61:47–54. Konishi H, Ogawa T, Kawahara S, Matsumoto S, Kiyama H. 2011. Continuous stress-induced dopamine dysregulation augments PAP-I and PAP-II expression in melanotrophs of the pituitary gland. Biochem Biophys Res Commun 407:7–12. Konishi H, Ogawa T, Nakagomi S, Inoue K, Tohyama M, Kiyama H. 2010. Id1, Id2 and Id3 are induced in rat melanotrophs of the pituitary gland by dopamine suppression under continuous stress. Neuroscience 169:1527– 1534. Maeda M, Tsuda M, Tozaki-Saitoh H, Inoue K, Kiyama H. 2010. Nerve injuryactivated microglia engulf myelinated axons in a P2Y12 signaling-dependent manner in the dorsal horn. Glia 58:1838–1846. Molander C, Grant G. 1985. Cutaneous projections from the rat hindlimb foot to the substantia gelatinosa of the spinal cord studied by transganglionic transport of WGA-HRP conjugate. J Comp Neurol 237:476–484. Molander C, Grant G. 1986. Laminar distribution and somatotopic organization of primary afferent fibers from hindlimb nerves in the dorsal horn. A study by transganglionic transport of horseradish peroxidase in the rat. Neuroscience 19:297–312. National Research Council, Guide for the Care and Use of Laboratory Animals, Washington DC: The National Academic Press, 1996.

Ogawa T, Kiryu-Seo S, Tanaka M, Konishi H, Iwata N, Saido T, Watanabe Y, Kiyama H. 2005. Altered expression of neprilysin family members in the pituitary gland of sleep-disturbed rats, an animal model of severe fatigue. J Neurochem 95:1156–1166. Ogawa T, Sei H, Konishi H, Shishioh-Ikejima N, Kiyama H. 2012. The absence of somatotroph proliferation during continuous stress is a result of the lack of extracellular signal-regulated kinase 1/2 activation. J Neuroendocrinol 24: 1335–1345. Ogawa T, Shishioh-Ikejima N, Konishi H, Makino T, Sei H, Kiryu-Seo S, Tanaka M, Watanabe Y, Kiyama H. 2009. Chronic stress elicits prolonged activation of alpha-MSH secretion and subsequent degeneration of melanotroph. J Neurochem 109:1389–1399. Shishioh-Ikejima N, Ogawa T, Yamaguti K, Watanabe Y, Kuratsune H, Kiyama H. 2010. The increase of alpha-melanocyte-stimulating hormone in the plasma of chronic fatigue syndrome patients. BMC Neurol 10:73. Takahashi K, Taguchi T, Itoh K, Okada K, Kawakita K, Mizumura K. 2005. Influence of surface anesthesia on the pressure pain threshold measured with different-sized probes. Somatosens Mot Res 22:299–305. Tanaka M, Nakamura F, Mizokawa S, Matsumura A, Nozaki S, Watanabe Y. 2003. Establishment and assessment of a rat model of fatigue. Neurosci Lett 352:159–162. Tokizane K, Konishi H, Yasui M, Ogawa T, Sasaki K, Minamino N, Kiyama H. 2013. Continuous stress promotes expression of VGF in melanotroph via suppression of dopamine. Mol Cell Endocrinol 372:49–56. Tsuda M, Beggs S, Salter MW, Inoue K. 2013. Microglia and intractable chronic pain. Glia 61:55–61. Tsuda M, Shigemoto-Mogami Y, Koizumi S, Mizokoshi A, Kohsaka S, Salter MW, Inoue K. 2003. P2X4 receptors induced in spinal microglia gate tactile allodynia after nerve injury. Nature 424:778–783. Tynan RJ, Naicker S, Hinwood M, Nalivaiko E, Buller KM, Pow DV, Day TA, Walker FR. 2010. Chronic stress alters the density and morphology of microglia in a subset of stress-responsive brain regions. Brain Behav Immun 24: 1058–1068.

Nagata K, Kiryu-Seo S, Maeda M, Yoshida K, Morita T, Kiyama H. 2010. Damage-induced neuronal endopeptidase is critical for presynaptic formation of neuromuscular junctions. J Neurosci 30:6954–6962.

Yasui M, Shiraishi Y, Ozaki N, Hayashi K, Hori K, Ichiyanagi M, Sugiura Y. 2012. Nerve growth factor and associated nerve sprouting contribute to local mechanical hyperalgesia in a rat model of bone injury. Eur J Pain 16:953– 965.

Nasu T, Taguchi T, Mizumura K. 2010. Persistent deep mechanical hyperalgesia induced by repeated cold stress in rats. Eur J Pain 14:236–244.

Zimmermann M. 1983. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16:109–110.

Month 2014

11