Role of Etanercept and Infliximab on Nociceptive Changes Induced by the Experimental Model of Fibromyalgia

Background: Fibromyalgia is a clinical condition that affects 1% to 5% of the population. No proper therapy has been currently found. It has been described that inflammation plays a central role in the nerve sensitizations that characterize the pathology. Methods: This paper aimed to evaluate the efficacy of etanercept and infliximab in the management of pain sensitization. Fibromyalgia was induced by three injections once a day of reserpine at the dose of 1 mg/kg. Etanercept (3 mg/kg) and infliximab (10 mg/kg) were administered the day after the last reserpine injection and then 5 days after that. Behavioral analyses were conducted once a week, and molecular investigations were performed at the end of the experiment. Results: Our data confirmed the major effect of infliximab administration as compared to etanercept: infliximab administration strongly reduced pain sensitization in thermal hyperalgesia and mechanical allodynia. From the molecular point of view, infliximab reduced the activation of microglia and astrocytes and the expression of the purinergic P2X7 receptor ubiquitously expressed on glia and neurons. Downstream of the P2X7 receptor, infliximab also reduced p38-MAPK overexpression induced by the reserpine administration. Conclusion: Etanercept and infliximab treatment caused a significant reduction in pain. In particular, rats that received infliximab showed less pain sensitization. Moreover, infliximab reduced the activation of microglia and astrocytes, reducing the expression of the purinergic receptor P2X7 and p38-MAPK pathway.


Introduction
Fibromyalgia is a clinical syndrome characterized by chronic, widespread pain, depression, and fatigue [1,2]. It affects approximately the 2-5.8% of people worldwide [3][4][5]. It is characterized by an inadequate transduction of nociceptive signaling and is a consequence of hypersensitivity to non-noxious stimuli [6,7]. Even though it is considered a non-inflammatory disorder [8][9][10], changes in inflammatory mediators [11][12][13][14][15] have been detected. Clinical and experimental evidence shows that inflammation has a key role in the development and perpetuation of chronic pain [16][17][18][19]. In particular, increased pro-inflammatory cytokine levels have been documented in the spinal cord of suffering animals, and their expression correlates with pain-like behaviors [20,21].

Experimental Timeline
In order to investigate the effects of TNF-α inhibitors, etanercept and infliximab, on fibromyalgia, rats were subcutaneously injected with reserpine 1 mg/kg for 3 consecutive days and were intraperitoneally injected with etanercept (3 mg/kg) or infliximab (10 mg/kg) after the last reserpine injection and 5 days after that ( Figure 1 Three days after the first reserpine injection, the animals showed hypersensitivity to 91 mechanical ( Figure 2A) and thermal ( Figure 2B,C) stimuli as compared to the sham group. 92 The reserpine group showed an increased hypersensitivity, also at different time points, 93 in particular at 5, 7, 14, and 21 days after the first reserpine injection. Etanercept and in-94

Effect of TNF-α Inhibitors on Pain Hypersensitivity Induced by Fibromyalgia
Three days after the first reserpine injection, the animals showed hypersensitivity to mechanical ( Figure 2A) and thermal ( Figure 2B,C) stimuli as compared to the sham group. The reserpine group showed an increased hypersensitivity, also at different time points, in particular at 5, 7, 14, and 21 days after the first reserpine injection. Etanercept and infliximab treatments importantly attenuated this overreaction (Figure 2A-C). In particular, this effect was prominent in infliximab-treated rats that, from 7 days from the reserpine injection, showed an increased withdrawal threshold and latency. No differences among the sham groups were detected; thus, no other results are shown about the sham + etanercept and sham + infliximab groups. Western blot analyses were conduced to investigate the activation of the expression 108 of the pain-related inflammatory mediators. Lumbar spinal cord tissues collected from the 109 reserpine group showed an increased expression of c-FOS ( Figure 3A) and nerve growth 110 factor (NGF) ( Figure 3B) as compared to the samples collected from the sham animals. 111 Differently, the samples collected from the etanercept and infliximab groups showed 112 strongly reduced expressions of both mediators. In particular, infliximab showed a 113 stronger effect of inhibition on c-FOS and NGF as compared to etanercept ( Figure 3A,B). 114 Figure 2. Efficacy of etanercept and infliximab administration on behavioural changes reserpine induced. Behavioural test: (A) Von Frey test, (B) hot plate test, (C) tail-flick test. For these analyses n = 5 animals for each group were employed. Repeated ANOVA analysis followed by a Bonferroni post-hoc test for multiple comparisons were employed to analyse the effect of treatments in reserpine model along time. A p-value <0.05 was considered significant. # p < 0.05 vs. vehicle, ## p < 0.01 vs. vehicle, *** p < 0.001 vs. sham, ### p < 0.001 vs. vehicle, $ p < 0.05 infliximab vs. etanercept.

Effect of TNF-α Inhibitors on Pain-Related Mediators Induced by Fibromyalgia
Western blot analyses were conduced to investigate the activation of the expression of the pain-related inflammatory mediators. Lumbar spinal cord tissues collected from the reserpine group showed an increased expression of c-FOS ( Figure 3A) and nerve growth factor (NGF) ( Figure 3B) as compared to the samples collected from the sham animals. Differently, the samples collected from the etanercept and infliximab groups showed strongly reduced expressions of both mediators. In particular, infliximab showed a stronger effect of inhibition on c-FOS and NGF as compared to etanercept ( Figure 3A,B). For these analyses n = 5 animals for each group were employed. Shown is a representative blot of lysates from five animals per group, together with a densitometric analysis normalized to housekeeping proteins. Results were analysed by one-way ANOVA followed by a Bonferroni post-hoc test for multiple comparisons. A p-value <0.05 was considered significant. # p < 0.05 vs. vehicle, ## p < 0.01 vs. vehicle, *** p < 0.001 vs. sham, $ p < 0.05 infliximab vs. etanercept.

Effect of TNF-α Inhibitors on Pro-Inflammatory Cytokines' Overexpression Induced by Fibromyalgia
Lumbar spinal cord tissues collected from the reserpine group showed an increased expression of pro-inflammatory cytokines as compared to the sham group ( Figure

Effect of TNF-α Inhibitors on Glia Activation Induced by Fibromyalgia
Immunofluorescence analysis showed increased GFAP ( Figure 5B) and Iba-1 ( Figure 5G) expressions in the lumbar spinal cord tissue of the reserpine group as compared to the sham group ( Figure 5A,F, respectively). Infliximab and etanercept administrations reduced both GFAP ( Figure 5C,D) and Iba-1 ( Figure 5H,I) expressions. In particular, this effect was prominent in the infliximab-treated rats for both the GFAP ( Figure 5E) and Iba-1 ( Figure 5J

Effect of TNF-α Inhibitors on Purinergic Receptor and p38-MAPK Expression Induced by Fibromyalgia
Western blot analysis showed an important increase in the P2X7 expression in lumbar spinal cord harvested from the reserpine group as compared to the sham group. Etanercept and infliximab administrations importantly reduced P2X7 levels ( Figure 6A). Additionally, Western blot analysis was employed to evaluate MAPK activation. In particular, p-38 was found phosphorylated in the reserpine group as compared to the sham group ( Figure 6B). Etanercept and infliximab treatments significantly reduced its phosphorylation ( Figure 6B), but infliximab strongly reduced both proteins' expressions. Western blot analysis showed an important increase in the P2X7 expression in 157 lumbar spinal cord harvested from the reserpine group as compared to the sham group. 158 Etanercept and infliximab administrations importantly reduced P2X7 levels ( Figure 6A). 159 Additionally, Western blot analysis was employed to evaluate MAPK activation. In 160 particular, p-38 was found phosphorylated in the reserpine group as compared to the 161 sham group ( Figure 6B). Etanercept and infliximab treatments significantly reduced its 162 phosphorylation ( Figure 6B), but infliximab strongly reduced both proteins' expressions. 163 164 Figure 6. Efficacy of etanercept and infliximab administration on P2X7 and phospo-p38 165 overexpression reserpine induced. Western Blot analysis of: (A) P2X7 and (B) phospo-p38 166 expressions. For these analyses n = 5 animals for each group were employed. Shown is a 167 representative blot of lysates from five animals per group, together with a densitometric analysis 168 normalized to housekeeping proteins. Results were analysed by one-way ANOVA followed by a 169 Bonferroni post-hoc test for multiple comparisons. A p-value <0.05 was considered significant. # p < 170 0.05 vs. vehicle, ## p < 0.01 vs. vehicle, *** p < 0.001 vs. sham, $ p < 0.05 infliximab vs. etanercept.  For these analyses n = 5 animals for each group were employed. Shown is a representative blot of lysates from five animals per group, together with a densitometric analysis normalized to housekeeping proteins. Results were analysed by one-way ANOVA followed by a Bonferroni post-hoc test for multiple comparisons. A p-value <0.05 was considered significant. # p < 0.05 vs. vehicle, ## p < 0.01 vs. vehicle, *** p < 0.001 vs. sham, $ p < 0.05 infliximab vs. etanercept.

Discussion
The data from this study, acquired using an experimental model of fibromyalgia, indicate that the administration of infliximab and etanercept displayed an analgesic effect.
We employed an experimental animal model of fibromyalgia, which was developed through the administration of reserpine, a biogenic amine depletor; long-lasting widespread nociceptive hypersensitivities were exhibited in rats [45][46][47][48][49]. It is noteworthy that the time course of pain-related behaviors was in parallel with the decrease in monoamine neurotransmitters after the reserpine treatment. Thus, monoamine depletion appears to cause nociceptive hypersensitivity in a rat reserpine-induced pain model, and the model could be useful to study pathological mechanisms of chronic, widespread pain such as with fibromyalgia. In our model of fibromyalgia, the animals showed hyperalgesia and allodynia and increased TNFR expression as compared to a sham group. This hypersensitivity to thermal and mechanical stimuli was reduced by etanercept and infliximab administrations. Both treatments showed important analgesic effects. Notably, the highly significant antinociceptive effects were observed in the infliximab-treated rats. Infliximab showed a strong potential to reduce fibromyalgia-evoked pain. Well in line with the literature, our data confirmed the major effect of infliximab administration as compared to etanercept: infliximab administration strongly reduced pain sensitization in thermal hyperalgesia and mechanical allodynia. These data are well in line with a reduced expression of the painrelated pro-inflammatory mediators and cytokines in the spinal cord. In particular, both infliximab and etanercept did not cross the blood-brain barrier; thus, the central effects seen here are consequences of the peripheral effects [50].
It has been described that astrocytes alone are not able to self-maintain this proinflammatory state but, in turn, activate microglia. Infliximab administration strongly reduced the activation of microglia and astrocytes, as compared to etanercept. Under this pathological condition, the activated microglia and astrocytes release a large amount of ATP, which activates specific purinoceptors, receptors ubiquitously expressed on glia and neurons [51,52]. ATP is a fundamental regulator of microglia activity [53,54], including cytokines' expression. It has been described that ATP is involved in the novo synthesis of pro-inflammatory cytokines from animal microglia through the activation of the P2X7 receptor. In response to the increased ATP, the ionotropic purinergic receptor induces Ca 2+ in microglia [51], which activates the pro-inflammatory cytokines' secretion. Downstream of the P2X7 receptor involves the MAP kinase activation [55,56]. Microglial p38-MAPK has been described as being involved in the development of neuropathic pain in studies on spinal cord injury [57,58] and peripheral nerve [59][60][61] injury. Infliximab administration significantly reduced P2X7 and p38 expression as compared to etanercept.
Overall, there is published evidence that infliximab does not bind rat TNF-α [62], while there is no evidence about etanercept [63]. This paper shows the indirect effect in rat models by anti-human-TNF-α antibody preparation such as etanercept and/or infliximab. The observed effects would be a "placebo" effect resulting from just giving an antibody in general, not specifically targeting TNF-α.
Our data showed the key role of the infliximab and etanercept in the development of fibromyalgia, its role in the activation of astrocytes and microglia, and its modulation of the signaling pathway involved in the modulation of pain.

Experimental Model
Reserpine administration was performed by subcutaneous injection of 1 mg/kg for 3 consecutive days [50]. Reserpine (Sigma-Aldrich, Saint Louis, MO, USA) was dissolved in distilled water with 0.5% acetic acid (vehicle). Sham animals received the same volume of vehicle but no reserpine administrations.

Treatment
Etanercept (3 mg/kg) was intraperitoneally injected the day after the last reserpine injection and on post-injection day 5. Infliximab (10 mg/kg) was intraperitoneally injected the day after the last reserpine injection and on post-injection day 5. The doses of etanercept and infliximab were based on previous studies [64].

Experimental Froups
Then, rats were randomly divided into several groups (n = 15): Sham: Rats were subcutaneously injected with vehicle instead of reserpine and treated with saline the day after the last reserpine injection and 5 days after.
Sham + etanercept: Rats were subcutaneously injected with saline instead of reserpine and intraperitoneally injected with etanercept.
Sham + infliximab: Rats were subcutaneously injected with saline instead of reserpine and intraperitoneally injected with infliximab.
Reserpine: Rats were administered with reserpine as previously described. Reserpine + etanercept: Rats were administered with reserpine as previously described and intraperitoneally injected with etanercept.
Reserpine + infliximab: Rats were administered with reserpine as previously described and intraperitoneally injected with infliximab.
Twenty-one days after reserpine injection, behavioural analysis were conducted, animals were sacrificed by isoflurane overdose (concentration to 5%) and L4-L6 area of spinal cord were collected for molecular analysis.

Von Frey Hair Test
Mechanical allodynia was evaluated on day 0 and 3, 5, 7, 14, and 21 days post-reserpine injection using a dynamic plantar von Frey hair esthesiometer, as already described [65]. The device is comprised of a force transducer equipped with a plastic tip. When pressure is applied to the tip, the force applied is calculated. The tip was applied to the plantar area, and a rising, upward force was exerted [66]. The withdrawal threshold was defined as the force, expressed in grams, at which the mouse removed its paw.

Hot Plate Test
On day 0 and 3, 5, 7, 14, and 21 days post-reserpine injection the hot plate test was performed. The hot-plate latency was calculated using a metal surface maintained at 53.6 • C. The rat was monitored, and the licking of a hind paw was set as the end point. Maximal latency accepted was 45 s [67].

Tail Flick Warm Water Test
On day 0 and 3, 5, 7, 14, and 21 days post reserpine injection, the tail-flick, warm water test was performed to evaluate pain threshold. The rat's tail was immersed in warm water (50 ± 0.5 • C), and the time between tail input and retraction was recordered. A maximum tail-flick latency of 10 s was employed to minimize tissue damage [67].

Western Blot Analysis
Western blot analysis was performed on lumbar spinal cord tissues. For this analysis, n = 5 animals for each group were employed. Cytosolic and nuclear extracts were prepared, with slight modifications of published procedures [68,69]. Tissues from each rat were suspended in extraction Buffer A, containing 0.2 mM PMSF, 0.15 mM pepstatin A, 20 mM leupeptin, and 1 mM sodium orthovanadate, homogenized at the highest setting for 2 min, and centrifuged at 12,000× g rpm for 4 min at 4 • C [70][71][72]. Supernatants represented the cytosolic fraction. The pellets, containing enriched nuclei, were resuspended in Buffer B, containing 1% Triton X-100, 150 mM NaCl, 10 mM Tris-HCl pH 7.4, 1 mM EGTA, 1 mM EDTA, 0.2 mM PMSF, 20 mm leupeptin, and 0.2 mM sodium orthovanadate [73]. After centrifugation for 10 min at 12,000 rpm at 4 • C, the supernatants contained the nuclear protein [74]. Protein concentrations were estimated by the Bio-Rad protein assay using bovine serum albumin as a standard [71]. Briefly, samples were heated to 100 • C for 9 of 13 5 min, and equal amounts of protein were separated on SDS-PAGE gel and transferred to nitrocellulose membrane [69,70]. The membranes were probed with specific antibodies, anti-c-FOS (sc-166940; Santa Cruz Biotechnology, Dallas, TX, USA), anti-NGF (sc-32300; Santa Cruz Biotechnology), anti-P2X7 ((Cell Signalling Technology, Danvers, MA, USA), or anti-p-p38 (sc-7983; Santa Cruz Biotechnology), in 1x PBS (phosphate-buffered saline), 5% w/v non-fat dried milk, and 0.1% Tween-20 at 4 • C, overnight [61,66]. To control the equal amounts of proteins, blots also were probed with an antibody against the b-actin protein (Santa Cruz Biotechnology). Signals were examined with an enhanced chemiluminescence (ECL) detection system reagent (Thermo Fisher, Waltham, MA, USA). The relative expression of the protein bands was quantified by densitometry with BIORAD ChemiDocTM XRS+ software and standardized to b-actin. The blot was stripped with glycine 2% and re-incubated several times to optimize the detection of the proteins and to visualize other proteins, minimizing the number of gels and transfers. The experiments were performed in triplicate and repeated three times with similar results.

Immunofluorescence Analysis
Immunofluorescence analysis was performed on lumbar spinal cord tissues. For this analysis n = 5 animals for each group were employed. N = 5 different fields from n = 5 different animals were evaluated. The areas between the lumbar vertebrae (L5-6) were harvested, fixed and decalcified [75]. In particular, the tissues were fixed for 24 h in a formaldehyde solution (10% in PBS) at room temperature and decalcified in Osteosoft solution (Merck Millipore, Milan, Italy). Next, the samples were dehydrated through a graded series of ethanol and xylene and embedded in BioPlast Plus (Bio Optica, Milan, Italy) [76,77]. Sections (5 µm in thickness) were prepared from tissues. After deparaffinization and rehydration, sections were boiled in 0.1 M citrate buffer for 1 min [78,79]. Non-specific adsorption was diminished by incubating the sections in 2% (v/v) normal goat serum in PBS for 20 min [80,81]. The sections were incubated with primary antibodies, anti-GFAP (Santa Cruz Biotechnology) or anti Iba-1 (Santa Cruz Biotechnology), overnight in a humidified chamber at 37 • C [74]. The sections were washed with PBS and were incubated with FITC-conjugated anti-rabbit Alexa Fluor-594 antibody (1:2000 v/v Molecular Probes, UK) for 1 h at 37 • C. The sections were analyzed using a Leica DM6 microscope (Leica Microsystems SpA, Milan, Italy) associated with Leica LAS X Navigator software (Leica Microsystems SpA, Milan, Italy). For immunofluorescence, the photographs were the outcomes of at least three independent experiments.

ELISA Analysis
The concentrations of TNF-α, IL-1β, IL-6, IL10, and MCP-1 were measured. Briefly, spinal cord tissues were homogenated in 1 mL PBS with 10 µL protease inhibitor at low speed for~20 seconds [72]. The samples were centrifuged at 14,000× g at 4 • C for 15 minutes; supernatants were employed, using respective ELISA kits according to the manufacturer's protocol, and analyzed using a microplate reader [61]. The values are expressed as pg/mL.

Statistical Evaluation
All values in the figures and text are expressed as mean ± standard error of the mean (SEM) of N number of animals. The results were analyzed by one-way ANOVA followed by a Bonferroni post hoc test for multiple comparisons. A p-value < 0.05 was considered significant; * p < 0.05 vs. sham; # p < 0.05 vs. vehicle; ** p < 0.01 vs. sham; ## p < 0.01 vs. vehicle; *** p < 0.001 vs. sham; ### p < 0.001 vs. vehicle; $ p < 0.05 infliximab vs. etanercept.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.