Pyridostigmine ameliorates pristane-induced arthritis symptoms in Dark Agouti rats

Objective Rheumatoid arthritis (RA) is a chronic inflammatory disorder. Pyridostigmine (PYR), an acetylcholinesterase (AChE) inhibitor, has been shown to reduce inflammation and oxidative stress in several animal models for inflammation-associated conditions. The present study aimed to investigate the effects of PYR on pristane-induced (PIA) in Dark Agouti (DA) rats. Method DA rats were intradermally infused with pristane to establish the PIA model, which was treated with PYR (10 mg/kg/day) for 27 days. The effects of PYR on synovial inflammation, oxidative stress, and gut microbiota were evaluated by determining arthritis scores, H&E staining, quantitative polymerase chain reaction, and biochemical assays, as well as 16S rDNA sequencing. Results Pristane induced arthritis, with swollen paws and body weight loss, increased arthritis scores, synovium hyperplasia, and bone or cartilage erosion. The expression of pro-inflammatory cytokines in synovium was higher in the PIA group than in the control group. PIA rats also displayed elevated levels of malondialdehyde, nitric oxide, superoxide dismutase, and catalase in plasma. Moreover, sequencing results showed that the richness, diversity, and composition of the gut microbiota dramatically changed in PIA rats. PYR abolished pristane-induced inflammation and oxidative stress, and corrected the gut microbiota dysbiosis. Conclusion The results of this study support the protective role of PYR in PIA in DA rats, associated with the attenuation of inflammation and correction of gut microbiota dysbiosis. These findings open new perspectives for pharmacological interventions in animal models of RA.

Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by synovial hyperplasia and inflammation, and cartilage or bone damage, which eventually lead to joint deformity and physical disability (1).Approximately 0.5-1% of the world's population is affected by RA and this population is at high risk of other systemic complications (2).Although the aetiology of RA is complex and not fully understood, the inflammatory response, such as cytokine and chemokine expression and excessive infiltration of inflammatory cells, plays an important role in the onset and progression of this disease (3).A large number of reports have shown a coexistence of inflammation and oxidative stress in patients and animal models for different diseases, including RA (4)(5)(6).The increased generation of reactive oxygen species caused by an imbalanced pro-/anti-oxidative system reportedly promotes the degradation of joint tissue and elicits and amplifies the abnormal inflammatory response (7,8).Likewise, the increased inflammation can also trigger oxidative stress and further aggravate the synovitis and bone or cartilage erosion (7).This suggests an interplay between inflammation and oxidative stress in RA.Nevertheless, clinical studies indicate that treatment targeting pro-inflammatory cytokines, such as tumour necrosis factor-α (TNF-α) or interleukin-6 (IL-6) could reduce the serum levels of oxidative stress and inflammatory markers, as well as attenuating the severity of arthritis (9)(10)(11), suggesting the crucial role of inflammation in the progression and treatment of RA.
RA is reportedly associated with dysbiosis of the gut microbiota (12).Thus, the composition, richness, and diversity of the gut microbiota are reported to dramatically change in people or animals with RA (13,14).Intriguingly, K/BxN or SKG mice, genetic models of RA that spontaneously develop arthritis, do not display severe disease when raised in germ-free conditions (15,16), whereas faecal microbiota transplantation, from patients or animals with arthritis, into those mice in the same germ-free conditions can induce arthritis (16,17).These findings suggest that the gut microbiota participates in the development of RA.Additional studies have confirmed that appropriate supplementation with probiotics or gut microbiota modulators can restore the composition of the microbiota and improve arthritis symptoms in rats (18)(19)(20).This highlights the importance of considering the status of the gut microbiota when evaluating the therapeutic effects of drugs for RA treatment.
Compelling evidence supports that improving vagal activity via acetylcholinesterase (AChE) inhibition or vagus nerve simulation ameliorates cardiovascular diseases (21,22), neurodegenerative disorders (23,24), and inflammatory diseases, such as pancreatitis, and systemic inflammation in endotoxic shock in rats (25,26).Based on these findings, it is reasonable and reliable to modulate the inflammatory response by enhancing vagal activity through AChE inhibition.Our previous studies showed that pyridostigmine (PYR), a classical reversible AChE inhibitor, exerted a significant anti-inflammatory effect on the animal models of cardiovascular diseases, including myocardial infarction and preeclampsia (27)(28)(29).However, little is known about the role of PYR in RA.Thus, we hypothesized that PYR could ameliorate the pristane-induced arthritis (PIA) manifestations in DA rats by modulating the inflammatory response and improving the gut microbiota.This study will determine the effects of PYR on experimental arthritis and provide novel insights for developing therapeutic strategies for RA treatment.

Animals
Dark Agouti (DA) rats, 8-10 weeks of age, weighing between 180 and 200 g, were provided by the Department of Biochemistry and Molecular Biology, School of Basic Medical Science, Xi'an Jiaotong University Health Science Center.All rats were housed under standard environmental conditions with a 12 h/12 h light/dark cycle, and fed with a normal chow diet and tap water ad libitum.Sixteen male rats were randomly divided into three groups: control group (CON, n = 5 rats), pristane-induced arthritis group (PIA, n = 5 rats), and pyridostigmine treatment group (PIA + PYR, n = 6 rats).The Institutional Animal Ethics Committee of the university examined and approved the experimental conditions (number 20201029).

Induction and evaluation of arthritis model and sample collection
To establish the experimental arthritis model, rats were intradermally injected with 150 μL pristane (2,6,10,14tetramethylpentadecane; ACROS Organics, Morris Plains, New jersey, USA) at the base of the tail under anaesthesia using isoflurane, and randomly distributed into the PIA group and PIA + PYR group; this day (d) was considered as d0.Animals in the control group received an intradermal injection of 150 μL 0.9% normal saline.Rats in the PIA + PYR group were treated with PYR (10 mg/kg/day) for 27 days via intragastric administration (gavage) after recovery from anaesthesia, while those in PIA and CON groups were given an equal volume of tap water.We adopted a macroscopic scoring system (arthritis score) to evaluate the development of arthritis and its severity every other day, as described previously (30).In brief, each metacarpophalangeal/metatarsophalangeal joint that is swollen and red will be given one point.The same is true for swollen or red interphalangeal joints.Wrist/ankle joints were given 1-5 points according to the extent of swelling.Therefore, the total score for the four paws in each rat ranged from 0 to 60 points.Two trained investigators assessed the arthritis score.We also determined the relative changes in body weight.At the end of the treatment period in this experiment, rats were killed with an overdose injection of pentobarbital sodium and the tissues were collected and stored in deep freezer at −80°C or fixed in 4% paraformaldehyde.Blood from the abdominal aorta was collected in ethylene diamine tetraacetic acid disodium (EDTANa 2 ) anticoagulant tubes, then centrifuged at 4°C and 3000 rpm for 15 min, and the resulting plasma was collected and stored at −80°C for further assays.

Histological analysis
The left hind paw was collected from each animal after it had been killed.Skin and muscle were removed from the paw before its fixation in 4% paraformaldehyde and decalcification in EDTA solution.The samples were embedded in paraffin and cut into 8-μm-thick tissue sections.These sections were stained with haematoxylin and eosin (H&E) and examined for synovial hyperplasia, and cartilage or bone erosion with a light microscope.

Measurement of oxidative stress markers in plasma
The enzymic activities of AChE (BC2025; Solarbio, Beijing, China), superoxide dismutase (SOD) (S0101S; Beyotime, Shanghai, China), and catalase (CAT) (S0051; Beyotime, Shanghai, China), and the concentrations of nitric oxide (NO) (S0021S; Beyotime, Shanghai, China) and malondialdehyde (MDA) (BC0025, Solarbio, Beijing, China) in plasma were measured using commercially available kits according to the manufacturers' instructions.The principle of these assays is that the addition of samples containing the aforementioned enzymes to the working solution results in the generation of coloured compounds.The depth of colour is dependent on the concentration and activity of the enzyme.Two separate assays with two replicates were carried out.The optical density (OD) of samples in each assay was read using the microplate reader at appropriate wavelengths, and data were averaged for each sample and each group.

Analysis of mRNA expression
Synoviums were immediately collected in nuclease-free tubes and stored at −80°C.One millilitre of Trizol (15596026; Invitrogen, Carlsbad, California, USA) was used to isolate the total RNA, then 500 μL of isopropanol was added to precipitate it and 1 mL of 75% ethanol was used to wash it.Finally, the total RNA was resuspended with 20 μL RNase-free water and its concentration was measured using a Nanodrop spectrophotometer.One microgram of total RNA with OD260/ 280 ratios of 1.8-2.0 was reversely transcribed into cDNA using the GoScript TM Reverse Transcription Mix, Oligo(dT) kit (A2790; Promega, Madison, Wisconsin, USA).The real-time quantitative polymerase chain reaction (RT-qPCR) was performed using the CFX Connect TM Real-Time PCR Detection System (Bio-Rad, Hercules, California, USA) with GoTaq®qPCR Master Mix (A6001; Promega, Madison, Wisconsin, USA).The relative expression of target genes, normalized to β-actin, was determined using the 2 −ΔΔct method.For each gene, two independent assays were carried out with three biological replicates for each sample.The list of primers used is shown in Table 1.

Analysis of gut microbiota composition
Rats faeces were collected in germ-free tubes for 16S rDNA sequencing and bioinformatic analysis at the end of the experiment on d27.Data generated by gene sequencing were read using an Illumina MiSeq/Novaseq machine (Illumina, San Diego, CA, USA), following the manufacturer's instructions.Cluster analysis of operational taxonomic units (OTUs) was performed based on the SILVA 132 database.Alpha and beta diversity analysis, principal coordinates analysis (PCoA), and analysis of similarities (ANOSIM) were used to describe the features of the intestinal microbiota in the different groups.

Statistical analysis
All data were analysed using one-way analysis of variance (ANOVA) with Tukey's multiple comparisons test in GraphPad Prism 8.0.2 software, except for those related to arthritis score and changes in body weight, which required repeated-measures ANOVA.All results were presented as mean ± sem.A level of p < 0.05 was considered for statistical significance.

Effect of pyridostigmine on arthritis symptoms in PIA rats
To investigate whether PYR could alleviate the symptoms of PIA, arthritic rats received oral administration of PYR at the dose of 10 mg/kg once per day.Arthritis scores and representative images of H&E staining of the ankle are shown in Figure 1.The administration of PYR significantly inhibited the circulating AChE activity in PIA rats (Supplementary Figure S1).The arthritis scores, and joint swelling and redness markedly decreased after PYR treatment (Figure 1A,C).PYR also had an ameliorative effect on body weight loss over the experimental period (Figure 1B).H&E staining revealed that PYR reduced the infiltration of inflammatory cells, synovial hyperplasia, and cartilage destruction compared with the PIA group (Figure 1D).This suggests that the experimental arthritis model, namely PIA, was successfully established and that PYR significantly improved the symptoms of PIA in DA rats.

Pyridostigmine attenuates arthritis in rat
Pyridostigmine attenuated inflammation and oxidative stress in PIA rats We evaluated the expression of inflammatory cytokines or matrix metalloproteinases (MMPs) in synovium and the levels of oxidative stress markers in plasma.The qPCR data showed that PYR significantly reduced the mRNA level of the synovial pro-inflammatory cytokines, namely IL-1β, IL-6, and TNF-α, but seemingly had no effect on the transcription level of anti-inflammatory cytokines IL-4 and IL-10 (Figure 2A-E).Increased mRNA levels of MMP3 and MMP13, which are associated with cartilage and bone matrix degradation in response to pristane injection, were markedly reduced by PYR intake (Figure 2F,G).We found a significant increase in NO and MDA levels in PIA rats compared with the control group.This was reduced by PYR treatment (Figure 2H,I).Meanwhile, the enzymic activities  of SOD and CAT also showed an increasing trend in PIA and a decreasing trend in response to PYR administration, but these changes were not statistically significant (Figure 2J,K).Collectively, these findings suggest that PYR treatment modulates the inflammatory response and oxidative stress.

Pyridostigmine changed the gut microbiota diversity and richness in PIA rats
As dysbiosis of the gut microbiota is reportedly involved in the occurrence and development of RA (12,31), we assessed the composition of the gut microbiota in rats with arthritis, with or without PYR treatment, through 16S rDNA sequencing of amplicon.The sequencing results were first analysed for β-diversity with PCoA (Figure 3A), which indicated clear differences in gut microbiota between the three groups.In addition, the analysis of similarities (ANOSIM) showed an evident separation between the CON group and the PIA group (ANOSIM, p = 0.009, r = 0.996) (Table S1).However, the difference in composition of the gut microbiota between the PIA group and PIA + PYR group was less pronounced (ANOSIM, p = 0.002, r = 0.405) (Table S1).Furthermore, Good's Coverage Index, which reflects the coverage of each sample library, suggested that the sample size was large enough to detect all sequences in these samples (Figure 3B).The Shannon and Simpson Index for community diversity was reduced in PIA rats, but was increased by PYR (Figure 3C,D).The community richness of the gut microbiota showed a tendency to increase in both the PIA and PIA + PYR groups, as indicated by the Chao 1 and ACE indices (Figure 3E,F).The trend towards further elevation in Chao 1 and ACE in the PIA + PYR group compared with the PIA group indicates that PYR treatment may increase the species richness of the gut microbiota.

The composition of the gut microbiota in PIA rats was altered by pyridostigmine treatment
According to the sequencing results, the representative sequences of each OTU in different groups were annotated to analyse the composition of the gut microbiota.Counting the number of identified entries, at each level of bacterial classification, we found a significant increase in the number of entries from phylum to species in the PIA group compared with the CON group (Figure 4A).PYR treatment further increased the number of species at each taxonomic level, but this increase was statistically significantly different only at the genus and species levels (Figure 4A).
Based on the above results, we examined the gut microbiota at each taxonomic level.The number of bacteria belonging to the Bacteroidetes phylum (Figure 4B), Bacteroidia class (Figure 4E), Bacteroidales order (Figure 4G), Prevotellacea family (Figure 4I), and Prevotellaceae_UCG-001 genus (Figure 4K) was markedly increased in the PIA group compared with the CON group, and this was reduced with PYR treatment.Meanwhile, the number of Firmicutes phylum (Figure 4C), Clostridia class (Figure 4F), Clostridiales order (Figure 4H), Lachnospiraceae family (Figure 4J), Lachnospiraceae_NK4A136_group genus (Figure 4L), and f__Lachnospiraceae_Unclassified genus (Figure 4M) was significantly decreased in the PIA group compared with the CON group, and these bacteria were also up-regulated after PYR treatment.Similar results were also found in the linear discriminant analysis effect size (LEfSe) analysis, which suggested an abundance of bacteria in the Bacteroidetes phylum and Prevotellaceae_UCG-001 genus in the PIA group, and an abundance of bacteria in the Firmicutes phylum in the CON group (Figure S2).Furthermore, we observed a significantly decreased Firmicutes/ Bacteroidetes (F/B) ratio, which is a feature of gut microbiota dysbiosis, in the PIA group, compared with the CON group (Figure 4D).The F/B ratio was increased after PYR treatment (Figure 4D), which shows that PYR could ameliorate the gut microbiota dysbiosis.
The functional profiles of the gut microbiota were predicted using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUST) tool.The OTUs were mapped to level 1 (kingdom), level 2 (phylum), and level 3 (class) of the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway.At level 1, compared with the CON group, the relative abundances of metabolism and genetic information processing were higher in the PIA group (Figure 5A).At level 2, the relative abundances of amino acid metabolism, replication, and repair, and energy metabolism, etc, was increased, while cell motility, signal transduction, and xenobiotic biodegradation and metabolism were significantly decreased when comparing the results of the CON and PIA groups (Figure 5B).PYR treatment improved these changes in these categories.At level 3, 50 categories had significantly changed in OTU number between the CON and PIA groups (Table S2).Of these, four had changed significantly when comparing the CON to the PIA group; these were steroid hormone biosynthesis, flavone and flavonol biosynthesis, amyotrophic lateral sclerosis, and prion diseases (Figure 5C-F).Thus, these data suggest that the metabolic pathways of bacteria in PIA rats were different from those in CON rats, and PYR could partially correct these metabolic disorders to normal levels.

Discussion
The present study aimed to investigate the effects of PYR on PIA.Our results showed that (i) PYR significantly ameliorated body weight loss, joint redness and swelling, and synovium hyperplasia and cartilage erosion in rats with arthritis; (ii) PYR decreased the expression of IL-1β, IL-6, TNF-α, MMP3, and MMP13 in synovium, and reduced the plasma levels of NO, MDA, SOD, and CAT in PIA rats; and (iii) PYR markedly improved the abundance, diversity, and composition of the gut microbiota in PIA rats.Collectively, these findings suggest that the beneficial effects of PYR on arthritis induced by pristane are associated with a reduction in inflammation and the improvement of gut microbiota dysbiosis.
RA is a chronic inflammatory disorder that mainly affects synovium and joints, and heavily impairs the quality of life of affected people.To better understand the pathogenesis of RA and test the efficacy of antiarthritis drugs, animal models are widely used in RA studies.On the basis that PIA in DA rat is a wellestablished animal model with similar clinical features and disease course to RA, as well as being in common use and having excellent reproducibility (32), we used this model in the current study.We observed significant joint swelling and redness, loss of body weight, and synovium hyperplasia or cartilage erosion in PIA rats, suggesting the success of the model.Encouragingly, PYR administration significantly inhibited the AChE activity and reduced the arthritis symptoms in DA rats.
As RA is a chronic inflammatory joint disease, inflammation is believed to play a crucial role in the development of RA (6,33).Previous studies showed that inflammatory markers, including IL-6, TNF-α, and C-reactive protein, are higher in synovial fluid or serum and correlate significantly with disease severity in RA patients (34,35).Accordingly, a growing body of literature indicates that anti-inflammatory measures can reduce oxidative stress and clinical manifestations in patients or animal models of RA (6,9,36).Similarly, in the current study, we confirmed that the increased expression of the pro-inflammatory cytokines IL-1β, IL-6, and TNF-α in synovium of rats with arthritis was reduced by PYR administration.Moreover, the elevated expression of MMP3 and MMP13, related to cartilage degradation or bone resorption, was also reduced after PYR treatment.Thus, we presume that the amelioration of arthritis symptoms may be associated with the antiinflammatory effects of PYR.
In 2006, Holmdahl's team reported an increased oxidative burst upon pristane injection in DA rats (37).In line with this study, we found an increase in the levels of MDA and NO, markers of oxidative stress, in PIA in DA rats.Moreover, SOD and CAT, two antioxidant enzymes that protect against free radical formation and oxidative damage, were increased in this model.However, a previous report showed that phytol (3,7,11,15-tetramethyl-2-hexadecene-1-ol), an oxidative burst inducer, prevented and ameliorated arthritis established in DA rats using pristane (37).Therefore, the increased oxidative stress markers in pristane-injected rats, compared with normal DA rats, may reflect a secondary response to increased inflammation.The reduction in oxidative stress markers by PYR in PIA seems not to play a causative role in improving arthritis symptoms in DA; rather, it is more likely to be an associated phenomenon.Further investigation could elucidate this observation.
Currently, compelling evidence indicates that gut microbiota dysbiosis is also strongly linked to the occurrence and development of RA (12,38).Under normal conditions, microorganisms are key players in the maintenance of homoeostasis of the host's intestinal and systemic immunity through diverse bioactive metabolites (39,40).In RA, the production of inflammatory cytokines, gut permeability, and the richness and diversity of the microbiota change dramatically, perturbing the homoeostasis of immunoregulation and subsequently aggravating the disease (41,42).Therefore, the normalization of gut dysbiosis could be considered as an effective means to ameliorate RA.In the current study, our data showed that PYR administration improved the diversity, abundance, and evenness of the gut microbiota, and attenuated the symptoms of arthritis in PIA rats.Besides, in agreement with other research in patients or animals with RA (20,38), PIA rats showed an increased abundance of prevotella in the gut microbiota.It has been reported that an increased abundance of prevotella may be a pathological factor that participates in human disease, including RA, by promoting Th17-cell-mediated mucosal inflammation and immune response (16,43).In contrast, Lachnospiraceae, a short-chain fatty-acid (SCFA)-producing family belonging to Clostridium cluster XIVa of the Firmicutes phylum (44), produces SCFAs with immunosuppressive and antiinflammatory effects (45).In this study, a reduced abundance of Lachnospiraceae and an increased abundance of prevotella were found in PIA rats compared with the control group.However, the PYR intervention increased the abundance of Lachnospiraceae and reduced that of prevotella.Furthermore, the function prediction of gut microbiota showed that PYR could adjust the metabolic pathways to normal levels.Thus, in combination with the results of inflammation and oxidative stress, we speculate that changes in the gut microbiota induced by PYR are associated with the ameliorative effects of PYR on PIA rats, and the detailed mechanisms are a worthwhile subject for further investigation.

Limitations
One limitation of the present study is that we did not explore the mechanisms of PYR.Indeed, in this experiment, we only explored the effect of PYR as a drug employed in many inflammation-associated conditions.Since PYR is intended to increase the level of ACh in vivo, determining the concentration of ACh upon PYR administration needs to be addressed.We also acknowledge that this study used a small sample size and did not perform a dose titration of PYR to capture the peak dose of its anti-arthritic effect.Further studies should address these concerns to provide more useful data.In addition, the effect of PYR on ACh nicotinic and muscarinic receptor expression in synovium could deepen our understanding of the mechanisms of PYR, as the activation of these receptors by ACh or its agonists plays a critical anti-inflammatory role.Further investigations should provide more information on this aspect of the use of PYR in PIA in DA rats.

Conclusion
The current investigation found that PYR, an AChE inhibitor, could ameliorate PIA symptoms in DA rats.This protective effect of PYR was associated with significant decreased inflammation and correction of the gut microbiota dysbiosis.Our findings open new perspectives for pharmacological interventions in RA animal models.

Figure 4 .
Figure 4. Oral administration of pyridostigmine (PYR) altered the gut microbiota composition in rats with pristane-induced arthritis (PIA).(A) Species numbers were identified at each level of taxonomic analysis.(B, C) Differences in the intestinal microbiota at the phylum level.(D) Firmicutes/Bacteroidetes (F/B) ratio.(E, F) Differences in the intestinal microbiota at the class level.(G, H) Differences in the intestinal microbiota at the order level.(I, J) Differences in the intestinal microbiota at the family level.(K-M) Differences in the intestinal microbiota at the genus level.(A-M): Control (CON), n = 5 rats; PIA, n = 5 rats; PIA + PYR, n = 6 rats.Data are shown as mean ± sem.*p < 0.05, **p < 0.01, ***p < 0.001 vs CON group; #p < 0.05 vs PIA group.

Figure 5 .
Figure 5. Functional profiles of the gut microbiota in the control (CON), pristane-induced arthritis (PIA), and pristane-induced arthritis plus pyridostigmine (PIA + PYR) groups.(A, B) The function of the gut microbiota in level 1 (kingdom) and level 2 (phylum) was predicted by the PICRUST tool.The results are shown as the percentage of sequencing entries mapped to every function category operational taxonomic unit (OTU) in the total entries.(C-F) In level 3 (class), the first four categories show a significant difference in the number of entries between the three groups.Data are shown as mean ± sem.*p < 0.05 vs CON group; #p < 0.05 vs PIA group.

Table 1 .
Primer sequences used in real-time quantitative polymerase chain reaction (RT-qPCR).