Sensory Neuron Expressed FcγRI Mediates Postinflammatory Arthritis Pain in Female Mice

Persistent arthritis pain after resolution of joint inflammation represents a huge health burden in patients with rheumatoid arthritis (RA). However, the underling mechanisms are poorly understood. We and other groups recently revealed that FcγRI, a key immune receptor, is functionally expressed in joint nociceptors. Thus, we investigated a potential role of sensory neuron expressed FcγRI in postinflammatory arthritis pain in a mouse model of collagen antibody-induced arthritis (CAIA). Here, we show that global deletion of Fcgr1 significantly attenuated mechanical hyperalgesia in the ankle and hind paw of female mice in both inflammatory and postinflammatory phases of CAIA. No obvious differences in cartilage destruction were observed after resolution of joint inflammation between genotypes. In situ hybridization (ISH) revealed that a larger proportion of dorsal root ganglion (DRG) neurons expressed Fcgr1 mRNA signal in the late phase of CAIA. Conditional deletion of Fcgr1 in primary sensory neurons produced similar analgesic effects without affecting joint swelling. Knockdown of Fcgr1 expression within DRG in the postinflammatory phase of CAIA alleviated persistent pain. Inflammation within DRG after resolution of joint inflammation in the CAIA model was evidenced by T cell and neutrophil infiltration and upregulated mRNA expression of numerous inflammatory mediators. Yet, such changes were not altered by genetic deletion of Fcgr1. We suggest that neuroinflammation within the DRG after resolution of joint inflammation might upregulate FcγRI signaling in DRG neurons. Sensory neuron expressed FcγRI thus merits exploration as a potential target for the treatment of arthritis pain that persists in RA patients in remission.


INTRODUCTION
Rheumatoid arthritis (RA) is a common chronic autoimmune disease affecting 1% of the population across the world. Chronic pain is a dominant clinical symptom in RA patients, and poses an enormous health burden. Although joint inflammation has been considered as a critical contributor to RA pain, about 10-40% of RA patients continue to report joint pain even after joint inflammation is successfully controlled (1). In animal models of RA, hyperalgesia during the early inflammatory phase is attenuated by conventional therapeutics (nonsteroidal anti-inflammatory drugs, gabapentin, etanercept, the partial mu opioid receptor agonist buprenorphine), but hyperalgesia in the late postinflammatory phase is only modestly reduced by gabapentin, indicating that the pathways targeted by these drugs likely do not drive the chronification of arthritis pain (2,3).
Despite significant advances in understanding RA pain mechanisms in the inflammatory phase, in fact, little effort has been undertaken to define the drivers of postinflammatory arthritis pain. A number of studies suggest that peripheral inflammation causes long-term alterations in nociceptive pathways at both peripheral and central levels (4,5). In the spinal cord, astrocytes and microglia were activated in mice of both sexes after resolution of inflammation (4). Yet, inhibition of spinal glia activity reversed postinflammatory arthritis pain only in males but not in females (4). In addition, an inflammatory environment was observed within the DRG after resolution of joint inflammation, which drives continuous sensitization of nociceptors (6,7). At a molecular level, voltage-gated Ca 2+ channel (VGCC) subunit a2d, P2X3, and certain cytokine receptors were upregulated in DRG neurons in the postinflammatory phase of collagen antibody-induced arthritis (CAIA) (5). In addition, RA is characterized by the production of autoantibodies and the formation of IgG immune complexes (IgG-IC). The level of certain autoantibodies remains elevated in subgroups of RA patients even after successful reversal of joint inflammation (8). FcgRI (also called CD64) is the sole high-affinity receptor that can bind both monomeric and polymeric IgG (9). FcgRI is most famously expressed in various immune cells (e.g., macrophages and T cells) and serves as a critical player in the regulation of immunity (10,11). Yet, we and other groups recently discovered that FcgRI is also expressed in subsets of nociceptive DRG neurons (12,13). Moreover, our recent study showed that neuronal FcgRI mediated arthritis pain via an inflammatory cell-independent mechanism during the inflammatory phase in mouse models of RA (14). Still, whether sensory neuron expressed FcgRI contributes to postinflammatory arthritis pain has remained entirely unexplored.
In the mouse model of CAIA, mechanical hyperalgesia persists even after resolution of joint inflammation (2). Thus, CAIA in mice faithfully recapitulates certain aspects of medically controlled RA or RA in remission in humans. Because of this feature, the CAIA model can be used to identify and evaluate candidate mechanisms driving postinflammatory arthritis pain. In this study, we test the hypothesis that sensory neuron expressed FcgRI mediates persistent joint pain after arthritis remission.

Animals
Given the low incidence (~50%) of CAIA induction in BALB/c male mice, only BALB/c female mice that were 2 to 4 months old and 20 to 30 g body weight were used for this study (4). Animals were housed under a 14-hour light/10-hour dark cycle with ad libitum access to food and water. Breeders of global Fcgr1 knockout (Fcgr1 -/-) mice were provided by Sjef Verbeek (Leiden University Medical Center, Leiden, The Netherlands). Fcgr1 -/-/BALB/c mice were obtained by backcrossing Fcgr1 -/-/C57BL/6 mice at least five generations with BALB/c mice. Fcgr1 fl/fl mice were previously generated by our laboratory using CRISPR/Cas9 genome editing (14). Specific deletion of Fcgr1 expression in primary sensory neurons was achieved by mating Fcgr1 fl/fl mice with a PirtCre mouse line (gift of Dr. Xinzhong Dong; Johns Hopkins University). This resulting mouse line was then backcrossed to the BALB/c genetic background at least five generations. All animal experimental procedures were approved by the Institutional Animal Care and Use Committee of Johns Hopkins University School of Medicine and were in accordance with the guidelines provided by the NIH and the International Association for the Study of Pain.
Collagen Antibody-Induced Arthritis CAIA was induced in female mice on a BALB/c background by intraperitoneal (i.p.) injection of an anti-CII arthritogenic cocktail (1.5 mg/mouse, 150 µl, Chondrex Inc, Woodinville, WA) on day 0 followed by i.p. injection of lipopolysaccharide (LPS, 25 µg/mouse, 100 µl in saline, Chondrex Inc, Woodinville, WA) on day 5 as described previously (2). Mice that receive saline i.p. on days 0 and 5 serve as controls. Arthritis score was used as an index of inflammation and disease activity, and evaluated by an investigator blinded to treatments and genotypes. Arthritis score was assigned based on a scoring protocol in which each swollen or red toe was given 1 point. A red or swollen base knuckle was given 1 point, a swollen ankle and/or wrist was given 5 points. The maximum score for each paw was 15 points, resulting in a maximum possible score of 60 points per mouse (2).

Pain Behavioral Assessment
All behavioral measurements were performed on awake, agematched littermates (2-4 months) by experimenters blinded to genotypes and treatments. Primary mechanical hyperalgesia in ankle joints was measured by applying ascending forces to the mouse ankle with calibrated electronic blunt forceps (Bioseb, Pinellas Park, FL) while mice were restrained with a cloth (14). The cutoff force was set at 350 g to avoid joint damage. The mechanical threshold was defined as the force at which the mouse withdrew its hindlimb forcefully or vocalized (14). The mechanical threshold in the joint was averaged over three measurements obtained at intervals of at least 5 min. Secondary mechanical hyperalgesia in the glabrous skin of hind paws was evaluated using von Frey monofilaments of different forces (0.04, 0.07, 0.16, 0.4, and 1.0 g). The frequency of paw withdrawal responses to ten applications of each filament was counted. Secondary thermal hyperalgesia in the glabrous hind paw skin was assessed by measuring withdrawal latency to noxious heat stimuli delivered using a radiant heat source (IITC Life Science Inc., Woodland Hills, CA). The cutoff latency was set at 15 s. Heat response latencies were averaged over three measurements obtained at intervals of at least 3 min.
Quantitative Real-Time PCR RNA was extracted from L3-L5 DRGs, synovium and lumbar spinal cord of mice using a RNeasy Lipid Tissue Mini Kit (Qiagen, Germantown, MD) and RNA quantity was measured using a NanoDrop 1000 (Thermo Fisher Scientific, Waltham, MA). RNA was then reverse-transcribed to complimentary DNA using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) according to the manufacturer's protocol. For relative quantification of mRNA, real time PCR was performed on a QuantStudio 3 real-time PCR system (Applied Biosciences Corp., Beverly Hills, CA) using the PowerUp SYBR Green Master Mix (Applied Biosciences Corp., Beverly Hills, CA). Each sample was run in duplicate. The expression levels of the target genes were quantified relative to the level of b-actin (Actb) gene expression using the 2 -DDCT method. The sequence of primers used was as follows: siRNA Knockdown ON-TARGETplus mouse CD64 siRNA SMARTPool (cat# L-040932-01) and non-targeting control pool (NC; cat# D-001810-10) were purchased from Dharmacon. siRNA was dissolved in 5% glucose and mixed with the in vivo transfection reagent (In vivo-jetPEI; N/P = 6; Polyplus Transfection Inc, New York, NY) and incubated at room temperature for 15 min according to the manufacturer's instructions. A total of 2 µg of siRNA or NC in 10 µl volume was injected intrathecally (i.t.) into mice at L4/5 spinal levels once a day on days 49, 52 and 55 after the induction of CAIA. A valid spinal puncture and intrathecal delivery of the mixture was confirmed by a reflexive tail flick after needle entry into the subarachnoid space. Pain-related behaviors were assessed 72 hrs after each injection. At 72 hrs after the last injection, L3-5 DRGs and lumbar spinal cord was collected and subjected to qPCR to test the efficiency of Fcgr1 knockdown.

Immunohistochemistry and In Situ Hybridization
Anesthetized mice were transcardially perfused with phosphatebuffered saline, followed by 4% paraformaldehyde. L3-L5 DRGs were dissected, cryopreserved in 30% sucrose and cryosectioned at 12 µm. After blocking, sections were incubated with primary antibodies (Table S1) overnight at 4°C. Sections were then washed and incubated with complementary secondary antibodies (Table S1) at room temperature for 1 hr. ISH for Fcgr1 mRNA was performed as described previously (14). SP6 transcribed antisense and T7 transcribed sense control probes were synthesized from mouse Fcgr1 (NM_010186) cDNA clone (MR225268, OriGene) using one set of primers (forward, 5′-ATTTAGGTGACACTATAGAATCCTCAATGCCAAGT GACCC-3′; reverse, 5′-GCGTAATACGACTCACTATAG GGCGCCATCGCTTCTAACTTGC-3′). The specificity of these probes was validated in DRG sections of Fcgr1 -/mice in our previous study (14). The probes were then labeled using a digoxigenin (DIG) RNA labeling Kit (Roche Diagnostics Corp., Indianapolis, IN). After pre-hybridization, sections were hybridized with probes (2 ng/µl) at 60°C overnight and washed with 0.2x SSC at 62°C. To combine ISH with IHC, tissue sections undergoing ISH were incubated with sheep anti-DIG antibody ( Table S1) and subjected to the standard IHC protocol as above. Images were captured using a confocal microscope (Nikon A1+; Nikon Metrology Inc, Brighton, MI). All images were analyzed using NIS elements in a blinded manner. At least three sections of L3-L5 DRGs per animal were analyzed for the quantification of DRG neuron subpopulations. To quantify Fcgr1 expression, cell diameters were measured in a given DRG section using the polyline tool. Quantification of total neuron numbers was performed by counting NeuN positive neurons with clear nuclei. The proportion of Fcgr1 positive neurons in each cell size population was calculated. For siRNA experiments, Fcgr1 expression was quantified by measuring mean signal fluorescence intensity per neuron. For analysis of immune cell infiltration in DRG, 3-4 sections per animal of L3-L5 DRGs were imaged and the number or fluorescent intensity of cells immunopositive for each marker (CD3, Ly6C/G, and CD68) per unit area was counted. For each marker within a group, all parameters for image acquisition and analysis were kept constant across images.

Histological Analysis
Total ankle joints were harvested from mice on day 56 after CAIA induction. Joints were decalcified in 0.5 mM ethylenediamine tetraacetic acid at 4°C for 3 weeks, then embedded in paraffin, and 5-mm coronal sections were cut. Sections were stained with Safranin-O and Fast Green for histological analysis. Destruction of the articular cartilage layers at the tibia and talus were scored on a scale from 0 (no erosion) to 3 (complete erosion of the calcified cartilage layer) (15). A total of 3 sections per joint from various depths were scored and results from all locations were averaged.

Statistical Analysis
Data are presented as means ± SEM. A two-tailed Student's t test was used to test the significance of differences between two groups. Comparisons for multiple groups or multiple time points were carried out using a two-way ANOVA for random measures or repeated measures followed by Bonferroni correction. P value less than 0.05 was considered significant. Sample sizes were chosen based on our pilot experiments, field conventions to accurately detect statistical significance, considering ethical animal use, experiment design, resource availability, and technical feasibility. Group size and statistical tests used for each comparison are noted in each figure legend. All statistical analyses were performed on GraphPad Prism 7.0 software.

Global Deletion of Fcgr1 Attenuates Arthrits Pain and Joint Swelling in Female Mice With CAIA
Although our recent study showed that FcgRI mediated arthritis pain in several inflammatory arthritis models (14,16), much less is known about its potential roles in arthritis pain and disease activity in the CAIA model. Considering that the CAIA model allows us to define postinflammatory mechanisms of arthritis pain, we compared pain related behaviors and arthritis scores between wildtype (Fcgr1 +/+ ) and Fcgr1 -/mice in both inflammatory and postinflammatory phases of CAIA. Consistent with previous studies (2,5), Fcgr1 +/+ mice subjected to CAIA developed obvious joint swelling and redness (indicated by arthritis score) on day 7, peaking around day 14 and gradually resolving by around day 42 ( Figures 1A, B). In contrast, mechanical hyperalgesia in both the ankle ( Figure 1C model (from day 6 to day 56; Figure 1). We refer to the period of hyperalgesia with visible signs of joint swelling as the 'inflammatory phase' (days  and that without obvious joint swelling as the 'postinflammatory phase' (days 42-56). Arthritis scores in Fcgr1 -/mice were markedly attenuated in the inflammatory phase compared to wildtype controls ( Figure 1B). Strikingly, despite apparently normal ankle and hind paw mechanical sensitivity prior to the induction of arthritis, global Fcgr1 -/mice also exhibited less mechanical hyperalgesia in the ankle and hind paw, compared with Fcgr1 +/+ littermates, throughout both the inflammatory and postinflammatory phases of CAIA ( Figures 1C-E).
FcgRI has been implicated in cartilage and bone damage in antigen-induced arthritis (AIA) and collagen II-induced arthritis (CIA) models (12,13). To determine whether the apparent antihyperalgesic effects of Fcgr1 knockout were attributable to a possible reduction in CAIA-induced cartilage destruction, we performed Safranin-O/fast green staining on mouse ankle joint sections ( Figure 2). Safranin-O/fast green staining analysis revealed obvious cartilage destruction in both genotypes on day 56 after CAIA induction ( Figure 2A). However, the extent of cartilage damage was not significantly different between genotypes in the CAIA model ( Figure 2B).

Conditional Knockout of Fcgr1 in Primary Sensory Neurons Alleviates Arthritis Pain in the CAIA Model
Given that neuronal FcgRI can regulate the excitability of primary sensory neurons (13,14), we next investigated whether neuronally expressed FcgRI contributes to postinflammatory arthritis pain in No obvious differences in the histology score for cartilage destruction were observed at day 56 after CAIA between genotypes. n = 4 -6 mice/group; *p < 0.05; **p < 0.01 versus control (Ctrl), two-way ANOVA followed by Bonferroni correction. n.s., not significant. the CAIA model. Fcgr1 fl/fl mice were crossed with the PirtCre line to selectively omit FcgRI expression from primary sensory neurons ( Figure 3A). The specific loss of FcgRI expression in primary sensory neurons was validated in our recent studies (14). In both inflammatory and postinflammatory phases of CAIA, PirtCre-Fcgr1 fl/fl mice displayed less mechanical hyperalgesia in the ankle and hind paw, and less heat hyperalgesia in the hind paw, compared to PirtCre negative controls (Figures 3B-E). However, unlike our findings in global Fcgr1 -/mice, sensory neuron-specific Fcgr1 deletion resulted in no significant differences in arthritis scores, compared to PirtCre negative controls, in the inflammatory phase of CAIA ( Figure 3F).
Interestingly, CAIA caused a significant downregulation of Pirt mRNA expression in the DRG on day 56 after challenge (Supplementary Figure 1). However, disruption of the Fcgr1 fl/fl alleles would have occurred earlier in life, and therefore should not be affected by this phenomenon. inflammation was sufficient to alleviate persistent pain in the CAIA model. To test this possibility, we delivered CD64 siRNA or a non-targeting siRNA negative control by i.t. injections into wildtype mice on days 49, 52 and 55 after CAIA induction, a time period after joint inflammation was resolved. Mechanical hyperalgesia was assessed 72 hrs after each injection ( Figure 4A). CD64 siRNA significantly increased mechanical threshold in the ankle of mice on all testing days, compared to the negative control ( Figure 4B). Similarly, CAIA mice treated with CD64 siRNA exhibited less secondary mechanical hyperalgesia in the hind paws than those treated with the negative control ( Figure 4C, D). In addition, we harvested L3-L5 DRGs after behavioral testing on day 58 of CAIA induction and found a significant decrease of Fcgr1 mRNA expression in the DRG but not in spinal cord after CD64 siRNA injection compared to negative control injection ( Figure 4E). ISH assay confirmed a significant reduction in Fcgr1 mRNA signal in DRG neurons in CD64 siRNA-treated mice compared to that in negative control-treated animals ( Figures 4F, G).

Fcgr1 mRNA Expression in Mouse DRG Is Upregulated in the Late Phase of CAIA
Our previous studies revealed that Fcgr1 mRNA expression was upregulated in the DRG during inflammation in both AIA and CIA models (14,16). However, whether such changes occur in the CAIA model remained unknown. Therefore, we performed qPCR on the DRG from wildtype mice to assay for the alterations of Fcgr1 mRNA expression on days 15 (inflammatory phase) and 56 (postinflammatory phase) after CAIA induction. On day 15 after immunization, the Fcgr1 mRNA expression level was significantly upregulated in the synovium and spinal cord but not in the DRG of mice with CAIA compared to vehicle treated animals ( Figure 5A). By contrast, on day 56 after CAIA induction, upregulated levels of Fcgr1 mRNA expression were observed in the DRG but not the synovium and spinal cord in the CAIA group compared to control mice ( Figure 5B). To further define the changes in Fcgr1 mRNA expression profile in the DRG at late stages of CAIA, and to determine whether these changes occurred within sensory neurons, we carried out ISH and IHC assays on the DRG on day 56 after CAIA induction. We revealed that a larger proportion of DRG neurons expressed Fcgr1 mRNA in CAIA mice compared control animals ( Figures 6A, B).
Furthermore, these alterations were statistically significant in medium and large diameter DRG neurons but not in small diameter neurons. (Figure 6C). The specificity of Fcgr1 ISH probes was validated in our recent studies (data not shown) (14).
In agreement with previous studies (5), the number of CGRP + neurons were not altered in the DRG at the late phase of CAIA ( Figure 6D). Among Fcgr1 mRNA expressing neurons, we also observed no significant alterations in the percentage of CGRP + neurons during the postinflammatory phase of CAIA ( Figure 6E). Surprisingly, however, the proportion of CGRP + DRG neurons that also expressed Fcgr1 mRNA was significantly reduced on day 56 after CAIA induction ( Figure 6F), suggesting that an increase in Fcgr1 mRNA expression may exclusively occur in non-CGRP expressing neurons.

FcgRI Does Not Contribute to Immune Cell Infiltration Within the DRG in the Postinflammatory Phase of CAIA
Considering that immune cell infiltration within the DRG has been implicated in postinflammatory arthritis pain (6, 7), we next asked whether FcgRI is necessary for immune cell infiltration within the DRG in the postinflammatory phase of CAIA. On day 56 after CAIA induction, IHC analysis showed that markers for T cells (CD3; Figures 7A, B), and neutrophils and monocytes (Ly6C/G; Figures 7A, C) were increased in the DRG whereas the marker for macrophages (CD68;  inflammatory mediators were observed between genotypes ( Figures 8A-D).

DISCUSSION
Arthritis pain often persists in a significant subpopulation of RA patients even after joint inflammation has resolved. Although central mechanisms of postinflammatoy arthritis pain are well defined (4,5), the studies on peripheral mechanisms are quite sparse. In the present study, we have focused on the late phase of the CAIA model when joint inflammation has resolved but arthritis pain persists. Consistent with prior findings in mouse AIA and K/BxN serum transfer models (6, 7), we have identified an inflammatory state within DRG after arthritis remission. We also provide new evidence for the involvement of sensory neuron expressed FcgRI in mechanical hypersensitivity in both inflammatory and postinflammatory phases of CAIA. FcgRI is widely expressed in immune cells (e.g., macrophages and T cells) and plays a key role in the pathogenesis of RA (17). In RA patients, FcgRI is expressed de novo in the synovium (18) and its expression level in neutrophils, monocytes and synovial Quantification showed an increase in the number of CD3 + (B; n = 8-10 mice/group) and Ly6C/G + (C; n = 6-10 mice/group) cells per unit area but not the mean fluorescent intensity of CD68 + (D; n = 6-7 mice/group) per unit area in the DRG of either genotype on day 56 after CAIA induction. Yet, no obvious differences in CD3 + or Ly6C/G + cell infiltration were observed between genotypes. *p < 0.05, **p < 0.01 versus Ctrl, two-way ANOVA followed by Bonferroni post hoc test. n.s., not significant. macrophages is upregulated (19)(20)(21). The pathogenicity of FcgRI has been confirmed in animal models of AIA and CIA (21)(22)(23). Our previous work has demonstrated the contribution of FcgRI to arthritis pain in several murine models of inflammatory arthritis (14,16). We and other groups have reported that FcgRI is also present in primary sensory neurons besides immune cells (13,24,25). Moreover, we showed that neuronally expressed FcgRI mediated the sensitization of joint nociceptors and arthritis pain during inflammation in animal models of RA through a mechanism independent of inflammation (14). The present study extended previous studies by showing that FcgRI signaling, particularly in primary sensory neurons, contributes to persistent arthritis pain observed after resolution of joint inflammation. First, global deletion of Fcgr1 alleviated mechanical hyperalgesia in the ankle and hind paws of mice in both inflammatory and postinflammatory phases of the CAIA model. Since cold hyperalgesia occurred at the same time as joint inflammation but did not persist beyond the signs of arthritis (2, 3), we did not evaluate those behaviors in this study. Second, the contribution of neuronal FcgRI to postinflammatory pain was validated by use of conditional Fcgr1 knockout mice, in which Fcgr1 is specifically deleted in primary sensory neurons. Indeed, conditional deletion of Fcgr1 produced similar antihyperalgesic effects as global Fcgr1 knockout. However, unlike our findings in global Fcgr1 -/mice, our conditional knockout data suggest that neuronal FcgRI might not be involved in joint swelling in the CAIA model, at least as reflected by arthritis score. Further work is warranted to define any contributions of neuronal FcgRI to less overt joint inflammatory processes in the setting of RA, using more highly sensitive histological and biochemical assays, including flow cytometry and hematoxylin and eosin stain (H&E) staining. Lastly, qPCR revealed that the upregulated Fcgr1 mRNA expression differed across anatomical loci over the course of CAIA. During the inflammatory phase of CAIA, we observed an upregulation of Fcgr1 mRNA expression in the synovium and spinal cord but not in the DRG. By contrast, upregulation of Fcgr1 mRNA expression was observed only in the DRG but not in the synovium and spinal cord after arthritis remission in the CAIA model. ISH analysis further supported the upregulation of F c g r 1 mRN A e x pr es si on i n DRG n eu ro n s i n t h e postinflammatory phase of CAIA. Moreover, these changes mainly occurred in medium and large diameter DRG neurons. Unlike during inflammatory stages of AIA, where there is an increase in the proportion of FcgRI expressing neurons that also express CGRP (14), in the late phase of the CAIA model, we did not observe such increases. In fact, the overall number of CGRP + cells also expressing FcgRI was reduced at this time point, suggesting that an increase in Fcgr1 mRNA expression may occur exclusively in non-CGRP expressing neurons. Considering that neuronal FcgRI is important to the regulation of the excitability of primary sensory neurons (14), it is possible that the observed upregulation of Fcgr1 mRNA expression in DRG neurons may contribute to continuous peripheral sensitization and postinflammatory arthritis pain. This notion was further supported by use of selective CD64 siRNA. Intrathecal delivery of CD64 siRNA into CAIA mice in the postinflammatory phase downregulated Fcgr1 mRNA expression in the DRG neurons and alleviated persistent arthritis pain. This is consistent with the possibility that the antihyperalgesic effect of CD64 siRNA is mediated, at least in part, by sensory neuron expressed FcgRI. Since FcgRI is also expressed in immune cells (17), however, we cannot rule out non-neuronal effects of CD64 siRNA in this experiment. In addition, considering that CD64 siRNA was administered through intrathecal delivery,we also cannot completely exclude a central action of CD64 siRNA. Cartilage and bone damage appears to be a key driver of arthritis pain in the absence of visible joint swelling (i.e., osteoarthritis pain) (26). Clinical studies suggest that the destruction of cartilage and bone persists in RA patients after resolution of joint inflammation (27,28). FcgRI has been implicated in cartilage destruction and bone erosion during the inflammatory phase in animal models of CIA and AIA (22,23). Given that FcgRI is a key immune receptor expressed in a variety of immune cells (29), FcgRI might indirectly regulate bone erosion via the recruitment and activation of immune cells during RA (30). In addition, FcgRI is also expressed in preosteoclasts and osteoclasts (31,32). Direct activation of FcgRI in these cells increases both osteoclast differentiation and function (31,32). Consistent with clinical studies, the present study showed obvious cartilage destruction in the postinflammatory phase of the CAIA model. However, FcgRI is not necessary for these anatomical changes since genetic deletion of Fcgr1 did not alter the degree of cartilage damage following arthritis remission.
Neuroinflammation within the DRG has been implicated in the development and maintenance of arthritis pain during the inflammatory phase (33,34). Recent studies revealed persistent immune cell infiltration in the DRG after joint inflammation was resolved in AIA and K/BxN serum transfer models (6,7). Moreover, increased numbers of activated resident macrophages in the DRG released TNF, which may drive continuous sensitization of primary sensory neurons and the maintenance of persistent joint pain following the remission of AIA (7). However, in the CAIA model, we did not observe obvious macrophage infiltration within the DRG after joint inflammation was resolved. By contrast, our results revealed infiltration of T cells and neutrophil and/or monocytes and elevated expression level of a variety of proinflammatory chemokines and cytokines in the DRG after arthritis remission. Although FcgRI is critical to the regulation of immune responses (11), the present study found that genetic deletion of Fcgr1 had no significant effects on immune cell infiltration or the extent of the upregulation of inflammatory mediators in the DRG after resolution of joint inflammation. Thus, it is unlikely that FcgRI signaling acts as a major upstream mechanism for immune cell infiltration and release of proinflammatory mediators in the DRG after arthritis remission. By contrast, FcgRI may function as a downstream target of proinflammatory mediators. These inflammatory mediators within the DRG might upregulate FcgRI expression in primary sensory neurons and/or enhance the affinity of neuronal FcgRI for IgG-IC, leading to peripheral sensitization and persistent arthritis pain after arthritis remission (35,36).
The endogenous activators of FcgRI during postinflammatory arthritis remain unclear. Conventional wisdom has held that autoantibodies revert to negative titers after RA patients achieve disease-modifying antirheumatic drug -free remission (37,38). However, a recent clinical study revealed no obvious association between disease remission and reversion to autoantibody negativity (8). Anti-citrullinated protein antibodies (ACPA) IgG levels were not significantly altered in a subpopulation of RA patients in remission (8). ACPA IgG together with their citrullinated antigens might form IgG-IC to activate joint nociceptors to trigger arthritis pain via their interaction with neuronally expressed FcgRI (39). Considering that FcgRI plays an important role in the regulation of substance P and CGRP release from DRG neurons induced by IgG-IC (12,24), neuronal FcgRI may contribute to postinflammatory arthritis pain through peripheral and central mechanisms by modulating the release of these pain mediators. Further investigations are needed to test this possibility.
This study has some limitations. We did not evaluate painrelated behaviors in male mice due to the low incidence of CAIA induction in males [4]. Considering that sexual dimorphism was observed in central mechanisms of postinflammatory arthritis pain [4], further investigations are warranted to assay for potential sex differences in the role of peripheral FcgRI in persistent joint pain after arthritis remission. Since Cre rather than inducible CreER mice were used in this study, the observed diminished mechanical pain hypersensitivity in Fcgr1 knockout mice after arthritis remission may be secondary to antihyperalgesic effects of FcgRI in the inflammatory phase of arthritis. Further investigation is required to determine whether inducible deletion of the Fcgr1 gene in primary sensory neurons after arthritis remission attenuates persistent joint pain.
In conclusion, our results provide novel evidence that FcgRI expressed in primary sensory neurons contributes to the maintenance of postinflammatory arthritis pain. Although FcgRI is not necessary for the development of the late state of neuroinflammation in the DRG, the prolonged neuroinflammation within the DRG after arthritis remission might upregulate FcgRI signaling in primary sensory neurons, leading to peripheral sensitization and persistent arthritis pain. We suggest that neuronal FcgRI may represent a new candidate therapeutic target for the treatment of persistent arthritis pain in RA patients with low disease activity and/or in remission.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.