Sensitization of knee-innervating sensory neurons by tumor necrosis factor-α activated fibroblast-like synoviocytes: an in vitro, co-culture model of inflammatory pain

Background Pain is a principle contributor to the global burden of arthritis with peripheral sensitization being a major cause of arthritis-related pain. Within the knee joint, distal endings of dorsal root ganglion neurons (knee neurons) interact with fibroblast-like synoviocytes (FLS) and the inflammatory mediators they secrete, which are thought to promote peripheral sensitization. Therefore, we investigated the communication between knee neurons and FLS in a co-culture system. Methods Inflammation in FLS (isolated from mouse patella) was induced by stimulation with tumor necrosis factor-α (TNF-FLS). Expression of relevant genes, secretion of cytokines and functional acid response was assessed using polymerase chain reactions, inflammatory antibody array blots and Ca2+ imaging respectively. Electrophysiology was performed on dissociated knee neurons in mono-culture, co-culture with control FLS or TNF-FLS, or supernatant derived from TNF-FLS cultures to determine electrical excitability and sensitivity to transient receptor potential (TRP) agonists. Two group comparisons were conducted using Student’s t-tests (distributed variable) or chi-sq tests (categorical variable). ANOVA followed by Tukey’s post-hoc test was performed for comparing more than two groups. Results Compared to control, TNF-FLS showed increased expression of interleukin-6 (Il-6, p < 0.01, unpaired t-test) and Il-1r1 (p < 0.05, unpaired t-test), and enhanced cytokine secretion. FLS also responded to acidic stimuli and the percentage of TNF-FLS acid responders increased in the pH range 6.0-5.0 (p < 0.001, chi-sq test). Co-culture of DRG neurons with TNF-FLS or supernatant derived from TNF-FLS depolarized the resting membrane potential (p < 0.01, ANOVA), increased spontaneous action potential firing (p < 0.05, ANOVA) and enhanced TRPV1 function (p < 0.05, ANOVA) of knee neurons consistent with a role for FLS in the sensitization of pain sensing nerves in arthritis. Conclusions TNF-FLS have a pro-nociceptive phenotype, displaying enhanced acid sensitivity and increased secretion of pro-inflammatory cytokines. Co-culture with TNF-FLS or supernatant from TNF-FLS, induces hyperexcitability of knee neurons and enhances TRPV1 function. Data from this study demonstrate the ability of FLS activated by TNF-α to promote neuronal sensitization, results that highlight the importance of both non-neuronal and neuronal cells to the development of pain in arthritis.


Freund's adjuvant, knee
Experiments conducted on SW982, a human tumor derived synoviocyte cell line, and primary FLS 126 from an RA patient, have demonstrated that a Ca 2+ influx occurs upon application of the TRPV1 127 agonist capsaicin in an acidic (pH 7.1) environment and upon application of the TRPA1 and TRPM8 agonist icilin (18). In addition to Trpv1, the human FLS cell lines SW982 and MH7A also of co-culture studies to understand neuron-synoviocyte communication within the arthritic joint.

154
However, these studies did not functionally assess modulation of DRG neuron excitability, which 155 is a key mechanism of peripheral sensitization and hence pain. We recently showed that human 156 osteoarthritic (OA) synovial fluid (a lubricating fluid largely secreted by FLS (1)) can cause female mice were used since in humans, females are at a higher risk for inflammatory pain (34).
bedding materials, a red shelter and enrichment. They were on a 12-hour/light dark cycle with food  Under anesthesia (100 mg/kg ketamine and 10 mg/kg xylazine, intra-peritoneally) mice were 187 injected intra-articularly through the patellar tendon into each knee with the retrograde tracer, fast 188 blue (FB, 1.5 l 2% in 0.9% saline, Polysciences) or into the left knee with 7.5 l CFA (10 mg/ml, 189 Chondrex). Vernier's calipers were used to measure knee width (as before (11)) pre-and 24-hour post-CFA injection.
Isolation and culture of FLS 24-hour after CFA injection into the knee, mice were killed by cervical dislocation and 195 decapitation. Knee joints were exposed by removing the skin, then the quadriceps muscles were 196 resected in the middle and pulled distally to expose the patellae. Patellae were then collected by 197 cutting through the surrounding ligaments, as described before (35), in phosphate-buffered saline (PBS) and then transferred into one well of a 24-well plate with FLS culture media containing: Mini Kit (Qiagen). 500 ng of the extracted RNA was used to synthesize cDNA using a High 225 Capacity cDNA RT kit (Applied Biosystems) following the manufacturer's guidelines, using a 226 T100 Thermal Cycler (Bio-Rad). The resultant cDNA was diluted to a 1:5 ratio with nuclease free 227 water and quantitative PCR (qPCR) was performed using a StepOnePlus Real Time PCR system,

228
following the manufacturer's guidelines on settings (Applied Biosystems) using TaqMan

257
Densitometry of the spots in the array membranes was performed using ImageJ software (NIH).
Briefly, the mean gray value of each spot was measured from all membranes using the same circular region of interest. The spots of interests were then background (average of all negative 260 control spots) subtracted and normalized to the positive control spots of the reference membrane 261 (control FLS media). Fold change was obtained by dividing normalized intensities of the 262 membrane of interest and the control membrane, analyte-by-analyte.  For co-culture studies, FLS were plated onto MatTek dishes and cultured for 24-hours with FLS plates, then DRG neurons were isolated as described above and plated on top of the FLS. Coculture plates were then kept in DRG culture medium for up to 24-hours for electrophysiogical    software (Wavemetrics) as described before (11) and shown in Figure 3D

381
Concurrently, to establish the pro-inflammatory phenotype of FLS, we used RT-qPCR to 382 determine the expression of the inflammatory genes Il-6, Il-1r1 and Cox-2, as well as the 383 constitutively expressed gene Cox-1. When control FLS (P5) were stimulated with 10 ng/ml TNF-

392
we did not find any changes in expression levels of the genes between Contra and Ipsi FLS across

537
However, a genome-wide study on DNA-methylation has shown that important differences exist 538 between knee and hip FLS, including of genes involved in IL-6 signaling (45). By using a cell-539 outgrowth method to culture mouse FLS, as previously described in humans (46) and rats (6), we 540 have avoided the biological ambiguity introduced by joint-to-joint variability. Using these FLS,

541
we investigated the expression of inflammatory genes Il-6, Il-1r1 and Cox-2, all of which have IL-1β increases Cox-2 and Il-6 expression (46); increased Il-6 expression is seen in FLS derived (6). Here we observed that neither the expression level of Il-6, Il-1r1, Cox-2, nor the level of 547 secreted cytokines were upregulated in FLS derived from CFA-injected knee. This is possibly 548 because the model used here is too brief to influence FLS gene expression. Indeed, evidence to 549 support this is that FLS derived from rats with longer (3-28 days) AIA-induced knee inflammation

555
We reasoned that FLS need more direct and prolonged exposure to inflammatory mediators to 556 show a robust inflammatory phenotype. TNF- is one such cytokine that is locally upregulated 557 within 3-hours of intra-plantar CFA injection in mice (12). It is also present in high concentrations

583
In order to understand the effector role of FLS in driving nociception through peripheral 584 sensitization, we set up a co-culture system combining FLS and knee neurons. Studying co-culture 585 of rat FLS and DRG neurons, von Banchet et al showed using immunohistochemistry that 586 bradykinin 2 receptor labelling (but not that of neurokinin 1 or TRPV1) was increased when DRG 587 neurons were cultured with healthy rat FLS (6), i.e. co-culture of DRG neurons with FLS can alter expression of genes associated with nociception. Therefore, we first verified that knee neurons in co-culture with control FLS do not show dysregulation of excitability or TRP agonist response.

590
Then we asked whether TNF-FLS modulates knee neuron function and found using whole-cell 591 patch clamp that 23% and 16% of knee neurons in neuron/TNF-FLS co-culture and neuron/TNF-592 FLS media (inflamed conditions) respectively evoked spontaneous AP compared to 6% in 593 neuron/control FLS and 0% in neuron mono-culture (healthy conditions). This suggests that TNF-  supplying the knee joint are exposed to a pro-inflammatory environment due to cell-contact and 633 soluble mediator release from immune cells. One such key effector in the joint is FLS which can respond to inflammatory mediators and also themselves secrete inflammatory soluble factors (2). a co-culture system. Therefore, we establish a direct inflammation-pain axis through FLS and DRG 639 neurons. This system can be adapted to investigate peripheral sensitization using FLS activated by