Evidence for purinergic signaling pathways in human rheumatoid and osteoarthritis synovial tissue

Abstract

Background

Inflammatory joint destruction releases intracellular components like ATP and triggers an acidic microenvironment. Detection of ATP and ischemia by sensory neurons leads to pain sensation, but little is known about the presence of purinoceptors and regulating factors in synovial tissue. The current study investigates the presence of ATP-receptor P2X5, acid-sensing ion channel ASIC3, and ATP degrading ectonucleotidases NTPDase1 (CD39) and NTPDase2 in human rheumatoid arthritis (RA) and osteoarthritis (OA) synovial tissue.

Methods

H&E and immunofluorescent staining on 8 rheumatoid arthritis (2 male, 6 female) and 8 osteoarthritis specimens (1 male, 7 female) were analyzed.

Results

Synovitis score was significantly higher in RA compared to OA (p=0.0002*, 7.38 ± 1.30 versus 2.75 ± 1.28 points, mean±SD). ASIC3 and P2X5 as well as NTPDase1 and NTPDase2 show a very high, significant correlation in synovial tissue of RA lining and sublining layer (r>93%, double staining in 11.2-24.6% of all cells, single staining in 0.5-10%). A high correlation for NTPDase1 and NTPDase2 was observed also in OA sublining layer (r>94%, double staining in 15.2-37.6% of cells, 0.8-8.1% single staining).

Conclusions

ASIC3 and P2X5 together with NTPDase1 and NTPDase2 indicate a tightly regulated purinergic microenvironment in RA and OA synovial tissue. Data suggests that nucleotide turnover, including ATP breakdown and subsequent adenosine generation, modulates inflammatory and nociceptive processes and purinergic signaling pathways in synovitis.

Keywords

ectoATPase CD39 antigen acid sensing ion channels arthritis inflammation rheumatoid osteoarthritis pain purinergic signal transduction synovitis

Introduction

Rheumatoid arthritis (RA) and osteoarthritis (OA) are characterized by synovitis, joint destruction and painful joint movement. Synovial tissue fibroblasts and macrophages are significantly involved in these destructive processes.^1,2 Purinergic signaling may play an important role in regulating local synovial tissue cell functions and pain sensation. In RA autoimmune inflammation increases cytokine release, creates an acidic microenvironment, reduces cell-cell contacts, and raises mechanical pressure due to swelling. These conditions can trigger the release or leakage of Adenosine Triphosphate (ATP) and other intracellular molecules.^2,3

In OA, degenerative and mechanical factors are crucial for cartilage and joint destruction while synovial tissue cell interactions remain not fully understood.^4,5 Chemical stimulation and increased membrane permeability during inflammation can further result in hypoxia and more secretion of inflammatory mediators and cytosolic components.

Synovial tissue can control the release of nucleotides like ATP via different channels, exocytosis or pores, and uncontrolled release may emerge during cell damage or death. The release may be triggered by different noxes such as mechanical stress, chemical changes, direct or indirect receptor activation or apoptosis.^6–10

Among these molecules, ATP may be particularly important for pain sensation and intercellular communication via purinergic signaling. It facilitates communication between chondrocytes and synovial cells.^11–15 ATP is directly involved in prostaglandin release, endothelial function regulation, blood flow regulation, and it facilitates purinergic pain signaling in the peripheral nervous system.^3,16–25

During muscle ischemia, ATP mediates ischemic pain by enhancing the sensitization of acid-sensing ion channel ASIC3 through P2X5 receptors.^26,27 The two receptors have been found assembled on the membranes of sensory neurons of inflamed mice and rat knee joints innervating the synovium and the muscles, and they were found on type B synoviocytes.^26–28 The presence of both receptors on human synovial cells has yet to be investigated.

Altered ATP levels are observed in inflamed tissue and various immunological diseases.²⁹ It is assumed that extracellular ATP concentrations temporarily increase significantly during inflammatory conditions, although its rapid hydrolysis by local ectonucleotidases makes detection difficult.³⁰ Purinoceptor modulation by regulating extracellular nucleotide concentrations is achieved by nucleotide-degrading enzymes, including ectonucleotidases like NTPDase1 and NTPDase2.^31,32

NTPDase1 (CD39) hydrolyzes ATP and ADP, it is found on the cell surface of activated B and T-lymphocytes and natural killer cells, suggesting it protects these cells from the cytotoxic effects of high extracellular ATP levels during inflammation.

In inflamed milieus, extracellular ATP is rapidly metabolized via CD39 (NTPDase1) to ADP/AMP and via CD73 (NTPDase2) to adenosine, with adenosine exerting well-established anti-inflammatory actions.³³

Increased activity of NTPDase1 has been observed in circulating lymphocytes in RA.^34–36 NTPDase1 positive regulatory T-cells are also known to play a crucial role in autoimmune inflammation.³⁷ NTPDase2 specifically hydrolyzes ATP and was shown to be important in the central and peripheral nervous system during peripheral neuron activation. It has not yet been investigated in synovial tissue.

According to the above considerations, the aim of the current study was to histologically examine evidence for purinergic signaling in inflamed human synovial tissue by investigating the presence and distribution of P2X5 and ASIC3 receptors as well as NTPDase1 and NTPDase2 and compare RA and OA tissues. It is hypothesized that RA shows higher inflammation in synovial tissue and that RA presents more cells with these receptors and enzymes than OA tissue.

Materials and methods

Patients and specimen

Synovial tissue remnants of eight RA patients and eight OA patients who had elective knee arthroplasty at the Orthopedic Department of the LMU and the Orthopedic Department of the Rheumazentrum Oberammergau were collected and analyzed. Knee arthroplasty was performed because patients experienced severe pain, limited range of motion, and a high-grade radiological joint destruction. RA was diagnosed years before surgery, using the American College of Rheumatology criteria based on Arnett´s revised criteria.^38,39 OA was diagnosed radiologically and clinically prior to surgery. Patients informed consent to was obtained, and the study was approved by the LMU´s ethics committee. Basic patient data is provided in Table 1. The study was conducted between 2013 and 2016 in accordance with the Declaration of Helsinki (1975), as revised in 2013. Samples were collected in 2013. The study has followed the relevant EQUATOR Network guideline STROBE.

Table 1.

Patient information.

Characteristics		Rheumatoid arthritis			Osteoarthritis
Male/Female/total	n	2	6	8	1	7	8
Age	years mean±SD	59.6	±	11.3	65.4	±	13.5
Duration of disease	years mean±SD	11.5	±	8.2	2.3	±	1.3
Steroids/DMARDS/NSAIDs	n	7	6	1	0	1	7
CRP level	mg/dL mean±SD	4.3	±	4.6	0.6	±	0.4
Leucocyte count	cells/dL mean±SD	9.4	±	2.2	7.7	±	1.5
Hemoglobin	mg/dL mean±SD	12.7	±	1.5	13.2	±	1.8
Synovitis score* (p=0,0002)	Points	7.38	±	1.30	2.75	±	1.28
Cells per ROI Lining Layer	number mean±SD	65	±	36	54	±	17
Cells per ROI Subsynovial Layer	number mean±SD	238	±	109	167	±	41
Subsynovial endothel cells	in %	12.4	±	12.1	20.6	±	16.0

Sterile tissue samples were collected in the operating theatre, placed in sterile medium for transport, and within a few hours, rinsed and diced into 1x1 cm pieces after removing fatty tissue. The patient identities were anonymized by assigning random numbers. Samples were embedded in Tissue-Tek® (Tissue-Tek® O.C.T.™ Compound) ensuring the synovial lining layer was positioned sideways, and snap-frozen en bloc. Using a cryostat, frozen sections of 7 µm were cut, with two sections mounted on each glass slide. The cryosections on slides were dried at room temperature, wrapped in aluminum foil, and stored at -20°C.

Hematoxylin-eosin and double immunofluorescence staining

Frozen sections were stained with hematoxylin-eosin (H&E) to pinpoint regions of interest, and Krenn´s standard synovitis score was used.⁴⁰ Double immunofluorescent staining assessed cells using specific antibodies. The Combitec Slide Stainer 4009 by Vogel facilitated this process.

Primary and secondary antibodies were tested using standard protocols. Frozen sections were incubated in 4°C paraformaldehyde (PFA) for 15 minutes, then washed three times for 5 minutes in washing buffer at room temperature. Next, sections were blocked with 5% bovine serum albumin in PBS for 30 minutes. After washing, sections were incubated overnight at 4°C with the first primary antibody diluted in PBS with 5% bovine serum albumin in a humid chamber. The following day, sections were washed and incubated with the first secondary antibody diluted in PBS for 30 minutes in a darkened room. After washing and blocking, sections were exposed to the second primary antibody for 1 hour at room temperature. They were then treated with 0.1% Triton X 100 for 15 minutes. Following another wash, sections were incubated with the secondary antibody for 30 minutes. Nuclei were stained with DAPI in PBS or a DAPI-containing mounting medium before sections were covered with glass, dried, and stored at 4°C in the dark until they were viewed and photographed the next day. Antibody validation was confirmed using standard tissue samples as suggested by the datasheets; negative controls omitted primary antibodies, and isotype controls were included where applicable.

Primary antibodies

P2X5 antibody rabbit-anti-human polyclonal IgG (1:200, Prestige Antibodies, Sigma-Aldrich¬¬, Munich, Germany)

HPA021948

ASIC3 antibody goat-anti-human polyclonal (1:200, Santa Cruz Biotechnology, USA), sc-21845

ENTPD1 antibody mouse-anti-human IgG (1:200, antibodies-online GmbH Aachen, Germany), ABIN1009103

ENTPD2 antibody rabbit-anti-human polyclonal IgG (1:50, Prestige Antibodies, Sigma-Aldrich, Munich, Germany), HPA017676

CD3 antibody (mouse) (Biorbyt Ltd. Cambridgeshire, UK), orb223668

CD14 antibody rabbit-anti-human polyclonal IgG (1:200, Prestige Antibodies, ¬¬Munich, Germany)

CD34 antibody (mouse) (Biorbyt Ltd. Cambridgeshire, UK), orb98267

CD68 antibody (mouse) (Biorbyt Ltd. Cambridgeshire, UK), orb223629

CD90 (Thy1) antibody mouse-anti-human (FITC conjugated, 1:200, antibodies-online GmbH Aachen, Germany)

Vimentin antibody (rabbit) (GeneTex INC. Irvine, USA), GTX100619

Secondary antibodies

Alexa Fluor 488 donkey-anti-goat IgG (H+L) (1:200, life technologies, Invitrogen, Carlsbad, USA), A-11055

Alexa Fluor 488 goat-anti-mouse IgG (H+L) (1:200, life technologies, Invitrogen)

Cy3-conjugated AffiniPure goat-anti-rabbit IgG (H+L) (1:200, Dianova, Jackson ImmunoResearch, Hamburg, Germany)

CyTM3- conjugated AffiniPure Anti-Rabbit IgG (H+L) and Fluorescein (FITC)-conjugated AffiniPure Anti-Mouse IgG (H+L) and CyTM3- conjugated AffiniPure Anti-Mouse IgG (H+L) (Jackson ImmunoResearch Laboratories INC. West Grove, USA)

DAPI (Hoechstfarbstoff) 8 Min

ImmunoSelect Antifading Mounting Medium (Dianova GmbH Hamburg, Germany), SCR-38448

Vectashield Antifade Mounting Medium with DAPI (Vector Laboratories)

Two sections per specimen underwent double immunofluorescent staining. To identify macrophages and fibroblasts, CD14 and CD90 markers were utilized for the double immunofluorescent staining on representative sections.

Microscopic analysis and scores

Regions of interest (ROI) with synovial lining layer (LL) and sublining layer (SL) were pinpointed, and digital imaging was carried out. Two independent observers analyzed the sections. The synovium underwent standard histological evaluation using Krenn et al.'s synovitis score, considering synovial lining cell layer enlargement, cell density, and inflammatory infiltrates (0–1: no synovitis, 2–4: low-grade synovitis, 5–9: high-grade synovitis).

Double immunofluorescent sections were observed with fluorescence microscopy in a darkened room using the Axioskop 40FL (Carl Zeiss Microscopy GmbH, Germany) with different specific filters and a mercury arc lamp (Zeiss, HBO 100) for excitation. Images were captured at 40x magnification using the AxioCam MRc5, managed by Axiovision 4.8.2 software (Zeiss, Germany). Visual data analysis was supported using the graphics program GIMP 2.8. The merge mode was utilized for presentations, and the contrast and brightness were adjusted. Cells were counted if their nuclei presented DAPI staining, or if the nucleus edges were clearly visible. Endothelial cells were identified by morphology in double immunofluorescent sections, and by CD34 in single-stained sections (data not shown). The number of positive and negative stained cells was recorded for each dye, layer, and ROI. Data was stored using Microsoft Exel.

Statistical analysis

The ratio of stained to unstained cells was determined. Data are presented as mean ± standard error of the mean. The Mann-Whitney U test was used for comparisons of quantitative antibody distribution and synovitis scores between RA and OA, employing semi-quantitative scores. P-values less than 0.05 were considered significant.

Correlations of antibody distribution and between immunohistochemical staining were assessed using Pearson´s and Spearman´s correlation methods along with graphical data processing. P-values less than 0.05 were considered significant and marked with an asterisk in tables. Non-significant results are not marked with an asterisk, a p-value was given when applicable.

Results

Patient characteristics

6 out of 8 patients were female in RA and 7 out of 8 in OA, with 7 out of 8 RA patients receiving corticosteroids, 6 receiving DMARDs, and 1 receiving NSAIDs, while in OA only 1 received DMARD and 7 out of 8 received NSAIDs (Table 1).

The synovitis score in H&E sections was significantly higher (p=0.0002*) in RA specimen (7.38 ± 1.30 points, mean ± SD) compared to OA specimen (2.75 ± 1.28 points mean ± SD) showing high-grade synovitis in RA and low-grade synovitis in OA. The total number of cells per ROI was marginally greater in RA compared to OA, with 65 ± 36 cells in RA and 54 ± 17 cells in OA within the synovial lining layer, and 238 ± 109 cells in RA versus 167 ± 41 cells in OA in the sublining layer. However, these differences did not reach statistical significance. The ratio was similar in double immunofluorescence sections.

Endothelial cells were observed in 12.4 ± 12.1% of cells in the sublining layer for RA and 20.6 ± 16.0% for OA, with no significant difference between the two groups (Table 1).

P2X5 and ASIC3

P2X5 and ASIC3 were identified to be present on local synovial tissue cells as well in RA (Figure 1) as in OA samples (Figure 2). Immunofluorescent staining for P2X5 and ASIC3 present within both layers LL and SL. The amount of cells staining for P2X5 and/or ASIC3 was not significantly different between RA and OA (Table 2).

Figure 1.

Double immunofluorescence labeling for ASIC3 and P2X5 receptors in RA synovial tissue sample. RA synovial tissue cells staining for ASIC3 (green), P2X5 (red) and DAPI (blue). Merged layer (a) shows cells staining for ASIC3 (green), P2X5 (red) and double (yellow). Primary images display cells staining for ASIC3 (green) and DAPI (blue) in panel (b), and P2X5 (red) in panel (c); Fluorescent cells are indicated by arrows.

Figure 2.

Double immunofluorescence labeling of ASIC3 and P2X5 receptors in OA synovial tissue sample. OA synovial tissue cells staining for ASIC3 (green), P2X5 (red) and DAPI (blue). Merged layer (a) shows cells staining for ASIC3 (green), P2X5 (red) and double (yellow). Primary images display cells staining for ASIC3 (green) and DAPI (blue) in panel (b), and P2X5 (red) in panel (c); Fluorescent cells are indicated by arrows.

Table 2.

Synovial cells in Lining Layer LL, Sublining Layer SL, staining for ASIC3 and P2X5, for NTPDase1 and NTPDase2 in RA and OA specimen (staining cells in % mean±SD).

Characteristics in %	Rheumatoid Arthritis						Osteoarthritis
Marker/Layer	LL			SL			LL			SL
ASIC3 solely, of all cells	7.8	±	9.3	7.1	±	6.2	7.5	±	12.6	10.3	±	7.0
P2X5 solely, of all cells	0.5	±	1.4	1.6	±	1.4	1.4	±	2.5	2.0	±	3.0
ASIC3+P2X5 double, of all cells	24.6	±	27.1	13.5	±	7.0	12.4	±	7.1	11.2	±	9.2
ASIC3 total, of all cells	32.4	±	25.2	20.5	±	10.6	20.0	±	9.8	21.5	±	8.8
ASIC3+P2X5 of all ASIC3	76			65			62			52
P2X5 total, of all cells	25.1	±	27.0	15.1	±	7.7	13.9	±	7.4	13.2	±	8.6
ASIC3+P2X5 of all P2X5	98			89			90			85
Correlation r ASIC3 to P2X5 *p>0.005	93.6*			98.8*
NTPDase1 solely	0.8	±	1.8	1.3	±	2.0	8.1	±	11.7	4.7	±	4.0
NTPDase2 solely	3.0	±	3.3	1.7	±	2.3	6.5	±	8.7	4.6	±	6.5
NTPDase1+2 double	37.6	±	23.0	32.0	±	17.6	15.5	±	10.7	15.2	±	11.1
NTPDase1 total, of all cells	38.4	±	23.5	33.3	±	16.6	23.6	±	12.9	19.9	±	10.3
NTPDase1+2 of all NTPDase1	98			96			66			77
NTPDase2 total, of all cells	40.5	±	23.2	33.7	±	16.9	22.0	±	10.1	19.8	±	9.6
NTPDase1+2 of all NTPDase2	93			95			71			77
Correlation r ENTPDase 1 to 2 *p<0.005	99.3*			99.9*						93.6*

In RA LL a total of 25.1% ± 27% cells stained for P2X5, 98% of these also stained for ASIC3. A total of 32.4% ± 25.2% of cells stained positive for ASIC3, 76% of these stained double for P2X5.

In RA SL a total of 15.1% ± 7.7% of cells stained positive for P2X5, 89% of these stained double for ASIC3. A total of 20.5% ± 10.6% of all cells stained positive for ASIC3, 65% out of these stained double for P2X5.

In RA, a highly significant positive correlation between P2X5 and ASIC3 positive, non-endothelial synovial cells was found in both layers (> 93% in LL, p = 0.0006; > 98% in SL, p = 0.0001; Pearson correlation coefficient suggests that cells co-express P2X5 and ASIC3.

In OA LL a total of 13.9% ± 7.4% of cells stained positive for P2X5, double staining for ASIC3 was seen in 90%. 20.0 ± 9.8% of cells stained positive for ASIC3, 62% of them stained double for P2X5.

In OA SL a total of 13.2% ± 8.6% of all cells stained positive for P2X5, 85% of these stained also for ASIC3. A total of 21.5% ± 8.8% of all cells stained positive for ASIC3, 52% of them stained double for P2X5.

In OA there was no significant correlation between P2X5 and ASIC3 in SL (p=0.0911) and LL (p=0.636).

NTPDase1 and NTPDase2

Synovial tissue cells in both RA and OA marked positive for NTPDase1 and NTPDase2. The amount of cells staining with NTPDase1 and/or NTPDase2 showed no significant difference between RA and OA (Table 2).

In RA LL a total of 38.4% ± 23.5% cells stained for ENTPD1, 98% of these stained double for ENTPDase2. A total of 40.5% ± 23.2% cells stained for NTPDase2, 93% of them stained double for ENTPDase1 (Figure 3).

Figure 3.

Double immunofluorescence labeling for NTPDase1 and NTPDase2 in RA synovial tissue sample. RA synovial tissue cells staining for NTPDase1 (green), NTPDase2 (red) and DAPI (blue). Merged layer (a) shows cells staining for NTPDase1 (green), NTPDase2 (red) and double (yellow). Primary images display cells staining for NTPDase1 (green) and DAPI (blue) in panel (b), and NTPDase2 (red) in panel (c).

In RA SL a total of 33.3% ± 16.6% of all cells stained positive for ENTPDase1, 96% of these stained double for ENTPDase2. A total of 33.7% ± 16.9% cells stained positive for NTPDase2, 95% of them stained double for NTPDase1.

There was a significant positive correlation between ENTPDase1 and ENTPDase2 in > 99% (p=0.0001) in RA LL and RA SL using the Pearson correlation coefficient. This suggests that cells co-express ENTPDase1 and ENTPDase2.

In OA LL a total of 23.6% ± 12.9% cells stained positive for NTPDase1, 66% of these stained double for NTPDase2. A total of 22.0% ± 10.1% cells stained positive for NTPDase2, 71% of these stained double for NTPDase1.

In OA SL a total of 19.9% ± 10.3% cells stained positive for NTPDase1, 77% of them stained double for NTPDase2. A total of 19.8% ± 9.6% cells stained positive for NTPDase2, 77% of them stained double for NTPDase1 (Figure 4).

Figure 4.

Double immunofluorescence labeling for NTPDase1 and NTPDase2 in OA synovial tissue sample. OA synovial tissue cells staining for NTPDase1 (green), NTPDase2 (red) and DAPI (blue). Merged layer (a) shows cells staining for NTPDase1 (green), NTPDase2 (red) and double (yellow). Primary images display cells staining for NTPDase1 (green) and DAPI (blue) in panel (b), and NTPDase2 (red) in panel (c).

There was a significant positive correlation between ENTPDase1 and ENTPDase2 in 94% in OA SL (p=0.0001).

Detailed percentages of cells staining solely or double for the different markers are provided in Table 2 and Figure 5.

Figure 5.

Distribution of markers in synovial tissue of RA and OA. Quantity and percentages (in %, mean±SD) of cells staining for: acid-sensing ion channel 3 (ASIC3), ATP-receptor P2X5, NTPDase1, NTPDase2 in the sublining layer (SL) and lining layer (LL) of human RA and OA synovial tissue specimen.

Further identification of synovial cell types using double staining with CD14, CD90, CD68, or vimentin was intended. Associations between CD14 paired with CD90 and ASIC3, and P2X5 with CD90 were present. However, a more precise identification and localization was not achievable using this histological method due to technical limitations.

Discussion

This study histologically explored the evidence for ATP-mediated purinergic signaling pathways suggesting an interplay of ASIC3 and P2X5 receptors in arthritic conditions in human rheumatoid arthritis and osteoarthritis synovial tissue. It also aimed to examine the local presence and location of ATP-degrading enzymes NTPDase1 and NTPDase2 assuming their importance in regulating purinergic signaling pathways.

There were more female than male participants in both patient groups because female RA and OA samples were more readily available at time of investigation. Furthermore, statistical power was limited by the small sample size and the ethical unavailability of non-arthritic control synovium. Future research could aim for a larger sample size and a more balanced distribution.

Synovial tissue inflammation confirmed a significantly higher level of synovitis in RA than in OA specimens, as anticipated. High-grade synovitis was found in RA and low-grade synovitis in OA. However, overall cell counts were only marginally and not significantly higher in RA than in OA. This could be attributed to samples being taken from end-stage disease joints of patients undergoing knee arthroplasty, which involved severe joint destruction, inflammation and limited treatment response over time.

In this study, ASIC3 and P2X5, as well as NTPDase1 and NTPDase2, were found present for the first time in human RA and OA synovial tissue cells within both the lining and sublining layers.

This finding suggests that the enhancement of ASIC3 by the P2X5 receptor is not confined to neural or muscular structures but may also influence local human synovial tissue cells directly.^26,41,42 ASIC3 is believed to facilitate hyaluronan release from articular chondrocytes in mice.¹⁵ Receptor activation on neural structures has led to central sensitization and hyperalgesia in arthritis and osteoarthritis models.

Increases in ASIC3 expression have been observed in joint afferents, as well as in the inflamed synovium of mice with osteoarthritis.^15,41–43 Sensory neurons innervating muscles use combinations of ASIC, P2X4, or P2X5 and TRPV1 receptors to detect metabolites such as ATP, lactate, and protons during muscle ischemia, which can cause severe pain 27.

The significant positive and very high correlation of P2X5 and ASIC3 on human rheumatoid synovial tissue cells in the present study highly suggests colocalization and a tight biological link and interplay between these receptors. The correlation was much higher in RA than in OA, pointing to a role in more severe synovitis, as suggested by the higher synovitis score.

This could correlate with increased ischemia and extracellular ATP in inflamed synovitis tissue in RA, leading to possibly upregulated P2X5, ASIC3, as well as NTPDase1, and NTPDase2 levels in RA synovial tissue. It remains unclear, whether upregulation of receptor units or ectonucleotidase quantity occurs on the individual cell membrane, which cannot be detected with the current method. Alternatively, receptors and enzymes may enhance their turnover capacity without increasing the total number of receptor units or cells, which also flees detection with the current study setup. Further research is needed to explore potential upregulation at the protein or turnover level. Final stages of disease with synovitis in both conditions may also contribute to similar results, while distribution in healthy controls remains unclear.

They may contribute to increased hyaluronan release, pannus formation and joint destruction. The study design does not allow for detailed evaluation of their consecutive effects due to the limited resolution of immunofluorescence microscopy. Surface contact between receptor units intracellular signal cascades and effects following ASIC3 receptor activation and possible purinergic enhancement via P2X5 remain subject for further investigation.

A significant and extremely high correlation between NTPDase1, and NTPDase2 in RA LL and SL, as well as in OA SL synovial tissue was found. This strongly suggests coexpression and a shared role in inflammation or pain signaling. It points to the need to tightly regulate extracellular nucleotide levels, balancing ATP signaling with adenosine-mediated resolution and other purins.

The current study found ASIC3 and P2X5, as well as NTPDase1 and 2 double staining on the same cells but could not clearly identify the cell types. More precise identification of cells was attempted using CD14, CD90, CD68 and vimentin. But limited selective markers and epitopes as well as technical limitations of immunofluorescent staining hindered clear identifications.^22,23 Further molecular studies are required to determine receptor and enzyme distribution on cell membrane level, cell identities, and intracellular signal cascades. Histological comparisons of RA and OA of peptidyl arginine deaminases and citrullinated proteins expression and distribution would be of interest for future studies. They play an additional role and could clarify how inflammatory microenvironments (hypoxia, extracellular ATP, cytokines) intersect with antigenic remodeling.

The current study found high evidence for local purinergic signaling pathways especially in RA and partially in OA synovial tissue. For the first time ASIC3 and P2X5 were found present in RA and OA synovial tissue and highly correlated in RA, supporting the idea of ischemia and ATP-mediated activation of these cells via purinergic enhancement of ASIC3 by P2X5.

Conclusion

ATP-degrading ectonucleotidases NTPDase1 and NTPDase2 were found present and colocalized on local synovial tissue cells in RA and OA, indicating a tightly regulated purinergic microenvironment in RA and OA synovial tissue. Data suggests that nucleotide turnover, including ATP breakdown and subsequent adenosine generation, modulates inflammatory and nociceptive processes and purinergic signaling pathways in synovitis.

Further research is needed to identify cell types with P2X5 and ASIC3, NTPDase1 and 2 using advanced molecular methods to understand enzyme distribution on membrane level and signaling pathways.

Footnotes

ORCID iDs

Veronika Wegener

Carolin Melcher

Christof Birkenmaier

Ethical considerations

This non interventional study was approved by the Ethics Committee of the Ludwig-Maximilians-University, Geschwister-Scholl-Platz 1, 80539 München. Approval date: 07.11.2012, approval number: 444-12.

Consent to participate

Informed consent to participate was written.

Author contributions

Veronika Wegener: study design, conduction, manuscript, Volkmar Jansson: study design, support, Christof Birkenmaier: Manuscript preparation, Bernd Wegener: study design, manuscript preparation, Carolin Melcher: Manuscript preparation.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by the Hans Sauer Stiftung, a german non-profit foundation promoting projects of science and research within the fields of natural environment and human health (grant number HSS 2011-035, title RESA). The funder had no role in the research design and conduction.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Manferdini

Paolella

Gabusi

, et al. From osteoarthritic synovium to synovial-derived cells characterization: synovial macrophages are key effector cells. Arthritis Res Ther 2016; 18: 83. https://doi.org/10.1186/s13075-016-0983-4

Ambarus

Noordenbos

de Hair

, et al. Intimal lining layer macrophages but not synovial sublining macrophages display an IL-10 polarized-like phenotype in chronic synovitis. Arthritis Res Ther 2012; 14(2): R74. https://doi.org/10.1186/ar3796

Becker

Rosa

Souza Vdo

, et al. Activities of enzymes that hydrolyze adenine nucleotides in platelets from patients with rheumatoid arthritis. Clin Biochem 2010; 43(13-14): 1096–1100. https://doi.org/10.1016/j.clinbiochem.2010.06.011

Xia

Zhang

, et al. Osteoarthritis pathogenesis: a review of molecular mechanisms. Calcif Tissue Int 2014; 95(6): 495–505. https://doi.org/10.1007/s00223-014-9917-9

Korkusuz

Dagdeviren

Eksioglu

, et al. Immunohistological analysis of normal and osteoarthritic human synovial tissue. Bull Hosp Jt Dis 2005; 63(1-2): 63–69.

Grafe

Schaffer

Rucker

. Kinetics of ATP release following compression injury of a peripheral nerve trunk. Purinergic Signal 2006; 2(3): 527–536. https://doi.org/10.1007/s11302-006-9018-y

Graff

Lazarowski

Banes

, et al. ATP release by mechanically loaded porcine chondrons in pellet culture. Arthritis Rheum 2000; 43(7): 1571–1579. https://doi.org/10.1002/1529-0131(200007)43:7<1571::AID-ANR22>3.0.CO;2-L

Sadananda

Kao

Liu

, et al.

Acid and stretch, but not capsaicin, are effective stimuli for ATP release in the porcine bladder mucosa: Are ASIC and TRPV1 receptors involved?

Eur J Pharmacol 2012; 683(1-3): 252–259. https://doi.org/10.1016/j.ejphar.2012.02.041

Garcia

Knight

. Cyclic loading opens hemichannels to release ATP as part of a chondrocyte mechanotransduction pathway. J Orthop Res 2010; 28(4): 510–515. https://doi.org/10.1002/jor.21025

10.

Cook

McCleskey

. Cell damage excites nociceptors through release of cytosolic ATP. Pain 2002; 95(1-2): 41–47. https://doi.org/10.1016/s0304-3959(01)00372-4

11.

D'Andrea

Calabrese

Grandolfo

. Intercellular calcium signaling between chondrocytes and synovial cells in co-culture. Biochem J 1998; 329(Pt 3): 681–687.

12.

Cotrina

Lin

Alves-Rodrigues

, et al. Connexins regulate calcium signaling by controlling ATP release. Proc Natl Acad Sci U S A 1998; 95(26): 15735–15740. https://doi.org/10.1073/pnas.95.26.15735

13.

Reinohl

Hoheisel

Unger

, et al. Adenosine triphosphate as a stimulant for nociceptive and non-nociceptive muscle group IV receptors in the rat. Neurosci Lett 2003; 338(1): 25–28. https://doi.org/10.1016/s0304-3940(02)01360-5

14.

Koeck

Schmitt

Baier

, et al. Predominance of synovial sensory nerve fibers in arthrofibrosis following total knee arthroplasty compared to osteoarthritis of the knee. J Orthop Surg Res 2016; 11: 25. https://doi.org/10.1186/s13018-016-0359-0

15.

Kolker

Walder

Usachev

, et al. Acid-sensing ion channel 3 expressed in type B synoviocytes and chondrocytes modulates hyaluronan expression and release. Ann Rheum Dis 2010; 69(5): 903–909. https://doi.org/10.1136/ard.2009.117168

16.

Loredo

Benton

. ATP and UTP activate calcium-mobilizing P2U-like receptors and act synergistically with interleukin-1 to stimulate prostaglandin E2 release from human rheumatoid synovial cells. Arthritis Rheum 1998; 41(2): 246–255. https://doi.org/10.1002/1529-0131(199802)41:2<246::AID-ART8>3.0.CO;2-I

17.

Caporali

Capecchi

Gamberucci

, et al. Human rheumatoid synoviocytes express functional P2X7 receptors. J Mol Med (Berl) 2008; 86(8): 937–949. https://doi.org/10.1007/s00109-008-0365-8

18.

Wirkner

Sperlagh

Illes

. P2X3 receptor involvement in pain states. Mol Neurobiol 2007; 36(2): 165–183. https://doi.org/10.1007/s12035-007-0033-y

19.

Burnstock

. Purinergic mechanisms and pain--an update. Eur J Pharmacol 2013; 716(1-3): 24–40. https://doi.org/10.1016/j.ejphar.2013.01.078

20.

Burnstock

. P2X ion channel receptors and inflammation. Purinergic Signal 2016; 12(1): 59–67. https://doi.org/10.1007/s11302-015-9493-0

21.

Burnstock

. Purinergic Mechanisms and Pain. Adv Pharmacol 2016; 75: 91–137. https://doi.org/10.1016/bs.apha.2015.09.001

22.

Burnstock

Arnett

Orriss

. Purinergic signaling in the musculoskeletal system. Purinergic Signal 2013; 9(4): 541–572. https://doi.org/10.1007/s11302-013-9381-4

23.

Burnstock

Boeynaems

. Purinergic signaling and immune cells. Purinergic Signal 2014; 10(4): 529–564. https://doi.org/10.1007/s11302-014-9427-2

24.

Dowd

McQueen

Chessell

, et al. P2X receptor-mediated excitation of nociceptive afferents in the normal and arthritic rat knee joint. Br J Pharmacol 1998; 125(2): 341–346. https://doi.org/10.1038/sj.bjp.0702080

25.

Chen

Zhang

Yang

, et al. LPS-induced dental pulp inflammation increases expression of ionotropic purinergic receptors in rat trigeminal ganglion. Neuroreport 2014; 25(13): 991–997. https://doi.org/10.1097/WNR.0000000000000193

26.

Birdsong

Fierro

Williams

, et al. Sensing muscle ischemia: coincident detection of acid and ATP via interplay of two ion channels. Neuron 2010; 68(4): 739–749. https://doi.org/10.1016/j.neuron.2010.09.029

27.

Light

Hughen

Zhang

, et al. Dorsal root ganglion neurons innervating skeletal muscle respond to physiological combinations of protons, ATP, and lactate mediated by ASIC, P2X, and TRPV1. J Neurophysiol 2008; 100(3): 1184–1201. https://doi.org/10.1152/jn.01344.2007

28.

Molliver

Immke

Fierro

, et al. ASIC3, an acid-sensing ion channel, is expressed in metaboreceptive sensory neurons. Mol Pain 2005; 1: 35. https://doi.org/10.1186/1744-8069-1-35

29.

Schetinger

Morsch

Bonan

, et al. NTPDase and 5'-nucleotidase activities in physiological and disease conditions: new perspectives for human health. Biofactors 2007; 31(2): 77–98. https://doi.org/10.1002/biof.5520310205

30.

Alvarado-Castillo

Harden

Boyer

. Regulation of P2Y1 receptor-mediated signaling by the ectonucleoside triphosphate diphosphohydrolase isozymes NTPDase1 and NTPDase2. Mol Pharmacol 2005; 67(1): 114–122. https://doi.org/10.1124/mol.104.006908

31.

Braun

Sevigny

Robson

, et al. Association of the ecto-ATPase NTPDase2 with glial cells of the peripheral nervous system. Glia 2004; 45(2): 124–132. https://doi.org/10.1002/glia.10309

32.

Corriden

Chen

Inoue

, et al. Ecto-nucleoside triphosphate diphosphohydrolase 1 (E-NTPDase1/CD39) regulates neutrophil chemotaxis by hydrolyzing released ATP to adenosine. J Biol Chem 2008; 283(42): 28480–28486. https://doi.org/10.1074/jbc.M800039200

33.

Antonioli

Pacher

Vizi

, et al. CD39 and CD73 in immunity and inflammation. Trends Mol Med 2013; 19(6): 355–367. https://doi.org/10.1016/j.molmed.2013.03.005

34.

Dos Santos Jaques

Becker

Souza

, et al. Activities of enzymes that hydrolyze adenine nucleotides in lymphocytes from patients with rheumatoid arthritis. Cell Biochem Funct 2013; 31(5): 395–399. https://doi.org/10.1002/cbf.2910

35.

Jimenez

Garcia

Adarmes

, et al. ATPase and Adpase activities in synovial membrane of equine metacarpophalangeal joint. Life Sci 2002; 70(20): 2445–2455. https://doi.org/10.1016/s0024-3205(02)01529-1

36.

Moncrieffe

Nistala

Kamhieh

, et al. High expression of the ectonucleotidase CD39 on T cells from the inflamed site identifies two distinct populations, one regulatory and one memory T cell population. J Immunol 2010; 185(1): 134–143. https://doi.org/10.4049/jimmunol.0803474

37.

Cope

Schulze-Koops

Aringer

. The central role of T cells in rheumatoid arthritis. Clin Exp Rheumatol 2007; 25(5 Suppl 46): S4–S11.

38.

Arnett

. Revised criteria for the classification of rheumatoid arthritis. Bull Rheum Dis 1989; 38(5): 1–6.

39.

Arnett

. Revised criteria for the classification of rheumatoid arthritis. Orthop Nurs 1990; 9(2): 58–64. https://doi.org/10.1097/00006416-199003000-00011

40.

Krenn

Morawietz

Burmester

, et al. Synovitis score: discrimination between chronic low-grade and high-grade synovitis. Histopathology 2006; 49(4): 358–364. https://doi.org/10.1111/j.1365-2559.2006.02508.x

41.

Ikeuchi

Kolker

Burnes

, et al. Role of ASIC3 in the primary and secondary hyperalgesia produced by joint inflammation in mice. Pain 2008; 137(3): 662–669. https://doi.org/10.1016/j.pain.2008.01.020

42.

Izumi

Ikeuchi

, et al. Local ASIC3 modulates pain and disease progression in a rat model of osteoarthritis. J Biomed Sci 2012; 19: 77. https://doi.org/10.1186/1423-0127-19-77

43.

Sluka

Rasmussen

Edgar

, et al. Acid-sensing ion channel 3 deficiency increases inflammation but decreases pain behavior in murine arthritis. Arthritis Rheum 2013; 65(5): 1194–1202. https://doi.org/10.1002/art.37862