The P2X7 receptor–inflammasome complex has a role in modulating the inflammatory response in primary Sjögren's syndrome

Authors


Abstract

Objective

Innate and adaptive immunity may contribute to gland dysfunction in patients with primary Sjögren's syndrome (pSS). The P2X7 receptor (P2X7R)–NLRP3 inflammasome complex modulates the release of the inflammatory cytokines IL-1β and IL-18. The presence of P2X7R in salivary glands suggests an interesting scenario for the initiation and amplification of the innate immune response in pSS. Therefore, the aim of this study was to assess the role of the P2X7R–NLRP3 inflammasome in pSS.

Subjects and Methods

Twenty-one consecutive patients with pSS according to the American–European Consensus Group criteria and 15 patients with sicca syndrome (i.e. without Sjögren's syndrome, non-SS) were enrolled in this study, together with six control (CTL) subjects. Expression of the P2X7R-NLRP3 platform and IL-18 was determined by real-time PCR and western blotting in gland specimens and peripheral lymphomonocytes; data were related to patients\x92 clinical, serological and histopathological characteristics. The presence of IL-18 was determined in gland and saliva samples.

Results

P2X7R expression was significantly higher in salivary glands from individuals with pSS than in those from non-SS and CTL subjects. Accordingly, the gene expression levels of the inflammasome components NLRP3, ASC and caspase-1 were significantly higher in pSS gland specimens, and this was paralleled by an increased expression of mature IL-18 in pSS saliva samples. The expression of both the P2X7R and the inflammasome components was a marker of disease-related glandular involvement, being increased in patients with anti-Ro/SSA positivity and correlated with focus score.

Conclusion

The results of this study suggest an involvement of the P2X7R–inflammasome–caspase-1–IL-18 axis in the development of pSS exocrinopathy. This finding provides the basis for studying the complex mechanisms underlying pSS, as well as for developing novel potential therapeutic strategies.

Abbreviations
pSS

primary Sjögren's syndrome

IFN

type-I interferon

ASC

apoptosis-associated speck-like protein containing a CARD

PYD

pyrin domain

IL-1β

interleukin-1β

IL-18

interleukin-18

P2X7R

P2X7 receptor

AECG

American and European Consensus Group

non-SS

non-sjogren syndrome

LSG

labial salivary gland

FS

focus score

CTL

control

ANA

antinuclear autoantibody

BSA

bovine serum albumin

Introduction

Primary Sjögren's syndrome (pSS) is a systemic autoimmune disease characterized by dysfunction and inflammation of the salivary and lachrymal glands, resulting in severe dryness of the eyes (xerophthalmia) and mouth (xerostomia). The exact cause of exocrine dysfunction in pSS remains unclear, but it has recently been suggested that both innate and adaptive immunity may contribute significantly to the glandular damage [1, 2]. Traditionally, the dominant role of the adaptive immune system in the pathogenesis of pSS has been attributed mainly to the presence of lymphocytic infiltrates in the glands and to positivity for specific autoantibodies such as anti-Ro/SSA (Sjögren syndrome A) and anti-La/SSB (Sjögren syndrome B). Recent evidence has highlighted the crucial role of the type-I interferon (IFN) signature in pSS, leading to growing interest in clarifying the contribution of the innate immune system at least in the initiation of the pathogenesis of pSS [1].

A critical component of the innate immune response is represented by inflammasome activation, triggered not only by microbial infection, but also by a wide range of both exogenous and endogenous noninfectious stimuli [3, 4]. Several inflammasomes have been described so far; amongst these, the most extensively characterized is the NLRP3 (Nod-like receptor family protein 3) inflammasome, activated by a number of diverse stimuli, including whole pathogens, microbial components and danger signals [5-7]. Upon activation, NLRP3 oligomerizes and recruits the adaptor protein apoptosis-associated speck-like protein containing a COOH-terminal caspase activation and recruitment domain (CARD) (ASC) through pyrin domain (PYD) interactions. In turn, procaspase-1 is recruited by ASC via CARD–CARD interactions, thus forming the NLRP3 inflammasome and leading to caspase-1 activation. This regulates the processing and secretion of proinflammatory mediators, such as the interleukins IL-1β and IL-18, which are able to cause tissue damage and maintain a state of chronic inflammation in several tissues and organs [8, 9]. In this complex scenario, the role of pyrin is debated: it was originally considered to be an anti-inflammatory molecule but more recent observations indicate that pyrin is an inflammasome modulator that is capable of activating caspase-1 [10, 11].

The purinergic P2X7 receptor (P2X7R) is an ATP-gated ion channel, which has an essential role in the innate immune response and is involved in the modulation of cell functions, such as growth and replication, apoptosis and neurotransmission [12-14]; the receptor also mediates several inflammatory responses in both circulating and resident cells [15, 16]. It has been demonstrated that P2X7R mediates activation of the NLRP3 inflammasome, leading to release of mature IL-1β and IL-18 [17, 18].

In addition to the systemic effects of IL-1β and IL-18, the latter is increased in serum and salivary glands of patients with pSS, where it is considered to exert a broad range of proinflammatory activities and to play a crucial role in the expansion and organization of infiltrative injuries [19, 20]. In salivary samples from patients with pSS, it appears that periductal CD68 +  macrophages and ductal epithelial cells represent the major sources of IL-18 [21].

The presence of P2X7R in salivary glands has already been characterized in relation to the process of saliva production [22, 23]; therefore, in the present study, we explored whether the P2X7R–inflammasome axis might have a role in the initiation and amplification of the innate immune response in pSS exocrinopathy.

Methods

Patients

Consecutive unselected female patients attending the Rheumatology Unit of the University Hospital in Pisa between November 2011 and April 2012 because of persistent (>3 months) subjective dry mouth and/or dry eye were enrolled in the study. The diagnosis of pSS was made according to the classification criteria of the American and European Consensus Group (AECG) [24], and all patients underwent minor salivary gland biopsy, which was evaluated according to the focus score (FS) by the same pathologist. Control salivary glands were obtained from subjects with suspected pSS but who did not fulfil the AECG criteria, that is,. having a FS of <1 in a labial salivary gland (LSG) biopsy (non-SS). Exclusion criteria were type 1 or type 2 diabetes, cancer, severe liver or kidney impairment and systemic inflammatory diseases. Some measurements were also compared with those obtained in preserved parotid specimens from six healthy control (CTL) subjects undergoing surgery for neoplasm of the neck area. A further group of 12 healthy volunteers provided a blood sample for lymphomonocyte isolation.

A personal medical history was obtained from all participants, including duration of sicca symptoms (e.g. xerophthalmia, xerostomia and parotid gland enlargement), smoking habits, glandular and extraglandular manifestations of the disease and ongoing treatments and comorbidities. Moreover, at the time of study entry, all subjects underwent a standardized evaluation for pSS, which included oral and ophthalmological examinations, laboratory tests and rheumatological assessment. Routinely used laboratory markers of disease activity (IgG, IgA, IgM, β2-microglobulin, complement C3 and C4 components and white blood cell count) were measured, as well as autoantibody profile [antinuclear antibody (ANA), anti-Ro/SSA, anti-La/SSB] and rheumatoid factor. None of the participants was positive for markers of hepatitis C or B virus (HCV/HBV).

The study protocol was approved by the ethics committee of the University of Pisa School of Medicine, and all participants gave their written informed consent.

Saliva sample collection and preparation

Unstimulated whole saliva was collected under standard conditions, using the draining method according to the AEGC criteria [24] in the morning (between 08.00 and 10.00) after an overnight fast; subjects were requested to refrain even from drinking water or chewing gum or sucking sweets for the previous 12 h [25]. To minimize the degradation of proteins, samples were kept on ice whilst being processed immediately. To remove debris and cells, samples were centrifuged at 21 000 g for 20 min at 4 °C. The amount of protein was estimated using an RC DC protein assay from Bio-Rad Laboratories (Segrate, Italy). Bovine serum albumin (BSA) was used as a reference standard.

Salivary gland samples

Minor salivary glands were obtained as part of routine diagnostic procedures when pSS was suspected. LSG biopsies were performed after infiltration of local anaesthetic. In all cases, part of the LSG specimen was fixed in neutral-buffered formalin for the assessment of the FS, and the remaining part was quickly frozen and stored at −80 °C. The formalin-fixed LSG specimens were processed (paraffin embedding, sectioning and haematoxylin and eosin staining) and evaluated by the same pathologist. If a diagnosis of focal lymphocytic sialadenitis was made, the FS was then determined according to the method of Greenspan and co-workers [26]. A focus was defined as an aggregate of ≥50 lymphocytes. The FS was reported as the number of foci per 4 mm2 of tissue, up to a maximum of 12 foci.

Parotid samples from CTL subjects were processed in the same manner.

Lymphomonocyte isolation

Mononuclear cells were isolated from fresh blood sample by density gradient centrifugation. Blood sample was collected in vacutainers containing sodium heparinate, diluted (1 : 1) with sterile Dulbecco's phosphate-buffered saline (D-PBS), slowly laid onto warm Histopaque 1077 (Sigma-Aldrich, Milan, Italy) and centrifuged at 400 g for 30 min at room temperature. To separate monocytes from lymphocytes, 1 ml of cell suspension was added to hyperosmotic Percoll solution (Sigma-Aldrich) in a final volume of 3 ml and centrifuged at 580 g for 15 min. Collected cells were added to D-PBS to a final volume of 15 ml and centrifuged at 350 g for 7 min, thus obtaining a monocyte-enriched suspension. To remove platelets and dead cells, 3 ml of iso-osmotic Percoll solution was overlaid with 1 ml of the monocyte-enriched suspension, and centrifuged at 350 g for 15 min; then, the supernatant was resuspended in D-PBS and centrifuged again at 350 g for 7 min, and the resultant pellet was stored at −80 °C.

RNA extraction and quantitative real-time PCR

Total RNA was extracted from a minor salivary gland biopsy and the frozen pellet of peripheral lymphocytes using the NucleoSpin RNA II kit (MACHEREY-NAGEL GmbH & Co., Duren, Germany). RNA (1 μg) was reverse-transcribed in a reaction volume of 20 μL using the High Capacity cDNA Reverse Transcription kit (Life Technologies, Monza, Italy). Real-time PCR was performed in triplicate using an Eco Real-Time System (Illumina Inc., San Diego, CA, USA) according to the standard protocol. Transcripts for P2X7R and inflammasome components were evaluated using TaqMan Gene Expression assays (Life Technologies): P2X7R, Hs00175721_m1; NRLP3, Hs00918082_m1; ASC, Hs00203118_m1; Pyrin, Hs00925524_m1; and Caspase 1, Hs00354832_m1.

Amplifications were normalized relative to GAPDH (Hs02758991_g1), and gene expression was quantified according to the ΔΔCT calculation, where CT is the threshold cycle and ΔΔ are the difference in threshold cycles for target and reference and the amount of the target gene, normalized to β-actin and relative to the calibrator (CTL subject) (given as 2-ΔΔCT).

Western blotting For Western blotting, 30 μg protein extracted from gland specimens and unstimulated whole saliva was diluted in sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) buffer and heated at 100 °C for 5 min. Samples and molecular weight markers were electrophoresed on Any kD Mini-Protean TGX gels (Bio-Rad, Hercules, CA) and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA, USA). After blocking using 5% BSA in tris-buffered saline and 0.05% Tween 20 (TTBS) for 1 h at room temperature, blots were washed three times in TTBS incubated overnight with primary antibodies against IL-18 (06-1115, Millipore, Temecula, CA, USA) and β-actin (sc-47778 Santa Cruz Biotechnology, Santa Cruz, CA, USA) both diluted 1 : 1000. The bands were detected by incubating the blot with species-specific secondary antibodies [goat anti-rabbit for IL-18 and procaspase-1, and goat anti-mouse for β-actin (IL-18, AP307P and β-actin, AP308P, Chemicon International, Temecula, CA, USA; procaspase-1, 06-503, Millipore, Darmstadt, Germany)], diluted 1 : 5000 for 1 h at room temperature, followed by enzymatic chemiluminescence (Immobilon Western, Millipore; Billerica, MA, USA; Temecula, CA, USA).

Nuclear and cytoplasmic extracts were prepared as described by Thiyagarajan et al. [27], using anti-NF-kB antibody (SC-109, Santa Cruz Biotechnology).

Statistical analysis

Results are presented as mean ± SD or median (range).The nonparametric Mann–Whitney test and Spearman's rank correlation coefficient were used for statistical comparisons. P-values <0.05 were considered statistically significant.

Results

Of the 36 LSG biopsies examined, 21 were diagnostic for pSS; the remaining 15 non-SS patients with sicca syndrome served as the reference group. All the non-SS patients had a FS <1 in LSG biopsies; by comparison, the mean FS in patients with pSS was 2.28 ± 1.33 (range, 1.0–5.0). The characteristics of the two study groups are shown in Table 1. None of the participants was positive for HCV/HBV serological markers, and no significant abnormalities were observed in routine biochemical and haematological tests. In addition, thyroid function was preserved in all subjects (data not shown).

Table 1. Characteristics of patients with Sjiögren syndrome (pSS) and sicca syndrome (non-SS)
 pSS (= 21)non-SS (= 15)
  1. Data are expressed as mean ± SD or number (%).

  2. ANA, antinuclear antibody.

  3. a

    = 0.05 versus non-SS.

Age (years)52.6 ± 14.854.0 ± 14.6
Disease duration (years)2.4 ± 2.12.6 ± 2.6
Smoker2 (9.5)3 (20)
Xerophthalmia21 (100)12 (80)
Xerostomia21 (100)11 (73)
Labial salivary gland focus score2.28 ± 1.330.0 ± 0.0
Salivary gland enlargement9 (42.9)2 (13)
Positive ocular test21 (100)8 (53)
ANA21 (100)6 (40)
Anti-Ro/SSA11 (52.4)0
Anti-La/SSB1 (4.8)0
Rheumatoid factor11 (52.4)3 (20)
C3 (mg L−1)102.9 ± 20.2115.3 ± 25.4
C4 (mg dL−1)20.2 ± 8.223.2 ± 7.2
IgG (mg dL−1)1432 ± 496a1145 ± 281
IgA (mg dL−1)215 ± 81200 ± 106
IgM (mg dL−1)158 ± 70137 ± 87
White blood cell count (n mmc−1)4946 ± 10156726 ± 4406
β2 microglobulin (μg L−1)2148 ± 5041926 ± 580

To clarify the possible involvement of the P2X7R in the local and/or systemic pathogenetic mechanisms of pSS, we measured P2X7R mRNA message in minor salivary glands and in peripheral monocytes of pSS and non-SS patients. As shown in Fig. 1, P2X7R mRNA was significantly more abundant in minor LSGs of patients with pSS, compared with either non-SS patients or CTL subjects. In peripheral lymphocytes from the three groups, we still observed a tendency, although not significant, towards a higher level of P2X7R expression in patients with pSS (P = 0.447).

Figure 1.

P2X7R mRNA expression in salivary gland samples (a) and peripheral lymphomonocytes (b) of patients with primary Sjögren's syndrome (pSS), patients with sicca syndrome (non-SS) and control subjects (CTL for gland samples in (a); HV for lymphomonocyte samples in (b). Data, normalized relative to the housekeeping gene GAPDH, are expressed as mean ± SD. *P < 0.004.

The next step was to assess whether there was a correlation between salivary P2X7R expression and the degree of inflammatory response in patients with pSS. In the whole study group, we observed a strong positive correlation between P2X7R mRNA message and FS in the minor salivary glands (R = 0.739, < 0.0001); a positive correlation was also observed between P2X7R mRNA expression and plasma levels of β2-microglobulin (R = 0.567, = 0.0003), which is an indirect index of lymphocytic activation in the gland (Fig. 2). In addition, P2X7R expression was significantly higher in patients with Ro-SSA (= 0.005), in ANA-positive subjects (< 0.0001) and in those with IgG levels >1600 mg dL−1 (= 0.013) (Fig. 3).

Figure 2.

Linear relationship between P2X7R mRNA expression and focus score (a) and plasma β2-microglobulin concentration (b) in pSS and non-SS patients. A representative image of inflammatory infiltrate without lymphocytic aggregates (c, magnification 4×) and histological section with a focus score of 4.88 (d, magnification 2.5×) are also shown.

Figure 3.

P2X7R mRNA expression in salivary gland samples from patients with and without Sjöogren's syndrome according to the presence of Ro-SSA autoantibodies (a, *P = 0.005); positivity for antinuclear antibody (ANA) (b, *P < 0.0001) and IgG level (c, *P = 0.013). Data, normalized relative to the housekeeping gene GAPDH, are expressed as mean ± SD.

No correlation was detected between P2X7R expression and other serological markers, including C3/C4 levels or presence of neutropenia or lymphocytopenia (defined as neutrophil count <1500 mm−3 and lymphocyte count <1000 mm−3, respectively).

Next, we explored the link between P2X7R and the NRLP3 inflammasome, to determine whether increased P2X7R expression and/or activity might translate into an activated inflammasome multimolecular platform. Thus, we measured the level of expression of NRLP3, ASC and caspase-1 in LSG biopsies. All these molecules as well as pyrin were more abundantly represented in pSS than in non-SS patients (Fig. 4). Weak correlations emerged between P2X7R expression and the inflammasome components NLRP3 (R = 0.234), ASC (R = 0.430) and caspase-1 (R = 0.325). A linear correlation was also found between pyrin and caspase-1 in patients with pSS (R = 0.955, = 0.011); in the same patients, but not in control subjects, P2X7R and pyrin expressions were also directly correlated with peripheral lymphomonocytes (R = 0.861, < 0.0001). However, of most clinical interest, there was a significant correlation between NLRP3 and FS in patients with pSS (R = 0.532, = 0.041).

Figure 4.

Expression of inflammasome components in patients with Sjögren's syndrome (pSS) and those with sicca syndrome (non-SS): NLRP3 (a); apoptosis-associated speck-like protein containing a CARD (ASC) (b); caspase-1 (c); and pyrin (d). Data, normalized relative to the housekeeping gene GAPDH, are expressed as mean ± SD. (e) Western blot for NF-κB expression in pooled proteins from pSS and non-SS is shown. *P = 0.005–0.0005 versus non-SS; °P = 0.07 versus non-SS.

Given that priming of NLRP3 activation is NF-κB dependent, we quantified NF-κB expression in gland specimens; no differences were observed between pSS and non-SS patients (Fig. 4).

Finally, because the inflammasome is a cytoplasmic multiprotein complex that activates caspase-1, leading to the processing and secretion of IL-18 (a powerful IFN-γ inducer, resulting in loss of local self-tolerance prior to developing gender-based autoimmunity), we evaluated protein expression of procaspase-1 in glands and measured IL-18 levels in gland and saliva specimens of all the patients. Procaspase-1 was more highly expressed in samples from pSS than from non-SS patients (Fig. 5); accordingly, even though we were unable to detect any expression of mature IL-18 in gland tissues, either in pSS or in non-SS individuals, P2X7R overexpression in LSGs was paralleled by an increased IL-18 presence in saliva samples from patients with pSS (Fig. 5). This finding indirectly supports the notion that activation of the inflammasome platform is mediated by the P2X7R in salivary glands in patients with pSS.

Figure 5.

Representative western blot showing procaspase-1 protein expression in gland specimens (a) and IL-18 protein expression in gland specimens (b) and saliva (c) of patients with Sjögren's syndrome (pSS), patients with sicca syndrome (non-SS) and healthy control subjects (CTL for gland samples; HV for saliva samples). Numbers (0–4) indicate the focus score. Densitometric analysis of IL-18 in saliva was performed in all participants (d) and refers to all the determinations; data are normalized relative to the housekeeping gene GAPDH and reported in arbitrary units. *P < 0.001 versus CTL and non-SS.

Discussion

There are two main novel findings of this study. First, the P2X7R is more highly expressed in salivary glands of patients with pSS than those of either non-SS patients with sicca syndrome or healthy individuals, and expression correlates with clinical, serological and histopathological features of the disease. Secondly, in patients with pSS, P2X7R expression is coupled to activation of the inflammasome complex and increased release of IL-18, suggesting a pathogenetic link between this receptor and the consequent focal sialadenitis via NLRP3 inflammasome activation.

Mammalian exocrine glands, such as pancreas, salivary and lacrimal glands, have different functions but similar morphology. The P2X7R is expressed in mouse and human parotid and submandibular acini and ducts [28-30], pancreatic ducts and lacrimal glands [31]. Recently, inflammasome activation has been described during acute pancreatitis, with the P2X7R upstream of the inflammasome and necessary for the development of pancreatic injury [32]. Caspase 1, the cytosolic enzyme central to the activation of the proinflammatory cytokine IL-1β, was earlier shown to be required for development of pancreatitis [33]. However, in other exocrine cells, such as salivary acinar cells, P2X7R stimulation leads formation of lytic pores, depolarization of the mitochondrial membrane and production of reactive oxygen species [34]. In murine salivary gland cells, activation of the P2X7R with extracellular nucleotides also stimulates the cleavage and release of α-fodrin, a cytoskeletal protein that acts as an autoantigen in the development of salivary gland dysfunction, enhances immune cell infiltration and triggers apoptosis [35]. Our observations support a relevant role of this receptor system in the pathophysiology of salivary gland inflammatory diseases, demonstrating the specific involvement of the P2X7R in the mechanisms initiating and/or maintaining chronic salivary gland inflammation in patients with pSS. Indeed, in these patients, the increased glandular P2X7R expression is significantly correlated with the entity of the lymphocytic infiltrate of the gland specimens; moreover, P2X7R expression is even higher in patients with a positive serological profile. Therefore, we tested the hypothesis that P2X7R-mediated activation of the inflammasome platform in pSS salivary glands could represent a novel and interesting role of the innate immune response in the pathogenesis of pSS. Thus, we examined the expression of the multiprotein complex consisting of NLRP3, ASC and caspase-1 in LSGs and found that all these components were present in abundance in patients with pSS, with a trend, although not significant, towards correlation with the levels of P2X7R expression. The linear relationship between either P2X7R or NLRP3 (the main component required for inflammasome activation) and FS (the major clinical indicator of the degree of the disease) seems to confirm the direct role of the P2X7R–inflammasome complex in the pathogenesis of immune-mediated damage that occurs in the salivary glands of patients with pSS, providing evidence of a link between the degree of local inflammation and the level of involvement of this complex. On the other hand, the lack of difference in gland NF-kB expression between patients and control subjects suggests the selective impairment of activation of the P2X7R-mediated NLRP3–ASC pathway in pSS.

Accordingly, pyrin, which is more highly expressed in salivary glands from patients with pSS, compared with non-SS and CTL subjects, appears to be a potential modulator of caspase-1 activation. However, the molecular pathway through which it is likely to exert this effect, probably involving activation of p38 MAPK by ribotoxic stress [36], needs further clarification. However, the interesting direct relationship between P2X7R expression and pyrin levels in lymphomonocytes of patients with pSS, but not in those of control subjects, further supports the link between these two systems.

As already mentioned, IL-18 is recognized as one of the main effectors of the gland inflammatory process during the pathogenesis of pSS [37]. The increase in levels of this key cytokine in the salivary samples of patients with pSS indirectly supports our working hypothesis, therefore suggesting that, in these patients, P2X7R activation increases in parallel with increasing disease activity and probably acts as a trigger for the subsequent recruitment of the NLRP3 inflammasome components and IL-18 cleavage and release [38]. This view is reinforced by the increased procaspase-1 expression found in protein extracted from patients with pSS and by the lack of histological detection of IL-18 in salivary gland samples, confirming that the IL-18 precursor is rapidly cleaved by caspase-1 and fully released with saliva as mature cytokine.

In conclusion, the results of this proof-of-concept study support a potential involvement of the P2X7R–inflammasome–caspase-1–IL-18 axis in the development of pSS exocrinopathy, shedding new light on the role of the innate immune response in patients with this condition. Our preliminary findings warrant validation with further studies. Nevertheless, these observations on the potential pathogenetic role of P2X7R in pSS may provide a rationale for novel targeted therapies for this disease.

Conflict of interest statement

None of the authors has any conflict of interests to declare.

Acknowledgements

This study was supported by a grant from the University of Pisa.

Ancillary