Article Text
PDF
Abstract
Objective To determine whether dysregulated production of cytokines downstream of interleukin (IL)-1 participates in the pathophysiology of cryopyrin-associated periodic syndromes (CAPS).
Methods Primary monocytes from patients with CAPS, unstimulated or after stimulation with lipopolysaccharide (LPS) and other Toll-like receptor (TLR) agonists, were examined for signs of stress and production of IL-1β, IL-1 receptor antagonist (IL-1Ra) and IL-6 in comparison with monocytes from patients with autoimmune diseases and from healthy donors.
Results Unstimulated CAPS monocytes showed mild signs of stress including elevated levels of reactive oxygen species and fragmented mitochondria. Stress signs were worsened by TLR stimulation and eventually led to protein synthesis inhibition with strong impairment of production of cytokines downstream of IL-1, such as IL-1Ra and IL-6. These defects were not detected in monocytes from autoimmune patients and healthy donors.
Conclusions The stress state of LPS-stimulated CAPS monocytes and the consequent inhibition of translation are likely to be responsible for the impaired production of IL-1Ra and IL-6. The deficient secretion of these cytokines coupled with increased IL-1β release explains the severity of the IL-1-related clinical manifestations and the predominant implication of innate immunity in CAPS.
Statistics from Altmetric.com
Introduction
Cryopyrin-associated periodic syndromes (CAPS) are autoinflammatory disorders linked to mutations of the inflammasome gene NLRP3 and characterised by severe inflammatory symptoms.1 ,2 Unlike autoimmune diseases where lymphocytes are involved, autoinflammatory diseases are exquisitely innate immunity disorders and the key cytokine is interleukin (IL)-1β.2 Although CAPS monocytes secrete more IL-1β than healthy monocytes,1 the overall increase is relatively small, raising the question of how a modest increase might cause the devastating IL-1-dependent symptomatology.2 The finding that lipopolysaccharide (LPS) brings forward IL-1β secretion by CAPS monocytes3 provided a partial explanation: the premature secretion of all the secretable IL-1β when control mechanisms such as IL-1 receptor antagonist (IL-1Ra) are not yet operative4 might explain the violent inflammatory manifestations of CAPS. Indeed, as is evident in patients genetically deficient in IL-1Ra,5 lack of IL-1Ra allows unopposed action of IL-1 with dramatic consequences. The rapid secretion of IL-1β is related to abnormal redox conditions of resting CAPS monocytes and their consequent aberrant redox response to LPS.3 We determined whether stress conditions or the subsequent stress responses also affect the production of cytokines downstream of IL-1β and impact on the pathophysiology of the disease.
Methods
Patients
Eight patients with CAPS with NLPR3 mutations, five patients with antineutrophil antibody-positive oligoarticular juvenile idiopathic arthritis (JIA) and six with systemic lupus erythematosus were enrolled (table 1). Eighteen age-matched healthy controls were studied in parallel.
Clinical characteristics and ongoing treatment of patients
Cell preparation and culture
Blood samples were taken and fresh monocytes were enriched and activated with LPS, R848 or zymosan (Sigma-Aldrich, Milan, Italy) as described previously.3 ,6
ELISA
Cytokines in supernatants were quantified by ELISA (R&D Systems, Minneapolis, MN, USA).
Cell death evaluation
Monocytes stained with annexin V-FITC/propidium iodide (Bender MedSystems, San Diego, CA, USA) were analysed by flow cytometry.3 Trypan blue staining was performed in parallel.
Intracellular reactive oxygen species (ROS) measurement
LPS-stimulated monocytes were loaded with 10 µM 2′7′-dichlorofluorescein diacetate (H2DCF-DA) (Life Technologies Italia, Monza, Italy). Fluorescence was measured in cell lysates and data normalised versus the protein content.3 ,6
Biosynthetic labelling
Monocytes were endogenously labelled with (35S) methionine/cysteine (Perkin Elmer, Monza, Italy) for 1 h at various times from LPS exposure.7 Cells were then lysed and aliquots of cell lysates precipitated in 25% cold trichloroacetic acid. Insoluble radioactivity was measured in β-counter.7
Electron microscopy
Cells were prepared as described elsewhere.8 Ultrathin sections were examined under a Morgagni 268D transmission electron microscope (FEI Company, Hillsboro, Oregon, USA).8 For each sample, 30 randomly taken cell sections in 10 different microscopic fields were recorded using a charge-coupled device camera (Soft Imaging System GmbH, Munster, Germany) and analysed with the AnalySIS software (Soft Imaging System) for the percentage of damaged mitochondria classified using a grading scale based on the mitochondrial area with intact cristae, as described previously.9 Thus, injury grading was categorised into three levels of morphological mitochondria damage: Mt-g3, severe; Mt-g2, moderate and Mt-g1, slight, corresponding to 0%, 1–50% and 51–75% of the area occupied by intact cristae.9
Statistical analysis
In electron microscopy experiments, Student t tests were used (significance level defined as p<0.05). In the other experiments differences between groups were evaluated using the non-parametric Kruskal–Wallis test. Post hoc analysis was performed with the non-parametric Mann–Whitney U test. The significance level was defined as p<0.01.
Results
Oxidative stress and mitochondria damage characterise CAPS monocytes
The presence of an altered redox potential and its effects on cell function were investigated in monocytes from patients with CAPS compared with autoimmune patients (table 1) and healthy subjects. Unstimulated CAPS monocytes exhibited higher levels of ROS, further increased by LPS and Toll-like receptor (TLR)7 and TLR8 agonist R848 (figure 1A). Antioxidant markers such as released free cysteine3 ,6 were also higher in unstimulated CAPS monocytes but decreased following LPS activation (figure 1B), indicating the insurgence of oxidative stress. Comparable levels of redox stress were found in monocytes from patients untreated or under treatment with IL-1 blockers. Unlike other case records,10 the rate of death of CAPS monocytes stimulated with LPS or with R848 or the TLR2-ligand zymosan was variable but comparable to healthy monocytes (figure 1C). However, stress signs were observed—namely, mitochondria alterations such as fragmentation and damage, measured as previously described,9 were significantly more abundant in unstimulated CAPS than in healthy monocytes and were increased by TLR agonists (figure 1D), supporting the view that mitochondria are an early target of ROS that cause imbalanced fission-fusion, fragmentation and dysfunction.11
Redox, mitochondria and protein synthesis alterations in monocytes from patients with cryopyrin-associated periodic syndromes (CAPS). (A) Intracellular reactive oxygen species (ROS) levels (expressed as mean±SD relative fluorescence units (RFU)/106 cells) in monocytes from healthy donors (HD) (n=13), patients with CAPS (n=6), patients with juvenile idiopathic arthritis (JIA) (n=4) or systemic lupus erythematosus (SLE) (n=4) cultured 1 h without (white columns) or with lipopolysaccharide (LPS) (grey columns). Three representative HD and CAPS were also stimulated with R848 (black column). (B) Cysteine release (µM/106 cells, mean±SD) in 18 h supernatants of unstimulated (white columns) and LPS-stimulated (grey columns) monocytes. (C) Percentage of living monocytes from HD and CAPS patients stimulated with LPS for various times (left panel) or with LPS, R848 or zymosan (ZYM) for 48 h (right panel) evaluated by flow cytometry of annexin V/propidium iodine-stained cells (mean±SD). Trypan blue exclusion assay performed in parallel showed that the percentage of cells excluding the dye at 48 h from LPS exposure were 65±17 and 60±20 for HD and CAPS patients, respectively. (D) Percentage (mean±SE) of mitochondria displaying increasing grading of damage (Mt-g1 to Mt-g3) measured as the percentage area with intact cristae in monocytes (Mo) and lymphocytes (Ly) from patients with CAPS and matched HD, untreated or treated with LPS or R848 for 9 h. Mt-g1: 51–75% of intact cristae; Mt-g2: 1–50% of intact cristae, Mt-g3: 0% of intact cristae. Insets show representative images of mitochondria in LPS-treated HD (1) and CAPS monocytes (2) (bars 0.5 μm); (3) shows a monocyte and a lymphocyte from the same patient with CAPS in close proximity (bars 5 μm). (E) Acid insoluble 35S-methionine/cysteine (met/cys) incorporation by monocytes from patients with CAPS, SLE or JIA exposed to LPS and incubated for 1 h with 35S-met/cys at the indicated time points. Data are expressed as percentage of 35S-met/cys incorporation by healthy monocytes (n=6) labelled in parallel (taken as 100%). In B–E, the number of analysed subjects is given in parentheses.
ROS and mitochondrial dysfunction may induce integrated stress responses with inhibition of protein translation.12 ,13 Accordingly, CAPS monocytes reacted to LPS-induced oxidative stress by slowing down protein synthesis (figure 1E).
TLR agonist-induced production of IL-1Ra and IL-6 is impaired in CAPS monocytes
To clarify the functional relevance of CAPS monocyte distress, secretion of cytokines downstream of IL-1 was assessed. Both IL-1Ra and IL-6 production were dramatically impaired in CAPS monocytes (figure 2A,B), as evident 9 h after TLR stimulation (figure 2D,E). In contrast, secreted IL-1β evaluated 18 h after LPS exposure was higher in CAPS monocytes from untreated patients and almost normalised by anti-IL-1 therapy in monocytes from treated patients (figure 2C), as described elsewhere.1 ,3 ,14 ,15 Moreover, confirming our previous observations,3 IL-1β secretion was accelerated in patients with CAPS, reaching a plateau at 3 h from TLR triggering and then stopped while IL-1β secretion by healthy monocytes was still increasing at 18 h (figure 2F). Unlike CAPS, monocytes from autoimmune patients released high levels of the three cytokines (figure 2A–C).
Monocytes from patients with cryopyrin-associated periodic syndromes (CAPS) are impaired in the production of interleukin-1 receptor antagonist (IL-1Ra) and IL-6. Concentrations of (A) IL-1Ra, (B) IL-6 and (C) IL-1β were determined in the supernatant of monocytes from healthy donors (HD) and patients with CAPS, systemic lupus erythematosus (SLE) or juvenile idiopathic arthritis (JIA) cultured for 18 h with lipopolysaccharide (LPS) (1 µg/ml). Closed symbols represent CAPS patients receiving anti-IL-1 treatment (treatment duration 3–12 months); open symbols represent CAPS patients out of treatment. (D–F) Time course analysis of the indicated cytokines secreted by monocytes from HD and CAPS subjects exposed to LPS for 3, 6, 9 and 18 h. (G) IL-1Ra and IL-1β secreted by monocytes obtained from the indicated CAPS subjects (table 1) before (pre) or 1 week after (post) the beginning of treatment with IL-1 blockers and stimulated for 18 h with LPS. (H) Monocytes from representative HD and CAPS patients were stimulated in parallel with LPS (n=3), R848 (n=3), zymosan (ZYM, n=1). Released IL-1Ra (left), IL-6 (middle) and IL-1β (right) were quantified by ELISA 18 h after exposure to the toll-like receptor ligands. In A–F, the number of analysed subjects is given in parentheses.
Cytokine secretion by monocytes from two patients with CAPS evaluated before or after starting treatment with IL-1 blockers showed that the treatment decreased IL-1β secretion, as expected,3 but did not influence IL-1Ra and IL-6 secretion which remained low (figure 2G).
Finally, IL-1Ra and IL-6 secretion were evaluated following stimulation of CAPS monocytes with R848 or zymosan. As shown in figure 2H, a dramatic decrease in both cytokines was seen, indicating that the early failure of CAPS monocytes is not restricted to TLR4 triggering but is induced by different TLR agonists.
Discussion
We have shown that TLR-stimulated CAPS monocytes undergo an early functional exhaustion, resulting in deficient production of IL-1Ra and IL-6. Abnormally elevated levels of ROS and antioxidants and of damaged mitochondria are already present in unstimulated CAPS monocytes and are increased upon TLR triggering. ROS in surplus cannot be detoxified due to the collapse of the antioxidant response and lead to mitochondrial damage. CAPS monocytes then shut down protein translation.
Attenuation of protein translation, which affects several different proteins, is a common stress response in stressed cells aimed at preventing cell death.12 ,13 In CAPS monocytes the decreased protein synthesis, while indeed maintaining cells alive in a percentage comparable to healthy monocytes, is conceivably responsible for the dramatically reduced secretion of IL-1Ra and IL-6.
All the described alterations are similar in CAPS monocytes from untreated and treated patients, indicating that redox defects are intrinsic in CAPS monocytes. Since anti-IL-1 therapy decreases IL-1β production,3 the stress state of CAPS monocytes is probably unrelated to the IL-1β levels.
Owing to the low basal level of ROS, healthy monocytes resist LPS/ROS-induced stress longer,16 preserving effective cell functioning with secretion of high levels of IL-6 and IL-1Ra. Similarly, monocytes from patients with autoimmune diseases do not display redox distress and secrete large amounts of both cytokines, ruling out the possibility that stress and low secretion of IL-1Ra and IL-6 are common traits of chronic inflammatory disorders.
Why and how the NLRP3 mutations cause cell stress is unclear and deserves further investigation. Notably, lymphocytes from patients with CAPS have normal mitochondria (figure 1D). Since NLRP3 is not expressed by lymphocytes,17 the data suggest that stress is restricted to cells actually expressing the mutated protein. Interestingly, increased ROS levels also characterise monocytes from patients with NLRP12 mutations responsible for a mild autoinflammatory disease18 and patients affected by TRAPS, an autoinflammatory syndrome associated with mutated TNFR1,19 which suggests that loss of redox homeostasis is a hallmark of monocytes from monogenic autoinflammatory syndromes expressing mutated proteins.20
The deficiency of IL-1Ra allows the amplification of IL-1-mediated effects.5 ,21 The low availability of IL-1Ra therefore explains the stronger inflammatory symptoms exhibited by patients with CAPS (and also those cases where the increase in IL-1β secretion is modest) compared with autoimmune patients whose monocytes may secrete more IL-1β than controls but accompanied by high levels of IL-1Ra.
Serum IL-6 derives in part from monocytes. In agreement with the different rate of IL-6 produced by CAPS and autoimmune monocytes (figure 2B), serum IL-6 is much lower in CAPS than in autoimmune diseases.22 However, IL-6 in CAPS serum is above the normal range,15 most likely due to other sources of IL-6 which, not expressing NLRP3, are not under stress and can produce normal levels of the cytokine.
In conclusion, our data identify a new pathogenic mechanism in CAPS that may contribute to the disease phenotype, which is mediated by the higher susceptibility of CAPS monocytes to TLR-induced stress.16 Decreased production of IL-1Ra and IL-6 coupled with increased IL-1β secretion explain the main implication of the innate immunity and gives reason for the dependency of monogenic autoinflammatory diseases on IL-1.
Acknowledgments
The authors thank the NCI (Biological Resources Branch) for the anti-IL-1β 3ZD monoclonal antibody.
References
Footnotes
-
Funding Telethon Italia (GGP09127), Compagnia San Paolo, MIUR, Associazione Italiana per la Ricerca sul Cancro (IG 10272). SC is the recipient of the ‘Young Investigators’ grant GR-2010-2309622 from the Italian Ministry of Health.
-
Competing interests None.
-
Patient consent Obtained.
-
Ethics approval Ethics approval was obtained from the ethical board of the G. Gaslini Institute.
-
Provenance and peer review Not commissioned; externally peer reviewed.