NLRP3 Inflammasome Activation in Dialyzed Chronic Kidney Disease Patients

To assess whether NLR pyrin domain-containing protein 3 (NLRP3) inflammasome, a multiprotein complex that mediates the activation of caspase-1 (CASP-1) and pro-inflammatory cytokines IL-18 and IL-1β, could be involved in the chronic inflammatory state observed in chronic kidney disease patients undergoing hemodialysis treatment (CKD-HD), we employed several biomolecular techniques including RT-PCR, western blot, FACS analysis, confocal microscopy and microarray. Interestingly, peripheral blood mononuclear cells from 15 CKD-HD patients showed higher mRNA levels of NLRP3, CASP-1, ASC, IL-1β, IL-18 and P2X7receptor compared to 15 healthy subjects. Western blotting analysis confirmed the above results. In particular, active forms of CASP-1, IL1-β and IL-18 resulted significantly up-regulated in CKD-HD versus controls. Additionally, elevated mitochondrial ROS level, colocalization of NLRP3/ASC/mitochondria in peripheral blood mononuclear cells from CKD-HD patients and down-regulation of CASP-1, IL1-β and IL-18 protein levels in immune-cells of CKD-HD patients stimulated with LPS/ATP in presence of mitoTEMPO, inhibitor of mitochondrial ROS production, suggested a possible role of this organelle in the aforementioned CKD-associated inflammasome activation. Then, microarray analysis confirmed, in an independent microarray study cohort, that NLRP3 and CASP-1, along with other inflammasome-related genes, were up-regulated in 17 CKD-HD patients and they were able to clearly discriminate these patients from 5 healthy subjects. All together these data showed, for the first time, that NLRP3 inflammasome was activated in uremic patients undergoing dialysis treatment and they suggested that this unphysiological condition could be possibly induced by mitochondrial dysfunction.


Introduction
Chronic kidney disease (CKD) is one of the leading clinical features of nephrology patients and it represents a major and growing challenge for healthcare systems. The prevalence rate of induction of inflammasome components including NLRP3 itself and pro-IL-1β. The second, or activating signal, is able to directly activate inflammasome assembly [41].
Although several reports have focused on inflammasome in various kidney diseases [42,43] no report has studied its role in chronic inflammation associated with hemodialysis. Therefore the aim of our study was to investigate the activation of NLRP3 inflammasome in peripheral blood mononuclear cells (PBMCs) from CKD-HD patients and the possible involvement of mitochondrial dysfunction in this process.

Patients and Controls
A total of 52 subjects [n: 32 hemodialyzed chronic kidney disease patients (CKD-HD) and n:20 healthy subjects (NORM)], after signing informed consent, were enrolled in our study.
Thirty subjects were included in the training-group (15 NORM and 15 CKD-HD) and 22 in the microarray-group (5 NORM and 17 CKD-HD).
Since we did not obtain sufficient number of cells for all the analysis and in order to avoid additional blood collection, western blot experiments and FACS analysis were performed in 10 NORM and 10 CKD-HD.
The main demographic and clinical features are summarized in Table 1.
To avoid confounding factors, all patients suffering from systemic autoimmune disorders, (e.g. SLE, vasculitis) infectious diseases, diabetes, chronic lung diseases, neoplasm, or inflammatory diseases and patients receiving antibiotics, corticosteroids, or non-steroidal antiinflammatory agents were excluded. No patients had symptomatic coronary artery diseases or a family history of premature cardiovascular diseases.
The study was carried out according to the Declaration of Helsinki and approved by Institutional Ethic Review Boards of the University Hospital of Verona, Italy and University Hospital "Policlinico di Bari", Bari, Italy. Universal master mix obtained from Kapa Biosystems included all reagents. The β-actin gene amplification was used as a reference standard to normalize the target signal. The comparative Ct method (ΔΔCt) was used to quantify gene expression, and the relative quantification was calculated as 2 -ΔΔCt . Amplification specificity was controlled by a melting curve analysis and the amount of mRNA target was evaluated using the comparative Ct method.

Mitochondrial reactive oxygen species (ROS) measurement
Freshly isolated PBMC were incubated with 5 μM MitoSOX Red mitochondrial superoxide indicator (Molecular Probes, Invitrogen, Carlsbad, CA), for 10 min at 37°C, in the dark, according to manufacturer's instructions. Autofluorescence of unstained cells were used as control for each sample. Approximately 40,000 gated events were acquired for each sample on a FACS-Canto (Becton Dickinson, San Jose, CA) and analyzed using FlowJo software (TreeStar, Ashland, OR). Dead cells and debris were excluded based upon forward scatter and side scatter measurements. All analyses were gated on PBMC, based on morphologic identification (forward scatter vs. side scatter). Percentage of Mitosox-positive cells (fluorescence signals at 580 nm) was calculated by subtracting the percentage of autofluorescence-positive cells.

Confocal microscopy
Freshly isolated PBMC from 3 NORM and 3 CKD-HD were spotted on poly-L-lysine-coated slides and incubated with Mitotracker deep red (Molecular Probes, Invitrogen, Carlsbad, CA), 100 nM for 30 min at 37°C according to manufacturer's instruction. The cells were washed and fixed with 4% paraformaldehyde (PFA). PFA was quenched with 50 mM NH 4 Cl. Cells were then permeabilized with PBS-0.1% Triton X-100. After blocking with 1% bovine serum albumin in PBS, the slides were incubated with anti-NLRP3 and anti-ASC antibodies (Abcam) for 1 hour. The slides were then extensively washed in PBS and incubated with Alexa Fluor 488 Goat Anti-Rabbit and Alexa Fluor 594 anti-mouse. Finally, nuclei were stained with DAPI. Negative controls were performed by omitting the primary antibodies (S1 Fig.). Images were collected using the SP5 confocal microscope from Leica Microsystems (Wetzlar, Germany).

Statistical analysis
Data are expressed as the mean ± standard deviation (SD). T-test and chi-square test were used to assess differences in clinical and demographic features. A value of p < 0.05 was considered to be statistically significant.
For microarray analysis, gene expression values for the 22,283 gene probe sets, scaled to the target intensity of 2,500, were log transformed. Probe sets unexpressed in the entire microarray cohort of patients were omitted. However, for our analysis we used only 14 gene probe sets (corresponding to 5 genes) encoding for the biological elements included in the NLRP3 inflammasome according to NOD-like receptor signaling pathway in KEGG pathway analysis (http:// www.genome.jp/kegg/pathway.html). Principal component analysis (PCA) was performed using Spotfire decision site 9.0 (www.spotfire.com). A p value cut-off of 0.005 was used to identify differentially expressed probe sets between the two groups.

Increased expression of genes encoding for NLRP3 inflammasome components and pro-inflammatory cytokines in PBMC from CKD-HD patients
RT-PCR demonstrated that mRNA levels of NLRP3, ASC, CASP-1 (inflammasome components) and pro-inflammatory cytokines IL-1β and IL-18 were higher in PBMC isolated from CKD-HD patients compared to NORM (Fig. 1).
P2X7 receptor (P2X7R), an ATP-gated cation channel that plays a key role in ATP-induced inflammasome activation [44] was up-regulated in patients.
These results reveal an enhanced inflammatory status in patients with severe renal failure undergoing dialysis treatment.

Increment of mitochondrial ROS level in CKD-HD compared to controls
To measure the mitochondrial ROS production in PBMC of CKD-HD patients, we used the MitoSOX Red, a fluorogenic dye for selective detection of superoxide in mitochondria of live cells.
CKD-HD patients showed a significant higher levels of mitochondrial ROS compared to NORM (Fig. 2).
This result was in line with our previous findings demonstrating a CKD-related mitochondrial dysfunction [45] and it suggests a possible link between this organelle and inflammasome activation.
Colocalization of NLRP3, ASC and mitochondria in PBMC from CKD-HD patients ROS are short-lived molecules and they can act as a signaling messenger only for a short distance [46]. Thus, to be effective NLRP3 should be localized in close proximity to mitochondria allowing an efficient sensing of the ROS produced by this organelle. To demonstrate the above mentioned co-localization and to evaluate the activation of NLRP3 inflammasome in CKD-HD patients, we performed confocal microscopy experiments using MitoTracker Deep Red FM, a far red-fluorescent dye that stains mitochondria in live cells. Interestingly, NLRP3 protein co-localizes with ASC and mitochondria in PBMC from CKD-HD patients whereas it remains in cytoplasmic granular structure in PBMC from NORM (Fig. 3).
These results show the activation of NLRP3 inflammasome in PBMC from CKD-HD patients with a possible involvement of mitochondria.

Caspase-1 activation and pro-inflammatory cytokines maturation in PBMC from CKD-HD patients
Since the NLRP3 inflammasome once activated causes the proteolytic cleavage of caspase-1 and the subsequent maturation of pro-inflammatory cytokines [36], we decided to measure the protein level of activated (cleaved) caspase-1 and mature IL1β and IL-18 in 10 CKD-HD patients and 10 NORM from the training-group.
As predictable, the protein levels of caspase and both cytokines were higher in CKD-HD patients compared to NORM (p<0.01) (Fig. 4).

Inhibitory effects of mitoTEMPO on caspase-1 activation and maturation of IL-1β and IL-18 induced by LPS/ATP in PBMC from CKD-HD
To better assess the role of mitochondria in inflammasome activation, we measured the level of CASP-1, IL-1β and IL-18 in PBMC from CKD-HD stimulated with LPS/ATP, Pathogen- associated molecular pattern (PAMP) and danger-associated molecular pattern (DAMP) signals able to activate NLRP3 inflammasome, in absence or presence of mitoTEMPO, an inhibitor of mitochondrial ROS production.
As shown in Fig. 5, mitoTEMPO caused a reduction of caspase-1 activation and cytokines maturation in CKD-HD.
These results confirmed the close association between inflammation and mitochondrial dysregulation in CKD-KD patients.

Microarray analysis confirmed a de-regulation of inflammasome genes in CKD-HD patients compared to healthy subjects
Microarray analysis revealed that 5 up to 15 gene probe sets were able to discriminate CKD-HD from NORM (cut-off of discrimination: p<0.005).
Principal component analysis (PCA) using the expression levels of the aforementioned gene probe sets was able to clearly discriminate in three dimensional space CKD-HD patients from NORM (Fig. 6B).

Discussion
In this study, performed by using classical bio-molecular methodologies (RT-PCR, western blot, FACS analysis, confocal microscopy) combined with an high-throughput technology (microarray), we found that the NLRP3 inflammasome was activated in immunocompetent peripheral cell lines isolated from uremic patients undergoing dialysis treatment. This result was in line with a previous study showing the inflammasome activation in kidney tissue of mice with renal impairment followed to unilateral ureteral obstruction. Compared with wild-type, Nlrp3 -/mice had less tubular injury, inflammation and fibrosis associated with a reduction in caspase-1 activation and maturation of IL-1β and IL-18 [47].
Additionally in vitro studies have shown that elevated levels of IL-1β and IL-18, produced during CKD, were able to promote renal tubulointerstitial fibrosis [48,49]. In particular, stimulation of primary cultures of human renal fibroblast with IL-1β for 24 h caused, collagen type 1 production and secretion of fibronectin and transforming growth factor-β (TGF-β).
Similarly, tubular proximal epithelial cells stimulated with IL-18 showed increase alphasmooth muscle actin (α-SMA) expression, type I collagen and fibronectin production in dosage-and/or time-dependent manners.
Additionally, in the current study, we found elevated production of mitochondrial ROS in PBMC from CKD-HD patients confirming mitochondrial impairment we previously reported [45]. In fact, using an innovative high-throughput technology, we discovered that several biological elements involved in the oxidative phosphorylation system (OXPHOS) and two key constituents of the mitochondrial complex IV (COXI and COXIV) were deregulated in CKD and HD patients compared to the healthy controls. Furthermore, complex IV activity, the terminal enzyme of the mitochondrial respiratory chain catalyzing the electron transfer from reduced cytochrome c to oxygen [57], resulted significantly lower in CKD-HD patients compared to healthy controls demonstrating a reduced activity of OXPHOS in this population.
It is conceivable that ROS produced by damaged organelles, in our patients' population, could be one of the abovementioned signal 1 in NLRP3 inflammasome activation.
In fact, in this paper, colocalization of NLRP3/ASC/mitochondria together with the inhibitory effect of mitoTEMPO (an inhibitor of mitochondrial ROS production) on CASP-1 maturation and cytokines activation in PBMC from CKD-HD patients confirm the involvement of this organelle.
This effect has been previously demonstrated through two biological/biochemical mechanisms that cause generation of ROS: inhibition of complex I or III of the mitochondrial respiratory chain and inhibition of mitophagy/autophagy resulting in the prolonged presence of damaged mitochondria [58][59][60].
Contrarily, Schreiber et al [61] have recently found ROS to be potent suppressors of inflammasome. In particular, Phox-generated ROS downregulate caspase 1, thereby keeping the inflammasome in check and limiting antineutrophil cytoplasmic antibody (ANCA)induced inflammation.
Notably, although extremely interesting, these unexpected data have been obtained, by a fashionable set of experiments, in a mouse model of ANCA-associated crescentic glomerulonephritis. However, because of the extreme complexity of the disease and the lack of etiopathogenetic information, no patients with this disease have been included in our study.
Moreover, microarray and principal component analysis, showed that 5 up to 15 gene probe sets were able to discriminate CKD-HD from NORM.
Interestingly, caspase recruitment domain-containing protein (CARD) 8 (also known as tumor CARD-containing antagonist of caspase nine-TUCAN), resulted down-regulated. This protein interacts physically with caspase-1 and negatively regulates caspase-1-dependent IL-1β expression [62,63]. A down-regulation of this protein has been reported in a model of human pancreatic islets isolated for transplantation exposed to oxidative stress. Authors suggested that an increased production of ROS and subsequent oxidative stress, through CARD8 downregulation, represents a possible mechanism by which high concentrations of glucose could kill beta cells [64].
However, limitations of this part of the study is the unusual bioinformatic analysis (because the low number of patients we did not perform false discovery rate evaluation) and the unbalance between patients and healthy subjects.
All together these data showed, for the first time, that NLRP3 inflammasome was activated in uremic patients undergoing dialysis treatment and they suggested that this unphysiological condition could be possibly induced by mitochondrial dysfunction.
Finally, NLRP3 inflammasome pathway could turn to be a valuable therapeutic target to minimize or avoid severe clinical complications in chronic kidney disease patients with advanced renal impairment.

Conclusions
In conclusion, the present study reveals, for the first time, that damaged mitochondria of uremic patients through an elevated production of ROS could be able to activate NLRP3 inflammasome representing a new deregulated biological machinery and a novel therapeutic target in CKD-HD patients.