Role of CARD9 in inflammatory signal pathway of peritoneal macrophages in severe acute pancreatitis

Abstract Previous studies revealed that caspase recruitment domain protein 9 (CARD9) was involved in severe acute pancreatitis (SAP) inflammation and that interfering with its expression in vivo could inhibit inflammation. However, the specific mechanism is unknown. This study aimed to discover the related signal pathways of CARD9 in macrophages. SiRNA interference technology was used in vivo and in vitro to detect CARD9‐related signal pathways in peritoneal macrophages. Furthermore, Toll‐like receptor 4 (TLR4) and membrane‐associated C‐type lectin‐1 (Dectin‐1) pathways in macrophages were activated specially to looking for the upstream signal path of CARD9. Results showed up‐regulation of CARD9 expression in peritoneal macrophages of SAP rats (P < .05). CARD9 siRNA alleviated inflammatory cytokines, and inhibited the phosphorylation of NF‐κB and p38MAPK in peritoneal macrophages in vivo or in vitro. Meanwhile, CARD9 siRNA reduced the concentration of CARD9 and Bcl10 in peritoneal macrophages, and TLR4 and Dectin‐1 took part in CARD9 signal pathways in macrophages. In conclusion, there is an inflammation signal pathway comprised of TLR4/Dectin‐1‐CARD9‐NF‐κB/p38MAPK activated in macrophages in SAP. Blockade of CARD9 expression in macrophages can effectively alleviate SAP inflammation.

Caspase recruitment domain protein 9 (CARD9) is highly expressed in macrophages. 4 It is an important adaptor protein, which can efficiently integrate the recognition signals of various innate immune receptors, and regulate the intracellular signal transmission to play an important role in the innate immunity. 5 Recently, articles about CARD9 have mainly focused on infectious inflammation, [6][7][8][9] but there have been no researches reporting on the relationship between CARD9 and SAP before our studies.
Previous studies by our group revealed that CARD9 in peripheral blood monocytes of patients with SAP increased significantly. 10 Furthermore, in our animal experiments, the mRNA and protein level of CARD9 in pancreatic tissue is higher in SAP rats, while adenovirus interfered with the expression of CARD9, which effectively alleviated the inflammatory response of pancreatic tissue. 11 Therefore, these previous studies demonstrated that CARD9 was involved in SAP inflammation and that interfering with its expression could inhibit inflammation. However, some problems still remain unsolved.
This study aimed to discover the related signal pathway of CARD9 in macrophages in vivo and in vitro via siRNA interference technology, and to explore the possible molecular mechanisms of CARD9 in SAP.
Standard deviation rats were divided into four groups: Control group (Control, n = 6), rats underwent a sham operation with nothing infused; SAP group (SAP, n = 18), 5.0% sodium taurocholate was retrograde perfusion into the pancreatic and biliary duct, the dosage of sodium taurocholate is 1 mL/kg bodyweight; siCARD9 + SAP group (siCARD9, n = 18), 1 × 10 9 PFUs of CARD9 siRNA (dissolved in 200 μL PBS) were injected into caudal vein, and SAP model was induced in the same manner as the SAP group 48 hours later; sicontrol + SAP group (sicontrol, n = 18), the same manner as the (siCARD9 + SAP) group, control siRNA instead of CARD9 siRNA. Following the treatments, rats were fast and free to drink water. Peripheral blood, ascitic fluid and pancreatic tissue were collected 3, 6 and 12 hours after modelling. Experimental animal part was approved by the Ethical and Research Committee of Shanghai Jiao Tong University, and was designed and implemented in strict accordance with 'the guidelines for the care and use of experimental animals in research'.

| Isolation and culture of peritoneal macrophages
At 3, 6 and 12 hours after the establishment of the model, peritoneal effusions from the SAP rat groups were extracted under an aseptic environment, and peritoneal lavage was performed with 25 mL precooled PBS two consecutive times. The control group rats were given peritoneal lavage directly. The peritoneal lavage fluid and ascites were mixed evenly and the peritoneal macrophages were separated by density gradient centrifugation at 4°C, centrifuged at 500 × g for 15 minutes, then at 120 × g for 3 minutes. Mixed precipitated cells were mixed with DMEM complete medium (10% foetal bovine serum, penicillin and streptomycin) and evenly distributed in 12-well plates. The plates were incubated under 5% CO 2 at 37°C. After 3 hours, the macrophages were washed twice with PBS to remove unattached cells. After 24 hours of incubation, RNA and protein were collected from cells lysed in Trizol and RIPA for real-time PCR and Western blot detection, respectively.

| Inhibiting CARD9 expression in vitro
The purified peritoneal macrophages were cultured for 24 hours and then mixed with adenoviruses expressing siRNA against CARD9. The concentration of the original viral solution was 1 × 10 11 PFU/mL. The Rats were randomly divided into two groups: Normal group (n = 6), rats without any treatment, and SAP model group (n = 36), using the previously described method for establishing SAP. Isolation and purification of peritoneal macrophages were performed as described above 6 hours after establishment of the model. After

| Polyinosine andβ-glucan stimulation of primary peritoneal macrophages
The rats were randomly divided into two groups: Normal group (n = 18), rats without any treatment, and the SAP model group (n = 30), using the previously described method for establishing SAP.

| Histological examinations
The pancreatic tissue was fixed with 10%formalin and made into wax block, then sectioned and stained with haematoxylin and eosin (H&E). The histological changes of the pancreatic tissue were observed by the pathologist under light microscope, and evaluated from the four aspects of inflammation, oedema, haemorrhage and necrosis. The score of each aspect was from 0 to 4 according to the severity, and the total score was pathological score. 11

| Real-time PCR
The total RNA was extracted from 1 × 10 6 fresh adherent peritoneal macrophages by Trizol one-step method in strict accordance with the instructions of the reagent (TaKaRa Bio Inc.). The mRNA expression of Toll-like receptor 4 (TLR4), membrane-associated C-type lectin-1 (Dectin-1), CARD9, TNF-α, IL-1β and IL-6 mRNA in peritoneal macrophages were detected by two-step PCR reaction procedure. GAPDH was used as internal reference gene for calculation. The gene-specific primers sequence in Table 1. The amplification conditions was: first, pre denaturation: 95°C for 10 minutes; second, PCR reaction: 39 cycles of 95°C for 15 seconds, 60°C for 45 seconds; the third step is decomposition stage: 95° 15 seconds + 60° 1 minute + 95° 15 seconds, collecting the fluorescence of each PCR extension period. 2 −△△Ct was used to express the relative level of related genes.

Sense
Antisense TA B L E 1 The gene-specific primers

| Immunoprecipitation
Interaction between CARD9 and B-cell lymphoma factor 10 (Bcl10) in peritoneal macrophages was detected by immunoprecipitation according to the manufacturer's instructions. Briefly, peritoneal macrophages (2 × 10 6 ) from the animal models were lysed in 500 μL cold RIPA buffer. Lysates were pre-cleared with uncoated Dynabeads (Invitrogen). Anti CARD9 (Santa Cruz, CA, USA) and BCL-10 antibody (CST) together with protein A/G agarose were used to immunoprecipitate BCL-10 and CARD9, respectively. Then, proteins were resolved by SDS-PAGE and analysed by Western blot to detect whether there was interaction between BCL-10 and CARD9. 13

| Electrophoretic Mobility Shift Assay (EMSA)
EMSA was to detect the transcription activity of NFκB.

| Statistical analysis
Results are presented as means ± SD. The statistical significance was assessed through a one-way analysis of variance followed by SNK using SPSS 21.0 software (IBM). All tests were two-tailed, and differences were considered significant at P < .05.

| Up-regulation of CARD9 expression in macrophages of SAP rats
Compared with the control group, the mRNA level of CARD9 in peritoneal macrophages in the SAP group increased significantly at 3, 6 and 12 hours after induction of SAP, and was highest at 12 hours (P < .05) (Figure 1). The expression of CARD9 protein in peritoneal macrophages from SAP rats was also higher than that in control group (P < .05) ( Figure 1). These data showed that CARD9 participates in the inflammatory responses. ( Figure 1). Control siRNA rats showed no differences in CARD9 expression compared to SAP rats (Figure 1). The results suggested that CARD9 gene silencing with siRNA by caudal vein injection could significantly inhibit CARD9 expression in peritoneal macrophages.
After extraction of peritoneal macrophages from the siCARD9 adenovirus-treated group and the sicontrol group, the expression of CARD9 mRNA and protein in the siCARD9 group was lower than in the SAP and sicontrol groups. Meanwhile, the expression of CARD9 in SAP and sicontrol groups was higher than in the blank group.
Furthermore, the siCARD9 group showed a reduction of up to 60% in CARD9 mRNA expression ( Figure 2B).

| CARD9 siRNA alleviates pancreatitis severity in SAP rats and inhibits inflammatory cytokines in peritoneal macrophages in vivo and in vitro
In our preliminary work, results showed that compared to the SAP group, CARD9 gene silencing caused a reduction in ascites volume and obvious amelioration of pancreatic injury. 11 In this study, as shown in Figure 3A we got the same results in pathological changes and pathological scores of pancreas in (siCARD9 + SAP) group at 6 and 12 hours.
It indicates that the moulding is successful and interfering with the expression of CARD9 in peritoneal macrophages in vivo can effectively reduce the inflammatory response of pancreatic tissue. TNF-α, IL-6 and IL-1β levels were correlated with the severity of SAP and pancreatic injury. In vivo, the mRNA level of these inflammatory factors in peritoneal macrophages in SAP group was higher than that in control group and in (siCARD9 + SAP) group at 6 and 12 hours. There was no significant difference between the SAP group and sicontrol group ( Figure 3B). In vitro, as shown in Figure 4A-C, the levels of TNF-α, IL-1β and IL-6 in cell culture medium from the si-CARD9 group were significantly lower than in the SAP and sicontrol groups. Meanwhile, the levels of TNF-α, IL-1β and IL-6 in the latter two groups were significantly higher than in the blank group. There were statistically significant differences in all comparisons (P < .05), especially for IL-1β and IL-6 at 6 hours (P < .01). These data confirming that CARD9 siRNA could reduce the secretion of inflammatory factors in peritoneal macrophages.

| Reduction of CARD9 inhibits phosphorylation of NF-κB and P38 in peritoneal macrophages in vivo and in vitro
To investigate whether CARD9 can regulate the signalling pathways of NF-κB and p38MAPK in peritoneal macrophages, we applied siRNA interference technology in vivo and in vitro. In vivo, as shown in Figure 5A, the phosphorylation of NF-κB and P38 in peritoneal macrophages was increased in SAP rats compared with control rats, but decreased in (siCARD9 + SAP) group compared with SAP group, which was consistent with the expression of CARD9 (P < .05). In vitro, as shown in Figure 4D, the phosphorylation of NF-κB and P38MAPK in peritoneal macrophages of each group also changed in coordination with cytokine levels. The phosphorylation of NF-κB and P38MAPK in peritoneal macrophages in the siCARD9 group was significantly lower than in the SAP and sicontrol groups (P < .05).
Thus, the data demonstrated that CARD9 acts as a activator in inducing phosphorylation of NF-κB and P38MAPK in peritoneal macrophages in SAP.

| Reduction of CARD9 inhibits transcriptional activity of NF-κB in peritoneal macrophages in vivo
To explore the effect of CARD9 on the transcriptional activity of NF-κB, we applied EMSA technology. As shown in Figure 3B Therefore, These results were interpreted to indicate that CARD9 regulates the transcriptional activity of NF-κB in peritoneal macrophages in SAP.

| Reduction of CARD9 reduces the concentration of CARD9 and Bcl10 in peritoneal macrophages of SAP rats
It has been reported that CARD9 binds to two proteins containing the CARD domain in macrophages, B-cell lymphoma factor 10 (Bcl10) and mucosa-associated tissue lymphoma translocation protein (MALT1), forming a CARD9/BCL10/MALT1 complex (CBM complex). CARD9 activates inflammatory signalling pathways, such as NF-κB and MAPK in the form of a CBM complex to participate in fungal, bacterial and viral reactions. 5 To determine if an interaction between CARD9 and Bcl10 occurred in peritoneal macrophages of SAP rats, we utilized immunoprecipitation assays. As shown in Figure 6A, the interaction of CARD9 and Bcl10 could be detected in the control group, SAP group, sicontrol group and siCard9 group.
The formation of the complex was greater in the SAP group and sicontrol group, lower in siCard9 group, which was consistent with the phosphorylation of NF-κB and P38 and transcriptional activity of NF-κB. These data clearly revealed that Card9 acts on downstream signalling by forming a CARD9/Bcl10 complex in peritoneal macrophages of SAP rats.

| The expression of TLR4 and Dectin1 increased in peritoneal macrophages of SAP rats
Several lines of evidence have been published showing that TLR4 and Dectin-1 are involved in CARD9 signal transduction in fungal diseases. In order to find the upstream signal of CARD9 in peritoneal macrophages of SAP rats, we detected the expression of TLR4 and Dectin-1. As shown in Figure 6B, protein expression of TLR4 and Dectin-1 in peritoneal macrophages of SAP rats was significantly greater than in the control group at 3 hours and remained high at 12 hours (P < .05). These data suggested that TLR4 and Dectin-1 are involved in peritoneal macrophage inflammation in SAP.

| TLR4 and Dectin-1 pathways in macrophages were activated by stimulation with polyinosine and β-glucan, respectively
In order to clarify the relationship between TLR4 and CARD9 in SAP peritoneal macrophages, we used polyinosine to specifically activate the TLR4 pathway of normal peritoneal macrophages and F I G U R E 5 In vivo CARD9 siRNA alleviates Inflammatory reactions in peritoneal macrophages. A, Phosphorylation of NF-κB and P38 in peritoneal macrophages from each group at the 3-h and 12-h time points. The molecular weight of the NF-κBp65 and P38 bands were 65 kD and 43 kD. B, Transcriptional activity of NF-κB in peritoneal macrophages from each group. #: SAP group vs siCARD9 group, P < .05 peritoneal macrophages treated with the siCard9 adenovirus. As shown in Figure 7, compared with the blank group, the expression of TLR4 was enhanced after polyinosine treatment. Meanwhile, the levels of TNF-α, IL-1β and IL-6, the phosphorylation of NF-κB and p38MAPK, and the expression of CARD9 were also increased (P < .05). the expression of TNF-α, IL-1β, IL-6 and the phosphorylation of NF-κB and p38MAPK in SAP rat peritoneal macrophages treated with the siCard9 adenovirus in vitro was lower than that in the SAP group and Poly group (P < .05).These data suggested that TLR4 is the upstream of CARD9 and could regulate the expression of CARD9, there may exist a TLR4-CARD9-NF-κB/p38MAPK signalling pathway in peritoneal macrophages in SAP. In addition, we found that the expression of TLR4 was lower in the siCARD9 group in vitro, consistent with the results of in vivo ( Figure 6B). We speculated that CARD9 could feedback regulate the expression of TLR4, and there may be exit a TLR4/-CARD9-NF-κB/p38MAPK signalling network in peritoneal macrophages in SAP. We then used β-glucan to specifically activate the Dectin-1 pathway in normal peritoneal macrophages and peritoneal macrophages treated with the siCard9 adenovirus. The results showed that the changes were the same as those observed after polyinosine stimulation ( Figure S1).

| D ISCUSS I ON
Our study detected a significant increase in the expression of  However, the relationship between TLR4 and CARD9 has not been plays an important role in natural immunity, which can resist fungal pathogens and bacterial (especially Mycobacterium) infection. [23][24][25] Evidence has shown that CARD9 is mainly involved in activation of the Dectin-1-induced NF-κB signalling pathway for innate anti-fungal immunity. 26,27 However, there has been no research reported to our knowledge about the relationship between CARD9 and Dectin-1 in the early stage of SAP under conditions of aseptic inflammation.
In our study, wu found that Dectin-1 is a participant in the activation of CARD9 in SAP, there may exist a Dectin-1-CARD9-NF-κB/ p38MAPK pathway.
In addition, we also found that the expression of TLR4 and Dectin-1 decreased after inhibiting CARD9 expression via RNA interference in SAP peritoneal macrophages in vitro, suggesting that the regulation of TLR4 and Dectin-1 on CARD9 is not unidirectional; rather, peritoneal macrophages may form an inflammation signalling network, which regulates PPRs. CARD9 may play an important role in this signalling network.

| CON CLUS ION
Early diagnosis and treatment of SAP patients are essential to reduce mortality and control disease exacerbation. The role of CARD9 in SAP peritoneal macrophages was verified repeatedly by animal and cell experiments. An inflammation signalling pathway comprised of TLR4/Dectin-1-CARD9-NF-κB/p38MAPK is likely active in peritoneal macrophages, in which CARD9 plays an important role.
Blockade of CARD9 expression can effectively alleviate SAP inflammation, and thus, may be an important therapeutic target for SAP treatment.

CO N FLI C T O F I NTE R E S T
No conflicts of interest exist.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data models are available from the corresponding author by request.