Innate Immune Molecule Surfactant Protein D Attenuates Sepsis-induced Acute Pancreatic Injury through Modulating Apoptosis and NF-κB-mediated Inflammation

Sepsis causes multiple-organ dysfunction including pancreatic injury, thus resulting in high mortality. Innate immune molecule surfactant protein D (SP-D) plays a critical role in host defense and regulating inflammation of infectious diseases. In this study we investigated SP-D functions in the acute pancreatic injury (API) with C57BL/6 Wild-type (WT) and SP-D knockout (KO) mice in cecal ligation and puncture (CLP) model. Our results confirm SP-D expression in pancreatic islets and intercalated ducts and are the first to explore the role of pancreatic SP-D in sepsis. CLP decreased pancreatic SP-D levels and caused severe pancreatic injury with higher serum amylase 24 h after CLP. Apoptosis and neutrophil infiltration were increased in the pancreas of septic KO mice (p < 0.05, vs septic WT mice), with lower Bcl-2 and higher caspase-3 levels in septic KO mice (p < 0.05). Molecular analysis revealed increased NF-κB-p65 and phosphorylated IκB-α levels along with higher serum levels of TNF-α and IL-6 in septic KO mice compared to septic WT mice (p < 0.01). Furthermore, in vitro islet cultures stimulated with LPS produced higher TNF-α and IL-6 (p < 0.05) from KO mice compared to WT mice. Collectively, these results demonstrate SP-D plays protective roles by inhibiting apoptosis and modulating NF-κB-mediated inflammation in CLP-induced API.


Acute pancreatic injury in septic SP-D KO and WT mice. Previous studies have shown the pan-
creas is vulnerable to injury during sepsis 24 . Therefore, we examined histopathological changes in pancreatic tissues from septic SP-D KO, WT and sham mice. As shown in Fig. 2, there were obvious pathological changes in the septic SP-D KO and WT mice 24 h after CLP treatment, but not in control mice ( Fig. 2A). The pancreatic tissues in both septic KO and WT mice exhibited characteristic edema, inflammation and necrosis of the acinar cells, but more severe injury was found in the SP-D KO mice compared with WT mice ( Fig. 2A). Quantitative analysis of the pancreatic histopathological score revealed that septic SP-D KO mice was higher compared with septic WT mice 24 h after CLP (Fig. 2B, p < 0.01), suggesting that SP-D KO mice were more sensitive to sepsis-induced API.
Increased serum amylase activity in septic SP-D KO and WT mice. Serum amylase activity is a commonly used biomarker to detect pancreatic injury. Amylase activity was slightly increased at 6 h post-CLP in the WT than sham mice, but it was higher in the septic SP-D KO mice compared to sham and WT mice (p < 0.05) (Fig. 3). Amylase activity was dramatically increased at 24 h for septic SP-D KO and WT mice compared to sham mice (p < 0.01). Septic SP-D KO mice demonstrate a higher level of amylase activity than septic WT mice (p < 0.01) (Fig. 3).

Apoptosis of the pancreas in septic SP-D KO and WT mice.
To explore the role of SP-D in pancreatic cell apoptosis in sepsis-induced API, the slides of pancreas were examined by the TUNEL method. As shown in Fig. 4, a number of apoptotic cells (including the pancreatic acinar cells, intercalated ducts cells, as well as some types of cells in islets) with brown nucleus were observed in the pancreases of septic, but not in sham mice. Quantitative analysis of apoptotic cells demonstrates that the number of TUNEL-positive cells was higher in pancreas from septic SP-D KO mice compared with septic WT mice 24 h after CLP (Fig. 4B, p < 0.01).
To investigate potential mechanisms for SP-D effects in sepsis-induced apoptosis, two apoptosis-related biomarkers were measured by Western blotting analysis in the pancreatic tissues of septic and sham mice. As shown in Fig. 4C,D, the expression levels of Bcl-2 or caspase-3 showed differences between septic SP-D KO mice and septic WT mice. Bcl-2 is an inhibitor of apoptosis and was used as a negative biomarker. The expression level of Bcl-2 was decreased in septic compared with sham mice (p < 0.01), but the amount of Bcl-2 in septic SP-D KO mice was lower than septic WT mice 24 h post-CLP (Fig. 4C, Scientific RepoRts | 5:17798 | DOI: 10.1038/srep17798 p < 0.01). In contrast, the expression level of caspase-3 (17 kDa band as active form), a biomarker of apoptosis, was increased in septic mice compared with sham mice; and the level of caspase-3 was higher in septic SP-D KO mice than septic WT mice at 6 h (p < 0.05) and 24 h (p < 0.05) after CLP (Fig. 4D). These results provide evidence SP-D plays a role in inhibiting apoptosis during sepsis-induced API.

Neutrophil infiltration in the pancreas of septic SP-D KO and WT mice. Inflammatory cell infil-
tration is also an important mechanism in the pathogenesis of sepsis-induced API. To evaluate pancreatic neutrophil infiltration, we used neutrophil-specific antibody to perform IHC analysis of pancreatic tissue. The results show that pancreas from septic SP-D KO mice display more neutrophil positive cells than septic WT mice (Fig. 5A). Quantitative analysis of neutrophils by light microscopy indicates the number of neutrophils in pancreas from septic SP-D KO mice increased significantly compared with septic WT mice (Fig. 5B). These data indicate SP-D inhibits neutrophil infiltration into septic pancreases.

NF-κB activation in the pancreas of septic SP-D KO and WT mice. Previous studies by Yoshida
et al. 18 demonstrated SP-D is involved in the regulation of NF-κ B signaling pathway. Phosphorylation of Iκ B-α is required for initiation of NF-κ B activation. Therefore, the levels of phosphorylated Iκ B-α (p-Iκ B-α ) and NF-κ B p65 in the pancreas were examined by Western blotting analysis. As shown in Fig. 6, there were significant increases of the levels of p-Iκ B-α and NF-κ B p65 in septic mice 12 and 24 h after CLP compared with control animals (Fig. 6, p < 0.01). Importantly, the levels of p-Iκ B-α and NF-κ B p65 in the pancreas from septic SP-D KO mice were higher when compared with the pancreas Pro-inflammatory cytokine levels in the sera of septic SP-D KO and WT mice. To evaluate systemic inflammation, the levels of circulating pro-inflammatory cytokines IL-6 and TNF-α in the serum were measured by ELISA. As shown in Fig. 7A,B, CLP resulted in significant expression of IL-6 and TNF-α in the sera of septic mice at 6 h and 24 h after surgery compared with sham mice. Furthermore, the levels of IL-6 and TNF-α were higher in the sera of septic SP-D KO mice than in septic WT mice (Fig. 7A,B, p < 0.05).  Productions of pro-inflammatory cytokines in the cultured islets stimulated with LPS. In order to investigate the regulation of inflammation in pancreas by SP-D we isolated and cultured pancreatic islets from SP-D KO and WT mice. Cultured islets were stimulated by adding LPS (5 μ g/mL) into the media (Fig. 8A,B). The phenotypes of cultured islets showed some differences before (Fig. 8A) and after (Fig. 8B) LPS stimulation. Twenty four hrs after LPS stimulation most islets from both SP-D KO and WT mice became smaller cell aggregates and exhibited apoptotic cell fragments in the culture medium compared to untreated islets. Pro-inflammatory cytokines IL-6 and TNF-α were measured in the conditioned media from islet cultures. After 12 h and 24 h exposure to LPS (5 μ g/mL), the levels of IL-6 and TNF-α were markedly increased (Fig. 8C,D) than sham group, and the levels of IL-6 and TNF-α were higher in the conditioned medium of islets from SP-D KO mice compared to WT mice (Fig. 8C,D). These data suggested that SP-D can directly play a protective role in the local islet tissue to regulate pro-inflammatory cytokine production. Apoptotic cells with brown nucleus were observed (arrows). Apoptotic cells were quantified by 8 randomly high power fields (HPF) at x400 magnification. The number of apoptotic cells was larger in septic SP-D KO mice than septic WT mice 24 h after CLP (B). Furthermore, the levels of two apoptotic biomarkers (Bcl-2 and caspase-3) were analyzed in the pancreas of septic WT and SP-D KO mice by Western blotting method. β -actin was used as an internal control to normalize protein levels. Lower Bcl-2 level (C) and higher activated caspase-3 (17 kDa band) level (D) were observed in the pancreas of septic SP-D KO mice compared with septic WT mice 6 h and 24 h post-CLP. Graphs represent the mean ± SEM, *p < 0.05, **p < 0.01. Scale bars = 50 μ m. (n = 6 mice/group).

Discussion
Previous studies have demonstrated that SP-D is expressed in the epithelial cells of mucosa surfaces in humans 10 . SP-D expression was also detected in several other organs and tissues from fetuses, children, and adults 25 . We are among the groups to confirm SP-D expression in pancreas and to localize SP-D expression in the islets and the intercalated ducts of mouse pancreas 25,26 . Our IHC analysis clearly demonstrates SP-D expression in the islets with SP-D KO mouse control. While the exact role(s) of pancreatic SP-D are poorly defined, in pregnancy, increased pancreatic SP-D expression correlates with increased ß-cell proliferation 26 . SP-D is known to play a crucial role in lung surfactant homeostasis 27 , innate host defense and modulating inflammation during pulmonary infections 9,28 . However, the results of the recent study indicate SP-D has a protective role during sepsis-induced API. These findings are consistent with the anti-inflammatory effects of SP-D in described endothelial cells 29 . They are also consistent with anti-inflammatory role posited for SP-D in response to cytokine stimulation of pancreatic islets 26 . Furthermore, cellular and molecular analyses revealed the mechanisms of SP-D through  inhibiting pancreatic cell apoptosis and regulating the NF-κ B-mediated inflammation in sepsis-induced API.
In the present study, we found that CLP-induced sepsis caused remarkable pancreatic injury 24 h after treatment in the septic mice compared with sham mice. Characteristic edema, inflammation, and necrosis of the acinar cells were observed in the pancreas by histopathological analysis, and pancreatic injury was worse in septic SP-D KO mice compared with septic WT mice. Although variation of histological  changes in the pancreas of patients reported 30 , the results of this study clearly indicate increased susceptibility of SP-D KO mice to sepsis-induced API.
Serum amylase level is a biomarker which is commonly used to detect acute pancreatic injury 31,32 . Pancreatic blood supply can be affected earlier in sepsis than other organs 33 thus increasing the possibility of pancreatic injury. In our study, serum amylase level is increased at 6 h after CLP, and significant increase in amylase activity in the serum occurred at 24 h in both WT mice and SP-D KO mice, with higher activity in septic SP-D KO vs. septic WT mice. These demonstrated that CLP-induced sepsis resulted in severe pancreatic injury 24 h after CLP; and SP-D expression in the pancreas and other organs may protect from sepsis-induced API. In the present study, we observed decreased SP-D level in the pancreas in septic mice 6 and 24 h post-CLP. The observations might be explained due to apoptosis and necrosis of a larger number of SP-D expressing cells in the pancreas of septic mice. Decreased SP-D level in the pancreas further lost SP-D protective roles in the host defense and anti-inflammatory regulation, as well as apoptosis 34 . Indeed, SP-D protective effects were also observed in other organs, like kidney 23 and liver 35 , in the early stage of sepsis-induced dysfunction.
Apoptosis has been found in clinical and experimental API 36 . In the present study, a number of apoptotic cells were observed in the pancreas of septic mice at 24 h after CLP. In previous studies, mild acute pancreatitis was found to be associated with extensive apoptosis while severe acute pancreatitis was noted to involve extensive severe necrosis but with a little apoptosis 37,38 . It was also found that induction of apoptosis around the necrotic tissues can reduce the severity in severe AP 39 . This study has observed only moderate necrosis in sepsis-induced acute pancreatic injury of both SP-D KO and WT mice but many apoptotic cells and increased apoptosis-related biomarkers were found. Moreover, septic SP-D KO mice showed more severe pancreatic injury compared to septic WT mice. Therefore, these indicate excessive apoptosis may be harmful to the function of the pancreas. Apoptosis is a form of programmed cell death, it is the consequence of several interlinked intracellular caspase proteins 40,41 . Caspase proteins belong to a family of intracellular cysteine protease, they play important roles in the initiation and execution of apoptosis. Both Bcl-2 and caspase-3 are important regulatory proteins of apoptosis. Bcl-2 acts as an inhibitor of apoptosis, overexpression of Bcl-2 will inhibit apoptosis. Caspase-3 is an executor of apoptosis, overexpression of caspase-3 will accelerate apoptosis 42,43 . In this study, apoptotic cells were identified in the pancreas evidenced by the changed levels of apoptosis regulatory proteins in the pancreas. These indicate that apoptosis of pancreatic cells occurred at the early phase of API in this study, which is consistent with other studies 36,44 . In addition, we observed that septic SP-D KO mice showed more severe apoptosis compared with septic WT mice, suggesting that SP-D had a protective effect to pancreatic apoptosis in sepsis-induced API. These findings were consistent with protective effects of SP-D against renal apoptosis in sepsis-induced acute kidney injury (AKI) 23 .
Nuclear factor-kappa B (NF-κ B) is a transcription factor that regulates the expression of many inflammatory genes in various infectious diseases 45 . In the pancreas, NF-κ B is a p65/p50 heterodimer as a predominant form 46 . Under normal conditions, NF-κ B dimers are blocked nuclear localization by inhibitory proteins (Iκ Bs) and are trapped in the cytoplasm as inactivated forms. When cells are stimulated with various factors, the inhibitory proteins are phosphorylated, which allows the NF-κ B to translocate into the nucleus, where it binds to relevant consensus sequence of NF-κ B target genes, rendering NF-κ B activation 47 . Activated NF-κ B can induce many genes transcription, including inflammatory and apoptotic responses 48 . In this study, both NF-κ B p65 and p-Iκ B-α levels in the pancreas were significantly increased in the septic compared to sham mice 6 h and 24 h post-CLP, but septic SP-D KO mice have more NF-κ B and p-Iκ B-α expression than septic WT mice. These data suggest that SP-D could modulate NF-κ B activation during the early phase of sepsis-induced API. Spontaneous activation of NF-κ B has been observed in alveolar macrophages of SP-D-null mice 18 and SP-D acts as an important modulator of pulmonary inflammation 49 . Recent studies suggested increased caspase-3 expression is positively associated with the NF-κ B p65 level in kidney tissues from septic patients 50 . Previous studies have shown that SP-D can bind to TLR4 51 and inhibit TLR4-mediated pro-inflammatory responses 52 . The loss of functional TLR4 lead to a blunted NF-κ B response, reduced pro-inflammatory mediator production, and decreased sensitivity to hyperoxia in mice 53 . Collectively, the findings from this and previous studies provide solid evidences that SP-D can ameliorate sepsis-induced API through regulating NF-κ B-mediated signaling pathway and inhibiting pancreatic apoptosis.
Inflammatory cell infiltration may also play a key role in the pathogenesis of sepsis-induced API. Neutrophil infiltration contribute to parenchymal cell injury and vascular dysfunction during liver injury 54 . Our IHC studies reveal intensive neutrophils in the interstitial capillaries and the pancreatic tissue in septic WT and SP-D KO mice. In contrast to septic WT, neutrophil infiltration increased significantly in the septic SP-D KO mice. Previous work found that SP-D KO mice were more susceptible to LPS-induced acute lung injury compared to WT mice, and intratracheal instilled SP-D can inhibit lung inflammation 55 .
Pro-inflammatory cytokines are important mediators in the pathogenesis of API 56 . As expected, in the present study CLP sepsis induced a serious systemic inflammatory response with significant increase of both pro-inflammatory cytokines TNF-α and IL-6 in the serum 6 h and 24 h after CLP. These data are supported by the increased NF-κ B and p-Iκ B-α levels in the pancreas of septic WT and SP-D KO mice. Our findings are consistent with previous studies that demonstrated increased TNF-α and IL-6 in the serum during polymicrobial sepsis by CLP 57 cultures stimulated by LPS demonstrated SP-D function in the modulation of inflammatory cytokine production in the islet cells.
In summary, our results demonstrate SP-D expression in pancreatic islets and intercalated duct. SP-D can decrease pro-inflammatory cytokine production and amylase activity in the serum, and inhibits pancreatic apoptosis thus attenuating API in the mouse sepsis model. The cellular and molecular mechanisms include SP-D inhibition of apoptosis and modulation of NF-κ B-mediated inflammation in the sepsis-induced API.

Materials and Methods
Animals. All of the experiments in this study were approved by Institutional Animal Care and Use Committee at SUNY Upstate Medical University (IACUC #270) and meet the National Institutes of Health and ARRIVE guidelines on the use of laboratory animals. The original SP-D KO mice (C57BL/6 background) were kindly provided by Dr. Hawgood's laboratory of the University of California, San Francisco. The SP-D KO mice were backcrossed at least 10 generations with C57BL/6 background mice 59,60 . All SP-D KO mice used in this study were bred in the animal core facility at SUNY Upstate Medical University 61 . Age-matched C57BL/6 WT mice were purchased from Jackson Laboratories (Bar Harbor, ME). There were no significant difference in phenotype between SP-D KO and matched WT mice. Age, sex and background-matched mice were housed in a temperature-controlled room at 24 °C under the pathogen-free conditions, maintained on water and food ad libitum before the study. Both male and female mice are used in this study.
Pancreatic islet isolation and culture. Islets from C57BL/6 WT and SP-D KO mice were isolated by collagenase digestion as described previously 62 . In brief, the pancreas of mouse was distended by injecting 3 ml collagenase VI solution (0.5 mg/ml)(Sigma, St. Louis. MO), and then digested in a water bath at 37 o C for 15 min. The islets were washed with Hank's solution, the collected islets were cultured in RPMI-1640 (Gibco, Grand island, NY) containing 10% FBS, 100 U/ml streptomycin and penicillin. Isolated islet cultures were exposed for 12 h and 24 h in the presence or the absence of LPS (5 μ g/mL). CLP model. Cecal ligation and puncture (CLP) was performed to induce polymicrobial sepsis as previously described 23,63 . In brief, mice (8-12 weeks) were anesthetized with intraperitoneal ketamine/xylazine (90 mg/kg ketamine, 10 mg/kg xylazine) injection. After the indicative of anesthsia, under aseptic conditions, a ventral midline incision (1-cm) was performed, the cecum was exposed, then ligated with a 5-0 silk suture at 1.3 cm position from distal to the base of the cecum, but avoiding intestinal obstruction. The cecum was punctured twice using a 22-gauge needle at top and bottom, respectively. The punctured cecum was squeezed to extrude about 1-mm 3 droplet of fecal material from the perforation sites and returned to the peritoneal cavity. The incision was closed in layers with 5-0 surgical sutures. Immediately after the procedure, mice were injected with prewarmed sterile saline (1 ml) subcutaneously to replace fluid lost. Pain medication was injected every 12 hrs after surgery (buprenorphine 0.05 mg/kg). For sham animals, mice underwent the same procedure, but no CLP.
Tissue harvesting. Mice were anesthetized, and sacrificed at 6 h and 24 h after CLP or sham surgery as previously described 23,63 . Blood samples were collected from the inferior vena cava by means of 1-ml syringe, and plasma was collected after centrifugation (4000 xg) at 4 o C for 5 min. Pancreatic tissues were removed and immediately frozen in liquid nitrogen or fixed in 10% formalin for further study. Cultured Islets treated with or without LPS (5 μ g/ml) were collected by centrifugation (15000 x g, 5 min) at 4 o C, and immediately frozen at − 20 o C.
Histopathological analysis. Formalin-fixed pancreatic tissues were embedded in paraffin and sectioned at about 4-μ m for routine histology from six mice for each group 23 . Slides were stained with hematoxylin and eosin (H&E). Histopathology of the pancreatic injury was qualitatively evaluated by two experienced investigators who were blinded as to the experimental groups. Pancreatic histological grading was scored according to the method described by Demols et al. 64 .
Biochemical measurement. Serum amylase (AMY) activity was measured to monitor the pancreatic injury by using a commercial kit (Sigma-Aldrich, St. Louis. MO). Levels of AMY were determined according to the manufacturer's instructions.

Apoptotic cell determination.
A TUNEL kit (Roche, Indianapolis, IN) was used to detect apoptosis in pancreatic cells from septic and sham mice as the described previously 23 . In brief, sections from paraffin-embedded pancreas were incubated at 60 °C for 20 min, then deparaffinized and rehydrated. Sections were digested for 10 minutes with 2 μ g/mL proteinase K, washed with PBS, incubated in 25 mmol/L cobalt chloride and so on according to the manufacturer's protocol. Apoptotic cells were quantified at high-power field (x 400 magnification) by Nikon ECLIPSE TE2000-U optical microscopy.
Immunohistochemistry (IHC). The parrffin-embedded pancreas specimens were sectioned about 4μ m for (IHC). Briefly, sections were de-waxed and rehydrated and put in 3% H 2 O 2 to block endogenous peroxidase activity for 10 minutes. Epitope retrieval was carried out by boiling in 10 mM citrate buffer (pH = 6.0) for 20 min. After the sections were incubated with 10% goat serum for 40 minutes at room temperature, they were added with SP-D antibody (1:500, Santa Cruz Biotechnology, Dallas, Texas) or Anti-Mouse neutrophils antibody (1:100, Hycult Biotech, the Netherlands) at 4 °C overnight. Subsequently, the sections were applied biotinylated secondary antibody (Vector Laboratories, Burlingame, CA) at 1:2000 dilution and incubated in a humidified chamber at 37 °C for 30 minutes. DAB was used as a chromogenic substrate to incubate for 3-10 minutes, and then the sections were counterstained for 1-3 minutes with hematoxylin. Slides were viewed by an Olympus microscope (Olympus AG, Zurich, Switzerland). The positive cells were counted at high-power field ( x 200 magnification).
Western blotting analysis. Frozen pancreatic tissues were homogenized in RIPA cell lysis buffer containing PMSF (Thermo Scientific Pierce, Rockford, Illinois) and cocktail of protease inhibitors (Roche, Indianapolis, IN), as well as aprotinin (Sigma-Aldrich, St. Louis, MO). All samples were centrifuged at 14,000xg for 10 min. Supernatant was recovered, and protein concentration of the lysate was determined by using a bicinchoninic acid (BCA) protein assay kit (Thermo Scientific, Rockford, IL) 65 . Total protein (100 mg) for each sample were resolved by reducing, electrophoresed on 12% SDS-PAGE gel electrophoresis and transferred onto PVDF membranes (Bio-Rad, Hercules, USA). The membrane was soaked in Tris-buffered saline (TBS) containing 5% defatted milk for 1 hour, and was then incubated with a primary antibody against SP-D (Santa Cruz Biotechnology, Dallas, Texas), or caspase-3 (Santa Cruz Biotechnology, Dallas, Texas), or Bcl-2 (Santa Cruz Biotechnology, Dallas, Texas), or p-IκB-α (Santa Cruz Biotechnology, Dallas, Texas), or NF-κ B p65 (Santa Cruz Biotechnology, Dallas, Texas) at 4 °C overnight. A β -actin antibody (Santa Cruz Biotechnology, Dallas, Texas) was used as internal control. After washing the membranes, a HRP-conjugated secondary antibody (Bio-Rad, Hercules, CA) was applied and incubated at room temperature for 1 hour. Washed again, the blots were exposed to X-ray film (Pierce Biochemicals, FL) for 2-3 min. The relative expression of protein was quantified after scanning of the films by the Quantity One software (Bio-Rad, Hercules, CA).
Cytokine determination in the serum and the conditioned media from islet cultures. The concentrations of IL-6 and TNF-α in serum and conditioned medium of islet cell cultures were measured using commercially available murine enzyme-linked immunosorbent assay (ELISA) kits in accordance with the manufacturer's instructions (Life Technologies, Frederick, MD) 23 .

Statistical analysis.
All the data are expressed as mean ± SEM. Results were analyzed by one-way analysis of variable (ANOVA) and Tukey's post-hoc tests with the Sigmastat software version 3.5 (Jandel Scientific, CA). If P values were less than 0.05, differences were considered as statistically significant.