Anti-inflammatory and anti-necrotic effects of lectins from Canavalia ensiformis and Canavalia brasiliensis in experimental acute pancreatitis

Lectins isolated from Canavalia ensiformis (ConA) and Canavalia brasiliensis (ConBr) are promising molecules to prevent cell death. Acute pancreatitis, characterized by acinar cell necrosis and inflammation, presents significant morbidity and mortality. This study has investigated the effects of ConA and ConBr in experimental acute pancreatitis and pancreatic acinar cell death induced by bile acid. Pancreatitis was induced by retrograde pancreatic ductal injection of 3% sodium taurocholate (Na-TC) in male Swiss mice. ConA or ConBr (0.1, 1 or 10 mg/kg) were intravenously applied to mice 1 h and 12 h after induction. After 24 h, the severity of pancreatitis was evaluated by serum amylase and lipase, histopathological changes and myeloperoxidase assay. Pancreatic acinar cells were incubated with ConA (200 µg/ml) or ConBr (200 µg/ml) and taurolithocholic acid 3-sulfate (TLCS; 500 µM). Necrosis and changes in mitochondrial membrane potential (ΔѰm) were detected by fluorescence confocal microscopy. Treatment (post-insult) with ConA and ConBr decreased pancreatic damage caused by retrograde injection of Na-TC in mice, reducing pancreatic neutrophil infiltration, edema and necrosis. In addition, ConA and ConBr decreased pancreatic acinar cell necrosis and depolarization of ΔѰm caused by TLCS. The inhibition of necrosis was prevented by the lectin domain blockade. In conclusion, ConA and ConBr markedly inhibited in vitro and in vivo damage, effects partly dependent on the interaction with mannose residues on acinar cells. These data support the potential application of these proteins for treatment of acute pancreatitis.


Introduction
Acute pancreatitis is a potentially severe inflammatory disease of the exocrine pancreas presenting significant morbidity and mortality. The most common cause is cholelithiasis, accounting for 35%-60% of cases [1]. Acute biliary pancreatitis is characterized by elevated pancreatic amylase and lipase, tissue necrosis and severe pancreatic inflammation, which may lead to a systemic inflammatory response syndrome, multiple organ dysfunction and death [2].
According to the revised 2012 Atlanta Classification of Acute Pancreatitis, the presence of necrosis and the number of organs affected by the subsequent inflammatory response determines the disease severity (mild, moderate, severe) and dictates the short-term and long-term management of the patients [3]. The severity of experimental pancreatitis is correlated with the extent of cell necrosis and inversely with apoptosis [4][5][6].
In isolated pancreatic acinar cells, bile acids such as taurolithocholic acid 3-sulfate (TLC-S), cause necrosis due to a sustained increase of cytosolic calcium ([Ca2 +]C) via activation of the Gpbar1 G-protein receptor [7] Ca2 + release from intracellular stores and inhibition of the sarco-endoplasmic reticulum Ca2 + -adenosine triphosphatase (SERCA) pump [8] leading to premature intracellular enzyme activation, cell vacuolization, mitochondrial depolarization and decrease in ATP production [9][10][11]. However, currently there is no specific therapy to treat the disease.
Leguminous lectins of the Diocleinae subtribe, a class of carbohydrate-binding proteins, have received increasing attention because of their various biological properties [12]. ConA and ConBr, isolated from the seeds of Canavalia ensiformis and Canavalia brasiliensis, possess binding specificity to residues of glucose/mannose and share 90% structural similarity, and are considered promising molecules since they possess a variety of biological functions, including neuroplasticity [13][14][15][16][17], modulation of cell death, and antiproliferative effects on tumor cells [18,19]. They promote apoptosis in human leukemic cells lines, in a mitochondrialdependent manner, while preserving healthy cells [20]. In addition, these lectins exert immunomodulatory and antinociceptive properties in experimental models in vivo [21][22][23][24].
However, the potential of these lectins to prevent cell necrosis and inflammation in acute pancreatitis has not been fully investigated. This study aimed to evaluate the effects of ConA and ConBr in the mice experimental acute biliary pancreatitis induced by taurolithocholic acid 3-sulphate (TLC-S) and assess the involvement of the lectin-carbohydrate interaction.

Lectin purification and dissolution
The lectins were purified from seeds of the leguminous plants Canavalia ensiformis (L.) DC. (Jack bean) (ConA) and Canavalia brasiliensis MART (ConBr) by affinity chromatography [25,26]) both possessing binding-specificity to D-glucose and D-mannose. After purification, the lectins were dialyzed in distilled water, lyophilized and dissolved in 0.9% NaCl (sterile saline) immediately prior to use.

Animals
Male Swiss mice (20-25 g), provided by the Animal Housing Facility of the Department of Physiology and Pharmacology of the Medical School of the Federal University of Ceará, were kept in ventilated racks maintained at 23 ± 2 °C, with a 12 h light/dark cycle and free access to a balanced diet (PURINA ® ) and purified water. All procedures were in accordance with the guidelines of the National Council for the Control of Animal Experimentation (CONCEA) and approved by the Animal Use Ethics Committee (CEUA) of the Federal University of Ceará (protocol n° 99/2013).

Experimental acute pancreatitis
Acute pancreatitis was induced by retrograde pancreatic ductal injection of 3% Na-TC (5 μL/min for 10 min via infusion pump, survival rate was 80% on average) in adult mice (20-25 g). The control groups received retrograde pancreatic ductal infusion of saline (Sal Group) alone or the surgical procedure without infusion, denominated Sham group [27], ConA or ConBr (0.1, 1 or 10 mg/kg i.v.) were applied as treatment to mice 1 h and 12 h after pancreatitis induction. After 24 h, the animals were sacrificed, and the blood was collected for estimation of pancreatic enzymes amylase and lipase and the pancreas removed for myeloperoxidase assay (MPO) and histopathological analysis.

Histopathological evaluation
The pancreas was fixed in 10% formalin, embedded in paraffin by standard methods, cut into sections of 5 μm through a microtome, and subsequently stained with hematoxylin-eosin (HE). The morphological alterations characteristic of acute pancreatitis were evaluated as previously [28]: edema, inflammatory infiltration and necrosis. The degree of edema was determined using a scale of 0 to 3 (0 = absent, 1 = interlobular edema, 2 = interlobular edema and moderate intralobular edema, and 3 = interlobular edema and severe intralobular edema). The presence of inflammatory infiltrate was also analyzed, using a grid of 0 to 3 (0 = absent 1 = scarce perivascular infiltration, 2 = moderate perivascular infiltration and scarce diffuse infiltration, and 3 = abundant diffuse infiltration). Parenchymal necrosis was analyzed by assigning scores from 0 to 3 (0 = absent, 1 = less than 15% of pancreatic cells involved, 2 = 15 to 35% of pancreatic cells involved, and 3 = more than 35% of cells involved). Evaluation was performed on 10 random fields (× 400) by 1 blinded investigator, calculating the means ± SEM (≥ 8 mice/group).

Amylase measurement
Serum amylase was determined by a colorimetric assay (Labtest ® , Brazil) according to the manufacturer's instructions. Blood samples were taken and centrifuged at 3500 rpm for 10 min as previously [30]. For the amylase assay, 500 μl of the substrate was added to the samples (10 μl). Theywere incubated in a water bath at 37 °C for 2 min, followed by addition of 500 μl of the color reagent and 4 ml of distilled water. After mixing and waiting 5 min, absorbance was determined at 660 nm.

Roles of ConA and ConBr on cellular necrosis and mitochondrial membrane potential (Δѱm): involvement of lectin domain
Cells were incubated for 1 h with ConA (200 µg/ml) or ConBr (200 µg/ml) and stimulated for 30 min with TLCS (500 µM). Necrotic cell death was detected by confocal microscopy (FluoViewTM 1000 -Olympus) using propidium iodide (PI; 1 µM: excitation 488 nm, emission 630-693 nm), cell membrane impermeable nucleic acid marker. In separate experiments changes in mitochondrial membrane potential (ΔѰm) were performed using the fluorescent probe tetramethylrodamine methyl ester (TMRM 100 nM-excitation/ emission: 543/550-650, 15 min incubation), the accumulation of which in mitochondria is driven by the highly negative inner mitochondrial membrane potential. Total cell number was detected using nuclear Hoechst 33,342 (50 µg/mL: excitation 364 nm, emission 405-450 nm) [10]. An average of 100 cells was analyzed from each test group. The percentage (%) of cells exhibiting mitochondrial depolarization, detected as a relative decrease in TMRM fluorescence, was calculated from total number of cells labeled with Hoechst 33,342 and TMRM.. Cell counts were performed in triplicate withiin 15 high-power fields. The involvement of the lectin domain on cellular necrosis was assessed by the prior incubation of ConA and ConBr (37 °C; 1 h) with 0.2 M of their binding sugar α -methyl-D-mannoside (α-MM).

Statistical analysis
The results were expressed as mean ± SEM (Standard Error of the Mean). Statistical analysis between groups was performed using Analysis of Variance ANOVA, followed by Bonferroni multiple comparisons test. The differences were considered statistically significant when P < 0.05. GraphPad Prism ® Software version 5.0 was used.

Effects of ConA and ConBr on histopathological changes of experimental acute pancreatitis
In vivo, lectins given after induction of pancreatitis protected the animals against acute pancreatitis caused by Na-TC. Thus ConA (10 mg/kg) and ConBr (10 mg/kg) protected the pancreatic tissue against histopathological alterations, comprising increased neutrophilic infiltration - (Fig. 1a), oedema (Fig. 1b) and tissue necrosis (Fig. 1c) caused by Na-TC. ConA and ConBr reduced the total histological changes induced by Na-TC by 43.6% and 71.8%, respectively, compared to controls (Fig. 1d). Photomicrographs show the histopathological changes caused by Na-TC (Fig. 1f) (black arrow: neutrophilic infiltrate; red arrow: edema) compared to the saline control group (Fig. 1e); ConA (Fig. 1g) and ConBr (Fig. 1h) protected pancreatic tissue in all parameters evaluated.

Role of ConA and ConBr on MPO activity in the Na-TC-induced experimental acute pancreatitis
Having observed the protective effects of lectins at a dose of 10 mg/kg on histopathological damage, a range of concentrations (0.1; 1 and 10 mg/kg) were analyzed on pancreatic myeloperoxidase changes. Both lectins decreased the sodium taurocholate Na-TC-induced elevation of myeloperoxidase in a dose-dependent manner; the results of this analysis confirmed that the 10 mg/kg dose of ConA and ConBr showed higher efficacy with both molecules decreasing the MPO levels by 94.4% and 98.1%, respectively, compared to Na-TC group, justifying its choice for subsequent analysis (Fig. 2a, b).

Effects of ConA and ConBr on amylase and lipase assay in the Na-TC-induced experimental acute pancreatitis
Na-TC administration increased amylase and lipase by 81.3% and 73.0%, respectively, compared to controls. ConA and ConBr treatment was protective, decreasing the elevated amylase by 42.0% and 31.4%, respectively. Similarly, the Na-TC-induced rise of lipase was reduced by 55.0% and 63.0% by ConA and ConBr treatment, respectively (Fig. 3a, b).

ConA and ConBr decrease necrosis and depolarization of mitochondrial membrane potential (Δѱm) induced by TLCS in pancreatic acinar cells
Pancreatic acinar cells that were treated with lectins prior to administration of TLCS showed a percentage of necrotic cells similar to controls. Application of TLCS increased acinar cell necrosis by 66.9% compared to control. Incubation with ConA and ConBr decreased this rise by 45.0% and 44.9%, respectively (Fig. 4a, b). In separate experiments, the intense mitochondrial depolarization of acinar cells induced by TLCS was reduced by 62.1% and 56.5% by prior incubation with ConA and ConBr, respectively (Fig. 5a, b).

Protective effects of ConA and ConBr on TLCS-induced necrosis is dependent on interaction with mannose residues
The anti-necrotic effects of the lectins were abolished by the association of both molecules with their specific sugar ligand α-methyl mannoside (α-MM). No statistical difference between lectin and TLCS groups was observed. In addition, application of α-MM per se did not induce acinar cell injury (Fig. 6).

Discussion
This study has shown for the first time that treatment with ConA and ConBr lectins is protective against the detrimental inflammatory, biochemical and histopathological changes that occur during biliary acute pancreatitis. It is likely that a significant component of this beneficial activity resides in local actions of these biological molecules, since both protected pancreatic acinar cells against mitochondrial dysfunction and necrosis caused by TLCS. The protective actions were dependent, at least in part, on specific interactions with the lectin domain.
Acute pancreatitis is a necro-inflammatory disorder the main causal factor of which is the presence of gallstones within the distal common bile duct, allowing consequent reflux of bile into the pancreatic duct [32,33] accounting for 30-60% of cases [34]. Retrograde administration of Na-TC into the pancreas of mice is a reliable, established experimental model of biliary acute pancreatitis, characterized by pancreatic inflammation with defined histopathological, inflammatory and biochemical changes [27]. Toxic precipitating agents such as bile acids cause acinar cell lesions; after intense stimulation with Na-TC the pancreatic tissue becomes swollen, with tissue necrosis and prominent leukocyte infiltration apparent [11]. In the present study, ConA and ConBr protected against histopathological damage of the pancreas caused by Na-TC, with significantly reduced edema, necrosis and neutrophil infiltration. In accord, both lectins decreased MPO levels elevated in biliary AP, consistent with protection against neutrophil infiltration that causes tissue damage and aggravates the inflammatory lesions [35]. A marked increase in serum pancreatic lipase and amylase levels at 24 h, characteristic of acinar cell damage during acute pancreatitis [36][37][38] was also reduced by treatment with ConA and ConBr.
Antinociceptive [22,24]; and anti-inflammatory [23,39]; properties of Diocleinae lectins, have previously been reported in animal experimental models. However, to-date . Data were expressed as mean ± standard error of the mean (SEM) of an experimental "n" of at least 8 animals. * p <0.05 vs. Sal group; # p < 0.05 vs. Na-TC group. One-way ANOVA followed by Bonferroni post-test there has been no study showing anti-inflammatory activities of ConA and ConBr. Our current results demonstrate novel protective actions of ConA and ConBr to combat inflammation in experimental acute pancreatitis. Since the lectins were applied after acute pancreatitis had been instigated, rather than as pretreatment, new possibilities for therapeutic application are suggested; an important finding since there is currently no specific treatment for this debilitating and sometimes fatal disease. Systemically administered lectins bind to pancreatic acinar cells and modulate their metabolism. However, it cannot be excluded that the protection may also involve other properties of the lectin proteins and indirect effects mediated through lectin-driven changes to the immune response [40].
Pancreatic acinar cell necrosis is considered the initiating event of cell damage fundamental for the onset of an inflammatory cascade in acute pancreatitis [41][42][43][44]. Bile acids cause local cell injury by the induction of sustained increases in cytosolic [Ca 2+ ] that reduce mitochondrial membrane potential and deplete cellular ATP, leading to acinar cell necrosis [5][6][7][8][9][10]. Mitochondrial dysfunction, involving formation of the permeability transition pore, is a core feature of acute pancreatitis. [11] In isolated acinar cells ConA and ConBr protected against mitochondrial dysfunction, partially inhibiting mitochondrial depolarization induced by TLCS. These data are supported by studies showing the involvement of these lectins with mitochondrial function in other cell types e.g. after associating with the mannose moiety residing on a cell membrane glycoprotein in a liver tumor cell line, ConA was internalized to mitochondria via clathrin-mediated endocytosis and promoted autophagic cell death [45]. In addition, ConA and ConBr promoted partial Δψm loss that triggered apoptosis in leukemic cells lines [20]. Consistent with an inhibition of Δψm depolarization induced by TLC-S, ConA and ConBr significantly prevented TLC-S-induced necrosis. Previously ConBr was found to protect against hippocampal cell death in mice induced by quinolinic acid, an action that involved inhibition of necrosis [46]. The surface of the pancreatic acinar cell contains several sugar residues including N-acetyl-glucosamine, galactose and possibly also N-acetyl-neuraminic acid in the apical region, and fucose, galactose, mannose and glucose residues in the basolateral region [47,48]. Molecules such as ConA and ConBr possess binding specificity to glucose/ mannose residues, suggesting binding to sugars present on the surface of the acinar cell as a possible mechanism of action of lectins. The non-catalytic domain or carbohydrate recognition domain (CRD) of the lectins, also called the lectin domain, is the site at which lectins bind specifically and reversibly to carbohydrates and other substances that contain sugar moieties. Many of the biological activities of lectins in general are related to an interaction with sugars present on the cell surface [49]. Proof of participation of the lectin domain in processes of cell recognition and interaction has been shown by inhibition of the lectin effect by binding to its specific sugar/glycoconjugate [22,23,24,38). ConA and ConBr are glucose/mannose ligands lectins which have   [23,50,51]. In this study, protective effects against TLC-S-induced cell death were abolished after blocking the ConA and ConBr lectin domain, indicating that their actions are, at least partly, due to a direct interaction of the lectin domain with pancreatic acinar cells.
Diocleinae lectins exhibit high degree of homology in their primary structures and share many biochemical and structural features, such as evolutionarily conserved regions that characterize this group [52]. Differences in their biological activity may reflect not only small changes in the amino acid sequence of the CRD, but also different conformations of the site itself and adjacent loops [52,53]. The comparative study of lectins from the same subfamily is useful to broaden knowledge about biological structure-activity relationships of these proteins. The lectins evaluated in this study study exhibit 99% structural similarity with only two amino acid residues different; glycine (Gly) 58 and Gly 70 in ConBr are replaced by aspartate (Asp) and alanine (Ala) in ConA, respectively, with none of these residues close to the carbohydrate binding site. Both are tetramers at neutral pH or above 5.5, and dimers below 5.5 [54,55]. ConA and ConBr applied under the same conditions of physiological pH would present as tetramers and thus retain their ability to bind to carbohydrate residues on the cell surface; similar activities of ConA and ConBr were observed in this study. This might explain small differences in biological activities previously reported despite the high structural similarity of these lectins [12,[55][56][57]. However, in the present study, ConA and ConBr exhibited a very similar pattern of results in all assays, in agreement with prior analyses with ConA and ConBr in leukemia cell lines [20].
In summary, our findings demonstrate for the first time that ConA and ConBr, applied after the initiation of biliary acute pancreatitis, are protective against inflammation and tissue damage. These actions are partly mediated via interaction of the lectin domain with sugar residues present on the acinar cells, leading to protection against mitochondrial dysfunction and acinar cell necrosis. Further studies are warranted to better understand the underlying mechanisms of action of these anti-inflammatory molecules and assess their potential for therapeutic development.

Data availability
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.

Compliance with ethical standards
Ethics approval All procedures were approved by the Animal Use Ethics Committee (CEUA) of the Federal University of Ceara (protocol n° 99/2013).

Consent to participate
All authors declare participation in the study.

Consent for publication
All authors declare consent for publication of study.