Ketamine and xylazine effects in murine model of acute pancreatitis

Abstract

Ketamine and xylazine (Ket/Xyl) are anesthetic agents that target neural pathways and are commonly used in combination in mouse studies. Since neural pathways can modulate acute pancreatitis severity, we asked if Ket/Xyl affect disease severity. C57BL/6 mice were treated with six hourly injections of cerulein to induce mild acute pancreatitis. Mice were also treated with and without ketamine, xylazine, and Ket/Xyl before pancreatitis induction in vivo and in vitro. Ket/Xyl pretreatment in vivo increased selected parameters of pancreatitis severity such as trypsin activity and edema; these effects were predominantly mediated by xylazine. Ket/Xyl also changed markers of autophagy. These in vivo effects of Ket/Xyl were not attenuated by atropine. The drugs had no little to no effect on pancreatitis responses in isolated pancreatic cells or lobules. These findings suggest that Ket/Xyl administration can have substantial effect on acute pancreatitis outcomes through nonmuscarinic neural pathways. Given widespread use of this anesthetic combination in experimental animal models, future studies of inflammation and injury using Ket/Xyl should be interpreted with caution.

NEW & NOTEWORTHY Ketamine and xylazine anesthetic agent administration before acute pancreatitis induction in mice lead to changes in pancreatitis responses independent of acute pancreatitis induction. Future studies should consider the potential effects of anesthesia administration when studying disease processes associated with inflammation and injury.

INTRODUCTION

The pancreas has sympathetic and parasympathetic neural innervations from vagal and spinal pathways that affect its endocrine and exocrine functions. These neural pathways enter the pancreas through splanchnic nerves and the celiac plexus, are primarily sensitive to capsaicin, and contain substance P and calcitonin gene-related peptide (CGRP). The parasympathetic pathway acts through the vagus nerve to regulate pancreatic exocrine secretion in humans and rodents. Studies of the sympathetic effects of α- or β-adrenergic agonists and antagonists on pancreatic exocrine secretion have been inconsistent. Parasympathetic excitatory pathways act through cholinergic muscarinic M3 and M1 receptors. Inhibitory parasympathetic pathways act through other substances like nitric oxide, vasointestinal peptide, gastrin-releasing peptide, and pituitary adenylate cyclase-activating polypeptide (1–4). Neural innervation has been associated with the parenchymal injury in acute pancreatitis. Activation of pancreatic sensory neurons has been shown to initiate and maintain a proinflammatory response that corresponds to degree of acute pancreatitis severity (5, 6). In addition, intraduodenal toxins have been shown to promote acute pancreatitis through neural connections between the duodenum and pancreas (7). Transient receptor potential (TRP) channels, including TRPV1 and TRPV4 that are present on pancreatic nerves and shown to modulate inflammation and pain in acute and chronic pancreatitis (5, 8, 9). Finally, the pain associated with chronic pancreatitis has been linked to the release of the neurotransmitters substance P and CGRP (5). Thus, a range of neural pathways have the potential to affect both physiologic pancreatic responses and pancreatitis.

Xylazine, an α2 agonist, and ketamine, which has less well-defined neural target, are often used in combination as anesthesia for surgical procedures in mice. These agents have mild side effects including enhanced anxiety-related behaviors with repeated use but are otherwise well tolerated (10–14). Rarely, their effects on thymocytes and ocular changes have prompted the use of other anesthetics (11, 14–16). Xylazine alone has been reported to affect gastrointestinal functions including a dose-dependent delay in small intestinal transit (17). The favorable overall safety and side effects profile of combining ketamine and xylazine has made them a preferred anesthetic for rodent models.

Given the neural innervations of the pancreas and its implications in acute pancreatitis, we questioned whether ketamine and xylazine could affect baseline pancreatic responses and the development or severity of acute pancreatitis in animal models. To our knowledge, this potential association has not been explored. Here, we treated mice with ketamine and/or xylazine before induction of acute pancreatitis to assess whether these drugs affect the markers of acute pancreatitis in mice. Ketamine/xylazine (Ket/Xyl) are typically short acting anesthetic agents used most often in rodent surgery. Our studies show that this anesthetic combination has prominent effects on baseline pancreatic function and select acute pancreatitis responses in the cerulein pancreatitis model.

METHODS

Animals Used

C57/BL6 female and male mice between 20 and 25 g were used in all experiments. All animal procedures and experiments were approved by the Veterans Affairs Institutional Animal Care and Use Committee (West Haven, CT).

Preparation of Isolated Pancreatic Lobules and Acini

For pancreatic lobules studies, after C57/BL6 female and male mice were euthanized using CO₂ inhalation, the pancreas was removed and mechanically separated into 5-mm diameter pancreatic fragments. Each pancreatic lobule was placed in a 24-well Falcon tissue culture plate (BD Biosciences, Franklin Lakes, NJ) and placed in 0.5 mL of incubation buffer including 10 mM HEPES, pH 7.4, 95 mM NaCl, 4.7 mM KCl, 0.6 mM MgCL₂, 1 mM NaH₂PO₄, 10 mM glucose, 2 mM glutamine, 0.1% bovine serum albumin, and 1× minimum essential medium amino acids (Gibco, Carlsbad, CA).

For pancreatic acini studies, C57B/BL6 mice were euthanized using CO₂ inhalation, and the pancreas was removed, weighed, and placed in oxygenated incubation buffer containing 10× Salts solution, 100× nonessential MEM-amino acids, 1 M NaH₂PO₄, 0.6 M MgCl₂, 1.3 M CaCl₂, 10 mM glucose, 2 mM l-glutamine, and 0.1% BSA, pH 7.4. The pancreas was then mechanically minced in buffer and incubated with collagenase buffer to digest the acini for 60 min in a shaking water bath at 37°C. Digested acini were filtered through a 200-µm mesh and washed before being placed in a 24-well Falcon tissue culture plate (BD Biosciences, Franklin Lakes, NJ). All reagents were purchased from Sigma (Sigma Aldrich, St. Louis, MO).

Treatment of Isolated Pancreatic Lobules and Acini

Lobules and acini were pretreated with ketamine/xylazine (Ket/Xyl) (10⁻⁵, 10⁻⁷, and 10⁻⁹ M) for 1 hr before induction of pancreatitis with cerulein hyperstimulation (10⁻⁷ M) and physiological cerulein (10⁻⁹ M) (West Haven VA Pharmacy, CT). After treatments, 50 µL of cell-free medium was removed for amylase secretion analysis. For lobules, the remaining medium containing the pancreatic lobules was homogenized with 10 strokes in a 2-mL glass Teflon homogenizer. The homogenized pancreatic lobules were then centrifuged at 500 g for 10 min. The supernatant was used to assay zymogen activation and amylase secretion. For acini, the remaining medium containing the pancreatic acini were placed in 1.5-mL Eppendorf tubes and centrifuged at 30 g for 60 s. About 50 µL of the supernatant was removed to assay secreted amylase.

In Vivo Pancreatitis Induction and Pretreatment (Ketamine, Xylazine, Atropine)

Mice were pretreated with subcutaneous injection of 0.1 mL/10 g body weight of 100 mg/kg ketamine and/or 10 mg/kg xylazine 5 min before cerulein injection (West Haven VA Pharmacy, CT). Ketamine/xylazine concentrations were consistent with concentrations of ketamine/xylazine used in our laboratory methodology for animal procedures. For atropine experiments, mice were treated with subcutaneous injection of 0.05 mg/kg atropine sulfate and saline 10 min before ketamine/xylazine (West Haven VA Pharmacy, CT). Studies have used anywhere from 0.15 to 0.3 mg/kg of atropine in veterinary practices as a preanesthetic or higher as organophosphate poisoning antidote without any significant safety concerns (18–20). Mice were then treated with six hourly intraperitoneal injections of 40 µg/kg cerulein in saline. At the time of tissue and blood collection, mice were treated with a subcutaneous injection of 0.1 mL/10 kg body weight of a combination of 100 mg/kg ketamine and 10 mg/kg xylazine just before euthanization by cardiac puncture and exsanguination. Pancreatic tissue was isolated and used for subsequent assays. Blood was placed in heparin-coated tubes and centrifuged at 2,000 g for 15 min (BD Biosciences, Franklin Lakes, NJ). Remaining supernatant (plasma) was removed and stored at −80°C for future assays.

Amylase Assay

Samples were thawed on ice and diluted 1:10 in deionized water for amylase assay. Amylase levels were determined using the commercial Phaebadas kit and read at 620 nm in duplicate samples (Phaebadas Amylase Test, Cambridge, MA; Magle Life Sciences, Lund, Sweden).

Zymogen Activation Assays

Supernatants from pancreatic lobules and acini and in vivo pancreatic samples were assayed for zymogen activity. Pancreatic samples were thawed on ice and homogenized manually with 10 strokes in a 2-mL glass Teflon homogenizer in zymogen assay buffer (50 mM Tris buffer, 150 mM NaCl, 1 mM CaCl₂, pH 8.1). Samples were centrifuged at 1,000 g for 1 min and supernatant was subsequently used. Samples of 20 µL were added to a 24-well Falcon tissue culture plate (BD Biosciences, Franklin Lakes, NJ) along with 50 µL of fluorometric enzyme substrate [40 mM final; Chymotrypsin (Suc-Ala-Ala-Pro-Phe-AMC) from Calbiochem Division, EMD chemicals, Gibbstown, NJ; Trypsin (Boc-Gln-Ala-Arg-MCA) from Peptides International, Louisville, KY] and 450 µL of zymogen assay buffer. Plates were read using fluorometric plate reader Flx800 (Biotek Instruments, Winnoski, VT) at excitation wavelength of 380 nm and emission wavelength of 440 nm for 20 measurements over 10 min. The slope of the resulting line was used to represent enzyme activity. For lobules and acini, the slope of the line was normalized to total amylase content. For pancreatic homogenates, the slope of the line was homogenized to total protein concentration. Protein concentration was obtained using Pierce 660 nm Protein Assay kit (Thermo Fisher Scientific, Waltham, MA).

Edema Measurements

Immediately upon removal of pancreatic tissue from animal, pancreatic pieces were placed on aluminum foil and measured to determine the wet weight. Pancreatic pieces were subsequently dried at 60°C for 48 h and re-weighed. Percent wet weight was determined using the following formula: 100 × (wet weight – dry weight)/wet weight.

Histology and Immunostaining

Pancreatic tissue pieces were placed into cassettes after removal from animals and fixed in normal buffered formalin for 48 h. Cassettes were washed in 70% ethanol and delivered to Yale Pathology Tissue Services Lab for paraffin embedding, sectioning, processing H&E, and immunostaining (TUNEL: S7101, MilliporeSigma, Burlington, MA, Ki67: 1:80; CRM325, Biocare Medical, Pacheco, CA, Ly6-G: 1:100; MCA771, ABD Serotec, Hercules, CA) and examined at ×40 magnification using Zeiss Axiophot microscope (Oberkochen, Germany). Histological markers were scored as described and expressed as immunoreactive cells per field (21). Images were recorded using Images digital camera (Spot Imaging, Sterling Heights, MI).

Western Blot

Frozen pancreatic samples were homogenized manually with 10 strokes in a 2-mL glass Teflon homogenizer in homogenization buffer of 0.3 mM sucrose, 10 mM Tris, pH 6.4, Complete Protease Inhibitor mini (Roche, Mannheim, Germany), 10 mM benzamidine, 0.1 mg/mL soybean trypsin inhibitor, and phosStop tablet (Roche, Mannheim Germany). Samples were centrifuged at 500 g for 10 min and the remaining supernatant was used for Western blot analysis. Protein concentration was assayed using Pierce Protein kit as aforementioned. Samples were diluted with deionized water to 25–50 μg for Western blot. Briefly, 6× SDS was added at 1:5 dilution and boiled at 95°C. Samples were run on electrophoresis gels (BioRad, Hercules, CA) and transferred to Immobilin-P membranes (Millipore, Billerica, MA). Membranes were blocked for 60 min at room temperature with blotto (5% milk powder in TBS-Tween: 1× TBS, 0.05% Tween 20) and washed three times with TBS-Tween before probing with primary antibody in 5% bovine serum albumin in TBS-Tween [LC3B: 1:750 (Cell Signaling Technology, Danvers, MA), p62: 1:750 (BD Biosciences, Franklin Lakes, NJ), Grp78/BiP 1:1,000 (Enzo Life Sciences, Farmingdale, NY)] at 4°C overnight. Membranes were subsequently washed with blotto and incubated at room temperature with secondary antibody (Sigma-Aldrich, St. Louis, MO) for 1 h. Membranes were washed with blotto and TBS-Tween and labeled bands were detected using Supersignal West Pico Chemiluminescence ECL reagent (Pierce, Thermo Fisher Scientific, Waltham, MA) and imaged with ChemiDoc Touch Imaging System (BioRad, Hercules, CA). Total lane protein was measured using Stain-Free imaging technology, which uses a proprietary trihalo compound to fluoresce proteins directly in gel, to normalize signals within each lane (BioRad).

Statistical Analysis

Data were analyzed using nonparametric statistical analysis including Mann–Whitney U test to compare between samples. Outliers in edema by wet weight were removed from final analysis as these values were deemed likely to be due to weight of pancreatic fat instead of pancreatic tissue. All statistically analyzed samples included at least four mice. Where appropriate, data were presented as mean values with a P value of ≤0.05 considered statistical significance. Data were analyzed using Prism Version 8 (Graphpad, San Diego, CA).

RESULTS

Ketamine/Xylazine Changes Parameters of Acute Pancreatitis in Male and Female Mice

Mice were treated with subcutaneous ketamine/xylazine (Ket/Xyl) injection 5 min before initiation of cerulein-induced acute pancreatitis. Pancreatitis markers were measured after 6 h from induction of acute pancreatitis. Ket/Xyl pretreatment, without induction of pancreatitis, had lower serum amylase levels than control mice (1.46 vs. 3.22, P = 0.0027; Fig. 1A). Ket/Xyl pretreatment also reduced serum amylase levels 6 h after cerulein induction of acute pancreatitis (1.27 vs. 1.98, P < 0.0001; Fig. 1A). Mice with only Ket/Xyl pretreatment had increased pancreatic trypsin (20.77 vs. 5.00, P < 01; Fig. 1B) and chymotrypsin (4.77 vs. 0.10, P = 0.018; Fig. 1C) activities compared with control mice at 6 h. When normalized to controls, there was no difference in cerulein-induced trypsin activity (2.88 vs. 2.81, P = 0.73; Fig. 1B) between control and Ket/Xyl pretreated mice with pancreatitis. However, there was a decrease in chymotrypsin in Ket/Xyl pretreated mice with pancreatitis compared with pancreatitis only (7.81 vs. 32.57, P = 0.020; Fig. 1C). Control mice pretreated with Ket/Xyl had more pancreatic edema (71.81 vs. 69.35, P = 0.022; Fig. 1D), but there was no difference in pancreatic edema after pancreatitis induction (1.07 vs. 1.09, P = 0.76; Fig. 1D). There was no statistically significant difference in these pancreatitis markers when comparing male to female mice.

Figure 1.

Ketamine/xylazine administration before inducing cerulein pancreatitis affects functional assays in male and female mice. In serum amylase at baseline and fold change over control (A), pancreatic trypsin activity at baseline and fold change over control (B), pancreatic chymotrypsin activity at baseline and fold change over control (C), and pancreatic edema by wet/dry weight at baseline and fold change over control (D). Means ± SD. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001. Female mice denoted in blue. Cer, cerulein; Ket/Xyl, ketamine/xylazine.

Acute pancreatitis histology markers also showed changes in acute pancreatitis in the pretreated mice compared with mice not pretreated with Ket/Xyl. The development of vacuoles with central clearing was defined as “clear vacuoles” and vacuoles with cellular components within was defined as “filled vacuoles.” Though the identity of these two morphologic forms of vacuoles is unclear, they likely represent distinct forms of autophagy. Pretreated (Ket/Xyl) mice without pancreatitis induction had fewer clear vacuoles (0.00 vs. 0.38, P < 0.001; Fig. 2, A and D) than control mice without pancreatitis induction. No filled vacuoles were detected at baseline for pretreated Ket/Xyl mice and control mice (Fig. 2, B and D); necrosis was also minimal for the two groups (Fig. 2, C and D). Mice treated with Ket/Xyl pretreatment with pancreatitis induction had more clear vacuoles (1.38 vs. 0.74, P = 0.020; Fig. 2, A and D), and more filled vacuoles (0.92 vs. 0.00, P = 0.004; Fig. 2, B and D) than mice without pretreatment with pancreatitis induction. There was no difference in necrosis in mice with and without pretreatment with pancreatitis induction (1.64 vs. 1.42, P = 0.66, Fig. 2, C and D). Taken together, both histology and functional pancreatitis markers show that pretreatment with Ket/Xyl 5 min before the induction of acute pancreatitis changed pancreatitis markers.

Figure 2.

Ketamine/xylazine administration before inducing cerulein pancreatitis changes histological disease features in male and female mice. Clear vacuole formation (arrow) at baseline and fold change over control (A), filled vacuole formation (arrow) at baseline and fold change over control (B), and necrosis at baseline and fold change over control (C). Representative histology of female mice at ×40 are shown in D. Means ± SD. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. Female mice denoted in blue. Cer, cerulein; Ket/Xyl, ketamine/xylazine.

Ketamine/Xylazine Pretreatment Affects Proliferation and Neutrophil Infiltration in Vivo

To assess whether mice pretreated with Ket/Xyl also experienced an increase in histologic apoptosis, additional immunohistochemical markers in male and female mice were measured. There was no statistically significant increase in apoptosis using TUNEL staining in mice pretreated with Ket/Xyl (0.53 vs. 0.03, P = 0.133; Fig. 3, A and B) at 6 h compared with Ket/Xyl controls. There was significantly less TUNEL staining in Ket/Xyl pretreated mice than in nonpretreated mice (6.02 vs. 0.33, P = 0.002; Fig. 3, A and B). Ket/Xyl pretreated mice had less neutrophil infiltration at 6 h than nonpretreated mice (0.38 vs. 5.56, P = 0.002; Fig. 3C). This difference remained at 24 h. At 24 h, pretreated mice also had less neutrophil infiltration than nonpretreated mice (8.55 vs. 28.88, P = 0.009; Fig. 3C). In addition, at 24 h, pretreated mice had dramatically increased levels of Ki67 suggesting increased cell proliferation compared with nonpretreated mice (22.15 vs. 5.58, P = 0.015; Fig. 3D). The increased Ki67 positivity was observed in both pancreatic acini cell and nonacinar cell types (Fig. 3D). Ki67 levels were also increased at baseline in Ket/Xyl pretreated mice without pancreatitis induction compared with no pretreatment (3.83 vs. 0.00, P = 0.029; Fig. 3D). Although only a single Ket/Xyl dose was given before cerulein-induced acute pancreatitis initiation, its effects were still apparent at 24 h.

Figure 3.

Ketamine/xylazine reduces apoptosis and neutrophil infiltration but increases a proliferative marker during pancreatitis. Pancreatic TUNEL levels (A) and staining in mice with and without ketamine/xylazine pretreatment (B). Representative immunohistochemistry for TUNEL staining at ×40 are shown in B. C: Ly6b staining scores for neutrophil infiltration. D: Ki67 staining scores for proliferation. Means ± SD. *P ≤ 0.05, **P ≤ 0.01. Cer, cerulein; Ket/Xyl, ketamine/xylazine. ≥2 male and female mice per testing condition.

Together these findings demonstrate that Ket/Xyl pretreatment increases select parameters of acute pancreatitis responses in vivo. Since these agents are thought to act on different cellular targets, we next examined whether one or both agents increased pancreatitis responses in vivo.

Ketamine or Xylazine Pretreatment Has Distinct Effects on Acute Pancreatitis Severity

To assess whether ketamine and xylazine have distinct effects on acute pancreatitis markers, these agents were given separately as a pretreatment in male and female mice having cerulein-induced acute pancreatitis. Mice pretreated with ketamine demonstrated an increase in serum amylase levels compared with mice treated with Ket/Xyl (6.21 vs. 1.90, P = 0.004; Fig. 4A). This increase was not seen in xylazine pretreated mice (2.19 vs. 1.90, P = 0.067; Fig. 4A). Mice pretreated with xylazine (5.73 vs. 2.19, P = 0.002), but not ketamine (6.21 vs. 5.71, P = 0.66), demonstrated an increase in serum amylase compared with mice without pretreatment with pancreatitis induction (Fig. 4A). In addition, the increased trypsin and chymotrypsin activities seen in Ket/Xyl treated were more prominent in xylazine pretreated mice than ketamine pretreated mice (Fig. 4, B and C). Ketamine pretreated mice showed much lower trypsin and chymotrypsin levels compared with Ket/Xyl pretreated mice (trypsin: 17.05 vs. 62.37, P = 0.004; chymotrypsin: 3.79 vs. 38.30, P = 0.004; Fig. 4, B and C). Xylazine pretreatment increased trypsin and chymotrypsin activities but the effects were much less than with Ket/Xyl pretreatment (trypsin: 41.24 vs. 62.37, P = 0.016; chymotrypsin: 26.27 vs. 38.30, P = 0.109; Fig. 4, B and C). Compared with controls, xylazine (trypsin: 41.24 vs. 14.05, P = 0.002; chymotrypsin: 26.27 vs. 4.99, P = 0.004) but not ketamine (trypsin: 17.05 vs. 14.05, P = 0.52; chymotrypsin (3.79 vs. 4.99, P = 0.79) pretreatment had higher trypsin and chymotrypsin activity (Fig. 4, B and C). Ketamine pretreated mice also showed a decrease in edema compared with Ket/Xyl pretreated mice, which was not seen with xylazine pretreatment (ketamine: 72.62 vs. 76.78, P = 0.016; xylazine: 75.66 vs. 76.78, P = 0.933; Fig. 4D). Edema was not significantly increased in ketamine (72.62% vs. 74.89%, P = 0.11) or xylazine (75.66% vs. 75.89%, P = 0.24) pretreated mice compared with no pretreatment (Fig. 4D). These data suggest that the increase in zymogen activation and decrease in serum amylase seen in Ket/Xyl pretreated mice is more similar to treatment with xylazine alone than ketamine alone. Xylazine, but not ketamine, also significantly affects pancreatitis markers compared with mice without pretreatment. In addition, with the exception of serum amylase levels in mice treated only with ketamine, the pancreatitis markers after ketamine and xylazine alone were less than pancreatitis markers after Ket/Xyl together was given.

Figure 4.

Ketamine and xylazine have distinct effects on cerulein-induced acute pancreatitis: (A) serum amylase secretion, (B) pancreatic trypsin activity, (C) pancreatic chymotrypsin activity, and (D) pancreatic edema by dry/wet weight. Means ± SD. *P ≤ 0.05, **P ≤ 0.01, ****P ≤ 0.0001. Cer, cerulein; Ket/Xyl, ketamine/xylazine; Xyl, xylazine; Ket, ketamine. ≥2 male and female mice per testing condition.

Histological findings were consistent with other pancreatitis severity markers. Xylazine-treated mice had similar levels of vacuoles (clear and filled) compared with ketamine/xylazine pretreated mice (clear vacuoles: 1.93 vs. 1.38, P = 0.067; filled vacuoles: 0.15 vs. 0.93, P = 0.200; Fig. 5, A and B). However, ketamine alone had fewer clear vacuoles and more filled vacuoles than Ket/Xyl pretreated mice (clear vacuoles: 0.68 vs. 1.38, P = 0.044; filled vacuoles: 2.18 vs. 0.93, P = 0.044; Fig. 5, A and B). Mice pretreated with xylazine (1.58 vs. 0.74, P = 0.047) but not ketamine (0.89 vs. 0.74, P = 0.96) had more clear vacuoles compared with mice without pretreatment (Fig. 5A). Mice pretreated with ketamine (2.38 vs. 0.33, P = 0.004), but not xylazine (0.15 vs. 0.33, P = 0.62), on the other hand, had more filled vacuoles compared with mice without pretreatment (Fig. 5B). There was no significant difference in necrosis in mice treated with either ketamine or xylazine compared with Ket/Xyl pretreated mice (Fig. 5D). Necrosis was not significantly changed in mice pretreated with ketamine or xylazine compared with mice without pretreatment. These findings similarly suggest that xylazine alone pretreatment is more similar to Ket/Xyl pretreatment than pretreatment with only ketamine. In addition, ketamine-only pretreatment led to distinct differences in acinar cell vacuoles compared with Ket/Xyl pretreatment.

Figure 5.

Ketamine and xylazine affect pancreatitis histology differently: clear vacuoles (A), filled vacuoles (B), and necrosis (C). Means ± SD. *P ≤ 0.05, **P ≤ 0.01. Cer, cerulein; Ket/Xyl, ketamine/xylazine; Xyl, xylazine; Ket, ketamine. ≥2 male and female mice per testing condition.

Ketamine and Xylazine Pretreatment Changes Autphagy Proteins

Acinar cell vacuolization during pancreatitis is often related to accumulation of organelles in autophagy pathways. Since there was an increase in vacuoles in the ketamine-only treated mice compared with Ket/Xyl pretreated mice, we sought to determine whether these differences could reflect changes in autophagy. Pancreatic levels of the autophagic markers LC3-2 and p62 were assessed using Western blot. Reduced (less effective) autophagic flux is characterized by an increase in both p62 and LC3-2 levels as p62 accumulates in autophagic vacuoles compared with control samples. On the other hand, effective autophagy is characterized by a relative increase over baseline in LC3-2 levels but a decrease in p62 levels because of its degradation within the autophagic pathway. To assess differences in p62 and LC3-2 levels, we examined the change in p62 and LC3-2 levels in cerulein-treated versus control mice. Compared with mice without any pretreatment, Ket/Xyl pretreated mice had higher p62 level (1.12 vs. 0.55, P = 0.029; Fig. 6, A–D), suggesting that Ket/Xyl pretreatment caused an accumulation of p62 after pancreatitis induction. This would be consistent with reduced autophagic flux. Xylazine-only treated mice also had higher p62 levels than control mice (1.42 vs. 0.55, P = 0.029; Fig. 6, A–D). Ketamine-only treated mice also had higher p62 levels compared with control mice (0.88 vs. 0.55, P = 0.029; Fig. 6, A–D) and ketamine-only mice tended to have more LC3-2 and p62 protein levels compared with other conditions (Fig. 6, A–D). These findings suggest that Ket/Xyl pretreatment, similar to xylazine pretreatment or ketamine pretreatment, causes changes in autophagic proteins would be consistent with reduced autophagic flux (a feature of acute pancreatitis). This is consistent with the overall changes seen with pancreatic severity markers and histology.

Figure 6.

Ketamine and xylazine affect autophagy markers differently. Integrated density of p62 (A) and LC3-2 (B) on Western blots. Fold over control integrated density from p62 (C) and LC3-2 (D) Western blots with and without pretreatment, xylazine pretreatment, or ketamine pretreatment. E: representative Western blot. Means ± SD. *P ≤ 0.05). Cer, cerulein; Ket/Xyl, ketamine/xylazine; Xyl, xylazine; Ket, ketamine.

Ketamine/Xylazine Administration in Supraphysiological Cerulein In Vitro Studies Did Not Increase Pancreatitis Severity in Acini and Lobule from Female Mice

Ketamine and xylazine are thought to act through neural pathways, some of which can be preserved in mechanically dissected pancreatic lobules. Furthermore, pancreatic acinar cells also have receptors that respond to cholinergic stimulation. To determine whether the increased pancreatitis severity observed in pretreated mice is preserved when vagal nerve stimulation is removed or when intrapancreatic neuronal connections are disrupted, we assessed supraphysiological cerulein in vitro pancreatitis responses with Ket/Xyl pretreatment of pancreatic lobules and acini. Female pancreatic preparations were used in these studies given evidence in in vivo studies that female mice may demonstrate greater Ket/Xyl effects compared with male mice. In isolated acini, Ket/Xyl pretreatment did not change zymogen activation or amylase secretion (Fig. 7, A–C). There was less trypsin activity in Ket/Xyl pretreated acini compared with cerulein-only treated acini (1.35 vs. 1.79, P = 0.029; Fig. 7A). Little effect of Ket/Xyl was seen in isolated pancreatic acini suggesting that the pancreatic acinar cell is not the primary target of Ket/Xyl effects seen in vivo. Administration of Ket/Xyl with physiologic cerulein is shown in Supplemental Figs. S1 and S2 (all Supplemental material is available at https://doi.org/10.6084/m9.figshare.14343227).

Figure 7.

Ketamine/xylazine treatment before cerulein treatment does not increase pancreatitis responses in acini: trypsin activity (A), chymotrypsin activity (B), and amylase secretion (C). Acini were collected from ≥2 male and female mice per testing condition. Means ± SD. Cer, cerulein; Ket/Xyl, ketamine/xylazine.

To determine whether the in vivo Ket/Xyl effects could be replicated in supraphysiological cerulein in vitro lobules that retain neural elements, lobules from female mice were pretreated with these agents. There was no significant difference in zymogen activation or amylase secretion in these Ket/Xyl-treated preparations (Supplemental Fig. S3, A–C). These findings suggest that neural pathways that remain on lobules do not account for the observed Ket/Xyl effects seen in vivo and that higher neural pathways likely mediate the effects of these anesthetics on pancreatitis.

Atropine Administration Does Not Decrease the Effects of Ketamine/Xylazine on Acute Pancreatitis Severity

Parasympathetic and sympathetic systems act through cholinergic receptors; cholinergic muscarinic pathways can modulate physiological and pancreatitis responses. To determine whether Ket/Xyl acts on this pathway, we coadministered the muscarinic antagonist atropine. Atropine did not significantly change the responses in Ket/Xyl-pretreated mice with respect to serum amylase and zymogen activation (Supplemental Fig. S4, A–C). Atropine and Ket/Xyl-pretreated mice, however, showed an increase in pancreatic edema compared with Ket/Xyl-pretreated mice without atropine (78.71 vs. 76.78, P = 0.024; Supplemental Fig. S4D). There were also no statistically significant differences in histology when atropine was added to Ket/Xyl-pretreated mice (Supplemental Figs. S5, A–D, and S6). Taken together, these findings suggest that the effects of Ket/Xyl are mediated in great part by nonmuscarinic neural pathways.

DISCUSSION

Ketamine/xylazine is a common analgesic/anesthetic used in animal models, including pancreatic disease models. We found that ketamine/xylazine treatment before cerulein-induced acute pancreatitis led to changes in acute pancreatitis outputs, with changes in some markers, such as zymogen activation and histology. Notably, the effects of ketamine/xylazine on the pancreas in unstimulated control conditions suggest that their effects in acute pancreatitis may be secondary to altered basal responsiveness. This finding suggests that it is likely these baseline effects will change select pancreatitis outputs regardless of the model that follows administration of ketamine/xylazine. Sex-dependent differences were not significantly observed. We also describe “clear” and “filled” vacuoles on morphology that are increased in quantity after ketamine/xylazine pretreatment. These vacuoles likely reflect morphologic differences in types or stages of autophagic events. Future studies should explore the identity of these vacuoles and their potential importance in the responses we have explored.

Ketamine is primarily an NMDA receptor antagonist but can also interact with other receptor classes such as opioid receptors. Xylazine, on the other hand, is an α2-adrenergic agonist that is similar to clonidine. TRPV1 and TRPV4 on primary sensory neurons have been associated with increased pancreatic inflammation upon activation, leading to development of experimental acute pancreatitis (5). Ketamine has been associated with increased TRPV1 sensitivity to pain induction via capsaicin (22). Our findings are consistent with effects of ketamine in part due to acting on TRPV channels. Ket/Xyl increased some measures of acute pancreatitis severity. Whether the increased severity we observed was due to potentiation of TRPV channels leading to increased pancreatic inflammation should be further assessed. Ketamine has also been associated with inhibiting calcium influx in L-type calcium channel (Cav1.2) and reducing smooth muscle contractility in the bladder (23). Cav1.2 has also been implicated in acute pancreatitis and increased Cav1.2 is associated with increased intracellular calcium in pancreatic acinar cells and can lead to acute necrotizing pancreatitis (24). Future studies should also assess whether ketamine’s mechanism of action in increased pancreatitis severity is associated with calcium channels such as Cav1.2.

Though we do not know of studies that have assessed ketamine and/or xylazine effects in acute pancreatitis severity, several studies have explored the effects of other anesthetics and pain-relieving agents on acute pancreatitis severity that share features with ketamine and/or xylazine. For example, opioids are often used as analgesic in the treatment of pain associated with acute pancreatitis. Morphine treatment after pancreatitis induction has been associated with increased pancreatic neutrophil infiltration and necrosis, gut permeabilization, and bacteremia in several models of acute pancreatitis including cerulein-induced acute pancreatitis (25). Morphine also delays macrophage infiltration and regeneration associated with resolution of acute pancreatitis. These findings are mediated through µ-opioid receptors (25). In addition, clonidine, an α2 agonist, has also been assessed in acute pancreatitis literature. Interestingly, clonidine inhibits both salivary and pancreatic secretion and has been explored as a potential effective treatment for patients with acute pancreatitis (26, 27). The combination of or the timing of ketamine/xylazine application may affect pancreatitis through mechanisms unique from other similar substances.

Ketamine/xylazine also decreased neutrophil infiltration and increased proliferation at 24 h in our study. Though there is conflicting evidence for the role of neutrophils in pancreatitis, models suggest that neutrophil infiltration correlates with the inflammatory response during acute pancreatitis. Several studies have found that neutrophil depletion results in less severe acute pancreatitis (28). In addition, ketamine/xylazine cocktail administered intravenously in sheep decreases packed cell volume from 30 to 90 min and decreases hemoglobin and neutrophils at 45 min compared with hematological parameters before ketamine/xylazine administration (29). In this study, there were no changes in lymphocytes, monocyte, or eosinophils (29). Taken together, it is possible that ketamine/xylazine leads to a depletion of neutrophils within mice, ultimately leading to a decrease in inflammatory response at 24 h after acute pancreatitis induction. This reduction in neutrophils could also promote continued injury, as early neutrophil infiltration may be important for the subsequent macrophage infiltration and tissue recovery. Proliferation and acinar regeneration are important recovery responses in cerulein-induced acute pancreatitis. The increase in proliferative cells at 24 h with Ket/Xyl pretreatment could reflect either improvement in recovery or persistent injury. At 6 h, however, while acute pancreatitis severity markers and histological findings suggest worsening pancreatitis with Ket/Xyl pretreatment, neutrophil infiltration was also low. Given the order of events associated with acute pancreatitis, it is possible, that at 6 h, the pancreatitis severity markers and histology predominantly reflect the initial acinar cell injury. Future studies should evaluate whether Ket/Xyl might impact specific inflammatory responses not examined here as well as parameters of pancreatic recovery at longer times.

Of the acinar cell responses associated with ketamine/xylazine in vivo, the effects on autophagic markers were prominent. Anesthetic agents have been associated with modulating autophagy (30, 31). Notably, ketamine/xylazine and ketamine alone have been associated with an upregulation of autophagy in skeletal muscles after general anesthesia (30, 31). α2 Agonists have also been known to increase cardiac autophagy by stimulating autophagic flux (32). In a system in which autophagy is already dysregulated, such as acute pancreatitis, ketamine/xylazine could further reduce autophagic flux. Such an effect is consistent with other parameters showing a Ket/Xyl-induced increase in pancreatitis severity.

Since ketamine/xylazine act on neural pathways, we sought to determine if they acted on acinar cells, which have cholinergic receptors, intrapancreatic nerves, or at higher neural levels. Our findings suggest that the effects of these agents are not directly on acinar cells or intrapancreatic ganglia (which have been shown to be functional in pancreatic lobules) (33, 34). However, in our study, ketamine/xylazine had little effect on cerulein-induced injury responses in pancreatic lobules. These findings strongly suggest that the enhancement of pancreatitis responses by ketamine/xylazine in vivo require an intact nervous system.

In in vivo studies, like the ketamine/xylazine combination, xylazine increased zymogen activation and decrease serum amylase levels. Ketamine, on the other hand, tended to decrease zymogen activation and increase serum amylase levels. Both ketamine pretreatment and xylazine pretreatment led to less severe acute pancreatitis compared with ketamine/xylazine combination pretreatment. These findings suggest that although xylazine may act similar to ketamine/xylazine, ketamine and xylazine have additive effects that increase its potency in acute pancreatitis severity.

Both the sympathetic and parasympathetic systems have cholinergic components. Atropine blocks the muscarinic cholinergic pathway but does not affect the nicotinic cholinergic pathway. Atropine administration in other models of acute pancreatitis such as alcoholic pancreatitis can reduce the severity of disease (35), however, we found that atropine did not modify the effects of ketamine/xylazine on pancreatitis severity. These findings suggest that the effects of ketamine/xylazine are not mediated by muscarinic pathways.

In summary, we found that ketamine/xylazine pretreatment inconsistently affects the basal pancreas and pancreatitis outcomes in the short term. This effect appears to be primarily mediated by xylazine but the addition of ketamine often had a synergistic effect on acute pancreatitis outcomes. The effects of these agents are likely mediated by neural pathways that are not muscarinic and reside outside the pancreas. Our findings are significant because of the importance of ketamine/xylazine as a primary anesthesia tool in animal models. Given the effects of ketamine/xylazine not only on immune but also on pancreatic acinar cells, studies using ketamine/xylazine, particularly in animal models of pancreatic disease should interpret their findings cautiously.

The use of ketamine/xylazine is not required in the cerulein model but the model provided a suitable preparation for examining the effects of these drugs in experimental pancreatitis. Future studies that use pancreatitis models that require the use of anesthesia, such as surgical models of pancreatitis, should consider systematically testing ketamine/xylazine effects on pancreatitis with other anesthetic options such as inhalants like isoflurane (14), to determine whether changes are anesthesia-related of pancreatitis model-related. In addition, future studies may further explore the biological relevance of the characteristics identified in this study. In addition, since drugs similar to ketamine and xylazine are considered in clinical acute pancreatitis treatment (27, 36), further assessment of the effects of ketamine and xylazine on pancreatitis should be conducted. For example, since ketamine/xylazine and xylazine inconsistently affects acute pancreatitis in our studies, exploring these effects further before assessing these agents in acute pancreatitis treatment should be warranted. Finally, since this combination is used when studying other forms of injury, its impact should be considered in other experimental settings.

DATA AVAILABILITY

All data generated or analyzed in this study are included in this published article.

SUPPLEMENTAL DATA

Supplemental data can be found at https://doi.org/10.6084/m9.figshare.14343227.

GRANTS

A National Institutes of Health Medical Student Fellowship was presented to M. Wang (DK007017) and a Veterans Administration Merit Award was presented to F. S. Gorelick (BX-003250).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

F.S.G. conceived and designed research; M.W. performed experiments; M.W. and F.S.G. analyzed data; M.W. and F.S.G. interpreted results of experiments; M.W. prepared figures; M.W. drafted manuscript; M.W. and F.S.G. edited and revised manuscript; M.W. and F.S.G. approved final version of manuscript.