Autoimmune pancreatitis (AIP) is a rare cause of chronic pancreatitis and mimics pancreatic cancer. Although there is strong interest in research, etiology and pathophysiology of AIP are still unknown. Therefore, we analyzed a total of 92 MRL/Mp-mice of either sex, which are prone to develop AIP, in four different age groups (8–12, 16–20, 24–28, and 32–40 wk). Using intravital fluorescence microscopy, histology, laboratory analysis, and Western blot, onset, severity, and pathophysiological mechanisms of AIP were evaluated. Female animals showed in vivo an age-dependent increase of intrapancreatic leukocyte accumulation, as well as a loss in functional capillary perfusion. In contrast, intrapancreatic inflammation in male mice was less pronounced and not age dependent. Furthermore, pancreatic tissue specimen of female animals exhibited major organ destruction with significantly higher values of mean pathological scores (1.5 ± 0.3 vs. ≤0.2; P < 0.05), as well as significantly increased CD4-, CD8-, CD11b-, and CD138-positive cells compared with male animals of the same age. Interestingly, there was a significant positive correlation between intravascular leukocyte adherence and the histopathological score of the pancreas, indicating a determining role of the innate immune system for the late onset of AIP. The present study shows that the onset of AIP is characterized by an inflammatory response and microcirculatory failure, most probably constituting initiators and propagators of this autoimmune disease.
- intravital microscopy
- perfusion failure
in recent years, the incidence of autoimmune pancreatitis (AIP), a benign form of chronic pancreatitis (CP), is accumulating and is getting recognized as a distinct entity worldwide (16). Furthermore, ∼2% of patients are operated on for a suspected malignant pancreatic mass (1, 21, 38, 39), and ∼22% of pancreatoduodenectomies for benign conditions (2, 14) are misdiagnosed, having instead an AIP. In the focus of etiology and pathogenesis, AIP is described as an enigma (38). Whereas CP is portrayed as a fibrotic inflammation, induced by alcohol abuse, gene mutations, special anatomic changes and/or obstructive duct lesions (23), AIP is a nonalcoholic, duct-destructive CP with periductal lymphoplasmacytic inflammation, obliterative phlebitis, and abundant IgG4-positive plasma cells (10, 15, 18, 44, 45). Clinical signs include mild abdominal pain, jaundice, recurrent episodes of acute pancreatitis, and new-onset diabetes mellitus (3, 20). Furthermore, AIP is frequently associated with other autoimmune diseases, such as Sjögren's syndrome, sclerosing extrahepatic cholangitis, and retroperitoneal fibrosis (15). In the majority, AIP affects middle-aged patients with Asian background but presents with increasing incidence in Western countries (16, 35). The present therapeutic gold standard for the treatment of AIP, and also a diagnostic criterion, seems to be its prompt response to steroid therapy (15, 16). Although AIP is rare, there is increasing significant clinical and experimental interest to understand the pathophysiology of this disease. In addition, sufficient experimental in vitro and in vivo models for the distinct analysis of AIP are lacking. Therefore, the present study was undertaken to gain new insights into the evolution of the immune response at different stages of AIP. By correlating the number of activated inflammatory cells and microcirculatory changes with the disease progression, we may delineate pathomechanisms responsible for the disease induction and progression in an animal model of mice prone to the development of AIP.
MATERIAL AND METHODS
A total of 92 MRL/Mp-mice of either sex were used for the study in four different age groups. The animals were housed in a facility with a 12-h light/dark cycle and had access to standard laboratory chow and water ad libitum. The experiments were conducted in accordance with the German legislation on protection of animals and the NIH Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council) and approved by the local animal care committee (LALLF M-V/TSO/722.3-3.1-026/07).
Microsurgical preparation of the pancreas.
For intravital fluorescence microscopy of the pancreas, animals were anesthetized with ketamine/xylazine (90/25 mg/kg body wt ip) and placed in supine position on a heating pad to maintain a constant body temperature of 37.5°C. A catheter (PE-28) in the right jugular vein served as a route for administration of fluorescent dyes. The model to study pancreatic microcirculation has previously been established by our group in rats and has now been transferred to mice (12, 28, 41, 42). After transverse laparotomy and dissection of the omentum from the greater curvature of the stomach, the tail and parts of the corpus of the pancreas were gently exteriorized on a plasticine stage, allowing ideal placement of the tissue, which guaranteed adequate homogeneous focus level and minimal respiratory movement for the microscopic procedure. To keep the exteriorized pancreas moist and to exclude effects of ambient oxygen, the pancreatic surface was covered with an oxygen-impermeable Saran wrap.
Experimental groups and protocol.
A total of 92 animals have been included in the study and have been randomly allocated into eight experimental groups. Animals were studied by intravital fluorescence microscopy in four different groups of each sex with the age of 8–12 (n = 10 males; n = 10 females), 16–20 (n = 10 males; n = 10 females), 24–28 (n = 10 males; n = 10 females) or 32–40 wk (n = 15 males; n = 13 females), followed by sampling of blood and tissue for subsequent laboratory analysis.
Intravital microscopy of the pancreas and microcirculatory analysis.
Anesthetized mice were placed under an intravital fluorescence microscope equipped with a 100-W mercury lamp (Axiotech vario; Zeiss, Jena, Germany) and allowed the analysis of pancreatic microcirculation and cell-to-cell interaction, using an ×20 and ×40 long-distance objective. Microscopic images were taken and recorded by a video system (S-VHS Panasonic AG 7350-E; Matsushita, Tokyo, Japan) for off-line evaluation via a charge-coupled device video camera (FK 6990A-IQ; Pieper, Berlin, Germany) and monitored on a television screen. Quantitative off-line analysis of the videotaped images was performed by means of a computer-assisted image analysis system (CapImage; Dr. Zeintl Software, Heidelberg, Germany). Contrast enhancement for assessment of capillary perfusion was achieved by intravenous injection of 0.1 ml 2% FITC-labeled dextran (MW 150 kDa; Sigma, Deisenhofen, Germany) and blue light epi-illumination (450–490 nm). Leukocyte-endothelial cell interaction was documented after in vivo staining of white blood cells by 1% rhodamine-6G (0.1 ml) and green light epi-illumination (510–560 nm) (28). Intravital microscopic analyses of the microcirculation included the determination of venular vessel diameter, red blood cell velocity (RBCV) both in capillaries and venules, and the functional capillary density (FCD) (12, 41, 42). Moreover, we assessed microvascular leukocyte count and flow behavior of leukocytes in postcapillary venules, classified according to their interaction with the endothelial lining as adherent, rolling, or free-flowing (nonadherent) cells (28).
Quantitative microcirculatory analysis.
Microvascular diameters (μm) were assessed in ten individual capillaries per observation field and eight observation fields per animal. Within these capillaries, RBCV was analyzed using the line-shift-diagram method (24). FCD as a parameter of nutritive perfusion was defined as the total length of red blood cell-perfused microvessels per observation area, given in centimeters per centimeters squared, and was assessed in eight randomly selected observation areas using CapImage software. In postcapillary venules, leukocyte flux was determined by counting all cells passing the vessel segment under investigation. Adherent leukocytes were defined in each vessel segment as cells that did not move or detach from the endothelial lining within an observation period of 20 s and were given as number of cells per millimeter squared of endothelial surface, calculated from diameter and length of the vessel segment studied, assuming cylindrical geometry. Rolling leukocytes were defined as those white cells moving at a velocity less than 2/5 of that of erythrocytes in the centerline of the microvessel and are given as percentage of nonadherent leukocytes passing through the observed vessel segment within 30 s.
Histology and immunohistochemistry.
The pancreas was excised after intravital microscopy and was fixed in 4% phosphate-buffered formalin for 3 days and then embedded in paraffin. From the paraffin-embedded tissue blocks, 2-μm sections were serially cut and stained with hematoxylin-eosin for assessment of routine histology. Histopathological evaluation of pancreatic lesions was performed under a light microscope as previously described (17). Severity of the lesions in each mouse was scored on a 0–4 grade, on the basis of the histopathological changes as follows: 0, pancreas without mononuclear cell infiltration, indicating almost normal tissue morphology and organ integrity; 1, mononuclear cell aggregation and/or infiltration within the interstitium without any parenchymal destruction; 2, focal parenchymal destruction with mononuclear cell infiltration; 3, diffuse parenchymal destruction but retained some intact parenchymal residue; 4, almost complete mononuclear cell infiltration of pancreatic tissue except pancreatic islets and destruction of acini being either destroyed or replaced with adipose tissue. To estimate the incidence of pancreatitis in MRL/Mp+/+ mice, pancreatic lesions that scored >2 were defined as positive for AIP.
For immunohistochemical demonstration of CD4, CD8, CD11b, and CD138, 4-μm serially cut sections were collected on poly-L-lysine-coated glass slides and were treated by microwave for antigen unmasking. After being equilibrated to room temperature, sections were incubated with a primary antibody for 18 h at 4°C [for CD4: rat monoclonal anti-CD4 (RM4-5), 1:100, Abcam, Cambridge, UK; for CD8: rat monoclonal anti-CD8 IgM, 1:100, Santa Cruz Biotechnology, Heidelberg, Germany; for CD11b: rat monoclonal anti CD-11b (M1/70.15), 1:100, Abcam; and for CD138: rat monoclonal anti-CD138 IgG, 1:500, BD Pharmingen, Heidelberg, Germany]. This was followed by the exposure of a secondary antibody (mouse anti-rat AP, 1:200, Santa Cruz Biotechnology) for 30 min. Fuchsin (DakoCytomation, Hamburg, Germany) was used as chromogen and sections were counterstained with hemalaun. CD4-, CD8-, CD11b-, and CD138-positive cells were assessed by counting cells with positive cellular staining and given as cells per millimeter squared.
Total RNA was isolated from homogenized pancreatic tissue by the guanidinium thiocyanate/phenol method according to Sparmann et al. (34). Afterward, RNA was reverse transcribed into cDNA by means of TaqMan Reverse Transcription Reagents (Applied Biosystems, Foster City, CA) and random hexamer priming. Target cDNA levels were analyzed by quantitative real-time PCR using TaqMan Universal PCR Master Mix and the following Assay-on-Demand mouse gene-specific fluorescently labeled TaqMan MGB probes in an ABI Prism 7000 sequence detection system (Applied Biosystems): Mm00801778_m1 (IFN-γ), Mm00434228_m1 (IL-1β), Mm00434256_m1 (IL-2), and Mm00446190_m1 (IL-6). Expression of the housekeeping gene GAPDH was monitored using TaqMan rodent GAPDH control reagents. PCR conditions were as follows: 95°C for 10 min, 55 cycles of 15 s at 95°C/1 min at 60°C. PCR reactions were performed in triplicate. The relative expression of each mRNA compared with GAPDH was calculated according to the equation ΔCt=Cttarget − CtGAPDH. For further calculation, the sample with the lowest ΔCt value, corresponding to the highest target gene expression level, was taken as reference. The relative amount of target mRNA in individual samples was expressed as 2−(ΔΔCt), where ΔΔCtsample = ΔCtsample − ΔCtreference.
Under ether inhalation anesthesia, animals repetitively had blood taken by puncture of the retrobulbar venous plexus to determine lipase, amylase, basic phosphatase, and glucose activities using a spectrophotometric analysis system (Cobas CIII; Roche Diagnostics, Mannheim, Germany). After intravital microscopic analysis of the pancreatic microcirculation by puncture of the vena cava, blood was sampled for subsequent analysis of systemic red and white blood cell counts, hemoglobin, and hematocrit using an automatic cell counter (Sysmex KX21; Sysmex, Norderstedt, Germany), as well as for the determination of the above mentioned serum parameters. In addition, concentrations of antinuclear antibodies (ANA) were measured by an ELISA kit for mouse ANA according to the manufacturer's instructions (ANA ELISA kit 5210; Alpha Diagnostics International, San Antonio, TX). The detection limit of this ANA ELISA is 0.25 μg/ml.
All data are given as mean ± SE. Data were analyzed for normality and equal variance across groups. Differences between groups were assessed using one-way ANOVA followed by the appropriate post hoc comparison test including correction of the α-error according to the Bonferroni probabilities for repeated measurements. Overall statistical significance was set at P < 0.05. Statistics were performed using the software package SigmaStat (Version 3.5; Jandel, San Rafael, CA).
Over the complete observation period in each of the observed experimental groups, animals did not reveal any signs of discomfort or disturbances in eating or drinking habits. Furthermore, animals did recover well after each of the repetitive blood withdrawals and could constantly gain weight over the observational time period of 40 wk (data not shown).
Intravital microscopic analysis showed an age- and sex-related increase of inflammation. Herein, the rolling fraction of leukocytes markedly increased in both sexes over time. Values of adherent leukocytes also increased time dependently, with significantly elevated intrapancreatic leukocyte accumulation in females (Fig. 1, A and B). Intrapancreatic inflammation in male animals was clearly less pronounced and did not depend on age (Fig. 1B). RBCV did not differ among the different age groups studied (data not shown). However, as seen by allocation to the histopathological score, there is a marked decrease of both RBCV and FCD of the pancreatic parenchyma in animals with an apparent development of AIP (Score ≥ 2; Fig. 2, A–C).
In the 8–12-wk-old animals hematoxylin-eosin-stained paraffin sections of pancreatic tissue specimen showed an intact pancreatic morphology (histopathological score 0 ± 0). In old animals, and particularly in female mice, histological sections presented with rarefied pancreatic parenchyma, massive leukocytic infiltration, and fatty degeneration. In the group of 24–40-wk-old mice, female animals presented with an incidence of AIP between 20–54% (Fig. 3C), whereas in male animals AIP could not be recognized at all (Fig. 3C). Furthermore, pancreatic tissue specimen of female animals showed significantly worse parenchymal destruction as given by a mean histopathological score of 1.5 ± 0.3 in females, whereas male mice revealed no AIP-related disintegration of tissue (Fig. 3C). There was a significant correlation between in vivo leukocyte adherence and the histological score (Fig. 4), implying some dependency between microvascular function and manifestation of disease.
To further characterize the inflammatory infiltration within the pancreatic tissue specimen, immunohistochemistry for CD4- and CD8-positive T lymphocytes, macrophages (CD11b), and plasma cells (CD138) was performed. Pancreatic tissue of female animals within the age of 32–40 wk revealed significantly increased CD4-, CD8-, CD11b-, and CD138-positive cells compared with male animals of the same age (Fig. 5).
RNA-expression analysis of inflammation-related cytokines in the group of 32–40-wk-old mice revealed that female animals expressed more IFN-γ, IL-2, and IL-6 mRNA in the pancreatic tissue than male mice (Fig. 6). These data are in agreement with the differences in pancreatic histopathology between both sexes.
Analysis of serum markers of pancreatitis and glucose could not demonstrate any predictable change for the onset of AIP. Determination of full blood cell counts after intravital microscopy showed physiological values for leukocytes, erythrocytes, hemoglobin, and hematocrit, and no differences could be observed either between sexes or between the different age groups (data not shown). Young animals (8–12 wk) did not present any detectable concentrations of ANA (detection limit 0.25 μg/ml), which correlates with the histopathology score of 0. In contrast, animals with the age of 24–28 wk revealed ANA concentrations about 70 μg/ml; especially, female animals with the age of 32–40 wk presented with highest values of ∼160 μg/ml (Fig. 7).
AIP is a recently more recognized entity of disease attributable to the mimicking of pancreatic cancer. Whereas diagnostic criteria and techniques have been described in a rather detailed way (7, 16, 19, 31), etiology and pathophysiology of human AIP are still unknown. Although the autoimmune entity of AIP is not clarified (23), there is some evidence that this might be caused by a systemic IgG4-complex disease (10). Moreover, autoantibodies against carbonic anhydrase II and ANAs are found, for instance (3, 4, 22). Though microcirculatory deteriorations are well established as being important in the initiation and progression of both acute pancreatitis and CP (12, 28, 40), pancreatic microcirculation has not systematically been studied in animals prone to autoimmune diseases.
The technique of intravital fluorescence microscopy employed in the present study proved itself useful in the simultaneous and serial in vivo analysis of the course of microvascular and cellular aspects and their relationship in consecutive stages of pancreatic inflammation (28). The present comprehensive analysis provides for the first time in vivo evidence that the late onset of AIP is likely triggered by an early inflammatory response of the innate immune system followed by a reduction of nutritive organ perfusion. This was accompanied by a significant increase of CD4+ and CD8+ T lymphocytes, as well as macrophages and activated B cells within the inflammatory foci of the pancreatic tissue, especially of old female animals. Furthermore, this process was associated with a marked increase of systemic concentrations of ANAs over time. Thus mice reveal some of the definitive criteria for AIP, underlining their genetic proneness for the development of autoimmune diseases.
Early changes in the immune system yielding in a manifestation of disease could be shown in many different illnesses, especially in patients with autoimmune disease. Herein, it could be shown that the upregulation of proinflammatory mediators and molecules at early stage represented a necessary step in the development of rheumatoid arthritis (11). In an animal model of atopic dermatitis, it could also be demonstrated that there is a continuous and progressive migration of activated inflammatory cells from secondary lymphoid organs into the skin that might lead to the development and maintenance of atopic dermatitis (8). Furthermore, patients suffering from inflammatory bowel disease presented with an immunoregulation, which varies with the course of disease. A very early onset of disease appears with an immune response that looks similar to an acute infectious process, whereas T cell function gets lost with progression of the disease (26). The activation of leukocytes and microvascular endothelium has also been demonstrated as a consistent early event in the evolution of acute pancreatitis, which correlates with the clinical severity of disease (29). Moreover, it has been proposed that pancreatitis occurs because of the excessive leukocyte activation and its severe consequences, i.e., lipid peroxidation (27, 33).
As detailed quantitative analysis confirmed that also the number of perfused capillaries significantly correlates with the severity of acute pancreatitis (5, 25, 29), this entity seems to also contribute to the initiation of AIP as given by the continuous decrease in functional capillary density over time in our study. This reduction comprises insufficient nutritional supply, especially of oxygen, to the tissue. In consequence, this could lead to a hypoxia-driven tissue collapse with consecutive organ remodeling (12), all representing conditions that might support the development of an AIP. Several mechanisms and mediators are associated with microvascular perfusion failure. Therefore, the results of the present and previously published studies suggest that cytokines (bradykinin, substance P, IFN-γ, and IL-6) and proteolytic enzymes (trypsinogen and cathepsin B) are highly involved in the complex pancreatitis-associated microcirculatory disorder (9, 13, 29). In the present study, proinflammatory cytokines hallmarking pancreatitis were found upregulated. Especially IL-1β was seen most elevated, a cytokine that is known to accelerate the inflammation cascade in acute pancreatitis (36). Furthermore, IFN-γ, IL-6, and IL-2 were found upregulated, all mediators that likely contribute to capillary perfusion failure, leukocyte adherence, and immunodepletion (6, 32, 37, 43).
The fact that a remarkable inflammatory response and perfusion failure preceded manifestation of AIP underlines the specific temporal pattern of the innate immune response and represents an essential precondition for the development of disease. Thus early inflammation and microcirculatory disorders should be an ongoing focus for future research to determine whether leukocyte activation and perfusion failure forecast the severity of disease. Follow-up studies, therefore, should focus on specific mechanisms of increased leukocyte endothelial interaction since it is also described that intercellular adhesion molecule-1 expression is increased in the hepatic microcirculation in MRL/lpr mice (30).
The study was in part supported by a grant of the Medical faculty of the University of Rostock (FORUN; to H. Sorg, R. Jaster, S. Ibrahim, and B. Vollmar).
The authors kindly thank Berit Blendow, Doris Butzlaff, Dorothea Frenz, and Kathrin Sievert-Küchenmeister (Institute for Experimental Surgery with Central Animal Care Facility, University of Rostock) for excellent technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society