Sex-dependent modulation of caerulein-induced acute pancreatitis in C57BL/6J mice

Abstract

Acute pancreatitis (AP) is a life-threatening condition, with a higher mortality rate in men than women and in which estrogens might play a protective role. This study aimed to investigate sex-dependent differences in a mouse model of caerulein-induced AP. Thirty-six C57BL/6J mice (19 females and 17 males) were treated intraperitoneally with phosphate-buffered saline or caerulein, and sacrificed 12 hours, 2 days, or 7 days after the last injection. Blood was collected for amylase, lipase, and glucose determination. Severity and extent of inflammation, apoptosis, and acinar to ductal metaplasia (ADM) in pancreatic tissue were scored histologically and total macrophages, major histocompatibility complex (MHC)-II+ cells, M2 macrophages, T and B cells, neutrophils, apoptosis, and ADM were marked immunohistochemically and quantified by digital image analysis. Serum amylase had a peak at 12 hours, without differences between the sexes. In females, pancreatitis reached a peak at 12 hours with a fast recovery while, in males, the peak was delayed to day 2 with residual apoptosis still present. Macrophages were the main inflammatory cell population, followed by T cells, B cells and neutrophils, without differences between sexes. In males, CD206+ cells and apoptosis were higher at both days 2 and 7, and cytokeratin-19+ (CK19+) ADM was higher at day 7 compared with females. The results of this study revealed a faster onset and resolution of caerulein-induced AP in female mice compared with male mice, supporting a sex-dependent modulation of acute pancreatitis.

Acute pancreatitis (AP) in humans is a sudden life-threatening condition caused by the release and activation of acinar enzymes leading to glandular self-digestion and potentially fatal systemic inflammatory states.^8,36 AP prevalence is increasing worldwide, and it has been linked to numerous risk factors and etiologies in human patients including smoke, alcohol consumption, gallstones, hyperlipidemia, hypercalcemia, drug reactions, and trauma.^26,28,55

In the last decades, sex has been extensively recognized as an important biological variable modulating both physiological and pathological processes in different organs, including the pancreas.^11,22,48 Although the overall incidence of pancreatitis seems not to differ significantly between sexes in humans, the rate of some forms of pancreatitis, as well as pancreatic-related mortality, is higher in males than females, with estrogens proposed as either potentially protective hormones in premenopausal women or as predisposing factors to the development of pancreatitis, particularly in cases induced by estrogen treatment.^8,48

Biomedical research has mainly relied upon animal models, particularly mice and rats, to get insights into different features of normal and diseased pancreas. ⁵² Different mechanical and chemical AP murine models have been developed with supramaximal caerulein-induced AP being one of the most used models to investigate pancreatic inflammatory conditions.^42,52 This cholecystokinin analogue oligopeptide induces the release of pancreatic acinar enzymes, causing a self-limiting dose-dependent AP.^30,40

Although the exact chain of events occurring after caerulein stimulation has not been fully elucidated yet, it has been demonstrated that this molecule causes pancreatic trypsinogen activation which rapidly leads to reactive oxygen species (ROS) formation, nuclear factor-kB (NF-kB) and intrinsic apoptosis pathways activation, and damage-associated molecular patterns (DAMPs) release with subsequent immune cells recruitment, pattern recognition receptors (PRRs)-mediated activation of immune cells, and production of cytokines.^17,20,49 Both innate and adaptive immune cells play a role in the pathogenesis of pancreatitis. ⁴⁹ Macrophages appear to be key players as they are involved in the development of acute inflammation as classically activated (M1) macrophages and in the reparative response as alternatively activated (M2) macrophages in more advanced phases of inflammation.^16,34 The subsequent pancreatic regeneration is unique among the body organs and occurs through a process known as acinar to ductal metaplasia (ADM) in which damaged acinar cells transition to ductal-like precursors expressing cytokeratin-19 (CK19), which are then capable of proliferation and differentiation to form new acini, re-establishing the normal pancreatic architecture. ²⁹

Animal studies investigating the role of sex in AP have resulted in contrasting results, with some studies highlighting sex-dependent differences in the extent and severity of pancreatitis and others failing to detect such differences.^{7,13,19,33,35} More investigations into the sex-related differences in AP and chronic pancreatitis are needed to improve prevention, clinical management, diagnostics, and therapeutics. The aim of this study was, therefore, to investigate sex-dependent differences in AP development and subsequent regeneration in a mouse model of caerulein-induced AP in C57BL6/J mice through a combination of clinical chemistry, histology, immunohistochemistry, and digital image analysis.

Materials and Methods

Animals

The use of animals in this study was approved by the internal organization for animal welfare of the University of Milan and by the Italian Ministry of Health (authorization number 970/2020-PR). A total of 36 (19 females and 17 males) C57BL/6J mice of 4 months of age were purchased from Jackson Laboratories (France) and housed at the animal facility of the Department of Pharmacological and Biomolecular Sciences at the University of Milan. Animals were allowed food and water access ad libitum and kept in temperature-controlled facilities on a 12-hour light and dark cycle.

Pancreatitis Induction

Mice were injected intraperitoneally with phosphate-buffered saline (PBS) (vehicle) (14 mice) or 70 μg/kg of caerulein (Sigma-Aldrich C9026) (20 mice) every hour, 8 times a day, for 2 consecutive days. ¹² Mice were sacrificed at 12 hours, 2 days, and 7 days after last injection (Table 1). Two nonmanipulated control mice were also included in the study and sacrificed after 12 hours together with treated animals. All mice were weighed daily before and after treatment until sacrifice.

Table 1.

Summary of treatment groups.

Time Point	Group	Male Number	Female Number	Total Mice
12 hours	Not manipulated	1	1	2
	Vehicle (PBS)	3	3	6
	Caerulein	3	4	7
Day 2	Vehicle (PBS)	2	2	4
	Caerulein	3	3	6
Day 7	Vehicle (PBS)	2	2	4
	Caerulein	3	4	7
Total mice		17	19	36

Abbreviations: PBS, phosphate-buffered saline.

Clinical Chemistry

After general anesthesia induction through subcutaneous administration of xylazine and ketamine (6 and 78 mg/kg, respectively), blood was collected through intracardiac puncture using a 23 G needle. Blood was collected in plain tubes (Eppendorf, Germany, Hamburg) and left to incubate for 30 minutes at room temperature to allow clot formation. Subsequently samples were centrifuged for 10 minutes at 2000 x g at 4°C (Eppendorf centrifuge 5415R, Eppendorf, Germany, Hamburg) and the harvested sera were frozen at −20°C until analysis. Glucose concentration (GOD-POD [glucose oxidase-peroxidase] method) and amylase (2-chloro-4-nitrophenyl maltotrioside [CNP-G3] method) and lipase (DGGR [1,2-o-dilauryl-rac-glycero-3-glutaric acid-(6'-methylresorufin) ester] method) activity were measured on sera with an automated spectrophotometer (BT3500, Biotecnica Instruments s.p.a. Rome, Italy) using reagents provided by the manufacturer of the instrument. Quality control procedures were performed before each batch of analysis with 2 levels of human-based control sera and with a human-based calibrator provided by the manufacturer of the instrument. To reduce the analytical variability of results, analyses were performed in batch within 10 days from sampling.

Histology

At sacrifice, the pancreas was carefully and gently separated from the duodenum, stomach, and spleen using blunt-ended forceps and scissors, placed in a cassette on a dry sponge, and gently spread out to ensure its flattening. The pancreas was subsequently fixed in 10% neutral-buffered formalin for 24 hours at room temperature, routinely processed for paraffin embedding, sectioned at 4 μm thickness, and stained with hematoxylin and eosin. Pancreas sections were blindly evaluated according to a newly designed semiquantitative grading system reported in Supplemental Table S1 by 2 pathologists (LB and CR) who reached a consensus. The examined findings included interstitial edema, interstitial inflammatory cell infiltration, acinar cell apoptosis, ADM, and extent of inflammation (ie, the amount of pancreatic tissue involved by inflammation), with a score ranging from 0 to 4 based on increasing severity. A total pancreatitis histological score was, then, calculated by adding together the interstitial edema, interstitial inflammatory cell infiltration, and apoptosis scores and multiplying the sum by the extent of inflammation score. No other organ was subjected to histologic analysis.

Immunohistochemistry

To profile the leukocytes’ populations, apoptosis and ADM, the following primary antibodies were applied: IBA1, F4/80 (total macrophages), major histocompatibility complex (MHC)-II, CD206, HO-1 (M2 macrophages), B220 (B lymphocytes), CD3ε (T lymphocytes), Ly6G (neutrophils), cleaved caspase-3 (apoptosis), and CK19 (ducts and ADM). Details on the applied antibodies are reported in Supplemental Table S2. For all markers, except for F4/80, 4 μm sections of pancreas underwent deparaffinization and heat-induced epitope retrieval (HIER) in a water bath for 40 minutes at 100°C (Dewax and HIER Buffer H, Thermo Scientific Lab Vision, cat. No. TA-999-DHBH). For F4/80 staining, slides were deparaffinized, rehydrated, rinsed in PBS, and unmasked with 1% trypsin (Sigma-Aldrich, T1426) for 30 minutes at 37°C. All slides were rinsed in PBS and placed in an autostainer (Lab Vision Autostainer 480S-2D Thermo Fischer Scientific) after application of Peroxidase-Anti-Peroxidase (PAP) pen (Liquid Daido Sangyo Co., Ltd.). Endogenous peroxidase activity was blocked by incubating sections with 3% H₂O₂ for 10 minutes. Slides were rinsed and incubated with PBS containing 10% normal goat serum (IBA1, CD206, HO-1, cleaved caspase-3, and CK19) or 10% normal rabbit serum (F4/80, CD3e, B220, Ly6G, and MHC-II), for 30 minutes at room temperature to prevent nonspecific background labeling and subsequently incubated for 1 hour at room temperature with primary antibodies. Sections were rinsed in PBS and incubated with a biotinylated secondary antibody (goat antirabbit, Vector Laboratories, USA, cat. No. BA-1000 for IBA1, CD206, HO-1, cleaved caspase-3, and CK19; rabbit antirat, Vector Laboratories, USA, cat. No. BA-4001 for F4/80, B220, MHC-II, and Ly6G; rabbit antigoat Vector Laboratories, USA, cat. No. BA-5000 for CD3e) and labeled through avidin-biotin-peroxidase procedure (VECTASTAIN Elite ABC-Peroxidase Kit Standard, Vector Laboratories, USA, cat. No.PK-6100). The immunoreaction was visualized with 3,3’-diaminobenzidine substrate (DAB, Peroxidase DAB Substrate Kit, Vector Laboratories, USA, cat. No. SK-4105). Sections were counterstained with Mayer’s hematoxylin, dehydrated in a graded alcohol series and coverslipped with resinous mounting medium. Adequate positive controls were used for all markers including samples of murine spleen (IBA1, F4/80, CD206, HO-1, MHC-II, B220, and CD3e), pancreas (CK19), lymph node (cleaved caspase-3), and a sample of preputial gland suppurative adenitis (Ly6G).

All slides were digitalized in 40× high-resolution mode (0.23 µm × pixel) through the NanoZoomer-XR Digital slide scanner (Hamamatsu, C12000) at the Unitech Nolimits facility (UNIMI, Milano) and were visualized using the NDP.view2 Viewing software (Hamamatsu, U12388-01).

Digital Image Analysis

The total pancreatic area and the pancreatitis area (ie, the area affected by pancreatitis) were manually outlined excluding peripancreatic adipose tissue, pancreatic lymph nodes, and fat-associated lymphoid clusters, using QuPath software. ² Briefly, after uploading digital slides in QuPath, all pancreatic tissue was manually outlined using the magic wand tool of QuPath. Next, the wand tool or the manual selection tool were used to outline areas of pancreas affected by inflammation (pancreatitis areas), ie, areas characterized by interstitial edema, inflammatory cells infiltration, apoptosis, and ADM. The percentage of pancreatitis area to total pancreas area was then calculated. In sections immunolabeled with IBA1, F4/80, MHC-II, CD206, HO-1, B220, CD3ɛ, and Ly6G, positive cells were detected using the positive cell detection function of QuPath, running a single threshold analysis after optimizing nucleus parameters, intensity parameters, and reducing cellular fragmentation by modifying sigma and median filter radius parameters. Data were exported in Excel and, for each marker, the number of positive cells per mm² of pancreatitis area was calculated. For cleaved caspase-3 and CK19, the positive area was calculated after running a color deconvolution through the pixel classifier tool (create thresholder function). A threshold that could avoid nonspecific signals was set and the positive area calculated. Data were exported in excel and the percent of positive area per pancreatitis area for each marker was calculated.

Statistical Analysis

Data were analyzed using Graph Pad Prism version 9.0 (GraphPad Software, San Diego, CA, USA). Nonmanipulated and PBS (vehicle) treated animals were considered together as controls in the analysis since no differences were identified between the 2 groups. Comparisons between the groups were performed using the Mann-Whitney U-test and Kruskal-Wallis test; P values < .05 were considered statistically significant. In addition, Spearman’s correlation tests were conducted to assess correlations between the score of inflammation and the percent of pancreatitis area, between the score of apoptosis and the percent of cleaved caspase-3/ positive area and between the score of ADM and the percent of CK19 positive area measured in QuPath.

Results

Clinical Data

After the first day of injections, caerulein-treated animals showed moderate discomfort, as demonstrated by orbital tightening, nose bulge, changed ear position, and scruffy coats. Caerulein-treated mice were also less active compared with vehicle-treated animals. All these signs were exacerbated following the second day of injections, at 12 hours. The animals returned to normal behavior by day 2 after the last injection. No differences were observed between males and females.

A significant reduction in body weight was observed in both caerulein-treated male and female mice after the first day of injections compared with vehicle-treated mice. The body weight of caerulein-treated female mice returned to basal levels 3 days after the last caerulein injection, while the body weight of caerulein-treated male mice took a longer duration (day 6) to return to basal levels (Fig. 1). A significant difference in body weights was observed between vehicle and treated female mice at day 0 (P < .0001), day 1 (P < .0001), and day 2 (P = .0006), and between vehicle and treated male mice at day 0 (P < .0001), day 1 (P < .0001), day 2 (P = .0003), and day 3 (P < .0159) after treatment. No apparent differences were detected in body weights of vehicle-treated and nonmanipulated control animals at all the time points analyzed. No significant changes were detected between caerulein-treated male and female mice at all time points analyzed.

Figure 1.

Body weight changes in acute pancreatitis in (a) male and (b) female mice. Body weights are expressed in percent (%) compared with day –1 weights. Data are expressed as mean values ± standard deviation. P = < .0001, ***P = .006, *P = .0159.

Clinical Chemistry

In both sexes, pancreatic amylases were higher at 12 hours after last caerulein administration as compared with vehicle-treated animals (P = .0082), with a progressive return to basal levels at days 2 and 7. The lipase activity and glucose concentration were not affected by caerulein administration. No significant differences were observed between male and female mice in amylase and lipase activity, and glucose concentration in all time points (Supplemental Fig. S1).

Histology

Results of the histologic evaluation are summarized in Table 2. Raw data are reported in Supplemental Table S3. Representative images are provided in Fig. 2.

Table 2.

Acute pancreatitis semiquantitative evaluation results summary. Results are expressed as median values. Total pancreatitis histological scores were calculated as the sum of edema, interstitial inflammatory cell infiltration, and apoptosis scores multiplied per the extent score. ADM was not used to calculate the total score since it was considered part of the healing process.

Group	Time Point	Examined Mice (n)	Sex (n)	Edema	Interstitial Inflammatory Cell Infiltration	Inflammatory Cell Types	Apoptosis	ADM	Extent of Inflammation	Total Pancreatitis Histology Score
Not manipulated	12 hours	1	Male (1)	0	0	n/a	0	0	0	0
		1	Female (1)	0	0	n/a	0	0	0	0
Control	12 hours	3	Male (3)	0	0	n/a	0	0	0	0
		3	Female (4)	0	0	n/a	0	0	0	0
	Day 2	2	Male (3)	0	0	n/a	0	0	0	0
		2	Female (3)	0	0	n/a	0	0	0	0
	Day 7	2	Male (3)	0	0	n/a	0	0	0	0
		2	Female (4)	0	0	n/a	0	0	0	0
Caerulein	12 hours	3	Male (3)	3	3	N, M, rare L, and P	1	0	2	14
		4	Female (3)	2.5	3.5	N, M, rare L, and P	0	0	3.5	21.5
	Day 2	3	Male (3)	2	3	N, M, rare L, and P	4	2	2	18
		3	Female (3)	1	2	N, M, rare L, and P	2	2	1	5
	Day 7	3	Male (3)	0	1	M, L, P, and rare N	3	3	1	4
		4	Female (3)	0	1	M, L, P, and rare N	0.5	2	1	1.5

Abbreviations: n/a, not applicable; N, neutrophils; M, macrophages; L, lymphocytes; P, plasma cells; ADM, acinar to ductal metaplasia.

Figure 2.

Caerulein-induced acute pancreatitis, mouse. Hematoxylin and eosin. (a and b) Acute pancreatitis 12 hours after last caerulein treatment. (a) Male mice are characterized by lower interstitial inflammatory cells compared with (b) female mice. (c and d) Acute pancreatitis on day 2 after last caerulein treatment. (c) Male mice have more extensive inflammation compared with (d) female mice. (e and f). Acute pancreatitis on day 7 after last caerulein treatment. Residual areas of pancreatitis, characterized by fibroplasia and mononuclear inflammation are seen in both (e) males and (f) females.

Lesions were identified in the pancreas of 20 of 20 (100%) caerulein-treated mice at all examined time points (12 hours, day 2, and day 7) and involved the exocrine pancreas and interstitium, but spared the endocrine pancreas and ducts. No lesions were observed in nonmanipulated or vehicle-treated mice. Pancreatitis was characterized by interstitial edema, interstitial infiltration by inflammatory cells, apoptosis of acinar cells, and the presence of tubular-like structures resembling pancreatic ducts (consistent with ADM) starting from day 2.

At 12 hours, in both sexes, there was moderate to severe interstitial edema with frequent individualization of pancreatic acini, interstitial infiltration of inflammatory cells mainly composed of macrophages and neutrophils admixed with rare lymphocytes and plasma cells, and rare apoptotic acinar cells (observed only in male mice). The extent of inflammation was higher in female mice than in male mice (Table 2).

At day 2, interstitial edema decreased in both sexes, while interstitial inflammatory cells and apoptosis, the latter involving pancreatic acinar cells and recruited inflammatory cells were more numerous in males compared with females (Fig. 3a, b; Fig. 4a, b). Early ADM was present at day 2 in both sexes, with no differences between the sexes (Fig. 5a, b).

Figure 3.

Caerulein-induced acute pancreatitis, female mouse. Day 2 after last caerulein treatment. A higher number of apoptotic cells (arrows, a–b) and cleaved caspase-3 immunolabeled cells (c, d) are observed in males (a, c) compared with females (b, d). (a, b) Hematoxylin and eosin. (c, d) Immunohistochemistry for cleaved caspase-3.

Figure 4.

(a) Semiquantitative assessment of apoptosis. Both sexes have high apoptotic scores at day 2, which are higher in males (M) than females (F). (b) Percentage of cleaved caspase-3 positive area measured using digital image analysis. Females exhibit lower apoptotic activity on day 7.

Figure 5.

Caerulein-induced acute pancreatitis, mouse. (a, b) Day 2 after last caerulein treatment. Ill-defined ducts still resembling pancreatic acini (arrows) and characterized by a mild expression of cytokeratin 19 (CK19) (b) consistent with acinar to ductal metaplasia (ADM) are evident. (c, d). Day 7 after last caerulein treatment. Well-defined, CK19 positive ducts, morphologically indistinguishable from normal intralobular and interlobular ducts, are present. (a, c) Hematoxylin and eosin. (b, d) Immunohistochemistry for CK19.

Figure 6.

(a) Semiquantitative assessment of acinar to ductal metaplasia (ADM). Both sexes show a progressive increase in ADM scores from 12 hours to day 7, with males (M) having higher scores than females (F) by day 7. (b) Percentage of cytokeratin 19 (CK19) positive area measured using digital image analysis. Males exhibit higher ADM on day 7.

At day 7, there was an almost complete recovery in both sexes with only residual scattered, patchy areas of pancreatitis characterized by fibroplasia, acinar cell atrophy, inflammatory mononuclear infiltration, and residual apoptosis, which was increased in males compared with females (Fig. 4a, b). In the residual affected areas, the presence of well-defined, neoformed pancreatic ducts was slightly higher in males than in female mice, indicative of ADM (Fig. 5c, d; Fig. 6a, b).

Based on the total pancreatitis histological score, females reached a peak of pancreatitis severity at 12 hours followed by a progressive decrease at later time points and almost complete resolution at day 7, while, in males, the peak of pancreatitis severity was delayed at day 2 with residual lesions (mainly apoptosis) still detectable at day 7 (Fig. 7a). Digital image analysis of the extent of inflammation confirmed that, although not statistically significant, at 12 hours, the percent of inflammation area was higher in females than males (64% in females and 33% in males), while, at day 2, the extent of inflammation was higher in males than females (mean of 24% in males and 8% in females). No differences were identified at day 7 (2.70% in males and 2.53% in females) (Fig. 7b). The percentage of pancreatitis measured via digital image analysis correlated with the histological semiquantitative score, with an r value of 0.9375 (Spearman’s correlation test) (Supplemental Fig. S2a).

Figure 7.

(a) Total pancreatitis histology score. Data are expressed as median and interquartile range (b) Extent of inflammation evaluated by digital image analysis in male (M) and female (F) mice. Data are expressed as mean values ± SEM. In male mice, the peak of pancreatitis is delayed compared with female mice (day 2 and 12 hours, respectively).

Immunohistochemistry

Inflammatory cells

The immunohistochemical analysis of immune cell infiltrates revealed that IBA1+ and F4/80+ cells were the most prominent inflammatory cell population in the pancreas of treated animals at all examined time points, followed by CD206+, HO-1+ macrophages and by MHC-II+ cells, Ly6G+ neutrophils, CD3ɛ+ T cells, and rare B220+ B cells.

IBA1+ and F4/80+ cells progressively increased from 12 hours to day 7 with no statistically significant differences between the sexes (Figs. 8 and 9). CD206+ cells progressively increased from 12 hours to day 7 in males with more numerous CD206+ cells on day 2 and statistically significant more numerous CD206+ cells at day 7 (P = .0439) (Fig. 10). In females, CD206+ cells constantly remained at low levels over time. HO-1+ cells in pancreatitis areas showed a different trend compared with CD206+ cells, with a progressive decrease of HO-1+ cells from 12 hours to day 7 in males and a decrease at day 2 followed by an increase at day 7 in females, without significant differences between sexes. MHC-II+ cells increased at day 7 in both males and females without overt differences between sexes (Fig. 8).

Figure 8.

Results of quantification by digital image analysis of (a) IBA1, (b) F4/80, (c) CD206, and (d) MHC-II immunopositive cells (IBA1, F4/80, CD206, and MHC-II per pancreatic area of inflammation). IBA1+, F4/80+, CD206+, and MHC-II+ cells progressively increased from 12 hours to day 7, and CD206+ cells are higher in males at day 7 than females. Data are expressed as mean values ± SEM. *P < .05.

Figure 9.

Caerulein-induced acute pancreatitis, at (a, b) 12 hours, (c, d) day 2, and (e, f) day 7, mouse. (a, c, e) Immunohistochemistry for IBA1. (b, d, f) Immunohistochemistry for F4/80. IBA1+ and F4/80+ macrophages progressively increase from 12 hours to day 7 after the last caerulein treatment.

Figure 10.

Caerulein-induced acute pancreatitis, mouse, immunohistochemistry for CD206. At both (a, b) day 2 and (c, d) day 7, male mice have more CD206+ macrophages (a, c) compared with females (b, d).

Ly6G+ cells decreased from 12 hours to day 7 in both sexes, with slightly higher Ly6G+ cells in males than females on day 2 (Fig. 11c, f, i; Fig. 12c). A proportion of the infiltrating polymorphonucleated cells, morphologically consistent with neutrophils, were not immunolabeled by Ly6G.

Figure 11.

Caerulein-induced acute pancreatitis at (a–c) 12 hours, (d–f) day 2, and (g–i) day 7, mouse. (a, d, g) Immunohistochemistry for CD3ε. (b, e, h) Immunohistochemistry for B220. (c, f, i) Immunohistochemistry for Ly6G. T lymphocytes and B lymphocytes progressively increase from 12 hours to day 7 after the last caerulein treatment. Neutrophils progressively decrease from 12 hours to day 7 after the last caerulein treatment.

Figure 12.

Number of (a) CD3, (b) B220, and (c) Ly6g positive cells measured using digital image analysis. There is a progressive increase in number of T and B cells, and a progressive decrease of Ly6G cells from 12 hours to day 7 after last caerulein administration. M, male; F, female.

CD3ε+ and B220+ cells increased in both sexes at day 7 without differences between sexes (Fig. 11, a, b, d,e, g, h; Fig 12a, b).

Apoptosis and acinar ductular metaplasia

The cleaved caspase-3+ area abruptly increased at day 2 in both sexes (higher in males than females) and subsequently decreased at day 7, where residual apoptosis was still present in males (Figs. 3c–d, 4).

Different patterns of CK19 expression were identified at 12 hours, day 2, and day 7 in treated animals. At 12 hours, there was a diffuse cytoplasmic immunolabeling of mild intensity of pancreatic acinar cells. On day 2, CK19 was expressed by multiple, tubular-like structures resembling pancreatic ducts that were lined by plumper cells and were less intensely immunolabeled compared with normal intralobular and interlobular pancreatic ducts (Fig. 5b). On day 7, CK19 immunolabeling was present in numerous newly formed ducts indistinguishable from intralobular and interlobular pancreatic ducts, based on their morphology and immunolabeling intensity (Fig. 5d). The CK19+ area progressively increased over time in both sexes, but at day 7, it was higher in males than in females (Fig. 6b).

The percent of CK19+ area and the percent of cleaved caspase-3+ area measured via digital image analysis correlated with the ADM histological score and apoptosis histological score, respectively, with r values of 0.8925 and 0.8416 (Spearman’s correlation test) (Supplemental Fig. S2b, c).

Discussion

The study aimed to investigate sex-related differences in AP development and resolution through the application of clinical chemistry, histology, and immunohistochemistry in a murine caerulein-induced pancreatitis model.

Both clinical chemistry and histologic results were indicative of an acute pancreatic inflammatory process followed by a quick, almost complete resolution in a 7-day interval in both male and female mice. In line with previous reports, peak of pancreatic amylases at 12 hours with return to baseline levels after 48 hours, indicated a fast pancreatic damage followed by a quick resolution, without sex-related differences.^1,13,32 No significant differences were observed in glucose concentration between control and treated animals in both sexes, indicating no impact of pancreatitis on glucose metabolism. The decreased glucose concentration observed in both male and female mice at 12 hours was attributed to treatment-induced stress, potentially exacerbated by a reduced food intake in the acute phase of pancreatitis. Lipases values were extremely variability, with elevated levels even in controls, making this marker unreliable in pancreatic damage assessment.

Caerulein administration was previously demonstrated to cause a dose-related, time-dependent, self-limiting pancreatitis, histologically characterized by variable interstitial edema, interstitial neutrophilic and mononuclear cells infiltration, pancreatic acinar disarray, degeneration, apoptosis, and ADM.^32,50,53 Although differences in caerulein doses, administration protocols, and evaluated time points exist in the literature, lesions observed in our study are consistent with other published reports.^32,50,53 Pancreatitis histology scores, as well as extent of inflammation measured by digital image analysis, increased at 12 hours and decreased at days 2 and 7 in female compared with male mice, suggesting a faster onset and a faster resolution of pancreatitis in female mice. This is further supported by a lower presence of CK19+ newly formed ducts at day 7 in females, suggestive of a more efficient reparative process in this sex. Although numerous studies on the caerulein model report a complete recovery of pancreatic architecture, in our study both male and female mice still had minimal residual pancreatic lesions at day 7. Residual lesions were also reported by some authors and could be attributed to differences in caerulein administration protocols and doses, caerulein commercial product specificities, or even mouse strain used since strain-dependent susceptibility to caerulein effects has been recognized.^{14,13,23,32,47,54}

The immunohistochemical analysis on the immune cell infiltration revealed no statistically significant differences between sexes in terms of total macrophages, neutrophils, and T and B cells. The 2 pan-macrophage markers IBA1 and F4/80 showed a similar trend of positivity in treated male and female mice, with IBA1 detecting slightly more numerous cells than F4/80, possibly because of its ability to mark a wider spectrum of myeloid cells, including dendritic cells. ⁹ IBA1+ and F4/80+ macrophages were the predominant inflammatory population observed in the pancreas of both treated male and female mice. Although quantification methods of immune cells hugely vary among studies, our data are consistent with what was previously reported in literature where F4/80+ macrophages were recognized as the most numerous cells in AP.^10,24,41,51 Our analysis revealed an increasing concentration of IBA1+ and F4/80+ macrophages from 12 hours to day 7, without relevant differences between sexes. Since the time of onset and resolution of AP were different between sexes, we hypothesized that this could be reflected by a different polarization of macrophages and we, thus, performed CD206 and HO-1 immunolabeling for M2 macrophages.

CD206+ macrophages have been reported to reach a peak at day 3 after caerulein administration, in both Balb/c and C57BL/6 mice, and to correlate with a pro-resolving antiinflammatory response. ⁵¹ We observed a different trend in M2 macrophages between sexes, with constant over time low levels of CD206+ macrophages in females, which was opposite to males which showed more M2 macrophages at both days 2 and 7. These results highlight sex-dependent differences in macrophage polarization, revealing a prolonged (and likely less efficient) pro-resolving response in males. In females, AP was, instead, almost completely resolved from day 2, supporting a higher capacity of dampening inflammation. Although both are considered markers of M2 polarization, in our study HO-1 yielded different results from CD206, confirming the existence, in the inflammatory context, of different subsets of alternatively activated macrophages. CD206 is in fact a C-type lectin involved in glycoproteins recognition, while HO-1 is an antiinflammatory enzyme expressed in response to hemorrhage and ROS production, and their selective expression in different macrophagic subtypes have already been reported.^5,37,38,45 Our study revealed a peak of MHC-II+ cells only at day 7 in both males and females. This is partially consistent with other reports demonstrating no to mild increase of MHC-II+ cells in the 12-hour, 2-day interval and a peak at day 3 in a caerulein-induced model and at day 7 in a pancreatic duct ligation model.^3,44

Ly6G+ cells were, overall, present in small numbers and characterized by a progressive decrease over time in both sexes, with males showing a slightly slower reduction than females, reflecting, once more, a more protracted inflammation in male mice. Our results are consistent with other reports highlighting higher neutrophil concentrations in early phases of pancreatitis followed by a progressive decrease.^32,39 Interestingly, not all polymorphonucleated cells morphologically consistent with neutrophils were immunolabeled for Ly6G in our study. The Ly6G- neutrophils could represent an immature pool of newly recruited neutrophils, characterized by a low expression of Ly6G.^6,25

B220+ cells and CD3ε+ cells were rare in both sexes and were characterized by a progressive increase from 12 hours to day 7. Although lymphocytes are thought to play a marginal role in pancreatitis immunopathogenesis and are, thus, seldomly investigated in caerulein-induced models of pancreatitis, a study using Rag^-/-mice, deficient of both T and B lymphocytes, demonstrated the development of an atypical pancreatic inflammation characterized by increase pancreatic damage and M1 macrophage infiltration in acute phases compared with wild-type mice, demonstrating a role of B and T cells in inflammatory modulation of pancreatitis. ¹⁰ Nevertheless, our data seem to support a marginal role of these cells in AP of wild-type mice, with no relevant differences observed between sexes.

Apoptosis has been recognized as an important protective mechanism against pancreatic oxidative stress-induced necrosis in the caerulein-induced pancreatitis model.^4,17,23,27 In our study, apoptosis increased at day 2 and subsequently decreased at day 7 in both sexes as reported in other studies. ⁵⁴ However, our results revealed that males had more apoptosis than females at all time points (particularly at day 2), suggesting a sex modulation of apoptosis. A receptor-mediated inhibitory influence of estrogens on apoptosis pathways has, furthermore, been recognized in both acinar and immune (innate and adaptive) cells.^15,18,21

As pancreatic regeneration occurs through ADM, CK19, a marker of ductal differentiation, has been proven useful in the investigation of pancreatic regeneration and postinflammatory recovery. ⁴³ At 12 hours, pancreatic acinar cells of both male and female mice were characterized by a diffuse mild cytoplasmic CK19 immunolabeling. Since no CK19 immunolabeling was observed in controls at the level of acinar cells, and CK19 mRNA expression was previously detected as soon as 12 hours after caerulein administration, the diffuse mild immunolabeling observed in pancreatic acinar cells was interpreted as early CK19 expression by acinar cells. ¹³ On day 2, CK19+ ill-defined ducts interpreted as acinar cells phenotypically transitioning to pancreatic ducts were labeled, consistent with ADM. At day 7, in the residual areas of pancreatitis, CK19 immunolabeling was present in numerous, well-differentiated ducts resembling normal intralobular and interlobular ducts, and the extent of the CK19+ area was higher in males than females, potentially suggesting a less efficient pancreatic acinar regeneration in males compared with females.

Overall, the results of our study support a quicker onset, a faster resolution, and more efficient reparative response in female C57BL/6J mice compared with males. The differences observed might be associated with sex hormones, since research indicates that estrogens can influence numerous inflammatory processes, decreasing inflammatory cells recruitment and modulating macrophage populations toward an antiinflammatory phenotype.^31,46 Few studies comparing male and female mice in caerulein-induced AP exist in literature, often focusing on knockout mice strains in specific experimental settings.^13,19 A study comparing caerulein-treated male and female corticotropin-releasing factor receptor 2-deficient (Crhr2^–/–) mice showed that necrosis, polymorphonuclear cell infiltration, and zymogen depletion were significantly increased in male Crhr2^–/– mice, whereas only necrosis and acinar vacuolization were significantly increased in females compared with sex-matched wild-type controls, thus hypothesizing a sex modulating role in pancreatitis pathogenesis. ¹⁹ Hagen and colleagues demonstrated that female mice deficient in LAT1, an amino acid transporter involved in pancreatic metabolism, had delayed regeneration compared with male LAT1 deficient mice, hypothesizing a sexual influence on pancreatic amino acid metabolism. ¹³ A study using autophagy deficient Atg5flox/flox mice demonstrated a lower incidence and severity of chronic pancreatitis in females compared with males, which was attributed to a higher efficacy of steroid-dependent ROS metabolism in female mice. ⁷ Nevertheless, it is worth mentioning that a study investigating sex differences in a caerulein-induced model of chronic pancreatitis in wild-type C57BL/6 mice failed to detect differences in male and female mice regarding both disease severity and recovery. ³³

In conclusion, the results of this study demonstrated a faster onset and resolution of caerulein-induced AP in female mice compared with male mice. Male mice were characterized by higher CD206+ M2 macrophages at days 2 and 7, more extensive apoptosis at days 2 and 7, and more prominent ADM at day 7 suggesting a protracted damage (or a less controlled inflammatory process) and a delayed (or less efficient) reparative response in this sex. Data presented in this study support a sex-dependent effect on development and resolution of pancreatitis in a caerulein-induced mouse model of AP and emphasize the importance of employing animals of both sexes in experimental settings.