Pancreatic Protein Tyrosine Phosphatase 1B Deficiency Exacerbates Acute Pancreatitis in Mice

Abstract

Acute pancreatitis (AP) is a common and devastating gastrointestinal disorder that causes significant morbidity. The disease starts as local inflammation in the pancreas that may progress to systemic inflammation and complications. Protein tyrosine phosphatase 1B (PTP1B) is implicated in inflammatory signaling, but its significance in AP remains unclear. To investigate whether PTP1B may have a role in AP, we used pancreas PTP1B knockout (panc-PTP1B KO) mice and determined the effects of pancreatic PTP1B deficiency on cerulein- and arginine-induced acute pancreatitis. We report that PTP1B protein expression was increased in the early phase of AP in mice and rats. In addition, histological analyses of pancreas samples revealed enhanced features of AP in cerulein-treated panc-PTP1B KO mice compared with controls. Moreover, cerulein- and arginine-induced serum amylase and lipase were significantly higher in panc-PTP1B KO mice compared with controls. Similarly, pancreatic mRNA and serum concentrations of the inflammatory cytokines IL-1B, IL-6, and tumor necrosis factor-α were increased in panc-PTP1B KO mice compared with controls. Furthermore, panc-PTP1B KO mice exhibited enhanced cerulein- and arginine-induced NF-κB inflammatory response accompanied with increased mitogen-activated protein kinases activation and elevated endoplasmic reticulum stress. Notably, these effects were recapitulated in acinar cells treated with a pharmacological inhibitor of PTP1B. These findings reveal a novel role for pancreatic PTP1B in cerulein- and arginine-induced acute pancreatitis.

Acute pancreatitis (AP) is the most common gastrointestinal cause for hospitalization in the United States and is associated with an overall mortality of up to 5%.1, 2 The incidence of AP increases in parallel with its triggering risk factors, such as alcohol abuse, duct obstruction, and obesity.3, 4, 5 AP arises with local inflammation on activation of proteolytic pancreatic enzymes within the acinar tissue and may lead to systemic inflammatory response and complications.6, 7 Specific therapy for AP is lacking, and understanding the molecular mechanisms underlying its pathogenesis is fundamental for management and therapy of the disease at early and late stages.

Different animal models have been generated to investigate AP pathogenesis, and one of the most common is cerulein-induced pancreatitis.⁸ Administration of cerulein at high concentrations causes secretion of lipase and amylase, death of acinar cells, and infiltration of inflammatory cells into the pancreas, which are also observed in human pancreatitis.9, 10 Cerulein action involves activation of NF-κB, promotion of oxidative stress, and the release of proinflammatory cytokines.11, 12 In addition, cerulein treatment alters pancreatic tyrosine phosphorylation and modulates the activities of protein tyrosine kinases and protein tyrosine phosphatases (PTPs).13, 14

The roles of PTPs in AP remain largely unexplored, but some studies demonstrate altered PTP expression and activity in rodent models of AP. Cerulein-induced AP is associated with increased expression of the SH2 domain-containing phosphatases SHP1 and SHP2.15, 16 In addition, protein tyrosine phosphatase 1B (PTP1B) expression increases during the early stages of cerulein-induced AP and is suppressed by pharmacological inhibitors of mitogen-activated protein kinases (MAPKs).16, 17 Moreover, T-cell protein tyrosine phosphatase (TCPTP) expression increases in the early stages of the disease, and pancreatic TCPTP deficiency mitigates cerulein-induced AP in mice.¹⁸ These studies implicate PTPs in AP, but additional investigation into the contribution of these enzymes to disease pathogenesis is warranted.

PTP1B (encoded by PTPN1) is an abundant, widely expressed nonreceptor phosphatase that is localized to the endoplasmic reticulum (ER).¹⁹ PTP1B is an established regulator of epidermal growth factor receptor,²⁰ integrin,21, 22 and insulin receptor²³ signaling. Whole-body PTP1B knockout (KO) mice exhibit improved systemic insulin sensitivity, enhanced glucose tolerance, and resistance to high-fat diet–induced obesity.24, 25 Tissue-specific PTP1B deletion helped define the functions of this phosphatase in many tissues, including muscle,²⁶ liver,²⁷ and brain.28, 29 In addition, several studies implicate PTP1B in pancreas function. A compound knockout mouse model demonstrates that PTP1B global deficiency mitigates the severe diabetes caused by insulin receptor substrate 2 deletion.³⁰ In this model, PTP1B deficiency increases islet area and delays diabetes.³⁰ Moreover, PTP1B modulates ER stress signaling in the glucose-responsive MIN6 β-cell line.³¹ Furthermore, mice with pancreatic PTP1B deficiency display comparable glucose tolerance to controls, but when subjected to robust and prolonged challenges, exhibit mild glucose intolerance.³² Herein, we investigated the effects of pancreatic PTP1B deficiency on experimental acute pancreatitis using cerulein- and arginine-induced models. Alterations in inflammatory response were determined in cerulein/arginine-treated versus nontreated control and pancreas PTP1B knockout mice, and the underlying molecular mechanisms investigated.

Materials and Methods

Animal Studies

All animal studies were conducted using male mice and rats, according to federal guidelines, and approved by the Institutional Animal Care and Use Committee at University of California Davis and Ethical Committee for Animal Experimentation and Wellbeing of the University of Valencia, respectively. Generation of pancreas PTP1B knockout mice was previously described.³² Mice were maintained on a 12-hour light-dark cycle in a temperature-controlled facility, with free access to food and water. Mice were fed standard laboratory chow (Purina laboratory chow, 5001) at weaning. Acute pancreatitis was induced in 10- to 12-week-old control and pancreas PTP1B knockout mice using cerulein and arginine. For cerulein-induced pancreatitis, mice were fasted overnight, then received 12 i.p. injections of cerulein (50 μg/kg body weight) at 1-hour intervals or dimethyl sulfoxide as a vehicle control. Animals were sacrificed at 14, 24, and 48 hours after the initial injection, and blood was collected to determine serum amylase and lipase using commercial kits (Sigma, St. Louis, MO), according to the manufacturer's instructions. In addition, serum concentrations of IL-1B, TNFA, and IL-6 were determined by multiplex electrochemiluminescence (Meso Scale Discovery, Rockville, MD). Another group of mice was used for arginine-induced AP, as previously described, with slight modifications.³³ Briefly, 10- to 12-week-old mice received a single i.p. injection of 5 g/kg body weight l-arginine monohydrochloride in 0.9% sodium chloride (pH 7.0). A control group was injected i.p. with 0.9% sodium chloride as a vehicle control. Mice were sacrificed at 48 and 72 hours after arginine injection. For rat studies, Wistar rats were placed under deep anesthesia with isoflurane before treatment with 3.5% sodium taurocholate in 0.9% sodium chloride. AP was induced by a retrograde infusion of the solution. At 1 and 6 hours after induction of pancreatitis, rats were anesthetized again and pancreata were harvested and snap frozen in liquid nitrogen.

Cell Studies

Murine pancreatic acinar 266-6 cells³⁴ were purchased from ATCC (Manassas, VA) and cultured in Dulbecco's modified Eagle's medium containing 25 mmol/L glucose, 10% fetal bovine serum, 50 U/mL penicillin, and 50 μg/mL streptomycin. Pharmacological inhibition of PTP1B was achieved by treating cells with 400 μmol/L PTP1B inhibitor [3-(3,5-dibromo-4-hydroxybenzoyl)-2-ethyl-N-(4-[(2-thiazolylamino)sulfonyl]phenyl)-6-benzofuran] sulfonamide (Cayman Chemical, Ann Arbor, MI),³⁵ for 2 hours before cerulein treatment (2 μmol/L for 24 hours).³⁶ For phosphatase activity assays, 266-6 cells were lysed in NP40 buffer (10 mmol/L Tris-HCl, pH 7.4, 150 mmol/L NaCl, 1 mmol/L EDTA, 1% NP-40, 5% glycerol, and proteases inhibitors) and PTP1B was immunoprecipitated (10 μg antibody/mg of lysate) using PureProteome beads (Millipore, Billerica, MA). Phosphatase activity was determined using p-nitrophenyl phosphate (5 mmol/L), as we previously described.³⁷ Phosphatase activity was monitored by measuring the absorbance of p-nitrophenyl phosphate at 405 nm using a plate reader (GloMax-Multi Detection System; Promega Corp., Madison, WI).

Biochemical Studies

Tissues and cells were lysed using radioimmunoprecipitation assay buffer (10 mmol/L Tris-HCl, pH 7.4, 150 mmol/L NaCl, 0.1% sodium dodecyl sulfate, 1% Triton X-100, 1% sodium deoxycholate, 5 mmol/L EDTA, 1 mmol/L NaF, 1 mmol/L sodium orthovanadate, and protease inhibitors). Lysates were clarified by centrifugation, and protein concentrations were determined using bicinchoninic acid protein assay kit (Pierce Chemical, Dallas, TX). Proteins were resolved by SDS-PAGE and transferred to polyvinylidene difluoride membranes. Immunoblotting of lysates was performed with antibodies for PTP1B (Millipore), pERK1/2 (Tyr202/Thr204), pp38 (Thr180/Tyr182), p38, pJNK (Thr183/Tyr185), JNK, pIKKα/β (Ser178/180), IKKα/β, pIκBα (Ser32), IκBα, pNF-κB p65 (Ser536), NF-κB p65, and NF-κB p50 (all from Cell Signaling Technology, Beverly, MA) and cleaved caspases 8, 9, and 3, SHP1, ERK1/2, pPERK (Thr980), PERK, peIF2α (Ser51), eIF2α, sXBP1, IRE1α, and tubulin were from Santa Cruz Biotechnology (Dallas, TX). Antibodies for pIRE1α (Ser724) were purchased from Abcam (Cambridge, MA). After incubation with the appropriate secondary antibodies, proteins were visualized using Luminata Forte Western HRP substrate (Millipore). Pixel intensities of immunoreactive bands were quantified using FluorChem Q Imaging software version 9900 (Alpha Innotech, San Jose, CA). For phosphorylated proteins, data were presented as phosphorylation level normalized to protein expression; and for nonphosphorylated proteins, as protein expression normalized to tubulin.

Total RNA was extracted from pancreas using TRIzol reagent (Invitrogen Carlsbad, CA). cDNA was generated using high-capacity cDNA synthesis kit (Applied Biosystems, Foster City, CA). Il-1β, Il-6, and tumor necrosis factor (Tnf)-α mRNA were assessed by SYBR Green quantitative real-time PCR (iCycler; BioRad, Hercules, CA) using the ΔCT method with appropriate primers (Table 1) and normalized to TATA-box binding protein.

Table 1.

Primer Sequences Used to Quantitate IL1b, IL6, TNFA, and TBP Expression

Gene	Forward primer	Reverse primer
Il1b	5′-AGCTTCAGGCAGGCAGTATC-3′	5′-AAGGTCCACGGGAAAGACAC-3′
Il6	5′-ACAACCACGGCCTTCCCTACTT-3′	5′-CACGATTTCCCAGAGAACATGTG-3′
Tbp	5′-TTGGCTAGGTTTCTGCGGTC-3′	5′-GCCCTGAGCATAAGGTGGAA-3′
Tnfa	5′-GACGTGGAACTGGCAGAAGAG-3′	5′-TGCCACAAGCAGGAATGAGA-3′

Histological Analysis

A portion of the pancreas was fixed in 4% paraformaldehyde overnight, embedded in paraffin, and sections (5 μm thick) were stained with hematoxylin and eosin to observe morphological changes. Initial histological analysis was performed by an experienced histopathologist (S.G.) in a blinded manner without knowledge of treatment groups. Histological scoring of pancreatic sections was performed to grade the extent of pancreatic parenchyma edema (0, no edema; 1, interlobular edema; 2, interlobular and moderate intralobular edema; 3, interlobular and severe intralobular edema), cell vacuolation (0, none; 1, <20% acini with vacuoles; 2, <50% acini; 3, >50% acini), inflammation (0, no inflammation; 1, inflammatory cells present at intralobular; 3, inflammatory cells present at interacini), and acinar cell necrosis (0, no necrosis; 1, <10% necrosis; 2, <40% necrosis; 3, >40% necrosis), as previously described.38, 39

Statistical Analysis

Data are expressed as means ± SEM. Single data point comparisons were performed using Tukey's-Kramer honestly significant difference analyses using JMP Pro program version 11 (SAS Institute Inc., Cary, NC). Differences were considered significant at P < 0.05.

Results

PTP1B Expression Increases in Rodent Models of Acute Pancreatitis

Pancreatic PTP1B protein expression was determined in two different mouse models of AP. Pancreatitis was induced in wild-type mice by repetitive injections of cerulein or a single injection of arginine, as detailed in Materials and Methods. Mice were sacrificed at different times after injection, and PTP1B protein expression was determined at 14, 24, and 48 hours after initial cerulein injection, and 48 and 72 hours after arginine injection (Figure 1A). Immunoblots of pancreas lysates demonstrated significant increase in PTP1B protein expression on cerulein and arginine administration (Figure 1, B and C). In addition, SHP1 protein expression was increased on induction of AP, consistent with published reports.15, 17 To ensure that these observations were not limited to mouse models, pancreatic PTP1B expression was determined in taurocholate-induced pancreatitis rat model and comparable increases in expression were also observed (Figure 1D). Together, these data demonstrate increased PTP1B protein expression in different rodent models of chemical-induced AP.

Figure 1.

Increased pancreatic PTP1B protein expression in rodent models of acute pancreatitis (AP). A: Schematic depicting time course for cerulein- and arginine-induced experimental AP in mice. Total pancreas lysates of wild-type mice that were administered cerulein (Cer; B) or arginine (Arg; C) and immunoblotted for PTP1B, SHP1, and tubulin. Representative immunoblots are shown. D: Total pancreas lysates of rats that were administered taurocholate (Taur) then immunoblotted for PTP1B, SHP1, and tubulin. Representative immunoblots are shown. Bar graphs represent expression of PTP1B normalized to tubulin. Data are presented as means ± SEM (B–D). n = 8 per group (B, Cer); n = 6 per group (C, Arg, and D, Taur). ^∗∗P < 0.01 versus Cer/Arg/Taur-treated animals. A.U., arbitrary units.

Pancreatic PTP1B Deficiency Exacerbates Cerulein- and Arginine-Induced Acute Pancreatitis

Increased pancreatic PTP1B expression in rodent models of AP prompted us to investigate its potential role in disease pathogenesis. To that end, we used mice with pancreas PTP1B deficiency (panc-PTP1B KO).³² Immunoblots of total pancreas lysates demonstrated efficient deletion of PTP1B in knockout mice compared with controls, as previously reported (Figure 2A).³² In addition, selective deletion of PTP1B in the pancreas of knockout mice was previously demonstrated.³² Accordingly, this mouse model is suitable for determining PTP1B contribution to pancreatitis. We evaluated the severity of cerulein- and arginine-induced pancreatitis in control and panc-PTP1B KO mice. Histological analysis of hematoxylin and eosin–stained pancreatic sections was performed to grade pathological changes, including edema, cell vacuolization, inflammation, and necrosis, as detailed in Materials and Methods. In control and panc-PTP1B KO mice, cerulein administration caused an increase in edema, cell vacuolization, inflammation, and necrosis at 24 and 48 hours after the first injection, with highest grades at 24 hours (Figure 2B and Table 2). Scores for edema and necrosis in panc-PTP1B KO were significantly higher than control mice at 24 hours. In addition, 48 hours after the first injection, panc-PTP1B KO exhibited higher scores of vacuolization, inflammation, and necrosis compared with control mice (Table 2). Consistent with histological findings, serum amylase and lipase, which are commonly used markers for AP diagnosis,40, 41 were significantly different between control and panc-PTP1B KO mice (Figure 2C). Under basal conditions, serum amylase and lipase were comparable between control and knockout mice. Cerulein and arginine administration led to significant increase in amylase and lipase but to a higher level in panc-PTP1B KO compared with control mice. Similar findings were observed in two additional independent cohorts of mice (n = 6 to 8 per group, data not shown). During acute pancreatitis, activation of NF-κB enhances the release of proinflammatory cytokines, such as IL-1B and IL-6 and TNFA. Pancreatic mRNA of Il-1β, Il-6, and Tnf-α increased after cerulein and arginine administration, and levels were significantly higher in panc-PTP1B KO than control mice (Figure 2D). Similarly, serum levels of IL-1B, IL-6, and TNFA were increased in control mice after cerulein administration but were significantly higher in panc-PTP1B KO than control mice (Figure 2E). Together, these data indicate that pancreatic PTP1B deficiency exacerbates cerulein- and arginine-induced AP in mice.

Figure 2.

Pancreatic PTP1B deficiency exacerbates cerulein (Cer)- and arginine (Arg)-induced pancreatic injury. A: Total pancreas lysates from control and panc-PTP1B knockout (KO) mice were immunoblotted for PTP1B and tubulin. Representative immunoblots are shown. B: Hematoxylin and eosin–stained sections of pancreas with cerulein-induced pancreatitis. In nontreated groups, control and panc-PTP1B KO mice appear normal with no pathological changes; necrosis is increased in knockout compared with focal area of necrosis in control mice at 24 hours after injection; vacuolization (arrows), inflammation (arrowheads), and necrosis (asterisks) are increased in knockout compared with control mice 48 hours after injection. C: Acute pancreatitis was induced by i.p. injections of cerulein and arginine and pancreatic injury determined by measuring serum amylase and lipase. D: Il-1β, Il-6, and Tnf-α mRNA levels were determined from at least six mice per group. E: Serum levels of IL-1B and IL-6, and TNFA, in control and knockout mice with cerulein-induced AP at the indicated times. Data are presented as means ± SEM (C–E). n = at least 5 mice per group (C–E). ^∗P < 0.05, ^∗∗P < 0.01 versus Cer/Arg-treated mice; ^†P < 0.05, ^††P < 0.01 versus panc-PTP1B KO mice. Scale bar = 50 μm (B). A.U., arbitrary units., KU, kilo units.

Table 2.

Histological Scoring of Pancreatic Tissues of Control and Panc-PTP1B KO Mice

Variable	Edema	Vacuolation	Inflammation	Necrosis
Control
Cer (−) (n = 11)	0.00 ± 0.00	0.27 ± 0.19	0.00 ± 0.00	0.00 ± 0.00
Cer (24 hours) (n = 13)	1.61 ± 0.36∗∗	0.69 ± 0.13∗∗	1.610 ± 0.36∗∗	0.10 ± 0.00
Cer (48 hours) (n = 11)	1.45 ± 0.24∗∗	0.54 ± 0.15	1.00 ± 0.30∗∗	0.18 ± 0.00
KO
Cer (−) (n = 10)	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00	0.00 ± 0.00
Cer (24 hours) (n = 12)	2.58 ± 0.19∗∗†	0.91 ± 0.08∗∗	2.33 ± 0.25∗∗	1.25 ± 0.13∗∗†
Cer (48 hours) (n = 11)	1.72 ± 0.23∗∗	1.00 ± 0.00∗∗††	2.36 ± 0.27∗∗††	1.18 ± 0.22∗∗††

Hematoxylin and eosin–stained pancreas sections were scored to grade the extent of acinar edema, cell vacuolation, inflammation, and acinar cell necrosis. Data are presented as means ± SEM from two independent cohorts.

^∗∗P < 0.01 indicates significant difference between mice without and with cerulein administration.

^†P < 0.05, ^††P < 0.01 indicate significant difference between control and panc-PTP1B KO mice.

Cer, cerulein; KO, knockout.

Pancreatic PTP1B Deficiency Increases Cerulein- and Arginine-Induced NF-κB Inflammation

The mechanisms of cerulein and arginine action involve activation of the nuclear transcription factor NF-κB, leading to increased cytokine expression in the pancreas and inflammation.⁴² NF-κB is activated early in AP in leukocytes and pancreatic acinar cells.⁴³ Proinflammatory cytokines, such as TNFA, activate IκB kinase complex (IKK) to phosphorylate inhibitor of NF-κB (IκB).⁴⁴ IκB phosphorylation triggers its ubiquitination and subsequent degradation, leading to the dissociation of NF-κB dimers (p65 and p50) and translocation to the nucleus for transcription activation.⁴⁵ In addition, several reports implicate PTP1B in modulating the inflammatory response.46, 47, 48, 49, 50 Accordingly, we evaluated NF-κB signaling in panc-PTP1B KO and control mice in cerulein- and arginine-induced pancreatitis. Cerulein and arginine induced IKKα, I_kBα, and NF-κB p65 phosphorylation and increased NF-κB p50 expression (Figure 3). Notably, cerulein- and arginine-induced IKKα, I_kBα, and NF-κB p65 phosphorylation and NF-κB p50 expression were increased in panc-PTP1B KO compared with control mice. These findings are in keeping with the increased pancreatic proinflammatory cytokines in cerulein- and arginine-treated knockout mice.

Figure 3.

Regulation of cerulein- and arginine-induced NF-κB inflammatory response by pancreatic PTP1B. Total pancreas lysates from control and panc-PTP1B knockout (KO) mice treated with cerulein (Cer; A) and arginine (Arg; B) were immunoblotted for pIKKα, pIκBα, pNF-κB p65, and their respective unphosphorylated proteins, NF-κB p50, PTP1B, and tubulin. Representative immunoblots are shown. C: Bar graphs represent normalized data for pIKKα/IKKα, pIκBα/IκBα, pNF-κB p65/NF-κB p65, and NF-κB p50/tubulin. Data are presented as means ± SEM (C). n = 6 to 8 animals per treatment (C). ^∗P < 0.05, ^∗∗P < 0.01 versus Cer/Arg-treated mice; ^†P < 0.05, ^††P < 0.01 versus panc-PTP1B KO mice. A.U., arbitrary units.

Pancreatic PTP1B Deficiency Regulates Cerulein- and Arginine-Induced MAPK Signaling

To further investigate the molecular basis for AP in panc-PTP1B KO mice, we determined activation of MAPKs. MAPKs, including ERK1/2, p38, and JNK1/2, are rapidly and transiently induced during experimental AP in rodents.⁵¹ This is believed to be a component of the cellular stress response in the onset of inflammation in the pancreas. In addition, JNK and ERK have been implicated in the up-regulation of PTP1B expression on cerulein treatment.¹⁶ Under basal conditions, MAPK activation was comparable between control and panc-PTP1B KO mice (Figure 4). However, cerulein and arginine administration led to increased phosphorylation of ERK1/2, p38, and JNK in control mice, and that was significantly higher in panc-PTP1B KO mice. These findings establish increased MAPK phosphorylation in pancreata of cerulein- and arginine-treated panc-PTP1B KO mice and are consistent with the elevated inflammatory response in these mice.

Figure 4.

Pancreatic PTP1B deficiency regulates cerulein- and arginine-induced mitogen-activated protein kinase signaling. Total pancreas lysates from control and panc-PTP1B knockout (KO) mice treated with cerulein (Cer; A) and arginine (Arg; B) were immunoblotted for pERK1/2, pp38, pJNK1/2, and their respective unphosphorylated proteins and tubulin. Representative immunoblots are shown. C: Bar graphs represent normalized data for pERK/ERK, pp38/p38, and pJNK/JNK. Data are presented as means ± SEM (C). n = 6 to 8 animals per treatment (C). ^∗P < 0.05, ^∗∗P < 0.01 versus Cer/Arg-treated mice; ^††P < 0.01 versus panc-PTP1B KO mice. A.U., arbitrary units.

Pancreatic PTP1B Deficiency Regulates Cerulein- and Arginine-Induced ER Stress

The ER is highly responsive to the energy status of the cell and plays an important role in the folding of newly synthesized proteins. Unfolded protein response is used by cells to counter the deleterious effects of ER stress and is triggered by transmembrane sensors, such as PKR-like ER-regulated kinase (PERK), that detect unfolded proteins in the ER and convey information through their cytosolic domain.⁵² More important, activation of ER stress is associated with AP⁵³ and treatment with ER chaperones mitigates the disease in animal models.⁵⁴ In addition, we previously demonstrated that PTP1B knockdown in MIN6 β-cells enhanced PERK and inositol requiring enzyme 1α (IRE1α) signaling.³¹ We evaluated activation of ER transmembrane proteins PERK and IRE1α and their downstream targets α-subunit of eukaryotic translation initiation factor 2 (eIF2α) and X-box binding protein 1 (XBP1), respectively. Consistent with previous reports,54, 55 cerulein and arginine administration induced ER stress in the pancreas (Figure 5). Notably, pancreatic PTP1B deficiency potentiated cerulein- and arginine-induced ER stress signaling, as assessed by increased PERK, eIF2α, and IRE1α phosphorylation and sXBP1 expression. The unfolded protein response is deployed as a compensatory mechanism to restore homeostasis, but if it fails then apoptosis commences. We assessed cerulein- and arginine-induced expression of initiator and effector caspases in control and knockout mice. Both cerulein and arginine induced procaspase 8, 9, and 3 cleavage in controls but that was significantly enhanced in panc-PTP1B KO mice (Figure 5). Collectively, these findings establish increased ER stress and cleaved caspase expression on pancreatic PTP1B deficiency during the early phase of cerulein/arginine-induced AP.

Figure 5.

Regulation of cerulein- and arginine-induced endoplasmic reticulum (ER) stress by PTP1B. Total pancreas lysates from control and panc-PTP1B knockout (KO) mice treated with cerulein (Cer; A) and arginine (Arg; B) were immunoblotted for pPERK, peIF2α, pIRE1, and their respective unphosphorylated proteins, cleaved caspases 8, 9, and 3 and tubulin. C: Representative immunoblots are shown. Bar graphs represent normalized data for pPERK/PERK, peIF2α/eIF2α, pIRE1/IRE1, caspase 8, 9, 3, and sXBP1/tubulin. Data are means ± SEM (C). n = 6 to 8 animals per treatment (C). ^∗P < 0.05, ^∗∗P < 0.01 indicate significant difference between untreated and Cer/Arg-treated mice; ^†P < 0.05, ^††P < 0.01 versus panc-PTP1B KO mice. A.U., arbitrary units.

PTP1B Pharmacological Inhibition Exacerbates Cerulein-Induced Inflammation in Acinar Cells

As a complementary approach, we investigated the effects of PTP1B pharmacological inhibition in murine pancreatic acinar 266-6 cell line. Cells were treated with PTP1B pharmacological inhibitor (PTP1Bi) without and with cerulein stimulation, as detailed in Materials and Methods. Consistent with findings from total pancreas lysates, PTP1B protein expression was increased in acinar cells on cerulein treatment (Figure 6A). Similarly, SHP1 and TCPTP expression was also increased in cerulein-treated acinar cells. In addition, PTP1B pharmacological inhibition did not alter PTP1B protein expression but significantly decreased its activity in cells without cerulein (1.00 ± 0.07 without PTP1Bi versus 0.66 ± 0.01 with PTP1Bi) and with cerulein treatment (3.71 ± 0.25 without PTP1Bi versus 1.93 ± 0.34 with PTP1Bi). Cerulein treatment increased IKKα, IκBα, and NF-κB p65 phosphorylation and elevated NF-κB p50 expression in acinar cells (Figure 6B). In addition, PTP1B inhibitor–treated cells exhibited increased NF-κB inflammatory response compared with nontreated cells on cerulein stimulation. Moreover, PTP1B pharmacological inhibition exacerbated cerulein-induced increase in MAPK activation and ER stress in acinar cells (Figure 6, C and D). These findings demonstrate that PTP1B pharmacological inhibition in acinar cells recapitulates the effects of its pancreatic deficiency in cerulein-induced AP.

Figure 6.

Pharmacological inhibition of PTP1B in acinar cells exacerbates cerulein (Cer)-induced inflammation. A: Murine pancreatic acinar 266-6 cells were pretreated with PTP1B inhibitor for 2 hours, then exposed to cerulein for 24 hours. Cell lysates were immunoblotted for PTP1B, SHP1, TCPTP, and tubulin. Representative immunoblots are shown. Cell lysates from control and cerulein-treated cells, with and without PTP1B inhibition were immunoblotted for pIKKα, pIκBα, pNF-κB p65 (B), pp38, pJNK1/2 (C), pPERK, peIF2α, caspase 3 (D), and respective unphosphorylated proteins and tubulin.

Discussion

The development of acute pancreatitis involves a complex cascade of events that occurs in acinar cells, but the underlying cellular and molecular mechanisms are not fully understood. In this study, we investigated the role of pancreatic PTP1B in experimental AP using rodent models. We established increased pancreatic PTP1B expression during the early stages of AP in mice and rats. In addition, pancreatic PTP1B deficiency exacerbated cerulein- and arginine-induced pancreatitis. This was associated with increased cerulein- and arginine-induced NF-κB inflammatory response and elevated MAPK signaling and ER stress in knockout mice. These findings define a previously unknown role for PTP1B in chemical-induced acute pancreatitis.

Alterations in gene and protein expression during the early stages of AP play a significant role in disease progression and severity.⁵⁶ In this regard, pancreatic PTP1B expression increased during experimental AP using cerulein and arginine in mice and taurocholate in rats, establishing that it is not specific to a particular chemical or rodent model. Similarly, PTP1B expression and activity increased in cerulein-treated acinar cells. These findings are consistent with a previous report of elevated PTP1B expression during the early stages of AP.¹⁷ In most cases, increased PTP1B protein expression correlates with increased enzymatic activity.57, 58 Indeed, in cerulein-treated acinar cells, increased PTP1B expression was mirrored by elevated enzymatic activity. However, several post-translational modifications dynamically regulate PTP1B activity⁵⁹ and alterations of PTP1B activity during AP in rodents and humans remain to be determined. Increased PTP1B expression is often, but not always, associated with states of elevated inflammation, such as obesity and diabetes.60, 61 PTP1B is induced by inflammation in vivo, and TNF-α treatment leads to increase in PTP1B mRNA and protein in insulin- and leptin-responsive tissues in mice, whereas TNF-α deficiency blocks diet-induced increase in expression.⁶² Of note, PTP1B expression during the early stages of AP is comparable to that of other PTPs, such as SHP1, SHP2,15, 16 and TCPTP.¹⁸ However, pancreatic TCPTP deficiency mitigates cerulein-induced AP in mice¹⁸ compared with the effects of PTP1B deficiency observed herein. Thus, additional investigations into the regulation of PTPs and their contribution to AP pathogenesis are warranted.

Genetic and pharmacological approaches helped define the role of PTP1B in AP. Pancreatic PTP1B deficiency exacerbated chemical-induced AP, as evidenced by pancreas histology, increased serum amylase and lipase, and enhanced serum and pancreatic IL-1B, TNFA, and IL-6.

A limitation of this study is that panc-PTP1B KO mice exhibit PTP1B deletion in the endocrine and exocrine pancreas,³² so the precise contribution of acinar cells cannot be delineated. However, the effects of PTP1B deficiency in vivo were recapitulated in inhibitor-treated acinar cells, suggesting that they were likely cell autonomous, although we cannot rule out effects of nonacinar cells. Moreover, we cannot rule out that PTP1B pharmacological inhibition did not alter activity of other phosphatase(s), although activity of the closely related TCPTP was not significantly altered in inhibitor-treated acinar cells (data not shown). Additional studies are needed to determine the effects of acinar cell–specific PTP1B deficiency and PTP1B pharmacological inhibition in vivo on the progression and development of AP.

The molecular mechanisms underlying pancreatic PTP1B actions in AP remain to be determined but may encompass direct and/or indirect modulation of inflammatory, MAPK, and ER stress responses. First, pancreatic PTP1B deficient mice exhibited increased proinflammatory cytokines that play a significant role in the development and severity of the disease.⁶³ TNF-α exacerbates acinar cell injury, IL-1β regulates the development of AP, and IL-6 is a major mediator of the acute-phase response. Consistent with increased proinflammatory cytokines, knockout mice exhibited enhanced cerulein- and arginine-induced NF-κB signaling. NF-κB inflammatory response is activated early in AP and plays an important role in disease pathogenesis.64, 65 Previous studies implicate PTP1B in the control of inflammatory processes but are somewhat inconsistent. Knockdown of PTP1B increases production of TNFA and IL-6 in TLR-triggered macrophages.⁴⁶ In addition, PTP1B deficient macrophages display increased inflammatory activity in vitro and in vivo.⁴⁷ Moreover, PTP1B deficiency exacerbates inflammation and accelerates leukocyte trafficking in vivo.⁴⁸ Furthermore, PTP1B deficiency accentuates the effects of proinflammatory stimuli in both rodent and human macrophages.⁴⁹ On the other hand, myeloid cell–specific PTP1B deficiency protects against high-fat diet– and lipopolysaccharide-induced inflammation.⁵⁰ The reason(s) for these differences is not apparent and could depend on the experimental systems and/or regulation of inflammatory signaling by PTP1B in a stimulus/tissue-dependent manner. At any rate, given the current findings, it is reasonable to stipulate that the effects of pancreatic PTP1B deficiency during experimental AP could be mediated, at least partly, through exacerbation of the inflammatory response. Second, pancreatic PTP1B deficiency affected MAPK signaling that is already implicated in pancreatitis. In particular, pancreatic PTP1B deletion was associated with increased cerulein- and arginine-induced MAPK activation, indicative of elevated stress, and is consistent with previous studies implicating MAPKs in AP.⁶⁶ Of note, studies using MAPK pharmacological inhibitors suggest that JNK and ERK regulate cerulein-induced PTP1B expression.¹⁶ The mechanism by which PTP1B deficiency potentiates MAPK signaling remains unknown, but can be indirect and related to increased inflammation. Last, ER stress is implicated in the pathophysiology of pancreatitis,53, 67 and attenuation of ER stress by chemical chaperones mitigates the disease in animal models.⁵⁴ Increased PERK and IRE1α activation in PTP1B knockout mice and inhibitor-treated acinar cells is consistent with previous findings of enhanced PERK and IRE1α signaling in MIN6 β-cells.³¹ Conceivably, pancreatic PTP1B can regulate ER stress through direct and/or indirect mechanisms. One scenario involves direct interaction and dephosphorylation of component(s) in ER stress signaling, such as PERK. Of note, PERK is a PTP1B substrate in fibroblasts⁶⁸ and adipocytes,⁶⁹ and PERK Tyr615 is a mediator of its interaction with PTP1B. In support of this scenario, we observed increased cerulein-induced PERK phosphorylation in knockout mice and inhibitor-treated acinar cells. However, we cannot exclude the possibility that PTP1B modulates ER stress indirectly. Indeed, PTP1B activates c-Src by dephosphorylating Tyr527,⁷⁰ and Src has been implicated in ER stress response.⁷¹

The salutary metabolic effects of PTP1B deficiency and pharmacological inhibition on insulin sensitivity and energy balance establish this enzyme as a potential therapeutic target for obesity and type 2 diabetes. However, the current findings define a role for pancreatic PTP1B in experimental AP and suggest that PTP1B likely acts in a cell type– and tissue-dependent manner to regulate signaling and the inflammatory response.