Myeloid‐specific dopamine D₂ receptor signalling controls inflammation in acute pancreatitis via inhibiting M1 macrophage

Abstract

Background and Purpose

Macrophage infiltration and activation is a critical step during acute pancreatitis (AP). We have shown that pancreas‐specific D₂ receptor signalling protects against AP severity. As it is unclear to what extent myeloid‐specific D₂ receptor mediates AP, we investigated the role of myeloid‐specific D₂ receptor signalling in AP.

Experimental Approach

Using wild‐type and LysM^+/creD₂ ^fl/fl mice, AP was induced by l‐arginine, caerulein and LPS. Murine bone marrow‐derived macrophages and human peripheral blood mononuclear cells (PBMCs) were isolated, cultured and then induced to M1 phenotype. AP severity was assessed by measurements of serum amylase and lipase and histological grading. Macrophage phenotype was assessed by flow cytometry and qRT‐PCR. NADPH oxidase‐induced oxidative stress and NF‐κB and NLRP3 inflammasome signalling pathways were also evaluated.

Key Results

We found that dopaminergic system was activated and dopamine reduced inflammatory cytokine expression in M1‐polarized macrophages from human PBMCs. Dopaminergic synthesis was also activated, but D₂ receptor expression was down‐regulated in M1‐polarized macrophages from murine bone marrows. During AP, myeloid‐specific D₂ receptor deletion worsened pancreatic injury, systematic inflammation and promoted macrophages to M1 phenotype. Furthermore, M1 macrophages from LysM^+/creD₂ ^fl/fl mice exhibited increased NADPH oxidase‐induced oxidative stress and enhanced NF‐κB and NLRP3 inflammasome activation. D₂ receptor activation inhibited M1 macrophage polarization, oxidative stress‐induced NF‐κB and NLRP3 inflammasome activation.

Conclusion and Implications

Our data for the first time showed that myeloid‐specific D₂ receptor signalling controls pancreatic injury and systemic inflammation via inhibiting M1 macrophage, suggesting D₂ receptor activation might serve as therapeutic target for AP.

Abbreviations

AP

acute pancreatitis

BMDMs

bone marrow‐derived macrophages

Cae

caerulein

DA

dopamine

DBH

dopamine β‐hydroxylase

DDC

dopa decarboxylase

NC

normal control

Quin

quinpirole

WT

wild‐type

What is already known

• Pancreas‐specific D₂ receptor signalling protects against the severity of acute pancreatitis (AP).

• Macrophages were activated in AP and the degree of macrophage activation can determine AP severity.

What this study adds

• Myeloid‐specific D₂ signalling controls pancreatic injury and systemic inflammation via inhibiting M1 macrophage during AP.

• Myeloid‐specific D₂ signalling regulates NADPH oxidase‐mediated NF‐κB and NLRP3 inflammasome activation in M1 macrophages.

What is the clinical significance

• Therapeutic approaches that activate D₂ signalling may represent a novel strategy for AP.

1. INTRODUCTION

Acute pancreatitis (AP) is an inflammatory disorder of the pancreas, which is associated with substantial morbidity and mortality, and is the leading cause of admission to hospital for gastrointestinal disorders in the United States and many other countries (Forsmark, Vege, & Wilcox, 2016; Lankisch, Apte, & Banks, 2015; Lee & Papachristou, 2019). Premature intracellular protease activation leads to acinar cell death, which initiates innate immune responses, resulting in migration of monocytes and neutrophils into the inflamed pancreas (Sendler et al., 2018; Xue, Sharma, & Habtezion, 2014). These infiltrating cells produce cytokines and inflammatory mediators, leading to the amplification of the local and systemic inflammation (Zheng, Xue, Jaffee, & Habtezion, 2013). The extent of monocyte infiltration and macrophage activation determines the severity of acute pancreatitis (Saeki et al., 2012; Wu et al., 2018). Thus, the strategies mediating monocyte differentiation and/or macrophage activation could be a potential tool for treating acute pancreatitis, which currently has no specific therapy.

Accumulating evidence suggests that the neurotransmitter dopamine plays an important regulatory role in innate immunity. Several studies suggest that dopamine modulates the functions of immune cells in an autocrine/paracrine manner through D₁ and D₂ receptors, which are present in almost all leukocytes (McKenna et al., 2002; Sarkar, Basu, Chakroborty, Dasgupta, & Basu, 2010). Either dopamine agonists or cell‐specific D₂ activation attenuates inflammation in models of sepsis or chronic neuroinflammation (Shao et al., 2013; Torres‐Rosas et al., 2014). Limited studies indicate a role of dopamine receptor signalling in macrophage polarization and activation (Bernton, Meltzer, & Holaday, 1988; Qin et al., 2015). A recent study showed that D₁ receptor signalling negatively regulates NLRP3 inflammasome activation in macrophage through cyclic AMP, which binds to NLRP3 and promotes its ubiquitination and degradation (Yan et al., 2015). NLRP3 inflammasome and NF‐κB activation have shown to play a critical role in mediating macrophage cytokine production and secretion (Awad et al., 2017; Liu et al., 2015; Wu et al., 2013). ROS production also has a role in maintaining M1 macrophage activation (Tan et al., 2016).

We previously found that systemic administration of dopamine markedly reduced pancreatic inflammation and also showed that the protective effects observed with dopamine were completely reversed by a D₂ antagonist, but not by a D₁ antagonist, indicating that dopamine specifically via D₂ signalling modulates the severity of acute pancreatitis (Han et al., 2017). Moreover, we found that systemic administration of D₂ agonist protected against local and systemic injury in two models of experimental acute pancreatitis, while pancreas‐specific D₂ deletion worsened pancreatic injury (Han et al., 2017). However, it is unclear to what extent innate immune cell‐specific D₂ signalling modulates pancreatitis severity. Since macrophages play a dominant role in mediating the severity of acute pancreatitis (Wu et al., 2018), we therefore, in this study, aimed to (a) examine the role of D₂ signalling, specifically derived from macrophages in acute pancreatitis and (b) determine the effects of a D₂ agonist on macrophage polarization and functions with an attempt to further evaluate its therapeutic potential for treating acute pancreatitis.

We firstly showed that the dopamine system was present in M1 macrophages and that it was functionally activated in both human and mouse. Using myeloid‐specific D₂ knockout mice, we found that innate immune cell‐specific D₂ signalling worsened in experimental acute pancreatitis, likely via inhibiting pro‐inflammatory M1 macrophage polarization. Furthermore, myeloid‐specific D₂ signalling affected NADPH oxidase‐mediated oxidative stress and promoted NF‐κB and NLRP3 inflammasome activation. Using isolated bone marrow‐derived macrophages (BMDMs) we showed that D₂ agonist inhibited M1 macrophage polarization, mitigated NADPH oxidase‐mediated oxidative stress and NF‐κB, and NLRP3 inflammasome activation. Finally, we found that D₂ agonist protected against two models of experimental acute pancreatitis via preventing pro‐inflammatory M1 macrophage polarization in the pancreas. Our novel findings provide preclinical evidence that systemic D₂ receptor activation is a potential therapeutic strategy for treating acute pancreatitis by targeting both pancreatic acinar injury and innate immune responses. These findings also open up a new therapeutic strategy for other severe inflammatory diseases.

2. METHODS

2.1. Mouse strains

Wild‐type (WT) C57BL6/J mice (RRID:IMSR_JAX:000664, 6–8 weeks, 20–22 g, male) were purchased from Shanghai SLAC Laboratory Animal Co Ltd (Shanghai, China). D₂ ^fl/fl mice RRID:MGI:5313431 were kindly gifted by Professor Jiawei Zhou from Institute of Neuroscience, State Key Laboratory of Neuroscience, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China. LysM^cre mice were purchased from Jackson Laboratories (Bar Harbor, USA). To generate myeloid‐specific D₂ receptor deletion mice (LysM^+/creD₂ ^fl/fl), D₂ ^fl/fl mice were crossed with LysM^cre/cre miceRRID:SCR_000729. All mice were housed under pathogen‐free conditions in individually ventilated cages with wood shavings as bedding material (four to six mice per cage), 12‐hr dark/light cycle at 22°C, and free access to water and standard rodent diet. All mice (20–22 g, male) were marked with an earmark and allocated into groups in a completely randomized manner using a randomized table (n = 6 per group). The pathologists but not the experimental operators were blinded to the experiment groups. All the experiments involving animals were conducted under the principle for replacement, refinement and reduction (the 3Rs) and according to the legislation on the protection of animals and were approved by the Animal Ethics Committee of Shanghai Jiao Tong University School of Medicine (SYXK 2013–0050, Shanghai, China). Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny, Browne, Cuthill, Emerson, & Altman, 2010; Mcgrath & Lilley, 2015) and with the recommendations made by the British Journal of Pharmacology.

2.2. Induction of experimental acute pancreatitis and treatments

Acute pancreatitis was induced with 10 hourly injections of caerulein (ceruletide;

100 μg·kg⁻¹, i.p.) and LPS (5 mg·kg⁻¹, i.p. administered immediately after the last injection of caerulein) or two hourly intraperitoneal injections of l‐arginine (4 g·kg⁻¹, 8%, pH = 7.0) as previously described (Dawra et al., 2007; Gironella et al., 2007; Hu et al., 2011). Controls received equal volume of normal saline (NS) injection. D₂ agonist quinpirole (10 mg·kg⁻¹) was injected intraperitoneally 0.5 hr before the first injection of caerulein or l‐arginine, and in l‐arginine‐induced acute pancreatitis model mice continued to receive quinpirole daily until tissue harvesting. In caerulein and LPS‐induced acute pancreatitis model, mice were killed at 12 hr after the first injection of caerulein. In l‐arginine‐induced pancreatitis, mice were killed at 72 hr after the first injection of l‐arginine. Serum, lung and pancreas were collected.

2.3. Histology and immunohistochemistry

Fresh specimens of murine pancreas were fixed in 4% paraformaldehyde, embedded in paraffin, and 4‐μm sections were processed for H&E or immunohistochemistry staining by standard procedures. Histological scores were quantified on haematoxylin–eosin (H&E) staining tissue sections by two experienced pathologists (Han et al., 2017; Hu et al., 2011; Van Laethem et al., 1995).

2.4. Murine pancreatic leukocyte isolation, murine bone marrow‐derived macrophages preparation, human peripheral blood cell isolation and in vitro cultures

Pancreatic leukocytes were isolated from fresh obtained pancreatic tissue by collagenase digestion as described previously (Hawkins, Gala, & Dunbar, 1996; Xue, Nguyen, & Habtezion, 2012). Briefly, pancreas was cut into pieces with scissors and then digested by 2 mg·ml⁻¹ collagenase type IV from Sigma‐Aldrich Chemical (St. Louis, MO, USA) in a shaking incubator at 37°C for 15 min. Tissue suspension was filtered through a 100 μm nylon mesh, and cell pellet was dipped in red blood cell lysing buffer for 5 min. The cells were centrifuged and collected for cell surface or intracellular staining.

Bone marrow‐derived macrophages was prepared from C57BL6/J mice, as described previously (Anthony, Kobayashi, Wermeling, & Ravetch, 2011; Ying, Cheruku, Bazer, Safe, & Zhou, 2013). Briefly, both femurs and tibias were used to collect bone marrow cells. After removal of red blood cells, bone marrow cells were re‐suspended in DMEM/Ham F‐12 medium with 10% FBS and 20 ng·ml⁻¹ rmM‐CSF. On Day 3, non‐adherent cells were removed and medium was replaced. CD11b⁺F4/80⁺ cells were confirmed by flow cytometry on Day 6. LPS (100 ng·ml⁻¹) and IFN‐γ (10 ng·ml⁻¹) were used to induce M1 macrophages, and IL4 (50 ng·ml⁻¹) was used to induce M2 macrophages on Day 6, with or without quinpirole (2.5, 5 and 10 μM) treatment at the same time for 24 hr.

Human peripheral blood from healthy volunteers were collected at Shanghai General Hospital with approval of the Local Ethics Committee (2017KY170) and written informed consent was received from participants prior to inclusion in the study. The entire study design and procedures were in accordance with the Declaration of Helsinki. Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll–Hypaque (GE Healthcare; Stockholm, Sweden) density gradient centrifugation; cells were cultured in medium with 10% FBS and 50 ng·ml⁻¹ rhM‐CSF for 5 days and then stimulated by 100 ng·ml⁻¹ LPS and 20 ng·ml⁻¹ IFN‐γ for 24 hr. Quinpirole (5 μM) was added at the same time with LPS and IFN‐γ.

2.5. Flow cytometry

All antibodies used for flow cytometry of murine cells were purchased from Thermo Fisher Scientific (Waltham, MA, USA). For surface staining, murine pancreatic leukocytes or isolated bone marrow‐derived macrophages were stained with FITC‐CD45 (11‐0451‐81, 1:100), PE/Cy7‐CD11b (25‐0112‐81, 1:100), and PE‐F4/80 (12‐4801, 1:100).

For intracellular TNFα staining, cells were incubated with or without LPS (100 ng·ml⁻¹) in the presence of brefeldin A (3 μg·ml⁻¹, Thermo Fisher Scientific, Waltham, MA, USA) at 37°C for 3 hr before surface staining. The cells were then fixed and permeabilized using the Intracellular Fixation & Permeabilization Buffer Set (Thermo Fisher Scientific, Waltham, MA, USA) following the instructions. Finally, the pancreatic leukocytes or isolated bone marrow‐derived macrophages were stained with the following intracellular antibodies: APC‐TNFα (17‐7321‐81, 1:100) and isotype control (17‐4301‐82, 1:200), APC‐iNOS (17‐5920, 1:100) and isotype control (17‐4321‐81, 1:200), and APC‐CD206 (17‐2061, 1:100) and isotype control (17‐4031‐82, 1:200).

Flow cytometry was performed on FACSCanto II (BD Biosciences, Franklin, NJ, USA) and analysed using FlowJo software (RRID:SCR_008520, Tree Star Inc., OR, USA). Cell sorting was performed on BD influx (BD Biosciences, Franklin, NJ, USA).

2.6. Measurement of intracellular ROS

The production of intracellular ROS in isolated bone marrow‐derived macrophages was determined by 2,7‐dichlorofluorescein diacetate (DCFH‐DA) assay. Briefly, isolated bone marrow‐derived macrophages were washed with PBS followed by incubation with 10 μM DCFH‐DA (Sigma‐Aldrich Chemical, MO, USA) for 30 min at 37°C. Then isolated bone marrow‐derived macrophages were washed twice with PBS and divided into two parts. After detached by gentle scraping, one part of bone marrow‐derived macrophage was imaged for cellular fluorescence by FACSCanto II (BD Biosciences, NJ, USA; Rosenblat, Volkova, & Aviram, 2010); after fixed by 4% paraformaldehyde and counterstained with 4′,6‐diamidino‐2‐phenylindole (DAPI; 1:1,000, Jackson, USA), another part of isolated bone marrow‐derived macrophages was imaged by laser scanning confocal microscope (LSCM; Zeiss, Germany; Jiang et al., 2015).

2.7. Western blotting

Total protein of isolated bone marrow‐derived macrophages was extracted as described previously (Han et al., 2017). Membrane and Cytosol Protein Extraction Kit from Beyotime Biotechnology (Shanghai, China) was used to extract cell membrane and cytosol protein following the instructions. Protein (40 μg per lane) was separated by 10% SDS‐PAGE and then electrotransferred to NC membranes (Millipore, Mass, USA). Membranes were incubated with primary antibodies against polyclonal tyrosine hydroxylase (TH, RRID:AB_2303165, 1:800), dopa decarboxylase (DDC, RRID:AB_2088975, 1:400), dopamine β‐hydroxylase (DBH, RRID:AB_2089347, 1:400), monoamine oxidase (MAO‐A, RRID:AB_2137260, 1:400), D₁ (RRID:AB_445306, 1:800) and D₂ receptors (RRID:AB_2094976, 1:400), monoclonal NF‐κBp65 (RRID:AB_10859369, 1:800), p‐NF‐κBp65 (RRID:AB_331284, 1:800), NLRP3 (RRID:AB_2722591, 1:800), caspase1 (RRID:AB_781816, 1:400), IL1β (RRID:AB_2715503, 1:800), p47^phox (RRID:AB_627987, 1:400) and β‐actin (1:800) overnight at 4°C. Membranes were then probed with goat anti‐rabbit or goat anti‐mouse IR‐Dye 700 or 800 cw labelled secondary antisera for 1 hr at 37°C. Finally, membranes were scanned and analysed using Odyssey IR scanner (LI‐COR, USA). To control for unwanted sources of variation, the relative expression of target proteins was normalized to β‐actin and the phosphorylation level of NF‐κB was compared with total NF‐κB. The mean values of the control group were set to 1; the values of other groups were normalized to control group values, presented as fold mean of the controls. The Immuno‐related procedures used comply with the recommendations made by the British Journal of Pharmacology (Alexander et al., 2018).

2.8. Quantitative reverse transcription PCR (qRT‐PCR)

mRNA transcripts in isolated bone marrow‐derived macrophages or pancreatic macrophages or human peripheral blood mononuclear cells were analysed by qRT‐PCR as described previously (Han et al., 2017). Primers are shown in Tables S1 and S2. Fold changes for each mRNA were calculated using the comparative CT (2^−ΔΔCT) method. To control for unwanted sources of variation, the mRNA levels were normalized to β‐actin mRNA. The mean values of the control group were set to 1; the values of other groups were normalized to control group values, presented as fold mean of the controls. Each target gene was analysed in triplicate in each experiment.

2.9. Immunofluorescence staining

Fresh specimens of pancreas were fixed in 4% neutral paraformaldehyde for 2 hr, dehydrated in 30% sucrose overnight, embedded in OCT, and sectioned at 4 μm thickness. Sections of pancreas were then incubated overnight at 4°C with a polyclonal antibody against TNFα (60291‐1‐Ig, Proteintech Biotechnology, Wuhan, China, 1:200) and F4/80 (RRID:AB_385952, 1:200). After being rinsed in PBS for three times, sections were incubated with a mixture of IgG/Cy3 donkey anti‐mouse antibody (RRID:AB_2315777, 1:500, Jackson, USA) and DayLight488 donkey anti‐rat antibody (RRID:AB_2340684, 1:500, Jackson, USA) for 45 min in the dark. Slides of isolated bone marrow‐derived macrophages were fixed by 4% paraformaldehyde and incubated overnight at 4°C with a polyclonal antibody against TH (1:200), DDC (1:200), DBH (1:200), MAO (1:200), D₁ (1:200), or D₂ (1:200). After being rinsed in PBS for three times, slides of isolated bone marrow‐derived macrophages were then incubated with DayLight488 (RRID:AB_2340684, 1:500, Jackson, USA) or IgG/Cy3 donkey anti‐rabbit antibody (RRID:AB_2340667, 1:500, Jackson, USA) or IgG/Cy3 donkey anti‐goat antibody (RRID:AB_2340411, 1:500, Jackson, USA). Sections of pancreas or slides of isolated bone marrow‐derived macrophages were then incubated with 1 μg·ml⁻¹ 4′,6‐diamidino‐2‐phenylindole (DAPI; 1:1,000, Jackson, USA) for 10 min after being rinsed in PBS. Results of immunofluorescence staining were analysed by laser scanning confocal microscope (LSCM; Zeiss, Germany).

2.10. ELISA

Blood samples were centrifuged at 400× g for 20 min at 4°C. The concentrations of amylase, lipase, TNFα, IL6, IL10, IL1β, and IL18 in serum or culture supernatant of isolated bone marrow‐derived macrophages were measured by ELISA according to the manufacturer's protocols (Westang bio‐tech Co, LTD, Shanghai, China).

2.11. Electrophoretic mobility shift assays

Electrophoretic mobility shift assay (EMSA) was conducted to detect the binding activity of NF‐κB as described before (Han et al., 2017).

2.12. Statistical analysis

Statistical analysis was undertaken for studies with six independent values. All data are presented as mean ± SEM. The distribution of data was assessed by Kolmogorov–Smirnov test at first. If data followed a Gaussian distribution, parametric tests (Student's t test for two groups or one‐way ANOVA for three or more groups) were carried out; for ANOVA, Bonferroni's post hoc test was performed for data with F at P < .05 and no significant variance inhomogeneity. If data were not normally distributed, non‐parametric tests (Mann–Whitney test for two groups or Kruskal–Wallis test with Dunn's posttest for three or more groups) were used by GraphPad Prism version 7.00 for Windows (RRID:SCR_002798, GraphPad Software, La Jolla, CA, USA, www.graphpad.com). A P value < .05 was considered statistically significant. The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis et al., 2018). The results were obtained from at least three independent experiments, with six independent values per group to ensure their reliability.

2.13. Materials

Caerulein (Cae; C9026), l‐arginine monohydrochloride (l‐Arg; A5131), LPS (L2880), and quinpirole (Quin; Q102) were purchased from Sigma‐Aldrich Chemical (St. Louis, MO, USA). rmM‐CSF (315‐02), IFNγ (315‐05), IL4 (214‐14) and rhM‐CSF (300‐25) were purchased from PEprotech (Rocky Hill, NJ, USA). Antibodies against DDC (sc‐99203), DBH (sc‐15318), MAO (sc‐20156), D₂ (sc‐7523), caspase1 (sc‐56036), and p47^phox (sc‐17844) were from Santa Cruz Biotechnology (Dallas, TX, USA). Antibodies against TH (2792), NF‐κBp65 (8242), phosphorylated NF‐κBp65 (3033), NLRP3 (15101) and IL1β(3A6) (12242) were from Cell Signaling Technology (Danvers, MA, USA). Antibodies against D₁ (RRID:AB_445306, ab20066) and Ly6G (RRID:AB_470492, ab25377) were from Abcam (Cambridge, MA, USA). Antibodies against TNFα (60291‐1‐Ig) and CD206 (18704‐1‐AP) were from Proteintech Biotechnology (Wuhan, China). Antibody against β‐actin (AF0003) was purchased from Beyotime Biotechnology (Shanghai, China). Antibody against F4/80 (GTX26640) was from GeneTex (San Antonio, TX, USA). Nuclear and Cytoplasmic Extraction Reagents was from Pierce (Rockford, IL, USA). The biotin‐labelled probe containing the NF‐κB consensus site was purchased from Beyotime Biotechnology (Shanghai, China). The Light Shift chemiluminescent EMSA kit was from Pierce (Rockford, IL, USA).

2.14. Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in https://www.guidetopharmacology.org/, the common portal for data from the UPHAR/BPS Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archive in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander et al., 2019).

3. RESULTS

3.1. Dopaminergic system is functionally activated and the DA receptor is present in M1 macrophages

It has been reported that dopamine inhibits pro‐inflammatory signals in LPS‐primed isolated bone marrow‐derived macrophages (Yan et al., 2015). However, the presence of dopaminergic system and dopamine receptors and the effects of dopamine in human and murine monocytes/macrophages remain undetermined. Therefore, we isolated peripheral blood mononuclear cells from healthy volunteers, cultured and stimulated the cells towards M1 phenotype by LPS (100 ng·ml⁻¹) and IFN‐γ (20 ng·ml⁻¹) with or without dopamine (500 μM and 750 μM) treatment, and examined the levels of enzymes involved in the synthesis of dopamine, (tyrosine hydroxylase, TH, and DOPA decarboxylase, DDC,) or catabolic enzymes (dopamine ß‐hydroxylase, DBH, and monoamine oxidase A, MAO‐A,) and dopamine receptors. We found that the enzymes TH, DDC, DBH, MAO‐B and dopamine D₂ receptor were up‐regulated, while MAO‐A and D₁ remained unchanged (Figure 1a). Dopamine also reduced mRNA expression of inflammatory cytokines (Tnfα, Il1β and Il6) induced by LPS and IFN‐γ (Figure 1b). Secondly, we isolated and cultured isolated bone marrow‐derived macrophages from mice and induced isolated bone marrow‐derived macrophages to M1 phenotype by LPS (100 ng·ml⁻¹) and IFN‐γ (10 ng·ml⁻¹; Murray et al., 2014). Similarly, we found that in M1 macrophages, the mRNA and protein levels of dopamine synthesis enzymes, TH and DDC, were significantly increased, whereas the levels of dopamine catabolic enzyme, MAO‐A was decreased while DBH was unchanged (Figure 1c,d). Interestingly, we found that D₁ and D₂ receptors were decreased in M1 macrophages (Figure 1e,f). Immunofluorescent staining for dopamine synthetase, catabolic enzymes and dopamine receptors also showed similar results (Figure S1). Collectively, those data suggested that dopaminergic system and receptor signalling are functionally activated in human and mouse M1 macrophages and dopamine may be a potent inhibitor of inflammation.

FIGURE 1.

Change of the dopaminergic system in macrophages in human and mouse. (a, b) Peripheral blood cells (PBMCs) from healthy volunteers were isolated by Ficoll–Hypaque density gradient centrifugation; cells were cultured in medium with 10% FBS and 50 ng·ml⁻¹ rhM‐CSF for 5 days and then stimulated by 100 ng·ml⁻¹ LPS and 20 ng·ml⁻¹ IFN‐γ for 24 hr; 500 and 750 μM dopamine was added at the same time with LPS and IFN‐γ. (a) The mRNA levels of Th, Ddc, Dbh, Mao‐a, Mao‐b, D₁ and D₂ by qRT‐PCR in human PBMCs from six different individuals. (b) The mRNA levels of Tnfα, Il1β and Il6 by qRT‐PCR in human PBMCs from six different individuals. (c–f) Bone marrow‐derived macrophages (BMDMs) were cultured with 20 ng·ml⁻¹ M‐CSF for 6 days. M1 macrophages were induced by 100 ng·ml⁻¹ LPS and 10 ng·ml⁻¹ IFN‐γ for 24 hr on Day 6. (c) qRT‐PCR of mRNA levels of Th, Ddc, Dbh, and Mao‐A in M1‐phenotype isolated BMDMs. (d) Immunoblot analysis of TH, DDC, DBH, and MAO proteins of M1‐phenotype BMDMs. (e) qRT‐PCR of mRNA levels of D₁ and D₂ in M1‐phenotype BMDMs. (f) Immunoblot analysis of D₁ and D₂ proteins of M1‐phenotype BMDMs. BMDMs were isolated from n = 6 mice; data were expressed as mean ± SEM. *P < .05 versus M0; ^# P < .05 versus LPS + IFN‐γ

3.2. Myeloid‐specific D₂ receptor deletion aggravates experimental acute pancreatitis by promoting M1 macrophage polarization

Since we have previously shown that pancreas‐specific D₂ deletion (Pdx1^+/creD₂ ^fl/fl) aggravates acute pancreatitis via pancreatic acinar cell inflammation and injury (Han et al., 2017). In addition we have shown that pancreatic acinar cell injury and the innate immune responses are independent but synergetic events that mediate the development and progression of acute pancreatitis (Lankisch et al., 2015; Lee & Papachristou, 2019). To further understand the contribution of innate immune cell‐specific D₂ signalling during acute pancreatitis, we generated myeloid‐specific D₂ knockout mice by crossing D₂ ^fl/fl mice with LysM^+/cre mice. RT‐PCR data confirm the successful deletion of D₂ in isolated bone marrow‐derived macrophages (Figure S2A,B). Histological analysis showed no significant change in the development of organs between wild‐type (WT) and LysM^+/creD₂ ^fl/fl mice (Figure S2C). We then induced acute pancreatitis in those mice by intraperitoneal injections of l‐arginine or caerulein and LPS. Histological analysis and pathological scores showed that the pancreas in LysM^+/creD₂ ^fl/fl mice was similar to that in the WT group, which was not treated with l‐arginine or caerulein and LPS (Figure 2a,c). During acute pancreatitis, LysM^+/creD₂ ^fl/fl mice exhibited more severe pancreatic injury as assessed by histological scores and more marked elevation of serum amylase and lipase compared to age‐ and sex‐matched WT mice (Figure 2a–d). In l‐arginine‐ and caerulein and LPS‐induced acute pancreatitis, serum levels of TNFα, IL1β and IL18 were similarly higher in LysM^+/creD₂ ^fl/fl mice (Figure 2e). Moreover, we found that during both acute pancreatitis models, myeloid‐specific D₂ deletion also aggravated pancreatitis‐associated lung injury as assessed by lung histology and had more infiltrated F4/80⁺ macrophage and Ly6G⁺ neutrophil (Figure 2f,g). Collectively, our data showed that myeloid‐specific D₂ deletion worsens pancreatic local and systemic inflammation/injury during two models of experimental acute pancreatitis, suggesting innate immune cell‐specific D₂ signalling may equally contribute to mediate acute pancreatitis severity.

FIGURE 2.

Myeloid‐specific D₂ deletion worsens pancreatic local and systemic injury during two models of AP. Two models of acute pancreatitis (AP) were induced in WT C57BL6/J and LysM^+/creD₂ ^fl/fl on C57BL6/J background mice in vivo. (a) Representative pictures of H&E‐stained pancreatic sections (200×) and histological scores in caerulein and LPS‐induced AP model. (b) Change in serum activity of amylase and lipase in caerulein and LPS‐induced AP model. (c) Representative images of H&E‐stained pancreatic sections (200×) and histological scores in l‐arginine‐induced AP model. (d) Change in serum activity of amylase and lipase in l‐arginine‐induced AP model. (e) ELISA of TNFα, IL1β and IL18 in serum of caerulein and LPS‐induced AP model (up) and l‐arginine‐induced AP model (down). (f) Representative pictures of H&E‐stained lung sections in caerulein and LPS‐induced AP model (up) and l‐arginine‐induced AP model (down). (g) Representative images of macrophage marker F4/80 (left) and neutrophil marker Ly6G (right) immunohistochemical staining in the lung (200×) of caerulein and LPS‐induced AP model. n = 6 per group; data were expressed as mean ± SEM. Cae, caerulein; l‐Arg, l‐arginine. Scale bar = 100 μm. *P < .05 WT‐AP versus WT‐Ctrl; ⁺ P < .05 LysM^+/creD₂ ^fl/fl‐Ctrl versus LysM^+/creD₂ ^fl/fl‐AP ; ^ƫ P < .05 LysM^+/creD₂ ^fl/fl‐AP versus WT‐ AP

Since we had previously found that the recruitment of inflammatory monocytes (CD11b⁺Ly6‐C^hi) into the pancreas was much greater than neutrophils, reflecting the dominant role of infiltrated monocytes in mediating disease severity (Wu et al., 2018) and that these recruited inflammatory monocytes are further differentiated/activated into macrophages we focused our study on macrophages. Firstly, we found that F4/80⁺ macrophages started to infiltrate into the pancreas as early as 6 hr, peaked at 12 hr and gradually decreased at 24 to 48 hr. Further flow cytometry analysis with pancreatic leukocytes revealed that at 12 hr, the number of TNFα⁺ or iNOS⁺ macrophages (M1 phenotype) markedly increased in the pancreas, while CD206⁺ macrophages (M2 phenotype) remains unchanged; at 24 hr, TNFα⁺ or iNOS⁺ macrophages started to decline while CD206⁺ macrophages reached the peak; at 48 hr, both M1 and M2 macrophages nearly returned to the baseline level (Figure S3).

Next, we examined whether myeloid‐specific D₂ receptor signalling mediates the phenotypes of infiltrated macrophages in the pancreas. Since we observed M1 macrophage peaked at 12 hr in caerulein and LPS‐induced pancreatitis, we isolated pancreatic leukocytes from WT and LysM^+/creD₂ ^fl/fl mice with caerulein and LPS‐induced pancreatitis at 12 hr after the induction and performed flow cytometry analysis. Compared to WT mice, LysM^+/creD₂ ^fl/fl mice revealed significantly increased number of TNFα⁺ or iNOS⁺ macrophages. Interestingly, LysM^+/creD₂ ^fl/fl mice exhibited a decrease in the number of CD206⁺ macrophages at 12 hr (Figure 3a). Similar findings were observed in l‐arginine‐induced pancreatitis at 72 hr after the induction (Figure 3b). These data suggest that myeloid‐specific D₂ receptor signalling may be associated with increased pro‐inflammatory M1 macrophage polarization in the pancreas during acute pancreatitis. To further elucidate the role of D₂ signalling on macrophage polarization, we isolated and cultured bone marrow‐derived macrophages from WT and LysM^+/creD₂ ^fl/fl mice, stimulated towards M1 phenotype by LPS (100 ng·ml⁻¹) and IFNγ (10 ng·ml⁻¹) and treated with quinpirole (5 μM) in vitro, and examined M1 macrophage‐specific genes (inos, Tnfα, Il1β and Il6). Compared to isolated bone marrow‐derived macrophages from WT mice, we found that the expression of inos and Il1β but not Tnfα and Il6 was significantly higher in LysM^+/creD₂ ^fl/fl mice (Figure 4a). Flow cytometry analysis also showed that after stimulation with LPS and IFN‐γ, isolated bone marrow‐derived macrophages from LysM^+/creD₂ ^fl/fl mice had more TNFα⁺ macrophages, but iNOS⁺ macrophages and CD206⁺ macrophages remain unchanged between WT and LysM^+/creD₂ ^fl/fl mice (Figure 4b). Moreover, quinpirole did not lead to down‐regulation of inos and Il1β mRNA levels or decrease of TNFα⁺ macrophage number in M1 macrophages from LysM^+/creD₂ ^fl/fl mice (Figure 4a,b), further suggesting that the anti‐inflammatory effect of quinpirole is regulated by D₂ receptor. Taken together, our in vivo and in vitro data indicate that myeloid‐specific D₂ receptor signalling alleviates experimental acute pancreatitis by inhibiting M1 macrophage polarization.

FIGURE 3.

Myeloid‐specific D₂ deletion mediates M1 macrophage polarization in vivo. Two models of acute pancreatitis (AP) were induced in WT C57BL6/J and LysM^+/creD₂ ^fl/fl on C57BL6/J background mice in vivo. (a) FCM of TNFα, iNOS, and CD206 expression in CD45⁺CD11b⁺F4/80⁺ macrophages from pancreas of caerulein and LPS‐induced AP model. (b) FCM of TNFα, iNOS, and CD206 expression in CD45⁺CD11b⁺F4/80⁺ macrophages from pancreas of l‐arginine‐induced AP model. n = 6 per group, data were expressed as mean ± SEM. Cae, caerulein; l‐Arg, l‐arginine. ^ƫ P < .05 LysM^+/creD₂ ^fl/fl‐AP versus WT‐AP

FIGURE 4.

Myeloid‐specific D₂ deletion mediates M1 macrophage polarization in vitro. BMDMs were isolated and cultured from WT C57BL6/J and LysM^+/creD₂ ^fl/fl mice. M1 macrophages were induced by 100 ng·ml⁻¹ LPS and 10 ng·ml⁻¹ IFN‐γ for 24 hr and were treated with 5‐μM quinpirole at the same time. (a) qRT‐PCR of mRNA levels of inos, Tnfα, Il1β and Il6 in M1‐phenotype BMDMs. BMDMs were isolated from six animals per group, data were expressed as mean ± SEM. (b) FCM of TNFα, iNOS, and CD206 expression in M1‐phenotype BMDMs. BMDMs were isolated from six animals per group, data were expressed as mean ± SEM. *P < .05 WT‐M1 versus WT‐M0; ^# P < .05 WT‐M1 treated with quinpirole (Quin) versus WT‐M1; ⁺ P < .05 LysM^+/creD₂ ^fl/fl‐M1 versus LysM^+/creD₂ ^fl/fl‐M0; ^ƫ P < .05 LysM^+/creD₂ ^fl/fl ‐M1 versus WT‐M1

3.3. Myeloid‐specific D₂ receptor deletion affects NADPH oxidase‐mediated NF‐κB and inflammasome activation in M1 macrophages

Next, we sought to examine whether myeloid‐specific D₂ receptor signalling modulates M1 macrophage functions. Several studies have documented that nicotinamide adenine dinucleotide phosphate (NADPH) oxidase is the main source of ROS that can influence macrophage functions (Choi, Aid, Kim, Jackson, & Bosetti, 2012). NADPH oxidase consists the membrane subunits p22^phox and gp91^phox and the cytoplasmic subunits p40^phox, p47^phox and p67^phox (Bedard & Krause, 2007), and translocation of cytoplasmic subunits p47^phox to the cytoplasmic side of plasma membrane is a surrogate marker of NADPH oxidase activation (Jiang et al., 2015). Using isolated bone marrow‐derived macrophages from WT and LysM^+/creD₂ ^fl/fl mice we found that, compared to WT mice, M1 macrophages from LysM^+/creD₂ ^fl/fl mice exhibited more p47^phox plasma membrane translocation (Figure 5a), indicating an increase in NADPH oxidase activation. As expected, ROS production in M1 macrophages from LysM^+/creD₂ ^fl/fl mice was significantly increased (Figure 5b,c). Accumulative evidence suggests that NADPH oxidase‐mediated ROS can lead to the activation of downstream signalling pathways, such as NF‐κB or inflammasome activation (Kono et al., 2000; Moon et al., 2016). Therefore, we examined the effects of myeloid‐specific D₂ receptor deletion on downstream NF‐kB and NLRP3 inflammasome activation. We found that compared to WT, M1 macrophages from LysM^+/creD₂ ^fl/fl mice exhibited an increase in NF‐κBp65 nucleus translocation as assessed by co‐localization of p65 with DAPI (Figure 5d). The protein levels of NLRP3, cleaved caspase1 and mature IL1β were similarly higher in M1 macrophages from LysM^+/creD₂ ^fl/fl mice (Figure 5e). However, quinpirole cannot inhibit p47^phox plasma membrane translocation, ROS production, NF‐κBp65 nucleus translocation, protein levels of NLRP3, cleaved caspase1, and mature IL1β in M1 macrophages from LysM^+/creD₂ ^fl/fl mice (Figure 5a–e). Consistently, the levels of IL1β and IL18 induced by LPS and IFN‐γ in the supernatant were significantly higher in M1 macrophages from LysM^+/creD_2f ^l/fl mice and quinpirole cannot inhibit IL1β and IL18 secretion in M1 macrophages from LysM^+/creD₂ ^fl/fl mice (Figure 5f). Taken together, these data suggest that myeloid‐specific D₂ signalling promotes NADPH oxidase‐mediated NF‐κB and NLRP3 inflammasome activation in M1 macrophages, and quinpirole cannot reverse the functions of M1 macrophages from LysM^+/creD₂ ^fl/fl mice.

FIGURE 5.

Myeloid‐specific D₂ receptor deletion affects NADPH oxidase‐mediated NF‐κB and inflammasome activation in M1 macrophages. Bone marrow‐derived macrophages (BMDMs) were isolated and cultured from WT and LysM^+/creD₂ ^fl/fl mice. M1 macrophages were induced by 100 ng·ml⁻¹ LPS and 10 ng·ml⁻¹ IFN‐γ and were treated with 5 μM quinpirole at the same time. (a) Immunofluorescent images of p47^phox in M1‐phenotype BMDMs. Scale bar = 50 μm. (b) FCM of intracellular ROS was determined by 2,7‐dichlorofluorescein diacetate (DCFH‐DA) staining in M1‐phenotype BMDMs. Immunofluorescent images of (c) ROS stained with DCFH‐DA and (d) NF‐κBp65 in M1‐phenotype BMDMs. Scale bar = 50 μm. (e) Immunoblot analysis of NLRP3, caspase1 and IL1β levels in M1‐phenotype BMDMs. (f) ELISA of IL1β and IL18 in supernatants of M1‐phenotype BMDMs. n = 6 per group; data were expressed as mean ± SEM. *P < .05 WT‐M1 versus WT‐M0; ^# P < .05 WT‐M1 treated with quinpirole (Quin) versus WT‐M1; ⁺ P < .05 LysM^+/creD₂ ^fl/fl‐M1 versus LysM^+/creD₂ ^fl/fl‐M0; ^ƫ P < .05 LysM^+/creD₂ ^fl/fl ‐M1 versus WT‐M1; ^nsno significance

3.4. D₂ receptor activation by quinpirole prevents M1 macrophage polarization

Since systemic administration of a D₂ agonist protects against two models of experimental acute pancreatitis and pancreas‐specific D₂ receptor signals control inflammation (Han et al., 2017). Our current data further suggest that myeloid‐specific D₂ signalling contributes to the reduction in the severity of acute pancreatitis and this is likely by inhibiting M1 macrophage. To further evaluate the therapeutic potential of D₂ agonists, we next examined whether D₂ agonists can mediate macrophage polarization in vitro. We isolated bone marrow‐derived macrophages, cultured and stimulated with LPS and IFN‐γ towards M1 phenotype and treated with or without the D₂ agonist quinpirole and analysed the phenotypes of macrophages by qRT‐PCR or flow analysis. We found that D₂ activation by quinpirole concentration‐dependently reduced the expression of all the M1 macrophage‐specific markers, including inos, Tnfα, Il1β, and Il6 (Figure 6a). Interestingly, treatment of D₂ agonist alone also resulted in significant decrease in the expression of inos, Tnfα and Il1β, but not Il6 (Figure S4A). Flow cytometry analysis revealed that the D₂ agonist concentration‐dependently reduced the number of TNFα⁺ or iNOS⁺ macrophages and lead to an increase in CD206⁺ macrophages (Figure 6b). We also found that D₂ activation alone did not influence the differentiation of isolated bone marrow‐derived macrophages (Figure S4B). Collectively, our data suggest that D₂ activation by quinpirole inhibits M1 macrophage polarization in vitro.

FIGURE 6.

D₂ receptor activation inhibits M1 macrophage polarization in vitro. M1 macrophages were induced by 100 ng·ml⁻¹ LPS and 10 ng·ml⁻¹ IFN‐γ and were treated with quinpirole (2.5, 5, and 10 μM) at the same time. (a) mRNA levels of inos, Tnfα, Il1β, and Il6 by qRT‐PCR in LPS and IFN‐γ‐stimulated BMDMs. (b) FCM of TNFα, iNOS and CD206 expression in LPS and IFN‐γ‐stimulated BMDMs. n = 6 per group, data were expressed as mean ± SEM. Scale bar = 50 μm. *P < .05 versus NC; ^# P < .05 versus LPS and IFN‐γ

3.5. D₂ receptor activation by quinpirole mitigates NADPH oxidase‐mediated NF‐κB and NLRP3 inflammasome activation in M1 macrophages

Next, we sought to examine whether D₂ activation modulates M1 macrophage functions. To evaluate the effect of quinpirole on p47^phox translocation and ROS generation, isolated bone marrow‐derived macrophages were treated with 2.5, 5 and 10 μM quinpirole at the time of stimulation with LPS (100 ng·ml⁻¹) and IFN‐γ (10 ng·ml⁻¹). We found that quinpirole concentration‐dependently reduced the plasma membrane p47^phox and increased cytoplasmic p47^phox (Figure 7a,b), indicating that D₂ agonist can mitigate against the effects of NADPH oxidase activation. As expected, ROS production was similarly reduced by quinpirole (Figure 7c,d). Compared to un‐stimulation, M1 macrophages (stimulated with LPS and IFNγ) had significantly increased NF‐κBp65 nucleus translocation as assessed by p65 co‐localization with DAPI. D₂ activation by quinpirole markedly ameliorated p65 phosphorylation assessed by western blot and NF‐κBp65 activation as assessed by EMSA (Figures 7e and S5A, B). Secretion of inflammatory cytokines, including TNFα and IL6 were similarly reduced, while IL10 unchanged (Figure S5C). Moreover, we found that D₂ activation by quinpirole also lead to a reduction of the mRNA and protein levels of NLRP3, cleaved caspase1 and mature IL1β, as well as the secretion of IL1β and IL18 (Figure 7f,g). Collectively, these data indicate that D₂ activation by quinpirole can mitigate NADPH oxidase‐mediated NF‐κB and NLRP3 inflammasome activation in M1 macrophages.

FIGURE 7.

D₂ receptor activation mitigates NADPH oxidase‐mediated NF‐κB and NLRP3 inflammasome activation in M1 macrophages in vitro. M1 macrophages were induced by 100 ng·ml⁻¹ LPS and 10 ng·ml⁻¹ IFN‐γ and were treated with quinpirole (Quin; 2.5, 5, and 10 μM) at the same time. (a) Immunofluorescent staining of p47^phox in M1‐phenotype bone marrow‐derived macrophages (BMDMs). Scale bar = 50 μm. (b) Immunoblot analysis of p47^phox levels in the membrane and cytoplasm of M1‐phenotype BMDMs. (c) FCM of intracellular ROS was determined by 2,7‐dichlorofluorescein diacetate (DCFH‐DA) staining in M1‐phenotype BMDMs. Immunofluorescent images of (d) ROS stained with DCFH‐DA and (e) NF‐κBp65 in M1‐phenotype BMDMs. Scale bar = 50 μm. (f) Immunoblot analysis of NLRP3, caspase1, and IL1β levels in M1‐phenotype BMDMs. (g) ELISA of IL1β and IL18 in supernatants of M1‐phenotype BMDMs. n = 6 per group; data were expressed as mean ± SEM. *P < .05 versus NC; ^# P < .05 versus LPS and IFN‐γ

3.6. D₂ receptor activation by quinpirole prevents pro‐inflammatory M1 macrophage polarization in the pancreas during experimental acute pancreatitis

Since we found that D₂ activation inhibits M1 macrophage polarization and its functions in vitro, we sought to examine whether D₂ activation inhibits pro‐inflammatory M1 macrophage infiltration in vivo. Quinpirole at 10 mg·kg⁻¹ was intraperitoneally administered before the induction of l‐arginine or caerulein and LPS‐induced acute pancreatitis models. Consistent with our previous report, we found that administration of quinpirole protected against two models of experimental acute pancreatitis (Han et al., 2017). We then sorted CD45⁺CD11b⁺F4/80⁺ macrophages from isolated pancreatic leukocytes in the control, acute pancreatitis and acute pancreatitis treated with quinpirole for gene expression analysis. We found that D₂ activation by quinpirole significantly down‐regulated M1 macrophage‐specific markers, including inos, Tnfα, Il1β and Il6 in both l‐arginine‐ and caerulein and LPS‐induced acute pancreatitis models (Figure 8a,b). Further flow cytometry analysis revealed that quinpirole lead to a decrease in the number of TNFα⁺ and iNOS⁺ macrophages (M1 phenotype) and an increase in the number of CD206⁺ macrophages (M2 phenotype; Figure 8c,d). Immunofluorescence staining confirmed that quinpirole lead to a reduction of TNFα⁺ and F4/80⁺ double‐positive cells (M1 macrophages; Figure S6). Taken together, our data suggest that D₂ activation ameliorates acute pancreatitis severity in vivo likely via preventing pro‐inflammatory M1 macrophage polarization in the pancreas.

FIGURE 8.

D₂ receptor activation prevents pro‐inflammatory M1 macrophage polarization in the pancreas in vivo. Pancreatic leukocytes from control and acute pancreatitis (AP) mice were isolated and sorted for CD11b⁺F4/80⁺ macrophages. mRNA levels of inos, Tnfα, Il1β, and Il6 by qRT‐PCR in CD45⁺CD11b⁺F4/80⁺ cells from pancreatic leukocytes during (a) caerulein and LPS‐induced AP model and (b) l‐arginine‐induced AP model. FCM of TNFα, iNOS, and CD206 expression in CD45⁺CD11b⁺F4/80⁺ cells from pancreatic leukocytes during (c) caerulein and LPS‐induced AP model and (d) l‐arginine‐induced AP model. n = 6 per group, data were expressed as mean ± SEM. *P < .05 versus Ctrl; ^# P < .05 versus AP

4. DISCUSSION

Many studies have demonstrated that peripheral dopamine plays an important role in immune systems. Acting on its receptors, dopamine or selective agonists have been reported to modulate the activation, proliferation and cytokine production in immune cells (Beck et al., 2004; Torres‐Rosas et al., 2014). Consistently, we found that dopaminergic system was functionally activated in macrophages and dopamine receptors were expressed in differentiated M1 macrophages in mouse and human. More importantly, we found that dopamine, as a potent inhibitor of inflammation, markedly reduced expression of pro‐inflammatory cytokines in M1‐polarized macrophages from human peripheral blood mononuclear cells, which emphasized the important clinical relevance of the current study.

Our previous study showed that dopamine markedly reduced pancreatic inflammation specifically via D₂ signalling and pancreas‐specific D₂ deletion worsened pancreatic injury, whereas systemic administration of D₂ agonist protected against local and systemic injury in two models of experimental acute pancreatitis (Han et al., 2017). It is well known that pancreatic acinar cell injury and innate immune responses are two important events that mediate the development and progression of acute pancreatitis (Lankisch et al., 2015; Lee & Papachristou, 2019). In the current complementary study, we showed that myeloid‐specific D₂ signalling equally contributed to a reduction in the severity of acute pancreatitis via inhibiting M1 macrophage polarization and functions. Moreover, we found that D₂ agonist can inhibit M1 macrophage polarization and functions in vitro and prevent pro‐inflammatory M1 macrophage polarization in the pancreas during experimental acute pancreatitis. The major implications for our novel findings are that dopamine or D₂ agonist could be a highly promising therapeutic tool for treating acute pancreatitis since they target pancreatic acinar injury and innate immune responses, which are two independent but synergetic pathophysiological events in the pathogenesis of acute pancreatitis (Sendler et al., 2018; Xue et al., 2014).

Accumulative evidence suggested that dopamine or dopamine receptor signalling controls inflammation and immune cell function (Beck et al., 2004; Torres‐Rosas et al., 2014). Macrophages are capable of synthesizing dopamine (Brown et al., 2003; Nguyen et al., 2011) and express dopamine receptors (McKenna et al., 2002), suggesting that dopamine or dopamine receptor signalling is functionally important for macrophage function. Our study systemically assessed the role of D₂ signalling in macrophages. We found that myeloid‐specific D₂ deletion promoted macrophages towards M1, while D₂ activation inhibits this. Consistently, M1 macrophage transformation in human macrophages was reported to be dependent on cAMP, a specific second messenger of D₁ signalling in dengue virus infection (Bystrom et al., 2008). Macrophages display highly plasticity and the phenotype of polarized M1‐M2 macrophages can, to some extent, be reversed in vitro and in vivo. Moreover, dynamic changes in macrophage activation are associated with the status of diseases, specifically M1 cells are implicated in initiating and sustaining inflammation while M2 cells are involved in resolution and tissue repair after acute injury (Sica & Mantovani, 2012). Limited data from our current study (Figure 8) suggested that D₂ agonist can promote macrophage towards M2 phenotype at the early stage of acute pancreatitis, which may in turn promote inflammation resolution. Future studies are required to systemically assess the effects of D₂ signalling on M2 macrophage polarization in the context of pancreatitis.

Several studies have documented that NADPH oxidase‐mediated oxidative stress can regulate macrophage polarization and function (Choi et al., 2012; West et al., 2011). Previous studies showed that in the kidney, D₂ receptors can inhibit NADPH oxidase activity and ROS production and maintain normal blood pressure (Yang et al., 2014). Consistently, we found that myeloid‐specific D₂ deletion enhanced NADPH oxidase activity and ROS production. NADPH oxidase‐mediated ROS can activate downstream NF‐κB or inflammasome activation, resulting in production of pro‐inflammatory cytokines (Kono et al., 2000; Moon et al., 2016). We found that myeloid‐specific D₂ deletion also promoted NAPDH oxidase‐mediated NF‐κB and inflammasome activation in M1 macrophages. Furthermore, we showed that at the mechanistic level, D₂ agonist inhibited NADPH oxidase activity, ROS‐mediated NF‐κB and inflammasome activation in M1 macrophages in vitro and prevents M1 macrophage polarization in the pancreas in vivo. Lastly, in mice bone marrow‐derived macrophages studies, we found that low concentration of dopamine mainly activated D₁ receptors, while a slightly high concentration of dopamine mainly activated D₂ (Figure S7A,B). High concentration of dopamine can significantly inhibit the inflammasome activation and can be blocked by D₂ antagonist (Eticlopride) but not D₁ antagonist (SCH‐23390; Figure S7C,D). These data indicated that dopamine (500 μM) exhibited an anti‐inflammatory role in isolated bone marrow‐derived macrophages via activation of D₂ receptors. A recent study also suggests that mitochondrial ROS can contribute to macrophage function via a subset of Toll‐like receptors (TLR1, 2 and 4; West et al., 2011). Future studies are required to examine whether D₂ signalling mediates macrophage activation via mitochondrial ROS.

In summary, we found that dopamine system was present and functionally active in mouse and human macrophages. Myeloid‐specific D₂ receptor deletion worsened acute pancreatitis likely via mediating M1 macrophage polarization and functions. D₂ activation inhibited M1 macrophage polarization in vitro and mitigated NADPH oxidase‐mediated oxidative stress, NF‐κB and NLRP3 inflammasome activation. D₂ agonist protected against in vivo models of acute pancreatitis by preventing pro‐inflammatory M1 macrophage polarization in the pancreas (Figure 9). Our novel findings suggest that dopamine or D₂ agonist could be a promising therapeutic strategy for treating acute pancreatitis. More broadly, it opens up new therapeutic strategy for other severe inflammatory diseases. Because we found that D₂ signalling pathway derived from pancreas (Han et al., 2017) and myeloid cells (in the current study) controls pancreatic inflammation and pancreatitis severity, in addition, pancreatic acinar cell injury and excessive innate immune responses are two independent but synergetic pathophysiological events in the pathogenesis of acute pancreatitis (Sendler et al., 2018; Xue et al., 2014), so further studies are required to determine which cellular sources (pancreas‐specific or myeloid‐specific) of D₂ receptors play more predominant role during acute pancreatitis.

FIGURE 9.

Schematic diagram summarized the effect of D₂ receptor signalling on macrophage polarization and functions in experimental acute pancreatitis. D₂ receptor activation inhibits M1 macrophages and oxidative stress‐induced NF‐κB and inflammasome activation, which therefore controls pancreatic and systemic inflammation during acute pancreatitis (AP)

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

G.H. and X.W. designed and conceived the study. G.H., X.W., L.W., X.H., and J.N. provided funding to support the study. X.H., J.N., Z.W., J.W., B.L., and X.Y. performed the experiments and collected and analysed the data. J.D. and C.C. provided technical support in the in vivo and in vitro experiments. R.W. participated in the intellectual discussions. X.H. and L.W. drafted the manuscript. G.H. and L.W. supervised the study and revised the manuscript. All the authors approved the final version of the manuscript.

DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design & Analysis, Immunoblotting and Immunochemistry and Animal Experimentation, and as recommended by funding agencies, publishers and other organisations engaged with supporting research.

Supporting information

Table S1. Primer sequences used for qRT‐PCR analysis (Mouse)

Table S2. Primer sequences used for qRT‐PCR analysis (Human)

Figure S1. Expression of dopaminergic synthesis and dopamine receptors in mouse M1 and M2 macrophages. Immunofluorescent staining of D₁, D₂, TH, DDC, DBH and MAO immunofluorescence analyses in M1‐phenotype BMDMs

Figure S2. Generation and confirmation of conditional myeloidspecific D₂ knockout mice. (A) Images of RT‐PCR to confirm the presence of LysM^cre and D₂ ^flox bands. WT, wild‐type; Hom, homozygote; Het, heterozygote. (B) mRNA levels of D₂ by qRT‐PCR in BMDMs. (C) Representative pictures of H&E stained pancreas, lung, kidney, liver, heart, spleen sections (200 ×) from WT C57BL6/J and LysM^+/creD₂ ^{fl/ fl} on C57BL6/J background mice. ƫ P < 0.05 vs WT‐M1, one‐way ANOVA followed by Dunnett's multiple comparisons test

Figure S3. The changes of infiltrated macrophages in caerulein and LPS‐induced AP model. Mice were killed at 0, 6, 12, 24 and 48 hr after the first caerulein injection. (A) Representative images of macrophage marker F4/80 immunohistochemical analyses in the pancreas (200 ×). (B) Pancreatic leukocytes were isolated and CD45⁺CD11b⁺F4/80⁺ macrophages were sorted. FCM of TNFα, iNOS and CD206 expression of macrophages in the pancreas. n = 6 per group, mean ± SEM. Scale bar = 100 μm. * P < 0.05 versus 0 hr, one‐way ANOVA followed by Dunnett's multiple comparisons test

Figure S4. The effect of D₂ agonist alone on BMDM phenotype. (A) mRNA levels of inos, Tnfα, Il1β and Il6 by qRT‐PCR in BMDMs. (B) Bone marrow cells were stimulated with 5 μM quinpirole in the process of BMDM differentiation. Flow cytometry (FCM) of macrophage purity on day 3 and day 6. n = 6 per group, mean ± SEM. * P < 0.05 vs NC, one‐way ANOVA followed by Dunnett's multiple comparisons test

Figure S5. D₂ receptor activation alleviates NF‐κB activation in M1 macrophages in vitro. (A) Immunoblot analysis of NFκBp65 levels in M1‐phenotype BMDMs. (B) EMSA analysis of NF‐κB binding ability. (C) ELISA of TNFα, IL6 and IL10 in supernatants of M1‐phenotype BMDMs. n = 6 per group, mean ± SEM. * P < 0.05 vs NC; # P < 0.05 vs LPS and IFN‐γ, one‐way ANOVA followed by Dunnett's multiple comparisons test

Figure S6. D₂ receptor activation decreases TNFα⁺ M1 macrophage in two models of experimental AP. (A) Representative micrographs of TNFα with macrophage marker F4/80 immunofluorescence analyses in pancreas during caerulein and LPS induced AP. Cae, caerulein. (B) Representative micrographs of TNFα with macrophage marker F4/80 immunofluorescence analyses in pancreas during L‐arginine‐induced AP. L‐Arg, L‐arginine

Figure S7. Effect of different concentration of dopamine (DA) on dopamine receptors and inflammasome activation in BMDMs. (A) qRT‐PCR of mRNA levels of D₁ and D₂ in BMDMs (at 24 hr after DA incubation). (B) qRT‐PCR of mRNA levels of D₁ and D₂ in LPS and IFN‐γ‐stimulated BMDMs. (C) Immunoblot analysis of NLRP3, caspase1 and IL1β levels of BMDMs. (D) ELISA of IL18 in supernatants of BMDMs. * P < 0.05 versus NC, ^# P < 0.05 versus LPS and IFN‐γ, ⁺ P < 0.05 versus DA (500 μM)

Han X, Ni J, Wu Z, et al. Myeloid‐specific dopamine D₂ receptor signalling controls inflammation in acute pancreatitis via inhibiting M1 macrophage. Br J Pharmacol. 2020;177:2991–3008. 10.1111/bph.15026