William John Martin, Odette Shaw, Xiao Liu, Stefanie Steiger, and Jacquie L. Harper

Transcription

1 ARTHRITIS & RHEUMATISM Vol. 63, No. 5, May 2011, pp DOI /art , American College of Rheumatology Monosodium Urate Monohydrate Crystal Recruited Noninflammatory Monocytes Differentiate Into M1-like Proinflammatory Macrophages in a Peritoneal Murine Model of Gout William John Martin, Odette Shaw, Xiao Liu, Stefanie Steiger, and Jacquie L. Harper Objective. To profile monosodium urate monohydrate (MSU) crystal recruited monocyte inflammatory function during the course of in vivo differentiation, in a murine model of peritoneal MSU crystal induced inflammation. Methods. C57BL/6J mice were injected intraperitoneally with MSU crystals, and the peritoneal cells were harvested at different time points. The MSU crystal recruited monocyte/macrophage population was analyzed for the expression of differentiation and activation markers, cytokine production following MSU crystal restimulation ex vivo and in vivo, expression of NLRP3-associated proteins (ASC, caspase 1) and pro interleukin-1 (proil-1 ), and phagocytic capacity. Results. Monocytes recruited 8 hours after MSU crystal stimulation (F4/80 low Gr-1 int 7/4 ) exhibited poor phagocytic capacity, expressed low levels of proil-1, and failed to produce proinflammatory cytokines in response to MSU crystal restimulation. In the absence of MSU crystal restimulation, differentiating monocytes produced low levels of transforming growth factor 1 ex vivo, and this was abrogated following MSU crystal restimulation. Over time these cells developed a proinflammatory phenotype in vivo, characterized by the production of IL-1, tumor necrosis factor, IL-6, CCL2 (monocyte chemotactic protein 1), and CXCL1 (cytokine-induced neutrophil chemoattractant) following ex vivo MSU crystal restimulation, and leading to IL-1 production and cell infiltration following MSU William John Martin, PhD, Odette Shaw, BSc (Hons), Xiao Liu, MSc, Stefanie Steiger, Dipl. Bioeng., Jacquie L. Harper, PhD: Malaghan Institute of Medical Research, Wellington, New Zealand. Address correspondence to Jacquie L. Harper, PhD, Malaghan Institute of Medical Research, PO Box 7060, Wellington 6242, New Zealand. jharper@malaghan.org.nz. Submitted for publication July 11, 2010; accepted in revised form January 6, crystal rechallenge in vivo. Proinflammatory function was associated with differentiation toward a macrophage phenotype (F4/80 high Gr-1 7/4 ), an increase in phagocytic capacity, and an increase in the expression of proil-1. Conclusion. MSU crystal recruited monocytes differentiate into proinflammatory M1-like macrophages in vivo. This proinflammatory macrophage phenotype is likely to play a key role in perpetuating inflammation in gouty arthritis in the presence of ongoing deposition of fresh MSU crystals. Acute gout is an intensely painful form of inflammatory arthritis caused by the formation of monosodium urate monohydrate (MSU) crystals in the joint and connective tissues. An acute attack of gout is characterized by intense pain and swelling and reddening of the skin around the affected area. Without treatment this inflammatory response will resolve naturally over 7 10 days. Interleukin-1 (IL-1 ) has been identified as a pivotal cytokine in gout and MSU crystal induced inflammation (1 3). More recently, studies have identified IL-1 production following activation of the NLRP3 inflammasome in resident macrophages as a key event in the initiation of an attack of acute gout (4 7). However, there is a growing body of evidence indicating that cells of the monocyte/macrophage lineage play a pivotal role not only in the initiation but also in the progression and resolution of acute gouty inflammation. It has been proposed that the self-limiting nature of an acute gout attack is linked to the differentiation state of the MSU crystal recruited monocyte population. In vitro studies indicate that isolated blood monocytes that are differentiated into macrophages switch from producing proinflammatory cytokines to producing 1322

2 M1 MACROPHAGE DIFFERENTIATION IN MSU-INDUCED INFLAMMATION 1323 the antiinflammatory cytokine transforming growth factor 1 (TGF 1) in response to MSU crystal stimulation (8 10). These data would indicate that monocytes drive and differentiated macrophages resolve inflammation in gout. However, other studies show that MSU crystals induce a classic proinflammatory response in both primary and bone marrow derived macrophages (3,4,11 14), indicating a key role for macrophages in the initiation of the inflammatory cascade. In addition, there is evidence to suggest that upon recruitment to the site of inflammation, early infiltrating monocytes do not exhibit a proinflammatory phenotype and therefore may not be driving the early phase of inflammation in gout (5). Together, these findings raise questions about the true functional phenotypes expressed by MSU crystal recruited monocyte/macrophage populations in vivo. There is extensive evidence illustrating the ability of cells from the monocyte/macrophage cell lineage to express a diverse range of activities in response to changes in their local environment (15). Two general populations of macrophages, M1 and M2, have been classified based on their environmental drivers and their resulting phenotype (16). While M1 macrophages exhibit a proinflammatory phenotype that is classically driven by lipopolysaccharide (LPS) and interferon-,m2 macrophages express an antiinflammatory/repair phenotype that is driven by Th2 cytokines and glucocorticoids. The clearly opposing functions of the M1 and M2 macrophage phenotypes highlight the need to provide a physiologically relevant immune environment for the investigation of monocyte differentiation, and emphasize the importance of investigating the differentiation of MSU crystal recruited monocytes in vivo. The aim of this study was to profile the differentiation pathway of MSU crystal recruited monocytes in vivo to determine whether the infiltrating monocyte population differentiates into a proinflammatory (M1- like) or antiinflammatory (M2-like) macrophage phenotype in a murine model of acute MSU crystal induced inflammation. MATERIALS AND METHODS Mice. Male C57BL/6J mice were bred and housed in a conventional animal facility at the Malaghan Institute of Medical Research. All mice used for the experiments were aged 8 10 weeks. All experimental procedures were approved by the Victoria University Animal Ethics Committee (VUAEC) in accordance with the VUAEC guidelines for the care of animals. Reagents. Uric acid, LPS, and anti-mouse -actin (catalog no. A5441) were from Sigma. Anti-mouse Gr-1 monoclonal antibody (catalog no ) was from BD Biosciences. Anti-mouse IL-1 (catalog no ) and anti-mouse caspase 1 (catalog no ) were from BioVision Research Products, antibodies to ASC (catalog no. NB ) were from Novus Biologicals, and anti-mouse CD80 (catalog no ) and anti-mouse CD86 (catalog no ) were from ebioscience. F4/80 (catalog no. MCA497B), CD206 (catalog no. MCA2235PE), and clone 7/4 (catalog no. MCA771F) antibodies were from AbD Serotec. Horseradish peroxidase (HRP) conjugated goat anti-rabbit secondary antibody (catalog no. sc2004) was from Santa Cruz Biotechnology, and HRP-conjugated goat anti-mouse secondary antibody (catalog no. P0447) was from Dako. Chemiluminescence kits were from Thermo Fisher Scientific. PVDF membrane and Lowry reagent were from Bio-Rad. Complete, mini EDTA-free protease inhibitor cocktail (catalog no ) was from Roche Applied Science. Bio-Plex multiplex arrays were from Bio-Rad. Mouse IL-1 and TGF enzyme-linked immunosorbent assay (ELISA) kits were from R&D Systems. The Limulus amebocyte cell lysate assay was from Associates of Cape Cod. Heparin was from Mayne Pharma. Low Cell Binding plates were from Nunc. Diff-Quik was from Dade Behring Diagnostics. Lympholyte was from Cederlane Laboratories. Fluo- Spheres fluorescent microspheres (0.5 m, yellow green fluorescent 505/515 nm, 2% solids), NuPAGE 4 12% Bis-Tris Gel, MES sodium dodecyl sulfate (SDS) buffer, GolgiStop, and media and media supplements were from Invitrogen. Preparation of MSU crystals. MSU crystals were prepared and characterized as previously described (5). All MSU crystals were determined to be endotoxin free by Limulus amebocyte cell lysate assay ( 0.01 endotoxin units/10 mg). MSU crystal induced peritonitis. Mice were injected intraperitoneally (IP) with MSU crystals suspended in phosphate buffered saline (PBS) (3 mg/0.5 ml). At different time points after MSU crystal administration, mice were killed by CO 2 inhalation. The peritoneal cavity was lavaged with 3 ml PBS containing 25 units/ml heparin. Cells collected by peritoneal lavage were washed 3 times in PBS and retained for subsequent experiments. In vivo MSU crystal restimulation. Mice were injected IP with MSU crystals. After 48 hours, mice were given a second IP injection of MSU crystals, and peritoneal cells were collected as described above. In vivo labeling of monocyte/macrophages. Four days prior to the administration of MSU crystals, mice were injected IP with PKH26 (0.5 M in 0.5 ml PBS) (Sigma-Aldrich) in accordance with the manufacturer s instructions. Cell morphology. Cells harvested from MSU crystal treated mice were spun onto a glass slide using a cytocentrifuge and stained with Diff-Quik. Differential cell counts were determined microscopically using standard histologic criteria. Flow cytometry. For fluorescent staining of cell surface markers, peritoneal cells were washed and resuspended in fluorescence-activated cell sorting (FACS) buffer (0.1% bovine serum albumin, 0.01% sodium azide in PBS, ph 7.4), stained with fluorescent antibodies to the surface markers F4/80, Gr-1, clone 7/4, CD206, CD80, CD86, and CD14, and then washed and resuspended in FACS buffer. Monocyte/ macrophage populations were sorted based on their expression

3 1324 MARTIN ET AL of F4/80, Gr-1, and clone 7/4. FACS was performed using a FACS Diva (BD Biosciences). In vitro restimulation assay. Peritoneal exudate cells from MSU crystal injected mice were incubated in complete RPMI 1640 (10% fetal bovine serum [FBS], 100 units/ml penicillin, 100 units/ml streptomycin, 2 mm glutamax) at cells/ml in 96-well plates. For experiments requiring the measurement of TGF, cells were incubated in Macrophage Serum-Free Media (Invitrogen). Peritoneal exudate cells were restimulated ex vivo with MSU crystals (200 g/ml, 16 hours, 37 C). Supernatants were collected and stored at 20 C for subsequent cytokine analysis by ELISA or cytokine bead array. For intracellular cytokine staining, peritoneal cells were collected from mice 36 hours after MSU crystal administration in vivo. Peritoneal exudate cells ( ) were cultured in 24-well Low Cell Binding plates in complete RPMI 1640 containing 1:1,500 GolgiStop. Cells were stimulated with 200 g/ml MSU crystals, and 6 hours later cells were harvested for intracellular cytokine analysis by flow cytometry. Phagocytosis assay. Peritoneal cells were transferred to 10-ml Falcon tubes (2 ml, cells/ml, RPMI 1640, 10% FBS), fluorescent beads were added (1:2,000, final concentration % bead solids), and the cell suspension was incubated with shaking for 30 minutes at 37 C. The cell suspension was layered over the top of 1 ml of FBS and centrifuged at 500g for 5 minutes to pellet the cells, leaving free beads at the serum medium interface. The supernatant was removed, and the cell pellet was stained with fluorescent antibodies to the surface markers F4/80, clone 7/4, and Gr-1 and analyzed by flow cytometry. Western blotting. Peritoneal cells were washed in PBS, and monocyte/macrophages were isolated using Lympholyte solution in accordance with the manufacturer s instructions. The monocytes were lysed with protease inhibitor cocktail in Tris EDTA buffer. Protein concentrations were determined using the Lowry method, and sample concentrations were adjusted to 100 g of total protein. Protein samples were loaded alongside -actin loading control. Protein bands were separated by SDS polyacrylamide gel electrophoresis under reducing conditions (NuPAGE 4 12% Bis-Tris Gel, MES SDS buffer, 35 minutes at 200V) and transferred onto a PVDF membrane. Nonspecific protein binding was blocked with 5% nonfat milk (PBS with 0.2% Tween 20), and the membranes were probed overnight at 4 C with primary antibodies specific to IL-1, caspase 1, ASC, or -actin. Membranes were washed and incubated with HRP-conjugated goat anti-rabbit or goat anti-mouse secondary antibody. Proteins were visualized on film using chemiluminescence, and protein expression was normalized to the -actin loading control. RESULTS Cytokine production by MSU crystal recruited monocytes following MSU crystal restimulation ex vivo. Previously, it has been shown that MSU crystal recruited peritoneal monocytes isolated from mice 8 hours after induction of inflammation did not produce proinflammatory cytokines following MSU crystal restimulation (5). To determine if the recruited monocyte Figure 1. Cytokine production by monosodium urate monohydrate (MSU) crystal recruited monocytes in response to MSU crystal restimulation ex vivo. A, Mice were injected intraperitoneally with MSU crystals (3 mg in 0.5 ml phosphate buffered saline [PBS]), and peritoneal cells were harvested by lavage with 3 ml PBS. Peritoneal exudate cells ( cells/ml) were restimulated ex vivo for 16 hours with MSU crystals (200 g/ml) (solid bars) or PBS (open bars), and culture supernatants were analyzed for proinflammatory cytokines by enzyme-linked immunosorbent assay (ELISA) or multiplex bead array. B, Peritoneal exudate cells were restimulated (solid bars) or not (open bars) with MSU crystals as in A but were cultured Macrophage Serum-Free Media. Culture supernatants were analyzed for transforming growth factor (TGF ) by ELISA. Values are the mean SEM. Results are representative of 3 separate experiments. P 0.05; P 0.001, by Student s t-test. IL-1 interleukin-1 ; TNF tumor necrosis factor ; KC cytokineinduced neutrophil chemoattractant; MCP-1 monocyte chemotactic protein 1; N naive mice.

4 M1 MACROPHAGE DIFFERENTIATION IN MSU-INDUCED INFLAMMATION 1325 Figure 2. Inflammatory responsiveness to MSU crystal rechallenge in vivo. Mice were injected intraperitoneally with MSU crystals (3 mg in 0.5 ml PBS). After 48 hours, mice received a second injection of MSU crystals, and peritoneal cells were harvested by lavage with 3 ml PBS. A C, Shown are recruited neutrophils (A), recruited F4/80 monocyte/macrophages (B), and IL-1 levels (C) in peritoneal lavage fluid. Graphs show data from mice following 1 injection of MSU crystals (open bars) and 2 injections of MSU crystals (solid bars). Values are the mean SEM. Combined results from 2 independent experiments are shown (n 10 mice per group). P 0.05; P 0.01; P 0.001, by Student s t-test. D, Peritoneal exudate cells ( cells/ml; 88% recruited monocyte/macrophages, 0% neutrophils) from mice 72 hours after MSU crystal injection were restimulated ex vivo for 6 hours with MSU crystals (200 g/ml) in the presence of GolgiStop, and F4/80 cells were analyzed by flow cytometry for expression of intracellular IL-1, IL-6, and TNF. See Figure 1 for definitions. population retained a noninflammatory phenotype over time, we isolated peritoneal exudate cells at different time points following MSU crystal administration in vivo and measured the levels of cytokines and chemokines produced by these cells following MSU crystal restimulation ex vivo. MSU crystal recruited cells failed to produce proinflammatory cytokines in response to MSU crystal restimulation ex vivo until 3 days after MSU crystal administration in vivo (Figure 1A), when neutrophils were absent. Cells isolated 4 hours after induction of inflammation in vivo did not produce TGF 1 either alone or following MSU crystal restimulation (Figure 1B). However, in the absence of MSU crystal restimulation, the recruited cell population exhibited increased TGF 1 production over time compared with naive resident cells. The increase in background TGF 1 production peaked at 16 hours, correlating with maximal neutrophil infiltration. Interestingly, this TGF 1 production was partially abrogated following MSU crystal restimulation ex vivo (Figure 1B). MSU crystal restimulation in vivo. Next, we investigated whether the emerging proinflammatory responses observed ex vivo translated into proinflammatory responsiveness in vivo. As we have shown previously, 48 hours after IP MSU crystal administration, neutrophils had been cleared and the proinflammatory cytokine levels had returned to baseline (5); however, the numbers of recruited monocyte/macrophages remained elevated (Figure 2B), and these cells comprised 86% of the total cell population. At this time point, mice received a second treatment of MSU crystals, and neutrophil infiltration and IL-1 levels in the peritoneal fluid were measured. In vivo restimulation with MSU

5 1326 MARTIN ET AL Figure 3. Differentiation profile of MSU crystal recruited monocytes in vivo. Mice were injected intraperitoneally with PKH26 (0.5 M in 0.5 ml PBS). A, On day 4, resident peritoneal cells were harvested by lavage with 3 ml PBS and stained for the macrophage surface marker F4/80. B, Mice were injected intraperitoneally with MSU crystals (3 mg in 0.5 ml PBS), and peritoneal cells were collected 24 hours later by lavage. C, Peritoneal cells from MSU-treated mice were harvested by lavage, and F4/80 cells were analyzed by flow cytometry for the expression of Gr-1, clone 7/4, CD206, and CD86. Values are the mean SEM (n 5 mice per group). Results are representative of at least 2 separate experiments. D, Monocyte/macrophages from naive mice (F4/80 high Gr-1 7/4 ) and from mice 8 hours (F4/80 low Gr-1 int 7/4 ) and 48 hours (F4/80 low Gr-1 7/4 ) after intraperitoneal injection of MSU crystals were sorted by fluorescence-activated cell sorting. Cells were spun onto slides by cytospin and stained with Diff-Quik. Bars 20 m; original magnification 100. MFI mean fluorescence intensity (see Figure 1 for other definitions). crystals after 48 hours triggered significantly elevated neutrophil infiltration and IL-1 production compared with that observed following a single treatment with MSU crystals (Figures 2A and C). Analysis of the monocyte profile showed that following a single dose of MSU crystals, the numbers of recruited F4/80 monocyte/macrophages were higher at 48 hours compared with the numbers of resident macrophages from naive unchallenged mice (Figure 2B). However, 48 hours after the second in vivo MSU crystal treatment, the numbers of recruited monocyte/macrophages dropped significantly. Differentiation of MSU crystal recruited monocytes into proinflammatory macrophages in vivo. To more clearly identify the source of proinflammatory cytokine production following MSU crystal restimulation ex vivo, peritoneal cells were isolated 72 hours after MSU crystal administration in vivo and then restimulated with MSU crystals, and intracellular cytokine production was analyzed by flow cytometry. At 72 hours, the total monocyte numbers were still elevated in vivo (mean SEM cells/ml, 88% monocyte/macrophages) compared with the numbers of naive resident macrophages (mean SEM cells/ml, 89% naive resident macrophages). At 72 hours, the cells that expressed the macrophage marker F4/80 and low levels of the hemopoietic marker Gr-1 also exhibited MSU crystal induced proinflammatory cytokine production (Figure 2D). Proinflammatory cytokine production by other cell populations (identified as CD3 lymphocytes and natural killer antigen 1.1 positive natural killer cells) was not observed (data not shown). These results indicated that cells expressing a more macrophage phenotype were the likely source of MSU crystal induced proinflammatory cytokine production. Resident macrophages have been shown to become adherent following MSU crystal activation in vivo (5), a common feature of activated macrophages in response to inflammatory stimuli (17). To confirm that the activated resident macrophage population was not detaching and being harvested in the peritoneal wash after initiation of MSU crystal induced inflammation, resident macrophages were first stained by IP injection of the phagocyte-specific dye PKH26 prior to in vivo administration of MSU crystals. Resident macrophages (F4/80 high PKH26 high ) (Figure 3A) were not identified in the peritoneal wash cells following MSU crystal administration in vivo (Figure 3B). This confirmed that, after in vivo administration of MSU crystals, the monocyte/ macrophage cells (F4/80 low PKH26 int ) harvested from the peritoneum originated from the MSU crystal

6 M1 MACROPHAGE DIFFERENTIATION IN MSU-INDUCED INFLAMMATION 1327 recruited monocyte population. PKH26 neutrophils were not observed (data not shown). To determine whether the observed development of a proinflammatory response was indeed associated with a change in the MSU crystal recruited monocyte toward a macrophage phenotype in vivo, we looked for changes in the expression of hemopoietic monocyte markers (Gr-1, clone 7/4), the M2-associated surface marker CD206, and surface activation markers (CD80, CD86) on in vivo recruited monocytes, as well as for changes in the morphology of these monocytes, over time. Surface marker staining and histologic analysis of the infiltrating monocyte population showed that increased cytokine production correlated with downregulation of the hemopoietic monocyte markers Gr-1 (recognizing lymphocyte antigen 6 complex, locus C on monocytes) and clone 7/4 and with up-regulation of the Figure 4. Profile of phagocytic function of MSU crystal recruited monocyte/macrophages. Mice were injected intraperitoneally with MSU crystals (3 mg in 0.5 ml PBS), and peritoneal cells were harvested by lavage with 3 ml PBS. A, Peritoneal exudate cells (1 ml, cells) were incubated with fluorescent beads for 30 minutes. Nonphagocytosed beads were removed, and cells were stained for F4/80, Gr-1, and clone 7/4 and analyzed by flow cytometry. Bead uptake is represented by FL-1 fluorescence. B, Shown is bead uptake in F4/80 high (region 1 [R1]) and F4/80 low (R2) monocyte/macrophages in peritoneal exudate cells after 72 hours. Results are representative of 3 separate experiments. MFI mean fluorescence intensity (see Figure 1 for other definitions). Figure 5. Profile of CD14, ASC, caspase 1, and proil-1 expression by differentiating MSU crystal recruited monocytes. Mice were injected intraperitoneally with MSU crystals (3 mg in 0.5 ml PBS), and peritoneal cells were harvested by lavage with 3 ml PBS. A, Monocyte/ macrophages were purified from peritoneal exudate cells. Protein bands were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, and membranes were probed with primary antibodies specific to IL-1, caspase 1, or ASC. The loading control was -actin. B, Expression of CD14 on F4/80 monocytes and macrophages was monitored by flow cytometry. Values are the mean SEM (n 5 mice per group). Results are representative of at least 2 separate experiments. rm resident macrophages from naive mice; MFI mean fluorescence intensity (see Figure 1 for other definitions). activation marker CD86 (Figure 3C), as well as with the development of classic macrophage morphology (Figure 3D). There was no significant change in surface expres-

7 1328 MARTIN ET AL sion of the M2 macrophage marker CD206 (Figure 3C) or the activation marker CD80 (data not shown). These results indicated that development of inflammatory responsiveness to MSU crystal restimulation was associated with differentiation of the recruited monocytes toward a macrophage phenotype in vivo. F4/80 expression and phagocytic capacity. Next, we investigated possible reasons to explain why the in vivo differentiated macrophages produced proinflammatory cytokines in response to MSU crystal restimulation, while the early-recruited monocytes did not. MSU crystal induced inflammation is linked with phagocytosis of crystals by macrophages, leading to activation of the inflammatory cascade (18). To determine whether newly recruited monocytes were less phagocytic, monocyte/macrophage populations harvested at different time points following in vivo MSU crystal administration were tested for the ability to phagocytose fluorescent beads. Bead uptake by MSU crystal recruited F4/80 monocyte/macrophages showed that early infiltrating monocytes were less phagocytic than naive resident macrophages, showing 6.25% and 42.8% bead-positive cells, respectively (Figure 4A). Over time, the percentage of bead-positive monocyte/ macrophages increased. These cells also began to exhibit the capacity to phagocytose more than 1 bead. By 3 days, the emerging F4/80 high population (R1) exhibited a phagocytic profile that was similar to that of naive resident macrophages (Figure 4B). These results indicated that the differentiation of the MSU crystal recruited monocyte population could be contributing to the development of a proinflammatory macrophage phenotype potentially through an increased capacity to phagocytose MSU crystals. NLRP3 inflammasome, proil-1, and CD14 expression in MSU crystal recruited monocyte/macrophages. The assembly and activation of the NLRP3 inflammasome are considered to play a pivotal role in MSU crystal induced inflammation, being primarily responsible for the cleavage of proil-1 by caspase 1 and the release of active IL-1 (18). Therefore, the expression of ASC, procaspase 1, and caspase 1 and the accumulation of proil-1 could directly affect the ability of MSU crystal recruited monocytes and macrophages to raise a proinflammatory response to MSU crystal restimulation. The expression of ASC was found to be comparable in naive resident macrophages and MSU crystal recruited monocyte/macrophages, indicating that assembly of the NLRP3 inflammasome was not affected. The expression of procaspase 1 and caspase 1 was comparable or higher in recruited monocytes compared with resident macrophages. However, compared with naive resident macrophages, early MSU crystal recruited monocytes expressed lower levels of proil-1 and IL-1 (Figure 5A). Expression of these proteins increased as the recruited monocyte population differentiated toward a macrophage phenotype, and by 3 days the differentiating macrophages expressed significantly higher levels of proil-1 and IL-1 as well as of procaspase 1 and caspase 1 compared with naive resident macrophages. CD14 deficiency has been associated with depleted pools of proil-1, decreased caspase 1 activation, and impaired MSU crystal recognition (7). In this study, CD14 expression by early infiltrating monocyte/macrophages was low compared with that by resident macrophages (Figure 5B). After 3 days, monocyte/macrophages exhibited higher levels of CD14 expression, although the level of CD14 expression was still low compared with that by resident macrophages. DISCUSSION Contrary to the findings of previous in vitro studies, our results indicate that during the early stages of MSU crystal induced inflammation, the newly recruited monocyte population exhibits a noninflammatory phenotype that develops into a proinflammatory M1-like phenotype in vivo. Increased responsiveness to MSU crystal restimulation is associated with differentiation of the recruited monocyte population (F4/80 low Gr- 1 int 7/4 ) toward a macrophage phenotype (F4/80 high Gr- 1 7/4 ) and morphology, an increase in phagocytic capacity, and an increase in the expression of proil-1. Reports of earlier in vitro studies investigating the effect of monocyte differentiation on responsiveness to MSU crystals propose that proinflammatory monocytes differentiate into noninflammatory or suppressive macrophages. However, those studies have predominantly used panels of monocyte/macrophage cell lines or primary monocytes differentiated in vitro over several days in sterile media (8 10,19,20). These in vitro environments differ significantly from the highly complex and dynamic inflammatory environment encountered by MSU crystal recruited monocytes in vivo. One possible reason for the apparent lack of a proinflammatory macrophage response in earlier in vitro studies is the absence of continual low-level Toll-like receptor (TLR) signaling by TLR ligands that is linked with priming of intracellular stores of proil-1 in vivo (3,21). Without

8 M1 MACROPHAGE DIFFERENTIATION IN MSU-INDUCED INFLAMMATION 1329 Figure 6. Phenotypic profile of in vivo monocyte-to-macrophage differentiation in MSU crystal induced inflammation (see Discussion). Ly-6C lymphocyte antigen 6 complex, locus C (see Figure 1 for other definitions). this natural priming event, cultured macrophages would be unable to raise a proinflammatory response to MSU crystals in vitro. A key strength of the current study is that MSU crystal recruited monocyte differentiation occurs during the course of the MSU crystal induced response in vivo. As such, this approach provides a more accurate representation of the natural process of monocyte-to-macrophage differentiation. In this study we confirm that early infiltrating monocytes are unresponsive to MSU crystal stimulation ex vivo. This finding provides further evidence that the resident macrophage, rather than the newly recruited monocytes, initiates the primary phase of inflammation in vivo (5). The lack of proinflammatory activity by early-recruited monocytes appears to be primarily associated with low levels of proil-1. The activation of the NLRP3 inflammasome and the cleavage of proil-1 by active caspase 1 to release active IL-1 are important steps in the initiation of MSU crystal induced inflammation (4 7). As a result, low levels of proil-1 would significantly block the ability of newly recruited monocytes to raise a proinflammatory response to MSU crystal restimulation, despite the presence of active caspase 1. The generation of proil-1 pools has been shown to be impaired in CD14-knockout macrophages (7). It is therefore possible that the low CD14 expression observed on MSU crystal recruited monocytes plays a key role in blocking proil-1 accumulation, resulting in a noninflammatory monocyte phenotype. This may be compounded by the poor phagocytic activity of earlyrecruited monocyte/macrophages. It has been proposed that CD14 rather than phagocytosis of crystals is important for IL-1 production (7). However, it is more likely that a certain degree of codependency exists, whereby CD14 signaling is necessary for proil-1 production but crystal phagocytosis is required to trigger activation of the NLRP3 inflammasome and release of active IL-1. Therefore, in the absence of either CD14 expression or phagocytic function, MSU crystal induced activation of monocyte/macrophages may not occur. Our data now show that MSU crystal recruited monocytes differentiate into an M1-like (proinflammatory) functional phenotype in vivo, as illustrated by the

9 1330 MARTIN ET AL ability of day 3 differentiated monocytes to produce the classic proinflammatory molecules following MSU crystal restimulation ex vivo. Our data also show that significant accumulation of proil-1 and procaspase 1, in association with increased CD14 expression and phagocytic capacity (22), may be priming the emerging macrophage phenotype for hyperresponsiveness to MSU crystal stimulation. It is interesting to note that elevated intracellular expression of the active form of IL-1 in MSU crystal recruited M1-like macrophages did not correlate with elevated IL-1 expression in vivo, indicating that active IL-1 was not being secreted. This provides evidence for the accumulation of a pool of active IL-1 in M1-like macrophages. The availability of this pool of IL-1 for immediate release upon MSU crystal stimulation may facilitate the rapid, exacerbated inflammatory response observed following MSU crystal restimulation in vivo. Contrary to the development of an M1-like macrophage phenotype, TGF 1 production by unstimulated MSU crystal recruited cells was observed ex vivo. However, the absence of the M2 surface marker (CD206) indicates that this TGF 1 production is not associated with the development of M2 macrophages. Instead, the observed increase in background TGF 1 production over time is likely associated with an increase in the ability of the differentiating monocyte/macrophages to phagocytose apoptotic neutrophils, a process strongly linked with TGF 1 production, neutrophil clearance, and resolution of acute inflammation (23 25). Whether TGF 1 production plays a role in the accumulation of intracellular IL-1 in the recruited M1-like macrophage, possibly by blocking cytokine secretion, remains to be determined. Interestingly, the ex vivo restimulation data indicate that MSU crystal activation can abrogate TGF 1 production linked with neutrophil clearance. At this time it is not clear how this occurs. However, it does illustrate that the differentiating recruited macrophage has the capacity to respond to MSU crystals and shut down TGF 1-driven resolution, independent of the ability to release proinflammatory cytokines. The in vivo rechallenge data provide further evidence that the emerging proinflammatory M1-like macrophage can override the process of resolution and is capable of initiating a second, exacerbated inflammatory insult. Interestingly, features of this secondary response appear to be more strictly regulated, leading to a more rapid shutdown of proinflammatory cytokine production and a significant drop in the accumulation of monocytes and macrophages over time. Nevertheless, it appears that in the first instance both the development of the proinflammatory M1-like macrophage phenotype in vivo and the continued deposition of fresh MSU crystals at the site of inflammation are necessary for overriding resolution and driving ongoing inflammation in an attack of gout. Historically, neutrophils have been widely viewed as the major driving force for orchestrating inflammation in and resolution of gout attacks. Both MSU crystals and the inflammatory cytokine environment have been shown to prolong neutrophil survival and stimulate neutrophil IL-8 (CXCL8) and superoxide production, factors that can contribute to ongoing inflammation (26 30). In contrast, neutrophil apoptosis and clearance by phagocytic cells are recognized as playing a key part in TGF 1 production and resolution (24,26,28,31,32). Based on our findings, the differentiation of recruited monocytes into M1-like macrophages may play an equally important role in directing the progression of inflammation in gout. These macrophages appear to be able to contribute to inflammation through the expression of a functional inflammatory phenotype, but they may also represent the key phagocytic cells involved in neutrophil clearance, thereby facilitating resolution. Although care needs to be taken when directly extrapolating the M1-like functional response ex vivo to the in vivo environment, the in vivo restimulation data provide strong evidence that the observed proinflammatory M1-like macrophage phenotype is functionally relevant in vivo. The results of this study unveil a new level of complexity in our understanding of the progression of gouty arthritis, whereby recruited monocytes, along with neutrophils, contribute to the progression of inflammation in and resolution of gout (Figure 6). In this model, resident macrophage activation initiates the inflammatory cascade, resulting in the recruitment of neutrophils and monocytes. Early MSU crystal recruited monocytes exhibit a nonresponsive phenotype in vivo. However, during this early phase, recruited neutrophils may respond to both MSU crystals and local mediators of inflammation to produce CXCL8 and drive selfrecruitment. Over time, the recruited monocyte population down-regulates expression of markers of hemopoiesis and becomes primed to respond to ongoing MSU crystal deposition via increased phagocytic capacity and expression of CD14, and via the accumulation of proil- 1, active caspase 1, and intracellular IL-1, consistent with differentiation into an M1-like macrophage. In the absence of ongoing inflammatory stimulation, these

10 M1 MACROPHAGE DIFFERENTIATION IN MSU-INDUCED INFLAMMATION 1331 phagocytic macrophages contribute to neutrophil clearance and resolution of the inflammatory response. However, the stimulation of these proinflammatory M1-like macrophages with fresh MSU crystals then overrides TGF 1-dependent resolution and drives a secondary wave of inflammation in vivo. This study serves to further emphasize the integral involvement of the monocyte/macrophage phenotype in the inflammation profile of a gout attack. Importantly, our data show for the first time that differentiation of MSU crystal recruited monocytes into a proinflammatory M1-like macrophage phenotype in vivo, combined with ongoing MSU crystal deposition, may play a significant role in abrogating resolution and perpetuating inflammation in gout. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Harper had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Martin, Harper. Acquisition of data. Martin, Shaw, Liu, Steiger. Analysis and interpretation of data. Martin, Shaw, Steiger, Harper. REFERENCES 1. Roberge CJ, Grassi J, De Medicis R, Frobert Y, Lussier A, Naccache PH, et al. Crystal-neutrophil interactions lead to interleukin-1 synthesis. Agents Actions 1991;34: So A, De Smedt T, Revaz S, Tschopp J. A pilot study of IL-1 inhibition by anakinra in acute gout. Arthritis Res Ther 2007;9: R Chen CJ, Shi Y, Hearn A, Fitzgerald K, Golenbock D, Reed G, et al. MyD88-dependent IL-1 receptor signaling is essential for gouty inflammation stimulated by monosodium urate crystals. J Clin Invest 2006;116: Martinon F, Petrilli V, Mayor A, Tardivel A, Tschopp J. Goutassociated uric acid crystals activate the NALP3 inflammasome. Nature 2006;440: Martin WJ, Walton M, Harper J. Resident macrophages initiating and driving inflammation in a monosodium urate monohydrate crystal induced murine peritoneal model of acute gout. Arthritis Rheum 2009;60: Pope RM, Tschopp J. The role of interleukin-1 and the inflammasome in gout: implications for therapy [review]. Arthritis Rheum 2007;56: Scott P, Ma H, Viriyakosol S, Terkeltaub R, Liu-Bryan R. Engagement of CD14 mediates the inflammatory potential of monosodium urate crystals. J Immunol 2006;177: Landis RC, Yagnik DR, Florey O, Philippidis P, Emons V, Mason JC, et al. Safe disposal of inflammatory monosodium urate monohydrate crystals by differentiated macrophages. Arthritis Rheum 2002;46: Yagnik DR, Evans BJ, Florey O, Mason JC, Landis RC, Haskard DO. Macrophage release of transforming growth factor 1 during resolution of monosodium urate monohydrate crystal induced inflammation. Arthritis Rheum 2004;50: Yagnik DR, Hillyer P, Marshall D, Smythe CD, Krausz T, Haskard DO, et al. Noninflammatory phagocytosis of monosodium urate monohydrate crystals by mouse macrophages: implications for the control of joint inflammation in gout. Arthritis Rheum 2000;43: Jaramillo M, Godbout M, Naccache PH, Olivier M. Signaling events involved in macrophage chemokine expression in response to monosodium urate crystals. J Biol Chem 2004;279: Jaramillo M, Naccache PH, Olivier M. Monosodium urate crystals synergize with IFN- to generate macrophage nitric oxide: involvement of extracellular signal-regulated kinase 1/2 and NF- B. J Immunol 2004;172: Chen L, Hsieh MS, Ho HC, Liu YH, Chou DT, Tsai SH. Stimulation of inducible nitric oxide synthase by monosodium urate crystals in macrophages and expression of inos in gouty arthritis. Nitric Oxide 2004;11: Murakami Y, Akahoshi T, Hayashi I, Endo H, Kawai S, Inoue M, et al. Induction of triggering receptor expressed on myeloid cells 1 in murine resident peritoneal macrophages by monosodium urate monohydrate crystals. Arthritis Rheum 2006;54: Stout RD, Suttles J. Functional plasticity of macrophages: reversible adaptation to changing microenvironments. J Leukoc Biol 2004;76: Martinez FO, Sica A, Mantovani A, Locati M. Macrophage activation and polarization. Front Biosci 2008;13: Barth MW, Hendrzak JA, Melnicoff MJ, Morahan PS. Review of the macrophage disappearance reaction. J Leukoc Biol 1995;57: Hornung V, Bauernfeind F, Halle A, Samstad EO, Kono H, Rock KL, et al. Silica crystals and aluminum salts activate the NALP3 inflammasome through phagosomal destabilization. Nat Immunol 2008;9: Chapman PT, Yarwood H, Harrison AA, Stocker CJ, Jamar F, Gundel RH, et al. Endothelial activation in monosodium urate monohydrate crystal induced inflammation: in vitro and in vivo studies on the roles of tumor necrosis factor and interleukin-1. Arthritis Rheum 1997;40: Landis RC, Haskard DO. Pathogenesis of crystal-induced inflammation. Curr Rheumatol Rep 2001;3: Netea MG, van de Veerdonk FL, Kullberg BJ, Van der Meer JW, Joosten LA. The role of NLRs and TLRs in the activation of the inflammasome. Expert Opin Biol Ther 2008;8: Leenen PJ, Jansen AM, van Ewijk W. Murine macrophage cell lines can be ordered in a linear differentiation sequence. Differentiation 1986;32: Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-, PGE2, and PAF. J Clin Invest 1998;101: Rose DM, Sydlaske AD, Agha-Babakhani A, Johnson K, Terkeltaub R. Transglutaminase 2 limits murine peritoneal acute goutlike inflammation by regulating macrophage clearance of apoptotic neutrophils. Arthritis Rheum 2006;54: Huynh ML, Fadok VA, Henson PM. Phosphatidylserine-dependent ingestion of apoptotic cells promotes TGF- 1 secretion and the resolution of inflammation. J Clin Invest 2002;109: Martin WJ, Grainger R, Harrison A, Harper JL. Differences in MSU-induced superoxide responses by neutrophils from gout subjects compared to healthy controls and a role for environmental inflammatory cytokines and hyperuricemia in neutrophil function and survival. J Rheumatol 2010;37: Hachicha M, Naccache PH, McColl SR. Inflammatory microcrystals differentially regulate the secretion of macrophage inflammatory protein 1 and interleukin 8 by human neutrophils: a possible mechanism of neutrophil recruitment to sites of inflammation in synovitis. J Exp Med 1995;182:

11 1332 MARTIN ET AL 28. Popa-Nita O, Naccache PH. Crystal-induced neutrophil activation. Immunol Cell Biol 2010;88: Abramson S, Hoffstein ST, Weissmann G. Superoxide anion generation by human neutrophils exposed to monosodium urate: effect of protein adsorption and complement activation. Arthritis Rheum 1982;25: Chia EW, Grainger R, Harper JL. Colchicine suppresses neutrophil superoxide production in a murine model of gouty arthritis: a rationale for use of low-dose colchicine. Br J Pharmacol 2008;153: Nishimura A, Akahoshi T, Takahashi M, Takagishi K, Itoman M, Kondo H, et al. Attenuation of monosodium urate crystal-induced arthritis in rabbits by a neutralizing antibody against interleukin-8. J Leukoc Biol 1997;62: Martin WJ, Harper JL. Innate inflammation and resolution in acute gout. Immunol Cell Biol 2010;88:15 9.