Triggering of NOD2 Receptor Converts... (PDF Download Available)

June 12, 2018 | Author: Anonymous | Category: Documents
Share Embed


Short Description

Aug 24, 2017 - 5. Lead Contact. *Correspondence: [email protected] ...... libraries was 400 bp. Bioi...

Description

Article

Triggering of NOD2 Receptor Converts Inflammatory Ly6Chigh into Ly6Clow Monocytes with Patrolling Properties Graphical Abstract

Authors Anne-Julie Lessard, Manon LeBel, Benoit Egarnes, ..., Alexandre Brunet, Serge Rivest, Jean Gosselin

Correspondence [email protected]

In Brief The signals that regulate the conversion of inflammatory monocytes into patrolling subset(s) remain unknown. Here, Lessard et al. demonstrate that triggering NOD2 transforms inflammatory Ly6Chigh monocytes into Ly6Clow monocytes that look and function like patrolling cells.

Highlights d

d

d

d

Triggering NOD2 converts Ly6C monocytes

Accession Number high

into Ly6C

low

patrolling

Converted Ly6Chigh monocytes express typical markers of patrolling monocytes Triggering NOD2 increases crawling of converted Ly6Chigh monocytes Triggering NOD2 converts human primary CD142+ CD16 into CD14± CD162+ monocytes.

Lessard et al., 2017, Cell Reports 20, 1830–1843 August 22, 2017 ª 2017 The Author(s). http://dx.doi.org/10.1016/j.celrep.2017.08.009

GSE101496

Cell Reports

Article Triggering of NOD2 Receptor Converts Inflammatory Ly6Chigh into Ly6Clow Monocytes with Patrolling Properties Anne-Julie Lessard,1,4 Manon LeBel,1,4 Benoit Egarnes,1 Paul Pre´fontaine,3 Peter The´riault,3 Arnaud Droit,2,3 Alexandre Brunet,2,3 Serge Rivest,2,3 and Jean Gosselin1,2,5,* 1Laboratory

of Innate Immunology, Centre de recherche du CHU de Que´bec-Universite´ Laval, Que´bec, QC G1V 4G2, Canada of Molecular Medicine, Universite´ Laval, Que´bec, QC G1V 4G2, Canada 3Centre de recherche du CHU de Que ´ bec, Universite´ Laval, Que´bec, QC G1V 4G2, Canada 4These authors contributed equally 5Lead Contact *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2017.08.009 2Department

SUMMARY

The signals that regulate the fate of circulating monocytes remain unknown. In the present study, we demonstrate that triggering of the NOD2 receptor by muramyl dipeptide (MDP) converts inflammatory Ly6Chigh monocytes into patrolling Ly6Clow monocytes. Administration of MDP to Nr4a1/ mice, which lack Ly6Clow monocytes, or to Ly6Clow-depleted mice led to the emergence of blood-patrolling monocytes with a profile similar to that of Ly6Clow monocytes, including high expression of CX3CR1 and LFA1. Using intravital microscopy in animal models of inflammatory diseases, we also found that converted Ly6Chigh monocytes patrol the endothelium of blood vessels and that their presence contributes to a reduction in the inflammatory response following MDP injection. Our results demonstrate that NOD2 contributes to the regulation of blood monocytes and suggest that it could be therapeutically targeted to treat inflammatory diseases. INTRODUCTION NOD2 receptor, a member of the NOD-like receptor family, is recognized as a cytosolic protein that responds to fragments of bacterial peptidoglycan such as muramyl dipeptide (MDP) to activate innate immunity to bacterial infection (Girardin et al., 2003a; Grimes et al., 2012; Inohara et al., 2000; Ogura et al., 2001). It was reported that viral single-stranded RNAs can also activate NOD2 to induce the production of type 1 IFNs through the adaptor protein IPS-1 (Sabbah et al., 2009). Although the antimicrobial functions of NOD2 are now supported by several studies, the role of NOD2 in the inflammatory response remains to be clarified. Different studies have reported that NOD2 contributes to initiate an inflammatory response because its triggering in response to bacterial and viral infections leads to the production of a wide range of inflammatory mediators (Coulombe et al., 2012; Kim et al., 2011; Macho Fernandez et al.,

2011). On the other hand, other reports suggest that NOD2 could play a significant role in the control of inflammation. Indeed, NOD2 polymorphisms are associated with the susceptibility to develop Crohn’s disease (CD), which is characterized by a chronic inflammation of the digestive tract (Hugot et al., 2001; Ogura et al., 2001; reviewed by Philpott et al., 2014). NOD2 variants were also linked to other inflammatory disorders, such as Blau syndrome and sarcoidosis (Caso et al., 2014; Kanazawa et al., 2005). Although the mechanism(s) remains to be defined, it is proposed that NOD2 variants in CD patients weaken bacterial clearance in the gut and consequently alter the composition of the gut microbiota. Such alterations may provoke massive recruitment of immune cells and aberrant inflammation (Inohara et al., 2003; Ogura et al., 2001; Zigmond et al., 2012). The lack of functional NOD2 may also fail to limit Toll-like receptor-induced production of inflammatory cytokines, causing an enhanced intestinal inflammation (Watanabe et al., 2006). Although the bulk of published studies on NOD2 functions have described mechanisms associated with its control of infections by microorganisms and its interactions with the immune response, how NOD2 can influence the inflammatory response remains to be elucidated. Monocytes, which express NOD2 (Ogura et al., 2001), are a subset of blood leukocytes that play a crucial role in the regulation of inflammation. They are rapidly recruited to lymphoid and non-lymphoid tissues in response to tissue injury or pathogens, where they produce cytokines and differentiate into dendritic cells and macrophages. In humans, blood monocytes are divided into three subsets according to the expression of CD14 and CD16. Counterparts to human monocyte subsets have now been characterized in mice. They are divided into at least two major subsets on the basis of their phenotype and functions (Auffray et al., 2009; Geissmann et al., 2003). The inflammatory monocytes, defined as Ly6Chigh CCR2+ CX3CR1low CD62L+, extravasate in inflamed tissues in a CCR2-dependent manner and thus contribute to local inflammation (Auffray et al., 2007; Geissmann et al., 2003; Huo et al., 2001; Serbina and Pamer, 2006). The second subset of monocytes is defined as Ly6Clow CCR2 CX3CR1high CD62L and referred to as patrolling monocytes because they patrol the blood vessels and can infiltrate tissue in response to damage or infection to initiate an early and transient inflammatory

1830 Cell Reports 20, 1830–1843, August 22, 2017 ª 2017 The Author(s). This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

response (Auffray et al., 2007; reviewed by Auffray et al., 2009). Patrolling monocytes are also involved in the later phase of inflammation and participate in wound healing. Our previous study showed that NOD2 activation by MDP, a known agonist of NOD2 (Girardin et al., 2003a), can reduce lung inflammation in mice infected with influenza virus (Coulombe et al., 2012). Treatment of infected mice with MDP was found to cause a selective increase in circulating monocytes, along with a transient recruitment of Ly6Chigh monocytes to the lungs of mice. This prompted us to further evaluate whether triggering of NOD2 could contribute to modulate biological functions of monocytes and consequently, the features of the inflammatory response. Here we attempted to characterize the role of NOD2 on monocyte plasticity and function in vivo. We observed that stimulation of NOD2 receptor with MDP induces the polarization of pro-inflammatory Ly6Chigh monocytes into patrolling Ly6Clow monocytes. Moreover, treatment of NR4A1-deficient mice, which lack patrolling Ly6Clow monocytes, with MDP was found to promote the frequency of Ly6Clow monocytes, which present phenotype and functions of typical circulating Ly6Clow monocytes. Together, our results provide insights on the capacity of NOD2 to regulate differentiation of Ly6Chigh monocytes, a mechanism that might contribute to regulate inflammation. RESULTS In Vivo Treatment with MDP Converts Ly6Chigh Monocytes into Ly6Clow Patrolling Monocytes Monocytes are recognized to play a crucial role during both homeostasis and inflammation, but the signals that drive monocyte differentiation and function remain unclear. Because NOD2 was suspected to play a role in inflammation, we wanted first to evaluate the effects of NOD2 triggering on monocyte subsets. A gating strategy to identify monocyte subsets by flow cytometry is presented in Figure S1. Intravenous (i.v.) administration of NOD2 agonist MDP to wild-type (WT) mice significantly increases the level of blood monocytes after 2 days of treatment to reach almost a 1.5-fold increase after 96 hr of stimulation, suggesting that MDP treatment might stimulate the exit of monocytes from the bone marrow (BM) (Figure 1A). As expected, such effects of MDP were abolished in Nod2/ mice. Levels of circulating Ly6Chigh monocytes in WT mice were not significantly affected, but numbers of Ly6Clow monocytes were found to progressively increase during treatment with MDP (Figures 1B and 1C). Ly6Cinter monocytes gradually increase and stabilize thereafter, supporting the transitory nature of this monocyte subset. Comparable results were obtained with two other agonists tested, N-glycolyl-MDP and L18-MDP (Figure S2) (Ermann et al., 2014; Feinen et al., 2014; Pandey et al., 2009). No effect was detected when using murabutide. We next took advantage of Nr4a1/ mice, which lack Ly6Clow monocytes (Hanna et al., 2011), to evaluate the potential of MDP treatment to generate Ly6Clow monocytes. We did not detect a significant increase in total blood monocyte levels following MDP treatment compared with controls (Figure 1A). However, we found that levels of Ly6Chigh monocytes, which represent approximately 85% of total blood monocytes in Nr4a1/ mice, were reduced by more

than 50% following MDP treatment compared with the controls (Figures 1B and 1C). On the other hand, the number of circulating Ly6Clow monocytes reaches an average 4-fold increase within 48 hr compared with vehicle-treated Nr4a1/ mice. Again, the general tendency shows an increase of Ly6Cinter after MDP treatment, reflecting a transient status of this monocyte subset. In order to evaluate the capacity of MDP to switch in vivo inflammatory Ly6Chigh monocytes into patrolling monocytes, we depleted monocytes in the circulation by administration of liposomes loaded with clodronate (Sunderko¨tter et al., 2004; Van Rooijen and Sanders, 1994) and monitored the kinetics of blood monocyte repopulation following treatment with the NOD2 agonist MDP. A schematic representation of the procedure is illustrated in Figure 2A. The majority of monocytes are eliminated by day 1 post-liposome injection (1% remaining), but patrolling monocytes remain absent and re-emerge only by day 6 postclodronate injection, which facilitates the visualization of Ly6Clow monocyte emergence by flow cytometry (Brunet et al., 2016). In clodronate-treated mice, we found that exclusively Ly6Chigh monocytes reached almost normal levels in 48 hr, and Ly6Clow monocytes remained absent (Figures 2B and 2C). In contrast, when mice are treated with MDP, Ly6Clow monocytes had started to repopulate the bloodstream at 72 hr and gradually increased up to 96 hr (Figures 2B and 2C). In addition, we also observed that the number of Ly6Cinter monocytes rapidly increased (Figure 2B) following treatment with MDP, suggesting that Ly6Chigh monocytes give rise to Ly6Clow monocytes through a cellular transition in Ly6Cinter phenotype. To confirm more directly the consequences of NOD2 triggering on Ly6Chigh monocyte conversion, we pulse-labeled mice with DiO-labeled liposomes (DiO-lipo) after clodronate-liposome-induced depletion of monocytes (Figure 2D). As expected, the large majority of re-emerged DiO-labeled monocytes have the Ly6Chigh phenotype (87.7%) at 72 hr, while the percentage of Ly6Clow gradually increased with time to reach 66.7% at 144 hr (Figure 2E). At 96 hr, we observed that a significant portion of the circulating DiOlabeled Ly6Chigh cells had shifted to Ly6Cint and Ly6Clow phenotype. Interestingly, at the later time, the number of DiO-positive Ly6Cint and Ly6Clow cells was increased to a level that could not have emerged only from the DiO-labeled Ly6Chigh in the circulation. In fact, we have reasons to believe that a significant number of DiO-labeled Ly6Chigh cells egressed the BM to spontaneously mature into Ly6Cint and consequently Ly6Clow monocytes. In mice treated with MDP, levels of DiO-labeled Ly6Clow monocytes appeared faster in the circulation to reach 72.9% of total blood monocytes after 120 hr of treatment (Figure 2E). Again, these results support the concept that Ly6Chigh might give rise to Ly6Clow monocytes in the circulation and show that triggering of NOD2 receptor accelerates such conversion of Ly6Chigh monocytes into the Ly6Clow phenotype. To further evaluate whether treatment with MDP can drive the switch of Ly6Chigh into Ly6Clow monocytes, we adoptively transferred CD45.1 BM cells into CD45.2 mice and evaluated their differentiation after MDP treatment. First, we confirmed that Ly6Chigh monocytes constituted the majority of CD45.1 grafted monocytes (94.8%) (Figure 2F). As shown in Figure 2G, flow cytometry allowed us to discriminate between CD45.1 and CD45.2 populations after transfer. After engraftment of CD45.1 cells, the

Cell Reports 20, 1830–1843, August 22, 2017 1831

Figure 1. MDP Treatment Promotes the Conversion of Inflammatory Ly6Chigh Monocytes into a Ly6Clow Patrolling Subset Wild-type, Nr4a1/, and Nod2/ mice (n = 3–6 mice/group) were treated daily with vehicle (saline 0.9%, i.v.) or MDP (10 mg/kg, i.v.), and blood was collected at indicated times after MDP treatment. (A) Absolute count of blood total monocytes following treatment with vehicle (full lines) or MDP (dashed lines) as measured by flow cytometry. (B) Flow cytometry dot plots for blood Ly6Chigh, Ly6Cinter, and Ly6Clow monocyte number from wild-type, Nr4a1/, and Nod2/ mice treated with vehicle or MDP (48 hr). (C) Kinetics of blood Ly6Chigh, Ly6Cinter, and Ly6Clow monocyte number following daily treatment with vehicle (full lines) or MDP (dashed lines) as measured by flow cytometry. Data are presented as mean ± SEM of three independent experiments. *p % 0.05, **p % 0.01, ***p % 0.001, and ****p % 0.0001 (unpaired t test).

proportion of CD45.1 Ly6Clow monocytes was drastically increased (15.1%), as early as 1 hr after MDP treatment (Figure 2H). At 24 hr after engraftment, MDP-treated mice showed increased Ly6Clow (23.9%) and especially Ly6Cinter (37.8%) compared with control mice, which respectively had 20.2% Ly6Clow and 7.9% Ly6Cinter. We also confirmed the conversion of adoptively transferred CD45.1 Ly6Chigh sorted monocytes. Although reduced in number, both CD45.1 Ly6Clow and Ly6Cinter monocytes were increased after MDP treatment (Figure 2I). We evaluated if the absence of chemokine receptors CCR2 and CX3CR1 on monocyte subsets might affect MDP-induced

1832 Cell Reports 20, 1830–1843, August 22, 2017

differentiation of Ly6Chigh into Ly6Clow monocytes. As expected, deficiency of CCR2 reduced the percentage of total recruited blood monocytes compared with WT controls (Figure S3A), because Ly6Chigh monocyte recruitment is strongly mediated through CCR2 receptor. Deletion of CX3CR1 did not alter the effects of MDP treatment on total blood monocytes. However, we observed that in both Ccr2/ and Cx3cr1/ mice, MDP treatment strongly increased the percentage of Ly6Clow monocytes (Figures S3B and S3C), supporting the intrinsic capacity of NOD2 to polarize monocytes from an inflammatory to a patrolling phenotype.

Figure 2. MDP Treatment Accelerates the Re-emergence of Ly6Clow Blood Monocytes following Clodronate Administration (A) Experimental design of clodronate liposomes (Clo-lipo) and MDP administration. Wild-type mice (n = 3–6 mice/group) were injected with PBS-liposomes (PBS-lipo) as control or Clo-lipo (i.v.) in order to deplete blood monocytes. MDP was injected daily (i.v.) 24 hr following clodronate administration and animals were sacrificed at indicated times.

(legend continued on next page)

Cell Reports 20, 1830–1843, August 22, 2017 1833

NOD2 Triggering Gives Rise to Ly6Clow Patrolling Monocytes in the BM and Spleen of Mice The percentage of Ly6Clow monocytes in the BM of mice is very low but significantly higher in spleen of mice (Swirski et al., 2009; van Furth and Cohn, 1968). We have evaluated whether MDP treatment of mice could also induce differentiation of Ly6Chigh into Ly6Clow monocytes in these two compartments. In BM, monocytes were identified as Lin CD117 CD115+ CD11b+ Ly6C+ or  and in spleen as Lin CD11b+ CD11c Ly6C±. Treatment with MDP increased Ly6Clow frequencies in BM after 48 hr, and in parallel, levels of Ly6Chigh decreased (Figures 3A and 3B). In spleen of MDP-treated mice, levels of Ly6Chigh monocytes rapidly decreased after 24 hr, but as observed in the circulation, numbers of Ly6Clow monocytes were found to increase following 48 hr of treatment (Figures 3C and 3D). While expressing NOD2, in vivo treatment with MDP did not affect phenotype of splenic pDC, cDC, and circulating neutrophils (Figures S4A and S4B). We also used a flow cytometry approach to discriminate between different progenitors associated with the monocyte maturation pathway, including LSK, MDPs, and cMOP (Figures S5A and S5B). Our results indicate that triggering of NOD2 by MDP (i.v.) does not affect monocyte precursors, thus supporting the intrinsic activity of MDP on monocyte differentiation. The effects of NOD2 triggering were also observed on primary human monocytes. Isolated human blood monocyte subsets were identified by cytometry on the basis of their expression of CD14 and CD16 markers and gated as P1, CD142+ CD16 (classical); P2, CD14+ CD16+ (intermediate); and P3, CD14± CD162+ (non-classical) (Figure 3E). Freshly isolated monocytes are composed mainly of more than 80% classical monocytes, 1%– 3% intermediate monocytes, and 5%–7% non-classical monocytes. The nature of non-classical monocytes was confirmed by phenotypic profiling with other surface markers. Indeed, CD43 and CX3CR1 expression is found in CD14± CD162+ monocytes (P3) in both untreated and treated samples. Substantial expression of CD1d was detected only on CD142+ CD16 (P1) and on CD142+ CD16+ (P2) monocytes (data not shown). Although in vitro cultures of monocytes seem to promote the fre-

quencies of intermediate monocytes, the percentages of nonclassical monocytes were not affected. On the other hand, after 24 hr of in vitro treatment with MDP, we found that percentages of non-classical monocytes (P3) increased to approximately 15% (Figure 3F), indicating that CD142+ CD16 monocytes switch into CD14± CD162+ monocytes when stimulated with MDP. The increase in intermediate monocytes following in vitro treatment may reflect the transition into the CD14± CD162+ phenotype. Characterization of Surface Antigen Expression and Biological Functions of Differentiated Ly6Clow Monocytes following MDP Treatment In addition to Ly6C and the chemokine receptor CX3CR1, the patrolling monocyte subset can be distinguished from inflammatory monocytes by expression of various antigens and by specific biological functions. First, we determined the phenotypic features of isolated Ly6Clow monocytes from MDP-treated (i.v.) Nr4a1/ mice and of converted monocytes isolated from naive WT mice pretreated with clodronate before MDP (i.v.) administration. In these clodronate-treated mice, only Ly6Chigh monocytes were restored in the circulation at 48 hr post-administration of clodronate. Flow cytometry analysis revealed that MDP-induced Ly6Clow monocytes isolated from clodronatetreated mice as well as Nr4a1/ mice express high levels of LFA1. Both of these Ly6Clow cells display typical marker expression characterizing patrolling monocytes, such as negative or low expression of CD62L, 7/4, and CD49b and positive expression of CD11c (Table S1). Surprisingly, expression of CX3CR1 and CCR2 on converted Ly6Clow monocytes was not modulated by MDP treatment in Nr4a1/ mice. In contrast to Ly6Chigh inflammatory monocytes, one major feature of Ly6Clow monocytes is their capacity of crawling on the lumina side of the blood vessels in steady-state conditions. To determine whether MDP-induced Ly6Clow monocytes can patrol the endothelium of the vessels, we undertook intravital microscopy of ear capillaries in Cx3cr1+/gfp mice. This mouse strain was used to facilitate tracking of patrolling monocytes

(B) Flow cytometry analysis of blood Ly6Chigh, Ly6Cinter, and Ly6Clow monocyte subsets in naive, PBS-lipo-, or Clo-lipo-injected mice following vehicle (saline) or MDP treatment. Animals were sacrificed at indicated times post-treatment, and absolute counts of re-emerging monocyte subsets were determined by flow cytometry. (C) Absolute values of total, Ly6Chigh, and Ly6Clow blood monocytes in clodronate (Clo-lipo)-injected mice treated daily with vehicle or MDP. Animals were sacrificed at indicated times following Clo-lipo administration. In naive mice, absolute count of total monocytes was 191,259.44 ± 10,125.5, Ly6Chigh monocytes was 106,419.01 ± 7,524.75, and Ly6Clow monocytes was 64,090.78 ± 4,525.83. (D) Experimental design of DiO-labeled liposomes and MDP treatment. DiO-labeled liposomes (DiO-lipo) or PBS-liposomes (control) were injected intravenously in wild-type mice 48 hr following Clo-lipo administration in order to label newly synthesized Ly6Chigh monocytes. Mice were treated daily with vehicle or MDP (i.v.) and sacrificed at indicated times. (E) Flow cytometry of Ly6Chigh, Ly6Cinter, and Ly6Clow monocyte subsets at indicated times following vehicle or MDP treatments after gating on all monocytes (top) or DiO-labeled monocytes (bottom). Data are presented as mean ± SEM of three independent experiments. (F) Flow cytometry analysis of CD45.1+ bone marrow-derived monocyte subsets before proceeding to engraftment. Grafted CD45.1 bone marrow monocytes were mostly Ly6Chigh cells (94.8%). (G) Flow cytometry analysis of CD45.1+ and CD45.2+ cells in blood of CD45.2 mice adoptively transferred with CD45.1 BM cells. (H) Total bone marrow cells (35 3 106 cells) were adoptively transferred (i.v.) into CD45.2 recipient mice. Animals were treated with vehicle or MDP (i.v.) 2 hr after adoptive transfer. Blood samples were collected at indicated time and analyzed by flow cytometry for the levels of Ly6Clow and Ly6Chigh CD45.1+ monocytes. Data are representative of three independent experiments with one mouse/experiment. *p % 0.05, **p % 0.01, and ***p % 0.001 (two-way ANOVA), MDP-treated compared with vehicle-treated animals. (I) Sorted Ly6Chigh CD45+ bone marrow-derived monocytes (2.5 3 106 cells) were adoptively transferred into CD45.2 recipient mice. Animals were treated with MDP (i.v.), and blood was collected 1 hr following treatment for flow cytometry analysis. Data are representative of two experiments performed with one mouse/ experiment.

1834 Cell Reports 20, 1830–1843, August 22, 2017

Figure 3. MDP Treatment Increases Ly6Clow Monocytes in Bone Marrow and Spleen of Mice (A and C) Flow cytometry gating strategies of total, Ly6Chigh, and Ly6Clow monocytes in (A) bone marrow and (C) spleens of wild-type mice. Bone marrow monocytes are gated as Live+LinCD117CD115+CD11b+ and spleen monocytes as Live+LinCD11b+CD11c. Monocyte subsets were analyzed according to their Ly6C expression in vehicle (control) or MDP (i.v.) treated mice (n = 5) and are presented as absolute number ± SEM of two independent experiments. (B and D) Frequencies of Ly6Clow and Ly6Chigh monocytes are shown in (B) bone marrow cells and (D) spleen cells of mice treated with vehicle or MDP (i.v.) at indicated times. (E) Gating strategies of CD142+ CD16 (P1, blue), CD14+ CD16+ (P2, red), and CD14± CD162+ (P3, green) monocyte subsets in freshly isolated, non-stimulated (cell cultures), or MDP-treated human mononuclear cells for 18 hr. (F) Frequencies (%) of blood CD142+ CD16 (P1) and CD14± CD162+ (P3) monocyte subsets from freshly isolated, non-stimulated, or MDP treated human mononuclear cells for 18 hr. Dot plots show the distribution and mean ± SEM (horizontal lines) of monocytes subset frequencies of three independent experiments. Each symbol represents an independent individual. *p % 0.05, **p % 0.01, ***p % 0.001, and ****p % 0.0001 (unpaired t test).

via intravital microscopy (Auffray et al., 2007; Michaud et al., 2013). When indicated, mice were injected with clodronate to completely eliminate monocytes prior to treatment with the NOD2 agonist MDP. As we mentioned above, patrolling monocytes start to re-emerge in circulation only by day 6 post-clodronate injection, whereas Ly6Chigh monocytes appear by day 2. We thus assume that the presence of Ly6Clow patrolling monocytes detected in blood by days 2–3 should arise from Ly6Chigh. Indeed, intravital observation revealed that following treatment with MDP (i.v.), a significant number of crawling cells expressing the GFP proteins is detected compared with the control groups

treated with a vehicle (Figures 4A and 4B; Movies S1, S2, S3, and S4). The average number of crawling patrolling cells under steady-state conditions was 4 cells/hour/mm2. In MDP-treated mice, this number increased to 88 cells/hour/mm2. At 18 and 96 hr post-clodronate administration, no patrolling cell was observed. In contrast, when these mice were treated with MDP, we counted 33 cells/hour/mm2, and flow cytometry analysis also confirmed that these crawling cells displayed a patrolling phenotype (Figures 4C and 4D). Monocyte subsets have distinct homing properties under inflammatory conditions (Geissmann et al., 2003). In the early

Cell Reports 20, 1830–1843, August 22, 2017 1835

Figure 4. MDP Treatment Induces Crawling of Differentiated Ly6Chigh Monocytes (A and B) Intravital imaging of monocytes in the dermis (ear) of Cx3cr1+/gfp mice treated with (A) vehicle or MDP (i.v., 72 hr post-treatment) and (B) Cx3cr1+/gfp mice injected with clodronate-liposomes (Clo-lipo) 24 hr prior to vehicle or MDP treatments. Images were recorded at indicated time following Clo-lipo administration. The scale bar represents 20 mm. See also Movies S1, S2, S3, and S4. (C and D) Flow cytometry analysis of Ly6Chigh, Ly6Cinter, and Ly6Clow blood monocytes from corresponding Cx3cr1+/gfp in vivo imaged mice. Data are presented as absolute numbers of monocyte subsets.

phase of inflammation, Ly6Chigh monocytes invade the inflamed tissues, but Ly6Clow monocytes largely remain in blood. We thus compared the recruitment of converted Ly6Clow monocytes following MDP treatment using thioglycollate (TG)-induced intraperitoneal (i.p.) inflammation model (Chen et al., 2008). Cx3cr1+/gfp mice were treated daily with MDP (i.v.) or with placebo for 2 days prior to i.p. administration of TG. As expected, a larger proportion of Ly6Clow monocytes was detected in blood of MDP-treated mice early after TG administration (1.5 and 5 hr) compared with the placebo group (Figure 5), many of them were found to roll along the endothelial surface (Figure 5A; Movies S5, S6, S7, S8, and S9). Indeed, the number of crawling patrolling monocytes was evaluated at 38 and 18 cells/hr/mm2 after 1.5 and 5 hr, respectively, post-TG injection. The proportion of crawling cells was found to increase at 106 and 202 cells/hour/ mm2, respectively, in mice pretreated with MDP. The migration of Ly6Clow monocytes in the inflamed peritoneum was also found to be more rapid after treatment with MDP, while the number of Ly6Chigh inflammatory monocytes was found to decrease compared with the placebo groups (Figure 5B). To further discriminate between Ly6Clow and Ly6Chigh monocyte subsets, we injected mice with a CCR2-specific antibody before intravital microscopy. We observed that the majority of crawling cells were GFP+, and most CCR2+ cells circulated rapidly inside the microvessels (Figure 5C). We also compared the cytokine profile in the peritoneal cavity of mice treated with MDP or with a vehicle. First, we did not detect any IL-10, TGF-b1, or TNF-a production in the peritoneum of mice treated i.v. with MDP alone.

1836 Cell Reports 20, 1830–1843, August 22, 2017

IL-6 levels were slightly detected in peritoneal exudates although not significant compared with vehicle-treated mice. In the early period post-TG administration, the concentrations of IL-6 in peritoneal exudates were significantly high in MDP-treated mice to gradually decrease thereafter. Concentration of TNF-a was significantly reduced at 10 hr post-TG administration. On the other hand, synthesis of suppressive IL-10 was found to be rapidly produced as well as TGF-b1, which was found to reach maximum levels at 5 hr and to slightly increase with time (Figure 5D). The release of these cytokines was markedly suppressed in peritoneal cavities of NOD2/ mice, confirming that decreased inflammation is due to the MDP-NOD2 interactions (data not shown). NOD2 Triggering Contributes to Reduce Inflammation Because MDP treatment drives differentiation of Ly6Chigh monocytes into Ly6Clow monocytes, we wanted to investigate whether differentiated Ly6Clow monocytes may affect the progression of the inflammatory response in vivo. We induced a septic response to create an acute inflammation in mice by administration of lipopolysaccharide (LPS) and measured the presence of Ly6Clow and Ly6Chigh monocytes as well as the production of pro- and anti-inflammatory cytokines in blood and in spleen of mice following treatment with MDP (i.v.) or with a placebo (Semaeva et al., 2010). As expected, the number of Ly6Clow monocytes significantly increased both in blood and in spleen following administration of MDP to vehicle mice (Figures 6A and 6B). Furthermore, in mice injected with LPS, treatment

Figure 5. Homing of Converted Monocytes in MDP-Treated Mice following Thioglycollate-Induced Peritonitis (A) Intravital imaging of peritoneal microvessels of Cx3cr1+/gfp mice treated daily with vehicle (top) or MDP (i.v.) (bottom) prior to intraperitoneal injection of thioglycollate. Images were recorded at indicated times following thioglycollate administration. The scale bar represents 20 mm. See also Movies S5, S6, S7, S8, and S9. (B) Numbers of Ly6Clow and Ly6Chigh monocytes were assessed by flow cytometry in peritoneal lavages performed on in vivo imaged Cx3cr1+/gfp mice at indicated times following thioglycollate administration. (C) Intravital imaging of peritoneal microvessels of Cx3cr1+/gfp mice treated daily with vehicle or MDP (i.v.) prior to intraperitoneal injection of thioglycollate. Animals were injected with 2.5 mg (i.v.) of CCR2 PE antibody prior to imaging. Images were recorded at 5 hr following thioglycollate administration. The scale bar represents 20 mm.

(legend continued on next page)

Cell Reports 20, 1830–1843, August 22, 2017 1837

Figure 6. MDP Treatment Increases Levels of Ly6Clow Monocytes and Reduces Inflammation in LPS-Treated Mice (A and B) Flow cytometry analysis (A) and absolute counts (B) of Ly6Chigh, Ly6Cinter, and Ly6Clow monocytes in blood (left) and spleen (right) of wild-type mice treated daily with MDP (48 hr). Mice were injected with LPS (1 mg) 24 hr following the last vehicle or MDP injection. Animals were sacrificed 48 hr after LPS administration. (C) Levels of TNF-a and IL-6 were determined in sera (left) and spleens (right) of mice. Data are presented as mean ± SEM of two independent experiments (n = 4 mice/group). *p % 0.05 (Mann-Whitney test) compared with indicated groups.

with MDP was found to reduce the number of inflammatory Ly6Chigh monocytes in blood and in the spleen of mice but also to markedly increase the number of Ly6Clow monocytes. In addition, the production of inflammatory cytokines TNF-a and IL-6 induced following LPS administration was significantly reduced in blood and in spleen of mice treated with MDP (Figure 6C). Using a serum transfer-induced arthritis model (Kouskoff et al., 1996; Monach et al., 2008) to mimic chronic inflammation, we also compared the severity of joint inflammation in arthritic mice treated daily with MDP (i.v.) or with a placebo. The results obtained showed that MDP treatment significantly attenuated joint inflammation as early as day 6 post-treatment, and such reduction gradually occurred thereafter. This also correlates

with the increased proportions of Ly6Clow monocytes in blood of treated mice (Figures S6A and S6B). Together, these data indicate that the increased frequencies of Ly6Clow monocytes induced following NOD2 triggering in mice might contribute to reduce the inflammatory response. Gene Encoding Profile by NOD2-Activated Ly6Clow and Ly6Chigh Monocytes Next, we wanted to evaluate gene profile in both subsets of monocytes following MDP treatment. We performed differential expression analysis on inflammatory and patrolling monocytes before and after treatment with MDP. Among the 11,998 genes that have normalized expression >1 transcript per million

(D) Levels of IL-6, TNF-a, IL-10, and TGF-b1 were determined in peritoneal lavages of in vivo imaged mice at indicated times following thioglycollate administration. No significant levels of IL-6, TNF-a, IL-10, and TGF-b1 were detected in MDP treated mice (data not shown). Data are presented as mean ± SEM of two independent experiments (n = 3 mice/group). *p % 0.05, **p % 0.01, and ***p % 0.001 (two-way ANOVA followed by Bonferroni test in B and Mann-Whitney test in D).

1838 Cell Reports 20, 1830–1843, August 22, 2017

Figure 7. Impact of MDP Treatment on Gene Expression in Ly6Chigh and Ly6Clow Monocyte Subsets (A) Venn diagram illustrating the overlap of the differentially expressed genes (p < 0.05 after Benjamini-Hochberg correction for multiple testing) in Ly6Chigh and Ly6Clow blood monocytes of MDP-treated mice (18 hr). The number in each section of the diagram represents the number of genes modulated by MDP treatment. (B) Heatmap displaying the log2 fold change in expression of 51 genes regulated in monocyte subsets following MDP treatment. Key genes are presented in red. Data are representative of two or more independent experiments.

functions (Kittan et al., 2013; Warren et al., 2016; Yaddanapudi et al., 2013). DISCUSSION

(TPM), 1,115 were commonly modulated in both monocyte subsets (Figure 7A) (p < 0.05) after Benjamini-Hochberg correction for multiple testing. Five hundred eighty-four and 2,058 genes were differentially expressed in Ly6Chigh and Ly6Clow monocytes, respectively. Among these, the mRNA expression profile of 51 selected genes supports the capacity of MDP to regulate gene expression in Ly6Chigh and Ly6Clow monocytes (Figure 7B). MDP treatment induced strong activation of IL-6 gene in Ly6Chigh monocytes. Because exogenous IL-6 has been recognized to have the capacity to exert anti-inflammatory properties (Scheller et al., 2011; Tilg et al., 1994; Xing et al., 1998; Yasukawa et al., 2003), its release following MDP treatment might contribute to control inflammatory response. We also found that mRNA expression of CXCR3 and its ligand CXCL10 was strongly suppressed in Ly6Chigh monocytes after MDP treatment. IL-1b mRNA expression was also suppressed in Ly6Chigh monocytes along with Axl gene expression, which provides a negative feedback signal to reduce excessive immune response (Rothlin et al., 2007; Sharif et al., 2006). NR4A1 and C/EBPb, which have recently been showed to contribute to control NR4A1 expression (Mildner et al., 2017), were both activated in Ly6Chigh monocytes after MDP treatment. M2 associated factor IRF4 and chemokine CCL22 (Ono et al., 2009) were found to be activated in Ly6Clow monocytes, whereas CCL17 and the suppressor SOCS2 genes were highly expressed in both monocyte subsets. The gene coding for SiglecE, a negative regulator of acute inflammation, was activated in Ly6Clow monocytes following administration of MDP, while M2 signature genes such as Arg1, Alox15, and Tgm2 were upregulated in Ly6Chigh monocytes, suggesting that these cells might engage in a differentiation pathway favoring M2-like

The signals that promote the differentiation and the fate of blood monocytes are largely unknown. In this study, we have demonstrated that in vivo stimulation of NOD2 receptor with MDP induces the emergence of circulating Ly6Clow patrolling monocytes, which presented strong biological similarities with endogenous Ly6Clow monocytes. Indeed, following triggering of NOD2 with MDP, a large proportion of inflammatory Ly6Chigh monocytes acquire the expression of surface markers that typically characterize patrolling monocytes. Our intravital microscopy observations have clearly shown that treatment of Cx3cr1+/gfp mice with MDP increased the number of patrolling Ly6Clow monocytes compared with naive mice under steadystate conditions. The most convincing results came from the use of Nr4a1/ mice, which lack patrolling Ly6Clow monocytes (Hanna et al., 2011). Indeed, following administration of MDP, we were able to detect by flow cytometry a significant proportion of blood monocytes, which were found to express typical markers and functions of patrolling subsets. Furthermore, these results clearly support the notion that Ly6Chigh monocytes give rise to patrolling Ly6Clow monocytes following treatment with MDP and therefore suggest that switching of inflammatory monocytes into patrolling phenotype is not due to a massive exit of Ly6Clow monocytes or a signal from the BM. This was also supported in experiments using mice pretreated with clodronate before MDP administration in order to deplete monocytes in the circulation. We detected the presence of circulating Ly6Clow as soon as 72 hr post-MDP treatment, although this monocyte subset begins to repopulate the bloodstream only on the sixth day after clodronate injection. Adoptive transfer of monocytes revealed that Ly6Chigh monocytes change into Ly6Clow patrolling monocytes when triggered by MDP agonist. These results demonstrated that ‘‘transformed’’ Ly6Clow monocytes can acquire not only the phenotype of patrolling monocytes but also their capacity to patrol endothelium of blood vessels. The effects of NOD2 triggering were also observed with primary human monocytes,

Cell Reports 20, 1830–1843, August 22, 2017 1839

as we found that CD142+ CD16 monocytes gave rise to a significant increase in patrolling CD14± CD162+ monocytes following treatment with MDP. We found that not all agonists that bind to NOD2 have the capacity to convert inflammatory monocytes into patrolling monocytes. Indeed, in the case of murabutide, we did not observe any conversion of inflammatory Ly6Chigh monocytes into patrolling Ly6Clow monocytes. That could be explained by the presence in the molecular structure of an amide and an ester residue on the glutamine instead of the carboxyl group, which seems very important for activation of other NOD2 agonists (Girardin et al., 2003b). It is imperative to characterize the domains of NOD2 receptor interacting with MDP (and other agonists) to identify the structural requirements involved in the conversion of Ly6Chigh monocytes into patrolling subset. Our findings show that modulation of Ly6Chigh monocyte phenotype and function can be induced through the triggering of NOD2 receptor, highlighting the contribution of this receptor in monocyte plasticity. In a previous study, Askenase et al. (2015) demonstrated that Ly6Chigh monocytes can be remodeled for regulatory functions by signals dependent on IL-12 and IFNg before they exit the BM. Our results are in part similar to those reported by Askenase et al. (2015). Indeed, we demonstrated that under a different signal, in our case NOD2 triggering, the fate of Ly6Chigh monocytes can be reprogrammed upon their recruitment to the inflamed tissue. The mechanisms that lead NOD2 signals to the conversion of Ly6Chigh monocytes remain to be clarified, however. The mRNA analysis indicates that NR4A1 is significantly increased in Ly6Chigh monocytes but modestly affected in Ly6Clow monocytes after in vivo treatment with MDP. However, although NR4A1 probably contributes to the NOD2 cascade in Ly6Chigh monocytes, we believe that another signaling mechanism could also contribute to the regulation of this monocyte subset. Indeed, conversion of Ly6Chigh into Ly6Clow monocytes was also observed in NR4A1-deficient mice. This raises the question of whether another signaling cascade inherent to Ly6Chigh monocytes might exist. Recently, it was demonstrated that C/EBPb contributes to control expression of NR4A1 in Ly6Clow monocytes and that C/EBPb might act upstream of NR4A1 (Mildner et al., 2017). Because we observed that MDP treatment increases C/EBPb expression in monocyte subsets, it is possible that activation of C/EBPb following NOD2 triggering binds to other promoters that participate to differentiation of Ly6Chigh monocytes in absence of NR4A1. This potential mechanism is currently under study. The precise role of NOD2 in the regulation of the inflammatory response is still ambiguous. It is not clear whether NOD2 variantrelated inflammation is due to an impaired capacity to control bacterial clearance or to regulate excessive inflammatory response. In our studies, we were able to demonstrate that treatment with MDP reduces the inflammatory response in murine model of systemic inflammation and also in an arthritis mouse model of chronic inflammation. In both models, increased levels of circulating Ly6Clow patrolling monocytes were detected, suggesting that the conversion of Ly6Chigh into Ly6Clow monocytes following NOD2 triggering might also contribute to control excessive inflammation. We suspect that Ly6Clow monocytes may differentiate in M2-like macrophages following their recruitment in the inflamed tissue in response to MDP treatment. This

1840 Cell Reports 20, 1830–1843, August 22, 2017

hypothesis is in line with our genomic analyses, indicating that after MDP treatment, Ly6Chigh monocytes show an upregulation of typical genes associated with an anti-inflammatory signature. Despite the role of NOD2 in host defense against microorganisms, this receptor appears to also regulate distinct cellular functions possibly by a transcriptional regulation of key innate immune genes. We thus propose that activation of NOD2 might lead to two distinct responses. First, in an ‘‘immediate’’ response following recognition of bacterial fragments or of viral RNAs, a rapid production of inflammatory factors is initiated to counter infection. Second, following sustained triggering of NOD2 by agonist(s), a ‘‘late’’ and distinct response is induced, leading to the emergence of patrolling monocytes, a mechanism that could contribute to regulate homeostasis in inflamed tissues. These sequential responses mediated by NOD2 need to be further investigated to better understand the role of NOD2 in inflammatory diseases. EXPERIMENTAL PROCEDURES Mice C57BL/6 (WT) mice were purchased from Charles River Laboratories and NOD2-deficient mice (Nod2/) from Jackson Laboratory. Mice deficient for NR4A1 (Nr4a1/) were kindly provided by Dr. Claude Rouillard (CHU de Que´bec Research Center, Universite´ Laval), and CCR2 (Ccr2/), Cx3cr1+/gfp, and GFP transgenic mice (Gfp+) were kindly provided by Dr. Serge Rivest (CHU de Que´bec Research Center, Universite´ Laval). All mice were bred in our animal facilities to generate control animal and littermates. Mice were 4–10 weeks old except as indicated. Experiments were conducted in accordance with the guidelines of animal research ethics boards of Laval University (approval number 15-109-2). Mouse Treatments N-acetyl-MDP (MDP; InvivoGen) was diluted in 0.9% saline and i.v. injected at 10 mg/kg except as indicated. Treatment was administrated daily and mice were sacrificed 48 hr following treatments, unless otherwise indicated. N-glycolyl-MDP, L-18 MDP, and murabutide (InvivoGen) were diluted in 0.9% saline and i.v. injected at 1.25, 2.5, 5, and 10 mg/kg. Mice were sacrificed 48 hr after drug administration. Control mice were injected with 0.9% saline for all experiments. In Vivo Depletion of Murine Blood Monocyte Subsets Blood mononuclear phagocyte depletion was achieved by i.v. administration of 0.1 mL of dichloromethylene-bisphononate (clodronate)-loaded liposomes (Clo-lipo; clodronate liposomes). PBS-loaded liposomes (PBS-lipo) were used as a negative control. Monocyte depletion was monitored by flow cytometry. 3,30 -Dioctadecyloxacarbocyanine perchlorate (DiO; Molecular Probes) lipophilic tracer was incorporated into PBS-liposomes at a concentration of 25 mg/mL for in vivo labeling of mononuclear phagocytes following the manufacturer’s instructions. Isolation of Human Mononuclear Cells and Treatment with MDP Plasma from healthy donors was isolated from peripheral blood by centrifugation. Mononuclear cells were isolated using Ficoll density gradient (Wisent) as previously reported (Lacerte et al., 2016) and treated with MDP (10 mg/mL) overnight at 37 C. Cells were stained for flow cytometry as described in Cell Isolation and Flow Cytometry Analysis. Classical, intermediate, and non-classical monocyte subsets were identified by flow cytometry as follows: CD142+ CD16 (classical) (P1); CD14+ CD16+ (intermediate) (P2); and CD14± CD162+ (non-classical) (P3). Cell Isolation and Flow Cytometry Analysis Blood samples were collected by cardiac puncture using Hank’s balanced salt solution (HBSS)-EDTA-coated syringes. BM cells were collected from both

femur and tibia of mice by flushing the BM with Dulbecco’s PBS (DPBS) (Wisent) supplemented with 10% FBS (Wisent). Harvested cells were filtered through a 40 mm cell strainer and washed with DPBS. Spleen cells were collected by mechanical disruption and digestion with 180 U/mL DNase I (Sigma-Aldrich) and 1 mg/ml collagenase (Sigma-Aldrich). Cells were passed through a 40 mm cell strainer and washed with HBSS-EDTA. Erythrocytes from all samples were lysed with red blood cell (RBC) lysis buffer. Peritoneal exudate cells (PECs) were isolated by centrifugation of peritoneal lavages obtained by injection (2 3 2 mL) of DPBS supplemented with 3% FBS into the peritoneal cavity. For details on flow cytometry experiments, see Supplemental Experimental Procedures. Two-Photon Intravital Microscopy Imaging Blood vessels were labeled with Qdot 750 (Qtracker 705, 5% in PBS; Invitrogen) administered through the tail vein. Animals were anesthetized with isoflurane and placed under a two-photon microscope using a special support. For intravital microscopy of the ear, skin was fixed on slide and covered with 0.9% saline. For peritoneal intravital microscopy, a small midline abdominal incision was made to expose the parietal peritoneum, and tissue was overlaid with a slide covered with 0.9% saline. Intravital microscopy was carried out using an MPE two-photon microscope. The two-photon Mai Tai DeepSee laser (Spectra-Physics) was tuned to 900 nm for all the experiments. Tissues were imaged using an Olympus Ultra 25 3 MPE water immersion objective (1.05 NA), with filter set bandwidths optimized for GFP (520–560 nm) and Qdot 705 (669–800 nm) imaging. Detector sensitivity and gain were set to achieve the optimal dynamic range of detection. Using Olympus Fluoview software (version 3.0a), images with resolution of 512 3 512 pixels were acquired for 15 min at 1.53 zoom factor and 2.5 frames/s. No Kalman filter was used, to avoid slowing down the acquisition speed. Adoptive Transfer of Monocytes Monocytes and sorted Ly6Chigh monocytes from the BM of CD45.1 WT mice were injected (i.v.) into congenic CD45.2 WT mice as previously detailed (Brunet et al., 2016). Mice were treated with MDP and levels of blood monocyte subsets measured at indicated times after engraftment. Library Preparation and Sequencing by RNA Sequencing Blood Ly6Chigh and Ly6Clow monocytes from vehicle or MDP-treated (18 hr) WT mice were enriched by cell sorting. Total RNA from sorted Ly6Chigh and Ly6Clow monocytes was isolated using EZ-10 DNAaway RNA miniprep (Bio Basic) according to the manufacturer’s protocol. For more details on RNA sequencing (RNaseq) experiments, see Supplemental Experimental Procedures. Bioinformatic Analyses of RNaseq All bioinformatics analyses were performed on RNA isolated from monocyte subsets and executed on Compute Canada clusters for RNaseq. See Supplemental Experimental Procedures for details. Murine Models of Inflammation Peritonitis was induced by i.p. injection of 1 mL of 3% sterile TG (Thermo Fisher Scientific). Briefly, mice were daily treated with vehicle or MDP for 48 hr. TG was injected 24 hr following the last MDP injection. Peritoneal exudates (PECs) were collected at indicated times following TG injection. Systemic inflammation was induced through administration of LPSs (LPS from E. coli; InvivoGen). Briefly, mice were treated daily with MDP for 48 hr, and LPS (1 mg diluted in 0.9% saline) was i.v. injected 24 hr following the last injection of MDP. Peripheral blood and spleen were collected 48 hr following LPS treatment. When indicated, levels of IL-6, IL-10, and TNF-a were determined in sera, peritoneal lavages, and spleen homogenates using the BD cytometric bead assay system (CBA flex set; BD Bioscience). Levels of TGF-b1 were assessed by ELISA (R&D Systems) in peritoneal lavages. Transient arthritis was induced by injection of arthritogenic serum from K/B 3 N mice at day 0 and day 2 (Monach et al., 2008). Mice were randomized on the basis of their arthritic scores and then treated daily with MDP. Arthritic scores were evaluated on the basis of visual evaluation as previously described (Brunet et al., 2016). Severity of arthritis in mice was monitored by evaluating the ankle

thickness of all limbs. Disease severity was recorded for each limb on a scale of 0–4 as follows: 0, no inflammation; 1, erythema or mild swelling of the tarsal or the ankle; 2, erythema and mild swelling extending from the ankle to the tarsal; 3, erythema and moderate swelling extending from the ankle to the metatarsal joint; and 4, erythema and severe swelling of the ankle, paw, and digits. Arthritic scores were assessed by two independent observers. The mean arthritic score represent the average score for mice of each group. Blood of mice was collected at indicated times for flow cytometry analysis. Statistical Analysis Statistical analyses were carried out with Prism version 6.0 (GraphPad Software) and were performed using unpaired t tests except as indicated. The level of significance was set at p % 0.05. ACCESSION NUMBER The accession number for the RNaseq data reported in this paper is GEO: GSE101496. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, six figures, one table, and nine movies and can be found with this article online at http://dx.doi.org/10.1016/j.celrep.2017.08.009. AUTHOR CONTRIBUTIONS A.-J.L. and M.L. carried out the experiments. P.T. conducted specific experiments. B.E. and A.B. conducted experiments and contributed to analysis of the data. P.P. conducted intravital microscopy experiments. A.D. carried out RNA sequencing and bioinformatics analyses. S.R. helped to conceive the study, analyzed the data, and revised the manuscript. J.G. conceived the study, designed experiments, interpreted data, and wrote the manuscript. ACKNOWLEDGMENTS ^ te´ for secretarial assistance, Lisa Auclert and Dominic We thank Pierrette Co Bastien for technical assistance, and Charles Joly Beauparlant for his contribution in bioinformatic analysis. This work was supported by grants from the Canadian Institutes of Health Research (CIHR) to J.G. (MOP-123440 and MOP142304). S.R. is supported by a Canadian Research Chair in Neuroimmunology (950-231099). Received: February 14, 2017 Revised: May 16, 2017 Accepted: July 28, 2017 Published: August 22, 2017 REFERENCES Askenase, M.H., Han, S.J., Byrd, A.L., Morais da Fonseca, D., Bouladoux, N., Wilhelm, C., Konkel, J.E., Hand, T.W., Lacerda-Queiroz, N., Su, X.Z., et al. (2015). Bone-marrow-resident NK cells prime monocytes for regulatory function during infection. Immunity 42, 1130–1142. Auffray, C., Fogg, D., Garfa, M., Elain, G., Join-Lambert, O., Kayal, S., Sarnacki, S., Cumano, A., Lauvau, G., and Geissmann, F. (2007). Monitoring of blood vessels and tissues by a population of monocytes with patrolling behavior. Science 317, 666–670. Auffray, C., Sieweke, M.H., and Geissmann, F. (2009). Blood monocytes: development, heterogeneity, and relationship with dendritic cells. Annu. Rev. Immunol. 27, 669–692. Brunet, A., LeBel, M., Egarnes, B., Paquet-Bouchard, C., Lessard, A.J., Brown, J.P., and Gosselin, J. (2016). NR4A1-dependent Ly6C(low) monocytes contribute to reducing joint inflammation in arthritic mice through Treg cells. Eur. J. Immunol. 46, 2789–2800.

Cell Reports 20, 1830–1843, August 22, 2017 1841

Caso, F., Costa, L., Rigante, D., Vitale, A., Cimaz, R., Lucherini, O.M., Sfriso, P., Verrecchia, E., Tognon, S., Bascherini, V., et al. (2014). Caveats and truths in genetic, clinical, autoimmune and autoinflammatory issues in Blau syndrome and early onset sarcoidosis. Autoimmun. Rev. 13, 1220–1229. Chen, D., Carpenter, A., Abrahams, J., Chambers, R.C., Lechler, R.I., McVey, J.H., and Dorling, A. (2008). Protease-activated receptor 1 activation is necessary for monocyte chemoattractant protein 1-dependent leukocyte recruitment in vivo. J. Exp. Med. 205, 1739–1746. Coulombe, F., Fiola, S., Akira, S., Cormier, Y., and Gosselin, J. (2012). Muramyl dipeptide induces NOD2-dependent Ly6C(high) monocyte recruitment to the lungs and protects against influenza virus infection. PLoS ONE 7, e36734. Ermann, J., Staton, T., Glickman, J.N., de Waal Malefyt, R., and Glimcher, L.H. (2014). Nod/Ripk2 signaling in dendritic cells activates IL-17A-secreting innate lymphoid cells and drives colitis in T-bet-/-.Rag2-/- (TRUC) mice. Proc. Natl. Acad. Sci. U S A 111, E2559–E2566. Feinen, B., Petrovsky, N., Verma, A., and Merkel, T.J. (2014). Advaxadjuvanted recombinant protective antigen provides protection against inhalational anthrax that is further enhanced by addition of murabutide adjuvant. Clin. Vaccine Immunol. 21, 580–586. Geissmann, F., Jung, S., and Littman, D.R. (2003). Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity 19, 71–82. Girardin, S.E., Boneca, I.G., Viala, J., Chamaillard, M., Labigne, A., Thomas, G., Philpott, D.J., and Sansonetti, P.J. (2003a). Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J. Biol. Chem. 278, 8869–8872. Girardin, S.E., Travassos, L.H., Herve´, M., Blanot, D., Boneca, I.G., Philpott, D.J., Sansonetti, P.J., and Mengin-Lecreulx, D. (2003b). Peptidoglycan molecular requirements allowing detection by Nod1 and Nod2. J. Biol. Chem. 278, 41702–41708. Grimes, C.L., Ariyananda, Lde.Z., Melnyk, J.E., and O’Shea, E.K. (2012). The innate immune protein Nod2 binds directly to MDP, a bacterial cell wall fragment. J. Am. Chem. Soc. 134, 13535–13537. Hanna, R.N., Carlin, L.M., Hubbeling, H.G., Nackiewicz, D., Green, A.M., Punt, J.A., Geissmann, F., and Hedrick, C.C. (2011). The transcription factor NR4A1 (Nur77) controls bone marrow differentiation and the survival of Ly6C- monocytes. Nat. Immunol. 12, 778–785. Hugot, J.P., Chamaillard, M., Zouali, H., Lesage, S., Ce´zard, J.P., Belaiche, J., Almer, S., Tysk, C., O’Morain, C.A., Gassull, M., et al. (2001). Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease. Nature 411, 599–603.

(2013). Cytokine induced phenotypic and epigenetic signatures are key to establishing specific macrophage phenotypes. PLoS ONE 8, e78045. Kouskoff, V., Korganow, A.-S., Duchatelle, V., Degott, C., Benoist, C., and Mathis, D. (1996). Organ-specific disease provoked by systemic autoimmunity. Cell 87, 811–822. ^ne, B., Brown, J.P., and Gosselin, J. Lacerte, P., Brunet, A., Egarnes, B., Duche (2016). Overexpression of TLR2 and TLR9 on monocyte subsets of active rheumatoid arthritis patients contributes to enhance responsiveness to TLR agonists. Arthritis Res. Ther. 18, 10. Macho Fernandez, E., Valenti, V., Rockel, C., Hermann, C., Pot, B., Boneca, I.G., and Grangette, C. (2011). Anti-inflammatory capacity of selected lactobacilli in experimental colitis is driven by NOD2-mediated recognition of a specific peptidoglycan-derived muropeptide. Gut 60, 1050–1059. Michaud, J.P., Bellavance, M.A., Pre´fontaine, P., and Rivest, S. (2013). Realtime in vivo imaging reveals the ability of monocytes to clear vascular amyloid beta. Cell Rep. 5, 646–653. Mildner, A., Schonheit, J., Giladi, A., David, E., Lara-Astiaso, D., LorenzoVivas, E., Paul, F., Chappell-Maor, L., Priller, J., Leutz, A., et al. (2017). Genomic characterization of murine monocytes reveals C/EBPbeta transcription factor dependence of Ly6C- cells. Immunity 46, 849–862.e7. Monach, P.A., Mathis, D., and Benoist, C. (2008). The K/BxN arthritis model. Curr. Protoc. Immunol. Chapter 15, Unit 15.22. Ogura, Y., Inohara, N., Benito, A., Chen, F.F., Yamaoka, S., and Nunez, G. (2001). Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes and activates NF-kappaB. J. Biol. Chem. 276, 4812–4818. Ono, K., Condron, M.M., and Teplow, D.B. (2009). Structure-neurotoxicity relationships of amyloid beta-protein oligomers. Proc. Natl. Acad. Sci. U S A 106, 14745–14750. Pandey, A.K., Yang, Y., Jiang, Z., Fortune, S.M., Coulombe, F., Behr, M.A., Fitzgerald, K.A., Sassetti, C.M., and Kelliher, M.A. (2009). NOD2, RIP2 and IRF5 play a critical role in the type I interferon response to Mycobacterium tuberculosis. PLoS Pathog. 5, e1000500. Philpott, D.J., Sorbara, M.T., Robertson, S.J., Croitoru, K., and Girardin, S.E. (2014). NOD proteins: regulators of inflammation in health and disease. Nat. Rev. Immunol. 14, 9–23. Rothlin, C.V., Ghosh, S., Zuniga, E.I., Oldstone, M.B., and Lemke, G. (2007). TAM receptors are pleiotropic inhibitors of the innate immune response. Cell 131, 1124–1136. Sabbah, A., Chang, T.H., Harnack, R., Frohlich, V., Tominaga, K., Dube, P.H., Xiang, Y., and Bose, S. (2009). Activation of innate immune antiviral responses by Nod2. Nat. Immunol. 10, 1073–1080.

Huo, Y., Weber, C., Forlow, S.B., Sperandio, M., Thatte, J., Mack, M., Jung, S., Littman, D.R., and Ley, K. (2001). The chemokine KC, but not monocyte chemoattractant protein-1, triggers monocyte arrest on early atherosclerotic endothelium. J. Clin. Invest. 108, 1307–1314.

Scheller, J., Chalaris, A., Schmidt-Arras, D., and Rose-John, S. (2011). The pro- and anti-inflammatory properties of the cytokine interleukin-6. Biochim. Biophys. Acta 1813, 878–888.

Inohara, N., Koseki, T., Lin, J., del Peso, L., Lucas, P.C., Chen, F.F., Ogura, Y., and Nu´n˜ez, G. (2000). An induced proximity model for NF-kappa B activation in the Nod1/RICK and RIP signaling pathways. J. Biol. Chem. 275, 27823–27831.

Semaeva, E., Tenstad, O., Skavland, J., Enger, M., Iversen, P.O., Gjertsen, B.T., and Wiig, H. (2010). Access to the spleen microenvironment through lymph shows local cytokine production, increased cell flux, and altered signaling of immune cells during lipopolysaccharide-induced acute inflammation. J. Immunol. 184, 4547–4556.

Inohara, N., Ogura, Y., Fontalba, A., Gutierrez, O., Pons, F., Crespo, J., Fukase, K., Inamura, S., Kusumoto, S., Hashimoto, M., et al. (2003). Host recognition of bacterial muramyl dipeptide mediated through NOD2. Implications for Crohn’s disease. J. Biol. Chem. 278, 5509–5512. Kanazawa, N., Okafuji, I., Kambe, N., Nishikomori, R., Nakata-Hizume, M., Nagai, S., Fuji, A., Yuasa, T., Manki, A., Sakurai, Y., et al. (2005). Early-onset sarcoidosis and CARD15 mutations with constitutive nuclear factor-kappaB activation: common genetic etiology with Blau syndrome. Blood 105, 1195– 1197. Kim, Y.G., Kamada, N., Shaw, M.H., Warner, N., Chen, G.Y., Franchi, L., and Nu´n˜ez, G. (2011). The Nod2 sensor promotes intestinal pathogen eradication via the chemokine CCL2-dependent recruitment of inflammatory monocytes. Immunity 34, 769–780. Kittan, N.A., Allen, R.M., Dhaliwal, A., Cavassani, K.A., Schaller, M., Gallagher, K.A., Carson, W.F., 4th, Mukherjee, S., Grembecka, J., Cierpicki, T., et al.

1842 Cell Reports 20, 1830–1843, August 22, 2017

Serbina, N.V., and Pamer, E.G. (2006). Monocyte emigration from bone marrow during bacterial infection requires signals mediated by chemokine receptor CCR2. Nat. Immunol. 7, 311–317. Sharif, M.N., Sosic, D., Rothlin, C.V., Kelly, E., Lemke, G., Olson, E.N., and Ivashkiv, L.B. (2006). Twist mediates suppression of inflammation by type I IFNs and Axl. J. Exp. Med. 203, 1891–1901. Sunderko¨tter, C., Nikolic, T., Dillon, M.J., Van Rooijen, N., Stehling, M., Drevets, D.A., and Leenen, P.J. (2004). Subpopulations of mouse blood monocytes differ in maturation stage and inflammatory response. J. Immunol. 172, 4410–4417. Swirski, F.K., Nahrendorf, M., Etzrodt, M., Wildgruber, M., Cortez-Retamozo, V., Panizzi, P., Figueiredo, J.L., Kohler, R.H., Chudnovskiy, A., Waterman, P., et al. (2009). Identification of splenic reservoir monocytes and their deployment to inflammatory sites. Science 325, 612–616.

Tilg, H., Trehu, E., Atkins, M.B., Dinarello, C.A., and Mier, J.W. (1994). Interleukin-6 (IL-6) as an anti-inflammatory cytokine: induction of circulating IL-1 receptor antagonist and soluble tumor necrosis factor receptor p55. Blood 83, 113–118. van Furth, R., and Cohn, Z.A. (1968). The origin and kinetics of mononuclear phagocytes. J. Exp. Med. 128, 415–435. Van Rooijen, N., and Sanders, A. (1994). Liposome mediated depletion of macrophages: mechanism of action, preparation of liposomes and applications. J. Immunol. Methods 174, 83–93. Warren, K.J., Fang, X., Gowda, N.M., Thompson, J.J., and Heller, N.M. (2016). The TORC1-activated proteins, P70S6K and GRB10, regulate IL-4 signaling and M2 macrophage polarization by modulating phosphorylation of insulin receptor substrate-2. J. Biol. Chem. 291, 24922–24930. Watanabe, T., Kitani, A., Murray, P.J., Wakatsuki, Y., Fuss, I.J., and Strober, W. (2006). Nucleotide binding oligomerization domain 2 deficiency leads to dysregulated TLR2 signaling and induction of antigen-specific colitis. Immunity 25, 473–485.

Xing, Z., Gauldie, J., Cox, G., Baumann, H., Jordana, M., Lei, X.F., and Achong, M.K. (1998). IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses. J. Clin. Invest. 101, 311–320. Yaddanapudi, K., Putty, K., Rendon, B.E., Lamont, G.J., Faughn, J.D., Satoskar, A., Lasnik, A., Eaton, J.W., and Mitchell, R.A. (2013). Control of tumorassociated macrophage alternative activation by macrophage migration inhibitory factor. J. Immunol. 190, 2984–2993. Yasukawa, H., Ohishi, M., Mori, H., Murakami, M., Chinen, T., Aki, D., Hanada, T., Takeda, K., Akira, S., Hoshijima, M., et al. (2003). IL-6 induces an anti-inflammatory response in the absence of SOCS3 in macrophages. Nat. Immunol. 4, 551–556. Zigmond, E., Varol, C., Farache, J., Elmaliah, E., Satpathy, A.T., Friedlander, G., Mack, M., Shpigel, N., Boneca, I.G., Murphy, K.M., et al. (2012). Ly6C hi monocytes in the inflamed colon give rise to proinflammatory effector cells and migratory antigen-presenting cells. Immunity 37, 1076–1090.

Cell Reports 20, 1830–1843, August 22, 2017 1843

Cell Reports, Volume 20

Supplemental Information

Triggering of NOD2 Receptor Converts Inflammatory Ly6Chigh into Ly6Clow Monocytes with Patrolling Properties Anne-Julie Lessard, Manon LeBel, Benoit Egarnes, Paul Préfontaine, Thériault, Arnaud Droit, Alexandre Brunet, Serge Rivest, and Jean Gosselin

Peter

Figure S1, Related to figure 1. Flow cytometry gating strategy Representative gating strategy for CD11b+ CD115+ monocytes and Ly6C monocyte subsets from WT mice as measured by flow cytometry. First, 123count eBeads™ are gated to determine absolute counting of cell population. After excluding the bead population, doublet discrimination is performed with a singlet gate (FSC-H/FSC-A dot blot). Next, neutrophils are identified as CD45+/CD11b+/Ly6G+ cells. After gating out the neutrophil cell population, monocytes were identified with CD45, CD11b and CD115 expression. Monocyte subsets were further subdivided in three populations based on the expression of Ly6C: Ly6Chigh, Ly6Cint and Ly6Clow correspond respectively to inflammatory, intermediate and patrolling monocytes.

***

20

0

40

20

0 0

1.25

2.5

5.0

10.0

Murabutide

*

60

Ly6Chigh monocytes (%)

40

L18-MDP

***

60

Ly6Chigh monocytes (%)

60

Ly6Chigh monocytes (%)

N-Glycolyl-MDP

60

Ly6Chigh monocytes (%)

N-Acetyl-MDP

40

20

0 0

Dose (mg/kg)

1.25

2.5

5.0

10.0

40

20

0 0

Dose (mg/kg)

1.25

2.5

5.0

10.0

0

Dose (mg/kg)

1.5

2.5

5.0

10.0

Dose (mg/kg)

***

20

10

0

30

20

10

0 1.25

2.5

5.0

10.0

1.25

2.5

5.0

10.0

30

20

10

0 0

1.25

2.5

5.0

10.0

0

Dose (mg/kg)

1.5

2.5

5.0

10.0

Dose (mg/kg)

***

***

***

40

20

0

60

Ly6Clow monocytes (%)

Ly6Clow monocytes (%)

Ly6Clow monocytes (%)

10

Dose (mg/kg)

60

40

20

0 1.25

20

0 0

Dose (mg/kg)

0

30

2.5

5.0

Dose (mg/kg)

10.0

60

Ly6Clow monocytes (%)

0

40

**

Ly6CInter monocytes (%)

30

40

Ly6CInter monocytes (%)

40

Ly6CInter monocytes (%)

Ly6CInter monocytes (%)

40

40

20

0 0

1.25

2.5

5.0

Dose (mg/kg)

10.0

60

40

20

0 0

1.25

2.5

5.0

Dose (mg/kg)

10.0

0

1.5

2.5

5.0

10.0

Dose (mg/kg)

Figure S2, Related to Figure 1.N-Acetyl-MDP, N-Glycolyl-MDP and L18-MDP, but not Murabutide, drive polarization of circulating Ly6Chigh monocytes into Ly6Clow monocytes. Wild type mice were treated with vehicle (white histogram) or increasing doses 1.25, 2.5, 5 and 10 mg/kg (black histograms) of NOD2 agonists. Animals were sacrificed 48 hours following treatment and percentages of Ly6Chigh, Ly6Cinter and Ly6Clow blood monocytes were analyzed by flow cytometry. Data are presented as mean ± SEM of two independent experiments (n=4 mice per dose). * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001 (Two-way Anova followed by Tuckey test).

Figure S3, Related to Figure 1. Absence of functional CCR2 and CX3CR1 receptors does not influence effects of MDP treatment. Flow cytometry analysis (left panels) and gating strategies (right panels) of blood monocytes from Ccr2-/- and Cx3cr1-/mice (n=6 mice/group) following daily treatment with vehicle (solid lines) or MDP (dashed lines) (i.v.). Results are presented as (a) percentages of total monocytes and (b) percentages of Ly6Chigh, Ly6Cinter and Ly6Clow monocyte subsets. Inset: wild-type mice treated with vehicule (solid line) or MDP (48 hours, dashed line) (i.v.) are shown to illustrate the increase in blood monocytes following MDP treatment. (c) Gating strategies are shown for Ly6Chigh, Ly6Cinter and Ly6Clow circulating monocytes from Ccr2-/- and Cx3cr1-/- mice. Data are presented as mean ± SEM of three independent experiments. * p ≤ 0.05 (unpaired t-test).

Figure S4, Related to Figure 3. MDP treatment does not modulate the frequencies of splenic dendritic cells and of blood neutrophils. (a) Gating strategies of plasmacytoid dendritic cells (pDC) and conventional dendritic cells (cDC) in spleen of wild-type mice (n=5 mice/group) treated with vehicle or MDP (i.v.) (48 hours). (b) Gating strategies of blood neutrophils of wild-type mice (n=5 mice/group) treated with vehicle or MDP (i.v.) (48 hours). Data are presented as absolute numbers ± SEM and representative of two independent experiments.

Figure S5, Related to Figure 3. MDP treatment does not affect expression levels of hematopoietic monocyte precursors. (a) Flow cytometry gating strategies of hematopoietic monocyte precursors (LSK, cMoP and MDPs) in bone marrow of wild-type mice (n=4 mice/group). (b) Frequencies (%) of LSK, MDPs and cMoP in wild-type mice (n=4 mice/group) treated with vehicle or MDP (i.v.) at indicated times. Data are presented as absolute numbers of monocyte precursors. Dot plots show the distribution and mean (± SEM, horizontal lines) of monocyte precursors of three independent experiments. Each symbol represents an independent individual.

b Mean arthritic score

a 3.0

Arthritic + vehicle Arthritic + MDP

2.5 2.0 ** ** ** ** ** *

1.5 1.0 0

2

Serum

4

6

8

Ratio Ly6Chigh/Ly6Clow Days

Arthritic + vehicle

Arthritic + MDP

3

2.1 ± 0.4

2.1 ± 0.2*

5

2.1 ± 0.2

0.7 ± 0.1 *

7

2.5 ± 0.3

0.4 ± 0.1 *

10 12 14 16 18 20

Days

Serum + MDP

Figure S6, Related to Figure 6. MDP treatment reduces joint inflammation of mice in serum transfer-induced arthritis. (a) Arthritis was induced in wild-type mice (n=5 mice/group) by administration of arthritogenic serum and animals were daily treated with vehicle (black curve) or MDP (grey curve). Arthritic scores were daily assessed as detailed in Methods section. (b) Ratios of Ly6Chigh/Ly6Clow monocytes in blood of arthritic mice at indicated times following arthritis induction. Results are presented as mean arthritic score ± SEM of three independent experiments. * p ≤ 0.05 and ** p ≤ 0.01 as compared to arthritic mice treated with vehicle. (unpaired-t-test).

Nr4a1-/-

Wild-type Control Receptor

Clo-lipo+ MDP

MDP

Ly6Clow monocytes

Ly6Chigh monocytes

Ly6Clow monocytes

Ly6Clow monocytes

CD11b

++ (19500)

++ (36350)

++ (26950)

++ (22800)

CD11c

+ (800)

- (< 250)

+ (3800)

+ (3400)

CD115

+ (1100)

+ (1400)

+ (2100)

+ (1150)

F4/80

++ (8400)

++ (7050)

+ (3300)

+ (4000)

CD62L

+ (1450)

++ (16600)

+ (1300)

+ (3800)

++ (40000)

+ (2050)

++ (6900)

++ (9500)

MHC class II

- (< 250)

- (< 250)

- (< 250)

- (< 250)

7/4

- (< 250)

+ (2000)

- (< 250)

- (< 250)

CD49b

+/- (< 350)

+ (3500)

- (< 250)

- (< 250)

LFA-1 (CD11a/CD18)

++ (5950)

++ (5050)

++ (6500)

++ (7100)

CX3CR1

++ (60800)

++ (28950)

++ (54800)

+ (1)

± (< 350)

++ (5010)

+ (1700)

++ (6400)

CD43

CCR2

Table S1, Related to Figure 4. Characterization of surface antigen expression of Ly6Clow monocytes following MDP treatment. The surface marker expression of blood Ly6Clow monocytes (grey) isolated from naïve, Clo-lipo (clodronateliposomes) and Nr4a1-/- mice was compared to those of Ly6Chigh monocytes from WT mice following treatment with MDP (n=6 mice/group). Expression profiles have been designated as undetectable (-) (5000 MFI) base on flow cytometry analysis. (1): CX3CR1 expression on monocyte subsets was determined using Cx3cr1+/gfp mice except for Nr4a1-/- mice where CX3CR1 antibody was used.

Supplemental Experimental Procedures Cell isolation and flow cytometry analysis. All cells were incubated with anti-CD16/32 (clone 93; BioLegend) to block Fc receptors. Mice monocytes were identified using CD45 (clone 30F11; BD Biosciences), Ly6G (clone 1A8; BD Biosciences), CD11b (clone M1/70; BD Biosciences), Ly6C (clone HK1.4; BioLegend) and CD115 (clone AFS98; BioLegend) antibodies. For the characterization of surface marker antigens of blood monocytes, F4/80 (clone BM8; BioLegend), CD43 (clone S7; BD Biosciences), CD62L (clone MEL-14, BioLegend), LFA-1 (clone H155-78; BioLegend), CD49b (clone DX5; BioLegend), MHC-II (clone M5/114.15.2; BD Biosciences), CD11c (clone HL-3; BD Biosciences), 7/4 (AbD Serotec), CCR2 (clone 475301, R&D system) and CX3CR1 (polyclonal, R&D system) antibodies were added to the previous panel. Spleen monocytes were identified using Ly6G, NK1.1 (clone PK136; BD Biosciences), CD3 (clone 17A2; eBiosciences), CD19 (clone eBio1D3; eBiosciences), CD11b, CD11c and Ly6C antibodies. Spleen dendritic cells were identified using CD45, CD11c, MHC-II, CD11b, B220 (clone RA3-6B2; BioLegend), and BST2 (clone eBio129c(129c); eBioscience) antibodies. Monocytes and precursors from bone marrow were identified using CD117 (clone 2B8; BD Biosciences), CD11b, Sca-1 (clone D7; BD Biosciences), Ly6G, NK1.1, CD115, Ly6C, CD135 (clone A2F10; eBiosciences), CD3 and CD19 antibodies. A viability dye (LIVE/DEAD Fixable; Molecular Probes) was added to the previous panels to discriminate live cells. Human blood monocytes were analyzed by flow cytometry based on their expression of CD14 and CD16 using CD14-PE-Cy7 (clone M5E2) and CD16A647 (clone 3G8) antibodies (BD Biosciences). Classical, intermediate and non-classical monocytes were specifically identified by selective gating strategy as follows: CD14++ CD16- (classical), CD14+ CD16+ (intermediate) and CD14± CD16++ (non-classical). All analyses were performed on BD LSR II or sorted on BD FACS Aria II. Data were analyzed with FACS Diva software (BD Biosciences). Library preparation and sequencing by RNA-seq. RNA quality was checked using a TapeStation 2200 (Agilent Technologies). RNA integrity number (RIN) for all samples was > 7.7. The Illumina TruSeq RNA sample preparation V2 kit (Illumina Inc.) was used to prepare mRNA sequencing libraries, according to manufacturer’s protocol. Three libraries were independently prepared for each condition. Briefly, 50 ng of total RNA was used for poly A mRNA selection using oligo-dT attached magnetic beads. Fragmented mRNA was used as template for cDNA synthesis by reverse transcriptase with random primers. The cDNA was further converted into double stranded DNA that was end-repaired to incorporate the specific index adaptor for multiplexing, followed by a purification step with Agencourt AMPure XP beads (Beckman Coulter) and an amplification for 15 cycles. The quality of final libraries were examined with a DNA screentape D1000 on a TapeStation 2200 and the quantification was done on the QBit 3.0 fluorometer (ThermoFisher Scientific) as well as by (q)PCR using KAPA library quantification (KAPABiosystems). Subsequently, RNA-seq libraries with unique index were pooled in equimolar ratio (14 samples/pool) and sequenced using both lanes of a rapid run flowcell on an HiSeq 2500 system at the Next-Generation Sequencing Platform (Genomics Center from CHU de Québec Research Center) for paired-ends 100 pb sequencing. The average insert size for the paired-end libraries was 400 bp. Bioinformatic analyses of RNAseq. Reads from HiSeq2500 Illumina were first processed with Trimmomatic v0.33 software to remove low quality with a ILLUMINACLIP value of 2:30:10, a TRAILING value of 30 and a MINLEN value of 36. Reads were then aligned of the mm10 reference genome using Tophat v2.0.14 and Bowtie v2.2.5 with default values for the parameters. The raw reads were splitted into two sequencing lanes, the aligned reads were thus merged and sorted using samtools v1.2 with default values. Duplicated reads were marked with Picard’s MarkDuplicate software v1.130 with the ASSUME_SORTED and CREATE_INDEX values set at true. Gene abundance and differential expression were calculated using the Cuffquant and Cuffdiff softwares from the Cufflinks suite v2.2. Gene counts were obtained with the htseq-count software v0.6.1p1 -s no and -m intersection-noempty parameters values. The annotation used for the quantification, differential expression analysis and gene count was downloaded from USCS on May 23, 2014. Gene counts were analyzed with PCA using the PCA function from the FactoMineR package.

View more...

Comments

Copyright © 2017 DOCIT Inc.