Regulation of macrophage migration in ischemic mouse hearts via an AKT2/NBA1/SPK1 pathway

The role of the AKT2/NBA1/SPK1 signaling cascade in macrophage migration regulation and post-ischemic cardiac remodeling was investigated. We determined that the AKT2/NBA1/SPK1 signaling cascade regulated macrophage migration. A novel role for NBA1 in macrophage migration was discovered. Elevated AKT2 phosphorylation, NBA1, SPK1 (along with phosphorylated SPK1) levels, macrophage recruitment, apoptosis, and fibrosis were found within the infarct area. Atorvastatin had a beneficial effect on cardiac remodeling following myocardial infarction by inhibiting AKT2/NBA1/SPK1-mediated macrophage recruitment, apoptosis, and collagen deposition while increasing angiogenesis in the infarct area. Atorvastatin-related protection of cardiac remodeling following myocardial infarction was abolished in SPK1-KO mice. The AKT2/NAB1/SPK1 pathway is a novel regulating factor of macrophage migration and cardiac remodeling after myocardial infarction.


INTRODUCTION
Improved reperfusion strategies have reduced the mortality rate following acute myocardial infarction (MI), but the prevalence of post-MI heart failure remains high. Macrophage infiltration contributes to ventricular remodeling after MI [1]. SPK1 has recently been implicated in the onset and development of post-ischemic remodeling [2]. However, the role of SPK1 in macrophage migration and cardiac remodeling is unclear. The PI3K/AKT pathway is involved in cardiac remodeling and macrophage migration [3]. The roles of SPK1, AKT, and the AKT substrate, NBA1 [4], in these processes are unclear. Pretreatment with atorvastatin was shown to reduce infarct size by an unknown mechanism [5]. We determined the effect of SPK1 and AKT/NBA1 signaling on macrophage migration and cardiac remodeling. The effects of atorvastatin on SPK1, AKT, and NBA1-mediated macrophage migration, macrophage recruitment, apoptosis, angiogenesis, fibrosis, and the mechanisms of cardiac remodeling were elucidated.
The association between AKT2 phosphorylation and NBA1 expression was examined in lipopolysaccharidechallenged peritoneal macrophages isolated from AKT2 KO (AKT2-/-) mice [6]. Limited migration was observed in lipopolysaccharide-stimulated peritoneal macrophages isolated from AKT2-/-mice ( Figure 3A). AKT2, P-AKT2, NBA1, SPK1, and P-SPK1 ( Figure 3B-3F) levels were decreased in peritoneal macrophages isolated from AKT2- ANA-1 cells in a trans-well assay. Cells were grown overnight, starved for 24 h and detached. Then, 5×10 5 cells were plated in the upper well in a serum-free RPMI 1640 medium containing 100 ng/ml LPS. After 16 h of incubation, cell migrating across the membrane were stained and counted. Four random fields were counted and the number of migrated cells was used as an index for migration. The experiment was repeated at least three times. (B-G) Protein expression in LPS induced ANA-1 cells. Whole cell lysates were prepared after cells were starved for 24 h and stimulated with 100 ng/ml LPS for 2 h. Immunoblots and graph shows total AKT1 (B), P-AKT1 (C), AKT2 (D), P-AKT2 (E), AKT3 (F), NBA1 (G) and GAPDH protein expression levels. Graph shows GAPDH normalized AKT1 (B), P-AKT1 (C), AKT2 (D), P-AKT2 (E), AKT3 (F), and NBA1 (G) levels. Data are presented as the mean ± SEM; n=3. * P < 0.05, ** P < 0.01 compared with Con (Control); NS=not significant. www.impactjournals.com/oncotarget /-mice with or without lipopolysaccharide stimulation. AKT2 phosphorylation might be an upstream signal for NBA1, SPK1, and P-SPK1.

Atorvastatin mediated post-MI cardioprotection via P-AKT2/NBA1/P-SPK1 inhibition
We used a mouse MI model to explore the mechanism underlying the protective role of atorvastatin [8]. Atorvastatin was administered (10 mg/kg/day) to mice for 1 week before and after the MI procedure. The role of SPK1 in atorvastatin-mediated cardiac protection during remodeling was also examined. AKT2 phosphorylation, NBA1, SPK1, SPK1 phosphorylation, F4/80 protein expression, F4/80 density, and hypertrophy marker ANP mRNA expression in the infarction area were promoted AKT2-/-mice. Primary cells enriched for macrophages were placed into 6-well plates (1×10 6 cells/ml) in DMEM plus 10% FBS. After 24 h incubation and starved overnight, cells were plated in the upper well and serum-free RPMI 1640 medium containing 100 ng/ml LPS was added to the bottom well. After 16 h of incubation, cells migrating across the membrane were stained and counted. Four random fields were counted and the number of migrated cells was used as an index for migration. (B-F) Protein expression in LPS induced murine peritoneal macrophages. Whole cell lysates were prepared after cells were stimulated with LPS. Immunoblots show total AKT2 (B), P-AKT2 (C), NBA1 (D), SPK1 (E), P-SPK1 (F), and GAPDH protein expression levels. Graph shows GAPDH normalized AKT2 (B), P-AKT2 (C), NBA1 (D), SPK1 (E), and P-SPK1 (F) levels. Data are presented as the mean ± SEM; n=3. * P < 0.05, ** P < 0.01 compared with WT without LPS; # P < 0.05, ## P < 0.01 compared with WT+LPS induced group; § P<0.05, § § P<0.01 vs SPK1-/-without ATV treatment group; NS=not significant. at day 7 after MI without treatment and diminished by atorvastatin treatment (Figure 7A-7G).
Echocardiographic measurements showed that atorvastatin treatment increased fractional shortening and decreased LVEDD and LVESD (Table 1). Hemodynamic parameters showed that atorvastatin treatment increased +dP/dt, −dP/dt, decreased LVEDP after isoproterenol induction, and decreased HW/ BW ( Table 2). Atorvastatin exerted cardioprotective function by inhibiting P-AKT2/NBA1/P-SPK1mediated regulation of macrophage recruitment in the infarction area.

Cardioprotective effects of atorvastatin were abolished in SPK1-KO mice
We examined the role of SPK1 in cardiac remodeling in WT and SPK1-/-animal models of MI. The death rate was higher in SPK1-/-MI group than in the WT counterpart at day 7 post-MI ( Figure 8A). Most SPK1-/-MI mice died from cardiac rupture. Atorvastatin treatment ameliorated cardiac remodeling and increased survival rate in the WT MI group. The death rate in the SPK1-/-MI group was not lowered with or without atorvastatin treatment ( Figure 8A). Figure 4: NBA1-siRNA decreased NBA1, SPK1, and P-SPK1 protein expressions, but no effect on AKT2 and P-AKT2 protein expression. First, 5×10 4 ANA-1 cells were infected by NBA1-siRNA infection solution (5×10 8 TU) for 72h, starved overnight and then stimulated with 100 ng/ml LPS for 2h. Whole cell lysates were prepared after cells were stimulated with LPS. Immunoblots show total AKT2 (A), P-AKT2 (B), NBA1 (C), SPK1 (D), P-SPK1 (E) and GAPDH protein expression levels. Graph shows GAPDH normalized AKT2 (A), P-AKT2 (B), NBA1 (C), SPK1 (D), and P-SPK1 (E) levels. Data are presented as the mean ± SEM; n=3. * P<0.05, ** P < 0.01 compared with Con without LPS; # P<0.05, ## P < 0.01 compared Con+LPS group; NS=not significant. www.impactjournals.com/oncotarget Cardiac rupture and heart failure were the cause of death in these mice.
The infarction area was larger in the SPK1-/-MI group compared to the WT MI group ( Figure 8B). Atorvastatin treatment limited the infarction area in the WT group, but not in the SPK1-/-MI group ( Figure  8B). Atorvastatin treatment inhibited P-AKT2 and NBA1 ( Figure 8C-8D) protein expression in the WT and SPK1-/-MI groups (P<0.05), but the difference between WT and SPK1-/-groups was not significant (P>0.05).
Fractional shortening decreased and LVESD increased 7 days after MI in SPK1-/-mice with or without atorvastatin treatment. Echocardiographic analysis demonstrated no difference in fractional shortening, LVEDD, or LVESD in SPK1-/-mice with or without atorvastatin treatment (Table 1). Hemodynamic parameters displayed a drop in +dP/dt and −dP/dt and a rise in LVEDP after MI in the SPK1-/-groups compared to the WT groups. There was no difference in +dP/dt, −dP/dt, LVEDP and HW/BW in SPK1-KO mice with or without atorvastatin treatment ( Table 2). . Data are presented as the mean ± SEM; n=3. * P < 0.05, ** P < 0.01 compared with Con without LPS; # P < 0.05, ## P < 0.01 compared with Con + LPS induced group; NS=not significant. www.impactjournals.com/oncotarget DISCUSSION AKT2 phosphorylation/NBA1/SPK1 phosphorylation was involved in macrophage migration and cardiac remodeling after MI. We demonstrated that SPK1 might serve as a beneficial cytokine during cardiac remodeling in MI animal model. Atorvastatin attenuated post-ischemic pathologic remodeling by suppressing the levels of phosphorylated AKT2, NBA1, and phosphorylated SPK1, macrophage recruitment, apoptosis, collagen deposition, and increased angiogenesis in the infarction area.
NBA1 is a primary component of the BRCA1 A complex, which contains Brca1/Bard1, Abra1, RAP80, BRCC36, and BRE. NBA1 is localized in both the nucleus and cytoplasm and helps to maintain BRE and Abra1 levels for BRCA1 recruitment to sites of DNA damage.
NBA1 also contributes to multiple cell-cycle-dependent processes [9]. Our study confirmed participation of NBA1 in macrophage migration and cardiac remodeling as a potential upstream signal for SPK1(P-SPK1) and a downstream signal of AKT2 phosphorylation. These findings favor the therapeutic potential of NBA1.
The effects of SPK1 on cardiac remodeling have been extensively investigated. Jin and colleagues revealed a cardioprotective effect of SPK1 in ischemic Data were obtained 7 days after MI. FS%, percent fractional shorting; LVEDD, LV end-diastolic dimension in mm; LVESD, LV end-systolic dimension in mm; * P < 0.05; ** P < 0.01 vs WT Sham; # P < 0.05, ## P < 0.01 vs WT+M; † P < 0.05, † † P < 0.01 vs WT+ MI+ ATV; § P > 0.05 vs SPK1-KO+MI. Base and 10 ng of isoproterenol (ISO) stimulated cardiac hemodynamic function were recorded. Data were obtained from mice 7 days after MI. HW/BW: heart weight/body weight. +dP/dt and −dP/dt: maximal 1st time derivatives of left ventricular pressure rise and fall, respectively. LVEDP: LV end diastolic pressure. * P < 0.05; ** P < 0.01 vs WT Con; # P < 0.05, ## P < 0.01 vs WT MI; † P < 0.05, † † P < 0.01 vs WT MI+ATV; § P > 0.05 vs SPK1-/-MI+ATV. www.impactjournals.com/oncotarget postconditioning [2]. Some studies suggested that a low dose of SPK1 inhibitor N, N-dimethylsphingosine (DMS) was cardioprotective in vitro [8]. The role of SPK1 in cardiac remodeling remains controversial. Macrophages are pivotal for wound healing with biological functions including cell debris phagocytosis, apoptosis induction, inflammatory cell and myofibroblasts recruitment, neovascularization regulation, and induction of scar formation [15]. Connective tissue formation is an essential process in the healing and repair of myocardial repair [16]. The fragile ventricular wall will undergo sudden rupture or heart failure in the absence of these connective tissues [17]. The role of macrophages in mediating the fibrotic response is complex. Excessive and prolonged infiltration of macrophages into the infarct myocardium was shown to be harmful [18]. Macrophage depletion resulted in a high mortality rate accompanied by increased left ventricular dilatation and wall thinning. Depletion of infiltrating macrophages impaired wound healing and increased mortality after myocardial injury [19].
We used a murine model of MI to explore the role of SPK1 in cardiac remodeling. Our data revealed that the survival rate of SPK1-KO MI mice was lower than WT MI mice due to the risk of cardiac rupture. The increased risk of cardiac rupture was attributed to diminished inflammation reaction, connective tissue content, and higher cell apoptosis 7 days after MI. Macrophage density in the infarction area was positively correlated with the levels of SPK1 (P-SPK1).
Atorvastatin treatment exceeded the benefits of lipid level reduction alone [20]. Numerous studies have demonstrated the beneficial role of atorvastatin in cardiac remodeling [21] and macrophage migration [7]. Our results indicated that atorvastatin might ameliorate cardiac remodeling by inhibiting myocardial infarction, macrophage recruitment, apoptosis, fibrosis, and increasing angiogenesis in the infarct area after MI injury. The protective function of statins against cardiac ischemic injury was partially attenuated in SPK1-KO mice.

Figure 8: ATV protected cardiac function by reducing myocardial infarction area, macrophage recruitment, apoptosis, fibrosis, and increasing vascular density and survival curve, but ATV's-offered protective effect on cardiac remodeling completely dampened in SPK1-KO mice.
WT and SPK1-/-mice were fed ATV (10 mg/kg/day) for 1 week before and after MI.
Survival curve was observed between WT MI, WT + ATV MI, SPK1-/-MI, SPK1-/-+ ATV MI groups. After MI 7 days, the death rate of SPK1-/-MI group is higher than that of WT MI group (A). ATV treatment could decrease the death rate of WT MI model, but did not overtly lower the death rate of SPK1-/-models (A). * P<0.05, ** P<0.01 vs WT sham without ATV group; # P < 0.05, ## P < 0.01 compared with WT MI without ATV treatment group; § P < 0.05, § § P < 0.01 compared with WT MI with ATV treatment group; ※ P>0.05 vs SPK1-/-MI without ATV treatment group. The infarction area of SPK1-/-MI group is bigger than that of WT MI group (B). ATV could limit infarction area of WT MI with ATV treatment group (B). But there was no difference in the infarction area in SPK1-/-MI mice with and without ATV treatment (B). * P < 0.05, ** P < 0.01 compared with WT MI without ATV treatment group. # P< 0.05, ## P < 0.01 compared with WT MI with ATV treatment group. ATV could inhibit P-AKT2 (C) and NBA1 (D) protein levels in WT and SPK1-/-MI animal model. There is no difference in P-AKT2 and NBA1 protein level between WT MI and SPK1-/-MI with and without ATV treatment. ATV could inhibit Data are presented as the mean ± SEM. * P < 0.05, ** P < 0.01 compared with WT sham without ATV sham group; # P < 0.05, ## P < 0.01 compared with WT MI without ATV treatment group; § P < 0.05, § § P < 0.01 compared with WT MI with ATV treatment group; † P < 0.05, † † P < 0.01 compared with SPK1-/-MI without ATV treatment group; NS=not significant.

Isolation of murine peritoneal macrophages
Peritoneal macrophages, WT or SPK1-/-, were isolated from mice as previously described [24]. Briefly, after mice were euthanized by cervical dislocation, the body was cleaned with 75 % ethanol and a small incision was made in the abdomen. Mice were injected with 10 mL 1×PBS through the peritoneal wall into the peritoneal cavity without puncturing the intestine or any other organ. Peritoneal fluid was collected into a 15 mL tube and centrifuged at 800 g at 4 °C for 5 min. The cell pellet was washed twice and resuspended in DMEM/F12 with 10% FBS, 50 μg/ml gentamicin, 50 μg/ml penicillin, and 50 ug/ml streptomycin. Cells were plated in a 60 mm petri dish and allowed to attach for 2 h in a humidified incubator with 5 % CO 2 at 37 °C. Non-adherent cells were removed by vigorously washing three times with cold PBS. Adherent cells were cultured for another 24 h for further experiments.

SPK1 and NBA1 lentivirus and cell transfection
Production of knockdown SPK1 and NBA1 lentivirus vectors was performed as previously described [8]. GFP-encoding lentiviral strains carrying the siRNA oligonucleotides that targeted 5′-GGCAGAGATAACCTTTAAA-3′ on SPK1 mRNA and 5′-CACTCTGTCCACTGTAAAT-3′ on NBA1 mRNA were designed and cloned into a GV248 vector (Gene Chem, Shanghai, China). A scramble siRNA with the sequence of 5′-TTCTCCGAACACGTGTCACGT-3′ was cloned in parallel to create the negative control lentiviral strain. The titer of the SPK1-siRNA and the negative control was approximately 8×10 8 TU/mL, and the NBA1-siRNA was 5×10 8 TU/mL. ANA-1 cells were transfected by seeding 3-5×10 4 /mL ANA-1 cells (Chinese Academy of Sciences, Shanghai, China) in 6-well plates and cultured for 24 h until 30-40% were fused. Cells were starved for 24 h. Lentiviral stocks were diluted to MOI 20 with enhanced infection solution (Gene Chem, Shanghai, China) containing polybrene (5 ug/mL) according to manufacturer instructions and added to the seeded cells for 12 h with FBS-free RPMI-1640. The virus-containing medium was then replaced with fresh RPMI-1640 medium containing 10% FBS. Reporter gene expression in the lentivirus was observed 3 days after transfection via green fluorescent protein (GFP). Fluorescence microscopy (IX-53; Olympus Corporation, Tokyo, Japan) was used to detect cells that expressed GFP, and the percentage of GFP-positive cells was used to measure the transfection efficiency of the cells.

Macrophage migration
Cell migration was performed in a 24-well transwell migration system (Corning) as previously described [25]. For cell migration assay, cells were pre-treated with atorvastatin (10 uM) for 24 h at 37 °C. Cells were harvested, suspended in serum-free medium, and added to the upper transwell chambers (Corning, USA) at a density of 5×10 5 cells per chamber. Serum-free medium with lipopolysaccharide (100 ng/mL) was added in the lower chamber. After 12 h incubation, the insert was removed, and cells on the upper surface were removed with a cotton swap. Cells on the lower surface were fixed with 4% paraformaldehyde in PBS for 30 min, permeabilized with 0.2% Triton X-100 in PBS for 5 min, and stained with DAPI. The stained cells were subsequently photographed and counted in each field using Image-Pro Plus Software. Total nuclei (DAPI staining, blue) in each field were counted by IP Lab Imagine Analysis Software (Version 3.5; Scanalytics, Fairfax, VA, USA). Results from different fields taken from the same slide were averaged and counted as one sample.

Animal MI model and atorvastatin treatment
The classical, ventilation-based method of MI in mice has been fully described [26]. Experiments were performed according to National Institutes of Health Guidelines on the Use of Laboratory Animals, and all procedures were approved by the Forth Military Medical University Committee on Animal Care. Mice were randomly assigned to eight groups: WT control, WT atorvastatin, WT MI, WT MI+atorvastatin, SPK1-/control, SPK1-/-atorvastatin, SPK1-/-MI, and SPK1-/-MI+atorvastatin groups. Mice were fed atorvastatin (10 mg/kg/day) for 7 days before the MI procedure. Mice were treated with atorvastatin for 7 days after MI. Briefly, mice were anesthetized with 3% isoflurane inhalation, intubated with a 20G intravenous catheter, and ventilated with a mixture of O 2 and 1.5-2% isoflurane using a rodent ventilator (Harvard Mini Vent 845). The stroke volume was 0.2 mL, and the respiratory rate was 120 breaths/min. Animals were placed in a supine position. A left thoracotomy was then performed through the 4 th intercostal space by transverse cutting of the pectoralis muscles to expose the thoracic cage. The thymus was retracted upward, and the left lung was partially collapsed. After the pericardium was opened, the left main descending coronary artery (LCA) was located and ligated with a 6-0 silk suture 2-3 mm from the origin. The ligation was confirmed successfully when the anterior wall of the left ventricle turned pale. The lungs were then inflated to displace air, and the thoracotomy site closed in layers. After 2-5 min ventilation with room air, the animal was gradually weaned from the respirator once spontaneous respiration resumed and remained in a supervised setting until fully conscious. The sham-treated animals underwent the same surgical procedures except the LCA was not occluded.

Echocardiography
We assessed in vivo cardiac function at 4 weeks after MI with an echocardiographic imaging system (Vevo 770, VisualSonic, Toronto, Canada). Mice were anesthetized with 1.5% isoflurane, and two-dimensional echocardiographic views of the mid-ventricular short axis were obtained at the level of the papillary muscle tips below the mitral valve. Left ventricular (LV) internal dimensions were measured and the LV fractional shortening (LVFS) was calculated as previously described [27]. A 1.4 French micro-manometer-tipped catheter (SPR-671; Millar Instruments Inc.) was inserted into the right carotid artery and advanced into the LV of mice that were lightly anesthetized (i.e., maintained spontaneous respirations) with tribromoethanol/amylene hydrate (2.5% w/v, 8 μL/g, injected intraperitoneally; Avertin) For in vivo hemodynamic measurements. Hemodynamic parameters, including heart rate, LV end-diastolic pressure, +dP/dt, and −dP/dt were recorded in closed-chest mode at baseline and in response to 10 ng isoproterenol administered via cannulation of the right internal jugular vein [28].

Immunoblotting
Immunoblots were performed as previously described [29]. LV tissue was homogenized in 10 volumes of lysis buffer (50 mM Tris-HCl, pH 7.4), 150 mM NaCl, 1 mM EDTA, 0.25% sodium deoxycholate, and 1% NP-40 with a protease inhibitor cocktail and phosphatase inhibitor cocktail. The homogenates were centrifuged at 15,000×g for 15 min to obtain the NP-40soluble supernatant and insoluble pellet. Equal amounts of proteins were subjected to SDS-PAGE and subsequently transferred to nitrocellulose membranes. Membranes were scanned with the Odyssey Infrared Imaging System (LI-COR). www.impactjournals.com/oncotarget

Masson's trichrome staining and infarct area calculation
Infarct area was examined by staining heart sections with the standard Masson's trichrome method as previously described [28]. Briefly, the mice hearts were excised post-anesthesia and rinsed with phosphatebuffered saline. Heart specimens were fixed for 1-3 days in 4% paraformaldehyde at 4 °C and embedded in paraffin. Serial sections of 5 μm were cut and placed on polylysinecoated glass slides. Tissue sections were deparaffinized and stained with Masson's trichrome reagent (Sigma-Aldrich) according to manufacturer instructions. Digital photographs were taken and quantified by color threshold measures using Image J software (NIH, Bethesda, MD, USA). The infarct area was measured as the ratio (%) of the infarct area divided by the entire LV area.

Immunohistochemical staining
Tissue sections (4-5 μm) were deparaffinized, dehydrated using a graded series of ethanol solutions, and stained with F4/80, CD31, and Ly6c. Endogenous peroxidase was inactivated with 3% hydrogen peroxide at room temperature for 20 min. Slides were soaked in 0.1 mol/L citrate buffer (pH 6.0) and placed in an autoclave at 121 °C for 2 min to retrieve antigens. After washing with PBS (pH 7.4), the sections were blocked with 1% BSA diluted in PBS at 37 °C for 30 min and incubated with anti-F4/80 (1:50, Abcam, Cambridge, UK) at 4 °C overnight. Sections were then rinsed with PBS, incubated with HRP-conjugated goat anti-mouse antibody and DAB (DAKO, Glostrup, Denmark), washed with distilled water, incubated with 0.5% PAS for 10 min in a dark chamber, and washed with distilled water for 3 min. All sections were counterstained with hematoxylin. All immunohistochemically stained sections were observed and photographed with an Olympus microscope (IX-70 OLYMPUS, Japan). Four representative fields within each section were randomly chosen and captured under 200X magnification. The integrated optical density (IOD) in each image was measured with the same setting for all slides, and the density was calculated as IOD/total area of each image.

Tunel assay
After myocardial infarction, the hearts were fixed in 4% paraformaldehyde in PBS for 24 h at room temperature. Fixed tissues were embedded in a paraffin block, and 4-5 μm slices were cut from each tissue block. Immunohistochemical detection of apoptotic cardiomyocytes was performed with an apoptosis detection kit (Boehringer Mannheim, Ridgefield, CT, U.S.A.) according to the manufacturer's instructions. Myocardial apoptosis was qualitatively analyzed by detection of DNA fragmentation (DNA ladders) and quantitatively analyzed by a terminal dUTP nick end-labeling (TUNEL) assay as described previously [31]. Assays were performed in a blinded manner.

Isolation of single heart cells
Single cells were isolated from mouse hearts according to previous methods [32]. Mice were sacrificed, mouse hearts were excised in whole, and placed in heparinized saline. The heart was finely minced into 1-2 mm pieces after removal of epicardial fatty tissue and the aorta. Blood was removed by repeated washing in heparinized saline. The tissue was digested with collagenase (1 mg/mL, Thermo Fisher Scientific), trypsin (0.1%, Gibco), and DNase I (10 μg/mL, Roche) in 10 mL RPMI media (Hyclone) for 1 h at 37 °C with occasional shaking. Released cells were separated from the remaining tissue by filtration through a 100 μm nylon cell strainer (BD Falcon), washed with R10 media (RPMI-1640 supplemented with 10% heat-inactivated FBS, 2 mM L-glutamine, 25 μM 2-mercaptoethanol, 1% penicillin/streptomycin), and placed on ice. These steps were repeated twice to digest the remaining tissue. Any residual solid tissue was treated with EDTA (2 mM) in digestion media for 10 min at 37 °C, collagenase (2 mg/ mL) in R10 media for 30 min at 37 °C, and released cells were filtered. Collected cells were pooled and pelleted at 300 g for 5 min at 4 °C. Single cell suspensions were cleaned in PBS and centrifuged at 500 g for 5 min at 4 °C. Cell viability (70-90%) was confirmed by the trypan blue exclusion method.

Flow cytometry
Cell flow cytometry was performed as previously described [33]. Cells were stained with fluorophoreconjugated antibodies for 30 min on ice and washed. Cells were fixed and permeabilized after surface staining for intracellular staining using cytofix/cytoperm buffers according to manufacturer's instructions (BD Biosciences) and stained with anti-Ly6C-FITC (BD Biosciences 553104) for 30 min at 4 °C. Cells were pelleted and resuspended in staining buffer. Flow cytometry was performed with an FACSAria™ II flow cytometer (BD Biosciences) and FlowJo software version 7.6.1 was used for analysis.

Statistical analysis
Data were expressed as the mean ± SD from at least four independent experiments or 4 mice per group. Statistical significance was determined by oneway ANOVA with Bonferroni correction for multiple comparisons or unpaired student t-tests. A P-value <0.05 was considered statistically significant.