CX3CR1 deficiency suppresses activation and neurotoxicity of microglia/macrophage in experimental ischemic stroke

Background Chemokine (C-X3-C motif) ligand 1 (CX3CL1)/ CX3C chemokine receptor 1 (CX3CR1) signaling is important in modulating the communication between neurons and resident microglia/migrated macrophages in the central nervous system (CNS). Although CX3CR1 deficiency is associated with an improved outcome following ischemic brain injury, the mechanism of this observation is largely unknown. The aim of this study was to investigate how CX3CR1 deficiency influences microglia/macrophage functions in the context of its protection following brain ischemia. Methods Wild-type (WT) and CX3CR1-deficient (CX3CR1-/-) mice were subjected to transient middle cerebral artery occlusion (MCAO) and reperfusion. The ischemic brain damage was monitored by rodent high-field magnetic resonance imaging. Neurological deficit was assessed daily. Neuronal apoptotic death and reactive oxygen species (ROS) production were analyzed by immunostaining and live imaging. Activation/inflammatory response of microglia/macrophage were assessed using immunohistochemistry, flow cytometry, 5-bromo-2-deoxyuridine labeling, cytokine ELISA, and real-time PCR. Results CX3CR1-/- mice displayed significantly smaller infarcts and less severe neurological deficits compared to WT controls, following MCAO. In addition, CX3CR1-/- MCAO mice displayed fewer apoptotic neurons and reduced ROS levels. Impaired CX3CR1 signaling abrogated the recruitment of monocyte-derived macrophages from the periphery, suppressed the proliferation of CNS microglia and infiltrated macrophage, facilitated the alternative activation (M2 state) of microglia/macrophages, and attenuated their ability to synthesize and release inflammatory cytokines. Conclusion Our results suggest that inhibition of CX3CR1 signaling could function as a therapeutic modality in ischemic brain injury, by reducing recruitment of peripheral macrophages and expansion/activation of CNS microglia and macrophages, resulting in protection of neurological function.


Background
Ischemic brain injury elicits an inflammatory response that involves activation of microglia and monocytederived macrophages in the central nervous system (CNS). Within the lesion, these cells adopt unique molecular phenotypes with either protective or detrimental effects on neuron survival [1]. Despite this basic understanding, the mechanisms controlling functional diversity in microglia/macrophages remain elusive but undoubtedly include differential signaling via chemokines, cytokines, and other cues present in the ischemic microenvironment.
The chemokine (C-X3-C motif ) ligand 1 (CX3CL1)/ CX3C chemokine receptor 1 (CX3CR1) signaling pathway has been shown to play an important role in the maintenance of neural-immune communication in health and disease [2,3]. Fractalkine (CX3CL1), a membrane-bound chemokine, is expressed predominantly by neurons in the healthy CNS [4,5] while its receptor (CX3CR1) is highly expressed on resident brain microglia and on peripheral immune cell populations including macrophages [6,7]. CX3CR1 selectively modulates microglia activity in response to its ligand CX3CL1 [6]. CX3CL1/CX3CR1 also regulates recruitment of circulating leukocytes to sites of injury [8,9].
Under neuroinflammatory conditions, loss of signaling via CX3CR1 has been shown to have either protective [10,11], or detrimental effects [12], or no effects at all [13,14]. These conflicting observations indicate the complexity and disease-specific regulation of neuron-microglia communication via CX3CL1 and CX3CR1. In regards to acute CNS injury models (transient and permanent brain ischemia, spinal cord injury), the collective data suggest that the absence of CX3CR1 significantly reduces ischemic damage and inflammation [15][16][17].
Although the processes of microglia activation, cytokine production, and phagocytosis have been studied in several animal models of ischemia injury in CX3CR1deficient (CX3CR1 −/− ) mice, it remains unclear as to how CX3CR1 deficiency alters the function of these cells, resulting in reduced damage during ischemic injury (highlighted by several controversial findings). Following induction of focal cerebral ischemia within CX3CR1 −/− mice, no difference was observed in the number of IL-1β-expressing microglia; rather, decreased leukocyte infiltration is involved in the development of smaller infarcts [15]. Conversely, in an animal model of spinal cord injury, modest but significantly more recruited monocytes (CD45 hi ) accumulate in the spinal cord of CX3CR1 −/− mice by 3 days post-injury [16]; however, a decrease in the number of CD11b + /Ly6C lo /iNOS + macrophages after 7 days post-injury was associated with reduced neuropathology and enhanced functional recovery in these CX3CR1-deficient mice [16]. To date, little is known about the CX3CL1/CX3CR1 pathway in the context of microglia activation following brain ischemia.
In the current study, using a murine model of transient middle cerebral artery occlusion (MCAO), we tested the consequences of the absence of CX3CR1 on microglia/macrophage proliferation/recruitment and on their neurotoxicity after brain ischemia. We find that loss of CX3CR1 signaling attenuates the monocyte recruitment, the expansion of resident microglia and newly recruited monocytes, and reduced their inflammatory capacity after MCAO, contributing to neuroprotection in CX3CR1deficient mice.

Animals
Breeding pairs of CX3CR1 −/− mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). Knockouts were generated by replacing the second exon of the CX3CR1 gene with the enhanced green fluorescent protein (GFP) reporter gene, and backcrossed for more than 10 generations to C57BL/6 [13]. Cells under control of the endogenous CX3CR1 locus (that is, microglia, macrophages, dendritic cells, and so forth) in homozygote CX3CR1 −/− mice are labeled with GFP and also lack CX3CR1 receptor function. Wild-type (WT) C57BL/6 mice were purchased from Taconic (Oxnard, CA, USA). All animals were housed in pathogen-free conditions at the animal facilities of the Barrow Neurological Institute, Phoenix, AZ, USA. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the Barrow Neurological Institute and performed according to the Revised Guide for the Care and Use of Laboratory Animals.

Transient middle cerebral artery occlusion and reperfusion
Adult male mice (aged 10 to 14 weeks, weight 24 to 27 g) were exposed to transient (90 minutes) focal cerebral ischemia induced by occlusion of the right middle cerebral artery using an intraluminal filament method [18,19]. The production of an infarct was confirmed by 2,3,5-triphenyltetrazolium chloride (TTC) [20]. The methodology for MCAO and TTC staining is detailed in Additional file 1: Supplementary Methods.

Neurological deficit assessment
Daily neurological deficit assessment was performed by investigators blinded to the control and MCAO groups as described previously [19]. Rating scale: 0 = no deficit, 1 = failure to extend left forepaw, 2 = decreased grip strength of left forepaw, 3 = circling to left by pulling the tail, and 4 = spontaneous circling.
Cerebral blood flow, apparent diffusion coefficient values, and T2-weighted images by rodent magnetic resonance imaging Magnetic resonance imaging (MRI) was performed on a 7-T small animal MRI, 30-cm horizontal-bore magnet, and BioSpec Avance III spectrometer (Bruker Daltonics Inc., Fremont, CA, USA). In order to assess whether the ischemia and reperfusion were induced successfully, single slice cerebral blood flow (CBF) images were acquired before MCAO, 1 hour after MCAO and 24 hours after reperfusion, using a Continuous Arterial Spin Labeling sequence. Multiple Segments Echo Planer Imaging sequences were used to acquire apparent diffusion coefficient (ADC) values 30 minutes after MCAO to assess the damage volume. T2weighted images were acquired 24 and 72 hours after MCAO to evaluate the development of ischemic lesions, using a Rapid Acquisition with Refocused Echoes sequence. Detailed methodologies for these imaging are available in Additional file 1: Supplementary Methods. MRI data were analyzed using the MED × 3.4.3 software package (Medical Numerics Inc., Germantown, MA, USA) on a LINUX workstation.

Reactive oxygen species assessment in vivo
Reactive oxygen species (ROS) generated in the brain were assessed in live mice by using the Xenogen IVIS200 imager (Caliper Life Sciences, Alameda, CA, USA) [21]. Briefly, mice were intraperitoneally injected with 200 mg/kg Luminol (Invitrogen, Carlsbad, CA, USA). After 10 minutes, bioluminescence images were captured with exposure time of 3 minutes. A region of interest tool was used to measure the chemiluminescent intensity of the whole brain. Data were collected as photons per second per centimeter squared using the Living Image software (Caliper Life Sciences).

Enzyme-linked immunosorbent assay
Brain homogenates were prepared from WT and CX3CR1 −/− mice 72 hours after MCAO. Animals were anesthetized and the brains were removed and immediately frozen in liquid nitrogen. The ipsilateral or contralateral hemisphere was homogenized in RIPA buffer (10 μl/mg brain, Sigma, St. Louis, MO, USA). Protein concentration was measured with bicinchoninic acid Protein Assay kit (Pierce, Appleton, WI, USA). The total protein concentration was adjusted to 1 mg/ml protein extract. The concentrations of IL-1β, IL-6, and TNF-α in brain homogenates were quantified by enzyme-linked immunosorbent assay (ELISA) kits (BioLegend, San Diego, CA, USA) and converted into pg/mg protein extract.

5-bromo-2-deoxyuridine labeling
In order to visualize proliferating mononuclear cells in CNS, mice were injected intraperitoneally with 5bromo-2-deoxyuridine (BrdU) (50 μg/g of mouse weight in saline, Sigma) immediately before the MCAO procedure and again 24, 48, and 60 hours after surgery. At 72 hours after surgery, mice were anesthetized with isoflurane and transcardially perfused with PBS. Brains were removed, and microglia and invading leukocyte isolation was performed according to the standardized protocol described below.

Isolation of mononuclear cells from the central nervous system
The isolation of microglia and invading leukocytes are based on discontinuous percoll gradients [22][23][24]. Briefly, fresh brain tissues were removed from mice and cut into~2 mm pieces and incubated in 10 mM Hepes/NaOH buffer (10 mM HEPES, 150 mM NaOH, 7 mM KCL, 1 mM MgCl 2 , 1 mM MgCl 2 , 0.36 mM CaCl 2 ) containing 1 mg/ml collagenase (Sigma) for 1 hour at 37°C. The tissues were dispersed with a syringe, filtered through a 100-mm wire mesh, and centrifuged at 2,000 rpm for 5 minutes at 4°C. After centrifugation, cell pellets were resuspended in 15 ml 30% Percoll (Amersham Biosciences, Piscataway, NJ, USA), and centrifuged against 70% Percoll in a 50-ml tube for 15 minutes. The cell monolayer at the 30 to 70% Percoll interface was collected and washed once for further staining.

Flow cytometry Cell phenotype
The number of microglia/macrophage, their proliferation properties and inflammatory cytokine secretion were analyzed by flow cytometry. Single cell suspensions prepared from brain tissues were stained with fluorescently labeled antibodies: APC-CD45, PerCP-Cy5.5-CD45, PE-Ly6G, BV421-Ly6G, PE-Cy7-CD11b, PerCP-Cy5.5-BrdU, PE-IL-1β, V450-IL-6, or APC-TNF-α at designed combination. All antibodies and the isotype controls were purchased from BD Biosciences, San Jose, CA, USA. The staining was performed according to the manufacturer's instructions and the details are available in Additional file 1: Supplementary Methods. After staining, samples were analyzed using a FACSAria Ι flow cytometer (BD Biosciences). To avoid the interference of GFP expressed on CX3CR1 −/− microglia/macrophage, fluors such as FITC and Alexafluor 488 that excite in the same spectrum at the same filter sets as GFP were excluded and the proper compensation was performed in flow cytometry. Subsequent data analyses were completed using FACSDiva software or FCS Express 4 software (BD Biosciences).

Cell sorting
Microglia/macrophages were enriched from the single-cell suspension of the ischemic hemisphere of MCAO mice using CD11b MicroBeads (Miltenyi Biotec, Auburn, CA, USA), and then sorted via APC-45/PE-Cy7-CD11b/BV421-Ly6G makers from the remaining cells by FACSAria I using Diva software (BD Biosciences). Purity of microglia and macrophages obtained by this approach reached 99%.

Quantitative real-time polymerase chain reaction analysis
Total RNA was extracted from the sorted microglia/ macrophages using the RNeasy Micro Kit (Qiagen, Germantown, MD, USA) according to the manufacturer's instructions; 1 μg was used to synthesize the first strand of cDNA using the Superscript First-Strand Synthesis System for real-time PCR (Invitrogen). PCR was performed on the Opticon 2 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) using corresponding primers (Table 1) and iQ TM SYBR Green Supermix (Bio-Rad). Cycle conditions included heating for 5 minutes at 95°C, followed by 40 cycles of 30 seconds at 95°C, 30 seconds at 60°C, and 60 seconds at 72°C. A melt curve analysis was performed to ensure specific amplification. For each target gene, relative levels of expression were normalized against housekeeping gene GAPDH of the same sample. The relative expression levels of the mRNAs were then reported as fold changes versus sham controls.

Statistical analysis
Results are presented as the means ± SEM. Statistical differences between two groups were evaluated by the twotailed unpaired Student's t-test. Multiple comparisons were performed with two-way analysis of variance accompanied by Bonferroni post-hoc test. Values of P < 0.05 were considered significant.
To assess the infarct volume, T2-weighted images were acquired at 24 and 72 hours following MCAO. The infarct volume in WT mice was 36.5 ± 5.7 mm 3 at 24 hours and increased to 45.8 ± 6.8 mm 3 at 72 hours. Within CX3CR1 −/− mice, infarct volume was 26.9 ± 5.7 mm 3 at 24 hours, and a modest but significant decrease (to 19.0 ± 4.9 mm 3 ) was observed at 72 hours ( Figure 1A). Although there was no significant difference in infarct volume at 24 hours between the two groups, CX3CR1 −/− mice showed markedly smaller infarct volume at 72 hours relative to WT mice (P < 0.01, Figure 1B). The infarct observed in MRI scan was confirmed by TTC staining (Additional file 1: Figure S1D).
To further assess the differential response to MCAO in CX3CR1 −/− and WT mice, the neurological deficit was assessed daily following MCAO. WT mice had an average clinical score of 2.4 ± 0.4 at 24 hours and 2.0 ± 0.3 at 72 hours, while CX3CR1 −/− mice had clinical scores of 2.0 ± 0.3 at 24 hours and 1.2 ± 0.2 at 72 hours indicative of the beginning of recovery at 72 hours in CX3CR1 −/− mice but not in WT mice ( Figure 1C).

CX3CR1 deficiency attenuates neuronal apoptosis after middle cerebral artery occlusion
Double staining with NeuN (neuron marker) and cleaved Caspase-3 (apoptotic marker) was performed to investigate whether CX3CR1-dependent differences observed in the size of ischemic damage at 72 hours following MCAO were associated with differential neuronal apoptosis in peri-infarct areas. More cleaved Caspase-3 positive cells were observed in WT mice compared to CX3CR1 −/− mice and co-localized with NeuN in most apoptotic cells (Figure 2A). The number of cleaved Caspase-3 positive neurons was 179.3 ± 15.6/mm 2 in WT mice and 95.1 ± 16.9/mm 2 in CX3CR1 −/− mice 72 hours after MCAO (P < 0.01, Figure 2B). To reveal the phenotype of the other apoptotic cells, double staining of cleaved Caspase-3 with Iba-1 or GFAP markers, respectively, were employed. These experiments revealed cleaved Caspase-3 positive staining within some microglia and astrocytes (data not shown).
Fewer microglia and macrophages in ipsilateral hemisphere of CX3CR1 −/− mice after middle cerebral artery occlusion Microglia activation plays an important role in the pathological progression after stroke, and is regulated by CX3CR1 [6,12]. To investigate the effects of CX3CR1 deficiency upon these processes, expression of microglia/ macrophage activation marker Iba-1 was examined at different sites of the brain in WT and CX3CR1 −/− mice 72 hours following MCAO ( Figure 3A). In the ipsilateral hemisphere of WT mice, the numbers of Iba-1 positive cells in hippocampus, striatum, cortex and peri-infract area (666.1 ± 50.0, 1132.8 ± 96.1, 730.9 ± 68.1, 489.4 ± 56.1/mm 2 ) were significantly higher than those in CX3CR1 −/− mice    Figure 3B).
Following ischemia, activated microglia/macrophage can potentially exert either a protective or detrimental effect, suggesting that these cells may acquire different phenotypes belonging to the classical (M1) or to the alternative (M2) active status. We found that activated M1-like Iba-1 + cells, which have shorter and thicker processes and bigger cell bodies, were visualized in the WT brain section, while ramified M2-like Iba-1 + cells were predominantly located in the CX3CR1 −/− brain ( Figure 3A). To evaluate their M1/M2 polarization, we sorted and purified microglia/macrophages (CD45 + /CD11b + /Ly6G − ) from the ischemic hemisphere of CX3CR1 −/− and WT MCAO mice brain. Using real-time PCR, we found that the levels of tested M1-and M2-type genes in Fluorescence Activated Cell Sorter (FACS)-sorted microglia/macrophages from WT mice were increased starting from 1 day after MCAO and further elevated by 3 days post-MCAO (data not shown). Within CX3CR1 −/− microglia/macrophages, the M2-type genes (Ym1, Mcr1) were significantly increased, whereas the M1-type gene (iNOS) was notably decreased compared to WT microglia/ macrophages when isolated at 72 hours after MCAO ( Figure 3E). These results suggest that deficiency of CX3CR1 may facilitate the alternative activation (M2 state) of microglia/macrophages in stroke.
CX3CR1 deficiency attenuates reactive oxygen species generation in brain after middle cerebral artery occlusion ROS, generated as by-products of cellular metabolism, have long been known to be a component of the inflammatory response after ischemia. To investigate whether CX3CR1 deficiency had effects upon ROS production, we assessed ROS levels in live mice using the Xenogen IVIS200 imager at 24 and 72 hours after MCAO. The chemiluminescence detection of ROS was performed in the brains, specifically in the ipsilateral hemisphere ( Figure 5A). The mean chemiluminescence intensity of brain in WT mice was 1,090.3 ± 127.4 p/s/cm 2 /sr (photons per second per centimeter squared per steradian) at 24 hours and 850.9 ± 81.7 p/s/ cm 2 /sr at 72 hours. Within CX3CR1 −/− mice, mean chemiluminescence intensity values were 1,350.9 ± 118.9 p/s/ cm 2 /sr at 24 hours and 641.7 ± 47.6 p/s/cm 2 /sr at 72 hours. At 24 hours post-ischemia, no differences in ROS levels were observed between the two groups (P > 0.05). At 72 hours post-ischemia, the ROS levels decreased significantly in CX3CR1 −/− mice compared to 24 hours (P < 0.01), while no change was observed in WT mice (P > 0.05) ( Figure 5B). In addition, the oxidative impairment of neurons was Figure 4 Suppressed proliferation of CD45 low /CD11b + /Ly6G − and CD45 hi /CD11b + /Ly6G − cells in the ipsilateral hemisphere of CX3CR1 −/− mice after middle cerebral artery occlusion. (A) Flow cytometry analysis of proliferation of Ly6Ggated CD45 low /CD11b + (microglia) and CD45 hi /CD11b + (macrophages/activated microglia) population with 5-bromo-2-deoxyuridine (BrdU) incorporation in the ipsilateral hemisphere of wild-type (WT) and CX3CR1 −/− mice 72 hours after middle cerebral artery occlusion (MCAO), compared to their contralateral (unlesioned) hemisphere controls. CD45 hi /CD11b + /Ly6G − cell proliferation was not shown in the contralateral due to their very low to undetectable presence. (B) Graph presents quantification of microglia/macrophage proliferation measured by flow cytometry. Data indicate a marked reduction in CD45 low /CD11b + /Ly6G − cell proliferation and a modest but significant reduction in CD45 hi /CD11b + /Ly6G − cell proliferation in the ipsilateral hemisphere of CX3CR1 −/− mice compared to WT mice 72 hours after MCAO. P < 0.0001 for genotype and localization, P = 0.0017 for interaction by two-way analysis of variance. *P < 0.05, **P < 0.01 by Bonferroni post-hoc tests. n = 4 per group.
immunohistochemically assessed by stain for lipid peroxidation with 4-HNE and damaged DNA with 8-OHdG ( Figure 5C). The number of stained cells in the CX3CR1 −/− mice was notably less compared with WT mice (P < 0.01) ( Figure 5D).

CX3CR1 deficiency impairs inflammatory signaling in microglia and macrophage in ischemic brain
To determine whether CX3CR1 deficiency is associated with changes in the expression of inflammatory mediators produced by activated macrophages and microglia, ELISA was used to screen injured brain homogenates for differences in cytokine production. Consistent with previous reports [25][26][27], expression of a subset of inflammatory cytokines (IL-1β, IL-6, and TNF-α) was increased after MCAO regardless of genotype. In WT mice, the amounts of IL-1β, IL-6, and TNF-α were 160.9 ± 8.4, 56.5 ± 6.1, 253.0 ± 22.9 pg/mg in the ipsilateral hemisphere and 80.7 ± 4.1, 19.2 ± 2.2, 101.9 ± 15.3 pg/mg in the contralateral hemisphere, respectively.

Discussion
The work presented here provides important in vivo evidence for the role of the CX3CL1/CX3CR1 signaling pathway in the activation and neurotoxicity of microglia/ macrophage in cerebral ischemia. Using CX3CR1 −/− mice, in which the CX3CR1 gene is substituted with the gene for the GFP on both alleles (CX3CR1 GFP/GFP ), and a murine model of transient brain ischemia, we find that CX3CR1 deficiency resulted in a decrease in the size of ischemic lesion, a decrease in the number of apoptotic cells (predominantly neurons), marked deregulation of post-ischemic brain inflammatory responses (including ROS and pro-inflammatory cytokines), a significant decrease in proliferation of microglia and infiltrating macrophages in the ischemic lesion, and a marked decrease in the levels of IL-1β/IL-6/TNF-α expressed by microglia/macrophage in ischemic brain. Collectively, these effects due to CX3CR1 deficiency correlate with improved neurological function following MCAO and suggest that blockade of CX3CR1/CX3CL1 signaling may provide neuroprotection against ischemic injury.
Absence of fractalkine receptor in CX3CR1 −/− mice has been previously reported to be protective in ischemia. Critically, significantly smaller (56%) infarcts and blood-brain barrier damage have been observed in CX3CR1 −/− animals compared with CX3CR1 +/− and WT mice 72 hours after a 60-minute MCAO [15]. CX3CR1 deficiency has also been shown to induce protection from a 30-minute MCAO starting as early as 24 hours [17]. In an experimental model of spinal cord injury, CX3CR1 −/− mice demonstrate neuroprotection and functional recovery 5 days post-injury [16]. In agreement with these previous studies on the role of the CX3CL1/CX3CR1 pathway in acute CNS injury, our work finds that mice lacking CX3CR1 are protected from ischemic injury 72 hours after a 90-minute MCAO. Interestingly, we find prevention of exacerbation of ischemic lesion was accompanied with reduced cleaved Caspase 3 positive neurons in the peri-infarct area at 72 hours after initial stroke in CX3CR1 −/− mice compared to WT mice, inconsistent with a previous report [15]. While these observed differences could be attributed to the differences in the experimental designs, the outcomes between our study and others are identical. Further, it may also imply that the effects of CX3CL1/CX3CR1 signaling may be dependent upon the context of the specific disease state. CX3CR1 deficiency has been suggested to attenuate tissue damage and improve recovery of function by reducing the recruitment and/or the activation of microglia and macrophages [15,[28][29][30]. We tested this hypothesis by isolating mononuclear cells from the injured brain of WT and CX3CR1 −/− mice, followed by quantifying CD45 + /CD11b + cells by flow cytometry. We find a significant decrease in CD45 + /CD11b + /Ly6Gmicroglia/macrophage in the ipsilateral hemisphere of CX3CR1 −/− mice vs WT mice, which was confirmed by immunohistochemical staining with Iba-1 in brain sections. To determine whether this was a result of suppressed microglia proliferation or decreased monocyte recruitment, the relative expression of CD45 has been used to differentiate resident microglia (CD45 low ) from recruited monocytes (CD45 hi ) [31]. CD45 low /CD11b + / Ly6Gcells (microglia) predominate in the contralateral side of both genotypes. However, by 72 hours postinjury, accumulation of both CD45 low (microglia) and CD45 hi (macrophage/activated microglia) CD11b + / Ly6Gcells dramatically decreased in the ischemic hemisphere of CX3CR1 −/− mice. Moreover, BrdU assays show that the proliferation of microglia (CD45 low ) is significantly impaired in CX3CR1 −/− MCAO mice. The proliferation of CD45 hi cells (including activated microglia) was also suppressed; however, not to the extent which could fully compensate for the decreased number of CD45 hi cells seen in CX3CR1 −/− MCAO mice compared to WT MCAO mice. We therefore speculate that the migration of monocyte-derived macrophages from the periphery was inhibited. These data together indicate that both microglia/macrophage proliferation and macrophage recruitment are attenuated by CX3CR1 deficiency within the MCAO model. Our results coincide with a previous study that demonstrated decreased leukocytes infiltration as a possible mechanism for the development of smaller infarcts in CX3CR1 −/− mice after a 60-minute MCAO [15]. Conversely, more CD45 hi cells were found accumulated in the injured spinal cords of CX3CR1 −/− mice. Although the role of this cell population is not clear (rather, this study revealed a distinct monocyte subset), CD11b + /Ly6C lo /iNOS + macrophages were associated with reduced neuropathology and enhanced functional recovery [16]. Indeed, it has long been thought that macrophage (microglia) are not a uniform cell population, such that specific environmental signals, including effector molecules, timing of activation, and degree of injury, may induce their different polarization states [32,33]. Therefore, phenotypically distinct monocyte populations appear to be associated with tissue pathology and repair in ischemia and, thus, warrant further study [16].
Previous reports indicated a marked increase in proinflammatory cytokine levels that peak 12 to 24 hours after ischemic injury [34]. Our results obtained from the WT mice are in agreement with these temporal dynamics of cytokines. As expected, an initial increase of TNF-α, IL-1β, and IL-6 was detected in the ipsilateral hemisphere after MCAO. However, contrary to the levels of proinflammatory cytokines observed in control samples, we show that the ablation of proliferating microglia cells and recruitment of monocytes, associated with CX3CR1 deficiency, resulted in a significant decrease in the levels of IL-1β, TNF-α, and IL-6, suggesting an important temporal deregulation of the proinflammatory cytokine response. Similar patterns of mRNA expression and temporal deregulations were observed in the NF-κB signaling pathway, a general marker of inflammatory response in cerebral ischemia [16]. Furthermore, using flow cytometry, we confirmed that the trends of proinflammatory cytokines in the ipsilateral hemisphere of the MCAO brain in CX3CR1 −/− mice are consistent with the levels secreted by CX3CR1 −/− microglia/macrophage isolated from MCAO brains. Therefore, it appears that microglia/macrophage may play an important role in the regulation/modulation of the proinflammatory responses in cerebral ischemia, and their functions are attenuated by CX3CR1 deficiency.
Variations of cell morphology or cell displacement are related to the specific activity of microglia/macrophage [32]. Following ischemia, activated microglia/macrophage can potentially exert either a protective or detrimental effect, suggesting that these cells may acquire different phenotypes belonging to the classical (M1) or to the alternative (M2) active status [35]. M1 activation is generally referred to as the pro-inflammatory and cytotoxic phenotype, characterized by nitric oxide, ROS, and proinflammatory cytokine production, while the M2 phenotype is an alternative activation state, associated with a fine tuning of inflammation, scavenging of debris, promotion of angiogenesis, tissue remodeling and repair [17,36]. Fumagalli and colleagues [17] reported that, in CX3CR1 −/− mice, protection from ischemia at early time points after injury is associated with a protective inflammatory milieu, characterized by the promotion of M2 polarization markers. We found that activated M1-like Iba-1 + cells, which have shorter and thicker processes and bigger cell bodies, were visualized in the WT brain section, while ramified M2-like Iba-1 + cells were predominantly located in the CX3CR1 −/− brain ( Figure 3A) with increased expression of M2 markers ( Figure 3E), implying that suppressed activation of microglia/macrophage by CX3CR1 deficiency may be responsible for the neuroprotective effects in the CX3CR1 −/− mice.

Conclusion
In summary, using a murine model of transient focal ischemia, we explored the consequences of CX3CR1 absence on microglia/macrophage proliferation/recruitment and on their activation after brain ischemia. When CX3CR1 signaling is abolished, neurological recovery is improved and neural pathology is reduced. Our data show that loss of CX3CR1 signaling attenuates the proliferation and inflammatory capacity of microglia/macrophages. Moreover, monocytes recruitment and/or the differentiation of newly recruited monocyte subsets are altered. Collectively, these data identify CX3CR1 as a putative therapeutic target for selectively reducing neurodestructive inflammatory cascades after stroke.