Human CX3CL1 was first cloned in 1997 and the mouse homolog in 1998. CX3CL1 is a relatively large protein consisting of an amino-terminal domain, a mucin-like stalk attached to a transmembrane region that connects the molecule to the plasma membrane, followed by the intracellular domain. CX3CL1 is biologically active either as a membrane-bound protein or as a soluble protein upon proteolytic cleavage from cell membranes. CX3CL1 actions are mediated through interaction with its unique G-protein coupled receptor CX3CR1 (previously called chemokine receptor CKRBRL1, RBS11, or V28). Both CX3CL1 and CX3CR1 are highly abundant in central nervous system (CNS) tissues; CX3CL1 is produced by neurons and CX3CR1 is present on microglial cells. In the periphery, CX3CL1 is produced mostly by endothelial cells and CX3CR1 is expressed on peripheral leukocytes. CX3CL1 plays a role in chemotaxis, cell adhesion, and cellular activation. Notably, CX3CL1 and CX3CR1 have been implicated in modulation of microglial activation and are involved in a wide spectrum of biological functions in various tissues and pathologies.
CX3CL1 Expression Patterns
CX3CL1 and its unique receptor CX3CR1 are abundant in CNS tissues. Unlike most chemokines, CX3CL1 is expressed as a transmembrane protein present predominantly in neurons and on the surface of endothelial cells of selected tissues including heart, lung, kidney, and pancreas. Both human coronary artery and umbilical vein endothelial cells (ECs) express CX3CL1 mRNA and protein; however, CX3CL1 mRNA has not been detected in brain vasculature in healthy mice or upon autoimmune inflammation (Sunnemark et al. 2005). Treatment of human umbilical vein endothelial cell cultures with recombinant CX3CL1 caused proliferation, migration and tube formation suggesting an effect on angiogenesis. In HUVEC cells CX3CL1 induced phosphorylation of ERK, Akt, and eNOS as well as an increase of nitric oxide (NO) production (Lee et al. 2006), indicative of the activating effects upon CX3CR1 engagement.
Proteolytic cleavage is essential for regulation of the available pools of membrane-bound versus soluble CX3CL1, and in turn for defining its adhesive versus chemotactic properties. In the healthy CNS, low levels of CX3CL1 are detected in a constitutive fashion. Upon inflammation, both CX3CL1 and its receptor are upregulated in the inflamed CNS tissues during EAE (Huang et al. 2006) as well as in several other pathologies (Cardona et al. 2006; Savarin-Vuaillat and Ransohoff 2007). More specifically, TNF significantly increased CX3CL1 mRNA (>100-fold) and protein expression, which was associated with increased shedding of CX3CL1 from the cell membranes (Hurst et al. 2009). CCL2 also appeared to regulate CX3CL1 functions by regulating receptor expression. More specifically, monocytes stimulation with CCL2 increased their adhesion to immobilized CX3CL1, in a p38 MAPK dependent pathway (Green et al. 2006). Although in vitro studies suggest that lipopolysaccharide (LPS) can regulate expression of both CX3CR1 and CX3CL1 (Mizuno et al. 2003), the exact signaling pathways and players involved in the transcriptional and posttranslational regulation of CX3CL1 by cytokines in vivo remain to be clarified.
Release of CX3CL1 from cellular membranes is mediated by the alpha-secretase activity of two disintegrins and metalloproteases (MMPs), ADAM10 (Gough et al. 2004) and ADAM17 (also known as tumor necrosis factor converting enzyme, TACE) (Garton et al. 2001). ADAM10 mediates constitutive and ionomycin-induced cleavage of CX3CL1, whereas ADAM17 catalyzes the inducible release of the soluble domain of CX3CL1 upon phorbol myristate acetate (PMA) stimulation. Most recently it was demonstrated that cathepsins S and MMP2 are also involved in CX3CL1 shedding. Of interest is the report of CX3CL1 release by apoptotic lymphocytes, providing the first demonstration of chemokine/chemokine-receptor activity in the mobilization of macrophages toward apoptotic cells, suggesting a mechanism by which macrophages infiltrate tissues containing apoptotic lymphocytes (Truman et al. 2008).
Dendritic cells (DCs) also express CX3CL1, which functions to control NK cell activation (Pallandre et al. 2008). Recently, a new role of CX3CL1 receptor in DCs differentiation was reported (Lyszkiewicz et al. 2011). Using competitive adoptive transfer experiments it was demonstrated that CX3CR1 expression on hematopoietic progenitors promotes development of DCs and myeloid cells. Interestingly, this model was performed transferring precursor cells into non-irradiated and irradiated recipients to delineate the real role of CX3CR1 under steady-state conditions, compared to inflammatory conditions induced by irradiation (Chen et al. 2007; Xun et al. 1994). It is proposed that irradiation might have a negative effect on the contribution of CX3CR1 signaling to the generation of DCs and monocytes/macrophages (Auffray et al. 2009; Lyszkiewicz et al. 2011). Based on the above, the role of CX3CR1 is mainly restricted to the steady-state generation of myeloid cells. Additionally, competitive adoptive transfer experiments were developed transferring CX3CR1-deficient precursors directly into the lymphoid target organs such as spleen and thymus, demonstrating similar competitive disadvantage when compared to CX3CR1-sufficient precursor cells. These data implicate CX3CR1 in developmental processes rather than homing of developing cells to lymphoid organs.
In mice the gene encoding CX3CL1 is located in chromosome 8. In humans, CX3CL1 is clustered on chromosome 16q13 together with monocyte derived chemokine (MDC, CCL22) and TARC (CCL17) (Hiroyama et al. 2001). The three human genes are separated by 6.3- and 28.5-kb intervening sequences, respectively, and reside in a head-to-tail orientation with respect to each other (GenBank acc. no. AC004382).
The CX3CL1 receptor (CX3CR1) is a membrane-bound protein and member of the G-protein coupled receptor family mainly produced by microglia. Outside the CNS, CX3CR1 resides mainly on specific leukocyte populations (Jung et al. 2000) including monocytes, NK cells, dendritic cells, and subpopulations of T cells.
CX3CL1/CX3CR1 in Microglial Function
Microglia are the resident macrophage of the CNS, which under resting conditions, constantly survey their microenvironment supporting their critical role in CNS homeostasis (Nimmerjahn et al. 2005). Under inflammatory conditions, microglial cells become activated and upregulate CD45, MHC and co-stimulatory molecule expression, phagocytic activity and are able to stimulate the proliferation of Th1 ( IFN-gamma-producing) and Th2 (Interleukin-4-producing) CD4+ T cell lines (Carson et al. 1998). Activated microglia are responsible for the removal of cellular debris and pathogens during CNS injury, inflammation, and infections. In vitro studies have suggested that phagocytosis of myelin by microglia and macrophages could enhance neuroinflammation based on the release of pro-inflammatory cytokines and nitric oxide by these cells (Williams et al. 1994). Additionally, it has been suggested a dual role for microglial phagocytosis based on the need of myelin debris removal for remyelination (Neumann et al. 2009; Chastain et al. 2011).
Meucci et al. in 2000 showed for the first time the neuroprotective functions of CX3CL1 using hippocampal neurons. CX3CL1 conferred neuronal protection from HIV-1 gp120 induced neurotoxicity. Subsequently it was demonstrated that CX3CL1 inhibited release of pro-inflammatory mediators by microglia both in vitro and in vivo. Two knockout mouse models for CX3CL1 and four for CX3CR1 have been reported. These mice develop normally, and under healthy conditions, microglia of CX3CR1-deficient mice did not differ phenotypically from wild type microglia. However upon inflammation or neurodegeneration a distinct phenotype linked to defective CX3CR1 signaling was revealed. In response to systemic lipopolysaccharide injections Cx3cr1− / − microglia became highly activated and released pro-inflammatory mediators such as IL-1 that correlated with neuronal damage. Detrimental effects of CX3CR1-deficiency were also observed in models of Parkinson’s disease, amyotrophic lateral sclerosis(Cardona et al. 2006), and most recently the absence of CX3CR1 was found to modulate beta-amyloid deposition in Alzheimer’s disease models (Lee et al. 2010). Therefore, CX3CR1 appeared to be a key molecule involved in modulation of microglial activation. Interestingly, in a model of transient focal ischemia, Cx3cl1−/− mice showed a 28% reduction in infaction size and lower mortality rate when compared to wild type mice.
Recently, it has been found that soluble CX3CL1 (sCX3CL1), released from mouse cortical neurons damaged by glutamate excitotoxicity, enhances phagocytic uptake of neuronal debris by microglia (Noda et al. 2011). In this study, phagocytosis-related factors expressed by microglia in response to sCX3CL1 such as the PS receptor MFG-E8, was identified as the opsonin involved in clearance of apoptotic cells (Fuller and Van Eldik 2008). It was found that the enhanced phagocytosis and the clearance of damaged neurons by CX3CL1-treated microglia in turn promoted neuronal survival. Another finding was the activation of intracellular signaling pathways by sCX3CL1, such as ERK and JNK MAPK, being the last one an important signaling to drive the expression of the antioxidant enzyme heme oxygenase-1 (HO-1) by microglia via Nrf2 nuclear translocation factor (Noda et al. 2011). Therefore, CX3CL1 plays an important neuroprotective role signaling through CX3CR1 on microglial cells.
CX3CL1/CX3CR1 in Peripheral Leukocytes
In peripheral blood CX3CR1 helps to distinguish “resident” monocytes (identified as LFA-1+, L-Sel–, Ly6C–, CCR2 – , CX3CR1 + ), whereas CCR2 marks the “inflammatory” monocyte subset (LFA-1–, L-Sel+, Ly6C+, CCR2 + , CX3CR1 – ) (Geissmann et al. 2008). During EAE, monocyte populations mirror this peripheral pattern (Saederup et al. 2010). It was demonstrated that CCR2 expression is controlled at a posttranscriptional stage in some Ly6Clo monocytes and NK and T cells. Furthermore, Ly6Chi monocytes fail to enter the CNS of CCR2-deficient mice during EAE and are mainly replaced by granulocytic cells, not Ly6Clo monocytes. These findings suggest that CCR2+ Ly6Chi monocytes initiate and maintain neuroinflammatory responses, while tissue remodeling is mediated by CX3CR1+ microglia. These studies provide further evidence of differential expression of CX3CR1 and CCR2 to distinguish monocyte/macrophage subsets and discriminate between resident microglia and infiltrating macrophages and DCs in the brain (Saederup et al. 2010; Mizutani et al. 2011).
Notably, multiple sclerosis (MS) patients show lower expression of CX3CR1 in peripheral NK cells when compared to healthy controls (Infante-Duarte et al. 2005) and a deficiency of CX3CR1+ PBMC correlated with disease activity. It was also shown the existence of distinct NK cell phenotypes depending on the magnitude of CX3CR1 expression (Hamann et al. 2011). The NK cell phenotypes defined by CX3CR1 expression differ in their cytolytic activity, cytokine profile, proliferative response, and their impact on monocyte functionality. Moreover, three different stages of NK cell maturation were identified based on the expression of CX3CR1 and CD56. While the CX3CR1neg CD56bright phenotype is exclusively characteristic for immature NK cells, the magnitude of CX3CR1 expression on CD56dim NK cells discriminates between intermediary CX3CR1neg CD56dim and fully mature CX3CR1high CD56dim NK cells. Hence, these studies demonstrates that CX3CR1 can be used in conjunction with CD56 and with other novel maturation markers to delineate NK cell phenotypes characteristic for the sequential stages of human NK cell maturation.
Human CX3CR1 in Chronic Inflammation
The studies using CX3CR1-deficient mice become of particular relevance for humans as two single nucleotide polymorphisms in the human CX3CR1 loci produce four allelic receptor variants. Most individuals carry CX3CR1V249/T280, and CX3CR1I249/M280 is present in >20% of the population. These changes decrease CX3CL1 affinity and correlate with protection from atherosclerosis (Nassar et al. 2008; McDermott et al. 2001, 2003; Moatti et al. 2001) making CX3CR1 a potential attractive target for therapeutic intervention in cardiac disease. Contrasting this protective effect in atherosclerosis, variant M280 receptor was reported to correlate with enhanced susceptibility to age-related macular degeneration (Chan et al. 2005). Interestingly, CX3CR1 clears its ligand from circulation and tissues acting as a scavenger receptor. Therefore, CX3CL1 elevation could possibly function as a biomarker to assess CX3CR1 dysfunction. CX3CL1/CX3CR1 have also been implicated in the pathology of autoimmune diseases such as systemic lupus erythematosus and patients with neuropsychiatric involvement showed higher serum levels of CX3CL1 (Yajima et al. 2005). Similar findings were reported in rheumatoid vasculitis and osteoarthritis patients (Klosowska et al. 2009).
CX3CL1/CX3CR1 in Cancer
Binding of CX3CL1 to CX3CR1 present on prostate cancer cells led to activation of anti-apoptotic signaling pathways, thereby enhancing survival and persistence of malignant cells. Therefore, CX3CR1 might support tumorigenic responses which translate into worse prognosis. In addition CX3CR1 signaling on tumor cells mediated cytotoxicity of malignant cells against endothelial cells expressing membrane-bound CX3CL1 and therefore enhancing pathology. CX3CR1 on tumor cells from patients with pancreatic ductal adenocarcinoma (PDAC) mediated chemotactic migration of PDAC cells toward CX3CL1 in vitro, as well as adhesion to neural cells expressing the ligand CX3CL1. High CX3CR1 expression was associated with perineural invasion and with earlier local tumor recurrence in PDAC patients correlating with worse prognosis (Marchesi et al. 2010). In a separate context, combination of CX3CL1 and IL2 gene transfer into tumor cells was shown to reduce tumor size and liver metastasis in an animal model of neuroblastoma. Similarly, expression of CX3CL1 on the surface of tumor cells, via a chimeric immunoglobulin-chemokine construct, reduced incidence and size of lymphoma (Lavergne et al. 2003). These results suggest that the oncogenic effects of CX3CL1 toward tumor cells vary depending on the predominant expression of CX3CL1 or the receptor by the malignant cell type.
CX3CL1 is a unique chemokine that plays important roles in myeloid cells and in CNS microglia. CX3CL1 functions go beyond the original proposed role of chemokines in cellular recruitment. CX3CL1/CX3CR1 are key modulators of microglial function, play important roles in the development of myeloid cells and expression levels appeared altered in inflammatory conditions including autoimmune inflammation, atherosclerosis, and cancer. CX3CL1/CX3CR1 provides an example of neuronal/microglial communication. Due to its unique peripheral pattern of expression, this chemokine/receptor pair might confer a CNS/peripheral communication system whose tight functions are now recognized as critical for myeloid cells and its effect in other cell types are yet to be defined.
- Klosowska K, Volin MV, Huynh N, Chong KK, Halloran MM, Woods JM. Fractalkine functions as a chemoattractant for osteoarthritis synovial fibroblasts and stimulates phosphorylation of mitogen-activated protein kinases and Akt. Clin Exp Immunol. 2009;156(2):312–9.CrossRefPubMedPubMedCentralGoogle Scholar
- Mizutani M, Pino A, Saederup N, Charo I, Ransohoff RM, Cardona AE. The fractalkine receptor but not CCR2 is present on microglia from embryonic development throughout adulthood. J Immunol. 2011. (in press).Google Scholar
- Nassar BA, Nanji AA, Ransom TP, Rockwood K, Kirkland SA, Macpherson K, et al. Role of the fractalkine receptor CX3CR1 polymorphisms V249I and T280M as risk factors for early-onset coronary artery disease in patients with no classic risk factors. Scand J Clin Lab Invest. 2008;68(4):286–91.CrossRefPubMedGoogle Scholar
- Xun CQ, Thompson JS, Jennings CD, Brown SA, Widmer MB. Effect of total body irradiation, busulfan-cyclophosphamide, or cyclophosphamide conditioning on inflammatory cytokine release and development of acute and chronic graft-versus-host disease in H-2-incompatible transplanted SCID mice. Blood. 1994;83(8):2360–7.PubMedGoogle Scholar