Richard M. Ransohoff, MD, Cleveland Clinic, Neuroinflammation Research Center, Department of Neurosciences, Lerner Research Institute, 9500 Euclid Avenue, Cleveland, OH 44195, USA. Tel: +1-216-444-0627 Fax: +1-216-444-7927 Email: email@example.com
Chemokines and their receptors play crucial roles in the trafficking of leukocytes and are of particular interest in the context of the unique inflammatory responses elicited in the central nervous system (CNS). The chemokine receptor CCR2 and its ligand CCL2 have been implicated in a wide range of immunobiological processes and neuropathologies, including recruitment of monocytes and regulation of bone marrow homeostasis, as well as multiple sclerosis, HIV-associated dementia, Alzheimer’s disease and neuropathic pain. Recently, powerful biological tools (CCR2-red fluorescent protein [RFP] knock-in mice) have been developed to analyze the functions of CCR2 in different cell populations, and intriguing results have emerged from those mice. The present review emphasizes CCR2/CCL2 as a key chemokine/chemokine receptor pair that controls the recruitment or retention of a key subset of mononuclear phagocytes in inflammation associated with host defense or disease. (Clin. Exp. Neuroimmunol. doi: 10.1111/j.1759-1961.2011.00024.x, January 2012)
Recent advances in immunobiology have enabled elucidation of many of the mechanisms underlying inflammation associated with host defense and immunopathologic disease. It is clear that monocyte infiltration into lesion sites is of great importance during the development of inflammatory disease, particularly chronic diseases. Recruitment of the monocyte into lesion sites is dependent on both adhesion molecules and the chemokine–receptor axis. In the periphery (tissues other than the CNS), atherosclerosis,1 rheumatoid arthritis,2 insulitis in type 1 diabetes mellitus,3 and adipose tissue inflammation in metabolic syndrome,4 among others, have been related to the monocyte infiltration of tissues. In addition, in the CNS, the infiltration of immune cells is closely associated with the primary pathology of both neuroinflammatory and neurodegenerative diseases.5 The CNS is an immune-privileged environment (where de novo immune responses involving the priming of naïve T cells with antigen does not occur) because of the lack of resident dendritic cells.6 However, recent studies have revealed that immune mechanisms, including innate, adaptive, and regulatory processes, are seen to at least some extent in every CNS disorder, including those with intact blood–brain barrier (BBB) integrity. Chemokines and chemokine receptors play important roles in linking systemic inflammation and CNS inflammation, and this may be a critical factor in exploring the development of therapeutic strategies.7–9
In the present review, we discuss the basic functions of CCR2, the major chemokine receptor expressed mainly on monocytes, and its involvement in both systemic and CNS inflammatory disease.
CCR2: structure, expression and associated ligands
Within the human genome, there are 3257 genes that are predicted to encode G-protein-coupled receptors (GPCRs). Human GPCRs consists of five subfamilies: rhodopsin like (type A), secretin like (type B), metabotrophic receptor (type C), ocular albinism proteins, and the Frizzled/Smoothened family. Chemokine receptors comprise a large branch of the class A rhodopsin-like family of cell surface GPCRs with seven transmembrane domains (GPCR-Ligand Database [GLIDA]: http://pharminfo.pharm.kyoto-u.ac.jp/services/glida/, accessed 20 June 2011; SEVENS database, http://sevens.cbrc.jp/, accessed 20 June 2011). Type A receptors are characterized by the presence of the DRY (Asp–Arg–Tyr) motif in the second intracellular domain, which provides a docking site for heterotrimeric G proteins10 (Fig. 1). Twenty-one chemokine receptors (which bind chemokines with high specificity and affinity and also exhibit signaling) and 53 corresponding chemokines have been characterized. Chemokines contain four positionally conserved cysteines and are divided into four families based on the organization of the first two cysteines near the N-terminus: C-X-C, C-C, C, and C-X3-C.11,12 Charo et al.11 discovered two specific receptors for monocyte chemoattractant protein (MCP)-1 (termed CCL2 in the systematic nomenclature) on human monocyte cell line and named them CCR2A and CCR2B (the receptors for chemokines are designated CCRs or CXCRs according the family of bound chemokines and then assigned numbers in the order of their cloning and characterization11). Wong et al.13 determined that the CCR2 gene was comprised of three exons spanning approximately 7 kb of genomic sequence. The CCR2 gene is highly conserved in mammals, including chimpanzees (Pan troglodytes), monkeys (Macaca mulatta), cows (Bos taurus), horses (Equus caballus), pigs (Sus scrofa), cats (Felis catus), dogs (Canis lupus familiaris), chickens (Gallus gallus), mice (Mus musculus), and rats (Rattus norvegicus).14–16 In human, the open reading frame is contained on two alternatively spliced exons that encode two distinct polypeptides that are 360 amino acids (CCR2A) and 374 amino acids (CCR2B) in length.11 The difference between CCR2A and CCR2B is mainly in the C-terminal of CCR2, which is important for receptor trafficking to the cell membrane. The carboxyl tail of CCR2A may cause retention of the entire CCR2 receptor domain in the cytoplasm.13 The C-terminal tail also contains serine and threonine residues susceptible to phosphorylation. Phosphorylation of these sites by GPCR receptor kinase (GRK) is important for receptor internalization and desensitization.17
The CCR2 protein consists of seven hydrophobic transmembrane domains linked by three intracellular and three extracellular loops, an extracellular N-terminus, and a cytoplasmic C-terminal tail (Fig. 1). The N-terminal extracellular region is the most important part for binding ligands with high affinity. The other three extracellular regions are also important for triggering further intracellular reactions.18,19
As mentioned above, there are two alternatively spliced forms of CCR2 activating different signaling pathways: CCR2A and CCR2B. The dominant form is CCR2B, which accounts for approximately 90% of CCR2 expression on the cell surface.11,13 The mRNA for both CCR2A and CCR2B is detectable in monocytes, blood-derived dendritic cells (DC), natural killer (NK) cells, and T lymphocytes, but not in resting neutrophils or eosinophils12 (Table 1). In addition, CCR2 mRNA has been detected in endothelial cells under inflammatory conditions20 and in lung fibroblasts.21 The expression of CCR2 protein is more restricted than that of the mRNA; new information from CCR2-RFP knock-in mice has shown that protein expression is seen only in basophils and monocytes, despite the presence of CCR2 message in lymphocytes and NK cells.22 Reports of CCR2 protein expression by CNS neurons, astrocytes, or endothelial cells in vivo using immunohistochemistry could not be confirmed by CNS tissue analysis of CCR2-RFP mice.22 Therefore, at a first approximation, the biology of CCR2 relates primarily to the action of its ligand towards hematopoietic cells during host defense, inflammation, and immunity.
Table 1. Subsets of leukocytes that express CCR2 mRNA and surface immunoreactive protein (as determined by flow cytometry assays)
Immunoreactive CCR2 protein
✓, confirmed; ?, unconfirmed or inconsistent; h, human; m, mouse.
✓ (h), ? (m)
B lymphocytes (immature)
Natural killer cells
Dendritic cells (immature)
The monocyte is the first cell type that was established to express CCR2. Blood monocyte expression of CCR2 is constitutive and uniform. However, when monocytes differentiate into macrophages (usually accompanied by cell activation), surface expression of CCR2 is downregulated.11,23 In addition, CCR2 expression helps distinguish two non-overlapping blood monocyte subpopulations, initially characterized as CX3CR1loCCR2+GR1+ inflammatory monocytes and CX3CR1hiCCR2−GR1−“resident” monocytes.24 The anti-Gr1 monoclonal antibody (mAb) recognizes both Ly6C and Ly6G isoforms and Ly6G is expressed on neutrophils, whereas Ly6C is expressed on neutrophils and subsets of monocytes, DC, NK cells, and lymphocytes.25 Therefore, the classification of monocytes is able to help establish that Ly6ChiCX3CR1loCCR2+CD62L+“inflammatory” monocytes readily infiltrate into inflammatory sites, whereas Ly6CloCX3CR1hiCCR2−CD62L−“resident” monocytes are less effective at transmigration, patrol the endothelial lumen scanning for inflammatory or pathogen-associated signals, and also repopulate resident tissue macrophages.26 These subsets correspond to two major human monocyte subsets, namely the CD14+CD16− and CD14loCD16+ monocytes, respectively, which share phenotype and homing potential with the mouse subsets.24
There are five known members of the human MCP family, which are designated CCL2, CCL8, CCL7, CCL13, and CCL16 and correspond to MCP-1, MCP-2, MCP-3, MCP-4, and human CC chemokine-4 (HCC-4), respectively. There are also four members in the mouse, namely Ccl2 (MCP-1), Ccl8 (MCP-2), Ccl7 (MCP-3), and Ccl12 (MCP-5). (Note, were needed, we distinguish between human and mouse chemokines by using uppercase letters for humans [CCL2 etc.] and lowercase letters for mice [Ccl2 etc.].) The human MCP-1, -2, and -3 are orthologs of mouse MCP-1, -2, and -3, respectively. No mouse ortholog has been detected for human MCP-4, and mouse MCP-5 has no ortholog in humans (Table 2). Other chemokines, such as MCP-2 (CCL8),27 MCP-3 (CCL7),28 MCP-4 (CCL13),29 HCC4 (CCL16),30 and mouse MCP-5 (Ccl12),31 also signal to CCR2.
Table 2. CCR2 ligands and their receptors
h, human; m, mouse; DARC, Duffy antigen receptor for chemokines; MCP, monocyte chemoattractant protein; HCC, human CC chemokine.
There are other important molecules that bind to CCL2, namely Duffy antigen (Duffy antigen receptor for chemokines; DARC) and D6. DARC, which binds promiscuously to many CC and CXC chemokines, is mainly expressed by erythrocytes and endothelial cells (ECs). This seven-transmembrane glycoprotein has a similar structure to GPCRs, but lacks the DRY motif, which is necessary to generate signals via G-proteins. In a proposed nomenclature, DARC is termed chemokine-internalizing (pseudo) receptor (CIPR-1) and has an important role in controlling tissue and plasma concentrations and the distribution of chemokines. Once a chemokine binds to DARC/CIPR-1 abluminally on ECs, the chemokine is internalized and transcytosed, and CIPR-1 supports its luminal immobilization on glycosaminoglycans (GAGs) and presentation to rolling leukocytes. Conversely, CIPR-1 on erythrocytes buffers plasma chemokine levels. D6, another pseudo-receptor for chemokines (CIPR-2), is mainly expressed by lymphatic ECs, where it binds approximately 12 CC chemokines. Persistence of chemokines in inflamed tissues impairs DC trafficking to lymph nodes and leads to tissue injury. CIPR-2 controls ambient tissue concentrations of chemokines mainly by internalizing and degrading its ligands.32,33
CCR2: signal transduction and regulation
The main ligand of CCR2 is CCL2 (MCP-1). Despite difference in amino acid sequence, chemokines adopt a common three-dimensional structure, known as the chemokine fold: a short unstructured N-terminal region and an extended N-loop region followed by three β-strands stabilized by intercysteine disulfide bonds and an α-helix (Fig. 1b).34
CCL2 forms dimers and higher-order oligomers in solution, which are associated with GAGs and contribute to ligand localization and signaling.35 Upon receptor engagement with basic residues of CCL2 oligomers associated with GAGs, an N-loop region of the ligand binds to the N-terminus of CCR2, whereas another undefined region binds to an extracellular loop of CCR2. Then, the short N-terminal region of the ligand binds to the intermembrane region of CCR2 (Fig. 1, inset). These last bindings are weaker than the first tethering between the N-loop of CCL2 and the N-terminus of CCR2, but are important to initiate G-protein signaling via CCR2. These reactions are not uniform for all chemokine receptors: ligand binding and activation of CCR1 is dependent upon the third extracellular loop despite high amino acid homology with CCR2.18,19 The N-terminal region of CCR2 is also responsible for low-efficiency infection of CCR2+ cells with macrophage-tropic HIV-1.36
Binding of the N-terminal segment of CCL2 to CCR2 induces dissociation of GDP from Gαi and replacement of GDP by GTP from the cytosol. The Gαi–GTP complex dissociates from the receptor and from the Gβγ heterodimer, and both complexes activate downstream physiological responses (Fig. 2). Gαi inhibits adenylate cyclase, whereas Gβγ activates phospholipase C to generate diacylglycerol and inositol 1,4,5-trisphosphate, leading to the release of intracellular calcium and activation of protein kinase C.37 The Gβγ subunits also activate phosphatidylinositol 3-kinase and protein kinase B, induce actin polymerization and pseudopod formation, and lead to monocyte migration.38
Activated CCR2 interacts directly with FROUNT, a unique clathrin heavy-chain repeat homology protein, and forms clusters at the cell front during chemotaxis.39 Interactions between CCR2 and FROUNT are necessary for homing of bone marrow-derived multipotent mesenchymal stem cells to the injured heart.40 In addition, FROUNT regulates directional migration of cultured human osteosarcoma cells via CCR5.41
CCR2 is expressed at the cell surface variably and under stringent regulation (Table 3). As one example, CCR2 mRNA is upregulated by stimulation with interleukin (IL)-2 in monocytes,42 NK cells,43 and T cells.44 IL-15, one of the Th1 cytokines, like IL-2, also upregulates CCR2 mRNA in T lymphocytes.45 Low-density lipoprotein and free cholesterol induce CCR2 protein expression on the cell surface of monocytes.46 IL-4 upregulates CCR2 protein expression in cultured lung fibroblasts in vitro.21 However, as noted above, CCR2 mRNA is not invariably translated to protein. In mouse T cells and NK cells, CCR2 mRNA is highly expressed (corresponding to the presence of the CCR2-RFP reporter), but no detectable CCR2 protein is present, either on the plasma membrane or in the cytoplasm.22
Table 3. CCR2 receptor regulation with different stimuli
Monocyte CCR2 protein expression and mRNA abundance are downregulated following stimulation with lipopolysaccharide, without reduction of nuclear transcription but with markedly reduced mRNA half-life47 in monocytes/macrophages48 and NK cells.42 The antioxidant pyrrolidine dithiocarbamate also rapidly inhibits CCR2 mRNA expression on human monocytes by decreasing transcript stability49 and tumor necrosis factor (TNF)-α downregulates both CCR2 mRNA and cell surface protein in human monocytes.50 Human monocytic THP-1 cells treated with IL-1α showed rapid downregulation of CCR2 mRNA and protein, reversible within 48 h.51 Conversely, in human umbilical vein endothelial cells (HUVEC) treated with IL-1β or TNF-α, CCR2 protein on the cell surface was upregulated.20 CCR2 downregulation by interferon (IFN) γ is a more contentious topic. For example, IFNγ has been shown to upregulate CCR2 protein expression in vitro in lung fibroblasts,21 but to downregulate mRNA expression on monocytes52 and to have no effects on neutrophils.53 Taken in the context of our present understanding of CCR2 protein expression, the effects of IFNγ on monocytes seem most likely to be physiologically relevant. In addition, the expression of both CCR2 mRNA and surface proteins differs in each cell population receiving the same stimulation.
Receptor internalization is an additional mechanism for regulating the reaction of chemokine receptors to ligands. When CCR2 is activated by CCL2 or other ligands, G-protein receptor kinases (GRKs) phosphorylate serine (Ser)/threonine (Thr) residues in the carboxyl tail and intracellular loop of CCR2, leading to the recruitment of β-arrestins, which promote internalization via the adaptor-related protein 2 (AP2)–clathrin complex and also serve as scaffolds for signaling intermediates17,54,55 (Fig. 2). Knockdown of clathrin or treatment with the dynamin inhibitor dynasore impair CCR2 internalization.56
Function of CCR2 in peripheral disease pathologies
Recruitment of monocytes
Monocytes are produced from hematopoietic stem cells in the bone marrow (BM). Inflammatory signals from the periphery cause monocytes to egress from the BM into the peripheral blood (by serum CCL2) or from the blood into draining lymph nodes (by the remote effect of chemokines released in lymph57), where they are recruited to sites of inflammation. Figure 3 shows schematic pictures of this migration.
In CCR2-deficient mice, there were no obvious deficits in growth, development, or fertility.58 Recruitment of Ly6Chi monocytes requires signals mediated by CCR2.59 Evaluation of monocyte mobilization from the BM showed that MCP-3 (CCL7) and, to a lesser extent, MCP-1 (CCL2) are the CCR2 agonists most critical for the release of monocytes from the BM to maintain the blood monocyte population.60 When mice are infected orally with Toxoplasma gondii, the inflammatory responses to the parasite rapidly recruits monocytes that establish a defensive wall within the villi of the ileum in the small intestine.61 Mice deficient for either CCR2 or CCL2 failed to recruit Ly6Chi inflammatory monocytes, resulting in amplified parasite burden and extensive intestinal necrosis with robust infiltration by neutrophils. When CCR2+ inflammatory monocytes are transferred to CCR2-deficient mice, they migrate to the infected site and alleviate disease progression. This was not the case in CCL2-deficient mice: CCR2+ inflammatory monocytes transferred to CCL2-deficient mice did not enter sites of infection and disease progressed.61 The number of BM monocyte progenitor cells is increased in CCR2-deficient mice, indicating they are unable to exit the BM efficiently.61
When CCR2-deficient mice are challenged with intraperitoneal thioglycollate, the recruitment of peritoneal macrophages is decreased, as also seen in CCL2-knockout mice.22,62 When CCR2-deficient mice are crossed with apolipoprotein E (ApoE)-knockout mice, atherosclerotic lesions markedly decrease.1 Similar results are observed in CCL2-deficient mice.63 The atheromatous lesions shrink when ApoE-deficient mice are treated with anti-CCL2 gene therapy.64 However, there are likely receptors other than CCR2 that respond to CCL2, at least on some cells: aortic smooth muscle cells from CCR2−/− mice responded to physiological concentrations of CCL2 and express tissue factor (TF) mRNA and protein.65 TF is a principal initiator of the clotting cascade.65 In an acute skeletal muscle injury model, CCR2-dependent recruitment of monocytes plays an important role in promoting muscle repair. In CCR2-deficient mice with muscle injuries, there is less infiltration of both Ly-6C+ and Ly-6C− subsets of monocyte populations, which are responsible for scavenging of debris (mainly Ly-6C+ population) and the expression of insulin-like growth factor (IGF)-1 for muscle repair (mainly the Ly-6C− population).66
Regulation of bone homeostasis through osteoclasts
CCR2-deficient mice have high bone mass owing to a decrease in the number, size, and function of osteoclasts and are resistant to osteoporosis induced by ovariectomy. These results suggest that CCR2 is involved in bone homeostasis through the regulation of the development and behavior of osteoclasts.67 Estrogen inhibits mouse Ccl2 mRNA expression, but progesterone has no effect on the expression of Ccl2 mRNA.68 If the lack of inhibition of CCL2 by estrogen contributes to post-menopausal osteoporosis, the CCL2–CCR2 axis could become one of the targets of osteoporosis therapy.
Scavenging ligands to control levels of chemokines
Chemokine levels in the serum and tissue are under strict control. Recent results indicate a scavenging role for CCR2 in vivo in CCL2 clearance.69 Similar to other chemokine receptor-knockout mice, such as CX3CR1-, CXCR2-, and CXCR3-knockout mice, CCR2-deficient mice exhibit significantly increased amounts of circulating ligand compared with wild-type mice.69 Given the complexity of chemokine/chemokine receptor biology and the ability of multiple ligands to signal CCR2, high levels of some circulating CCR2 ligands, such as Ccl3, act on alternative CC receptors, including CCR1. In particular, CCR2-deficient peripheral blood and resident peritoneal cells exhibit reduced binding capacity and biologic responses to the CCR1 ligand Ccl3, suggesting that elevated levels of the CCR2 ligand Ccl3 had downregulated CCR1.69
CCR2 in brain pathologies
In the healthy CNS, hematopoietic cells reside in restricted regions. Macrophages reside only in the perivascular space, meninges, and choroid plexus.70 Microglia comprise the only hematopoietic cell population in the CNS parenchyma. Remarkably, the origin of mouse monocytes/macrophages in the periphery and microglia in the CNS is different. Microglial progenitors were born between embryonic day (E) 7 and E7.5 in the yolk sac, whereas the precursors of peripheral monocytes/macrophages were born on E7.5 or later in the yolk sac, the aorta–gonad–mesonephros (AGM) region, and in the fetal liver.71 Only the precursor cells born on E7–7.5 were found in brain rudiments and, thereafter throughout life, microglia are replaced by local proliferation, not from the periphery or BM. In contrast, monocytes in the blood and tissue/organ macrophages arise from the BM after birth. The CNS microglia express CX3CR1 and have not been shown to express CCR2 in vivo;22,72 CCR2-positive cells in the CNS parenchyma enter under pathological conditions. Thus, the chemoattractant functions of CCR2 are exerted only under pathological circumstances.
Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease. Inflammation plays an essential role in the pathogenesis of MS. In MS autopsy sections, hypertrophic astrocytes primarily express CCL2 within active lesions and periplaque white matter.73 Maximum staining for CCL2 in chronic active MS lesions is seen on the inner rim of active lesion edges adjacent to the hypocellular lesion center and on the outer rim of the hypercellular leading edge.74 CCL2 is expressed within hypertrophic astrocytes in non-inflamed submeningeal regions near chronic active lesions.75 This “doughnut”-like expression of CCL2 has also been described in a CCL2 transgenic model.76 The desensitization of CCR2 with binding to CCL2 may result in the inability of monocytes to enter the maximal expression site of CCL2.23 Significantly decreased levels of CCL2 in the cerebrospinal fluid (CSF) in MS, clinically isolated syndromes (CIS), and acute disseminated encephalomyelitis (ADEM) compared with control patients with other inflammatory and non-inflammatory neurological diseases have also been reported and proposed to be due to CCL2 being consumed by infiltrating cells.77–82 The reduced levels of CCL2 in the CSF appear to be relatively specific to CNS inflammatory demyelinating diseases, as opposed to other CNS inflammatory diseases.79
In rodents, MS can be mimicked by inducing experimental autoimmune encephalomyelitis (EAE) after immunization with a myelin peptide and adjuvant.83 However, CCR2-deficient mice are relatively resistant to EAE induction: early studies reported the inability to induce EAE in C57BL/6 mice that lacked CCR2.83 That group used mice that were 12–16 weeks of age and immunized with 100 μg MOGp35–55, which was sufficient to induce EAE in wild-type, but not CCR2-mutant, mice.84 Recently, we immunized wild-type and CCR2-deficient C57BL/6 mice with MOGp35–55 and found that more than one-third of immunized CCR2-deficient mice developed signs of EAE, whereas nearly 90% of immunized CCR2-positive mice developed signs (R. Yamasaki et al., unpubl. data, 2010). The peak scores of EAE were lower in CCR2-deficient mice than in wild-type EAE mice, and onset was delayed. These results are compatible with those of a previous report,22 in which lesions were smaller and the infiltrated cell numbers were lower in CCR2RFP/RFP-knockin mice (without any CCR2) than in mice with functional CCR2.
Markers such as F4/80, CD11b, and Iba1 have been used to identify the macrophages in inflamed CNS lesions. However, these markers cannot distinguish monocytes from microglia.85
Because there are no CCR2-positive cells found in normal CNS parenchyma and only microglia express CX3CR1 in the CNS, the CCR2-RFP reporter signal can be used to distinguish the infiltrated cells (CCR2-RFP)22 from resident microglia (CX3CR1-GFP)72 in the inflamed CNS lesion. In CCR2-RFP/CX3CR1-GFP mice with EAE, RFP+ monocytes and activated green fluorescent protein (GFP)-positive microglia have very similar morphology and CD11b staining.22 Nevertheless, microglia and monocytes were easily discriminated because of the GFP and RFP signals (Fig. 4). CCR2-RFP+ cells infiltrated more in CX3CR1GFP/+CCR2RFP/+, which have normal CX3CR1 and CCR2 function and are also labeled by GFP and RFP, than CX3CR1GFP/+CCR2RFP/RFP, which have normal CX3CR1 function without CCR2 function (R. Yamasaki et al., unpubl. data, 2010). From these experiments, it appears that CCR2 affects the development and severity of EAE, but is not indispensable for the induction of EAE.
Another study reported susceptibility to EAE in multiple mouse strains (C57/J129, C57BL/6 and BALB/c) using a high dose of antigen.86 These studies showed altered CNS-infiltrating cell populations in EAE lesions of CCR2−/− mice, and our recent data also showed the predominance of neutrophils infiltration instead of monocytes in CCR2−/− mice with EAE.22 Infiltration of neutrophils in inflammatory lesions of CCR2−/− mice is a common feature of several disease models.61,86,87 Adoptively transferred, primed T cells isolated from CCR2−/− mice induce EAE in CCR2+/+ mice, but primed T cells from wild-type mice transferred to CCR2−/− mice failed to induce EAE.88 Similar results have been obtained in CCL2-deficient mice, which were also relatively resistant to EAE, due to deficient monocyte recruitment to the CNS.89 These results suggest that the effective recruitment of monocytes is dependent on the CCL2–CCR2 axis, and that the blockade of this axis leads to the alleviation of immune-mediated disorders.
In summary, the role of CCR2 in demyelinating disease in CNS inflammation seems related to the recruitment of monocytes to inflammatory sites from the BM. If there is no functional CCR2, monocytes cannot exit the BM and the rapid infiltration of monocytes from the blood stream to lesions is also impaired. The altered environment of the lesion induced by the lack of robust monocytic infiltration leads to an increase in other inflammatory myeloid cell populations, associated with late onset and milder signs of disease.
HIV-associated dementia (HAD) is believed to be a mononuclear phagocyte (MP)-mediated disorder of AIDS patients.90 Both CD4+ T cells and MPs are targets for HIV-1 infection and MPs in particular serve as “Trojan horses” for the dissemination of virus in multiple organs, including the brain. CCL2 inhibits HIV-1 infection of MPs in vitro, but the mechanisms involved are unknown. A promoter polymorphism of CCL2 (MCP-1 −2578G allele) had 50% reduced acquisition of HIV-1 but, after infection, disease progression accelerated and there was a 4.5-fold increased risk of HAD.91 Interestingly, the −2578G allele is associated with increased CSF levels of CCL2, which is a predictive biomarker for HAD.92 Conceivably, recruitment of HIV-infected monocytes to the CNS is involved (CSF monocytes are >70% CCR5-positive cells, which are readily infected with HIV primary isolates). Increased CCL2 is also an indicator of reactive astrocytosis because astrocytes are the major CNS cellular source of CCL2.93
Alzheimer’s disease (AD), the most common neurodegenerative disorder of the elderly, is characterized by the presence of senile plaques of β-amyloid (Aβ), particularly in the cortex and hippocampus.94 In AD brains, macrophages and activated microglia accumulate around Aβ deposits95 and may be implicated in the clearance and processing of Aβ before the formation of visible Aβ deposits.96 Indeed, the elimination of such accumulation, as occurred in the AD model of Tg2576 mice with CCR2 deficiency, led to the acceleration of early disease progression and was associated with increased mortality in these mice, possibly due to amyloid angiopathy.97 In the Tg2576 mouse brain, accumulation of Aβ correlated with Ccr2 gene dosage, suggesting that the clearance of Aβ in the early stages of the disease is important for controlling disease progression.97 In support of a partially protective role for macrophages and microglia, overexpression of IL-1β in the brain of amyloid precursor protein/presenilin1− (APP/PS1−) mice markedly induced CCL2 expression in the brain, therefore increasing monocyte accumulation and decreasing AD-like pathology.98 In contrast, microglial ablation in two distinct APP transgenic mouse strains had no effect on the accumulation of amyloid plaques or neuritic dystrophy.99 Therefore, it remains contentious whether monocytes, macrophages, and microglia play positive, negative, or neutral roles in the processing, deposition, and clearance of Aβ from the brain. In contrast, perivascular macrophages clear the Aβ along the perivascular spaces without activated microglia or astrocyte.100
Multiple roles in pain
CCR2-deficient mice are resistant to neuropathic pain101 and, remarkably, both inflammatory pain (elicited by injecting irritants) and neuropathic pain (induced by nerve injury) are about equally affected. Initial evaluation of affected wild-type mouse tissues demonstrated upregulation of CCR2 in the dorsal root ganglia (DRG) in the injured nerve and in the inflamed injection site.101 It seems plausible that pain pathways are modulated by the presence of CCR2+ inflammatory monocytes and that loss of CCR2 signaling ameliorates pain by dampening macrophage-mediated inflammation. Concordant results were obtained by studying transgenic mice that overexpressed CCL2 in the CNS under the control of the astrocyte-specific glial fibrillary acidic protein (GFAP) promoter.102 CCL2 was upregulated in DRG neurons after partial ligation of the sciatic nerve,103 suggesting monocytic involvement in neuropathic pain. However, the surprising demonstration that DRG neurons upregulated both the CCL2 ligand and the CCR2 receptor in the context of chronic nerve root damage complicates this interpretation and suggests autocrine effects towards neurons. Within DRG neurons, CCL2 was packaged in synaptic-like vesicles and its action at CCR2 led to pain-promoting responses, including sensitization of the transient receptor potential vanilloid 1 (TRPV1) ion channel.104–106 Electron microscopy confirmed that the CCL2 was released from glomerular boutons and secretory vesicles in the dorsal horn spinal cord (DHSC) and CCL2 mostly originates from TRPV1-positive nociceptive fibers.107 These investigators also succeeded in attenuating allodynia in rats using CCR2 antagonists.107 Together, these results indicate that CCR2 plays a complex, multifarious role in pain states, both through its expected function on inflammatory monocytes and via its startling and novel properties as an inducible neuromodulatory receptor on DRG neurons.101 Despite this evidence for inducible CCR2 expression in the rodent peripheral nervous system (PNS), previous reports of CCR2 expression by CNS neurons and astrocytes in vivo using immunohistochemistry or receptor autoradiography could not be confirmed by analysis of CNS tissues from CCR2-RFP mice.22
The expression of CCR2 on particular cell populations remains to be elucidated and studying CCR2-RFP mice will help clarify this cellular expression.
CCR2: therapeutic applications
Mononuclear macrophages are important for the pathogenesis of immune-mediated and inflammatory neurological disorders. CCR2/CCL2 is a key chemokine/chemokine receptor pair that controls the recruitment of mononuclear phagocytes in inflammatory diseases. According to accumulated preclinical data, CCL2/CCR2 is a promising target for the therapeutic reduction in inflammatory cell infiltrates and alleviation of tissue damage. Investigations into the effects of the administration of CCR2 antagonist are in the early phases of clinical trials in both peripheral systemic inflammatory conditions and neuroinflammatory disorders, including rheumatoid arthritis (RA), atherosclerotic cardiovascular disease, metabolic syndrome, IgA nephropathy, and MS.108–110 However, to date there has been no definitive proof-of-principle for anti-CCR2 therapeutics. For example, Merck identified the CCR2 inhibitor MK-0812, which inhibited chemokine binding with an IC50 of 5 nmol/L.111 MK-0812 then entered Phase II clinical trials for both RA and MS, but failed to show any improvement in the treatment of either RA or MS.111 Other companies are also testing several compounds, but, until recently, there have note been so many favorable results.111 This slow progress may spring from several reasons: the nature of the chemokines themselves (small amounts of chemokine could still have ample effect to activate the receptor), differences in the immune systems between humans and animals, the complexity and redundancy of chemokine/chemokine receptor interactions, and our limited understanding of complex chronic inflammatory diseases.109 Innovative technologies that may assist in effectively targeting CCL2/CCR2 in neuroinflammation include the use of interfering RNA to silence specific genes, RNA oligonucleotides, or dominant negative mutants of chemokine ligands to eliminate receptor function, but their successful application will require new knowledge and hard work.112
Understanding the mechanisms that govern the expression and regulation of chemokines and their receptors is crucial in defining the potential of chemokine receptor blockade for therapeutic purposes. Here we have emphasized the importance of CCR2 in monocyte trafficking to the inflamed CNS, where monocytes have widespread and varied effects. CCR2 and its ligands link immunobiological monocytes to neuropathological processes. CCR2 also has important roles in the development of peripheral inflammatory disorders, including RA, or metabolic disorders. Although the function of monocytes varies depending on the environment they infiltrate, unveiling the function of CCR2 on monocytes and other inflammatory cells may lead to the development of new therapies.
The Ransohoff laboratory has been supported by grants from the US National Institutes of Health, the Charles A. Dana Foundation, and the Nancy Davis Center Without Walls; and grants and fellowships from the National Multiple Sclerosis Society, the Robert Packard Foundation for ALS Research at Johns Hopkins University, Boye Foundation, Williams Family Fund for MS Research and ChemoCentryx, Inc. The authors thank Dr Chris Nelson for constructive critical review of the manuscript. In addition, the authors thank past and present members of the Ransohoff laboratory for their dedication and hard work. The authors apologize to those whose work was not cited due to space limitations.