The chemokine receptor CCR5 in the central nervous system
Research highlights
▶ The chemokine receptor CCR5 is expressed in microglia, astrocytes and neurons. ▶ CCR5 expression can control the progression of certain infections in the CNS. ▶ Emerging evidence suggests a role of CCR5 also in neuroprotective mechanisms.
Introduction
Chemokines are small proteins (70–90 aminoacids), divided into four different groups according to the position of conserved cysteines (C) in their sequence: CXC or alpha, CC or beta, C or gamma and CXC3 or delta chemokines (Rossi and Zlotnik, 2000). The word “chemokine” is derived from their function as chemotactic cytokines, resulting in activation and migration of leukocytes toward sites of inflammation. This process is mediated by the interaction of chemokines with their respective chemokine receptor. There have been 50 chemokines and 20 chemokine receptors identified to date (Domanska et al., 2010).
Chemokine receptors belong to the seven transmembrane G-protein-coupled receptor (GPCR) family. Receptor nomenclature is derived from their binding of chemokines: CC chemokines bind to CC receptors (CCR), CXC chemokines to CXCRs, and similarly for C and CX3C receptors. Within one subfamily, usually more than one chemokine can activate a given chemokine receptor, but in a few cases specific receptor/ligand pairs have been documented (Domanska et al., 2010).
Expression and activation of chemokine receptors is mostly related to the control of leukocyte migration. However, chemokine receptors also participate in the regulation of many other physiological and pathological processes, such as angiogenesis or metastasis formation (Rossi and Zlotnik, 2000). Also, there is increasing evidence for a role of chemokine receptors in the central nervous system (Bajetto et al., 2001, Cartier et al., 2005, Rostene et al., 2007).
The CCR5 chemokine receptor is mainly expressed in memory and effector T-lymphocytes, NK cells, monocytes, macrophages and immature dendritic cells. In these cell types it regulates chemotaxis and cell activation (Balistreri et al., 2007, Mueller and Strange, 2004, Oppermann, 2004), through the interaction with the following β-chemokines: CCL3 (MIP-1alpha), CCL4 (MIP-1beta), CCL5 (RANTES), CCL8 (MCP-2), CCL11 (eotaxin), CCL14a (HCC-1), CCL16 (HCC-4) (Alexander et al., 2008).
In 1996, five groups from different laboratories published at the same time their studies showing that the chemokine receptor CCR5 is a major co-receptor for the entry of HIV (human immunodeficiency virus) into target cells (Alkhatib et al., 1996, Choe et al., 1996, Deng et al., 1996, Doranz et al., 1996, Dragic et al., 1996). The interest for CCR5 was sparked by the discovery that a particular allelic alteration in the human CCR5 gene provided protection against HIV infection. Indeed, it was observed that some individuals had remained uninfected in spite of multiple exposures to HIV (Clerici et al., 1992, Langlade-Demoyen et al., 1994, Rowland-Jones and McMichael, 1995). The careful analysis of the CCR5 gene in isolated CD4+ T cells led to the discovery of a 32 base pair deletion (CCR5Δ32), which results in a truncated protein (215 instead of 352 amino acids), thought to be devoid of intrinsic activity (Samson et al., 1996b). Indeed, the deletion spans nucleotides 794–825 corresponding to the second extracellular loop of the receptor, which precludes membrane insertion of the mutant protein (Liu et al., 1996). Individuals carrying both alleles of this mutation do not show any particular phenotype, appear healthy and are significantly protected from HIV, due to the virus not being able to access its co-receptor CCR5 for cellular entry. Although heterozygous individuals can be infected by HIV, the progression of infection is slower and their viral load is decreased (Liu et al., 1996, Samson et al., 1996b). These discoveries have focused research on the mechanisms of activation, signalling and function of this particular chemokine receptor, motivated by the potential discovery of drugs targeting CCR5 for use as HIV therapies.
Despite the beneficial effect of CCR5-deficiency in the case of HIV infection, other consequences of the absence of CCR5 have not been fully elucidated, especially concerning possible adverse effects in the central nervous system (CNS). Indeed, increasing evidence indicates the expression of chemokines and their receptors in the CNS. It appears that chemokines are involved in brain development, neuronal signalling and synaptic transmission (reviewed in Bajetto et al., 2001, Cartier et al., 2005, Domanska et al., 2010, Rostene et al., 2007). Similarly, immunohistochemical and in vitro studies have revealed a constitutive expression of CCR5 in the CNS; however its function in the brain is still poorly understood. Furthermore, recent studies have proposed that a non-functional CCR5 receptor could influence response to CNS pathologies and even confer a certain vulnerability to the progression of particular diseases.
In this review, we summarize the present data concerning the distribution and the physiological functions of CCR5 in the CNS, and its involvement in neurological diseases.
The CCR5 gene is located on chromosome 3p21 (Liu et al., 1996) and consists of 4 exons and only 2 introns. Exon 4 contains the open reading frame, but a complex splicing in the 5′-UTR and in the 4 exons leads to multiple CCR5 transcripts (Mummidi et al., 1997). The gene codes for a protein consisting of 352 amino acids, with a resulting molecular mass of 40.6 kDa (Samson et al., 1996a). Transcription is regulated by two different promoters, which are upstream and downstream of exon 1 (Mummidi et al., 1997). The downstream promoter includes the ‘intronic’ region between exons 1 and 3. It contains a pair of consensus TATA elements and potential binding sites for several transcription factors, such as STAT, NF-κB, AP-1, NF-AT, and CD28RE (Liu et al., 1998). The transcriptional regulation of the CCR5 gene seems to be cell specific and there are data indicating that NFκB subunits (Bream et al., 1999, Liu et al., 1998) and/or the cAMP/CREB pathways (Kuipers et al., 2008) are mainly involved. A strong homology has been described between the 5′-UTR of the human CCR5 and that of mouse and rat, suggesting similarly conserved regulatory regions. Concerning the coding sequence, human CCR5 has 82% homology with the mouse (Boring et al., 1996) and 98% with the rhesus (Margulies et al., 2001). Despite these differences, human β-chemokines CCL3, CCL4, and CCL5 show very similar IC50 values for the human rhesus and mouse CCR5 receptor (Saita et al., 2007).
By contrast, small organic molecule antagonists developed for the human CCR5 receptor (see also below) appear to be species-specific. This is true for TAK-779, an experimental compound, which shows similar affinity for rhesus and human CCR5, but only a very low affinity for the murine CCR5 receptor (Saita et al., 2007). In the same way, the CCR5 antagonist maraviroc, used for HIV therapy has at least 1400-fold lower affinity interaction for the mouse, rat and dog CCR5 (Pfizer, 2007).
Several allelic variants of the CCR5 gene have been identified in different populations (Ansari-Lari et al., 1997, Carrington et al., 1997). These mutations occur in both the coding sequence and the promoter region (Kostrikis et al., 1999, McDermott et al., 1998), and are able to alter either the extent of transcript expression (McDermott et al., 1998) or the functionality of the receptor (Blanpain et al., 2000). However, the most known and studied mutation is the deletion of 32 base pairs in the coding sequence; this polymorphism has a surprisingly high frequency in certain populations.
Approximately 10% of the Caucasian population is heterozygous for the CCR5Δ32 mutation and 1–2% is homozygous (Libert et al., 1998, Martinson et al., 1997). This allele displays a particular geographic distribution. In fact, it is common almost exclusively in Caucasians, with the highest allele frequencies present among North eastern European populations (Finnish, Icelandic and Mordvinian around 16%), and the lowest in the South of Europe.
The origins of this mutation and the reasons for its particular geographic distribution have been the subject of many studies. Multiple evidences indicate that the CCR5Δ32 allele probably arises from a unique mutation event (Galvani and Novembre, 2005, Libert et al., 1998). It has been further hypothesized that this mutation occurred in Scandinavia, among Viking populations. Subsequently it has been disseminated in the rest of Europe during their expansion in the eighth to the tenth centuries. Indeed, areas of highest frequencies coincide with the territories, where the Viking influence was significant (Lucotte, 2001, Lucotte and Dieterlen, 2003).
Another hypothesis proposes that the mutation originally generated among Uralic-speaking populations (corresponding grossly to actual Baltic countries) (Balanovsky et al., 2005). In any event, the evidence suggests that the mutation has originated recently, and has been affected by natural selection. Indeed, it has been estimated that without any selection pressure, the mutation would have appeared approximately 125,700 years ago. However, studies have indicated that the mutation is more recent, and in spite of limited accuracy, the origin should be between 700 and 7000 years BP (Hummel et al., 2005, Libert et al., 1998, Sabeti et al., 2005, Slatkin and Bertorelle, 2001, Stephens et al., 1998). If this theory is true, it means that the mutation frequency has been enhanced in ancestral Caucasian populations by a natural selective factor. These evaluations also explain the reason why the CCR5Δ32 is not present among Africans and most Asian populations. In fact, it has been estimated that Europeans segregated from Asian populations approximately 35,000 years ago, and from African populations around 120,000 years ago (Cavalli-Sforza et al., 1988).
The rapid evolution of the presence of the mutation in the European population appears to be due to the fact that it may have conferred a selective advantage against some agent, such as a pathogen (Galvani and Novembre, 2005). Several positive selection criteria have been proposed, including the resistance to smallpox and plague infections. Indeed, a catastrophic event such as the bubonic plague which devastated millions of people during the Middle Age certainly had the potential of applying such selective pressure. Although attractive, this hypothesis has not been confirmed (Galvani and Slatkin, 2003, Mecsas et al., 2004). Another recent provocative theory suggests that the CCR5Δ32 allele was more frequent before Roman invasions, but had conferred increased vulnerability towards pathogenic agents or zoonoses carried by the Roman legions (Faure and Royer-Carenzi, 2008). However, so far, the real causes of the spread or decrease of the CCR5Δ32 frequency is still a matter of debate.
In 1998, CCR5-deficient mice were generated by targeted deletion of the ccr5 gene (Zhou et al., 1998). Similar to humans, these mice do not show a spontaneous phenotype. However, they have been extensively studied in a large number of disease models. Results obtained in the central nervous system are described in detail below and in Table 3.
The most evident consequence of CCR5-deficiency in mice is a decreased host defence against certain types of infections associated with more severe outcomes (Dawson et al., 2000, Khan et al., 2006, Luangsay et al., 2003, Machado et al., 2005, Zhou et al., 1998). In contrast, a reduced inflammation due to the absence of CCR5 has been described to be beneficial in other pathological conditions, such as hepatic fibrosis (Seki et al., 2009), pulmonary emphysema (Bracke et al., 2007, Ma et al., 2005) or atherosclerosis (Braunersreuther et al., 2007, Potteaux et al., 2006, Zernecke et al., 2006).
The chemokine receptor CCR5 is expressed in memory and effector T-lymphocytes, NK cells, monocytes, macrophages and immature dendritic cells (Balistreri et al., 2007, Mueller and Strange, 2004, Oppermann, 2004). Upon binding of ligands, various cascades are initiated which regulate CCR5 signalling (Fig. 1) and cell surface expression.
Signalling: The dissociation of the G protein heterotrimer into α and βγ subunits results in the inhibition of adenylyl cyclase (AC) by the α subunit (Zhao et al., 1998), and activation of phospholipase Cβ (PLCβ) and phosphoinositide 3-kinase (PI-3K) by the βγ subunits. This latter mechanism (βγ subunit dissociation) is principally responsible for chemotaxis (Neptune and Bourne, 1997). Specifically, Gβγ subunit activation of PLCβ results in the production of diacylglycerol (DAG) and inositol-1,4,5-triphosphate (InsP3), ultimately releasing intracellular Ca2+ and activating protein kinase C (PKC). The release of intracellular Ca2+ stimulates proline rich tyrosine kinase 2 (PYK2), which is responsible for cell motility and migration, as well as for activation of the mitogen-activated protein (MAP) kinases ERK1/2 (extracellular signal-regulated kinase), p38 and JNK (c-Jun N-terminal kinase), which are important for T-cell proliferation and expression of cytokines (Dairaghi et al., 1998, Del Corno et al., 2001, Ganju et al., 1998, Kraft et al., 2001, Misse et al., 2001, Wong et al., 2001). Activation of PI-3K by Gβγ subunits leads to stimulation of protein kinase B (PKB/Akt) (Burgering and Coffer, 1995), which is involved in cell survival (Downward, 2004). Similarly, Rho GTPase activation can occur in this way, thereby regulating the structuring of the actin cytoskeleton, as well as cellular adhesion, polarity and motility (Oppermann, 2004).
Cell surface expression: The regulation of CCR5 expression involves desensitization, internalization and recycling or degradation of the receptor. These mechanisms are initiated by the phosphorylation of the receptor/ligand complex in a heterologous manner by PKC, or by homologous phosphorylation through G protein coupled receptor kinases (Oppermann, 2004). The phosphorylated complex then binds to the regulatory proteins β-arrestins, which initiate internalisation of the CCR5 receptor by way of clatherin coated pits (Goodman et al., 1996, Laporte et al., 1999). Endocytosis of the receptor/ligand complex can however occur independently of a β-arrestin/clatherin pathway (Kraft et al., 2001). Once the ligand is dissociated or degraded, CCR5 proteins are recycled by perinuclear endosomes and appear at the cell membrane again, dephosphorylated (Pollok-Kopp et al., 2003, Signoret et al., 1998).
Although current antiretroviral therapies have proven effective, resulting in the long term reduction of viral RNA to undetectable levels with ongoing long-term treatment, rapid viral rebound occurs in patients when treatment is stopped. This phenomenon has been attributed to viral reservoirs. In particular the CNS and gut associated lymphoid tissues have been identified as the main HIV-1 viral reservoirs (Chun et al., 2008, Finzi et al., 1999, Poles et al., 2006, Walker et al., 2008). In addition, drug toxicity and viral resistance resulting from the essential long term use of antiretrovirals have prompted the search for possible alternative therapies for HIV. During the last few decades drugs targeting the host (CCR5 co-receptor), rather than the virus have been developed. These CCR5 antagonists hinder HIV adherence to host target cells through prevention of interaction between the HIV gp-120 glycoprotein and the host CD4+ T cell and macrophage CCR5 receptors. There have been four CCR5 antagonists developed which have been of importance: aplaviroc (AK602 and GSK-873140, GlaxoSmithKline), maraviroc (Celsentri®, Pfizer), vicriviroc (SCH 417690 and SCH-D, Schering-Plough) and cenicriviroc (TBR-652 formerly TAK-652, Tobira therapeutics Inc.) (Emmelkamp and Rockstroh, 2007, Klibanov et al., 2010). Aplaviroc is no longer being developed due to the discovery in 2005 of patients developing severe hepatic toxicity after administration of the drug (Emmelkamp and Rockstroh, 2007). The remaining three compounds are in different stages of development. Currently, maraviroc is the only CCR5 antagonist available on the market and was approved by the FDA in 2007. Vicriviroc successfully completed phase II clinical trials and is undergoing phase III tests, while cenicriviroc is currently undergoing phase II clinical trials in the US and Argentina (Emmelkamp and Rockstroh, 2007, Klibanov et al., 2010).
As mentioned above, the CNS plays a major role as a viral reservoir for HIV-1 and HIV-1 related neurological disorders are well recognised in patients (see section below). Therefore, it is relevant to understand to what extent CCR5 antagonists can permeate the blood brain barrier and be distributed in the CNS. In this context, it is also crucial to identify the possible side effects of CCR5 antagonists, in particular on neurological functions (Husstedt et al., 2009, Stephenson, 2007, Telenti, 2009). Only few data exist in literature concerning the distribution and effects of maraviroc and vicriviroc in the CNS. This is briefly discussed below:
Maraviroc: Following intravenous infusion in rats, only 10% of the free plasma concentration of maraviroc was found in cerebrospinal fluid (CSF) and 25% in the brain. This limited distribution seems to be due to the affinity of the compound for the efflux transporter P-glycoprotein (Walker et al., 2008). Similarly, low concentrations of maraviroc were found in the CSF of HIV-1-infected patients. However, this low concentration of maraviroc still reaches approximately its EC90, which would result in effective viral inhibition in the CSF (Melica et al., 2010, Tiraboschi et al., 2010, Yilmaz et al., 2009).
Vicriviroc: Good CNS penetration was reported for vicriviroc (Tagat et al., 2004). Due to concerns about the potential of vicriviroc to prolong the QT/corrected QT (QTc) interval, an investigatory study was recently performed. The results showed no QTc prolongation and no adverse cardiac or epileptogenic effects in healthy individuals receiving dosages higher than those necessary for HIV therapy (O’Mara et al., 2010).
An alternative approach to CCR5-based therapy of HIV would be a combination of gene and cell therapy. Indeed, a transplantation of CCR5-deficient bone marrow has been recently performed in a HIV patient with promising results (Hutter et al., 2009). Thus, there is emerging interest in CCR5 gene knock-down in bone marrow cells, which might then be used for autologous transplantation. Such an approach would avoid knock-down or inhibition of CCR5 in neural cells.
Section snippets
Distribution
The expression of CCR5 has been detected in total brain samples, but also specifically localized in several brain regions (Table 1).
Immunohistochemical analyses have revealed that CCR5 can be constitutively expressed not only in microglia cells, but also in astrocytes and neurons of human, baboon, macaque, rat and mouse samples (Table 2). However, it appears that not all the neurons of a given brain region express CCR5, but only subpopulations of neuronal cells (Westmoreland et al., 2002).
Conclusions and perspectives
The perception of CCR5 and its ligands has undergone extensive changes over the last decade. While initially considered a purely immune-inflammatory system, involved in leukocyte trafficking and pathogen clearance during infections, it is now becoming increasingly clear that its biological role is much broader. As discussed in the review, this is particularly true for its roles in the central nervous system.
Upon the discovery that the homozygous deletion of CCR5 is relatively frequent in humans
Conflict of interest
The authors declare no conflict of interest.
Acknowledgment
The work was funded by grants from the Swiss National Foundation no. 125115 (Division Biologie et Médecine) to KHK.
References (216)
- et al.
Aging sensitizes mice to behavioral deficits induced by central HIV-1 gp120
Neurobiol. Aging
(2008) - et al.
Brain-derived neurotrophic factor modulates expression of chemokine receptors in the brain
Brain Res.
(2008) - et al.
Chemokines and their receptors in the central nervous system
Front. Neuroendocrinol.
(2001) - et al.
CCR5 deficiency decreases susceptibility to experimental cerebral malaria
Blood
(2003) - et al.
The CCR5 deletion mutation fails to protect against multiple sclerosis
Hum. Immunol.
(1997) - et al.
Multiple nonfunctional alleles of CCR5 are frequent in various human populations
Blood
(2000) - et al.
Molecular cloning and functional expression of murine JE (monocyte chemoattractant protein 1) and murine macrophage inflammatory protein 1alpha receptors: evidence for two closely linked C-C chemokine receptors on chromosome 9
J. Biol. Chem.
(1996) - et al.
Cellular expression of functional chemokine receptor CCR5 and CXCR4 in human embryonic neurons
Neurosci. Lett.
(2001) - et al.
Novel alleles of the chemokine-receptor gene CCR5
Am. J. Hum. Genet.
(1997) - et al.
Chemokine receptors in the central nervous system: role in brain inflammation and neurodegenerative diseases
Brain Res. Brain Res. Rev.
(2005)