https://www.abbs.info e-mail:[email protected] ISSN 0582-9879                                    ACTA BIOCHIMICA et BIOPHYSICA SINICA 2003, 35(9):779–788                                    CN 31-1300/Q

Mini Review

CC Chemokine Receptor-coupled Signalling
Pathways

NEW David C., WONG Yung H.*

( Department
of Biochemistry, the Molecular Neuroscience Center, and the Biotechnology Research
Institute, Hong Kong University of Science and Technology, Clear Water Bay,
Kowloon, Hong Kong, China )

Abstract       The
isolation and characterization of multiple CC chemokine receptors (CCRs) in a
wide range of tissues and cells signifies the functional diversity of CC
chemokines. The realization that multiple chemokines activate individual
receptors and that some chemokines are functional at several different CCRs,
indicates that interplay between a complex network of intracellular pathways is
required for the full expression of the physiological function of each ligand.
In different cellular environments, chemokines can regulate distinct second
messengers or even positively or negatively regulate the same signal
transduction pathway. The specific interactions between many signalling
molecules have been discerned in an increasing number of cellular systems and
this information is being used to explain the physiological actions of
chemokines. This review will attempt to summarize recent research by many
groups that has revealed numerous subtleties of the CC chemokine-coupled
signalling pathways.

Key words   CC
chemokine receptor (CCR); chemokines; chemotaxis; G protein; G protein-coupled
receptor (GPCR)

The chemokines are a family of proinflammatory cytokines
that act through cell surface receptors to regulate numerous routine
physiological and pathophysiological processes, including hematopoiesis, T-cell
activation, angiogenesis, inflammatory diseases as well as HIV-1
infection[1,2]. These small peptides are typically composed of 70－130 amino acids and are characterized by the presence
of two disulphide bonds formed between four conserved cysteine residues.
Chemokines are classified into four subfamilies according to the pattern of
conserved cysteines in their amino acid sequences. They include at least
twenty-six CC chemokines, seventeen CXC chemokines, two C chemokines and one CX3C
chemokine[3－5].

Chemokines exert their effects through at least
nineteen G protein-coupled receptors (GPCRs). The nomenclature of the chemokine
receptors follows the notation used for the chemokine subfamilies and they are
termed CCR1－10 (CC chemokine receptor 1－10), CXCR1－6, XCR1
and CX3CR1[4,6]. A remarkable feature of the chemokine receptors is their
relative lack of selectivity in ligand binding, with many chemokine receptors
binding more than one chemokine with high affinity. For example, eleven
chemokines are reported to bind to the CCR1 receptor, including MIP-1α (
macrophage inflammatory protein 1α), MIP-1β, MIP-1δ, RANTES (regulated on
activation normal T cell expressed and secreted), MCP-1 (monocyte chemotactic
peptide 1), MCP-2, MCP-3, MCP-4, Lkn-1 (leukotactin-1), MPIF-1 (myeloid
progenitor inhibitory factor 1) and HCC-1 (hemofiltrate CC chemokine 1), with
varying affinities and acting with different degrees of agonism. Similarly,
individual chemokines act as ligands for different receptors. For example,
MCP-3 acts as a ligand for CCR1, CCR2, CCR3 and CCR5[4,7]. This promiscuity and
the apparent redundancy of signalling that may arise poses many questions as to
the control of chemokine signalling in different tissues expressing different
combinations of chemokines, receptors and effectors.

Despite the apparent complexities of the chemokine signalling
systems, the importance of individual chemokine receptors is gradually emerging
from detailed studies on knockout mice, targeted gene disruption and the
application of specific chemokine antagonists[6]. As an example, CCR1 knockout
mice have been reported to have disordered trafficking and proliferation of
myeloid progenitor cells and to display impaired inflammatory responses to a
variety of stimuli. Control of the CCR1 signalling system was demonstrated to
have clinical significance as CCR1 knockout mice display significantly reduced
rejection responses to cardiac allografts[8]. This suggests that a strategy of
blocking CCR1 signalling pathways may be useful in preventing rejection of
transplanted tissues.

CCR5 has generated widespread interest because of its
role as a co-receptor for HIV[9]. The identification of a naturally occurring
mutant of this receptor, CCR5Δ32, and observations that homo and heterozygotes
for this mutant have increased resistance to HIV infection and the development
of AIDS[10] has highlighted the potential benefits to human health that could
accrue from controlling the ability of CCR5 to bind ligands.
The CC chemokine receptors (CCRs) are thought to predominantly signal via
Gi/o-coupled heterotrimeric G proteins to inhibit the activity of adenylyl
cyclase (AC) and to regulate Ca²⁺ flux, however there is increasing
evidence that CCRs are able to couple to a wider spectrum of G proteins to
potentially influence the activity of numerous intracellular signalling
pathways and gene transcription events. If we wish to exert a degree of control
over chemokine signalling it will be necessary to unravel the complexities of
promiscuity and apparent redundancy in chemokine expression, receptor binding
and signalling. In this review we will outline recent efforts to delineate the
intracellular pathways that define the CCR system.
The reader is directed to Table 1 for an overview of the signalling events
activated by the individual receptors. In addition, Fig.1 graphically
integrates the information presented. Throughout the text we shall identify
chemokine ligands by their common name but upon their first use we will also
include the official nomenclature[4].

Table 1  
CCR receptor activation of intracellular signalling pathways

The activities induced or
inhibited by individual CCRs are presented along with the G protein mediators
and the relevant tissues or cell types. Activites identified with non-selective
chemokines are not included. ND, no data available. KCa, Ca²⁺-dependent
K⁺ channels; Kir, G protein-gated inwardly rectifying potassium
channels.

Fig.1
Intracellular signalling pathways activated by CC chemokine receptors

The CC chemokine receptors are
grouped together according to their G protein-coupling specificity. For
descriptions of pathways modulated by individual CCRs, the reader should
consult the text and Table 1. The diagram identifies the heterotrimeric G
protein subunits that have been shown to couple with CCRs, as well as the
signalling pathways that these interactions modulate. Dashed lines indicate
that the role of intermediate proteins has not yet been elucidated. The experimental
evidence identifying individual pathways and the interactions between their
intermediates is described in the text.

1    Regulation of adenylyl
cyclase

Adenylyl cyclase
(AC) is an enzyme which catalyses the synthesis of cyclic adenosine
monophosphate (cAMP) from ATP. All nine AC isozymes are activated by
interaction with Gαs subunits. However, each type of AC isozyme exhibits
different sensitivity to Gαi, Gαz and Gαo. Type 1－6 ACs can be inhibited by Gαi1－3 and Gαz. However, GαoA/B only inhibit type 1 AC but not types 2, 4－6 AC. In addition to Gα subunits,
Gβγ subunits are also known to play important roles in the regulation of AC
activity. Type 1 AC is inhibited by Gβγ. On the other hand, Gβγ stimulates
types 2, 4 and 7 AC in the presence of activated Gαs subunits[11] or following
phosphorylation by protein kinase C (PKC)[12]. Thus activation of Gi/o-family
heterotrimeric proteins by CCRs may inhibit or activate AC depending on the
tissue-specific pattern of expression of receptors, G proteins and AC isozymes,
as well as the relative strength of the Gα and Gβγ-mediated signals.

In human
embryonic kidney 293 cells (HEK-293) stably expressing CCR1, challenging cells
with a variety of CCR1 ligands induces potent inhibition [EC50 values for
MIP-1α (also known as CCL3) and RANTES (CCL5) approximately 0.1 nmol/L] of
forskolin-stimulated cAMP accumulation[13]. CCR2a also inhibits AC activity
with equal potencies when cells are challenged with MCP-1 (CCL2). Furthermore,
the CCR2 mediated responses are sensitive to pretreatment of the cells with
pertussis toxin (PTX), indicating that the CCR2 mediated inhibition of AC
proceeds via a Gi/o-coupled pathway[13]. CCR2b similarly mediates inhibition of
AC in response to MCP-1[14].

CCR3-mediated
signalling events in induced eosinophil acute myelogenous leukaemia (AML) cell
lines are known to be sensitive to PTX, implying that Gi/o-coupled pathways
transduce CCR3 signals[15]. However, when the ability of eotaxin (CCL11), a
selective CCR3 agonist, to inhibit forskolin-stimulate AC activity was tested,
it was observed that CCR3 did not induce inhibition of AC in AML cells[16].
These cells do express CCR3 and respond to eotaxin in a number of physiological
and biochemical assays[16], suggesting to the authors that CCR3 is coupled to a
PTX-sensitive pathway that does not involve Gi proteins. However, considering
the previously discussed sensitivity of AC isozymes to Gαi/o subunits, it is
also possible that this cell type does not express AC isoforms that respond to
Gαi and/or Gαo. There are currently no reports of the inhibition of AC by CCR3
in heterologous systems. Although, it has been observed that eotaxin promotes
adhesion, aggregation and migration of interleukin-stimulated T lymphocytes,
and these CCR3-mediated events were completely blocked by a specific protein
kinase A (PKA) inhibitor, H89[17]. Even though cAMP levels were not directly
measured in the study, this indicates that CCR3 may induce increases in cAMP
levels in certain cell types and that the activity of the downstream pathways
activated by cAMP is required for CCR3-mediated cellular events to occur. It is
not known whether this occurs via activation of Types 2, 4 or 7 AC by Gβγ
subunits released from Gi/o heterotrimeric complexes to potentiate signals from
Gs-coupled receptors, or whether cAMP levels are increased by an as yet
undocumented interaction of CCR3 and Gs.

A similar
phenomenon may be operating in platelets, which express CCR4. In these cells,
the CCR4 selective ligand macrophage-derived chemokine (MDC) (CCL22) induces Ca²⁺
mobilization and platelet aggregation but does not inhibit the
prostaglandin-induced cAMP accumulation[18]. In hippocampal neurons, MDC
promoted accumulation of the cAMP responsive element binding protein (CREB) to
the nucleus[19], and in rat cerebellar slices MDC increased the cellular levels
of phosphorylated CREB[20]. This suggests that activation of CCR4 is also able
to promote increases in AC activity resulting in activation of downstream
effectors. However, it has been reported that chemokine-induced phosphorylation
of CREB is blocked by U0126, an inhibitor of the mitogen activated protein
kinase (MAPK) pathway intermediate MAPK/ERK kinase (MEK)[21], so it is not at
all clear whether CCR4 is able to modulate AC activity. It is known that CCR4
couples to Gi/o family proteins as the MDC-induced Ca²⁺ currents in
G1A1 cells expressing CCR4 cells are PTX sensitive[22].

Expression of
CCR5 in CHO or NG-108-15 cells results in cell sensitivity to the ligands
RANTES or MIP-1β (CCL4) in a PTX-dependent manner[23,24]. This study also noted
that overexpression of Gαi2 strongly potentiated the CCR5 mediated inhibition
of AC activity, providing further evidence that subunits of the Gαi family are
able to mediate CCR5 responses[23].

There are
currently no reports demonstrating the coupling of CCR6-10 to changes in
cellular cAMP levels either in endogenous or heterologous systems. However,
GTPγS incorporation assays combined with antibody capture techniques
demonstrated that liver and activation-regulated chemokine (LARC) (CCL20)
activation of CCR6 promotes guanine nucleotide exchange on Gαi, Gαo and Gαz
subunits and that Epstein-Barr virus-induced receptor ligand chemokine (ELC)
(CCL19) activation of CCR7 promotes exchange on Gαi subunits in
interleukin-2-activated natural killer (IANK) cells[25]. This coupling might be
expected to allow CCR6 and CCR7 to regulate AC activity, although as was
observed for CCR3 and CCR4 this may depend on the co-expression of these
receptors in cells expressing appropriate AC isozymes.
In cells endogenously expressing CC chemokine receptors, there are reports that
CC chemokines are able to both positively and negatively regulate the
AC/cAMP/PKA/CREB pathway. Unfortunately, many of the chemokines are specific
for several receptors and thus it is often not possible to be precise about
which chemokine receptor has been activated. For example, MIP-1α increases cAMP
levels in the human growth factor-dependent cell line M07e, whereas RANTES is
without effect[26]. Addition of a membrane soluble form of cAMP to the cell
growth medium was shown to mimic the growth suppressive effects of MIP-1α on
M07e cells. MIP-1α is a potent agonist for CCR1 and CCR5 and RANTES is a ligand
at CCRs 1, 3 and 5[4]. Coupled with a lack of information on the PTX responses
of the cells, it is unclear which receptor is acting through which G proteins
to modulate cAMP levels. Indeed, it is possible that the cAMP levels measured
arise from competing responses elicited by different chemokine receptors.
Supporting a role for CC chemokine-induced increases in cAMP is evidence that
the cell polarization and adhesion effects on T lymphoblasts induced by RANTES
and MIP-1α are mimicked by cAMP agonists and inhibited by PKA inhibitors[27].
The chemokine-induced effects are abolished by PTX pre-treatment of the cells,
establishing that Gi/o-coupled proteins, probably Gβγ subunits, are able to
promote cAMP formation, which results in physiological activation of the cells.

In contrast,
MIP-1α and RANTES both inhibit forskolin-stimulated cAMP formation in human
hepatoma and endothelial cells[28], and RANTES decreases cAMP levels and PKA
activity in human astrocytes[29]. In astrocytes, RANTES is a powerful
stimulator of the production of proinflammatory mediators and this property is
mimicked by the PKA inhibitor H89[29]. In human monocytes, the chemoattractant
properties of MIP-1α and RANTES are significantly inhibited by treatment with
forskolin and IBMX, both of which raise intracellular cAMP levels[30].

There are also a
number of virally-encoded proteins that have homology to chemokine receptors
and bind CC chemokine ligands. US28 is a GPCR encoded by human cytomegalovirus
(HCMV) with 30% amino acid identity to CCR1 and recognises several CCR1
ligands[4]. HCMV infection can cause vascular disease and this is dependent on
the expression of US28, which promotes the activation of a number of signal
transduction pathways, including increasing the transcriptional activity of
CREB[31]. US28 and the related HCMV-encoded CCR homologue UL33 also demonstrate
constitutive CREB activity in the absence of chemokines[32]. Expression and
signalling by HCMV-encoded chemokine receptors leads to smooth muscle cell
migration in response to CC chemokines[33]. Therefore, these receptors are
potential therapeutic targets for the treatment of HCMV-accelerated vascular
diseases.

2    Regulation of phospholipase
C-coupled pathways

In mammalian cells, the enzyme
phosphoinositide-specific phospholipase C (PLC) catalyses the hydrolysis of
phosphatidylinositol to inositol-trisphosphates (IP3) and diacylglycerol (DAG).
These second messengers trigger numerous downstream events, including the
mobilization of Ca²⁺ from intracellular stores and the activation of
PKC. Thus far, three subfamilies of PLC isozyme, designated as PLC-β, -δ, and
-γ, have been identified. Of these, only PLC-β(1－4)
isozymes are known to be activated by Gαq, Gα11, Gα14, Gα15, and Gα16 subunits.
In addition PLC-β2 and PLC-β3 can be activated by Gβγ subunits[34]. This
sensitivity to Gβγ subunits allows PLC-β to be stimulated by activation of Gi/o
heterotrimeric complexes. In the complex signal transduction networks involving
GPCRs there are numerous examples where Gi-linked receptors activate PLC via
Gβγ subunits (for example, reference [35]). In addition, the augmentation of
Gq/11-dependent signals by Gi/o-linked receptors has been described as a means
of PLC-β activation. This requires preactivation of PLC-β by a Gq/11-coupled
receptor along with Gβγ stimulation arising from Gi/o-coupled receptor
activation[36].

The activity of chemokine receptors is most commonly
assayed by measuring changes in intracellular Ca²⁺ levels following
application of agonists. Mobilization of Ca²⁺ from intracellular
stores indicates that PLC-β has been activated, although it does not define
which G protein family has transduced the chemokine message from receptor to
effector. However, increases in intracellular Ca²⁺ levels do not necessarily
prove that the PLC/IP3 pathway has triggered Ca²⁺ release from
intracellular stores. An alternative explanation would be the
chemokine-mediated opening of channels in the outer cell membrane allowing Ca²⁺
flux into the cell. This possibility has been demonstrated experimentally and
will be explored in this and subsequent sections. In addition, CC chemokines
are able to promote Ca²⁺ fluxes via the cyclic adenosine
diphosphate-ribose (cADPR)/ryanodine receptor pathway and the sphingosine
kinase/sphingosine 1-phosphate cascade[37].

CCR1 receptors transiently expressed in COS-7 cells do
not induce IP3 formation following challenge with MIP-1α, suggesting that CCR1
does not activate PLC-β pathways via endogenous Gq/11 family proteins, and that
endogenous PLC-β1 is not activated by Gβγ subunit released from Gi/o proteins.
Upon co-transfection with Gα14, but not Gα16, MIP-1α is able to induce IP3
formation. When COS-7 cells were co-transfected with CCR1 and PLC-β2, IP3
formation was induced by MIP-1α in a PTX sensitive manner[38], demonstrating
that in the presence of appropriate Gα subunits and effector isozymes CCR1 can
activate PLC-β via different pathways. In an HL-60 cell line endogenously
expressing several CCRs, the application of different chemokines with
overlapping but not identical selectivities identified a CCR1-mediated increase
in intracellular Ca²⁺ levels[39]. The nature of the transducing G
protein was not identified. Interestingly, it has been observed that the rise
in intracellular Ca²⁺ levels following activation of CCR1 expressed
in HEK-293 cells is completely inhibited by the PLC inhibitor U73122,
indicating that the Ca²⁺ is released from intracellular pools.
However, it was noticed that in the absence of extracellular Ca²⁺,
the intracellular levels induced by CCR1-mediated events were considerably
reduced[40]. This implies that CCR1 is able to promote the influx of Ca²⁺
into a cell by activating Ca²⁺ channels (see below) as well as by
promoting the release of Ca²⁺ from IP3-sensitive intracellular
pools. Other studies have confirmed the CCR1-mediated activation of a
PLC/IP/DAG/PKC pathway in human osteogenic sarcoma (HOS) cells stably
expressing CCR1. In this cell line, the chemokine Lkn-1 (CCL15) promotes IP3
formation and redistribution of PKCδ from the cytosol to the membrane.
Furthermore, Lkn-1 increases the binding activity of the transcriptional
regulator NF-κB, and this activation is dependent on Gi/o signalling, PLC and
PKC activity[41]. In HOS cells, Lkn-1 is a powerful chemoattractant, whose
activity is dependent on the synthesis of new proteins, as demonstrated by the
inhibition of chemotaxis by cycloheximide and actinomycin D[41]. This indicates
that the PLC/PKC pathway activated by Lkn-1 ultimately leads to gene
transcription and translation, possibly by activation of NF-κB.
Similarly, in COS-7 cells CCR2a and CCR2b promote IP3 formation through
Gβγ-subunit release from Gi/o proteins in the presence of PLC-β2[38]. In
HEK-293 cells, PTX-sensitive Ca²⁺ mobilization by CCR2a activation
does not require co-transfection of PLC-β2[13], presumably because Gβγ subunit
sensitive PLC-β isozymes are endogenously expressed in this cell type. CCR2b,
but not CCR2a, is also able to promote IP3 formation via G14 and G16
proteins[38].
Activation of CCR3 receptors by the specific ligand eotaxin induces calcium
transients in AML cells and cultured Purkinje cells. The calcium release in
both of these cell types is completely inhibited by pre-treatment of the cells
with PTX[16,20]. This indicates that in these cell types CCR3 does not
functionally couple to Gq/11-family proteins, however we have recently observed
that in heterologous systems CCR3 can promote IP3 formation when co-expressed
with either Gα14 or Gα16 (unpublished data). It is predicted that CCR3 and Gα14
and Gα16 will be co-expressed in lymphoid cell lines[4,42,43] but PTX-resistant
activation of PLC-β-coupled pathways has not yet been demonstrated.

Activation of CCR4 leads to Ca²⁺
mobilization in many cell types, including platelets[18], glomerular
podocytes[44], CD4+ T lymphocytes[45] and IANK cells[46]. The G protein
involvement in these Ca²⁺ responses has not been investigated. In
IANK cells the CCR4-mediated chemotaxis is PTX-sensitive, but a direct link
between Ca²⁺ levels and cell motility was not demonstrated[46]. In
cultured cerebellar Purkinje cells, MDC-induced Ca²⁺ levels are
reduced by approximately 50% following PTX treatment[20]. Surprisingly, in the
same study the response to thymus- and activation-related chemokine (TARC) (CCL17),
another fairly selective CCR4 agonist, was completely insensitive to PTX
treatment. This data suggests that CCR4 is linked to PLC-β activation by more
than one pathway and that CCR4 ligands do not necessarily induce all pathways
with equal potency and efficacy.

In COS-7 cells transiently expressing CCR5, a variety
of known agonists promote IP3 accumulation[47]. In CHO-K1 cells stably
expressing CCR5, MIP-1β stimulates intracellular Ca²⁺ release, as
measured in an aequorin-based functional assay. PTX pretreatment of the cells
completely abolished the response, indicating the involvement of Gi/o proteins
in CCR5-mediated activation of PLC-β-coupled pathways[24]. Interestingly,
several HIV strains are known to induce intracellular inositol phosphate formation
in neurons[48]. While it is known that CCR5 activation of Gi/o-coupled pathways
is not require for HIV infection[49], these authors hypothesize that viral
induction of chemokine signalling pathways may play a role in the development
and progression of HIV-associated dementia.
CCR6 expressed in HEK-293 cells respond to their selective ligand LARC with
mobilization of intracellular Ca²⁺[50]. LARC also induces Ca²⁺
mobilization in peripheral blood lymphocytes[51], dendritic cells[52], memory T
cells[53]. In cultured cerebellar Purkinje cells, the LARC-induced signalling
is abolished by PTX pretreatment of the cells[20]. The ability of CCR6 to
couple to Gq/11 family subunits has not been thoroughly investigated although
LARC is known to induce binding of GTPγS to Gαq in IANK cells[25]. In the same
study, the selective CCR7 ligand ELC also activated nucleotide exchange on Gαq
subunits, and in lymophocytes and dendritic cells, ELC induces Ca²⁺
mobilization[54,55]. In cultured cerebellar Purkinje cells, CCR7-induced Ca²⁺
fluxes are partially reduced by PTX pretreatment[20] and this is in agreement
with the previously observed Gαq activation by CCR7. However, we have recently
observed that CCR7 is unable to promote IP3 formation through activation of
endogenous Gq/11 proteins in COS-7 cells and furthermore, is unable to
functionally couple to G14 or G16 proteins (unpublished data).

CCRs8, 9 and 10 also promote increases in
intracellular Ca²⁺ levels[44,46,56] in cells endogenously expressing
these receptors. In cultured cerebellar Purkinje cells, CCR8-induced Ca²⁺
fluxes were not affected by PTX treatment of the cells indicating the lack of
involvement of Gi/o-coupled pathways[20]. This data correlates well with
results that demonstrate that CCR8 is able to functionally couple to Gα16 in
transfected HeLa cells to enhance the activity of NF-κB, a transcriptional
regulator[57].

The viral CC chemokine receptor homologue US28
activates Ca²⁺ flux when expressed in HEK-293 cells and challenged
with the chemokines RANTES or MCP-3. It was determined that US28 is able to
trigger Ca²⁺ flux via Gαi or Gα16 subunits[58]. Furthermore, human
fibroblasts infected with HCMV respond to RANTES with increased Ca²⁺
flux, while HCMV with a disrupted US28 open reading frame does not induce the
same response[59]. When the human herpesvirus 6 protein UL12, a CC chemokine
GPCR homologue, is expressed in human cells it responds to a number of CC
chemokines with increased Ca²⁺ flux[60], as does the E1 protein
encoded by the equine herpesvirus 2[61].

3    Ion channel regulation

Two groups of Ca²⁺
channels, voltage-dependent Ca²⁺ channels (VDCCs) and second
messenger-operated Ca²⁺ channels (SMOCs), are regulated by
GPCR-coupled pathways. Five types of VDCCs have been cloned or identified,
designated as L-, N-, P-, Q-, and R-types. L-, N- and P/Q-type VDCCs are
inhibited by Gi/o family proteins while N- and P/Q-type VDCCs can be stimulated
by Gs proteins[62] but Go appears to be more efficient than Gi[63]. Moreover, Gβγ subunits have been
suggested to inhibit N-type channels by a PLC-β stimulated PKC pathway[64].
Thus, it might be expected that stimulation of CC chemokine receptors may lead
to the inhibition of N-, L-, and P/Q-type VDCCs via Gαi/o- and Gβγ-mediated pathways. In
addition, it has been shown that intracellular inositol phosphates can trigger
the opening of SMOCs[65], so it is possible that CCRs induce the opening of
SMOCs via a PLC-dependent pathway.

Indeed, CCR4 and CCR5
are able to mediate the PTX-sensitive inhibition of N-type VDCCs in transiently
transfected cells as well as in subpopulations of neuronal cells[22]. The
physiological relevance of this inhibition is not clear but may provide a link
between chemokines released in the periphery and the central nervous system.

A far more common observation is
the chemokine-induced activation of Ca²⁺ channels, whose activities
have been shown in a number of studies to account for the majority of the
intracellular Ca²⁺[40,66]. When the kinetics of the changes in Ca²⁺
levels are closely monitored there often appears to be a biphasic mobilization.
In T cells, RANTES produces a biphasic mobilization of Ca²⁺ levels
with the initial transient rise due to intracellular release followed by a
sustained influx of Ca²⁺ from the extracellular environment[67]. In
monocytes, MCP-1, a CCR2 agonist, induces increases in Ca²⁺ levels,
which are dependent on extracellular Ca²⁺ levels and are partially
inhibited by the Ca²⁺ channel blocker Ni²⁺. Thapsigargin,
which empties intracellular Ca²⁺ stores, also partially blocks Ca²⁺
mobilization, but when combined with Ni²⁺ completely blocks the
actions of MCP-1[68]. Furthermore, the MCP-1-induced Ca²⁺ levels are
sensitive to PTX but not cholera toxin (CTX), suggesting that the effects were
predominantly Gi/o-mediated. However, the Ca²⁺ level increases
induced by MCP-2, a broad spectrum CC chemokine, were sensitive to CTX but not
PTX, implicating a role for Gs proteins[68].

Neurotransmitters and
hormones are known to regulate many kinds of potassium channels directly via Gα and Gβγ subunits or
indirectly via second messengers[69]. Chemokine activation of CCR5 on
macrophages opens calcium-activated potassium channels[70] and on T cells opens
voltage-gated potassium channels[71].

4   Regulation of MAPK pathways

GPCRs regulate
cell proliferation and differentiation through a family of MAPKs. These
serine/threonine protein kinases are capable of phosphorylating several
transcription factors, thereby regulating subsequent transcriptional events.
There are at least three subtypes of MAPK, the extracellular signal-regulated
kinases (ERKs) are mainly stimulated by growth factors, whereas c-Jun
NH2-terminal kinases (JNKs, also referred to as the stress activated protein
kinases, SAPKs) and p38 MAPK are more responsive to cellular stress[72].

In astrocytes,
RANTES induces the phosphorylation of MEK, ERK1/2 as well as p90 ribosomal S6
kinases (RSK) and CREB. This cascade of events is inhibited by U0126, a MEK
inhibitor, and dominant negative mutants of RSK and CREB blocked the
transcriptional activation of chemokine synthesis[21]. RANTES also induces
Raf-1 kinase activity and this is sufficient to induce the transcription of
proinflammatory chemokines. The PKA inhibitor H-89 is also able to promote
chemokine synthesis[29]. This suggested to the authors that RANTES signalling
induces the activation of a Raf/MEK/ERK pathway resulting in increased
synthesis of chemokines, which may prolong the inflammatory response, and that
activation of this MAPK kinase pathway works in concert with RANTES-induced
inhibition of AC. As RANTES is a potent ligand for CCRs1, 3 and 5 the receptor
mediating these events was not identified. In HOS cells expressing CCR1, Lkn-1
activation of ERK1/2 is PTX-sensitive and required the activity of PLC, PKCδ
and Ras. Activation of this cascade by Lkn-1 induces the immediate early genes
c-fos and c-myc, which may regulate the Lkn-1-mediated cell cycle
progression[73].

Activation of
CCR2b, but not CCR2a, induces rapid induction of ERK activity in a PKC/MEK
dependent manner. In human monocytes and transiently transfected cells, this
effect is partially PTX-sensitive[74] and may involve an element of mediation
by G14 and G16 proteins (unpublished data). Conversely, in murine peritoneal
macrophages and skin fibroblasts the MCP-1-induced ERK phosphorylation is
completely PTX-sensitive[75,76]. MCP-1 challenge of murine peritoneal
macrophages also demonstrated the MCP-1-mediated phosphorylation of JNK and
activation of the c-jun transcription factor[75].

CCR3-mediated
phosphorylation and activation of ERK1/2 and p38, but not JNK, in eosinophils
is required for cellular chemotaxis[77]. In rat basophilic leukemia (RBL-2H3)
mast cells CCR3-mediated activation of MAPK pathways is dependent on Rac
activity, which correlates with CCR3-mediated actin reorganization known to be
necessary for cell motility[78].

There are also
reports that CCR4-mediated activation of ERK1/2 in hippocampal neurons
partially protects the cells against the neurotoxicity of the HIV-1 envelope
protein gp120[19]. In contrast CCR5 is thought to potentiate the
neurodegenerative effects of HIV infection. When CCR5 was activated by its
ligand MIP-1β along with co-stimulation of T cells and glioma cells by gp120,
both p38 and JNK pathways were activated but stimulation by MIP-1β alone
resulted in activation of ERK1/2. Selective inhibition of p38 prevented the
gp120 induction of MMP-9, a protein thought to be involved in the development
of HIV-induced dementia, suggesting that a strategy of targeting p38 may have
clinical benefits[79]. CCR8 may also provide an element of protection against
external challenges that induce apoptosis by activating ERK1/2 in mouse thymic
lymphoma cell lines via Gi/o proteins[80].

5    Regulation
of phospholiase A2 activity

Arachidonic acid
(AA) acts as a second messenger for receptors in the superfamily of GPCR and
receptor tyrosine kinases. AA hydrolysis from phospholipids is regulated by the
cytosolic form of phospholipase A2 (cPLA2) in response to hormones[81]. AA is a
precursor of proinflammatory eicosanoids[82] and is therefore of considerable
relevance to the study of chemokine action.

The selective
CCR1 ligand MPIF-1 (CCL23) promotes rapid AA release from monocytes and the
activity of PLA2 is required for the formation of filamentous actin formation,
an essential step for cellular motility[40]. Monocyte chemotaxis induced by
several CC chemokines is also inhibited following antisense inhibition of cPLA2
expression[83]. In this study, the chemotaxis induced by MCP-1 was completely
inhibited by antisense knockout of cPLA2, suggesting that AA formation is an
absolute requirement for CCR2-induced monocyte chemotaxis.

6 Concluding remarks

The G
protein-coupled CC chemokine receptors are known to modulate a series of effectors
through G proteins and second messengers. CCR-induced regulation of some of
these effectors including AC, PLC-β, and MAPK pathways has been confirmed in
both heterologous systems and cells endogenously expressing CC chemokine
receptors. There is also considerable evidence for the involvement of Ca²⁺,
K+, phosphoinositide 3-kinase[84] and small GTPases[37] in the transduction of
chemokine signalling but the role of individual G protein subunits has often
not been clearly identified. Since many of the components of these pathways
have been cloned, the putative signalling pathways linking CCRs to their
effectors can be reconstituted and examined in a defined environment. This has
undoubtedly enabled the elucidation of many of the subtleties of CCR signalling
and will allow further refinement of our understanding of many of the puzzling
aspects of these signalling events. This approach must always be complemented
by examination of signalling pathways in endogenous tissues as overexpression
of various signalling components may allow measurement of activities that do
not occur naturally, due to different levels of expression or spatial and
temporal separation. For example, in this review we have presented examples
where CCRs have differing effects on AC and Ca²⁺ channels, depending
on the subtype of effector isozyme present in particular tissues.

A number of
CCR-coupled transduction pathways are yet to be described adequately. For
example, the ability of some CCRs to couple to G16 proteins opens up a number of
new signalling possibilities, including activation of signal transducers and
activators of transcription (STAT)-dependent pathways[85] and NF-κB
activation[57].

Given the number
of signalling pathways that have been demonstrated to mediate CC chemokine-induced
events, it will be particularly important to establish which pathway or, more
likely, which combination of pathways are actually required to regulate a
physiological event in a given cell type. As an example of the current
uncertainty in the relevance of individual pathways, the presented data on
human monocytes is instructive. In this cell type, the chemoattractant
properties of MIP-1α and RANTES are significantly inhibited by treatment with
forskolin and IBMX, implying a vital role for CCR regulation of AC-coupled
pathways[30]. In addition, antisense knockout of cPLA2 reduces MIP-1α and
RANTES-induced chemotaxis by approximately 75%[83]. This suggests that either
there is a degree of overlap between these two pathways or that they operate
independently and both are required to trigger the biochemical mechanisms that
ultimately regulate the multitude of cellular events required for cell
motility.

The studies
described suggest that controlling CC chemokine signalling activity may be
beneficial to the treatment of numerous pathophysiological states, and indeed
CCR ligands are being actively investigated for their therapeutic applications.
For example, a number of non-peptide CCR1 antagonists have been described with
applications in the treatment of multiple sclerosis and improving the success
of organ transplants[6]. The increasing use of the CCRs in high-throughput drug
screening protocols and the availability of highly promiscuous chimeric G
protein adaptors[86] (Wong et al., submitted for review; Hazari et al.,
submitted for review), will certainly facilitate the discovery of new ligands
that may increase our ability to control CC chemokine signal transduction
cascades.

CCRs clearly
activate a network of signalling pathways many of which have been shown to play
vital roles in the regulation of physiological and pathophysiological events.
Unravelling the interplay between these interwoven pathways is undoubtedly a
daunting task but one which will result in a much fuller understanding of how
we may be able to regulate many physiological processes

References

1    Hasegawa
H, Fujita S. Chemokines and lymphocytes: The role of chemokines and their
receptors in the immune system. Cell Mol Biol, 2001, 47(4): 599－607

2     Stantchev
TS, Broder CC. Human immunodeficiency virus type-1 and chemokines: Beyond
competition for common cellular receptors. Cytokine Growth Factor Rev, 2001,
12(2-3): 219－243

3     Gale
LM, McColl SR. Chemokines: Extracellular messengers for all occasions?
BioEssays, 1999, 21(1): 17－28

4     Murphy
PM, Baggiolini M, Charo IF, Hébert CA, Horuk R, Matsushima K, Miller LH et al.
International Union of Pharmacology. XXII. Nomenclature for chemokine
receptors. Pharmacol Rev, 2000, 52(1):145－176

5     Fernandez
EJ, Lolis E. Structure, function, and inhibition of chemokines. Ann Rev
Pharmacol Toxicol, 2002, 42: 469－499

6     Horuk
R. Chemokine receptors. Cytokine Growth Factor Rev, 2001, 12(4): 313－335

7     Chou
CC, Fine JS, Pugliese-Sivo C, Gonsiorek W, Davies L, Deno G, Petro M et al.
Pharmacological characterization of the chemokine receptor, hCCR1 in a stable
transfectant and differentiated HL-60 cells: antagonism of hCCR1 activation by
MIP-1β.
Br J Pharmacol, 2002, 137(5): 663－675

8     Gao
JL, Wynn TA, Chang Y, Lee EJ, Broxmeyer HE, Cooper S, Tiffany HL et al.
Impaired host defense, hematopoiesis, granulomatous inflammation and type
1-type 2 cytokine balance in mice lacking CC chemokine receptor 1. J Exper Med,
1997, 185(11): 1959－1968

9     Deng
HK, Unutmaz D, KewalRamani VN, Littman DR. Expression cloning of new receptors
used by simian and human immunodeficiency viruses. Nature, 1997, 388(6639): 296－300

10   Liu R, Paxton WA, Choe S, Ceradini D, Martin SR, Horuk
R, MacDonald ME et al. Homozygous defect in HIV-1 coreceptor accounts for
resistance of some multiply-exposed individuals to HIV-1 infection. Cell, 1996,
86(3): 367－377
11   Taussig R, Zimmermann G.
Type-specific regulation of mammalian adenylyl cyclases by G protein pathways.
Ad Second Messenger Phosphoprotein Res 1998, 32: 81－98

12   Tsu RC, Wong YH. Gi-mediated stimulation of type II
adenylyl cyclase is augmented by Gq-coupled receptor activation and phorbol
ester treatment. J Neurosci, 1996, 16(4): 1317－1323

13   Myers SJ, Wong LM, Charo IF. Signal transduction and ligand
specificity of the human monocyte chemoattractant protein-1 receptor in
transfected embryonic kidney cells. J Biol Chem, 1995, 270(11): 5786－5792

14   Monteclaro FS, Charo IF. The amino-terminal domain of
CCR2 is both necessary and sufficient for high affinity binding of monocyte
chemoattractant protein 1. Receptor activation by a pseudo-tethered ligand. J
Biol Chem, 1997, 272(37): 23186－23190

15   Tenscher K, Metzner B, Schopf E, Norgauer J, Czech W.
Recombinant human eotaxin induces oxygen radical production, Ca²⁺-mobilization,
actin reorganization, and CD11b upregulation in human eosinophils via a
pertussis toxin-sensitive heterotrimeric guanine nucleotide-binding protein.
Blood, 1996, 88(8): 3195－3199

16   Zimmermann N, Daugherty BL, Stark JM, Rothenberg ME.
Molecular analysis of CCR-3 events in eosinophilic cells. J Immunol, 2000,
164(2): 1055－1064
17   Jinquan T, Quan S, Feili G,
Larsen CG, Thestrup-Pedersen K. Eotaxin activates T cells to chemotaxis and
adhesion only if induced to express CCR3 by IL-2 together with IL-4. J Immunol,
1999, 162(7): 4285－4292

18   Kowalska MA, Ratajczak MZ, Majka M, Jin J, Kunapuli S,
Brass L, Poncz M. Stromal cell-derived factor-1 and macrophage-derived
chemokine: 2 chemokines that activate platelets. Blood, 2000, 96(1): 50－57

19   Meucci O, Fatatis A, Simen AA, Bushell TJ, Gray PW,
Miller RJ. Chemokines regulate hippocampal neuronal signaling and gp120
neurotoxicity. Proc Nat Acad Sci USA, 1998, 95(24): 14500－14505

20   Gillard SE, Lu M, Mastracci RM, Miller RJ. Expression of
functional chemokine receptors by rat cerebellar neurons. J Neuroimmunol, 2002,
124(1-2): 16－28

21   Zhang Y, Zhai Q, Luo Y, Dorf ME. RANTES-mediated
chemokine transcription in astrocytes involves activation and translocation of
p90 ribosomal S6 protein kinase (RSK). J Biol Chem, 2002, 277(21): 19042－19048

22   Oh SB, Endoh T, Simen AA, Ren D, Miller RJ. Regulation
of calcium currents by chemokines and their receptors. J Neuroimmunol, 2002,
123(1-2): 66－75
23   Zhao J, Ma L, Wu YL, Wang
P, Hu W, Pei G. Chemokine receptor CCR5 functionally couples to inhibitory G
proteins and undergoes desensitization. J Cell Biochem, 1998, 71(1): 36－45

24   Blanpain C, Wittamer V, Vanderwinden JM, Boom A,
Renneboog B, Lee B, Le Poul E et al. Palmitoylation of CCR5 is critical for
receptor trafficking and efficient activation of intracellular signaling
pathways. J Biol Chem, 2001, 276(26): 23795－23804
25   Al-Aoukaty A, Rolstad B,
Giaid A, Maghazachi AA. MIP-3alpha, MIP-3β and fractalkine induce the
locomotion and the mobilization of intracellular calcium, and activate the
heterotrimeric G proteins in human natural killer cells. Immunology, 1998,
95(4): 618－624
26   Mantel C, Aronica S, Luo Z,
Marshall MS, Kim YJ, Cooper S, Hague N et al. Macrophage inflammatory protein-1α enhances growth factor-stimulated
phosphatidylcholine metabolism and increases cAMP levels in the human growth
factor-dependent cell line M07e, events associated with growth suppress