![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 62, Issue 3, 729-736, September 2002
and Its
CCR1 Receptor
Département Récepteurs et Protéines Membranaires, Centre National de la Recherche Scientifique UPR 9050 and IFR 85, Ecole Supérieure de Biotechnologie de Strasbourg, Illkirch, France (S.Z., J.-L.G.); and Serono Pharmaceutical Research Institute, Plan-les-Ouates, Geneva, Switzerland (A.C.)
 |
Abstract |
|---|
 Top
 Abstract
 Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Macrophage inflammatory peptide-1
(MIP-1)/CC-chemokine receptor
ligand 3 is an 8-kDa peptide that induces chemotaxis of various
lymphocytes to sites of inflammation through interaction with the G
protein-coupled chemokine receptors CCR1 and CCR5. We recently
described the preparation of a photoactivatable derivative of MIP-1
labeled with a benzophenone group at the extreme N-terminal end, which
is a determinant for the agonist character of chemokines. Benzophenone-MIP-1 is a full agonist that specifically and
covalently labels CCR1 and CCR5 receptors upon irradiation. In the
present study, we use enzymatic and chemical cleavage methods on
wild-type and mutated CCR1 receptors to show that the N terminus of the chemokine MIP-1 interacts in a specific manner with the second extracellular loop of the CCR1 receptor, within a segment comprising amino acids 178 to 194. This is the first report on the direct identification of a contact point between the N terminus of a chemokine
and its membrane-bound receptor. The work shows that the part of
chemokines that is endowed with agonist properties interacts with
extracellular parts of the receptor rather than the transmembrane core
of the protein.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Chemokines
constitute a large family of 8- to 12- kDa small proteins,
characterized by a well-conserved three-dimensional structure involving
two highly conserved cysteine bridges (Rollins, 1997
). Chemokines are
responsible for the orchestration of leukocyte recruitment to sites of
inflammation or lymphoid tissue (Rollins, 1997). Their receptors belong
to the rhodopsin-like group of G-protein coupled receptors. So far,
approximately 20 chemokine receptors and 50 chemokines have been
identified. They can be subdivided into four groups, based on position
of cysteine residues (Murphy et al., 2000).
The human homolog of the CC-chemokine macrophage inflammatory protein
1 (MIP-1) (Nakao et al., 1990), also known as LD78 or
CC-chemokine receptor ligand 3 (Murphy et al., 2000), belongs to the
largest group of chemokines, characterized by two adjacent cysteines in
the N terminus.
MIP-1 expression has been identified in peripheral blood monocytes
and different lymphocyte derived cell-lines (Nakao et al., 1990; Menten
et al., 1999). It is found at sites of inflammation, such as wounded
tissue (DiPietro et al., 1998). The folding of the peptide backbone
determined by NMR is very similar to that of other chemokines
(Czaplewski et al., 1999), such as MIP-1
, RANTES and MCP-1: a
relatively flexible N-terminal sequence of 10 amino acids followed by
the so-called N-loop, a helical turn, a three-stranded -sheet, and a
C-terminal -helix. The whole structure is stabilized by the two
disulfide bridges linking the N terminus and the N-loop to the
-sheet at -strand 3 and to the loop connecting -strands 1 and
2 (Czaplewski et al., 1999).
MIP-1 mediates its effects through binding to CCR1 and CCR5
receptors.
CCR1 receptors are expressed on lymphocytes such as
monocytes and peripheral blood monocytes (Neote et al., 1993; Su et
al., 1996). Apart from MIP-1, this receptor interacts with several other CC-chemokines such as RANTES, leukotactin, MCP1-4, myeloid progenitor inhibitory factor-1, and LD78 (Menten et al., 1999; Murphy et al., 2000).
and CCR1 is involved
in chemotaxis of peripheral blood monocytes in vitro (Su et al., 1996;
Hesselgesser et al., 1998).
Knockout mice have been generated for both MIP-1 (Cook et al., 1995)
and the CCR1 receptor (Gao et al., 1997). In both cases, gene
disruption is nonlethal, but a reduced immunological response by cell
infiltration is observed in general when the two strains of mice are
challenged in different ways. Better understanding of MIP-1-CCR1
receptor interactions may thus be used as a basis for development of
new agents for the treatment of a variety of inflammatory diseases such
as asthma, arthritis, or multiple sclerosis.
Very little is known about the topology of the complex formed between
MIP-1 and the CCR1 receptor. The effect of a few structural changes,
such as point mutagenesis of MIP-1, on the interaction with CCR1
have been investigated (Koopmann and Krangel, 1997; Czaplewski et al.,
1999).
One powerful approach to study receptor-ligand interactions is
affinity- or photoaffinity-labeling. By analysis of the covalently labeled receptor, it has been possible to identify interaction regions
or even amino acids for various peptide ligands such as substance P
(Girault et al., 1996), angiotensin (Boucard et al., 2000),
cholecystokinin (Dong et al., 1999), and vasopressin (Cotte et al.,
1998). In most cases, the preferred photoactivatable group was
benzophenone. Benzophenone incorporation can take place at many
different types of amino acids (Kotzyba-Hibert et al., 1995). The
reaction involves a diradical intermediate that is not deactivated by
water and that reacts with the target protein within ~3-Å distance (Dorman and Prestwich, 1994; Kotzyba-Hibert et al., 1995).
We described previously the preparation and pharmacological
characterization of MIP-1 derivatized with benzophenone
(BP-MIP-1) and fluorescein at the N-terminal position (Zoffmann et
al., 2001). We showed that BP-MIP-1 is specifically cross-linked to
either CCR1 or CCR5 receptor upon irradiation (Zoffmann et al., 2001). In the present work, we use enzymatic and chemical methods to identify
the site of covalent incorporation of BP-MIP-1 in the CCR1 receptor.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
The N-terminal photoactivatable chemokine
125I-BP-MIP-1 was prepared as described
previously (Zoffmann et al., 2001). The Chinese hamster ovary (CHO)
cell line stably expressing human CCR1 with a hemagglutinin tag
(HA-sequence: YPYDVPYASLRS) introduced at the N terminus was prepared
by Christine Power (Solari et al., 1997).
Reagents and Materials. pIRES-Neo plasmid was purchased from BD Biosciences Clontech (Le pont de Claix, France); QuikChange site-directed mutagenesis kit was purchased from Stratagene (Amsterdam Netherlands); solutions for cell-culture were purchased from Invitrogen (Cergy Pontoise France). The Polytron PT1200 sonicator was from Kinematica (Lucerne, Switzerland). Trypsin, Glu-C, Lys-C, Arg-C (all sequencing grade), protease inhibitor cocktail, bacitracin, chymostatin, leupeptin were purchased from Roche Diagnostics (Meylan France); phosphoramidon and bestatin were purchased from ICN Biochemicals (Aurora OH). The Philips HPR 125-W lamp was from Philips AG lightning (Zurich, Switzerland), and Microsep concentrators were from Pall Life Sciences, Bioblock Scientific (Illkirch, France). Discontinuous polyacrylamide gels were from Novex, ICT AG (Basel, Switzerland), and the Spectra/Por Cellulose Ester DispoDialyzer was from Spectrum, Struers Kebolab (Albertslund, Denmark).
Introduction of Point Mutations in the CCR1 Receptor. The CCR1 gene was inserted in the KS-Bluescript vector by use of the sites BamHI and XhoI, and the mutations Phe149 to asparagine (TM4) and Leu205 to glycine (TM5) were introduced by use of the QuikChange site-directed mutagenesis kit. The vector was multiplied in the dam-methylation positive Escherichia coli strain XL1-Blue. For each mutation, two complementary mutated oligonucleotides were designed. The whole plasmid was replicated by polymerase chain reaction with the Pfu-turbo DNA polymerase. The WT template DNA was removed by digestion with the dam-methylation recognizing endopeptidase DpnI and the nicked mutated plasmids were transfected into the E. coli strain XL1-Blue. The genes for CCR1, CCR1[F149N], and CCR1[L205G] were excised with BamHI and NotI and introduced in the pIRES-Neo plasmid, opened with EcoRV and NotI.
Cell Membrane Preparation.
A 60-liter suspension culture of
CCR1 expressing CHO cells was grown to an approximate density of
106 cells/ml in Dulbecco's modified
Eagle's/Ham's F-12 medium supplied with 10% fetal calf serum, 2 mM
glutamine, and 100 units/ml penicillin/streptomycin. Cell membranes
were prepared by sonication with a Polytron PT1200 sonicator in a small
volume of lysis buffer (HEPES, pH 7.4, 1 mm EDTA, 10 mM
MgCl2 supplied with protease inhibitor cocktail), followed by centrifugation at 500g for 30 min to remove cell
debris. Membranes were recovered from the supernatant by centrifugation at 48,000g for 30 min. The resulting pellets were
reconstituted in a small volume of breaking buffer, aliquoted, and
stored at 
80°C until use. Stored under these conditions, the
receptors retained their binding activity for several months. The
protein content was estimated by use of a colorimetric assay.
Photoaffinity Labeling of CCR1 with
125I-BP-MIP-1.
For photoaffinity labeling of
receptors, 1 mg membrane protein (30 µg/µl) was diluted in 2 ml of
Rosen A buffer (50 mM HEPES, 1 mM CaCl2, 5 mM
MgCl2, 0.5% BSA, pH 7.2).
125I-BP-MIP-1 was added to a final
concentration of 2 nM and the mixture incubated at room temperature for
2 h. Membranes were irradiated at 365 nm with a Philips HPR 125-W
lamp at 6-cm distance while kept on ice with continuous stirring. After
30 min irradiation, the membrane fraction was separated from unbound
ligand by centrifugation at 20,000g for 25 min at 4°C and
washed once in 2 ml of ice-cold Rosen A buffer.
Alkylation and Purification of CCR1-125I-BP-MIP-1
Complex.
For reduction and alkylation of the free cysteines, the
membrane pellet was resuspended in 200 µl of 0.1 M Tris, pH 8.5, containing 8 M urea, sonicated, and flushed with nitrogen.
-mercaptoethanol (10 µl) was added and the reaction incubated at
40°C for 16 h under nitrogen atmosphere. A 200-µl solution of
24 mg of iodoacetamide and 30 mg of Tris-base was added, pH adjusted to
9 with HCl; the reaction incubated at room temperature for 15 min in
the dark and a further 10 µl of -mercaptoethanol was added to
quench unreacted iodoacetamide. Afterward, the reaction mixture was
desalted by two consecutive dilutions and concentration steps using
Microsep concentrators (molecular mass >10-kDa cut-off).
Finally, the sample was separated on polyacrylamide gel. A small part
of the undried gel was autoradiographed at 80°C for 2 h and
the position of the labeled receptor used to identify the position of
the complex in the rest of the undried and unstained gel. For
endopeptidase and NH4OH cleavage, the peptide was
extracted by passive elution at room temperature for 24 to 48 h in
1.5 ml of 0.1 M NH4HCO3 buffer, pH 8.5. To remove salt and SDS traces from the gel and to
exchange buffer, the extract was dialyzed in a DispoDialyzer with a
molecular mass cut-off at 25 kDa against 10 mM
NH4HCO3 or 10 mM
Tris-buffer, and the peptide concentrated 10 times under vacuum or in a
Microsep concentrator.
Protease Digestion, CNBr and NH4OH Fragmentation of
the CCR1-125I-BP-MIP-1 Complex.
For the Glu-C
cleavage, the receptor-ligand complex was incubated at 25°C for
18 h in 20 µl 25-50 mM
NH4HCO3, pH 7.8, with 2 µg of Glu-C. For trypsin cleavage,
CCR1-125I-BP-MIP-1 complex was incubated in 20 µl of 100 mM Tris, pH 8.5, containing 2 µg trypsin for 18 h at
37°C. After reaction, the samples were analyzed by SDS-PAGE on a 12%
Tris-tricine gel or a Novex 10 to 20% Nupage gel, followed by autoradiography.
covalent complex was
cleaved in gel with CNBr (Grutter et al., 2000). The isolated gel
fragment containing the ligand receptor complex was cut into small
pieces, washed twice for 20 min in a 1:1 mixture of water and
acetonitrile and dried to approximately half the initial size. CNBr
solution (0.7 ml or 150 mM) in 70% formic acid was added in several
steps to allow the gel to absorb the reagent and finally cover the
reconstituted gel pieces. The reaction was allowed to proceed in the
dark for 3 days under argon atmosphere. To eliminate the CNBr and acid, the samples were repeatedly dried under vacuum and reconstituted seven
times in 1 ml of water. Finally, peptides were extracted from gel
pieces in 2 ml of 100 mM
NH4HCO3 solution for
24 h at room temperature. The recovery of radioactivity was 80 to
90%. Extract was concentrated on a Microsep concentrator (3-kDa
cut-off). After reaction, the samples were analyzed by SDS-PAGE on a
12% Tris-tricine gel, followed by autoradiography.
For cleavage with NH4OH, the receptor ligand
complex was prepared as for protease digest. A hydroxylamine solution
at pH 9 in 8 M urea was prepared according to the method of Allen
(1989), with urea replacing 6 M guanidinium chloride.
NH4OH (25 µl) was added to the labeled receptor
and the reaction was incubated at 45°C overnight. The content of salt
was reduced 20-fold by use of a Microsep concentrator (10 kDa cut-off).
The reaction products were analyzed by SDS-PAGE on a 10% Tris-glycine
gel, followed by autoradiography.
Protease Digestion, CNBr and NH4OH Fragmentation of
125I-BP-MIP-1.
For protease digestion and CNBr
cleavage of 125I-BP-MIP-1 the ligand was
reduced, alkylated, purified by gel electrophoresis and the gel
containing the radioactivity isolated as described for the complex. The
ligand was extracted in
NH4HCO3 buffer,
concentrated under vacuum, and digested with Glu-C and trypsin, as
described for the receptor ligand complex. For Lys-C, the incubation
conditions were 1 µg of enzyme in 20 µl of 25 mM Tris-HCl and 1 mM
EDTA, pH 8.5, for 18 h. The conditions used for fragmentation with
Arg-C were 1 µg of enzyme in a total of 100 µl of 100 mM Tris-HCl,
10 mM CaCl2, 5 mM dithiothreitol, and 0.5 mM
EDTA supplied with 10 µl of Arg-C activation solution and
incubation for 18 h at 37°C.
ligand was performed in gel following the
same procedure as for the photoaffinity-labeled receptor. For the
cleavage with NH4OH, a sample of
125I-BP-MIP-1 was incubated with
NH4OH using the same protocol as for the complex.
The resulting fragments from CNBr cleavage, NH4OH cleavage, and protease digest were separated by SDS-PAGE and the gels
were analyzed by autoradiography.
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Photoaffinity Labeling and Partial Purification of CCR1.
Photoaffinity labeling of the recombinant CCR1 receptor is carried out
with a radioactive photoactivatable benzophenone derivative of the
chemokine MIP-1, 125I -BP-MIP-1, on
membrane preparations. This high-affinity agonist covalently binds in a
specific manner to CCR1 to form a complex with an apparent molecular
mass of 54 kDa, which represents the addition of 8 kDa of one MIP-1
into CCR1 receptor (46 kDa). The ratio between unincorporated ligand
and cross-linked complex is approximately 10 to 1 for receptors in
membrane preparations, which is comparable with the values found in
assays with intact cells (Zoffmann et al., 2001).
Experimental Strategy Used to Identify the Site of MIP-1
Incorporation.
The general approach used here to determine the
site of MIP-1 incorporation consists of identifying the smallest
sizable radioactive fragment that corresponds to a plausible
subfragment of the CCR1-MIP-1 complex, but not to a subfragment of
MIP-1 alone. The size of the fragment is then compared with the size of all expected digestion products. A combination of cleavage methods
yielding receptor fragments of different sizes is then used to reduce
the number of possibilities to a single domain of the receptor.
is a 69-amino acid
photoactivatable ligand to which no modifications were made to remove
existing enzymatic or chemical cleavage sites. Thus, only enzymatic or
chemical methods that do not cleave MIP-1 between the BP group and
the radioiodinated tyrosine(s) could be used for this approach.
Pattern of Enzymatic and Chemical Cleavage of MIP-1 Ligand with
Glu-C, Trypsin, Lys-C, Arg-C Proteases, and CNBr.
To analyze the
fragmentation pattern of the reduced and alkylated BP-MIP-1, the
peptide was submitted to cleavage with Glu-C, trypsin, Lys-C and Arg-C
endopeptidases, or CNBr. Migration of the cleavage products is shown in
Fig. 1. Using separation on SDS-PAGE, it
is possible to identify peptides with molecular masses as low as ~2
kDa.
|
are
radiolabeled upon radioiodination, the expected size of radioactive fragments obtained after Glu-C fragmentation is 3.6 (Y15 and Y28) and
1.5 kDa (Y62). We observe only one band at 3.4 kDa (Fig. 1A).
The Lys-C digestion is expected to result in two radiolabeled peptides
with molecular masses at 4.4 (Y15 and Y28) and 1 kDa (Y62) (cleavage at
positions 37, 45, and 61). A unique band at 4 kDa is observed (Fig.
1B). The Glu-C and Lys-C indicate that 125I is
present on at least one of the two tyrosines Y15 and Y28 at the
N-terminal fragment.
Trypsin digestion results in a broad band positioned between 3.5 and
2.5 kDa. For this digestion, three peptides with the respective sizes
of 2.2 (Y15), 2.3 (Y28), and 1.0 kDa (Y62) were expected (Fig. 1C). In
addition, the Arg-C digest was predicted to result in three bands at
2.2 (Y15), 3.3 (Y28), and 2.6 kDa (Y62). Only a band of ~3.0 kDa is
seen (Fig. 1D). Both trypsin and Arg-C are expected to cleave at
position 18 between the two tyrosines Y15 and Y28. The detectable bands
show similar apparent size for the two digests. The apparent molecular
mass is slightly larger than expected for all labeled bands from
trypsin, and for the N-terminal part of Arg-C, where the size of the
observed fragment better corresponds to the size of the Y28
radiolabeled fragment. The size of each fragment, however, is near the
detection limit, and slight changes in migration may occur. From the
fragmentation of BP-MIP-1 with Arg-C and trypsin, is it not possible
to conclude whether iodine is present at each of the tyrosines (Y15 and
Y28) or only at one of them.
CNBr treatment under acidic conditions does result, as expected, in no
cleavage, because no methionines are present in BP-MIP-1 (Fig. 1E).
Thus, from all digestion analyses, it can be concluded that
125I labels Y15 and/or Y28. Iodination of Y62 is
not crucial for the analysis and it is difficult to conclude whether it
is present or not.
Fragmentation of the CCR1-BP-MIP-1 Complex.
The Glu-C
endopeptidase (Staphylococcus aureus endopeptidase V8)
hydrolyzes peptide bonds after glutamate residues. As can be seen in
Fig. 2, digestion of the receptor-ligand
complex is not complete; such a pattern has been observed in four
independent cleavage experiments with this enzyme. The smallest clearly
detectable peptide on the autoradiography of Glu-C digest has an
apparent molecular mass of 3.5 to 4.5 kDa. Compared with digestion of
the ligand alone, the smallest band detected here is larger by up to
1.5 kDa indicating that a small fragment of the receptor protein is
covalently attached to the radioactive MIP-1 fragment. Assuming that
BP-MIP-1 does not link to intracellular parts of the receptor, putative small-sized receptor fragments that may correspond to the 3.5- to 4.5-kDa band that results from Glu-C cleavage are: 13 to 29 (N-terminal), 178-194 (E2), or 273-287 (E3). Their position in the
receptor structure is illustrated in Fig.
4.
|
13 to 2 in the HA sequence) aa 27-30 (N-terminal), 94 to
106 (E1), and 174 to 196 (E2). Their position in the expected
fragmentation pattern for the cleavage of CCR1 with trypsin is shown in
Fig. 4.
CNBr Cleavage.
Because of the acid labile nature of the
covalent bond between benzophenone and MIP-1, the conditions
required to cleave with CNBr were optimized by adjusting the incubation
time and pH of the reaction to selectively cleave methionine peptide
bonds before releasing the BP group. Figure
3 shows the result of incubation in 70%
formic acid for 24 h in the presence (lane 1) and absence (lane 2)
of CNBr. The amount of free ligand released by acid treatment is high,
but weak bands with apparent masses of 14, 16, and 20 kDa, which are
larger than the ligand alone, are clearly detected in three independent
experiments. These bands are not seen when the sample is incubated in
acid without CNBr. The observed size and relative intensity of these
fragments were very reproducible. Furthermore, bands around this size
are well separated on a 12% Tricine gel. The use of molecular mass
markers at 10, 14.4, 17.0, and 20 kDa allow accurate estimate of
apparent molecular masses. The contribution from the receptor to the
14-kDa band is thus expected to be about 6 kDa. This fits with the
expected fragment 106 to 166 (TM3-4) or the fragment 167 to 215 (E2 + TM5), as indicated in Fig. 4.
|
|
). Therefore the weak bands with
apparent sizes of 17 and 20 kDa (Fig. 3) could result from partial
cleavage. In particular, if the 14-kDa band corresponds to segment 167 to 215, the size of higher molecular mass bands could be accounted for
by partial cleavage at position 166, leading to a 20.4-kDa fragment
(106-215), and at position 215, leading to a 17-kDa fragment
(167-245). In contrast, a similar reasoning for cross-linking to the
fragment containing E1 + TM2 (106-166) would result in partial
cleavage bands at 20.4 (aa 106-215) and 18.9 kDa (aa 69-166), which
do not correspond to the experimentally detected fragment sizes. Our
results thus support cross-linking of BP-MIP-1 to fragment 167-215.
The N-terminal parts of the receptor that include aa 27-29, and 13
to 2 (in HA-tag), two potential segments from the trypsin and Glu-C
cleavages, have calculated weights at 16.5 and 23 kDa for a noncomplete
cleavage at the C-terminal, which do not fit with the CNBr cleavage pattern.
Despite purification attempts carried out to get mass spectrometry
analysis, the amount of available material was too low to get a
detectable signal that could lead to a direct confirmation of a part of
the receptor as being photoaffinity-labeled. We therefore used a
site-directed mutagenesis approach to support one receptor sequence as
interacting with the N-terminal of BP-MIP-1.
Cleavage with NH4OH of WT and Point-Mutated
Receptors.
Two mutant receptors were designed, each with a unique
Asn-Gly cleavage site for hydroxylamine introduced after positions 150 (TM4) and 205 (TM5) (Fig. 6). Both
receptors (F149N and L205G) are expressed in HEK293 cells at a level
comparable with that of wild-type CCR1 (data not shown). The mutations
did not significantly change the BP-MIP-1 binding affinity, and
BP-MIP-1 was incorporated in the mutants at a level similar to that
of wild-type receptor (data not shown). Migration on a gel was
identical for the mutant and wild-type receptors expressed in HEK293
cells and the wild-type receptor expressed in CHO cells used for the
Glu-C, trypsin, and CNBr cleavages. Both L205G and F149N are cleaved by
NH4OH, albeit incompletely (Fig.
5). However, for both mutated receptor-
BP-MIP-1 complexes, a band migrating at an apparent molecular mass
of 33 kDa was detected. This band was not present in the WT samples. Cleavage of the mutated receptors by NH4OH is
expected to produce two fragments. In the F149N mutant, the calculated
molecular masses expected are 25.5 kDa for the N-terminal part (aa
1-150) and 32.5 kDa for the C-terminal part (aa 151-354) when the
mass of MIP-1 is included. The experimental mass fits with the size
of the C-terminal fragment. In the L205G mutant, the similarly
calculated molecular masses of the two receptor fragments after
cleavage are 32 kDa for the N-terminal part (aa 1-205) and 26 kDa for
the C-terminal part (aa 206-354). For this mutant, the experimental
weight fits with the N-terminal part. The overlap between the
identified regions for the two mutants extends from the beginning of
TM4 to the middle of TM5 (aa 151-205), as illustrated in Fig. 4.
|
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
The recent elucidation of rhodopsin structure (Palczewski et al.,
2000), a protein from the same group as chemokine, tachykinin, cholecystokinin, and angiotensin receptors, sheds new light on N-terminal domain and extracellular loop folding. Indeed, most of these
receptors contain extracellular domains of comparable length. In
addition, they all contain a disulfide bridge linking E1 to E2. It is
thus plausible that these receptors exhibit similar folding of their
external parts, which, according to rhodopsin structure, would limit
the access of large ligands to transmembrane segments.
Compared with most ligands interacting with G protein-coupled
receptors, chemokines are characterized by a large size and a conserved
three-dimensional structure (Rollins, 1997). Their interactions with
chemokine receptors therefore may take place at the level of a large
area and involve multiple contacts between the ligand and the receptor,
as was shown for the interaction of -bungarotoxin with the nicotinic
acetylcholine receptor (Maelicke and Conti-Tronconi, 1989).
Analyses of chemokine interactions with receptors included two separate
aspects, namely chemical specificity of receptor recognition and
receptor activation (Wells et al., 1996; Crump et al., 1997). Receptor
recognition is mainly driven by an extended hydrophobic groove
delineated by the second and third -sheet strands (Clubb et al.,
1994; Mizoue et al., 1999; Mayer and Stone, 2000; Ye et al., 2000) as
well as the N-terminal end and the N-loop (Lowman et al., 1996) of the
chemokine. On the receptor side, the N terminus has been identified as
being counterpart in the interaction. The interaction is likely to
involve other parts of the receptor as well, because the structure of
rhodopsin demonstrates a three dimensional structure, whereas the
extracellular loops of the receptor are near the N-terminal end. For
the interaction between MIP-1 and the CCR1 receptor, all available
data indicate that it involves several parts of the receptor, including
the N terminus and E3 loop (Monteclaro and Charo, 1996; Su et al.,
1996; Pease et al., 1998).
Receptor activation mostly relies on residues situated in the N
terminus of the chemokines (Gong et al., 1996; Solari et al., 1997;
Laurence et al., 2000; Townson et al., 2000; Zoffmann et al., 2001),
whereas the corresponding parts of the receptor that interact with the
chemokine are poorly identified.
Based on the structure-activity relationship analyses carried out on
chemokines, a two-domain model of ligand interactions with the receptor
has been proposed (Wells et al., 1996; Crump et al., 1997; Hesselgesser
et al., 1998). According to this model, the residues of the chemokine
that confer binding selectivity or specificity may be distinguished
from those that confer agonist versus antagonist potency.
In this work, we use a photoactivatable chemokine carrying the
photochemically reactive group at the N terminus, within a domain of
the ligand that is involved in receptor activation rather than receptor
recognition. From separate cleavages carried out with Glu-C, trypsin or
CNBr of covalently labeled CCR1, we obtain data that support more than
one candidate for interaction with the N terminus of the chemokine as
presented in Fig. 4. However, when combined, the cleavage data converge
toward labeling occurring in the extracellular loop 2 of CCR1, within a
17-amino acid sequence comprising amino acids 178 to 194. To confirm
this potential incorporation site, we further analyze the cleavage
pattern of mutant receptors containing hydroxylamine sites in TM4 or
TM5 (Fig. 4). This approach, which consists of adding new sites rather
than removing existing ones, is based on the following arguments: 1)
because chemical or enzymatic cleavage reactions may not be
quantitative, removal of existing sites may not be conclusive. In
contrast, introduction of new cleavage sites is expected to lead to
positive cleavage results, for hydroxylamine in particular, which has
no consensus cleavage site on the WT CCR1 receptor; 2) the mutations
requested to create hydroxylamine sites are minor and located at some
distance from the site of chemokine labeling. They are less likely to
modify ligand-receptor interaction in contrast to mutations removing Glu-C or trypsin sites that are located near the MIP-1 incorporation site; and 3) removal or introduction of CNBr sites was ruled out because of the poor efficacy of CNBr cleavage on the WT receptor.
The hydroxylamine cleavage data on the mutant receptors, in combination
with the chemical and enzymatic cleavage data obtained on WT receptors,
unambiguously establish that the N terminus of MIP-1 interacts
within a segment comprising residues 178 to 194. The corresponding
sequence in rhodopsin is situated entirely in the E2 (Palczewski et
al., 2000), probably reflecting the situation in the CCR1 because the
length of the extracellular domains is highly conserved between the two receptors.
In association with previous work describing that fluorescein,
incorporated at the same N-terminal position as benzophenone in
MIP-1, is fully accessible to buffer-soluble fluorescence quenchers,
such as iodine (Zoffmann et al., 2001), the present data support the
view that the N-terminal of receptor-bound MIP-1, as well as the
labeled part of the E2 loop of the receptor, are located at the
exterior of the receptor.
According to the location of the photoactivatable group on the chemokine, the E2 loop would contribute interaction points that are responsible for the agonist activity of chemokines, whereas other interactions, in particular those with the N terminus and the E3 loop of the chemokine receptor, would contribute residues participating to the chemical specificity of receptor recognition (i.e., that are specific to chemokine recognition independently of their agonist or antagonist character).
In conclusion, the present data pinpoint the interaction between the
second extracellular loop of CCR1, and a portion of MIP-1 that
determines the agonist character of the ligand rather than its
recognition specificity. This result does not provide support to the
idea that agonists must have access to the transmembrane core of the
protein. An alternative activation mechanism, where the whole receptor
molecule undergoes structural changes during the transition leading to
receptor activation, should be considered. According to this hypothesis
the conformational transition would be global and would imply all
intracellular, transmembrane, and extracellular domains of the protein.
Accordingly, agonists would be able to selectively recognize the active
state of the receptor by interacting with any of these domains,
depending on their size, on the structure of their binding site, and on
accessibility of these regulatory sites.
| |
Acknowledgments |
|---|
We thank Amanda Proudfoot (Serono Pharmaceutical Research
Institute) for the kind provision of MIP-1, Christine Power for advice in construction of the CCR1-mutants and Manuela Lima for molecular biology and cell-culture. We also thank Professor Thue Schwartz for his support, and Gerardo Turcatti and Tim Wells for stimulating discussion.
| |
Footnotes |
|---|
Received March 4, 2002; Accepted June 17, 2002
This work was supported by GlaxoWellcome (fellowship to S.Z.), the Ph.D. study council, Faculty of Health, Copenhagen University (S.Z.), the European Molecular Biology Organization (Fellowship to S.Z.), the Centre National de la Recherche Scientifique, the Association pour la Recherche sur le Cancer, the Agence Nationale pour la Recherche sur le SIDA, and the Université Louis Pasteur.
Address correspondence to: Jean-Luc Galzi Département Récepteurs et Protéines Membranaires, CNRS UPR 9050 and IFR 85, Ecole Supérieure de Biotechnologie de Strasbourg, Boulevard Sébastien Brant, F-67400 Illkirch, France. E-mail: galzi{at}esbs.u-strasbg.fr
| |
Abbreviations |
|---|
aa, amino acid;
Arg-C, Clostridium
histolyticum endoproteinase;
BP, benzophenone;
CCR, CC-chemokine receptor;
CHO, Chinese hamster ovary;
E1, E2, and E3,
extracellular loops 1, 2, and 3;
Glu-C, Staphylococcus
aureus endopeptidase V8;
HA, hemagglutinin tag;
HEK, human
embryonic kidney;
LD78, tonsillar lymphocyte LD78 protein;
Lys-C, Lysobacter enzymogenes endopeptidase;
MCP, monocyte chemotactic protein;
MIP, macrophage inflammatory protein;
PAGE, polyacrylamide gel electrophoresis;
RANTES, regulated on
activation normal T-cell expressed and secreted;
TM, transmembrane
domain;
WT, wild-type.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
for an inflammatory response to viral infection.
Science (Wash DC)
269:
1583-1585, MIP-1, and RANTES. Characterization of active disaggregated chemokine variants.
J Biol Chem
274:
16077-16084 as a critical macrophage chemoattractant in murine wound repair.
J Clin Invest
101:
1693-1698[Medline]..
J Biol Chem
272:
10103-10109 can function as a monomer and depends on Phe13 for receptor binding.
Biochemistry
39:
3401-3409[CrossRef][Medline]. isoform of MIP-1 is the most potent CCR5 agonist and HIV-1-inhibiting chemokine.
J Clin Invest
104:
R1-5[Medline]. receptor, confers chemokine selectivity. Evidence for a two-step mechanism for MCP-1 receptor activation.
J Biol Chem
271:
19084-92 and eotaxin binding, respectively, but a second domain is essential for efficient receptor activation.
J Biol Chem
273:
19972-19976P increases receptor affinities and HIV Inhibition.
J Biol Chem
275:
39254-39261 ligands and interactions with chemokine receptors CCR1 and CCR5.
J Med Chem
44:
215-222[CrossRef][Medline].
This article has been cited by other articles:
|
|
 |
|
  G. Kleinau, M. Claus, H. Jaeschke, S. Mueller, S. Neumann, R. Paschke, and G. Krause Contacts between Extracellular Loop Two and Transmembrane Helix Six Determine Basal Activity of the Thyroid-stimulating Hormone Receptor J. Biol. Chem., January 5, 2007; 282(1): 518 - 525. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Vaidehi, S. Schlyer, R. J. Trabanino, W. B. Floriano, R. Abrol, S. Sharma, M. Kochanny, S. Koovakat, L. Dunning, M. Liang, et al. Predictions of CCR1 Chemokine Receptor Structure and BX 471 Antagonist Binding Followed by Experimental Validation J. Biol. Chem., September 15, 2006; 281(37): 27613 - 27620. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gupta, S. Schulz-Maronde, C. Kutzleb, R. Richter, W.-G. Forssmann, A. Kapp, U. Forssmann, and J. Elsner Cloning, expression, and functional characterization of cynomolgus monkey (Macaca fascicularis) CC chemokine receptor 1 J. Leukoc. Biol., November 1, 2005; 78(5): 1175 - 1184. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Sachon, G. Bolbach, G. Chassaing, S. Lavielle, and S. Sagan Cgamma H2 of Met174 Side Chain Is the Site of Covalent Attachment of a Substance P Analog Photoactivable in Position 5 J. Biol. Chem., December 20, 2002; 277(52): 50409 - 50414. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
