| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Vascular Biology, SmithKline Beecham Pharmaceuticals, The Pinnacles, Harlow, Essex, United Kingdom; and Departments of
Gene Expression Sciences,
Protein Biochemistry, and
§
Molecular and Cellular Immunology, SmithKline Beecham, King of Prussia, PA 19406
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MCP-1,2 MCP-2, MCP-3, MIP-1
, MIP-1ß, RANTES, and
eotaxin represent the most fully characterized CC chemokines in terms
of their biologic function and the receptor types that mediate their
actions (2, 6). Chemokine receptors (CCRs) are seven-transmembrane G
protein-coupled receptors that are sensitive to inhibition by pertussis
toxin. Bioinformatics and degenerate PCR strategies have led to the
identification of new chemokines and novel chemokine receptors. For
example, MIP-3/liver and activation-regulated chemokineliver and
activation-regulated chemokine (7, 8), MIP-3ß/EBI-1 (7, 9), thymus
and activation-regulated chemokine (10), MCP-4 (11, 12),
DC-CK1/pulmonary and activation-regulated chemokine (13, 14),
and CKß-8 (15, 16) are six new CC chemokines identified via these
methodologies. As such, the biologic activity is always characterized
through the use of recombinant chemokine along with the assumption that
the recombinantly expressed form is representative of the
physiologically relevant form that exists in vivo. A certain amount of
assurance can be derived from demonstrating that the recombinant
secreted protein exhibits appropriate biologic activity when used in
low nanomolar concentrations. However, it is important to note that
relatively small alterations in structure can significantly influence
biologic activity. Truncation, for example, is known to significantly
alter CC chemokine potency, as exemplified by work on RANTES (17). In
addition, truncation of the 77-amino acid form of IL-8 to the 72-amino
acid form leads to an increase in IL-8 affinity and efficacy (18), and
a single amino acid mutation in IL-8 completely alters its profile of
leukocyte activation by introducing CC chemokine activity (19).
CKß-8 differs from other CC chemokines in the long sequence of amino
acids preceding the conserved cysteine pair. The coding region carried
by CKß-8 cDNA clones encodes a predicted signal sequence of 21 amino
acids, followed by a 99-amino acid sequence that includes 32 amino
acids in front of the CC pair. Other reported CC, CXC, C, and CXXXC
chemokines typically contain only 5–15 amino acids between the signal
sequence and conserved cysteines. Thus, CKß-8 is a larger than normal
novel CC chemokine consisting of 99 amino acids, with greatest homology
to MIP-1. It has been shown to act weakly on monocytes (15, 16) and
eosinophils, but has also been demonstrated to inhibit hemopoietic
progenitors (15), a characteristic of many CC chemokines. We have
prepared a recombinant 99-amino acid form of CKß-8 from baculovirus
and confirmed that it is weakly active as a chemotactic factor for
human monocytes. We have also observed that in some protein
preparations, several shorter length forms of this chemokine can be
identified. To investigate their potential physiologic relevance we
have individually expressed the three major forms and compared their
biologic activities. The results clearly demonstrate that N-terminal
truncation of CKß-8 not only has a dramatic effect on its biologic
potency, but also influences the overall profile of leukocyte
activation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant human chemokines were purchased from R & D Systems, Europe Ltd. (Abingdon, U.K.). All other reagents were analytical grade.
Generation of 1–99, 24–99, and 25–99 forms of CKß-8 in a baculovirus expression system
A plasmid obtained from Human Genome Sciences (15) comprising the 580 bp of CKß-8 cDNA was cloned between the EcoRI and XhoI sites of pBluescript IISK- (Stratagene, La Jolla, CA). The 580-bp cDNA fragment was released by digestion of the plasmid with EcoRI and XhoI (with the XhoI overhang filled in using T4 DNA polymerase) and was subcloned into the EcoRI and SmaI sites of baculovirus transfer vector pVL1392 (Invitrogen, San Diego, CA). Recombinant baculovirus was generated in vivo in Sf21 cells using Baculogold (PharMingen, San Diego, CA) viral DNA. Sf21 cells were grown in IPL-41 insect medium (catalog no. 96-5104, Life Technologies, Gaithersburg, MD) supplemented with 10% FBS to approximately 1.2–1.5 x 106 cells/ml and infected by addition of amplified virus directly to the culture. After 18–24 h, cells were spun out and resuspended in serum-free culture medium. Growth was continued for 72–96 h, and the secreted chemokine was purified directly from the medium. N-terminal sequencing and matrix-assisted laser desorption mass spectrometry (MALDI-MS) of the purified CKß-8 revealed multiple forms. Some preparations of purified protein resulting from this procedure were found to contain mainly the 1–99 form of the chemokine, whereas other preparations were found to contain a mixture of the 1–99, 24–99, 25–99, and other unidentifiable forms.
Baculovirus transfer vectors designed for expression and secretion of
the truncated forms of CKß-8 encoding amino acids 24–99 and 25–99
were constructed using PCR and subcloned to delete the cDNA coding
region corresponding to amino acids 1–23 and 1–24, respectively,
leaving the truncated CKß-8 in-frame with the signal sequence. The
24–99 vector sequence at the signal sequence/24 junction was
 |
 |
 |
 |
Purification and characterization of recombinant proteins
Baculovirus-infected cell culture medium (10 l) was adjusted to pH 6.0 and a conductivity of 3.0 mS. In some experiments PMSF and benzamidine (1 mM of both) were added to the medium. The chemokine was captured from medium on S-Sepharose FF (Pharmacia, Piscataway, NJ; 2.6 x 11 cm) equilibrated with 20 mM Na phosphate, pH 6.0. The column was washed with 20 mM sodium phosphate, pH 7.0 (buffer P), and the chemokine was eluted with buffer P containing 1 M NaCl. The fractions containing chemokine were pooled (180 ml) and concentrated to 4.5 mg/ml using Amicon YM3 (Amicon, Beverley, MA). The concentrated chemokine solution was applied to Superdex 75 (Pharmacia; 5 x 100 cm) equilibrated with buffer P and eluted with same buffer. In all cases, CKß-8 was eluted as a monomer from Superdex 75, which was previously calibrated with standard mixture of known m.w. proteins. Fractions containing CKß-8 were pooled, concentrated 10-fold to 1.3 mg/ml, and stored -80°C. The endotoxin level of the final product was about 0.25 EU/ml, and the final yield of CKß-8 was approximately 40 mg from 10 l of medium.
The purity of CKß-8 was determined by SDS-PAGE (15%) and analytical
reverse phase HPLC. Approximately 0.1 mg of CKß-8 was applied to a
Vydac C4 (Hesperia, CA) column (4.4 x 250 mm) and
eluted with a gradient of 20–35% acetonitrile in 0.1%
trifluoroacetic acid for 45 min at 1 ml/min. For MALDI-MS analysis, 5
µl of sample was mixed with 10 µl of matrix solution, which was a
saturated solution of -cyano-4-hydroxycinnamic acid (HCCA) in 1%
trifluoroacetic acid in 40% acetonitrile. One microliter of sample
solution was spotted on the sample stage and dried. Mass data were
obtained on a Ciphergen MassPhoresis unit (Iphergen Biosystems,
Palo Alto, CA) and were analyzed using Ciphergen MassPhoresis software
version 1.2. Bovine insulin was used as an external calibrant (MH +5,
734). N-terminal amino acid sequence analysis was performed on an
Applied Biosystems 470A gas phase protein sequencer (Foster City, CA)
equipped with a Beckman 126/166 system (Palo Alto, CA) for on-line
phenylthiohydantain (PTH) analysis. Data were acquired using
System Gold chromatography software (Beckman Instruments, Fullerton,
CA). Samples were spotted directly onto polybrene-coated GF/C
filters (Applied Biosystems), and standard Applied Biosystems
sequencing cycles were used.
Leukocyte isolation
For eosinophil purification, human granulocytes were isolated from heparin-anticoagulated venous blood from allergic volunteers by Percoll (1.090 g/ml) gradient centrifugation at room temperature. After centrifugation, all procedures were conducted at 4°C to minimize cell activation. RBC were removed by hypotonic lysis followed by removal of CD16-positive cells (neutrophils) using an immunomagnetic bead technique (20). Eosinophil purity, based on examination of Diff-Quik-stained cytocentrifugation preparations, was 99 ± 1%, and viability (based on toludine blue dye exclusion) was >98%. Human monocytes were prepared exactly as previously described (12).
Calcium mobilization and chemotaxis experiments
Eosinophil chemotaxis experiments were performed using a modified Boyden chamber technique as described previously (21). Briefly, 25 µl of PAGCM buffer (147 mM NaCl, 5 mM KCl, and 20 mM PIPES (pH 7.4) with 0.03% human serum albumin, 0.1% glucose, 1 mM CaCl2, and 1 mM MgCl2) or various concentrations of the stimuli in the same buffer were placed in triplicate in the lower chamber. A 5-µm pore size polycarbonate membrane (Nucleopore Corp., Cabin John, MD) separated the upper and lower chambers. Eosinophils (2 x 106 cell/ml) suspended in PAGCM were placed in each well of the upper chamber. The chamber was then incubated for 30 min at 37°C in 5% CO2/air, after which the chamber was disassembled. The membrane was removed and washed in PBS to remove the nonmigrating eosinophils from the upper surface, scraped, and then stained with Diff-Quick stain (Gaminor, Abingdon, U.K.). Eosinophils from four high power fields of triplicate wells were identified and counted. Monocyte chemotactic activity was measured as described previously (12). To assess whether CKß-8 induced a chemokinetic rather than a chemotactic effect on leukocyte migration, some experiments were conducted in the absence or the presence of CKß-8 in the upper chamber of the apparatus.
Analysis of intracellular calcium transients in eosinophils was conducted exactly as outlined previously (21). Similarly, monocyte calcium mobilization studies using fura-2-loaded cells followed a previously described protocol (12). In all cases the fluorescence was monitored continuously.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When full-length CKß-8 was expressed in a baculovirus system, the major form of purified product was the 1–99 form, which was confirmed by N-terminal sequencing to begin with the amino acid sequence RVTKDEA. MALDI-MS confirmed the m.w. as 11,361, which was consistent with the predicted N-terminal amino acids and agrees with the previously published form (15, 16). However, it was noted that when CKß-8 was purified from the same expression system without the presence of protease inhibitors (1 mM PMSF and 1 mM benzamidine), only shorter forms of the chemokine were identified by MALDI-MS. Preliminary functional studies indicated that the mixture of short CKß-8 variants was significantly more potent than the full-length chemokine. N-terminal sequencing of these shorter forms identified the most abundant form as beginning at position 25, and less abundant forms beginning at positions 24, 23, and 18. The 25–99, 24–99, 23–99, and 18–99 forms were then expressed individually as recombinant molecules using baculovirus. Subsequent calcium mobilization assays on crude conditioned medium containing each of these forms showed CKß-825–99 to be significantly more potent than the 24–99, 23–99, 18–99, and 1–99 forms (data not shown). We therefore purified the 25–99 and 24–99 recombinant forms for further analysis of biologic activity.
The preparations of CKß-81–99,
CKß-825–99, and CKß-824–99 obtained were
homogeneous as demonstrated by analytical reverse phase HPLC and as a
single band by SDS-PAGE (data not shown). Fig. 1
shows the reverse phase HPLC and
MALDI-MS traces obtained for the purified forms of
CKß-81–99 and CKß-825–99 used in the
functional studies. Furthermore, the N-terminal sequences for
CKß-81–99, CKß-824–99, and
CKß-825–99 were as expected (i.e., RVTKD-, DRFHA-, and
RFHAT-), as were the molecular masses of 11,361, 8,723, and 8,618 Da,
respectively. The alignments of the three different forms of CKß-8
are provided in Fig. 2.
CKß-825–99 has highest homology with MIP-1 (53%
identity in amino acid sequence).
|
|
The three different forms of CKß-8 were compared for their
abilities to stimulate a rise in [Ca2+]i in
freshly prepared human monocytes (Fig. 3)
and eosinophils (Fig. 4). In agreement
with others (15), addition of CKß-81–99 induced a
transient rise in monocyte [Ca2+]i that was
dose dependent, with a half-maximal effective concentration
(EC50) of 150 ± 19 nM (n = 3). In
stark contrast, CKß-825–99 was significantly more potent
in this assay, with an EC50 value of 0.38 ± 0.04 nM
(n = 4), which compared favorably to that observed for
MIP-1, i.e., 0.27 ± 0.03 nM (n = 4).
CKß-824–99 was intermediate among the three forms of
CKß-8 evaluated in this assay (Fig. 3).
|
|
). CKß-81–99 was essentially inactive in eliciting a
calcium transient in eosinophils (EC50 = >100 nM), in
complete agreement with others (15). CKß-825–99, on the
other hand, was capable of stimulating a rise in
[Ca2+]i that was larger in magnitude than
that produced by eotaxin, a known activator of eosinophils.
CKß-824–99, although able to promote a calcium transient
in eosinophils, was less potent than CKß-825–99 (Fig. 4). Although MIP-1 was also observed to promote a calcium transient
in eosinophils, the magnitude of response was much smaller than that
produced by either eotaxin or CKß-825–99. The
EC50 values for CKß-81–99,
CKß-824–99, CKß-825–99, eotaxin, and
MIP-1 in this assay were >100, 9 ± 5.0, 2 ± 0.5,
0.5 ± 0.25, and 3 ± 0.24 nM, respectively. Thus, in
monocytes and eosinophils, the shortest form of CKß-8 was the most
potent variant of this chemokine for calcium mobilization. CKß-8 induced monocyte and eosinophil chemotaxis
In keeping with calcium mobilization studies,
CKß-825–99 was >3 orders of magnitude more potent than
CKß-81–99 in monocyte chemotaxis assays (Fig. 5). Indeed, with an EC50
value of 54 ± 5 pM, CKß-825–99 would appear to be
one of the most potent chemokine monocyte chemoattractants yet
characterized. For comparison, the EC50 values for
CKß-81–99 and MIP-1 in this assay were 80 ± 7
and 0.25 ± 0.06 nM, respectively (Fig. 5).
|
) that was smaller in magnitude than
that produced by eotaxin. This result was also observed for MIP-1.
Interestingly, these two chemokines share the same receptor, CCR1,
suggesting that CCR1 may not be as tightly linked to a chemotactic
response in human eosinophils as the CCR3 receptor.
|
Stimulation of chemokine receptors with an agonist typically
renders the same receptor refractory to further stimulation with the
same or another agonist that binds the same receptor. We have used this
phenomenon in an attempt to characterize receptor usage by
CKß-825–99. Chemokine concentrations that gave a maximal
[Ca2+]i were used throughout, and the results
of these cross-desensitization experiments for both monocytes and
eosinophils are shown in Figs. 7 and
8, respectively.
|
or
RANTES (Fig. 7, C and I). These data indicate
that CKß-825–99 must recognize at least CCR1
and CCR5. However, prior stimulation with either MIP-1 or RANTES
only partially abolished the
CKß-825–99-mediated response (Fig. 7, B and H), suggesting the presence of a potential
novel receptor that does not bind MIP-1 or RANTES. This additional
receptor may be common for both MCP-1 and MCP-3, since partial
desensitization was observed for these chemokines when used in
conjunction with CKß-825–99. It should be
noted, and is in agreement with others, that MCP-1 and MIP-1 did not
desensitize each other in these experiments and as such acted as a
suitable control for nondesensitizing chemokines (data not shown).
Pre-exposure of eosinophils to CKß-825–99
completely desensitized the cell to a second addition of
CKß-825–99 (Fig. 8A). Similarly,
CKß-825–99 and MIP-1 completely
cross-desensitized each other (Fig. 8, C and D),
suggesting that MIP-1 and CKß-8 share a common receptor on
eosinophils, probably CCR1. RANTES also partially cross-desensitizes
CKß-8 (Fig. 8, G and H), while MCP-3 (Fig. 8, A and B) failed to desensitize CKß-8, even
though both RANTES and MCP-3 are reported to bind CCR1. This may
reflect the fact that both RANTES and MCP-3 are weaker agonists at the
CCR1 receptor than CKß-8 and therefore cannot completely desensitize
the receptor or that RANTES and MCP-3 can bind CCR3 in addition to
CCR1, thus affording a second receptor for these chemokines (22). This
is further supported by the observation that eotaxin and CKß-8 failed
to completely cross-desensitize each other in the eosinophil.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, with the
majority of identity lying between the 25–99 amino acid region (53%
identity). By simply removing the first 24 amino acids of CKß-8 we
have observed increases in potency of >2 and 3 orders of magnitude for
the stimulation of monocyte calcium mobilization and chemotaxis,
respectively. Removal of the initial 23 amino acids, on the other hand,
provided only an intermediate improvement in potency. Indeed,
CKß-825–99 is currently the most potent
chemokine monocyte chemoattractant we have yet encountered. Truncation
similarly transformed an essentially inactive chemokine on eosinophils
to one that compared favorably with eotaxin in stimulating calcium
transients. Interestingly, while CKß-825–99
was able to elicit a rise in eosinophil
[Ca2+]i, it failed, unlike eotaxin, to induce
a substantial chemotactic response. This failure to promote strong
chemotaxis is shared by other chemokines that act solely on the CCR1
receptor (i.e., MIP-1), but not by RANTES and MCP-3, which act on
both the CCR1 and CCR3 receptors (22). This suggests that the CCR1
receptor in eosinophils may not function as the primary chemotactic
receptor in these cells.
Calcium studies with CKß-8 have provided an initial insight into its
receptor usage. Desensitization studies with MIP-1 indicate that it
and CKß-8 share the same receptor on both monocytes and eosinophils.
The CCR1 receptor would represent the most obvious candidate common for
both monocytes and eosinophils, but CCR5 could be another monocyte
receptor for CKß-825–99. The lack of
cross-desensitization with eotaxin certainly eliminates an involvement
of the CCR3 receptor in eosinophils (22). Furthermore, the ability to
desensitize RANTES is deemed further evidence for CKß-8 operating via
the CCR1 receptor on these two cell types. However, in both monocytes
and eosinophils RANTES failed to desensitize CKß-8, suggesting that
either RANTES does not completely desensitize all CCR1 receptors, or
that CKß-8 may operate through an additional novel receptor (16).
Another receptor for CKß-8 on monocytes was suggested by the partial
desensitization of MCP-1 and MCP-3. Clarification must await results
from direct binding studies using radiolabeled chemokine ligands on
both leukocytes and transfected cell populations.
The cDNA clone of CKß-8 used in this and previous (15) studies was isolated from an aortic endothelial library. Rescreening of a large database of human expressed sequence tags (ESTs) has allowed identification of seven additional CKß-8 ESTs. Two of the ESTs (one each from aortic endothelial and serum treated smooth muscle libraries) appeared to encode the same sequence as the cDNA used in this study. The other five ESTs (from activated neutrophil, thymus, and serum-treated smooth muscle libraries) had insertions of 51–54 additional nucleotides in the coding sequence (data not shown). These insertions were within the codon for amino acid 25 of the 1–99 form and potentially could lead to secretion of a form with an even longer N-terminus than the 1–99 form. Confirmatory DNA sequencing of the EST clones and expression of protein from those clones are required to explore this possibility.
The N-terminal regions of most chemokines contain 5–15 amino acids preceding the first conserved cysteine in the secreted proteins. CKß-81–99 is exceptional among CC chemokines in the length of this region (32 amino acids preceding the first cysteine). This is reminiscent of the extended N-terminus of platelet basic protein, a CXC chemokine that has 28 amino acids in this position. The observation that truncation of this region in CKß-8 leads to a dramatic change in activity is consistent with findings previously reported for some other chemokines. For example, truncations of the CXC chemokines platelet basic protein and IL-8 show increased biologic activities (18, 23, 24), while deletions and extensions of the CC chemokines MCP-1 and RANTES have been described that block biologic activity and convert the molecules into receptor antagonists (17, 25, 26, 27, 28).
Examination of DNA sequence databases revealed several other cDNA sequences that encode potential chemokines with very long N-termini. The human sequences in GenBank accession numbers z70292 and z70293 encode a chemokine (CC-2) that is very closely related to the long form of CKß-8 (CC-2 is also designated MIP-5 and HCC2) (29, 30). At the amino acid level, CC-2 is 90% identical with CKß-8 in the signal sequence, 75% identical in the 1–24 region of the secreted protein, and 34% identical in the remainder of the molecule (23 amino acids identical between the 25–99 region of CKß-8 and the 25–92 region of CC-2). All six cysteines in the putative secreted forms of these proteins are located at corresponding positions, including two extra cysteines that are not among the four typical cysteines found in most chemokines. Moreover, the CC-2 coding sequence appears in two different GenBank entries. In one of them (z70292), it appears upstream of chemokine CC-1 (HCC-1) (31) in a dicistronic messenger RNA. In the other entry (z70393), CC-2 appears upstream of a variant of CC-1 (called CC-3) that is identical with CC-1 except that it encodes an extra 17 amino acids in the region more or less corresponding to the 1–24 region of CKß-8. CC-1 has 15 amino acids between the signal sequence and the conserved cysteines, while CC-3 has 32 amino acids in that region. Thus, CC-3 may be yet a further example of a chemokine that can be secreted in a form with an unusually long N-terminus. In addition, a murine chemokine genomic clone encoding chemokine C10 has been described that has a novel second exon that places 16 extra amino acids in between the signal sequence region and the conserved ß-chemokine region (32). This, too, could potentially lead to secretion of a form with a long N-terminal extension.
These examples may define a new subfamily of chemokines that are secreted in an inactive pro form that might be activated by proteolytic removal of the N-terminal extension. However, it remains to be demonstrated whether any of the long forms or short forms can be detected in tissues and associate with physiologic responses. Additional experiments are underway in an attempt to address whether CKß-825–99 can be generated in vivo.
|
Footnotes |
|---|
2 Abbreviations used in this paper: MCP, monocyte chemoattractant protein; MIP, macrophage inflammatory protein; CCR, chemokine receptor; MALDI-MS, matrix-assisted laser desorption mass spectrometry; [Ca2+]i, cytosolic free calcium concentration; EC50, half-maximal effective concentration; EST, expressed sequence tag.
Received for publication December 22, 1997. Accepted for publication July 27, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and MIP-3ß. J. Immunol. 158:1033.[Abstract]
/LD78 and chemotactic for T lymphocytes, but not for monocytes. J. Immunol. 159:1140.[Abstract]
gt11 expression library. Blood 73:1498.This article has been cited by other articles:
|
|
 |
|
  Z. Miao, B. A. Premack, Z. Wei, Y. Wang, C. Gerard, H. Showell, M. Howard, T. J. Schall, and R. Berahovich Proinflammatory Proteases Liberate a Discrete High-Affinity Functional FPRL1 (CCR12) Ligand from CCL23 J. Immunol., June 1, 2007; 178(11): 7395 - 7404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Vergote, G. S. Butler, M. Ooms, J. H. Cox, C. Silva, M. D. Hollenberg, J. H. Jhamandas, C. M. Overall, and C. Power Proteolytic processing of SDF-1{alpha} reveals a change in receptor specificity mediating HIV-associated neurodegeneration PNAS, December 12, 2006; 103(50): 19182 - 19187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Berahovich, Z. Miao, Y. Wang, B. Premack, M. C. Howard, and T. J. Schall Proteolytic Activation of Alternative CCR1 Ligands in Inflammation J. Immunol., June 1, 2005; 174(11): 7341 - 7351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Forssmann, H.-J. Magert, K. Adermann, S. E. Escher, and W.-G. Forssmann Hemofiltrate CC chemokines with unique biochemical properties: HCC-1/CCL14a and HCC-2/CCL15 J. Leukoc. Biol., September 1, 2001; 70(3): 357 - 366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Zabel, W. W. Agace, J. J. Campbell, H. M. Heath, D. Parent, A. I. Roberts, E. C. Ebert, N. Kassam, S. Qin, M. Zovko, et al. Human G Protein-coupled Receptor GPR-9-6/CC Chemokine Receptor 9 Is Selectively Expressed on Intestinal Homing T Lymphocytes, Mucosal Lymphocytes, and Thymocytes and Is Required for Thymus-expressed Chemokine-mediated Chemotaxis J. Exp. Med., November 1, 1999; 190(9): 1241 - 1256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. White, J. M. Lee, K. Dede, C. S. Imburgia, A. J. Jurewicz, G. Chan, J. A. Fornwald, D. Dhanak, L. T. Christmann, M. G. Darcy, et al. Identification of Potent, Selective Non-peptide CC Chemokine Receptor-3 Antagonist That Inhibits Eotaxin-, Eotaxin-2-, and Monocyte Chemotactic Protein-4-induced Eosinophil Migration J. Biol. Chem., November 17, 2000; 275(47): 36626 - 36631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Burns, D. J. Dairaghi, M. Deitz, M. Tsang, and T. J. Schall Comprehensive Mapping of Poxvirus vCCI Chemokine-binding Protein. EXPANDED RANGE OF LIGAND INTERACTIONS AND UNUSUAL DISSOCIATION KINETICS J. Biol. Chem., January 18, 2002; 277(4): 2785 - 2789. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
