B cell- and monocyte-activating chemokine (BMAC), a novel non-ELR α-chemokine

Abstract

A novel alpha-chemokine, designated KS1, was identified from an EST database of a murine immature keratinocyte cDNA library. The EST has 94% similarity to a recently cloned human gene, BRAK, that has no demonstrated function. Northern analysis of mouse and human genes showed detectable mRNA in brain, intestine, muscle and kidney. Tumour panel blots showed that BRAK was down-regulated in cervical adenocarcinoma and uterine leiomyoma, but was up-regulated in breast invasive ductal carcinoma. KS1 bound specifically to B cells and macrophages, as well as two B cell lines, CESS and A20, and a monocyte line, THP-1. KS1 showed no binding to naive or activated T cells. In addition, KS1 stimulated the chemotaxis of CESS and THP-1 cells but not T cells. The s.c. injection of KS1 creates a mixed inflammatory response in Nude and C3H/HeJ mice. The above data indicates that KS1 and its human homologue represents a novel non-ELR alpha-chemokine that may have important roles in trafficking of B cells and monocytes. We propose the name B cell- and monocyte-activating chemokine (BMAC) for this molecule to reflect the described biological functions.

Full text

International Immunology, Vol. 12, No. 5, pp. 677–689 © 2000 The Japanese Society for Immunology
B cell- and monocyte-activating chemokine
(BMAC), a novel non-ELR α-chemokine
Matthew A. Sleeman, Jonathon K. Fraser, James G. Murison, Sharon L. Kelly,
Ross L. Prestidge, David J. Palmer, James D. Watson and
Krishnanand D. Kumble
Genesis Research and Development Corp. Ltd, PO Box 50, Auckland, New Zealand
Keywords: chemotaxis, inflammation, migration, nude mice, tumour
Abstract
A novel α-chemokine, designated KS1, was identified from an EST database of a murine immature
keratinocyte cDNA library. The EST has 94% similarity to a recently cloned human gene, BRAK,
that has no demonstrated function. Northern analysis of mouse and human genes showed
detectable mRNA in brain, intestine, muscle and kidney. Tumour panel blots showed that BRAK
was down-regulated in cervical adenocarcinoma and uterine leiomyoma, but was up-regulated in
breast invasive ductal carcinoma. KS1 bound specifically to B cells and macrophages, as well as
two B cell lines, CESS and A20, and a monocyte line, THP-1. KS1 showed no binding to naive or
activated T cells. In addition, KS1 stimulated the chemotaxis of CESS and THP-1 cells but not
T cells. The s.c. injection of KS1 creates a mixed inflammatory response in Nude and C3H/HeJ
mice. The above data indicates that KS1 and its human homologue represents a novel non-ELR
α-chemokine that may have important roles in trafficking of B cells and monocytes. We propose
the name B cell- and monocyte-activating chemokine (BMAC) for this molecule to reflect the
described biological functions.
Introduction
Chemokines are a large family of small peptides that are
involved in the trafficking of leukocytes around the body. They
consist of proteins between 8 and 12 kDa in size with a
number of conserved cysteines that form two disulphide
bridges (1–3). The chemokine superfamily is currently classi-
fied with respect to the number and position of the first
cysteines, CC, CXC, CX
3
C and C. The two main groups are
(i) the CXC or α-chemokines, defined by a single amino
acid separating the first two cysteines, and (ii) the CC or
β-chemokines, with the first two cysteines being contiguous.
The α-chemokines can be further subdivided into two groups
depending on whether they contain a Glu–Leu–Arg (ELR)
motif immediately prior to the first cysteine (1). Initially, these
molecules were thought to only be involved in stimulating an
inflammatory response by promoting chemotaxis of leuko-
cytes from the peripheral blood to sites of inflammation. IL-8
was one of the first chemokines identified and was shown to
promote neutrophil migration (4–6). Since then the chemokine
family has grown to ⬎50 members (http://cytokine.medic.kum-
amoto-u.ac.jp/CFC/CK/chemokine.html) with every leukocyte
Correspondence to: M. Sleeman
Transmitting editor: M. Feldmann Received 10 September 1999, accepted 28 January 2000
population having its own particular subset of chemokines and
chemokine receptors. The non-ELR α-chemokines currently
consist of six members whose chemotactic functions are
highly diverse (7–13), in contrast to the neutrophil migration-
promoting ELR chemokines, and are of great interest for their
therapeutic potential in areas other than leukocyte migration.
PF-4, IP-10 and Mig have all been shown to have anti-
angiogenic properties in a range of tumour models (14–16).
SDF-1α has been shown to competitively block viral entry in
HIV strains that uniquely use CXCR4 as their co-receptor for
infection (17,18). The potential of this therapeutic approach
is supported by the observation that high circulating levels of
β-chemokines can confer a degree of immunity on those
exposed to HIV (19). Recently, a non-ELR α-chemokine,
BRAK, was identified by screening human EST databases
(20). Function has yet to be assigned for this molecule,
although it has been postulated to have a role in oncogenesis.
We have identified the murine homologue of this gene, the
responding cell types for this new chemokine and propose a
new name that reflects its biological activity.

678 B cell- and monocyte-activating chemokine (BMAC)
Methods
Chemicals and reagents
Recombinant human stromal derived factor-1α and human
IL-2 were purchased from PeproTech (Rocky Hill, NJ). The
following primary anti-murine antibodies were obtained from
PharMingen (San Diego, CA), I-A
k
(A
α
k
) biotin (clone 11-5.2),
CD19–FITC (Clone 1D3), CD4–FITC (clone RM4-5), CD8a–
FITC (clone 53-6.7), rat IgG2a–FITC (R35-95) and mouse
IgG2b–biotin (clone 49.2). The secondary antibody goat anti-
human IgG–phycoerythrin (PE) and streptavidin–PE were
purchased from Southern Biotechnology Associates
(Birmingham, AL), and the streptavidin–alexa 488 from
Molecular Probes (Eugene, OR).
Bioinformatic analysis
An oligo-d(T)-primed directionally cloned murine immature
keratinocyte cDNA library was constructed from poly(A)
⫹
RNA using a ZAP express cDNA kit (Stratagene, La Jolla,
CA) following the manufacturer’s protocol. The library was
mass excised and colonies randomly selected for sequencing.
High-throughput single-pass sequence from the 5⬘ end of the
clones was obtained on ABI377 sequencers (Perkin Elmer,
Foster City, CA). Novel sequences were analysed using
BLAST (21), Prosite (Swiss Institute of Bioinformatics,
University of Geneva) and SignalP V1.1 (Center for Biological
Sequence Analysis, Technical University of Denmark), and
the Phylip package (University of Washington) to define
similarities to known gene families or motifs.
Sequence and cloning of KS1
The full-length sequence of KS1 was obtained by subcloning
and sequence primer walking. The coding region, without the
predicted signal sequence, was PCR amplified using Klentaq
polymerase (Clontech, Paolo Alto, CA) and KS1 as template,
using the following sequences 5⬘-CATGCCATGGCGTCCAA-
GTGTAAGTGTTCCCGGAAGGGG-3⬘ and 5⬘-CATGCCATGG-
CTAATGGTGGTGATGGTGATGTTCTTCGTAGACCCTGCGC-
TTCTC-3⬘ as forward and reverse oligonucleotides respect-
ively. The product was purified using a PCR purification kit
(Qiagen, Valencia, CA), digested with
Nco
I and ligated into
pET16B (Novagen, Madison, WI) to obtain the sequence in
frame with the C-terminal (His)
6
tag. In addition to this
we cloned the full-length coding region into a eukaryotic
expression vector, pIGFc, using 5⬘-GGAATTCCATGAGGCT-
CCTGGCGGCCGCGCTGCTC-3⬘ and 5⬘-ACGGATCCACTTA-
CCTGTTTCTTCGTAGACCCTGCGCTTCTCGTT-3⬘ as forward
and reverse primers respectively. PCR products were pre-
pared as above and ligated into pIGFc to obtain the sequence
in-frame with human IgG1 Fc present in the vector. All
constructs were confirmed by automated sequencing.
Northern analysis
KS1 probe was PCR amplified using Taq polymerase (Qiagen)
with 5⬘-ACGCGTCGACATGAGGCTCCTGGCGGC-3⬘ and 5⬘-
TCGTCCAGATCTTTCTTCGTAGACCCTGCGCTT-3⬘ as for-
ward and reverse oligonucleotides respectively. BRAK probe
was PCR amplified from human keratinocyte cDNA using Taq
polymerase (Qiagen) with 5⬘-ACGCGTCGACATGAGGCTCC-
TGGCGGCCGCGCTGCTC-3⬘ and 5⬘-ATAAGATCTTTCTTCG-
TAGACCCTGCGCTTC-3⬘ as forward and reverse
oligonucleotides respectively. Probe identity was confirmed
by sequencing. PCR products were labelled with [α-
32
P]dCTP
(3000 Ci/mmol, NEN/Life Science products, Boston, MA)
using 25 ng of DNA in a Rediprime II random-primed labelling
system (Amersham Pharmacia, Piscataway, NJ). Human mul-
tiple tissue northern blots (Clontech) were hybridized with a
300 bp PCR product (nucleotides 1–300 bp of the BRAK
coding sequence) following the manufacturer’s protocol
(Clontech). A human tumour panel blot (Invitrogen, Calsbad,
CA) was hybridized with the probe prepared as described
above, in 6⫻SSC buffer, 2⫻Denhardt reagent, 2% SDS, 120
µg heparin and 100 µg yeast tRNA (Boehringer Mannheim,
Mannheim, Germany) at 65°C for 18 h. RNA for mouse tissue
blots was isolated using Trizol reagent (Life Technologies,
Grand Island, NY) and 20 µg total RNA loaded per lane in a
1% formaldehyde agarose gel, transferred to Hybond N⫹
membrane (Amersham) and hybridized with the radiolabelled
PCR product (nucleotides 1–300 bp of the KS1 coding
sequence). Mouse tissue blots were hybridized as described
for the tumour panel blots. All blots were washed under
stringent conditions as specified by the manufacturers or by
standard protocols (22). Northern blots were exposed to X-ray
film at –80°C and developed at various times up to 7 days.
Both the tumour panel blot and human tissue blots were re-
probed, as described previously, with a 500 bp β-actin probe
as a loading control.
Expression and purification of recombinant KS1
A C-terminal (His)
6
tag fusion protein of KS1 was expressed
in BL21(DE3) Escherichia coli cells (Novagen). One litre
cultures were induced at an OD
600
of 0.5 with 1 mM IPTG
and harvested after 3 h. All subsequent procedures were
performed on ice. The pellet was re-suspended in lysis
buffer (20 mM Tris–HCl, pH 8.0, 1 mM PMSF, 10 mM β-
mercaptoethanol, 1% NP-40) and sonicated using a Virsonic
ultrasonicator (Virtis, Gardiner, NY) fitted with the miniprobe
at 20% output for 4⫻15 s bursts with 15 s intervals. The
sonicate was centrifuged in a JA20 rotor at 18,000 r.p.m.
for 10 min at 4°C. The resultant pellet was washed twice
for 1 h each in lysis buffer containing 0.5% CHAPS and
solubilized in 20 mM Tris–HCl, pH 8.0, containing 6 M
guanidine–HCl and 0.5 M NaCl. The (His)
6
fusion protein
was isolated by chromatography using nickel chelating
Sepharose FF resin (0.5 ml column; Pharmacia). After
loading, the column was washed sequentially with 20
volumes of binding buffer (6 M urea, 0.5 M NaCl and 20
mM Tris–HCl, pH 8.0), 20 volumes of 0.5% sodium
deoxycholate in binding buffer and 20 volumes of binding
buffer containing 20 mM imidazole. The protein was eluted
with 10 volumes 300 mM imidazole in binding buffer.
The eluate was dialysed against binding buffer and re-
chromatographed as above. Fusion protein in the eluate
was then refolded by dialysis against 1 l of 4 M urea,
20 mM Tris–HCl, pH 7.5, overnight while 1 l of 20 mM
Tris–HCl, pH 7.5, was pumped into the dialysis beaker
at a rate of 1 ml/min. The refolded protein was finally
dialysed against 20 mM Tris–HCl, pH 7.5, containing 10%
(w/v) glycerol. Preparations obtained were ⬎95% pure as
determined by SDS–PAGE using FragmeNT Analysis

B cell- and monocyte-activating chemokine (BMAC) 679
Fig. 1. (A) The nucleotide sequence of KS1 cDNA is shown along
with the deduced amino acid sequence using the single letter code.
The 5⬘ untranslated region is indicated by negative numbers. The
underlined N-terminal amino acids represent the predicted leader
sequence and the stop codon is denoted by ‘***’. The poly-adenylation
signal is marked by a double underline. The sequence data is
available from GenBank under accession no. AF144754. (B)
Comparison of the complete open reading frame of KS1 with its
human homologue BRAK and with the mouse α-chemokines mCrg-
2, mMig, mSDF-1, mBLC, mMIP-2, mKC and mLIX. An additional five
residues are present in KS1 and BRAK between cysteine 3 and
cysteine 4 that have not previously been described for chemokines.
(C) A phylogenetic tree of KS1 was constructed against current
murine α-chemokines using Phylip software version 3.57c, and
programs protdist and neighbour joining. The figure represents the
degree of divergence between each of the family members. The
branch lengths are proportional to the numbers of substitutions,
based on the amino acid homology, the level of conservation between
the different amino acid residues and the rate of evolution. GenBank
accession nos for the sequences are (from top to bottom): L12030,
M34815, M86829, U27267, J04596, X53798, AF044196, AF073957
and AF144754.

680 B cell- and monocyte-activating chemokine (BMAC)
Package (Molecular Dynamics, Sunnyvale, CA). Endotoxin
contamination of purified KS1 was determined using a
limulus amebocyte lysate assay kit (Biowhittaker, Walkersville,
MD). Endotoxin levels were ⬍0.1 ng/µg of protein. Internal
amino acid sequencing was performed on tryptic peptides
of KS1 by the Protein Sequencing Unit at the University of
Auckland, New Zealand.
An Fc fusion protein was produced by expression in HEK
293 T cells. Using 35 µg of KS1pIGFc DNA to transfect 6⫻10
6
cells/flask, 200 ml of KS1 Fc-containing supernatant was
produced. The Fc fusion protein was isolated by chromato-
Fig. 2. (A). Northern blot analysis of KS1 mRNA in various murine tissues. The upper panel shows the level of expression in these tissues,
whereas the lower panel illustrates equal loading of total RNA from these tissues. The position of the 1.35 kb RNA marker is indicated as is
the position of 1.6 kb KS1; 28S and 18S ribosomal bands are also indicated. (B) Northern blot of BRAK mRNA in various human tissues as a
comparison with murine expression. Human multiple tissue blots were purchased from Clontech. (C) Northern blot analysis of BRAK mRNA in
tumor versus normal tissue. Tumor panel blots were purchased from Invitrogen. The northern blot directly compares four different tumors with
their respective normal tissue. The upper panels of the human blots shows the level of expression of BRAK, whereas the lower panel
demonstrates the level of β-actin expression. The position of the 1.4 kb RNA marker and 1.8 kb BRAK message is indicated
graphy using an Affiprep Protein A resin (0.3 ml column;
BioRad, Hercules, CA). After loading, the column was washed
with 15 ml of PBS, followed bya5mlwashof50mMNa
citrate, pH 5.0. The protein was then eluted with 6 column
volumes of 50 mM Na citrate, pH 2.5, collecting 0.3 ml
fractions in tubes containing 60 µlof20mMTris–HCl, pH 7.5.
Fractions were analyzed by SDS–PAGE and pooled.
Cell isolation and culture
Murine spleens, thymus, peripheral lymph node and bone mar-
row cells for flow cytometric analysis were obtained from C3H/

B cell- and monocyte-activating chemokine (BMAC) 681
HeJ mice, erythrocytes were lysed using ACK lysis buffer (0.15
MNH
4
Cl, 1 mM KHCO
3
and 0.1 mM Na
2
EDTA). Peritoneal
exudate cells (PEC) were obtained by i.p. lavage from C3H/
HeJ mice. In brief, euthanized mice were injected with 2⫻4ml
volumes of 0.02% EDTA/PBS into the peritoneal cavity using an
18 gauge needle. Cells were then drawn out from the peritoneal
Fig. 3. Analysis of purified (His)
6
KS1 fusion protein by SDS–PAGE.
Protein was resolved on a 12% acrylamide denaturing gel and stained
with Coomassie blue. Lane 1, mol. wt standards; lane 2, 5 µgof
purified (His)
6
KS1 protein.
Fig. 4. Flow cytometric analysis of KS1Fc binding to murine splenocytes, peripheral lymph node cells and thymocytes. Cells were labelled
with KS1Fc or negative control protein EGBFc and visualized with a two-step staining procedure using goat anti-human–PE. Ten thousand
gated events were analysed for each of the experiments. (A) Binding of KS1Fc was detected on murine splenocytes and (B) peripheral lymph
node cells as compared with the negative control, EGBFc. (C) Alternatively, KS1Fc showed no binding to thymocytes when compared to
negative control. The phenotype of the KS1Fc
⫹
splenocytes was determined using two-color analysis with the following antibody markers. (D)
Murine splenocytes were double positive for KS1Fc and CD19 (D) but not for CD4 (E) or CD8a (F) cells. KS1Fc (solid line), EGBFc (dotted line).
cavity, pelleted and washed in PBS prior to further analysis.
Murine IL-2-activated T cells were cultured as described below.
Briefly, splenocytes were activated with 2 µg/ml concanavalin
A (Con A) (Sigma, St Louis, MO) in the presence of 5% FBS in
DMEM supplemented with 2 mM
L
-glutamine (Sigma), 1 mM
sodium pyruvate (Life Technologies), 0.77 mM
L
-asparagine
(Sigma), 0.2 mM arginine (Sigma), 160 mM penicillin G (Sigma),
70 mM dihydrostreptomycin sulfate (Boehringer Mannheim)
and 50 µM 2-mercaptoethanol, for 3 days followed by addition
of recombinant human IL-2 (PeproTech) at 10 ng/ml. Cytokine
was added at 3 day intervals for 9–21 days. Peripheral blood
mononuclear cells (PBMC) were isolated in heparin (10U/ml)
containing tubes from human donors and purified on a Ficoll-
Hypaque (Pharmacia) gradient by centrifugation at 900 g for
20 min with no brake. PBMC were aspirated from the interface,
washed and re-suspended in HBSS, 20 mM HEPES, 0.5% BSA
and used directly for assays. Human IL-2-activated T cells were
cultured as described below. Briefly, PBMC were activated with
0.1% phytohemagglutinin (PHA) (Gibco/BRL) in the presence
of 5% FBS in RPMI supplemented with 2 mM
L
-glutamine
(Sigma), 160 mM penicillin G (Sigma), 70 mM dihydrostrepto-
mycin sulphate (Boehringer Mannheim) and 50 µM 2-mer-
captoethanol, for 3 days followed by addition of recombinant
human IL-2 (PeproTech) at 10 ng/ml. CESS, THP-1 and Jurkat

682 B cell- and monocyte-activating chemokine (BMAC)
Fig. 5. Flow cytometric analysis of KS1Fc binding to monocytes from PEC. Ten thousand gated events were analysed for each of the
experiments. Forward (FS) versus side scatter (SS) histogram of PEC (A). All subsequent histograms were gated on region H. Cells from this
region were CD19
–
(B) [CD19 (solid line), rIgG2a (dotted line)] and MHC class II
⫹
(C) [MHC class II (solid line), mIgG2a (dotted line)]. Two-
colour analysis shows that MHC class II
⫹
cells are positive for KS1Fc (D) and not the control protein, EGBFc (E). The matching isotype for
MHC class II, mIgG2a, showed no non-specific binding (F).
cells were maintained in complete RPMI as described previ-
ously, whereas A20 cells were grown in 5% FBS in DMEM
supplemented with 2 mM
L
-glutamine (Sigma), 1 mM sodium
pyruvate (Life Technologies), 0.77 mM
L
-asparagine (Sigma),
0.2 mM arginine (Sigma), 160 mM penicillin G (Sigma) and 70
mM dihydrostreptomycin sulphate (Boehringer Mannheim).
Flow cytometric binding studies
Binding of KS1 to cells was tested in the following manner. Cells
(5⫻10
5
) were resuspended in 3 ml of wash buffer (2% FBS and
0.2% sodium azide in PBS) and pelleted at 4°C, 200 g for 5
min. Ig Fc receptors were blocked with 1% goat serum in
wash buffer for 30 min on ice. Cells were washed, pelleted, re-
suspended in 50 µl of KS1Fc at 10 µg/ml and incubated for
30 min on ice. After incubation the cells were prepared as
before and resuspended in 50 µl of goat anti-human IgG–PE at
1 µg/ml and incubated for 30 min on ice. Cells were washed
and resuspended in 250 µl of wash buffer containing 40 ng/ml
propidium iodide (Sigma) to exclude any dead cells. A purified
Fc tagged plant protein (EGBFc) was used, at 10 µg/ml, as a
negative control in place of KS1Fc to determine non-specific
binding. For two-colour staining, cells were incubated with one
of the following antibodies prior to staining with KS1Fc or
EGBFc, anti-CD4–FITC, anti-CD8a–FITC, anti-CD19–FITC and
rat IgG2a–FITC at 10 µg/ml. Biotinylated Ia
k
and its isotype
control, mouse IgG2b–biotin, were used to identify MHC class
II
⫹
cells and detected using streptavidin–alexa 488. Ten thou-
sand gated events were analysed on a log scale using a FITC,
PE and propidium iodide filter arrangement with peak trans-
mittance at 525, 575 and 675 nm respectively with a bandwidth
of 10 nm on an Elite cell sorter (Coulter, Hileah, FL). Todetermine
KS1Fc binding to human cell lines CESS, THP-1 and Jurkat,
and to reduce level of non-specific binding, both KS1Fc and
control protein EGBFc were biotinylated using the Sigma
biotinylation kit (Sigma BK-101) as described in the manufac-
turer’s protocols. Human cells were labelled with KS1Fc–biotin
or EGBFc–biotin as described previously and then detected
with streptavidin–PE. Cold competition was performed by add-
ing various concentrations of (His)
6
KS1 at 4°C as a competitor
prior to labelling with KS1Fc. An equivalent concentration of
(His)
6
GV14B, an identically expressed unrelated bacterial pro-
tein, was used as control in competition experiments.
Chemotaxis assays
Cell migration in response to KS1 was tested using a 48-well
Boyden chamber (Neuro Probe, Cabin John, MD) as described

B cell- and monocyte-activating chemokine (BMAC) 683
Fig. 6. Flow cytometric analysis of KS1Fc binding to murine and human cell lines. Murine cells were labelled with KS1Fc or negative control
protein EGBFc and visualized with a two-step staining procedure using goat anti-human IgG–PE. Human cell lines were labelled with KS1Fc–
biotin or negative control EGBFc–biotin and visualized with a two-step staining procedure using streptavidin–PE. Ten thousand gated events
were analysed for each of the experiments. Enhanced KS1Fc binding was detected on A20 (A), CESS (B) and THP-1 (C) cells but not on Con
A IL-2-activated T cells (D) or Jurkat T cells (E). KS1Fc (solid line), EGBFc (dotted line).
Fig. 7. Cold competition of KS1Fc binding with increasing
concentrations of (His)
6
KS1. KS1Fc binding on murine splenocytes
was inhibited by increasing concentrations of (His)
6
KS1 protein,
whereas it was not influenced by the negative control protein,
(His)
6
GV14B. Ten thousand gated events were analysed for each
experiment. Values are the geometric mean channel fluorescence ⫾
SD obtained for duplicate samples and are representative of two
individual experiments.
in the manufacturer’s protocol. In brief, agonists were
diluted in HBSS, 20 mM HEPES, 0.5% BSA and added to
the bottom wells of the chemotactic chamber. Cells were re-
suspended in the same buffer at 3⫻10
5
cells/50 µl. Top and
bottom wells were separated by a PVP-free polycarbonate filter
with a 5 µm pore size for CESS and THP-1 cells or 3 µm pore
size for splenocytes and lymphocytes. Cells were added to the
top well and the chamber incubated for 2 h for THP-1 and 4 h
for CESS cells, splenocytes and lymphocytes in a 5% CO
2
humidified incubator at 37°C. After incubation the filter was
fixed and cells scraped from the upper surface. The filter was
then stained with Diff-Quik (Dade Behring Diagnostics, Deer-
field, IL) and the number of migrating cells counted in five
randomly selected high-power fields. The results are
expressed as a migration index defined as: migration index ⫽
no. of test migrated cells/no. of control migrated cells. Assays
were repeated in triplicate.
In vivo experiments
BalbcByJ Hfh11 nu/nu (Nude) andC3H/HeJinbredmicestrains
used for all experiments were maintained in house. C-terminal
(His)
6
KS1 (20 µg) was injected into the left footpads of either
Nude or C3H/HeJ mice in triplicate. The right foot of each animal
was injected with an equal volume of 20 mM Tris–HCl, pH 7.5.

684 B cell- and monocyte-activating chemokine (BMAC)
Fig. 8. Chemotactic activities of recombinant KS1. Chemotaxis assays were performed using a 48-well modified Boyden chamber with varying
concentrations of KS1 (j)orSDF1α (u). Five randomly selected regions per well were chosen and the number of cells counted under high-
power field microscopy. The migration index was calculated by the following formula: migration index ⫽ no. of test migrated cells/no. of control
migrated cells. (A) Chemotactic activity of murine spleen cells. (B) Chemotactic activity of the B cell line CESS, left y-axis is migration to KS1
whereas right y-axis is migration to SDF1α. (C) Chemotactic activity of the monocyte leukemia cell line THP-1. (D) Chemotactic activity of PHA
IL-2-activated T cells and (E) chemotactic activity of Con A IL-2-activated murine T cells. Values are the mean migration index ⫾ SEM obtained
for triplicate wells and are representative of two individual experiments.
Mice were sacrificed after 18 h, and feet dissected and fixed
in 3.7% formol saline. All tissues were sectioned and stained
with haemotoxylin & eosin. Histology was performed at Agro-
Quality (Auckland, NZ). Photomicrography was performed on
a Leica compound microscope and images prepared using
Adobe Photoshop.

B cell- and monocyte-activating chemokine (BMAC) 685
Fig. 9. Inflammatory response of Nude and C3H/HeJ mice upon s.c. injection of KS1. Each group of three mice was injected with (His)
6
KS1
per group. Mice were injected in the left footpad with (His)
6
KS1. Equal volumes of Tris buffer were injected into the right footpads as controls.
Feet were biopsied after 18 h and haemotoxylin & eosin sections prepared. (A) Nude mice demonstrate a mixed inflammatory response upon
injection with KS1, as indicated by an arrow. (B) No inflammation is apparent in the footpad injected with Tris buffer control. (C and D) High-
power magnification (⫻100 objective) of cells in Nude mouse inflammation indicates the presence of monocytes, mononuclear cells and
polymorphonuclear cells. (E) A mixed inflammatory response was also present in C3H/Hej mice as denoted by an arrow. (F) No inflammation
was detectable in footpads injected with the negative control. Abbreviations: mu, muscle; v, vein; e, epidermis; d, dermis; mφ, monocyte; mn,
mononuclear; and pmn, polymorphonuclear cells. Scale bar ⫽ 50 µm.
Results
Identification of KS1 cDNA sequence
A directionally cloned cDNA library was constructed from
immature murine keratinocytes and submitted for high-
throughput sequencing. Sequence data from a clone desig-
nated KS1 showed 35% identity over 74 amino acids with rat
macrophage inflammatory protein (MIP)-2B and 32% identity
over 72 amino acids with its murine homologue. The insert of
1633 bp (Fig. 1A) contained an open reading frame of 300
bp with a 5⬘ untranslated region of 202 bp and a 3⬘ untrans-
lated region of 1161 bp (this sequence is available from
GenBank under accession no. AF144754). A poly-adenylation
signal of AATAAA is present 19 bp upstream of the poly(A)
tail. The predicted mature polypeptide is 77 amino acids in
length containing four conserved cysteines with no ELR motif.

686 B cell- and monocyte-activating chemokine (BMAC)
The putative signal peptide cleavage site between Gly22 and
Ser23 was predicted by the hydrophobicity profile. The full-
length sequence was then screened against the EMBL data-
base using the BLAST program, and showed 92.6, 94 and
93.6% identity at the nucleotide level with human EST clones
AA643952, AA865643 and HS1301003 respectively. A
recently described human α-chemokine, BRAK, has 94%
identity with KS1 at the protein level (20). The alignment of
KS1, BRAK and other murine α-chemokines is shown in
Fig. 1B. The phylogenetic relationship between KS1 and other
α-chemokine family members was determined using the
Phylip package (Fig. 1C). KS1 and BRAK demonstrate a high
degree of divergence from the other α-chemokines supporting
the relatively low homology shown in the multiple alignment.
Tissue expression of KS1 and BRAK
Tissue distribution of KS1 by Northern hybridization showed
high expression in brain, ovary, lung and muscle, with low
levels of expression in bone marrow. The transcript size on
the Northern blot of 1.6 kb was the same size as the full-
length cDNA sequence. BRAK was highly expressed in small
intestine, colon and kidney, with moderate to low levels in
liver, spleen, thymus, placenta, brain and pancreas. BRAK
mRNA could also be detected in skeletal muscle and heart.
Expression could not be detected in ovary, testis or prostate.
The transcript size of BRAK was ~1.8 kb, which is similar to
KS1 (Fig. 2B). As non-ELR α-chemokines have been implic-
ated as having angiostatic properties, BRAK expression levels
were tested in a variety of tumours and compared to normal
tissue. BRAK was expressed in normal uterine and cervical
tissue, whereas it was completely down-regulated in their
respective tumours, uterine leiomyoma and cervical aden-
ocarcinoma (Fig. 2C). Conversely, BRAK was expressed in
breast tissue but was up-regulated in breast invasive ductal
carcinoma (Fig. 2C).
Recombinant expression of KS1
Recombinant C-terminal (His)
6
KS1 was a homogenous protein
with an apparent molecular mass of 15 kDa (Fig. 3). Internal
sequencing of the 15 kDa protein gave the peptide sequence
WYNAWNEK, confirming that the observed sequence is ident-
ical to that predicted from the cDNA sequence. The isoelectric
point was predicted to be 10.26 using DNASIS (Hitachi
Software Engineering, Yokohama, Japan). Recombinant
KS1Fc, expressed and purified using Protein A-affinity column
chromatography, revealed a protein with a molecular mass
of 43 kDa corresponding to the predicted size plus the Fc
fusion tag (data not shown).
Flow cytometric analysis of KS1 binding
Fc tagged KS1 (KS1Fc) was used to determine the cell types
which express the receptor for this chemokine. KS1Fc bound
to 54% of splenocytes and 9% peripheral lymph node cells
(Fig. 4A and B). No positive binding was identified in thymo-
cytes (Fig. 4C). Dual labelling experiments with antibodies to
cell surface antigens showed that KS1Fc bound B cells in
spleen (Fig. 4D) but not CD4 or CD8 T cells (Fig. 4E and F).
KS1Fc also bound to the B cells in peripheral lymph node
cells but not the T cells (data not shown). The matched
isotype control for CD19, CD8 and CD4, rIgG2a–FITC, showed
no positive labelling (data not shown).
Additionally, we screened peritoneal exudate cells (PEC)
to determine whether KS1Fc bound monocytes. Forward and
side scatter histograms from PEC were used to identify the
monocyte population in region H (Fig. 5A). Cells in region H
were CD19
–
(Fig. 5B), but were MHC class II
⫹
(Fig. 5C)
indicating that they were monocytes and not B cells. Dual
labelling experiments showed that all the cells in region H
were double positive for MHC class II and KS1Fc (Fig. 5D).
The control protein, EGBFc, showed no binding to the MHC
class II
⫹
cells from region H (Fig. 5E). The matched isotype
control for MHC class II, mIgG2a–biotin, showed no positive
labelling (Fig. 5F).
As many non-ELR chemokines stimulate activated T cells
we analysed KS1Fc binding to Con A-activated splenocytes
grown in the presence of 10 ng/ml IL-2 for 9 days. All cells
were positive for the activation marker CD69, and consisted
of 63% CD4 cells and 37% CD8 cells (data not shown).
KS1Fc showed no positive binding to these cells (Fig. 6D).
KS1Fc also bound to the murine B cell line, A20 (Fig. 6A),
and the human B lymphoblastoid cell line, CESS (Fig. 6B).
Additionally, KS1Fc bound to the monocyte leukemia cell line
THP-1 (Fig. 6C) but not Jurkat T cells (Fig. 6E). Preliminary
analysis identifies B cells and monocytes as responsive cells
for KS1. To demonstrate specificity (His)
6
KS1 was used in
cold competition with KS1Fc against murine splenocytes.
Increasing concentrations of (His)
6
KS1 reduced the level of
binding of KS1Fc (Fig. 7), as demonstrated by a decrease in
the mean channel fluorescence, to murine splenocytes. An
equivalent concentration of a non-specific (His)
6
-tagged pro-
tein, GV14B, showed no decrease in mean channel fluores-
cence when co-incubated with KS1Fc (Fig. 7). The ability of
(His)
6
KS1 to competitively inhibit the binding of KS1Fc vali-
dates the hypothesis that this reagent bound via the KS1
receptor.
(His)
6
KS1 induces chemotaxis in B cells and monocytes
Flow cytometric analysis revealed that KS1 specifically bound
to B cells and monocytes. We determined whether KS1 could
stimulate the chemotaxis of these cells using a modified
Boyden chamber. (His)
6
KS1 induced a concentration-
dependant migration in murine splenocytes, with optimal
activity at 100 ng/ml (Fig. 8A). In addition, (His)
6
KS1 stimulated
the migration of the B lymphoblastoid cell line, CESS, and
the monocyte line, THP-1 (Fig. 8B and C). However, unlike
SDF-1α, KS1 did not stimulate the migration of either human
or murine activated T cells (Fig. 8D and E).
(His)
6
KS1 induces inflammation in vivo
To determine whether KS1 was active in vivo and whether
T cells are required for an inflammatory response we injected
Nude mice s.c. with (His)
6
KS1. Histological examination of
mouse footpads injected s.c. with (His)
6
KS1 showed a leuko-
cyte infiltrate (Fig. 9A). The inflammation was of a mixed
phenotype with evidence of mononuclear cells and polymor-
phonuclear cells (Fig. 9C and D). No obvious inflammation
was apparent in the feet of mice injected with Tris, the buffer
excipient (Fig. 9B). To confirm that this inflammation was due
to (His)
6
KS1 and not endotoxin we repeated the experiment

B cell- and monocyte-activating chemokine (BMAC) 687
in LPS-insensitive C3H/HeJ mice. (His)
6
KS1-injected footpads
from these mice showed a similar inflammatory response to
the Nude mice (Fig. 9E) with the buffer excipient-injected
footpads having no marked inflammation (Fig. 9F).
Discussion
We have identified some of the biological activities of a
novel non-ELR α-chemokine, KS1, and described its tissue
distribution. The cDNA was similar to a recently cloned human
gene called BRAK (20). Homology of KS1 to BRAK was 94%
at the protein level, indicating it as the murine homologue of
this gene. To date no known function has been described for
BRAK. KS1 and BRAK appear to be distant relatives of the
non-ELR α-chemokines as was shown by their phylogenetic
relationship. There are five additional residues (KS1, SMSRY
and BRAK, SVSRY), between cysteines 3 and 4 of the
conserved cysteines, which is not consistent with the pre-
dicted Prosite motif for α-chemokines. The predicted N-
terminus of the mature protein upstream of the first cysteine
has only two residues in contrast to other non-ELR α-chemok-
ines. Furthermore, KS1 has a lysine in place of an arginine
immediately prior to the first cysteine. The conservation of a
highly basic residue, typically arginine, prior to the first
conserved cysteine has been postulated as a requirement
for binding to the receptor (1). The amino acid substitutions
between KS1 and BRAK in the mature peptide are conservat-
ive, indicating that these differences are likely to be insigni-
ficant.
KS1 tissue distribution in mouse and BRAK in human is
unusual for α-chemokines in that it is highly expressed in
normal non-lymphoid tissues. Although expression levels are
different between mouse and human, KS1 and BRAK are
expressed in brain and muscle. Differences between mouse
and human expression profiles have been described for other
chemokines (23–25), and are thought to reflect pathological
changes of the particular donor. In contrast to the reported
expression profile for BRAK (20), we found BRAK was
expressed at higher levels in small intestine, colon and kidney.
Additionally, the predominant band ran at 1.8 kb rather than
2.5 kb as reported earlier (20), raising the possibility of splice
variance. Alternative splicing has previously been described
as a property of some chemokines, for example LARC/
MIP3α (23).
As non-ELR α-chemokines have been shown to have
angiostatic function (15,16,26,27), we investigated BRAK
expression in normal versus tumour tissue. The blot revealed
that BRAK is expressed in non-malignant breast, uterine and
cervical tissues. With good expression in human breast and
kidney it is not surprising that BRAK was identified from breast
and kidney EST. From the tumour expression data we saw
two patterns emerge, either BRAK was up-regulated in breast
invasive ductal carcinoma or, in the case of uterine leiomyoma
and cervical adenocarcinoma, BRAK mRNA was undetect-
able. The reason for the disparate trends in expression levels
from the different tumours is unclear at this stage and may
be related to differences in cancer pathology. Furthermore,
these biopsies were likely derived from a single patient and
may not reflect the majority of cases. Nevertheless, it would
be of interest to determine whether BRAK added directly to
tumour models of these cancers could alter malignancy.
KS1 had a predicted size of 9.4 kDa; however, the purified
(His)
6
KS1 protein had an apparent size of 15 kDa which could
not be accounted for by the additional histidine residues. This
discrepancy between the predicted and apparent size has
previously been reported for a number of chemokines (28,29),
and is thought to be due to the highly basic nature of
these proteins.
The high degree of homology between KS1 and BRAK
suggested that they would be active on both mouse and
human cell types. This was demonstrated by flow cytometric
analysis which showed that KS1 binds directly to mouse and
human B cells and monocytes. Furthermore, KS1 induced
chemotaxis on both mouse and human cells. This phenom-
enon of rodent chemokines stimulating human cells has been
previously described for a number of different chemokines
(30–32). We clearly defined B cells and monocytes, and not
T cells, as target cells for KS1 by binding studies. The only
other non-ELR α-chemokine to bind to these cells is SDF-1α;
however, SDF-1α also binds to T cells (17,18,33). We were
able to confirm that KS1 stimulates B cells and monocytes
but not T cells in migration assays using a range of different
cell types. In the case of splenocytes and THP-1 cells, SDF-
1α and KS1 stimulated equivalent levels of migration; however,
CESS cells were 15-fold more responsive to SDF1α than KS1.
As the in vitro data indicated that B cells and monocytes
respond to KS1, we tested its inflammatory properties by
injecting Nude mice with the protein. Mice injected s.c. with
(His)
6
KS1 showed a mixed inflammatory response. As Nude
mice have no T cells this supported the in vitro data that
(His)
6
KS1 promotes extravasation of cells other than T cells. As
seen with the Nude mice, C3H/HeJ also had an inflammatory
response to (His)
6
KS1, demonstrating that the response in the
Nude mice was (His)
6
KS1 specific and not due to endotoxin.
Although we have demonstrated the ability of KS1 to stimulate
chemotaxis of B cells and monocytes, we do not rule out the
possibility that other haemopoetic or non-haemopoetic cells
might respond to KS1.
The majority of non-ELR α-chemokines have been shown
to be chemotactic for activated T cells; however, KS1 did not
cause the migration of these cells. Therefore, this raises the
question of which receptor does KS1 utilize? There are
currently only three known chemokine receptors that bind
non-ELR α-chemokines: CXCR3, the receptor for I-TAC (12),
Mig and IP-10 (34); CXCR4, the receptor for SDF-1α (17,18);
and CXCR5, the receptor for BCA-1 (13). As we have shown
that KS1 does not stimulate T cells it is unlikely that it is
binding via CXCR4. Furthermore, it is unlikely to bind
via CXCR3, a receptor on activated T cells, as we can
demonstrate no activity on Con A IL-2-activated T cells. This
then leaves CXCR5; however, this receptor has only been
demonstrated on B cells and not monocytes. Therefore, the
likelihood of KS1 acting via a novel receptor merits further
investigation.
The biological function of a novel chemokine, initially identi-
fied as KS1, is described. KS1 has a broad expression in
non-lymphoid tissue, altered expression levels in tumours and
a role in trafficking of B cells and monocytes. Therefore, we
propose the name B cell- and monocyte-activating chemokine

688 B cell- and monocyte-activating chemokine (BMAC)
(BMAC) for this molecule to reflect its described biological
functions
Acknowledgements
We are grateful to Dr Paul Tan for his helpful comments with the
manuscript, Dr Matthew Glenn and Dr Ilkka Havukkala for mainten-
ance of the EST database, and Stewart Whiting for managing the
animal facility. We are also grateful to Dr Annette McGrath for
assistance with the bioinformatic analysis. We also appreciate the
support of the Functional Genomics group at Genesis Research
and Development Corp. Ltd, Auckland, NZ and Immunex Corp.,
Seattle, WA.
Abbreviations
BMAC B cell- and monocyte-activating chemokine
Con A concanavalin A
ELR Glu–Leu–Arg
MIP macrophage inflammatory protein
PBMC peripheral blood mononuclear cells
PE phycoerythrin
PEC peritoneal exudate cells
PHA phytohaemagglutinin
References
1 Baggiolini, M., Dewald, B. and Moser, B. 1997. Human
chemokines: an update. Annu. Rev. Immunol. 15:675.
2 Rollins, B. J. 1997. Chemokines. Blood 90:909.
3 Gale, L. M. and McColl, S. 1999. Chemokines: extracellular
messengers for all occasions? BioEssays 21:17.
4 Walz, A., Peveri, P., Aschauer, H. and Baggiolini, M. 1987.
Purification and amino acid sequencing of NAF, a novel neutrophil
activating factor produced by monocytes. Biochem. Biophys.
Res. Commun. 149:755.
5 Yoshimura, T., Matsushima, K., Tanaka, S., Robinson, E. A.,
Appella, E., Oppenheim, J. J. and Leonard, E. J. 1987. Purification
of a human monocyte derived neutrophil chemotactic factor that
has peptide sequence similarity to other host defense cytokines.
Proc. Natl Acad. Sci. USA 84:9233.
6 Schroder, J.-M., Mrowietz, U., Morita, E. and Christophers, E.
1987. Purification and partial biochemical characterization of a
human monocyte derived, neutrophil-activating peptide that lacks
interleukin 1 activity. J. Immunol. 139:3474.
7 Deuel, T. F., Senior, R. M., Chang, D., Griffin, G. L., Henrikson, R.
L. and Kaiser, E. T. 1981. Platelet factor 4 is chemotactic for
neutrophils and monocytes. Proc. Natl Acad. Sci. USA 78:4584.
8 Luster, A. D., Unkeless, J. C. and Ravetch, J. V. 1985. Gamma-
interferon transcriptionally regulates an early-response gene
containing homology to platelet proteins Nature 315:672.
9 Luster, A. D. and Ravetch, J. V. 1987. Biochemical characterization
of a γ-interferon inducible cytokine (IP-10). J. Exp. Med. 166:1098.
10 Farber, J. M. 1993. HuMIG: a new human member of the
chemokine family of cytokines. Biochem. Biophys. Res.
Commun. 192:223.
11 Nagasawa, T., Kikutani, H. and Kishimoto, T. 1994. Molecular
cloning and structure of a pre-B-cell growth-stimulating factor.
Proc. Natl Acad. Sci. USA 91:2305.
12 Cole, K. E., Strick, C. A., Paradis, T. J., Ogborne, K. T., Loetscher,
M., Gladue, R. P., Lin, W., Boyd, J. G., Moser, B., Wood, D. E.,
Sahagan, B. G. and Neote, K. S. 1998 Interferon-inducible T cell
alpha chemoattractant (I-TAC): a novel non-ELR CXC chemokine
with potent activity on activated T cell through selective high
affinity binding to CXCR3. J. Exp. Med. 187:2009.
13 Legler, D. F., Loestcher, M., Roos, R. S., Clark-Lewis, I., Baggiolini,
M. and Moser, B. 1998. B cell-attracting chemokine 1, a human
CXC chemokine expressed in lymphoid tissues, selectively
attracts B lymphocytes via BLR1/CXCR5. J. Exp. Med. 187:655.
14 Maione, T. E., Gray, G. S., Petro, J., Hunt, A. J., Donner, A. L.,
Bauer, S. I., Carson, H. F. and Sharpe, R. J. 1990. Inhibition of
angiogenesis by recombinant human platelet factor-4 and related
peptides. Science 247:77.
15 Angiolillo, A. L., Sgadari, C., Taub, D. D., Liao, F., Farber, J. M.,
Maheshwari, S., Kleinman, H. K., Reaman, G. H. and Tosato, G.
1995. Human interferon-inducible protein 10 is a potent inhibitor
of angiogenesis in vivo. J. Exp. Med. 182:155.
16 Sgadari, C., Farber, J. M., Angiolillo, A. L., Liao, F., Teruya-
Feldstein, J., Burd, P. R., Yao, L., Gupta, G., Kanegane, C. and
Tosato, G. 1997. Mig, the monokine induced by interferon γ,
promotes tumour necrosis in vivo. Blood 89:2635.
17 Bleul, C. C., Farzan, M., Choe, H., Parolin, C., Clark-Lewis, I.,
Sodroski, J. and Springer, T. A. 1996. The lymphocyte
chemoattractant SDF-1 is a ligand for LESTR/fusin and blocks
HIV-1 entry. Nature 382:829.
18 Oberlin, E., Amara, A., Bachelerie, F., Bessia, C., Virelizier, J. L.,
Arenzana-Seisdedos, F., Schwartz, O., Heard, J. M., Clark-Lewis,
I., Legler, D. F., Loetscher, M., Baggiolini, M. and Moser, B.
1996. The CXC chemokine SDF-1 is the ligand for LESTR/fusin
and prevents infection by T-cell-line-adapted HIV-1. Nature
382:833.
19 Zagury, D., Lachgar, A., Chams, V., Fall, L. S., Bernard, J., Zagury,
J. F., Bizzini, B., Gringeri, A., Santagostino, E., Rappaport, J.,
Feldman, M., O’Brien, S. J., Burny, A. and Gallo, R. C. 1998. C–
C chemokines, pivotal in protection against HIV type 1 infection.
Proc. Natl Acad. Sci. USA 95:3857.
20 Hromas, R., Broxmeyer, H. E., Kim, C., Nakshtri, H.,
Christopherson, K., II, Azam, M. and Hou, Y.-H. 1999. Cloning of
BRAK, a novel divergent CXC chemokine preferentially expressed
in normal versus malignant cells. Biochem. Biophys. Res.
Commun. 255:703.
21 Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D.
J. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403.
22 Sambrook, J., Fritsch, E. F. and Maniatis, T. 1989. Molecular
Cloning: A Laboratory Manual, 2nd edn, p. 7.39. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, NY.
23 Tanaka, Y., Imai, T., Baba, M., Ishikawa, I., Uehira, M., Nomiyama,
H. and Yoshie, O. 1999. Selective expression of liver and
activation-regulated chemokine (LARC) in intestinal epithelium in
mice and humans. Eur. J. Immunol. 29:633
24 Hromas, R., Gray, P. W., chantry, D., godiska, R., Krathwohl, M.,
Fife, K., Bell, G. I., takeda, J., Aronica, S., Gordon, M., Cooper,
S., Broxmeyer, H. E. and Klemsz, M. J. 1997. Cloning and
characterization of exodus, a novel beta-chemokine. Blood
89:3315.
25 Hieshima, K., Imai, T., Opdenakker, G., van Damme, J., Kusuda,
J., tei, H., Sakaki, Y., Takatsuki, K., Miura, R., Yoshie, O. and
Nomiyama, H. 1997. Molecular cloning of a novel CC chemokine
liver and activation-regulated chemokine (LARC) expressed in
liver. Chemotactic activity for lymphocytes and gene localization
on chromosome 2. J. Biol. Chem. 272:5846.
26 Tannenbaum, C. S., Tubbs, R., Armstrong, D., Finke, J. H.,
Bokowski, R. M. and Hamilton, T. A. 1998. The CXC chemokine
IP-10 and Mig are necessary for IL-12 mediated regression of
the mouse RENCA tumour. J. Immunol. 161:927.
27 Kanegane, C., Sgadari, C., Kanegane, H., Teruya-Feldstein, J.,
Yao, L., Gupta, G., Farber, J. M., Liao, F., Liu, L. and Tosato, G.
1998. Contribution of the CXC chemokines IP-10 and Mig to the
antitumor effects of IL-12. J. Leuk. Biol. 64:384.
28 Richmond, A., Balentien, E., Thomas, H. G., Flaggs, G., Barton,
D. E., Spiess, J., Bordoni, R., Francke, U. and Derynck, R.
1988. Molecular characterization and chromosomal mapping of
melanoma growth stimulatory activity, a growth factor structurally
related to β-thromboglobulin. EMBO J. 7:2025.
29 Liao, F., Rabin, R. L., Yannelli, J. R., Koniaris, L. G., Vanguri, P.
and Farber, J. M. 1995 Human Mig: biochemical and functional
characterization. J. Exp. Med. 182:1301.
30 Wuyt, A., Haelens, A., Proost, P., Lenaerts, J. P., Conings, R.,
Opdenakker, G. and van Damme, J. 1996. Identification of mouse
granulocyte chemotactic protein-2 from fibroblasts and epithelial
cells. Functional comparison with natural KC and macrophage
inflammatory protein-2. J. Immunol. 157:1736.
31 Abdullah, F., Ovadia, P., Feuerstein, G., Neville, L. F., Morrison,

B cell- and monocyte-activating chemokine (BMAC) 689
R., Mathiak, G., Whiteford, M. and Rabinovici, R. 1997. The
novel chemokine mob-1: involvement in adult respiratory distress
syndrome. Surgery 122:303.
32 Patel, V. P., Kreider, B. L., Li, Y., Li, H., Leung, K., Salcedo, T.,
Nardelli, B., Pippalla, V., Gentz, S., Thotakura, R., Parmelee,
D., Gentz, R. and Garotta, G. 1997. Molecular and functional
characterization of two novel human C–C chemokines as inhibitors
of two distinct classes of myeloid progenitors. J. Exp. Med.
185:1163.
33 Bleul, C. C., Fuhlbrigge, R. C., Casasnovas, J. M., Aiuti, A. and
Springer, T. A. 1996. A highly efficacious lymphocyte
chemoattractant, stromal cell-derived factor 1 (SDF-1). J. Exp.
Med. 184:1101.
34 Weng, Y., Siciliano, S. J., Waldburger, K. E., Sirotina-Meisher, A.,
Staruch, M. J., Daugherty, B. L., Gould, S. L., Springer, M. S. and
DeMartino, J. A. 1998. Binding and functional properties of
recombinant and endogenous CXCR3 chemokine receptors.
J. Biol. Chem. 273:18288.