Cytokine RANTES, released by thrombin-stimulated platelet is a potent attractant for human eosinophils

Abstract

Thrombin stimulation of human platelets results in the release of a preformed proteinaceous human eosinophil (Eo)-chemotactic activity. By the use of different high-performance liquid chromatography techniques, two Eo-chemotactic polypeptides (EoCPs), tentatively termed EoCP-1 and EoCP-2, were purified to homogeneity. Upon SDS-PAGE analysis, these chemotaxins showed molecular masses near 8 kD. NH2-terminal amino acid sequence analysis revealed identical sequences for both EoCP-1 and EoCP-2, which are also identical to that of RANTES, a cytokine that structurally belongs to the interleukin 8 superfamily of leukocyte selective attractants, and that is known to be a "memory-type" T lymphocyte-selective attractant. In the major Eo chemotaxin, EoCP-1, the residues 4 and 5, which in EoCP-2 were found to be serine residues, could not be identified. Electrospray mass spectrometry (ESP-MS) of EoCPs revealed for EoCP-2 a molecular mass of 7,862.8 +/- 1.1 daltons, which is 15.8 mass units higher than the calculated value of RANTES, indicating that EoCP-2 is identical to the full-length cytokine, and oxygenation, probably at methionine residue number 64, has taken place. Upon ESP-MS, EoCP-1 showed an average molecular mass of 8,355 +/- 10 daltons, suggesting O-glycosylation at these serine residues. Both natural forms of RANTES showed strong Eo-chemotactic activity (ED50 = 2 nM) with optimal chemotactic migration at concentrations near 10 nM, however, there were no significant migratory responses with human neutrophils. Chemotactic activity of RANTES for human Eos could be confirmed using recombinant material, which has been found to be as active as the natural forms. Since RANTES gene expression has been detected in activated T lymphocytes, and recombinant RANTES was shown to be a "memory" T lymphocyte-selective attractant, it is now tempting to speculate about an important role of RANTES in clinical situations such as allergene-induced late-phase skin reactions in atopic subjects or asthma, where in affected tissues both memory T cells and Eos are characteristic.

Full text

Brief De~nltive Report
Cytokine RANTES Released by Thrombin-stimulated
Platelets Is a Potent Attractant for Human E inophils
By Yoshikazu Kameyoshi,* Albrecht D6rschner,~
Anthony I. MaUet,S Enno Christophers,* and Jens-M. Schr6der*
From the *Department of Dermatology, University of Kiel, D-W-2300 Kiel, Germany;
*Beiersdorf AG, Hamburg~ Germany; and the $Institute of Dermatology, St. Thomas' Hospital,
London, United Kingdom
Summary
Thrombin stimulation of human platelets results in the release of a preformed proteinaceous
human eosinophil (Eo)-chemotactic activity. By the use of different high-performance liquid
chromatography techniques, two Eo-chemotactic polypeptides (EoCPs), tentatively termed EoCP-1
and EoCP-2, were purified to homogeneity. Upon SDS-PAGE analysis, these chemotaxins showed
molecular masses near 8 kD. NH2-terminal amino acid sequence analysis revealed identical
sequences for both EoCP-1 and EoCP-2, which are also identical to that of R.ANTES, a cytokine
that structurally belongs to the interleukin 8 superfamily of leukocyte selective attractants, and
that is known to be a
"memory-type"
T lymphocyte-selective attractant. In the major Eo
chemotaxin, EoCP-1, the residues 4 and 5, which in EoCP-2 were found to be serine residues,
could not be identified. Electrospray mass spectrometry (ESP-MS) of EoCPs revealed for EoCP-2
a molecular mass of 7,862.8 _+ 1.1 daltons, which is 15.8 mass units higher than the calculated
value of RANTES, indicating that EoCP-2 is identical to the full-length cytokine, and oxygenation,
probably at methionine residue number 64, has taken place. Upon ESP-MS, EoCP-1 showed
an average molecular mass of 8,355 _+ 10 daltons, suggesting O-glycosylation at these serine
residues. Both natural forms of RANTES showed strong Eo-chemotactic activity (EDs0 -- 2
nM) with optimal chemotactic migration at concentrations near 10 nM, however, there were
no significant migratory responses with human neutrophils. Chemotactic activity of RANTES
for human Eos could be confirmed using recombinant material, which has been found to be
as active as the natural forms. Since R.ANTES gene expression has been detected in activated
T lymphocytes, and recombinant RANTES was shown to be a
"memory"
T lymphocyte-selective
attractant, it is now tempting to speculate about an important role of R.ANTES in clinical situations
such as allergene-induced late-phase skin reactions in atopic subjects or asthma, where in affected
tissues both memory T cells and Eos are characteristic.
I
nflammatory diseases are characterized histologically by im-
migration of different leukocyte subtypes, i.e., neutrophils,
monocytes/macrophages, T lymphocyte subsets, or eosinophils
(Eos). Tissue-oriented migration of certain inflammatory ceU
types assumes, apart from expression of adherence proteins
and other in vivo migration-fadlitating factors, the existence
of locally produced cell-selective chemotactic factors. Well-
described chemotaxins such as C5a or leukotriene B4 are
chemotactic for a wide variety of leukocytes and do not show
this specificity. Apart from these pan-leukotactic factors, re-
cently a number of leukocyte-selective chemotaxins have been
discovered.
ID8 represents one of these selective leukocyte attractants,
which is chemotactic for neutrophils (1) and T lymphocytes
(2), but not for monocytes and Eos (3, 4). IL-8 comprises
a polypeptide mediator that is a member of a superfamily
of structurally related low molecular weight cytokines, which
contain four cysteines at identical relative positions with a
conserved -Cys-X-Cys-(C-X-C) or -Cys-Cys- (C-C-) motif.
Members of the C-X-C-subfamily, such as IL-8 (3-5), mela-
noma growth-stimulatory activity (MGSA/gro oe) (6, 7), and
neutrophil-activating protein 2 (NAP-2) (8), are potent
chemotactic factors for neutrophils and in part are also known
to be chemotactic for lymphocytes, however, not for mono-
cytes or Eos.
Members of the C-C-subfamily, monocyte chemotactic pro-
tein I(MCP-1) (9)/monocyte-chemotactic and -activating factor
(MCAF) (10), and the cytokine RANTES (11) represent
chemotactic factors for monocytes, but not for neutrophils.
Other members of this subfamily, the human equivalents of
587 J. Exp. Med. 9 The Rockefeller University Press 9 0022-1007/92/08/0587/06 $2.00
Volume 176 August 1992 587-592

macrophage inflammatory protein lc~ and 13, are chemotactic
for T lymphocyte subsets, cytotoxic T lymphocytes and naive
T lymphocytes, respectively, whereas RANTES shows an ap-
parent selectivity for memory T lymphocytes (12).
So far members of the II-8 superfamily have not been
reported to be Eo-chemotactic factors, and hence we address
the question whether the Ib8 supergene family of low mo-
lecular mass (6-10 kD) cytokines also contains Eo-selective
attractants. We report here that platelets stimulated with
thrombin represent a source of Eo-chemotactic proteins, which
structurally belong to the same superfamily of host defense
cytokines as a number of other cell-selective chemoattractant
cytokines do.
Materials and Methods
Recombinant Cytoleines.
Recombinant cytokines RANTES and
IL-8 were purchased from Pepro Tech Inc. (Rocky Hill, NJ). Purity
was >98% as shown by a single line upon SDS-PAGE analysis.
Isolation of Eosinophils and Neutrophils.
Blood was taken from
healthy donors or subjects with a mild eosinophilia (5-10% of pe-
ripheral blood leukocytes). Eos and neutrophils were isolated from
acidic citrate dextran-treated blood with the use of discontinuous
Percoll density gradient centrifugation as previously described (13).
Yielded cell preparations were >85% pure for Eos and >95% for
neutrophils.
Production of Platelet-den'ved EoCPs.
Platelet-rich phsma obtained
from citrate/dextran blood was centrifuged at 2,000 g for 30 rain.
Phtelet pellets were washed twice with PBS containing 10 mM
EDTA and resuspended in PBS. Platdet suspensions were incubated
at 37~ for 30 rain in the presence of 2 U/m1 thrombin (Sigma
Chemical Co., Munich, FRG). After centrifugation at 4~ cell-
free supernatants were collected and stored below -70~ until
further use.
Purification of EoCPs.
EoCPs were purified by HPLC methods
similar to those used for purification of NAP-1/IL-8 (3), MGSA/gro
(6), and platelet-derived neutrophil attractant (NAP-4) (14). Pooled
superuatants of thrombin-stimulated phtelets acidified with tri-
fluoroacetic acid (TFA) to pH 3 were concentrated over filters (YM5;
Amicon Corp., Danvers, MA) and applied to a preparative wide-
pore reversed-phase (RP)-8 HPLC column (300 x 7 #m C8
Nucleosfl, 250 x 12.5 mm; Macherey-Nagel, Dfiren, FRG). Pro-
teins were eluted using a gradient of acetonitrih containing 0.1%
TFA. Fractions active in an Eo chemotaxis system off RP-8 HPLC
were pooled, concentrated by lyophilization, and applied to a TSK-
2000 size exclusion HPLC column (600 x 8 mm; LKB, Bromma,
Sweden) previously equilibrated with 0.1% TFA. Proteins were
duted with the same solvent. Eo-chemotactic fractions off TSK-
2000 HPLC were pooled and applied to a wide-pore CN-propyl
RP HPLC column (5/zm, 250 x 4 mm; J. T. Baker, Gross Gerau,
FRG) previously equilibrated with 0.1% aqueous TFA. Protein elu-
tion was performed with a gradient of n-propanol containing 0.1%
TFA. Thereafter, fractions off CN-propyI-RP-HPLC, which were
active in the Eo chemotaxis assay system, finally were applied to
a narrow-pore RP-18-HPLC column (100 x 7/zm, C18 Nucleosil,
250 x 10 mm; Macherey-Nagel), previously equilibrated with 0.1%
aqueous TFA containing 10% (vol/vol) acetonitrile, and polypep-
tides were duted with a gradient of acetonitrih containing 0.1%
TFA.
SDS-PAGE.
SDS-PAGE analysis was performed with the Phast
system (Pharmacia, Freiburg, FRG) using high-density gels ac-
cording to the manufacturer's instructions. CNBr-cleavage prod-
ucts of myoglobin (Sigma Chemical Co.) as well as recombinant
human Ser72-Ib8 served as molecular mass standards.
Chemotaxis Assays.
Eo chemotactic activity was measured in
blind-well Boyden chambers as previously described (13). In some
experiments Eos were determined microscopically using a modi-
fication of Boyden's method as previously described in detail (3,
13). Chemotactic activity is expressed as chemotactic index, calcu-
lated as: stimulated migration/random migration.
Amino Acid Sequence Analysis.
Underivatized samples were ana-
lyzed using a gas phase sequencer (4701; Applied Biosystems, Inc.,
Foster City, CA) with on-line HPI.C analysis of the phenylthio-
hydantoin derivatives.
Electrospray Mass Spectrometry (ESP-MS).
For ESP-MS analysis,
a Trio-2 quadrupole (VG Biotech-Fisons Instruments
LTD,
Man-
chester, UK) was used. The peptides were introduced into the ESP-
MS ion source as a solution in 50:50:1 methanol/water/acetic acid
(vol/vol/vol) at a flow rate of 5 #l/min. The electrospray needle
was held at 4 kV relative to the source. As an internal mass calibrant
ubiquitine was used.
Results and Discussion
To test the hypothesis that 8-1O-kD Eo-specific or -selective
attractants would exist, we originally used supernatants of
PBMC preparations that were stimulated with a mixture of
bacterial LPS and PHA, since these cell preparations are known
to produce a number of IL-8-1ike chemotactic cytokines such
as IL-8 itself (3-5, 15), MCP-1/MCAF (16), MGSA/gro c~
(6), and apparently also other chemotactic members of the
C-C-family (12).
In initial experiments we detected 6-10-kD Eo-chemotactic
activity in supernatants of stimulated PBMC (data not shown).
To determine its origin purified monocytes were stimulated
with LPS in one experimental series and lymphocyte prepa-
rations were treated with mitogens in another series. To our
surprise, in the monocyte supernatants, which contained high
amounts of neutrophil-chemotactic II:8, no Eo-chemotactic
activity could be detected. Instead Eo-chemotactic activity
was found in supernatants of PHA-stimulated monocyte-
depleted lymphocyte preparations (data not shown).
Because these lymphocyte preparations contained various
numbers of contaminating platelets, we tested whether these
cells are a possible source of Eo-chemotactic activity. Platelets
are known to produce members of the II.-8 supergene family
such as platelet basic protein and its truncation products, con-
nective tissue-activating peptide III (CTAP III) as well as
/3-thromboglobulin (17), which by enzymatic cleavage form
neutrophil-chemotactic NAP-2 (18). Moreover, platelet factor
4 (PF-4) (19), as well as a structurally related molecule we
tentatively termed NAP-4 (14), are stored in platelets. There-
fore, it seemed to be an attractive working hypothesis to as-
sume that platelets also are a source of other members of the
6-10-kD cytokine family.
When lysates of platelets were tested for Eo-chemotactic
activity in vitro, a high titer of the activity was observed (data
not shown). To investigate whether this Eo-chemotactic ac-
tivity is released by physiological stimulation, platelets were
incubated with thrombin (2 U/ml) for 30 min and superna-
tants were analyzed for Eo-chemotactic activity. As shown
588 Eosinophil Chemotactic RANTES

oPo
/\
/ \
o~O~ ~
10 .3 l0 '2 16' ld ~ "
PAF
I~eciprocal dilution
Figure 1. Eo-chemotactic activity in supernatants of thrombin-
stimulated platelets. Platelets were incubated at 37~ for 15 rain in the
presence of thrombin (2 U/rrd). Cell-free supematants were collected, diluted
in PBS/BSA, and assayed in the Boyden chamber system for Eo-chemotactic
activity. The Eo chemotaxin PAF (100 riM) served as control. A typical
experiment is shown.
in Fig. 1, a dose-dependent bell-shaped Eo-chemotactic re-
sponse was observed. Since platelets are known to produce
PAF, a potent Eo chemotaxin of low molecular weight (20),
the relative molecular mass of this Eo-cbemotactic activity
was determined. By TSK-2000 size exclusion HPLC, the
majority of Eo-chemotactic activity was detected in fractions
corresponding to the molecular mass range between 5 and
15 kD (data not shown). To purify the Eo-chemotactic poly-
peptide(s) (EoCPs) present in supernatants of thrombin-
stimulated platelets, we used similar HPLC methods as those
with which we purified NAP-1/IL-8 (3), MGSA/gro (6), and
platelet-derived neutrophil attractant 4 (NAP-4) (14).
At the first step, pooled platelet supernatants were chro-
matographed on a preparative RP-8 HPLC column followed
by a TSK-2000 size exclusion HPLC (Fig. 2 A). Eo-chemotactic
material was further purified by wide-pore cyanopropyl HPLC
and finally purified by narrow-pore RP-18 HPLC (Fig. 2 B).
The last purification step led to a broad peak of Eo-chemotactic
activity corresponding to two peaks (EoCP-1 and EoCP-2)
absorbing at 215 nm (Fig. 2 B). The presence of a single silver-
stained band upon SDS-PAGE analysis for each EoCP, which
showed a mobility somewhat higher than that of authentic
72-residue II.-8 (molecular weight of 8,532), indicated that
the material in these EoCP preparations are 8-kD polypep-
tides (Fig. 2 C).
NHvterminal amino acid sequencing revealed a single se-
quence of 16 or 32 residues for each EoCP:
EoCP-1, SPYXXDTTPXXFAYIA
EoCP-2, SPYSS DTTPXXFAYIARPLPRAXXXEYFYXXG
R.ANTES (21), SPYSS DTTPCCFAYIARPLPRAHIKEYFYTSG...
The only sequence obtained for both EoCPs in several in-
vestigations is identical to that reported for the cytokine
RANTES.
The molecular weight of EoCP-1 and EoCP-2 was deter-
mined by ESP-MS (Fig. 3), and obtained values were 8,355
_+ 10 and 7,862.8 _ 1.1, respectively. Although both EoCPs
showed the same amino acid sequence, EoCP-1 had a molec-
ular mass approximately 500 mass units higher than that of
Figure 2. Purification of EoCPs. Eo-chemotactic activity detected in
supernatants of thrombin-stimulated platelets was purified by a series of
HPLCs. The shaded area represents Eo-chemotactic activity in HPLC frac-
tions. Experimental conditions are detailed in Materials and Methods. (A)
TSK-2000 size exclusion I--IPLC of Eo-chemotactic fractions off RP-8-HPLC.
(B) Narrow pore R.P-18-HPLC of an EoCP preparation purified by CN-
RP-HPLC. A representative purification is shown. (C) SDS-PAGE of
purified EoCPs. SDS-PAGE analysis was performed with the Phast System
and proteins were silver stained. In lanes 1 and 7, CNBr cleavage products
of myoglobin were applied as Mr standards, whereas lanes 2 and 6 con-
mined 72-residue rlL-8, and lane 3 contained rKANTES. In lane 4, EoCP-2
was applied, and lane 5 contained EoCP-1.
EoCP-2. This difference could be attributed to O-glycosylation
presumably at serine residues 4 and 5, where sequence anal-
ysis of EoCP-1 failed. N-glycosylation sites are known to be
absent from RANTES (21). The measured molecular mass
of EoCP-2 agrees well with the calculated molecular mass
of R.ANTES. RANTES contains four cysteine amino acids,
and these are thought to exist in the reduced form, the pep-
tide being folded with two disulphide bridges, as has been
proven by 1H-NMR-analysis as well as x-ray crystallographic
analysis for other members of the KANTES/IL-8 family (22).
This reduces the calculated molecular weight from 7,851.06
to 7,847.03. This value differs from the measured value by
15.8 units, which can be accounted for by the assumption
of oxidation having taken place on the single methionine res-
idue number 67 in RANTES. Oxidation of methionine is
very common, especially in the case of samples that have been
exposed to the atmosphere for any length of time before
analysis.
589 Kameyoshi et al. Brief Definitive Report

o~
E
E
r
>
r~
100
75
50
1311.oE 1572.8 E
1123.8 E
~176
u952.41071 12u421
779.4
,h,
500 750 1000 1250 1500
mass/charge
1965IE
L
zx. hL ., ,,..• .
1750 2000
Figure 3. ESP-MSofEoCP-2. The calcuhted molecular mass for EoCP-2
(E), corrected for Ubiquitin (U) as an internal standard, is 7,862.8 + 1.1
chitons. The charge states shown are +4 (1,965.6), +5 (1,572.8), +6
(1,311.0), and +7 (1,123.8).
Both EoCP-1 and EoCP-2 showed similar dose-dependent
Eo-chemotactic activity in the Boyden chamber system (Fig.
4 a), indicating that derivatization of serine residues 4 and
5 does not affect Eo-chemotactic activity either in potency
(EDs0) or efficacy (percent input migrating Eos upon op-
timal stimulation doses). Moreover, Eo-chemotactic activity
of RANTES could be confirmed with recombinant mate-
rial: rRANTES showed Eo-chemotactic activity at similar
doses than found for both natural forms (Fig. 4 b). In con-
trast with this, neither EoCP induced significant migratory
response of human neutrophils in the Boyden chamber system
(data not shown).
RANTES originally was identified as an apparently T
cell-spedfic inducible gene, which was found to be expressed
by cultured T cell lines that were antigen spedfic and growth
factor dependent (21). Furthermore, RANTES mRNA ex-
pression has been found to be inducible in PBL by antigen
or mitogen stimulation. It is therefore likely, but yet not
proven, that supernatants of mitogen-stimulated PBL con-
tain Eo-chemotactic RANTES.
The Eo is one of the predominant cell types found in late-
phase reactions or at the inflammatory sites in allergic dis-
eases (23, 24). Release of cytotoxic cationic Eo granule pro-
teins as well as synthesis of peptido-leukotrienes are believed
to contribute to such hypersensitivity diseases. Apart from
other cytokines such as II.-5 (25), GM-CSF (25), and "lym-
phocyte chemotactic factor LCF" (26), RANTES has been
identified as an Eo-selective attractant in vitro and might par-
"o
u
:G
L)
"S
E
0.I
r
(.3
a
5
7
{3 4 /.
~.3 2
o*'
I
0'.1 i 10 1()0
EoCP (nM)
9 41 - c HI d l i lb 160
PAF PAF
rRANTES (nM)
Figure 4. Eo-chemotactic activity of EoCPs and rRANTES. (a) Both
RP-18-HPLC-purified fractions, EoCP-1 (Q) and EoCP-2 (O), were ana-
lyzed for Eo-chemotactic activity. Results are expressed as mean + SE
from six (EoCP-1) and five (EoCP-2) experiments. (b) Eo-ehemotactic ac-
tivity ofrRANTES. Results are expressed as mean _+ SE from five experi-
ments. PAF (100 nM) served as control.
ticipate in recruitment of Eos to T cell-mediated hypersensi-
tivity reactions in vivo. Since it is known that rRANTES
attracts T lymphocytes of the memory type (CD45RO +)
(11), RANTES may be a common mechanism in diverse im-
munological reactions, which culminate in the emigration
of T lymphocytes of this phenotype and Eos from the circu-
lation into sites of inflammation. The allergen-induced late-
phase skin reaction in atopic subjects, in which both memory
T cell infiltration and Eo accumulation are characteristic (24,
27), may represent a possible clinical example.
Our finding that platelets release upon stimulation Eo-
chemotactic RANTES serves as additional evidence for the
recent understanding that platelets contribute to inflamma-
tory reactions (reviewed in reference 28). From guinea pig
models it was suggested that platelets are a prerequisite com-
ponent in allergic asthma, since platelet depletion reduced
Eo infiltration into the lung after PAF or allergen exposure
to sensitized animals (29). Moreover, bronchial Eo accumula-
tion was reduced without a significant change in neutrophil
infiltration after antigen challenge in thrombocytopenic-
allergic rabbits compared with control animals (30). It is there-
fore tempting to speculate that RANTES released from
platelets might play a role for selective Eo infiltration after
antigen challenge in these situations.
We gratefully acknowledge Jutta Quitzau and Marlies Brandt for their excellent technical assistance and
Ilse Brandt for editorial help.
Part of this work was supported by Deutsche Forschungsgemeinschaft (grant Schr 305/1-2).
590 Eosinophil Chemotactic ILANTES

Address correspondence to J.-M. Schr/Sder, Department of Dermatology, University of Kid, Schitten-
helmstraSe 7, D-W-2300, Kid, Germany.
Received for publication I April 1992 and in re,,ised form 1 June 1992.
References
1. Baggiolini, M., A. Walz, and S.L. Kunkel. 1989. Neutrophil-
activating peptide-1/interleukin 8, a novel cytokine that acti-
vates neutrophils.
J. Clin. Invest.
84:1045.
2. Gronhoj-Larsen, C., A.O. Anderson, E. Appella, J.J. Oppen-
heim, and K. Matsushima. 1989. The neutrophil-activating
protein (NAP-l) is also chemotactie for T lymphocytes.
Science
(Wash. DC).
243:1464.
3. Schr6der, J.-M., U. Mrowietz, E. Morita, and E. Christophers.
1987. Purification and partial biochemical characterization of
a human monocyte-derived neutrophil-activating peptide that
lacks Interleukin 1 activity.
J. Immunol.
139:3474.
4. Yoshimura, T., K. Matsushima, S. Tanaka, E.A. Robinson,
E. Appella, J.J. Oppenhaim, and E.J. Leonard. 1987. Puri-
fication of a human monocyte-derived neutrophil chemotactic
factor that shares sequence homology with other host defense
cytokines.
Proc Natl. Acad. Sci. USA.
84:9233.
5. Van Damme, J., J. Van Beeumen, G. Opdenakker, and A. Bil-
liau. 1988. A novel, NHz-terminal sequence-characterized
human monokine possessing neutrophil chemotactic, skin-
reactive and granulocytosis-promoting activity.
J. Ext~ Med.
167:1364.
6. Schr6der, J.-M., N. Persoon, and E. Christophers. 1990.
Lipopolysaccharide-stimulated human monocytes secrete apart
from neutrophil-activating peptide 1/interleukin 8 a second
neutrophil-activating protein: NHz-terminal amino acid se-
quence identity with melanoma growth stimulatory activity.
J. Ext~ Med.
171:1091.
7. Moser, B., I. Clark-Lewis, R. Zwahlen, and M. Baggiolini.
1990. Neutrophil-activating properties of the melanoma
growth-stimulatory activity.
J. Extx Med.
171:1797.
8. Walz, A., B. Dewald, V. yon Tscharuer, and M. Baggiolini.
1989. Effects of the neutrophil-activating peptide NAP-2,
platelet basic protein, connective tissue-activating peptide III,
and platelet factor 4 on human neutrophils.
J. Exlx ivied.
170:1745.
9. Matsushima, K., C. G. Larsen, G. DuBois, and J.J. Oppen-
heim. 1989. Purification and characterization of a novel mono-
cyte chemotactic and activating factor produced by human my-
elomonocytic cell line.
J. Ex F Med.
169:1485.
10. Yoshimura, T., N. Yuhki, S.K. Moore, E. Appella, M.I.
Lerman, and E.J. Leonard. 1989. Human monocyte chemoat-
tractant protein-1 (MCP-1): fuR-length cDNA cloning, expres-
sion in mitogen-stimulated blood mononuclear leukocytes, and
sequence similarity to mouse competence gene JE.
FEBS (Fed.
Eur. Biochem. Soc) Lett.
244:487.
11. Schall, T.J., K. Bacon, K.J. Toy, and D.V. Goeddel. 1990. Selec-
tive attraction of monocytes and T lymphocytes of the memory
phenotype by cytokine RANTES.
Nature (Lond.).
347:669.
12. Schall, T.J. 1991. Biology of the R.ANTES/SIS cytokine family.
Cytokine.
3:165.
13. Morita, E., J.-M. Schr6der, and E. Christophers. 1990.
Identification of a novel and highly potent eosinophil che-
motactic lipid in human eosinophils treated with arachidonic
acid.
J. Immunol.
144:1893.
14. Schr6der, J.-M., M. Sticherling, N.-L.M. Persoon, and E.
Christophers. 1990. Identification of a novel platelet-derived
neutrophil-chemotactic polypeptide with structural homology
to platelet factor 4.
Biochera. Biophys. Res. Coramun.
172:898.
15. Walz, A., and M. Baggiolini. 1989. A novel cleavage product
of fl-thromboglobulin formed in cultures of stimulated
mononuclear cells activate human neutrophils.
Biochem. Bio-
phys. Res. Commun.
159:969.
16. Yoshimura, T., E.A. Robinson, S. Tanaka, E. Appella, and E.J.
Leonard. 1989. Purification and amino acid analysis of two
human monocyte chemoattractants produced by phytohemag-
ghtinin-stimuhted human blood mononuclear leukocytes, f
Immunol.
142:1956.
17. Castor, C.W., J.W. Miller, and D.A. Walz. 1983. Structural
and biological characteristics of connective tissue activating pep-
tide (CTAP-III), a major human platelet-derived growth factor.
Pro~ Natl. Acad. Sci. USA.
80:765.
18. Walz, A., and M. Baggiolini. 1990. Generation of the
neutrophil-activating peptide NAP-2 from platelet basic pro-
tein or connective tissue-activating peptide III through mono-
cyte proteases.
Biochem. Biophys. Res. Comraun.
171:449.
19. Deuel, T.E, P.S. Keim, M. Farmer, and ILL. Heinrikson.
1977.
Amino acid sequence of human platelet factor 4.
Proc Natl.
Acad. Sci. USA.
74:2256.
20. Wardlaw, A.J., R. Moqbd, O. Cromwell, and A.R Kay. 1986.
Platdet activating factor. A potent chemotactic and chemoki-
netic factor for eosinophils.
J. Clin. Invest.
78:1701.
21. Schall, T.J., J. Jongstra, RJ. Dyer, J. Jorgensen, C. Claylberger,
M.M. Davis, and A.M. Krensky. 1988. A human T ceU-specific
molecule is a member of a new gene family.
J. Immunol.
141:1018.
22. Clore, G.M., and A.M. Gronenborn. 1992. NMR and X-Ray
Analysis of the three-dimensional structure of interleukin-8.
In
Interleukin-8 (NAP-l) and Related Chemotactic Cytokines.
Cytokines, Vol. 4. M. Baggiolini, and C. Sorg, editors. Karger,
Basel. 18-40.
23. Gleich, G.J. 1990. The eosinophil and bronchial asthma: cur-
rent understanding.
J. Allergy Clin. Immunol.
85:422.
24. Frew, A.J., and A.R Kay. 1990. Eosinophils and T-Lymphocytes
in late phase allergic reactions.J.
Allergy Clin. Iramunol.
85:533.
25. Wang, J.M., A. Rambaldi, A. Biondi, Z.G. Chen, C.J. San-
derson, and A. Mantovani. 1989. Recombinant human inter-
leukin 5 is a selective eosinophil chemoattractant.
Eur. j. Im-
munol.
19:701.
26. Rand, T.H., W3g r. Cruikshank, D.M. Center, and P.F. WeUer.
1991. CD4-mediated stimulation of human eosinophils: lym-
phocyte chemoattractant factor and other CD4-binding ligands
elicit eosinophil migration.
J. Ex F Med.
173:1521.
27. Frew, A.J., and A.B. Kay. 1991. UCHL1 + (CD45RO +)
"memory" T cells predominate in the CD4 + cellular infiltrate
associated with allergen-induced late-phase skin reactions in
atopic subjects.
Clin. ExF Immunol.
84:270.
28. Page, C.P. 1989. Platelets as inflammatory cells.
Immunophar-
macology.
17:51.
591 Kameyoshi et al. Brief Definitive Report

29. Lellouch-Tubiana, A., J. Lefort, M.-T. Simon, A. Pfister, and
B.B. Vargaftig. 1988. Eosinophil recruitment into guinea pig
lungs after PAF-acether and allergen administration. Modula-
tion by prostacyclin, platelet depletion, and selective antago-
nists.
Am. Rev. Respir. Dis.
137:948.
30. Coyle, A.J., C.E Page, L. Atkinson, K. Flanagan, and W.J.
Metzger. 1990. The requirement for platelets in allergen-
induced late asthmatic airway obstruction. Eosinophil infiltra-
tion and heightened airway responsiveness in allergic rabbits.
Am. Rev. Respir. Dis.
142:587.
592 Eosinophil Chemotactic R.ANTES