Content uploaded by Christopher Haslett
Author content
All content in this area was uploaded by Christopher Haslett
Content may be subject to copyright.
Thrombospondin
Cooperates
with
CD36
and
the
Vitronectin
Receptor
in
Macrophage
Recognition
of
Neutrophils
Undergoing
Apoptosis
John
Savill,
Nancy
Hogg,*
Yi
Ren,
and
Christopher
Haslett
Department
of
Medicine,
Royal
Postgraduate
Medical
School,
Hammersmith
Hospital,
Du
Cane
Road,
London
W12
ONN,
United
Kingdom;
and
*Macrophage
Laboratory,
Imperial
Cancer
Research
Fund,
Lincoln's
Inn
Fields,
London
WC2A
3PX,
United
Kingdom
Abstract
We
have
investigated
the
cell
surface
recognition
mechanisms
used
by
human
monocyte-derived
macrophages
(MO
)
in
phago-
cytosis
of
intact
aging
human
neutrophils
(PMNs)
undergoing
apoptosis.
This
study
shows
that
the
adhesive
protein
thrombo-
spondin
(TSP)
was
present
in
the
interaction,
both
associated
with
the
MO
surface
and
in
solution
at
a
mean
concentration
of
0.59
,g/ml.
The
interaction
was
inhibited
by
treatment
of
MO
(but
not
aged
PMN)
with
cycloheximide,
but
could
be
"res-
cued"
by
replenishment
with
exogenous
TSP.
Under
control
conditions,
MO
recognition
of
aged
PMNs
was
specifically
po-
tentiated
by
purified
platelet
TSP
at
5
,g/ml,
present
either
in
the
interaction
or
if
preincubated
with
either
cell
type,
suggest-
ing
that
TSP
might
act
as
a
"molecular
bridge"
between
the
two
cell
types.
In
support,
both
aged
PMN
and
MO
were
found
to
adhere
to
TSP,
and
phagocytosis
of
aged
PMN
was
specifically
inhibited
by
(a)
excess
soluble
TSP;
(b)
antibodies
to
TSP
that
also
inhibit
TSP-mediated
adhesion
to
aged
PMN;
and
(c)
down-regulation
of
M4
receptors
for
TSP
by
plating
MO
on
TSP-coated
surfaces.
Furthermore,
inhibition
with
mAbs
/
Arg-Gly-Asp-Ser
peptide
of
the
candidate
MO
receptors
for
TSP,
CD36,
and
a,83
exerted
synergistic
effects
on
both
MO
recognition
of
aged
PMN
and
MO
adhesion
to
TSP,
indicating
that
"two
point"
adhesion
of
TSP
to
these
MO
structures
is
involved
in
phagocytosis
of
aged
PMN.
Our
findings
indicate
newly
defined
roles
for
TSP
and
CD36
in
phagocytic
clearance
of
senescent
neutrophils,
which
may
limit
inflammatory
tissue
injury
and
promote
resolution.
(J.
Clin.
Invest.
1992.90:1513-
1522.)
Key
words:
adhesion
*
integrins
*
resolution
*
inflamma-
tion
*
leukocytes
Introduction
In
inflammation,
neutrophils
and
their
contents
can
injure
tis-
sue
and
may
generate
chemotactic
factors
with
potential
to
amplify
leukocyte
infiltration
(
1,
2,
3).
Safe
removal
of
neutro-
phils
from
the
perturbed
site
is,
therefore,
a
prerequisite
of
reso-
lution
of
inflammation.
Uptake
of
intact
senescent
neutrophils
by
macrophages
at
resolving
inflamed
sites
was
recognized
Address
correspondence
to
Dr.
John
Savill,
Ph.D.,
MRCP,
Renal
Unit,
Department
of
Medicine,
Royal
Postgraduate
Medical
School,
Ham-
mersmith
Hospital,
Du
Cane
Road,
London
W12
OHS,
United
King-
dom.
Received
for
publication
25
March
1991
and
in
revised
form
23
March
1992.
many
years
ago
(4),
and,
for
example,
has
been
shown
to
be
the
dominant
mechanism
of
neutrophil
clearance
from
the
experi-
mentally
inflamed
peritoneum
(5,
6).
However,
only
recently
have
the
mechanisms
involved
in
this
cellular
interaction
come
under
study.
Building
on
earlier
work
(7),
we
showed
that
human
neutrophils
isolated
from
blood
or
inflamed
sites
and
aged
for
-
24
h
in
culture
undergo
morphological
and
bio-
chemical
changes
typical
of
programmed
cell
death
(apopto-
sis),
a
process
characterized
by
evidence
of
endogenous
endo-
nuclease
activation
(
8-1
1
).
Apoptosis
in
aging
neutrophils
de-
termined
recognition
and
uptake
of
the
intact
senescent
cell
by
human
monocyte-derived
macrophages
and
human
macro-
phages
isolated
from
inflamed
sites
in
vivo.
As
apoptotic
neu-
trophils
retained
their
membrane
integrity
and
did
not
release
toxic
contents,
these
processes
appear
to
represent
a
mecha-
nism
for
disposal
of
intact
senescent
neutrophils
in
a
manner
likely
to
limit their
toxic
potential
(8).
In
the
present
study,
we
examine
the
cell
surface
mechanisms
used
by
macrophages
in
recognition
of
apoptotic
neutrophils.
Previously,
we
found
that
the
macrophage
aCf33
integrin,
or
vitronectin
receptor
(VnR),'
has
a
role
in
recognition
of
aged
neutrophils
(12).
The
interaction
depended
on
the
divalent
cations
Ca"+
and
Mg",
and
was
specifically
inhibited
by
(a)
the
tetrapeptide
Arg-Gly-Asp-Ser
(RGDS)
at
1
mM,
but
not
the
control
peptide
Arg-Gly-Glu-Ser
(RGES);
(b)
by
soluble
and
substrate-bound
fibronectin
(Fn)
and
vitronectin
(Vn),
but
not
fibrinogen
or
type
IV
collagen,
which
also
bear
the
RGD
recognition
sequence;
and
(c)
by
monoclonal
antibodies
to
both
subunits
of
the
aCf33
integrin,
which
is
expressed
by
the
MO
not
the
apoptotic
neutrophil
(
12-14).
It
was
concluded
that
the
VnR
played
a
direct
role
in
the
recognition
mecha-
nism,
a
new
function
for
this
receptor
in
addition
to
its
known
roles
in
cell
anchorage
to
matrix
proteins
(
13
).
However,
MO
recognition
of
aged
neutrophils
was
also
directly
modulated
by
pH
and
specifically
inhibited
by
millimolar
concentrations
of
aminosugars
and
basic
aminoacids
in
a
charge-dependent
fash-
ion
(15),
and
such
observations
were
not
explained
by
the
molecular
interactions
of
the
MO
VnR.
Cationic
amino
sugars
and
basic
amino
acids
also
inhibit
the
trypsin-sensitive
"lectin-like"
property
displayed
by
acti-
vated
platelets
in
agglutination
of
fixed
platelets
and
fixed
tryp-
sinized
erythrocytes
(
16).
Subsequently,
the
activated
platelet
"lectin"
was
shown
to
be
thrombospondin
(TSP)
(
17).
TSP
is
a
"multifunctional"
trimeric
adhesive
molecule
of
-
450
kD
capable
of
binding
to
a
wide
range
of
macromolecules
and
many
cell
types
(
18-22).
TSP
has
been
implicated
as
a
"molec-
1.
Abbreviations
used
in
this
paper:
EIgG,
ox
red
erythrocytes
opson-
ized
with
rabbit
IgG;
Fn,
fibronectin;
MO,
monocyte-derived
macro-
phages;
RGDS,
tetrapeptide
Arg-Gly-Asp-Ser;
RGES,
control
peptide
Arg-Gly-Glu-Ser;
TSP,
thrombospondin;
Vn,
vitronectin;
VnR,
av,3
vitronectin
receptor.
Thrombospondin
in
Phagocytosis
ofAged
Neutrophils
1513
J.
Clin.
Invest.
©
The
American
Society
for
Clinical
Investigation,
Inc.
0021-9738/92/10/1513/10
$2.00
Volume
90,
October
1992,
1513-1522
ular
bridge"
mediating
adhesive
interactions
between
activated
platelets
and
other
cells,
including
erythrocytes
and
monocyte/
macrophages
(23-25).
However,
the
mechanisms
by
which
TSP
binds
to
cells
are
incompletely
understood
and
potentially
complex,
since
there
is
evidence
that
a
number
of
cell
surface
molecules
may
be
involved,
including
sulfatides,
proteoglycans
and
proteins,
including
al33
and
the
88-kD
monomer
GPIV/
CD36
(23,
25-28).
Although
first
isolated
from
the
a-granule
of
the
platelet,
TSP
has
been
found
to
be
synthesized
and
se-
creted
by
a
wide
range
of
cell
types
in
culture
(20-22),
includ-
ing
both
macrophages
and
neutrophils
(29,
30),
and
there
is
evidence
that
each
cell
type
may
bear
receptors
for
TSP
(28,
31-34).
Indeed,
macrophages
may
have
TSP
associated
with
their
surfaces
(33,
34).
Furthermore,
TSP
has
been
shown
to
be
a
transient
component
of
the
inflammatory
extracellular
matrix
of
healing
wounds
present
at
the
time
in
which
neutro-
phils
are
removed
from
the
tissue
(35),
and
has
also
been
dem-
onstrated
in
lavage
fluid
from
inflamed
lungs
(36).
Conse-
quently,
in
the
present
study,
we
sought
evidence
that
TSP
and
TSP-binding
structures
were
involved
in
MO
recognition
of
aged
neutrophils.
Our
data
indicate
that
thrombospondin
can
act
as
a
"molecular
bridge"
between
the
aged
neutrophil
and
the
MO,
where
the
molecule
appears
to
bind
to
both
GPIV/
CD36
and
the
aCl33
integrin,
indicating
hitherto
unrecognized
functions
for
both
TSP
and
GPIV/CD36.
Methods
Materials
All
chemicals
were
from
Sigma
Chemical
Co.
(St.
Louis,
MO),
unless
otherwise
indicated;
culture
media
(HBSS,
Iscove's
DME)
and
supple-
ments
(
100
U/liter
of
penicillin
and
streptomycin)
were
from
Gibco
Laboratories
(Grand
Island,
NY);
and
sterile
tissue
culture
plasticware
were
from
Falcon
Plastics
(Cockeysville,
MD).
Peptides
and
proteins
The
tetrapeptides
RGDS
and
RGES
were
obtained
from
Peninsula
Laboratories,
Inc.,
(Belmont,
CA),
human
albumin
(fraction
V)
was
from
Sigma
and
fibronectin
from
Calbiochem
Corp.
(San
Diego,
CA).
TSP
was
purified
from
thrombin-stimulated
human
platelets
by
hepa-
rin-Sepharose
affinity
chromatography
followed
by
"sizing"
on
a
4B-
Sepharose
column
by
standard
methods
(18).
Particular
care
was
taken
to
use
only
those
fractions
in
which
the
protein
ran
as
a
single
band
on
reducing
SDS-PAGE
assessed
by
silver
staining
and
immuno-
blotting
with
a
mAb
specific
for
TSP
(TSP-B7;
Sigma).
Contamination
with
Vn
was
quantified
at
<
0.1
%
by
specific
ELISA
(data
courtesy
of
Dr.
W.
Bennett).
Proteins
and
peptides
were
dialyzed
in
HBSS
be-
fore
use.
Monoclonal
antibodies
Dr.
V.
Dixit
kindly
provided
a
panel
of
IgG1
mAbs
specific
for
TSP,
namely
C6.7,
A6.
1,
A2.5,
and
D4.6,
each
of
which
has
been
shown
to
inhibit
various
TSP-related
phenomena
(37-39);
Drs
R.
Nachman
and
R.
Silverstein
provided
the
TSP-specific
mAb
11.2
(24).
CD36
mAbs
were
SM4,
an
IgM
(40,
41)
and
ClMegl,
an
IgG,
(40)
(a
gift
from
Dr.
G.
Pilkington).
The
aC(3
mAb
was
23C6
(42)
(a
gift
from
Dr.
M.
Horton).
Control
mAbs
were
P3
(43)
an
irrelevant
IgG,
(a
gift
from
Dr.
P.
Morganelli),
P112
(44)
an
IgM
that
recognizes
the
GPII-
bIIIa
complex,
28
(40,
44)
an
IgM
recognizing
CDI
5,
and
OX7
(Sero-
tec,
Banbury,
Oxon,
U.K.),
an
IgG,
that
recognizes
Thyl.
1.
Antibodies
were
used
as
purified
immunoglobulin
(23C6,
11.2,P3)
or
ascites
(C6.7,
A2.5,
A6.
1,
D4.6,
SMU,
ClMegl,
23C6,
P112,
OX7)
diluted
in
HBSS
as
described.
Cells
Neutrophils
(>
98%
pure
May-Giemsa)
were
isolated
from
fresh,
ci-
trated
normal
human
blood
by
dextran
sedimentation
and
plasma-Per-
coll
(Pharmacia
Fine
Chemicals,
Piscataway,
NJ),
aged
in
tissue
cul-
ture
for
-
28
h
in
Iscove's
DME
with
10%
autologous
platelet-rich,
plasma-derived
serum,
and
apoptosis
verified
by
oil-immersion
light
microscopy
of
May-Giemsa
stained
cytopreps,
exactly
as
previously
described
(8,
15).
Human
monocyte-derived
macrophages
(M+)
were
prepared
by
standard
methods
from
adherent
PBMC
by
culture
for
5-7
d
in
24-well
plates
in Iscove's
medium
with
10%
autologous
serum
as
described
(45,
8).
No
contaminating
platelets
attached
to
M4
could
be
detected
either
by
careful
microscopic
inspection
or
by
immunofluores-
cence
using
mAb
P112
specific
for
platelet
GPIIbIIIa.
For
demonstra-
tion
of
cell
surface
TSP
by
flow
cytometry
(see
below),
monocytes
were
purified
by
elutriation
(46)
and
matured
in
nonadherent
culture
in
teflon-lined
vessels
for
5-7
d.
Ox
erythrocytes,
opsonized
with
polyclo-
nal
rabbit
anti-ox
IgG
(gift
from
Dr.
D.
Grennan)
were
prepared
as
described
(
15).
Assay
and
localization
of
TSP
TSP
in
solution
in
medium
was
assayed
by
ELISA
exactly
as
described
(34),
using
rabbit
polyclonal
immunoglobulin
specific
for
TSP
(
1:5,000
in
PBS;
a
gift
from
Dr.
N.
Hunter);
this
assay
was
sensitive
to
a
lower
limit
of
50
ng/ml.
TSP
was
sought
on
the
surface
of
cultured
MO
and
neutrophils
by
immunofluorescence
flow
cytometry,
using
a
standard
protocol
of
labeling
and
three
washes
at
the
end
of
each
step,
as
described
(
12).
Binding
of
C6.7
(
1:100
dilution
in
PBS/0.1%
BSA)
was
compared
with
an
identical
dilution
of
OX7
ascites
as
a
control,
detected
using
FITC-conjugated
Fab'2
sheep
anti-mouse
immunoglobu-
lin
antibody
(
1:30;
Sigma).
Aged
neutrophils
were
studied
by
the
same
techniques.
Interaction
assays
This
microscopically
quantified
phagocytic
assay
has
been
described
and
illustrated
in
detail
before
(8,
15).
Briefly,
aged
neutrophils
were
washed
once
in
HBSS
and
suspended
in
HBSS.
2.5
X
106
aged
PMN
in
300
M1
HBSS
were
added
to
each
washed
well
of
MO
and
incubated
at
37°C
in
a
5%
CO2
incubator
at
pH
7.4.
In
view
of
our
previous
findings
of
the
importance
of
pH
(
15),
particular
care
was
taken
to
ensure
that
the
pH
of
HBSS
was
adjusted
and
maintained
correctly.
At
the
end
of
the
interaction
period,
the
wells
were
washed
in
saline
at
4°C,
fixed
with
2%
glutaraldehyde
in
PBS,
stained
for
myeloperoxidase
and
then
the
proportion
of
MO
ingesting
neutrophils
counted
by
inverted
light
microscopy,
exactly
as
described
(8,
15).
In
keeping
with
our
previous
description
ofthis
assay,
the
proportion
of
M4
recognizing
aged
neutro-
phils
varied
between
donors,
so
to
compare
data
between
experiments
in
which
MO
from
different
donors
were
used,
data
are
presented
as
percentage
of
the
mean
of
control
in
each
of
the
relevant
experiments,
as
described
(
15).
However,
the
absolute
percentage
of
M4
recognizing
aged
neutrophils
in
the
relevant
series
of
experiments
is
also
given
in
the
figure
legend.
Uptake
of
opsonized
ox
erythrocytes
(EIgG)
was
determined
by
similar
means,
as
described
(
15).
In
all
experiments,
>
95%
of
Mk
took
up
EIgG.
Effects
of
proteins
in
solution.
TSP
or
human
albumin
were
in-
cluded
in
the
interaction
medium
at
the
desired
concentration
in
HBSS
and
interactions
performed
in
HBSS
alone
served
as
control.
In
prein-
cubation
experiments,
proteins
at
S
ug/ml
were
incubated
with
washed
aged
neutrophils
or
M4
at
37°C
for
15
min
(HBSS
alone
being
used
in
"control"
experiments),
washed
and
then
incubated
with
untreated
MO
or
neutrophils
as
appropriate
under
the
standard
conditions
of
the
assay.
Effects
of
monoclonal
antibodies.
These
were
determined
as
previ-
ously
described
(12,
15).
Briefly,
adherent
MO
cultured
in
24-well
plates
were
washed,
300
Ml
of
mAb
at
desired
concentration
in
HBSS
(or
HBSS
alone
as
a
control)
was
added
to
each
well.
The
plates
were
incubated
for
15
min
at
4°C,
followed
by
addition
of
2.5
x
106
washed
aged
neutrophils
(or
EIgG)
in
100
Al
of
warm
HBSS
and
then
interac-
1514
J.
Savill,
N.
Hogg,
Y.
Ren,
and
C.
Haslett
tion
for
30
min
under
standard
conditions.
In
no
experiment
did
mAbs
affect
cell
viability
as
assessed
by
trypan
blue
exclusion.
In
one
series
of
experiments
designed
to
determine
the
cellular
localization
of
the
effect
of
mAb
SMO
either
aged
neutrophils
(at
2.5
X
106/ml)
or
adherent
MO
were
preincubated
with
mAb
SM40
(
1:25
dilution)
for
30
min
at
4VC,
before
washing
twice
in
HBSS
and
interaction
with
untreated
cells
of
the
opposite
type
under
standard
conditions.
These
effects
were
com-
pared
with
those
of
presence
of
mAb
at
the
same
concentration
in
the
interaction
medium,
without
the
usual
15-min
M4$
preincubation.
It
should
be
noted
that
in
previous
work,
a
large
number
of
murine
mAbs
of
varying
isotype
have
been
examined
in
this
assay
in
the
form
of
intact
Ig
(either
purified
or
as
hybridoma
ascites)
(
12,
15).
Except
for
mAbs
to
afl3,
none
has
inhibited,
including
mAbs
to
MO
and
neutrophil
surface
structures,
indicating
that
nonspecific
"steric"
ef-
fects
are
unlikely
to
be
the
reason
for
the
blocking
effects
of
mAb
in
this
system.
Furthermore,
no
mAb
(including
those
used
in
this
study)
have
promoted
M40
phagocytosis
of
freshly
isolated
neutrophils
in
a
manner
suggesting
that
murine
mAbs
might
opsonize
neutrophils
for
M4,
thus
complicating
the
interpretation
of
these
data.
Effects
of
plating
M4
on
tissue
culture
surfaces
treated
with
pro-
teins.
TSP
was
adsorbed
to
the
bottom
of
tissue
culture-treated
wells
by
standard
methods
(47-49,
27).
50
Al
of
a
solution
of
TSP
or
human
albumin
at
80
tg/
ml
in
HBSS
was
incubated
in
each
well
of
a
96-well
flat-bottomed
Falcon
tissue
culture
plate
for
2
h
at
room
temperature
and
washed
twice
in
HBSS
and
"control"
wells
were
incubated
with
HBSS
alone.
Mature
M4
were
"mobilized"
from
adherent
culture
by
methods
shown
by
immunofluorescence
flow
cytometry
to
"strip"
them
of
surface-associated
TSP
(data
not
shown)
to
facilitate
modula-
tion
of
receptors
by
substrate-bound
proteins.
Thus
MO
were
incubated
with
5
mM
EDTA
in
HBSS
with
no
Ca"+
or
Mg"+
for
15
min
at
4VC,
detached
from
the
plate
by
pipetting,
washed
twice
in
warm
HBSS,
suspended
at
5
X
105/ml
and
50
,l
of
suspension
added
to
each
well.
The
cells,
which
formed
a
"subconfluent"
monolayer,
underwent
re-
ceptor
modulation
(or
not)
during
incubation
at
370C
for
30
min.
The
medium
was
then
aspirated
and
replaced
with
100
Ml
of
a
suspension
of
aged
neutrophils
in
HBSS
at
2.5
x
106/ml,
or
EIgG
at
5
x
106/ml
for
a
30-min
interaction.
Counting
of
M40
by
inverted
microscopy
showed
that
<
10%
of
Mk
originally
added
to
the
well
were
lost
during
these
steps,
indicating
that
adherent
MO
were
not a
subpopulation.
Assay
of
adhesion
of
aged
neutrophils
to
latex
beads
In
preliminary
experiments,
apoptotic
PMN
did
not
spread
on
tissue
culture-treated
surfaces
or
hemocytometer
glass,
remaining
spherical
and
resisting
mechanical
displacement only
weakly
(Whyte,
M.K.B.,
J.S.
Savill,
and
C.
Haslett;
unpublished
data).
Consequently,
we
sought
adhesion
to
protein-coated
0.995-,Mm
latex
beads
(LBl
1;
Sigma)
by
minor
modifications
of
standard
methods
(50,
51
),
since
this
does
not
require
neutrophil
spreading.
A
300-Ml
aliquot
of
a
10%
stock
suspen-
sion
of
latex
beads
was
added
to
a
1.5-ml
polythene
Eppendorf
tube,
pelleted
and
washed
three
times
in
0.9%
saline,
and
then
incubated
with
either
TSP
or
human
Fn
(Calbiochem)
at
100
gg/ml
in
0.9%
saline
with
Tris
HCl
20
mM,
pH
7.4,
for
10
min
at
room
temperature.
The
bead
suspension
was
then
sonicated
with
a
probe
sonicator
(60
W)
to
break
up
microclumps,
centrifuged
and
resuspended
in
10
mg/ml
human
albumin
in
HBSS,
then
incubated
for
a
further
10
min
at
room
temperature.
Beads
were
pelleted
and
washed
three
times
in
HBSS
before
resuspension
in
500
Ml
HBSS;
i.e.,
a
6%
suspension.
TSP
coating
was
confirmed
by
positive indirect
immunofluorescence
with
mAb
11.2
at
2.5
Ag/ml
and
negative
staining
with
irrelevant
mAb
P3
at
the
same
concentration.
Aged
neutrophils
were
washed
and
resuspended
at
50
X
106/ml
in
HBSS,
and
225
,l
were
added
to
each
test
tube
(Falcon
LP3
polythene
tubes).
In
some
experiments
TSP
mAbs
C6.7,
A2.5,
A6.
1,
and
D4.6
were
included
in
the
medium
at
a
dilution
of
1
in
25
of
ascites.
25
ml
of
freshly
prepared
6%
beads
was
added
to
the
cell
suspen-
sion
to
achieve
a
final
bead
concentration
of
0.6%,
and
the
tube
was
shaken
in
a
370C
water
bath
at
110
beats/min
for
15
min,
supple-
mented
by
manual
agitation
at
7
min.
In
preliminary
experiments
under
these
conditions,
TSP
mAbs
at
the
concentration
described
above
did
not
aggregate
either
beads
or
aged
neutrophils
alone,
indicat-
ing
that
inhibitory
effects
on
bead
binding
were
unlikely
to
be
by
such
"artefactual"
mechanisms
that
might
have
reduced
the
area
of
inter-
face
between
beads
and
cells.
The
cells
were
fixed
by
addition
of
250
M1
of
2%
glutaraldehyde
in
PBS
for
10
min,
and
then
cells
separated
from
unbound
beads
(which
are
of
lower
density)
by
four
consecutive
centrif-
ugations
at
200
g
for
4
min
with
washing
in
0.9%
saline,
before
finally
suspending
beads
in
250
Ml
0.9%
saline
and
placing
50
M1
on
a
micro-
scope
slide.
The
cells
were
inspected
by
oil
immersion
light
microscopy
after
placing
a
cover
slip
on
the
drop.
500
neutrophils
were
examined
to
determine
the
percentage
of
cells
binding
beads,
and
then
the
number
of
beads
on
each
of
100
cells
with
beads
were
counted.
Assay
of
Mq
adhesion
Established
methods
(47,
27),
with
minor
modifications,
were
used.
Tissue
culture
wells
in
96-well
flat
bottomed
Falcon
plates
were
incu-
bated
with
gelatin
(Sigma)
at
30
mg/ml
in
water
for
2
h
at
370C,
aspirated
to
dryness
and
then
thoroughly
dried
in
an
oven
at
40C
for
a
further
2
h.
The
wells
were
then
cooled,
and
50
Ml
of
TSP
or
Fn
(Calbio-
chem)
at
40
Mg/ml
incubated
in
each
well
for
2
h
at
room
temperature.
HBSS
alone
was
used
as
a
control.
The
protein
suspension
was
then
aspirated,
the
wells
washed
twice
with
HBSS,
and
the
remaining
me-
dium
was
aspirated.
Mature
(5
to
7
d)
MO
were
released
from
adherent
tissue
culture
by
incubation
with
5
mM
EDTA
in
the
cold,
washed
as
above,
and
suspended
at
106/ml
in
HBSS
alone
or
HBSS
with
inhibi-
tors/controls.
The
following
inhibitors
or
controls
were
dissolved
in
HBSS
and
used
at
the
stated
concentration,
either
alone
or
in
combina-
tion:
RGDS
and
RGES
at
2.2
mM;
the
CD36
mAb
SMO,
the
a!f33
mAb
23C6,
and
the
control
mAbs
P112
and
OX7
were
all
in
the
form
of
ascitic
fluid
diluted
1:25.
From
these
cell
suspensions,
50-,Ml
aliquots
were
immediately
added
to
protein-coated
wells
for
45
min
at
room
temperature
followed
by
washing
with
0.9%
saline
at
4°C.
The
wells
were
fixed
with
2%
glutaraldehyde
in
PBS,
and
the
number
of
adherent
M4
in
10
randomly
selected
microscope
fields
counted
using
an
in-
verted
instrument
with
a
40X
objective.
Treatment
of
PMNs
and
Mk
with
cycloheximide
PMNs
isolated
from
the
same
donor
were
cultured
either
under
stan-
dard
conditions
(see
above),
or
in
conditions
that
were
identical,
ex-
cept
for
the
presence
of
cycloheximide
at
5
x
10-6
M.
Before
interac-
tion
with
MO
under
standard
conditions,
PMNs
were
washed
three
times
to
ensure
removal
of
cycloheximide,
and
viability
was
deter-
mined
by
trypan
blue
exclusion.
Adherent
MO
in
24-well
plates
were
washed
twice
in
HBSS,
then
incubated
for
6
h
in
5%
CO2
at
37°C
in
Iscove's
DME
as
a
control,
or
with
the
same
medium
containing
cycloheximide
at
2.5
X
10-6
M.
At
the
end
of
this
period,
aliquots
of
medium
were
taken
for
measurement
of
TSP
concentration
by
ELISA
(see
above).
The
wells
were
then
washed
three
times
and
interacted
with
aged
neutrophils
or
EIgG
under
standard
conditions.
In
some
experiments,
the
interaction
medium
contained
TSP,
Fn,
or
human
albumin
at
4
Mg/ml.
Results
Macrophage-synthesized
TSP
participates
in
phagocytosis
of
apoptotic
neutrophils.
We
first
investigated
whether
TSP
was
present
in
the
standard
interaction
assay
in
which
106
MO
were
coincubated
with
2.5
X
106
aged
PMN
in
300
,l
of
serum-free
medium
for
30 min.
In
keeping
with
previous
studies
(29,
33,
34),
TSP
could
be
demonstrated
in
two
forms.
First,
TSP
was
detected
in
medium
at
the
end
of
the
interaction
at
a
concen-
tration
of
0.59±0.14
,g/ml
(n
=
6);
second,
low
levels
of
TSP
were
detected
on
the
surface
of
MO
by
immunofluorescence
flow
cytometry
(Fig.
1).
However,
TSP
could
not
be
demon-
strated
by
the
same
techniques
on
the
surface
of
aged
PMN
(data
not
shown).
Furthermore,
we
confirmed
that
the
capac-
Thrombospondin
in
Phagocytosis
ofAged
Neutrophils
1515
Figure
1.
Expression
of
surface
TSP
by
Mo.
Typical
fluorescence
histograms
from
flow
cytometry
of
MO,
showing
binding
of
the
IgG,
anti-TSP
mAb
C6.7
compared
with
positive
control
CD36
mAb
and
negative
con-
__-~~~-~-~-~~~-
trol
OXI
mAb.
*,
anti-
|
0
1=
|0
1
000
thrombospondin;
m,
1190
100
1000''
negative
control
OX
1;
Mean
Channel
Fluorescence
a,
CD36.
ity
of
MO
to
synthesize
TSP
was
comparable
to
that
previously
reported
(29),
and
was
over
40-fold
greater
than
that
of
aged
PMN
(Table
I).
Finally,
to
determine
if
TSP
might
play
a
role
in
Mi
recognition
of
aged
PMN,
we
studied
whether
the
inter-
action
was
influenced
by
manipulation
of
MO
TSP
synthesis
with
2.5
pM
cycloheximide
(Table
II).
This
treatment
mark-
edly
reduced
the
proportion
of
MO
taking
up
aged
PMNs
but
did
not
affect
uptake
of
EIgG.
Furthermore,
although
cyclo-
heximide
treatment
may
be
expected
to
inhibit
synthesis
of
a
wide
range
of
proteins,
"adding
back"
of
TSP
at
4
,g/ml
(but
not
Fn
or
albumin)
to
cycloheximide-inhibited
MO
partially
"rescued"
the
capacity
to
ingest
aged
PMN
(Table
II).
Aging
of
freshly
isolated
PMNs
for
22
h
in
the
presence
of
5
mM
cyclo-
heximide
did
not
inhibit
their
subsequent
recognition
by
Mo.
Thus,
treated
PMNs
were
taken
up
by
55.9±4.2%
of
Mo,
com-
pared
with
48.1±4.7%
uptake
of
untreated
PMNs
by
Mo
(n
=
4).
These
data
indicated
that
macrophage-synthesized
TSP
plays
a
role
in
recognition
of
aged
PMN.
Exogenous
TSP
modulates
macrophage
recognition
of
apoptotic
neutrophils.
To
examine
whether
TSP
might
be
play-
ing
an
adhesive
role
in
the
interaction,
exogenous
TSP
was
included
in
the
interaction
medium
(Fig.
2).
At
concentrations
up
to
a
peak
of
5
jg/ml,
added
TSP
increased
the
proportion
of
M4
ingesting
aged
PMN.
This
effect
was
specific
as
albumin
(Fig.
2),
Vn
and
Fn
(
12)
at
the
same
concentrations
had
no
effect,
and
the
potentiating
effect
of
added
TSP
was
completely
inhibitable
by
mAb
against
TSP
(Table
III).
Furthermore,
TSP
did
not
induce
Mo
recognition
of
freshly
isolated
PMN.
This
indicated
that
although
Mo
have
TSP
associated
with
their
surface,
availability
of
"extra"
TSP
in
solution
appears
to
po-
Table
I.
Release
of
TSP
into
Medium
Over
6
h:
Mq
Compared
with
Aged
PMN
Cell
Conditions
TSP
release
(pig/106
cells/6
h)
Mo
Control
1.84
Mo
Cycloheximide
0.14*
Aged
PMN
Control
0.04*
*
P
<
0.05
compared
with
TSP
release
by
Mo
under
control
condi-
tions.
Data
are
means
of
six
observations.
Over
the
6-h
period
in
each
set
of
conditions,
each
cell
type
remained
>98%
viable
by
trypan
blue
exclusion.
The
inhibitory
effect
of
2.5
AtM
cycloheximide
indi-
cates
that
the
majority
of
TSP
released
under
control
conditions
had
been
newly
synthesized.
Table
IL
Effect
on
Aged
Neutrophil
Recognition
of
Treating
Mk
with
2.5
,M
Cycloheximide
for
6
h
(n
=
12):
Partial
"Rescue"
by
TSP
Protein
in
Aged
neutrophil
Preincubation
interaction
recognition
EIgG
uptake
(at
4
pg/ml)
(%
of
control±SE)
(%
of
MO)
Control
100.0±1.26
>95%
Cycloheximide
22.2±1.05
>95%
Cycloheximide
Albumin
21.5±1.22
>95%
Cycloheximide
Fn
28.4±1.51
>95%
Cycloheximide
TSP
61.1±2.75*
>95%
In
these
experiments,
57.3±5.3%
of
MO
recognized
aged
neutrophils
under
control
conditions.
*
P
<
0.05
relative
to
cycloheximide-
treated
M+.
tentiate
the
interaction.
Indeed,
it
appeared
that
the
potentiat-
ing
effect
of
5
ug/ml
TSP
could
be
exerted
at
either
cell
surface,
since
preincubation
of
either
cell
type
with
TSP
enhanced
sub-
sequent
interaction
under
standard
conditions
(Fig.
3).
By
contrast,
at
higher
concentrations,
TSP
specifically
inhibited
MO
recognition
of
aged
neutrophils;
albumin
had
no
effect
at
100
ugg/ml,
and
TSP
at
this
concentration
did
not
affect
My
uptake of
a
control
particle,
the
IgG-opsonized
erythrocyte.
Although
these
results
suggested
that
the
role
of
TSP
in
the
interaction
might
be
to
form
a
"molecular
bridge"
between
Mq
and
aged
PMN,
further
supportive
evidence
would
come
from
demonstration
that
(a)
TSP
can
support
adhesion
by
both
cell
types;
and
(b)
inhibition
of
TSP
binding
to
either
cell
type
inhibits
the
interaction.
TSP
supports
adhesion
to
apoptotic
neutrophils.
TSP
can
mediate
adhesion
of
cells
of
monocyte/macrophage
lineage
(24,
27,
34).
However,
it
has
not
been
known
whether
TSP
could
support
adhesion
by
apoptotic
PMNs,
which
would
be
a
O
200-
C
0
S-
c
0
gi
-
100-
0
I0
01
-Q
0
control
42SD
°
I
(n=101
1
--I
0.1
0.14
1
4i
5
10
140
100
(Thrombospondin]
pg/ml
Figure
2.
Effect
of
TSP
in
solution
upon
M4
recognition
of
aged
neutrophils.
Each
point
represents
mean±SE
of
10
observations
made
in
experiments
where
44.8±4.3%
of
M4
recognized
aged
neutrophils
under
control
conditions.
At
100
,g/ml,
TSP
failed
to
affect
MO
up-
take
of
EIgG
(open
square).
The
dotted
lines
represent
±
2
SD
of
159
observations
of
My
recognition
of
aged
neutrophils
made
under
control
conditions
in
previous
work
to
determine
the
range
of
in-
traexperimental
variability
of
the
assay
(
15
).
1516
J.
Savill,
N.
Hogg,
Y.
Ren,
and
C.
Haslett
-j
Table
III.
Specificity
of
Potentiating
Effect
of
TSP
on
MO
Recognition
of
Aged
Neutrophils:
Specific
Inhibition
by
Anti-TSP
mAb
and
Failure
to
Induce
Recognition
of
Freshly
Isolated
PMN
Nature
of
Protein
in
mAb
in
PMN
interaction
interaction
Recognition
(%
of
control±SE)
Aged
TSP
5
jig/ml
None
152.1±6.3
Aged
TSP
5
,g/ml
Anti-TSP
53.7±6.3*
Aged
TSP
5
gg/ml
Control
154.3±4.6
Fresh
None
None
3.4±0.3
Fresh
TSP
5
gg/ml
None
2.8±0.4
In
these
experiments,
39.0±3.4%
(n
=
9)
of
MO
recognized
aged
PMNs
under
control
conditions.
The
anti-TSP
MAb
was
C6.7
at
a
1:25
dilution
of
ascites;
the
control
mAb
was
OX7
at
the
same
dilution.
*
P
<
0.05;
in
separate
experiments,
this
concentration
of
C6.7
was
shown
not
to
impair
M,0
ingestion
of
EIgG.
prerequisite
for
a
possible
"bridging"
role
for
TSP
in
the
inter-
action.
Apoptotic
PMNs
adhered
to
latex
beads
coated
with
TSP
to
a
greater
degree
than
beads
coated
with
a
Fn,
used
as
an
adhesive
protein
control
(Table
IV).
Furthermore,
binding
of
apoptotic
PMN
to
TSP-coated
beads
was
specifically
inhibited
by
three
mAbs
(C6.7,
A6.
1,
and
A2.5)
against
TSP,
but
not
by
a
fourth
TSP
mAb,
D4.6.
These
data
indicate
that
TSP
can
support
adhesion
to
apoptotic
PMNs.
mAbs
to
TSP
inhibit
the
interaction.
Surface
TSP
was
not
detected
on
aged
PMN
(see
above),
indicating
that
if
TSP
were
to
"bridge"
macrophage
to
apoptotic
neutrophil,
then
either
TSP
in
solution
or
TSP
attached
to
the
MO
surface
must
bind
to the
apoptotic
neutrophil.
Therefore,
in
the
absence
of
added
TSP,
mAbs
to
TSP
shown
to
inhibit
TSP
binding
to
LUl
en
+1
-
0
u
0
c
a
a
E
~0
to
z
a-
-n
a
0.
150-
100-
50
-
0o
T
Lii
Aib
TSP
Table
IV.
Specific
Binding
of
TSP-coated
Beads
to
Aged
PMN.
Specific
Inhibition
by
Anti-TSP
mAbs
C6.
7,
A6.
1,
and
A2.5,
but
not
by
D4.6
mAb
present
in
Protein
on
bead
bead-cell
interaction
Beads
binding
per
cell
(mean±SE,
n
=
6)
TSP
-
3.26±0.07
TSP
C6.7
(anti-TSP)
0.43±0.08*
TSP
A6.
1
(anti-TSP)
0.38±0.07*
TSP
A2.5
(anti-TSP)
0.46±0.07*
TSP
D4.6
(anti-TSP)
3.51±0.26
TSP
P112
(control)
3.19±0.23
Fn
-
0.17±0.03
Monoclonal
antibodies
were
used
as
a
1:25
dilution
of
ascites.
*
P
<
0.05
compared
with
TSP
bead
binding
to
aged
PMN
under
control
conditions.
apoptotic
PMNs
might
be
expected
to
inhibit
the
interaction.
Specific
inhibition
of
M4
ingestion
of
aged
PMN
was
observed
for
the
three
mAbs,
C6.7,
A6.1,
and
A2.5, while
mAb
D4.6
affected
neither
bead
binding
nor
the
interaction
(Fig.
4).
No
mAb
from
this
panel
inhibited
MO
ingestion
of
EIgG.
Further-
more,
mAb
3E3,
which
recognizes
the
cell-binding
domain
of
Fn
and
inhibits
Fn
adhesion
to
fibroblasts
(52),
failed
to
in-
hibit
the
interaction
(data
not
shown),
confirming
that
the
inhibitory
effect
of
anti-TSP
mAbs
was
not
some
general
prop-
erty
of
mAbs
against
adhesive
proteins.
Therefore,
these
exper-
iments
provided
direct
evidence
in
support
of
a
"physiologi-
cal"
bridging
role
for
TSP
in
the
interaction.
Inhibition
of
macrophage
receptors
for
TSP
specifically
im-
pairs
recognition
of
apoptotic
neutrophils.
A
further
strategy
to
examine
whether
TSP
has
a
bridging
role
in
the
interaction
would
be
to
manipulate
MO
surface
adhesion
receptors
for
TSP.
Attachment
of
macrophages
to
substrates
coated
with
ligand
so
that
receptors
become
modulated
to
the
underside
of
the
cell
is
a
well-established
method
of
down-regulating
ligand
100
oz
Z _
C
S
a
_f
C
_P
u
to
w
E
C
l_
m-
PMN
Msl
Protein
in
preincubn
preincubn
interaction
ln='5
(n=5)
(n=10)
Figure
3.
Localization
of
potentiating
effect
of
5
jig/ml
TSP.
In
these
experiments,
where
38.1±0.7%
of
MO
recognized
aged
neutrophils
under
control
conditions,
aged
neutrophils
(left
panel)
or
M4
(center
panel)
were
preincubated
with
either
TSP
or
human
albumin
(Alb),
and
the
effects
compared
with
presence
of
the
proteins
at
the
same
concentration
in
the
interaction
(right
panel).
The
dotted
lines
are
±
2
SD
of
control
observations,
as
defined
in
Fig.
2.
o
0-
Medium
alone
U
a
a
a
A2.5
A6.1I
C6.7
D
4.6
-.-Anti-TSP
mAbs-
OXI
Contu
rol
Figure
4.
Monoclonal
antibodies
against
TSP
inhibit
MO
recognition
of
aged
neutrophils
of
mAbs.
In
these
experiments,
mAbs
were
used
at
a
1:25
dilution
(the
same
concentration
optimally
employed
in
in-
hibition
of
TSP-coated
latex
bead
adhesion
to
aged
PMN).
In
this
series
of
experiments,
37.2±2.9%
of
M)
recognized
aged
PMN
under
control
conditions.
No
mAb
affected
EIgG
uptake
(closed
squares).
Thrombospondin
in
Phagocytosis
ofAged
Neutrophils
1517
11,
_
binding
at
the
"free"
surface
of
the
cell
(49,
53).
For
example,
we
have
reported
that
attachment
of
M)
to
surfaces
treated
with
Vn
but
not
fibrinogen
or
type
IV
collagen
inhibited
subse-
quent
recognition
of
aged
neutrophils,
in
keeping
with
a
role
for
MO
vitronectin
receptor
(ar,3)
in
the
interaction,
Conse-
quently,
we
examined
the
effect
of
allowing
MO
to
modulate
surface
receptors
for
TSP
by
attachment
to
tissue
culture-
treated
surfaces
coated
with
TSP.
This
maneuver
resulted
in
strong
inhibition
of
MO
phagocytosis
of
aged
PMNs
(to
19.8±3.2%
of
control,
n
=
11,
in
a
series
of
experiments
where
57.5±1.8%
of
MO
took
up
aged
PMN
in
control
conditions).
There
was
no
effect
on
ingestion
of
EIgG.
In
addition,
treat-
ment
of
the
substrate
with
the
control
protein
albumin
had
no
effect.
Therefore,
this
experiment
appeared
to
confirm
a
role
for
TSP-binding
structures
on
the
MO
in
recognition
of aged
PMNs.
This
result
would
be
expected
from
the
reported
specificity
of
a0033
for
TSP
(27).
Nevertheless,
it
was
also
compatible
with
a
role
for
other
MO
TSP
receptors,
although
our
previous
ob-
servation
that
fucoidan
failed
to
inhibit
the
interaction
indi-
cated
that
involvement
of
MO
heparan
sulfate
proteoglycan
was
unlikely
(26,
54).
However,
we
had
not
previously
sought
a
role
for
another
MO
adhesion
receptor
with
specificity
for
TSP,
the
88-kD
monomer,
CD36
(25,
27).
An
IgM
mAb
to
CD36,
SMO
(40)
inhibited
MO
recognition
of
aged
neutro-
phils,
but
not
of
EIgG
(Fig.
5
A).
A
number
of
isotype
control
mAbs,
including
mAbs
binding
to
either
cell
type,
had
no
effect
(data
for
one
mAb
shown
in
Fig.
5
A),
indicating
that
the
inhibitory
effect
of
SMO
was
highly
unlikely
to
be
caused
by
some
nonspecific
"steric"
effect
of
IgM
binding
to
cell
surfaces.
Furthermore,
an
IgG,
mAb
to
CD36,
ClMegl
(40),
also
specif-
ically
inhibited
phagocytosis
of
aged
neutrophils
(to
41.4±5.3%
of
control,
n
=
6;
no
effect
on
EIgG
uptake
was
seen).
The
inhibitory
effect
of
anti-CD36
mAbs
was
exerted
at
the
MO
surface;
since
preincubation
of
MO
but
not
of
neutro-
phils
with
CD36
mAbs
inhibited
recognition
(Fig.
5
B)
and
CD36
mAbs
did
not
bind
to
aged
neutrophils
as
assessed
by
immunofluorescence
flow
cytometry
(data
not
shown).
In
par-
allel
experiments
the
SMO
CD36
mAb
bound
specifically
to
MO
(Fig.
1).
Therefore,
these
data
indicate
a
hitherto
unex-
pected
role
for
CD36
in
the
interaction.
Synergistic
inhibition
of
the
interaction
by
mAbs
to
the
vi-
tronectin
receptor
and
CD36.
Previously,
we
reported
that
mAbs
to
both
subunits
of
macrophage
a143
specifically
inhib-
ited
MO
recognition
of
aged
neutrophils
by
>
70%
(12).
How-
ever,
in
the
current
series
of
experiments
a
similar
degree
of
inhibition
was
observed
with
the
CD36
mAb
SM4
(Fig.
5).
These
observations
are
compatible
with
the
possibility
that
MO
CD36
and
a^f,3
function
together
in
recognition
of
aged
neutro-
phils.
Thus,
we
studied
the
effects
of
a
combination
of
anti-
CD36
and
anti-av,33
mAbs,
each
at
a
concentration
causing
only
weak
inhibition
of
the
interaction
when
used
individually.
The
combination
exerted
a
strong
inhibitory
effect
on
MO
rec-
ognition
of
aged
neutrophils
which
was
synergistic
rather
than
merely
additive,
in
keeping
with
a
cooperative
role
for
the
two
receptors
in
recognition
of
the
aged
PMN
(Fig.
6).
This
effect
was
specific
because
combination
of
either
mAb
with
an
irrele-
vant
isotype
control
mAb
did
not
result
in
synergistic
inhibi-
tion
and
EIgG
recognition
was
not
affected.
Macrophage
adherence
to
TSP
is
mediated
by
a
"two-
point"
mechanism.
Taken
together,
the
data
above
suggest
that
in
promoting
recognition
of
aged
neutrophils,
TSP
may
bind
to
A
100-
0
0
u1
a
c
CS
0
c
0II
C
0'
Cr
.S
control
IgM
(CDIS
I
S
---
--
--
--6---
--
SMA
(C
361oo
11250
B
11125
MA
ingesting
1lso
1125
dilution
of
ascites
aged
pmn
(a
of
contralI
SM6
(C036.lqMl
control
1gM
(COISI
antibody
present
throughout
interaction
control
pmn
pre-incubated
with
ab
Sat6
control
M6
pre-incubated
with
ab
J
Figure
5.
(A)
Concentration-dependent
inhibition
by
CD36
mAb
SMO
on
MO
recognition
of
aged
neutrophils,
and
(B)
functional
lo-
calization
of
inhibitory
effect
to
M4
surface.
In
(A),
each
point
rep-
resents
the
mean±SE
of
12
observations
made
in
experiments
where
under
control
conditions,
39.1±2.3%
of
M4
recognized
aged
neutro-
phils.
The
highest
concentration
of
SM4
had
no
effect
on
the
propor-
tion
of
M)
taking
up
EIgG
(closed
square).
The
control
mAb
was
the
isotype
matched
IgM
CD15
mAb,
28.
The
dotted
lines
are
as
defined
in
Fig.
2.
In
(B),
preincubation
of
M)
but
not
neutrophil
with
a
1:25
dilution
of
SM4
(but
not
by
control
mAb
28)
for
30
min
at
4VC
in-
hibited
the
interaction
(open
bars:
mean±SE,
n
=
6)
to
a
degree
sim-
ilar
to
that
when
the
mAb
was
included
only
in
the
interaction
me-
dium.
Under
control
conditions,
48.9±5.2%
of
MO
recognized
aged
neutrophils.
both
a,,03
and
CD36
on
the
macrophage.
Although
inhibitor
studies
reported
by
others
indicate
that
some
cell
types
may
use
a
single
receptor
mechanism
to
adhere
to
TSP
(for
example,
an
RGD-inhibitable
integrin
[27]),
there
are
examples
of
cells
which
use
a
"two-point"
mechanism
(55).
Neither
the
tetra-
1518
J.
Savill,
N.
Hogg,
Y.
Ren,
and
C.
Haslett
100
-
Z
-
0,
-
Q
0a
0-
_E
C
_D
a
C_
-a-
ri
a
aa
a
a
medium
alone
-I
SM6
Z3C6
+23C5
F':-
SM6
23C6
+P3
+P112
Figure
6.
Synergistic
inhibition
of
M4
recognition
of
aged
neutrophils
by
combination
of
SMO
(CD36
mAb)
and
23C6
(VnR
mAb)
n
=
6.
SMO
was
used
at
a
dilution
of
1
in
100
ascites
in
HBSS,
and
23C6
at
5
ttg/ml
in
HBSS.
Each
mAb
induced
weak
but
specific
inhibition
of
aged
neutrophil
recognition
(open
bars)
and
there
was
no
effect
upon
M40
uptake
of
EIgG
(open
squares).
"SMo
+
23C6"
signifies
HBSS
in
which
mAb
SMO
was
present
at
a
final
dilution
of
1
in
100
and
23C6
was
also
present
at
a
final
concentration
of
5
gg/ml.
The
two
right
hand
bars
are
controls
for
this
combination
with
mAb
P3
at
a
final
concentration
of
5
jsg/ml
and
PI
12
ascites
at
a
final
dilution
of
1
in
100
ascites.
In
this
series
of
experiments,
64.2±1.2%
of
M)k
recognized
aged
neutrophils
under
control
conditions.
peptide
RGDS
nor
the
aVl33
mAb
23C6
inhibited
MO
binding
to
TSP
(Table
V),
although
both
inhibit
macrophage
recogni-
tion
of
aged
neutrophils.
Similarly,
mAb
to
CD36
failed
to
inhibit
MO
adhesion
to
TSP.
However,
the
combination
of
CD36
mAb
with
either
af3
mAb
or
RGDS
(but
not
RGES)
resulted
in
specific
inhibition
of
MO
to
TSP.
The
failure
of
these
inhibitors
to
block
MO
binding
to
Fn
(56)
was
used
as
a
control.
These
data
therefore
indicated
that
MO
adhesion
to
TSP-coated
surfaces
involved
cooperative
binding
by
both
CD36
and
the
RGD-inhibitable
a433
vitronectin
receptor
inte-
grin.
Table
V.
Evidence
for
Two-point
Adhesion
of
M4
via
CD36
and
af33
to
TSP-coated
Gelatin
Macrophage
binding
(expressed
as
%
of
control
for
each
adhesion)
Inhibitor(s)
present
in
medium
during
adhesion
assay
TSP-coated
gelatin
Fn-coated
gelatin
None
(Control)
100.0±2.6
100.0±1.9
SM0
(CD36
mAb)
97.0±7.4
100.2±4.7
RGDS
95.5±4.2
105.3±14.6
RGES
99.4±6.3
91.5±6.8
23C6
(avf3
mAb)
86.1±6.6
96.6±1.9
RGDS
+
SMO
17.4±4.1*
98.7±8.1
RGES
+
SMO
106.6±4.2
96.1±3.6
RGDS
+
P112
(control
mAb)
95.8±2.3
103.4±8.3
SMO
+
23C6
14.9±1.3*
98.1±3.6
SM4
+
OX7
(control
mAb)
104.7±7.3
95.8±3.7
Gelatin
alone
15.5±
1.0
Monoclonal
antibodies
were
used
as
a
1:25
dilution
of
ascites,
and
tetrapeptides
at
2.2
mM:
When
combined,
the
same
final
concentra-
tions
were
achieved.
In
these
experiments
(n
=
9),
686±51
My
ad-
hered
per
10
wells
to
TSP-coated
gelatin;
863±112
M0
adhered
per
10
wells
to
Fn-coated
gelatin.
*
P
<0.05
compared
with
TSP
bead
binding
to
aged
PMN
under
control
conditions.
Discussion
In
this
study,
we
have
examined
the
cell
surface
recognition
mechanisms
used
by
human
MO
in
phagocytosis
of
intact
ag-
ing
human
neutrophils
that
have
undergone
apoptosis,
a
po-
tentially
"injury
limiting"
mode
of
neutrophil
removal
operat-
ing
at
inflamed
sites
(8).
Our
conclusions
are
that
macrophage-
synthesized
TSP
mediates
the
interaction
by
forming
a
molecular
bridge
between
the the
cell
types.
Furthermore,
at
the
MO
surface
the
data
indicate
that
TSP
binds
to
both
CD36
and
the
a433
"vitronectin"
receptor.
There
is
evidence
that
TSP
mediates
platelet-platelet,
plate-
let-erythrocyte,
and
monocyte-platelet
adhesion
by
acting
as
a
"molecular
bridge"
between
cells
(
17,
21,
24,
25).
These
inter-
actions
may
be
potentiated
by
supplementing
available
TSP
by
addition
of
micromolar
concentrations
of
exogenous
TSP,
and
inhibited
by
higher
concentrations
which
saturate
TSP-binding
sites.
TSP-mediated
agglutination
of
erythrocytes
by
activated
platelets
has
also
been
reported
to
be
inhibited
by
millimolar
concentrations
of
aminosugars
and
basic
aminoacids
(
16,
17).
This
suggests
that
a
bridging
role
for
TSP
should
be
sought
in
MO
recognition
of
aged
PMN,
since
we
found
that
this
interac-
tion
was
also
inhibited
by
similar
concentrations
of
these
cat-
ionic
molecules
(
15).
The
data
obtained
in
this
study
support
the
possibility
that
MO-synthesized
TSP
has
a
bridging
role
in
this
phagocytic
interaction.
First,
TSP
is
present,
both
in
solu-
tion
and
associated
with
the
MO
surface.
Second,
the
interac-
tion
is
inhibited
by
reducing
supply
of
TSP
in
the
interaction
(by
treating
MO
with
cycloheximide)
and
"rescued"
by
replen-
ishing
TSP
supply.
Third,
the
interaction
is
specifically
poten-
tiated
by
TSP,
by
an
effect
which
preincubation
studies
show
can
be
exerted
at
the
surface
of
each
cell
type.
Fourth,
inhibi-
tion
of
TSP-mediated
adhesion
to
aged
PMN
by
mAbs
to
TSP
inhibits
the
interaction.
Fifth,
high
concentrations
of
soluble
TSP
specifically
inhibit,
and
sixth,
so
does
attachment
of
mac-
rophages
to
surfaces
coated
with
TSP,
a
maneuver
known
to
down-regulate
expression
of
adhesion
receptors
at
the
free
sur-
face
of
the
cell.
Finally,
supplementation
of
existing
TSP
sup-
ply
by
including
the
protein
in
the
interaction
medium
at
con-
centrations
around
5
,ug/ml
specifically
potentiated
recogni-
tion.
It
therefore
appears
reasonable
to
conclude
that
TSP
plays
a
hitherto
unexpected
bridging
role
in
MO
recognition
of
aged
neutrophils.
This
represents
an
important
new
addition
to
the
functional
repertoire
of
TSP,
and
the
first
defined
molecular
link
between
the
surfaces
of
the
apoptotic
PMN
and
the
MO.
Clearly,
to
understand
the
precise
role
played
by
TSP
in
the
interaction,
it
will
be
necessary
to
define
the
receptors
for
TSP
on
both
cell
types.
At
present,
the
mechanisms
by
which
TSP
binds
to
aged
neutrophils
are
unknown.
Our
current
and
pre-
vious
data
(
12)
indicate
that
CD36
and
Avf3
play
their
roles
at
the
surface
of
the
Mo,
since
in
both
cases
the
inhibitory
effects
of
mAbs
were
localized
to
the
Mo,
and
neither
receptor
was
detectable
on
the
surface
of
aged
PMNs
by
flow
cytometry.
Furthermore,
our
previous
finding
that
fucoidan
fails
to
inhibit
the
interaction
(
15)
suggests
that
neither
heparan-sulfate
pro-
teoglycan
nor
sulfatides
are
involved
(23,
26,
54).
Since
freshly
isolated
PMNs
are
not
recognized
by
MI
(8),
it
appears
likely
that
the
mechanisms
by
which
TSP
binds
to
fresh
PMN
(31,
32)
will
prove
to
be
different
from
those
involved
in
TSP
bind-
ing
to
aged
PMN.
Indeed,
it
is
of
interest
that
we
found
that
the
anti-TSP
mAb
A2.5
inhibited
the
adherence
of
apoptotic
PMN
to
TSP,
while
others
found
no
effect
upon
the
adherence
of
Thrombospondin
in
Phagocytosis
ofAged
Neutrophils
1519
freshly
isolated
PMN
to
TSP
(32).
Ongoing
work
on
the
alter-
ations
occurring
in
the
PMN
surface
during
apoptosis
will
ex-
amine
possible
changes
in
the
nature
of
the
adhesive
interac-
tion
with
TSP.
However,
the
current
study
advances
understanding
of
the
molecular
interactions
occurring
at
the
MO
surface
during
phagocytosis
of
apoptotic
cells.
First,
the
inhibitory
effects
of
monoclonal
antibodies
to
CD36
indicate
a
newly-identified
role
in
recognition
of
aged
PMNs
for
this
88-kD
MO
surface
molecule,
the
physiological
function
of
which
has
been
ob-
scure.
The
degree
of
inhibition
observed
(>
80%)
is
compara-
ble
to
that
reported
earlier
for
mAbs
to
aC133
(
12).
This
suggests
that
the
two
receptors
work
in
concert
rather
than
mediating
two
independent
phagocytic
pathways,
and
this
possibility
was
supported
by
synergistic
inhibition
of
the
interaction
by
combi-
nation
of
low
concentrations
of
mAbs
to
CD36
and
a,#3.
Sec-
ond,
although
there
has
been
controversy
as to
whether
either
CD36
or
a!f33
truly
represent
cellular
adhesion
receptors
for
TSP
(57-59),
our
studies
of
MO
binding
to
TSP
indicate
that
the
two
structures
participate
in
a
"two-point"
adhesion
mecha-
nism
for
TSP.
This
combination
of
TSP
receptors
is
different
from
those
previously
reported
to
mediate
two-point
adhesion
to
TSP
by
other
cell
types,
but
is
consistent
with
the
principle
established
by
these
workers
(55).
Furthermore,
interference
with
two-point
MO
adhesion
to
TSP
provides
an
explanation
for
the
individual
and
synergistic
inhibitory
effects
of
mAbs
to
CD36
and
avf3
upon
aged
PMN
ingestion.
However,
the
data
appear
to
raise
an
interesting
paradox.
Whereas
inhibition
of
aged
PMN
ingestion
occurs
after
blockade
of
either
MO
recep-
tor
for
TSP,
such
blockade
does
not
inhibit
MO
adhesion
to
TSP,
which
requires
simultaneous
blockade
of
both
CD36
and
aCf33.
A
possible
explanation
is
that
the
conformation
of
sub-
strate-bound
TSP,
as
well
as
the
substrate
itself,
may
allow
more
avid
binding
of
MO
than
the
circumstances
encountered
in
TSP-mediated
"bridging"
to
the
apoptotic
PMN.
The
result
would
be
that
disruption
of
one
receptor
class
is
insufficient
to
promote
MO
displacement
from
a
TSP-coated
substrate,
while
it
is
adequate
to
prevent
TSP-mediated
phagocytosis
of
aged
PMN.
It
can
be
further
speculated
that
a
specific
conformation
of
TSP
presented
by
MO
CD36
and
af#3
is
required for
recogni-
tion
and
ingestion
of
apoptotic
PMN.
The
candidate
TSP
epi-
topes
binding
to
aCf33
and
CD36
are
thought
to
be
located
close
to
the
carboxyl
terminus
of
each
subunit
of
the
trimeric
mole-
cule
(
19,
20,
22),
offering
possibilities
of
dual
receptor
interac-
tion
with
a
single
TSP
molecule.
Further
study
will
be
needed
to
define
the
stochiometry
and
precise
molecular
mechanisms
of
interactions
between
aC#3,,
CD36,
and
TSP
occurring
at
the
MO
surface
during
recognition
of
apoptotic
PMNs.
It
is
important
to
consider
how
the
proposed
recognition
mechanism
might
operate
at
the
inflamed
site
in
vivo.
The
potentiating
effect
of
low
concentrations
of
TSP
suggests
that
supply
of
this
protein
may
be
at
a
premium
in
our
standard
"no
added
protein"
interaction
assay,
but
there
is
evidence
that
TSP
is
in
plentiful
supply
soon
after
injury
of
a
tissue
(35),
presumably
being
secreted
by
platelets
and
a
range
of
other
cell
types.
Indeed,
the
transient
presence
of
the
molecule
in
the
extracellular
matrix
of
the
healing
skin
wound
corresponds
with
the
time
PMNs
are
being
removed
from
this
and
other
sites
of
self-limited
inflammation
(60).
Although
the
Avf3
inte-
grin
was
originally
thought
to
be
specific
for
vitronectin
(
13,
61
),
this
receptor
has
specificity
for
RGD-bearing
proteins
in
addition
to
TSP,
namely
Fn
and
fibrinogen
(27,
62-65).
Fn
has
potential
as
a
candidate
for
bridging
aged
PMN
to
MO,
since
it
may
mediate
agglutination
of
trypsinized
erythrocytes
which
can
be
inhibited
by
aminosugars
and
basic
aminoacids,
and
because
both
macrophages
and
neutrophils
may
synthe-
size
the
protein
(66).
However,
by
contrast
with
TSP,
Fn
did
not
enhance
recognition
of
aged
neutrophils
by
normal
(
12)
or
cycloheximide-inhibited
MO,
nor
did
a
mAb
to
the
cell-bind-
ing
domain
of
fibronectin
inhibit
recognition.
Similarly,
nei-
ther
Vn
nor
fibrinogen
potentiated
the
interaction
(
12).
There-
fore,
the
in
vitro
data
mitigate
against
a
bridging
role
for
ligands
of
the
avi3
other
than
TSP,
and
such
a
role
also
seems
unlikely
on
the
grounds
that
these
molecules
do
not
interact
with
CD36
(28, 57).
However,
the
data
suggest
the
speculative
possibility
that
in
addition
to
local
changes
in
pH,
or
in
concentrations
of
charged
molecules
(
15),
soluble
or
matrix-bound
Fn
or
Vn,
or
proteolytic
fragments
of
these
molecules,
may,
in
certain
cir-
cumstances,
inhibit
aged
PMN
removal
by
interfering
with
TSP
binding
to
M4
a,33,
thus
favouring
release
of
PMN
con-
tents
from
dying
cells
and
persistence
of
tissue
injury.
To
conclude,
human
monocyte-derived
macrophage
recog-
nition
of
aging
neutrophils
undergoing
apoptosis
involves
hith-
erto
unknown
functions
for
TSP
and
the
macrophage
surface
molecule
CD36.
The
data
in
this
report
indicate
that
TSP
acts
as
a
molecular
bridge
between
apoptotic
neutrophil
and
macro-
phage.
At
the
macrophage
surface,
the
evidence
suggests
that
TSP
forms
an
adhesive
complex
involving
macrophage
CD36
and
the
af33
integrin.
Further
study
of
the
mechanisms
by
which
TSP
binds
to
apoptotic
neutrophils
will
provide
a
new
approach
to
the
identification
of
the
surface
changes
that
occur
in
apoptotic
cells
to
lead
to
their
recognition
as
"senescent-
self."
This
study
also
implies
that
proteins
that
are
frequently
incorporated
into
the
extracellular
matrix
at
inflamed
sites
may
play
an
important
role
in
regulation
of
macrophage
recog-
nition
of
apoptotic
neutrophils,
a
cellular
interaction
that
has
potential
to
limit
the
toxic
potential
of
neutrophils
in
inflamma-
tion.
Acknowlednments
We
are
particularly
indebted
to
Dr.
R.
L.
Nachman
and
Dr.
R.
L.
Silverstein
(Cornell
University
Medical
College,
New
York)
for
their
gift
of
antibody
11.2
and
advice
on
preparation
of
TSP,
with
which
Dr.
Joan
Dawes
(SNBTS,
Edinburgh,
U.K.),
Dr.
Neil
Turner
and
Dr.
Ke-
vin
Davies
(Department
of
Medicine,
Royat
Postgraduate
Medical
School)
and
Ms.
Mandy
Baker
(Department
of
Clinical
Pharmacol-
ogy,
Royal
Postgraduate
Medical
School,
London)
also
gave
invalu-
able
assistance.
Dr.
Vishva
Dixit
(University
of
Michigan,
Ann
Arbor,
MI)
is
thanked
for
helpful
discussions
and
provision
of
mAbs
A2.5,
A6.
1,
C6.7,
and
D4.6;
Dr.
N.
Hunter
(Scottish
National
Blood
Trans-
fusion
Service,
Edinburgh,
U.K.)
gave
polyclonal
antibody
to
TSP.
The
continued
assistance
of
Dr.
M.
Horton
(St
Bartholomew's
Hospital,
London),
who
gave
mAb
23C6,
and
the
kind
gifts
of
mAb
ClMegl
from
Dr.
Glenn
Pilkington
(University
of
Melbourne,
Melbourne,
Australia);
and
mAb
P3
from
Dr.
P.
Morganelli
(Dartmouth
Medical
School,
Hanover
MA)
are
gratefully
acknowledged.
Dr.
Ian
Dransfield
made
helpful
comments
on
the
manuscript,
and
Dr.
W.
Bennett
(Zool-
ogy
Department,
Oxford
University)
provided
assays
for
vitronectin.
The
Medical
Research
Council
of
Great
Britain
and
the
Wellcome
Trust
provided
financial
support.
John
Savill
has
been
supported
by
a
Medical
Research
Council
(MRC)
Training
Fellowship
and
by
a
Well-
come
Trust
Senior
Research
Fellowship
in
Clinical
Science,
and
Chris-
topher
Haslett
was
a
MRC
Senior
Clinical
Fellow.
Nancy
Hogg
is
sup-
ported
by
the
Imperial
Cancer
Research
Fund.
Yi
Ren
is
supported
by
the
Wellcome
Trust.
1520
J.
Savill,
N.
Hogg,
Y.
Ren,
and
C.
Haslett
References
1.
Henson,
P.
M.,
and
R.
B.
Johnston.
1987.
Tissue
injury
in
inflammation.
Oxidants,
proteinases,
and
cationic
proteins.
J.
Clin.
Invest.
79:669-674.
2.
Malech,
H.
D.,
and
J.
I.
Gallin.
1988.
Neutrophils
in
human
diseases.
N.
Engl.
J.
Med.
317:687-694.
3.
Weiss,
S.
J.
1989.
Tissue
destruction
by
neutrophils.
N.
Engl.
J.
Med.
320:365-375.
4.
Metchnikoff,
E.
1891.
Lecture
VII.
In
Lectures
on
the
Comparative
Pathol-
ogy
of
Inflammation.
F.
A.
Starling
and
E.
H.
Starling,
translators.
Dover
Publi-
cations
Inc.,
New
York.
106.
5.
Sanui,
H.,
S-i.
Yoshida,
K.
Nomoto,
R.
Ohhara,
and
Y.
Adichi.
1982.
Peritoneal
macrophages
which
phagocytose
autologous
polymorphonuclear
leu-
kocytes
in
guinea-pigs.
I:
Induction
by
irritants
and
micro-organisms
and
inhibi-
tion
by
colchicine.
Br.
J.
Exp.
Pathol.
63:278-285.
6.
Chapes,
S.
K.,
and
S.
Haskill.
1983.
Evidence
for
granulocyte-mediated
macrophage
activation
after
C.
parvum
immunization.
Cell.
Immunol.
75:367-
377.
7.
Newman,
S.
L.,
J.
E.
Henson,
and
P.
M.
Henson.
1982.
Phagocytosis
of
senescent
neutrophils
by
human
monocyte-derived
macrophages
and
rabbit
in-
flammatory
macrophages.
J.
Exp.
Med.
156:430-442.
8.
Savill,
J.
S.,
A.
H.
Wyllie,
J.
E.
Henson,
M.
J.
Walport,
P.
M.
Henson,
and
C.
Haslett.
1989.
Macrophage
phagocytosis
of
aging
neutrophils
in
inflammation.
Programmed
cell
death
in
the
neutrophil
leads
to
recognition
by
macrophages.
J.
Clin.
Invest.
83:865-875.
9.
Wyllie,
A.
H.
1980.
Glucocorticoid-induced
thymocyte
apoptosis
is
asso-
ciated
with
endogenous
endonuclease
activation.
Nature
(Lond.).
284:555-556.
10.
Wyllie,
A.
H.,
J.
F.
R.
Kerr,
and
A.
R.
Curie.
1980.
Cell
death:
the
significance
of
apoptosis.
Int.
Rev.
Cytol.
68:251-306.
1
1.
Cohen,
J. J.
1991.
Programmed
cell
death
in
the
immune
system.
Adv.
Immunol.
50:55-85.
12.
Savill,
J.,
I.
Dransfield,
N.
Hogg,
and
C.
Haslett.
1990.
Vitronectin
recep-
tor
mediated
phagocytosis
of
cells
undergoing
apoptosis.
Nature
(Lond.).
343:170-173.
13.
Pytela,
R.,
M.
D.
Pierschbacher,
and
E.
Ruoslahti.
1985.
A
125/115-kDa
cell
surface
receptor
specific
for
vitronectin
interacts
with
the
arginine-glycine-
aspartic
acid
sequence
derived
from
fibronectin.
Proc.
Natl.
Acad.
Sci.
USA.
82:5766-5770.
14.
Krissansen,
G.
W.,
M.
J.
Elliott,
C.
M.
Lucas,
F.
C.
Stomski,
M.
C.
Berndt,
D.
F.
Cheresh,
A.
F.
Lopez,
and
G.
F.
Burns.
1990.
Identification
of
a
novel
integrin
,B
subunit
expressed
on
cultured
monocytes
(macrophages).
J.
Biol.
Chem.
265:823-830.
15.
Savill,
J.
S.,
P.
M.
Henson,
and
C.
Haslett.
1989.
Phagocytosis
of
aged
human
neutrophils
by
macrophages
is
mediated
by
a
novel
"charge
sensitive"
recognition
mechanism.
J.
Clin.
Invest.
84:1518-1527.
16.
Gartner,
T.
K.,
D.
C.
Williams,
F.
C.
Minion,
and
D.
R.
Phillips.
1978.
Thrombin-induced
platelet
aggregation
is
mediated
by
a
platelet
plasma
mem-
brane-bound
lectin.
Science
(Wash.
DC).
200:1281-1283.
17.
Jaffe,
E.
A.,
L.
L.
K.
Leung,
R.
L.
Nachman,
R.
I.
Levin,
and
D.
F.
Mosher.
1982.
Thrombospondin
is
the
endogenous
lectin
of
human
platelets.
Nature
(Lond.).
295:246-248.
18.
Lawler,
J.
W.,
H.
S.
Slater,
and
J.
E.
Coligan.
1978.
Isolation
and
character-
ization
of
a
high
molecular
weight
glycoprotein
from
human
blood
platelets.
J.
Biol.
Chem.
253:8609-8616.
19.
Lawler,
J.,
and
R.
0.
Hynes.
1986.
The
structure
of
human
thrombospon-
din,
an
adhesive
glycoprotein
with
multiple
calcium-binding
sites
and
homolgies
with
several
different
proteins.
J.
Cell
Biol.
103:1635-1648.
20.
Lawler,
J.
1986.
The
structural
and
functional
properties
of
thrombospon-
din.
Blood.
67:1197-1209.
21.
Silverstein,
R.
L.,
L.
L.
K.
Leung,
and
R.
L.
Nachman.
1986.
Thrombo-
spondin:
a
versatile
multifunctional
glycoprotein.
Arteriosclerosis.
6:245-253.
22.
Frazier,
W.
A.
1987.
Thrombospondin:
a
modular
adhesive
glycoprotein
of
platelets
and
nucleated
cells.
J.
Cell
Biol.
105:625-632.
23.
Roberts,
D.
D.,
D.
M.
Haverstick,
V.
M.
Dixit,
W.
A.
Frazier,
S.
A.
Santoro,
and
V.
Ginsburg.
1985.
The
platelet
glycoprotein
thrombospondin
binds
specifically
to
sulfated
glycolipids.
J.
Biol.
Chem.
260:9405-941
1.
24.
Silverstein,
R.
L.,
and
R.
L.
Nachman.
1987.
Thrombospondin
binds
to
monocytes-macrophages
and
mediates
platelet
monocyte
adhesion.
J.
Clin.
In-
vest.
79:867-874.
25.
Silverstein,
R.
L.,
A.
S.
Asch,
and
R.
L.
Nachman.
1989.
Glycoprotein
IV
mediates
thrombospondin-dependent
platelet-monocyte
and
platelet-U937
cell
adhesion.
J.
Clin.
Invest.
84:546-552.
26.
Sun,
X.,
D.
F.
Mosher,
and
A.
Rapraeger.
1989.
Heparan
sulfate-mediated
binding
of
epithelial
cell
surface
proteoglycan
to
thrombospondin.
J.
Biol.
Chem.
264:2885-2889.
27.
Lawler,
J.,
R.
Weinstein,
and
R.
0.
Hynes.
1988.
Cell
attachment
to
thrombospondin:
the
role
of
Arg-Gly-Asp,
calcium,
and
integrin
receptors.
J.
Cell
Biol.
107:2351-2361.
28.
Asch,
A.
S.,
J.
Barnwell,
R.
L.
Silverstein,
and
R.
L.
Nachman.
1987.
Isolation
of
the
thrombospondin
membrane
receptor.
J.
Clin.
Invest.
79:1054-
1061.
29.
Jaffe,
E.
A.,
J.
T.
Ruggiero,
and
D.
J.
Falcone.
1985.
Monocytes
and
macrophages
synthesize
and
secrete
thrombospondin.
Blood.
65:79-84.
30.
Kreis,
C.,
M.
La
Fleur,
C.
Menard,
R.
Paquin,
and
A.
D.
Beaulieu.
1989.
Thrombospondin
and
fibronectin
are
synthesized
by
neutrophils
in
human
in-
flammatory
joint
disease
and
a
rabbit
model
of
in
vivo
neutrophil
activation.
J.
Immunol.
143:1961-1968.
31.
Mansfield,
P.
J.,
L.
A.
Boxer,
and
S.
J.
Suchard.
1990.
Thrombospondin
stimulates
motility
of
human
neutrophils.
J.
Cell
Biol.
11
1:3077-3086.
32.
Suchard,
S.
J.,
M.
J.
Burton,
V.
M.
Dixit,
and
L.
A.
Boxer.
1991.
Human
neutrophil
adherence
to
thrombospondin
occurs
through
a
CD1
1
/CDI
8-inde-
pendent
mechanism.
J.
Immunol.
146:3945-3952.
33.
Riser,
B.
L.,
R.
Mitra,
D.
Perry,
V.
Dixit,
and
J.
Varani.
1989.
Monocyte
killing
of
human
squamous
epithelial
cells:
role
forthrombospondin.
CancerRes.
49:6123-6129.
34.
Varani,
J.,
L.
Stoolman,
T.
Wang,
L.
Schuger,
C.
Flippen,
M.
Dame,
K.
J.
Johnson,
R.
F.
Todd,
U.
S.
Ryan,
and
P.
A.
Ward.
1991.
Thrombospondin
production
and
thrombospondin-mediated
adhesion
in
U937
cells.
Exp.
Cell
Res.
195:177-182.
35.
Raugi,
G.
J.,
J.
E.
Olerud,
and
A.
M.
Gown.
1987.
Thrombospondin
in
early
human
wound
tissue.
J.
Invest.
Dermatol.
89:551-554.
36.
Idell,
S.,
R.
Maunder,
A.
M.
Fein,
H.
I.
Switalska,
G.
P.
Tuszynski,
J.
McLarty,
and
S.
Niewiaroski.
1989.
Platelet-specific
a-granule
proteins
and
thrombospondin
in
boronchoalveolar
lavage
in
the
adult
respiratory
distress
syn-
drome.
Chest.
96:1125-1132.
37.
Dixit,
V.
M.,
D.
M.
Haverstick,
K.
M.
O'Rourke,
S.
W.
Hennessy,
G.
A.
Grant,
S.
A.
Santoro,
and
W.
A.
Frazier.
1985.
A
monoclonal
antibody
against
human
thrombospondin
inhibits
platelet
aggregation.
Proc.
Nall.
Acad.
Sci.
USA.
82:3472-3476.
38.
Dixit,
V.
M.,
N.
J.
Galvin,
K.
M.
O'Rourke,
and
W.
A.
Frazier
1986.
Monoclonal
antibodies
that
recognize
calcium-dependent
structures
of
human
thrombospondin.
J.
Biol.
Chem.
261:1962-1968.
39.
Majack,
R.
A.,
L.
V.
Goodman,
and
V.
M.
Dixit.
1988.
Cell
surface
thrombospondin
is
functionally
essential
for
vascular
smooth
muscle
prolifera-
tion.
J.
Cell
Biol.
106:415-422.
40.
Knapp,
W.,
B.
Dorken,
W.
R.
Gilks,
E.
P.
Rieber,
R.
E.
Schmidt,
H.
Stein,
and
A.
E.
G.
Kr.
von
dem
Borne,
editors.
1989.
Leucocyte
Typing
IV.
White
Cell
Differentiation
Antigens.
Oxford
University
Press,
Eynsham,
Oxon,
England.
41.
Hogg,
N.,
S.
MacDonald,
M.
Slusarenko,
and
P.
C.
L.
Beverley.
1984.
Monoclonal
antibodies
specific
for
human
monocytes,
granulocytes
and
endothe-
lium.
Immunology.
53:753-767.
42.
Davies,
J.,
J.
Warwick,
N.
Totty,
R.
Philp,
M.
Helfrich,
and
M.
Horton.
1989.
The
osteoclast
functional
antigen,
implicated
in
the
regulation
of
bone
resorption,
is
biochemically
related
to
the
vitronectin
receptor.
J.
Cell
Biol.
109:1817-1826.
43.
Morganelli,
P.
M.,
and
P.
M.
Guyre.
1988.
IFN--y
plusglucorticoids
stimu-
late
expression
of
a
newly-identified
human
mononuclear
phagocyte
antigen.
J.
Immunol.
140:2296-2304.
44.
Bai,
Y.,
H.
Durbin,
and
N.
Hogg.
1984.
Monoclonal
antibodies
specific
for
platelet
glycoproteins
react
with
human
monocytes.
Blood.
64:139-146.
45.
Musson,
R.
A.
1983.
Human
serum
induces
maturation
of
human
mono-
cytes
in
vitro.
Am.
J.
Pathol.
11
1:331-340.
46.
Doherty,
D.
E.,
C.
Haslett,
M.
G.
Tonnesen,
and
P.
M.
Henson.
1987.
Human
monocyte
adherence:
a
primary
effect
of
chemotactic
factors
on
the
monocyte
to
stimulate
adherence
to
human
endothelium.
J.
Immunol.
136:1762-1771.
47.
Bevilacqua,
M.
P.,
D.
Armai,
M.
W.
Mosseson,
and
C.
Bianco.
1981.
Receptors
for
cold-insoluble
globulin
(plasma
fibronectin)
on
human
mono-
cytes.
J.
Exp.
Med.
153:42-60.
48.
Wright,
S.
D.,
and
B.
C.
Meyer.
1985.
Fibronectin
receptor
of
human
macrophages
recognizes
the
sequence
Arg-Gly-Asp-Ser.
J.
Exp.
Med.
162:762-
767.
49.
Wright,
S.
D.
1986.
Methodsforthestudyofreceptor-mediatedphagocyto-
sis.
Methods
Enzymol.
132:204-221.
50.
Grinnell,
F.
1980.
Fibroblast
receptor
for
cell-substratum
adhesion:
stud-
ies
on
the
interaction
of
baby
hamster
kidney
cells
with
latex
beads
coated
by
cold
insoluble
globulin
(plasma
fibronectin).
J.
Cell
Biol.
86:104-112.
51.
Haslett,
C.,
L.
A.
Guthrie,
M.
M.
Kopaniak,
R.
B.
Johnston,
and
P.
M.
Henson.
1985.
Modulation
of
multiple
neutrophil
functions
by
preparative
methods
or
trace
concentrations
of
bacterial
lipopolysaccharide.
Am.
J.
Pathol.
119:101-110.
52.
Pierschbacher,
M.
D.,
E.
G.
Hayman,
and
E.
Ruoslahti. 1981.
Location
of
the
cell-attachment
site
in
fibronectin
with
monoclonal
antibodies
and
proteo-
lytic
fragments
of
the
molecule.
Cell.
26:259-267.
53.
Michl,
J.,
M.
M.
Pieczonka,
J.
C.
Unkeless,
and
S.
C.
Silverstein.
1979.
Effects
of
immobilized
immune
complexes
on
Fc
and
complement
receptor
func-
tion
in
resident
and
thioglycollate-elicited
mouse
peritoneal
macrophages.
J.
Exp.
Med.
150:607-621.
54.
Murphy-Ullrich,
J.
E.,
and
D.
F.
Mosher.
1987.
Interaction
of
Thrombo-
Thrombospondin
in
Phagocytosis
ofAged
Neutrophils
1521
spondin
with
endothelial
cells:
receptor-mediated
binding
and
degradation.
J.
Cell
Biol.
105:1603-1611.
55.
Asch,
A.
S.,
J.
Tepler,
S.
Silbinger,
and
R.
L.
Nachman.
1991.
Cellular
attachment
to
thrombospondin.
Co-operative
interactions
between
receptor
sys-
tems.
J.
Biol.
Chem.
266:1740-1745.
56.
Ruoslahti,
E.
1988.
Fibronectin
and
its
receptors.
Annu.
Rev.
Biochem.
57:375-413.
57.
McGregor,
J.
L.,
B.
Catimel,
S.
Parmentier,
P.
Clezardin,
M.
Dechavanne,
and
L.
L.
K.
Leung.
1989.
Rapid
purification
and
partial
characterization
of
human
platelet
glycoprotein
IIb.
J.
Biol.
Chem.
264:501-506.
58.
Sun,
X.,
and
D.
F.
Mosher.
1991.
Ca2"-sensitivebindingofthrombospon-
din
to
U937
cells
is
due
to
the
formation
of
calcium
precipitate
in
the
binding
medium.
J.
Clin.
Invest.
87:171-176.
59.
Oquendo,
P.,
E.
Hundt,
J.
Lawler,
and
B.
Seed.
1989.
CD36
directly
mediates
cytoadherence
of
plasmodium
falciparum-parasitized
erythrocytes.
Cell.
58:95-101.
60.
Haslett,
C.,
and
P,
M.
Henson.
1988.
Resolution
of
inflammation.
In
The
Molecular
and
Cellular
Biology
of
Wound
Repair.
R.
A.
F.
Clark
and
P.
M.
Henson,
editors.
Plenum
Publishing
Corp.,
New
York.
185-21
1.
61.
Pytela,
R.,
M.
D.
Pierschbacher,
M.
H.
Ginsberg,
E.
F.
Plow,
and
E.
Ruoslahti.
1986.
Platelet
Membrane
glycoprotein
IIbIIIa:
member
of
a
family
of
Arg-Gly-Asp-specific
adhesion
receptors.
Science
(Wash.
DC).
232:1559-1561.
62.
Conforti,
G.,
A.
Zanetti,
P.-R.
Ivonne,
D.
Quaglino,
P.
Neyroz,
and
E.
Dejana.
1990.
Modulation
of
vitronectin
receptor
binding
by
membrane
lipid
composition.
J.
Biol.
Chem.
265:4011
4019.
63.
Charo,
I.
F.,
L.
S.
Bekeart,
and
D.
R.
Phillips.
1987.
Platelet
Glycoprotein
IlbIlIa-like
proteins
mediate
endothelial
cell
attachment
to
adhesive
proteins
and
the
extracellular
matrix.
J.
Biol
Chem.
262:9935-9938.
64.
Cheresh,
D.
A.
1987.
Human
endothelial
cells
synthesize
and
express
an
Arg-Gly-Asp-directed
adhesion
receptor
involved
in
attachment
to
fibrinogen
and
von
Willebrand
factor.
Proc.
Natl.
Acad.
Sci.
USA.
84:6471-6475.
65.
Charo,
I.
F.,
L.
Nannizzi,
J.
W.
Smith,
and
D.
A.
Cheresh.
1990.
The
Vitronectin
receptor
a,83
binds
fibronectin
and
acts
in
concert
with
a581
in
pro-
moting
cellular
attachment
and
spreading
on
fibronectin.
J.
Cell
Biol.
11
1:2795-
2800.
66.
Yamada,
K.
M.,
S.
S.
Yamada,
and
I.
Pastan.
1975.
The
majorcell
surface
glycoprotein
of
chick
embryo
fibroblasts
is
an
agglutinin.
Proc.
Nat!.
Acad.
Sci.
USA.
72:3158-3162.
1522
J.
Savill,
N.
Hogg,
Y.
Ren,
and
C.
Haslett
