Role of TNF priming and adhesion molecules in neutrophil recruitment to intravascular immune complexes.

Abstract

Neutrophils play an important role in immune complex (IC)-mediated diseases, but the mechanisms underlying their recruitment to sites of IC deposition remain largely undefined. Furthermore, neutrophils encounter cytokines that prime their effector functions, yet the physiological relevance of priming to neutrophil functions is unclear. Using intravital microscopy, we demonstrate that TNF treatment of neutrophils ex vivo significantly increased their adhesion in a model of intravascular ICs deposited in the cremaster muscle. Notably, TNF priming had no effect on neutrophil adhesion in the absence of ICs. Analyses of relevant knockout mice and neutrophil reconstitution revealed a critical role for FcgammaRs and the CD18 integrin Mac-1 in IC-mediated neutrophil adhesion. Furthermore, ICAM-1, a major Mac-1 ligand constitutively expressed on unactivated endothelium, significantly contributed to this process. These data suggest that TNF priming promotes FcgammaR interaction with intravascular ICs, leading to the binding of Mac-1 to ICAM-1 and subsequent neutrophil arrest.

Full text

Page 1

Role of TNF priming and adhesion molecules in neutrophil
recruitment to intravascular immune complexes
Michael Lauterbach, Peter O’Donnell, Kenichi Asano, and Tanya N. Mayadas1
Center of Excellence in Vascular Biology, Department of Pathology, Brigham and Women’s Hospital and Harvard
Medical School, Boston, Massachusetts, USA
Abstract:
immune complex (IC)-mediated diseases, but the
mechanisms underlying their recruitment to sites
of IC deposition remain largely undefined. Fur-
thermore, neutrophils encounter cytokines that
prime their effector functions, yet the physiologi-
cal relevance of priming to neutrophil functions is
unclear. Using intravital microscopy, we demon-
strate that TNF treatment of neutrophils ex vivo
significantly increased their adhesion in a model of
intravascular ICs deposited in the cremaster mus-
cle. Notably, TNF priming had no effect on neu-
trophil adhesion in the absence of ICs. Analyses of
relevant knockout mice and neutrophil reconstitu-
tion revealed a critical role for Fc?Rs and the
CD18 integrin Mac-1 in IC-mediated neutrophil
adhesion. Furthermore, ICAM-1, a major Mac-1
ligand constitutively expressed on unactivated en-
dothelium, significantly contributed to this pro-
cess. These data suggest that TNF priming pro-
motes Fc?R interaction with intravascular ICs,
leading to the binding of Mac-1 to ICAM-1 and
subsequent neutrophil arrest. J. Leukoc. Biol. 83:
1423–1430; 2008.
Neutrophils play an important role in
Key Words: adhesion molecules ? Fc receptors ? cytokines ? cell
trafficking
INTRODUCTION
Neutrophils are critical immune effector cells that cause tissue
damage in immune complex (IC)-mediated diseases, such as
immune-mediated glomerulonephritis, arthritis, and other au-
toimmune disorders. ICs form repeatedly in the blood in re-
sponse to foreign or self-antigens, tissue injury, or infection.
However, excessive accumulation of ICs within the vasculature
and surrounding tissue underlies the pathogenesis of a variety
of human diseases [1]. IC deposition in tissue is associated
with neutrophil accumulation, and neutrophil adhesion to ICs
triggers robust reactive oxygen species generation, primary
granule release, and generation of cytokine and lipid media-
tors, which can contribute to tissue inflammation [1, 2]. Thus,
a tight regulation of neutrophil recruitment at sites of IC
deposition is required to limit IgG-mediated tissue injury.
Leukocyte adhesion receptors on neutrophils and the endo-
thelium, such as the selectins, ?2 integrins, and members of
the IgG superfamily (ICAM family members), have been im-
plicated in neutrophil recruitment during an inflammatory re-
sponse. However, an understanding of the molecular require-
ments for neutrophil recruitment in the context of IC deposition
is still in its infancy. Fc?Rs, receptors for IgG-containing ICs,
contribute to the pathogenesis of several immune-mediated
diseases in mice, including nephrotoxic nephritis, lupus ne-
phritis, autoimmune skin diseases, and arthritis [2, 3]. In all of
these models, Fc? deficiency is associated with a reduction in
neutrophil accumulation. Neutrophil recruitment in these mod-
els may result from the engagement of ICs by Fc?R on mast
cells and macrophages, which leads to the release of endothe-
lial-activating agonists chemokines and cytokines. Subsequent
endothelial cell activation, IC-mediated activation of comple-
ment, and subsequent generation of anaphylatoxins C3a and
C5a may mobilize neutrophils, and/or Fc?Rs on neutrophils
may directly facilitate recruitment to deposited ICs [1, 4, 5].
Mice deficient in the leukocyte-specific CD18 integrin Mac-1
are protected from IgG-mediated diseases such as acute anti-
glomerular basement membrane nephritis and bullous pemphi-
goid [6, 7]. Mac-1 deficiency in these Fc?R-dependent models
is also associated with reduced neutrophil accumulation [6, 7].
This may be attributed to compromised Fc?R-mediated adhe-
sion, as some Fc?R functions rely on Mac-1-dependent signal
transduction [7–9]. It is also possible that the absence of Mac-1
interaction with its ligand complement fragment C3bi may limit
neutrophil cytotoxicity and thus, curb subsequent neutrophil
accumulation, and/or the absence of Mac-1 interaction with
ICAM-1 on the activated endothelium may mediate neutrophil
recruitment. Thus, although Fc?R or Mac-1 deficiency results
in reductions in neutrophil accumulation in various IgG-me-
diated models of disease, the mechanisms responsible for
neutrophil recruitment remain largely undefined.
Cytokines such as TNF, elevated in autoimmune diseases
[10], not only activate the endothelium but also have the
potential to “prime” neutrophil responses. The definition of
priming is that it itself does not result in the desired response,
but the process significantly amplifies neutrophil responsive-
1Correspondence: Center of Excellence in Vascular Biology, Department of
Pathology, Brigham and Women’s Hospital and Harvard Medical School, 77
Avenue Louis Pasteur, New Research Building 7520, Boston, MA 02115,
USA. E-mail: tmayadas@rics.bwh.harvard.edu
Received June 21, 2007; revised February 19, 2008; accepted February 20,
2008.
doi: 10.1189/jlb.0607421
0741-5400/08/0083-1423 © Society for Leukocyte Biology
Journal of Leukocyte Biology
Volume 83, June 2008
1423

Page 2

ness to subsequent external stimuli. This may serve to augment
the inflammatory response [11–13]. Priming of neutrophils,
which occurs within seconds/minutes, likely precedes cyto-
kine-mediated activation of the endothelium, which principally
results from the transcriptional up-regulation of leukocyte ad-
hesion receptors and cytokines/chemokines, a process that
occurs over hours. Priming likely represents a mechanism to
localize neutrophil effector functions to sites of inflammation,
thus limiting the damaging potential of neutrophils. The failure
to limit cell priming may contribute to disease pathogenesis,
yet definitive evidence that priming enhances neutrophil re-
sponses in vivo remains an elusive goal in the field.
Here, we examined the contribution of neutrophil priming to
neutrophil recruitment to intravascularly deposited ICs in vivo
and evaluated the potential contributions of Fc?Rs and Mac-1
on neutrophils as well as the endothelial ligand for Mac-1,
ICAM-1, to this process. Our experimental approach included
the analysis of knockout mouse strains and neutrophil recon-
stitution in an in vivo model of IC deposition in the exteriorized
cremaster microcirculation that is amenable to intravital mi-
croscopy (IVM).
MATERIALS AND METHODS
Mice
Six- to 12-week-old mice were used for the experiments. C57Bl/6 mice were
used as controls for knockout mice. Fc? chain?/?on a C57Bl/6 strain
(backcrossed 12 generations) was from Taconic (Germantown, NY, USA), and
ICAM-1?/?on a C57Bl/6 strain (backcrossed 10 generations) was from The
Jackson Laboratory (Bar Harbor, ME, USA). Mac-1?/?[14], backcrossed eight
generations to C57Bl/6, was bred and maintained in our animal facility. All
mice were housed in a virus antibody-free facility at Longwood Medical
Research Center, Brigham and Women’s Hospital (Boston, MA, USA). The
Institutional Animal Care and Use Committee approved the experimental
protocols used in this study.
Preparation and administration of preformed,
soluble ICs (sICs)
sICs were made using BSA/polyclonal rabbit anti-BSA (Sigma-Aldrich, St.
Louis, MO, USA) as described previously [5] and injected into the cannulated
femoral artery [15]. IC deposition was verified as described previously [5].
Preparation and injection of bone marrow-
derived neutrophils (BMN)
BMN were isolated from donor animals using three-layer Percoll-gradient
centrifugation (75%, 65%, 55%) [16], pooled, and counted. Isolated neutro-
phils were stained using the fluorophore 5-chloromethylfluorescein diacetate
(CMFDA; Celltracker, Carlsbad, CA, USA), according to the manufacturer’s
protocol. After staining, 2.5 ? 107cells were washed, treated with TNF (200
ng/ml recombinant murine TNF-? for 15 min), washed again, and injected into
the cannulated femoral artery [15] after sIC injection.
Flow cytometry of isolated BMN
Cells (2?106) for each sample were washed with PBS, then TNF-treated in 100
?L PBS, or left untreated. After 15 min, the cells were washed with PBS-0.1%
BSA and labeled with CD11b-allophycocyanin (APC) antibody (eBioscience,
San Diego, CA, USA), CD32/CD16-FITC antibody (BD PharMingen, San
Diego, CA, USA), or matching isotypes. A second set of cells was similarly
prepared but stained with propidium iodide (BD PharMingen) and Gr-1-FITC
antibody (BD PharMingen). The samples were analyzed with a BD flow
cytometer (BD FACSCalibur flow cytometer, four-color).
To assess potential differences in Mac-1 surface expression between pe-
ripheral blood neutrophils (PBN) and BMN, 2.5 ? 107CMFDA-stained BMN
were injected into wild-type (WT) mice, and their peripheral blood was
collected 45 min after injection of BMN. Following lysis of RBCs with
repetitive 4°C H2O washes, cells were stained with Gr-1-PE (BD PharMingen)
and CD11b-APC (eBioscience) or appropriate isotype controls.
IVM
Leukocyte recruitment in cremaster muscle venules was evaluated in mice
within 45 min of a single i.v. injection of unlabeled ICs or controls. Mice (n?5
per group) were anesthetized with ketamine (90 mg/kg), xylazine (18 mg/kg),
and atropine (0.24 mg/kg). A femoral artery line was placed, which took 15
min. sICs were injected through the femoral line prior to cremaster surgery.
The cremaster was prepared as described previously [14] and took 15–20 min.
A reflecting mirror was positioned underneath the cremaster for reflective light
oblique transillumination (RLOT) [17]. The mouse was transferred to an
upright saline immersion intravital microscope (Mikron Instruments, Glendale,
CA, USA). In experiments with neutrophil reconstitution, BMN were injected
through the femoral line and allowed to circulate and adapt for 15 min before
recording. Fluorescent signals were generated by a xenon lamp, video-syn-
chronized with a stroboscope (model 315-T, Colorado video), and detected
using a silicon-intensified camera (Hamamatsu Photonics, Japan). Signals were
transferred through an on-line image processor (Argus 20, Hamamatsu Pho-
tonics, Japan). Transilluminated light (RLOT) was captured by a charged-
coupled device camera (Sony, Japan). Video signals from both cameras were
digitized (Moviebox DV, Pinnacle Systems, Inc., Mountain View, CA, USA)
and stored on a standard personal computer. The stored video files were then
analyzed using ImageJ [National Institutes of Health (NIH), Bethesda, MD,
USA] with Quicktime plug-in.
Four to five venules were analyzed over a 20-min time period. During the
experiment, the cremaster muscle was immersed with warmed, normal saline
solution (0.9%). The use of normal saline instead of lactated Ringer’s solution
or Krebs Henseleit buffer had no detectable effect on leukocyte behavior (data
not shown). At completion of the IVM experiment, blood was sampled from the
retro-orbital plexus and collected into EDTA containing Eppendorf tubes to
obtain total and differential leukocyte counts.
Peripheral leukocyte counts
Total leukocyte counts were determined with a Coulter counter (Coulter
Electronics, Fullerton, CA, USA). Differential counts (100 cells) were per-
formed on Wright-Giemsa’s stained blood smears.
Analysis of intravital microscopy data
Leukocyte rolling flux and BMN rolling flux were defined as the number of
leukocytes rolling past a perpendicular point in a vessel over 60 s. These were
converted to the leukocyte rolling flux fraction (of total leukocyte flux) and
BMN rolling flux fraction, as described [18]. Rolling cell flux fraction repre-
sents the percentage of rolling cells of the total flux (calculated from the
systemic total leukocyte count) of leukocytes passing through the vessel. BMN
rolling flux fraction of local BMN flux describes the percentage of rolling BMN
to noninteracting BMN in the vessels of interest. The mean blood flow velocity
(Vmean) was calculated as centerline velocity/1.6, and wall shear rate was
calculated as 8 ? (Vmean/venule diameter) [19, 20]. Leukocyte rolling veloc-
ities were measured by tracking single leukocytes (10/venule) over several
frames and calculating distance moved per unit time (?m/s). Adherent leuko-
cytes were defined as cells remaining stationary for 30 s and were expressed
per mm2venule.
Statistical analysis
IVM data were presented as means ? SEM or median ? SEM if nonparametric.
Statistical analysis was performed with Sigma Stat (SPSS Science Inc., Chi-
cago, IL, USA). Statistical significance of differences between groups was
tested with a Student’s t-test or a one-way ANOVA where appropriate. A
multiple-comparison procedure was performed using the Holm-Sidak method.
Statistical significance was accepted at P ? 0.05.
1424Journal of Leukocyte Biology
Volume 83, June 2008
http://www.jleukbio.org

Page 3

RESULTS
TNF priming of neutrophils increases their
adhesion in the context of intravascular IC
deposits
Our previous work established a model of intravascular IC
deposition that is amenable to IVM [5]. In this model, sICs
(BSA, rabbit anti-BSA), injected i.v., deposited locally in the
exteriorized cremaster microcirculation and promoted slow leu-
kocyte rolling on P-selectin, and increased adhesion to ICs [5].
Our first objective was to determine whether TNF priming of
neutrophils enhanced their interaction with venules in which
ICs were deposited. Neutrophils were isolated from the bone
marrow compartment of WT mice, fluorophore CMFDA-la-
beled, and primed ex vivo with vehicle or TNF. Bone marrow-
derived neutrophils (BMN), 95% pure (assessed by flow cy-
tometry using Gr-1 staining) and 97% viable (assessed by
propidium iodide exclusion), expressed Mac-1 and Fc?Rs as
assessed by flow cytometry. Treatment with murine TNF for 15
min doubled the surface expression of Mac-1 (Fig. 1A) but did
not alter Fc?R expression levels compared with vehicle-treated
BMN (Fig. 1A). The lowest TNF dose to achieve the maximum
Fig. 1. Effect of TNF priming on neutrophil Fc?R and Mac-1 expression and IC-induced adhesion in vivo. (A) Flow cytometric analysis of Fc?RII/III (left panel)
and Mac-1 (CD11b) expression (right panel) in PBN (dotted line), BMN (gray line), and 200 ng/mL TNF-treated BMN (black line). (B) Mac-1 (CD11b) surface
expression in wild-type BMN following stimulation with the indicated doses of TNF (ng/ml) is shown (left panel). The lowest TNF concentration to achieve maximum
Mac-1 expression was 200 ng/mL. (Right panel) Mac-1 expression, before (–) and after (?) 200 ng/ml TNF treatment, was evaluated in Fc? deficient (?/?) and
in Mac-1-deficient neutrophils, which served as a control for nonspecific staining of the antibody. Histograms show expression levels after gated acquisition, with
the gate being set to side-scatter/forward-scatter characteristics of neutrophils combined with a high Gr-1 signal. (C) Analysis of endogenous leukocytes and
fluorophore-labeled, TNF-primed BMN by IVM. Mice given sICs or PBS alone (sIC, –/?), were i.v.-injected with fluorophore-labeled, naı ¨ve or TNF-treated BMN
(BMN/TNF, –/?), and the rolling velocity and adhesion of the endogenous, unlabeled leukocytes (left two panels) and fluorophore-labeled, adoptively transferred
BMN (right two panels) were evaluated. Adherent cells are expressed as cells per mm2of the endothelial surface assuming cylindrical geometry of the vessel. *,
P ? 0.05; n ? 5 per group. (Bottom) Representative pictures of a cremaster postcapillary venule of a WT recipient i.v. given TNF-primed BMN in bright-field
illumination (left panel), stroboscopic FITC illumination (middle panel), and a merged picture of bright-field and FITC illumination (right panel) are shown. Black
arrowheads identify endogenous leukocytes, and white arrowheads point to fluorophore-labeled BMN.
Lauterbach et al.
TNF/Mac-1 in immune complex-induced neutrophil recruitment 1425

Page 4

surface expression of Mac-1 on BMN was 200 ng/ml (Fig. 1B).
This TNF dose was henceforth used to prime BMN prior to
their injection into recipient mice as described below. BMN
from Fc??/?mice exhibited Mac-1 surface expression that was
similar to WT mice and likewise, inducible with TNF treatment
(Fig. 1B). As expected, Mac-1 expression was undetectable in
BMN from Mac-1?/?mice (Fig. 1B).
Vehicle- or TNF-treated, fluorophore-tagged BMN were in-
troduced through a femoral line into WT mice that had been
given sIC i.v., followed by preparation of the cremaster for
IVM. TNF-treated BMN were washed prior to introducing them
into the animals (described in Materials and Methods); thus,
TNF itself was not introduced into the circulation. Neutrophil
rolling and adhesion of endogenous leukocytes and injected
fluorophore-labeled BMN were evaluated in the same animal
(Fig. 1C). Endogenous leukocytes exhibited no change in the
number of cells rolling, but displayed a decrease in leukocyte
rolling velocity, and an increase in adhesion in response to
deposited sICs (Fig. 1C) as described previously [5]. Similarly,
deposited sIC did not increase the number of rolling BMN (not
shown) but reduced the leukocyte rolling velocity, which was
particularly evident when the BMN were TNF-primed. In-
creased adhesion of vehicle-stimulated BMN was observed in
mice receiving sICs compared with control mice. However, the
number of BMN adherent to ICs was significantly fewer com-
pared with the endogenous neutrophils, suggesting that BMN
were less responsive than PBN to IC stimulation. This is
consistent with a published report [21] and our own unpub-
lished observations that PBN are more reactive than BM coun-
terparts to activating stimuli. BMN pretreated with TNF had
significantly increased adhesion only in the presence of ICs.
Importantly, no increase in adhesion of TNF-primed BMN was
observed in the absence of deposited sICs (Fig. 1C), which
indicated that TNF priming enhanced neutrophil adhesion
selectively in the context of intravascular IC deposition.
Neutrophil recruitment to intravascular ICs is
significantly reduced in primed Fc??/?BMN
We have shown previously that Fc?R deficiency in mice ab-
rogates IC-induced, slow rolling on P-selectin and subsequent
adhesion [5]. Here, we directly examined the role of activating
Fc?Rs on neutrophils in recruitment of TNF-primed neutro-
phils. Fc? chain-deficient mice (Fc??/?) lack all activating
Fc?Rs but retain expression of the inhibitory Fc?RIIB. Con-
sistent with this, flow cytometry analysis of Fc??/?BMN using
an antibody that recognizes all Fc?Rs revealed a 60% reduc-
tion in Fc?R expression. The remaining fluorescence intensity
was the result of retained Fc?RIIB expression (data not
shown), as previously reported [22]. BMN isolated from WT
and Fc??/?mice were labeled, TNF-treated, and injected into
WT recipients given sIC. The rolling BMN flux fraction was
similar in mice reconstituted with WT or Fc??/?BMN. The
slow rolling velocity characteristically observed in the pres-
ence of ICs was similar in TNF-treated Fc??/?BMN and WT
BMN, which may be attributed to the retained Fc?RIIB ex-
pression in Fc? chain?/?BMN (Fig. 2); eliminating activating
Fc?Rs and Fc?RIIB voids the slow rolling in response to
intravascular sIC [5]. We also cannot rule out the possibility
that TNF-treated Fc?R?/?BMN may have alternate mecha-
nisms for promoting slow rolling compared with PBN. Impor-
tantly, firm adhesion of TNF-treated Fc??/?BMN was severely
altered compared with similarly treated WT BMN (Fig. 2).
Endogenous neutrophils in the WT hosts reconstituted with
Fc??/?BMN efficiently adhered to the vessel wall of mice
(data not shown), which demonstrates that IC deposition was
adequate for promoting BMN adhesion. Thus, Fc?Rs on neu-
trophils play an important role in the interaction of TNF-
primed neutrophils with the vessel wall.
Neutrophil recruitment to intravascular ICs is
reduced significantly in Mac-1?/?and ICAM-1?/?
mice
Next, we assessed the contribution of Mac-1 to neutrophil
behavior in the context of intravascular IC deposition. First,
studies were conducted in WT and Mac-1-deficient mice (Mac-
1?/?) following sIC injections. Leukocyte rolling velocity sig-
nificantly increased in Mac-1?/?compared with WT mice,
suggesting a role for MAC-1 in IC-induced slow rolling. Fur-
thermore, firm adhesion was decreased significantly in Mac-
1?/?mice. Rolling cell flux fraction was comparable in WT
and Mac-1?/?animals (Fig. 3).
To explore the possibility that the Mac-1-dependent in-
crease in neutrophil adhesion requires its interaction with its
endothelial ligand ICAM-1, we evaluated mice deficient in
ICAM-1 (ICAM-1?/?). ICAM-1?/?mice showed an increase
in total leukocyte, neutrophil, and lymphocyte counts (data not
Fig. 2. Contribution of neutrophil Fc?R to
IC-induced neutrophil adhesion. Fluoro-
phore-labeled, TNF-treated
?/?) or Fc? chain-deficient (Fc?, ?/?)
BMN were injected into WT recipients
given sIC. The left panel shows the rolling
velocity of injected BMN; the middle panel,
BMN rolling flux fraction; and the right
panel, the number of adherent BMN. *, P ?
0.05; n ? 5 per group.
WT(Fc?,
1426Journal of Leukocyte Biology
Volume 83, June 2008
http://www.jleukbio.org

Page 5

shown), as previously reported [23]. ICAM-1 deficiency had no
significant effect on IC-induced slow leukocyte rolling velocity
(Fig. 4). However, firm adhesion of endogenous leukocytes was
reduced by 60% in ICAM-1?/?mice compared with WT mice
counterparts.
Evaluation of the role of neutrophil Mac-1 and
endothelial cell ICAM-1 in IC-induced adhesion
of TNF-primed neutrophils
ICAM-1 is present on endothelial cells as well as circulating
leukocytes. To determine the contribution of endothelial
ICAM-1 to the phenotype in ICAM-1-deficient mice, we re-
constituted ICAM-1-deficient mice with naive or TNF-primed
WT neutrophils. Results obtained upon reconstitution of WT
recipients with WT neutrophils from Figure 1 are shown for
comparison (Fig. 5A). IC-induced BMN rolling velocity and flux
fraction were not significantly different between groups. Adhesion
of naı ¨ve WT neutrophils to WT and ICAM-1 recipient mice was
comparable. However, after TNF priming, significantly enhanced
BMN adhesion was observed in WT compared with ICAM-1?/?
recipients (Fig. 5A). Thus, endothelial ICAM-1 plays a role in
IC-induced firm adhesion of TNF-primed neutrophils.
To examine the contribution of Mac-1 on neutrophils in
TNF-primed cell adhesion to ICs, BMN from Mac-1?/?mice
were vehicle- or TNF-treated and injected into WT recipients
given sICs. A trend toward an increase in rolling velocity in
Mac-1?/?BMN was observed, although this did not approach
statistical significance. Importantly, adhesion of Mac-1?/?
BMN was reduced significantly compared with WT BMN (Fig.
5B), a reduction that was comparable with that observed following
Fc??/?BMN reconstitution of WT mice (Fig. 2). Thus, the study
of endogenous Mac-1?/?neutrophils and TNF-treated Mac-1?/?
BMN suggests that this ?2integrin on neutrophils plays an im-
portant role in regulating IC-induced neutrophil adhesion.
Next, we specifically addressed whether endothelial ICAM-1
interactions with Mac-1 are contributing to the observed IC-
mediated responses. TNF-primed WT or Mac-1?/?BMN were
injected into ICAM-1?/?recipient mice given sICs, and leu-
kocyte rolling velocity, flux fraction, and adhesion were mea-
sured. Adhesion of WT neutrophils in ICAM-1?/?recipients
was similar to that observed in the group of animals analyzed
in Figure 5A as expected. A further, significant decrease in
firm adhesion was observed in ICAM-1?/?recipients receiving
Mac-1?/?BMN (Fig. 5C). Indeed, ICAM-1?/?recipient mice
given Mac-1-deficient BMN exhibit a greater decrease in ad-
hesion (73%) compared with those given WT BMN (42%) or
when WT recipient mice are given Mac-1-deficient BMN
(53%). This suggests that the adhesion observed following IC
deposition is not exclusively the result of Mac-1 interaction
with ICAM-1 but includes the interaction of these receptors
with other ligands.
DISCUSSION
A key step in the initiation and progression of many IC-
mediated diseases is the recruitment of neutrophils to depos-
ited ICs [1]. Delineating the requirements for IC-induced neu-
Fig. 3. Analysis of Mac-1-deficient mice
given sICs. Mac-1 deficient (?/?) mice and
WT counterparts (?/?) were injected with
sICs and prepared for IVM. Leukocyte roll-
ing velocity, rolling cell flux fraction, and
the number of adherent cells are shown. *,
P ? 0.05; n ? 5 per group.
Fig. 4. IC-induced leukocyte recruitment
is reduced in ICAM-1-deficient mice (?/?),
which were analyzed following injection of
sIC. For purposes of comparison, the data
for WT mice are presented from Figure 3.
Leukocyte rolling velocity, rolling cell flux
fraction, and adherent cells are depicted. *,
P ? 0.05; n ? 5 per group.
Lauterbach et al.
TNF/Mac-1 in immune complex-induced neutrophil recruitment1427

Page 6

Fig. 5. Analysis of the contribution of endothelial ICAM-1 and neutrophil Mac-1 to IC-induced adhesion of TNF-primed neutrophils. (A) Vehicle- or TNF-treated
BMN were injected into ICAM-1-deficient recipient (REC) mice (ICAM-1Rec, ?/?) given sICs. For purposes of comparison, the dataset for the WT group (referred
to here as ICAM-1Rec, ?/?) is duplicated from Figure 1B. *, P ? 0.05; n ? 5 per group. (B) WT (WTRec, ?/?) mice injected with sICs were prepared for IVM,
and TNF-primed, fluorophore-labeled WT (Mac-1, ?/?) or Mac-1-deficient (Mac-1, ?/?) BMN were delivered i.v. *, P ? 0.05; n ? 5 per group. (C) ICAM-1?/?
mice (ICAM-1Rec, ?/?) were injected with sICs and labeled TNF-primed Mac-1?/?or Mac-1?/?BMN. *, P ? 0.05; n ? 5 per group.
1428Journal of Leukocyte Biology
Volume 83, June 2008
http://www.jleukbio.org

Page 7

trophil recruitment is critical for understanding the mecha-
nism(s)of leukocyteaccumulation
inflammatory and autoimmune diseases. Our previous study
demonstrated that complement C1q-dependent, intravascular
IC deposition in the cremaster leads to slow leukocyte rolling
and increased adhesion and that a deficiency in Fc?Rs in mice
abrogates these processes [5]. Here, our study used knockout
mice and neutrophil reconstitution approaches to show that
TNF priming significantly increased neutrophil adhesion only
in the context of intravascular ICs and demonstrates a critical
role for neutrophil-activating Fc?Rs and a CD18 integrin,
Mac-1, in recruitment of TNF-primed neutrophils to intravas-
cularly deposited ICs. Furthermore, endothelial ICAM-1,
which is a major ligand for neutrophil Mac-1, played an
essential role in IC-mediated neutrophil arrest.
The paradigm for leukocyte recruitment during an inflam-
matory response has been primarily established from studies of
IVM following injection of cytokines in tissues of interest.
Unlike the requirement in these models for cytokine-inducible
endothelial adhesion molecules and chemokines, neutrophil
recruitment following strictly intravascular IC deposition ap-
pears not to require overt endothelial cell activation. Instead, it
relies on the presence of constitutively present adhesion re-
ceptors such as neutrophil Fc?Rs and Mac-1 and endothelial
ICAM-1, and potentially “priming” stimuli such as TNF. TNF
is a pleiotropic cytokine shown to activate endothelial cells and
neutrophils. Previous analyses of leukocyte recruitment have
most often been conducted several hours following intrascrotal
injections of TNF and/or other inflammatory mediators. This leads
to endothelial activation and adhesion receptor and chemokine
expression that promote robust leukocyte-vessel wall interactions
[24], which precludes the analyses of the role of TNF priming of
neutrophils in neutrophil recruitment. Our studies show for the
first time that neutrophil priming has relevance in vivo and in
particular, contributes to leukocyte recruitment.
The increase in adhesion of TNF-primed BMN in response
to intravascular ICs provides strong evidence that TNF can
modulate Fc?R function in vivo and that this has consequences
for leukocyte recruitment. Previous studies show that inflam-
matory mediators such as C5a, TNF, and IFN-? can up-
regulate Fc?RIIIs on monocytes and macrophages [25–27].
However, Fc?R surface expression on neutrophils did not
change following TNF treatment. The mechanism for enhanced
adhesion could thus be related to TNF-induced changes in
Fc?R affinity for ligand. In support of this, TNF treatment of
human neutrophils did not increase surface expression of
Fc?Rs but induced their clustering leading to cytoskeletal
rearrangement and subsequent neutrophil binding to IgG in
vitro [28]. Similarly, stimulation with the neutrophil chemoat-
tractant fMLP significantly increased Fc?R engagement of
IC-opsonized targets in vitro [29]. Thus, although Fc?Rs are
capable of binding ICs, their affinity for ligand may be in-
creased by exogenous priming/activating stimuli, in a manner
analogous to that described for ?2integrins [8]. Indeed, Fc?R
affinity modulation may represent another layer of regulation of
Fc?R function in addition to those already described. Fc?R
function is regulated by cytokine-induced increase in their
surface expression levels. Cytokine regulation of IgG class-
inIgG-mediated
switching indirectly regulates Fc?R function activity as Fc?Rs
are known to differentially engage specific IgG isotypes [3].
Previous studies suggest that the mechanism of TNF-in-
duced priming may be an increase in receptor expression or
post-receptor events [30, 31]. For example, TNF priming of the
fMLP-mediated oxidative burst in neutrophils may be attrib-
uted to TNF-induced increases in cell surface expression of
fMLP receptors and/or increased phosphorylation of the
NADPH oxidase phox components [32]. In our studies, TNF
priming increased Mac-1 surface expression (Fig. 1, A and B)
[28], but this itself did not enhance the number of neutrophils
adherent to the vessel wall (Fig. 1C). The increase in adhesion
was only evident when a secondary stimulus, ICs, was present
within the vasculature (Fig. 1C). TNF priming likely modulates
more than Mac-1 expression to promote adhesion, as an in-
crease in integrin expression alone is not sufficient for ligand
binding. Additional changes in integrin affinity and/or stimu-
lation of downstream signaling pathways (e.g., protein kinase
C-dependent pathways) are needed. Regardless of the mecha-
nism of TNF priming, our studies do show that TNF priming is
relevant in the context of IC- mediated neutrophil recruitment.
This is important data, as although we assume that neutrophil
priming plays a physiological role in inflammatory responses,
evidence of this in vivo has not been presented previously.
Crosstalk between Fc?Rs and Mac-1 has been implicated in
IC-induced neutrophil adhesion in vitro. Similarly to GPCRs,
Fc?Rs switch Mac-1 from a low- to a high-affinity/avidity state
that binds ligands [8, 9, 33]. Here, Fc?R engagement of
intravascular ICs may trigger the activation of Mac-1 and
subsequent binding to its ligand ICAM-1 present on the endo-
thelium. At a molecular level, this may occur through Fc?R-
induced phosphorylation of L-plastin, a leukocyte-specific, ac-
tin-bundling protein that is involved in Mac-1 activation [34]
and/or Mac-1 clustering [33]. Mac-1 is critical for establishing
shear-resistant contacts with the ICs [4, 7]. We propose that
Mac-1 interaction with ICAM-1 strengthens the adhesive con-
tact initiated by Fc?R/IC interactions and that this is required
for firm arrest.
Our studies also suggest that Mac-1 and ICAM-1 interac-
tions with other binding partners may contribute to IC-induced
adhesion, as a deficiency in both leads to a greater defect in
neutrophil adhesion than either one alone. Notably, elimination
of neutrophil Mac-1 and endothelial ICAM-1 basically abro-
gates adhesion, suggesting that these two proteins are respon-
sible for all detectable IC-induced adhesive activity of TNF-
primed neutrophils. Our finding that ICAM-1 on the unstimu-
lated endothelium supports IC-mediated neutrophil adhesion is
surprising, as there are limited data about the contribution of
ICAM-1 on “unactivated” endothelium to neutrophil recruit-
ment [35]. Complement activation by ICs could also generate a
Mac-1 ligand, complement C3 that could support cell adhe-
sion. However, although complement C1q activated by the
classical complement pathway is required for permeability-
induced IC deposition, complement C3 does not play a role in
neutrophil recruitment to intravascular ICs in this model [5].
In conclusion, our analyses revealed a role for TNF priming
of neutrophils in selectively enhancing neutrophil adhesion to
intravascular ICs. Priming of neutrophil functions by cytokines
might contribute to the pathology of IgG-mediated disease. For
Lauterbach et al.
TNF/Mac-1 in immune complex-induced neutrophil recruitment1429

Page 8

example, in transfusion-related acute lung injury, characterized
by antibody- and neutrophil-priming stimuli, cytokine priming of
neutrophil functions may enhance disease severity [36]. Impor-
tantly, we provided evidence that Fc?R interaction with ICs leads
to Mac-1 interaction with ICAM-1 on the endothelium, steps that
are required for firm neutrophil arrest in the context of intravas-
cular IC deposition. Our finding that the molecular components
required for IC-induced recruitment are constitutively present in
the vascular system (i.e., Fc?R, Mac-1, and ICAM-1) suggests
that they may constitute a potent pathway of neutrophil recruit-
ment in the absence of overt inflammation. This could represent
an early pathogenic event in IC-induced disease characterized by
circulating ICs such as vasculitides, systemic lupus erythemato-
sus, and some forms of glomerulonephritis.
ACKNOWLEDGMENTS
This study was supported by NIH grants RO1AR050800 and
RO1HL065095 to T. N. M. M. L. was partially supported by
the Else Kroener-Fresenius Foundation.
REFERENCES
1. Jancar, S., Sanchez Crespo, M. (2005) Immune complex-mediated tissue
injury: a multistep paradigm. Trends Immunol. 26, 48–55.
2. McKenzie, S. E., Schreiber, A. D. (1998) Fc ? receptors in phagocytes.
Curr. Opin. Hematol. 5, 16–21.
3. Nimmerjahn, F., Ravetch, J. V. (2006) Fc? receptors: old friends and new
family members. Immunity 24, 19–28.
4. Coxon, A., Cullere, X., Knight, S., Sethi, S., Wakelin, M. W., Stavrakis, G.,
Luscinskas, F. W., Mayadas, T. N. (2001) Fc ? RIII mediates neutrophil
recruitment to immune complexes. a mechanism for neutrophil accumu-
lation in immune-mediated inflammation. Immunity 14, 693–704.
5. Stokol, T., O’Donnell, P., Xiao, L., Knight, S., Stavrakis, G., Botto, M., von
Andrian, U. H., Mayadas, T. N. (2004) C1q governs deposition of circu-
lating immune complexes and leukocyte Fc? receptors mediate subse-
quent neutrophil recruitment. J. Exp. Med. 200, 835–846.
6. Liu, Z., Zhao, M., Li, N., Diaz, L. A., Mayadas, T. N. (2006) Differential
roles for ?2 integrins in experimental autoimmune bullous pemphigoid.
Blood 107, 1063–1069.
7. Tang, T., Rosenkranz, A., Assmann, K. J., Goodman, M. J., Gutierrez-
Ramos, J. C., Carroll, M. C., Cotran, R. S., Mayadas, T. N. (1997) A role
for Mac-1 (CDIIb/CD18) in immune complex-stimulated neutrophil func-
tion in vivo: Mac-1 deficiency abrogates sustained Fc? receptor-depen-
dent neutrophil adhesion and complement-dependent proteinuria in acute
glomerulonephritis. J. Exp. Med. 186, 1853–1863.
8. Jones, S., Brown, E. (1996) Functional cooperation between Fc? receptors
and complement receptors in phagocytes. In Human IgG Fc Receptors (J.
van de Winkel, P. Capel, eds.), Austin, TX, USA, R. G. Landes, 149–163.
9. Ross,G.D.(2000)Regulationoftheadhesionversuscytotoxicfunctionsofthe
Mac-1/CR3/?M?2-integrin glycoprotein. Crit. Rev. Immunol. 20, 197–222.
10. Feldmann, M., Pusey, C. D. (2006) Is there a role for TNF-? in anti-
neutrophil cytoplasmic antibody-associated vasculitis? Lessons from other
chronic inflammatory diseases. J. Am. Soc. Nephrol. 17, 1243–1252.
11. Downey, G. P., Fukushima, T., Fialkow, L., Waddell, T. K. (1995) Intra-
cellular signaling in neutrophil priming and activation. Semin. Cell Biol.
6, 345–356.
12. Swain, S. D., Rohn, T. T., Quinn, M. T. (2002) Neutrophil priming in host
defense:roleofoxidantsasprimingagents.Antioxid.RedoxSignal.4,69–83.
13. Sheppard, F. R., Kelher, M. R., Moore, E. E., McLaughlin, N. J., Banerjee,
A., Silliman, C. C. (2005) Structural organization of the neutrophil
NADPH oxidase: phosphorylation and translocation during priming and
activation. J. Leukoc. Biol. 78, 1025–1042.
14. Coxon, A., Rieu, P., Barkalow, F. J., Askari, S., Sharpe, A. H., von
Andrian, U. H., Arnaout, M. A., Mayadas, T. N. (1996) A novel role for the
? 2 integrin CD11b/CD18 in neutrophil apoptosis: a homeostatic mech-
anism in inflammation. Immunity 5, 653–666.
15. Lauterbach, M., Horstick, G., Kempf, T., Weilemann, L. S., Munzel, T.,
Kempski, O. (2006) Anti-inflammatory treatment with standardized human
serum protein solution reduces local and systemic inflammatory response
after hemorrhagic shock. Eur. Surg. Res. 38, 399–406.
16. Hirahashi, J., Mekala, D., Van Ziffle, J., Xiao, L., Saffaripour, S., Wagner,
D. D., Shapiro, S. D., Lowell, C., Mayadas, T. N. (2006) Mac-1 signaling
via Src-family and Syk kinases results in elastase-dependent thrombohe-
morrhagic vasculopathy. Immunity 25, 271–283.
17. Mempel, T. R., Moser, C., Hutter, J., Kuebler, W. M., Krombach, F. (2003)
Visualization of leukocyte transendothelial and interstitial migration using
reflected light oblique transillumination in intravital video microscopy. J.
Vasc. Res. 40, 435–441.
18. Jung, U., Norman, K. E., Scharffetter-Kochanek, K., Beaudet, A. L., Ley,
K. (1998) Transit time of leukocytes rolling through venules controls
cytokine-induced inflammatory cell recruitment in vivo. J. Clin. Invest.
102, 1526–1533.
19. Stokes, K. Y., Clanton, E. C., Gehrig, J. L., Granger, D. N. (2003) Role of
interleukin 12 in hypercholesterolemia-induced inflammation. Am. J.
Physiol. Heart Circ. Physiol. 285, H2623–H2629.
20. Lauterbach, M., Horstick, G., Plum, N., Weilemann, L. S., Munzel, T.,
Kempski, O. (2006) Shunting of the microcirculation after mesenteric
ischemia and reperfusion is a function of ischemia time and increases
mortality. Microcirculation 13, 411–422.
21. Klut, M. E., Whalen, B. A., Hogg, J. C. (1997) Activation-associated
changes in blood and bone marrow neutrophils. J. Leukoc. Biol. 62,
186–194.
22. Takai, T., Li, M., Sylvestre, D., Clynes, R., Ravetch, J. V. (1994) FcR ? chain
deletion results in pleiotrophic effector cell defects. Cell 76, 519–529.
23. Dunne, J. L., Collins, R. G., Beaudet, A. L., Ballantyne, C. M., Ley, K.
(2003) Mac-1, but not LFA-1, uses intercellular adhesion molecule-1 to
mediate slow leukocyte rolling in TNF-?-induced inflammation. J. Immu-
nol. 171, 6105–6111.
24. McIntyre, T. M., Prescott, S. M., Weyrich, A. S., Zimmerman, G. A. (2003)
Cell-cell interactions: leukocyte-endothelial interactions. Curr. Opin. He-
matol. 10, 150–158.
25. Ravetch, J. V. (2002) A full complement of receptors in immune complex
diseases. J. Clin. Invest. 110, 1759–1761.
26. Shushakova, N., Skokowa, J., Schulman, J., Baumann, U., Zwirner, J.,
Schmidt, R. E., Gessner, J. E. (2002) C5a anaphylatoxin is a major
regulator of activating versus inhibitory Fc?Rs in immune complex-
induced lung disease. J. Clin. Invest. 110, 1823–1830.
27. Guyre, P. M., Morganelli, P. M., Miller, R. (1983) Recombinant immune
interferon increases immunoglobulin G Fc receptors on cultured human
mononuclear phagocytes. J. Clin. Invest. 72, 393–397.
28. Reumaux, D., Hordijk, P. L., Duthilleul, P., Roos, D. (2006) Priming by
tumor necrosis factor-? of human neutrophil NADPH-oxidase activity
induced by anti-proteinase-3 or anti-myeloperoxidase antibodies. J. Leu-
koc. Biol. 80, 1424–1433.
29. Nagarajan, S., Venkiteswaran, K., Anderson, M., Sayed, U., Zhu, C.,
Selvaraj, P. (2000) Cell-specific, activation-dependent regulation of neu-
trophil CD32A ligand-binding function. Blood 95, 1069–1077.
30. O’Flaherty, J. T., Rossi, A. G., Redman, J. F., Jacobson, D. P. (1991)
Tumor necrosis factor-? regulates expression of receptors for formyl-
methionyl-leucyl-phenylalanine, leukotriene B4, and platelet-activating
factor. Dissociation from priming in human polymorphonuclear neutro-
phils. J. Immunol. 147, 3842–3847.
31. Tennenberg, S. D., Solomkin, J. S. (1990) Activation of neutrophils by
cachectin/tumor necrosis factor: priming of N-formyl-methionyl-leucyl-
phenylalanine-induced oxidative responsiveness via receptor mobilization
without degranulation. J. Leukoc. Biol. 47, 217–223.
32. Dewas, C., Dang, P. M., Gougerot-Pocidalo, M. A., El-Benna, J. (2003)
TNF-? induces phosphorylation of p47(phox) in human neutrophils: par-
tial phosphorylation of p47phox is a common event of priming of human
neutrophils by TNF-? and granulocyte-macrophage colony-stimulating
factor. J. Immunol. 171, 4392–4398.
33. Jongstra-Bilen, J., Harrison, R., Grinstein, S. (2003) Fc?-receptors induce
Mac-1 (CD11b/CD18) mobilization and accumulation in the phagocytic
cup for optimal phagocytosis. J. Biol. Chem. 278, 45720–45729.
34. Wang, J., Brown, E. J. (1999) Immune complex-induced integrin activa-
tion and L-plastin phosphorylation require protein kinase A. J. Biol.
Chem. 274, 24349–24356.
35. Foy, D. S., Ley, K. (2000) Intercellular adhesion molecule-1 is required
for chemoattractant-induced leukocyte adhesion in resting, but not in-
flamed, venules in vivo. Microvasc. Res. 60, 249–260.
36. Bux, J. (2005) Transfusion-related acute lung injury (TRALI): a serious
adverse event of blood transfusion. Vox Sang. 89, 1–10. mention.
1430 Journal of Leukocyte Biology
Volume 83, June 2008
http://www.jleukbio.org