FcR-Bearing Myeloid Cells Are Responsible for Triggering
Murine Lupus Nephritis1
Amy Bergtold,* Anamika Gavhane,†Vivette D’Agati,‡Michael Madaio,§and Raphael Clynes2†
Lupus glomerulonephritis is initiated by deposition of IgG-containing immune complexes in renal glomeruli. FcR engagement by
immune complexes (IC) is crucial to disease development as uncoupling this pathway in FcR??/?abrogates inflammatory re-
sponses in (NZB ? NZW)F1mice. To define the roles of FcR-bearing hemopoietic cells and of kidney resident mesangial cells in
pathogenesis, (NZB ? NZW)F1bone marrow chimeras were generated. Nephritis developed in (NZB ? NZW)F1mice expressing
activating FcRs in hemopoietic cells. Conversely, recipients of FcR??/?bone marrow were protected from disease development
despite persistent expression of FcR? in mesangial cell populations. Thus, activating FcRs on circulating hemopoietic cells, rather
than on mesangial cells, are required for IC-mediated pathogenesis in (NZB ? NZW)F1. Transgenic FcR??/?mice expressing
FcR? limited to the CD11b?monocyte/macrophage compartment developed glomerulonephritis in the anti-glomerular basement
disease model, whereas nontransgenic FcR??/?mice were completely protected. Thus, direct activation of circulating FcR-bearing
myeloid cells, including monocytes/macrophages, by glomerular IC deposits is sufficient to initiate inflammatory responses. The
Journal of Immunology, 2006, 177: 7287–7295.
tis. Studies of acute murine models of Ab-mediated inflammation
in the skin (1–4), joints (5–12), lungs (13), kidneys (14–19), and
peritoneum (20, 21) in gene-deficient mice permit the general con-
clusion that the coordinate expression of activating and inhibitory
FcRs on effector cells regulates inflammatory responses. Comple-
ment components including C5a contribute directly as chemoat-
tractants and as inducers of preferential up-regulation of activating
FcRs on effector cells (20, 22–24).
The initial events following IC deposition in the tissues include
the local activation of complement and the triggering of tissue-
resident cells though their Fc and complement receptors. The re-
sultant collective action of locally produced chemokines, cyto-
kines, and small molecule mediators of inflammation activates
endothelial cells and promotes the adhesion and diapedesis of ac-
tivated bloodborne effectors, including monocytes and neutrophils,
into the tissue. In this scenario, the recruitment of circulating cel-
lular effectors is expected to occur as a consequence of local ac-
tivation of resident tissue cells. The importance of resident cells
including tissue macrophages and mast cells in the initiation of the
mmune complex (IC)3deposition in tissue contributes to
many autoimmune disease states including systemic vascu-
litis, arthritis, blistering skin diseases, and glomerulonephri-
inflammatory cascade and subsequent recruitment of circulating
neutrophils has been demonstrated in the joints (25–27) and in
Arthus reactions in the lungs (13, 22, 28), peritoneum (21, 29, 30),
and skin (3).
In the kidney, the relevant resident cell that would be expected
to initiate the inflammatory response to ICs deposited in glomeruli
is the mesangial cell (MC). MC activation contributes directly to
glomerular pathogenesis through proliferation and collagen depo-
sition and indirectly through the production of the inflammatory
mediators (31, 32) cytokines and chemokines (31, 33). Indeed,
FcRs are expressed on cultured rodent and human MC (34–36),
and Fc?R cross-linking on cultured MC induces matrix deposition
(34) and the production of inflammatory mediators including che-
mokines (37, 38) and cytokines (39). Numerous studies have im-
plicated Fc?R cross-linking on MC as a proximal and key step in
IC-mediated nephritis, yet few studies have directly demonstrated
mesangial expression of Fc?R in vivo. Low-level expression of the
inhibitory Fc?RIIB on glomerular cells was detectable by immu-
nohistochemistry (17), but other studies have failed to detect Fc?R
at the RNA level (40). FcR??/?mice are protected from the de-
velopment of nephritis despite IC mesangial deposition (18, 19).
However, a requisite inflammatory role of FcRs on MC in vivo
Recent work in the anti-glomerular basement membrane (anti-
GBM) model suggests instead that circulating hemopoietic cells
directly engage immune deposits in the mesangium, initiating the
inflammatory response without prior recruitment by FcR-engaged
MC. Transferred wild-type (WT) neutrophils become activated in
FcR??/?hosts bearing IC mesangial deposits, arguing that acute
injury can be initiated by FcR cross-linking-circulating neutrophils
(41). In bone marrow (BM) chimeras (42) using FcR??/?and
FcR??/?donors and recipients, anti-GBM nephritis required FcR-
bearing cells in the hemopoietic compartment, suggesting that MC
FcR engagement is not necessary for the induction of the IC-me-
diated inflammatory responses (42). Although these short-term
acute models provide important mechanistic insights, the sponta-
neous nephritis model in (NZB ? NZW)F1mice most closely
approximates pathogenetic mechanisms mediating human lupus
nephritis. We have addressed the role of FcR? in intrinsic renal
*Integrated Program in Cellular, Molecular, and Biophysical Studies,†Department of
Microbiology and Medicine,‡Department of Pathology, Columbia University, Col-
lege of Physicians and Surgeons, New York, NY 10032; and§Department of Med-
icine, University of Pennsylvania, School of Medicine, Philadelphia, PA 19104
Received for publication December 20, 2005. Accepted for publication September
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by the Immunology Training Program (T32 AI 07525 (to
A.B.), the National Institutes of Health/National Institute of Allergy and Infectious
Diseases RO3AR45764, and by an Investigator Award of the Arthritis Foundation
2Address correspondence and reprint requests to Dr. Raphael Clynes, Columbia Uni-
versity, College of Physicians and Surgeons, P & S Building, Room 8-510, 630 West
168th Street, New York, NY 10032. E-mail address: email@example.com
3Abbreviations used in this paper: IC, immune complex; MC, mesangial cell; GBM,
glomerular basement membrane; WT, wild type; BM, bone marrow; PAS, periodic
acid-Schiff; SLE, systemic lupus erythematosus.
The Journal of Immunology
Copyright © 2006 by The American Association of Immunologists, Inc.0022-1767/06/$02.00
cells or hemopoietic cells in the spontaneous NZB/NZW lupus
nephritis model and find that mesangial FcR expression is not re-
quired for disease development. Furthermore, we have partially
reconstituted anti-GBM nephritis in FcR??/?by transgenic ex-
pression of FcR? in the monocyte/macrophage compartment, im-
plicating direct activation of this FcR-bearing cellular subset in the
initiation of the inflammatory phase of IC-mediated nephritis.
Thus, direct activation of FcR-bearing monocyte/macrophages is
sufficient to induce inflammatory responses in response to glomer-
ular IC deposition.
Materials and Methods
(NZB ? NZW)F1FcR??/?mice were generated from an intercross of
NZB FcR??/?male and NZW FcR??/?female mice (18). To generate
BM chimeras, 10 ? 106BM cells obtained from 3-wk-old (NZB ?
NZW)F1FcR??/?and FcR??/?mice (The Jackson Laboratory) were in-
jected i.v. into the tail vein of lethally irradiated recipients (1000 rad ? 1
dose). Chimeric mice were given oral ciprofloxacin in the water ad libitum
for 14 days after reconstitution and followed for the development of pro-
teinuria weekly for 9 mo. Proteinuria was read using Urostix for the NZB/
NZW mice and scored positive if 2? measurements (?250 mg/dl) were
recorded for two successive readings. A subset of mice was sacrificed at 6
mo for histopathological analysis of the kidney. These studies were re-
viewed and approved by the Institution Animal Care and Use Committee
of Columbia University.
CD11b-? Tg?mice were generated after injection of oocytes obtained
from FcR??/?mice. The transgenic construct was generated by insertion
of the murine FcR? cDNA (550-bp fragment) as an EcoRI fragment (43)
into pB203 (a gift from Dr. D. G. Tenen, Harvard Medical School, Boston,
MA; see Ref 44) containing the 1.7-kb 5?-flanking sequences of the mouse
CD11b promoter and the 3?-flanking region from the human growth hor-
mone gene. A NotI/HindIII fragment (containing 5?-CD11bpromoter-FcR?
cDNA-hGH-3?) was injected into the oocytes and three founder lines har-
boring the transgene were further analyzed for expression. Of these three
founders, only one (line 14) expressed the FcR? chain in peritoneal
Accelerated anti-GBM nephritis
Mice were immunized with 100 ?g of sheep IgG in CFA 3 days before i.v.
injection of 150 ?l of specific sheep anti-mouse GBM sera. Urine was
obtained daily and blood obtained on the day before injection with anti-
GBM sera and then at the time of sacrifice 7 days later. Urine samples were
diluted in PBS and protein quantified by the Bradford method (Bio-Rad)
using an ELISA plate reader at OD570.
Anti-dsDNA and soluble immune complex ELISAs
Diluted serum (1/100) from 6- to 7-mo-old NZB/NZW-??/?and NZB/
NZW-??/?mice were added to ELISA plates coated with C1q (Sigma-
Aldrich) for detection of ICs (45, 46) and to dsDNA-coated plates (United
Biotech) for detection of Abs to chromatin. After washing away unbound
serum, rat anti-mouse IgG (BD Pharmingen) was added. Alkaline phos-
phatase-conjugated AKP polyclonal anti-rat IgG (BD Pharmingen) was
used as secondary Ab. After incubation with p-nitrophenyl phosphate sub-
strate, the samples were read spectrophotometrically at 405 nm with an
ELISA reader (Molecular Devices).
Immunofluorescence and immunohistochemistry
For histological analysis, formalin-fixed sections were stained with H&E or
periodic acid-Schiff (PAS). To detect IC deposition, paraformaldehyde- or
acetone-fixed cryosections were stained with (1/1000 diluted) FITC goat
anti-mouse C3 and IgG (Valeant Pharmaceuticals). To detect FcR?, a poly-
clonal anti-FcR? rabbit IgG (gift from Dr. J. Ravetch, The Rockefeller
University) or rat anti-Mac-1 (clone C71/16; BD Pharmingen) followed by
rabbit anti-rat IgG Alexa594 (Molecular Probes). Biotinylated goat anti-
rabbit IgG, followed by either streptavidin-FITC or streptavidin-HRP was
used for detection. A Nikon Eclipse 600 microscope equipped with a RT
Spot digital camera was used for imaging.
Renal pathological assessment
PAS sections were prepared from WT, FcR??/?, and FcR??/?CD11b-?
Tg?kidneys on day 7 after induction of accelerated glomerulonephritis
(five per group). Slides were examined in a blinded fashion by one of us
(V. d’A.). Severity of the following seven categories of histological activity
were semiquantitatively graded as follows: glomerular fibrinoid necrosis
0–4, endocapillary hypercellularity 0–4, glomerular leukocyte infiltration
0–4, crescents 0–4, tubular degeneration 0–4, casts 0–4, and interstitial
inflammation 0–4.The cumulative pathological score is the sum of all
seven categories and has a possible range of 0–28.
MC and NK culture
Glomeruli were isolated with successive sieving (47). Kidneys were
minced with scissors and tissue fragments were passed through a no. 60
mesh sieve (Fisher Scientific) and then sequentially passed through no. 100
and no. 200 sieves. Glomeruli were digested with 0.1% collagenase type IV
(Sigma-Aldrich) and 0.1% trypsin (Invitrogen Life Technologies) for 30
min at 37°C before plating in 24 wells in DMEM/10% FCS. Cells were
passaged in D-valine-substituted medium to eliminate fibroblasts. After 2
wk in culture, cells exhibited a stellate morphology and were replated.
Immunostains were smooth muscle actin-positive, weakly 2.4G2?and
Mac-1?, confirming their MC origin. RNA was prepared from MCs using
TRIzol and cDNA was generated using the cloned avian myeloblastosis
virus first-strand synthesis kit according to the manufacturer’s protocol
(Invitrogen Life Technologies). Primer sequences for RT-PCR amplifica-
tion (30 cycles) of FcR? were as follows: 5?-CCAGGATGATCTC
AGCCG-3? and 5?-ACAGTAGAGTAGGGTAAG-3?. These primers am-
plify a 137-bp band corresponding to exons 1 and 2 of the ? subunit. The
band is not amplified in genomic DNA due to intervening intronic se-
quences. The housekeeping gene, HPRT, was amplified from cDNA using
the following primer sequences: 5?-AGCTACTGTAATGATCAGTCA
ACG-3? and 5?-AGAGGTCCTTTTCACCAGCA-3?. For assessing MC
chimerism, genomic DNA was subjected to PCR analysis using the fol-
lowing primer sequences: neo, CTCGTGCTTTACGGTATCGCC; ?-1,
TATAGCTGCCTT. Annealing temperature was 62°C. Knockout and WT-
amplified products were 260 and 224 bp, respectively.
Hemopoietic chimerism was assessed in cultured NK cells obtained af-
ter isolation of the adherent cell population from a 14-day culture of nylon
wool nonadherent splenocytes grown in IL-2 (10,000 U/ml). Flow cyto-
metric analysis used anti-NK1.1 PE and 2.4G2-FITC (BD Pharmingen).
Murine NK cells do not express FcRIIb (48) and thus the anti-FcRII/III
mAb (2.4G2) recognizes only FcRIII on these cells.
Western blot analysis of FcR? expression
Protein extracts were obtained from B cells, T cells, NK cells, and neu-
trophils were immunoblotted with polyclonal rabbit anti-mouse FcR?
chain IgG and anti-?-actin Abs. Neutrophils were obtained from thiogly-
colate- elicited peritoneal exudates (4 h after i.p. injection of thioglycolate)
after GR-1?bead selection (Miltenyi Biotec). Adherent peritoneal macro-
phages were obtained from thioglycolate-elicited exudates (72 h after i.p.
injection). B and T cells were obtained from CD43?and CD3?splenocyte
populations, respectively. All cell populations were lysed in TBS buffer
that contained 1% Triton X-100, 2 mM EDTA, and complete mini-protease
Rabbit IgG-opsonized SRBCs were prepared with subagglutinating quan-
tities of rabbit anti-sheep RBC IgG (MP Biomedicals). After washing away
free Ab, IgG-opsonized RBCs were added to adherent macrophages for 1 h
at 37°C. Unphagocytosed RBCs were removed by osmotic lysis, and
phagocytosis plates were fixed with PBS/0.25% glutaraldehyde before mi-
Blood albumin and urea nitrogen measurements
Blood samples were read by the Clinical Chemistry Laboratory of the
Irving Clinical Research Center at Columbia-Presbyterian Hospital.
BM chimeric NZB/NZW mice reveal a requirement for
FcR?-expressing hemopoietic cells for nephritis development
(NZB ? NZW)F1female mice develop a uniformly fatal rapidly
progressive IC nephritis heralded by the serological appearance of
anti-chromatin IgG autoantibodies at 4–6 mo of age. Disease pro-
gression is swift, with a median survival of 180 days. However, in
(NZB ? NZW)F1FcR ??/?female mice, IgG autoantibodies oc-
cur with equivalent titers and are deposited similarly in the kidney,
7288FcR-BEARING MYELOID CELLS TRIGGER LUPUS NEPHRITIS
but induce little subsequent inflammation (18). Disease progres-
sion is markedly attenuated with median survival of ?400 days
with many animals living a normal life span. To distinguish the
role of FcR-bearing hemopoietic cells from FcR-bearing renal cell
in the development of the effector response in (NZB ? NZW)F1
nephritis, BM transplants between FcR??/?and FcR??/?mice
Hemopoietic reconstitution of lethally irradiated recipients was
assessed by immunophenotypic analysis of NK populations from
representative mice 4 mo postreconstitution. NK populations from
WT BM recipients uniformly expressed Fc?RIII, whereas FcR??/?
recipients were Fc?RIII negative, consistent with complete or near-
complete hemopoietic reconstitution by donor marrow (Fig. 1A). Glo-
in FcR??/?3 FcR??/?reciprocal chimeric mice, indicating that in-
trinsic glomerular cells in the kidney, presumably MC, remained
recipient derived (Fig. 1B) 4 mo after transplant. As previously
seen in nontransplanted NZB/NZW, FcR? deficiency has little im-
pact on the development of autoantibodies and their glomerular
deposition when assessed 6 mo after transplant. All experimental
groups developed anti-chromatin
C1q-binding activity (indicative of the presence of circulating ICs
and/or anti-C1q autoantibodies (45, 46)) regardless of the FcR?
genotype status of the host or recipient (Fig. 2, A and B). Circu-
lating ICs were deposited in the glomeruli in a similar fashion as
assessed by immunofluorescence studies demonstrating equivalent
IgG and complement glomerular deposition in all experimental
groups (Fig. 2C).
Although the afferent limb of autoimmunity was intact regard-
less of the FcR? genotype, the efferent response required activat-
ing FcR-expressing hemopoietic cells. Proteinuria occurred in 90–
100% of the NZB/NZW recipients of WT BM regardless of the
host genotype, whereas proteinuria occurred in only 10–20% of
recipients of NZB/NZW FcR??/?marrow (Fig. 3A). Thus, WT
hemopoietic cells can transfer disease susceptibility to FcR??/?
hosts and conversely FcR??/?BM-derived cells limit disease de-
velopment in WT NZB/NZW hosts, indicating that hemopoietic
expression of activating FcRs is both necessary and sufficient for
the development of nephritis.
mopoietic reconstitution whereas intrinsic renal cells remain recipient de-
rived. A, Six- to 8-wk-old irradiated (1000 cGy) NZB/NZW ??/?and ??/?
mice were reconstituted with 5 ? 106BM cells obtained from either NZB/
NZW ??/?or ??/?3-wk-old mice. Hemopoietic reconstitution was as-
sessed by immunophenotyping NK cell populations of mixed chimeras
(??/?3??/?and ??/?3??/?). Murine NK cells express FcRIII as their
sole Fc receptor and thus anti-FcRII/III mAb 2.4G2 binding reflects ex-
pression of the FcR?-dependent FcRIII. NK cells were ?98% donor de-
rived. B, Frozen renal sections of mice in A were stained with a polyclonal
rabbit anti-FcR? IgG and counterstained with hematoxylin. FcR? was de-
tected in ??/?3??/?but not in ??/?3??/?BM chimeras, indicating that
intrinsic renal cells remained predominantly recipient-derived 4 mo after
(NZB ? NZW)F1BM chimeras show complete donor he-
of FcR? status. A and B, Anti-dsDNA IgG and circulating IC detected in 1/100 dilutions of sera from BM chimeric mice 6 mo after transplant.
Anti-chromatin IgG, IgMs, and circulating IC were similar in all groups (ANOVA, p values of 0.404, 0.517, and 0.240, respectively). Notably, however,
??/?3??/?trended toward higher titers than that of ??/?3??/?(in all cases, except for IgG1 antichromatin, data not shown). C, Immunostains of fixed
frozen renal sections using anti-mouse IgG and anti-mouse complement revealed glomerular deposition in both mixed BM chimeras. Representative
examples of five mice per group are shown.
NZB/NZW BM chimeras develop similar levels of anti-chromatin autoantibodies, circulating IC, and glomerular IC deposition regardless
7289The Journal of Immunology
Histological analysis revealed fulminant glomerulonephritis in
recipients of WT BM with glomerular hypertrophy, mesangial and
endocapillary hypercellularity, neutrophilic and monocytic infil-
tration, necrosis, crescent formation, and sclerosis (Fig. 3B). In
contrast, disease protection was seen in ??/?BM recipients with
histological changes limited to mesangial expansion and mild en-
docapillary and mesangial hypercellularity.
These data suggest that FcR expression by resident cells of the
kidney does not critically contribute to the initiation of the effector
response in IC-triggered nephritis. The immunohistochemical analy-
sis of FcR? expression in Fig. 1 was performed on mice 4 mo post-
transplant and before disease onset. To confirm that MC remained
genotypically host derived at 6–9 mo, the time point when proteinuria
became evident, MC were isolated and additional immunostains were
at 6 mo posttransplant, many dual-positive FcR??Mac-1?cells were
present, indicative of infiltrating myeloid lineage cells. In healthy
FcR??/?3FcR??/?chimeric mice sacrificed 9 mo posttransplant,
immunostaining demonstrated persistent glomerular ? expression,
presumably in Mac-1?MC. In an additional experimental approach
to establish the donor vs host FcR? status of MC, MC populations
were enriched from disrupted glomeruli of chimeric mice at 6 mo
posttransplant in nephritic FcR??/?3FcR??/?mice and at 9 mo
from non-nephritic FcR??/?3FcR??/?mice and assessed for the
presence of WT and knockout FcR? alleles by genomic PCR. En-
riched MC populations exhibited typical stellate morphology and
were smooth-muscle actin positive (data not shown). PCR analysis of
genomic DNA of these MC populations (Fig. 3, C, insets, and D) at
6 mo indicated that genotypically these populations continued to in-
clude predominantly host-derived MC. Notably, however, in glomer-
ular cultures obtained from FcR??/?3FcR??/?mice at 9 mo
posttransplant, there was also PCR evidence for donor-derived
contributions consistent with replacement of some MC with BM-
derived precursors, although this was not evident by immunohis-
tochemistry. One possible explanation for the discrepancy between
the immunohistochemical data and the PCR data are that the
measurements were recorded weekly in BM chimeric mice posttransplant. All recipients of FcR??/?BM developed proteinuria by 12 mo posttransplant,
regardless of FcR? genotypic status of the recipient (??/?3??/?(n ? 17); ??/?into ??/?(n ? 10), mean age of onset, respectively, 198 ? 40 days and
186 ? 40 days). Conversely, proteinuria occurred rarely in recipients of FcR??/?BM during the 12-mo observation period (3 of 23 ??/?3??/?mice and
in 1 of 10 ??/?3??/?mice]. Two-sided Fisher’s exact ??/?3??/?vs ??/?3??/?, p ? 2.5 ? 10?8. B, H&E-stained sections demonstrate glomeru-
losclerosis and crescent formation in ??/?3??/?and ??/?3??/?kidneys. Glomerular hypertrophy and end-stage fibrotic changes were consistently found
in this group of mice when proteinuria became evident. Histological changes in ??/?3??/?and ??/?3??/?mice 6 mo posttransplant were minimal and
included mesangial thickening but little sclerosis and minimal increased cellularity. C, ??/?3??/?chimeras: immunofluorescence staining for Mac-1 (red) and
FcR? (green) reveals persistent FcR? expression and lack of infiltrating Mac-1?cells in non-nephritic animals at 9 mo posttransplant. Inset, PCR analysis
of isolated DNA from MC populations isolated from glomeruli reveal persistence of host FcR??alleles. ??/?3??/?chimeras: infiltrating dual-positive
FcR??, Mac-1?cells are seen in nephritic animals at 6 mo posttransplant, consistent with monocyte/macrophage infiltration of glomeruli. Inset, PCR
analysis of MC DNA reveals persistence of the disrupted host-derived FcR? allele in FcR??/?recipients. D, Enriched MC cultures were obtained from
five chimeric mice and two control nontransplanted C57BL/6 ??/?and ??/?mice are shown. ??/?3??/?and ??/?3??/?chimeric mice were sacrificed
at 9 and 6 mo, respectively. The ??/?3??/?MC PCRs indicate a predominance of host-derived MC, whereas by 9 mo PCR evidence for replacement
by donor-derived sources became more evident.
Lupus nephritis requires activating FcR expression in the hemopoietic compartment but not in renal intrinsic MC. A, Urinary protein
7290FcR-BEARING MYELOID CELLS TRIGGER LUPUS NEPHRITIS
enriched MC populations might have included other cells, includ-
ing contaminating leukocytes that were detectable by these sensi-
tive PCR assays. Taken together, these data indicate that FcR ex-
pression of MC is neither necessary nor sufficient to initiate an
inflammatory nephritic response to IC. Rather, inflammation oc-
curs in NZB/NZW as a direct consequence of FcR engagement on
Lineage-specific transgenic reconstitution of FcR? in monocytes/
macrophages partially restores the ability to develop nephritis
To genetically determine the role of myeloid cells by transgenic
manipulation of FcR??/?mice, we turned to the anti-GBM ne-
phritis model in the C57BL/6 FcR??/?background. In previous
studies in the autologous anti-GBM disease model, activating FcRs
expressed on hemopoietic cells were found to be required for dis-
ease development (42). To assess the specific contributions of
FcR-bearing myeloid cells, transgenic mice expressing FcR?
driven by the CD11b promoter (Fig. 4A) were generated in
FcR??/?C57BL/6 mice (CD11b-? Tg?). Three Tg?founder
mice were analyzed for functional expression of FcR? in perito-
neal macrophages. One of theses transgenic founder lines (line 14)
exhibited FcR? expression in peritoneal macrophages, but not in B
cells, T cells, NK cells, or neutrophils (Fig. 4B). Functional ex-
pression in peritoneal macrophages was shown by restored FcR-
mediated phagocytosis in CD11b-? Tg?(Fig. 4C). Lack of ex-
pression of FcR? in MC of CD11b-? Tg?was demonstrated by
RT-PCR analysis of RNA obtained from cultured MC. CD11b-?
Tg?MC did not express detectable FcR? at the RNA level neither
in the resting state nor after stimulation with IC or IFN-? for either
6 h (data not shown) or 24 h (Fig. 4D).
The transgenic CD11b-? Tg?mice provided a unique opportu-
nity to address the singular role of FcR-bearing monocytes/mac-
rophages to the development of nephritis. Mice were immunized
with sheep IgG in CFA 3 days before i.v. administration of specific
sheep anti-GBM sera (Fig. 5). Severe proteinuria, hypoalbumine-
mia, and uremia developed in all WT C57BL/6 mice by day 7
whereas FcR??/?mice, as expected, were completely protected
from disease development (Fig. 5). In contrast, CD11b-? Tg?
mice developed moderate proteinuria and consequent hypoalbu-
minemia. Histopathological assessment of H&E-stained renal sec-
tions was consistent with the induction of mild glomerulonephritis
in CD11b-? Tg?, with increased glomerular cellularity noted (Fig.
6, A and B). This likely reflects myeloid cell expression of FcR?
rather than IC-induced activation of the CD11b-? promoter in MC
as cultured CD11b-? Tg?. MC did not demonstrate IFN-?- or
IC-induced FcR? expression (Fig. 4D). Severity of histological
activity was semiquantitatively graded using seven criteria (glo-
merular fibrinoid necrosis, 0–4; endocapillary hypercellularity,
0–4; glomerular leukocyte infiltration, 0–4; crescents, 0–4; tubu-
lar degeneration, 0–4; casts, 0–4; and interstitial inflammation,
0–4). Average scores for the groups for each of the seven cate-
gories were, respectively: WT (2.8, 3.6, 3.6, 0.6, 4.0, 4.0, 1.0);
CD11b-? Tg?(0, 2.1, 1.5, 0, 2.1, 1.9, 0); and FcR??/?(0, 0.3, 0, 0,
0, 0.2, 0). Cumulative pathological scores were 20 ? 1.2 (mean ?
SD), 2.6 ? 1.3, and 6.6 ? 1.5, for WT, FcR??/?, and CD11b-?
Tg?animals, respectively. Thus, CD11b-? Tg?animals devel-
oped an intermediate level of glomerulonephritis manifested as
increased proteinuria and histological evidence of mildly increased
glomerular endocapillary cellularity and leukocyte infiltration.
To determine whether the increased cellularity and leukocyte
infiltration were due to the glomerular recruitment of Mac-1?-
circulating monocytes/macrophages, immunostaining of renal sec-
tions was performed (Fig. 6, C and D). All three groups of mice
showed similar levels of glomerular mouse anti-sheep IgG depo-
sition, confirming that the failure to develop fulminant nephritis in
FcR??/?was not due to differences in the production and depo-
sition of anti-sheep IgG in the kidney. In the absence of activating
FcR in FcR??/?, there was no evidence of infiltrating Mac-1?
macrophages despite deposition of ICs. In WT mice, Mac-1?-in-
filtrating cells were prominent. Macrophage influx was evident as
well in CD11b-? Tg?animals, indicating that reconstitution of
activating FcR expression in CD11b/Mac-1?cells was sufficient to
restore their direct recruitment and activation in glomeruli, with
injurious consequences manifested by proteinuria.
These studies provide the rationale for the systemic delivery of
FcR-targeted therapeutics in lupus. Previous work has shown that
activating FcRs are required for nephritis pathogenesis in the au-
tologous and heterologous anti-GBM models and in spontaneous
promoter reconstitutes Ab-mediated phagocytosis in FcR??/?macro-
phages (M?). A, CD11b-FcR? construct: the murine FcR? cDNA was
inserted between the 1.7-kb 5?-flanking sequences of the human CD11b
promoter and the 3?-flanking region from the human growth hormone
gene. Three founder mice were generated, of which one expressed FcR?
in macrophages. B, Western blot analysis of FcR? expression in
CD11b? Tg?mice: whole-cell extracts were run on denaturing gels and
blotted, and the FcR? chain was detected with a rabbit anti-mouse FcR?
polyclonal Abs. Blots were stripped and reprobed with anti-?-actin
polyclonal Abs for loading controls. FcR? expression in CD11b-? Tg?
mice was seen in peritoneal macrophages but not in neutrophils, NK
cells, B cells, or T cells. C, Phagocytosis assays: rabbit IgG-opsonized
SRBCs were added to adherent peritoneal macrophages from FcR??/?
and CD11b-? Tg?mice. No phagocytosis was observed by FcR??/?
macrophages after 1 h, whereas most CD11b-? Tg?macrophages had
ingested several RBCs. D, FcR? expression in cultured MC: RT-PCR
analysis using cDNA from WT, CD11b-? Tg?, and FcR??/?MC dem-
onstrates lack of FcR? RNA expression in both FcR??/?and CD11b-?
Tg?mice in resting cells or after 24 h of stimulation with either IFN-?
(1000 U/ml) or IC (50/10 ?g/ml rabbit anti-OVA/OVA). HPRT served
as a housekeeping gene control. PMN, Polymorphonuclear cells.
Targeted re-expression of FcR? by the human CD11b
7291 The Journal of Immunology