Overexpressed angiotensin-converting enzyme in neutrophils suppresses glomerular damage in crescentic glomerulonephritis

Abstract

While angiotensin-converting enzyme (ACE) regulates blood pressure by producing angiotensin II as part of the renin-angiotensin system, we recently reported that elevated ACE in neutrophils promotes an effective immune response and increases resistance to infection. Here, we investigate if such neutrophils protect against renal injury in immune complex (IC)-mediated crescentic glomerulonephritis (GN) through complement. Nephrotoxic serum nephritis (NTN) was induced in wild-type and NeuACE mice that overexpress ACE in neutrophils. Glomerular injury of NTN in NeuACE mice was attenuated with much less proteinuria, milder histological injury, and reduced IC deposits, but presented with more glomerular neutrophils in the early stage of the disease. There were no significant defects in T and B cell functions in NeuACE mice. NeuACE neutrophils exhibited enhanced IC uptake with elevated surface expression of FcγRII/III and complement receptor CR1/2. IC uptake in neutrophils was enhanced by NeuACE serum containing elevated complement C3b. Given no significant complement activation by ACE, this suggests that neutrophil ACE indirectly preactivates C3 and that the C3b-CR1/2 axis and elevated FcγRII/III play a central role in IC elimination by neutrophils, resulting in reduced glomerular injury. The present study identified a novel renoprotective role of ACE in glomerulonephritis; elevated neutrophilic ACE promotes elimination of locally formed ICs in glomeruli via C3b-CR1/2 and FcγRII/III, ameliorating glomerular injury.

NEW & NOTEWORTHY We studied immune complex (IC)-mediated crescentic glomerulonephritis in NeuACE mice that overexpress ACE only in neutrophils. Such mice show no significant defects in humoral immunity but strongly resist nephrotoxic serum nephritis (less proteinuria, milder histological damage, reduced IC deposits, and more glomerular neutrophils). NeuACE neutrophils enhanced IC uptake via increased surface expression of CR1/2 and FcgRII/III, as well as elevated serum complement C3b. These results suggest neutrophil ACE as a novel approach to reducing glomerulonephritis.

INTRODUCTION

Glomerulonephritis (GN) is a group of inflammatory diseases of glomeruli that includes lupus nephritis, IgA nephritis, antineutrophil cytoplasmic autoantibody-associated crescentic GN, and antiglomerular basement membrane (GBM) GN. GN is a major cause of chronic kidney disease and is an important cause of morbidity and mortality (1, 2). GN comprises 49.2% of biopsy-proven glomerular diseases (3); it affects 0.12% of the general population and 1.2% of the older population (1). Of GN, immune complex (IC)-mediated GN, such as lupus nephritis and IgA nephropathy, exhibits glomerular crescents in variable degrees.

GN is caused by an inflammatory response, typified by neutrophils particularly early in disease, to tissue deposits of immunoglobulins (Ig) against endogenous or exogenous antigens (4, 5). ICs elicit both a cellular response and activate the complement cascade (6, 7). Immunosuppressive agents are widely used to treat GN. However, remission and mortality rates are still unacceptably high (8).

Angiotensin-converting enzyme (ACE) is a carboxyl dipeptidase that cleaves angiotensin I to angiotensin II as part of the renin-angiotensin system (RAS) (9, 10). As opposed to renin, ACE cleaves many different peptide substrates and influences a wide range of physiological processes in addition to its well-known role in the cardiovascular system (11–13). Recently, our group reported (14) that mice overexpressing ACE in myeloid lineage cells are substantially more resistant to bacterial infection (15) and tumor growth (16) by promoting an inflammatory response. In mice termed NeuACE, ACE is overexpressed by neutrophils. Such mice have enhanced resistance to infection (15).

We hypothesized that ACE in neutrophils could reduce glomerular injury in IC-mediated GN through complement. We used a mouse model of nephrotoxic serum nephritis (NTN) (17, 18) in wild-type (WT) and NeuACE mice to study the role of neutrophil ACE in the pathogenesis of crescentic GN. This model generates glomerular pathology by the formation of rabbit-mouse IgG ICs and rabbit anti-GBM antibodies. NeuACE mice have reduced proteinuria, histological glomerular damage, and attenuated IC deposition compared with WT mice. In NeuACE mice, more glomerular neutrophils are observed in the early stage of NTN. Mechanistically, the ACE-overexpressing neutrophils take up ICs better than WT neutrophils. NeuACE neutrophils express elevated levels of complement receptor 1/2 (CR1/2; also known as CD21/CD35) and FcγRII/III (CD16/CD32). Furthermore, we found that the serum complement C3b level was increased in NeuACE mice compared with WT mice, although ACE itself does not have any complement activity. This study showed that overexpressed ACE in neutrophils contributes to the rapid elimination of locally formed ICs in glomeruli via a C3b-CR1/2 axis and FcγRII/III, ameliorating glomerular injury in crescentic GN.

METHODS

Mice

WT (C57Bl/6) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). NeuACE mice have increased ACE expression in neutrophils and a C57Bl/6 background, as described elsewhere (15). Briefly, mice were established using a transgenic gene (Tg) construct, as provided in Supplemental Fig. S1A, which was injected into C57BL/6-derived fertilized ova. The primers for the ACE NH₂-terminus (forward: 5′- CAATGGTACAGAAGGGCTGG-3′ and reverse: 5′- CACGTGGCCCATCTCGTG-3′) and internal ribosome entry site-Tomato (forward: 5′- GAAGCCGCTTGGAATAAGGC-3′ and reverse: 5′- CTTTGATGACCTCCTCGCCC-3′) were used to screen for Tg-positive mice by PCR. In one founder, analysis of ACE expression by flow cytometry of leukocytes revealed Tg expression only by neutrophils. Heterozygous mice of this founder were bred to homozygosity and are referred to as NeuACE mice. Previous analysis showed that NeuACE mice have plasma angiotensin II levels equivalent to those in WT mice (15). All mice were bred with 12:12-h day-night cycles and allowed free access to food and water. Female 8- to 12-wk-mice were used for experiments unless otherwise mentioned. All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee of Cedars-Sinai Medical Center (Protocol No. 007158).

Reagents and Antibodies

The Cytofix/Cytoperm kit with Golgi Stop (Cat. No. 554715) was purchased from BD Bioscience (Franklin Lakes, NJ). The mouse neutrophil isolation kit (Cat. No. 130-097-658) and mouse peritoneal macrophage isolation kit (Cat. No. 130-110-434) were purchased from Miltenyi Biotec (San Diego, CA). Incomplete Freund’s adjuvant (Cat. No. F5506) was purchased from Sigma-Aldrich (St. Louis, MO). Anti-PE/Cy7-conjugated CD11b (M1/70, No. 101215), APC-conjugated anti-Ly-6G (1A8, No. 127613), Pacific blue-conjugated anti-Ly-6C (HK1.4, No. 128013), APC-conjugated anti-F4/80 (BM8, No. 123115), FITC-conjugated anti-CD35/CD21 (7E9, No. 123407), APC/Cy7-conjugated anti-CD16/CD32 (93, No. 101325), FITC-conjugated anti-CD64 (X54-5/7.1, No. 139315), 7-aminoactinomycin D (7-ADD), and Flash Phalloidin Green 488 (No. 424201) were all purchased from BioLegend (San Diego, CA). Purified anti-CR1/CR2 (7E9, HM1112) was purchased from MyBiosource (San Diego, CA). APC-conjugated anti-IL-1β pro-form (NJTEN3, No. 50–112-4069), PE-conjugated anti-CD45 (30 F-11, No. 12–0451-82), purified anti-CD16/CD32 (93, No. 16–0161-82), CellROX green (No. C10444) for oxidative stress detection, SYTOX red (No. 50–113-7614), goat anti-mouse IgG (H + L) superclonal secondary antibody Alexa Fluor 488 conjugated (No. A28175), mouse anti-rabbit IgG (H + L) (No. 31213), and goat anti-rabbit IgG (H + L) highly cross-adsorbed secondary antibody Alexa Fluor 568 conjugated (No. A-11036) were purchased from Thermo Fisher Scientific (Waltham, MA). Normal rabbit IgG (No. 011-000-002) was purchased from Jackson ImmunoResearch Laboratories (West Grove, PA). Anti-ACE goat IgG (No. AF1513-SP) was purchased from R&D Systems (Minneapolis, MN). Anti-actin antibody (No. A2066) was purchased from Sigma-Aldrich. Goat anti-rabbit IgG (H&L) (No. 611–1102) and horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (No. 611-103-122) were purchased from Rockland Immunochemicals (Limerick, PA). The isotype-matched control for each antibody was purchased from the same company.

Induction of NTN

Nephrotoxic serum (NTS) was generated by immunizing rabbits with glomeruli from Sprague-Dawley rats (Lampire Biological Laboratory, Pipersville, PA) (19) and was heat inactivated at 56°C for 30 min. Based on our preliminary data, proteinuria had a within-group SD of 0.475. With a sample size of 15 per group, we will have 80% power to detect differences in means between groups of 0.5, assuming a two-sample t test with a 0.05 level of significance. On day 5, WT and NeuACE mice were immunized by intraperitoneal injection of 0.5 mg normal rabbit IgG emulsified with incomplete Freund’s adjuvant. On day 0, mice were intravenously injected with 50 µL of NTS or PBS. On day 7, urine samples were collected, and all mice were euthanized. Blood samples were collected, and kidneys were harvested after perfusion of 10-mL PBS from the left ventricle. In some experiments, mice were treated with ramipril (10 mg/kg/day) (15, 20, 21), losartan (20 mg/kg/day) (22, 23), vehicle (water) by oral gavage, or HOE-140 (100 µg/kg/day) (24) by subcutaneous injection every day for the indicated periods shown in the figures.

Proteinuria and Histological Evaluation

To obtain the protein-to-creatinine ratio (mg/mg) in urine, protein was measured with a protein assay dye (No. 500-0006, Bio-Rad, Hercules, CA), and urinary creatinine was measured using a Creatinine Assay Kit (No. DICT-500, BioAssay Systems, Hayward, CA). Two-micrometer-thick sections from paraffin-embedded kidney were deparaffinized and rehydrated, followed by periodic acid-methenamine silver stain or periodic acid-Schiff stain. Fibrinoid necrosis, a precursor lesion for crescents, defined by finding GBM rupture, fibrin deposition, and karyorrhexis (25–28), and crescents were assessed in all glomeruli on one section for each mouse using periodic acid-methenamine silver and PAS stains, respectively. Representative images were captured using a light microscope (Nikon Eclipse 50i, Nikon, Tokyo, Japan).

Detection of Glomerular Macrophages and Neutrophils

For immunohistochemistry of glomerular macrophages, formalin-fixed paraffin-embedded kidney sections were heated in 0.01 M citrate buffer (pH 6.0) for 30 min for antigen retrieval and incubated with 3% hydrogen peroxidase for 10 min to block intrinsic peroxidases. After being blocked with 1% BSA and a Streptavidin/Biotin blocking kit (No. SP-2002, Vector Laboratories, Burlingame, CA) for 30 min and 15 min each, sections were incubated for 24 h at 4°C with biotin anti-Mac-2 (No. 125404, BioLegend), followed by incubation with HRP streptavidin (No. SA-5004, Vector Laboratories) for 60 min at room temperature. HRP was visualized by reaction with ImmPACT DAB Peroxidase (HRP) Substrate (No. SK-4105, Vector Laboratories) followed by periodic acid-Schiff staining. To detect neutrophils, an esterase reaction was performed on paraffin-embedded kidney sections as previously described (5), followed by nuclei staining with hematoxylin solution (No. 3530, Ricca Chemical, Arlington, TX).

Immunofluorescence Experiments for Mouse IgG and Rabbit IgG

For immunofluorescence, 4-μm-thick frozen sections were fixed in acetone for 3 min, followed by blockade with 1% BSA for 30 min before overnight incubation at 4°C with goat anti-mouse IgG (H + L) superclonal secondary antibody Alexa Fluor 488 and goat anti-rabbit IgG (H + L) highly cross-adsorbed secondary antibody Alexa Fluor 568. Immunofluorescence images were obtained with a fluorescence microscope (BZ-X710, Keyence, Osaka, Japan). Semiquantifications (0 to 4+) for rabbit IgG and mouse IgG were performed in 10 high-power fields (magnification: ×400) per kidney in a blinded manner.

Transmission Electron Microscopy

Samples fixed in 3% glutaraldehyde were postfixed in OsO₄ and embedded in epoxy resin. Ultrathin sections were collected on carbon-coated formvar grids and stained with uranyl acetate and lead citrate. Images were obtained using a transmission electron microscope (Hitachi 7700, Hitachi, Santa Clara, CA).

Serum Mouse IgG against Rabbit IgG and Recall Assay to Rabbit IgG in Splenocytes

Mouse serum-specific IgG against rabbit IgG was measured by ELISA. Ninety-six-well plates were coated with 10 μg/mL normal rabbit IgG at 4°C overnight. The plate was washed with Tris-buffered saline with 0.05% Tween 20 and then blocked with 1% BSA at room temperature for 1 h. Serial dilutions of serum samples from NTN mice were applied on the plate and incubated at room temperature for 1 h. After being washed with Tris-buffered saline with Tween 20, mouse IgG bound to rabbit IgG was detected by anti-mouse IgG peroxidase conjugated. Relative titers were measured and expressed at an optical density of 450 nm. For the recall assay, 14 days after immunization with rabbit IgG, splenocytes were isolated from WT and NeuACE mice. Cells were seeded at 5.0 × 10⁶ cells per well in a 12-well plate and incubated with 100 μg/mL rabbit IgG for 24 h. Supernatants were collected, and concentrations of mouse IL-6 and interferon (IFN)-γ were measured using ELISA kits (No. 88-7064-77, eBioscience, San Diego, CA, and No. DY485-05, R&D Systems, respectively).

Mouse Primary Cell Isolation

Mouse bone marrow cells were obtained from the tibia and femur as described in a previous report (29). Briefly, leukocytes were flushed from the femur and tibia by a syringe with cell culture medium. The cell suspension was filtered through a 70-μm cell strainer, and cells were washed once with cell culture medium. Cells were treated with 1× red blood cell lysis buffer (No. 00-4333-57, ThermoFisher Scientific) at room temperature for 5 min. After red blood cell lysis, cells were washed twice with cell culture media and used as bone marrow leukocytes.

Neutrophils were isolated from bone marrow leukocytes with a neutrophil isolation kit. The purity of isolated neutrophils was checked by flow cytometry, and only neutrophils with >95% (CD11b⁺Ly-6G⁺) purity were used for experiments (Supplemental Fig. S2C).

Peritoneal macrophages were prepared as previously reported (30) with some modifications. Mice were intraperitoneally injected with 2.5 mL of 3% thioglycollate, and peritoneal leukocytes were harvested from the peritoneal cavity by washing with cell culture medium using a syringe 84–96 h after the injection. The cell suspension was centrifuged at 300 rcf for 5 min, and erythrocytes in the cell pellets were then lysed with 1× red blood cell lysis buffer at room temperature for 5 min. After a one-time wash with culture medium, macrophages were isolated by a peritoneal macrophage isolation kit. Macrophages with >90% (CD11b₊F4/80⁺) purity by flow cytometry were used for experiments (Supplemental Fig. S2D).

IC Formation

Rabbit IgG-mouse IgG ICs were prepared as previously reported (31) with some modifications. First, an immunodiffusion plate (No. PI31111, Thermo Fisher Scientific) was used to determine the equivalence point of normal rabbit IgG and mouse anti-rabbit IgG (H + L). After the preliminary experiments, it was decided to mix the same amounts (1:1 ratio in µg) of rabbit IgG and mouse anti-rabbit IgG (H + L) and incubate them at 4°C for 30 min. FITC-conjugated mouse anti-rabbit IgG (H + L) (No. 31584, Thermo Fisher Scientific) was also used for IC formation in some experiments. The final concentration of ICs was 200 µg (100 µg rabbit IgG and 100 µg mouse IgG) per milliliter in each sample.

Neutrophil Adhesion to ICs

A 96-well plate was coated with preformed ICs (200 µg/mL) at room temperature for 2 h. Neutrophils (1.0 × 10⁶ neutrophils/mL) were seeded on the 96-well plate and incubated at 37°C for 30 min. After a wash with PBS to remove nonadherent cells, adherent cells were fixed with 4% paraformaldehyde and stained with Flash Phalloidin Green 488. The adherent cells were counted under a microscope.

Flow Cytometry

Cell surface markers and intracellular cytokines were analyzed by a flow cytometer (LSR-II, BD Biosciences, Franklin Lakes, NJ) with the fluorochrome-conjugated monoclonal antibodies described in Reagents and Antibodies. Except for the samples analyzed for CD16/CD32, cells were initially incubated with FcR blocker (anti-CD16/32) at 4°C for 10 min. For surface marker staining, cells were incubated with antibody in PBS with 2% FBS at 4°C for 30 min. Intracellular cytokine staining was performed by a Cytofix/CytoPerm Kit. Briefly, cells stained with antibody for a surface marker were fixed and permeabilized at 4°C for 20 min. Cells were incubated with antibody for a cytokine at 4°C for 30 min. Dead cells were excluded by forward scatter, side scatter, and/or 7-ADD gating. A detailed design of the flow cytometry analysis and quality check of purified cells is provided in Supplemental Fig. S2. All data were analyzed by BD FACS Diva (BD Bioscience) or FlowJo (Tree Star, Ashland, OR).

IC Uptake Assay by Neutrophils and Macrophages

Neutrophils (1.0 × 10⁶) or macrophages (1.0 × 10⁶) were incubated with preformed FITC-labeled ICs at 37°C for 30 min. After the indicated incubation times shown in the figures, cells were harvested. After being washed with PBS and 2% FBS, cells were stained with several fluorochrome-conjugated antibodies for flow cytometry analysis. IC-uptaking cells were measured as FITC⁺ cells in the neutrophil (CD11b⁺Ly-6G⁺Ly-6C⁺) or macrophage (CD11b⁺F4/80⁺) gate.

Characterization of IC-Treated Immune Cells

For cytokine production, macrophages (1.0 × 10⁷ macrophages/mL) were treated with ICs at 37°C for 6 h. Protein transport was blocked with GolgiStop (No. 554724, BD Bioscience). Intracellular cytokine was detected using the Cytofix/CytoPerm Kit. For the reactive oxygen species (ROS) production assay, neutrophils (1.0 × 10⁷ neutrophils/mL) and macrophages (1.0 × 10⁷ macrophages/mL) were treated with ICs at 37°C for 30 min. Mitochondrial ROS production was monitored by CellROX green using flow cytometry. The production of IL-1β and TNF-α was measured by ELISA kits (Mouse IL-1 beta/IL-1F2 DuoSet ELISA, No. DY401-05, R&D Systems, and Mouse TNF alpha ELISA Ready-SET-Go Kit, No. 50–112-8954, ThermoFisher Scientific). All procedures followed the protocols in manuals supplied with the reagents.

Assay for Neutrophil Extracellular Traps

The formation of neutrophil extracellular traps (NETs) was analyzed as previously described (32, 33) with minor modifications. Briefly, after isolation of bone marrow neutrophils from a male WT or NeuACE mouse using a density gradient method (34), NET formation was induced with 1.0 × 10⁵ neutrophils seeded in 96-well clear flat-bottom plate. Unstimulated cells or cells added with the treatment of interest (5 µM ionomycin, Cayman Chemical, Immunocomplex) were incubated at 37°C for 3 h followed by DNA staining for 5 min at room temperature with SYTOX green (final concentration: 5 µM). The plate was then centrifuged at 400 g for 5 min, and three-quarters of the supernatant was carefully removed from each well to reduce background fluorescence. Images of each well (n = 5) were obtained on a Keyence BZ-9000 microscope at ×2 magnification. Total DNA area was automatically calculated using Keyence BioAnalyzer software. The DNA area normalized with the unstimulated group was calculated. Also, the fluorescent intensity of each well was measured with a plate reader (SpectraMax M2, Molecular Devices, Sunnyvale, CA) at excitation and emission wavelengths of 488 and 525 nm, respectively.

Western Blot Analysis

Whole kidney protein extracts in radioimmunoprecipitation assay (RIPA) buffer containing protease and phosphatase inhibitors (Cat. No. 78442, ThermoFisher Scientific) were denatured, resolved by SDS-PAGE, and transferred onto polyvinylidene difluoride membranes (Millipore Immobilon-FL, EMD Millipore, Billerica, MA). Membranes were blocked (Odyssey blocking buffer, Lincoln, NE) and then probed with each of the following specific primary antibodies: chemokine (C-X-C motif) ligand (CXCL)1 antibody (1:400, Cat. No. LS-C295614, Lifespan Biosciences, Seattle, WA), CXCL2 antibody (1:1,000, Cat. No. LS-C408070, Lifespan Biosciences), and CXCL5 antibody (1:1,000, Cat. No. LS-C212192, Lifespan Biosciences). After being washed, membranes were incubated with the appropriate HRP-labeled secondary antibody. β-Actin (1:5,000, Cat. No. A3854, Sigma-Aldrich) was used for a loading control.

Serum Complement C3 and C3b and Complement Activity of Recombinant ACE

Serum concentrations of complement C3 and C3b were measured by ELISA kits (GWB-7555C7, Genway Biotech, San Diego, CA; and MBS269569, MyBiosource, respectively). The Wieslab Complement System Screen (NC1372880, ThermoFisher Scientific) was used to measure classical, alternative, and lectin complement activity in recombinant human ACE (No. 790608, BioLegend).

Statistical Analyses

All statistical analyses were performed using Prism 7 (Graphpad Software, San Diego, CA). Data are presented as means ± SE. Comparison of two groups was performed using the Mann–Whitney test. Differences among more than two groups were compared by the Kruskal–Wallis test, followed by Dunn's multiple comparison test. For all tests, two-tailed P < 0.05 was considered statistically significant.

RESULTS

Overexpressed ACE in Neutrophils Ameliorates Glomerular Injury in IC-Mediated Crescentic GN

NTN was studied in NeuACE mice (characterized as provided in Supplemental Fig. S1) and WT mice. NTN is composed of two phases (18): an early heterologous phase and a late autologous phase. The heterologous phase is caused mainly by direct binding of rabbit IgG against glomerular antigens including GBMs, but there are no discrete IC deposits. First, we examined proteinuria in the heterologous phase. Twenty-four hours after NTS injection, both WT and NeuACE mice showed very mild proteinuria (WT mice at 24 h: 4.6 ± 0.3, NeuACE mice at 24 h: 4.2 ± 0, and naïve WT mice: 2.8 ± 0.2), with no significant differences between WT and NeuACE mice (Fig. 1A). The autologous phase develops 5–6 days after NTS injection and is characterized by heavy proteinuria and massive IC deposits, as well as by direct binding of anti-GBM antibodies. Seven days after GN induction, NeuACE mice showed less severe proteinuria than WT mice (WT male mice: 32.1 ± 2.4, WT female mice: 35.6 ±1.8, NeuACE male mice: 18.8 ± 2.4, and NeuACE female mice: 14.8 ± 3.5), indicating attenuated glomerular injury in NeuACE mice (Fig. 1B). There were no significant sex differences in the level of proteinuria in both WT mice and NeuACE mice. Histological analysis of WT mice documented significant glomerular injury with fibrinoid necrosis (a precursor lesion for crescents) and frank crescent formation in 74.2 ± 8.2% and 4.2 ± 1.1% of glomeruli, respectively (Fig. 1, C–E). In contrast, NeuACE mice presented with far less glomerular damage as indicated by a marked reduction in fibrinoid necrosis (7.3 ± 1.7%) and crescent formation (0.2 ± 0.1%). Neutrophils and macrophages are critical in the development and progression of GN (35, 36), and more glomerular neutrophils were detected by esterase staining as early as 2 h after injection of nephrotoxic serum in NeuACE mice versus WT mice (Fig. 1, F and G). However, by 24 h, the neutrophil number was reduced and no longer significantly different. In addition, no significant difference in the number of interstitial neutrophils was observed between WT and NeuACE mice both 2 h and 7 days after GN induction (Fig. 1, H and I). When immunohistochemistry for Mac-2 was used to quantify glomerular macrophages (37–41) at 7 days post-NTS injection, NeuACE mice had roughly half the number of such inflammatory cells compared with WT mice (Fig. 1, J and K). Thus, these data show a very significant reduction of glomerular damage from NTN in NeuACE mice.

Figure 1.

Overexpressed angiotensin-converting enzyme in neutrophils ameliorates glomerular injury in immune complex-mediated crescentic glomerulonephritis. A: proteinuria in the early phase of nephrotoxic serum nephritis (NTN), 24 h after nephrotoxic serum (NTS) injection. Each point represents data from one mouse [naïve wild-type (WT) mice: 2.8 ± 0.2, WT mice at 24 h: 4.6 ± 0.3, and NeuACE mice at 24 h: 4.2 ± 0.4]. *P < 0.05; ***P = 0.0009. NS, no significance. n = 15 for each group. B: proteinuria on day 7 of NTN (WT male mice: 32.1 ± 2.4, WT female mice: 35.6 ± 1.8, NeuACE male mice: 18.8 ± 2.4, and NeuACE female mice: 14.8 ± 3.5). **P < 0.005; ****P < 0.0001. n = 15 for each group. C–E: histological assessment of glomerular injury in NTN mice on day 7. n = 15 for each group. C: glomeruli with fibrinoid necrosis stained with periodic acid-methenamine silver (WT mice: 74.2 ± 8.2% and NeuACE mice: 7.3 ± 1.7%). ****P < 0.0001. D: the same assay was performed for crescents with periodic acid-Schiff stain (WT mice: 4.2 ± 1.1% and NeuACE mice: 0.2 ± 0.1%). **P < 0.003. E: representative image of glomeruli with or without fibrinoid necrosis and crescent formation. F: neutrophil infiltration into glomeruli from WT mice and NeuACE mice with NTN 2 or 24 h after NTS injection (0.35 ± 0.11 per glomerulus in the 2-h WT group vs. 0.80 ± 0.15 per glomerulus in the 2-h NeuACE group, ****P < 0.0001; 0.24 ± 0.08 per glomerulus in the 24-h WT group vs. 0.29 ± 0.03 per glomerulus in the 24-h NeuACE group, P = 0.93). n = 10 for each group. G: representative image of esterase-positive cells in the glomerulus 2 h after NTS injection. Arrows indicate neutrophils. H and I: interstitial neutrophils. H: esterase-positive cells in interstitial areas [50 high-power fields (HPF)/mouse, n = 10 mice for each group] were evaluated 2 h and 7 days after NTS injection. I: representative image of esterase-positive cells in the interstitium on day 7. Arrows indicate neutrophils. J and K: macrophage infiltration to glomeruli. J: Mac-2-positive cells in glomeruli from WT mice (n = 15) and NeuACE mice (n = 15) on day 7 (WT mice: 2.22 ± 0.21 and NeuACE mice: 0.97 ± 0.14). ***P = 0.0002. K: representative image of Mac-2-positive cells in the glomerulus. All scale bars = 25 µm.

The NTN model, particularly during the autologous phase, is of crescentic GN characterized by robust IC deposition and linear binding of IgG along GBMs, which shares the pathological features of human proliferative lupus nephritis and anti-GBM GN. Immunofluorescent microscopy showed that glomeruli from both groups of animals had abundant deposition of rabbit IgG (Fig. 2, A–C). However, there was significantly less granular deposition of glomerular rabbit and mouse IgG in NeuACE mice than was observed in the glomeruli of WT mice. These findings were verified by ultrastructural experiments that showed less subendothelial and mesangial ICs in glomeruli from NeuACE mice than in those from WT mice (Fig. 2, D and E).

Figure 2.

Lesser glomerular immune complex deposition in NeuACE mice. A: rabbit IgG and mouse IgG staining of kidney tissues from wild-type (WT) mice (n = 10) or NeuACE mice (n = 10) with nephrotoxic serum nephritis on day 7. Scale bars = 25 µm. B and C: staining of rabbit IgG (B) and mouse IgG (C) was evaluated semiquantitatively (rabbit IgG: 4 ± 0.0 in WT mice vs. 3.12 ± 0.24 in NeuACE mice, **P = 0.0031; mouse IgG: 3.65 ± 0.20 in WT mice vs. 2.40 ± 0.31 in NeuACE mice, **P = 0.0038). D and E: representative electron microscopy images. WT mice showed more abundant electron-dense immune complex deposits (*) in the glomerular capillary subendothelial space (D) and in the mesangial area (E) than NeuACE mice. Uranyl acetate and lead citrate stain was used. Original magnification: ×19,000. Scale bars = 2 µm. F: autologous murine IgG (anti-rabbit IgG) levels were determined by ELISA from serum samples from WT (n = 12) or NeuACE (n = 10) mice on day 7 of nephrotoxic serum nephritis. Titers in serial dilutions of serum are expressed in arbitrary units [optical density (OD) 450 nm] (×100 dilution: 0.82 ± 0.02 in WT mice vs. 0.85 ± 0.03 in NeuACE mice, P = 0.30; ×1,000 dilution: 0.57 ± 0.03 in WT mice vs. 0.63 ± 0.04 in NeuACE mice, P = 0.23; and ×10,000 dilution: 0.23 ± 0.02 in WT mice vs. 0.30 ± 0.03 in NeuACE mice, P = 0.12). G: recall assay to rabbit IgG in splenocytes. Splenocytes were isolated from WT mice (n = 6) or NeuACE mice (n = 6) after 10 days of immunization of rabbit IgG. Cells were incubated with or without rabbit IgG for 24 h. Supernatants were collected to measure interferon (IFN)-γ by ELISA (WT mice: 0.0 ± 0.0 pg/mL with no rabbit IgG vs. 314.2 ± 27.6 pg/mL with rabbit IgG, *P < 0.05; NeuACE mice: 0.0 ± 0.0 pg/mL with no rabbit IgG vs. 429.2 ± 27.8 pg/mL with rabbit IgG, ***P = 0.0007). H: the same assay as shown in G was done for IL-6 production (WT mice: 19.3 ± 0.75 pg/mL with no rabbit IgG vs. 320.7 ± 7.4 pg/mL with rabbit IgG, *P = 0.014; NeuACE mice: 23.0 ± 0.42 pg/mL with no rabbit IgG vs. 432.6 ± 16.1 pg/mL with rabbit IgG, *P = 0.0123).

In the NTN protocol, rabbit IgG priming induces mouse serum IgG against rabbit IgG; levels of this antibody were assessed in both NeuACE and WT mice and were found to be comparable (Fig. 2F). In addition, ex vivo stimulation of splenocytes from primed NeuACE mice with rabbit IgG showed robust expression levels of IFN-γ (Fig. 2G), a T helper (Th)1 cytokine (42), and IL-6 (Fig. 2H), an important cytokine for B cell development and antibody production (43). These results suggest that no significant defects exist at least in the humoral immunity of NeuACE mice. Thus, our data strongly suggest that the reduced amount of rabbit and mouse IgG in the glomeruli following the NTN protocol was due to other phenotypic features of NeuACE mice apart from the ability to generate an antibody response.

Increased IC Uptake in NeuACE Neutrophils

Given that deposition of glomerular ICs was reduced in NeuACE mice, we next compared the affinity and uptake of ICs by bone marrow neutrophils from NeuACE and WT mice. This was studied using two approaches. First, an adhesion assay was performed in which neutrophils from naive mice were seeded onto plates precoated with ICs. Adherent cells were then identified by staining with phalloidin-FITC and counted microscopically. NeuACE neutrophils showed a significantly greater number of adherent cells than WT neutrophils (Fig. 3A). In the second approach, neutrophils from naive mice were incubated with ICs preformed from rabbit IgG and fluorescently labeled mouse IgG for 30 min at 37°C. After being washed, cells were then analyzed by flow cytometry, which showed that NeuACE neutrophils (Fig. 3B), but not NeuACE macrophages (Supplemental Fig. S3A) or NeuACE monocytes (Supplemental Fig. S3D), ingested significantly more ICs than WT cells. This assay was not associated with cell death because at the end of the assay (30 min) ∼90% of both NeuACE and WT neutrophils were viable (Fig. 3C). However, a difference in viability was observed when the assay was extended from 30 to 120 min. At that point, the viability of NeuACE neutrophils was >80%, whereas WT cells demonstrated ∼64% viable cells. Neutrophil death can be associated with NET production (44). When NETs were measured in response to ICs, no significant degree of NETs was made by either NeuACE or WT neutrophils (Supplemental Fig. S4). A previous publication (15) indicated that ROS plays a major role against bacterial infection in NeuACE mice. Therefore, we measured ROS production by naive neutrophils following exposure to ICs (Fig. 3D). Greater than 90% of both NeuACE and WT cells produced ROS. When ROS production was quantitated by flow and expressed as mean fluorescent intensity, this analysis showed significantly more production of ROS by NeuACE neutrophils than WT cells but not NeuACE macrophages (Supplemental Fig. S3C). In addition, we measured IL-1β and TNF-α, major inflammatory cytokines (45, 46). There was a slight but significant increase in IL-1β production from neutrophils (Fig. 3E) but not from macrophages (Supplemental Fig. S3B). There was no significant difference in the production of TNF-α from neutrophils (Fig. 3F). These results suggested that the efficient removal of ICs in glomeruli by NeuACE neutrophils overcomes the proinflammatory effect of a transiently increased number of glomerular neutrophils.

Figure 3.

NeuACE neutrophils show enhanced immune complex (IC) uptake. A: neutrophil adhesion on IC-coated plates. Neutrophils were seeded on a plate precoated with ICs (mouse IgG and rabbit IgG) and incubated at 37°C for 30 min. Neutrophils were fixed with 4% paraformaldehyde. After being washed, cells were stained with Flash Phalloidin Green 488. The number of adherent neutrophils was counted under a microscope [without coating: 13.7 ± 0.8 cells/high-power field (hpf) in NeuACE mice and 12.7 ± 1.5 cells/hpf in wild-type (WT) mice, no significance (NS); with coating: 146.2 ± 5.1 cells/hpf in NeuACE mice and 101.2 ± 12.9 cells/hpf in WT mice, ***P = 0.0006]. n = 6 for each group. B: IC uptake assay in neutrophils. Neutrophils were incubated with preformed ICs (FITC-labeled mouse IgG and rabbit IgG) at 37°C for 30 min. After a wash, IC uptake by neutrophils (FITC-positive population in CD11b+Ly-6G+/hiLy-6Cdim/+gate) was detected by flow cytometry. IC-FITC mean fluorescent intensity (MFI) was calculated from the histogram (vehicle: 1.1 ± 0.1-fold MFI in NeuACE mice and 1.0 ± 0.0-fold MFI in WT mice, NS; with IC: 7.4 ± 0.2-fold MFI in NeuACE mice and 5.8 ± 0.1-fold MFI in WT mice, ****P < 0.0001). n = 8 for each group. C: live/dead assay in IC-treated neutrophils. Neutrophils were incubated with preformed ICs at 37°C for 0, 30, 60, and 120 min. The percentage of dead cells was detected in the neutrophil population by flow cytometry (naïve: 98.1 ± 0.7% in WT mice vs. 98.1 ± 0.5% in NeuACE mice, P = 0.98; 30 min: 97.8 ± 0.5% in WT mice vs. 98.2 ± 0.5% in NeuACE mice, P = 0.55; 60 min: 93.6 ± 1.7% in WT mice vs. 96.9 ± 0.5% in NeuACE mice, P = 0.09; 90 min: 63.9 ± 5.1% in WT mice vs. 83.3 ± 6.2% in NeuACE mice, *P = 0.04). n = at least 4 for each group. D: reactive oxygen species (ROS) production in IC-treated neutrophils. Neutrophils were treated with ICs at 37°C for 30 min. ROS production was detected by flow cytometry. ROS MFI was calculated from the histogram (vehicle: 0.6 ± 0.2-fold MFI in WT mice vs. 0.7 ± 0.2-fold MFI in NeuACE mice, NS; with IC: 7.4 ± 0.4-fold MFI in WT mice vs. 11.7 ± 0.5-fold MFI in NeuACE mice, ****P < 0.0001). n = 8 for each group. E: IL-1β production from IC-treated neutrophils (217.1 ± 11.0 pg/mL in WT mice vs. 265.4 ± 17.3 pg/mL in NeuACE mice, *P = 0.02). n = 21 for each group. F: TNF-α production from a similar experiment to that shown in E (78.3 ± 9.2 pg/mL in WT mice vs. 66.9 ± 7.5 pg/mL in NeuACE mice, NS). n = 15 for each group.

Increased Expression of CR1/2 and FcγRII/III Contributes to Effective Immune Complex Elimination by NeuACE Neutrophils

To investigate the mechanism underlying enhanced IC uptake by NeuACE neutrophils, we examined cell surface expression of CR1/2 and FcγRII/III (CD16/CD32) on both WT and NeuACE cells. When measured using flow cytometry, NeuACE neutrophils expressed increased amounts of both types of receptors (Fig. 4A). The difference was particularly marked for CR1/2 in neutrophils but not in monocytes or macrophages (Supplemental Fig. S5). In contrast, expression of FcγRI (CD64) by NeuACE neutrophils was equivalent to that of WT neutrophils. Furthermore, the use of antibodies to block CR1/2 and/or FcγRII/III attenuated IC uptake by both NeuACE and WT neutrophils (Fig. 4B), and it eliminated the increased IC uptake of NeuACE cells. In fact, treatment of cells with anti-CR1/2 nearly eliminated all cellular uptake of ICs.

Figure 4.

Complement C3b-mediated removal of immune complexes (ICs) is upregulated in NeuACE mice. A: receptor expression profile (CR1/CR2, FcγRII/III, and FcγRI) on neutrophils by flow cytometry [CR1/2: 1.4 ± 0.0-fold mean fluorescent intensity (MFI) in NeuACE mice vs. 1.0 ± 0.3-fold MFI in wild-type (WT) mice, ****P < 0.0001; FcγRII/III: 1.1 ± 0.0-fold MFI in NeuACE mice vs. 1.0 ± 0.0-fold MFI in WT mice, *P = 0.0265; FcγRI: 1.0 ± 0.0-fold MFI in NeuACE mice vs. 1.0 ± 0.0-fold MFI in WT mice, no significance (NS)]. n = 5 for each group. B: effects of CR1/2 blockade and/or FcγRII/III blockade on IC uptake in neutrophils. Preincubated neutrophils with anti-CR1/2 blocking antibody and/or anti-FcγRII/III antibody at room temperature for 10 min were incubated with ICs at 37°C for 30 min. IC uptake by neutrophils was measured by flow cytometry. The results are expressed as fold MFIs against the condition of WT neutrophils with anti-CR1/2 antibody and anti-FcγRII/III antibody (no blocking antibodies: 5.7 ± 0.1-fold MFI in WT mice vs. 7.3 ± 0.3-fold MFI in NeuACE mice, ****P < 0.0001; anti-FcγRII/III antibody only: 2.9 ± 0.1-fold MFI in WT mice vs. 3.0 ± 0.1-fold MFI in NeuACE mice; anti-CR1/2 antibody only: 1.4 ± 0.1-fold MFI in WT mice vs. 1.0 ± 0.1-fold MFI in Neu ACE mice, NS; anti-FcγRII/III antibody and anti-CR1/2 antibody: 1.2 ± 0.1-fold MFI in WT mice vs. 1.2 ± 0.1-fold MFI in NeuACE mice, NS). n = 5 for each group. C: expression levels of chemokine (C-X-C) ligand (CXCL)1, CXCL2, and CXCL5 in kidneys with nephrotoxic serum nephritis (NTN; day 7) from WT mice (n = 3) and NeuACE mice (n = 3) by Western blot. D: serum levels of complement C3 from WT mice (n = 16) and NeuACE mice (n = 13) by ELISA (1,683.0 ± 64.3 µg/mL in WT mice vs. 1,951.0 ± 160.7 µg/mL in NeuACE mice, NS). E: serum levels of complement C3b from WT mice (n = 16) and NeuACE mice (n = 13) by ELISA (18.3 ± 3.2 ng/mL in WT mice vs. 45.2 ± 6.1 ng/mL in NeuACE mice, ***P = 0.0003). F: effect of serum from WT mice and NeuACE mice on IC uptake by neutrophils. Neutrophils from WT mice or NeuACE mice were incubated with ICs in culture media supplemented with 5% serum from WT mice or NeuACE mice at 37°C for 30 min. IC uptake by neutrophils was detected by flow cytometry, and IC-FITC MFI was calculated from the analysis. Lane 1, 1.0 ± 0.0-fold MFI; lane 2, 1.3 ± 0.1-fold MFI; lane 3, 3.2 ± 0.2-fold MFI; lane 4, 3.9 ± 0.1-fold MFI; lane 5, 4.0 ± 0.2-fold MFI; lane 6, 4.7 ± 0.2-fold MFI. Lanes 1 vs. 2, *P < 0.05; lanes 3 vs. 4, **P < 0.01; lanes 5 vs. 6, **P < 0.01; lanes 3 vs. 5, **P < 0.01; lanes 4 vs. 6, **P < 0.01. n = 5 for each group. G: effects of angiotensin-converting enzyme inhibitor or bradykinin 2 receptor antagonist on CR1/2 and FcγRII/II of neutrophils. NeuACE mice were treated with ramipril (5 mg/kg/day, every day, orally), HOE-140 (100 µg/kg/day, every day, subcutaneously), or vehicle only for 3 wk and euthanized to collect neutrophils to analyze by flow cytometry. Expression of CR1/2 and FcγRII/III was expressed as MFI (CR1/2: 1.0 ± 0.1-fold MFI with vehicle, 0.6 ± 0.1-fold MFI with ramipril, and 0.8 ± 0.1-fold MFI with HOE-140, *P < 0.05 for ramipril vs. vehicle; FcγRII/III: 1.0 ± 0.0-fold MFI with vehicle, 0.8 ± 0.1-fold MFI with ramipril, and 1.0 ± 0.0-fold MFI with HOE-140, *P < 0.05 for ramipril vs. vehicle). n = 5 for each group. H: effects of ramipril, an angiotensin-converting enzyme inhibitor, or losartan, an angiotensin receptor blocker, on NTN mice. WT NTN mice and NeuACE NTN mice were treated with ramipril (10 mg/kg/day) or losartan (20 mg/kg/day) from day 2 to day 7 every day by oral gavage. Proteinuria was measured on day 7 (WT mice with vehicle: 35.3 ± 3.0; WT mice with ramipril 25.0 ± 2.9; WT mice with losartan: 20.7 ± 2.6; NeuACE mice with vehicle: 14.4 ± 2.3; NeuACE mice with ramipril: 9.7 ± 2.6; NeuACE mice with losartan: 9.0 ± 1.4). *P < 0.05; ***P = 0.0008. n = 10 for each group.

Recently, we reported that NeuACE neutrophils have a better migratory capacity (47). In NTN NeuACE mice, the major neutrophil chemokines (CXCL1, CXCL2, and CXCL5) in the kidneys were decreased (Fig. 4C) compared with NTN WT mice, whereas the chemokine receptors [chemokine (C-X-C motif) receptor (CXCR)2 and CXCR4] on neutrophils were comparable to WT mice (Supplemental Fig. S6), suggesting that increased FcγRs (4, 48) and CR1/2 are the major determinants of leukocyte influx, and the decreased chemokines are subsequent events.

It is known that ICs can be cleared by binding to C3b and, subsequently, the ICs-C3b-CR1 complex (49, 50). We examined the levels of complement C3 and C3b in the sera of naive NeuACE and WT mice (Fig. 4, D and E). By ELISA, levels of C3 were equivalent between the two strains. In contrast, NeuACE mice presented with levels of C3b that averaged more than twice those present in WT mice. Next, to investigate the contribution of serum C3b levels in the clearing of ICs, we performed IC uptake assays testing various combinations of mouse serum and neutrophils from naive NeuACE and WT mice (Fig. 4F). In this assay, flow cytometry data of IC uptake by WT neutrophils in the presence of heat-inactivated FBS was set as the control value (Fig. 4F, 1). In all instances, neutrophils from NeuACE mice took up significantly more ICs (Fig. 4F, 1 vs. 2, 3 vs. 4; and 5 vs. 6). However, there were marked differences depending on whether NeuACE or WT serum was used, with serum derived from NeuACE mice resulting in higher levels of IC uptake irrespective of whether the neutrophils tested were from WT or NeuACE mice (Fig. 4F, 3 vs. 5 and 4 vs. 6). Although these experiments were performed under in vitro conditions, these data imply that in NeuACE mice, increased surface expression of receptors such as CR1/2 in neutrophils and the increased presence of complement C3b in sera combine to increase neutrophil uptake of ICs. We also tested the effect of an ACE inhibitor on NeuACE neutrophil receptor expression. The elevated surface expression of CR1/2 and FcγRII/III was reduced by an ACE inhibitor (Fig. 4G). Bradykinin and angiotensin I are well-known physiological substrates of ACE (51), and we tested whether the increased expression levels of CR1/2 and FcgRII/III were mediated by bradykinin pathway using HOE-140, a bradykinin 2 receptor inhibitor (15, 24). However, HOE-140 did not reduce the expression level of both receptors. These results suggested that the increased expression of the receptors on neutrophils is mediated by enzymatic activity of ACE but not by the bradykinin pathway. Finally, in vivo, we examined whether the reduced glomerular injury in NTN NeuACE mice disappeared with the treatments of ramipril, an ACE inhibitor (20, 21), or losartan, an angiotensin receptor blocker (22, 23). In NeuACE mice, there were no significant differences in proteinuria among vehicle-only, ramipril, and losartan groups (Fig. 4H), although a mild trend of reduced proteinuria with ramipril or losartan was observed. On the other hand, ramipril or losartan treatment significantly reduced the proteinuria of NTN in WT mice. Furthermore, we measured complement activity of recombinant human ACE using a Complement System Screen kit with appropriate controls, but no significant complement activity was detected.

Taken together, the present study shows that elevated ACE in neutrophils reduces glomerular injury in a model of crescentic GN (Fig. 5) via 1) preactivation of complement C3 to generate C3b, 2) NeuACE neutrophil capture of locally formed ICs by increased CR1/2 and FcγRII/III, and 3) efficient elimination of ICs. This analysis indicates a novel role of neutrophilic ACE in GN and perhaps a new therapeutic direction in treatment approaches.

Figure 5.

Role of neutrophilic angiotensin-converting enzyme (ACE) in immune complex (IC)-mediated crescentic glomerulonephritis. The model depicts the ameliorated glomerular injury in NeuACE mice. Mouse IgG against rabbit IgG is produced upon immunization. Neutrophilic ACE contributes preactivation of complement C3 to generate serum C3b in NeuACE mice. Once nephrotoxic serum is injected, mouse IgG-rabbit IgG ICs are formed. ICs are efficiently captured by NeuACE neutrophils via increased surface CR1/2 and FcγRII/III and eliminated in the reticuloendothelial system, leading to reduced glomerular damage.

DISCUSSION

Glomerular injury in NTN was attenuated in the mice overexpressing ACE in neutrophils (NeuACE mice) revealing a previously unrecognized role of neutrophilic ACE in IC-mediated GN. NeuACE mice rapidly mobilize neutrophilic infiltration into the glomeruli with effective clearing of ICs resulting in significantly glomerular damage as measured by proteinuria and histology. Mechanistically, NeuACE mice have elevated serum complement C3b (a CR1 ligand) and NeuACE neutrophils have increased surface expression of CR1/2 and FcγRII/III, facilitating efficient elimination of locally formed ICs in NTN.

ACE is a zinc-dependent dicarboxypeptidase that plays a major role in blood pressure control by converting angiotensin I to angiotensin II (11). However, ACE expressed by myelocyte lineage cells also appears important for effective immune function during infection (15, 47), tumor growth (52), atherosclerosis (53), and Alzheimer’s disease (51). For example, NeuACE neutrophils have an enhanced oxidative response and antibacterial activity, thus endowing these mice with increased resistance to bacterial infection by methicillin-resistant Staphylococcus aureus (MRSA), Pseudomonas aeruginosa, and Klebsiella pneumoniae (15, 47). Mice expressing increased ACE in monocytes/macrophages (ACE10/10 mice) are remarkably resistant to melanoma due to the interaction of myeloid and lymphoid cells in adaptive immunity (52). Because of an increased ability of ACE-overexpressing macrophages to resolve the inflammatory challenge of elevated lipids, apolipoprotein E knockout ACE10/10 mice exhibit significantly less atherosclerosis compared with apolipoprotein E knockout mice (53) with WT levels of ACE. Finally, Alzheimer’s disease-prone mice with increased ACE expression in macrophages show reduced amyloid-β protein in the brain and retained cognitive function (51).

Here, we report the role of neutrophil ACE in GN. NTN is a well-known murine model of crescentic GN caused by anti-GBM antibody as well as ICs formed in situ in glomeruli (18). NTN in mice is similar in phenotype and pathology to human anti-GBM GN and proliferative lupus nephritis. Neutrophils and macrophages are the major effector cells in GN (54, 55). In NTN, neutrophils are recruited to glomeruli within a few hours after nephrotoxic serum challenge (56), whereas macrophages infiltrate into glomeruli at about day 7 (57). NeuACE neutrophils efficiently eliminate ICs in the glomeruli, leading to less glomerular damage despite NeuACE neutrophils having a basically proinflammatory phenotype with somewhat more ROS and inflammatory cytokine production.

There are two phases of NTN (18): a heterologous phase and an autologous phase. The initial heterologous phase begins within 24 h after NTS injection of rabbit IgG against glomerular antigens, and glomerular injury is caused by direct binding of nephrotoxic IgG. The glomerular injury is characterized by unremarkable light microscopic morphology, mild proteinuria, and linear binding of rabbit IgG, but there is no or at most minimal IC deposition by immunofluorescence study. The autologous phase develops 5–6 days after nephrotoxic serum injection and is characterized by severe crescentic GN, heavy proteinuria, and massive IC deposition (granular rabbit and mouse IgG). The severity of glomerular injury is correlated with the amount of IC deposition in NTN with the heterologous phase having mild proteinuria and the autologous phase having heavy proteinuria. Here, we confirmed minimal proteinuria (Fig. 1A) in the heterologous phase (24 h after NTS injection) of both WT and NeuACE mice with NTN. In contrast, our data clearly demonstrate that in the autologous phase, NeuACE mice have less IC deposits and less proteinuria compared with WT mice with abundant IC deposition.

In previous analyses of bacterial infection (15) and tumor growth (52), increased disease resistance was dependent on ACE activity and was reverted to a WT phenotype with an ACE inhibitor but was independent of angiotensin II and thus not affected by the angiotensin II receptor antagonist losartan. Thus, in the NTN model, we expected that the ACE inhibitor would increase the proteinuria of NeuACE mice to that of WT mice with vehicle. However, the data showed no significant difference among NeuACE mice treated with vehicle, ramipril, or losartan (Fig. 4H). These results suggest that ACE acts as a double-edged sword for angiotensin II-mediated proinflammatory effects and anti-inflammatory function via increased C3b-CR1/2 and FcγRII/III. Ramipril likely blocked both proinflammatory and anti-inflammatory functions of ACE, and therefore we could not observe a significant change of glomerular injury in NeuACE mice.

Various insults, including infection and inflammation, increase ACE expression in neutrophils (15). For example, bacterial infection increased ACE expression in neutrophils from the blood and spleen. In our NTN model, we have not investigated all aspects of ACE biology including what is the precise peptide responsible for the NeuACE phenotype or precisely how ACE regulates C3b-CR1/2 and FcγRII/III expression. Nonetheless, our findings could potentially lead to clinical application in the future via either small molecules that enhance neutrophil ACE expression or new ACE substrates that upregulate C3b, neutrophil CR1, and neutrophil FcγRs.

In summary, NeuACE mice exhibit ameliorated glomerular injury in IC-mediated crescentic GN. Neutrophils from these mice efficiently eliminate focally formed ICs via increased expression of CR1/2 and FcγRII/III together with elevated serum complement C3b. This study identifies a novel role of neutrophil ACE as renoprotective in IC-mediated GN.

Perspectives and Significance

NeuACE mice exhibit ameliorated glomerular injury in IC-mediated crescentic GN. Neutrophils from these mice efficiently eliminate focally formed ICs via increased expression of CR1/2 and FcγRII/III together with elevated serum complement C3b. This study identifies a novel renoprotective role of neutrophil ACE in GN.

SUPPLEMENTAL DATA

Supplemental Figs. S1–S5: https://doi.org/10.6084/m9.figshare.20353167.v1.

GRANTS

M.Y. was supported by supported by American Heart Association Scientist Development Grant 17SDG33660947, National Center for Advancing Translational Sciences University of California-Los Angeles Clinical and Translational Science Institute (CTSI) Grant KL2TR001882, and a Cedars-Sinai CTSI Clinical Scholars grant. K.E.B. was supported by National Institutes of Health Grants R01AI134714-04, P01HL129941-05, and R01AI164519-01. N.N. and K.S. were supported by National Institutes of Health Grant R01HL130353-01.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.S., N.T., D-Y.C., N.N., H.N., K.S., M.A., K.E.B., and M.Y. conceived and designed research; S.S., N.T., D-Y.C., N.N., D.N.L., E.A.B., and M.Y. performed experiments; S.S., N.T., D-Y.C., H.N., K.S., M.A., K.E.B., and M.Y. analyzed data; S.S., M.Y., and N.T. interpreted results of experiments; S.S., M.Y., and N.T. prepared figures; M.Y. drafted manuscript; M.Y. edited and revised manuscript; S.S., N.T., D-Y.C., N.N., H.N., D.N.L., E.A.B., K.S., M.A., K.E.B., and M.Y. approved final version of manuscript.