C57BL/6 background Lyn^-/- mice (L^-/-) (20) and Itgam^tm1Myd mice (Itgam^-/-, M^-/-) (23) were crossed to generate Lyn^-/-Itgam^-/- double deficient mice (LM^-/-) and confirmed by genotyping. Mice were bred under specific pathogen-free conditions at the Monash University Animal Research Platform (Clayton, Australia) and housed at the Monash University Intensive Care Unit (Melbourne, Australia) for long-term studies. C57BL/6 control mice were obtained from Monash Animal Services (Clayton, Australia). Research was performed in accordance with the Australian Code for the Care and Use of Animals for Scientific Purposes and in agreement with NHMRC animal welfare guidelines. Approval for animal experimentation was obtained from the Alfred Research Alliance (E/1377/2013/M, E/1688/2016/M) and Monash Medical Centre B (2018/19) animal ethics committees.

Single cell suspensions of spleen were prepared by extrusion. Saline-perfused kidneys were digested in RPMI + 0.1 mg/ml Liberase TL (Roche, 5401020001) + 0.1 mg/ml DNAse I at 37°C then mechanically dissociated to form a single cell suspension. Cells were treated with ACK red blood cell lysis buffer, filtered, counted and Fc receptors were blocked with anti-FcγRII/III F(ab’)₂ (2.4G2, in-house). Cells were then stained with fluorophore-conjugated mAbs and acquired on an LSRFortessa (BD Biosciences) with FACSDiva™ software (BD Biosciences) and analyzed using FlowJo Software (BD Bioscience). Dead cells were excluded based on uptake of propidium iodide or FluoroGold. Splenic cell populations were defined as: erythroblasts: Ter119⁺CD71^±; monocytes CD115⁺Ly6G^-; neutrophils: Ly6G⁺CD115^-; CD4⁺ T cells: CD4⁺CD8α^-; CD8⁺ T cells: CD8α⁺CD4^-. B cells: B220⁺CD138^low; plasma cells: B220^low-midCD138^high. Monocytes were further classified as conventional/inflammatory: Ly6C⁺CD62L⁺; non-conventional/patrolling: Ly6C^-CD62L^-. CD4⁺ and CD8⁺ T cells were further classified as effector cells: CD44⁺CD62L^-. B cells were further subdivided as follicular: CD23⁺CD21^low; marginal zone: CD23^-CD21⁺; or transitional: CD23^-CD21^-. Kidney lymphocytes were pre-gated CD45⁺CD11b^-CD11c^- and defined as: γδ T cells: CD3ϵ⁺γδTCR⁺; CD4⁺ T cells: γδTCR^-CD3ϵ⁺CD4⁺; CD8⁺ T cells: γδTCR^-CD4^-CD3ϵ⁺CD8α⁺; B cells: γδTCR^-CD4^-CD8α^-CD19⁺. Kidney granulocytes were pre-gated CD45⁺F4/80^-CX3CR1^- and defined as: eosinophils: SiglecF⁺Ly6G^-; neutrophils: Ly6G⁺SiglecF^-. Kidney mononuclear phagocytes were pre-gated CD45⁺F4/80⁺CX3CR1⁺ and defined as: resident macrophage-like phagocytes: CD11c⁺Ly6C^-; inflammatory macrophages: CD11c⁺Ly6C⁺. Kidney endothelium was defined as CD45^-CD326^-CD31⁺. Absolute cell numbers from spleen were determined from total cell counts and cell proportions identified by flow cytometry; cell numbers from kidney were determined by flow cytometry gated event count outputs and proportion of whole tissue acquired. Expression of activation and phenotypic markers (CD11b, CD62L, FcγRII/III, IgM) were determined by normalizing the geometric mean fluorescence intensity (gMFI) by dividing the value of each sample by the mean value of the C57BL/6 group within each experiment, allowing pooling of data from multiple experiments. Negative expression controls for activation markers were derived from gating on the population of C57BL/6 splenocytes that do not express the marker.

Kidney histopathology was conducted as previously described (22). Briefly, formalin-fixed, paraffin-embedded sections of 3 µm thickness were stained with hematoxylin and eosin or Periodic Acid-Schiff (PAS) for histopathological analysis. PAS-stained slides were imaged by Aperio Scanscope AT Turbo (Leica Biosystems). Mean glomerular cross-sectional area was quantified by polygon tracing using ImageJ software (NIH, 1.52d). Mean disease score criteria were assessed blinded as follows; 0 – Normal glomerular cellularity and morphology; 1 – Mild cellular expansion and/or early-stage disrupted glomerular morphology which includes lobularity and/or Bowman’s space enlargement; 2 – Advanced cellular expansion with distinct disruption to glomerular morphology, pyknosis and/or karyorrhexis; 3 – Severe cellular expansion/consolidation with advanced lobularity or fragmentation and/or enlargement of the Bowman’s space with cellular infiltration (with or without crescent formation) and/or peri-glomerular cellular expansion and/or evidence of segmental necrosis; 4 – End-stage glomerular destruction, progressive or complete loss of cellularity and/or distinct glomerular morphology. At least 30 glomeruli were assessed for each kidney.

Ageing cohorts of sex-matched Lyn^-/- (n=18) and Lyn^-/-Itgam^-/- (n=19) mice received daily condition checks and were euthanized when their condition had deteriorated to predefined ethical checkpoints set by researchers in consultation with veterinary staff. Lifespan in days were recorded at these time-points and graphed as a Kaplan-Meier survival curve.

Detection of autoantibodies was conducted as previously described (22). Briefly, 96-well plates (Maxisorp, Nunc) were coated with calf thymus DNA (Sigma-Aldrich). After incubation of serum samples and reference serum, goat anti-mouse IgG(H+L)-HRP (Southern Biotech) detection antibody followed by TMB chromogenic substrates A and B (BD Biosciences) elicited a colorimetric change which was recorded by a MultiSkan GO microplate spectrophotometer (Thermo Fisher Scientific). Relative titers were determined by plotting sample optical density against the reference serum-derived standard curve.

RNA was extracted from frozen kidney tissue by trizol and chloroform separation with genomic DNA removal by DNA-free DNA removal kit (Thermo Fisher). cDNA was generated by FireScript RT kit (Solis Biodyne) as per manufacturer’s instructions. RT-PCR was performed with validated primers for Gapdh (housekeeping control) (24), Il1b (25), Tnfa (Fwd 5’ CCCTCACACTCAGATCATCTTCT 3’, Rev 5’ GCTACGACGTGGGCTACAG 3’), Ccl2 (26), Cxcl1 (27) and PowerUp SYBR Green Master Mix (Applied Biosystems) for 40 cycles under single-plex conditions (QuantStudio 6, Life Technologies). cDNA-specific amplification was ensured by inclusion of reverse transcriptase-negative and template-negative controls. Samples were run in triplicate and cycle threshold values were determined by automatic threshold analysis (QuantStudio Software), with the 2^-ΔΔCt method used to calculated relative gene expression from the average threshold of each sample, relative to the average of the C57BL/6 group.

Formalin-fixed paraffin-embedded kidney sections (3 µm) underwent antigen-retrieval in DAKO antigen-retrieval buffer (Agilent). Sections were blocked with PBS/1% BSA, followed by incubation with rat anti-mouse CD45-biotin, then anti-rat-HRP secondary antibody (both in PBS/0.2% BSA). Chromogenic substrate (DAKO diaminobenzidine chromogen solution; Agilent) was added to sections followed by counter-staining with hematoxylin and coverslip mounting with DPX (Sigma-Aldrich). Sections were imaged using an Olympus BX-51 light microscope at 20x objective magnification (Olympus Australia), capturing at least six images of each kidney cortex. Images were analyzed using ImageJ software by isolating the red channel and setting the color threshold to 120 to visualize punctate staining. CD45+ staining was quantified using the polygon tool to trace each glomerulus and measuring the area fraction. A minimum of 30 glomeruli were analyzed per kidney, with the mean taken to represent each sample.

Saline-perfused kidneys were fixed in PLP fixative (PBS + 1.37%w/v L-lysine + 2%w/v paraformaldehyde + 0.22%w/v sodium periodate), infused with 30%w/v sucrose solution then embedded and snap frozen in OCT. 7 μm kidney sections were cut and mounted on SuperFrost Plus slides, rehydrated in PBS, blocked (4% BSA + 0.05% sodium azide in PBS) then stained with 20 μg/mL goat anti-mouse IgG-AF488 (In vivogen, A11029). Stained sections were mounted with antifade (4%w/v n-propyl gallate in 90% glycerol + 10% PBS), cover-slipped and imaged using the Nikon Eclipse TE2000-U Inverted Fluorescence Microscope with DS-Ri2 camera and 470/40 filter. IgG deposition was quantified in a blinded fashion with ImageJ FIJI, using the polygon tool to trace the glomerular tissue and measuring the mean gray value. A minimum of 30 glomeruli were analyzed per kidney, with the mean taken to represent each sample.

Glomerular trafficking of neutrophils and Ly6C⁺ monocytes was assessed by intravital multiphoton microscopy using a modification of a previously published technique (28–30). In brief, mice were anesthetized with ketamine hydrochloride (150 mg kg^-1)/xylazine hydrochloride (10 mg kg^-1), a catheter was inserted into the jugular vein, and the left kidney exteriorized via a dorsal incision. The intact kidney was immobilized in a heated well incorporated in a custom-built stage, bathed in normal saline and cover-slipped. The renal microvasculature was visualised using an FVMPE-RS multiphoton microscope (Olympus Australia, Notting Hill, Vic.), equipped with a 25X 1.05 NA water-immersion objective and an InSight X3 laser (Spectra-Physics, Milpitas, CA) tuned to 800 nm. Two recordings of superficial glomeruli were made for 30 min in each mouse, recording images every 30 s, collecting z-stack images of ~100 µm depth at a step-size of 3 μm. Emitted fluorescence was collected in non-descanned detectors with 410-455 nm, 495-540 nm and 575-645 nm emission filters. The glomerular microvasculature was labelled with AlexaFluor (AF) 594-anti-CD31 (clone MEC 13.3, BioLegend, San Diego, CA; 2 μg), adherent neutrophils with AF488-anti-Ly6G (1A8, eBioscience, Scoresby, Vic.; 2 μg) and Ly6C⁺ monocytes with AF405-anti-Ly6C (HK1.4, Novus; 3 μg), all administered i.v. immediately prior to imaging. Glomeruli were located on the basis of anti-CD31 staining, and the adhesion (cells/glomerulus/30 min) and duration of intraglomerular retention (dwell; min) determined for neutrophils and Ly6C⁺ monocytes using Imaris (Bitplane, Zurich).

Statistical significance was determined using Mann-Whitney non-parametric U test (for two groups) or Kruskal-Wallis H test followed by Dunn’s multiple comparisons test (for three groups). The Itgam^-/- group, acting as a reference control, was excluded from multiple comparison tests. To confirm an intermediate phenotype, specific comparisons between the Lyn^-/- and C57BL/6 or Lyn^-/-Itgam^-/- groups were also conducted using Mann-Whitney non-parametric U test. Horizontal bars represent median ± IQR where the data are displayed on a linear scale, or geometric mean ± geometric SD where logarithmic data are displayed. Significance by Mann-Whitney test (represented by #) and Dunn’s multiple comparisons test (represented by *) is denoted by p > 0.05 (not significant) ns or not stated, p < 0.05 # or *, p < 0.01 ## or **, p < 0.001 ### or ***, p < 0.0001 #### or ****. Significance of the survival study was determined by log-rank (Mantel-Cox) and Gehan-Breslow-Wilcoxon tests where p < 0.05 is significant. Statistical analysis was conducted by GraphPad Prism software (version 9.0.1).

Given the observations of increased susceptibility to SLE in individuals bearing ITGAM SNPs, we first determined whether deficiency of CD11b in mice was sufficient to drive changes in the immune system consistent with development of autoimmunity. Itgam^-/- mice were assessed at 36 weeks of age, a time point where autoimmune disease development is severe in lupus-prone Lyn^-/- mice (22, 31). Aged Itgam^-/- mice exhibited splenomegaly (Figure 1A), due to expansion of splenic erythroblasts, monocytes, and neutrophils (Figures 1B–D). While CD4⁺ T cells, CD8⁺ T cells and B cells were unchanged (Figures 1E–G), plasma cells were significantly expanded in Itgam^-/- mice (Figure 1H), although this correlated with only slight increases in autoantibody titers (Figure 1I). Histopathological assessment of kidneys revealed healthy glomeruli with no indication of cellular or morphological changes or enlargement of glomeruli in 36-wk old Itgam^-/- mice (Figure 1J). These findings indicate that while some key effector immune cell populations are expanded in 36-week-old Itgam^-/- mice, no overt autoimmunity or immunopathology is evident at this time.

Lupus-prone Lyn^-/- mice exhibit intrinsic CD11b dysregulation, with neutrophils showing upregulated CD11b from a pre-disease age (12 weeks) and throughout early- (24 weeks) and late-stage (36 weeks) disease (Supplementary Figure 1A) (22), suggesting that CD11b may contribute to early and ongoing disease processes. Therefore, to establish whether CD11b influences the development of inflammation and autoimmune disease on an autoimmune-susceptible genetic background, Lyn^-/-Itgam^-/- mice (LM-/-) were generated and assessed. Initial survival studies revealed that Lyn^-/-Itgam^-/- mice had significantly poorer survival (median 332 days) compared to Lyn^-/- mice (median 414.5 days) (Figure 2A). Histopathological analysis of kidneys at 36 weeks of age, where disease is typically maximal in Lyn^-/- mice (22), showed similarly severe glomerular damage in both Lyn^-/- and Lyn^-/-Itgam^-/- mice (Supplementary Figure 1B). However, assessment of renal pathology at 24 weeks of age, which captures early and progressing disease stages, showed that Lyn^-/-Itgam^-/- mice already exhibited significantly advanced glomerular enlargement (median area 4885 µm, IQR 4198-5849), lobularity and cellular expansion and periglomerular dysplasia (median disease score 2.029, IQR 1.676-2.612) well beyond the mild disease seen in Lyn^-/- mice (median glomerular area 3590 µm, IQR 3235-4208; median disease score 0.894, IQR 0.600-1.611) (Figures 2B–D). Systemic indices of autoimmunity and inflammation were also assessed, and at 24 weeks of age, autoantibody titers were significantly elevated in Lyn^-/-Itgam^-/- mice compared to Lyn^-/- mice (Figure 2E), and both splenomegaly (Figure 2F) and lymphadenopathy (Figure 2G) were pronounced in Lyn^-/-Itgam^-/- mice only. Collectively, these findings indicate that CD11b-deficiency accelerates systemic autoimmunity and renal disease progression in Lyn^-/- mice.

To assess the cellular impacts of CD11b-deficiency, flow cytometry was performed on splenocytes. Lyn^-/- mice with late-stage disease typically present with expanded splenic myeloid cell populations (22, 32), however at 24 weeks, neutrophils (Ly6G+CD115-) were only expanded in Lyn^-/-Itgam^-/- but not Lyn^-/- mice (Figures 3A, B), with both groups exhibiting downregulation of CD62L (Figure 3C). CD11b-deficiency further exacerbated expansion of splenic monocytes (CD115+Ly6G-) in Lyn^-/- mice (Figures 3A, D), with both Lyn^-/- and Lyn^-/-Itgam^-/- mice exhibiting a similar skewing toward CD62L-negative patrolling monocytes (Figure 3E). Interestingly, expression of FcγRII/III on these patrolling monocytes, which has implications for clearance of immunogenic cellular-debris and immune complexes, was progressively downregulated in Lyn^-/- mice, yet more markedly lost on Lyn^-/-Itgam^-/- cells (Figure 3F). Together, this shows that CD11b-deficiency in Lyn^-/- mice further promotes myeloid cell dysregulation.

Assessment of the T cell compartment revealed that Lyn^-/- mice displayed a typical age-associated T cell deficit, however, in 24-week-old Lyn^-/-Itgam^-/- mice, this was observed only in the CD8+ compartment and not the CD4+ compartment (Figures 3G-I). Skewing of the CD4+ T cell population towards effector cells (CD44^hiCD62L^-), which is typical in Lyn^-/- mice with disease progression (22), was further enhanced in Lyn^-/-Itgam^-/- mice (Figures 3J, K), with a similar trend for CD8+ effector T cells (Figure 3L), indicating that CD11b-deficiency can further promote T cell activation phenotypes in Lyn^-/- mice.

Assessment of the B cell compartment revealed that Lyn^-/-Itgam^-/- mice exhibited splenic B lymphopenia (Supplementary Figures 2A, B) and maturational defects (Supplementary Figure 2C) typical of Lyn^-/- mice. Similarly, plasmacytosis was pronounced in Lyn^-/-Itgam^-/- mice (Supplementary Figures 2A, D), with a typical Lyn^-/- phenotype; hyper-IgM and augmented FcγRII expression (Supplementary Figures 2E, F). Together, these data indicate that loss of CD11b has no overt influence on the intrinsic B cell and plasma cell defects in Lyn^-/- mice.

Given that CD11b-deficiency accelerates autoimmunity and lupus nephritis in Lyn^-/- mice (Figure 2), we next examined the inflammatory responses in the kidney of 24-week-old mice. Flow cytometry was used to identify and quantify kidney immune cell populations (Supplementary Figure 3A). Total CD45+ leukocyte numbers in Lyn^-/- kidneys were elevated above those in healthy C57BL/6 mice with a trend towards further expansion in Lyn^-/-Itgam^-/- mice (Supplementary Figure 3A). In comparison to healthy controls, Lyn^-/- mice exhibited an expansion of γδ, CD4+ and CD8+ T cell subsets (Figures 4A–C), with almost no detection of intrarenal B cells (Figure 4D). A similar expansion in kidney lymphocytes was observed in Lyn^-/-Itgam^-/- mice, with the notable further enrichment of γδ T cells, and marked B cell infiltration (Figures 4A–D). Eosinophils were barely detected in kidneys across all groups (Figure 4E) while neutrophils were equally expanded in both Lyn^-/- and Lyn^-/-Itgam^-/- kidneys (Figure 4F). Numbers of kidney-resident macrophage-like phagocytes (gated as CD45+F4/80+CX3CR1+CD11c+Ly6C-) were similar across all groups (Figures 4G, H), while infiltrating inflammatory macrophages (gated as CD45+F4/80+CX3CR1+CD11c+Ly6C+), were enriched in Lyn^-/- kidneys but significantly further expanded in Lyn^-/-Itgam^-/- kidneys (Figures 4G, I). Together, these data indicate that the absence of CD11b enhances immune cell infiltration in Lyn^-/- kidneys.

To examine this further, local inflammatory cytokine production in the kidney was measured by qRT-PCR. The inflammatory profile of the kidney in Lyn^-/- mice was typified by increases in the proinflammatory cytokines Il1 and Tnfa and the monocyte-recruiting Ccl2, and neutrophil-recruiting Cxcl1 chemokines (Figure 4J). This was also observed in Lyn^-/-Itgam^-/- kidneys, yet with further significant increases in the expression of Tnfa and Ccl2 above Lyn^-/- mice (Figure 4J). These observations provide evidence that loss of CD11b leads to enhanced inflammatory responses in the kidneys of Lyn^-/- mice.

Given that Lyn^-/-Itgam^-/- mice exhibited increased immune cells in the kidney which counters our previous observations that CD11b is required for the recruitment of cells to the kidney in acute inflammatory settings (28), we investigated renal leukocyte trafficking in Lyn^-/-Itgam^-/- kidneys at 24 weeks of age. Immunohistochemistry staining of CD45+ leukocytes revealed that infiltration in the kidney was largely focused to the glomeruli, or peri-glomerular areas with notable tissue dysplasia, which was most frequently observed in Lyn^-/-Itgam^-/- mice (Figure 5A). When quantified, Lyn^-/- mice exhibited a small increase in glomerular leukocyte infiltration at this time-point, however this was elevated significantly further in Lyn^-/-Itgam^-/- mice (Figure 5B). Glomerular immune-complex deposition assessment showed similar elevation in both Lyn^-/- and Lyn^-/-Itgam^-/- kidneys (Supplementary Figure 4), indicating that this is not driving the increased kidney leukocyte infiltration observed in Lyn^-/-Itgam^-/- mice.

Intravital multiphoton microscopy was employed to visualize leukocyte recruitment and interaction with the vasculature in the glomeruli of 24-week-old mice. The number of adherent Ly6G+ neutrophils in glomerular capillaries did not significantly differ across all groups (Figures 5C, D, Supplementary Video 1). Neutrophil retention (dwell time), which serves as a sensitive readout of the degree of glomerular inflammation (28–30), was mildly elevated in Lyn^-/- kidneys compared to C57BL/6 kidneys, yet significantly further increased in Lyn^-/-Itgam^-/- mice (Figure 5E, Supplementary Video 1). Intraglomerular adhesion and retention of Ly6C+ inflammatory monocytes, which are less inclined to patrol the vasculature but can give rise to inflammatory kidney macrophages (33), was rare in C57BL/6 and Itgam^-/- mice, yet was similarly enhanced in both Lyn^-/- and Lyn^-/-Itgam^-/- kidneys (Figures 5F–H, Supplementary Video 2). Together these findings demonstrate that in Lyn-deficient mice, CD11b is not required for intraglomerular adhesion of neutrophils and monocytes, and that the combined absence of Lyn and CD11b leads to increased neutrophil retention in glomeruli.

To further explore the mechanisms of inflammatory leukocyte trafficking, the induction of adhesion and trafficking molecules in the kidney was assessed by flow cytometry. On CD31+ renal microvascular endothelial cells, expression of VCAM-1 was ubiquitous in all groups, however ICAM-1 was significantly upregulated in Lyn^-/- and Lyn^-/-Itgam^-/- kidneys, albeit similarly in both groups (Supplementary Figure 3B), indicating comparable activation of the renal endothelium in Lyn^-/- and Lyn^-/-Itgam^-/- mice. Assessment of the β2-pairing integrin family on renal immune cells revealed high surface expression of CD11a and an absence of CD11c on kidney neutrophils, regardless of CD11b expression (Supplementary Figure 3C). Ly6C+ inflammatory kidney macrophages (which are gated CD11c+), exhibited a marked upregulation of CD11a, dependent on loss of Lyn but not Itgam (i.e., on Lyn^-/- and Lyn^-/-Itgam^-/- cells) (Figures 5I, J), and this phenotype was also observed in the peripheral precursor Ly6C+ monocytes (Figure 5K). Altogether, these findings demonstrate that during chronic inflammation, CD11b is dispensable for leukocyte recruitment to the diseased kidney and suggest that CD11a is the dominant trafficking integrin in Lyn^-/- mice.