Lrrk2 G2019S mutation incites increased cell-intrinsic neutrophil effector functions and intestinal inflammation in a model of infectious colitis

Abstract

Parkinson's Disease (PD) is a neurodegenerative disorder often preceded by gastrointestinal dysfunction. Mutations in leucine-rich repeat kinase 2 (LRRK2) are known risk factors for both PD and inflammatory bowel disease (IBD), suggesting a link between PD and the gastrointestinal tract. Using single-cell RNA-sequencing and spectral flow cytometry, we demonstrated that the Lrrk2 Gly2019Ser (G2019S) mutation is associated with an increased neutrophil presence in the colonic lamina propria during Citrobacter rodentium infection. This concurred with a Th17 skewing, upregulated Il17a, and greater colonic pathology during infection. In vitro experiments showed enhanced kinase-dependent neutrophil chemotaxis and neutrophil extracellular trap (NET) formation in Lrrk2 G2019S mice compared to wild-type counterparts. Our results add to the understanding of LRRK2-driven immune cell dysregulation and its contribution to PD, offering insights into potential biomarkers for early diagnosis and intervention in PD.

Fig. 1. Lrrk2 G2019S mutation does not interfere with C. rodentium colonization and clearance.

A Experimental design. Male and female Lrrk2 G2019S and WT mice (male-to-female ratio = 0.67) were gavaged once with ~1 × 10⁹ colony-forming units (CFUs) of C. rodentium. Bacterial fecal shedding was quantified on days 4, 7, or 8, 12, 21, and 28 after infection. B CFUs per g of feces. Data are represented as mean ± SD and analyzed by two-way ANOVA with Sidak post-test. One representative of two independent experiments is presented. n = 5 mice per group. C Comparison fecal CFU C. rodentium burden in WT and G2019S mice on day 7 of infection, separated by sex. Data are represented as mean ± SD and analyzed by t-test. Ten independent experiments are presented. n = 64–65 mice per group. Dashed lines indicate the limit of detection (LOD).

Fig. 2. Single cell RNA sequencing (scRNAseq) and flow cytometry show increased neutrophil presence in colonic lamina propria of Lrrk2 G2019S mice following infection.

Male and female Lrrk2 G2019S and WT mice were gavaged once with ~1 × 10⁹ CFUs of C. rodentium, and colons were harvested. Immune cells were isolated for scRNAseq and flow cytometry. A Unsupervised clustering of immune cells in WT or G2019S uninfected and infected mice. The scRNAseq datatset contains 13,892 cells total sequenced at a depth of ~20,000 genes per cell, depicted on respective UMAP plots. B Dotplot depicting identifying markers used to annotate each immune cluster based off literature. C Dotplot depicting average Lrrk2 expression across all immune cell clusters, split by condition. D Proportional analysis conducted through permutation test comparing G2019S-infected and WT-infected conditions. Significance cut off: FDR < 0.05. scRNAseq was obtained from pooled conditions of n = 3 mice per group (male-to-female ratio = 0.5). One independent experiment is presented. E Flow cytometry results of immune cell clusters projected onto t-SNE plots. F Percent Ly6G+ neutrophils isolated from CD45+ immune cells of the colonic lamina propria. Data are represented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. n = 5–6 mice per group (male-to-female ratio = 0.9). ***p < 0.001. G MPO enzymatic assay results of colonic tissue sample measured through O.D. and normalized over gram of colonic tissue. Data are represented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. n = 4-6 mice per group (male-to-female ratio = 0.42). *p < 0.05. One representative of two independent experiments is presented.

Fig. 3. Differentially expressed gene analysis indicates neutrophils of Lrrk2 G2019S-infected mice have an increased pro-inflammatory profile.

Male and female Lrrk2 G2019S and WT mice (male-to-female ratio = 0.5) were gavaged once with ~1 × 10⁹ CFUs of C. rodentium. A Differential gene expression (DEG) analysis comparing G2019S and WT-infected mice for each immune cell cluster calculated using Seurat differential expression analysis by non-parametric Wilcoxon rank sum test with Bonferroni correction. Adjusted P value cut off = 0.05, Q value cut off = 0.2. Upregulated genes are depicted as red in the heatmap, and downregulated genes are depicted in blue. B Pathway enrichment analysis completed on DEGs obtained by comparing G2019S and WT-infected neutrophils. DEGs were run against Gene ontology (GO) biological pathways database using ClusterProfiler V4.12.6 (RRID:SCR_016884). P value cut off = 0.05, Q value cut off = 0.2. Top 30 GO term categories are depicted as a Treeplot (Left). Cnetplots depict association of differentially expressed gene hits, and which GO term pathways they correspond to (Right). scRNAseq was obtained from pooled conditions of n = 3 mice per group. One independent experiment is presented.

Fig. 4. Lrrk2 G2019S mutation leads to increased colon morphological damage after C. rodentium infection.

Male and female Lrrk2 G2019S and WT mice (male-to-female ratio = 0.43) were infected once with ~1 × 10⁹ CFU of C. rodentium. Colons were harvested 7 days after infection, fixed in 10% formalin for 24 h, washed in 70% ethanol, and diaphanized in xylene before embedding in paraffin. Sections were stained with Hematoxylin and Eosin (H&E) and scored for histopathology. A Total histopathology score. B H&E representative images. Data are presented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. **p < 0.01. n = 7–11 mice per group. Two independent experiments were done and showed as a pool. White bars represent 100 µm. Symbols: blue star, edema; green arrow, inflammatory infiltrate; red arrow, erosion; yellow star, goblet cell loss. Please, also see Supplementary Fig. 8.

Fig. 5. Lrrk2 G2019S mutation leads to a cell-intrinsic increase in neutrophil effector function that is kinase-dependent.

Whole blood, femurs, tibias, and humeri were collected from male and female Lrrk2 G2019S and WT mice (male-to-female ratio = 0.55). A, B Bone marrow neutrophil migration quantification and representative images. Ly6G-isolated neutrophils were isolated from the bone marrow and incubated for 1 h with MLi-2 (100 nM), LRRK2in1 (1 μM), or DMSO (vehicle control) at 37 °C. Cells were seeded at 2 × 10⁵ cells per 5 μm transwell and allowed to migrate towards C. rodentium (MOI = 100) or PBS control for 1 h at 37 °C with 5% CO₂. Four random fields of view were obtained per membrane, quantified by a blind experimenter, and averaged per mouse. Data are presented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. *p < 0.05, **p < 0.01, n = 5–7 mice per group. A pool of two independent experiments is presented, normalized to WT untreated, uninfected controls for each respective experiment. Scale bar = 50 μm. C, D Blood neutrophil NETosis. Ly6G-isolated neutrophils were added at 1 × 10⁵ to a 24-well plate with coverslips. Cells were treated with MLi-2 (100 nM) or DMSO (vehicle control) for 1 h at 37 °C with 5% CO_2. Neutrophils were stimulated with C. rodentium and incubated for 2 h at 37 °C with 5% CO₂. Cells were stained for DAPI (blue) and Phalloidin (Alexa Fluor 488). Nine random fields of view from each condition were gathered, quantified results are shown as the percentage of NETosed cells. Data are presented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. ****p < 0.0001. n = 6–7 mice per group. A pool of two independent experiments is presented. Scale bar = 20 μm.

Fig. 6. Colonic infection in Lrrk2 G2019S mice initiates a skewing towards Th17 CD4⁺ T cells.

Male and female Lrrk2 G2019S and WT mice were gavaged once with ~1 × 10⁹ CFUs of C. rodentium, and colons were harvested. Immune cells were isolated for scRNAseq. A Re-clustering of CD4 T cell population based on effector T cell identities, depicted in a UMAP plot. B CD4 T cell subset markers used to identify effector cells based on literature. scRNAseq was obtained from pooled conditions of n = 3 mice per group (male-to-female ratio = 0.5). One independent experiment is presented. C Percent Th17 (RORgt+) CD4 T cells of CD45+ cells isolated from the colonic lamina propria across conditions. Data are presented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. **p < 0.01, ****p < 0.0001, n = 5–6 mice per group. One representative of two independent experiment is presented (male-to-female ratio = 0.9). D Dotplot depicting average expression of Il17a, Il17f, and Rorc transcripts within Th17 effector subset. E Il17a expression fold change of sorted CD45+ immune cells from the lamina propria measured through quantitative PCR (qPCR). One representative of two independent experiment is presented. n = 3 mice per group (male-to-female ratio = 0.5). Data are presented as mean ± SD and analyzed by two-way ANOVA with Fisher’s LSD post-test. *p < 0.05. All qRT-PCR data were normalized to Gapdh using the ΔΔCt method. F Left: Ratio of absolute counts of Th17 effector cells over T naïve cells in scRNAseq dataset. Right: Ratio of Th17 effector cells over T naïve cells in CD45+ lamina propria populations in flow cytometry data.