Genetic Coding Variant in GPR65 Alters Lysosomal pH and Links Lysosomal Dysfunction with Colitis Risk

Abstract

Although numerous polymorphisms have been associated with inflammatory bowel disease (IBD), identifying the function of these genetic factors has proved challenging. Here we identified a role for nine genes in IBD susceptibility loci in antibacterial autophagy and characterized a role for one of these genes, GPR65, in maintaining lysosome function. Mice lacking Gpr65, a proton-sensing G protein-coupled receptor, showed increased susceptibly to bacteria-induced colitis. Epithelial cells and macrophages lacking GPR65 exhibited impaired clearance of intracellular bacteria and accumulation of aberrant lysosomes. Similarly, IBD patient cells and epithelial cells expressing an IBD-associated missense variant, GPR65 I231L, displayed aberrant lysosomal pH resulting in lysosomal dysfunction, impaired bacterial restriction, and altered lipid droplet formation. The GPR65 I231L polymorphism was sufficient to confer decreased GPR65 signaling. Collectively, these data establish a role for GPR65 in IBD susceptibility and identify lysosomal dysfunction as a potentially causative element in IBD pathogenesis with effects on cellular homeostasis and defense.

Figure 1. Functional Genomic Analysis Identifies Genes Involved in Autophagy-Dependent Intracellular Pathogen Defense

(A) Schematic of high-throughput genetic screens. HeLa or HeLa-GFP-LC3 cells were transfected with an siRNA library and infected with bioluminescent S. Typhimurium (intracellular bacterial replication) or DsRed-labeled S. Typhimurium (LC3 colocalization). (B) Mean effect size of siRNAs in intracellular bacterial replication assay (left) and LC3 colocalization assay (right). n = 2 independent experiments. (C) Representative confocal micrographs of selected siRNA-treated HeLa cells infected for 1 hr with DsRed-labeled S. Typhimurium. Scale bars represent 7.5 μm. (D) Validated genes that scored in LC3 colocalization and intracellular bacterial replication assays using single siRNAs for a given gene. (E) Heatmap illustration of effect size scores of confirmed hits. (F) Schematic of GPR65 locus with identified IBD risk SNPs from the IBD fine-mapping project. See also Figure S1 and Tables S1, S2, and S3–S5.

Figure 2. Gpr65^−/− Mice Are More Susceptible to Bacterial-Induced Colitis

(A) Representative H&E-stained sections of distal colon tissue are shown from untreated WT and Gpr65^−/− mice (20× magnification) (n = 4 mice per genotype). (B) Representative H&E-stained sections of distal colon tissue are shown from infected WT and Gpr65^−/− mice (20× magnification) (n = 8 mice per genotype). (C) Histological score for inflammation in colon tissues 14 days after C. rodentium infection. Data shown as mean ± SD; n = 5/group. **p < 0.01 (Mann-Whitney U test). (D) WT and Gpr65^−/− mice were orally infected with C. rodentium and bacterial numbers (CFU) were measured in the cecum, colon, feces, and liver 14 days after infection. Data are means + SEM (n = 10 [WT], 10 [Gpr65^−/−]; 2 experiments). **p < 0.01 (Mann-Whitney U test). (E) Immunohistochemistry image of CD3 staining on WT and Gpr65^−/− colon tissue 14 days after infection with C. rodentium. See also Figure S2.

Figure 3. Gpr65 Expression in Non-hematopoietic and Hematopoietic Cells Limits C. rodentium Infection

(A) Bone marrow chimeric mice were orally infected with C. rodentium and bacterial numbers (CFU) were measured in the colon at 11 days after infection. Data are means + SEM (n = 11 [WT → WT], n = 6 [WT → Gpr65^−/−], n = 10 [Gpr65^−/− → WT], n = 4 [Gpr65^−/−/Gpr65^−/−]). *p < 0.05, **p < 0.01, ****p < 0.0001 (unpaired t test). (B) Colon length from bone marrow chimeric mice infected with C. rodentium for 11 days. Data are means + SEM. **p < 0.01 (unpaired t test). (C) Representative H&E-stained sections of distal colon tissue are shown from infected bone marrow chimeric mice at 11 days after infection (20× magnification). (D) Cytokine expression in C. rodentium-infected mice, as quantified by qRT-PCR. Relative mRNA levels of the indicated cytokine are shown. *p < 0.05; **p < 0.01 (unpaired t test). (E) Secretion of cytokines from colon tissues 11 days after infection with C. rodentium. Data are means + SEM. *p < 0.05 (unpaired t test). See also Figure S3.

Figure 4. Cells Lacking GPR65 Exhibit Impaired Antibacterial Autophagy and Accumulation of Aberrant Degradative Compartments

(A) S. Typhimurium intracellular replication in GPR65-null HeLa cells stably expressing either an empty vector or GPR65 WT. Bars represent means ± SD from n = 4. ***p < 0.001 (unpaired t test). (B) CFU of intracellular Listeria ΔactA recovered from WT or Gpr65^−/− BMDMs at 3 hr after infection. ***p < 0.001 (unpaired t test). (C) Confocal micrographs of endogenous LC3-S. Typhimurium colocalization in GPR65-null HeLa cells stably expressing either an empty vector or GPR65 WT at 1 hr after infection. Scale bars represent 7.5 μm. (D) Quantification of colocalization of S. Typhimurium and endogenous LC3 in indicated cells 1 hr after infection. Data shown represent means + SEM from n = 3 independent experiments. *p < 0.05 (unpaired t test). (E) Quantification of aberrant endogenous LC3 accumulation at sites of S. Typhimurium-LC3 colocalization in cells shown in Figure S3D. Data shown represent means + SEM of n = 3 independent experiments. ****p < 0.0001 (unpaired t test). (F) Quantification of the percentage of BMDMs displaying more than two aberrant phagosomes as shown in (B). Results are expressed as percent of cells with a defined type of phagosome in the cell population + SEM. **p < 0.01 (t test). (G) Representative EM micrographs of WT and Gpr65^−/− BMDMs showing phagosomal morphology. Abbreviations are as follows: ER, endoplasmic reticulum; G, Golgi; M, mitochondria; P, phagosomes; P*, aberrant phagosomes; PM, plasma membrane. Scale bars represent 1 μm. (H) WT and Gpr65^−/− BMDMs were treated with S. aureus-coated bioparticles conjugated to the pH indicator dye for the indicated times. Number of puncta were normalized to WT levels at each time point. n = 3 independent experiments. (I) Representative EM micrographs of the lysosomal subpopulations observed in WT and GPR65-null HeLa cells. Abbreviations and symbols are as follows: #, typical lysosome; *, lysosome with a partial degradation defect; **, aberrant lysosome with a degradative defect; ER, endoplasmic reticulum; G, Golgi; PM, plasma membrane. Scale bars represent 500 nm. (J) Quantification of each lysosome subpopulation shown in (I). Results are expressed as the number lysosomes per cell + SEM. ****p < 0.0001 (t test). See also Figure S4.

Figure 5. GPR65 I231L-Expressing Cells Exhibit Aberrant Lysosomal Function and Increased Bacterial Replication

(A) Relative Lysotracker fluorescence in GPR65-null HeLa cells stably expressing either an empty vector, GPR65 WT, or GPR65 I231L. Cells were left untreated or treated with 100 nM bafilomycin A1 for 2 hr. (B) Representative micrographs of GPR65-null HeLa cells stably expressing an empty vector, GPR65 WT, or GPR65 I231L stained with Lysotracker. Cells were left untreated or treated with 10 μg/mL E64d and pepstatin A for 3 hr prior to staining. (C) Quantification of cells imaged in (B) with aberrant lysosomal localization. (D) Relative DQ-BSA fluorescence in GPR65-null, GPR65 WT, or GPR65 I231L HeLa cells. Cells were left untreated or treated with 100 nM bafilomycin A1 for 2 hr. (E) Fraction of DQ-BSA-positive fluorescent beads compared to total internalized beads in GPR65-null, GPR65 WT, or GPR65 I231L HeLa cells. (F) HeLa cells treated with DQ BSA-conjugated beads. AF594 (red) was used to monitor total uptake of beads into cells. DQ-BSA (green) was used to monitor phagolysosomal activity. (G) GPR65-null HeLa cells stably expressing an empty vector or GPR65-V5 WT were treated with buffers at the indicated pH for 30 min and cAMP levels were measured. Average of n = 3 independent experiments. Error bars represent + SEM. For all panels, *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA with multiple comparisons. See also Figures S5 and S6.

Figure 6. GPR65 I231L Impairs Lipid Droplet Turnover

(A) Representative micrographs of cells treated for 4 hr with complete medium, HBSS, or HBSS and 200 nM bafilomycin A1 and stained with BODIPY. n = 4 independent experiments. (B and C) Quantification of images as shown in (A). Results are representative of n = 4 independent experiments. Bars show mean + SEM.

Figure 7. Impaired Lysosomal Function in GPR65 I231L-Expressing Cells Results in Increased Bacterial Replication

(A) S. Typhimurium intracellular replication in GPR65-null HeLa cells stably expressing either GPR65 WT or GPR65 I231L. Bars represent means ± SD from n = 4. (B) Fraction of DQ-BSA-positive fluorescent beads compared to total internalized beads in GPR65 WT cells or GPR65 I231L cells stably expressing either ATP6V1D or ATP6V1E1. (C) S. Typhimurium intracellular replication in GPR65 WT cells or GPR65 I231L cells stably expressing either ATP6V1D or ATP6V1E1. Bars represent means ± SD from n = 3 independent experiments. (D) Intracellular S. Typhimurium replication in IBD patient-derived lymphoblasts expressing the ancestral SNP (AA) or the risk allele (CC). For all panels, **p < 0.01, ***p < 0.001 (unpaired t test in A and D, one-way ANOVA with multiple comparisons in C).