Lysosomal LRRC8 complex impacts lysosomal pH, morphology, and systemic glucose metabolism

Abstract

The lysosome integrates anabolic signaling and nutrient sensing to regulate intracellular growth pathways. The leucine-rich repeat-containing 8 (LRRC8) channel complex forms a lysosomal anion channel and regulates PI3K-AKT-mTOR signaling, skeletal muscle differentiation, growth, and systemic glucose metabolism. Here, we define the endogenous LRRC8 subunits localized to a subset of lysosomes in differentiated myotubes. We show that LRRC8A affects leucine-stimulated mTOR; lysosome size; number; pH; expression of lysosomal proteins LAMP2, P62, and LC3B; and lysosomal function. Mutating an LRRC8A lysosomal targeting dileucine motif sequence (LRRC8A-L706A;L707A) in myotubes recapitulates the abnormal AKT signaling and altered lysosomal morphology and pH observed in LRRC8A knockout cells. In vivo, LRRC8A-L706A;L707A knock-in mice exhibit increased adiposity, impaired glucose tolerance and insulin resistance associated with reduced skeletal muscle PI3K-AKT-mTOR signaling, glucose uptake, and impaired incorporation of glucose into glycogen. These data reveal a lysosomal LRRC8-mediated metabolic signaling function regulating lysosomal function, systemic glucose homeostasis, and insulin sensitivity.

Fig. 1.. Endogenous LRRC8 proteins are present in lysosomes.

(A) Schematic representation for the generation of Cre-inducible LAMP1-RFP-Flag-TEV-HA–expressing mouse. (B) Primary skeletal muscle myotubes isolated from Cre-inducible LAMP1-RFP-Flag-TEV-HA mice and C57BL/6 mice were transduced with Ad-CMV-Cre virus. Lyso-IP was performed with anti-HA magnetic beads and WB was performed for different LRRC8 complex subunits (LRRC8A/B/C/D/E); Input (postnuclear supernatant), FT (unbound protein), Wash (lysosome-bound beads washed with wash buffer), Triton X-100 treated (supernatant and pellet), and Elute (intact lysosomes bound with beads). WB of organelle-specific markers was performed to confirm the purity of isolated lysosome; Cathepsin D (lysosomal lumen protein), ER (Calreticulin), Golgi (Golgin-97), SDHA (Mito-ComplexII), Na-K ATPase (membrane marker), Lysosome (LAMP1), HA (epitope tag on expressed LAMP1), and Actin (loading control). (C) Schematic representation of the generation of 3×Flag-tagged LRRC8A knock-in (KI) mouse using the CRISPR-Cas9 approach. WB with anti-Flag antibody in isolated tibialis muscle of an LRRC8A-3×Flag (KI) mouse. GAPDH and Ponceau used as loading control (right side). (D) Lyso-IP performed on skeletal muscle tissue of LRRC8A-3×Flag-KI and WT control mice using anti-Flag magnetic beads. WB of Lyso-IP samples of LRRC8A-3×Flag (KI) in and WT for Flag, LRRC8A, LAMP1, Cathepsin D, ER, Golgi, and GAPDH protein. Densitometry quantification of LAMP1 and Cathepsin D shown below the images. (E) Live-cell confocal imaging showing colocalization of LysoTracker Red–stained lysosomes. White arrowheads indicate that nontransfected cells stain with LysoTracker Red only. Inset image shows that individual lysosome colocalizes with LRRC8A-ALFA. Scale bar, 10 μm. A line scanning intensity graph drawn between individual lysosome and LRRC8A-ALFA (lower side). (F) Stimulated emission depletion (STED) super-resolution imaging showing the localization of LRRC8A in the membrane and LAMP1-positive lysosomes (green). Inset images show an enlarged view of LRRC8A colocalization with lysosomes. Scale bars, 2 and 1 μm (inset). (C) was created in BioRender. Sah, R. (2025); https://BioRender.com/e33yijx.

Fig. 2.. LRRC8A is required for leucine-stimulated mTOR signaling.

(A) Western blot of LRRC8A, p-P70 S6k, P70 S6K, p-S6, S6, and β-actin in WT and LRRC8A KO C2C12 myotube after leucine (5 mM) stimulation for 15 min. (B) Densitometry quantification of leucine-stimulated signaling WB of (A). (C) WB of LRRC8A, β-actin, p-mTOR, p-AKT2, AKT2, p-P70 S6k, P70 S6K, p-S6, S6, protein in WT (LRRC8A^flfl + Ad-CMV-EGFP), and LRRC8A KO (LRRC8A^flfl + Ad-CMV-Cre-EGFP) primary myotube after leucine (5 mM) stimulation for 15 min. (D) Densitometry quantification of leucine-stimulated signaling WB of (C). (E) WB of LRRC8A, β-actin, and mTOR (pS6 and S6) signaling target protein in WT C2C12-sh-SCR and LRRC8A KD C2C12 myotube after leucine (5 mM) stimulation for 15 min. (F) Densitometry quantification of leucine-stimulated signaling WB (E). Statistical significance between the indicated values were calculated using statistical significance between the indicated group calculated with one-way ANOVA, Tukey’s multiple comparisons test. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. ns, not significant. n = 3 independent experiments.

Fig. 3.. LRRC8A depletion increases lysosomal size, alters morphology and autophagic marker protein expression, and decreases pH.

(A) Transmission electron microscopy (TEM) images of C2C12 WT and LRRC8A KO myotubes. Inset TEM image shows lysosome (red arrow) and mitochondria (black hollow arrow). Quantification of lysosome area and circularity index (WT = 73, LRRC8A KO = 265 lysosomes) (n = 2 independent experiments). Scale bar, 2 μm. (B) Reads per kilobase million (RPKM) for autophagy and lysosome biogenesis-related genes in C2C12 myotubes, 3T3-F442A adipocytes, and HUVECs. (C and D) WB of LRRC8A, LAMP1, LAMP2, β-actin, and GAPDH protein in WT and LRRC8A KO C2C12 myotube (C) and WT (LRRC8A^flfl + Ad-CMV-EGFP) and LRRC8A KO (LRRC8A^flfl + Ad-CMV-Cre-EGFP) primary myotubes, respectively (D); densitometry quantification below. (E) WB of LRRC8A, autophagy marker, and GAPDH protein in WT and LRRC8A KO C2C12 myotubes; densitometry quantification below. (F) Ratiometric images of Lysosensor-labeled WT and LRRC8A KO myotube. Scale bar, 100 μm. (G) Ratiometric images of WT C2C12 myotubes labeled with Lysosensor and incubated with different pH buffers. (H) pH standard curve plotted by using ratiometric image intensity at different pH buffers in WT C2C12 myotubes (n = 5, fluorescent images). Inset shows Lysosensor dye fluorescence at different pH buffers. (I) Lysosomal pH of WT and LRRC8A KO myotubes, which were determined from the nonlinear least squares fit to the pH calibration curve [total field of view 18 (WT) and 18 (KO), collected from six dishes per condition]. Statistical significance for (A), calculated by Mann-Whitney test. Error bars represent SD. [(B) to (E) and (I)] Significance calculated using a two-tailed Student’s t test. Error bars in (B) to (E) represent ±SEM, whereas those in (I) represent SD. One-way ANOVA used for (H), error bars represent SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 4.. Lysosomal targeting sequence knock-in mutation LRRC8A-L706A;L707A selectively depletes LRRC8A in lysosomes in vivo.

(A) CRISPR-Cas9–based approach to deplete lysosomal LRRC8 channels in vivo by introducing L706A;L707A double point mutations into the previously generated LRRC8A-3×Flag-KI mouse to generate LRRC8A-L706A;L707A-3×Flag KI (LL:AA) mice. (B) Whole-cell patch-clamp recordings of primary myoblasts isolated from LRRC8A-3×Flag (KI-control) and LRRC8A-L706A;L707A-3×Flag (LL:AA KI) showing current-time (left, at +100 and −100 mV) and current-voltage (right) relationships under isotonic (300 mosmol) and hypotonic (210 mosmol) conditions, followed by the application of 10 μM DCPIB. Voltage protocol applied was a voltage ramp −100 to +100 mV. (C) Mean outward current in isotonic and hypotonic conditions recorded from KI-control (n = 5) and LL:AA KI (n = 5) myoblasts. (D) Schematic representation for generating Cre-inducible LAMP1-RFP-Flag-TEV-HA;LRRC8A-L706A;L707A-3×Flag KI mice by crossing CAG-loxP-STOP-loxP-LAMP1-RFP-Flag-TEV-HA mice with LRRC8A-L706A;L707A-3×Flag KI mice. Primary skeletal muscle isolated from Cre-inducible LAMP1-RFP-Flag-TEV-HA mice and Cre-inducible LAMP1-RFP-Flag-TEV-HA;LRRC8A-L706A;L707A-3×Flag KI mice, transduced with Ad-CMV-Cre and lyso-IP performed with anti-HA beads. WB performed for LRRC8A, HA, Na/K ATPase, Cathepsin D, LAMP1, ER (Calreticulin), and β-actin. The ratios of LRRC8A/HA (input) or LRRC8A/LAMP1 (input) proteins in the Lyso-IP lane of WT-LAMP1 and LL:AA-LAMP1 are shown below. (E) Confocal imaging shows colocalization of LAMP1-positive lysosomes (green) with transiently expressed LRRC8A-ALFA (red) or LL:AA-ALFA in LRRC8A KO C2C12 myoblast cells. Inset image shows that individual lysosome colocalizes with LRRC8A-ALFA. Scale bars, 5 and 2 μm (inset). (F) Pearson’s relation for colocalization of LRRC8A-ALFA or LL:AA-ALFA with LAMP1 (LRRC8A-ALFA = 11, LL:AA-ALFA = 12, cropped area from five to six different cells). For (C) and (F), statistical tests for significance between the indicated values were carried out using a two-tailed Student’s t test. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (A) was created in BioRender. Sah, R. (2025); https://BioRender.com/b83qa0k.

Fig. 5.. Bulk RNA transcriptome analysis of lysosomal LRRC8A-depleted myotubes reveals alterations in phagosome, inflammation, and metabolism-associated genes.

(A) Heatmap analysis of the top 30 differentially expressed genes derived from RNA transcript of LRRC8A-3×Flag KI (n = 5) and LL:AA-3×Flag KI primary myotubes (n = 5). (B) Reads per kilobase million for selected gene of complement cascade, lysosomal cathepsin and phagocytic marker–associated genes. Fold change for phagocytic genes is displayed above the bar graph. (C and D) Ingenuity pathway analysis (IPA) of canonical pathways showing altered cellular signaling (C) and altered development, endocrine, and metabolic pathways (D) in LL:AA-3×Flag KI (n = 5) myotubes compared to LRRC8A3×Flag-KI (n = 5). For analysis with IPA, a fold change of ≥1.5 and a false discovery rate of <0.05 were used for significant differentially expressed genes. (E) Interaction networks identified by IPA shows affected cellular signaling pathways in LL:AA-3×Flag, including PI3K cascade (left), formation of phagosomes (center), and PIP3-activated AKT signaling (right). Network-associated gene names and line symbol are indicated by a predicted legend box (bottom).

Fig. 6.. Lysosomal LRRC8A-depleted myotubes phenocopy LRRC8A KO myotubes with respect to lysosomal morphology, pH, and intracellular signaling.

(A) Fluorescence image of LysoTracker Red–stained image of KI-control and lysosomal-depleted LRRC8A (LL:AA) primary muscle myotubes. Scale bar, 10 μm. Lysosome surface area quantified in KI-control (n = 2360 lysosomes) and LL:AA myotube (n = 1813 lysosomes) images. Error bars represent SD. (B) Western blot of LRRC8A, LAMP1, LAMP2, β-actin, and GAPDH protein in KI-control and LL:AA primary myotube. Densitometry quantification below. (C) Ratiometric (Ex340/Ex380) images of Lysosensor-stained images of KI-control and LL:AA primary myotubes. Scale bar, 100 μm. Lysosomal pH values of KI-control and LL:AA primary myotubes that were determined from the nonlinear least squares fit to the pH calibration curve [total field of view 11 (KI) and 8 (LL:AA), collected from six dishes per condition] (shown below). Error bars represent SD. (D) WB of LRRC8A, Flag, pAS160, AS160, pAKT2, AKT2, pAKT1, AKT1, and GAPDH protein in KI-control and LL:AA primary myotubes under basal conditions. (E) Densitometry quantification of (D). (F) WB of pAKT2, AKT2, pAKT1, AKT1, and GAPDH protein of KI-control and LL:AA primary myotubes stimulated with 0 and 10 nM insulin for 15 min. (G) Densitometry quantification of (F). Statistical significance for lysosome area (A) was calculated by Mann-Whitney test. For (B), (C), and (E), statistical tests for significance between the indicated values were carried out using a two-tailed Student’s t test. For (G), statistical significance between the indicated group was calculated with one-way ANOVA, Tukey’s multiple comparisons test. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n = 3 independent experiments.

Fig. 7.. Lysosomal alkalinization restores insulin signaling in LRRC8A KO C2C12 myotubes.

(A and B) Western blots of LRRC8A, pAKT2, AKT2, and GAPDH in LRRC8A KO myotubes, pretreated with either HCQ (25 μM) or Baf A1 (10 nM) for 4 hours and then stimulated with insulin (10 nM) for 15 min. Densitometry quantification of (A) and (B) WB shown below for each. Statistical significance between the indicated group calculated with one-way ANOVA, Tukey’s multiple comparisons test. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n = 3 independent experiments.

Fig. 8.. Lysosomal LRRC8A-depleted mice exhibit increased adiposity, impaired glucose tolerance and insulin resistance, and decreased glucose uptake.

(A) Glucose tolerance test (GTT) of KI-control (n = 9) and LL:AA KI (n = 12) mice raised on chow diet for 20 to 22 weeks. (B) Six-hour fasting glucose during GTT. (C) Body weight. (D to F) Insulin tolerance test (ITT) of KI-control (n = 9) and LL:AA KI (n = 12) mice raised on chow diet for 22 to 24 weeks (D), fasting glucose after 4 hours (E), and body weight (F). (G) NMR measurement of absolute fat mass, % fat mass, absolute lean mass, % lean mass, and body weight of KI-control (n = 14) and LL:AA KI (n = 12) mice raised on chow diet for 38 to 40 weeks. (H) Average glucose-infusion rate during the euglycemic-hyperinsulinemic clamp period of KI-control (n = 5) and LL:AA KI (n = 7) mice on chow diet for 32 to 36 weeks, and mean glucose-infusion rate during the entire clamp period (70 to 120 min) on the right. (I) Hepatic glucose production at baseline and during the euglycemic-hyperinsulinemic clamp period. (J to Q) Glucose uptake determined from radiolabeled 2-deoxyglucose (2-DG) uptake in gastrocnemius (J), tibialis (K), soleus (L) muscle, liver (M), heart (N), inguinal white adipose tissue [iWAT, (O)], gonadal white adipose tissue [gWAT, (P)], and brown adipose tissue [BAT, (Q)] during the traced clamp period. Data were presented as mean ± SEM. Statistical test two-way ANOVA was used for (A), (D), and (H) (P value in the bottom corner of the graph). Error bars represent mean ± SEM. Two-tailed Student’s t test were done for all other data. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig. 9.. Lysosomal LRRC8A depletion impairs skeletal muscle insulin-PI3K-mTOR in vivo.

(A) Western blots of Flag, p-AS160, AS160, p-AKT2, AKT2, p-mTOR, p-P70 S6K, P70 S6K, p-S6, S6, and GAPDH in soleus muscle from LRRC8A-3×Flag KI and LL:AA-3×Flag KI mice following in vivo insulin stimulation (5 μl from 100 IU/ml Humulin R insulin via inferior vena cava) for 10 min. (B) Densitometry quantification of (A). Statistical significance between the indicated group calculated with one-way ANOVA, Tukey’s multiple comparisons test. Error bars represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. n = 3 independent experiments.