Hepatic LDL receptor-related protein-1 deficiency alters mitochondrial dynamics through phosphatidylinositol 4,5-bisphosphate reduction

Abstract

The LDL receptor-related protein 1 (LRP1) is a multifunctional transmembrane protein with endocytosis and signal transduction functions. Previous studies have shown that hepatic LRP1 deficiency exacerbates diet-induced steatohepatitis and insulin resistance via mechanisms related to increased lysosome and mitochondria permeability and dysfunction. The current study examined the impact of LRP1 deficiency on mitochondrial function in the liver. Hepatocytes isolated from liver-specific LRP1 knockout (hLrp1^-/-) mice showed reduced oxygen consumption compared with control mouse hepatocytes. The mitochondria in hLrp1^-/- mouse livers have an abnormal morphology and their membranes contain significantly less anionic phospholipids, including lower levels of phosphatidylethanolamine and cardiolipin that increase mitochondrial fission and impair fusion. Additional studies showed that LRP1 complexes with phosphatidylinositol 4-phosphate 5-kinase like protein-1 (PIP5KL1) and phosphatidylinositol 4-phosphate 5-kinase-1β (PIP5K1β). The absence of LRP1 reduces the levels of both PIP5KL1 and PIP5K1β in the plasma membrane and also lowers phosphatidylinositol(4,5) bisphosphate (PI(4,5)P₂) levels in hepatocytes. These data indicate that LRP1 recruits PIP5KL1 and PIP5K1β to the plasma membrane for PI(4,5)P₂ biosynthesis. The lack of LRP1 reduces lipid kinase expression, leading to lower PI(4,5)P₂ levels, thereby decreasing the availability of this lipid metabolite in the cardiolipin biosynthesis pathway to cause cardiolipin reduction and the impairment in mitochondria homeostasis. Taken together, the current study identifies another signaling mechanism by which LRP1 regulates cell functions: binding and recruitment of PIP5KL1 and PIP5K1β to the membrane for PI(4,5)P₂ synthesis. In addition, it highlights the importance of this mechanism for maintaining the integrity and functions of intracellular organelles.

Keywords: cardiolipin; inositol phosphate; lipoprotein receptor; lipoprotein receptor-related protein (LRP); liver metabolism; respiration.

Figure 1

Hepatic LRP1 deficiency reduces mitochondrial respiration and ATP production.A, primary hepatocytes isolated from wild-type (solid symbols, WT) or hLrp1^−/− mice (open symbols, LRP1 KO) (n = 4) were incubated in medium containing 25 mM glucose, 1 mM sodium pyruvate, and 4 mM Glutamax to determine basal oxygen consumption rates. Oligomycin, FCCP, and rotenone/antimycin were added to measure ATP production, maximum and spare respiration, and coupling efficiency. The data were used to determine basal respiration rate (B), ATP production (C), proton leak (D), maximum respiration rate (E), spare respiration (F), and coupling efficiency (G). The data are reported as mean ± SD and were evaluated by two-tailed Student’s t-test for statistical significance as indicated.

Figure 2

Hepatic LRP1 deficiency reduces (A) pyruvate/malate- and (B) succinate-mediated mitochondrial respiration. Primary hepatocytes isolated from wild-type (solid bars, WT) or hLrp1^−/− mice (open bars, LRP1 KO) (n = 4) were incubated in medium containing (A) pyruvate and malate or (B) succinate to determine basal oxygen consumption rates. State 2 respiration was assessed in the presence of the indicated substrates. State 3 respiration was assessed with ADP addition. State 4o respiration was assessed based on oxygen consumption in the presence of oligomycin, and state 3u respiration was assessed in the presence of FCCP. The data are reported as mean ± SD and were evaluated by two-tailed Student’s t-test with statistical significance as indicated.

Figure 3

Hepatic LRP1 deficiency reduces mitochondria complex I and complex II levels. Liver lysates were prepared from wild-type (WT, filled symbols) and hLrp1^−/− (LRP1 KO, open symbols) mice for western blot analysis with antibody cocktail against rodent oxidative phosphorylation proteins. The left panel shows representative western blot image with molecular size markers and the identification of complex I through V (C-I through C-V) as indicated. The right panel shows the relative expression levels of each mitochondria complex in seven different liver preparations in each group. The data are reported as mean ± SD and were evaluated by two-tailed Student’s t-test for statistical significance as indicated.

Figure 4

LRP1 deficiency has no impact on mitochondria number but enhances calcium accumulation and induces membrane potential loss.A, mitochondria number was determined based on mitochondrial DNA content in the livers of eight wild-type (WT) and eight hLrp1^−/− (LRP1 KO) mice. B, mitochondria were isolated from WT (solid symbols and bars) and hLrp1^−/− (open symbols and bars) mice and incubated with the calcium indicator Calcium Green-5N prior to the addition of 75 μM CaCl₂ at the time indicated by the arrow. Calcium levels in the medium were determined based on Calcium Green-5N fluorescence over time. C, membrane potential was assessed by absorbance measurement after incubation with the membrane voltage indicator safranin O. The data are reported as mean ± SD from four separate preparations and were evaluated by two-tailed Student’s t-test for statistical significance. ∗ indicates difference from wild type at p = 0.01.

Figure 5

Hepatic LRP1 deficiency reduces anionic phospholipid content in the mitochondria. Lipids in the mitochondria isolated from wild-type (solid bars) and hLrp1^−/− (open bars) mouse livers were extracted with chloroform:methanol:water (2:1:3, v/v/v) and then applied to thin layer chromatography plates for phospholipid separation. Migration of the phospholipids was compared with standards. Phospholipid spots on the chromatography plates corresponding to phosphatidylethanolamine (PE), cardiolipin (CL), phosphatidylinositol (PI), phosphatidylserine (PS), and phosphatidylcholine (PC) were scraped and quantified by phosphorus measurements. The data are reported as mean ± SD from N = 3 mice in each group. The data were evaluated by two-tailed Student’s t-test for statistical significance. # indicates difference from wild type at p < 0.01.

Figure 6

Hepatic LRP1 deficiency alters mitochondria morphology. Liver sections from wild-type (WT) and hLrp1^−/− (LRP1 KO) mice were subjected to electron microscopy analysis of the mitochondria. A, shows representative images of the liver sections from WT and KO mice. Mitochondria length and shape were determined by measuring ∼40 mitochondria from each mouse (N = 3 per group) and then averaged. B, cristae number and C, cristae density were determined from the electron microscopy images. The data are reported as mean ± SD (N = 8) and were evaluated by two-tailed Student’s t-test for statistical significance. # indicates difference from wild type at p < 0.01.

Figure 7

Hepatic LRP1 deficiency promotes mitochondrial fission and inhibits mitochondrial fusion in the liver. Liver homogenates from wild-type (WT, filled symbols) and hLrp1^−/− (LRP1 KO, open symbols) mice (N = 8) were prepared for western blot analysis of MFN2, OPA1, DRP1, and LRP1 expression levels using β-actin levels as the loading control. The images were digitalized by scanning, and quantitative measurements were performed using Image J software. Expression levels of the mitochondrial proteins were normalized to wild-type levels. # indicates statistically significant differences from wild type at p ≤ 0.01.

Figure 8

Hepatic LRP1 deficiency reduces PI(4,5)P₂levels by reducing the levels of PIP5K1B and PIP5KL1.A, PI(4,5)P₂ levels in liver extracts obtained from WT and LRP1 KO mice (N = 7) were determined by ELISA. The data were evaluated for statistical significance by two-tailed Student’s t-test. # indicates differences from WT group at p < 0.01. B, liver membrane fractions isolated from WT and LRP1 KO mice were subjected to western blot analysis for expression of PIP5K1A, PIP5K1B, and PIP5KL1 using ATP1A1 as loading control. The images were digitalized and quantified by Image J. # indicates differences at p < 0.05. C, liver lysates from wild-type mice were subjected to immunoprecipitation (IP) with antibodies against LRP1 or PIP5K1B, using nonspecific IgG as a negative control. The immunoprecipitates were analyzed by western blot (WB) analysis of LRP1, PIP5K1B, and PIP5KL1 as indicated. In, input used for immunoprecipitation. The bands identified as LRP1, PIP5K1B, and PIP5KL1 are outlined in the box.

Figure 9

Schematic diagram depicting the mechanism underlying the role of LRP1 in PI(4,5)P₂and cardiolipin synthesis. The diagram illustrates LRP1, located at the plasma membrane, binds PIP5KL1, and recruits PIP5K1β to the membrane to catalyze PI(4,5)P₂ synthesis from phosphatidylinositol (PI) and phosphatidylinositol-4-phosphate (PI4). The PI(4,5)P₂ is hydrolyzed to diacylglycerol (DAG) by phospholipase C (PLC) and then converted to phosphatidic acid (PA), which can be translocated to the endoplasmic reticulum for synthesis of CDP-DAG in the presence of CDP-DAG synthase-1 and -2 (CDS1/CDS2). The CDP-DAG can be translocated from the endoplasmic reticulum to the mitochondria where it can be combined with phosphatidylglycerol (PG) for cardiolipin synthesis.