Loss of hepatic chaperone-mediated autophagy accelerates proteostasis failure in aging

Chaperone-mediated autophagy (CMA), a cellular process that contributes to protein quality control through targeting of a subset of cytosolic proteins to lysosomes for degradation, undergoes a functional decline with age. We have used a mouse model with liver-specific defective CMA to identify changes in proteostasis attributable to reduced CMA activity in this organ with age. We have found that other proteolytic systems compensate for CMA loss in young mice which helps to preserve proteostasis. However, these compensatory responses are not sufficient for protection against proteotoxicity induced by stress (oxidative stress, lipid challenges) or associated with aging. Livers from old mice with CMA blockage exhibit altered protein homeostasis, enhanced susceptibility to oxidative stress and hepatic dysfunction manifested by a diminished ability to metabolize drugs, and a worsening of the metabolic dysregulation identified in young mice. Our study reveals that while the regulatory function of CMA cannot be compensated for in young organisms, its contribution to protein homeostasis can be handled by other proteolytic systems. However, the decline in the compensatory ability identified with age explains the more severe consequences of CMA impairment in older organisms and the contribution of CMA malfunction to the gradual decline in proteostasis and stress resistance observed during aging.

In the absence of liver CMA, metabolic dysregulation is evident from an early age but problems in protein quality control are masked due to effective compensation by other proteolytic systems. However, the loss of compensatory mechanisms that occurs with age or stress (i.e. diet-induced obesity) causes a progressive worsening of the phenotype in CMA-deficient mice. The age-related hepatic functional decline is caused not only by gradual metabolic dysfunction but also by the additive deterioration in proteostasis and the ensuing accumulation of damaged proteins and aggregates. Pharmacological interventions aimed at prolonging the period of compensation may be effective in delaying the consequences of CMA failure.

Extended Experimental Procedures
Animal diets and treatments. Where indicated, mice were starved with water ad libitum. Mice of all genotypes were kept on a regular 12h dark/light cycle and a standard chow diet (LabDiet #5058), except for the experimental subgroup maintained on a high-fat diet (HFD, Research Diets D12492, 60% kcal% fat) for 16 weeks. Where indicated, leupeptin (20mg/kg b.w.; Fisher Scientific) or saline (in controls) was injected i.p. 2h before tissue harvesting. Acetaminophen (375mg/kg) was injected i.p. 24h before tissue collection. To induce mild oxidative stress, mice were treated with two lowdose i.p. injections of paraquat for two consecutive days one week prior to sacrifice (4mg/kg b.w.; Sigma) and one higher dose 24h before tissue harvesting (40mg/kg b.w.; Sigma). All treatments and procedures followed the National Institutes of Health guidelines for animal care.
Hepatocyte isolation. Primary hepatocytes were isolated from livers of non-fasted control and liverspecific L2AKO mice by the Marion Bessin Liver Research Center Core and grown in culture as previously described (Matsuda et al. 2001;Edwards et al. 2013). Briefly, mice were anesthetized and livers were perfused and digested via the portal vein with a peristaltic pump. The livers were removed, the capsule peeled off, and the hepatocytes were dispersed in digestion media, followed by filtration through gauze. Digestion was stopped by the addition of RPMI media (Sigma-Aldrich) supplemented with 5% (v/v) Newborn Calf Serum (NCS, Hyclone), 10mM HEPES (pH 7.4), and 1% penicillin/streptomycin/fungizone (Invitrogen). Cell viability at the time of isolation was determined by trypan blue and was normally between 90-95%. The cells were plated on 35-and 60-mm dishes coated with mouse type I collagen in the above media at a density of 0.5 x 10 6 cells or 1.5 x 10 6 cells per dish, respectively. All cells were maintained at 37°C with 5% CO 2 .

Zoxazolamine-induced paralysis test.
We analyzed the clearance time of zoxazolamine, a muscle relaxant metabolized by the liver, as an index of hepatic function. We gave mice a single i.p. injection of zoxazolamine (150mg/kg) in olive oil, placed them on their backs and recorded the time required to regain the righting reflex after the paralysis induced by this compound .

Morphometric analysis and electron microscopy.
Electron microscopy for liver was done after fixation of liver blocks (1 mm 3 in size) with 2% paraformaldehyde and 2% gluteraldehyde in 0.1M sodium cacodylate buffer followed by post-fixation staining with 1% osmium tetroxide and 1% uranyl acetate. After dehydratation, resin embedding and ultrathin sectioning samples were viewed on a JEOL 1200EX transmission electron microscope at 80 kV. Morphometric analysis was performed in micrographs using Image J software and classification of autophagic vacuoles was done following the standard criteria (Singh et al. 2009) as follows: autophagosomes were distinguished as double membrane vesicles with content of similar density as the surrounding cytosol and comprised often of recognizable cellular structures; autophagolysosomes were identified as single or partially double membrane vesicles of content of lower density than the surrounding cytosol and comprised of amorphous content or partially degraded cellular structures. In both cases, the limiting membrane had to be denuded of ribosomal particles.

Measurement of proteasome activities.
Catalytic activities of the proteasome were determined as previously described (Liggett et al. 2010;Pickering et al. 2010). Briefly, liver homogenates were prepared in 0.25M sucrose and diluted 1:2 in reaction buffer. Protein concentration was quantified by Lowry assay and 0.01-0.05mg of homogenate was used. The volume was brought up to 90μl in ice-cold reaction buffer followed by the addition of 10μl of substrate stock solution. Each substrate stock consisted of a 10x solution made in reaction buffer, containing AMC-tagged fluorogenic peptides to measure trypsin-like, chymotrypsin-like, or caspase-like proteasome catalytic activities.
Plates were incubated at 37°C for 5 min and fluorescence readings were taken at 10 min intervals using an excitation wavelength of 350nm and an emission of 440nm for 2h. Fluorescence units were converted to moles of free AMC, with reference to an AMC standard curve of known amounts of AMC, following subtraction of background fluorescence. In some wells, Lactacystin (1μM) was added as a negative control.

Analysis of Protein oxidation. Levels of oxidized proteins were determined using the OxyBlot
Oxidized Protein Detection Kit (Chemicon International). Briefly, carbonyl groups of oxidized proteins were derivatized to 2,4-dinitrophenylhydrazone by reaction with 2,4-dinitrophenylhydrazine and detected by immunoblot with an antibody specific for the dinitrophenyl (DNP) derivatized groups.
Fractions to be used for oxidized protein detection were supplemented with 50mM DTT after preparation from mouse livers and stored at -80°C until use. Densitometric quantification of the immunoblotted membranes was performed using unsaturated images taken in the LAS-3000 Imager and with ImageJ software (NIH). For two-dimensional oxyblots, following sample rehydration and isoelectric focusing, the strips were incubated in 2N HCl with 10mM DNP at 25°C for 20 min, washed with 2M Tris containing 30% glycerol for 15 min at RT, and incubated with DTT and Iodoacetamide equilibration buffers. After SDS-PAGE was performed to separate proteins by size, immunoblot analysis was carried out using an antibody specific for DNP moieties using the Oxyblot protein detection kit.

Protein aggregation.
For the filter retardation assay, liver homogenates (200μg) were resuspended in 500μl of 50mM Tris pH8, 100mM NaCl, 5mM MgCl 2 , 0.5% NP-40 and protease inhibitors. After incubation on ice for 30 min, cells were centrifuged at 16,000g for 10 min at 4°C to pellet aggregated proteins as previously described (Massey et al. 2008). Pellets were resuspended in 200μl of Tris Buffer with 4% SDS and 100mM DTT, vortexed, and boiled at 100°C for 5 min. The samples were filtered through a 0.45μm nitrocellulose membrane in a BioDot Blot apparatus (Bio-Rad). After filtration, the aggregates caught in the membrane were assessed by immunoblot using antibodies against ubiquitinated proteins.

In vivo measurement of reactive oxygen species.
Mice were anesthetized with an i.p. injection of a mixture of ketamine and xylazine (6.6:1) and imaged using an In Vivo Imaging System (IVIS, Kodak Image Station 400MM PRO, Carestream Health) before any treatments to assess baseline autofluorescence. Mice were treated by tail vein injection with a fluorogenic probe that detects reactive oxygen species (ROS), purchased from Molecular Probes (CellROX Deep Red Reagent C) (Scharf et al. 2013). Mice were imaged 20 min after injection, followed by opening of their abdominal cavity to image their organs in situ. Lastly, organs were dissected and removed from the body and Isolation of nuclear fractions. Nuclear fractions were isolated with the NE-PER lysis Kit (Pierce) following manufactures' directions or by a NP-40 based lysis protocol (Cuervo et al. 1998). Briefly, cells were washed with PBS and lysed by incubation on ice in a NP-40 buffer (50mM Tris-HCl, pH 7.6; 150mM NaCl; 20mM NaF, 1mM EDTA, 1mM EGTA, 0.5% NP-40, 10% Glycerol). The nuclear pellet was collected by centrifugation and lysed in a high-salt buffer (50mM Tris-HCl, pH 7.6; 500mM NaCl; 20mM NaF, 1mM EDTA, 1mM EGTA, 1% NP-40, 10% Glycerol) by sonication with 3 burst/5 sec each. The nuclear pellet was collected by centrifugation a 15,000g for 5 min. Enrichment and purity of the fractions were evaluated by analysis of levels of γH2A (nuclear) and GAPDH (cytosol) in the samples.
Quantitative Proteomics. Comparative proteomics of lysosomes from Ctr and L2AKO mice was performed as described before (Schneider et al. 2014). Briefly, lysosomes active for CMA were isolated from 24 hour-starved Ctr and L2AKO mice treated or not with leupeptin two hours before isolation. Three different sets of lysosomes from three different animals were separately analyzed for purity, integrity, electrophoretic patterning and enrichment in markers of CMA lysosomes by immunoblot. Quantitative proteomics analysis was performed using iTRAQ multiplex (Applied Biomics) in the three animals under the four different conditions: Ctr mice untreated, Ctr mice treated with leupeptin, L2AKO mice untreated and L2AKO mice treated with leupeptin. For each protein hit the average ratio(s) for the protein, the number of peptide ratios that contributed and the geometric standard deviation were determined. Values in the three experimental groups were compared to untreated Ctr and are represented as the average of folds (lysosomes isolated from untreated Ctr mice are given a value of 1). CMA substrate proteins were defined as those for which leupeptin treatment resulted in increase in lysosomal levels >20% and with a reduction in leupeptin response of >20% in the L2AKO.
Other methods. Cell viability was determined by using the CellTiter-Blue Cell Viability Assay kit from Promega (Madison, WI). Protein concentration was determined by the Lowry method (Lowry et al. 1951) using bovine serum albumin as a standard. For immunoblotting, protein concentration was determined by the Lowry method (Lowry et al. 1951) using bovine serum albumin as a standard. After SDS-PAGE, gels were transferred to nitrocellulose membranes using a Mini-TransBlot SD wet transfer cell (Bio-Rad, Richmond, VA) and immunoblotting was performed following standard procedures (Towbin et al. 1979). Proteins recognized by the specific antibodies were visualized by chemiluminescence (RenaissanceR; PerkinElmer Life and Analytical Sciences) using peroxidase conjugated secondary antibodies in a LAS-3000 Imaging System (Fujifilm). Densitometric quantification of the immunoblotted membranes was performed with ImageJ software. Densitometric quantification of the immunoblotted proteins was performed from the TIFF images generated by detection of the chemiluminescent signal after subtraction of background. Although different exposures of each membrane were captured, we quantified those in which none of the bands were saturated. If in order to visualize low abundant proteins, we needed to use exposure times in which some bands were saturated. In this case, two different exposure times were quantified and a common unsaturated band in both exposures was utilized for normalization. Where applicable, purified protein was loaded on the same gel as an input and used for normalization across experiments.

Statistical analysis.
All numerical results are reported as mean + standard error of the mean (s.e.m.) and represent data from a minimum of three independent experiments unless otherwise stated. We determined the statistical significance of the difference between experimental groups in instances of single comparisons by the two-tailed unpaired Student's t-test with the Sigma Plot