Proteostasis failure and cellular senescence in long‐term cultured postmitotic rat neurons

Abstract Cellular senescence, a stress‐induced irreversible cell cycle arrest, has been defined for mitotic cells and is implicated in aging of replicative tissues. Age‐related functional decline in the brain is often attributed to a failure of protein homeostasis (proteostasis), largely in postmitotic neurons, which accordingly is a process distinct by definition from senescence. It is nevertheless possible that proteostasis failure and cellular senescence have overlapping molecular mechanisms. Here, we identify postmitotic cellular senescence as an adaptive stress response to proteostasis failure. Primary rat hippocampal neurons in long‐term cultures show molecular changes indicative of both senescence (senescence‐associated β‐galactosidase, p16, and loss of lamin B1) and proteostasis failure relevant to Alzheimer's disease. In addition, we demonstrate that the senescent neurons exhibit resistance to stress. Importantly, treatment of the cultures with an mTOR antagonist, protein synthesis inhibitor, or chemical compound that reduces the amount of protein aggregates relieved the proteotoxic stresses as well as the appearance of senescence markers. Our data propose mechanistic insights into the pathophysiological brain aging by establishing senescence as a primary cell‐autonomous neuroprotective response.

Data S1 Experimental Procedures.

Immunoblotting
Cells were lysed by SDS sample buffer containing β-mercaptoethanol. Whole cell extracts were loaded onto 4-14% SDS-PAGE gels and blotted on a PVDF membrane (Millipore). After blocking in 5% skim milk, 3%BSA, or Blocking One (Nacalai) at room temperature, the blots were incubated O/N with specific antibodies, probed with horseradish peroxidase-linked secondary antibodies (GE Health Care) for 1 h, and detected by ECL reagents (GE Health Care).

Detergent-insoluble protein fractionation and quantification
Triton-insoluble fractions were prepared as previously described (Bartlett et al., 2011). In brief, cells were extracted in ice-cold lysis buffer (50 mM NaCl, 5 mM EDTA, 0.1%SDS, 1% Triton X-100, 10 mM Tris-HCl (pH7.4), 1 mM Na3VO4, 1 x phosphatase inhibitor cocktail, 30 mM β-Glycerophosphate), centrifuged at 14,000 rpm for 10 min, and the supernatants were collected as the soluble fraction. The remaining pellets were extracted with 2% SDS-containing sample buffer (insoluble fraction). The soluble and insoluble fractions were subjected to SDS-PAGE to detect actin and ubiquitinconjugates with specific antibodies, respectively. Soluble actin was used as a loading control (Dai et al., 2015), and relative amounts of actin in each soluble fraction dictated the respective volumes of the insoluble fractions to be loaded. After detection of the ubiquitin conjugates, the PVDF membrane with the blot of the insoluble fraction was rinsed in TNT buffer (20 mM Tris-HCl (pH7.5), 140 mM NaCl, 0.05% Tween 20 (vol/vol)) and further subjected to Coomassie staining to determine total amounts of the detergent-insoluble proteins. The immunoreactive signals and levels of the total insoluble proteins were quantified by using Image J software.

ELISA for measurement of Aβ42
Supernatants were collected every seven days after initiating the neuronal cultures and stored at -80°C until use. Secreted Aβ42 across the LTC of PHNs was analyzed in accordance with the manufacturer's instructions for the Human/Rat Amyloid (42) ELISA kit from Wako (Wako,.

Immunofluorescence
Cells growing on coverslips were fixed with 4% paraformaldehyde (PFA) at room temperature for 15 min, or methanol on ice for 15 min. PFA-fixed cells were permeabilized by 0.5% triton buffer (20 mM HEPES (pH7.4), 50 mM NaCl, 3 mM MgCl2, 0.3 M sucrose, 0.5% Triton X-100 (vol/vol)) for 5 min. After two rinses with PBS, cells were blocked in PBS containing 0.1% skim milk (wt/vol) and 0.1% bovine serum albumin (wt/vol) for 0.5 to 1 h. After that, cells were incubated with primary antibodies for 1 h at room temperature. After three PBS washes, cells were incubated with secondary antibodies conjugated with Alexa488 or Cy3 for 1 h at room temperature. Nuclei were counterstained with 1 μg/mL 4',6-diamidino-2-phenylindole (DAPI). For Thio-S staining, blocked samples were incubated with 0.025% Thio-S (wt/vol) for 5 min followed by three 5 min washes in 80% ethanol and a 5 min wash in PBS. After that, samples were incubated with anti-MAP2 as a primary antibody and then the appropriate secondary antibody. Fluorescence images were acquired using a high-resolution microscope DeltaVision Elite (GE Health Care) and were processed to projections of z-sections. The obtained images were subsequently composed and edited in softWoRx software, in which the background was reduced by brightness and contrast adjustments applied to the whole images. Fluorescence intensity of selected proteins for MAP2-or NeuN-positive neurons was determined using Image J software and the mean of the relative fluorescence was represented as MFI. To quantify nuclear levels of REST, individual nuclei in MAP2-positive neurons were manually outlined and the mean of relative fluorescence in the reference channel (Alexa Fluor 488) for the specified nuclear region was measured using Image J.

RT-qPCR
Total RNA was prepared from cells using an RNeasy Mini kit (Qiagen) and transcribed into cDNA using AMV reverse transcriptase (Life Sciences) and an oligo-dT primer.
Autophagic flux assay 10 nM bafilomycin was used to inhibit autophagy for 4 h at 37°C and to detect accumulation of autophagosomes by immunoblotting of LC-3 and immunofluorescence of p62 (see Garcia-Prat et al., 2016).

De novo protein synthesis assay
Newly synthesized proteins were detected with a Click-iT HPG Alexa Fluor 488 Protein Synthesis Assay Kit (Thermo Fischer Scientific) according to the manufacturer's protocol. In brief, rat primary hippocampal neurons were maintained in Neurobasal media supplemented with B27 in the absence or presence of 100 nM rapamycin or 100 nM CHX from 4 DIV to 14 DIV (i.e., for 10 days). At 14 DIV, the media was replaced with L-methionine-free media containing 50 μM Click-iT-labeled HPG, and the cells were incubated for 30 min. The cells were then rinsed and fixed with 4% PFA for 15 min. Following Click-iT reaction with Alexa Fluor 488 dye, and counterstaining with NuclearMask, images were acquired using DeltaVision Elite, and were analyzed by Image J.

Lentiviral infection
293T cells were maintained in DMEM (Nissui) supplemented with 10% FBS (GIBCO), 2 mM L-glutamine (Sigma), 96 U ml -1 Penicillin (Sigma), and 72 U ml -1 Streptomycin (Sigma). 293T cells were transfected with Lipofectamine 2000 (Invitrogen) for shRNA knockdown or Polyethylenimine Max (Polysciences, Inc.) for cDNA expression with 9 μg DNA following the manufacturer's instructions. Thirty-six hours after transfection, the viral supernatant was collected and concentrated using Amicon Ultra-15 100K centrifugal filters (Millipore), yielding aliquots of lentiviral particles, which were stored at -80°C until use. At 3 DIV, hippocampal neurons were subjected to lentiviral infection for 24 h by replacing half of the medium with fresh Neurobasal B27 medium containing the lentiviral particles. 24 h after infection, the medium was changed to treat the neurons with AraC for an additional 24 h.

Statistical analysis
We conducted two-tailed unpaired t-tests, one-way ANOVA, Two-way ANOVA, and Mann-Whitney tests using GraphPad Prism software. Bonferroni or Tukey's multiple comparisons were used for post-hoc tests following ANOVA. Alternatively, cells were exposed to 50 mM KCl for an additional 12 h with APV, followed by fixation (12 h Figure S5. LTC-PHNs are resistant to apoptosis (A), (B) Fragmented or condensed nuclei of (apoptotic) neurons that were treated with 16 μM etoposide (A) or 100 μM H2O2 (B) at 7 or 28 DIV were counted. 153 to 543 nuclei counterstained by DAPI were analyzed. Stress treatments were performed as shown in Fig. 6A. The means ± SEM of three independent experiments are shown. Arrows indicate condensed nuclei of apoptotic cells, while arrowheads represent fragmented nuclei. Scale bar, 15 μm. Two-way ANOVA for statistical analyses (*p < 0.04; **p < 0.008). (C) PCNs at 14 or 28 DIV were exposed to either DMSO (0 μM) or etoposide (ETOP) (5 -40 μM). Cell survival of young (blue) and senescent cells (orange) without stress treatment (0 μM) was defined as 100%. Mann-Whitney U-test for (A); Two-way ANOVA for (C) (*p = 0.0056; **p = 0.0032; ***p = 0.0011). Post-mitotic hippocampal and cortical neurons display changes associated not only with aging brain, but also with conventional cellular senescence after long-term cultures. Proteostasis failure, possibly due to defective autophagy, is a potential contributor to the age-associated changes. Pharmacological interventions in proteostasis (e.g. rapamycin and EPPS) rescue cellular senescence in post-mitotic neurons. This senescence phenomenon in post-mitotic neurons is accompanied by stress resistance, which may serve as a safeguard against neurodegeneration under age-associated stress, thereby conferring life-lasting survival of neurons during physiological aging.