Cell culture
PC12 cells were cultured in high glucose DMEM (Gibco) with 10% heat inactivated horse serum (Gibco), 5% fetal bovine serum (Atlanta Biologicals) and 1% Penicillin/Streptomycin (Gibco). Cells were treated with 0.1 ug/mL nerve growth factor (Bon Opus) for ~48 hours in media containing 1% heat inactivated horse serum and 1% antibiotic to induce differentiation. All cells were incubated at 37°C in 5% CO2. Cells below passage 10 were used for all experiments.
Plasmids
EGFP-hArgonaute2 (eGFP-Ago2) was purchased from Addgene (plasmid # 21981) and was prepared in the laboratory of Philip Sharp (MIT). G3BP1-GFP was purchased from Addgene (plasmid # 119950) and was prepared in the laboratory of Jeffrey Chao (Friedrich Miescher Institute). siRNA(PLCβ1) was from Dharmacon (smartpool # L-092936-02-0005). Plasmid transfections and siRNA knockdowns were done using Lipofectamine 3000 (Invitrogen) in antibiotic free media. Medium was changed to one containing antibiotic (1% Penicillin/Streptomycin) 4-12 hours post-transfection.
Stress conditions
Cells were exposed to carbachol or isoproterenol stimulation at a final concentration of 5 µm for 10 minutes. For hypo-osmotic stress, the culture medium was diluted from 300 mOsm to 150 mOsm with 50% water for 5 minutes. For heat shock, cells were incubated at 42°C for 1 hour.
Number and Brightness (N&B) measurements
N&B defines the number of photons associated with a diffusing species by analyzing the variation of the fluorescence intensity in each pixel in the cell image, and full details about the method and analysis can be found in [17]. In these studies, we collected ~100 cell images viewing either free eGFP (control) or eGFP-G3BP1 at a 66 nm/pixel resolution and at a rate of 4 µs/pixel (see [12]). Number and Brightness (N&B) data were analyzed with SimFC4 software (www.lfd.uci.edu) where the brightness, B, in each pixel refers to the ratio of the variance, σ, over the average fluorescence intensity <k>
where n is the number of fluorophores. The variance in each pixel is obtained by rescanning the cell image ~100 times. The average fluorescence intensity, <k>, is directly related to the molecular brightness, ε, in units of photons per second per molecule.
Particle analysis
Samples were imaged with a 100X/1.49 oil TIRF objective to microscopically count the number of particles per µm2 formed under different conditions. For each condition, 10 to 20 cells were randomly selected, and z-stack measurements were taken (1.0 µm/frame). Analysis was performed with ImageJ and Fiji ImageJ software either by thresholding before analyzing, and averaging the number of particles per frame per measurement or by combining z-stack measurements to generate a 3D image for each sample before analyzing the number of particles per sample and averaging the results. Both methods produced identical results.
Mass spectrometry
Ago2 stress granules were purified from PC12 cells electroporated with eGFP-Ago2 then differentiated with NGF for ~48 hours using the protocol described by Wheeler et al.. [43]. Six 150 mm dishes of cells were used per condition. Immunoprecipitation was done using GFP antibody (Santa Cruz Biotechnology, sc-9996) with protein A dynabeads and elution buffer (Invitrogen,cat #10006D). Following short gel electrophoresis and band excision, mass spectrometry was performed at UMass Medical School Mass Spectrometry Facility using the methodology previously described [12].
eCLIP transcript sequences
PC12 cells were transfected with eGFP-Ago2 using electroporation (Bio-Rad Gene Pulser Xcell) with one pulse in a 0.4 cm gap cuvette (Bio-Rad, cat #1652088) in complete media and plated at ~50% confluency to leave room for neurite growth. Cells were stressed and crosslinked using 254 nm ultraviolet light. Treated cells were collected from at least four 150 mm culture dishes per sample condition. Lysis was done on fresh, non-frozen cell pellets using 100 µL lysis buffer supplemented with RNase inhibitor and protease inhibitor tablets (Roche). A 10 µL portion of each lysate was used to extract total RNA (Zymo Clean and Concentrator) which was then tested for RNA integrity using RNA Fragment Analyzer services at University of Massachusetts Medical School Molecular Biology Core Lab. The remaining lysate was immunoprecipitated using mouse Ago2 antibody. Subsequent western blots, RNA cleanup and library construction was done using Eclipse Bioinnovation protocol v1.01R, storing samples at -80°C overnight when necessary. Sample quantity was evaluated by qPCR. Libraries were sequenced at UMass Medical School Deep Sequencing Core using Illumina HiSeq 4000 system. All reagents were supplied by Eclipse Bioinnovations unless otherwise specified.
RT-PCR: For each condition, one 100 mm dish was used for total RNA extraction. Cells were differentiated and stressed before isolating RNA using RNeasy Mini Kit (Qiagen, Cat #74104) with DNase digestion (Qiagen, cat #79254). CDNA synthesis was done using MiniAmp Thermal Cycler (Applied Biosystems). Purified RNA sample concentrations were measured by Nanodrop and 1 µg RNA was incubated with dNTP (New England Biolabs, cat #N0447S), random hexamers (Thermo Scientific, cat #SO142), and nuclease free water (Invitrogen, cat #AM9930), at 65°C for 5 minutes. After the first incubation, 5x First Strand Buffer, DTT (Invitrogen cat #18080093) and RNase inhibitor (Invitrogen, cat #AM2694) were added to the mixture and incubated at 25°C for 2 minutes. After the second incubation, 1 µL Superscript III Reverse Transcriptase (Invitrogen, cat #18080093) was added and the mixture was incubated at 25°C for 10 minutes, 42°C for 50 minutes, and 70°C for 15 minutes. CDNA was stored at -20°C overnight or used immediately at a dilution of 1:20 in nuclease free water for RT-PCR experiments with Luna Universal Master Mix (New England Biolabs, cat #M3003L), nuclease free water, and custom ATP5f1b and CHGb primers (ITD). For positive and negative controls, 18S ribosomal subunit and tau primers were used (see Supplemental).
Western blotting: We used an establish procedure [40] with modifications. Samples were grown in 100 mm dishes plates and were treated with MG132 at a final concentration of 250 nM 24 hours before collection. Cells were stressed and then collected in 250 µl of NP-40 lysis buffer [150 mM NaCl, 50 mM Tris (pH=8.0), 1% NP40, protease inhibitor]. The primary antibodies used included: anti-ATP5f1b (Abcam, ab117991), anti-Chgb (Abcam, ab150354), anti-actin (Santa Cruz, sc-47778).
Fluorescence lifetime imaging measurements (FLIM)
Phase modulation FLIM measurements were performed on the dual-channel confocal fast FLIM (Alba version 5, ISS Inc.) using a two-photon titanium-sapphire laser and a Nikon Eclipse Ti-U inverted microscope as described[40]. The lifetime of the laser was calibrated each time before experiments by measuring the lifetime of Atto 425 in water with a lifetime of 3.61 ns at 80, 160, and 240 MHz. The samples were excited at 740 nm for intrinsic measurements, and emission spectra were collected through a 525/50 bandpass filter. For each measurement, the data were acquired until the photon count was >100.
Statistical analysis
Data were analyzed with the Sigma Plot 14 statistical packages, which included the student’s t test and one-way analysis of variance (ANOVA).