Cell culture and chemical treatment
HeLa and HEK293T cells (purchased from ATCC), and MEFs (a gift from Byung-Yoon Ahn, Korea University, Seoul, Korea) were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Capricorn Scientific) containing 10% fetal bovine serum (Capricorn Scientific) and 1% penicillin/streptomycin (Capricorn Scientific). HeLa cells stably expressing CFTR-ΔF50825 were cultured in the normal growth media containing 0.4 mg ml−1 G418.
To induce the formation of an aggresome containing misfolded polypeptides, cells were treated with either MG132 (5 μM; Calbiochem) or, as a negative control, DMSO (BioShop) for 12 h (for IP and immunostaining) or for 16 h (for the TUNEL assay). Where indicated, the cells were also treated with puromycin (10 μg ml−1 for IPs or 1 μg ml−1 for immunostaining) for 1 h, okadaic acid (30 nM) for 12 h, or nocodazole (1 μM) for 12 h before cell harvesting. For line-scan confocal microscopy, the cells were treated with MG132 for < 2 h before visualization.
Plasmocin (Invivogen) was employed to minimize mycoplasma contamination, and the MycoAlert PLUS Mycoplasma Detection Kit (Lonza) was used for mycoplasma detection.
The following plasmids have been described previously: p3xFLAG-CMV™-7.1 (renamed to p3xFLAG in this study; Sigma); pCMV-Myc (Clontech); pRSET A (Invitrogen); pcDNA3-FLAG, pcDNA3-FLAG-CTIF-WT, pCMV-Myc-CTIF, pEGFP-C1-CTIF, pmCMV-GPx1-Norm or -Ter, pGEX-6p-1, and phCMV-MUP24; p3xFLAG-eEF1A1, pCMV-Myc-GST, and pCMV-Myc-DCTN125; pCMV-Myc-UPF1-WT and pCMV-Myc-UPF1(G495R/G497E) renamed to pCMV-Myc-UPF1-HP in this study28; pCMV-Myc-PNRC2, pcDNA3-FLAG-UPF1-WT, and pcDNA3-FLAG-UPF1-(G495R/G497E) renamed to pcDNA3-FLAG-UPF1-HP in this study35; pcDNA3-FLAG-SMG5 and pcDNA3-FLAG-SMG659; pCMV-Myc-DCP1A60; GFP-CFTR-ΔF50817; and pcDNA3.1-CBP80-HA61.
To construct p3xFLAG-GPx1-Norm or -Ter, the GPx1-Norm or -Ter fragment was amplified by PCR using pmCMV-GPx1-Norm or -Ter as a template and two oligonucleotides: 5′-CATGCCATGGCCATGTCTGCTGCTCGGCTCTCCGCGGTGG-3′ (sense) and 5′-GCTCTAGATTAGGGGTTGCTAGGCTGCTTGGACAGC-3′ (antisense). The PCR-amplified and Klenow-filled NcoI/XbaI fragment was ligated into a Klenow-filled HindIII/XbaI fragment of p3xFLAG.
To construct pCMV-Myc-UPF1-HP-12A, the 12 serine or threonine residues (positions 10, 28, 1038, 1041, 1046, 1050, 1055, 1073, 1078, 1089, 1096, and 1116) experimentally proved to be phosphorylated by SMG131 were replaced by alanines in pCMV-Myc-UPF1-HP.
To construct pCMV-Myc-UPF1-HP-E3 mut, the fragment of UPF1 cDNA harboring substitutions S124A/N138A/T139A was synthesized in vitro. Then, the corresponding region in pCMV-Myc-UPF1-HP was replaced by the in vitro–synthesized UPF1 fragment.
To generate pcDNA3-FLAG-UPF1-HP-12A, a Klenow-filled HindIII/NotI fragment of pCMV-Myc-UPF1-HP-12A was ligated into a Klenow-filled Acc65I fragment of pcDNA3-FLAG.
To construct pGEX-6p-1-CTIF, a BamHI/EcoRI fragment of pSK(-)-CTIF24 was ligated to a BamHI/EcoRI fragment of pGEX-6p-1.
To construct pGEX-6p-1-eEF1A1, a Klenow-filled BamHI/SalI fragment of p3x-FLAG-eEF1A125 was ligated into a Klenow-filled NotI/SalI fragment of pGEX-6p-1.
To construct pGEX-6p-1-DCTN1, an EcoRI/BspEΙ fragment containing the 5′ half of DCTN1 cDNA was amplified by PCR with pCMV-Myc-DCTN1 as a template and two oligonucleotides: 5′-CCGGAATTCATGGCACAGAGCAAGAGGCACGTGTAC-3′ (sense) and 5′-TGGCCTGCAGCAGGCTCAGCGAGTAC-3′ (antisense). pCMV-Myc-DCTN1 was digested with BspEΙ and NotΙ to obtain the 3′ half of DCTN1 cDNA. The resulting two fragments were ligated to an EcoRI/NotΙ fragment of pGEX-6p-1.
To construct pRSET A-UPF1, a Klenow-filled BamHI/EcoRI fragment of pCMV-Myc-UPF1-WT was ligated to a Klenow-filled HindIII/EcoRI fragment of pRSET A.
DNA or siRNA transfection
For IP experiments or conventional confocal microscopy, HeLa cells or HEK 293T cells were transfected with a plasmid using calcium phosphate or Lipofectamine 2000 (Invitrogen). MEFs were transfected by means of Lipofectamine 3000 (Invitrogen). For specific downregulation of endogenous proteins, cells were transfected with 100 nM in vitro–synthesized siRNAs via Oligofectamine (Life Technologies) or Lipofectamine 3000 (Invitrogen). The siRNA sequences used in this study are listed in Supplementary Data 1.
Quantitative reverse-transcription PCR (qRT-PCR)
Total-RNA samples were purified with the TRIzol Reagent (Life Technologies) and analyzed by qRT-PCRs. The total-RNA samples were converted into complementary DNA (cDNA) using RevertAid Reverse Transcriptase (Thermo Scientific). qRT-PCR analyses were carried out with gene-specific oligonucleotides and the Light Cycler 480 SYBR Green I Master Mix (Roche) on a Light Cycler 480 II machine (Roche). The following gene-specific oligonucleotides for amplification of mRNAs were used in this study: 5′-GGACTACAAAGACCATGACG-3′ (sense) and 5′-CTTCTCACCATTCACCTCGCACTT-3′ (antisense) for amplification of FLAG-GPx1-Norm or -Ter mRNAs; 5′-TGGCAAATTCCATGGCACC-3′ (sense) and 5′-AGAGATGATGACCCTTTTG-3′ (antisense) for amplification of GAPDH mRNAs; 5′-CAACACCCCAACATCTTCG-3′ (sense) and 5′-CTTTCCGCCCTTCTTGGCC-3′ (antisense) for amplification of FLuc RNAs; and 5′-CTGATGGGGCTCTATG-3′ (sense) and 5′-TCCTGGTGAGAAGTCTCC-3′ (antisense) for amplification of MUP mRNAs.
To validate efficient removal of cellular RNAs after RNase A treatment under our conditions (Supplementary Figs. 4a and 8c), total-RNA samples purified from the extracts either treated or not treated with RNase A before IPs were mixed with equal amounts of in vitro–synthesized FLuc RNAs (10 pg) as a spike-in to adjust the data for differences among RNA preparations. Then, the relative amounts of endogenous GAPDH mRNA (normalized to FLuc RNA) were compared. Microsoft Excel (Microsoft) was used for analyzing of quantitative real-time RT-PCR data.
Antibodies against the following proteins (used for immunostaining, IPs, and western blotting) have been described previously: CTIF and CBP8024; eIF4A3, UPF1, and UPF262; UPF3B and SMG663; phospho-S1078-UPF1 and phospho-S1096-UPF162; and PNRC235.
Antibodies against the following proteins were purchased [listed in the format “protein name (catalog number, supplier, dilution fold)”]: FLAG (DYKDDDDK; 14793, Cell Signaling Technology, 1:50, or A8592, Sigma, 1:5000), HA (3724, Cell Signaling Technology, 1:200), Myc (9E10; OP10L, Calbiochem, 1:5000, or 2272, Cell Signaling Technology, 1:200), GFP (sc-9996, Santa Cruz Biotechnology, 1:50), DCTN1 (p150glued; 610474, BD Biosciences, 1:250), eEF1A1 (CBP-KK1; EF1α; 05-235, Merck Millipore, 1:1000), DCP1A (D5444, Sigma, 1:1000), Y14 (RBM8A; MAB2484, Abnova, 1:1000), SMG1 (A300-394A, Bethyl Laboratories, 1:1000), SMG5 (ab33033, Abcam, 1:500), SMG7 (A302-170A, Bethyl Laboratories, 1:2000), ATM (A300-299A, Bethyl Laboratories, 1:1000), DNA-PKcs (A300-517A, Bethyl Laboratories, 1:1000), p-(S/T)Q ATM/ATR substrate (2851, Cell Signaling Technology, 1:1000), puromycin (12D10; MABE343, Merck Millipore, 1:10000 for western blotting or 1:400 for immunostaining), eIF3b (sc-16377, Santa Cruz Biotechnology, 1:1000), γ-tubulin (sc-17788, Santa Cruz Biotechnology, 1:20), α-tubulin (sc-53030, Santa Cruz Biotechnology, 1:100), GST (A190-122A, Bethyl Laboratories, 1:8000), His (27-4710-01, GE Healthcare, 1:3000), β-actin (A5441, Sigma, 1:10,000), and GAPDH (LF-PA0212, AbFrontier, 1:10,000).
Immunostaining followed by conventional confocal microscopy
Immunostaining was performed on HeLa cells as described elsewhere25. The cells were fixed and permeabilized with 3.65–3.8% formaldehyde (Sigma, cat. # F8775) and 0.5% Triton X-100 (Sigma), respectively. Then, the fixed cells were incubated with 1.5% bovine serum albumin (BSA; BovoStar) for 1 h for blocking and next incubated in phosphate-buffered saline (PBS) containing 0.5% BSA with a primary antibody solution. After that, the cells were incubated in PBS containing 0.5% BSA with the secondary antibodies conjugated to either Alexa Fluor 488 or rhodamine. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; Biotium). LSM 510 Meta, LSM 700, or LSM 800 (Carl Zeiss) was used for visualization. Immunostained cell images were analyzed by Zeiss LSM Image Browser and Zen 2.1 (black; Carl Zeiss) and organized by Adobe Photoshop (Adobe) was used.
Quantitation of cells containing aggresome
In most cases, more than 100 cells were analyzed in each experiment of two (Fig. 2b,c; Supplementary Fig. 7b) or three (Fig. 7b; Supplementary Fig. 13b) biological replicates. For Fig. 4c, more than 50 cells were analyzed in each experiment. Two experienced independent investigators counted and rated the cells in a blinded way. For quantitation of cells containing aggresome or dispersed aggregates, cell images were analyzed by Zeiss LSM Image Browser (Carl Zeiss) and number of cells containing aggresome or dispersed aggregates was analyzed using Microsoft Excel (Microsoft).
Total-cell protein samples were separated by electrophoresis in an SDS-polyacrylamide gel, transferred to Hybond ECL nitrocellulose (Amersham), and probed with a primary antibody. After that, the following secondary antibodies were used to detect the primary antibody: a horseradish peroxidase (HRP)-conjugated goat α-mouse IgG antibody (cat. # AP124P, MilliporeSigma), HRP-conjugated goat α-rabbit IgG antibody (cat. # AP132P, MilliporeSigma), or HRP-conjugated rabbit α-goat IgG antibody (cat. # A5420, MilliporeSigma). Western blot images were obtained on a chemiluminescence imaging system (Amersham Imager 600, GE Healthcare). For organizing western blotting images, Adobe Photoshop (Adobe) was used. Signal intensities of western blot bands were quantitated in the ImageJ software (National Institutes of Health, Bethesda, MD).
Cell pellets were resuspended in NET-2 buffer [50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride (PMSF; Sigma), 2 mM benzamidine hydrochloride (Sigma), 0.05% NP-40 (IGEPAL® CA-630; Sigma), 10 mM sodium fluoride (Sigma), and 0.25 mM sodium orthovanadate (Sigma)]. The resuspended cells were sonicated and centrifuged at 13,800 × g for 10 min at 4 °C. For preclearing, protein G agarose 4B beads (Incospharm) were mixed with the cell extracts for 1 h at 4 °C. The precleared samples were incubated with various antibody-bound beads for 3 h at 4 °C. The beads were washed with NET-2 buffer five times, and the bead-bound proteins were eluted with 2× sample buffer [10% β-mercaptoethanol, 4% SDS, 100 mM Tris-HCl (pH 6.8), 15% glycerol, and 0.008% bromophenol blue].
The in vitro GST pull-down assay
This assay was carried out using recombinant GST, GST-CTIF, GST-eEF1A1, GST-DCTN1, and 6xHis-UPF1. Escherichia coli BL21(DE3)pLysS cells were transformed with the plasmid encoding either a GST-fusion protein or 6xHis-UPF1. The recombinant protein was induced by the addition of 0.5 mM isopropyl β-d-1-thiogalactopyranoside and additional incubation for 2 or 3 h. After that, the cells were harvested and resuspended in lysis buffer [50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.5% Triton X-100, 1 mM dithiothreitol (DTT), 10% (v/v) glycerol, 2 mM benzamidine, and 1 mM PMSF] and sonicated. The E. coli lysate expressing a GST-fusion protein was mixed with the lysate expressing 6xHis-UPF1. The mixture was incubated in incubation buffer [10 mM HEPES (pH 7.4), 1.5 mM magnesium acetate, 150 mM potassium acetate, 2.5 mM DTT, and 0.05% NP-40] for 30 min at 4 °C. Then, the glutathione Sepharose 4B resin was added to the mixture and incubated for 2 h. The protein-bound resin was washed four times with incubation buffer. The resin-bound proteins were resolved by SDS-PAGE followed by western blotting.
Line-scan confocal microscopy and image acquisition
Multicolor video rate line-scan confocal microscopy was performed for single-particle tracking in live cells43. GFP-CTIF and SiR-conjugated tubulin (SiR-tubulin, Spirochrome, SC002) were excited by two lasers (Cobolt MLDTM 488 nm, 60 mW; Cobolt MLDTM 638 nm, 100 mW), respectively. The incident beams and emission from the fluorophores were reflected and transmitted by a dichroic beam splitter (Semrock, Di01-R405/488/561/635). Two galvano scanning mirrors (Cambridge Technology 6231H, 15 mm) scanned the field of view with the linear focus and reproduced the focal image in the camera. A high NA objective (Olympus UPlanSApo 100×, NA = 1.4, oil immersion) in an inverted microscope (Olympus IX51) focused the excitation lasers and gathered the emitted photons. The emitted light was filtered by wavelengths (Chroma, 59007m for SiR and Semrock FF01-520/35-25 for GFP) using a DV2 multichannel imaging system (Photometrics) that was placed in front of the electron-multiplying charged-couple device (EMCCD) camera (Andor iXon Ultra). All images were recorded by the SOLIS imaging software (Andor) every 100 ms, while the recording and scanning were synchronously controlled by a LabView script. To maintain the viability of cells during fluorescence imaging, the imaging was performed in a CO2 incubation chamber at 37 °C (Live Cell Instrument). To colocalize the SiR-tubulin (microtubule) and GFP-CTIF particles, the image of fluorescent beads on the imaging plane was acquired as a reference for the mapping procedure.
Analysis of GFP-CTIF aggregates migrating toward the aggresome
The position of GFP-CTIF aggregates was determined with nanometer accuracy in DiaTrack 3.03 software64. The data were analyzed by means of Origin Pro 8 (OriginLab) and MATLAB script (2017a, Mathworks). The translocation rate of each spot was calculated from its trajectory with a continuous motion except for stalled segments. Trajectories moving less than or equal to 5 pixels (118 nm per pixel) for 100 ms (time resolution) were analyzed. To determine the frequency of GFP-CTIF aggregates moving into the aggresome, the first 150 frames were analyzed. From the signals, we subtracted background noise and filtered the data by means of the convolution function in ImageJ (National Institutes of Health). The distribution of the angle (θ) values was obtained from the trajectories with both the distance of greater than or equal to 9 pixels and lifetime longer than 5 points (500 ms). The histogram was produced by automatic splitting of the data range into bins of equal size in Origin Pro 8.
The TUNEL assay
The In Situ Cell Death Detection Kit (Roche) was utilized for the TUNEL assay. In brief, cells were fixed with 3.65–3.80% formaldehyde for 1 h. Endogenous peroxidase activity was inhibited by the addition of 3% hydrogen peroxide (Sigma). The cells were permeabilized with 0.1% Triton X-100. Then, DNA strand breaks were labeled with tetramethylrhodamine red. Nuclei were stained with DAPI. The cells were visualized under a Zeiss confocal microscope (LSM 510 Meta and LSM700, Carl Zeiss). For quantitation of apoptotic cells, more than 100 cells were analyzed for each experiment on three biological replicates.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.