TBC1D14 regulates autophagy via the TRAPP complex and ATG9 traffic

Abstract Macroautophagy requires membrane trafficking and remodelling to form the autophagosome and deliver its contents to lysosomes for degradation. We have previously identified the TBC domain‐containing protein, TBC1D14, as a negative regulator of autophagy that controls delivery of membranes from RAB11‐positive recycling endosomes to forming autophagosomes. In this study, we identify the TRAPP complex, a multi‐subunit tethering complex and GEF for RAB1, as an interactor of TBC1D14. TBC1D14 binds to the TRAPP complex via an N‐terminal 103 amino acid region, and overexpression of this region inhibits both autophagy and secretory traffic. TRAPPC8, the mammalian orthologue of a yeast autophagy‐specific TRAPP subunit, forms part of a mammalian TRAPPIII‐like complex and both this complex and TBC1D14 are needed for RAB1 activation. TRAPPC8 modulates autophagy and secretory trafficking and is required for TBC1D14 to bind TRAPPIII. Importantly, TBC1D14 and TRAPPIII regulate ATG9 trafficking independently of ULK1. We propose a model whereby TBC1D14 and TRAPPIII regulate a constitutive trafficking step from peripheral recycling endosomes to the early Golgi, maintaining the cycling pool of ATG9 required for initiation of autophagy.

The clarified lysate was incubated with 30 ul protein G-sepharose to pre-clear the lysate for half an hour, then with the antibody-bead complexes for two hours at 4 °C.
In both cases, the beads were washed three times with lysis buffer, then resuspended in 30 μl 2 x Laemmli Sample Buffer (LSB) and subjected to SDS-PAGE.
For figure EV2E, HEK293A cells overexpressing myc-TBC1D14 or transfected with empty vector (pcDNA3.1) and co-transfected with plasmids encoding either GFP or GFP-Golgin 84 were lysed in RAB lysis buffer (20 mM Na HEPES ph 7.4, 150 mM NaCl, 5 mM MgCl2, 1 mM DTT, 0.3% Triton X-100 plus PI), subject to GFP-Trap IP and immunoblotted for GFP, RAB1B and myc. The same conditions were used in Figure EV2F, with HEK293A cells transfected with the indicated siRNA duplexes and co-transfected with GFP-Golgin 84 and blotting for GFP, RAB1B, TBC1D14 and TRAPP subunits.

BioID analysis
An adapted version of the original Bio ID protocol (Roux et al., 2012) was used to identify TRAPP subunits proximal to TBC1D14. 4 x 15 cm dishes (approximately 20 million cells/plate) of HEK293A cells were transfected with myc-BioID or myc-BioID-TBC1D14. After transfection, the cells were recovered in DMEM + 50 μM Biotin.
The following day, the cells were harvested in PBS and lysed in 1.5 ml TNTE + PI and centrifuged to pellet insoluble components. The lysates were split and denatured at room temperature with 1% SDS (D) or not, by adding an equivalent volume of TNTE + PI (ND). The lysates were incubated with streptavidin-agarose (Pierce) for 1 hour at room temperature to precipitate biotinylated proteins. Bound proteins were eluted in 2x LSB + 3 mM Biotin. For mass spectrometry analysis, the samples were subjected to SDS-PAGE, stained with colloidal Coomassie (Invitrogen), lanes excised and analysed by mass spectrometry. For Western blot analysis, the eluted proteins were blotted with anti-myc, streptavidin-HRP and TRAPP antibodies.

Mass spectrometry analysis
For mass spectrometry analysis, the samples were subjected to SDS-PAGE, stained with colloidal Coomassie (Invitrogen) and eight bands covering the entire molecular weight range were excised. In-gel trypsin digestion was performed overnight at 37 °C using a Perkin Elmer Janus liquid handling system. Peptide extracts were acidified to 0.1% TFA and subjected to LC-MS analysis using an Ultimate3000 nano HPLC system connected to a Q-Exactive mass spectrometer (Thermo Scientific).
The instrument was operated in a top 10 data dependent acquisition mode. Raw mass spectrometry data was processed using MaxQuant version 1.3 (Cox & Mann, 2008). Data was searched against a Uniprot FASTA database containing human sequences and intensity based absolute quantification (iBAQ) (Schwanhausser et al., 2011) was used for label free quantification. iBAQ values were log 10 transformed and missing values were imputed by drawing from a random distribution using default settings (width 0.3, down shift 1.8) in Perseus software version 1.4.0.2. iBAQ scatter plots were visualised using R software.

Size exclusion chromatography
For figure EV2D, approximately 2 × 107 HEK293T cells were homogenised using a syringe and needle in 500 μl homogenisation buffer, the nuclei removed by 2 × 3 minute centrifugations at 2500 rcf and 4 °C, and membranes removed by ultracentrifugation for 1 hour at 100,000 rcf. The resulting supernatant was used in size exclusion chromatography with a Superose 6 column (GE Healthcare), with an elution volume of 1.2 column volumes and collecting 0.5 ml fractions. Proteins from fractions were concentrated using 10 μl Strataclear beads (Stratagene) and bound proteins eluted in 30 μl 2 x LSB. Fractions 19-40 were subjected to immunoblot analysis with the indicated antibodies, alongside 1% of the input.