HeLa cells were cultured in DMEM (Hyclone) and 10% fetal bovine serum (FBS, Gibco) supplemented with 1% penicillin-streptomycin (Gibco) at 37°C and 5% CO₂. For starvation, cells were washed with PBS three times and incubated with Earle’s balanced salt solution (EBSS, Gibco) for 2–h at 37°C. Transfection of plasmids was carried out with Lipofectamine 3000 (Invitrogen) according to the manufacturer’s protocols. Transfection of small interfering RNAs (siRNAs) was carried out with Lipofectamine RNAi MAX (Invitrogen). To knockdown CK1δ, double-stranded siRNAs were purchased from GenePharma. The following sequences were used: human CK1δ siRNA 5′-CGACCUCACAGGCCGACAATT-3′ and control siRNA 5′-UUCUCCGAACGUGUCACGUTT-3′.

Yeast cells were grown at 25°C in yeast extract peptone dextrose media (YPD; 1% yeast extract, 2% peptone, and 2% dextrose) or synthetic minimal media (SMD; 0.67% yeast nitrogen base, 2% dextrose, and auxotrophic amino acids, as needed). To induce starvation, yeast cells were transferred to SD(–N) medium (0.17% yeast nitrogen base without amino acids and 2% dextrose) or treated with 400 ng/ml rapamycin.

Total RNA was extracted from CK1δ-knockdown HeLa cells using the Cultured CellTotal RNA Extraction Kit (TIANGEN), and cDNA was reverse-transcribed using the FastQuant RT Kit(TIANGEN). Quantitative PCR was carried out on a Step One PlusTM RealTime PCR system using SuperReal PreMix Plus (TIANGEN). Data were normalized to the expression level of β-actin. Results are representative of at least three experiments. The following primers were used: F-CK1 Delta, 5′-CTCCGTGTTCCGTTTC-3′; R-CK1 Delta, 5′-TGCTACTCGCCATCCT-3′; F-GAPDH, 5′-GGCATCCTGGGCTACACTGA-3′; R-GAPDH, 5′-GTGGTC GTTGAGGGCAATG-3′.

Total proteins were extracted from HeLa cells with RIPA Lysis Buffer (Solarbio) supplemented with 1 mM PMSF, and incubated for 30 min on 4°C. Cell lysates were centrifuged at 12000 g for 30 min at 4°C. Supernatants were separated by SDS-PAGE and transferred onto a PVDF membrane, followed by incubation with primary and secondary antibodies (Table 1); then, the PVDF membrane was visualized using an ECL kit (Millipore). Results are representative of at least three experiments. The relative levels of p62 and LC3-II were normalized to β-actin levels.

HeLa cells grown on cover slips were fixed with 4% paraformaldehyde for 20 min. Cells were permeabilized in 10 μg/mL digitonin (Sigma) for 15 min at 25°C. After blocking with 5% goat serum for 60 min at 25°C, cells were incubated with the indicated primary antibodies overnight at 4°C. After three washes with PBS, cells were stained with FITC or Rhodamine-labeled secondary antibodies for 1 h at room temperature. Nuclei were stained with 0.2% Hoechst 33258 for 30 min. Images were acquired using a FV3000 confocal microscope (Olympus). The colocalization was measured using the ImageJ software.

HeLa cells were starved in EBSS for 2 h, treated with 100 nM bafilomycin A1 for 6 h, and suspended in ice-cold homogenization buffer (composed of 20 mM HEPES, 0.22 M mannitol, 0.07 M sucrose, and protease inhibitors; pH 7.4). Cells were then passed 10 times through a 27-gauge needle using a 1 ml injection syringe, and the post-nuclear supernatant (PNS) was recovered by centrifugation at 4500 g and at 4°C, for 10 min. The supernatant was then centrifuged again at 100,000 g for 30 min to acquire the pellet fraction. The pellet was resuspended in homogenization buffer and then treated with 100 μg/ml proteinase K (ProK) with or without 0.5% Triton X-100. After incubation for 20 min on ice, 10% Trichloroacetic acid (TCA) was added, and the samples were centrifuged at 15,000 g for 10 min. The pellet was washed with ice-cold acetone, resuspended in SDS-PAGE sample buffer, and boiled at 100°C for 10 min. Proteinase K digestion products were detected by immunoblotting.

To analyze phagophore formation, yeast cells were fixed with 3.7% formaldehyde at 25°C for 30 min, and then visualized by super-resolution structured illumination microscopy (SIM), at 25°C, on an Applied Precision DeltaVision OMX Super Resolution System using an Olympus UPlanSApo 100 × 1.4 NA oil objective. The data were acquired and processed using Delta Vision OMX Master Control software and SoftWoRx reconstruction and analysis software.

The significance of differences was evaluated by an unpaired two-tailed t test. Data were analyzed for statistical significance after at least three repeated experiments. Threshold for statistical significance for each test was set at 95% confidence (p < 0.05). Error bars represent SEM. ^∗p < 0.05, ^∗∗p < 0.01. ^∗∗∗p < 0.001.

To determine whether CK1δ is essential for autophagy, we used siRNAs to deplete HeLa cells of CK1δ, and examined the autophagic flux by monitoring the level of p62, an autophagy substrate (Klionsky et al., 2016). As shown in Supplementary Figure S1, CK1δ was efficiently depleted. CK1δ depletion resulted in increased levels of p62 (Figures 1A,B). Immunostaining of p62 also showed that, the number of p62 puncta was higher in CK1δ-depleted cells than that in control cells under starvation conditions (Figures 1D,E), suggesting the blockade of autophagic flux.

Following the induction of autophagy, the non-lipidated form of LC3 (LC3-I) is conjugated to PE to form the lipidated LC3-II form, which associates with autophagic membrane structures and forms punctate structures. Analysis of LC3-II levels serves as a useful assay to determine which step of autophagy is affected (Klionsky et al., 2016). We examined the levels of LC3 II in CK1δ-depleted cells under starvation conditions with or without bafilomycin A1, an autophagosome-lysosome fusion inhibitor. As shown in Figures 1A,C, CK1δ depletion resulted in increased levels of LC3-II, indicating that the defect in autophagic flux occurs after LC3 lipidation. The addition of bafilomycin A1 significantly increased the LC3-II levels in CK1δ-depleted cells, suggesting that autophagosome-lysosome fusion is not blocked by CK1δ depletion. LC3 immunostaining revealed that the number of LC3 puncta was increased in CK1δ-depleted cells (Figures 1F,G). The non-lipidated form of LC3-I is diffuse, and only the lipidated LC3-II form associates with autophagic membrane structures; therefore, these data indicate that LC3-II-positive structures accumulate in CK1δ-depleted cells. These data are consistent with results in temperature-sensitive yeast hrr25-5 mutant cells; in a previous study, we observed numerous punctate GFP-Atg8 structures in the hrr25-5 mutant, despite the decreased autophagic activity in these cells (Wang et al., 2015).

Collectively, these results indicate that CK1δ depletion impairs autophagic flux downstream of LC3 lipidation and leads to the accumulation of LC3-II-positive structures.

To determine which step of the autophagy pathway is affected by CK1δ depletion, we examined the stage of the autophagic structures observed in CK1δ-depleted cells and hrr25-5 mutant cells. The tandem red fluorescent protein (RFP)-green fluorescent protein (GFP)-LC3 reporter was transfected into CK1δ-depleted cells and control cells as previously described. The GFP fluorescence signal is quenched in acidified compartments; accordingly, prior to fusion with the lysosome, the isolation membranes and immature autophagosomes decorated with RFP-GFP-LC3 are visible as yellow puncta, whereas red puncta represent acidified autolysosomes (Kimura et al., 2007). We found that after 2 h of starvation, larger numbers of yellow puncta had accumulated and very few red puncta had formed in CK1δ-depleted cells (Figures 2A–C). These results indicate that CK1δ depletion causes a defect in the progression of isolation membranes to autolysosomes, and that accumulated LC3 puncta in CK1δ-depleted cells represent isolation membranes or autophagosomes, not autolysosomes.

To verify that the LC3-positive structures in CK1δ-depleted cells represented isolation membranes or autophagosomes, we performed a protease protection assay, which is based on the accessibility of GFP-LC3 or p62, sequestered in autophagosomes or not, to protease (Klionsky et al., 2016): when GFP-LC3 and p62 are sequestered by sealed autophagosomes, they are inaccessible to the protease; by contrast, GFP-LC3 and p62 in isolation membranes are accessible to the enzyme. We found that the sensitivity of GFP-LC3 and p62 to ProK was enhanced in CK1δ-depleted cells (Figures 2D,E). These results confirm that the LC3-positive structures in CK1δ-depleted cells are isolation membranes, revealing a role for CK1δ in autophagosome completion.

Consistent with the results in CK1δ-depleted cells, SIM revealed that elongated isolation membranes were accumulated in the temperature-sensitive hrr25-5 mutant at 37°C, a non-permissive temperature. As shown in Figures 2F,G, the number of isolation membranes was significantly increased in the hrr25-5 mutant compared with that in the wild type. The percentage of cells with isolation membrane was also significantly higher in the hrr25-5 mutant (Figure 2H). Together, these findings show that CK1δ depletion or the hrr25 mutation results in a defect in the progression of the autophagy pathway from the formation of isolation membranes to sealed autophagosomes.

Upon autophagy induction, ATG proteins are recruited to the autophagosome formation site in a hierarchical order. Most ATG proteins disassociate from the autophagic membrane structures as autophagosomes close, whereas the lipidated form of LC3/Atg8 associates with autophagic structures at all stages (Itakura and Mizushima, 2010; Lamb et al., 2013). Failure of isolation membranes to seal and form closed autophagosomes results in ATG proteins, which act during the early stages of autophagy, becoming trapped in the isolation membranes along with LC3. Accordingly, we examined the colocalization or association of LC3 puncta with multiple ATG proteins, including ATG9A, ATG14L, and ATG16L1, by immunofluorescence. As shown in Figure 3, the association of LC3 with all three ATG proteins increased in CK1δ-depleted cells although the numbers of ATG9A, ATG14L, and ATG16L1 puncta in CK1δ-depleted cells were comparable to those in control cells (Supplementary Figure S2). These data are consistent with the hypothesis that CK1δ regulates autophagosome completion.