Acetylation of SCFD1 regulates SNARE complex formation and autophagosome-lysosome fusion

ABSTRACT

SCFD1 (sec1 family domain containing 1) was recently shown to function in autophagosome-lysosome fusion, and multiple studies have demonstrated the regulatory impacts of acetylation (a post-translational modification) on macroautophagy/autophagy. Here, we demonstrate that both acetylation- and phosphorylation-dependent mechanisms control SCFD1ʹs function in autophagosome-lysosome fusion. After detecting a decrease in the extent of SCFD1 acetylation under autophagy-stimulated conditions, we found that KAT2B/PCAF catalyzes the acetylation of residues K126 and K515 of SCFD1; we also showed that these two residues are deacetylated by SIRT4. Importantly, we showed that AMPK-controlled SCFD1 phosphorylation strongly disrupts the capacity of SCFD1 to interact with KAT2B, thus ensuring that the SCFD1 acetylation level remains low. Finally, we demonstrated that SCFD1 acetylation inhibits autophagic flux, specifically by blocking STX17-SNAP29-VAMP8 SNARE complex formation. Thus, our study reveals a mechanism through which phosphorylation and acetylation modifications of SCFD1 mediate SNARE complex formation to regulate autophagosome maturation.ACLY: ATP citrate lyase; CREB: cAMP responsive element binding protein; EBSS: nutrient-deprivation medium; EP300: E1A binding protein p300; KAT5/TIP60: lysine acetyltransferase 5; HOPS: homotypic fusion and protein sorting; MS: mass spectroscopy; SCFD1: sec1 family domain containing 1; SM: Sec1/Munc18; SNARE: soluble N-ethylmaleimide-sensitive factor attachment protein receptor; UVRAG: UV radiation resistance associated

KEYWORDS:

• Autophagosome
• autophagy
• lysosome
• SCFD1
• SNARE

Introduction

Autophagy is a central cellular process that overcomes cellular unfavorable conditions occurring when cells experience stresses such as nutrient deprivation, protein misfolding, and organelle damage [1–3].

Upon autophagy induction, a double membrane transient structure, a phagophore, can sequester substrates such as damaged mitochondria, protein aggregates, and pathogens; the phagophore matures into an autophagosome, and the autophagosomes subsequently fuse with lysosomes to enable degradation of these cellular substrates [4]. After degradation, the breakdown products are released from autolysosomes and can be used to produce cellular components or energy supplements [5].

It is known that autophagosome-lysosome fusion is mediated by the soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex, which comprises the proteins STX17, SNAP29, and VAMP8 [6]. STX17 recruits and forms a binary complex with the cytosolic protein SNAP29 on the autophagosome, and this complex then forms a tertiary SNARE complex with the lysosome SNARE protein VAMP8. This STX17-SNAP29-VAMP8 complex pulls the autophagosome and lysosome together to achieve autophagic membrane fusion [7,8]. SCFD1 is a Sec1/Munc18(SM) like protein. Neuronal SM proteins are well studied; for example, neuro STXBP1/Munc18-1 stabilizes SNARE assembly [9], and deleting Stxbp1 in mice inhibits secretion of the neurotransmitter γ-aminobutyric acid [10]. The yeast ortholog of SCFD1 Sly1 binds to the yeast SNARE protein Sed5, which regulates ER-Golgi trafficking [11]. In mammalian cells, SCFD1 interacts with STX17 and VAMP8. Overexpressing SCFD1 promotes STX17-SNAP29-VAMP8 complex assembly, and deletion of SCFD1 in U2OS cells was shown to inhibit autophagosome maturation [12].

Acetylation and phosphorylation are posttranslational modifications that areessential for the regulation of autophagy [13–15]. For example, the acetylation of ATG5, ATG7, and LC3 has been shown to inhibit autophagy [15], and the phosphorylation of ULK1, ATG9, BECN1 (beclin 1), and PIK3C3/VPS34 regulates autophagy initiation [16,17]. Phosphorylation of UVRAG (UV radiation resistance associated) enhances its interaction with RUBCN/RUBICON, thereby inhibiting autophagosome maturation [18].

Some autophagy genes are regulated by both acetylation and phosphorylation. For example, EP300/p300-mediated acetylation of PIK3C3/VPS34 inhibits its kinase activity and therefore represses autophagy initiation [19]. KAT2B/PCAF has been identified as a lysine acetyltransferase enzyme that acetylates H3 histones and other proteins including various transcription factors and cytoskeletal components [20–22]. For example, KAT2B-mediated acetylation of PGK1 at residue K323 promotes its enzymatic activity and also increases cancer cell proliferation [23]. KAT2B/PCAF has also been shown to stabilize ACLY (ATP citrate lyase) under high glucose conditions to promote lipid biosynthesis [24]. However, how KAT2B/PCAF regulates autophagy remains unclear.

Here, we detected differential acetylation and phosphorylation of SCFD1 upon stimulation of autophagy with torin 1 or EBSS. Residues S303 and S316 of SCFD1 were phosphorylated upon autophagy activation, and phosphorylated SCFD1 promoted STX17-SNAP29-VAMP8 SNARE complex formation. SCFD1 was deacetylated at residues K126 and K515 by SIRT4 under nutrient deprivation conditions. When nutrients were sufficient, autophagy activity was low, and SCFD1 could be acetylated by KAT2B. Our data also show that phosphorylation blocks the SCFD1-KAT2B/PCAF interaction, which is consistent with the low acetylation level of SCFD1 when SCFD1is phosphorylated. Taken together, our results indicate that SCFD1 is phosphorylated during autophagy, which decreases the interaction between SCFD1 and the acetyltransferase enzyme KAT2B/PCAF; therefore, in the absence of acetylation activity SCFD1 promotes STX17-SNAP29-VAMP8 complex formation and autophagosome-lysosome fusion.

Results

Nutrient deprivation reduces the extent of SCFD1 acetylation

SCFD1 is known to promote autophagosome-lysosome fusion [12], and we examined whether SCFD1 harbors acetylation modifications by transfecting HEK293T cells with a FLAG-SCFD1 construct and measuring acetylation using an antibody against acetylated lysine. FLAG-SCFD1 was clearly acetylated under nutrient-rich conditions, but the extent of this acetylation was dramatically decreased under autophagy-stimulating conditions, namely treatment with an MTOR inhibitor (torin 1) and growth in nutrient-deprivation medium (EBSS) (Figure 1(a,b)). To investigate the pattern of SCFD1 acetylation upon autophagy induction, we treated HEK293T cells stably expressing FLAG-SCFD1 with torin 1 for different durations; this treatment reduced the acetylation level of SCFD1 in a time-dependent manner (Figure 1(c,d)).

Figure 1. SCFD1 is deacetylated during autophagy. (a) The acetylation level of SCFD1 decreases upon torin 1- and EBSS-mediated autophagy induction in HEK293T cells. HEK293T cells stably expressing FLAG-SCFD1 were stimulated with EBSS for 1 h or with 100 nM torin 1 for 3 h before cell lysates were immunoprecipitated with FLAG beads, followed by immunoblotting with the indicated antibodies. The FLAG-SCFD1 acetylation level was determined by immunoblotting with an anti-acetyl-lysine antibody. (b) Quantification of the bands in Figure 1(a) by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). ****p < 0.0001. (c) Treatment with the MTOR inhibitor torin 1 leads to reduced SCFD1 acetylation in a time-dependent manner. HEK293T cells stably expressing FLAG-SCFD1 were stimulated with 100 nM torin 1 for 0, 1, 2, 3, or 4 h. The cell lysates were immunoprecipitated with FLAG beads. Immunoblotting was performed to examine the levels of LC3, SQSTM1, TUBB, FLAG-SCFD1, and FLAG-SCFD1 acetylation, using antibodies against LC3, SQSTM1, TUBB, FLAG, and acetyl-lysine, respectively. (d) Quantification of the bands in Figure 1(c) by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). *** p < 0.001, ****p < 0.0001. (e) Serum deprivation decreases the acetylation level of SCFD1. HEK293T cells stably expressing FLAG-SCFD1 were cultured in full medium, cultured in EBSS medium for 1, 2, or 3 h, or cultured in full medium for 2 h after 1 h of EBSS starvation. The SQSTM1 and LC3 levels were analyzed by immunoblotting. The acetylation level of FLAG-SCFD1 was assessed using an antibody against acetyl-lysine. (F) Quantification of the bands in Figure 1E by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). ****p < 0.0001. (G) TSA/NAM treatment increases SCFD1 acetylation. HEK293T cells stably expressing FLAG-SCFD1 were treated with 500 nM TSA together with 5 mM NAM for 0, 8, 12, or 16 h. Cell lysates were analyzed by immunoblotting with the indicated antibodies. (h) Identification of SCFD1 K126 acetylation by mass spectrometry. Pellets were prepared from an HEK293T cell line stably overexpressing FLAG-SCFD1; FLAG-SCFD1 was purified using FLAG beads, followed by mass spectrometry analysis. (I) Identification of SCFD1 K330 acetylation by mass spectrometry. (j) Identification of SCFD1 K515 acetylation by mass spectrometry. (k) Identification of the essential acetylation site of SCFD1. HEK293T cells were transfected with FLAG-WT SCFD1, FLAG-SCFD1^K126R, FLAG-SCFD1^K330R, FLAG-SCFD1^K515R, or FLAG-SCFD1[3KR]. Cell lysates were prepared for immunoprecipitation using anti-FLAG beads, followed by immunoblotting with the indicated antibodies. (l) Quantification of the bands in Figure 1(k) by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). *** p < 0.001, ****p < 0.0001.

We also cultured cells in EBSS medium for different durations and found that starvation conditions reduced the SCFD1 acetylation level in a time-dependent manner. The reduction in SCFD1 acetylation was diminished by the addition of fetal bovine serum (FBS) to the medium (Figure 1(e,f)). We also co-treated cells with two broad spectrum deacetylase inhibitors (NAM, a SIRT family deacetylase inhibitor and TSA, an HDAC family deacetylase inhibitor). The acylation level of SCFD1 was dramatically increased upon TSA-NAM treatment (Figure 1(g)). This result narrows the relevant deacetylase enzyme down to a member of the SIRT or HDAC family.

To identify the acylation sites of SCFD1, we performed a mass spectroscopy (MS) analysis of purified SCFD1. From this analysis, three putative acetylation sites, K126, K330, and K515, were identified (Figure 1(h-j)). To validate these acetylation sites, we generated constructs expressing FLAG fusions with four potential acetylation-deficientproteins by replacing the lysines with arginines (separately or all three residues together). We transfected HEK293T cells with WT FLAG-SCFD1 and the mutant variants. The SCFD1^K126R and SCFD1^K515R variants showed dramatically less SCFD1 acylation compared with WT SCFD1 and the SCFD1^K330R variant (Figure 1(k.l)). Taken together, we conclude that SCFD1 has acetylation modifications and that the main acetylated residues are K126 and K515.

SCFD1 is acetylated by KAT2B/PCAF

To identify the acetyltransferase modifying SCFD1 at residues K126 and K515, we transfected four HA-tagged acetyltransferases, CREB (cAMP responsive element binding protein), EP300 (E1A binding protein p300), KAT2B, and KAT5/TIP60 (lysine acetyltransferase 5) together with FLAG-SCFD1 into HEK293T cells. We found overexpression of KAT2B but no other tested acetyltransferases increased the acetylation level of FLAG-SCFD1 (Figure 2(a,b)). Knocking down KAT2B/PCAF but not KAT5 decreased the acetylation level of FLAG-SCFD1 (Figure 2(c,d)). Autophagy can be induced by culturing cells in serum starvation medium (EBSS) or by adding the MTOR inhibitor torin 1 into full medium. We found that the interaction of SCFD1 and KAT2B/PCAF was reduced when autophagy was simulated by torin 1 or EBSS treatment (Figure 2(e,f)).

Figure 2. SCFD1 is acetylated by KAT2B/PCAF. (a) KAT2B/PCAF overexpression increases the extent of SCFD1 acetylation. HEK293T cells that stably express FLAG-SCFD1 were transfected with the empty vector, CPB, EP300, KAT2B/PCAF, or KAT5/TIP60, and cell lysates were prepared for immunoprecipitation using anti-FLAG beads, followed by immunoblotting with the indicated antibodies. (b) Quantification of the bands in Figure 2(a) by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). ****p < 0.0001. (c) The acetylation level of FLAG-SCFD1 decreases upon knockdown of KAT2B/PCAF. HEK293T cells were transfected with an siRNA for knockdown endogenous KAT2B/PCAF (siKAT2B/PCAF), an siRNA for knockdown of endogenous KAT5/TIP60 (siKAT5), or a scramble control siRNA (siNC). These cells were transfected with a FLAG-SCFD1^WT plasmid. Cell lysates were prepared for immunoprecipitation using anti-FLAG beads, followed by immunoblotting with the indicated antibodies. (d) Quantification of the bands in Figure 2(c) by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). ****p < 0.0001. (e) No SCFD1-KAT2B/PCAF interaction is detected upon torin 1 or EBSS treatment. HEK293T cells were transfected with FLAG-SCFD1 and HA-KAT2B/PCAF. At 24 h after transfection, these cells were stimulated with EBSS for 1 h or 100 nM torin 1 for 3 h before cell lysates were immunoprecipitated with a FLAG antibody, followed by immunoblotting with the indicated antibodies. (f) Quantification of the bands in Figure 2E by ImageJ. The levels of HA-KAT2B/PCAF were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). ****p < 0.0001. (g) Chemical inhibition of KAT2B/PCAF activity decreases SCFD1 acetylation. HEK293T cells that stably express FLAG-SCFD1 were treated with 100 μm of the KAT2B/PCAF-specific inhibitor Garcinol or 10 μm of the EP300-specific inhibitor C646 for 4 h. Cell lysates were immunoprecipitated with FLAG beads, followed by immunoblotting with the indicated antibodies. (h) KAT2B/PCAF acetylates SCFD1. Recombinant His-SCFD1 purified from E. coli cells was used for in vitro acetylation assays with recombinant GST-HAT. (i) KAT2B/PCAF acetylation-deficient mutants cannot be acetylated. Acetylation of recombinant His-WT SCFD1 or its acetylation deficient variants (His-SCFD1[2KR] and His-SCFD1[3KR]) was assessed by performing in vitro assays with GST-HAT for the indicated times. SCDF1 acetylation was analyzed by immunoblotting with an anti-acetyl-lysine antibody.

Moreover, treatment of HEK293T cells overexpressing FLAG-SCFD1 with the KAT2B/PCAF-specific inhibitor Garcinol, but not the EP300-specific inhibitor C646, dramatically reduced SCFD1 acetylation, indicating that KAT2B/PCAF may be involved in SCFD1 acetylation (Figure 2(g)). To confirm whether SCFD1 is directly acetylated by KAT2B/PCAF, we purified recombinant KAT2B/PCAF-HAT from Escherichia coli and performed an in vitro acetylation assay. SCFD1 was acetylated by KAT2B/PCAF in vitro (Figure 2(h)). Furthermore, the in vitro acetylation assay also revealed that SCFD1 acetylation by KAT2B/PCAF was diminished in cells expressing SCFD1[2KR] (double mutant) and SCFD1[3KR] (triple mutant) (Figure 2(i)). Therefore, these data indicate that KAT2B/PCAF acetylates SCFD1 at K126 and K515.

SCFD1 is deacetylated by SIRT4

To identify which deacetylase regulates SCFD1 acetylation on K126 and K515, we treated cells with the SIRT family deacetylase inhibitor NAM and the HDAC family deacetylase inhibitor TSA, separately or together. The SIRT inhibitor NAM had a greater effect on SCFD1 acetylation than the HDAC inhibitor TSA (Figure 3(a)). This result suggests that SIRT family members are the potential deacetylases of SCFD1. To test this hypothesis, we co-transfected HEK293T cells with FLAG-SCFD1 and MYC-SIRT1 to SIRT7, finding that only SIRT4 overexpression could reduce SCFD1 acetylation levels (Figure 3(b,c)).

Figure 3. SIRT4 deacetylates SCFD1. (a) Treatment with the SIRT inhibitor NAM increases the extent of SCFD1 acetylation. HEK293T cells transfected with FLAG-SCFD1 were treated with 5 mM of the SIRT inhibitor NAM, 500 nM of the HDAC inhibitor TSA, or 5 mM NAM and 500 nM TSA together for 12 h. Cell lysates were immunoprecipitated with FLAG beads, followed by immunoblotting with the indicated antibodies. The acetylation level of FLAG-SCFD1 was assessed using an antibody against acetyl-lysine. (b) SIRT4 deacetylates SCFD1. HEK293T cells that stably express FLAG-SCFD1 were transfected with the MYC-SIRT1, MYC-SIRT2, MYC-SIRT3, MYC-SIRT4, MYC-SIRT5, MYC-SIRT6, or MYC-SIRT7 plasmid. At 24 h after transfection, cell lysates were prepared for immunoprecipitation with FLAG antibody. The acetylation level of FLAG-SCFD1 was assessed using an antibody against acetyl-lysine. (c) Quantification of the bands in Figure 3(b) by ImageJ. The levels of ACE-lys were quantified by ImageJ and normalized with FLAG-SCFD1. Data are shown as mean ± SEM (n = 3). ****p < 0.0001. (d) FLAG-SCFD1 physically interacts with SIRT4. HEK293T cells that stably express FLAG-SCFD1 were transfected with the MYC-SIRT1, MYC-SIRT2, MYC-SIRT3, MYC-SIRT4, MYC-SIRT5, MYC-SIRT6, or MYC-SIRT7 plasmid, followed by anti-FLAG immunoprecipitation and immunoblotting for the indicated proteins. (e) SCFD1 interacts with SIRT4 in vitro. In vitro GST-affinity-isolation assay using recombinant His-SCFD1 and GST-SIRT4. (f) A SIRT4-deacetylase-dead mutant cannot deacetylate SCFD1. HEK293T cells stably expressing FLAG-SCFD1 were transfected with an siRNA for knockdown (KD) of endogenous SIRT4. To restore SIRT4 expression, these KD cells were transfected with an siRNA-resistant plasmid for expression of WT SIRT4 or a SIRT4-deacetylase-dead mutant variant. Cell lysates were immunoprecipitated with a FLAG antibody. The acetylation level of FLAG-SCFD1 was assessed using an antibody against acetyl-lysine. (g) SCFD1 is acetylated at K126 and K515. HEK293T cells transfected with FLAG-WT SCFD1 or FLAG-SCFD1[2KR] were co-transfected with an siRNA for KD of endogenous SIRT4, followed by anti-FLAG immunoprecipitation and immunoblotting for the indicated proteins.

Next, we examined the potential interaction between SCFD1 and SIRT4. We over-expressed FLAG-SCFD1 and MYC-SIRT1 to SIRT7 in HEK293T cells. A co-immunoprecipitation assay showed that MYC-SIRT4 and FLAG-SCFD1 readily interacted (Figure 3(d)). We next purified recombinant SIRT4 and SCFD1 from E. coli and performed an in vitro affinity-isolation assay, which confirmed this interaction (Figure 3(e)). Furthermore, when we knocked-down SIRT4 using a specific siRNA, the SCFD1 acetylation level in these cells was higher compared with that in cells overexpressing wild-type SIRT4 (Figure 3(f)). Consistent with these findings, there was no further increase in the acetylation of SCFD1[2KR] (acylation-deficient mutant) when SIRT4 was knocked down compared with WT SCFD1 (Figure 3(g)).

SCFD1 acetylation impairs autophagy flux

SCFD1 is a positive regulator of autophagosome-lysosome fusion. Given the observation that the SCFD1 acetylation level changed when autophagy was induced, we hypothesized that SCFD1 acetylation may regulate autophagy. To test this hypothesis, we expressed WT SCFD1, SCFD1[2KR] (de-acetylation mimic), and SCFD1[2KQ] (acetylation mimic) in stable cell lines. The cells were cultured in rich medium in the absence or presence of the MTOR inhibitor torin 1 followed by western blotting to analyze autophagic flux. The levels of LC3 and the autophagy substrate SQSTM1/p62 were dramatically higher in SCFD1 knockdown cells, and these levels could be efficiently rescued in cells expressing SCFD1[2KR], but not SCFD1[2KQ] (Figure 4(a)).

Figure 4. SCFD1 acetylation inhibits autophagy flux. (a) SCFD1 deacetylation promotes autophagic flux. U2OS cells were transfected with an siRNA for knockdown endogenous SCFD1. To restore SCFD1 expression, these cells were transfected with an siRNA-resistant WT SCFD1, SCFD1[2KR] (deacetylation mimic variant), or SCFD1[2KQ] (acetylation mimic variant) plasmid. The cells were treated with torin 1 for 3 h or left untreated. Cell lysates were analyzed by immunoblotting with the indicated antibodies. (b) SCFD1 acetylation inhibits GFP-LC3 cleavage. GFP-LC3-overexpressing U2OS cells were transfected with the FLAG-WT SCFD1, FLAG-SCFD1[2KR] (deacetylation mimic variant), or FLAG-SCFD1[2KQ] (acetylation mimic variant) variant and the cells were treated with 100 nM torin 1 or left untreated, followed by immunoblotting. (c) Quantification of the bands in Figure 4(b) by ImageJ. The levels of free GFP were quantified by ImageJ and normalized with tubulin. Data are shown as mean ± SEM. ****p < 0.0001. (d) SCFD1 deacetylation promotes autophagosome maturation. U2OS cells that stably express mRFP-GFP-LC3 were transfected with FLAG-WT SCFD1, FLAG-SCFD1[2KR], or FLAG-SCFD1[2KQ], and then treated with 100 nM torin 1for 3 h. After treatment, cells were fixed with 4% PFA followed by fluorescence microscopy. (e) Quantification of the proportion of mRFP⁺-GFP^–LC3 puncta among the total complement of autophagic vesicles per cell. At least 30 cells were examined for each group. Data are shown as means ± SEM. Statistical significance was evaluated by one-way ANOVA; ****p < 0.0001. (f) SCFD1 deacetylation promotes LC3 and LAMP2 colocalization. U2OS cells transfected with an siRNA for knockdown of endogenous SCFD1. To restore SCFD1 expression, these cells were transfected with siRNA-resistant FLAG-WT SCFD1, FLAG-SCFD1[2KR], or FLAG-SCFD1[2KQ]. The cells were stimulated with 100 nM torin 1 for 3 h, followed by fluorescence microscopy analysis to observe the co-localization of LC3 overlap with LAMP2.

We also analyzed autophagic flux in these cells by performing a GFP-LC3 cleavage assay. In GFP-LC3 expressing cells with autophagy induction, GFP-LC3 was delivered to lysosomes for degradation. Because GFP is more resistant to degradation than LC3 in lysosomes, we observed the accumulation of free GFP by western blotting [25] under normal autophagy flux conditions. Comparable with the autophagy flux assay results, GFP turnover was dramatically inhibited in SCFD1[2KQ] cells. This phenotype was neutralized by SCFD1[2KR] expression (Figure 4(b,c)). This result indicates SCFD1[2KQ] (acetylation mimic) may inhibit autophagy flux.

Next, we employed the mRFP-GFP-LC3 assay to evaluate autophagosome-lysosome fusion activity. When autophagosomes fuse with lysosomes, the GFP signal is quenched and a red signal is observed (GFP^–mRFP⁺-LC3). If the autophagosome is not fused with a lysosome, yellow dots are observed (GFP⁺-mRFP⁺-LC3). The results of these assays showed that the extent of autophagosome-lysosome fusion was higher in SCFD1[2KR]-expressing cells compared with that in WT SCFD1 cells (Figure 4(d,e)). Consistent with this, the colocalization of endogenous LC3 (autophagosome indicator) and LAMP2 (lysosome indicator) was higher in cells expressing the SCFD1[2KR] mutant than in those expressing WT SCFD1(Figure 4(f)). We conclude that SCFD1 acetylation impairs autophagosome-lysosome fusion.

SCFD1 deacetylation promotes SNARE complex formation

Given that SCFD1 interacts with the STX17-SNAP29-VAMP8 SNARE complex, SCFD1 acetylation may inhibit autophagosome maturation. We hypothesized that acetylation of sites K126 and K515 of SCFD1 may regulate SNARE complex formation. To test this hypothesis, we treated HEK293T cells co-expressing GFP-SCFD1 and FLAG-VAMP8 with NAM (SIRT family deacetylase inhibitor). The STX17-VAMP8 interaction decreased when SCFD1 was acetylated (Figure 5(a)). The interaction between STX17-VAMP8 was additionally assessed by overexpressing the deacetylation mimic mutant SCFD1[2KR] (Figure 5(b)). Furthermore, we observed that overexpression of the deacetylated SCFD1[2KR/3KR] mutants increased the interaction of SCFD1 with STX17 and VAMP8 in vitro (Figure 5(c,d)).

Figure 5. SCFD1 acetylation inhibits SNARE complex formation. (a) The extent of the STX17-VAMP8 interaction is reduced upon inhibition of SIRT4-mediated SCFD1 deacetylation. HEK293T cells were transfected with GFP-SCFD1 and FLAG-VAMP8, and these cells were treated with 5 mM of the SIRT inhibitor NAM for 12 h (or left untreated). Cell lysates were immunoprecipitated with FLAG beads, followed by immunoblotting with the indicated antibodies. (b) SCFD1 acetylation reduces the extent of the STX17-VAMP8 interaction. HEK293T cells overexpressing Flag-VAMP8 and HA-STX17 were transfected with plasmids for the expression of GFP-SCFD1, GFP-SCFD1[2KR], or GFP-SCFD1[2KQ]. Cell lysates were immunoprecipitated with FLAG beads, followed by immunoblotting with the indicated antibodies. (c) Recombinant GST-ZZ-FLAG-STX17 was bound to GST beads, and then co-incubated with recombinant His-WT SCFD1, His-SCFD1[2KR], or His-SCFD1[2KQ]. After TEV cleavage, the interaction was analyzed by immunoblotting with FLAG or His antibodies. (d) Recombinant GST-ZZ-FLAG-VAMP8 was co-incubated with recombinant His-WT SCFD1, His-SCFD1[2KR], or His-SCFD1[2KQ]. Immunoblotting was carried out to monitor the interaction. (e) Colocalization of FLAG-STX17 and LAMP2 in cells expressing WT SCFD1, SCFD1[2KR], or SCFD1[2KQ]. U2OS cells expressing WT SCFD1, SCFD1[2KR], or SCFD1[2KQ]. were transfected with the FLAG-STX17 plasmid; cells were fixed with 4% PFA, stained with antibodies against FLAG or LAMP2, and imaged by confocal fluorescence microscopy. (f) Quantitative analysis of the FLAG-STX17 overlap with LAMP2 (%). At least 20 cells were examined for each group. Data are shown as mean ± SEM. ****p < 0.0001. (g) Confocal microscopy analysis of FLAG-VAMP8 and LC3 in cells expressing WT SCFD1, SCFD1[2KR], or SCFD1[2KQ]. U2OS cells expressing WT SCFD1, SCFD1[2KR], or SCFD1[2KQ] were transfected with the FLAG-VAMP8 plasmid; cells were fixed with 4% PFA, stained with antibodies against FLAG or LC3, and imaged by confocal fluorescence microscopy. (h) Quantitative analysis of the extent of the FLAG-VAMP8 overlap with LC3 (%). At least 20 cells were counted. Data are shown as means ± SEM. ****p < 0.0001, according to one-way ANOVA.

To determine if SCFD1 acetylation regulates colocalization of STX17 and LAMP2, we transfected U2OS cells with GFP-SCFD1, GFP-SCFD1[2KR], and GFP-SCFD1[2KQ]; colocalization of FLAG-STX17 and LAMP2 was analyzed in these cells upon torin 1 treatment. The intensity of the signal for colocalization of FLAG-STX17 with LAMP2 was stronger for cells expressing GFP-SCFD1[2KR] than for cells expressing GFP-SCFD1 (Figure 5(e,f)). Similarly, we observed that colocalization of FLAG-VAMP8 and endogenous LC3 was lower in GFP-SCFD1[2KQ] cells and could be rescued by overexpression of GFP-SCFD1[2KR] (Figure 5(g,h)). In summary, these data demonstrate that upon stimulation, SCFD1 is deacetylated by SIRT4, and that deacetylated SCFD1 promotes SNARE complex formation, thereby increasing autophagosome-lysosome fusion.

SCFD1 phosphorylation promotes its deacetylation

We observed that the phosphorylation level of FLAG-SCFD1 was increased by torin 1 or by starvation medium (EBSS) treatment (Figure 6(a)). Next, we conducted a MS analysis, which identified S303 and S316 as putative SCFD1 phosphorylation sites (Figure 6(b,c)). We replaced the serine (S) 303 and S316 residues with alanine (A) to mimic dephosphorylated SCFD1. The phosphorylation level of the SCFD1^S303A and SCFD1^S316A variants was lower compared with that of WT SCFD1. When we mutated S303 and S316 together (SCFD1[2A]), no phosphorylation was observed. These results suggest both sites (S303 and S316) are essential for SCFD1 phosphorylation (Figure 6(d)).

Figure 6. SCFD1 phosphorylation promotes its deacetylation. (a) The phosphorylation level of FLAG-SCFD1 increases upon autophagy induction. HEK293T cells that stably express FLAG-SCFD1 were stimulated with EBSS for 1 h or 100 nM torin 1 for 3 h before cell lysates were immunoprecipitated with FLAG beads, followed by immunoblotting with the indicated antibodies. The FLAG-SCFD1 phosphorylation level was detected by immunoblotting with an anti-S/T phosphorylation antibody. (b) Identification of SCFD1 S303 phosphorylation by mass spectrometry. (c) Identification of SCFD1 S316 phosphorylation by mass spectrometry. (d) The S303 and S316 residues are essential for SCFD1 phosphorylation. HEK293T cells were transfected with the FLAG-WT SCFD1, FLAG-SCFD1^S303A, FLAG-SCFD1 ^S316A, or FLAG-SCFD1[2A] plasmid. Cell lysates were prepared for immunoprecipitation with FLAG beads. The FLAG-SCFD1 phosphorylation level was monitored with immunoblotting using the indicated antibodies. (e) HEK293T cells were transfected HA-AMPK, MYC-ULK1, or MYC-MTOR together with FLAG-SCFD1. After 24 h, cell lysates were immunoprecipitated with FLAG-beads. FLAG or MYC tagged proteins were then analyzed by immunoblotting. (f) HEK293T cells were transfected with FLAG-AMPK, and cell lysates were prepared for immunoprecipitation using anti-SCFD1 antibody, followed by immunoblotting. (G) Recombinant His-WT SCFD1 and His-SCFD1[2A] proteins were purified from E. coli. Recombinant AMPK was purchased form Genescript. ³²P-autoradiograms were obtained to determine the extent of SCFD1 phosphorylation. Coomassie staining was used to normalize the SCFD1 protein level. (h) SCFD1 phosphorylation inhibits its acetylation. HEK293T cells were transfected with the FLAG-WT SCFD1, FLAG-SCFD1^S303A, FLAG-SCFD1^S316A, FLAG-SCFD1[2A], FLAG-SCFD1^S303D, FLAG-SCFD1^S316D, or FLAG-SCFD1[2D] plasmid. Cell lysates were prepared for immunoprecipitation with FLAG beads, followed by immunoblotting analyzed with the indicated antibodies. (i) SCFD1 dephosphorylation promotes its acetylation. HEK293T cells expressing HA-KAT2B/PCAF were transfected with the FLAG-WT SCFD1, FLAG-SCFD1[2A] or FLAG-SCFD1[2D] plasmid. After transfection, these cells were stimulated with EBSS for 1 hor left untreated. Cell lysates were prepared for immunoprecipitation with FLAG beads, followed by immunoblotting analysis with the indicated antibodies. (j) SCFD1 phosphorylation promotes VAMP8 and STX17 interaction. HEK293T cells expressing FLAG-VAMP8 were transfected with the GFP-SCFD1, SCFD1[2A], or SCFD1[2D] plasmid. Cell lysates were prepared for immunoprecipitation with FLAG beads, followed by immunoblotting with the indicated antibodies. (k) Recombinant GST-ZZ-FLAG-VAMP8 was bound to GST beads, and then co-incubated with recombinant His-WT SCFD1, His-SCFD1^S303A, His-SCFD1^S316A, His-SCFD1[2A], His-SCFD1^S303D, His-SCFD1^S316D, or His-SCFD1[2D]. After TEV cleavage, the interaction was analyzed by immunoblotting with FLAG or His antibodies. (l) Recombinant GST-ZZ-FLAG-STX17 was bound to GST beads, and then co-incubated with recombinant His-WT SCFD1, His-SCFD1^S303A, His-SCFD1^S316A, His-SCFD1[2A], His-SCFD1^303D, His-SCFD1^S316D, or His-SCFD1[2D]. The interaction was analyzed by immunoblotting with FLAG or His antibodies.

Next, we performed experiments to identify the kinase that phosphorylates SCFD1. We observed an interaction between FLAG-SCFD1 and HA-AMPK (Figure 6(e)). Overexpression of FLAG-AMPK in HEK293T cells increased the SCFD1 phosphorylation level (Figure 6(f)). We also generated recombinant proteins and performed in vitro kinase assays. WT SCFD1 but not its dephosphorylation mimic mutant SCFD1[2A] could be phosphorylated by AMPK in vitro (Figure 6(g)). Taken together, these results suggest that AMPK is the kinase that phosphorylates SCFD1 at residues S303 and S316.

We also observed that the acetylation level of SCFD1 was lower in HEK293T cells expressing the phosphorylation mimic mutants (SCFD1^S303D, SCFD1^S316D, and SCFD1[2D]). In contrast, a dramatically higher acetylation level was observed in cells expressing the phospho-dead variants of SCFD1 (SCFD1^S303A, SCFD1^S316A, and SCFD1[2A]) (Figure 6(h)).

HA-KAT2B/PCAF was co-precipitated with SCFD1 in cell extracts from HEK293T cells overexpressing both proteins. The interaction of the SCFD1 phosphorylation mimic mutant SCFD1[2D] with HA-KAT2B/PCAF was dramatically lower compared with that of WT SCFD1 (Figure 6(i)). Because acetylation of SCFD1 impairs SNARE complex formation and the phosphorylation-dead mimic SCFD1 mutant (SCFD1[2A]) has higher SCFD1 acetylation levels, we hypothesized that SCFD1 phosphorylation affects SNARE complex formation. To test this, we co-expressed FLAG-VAMP8 with the WT SCFD1, SCFD1[2A], and SCFD1[2D] variants in HEK293T cells; the interaction of FLAG-VAMP8 and endogenous STX17 was analyzed by co-immunoprecipitation. Expression of the SCFD1 phosphorylation mimic mutant increased FLAG-VAMP8 and STX17 interaction (Figure 6(j)).

In addition, we purified recombinant WT SCFD1 and the phosphorylation and phosphorylation-dead mimic mutants form E. coli and performed in vitro binding assays. The phosphorylation mimic forms of SCFD1 (SCFD1^S303D, SCFD1^S316D, SCFD1[2D]) had increased interactions with STX17 and VAMP8 in vitro (Figure 6(k,l)). Taken together, these results further indicate that SCFD1 phosphorylation decreases the interaction between SCFD1 and KAT2B/PCAF, thus decreasing SCFD1 acetylation.

SCFD1 phosphorylation increases autophagosome maturation

Next, we explored the function of SCFD1 phosphorylation in autophagy. The LC3 and SQSTM1 protein levels were lower in cells overexpressing SCFD1[2D] than in those overexpressing WT SCFD1, indicating that the autophagy flux was higher (Figure 7(a)). Consistent with this, there was also more cleavage of GFP in the SCFD1[2D]⁻overexpressing cells (Figure 7(b)). We also examined the autophagy activity by performing an mRFP-GFP-LC3 assay, and the data confirmed that SCFD1[2D] increased autophagy flux, as indicated by the increased number red dots (Figure 7(c)). Consistent with this, the number of autophagy substrate SQSTM1 puncta was dramatically increased in U2OS cells expressing SCFD1[2A] (Figure 7(d)). Thus, we can conclude that SCFD1 phosphorylation increased autophagosome maturation.

Figure 7. SCFD1 phosphorylation promotes autophagosome maturation. (a) SCFD1 phosphorylation promotes autophagic flux. U2OS cells were transfected with an siRNA for knockdown of endogenous SCFD1. To restore SCFD1 expression, these cells were transfected with an siRNA-resistant WT SCFD1, SCFD1[2A] (dephosphorylation mimic variant), or SCFD1[2D] (phosphorylation mimic variant) plasmid. The cells were treated with torin 1 for 3 h or left untreated. Cell lysates were analyzed by immunoblotting with the indicated antibodies. (b) SCFD1 phosphorylation promotes GFP-LC3 cleavage. GFP-LC3-overexpressing U2OS cells were transfected with the FLAG-WT SCFD1, FLAG-SCFD1[2A] (dephosphorylation mimic variant), or FLAG-SCFD1[2D] (phosphorylation mimic variant) variant, and the cells were treated with 100 nM torin 1 for 3 h (or left untreated), followed by immunoblotting with the indicated antibodies. (c) An mRFP-GFP-LC3 assay was performed to monitor the autophagosome maturation levels in WT SCFD1-, SCFD1[2A]-, or SCFD1[2D]-expressing U2OS cells. U2OS cells that stably expressed mRFP-GFP-LC3 were transfected with the FLAG-WT SCFD1, FLAG-SCFD1[2A], or FLAG-SCFD1[2D] plasmid, and then treated with 100 nM torin 1 for 3 h. After treatment, cells were fixed with 4% PFA followed by fluorescence microscopy. (d) Quantification of the SQSTM1 puncta in WT SCFD1-, SCFD1[2A]-, or SCFD1[2D]-expressing U2OS cells. At least 30 cells were quantified. Data are shown as means ± SEM. ****p < 0.0001; *** < 0.001, according to one-way ANOVA.

Discussion

Autophagy is a cellular process involving multiple steps including autophagosome formation, autophagosome maturation, and autophagosome-lysosome fusion [1]. SCFD1 regulates STX17-SNAP29-VAMP8 complex formation and promotes autophagosome lysosome fusion [12]. In the present study, we found that acetylation and phosphorylation modifications of SCFD1 are required for its autophagy-related functions, and we found an interdependence of these SCFD1 modifications: phosphorylated SCFD1 has less acetylation and we showed that phosphorylation of SCFD1 increases the extent of STX17-SNAP29-VAMP8 complex formation, which in turn enhances autophagosome maturation.

The SCFD1 protein resembles Sec1/STXBP1/Munc18 (SM), and previous studies have shown that SM proteins are required for SNARE-mediated membrane fusion [18,19]. Multiple studies have detected interactions between STXBP1 and the SNARE protein STX1 or SNARE complex [20–23]. It has been shown that STXBP1 assists in the assembly of the SNARE complex. SM proteins contain three domains that fold into a conserved arch-shaped architecture [26–32]: Domain 1 consists of five β-strands surrounded by five α-helices and is located at one head of the arch structure. Domain 3 is divided into domains 3a and 3b, and domain 3a constitutes the other head of the arch. The neuronal Munc18-1 interacts with STX1 through the two head domains, domain 1 and domain 3a [31].

The homotypic fusion and protein sorting (HOPS) complex is essential for autophagosome-lysosome fusion [33]. This complex consists of VPS11, VPS16, VPS18, VPS33, VPS39, and VPS41 [34,35]. Like SCFD1, VPS33 is a Sec1/Munc18-like protein, maybe VPS33 and SCFD1 work together to promote SNARE complex formation and autophagosome-lysosome fusion.

Our immunoblotting and immunostaining data showed that KAT2B/PCAF-mediated acetylation of SCFD1 restricts the formation of the STX17-SNAP29-VAMP8 SNARE complex, and that this in turn limits autophagosome-lysosome fusion (Figures 4 and 5). Moreover, we provide evidence that autophagosome maturation is expedited by the SIRT4-mediated deacetylation of SCFD1 (Figures 4 and 5). The SCFD1 structure predicted by AlphaFold2 has a typical arch (U-shaped) architecture similar to that of SM proteins (Figure S1). The SCFD1 K126 acetylation site is located in the head region of domain 1 and may be involved in binding with the N-terminal peptide of STX1. However, the other acetylation site, K515, is located at the linker region connecting domain 3b and domain 2, which is far away from the reported SNARE binding sites (Figure S1). Sequence alignment showed that both acetylation sites are highly conserved (Figure S2), suggesting that the acetylation modification of SCFD1 may be a conserved mechanism for SNARE regulation among different species. Thus, more structural data are needed to help us better understand i) whether acylation modification changes the conformation of SCFD1 and ii) how this change affects SNARE binding and subsequent SNARE complex formation and autophagosome-lysosome fusion.

Acetylation has recently been shown to regulate autophagy [13,36,37]. Specifically, autophagy initiation is inhibited upon the EP300-mediated acetylation of known autophagy proteins including PIK3C3/VPS34, ATG5, and ATG12 [11,23]. It is also known that GSK3 (glycogen synthase kinase 3) activates KAT5/TIP60 via phosphorylation, and that active KAT5/TIP60 can acetylate ULK1 to activate autophagy [24]. Finally, KAT2B/GCN5 has been shown to inhibit autophagy by acetylating the master lysosomal biogenesis regulator TFEB [25]. Our study extends the understanding of the multi-faceted impacts of acetylation on autophagy by experimentally establishing that SCFD1 is phosphorylated at S303 and S316 under stressful conditions and demonstrating that SCFD1 phosphorylation results in reduced acetylation. Specifically, we showed that phosphorylated SCFD1 is preferentially deacetylated. Ultimately, deacetylated SCFD1 promotes STX17-SNAP29-VAMP8 SNARE complex formation and thereby promotes autophagosome-lysosome fusion.

Materials and methods

Antibodies

Rabbit polyclonal anti-STX17 antibody (Sigma-Aldrich, HPA001204), rabbit monoclonal anti-VAMP8 antibody (Abcam, ab76021), mouse monoclonal anti-human SQSTM1 antibody (Novus, H00008878-M01), mouse monoclonal anti-human LAMP2 antibody (Santa Cruz, sc-18,822), rabbit polyclonal anti-LC3B antibody (Sigma-Aldrich, L7543), rabbit polyclonal anti-FLAG antibody (Sigma-Aldrich, F7425), mouse monoclonal anti-TUBB/tubulin beta antibody (Sigma-Aldrich, T8328), rabbit polyclonal anti-SCFD1 antibody (Proteintech, 12,569-1-AP), rabbit polyclonal anti-acetylated-lysine antibody (Cell Signaling Technology, 9441), chicken polyclonal anti-GFP antibody (Abcam, ab13970), goat anti-mouse IgG (H + L), HRP (Proteintech, SA00001-1), goat anti-rabbit IgG (H + L), HRP (Proteintech, SA00001-2), goat anti-mouse IgG (H + L), Alexa Fluor 488 (Thermo Fisher Scientific, A-11029), goat anti-rabbit IgG (H + L), Alexa Fluor 488 (Thermo Fisher Scientific, A-11034), goat anti-mouse IgG (H + L), Alexa Fluor 647 (Thermo Fisher Scientific, A-21235), goat anti-rabbit IgG (H + L), Alexa Fluor 647 (Thermo Fisher Scientific, A-21245), goat anti-mouse IgG (H + L), Alexa Fluor 594 (Thermo Fisher Scientific, A32742), goat anti-rabbit IgG (H + L), Alexa Fluor 594 (Thermo Fisher Scientific, A-11012).

Chemicals and reagents

Torin 1 (Sigma-Aldrich, 475,991), IPTG (Promega, V3955), EASYpack Protease Inhibitor Cocktail (Sigma-Aldrich, 5,892,970,001), Lipofectamine 3000 (Thermo Fisher Scientific, L3000015), 3FLAG Peptide (Sigma-Aldrich, F4799), Trizol (Solarbio, M8018), ANTI-FLAG M2 Affinity Gel (Sigma-Aldrich, A2220), Earle’s basic salt solution (Thermo Fisher Scientific, 1,816,327), Glutathione Sepharose 4 Fast Flow (GE Healthcare, GE17-5132-01).

Plasmids, cell culture, and transfection

Construction of the FLAG-SCFD1, FLAG-VAMP8, and FLAG-STX17 plasmids was described previously [12]. HA-CBP (HA-CREB) and HA-P300 (HA-EP300) were gifted by Dr. Qiming Sun of Zhejiang University. HA-PCAF (HA-KAT2B) and HA-TIP60 (HA-KAT5) were gifted by Dr. Ting Liu of Zhejiang University. MYC-SIRT1-7 was gifted by Dr. Wei Liu of Zhejiang University. All point mutations were generated using a Takara PCR kit (R010A) and verified by sequencing. The following primers were used to generate point mutations: K126R, 5’-CTGCTATTTCAAGAAGTAGACTGGAAGATATTGCAAAT-3’ (forward) and 3’- ATTTGCAATATCTTCCAGTCTACTTCTTGAAATAGCAG-5’ (reverse); K126Q, 5’-CTGCTATTTCAAGAAGTCAACTGGAAGATATTGCAAAT-3’ (forward) and 3’-ATTTGCAATATCTTCCAGTTGACTTCTTGAAATAGCAG-5’ (reverse); K330R, 5’-ATTTTGGCAAAAACATAGAGGAAGTCCATTCCCAGAAG-3’ (forward) and 3’-CTTCTGGGAATGGACTTCCTCTATGTTTTTGCCAAAAT-5’ (reverse); K330Q, 5’-ATTTTGGCAAAAACATCAAGGAAGTCCATTCCCAGAAG-3’ (forward) and 3’-CTTCTGGGAATGGACTTCCTTGATGTTTTTGCCAAAAT-5’ (reverse); K515R, 5’-TGGCAGCACTACCACTAGACCAATGGGTCTTTTATCAC-3’ (forward) and 3’-GTGATAAAAGACCCATTGGTCTAGTGGTAGTGCTGCCA-5’(reverse); K515Q, 5’-TGGCAGCACTACCACTCAACCAATGGGTCTTTTATCAC-3’ (forward) and 3’-GTGATAAAAGACCCATTGGTTGAGTGGTAGTGCTGCCA-5’ (reverse).

HEK293T and U2OS cells (American Type Culture Collection; HTB-96 and CRL-11268) were grown in Dulbecco modified eagle medium (DMEM; Gibco, 41,966,052) containing 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (P/S). According to the manufacturer’s recommendations, cells were transfected with the appropriate plasmid for 24–48 h using Lipofectamine 3000. The following siRNAs were used: KAT2B/PCAF siRNA: GCAGATACCAAACAAGTTTA; KAT5/TIP60 siRNA: GAGAAAGAATCAACGGAAG;and SIRT4 siRNA: CCAGACTACAGGTCAGAAA.

Immunoprecipitation and western blotting

After transfecting HEK293T cells with the indicated plasmids, the cell pellets were collected and lysed for 30 min in TAP buffer (20 mM Tris-HCl, pH 7.4, 0.5% NP40 [Sigma, 1,175,499,001], 150 mM NaCl, 1 mM NaF, 1 mM Na₃VO₄, 1 mM EDTA). After centrifuging the lysate at 4000 x g for 3 min, the supernatant was incubated with the indicated beads for 6 h. The beads were washed three times after incubation and eluted with FLAG peptide. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels with a concentration of 6%–15%were used to separate proteins. The proteins were then transferred to 0.22-µm polyvinylidene fluoride (PVDF) membranes for the indicated times at 80 V, then blocked in skim milk in PBS (50 mM Tris-HCl, pH 7.4).

Mass spectrometry analysis

To identify the acetylation and phosphorylation sites of SCFD1, FLAG-SCFD1 was separated by SDS-PAGE. Gel bands containing FLAG-SCFD1 were excised and digested by trypsin at 37°C for 16 h. The digested peptides were loaded onto a nano C18 column (1.8 mm, 0.15 × 1.00 mm) with the Thermo Fisher Easy-nLC 1000 high performance liquid chromatography (HPLC) system as described previously [12]. In brief, solvent A contained 0.1% formic acid and solvent B contained 100% acetonitrile. The elution gradient was from 4% to 18% in 182 min, 18% to 90% solvent B in 13 min at a flow rate of 300 nL/min. Mass spectrometry analysis was carried out at the AIMSMASS Co., Ltd. (Shanghai, China) in the positive-ion mode with automated data-dependent MS/MS analysis with full scans (350–1600 m/z) acquired using FTMS at a mass resolution of 30,000, and the ten most intense precursor ions were selected for MS/MS.

Confocal fluorescence microscopy

U2OS cells were plated onto coverslips, rinsed with cold PBS, and fixed for 10 min in 4% paraformaldehyde. The fixed coverslips were then blocked with PBS containing 5% BSA (Sigma-Aldrich, A7030) for 1 h and subsequently incubated with primary antibody overnight at 4°C. The coverslips were rinsed three times with PBS and then incubated with the secondary antibody for 1 h. After three washes, cells were mounted (Prolong Gold, Invitrogen, P36930) on slides. Cells were observed with a laser scanning confocal microscope (Olympus FV3000).

Immunofluorescence was analyzed for U2OS cells expressing WT SCFD1, SCFD1[2KR], or SCFD1[2KQ], which were stained with antibodies specific for FLAG, LC3, or LAMP2.

For the comparisons in 4 F, 5E, and 5 G, one-way ANOVA was used to assess significance; ****p < 0.0001; Data are presented as means ± SEM. At least 20 cells were evaluated per group.

The data in Figures 4(f), 5(e), and 5(g) are the count data for the number of cells in which colocalization was detected. Specifically, the data in Figure 4(f) are based on counting of puncta per cell (at least 20 cells were examined for each group), as follows:

Number of puncta positive for both LC3 and LAMP2

Number of puncta positive for LAMP2

The data in Figure 5(e) are based on counting of puncta per cell (at least 20 cells were examined for each group), as follows:

Number of puncta positive for both FLAG-STX17 and LAMP2

Number of puncta positive for LAMP2

The data in Figure 5(g) are based on counting of puncta per cell (at least 20 cells were examined for each group), as follows:

Number of puncta positive for both FLAG-VAMP8 and LC3

Number of puncta positive for LC3

Protein purification

The His-tagged WT SCFD1 plasmid was transformed into E. coli BL21cells, which were grown in LB media and protein expression was induced by adding IPTG to a final concentration of 0.3 mM. After lysis and centrifugation, the supernatant was transferred to a fresh tube and incubated overnight with Ni-NTA beads (Thermo Fisher Scientific, R90115) at 4°C. The beads were then washed three times with washing buffer, and the beads were incubated for 4 h with 50 mM imidazole at 4°C. The recombinant protein was obtained by centrifugation at 200 g for 10 min. The GST-ZZ-FLAG-STX17 and GST-ZZ-FLAG-VAMP8 plasmids were transformed into competent E. coli BL21 cells. FLAG-STX17 and FLAG-VAMP8 were purified using suitable affinity beads followed by cleavage with tobacco etch virus (TEV) protease (New England Biolabs, P8112S). The purity of each recombinant protein was assessed by SDS-PAGE and Coomassie blue staining.

In vitro acetylation assay

FLAG-WT SCFD1, FLAG-SCFD1[2KR], and FLAG-SCFD1[3KR] were incubated with recombinant KAT2B-HAT in reaction buffer (20 mM Tris-Cl, pH 8.0, 20% glycerol, 100 mM KCl, 1 mM DTT, 0.2 mM EDTA, 10 mM TSA, and 10 mM nicotinamide). The proteins were separated by SDS-PAGE after the treatment, and the acetylation level was determined using an antibody against acetylated lysine.

mRFP-GFP-LC3 assay

The mRFP-GFP-LC3 assay was performed using the previously described tandem fluorescent-tagged LC3 reporter protein [38]. Briefly, LC3 was fused with both mRFP and GFP. When an autophagosome fuses with a lysosome to form an autolysosome, the acidic pH of the lysosome lowers the pH of the newly formed autolysosome vesicle; for the tandem fluorescent-tagged LC3 reporter, the stability of the mRFP fluorescence signal is unaffected by this change but the intensity of the GFP fluorescence signal is dramatically reduced. Accordingly, a punctum detected as yellow (GFP positive, RFP positive) represents an autophagosome, whereas a red punctum (GFP negative, RFP positive) indicates an autolysosome.

Cells stably expressing FLAG-WT SCFD1, FLAG-SCFD1[2KR], or FLAG-SCFD1[2KQ] were seeded onto coverslips, transfected with the mRFP-GFP-LC3 plasmid, and treated with or without the MTOR inhibitor torin 1. The coverslips were fixed with paraformaldehyde and analyzed by fluorescence microscopy.

We assessed the proportion of autolysosomes (i.e., GFP negative, RFP positive) puncta among the total complement of autophagic vesicles present in each cell, as follows:

Number of autolysosomes (GFP negative, RFP positive)

Number of autophagosomes (GFP positive, RFP positive) + autolysosomes (GFP negative/RFP positive)

(For each examined cell, the number of red puncta was divided by the sum of red puncta + yellow puncta). Statistical significance was evaluated by one-way ANOVA; ***p < 0.001, ****p < 0.0001; Data are presented as means ± SEM. At least 30 cells were evaluated per group.

In vitro affinity-isolation assay

GST-SIRT4 was transformed into E. coli BL21 cells, which were grown in LB buffer media. The cells were extracted and GST-SIRT4 was bound to glutathione-Sepharose 4B beads at 4°C for 3 h. After washing the beads, recombinant His-WT SCFD1 was added to the system and rotated for 2 h at 4°C. The beads were then pelleted at 4°C and the supernatant was discarded. After washing, SDS lysis buffer was added to the GST beads, and western blotting was performed to identify the bound proteins.

Statistical analyses

Statistical significance was evaluated by one-way ANOVA; ***p < 0.001, ****p < 0.0001; Data are presented as means ± SEM. At least 20 cells were evaluated per group.

Materials and data availability

The reagents and data of this study are available from the lead contact, Rong Liu, on reasonable request.

Acknowledgments

We thank for the support by National center for international research on animal gut nutrition and Jiangsu collaborative innovation center of meat production and processing.This work in Liu Lab was supported by the National Natural Science Foundation of China (Grant Nos. 91954115 and 31771532); the Jiangsu Natural Science Funds for Distinguished Young Scholar (Grant No. BK20170025); and the “Shuang chuang”, “Six talent peaks”, and “333” projects in Jiangsu Province. K.R. Mei was supported by National Science Foundation of China (91954112 and 31900501) and the Young Elite Scientists Sponsorship Program by Tianjin (TJSQNTJ-2020-19).

Disclosure statement

The authors declare that they have no conflict of interest.

Supplementary material

Supplemental data for this article can be accessed here

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [91954115 and 31771532]; The Jiangsu Natural Science Funds for Distinguished Young Scholar [Grant No. BK20170025]; and the “Shuang chuang”, “Six talent peaks”, and “333” projects in Jiangsu Province. K.R. Mei was supported by National Science Foundation of China [91954112 and 31900501] and the Young Elite Scientists Sponsorship Program by Tianjin [TJSQNTJ-2020-19].