Proteasome inhibition suppresses the phosphorylation of SQSTM1 S403

Since SQSTM1 S403 phosphorylation is essential for the aggrephagy of ubiquitinated proteins [25], we studied the influence of proteasome inhibition on SQSTM1 S403 phosphorylation. We utilized the proteasome inhibitor MG132 and the autophagic flux inhibitor Bafilomycin A1 to reduce proteasome activity and block autophagic flux, either alone or in combination and measured the phosphorylation level of SQSTM1 S403. As depicted in Fig. 1 A, we discovered that in Hela cells, both suppression of proteasome activity and inhibition of autophagy can induce a substantial rise in SQSTM1 S403 phosphorylation. The combination inhibition of proteasome activity and autophagy had no further effect on the accumulation of S403 phosphorylation as compared to the therapy alone, indicating that inhibition of proteasome and blocking of autophagy may share the same fundamental mechanism for regulating the phosphorylation of SQSTM1 S403. Matsumoto et al. have demonstrated that phosphorylated S403 can be quickly eliminated by the autophagy pathway [25]. Sha and colleagues have reported that the initiation of autophagy is postponed following proteasome inhibition [28]. The aggresome formation of misfolded protein prior to the activation of autophagy emerges as a crucial strategy for cellular defense against proteotoxic crises [28]. As reported in our previous study [20], autophagosome production in cultured cells is inhibited in the early stage of proteasome inhibition (Fig. 1A–D). Therefore, the increase of S403 phosphorylation caused by a proteasome inhibitor may result from autophagy inhibition. Intriguingly, we discovered that inhibiting autophagy in AD293 greatly increased the accumulation of phosphorylated SQSTM1S403 relative to proteasome inhibition. Moreover, we also showed that the proteasome inhibitor can considerably diminish the accumulated S403 phosphorylation caused by autophagy inhibition (Fig. 1B). These data show that Hela cells lack a molecular mechanism that underlies the suppression of SQSTM1 S403 phosphorylation in response to inhibition of the proteasome in AD293 cells.

Fig. 1

SQSTM1 S403 phosphorylation does not affect the autophagic degradation of ubiquitinated proteins during proteasome inhibition. A–B Hela cells and AD293 cells were treated with MG132 (2 μM) and Bafilomycin A1 (25 nM), alone or in combination for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. Data are mean ± SEM of three independent experiments; **P < 0.01, ***P < 0.001, NS no significance. C Hela cells stably expressing GFP-LC3 were treated with MG132 (1 μM) and Bafilomycin A1 (25 nM), alone or in combination for 14 h. The cells were fixed, the images were captured with confocal microscope. Scale bar: 10 μm. D Quantitative analysis of results in C. About 25 cells from three independent experiments were scored for each group. Data are mean ± SEM; ***P ≤ 0.001. E Hela cells were treated with MG132 (2 μM) and Bafilomycin A1 (25 nM), alone or in combination for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. F-G Hela cells were treated with MG132 (2 μM) alone, or combined with BX-795 (5 μM)/CX-4945 (10 μM)/SBI-0206965 (10 μM) for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. Data are mean ± SEM of three independent experiments; **P < 0.01, ***P < 0.001, NS no significance. H SQSTM1 knockout AD293 cells stably expressing indicated constructs were treated with or without MG132 (2 μM) for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies

SQSTM1 S403 phosphorylation does not affect the autophagic degradation of ubiquitinated proteins during proteasome inhibition

Next, we examined whether the phosphorylation of SQSTM1 S403 influences the autophagic degradation of ubiquitinated proteins during proteasome suppression. As depicted in Fig. 1E, in Hela cells, blocking autophagy did not increase the accumulation of polyubiquitinated proteins compared to inhibiting the proteasome alone. According to previous research, TBK1, CK2, and ULK1 are the primary kinases that phosphorylate SQSTM1 S403 [25,26,27]. We attempted to suppress SQSTM1 S403 phosphorylation using targeted inhibitors for these kinases. As depicted in Fig. 1F, TBK1 inhibitor BX-795, and CK2 inhibitor CX-4945 could substantially reduce the phosphorylation level of SQSTM1 S403 enhanced by proteasome inhibitors, but not the ULK1 inhibitor, suggesting that ULK1 is not involved in the regulation of SQSTM1 S403 phosphorylation under our treatment conditions. However, we did not observe the further accumulation of ubiquitinated proteins in Hela cells cotreated with CK2 or TBK1 inhibitor and a proteasome inhibitor (Fig. 1G). These results demonstrate that although proteasome inhibitors can significantly increase SQSTM1 S403 phosphorylation, this does not cause the autophagic degradation of ubiquitinated proteins in Hela cells. To validate further the nonfunctional involvement of SQSTM1 S403 phosphorylation in the degradation of ubiquitinated proteins during proteasome inhibition, In SQSTM1 knockout AD293 cells, we re-expressed wildtype SQSTM1, the non-phosphorylatable (S403A) mutant, and the phosphomimetic (S403E) mutant. Then we evaluated the impact of these mutants on ubiquitinated protein clearance. As depicted in Fig. 1H, neither S403A nor S403E mutation significantly altered the number of ubiquitinated proteins during proteasome suppression. These findings suggest that SQSTM1 S403 phosphorylation does not affect the autophagic degradation of ubiquitinated proteins when the proteasome is blocked.

Phosphorylation of SQSTM1 S403 suppresses the aggresome formation of ubiquitinated proteins during proteasome inhibition

Next, we explored whether SQSTM1 S403 phosphorylation influences the proteasome inhibition-induced aggresome generation of ubiquitinated proteins. According to previous research, the rate of aggresome production in Hela cells was slower than AD293 and A375 cells [6, 20]. Intriguingly, we discovered that blocking SQSTM1 S403 phosphorylation with a CK2 inhibitor promoted the aggresome synthesis of ubiquitinated proteins in Hela cells (Fig. 2A, B), as evidenced by K48-linked polyubiquitinated proteins (UB-K48), a well mark of proteasome inhibition-induced aggresome [6, 20, 29]. Besides, we also observed the SQSTM1 S403A overexpression significantly enhanced the aggresome formation compared to SQSTM1 WT or SQSTM1 S403E (Fig. 2C, D). In SQSTM1 knockout cells that re-expressed wildtype or mutant SQSTM1, we discovered that re-expressing S403E decreased aggresome formation during proteasome suppression (Fig. 2E, F). These data indicate that non-phosphorylation of SQSTM S403 is necessary for aggresome formation during proteasome suppression.

Fig. 2

Phosphorylation of SQSTM1 S403 suppresses the aggresome formation of ubiquitinated proteins during proteasome inhibition. A Hela cells were treated with MG132 (1 μM) alone or combined with CX-4945 (5 μM) for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. B Quantitative analysis of results in (A). C HeLa cells were transfected with plasmids expressing FLAG-SQSTM1 wildtype (WT), FLAG-SQSTM1 (403A), FLAG-SQSTM1 (403E) for 24 h, and then treated with 1 μM MG132 for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 (Red) and anti-FLAG (Green) antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. D Quantitative analysis of results in C. E SQSTM1 knockout AD293 cells stably expressing indicated constructs were treated with 1 μM MG132 for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 (Red) and anti-FLAG (Green) antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. F Quantitative analysis of results in E. For B, D, and F, at least 50 cells were randomly selected from each group to score for aggresomes. Data are mean ± SEM of three independent experiments. *P < 0.05, **P < 0.01

Phosphorylation of SQSTM1 S403 promotes its autophagic sequestration

According to previous work, autophagic sequestration of SQSTM1 prevented the formation of ubiquitinated protein aggresomes [20]. We assume that S403 phosphorylation may exacerbate the autophagic sequestration of SQSTM1, which suppresses the aggresome formation. Therefore, we examined the co-localization of mutant SQSTM1 and autophagosome and found that GFP-LC3B, a well marker of autophagosome, robustly co-localized with SQSTM1 S403E and UB-K48 in micro-aggregates (Fig. 3A). In contrast, the GFP-LC3B signals were weak in SQSTM1 S403A and UB-K48-positive aggresomes (Fig. 3A). Additionally, we discovered that Wortmannin, an inhibitor of autophagy initiation, can restore SQSTM1 S403E-induced defection of aggresome formation (Fig. 3B, C). These findings indicate that SQSTM1 S403 phosphorylation may interfere with the proteasome inhibition-induced aggresome formation by aggravating the autophagic sequestration of SQSTM1 and ubiquitinated proteins.

Fig. 3

Phosphorylation of SQSTM1 S403 promotes its autophagic sequestration. A HeLa cells stably expressing EGFP-LC3B were transfected with plasmids expressing FLAG-SQSTM1(S403A) or FLAG-SQSTM1(S403E) for 24 h, and then treated with MG132 (1 μM) for 14 h. The cells were then fixed and analyzed by immunostaining with anti-UB-k48 (red) and anti-FLAG (meganta) antibodies. Scale bar: 10 μm. B SQSTM1 knockout AD293 cells stably expressing FLAG-SQSTM1(S403E) were treated with 1 μM MG132 alone or combined with Wortmannin (5 μM) for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. C Quantitative analysis of results in B. At least 50 cells were randomly selected from each group to score for aggresomes. Data are mean ± SEM of three independent experiments. *P < 0.05, **P < 0.01

Phosphorylation of SQSTM1 T269/S272 inhibits the S403 phosphorylation during proteasome inhibition

Our previous studies have revealed that compared to AD293 cells and A375 cells, the SQSTM1 T269/S272 phosphorylation was limited in Hela cells due to the much lower expression of p38γ/δ that can phosphorylate SQSTM1 at T269/S272 under proteasome inhibition [6, 20]. Therefore, we speculate that the failure of proteasome inhibitor-induced suppression of SQSTM1 S403 phosphorylation in Hela cells may be caused by the ineffective SQSTM1 T269/S272 phosphorylation. To test this hypothesis, we first examined the phosphorylation levels of these sites in AD293 cells and Hela cells after being treated with proteasome inhibitors and autophagy blockers. As shown in Fig. 4A, B, when proteasome was suppressed alone or with Bafilomycin A1 in AD293 cells, the robust signal of T269/S272 phosphorylation matched the weak signal of S403 phosphorylation. Contrarily, in Hela cells, the phosphorylation level of T269/S272 did not change significantly; the phosphorylation of S403 remained at a high level, regardless of inhibition of proteasome alone or/and blocking autophagy (Fig. 4A, B). These results implicate SQSTM1 T269/S272 phosphorylation may inhibit its S403 phosphorylation. Next, we overexpressed SQSTM1 T269A/S272A and SQSTM1 T269E/S272D in Hela cells and immunofluorescence revealed the mutants' S403 site phosphorylation. We observed that after inhibiting proteasome activity and autophagic flux, the signal of S403 phosphorylation in the T269E/S272D mutant was significantly reduced compared with that in T269A/S272A mutant (Fig. 4C, D). In addition, we found overexpressed constitutively active mutants, p38γ (D179A) and p38δ (F324S), which could phosphorylate SQSTM1 T269/S272 in response to proteasome inhibition [6, 20], can significantly enhance the phosphorylation of SQSTM1 T269/S272 and inhibit the phosphorylation of SQSTM1 S403 during proteasome inhibition, rather than the kinase-dead mutants, p38γ (K56R) and p38δ (K54R) (Fig. 4E, F). SQSTM1 mutants re-expressed in SQSTM1 knockout cells were examined for S403 phosphorylation to confirm the effect of SQSTM1 T269/S272 phosphorylation. As shown in Fig. 4G, H, compared with the wildtype SQSTM1 and T269E/S272D mutants, the phosphorylation level of S403 in the T269A/S272A mutant was significantly increased during proteasome inhibition. Besides, we also observed that Doramapimod (Doram), a p38γ/p38δ inhibitor, can enhance the phosphorylation of S403 in AD293 and A375 cells under proteasome inhibition (Fig. 4I–K). These results indicate that proteasome inhibition induced SQSTM1 T269/S272 phosphorylation suppresses its S403 phosphorylation.

Fig. 4

Phosphorylation of SQSTM1 T269/S272 inhibits the S403 phosphorylation during proteasome inhibition. A AD293 and Hela cells were treated with MG132 (2 μM), and Bafilomycin A1 (25 nM), alone or in combination for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. B Quantitative analysis of results in A. C HeLa cells were transfected with plasmids expressing FLAG-SQSTM1(T269A/S272A) or FLAG-SQSTM1(T269E/S272D) for 24 h, and then treated with MG132 (1 μM) and Bafilomycin A1 (25 nM) for 14 h. The cells were then fixed and analyzed by immunostaining with phospho-SQSTM1 (S403)-specific antibodies (Red) and anti-FLAG (Meganta) antibodies. Nuclei were stained with DAPI (blue). Scale bar: 10 μm. D Quantitative analysis of results in C, at least 25 cells from three independent experiments were scored for each group. E HeLa cells were transfected with plasmids expressing p38γ or p38δ mutants for 24 h, and then treated with MG132 (2 μM) for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. F Quantitative analysis of results in E. G SQSTM1 knockout AD293 cells stably expressing indicated constructs were treated with or without MG132 (2 μM) for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. H Quantitative analysis of results in G. I AD293 cells and A375 cells were treated with MG132 (2 μM), and Doramapimod (50 μM), alone or in combination for 14 h. The whole-cell lysates were subjected to western blot analysis with indicated antibodies. J, K Quantitative analysis of results in I. Data are mean ± SEM of three independent experiments; *P < 0.05, **P < 0.01, ***P < 0.001, NS no significance

Suppressing S403 phosphorylation rescues the defective aggresome formation caused by unphosphorylated SQSTM1 (T269/S272)

We predicted that SQSTM1 S403 phosphorylation would limit the aggresome formation of ubiquitinated proteins and suppress T269/S272 phosphorylation-enhanced aggresome formation. Next, we examined the impact of S403 phosphorylation on T269/S272 phosphorylation-regulated aggresome production. First, we discovered that a mutation of S403A can repair the impairment in the aggresome formation of ubiquitinated proteins caused by the T269A/S272A mutant (Fig. 5A, B). We also discovered that inhibiting CK2 can rescue this impairment (Fig. 5C, D). Furthermore, we studied the effect of CK2 inhibition on the defection of aggresome formation mediated by Doram and showed that the defection may be successfully reversed by CK2 inhibitors in AD293 and A375 cells (Fig. 5E–H). Inhibition of S403 phosphorylation appears to be the crucial mechanism for T269/S272 phosphorylation to increase the development of ubiquitinated protein aggresomes in response to proteasome inhibition.

Fig. 5

Suppressing S403 phosphorylation rescues the defective aggresome formation caused by unphosphorylated SQSTM1 (T269/S272). A SQSTM1 knockout AD293 cells stably expressing FLAG-SQSTM1(WT), FLAG-SQSTM1(T269A/S272A), or FLAG-SQSTM1(T269A/S272A/S403A) were treated with 1 μM MG132 for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 (Red) and anti-FLAG (Green) antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. B Quantitative analysis of results in A. C SQSTM1 knockout AD293 cells stably expressing FLAG-SQSTM1(T269A/S272A) were treated with 1 μM MG132 alone or combined with CX-4945 (5 μM) for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 (Red) and anti-FLAG (Green) antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. D Quantitative analysis of results in C. E AD293 cells were treated with MG132 (1 μM), Doramapimod (50 μM), and CX-4945 (5 μM), alone or in combination for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 (Green) antibodies. Scale bar: 20 μm. F Quantitative analysis of results in E. G A375 cells were treated with MG132 (1 μM), Doramapimod (50 μM), and CX-4945 (5 μM), alone or in combination for 14 h. The aggresome formation was analyzed by immunostaining with anti-UB-K48 (Red) and anti-SQSTM1 (Green) antibodies. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. H Quantitative analysis of results in G. For quantitative analysis of aggresome formation, at least 50 cells were randomly selected from each group to score for aggresomes. Data are mean ± SEM of three independent experiments. **P < 0.01

Suppressing S403 phosphorylation protects cells from proteasome inhibition-induced cell death

Since aggresome formation protects against cell death caused by proteasome inhibition, we wondered if blocking S403 phosphorylation might protect cells against proteasome inhibitor-induced cell death. To investigate this, we first studied the effect of SQSTM1 mutants on the cell viability of SQSTM1 knockout cells after being treated with proteasome inhibitors. Figure 6A shows that the S403E mutant caused more cell damage than S403A. Besides, we also found S403A mutation could decrease the cell damage aggravated by T269A/S272A mutation (Fig. 6A). Secondly, as the suppressor of S403 phosphorylation, CX-4945 could reverse the damage of AD293 cells aggravated by Doram (Fig. 6B). We also found that CX-4945 reduces cell damage caused by Bortezomib (Fig. 6C, D), another proteasome inhibitor, combined with Doram. These results suggest that inhibition of SQSTM1 S403 phosphorylation is conducive to cell defense against proteotoxic crisis during proteasome inhibition.

Fig. 6

Suppressing S403 phosphorylation protects cells from proteasome inhibition-induced cell death. A SQSTM1 knockout AD293 cells stably re-expressing FLAG-SQSTM1(WT or mutants), were treated with 2 μM MG132 for 36, and then examined the cell viability with CCK-8 assay. B AD293 cells were treated with MG132 (1 μM), Doramapimod (25 μM), and CX-4945 (1 μM), alone or in combination for 24 h, the cell viability was examined with CCK-8 assay. C, D AD293 cells were treated with Bortezomib (20 nM), Doramapimod (25 μM), and CX-4945 (1 μM), alone or in combination for 24 h, and then the images were acquired by brightfield microscopy (C), the cell viability analyzed with CCK-8 assay (D). Scale bar: 100 μm. For A, B and D, data are mean ± SEM of three independent experiments. *P < 0.05, **P < 0.01