The Cullin3-Rbx1-KLHL9 E3 ubiquitin ligase complex ubiquitinates Rheb and supports amino acid-induced mTORC1 activation

Abstract

Mechanistic target of rapamycin complex 1 (mTORC1) is recruited to the lysosomal membrane by the active Rag heterodimer, where mTORC1 interacts with active Rheb for its activation. It has been shown that polyubiquitination of Rheb is crucial for enhancing its interaction with mTORC1 on the lysosome. However, the specific ubiquitin ligases for Rheb, which promotes mTORC1 activation, remain elusive. We report that the CUL3-RBX1-KLHL9 E3 ubiquitin ligase complex is translocated to the lysosome and ubiquitinates Rheb in response to amino acid stimulation. KLHL9 serves as an essential adaptor for CUL3-RBX1 to target Rheb on the lysosome. Deleting either CUL3, RBX1, or KLHL9 diminishes Rheb ubiquitination and reduces amino acid-induced mTORC1 activation without impacting lysosomal mTORC1 localization or Akt activity. Thus, the CUL3-RBX1-KLHL9 complex functions as a mTORC1 activator by acting as an E3 ubiquitin ligase for Rheb and supports amino acid-induced mTORC1 activation.

Keywords: CP: Cell biology; CUL3; KLHL9; Rheb; lysosome; mTORC1; ubiquitination.

Figure 1.. The CUL-RBX1 E3 ubiquitin ligase stimulates Ub-Rheb accumulation and mTORC1 activity specifically

(A) Ablation of RBX1 diminishes amino acid-induced mTORC1 activation in EMSCs. Three gRNAs designed to target distinct exon regions of RBX1 were used. Cells were starved in amino acid-deprived medium for 50 min and treated with 1× amino acids for 10 min. gRNA targeting GFP was used as a control. Note that RBX1–3 gRNA failed to ablate RBX1. (B) Ablation of RBX1, but not RBX2, diminishes amino acid-induced mTORC1 activation in EMSCs. Cells were treated as in (A). Levels of phosphorylated S6K1 in the indicated cells (amino acid-replete conditions) were quantified and expressed as the ratio of pS6K/S6K1. *p < 0.05, **p < 0.01, mean ± SD, n = 4. (C) Ablation of RBX1 diminishes CQ/MG-induced mTORC1 activation in Ragulator-deficient cells. RBX1 is ablated in p18/LAMTOR1 knockout (KO) EMSCs. Cells were treated with CQ (50 μM) or/and MG (20 μM) in amino acid-free medium for 45 min. (D) Ablation of RBX1 diminishes Rheb ubiquitination. RBX1 KO EMSCs were lysed under amino acid-replete conditions in RIPA buffer containing the deubiquitinase inhibitor (100 mM) N-ethylmaleimide (NEM). The lysates were subjected to SDS-PAGE, and levels of Rheb and Ub-Rheb were monitored with a Rheb antibody. Non-Ub-Rheb and Ub-Rheb are indicated. (E) MLN4924 specifically inhibits mTORC1 activity induced by amino acids in EMSCs. Cells were treated with the indicated concentrations of MLN4924 for 16 h. Cells were amino acid starved for 50 min and re-stimulated with 1× amino acids for 10 min. Levels of phosphorylated S6K1 in the indicated cells (amino acid-replete conditions) were quantified and expressed as the ratio of pS6K/S6K1. ****p < 0.0001, mean ± SD, n = 4. (F) MLN4924 inhibits mTORC1 activity induced by amino acids in both control and DEPDC5 KO HEK293T cells. Cells were treated with MLN4924 (1 μM) as in (E). Cells were amino acid starved for 50 min and re-stimulated with 1× amino acids for 10 min. The effect of DEPDC5 ablation was confirmed by the high levels of mTORC1 activity resistant to amino acid starvation. (G) MLN4924 diminishes levels of Ub-Rheb in HEK293T cells. Cells were treated with MLN4924 (1 μM) for 16 h and lysed as in (D), and the lysates were subjected to immunoblotting with a Rheb antibody.

Figure 2.. The CUL3-RBX1 E3 ubiquitin ligase stimulates Ub-Rheb accumulation and mTORC1 activity specifically

(A) Ablation of CUL3 diminishes mTORC1 activity induced by amino acids. CUL3 was ablated by using two distinct sgRNAs in EMSC cells. Cells were treated as in Figure 1A. Levels of phosphorylated S6K1 in the indicated cells (amino acid-replete conditions) were quantified and expressed as the ratio of pS6K/S6K1. ***p < 0.001, mean ± SD, n = 3. (B) Ablation of CUL3 diminishes levels of Ub-Rheb. EMSC cells under amino acid-replete conditions were lysed with RIPA buffer containing 100 mM NEM. Cell lysates were subjected to immunoblotting with a Rheb antibody. (C) CUL3 and RBX1 interact with Rheb. HEK293T cells were expressed with HA-RBX1, Myc-CUL3, and FLAG-Rheb or FLAG-Rap2a as a control. 48 h after transfection, cell lysates were subjected to immunoprecipitation (IP) with FLAG antibody (FLAG antibody-conjugated agarose). CoIPed HA-RBX1 and Myc-CUL3 were monitored with HA and Myc antibodies, respectively. (D) Ablation of CUL3 has little effect on amino acid-induced lysosomal mTOR localization. EMSCs with gRNAs targeting CUL3 or GFP as a control were starved of amino acid for 50 min and treated with 1× amino acids for 10 min. Cells were fixed with 4% paraformaldehyde and stained with mTOR antibody (red) and LAMP2 (green) antibody. The scale bar indicates 10 μm. Pearson’s correlation coefficient R values were determined to quantify the colocalization of mTOR and LAMP2. **p < 0.01, ****p < 0.0001, mean ± SD; n = 8, 3, 6, and 19 (images) and 55, 18, 25, and 84 (cells).

Figure 3.. KLHL9 acts as an adopter protein for the CUL3-RBX1 E3 ubiquitin ligase complex to interact with Rheb

(A) KLHL9 specifically interacts with Rheb. HEK293T cells were expressed with Myc-Rheb and FLAG-KLHL9 or FLAG-Metap2 as a control. The cell lysates were subjected to IP with FLAG antibody. CoIPed Myc-Rheb was monitored with Myc antibody. (B) Rheb specifically interacts with KLHL9. HEK293T cells were expressed with FLAG-KLHL9 and Myc-Rheb or Myc-Rap2 as a control. The cell lysates were subjected to IP with Myc antibody. CoIPed FLAG-KLHL9 was monitored with FLAG antibody. (C) Rheb interacts with KLHL9 but not with KLHL22. HEK293T cells were expressed with Myc-Rheb and FLAG-Metap2, FLAG-KLHL9, or FLAG-KLHL22. Cell lysates were subjected to IP with FLAG antibody. CoIPed Myc-Rheb was monitored. (D) Endogenous KLHL9 interacts with endogenous Rheb. HEK293T cells were lysed and subjected to IP with KLHL9 antibody. CoIPed endogenous Rheb, CUL3, and RBX1 were monitored. (E) Endogenous Rheb interacts with endogenous KLHL9. EMSC lysates were subjected to IP with control, Rheb, or Rab7 antibody. CoIPed endogenous KLHL9 was monitored. An asterisk indicates a non-specific band. (F) The kelch domain of KLHL9 mediates the interaction between KLHL9 with Rheb. The indicated various FLAG-KLHL9s with truncation or deletion mutants were expressed with Myc-Rheb and Myc-CUL3 in HEK293T. Cell lysates were subjected to IP with FLAG antibody, and coIPed Myc-Rheb and Myc-CUL3 were monitored. (G) KLHL9 is required for the interaction between Rheb and CUL3. Myc-CUL3 and FLAG-Rheb or FLAG-Rap2a (control) were expressed in KLHL9-ablated or control HEK293T cells. Cell lysates were subjected to IP with FLAG antibody, and coIPed Myc-CUL3 was monitored. (H) KLHL9 enhances the interaction between Rheb and CUL3. HEK293T cells were expressed with the indicated proteins. The effect of ectopic expression of KLHL9 on the interaction between Myc-Rheb and FLAG-CUL3 was determined. Myc-Rap2a was included as a control.

Figure 4.. CUL3-RBX1-KLHL9 acts as an E3 ubiquitin ligase for Rheb

(A) KLHL9 supports the accumulation of Ub-Rheb. Myc-Rheb and HA-ubiquitin were expressed in HEK293T cells in the presence or absence of FLAG-KLHL9. Levels of Ub-Rheb were monitored in the lysates and IPed Myc-Rheb. (B) MLN4924 diminishes KLHL9-induced Ub-Rheb accumulation. HEK293T cells were expressed with FLAG-KLHL9, Myc-Rheb, and HA-ubiquitin and treated with MLN4924 (1 μM) for 16 h. Cells were lysed in RIPA buffer containing NEM (100 mM). Levels of Ub-Rheb were monitored in the lysates with a Myc antibody or in the IPed Myc-Rheb with an HA antibody. (C) Ablation of KLHL9 diminishes Ub-Rheb. Endogenous KLHL9 was ablated by two distinct sgRNAs in HEK293T cells under amino acid-replete conditions. Cells were lysed with RIPA buffer containing NEM (100 mM), and endogenous Ub-Rheb was monitored by a Rheb antibody. (D) KLHL9 supports amino acid-induced Ub-Rheb accumulation. The indicated EMSCs were starved of amino acids for 50 min and re-stimulated with 1× amino acids for 10 min. Endogenous Ub-Rheb was monitored. (E) KLHL9-induced polyubiquitination of Rheb was abolished in the Rheb 4Rs mutants. HEK293T cells were expressed with Myc-KLHL9 and FLAG-Rheb or the FLAG-Rheb 4Rs mutant (K109/135/151/178R). Cells were lysed with RIPA buffer containing NEM (100 mM) and subjected to IP with a FLAG antibody. Levels of Ub-Rheb were monitored by the Rheb antibody. (F) The CUL3-RBX1-KLHL9 complex ubiquitinates Rheb in vitro. HEK293T cells were expressed with FLAG-KLHL9, Myc-CUL3, and HA-RBX1, and the CUL3-RBX1-KLHL9 E3 ubiquitin complex was purified by FLAG antibody and eluted with FLAG peptides. Non-ubiquitinated glutathione S-transferase (GST)-Rheb and GST alone (control) was generated in E. coli and used as substrates. The in vitro ubiquitin reaction was performed by adding recombinant UBE1 (E1), UbcH5a (E2), and ubiquitin at 37°C for 2 h, and the reaction was terminated by Laemmli buffer. The ubiquitinated Rheb was determined by the Rheb antibody. (G) Kelch domain-mediated KLHL9-Rheb binding is required for CUL3-RBX1-KLHL9-dependent Rheb ubiquitination in vitro. The CUL3-RBX1 complex containing the KLHL9 delta Kelch domain was purified, and in vitro ubiquitin assays were performed as in (F).

Figure 5.. Lysosomal localization of Rheb is required for its binding to the CUL3-RBX1-KLHL9 complex

(A) The CUL3-RBX1-KLHL9 complex is expressed on the lysosome. HA-tagged TMEM192, a lysosomal membrane protein, was stably expressed in HEK293T cells. Lysosomes were immunopurified with HA-antibody-conjugated beads, and the indicated lysosomal and non-lysosomal proteins were monitored. (B) Dissociation of Rheb from the membrane by treatment with FTI-277, a farnesyl transferase inhibitor, diminishes its interaction with KLHL9. HEK293T cells expressing the indicated proteins were treated with FTI-277 (20 μM) for 24 h. Cell lysates were subjected to IP with Myc antibody, and coIPed FLAG-KLHL9 was monitored. (C) Dissociation of Rheb from the membrane by treating FTI-277 diminishes its ubiquitination. HEK293T cells expressed with Myc-Rheb or Myc-Rap2a were treated with FTI-277 as in (B). Cell lysates were subjected to IP with goat Myc antibody, and levels of ubiquitinated Myc-Rheb were monitored with mouse Myc antibody. (D) Dissociation of Rheb from the membrane by the C181S mutation disrupts Rheb-KLHL9 interaction. HEK293T cells were expressed with FLAG-KLHL9 and Myc-Rheb or the Myc-Rheb C181S mutant. Cell lysates were subjected to IP with Myc antibody, and coIPed FLAG-KLHL9 was monitored. (E) KLHL9-induced Rheb ubiquitination is mitigated in the Rheb C181S mutant. HEK293T cells were expressed with FLAG-KLHL9 and Myc-Rheb or the Myc Rheb C181S mutant, and cell lysates were subjected to IP with Myc antibody. Levels of Ub-Rheb were monitored with Myc antibody as in (C). (F) Tethering Rheb to lysosomes but not the ER enhances Rheb-KLHL9 interaction. The Rheb C181S mutant was fused with the lysosomal target signal peptides from LAMTOR1 or the ER target signal peptides from calreticulin with the KDEL ER retention signal to establish Lyso-myc-Rheb C181S and ER-myc-Rheb C181S. The indicated proteins, including those Rheb mutants, were expressed in HEK293T cells, and cell lysates were subjected to IP with Myc-Rheb. Levels of co-IPed FLAG-KLHL9 were monitored. (G) Tethering Rheb to lysosomes but not ERs enhances Rheb ubiquitination. HEK293T cells were expressed with the indicated Myc-Rheb or Rheb mutants. Levels of Ub-Rheb were monitored with Myc antibody.

Figure 6.. Amino acids enhance lysosomal CUL3-RBX1-KLHL9 complex localization and its interaction with Rheb

(A) Amino acids increase lysosomal CUL3-RBX1-KLHL9 expression. HEK293T cells stably expressing HA-TMEM192 were used for immune-purifying lysosomes in the presence or absence of amino acids in the medium. The indicated proteins, including the subunits of the CUL3-RBX1-KLHL9 complex, were monitored. (B) Amino acids increase lysosomal localization of KLHL9. HEK293T cells expressing FLAG-KLHL9 were starved and stimulated with amino acids for 15 min. The cells were fixed, and double immunofluorescence staining with FLAG antibody and LAMP2 antibody was performed. The scale bar indicates 10 μm. See Figure S3C for the quantification of colocalization. (C) Amino acids enhance the interaction between Rheb and KLHL9. HEK293T cells were expressed with Myc-Rheb or Myc-Rap2a as a control, and cell lysates were subjected to IP with Myc antibody. Levels of coIPed endogenous KLHL9 were monitored. Cells were starved of amino acids for 50 min and stimulated with 1× amino acids for 15 min. (D) Dissociation of Rheb from the lysosome diminishes lysosomal KLHL9 expression. Lysosomes were immunopurified as in (A) in the presence or absence of FTI-277 treatment in amino acid-replete HEK293T cells. Levels of coIPed endogenous Rheb and KLHL9 with the lysosome were monitored. (E) Ablation of Rheb diminishes lysosomal KLHL9 expression. Lysosomes were immunopurified as in (A) from amino acid-replete HEK293T cells lacking endogenous Rheb expression. Levels of coIPed endogenous Rheb and KLHL9 with the lysosome were determined.

Figure 7.. KLHL9 positively contributes to amino acid-induced mTORC1 activation

(A) Ablation of KLHL9 inhibits amino acid-induced mTORC1 activation specifically. Endogenous KLHL9 was ablated by two distinct sgRNAs in EMSCs. Cells were starved of amino acids for 50 min and restimulated with 1× amino acids for 10 min. Levels of phosphorylated S6K1 in the indicated cells (amino acid-replete conditions) were quantified and expressed as the ratio of pS6K/S6K1. **p < 0.01, mean ± SD, n = 4. (B) Ablation of KLHL9 inhibits amino acid-induced mTORC1 activation. Similar experiments shown in (A) were conducted in HEK293T cells. (C) Increased KLHL9 expression enhanced amino acid-induced mTORC1 activation. FLAG-KLHL9 was stably expressed at different levels in HEK293T cells, and amino acid sensitivity for mTORC1 activation was determined. Levels of phosphorylated S6K1 in the indicated cells (amino acid-replete conditions) were quantified and expressed as the ratio of pS6K/S6K1. **p < 0.01, mean ± SD, n = 5. (D) Increased KLHL9 expression enhanced the interaction between Rheb and mTOR. HEK293T cells were expressed with the indicated constructs. HA-Rheb was IPed with HA antibody, and coIPed endogenous mTOR was monitored. The intensity of coIPed endogenous mTOR and IPed HA-Rheb was quantified, and the values (mTOR/HA-Rheb) were compared in the presence or absence of FLAG-KLHL9 expression. Data are expressed as fold increase; ***p < 0.001, mean ± SE, n = 3. (E) Model for CUL3-RBX1-KLHL9-induced Rheb ubiquitination and mTORC1 activation. Under growth factor- and amino acid-limited conditions, Rheb is inactivated by the GAP activity of TSC2. Simultaneously, ATXN3, a deubiquitinase of Rheb, is recruited on the lysosome by the inactive Rag small GTPases and deubiquitinates Rheb, leading to elimination of the interaction between Rheb and mTORC1 and blockade of Rag-independent mTORC1 lysosomal localization. Under growth factor- and amino acid-replete conditions, growth factor-induced Akt activation renders Rheb active by inhibiting the GAP activity of TSC2. Amino acid sufficiency converts inactive Rag small GTPases to active, which induces dissociation of ATXN3 from the lysosome, placing Rheb in permissive conditions for its ubiquitination, but recruits mTORC1 to the lysosome. Simultaneously, amino acids, but not growth factors, trigger lysosomal localization of the CUL3-RBX1-KLHL9 ubiquitin ligase complex and its interaction with Rheb on the lysosome, where the CUL3-RBX1-KLHL9 ubiquitin ligase complex polyubiquitinates Rheb. Ub-Rheb may form a multimer with non-ubiquitinated Rheb, facilitating its interaction with mTORC1 and contributing to growth factor- and amino acid-induced canonical mTORC1.