Mdm20 Modulates Actin Remodeling through the mTORC2 Pathway via Its Effect on Rictor Expression

Abstract

NatB is an N-terminal acetyltransferase consisting of a catalytic Nat5 subunit and an auxiliary Mdm20 subunit. In yeast, NatB acetylates N-terminal methionines of proteins during de novo protein synthesis and also regulates actin remodeling through N-terminal acetylation of tropomyosin (Trpm), which stabilizes the actin cytoskeleton by interacting with actin. However, in mammalian cells, the biological functions of the Mdm20 and Nat5 subunits are not well understood. In the present study, we show for the first time that Mdm20-knockdown (KD), but not Nat5-KD, in HEK293 and HeLa cells suppresses not only cell growth, but also cellular motility. Although stress fibers were formed in Mdm20-KD cells, and not in control or Nat5-KD cells, the localization of Trpm did not coincide with the formation of stress fibers in Mdm20-KD cells. Notably, knockdown of Mdm20 reduced the expression of Rictor, an mTORC2 complex component, through post-translational regulation. Additionally, PKCαS657 phosphorylation, which regulates the organization of the actin cytoskeleton, was also reduced in Mdm20-KD cells. Our data also suggest that FoxO1 phosphorylation is regulated by the Mdm20-mTORC2-Akt pathway in response to serum starvation and insulin stimulation. Taken together, the present findings suggest that Mdm20 acts as a novel regulator of Rictor, thereby controlling mTORC2 activity, and leading to the activation of PKCαS657 and FoxO1.

Fig 1. Mdm20 deficiency suppresses cell growth.

A. Growth curve for Mdm20-KD and Nat5-KD HEK293 cells. The left panel shows the effectiveness of the siRNAs (control, Mdm20, and Nat5) 72 h post-transfection. After the transfection of cells with control, Mdm20, and Nat5 siRNA oligonucleotides, the number of cells was counted at the indicated time points and compared with the number of control cells. The data represent the mean ± the S.D. (n = 5). ^★P<0.05, ^★★P<0.005 and ^★★★P<0.0005 indicate statistical significance between control and Mdm20-KD-1 cells. *P<0.005 and **P<0.0005 indicate statistical significance between control and Mdm20-KD-2 cells. B. Mdm20- and Nat5-KD HeLa cells displayed suppressed cell growth. Left panel: Western blot showing the effectiveness of the siRNAs (control, Mdm20 and Nat5). Right panel: Evaluation of cell growth using a similar method to A. The data represent the mean ± the S.D. (n = 5). *P<0.02 and **P<0.0001 indicate statistical significance between control and Mdm20-KD cells. C. Representative images of HEK293 cell growth at 0, 24, 48 and 72 h after siRNA transfection (Scale bar: 100 μm.).

Fig 2. Effect of NatB deficiency on cell cycle progression.

A. Cell cycle progression distribution in Fucci-expressing HeLa cells after transfection with control, Mdm20-, and Nat5 siRNA oligonucleotides. The upper panel (green: a, b, and c) and middle panel (red: d, e, and f) show S/G2/M and G1 phases, respectively. The G1/S interphase is shown as a merged image of the cells in panels g, h, and i. (Scale bar: 50 μm.) B. Comparison of cell cycle progression in Mdm20- and Nat5-KD cells. The percentage of cells in each cell cycle phase (green: S, G2 and M phase; red: G1 phase, and orange: G1/S interphase) was calculated as the total number of cells. The data represent the mean ± the S.D. (n = 3). *P<0.02 and **P<0.005.

Fig 3. Mdm20 modulates cellular motility via its effects on the actin cytoskeleton.

A. Cell migration assay using Mdm20-KD HeLa (left) and HepG2 cells (right). Seventy-two hours after transfection with two types of Mdm20 siRNA oligonucleotides, cells were applied to a Transwell chamber and incubated for 6 h. Non-migrating cells were then removed, and migrating cells were counted. The data represent the mean ± the S.D. (n = 3). *P<0.01, **P<0.005, and ***P<0.002. B. Immunohistochemistry showing the cellular localization of actin (upper panels: a-c) and paxillin (lower panels: d-f) in HeLa cells transfected with siRNA oligonucleotides. Merged images of cells counterstained with DAPI (blue) are shown in panels g, h, and i. The cells were fixed and the images were captured 72 hr post-transfection. The arrows indicate stress fibers. (Scale bar: 20 μm.)

Fig 4. NatB affects Trpm assembly but is not linked to actin fiber formation.

A. Distribution of actin and Trpm in Mdm20- and Nat5-KD HeLa cells. After fixing HeLa cells with paraformaldehyde, the cells were immunostained with phalloidin (green: a, e, and i) and anti-Trpm (red: b, f, and j) antibodies. Merged images of cells counterstained with DAPI (blue) are shown in panels c, g, and k. Low magnification (b, f, and j) and partial zoom up magnification views (d, h, and l) of the indicated regions (dotted line squares) are shown. Arrows indicate the vacuoles. (Scale bar: 20 μm.) B. Comparison of the number of vacuoles in Mdm20- and Nat5-KD cells. Cells were classified into three categories based on the number of vacuole-like structures formed, category 1 (0 to 2 (white area)), category 2 (3 to 5 (light gray area)) and category 3 (over 5 (dark gray area)). The relative ratios of the three categories were calculated based on total cell number. The data represent the mean ± the S.D. (n = 5). C. Western blots for Trpm and cofilin in HeLa cells transfected with siRNA oligonucleotides.

Fig 5. Mdm20 modulates mTORC2 activity through Rictor expression.

A. Western blots for Akt and PKCα with a focus on the phosphorylated forms. HEK293 cells were transfected with control, Mdm20, and Nat5 siRNAs and after 72 h, the transfected cells were harvested and cell lysates were prepared and processed for Western blot analysis. The antibodies used are indicated. B. Western blotting was used to analyze mTORC2 activity. (a) Western blots using the indicated antibodies. (b) The amounts of Rictor and Raptor were calculated relative to those of cell extracts from control siRNA-transfected cells. The data represent the mean ± the S.D. (n = 5). *P<0.0001 and **P<0.01. C. Distribution of actin and paxillin in Mdm20-, Nat5-, and Rictor-KD HeLa cells. After fixing the HeLa cells with paraformaldehyde, the cells were immunostained with phalloidin (green) and anti-paxillin (red) antibodies and were then counterstained with DAPI (blue). (Scale bar: 20 μm.)

Fig 6. Reduction in Rictor expression in Mdm20-KD cells is related to post-translational but not transcriptional regulation.

A. Western blots showing the Rictor and Nat5 protein in Mdm20-, Nat5- and Rictor-KD cells treated with MG132 (0.2 μM) or NH₄Cl (7.5 mM). B. Quantitative real-time PCR for Nat5 and Rictor mRNA expression was performed using reverse transcription products obtained from control, Mdm20-, Nat5-, and Rictor-KD cells. GAPDH mRNA was used as an internal control for this assay. The data represent the mean ± the S.D. (n = 3). *P<0.01 and **P<0.0005.

Fig 7. Effect of reduced levels of Mdm20 on FoxO1 localization and phosphorylation in Mdm20-KD cells.

A. Immunohistochemistry against FoxO1 in Mdm20-, Nat5- and Rictor-KD HEK293 cells. Panels a, d, g and j are immunostained with anti-FoxO1 antibody; these images are merged with DAPI (blue) staining in panels b, e, h, and k. Low magnification (a, d, g, and j) and partial zoom up views (c, f, i, and l) of the indicated area (dotted line squares) are shown. (Scale bar: 20 μm.) B. Western blots of pFoxO1 to evaluate its induction in response to insulin stimulation. Control, Mdm20-KD, Nat5-KD, and Rictor-KD HEK293 cells were serum starved (0.2% FBS) for 20 h, and were then treated with insulin (100 nM) for 30 min. Note that the pAkt bands are less intense in extracts of serum starved cells and enhanced in the extracts of insulin stimulation after serum starvation. Actin was used as a loading control. The amounts of pFoxO1 ^T24 relative to total FoxO1 protein and the amounts of pAkt^S473 relative to total Akt protein were calculated relative to those in the cell extracts from control siRNA-transfected cells.

Fig 8. Novel regulation of actin remodeling by Mdm20 through the mTORC2 pathway.

Mdm20 (magenta) is required for actin (green) cable stabilization through the N-acetylation of Trpm (light blue), which is mediated by NaB activity and involved in vacuole inheritance through actin-Trpm fibers. However, the present data show that Mdm20 deficiency suppresses mTORC2 activity by reducing Rictor (orange) expression, which may be regulated during de novo protein synthesis by the Nat5 (blue)-independent pathway. Because mTORC2 activates PKCα (light purple) and regulates actin (green) organization and cellular motility, these data indicate that Mdm20 modulates actin remodeling and cellular motility through the mTORC2 pathway. Additionally, Mdm20 regulates not only pAkt (pink), but also pFoxO1 (dark purple), through its effects on mTORC2 activity in response to serum starvation and insulin stimulation. Our findings show that Mdm20 acts as novel regulator of mTORC2 via its ability to regulate Rictor expression and is independent of NatB.