IBMPFD Disease-Causing Mutant VCP/p97 Proteins Are Targets of Autophagic-Lysosomal Degradation

Abstract

The ubiquitin-proteasome system (UPS) degrades soluble proteins and small aggregates, whereas macroautophagy (autophagy herein) eliminates larger protein aggregates, tangles and even whole organelles in a lysosome-dependent manner. VCP/p97 was implicated in both pathways. VCP/p97 mutations cause a rare multisystem disease called IBMPFD (Inclusion Body Myopathy with Paget's Disease and Frontotemporal Dementia). Here, we studied the role IBMPFD-related mutants of VCP/p97 in autophagy. In contrast with the wild-type VCP/p97 protein or R155C or R191Q mutants, the P137L mutant was aggregate-prone. We showed that, unlike commonly studied R155C or R191Q mutants, the P137L mutant protein stimulated both autophagosome and autolysosome formation. Moreover, P137L mutant protein itself was a substrate of autophagy. Starvation- and mTOR inhibition-induced autophagy led to the degradation of the P137L mutant protein, while preserving the wild-type and functional VCP/p97. Strikingly, similar to the P137L mutant, other IBMPFD-related VCP/p97 mutants, namely R93C and G157R mutants induced autophagosome and autolysosome formation; and G157R mutant formed aggregates that could be cleared by autophagy. Therefore, cellular phenotypes caused by P137L mutant expression were not isolated observations, and some other IBMPFD disease-related VCP/p97 mutations could lead to similar outcomes. Our results indicate that cellular mechanisms leading to IBMPFD disease may be various, and underline the importance of studying different disease-associated mutations in order to better understand human pathologies and tailor mutation-specific treatment strategies.

Fig 1. VCP/p97 mutant P137L induced autophagy.

(A) Primary structure of VCP/p97 protein and IBMPFD-related mutations. Major protein domains were shown. Bold, P137 mutant; underlined, R155, R191 and A232 mutants. (B) HEK293T cells were transfected with empty vector (CNT), MYC-tagged wild type (WT) or mutant VCP/p97 (P137L) and western blot analysis was performed using MYC, LC3 and ACTIN antibodies. ACTIN was used as loading control. LC3-I and LC3-II band intensity was quantified on scans of immunoblot films using Image J software. LC3-II/LC3-I ratios were determined after subtraction of non-band containing area intensities, and normalized to CNT condition. (C and D) WT or P137L mutant VCP/p97 were co-transfected with GFP-tagged LC3 in HEK293T cells and dot formation was visualized (C) and quantified (D). Data were shown as mean ± SD of independent experiments (n = 3) *** p<0.01.

Fig 2. VCP/p97 P137L mutant induced both autophagosome and autolysosome formation in HEK293T cells.

(A) Cells expressing empty vector (CNT), wild type (WT) or mutant VCP/p97 (P137L) were co-transfected with RFP-GFP-LC3 fusion construct and dot formation was visualized under fluorescent microscope. Arrow heads indicate autophagosomes and arrows indicate autolysosomes. (B) The graph shows the number of autophagosomes and autolysosomes in WT and P137L mutant expressing cells. Data were shown as mean ± SD of independent experiments (n = 3). **, p<0.05; ***, p<0.01 (C) Cells were transfected with CNT, and MYC-tagged WT or P137L mutant in the absence or presence of lysosomal inhibitors E64d and Pepstatin A (PepA) and western blot analysis was performed. ACTIN was used as loading control. ImageJ software was used for the quantification of band intensities.

Fig 3. VCP/p97 P137L mutant induced both autophagosome and autolysosome formation in U2OS and PC-12 cells.

(A and C) Cells expressing wild type (WT), mutant(P137L or R155C) VCP/p97 were co-transfected with RFP-GFP-LC3 fusion construct and dot formation was visualized in U2OS (A) and PC-12 (C) cells. Arrow heads indicate autophagosomes whereas arrows indicate autolysosomes. (A and C) The graphs show the number of autophagosomes and autolysosomes in U2OS (B) and PC-12 (D) cells. Data were shown as mean ± SD of independent experiments (n = 3). * p<0.05, ** p<0.01, *** p<0.001.

Fig 4. Mutant VCP/p97 colocalized with the autophagic vesicle marker LC3 during starvation.

HEK293T (A-C), U2OS (D-F) and PC-12 (G-I) cells expressing wild type (WT) or mutant VCP/p97 (P137L) were co-transfected with GFP-LC3 and colocalization was analyzed by confocal microscopy under basal/non-starved (NON-STV) (A, D, G) or starved (STV) (B, E, H) conditions. Overexpressed MYC-VCP was indirectly immunostained using anti-MYC antibodies. Arrowheads indicate colocalized dots in (A,B),(D,E) and (G,H). Quantification of co-localized dots in HEK293T (C), U2OS (F) and PC-12 (I) cells. Minimum 40 cells were counted under each condition, and quantification was expressed as a percentage of colocalization positive cells within a total transfected cell population. Data were shown as mean ± SD of independent experiments (n = 3). * p<0.05, ** p<0.01.

Fig 5. Mutant VCP/p97 colocalized with the lysosome/autolysosome marker LAMP1 during starvation.

HEK293T (A-C), U2OS (D-F) and PC-12 (G-I) cells expressing wild type (WT) or mutant VCP/p97 (P137L) were co-transfected with LAMP1-RFP and colocalization was analyzed by confocal microscopy under basal/non-starved (NON-STV) (A, D, G) or starved (STV) (B, E, H) conditions. Overexpressed MYC-VCP was indirectly immunostained using anti-MYC antibodies. Arrowheads indicate colocalized dots in (A,B),(D,E) and (G,H). Quantification of co-localized dots in HEK293T (C), U2OS (F) and PC-12 (I) cells. Minimum 40 cells were analyzed under each condition, and quantification was expressed as a percentage of colocalized positive cells within the total transfected cell population. Data were shown as mean ± SD of independent experiments (n = 3). * p<0.05, ** p<0.01.

Fig 6. P137L mutant VCP/p97, but not the WT VCP/p97 or R155C and R191Q mutants, was a target of starvation-induced autophagy.

(A) HEK293T cells were transfected with empty vector (CNT), MYC-tagged wild type (WT) or mutant VCP/p97 (P137L) and then starved for indicated time points. (B and C) HEK293T cells and (D) U2OS cells expressing empty vector (CNT), wild type (WT) or mutant (P137L) VCP/p97 were starved for 40 min in the absence or presence of lysosomal inhibitors E64d/Pepstatin A (E64D/PEPA) or chloroquine (CQ). (E) HEK293T cells and (F) U2OS cells expressing other VCP/p97 mutants (R155C) and (R191Q) were starved for 40 min in the absence or presence of lysosomal inhibitor chloroquine (CQ). (G) U2OS cells expressing wild type (WT) or mutant VCP/p97 (P137L) and (R155C) were starved for 40 min in the absence or presence of lysosomal inhibitor chloroquine (CQ). (H) U2OS cells were treated with carrier (DMSO) or Torin-1 (TRN) and western blot analysis was performed by using LC3 antibody. (I) Cells were transfected with wild type (WT) and mutant (P137L or R155C) VCP/p97 and treated with TRN in the presence or absence of CQ. ACTIN was used as loading control. Image J software was used for the quantification of band intensities. Western blot analysis was performed by using MYC and ACTIN antibodies. ACTIN was used as loading control. Image J software was used for the quantification of band intensities.

Fig 7. Another IBMPFD disease-associated mutant, the G157R mutant, was an autophagy target.

(A-C) WT or mutant VCP/p97 constructs were co-transfected to HEK293T cells with GFP-tagged LC3 and autophagosomal dot formation was visualized (A) and quantified (B). Data were shown as mean ± SD of independent experiments (n = 3) ** p<0.01, *** p<0.001. (C) Example of an immunoblot showing the expression of WT or mutant VCP/p97 proteins in A and B. (D) HEK293T cells expressing VCP/p97 R93C or G157R mutants were co-transfected with RFP-GFP-LC3 fusion construct and dot formation was visualized under fluorescent microscope. Arrow heads indicate autophagosomes (yellow) and arrows indicate autolysosomes (red). (E) Quantification of autophagosome (APG) and autolysosome (ALY) numbers in cells expressing R93C or G157R mutants. Data were shown as mean ± SD of independent experiments (n = 3). *, p<0.05; ***, p<0.01 (F) Cells were transfected with MYC-tagged R93C or G157R mutant VCP/p97 constructs and starved (STV) or not in the absence or presence of the lysosomal inhibitor chloroquine (CQ, 10 μM, 40 min) and immunoblot analysis was performed. ACTIN was used as loading control. For LC3-II/ACTIN ratios band intensity ratios quantifications with the ImageJ software was used. (G and H) HEK293T expressing WT or mutant (R93C and G157R) VCP/p97 were fractionated, and subcellular distribution of soluble (sol.) and insoluble (insol.) proteins were analyzed by immunoblotting.

Fig 8. Model of P137 mutant vs WT VCP/p97 degradation.

Following autophagy stimulation by starvation or Torin-1, aggregates containing VCP/p97 mutant P137L were sequestered in autophagosomes and delivered to autolysosomes for degradation. A soluble fraction of the mutant protein could interact with WT VCP/p97protein and form hetero-hexamers interfering with the function of the WT protein. On the other hand, a major fraction of the WT VCP/p97was not colocalizing with autophagosomes and autolysosomes, and it was not significantly degraded by the autolysosomal activity. Therefore, autophagy activation led to a selective degradation of the mutant protein, while the WT counterpart was spared.