PEX2 is the E3 ubiquitin ligase required for pexophagy during starvation

Abstract

Peroxisomes are metabolic organelles necessary for anabolic and catabolic lipid reactions whose numbers are highly dynamic based on the metabolic need of the cells. One mechanism to regulate peroxisome numbers is through an autophagic process called pexophagy. In mammalian cells, ubiquitination of peroxisomal membrane proteins signals pexophagy; however, the E3 ligase responsible for mediating ubiquitination is not known. Here, we report that the peroxisomal E3 ubiquitin ligase peroxin 2 (PEX2) is the causative agent for mammalian pexophagy. Expression of PEX2 leads to gross ubiquitination of peroxisomes and degradation of peroxisomes in an NBR1-dependent autophagic process. We identify PEX5 and PMP70 as substrates of PEX2 that are ubiquitinated during amino acid starvation. We also find that PEX2 expression is up-regulated during both amino acid starvation and rapamycin treatment, suggesting that the mTORC1 pathway regulates pexophagy by regulating PEX2 expression levels. Finally, we validate our findings in vivo using an animal model.

Figure 1.

Overexpression of the E3 ubiquitin ligase PEX2 decreases the number of peroxisomes. (A) HeLa cells were transfected with PMP34 or with one of the peroxisomal E3 ubiquitin ligases (PEX2, PEX10, and PEX12) fused to GFP. The cells were grown for 48 h, then fixed and stained for the peroxisomal membrane protein PMP70. Bars, 20 µm. (B) Box plot of the peroxisomal density of cells expressing the constructs in A. The peroxisomal density was calculated by quantifying the number of PMP70 puncta structures from Z-stack confocal images of the entire cell by and dividing by the cell volume. (C) Box plot of the total PMP70 fluorescence intensity from the entire cell. Both B and C are quantification of 150 cells from three independent trials. **, P < 0.01; ***, P < 0.001; ns (not significant), P > 0.05. In B and C, boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation.

Figure 2.

PEX2-mediated peroxisome loss requires the RING-finger domain. (A) Schematic of PEX2 and its deletion mutant constructs. RING, RING-finger domain; TM, transmembrane domain. Number above the image represents residue number. (B) HeLa cells were transfected with various PEX2-GFP constructs in A and fixed and stained for the peroxisomal membrane protein PMP70 after 48 h. Bars, 20 µm. (C) Box plots of the peroxisomal density of cells expressing the constructs in B. Plotted are a total of 150 cells from three independent trials. **, P < 0.01; ns (not significant), P > 0.05. Boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation.

Figure 3.

PEX2-mediated peroxisome loss is an autophagic process. (A) ATG5^+/+ and ATG5^−/− MEF cells were mock transfected, transfected with PMP34-GFP, or transfected with PEX2-GFP as indicated. 48 h later, the cells were fixed, stained, and imaged for the peroxisomal membrane protein PMP70. Bars, 20 µm. (B) Box plot of the peroxisomal density of cells in A. Plotted are a total of 150 cells from three independent trials. **, P < 0.01; ns, not significant. boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation. (C) Immunofluorescent image of HeLa cells transfected with PMP34-GFP or PEX2-GFP (green) and probed for endogenous NBR1 (red). Panels on the right show enlargement of the region of interest (white box). The channels are shown individually with the merge as indicated. (D) To detect autophagosomes, cells were cotransfected with mCherry-LC3. (E) To detect lysosomes, cells were cotransfected with Lamp1-GFP. Bars, 20 µm. The region of interest (white box) is magnified and each channel is shown separately along with the merged image on the right side in C–E. The pseudocolor used for each probe in the images is indicated by the color of the name in each panel.

Figure 4.

PEX2-mediated pexophagy requires NBR1. (A) Representative images of HeLa cells treated with nontargeting siRNA (siCtrl), or with siRNA against p62 or NBR1. 48 h after siRNA transfection, these cells were transfected with PEX2-FLAG, and 24 h afterward they were fixed and stained for the peroxisomal membrane protein PMP70 and with an antibody against the FLAG epitope. “mock” and “mock-transfected” are cells treated with transfection agent, but not with siRNA or plasmids, respectively. Bars, 20 µm. (B) Box plot of the PMP70 puncta density of cells in A. Plotted are a total of 50 cells from three independent trials. **, P < 0.01; ns (not significant), P > 0.05. Boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation.

Figure 5.

Peroxisomes are selectively degraded during amino acid starvation. (A) Immunoblot of HeLa cell lysates incubated in HBSS between 0 and 24 h as indicated. Shown are blots probed for the peroxisomal marker PMP70, the ribosomal marker PO, the mitochondrial marker TOMM20, and the cytosolic marker GAPDH. (B) Line graphs of the quantification of PMP70, PO, and TOMM20 blots relative to GAPDH and normalized against 0 h. Shown is the mean ± standard deviation from three independent experiments. (C) Immunofluorescent image of PMP70-stained HeLa cells grown in regular growth media (DMEM) or in HBSS media for 24 h. Bars, 20 µm. (D) Box plots of the peroxisomal density from 150 cells from three trials as shown in C. (E) HeLa cells were grown in either the growth media DMEM or the amino acid depletion media HBSS for time indicated and stained for PMP70 (red) and NBR1 (green). Panels on the right show enlargement of the region of interest (white box). The channels are shown individually with the merge as indicated. Bar, 20 µm. (F) Box plot of Manders’ coefficients showing the percentage of NBR1 that is colocalized with PMP70-positive puncta. Plotted are 150 cells from three trials. *, P < 0.05; ***, P < 0.001; ns (not significant), P > 0.05. Boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation.

Figure 6.

PEX2 is required for pexophagy during both amino acid starvation and basal turnover. (A) HeLa cells were transfected with nontargeting siRNA (siCtrl) or with siRNA against PEX2, PEX10, or PEX12 for 48 h, and then grown in either regular DMEM media or amino acid starved in HBSS media for an additional 16 h. Shown are the representative fluorescent images of cells immunostained for PMP70. Bars, 20 µm. (B) Box plot of the peroxisomal density of cells after amino acid depletion. Plotted are a total of 150 cells from three independent trials. Only those with P < 0.05 when compared with either siCtrl DMEM or siCtrl HBSS are indicated. *, P < 0.05; **, P < 0.01. Boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation.

Figure 7.

PEX5 and PMP70 are ubiquitinated by PEX2. (A) HEK293 cells stably expressing a HA-ubiquitin (HA-Ub) construct under the control of a tamoxifen-inducible promoter were treated with either DMSO or tamoxifen for 24 h. 4 h before lysis, cells were depleted of amino acids in HBSS media, either in the absence or presence of proteasomal inhibitor MG132 (10 µM) and the lysosomal inhibitor chloroquine (10 nM). Ubiquitinated proteins were immunoprecipitated with an antibody against the FLAG epitope and lysates (input) and immunoprecipitation (IP:HA) were blotted for PMP70, PEX5, catalase, and HA as indicated. (B) As in A, but before induction of HA-Ub, cells were mock treated or transfected with nontargeting siRNA (siCtrl) or siRNA against PEX2, PEX10, or PEX12. 4 h before lysis, cells were treated grown in HBSS media, either in the presence or absence of the proteasomal inhibitor MG132, and analyzed as in A. (C) Peroxisomes isolated from cells expressing PEX2-FLAG show an increase in ubiquitinated proteins. The HA-Ub HEK293 cell line was transfected with either an empty FLAG vector or PEX2-FLAG. 4 h before lysis, the cells were treated with the proteasomal inhibitor MG132 (10 µM). After treatment, cell lysates were collected (WCL), the total membrane (TM) fraction was separated (TM), and the TM fraction was further separated into light (lig), medium (med), heavy (hea), and pellet (pel) fractions. Each fraction was then analyzed by immunoblot using antibodies against HA, PMP70 (peroxisome membrane), Mfn2 (mitochondrial outer membrane), calnexin (ER membrane), and FLAG as indicated. (D) Bar graph showing the relative HA signal >48 kD from the total membrane fraction (TM) and the heavy fraction (heavy) lanes from three independent experiments. Means and standard deviations are shown. *, P < 0.05; ns (not significant), P > 0.05.

Figure 8.

PEX2 protein levels increase but mRNA is unchanged during amino acid starvation or mTORC1 inhibition. (A) HeLa cells were depleted of amino acids for 0–24 h as indicated in HBSS and analyzed by immunoblotting for PEX2, PMP70, and GAPDH. Line graphs show the amount of each protein relative to GAPDH and normalized to the start of HBSS treatment (0 h). n = 3. (B) HeLa cells were treated with 5 µM rapamycin for 0–24 h as indicated to inhibit mTOR and cell lysates were analyzed by immunoblotting for PEX2, PMP70, and GAPDH. Line graphs show the amount of each protein relative to GAPDH and normalized to 0 h. n = 3. (C) Immunofluorescent images of PMP70 in HeLa cells treated with 5 µM rapamycin for 0–48 h. Bars, 20 µm. (D) Box plot of the peroxisomal density of cells in (C). Plotted are a total of 150 cells from three independent trials. Boxes show the 25th, 50th, and 75th percentiles, and lines show one-way standard deviations. Points represent all cells that did not fall within one standard deviation. (E) Quantitative PCR of PEX2 mRNA in HeLa cells grown in DMEM, HBSS, or 5 µM rapamycin for 1 h. Also shown is the PEX2 mRNA level in cells treated with a nontargeting siRNA (siCtrl) or treated with two different siRNA constructs against PEX2. The means and standard deviations of three independent trials are shown. (F) Western blots of PEX2, PMP70, TOMM20, and GAPDH from primary rat hepatocytes after 0, 1, 4, or 6 h treatment in HBSS media. The line graphs represent the protein levels of PEX2, PMP70 and TOMM20 relative to GAPDH loading control. Each point represents the mean of three different trials with standard deviations shown. *, P < 0.05; **, P < 0.01.

Figure 9.

Peroxisomes are degraded and PEX2 protein levels elevated in a mouse model of amino acid starvation. (A) Mice were fed a control diet containing 18% protein (wt/wt) or a protein-restriction diet containing 1% protein (wt/wt). Representative images of fluorescent immunohistochemistry images thin slices of the livers stained with PMP70 (green) or DAPI (blue) as a marker of the nucleus. Images from two different mice for each diet conditions are shown. Bars, 20 µm. (B) Whole-liver lysates from three different mice for each condition were separated by SDS-PAGE and blotted for PEX2, catalase, PMP70, TOMM20, and GAPDH. (C) Immunoblot of liver lysate of 1% diet animals treated with chloroquine compared with control. (D) Bar graphs represent the relative expression of PMP70/GAPDH ratio in B and C normalized to the 18% diet control animals. Means and standard deviations are shown. *, P < 0.05; **, P < 0.01; ns (not significant), P > 0.05.

Figure 10.

A model of pexophagy activation during amino acid starvation. (A) In cells grown in normal growth media, mTORC1 maintains a low PEX2 expression level by promoting PEX2 degradation. A low level of PEX2 is maintained to allow for ubiquitination (Ub) of PEX5 for its recycling from the peroxisomal membrane. mTORC1 also inhibits the activation of autophagy. (B) During amino acid depletion conditions, mTORC1 is inactivated, releasing its inhibition of both autophagy and the expression levels of PEX2. Autophagosome formation increases, and PEX2 levels are elevated. The increase in PEX2 on peroxisomes results in increased ubiquitination of at least two peroxisomal proteins, PEX5 and PMP70, which allows for the recruitment of the autophagy receptor NBR1. NBR1 then targets these ubiquitinated peroxisomes to autophagosomes for degradation.