Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jul 3;42(13):e113033.
doi: 10.15252/embj.2022113033. Epub 2023 Mar 10.

A mitochondrial SCF-FBXL4 ubiquitin E3 ligase complex degrades BNIP3 and NIX to restrain mitophagy and prevent mitochondrial disease

Yu Cao  1   2   3 , Jing Zheng  2   3 , Huayun Wan  2   3 , Yuqiu Sun  2   3   4 , Song Fu  2   3   5 , Shanshan Liu  2   3 , Baiyu He  2   3   6 , Gaihong Cai  2 , Yang Cao  2 , Huanwei Huang  2 , Qi Li  2   4 , Yan Ma  2   4 , She Chen  2   4 , Fengchao Wang  2   4 , Hui Jiang  1   2   3   4
Affiliations

A mitochondrial SCF-FBXL4 ubiquitin E3 ligase complex degrades BNIP3 and NIX to restrain mitophagy and prevent mitochondrial disease

Yu Cao et al. EMBO J. .

Abstract

Mitophagy is a fundamental quality control mechanism of mitochondria. Its regulatory mechanisms and pathological implications remain poorly understood. Here, via a mitochondria-targeted genetic screen, we found that knockout (KO) of FBXL4, a mitochondrial disease gene, hyperactivates mitophagy at basal conditions. Subsequent counter screen revealed that FBXL4-KO hyperactivates mitophagy via two mitophagy receptors BNIP3 and NIX. We determined that FBXL4 functions as an integral outer-membrane protein that forms an SCF-FBXL4 ubiquitin E3 ligase complex. SCF-FBXL4 ubiquitinates BNIP3 and NIX to target them for degradation. Pathogenic FBXL4 mutations disrupt SCF-FBXL4 assembly and impair substrate degradation. Fbxl4-/- mice exhibit elevated BNIP3 and NIX proteins, hyperactive mitophagy, and perinatal lethality. Importantly, knockout of either Bnip3 or Nix rescues metabolic derangements and viability of the Fbxl4-/- mice. Together, beyond identifying SCF-FBXL4 as a novel mitochondrial ubiquitin E3 ligase restraining basal mitophagy, our results reveal hyperactivated mitophagy as a cause of mitochondrial disease and suggest therapeutic strategies.

Keywords: BNIP3/NIX; FBXL4; mitochondrial disease; mitophagy; ubiquitin-proteasome pathway.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Figure 1
Figure 1. Mitochondria‐targeted CRISPR‐Cas9 screen for mitophagy regulators

  1. Schematic of the mitoQC reporter.

  2. Schematic of the screen process of mitophagy regulators.

  3. Volcano plot of the screen result. The top hits are highlighted in red.

  4. Verification of screen hits by mtKeima‐based FACS analysis of mitophagy. Upper: representative FACS analysis of mitophagy in HeLa cells expressing sgNTC and sgFBXL4. Bottom: quantitative analysis of mitophagy levels in HeLa cells expressing the indicated sgRNAs. The percentage of cells with high mitophagy level is indicated in red. Two independent sgRNAs were used for each gene. NTC: nontargeting control.

  5. Imaging analysis of mitophagy levels in the indicated HeLa cells. Left: sequencing result of WT and FBXL4‐KO cells and representative mitophagy images, white arrows point to mitolysosomes (FBXL4‐KO cells were not labeled because of the large numbers of mitolysosomes); middle: percentage of cells with mitophagy (mitolysosome‐positive); right: quantification of mitolysosome/mitochondria area. n: number of cells analyzed (middle); number of imaging areas (20–30 cells/area) analyzed (right).

  6. FACS analysis of mitophagy levels in the indicated HeLa cells. Left: representative FACS results; right: quantitative analysis of mitophagy levels.

Data information: Data are mean + SD from three biological replicates (D–F). Statistics: two‐tailed unpaired Student's t‐test (D–F); ***P < 0.001. Source data are available online for this figure.
Figure EV1
Figure EV1. FBXL4 deficiency hyperactivates mitophagy. Related to Fig 1
  1. A

    Verification of screen hits by mitoQC‐based FACS analysis of mitophagy. Left: representative FACS analysis; right: quantitative analysis of mitophagy levels in HeLa cells expressing the indicated sgRNAs. Two independent sgRNAs were used for each gene. NTC: nontargeting control.

  2. B

    Schematic of the pH‐dependent dual excitation of mtKeima.

  3. C

    Live cell imaging analysis of mitolysosomes. WT and FBXL4‐KO HeLa cells expressing mtKeima and LAMP1‐YFP were subject to live cell confocal imaging.

  4. D

    Immunoblot analysis of the indicated HeLa cells. FBXL4‐KO HeLa cells were infected with lentiviruses expressing sgNTC, sgBECLIN1, or sgFIP200.

  5. E

    FACS analysis of mitophagy levels in the indicated HeLa cells. The same cells in (D) were analyzed. Left: representative FACS analysis; right: quantitative analysis.

  6. F

    Schematic of a cleavage‐based mitophagy reporter. The linker between mCherry and mitochondrial targeting sequence is sensitive to lysosomal hydrolases.

  7. G

    Immunoblot analysis of reporter cleavage in the indicated HeLa cells. Left: representative immunoblot. Arrow points to the cleaved mCherry. Right: quantitative analysis of mCherry cleavage.

  8. H, I

    Immunoblot analysis of mitochondrial proteins in the indicated HeLa cells.

Data information: Data are mean + SD from three biological replicates (A, E, G). Statistics: two‐tailed unpaired Student's t‐test (E, G); ***P < 0.001. Source data are available online for this figure.
Figure 2
Figure 2. FBXL4 deficiency post‐transcriptionally upregulates BNIP3 and NIX to hyperactivate mitophagy
  1. A

    Volcano plot of the screen result of mitophagy regulators in FBXL4‐KO HeLa cells. Mitochondrial outer‐membrane protein MTCH2 and mitophagy receptors BNIP3 and NIX are highlighted in red.

  2. B

    Immunoblot analysis of the indicated HeLa cells. WT and FBXL4‐KO HeLa cells were rescued with vector or FBXL4‐FLAG.

  3. C

    qPCR analysis of BNIP3 and NIX in the indicated HeLa cells. The same cells in (B) were analyzed.

  4. D

    Immunoblot analysis of the indicated HeLa cells. B: BNIP3; N: NIX. Two clones were shown for each double and triple knockout cell.

  5. E, F

    Representative live cell imaging (E) and quantitative analysis (F) of mitophagy levels in the indicated HeLa cells. n: number of cells analyzed (left); number of imaging areas analyzed (right).

  6. G

    Representative FACS analysis of mitophagy levels in the indicated HeLa cells.

  7. H

    FACS‐based quantitative analysis of mitophagy levels in the indicated HeLa cells.

Data information: Data are mean + SD from three biological replicates (C, F, H). Statistics: two‐tailed unpaired Student's t‐test (C, F, H); **P < 0.01; ***P < 0.001. Source data are available online for this figure.
Figure EV2
Figure EV2. MTCH2 regulates the protein levels of BNIP3 and NIX and controls mitophagy in FBXL4‐null cells. Related to Fig 2
  1. A, B

    Representative FACS analysis (A) and quantitative analysis (B) of mitophagy levels in the indicated HeLa cells. Two independent sgRNAs for MTCH2 were used. Data are mean + SD from three biological replicates. Statistics: two‐tailed unpaired Student's t‐test; ***P < 0.001.

  2. C

    Immunoblot analysis of the indicated HeLa cells. *Nonspecific bands.

Source data are available online for this figure.
Figure 3
Figure 3. FBXL4 is an integral mitochondrial outer‐membrane protein

  1. Determination of the submitochondrial localization of FBXL4 by Protease K and trypsin digestion. Mitochondria from HeLa FBXL4‐FLAG stable line were purified and stored as intact mitochondria or treated with hypotonic swelling buffer or lysed with Triton X‐100 buffer. Different mitochondrial preparations were then digested with Protease K or trypsin.

  2. Analysis of the association of FBXL4 with membrane by alkaline carbonate extraction. Purified mitochondria from HeLa FBXL4‐FLAG stable line were treated with alkaline carbonate buffer at the indicated pH and then centrifuged to collect the supernatant and the pellet fractions. S: supernatant; P: pellet.

  3. “DAS” prediction of the transmembrane domain (TMD) of FBXL4. The red box highlights the putative TMD at FBXL4 N terminus.

  4. Analysis of the TMD and the flanking sequences of FBXL4, Tom20, and Tom70. Positively charged residues are highlighted in red; negatively charged residues are highlighted in green. H.s.: Homosapien; R.n.: Rattus norvegicus.

  5. Identification of the mitochondrial targeting sequence of FBXL4. Indicated FBXL4 N‐terminal sequences were fused to RFP and expressed in HeLa cells to analyze their subcellular localization.

  6. Determination of the submitochondrial localization of FBXL4(1–23)‐RFP by Protease K digestion. Experiments were performed similarly as (A).

  7. Schematic of FBXL4 N‐anchored to mitochondrial outer membrane.

Source data are available online for this figure.
Figure 4
Figure 4. FBXL4 forms an SCF‐FBXL4 ubiquitin E3 ligase complex to ubiquitinate and degrade BNIP3 and NIX

  1. Alignment of F‐box sequences. Critical residues for Skp1 binding are highlighted in red.

  2. Immunoprecipitation analysis of the FBXL4‐Skp1‐Cullin1 (SCF‐FBXL4) complex. HeLa cells were infected with lentiviruses expressing the indicated genes. ∆F: the F‐box deletion mutant of FBXL4‐FLAG; 4A: FBXL4‐4A mutant as shown in (A).

  3. Immunoprecipitation analysis of the FBXL4‐UBXD8‐VCP complex. HeLa cells expressing vector or FBXL4‐FLAG were treated with DMSO or MG132 (20 μM) for 8 h.

  4. Schematic of the SCF‐FBXL4‐UBXD8‐VCP complex at mitochondrial outer membrane.

  5. Immunoprecipitation analysis of the association of FBXL4 with BNIP3 and NIX. FBXL4‐KO HeLa cells were rescued with GFP, FBXL4‐FLAG, or FBXL4(∆F)‐FLAG.

  6. Immunoprecipitation analysis of the ubiquitination of BNIP3. FLAG‐BNIP3 (knockin), FBXL4‐KO, and HA‐ubiquitin (Ub) HeLa cells were rescued with WT or ∆F FBXL4 and treated with DMSO or MG132 (20 μM) for 8 h.

  7. Immunoprecipitation analysis of the ubiquitination of NIX. FLAG‐NIX (knockin), FBXL4‐KO, and HA‐Ub HeLa cells were treated the same as (F).

  8. Immunoblot analysis of the indicated HeLa cells. FBXL4‐KO HeLa cells were rescued with vector, FBXL4‐FLAG, FBXL4(4A)‐FLAG, or FBXL4(∆F)‐FLAG.

Source data are available online for this figure.
Figure 5
Figure 5. Pathogenic mutations of FBXL4 disrupt SCF‐FBXL4 complex assembly and impair the degradation of BNIP3 and NIX to hyperactivate mitophagy

  1. Schematic of FBXL4 domain organization and predicted structure by AlphaFold2.

  2. Immunoblot analysis of the indicated HeLa cells. WT and FBXL4‐KO HeLa cells were rescued with vector, WT or mutant FBXL4‐FLAG.

  3. FACS‐based analysis of mitophagy levels in the indicated HeLa cells. The same HeLa cells in (B) were used. Left: representative FACS analysis; right: quantitative analysis of mitophagy levels. The dashed red line marks the mitophagy level in FBXL4‐KO cells rescued with FBXL4‐FLAG.

  4. Immunoprecipitation analysis of FBXL4‐substrate interaction. FBXL4‐KO HeLa cells were rescued with vector, FBXL4(∆F)‐FLAG or mutant FBXL4(∆F)‐FLAG.

  5. Immunoprecipitation analysis of FBXL4‐Skp1 interaction. FBXL4‐KO HeLa cells were rescued with vector, WT or mutant FBXL4‐FLAG.

  6. Analysis of the mass spectrometry results of immunoprecipitation with WT or mutant FBXL4‐FLAG. FBXL4‐KO HeLa cells expressing WT, V140A or I551N FBXL4‐FLAG were subject to DSP crosslinking and anti‐FLAG immunoprecipitation. The immunoprecipitates were analyzed by mass spectrometry. The Mascot score of immunoprecipitated proteins was plotted. Because of space limitation, six common proteins with Mascot score between 5,000 and 20,000 were removed. Cullin1, which has decreased interaction with both FBXL4 mutants, is highlighted in red.

  7. Immunoprecipitation analysis of the integrity of the SCF‐FBXL4 complex. Upper: FBXL4‐KO HeLa cells expressing vector, WT or mutant FBXL4‐FLAG were crosslinked with DSP and subject to anti‐FLAG immunoprecipitation; lower: quantification of the ratio of Cullin1/FLAG band intensities.

Data information: Data are mean + SD from three biological replicates (C, G). Statistics: two‐tailed unpaired Student's t‐test (C, G); *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant. Source data are available online for this figure.
Figure 6
Figure 6. Fbxl4 deficiency increases the protein levels of BNIP3 and NIX and hyperactivates mitophagy in vivo in mice

  1. Viability of the indicated mice. n: number of mice analyzed.

  2. Immunoblot analysis of organs from the indicated mice at P0 of age. Elevated BNIP3 and NIX levels are highlighted by red boxes; reduced mitochondrial proteins are highlighted by blue boxes. Three mice for each genotype were analyzed. Protein samples from HeLa cells treated with DMSO or Actinomycin D (ActD) were used as positive control for PARP cleavage.

  3. Schematic of AAV‐based gene knockout and mitophagy examination. Upper: schematic of the bicistronic AAV(mtKeima)‐sgRNA vector that simultaneously expresses mtKeima and sgRNA; Middle: schematic of delivering AAV9(mtKeima)‐sgNTC or AAV9(mtKeima)‐sgFBXL4 to the Cas9‐knockin mice to generate the control mice and hepatic Fbxl4‐KO mice; below: immunoblot analysis of the indicated mouse liver at 45 days after AAV delivery. Three mice for each genotype were analyzed. Data are mean + SD from three biological replicates. Statistics: two‐tailed unpaired Student's t‐test; *P < 0.05.

  4. Live cell confocal imaging of mitophagy level in the indicated mouse liver at 45 days after AAV delivery. Left: representative images. The dashed line indicates cell boundary. Right: quantitative analysis of mitophagy levels in hepatocytes. n: number of cells analyzed. Data are mean ± SD. Statistics: two‐tailed unpaired Student's t‐test; ***P < 0.001.

Source data are available online for this figure.
Figure 7
Figure 7. Knockout of either Bnip3 or Nix rescues the Fbxl4 −/− mice

  1. Viability of the indicated mice. n: number of mice analyzed.

  2. Body weight of the indicated male mice at P30. Data are mean ± SD. Statistics: one‐way ANOVA with the Tukey–Kramer test; *P < 0.05; ***P < 0.001.

  3. Immunoblot analysis of liver samples from the indicated mice at P0 of age. Left: immunoblot analysis, three mice for each genotype were analyzed; right: quantification of the ratio of mitochondrial protein/actin band intensities. Data are mean + SD from three biological replicates. Statistics: two‐tailed unpaired Student's t‐test; *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant.

  4. Metabolite analysis of liver samples from the indicated mice at P0 of age. Three to five mice for each genotype were analyzed.

Source data are available online for this figure.
Figure EV3
Figure EV3. Overexpression of DN‐Cullin1 causes the accumulation of BNIP3 and NIX to hyperactivate mitophagy. Related to Fig 4
  1. A

    Immunoblot analysis of the indicated HeLa cells. WT and FBXL4‐KO HeLa cells were infected with lentivirus expressing TO‐DN‐Cullin1. Cells were then treated with doxycycline (Dox, 2 μg/ml) for the indicated time.

  2. B

    qPCR analysis of the indicated HeLa cells. The same cells in (A) were used. Dox (2 μg/ml) was treated for 12 h.

  3. C, D

    Representative FACS analysis (C) and quantitative analysis (D) of mitophagy levels in the indicated HeLa cells. WT and FBXL4‐KO HeLa cells were infected with lentivirus expressing TO‐DN‐Cullin1. Cells were then treated with Dox (2 μg/ml) for 48 h.

Data information: Data are mean + SD from three biological replicates (B, D). Statistics: two‐tailed unpaired Student's t‐test (B, D); ns: not significant; ***P < 0.001. Source data are available online for this figure.
Figure EV4
Figure EV4. Pathogenic mutations of FBXL4 impair substrate ubiquitination. Related to Fig 5
Immunoprecipitation analysis of the ubiquitination of BNIP3 by FBXL4 pathogenic mutants. FLAG‐BNIP3 (knockin), FBXL4‐KO, and HA‐Ub HeLa cells were rescued with WT or mutant FBXL4 and treated with DMSO or MG132 (20 μM) for 8 h.Source data are available online for this figure.
Figure EV5
Figure EV5. Tissue‐specific contribution of Bnip3 and Nix to mitochondrial mass reduction in Fbxl4 −/− mice. Related to Fig 7
  1. A, B

    Immunoblot analysis of heart (A) and kidney (B) from the indicated mice at P0 of age. Three mice for each genotype were analyzed. Data are mean + SD from three biological replicates (A, B). Statistics: two‐tailed unpaired Student's t‐test; *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant.

Source data are available online for this figure.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Allen GFG, Toth R, James J, Ganley IG (2013) Loss of iron triggers PINK1/Parkin‐independent mitophagy. EMBO Rep 14: 1127–1135 - PMC - PubMed
    1. Alsina D, Lytovchenko O, Schab A, Atanassov I, Schober FA, Jiang M, Koolmeister C, Wedell A, Taylor RW, Wredenberg A et al (2020) FBXL4 deficiency increases mitochondrial removal by autophagy. EMBO Mol Med 12: e11659 - PMC - PubMed
    1. Andreux PA, Blanco‐Bose W, Ryu D, Burdet F, Ibberson M, Aebischer P, Auwerx J, Singh A, Rinsch C (2019) The mitophagy activator urolithin A is safe and induces a molecular signature of improved mitochondrial and cellular health in humans. Nat Metab 1: 595–603 - PubMed
    1. Bai C, Sen P, Hofmann K, Ma L, Goebl M, Harper JW, Elledge SJ (1996) SKP1 connects cell cycle regulators to the ubiquitin proteolysis machinery through a novel motif, the F‐box. Cell 86: 263–274 - PubMed
    1. Barøy T, Pedurupillay CRJ, Bliksrud YT, Rasmussen M, Holmgren A, Vigeland MD, Hughes T, Brink M, Rodenburg R, Nedregaard B et al (2016) A novel mutation in FBXL4 in a Norwegian child with encephalomyopathic mitochondrial DNA depletion syndrome 13. Eur J Med Genet 59: 342–346 - PubMed

Publication types

MeSH terms