A UBH-UBX module amplifies p97/VCP's unfolding power to facilitate protein extraction and degradation

Abstract

The p97-UFD1L-NPLOC4 ATPase unfolds numerous proteins for proteasomal degradation, but whether it suffices to pull proteins out of lipid bilayer remains unclear. Here, we identify a conserved ubiquitin-binding helix (UBH) in many UBX-containing p97 adapters, including FAF2, across yeast, plants, and metazoans. The UBH-UBX substantially facilitates the engagement of ubiquitinated substrates with p97-UFD1L-NPLOC4, and enhances p97 motor's working ATPase and unfolding activities by approximately twofold. Using purified p97-UFD1L-NPLOC4-FAF2^UBH-UBX, we reconstitute membrane protein extraction from the ER and mitochondria, establishing p97-UFD1L-NPLOC4-FAF2 (p97-UNF) as a power-enhanced unfoldase. Deleting UBH or disrupting UBH-ubiquitin interaction impairs substrate targeting, reduces p97-UNF's working ATPase and unfolding activities, and abolishes membrane protein extraction and degradation. We propose that UBH-UBX module amplifies p97's mechanical output power, enabling the removal of challenging substrates from large assemblies and ensuring rapid responses to protein misfolding or regulatory signals in diverse physiological processes.

Fig. 1. Identification of an α-helical domain in FAF2 necessary for protein extraction and degradation from the ER.

A Schematic representation of factors involved in p97-mediated CD4 extraction. Inhibitors and their specific targets are indicated. The ubiquitin is indicated in gray. B Analysis of p97-mediated CD4 extraction from the ER in cells with or without FAF2. Cells expressing CD4 and Vpu were treated with 20 μM CB5083, 25 μM MG132, or both for 3 h (or DMSO as a control), and analyzed by immunoblotting. Arrowheads denote p97-extracted, deglycosylated CD4 species. C A cell fractionation experiment to analyze CD4 extraction from the ER into the cytosol. Wild-type and FAF2^KO cells expressing Vpu, CD4 and FLAG-tagged FAF2 or empty vector were treated with or without MG132 and semi-permeabilized by digitonin. The cytosolic (cyto) and membrane (mem) fractions were analyzed by immunoblotting. Arrowheads denote deglycosylated, extracted CD4 in the cytosol. D Analysis of p97-mediated TCRα extraction in WT and FAF2^KO cells. Cells expressing TCRα-AsnTag (with an AsnTag inserted into the ER luminal domain) were either treated with MG132 or co-treated with MG132 and CB5083, and analyzed by immunoblotting. Arrowheads denote extracted and deglycosylated TCRα, which is deamidated and detected by the α-AspTag antibody. E Domain organization (top) and AlphaFold-predicted structural model (bottom) of FAF2. F Analyzing CD4 stabilization in wild-type and FAF2^KO cells expressing WT FAF2 or FAF2 variants lacking specific domains. FAF2^KO cells expressing Vpu and CD4 were transfected with plasmids encoding WT FAF2 or FAF2 variants missing indicated domains and analyzed by immunoblotting. Arrowheads denote stabilized CD4. (G) Quantification of CD4 stabilization from lane 3 to 10 in (F). The intensities of CD4 are normalized to those in lane 4 and the error bars represents the ± SEM of 4 independent experiments. (H) Analysis of CD4 stabilization and extraction in wild-type and FAF2^KO cells expressing wild-type FAF2 or FAF2 truncations lacking UBA (ΔUBA), membrane anchor (ΔM), UAS (ΔUAS), or UBX (ΔUBX). Cells expressing Vpu and CD4 were transfected with wild-type FAF2 or indicated truncation mutants, treated with or without MG132, and analyzed by immunoblotting. The arrowheads denote deglycosylated, extracted CD4. I As in C except that FAF2^KO cells expressing FAF2 or FAF2 lacking HD or UBX domains were analyzed. The arrowhead denotes extracted CD4 species in the cytosol.

Fig. 2. The HD serves as a UBH conserved in UBX-containing p97 cofactors across eukaryotes.

A Diagram of the experiment for analyzing the interaction between HD and ubiquitin chains. The purified 14xHis-2xStrepII-tagged HD is analyzed by Coomassie blue staining (left), immobilized on the Strep-Tactin sepharose, and incubated with monoubiquitin (6xHis-TEV-tagged) and unanchored ubiquitin chains for Strep-Tactin pulldown assay. B Input and HD-bound materials from (A) were analyzed by immunoblotting with indicated antibodies. C Quantification of ubiquitin chains bound to FAF2’s HD. The intensities of ubiquitin chains in the elution fractions were quantified and normalized to those in the input. Dots represent individual data points from three independent experiments. Data are presented as mean values ± SEM. D Sequence alignment of the HDs of selected UBX-containing proteins from indicated species. The amino acid positions of human FAF2’s HD and two putative ubiquitin-binding sites (UBSs) are indicated on the top. E A TALON pulldown experiment to analyze interaction between ubiquitin chains (non-tagged) and His-SUMO-tagged HDs of indicated UBX-containing proteins from various species. Samples were analyzed by immunoblotting and Ponceau S staining. F Diagram of FAF1, FAF1 inserted with a hydrophobic membrane anchor (FAF1^MA) or FAF1^MA truncations lacking UBA (ΔUBA), UBL (ΔUBL), or both (ΔUBAΔUBL). G Rescuing experiment analyzing the effect of FAF1 with a hydrophobic membrane anchor (FAF1^MA) on CD4 extraction. FAF2^KO cells expressing Vpu and CD4 were transfected with empty vector, FAF2, FAF1, or FAF1^MA plasmid, and analyzed by immunoblotting. Arrowheads denote extracted, deglycosylated CD4 species. The asterisk indicates a minor CD4 population likely caused by failure in ER translocation. H Analysis of CD4 stabilization and extraction in FAF2^KO expressing FAF1, FAF1^MA and FAF1^MA truncations lacking UBA (ΔUBA), UBL (ΔUBL), or both (ΔUBAΔUBL). Cells were treated with or without MG132 before analysis by immunoblotting. The arrowheads indicate extracted species. The asterisk indicates a minor non-translocated CD4 species.

Fig. 3. UBH-ubiquitin binding is necessary for extraction and degradation of diverse membrane proteins from the ER.

A AlphaFold-predicted structural model of FAF2’s HD in complex with three ubiquitin molecules. The putative UBS1 and UBS2, and the hydrophobic patch L8-I44-V70 of ubiquitin are indicated. B Purified HD and indicated UBS mutants (2xStrepII-tagged) were immobilized on the Strep-Tactin beads and incubated with monoubiquitin and unanchored ubiquitin chains (6xHis-TEV-tagged). Input and HD-bound materials were analyzed by immunoblotting and Ponceau S staining. C Quantification of band intensities of ubiquitin chains bound to HD and UBS mutants in (B). D As in B except that ubiquitinated sfGFP was analyzed. E Schematic representation of the experimental flow (left) and steps involved in p97-mediated extraction of ERAD substrates (right). The Asn (N) linked to the glycan is deamidated to Asp (D) after extraction. F Analysis of CD4 extraction in FAF2^KO cells expressing FAF2, FAF2^{L289N/Y297H/L301N}, FAF2^Y297H/L301N, FAF2^Y297H and FAF2^L301N. Blue and red arrowheads denote CD4 population resided in the ER and extracted into the cytosol, respectively. The asterisk indicates a minor non-translocated CD4 population, which is FAF2-independent. G Quantification of extracted CD4 in (F) (n = 3 independent experiments). Data are presented as mean values ± SEM. H As in F except that FAF2, FAF2^L289A, FAF2^L289N, FAF2^Y297A/L301A and FAF2^Y297H/L301N were analyzed. I Quantification of extracted, deglycosylated CD4 in (H) (n = 3 independent experiments). Data are presented as mean values ± SEM. J Immunoblotting analysis of TCRα-AsnTag extraction in WT and FAF2^KO cells expressing vector, wild-type FAF2 or its UBS mutants. The red arrowheads denote deglycosylated, deamidated, p97-extracted species. K Quantification of extracted, deglycosylated TCRα in (J) (n = 3 independent experiments). Data are presented as mean values ± SEM. (L) ABCG2-F208S-AsnTag was analyzed by immunoblotting in FAF2^KO cells expressing vector, wild-type FAF2 or its UBS mutants. The red arrowheads denote deglycosylated, deamidated, p97-extracted species. M Quantification of extracted ABCG2-F208S in (L) (n = 3 independent experiments). Data are presented as mean values ± SEM.

Fig. 4. UBH-ubiquitin association enhances targeting of ubiquitinated proteins to p97-UFD1L-NPLOC4.

A A GST pulldown assay to analyze interaction between FAF2’s UBX domain and p97-UN. GST-tagged UBX was incubated with purified UN, p97, or p97-UN. Bound materials were analyzed by Coomassie staining. B A co-IP experiment to analyze interactions between p97-UN and FAF2 or its UBH mutant Y297H/L301N. The diagram of experiment is shown on the left. The FLAG-tagged FAF2 and its mutant lacking UBA and MA domains (ΔUBAΔM) were immobilized on the α-FLAG agarose and incubated with p97-UN. Bound materials were analyzed by Coomassie staining and immunoblotting for UFD1L. C Diagram of the experiment for analyzing the recruitment of ubiquitinated sfGFP to p97 alone, or p97 complexed with UFD1L-NPLOC4 (UN) and/or FAF2 lacking UBA and MA domains (F^ΔUBAΔM). D The Myc-tagged p97 complexed with F^ΔUBAΔM, UN, or UNF^ΔUBAΔM was immobilized on the α-Myc beads and incubated with ubiquitinated sfGFP. The input and bead-bound materials were analyzed by either immunoblotting for sfGFP, UFD1L, NPLOC4 and FAF2, or Ponceau S staining for p97. E As in C, D, with the UBH mutant Y297H/L301N (F^{ΔUBAΔM/Y297H/L301N}) analyzed. F A co-IP experiment to analyze interaction between CD4 and p97 in FAF2^KO cells expressing either WT FAF2 or FAF2 variants lacking indicated domains. Samples were analyzed by immunoblotting either directly or after α-FLAG IPs for recovery of FLAG-tagged CD4. WT cells expressing Vpu and CD4 were used as a positive control. G Quantification of CD4-bound, endogenous p97 in (F). The error bars represent the ± SEM of three independent experiments.

Fig. 5. Conserved UBH-UBX modules enhance p97-UN’s working ATPase activity and accelerate substrate unfolding in vitro.

A Time course of NADH consumption to analyze ATPase activities of FAF2, p97 and p97 complexed with UN, UNF^ΔUBAΔM, or UNF^{ΔUBAΔM/Y297H/L301N} in the presence or absence of polyubiquitin chains. The concentration of p97 hexamer in each reaction was 150 nM. B ATPase activities (V₀/[p97 hexamer]₀) of p97 and its complexes in (A) (mean ± SEM, n = 6 independent experiments). ATPase activity was expressed as the maximum reaction rate per p97 hexamer. C As in (A), except that the purified UBH-UBX domain of FAF1 or FAF2 and their mutants (L289N/Y297H/L301N, V507N/Y515H/L519N) were used for analysis. The concentration of p97 hexamer in each reaction was 50 nM. D ATPase activities (V₀/[p97 hexamer]₀) of p97 and its complexes in (C) (mean ± SEM, n = 3 independent experiments). E ATPase activities of p97-UN (empty circle), p97-UN-FAF1^UBH-UBX (filled triangle) and p97-UN-FAF2^UBH-UBX (filled circle) measured at various concentration of ATP in the presence of 70 μg/mL polyubiquitin chains (mean ± SEM, n = 3 independent experiments). Data were fitted to the Hill equation. Fitting was performed over the range of 0-3 mM ATP for p97-UN, and 0-6 mM ATP for p97-UN-FAF1^UBH-UBX and p97-UN-FAF2^UBH-UBX. F Apparent kinetic parameters derived from Hill equation fits of the ATPase velocity data shown in (E) (mean ± SEM, n = 3 independent experiments). G Unfolding of polyubiquitinated mEOS3.2 by p97 alone or p97 complexed with UN, UNF^ΔM, UNF^ΔUBAΔM or UNF^UBH-UBX. The fluorescent intensity of mEOS3.2 (green) was monitored over 32 minutes after adding ATP (reactions without ATP served as controls). AU stands for arbitrary units. H Relative rates of mEOS3.2 unfolding in (G) (mean ± SEM, n = 3 independent experiments). Initial velocities were calculated by fitting a linear regression to the curve from 3 to 7 minutes in (G). Statistical significance was calculated using one-way ANOVA. ns, not significant. AU stands for arbitrary units. (I) As in (G), except that the purified UBH-UBX domain of FAF2 or FAF1 and their mutants (L289NY297H/L301N, V507N/Y515H/L519N) were used for the analysis of unfolding of polyubiquitinated mEOS3.2 (green). AU stands for arbitrary units. (J) Relative rates of mEOS3.2 unfolding in (I) (mean ± SEM, n = 3 independent experiments). Initial velocities were calculated by fitting a linear regression to the curve from 3 to 7 minutes in (I). Statistical significance was determined by one-way ANOVA. ***, p < 0.001. AU stands for arbitrary units.

Fig. 6. UBH-UBX enables membrane protein extraction from the ER by p97-UNF in vitro.

A Schematic representation of the in vitro reconstituted CD4 extraction assay. The microsomes containing mCD4 (FLAG-tagged) and Vpu were isolated from FAF2^KO cells and incubate with SCF^βTrCP for ubiquitination of mCD4. The microsomes were then sedimented, washed, resuspended and incubated with p97-UN, p97-UNF, or p97-UNF^{ΔM/Y297H/L301N}, plus GroEL-D87K which was used to capture unfolded mCD4 and ubiquitin, thereby preventing their aggregation. The reaction mixture was subjected to centrifugation to separate the microsomal pellet and the supernatant. The extracted mCD4 in the supernatant was recovered by α-FLAG IPs. B The incubated total reaction mixture described in (A) were analyzed by Coomassie staining for indicated proteins, and by immunoblotting for BAP31 (as a microsomal control). C The extracted mCD4 recovered by α-FLAG IP was analyzed by immunoblotting. Arrowhead indicates extracted CD4 modified with three ubiquitin molecules. D As in C except that mCD4 containing a single lysine at position 453 was analyzed. Arrowhead indicates mCD4 with a single ubiquitin chain of 3 molecules. E The extracted mCD4-K453 recovered by α-FLAG IP was analyzed by immunoblotting. The purified UBH-UBX domain of FAF2 was used with p97-UN. Arrowhead indicates extracted mCD4-K453 modified with a single ubiquitin chain of three molecules.

Fig. 7. UBH-ubiquitin interaction regulates pexophagy and outer mitochondria membrane associated degradation.

A Schematic representation of the RFP-GFP-SKL biosensor to measure pexophagy flux. B FACS-based analysis of pexophagy in wild-type, FAF2^KO and FAF2^KO cells stably expressing FAF2 or FAF2^{L289N/Y297H/L301N}. All cells stably expressed RFP-GFP-SKL, and pexophagy-positive clusters are indicated. C Quantitative analysis of the pexophagy-positive clusters in (B) (n = 3 independent experiments). Data was presented as mean values  ±  SEM. Statistical significance was calculated using one-way ANOVA. ***, p < 0.001; ****, p < 0.0001. D Immunoblotting analysis of endogenous ABCD3 in wild-type cells and FAF2^KO cells stably expressing empty vector, FAF2 or FAF2^{L289N/Y297H/L301N}. E Quantification of ABCD3 in indicated cells in (D). The error bars represent the ± SEM of three independent experiments. Statistical significance was calculated using one-way ANOVA. ***, p < 0.001; ****, p < 0.0001. F A cycloheximide (CHX) chase experiment to analyze MCL1 turnover in FAF2^KO cells expressing wild-type FAF2 or FAF2^{L289N/Y297H/L301N}. G Quantification of results in (F). The error bars stand for ± SEM from three independent experiments. H Immunoblotting analysis of Mfn1 in FAF2^KO cells stably expressing empty vector, wild-type FAF2, FAF2^ΔUBA or FAF2^{L289N/Y297H/L301N}. I Quantification of results in (H). The error bars stand for ± SEM from three independent experiments. Statistical significance was calculated using two-tailed Student t test. **, p < 0.01. J Working model of p97-UNF-mediated protein extraction from the ER, OMM, peroxisome and protein complexes.