Skip to main content
PMC home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2007 May 11;104(21):8785–8790. doi: 10.1073/pnas.0702966104

Studies on peptide:N-glycanase–p97 interaction suggest that p97 phosphorylation modulates endoplasmic reticulum-associated degradation

Gang Zhao *,†, Xiaoke Zhou *,†, Liqun Wang *,†, Guangtao Li †, Hermann Schindelin *,†,‡,§, William J Lennarz †,§
PMCID: PMC1885580  PMID: 17496150

Abstract

During endoplasmic reticulum-associated degradation, the multifunctional AAA ATPase p97 is part of a protein degradation complex. p97 associates via its N-terminal domain with various cofactors to recruit ubiquitinated substrates. It also interacts with alternative substrate-processing cofactors, such as Ufd2, Ufd3, and peptide:N-glycanase (PNGase) in higher eukaryotes. These cofactors determine different fates of the substrates and they all bind outside of the N-terminal domain of p97. Here, we describe a cofactor-binding motif of p97 contained within the last 10 amino acid residues of the C terminus, which is both necessary and sufficient to mediate interactions of p97 with PNGase and Ufd3. The crystal structure of the N-terminal domain of PNGase in complex with this motif provides detailed insight into the interaction between p97 and its substrate-processing cofactors. Phosphorylation of p97's highly conserved penultimate tyrosine residue, which is the main phosphorylation site during T cell receptor stimulation, completely blocks binding of either PNGase or Ufd3 to p97. This observation suggests that phosphorylation of this residue modulates endoplasmic reticulum-associated protein degradation activity by discharging substrate-processing cofactors.

Keywords: PUB domain, Ufd3, endoplasmic reticulum-associated protein degradation

Endoplasmic reticulum-associated protein degradation (ERAD) is a major component of the quality control system in the protein secretory pathway. It is responsible for the removal of misfolded proteins and unassembled protein subunits (1, 2). Once (glyco-)proteins are recognized for degradation in the endoplasmic reticulum, they are routed to the cytosol via the putative retrotranslocon, polyubiquitinated, and degraded by the ubiquitin–proteasome system (3). It has been proposed that ERAD substrates are threaded mechanically through the retrotranslocon by the chaperone-like protein p97 (CDC48 in yeast) (4), which is also known as VCP (valosin-containing protein). p97 belongs to the AAA ATPase family of proteins (5), and its sequence is highly conserved in eukaryotes. RNAi and mutagenesis experiments have confirmed the essential role of p97 in the proteasome-mediated degradation of misfolded proteins (6). Structural studies (7) revealed that p97 has an N-terminal domain (N), two AAA ATPase domains (D1 and D2), and a short disordered C-terminal region (C). In addition to its role in ERAD, p97 has been found to participate in other cellular processes, including DNA repair, cell cycle control, and membrane trafficking (4, 8). Structural and biochemical studies have suggested that p97 fulfills its various functions by interacting with a wide spectrum of cofactors (9–12).

Most of its cofactors interact with the N-terminal domain of p97, with the exceptions of Ufd2, Ufd3, and peptide:N-glycanase (PNGase), which interact with the C-terminal region of p97 (9, 12, 13). The cofactors can be divided into substrate-recruiting cofactors, which regulate the substrate specificity of p97, and substrate-processing cofactors, which determine the fate of substrates (12). The substrate-recruiting cofactors include the Ufd1-Npl4 heterodimer and proteins containing the UBX (ubiquitin regulatory X) domain. Ufd1-Npl4 functions in ERAD (10), nuclear envelope formation (14), and spindle disassembly after mitosis (15). One of the best characterized UBX-containing proteins is p47. The complex formed between p47 and p97 mediates the reassembly of the Golgi membrane and nuclear envelope following cell division (16, 17). Compared with the extensively studied substrate-recruiting cofactors, the functions of the substrate-processing cofactors are less well defined. Recently, Rumpf and Jentsch (12) showed that, depending on whether p97 interacts with Ufd2 or with Ufd3, the p97 substrates are either polyubiquitinated and degraded, or deubiquitinated and stabilized.

Recently, we showed that p97 organizes a degradation complex by simultaneously interacting with AMFR (an endoplasmic reticulum membrane-associated E3 ubiquitin ligase), Y33K (a UBX domain-containing protein of unknown function), and cytoplasmic PNGase (13). PNGase is responsible for de-N-glycosylating misfolded glycoproteins before their degradation by the proteasome (13, 18). PNGase is present in all eukaryotes ranging from yeast to mammals. The core domain of PNGase consists of a transglutaminase-like domain, a zinc-binding domain, and a RAD4/XPC-like motif that tightly interacts with RAD23/HR23 (19, 20). In addition to the core catalytic domain, in higher eukaryotes PNGase has acquired two additional domains, one each at the N and C termini. Structural and biochemical studies on the C-terminal domain of mouse PNGase (mPNGase) have demonstrated that it binds oligo-mannose carbohydrates; this interaction enhances substrate binding and processing by mouse PNGase (21). The N-terminal domain of mPNGase harbors a PUB domain (PNGase/UBA or UBX-containing proteins), which was proposed to engage in protein–protein interactions (22). In fact, the PUB domain was recently shown to interact with p97. However, unlike most of the other cofactors, the PUB domain of mPNGase interacts with the C-terminal half (residues 459–806) of p97 (13).

In this report, we demonstrate that the PUB domain binds to a previously undetected protein–protein interaction module of p97, contained within its last 10 residues (p97-C10); this module also interacts with Ufd3 and possibly other cofactors. Phosphorylation of the penultimate tyrosine residue of p97 (Tyr-805 in mouse p97) abolishes p97's interaction with both PNGase and Ufd3. The crystal structure of the PUB domain in complex with p97-C10 reveals that the C terminus of p97 fits into a conserved and positively charged groove on the surface of the PUB domain. These results open up new possibilities to elucidate the roles of p97 in various cellular processes and to study proteins that possess the PUB domain, an evolutionarily new domain.

Results and Discussion

The C Terminus of p97 Harbors a Protein–Protein Interaction Motif.

Previously, we reported that the PUB domain of mouse PNGase interacts with the D2-C fragment (residues 459–806) of mouse p97 (13). To further narrow down the region of p97 that interacts with the PUB domain, various truncations of p97 were prepared and their interactions with PNGase were studied. As shown in supporting information (SI) Fig. 5, the last 40 residues of p97 (p97-C40) are both necessary and sufficient for binding to PNGase. Isothermal titration calorimetry (ITC) experiments showed that p97-C40 interacts with the PUB domain of PNGase with an apparent KD of 11.1 μM, an affinity similar to that observed with the full-length protein (Table 1). A notable feature of the C-terminal region of p97 is a cluster of acidic residues at the C-terminal end (SI Fig. 6A). To test whether this sequence is involved in the PUB domain interaction, two peptides corresponding to either the last 13 (p97-C13) or 10 (p97-C10) residues were synthesized, and their interactions with the PUB domain were characterized by ITC (Fig. 1B and Table 1). The results revealed that the binding affinities of both p97-C13 and p97-C10 to the PUB domain are similar to that of p97-C40 and full-length p97 (Table 1). These data indicate that only the C-terminal 10 residues of p97 are necessary for interaction with the PUB domain of PNGase. Further experiments demonstrated that this protein–protein interaction motif also mediates p97's interaction with Ufd3, albeit with a slightly lower affinity than that with the PUB domain (Table 1). Because Ufd2 competes with Ufd3 for p97 interaction (12), it is possible that Ufd2 also interacts with the C terminus of p97, and all three proteins, Ufd2, Ufd3, and PNGase, may compete with each other for p97 interaction. Another scenario would predict a different binding site for Ufd2, and the Ufd2 and Ufd3 competition would simply result from spatial overlap of both proteins when bound to p97.

Table 1.

ITC analyses of p97–PNGase interactions

Injectant Cell n Kd, μM ΔH, kcal/mol
PNG p97 0.27 ± 0.07 16.7 ± 1.3 −50.5 ± 13.9
PUB p97 0.88 ± 0.03 5.6 ± 0.3 −12.9 ± 0.6
PUB-R55E p97 n.d.
p97-C40 PUB 0.76 ± 0.03 11.1 ± 1.1 −8.9 ± 0.6
p97-C13 PUB 0.90 ± 0.01 3.2 ± 0.12 −14.2 ± 0.3
p97-C10 PUB 0.91 ± 0.01 3.6 ± 0.1 −13.5 ± 0.2
Phos-p97-C10 PUB n.d.
p97-C10 PUB-Y38F 1.15 ± 0.05 39.1 ± 5.3 −18.7 ± 1.5
p97-C10 PUB-K50E 0.77 ± 0.13 32.8 ± 4.1 −14.9 ± 3.2
p97-C10 PUB-R55E n.d.
p97-C10 PUB-N58D n.d.
p97-C10 PUB-R64E 0.83 ± 0.1 140.8 ± 51 −24.8 ± 1.7
p97-C10 Ufd3 330–794 0.86 ± 0.1 34.5 ± 2.5 −9.9 ± 1.4
Phos-p97-C10 Ufd3 330–794 n.d.
Open in a new tab

n.d., no binding detected.

Fig. 1.

Fig. 1.

Open in a new tab

Binding of p97 to the PUB domain of mouse PNGase. (A) GST pull-down assay. GST-PUB immobilized on glutathione beads was incubated with His-tagged p97 variants as indicated. Bound proteins were detected by Western blot with a monoclonal anti-His-tag antibody. (B) Interaction studies by ITC. Binding isotherms are shown for the last 10 residues of p97 (p97-C10) to WT (○) and the Arg55Glu variant of the PUB domain (▵) as well as for the Tyr-805 phosphorylated form of p97-C10 to WT PUB domain (□).

One p97 Hexamer Binds Two PNGase Molecules.

During the protein interaction experiments, we observed that the stoichiometry of the interaction between full-length, hexameric p97 and full-length PNGase is significantly greater than 1:1. After the PNGase–p97 complex was purified by size exclusion chromatography, resolved on SDS/PAGE, and stained with Coomassie blue, the molecular quantities of PNGase and p97 in the complex were estimated. Assuming that PNGase and p97 proteins stain equally well with Coomassie blue, a p97 to PNGase molar ratio of 3.2–3.5:1 was obtained (data not shown). A similar stoichiometry was seen in an ITC experiment by titrating p97 with full-length PNGase (Table 1). These data suggest that p97 interacts with PNGase in a 3:1 molar ratio. Considering that p97 forms a stable hexamer, this ratio corresponds to one p97 hexamer binding two PNGase monomers. Because the isolated PUB domain interacts with full-length p97 and p97-C10 in a 1:1 molar ratio based on our biochemical and cocrystallization data (Table 1 and see below), the stoichiometry of the interaction of the full-length proteins is likely constrained by steric effects imposed by the core and/or C-terminal domains of mPNGase.

Crystal Structure of the mPNGase PUB Domain.

The N-terminal domain (residues 12–111) of mPNGase was crystallized in space group C2 containing one molecule per asymmetric unit (Table 2). The structure was determined by single isomorphous replacement with anomalous scattering and was refined at 1.7-Å resolution (Table 3) to an R factor of 14.7% (Rfree = 17.7%). All residues with the exception of Ser-111 are well defined, and the resulting model is characterized by very good stereochemistry. The PUB domain adopts a compact fold (Fig. 2A), which mainly consists of a left-handed antiparallel four-helical bundle (H1, H2, H4, and H5). Helices H2 and H4 are linked by an extended connection, which is comprised of a 310 helix, a β strand (β1), and the short H3 helix. A twisted antiparallel sheet is formed by β1 and two other strands located between H4 and H5.

Table 2.

Data collection and structure solution

PUB Hg Complex
Space group C2 C2 C2
Cell dimensions, Å and ° 66.64, 52.04, 34.95, 90, 115.25, 90 66.6, 52.2, 35.2, 90, 115.5, 90 65.4, 52.9, 35.0, 90, 115.1, 90
Resolution limits, Å 50–1.7 50–2.5 50–2.28
Rsym 0.055 (0.106) 0.114 (0.145) 0.061 (0.153)
Mean redundancy 4.1 1.8 4.0
Completeness 0.950 (0.568) 0.970 (0.938) 0.996 (0.994)
〈I/sigI〉 30.5 (6.6) 8.5 (6.8) 16.3 (7.1)
Heavy atom sites 2 —
Phasing power (centric/acentric) 0.90/1.25 —
Cullis R factor 0.59 —
FOM 0.49 —
Open in a new tab

Rsym = ΣΣ |I − 〈I〉|/Σ 〈I〉, where I is the ith measurement and 〈I〉 is the weighted mean of all measurements of I. 〈I/sigI〉 indicates the average of the intensity divided by its standard deviation. Numbers in parentheses refer to the respective highest resolution data shell in each dataset. Single isomorphous replacement and anomalous scattering phasing was performed to 2.5-Å resolution. Phasing power is the mean value of the heavy atom structure factor amplitude divided by the lack of closure for isomorphous/anomalous differences. RCullis is the lack of closure divided by the absolute of the difference between FPH and FP for isomorphous differences of centric data. FOM is the figure of merit given for all reflections.

Table 3.

Refinement statistics

PUB Complex
Resolution limits, Å 20–1.7 20–2.28
Number of reflections 10,843 4,692
Number of protein/solvent atoms 0.148 (0.177) 0.195 (0.257)
Deviations from ideal values in
        Bond distances, Å 0.014 0.006
        Bond angles, ° 1.525 0.935
        Chiral volumes, Å3 0.100 0.062
        Planar groups, Å 0.006 0.002
        Torsion angles, ° 5.34, 36.81, 11.87 4.70, 36.97, 13.72
Ramachandran statistics 0.955/0.045/0/0 0.956/0.044/0/0
Average B factor, Å2 (protein/solvent) 12.4/15.5 23.2/29.1
Open in a new tab

Rcryst = Σ||Fo| − |Fc||/Σ|Fo|, where Fo and Fcare the observed and calculated structure factor amplitudes. Rfree is the same as R for 5% of the data randomly omitted from refinement. Ramachandran statistics indicate the fraction of residues in the most favored, additionally allowed, generously allowed, and disallowed regions of the Ramachandran diagram as defined by PROCHECK (50).

Fig. 2.

Fig. 2.

Open in a new tab

Structure of the PUB domain. (A) Ribbon diagram (stereo representation) of the mouse PNGase PUB domain. Helices are shown in blue, β strands in green, and loops in gray. Secondary structure elements and the termini are labeled. Figs. 2–4 were generated with PyMol (48). (B) Surface representation of the PUB domain with electropositive patches in blue and electronegative areas in red. The electrostatic surface potential was calculated with APBS (49).

A surface representation of the roughly spherical molecule reveals a shallow groove with a width of 8 Å and a length of 16 Å in which a hydrophobic pocket is embedded (Fig. 2B). The top half of the groove is surrounded by highly conserved residues that form an extensive hydrogen bond network with 12 water molecules and a glycerol molecule that is coordinated by Arg-55, Asn-58, and His-86 of the PUB domain. A calculation of the electrostatic potential of the PUB domain reveals that positively charged residues are clustered around the groove. Not surprisingly, this structure can be superimposed with the recently published structure of the human PNGase PUB domain (23) resulting in a rms deviation in Cα positions of 0.4 Å.

The PUB domain has also been found in several other proteins present in plants and mammals (22) (SI Fig. 6B). Based on the crystal structure, we can demonstrate that a number of the conserved residues of the PUB domain have important structural roles. For example, the side chains of Leu-35, Ile-42, Leu-65, Leu-75, Phe-80, and L98 contribute to the hydrophobic core of the PUB domain, whereas Pro-46, Gly-79, and Pro-90 are important for maintaining the conformations of loop regions. The indispensable role of Gly-79 and Phe-80 has been demonstrated in our previous mutagenesis study (13). In addition, several conserved hydrophilic residues, including Asn-41, Arg-55, and Asn-58, are of functional importance and, as shown below, are directly involved in interactions with p97.

A search for structural homologues of the PUB domain with the DALI server revealed close matches with a hypothetical protein from Vibrio cholerae (Vc1899), to which no function has been assigned yet, and with the orange carotenoid protein with Z scores of 5.6 and 5.3, respectively. These are followed by several members of the WH (winged-helix) family of DNA binding proteins, including Cdc6 and Orc2 with Z scores of 5.0 and 4.0, respectively. The overall architecture of the PUB domain largely mimics the WH topology, except that H3 and H4 of the PUB domain are shorter and in a different orientation relative to the β sheet compared with typical WH domains (Fig. 3). WH proteins are primarily involved in DNA recognition (24) and they bind to their targets via a positively charged surface. However, the location of this positively charged region and the one in the PUB domain differ (Fig. 3), and it is therefore unclear whether the PUB domain evolved from an ancestral DNA-binding domain.

Fig. 3.

Fig. 3.

Open in a new tab

Relationship of the PUB domain and WH DNA-binding domains of Cdc6 (1FFN, residues 275–388) and Orc2 (1W5S, residues 300–409). The homologous segments between the PUB and WH domains are shown in solid blue, whereas divergent regions are rendered transparent. The p97 binding site for the PUB domain and the DNA binding sites of Cdc6 and Orc2 are indicated by arrows.

Crystal Structure of the PUB Domain in Complex with a p97-Derived Peptide.

To further investigate the interaction between p97 and PNGase, we cocrystallized the mPNGase PUB domain with the p97-C10 peptide (Tables 2 and 3). The last four residues of the peptide, with the sequence Asp-Leu-Tyr-Gly (residues 803–806 of p97), are well defined in the electron density (Fig. 4). In addition, weak density is present for the preceding residue (Asp-802), and this residue has been tentatively included in the model. Binding of the p97 peptide does not perturb the overall structure of the PUB domain as evidenced by the fact that the two structures can be superimposed with a rms deviation of 0.33 Å in Cα positions. Notable differences between the apo-PUB domain and the PUB domain in the complex are only observed for the side chains of Glu-26, Lys-33, and Arg-64.

Fig. 4.

Fig. 4.

Open in a new tab

Interaction between the PUB domain and the C-terminal residues of p97. (A) Surface representation of the PUB domain with the peptide (carbon, nitrogen, and oxygen atoms in yellow, blue, and red, respectively) in stick representation. The surface of the PUB domain is colored according to sequence homology with highly conserved residues in pink, moderately conserved residues in orange, and nonconserved residues in gray. (B) 2Fo − Fc omit electron density map of the p97 C-terminal five residues contoured at a level of 1 times the rms deviation. (C) Detailed view of the interactions between the PUB domain and the peptide. The PUB domain is shown in a ribbon diagram with the p97 interacting residues shown in stick representation. The p97 peptide is colored as in A. Hydrogen bonds are shown as red dashed lines, and water molecules, which mediate interactions, are shown as red spheres.

As expected from the electrostatic surface potential of the PUB domain, p97-C10 binds to the positively charged groove on the surface of the PUB domain formed by H2-H4 and β1. Complex formation is mediated by both ionic and hydrophobic interactions. The C-terminal carboxylate of the peptide interacts with the side chain of Arg-55, which is highly conserved and only replaced by a lysine in the PUB domain of a mouse protein (UBXD1) of unknown function (25) (SI Fig. 6B). Extensive hydrogen bonded interactions are formed between residues Asp-803, Leu-804, Tyr-805, and Gly-806 of p97 and residues Tyr-38, Asn-41, Arg-55, and Asn-58 of the PUB domain. To probe the importance of the ionic interactions, several basic residues around the p97 binding site of the PUB domain were mutated to glutamate and their effects on the p97 interaction were analyzed in ITC experiments. The K50E and R64E mutants of the PUB domain showed reduced affinities, whereas the R55E and the N58D mutants completely lost their ability to interact with p97-C10 and full-length p97 (Table 1). This is in agreement with the high degree of sequence conservation of these residues, especially that of Arg-55 (SI Fig. 6B).

The second and third residues at the C terminus of p97, which are leucine and tyrosine in all eukaryotes (SI Fig. 6A), are engaged in hydrophobic interactions with the PUB domain. Leu-804 of p97 is engaged in van der Waals interactions with Tyr-38 and Leu-34 of the PUB domain, and its main chain oxygen atom hydrogen bonds with the side chain of Tyr-38 (Fig. 4). Interestingly, the side chain of Tyr-805 snugly fits into the aforementioned hydrophobic pocket formed by residues Tyr-38, Asn-41, Lys-50, Tyr-51, Ser-53, and Ile-54 of the PUB domain (Fig. 4 A and B). A water molecule is also present in the Tyr-805 binding pocket and bridges the phenolic hydroxyl group of Tyr-805 to the surrounding residues. The critical location of Tyr-805 in the complex structure prompted us to further explore the possibility that modification of this residue might abolish complex formation. Tyr-805 was mutated to either Phe or Glu in full-length p97, and its interaction with mPNGase was tested. The GST pull-down assay showed that only WT His-tagged p97, but neither the Y805E nor the Y805F mutants, were bound to GST-mPNGase (1–130) (Fig. 1A). This result confirmed the indispensable role of the C-terminal motif, especially Tyr-805, of p97 in its interaction with the PUB domain.

Tyr-805 Phosphorylation Abolishes the p97–PUB Domain Interaction.

Phosphorylation of p97 has been observed during sperm capacitation (26), DNA repair (27), transitional endoplasmic reticulum assembly (28), and T cell receptor activation (29). In addition, phosphorylation of p97 has been shown to affect its association with ubiquitinated proteins and its nuclear localization (30, 31). Interestingly, the primary site of phosphorylation during T cell stimulation is the highly conserved residue Tyr-805 (32). In light of the pivotal role of Tyr-805 in interaction of p97 with PNGase, we compared the binding affinity of phosphorylated and nonphosphorylated p97-C10 peptides to the PUB domain. In agreement with our hypothesis, phosphorylation of Tyr-805 completely abolished p97-C10's interaction with the PUB domain (Fig. 1 and Table 1). Furthermore, phosphorylation of Tyr-805 also eliminated p97-C10's interaction with Ufd3 (Table 1). Although structural details of the p97–Ufd3 interaction are unavailable, we speculate that p97 interacts with the PUB domain and with Ufd3 via a similar mechanism, even though there is no obvious sequence similarity between both of these p97 binding partners.

Does Tyrosine Phosphorylation of p97 Modulate ERAD Activity?

Proper ERAD activity is essential for homeostasis and cell survival. ERAD malfunctions have been associated with various pathophysiological conditions, such as Alzheimer's and Parkinson's disease, as well as with cystic fibrosis (33, 34). In addition, it has been shown that the ERAD pathway is “hijacked” by human CMV during viral infection. The virus compromises the host's immune system by encoding two proteins, US2 and US11, which bind to the MHC class 1 heavy chain and direct it to the ERAD pathway for degradation (35, 36). The biological and potential therapeutical importance of ERAD emphasizes how important it is to have a clear understanding of its regulation.

p97 has an essential role in the ERAD process, in which it extracts unfolded substrates through the putative retrotranslocon and recruits ERAD components to form a degradation complex (11, 13). It has been proposed that the p97 substrates are first recognized by substrate-recruiting cofactors and then processed by substrate-processing cofactors (12). We identified a highly conserved binding motif of p97 that mediates its interaction with at least two substrate-processing cofactors, Ufd3 and PNGase. Furthermore, tyrosine phosphorylation of this motif abolished the interaction of p97 with these cofactors. Based on these observations, we hypothesize that tyrosine phosphorylation/dephosphorylation of p97 may dynamically modulate ERAD activity, affecting both the deglycosylation activity of PNGase and the ubiquitination functions of the Ufd proteins downstream. In this context, it is interesting to note that p97 has been identified as a substrate of protein-tyrosine phosphatases, such as PTPH1 (PTPN3) (37) and PTPN22 (38). Because protein-tyrosine phosphatases, including PTP1B, potentiate IRE1 signaling during endoplasmic reticulum stress (39), we speculate that, under these conditions, p97 is dephosphorylated so as to recruit components of the degradation complex to p97, including PNGase and the Ufd proteins.

Unassembled T cell receptor α and CD3δ are well characterized substrates of the ERAD pathway and have been used as an established model system in protein degradation studies (40). Although it has been observed that Tyr-805 contributes to >90% of p97 tyrosine phosphorylation during T cell receptor activation (32), the rationale for p97 phosphorylation has been unclear so far. Based on our hypothesis, it is possible that the T cell receptor activation-induced p97 tyrosine phosphorylation down-regulates ERAD activity. As a result, more T cell receptor subunits and CD3δ molecules could survive protein degradation and thereby assemble on the cell surface.

The C-terminal protein-binding motif of p97 may be used by other cofactors to regulate p97's function and/or direct it to various cellular processes. In fact, the PUB domain has been found in several proteins with unknown functions (22), such as UBXD1 in mammals and AtPUB1 through AtPUB4 in Arabidopsis thaliana (SI Fig. 6B). UBXD1 is intriguing in that, in addition to the PUB domain, it also contains a UBX domain, which was shown to interact with the N-terminal region of p97 (9, 41). Possessing two p97 interacting domains, each binding to a different part of p97, UBXD1 would be expected to constrain any conformational changes of p97 (if the two domains bind to the same p97 molecule) or cross-link multiple p97 molecules into a supramolecular assembly. In addition, the fact that p97 employs a rather short and unstructured motif to interact with proteins with no obvious sequence similarities (PUB domain, Ufd3, and possibly Ufd2) implies the versatility of this motif and raises the possibility of additional protein interaction partners.

Materials and Methods

Protein Expression and Purification.

The gene encoding the N-terminal domain of mouse PNGase (residues 12–111) was amplified from full-length mPNGase by PCR and was cloned into the NdeI/SapI sites of the pTYB1 vector (pTYB1-PUB) (New England Biolabs, Beverly, MA). All mutations were introduced with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) by using pTYB1-PUB as template. The constructs of mouse p97 were amplified from pQE9-p97 (13) and ligated into the pET28a vector (Novagen, San Diego, CA). Full-length p97 was expressed in M15 cells, whereas all other proteins were expressed overnight in BL21DE3 Codon Plus RIL cells at 15°C after induction with 0.3 mM isopropyl β-d-thiogalactopyranoside at an A600 of 0.6–0.8. Purification of all proteins followed the same protocol, including either chitin (New England Biolabs, Ipswich, MA) or Ni-nitrilotriacetic acid superflow (Qiagen, Valencia, CA) affinity chromatography, followed by MonoQ anion exchange and Superdex 200 size exclusion chromatography (Amersham, Piscataway, NJ).

Protein–Protein Interaction Studies.

ITC measurements were carried out by using a VP-ITC microcalorimeter (MicroCal, Northampton, MA). Before each experiment, the proteins and peptides were dialyzed overnight at 4°C against a buffer containing 20 mM Tris·HCl (pH 8.5), 150 mM NaCl, 2% glycerol, and 1 mM β-mercaptoethanol. Proteins at concentrations of ≈15 μM were titrated with 0.2 mM of either peptides (Anaspec, San Jose, CA) or proteins at 25°C. The binding parameters were calculated by using MicroCal Origin (version 7.0) by fitting the data to a single site binding model. GST pull-down experiments were performed with purified proteins. Two micrograms of GST-mPNGase (1–130) bound to 8 μl of GSH-agarose beads were mixed with 4 μg of either WT or mutated His6-tagged p97 in 0.5 ml of binding buffer [1× PBS, 1% Triton X-100, 1 mM MgCl2, 1 mM ATP, 5 mM DTT, 5% glycerol, 4 mM PMSF (pH 7.4)]. The binding experiments were performed at 4°C for 1 h, after which the beads were washed three times with binding buffer. Bound proteins were eluted with SDS sample buffer and analyzed by SDS/PAGE followed by Western blot analysis with a monoclonal antibody against His-tag.

Crystallization and Structure Determination.

The PUB domain was crystallized by hanging drop vapor diffusion against a reservoir solution containing 18–22% PEG 5000 MME and 14–18% Tacsimate (Hampton Research, Aliso Viejo, CA). The heavy atom derivative was prepared by soaking the crystals in mother liquor containing 2.5 mM sodium ethylmercurithiosalicylate for 20 min. After mixing the PUB domain and p97-C10 in a 1:2 molar ratio, the resulting complex was crystallized under similar conditions. The native dataset was collected with a Rigaku (Tokyo, Japan) RU-H3R rotating anode x-ray generator operating at 50 kV and 100 mA and equipped with confocal multilayer optics and a Rigaku RAXIS-IV++ detector. The mercury derivative and the complex datasets were collected on beam line X26C of the National Synchrotron Light Source at Brookhaven National Laboratory. All diffraction data were collected at 100 K, indexed, integrated, and scaled with HKL2000 (42).

The structure was determined by the single isomorphous replacement and anomalous scattering method. Phase determination and density modification were carried out with Solve/Resolve (43, 44) to 2.5 Å including building of a partial structure. The resulting model was input into ARP/Warp (45), which was able to build 98 of the 100 residues. The protein model was completed manually with the aid of the program O (46) and was refined with Refmac (47). Water molecules were added automatically with ARP/Warp.

Supplementary Material

Supporting Figures
pnas_0702966104_index.html (2.4KB, html)

Acknowledgments

This work was supported by National Institutes of Health Grants GM33814 (to W.J.L.) and DK54835 (to H.S.).

Abbreviations

ERAD

endoplasmic reticulum-associated protein degradation

PNGase

peptide:N-glycanase

mPNGase

mouse PNGase

ITC

isothermal titration calorimetry.

Footnotes

The authors declare no conflict of interest.

Data deposition: The atomic coordinates and structure factors have been deposited in the Protein Data Bank, www.pdb.org (PDB ID code www.pdb.org (PDB ID code 2HPJ for the apo-structure and 2HPL for the complex).

This article contains supporting information online at www.pnas.org/cgi/content/full/0702966104/DC1.

References

  • 1.Hampton RY. Curr Opin Cell Biol. 2002;14:476–482. doi: 10.1016/s0955-0674(02)00358-7. [DOI] [PubMed] [Google Scholar]
  • 2.Meusser B, Hirsch C, Jarosch E, Sommer T. Nat Cell Biol. 2005;7:766–772. doi: 10.1038/ncb0805-766. [DOI] [PubMed] [Google Scholar]
  • 3.Huibregtse JM. Cell. 2005;120:2–4. doi: 10.1016/j.cell.2004.12.029. [DOI] [PubMed] [Google Scholar]
  • 4.Wang Q, Song C, Li CC. J Struct Biol. 2004;146:44–57. doi: 10.1016/j.jsb.2003.11.014. [DOI] [PubMed] [Google Scholar]
  • 5.Neuwald AF, Aravind L, Spouge JL, Koonin EV. Genome Res. 1999;9:27–43. [PubMed] [Google Scholar]
  • 6.Wojcik C, Yano M, DeMartino GN. J Cell Sci. 2004;117:281–292. doi: 10.1242/jcs.00841. [DOI] [PubMed] [Google Scholar]
  • 7.Brunger AT, DeLaBarre B. FEBS Lett. 2003;555:126–133. doi: 10.1016/s0014-5793(03)01107-4. [DOI] [PubMed] [Google Scholar]
  • 8.Pye VE, Dreveny I, Briggs LC, Sands C, Beuron F, Zhang X, Freemont PS. J Struct Biol. 2006;156:12–28. doi: 10.1016/j.jsb.2006.03.003. [DOI] [PubMed] [Google Scholar]
  • 9.Dreveny I, Pye VE, Beuron F, Briggs LC, Isaacson RL, Matthews SJ, McKeown C, Yuan X, Zhang X, Freemont PS. Biochem Soc Trans. 2004;32:715–720. doi: 10.1042/BST0320715. [DOI] [PubMed] [Google Scholar]
  • 10.Alzayady KJ, Panning MM, Kelley GG, Wojcikiewicz RJ. J Biol Chem. 2005;280:34530–34537. doi: 10.1074/jbc.M508890200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Richly H, Rape M, Braun S, Rumpf S, Hoege C, Jentsch S. Cell. 2005;120:73–84. doi: 10.1016/j.cell.2004.11.013. [DOI] [PubMed] [Google Scholar]
  • 12.Rumpf S, Jentsch S. Mol Cell. 2006;21:261–269. doi: 10.1016/j.molcel.2005.12.014. [DOI] [PubMed] [Google Scholar]
  • 13.Li G, Zhao G, Zhou X, Schindelin H, Lennarz WJ. Proc Natl Acad Sci USA. 2006;103:8348–8353. doi: 10.1073/pnas.0602747103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Hetzer M, Meyer HH, Walther TC, Bilbao-Cortes D, Warren G, Mattaj IW. Nat Cell Biol. 2001;3:1086–1091. doi: 10.1038/ncb1201-1086. [DOI] [PubMed] [Google Scholar]
  • 15.Cao K, Nakajima R, Meyer HH, Zheng Y. Cell. 2003;115:355–367. doi: 10.1016/s0092-8674(03)00815-8. [DOI] [PubMed] [Google Scholar]
  • 16.Kondo H, Rabouille C, Newman R, Levine TP, Pappin D, Freemont P, Warren G. Nature. 1997;388:75–78. doi: 10.1038/40411. [DOI] [PubMed] [Google Scholar]
  • 17.Uchiyama K, Kondo H. J Biochem (Tokyo) 2005;137:115–119. doi: 10.1093/jb/mvi028. [DOI] [PubMed] [Google Scholar]
  • 18.Hirsch C, Blom D, Ploegh HL. EMBO J. 2003;22:1036–1046. doi: 10.1093/emboj/cdg107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lee JH, Choi JM, Lee C, Yi KJ, Cho Y. Proc Natl Acad Sci USA. 2005;102:9144–9149. doi: 10.1073/pnas.0502082102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Zhao G, Zhou X, Wang L, Li G, Kisker C, Lennarz WJ, Schindelin H. J Biol Chem. 2006;281:13751–13761. doi: 10.1074/jbc.M600137200. [DOI] [PubMed] [Google Scholar]
  • 21.Zhou X, Zhao G, Truglio JJ, Wang L, Li G, Lennarz WJ, Schindelin H. Proc Natl Acad Sci USA. 2006;103:17214–17219. doi: 10.1073/pnas.0602954103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Suzuki T, Park H, Till EA, Lennarz WJ. Biochem Biophys Res Commun. 2001;287:1083–1087. doi: 10.1006/bbrc.2001.5688. [DOI] [PubMed] [Google Scholar]
  • 23.Allen MD, Buchberger A, Bycroft M. J Biol Chem. 2006;281:25502–25508. doi: 10.1074/jbc.M601173200. [DOI] [PubMed] [Google Scholar]
  • 24.Gajiwala KS, Burley SK. Curr Opin Struct Biol. 2000;10:110–116. doi: 10.1016/s0959-440x(99)00057-3. [DOI] [PubMed] [Google Scholar]
  • 25.Carim-Todd L, Escarceller M, Estivill X, Sumoy L. Biochim Biophys Acta. 2001;1517:298–301. doi: 10.1016/s0167-4781(00)00248-7. [DOI] [PubMed] [Google Scholar]
  • 26.Ficarro S, Chertihin O, Westbrook VA, White F, Jayes F, Kalab P, Marto JA, Shabanowitz J, Herr JC, Hunt DF, et al. J Biol Chem. 2003;278:11579–11589. doi: 10.1074/jbc.M202325200. [DOI] [PubMed] [Google Scholar]
  • 27.Livingstone M, Ruan H, Weiner J, Clauser KR, Strack P, Jin S, Williams A, Greulich H, Gardner J, Venere M, et al. Cancer Res. 2005;65:7533–7540. doi: 10.1158/0008-5472.CAN-04-3729. [DOI] [PubMed] [Google Scholar]
  • 28.Lavoie C, Chevet E, Roy L, Tonks NK, Fazel A, Posner BI, Paiement J, Bergeron JJ. Proc Natl Acad Sci USA. 2000;97:13637–13642. doi: 10.1073/pnas.240278097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Egerton M, Ashe OR, Chen D, Druker BJ, Burgess WH, Samelson LE. EMBO J. 1992;11:3533–3540. doi: 10.1002/j.1460-2075.1992.tb05436.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Madeo F, Schlauer J, Zischka H, Mecke D, Frohlich KU. Mol Biol Cell. 1998;9:131–141. doi: 10.1091/mbc.9.1.131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Klein JB, Barati MT, Wu R, Gozal D, Sachleben LR, Jr, Kausar H, Trent JO, Gozal E, Rane MJ. J Biol Chem. 2005;280:31870–31881. doi: 10.1074/jbc.M501802200. [DOI] [PubMed] [Google Scholar]
  • 32.Egerton M, Samelson LE. J Biol Chem. 1994;269:11435–11441. [PubMed] [Google Scholar]
  • 33.Vij N, Fang S, Zeitlin PL. J Biol Chem. 2006;281:17369–17378. doi: 10.1074/jbc.M600509200. [DOI] [PubMed] [Google Scholar]
  • 34.Yoshida H. FEBS J. 2007;274:630–658. doi: 10.1111/j.1742-4658.2007.05639.x. [DOI] [PubMed] [Google Scholar]
  • 35.Wiertz EJ, Jones TR, Sun L, Bogyo M, Geuze HJ, Ploegh HL. Cell. 1996;84:769–779. doi: 10.1016/s0092-8674(00)81054-5. [DOI] [PubMed] [Google Scholar]
  • 36.Wiertz EJ, Tortorella D, Bogyo M, Yu J, Mothes W, Jones TR, Rapoport TA, Ploegh HL. Nature. 1996;384:432–438. doi: 10.1038/384432a0. [DOI] [PubMed] [Google Scholar]
  • 37.Zhang SH, Liu J, Kobayashi R, Tonks NK. J Biol Chem. 1999;274:17806–17812. doi: 10.1074/jbc.274.25.17806. [DOI] [PubMed] [Google Scholar]
  • 38.Wu J, Katrekar A, Honigberg LA, Smith AM, Conn MT, Tang J, Jeffery D, Mortara K, Sampang J, Williams SR, et al. J Biol Chem. 2006;281:11002–11010. doi: 10.1074/jbc.M600498200. [DOI] [PubMed] [Google Scholar]
  • 39.Gu F, Nguyen DT, Stuible M, Dube N, Tremblay ML, Chevet E. J Biol Chem. 2004;279:49689–49693. doi: 10.1074/jbc.C400261200. [DOI] [PubMed] [Google Scholar]
  • 40.Tiwari S, Weissman AM. J Biol Chem. 2001;276:16193–16200. doi: 10.1074/jbc.M007640200. [DOI] [PubMed] [Google Scholar]
  • 41.Buchberger A, Howard MJ, Proctor M, Bycroft M. J Mol Biol. 2001;307:17–24. doi: 10.1006/jmbi.2000.4462. [DOI] [PubMed] [Google Scholar]
  • 42.Otwinowski Z, Minor W. Processing of X-Ray Diffraction Data Collected in Oscillation Mode. New York: Academic; 1997. [DOI] [PubMed] [Google Scholar]
  • 43.Terwilliger TC, Berendzen J. Acta Crystallogr D Biol Crystallogr. 1999;55:849–861. doi: 10.1107/S0907444999000839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Terwilliger TC. Acta Crystallogr D Biol Crystallogr. 2000;56:965–972. doi: 10.1107/S0907444900005072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Perrakis A, Morris R, Lamzin VS. Nat Struct Biol. 1999;6:458–463. doi: 10.1038/8263. [DOI] [PubMed] [Google Scholar]
  • 46.Jones TA, Zou JY, Cowan SW, Kjeldgaard M. Acta Crystallogr A. 1991;47:110–119. doi: 10.1107/s0108767390010224. [DOI] [PubMed] [Google Scholar]
  • 47.Murshudov GN, Vagin AA, Dodson EJ. Acta Crystallogr D Biol Crystallogr. 1997;53:240–255. doi: 10.1107/S0907444996012255. [DOI] [PubMed] [Google Scholar]
  • 48.DeLano WL. The PyMOL User's Manual. San Carlos, CA: DeLano Scientific; 2002. [Google Scholar]
  • 49.Baker NA, Sept D, Joseph S, Holst MJ, McCammon JA. Proc Natl Acad Sci USA. 2001;98:10037–10041. doi: 10.1073/pnas.181342398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Morris AL, MacArthur MW, Hutchinson EG, Thornton JM. Proteins. 1992;12:345–364. doi: 10.1002/prot.340120407. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Figures
pnas_0702966104_index.html (2.4KB, html)
pnas_0702966104_1.pdf (144.4KB, pdf)
pnas_0702966104_2.pdf (2.7MB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES