**IBMPFD and p97, the Structural and Molecular Basis for Functional Disruption**

Wai-Kwan Tang and Di Xia *Laboratory of Cell Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA* 

#### **1. Introduction**

154 Neuromuscular Disorders

Sato S, Hirakata M, Kuwana M, Suwa A, Inada S, Mimori T, Nishikawa T, Oddis CV, Ikeda

Tagoff IN, Johnson AE, Miller FW. (1990). Antibody to signal recognition particle in

Takada K, Nagasaka K, Miyasaka N. (2005). Polymyositis / dermatomyositis and interstitial

Wilkes MR, Sereika SM, Fertig N, Lucas MR, Oddis CV. (2005). Treatment of antisynthetase-

Yamasaki Y, Yamada H, Yamasaki M, Ohkubo M, Azuma K, Matsuoka S, Kurihara Y,

*Rheumatol (Oxford)*, Vol. 46, No. 1, (Jun 2006), pp. 124-130, ISSN 1462-0324.

*Autoimmunity*, Vol. 38, No. 5, (Aug 2005), pp. 383-392, ISSN 0891-6934. Takada K, Kishi J, Miyasaka N. (2007). Step-up versus primary intensive approach to the

2005), pp.1571-1576, ISSN 0004-3591.

(Aug 2005), pp. 2439-2446, ISSN 0004-3591.

123-130, ISSN 1439-7595.

3591.

Y. (2005). Autoantibodies to a 140-kd polypeptide, CADM-140, in Japanese patients with clinically amyopathic dermatomyositis. *Arthritis Rheum*, Vol.52, No. 5, (May

polymyositis. *Arthritis Rheum*, Vol. 33, No. 9, (Sep 1990), pp.1361-1370, ISSN 0004-

lung disease: a new therapeutic approach with T-cell-specific immunosuppressants.

treatment of interstitial pneumonia associated with dermatomyositis/ polymyositis: a retrospective study. *Mod Rheumatol*, Vol. 17, No. 2, (Apr 2007), pp.

associated interstitial lung disease with tacrolimus. *Arthritis Rheum*, Vol. 52, No. 8,

Osada H, Satoh M, Ozaki S. (2007) Intravenous cyclophosphamide therapy for progressive interstitial pneumonia in patients with polymyositis/dermatomyositis.

Inclusion body myopathy associated with Paget's disease of the bone and frontotemporal dementia (IBMPFD, OMIM 167320) is an inherited, autosomal dominant, adult onset multidisorder, which affects the muscle, bone and the brain (Watts et al., 2004). It is a rare condition with unknown worldwide prevalence. Affected individuals may display one or a combination of the following three symptoms which, however, are generally not recognized until patients are in their 40s or 50s (Weihl, 2011). (1) IBM (Inclusion body myopathy): About 90% of all patients develop proximal and distal muscle weakness initially with atrophy of the pelvic and shoulder girdle muscles (Kimonis et al., 2000; Kimonis et al., 2008b; Kovach et al., 2001; Watts et al., 2003). Cellular inclusion bodies and rimmed vacuoles are commonly found in these muscle tissues (Kimonis et al., 2008a; Kimonis et al., 2000; Watts et al., 2004). Characteristically, two proteins are most frequently found co-localized with the inclusion, ubiquitin and TDP-43 (TAR DNA binding protein-43) (Ritson et al., 2010; Weihl et al., 2008). Ubiquitin is a signaling molecule that directs protein substrates into a variety of cellular pathways, including protein degradation. Misfolded or unwanted proteins are labeled with ubiquitin, mostly in the form of polyubiquitin, and targeted for degradation (Clague and Urbe, 2010). TDP-43, on the other hand, is believed to be a substrate itself for either proteasome or autophagal degradation (Caccamo et al., 2009; Wang et al., 2010). Detection of these proteins in the inclusions suggests impairments in the protein degradation pathways. (2) PDB (Paget's disease of the bone): About half of IBMPFD patients develop PDB, which is caused by an imbalance in the activities between osteoblasts and osteoclasts. (3) FTD (Frontotemporal dementia): Only ~30% of patients develop FTD, which is characterized by language and/or behavioral dysfunction. Interestingly, clinical manifestation of these symptoms is rather random, and has no clear-cut correlation with family history or mutations. Even within isolated families bearing the same genetic mutation, individuals can exhibit different symptoms. These heterogeneities in clinical presentations cause frequent misdiagnoses of IBMPFD patients (van der Zee et al., 2009). Accurate diagnosis of IBMPFD often requires molecular genetic testing, in addition to a combined clinical diagnosis of myopathy, PDB and FTD.

IBMPFD and p97, the Structural and Molecular Basis for Functional Disruption 157

attached to the periphery of the D1 ring, making one ring appear larger than the other. The central pore does not run unrestricted through the hexamer, but has a narrowing or a

constriction point that is formed by a bound zinc ion (DeLaBarre and Brunger, 2003).

Fig. 1. **Structure of p97** (A) Domain organization of the full-length p97. (B) Ribbon

**5. Wild type vs. IBMPFD mutant p97: Structural characteristics** 

positions representing twenty mutations are presented.

**4. Mutations in p97 associated with IBMPFD** 

domains (N-D1-interface, Fig. 1C).

representation of the crystal structure of p97 based on PDB: 1OZ4. Domains are color-coded using the scheme in (A) and two views are presented. (C) Locations of IBMPFD mutations are shown in the context of the p97 N-D1 structure (PDB: 1E32), which has the D1-domain bound with ADP. The IBMPFD mutations are represented by yellow balls. Thirteen

So far, only single amino acid substitutions in p97 have been identified from all the clinical IBMPFD specimens examined. Altogether, twenty missense mutations found in 13 different amino acid positions in p97 have been reported to be associated with the disease and the majority of them involve substitutions of arginine residues (Table 1) (http://www.molgen.ua.ac.be/FTDMutations). While more than half of these mutations are located in the N-domain (Ile27, Arg93, Arg95, Pro137, Arg155, Gly157 and Arg159), a few are found in the N-D1 linker region between the N- and D1-domains (Arg191 and Leu198) and in the D1 domain (Ala232, Thr262, Asn387 and Ala439). None has been found in the D2-domain. Among these, mutations at Arg155 are the most frequently observed in patients (Table 1). Interestingly, mapping these mutations onto the three-dimensional p97 structure in the ADP-bound form revealed that they all clustered at the interface between the N- and D1-

Changes in structure as a result of amino acid mutations can lead to a global disruption of the protein folding, resulting rapid clearance by cellular stress response mechanisms, or to localized structural changes that cause complete loss of the protein function, or to subtle conformational changes that alter the function of the protein. Structural changes in mutant

#### **2. What is p97?**

In the year 2000, IBMPFD was recognized as a genetically distinct clinical syndrome (Kimonis et al., 2000) and was subsequently linked to heterozygous missense mutations in a highly abundant cellular protein called p97 (also called valosin-containing protein, VCP) (Watts et al., 2004). P97 belongs to the family of AAA+ proteins (ATPases Associated with various cellular Activities), which use the energy from hydrolyzing ATP to drive mechanical work necessary for a host of functions including homotypic membrane fusion, cell cycle regulation and protein degradation (Wang et al., 2004; Woodman, 2003; Ye, 2006). The multi-functionality of p97 is consistent with its embryonic lethality when the p97 gene or its homologs are disrupted or knocked-out in the mouse, in yeast, in trypanosomes, and in Drosophila (Frohlich et al., 1991; Lamb et al., 2001; Leon and McKearin, 1999; Muller et al., 2007). Moreover, the functional versatility of p97 appears to lie in its ability to interact with a large variety of adaptor proteins. For instance, binding to the protein p47 incorporates p97 in the membrane fusion pathway (Kondo et al., 1997), whereas the p97-Ufd1-Npl4 complex participates in ER associated degradation (ERAD) (Richly et al., 2005). So far, more than twenty adaptor proteins have been identified that interact with p97 (Madsen et al., 2009), but detailed molecular mechanisms of these interactions remain elusive.

#### **3. Structure of p97**

Structurally, p97 is a homo-hexamer, each subunit (806 residues) consisting of three domains: a unique N-terminal domain (N-domain) followed by two conserved AAA+ ATPase domains (D1- and D2-domain) in tandem (DeLaBarre and Brunger, 2003; Huyton et al., 2003) (Fig. 1A). The N-domain (residues 1-184) contains two sub-domains, an Nterminal double -barrel and a C-terminal four-stranded -barrel, and is responsible for interacting with most adaptor proteins as well as with protein substrates. Both the D1- (residues 211-463) and D2-domains (residues 483-762) are typical AAA+ ATPase domains comprised of two sub-domains: a large N-terminal RecA-like domain with an / fold and a smaller C-terminal -helical bundle domain. The D1-domain is essential for hexamerization of p97 subunits (Wang et al., 2003) and the hexameric ring formation is predominantly mediated through interactions between the RecA-like sub-domains (Fig. 1B). However, unlike many members of the AAA+ family proteins such as the *E. coli* ClpA unfoldase, the hexamerization of p97 subunits does not require the binding of nucleotide (ADP or ATP) at D1-domains, though it has been shown that nucleotide binding does accelerate p97 hexamer formation (Singh and Maurizi, 1994; Wang et al., 2003). Most of the ATPase activities of p97 involve the D2-domain, presumably required for the processing of substrates (Song & Li, 2003).

Connecting the domains are loops that have been shown to play important functions. The N-D1 loop is 27 residues long and is embedded at the interface between the N-domain and the D1-domain. The short peptide stretch (residues 763-806) immediately following the D2 domain is another region for adaptor protein binding. Although not as common as the Ndomain, this C-terminal tail has been shown to interact with a number of proteins, such as Ubxd1 (Allen et al., 2006; Madsen et al., 2008). Similar to other Type-II AAA+ assemblies, the p97 assembly was revealed by electron microscopy (EM) and crystallography to be a twotiered concentric ring encircling a central pore or axial channel (Fig. 1B). The N-domains are

In the year 2000, IBMPFD was recognized as a genetically distinct clinical syndrome (Kimonis et al., 2000) and was subsequently linked to heterozygous missense mutations in a highly abundant cellular protein called p97 (also called valosin-containing protein, VCP) (Watts et al., 2004). P97 belongs to the family of AAA+ proteins (ATPases Associated with various cellular Activities), which use the energy from hydrolyzing ATP to drive mechanical work necessary for a host of functions including homotypic membrane fusion, cell cycle regulation and protein degradation (Wang et al., 2004; Woodman, 2003; Ye, 2006). The multi-functionality of p97 is consistent with its embryonic lethality when the p97 gene or its homologs are disrupted or knocked-out in the mouse, in yeast, in trypanosomes, and in Drosophila (Frohlich et al., 1991; Lamb et al., 2001; Leon and McKearin, 1999; Muller et al., 2007). Moreover, the functional versatility of p97 appears to lie in its ability to interact with a large variety of adaptor proteins. For instance, binding to the protein p47 incorporates p97 in the membrane fusion pathway (Kondo et al., 1997), whereas the p97-Ufd1-Npl4 complex participates in ER associated degradation (ERAD) (Richly et al., 2005). So far, more than twenty adaptor proteins have been identified that interact with p97 (Madsen et al., 2009),

Structurally, p97 is a homo-hexamer, each subunit (806 residues) consisting of three domains: a unique N-terminal domain (N-domain) followed by two conserved AAA+ ATPase domains (D1- and D2-domain) in tandem (DeLaBarre and Brunger, 2003; Huyton et al., 2003) (Fig. 1A). The N-domain (residues 1-184) contains two sub-domains, an Nterminal double -barrel and a C-terminal four-stranded -barrel, and is responsible for interacting with most adaptor proteins as well as with protein substrates. Both the D1- (residues 211-463) and D2-domains (residues 483-762) are typical AAA+ ATPase domains comprised of two sub-domains: a large N-terminal RecA-like domain with an / fold and a smaller C-terminal -helical bundle domain. The D1-domain is essential for hexamerization of p97 subunits (Wang et al., 2003) and the hexameric ring formation is predominantly mediated through interactions between the RecA-like sub-domains (Fig. 1B). However, unlike many members of the AAA+ family proteins such as the *E. coli* ClpA unfoldase, the hexamerization of p97 subunits does not require the binding of nucleotide (ADP or ATP) at D1-domains, though it has been shown that nucleotide binding does accelerate p97 hexamer formation (Singh and Maurizi, 1994; Wang et al., 2003). Most of the ATPase activities of p97 involve the D2-domain, presumably required for the processing of

Connecting the domains are loops that have been shown to play important functions. The N-D1 loop is 27 residues long and is embedded at the interface between the N-domain and the D1-domain. The short peptide stretch (residues 763-806) immediately following the D2 domain is another region for adaptor protein binding. Although not as common as the Ndomain, this C-terminal tail has been shown to interact with a number of proteins, such as Ubxd1 (Allen et al., 2006; Madsen et al., 2008). Similar to other Type-II AAA+ assemblies, the p97 assembly was revealed by electron microscopy (EM) and crystallography to be a twotiered concentric ring encircling a central pore or axial channel (Fig. 1B). The N-domains are

but detailed molecular mechanisms of these interactions remain elusive.

**2. What is p97?** 

**3. Structure of p97** 

substrates (Song & Li, 2003).

attached to the periphery of the D1 ring, making one ring appear larger than the other. The central pore does not run unrestricted through the hexamer, but has a narrowing or a constriction point that is formed by a bound zinc ion (DeLaBarre and Brunger, 2003).

Fig. 1. **Structure of p97** (A) Domain organization of the full-length p97. (B) Ribbon representation of the crystal structure of p97 based on PDB: 1OZ4. Domains are color-coded using the scheme in (A) and two views are presented. (C) Locations of IBMPFD mutations are shown in the context of the p97 N-D1 structure (PDB: 1E32), which has the D1-domain bound with ADP. The IBMPFD mutations are represented by yellow balls. Thirteen positions representing twenty mutations are presented.
