Valosin-containing protein (p97) is a regulator of endoplasmic reticulum stress and of the degradation of N-end rule and ubiquitin-fusion degradation pathway substrates in mammalian cells.

Abstract

Valosin-containing protein (VCP; p97; cdc48 in yeast) is a hexameric ATPase of the AAA family (ATPases with multiple cellular activities) involved in multiple cellular functions, including degradation of proteins by the ubiquitin (Ub)-proteasome system (UPS). We examined the consequences of the reduction of VCP levels after RNA interference (RNAi) of VCP. A new stringent method of microarray analysis demonstrated that only four transcripts were nonspecifically affected by RNAi, whereas approximately 30 transcripts were affected in response to reduced VCP levels in a sequence-independent manner. These transcripts encoded proteins involved in endoplasmic reticulum (ER) stress, apoptosis, and amino acid starvation. RNAi of VCP promoted the unfolded protein response, without eliciting a cytosolic stress response. RNAi of VCP inhibited the degradation of R-GFP (green fluorescent protein) and Ub-(G76V)-GFP, two cytoplasmic reporter proteins degraded by the UPS, and of alpha chain of the T-cell receptor, an established substrate of the ER-associated degradation (ERAD) pathway. Surprisingly, RNAi of VCP had no detectable effect on the degradation of two other ERAD substrates, alpha1-antitrypsin and deltaCD3. These results indicate that VCP is required for maintenance of normal ER structure and function and mediates the degradation of some proteins via the UPS, but is dispensable for the UPS-dependent degradation of some ERAD substrates.

Full text

Page 1

Molecular Biology of the Cell
Vol. 17, 4606–4618, November 2006
Valosin-containing Protein (p97) Is a Regulator of
Endoplasmic Reticulum Stress and of the Degradation of
N-End Rule and Ubiquitin-Fusion Degradation Pathway
Substrates in Mammalian Cells
Cezary Wo ´jcik,*†Maga Rowicka,‡Andrzej Kudlicki,‡Dominika Nowis,*
Elizabeth McConnell,* Marek Kujawa,§and George N. DeMartino†
*Department of Anatomy and Cell Biology, Indiana University School of Medicine, Evansville, IN 47712;
Departments of†Physiology and‡Biochemistry, University of Texas Southwestern Medical Center, Dallas, TX
75390; and§Department of Histology and Embryology, Medical University of Warsaw, 02-004 Warsaw, Poland
Submitted May 18, 2006; Revised August 3, 2006; Accepted August 7, 2006
Monitoring Editor: Jeffrey Brodsky
Valosin-containing protein (VCP; p97; cdc48 in yeast) is a hexameric ATPase of the AAA family (ATPases with multiple
cellular activities) involved in multiple cellular functions, including degradation of proteins by the ubiquitin (Ub)–
proteasome system (UPS). We examined the consequences of the reduction of VCP levels after RNA interference (RNAi)
of VCP. A new stringent method of microarray analysis demonstrated that only four transcripts were nonspecifically
affected by RNAi, whereas ?30 transcripts were affected in response to reduced VCP levels in a sequence-independent
manner. These transcripts encoded proteins involved in endoplasmic reticulum (ER) stress, apoptosis, and amino acid
starvation. RNAi of VCP promoted the unfolded protein response, without eliciting a cytosolic stress response. RNAi of
VCP inhibited the degradation of R-GFP (green fluorescent protein) and Ub-G76V-GFP, two cytoplasmic reporter proteins
degraded by the UPS, and of ? chain of the T-cell receptor, an established substrate of the ER-associated degradation
(ERAD) pathway. Surprisingly, RNAi of VCP had no detectable effect on the degradation of two other ERAD substrates,
?1-antitrypsin and ?CD3. These results indicate that VCP is required for maintenance of normal ER structure and function
and mediates the degradation of some proteins via the UPS, but is dispensable for the UPS-dependent degradation of some
ERAD substrates.
INTRODUCTION
Valosin-containing protein VCP (p97; cdc48 in yeast) is a
hexameric type II ATPase of the AAA family (ATPases with
multiple cellular activities) that mediates disparate cellular
functions, including endoplasmic reticulum-associated deg-
radation (ERAD) via the ubiquitin–proteasome system (UPS)
(Woodman, 2003; Dreveny et al., 2004; Wang et al., 2004;
Bar-Nun, 2005; Halawani and Latterich, 2006). VCP interacts
with at least 30 different cellular proteins, some of which
may differentially mediate its functions. For example, VCP
forms a complex with the Ufd1–Npl4 heterodimer that is
essential for its role in ERAD, a process by which constituent
and transient endoplasmic reticulum (ER) proteins are re-
moved from the ER and degraded in the cytosol by the 26S
proteasome. A general model for the role of VCPUfd1–Npl4in
ERAD involves binding of polyubiquitinated ERAD sub-
strates at the cytoplasmic face of the ER membrane both
before and after substrate ubiquitination, followed by a com-
plete substrate extraction/dislocation and its transfer to the
26S proteasome (Meyer et al., 2000, 2002; Dai and Li, 2001;
Rabinovich et al., 2002; Elkabetz et al., 2003; Ye et al., 2003).
Substrate dislocation from the ER could result from mechan-
ical stress transmitted to the bound substrate by conforma-
tional changes in VCP during cycles of VCP-catalyzed ATP
hydrolysis (Zhang et al., 2000; Wang et al., 2004).
ERAD is a component of a coordinated cellular response
to ER stress, termed the unfolded protein response (UPR)
(Harding et al., 2002; Ma and Hendershot, 2002; Kostova and
Wolf, 2003; Sitia and Braakman, 2003). UPR can be promoted
by the buildup of unfolded proteins in the ER and consti-
tutes a mechanism to reduce this burden. UPR acutely re-
duces translation of new proteins, followed by increased
expression of chaperones to aid folding of existing proteins
and enhanced elimination of proteins that cannot be re-
folded. In mammals, apoptosis is initiated if ER stress is not
relieved. A critical UPR pathway is initiated by activation of
IRE-1, an ER membrane endonuclease that splices XBP-1
mRNA (Yoshida et al., 2001). Translation of spliced XBP-1
mRNA promotes transcriptional activation of genes for UPR,
including those required for ERAD (Sriburi et al., 2004).
VCP is required for the fusion of ER and Golgi membranes
(Latterich et al., 1995; Rabouille et al., 1995; Patel et al., 1998).
This article was published online ahead of print in MBC in Press
(http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E06–05–0432)
on August 16, 2006.
Address correspondence to: Cezary Wo ´jcik (cwojcik@iupui.edu) or
George N. DeMartino (george.demartino@utsouthwestern.edu).
Abbreviations used: AAA, ATPases with multiple cellular activities;
BFA, brefeldin A; ERAD, endoplasmic reticulum-associated degra-
dation; GFP, green fluorescent protein; R-GFP, GFP with the N-
terminal Met replaced by Arg; siRNA, small interfering RNA;
?TCR, ? chain of the T-cell receptor; Ub-G76V-GFP, GFP with a
noncleavable, mutated (Gly76Val) Ub attached to its N terminus;
UFD, ubiquitin-fusion degradation; UPR, unfolded protein re-
sponse; UPS, ubiquitin-proteasome system; VCP, valosin-contain-
ing protein.
4606 © 2006 by The American Society for Cell Biology

Page 2

RNA interference (RNAi) of VCP in HeLa cells results in the
formation of large intracellular vacuoles, likely derived from
ER (Wojcik et al., 2004b). Thus, VCP seems to be required for
normal ER function, whereas reduced VCP content seems to
induce ER stress, perhaps as consequence of reduced con-
stitutive ERAD and/or by disturbing the fusion of ER mem-
branes. RNAi of VCP also caused a general increase in
polyubiquitinated cellular proteins, indicative of impaired
UPS function (Wojcik et al., 2004b). However, it is unclear
whether this effect reflects the quantitative significance of
ERAD to overall cellular protein degradation or whether
VCP mediates UPS-dependent degradation of non-ERAD
substrates, as shown previously for several individual pro-
teins (Johnson et al., 1995; Ghislain et al., 1996; Dai et al., 1998;
Dai and Li, 2001). To gain insight to these various issues, we
have analyzed altered transcription profiles in mammalian
cells subjected to RNAi of VCP and directly determined the
role of VCP in UPR and in UPS-dependent degradation of
specific ERAD and non-ERAD substrates. Our results dem-
onstrate that VCP mediates multiple aspects of ER structure
and function and multiple aspects of UPS function. Surpris-
ingly, however, VCP is not required for the degradation of
all ERAD substrates.
MATERIALS AND METHODS
Antibodies, Reagents, and Plasmids
Anti-hemagglutinin (HA) monoclonal antibody (mAb) was from Covance
(Princeton, NJ), anti-ubiquitin mAb was from Santa Cruz Biotechnology
(Santa Cruz, CA), anti-VCP mAb was from BD Biosciences (Franklin Lakes,
NJ), anti-green fluorescent protein (GFP) mAb was from Roche (Alameda,
CA), and anti-KDEL antibody detecting BiP and GRP94 as well as anti-hsp70
antibody were from Stressgen Biotechnologies (Victoria, British Columbia,
Canada). The plasmids encoding ubiquitin (Ub)-R-GFP and Ub-G76V-GFP
were derived from pEGFP-N1 (Dantuma et al., 2000), HA-?CD3 and HA-?
chain of the T-cell receptor (?TCR) were on pcDNA 3.1 (Yang et al., 1998; Yu
and Kopito, 1999), and ?1-antirtrypsin Hong Kong mutant was on pCMV
(Hosokawa et al., 2003). All the remaining reagents were from Sigma-Aldrich
(St. Louis, MO).
Cell Culture and Establishment of Stable Cell Lines
HeLa cells were grown in Advanced DMEM (Invitrogen, Carlsbad, CA)
supplemented with GlutaMAX, antibiotic/antimycotic solution, and 2% fetal
bovine serum (Gemini Bioproducts, Woodland, CA). Plasmids used for trans-
fection were sequenced using CEQ 2000XL DNA analysis system (Beckman
Coulter, Fullerton, CA). Transfection was carried on using Lipofectamine
2000 according to the manufacturer’s instructions (Invitrogen). After transient
transfection, HeLa cells were used for the production of stable cell lines by
selection with Geneticin (Invitrogen). All clones that expressed a given pro-
tein showed accumulation of that protein after inhibition of the proteasome
(our unpublished data). One clone from each group (UbG76VGFP, Ub-R-GFP,
?CD3, and ?1AT) with the highest basal expression level was selected for the
study. None of these clones differed from the nontransfected or mock-trans-
fected controls in morphology, growth characteristics, or time- and dose-
dependent sensitivity to proteasome inhibitors (our unpublished data).
RNA Interference
Small interfering RNAs (siRNAs) were obtained by chemical synthesis using
2?-ACE chemistry (Hartsel et al., 2005) from Dharmacon RNA Technologies
(Lafayette, CO). siRNAs were 2? deprotected, desalted, purified by PAGE,
and duplexed by the manufacturer. The mass of each siRNA was verified by
matrix-assisted laser desorption ionization/time of flight mass spectrometry.
After shipment in a dry form, the siRNAs were suspended in the 1? universal
buffer (20 mM KCl, 6 mM HEPES-KOH, pH 7.5, and 0.2 mM MgCl2) at a 20
?M concentration, aliquoted, and frozen at ?20°C for further use. Two
siRNAs were obtained targeting VCP and one siRNA targeting enhanced
green fluorescent protein (EGFP), a protein not found in HeLa cells. The first
siRNA (positions 599–619 of human VCP mRNA; accession number
NM_007126), called VCP-2, has been used previously, and it was originally
selected from five different siRNAs based on its efficiency (Wojcik et al., 2004a,
b). The second siRNA targeting VCP (positions 480–500), called VCP-6, was
designed using Dharmacon’s Web site siRNA design center (Reynolds et al.,
2004). As control for nonspecific effects of RNAi, we have designed and used
the siRNA targeting EGFP (positions 1101–1018 of CVU55763, preceded by
AA). RNAi was performed by single Oligofectamine-mediated transfection (In-
vitrogen) as described previously (Wojcik et al., 2004a, b). Cells were collected
72 h after the transfection. HeLa cells mock transfected with EGFP siRNA served
as a control.
Transmission Electron Microscopy
HeLa cells were grown on glass slides and submitted to RNAi of VCP by
using either VCP-2 or VCP-6. Three days after RNAi, cells were fixed in 2%
glutaraldehyde in a cacodylate buffer supplemented with 5 mM CaCl2, post-
fixed with OsO4in the cacodylate buffer supplemented with CaCl2and
K4[Fe(CN)6], and then dehydrated with ethanol and acetone and embedded
in LR White resin (Sigma-Aldrich). Resin blocks were cut, mounted on Form-
var carbon-coated grids, counterstained with lead citrate and uranyl acetate,
and observed in a Jeol JEM-100S electron microscope (Jeol, Tokyo, Japan).
Immunofluorescence Microscopy
HeLa cells were grown in Lab-Tek two-chamber slides (Nunc Nalgene, Na-
perville, IL). After 16-h treatment with 10 ?g/ml tunicamycin or 5 ?M
brefeldin A (BFA) either alone or in combination with 10 ?M MG132, cells
were fixed with 2% formaldehyde in phosphate-buffered saline for 30 min,
quenched in 50 mM NH4Cl, permeabilized in 0.1% Triton X-100, washed
twice for 15 min each with Tris-buffered saline (TBS), pH 7.6, supplemented
with 0.1% bovine serum albumin and 0.1% fish gelatin, and incubated with
anti-polyubiquitin FK1 mAb (BIOMOL Research laboratories, Plymouth
Meeting, PA) diluted in the same buffer containing Tween 20 for 2 h. After
three 15-min washes in TBS with 0.1% bovine serum albumin and 0.1% fish
gelatin, the cells were incubated with secondary rhodamine-conjugated anti-
mouse F(ab?)2fragment (Jackson ImmunoResearch Laboratories, West Grove,
PA). Cells were then stained with 100 nM Yo-Pro1 iodide (Invitrogen). After
two washes in TBS, cells were mounted using Gel/Mount (Biomeda, Foster
City, CA). Slides were observed using the 60? Plan Apo objective of a Nikon
Eclipse TE2000-U epifluorescence microscope. Images were acquired using
the CoolSNAP ES charge-coupled device camera operated by the MetaMorph
6.3 software (Fryer Company, Cincinnati, OH).
RT-PCR
RNA was isolated using the modified method of Chomczynski (Chomczynski
and Sacchi, 1987) from HeLa cells 72 h after transfection with two different
siRNAs targeting VCP (VCP-2 and VCP-6) or from cells treated for 6 h with
10 ?M MG132, 10 ?g/ml tunicamycin, and 5 ?M brefeldin A (all from
Calbiochem, San Diego, CA). RT-PCR was performed with the OneStep kit
(QIAGEN, Valencia, CA) by using the pairs of primers amplifying the genes
of interest, as indicated in Table 1. For the XBP-1 transcript, the primers
amplify the region that includes the 26-base pair deletion dependent on IRE-1
endonuclease activity (Yoshida et al., 2001). The number of cycles was ad-
justed to obtain a linear range of reaction products. Gels were scanned with
the Kodak 4000MM Image Station (Eastman Kodak, Rochester, NY).
SDS-PAGE and Western Blotting
SDS-PAGE and Western blotting was conducted for indicated proteins as
described previously (Wojcik et al., 2004b). Primary antibodies were detected
using horseradish peroxidase-conjugated anti-mouse and anti-rabbit antibod-
ies from Jackson ImmunoResearch Laboratories. Horseradish peroxidase was
detected using the ECL Advance kit (GE Healthcare, Piscataway, NJ). Images
were acquired with Kodak 4000MM Image Station. Densitometry was per-
formed using Image Quant version 5.2 (GE Healthcare).
Preparation of RNA and Hybridization of Spotted
Microarrays
Preliminary experiments established 72 h posttransfection as the optimal time
for RNAi of VCP (Wojcik et al., 2004a, b) Therefore, total RNA was isolated
72 h after transfection with siRNA. We used the modified method of Chom-
czynski after lysis in TRIzol (Invitrogen) (Chomczynski and Sacchi, 1987). The
quality of RNA was assessed by spectrophotometric analysis (measuring
OD260/280ratio) and by BioAnalyzer (Agilent Technologies, Palo Alto, CA).
Samples were transferred to the Microarray Core Facility at University of
Texas Southwestern Medical Center (Dallas, TX) where microarray experi-
ments were performed using human 35k spotted oligonucleotide arrays ver-
sion 3.0.2 (Operon Biotechnologies, Huntsville, AL). The samples were fluo-
rescently labeled with either Cy3-dCTP or Cy5-dCTP (GE Healthcare). The
labeled probes were mixed with preheated ASAP hybridization buffer
(PerkinElmer Life and Analytical Sciences, Boston, MA), and hybridized to an
oligo array according to the manufacturer’s instruction. The slides were
washed with SSC buffer from low to high stringency and scanned by GenePix
scanner (Molecular Devices, Sunnyvale, CA) at 532 nm (Cy3) and 635 nm
(Cy5). The Cy3 and Cy5 scans for each slide were superimposed, and the
fluorescent ratio for each spot was obtained.
Design of Microarray Study
Transfection with any siRNA may induce certain sequence-independent ef-
fects, regardless of the lack or presence of specific effects (Sledz et al., 2003). To
VCP in ER Stress
Vol. 17, November 2006 4607

Page 3

increase specificity of analysis, we used two independent siRNAs against
VCP and one against, an irrelevant control target (EGFP), and compared
results with those from cells treated with Oligofectamine only. RNA isolated
from each experimental group was divided equally in two, with one-half of
treated samples and one-half control samples labeled with Cy3 and Cy5, to
randomize effects of the dye bias (Dobbin et al., 2003; Rosenzweig et al., 2004).
To minimize other unknown variables, repeat experiments were performed
on different days by using identical reagents. In total, we generated six
microarray data sets for VCP knockdown by using VCP-2, four microarray
data sets by using VCP-6, and four microarray data sets by using siRNA
targeting EGPF.
Data Processing
The preliminary data analysis (flagging low-quality spots) was performed
using Gene Pix 3.0 Prosoftware (Molecular Devices). Local Lowess normal-
ization was done using Gene Traffic Duo software (Iobion Informatics, La-
Jolla, CA). Before further analysis, we rejected all spots flagged by Gene
Traffic software. Further analysis was conducted using in-house custom C??
programs and Perl scripts developed for this project. We rejected all points
with raw signal intensity ?100 (in either red or green channel) and any point
with sum of normalized green and red channel intensities smaller than 25 (to
avoid meaningless high fold ratios). To make fold ratio distribution closer to
Gaussian (normal), we converted fold ratios to logarithmic scale. After loga-
rithm transformation, for each data set, we subtracted the average measure-
ment from each probe measurement, thus setting the average log-ratio in each
experiment to zero. We then performed principal component analysis, in-
spected the data, and observed that they did not reveal any structure. Because
the data did not contain distinct clusters, and the distribution was close to
normal, we examined outliers to identify transcripts of genes significantly
changed during our experiments. We computed mean values and standard
deviations of measurement for each experiment group (xexpand ?exp) as well
as a mean value and a SD of each transcript measurement during each
knockdown (xtrand ?tr). For ideal measurement, we would obtain outliers
with 95% confidence by selecting as altered those transcripts that are more
than ?expdistant from the average during this experiment:
?xtr? xexp? ? 2?exp
However, because the individual transcripts ratios contain errors of their
own, even if the average ratio measured for a given transcripts, xtr, lies farther
than 2?expfrom the average result in this experiment, xexp, the measurement
still may be not a true outlier.
For example, if ?tr? 5?exp, the accuracy of the transcript log-ratio mea-
surement is clearly not sufficient to determine whether it is altered signifi-
cantly. To take into account a varying accuracy of individual transcript
log-ratio measurements, we used a conservative criterion of defining a tran-
script as up- or down-regulated only if its log-ratio together with its 95%
confidence interval (corresponding to ca. 2?) lies in the outlier region:
?xtr? xexp? ? 2?tr? 2?exp
Thus, our method eliminates transcripts with high fold ratios, which may
havebeencausedbyhighexperimentalerror.Becauseweperformedatleastfour
replicates of each experiment, we also used information about variation between
experiments to judge the quality and reproducibility of the results. We define a
transcriptasup-regulatedifitssignificance(i.e.,1minusprobability)ofbeingup-
or down-regulated is ?0.05. The significance is computed as follows:
SUP(xtr) ? 1/2{1 ? erf((xexp? 2?exp? xtr)/(?2?tr))}
where erf is the error function. The probability of a gene being down-
regulated is computed analogously. To define transcripts that remain un-
changed during knockdown, we assumed the measured transcript log-ratio,
together with its 95% confidence interval, lies within 95% confidence interval
of the measured mean log-ratio during this experiment. The significance is
computed as follows:
SSTABLE(xtr) ? 1/2 ? 1/4{1 ? erf((xtr? 2?exp? xexp)/(?2?tr))}
? 1/4{1 ? erf((xtr? 2?exp? xexp)/(?2?tr))}
RESULTS
RNAi Causes Few Nonspecific Effects on the
Transcriptome Level
To gain insight into cellular roles of VCP, we determined the
effects of decreased levels of VCP on the gene expression
profile of HeLa cells subjected to RNAi of VCP. To discrim-
inate between specific and off-target effects, we used two
different siRNAs against VCP and a control siRNA against
EGFP that does not match any sequence in the human
genome (Jackson et al., 2003). Expression profiles of cells
transfected with all three siRNAs were compared with con-
trol cells treated with the transfection reagent alone. Tran-
scription profiles were analyzed using the strict criteria out-
lined in Materials and Methods. From 34,993 spots on the
microarrays, 9637 produced high-quality data in all 12 hy-
bridizations and were therefore analyzed further. The com-
plete results of the microarray experiments have been depos-
ited in the public Arrayexpress database (http://www.ebi.ac.
uk/arrayexpress, accession no. E-MEXP-817). Transcripts that
were up- or down-regulated after RNAi of VCP with both
siRNAs with at least 95% confidence level and were not sig-
nificantly up- or down-regulated after siRNA of EGFP were
considered specific for the VCP knockdown. In contrast, tran-
scripts that were altered after transfection with siRNAs for
both VCP and EGFP were considered nonspecific, probably
reflectingasequence-independentcellularresponsetosiRNAs.
Only one transcript was down-regulated and three transcripts
were up-regulated after transfection by each of these three
siRNAs (Table 2). These results indicate that siRNA transfec-
tion per se causes remarkably few nonspecific changes in the
geneexpressionprofile.Thisconclusion,however,islimitedby
the single time point of our analysis. Thus, it is possible that
some transcripts were altered at earlier times but returned to
normal levels after 72 h.
RNAi of VCP Down-Regulates Expression of a Limited
Number of Transcripts
RNAi of VCP resulted in a 2.2-fold down-regulation of the
VCP transcript, verifying the effectiveness of RNAi against
this target (Table 3). Six other transcripts were down-regu-
Table 1. Primers used for semiquantitative RT-PCR
Gene nameAccession no.Forward primer Reverse primer
XBP1
VCP
?5
Rpt2
GADD45
BiP
GDF15
ATF3
CTH
IL18
Actin
AB076384
NM?007126
BC057840
NM?002802
NM?001924
X87949
NM?004864
NM?004024
NM?153742
NM?001562
NM?001101
CCTTGTAGTTGAGAACCAGG
TGGAGTTCAAAGTGGTGGAAA
GCTTCGAAATAAGGAACGCA
AACCAAACCTCAGCCACTTTC
TTTTGCTGCGAGAACGACAT
ACGTGGAATGACCCGTCTGT
AAGAACTCAGGACGGTGAATG
TCCTGGGTCACTGGTGTTTGA
TTGCCCAGTTCCTGGAATCTA
ATGGCTGCTGAACCAGTAGAA
TTCCTTCCTGGGCATGGAGT
GGGGCTTGGTATATATGTGG
ATGGCAGGAGCATTCTTCTCA
ATTGTCACTGGAGACTCGGAT
TTAAGGCCATCAGACCAGCTT
ACTGGAACCCATTGATCCAT
ATGAAGTGTTCCATGACACGC
AAGAACTCAGGACGGTGAATG
TTCTTGTTTCGGCACTTTGC
TGCTGCCTTCAAAGCTTGAT
AATAAATATGGTCCGGGGTG
ATCCACATCTGCTGGAAGGT
C. Wo ´jcik et al.
Molecular Biology of the Cell4608

Page 4

lated in response to RNAi of VCP. Two of the down-regu-
lated sequences encode unknown proteins, whereas the oth-
ers encode proteins that do not seem to be functionally
related. Two are receptors located at the plasma membrane,
whereas two are cytosolic proteins. One cytosolic protein
contains an F-box and therefore is a putative component of
an SCF-type ubiquitin ligase complex. To further analyze the
specificity of these effects, we compared the sequences of
each down-regulated transcript with the sequences of each
VCP siRNA. Partial matching of both siRNAs with the tran-
scripts was detected (Table 4), raising the question of
whether down-regulation of these transcripts might result
from off-target effects (Jackson et al., 2003). Additional anal-
ysis of possible off-target sequences by using the Web engine
http://rnai.cs.unm.edu/offTarget (Qiu et al., 2005) revealed
11 possible off-target transcripts for the VCP-2 siRNA with
an off-target score ?35, but only one off-target transcript for
the VCP-6 siRNA with an off-target score ?35. This differ-
ence may be due to the different algorithms used to design
these siRNAs (Elbashir et al., 2001; Reynolds et al., 2004).
Nevertheless, none of the possible off-target sequences identi-
fied by this search corresponded to the transcripts down-reg-
ulated by both VCP siRNAs. Moreover, none of the predicted
11 off-target transcripts for VCP-2 was found among the eight
Table 2. Transcripts nonspecifically up- or down-regulated after transfection with any tested siRNA
NameId FoldSign. DescriptionLocalization
SERPINE2
NRG1
IDS
CABLES1
NM?006216
NM?013959
NM?000202
NM?138375
2.9
1.9
2.1
4.6 ? 10?6
4.5 ? 10?6
3.0 ? 10?10
0.010
Protease nexin 1
Neuregulin 1
Iduronate 2 sulfatase
Cdk5 and Abl enzyme substrate
EC
PM
L
N, membranes
?1.5
Name, gene name; Id, systematic id (RefSeq for mRNA or GenBank accession no.); Fold, average fold change for all VCP knockdown
experiments; Sign., significance (as explained in Materials and Methods) that a given transcript is more than 1.96 ? down-regulated and stable
during control knockdown; PM, plasma membrane; N, nucleus; L, lysosomes; and EC, extracellular, secreted.
Trancripts up- or down-regulated by RNAi of VCP and control EGFP. Transfection of HeLa cells with siRNAs was conducted as described
in Materials and Methods.
Table 3. Transcripts specifically down-regulated after RNAi of VCP with two different siRNAs
Name IdFold Sign.Description Localization
— XM?378620
?2.31.9 ? 10–5Hypothetical protein, predicted by PSORTII to be mitochondrial
(60.9%), nuclear (34.8%), or ER (4.3%)
Valosin-containing protein
Hypothetical protein, predicted by PSORTII to be nuclear
(47.8%) or mitochondrial (34.8%)
?-2Aadrenergic receptor
? subunit of platelet-activating factor acetylhydrolase, isoform Ib
Vasoactive intestinal polypeptide receptor 2
F-box only protein 5
?
VCP
—
NM?007126
AK091343
?2.2
?1.7
0.02
0.010
C, N
?
ADRA2A
PAFAH1B2
VIPR2
FBP5
NM?000681
BC001774
NM?003382
NM?012177
?1.5
?1.5
?1.4
?1.4
0.03
0.010
4.5 ? 10-3
0.05
PM
C
PM
C, N
Name, gene name; Id, systematic id (RefSeq for mRNA or GenBank accession no.); Fold, average fold change for all VCP knockdown
experiments; Sign., significance (as explained in Materials and Methods) that a given transcript is more than 1.96 ? down-regulated and stable
during control knockdown; PM, plasma membrane; C, cytoplasm; N, nucleus; L, lysosomes; EC, extracellular, secreted; and ?, unknown.
Transcripts specifically down-regulated by RNAi of VCP.
Table 4. Alignment of the two different siRNAs targeting VCP with transcripts specifically down-regulated after RNAi of VCP
XM?378620
VCP2 siRNA
AK091343
VCP2 siRNA
ADRA2A
VCP2 siRNA
PAFAH1B2
VCP2 siRNA
VIPR2
VCP2 siRNA
FBP5
VCP6 siRNA
TCTCGTGTTTGATTATATCCTG
TGTAGGGTATGATGACAT—TG
TGTAGCCTA–ATGA-ATTG
TGTAGGGTATGATGACATTG
-GTTGAG-ATCATGTCATTG
TGTAGGGTATGATGACATTG
TGTGGTGTATGA-G-CATTG
TGTAGGGTATGATGACATTG
TGTAGGGTTTG–GACA—G
TGTAGGGTATGATGACATTG
-GTAAAACCTGATGACATTG
TAACCTTCGTGTACGCCTA
XM?378620
VCP6 siRNA
AK091343
VCP6 siRNA
ADRA2A
VCP6 siRNA
PAFA
VCP6 siRNA
VIPR2
VCP6 siRNA
FBP5
VCP2 siRNA
TTACCATA-TGATACAGCCTA
TAACCTTCGTG-TAC-GCCTA
TAACCTAGCAGTACTCC-A
TAACCTTCGTGTACGCCTA
TCCCCTTCTTCTTCACCTA
TAACCTTCGTGTACGCCTA
TATTCCTTCGCCCACGCATT
TAA-CCTTCGTGTACGCCTA
TCACCTTGGTTTGCAAAACCCAT
TAACCTTCGTGT—–ACGCCT
TCAACTTC-TGGA-GTCTA
TGTAGGGTATGATGACATTG
Alignment of the two different siRNAs targeting VCP (denoted VCP-2 and VCP-6 siRNAs) with down-regulated transcripts.
VCP in ER Stress
Vol. 17, November 20064609

Page 5

transcripts down-regulated by VCP-2 alone (our unpublished
data). These results suggest that RNAi of VCP results in de-
creased expression of a minimal number of genes. Most if not
all down-regulated transcripts presented in Table 3 are specif-
ically down-regulated in response to low cellular VCP levels.
RNAi of VCP Up-Regulates Expression of Multiple
Transcripts
RNAi of VCP specifically up-regulated transcripts of 28
genes (Table 5). The up-regulation of three transcripts,
HERP, INSIG2, and SAT, was detected independently by
two different oligonucleotides present in the microarray. In
contrast to the identified down-regulated genes, many up-
regulated transcripts encode functionally or structurally re-
lated proteins. For example, 12 transcripts (46%) encode
proteins that are either secreted or reside in compartments
of the secretory pathway. Moreover, many of these proteins
are known to be up-regulated by ER stress and/or may
participate in the unfolded protein response. Four other
cytosolic proteins associate with the cytoplasmic face of
cellular membranes. Ten transcripts are known to be in-
volved in various forms of cellular stress, including ER
stress and oxidative stress, and five transcripts encode pro-
teins involved in apoptosis. Interestingly, none of the up-
regulated transcripts is known to physically interact with
VCP and with the exception of one putative ubiquitin-spe-
cific hydrolase, none is a recognized component of the UPS.
We performed semiquantitative RT-PCR for selected tran-
scripts to verify their up-regulation by RNAi of VCP (Figure
1A). The mRNA levels of GADD45, GDF15, ATF3, CTH, and
IL18 were increased by each siRNA against VCP (Figure
1A). These results are in excellent accord with the microar-
ray data and strongly support the conclusion that RNAi of
VCP up-regulates expression of these transcripts. We have
Table 5. Transcripts specifically up-regulated after RNAi of VCP with two different siRNAs
Name Id FoldSign. Description Localization
GDF15 NM?0048644.60.050 Growth differentiation factor 15 (macrophage
inhibitory cytokine 1, MIC1)
Very low density lipoprotein receptor
Activation transcription factor 3, involved in stress
responses
Harakiri, BCL2 interacting protein, involved in
apoptosis
Cystathionase, enzyme of transsulfuration
pathway, increased in oxidative stress
Diamine acetyltransferase/Spermidine/spermine
N(1)-acetyltransferase
Myosin VB mRNA
Transforming growth factor, ?2
Ubiquitin-specific protease 53, according to
PSORTII nuclear (73.9%)
Growth arrest and DNA damage-inducible/DDIT1
WD repeat domain, phosphoinositide interacting,
involved in autophagy
interleukin 18
Homocysteine-responsive ER-resident Ub like
Hypothetical protein NP?060840, according to
PSORTII transmembrane (100%)
carbohydrate (chondroitin 4) sulfotransferase 11
Biliverdin reductase B/Flavin reductase
Ephrin type-A receptor 2
Immediate early response 3 (IER3), involved in
apoptosis
Solute carrier family 2, facilitiated glucose
transporter member 1
Protein kinase C, ? type
intercellular adhesion molecule 2
KIAA0265, has 3 Kelch motifs, according to
PSORTII cytoplasmic (73.9%)
MYC associated factor X (MAX), transcript variant
4, transcription factor, binds ATF3 and GADD45
promoters
Insulin induced gene 2, block processing of SREBP
by binding SCAP
Epithelial membrane protein 1
Tryptophanyl-tRNA synthetase
Glycoprotein (transmembrane) nmb transcript
variant 2
EC
VLDLR
ATF3
NM?003383
NM?004024
3.3
3.1
2.4 ? 10?6
1.4 ? 10?4
PM
N
HRK NM?003806 2.90.046 C, membrane-associated
CTHNM?153742 2.70.020C
SSATNM?002970 2.64.7 ? 10?5
C
Myosin VB
TGF?2
USP53
L29143
NM?003238
AB037771
2.5
2.4
2.3
2.6 ? 10?6
0.035
0.049
C, membrane-associated
EC
?
GADD45
WIPI1, Atg18
NM?001924
NM?017983
2.2
2.2
0.002N
Autophagosome6.0 ? 10?6
IL18
HERP, MIF1
—
NM?001562
NM?014685
NP?060840
2.2
2.1
2.1
0.031EC
ER
?
1.3 ? 10?4
0.001
CHST11
BLVRP
EPHA2/ECK
IER3
NM?018413
NM?000713
NM?004431
NM?003897
2.0
1.9
1.9
1.9
0.003
0.025
0.035
0.047
Golgi
C
PM
N (PML bodies)
SLC2A1, GLUT1NM?0065161.80.011 PM
PKCA, PRKCA
ICAM2
—
NM?002737
NM?000873
D87454
1.8
1.8
1.8
0.003
0.038
0.003
C, membrane-associated
PM
?
MAXNM?145114 1.80.013N
INSIG2 NM?0161331.80.048 ER
EMP1
WARS
GPNMB
NM?001423
NM?004184
NM?002510
1.7
1.7
1.6
0.034
0.002
0.036
PM
C
PM
Name, gene name; Id, systematic id (RefSeq for mRNA or GenBank accession no.); Fold, average fold change for all VCP knockdown
experiments; Sign., significance (as explained in Materials and Methods) that a given transcript is more than 1.96 ? down-regulated and stable
during control knockdown; PM, plasma membrane; C, cytoplasm; N, nucleus; L, lysosomes; EC, extracellular, secreted; and ?, unknown.
Transcripts specifically up-regulated by RNAi of VCP.
C. Wo ´jcik et al.
Molecular Biology of the Cell4610

Page 6

also performed semiquantitative RT-PCR for selected ER
stress-related transcripts that were not detected by our
screen, in particular BiP, the ?5 subunit of the 20S protea-
some, and the S4/Rpt2 subunit of the PA700. Although
mRNA levels of ?5 were not altered, those of BiP and
S4/Rpt2 were increased. These transcripts are present in our
microarray data set, but they did not pass our strict criteria
for significant change. BiP was rejected because one of four
control microarrays yielded a flagged spot, even though it
was significantly increased (2.3-fold, significance of 1.3 ?
10?3) in other RNAi experiments. S4/Rpt2 (NM_002802)
showed a consistent pattern of expression characteristic of
specifically up-regulated genes, but the low magnitude of
change (average 1.4-fold difference) did not meet our criteria.
Several other AAA protein subunits of the 26S proteasome
showed similar changes (see original data).
RNAi of VCP Induces the UPR
VCP has an established but incompletely defined role in
various aspects of ERAD (Bar-Nun, 2005; Romisch, 2005).
However, the role of VCP in ERAD has been studied more
extensively in yeast than in mammals. We previously dem-
onstrated that RNAi of VCP in HeLa cells induces an accu-
mulation of polyubiquitinated proteins and promotes exten-
sive cellular vacuolization due to swelling of the endoplasmic
reticulum (Wojcik et al., 2004b). Transmission electron micros-
copy of HeLa cells subjected to RNAi of VCP confirms and
extends the latter finding, demonstrating the presence of dis-
tended vacuole-like membrane-limited compartments proba-
bly corresponding to swollen ER cisternae identified previ-
ously by the immunofluorescent labeling (Figure 2, C and D).
Surprisingly, the lumen of the distended cisternae has a very
low electron density, suggesting an osmotic mechanism of
Figure 1.
siRNA as described in Materials and Methods. (A) RT-PCR was performed for the indicated messages from cells subjected to RNAi and from
cells treated for 6 h with 10 ?M MG132, 10 ?g/ml tunicamycin, and 5 ?M brefeldin A. Arrows show spliced and unspliced transcripts of XBP-1.
(B) Western blotting was performed for the indicated proteins from cells subjected to RNAi against VCP and from cells treated with 10 ?M
MG132 for 6 h. (C) Effect of RNAi on levels of VCP and polyubiquitinated proteins. HeLa cells were subjected to RNAi of VCP with the
indicated siRNAs or treated with 10 ?M MG132 for 6 h. Cell lysates were Western blotted for VCP (top) or polyubiquitin (bottom). Blots were
quantified, and data are expressed as a percentage of control values. Results represent means ? SEM from five independent experiments.
RNAi of VCP induces ER stress and induction of UPR. HeLa cells were subjected to RNAi of VCP by using either VCP-2 or VCP-6
VCP in ER Stress
Vol. 17, November 20064611

Page 7

swelling rather than one caused by an accumulation of protein
aggregates within the ER lumen as a consequence of inhibited
ERAD. In contrast, proteasome inhibition with MG132 induces
the formation of electron-dense cytosolic aggregates (Figure
2B) as observed previously (Wojcik et al., 1996). Nevertheless,
the dramatic alterations in cell morphology caused by RNAi of
VCP are probably associated with induction of ER stress and
promotion of the UPR. Interestingly, vacuolization is not
caused by all inducers of ER stress. For example, brefeldin A
promotes vacuolization, whereas tunicamycin does not (Figure
3). These data suggest that the induction of UPR by RNAi of
VCP may reflect a BFA-sensitive function of VCP in membrane
fusion rather than an effect of VCP on accumulation with
unfolded proteins.
To determine whether RNAi of VCP specifically induced
UPR, we assayed IRE-1–dependent splicing of mRNA en-
coding the transcription factor XBP-1, a diagnostic feature of
UPR (Yoshida et al., 2001). RNAi of VCP by each siRNA
reduced VCP protein levels by more than 85% and consis-
tently induced splicing of XBP-1 mRNA (Figure 1A). Al-
though the magnitude of RNAi-induced XBP-1 splicing was
somewhat variable in different experiments, it was similar to
that promoted by the proteasome inhibitor MG132, but not
as great as that caused by established inducers of ER stress
such as tunicamycin or brefeldin A. RNAi of VCP caused a
twofold increase in the levels of polyubiquitinated proteins,
and increased levels of BiP, an ER chaperone whose expression
increases as part of the UPR, but it had no effect on levels of
cytosolic hsp70, which was increased by MG132 (Figure 1B).
RNAi of VCP had no effect on the levels of phosphorylated
eukaryotic initiation factor 2? (eIF2?), which was decreased by
the treatment with MG132. Phosphorylation of eIF2? is a
Figure 2.
as described in Materials and Methods, or treated with MG132 for 6 h before processing for transmission electron microscopy. (A) Control cells.
(B) MG132-treated cells. Arrows show aggregation of high-electron-density cytosolic material. (C) RNAi of VCP by using VCP-2 siRNA. (D)
RNAi of VCP by using VCP-6 siRNA.
RNAi of VCP promotes vacuolization of cells. HeLa cells were transfected with siRNAs directed against EGFP (control) or VCP
C. Wo ´jcik et al.
Molecular Biology of the Cell 4612

Page 8

downstream event in multiple stress signaling pathways, in-
cluding ER stress and UPR activation (Wek et al., 2006). The
decrease in eIF2a phosphorylation after proteasome inhibition
may be a consequence of reduced degradation of GADD34, a
known phosphatase of eIF2a, or of a direct inhibition of PERK
(Nawrocki et al., 2005). These results directly demonstrate that
reduction of VCP levels by RNAi induces some, but not all
features of UPR, without inducing a cytosolic stress response.
In contrast, proteasome inhibition or induction of ER stress by
either tunicamycin or BFA did not induce up-regulation of
VCP mRNA or protein (Figure 1A).
RNAi of VCP Differentially Affects Degradation of
Various UPS Substrates
VCP is implicated in the degradation of both cytosolic and
ER proteins (Dai et al., 1998; Dai and Li, 2001; Ye et al., 2001).
To further examine the role of VCP in the degradation of
various cellular proteins, we engineered HeLa cell lines
that stably express five established substrates of the UPS:
Ub-G76V-GFP, a cytosolic substrate of the ubiquitin-fusion
degradation (UFD) pathway (Johnson et al., 1995; Dantuma et
al., 2000); R-GFP, a rapidly degraded cytosolic substrate of
the N-end rule pathway (Bachmair et al., 1986; Dantuma et
al., 2000); ?TCR and ?CD3, two different ER transmembrane
subunits of the T-cell receptor subject to ERAD (Yu et al.,
1997; Yang et al., 1998; Yu and Kopito, 1999; Tiwari and
Weissman, 2001); and ?1-antitrypsin Hong Kong mutant, a
misfolded lumenal ER protein subject to ERAD (Hosokawa
et al., 2001; Hosokawa et al., 2003) (Figure 4A). Treatment of
each cell line with MG132 caused time-dependent accumu-
lation of the respective protein, thereby confirming the role
of the UPS in its degradation (Figure 4). Because overexpres-
sion of misfolded proteins in the ER may promote ER stress
and constitutive UPR, we examined the status of XBP-1
splicing in each cell line. XBP-1 splicing did not differ sig-
nificantly in any cell line compared with the parental non-
transfected HeLa cells (Figure 4B). Moreover, these cells had
normal morphology and growth characteristics (our unpub-
lished data). Thus, chronic expression of transfected proteins
did not seem to induce features of ER stress.
Figure 3.
is not always associated with cell vacuolization.
Immunofluorescence images of HeLa cells labeled
by the FK1 anti-ubiquitin antibody (red) and the
nuclear dye YoPro iodide (green). Cells were sub-
mitted to a 16 h treatment with 5 ?M BFA and 10
?g/ml tunicamycin, either alone or in combination
with 10 ?M MG132. Although both MG132 and
BFA induce cell vacuolization, probably by disten-
sionofERcistaernae(arrows),tunicamycininduces
an elongated cellular phenotype without inducing
vacuoles, which can be induced by a cotreatment
with MG132.
Pharmacological induction of ER stress
VCP in ER Stress
Vol. 17, November 2006 4613

Page 9

Figure 4.
in this study. (B) Overexpression of UPS substrates does not induce UPR in stable cell lines as assessed by XBP-1 splicing. (C) HeLa cells stably
Differential effects of RNAi of VCP on the degradation of UPS substrates. (A) Schematic representation of five UPS substrates used
C. Wo ´jcik et al.
Molecular Biology of the Cell 4614

Page 10

Each cell line was subjected to RNAi of VCP with two
different siRNAs (Figure 4C). RNAi of VCP increased levels
of polyubiquitinated cellular proteins and promoted UPR in
each cell line, similarly to the effects in nonengineered cells
(our unpublished data), but it had divergent effects on the
degradation of the respective expressed proteins. RNAi of
VCP caused a four- to sixfold accumulation of the cytosolic
proteins R-GFP and UbG76VGFP. Analogous results with
different UFD and N-end rule substrates have been reported
previously in yeast expressing mutant VCP homolog
(Cdc48) (Bachmair et al., 1986; Johnson et al., 1995; Ghislain et
al., 1996). RNAi of VCP also inhibited degradation of ?TCR,
in accord with the established role of VCP in ERAD (Figure
4C). Proteasome inhibition with MG132 caused selective
accumulation of a faster migrating (?29-kDa) form of ?TCR,
probably corresponding to the deglycosylated form, as de-
scribed previously (Yu et al., 1997; Huppa and Ploegh, 1997;
Yang et al., 1998; Yu and Kopito, 1999). In contrast, RNAi of
VCP caused accumulation of the slower migrating (?38-
kDa) glycosylated ?TCR. These results suggest that MG132
inhibited degradation of ?TCR after its extraction from the
ER, whereas RNAi of VCP inhibited degradation of ?TCR by
preventing its extraction from the ER. In surprising contrast
to ?TCR, RNAi of VCP did not affect the levels of two other
ERAD substrates, ?1-antitrypsin and ?CD3. These results
support the conclusion that VCP mediates proteolysis in
multiple cellular compartments and suggest that certain es-
tablished ERAD substrates can be processed by VCP-inde-
pendent mechanisms.
DISCUSSION
VCP (aka p97 or Cdc48 in yeast) is an AAA ATPase that has
been implicated in numerous and diverse cellular functions
(Woodman, 2003; Dreveny et al., 2004; Wang et al., 2004;
Bar-Nun, 2005; Halawani and Latterich, 2006). Despite con-
siderable work, a comprehensive view of its biological
role(s) remains elusive. To gain insight into the cellular roles
of VCP, we have conducted a microarray analysis to identify
transcripts altered in response to RNAi of VCP in HeLa cells.
We analyzed the data by a stringent method designed to
reduce spurious positives that devalue some microarray
results. This analysis demanded comparable effects from
each of two different siRNAs in each of four different exper-
iments to designate a transcript as altered. Although these
stringent requirements probably resulted in elimination of
some authentic responses (e.g., BiP and some proteasome
subunits; see below), it produced a tractable list of affected
transcripts. The altered expression of multiple selected tran-
scripts was verified by independent methodology.
Our microarray analysis identified ?30 genes whose ex-
pression is altered by RNAi of VCP. Although these proteins
have diverse functions and many would not necessarily
have been anticipated to emerge in this screen, most have
plausible connections to known or suspected functions of
VCP, including ERAD and ER stress. For example, the most
up-regulated transcript is growth differentiation factor 15
(GDF15), up-regulated 4.6?; GDF15 is induced in different
tissues after multiple types of chemical and physical injury,
including oxidative stress and heat shock (Hsiao et al., 2000;
Zimmers et al., 2005). Several proteins involved in choles-
terol homeostasis also were up-regulated, including the very
low density lipoprotein receptor and insulin-induced pro-
tein 2. Up-regulation of genes involved in cholesterol uptake
and cholesterol biosynthetic pathway have been shown to be
induced by ER stress (Werstuck et al., 2001). Activating
transcription factor 3, a transcription factor activated by
various stress stimuli, including ER stress and proteasome
inhibition (Wek et al., 2006), was up-regulated 3.1-fold. Ho-
mocysteine-responsive ER-resident ubiquitin like protein, a
transmembrane ER ubiquitin-like protein shown previously
to be involved in ERAD (van Laar et al., 2001; Hori et al.,
2004), was up-regulated more than twofold. Contrary to
other reports (Sledz et al., 2003), we have not observed an
interferon-like response to introduction of siRNAs; only four
transcripts out of the 34,993 analyzed by the microarrays
were affected by RNAi in a sequence-independent manner,
thereby validating the specificity of the VCP knockdown.
The most extensively studied function of VCP involves
ERAD, a process by which resident and transient ER pro-
teins can be selectively or constitutively degraded in the
cytoplasm by the UPS (Tsai et al., 2002; Meusser et al., 2005;
Romisch, 2005; Bar-Nun, 2005). VCP in conjunction with
Ufd1 and Npl4 may couple ATP hydrolysis and polyubiq-
uitin chain binding properties to power extraction of pro-
teins from the ER for delivery to and degradation by the 26S
proteasome. Because this process probably involves multi-
ple proteins whose functions are mechanistically linked, we
were somewhat surprised not to identify many components
of the UPS as altered transcripts. USP53, a poorly-character-
ized ubiquitin-specific protease, was the only specifically
up-regulated UPS component, whereas Emi1, an F-box pro-
tein and thus a putative E3 ubiquitin ligase, was the only
down-regulated UPS component. Emi1 is a known regulator
of progression through mitosis (Reimann et al., 2001). There-
fore, its down-regulation may be involved in severe mitotic
abnormalities caused by RNAi of VCP (Wojcik et al., 2004b).
Down-regulation of the ? subunit of acetylhydrolase, a pro-
tein involved in control of intracellular microtubule-depen-
dent motility (Arai, 2002), may be related to the previously
described defect in the formation of aggresomes that are
linked to altered UPS function (Wojcik et al., 2004b). Our
analysis may have failed to detect some UPS components, such
as the subunits of the PA700(19S) proteasome regulatory com-
plex, whose altered expression fell short of our strict criteria.
Alteration of PA700 content may be particularly significant
because PA700 seems to have some similar functions as those
attributed to VCP during ERAD (see below).
Previous work from many investigators has indicated a role
for VCP in other aspects of ER and Golgi function, including
mediation of homotypic membrane fusion (Latterich et al.,
1995; Patel et al., 1998; Rabouille et al., 1998). Here, we provide
direct evidence that RNAi of VCP not only induces UPR but
also alters cellular ultrastructure. We also observed an up-
regulation of several genes involved in ER stress. Despite the
lackofincreasedeIF2?phosphorylation(Weketal.,2006),three
genes involved in the response of cells to protein starvation are
induced: the biosynthetic enzyme tryptophanyl-tRNA syn-
thetase; Atg18 required for autophagy; and spermidine/
spermine N1-acetyltransferase, a rate-limiting enzyme in the
catabolic pathway of polyamine metabolism. Several up-regu-
lated transcripts are different gene products involved in apo-
ptosis, a process induced by VCP knockdown (Wojcik et al.,
2004b). Proapoptotic transcripts included GADD45A, Harakiri,
Figure 4 (cont).
RNAi with VCP-2 and VCP-6 siRNAs or treated with 10 ?M MG132
for 6 h. Cell extracts were subjected to Western blotting for indicated
proteins. (D) Analysis from C was conducted in four independent
experiments. Arrows on C indicate bands that were used for den-
sitometric quantification. Mean values ? SEM are expressed as a
percentage of control.
expressing indicated proteins were subjected to
VCP in ER Stress
Vol. 17, November 20064615

Page 11

EPHA2, and MAX. Oxidative stress also triggers UPR through
a novel pathway involving inactivation of the VCP ATPase
activity by the oxidative modification of Cys522 (Noguchi et al.,
2005). Thus, RNAi of VCP may mimic the cellular effects of
oxidative inactivation of VCP. In fact, we observed up-regula-
tion of transcripts known to be induced by oxidative stress,
such as SSAT (Chopra and Wallace, 1998), cytosolic flavin
reductase (Sedlak and Snyder, 2004), and CTH (cystathionin-
?-lyase) (Ishii et al., 2004). These effects may result directly from
decreased VCP levels or may be secondary to oxidative stress
resulting from accumulation of misfolded ERAD substrates
(Haynes et al., 2004).
VCP was required for degradation by the UPS of two
cytoplasmic proteins, R-GFP and Ub-G76V-GFP. These re-
sults are in accord with findings in yeast, where VCP ho-
molog (Cdc48) has been shown to be required for the deg-
radation of substrates of the N-end rule and UFD pathways
(Johnson et al., 1995; Ghislain et al., 1996). RNAi of VCP also
inhibited degradation of a prototypical ERAD substrate,
?TCR. Accumulation of the fully glycosylated form of ?TCR
indicates that lack of VCP delays its extraction from the ER
membrane. In contrast, we failed to detect altered degrada-
tion of two other ERAD substrates, the lumenal ?1-antitryp-
sin and the transmembrane ?-CD3 glycoproteins. The reason
for the different sensitivity of these ERAD substrates to VCP
depletion is unclear. It is possible that the localization (lu-
menal ?1-antitrypsin versus transmembrane ?TCR and
?CD3) of proteins, the size of the domain in each compart-
ment (short cytosolic portion in ?TCR and large cytosolic
portion in ?CD3), and the nature of polyubiquitin chain
linkages determine functional interactions with VCP. Previ-
ous work demonstrated that when cells are treated with
proteasome inhibitors, at least some ?TCR is exported to
and accumulates in the cytosol, whereas ?CD3 remains ER
associated (Yang et al., 1998; Tiwari and Weissman, 2001).
Regardless, the current results suggest that VCP is not re-
quired for retrotranslocation of all ERAD substrates. Al-
though it has been recognized that ERAD proceeds by dif-
ferent pathways depending upon the localization of
misfolded domains (Taxis et al., 2003; Vashist and Ng, 2004),
it is assumed that different pathways of proteasome-depen-
dent ERAD converge at a common step requiring VCP (Bar-
Nun, 2005). Our results suggest that even that step may not
be common to all ERAD, and therefore they are in accord
with suggestions of others (Romisch, 2005). Emerging evi-
dence demonstrates that certain ERAD substrates can be
removed from the ER directly by PA700 (19S), the regulatory
cap of the 26S proteasome (Lee et al., 2004). This VCP-
independent process may involve binding of PA700 directly
the Sec61 retrotranslocation channel (Kalies et al., 2005).
PA700 contains a heterohexameric ring of AAA ATPases
and therefore may share important functional features of
VCP in ERAD (Glickman et al., 1998; DeMartino and Slaugh-
ter, 1999; Zhang et al., 2000). Moreover, retrotranslocation of
the cholera toxin A1 chain, which hijacks the retrotransloca-
tion pathway, does not require active VCP (Kothe et al.,
2005). Thus, the role of VCP in ERAD is considerably more
complex than envisioned by established models (Tsai et al.,
2002; Bar-Nun, 2005). Our results may reflect such complex-
ity whereby VCP functions as a partition for the fate of
polyubiquitinated proteins (Halawani and Latterich, 2006).
Thus, VCP may promote degradation of certain proteins but
deubiquitination and salvage of others. The molecular basis
for these distinctions will require additional work.
ACKNOWLEDGMENTS
We acknowledge the generous gifts of pEGFP-N1-Ub-G76V-GFP and pEGFP-
N1-Ub-R-GFP plasmids from Dr. Maria Masucci (Karolinska Institutet, Stock-
holm, Sweden), pcDNA3.1-HA-?CD3 from Allan Weissman (National Insti-
tutes of Health, Bethesda, MD), pCDNA3.1-HA-?-TCR from Dr. Ron Kopito
(Stanford University, Stanford, CA), and pCMV-?1-antitrypsin Hong Kong
from Dr. Nobuko Hosokawa (Kyoto University, Kyoto, Japan). This work was
supported by American Heart Association, Texas Affiliate Grant 0365148Y (to
C.W.), Biomedical Research Grant from Indiana University School of Medi-
cine 22–812-57 (to C.W.), American Cancer Society Grant IRG-84-002-22 (to
C.W.), National Institutes of Health Grant DK-46181 (to G.N.D.), and the
Welch Foundation (to G.N.D.). D.N. is on temporary leave from Department
of Immunology, Medical University of Warsaw, Warsaw, Poland.
REFERENCES
Arai, H. (2002). Platelet-activating factor acetylhydrolase. Prostaglandins
Other Lipid Mediat. 68–69, 83–94.
Bachmair, A., Finley, D., and Varshavsky, A. (1986). In vivo half-life of a
protein is a function of its amino-terminal residue. Science 234, 179–186.
Bar-Nun, S. (2005). The role of p97/Cdc48p in endoplasmic reticulum-asso-
ciated degradation: from the immune system to yeast. Curr. Top. Microbiol.
Immunol. 300, 95–125.
Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation
by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Bio-
chem. 162, 156–159.
Chopra, S., and Wallace, H. M. (1998). Induction of spermidine/spermine
N1-acetyltransferase in human cancer cells in response to increased produc-
tion of reactive oxygen species. Biochem. Pharmacol. 55, 1119–1123.
Dai, R. M., Chen, E., Longo, D. L., Gorbea, C. M., and Li, C. C. (1998).
Involvement of valosin-containing protein, an ATPase Co-purified with I?B?
and 26 S proteasome, in ubiquitin-proteasome-mediated degradation of I?B?.
J. Biol. Chem. 273, 3562–3573.
Dai, R. M., and Li, C. C. (2001). Valosin-containing protein is a multi-ubiquitin
chain-targeting factor required in ubiquitin-proteasome degradation. Nat.
Cell Biol. 3, 740–744.
Dantuma, N. P., Lindsten, K., Glas, R., Jellne, M., and Masucci, M. G. (2000).
Short-lived green fluorescent proteins for quantifying ubiquitin/proteasome-
dependent proteolysis in living cells. Nat. Biotechnol. 18, 538–543.
DeMartino, G. N., and Slaughter, C. A. (1999). The proteasome, a novel
protease regulated by multiple mechanisms. J. Biol. Chem. 274, 22123–22126.
Dobbin, K., Shih, J. H., and Simon, R. (2003). Statistical design of reverse dye
microarrays. Bioinformatics. 19, 803–810.
Dreveny, I., Pye, V. E., Beuron, F., Briggs, L. C., Isaacson, R. L., Matthews, S. J.,
McKeown, C., Yuan, X., Zhang, X., and Freemont, P. S. (2004). p97 and close
encounters of every kind: a brief review. Biochem. Soc. Trans. 32, 715–720.
Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., and Tuschl,
T. (2001). Duplexes of 21-nucleotide RNAs mediate RNA interference in
cultured mammalian cells. Nature 411, 494–498.
Elkabetz, Y., Shapira, I., Rabinovich, E., and Bar-Nun, S. (2003). Distinct steps
in dislocation of luminal ERAD substrates: roles of ER-bound p97/Cdc48p
and proteasome. J. Biol. Chem. 279, 3980–3989.
Ghislain, M., Dohmen, R. J., Levy, F., and Varshavsky, A. (1996). Cdc48p
interacts with Ufd3p, a WD repeat protein required for ubiquitin-mediated
proteolysis in Saccharomyces cerevisiae. EMBO J. 15, 4884–4899.
Glickman, M. H., Rubin, D. M., Coux, O., Wefes, I., Pfeifer, G., Cjeka, Z.,
Baumeister, W., Fried, V. A., and Finley, D. (1998). A subcomplex of the
proteasome regulatory particle required for ubiquitin-conjugate degradation
and related to the COP9-signalosome and eIF3. Cell 94, 615–623.
Halawani, D., and Latterich, M. (2006). p97, the cell’s molecular purgatory?
Mol. Cell 22, 713–717.
Harding, H. P., Calfon, M., Urano, F., Novoa, I., and Ron, D. (2002). Tran-
scriptional and translational control in the Mammalian unfolded protein
response. Annu. Rev. Cell Dev. Biol. 18, 575–599.
Hartsel, S. A., Kitchen, D. E., Scaringe, S. A., and Marshall, W. S. (2005). RNA
oligonucleotide synthesis via 5?-silyl-2?-orthoester chemistry. Methods Mol.
Biol. 288, 33–50.
Haynes, C. M., Titus, E. A., and Cooper, A. A. (2004). Degradation of mis-
folded proteins prevents ER-derived oxidative stress and cell death. Mol. Cell
15, 767–776.
C. Wo ´jcik et al.
Molecular Biology of the Cell 4616

Page 12

Hori, O., et al. (2004). Role of Herp in the endoplasmic reticulum stress
response. Genes Cells 9, 457–469.
Hosokawa, N., Tremblay, L. O., You, Z., Herscovics, A., Wada, I., and Nagata,
K. (2003). Enhancement of endoplasmic reticulum (ER) degradation of mis-
folded Null Hong Kong ?1-antitrypsin by human ER mannosidase I. J. Biol.
Chem. 278, 26287–26294.
Hosokawa, N., Wada, I., Hasegawa, K., Yorihuzi, T., Tremblay, L. O.,
Herscovics, A., and Nagata, K. (2001). A novel ER alpha-mannosidase-like
protein accelerates ER-associated degradation. EMBO Rep. 2, 415–422.
Hsiao, E. C., Koniaris, L. G., Zimmers-Koniaris, T., Sebald, S. M., Huynh, T. V.,
and Lee, S. J. (2000). Characterization of growth-differentiation factor 15, a
transforming growth factor beta superfamily member induced following liver
injury. Mol. Cell. Biol. 20, 3742–3751.
Huppa, J. B., and Ploegh, H. L. (1997). The alpha chain of the T cell antigen
receptor is degraded in the cytosol. Immunity 7, 113–122.
Ishii, I., Akahoshi, N., Yu, X. N., Kobayashi, Y., Namekata, K., Komaki, G.,
and Kimura, H. (2004). Murine cystathionine gamma-lyase: complete cDNA
and genomic sequences, promoter activity, tissue distribution and develop-
mental expression. Biochem. J. 381, 113–123.
Jackson, A. L., Bartz, S. R., Schelter, J., Kobayashi, S. V., Burchard, J., Mao, M.,
Li, B., Cavet, G., and Linsley, P. S. (2003). Expression profiling reveals off-
target gene regulation by RNAi. Nat. Biotechnol. 21, 635–637.
Johnson, E. S., Ma, P. C., Ota, I. M., and Varshavsky, A. (1995). A proteolytic
pathway that recognizes ubiquitin as a degradation signal. J. Biol. Chem. 270,
17442–17456.
Kalies, K. U., Allan, S., Sergeyenko, T., Kroger, H., and Romisch, K. (2005).
The protein translocation channel binds proteasomes to the endoplasmic
reticulum membrane. EMBO J. 24, 2284–2293.
Kostova, Z., and Wolf, D. H. (2003). For whom the bell tolls: protein quality
control of the endoplasmic reticulum and the ubiquitin-proteasome connec-
tion. EMBO J. 22, 2309–2317.
Kothe, M., Ye, Y., Wagner, J. S., De Luca, H. E., Kern, E., Rapoport, T. A., and
Lencer, W. I. (2005). Role of p97 AAA-ATPase in the retrotranslocation of the
cholera toxin A1 chain, a non-ubiquitinated substrate. J. Biol. Chem. 280,
28127–28132.
Latterich, M., Frohlich, K. U., and Schekman, R. (1995). Membrane fusion and
the cell cycle: Cdc48p participates in the fusion of ER membranes. Cell 82,
885–893.
Lee, R. J., Liu, C. W., Harty, C., McCracken, A. A., Latterich, M., Romisch, K.,
DeMartino, G. N., Thomas, P. J., and Brodsky, J. L. (2004). Uncoupling
retro-translocation and degradation in the ER-associated degradation of a
soluble protein. EMBO J. 23, 2206–2215.
Ma, Y., and Hendershot, L. M. (2002). The mammalian endoplasmic reticulum
as a sensor for cellular stress. Cell Stress. Chaperones 7, 222–229.
Meusser, B., Hirsch, C., Jarosch, E., and Sommer, T. (2005). ERAD: the long
road to destruction. Nat. Cell Biol. 7, 766–772.
Meyer, H. H., Shorter, J. G., Seemann, J., Pappin, D., and Warren, G. (2000). A
complex of mammalian ufd1 and npl4 links the AAA-ATPase, p97, to ubiq-
uitin and nuclear transport pathways. EMBO J. 19, 2181–2192.
Meyer, H. H., Wang, Y., and Warren, G. (2002). Direct binding of ubiquitin
conjugates by the mammalian p97 adaptor complexes, p47 and Ufd1-Npl4.
EMBO J. 21, 5645–5652.
Nawrocki, S. T., Carew, J. S., Dunner, K., Jr., Boise, L. H., Chiao, P. J., Huang,
P., Abbruzzese, J. L., and McConkey, D. J. (2005). Bortezomib inhibits PKR-
like endoplasmic reticulum (ER) kinase and induces apoptosis via ER stress in
human pancreatic cancer cells. Cancer Res. 65, 11510–11519.
Noguchi, M., Takata, T., Kimura, Y., Manno, A., Murakami, K., Koike, M.,
Ohizumi, H., Hori, S., and Kakizuka, A. (2005). ATPase activity of p97/
Valosin-containing protein is regulated by oxidative modification of the evo-
lutionally conserved 522nd cysteine residue in Walker A motif. J. Biol. Chem.
280, 41332–41341.
Patel, S. K., Indig, F. E., Olivieri, N., Levine, N. D., and Latterich, M. (1998).
Organelle membrane fusion: a novel function for the syntaxin homolog Ufe1p
in ER membrane fusion. Cell 92, 611–620.
Qiu, S., Adema, C. M., and Lane, T. (2005). A computational study of off-target
effects of RNA interference. Nucleic Acids Res. 33, 1834–1847.
Rabinovich, E., Kerem, A., Frohlich, K. U., Diamant, N., and Bar-Nun, S.
(2002). AAA-ATPase p97/Cdc48p, a cytosolic chaperone required for endo-
plasmic reticulum-associated protein degradation. Mol. Cell. Biol. 22, 626–
634.
Rabouille, C., Kondo, H., Newman, R., Hui, N., Freemont, P., and Warren, G.
(1998). Syntaxin 5 is a common component of the NSF- and p97-mediated
reassembly pathways of Golgi cisternae from mitotic Golgi fragments in vitro.
Cell 92, 603–610.
Rabouille, C., Levine, T. P., Peters, J. M., and Warren, G. (1995). An NSF-like
ATPase, p97, and NSF mediate cisternal regrowth from mitotic Golgi frag-
ments. Cell 82, 905–914.
Reimann, J. D., Freed, E., Hsu, J. Y., Kramer, E. R., Peters, J. M., and Jackson,
P. K. (2001). Emi1 is a mitotic regulator that interacts with Cdc20 and inhibits
the anaphase promoting complex. Cell 105, 645–655.
Reynolds, A., Leake, D., Boese, Q., Scaringe, S., Marshall, W. S., and
Khvorova, A. (2004). Rational siRNA design for RNA interference. Nat.
Biotechnol. 22, 326–330.
Romisch, K. (2005). Endoplasmic reticulum-associated degradation. Annu.
Rev. Cell Dev. Biol. 21, 435–456.
Rosenzweig, B. A., Pine, P. S., Domon, O. E., Morris, S. M., Chen, J. J., and
Sistare, F. D. (2004). Dye bias correction in dual-labeled cDNA microarray
gene expression measurements. Environ. Health Perspect. 112, 480–487.
Sedlak, T. W., and Snyder, S. H. (2004). Bilirubin benefits: cellular protection
by a biliverdin reductase antioxidant cycle. Pediatrics 113, 1776–1782.
Sitia, R., and Braakman, I. (2003). Quality control in the endoplasmic reticu-
lum protein factory. Nature 426, 891–894.
Sledz, C. A., Holko, M., de Veer, M. J., Silverman, R. H., and Williams, B. R.
(2003). Activation of the interferon system by short-interfering RNAs. Nat.
Cell Biol. 5, 834–839.
Sriburi, R., Jackowski, S., Mori, K., and Brewer, J. W. (2004). XBP 1, a link
between the unfolded protein response, lipid biosynthesis, and biogenesis of
the endoplasmic reticulum. J. Cell Biol. 167, 35–41.
Taxis, C., Hitt, R., Park, S. H., Deak, P. M., Kostova, Z., and Wolf, D. H. (2003).
Use of modular substrates demonstrates mechanistic diversity and reveals
differences in chaperone requirement of ERAD. J. Biol. Chem. 278, 35903–
35913.
Tiwari, S., and Weissman, A. M. (2001). Endoplasmic reticulum (ER)-associ-
ated degradation of T cell receptor subunits. Involvement of ER-associated
ubiquitin-conjugating enzymes (E2s). J. Biol. Chem. 276, 16193–16200.
Tsai, B., Ye, Y., and Rapoport, T. A. (2002). Retro-translocation of proteins
from the endoplasmic reticulum into the cytosol. Nat. Rev. Mol. Cell Biol. 3,
246–255.
van Laar, T., van der Eb, A. J., and Terleth, C. (2001). Mif 1, a missing link
between the unfolded protein response pathway and ER-associated protein
degradation? Curr. Protein Pept. Sci. 2, 169–190.
Vashist, S., and Ng, D. T. (2004). Misfolded proteins are sorted by a sequential
checkpoint mechanism of ER quality control. J. Cell Biol. 165, 41–52.
Wang, Q., Song, C., and Li, C. C. (2004). Molecular perspectives on p97-VCP:
progress in understanding its structure and diverse biological functions. J.
Struct. Biol. 146, 44–57.
Wek, R. C., Jiang, H. Y., and Anthony, T. G. (2006). Coping with stress: eIF2
kinases and translational control. Biochem. Soc. Trans. 34, 7–11.
Werstuck, G. H., et al. (2001). Homocysteine-induced endoplasmic reticulum
stress causes dysregulation of the cholesterol and triglyceride biosynthetic
pathways. J. Clin. Invest. 107, 1263–1273.
Wojcik, C., Fabunmi, R., and DeMartino, G. N. (2004a). Modulation of gene
expression by RNAi. In: Hypertension. Methods and Protocols, ed. J. P.
Fennell and A. H. Baker, Totowa, NJ: Humana Press, 381–394.
Wojcik, C., Schroeter, D., Wilk, S., Lamprecht, J., and Paweletz, N. (1996).
Ubiquitin-mediated proteolysis centers in HeLa cells: indication from studies
of an inhibitor of the chymotrypsin-like activity of the proteasome. Eur. J. Cell
Biol. 71, 311–318.
Wojcik, C., Yano, M., and DeMartino, G. N. (2004b). RNA interference of
valosin-containing protein (VCP/p97) reveals multiple cellular roles linked to
ubiquitin/proteasome-dependent proteolysis. J. Cell Sci. 117, 281–292.
Woodman, P. G. (2003). p97, a protein coping with multiple identities. J. Cell
Sci. 116, 4283–4290.
Yang, M., Omura, S., Bonifacino, J. S., and Weissman, A. M. (1998). Novel
aspects of degradation of T cell receptor subunits from the endoplasmic
reticulum (ER) in T cells: importance of oligosaccharide processing, ubiquiti-
nation, and proteasome-dependent removal from ER membranes. J. Exp.
Med. 187, 835–846.
Ye, Y., Meyer, H. H., and Rapoport, T. A. (2001). The AAA ATPase Cdc48/p97
and its partners transport proteins from the ER into the cytosol. Nature 414,
652–656.
Ye, Y., Meyer, H. H., and Rapoport, T. A. (2003). Function of the p97-Ufd1-
Npl4 complex in retrotranslocation from the ER to the cytosol: dual recogni-
VCP in ER Stress
Vol. 17, November 2006 4617

Page 13

tion of nonubiquitinated polypeptide segments and polyubiquitin chains.
J. Cell Biol. 162, 71–84.
Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., and Mori, K. (2001). XBP1
mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to
produce a highly active transcription factor. Cell 107, 881–891.
Yu, H., Kaung, G., Kobayashi, S., and Kopito, R. R. (1997). Cytosolic degra-
dation of T-cell receptor alpha chains by the proteasome. J. Biol. Chem. 272,
20800–20804.
Yu, H., and Kopito, R. R. (1999). The role of multiubiquitination in dislocation
and degradation of the alpha subunit of the T cell antigen receptor. J. Biol.
Chem. 274, 36852–36858.
Zhang, X., et al. (2000). Structure of the AAA ATPase p97. Mol. Cell 6, 1473–1484.
Zimmers, T. A., Jin, X., Hsiao, E. C., McGrath, S. A., Esquela, A. F., and
Koniaris, L. G. (2005). Growth differentiation factor-15/macrophage inhibi-
tory cytokine-1 induction after kidney and lung injury. Shock 23, 543–548.
C. Wo ´jcik et al.
Molecular Biology of the Cell4618