Investigating the role of the Caenorhabditis elegans unfolded protein response
in immunity and development
by
MAssACHU
i
Claire E. Richardson
B.A. Biology
Carleton College, 2006
SUBMITTED TO THE DEPARTMENT OF BIOLOGY
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
ARCHIVES
DOCTOR OF PHILOSOPHY
AT THE
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
FEBRUARY 2011
0 2011 Claire E. Richardson. All rights reserved.
The author hereby grants to MIT permission to reproduce and to distribute publicly paper
and electronic copies of this thesis document in whole or in part in any medium now
known or hereafter created.
Al
Signature of the Author:
Department of Biology
16 September 2011
Certified by:
Dennis Kim
Associate Professor of Biology
Thesis Supervisor
Accepted by:
Lx
Alan D. Grossman
Professor of Biology
Chairman of the Graduate Committee
I
Investigating the role of the Caenorhabditis elegans unfolded protein response
in immunity and development
by
Claire E. Richardson
Submitted to the Department of Biology on
November 21, 20011 in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
Abstract
Proteins destined for the secretory pathway are folded, posttranslationally
modified, and assembled into complexes in the endoplasmic reticulum (ER). To maintain
ER proteostasis, the rate of nascent peptide influx into the ER must be matched with the
rate of protein folding and export. An imbalance between peptide influx and ER folding
capacity activates a conserved set of signal transduction pathways termed the ER
Unfolded Protein Response (UPR), which function to restore ER proteostasis. In
metazoans, the UPR is controlled by three signaling pathways, controlled by the ER
localized transmembrane sensors IRE-1, PERK/PEK-1, and ATF6/ATF-6. The molecular
mechanisms and output of the UPR have been defined largely by exogenously inhibiting
ER protein folding, either chemically or through overexpression of unfoldable mutant ER
proteins, while genetic studies have implicated essential functions for UPR signaling in
normal development and in the pathogenesis of disease.
This work defines an essential role for the UPR in Caenorhabditiselegans in
protection against host immunity and maintenance of ER proteostasis during
development. In Chapter Two, I show that the IRE-1 -XBP- 1 pathway is activated by
infection with the bacterial pathogen Pseudomonasaeruginosaand is essential for larval
development in the presence of pathogen. Through genetic analyses, I demonstrate that
immune activation is necessary and sufficient to activate the IRE-1 -XBP- 1 pathway, and
that the function of the IRE-1 -XBP- 1 pathway during infection is to protect against the
host immune response. In Chapter Three, I present evidence suggesting that the IRE-I
and PEK- 1 negative feedback loops function constitutively to maintain ER proteostasis,
even during growth under standard, "unstressed" conditions. Together, these studies
highlight the integral role of UPR signaling in C. elegans physiology, and future work,
described in Chapters Four and Five, will use genetic approaches to further define the
molecular mechanisms underlying this requirement for UPR activity.
Thesis Supervisor: Dennis Kim
Title: Associate Professor of Biology
4
Acknowledgments
First, I would like to thank Dennis for being a phenomenal advisor. I have been
continuously impressed by and grateful for his enthusiasm, brilliance, patience, and
humor.
I would like to thank all the past and present members of the Kim lab who have
shared their time and knowledge with me while creating a friendly and entertaining
environment in which to work: Annie, Chi-Fong, Dan, Doug, Howard, Jen, Josh, Julie,
Kirthi, Matt M., Matt Y., Odile, Pete, Rob, Stephanie C., Stephanie K., Tristan, and Zo8.
Special thanks to Kirthi and Matt Y. for reading thesis chapters.
I would like to thank the members of my thesis committee for their input and
support. Peter Reddien has been a member of my committee since the beginning, Graham
Walker provided helpful feedback during the early years of the project, and Chris Kaiser
has contributed useful insights in the last year of my graduate work. Thank you also to
Sue Lindquist and Gary Ruvkun for their participation in my thesis defense.
My friends during graduate school have contributed so much to my life outside of
the lab, and I would like to thank them for all of the good times we have shared over the
years. Thank you in particular to my roommates Michelle and Nozomi for four and six
years, respectively, of levity, harmony, and support.
I would also like to thank my family for their support and love.
6
Table of Contents
Title Page .........................................................................................................................
1
Abstract ............................................................................................................................
3
A cknow ledgm ents.......................................................................................................
5
Table of Contents ......................................................................................................
7
C hapter O ne: Introduction............................................................................................
11
Protein folding in the endoplasm ic reticulum ................................................................
14
Figure One ..............................................................................................................
17
The U PR alleviates perturbations in ER protein folding .............................................
21
M echanism s of U PR activity .........................................................................................
22
IR E l .............................................................................................................................
22
Figure Tw o..............................................................................................................25
PERK ...........................................................................................................................
29
A T F 6 ............................................................................................................................
31
Physiological requirem ents for U PR function.............................................................
23
In yeast .........................................................................................................................
34
In mam mals ..................................................................................................................
35
Innate im munity and the U PR in mam m als..................................................................
39
Mam m alian innate im munity ..................................................................................
39
The UPR in m amm alian innate imm unity ...............................................................
40
The UPR in m etabolic disease ..................................................................................
42
The U PR in Caenorhabditiselegans .............................................................................
46
Table One ................................................................................................................
Caenorhabditiselegans im munity and the UPR ...........................................................
48
50
Sum mary...........................................................................................................................53
References.........................................................................................................................55
Chapter Two: An essential role for XBP-1 in host protection against
im m une activation in Caenorhabditis elegans ..........................................................
69
Sum m ary ...........................................................................................................................
71
Results and D iscussion .................................................................................................
73
Figure One ........................................................................................................
75
Figure Tw o.........................................................................................................
79
Figure Three......................................................................................................
83
Figure Four............................................................................................................87
M aterial and M ethods ....................................................................................................
90
Supplem entary Tables....................................................................................................
94
Supplem entary Table One..................................................................................
94
Supplem entary Table Tw o...............................................................................
95
Supplem entary Figures .................................................................................................
96
Supplem entary Figure One ...............................................................................
97
Supplem entary Figure Tw o................................................................................
99
Supplem entary Figure Three................................................................................101
Supplem entary Figure Four .................................................................................
103
Supplem entary Figure Five..................................................................................105
Supplem entary Figure Six....................................................................................107
A cknow ledgm ents............................................................................................................108
References........................................................................................................................109
Chapter Three: Physiological IRE-1-XBP-1 and PEK-1 signaling in
endoplasmic reticulum homeostasis in Caenorhabditis elegans...............113
Sum mary ..........................................................................................................................
115
Introduction......................................................................................................................117
Results ..............................................................................................................................
120
Elevated constitutive IRE-I activity in C elegans xbp-1 mutants ............................. 120
Figure One ...........................................................................................................
123
Increased constitutive PEK-1 activation in C. elegans xbp-1 mutants ....................... 124
Figure Tw o...........................................................................................................127
The effects of innate immune activation on ER stress levels in the xbp-1
M utant .........................................................................................................................
8
128
Figure Three.........................................................................................................131
A temperature-dependent requirement for XBP- 1 and PEK- 1 activities in
C. elegans developm ent and survival .........................................................................
132
Figure Four............................................................................................................135
XBP-1 and PEK- 1 maintain intestinal cell homeostasis during ER stress
caused by basal and induced innate imm unity............................................................136
Figure Five ............................................................................................................
139
Thermal stress necessitates UPR signaling for survival in C. elegans ....................... 141
The PM K -1 pathw ay protects against exogenous ER stress.......................................142
Figure Six...............................................................................................................145
D iscussion........................................................................................................................146
Figure Seven ..........................................................................................................
149
M aterials and M ethods.....................................................................................................152
A cknow ledgm ents............................................................................................................155
References........................................................................................................................156
Chapter Four: A genetic screen for factors contributing to lethality in
an X BP-1-deflcient strain during im m une activation.....................................159
Sum mary..........................................................................................................................161
Introduction......................................................................................................................163
Figure One .............................................................................................................
Results ..............................................................................................................................
165
169
Figure Tw o.............................................................................................................171
Table One ...............................................................................................................
173
Figure Three...........................................................................................................175
D iscussion and Future D irections....................................................................................176
M aterials
h
.....................................................................................................
References........................................................................................................................181
Chapter Five: C onclusion and Future D irections.......................................................183
Sum mary..........................................................................................................................185
9
179
Table One...............................................................................................................189
Future Directions .............................................................................................................
190
Observing the effects of UPR depletion on sub-cellular morphology........................190
Exam ining tissue specific requirem ents for the UPR .................................................
192
Identifying molecular mechanisms that lead to xbp-1 lethality in the presence
of Pseudomonas aeruginosa.......................................................................................193
Identifying the molecular mechanisms that lead to xbp-1; pek-1 temperaturesensitive lethality ........................................................................................................
194
Characterizing the role of ATF-6 in basal physiology and protection against
im mune activation.......................................................................................................195
Figure One .............................................................................................................
References........................................................................................................................200
10
199
Chapter One
Introduction
12
The endoplasmic reticulum (ER) is an organelle composed of a membrane
network enclosing a space that is topologically distinct from the cytoplasm. As the
beginning of the secretory pathway, the ER imports and folds nascent peptides before
they are trafficked via vesicles through the Golgi to the extracellular space, the plasma
membrane, or the lysosome. In an average eukaryotic cell, approximately one-third of the
total protein folding occurs in the ER. This fraction varies with the level of secreted
protein production, however, which is tremendously dynamic. Secreted protein
production fluctuates within individual cells as they react to changes in the environment,
such as when a macrophage perceives the presence of an extracellular pathogen and
responds by increasing production and secretion of cytokines. In metazoans, the level of
secreted protein production also differs widely between cell types, with specialized
secretory cells such as the mammalian plasma cell secreting hundreds to thousands of
proteins per second. The need to adjust ER protein folding capacity to match this
dynamic requirement for ER function is met by conserved signaling pathways that sense
the status of ER protein folding and initiate cellular compensatory responses. These
pathways, collectively termed the Unfolded Protein Response (UPR), are essential in all
eukaryotes. After a brief description of the features of protein folding in the ER, this
chapter will review the mechanism of UPR function and discuss its demonstrated roles in
the physiology of Saccharomyces cerevisiae,mammals, and Caenorhabditiselegans,
focusing in particular on the interaction between the UPR and immunity.
Protein Folding in the Endoplasmic Reticulum
A number of essential features are shared between protein folding in the ER and
protein folding in the cytoplasm, so it is worth first considering the general features of
eukaryotic protein folding. An essential challenge for the eukaryotic cell is that, although
the native conformation of all proteins is also the most thermodynamically favorable
state, only a small fraction of eukaryotic proteins spontaneously fold into the native
conformation in vitro (reviewed in Frydman 2001). Even under ideal folding conditions
in vitro, the majority of eukaryotic proteins become trapped in intermediate
conformations of local energetic minima. Conditions in vivo further complicate protein
folding for several reasons. First, peptide folding is faster than translation, so folding
proceeds at the N-terminus of a nascent peptide as it emerges from the ribosome, even
though correct folding of a domain is contingent upon the completed synthesis of that
domain. Protein folding in vivo is further confounded by the high protein concentration
within the cell, as hydrophobic regions of nascent peptides tend to interact
intermolecularly to form aggregates.
Cells overcome these obstacle to protein folding through use of several distinct
families of proteins collectively termed molecular chaperones. Chaperones are abundant
and essential within every eukaryotic protein folding compartment, and the major
chaperone families, including the Hsp70s and the chaperonins, are conserved throughout
eukaryotes and prokaryotes. The ER-localized Hsp70 chaperone, termed Grp78 (glucoseregulated protein 78)/BiP (binding immunoglobulin protein) in mammals and Kar2p in
yeast, is a particularly central player in ER protein folding. Like its Hsp70 family
homologs localized to the cytoplasm, BiP/Kar2p uses ATP catalysis, which is promoted
by J-domain-containing co-chaperones, to fuel repeated cycles of binding and release to
hydrophobic stretches of peptide (reviewed in Gething 1999). This ATP consumption
makes ER protein folding energetically costly, and depletion of ATP stores within a cell
can inhibit folding of secretory proteins (Gething 1999). The activity of BiP does not
directly facilitate protein folding into the native conformation, but rather it functions to
oppose both intramolecular misfolding of peptides and intermolecular interactions with
other unfolded peptides that would lead to aggregate formation.
Several features of ER protein folding are distinct from protein folding in the
cytoplasm. First, proteins destined for the secretory pathway arc marked by the
genetically encoded Signal Sequence - a sequence of 5-10 hydrophobic amino acids on
the N-terminus. As this sequence emerges from the ribosome during translation, it is
recognized and bound by the signal recognition particle (SRP), an abundant cytosolic
ribonuclear protein. This stalls translation and targets the entire complex to the ER
membrane. At the membrane, translation continues while the Sec61 translocation
complex is co-translationally imports the nascent peptide into the ER (Figure 1). Integral
membrane proteins are retained in the ER membrane at this stage, with the extracellular
domains facing into the ER lumen.
Figure 1. Protein folding in the Endoplasmic Reticulum. (Adapted from Brodsky and
Skach, 2011).
OST - oligosaccharyltransferase
BiP - binding immunoglobulin protein, also called Grp78 (glucose-regulated protein 78)
in mammals and Kar2p in S. cerevisiae
CNX - calnexin
Glc I and II - glucosidases I and II
UGT 1 - UDP-Glc glycoprotein glycosyltransferase
CRT - calreticulin
ERp57 - A protein disulfide isomerase, also called GRP58 (glucose regulated protein)
Erol - ER oxidoreductin 1
Man I - mannosidase I
EDEM - ER degradation enhancing a-mannosidase-like protein, also called Html
(homologous to mannosidase 1) in S. cerevisiae
ERAD - ER associated degradation
Figure I
to Goigi
Cytoplasm
Endoplasmic
reficulum
1",PDI
Ok 11
\EDEM
Man I
Old
Merooidases
GIG 11
IERAD
Sk4%
*Nkaceylglucosamine
*mannose
A glUCOSe
A second feature of ER protein folding is that, during peptide import into the ER,
oligosaccharyltransferase (OST) cotranslationally glycosylates most ER proteins with a
large core oligosaccharide (GlcNAc 2Man9 Glc 3) on an asparagine residue that is part of a
consensus Asn-X-Ser/Thr sequence. In addition to improving solubility and altering the
inherent folding propensities, this glycosylation is the central mechanism by which the
ER controls the trajectory of ER client proteins (reviewed in Aebi et al. 2010).
Glucosidases I and II rapidly remove the terminal two glucose residues from the
oligosaccharide, and the resulting glycosyl group identifies a client peptide as a substrate
for the ER resident lectin binding chaperones calnexin (CNX) and calreticulin (CRT),
which promote folding (Schrag et al. 2001; Thomson et al. 2005). At this point, a
balanced rate of removal and replacement of the final glucose serves as a timer for the
folding of each client. When the glucose is removed, the oligosaccharide is no longer
recognized and bound by CNX or CRT. If the client is folded, it can exit the ER; if the
client is incompletely folded, it's glycosyl group may be reglucosylated by UGT (UDPGlc glycoprotein glycosyltransferase), which recognizes both the GlcNAC2Man9 and
exposed hydrophobic sequences (Aebi et al. 2010). This allows for another cycle of
interaction with CNX or CRT. Slow acting Mannosidase I (ERManI/Mns I) competes
with UGT to modify the GlcNAC2Man9 glycosylation, however, and once a mannose is
removed, the client exits the CNX CRT cycle and becomes a target of ER-associated
degradation (ERAD).
Through ERAD, unfolded or terminally misfolded clients are translocated from
the ER to the cytosol, where they are degraded by the proteasome (reviewed in Schr6der
and Kaufman 2005). Many ERAD effectors have been identified, and the subset involved
in shuttling a particular misfolded ER client through ERAD is dependent on the structure
and location (lumen or membrane) of that client. After removal of the first mannose,
other mannosidases catalyze the removal of additional mannose residues. Once two to
four mannose residues have been trimmed, the client becomes a ligand for a set of
ERAD-dedicated lectins, such as EDEM (ER degradation enhancing a-mannosidase-like
protein), which direct the client to the ER membrane to be ubiquitinated and exported to
the cytoplasm (reviewed in Hedge and Ploegh 2010).
A third feature of protein folding unique to the ER is that most ER clients contain
one or more intramolecular disulfide bonds between cysteine residues in their folded
conformations (reviewed in Sevier and Kaiser 2008). These disulfide bonds protect
secreted proteins from denaturing in the harsh extracellular environment. They are
formed by a series of dithiol disulfide transfer reactions. Client proteins are oxidized by
protein disulfide isomerases (PDIs), which are recycled through oxidation by the
membrane-associated protein EroIp (ER oxidoreductin 1). EroIp is oxidized in turn by
unloading the electrons to 02, forming H2 0 2 . This production of H20 2 makes the ER the
second largest source of reactive oxygen species (ROS) in the cell after the mitochondria.
Just as folding a nascent peptide involves sampling conformations to increase
thermodynamic stability, formation of the correct disulfide bonds likewise involves
sampling possible disulfide bonds between the various cysteine residues within a protein.
A client protein will therefore undergo several rounds of oxidation and reduction by PDIs
(Sevier and Kaiser 2008).
A fourth unique aspect of ER protein folding is that it is dependent on a high
calcium concentration. The cell sequesters calcium (Ca2) in the ER: whereas the average
calcium concentration in the cytosol is 0.1 pM, the ER calcium concentration reaches 5
mM (Schr6der and Kaufman 2005). The majority of ER chaperones require calcium,
which they bind with low affinity, to function. ER calcium also functions as a second
messenger in a variety of signaling pathways, which induce the rapid release of calcium
from the ER into the cytosol. This use of calcium is necessarily coupled with a temporary
reduction in ER protein folding until a high calcium concentration is restored.
The assorted components of ER protein folding are organized through physical
associations with each other. BiP/Kar2p association with the Sec61 complex is required
for peptide translocation into the ER ensuring that, as in the cytoplasm, translation of
each mRNA molecule is coupled with the interaction between the resulting nascent
peptide and chaperones (Brodsky et al. 1995; Dierks et al. 1996). The Sec61 translocon is
also associated with both OST, the enzyme that catalyzes N-linked glycosylation, and
CNX, which associates with the newly glycosylated client before translation is completed
(reviewed in Johnson and van Waes 1999). CNX and CRT physically interact with the
PDI ERp57/GRP58 to promote disulfide bond formation and isomerization. Likewise,
BiP associates with another PDI, P5, and the interaction between BiP and a client is
strengthened when an aberrant disulfide bond is transferred from the client to P5
(reviewed in Brodsky and Skatch 2011). These associations organize ER protein folding,
increasing efficiency and preventing the presence of unattended clients, thereby avoiding
the formation of aggregates.
The UPR alleviates perturbations in ER protein folding
In addition to associating with each other to promote folding of each ER client
protein, ER chaperones monitor the overall status of ER protein folding. When the
folding requirement of the ER exceeds the folding capacity, the depletion of excess
chaperones results in the activation of the Unfolded Protein Response (UPR). UPR
signaling activates mechanisms to restore the balance between ER load and ER folding
capacity, thus functioning as a negative feedback loop. The presence of such a system to
monitor ER status is critical because, as the ER is isolated from the cytoplasm by the ER
membrane, ER proteostasis is distinct from that of the cytoplasm. Conditions that
activate the UPR are collectively termed "ER stress." The consequences of and responses
to ER stress have been best studied using exogenous agents that inhibit ER protein
folding. These include calcium ionophores, which disrupt the function of calciumdependent chaperones, reducing agents, which inhibit oxidative protein folding, and
tunicamycin, which blocks N-linked glycosylation.
The UPR is highly conserved, and the most ancient pathway, activated by the ERlocalized integral membrane protein IRE1/Irelp (inositol-requiring enzyme 1), has been
found in all eukaryotes queried to date. In animals, the UPR has been expanded to
include two additional signaling pathways, which are activated by the ER-localized
integral membrane proteins PERK/PEK-1 (PKR-like ER kinase) and ATF6 (activating
transcription factor 6). All three of these proteins act as sensors of ER stress (Figure 2).
Together, they initiate a multipronged response that includes by expanding the ER to
accommodate the increased protein load, increasing folding capacity by elevating
production of ER chaperones and foldases, and decreasing the accumulation of misfolded
proteins through increased production of components of ERAD. In animals, the UPR also
lowers the rate of influx of nascent peptides into the ER by two mechanisms: the PERK
branch reduces global translation (Harding et al. 1999), and the IRE 1 branch reduces
translation specifically of secretory proteins (Hollien et al. 2006). If these mechanisms
are insufficient to alleviate ER stress, chronic UPR activation in mammals induces
apoptosis.
Mechanisms of UPR activity
IRE]
The existence of the UPR has been appreciated since the 1980s, when it was
observed that treatment with drugs to inhibit ER protein folding or overexpression of
unfoldable secretory proteins caused an increased production of a characteristic set of
proteins that were localized to the secretory pathway (Kozutsumi et al. 1988). One of the
most prominent of these was BiP/Kar2p, suggestive that this response was protective
(reviewed in Gething et al. 1992; Lee 1987). Screens for mutants defective in this
response, and consequently more sensitive to ER stress, led to the first identification of
IRE] in S. cerevisiae (Cox et al. 1993; Mori et al. 1993). Mammals have two paralogs of
Ire]: IREla, which is expressed ubiquitously, and IRE1I, which is expressed in the
intestine (Bertolotti et al. 2001; Tirasophon et al. 1998; Wang et al. 1998).
Irelp is an ER-localized type I integral membrane protein with a kinase domain
and an endoribonuclease domain on its cytosolic side (Cox et al. 1993; Mori et al. 1993).
In the inactive state, Irelp monomers associate with BiP/Kar2 through a hydrophobic
region in the ER luminal domain (Bertolotti et al. 2000; Okamura et al. 2000).
Accumulation of unfolded ER clients titrates BiP/Kar2p away from IRE 1/Ire Ip, allowing
the IRE1/Irelp monomers to dimerize (Bertolotti et al. 2000; Okamura et al. 2000). This
dissociation of BiP from IREI may be a sufficient signal from the ER luminal domain to
induce IRE 1 activation in animals, as mutant versions of mammalian IRE 1a that have
low binding affinity for BiP are constitutively activated (Oikawa et al. 2009). In S.
cerevisiae, however, the Ire Ip dimers form an extra binding pocket in the ER luminal
side, similar to an MHC (major histocompatibility complex) groove, that appears to be
incompletely formed in animal IRE 1 (Credle et al. 2005). Kar2p dissociation from S.
cerevisiaeIreIp is not sufficient for Ire Ip activation, but it permits dimerization to reveal
this hydrophobic pocket (Okamura et al. 2000; Pincus et al. 2010). Mutation of residues
within the MHC-like pocket abrogate Ire 1p activation, indicating that this pocket may
directly bind unfolded proteins in order for Irelp to be activated (Credle et al. 2005). In
both yeast and animals, IRE1/Irelp dimerization leads to oligomerization into higher
order clusters, which permits transautophosphorylation by the IRE 1/Ire 1p cytoplasmic
Ser/Thr kinase domain (Shamu et al. 1996; Li et al. 2010; Korennykh et al. 2009).
Figure 2. Overview of the mammalian UPR.
ATF6 - activating transcription factor 6
GLS - Golgi-localization sequence
IRE1 - inositol-requiring enzyme 1
PERK - Protein kinase RNA-activated (PKR)-like ER kinase
KD - kinase domain
ED - endoribonuclease domain
SIP - site 1 protease
S2P - site 2 protease
TRAF2 - tumor necrosis factor receptor (TNFR)-associated factor 2
ASKI - apoptosis signal-regulated kinase 1
NF-KB - nuclear factor wB
JNK - cJun-N-terminal kinase
Nrf2 - nuclear factor erythroid-derived 2-related factor 2
eIF2a - eukaryotic initiation factor 2 alpha
uORF - upstream open reading frame
ATF4 - activating transcription factor 4
ERSE - ER stress response element
UPRE - unfolded protein response element
ERAD - ER associated degradation
CHOP - C/EBP homologous protein, also called GADD 153 (growth arrest- and DNA
damage-inducible gene 153)
GADD34 - growth arrest- and DNA damage-inducible gene 34
Figure 2
No ER stress
ER stress
Cytoplasm
:2
IL
p
0
'P
2P
3RF5
NF-KB
$
Nrf2
Inflammation.
Apo
.
Vtf
............
U
Jr
ERSE
L
InflarnnlatM
ImmtaMy
Apoptosis
UPRE
Chaperones, Foldases,
upid biosynthesis, ERAD
Nucleus
UPRE
HOP
GADD34
,
Oxtcatiwe stress
response
Oxusative stress
response
apoptosis
The oligomerization of IRE l/Ire Ip causes activation of its endoribonuclease
domain, the primary function of which is to catalyze the non-canonical splicing of a
specific mRNA that encodes a basic leucine zipper (bZIP) transcription factor. This
transcription factor, which is functional as a homodimer, is encoded by HAC1
(homologous to ATF/CREB 1) in S. cerevisiae and XBP1 (X-box binding protein 1) in
animals. S. cerevisiae Irelp cleaves the HAC1 mRNA at two sites to remove a 252-nt
intron located in the loop region of a bipartite stem loop structure, and the exons are
joined by tRNA ligase Rlgl p (Sidrauski et al. 1996). This splicing is required for the
efficient translation of HA C1 into HacIp (Chapman et al. 1997; Cox et al. 1996).
The regulation of XBP1 in metazoans is somewhat different. The intron removed
by IRE 1 is short - 26 bp in mammals, 23 in C. elegans - and the removal of the intron
changes the translational reading frame (Calfon et al. 2002; Lee et al. 2002; Yoshida et al.
2001). Both the IRE 1-spliced and unspliced mRNAs are translated into XBPls and
XBPlu, respectively, but while both proteins contain the bZIP DNA binding domain,
only XBPls contains a functional transactivation domain at the C-terminus. The Cterminus of XBP 1u is as conserved as that of XBP Is, though, indicative of functional
significance. Indeed, XBP lu performs at least two functions: First, during translation of
XBPlu mRNA, a hydrophobic region in the emerging XBP 1u protein localizes the
mRNA-ribosome-peptide complex to the ER membrane to facilitate IRE 1-mediated
splicing (Yanagitani et al. 2009). Second, fully translated XBPlu can dimerize with
XBPs to inhibit both its nuclear localization and its ability to activate transcription (Lee et
al. 2003; Yoshida et al. 2006). Transcription of the XBP1 gene is induced by ER stress
and continues to rise even after IRE 1 is inactivated, so the resulting increase in
production of XBP 1u may function to turn off the XBP Is-mediated transcription once the
ER stress is resolved (Yoshida et al. 2006). Recently, two studies demonstrated that
XBPls physically interacts with phosphatidylinositol 3-kinase (P13K), a component of
the insulin signaling pathway, and that this interaction enhances XBP 1s function by
promoting its nuclear localization when the insulin signaling pathway is activated
(Yoshida et al. 1998; Park et al. 2010; Winnay et al. 2010). This may serve to amplify the
signaling of this branch of the UPR in preparation for the cellular responses to insulin,
which include cell growth and increased protein synthesis (reviewed in Nandi et al.
2004). The functional significance of this link between insulin signaling and UPR
activation will be discussed in a later section.
S. cerevisiae HacIp binds the UPR element (UPRE), a 22 bp sequence sufficient
to confer inducibility to ER stress (Mori et al. 1992; Kohno et al. 1993; Chapman et al.
1997; Cox et al. 1996). Transcriptional profiling comparing the WT versus HAC1 null
response to two drugs that inhibit ER protein folding, DTT and tunicamycin, identified
381 genes that are differentially regulated, about 250 of which are induced by ER stress
in a Hac Ip-dependent manner (Travers et al. 2000). These genes encode proteins that
function directly in ER protein folding, including KAR2, PDI,and ERO], as well as
proteins involved in phospholipid biogenesis, ERAD, and later stages of the secretory
pathway (Travers et al. 2000). Importantly, although HA C1 is required to induce these
genes in response to ER stress, many are transcribed during basal growth through HA Clindependent mechanisms. For example, the KAR2 promoter contains not only an UPRE
but also a separate GC-rich region sufficient for basal KAR2 expression and a heat shock
element (HSE), making it inducible by Hsflp (Mori et al. 1992; Kohno et al. 1993).
Hsflp regulates the cytosolic response to excess unfolded proteins, making this dual
regulation of KAR2 expression one of the few known instances of cross-talk between
cytosolic and ER proteostasis. Because of such alternate mechanisms of transcriptional
regulation of genes encoding UPR effectors, although many of the transcriptional targets
of Haclp are essential for viability, HACI itself is not.
In mammals, comparison between the promoters of known UPR response genes
led to the identification of the cis-acting ER stress response element (ERSE I), a 19 bp
sequence that is structurally distinct from the yeast UPRE (Yoshida et al., 1998; Roy et
al. 1999). XBP 1 binds the ERSE I and several other promoter motifs identified
subsequently (Wang et al. 2000; Yoshida et al. 2001). Through transcriptional profiling
and ChIP studies, XBP 1 has been demonstrated to regulate transcription of hundreds of
genes both in Caenorhabditiselegans and across diverse mammalian cell types, including
MEFs, B cells, skeletal muscle cells, and hepatocytes (Acosta-Alvear et al. 2007; Lee et
al. 2003; Lee et al. 2008; Shaffer et al. 2004; Shen et al. 2005; Yoshida et al. 2003).
Similar to the transcriptional targets of Hac lp, the products of these XBPJ-regulated
genes include proteins involved in ER folding, glycosylation, ERAD, and other steps in
the secretory pathway (Lee et al. 2003; Yoshida et al. 2003; Shaffer et al. 2004; Shen et
al. 2005; Acosta-Alvear et al. 2007). They also include proteins that are regulated by
XBPI in a tissue-specific manner, many of which function in pathways apparently
unrelated to ER stress such as development (Shaffer et al. 2004; Acosta-Alvear et al.
2007; Lee et al. 2008).
In addition to splicing HA C1/XBP1 mRNA, IRE 1 performs a number of other
functions. In both mammals and Drosophila, it degrades a subset of mRNAs encoding
secretory proteins, using the cotranslational import of nascent peptides to recognize the
appropriate mRNAs. This process, termed regulated IRE 1-dependent decay (RIDD),
reduces ER stress by decreasing the influx of new proteins to the ER (Hollien et al. 2006;
Hollien et al. 2009; Han et al. 2009). Furthermore, mammalian IREl has been shown to
interact with several other cellular signaling pathways independent of its
endoribonuclease activity. It binds to the proapoptotic BCL-2 family members BAX and
BAK, which promotes its activation (Hetz et al. 2006). IRE 1 also forms a complex with
TRAF2 (TNFR-associated factor 2) and ASKI (apoptosis signal-regulated kinase 1) to
activate the downstream mitogen activated protein kinase JNK (cJun-N-terminal kinase)
(Urano et al. 2000; Nishitoh et al. 2002). This can lead to induction of apoptosis, a last
resort when mammalian cells face overwhelming ER stress. The IRE 1-TRAF2 complex
has also been shown to recruit and activate IKK (IKB kinase) (Hu et al. 2006).
Phosphorylation of IKB (NF-KB inhibitor) promotes its own degradation, thereby
permitting activation of NF-KB (nuclear factor KB). The significance of the ER stressinduced activation of this central regulator of the inflammatory response is not well
understood, but it will be discussed further in a later section.
PERK
The ER luminal domain of PERK is homologous to that of IRE 1, and PERK
activation is likewise repressed by the association of its ER luminal domain with BiP
(Bertolotti et al. 2000; Chen et al. 2002). Indeed, the ER luminal domains of IRE 1 and
PERK can be interchanged without affecting the rate of UPR induction during ER stress
(C Y Liu et al. 2000). The cytosolic side of PERK contains a kinase that has been shown
to phosphorylate two targets: eIF2a (eukaryotic initiation factor 2 alpha) and NRF2
(nuclear factor erythroid-derived 2-related factor 2) (Cullinan et al. 2003; Harding et al.
1999).
PERK is one of four known kinases that phosphorylate eIF2ct. The other three
eIF2a kinases are also activated by stress responses - Gcn2 responds to amino acid
deprivation, PKR (protein kinase RNA-activated) to dsRNA (indicative of viral
infection), and HRI (heme-regulated eIF2a kinase) to depletion of intracellular heme leading Harding et al. to designate this portion of the UPR as the integrated stress
response (ISR) (Harding et al. 2000; Harding et al. 2003). Phosphorylation of eIF2a
inhibits the eIF-2B guanylate exchange factor, which causes inhibition of translation
initiation of ribosomes, leading to reduced translation of most mRNAs (Schr5der and
Kaufman 2005). Conversely, as this decreased translation initiation efficiency increases
the rate of ribosomes skipping inhibitory upstream open reading frames (uORFs), it also
effects increased translation of uORF-containing mRNAs, the most important of which is
the mRNA encoding the bZIP transcription factor ATF4 (Vattem et al. 2004). ATF4
activates transcription of genes that protect cells against oxidative stress, and disruption
of PERK-mediated eIF2a phosphorylation or loss of functional A TF4 causes increased
sensitivity of cells to death from acute ER stress (Harding et al. 2003; Scheuner et al.
2001). ATF4 also promotes apoptosis during extended ER stress by activating
transcription of the pro-apoptotic transcription factor CHOP (C/EBP homologous
protein) (Harding et al. 2003). In mammalian cells experiencing such prolonged ER
stress, PERK signaling remains elevated, whereas the level of IRE 1-mediated XBP1
mRNA splicing declines, perhaps because the cells switch from a program designed to
alleviate ER stress to one that promotes apoptosis (Lin et al. 2007). Phosphorylation of
eIF2a can also activate NF-KB, either by causing disruption of the NF-KB-IKB complex,
by decreasing transcription of IKB, or through a combination of the two mechanisms
(Jiang et al. 2003; Deng et al. 2004).
Nrf2, another substrate of PERK, is a bZIP transcription factor that associates in
the cytoplasm with Keap 1 (Kelch ECH associating protein 1), which sequesters Nrf2
through association with the actin cytoskeleton and also enhances Nrf2 ubiquitination and
degradation (Kensler et al. 2007). Phosphorylation by PERK disrupts the interaction
between Nrf2 and Keap 1, allowing Nrf2 to translocate to the nucleus and activate
transcription of antioxidant and oxidant-detoxifying enzymes such as glutathione Stransferase (reviewed in Zhang and Kaufman 2008). As with the phosphorylation of
eIF2a, Nrf2 is phosphorylated by a number of other signaling pathways, including the
ERK, p38, and JNK MAPKs (mitogen-activated protein kinases), PI3K, and PKC
(protein kinase C), making it a central node in a variety of stress responses (Kensler et al.
2007). Consequently, whereas PERK -cells exhibit increased ROS only when ER stress
is induced, NrJ2- cells exhibit elevated levels of ROS constitutively (Cullinan et al.
2004; Harding et al. 2003; Leung et al. 2003).
A TF6
ATF6, a type II integral membrane protein, contains a bZIP transcription factor
domain its cytosolic portion (Haze et al. 1999; Yoshida et al. 1998). C. elegans possesses
a single ATF6, and mammals have two paralogs, ATF6a and ATF6p (Yoshida et al.
2003; Shen et al. 2005). Although ATF6 shares no homology with IREI or PERK,
activation of ATF6 is also initiated when unfolded ER proteins titrate BiP away from a
binding site on the ATF6 ER luminal domain (Shen et al. 2002). Removal of BiP reveals
two Golgi localization sequences which direct ATF6 translocation to the Golgi (Shen et
al. 2002; Chen et al. 2002). In the Golgi, ATF6 is cleaved first in the luminal domain by
the serine protease Si P (Site-I protease), and then within the transmembrane domain by
the metalloprotease S2P (Site-2 protease), to release the active transcription factor (Chen
et al. 2002; Haze et al. 1999). Additionally, while ATF6 is localized to the ER, it is
negatively regulated by intramolecular and intermolecular disulfide bonds formed
between two conserved cysteine residues in the luminal domain. These disulfide bonds
are reduced during conditions of ER stress, and only the reduced monomeric form of
ATF6 reaches the Golgi (Nadanaka et al. 2007). In addition to ATF6a and ATF6p, a
number of more distantly related ATF6 homologs are activated by a similar mechanism
in mammalian cells, including OASIS, CREBH, CREB3, and CREB4 (review in Ron and
Walter 2007). The function of these proteins is not well understood, but they appear to
modulate the ER stress response in specific tissues (Ron and Walter 2007).
Like XBP 1, ATF6 binds to ERSE promoter sequences to activate transcription,
and it has been difficult to predict which transcription factor activates each gene with
promoter ERSEs (Haze et al. 1999; Yoshida et al. 1998; Yoshida et al. 2001). For
example, a study examining UPR induction in A TF6a-l and XBP1-I- mouse embryonic
fibroblasts (MEFs) found that ATF6a and not XBP 1 was important for the induction of
major ER chaperones, including BiP and CRT, as well as components of ERAD
(Yamamoto et al. 2007). Another study showed that overexpression of ATF6a, but not
ATF6p or XBP 1, was sufficient to promote lipid biosynthesis and ER expansion under
stress conditions (Bommiasamy et al. 2009). Like IRE1 and PERK, ATF6 activation can
lead to activation of NF-KB, though the mechanism behind this interaction is unclear
(Yamazaki et al. 2009). A subset of ER stress-induced transcriptional targets contain a
promoter element, the mammalian UPRE, to which ATF6a-XBP1 heterodimers, but not
ATF6a or XBP1 homodimers, binds with high affinity (Wu et al. 2007; Yamamoto et al.
2007). In addition, ATF6a heterodimerizes with ATF4, and this heterodimer is important
for inducing transcription of CHOP (Ma et al. 2002). These interactions between ATF6,
XBP 1, and ATF4 coordinate functionality between the three branches of the UPR.
Physiological requirements for UPR function
Studies examining the mechanisms of UPR function have been predominantly
performed by exposing isolated cells in culture to drugs that disrupt ER protein folding.
Indeed, in the initial work characterizing the UPR in yeast, Ire 1p-medated splicing of the
HAC] mRNA was observed in response to harsh drug treatments or with expression of
unfoldable secretory proteins, whereas it was not detected during standard growth
conditions (Cox et al. 1996; Spear et al. 2003). Furthermore, IreIp is not required for
growth of S. cerevisiae under standard growth conditions. Because of these observations,
the UPR was originally considered a response to extreme conditions of stress in the ER.
Through further genetic and molecular characterization of the UPR, however, it has
become apparent that the UPR performs critical functions in the normal physiology of
'unstressed" cells.
IN YEAST
In contrast with the initial work characterizing the regulation of HA C], more
recent studies have reported that between 1%and 8% of HA C1 mRNA is spliced in
exponentially growing S. cerevisiae under standard growth conditions (Schroder et al.
2000; Bicknell et al. 2007). This basal UPR activation has been demonstrated to perform
several functions. Exposure to exogenous ER stress or deletion of HAC1 resulted in a
partially penetrant defect in cytokinesis, suggestive that one function of the constitutive
UPR activity is to permit adaptation to the fluctuating demand for ER function during cell
cycle progression (Bicknell et al., 2007). Schroder et al. found that another function of
basal UPR activation is to repress diploid sporulation and pseudohyphal growth, both of
which are differentiation responses to nitrogen starvation (Schroder et al., 2000). They
found that the low level of basal Ire Ip activity is abrogated during nitrogen starvation,
possibly due to the concurrent general decrease in translation (Schroder et al., 2000).
Further, they showed that, to negatively regulate the sporulation response, Hac lp directly
associates with the Rpd3p-Sin3p histone deacetylase complex (HDAC), which was
previously known to inhibit transcription of early meiotic genes (EMGs) (Schroder et al.
2004). The Rpd3p-Sin3p HDAC complex is involved in the regulation of multiple other
processes, including mating type switching, the response to nutrient starvation,
phospholipid biosynthesis, and the response to osmotic stress (reviewed in Silverstein and
Ekwall 2005). It will be interesting to determine the extent to which HacIp is involved in
HDAC activity, as this interaction could both reduce ER stress through the regulation of
metabolic processes and use the sensing of ER stress to inform metabolic and
differentiation decisions.
To systematically determine which processes protect against ER stress, Jonikas et
al. screened a deletion library of ~4,500 S. cerevisiae strains for induction of a Hac Ipresponsive GFP reporter (Jonikas et al. 2009). Of the 398 genes whose deletion caused
upregulation of the reporter, just 56 have been shown to be induced by the UPR,
emphasizing that the study of UPR targets provides an important but incomplete
description of the processes required for the maintenance of ER homeostasis. Further, by
analyzing the synthetic effects on reporter induction of 340 of the hits (over 60,000
strains), they categorized the hits into functional groups (Jonikas et al., 2009). Future
studies to characterize the activities of these functional groups and how they impact ER
protein folding will likely provide insight into how ER protein folding is impacted by
diverse cellular processes.
IN MAMMALS
The function of the UPR in mammals is complex, varying considerably between
different tissues and cell types. The initial phenotypic characterizations of knockout mice
for individual branches of the UPR demonstrated a requirement for the UPR in highly
secretory cell types. For example, knockouts of IRE la or XBP1 are embryonic lethal due
to liver hypoplasia (Reimold et al. 2001; Urano et al. 2000). PERK knockout mice are
viable through embryogenesis, but they rapidly lose pancreatic islet cells, which results in
lethality (Zhang et al. 2002). Similarly, humans with mutations in the locus encoding
PERK develop Wolcott-Rallison syndrome, one of the characteristics of which is early
onset insulin-dependent diabetes (Deldpine et al. 2000). Likewise, although single
knockouts of ATF6a or ATF6p develop normally, double-knockout mice exhibit
embryonic lethality (Yamamoto et al. 2007).
A well studied instance of the UPR functioning in mammalian development is the
requirement for the IRE 1 pathway in the production of cells of the adaptive immune
system, especially in plasma cells. Plasma cells, a differentiated subset of B cells, are an
extreme example of specialized secretory cells - each cell can produce and secrete
millions of antibodies per minute into the blood. Indeed, XBP 1 was initially identified
not as a component of the UPR, but rather as a transcription factor that bound the
promoters of genes encoding the MHC II and that was highly expressed in myeloma cells
(Liou et al. 1990; Reimold et al. 1996). Tissue-specific knockout of XBP1 from the
hematopoietic lineage showed that it is essential for the development of B cells into
antibody secreting plasma cells (Reimold et al. 2001). In further analyses of the role of
XBP 1 in plasma cell development, it was found that IRE 1 mediated XBP1 mRNA
splicing is induced when mature B cells are stimulated with the signals sufficient to cause
differentiation into plasma cells (Iwakoshi et al. 2003). Transcriptional profiling in
XBP 1-deficient B cells showed that here, as in other settings, XBP1 regulates
transcription of genes whose products are involved in ER expansion and protein folding
(Shaffer et al. 2004). Using a GFP reporter for IRE 1-mediated XBP1 splicing in the
hematopoietic lineage, Brunsing et al. showed that this branch of the UPR is also
activated in pro-B cells, double positive thymic T cells, and cytotoxic T cells, suggestive
that it plays a role in multiple cells of the adaptive immune response (Brunsing et al.
2008). Although the hematopoietic XBP1 knockout mice had normal levels of B and T
cells overall, IRE la-deficient B cells were unable to develop beyond the pro-B cell state
(the stage at which an incomplete version of an antibody is produced and displayed on
the cell surface), indicating that the essential function of IRE 1 in pro-B cells is XBP 1independent (Zhang et al. 2005; Reimold et al. 200 1).
The function of the mammalian UPR in vivo is somewhat controversial, even in
the cells in which the UPR has been most thoroughly studied - hepatocytes and plasma
cells. The requirement for UPR function in these secretory cell types has generally been
rationalized through the inference that as a cell develops into a secretory cell, the UPR
senses the increased production of secreted proteins and adapts the ER to accommodate
that increase. There has been speculation, however, that this model may not entirely
account for the activity of the UPR. For instance, Lee et al. have suggested that the IRElXBP 1 pathway performs metabolic functions in hepatocytes in the absence of ER stress.
They showed that deletion of XBPI from fully developed hepatocytes in mice resulted in
decreased levels of plasma triglyceride, cholesterol, and free fatty acids, and that XBPls
directly binds the promoters of lipogenic genes in hepatocytes (Lee et al. 2008). They
observed no activation of either ATF6 or PERK and no alteration in ER morphology in
the XBP1- - hepatocytes, leading them to suggest that this lipogenesis-promoting activity
of XBP1 is independent of ER stress.
On the other hand, the Mori lab observed that A TF6a knockout mice developed
liver steatosis and lipid droplet formation when exposed to pharmacologically induced
ER stress as adults (Yamamoto et al. 2010), and the Ron lab found that mice with
diminished eIF2a phosphorylation were protected against hepatosteatosis induced by
high fat diet (Oyadomari et al. 2008). Therefore, all three branches of the UPR influence
the development of liver steatosis - the ATF6 branch protects against it, whereas chronic
activation of the PERK or IRE 1-XBP 1 pathways contributes to the development of fatty
liver. It is likely, therefore, that all three branches are activated in the liver by ER stress,
but the cause of that stress is still unknown.
Even in plasma cell development, long used as the canonical example of UPR
function in mammalian development, several surprising features of XBP 1 activity have
been identified. First, it was thought that IRE 1 was activated during plasma cell
differentiation in response to the enormous elevation in antibody flux through the
secretory pathway, but in fact it is not dependent on the production of secreted antibodies
(Skalet et al. 2005; Hu et al. 2009). Second, although it was long assumed that XBP1
activation served to expand ER folding capacity so that the required level of antibodies
could be produced, it appears that antibody secretion in XBP1- plasmablasts (developing
plasma cells) is unimpaired, and only more subtle differences in lipid synthesis and
glycosylation have been observed (McGehee et al. 2009). These recent findings indicate
that even this well-studied instance of UPR function will require further investigation.
Taken together, these genetic analyses showed that the UPR is not only a stress
response; rather, it is an integral signaling pathway in development, be it during cell
division in S. cerevisiae or differentiation of specialized secretory cells in mammals.
Linking these studies together, it appears that the UPR functions under physiological
conditions to adapt the ER at times when cells require an increased capacity of protein
folding in the secretory pathway. Still, it was unclear, particularly in the metazoan
models, what specific cellular processes instigate the requirement for the UPR.
Considerable evidence in mammals suggests that one such process which necessitates the
UPR is the innate immune response.
Innate immunity and the UPR in mammals
MAMMALIAN INNA TE IMMUNITY
Unlike the adaptive immune response, in which genome rearrangement permits an
individual to tailor its defenses to the pathogens it encounters, the innate immune system
is composed of the cells and mechanisms which protect the host against infection in a
non-specific manner. It includes not only designated immune cell types but also immune
mechanisms that can be activated in numerous additional tissues. Two of the major
functions performed by the innate immune response are 1) to sense and impede invading
pathogens and 2) to orchestrate an appropriate overall response within the organism
through paracrine and endocrine signaling to other cells and tissues.
To sense pathogens, mammalian cells use a variety of pathogen recognition
receptors (PRRs) that each bind specific pathogen associated molecules. These are
expressed particularly in cells likely to be exposed to invading pathogens, including
macrophages, dendritic cells, and epithelial cells. They are often expressed on the cell
membrane, ready to be activated by extracellular pathogens. Engagement of PRRs leads
to the activation of signaling cascades, such as the NF-KB and p38 MAP kinase
pathways, which initiate transcriptional responses. These signaling pathways are highly
conserved: the NF-KB pathway mediates components of the immune response in D.
melanogaster,and the p38 pathway is the primary immune response pathway in C.
elegans, as will be discussed in depth later. The genes upregulated by these signaling
pathways encode secreted proteins that effect the two functions of the innate immune
response. The task of killing pathogens is taken up by antimicrobial peptides, such as
lysozymes and defensins, while cytokines are used to signal to other cells, further
activating the cellular immune responses and causing inflammation.
THE UPR IN MAMMALIAN INNA TE IMMUNITY
Macrophages are phagocytic cells that not only produce antimicrobial effectors
and cytokines but also present antigens to be recognized by cells of the adaptive immune
response. Martinon et al. found that stimulation of mouse macrophages with TLR
agonists including LPS and flagellin induced IRE 1 phosphorylation and IRE 1-mediated
XBP1 splicing but did not detectably increase PERK phosphorylation or ATF6 cleavage
(Martinon et al. 2010). TLR engagement signals through the NF-KB, p38, and JNK
pathways to induce production of secreted antimicrobial effectors and cytokines, but
pharmacological inhibition of any of these signaling pathway individually was not
sufficient to prevent IREl activation. Instead, they found that IREl activation was
dependent on activation of NADPH oxidase 2, which is activated by TLR signaling to
produce ROS. Macrophages use ROS both as an antimicrobial effector that kills
phagocytosed pathogens and as a signaling molecule required for production of a number
of proinflammatory cytokines (Bae et al. 2009). It is unclear whether ROS activates IREl
by interfering with oxidative protein folding, by elevating ER load through increased
cytokine production, or through some combination of the two. An intriguing aspect of the
TLR-induced IRE 1 activation is that it does not lead to increased production of typical
UPR effectors. In fact, both Martinon et al. and Woo et al. found that stimulation with
TLR agonists diminished production of UPR effectors upon subsequent challenge with
tunicamycin (Woo et al. 2009; Martinon et al. 2010). Based on these observations, both
groups speculated that the UPR may be specifically depressed instead of activated by this
physiological source of ER stress in order to prevent PERK-mediated apoptosis and
maximize cytokine production.
Intestinal epithelial cells (IECs) secrete digestive enzymes and absorb nutrients,
and they also maintain the perennial intestinal flora in check and serve as the first line of
defense against more malignant intestinal pathogens. These latter two activities are
performed by Paneth and goblet cells, which produce and secrete antimicrobial effectors
and mucin, respectively, into the intestinal lumen. Kaser et al. found that mice lacking
XBP 1 in the IEC lineage exhibited near complete eradication of the Paneth and goblet
cells (Kaser et al. 2008). Closer examination of the Paneth cell line showed that they
were generated in the IEC XBP-'-mice but then rapidly eliminated by apoptosis. Further,
the IEC XBP1-- mice show increased susceptibility to infection with an intestinal
bacterial pathogen, Listeria monocytogenes, and they also developed intestinal
inflammation reminiscent of inflammatory bowel disease (IBD). Similarly, a mouse
knockout of IRE IfP, the expression of which is restricted to the stomach, small intestine,
and colon, exhibited elevated constitutive ER stress in the colon and an increased
susceptibility to DSS (dextran sulfate sodium)-induced colitis (Bertolotti et al. 2001).
Furthermore, Kaser et al. demonstrated a link between single nucleotide polymorphisms
(SNPs) in human XBP1 and IBD (Kaser et al. 2008). One interpretation of this phenotype
is that the IRE 1-XBP 1 pathway functions in Paneth and goblet cells to protect against the
high secretory load engendered by production of innate immune effectors (Kaser et al.
2008). Reduced function of these cell types would permit the intestinal flora to expand,
which would further activate secretion of immune effectors in remaining Paneth and
goblet cells, resulting in a positive feedback loop that eliminates these cell populations
through apoptosis. The expansion of the intestinal flora, the loss of IEC immune cells, or
a combination of the two would be sensed by both IECs and intestine-associated immune
cells, which would then activate an inflammatory response.
THE UPR IN ME TABOLIC DISEASE
In animals, cellular and organismal metabolism are coordinated through insulin
signaling. The hormone insulin, which is synthesized and secreted from pancreatic beta
cells into the circulatory system in mammals, functions as the "fed" signal in animal
metabolism. It binds to and activates the insulin receptor, a receptor tyrosine kinase that
is expressed in an array of tissues including liver, adipose, and muscle. This leads to
activation of insulin-receptor substrate 1 (IRS 1), which mediates responses that are
tailored for each cell type. These include increasing uptake of glucose by adipose and
muscle cells, inhibiting hepatic gluconeogenesis, and promoting adipocyte lipogenesis. A
disruption in insulin signaling in peripheral tissues, which can result from genetic
predisposition or chronic over-nutrition, is associated with metabolic disease, a collection
of pathologies including obesity, type 2 diabetes, and chronic inflammation.
Inflammation is a cytokine-mediated response to injury or infection that involves
vasodilation, swelling, and stimulation of immune cells. Chronic inflammation, a state in
which the systemic levels of proinflammatory cytokines are elevated, is well established
as a component of metabolic disease in both humans and mice (reviewed in Baker et al.
2011). The inflammatory state is intimately linked with metabolic state because both the
liver and adipose tissues - two tissues that play central roles in the regulation of
metabolism - are also major sites of proinflammatory cytokine production and secretion.
These tissues increase production of proinflammatory cytokines in response to excess
nutrition, and an increasing body of evidence suggests that the resulting inflammatory
state contributes to the pathology of metabolic disease (Baker et al. 2011).
Results from a number of studies have suggested that UPR signaling is a key
player in metabolic disease. In humans, a positive correlation was observed between body
mass index and ER stress markers in adipose tissue, and patients who experienced
substantial weight loss due to gastric by-pass surgery exhibited a decrease in ER stress
markers from both liver and adipose tissue (Gregor et al. 2009; Sharma et al. 2008). Mice
lacking one allele of XBP1 (XBP14- mice) fed a high-fat diet developed insulin resistance
and obesity to a greater extent than WT, and they exhibited elevated levels of PERK
activation in the liver (Ozcan et al. 2004). Furthermore, in leptin deficient (ob/ob) mice, a
genetic model for obesity, IRE 1 and PERK activation were both abnormally elevated in
liver and adipose tissues (Ozcan et al. 2004). Treatment of ob/ob mice with oral doses of
4-phenyl butyric acid (PBA), a chemical chaperone that improves ER protein folding,
concomitantly reduced both UPR activation in hepatocytes and blood levels of glucose
and insulin without affecting body weight (Ozcan et al. 2006). These data point to a
causative relationship between metabolic syndrome and ER stress in liver and adipose
tissues.
The mechanism by which metabolic syndrome causes ER stress is unclear, but it
has been speculated that it may result from the elevated requirement for synthesis of
lipids and lipoproteins in the liver and adipose tissues (reviewed in Hotamisligil 2010).
Alternative hypotheses posit that increased lipid accumulation, increased levels of
glucose, or cytokine signaling in response to excessive blood glucose levels may lead to
calcium release from the ER and production of ROS, thereby inhibiting ER protein
folding (Zhang and Kaufman 2008). Although studies in cell culture have demonstrated
that each of these potential sources of ER stress is sufficient to activate the UPR,
determining the degree to which they account for UPR activation in metabolic syndrome
in vivo presents a significant challenge.
Recent evidence suggests that the non-canonical UPR mediator cAMP response
element-binding protein H (CREBH), an ATF6 homolog expressed specifically in the
liver, may function in a positive feedback loop to exacerbate metabolic syndrome.
CREBH was activated, along with the transcription of genes regulated by the UPR, in
mice injected with the proinflammatory cytokine IL-6, and it was also activated in a
mouse genetic model for insulin resistance (Zhang et al. 2006; Lee et al. 2010).
Transcriptional profiling showed that CREBH regulates transcription of gluconeogenesis
and inflammatory genes (Lee et al., 2010). Importantly, RNAi knockdown of CREBH
was protective against metabolic disease, causing reduced blood glucose levels and
decreased production of components of the acute-phase inflammatory response (Zhang et
al. 2006).
The UPR might function upstream of inflammation in metabolic disease through
activation of the JNK, NF-KB, or a combination of both. JNK intersects with insulin
signaling by phosphorylating Irsl on a conserved serine residue (Ser307 in mice)
(Hirosumi et al. 2002). This inhibits the activating phosphorylation of Irs 1 by the insulin
receptor, resulting in insulin resistance in cultured cells (Hirosumi et al., 2002).
Surprisingly, however, knock-in mice expressing Irs1 in which Ser307 is mutated to
alanine exhibit enhanced insulin resistance when fed a high-fat diet (Copps et al. 2010),
suggesting that this pathway may not be the central mechanism by which UPR signaling
leads to metabolic syndrome in vivo.
As described previously, both the IRE 1 and PERK pathways positively regulate
the activation of the NF-KB pathway. Although originally described as a central mediator
of the innate immune response, the NF-KB pathway can be activated in most cell types in
response to inflammatory signaling and stress, and it has been shown to promote the
development of metabolic syndrome through several mechanisms. In the hypothalamus,
which responds to insulin and leptin signaling by negatively regulating food intake,
elevated NF-KB signaling through overexpression of IKKp in mice led to greater weight
gain than WT, whereas knockout of IKKp protected mice against developing obesity and
symptoms of metabolic syndrome (Zhang et al. 2008). Expression of constitutivelyactivated IKKP resulted in defective insulin signaling in both hepatocytes and muscle
cells, accompanied by systemic insulin resistance and glucose intolerance (Cai et al.
2005). A direct transcriptional target of NF-KB is the noncanonical IKK kinase IKK,
which both contributes to late phase NF-i<B activity and promotes NF-KB-independent
interferon signaling (Adli et al. 2006; Tenoever et al. 2007). IKKs was shown to be
induced in adipocytes and fat-infiltrating macrophages in mice fed a high-fat diet, and
IKKC7 mice were dramatically protected against the development of diet induced obesity
(Chiang et al. 2009). Although these data provide strong evidence suggesting that NF-KB
functions upstream of metabolic disease, there is little functional data implicating the
UPR upstream of NF-KB in metabolic disease in vivo. In one intriguing study, treatment
of mice fed on a high-fat diet with the chemical chaperone tauroursodeoxycholic acid
(TUDCA) suppressed the elevation in PERK and NF-KB activation observed in the
hypothalamus of untreated mice (Zhang, et al. 2008). Further work will be required to
more fully address the functional link between ER stress and NF-KB in metabolic disease.
The UPR in Caenorhabditis elegans
C. elegans has proved an attractive model in which to study the UPR for several
reasons. As C. elegans is a multicellular organism with distinct tissues, so the function of
the UPR can be studied in cell types analogous to those found in mammals, including
intestine, muscle, and neuron. However, C. elegans is also small, with a virtually
invariant and fully mapped cell lineage, hermaphroditic, and genetically tractable,
making it a particularly amenable system in which to examine the function of genes in
the context of whole-organism physiology. C. elegans was initially used to study the
UPR as part of the search for the metazoan homolog of HA C1. Through a screen for
mutants unable to activate transcription of hsp-4 (one of two C. elegans paralogs of BiP),
Calfon et al. were among the first groups to independently identify xbp-1 as the HAC1
homolog (Calfon et al. 2002). Subsequent work has shown that strains carrying loss of
function mutations in any of the three UPR sensors are all fully viable, but they each
exhibit increased sensitivity to exogenously induced ER stress, indicating that they are all
functional components of the C. elegans UPR (Bischof et al. 2008; Shen et al. 2001; Shen
et al. 2005).
To directly compare the individual contribution of each branch to the UPR, Shen
et al. performed a microarray study comparing the transcriptional profiles of loss-offunction mutants in ire-i, pek-1, atf-6, and xbp-1. They found that each branch was
important for the transcription of between 40 and 250 genes, many of which have known
function in ER or secretory pathway maintenance (Shen et al. 2005). A number of these
transcriptional targets were found to be regulated by more than one branch of the UPR,
demonstrating a redundancy in function between the branches that has not been directly
investigated in mammalian cells except for a handful of transcriptional targets. Additional
functional analyses have demonstrated the importance of particular transcriptional targets
of the UPR in protecting against both exogenous and physiological ER stress (Table 1).
Furthermore, they showed that combining deficiencies between any two branches of the
UPR resulted in early larval arrest (Shen et al. 2001; Shen et al. 2005). Whether loss of
one branch caused compensatory activation in the remaining branches was unclear, as
was the cause of lethality. The simplest explanation for the lethality is that the UPR is
required constitutively to sense and respond to ER stress throughout the organism, but
there could also be a more specific role for the UPR in producing factors that are required
for larval development.
Several studies have elucidated roles of the IRE-1 -XBP- 1 pathway in C. elegans
in more detail. In one, neurons deficient for IRE-I or XBP- 1 were shown to accumulate
ionotropic glutamate receptor subunits in the ER in the absence of exogenous ER stress
(Shim et al. 2004). The phenomenon of WT proteins aggregating in the ER of neurons in
which the UPR has been compromised is of particular interest because a common feature
between a number of neuropathologies is the production of a mutant secretory protein
Table 1. Functionally validated UPR effectors in C. elegans*
gene name
function
hsp-3
One of two functional homologs of BiP
apy-]
Hydrolyzes nucleoside diphosphates
generated by glycosyltransferases
ero-i
Oxidizes PDI to promote disulfide bond
D fxdzs
oprmti
fdebn
formation
sel-i
cdc-48. 1,
cdc-48.2
ufd-I
npl-4.2
hrd-1,
hrd/-i,
marc-6
jamp-i
dnj-7
hut-I
rnf121
Regulated by UPR
componentsc
ire-i, xbp-I
loss-of-function phenotype
Reference
synthetic larval lethal with xbp-i
RNAi depletion caused reduced
brood size and lifespan, slow larval
development, disrupted muscle and
parnyx
Kapulkin et al., 2005
Uccelletti et al., 2008
ire-i, xbp-i
RNAi depletion reduces lifespan
Harding et al., 2003
ER-associated degradation
ire-i, xbp-I
ER-associated degradation
ire-i, xbp- I
partially synthetic larval lethal with
xbp-i (10% lethality)
increased sensitivity to TM
Urano et al., 2002; Ye et
al., 2004
Mouysset et al., 2006
ER-associated degradation
ER-associated degradation
Paralogous E3 ubiquitin ligases in ERasoitddgaainnone
associated degradationnot
ER-associated degradation
none detected
none detected
ire-i, xhp-i,
detected,
tested
none detected
Mouysset et al., 2006
Mouysset et al., 2006
cotranslational degradation of ERlocalized proteins
Transports UDP-Gal into ER
ire-, xbp-I
increased sensitivity to TM
increased sensitivity to TM
synthetic larval lethal, increased
sensitivity to exogenously induced
ER stress
increased sensitivity to TM
synthetic larval lethal with ire-i
Tcherpakov et al., 2008
Oyadomari et al., 2006
ire-i, xbp- I
ts larval lethal
Dejima et al., 2009
E3 ubiquitin ligase in ER-associated
nn
eetdicesdsniiiyt
R
Sasagawa et al., 2007
MDrme
idlocsdegradationed reduced
TM, Tunicamycin; ts, temperature-sensitive
*Depletion of gene leads to elevated IRE- I activation, as measured by expression of BiP using the Phsp-4.. GFP(zcls4) reporter.
'iRegulation of expression from microarray study by Shen et al., 2005.
of expression as determined by Ucelletti et al.,nRegulation
2008.
l,21
that forms aggregates in the ER (reviewed in Soto and Estrada 2008). Such aggregates
have been shown to cause chronic ER stress, suggestive that they may compromise the
UPR such that glutamate receptors and other synaptic components would likewise fail to
exit the ER, leading to impaired neural function (Soto and Estrada, 2008).
Another recent study by Henis-Korenbilt et al. linked the IRE- 1-XBP- 1 pathway
to insulin signaling in C. elegans (Henis-Korenblit et al. 2010). They found that strains
carrying hypomorphic alleles of the insulin receptor daf-2, which were known to exhibit
dramatically extended lifespans and have enhanced resistance to various forms of stress,
likewise had enhanced resistance against ER stress. This enhanced resistance was
completely dependent on the IRE- 1-XBP- 1 branch of the UPR, but paradoxically, both
the basal and inducible levels of IRE- 1-XBP- 1 pathway activation were markedly
reduced in the daf2 mutant. Nevertheless, these data rule out a model in which reduced
DAF-2 function leads to protection against ER stress through hyperactivation of the UPR,
indicating instead that the daf2 mutant is protected against ER stress through some
mechanism that functions synergistically with the IRE- 1-XBP- 1 pathway. The reduced
IRE- 1-XBP- 1 pathway activation in the daf-2 mutant was dependent upon the FOXO
transcription factor DAF-16, which is negatively regulated by DAF-2, but loss of daf-16
only slightly suppressed the enhanced resistance of the daf2 mutant to ER stress (HenisKorenbilt et al. 2010). These results indicate that some other factor regulated by daf-2 is
involved in the synergistic interaction with the IRE- 1-XBP- 1 pathway. Further
investigation will be required to obtain a more complete understanding of this interaction
between insulin signaling and the UPR in C. elegans; perhaps such studies will contribute
to the understanding of the interaction between the UPR and metabolic disease in
mammals.
Caenorhabditis elegans immunity and the UPR
The C. elegans immune response is simple relative to that of mammals,
representing ancient conserved mechanisms of pathogen resistance. As there are no
specialized immune cells, signaling pathways that regulate immunity are activated within
the tissues at the site of infection. For most C. elegans-pathogen interactions described,
which include both human and indigenous pathogens and represent bacteria, fungi, and
viruses, the site of infection is either the intestine or the hypodermis. A number of
conserved signaling pathways have been described to play a role in immunity, many of
which are also involved in resistance to other forms of stress responses or development,
underscoring the ancestral origins of immune signaling (Garsin et al. 2003; GravatoNobre et al. 2005; Inoue et al. 2005; Shapira et al. 2006; Tong et al. 2009).
For the majority of pathogens studied, the principal regulator of the C. elegans
immune response is the p38/PMK-1 MAP kinase pathway (Kim et al. 2002). The role of
the PMK- 1 pathway in C. elegans immunity was identified through a genetic screen for
mutants with enhanced susceptibility to the human pathogen Pseudomonas aeruginosa
PA 14. Whereas a WT population of C. elegans infected with P. aeruginosaPA 14 dies
with an LT50 of about 4 days, the LT50 forpmk-1 loss-of-function mutants is
approximately 2 days, underscoring the central role of the PMK- 1 pathway in defense
against this pathogen. As in mammals, the upstream activator of PMK- 1 is the MAPKKK
ASKl/NSY-1, which signals through MAPKKs MKK(3 and 6)/SEK-1 (Kim et al.,
2002). Whereas mammalian p38 is activated in response to infection through known
PRRs and cytokine receptors, however, the most upstream molecules responsible for
sensing infection to activate PMK- 1 are unknown. The mammalian the p38 pathway,
together with two other MAPK pathways and the NF-KB pathway, is the central regulator
of the inflammatory response in both specialized immune cells and a diversity of other
cells, from endothelial cells to neurons (reviewed in Saklatvala 2004). It also plays a role
in responding to a variety of other extracellular stimuli including UV light, heat, osmotic
shock, and growth factors (Zarubin et al. 2005). Likewise, the PMK- 1 pathway in C.
elegans not only regulates the immune response, but it has also been demonstrated to
protect against diverse sources of stress, including sterile wounding and oxidative stress
(Pei et al. 2008; Pujol et al. 2008).
To identify the immune effectors regulated by the PMK- 1 pathway, Troemel et al.
performed a transcriptional profiling study of both WT C. elegans in response to
infection with P. aeruginosaPA14 and a pmk-1 loss-of-function mutant (Troemel et al.
2006). Troemel et al. identified 86 genes whose expression was dependent on pmk-1, and
25% of these were among the approximately 300 genes that are upregulated upon
infection with P. aeruginosa PA14. The list of immune effectors included C-type lectins,
ShK toxins, and detoxifying enzymes such as glutathione-S-transferases. As in the
mammalian innate immune response, many of the C. elegans immune effectors are
secretory proteins. Although stringent criteria for identifying transcriptional changes may
account for some of the difference between genes regulated by the PMK- 1 pathway and
genes induced by exposure to pathogen, it is likely that there are likely other pathways
involved in the transcriptional response to P. aeruginosaPA14 that may be less
functionally important for resistance to infection.
In a genetic screen for genes involved in the protective response against Bacillus
thuringiensistoxin Cry5B, a pore-forming toxin that targets intestinal cells of insects and
nematodes, the Aroian lab identified a mutant predicted to be deficient in N-linked
glycosylation (Bischof et al. 2008). As such a mutation would cause ER stress, they
hypothesized that Cry5B might be an additional source of ER stress, and they therefore
tested the role of the UPR in protection against Cry5B. Indeed, they found that exposure
to Cry5B caused activation of the IRE- 1-XBP- 1 pathway in the intestine, and that both
the IRE-1 and ATF-6 pathways were important for survival in the presence of Cry5B.
Further, they showed that the PMK- 1 pathway, a known regulator of the response to
Cry5B, was genetically upstream of the IRE-1-XBP-1 response to Cry5B (Huffman et al.
2004; Bischof et al. 2008). They therefore postulated that the UPR acted as a downstream
component of the PMK- 1-mediated immune response, either to prepare the intestinal
cells to accommodate an increase in ER protein load as immune effectors are produced or
to increase lipid biosynthesis to repair the damaged cell membrane (Bischof et al. 2008).
The p38 pathway had been linked both upstream and downstream to the UPR in
mammalian cells (Devries-Seimon et al. 2005; Luo et al. 2002; Nguyen et al. 2004;
Ranganathan et al. 2006), but this was the first functional study linking the p38 pathway
to the UPR in any system, and future studies to further characterize the interaction
between PMK- 1 and the UPR in defense against Cry5B may prove relevant for
understanding this interaction in mammals.
Summary
The activation of the UPR and its direct downstream effects to restore ER
homeostasis have been well characterized in both S. cerevisiae and mammalian cell
types, and much work has gone into identifying the physiological roles for the UPR. In
yeast, the UPR is activated not only by adverse environmental conditions that negatively
impact ER protein folding, but also during cell division, when the ER must be expanded
prior to being divided between the mother and daughter cells. In mammals, genetic
studies have identified a requirement for individual branches of the UPR in the
development of highly secretory cell types, although the nature of this requirement
remains incompletely understood. Comparatively little was known about the role of the
UPR in cells whose primary function is not protein secretion. Furthermore, genetic
investigation of the UPR has predominantly been performed by manipulating one branch
at a time, so the extent to which the three branches can functionally compensate for each
other has been unclear. Studies in C. elegans have defined the overlapping transcriptional
and functional output of the UPR and also linked UPR activity to neuronal function,
insulin signaling, and immunity.
The following work further defines physiological functions of the UPR in C.
elegans. In Chapter Two, the IRE-1-XBP- 1 branch of the UPR is shown to protect the
host against pathogen-induced immune activation. In Chapter Three, evidence is
presented to suggest that the PEK- 1 and IRE- 1 branches of the UPR play functionally
redundant roles in the constitutive maintenance of ER homeostasis during both immune
activation and basal growth conditions. Chapter Four outlines the initiation of a genetic
screen to identify the mechanism by which the IRE-I pathway protects against immune
activation. Finally, future approaches to elucidating the function of the UPR through
studies in C. elegans are presented in Chapter Five.
References
Acosta-Alvear, D., Zhou, Yiming, Blais, A., Tsikitis, M., Lents, N. H., Arias, C., Lennon,
C. J., Kluger, Y., and Dynlacht, B. D. (2007). XBPl controls diverse cell type- and
condition-specific transcriptional regulatory networks. Molecular Cell 27, 53-66.
Adli, M., and Baldwin, A. S. (2006). IKK-i/IKKepsilon controls constitutive, cancer cellassociated NF-kappaB activity via regulation of Ser-536 p65/RelA phosphorylation.
The Journal of Biological Chemistry 281, 26976-84.
Aebi, M., Bernasconi, R., Clerc, S., and Molinari, M. (2010). N-glycan structures:
recognition and processing in the ER. Trends in Biochemical Sciences 35, 74-82.
Bae, Y. S., Lee, J. H., Choi, S. H., Kim, S., Almazan, F., Witztum, J. L., and Miller, Y. I.
(2009). Macrophages generate reactive oxygen species in response to minimally
oxidized low-density lipoprotein: toll-like receptor 4- and spleen tyrosine kinasedependent activation of NADPH oxidase 2. Circulation Research 104, 210-8, 21p
following 218.
Baker, R. G., Hayden, M. S., and Ghosh, S. (2011). NF-KB, inflammation, and metabolic
disease. Cell Metabolism 13, 11-22.
Bertolotti, A., Wang, X., Novoa, I, Jungreis, R., Schlessinger, K., Cho, J. H., West, A. B.,
and Ron, D. (2001). Increased sensitivity to dextran sodium sulfate colitis in
IRElbeta-deficient mice. The Journal of Clinical Investigation 107, 585-593.
Bertolotti, A., Zhang, Y, Hendershot, L. M., Harding, H. P., and Ron, D. (2000).
Dynamic interaction of BiP and ER stress transducers in the unfolded-protein
response. Nature Cell Biology 2, 326-332.
Bicknell, A. A., Babour, A., Federovitch, C. M., and Niwa, M. (2007). A novel role in
cytokinesis reveals a housekeeping function for the unfolded protein response. The
Journal of Cell Biology 177, 1017-1027.
Bischof, L. J., Kao, C.-Y., Los, F. C. 0., Gonzalez, M. R., Shen, Z., Briggs, S. P., Goot,
F. G. van der, and Aroian, R. V. (2008). Activation of the Unfolded Protein
Response is required for defenses against bacterial pore-forming toxin in vivo. PLoS
Pathogens 4, e100176. doi: 10.1371/journal.ppat.1000176.
Bommiasamy, H., Back, S.H., Fagone, P. L., Kyungho, M. S., et al. (2009). ATF6alpha
induces XBP 1-independent expansion of the endoplasmic reticulum. Journal of Cell
Science 122, 1626-1636.
Brodsky, J. L., Goeckeler, J., and Schekman, R. (1995). BiP and Sec63p are required for
both co- and posttranslational protein translocation into the yeast endoplasmic
reticulum. Proceedings of the National Academy of Sciences 92, 9643-9646.
Brunsing, R., Omori, S. A., Weber, F., Bicknell, A., Friend, L., Rickert, R., and Niwa, M.
(2008). B- and T-cell development both involve activity of the unfolded protein
response pathway. The Journal of Biological Chemistry 283, 17954-17961.
Cai, D., Yuan, M., Frantz, D. F., Melendez, P. A., Hansen, L., Lee, Jongsoon, and
Shoelson, S. E. (2005). Local and systemic insulin resistance resulting from hepatic
activation of IKK-beta and NF-kappaB. Nature Medicine 11, 183-190.
Calfon, M., Zeng, H. Q., Urano, F., Till, J. H., Hubbard, S. R., Harding, H. P., Clark, S.
G., and Ron, D. (2002). IRE1 couples endoplasmic reticulum load to secretory
capacity by processing the XBP-l mRNA. Nature 415, 92-96.
Chapman, R. E., and Walter, P. (1997). Translational attenuation mediated by an mRNA
intron. Current Biology: 7, 850-859.
Chen, X., Shen, J., and Prywes, R. (2002). The luminal domain of ATF6 senses
endoplasmic reticulum (ER) stress and causes translocation of ATF6 from the ER to
the Golgi. The Journal of Biological Chemistry 277, 13045-52.
Chiang, S.-H., Bazuine, M., Lumeng, C. N., Geletka, L. M., Mowers, J., et al. (2009).
The protein kinase IKKepsilon regulates energy balance in obese mice. Cell 138,
961-975.
Copps, K. D., Hancer, N. J., Opare-Ado, L., Qiu, W., Walsh, C., and White, M. F.
(2010). Irs1 serine 307 promotes insulin sensitivity in mice. Cell Metabolism 11, 8492.
Cox, J. S., Shamu, C. E., and Walter, P. (1993). Transcriptional induction of genes
encoding endoplasmic reticulum resident proteins requires a transmembrane protein
kinase. Cell 73, 1197-1206.
Cox, J. S., and Walter, P. (1996). A novel mechanism for regulating activity of a
transcription factor that controls the unfolded protein response. Cell 87, 391-404.
Credle, J. J., Finer-Moore, J. S., Papa, F. R., Stroud, R. M., and Walter, P. (2005). On the
mechanism of sensing unfolded protein in the endoplasmic reticulum. Proceedings
of the National Academy of Sciences of the United States of America 102, 1877311784.
Cullinan, S. B., Zhang, D., Hannink, M., Arvisais, E., Kaufman, R. J., and Diehl, J. A.
(2003). Nrf2 Is a direct PERK substrate and effector of PERK-dependent cell
survival. Molecular and Cellular Biology 23, 7198-7209.
Cullinan, S. B., and Diehl, J. A. (2004). PERK-dependent activation of Nrf2 contributes
to redox homeostasis and cell survival following endoplasmic reticulum stress. The
Journal of Biological Chemistry 279, 20108-20117.
Darom, A., Bening-Abu-Shach, U., Broday, L. (2010). RNF-121 is an endoplasmic
reticulum-membrane E3 ubiquitin ligase involved in the regulation of p-integrin.
Molecular Biology of the Cell 21, 1788-1798.
Dejima, K., Murata, D., Mizuguchi, S., Nomura, K. H., Gengyo-Ando, K., Mitani, S.,
Kamiyama, S., Nishihara, S., Nomura, K. (2009). The ortholog of human solute
carrier family 35 member B 1 (UDP-galactose transporter-related protein 1) is
involved in maintenance of ER homeostasis and essential for larval development in
Caenorhabditiselegans. The FASEB Journal 23, d.o.i. 10.1096/fj.08-123737.
Deldpine, M., Nicolino, M., Barrett, T., Golamaully, M., Lathrop, G. M., and Julier, C.
(2000). EIF2AK3, encoding translation initiation factor 2-alpha kinase 3, is mutated
in patients with Wolcott-Rallison syndrome. Nature Genetics 25, 406-409.
Deng, J., Lu, P. D., Zhang, Y., Scheuner, D., Kaufman, R. J., Sonenberg, N., Harding, H.
P., and Ron, D. (2004). Translational repression mediates activation of nuclear
factor kappa B by phosphorylated translation initiation factor 2. Molecular and
Cellular Biology 24, 10161-10168.
Devries-Seimon, T., Li, Y., Yao, P. M., Stone, E., Wang, Y., Davis, R. J., Flavell, R., and
Tabas, I. (2005). Cholesterol-induced macrophage apoptosis requires ER stress
pathways and engagement of the type A scavenger receptor. The Journal of Cell
Biology 171, 61-73.
Dierks, T. et al. (1996). A microsomal ATP-binding protein involved in efficient protein
transport into the mammalian endoplasmic reticulum. The EMBO Journal 15, 69316942.
Frydman, J. (2001). Folding of newly translated proteins in vivo: the role of molecular
chaperones. Annual review of Biochemistry 70, 603-647.
Garsin, D. A., Villanueva, J. M., Begun, J., Kim, D. H., Sifri, C. D., Calderwood, S. B.,
Ruvkun, G., and Ausubel, F. M. (2003). Long-lived C. elegans daf-2 mutants are
resistant to bacterial pathogens. Science 300, 1921.
Gething, M. J. (1999). Role and regulation of the ER chaperone BiP. Seminars in Cell &
Developmental Biology 10, 465-472.
Gething, M. J., and Sambrook, J. (1992). Protein folding in the cell. Nature 355, 33-45.
Gravato-Nobre, M. J., Nicholas, H. R., Nijland, R., O'Rourke, D., Whittington, D. E.,
Yook, K. J., and Hodgkin, J. (2005). Multiple genes affect sensitivity of
Caenorhabditiselegans to the bacterial pathogen Microbacteriumnematophilum.
Genetics 171, 1033-1045.
Gregor, M. F., Yang, L., Fabbrini, E., Mohammed, B. S., Eagon, J. C., Hotamisligil, G.
S., and Klein, S. (2009). Endoplasmic reticulum stress is reduced in tissues of obese
subjects after weight loss. Diabetes 58, 693-700.
Han, D., Lerner, A. G., Vande Walle, L., Upton, J.-P., Xu, W., Hagen, A., Backes, B. J.,
Oakes, S. A., and Papa, F. R. (2009). IRElalpha kinase activation modes control
alternate endoribonuclease outputs to determine divergent cell fates. Cell 138, 562575.
Harding, H. P., Zhang, Y., and Ron, D. (1999). Protein translation and folding are
coupled by an endoplasmic-reticulum-resident kinase. Nature 397, 271-274.
Harding, H. P., Zhang, Y., Zeng, H., Novoa, I., Lu, P. D., et al. (2003). An integrated
stress response regulates amino acid metabolism and resistance to oxidative stress.
Molecular Cell 11, 619-633.
Haze, K., Yoshida, H., Yanagi, H., Yura, T., and Mori, K. (1999). Mammalian
transcription factor ATF6 is synthesized as a transmembrane protein and activated
by proteolysis in response to endoplasmic reticulum stress. Molecular Biology of the
Cell 10, 3787-3799.
Hedge, R. S., and Ploegh, H. (2010). Quality and quantity control at the endoplasmic
reticulum. Current Opinion in Cell Biology 22, 437-446.
Henis-Korenblit, S., Zhang, Peichuan, Hansen, M., McCormick, M., Lee, S.-J., Cary, M.,
and Kenyon, C. (2010). Insulin/IGF-1 signaling mutants reprogram ER stress
response regulators to promote longevity. Proceedings of the National Academy of
Sciences of the United States of America 107, 9730-9735.
Hetz, C., Bernasconi, P., Fisher, J., Lee, A.-H., Bassik, M. C., et al. (2006). Proapoptotic
BAX and BAK modulate the unfolded protein response by a direct interaction with
IRElalpha. Science 312, 572-576.
Hirosumi, J., Tuncman, G., Chang, L., Gbrgian, C. Z., Uysal, K. T., Maeda, K., Karin, M.,
and Hotamisligil, G. S. (2002). A central role for JNK in obesity and insulin
resistance. Nature 420, 333-336.
Hollien, J., Lin, J. H., Li, H., Stevens, N., Walter, P., and Weissman, J. S. (2009).
Regulated Ire 1-dependent decay of messenger RNAs in mammalian cells. The
Journal of Cell Biology 186, 323-33 1.
Hollien, J., and Weissman, J. S. (2006). Decay of endoplasmic reticulum-localized
mRNAs during the unfolded protein response. Science 313, 104-107.
Hotamisligil, G. S. (2010). Endoplasmic reticulum stress and the inflammatory basis of
metabolic disease. Cell 140, 900-917.
Hu, C.-C. A., Dougan, S. K., McGehee, A. M., Love, J. C., and Ploegh, H. L. (2009).
XBP- 1 regulates signal transduction, transcription factors and bone marrow
colonization in B cells. EMBO Journal 28, 1624-1636.
Hu, P., Han, Z., Couvillon, A. D., Kaufman, R. J., and Exton, J. H. (2006). Autocrine
tumor necrosis factor alpha links endoplasmic reticulum stress to the membrane
death receptor pathway through IRE 1alpha-mediated NF-kappaB activation and
down-regulation of TRAF2 expression. Molecular and Cellular Biology 26, 30713084.
Huffman, D. L., Abrami, L., Sasik, R., Corbeil, J., Goot, F. G. van der, and Aroian, R. V.
(2004). Mitogen-activated protein kinase pathways defend against bacterial poreforming toxins. Proceedings of the National Academy of Sciences of the United
States of America 101, 10995-1000.
Inoue, H., Hisamoto, N., An, J. H., Oliveira, R. P., Nishida, E., Blackwell, T. K., and
Matsumoto, K. (2005). The C. elegans p38 MAPK pathway regulates nuclear
localization of the transcription factor SKN-1 in oxidative stress response. Genes &
development 19, 2278-2283.
Iwakoshi, N. N., Lee, A.-H., Vallabhajosyula, P., Otipoby, K. L., Rajewsky, K., and
Glimcher, L. H. (2003). Plasma cell differentiation and the unfolded protein
response intersect at the transcription factor XBP- 1. Nature Immunology 4, 321-329.
Jiang, H.-Y., Wek, S. A., McGrath, B. C., Scheuner, D., Kaufman, R. J., Cavener, D. R.,
and Wek, R. C. (2003). Phosphorylation of the a subunit of eukaryotic initiation
factor 2 is required for activation of NF- KB in response to diverse cellular stresses.
Molecular and Cellular Biology 23, 5651-5663.
Johnson, A. E., and Waes, M. A. van (1999). The translocon: a dynamic gateway at the
ER membrane. Annual Review of Cell and Developmental Biology 15, 799-842.
Jonikas, M. C., Collins, S. R., Denic, V., Oh, E., Quan, E. M., et al. (2009).
Comprehensive characterization of genes required for protein folding in the
endoplasmic reticulum. Science 323, 1693-1697.
Kapulkin, V., Hiester, B. G., Link, C. D. (2005). Compensatory regulation among ER
chaperones in C. elegans. FEBS Letters 579, 3063-3068.
Kaser, A., Lee, A.-H., Franke, A., Glickman, J. N., Zeissig, S., et al. (2008). XBP1 links
ER stress to intestinal inflammation and confers genetic risk for human
inflammatory bowel disease. Cell 134, 743-756.
Kensler, T. W., Wakabayashi, N., and Biswal, S. (2007). Cell Survival Responses to
Environmental Stresses Via the Keap 1-Nrf2-ARE Pathway. Annual Review of
Pharmacology and Toxicology 47, 89-116.
Kim, D. H., Feinbaum, R., Alloing, G., Emerson, F. E., Garsin, D. A., et al. (2002). A
conserved p38 MAP kinase pathway in Caenorhabditis elegans innate immunity.
Science 297, 623-626.
Kohno, K., Normington, K., Sambrook, J., Gething, M. J., and Mori, K. (1993). The
promoter region of the yeast KAR2 (BiP) gene contains a regulatory domain that
responds to the presence of unfolded proteins in the endoplasmic reticulum.
Molecular and Cellular Biology 13, 877-890.
Korennykh, A. V., Egea, P. F., Korostelev, A. A., Finer-Moore, J., Zhang, C., Shokat, K.
M., Stroud, R. M., and Walter, P. (2009). The unfolded protein response signals
through high-order assembly of Ire 1. Nature 457, 687-693.
Kozutsumi, Y., Segal, M., Normington, K., Gething, M. J., and Sambrook, J. (1988). The
presence of malfolded proteins in the endoplasmic reticulum signals the induction of
glucose-regulated proteins. Nature 332, 462-464.
Lee, A.S. (1987). Coordinated regulation of a set of genes by glucose and calcium
ionophores in mammalian-cells. Trends in Biochemical Sciences 12, 20-23.
Lee, A.-H., Iwakoshi, N. N., and Glimcher, L. H. (2003). XBP-1 regulates a subset of
endoplasmic reticulum resident chaperone genes in the unfolded protein response.
Molecular and Cellular Biology 23, 7448-7459.
Lee, A.-H., Scapa, E. F., Cohen, D. E., and Glimcher, L. H. (2008). Regulation of hepatic
lipogenesis by the transcription factor XBP1. Science 320, 1492-1496.
Lee, K., Tirasophon, W., Shen, X., Michalak, M., Prywes, R., Okada, T., Yoshida, H.,
Mori, K., and Kaufman, R. J. (2002). IRE 1-mediated unconventional mRNA
splicing and S2P-mediated ATF6 cleavage merge to regulate XBP 1 in signaling the
unfolded protein response. Genes & Development 16, 452-466.
Lee, M.-W. Chanda, D., Yang, J., Oh, H., Kim, S. S., et al. (2010). Regulation of hepatic
gluconeogenesis by an ER-bound transcription factor, CREBH. Cell Metabolism
11, 331-339.
Leung, L., Kwong, M., Hou, S., Lee, C., and Chan, J. Y. (2003). Deficiency of the Nrfl
and Nrf2 transcription factors results in early embryonic lethality and severe
oxidative stress. The Journal of Biological Chemistry 278, 48021-48029.
Li, H., Korennykh, A. V., Behrman, S. L., and Walter, P. (2010). Mammalian
endoplasmic reticulum stress sensor IREl signals by dynamic clustering.
Proceedings of the National Academy of Sciences of the United States of America
107, 16113-16118.
Lin, J. H., Li, H., Yasumura, D., Cohen, H. R., Zhang, C., Panning, B., Shokat, K. M.,
Lavail, M. M., and Walter, P. (2007). IRE1 signaling affects cell fate during the
unfolded protein response. Science 318, 944-949.
Liou, H. C., Boothby, M. R., Finn, P. W., Davidon, R., Nabavi, N., Zeleznik-Le, N. J.,
Ting, J. P., and Glimcher, L. H. (1990). A new member of the leucine zipper class of
proteins that binds to the HLA DR alpha promoter. Science 247, 1581-1584.
Liu, C Y, Schr6der, M, and Kaufman, R. J. (2000). Ligand-independent dimerization
activates the stress response kinases IRE 1 and PERK in the lumen of the
endoplasmic reticulum. The Journal of Biological Chemistry 275, 24881-24885.
Luo, S., and Lee, Amy S (2002). Requirement of the p38 mitogen-activated protein
kinase signalling pathway for the induction of the 78 kDa glucose-regulated
protein/immunoglobulin heavy-chain binding protein by azetidine stress: activating
transcription factor 6 as a target for stress-induced phosphorylation. The
Biochemical Journal 366, 787-795.
Ma, Y., Brewer, J. W., Diehl, J. A., and Hendershot, L. M. (2002). Two distinct stress
signaling pathways converge upon the CHOP promoter during the mammalian
unfolded protein response. Journal of Molecular Biology 318, 1351-1365.
Martinon, F., Chen, X., Lee, A.-H., and Glimcher, L. H. (2010). TLR activation of the
transcription factor XBP1 regulates innate immune responses in macrophages.
Nature Immunology 11, 411-418.
McGehee, A. M., Dougan, S. K., Klemm, E. J., Shui, G., Park, B., et al. (2009). XBP-1deficient plasmablasts show normal protein folding but altered glycosylation and
lipid synthesis. Journal of Immunology 183, 3690-3699.
Mori, K, Ma, W., Gething, M. J., and Sambrook, J. (1993). A transmembrane protein
with a cdc2+/CDC28-related kinase activity is required for signaling from the ER to
the nucleus. Cell 74, 743-756.
Mori, K, Sant, A., Kohno, K, Normington, K., Gething, M. J., and Sambrook, J. F.
(1992). A 22 bp cis-acting element is necessary and sufficient for the induction of
the yeast KAR2 (BiP) gene by unfolded proteins. The EMBO Journal 11, 2583-2593.
Mouysset, J., Kahler, C., Hoppe, T. A. (2006). conserved role of Caenorhabditiselegans
CDC-48 in ER-associated protein degradation. Journal of Structural Biology 156,
41-49.
Nadanaka, S., Okada, T., Yoshida, H., and Mori, K. (2007). Role of disulfide bridges
formed in the luminal domain of ATF6 in sensing endoplasmic reticulum stress.
Molecular and Cellular Biology 27, 1027-1043.
Nandi, A., Kitamura, Y., Kahn, C. R., and Accili, D. (2004). Mouse models of insulin
resistance. Physiological Reviews 84, 623-647.
Nguyen, D. T. Kebache, S., Fazel, A., Wong, H. N., Jenna, S., et al. (2004). Nckdependent activation of extracellular signal-regulated kinase- 1 and regulation of cell
survival during endoplasmic reticulum stress. Molecular Biology of the Cell 15,
4248-4260.
Nishitoh, H., Matsuzawa, A., Tobiume, K., Saegusa, K., Takeda, K., Inoue, K., Hori, S.,
Kakizuka, A., and Ichijo, H. (2002). ASKI is essential for endoplasmic reticulum
stress-induced neuronal cell death triggered by expanded polyglutamine repeats.
Genes & Development 16, 1345-1355.
Oikawa, D., Kimata, Y., Kohno, K., and Iwawaki, T. (2009). Activation of mammalian
IRE 1alpha upon ER stress depends on dissociation of BiP rather than on direct
interaction with unfolded proteins. Experimental Cell Research 315, 2496-2504.
Okamura, K., Kimata, Y, Higashio, H., Tsuru, A., and Kohno, K. (2000). Dissociation of
Kar2p/BiP from an ER sensory molecule, Ire lp, triggers the unfolded protein
response in yeast. Biochemical and Biophysical Research Communications 279,
445-450.
Oyadomari, S., Harding, H. P., Zhang, Y., Oyadomari, M., and Ron, D. (2008).
Dephosphorylation of translation initiation factor 2alpha enhances glucose tolerance
and attenuates hepatosteatosis in mice. Cell Metabolism 7, 520-532.
Oyadomari, S., Yun, C., Fisher, E. A., Kreglinger, N., Kreibich, G., Oyadomari, M.,
Harding, H. P., Goodman, A. G., Harant, H., Garrison, J. L., Taunton, J., Katze, M.
G., Ron, D. (2006). Cotranslational degradation protects the stressed endoplasmic
reticulum from protein overload. Cell 126, 727-739.
Ozcan, U., Cao, 0., Yilmaz, E., Lee, A.-H., Iwakoshi, N. N., et al. (2004). Endoplasmic
reticulum stress links obesity, insulin action, and type 2 diabetes. Science 306, 457461.
Ozcan, U., Yilmaz, E., Ozcan, L., Furuhashi, M., Vaillancourt, E., Smith, R. 0., Girgun,
C. Z., and Hotamisligil, G. S. (2006). Chemical chaperones reduce ER stress and
restore glucose homeostasis in a mouse model of type 2 diabetes. Science 313,
1137-1140.
Park, S. W., Zhou, Y., Lee, J., Lu, A., Sun, C., Chung, J., Ueki, K., and Ozcan, U. (2010).
The regulatory subunits of P13K, p85alpha and p85beta, interact with XBP-l and
increase its nuclear translocation. Nature Medicine 16, 429-437.
Pei, B., Wang, S., Guo, X., Wang, Jun, Yang, G., Hang, H., and Wu, L. (2008). Arseniteinduced germline apoptosis through a MAPK-dependent, p53-independent pathway
in Caenorhabditiselegans. Chemical Research in Toxicology 21, 1530-1535.
Pincus, D., Chevalier, M. W., Arag6n, T., Anken, E. van, Vidal, S. E., El-Samad, H., and
Walter, P. (2010). BiP binding to the ER-stress sensor Irel tunes the homeostatic
behavior of the unfolded protein response. PLoS Biology 8, e1000415. doi:
10.137 1/journal.pbio. 1000415.
Pujol, N., Cypowyj, S., Ziegler, K., Millet, A., Astrain, A., Goncharov, A., Jin, Y.,
Chisholm, A. D., and Ewbank, J. J. (2008). Distinct innate immune responses to
infection and wounding in the C. elegans epidermis. Current Biology 18, 481-489.
Ranganathan, A. C., Zhang, L., Adam, A. P., and Aguirre-Ghiso, J. A. (2006). Functional
coupling of p38-induced up-regulation of BiP and activation of RNA-dependent
protein kinase-like endoplasmic reticulum kinase to drug resistance of dormant
carcinoma cells. Cancer Research 66, 1702-1711.
Reimold, A. M., Iwakoshi, N. N., Manis, J., Vallabhajosyula, P., Szomolanyi-Tsuda, E.,
et al. (2001). Plasma cell differentiation requires the transcription factor XBP-1.
Nature 412, 300-307.
Reimold, A. M., Ponath, P. D., Li, Y. S., Hardy, R. R., David, C. S., Strominger, J. L.,
and Glimcher, L. H. (1996). Transcription factor B cell lineage-specific activator
protein regulates the gene for human X-box binding protein 1. Journal of
Experimental Medicine 183, 393-401.
Ron, D., and Walter, P. (2007). Signal integration in the endoplasmic reticulum unfolded
protein response. Nature Reviews Molecular Cell Biology 8, 519-529.
Roy, B., and Lee, Amy S. (1999). The mammalian endoplasmic reticulum stress response
element consists of an evolutionarily conserved tripartite structure and interacts with
a novel stress-inducible complex. Nucleic Acids Research 27, 1437-1443.
Saklatvala, J. (2004). The p38 MAP kinase pathway as a therapeutic target in
inflammatory disease. Current Opinion in Pharmacology 4, 372-377.
Sasagawa, Y., Yamanaka, K., Ogura, T. (2007). ER E3 ubiquitin ligase HRD-1 and its
specific partner chaperone BiP play important roles in ERAD and developmental
growth in Caenorhabditiselegans. Genes to Cells 12, 1063-1073.
Scheuner, D., Song, B., McEwen, E., Liu, C., Laybutt, R., Gillespie, P., Saunders, T.,
Bonner-Weir, S., and Kaufman, R. J. (2001). Translational control is required for the
unfolded protein response and in vivo glucose homeostasis. Molecular Cell 7, 11651176.
Schrag, J. D., Bergeron, J. J., Li, Y, Borisova, S., Hahn, M., Thomas, D. Y., and Cygler,
M. (2001). The Structure of calnexin, an ER chaperone involved in quality control
of protein folding. Molecular Cell 8, 633-644.
Schrider, M., Chang, J. S., and Kaufman, R. J. (2000). The unfolded protein response
represses nitrogen-starvation induced developmental differentiation in yeast. Genes
& Development 14, 2962-2975.
Schr~der, M., and Kaufman, R. J. (2005). The mammalian unfolded protein response.
Annual Reviews in Biochemistry 74, 739-789.
SchrOder, M., Clark, R., Liu, C. Y., and Kaufman, R. J. (2004). The unfolded protein
response represses differentiation through the RPD3-SIN3 histone deacetylase. The
EMBO Journal 23, 2281-2292.
Sevier, C. S., and Kaiser, C. A. (2008). Erol and redox homeostasis in the endoplasmic
reticulum. Biochimica et Biophysica Acta 1783, 549-556.
Shaffer, A. L. Shapiro-Shelef, M., Iwakoshi, N. N. Zhao, H., Yu, X., et al. (2004). XBP1,
downstream of Blimp-1, expands the secretory apparatus and other organelles, and
increases protein synthesis in plasma cell differentiation. Immunity 21, 81-93.
Shamu, C. E., and Walter, P. (1996). Oligomerization and phosphorylation of the Irelp
kinase during intracellular signaling from the endoplasmic reticulum to the nucleus.
EMBO Journal 15, 3028-3029.
Shapira, M., Hamlin, B. J., Rong, J. M., Chen, K., Ronen, M., and Tan, M.-W. (2006). A
conserved role for a GATA transcription factor in regulating epithelial innate
immune responses. Proceedings of the National Academy of Sciences 103, 1408614091.
Sharma, N. K. Das, S. K., Mondal, A. K., Hackney, 0. G. Chu, W. S. et al. (2008).
Endoplasmic reticulum stress markers are associated with obesity in nondiabetic
subjects. The Journal of Clinical Endocrinology and Metabolism 93, 4532-4541.
Shen, J., Chen, X., Hendershot, L., and Prywes, R. (2002). ER stress regulation of ATF6
localization by dissociation of BiP/GRP78 binding and unmasking of Golgi
localization signals. Developmental Cell 3, 99-111.
Shen, X., Ellis, R. E., Lee, K., Liu, C.-Y., Yang, K., et al. (2001). Complementary
signaling pathways regulate the unfolded protein response and are required for C.
elegans development. Cell 107, 893-903.
Shen, X., Ellis, R. E., Sakaki, K., and Kaufman, R. J. (2005). Genetic interactions due to
constitutive and inducible gene regulation mediated by the unfolded protein
response in C. elegans. PLoS Genetics 1, e37. doi:10.1371/journal.pgen.0010037.
Shim, J., Umemura, T., Nothstein, E., and Rongo, C. (2004). The unfolded protein
response regulates glutamate receptor export from the endoplasmic reticulum.
Molecular Biology of the Cell 15, 4818-4828.
Sidrauski, C., Cox, J. S., and Walter, P. (1996). tRNA ligase is required for regulated
mRNA splicing in the unfolded protein response. Cell 87, 405-413.
Silverstein, R. A., and Ekwall, K. (2005). Sin3: a flexible regulator of global gene
expression and genome stability. Current Genetics 47, 1-17.
Skalet, A. H., Isler, J. A., King, L. B., Harding, H. P., Ron, D., and Monroe, J. G. (2005).
Rapid B cell receptor-induced unfolded protein response in nonsecretory B cells
correlates with pro- versus antiapoptotic cell fate. The Journal of Biological
Chemistry 280, 39762-39771.
Soto, C., and Estrada, L. D. (2008). Protein misfolding and neurodegeneration. Archives
of Neurology 65, 184-189.
Spear, E. D., and Ng, D. T. W. (2003). Stress tolerance of misfolded carboxypeptidase Y
requires maintenance of protein trafficking and degradative pathways. Molecular
Biology of the Cell 14, 2756-2767.
Tcherpakov, M., Broday, L., Delaunay, A., Kadoya, T., Khurana, A., ErdjumentBromage, H., Tempst, P., Qiu, X.-B., DeMartino, G. N., Ronai, Z. (2008). JAMP
optimizes ERAD to protect cells from unfolded proteins. Molecular Biology of the
Cell 19, 5019-5028.
Tenoever, B. R., Ng, S.-L., Chua, M. A., McWhirter, S. M., Garcia-Sastre, A., and
Maniatis, T. (2007). Multiple functions of the IKK-related kinase IKKepsilon in
interferon-mediated antiviral immunity. Science 315, 1274-1278.
Thomson, S. P., and Williams, D. B. (2005). Delineation of the lectin site of the
molecular chaperone calreticulin. Cell Stress & Chaperones 10, 242-251.
Tirasophon, W, Welihinda, A. A., and Kaufman, R. J. (1998). A stress response pathway
from the endoplasmic reticulum to the nucleus requires a novel bifunctional protein
kinase/endoribonuclease (Irelp) in mammalian cells. Genes & Development 12,
1812-1824.
Tong, A., Lynn, G., Ngo, V., Wong, D., Moseley, S. L., et al. (2009). Negative regulation
of Caenorhabditiselegans epidermal damage responses by death-associated protein
kinase. Proceedings of the National Academy of Sciences of the United States of
America 106, 1457-1461.
Travers, K. J., Patil, C. K., Wodicka, L., Lockhart, D. J., Weissman, J. S., and Walter, P.
(2000). Functional and genomic analyses reveal an essential coordination between
the unfolded protein response and ER-associated degradation. Cell 101, 249-25 8.
Troemel, E. R., Chu, S. W., Reinke, V., Lee, S. S., Ausubel, F. M., and Kim, D. H.
(2006). p3 8 MAPK regulates expression of immune response genes and contributes
to longevity in C. elegans. PLoS Genetics 2, 1725-1739.
Uccelletti, D., Pascoli, A., Farina, F., Alberti, A., Mancini, P., Hirschberg, C. B.,
Palleschi, C. (2008). APY- 1, a novel Caenorhabditiselegans apyrase involved in
unfolded protein response signaling and stress responses. Molecular Biology of the
Cell 19, 1337-1345.
Urano, F., Calfon, M., Yoneda, T., Yun, C., Kiraly, M., Clark, S. G., Ron, D. (2002). A
survival pathway for Caenorhabditiselegans with a blocked unfolded protein
response. Journal of Cell Biology 158, 639-646.
Urano, F., Wang, X., Bertolotti, A., Zhang, Y., Chung, P., Harding, H. P., and Ron, D.
(2000). Coupling of stress in the ER to activation of JNK protein kinases by
transmembrane protein kinase IRE 1. Science 287, 664-666.
Vattem, K. M., and Wek, R. C. (2004). Reinitiation involving upstream ORFs regulates
ATF4 mRNA translation in mammalian cells. Proceedings of the National Academy
of Sciences of the United States of America 101, 11269-11274.
Wang, X. Z., Harding, H. P., Zhang, Y., Jolicoeur, E. M., Kuroda, M., and Ron, D.
(1998). Cloning of mammalian Irel reveals diversity in the ER stress responses. The
EMBO Journal 17, 5708-5717.
Wang, Y, Shen, J, Arenzana, N., Tirasophon, W, Kaufman, R. J., and Prywes, R. (2000).
Activation of ATF6 and an ATF6 DNA binding site by the endoplasmic reticulum
stress response. The Journal of Biological Chemistry 275, 27013-27020.
Winnay, J. N., Boucher, J., Mori, M. A., Ueki, K., and Kahn, C. R. (2010). A regulatory
subunit of phosphoinositide 3-kinase increases the nuclear accumulation of X-boxbinding protein-1 to modulate the unfolded protein response. Nature Medicine 16,
438-445.
Woo, C. W., Cui, D., Arellano, J., Dorweiler, B., Harding, H. P., Fitzgerald, K. A., Ron,
D., and Tabas, I. (2009). Adaptive suppression of the ATF4-CHOP branch of the
unfolded protein response by toll-like receptor signalling. Nature Cell Biology 11,
1473-1480.
Wu, J., Rutkowski, D. T., Dubois, M., Swathirajan, J., Saunders, T., Wang, J., Song, B.,
Yau, G. D.-Y., and Kaufman, R. J. (2007). ATF6alpha optimizes long-term
endoplasmic reticulum function to protect cells from chronic stress. Developmental
Cell 13, 351-364.
Yamamoto, K., Sato, T., Matsui, T., Sato, M., Okada, T., Yoshida, H., Harada, A., and
Mori, K. (2007). Transcriptional induction of mammalian ER quality control
proteins is mediated by single or combined action of ATF6alpha and XBP .
Developmental Cell 13, 365-376.
Yamamoto, K., Takahara, K., Oyadomari, S., Okada, T., Sato, T., Harada, A., and Mori,
K. (2010). Induction of liver steatosis and lipid droplet formation in ATF6alphaknockout mice burdened with pharmacological endoplasmic reticulum stress.
Molecular Biology of the Cell 21, 2975-2986.
Yamazaki, H., Nobuhiko, H., Hayakawa, K., Tagawa, Y., Maro, 0., et al. (2009).
Activation of the Akt-NF-kappaB pathway by subtilase cytotoxin through the ATF6
branch of the unfolded protein response. Journal of Immunology 183, 1480-1487.
Yanagitani, K., Imagawa, Y., Iwawaki, T., Hosoda, A., Saito, M., Kimata, Y., and
Kohno, K. (2009). Cotranslational targeting of XBP1 protein to the membrane
promotes cytoplasmic splicing of its own mRNA. Molecular Cell 34, 191-200.
Ye, Y., Shibata, Y., Yun, C., Ron, D., Rapaport, T. A. (2004). A membrane protein
complex mediates retro-translocation from the ER lumen into the cytosol. Nature
429, 841-847.
Yoshida, H., Haze, K., Yanagi, H., Yura, T., and Mori, K. (1998). Identification of the
cis-Acting Endoplasmic Reticulum Stress Response Element Responsible for
Transcriptional Induction of Mammalian Glucose-regulated Proteins: involvement
of basic leucine zipper transcription factors. Journal of Biological Chemistry 273,
33741-33749.
Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., and Mori, K. (2001). XBP1 mRNA is
induced by ATF6 and spliced by IRE 1 in response to ER stress to produce a highly
active transcription factor. Cell 107, 881-891.
Yoshida, H., Oku, M., Suzuki, M., and Mori, K. (2006). pXBP1(U) encoded in XBP1
pre-mRNA negatively regulates unfolded protein response activator pXBP l(S) in
mammalian ER stress response. The Journal of Cell Biology 172, 565-575.
Yoshida, H., Matsui, T., Hosokawa, N., Kaufman, R. J., Nagata, K., and Mori, K. (2003).
A time-dependent phase shift in the mammalian unfolded protein response.
Developmental Cell 4, 265-271.
Zarubin, T., and Han, J. (2005). Activation and signaling of the p38 MAP kinase
pathway. Cell Research 15, 11-18.
Zhang, K., Wong, H. N., Song, Benbo, Miller, C. N., Scheuner, D., and Kaufman, R. J.
(2005). The unfolded protein response sensor IRE lalpha is required at 2 distinct
steps in B cell lymphopoiesis. The Journal of Clinical Investigation 115, 268-28 1.
Zhang, K., and Kaufman, R. J. (2008). From endoplasmic-reticulum stress to the
inflammatory response. Nature 454, 455-462.
Zhang, P., McGrath, B., Li, S., Frank, A., Zambito, F., et al. (2002). The PERK
eukaryotic initiation factor 2 kinase is required for the development of the skeletal
system, postnatal growth, and the function and viability of the pancreas. Molecular
and Cellular Biology 22, 3864-3874.
Zhang, W., Feng, D., Li, Y., Lida, K., McGrath, B., and Cavener, D. R. (2006). PERK
EIF2AK3 control of pancreatic beta cell differentiation and proliferation is required
for postnatal glucose homeostasis. Cell Metabolism 4, 491-497.
Zhang, X., Zhang, G., Zhang, H., Karin, M., Bai, H., and Cai, D. (2008). Hypothalamic
IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance and
obesity. Cell 135, 61-73.
Chapter Two
An essential role for XBP-1 in host protection against immune
activation in Caenorhabditis elegans
Claire E. Richardson, Tristan Kooistra, Dennis H. Kim
T. K. performed early experiments probing the role of the UPR in adult immunity and
assisted with the generation of the data provided in response to reviewers. Data in all
figures was generated by C.E.R.
Originally published in Nature 463, 1092-1095, February 25, 2010.
70
SUMMARY
The detection and compensatory response to the accumulation of unfolded
proteins in the endoplasmic reticulum (ER), termed the Unfolded Protein Response
(UPR), represents a conserved cellular homeostatic mechanism with important roles in
normal development and in the pathogenesis of disease (Schr6der et al. 2005). The IRE 1XBPl/Haclp pathway is a major branch of the UPR that has been conserved from yeast
to human (Cox et al. 1996; Calfon et al. 2002; Mori et al. 1996; Shen et al. 2001; Yoshida
et al. 2001). XBP-1 is required for the differentiation of the highly secretory plasma cells
of the mammalian adaptive immune system (Iwakoshi et al. 2003; Reimold et al. 2001),
but recent work also points to reciprocal interactions between the UPR and other aspects
of immunity and inflammation (Kaser et al. 2008; Todd et al. 2008; Zhang et al. 2006).
We have been studying innate immunity in the nematode Caenorhabditiselegans, having
established a key role for a conserved PMK- I p38 mitogen-activated protein kinase
(MAPK) pathway in mediating resistance to microbial pathogens (Kim et al. 2002). Here,
we show that during C. elegans development, XBP- 1 has an essential role in protecting
the host during activation of innate immunity. Activation of the PMK- 1-mediated
response to infection with Pseudomonas aeruginosainduces the XBP- 1-dependent UPR.
Whereas a loss-of-function xbp-] mutant develops normally in the presence of relatively
non-pathogenic bacteria, infection of the xbp-] mutant with P. aeruginosaleads to
disruption of ER morphology and larval lethality. Unexpectedly, the larval lethality
phenotype on pathogenic P. aeruginosais suppressed by loss of PMK-1-mediated
immunity. Furthermore, hyperactivation of PMK- 1 causes larval lethality in the xbp- 1
mutant even in the absence of pathogenic bacteria. Our data establish innate immunity as
a physiologically relevant inducer of ER stress during C. elegans development and
suggest that an ancient, conserved role for XBP- 1 may be to protect the host organism
from the detrimental effects of mounting an innate immune response to microbes.
RESULTS AND DISCUSSION
We began our analysis of the role of the UPR in C. elegans immunity by asking if
the UPR is activated upon exposure of C. elegans larvae to P. aeruginosastrain PA 14, a
human opportunistic pathogen that can also infect and kill C. elegans (Tan et al. 1999).
First, we assessed expression of the transgenic transcriptional reporter Phsp4:.:GFP(zcls4)(the C. elegans hsp-4 gene encodes a homolog of mammalian
BiP/GRP78), which reflects activation of the IRE- 1-XBP- 1 branch of the UPR (Calfon et
al. 2002). After exposure of C. elegans larvae to P. aeruginosa,GFP expression was
induced in the intestine, the site of infection (Figure 1a). We confirmed the pathogeninduced activation of the endogenous hsp-4 gene by quantitative RT-PCR (qRT-PCR)
(Figure Ib).
The mechanism by which IRE-I activates XBP- 1 in response to ER stress is
conserved from yeast to mammals (Calfon et al. 2002; Cox et al. 1996; Shen et al. 2001;
Yoshida et al. 2001). Upon accumulation of unfolded proteins in the ER, the integral ERmembrane protein IRE-I activates XBP-1 by alternative splicing of the xbp-I mRNA,
thereby changing the reading frame. The activated "spliced form" of XBP- 1 regulates
expression of genes involved in ER homeostasis, such as those encoding chaperones.
Using qRT-PCR to quantify the IRE-1-mediated splicing of xbp-1 mRNA after exposure
to P. aeruginosa,we found there was a marked induction of IRE-I-spliced xbp-1
transcript within 4 h of exposure to pathogen (Figure Ic).
Figure 1. PMK-1 p38 MAPK-dependent activation of the IRE-1-XBP-1-dependent
UPR by P. aeruginosa infection. (a) Fluorescence and DIC microscopy of WT and
mutant larvae carrying the Phsp-4::GFP(zcs4)transgene 24 h after egg lay on indicated
treatments (Scale bar = 100ptm). (b) qRT-PCR of endogenous hsp-4 mRNA levels. (c)
qRT-PCR analysis of total and spliced xbp-1 mRNA levels. Values represent fold change
relative to WT on OP50 ± s.e.m. (n = 4 independent experiments, *P < 0.05 and ***P <
0.001, two-way ANOVA with Bonferroni post test).
Relative xbp-1 mRNA levels
3z
z
0-
ON'
Relative hsp-4 mRNA levels
tunicamycin
R aeruginosa
E coli
"H
(O
T!
We hypothesized that the rapid activation of the IRE-1 -XBP- 1 pathway after P.
aeruginosaexposure might in fact be caused by induction of the host immune response.
A conserved p38 MAPK pathway mediates innate immunity in organisms ranging from
C. elegans to humans (Dong et al. 2002; Kim et al. 2002). We have previously shown
that loss-of-function of the p38 MAPK ortholog PMK-1 results in markedly enhanced
susceptibility to killing of C. elegans by P. aeruginosa.In the pmk-] (km25) mutant, we
found no increase in IRE-I-spliced XBP-1 mRNA and no induction of the Phsp-4.::GFP
reporter in response to P. aeruginosa(Figure 1). RNAi of sek-1, encoding a MAPKK
required for PMK-1 activation (Kim et al. 2002), also abrogated induction of Phsp4.:GFP expression in response to P. aeruginosainfection (Supplementary Figure la).
Consistent with the induction of XBP- 1 splicing arising as a consequence of a
transcriptional response to infection, we found that a mutant, atf-7(qd22 qd130), carrying
a mutation in a conserved transcription factor that abrogates the pathogen-induced
expression of genes regulated by the PMK- I pathway (Shivers et al. 2010), also blocked
induction of the Phsp-4::GFPreporter in response to P. aeruginosaPA14
(Supplementary Figure Ib). Activation of the IRE-1-XBP-1 pathway in response to
tunicamycin (an N-glycosylation inhibitor), however, occurred in the pmk-] mutant as in
the wild-type strain N2 (WT), as reported by Bischof et al. in their study of PMK-1dependent activation of the IRE-I -XBP- 1 pathway in the C. elegans response to a poreforming toxin from Bacillus thurigiensis(Shen et al. 2005). Whereas our data show that
activation of the IRE-i -XBP- 1 pathway by P. aeruginosainfection is dependent on the
PMK- 1 pathway, we did not observe a reciprocal dependence of activation of the PMK- 1
pathway on XBP-1 (Supplementary Figure 2).
We next asked if XBP- 1 is required for resistance to P. aeruginosa,focusing on
its role during larval development. Synchronized eggs of WT and mutants deficient in
each of the three branches of the UPR were propagated on plates with P. aeruginosa
PA14 as the only food source and development was monitored over time. Nearly all the
eggs of the WT strain developed to at least the fourth larval stage (L4) in this assay,
comparable to their growth and development on the relatively non-pathogenic bacterial
strain and standard laboratory food source Escherichiacoli OP50 (Figure 2a). In contrast,
the xbp-1 (zc 12) mutant on P. aeruginosaexhibited severely attenuated larval
development and growth, as measured by the rate of progression between molts, size of
larvae, and ability of larvae to reach the L4 stage by 72 h (Figure 2a), eventually resulting
in larval lethality. Loss-of-function mutations in the each of the other two major branches
of the UPR, atf-6(ok551) and pek-1(ok275) (Shen et al. 2001; Rahme et al. 1995), did not
affect larval development in the presence of P. aeruginosa(Figure 2a). These data
implicate a specific and essential role for the IRE-1-XBP-1 branch of the UPR in C.
elegans development in the presence of P. aeruginosa.
Figure 2. XBP-1 is required for C. elegans development and survival on P.
aeruginosa. (a) Development of indicated mutants to the L4 larval stage or older after 3 d
at 250 C on either P. aeruginosaPA14 or E. coli OP50, or (b) mixture of PA14 and E.
coli HB1O1. Plotted is mean ± s.d. (n = 3-4 plates, ***P<0.001, two-way ANOVA with
Bonferroni post test in (a), and n = 4 plates, *** p<0.001 two tailed t-test in (b)). Results
in (a) and (b) are representative of 3 independent experiments. (c) Transmission electron
microscopy of L3 larvae at 60,000x magnification (Scale bar = 500 nm). (d)
Fluorescence and DIC microscopy overlay of larvae 40 h after egg-lay on P. aeruginosa
PA14 expressing GFP.
M E.coil
ED P. aeruginosa
1.0
-
7
0.5-
Figure 2
E. coli +
P. aeruginosa
+ 1.0
-0
U-
0.0
WT
pek-1 atf-6 xbp-1
WT
S
*
0.0-T -+
WT xbp-1
xbo-1
Uk-1(km
xbp-1 (zc 12)
The xbp-1 developmental phenotype was alleviated in the presence of the gacA
mutant of P. aeruginosaPA14 (Supplementary Figure 3a), which has diminished
pathogenicity in C. elegans and mammals (Tan et al. 1999; Y. Zhang et al. 2005). On a
lawn of wild-type P. aeruginosaPA14 mixed with relatively non-pathogenic E. coli, the
xbp-1 mutant again exhibited the severely attenuated developmental phenotype that was
observed in the presence of P. aeruginosaalone, suggestive that the pathogenicity of P.
aeruginosa,not a nutritional deficiency, caused the xbp-1 developmental phenotype
(Figure 2b and Supplementary Figure 3b). Behavioral avoidance of pathogenic bacteria
(Reddy et al. 2009; Pujol et al. 2001) can influence survival (Shivers et al. 2009;
Rutkowski et al. 2006; Styer et al. 2008), but we found that the development of WT was
unaffected by experimental conditions in which the P. aeruginosacould not be avoided,
demonstrating that the inability of the xbp-I mutant to develop on P. aeruginosawas not
due to a defect in behavioral avoidance (data not shown). In addition, the attenuated
development and death of the xbp-1 mutant on P. aeruginosawas also unaffected by the
blocking of cell necrosis (Supplementary Table 1).
WT and xbp-1 mutant larvae exhibit characteristic ribosome-studded tubular
structures of the ER by transmission electron microscopy when propagated on relatively
non-pathogenic E. coli OP50 (Figure 2c). The ER architecture of WT worms was not
markedly perturbed upon exposure to P. aeruginosa(Figure 2c). In contrast, xbp-1(zc12)
larvae propagated on P. aeruginosaPA14 revealed disruption in ER morphology of xbp1 mutant worms upon pathogen exposure (Figure 2c), with localized regions of dilated
ER lumen, comparable to that induced by tunicamycin treatment of xbp-](zc12) worms
(Supplementary Figure 4). Such changes in ER morphology are consistent with ER stress
and perturbation (Kim et al. 2004), and strongly suggest that a disruption in ER
homeostasis is induced by PA14 infection in the xbp-1 mutant.
We considered whether the developmental lethality of the xbp-1 mutant might be
due to the toxicity of PA14 arising from accelerated infection due to diminished
resistance to P. aeruginosain the xbp-] mutant. Because activation of the PMK- 1
pathway results in the splicing of XBP- 1 upon P. aeruginosaexposure, we first asked
whether XBP- 1 confers a survival benefit against P. aeruginosaby facilitating resistance
to intestinal infection. Diminished activation of PMK-1-dependent innate immunity is
accompanied by an increased rate of P. aeruginosaaccumulation in the intestine (Kim et
al. 2002)(Figure 2d). If XBP- 1 functioned downstream of PMK- 1 to promote the immune
response to P. aeruginosa,we would anticipate that xbp-] deficiency would lead to an
increased rate of P. aeruginosaaccumulation relative to WT. Interestingly, in spite of the
severely attenuated development and survival of the xbp-1 mutant larvae on P.
aeruginosa,the kinetics of accumulation of a PA14-derived strain of P. aeruginosa
expressing GFP in the xbp-1(zc12) mutant were comparable to that observed for WT
(Figure 2d). Although the pmk-] (km25) mutant exhibited an increased rate of
accumulation of P. aeruginosa(Figure 2d), larval development and survival was
markedly greater than that observed for the xbp-] mutant (Figures 3a and 3b, and
Supplementary Figure 5). These data decouple enhanced susceptibility to P. aeruginosa
infection from the attenuated larval development phenotype of the xbp-1 mutant and
suggest that while activation of XBP-l by P. aeruginosainfection is PMK- 1-dependent,
the primary protective role of XBP- 1 is not to facilitate the killing of P. aeruginosa.
Figure 3. Suppression of the P. aeruginosa-induced larval lethality phenotype of
xbp-1(zc12) by pmk-1(km25). (a) DIC microscopy of representative worms of each
genotype 3 d after egg-lay on P. aeruginosaPA14 (Scale bar = 100 [tm). (b)
Development of indicated mutants to the L4 larval stage or older after 3 d at 250 C on
either P. aeruginosaPA14 or E. coli OP50. Plotted is the mean
±
s.d. (n = 3-4 plates,
***P < 0.001, two-way ANOVA with Bonferroni post test). Representative results of 3
independent experiments.
Figure 3
xbp-1(zc12)
xbp -1(zc12);
pmk- 1(km25)
pmk-1(km25)
M E.coli
M P. aeruginosa
WT
pmk-1 xbp-1 xbp-1;
pmk-1
In view of these data, we considered whether the XBP- 1-dependent UPR
functions in this setting to protect against the ER stress induced by the immune response
itself. If this were the case, then diminishing the immune response might actually
improve the survival of the xbp-I mutant on P. aeruginosa.To test this hypothesis, we
examined the larval development of a xbp-1(zc12);pmk-1(km25) double mutant.
Strikingly, the xbp-1;pmk-1 double mutant showed markedly increased development and
survival relative to the xbp-I mutant, to a degree that was comparable to the development
and survival of the pmk-] mutant (Figures 3a and 3b). RNAi of sek-1, and RNAi of atf-7
in the xbp-](zc]2) mutant each alleviated the attenuated development phenotype in the
presence of P. aeruginosa(Supplementary Figures 6a and b). These data indicate that
activation of the PMK- 1-dependent transcriptional innate immune response to pathogen
is indeed detrimental to survival during development in the absence of the IRE-1 -XBP- 1
pathway, and that the principal mechanism by which XBP- 1 promotes development and
survival during infection with P. aeruginosais by protecting against the innate immune
response.
To separate further the effect of toxicity caused by pathogenic P. aeruginosafrom
that caused specifically by immune activation on the xbp-1 developmental phenotype, we
next asked if the xbp-I mutant would be compromised in the setting of hyperactivation of
the PMK- 1-mediated immune response in the absence of P. aeruginosa.RNAi against
vhp-1, which encodes a MAPK phosphatase that negatively regulates PMK-1, results in
hyperactivation of PMK- 1 (Troemel et al. 2006). We observed that RNAi of vhp-1
resulted in sustained induction of the Phsp-4::GFP(zcIs4)reporter with the relatively
non-pathogenic, standard laboratory food source E. coli OP50 as the only bacteria present
in the assay (Figure 4a). RNAi-mediated knockdown of vhp-1 had little effect on
development of WT (Figure 4b). In contrast, RNAi of vhp-] severely impaired
development of the xbp-](zc12) mutant in the absence of pathogenic bacteria (Figure 4b)
in a manner similar to that observed for the xbp-1 mutant when propagated on P.
aeruginosa(Figure 2a). Both the induction of hsp-4 and the larval developmental
phenotype of the xbp-1 mutant resulting from RNAi of vhp-1 were suppressed by pmk-]
(Figures 4a and 4b), demonstrating that the observed effects of RNAi of vhp-1 were
mediated by the PMK-1-dependent immune response.
Figure 4. XBP-1 is required for development and survival during pathogenindependent constitutive activation of PMK-1. (a) Fluorescence and DIC microscopy
of WT and xbp- 1 larvae carrying the Phsp-4:.:GFP(zcIs4)transgene, which have been
treated with control (empty vector) or vhp-] RNAi, 24 h after egg lay. (b) Development
to the L4 larval stage or older after 3 d at 25'C of WT, xbp-1, and xbp-1,pmk-1 mutants
that have been subjected to RNAi of vhp-1 and propagated on E. coli OP50.
Plotted is mean t s.e.m (n = 3 independent experiments, ***P < 0.001, two-way ANOVA
with Bonferroni post test).
Figure 4
a
Control RNAi
vhp-1 RNAi
WT
pmk-1
b
RNAi
M
control
M vhp-1
WT
xbp-1 xbp-1;
pmk-1
Our data suggest that the XBP- 1-mediated UPR plays an essential role during C.
elegans larval development by protecting against the activation of innate immunity.
Transcriptional profiling of the C. elegans response to infection with P. aeruginosa
(Shapira et al. 2006; Kaufman et al. 2002) has revealed the induction of at least 300 genes
(Shapira et al. 2006), nearly half of which are predicted to be trafficked through the ER.
Although this innate immune response promotes pathogen resistance, retarding the
intestinal accumulation of P. aeruginosaand substantially increasing the fraction of the
population that survives through larval development (as illustrated by the survival of WT
relative to the pmk-1 (km25) mutant (Figure 3 and Supplementary Figure 5)), our data
suggest that innate immunity also constitutes a physiologically relevant source of ER
stress that necessitates the compensatory activity of the UPR in C. elegans larval growth
and development. An ancient role for the UPR may be to facilitate the development and
survival of metazoan cell types that are exposed to the environment and mount a
defensive, secretory response to microbial pathogens and abiotic toxins (Shen et al. 2001;
Kaufman et al. 2002). Our data not only support this hypothesis, but more specifically
suggest that in fact the essential role of XBP-1 lies not in the neutralization of these
environmental insults, but instead in conferring protection against the potentially lethal
ER stress that is induced by this response. Our findings support the idea that the
microbiota of multicellular organisms may represent the most commonly encountered
and physiologically relevant inducer of such ER stress. Consistent with this hypothesis is
the Paneth cell death observed in XBP1 intestinal knockout mice (Zhang et al. 2006).
Evolutionary conservation of the function of XBP- 1 that we have defined in C. elegans
would implicate the UPR in protection against ER toxicity caused by activation of
immune and inflammatory pathways in mammals, where the UPR may have a critical
role in the pathogenesis of infectious and inflammatory autoimmune diseases.
MATERIALS AND METHODS
C. elegans strains
C. elegans strains were maintained as described (Brenner 1974). Strains used in this study
were N2 Bristol WT strain, SJ4005 Phsp-4::.GFP(zcIs4),ZD305 pmk-1(km25);Phsp4..GFP(zcIs4), ZD441 atf-7(qd22 qd130); Phsp-4.:GFP(zcIs4),ZD361 xbp-](zc12),
RB545 pek-1(ok275), RB772 atf-6(ok551), ZD363 xbp-1(zc12);pmk-1(km25), ZD416
xbp-](tm2457), ZD417 xbp-1(tm2457);pmk-1(km25), ZD418 xbp-1(tm2482) and ZD419
xbp-1(tm2482),pmk-1(km25). To make ZD361 and ZD363, SJ17 xbp-1(zc12) was
outcrossed twice, removing Phsp-4:.:GFP(zcls4), then crossed to ZD39
pT24B8.5:.:GFP(agls9)III;pmk-1(km25)to isolate a xbp-1(zc12);pmk-1(km25) double
mutant. This xbp-1(zc12),pmk-1(km25) double mutant was backcrossed to ZD39 to
isolate 4X back-crossed xbp-](zc12);pmk-1(km25) and xbp-] (zc12). To make ZD416,
ZD417, ZD418 and ZD419, strains carrying the xbp-1(tm2457) and xbp-1(tm2482) alleles
were outcrossed four times, then crossed to ZD39 pT24B8.5:.GFP(agIs9)III;pmk1(km25), from which single xbp-1 and double xbp-1;pmk-1 strains were isolated.
Preparation of P. aeruginosa treatment plates
P. aeruginosa(strain PA14) was grown in Luria Broth (LB). All PA14 plates were
prepared as described (Tan et al. 1999), with the following changes: for RNA preparation
and electron microscopy, 25 t overnight culture was seeded onto 10 cm plates. To make
plates with lawns of mixed bacteria, E. coli HB 101 containing an ampicillin-resistant
plasmid encoding GFP was grown in LB with 50 [tg/ml ampicillin in parallel to P.
aeruginosa,and the two cultures were mixed in equal parts before seeding 10 [il of the
mix onto plates containing 25 tg/ml carbenicillin. The presence of E. coli in the lawns at
the end of assays was confirmed by the expression of GFP. To visualize intestinal
accumulation of P. aeruginosa,P. aeruginosaPA14 containing an ampicillin-resistant
plasmid encoding GFP was grown in LB with 50 [tg/ml ampicillin, and 10 tl of the
culture was seeded onto plates containing 25 jig/ml carbenicillin and spread to cover to
surface of the plate (to prevent the growth of patches of the lawn from which the GFP
plasmid had been lost). For tunicamycin treatments, tunicamycin was added to E. coli
plates at 5 jig/ml final concentration.
Microscopy
Light microscopy images were acquired with an Axioimager Z 1 microscope using worms
anesthetized in 0.1% sodium azide. For transmission electron microscopy, synchronized
LIs were grown on E. coli (OP50) or P. aeruginosa(PA 14) at 250 C to the L3 stage
(Figure 2c), or as described in figure legend (Supplementary Figure 4), then fixed as
described (Troemel et al. 2008). The WT and xbp-1(zc12) strains were grown in parallel
on P. aeruginosaand fixed twice in independent experiments with similar results.
RNA Extraction, cDNA Preparation, and Quantitative RT-PCR
For RNA analysis, LI larvae were synchronized by hypochlorite treatment, washed onto
OP50 plates and grown at 20'C for 23 h on OP50, then washed in M9 onto treatment
plates. After 4 h, worms were washed off plates and frozen in liquid nitrogen. Total RNA
was isolated using Tri Reagent (Ambion), DNA contamination was removed with DNAfree treatment (Ambion). cDNA was synthesized with RETROscript kit (Ambion) using
oligo (dT) primers. Quantitative real-time PCR reactions were performed in triplicate
using FastStart SYBR green master mix (Roche) with a Realplex mastercycler
(Eppendorf). Relative mRNA was determined with the AACt method using expression of
snb-1 (Figure lb) or the mean expression of snb-1, act-], and tba-1 as a normalization
control (Figure Ic, Supplementary Figure 2) (Troemel et al. 2006; Pocock et al. 2008;
Hoogewijs et al. 2008). Biological replicates were compared as described (Willems et al.
2008). Primer efficiencies were tested with dilution series, and primer sequences are
provided in Supplementary Table 2.
Development assays
For development assays, E. coli OP50 plates were prepared in parallel with PA14 plates.
Strains were egg laid in parallel onto 4 plates each of PA14 and OP50 (at least 80 eggs
for each strain and treatment), and fraction of worms growing to at least the L4 larval
stage ("L4+") between the plates was averaged. The L4 stage of development was chosen
due to ease of scoring vulval development. RNAi by feeding of bacteria (Timmons et al.
1998) was performed as described by propagating stains on E. coli HT 115 carrying the
RNAi clones as indicated at 20'C, and adult progeny were used to lay eggs in
experiments, which were performed at 25C. The vhp-1 RNAi clone and clones of genes
required for necrosis were from the Ahringer library, and the sek-1 RNAi clone was from
the Vidal library (Kamath et al. 2003; Rual et al. 2004). A segment of the atf-7 coding
region, 771 bp region of atf-7 corresponding to bases 11,532 to 12,303 with respect to
cosmid C07G2, was amplified by PCR and subeloned into the Fire vector L4440 (Shivers
et al. 2010). The sequences of all clones were confirmed prior to use.
Statistics
Statistical analyses were performed with GraphPad Prism (GraphPad Software, Inc.).
SUPPLEMENTARY TABLES
Supplementary Table 1. RNAi of genes required for necrosis does not rescue
development of the xbp-1 mutant on P aeruginosa PA14
RNAi clone
gene
Necrosis
Gene description
phenotype
(ref)
Phenotype in development assay
WT
xbp-1(zc12)
E. coli P aeruginosa E. coli P aeruginosa
all L4+
all L4+
all L4+
0 L4+
L4440
empty
F20B6.2
vha-12
1
V-ATPase
0 L4+
0 L4+
0 L4+
0 L4+
ZK637.8
unc-32
1
V-ATPase
0 L4+
0 L4+
0 L4+
0 L4+
T03E6.7
cpl-1
2, 3
lysosomal cysteine
protease
0 L4+
0 L4+
0 L4+
0 L4+
C25B8.3
cpr-6
3
NC
NC
NC
NC
Y39B6A.20
asp-]
2, 3
lysosomal cysteine
protease
aspartic peptidase
NC
NC
NC
NC
H22K11.1
asp-3
2,3
aspartic peptidase
NC
NC
NC
NC
R12H7.2
asp-4
2
aspartic peptidase
NC
NC
NC
NC
LLC 1.1
tra-3/clp-5
2, 3
calpain
NC
NC
NC
NC
WO5Gl1.4
3
calpain
NC
NC
NC
NC
NC - No Change from L4440 control. Obersed abnormal developmental phenotypes were consistant with
previous reports for vha-12, unc-32, and cpl-1.
1. Wong et al. 2007
2. Syntichaki et al. 2002
3. Luke et al., 2007
Supplementary Table 2. Primer sequences used for quantitative RT-PCR
Gene
Sequence name
xbp-] total
R 74.3
xbp-1 spliced
R 74.3a
F43E2.8
hsp-4
K08D8.5
C1 7H]2.8
T24B8.5
C32H11. 12
F55G11.2
F49F].6
C32H]1. 1
T10H9.4
snb-]
act-]
T04C2.6
tba-1
F26E4.8
Forward primer
ccgatccacctccatcaac
tgcctttgaatcagcagtgg
agttgaaatcatcgccaacg
atgctaccgaatgcttttcc
tgtcatttcaatggaggatattgt
agaccatcatgccettcact
gtgtccaacacaacctgcat
ggttctccagacgtgttcact
ccatcaactacgccaaagc
aaatctatcacctggatcacga
ccggataagaccatcttgacg
cttgggtatggagtccgcc
gtacactccactgatctctgctgacaag
Reverse primer
accgtctgetccttcctcaatg
accgtctgctccttcctcaatg
gcccaatcagacgcttgg
tccttgggtgtagtttccaa
tgatggagttggaggatattga
gtaacgcagacaccacaggt
catgtttccattcacctgga
aatgaacgttagtgggagtgc
tccggtggatagaaggtgtt
ccttgatatttcgtccateg
gacgacttcatcaacctgagc
ttagaagcacttgcggtgaac
ctctgtacaagagacaaacagccatg
SUPPLEMENTARY FIGURES
Supplementary Figure 1. SEK-1 and ATF-7 are required for P. aeruginosa- induced
activation of the IRE-1-XBP-1 pathway. (a) L4 larvae of WT worms carrying the
Phsp-4:.:GFP(zcIs4)transgene were propagated on E. coli strain HT 115 carrying the
empty L4440 vector (control RNAi) or a sek-1 RNAi vector as indicated, and adult
progeny were used to lay eggs onto treatment plates. (b) WT and atf-7(qd22 qd]30)
strains carrying the Phsp-4.::GFP(zcls4)transgene were used to lay eggs onto treatment
plates. For both (a and b), treatment plates were P. aeruginosa(strain PA14), E. coli
(strain OP50), or tunicamycin (5 Rg/ml final concentration added to OP50 plates).
Fluorescence and DIC microscopy images were acquired 24 h after eggs were laid (Scale
bar = 100 pim).
.. ....
..
....
..
...
tunicamycin
R aeruginosa
...
......
.....
......
.........
E coli
tunicamycin
R aeruginosa
E coli
C
(O
"i
*0
C)
Supplementary Figure 2. Activation of the PMK-1-dependent transcriptional
response is not dependent on XBP-1. Quantitative RT-PCR of eight genes known to be
induced by exposure to P. aeruginosa(Troemel et al., 2006) in populations of WT, pmk1(km25), and xbp-1(zc]2) L3 larvae after 4 h exposure to P. aeruginosa.Values represent
fold change relative to WT on OP50
s.d. Expression was normalized to that of the
average of snb-1, act-] and tba-1 (n
3 reactions).
Supplementary Figure 2
1000
100 1
MWT
Mpmk-1
Mxbp-1
10
1
,C'
4
0.1
CV
4
0.01
0.001
0-
CV
4
U
Supplementary Figure 3. Requirement for xbp-1 in C. elegans development on P.
aeruginosa is not due to an xbp-1-specific nutritional deficiency. At least 130 eggs of
WT or xbp-1(zc12) were laid onto plates seeded with (a) P. aeruginosaPA14 mixed
equally with E. coli (HB 101 -GFP), in a replicate experiment of Figure 2b, or (b) P.
aeruginosaPA 14 gacA. Development to the L4 stage was monitored for 3 d. Shown in
(b) are representative results of 2 independent experiments. Plotted is the mean ± s.d. (n
= 4 plates, ***P < 0.001, two-tailed t-test).
Supplementary Figure 3
E. coli +
z P. aeruginosa
ED
M P.aeruginosa gacA
+ 1.0-
0.5-I/
0.0-
WT xbp-1
WT xbp-1
Supplementary Figure 4. Transmission electron microscopy of WT and xbp-1 larva
treated with tunicamycin. Synchronized WT or xbp-1(zc12) Lls were grown on E. coli
(OP50) for 15 h at 20'C, then transferred to plates with tunicamycin (5 [tg/ml final
concentration added to OP50 plates) for 10 h at 25'C. Images show representative larvae
at 60,000x magnification (Scale bar = 500 nm).
Supplementary Figure 4
WT
xbp-1
Supplementary Figure 5. Suppression of the larval lethality phenotype on P.
aeruginosa of multiple xbp-1(f) alleles bypmk-1(km25). At least 120 eggs/strain were
laid in parallel onto plates seeded with P. aeruginosaPA14 or E. coli OP50, and
development to the L4 stage was monitored for 3 d. Plotted is the mean ± s.d. (n = 4
plates, *P < 0.05, two-way ANOVA with Bonferroni post test). (a) Shown is a replicate
of the experiment in Fig 3b. (b) Representative results of 2 independent experiments.
Supplementary Figure 5
a
b
M E. coli
M E.coli
[] P. aeruginosa
E] P. aeruginosa
1.0-
+1.0C:C
-0.5-
0
0
0.5-
LL
u..
0.0-
0.0
(V
AIV
CV
Supplementary Figure 6. Suppression of the larval lethality phenotype on P.
aeruginosa by sek-1 and atf-7 RNAi. L4 larvae of WT and xbp-1(zc12) strains were
propagated on E. coli strain HT 115 carrying the empty L4440 vector (control RNAi) or a
(a) sek-1 or (b) atf-7 RNAi vector as indicated, and adult progeny were used to lay eggs
onto P. aeruginosaPA14 plates. At least 80 eggs were laid in parallel for each
strain/treatment and development to the L4 larval stage was monitored for 3 d. Results
shown are representative of 2 independent experiments, plotted is mean ± s.d. (n = 3-4
plates, *P < 0.05, ***P < 0.001, two-way ANOVA with Bonferroni post test).
Supplementary Figure 6
a
b
M
M control RNAi
[]
+ 1.0-
+ 1.00
control RNAi
[] atf-7RNAi
sek-1 RNAi
Q50
0.5-
-
I
U-
0.0
0.5-
L
0.0
WT
xbp-1
WT
xbp-1
ACKNOWLEDGMENTS
We thank M. McKee for technical assistance with transmission electron microscopy
experiments, and E. Hartwieg and G. Voeltz for advice in the interpretation of electron
microscopy images. We thank T. Stiernagle and the Caenorhabditis Genetics Center
(supported by the NIH), and S. Mitani and the National Bioresource Project of Japan for
strains. T.K. was supported by summer research fellowships from the Howard Hughes
Medical Institute. This work was supported by NIH grant RO 1-GM084477, a Career
Award in the Biomedical Sciences from the Burroughs Wellcome Fund, and an Ellison
Medical Foundation New Scholar Award (to D.H.K.).
REFERENCES
Brenner, S. (1974). The genetics of Caenorhabditiselegans. Genetics 77, 71-94.
Calfon, M., Zeng, H. Q., Urano, F., Till, J. H., Hubbard, S. R., Harding, H. P., Clark, S.
G., and Ron, D. (2002). IREl couples endoplasmic reticulum load to secretory
capacity by processing the XBP-1 mRNA. Nature 415, 92-96.
Cox, J. S., and Walter, P. (1996). A novel mechanism for regulating activity of a
transcription factor that controls the unfolded protein response. Cell 87, 391-404.
Dong, C., Davis, R. J., and Flavell, R. A. (2002). MAP kinases in the immune response.
Annual Reviews of Immunology 20, 55-72.
Hoogewijs, D., Houthoofd, K., Matthijssens, F., Vandersompele, J., and Vanfleteren, J.
R. (2008). Selection and validation of a set of reliable reference genes for
quantitative sod gene expression analysis in C. elegans. BMC Molecular Biology 9.
doi:10. 1186/1471-2199-9-9.
Iwakoshi, N. N., Lee, A.-H., Vallabhajosyula, P., Otipoby, K. L., Rajewsky, K., and
Glimcher, L. H. (2003). Plasma cell differentiation and the unfolded protein
response intersect at the transcription factor XBP-1. Nature Immunology 4, 321-329.
Kamath, R. S. et al. (2003). Systematic functional analysis of the Caenorhabditiselegans
genome using RNAi. Nature 421, 231-237.
Kaser, A. et al. (2008). XBP 1 links ER stress to intestinal inflammation and confers
genetic risk for human inflammatory bowel disease. Cell 134, 743-756.
Kaufman, R. J., Scheuner, D., SchrOder, M., Shen, X., Lee, K., Liu, C. Y., and Arnold, S.
M. (2002). The unfolded protein response in nutrient sensing and differentiation.
Nature Reviews Molecular Cell Biology 3, 411-421.
Kim, D. H. et al. (2002). A conserved p38 MAP kinase pathway in Caenorhabditis
elegans innate immunity. Science 297, 623-626.
Kim, D. H., Liberati, N. T., Mizuno, T., Inoue, H., Hisamoto, N., Matsumoto, K., and
Ausubel, F. M. (2004). Integration of Caenorhabditiselegans MAPK pathways
mediating immunity and stress resistance by MEK- 1 MAPK kinase and VHP- 1
MAPK phosphatase. Proceedings of the National Academy of Sciences 101, 1099010994.
Luke, C. J. et al. (2007). An intracellular serpin regulates necrosis by inhibiting the
induction and sequelae of lysosomal injury. Cell 130, 1108-1119.
Mori, K., Kawahara, T., Yoshida, H., Yanagi, H., and Yura, T. (1996). Signalling from
endoplasmic reticulum to nucleus: transcription factor with a basic-leucine zipper
motif is required for the unfolded protein-response pathway. Genes Cells 1, 803817.
Pocock, R., and Hobert, 0. (2008). Oxygen levels affect axon guidance and neuronal
migration in Caenorhabditiselegans. Nature Neuroscience 11, 894-900.
Pujol, N. et al. (2001). A reverse genetic analysis of components of the Toll signaling
pathway in Caenorhabditiselegans. Current Biology 11, 809-821.
Rahme, L. G., Stevens, E. J., Wolfort, S. F., Shao, J., Tompkins, R. G., and Ausubel, F.
M. (1995). Common virulence factors for bacterial pathogenicity in plants and
animals. Science 268, 1899-1902.
Reddy, K. C., Anderson, E. C., Kruglyak, L., and Kim, D. H. (2009). A polymorphism in
npr-1 is a behavioral determinant of pathogen susceptibility in C. elegans. Science
323, 382-384.
Reimold, A. M. et al. (2001). Plasma cell differentiation requires the transcription factor
XBP-1. Nature 412, 300-307.
Rual, J. F. et al. (2004). Toward improving Caenorhabditiselegans phenome mapping
with an ORFeome-based RNAi library. Genome Research 14, 2162-2168.
Rutkowski, D. T. et al. (2006). Adaptation to ER Stress is mediated by differential
stabilities of pro-survival and pro-apototic mRNAs and proteins. PLOS Biology 4,
e374. doi: 10.1371/joumal.pbio.0040374.
SchrOder, M., and Kaufman, R. J. (2005). The mammalian unfolded protein response.
Annual Reviews in Biochemistry 74, 739-789.
Shapira, M., Hamlin, B. J., Rong, J. M., Chen, K., Ronen, M., and Tan, M.-W. (2006). A
conserved role for a GATA transcription factor in regulating epithelial innate
immune responses. Proceedings of the National Academy of Sciences 103, 1408614091.
Shen, X. et al. (2001). Complementary signaling pathways regulate the unfolded protein
response and are required for C. elegans development. Cell 107, 893-903.
Shen, X., Ellis, R. E., Sakaki, K., and Kaufman, R. J. (2005). Genetic interactions due to
constitutive and inducible gene regulation mediated by the unfolded protein
response in C. elegans. PLOS genetics 1, e37. doi:10.1371/joumal.pgen.0010037.
Shivers, R. P., Kooistra, T., Chu, S. W., Pagano, D. J., and Kim, D. H. (2009). Tissuespecific activities of an immune signaling pathway regulate physiological responses
to pathogenic and nutritional bacteria in C. elegans. Cell Host and Microbe 6, 321330.
Shivers, R. P. et al. (2010). Phosphorylation of the conserved transcription factor ATF-7
by PMK-1 p38 MAPK regulates innate immunity in Caenorhabditiselegans. PLoS
genetics 6, e1000892. doi:10.1371/journal.pgen.1000892.
Styer, K. L., Singh, V., Macosko, E., Steele, S. E., Bargmann, C. I., and Aballay, A.
(2008). Innate immunity in Caenorhabditiselegans is regulated by neurons
expressing NPR-1/GPCR. Science 322, 460-464.
Syntichaki, P., Xu, K., Driscoll, M., and Tavernarakis, N. (2002). Specific aspartyl and
calpain proteases are required for neurodegeneration in C.elegans. Nature 419, 939944.
Tan, M.-W., Mahajan-Miklos, S., and Ausubel, F. M. (1999). Killing of Caenorhabditis
elegans by Pseudomonas aeruginosaused to model mammalian bacterial
pathogenesis. Proceedings of the National Academy of Sciences 96, 715-720.
Timmons, L., and Fire, A. (1998). Specific interference by ingested dsRNA. Nature 395,
854.
Todd, D. J., Lee, A.-H., and Glimcher, L. H. (2008). The endoplasmic reticulum stress
response in immunity and autoimmunity. Nature Reviews Immunology 8, 663-674.
Troemel, E. R., Chu, S. W., Reinke, V., Lee, S. S., Ausubel, F. M., and Kim, D. H.
(2006). p3 8 MAPK regulates expression of immune response genes and contributes
to longevity in C. elegans. PLOS genetics 2, 1725-1739.
Troemel, E. R., Felix, M.-A., Whiteman, N. K., Barriere, A., and Ausubel, F. M. (2008).
Microsporidia are natural intracellular parasites of the nematode Caenorhabditis
elegans. PLOS Biology 6, e309. doi:10. 1371 /journal.pbio.0060309.
Willems, E., Leyns, L., and Vandesompele, J. (2008). Standardization of real-time PCR
gene expression data from independent biological replicates. Analytical
Biochemistry 379, 127-129.
Wong, D., Bazopoulou, D., Pujol, N., Tavernarakis, N. and Ewbank, JJ. (2007). Genomewide investigation reveals pathogen-specific and shared signatures in the response of
Caenorhabditiselegans to infection. Genome Biology 8, RI 94. doi: 10.1186/gb2007-8-9-r194.
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.
Zhang, K., Shen, X., Wu, J., Sakaki, K., Saunders, T., Rutkowski, D. T., Back, S. H., and
Kaufman, R. J. (2006). Endoplasmic reticulum stress activates cleavage of CREBH
to induce a systemic inflammatory response. Cell 124, 587-599.
Zhang, Y., Lu, H., and Bargmann, C. I. (2005). Pathogenic bacteria induce aversive
olfactory learning in Caenorhabditiselegans. Nature 438, 179-184.
Chapter Three
Physiological IRE-1-XBP-1 and PEK-1 signaling in
endoplasmic reticulum homeostasis in Caenorhabditis elegans
Claire E. Richardson, Stephanie Kinkel, and Dennis H. Kim
S. K. performed the experiments presented in Figure 2.
Originally published in PLoS Genetics, November 2011.
114
SUMMARY
An excess of unfolded proteins in the endoplasmic reticulum (ER stress)
activates the Unfolded Protein Response, a compensatory signaling response that
is mediated by the IRE-1, PERK/PEK-1, and ATF-6 pathways in metazoans.
Genetic studies have implicated roles for UPR signaling in animal development
and disease, but the function of the UPR under physiological conditions, in the
absence of chemical agents administered to induce ER stress, is not well
understood. In this chapter, we show that in Caenorhabditiselegans, XBP-1
deficiency results in constitutive ER stress, reflected by increased basal levels of
IRE-1 and PEK-1 activity under physiological conditions. We define a dynamic,
temperature-dependent requirement for XBP-1 and PEK-1 activities that
increases with immune activation and at elevated physiological temperatures in
C. elegans. Our data suggest that the negative feedback loops involving the
activation of IRE-i-XBP-1 and PEK-1 pathways serve essential roles not only at
the extremes of ER stress, but also in the maintenance of ER homeostasis under
physiological conditions.
116
INTRODUCTION
The accumulation of misfolded proteins in the endoplasmic reticulum (ER), also
known as ER stress, activates the Unfolded Protein Response (UPR), which upregulates
the synthesis of chaperones such as BiP and components of ER-associated degradation
(ERAD), promotes ER expansion, and attenuates translation (Mori 2009; Ron et al. 2007;
Schr~der et al. 2005). The UPR is conserved from yeast to humans and in metazoans is
comprised of three branches, mediated by the transmembrane ER luminal sensors IRE- 1,
PERK/PEK-1, and ATF-6 (Mori 2009, Ron et al. 2007, Schrider et al., 2005). In
response to ER stress, IRE-I oligomerizes, activating an endoribonuclease domain that
splices the mRNA of xbp-1 to enable the generation of the activated form of the XBP- 1
transcription factor (Calfon et al. 2002; Shamu et al. 1996; X. Shen et al. 2001; Yoshida
et al. 2001). PERK phosphorylates the translation initiation factor eIF-2a, causing global
translational attenuation that diminishes the secretory load to the ER (Harding et al.
2000a). In addition, phosphorylation of eIF-2 a selectively increases the translation of
ATF4, a transcription factor that regulates stress responses (Harding et al. 2000b). ATF-6
undergoes proteolysis, releasing the cytosolic domain of ATF-6, which functions as a
transcription factor that translocates to the nucleus and activates transcription of UPR
genes (Yoshida et al. 2000).
Genetic studies suggest essential roles for UPR signaling in animal development.
In mice, genetic studies focused on either the IRE-i -XBP- 1 or the PERK pathway have
shown that each functions in the development of specialized cell types, including plasma
cells, pancreatic beta cells, hepatocytes, and intestinal epithelial cells (Lee et al. 2008;
Kaser et al. 2008; Schr6der et al. 2005; Shaffer et al. 2004; Todd et al. 2008; Zhang et al.
2006). In Caenorhabditiselegans, mutants deficient in any one of the three branches of
the UPR are viable, but combining a deficiency in the IRE-i -XBP- 1 pathway with lossof-function mutations in either the ATF-6 or PEK- 1 branch has been reported to result in
larval lethality (Shen et al. 2001, Shen et al. 2005). These studies suggest that the UPR is
required for animal development, but the nature of this essential role has not been
defined.
The experimental analysis of UPR signaling both in yeast and in mammalian cells
has been greatly facilitated by the use of chemical agents that induce ER stress, such as
the N-linked glycosylation inhibitor tunicamycin, the calcium pump inhibitor
thapsigargin, and the reducing agent dithiothreitol (DTT). However, the activation of the
UPR under physiological conditions is less well understood (Rutkowski et al. 2010).
Constitutive IRE- 1 activity has been observed in diverse types of mammalian cells,
particularly with high secretory activity or in the setting of increased inflammatory
signaling (Lee et al. 2008; Reimold et al. 2001; Todd et al. 2008; Zhang et al. 2006).
These studies suggest critical roles for IRE-1 -XBP- 1 signaling in physiology and
development, some of which have been proposed to be independent of its role in
maintaining protein folding homeostasis in the ER (Hu et al. 2009; Lee et al. 2008;
Martinon et al. 2010).
In Chapter Two, we showed that XBP-1 is required for C. elegans larval
development on pathogenic Pseudomonas aeruginosa,conferring protection to the C.
elegans host against the ER stress caused by its own secretory innate immune response to
infection. Our study established that the innate immune response to microbial pathogens
represents a physiologically relevant source of ER stress that necessitates XBP- 1.
We sought to better understand the consequences of UPR deficiency under
physiological conditions during C. elegans larval development. We describe our studies
that suggest that even in the absence of ER stress induced by exogenously administered
chemical agents, the IRE-1 -XBP- 1 pathway, in concert with the PEK- 1 pathway,
functions in a homeostatic loop that is under constitutive activation during C. elegans
larval development. Our data implicate an essential role for the UPR in ER homeostasis,
not only in the response to toxin-induced ER stress, but also under basal physiological
conditions.
119
RESULTS
Elevated constitutive IRE-1 activity in C. elegans xbp-1 mutants
The detection of IRE-I activity provides a sensitive and responsive measure of
ER stress. Most methods to measure IRE-I activity require functional IRE-1 -XBP- 1
output, relying on either detection of the activated spliced form of the xbp-1 mRNA or
the transcriptional activity of the resulting XBP- 1 protein. In order to follow IRE-I
activity in the absence of a functional XBP- 1 protein, we utilized the C. elegans xbp1(zc12) mutant, which has a C->T mutation that results in an early premature stop codon
(Calfon et al. 2002). We reasoned that we could detect IRE-I-mediated splicing of xbp1(zc12) mRNA by quantitative RT-PCR (qPCR), as we did in Chapter Two for wild type
xbp-1 mRNA, as a measure of IRE- 1 activity.
We anticipated, however, that the xbp-1(zc12) mRNA might be degraded by
nonsense-mediated decay (NMD) (Rehwinkel et al. 2006), which would reduce the
abundance of xbp-1(zc12) mRNA (Figure 1a). Thus, we constructed a strain carrying
xbp-1(zc12) and a null allele of smg-2, the C. elegans homolog of the NMD component
Upfl (Page et al. 1999). Indeed, we observed that the level of xbp-1 mRNA in the xbp1(zc12) mutant was markedly diminished compared with the level of xbp-1(zc12) mRNA
in the smg-2(qd101), xbp-1(zc12) mutant (Figure lb). These data confirmed that xbp1(zc12) mRNA is a substrate for the NMD pathway, but that inhibition of NMD permits
detection of xbp-1(zc12) mRNA. As expected from the predicted truncated protein
product made from translation of the xbp-1(zc12) mRNA (Figure la), loss of NMD had
no effect on the null phenotype of the xbp-I(zc12) allele, as assessed by the effect of the
smg-2(qd101) mutation on expression of an xbp-1-regulated gene, the C. elegans BiP
homolog hsp-4 (Figure 1b). We next examined the level of WT xbp-] mRNA in the sng2(qd]01) mutant, and we observed that NMD inhibition increased the level of xbp-1
mRNA 2-fold relative to WT C. elegans (Figure 1c), which suggests that the NMD
complex may function to decrease the level of WT xbp-1 mRNA. This observation is
consistent with a prior report suggesting that stress-induced genes may be NMD targets
(Gardner 2008), although we hypothesize that the relatively early termination codon
present in the xbp-1 mRNA prior to IRE-I-mediated splicing may also contribute to
recognition and degradation by the NMD pathway. Consistent with this explanation, after
exposing both the WT and smg-2(qd101) strains to tunicamycin for 4 h, the level of IRE1-spliced xbp-1 mRNA was similar between the two strains (Figure Ic). Furthermore, the
loss of NMD did not increase the lethality of either the WT strain or xbp-1(zc12) mutant
when grown in the presence of tunicamycin (Figure Id). Although loss of NMD caused
neither increased larval lethality during growth in the presence of tunicamycin nor
elevated IRE-I-mediated xbp-1 splicing after exposure to tunicamycin, we cannot
exclude the possibility that loss of NMD does cause a low level of increased ER stress
which our assays were too insensitive to detect.
Figure 1. XBP-1 deficiency results in a dramatic increase in IRE-1 activity. (a)
Detection of xbp-1 mRNA splicing by IRE-I in the C. elegans xbp-1(zc]2) mutant.
Schematic of the unspliced and spliced xbp-1 mRNA, noting the position of the
premature termination codon present in the xbp-1(zc12) allele. The smg-2(qd101)
mutation results in inactivation of the NMD pathway, stabilizing the xbp-1(zc12) mRNA
for detection. (b) Quantitative real-time PCR measurements of mRNA levels in the xbp](zcl2) and smg-2(qd101); xbp-1(zc12) mutants relative to the smg-2(qd101) mutant
synchronized in the L3 stage. (c) Quantitative real-time PCR measurements of levels of
total and spliced xbp-1 mRNA in the smg-2(qdl 01) strain relative to WT grown to the L3
stage and then shifted to plates with or without tunicamycin (5 tg/mL) for 4 h. (d) Larval
development and survival assay showing the proportion of animals of each of the
indicated strains that reach the indicated stage after 4 d of development from eggs laid on
plates containing tunicamycin (2.5 [tg/mL) at 16'C. (e) Quantitative real-time PCR
measurements of levels of total and spliced xbp-1 mRNA in the smg-2(qd101); xbp1(zc12) strain relative to the smg-2(qd101) strain treated as in c. (In b, c, and e, values
represent fold change ± s.e.m., n = 3 independent experiments, *P < 0.05, ***P < 0.001,
two-way ANOVA with Bonferroni post test).
Figure 1
m xb-I to
ZC12
C3 hwp4
~-lu mRNA
IRE-I
(qdl;l)
=c12
xbp-ls mRNA
f=214
158
155
134
D3L4
spltice
TM:
+
-
wi
-
+
srrv2qdo1)
TM:
wVn-2(qdl0l) srrq-2(qdl0l):rhp-~zC12m
mL1-L3
M dead
Comparing levels of IRE-I activity between smg-2(qdl 01) and smg-2 (qdl 01);
xbp-1(zc12) animals, we observed a dramatic elevation in the level of spliced xbp-1
mRNA in the smg-2(qd101); xbp-1(zc12) strain (Figure le). To provide a measure for
comparison, the basal elevation of spliced xbp-1 mRNA in the smg-2(qdl 01); xbp1(zc 12) mutant far exceeded the level of spliced xbp-] mRNA in the smg-2(qd101)
mutant even after administration of tunicamycin. Treating the smg-2(qd101); xbp-1(zc12)
strain with tunicamycin resulted in only a minor additional increase in spliced xbp-1
mRNA compared with the magnitude of elevation in spliced xbp-1 in that strain under
standard growth conditions (Figure l e). These data show that XBP- 1 deficiency results in
a dramatic increase in IRE-I activity, even in the absence of exogenously administered
agents such as tunicamycin.
Increased constitutive PEK-1 activation in C. elegans xbp-1 mutants
If the elevated level of IRE-I activity observed in the smg-2(qd101); xbp-](zc]2)
mutant were indicative of increased ER stress due to loss of xbp-1, we might anticipate
compensatory activation of the PEK- 1 and/or ATF-6 pathways in the absence of XBP- 1.
We therefore sought to determine levels of PEK- 1 activity in an xbp-1 mutant through the
detection of eIF-2a phosphorylation. In particular, these measurements would provide an
additional measure of ER stress in the xbp-1 mutant that is not dependent on the
inactivation of the NMD pathway and its aforementioned experimental caveats.
Antibodies raised against mammalian eIF-2a and specifically phosphorylated eIF-2a (PeIF-2a) cross-react with the highly homologous C. elegans protein (Hamanaka et al.
2005; Nukazuka et al. 2008). We detected a single band in immunoblots using these
antibodies with lysates from WT C. elegans (Figure 2a). We observed that elF-2a
phosphorylation was induced by a 4 h exposure to a high dose of tunicamycin in a PEK1-dependent manner (Figure 2a). eIF-2a phosphorylation appears to be a less sensitive
measure of ER stress than IRE-I-mediated xbp-1 mRNA splicing, as we did not observe
a significant increase in eIF-2a phosphorylation in response to standard doses of
tunicamycin sufficient to induce xbp-1 mRNA splicing (Figures 1c and 1e).
We next determined PEK-1 activity under basal physiological conditions in the
xbp-1 mutant. Paralleling our observations of increased xbp-] mRNA splicing in the xbp1 mutant, we saw induction of PEK- 1-mediated eIF-2a phosphorylation relative to WT in
the absence of exogenously administered agents to induce ER stress at 16'C (Figure 2b).
The magnitude of the effect of XBP- 1 deficiency on PEK- 1 activity was comparable to
the induction of PEK- 1 in WT by treatment with high-dose tunicamycin. Taken together,
the increase in levels of IRE-I and PEK-l activities in the xbp-1 mutant suggests that
XBP- 1 deficiency is accompanied by a marked increase in constitutive ER stress under
basal physiological conditions.
125
Figure 2. XBP-1 deficiency increases PEK-1 dependent phosphorylation of eIF2a.
(a) PEK- 1 dependent phosphorylation of eIF2a is induced by high dose (50ug/ml)
tunicamycin. Western blot of P-eIF2a, total eIF2a and tubulin in WT and pek-1(ok2 75)
strains grown to the L4 stage and then shifted to plates with or without tunicamycin (50
[tg/mL) for 4 hours. (b) PEK- 1 dependent phosphorylation of eIF2a is induced by XBP- 1
deficiency. Western blot of P-eIF2ct, total eIF2a and tubulin in WT, xbp-1(tm2482), xbp1(tm2482);pek-1(ok2 75) and pek-1 (ok2 75) strains grown to the L4 stage.
Figure 2
WT
-
pek-1
+
-
+
Tunicamycin
P-elF2a
Total eIF2a
P-tubulin
P-elF2a
mel
am
-
mm
m Alo -s
Total eIF2a
-tubulin
The effects of innate immune activation on ER stress levels in the xbp-1 mutant
In Chapter Two, we showed that the activation of innate immunity by infection
with pathogenic P. aeruginosainduces ER stress, and that XBP- 1 serves an essential role
in protecting the host against the detrimental effects of immune activation. Our prior
ultrastructural analysis of the ER in xbp-1 mutants suggested that disruption of ER
homeostasis contributes to this phenotype. One explanation for these observations is that
ER homeostasis in the xbp-1 mutant might be minimally perturbed under basal
physiological conditions but have a pronounced sensitivity to ER stress from endogenous
(e.g. immune activation) or exogenous (e.g. tunicamycin) sources. However, the data in
Figures 1 and 2 suggest that even during physiological growth and development, XBP- 1
deficiency results in a marked elevation in levels of basal ER stress. We hypothesized,
therefore, that under these circumstances, the activation of innate immunity might further
increase ER stress levels.
The smg-2(qd101); xbp-1(zc12) strain provided the opportunity to assess levels of
ER stress caused by immune activation in the setting of XBP- 1 deficiency. Whereas a 4 h
exposure of the WT strain to P. aeruginosaPA14 causes a two-fold increase in spliced
xbp-1 mRNA relative to exposure to the relatively non-pathogenic bacterial food
Escherichiacoli OP50 (Chapter 2, Figure 3a), we observe a blunted response to P.
aeruginosainfection in the smg-2(qd101) mutant (Figure 3a). This observation is likely
due to the 5-fold elevation in spliced xbp-1 mRNA levels in the smg-2(qd101) mutant
(Figure 1c), which may buffer the ER from the stress caused by pathogen-induced
immune activation. Nevertheless, we observed that the level of spliced xbp-1 mRNA in
the smg-2(qd101); xbp-1(zc12) mutant was increased by a 4 h exposure to P. aeruginosa
relative to the smg-2(qd1); xbp-](zc12) mutant treated in parallel with E. coli (Figure
3a). Specifically, under these treatment conditions, the level of spliced xbp- 1 mRNA in
the smg-2(qd]01); xbp-](zc12) mutant was 20-fold greater than that of the smg-2(qd101)
mutant in the absence of additional stress, whereas exposure to P. aeruginosaincreased
the level of spliced xbp-1 mRNA to over 25-fold that of the smg-2(qd]01) mutant
(Figure 3a). The total amount of xbp-1 mRNA was unchanged between smg-2(qd101)
and smg-2(qd101); xbp-1(zc12) strains, indicating that the increase in spliced xbp-1
mRNA is due to increased IRE-I activation. Also, because we observed UPR induction
in both in the WT strain, where NMD is functional, and in the smg-2; xbp-1 strain, where
NMD is absent, we infer that NMD does not play a role in the activation of IRE-I in
response to P. aeruginosainfection.
We observed persistent elevation of spliced xbp-1 mRNA after an 11 h exposure
to P. aeruginosa,above and beyond the elevated basal levels of spliced xbp-1 mRNA in
the xbp-1 mutant, suggesting that IRE-I activity is not attenuated under conditions of
physiological ER stress (Figure 3b).
Figure 3. Pathogen-induced immune activation exacerbates ER stress levels in XBP1 deficiency in C. elegans. Quantitative real-time PCR measurements of levels of total
and spliced xbp-1 mRNA in the indicated strains grown from synchronized Li s for 23 h
at 20'C and then shifted to plates with or without P. aeruginosaPA14 at 25'C for (a) 4 h
or (b) 11 h. Values represent fold change + s.e.m. (n = 2 independent experiments, ***P
<0.001, two-way ANOVA with Bonferroni post test). The WT strain exposed to either
treatment was normalized to WT without P. aeruginosa;all other strains were normalized
to smg-2(qd101) without P. aeruginosa.
Figure 3
total
spliced
P. aeruginosa: -
+
-
+
WT
+
--
smg-2(qdl0l)
I
total
spliced
+
P. aeruginosa: -
+
+
0
AWT
smg-2(qdl0l)
-
+
Previously, we established that the ER stress induced by exposure to P.
aeruginosa,as well as the lethality of the xbp-1 mutant during infection by P.
aeruginosa,are suppressed by a loss-of-function mutation in pmk-1, which encodes a
conserved p38 mitogen-activated protein kinase (MAPK) that regulates innate immunity
in C. elegans (Kim et al. 2002). Our interpretation of these data was that loss of PMK- I
activity diminished the secretory load on the ER by attenuating the innate immune
response. In support of this interpretation, we found that the pathogen-induced increase in
spliced xbp-] mRNA in smg-2(qd101); xbp-l(zc]2) was suppressed in the smg-2(qd101);
xbp-1(zc12); pmk-1(km25) mutant, although the basal levels on E. coli OP50 nevertheless
remained markedly elevated (Figure 3a). These data provides quantitative support for a
model in which the activation of PMK- 1-mediated innate immunity is a physiologically
relevant source of ER stress, which in XBP- 1-deficient animals exacerbates an already
elevated level of ER stress to cause larval lethality.
A temperature-dependent requirement for XBP-1 and PEK-1 activities in C. elegans
development and survival
What are the functional consequences of the elevated ER stress present in the xbp1 mutant? The xbp-1 mutant, while viable, exhibits increased sensitivity to exogenously
administered ER stress as well as physiological ER stress from immune activation
(Chapter 2; Bischof et al. 2008). Inactivation of both xbp-1 andpek-1 was previously
reported to result in larval arrest when propagated at 20'C (Shen et al. 2005). Our
observations of constitutive ER stress in the xbp-1 mutant and increased PEK- 1 activity
suggest a compensatory functional role for pek-1, and thus we sought to further
characterize the larval arrest phenotype of the xbp-1; pek-1 mutant.
Surprisingly, we observed that the xbp-1 (tm2482); pek-1 (ok2 75) double mutant
exhibited temperature-dependent viability over the physiological temperature range of C.
elegans (Figure 4a). The larval development of the xbp-1 (tm2482); pek-1 (ok2 75) mutant
was similar to that of WT at 16'C. At 20'C, however, approximately half of xbp1(tm2482); pek-1(ok2 75) eggs developed to become gravid adults, while the remainder
arrested during larval development at the L2 and L3 stages. These arrested larvae died
over the course of several days with intestinal degeneration as previously described (Shen
et al., 2005). At temperatures greater than 23'C, larval lethality was 100%. At 25'C,
100% of the population died in the L 1 and L2 stages after just 2 days. The physiological
temperature range for propagation of C. elegans in the laboratory is generally 15'C to
25 0 C, with optimal reproduction at 20'C. Thus, the observed temperature dependence is
observed not at "heat shock" temperatures, but rather, well within the range of
physiological temperatures for C. elegans.
Figure 4. Temperature-sensitive lethality of the xbp-1; pek-J double mutant. (a)
Development of the xbp-1(tm2482); pek-1 (ok2 75) mutant across physiological
temperatures. Values represent average fraction of eggs developed to the indicated stage
- s.e.m. (n = 3 independent experiments at 20'C, 2 independent experiments at all other
temperatures; ***P < 0.001, two-way ANOVA with Bonferroni post test). (b) Lifespan
of indicated strains grown at 16'C to the L4 stage, then shifted to either 16'C to 25'C.
Results are representative of two independent experiments.
Figure 4
1.0-
_0
,,,
U)
* L4+
M gravid
Cu
C
.
fC )
U
co E-
0.5-
L0
C)
0 >
co _0
U-
0.0-r
15
20
25
Temperature (*C)
b
9
c
0
1600
C
U-
0
5
10
Time (d)
xbp-1(tm2482)
,*- pek-1(ok275)
xbp- 1(tm2482);
pek- 1(ok275)
xbp-1(zc 12);
pek- 1(tm629)
C:
0
U-
0
5
10
Time (d)
The temperature dependence of xbp-1 (tm2482); pek- (ok2 75) lethality permitted
the investigation of whether the larval lethality of the xbp-1; pek-1 mutant is due to a
requirement for XBP- I and PEK- 1 at a specific stage of development, or whether the
activities of XBP-1 and PEK- 1 are required constitutively for viability at other life stages.
Specifically, we propagated two xbp-1; pek-1 mutants comprised of different mutant
alleles at 16'C until the animals reached the L4 larval stage, then either maintained the
mutants at 16'C or shifted them to 25'C to monitor survival. When shifted to 25'C, the
xbp-1; pek-1 double mutants exhibited a sharp decrease in survival as compared with the
strains maintained at 16'C (Figures 4b). These observations suggest that the activity of
either XBP-1 or PEK- 1 is not specifically required at a particular developmental stage;
instead, the constitutive activities of XBP- 1 and PEK- 1 are required for survival at
physiological temperatures.
XBP-1 and PEK-1 maintain intestinal cell homeostasis during ER stress caused by
basal and induced innate immunity
Although we observed in Chapter Two that pek-1 and atf-6 single mutants did not
exhibit larval lethality in the presence of pathogenic bacteria, our data presented in this
paper suggest that PEK- 1 functions in parallel to XBP- 1 under physiological conditions
in C. elegans to maintain ER homeostasis. Because the xbp-1; pek-1 mutant is viable at
least to adulthood at 16'C, we were able to ask whether PEK- 1 contributes to protection
against immune activation in the absence of XBP-1. Populations of synchronized eggs
were grown at 16'C with P. aeruginosaas the only food source and development was
monitored over time. P. aeruginosahas been shown to exhibit markedly diminished
pathogenicity to C. elegans adults at 16'C relative to at 25'C (Kurz et al. 2007), and we
found this to also be the case during larval development. Specifically, the pmk-i mutant
was able to complete larval development on P. aeruginosaat 16'C (Figure 5a), whereas
only half of the pmk-] eggs grown on P. aeruginosadevelop to the L4 stage at 25 C
(Chapter 2), indicating that immune activation is less important for development in the
presence of P. aeruginosagrown at 16'C than it is at 25'C. Likewise, the larval
development of the xbp-1 mutant, which is severely compromised on P. aeruginosaat
25'C (Chapter 2), was equivalent to that of WT at 16'C (Figure 5a). Both the diminished
pathogenicity of P. aeruginosaat 160 C and the aforementioned temperature-sensitive
requirement for UPR function may contribute the survival of the xbp-] mutant at 16'C.
Nevertheless, even under these conditions, the xbp-1(tm2482); pek-](ok275) mutant
exhibited complete larval lethality on P. aeruginosaat 16'C, reminiscent of the larval
lethality of xbp-1 on P. aeruginosagrown at 25'C. Eliminating PMK-1-mediated
immunity completely rescued this larval lethality (Figure 5a), demonstrating that PEK- 1
functions with XBP- 1 to protect against PMK- 1-mediated immune activation during
larval development.
Figure 5. XBP-1 and PEK-1 each protect against elevated physiological temperature
and immune activity. (a) Development of indicated mutants from eggs to the L4 larval
stage or older after 4 d at 16'C on P. aeruginosaPA14. Values represent mean t s.d.
from 1 of 2 representative experiments (n = 4 plates with 20-50 eggs each, ***P < 0.001,
one-way ANOVA with Bonferroni post test). (b) Development of xbp-1(tm2482); pek1(ok275) and xbp-1(tm2482); pmk-](km25);pek-](ok275) mutants from eggs on E. coli
OP50. Values represent average fraction of eggs developed to the indicated stage
±
s.e.m. (n = 2 independent experiments, **P < 0.01, ***P < 0.001, two-way ANOVA
with Bonferroni post test). (c) Development of indicated mutants from eggs to the L4
larval stage or older after 2 d at 27'C on E. coli OP50. Values represent mean
±
s.e.m. (n
= 3 independent experiments, * * *P < 0.001, one-way ANOVA with Bonferroni post
test). (d) Development of indicated mutants from eggs to the L4 larval stage or older after
2 d at 27'C on P. aeruginosaPA 14. Values represent mean ± s.d. from 1 of 2
representative experiments (n = 3-4 plates with 20-60 eggs each, ***P < 0.001, one-way
ANOVA with Bonferroni post test).
Figure 5
R aeruginosa, 160C
E. coli
1.0.
C
xbp-1;pek-1
M
xbp-1; pmk-1; pek-1
0.4-
0.5
co
*
0.2-
~0.0
0.0-
P9
23
25
Temperature (0C)
k
R aeruginosa, 270C
E. coli, 27*C
1.0
1.0-
0.0
J.A\
9~
0.0-
94
9
Larval arrest of xbp-1; pek-1 mutants has been reported to be accompanied by
evidence of intestinal degeneration, including the appearance of vacuoles and lightreflective aggregates in intestinal cells, degradation of intestinal tissues, and distention of
the intestinal lumen (Shen et al. 2005). We observe similar morphology not only in xbp1; pek-] larvae at 23'C on E. coli OP50, but also at 16'C on P. aeruginosa.The similar
appearance between xbp-1; pek-] larvae dying either at 16'C on pathogenic bacteria or at
23'C on E. coli OP50 led us to consider whether ER stress arising from intestinal innate
immune activation might contribute in a similar manner to both conditions. We have
previously characterized PMK- 1-mediated innate immunity and noted both basal and
induced components to immunity regulated by PMK- 1 (Troemel et al. 2006). We
therefore hypothesized that basal immune activity under standard growth conditions
could present a low level of ER stress that is severely exacerbated in the absence of intact
physiological UPR function, leading to larval lethality of the xbp-1; pek-1 mutants even
in the absence of pathogen infection. Consistent with this hypothesis, we observed that
pmk-I loss-of-function was able to partially suppress the larval lethality of the xbp-1;
pek-1 double mutant at 23'C and 25'C (Figure 5b).
One explanation for the temperature-sensitive lethality of the xbp-1; pek-1 mutant
might have been that increased temperature leads to increased PMK- 1 pathway
activation, perhaps as the "non-pathogenic" E. coli OP50 becomes slightly pathogenic.
However, the temperature-sensitive lethality is not abrogated by loss of PMK-1; instead,
the xbp-1; pmk-1; pek-] mutant exhibits larval lethality at a temperature several degrees
higher than the xbp-1; pek-] mutant (Figure 5b). Furthermore, the temperature-sensitive
larval lethality of the xbp-1; pek-1 mutant on E. coli OP50 was not suppressed by the
presence of the bacteriostatic drug ampicillin (data not shown). These data indicate that
basal immune activation and temperature are distinct sources of ER stress that function in
parallel during growth on E. coli OP50.
Thermal stress necessitates UPR signaling for survival in C. elegans
The temperature-dependent larval lethality of the xbp-1; pek-1 mutant over a
physiological temperature range suggested that UPR signaling might be required for
survival in response to thermal stress. Indeed, we observed that the xbp-1 mutant
exhibited larval lethality when grown at 27'C, an elevated temperature at which WT N2
C. elegans exhibits a reduced brood size and increased dauer formation (Figure 5c).
Similar to our observation that depletion of basal immunity rescued the development of
the xbp-1; pek-1 mutant when propagated on E. coli OP50, the temperature-sensitive
lethality in the xbp-1 mutant was suppressed in the xbp-1; pmk-] double mutant (Figure
5c).
Unlike the xbp-] mutant, the development of the pek-1 mutant at 27'C was
similar to WT. This is reminiscent of our observation in Chapter Two that the pek-1
mutant did not exhibit the larval lethality found in xbp-] when grown on P. aeruginosaat
25'C. However, we next grew the pek-] mutant on P. aeruginosaat 27'C, reasoning that
the elevated temperature would not only increase the ER stress caused by basal growth,
but also enhance the pathogenicity of the P. aeruginosaand thereby increase the immune
response relative to that at 25'C. Indeed, the pmk-] mutant exhibited 100% larval
lethality on P. aeruginosaat 27'C (Figure 5d), as compared with the 50% lethality we
have previously reported for the pmk-] mutant on P. aeruginosaat 25'C (Chapter 2). The
increased susceptibility of this immune-deficient mutant to P. aeruginosaat 27'C relative
to 25'C indicates that the increased temperature causes an increase in P. aeruginosa
pathogenicity. On P. aeruginosaat 27'C, the pek-1 mutant exhibited larval lethality
relative to the WT strain grown (Figure 5d). These data further suggest that PEK- 1
functions in parallel with XBP- 1 to protect C. elegans against the ER stress caused by
immune activation.
The PMK-1 pathway protects against exogenous ER stress
We showed in Figures 5b and 5c that loss of PMK-1 improves larval development
of the xbp-1; pek-1 mutant and the xbp-1 mutant, respectively, in the absence of
infection. We suggested that the mechanism behind this phenomenon is that the
previously described basal immune activity through the PMK- 1 pathway (Troemel et al.
2006) is a source of ER stress. We also considered the possibility, however, that the
PMK- 1 pathway plays an immunity-independent role in exacerbating ER stress in the
setting of UPR depletion. To test this possibility, we examined the ability of WT and
UPR mutants to develop in the presence of tunicamycin with or without functional pmk-I
(Figure 6). We found that the pmk-i mutant exhibited severe sensitivity to development
in the presence of tunicamycin. In fact, the pmk-] mutant exhibited greater lethality at a
lower dose of tunicamycin than either the xbp-1 orpek-1 single mutants (Figure 6). These
data suggest that the PMK-1 pathway influences ER stress in two ways: First, during
infection or under standard growth conditions in the setting of UPR depletion, activation
of the PMK- 1 pathway generates an increased secretory load that contributes to ER
stress. When ER stress is induced exogenously with tunicamycin, however, PMK- 1
pathway activity is protective against larval lethality.
Figure 6. PMK-1 protects against exogenously induced ER stress. Strains were egg
laid on 4-5 plates seeded with E. coli OP50 and containing the indicated concentration of
tunicamycin. Eggs were counted, and stages were scored after 4 d at 16'C. Alleles used
were pmk-i (km25), xbp-i(tm2482), and pek-1(ok275).
Figure 6
pmk-1
n=215 405 230
WT
n=237 441 294
CD,
0
c
0
U-
1.0-1
1.0-
0.5-I
m L4+
m L1-L3
0.5-
] dead
0.0-
0.0-
xbp-1; pmk-1
n=150 336 247
xbp-1
n=189 349 148
1.0-
1.0-
0.5-
0.5-
0.01w-
0.0-
-
0)
0
0
I-
U-
0*
'v,* t
pmk-1; pek-1
n =194 349 143
pek-1
226 448 161
0
c
0
U-
xbp-1; pmk-1; pek-1
n=180 385 147
xbp-1; pek-1
n= 175 365 121
()
a)
1.0-11
]F
1.0-
4-
0
c
0
0.5-n
-1
1
1
0.5-
Cu
LL
0. 0-
-m
Tunicamycin (Rg/ml)
0.0-
-
Tunicamycin ([tg/ml)
DISCUSSION
We have shown that the IRE-1 -XBP- 1 and PEK- 1 pathways function together to
maintain ER homeostasis in C. elegans under physiological conditions. We found that
XBP- 1 deficiency results in marked activation of both IRE-I and PEK-1, reflecting
constitutive ER stress. Activation of innate immunity mediated by PMK-1 p38 MAPK
further exacerbated the constitutive ER stress in the xbp-1 mutant. To investigate the
physiological roles of UPR signaling as well as the compensatory activity between
distinct UPR pathways, we examined both the individual and the combined effects of
XBP- 1 and PEK- 1 deficiency in vivo. We found that the xbp-1; pek-1 double mutant
exhibited temperature-sensitive lethality that was independent of developmental stage.
Compared with the xbp-1; pek-1 mutant, the xbp-1; pmk-1; pek-] mutant had moderately
increased survival during larval development on non-pathogenic bacteria, when there is a
low level of PMK- 1-mediated basal immune activity, and dramatically increased survival
on pathogenic P. aeruginosa,when the PMK- 1-mediated immune response is induced.
We further showed that both XBP- 1 and PEK- 1 are required for full protection against
the combined stress of immune activation and that of growth at elevated physiological
temperatures, confirming that these two branches of the UPR function together to protect
against physiological ER stress.
Our observation of dramatically elevated levels of IRE-I and PEK- 1 activity in
the setting of XBP- 1 deficiency, under standard growth conditions in the absence of
exogenous agents to induce ER stress, provides strong evidence for homeostatic activity
of the IRE-1 -XBP- 1 signaling pathway under laboratory conditions that are presumably
physiological (Figure 7a), and not merely at the extremes of ER stress induced by
pharmacological treatment or in specialized secretory cell types. Our data also reveal a
dynamic requirement for UPR signaling in survival that increases with both temperature
and increased secretory activity as is induced by immune activation (Figure 7b).
Interestingly, the temperature dependent role for the IRE- 1 and PEK- 1 pathways is
manifest at physiological temperatures optimal for C. elegans development and
fecundity, far from commonly utilized "heat shock" conditions (Figure 7a). We speculate
that this temperature dependence may be due to altered secretory load at higher
temperature or increased tendency for proteins to aggregate in the ER in the absence of
intact chaperone production.
Importantly, our data suggest that immune activation is not necessarily the
primary source of the requirement for the UPR during larval development in the absence
of infection. We found that, although loss of basal PMK- 1 pathway activation partially
suppressed the temperature-sensitive larval lethality of the xbp-1; pek-] mutant, the xbp]; pmk-1; pek-1
mutant nevertheless exhibited almost complete larval lethality at 25'C.
Further, using our smg-2(qd101); xbp-1(zc12) strains, we observed high constitutive IRE1-mediated xbp-1 splicing in the xbp-1; pmk-I mutant that was not significantly different
under these experimental conditions from that of the xbp-1 mutant (Figure 3a). These
results indicate that the UPR has an essential role during development in protection
against processes other than immune activation. Identification of these processes will
likely lead to increased understanding of conserved physiological roles of the UPR.
Figure 7. Maintenance of ER homeostasis through activation of the IRE-1 and PEK1 pathways under basal physiological conditions during development. (a) The
increase in both IRE-I and PEK- 1 activities in XBP- 1 deficiency in the absence of
exogenous compounds to impose ER stress, combined with the temperature-sensitive
lethality of the UPR mutants, implies that UPR signaling maintains ER homeostasis not
only in response to the extremes of ER stress, but also under basal physiological
conditions. (b) Infection, basal growth and development, and elevated physiological
temperature all contribute to ER stress, leading to lethality of UPR mutants as indicated
by dashed lines.
Figure 7
a
Basal physiology
immurity
Exogenious
IRE-1PEK-1
XBP-1
-
Infecbon
-
Basal Phvsiokny
-pek-
Iethal
- xbo-1Iethal
---
-
- xbp-pek-1
lethal
2T
25
Temperature (*C)
In mice, Xbp] deficiency in intestinal epithelial cells (IEC) resulted in marked
intestinal inflammation that may contribute to the observed activation of not only IRE 1
but also of PERK, as measured by expression of one of its downstream effectors, CHOP
(Kaser et al. 2008). In mammals, the transcription factor CHOP promotes apoptosis of
mammalian cells that experience prolonged ER stress (Zinszner et al. 1998), and indeed,
the majority of Paneth cells underwent apoptosis in the XBPJ1 IECs. Our observations
are consistent with the idea that XBPF IECs may be predisposed to detrimental
consequences of additional ER stress caused by intestinal inflammation because of
disregulation of basal ER homeostasis due to XBP- 1 deficiency.
Our observations that PEK- 1, in concert with XBP- 1, function to protect against
ER stress from immune activation differ from observations in mouse macrophages, in
which TLR stimulation was shown to activate IRE 1, but PERK activation was reported to
be suppressed rather than elevated (Woo et al. 2009; Martinon et al. 2010). This
difference may be because of roles for XBP-1 in macrophages that extend beyond its
function in maintaining ER homeostasis. Indeed, when stimulated by TLRs in
macrophages, the IRE -XBP I pathway was shown to induce expression of immune
effectors rather than typical UPR genes, suggestive that the IRE 1-XBP 1 pathway may
have been co-opted in macrophages to promote macrophage-specific function
independent of the UPR (Martinon et al. 2010).
Our data support the idea that UPR signaling does not function simply in response
to the extremes of ER stress, as when induced by tunicamycin or by the elevated
secretory load of specialized cells such as plasma cells, but instead, as a critical pathway
in the maintenance of ER homeostasis during normal growth and development in C.
elegans. The diverse and dramatic consequences of XBP-1 deficiency on development
and disease, taken together with our observations on the effect of XBP- 1 deficiency on
basal ER stress levels, underscore the critical role of ER homeostasis and homeostatic
UPR signaling in both normal physiology and disease.
MATERIALS AND METHODS
Strains
C. elegans strains were constructed and propagated according to standard
methods on E. coli OP50 at 16'C (Brenner 1974). The smg-2(qd10J) allele was isolated
by K. Reddy and contains a C->T nonsense mutation at nucleotide 1189 of the spliced
transcript. The following strains were used in the study: N2 Bristol, ZD627 smg2(qd101), ZD607 smg-2(qd101); xbp-1 (zc12), ZD605 smg-2(qd101); xbp-1(zc12); pmk](km25), KU25pmk-1(km25), RB545 pek-1(ok275), ZD5l xbp-1(tm2482);pek1(ok275), ZD524 xbp-1(zc12); pek-1(tm629), ZD496 xbp-1(tm2482); pmk-](km25); pek1(ok2 75). All of the alleles used are predicted to be null alleles regardless of temperature.
Specifically, xbp-1(tm2482) is a 202 bp deletion from nt 231 that causes a frame-shift.
The xbp-1(zc12) allele, a nonsense mutation that changes Q34 to an ochre stop. The two
alleles exhibit an equivalent phenotype in every assay tested (Chapter 2, this Chapter, and
data not shown). The pek- 1(ok2 75) allele is a 2013 bp deletion and the pek-] (tim 629)
allele is a 1473 bp deletion, both of which remove the PEK- 1 transmembrane domain and
are therefore likely null alleles (Shen et al., 2001). Double mutants were made between
xbp-1 and pek-] by crossing strains marked with GFP: xbp-1(III);
pT24B8.5.::GFP(agJs220)(X)and p T28B8.4:: GFP(agIs219)(III); pek-1(X). GFPnegative F2s were singled, propagated, and genotyped by PCR.
RNA isolation and quantitative RT-PCR
For the experiment in Figures lb, 1c, and le, LI larvae were synchronized by
hypochlorite treatment, washed onto E. coli OP50 plates, and grown for 40 h at 16'C to
the L3 stage, when they were washed in M9 to plates containing E. coli OP50 or E. coli
OP50 with 5 [Ig/ml tunicamycin. For the experiments in Figure 3, strains were grown and
treated as described in Chapter Two. Specifically, LI larvae were synchronized by
hypochlorite treatment, washed onto E. coli OP50 plates and grown at 20'C for 23 h,
then washed in M9 onto treatment plates. After incubation at 25C for indicated times,
worms were washed off plates and frozen in liquid nitrogen. For P. aeruginosatreatment,
P. aeruginosastrain PA14 was grown in Luria Broth (LB), and 25 tl overnight culture
was seeded onto 10 cm NGM plates. Plates were incubated first at 37 C for 1 d, then at
room temperature for 1 d. All RNA extraction, cDNA preparation, qRT-PCR methods
and specific primers to detect xbp-] mRNA were as described in Chapter 2.
Immunoblotting
For all immunoblots, strains were synchronized by hypochlorite treatment and
washed onto E.coli OP50 plates for growth until the L4 stage. For the experiment in
Figure 2a, strains were grown at 20'C and L4 worms were then washed in M9 onto
treatment plates for incubation at 25'C for 4 hours. For the experiment in Figure 2b,
strains were grown at 16'C until the L4 stage and harvested without treatment. All strains
were collected and rinsed 2 times in M9. Worm pellets were resuspended in an equal
volume of 2x lysis buffer containing 4% SDS, loomM Tris Cl, pH 6.8, and 20%
Glycerol. After boiling for 15 minutes with occasional vortexing to aid in dissolution,
lysates were clarified by centrifugation. Protein samples (50ug of total lysate loaded per
lane) were separated by SDS-PAGE and transferred to a nitrocellulose membrane (Biorad). Western blots were blocked in 5% milk in PBST and probed with (1:10,000) anti-
eIF2a (Nukazuka et al. 2008), (1:1,000) anti-phospho-eIF2a (Cell Signaling
Technology), or (1:10,000) anti-tubulin (E7 Developmental Hybridoma Bank, Iowa
City). All primary antibodies were diluted in 5% milk in PBST. Following incubation
with anti-rabbit or anti-mouse IgG antibodies conjugated with horseradish peroxidase
(HRP) (Cell Signaling Technology), signals were visualized with chemiluminescent HRP
substrate (Amersham).
Survival/development assays
For all development assays, strains were egg laid on 4-5 prepared plates for no
more than 3 h (at least 110 eggs for each strain and treatment). Development was
monitored daily for 4 d for experiments conducted at 16'C and 3 d for experiments
conducted at all other temperatures. Experiments monitoring development on E. coli
OP50 were performed on 6 cm NGM plates. P. aeruginosaPA 14 plates were prepared as
described (Kim et al. 2002). For development in the presence of tunicamycin,
tunicamycin was added to plates seeded with OP50 1 d before use.
ACKNOWLEDGMENTS
We thank the CaenorhabditisGenetics Center, the C. elegans Gene Knockout
Consortium, and S. Mitani and the National Bioresource Project of Japan for providing
strains. This work was supported by NIH grant RO 1-GM084477 (to D.H.K).
REFERENCES
Bischof, L. J., Kao, C.-Y., Los, F. C. 0., Gonzalez, M. R., Shen, Z., Briggs, S. P., Goot,
F. G. van der, and Aroian, R. V. (2008). Activation of the Unfolded Protein
Response is required for defenses against bacterial pore-forming toxin in vivo. PLOS
pathogens 4, e100176. doi: 10.1371/journal.ppat.1000176.
Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 71-94.
Calfon, M., Zeng, H. Q., Urano, F., Till, J. H., Hubbard, S. R., Harding, H. P., Clark, S.
G., and Ron, D. (2002). IRE1 couples endoplasmic reticulum load to secretory
capacity by processing the XBP-1 mRNA. Nature 415, 92-96.
Gardner, L. B. (2008). Hypoxic inhibition of nonsense-mediated RNA decay regulates
gene expression and the integrated stress response. Molecular and Cellular Biology
28, 3729-3741.
Hamanaka, R. B., Bennett, B. S., Cullinan, S. B., and Diehl, J. A. (2005). PERK and
GCN2 contribute to eIF2alpha phosphorylation and cell cycle arrest after activation
of the unfolded protein response pathway. Molecular Biology of the Cell 16, 54935501.
Harding, H. P., Novoa, I., Zhang, Y., Zeng, H., Wek, R., Schapira, M., and Ron, D.
(2000b). Regulated translation initiation controls stress-induced gene expression in
mammalian cells. Molecular Cell 6, 1099-1108.
Harding, H. P., Zhang, Y., Bertolotti, A., Zeng, H., and Ron, D. (2000a). PERK is
essential for translational regulation and cell survival during the unfolded protein
response. Molecular Cell 5, 897-904.
Hu, C.-C. A., Dougan, S. K., McGehee, A. M., Love, J. C., and Ploegh, H. L. (2009).
XBP- 1 regulates signal transduction, transcription factors and bone marrow
colonization in B cells. EMBO Journal 28, 1624-1636.
Inoue, H., Hisamoto, N., An, J. H., Oliveira, R. P., Nishida, E., Blackwell, T. K., and
Matsumoto, K. (2005). The C. elegans p38 MAPK pathway regulates nuclear
localization of the transcription factor SKN-1 in oxidative stress response. Genes
and Development 18, 2278-2283.
Kaser, A. et al. (2008). XBP1 links ER stress to intestinal inflammation and confers
genetic risk for human inflammatory bowel disease. Cell 134, 743-756.
Kim, D. H. et al. (2002). A conserved p38 MAP kinase pathway in Caenorhabditis
elegans innate immunity. Science 297, 623-626.
Kurz, C. L., Shapira, M., Chen, K., Baillie, D. L., and Tan, M.-W. (2007).
Caenorhabditiselegans pgp-5 is involved in resistance to bacterial infection and
heavy metal and its regulation requires TIR-1 and p38 map kinase cascade.
Biochemical and Biophysical Research Communications 363, 438-443.
Lee, A.-H., Scapa, E. F., Cohen, D. E., and Glimcher, L. H. (2008). Regulation of hepatic
lipogenesis by the transcription factor XBPl. Science 320, 1492-1496.
Martinon, F., Chen, X., Lee, A.-H., and Glimcher, L. H. (2010). TLR activation of the
transcription factor XBP1 regulates innate immune responses in macrophages.
Nature Immunology 11, 411-418.
Mori, Kazutoshi (2009). Signaling pathways in the unfolded protein response:
development from yeast to mammals. Journal of Biochemistry 146, 743-750.
Nukazuka, A., Fujisawa, H., Inada, T., Oda, Y., and Takagi, S. (2008). Semaphorin
controls epidermal morphogenesis by stimulating mRNA translation via eIF2U in
Caenorhabditiselegans. Genes and Development 22, 1025-1036.
Page, M. F., Carr, B., Anders, K. R., Grimson, A., and Anderson, P. (1999). SMG-2 is a
phosphorylated protein required for mRNA surveillance in Caenorhabditiselegans
and related to Upflp of yeast. Molecular and Cellular Biology 19, 5943-595 1.
Rehwinkel, J., Raes, J., and Izaurralde, E. (2006). Nonsense-mediated mRNA decay:
target genes and functional diversification of effectors. Trends in Biochemical
Sciences 31, 639-646.
Reimold, A. M. et al. (2001). Plasma cell differentiation requires the transcription factor
XBP-1. Nature 412, 300-307.
Richardson, C. E., Kooistra, T., and Kim, D. H. (2010). An essential role for XBP-1 in
host protection against immune activation in C. elegans. Nature 463, 1092-1095.
Ron, D., and Walter, P. (2007). Signal integration in the endoplasmic reticulum unfolded
protein response. Nature Reviews Molecular Cell Biology 8, 519-529.
Rutkowski, D. T., and Hegde, R. S. (2010). Regulation of basal and cellular physiology
by the homeostatic unfolded protein response. Journal of Cell Biology 189, 783-794.
Schr6der, M., and Kaufman, R. J. (2005). The mammalian unfolded protein response.
Annual Reviews in Biochemistry 74, 739-789.
Shaffer, A. L. et al. (2004). XBP1, downstream of Blimp-1, expands the secretory
apparatus and other organelles, and increases protein synthesis in plasma cell
differentiation. Immunity 21, 81-93.
Shamu, C. E., and Walter, P. (1996). Oligomerization and phosphorylation of the IreIp
kinase during intracellular signaling from the endoplasmic reticulum to the nucleus.
EMBO Journal 15, 3028-3029.
Shen, X. et al. (2001). Complementary signaling pathways regulate the unfolded protein
response and are required for C. elegans development. Cell 107, 893-903.
Shen, X., Ellis, R. E., Sakaki, K., and Kaufman, R. J. (2005). Genetic interactions due to
constitutive and inducible gene regulation mediated by the unfolded protein
response in C. elegans. PLOS genetics 1, e37. doi:10.1371/joumal.pgen.0010037.
Todd, D. J., Lee, A.-H., and Glimcher, L. H. (2008). The endoplasmic reticulum stress
response in immunity and autoimmunity. Nature Reviews Immunology 8, 663-674.
Troemel, E. R., Chu, S. W., Reinke, V., Lee, S. S., Ausubel, F. M., and Kim, D. H.
(2006). p3 8 MAPK regulates expression of immune response genes and contributes
to longevity in C. elegans. PLOS genetics 2, 1725-1739.
Woo, C. W., Cui, D., Arellano, J., Dorweiler, B., Harding, H. P., Fitzgerald, K. A., Ron,
D., and Tabas, I. (2009). Adaptive suppression of the ATF4-CHOP branch of the
unfolded protein response by toll-like receptor signalling. Nature Cell Biology 11,
1473-1480.
Yoshida, H., Matsui, T., Yamamoto, A., Okada, Tetsuya, and Mori, Kazutoshi (2001).
XBP1 mRNA is induced by ATF6 and spliced by IREl in response to ER stress to
produce a highly active transcription factor. Cell 107, 881-891.
Yoshida, H., Okada, T, Haze, K., Yanagi, H., Yura, T., Negishi, M., and Mori, K (2000).
ATF6 activated by proteolysis binds in the presence of NF-Y (CBF) directly to the
cis-acting element responsible for the mammalian unfolded protein response.
Molecular and Cellular Biology 20, 6755-6767.
Zhang, W., Feng, D., Li, Y., Lida, K., McGrath, B., and Cavener, D. R. (2006). PERK
EIF2AK3 control of pancreatic beta cell differentiation and proliferation is required
for postnatal glucose homeostasis. Cell Metabolism 4, 491-497.
Zinszner, H., Kuroda, M., Wang, X., Batchvarova, N., Lightfoot, R. T., Remotti, H.,
Stevens, J. L., and Ron, D. (1998). CHOP is implicated in programmed cell death in
response to impaired function of the endoplasmic reticulum. Genes and
Development 12, 982-995.
Chapter Four
A genetic screen for factors contributing to lethality in an
XBP-1-deficient strain during immune activation
Claire E. Richardson, Kirthi C. Reddy, Dennis H. Kim
K.C.R. performed the crosses between N2 xbp-1 and HW xbp-1 with and without
suppressor mutants.
160
SUMMARY
A requirement for the UPR has been demonstrated under a variety of conditions
in Saccharomyces cerevisiae, Caenorhabditiselegans and mammals, but the mechanisms
underlying that requirement in each case are incompletely understood. Further
investigations into the upstream activators and downstream effectors of the UPR are
essential for understanding its complex physiological activities. In Chapter Two, we
showed that activation of the innate immune response in C. elegans is a source of ER
stress, and that the IRE-1 -XBP- 1 pathway is required for protection against immune
activation. In the work described in this chapter, we set out to identify the mechanisms
that lead to death during immune-induced ER stress in the absence of XBP- I-mediated
protection. To this end, we performed a forward genetic screen for mutants that
suppressed the xbp-] lethality during larval development on Pseudomonas aeruginosa,
from which we isolated 24 suppressors. As none of the causative mutations have yet been
identified, we will first discuss the types of mutants we might expect to isolate, and then
we will describe the progress of and future directions for the screen.
162
INTRODUCTION
In Chapter Two, we described a requirement for the IRE-1 -XBP- 1 branch of the
UPR in C. elegans in protection against immune activation. We showed that xbp-1 lossof-function mutants exhibit larval lethality in the presence of the pathogenic bacteria
Pseudomonas aeruginosa(strain PA 14), and that activation of the PMK-1 MAP kinase,
the central regulator of the immune response to P. aeruginosa(Kim et al., 2002), is both
necessary and sufficient to effect this lethality. Depletion of ATF-7, a bZIP transcription
factor that mediates immune activation through phosphorylation by PMK- 1 (Shivers et
al., 2010), likewise suppresses the xbp-I larval lethality in the presence of P. aeruginosa.
This indicates that the increased transcription of one or more immune effectors is the
source of ER stress, leading us to hypothesize that the production and secretion of
immune effectors causes ER stress by increasing the cumulative ER load.
In this chapter, we describe a genetic screen to identify additional suppressors of
the pathogen-induced xbp-1 larval lethality. We conducted this screen by growing the F2s
from an EMS mutagenized population of xbp-1 worms on P. aeruginosaPA14 and
isolating the individuals that developed to the L4 stage. Having retested the progeny of
these isolates in population based assays for suppression of the xbp-] pathogen-induced
larval lethality, we are currently working toward identifying the mutations that cause the
suppression. Aside from isolating mutants that cause abolished or depleted function of
the PMK-1 pathway, we anticipate that the screen isolates will suppress the xbp-1
lethality through one of two general mechanisms: either by reducing ER stress upstream
of UPR activation or by protecting against ER stress-induced lethality in parallel to or
downstream of xbp-1 (Figure 1).
Figure 1. Model for how mutations isolated from screen might suppress larval
lethality in the xbp-1 mutant growing on P. aeruginosa. Suppression of the xbp-1
larval lethality could be effected by reduction-of-function of genes that cause ER stress
upstream of UPR activation (A), negatively regulate PEK- 1 or ATF-6 (B), affect ER
homeostasis independent of the UPR (C), or inhibit lethality downstream of ER stress
(D). Of note, the screen was designed such that mutations that result in gain-of-function
could also be isolated (not diagrammed).
FIgure i
IIEl
Larv_
iottiality
-
PEKi1 ATP-6
E
Dtsrupted Eft
homesoM
Incraftd
ER load
DI
C
A
immnunity
Suppressor mutations that decrease ER stress upstream of UPR activation will
likely cause a reduction in ER load (A in Figure 1). Such mutants might disrupt either the
function of other immune pathways that are activated in parallel to PMK- 1 or the
production of secretory proteins that are made in the intestine independent of the immune
response, such as digestive enzymes or membrane proteins involved in nutrient uptake.
Although we expect such mutations would disrupt a signaling pathway which regulates
production of multiple secretory proteins, it is possible that we will isolate an allele that
disrupts production of a single secretory protein that has an exceptionally high impact on
ER function.
Few physiological upstream activators of ER stress have been established.
Identifying such upstream activators of the UPR would be significant because, although
the IRE- 1-XBP- 1 pathway is known to be essential for development of numerous
metazoan cell types, the cellular functions that it protects against are widely unknown. In
C. elegans, other known inducers of the UPR include elevated temperature and hypoxia
(Anderson et al., 2009; Galbadage & Hartman, 2008). Mutations that cause reduced
global translation, such as a reduction-of-function allele of an arginyl-tRNA synthetase,
diminish hypoxia-induced UPR activation (Anderson et al., 2009). Since lowering overall
translation is also likely to reduce UPR activation during immune activation, this screen
may isolate additional mutations that suppress xbp-1 larval lethality on pathogen through
this mechanism.
The second class of mutants, wherein the pathogen-induced lethality of the xbp-1
mutant would be suppressed through a process that functions in parallel to or downstream
of XBP- 1, would also be of interest. One mechanism by which this could occur is if the
suppressor mutation inhibited the function of proteins that normally negatively regulate
the activity of PEK- 1 or ATF-6 (B in Figure 1). For example, in mice, an eventual
downstream effect of PERK activation is increased production of GADD34/PP1R15A,
which recruits a catalytic subunit of protein phosphatase 1 (PP 1c) to dephosphorylate
eIF2a, counteracting the effect of PERK to restore global translation.
Alternatively, suppressors in the second class might affect ER homeostasis
through a mechanism independent of PEK-1 or ATF-6 (C in Figure 1). For example, in S.
cerevisiae, exposure to exogenous agents that cause ER stress leads to ER expansion, and
HAC] mutants exhibit diminished ER expansion (Schuck et al., 2009). This requirement
for the UPR in ER stress-induced ER expansion can be circumvented, however, through
activation of the Ino2/4 transcription factor complex, which activates transcription of
genes involved in lipid biosynthesis. The Ino2/4 complex is activated by deletion of
OPI1, which encodes a negative regulator of Ino2p, and deletion of OPI] partially
suppresses the sensitivity of HAC] mutants to ER stress. These results suggest that ER
membrane expansion itself reduces ER stress independently of the UPR, presumably by
reducing the concentration of proteins in the ER (Schuck et al., 2009). Also in S.
cerevisiae, constitutive activation of Hsflp suppresses the sensitivity of an IRE] deletion
strain to treatment with tunicamycin or expression of a misfolding vacuolar enzyme,
carboxypeptidase (CPY*) (Liu & Chang, 2008). Hsflp may protect against ER stressinduced lethality through increasing resistance in the cytoplasm to ROS, promoting
ERAD, or directly upregulating production of a subset of UPR effectors. (Hahn et al.,
2004; Harding et al., 2003; Kohno et al., 1993; Liu & Chang, 2008).
In C. elegans, Marciniak et al. found that RNAi knockdown of ero-1 results in
enhanced survival in the presence of tunicamycin (Marciniak et al., 2004). This result is
surprising, as the IRE- 1-XBP- 1 pathway upregulates production of ERO- 1 upon
tunicamycin-induced ER stress (Shen et al., 2005). However, by promoting oxidative
protein folding, ERO- 1 also produces oxidative stress with potentially damaging effects.
Expression of ero-1 is reduced but not abrogated in the xbp-1 mutant; perhaps further
depletion of ero-1 may improve survival of the xbp-1 mutant during immune activation.
In all of the previous examples, the suppressor mutations would reduce ER stress
in the xbp-1 strain during development on P. aeruginosa.It may also be possible,
however, to improve survival through a process that occurs downstream of ER stress such
that ER stress would not be affected (D in Figure 1). We considered this possibility in
Chapter Two, when we hypothesized that, as the intestinal cells of the xbp-1 mutant
developing on P. aeruginosatake on an appearance consistent with necrosis, inhibiting
necrotic cell death might improve survival. Although we did not observe suppression of
the xbp-1 pathogen-induced lethality through RNAi against genes involved in necrosis,
we may isolate suppressor mutants with this screen that function through an analogous
mechanism to prevent larval lethality.
RESULTS
To identify genes involved in the xbp-] larval lethality on P. aeruginosaPA14,
we performed EMS mutagenesis in a strain with the xbp-1(tm2482) deletion allele and
screened for mutants that suppressed the xbp-1 lethality during development on P.
aeruginosa (Figure 2). To make mutations that inhibit activation of the PMK- 1-mediated
immune response immediately identifiable, our starting strain also carried the qdIs8
transgene, in which a promoter responsive to PMK- 1 activation is fused to GFP
(pT24B8.5.:GFP). This transgene exhibits a basal level of intestinal expression in C.
elegans during growth on non-pathogenic bacteria, and exposure to P. aeruginosa causes
increased intestinal reporter expression.
xbp-1(tm2482);qdIs8 worms were subjected to EMS mutagenesis, and their F2
progeny were screened for the ability to progress through larval development in the
presence of P. aeruginosa.We screened 10,000 haploid genomes and isolated 208
mutants from 24 independent pools, which we transferred to non-pathogenic bacteria to
recover. The 139 mutants which produced progeny were next screened in quantitative
population-based assays for suppression of xbp-1 lethality during development on P.
aeruginosa.Mutants that passed this second round of screening were subjected to a
similar third round, and they were also checked for fluorescence of the qdls8 reporter.
Finally, 24 isolates from 15 independent pools were identified that consistently improved
larval development of the xbp-] mutant on P. aeruginosa,and the strength of each was
annotated based on the quantitation from the second and third round of screening (Table
1).
Figure 2. Screen for mutations that suppress the pathogen-induced larval lethality
of the xbp-1 mutant.
Figure 2
EMS
xbp-1(672482);
pT24B&5:FPiGqdtsM)
Gravid F1 hermaphrodtles
Grow F2 eggs on
Paeruglhosa PA14, 25'C
Select for survival through larval developret
Low Iluorescence IndIcates loss of PMK-1 pathway
In two of the mutants, the basal expression of the qdIs8 reporter was diminished,
as was its induction by pathogen exposure, suggestive that the xbp-1 suppression
observed in each of these is caused by a mutation that abrogates PMK- 1 pathway
function. Expression of the qdIs8 reporter was unaffected in each of the remaining 22
mutants, indicating the mutation(s) responsible for the suppression do not affect the
PMK- 1 pathway.
Six mutants were selected for further characterization. Two of these were found to
be dominant, and four were recessive (Table 1). Complementation testing between the
four recessive mutants indicated that they make up three complementation groups (Figure
3).
To identify the causative mutations, we attempted to use single nucleotide
polymorphism (SNP)-based mapping using the CB4856 Hawaiian strain (Davis et al.,
2005; Wicks et al., 2001). This approach was unsuccessful, however, due to a genetic
interaction between N2 and CB4856 that was unrelated to the suppressor mutations but
dependent upon loss of IRE- 1-XBP- 1 signaling. Therefore, we will use alternative
methods to identify the suppressor mutations, and we will also further characterize the
nature of this interaction between N2 and CB4856.
Table 1: Screen isolates
Allele
name
Screen
isolate
(pool.isolate)
Strength of
suppression
qd202
qd222
1.3
2.6
++
+++
qd217
qd201
qd206
qd212
qd204
qd203
qd205
4.1
6.2
7.3
8.1
8.4
8.6
8.7
+
+
++
+
+++
+++
+*
qd2l1
10.2
++
qd216
Comments
dominant
GFP dark
GFP dark
recessive
10.3
+++
qd199
11.1
+++
qd214
11.13
+++
qd200
qd215
14.1
14.7
+++
++
qd213
qd198
17.5
18.1
+
+
recessive
qd218
qd197
18.4
19.4
++
+++
recessive
qd209
21.5
+
qd219
22.4
++
qd207
24.1
+
qd208
24.3
++
qd210
24.5
+
dominant
recessive
*High variability in suppression within population,
indicative of incomplete penetrance or a heterozygous
dominant mutation.
Figure 3. Complementation between six mutants isolated from the screen.
Complementation testing was performed by mating xbp-1(tm2482); qdls8 with screen
isolates to generate xbp-1(tm2482); qdls8; suppressormutation/+ F 1 males. These were
then mated with hermaphrodites of other screen isolates, and the progeny were grown on
P. aeruginosaPA14. Unless the causative mutation was on the X chromosome, the cross
progeny between two screen isolates with strong suppression phenotypes would exhibit
50% larval lethality if the mutations failed to complement and 100% larval lethality if
they complemented. None of the causative mutations were on the X chromosome, which
would be indicated by either 100% larval development (if the two strains failed to
complement) or 100% larval development of exclusively the male F Is (if the two strains
complemented and the suppressor mutation in the hermaphrodite was on the X
chromosome).
Figure 3
Cl
qd197
qd198
Ad200
Qd203
SComplements
qdl98/+
Falls to
qd200/+
ca mi ement
DISCUSSION AND FUTURE DIRECTIONS
Through this relatively small genetic screen, we isolated 24 mutants that
suppressed the lethality of an xbp-1 mutant during larval development on P. aeruginosa,
and we are currently working to identify the causative mutations. Because we have
previously shown that immune activation through the PMK-1 pathway is both necessary
and sufficient to cause larval lethality in the xbp-1 mutant, we anticipated that we would
identify mutants that disrupted the PMK- 1 pathway. Indeed, two of the screen isolates
exhibited diminished expression of a fluorescent reporter for PMK- 1 activity, indicating
that the causative mutation is likely a reduction-of-function allele of one of the PMK- 1
pathway genes. This verifies that the screen design could isolate bona fide suppressors of
the xbp-1 mutant phenotype. Efforts are currently underway to identify the causative
mutation for three of the strongest suppressors.
We designed the screen such that we could isolate recessive, loss-of-function
alleles, but of the six mutants we checked for dominant versus recessive inheritance, we
found that two were dominant. Dominant inheritance may result from haploinsufficiency,
but it could also be caused by a gain-of-function mutation. Therefore, while some of our
mutants may disrupt function of proteins that promote ER stress or inhibit UPR function,
others may hyperactivate proteins that protect against ER stress.
We will further characterize the suppressors by ascertaining whether they
alleviate ER stress. In Chapter Three, we showed that an xbp-1 mutant experiences
elevated ER stress under basal growth conditions, and this stress was further exacerbated
by exposure to P. aeruginosaor tunicamycin. We will use the smg-2; xbp-1(zc]2) system
described in Chapter Three to measure the levels of IRE-I-mediated splicing of
nonfunctional xbp-1 mRNA in strains also carrying the suppressors and compare it
against that of the smg-2; xbp-1(zc12) mutant alone. The suppressors could reduce the
level of ER stress basally or exclusively in the context of exposure to pathogen, so we
will distinguish between these two possibilities by comparing xbp-1 splicing in the strains
grown on non-pathogenic bacteria against the strains exposed to P. aeruginosaor
tunicamycin. Further, we will perform western blots against phosphorylated eIF2a to
measure the levels of PERK activation in each of the suppressors.
We will also determine whether the screen isolates can suppress other aspects of
the xbp-] phenotype. It will be particularly interesting to cross the suppressor mutations
into the xbp-1; pek-] double mutants, which exhibit enhanced larval lethality in the
presence of P. aeruginosarelative to xbp-] as well as intrinsic temperature sensitive
lethality at any stage of development (Chapter 3). If the isolates do not suppress the
pathogen-induced larval lethality of the xbp-1; pek-] double mutants, it will indicate that
suppression of the pathogen-induced xbp-] lethality occurs through the PEK-1 pathway.
If the pathogen-induced lethality but not the temperature-sensitive lethality of the xbp-1;
pek-] double mutant is suppressed, it would indicate that either the genetic pathway or
the tissue affected by the suppressor mutation is not relevant in the temperature-sensitive
lethality of the xbp-1; pek-1 double mutant. Intriguingly, although loss of PMK-1
pathway function can partially suppress the temperature sensitive lethality of the xbp-] ;
pek-1 mutant, preliminary evidence suggests that the tissue involved in the lethality
induced by exposure to P. aeruginosa,the intestine, may not be the relevant tissue in the
temperature-sensitive lethality (data not shown). We are interested in investigating the
differences between the immune-induced and temperature-induced lethality in the UPR
mutants, as will be described further in Chapter Five.
MATERIALS AND METHODS
ZD459 xbp-] (tm2482); pT24B8.5..GFP(qdls8)L4 worms were mutagenized with
EMS as described (Jorgensen & Mango, 2002). After overnight recovery, the
mutagenized worms were picked into 24 pools of 20 worms/pool, allowed to egg lay for
8 h, then removed. The approximately 5,000 worms were grown at 20'C for three days.
The F2 eggs were isolated with hypochlorite treatment and dropped onto 6 cm NGM
plates that had been seeded with 25 pl P. aeruinosa PA14 and grown as described (Kim
et al., 2002). After 2 d at 25'C, up to 12 L4s from each pool were transferred from the P.
aeruginosaPA14 plates to a plate seeded with E. coli OP50 and allowed to recover for 1
d, by which time most were gravid. Finally, each mutant was singled to an individual
plate in a drop of 1 part bleach, 1 part 1 M KOH (killing residual P. aeruginosaand the
F2 worm but not the eggs).
Retesting the mutants with population based assays was performed as described in
Chapter Two. Each strain was egg laid onto 2 or 4 plates seeded with P. aeruginosafor
the first and second round of retesting, respectively.
To map the mutations, the xbp-1(tm2482) and pT24B8.5::GFP(qdls8)alleles
were put into the CB4856 genetic background by crossing CB4856 males to ZD459
hermaphrodites nine times. To circumvent the zeel-1,peel-] incompatibility, CB4856
males were crossed to Fls from CB4856 X ZD459 at several rounds. The qdIs8 array was
subsequently removed in a tenth round of crossing with CB4856 to make ZD508 ("xbpI(tm2482) HW").
We attempted to map three strains using single nucleotide polymorphism mapping
with ZD508, but no N2 enrichment was observed at any locus. Instead, we found that a
particular locus was enriched for CB4856 after every mapping cross. It was subsequently
observed that even crossing N2 xbp-1(tm2482) with CB4856 xbp-1(tm2482) or N2 ire1(v33) with CB4856 ire-1(v33) resulted in an enrichment for the CB4856 SNP at this
locus in the absence of selection.
REFERENCES
Anderson, L. L., Mao, X., Scott, B. A., & Crowder, C. M. (2009). Survival from hypoxia
in C. elegans by inactivation of aminoacyl-tRNA synthetases. Science 323, 630-633.
Galbadage, T., & Hartman, P. S. (2008). Repeated temperature fluctuation extends the
life span of Caenorhabditiselegans in a daf-16-dependent fashion. Mechanisms of
Ageing and Development 129, 507-514.
Hahn, J.-S., Hu, Z., Thiele, D. J., & Iyer, V. R. (2004). Genome-wide analysis of the
biology of stress responses through heat shock transcription factor. Molecular and
Cellular Biology 24, 5249-5256.
Harding, H. P., Zhang, Y., Zeng, H., Novoa, I., Lu, P. D., Calfon, M., Sadri, N., et al.
(2003). An integrated stress response regulates amino acid metabolism and
resistance to oxidative stress. Molecular Cell 11, 619-633.
Jorgensen, E. M., & Mango, S. E. (2002). The art and design of genetic screens:
Caenorhabditiselegans. Nature Reviews Genetics 3, 356-469.
Kim, D. H., Feinbaum, R., Alloing, G., Emerson, F. E., Garsin, D. A., Inoue, H., TanakaHino, M., et al. (2002). A conserved p38 MAP kinase pathway in Caenorhabditis
elegans innate immunity. Science 297, 623-626.
Kohno, K., Normington, K., Sambrook, J., Gething, M. J., & Mori, K. (1993). The
promoter region of the yeast KAR2 (BiP)gene contains a regulatory domain that
responds to the presence of unfolded proteins in the endoplasmic reticulum.
Molecular and Cellular Biology 13, 877-890.
Liu, Y., & Chang, A. (2008). Heat shock response relieves ER stress. The EMBO Journal
27, 1049-1059.
Marciniak, S. J., Yun, C. Y., Oyadomari, S., Novoa, I., Zhang, Y., Jungreis, R., Nagata,
K., et al. (2004). CHOP induces death by promoting protein synthesis and oxidation
in the stressed endoplasmic reticulum. Genes & Development 18, 3066-3077.
Schuck, S., Prinz, W. A., Thorn, K. S., Voss, C., & Walter, P. (2009). Membrane
expansion alleviates endoplasmic reticulum stress independently of the unfolded
protein response. The Journal of Cell Biology 187, 525-536.
Shen, X., Ellis, R. E., Sakaki, K., & Kaufman, R. J. (2005). Genetic interactions due to
constitutive and inducible gene regulation mediated by the unfolded protein
response in C. elegans. PLoS Genetics 1, e37. doi:10.1371/journal.pgen.0010037.
Shivers, R. P., Pagano, D. J., Kooistra, T., Richardson, C. E., Reddy, K. C., Whitney, J.
K., Kamanzi, 0., et al. (2010). Phosphorylation of the conserved transcription factor
ATF-7 by PMK-1 p38 MAPK regulates innate immunity in Caenorhabditiselegans.
PLoS Genetics, 6, e1000892. doi:10.1371/joumal.pgen.1000892.
Chapter Five
Conclusion and Future Directions
Claire E. Richardson, Stephanie Kinkel, Kirthi Reddy, Douglas J. Cattie,
Dennis H. Kim
S.K. synthesized and injected the xbp-1 tissue specific expression constructs, isolated the
lines, and synthesized the epitope-tagged ATF6 constructs.
D.J.C. performed the screen for mutants that suppress the xbp-]; pek-] temperaturesensitive lethality and the screen for mutants that suppress the lethality of xbp-I with atf6 RNAi with assistance from K.R. and C.E.R.
184
SUMMARY
By sensing the status of ER protein folding and initiating cellular compensatory
responses, the Unfolded Protein Response both adjusts ER capacity to match the dynamic
needs for ER function and integrates with other signaling pathways to affect overall
cellular function. Much has been learned about the molecular mechanisms of UPR
signaling, primarily through biochemical studies in Saccharomyces cerevisiae and
mammalian cell culture. Furthermore, genetic studies have implicated a requirement for
the UPR in multiple aspects of physiology from yeast to mammals, including
differentiation, starvation, hypoxia, and immunity. Still, a number of central questions
regarding UPR function remain incompletely answered. First, the sources of ER stress
under the diverse circumstances in which the UPR has been genetically implicated are
incompletely understood at the molecular level. Second, the functional significance of
each of the downstream components of UPR signaling is unclear and likely varies
between different cell types and conditions. Third, the metazoan UPR is a negative
feedback loop with three parallel branches, which presents an inherent challenge in
interpreting genetic studies. Namely, in studies conducted by depleting individual
branches of the UPR, it has been difficult to determine which phenotypic consequences
result directly from the depletion and which result from compensatory activation of the
other two branches. The work described in this thesis used the nematode Caenorhabditis
elegans as a simple experimental organism in which to address the sources of ER stress,
the functionally important downstream effectors of the UPR, the interaction between the
UPR branches.
The three branches of the UPR are well conserved between C. elegans and
mammals, and there are several advantages to examining the UPR in C. elegans that
make such studies a useful complement to the work in mammals. First, in mice, genetic
ablation of each UPR branch leads to either embryonic or early postnatal lethality,
whereas in C. elegans, mutants lacking each individual branch are viable and
phenotypically similar to WT under standard growth conditions. These strains can
therefore be studied under a wide variety of environmental conditions to identify the
sources and consequences of ER stress, which can then be useful in thinking about how
UPR depletion causes lethality in mammals. Second, whereas many mammalian studies
probing the function of the UPR have used cell culture, studies in C. elegans address
UPR function in the context of whole-organism physiology. Because of its relatively
simple physiology, C. elegans is a more feasible system than mammals in which to
deconvolute the direct effect of UPR depletion from the secondary and tertiary effects.
Third, because C. elegans is highly genetically tractable, it is amenable to both genetic
screens and hypothesis-driven searches for factors that interact with the UPR.
Work on this thesis began with the hypothesis that the UPR may function in the
immune response in C. elegans. It has led to the identification and description of a
functionally overlapping role for the IRE- 1 and PEK- I branches of the UPR in protection
against immune activation, thermal stress, and constitutive stress during growth under
standard laboratory conditions.
We found that activation of the PMK- 1 MAP kinase-mediated immune response
is necessary and sufficient to induce larval lethality in the absence of XBP-1, and we
subsequently demonstrated that PEK- 1 was likewise required for survival during immune
activation. These results suggest that immune activation, which involves increased
production of secreted immune effectors, represents a source of ER stress that
necessitates UPR function. We speculate that the UPR may play a similar role in the
mammalian intestinal epithelium, where XBP] deletion results in loss of Paneth and
goblet cells and the development of inflammatory bowel disease (IBD), and in tissues
subject to chronic inflammation during metabolic disease (Kaser et al., 2008; Ozcan et
al., 2004). Given this role for the UPR in C. elegans, it is of particular significance to
note that the intestinal microbiota influences the development of IBD in both mice and
humans (Sartor, 2008), and it will be of interest to determine if host-microbe interactions
modulate ER stress in the intestinal epithelium.
In mouse macrophages, activation of the innate immune response likewise causes
activation of the IRE-1 branch of the UPR (Martinon et al., 2010), but the role of IREl
activation appears to be distinct from that in C. elegans: In C. elegans, activation of XBP1 in response to infection with the bacterial pathogen Pseudomonasaeruginosaleads to
increased mRNA levels of known UPR effectors without affecting transcription of
immune effectors, and loss of XBP- 1 does not influence susceptibility to P. aeruginosa.
In contrast, TLR stimulation in macrophages appears to dampen the production of UPR
effectors through the IRE 1 and PERK pathways, with XBP 1 directly promoting
transcription of genes involved in immunity rather than the UPR (Martinon et al., 2010;
Woo et al., 2009). Perhaps in these more specialize immune cells, there are mechanisms
that modulate UPR activity downstream of the three UPR sensors to promote macrophage
function and inhibit certain features of UPR activation, such as inhibition of protein
synthesis, that would limit immune activation.
We also found that xbp-1 loss-of-function mutants experience constitutive
elevated ER stress, resulting in increased PEK- 1 activation. By examining the synthetic
sensitivity of a xbp-]; pek-1 double mutant to lethality at elevated physiological
temperatures, we show that this PEK- 1 activation is functionally important in protecting
C. elegans against the constitutive ER stress manifested under standard growth
conditions. Similarly, in mammals, activation of the PERK and ATF6 pathways has been
observed in XBP1-/- mutants (Kaser et al., 2008; Ozcan et al., 2004; Zhang et al., 2011).
Our finding in C. elegans suggest that hyperactivation of these other two branches may
substantially contribute to the XBP1-/- phenotype. Because sustained activation of
mammalian PERK promotes apoptosis, the effect of this compensatory activation could
in fact be deleterious rather than protective for tissue survival.
In summary, we have identified a number of situations in which UPR deficiency
leads to larval lethality in C. elegans, shown in Table 1. Future work guided by the
findings presented here should seek to more fully elucidate the mechanisms that regulate
the requirement for and function of the UPR in C. elegans physiology. The following
sections will highlight several of the immediate directions such research might pursue.
Table 1. Temperature of larval lethality in
UPR mutants with varying immune function
Strain
Level of immune function
basal
activatedo
none*
0
0
xbp-1
> 27 C
27 C
250 C
pek-]
> 27 0 C > 27 0 C
270 C
0
0
atf-6
> 27 C > 27 C
> 27 0 C
23'C
160 C
25 0 C
xbp-];pek-]
xbp-];atf-6
16'C
16'C
ND
0
0
pek-];atf-6
> 27 C > 27 C
> 27 0 C
ND - not determined
* pmk-1(lf) mutant
* growth on Escherichiacoli OP50
growth on Pseudomonas aerugonisaPA14
FUTURE DIRECTIONS
Observing the effects of UPR depletion on sub-cellular morphology
In Chapter Two, we present electron microscopic images of intestinal cells of WT
and xbp-1 mutant C. elegans grown on P. aeruginosa,in the presence of tunicamycin, or
under standard growth conditions. We observed a striking defect in ER morphology in
the xbp-] mutants exposed to P. aeruginosaor tunicamycin that was characterized by
disorganization of the ER sheets, distention of the ER lumen, and a qualitative reduction
in ER size. The xbp-I mutant grown under standard conditions did not exhibit such gross
defects in ER architecture; however, we did not acquire enough images to determine
whether there were more subtle changes relative to WT. Based on the data presented in
Chapter Three which showed hyperactivation of IRE- 1 and PEK- 1 in an xbp-1 mutant
growing under basal conditions, we might expect to observe some accompanying
morphological defects in the ER. Notably, although we showed in Chapter Three that
PEK-1 activity does partially compensate for loss of xbp-1 function, the fact that both
IRE- 1 and PEK- 1 remain constitutively activated in the xbp-1 mutant indicates that there
is still unresolved ER stress. On the other hand, the ER structure might be resistant to
substantial levels of ER stress, altering only under conditions of sufficient ER stress to
cause lethality. It would therefore be useful to perform a more exhaustive EM analysis of
the xbp-1 mutant, including examination of tissues other than the intestine. Furthermore,
EM analysis of the xbp-1; pek-1 mutant may be informative in comparing how the
presence of PEK- 1 affects the cellular changes that result from ER stress.
In mammalian and yeast cells, changes in ER structure have been observed in a
variety of ER stress conditions. Expression of disease-causing mutant membrane
proteins, such as AF508 CFTR (cystic fibrosis transmembrane conductance regulator,
causes cystic fibrosis) and PMP-22TR-J (peripheral myelin protein 22, associated with
neuropathies), leads to changes in ER architecture. These changes involve the
reorganization of the ER, which is normally a mix of RER (rough ER - associated with
ribosomes) sheets and SER (smooth ER - involved in lipid synthesis) tubules, into
aberrant structures collectively termed OSER (organized smooth ER) (Dickson et al.,
2002; Okiyoneda et al., 2004). OSER can take a number of distinct forms, including long
cisternae of smooth ER, concentric whorls, and stacks of repeating cubic or hexagonal
symmetry (Borgese et al., 2006). Overexpression of specific WT membrane proteins, like
cytochrome P450 (Vergeres et al., 1993) and HMG CoA reductase (Wright et al., 1990)
also induces OSER formation. The formation of OSER is dependent upon UPR induction
in yeast (Schuck et al., 2009) and likely also in animals (Bommiasamy et al., 2009), so
we do not expect to see such structures in the C. elegans UPR mutants. It would be
interesting, though, if immune activation could cause OSER formation in the WT strain.
As C. elegans immune activation leads to increased production of numerous secreted and
membrane-bound proteins, appearance of OSER during immune activation would expand
the model for OSER formation, which generally involves retention and dimerization of
one membrane protein in the ER (Snapp et al., 2003).
Furthermore, we might observe changes in cellular structure apart from the ER.
One possible cause of lethality in the UPR mutants is that disruption of the intestinal cell
secretory pathway inhibits nutrient intake, leading to starvation. If this is the case, we
might observe an increase in autophagosome formation (Levine & Kroemer, 2008).
Interestingly, in mice, expression of mutant SOD 1 in XBP1 -/- neurons, but not neurons
with intact XBP1, caused an increased appearance of autophagosomes through an
unknown mechanism (Hetz & Glimcher, 2009). Alternatively, if the UPR-deficient cells
in our strains are undergoing necrosis, we might observe plasma membrane rupture,
swelling of mitochondria, and/or perinuclear clustering of organelles (Golstein &
Kroemer, 2007). We hypothesize that the C. elegans UPR mutants will exhibit
ultrastructural changes distinct from the WT strain that are indicative of the pathology
leading to lethality.
Examining tissue specific requirements for the UPR
To better understand the mechanisms behind the physiological requirements for
the UPR that we have described, we intend to investigate the cells and tissues in which
the UPR components are required. We therefore fused tissue-specific promoters to the
xbp-1 genomic sequence, and then injected these constructs into the xbp-1(tm2482)
mutant. Our preliminary data suggest that expression of xbp-1 in the intestine is sufficient
to rescue larval development during infection with P. aeruginosa.Similarly, Bischof et
al. showed that expression of xbp-1 exclusively in the intestine was sufficient to rescue
the PMK- I-mediated lethality of the xbp-1 mutant exposed to Bacillus thuringiensistoxin
Cry5B (Bischof et al., 2008). We also introduced the tissue-specific expression
transgenes into the xbp-1; pek-1 double mutant. We will use the resulting strains to
determine if the larval lethality of the xbp-1 mutant originated in the same tissue(s) as the
larval lethality of the xbp-1; pek-1 double mutant during immune activation or at elevated
temperatures (Table 1). Intriguingly, our preliminary results suggest that xbp-1 is
required in different tissues to rescue these four sensitivities. Further work will be
required to fully characterize the tissue specific requirement for xbp-1. If distinct tissues
are indeed involved in xbp-] versus xbp-1; pek-] on P. aeruginosaversus at elevated
temperatures, then distinct molecular mechanisms may be involved as well.
Identifying molecular mechanisms that lead to xbp-1 lethality in the presence of
Pseudomonas aeruginosa
What causes lethality in the xbp-1 mutant developing in the presence of pathogen?
We generated and addressed several hypotheses to answer this question, some of which
are detailed in Chapter Two, but this approach has not yet generated informative results.
For example, because the intestinal cells of xbp-1 mutants developing in the presence of
pathogen had an appearance consistent with necrotic cell death, we assessed whether
RNAi targeted against genes encoding proteins that function to promote the cascade of
events involved in necrosis, which include release of calcium from the ER and
acidification of the cytoplasm, could suppress the pathogen-induced xbp-1 lethality. None
of these knockdown experiments resulted in suppression of the xbp-1 larval lethality, but
this might be due to technical problems, such as if the RNAi knockdown was insufficient
to inhibit the function of the gene products, or confounding biology, such as if the gene
products are also involved in functions that promote survival or ER function (Chapter 2).
We therefore decided to take un unbiased approach by performing a genetic
screen for mutants that suppress the lethality of xbp-1 mutants during development on P.
aeruginosa,and the progress and future directions of this screen are discussed in Chapter
Four. Briefly, this screen might identify pathways that cause ER stress upstream of the
UPR or influence the response to ER stress downstream of or parallel to XBP- 1. In light
of our preliminary results with tissue specific rescue, it will be particularly interesting to
determine whether the suppressors also protect the xbp-1; pek-1 double mutant against
ER stress induced by immune activation or elevated temperatures. If the suppressor
mutant we have identified influences the UPR in diverse cell types and in both the xbp-1
and the xbp-]; pek-1 mutant, it likely has a general function that influences ER
homeostasis independent of the particular UPR mutant background and screen conditions
used to isolate it.
Identifying molecular mechanisms that lead to xbp-1; pek-1 temperature-sensitive
lethality
We have several reasons to suspect that the mechanisms leading to death in the
xbp-1; pek-] mutant grown at elevated physiological temperatures is different from that
of the xbp-1 mutant during development on pathogen. First, whereas the pathogeninduced lethality of xbp-1 or xbp-1; pek-1 is completely suppressed by loss of the PMK-1
pathway, the temperature sensitive lethality of xbp-]; pek-1 is only partially suppressed,
indicating that mechanisms other than immune activation are involved. Second, as
described above, our preliminary results with tissue specific rescue suggest that xbp-1 is
required in different tissues to rescue survival of the xbp-1 mutant on pathogen versus the
xbp-1; pek-1 mutant at elevated temperature. Third, the xbp-1; pek-1 mutant has a
pleiotropic phenotype when grown at 23'C distinct from that of xbp-1 on pathogen. The
xbp-1 mutant on pathogen arrests in the L2 and L3 stages, becomes thin and pale, and
exhibits intestinal morphology consistent with necrotic cell death. It was previously
reported that the xbp-1; pek-1 mutant arrests at the L2 larval stage with intestinal cell
necrosis (Shen et al., 2005), suggesting that the mechanisms leading to death may be
similar. However, although we observed L2 and L3 individuals with morphology
consistent with intestinal necrosis, we also observed a small proportion of individuals
with several other distinct morphologies, indicative of distinct mechanisms in effect that
do not function in pathogen induced xbp-] lethality. We will continue to characterize
these distinct morphologies of the xbp-1; pek-1 mutants during growth at 23'C.
Because of this evidence suggesting that the mechanisms leading to lethality in
the xbp-1; pek-1 mutant at elevated temperature are to some extent distinct from those
that cause lethality in the xbp-1 mutant on pathogen, a screen has been performed to
suppress the temperature sensitive larval lethality of the xbp-]; pek-1 mutant. The screen
design was similar to that of the screen for suppressors of pathogen-induced xbp-1
lethality, and 8 strong suppressors have been isolated. These will be characterized further
in a manner similar to that of the isolates from the screen for suppressors of the xbp-1
pathogen-induced lethality.
Characterizing the role of ATF-6 in basal physiology and protection against
immune activation
In the work presented in this thesis, we focus on the IRE- 1 and PEK- 1 branches of
the UPR, having failed to identify a defect in the atf-6 (lf) mutant during development in
the presence of pathogen. However, we found the xbp-]; atf-6 double mutant to be
inviable due to completely penetrant larval lethality, similar to the phenotype previously
reported (Shen et al., 2005). We therefore infer that, although ATF-6 does not protect
against PMK- 1 activation, it performs an essential role in parallel with XBP- 1 in larval
development.
Curiously, we did not observe increased activation of IRE-I when ATF-6 was
depleted by RNAi (data not shown). To perform the converse experiment, we intend to
develop an assay to monitor ATF-6 activation. To this end, epitope-tagged sequences of
atf-6 have been constructed and will be used to rescue an atf-6 mutant (which exhibits
sensitivity to tunicamycin relative to WT (Bischof et al., 2008)). As ATF-6 activation
involves cleavage of the protein to release the cytosolic transcription factor domain,
activation of epitope-tagged ATF-6 can be observed using western blots to monitor the
proportion of cleaved versus full-length ATF-6. We will monitor ATF-6 proteolytic
cleavage in the strain grown under basal conditions, exposed to either P. aeruginosaor
tunicamycin. We anticipate that exposure to tunicamycin will lead to increased ATF-6
activation but exposure to P. aeruginosamay not. Because we suspect that ATF-6
functions to relieve the PMK- 1-independent basal ER stress in an xbp-1 mutant, however,
we expect to observe a constitutive increase in ATF-6 proteolytic cleavage in the xbp-1
mutant relative to WT, and an even greater increase in the xbp-1; pek-] mutant. An
alternative method to monitor ATF-6 activation would be to measure its transcriptional
output. Using the microarray data collected from WT versus atf-6 mutants from Shen et
al., 2005, we could verify a transcriptional target or transcriptional targets of atf-6 with
quantitative RT-PCR (Shen et al., 2005). This would allow us to develop a fluorescent
transcriptional reporter that we could use to visualize atf-6 activation in vivo.
To better understand the requirement for XBP- I or ATF-6 in larval development,
a genetic approach is also being taken to identify mutants that suppress the lethality when
both pathways are depleted (Figure 1). We performed a genetic selection by
mutagenizing the xbp-I(tm2482) mutant with EMS, then exposing the FIs to RNAi
against atf-6. F2s were also propagated in the presence of atf-6 RNAi, and those that
survived through larval development were isolated. Because survival could be effected
through loss of function of the RNAi pathway, a secondary characterization was
performed to ensure that isolates not only suppressed lethality in the presence of atf-6
RNAi but also exhibited sensitivity to unc-22 RNAi (which causes twitching and
paralysis). Through this secondary screening, it was established that several of the
suppressors were RNAi-defective. There are four strong suppressors that are proficient
for RNAi, and future experiments will be performed to identify the causative mutations.
In mammalian cells, ATF-6 is activated by exogenous agents that induce ER
stress, similar to IRE-I and PEK-1. Likewise, the transcriptional targets of ATF-6 include
such canonical UPR effectors as BiP, CRT, enzymes involved in lipid biosynthesis
(Bommiasamy et al., 2009; Haze et al., 1999; Yamamoto et al., 2007; Yoshida, et al.,
2001). Little is known, however, about the physiological roles of ATF-6. Therefore, our
studies to further define the interaction between ATF6 and XBP- 1 in C. elegans may
provide valuable insight into the function of this well conserved pathway.
Figure 1. Screen for mutations that suppress the larval lethality caused by
simultaneous depletion of xbp-1 and atf-6.
Figure 1
EMS
x p- I h 2482)
F1 hermaphrodites
atf-6 RNAI
Select for survival through larval development
unc-22 RNAI
MO,
Mutant exhibits Unc phenotype mutation affects UPR
Mutant does not exhtft Unc phenotype Mutation r upts RNAl
REFERENCES
Bischof, L. J., Kao, C.-Y., Los, F. C. 0., Gonzalez, M. R., Shen, Z., Briggs, S. P., Goot,
F. G. van der, et al. (2008). Activation of the Unfolded Protein Response is required
for defenses against bacterial pore-forming toxin in vivo. PLoS Pathogens 4,
e100176. doi: 10.1371/journal.ppat.1000176.
Bommiasamy, H., Back, S. H., Fagone, P., Lee, K., Meshinchi, S., Vink, E., Sriburi, R.,
et al. (2009). ATF6alpha induces XBP 1-independent expansion of the endoplasmic
reticulum. Journal of Cell Science 122, 1626-1636.
Borgese, N., Francolini, M., & Snapp, E. (2006). Endoplasmic reticulum architecture:
structures in flux. Current Opinion in Cell Biology 18, 358-364.
Dickson, K. M., Bergeron, J. J. M., Shames, I., Colby, J., Nguyen, D. T., Chevet, E.,
Thomas, D. Y., et al. (2002). Association of calnexin with mutant peripheral myelin
protein-22 ex vivo: a basis for "gain-of-function" ER diseases. Proceedings of the
National Academy of Sciences of the United States of America 99, 9852-9857.
Golstein, P., & Kroemer, G. (2007). Cell death by necrosis: towards a molecular
definition. Trends in Biochemical Sciences 32, 37-43.
Haze, K., Yoshida, H., Yanagi, H., Yura, T., & Mori, Kautoshi. (1999). Mammalian
transcription factor ATF6 is synthesized as a transmembrane protein and activated
by proteolysis in response to endoplasmic reticulum stress. Molecular Biology of the
Cell 10, 3787-3799.
Hetz, C., & Glimcher, L. H. (2009). Fine-tuning of the unfolded protein response:
Assembling the IRElalpha interactome. Molecular Cell 35, 551-561.
Kaser, A., Lee, A.-H., Franke, A., Glickman, J. N., Zeissig, S., Tilg, H., Nieuwenhuis, E.
E. S., et al. (2008). XBPI links ER stress to intestinal inflammation and confers
genetic risk for human inflammatory bowel disease. Cell 134, 743-756.
Levine, B., & Kroemer, G. (2008). Autophagy in the pathogenesis of disease. Cell 132,
27-42.
Martinon, F., Chen, X., Lee, A.-H., & Glimcher, L. H. (2010). TLR activation of the
transcription factor XBP 1 regulates innate immune responses in macrophages.
Nature Immunology 11, 411-418.
Okiyoneda, T., Harada, K., Takeya, M., Yamahira, K., Wada, I., Shuto, T., Suico, M. A.,
et al. (2004). Delta F508 CFTR pool in the endoplasmic reticulum is increased by
calnexin overexpression. Molecular Biology of the Cell 15, 563-574.
Ozcan, U., Cao, Q., Yilmaz, E., Lee, A.-H., Iwakoshi, N. N., Ozdelen, E., Tuncman, G.,
et al. (2004). Endoplasmic reticulum stress links obesity, insulin action, and type 2
diabetes. Science 306, 457-461.
Sartor, R. B. (2008). Microbial influences in inflammatory bowel diseases.
Gastroenterology 134, 577-594.
Schuck, S., Prinz, W. A., Thorn, K. S., Voss, C., & Walter, P. (2009). Membrane
expansion alleviates endoplasmic reticulum stress independently of the unfolded
protein response. The Journal of Cell Biology 187, 525-536.
Shen, X., Ellis, R. E., Sakaki, K., & Kaufman, R. J. (2005). Genetic interactions due to
constitutive and inducible gene regulation mediated by the unfolded protein
response in C. elegans. PLoS Genetics 1, e37. doi:10.1371/joumal.pgen.0010037.
Snapp, E. L., Hegde, R. S., Francolini, M., Lombardo, F., Colombo, S., Pedrazzini, E.,
Borgese, N., et al. (2003). Formation of stacked ER cisternae by low affinity protein
interactions. The Journal of Cell Biology 163, 257-269.
Vergeres, G., Yen, T., Aggeler, J., Lausier, J., & Waskell, L. (1993). A model system for
studying membrane biogenesis. Overexpression of cytochrome b5 in yeast results in
marked proliferation of the intracellular membrane. Journal of Cell Science 106,
249-259.
Woo, C. W., Cui, D., Arellano, J., Dorweiler, B., Harding, H. P., Fitzgerald, K. A., Ron,
D., et al. (2009). Adaptive suppression of the ATF4-CHOP branch of the unfolded
protein response by toll-like receptor signalling. Nature Cell Biology 11, 1473-1480.
Wright, R., Keller, G., Gould, S. J., Subramani, S., & Rine, J. (1990). Cell-type control of
membrane biogenesis induced by HMG-CoA reductase. The New Biologist, 2, 915921.
Yamamoto, K., Sato, T., Matsui, T., Sato, M., Okada, T., Yoshida, H., Harada, A., et al.
(2007). Transcriptional induction of mammalian ER quality control proteins is
mediated by single or combined action of ATF6alpha and XBP 1. Developmental
Cell 13, 365-376.
Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., & Mori, Kazutoshi. (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.
Zhang, K., Wang, S., Malhotra, J., Hassler, J. R., Back, S. H., Wang, G., Chang, L., et al.
(2011). The unfolded protein response transducer IRE 1a prevents ER stress-induced
hepatic steatosis. The EMBO Journal 30, 1357-1375.