Physiol Rev 92: 689 –737, 2012
doi:10.1152/physrev.00028.2011
MAMMALIAN MAPK SIGNAL TRANSDUCTION
PATHWAYS ACTIVATED BY STRESS AND
INFLAMMATION: A 10-YEAR UPDATE
John M. Kyriakis and Joseph Avruch
Molecular Cardiology Research Institute, Department of Medicine, Tufts Medical Center, and Diabetes Research
Laboratory, Medical Services and Department of Molecular Biology, Massachusetts General Hospital, and
Department of Medicine, Harvard Medical School, Boston, Massachusetts
L
I.
II.
III.
IV.
V.
INTRODUCTION
THE ERK, JNK, AND P38 MAPK CORE...
SIGNALING COMPONENTS THAT...
BIOLOGICAL FUNCTIONS OF THE...
CONCLUDING REMARKS
689
690
703
710
728
I. INTRODUCTION
Mitogen-activated protein kinase (MAPK) signal transduction pathways are ubiquitous and highly evolutionarily
conserved mechanisms of eukaryotic cell regulation. The
multiple MAPK pathways present in all eukaryotic cells
enable coordinated and integrated responses to diverse
stimuli. These stimuli include hormones, growth factors,
cytokine, agents acting through G protein-coupled receptors, transforming growth factor (TGF)-␤-related agents
that through Ser-Thr kinase receptors, pathogen-associated
molecular patterns (PAMPs) and danger-associated molecular patterns (DAMPs) that recruit pattern recognition receptors (PRRs), and environmental stresses. MAPK pathways, once activated, exert major effects on cell physiology.
Thus MAPKs orchestrate the recruitment of gene transcription, protein biosynthesis, cell cycle control, apoptosis, and
differentiation (33, 152, 232, 235).
The three-tiered so-called “core signaling module” is a defining feature of MAPK pathways. Within this MAPKs are
phosphorylated and activated via simultaneous Tyr and
Thr phosphorylation within a distinct, and evolutionarily
conserved Thr-X-Tyr motif in the kinase subdomain VIII
activation loop. This phosphorylation is catalyzed by members of the dual-specificity MAPK kinases (MAP2Ks, also
called MEKs or MKKs). The MAP2Ks are activated by
Ser/Thr phosphorylation, again within a conserved motif in
kinase subdomain VIII. This phosphorylation is catalyzed
by a bewilderingly diverse group of protein kinase families
referred to collectively as MAPK kinase kinases (MAP3Ks).
MAPK core signaling modules are themselves regulated by
a poorly understood array of upstream components (33,
152).
MAPKs are proline directed. Thus they phosphorylate Ser/
Thr residues only if they are followed immediately by Pro
residues. It is important to note, however, that proline directedness itself is insufficient to account for the remarkable
substrate selectivity of the different MAPK subgroups. A
notable property of the physiological substrates of the
MAPKs is the presence of specific MAPK docking sites.
Interestingly, these docking sites often reside at considerable distance in the primary sequence from the MAPK phosphoacceptor site. These docking sites permit a strong and
selective interaction between specific MAPK and their bona
fide substrates. Not surprisingly, the docking sites on
MAPK substrates are themselves recognized by comple-
689
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Kyriakis JM, Avruch J. Mammalian MAPK Signal Transduction Pathways Activated by
Stress and Inflammation: A 10-Year Update. Physiol Rev 92: 689 –737, 2012;
doi:10.1152/physrev.00028.2011.—The mammalian stress-activated families of
mitogen-activated protein kinases (MAPKs) were first elucidated in 1994, and by
2001, substantial progress had been made in identifying the architecture of the
pathways upstream of these kinases as well as in cataloguing candidate substrates. This information remains largely sound. Nevertheless, an informed understanding of the physiological and
pathophysiological roles of these kinases remained to be accomplished. In the past decade, there
has been an explosion of new work using RNAi in cells, as well as transgenic, knockout and
conditional knockout technology in mice that has provided valuable insight into the functions of
stress-activated MAPK pathways. These findings have important implications in our understanding
of organ development, innate and acquired immunity, and diseases such as atherosclerosis,
tumorigenesis, and type 2 diabetes. These new developments bring us within striking distance of
the development and validation of novel treatment strategies. Herein we first summarize the
molecular components of the mammalian stress-regulated MAPK pathways and their regulation as
described thus far. We then review some of the in vivo functions of these pathways.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
mentary docking motifs on the cognate MAPKs. This confers
a high degree of substrate specificity that would not be possible
if proline directedness were the sole criterion for MAPK substrate recognition (15, 74, 136, 137, 138, 288, 289, 341).
engage PRRs [including those of the IL-1 receptor/toll-like
receptor (TLR) family]. As is discussed below, these mechanisms of ERK activation play an unexpectedly important role
in innate immunity and inflammation.
In contrast to the high degree of MAPK substrate selectivity,
individual elements within MAPK core signaling modules
frequently function promiscuously in several pathways and
are often subject to regulation by multiple stimuli (TABLE 1).
The ERKs are encoded by two genes, ERK1 and ERK2
(TABLE 1). As with most conventional MAPKs, the ERKs
require dual Tyr and Thr phosphorylation at two closely
spaced residues in the activation loop of subdomain VIII:
Thr185-Glu-Tyr187 (ERK2) or Thr203-Glu-Tyr205 (ERK1)
(229).
II. THE ERK, JNK, AND P38 MAPK CORE
PATHWAYS AND THEIR TARGETS:
ACTIVATION BY ENVIRONMENTAL
STRESSES AND INFLAMMATORY
MEDIATORS
A. The ERKs: Not Just Effectors for
Mitogenic Signaling
Mammals express multiple MAPK pathways. The majority
of these are, along with the nuclear factor-␬B (NF-␬B) pathway, recruited by stress and inflammatory stimuli rather
than by mitogens. Accordingly, these pathways represent a
substantial trove of potentially important targets for novel
anti-inflammatory drugs.
The extracellular signal-regulated kinases (ERKs-1 and -2;
MAPKs-3 and -1, respectively) were the first mammalian
MAPKs to be identified. The ERKs are most familiar as insulin
and mitogen-activated MAPKs recruited by agonists that engage the Ras protooncoprotein. Ras, in turn, recruits the Raf
family of MAP3Ks. The Rafs activate two ERK-specific
MAP2Ks: MEK1 and MEK2 (MAP2Ks-1 and -2, respectively). The biochemistry, biology, and regulation of Ras-dependent ERKs activation have been extensively reviewed (192,
229).
In many instances, the ERKs can also be activated, in a manner
independent of Ras, by proinflammatory stimuli including cytokines of the tumor necrosis factor (TNF) family, by PAMPs,
such as lipopolysaccharide produced by invading microbial
pathogens, and by DAMPs, endogenously produced danger
signals such as oxidized low-density lipoprotein (LDL) in atherosclerosis and crystalline uric acid in gout. PAMPs/DAMPs
690
B. The c-Jun NH2-Terminal Kinases, a MAPK
Group Activated by Growth Factors,
Environmental Stresses, and
Proinflammatory Stimuli
The c-Jun NH2-terminal kinases (JNKs) (TABLE 1) were
initially identified as a protein kinase activity purified from
cycloheximide-treated rat liver. The purified kinase was
able to phosphorylate the c-Jun transcription factor at
Ser63 and -73, sites key to the regulation of c-Jun and
activator protein-1 (AP-1) transactivation function (245).
Of note, the liver kinase phosphorylated c-Jun at a substantially higher rate than that observed for the ERKs. Independently, JNKs were identified as a Jun kinase activity that
adhered to immobilized c-Jun (59, 163–165, 225, 320).
JNKs, while activated by mitogens, are also vigorously activated by a variety of environmental stresses (heat shock,
ionizing radiation, oxidants), genotoxins (topoisomerase
inhibitors and alkylating agents), ischemic reperfusion injury, mechanical shear stress, vasoactive peptides, proinflammatory cytokines, PAMPs/DAMPs and, as expected
for a kinase activated in liver by cycloheximide, translational inhibitors (cycloheximide and anisomycin) (59, 165,
320).
The JNKs are also activated by tunicamycin. This is conceptually important inasmuch as tunicamycin, which inhibits Nlinked protein glycosylation, causes the accumulation of misfolded proteins exclusively within the lumen of the endoplasmic reticulum (ER). This leads to ER stress, and indeed,
tunicamycin serves as a model for ER stress induction. The
ability of an ER stress to activate the cytosolic JNK (165)
implicated the JNKs as important effectors for the ER stress
response. As discussed in section IVE, high-fat diet can trigger
ER stress in vivo, and this may play a critical role in triggering
insulin resistance and high-fat diet-induced inflammation.
The JNKs are encoded by three genes JNKs 1–3, also called
MAPKs 8, 9, and 10, respectively (59, 163–165) (TABLE 1).
Each JNK isoform contains the MAPK characteristic ThrX-Tyr phosphoacceptor loop in kinase subdomain VIII. For
JNKs, the sequence is Thr183-Pro-Tyr185. Each JNK gene
is subject to differential hnRNA splicing, both within the
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Despite the apparent ability of MAPK components to function in multiple pathways, the selectivity of MAPK pathways as well as their efficiency must be preserved. In this
capacity, scaffold proteins function to sequester select
MAPK pathway elements, thereby maintaining a critical
degree of pathway integrity and signaling efficiency. In
some cases, scaffold proteins are distinct polypeptides that
bind specific MAPK pathway components. Alternatively,
core MAPK signaling components themselves also possess
intrinsic scaffolding properties (203).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
Table 1 Nomenclature for mammalian MAPK pathway components
Name
Alternate Names
Substrates/Effectors
Human Genome Accession Number
Mouse Genome Accession Number
MAPK nomenclature
ERK1
ERK2
JNK1
isoforms:
JNK1␤2
JNK1␣2
JNK1␤1
JNK1␣1
JNK2
isoforms:
JNK3␤2
JNK3␣2
JNK3␤1
JNK3␣1
p38␣
p38␤
p38␥
p38␦
RSK, MNKs, MSKs, Elk1
RSK, MNKs, MSKs, Elk1
c-Jun, Jun D, ATF2, Elk1
NP_002737.2
NP_620407.1
NP_620637.1 (JNK1␣2)
NP_036082.1
NP_036079.1
NP_057909.1 (JNK1␣2)
c-Jun, JunD, ATF2, Elk1
NP_002743.3 (JNK2␣2)
NP_997575.2 (JNK2␣2)
c-Jun, JunD, ATF2, Elk1
NP_620448.1 (JNK3␣2)
NP_001075036.1 (JNK3␣2)
MK2/3, MSKs, ATF2,
Elk1, MEF2C
MK2/3, MSKs, ATF2
NP_001306.1
NP_001161980.1
NP_002742.3
NP_034991.4
ATF2
NP_002960.2
NP_038899.1
ATF2
NP_002745.1
NP_036080.2
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
isoforms:
JNK2␤2
JNK2␣2
JNK2␤1
JNK2␣1
JNK3
Mapk3, p44-MAPK
Mapk1, p42-MAPK
Mapk8, SAPK-␥,
SAPK1c
isoforms:
SAPK-p54␥1
SAPK-p54␥2
SAPK-p46␥1
SAPK-p46␥2
Mapk9, SAPK-␣,
SAPK1a
isoforms:
SAPK-p54␣1
SAPK-p54␣2
SAPK-p46␣1
SAPK-p46␣2
Mapk10, SAPK-␤,
SAPK1b
isoforms:
SAPK-p54␤1
SAPK-p54␤2
SAPK-p46␤1
SAPK-p46␤2
Mapk14, SAPK2a,
CSBP1
Mapk11, SAPK2b,
p38-2
Mapk12, SAPK3
ERK6
Mapk13, SAPK4
MEK nomenclature
MEK1
MEK2
MKK4
MKK7
MKK3
MKK6
Map2k1, MKK1
Map2k2, MKK2
Map2k4, SEK1,
JNK kinase
(JNKK)-1, MEK4,
SAPK-kinase
(SKK)-1
Map2k7, JNKK2,
MEK7, SKK4
Map2k3, MEK3,
SKK2
Map2k6, MEK6,
SKK3
ERK1, ERK2
ERK1, ERK2
JNKs
NP_002746.1
NP_109587.1
NP_003001.1
NP_032953.1
NP_075627.2
NP_033183.1
JNKs
NP_660186.1
NP_001157644.1
p38 s
NP_659731.1
NP_032954.1
p38 s
NP_002749.2
NP_036073.1
MAP3K nomenclature
Raf-1
A-Raf
B-Raf
MEKK1
MEKK2
Map3k1
Map3k2
MEK1, MEK2
MEK1, MEK2
MEK1, MEK2
MKK4, MKK7
MKK4, MEK1
NP_002871.1
NP_001645.1
NP_004324.2
NP_005912.1
NP_006600.3
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
NP_084056.1
NP_033833.1
NP_647455.2
NP_036075.2
NP_036076.2
691
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
Table 1—Continued
Name
Alternate Names
Map3k3
MEKK4
Map3k4, MTK1
(MAP three
kinase-1)
Map3k5
Map3k6
Map3k7
Map3k8, Cot
Map3k9
Map3k10, MST
Map3k11, SPRK,
PTK1
Map3k12, MUK,
ZPK
PSK (?a splicing
isoform of TAO2)
ASK1
ASK2
TAK1
Tpl-2
MLK1
MLK2
MLK3
DLK
TAO1/2
Human Genome Accession Number
Mouse Genome Accession Number
MKK4, MEK1, ?MKK3,
?MKK6
MKK4, MKK3, MKK6
NP_976226.1
NP_036077.1
NP_005913.2
NP_036078.2
MKK4, MKK3, MKK6
MKK4, MKK3, MKK6
MKK4, MKK3, MKK6
MEK1, SEK1
MKK4, MKK7
MKK4, MKK7
MKK4, MKK7
NP_005914.1
NP_004663.3
NP_033342.1
NP_005195.2
NP_149132.2
NP_002437.2
NP_002410.1
NP_032606.4
NP_057902.3
NP_663304.1
NP_031772.1
NP_001167578.1
NP_001074761.1
NP_071295.2
MKK4, MKK7
NP_001180440.1
NP_033608.3
MKK3, ?MKK4/MKK7
(PSK only)
NP_065842.1 (TAO1)
NP_659074.2 (TAO1)
GCK/PAK nomenclature
GCK
GCKR
GLK1
HPK1
NIK
PAK1
PAK2
PAK3
MAPK-kinasekinase-kinase-2
(Map4k2)
Map4k5, kinase
homologous to
Ste20 (KHS)
Map4k3
Map4k1
Map4k4, NIK (not
to be confused
with NF-␬Binducing kinase),
HPK1/GCKrelated kinase
(HGK)
␣PAK
␥PAK, hPAK65
␤PAK
MEKK1, MLK3
NP_004570.2
NP_033032.1
MEKK1
NP_006566.2
NP_077237.2
?
MEKK1, MLK3
MEKK1
NP_003609.2
NP_009112.1
NP_663719.1
NP_001074826.1
NP_032305.2
NP_032722.2
?
?
Raf-1
NP_001122092.1
NP_002568.2
NP_001121640.1
NP_035165.2
NP_796300.1
NP_001181975.1
Included are commonly accepted names found in the primary literature. Genome Project nomenclature gene
designations are indicated in italics. Not all of the names listed are included in the text. Accession numbers are
from UniGene and link to additional genomic information.
catalytic domain at a region spanning subdomains IX and X
(resulting in ␣- and ␤-JNKs, respectively, TABLE 1), and at
the extreme COOH terminus to produce 46- and 54-kDa
polypeptides (respectively, types 1 and 2 JNKs, TABLE 1).
Accordingly, 12 JNK polypeptides exist. The functional significance of the type 1 and type 2 isoforms is unclear. The ␣
and ␤ JNKs apparently differ modestly in their affinities for
substrates (47, 59, 137, 165, 320).
C. The p38 MAPKs Are Activated by
Stresses and Proinflammatory Stimuli
protein in extracts of endotoxin-treated cells. cDNA cloning revealed that p38␣ was remarkably closely related to the
Streptomyces cerevesiae osmosensing MAPK HOG1. Thus,
as with Hog1p, the phosphoacceptor motif of the p38s has
the sequence Thr-Gly-Tyr (104, 108). p38␣ was also identified independently and contemporaneously as a stressand IL-1-activated kinase that could phosphorylate and activate MAPK-activated protein kinase-activated protein kinase-1 (MK2; sect. IID1). MK2 is a member of a Ser/Thr
kinase family that, among other substrates, phosphorylates
and activates the small heat shock protein Hsp27 (80, 237).
The first p38 MAPK identified, the ␣ isoform, was isolated
on anti-phosphotyrosine immunoaffinity beads as a P-Tyr
There are four p38 genes (TABLE 1): p38␣-␦, also called
MAPK-14 (p38␣), -11 (p38␤), -12 (p38␥), and -13
692
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
MEKK3
Substrates/Effectors
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
(p38␦) (80, 98, 104, 132, 133, 170, 171, 194, 237, 274,
310), all of which are preferentially activated in situ by
environmental stresses, inflammatory cytokines, PAMPs,
and DAMPs.
D. Key Substrates of the ERKs, JNKs,
and p38s
The ERKs, JNKs, and p38s phosphorylate both other protein kinases and transcription factors, highlighting the importance of these MAPKs to inflammatory and stress responses.
1. Protein kinases
A) RIBOSOMAL S6 KINASES (RSKS), SELECTIVE SUBSTRATES FOR THE
ERKS.
Rsks are a family of Ser/Thr kinases that contain two
tandem protein kinase domains: the COOH-terminal kinase domain is in the calcium-calmodulin kinase subfamily,
FIGURE 1.
The Rsks can also be activated, selectively in dendritic cells
and possibly macrophages, through an alternative mechanism. Here, TLRs trigger activation of p38 which, in turn,
stimulates the activities of MKs-2 and -3 (see below). Genetic and biochemical evidence shows that these kinases
phosphorylate Ser380 (human Rsk1), which is phosphorylated in cells other than dendritic cells by the ERK-activated
COOH-terminal catalytic domain. Thus this MK2/3-mediated phosphorylation bypasses the necessity for phosphorylation of Rsk at the COOH-terminal kinase domain
(Ser573). However, ERK phosphorylation of Ser363 in the
MAPK regulation of downstream protein kinases.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
693
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Contemporaneously, p38␣ was isolated as a targeet for a
class of experimental pyridinyl-imidazole anti-inflammatory drugs, the most extensively characterized of which is
the compound SB203580 (171). In vitro assays demonstrated that only p38␣ and p38␤ are inhibited by SB203580
and its relative compounds. p38␥ and p38␦ are not at all
affected by these drugs in vitro or in situ (98). The reason
for the selective inhibition of p38␣ and p38␤ by SB203580
is that Thr106 resides in the hinge of the ATP binding
pocket of p38␣ (and p38␤). In this pocket, Thr106 interacts
with a fluorine atom, the SB203580. This positions the drug
such that it interacts with His107 and Leu108 of the ATP
binding pocket-blocking ATP binding. Substitution of the
more bulky amino acids found in p38␥ or p38␦ (Met, and
Pro or Phe, respectively, in both cases) for Thr106 abolishes
SB203580 binding. Conversely, if Thr, corresponding to
Thr106, is used to replace the bulky amino acids of p38␥,
p38␦ (or even of JNK1), the resulting mutants are rendered
at least partially sensitive to SB203580 (42, 75, 110, 246).
and the NH2-terminal kinase domain is in the AGC subfamily. Rsks are so named for their phosphorylation of the
small ribosomal protein S6, in Xenopus oocytes. However,
Rsks do not represent the major somatic cell S6 kinases
(these are the S6Ks). The Rsks were among the first MAPK
substrates identified, and demonstration of ERK activation
of Rsk was a major breakthrough in the elucidation of
MAPK pathways. JNKs and p38s do not regulate the Rsks.
Rsks are activated through the combined activity of ERKs
and 3-phosphoinositide-dependent kinase-1 (PDK1). ERKs
activate the COOH-terminal catalytic domain by phosphorylating a Ser/Thr residue in the activation loop (e.g.,
Ser573 in human Rsk1) and participate in the activation of
the NH2-terminal catalytic domain by phosphorylating a
Ser-Pro site (e.g., Ser363 in human Rsk1) in the “turn”
motif beyond the catalytic domain of the NH2-terminal
kinase. The activated COOH-terminal catalytic domain
then phosphorylates Ser380 (in a Phe-Ser-Phe motif) located further downstream of the NH2-terminal kinase domain. This provides a docking site for PDK1, which phosphorylates Rsk1 at Ser222 in the activation loop of the
NH2-terminal catalytic domain, a final phosphorylation
event required for full activation of Rsk. It is the NH2terminal kinase domain that is likely responsible for phosphorylation of Rsk substrates. (FIGURE 1). Phosphorylation
of Rsk substrates is likely mediated by the Rsk the NH2terminal kinase domain (49, 231).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
MK2 is phosphorylated at Thr24, Thr222, and Ser272 and
activated by p38␣ and p38␤ (but not by p38␥ or p38␦). In
line with the regulation of MK-2, -3, and -5 as well as
Hsp27 by p38␣ and p38␤, these processes are inhibited by
pyridinyl-imidazole p38␣/␤ inhibitors such as SB203580
(14, 123, 168, 193).
This alternative mechanism apparently functions selectively
in dendritic cells (although it may also function in macrophages) and is necessary for TLR-stimulated proinflammatory dendritic cell phagocytosis and pinocytosis of antigens,
as well as for later induction of IL-10 and for resolution of
inflammation (348) (see sect. IVC1).
C) MAPK-INTERACTING KINASES. A site of translational regulation for many mRNAs is the the N7-methylguanosine 5=
cap structure. Eukaryotic initiation factor (eIF)-4E (eIF4E), an N 7-methylguanosine-binding protein, tethers
mRNAs onto eIF-rG, a scaffold protein, which also binds
eIF-4A, an RNA helicase. The eIF-4A polypeptide, together with eIF-4B, an RNA binding protein, unwinds
secondary structure in the mRNA 5= untranslated region.
This facilitates the scanning of the mRNA by the 40S
ribosomal complex which moves along the mRNA to the
ATG translational initiation site. The complex of eIF-4A,
-B, -G, and -E is referred to collectively as eIF-4F (272).
B) MITOGEN-ACTIVATED PROTEIN KINASE-ACTIVATED PROTEIN KINASES (MK)-2, -3, AND -5.
MAPK-activated kinase-2 (MK2) the
structurally related MK3 and MK5 (also called p38-regulated and activated kinase, PRAK) are a small family of
Ser/Thr kinases strongly activated by stresses, proinflammatory cytokines, and PAMPs/DAMPs. MK2/3 both consist of an NH2-terminal regulatory domain and a COOHterminal kinase domain. They are unrelated to the Rsks.
Recombinant MK-2, -3, and -5 phosphorylate the small
heat shock protein Hsp27 in vitro (123, 168, 193, 205, 269,
275, 276). However, this substrate promiscuity may not
occur in situ inasmuch as only endogenous MK2 can robustly phosphorylate Hsp27, while MK3 does so much
more weakly. Endogenous MK5 cannot phosphorylate
Hsp27 at all. MK2 can also phosphorylate a number of
mRNA binding proteins implicated in the signal-induced,
posttranscriptional stabilization of mRNAs, notably those
encoding proinflammatory cytokines (87, 233, 234, 265).
This is discussed in detail in section IVC5.
When not phosphorylated, Hsp27 aggregates into highmolecular-weight multimers. It is these complexes that
serve as molecular chaperones. Hsp27 phosphorylation,
catalyzed by MK2/3 at Ser15, Ser78, Ser82, and Ser90,
coincides with the dissociation of Hsp27 into monomers
and dimers and redistribution of Hsp27 to the actin cytoskeleton. Hsp27 redistribution may lead to the reorganization of F-actin into stress fibers, thus affecting cell
motility (123, 168).
As is discussed in section IVC5, MK2 can also phosphorylate a number of polypeptides that bind to mRNAs and
modulate their stability. This process enables p38, through
MK2, to regulate posttranscriptionally the in vivo and in
situ expression of several proinflammatory cytokine genes
including that for TNF (57, 87, 185, 228, 233, 234, 238,
265).
Finally, as noted above, MK2 and MK3 are required for
optimal activation, by TLRs, of dendritic cell Rsk and for
consequent dendritic cell phagocytic function (348).
694
The translational repressor protein 4E-binding protein-1
(4E-BP1 also referred to as phosphorylated heat- and acidstable protein regulated by insulin, PHAS-I) negatively regulates the eIF-4E-eIF-4G interaction. Insulin and mitogenstimulated phosphorylation of 4E-BP1, catalyzed by mammalian target of rapamycin (mTOR), results in the
dissociation of 4E-BP1 from eIF-4E and the release of eIF4E. As a consequence, eIF-4E is incorporated into the
eIF-4F complex, thereby enabling translational initiation
(272).
Insulin, mitogens, and environmental stresses also stimulate
the regulatory phosphorylation of eIF-4E at Ser209. Evidence has been presented indicating that this phosphorylation contributes to eIF-4E’s oncogenic potential (86, 297,
318), but its specific role in translational regulation remains
uncertain (21, 272).
The likely eIF-4E Ser209 kinases are two closely related
Ser/Thr kinases MAPK interacting kinases (MNKs)-1 and
-2. MNKs, as their names suggest, interact with and are in
vitro and in vivo substrates of MAPKs. Both ERKs 1/2 (in
response to insulin and mitogens) and by the p38s (in response to cytokines and stress) can catalyze phosphorylation and activation of the MNKs (85, 311, 312), indicating
that, as with AP-1 (see below), MNKs are a site of integration of stress and mitogenic signaling pathways. Given their
role in modulating eIF4E phosphorylation, which has been
linked to tumorigenesis (86, 297, 318), it is reasonable to
suggest that MNKs promote tumor development (FIGURE 1).
D) MSK1/2. Mitogen- and stress-activated protein kinases
(MSK)-1 and -2 are a family of Ser/Thr protein kinases that,
like the Rsks, possess two tandem protein kinase domains.
The activity of MSK1 is activated by both the ERKs and p38
MAPKs (56).
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
NH2-terminal domain is still required, and indeed, both
ERK and p38 MK2/3 are necessary for optimal TLR activation of dendritic cell Rsk. In this dendritic cell system,
MK2/3-catalyzed phosphorylation of Rsk Ser380 still provides the PDK1 recognition site which facilitates Ser222
phosphorylation (348).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
2. Transcription factors
A) CHOP/GADD153. The bZIP, CCAAT enhancer binding protein (c/EBP) family transcription factor C/EBP homologous
protein (CHOP), also called growth arrest and DNA damage-155 (GADD153) (214), is induced at the transcriptional level and functionally activated in response to genotoxic and inflammatory stimuli. The transcriptional regulatory functions of CHOP/GADD153 are activated by
stimulus-induced phosphorylation of Ser78 and Ser81.
CHOP/GADD153 acts as both a transcriptional repressor
(of certain cAMP-regulated genes) and a transcriptional activator (of a subset stress-induced genes). Genotoxin-induced cell cycle arrest at G1/S is mediated in part by CHOP/
GADD153. This is an important process, as cell cycle arrest
permits DNA repair prior to S phase, thereby preserving the
integrity of the genome. Inasmuch as p38␣, but not JNK or
ERK, phosphorylates CHOP/GADD153 at Ser78 and
Ser81 in vivo and in vitro, it is likely that p38␣ is a stressactivated regulator of CHOP/GADD153 function (307)
(FIGURE 2).
B) NFAT.
The nuclear factor of activated T cells (NFAT)
family of transcription factors is distantly related to Rel/
NF-␬B. In resting cells, NFATs are phosphorylated [by casein kinase-I␣ and, possibly, glycogen synthase kinase
(GSK)-3] at five or six sites. This masks the NFAT nuclear
localization signal resulting in the retention of NFATs in the
cytosol. Stimulus-dependent Ca2⫹ entry activates the Ca2⫹dependent phosphatase calcineurin (phosphatase 2B). Calcineurin then dephosphorylates NFATs, unmasking
NFAT’s nuclear localization signal thereby enabling NFAT
nuclear translocation. NFAT DNA binding affinity is also
enhanced by dephosphorylation. NFAT and AP-1 sites are
often closely apposed in many promoters. This favors the
cooperative binding and synergistic trans activation of numerous genes (IL-2, IL-4, IL-5, and CD40L are examples)
(113, 184). The immunosuppressants FK506 and cyclosporin-A are potent calcineurin inhibitors, which accounts
for essentially all of their biological activity (256).
The Ca2⫹-mediated nuclear translocation of NFATs is
strongly inhibited by serum factors. Neither casein kinase
nor GSK3 is serum-stimulated. Accordingly, it was suggested that serum-dependent inhibition of Ca2⫹-mediated
NFAT translocation was mediated by serum-responsive kinase cascades. Inasmuch as nuclear translocation is essential to NFAT function, these kinases are considered NFAT
inhibitors (36, 113). In line with this, the JNKs phosphorylate NFAT4 (also called NFATc3) at Ser163 and Ser165,
coinciding with an inhibition of stimulus-induced NFAT4
nuclear translocation (36) (FIGURE 2A).
Coadministration of phorbol myristate acetate (PMA)
and ionomycin stimulates JNK-dependent phosphorylation of T lymphocyte NFAT2 (also called NFATc1) at
Ser117 and -172. As with NFAT4, this phosphorylation
inhibits or delays the accumulation of NFAT2 in the
nucleus. This inhibition is due to the ability of JNK phosphorylation of NFAT2 to prevent the binding of calcineurin, an obligate step in NFAT2 activation (37).
NFAT2 activity is critical to the polarization of TH cells
to the TH2 effector phenotype, consistent with the idea
that JNK activation biases TH polarization to the TH1
trajectory at least in part through NFAT2 inhibition. In
line with this, disruption of JNK1 or both JNK1 and -2
leads to the preferential accumulation of TH2 cells (37,
66, 67) (see sect. IVC2).
C) AP-1. The JNKs and p38s are the major Ser/Thr kinases
responsible for the recruitment of the heterodimeric transcription factor activator protein-1 (AP-1) (147, 166); however, the ERKs also function to regulate AP-1 (FIGURE 2, A
AND C). AP-1 dimers consist of bZIP transcription factors of
the Jun group, primarily c-Jun and JunD, paired either with
members of the fos (usually c-Fos) or activating transcription factor (ATF, usually ATF2) families. Characteristics of
bZIP transcription factors are leucine zippers that enable
homo- and heterodimerization AP-1 components are most
commonly observed as Jun-Jun, Jun-Fos, or Jun-ATF
dimers (143, 260, 300, 301).
AP-1, via Jun family components, binds to the tetradecanoylphorbol acetate (TPA) response element (TRE). Insofar
as ATFs, including ATF2, in the CREB subfamily of bZIP
transcription factors, AP-1 heterodimers containing ATFs
can bind both to the TRE and to the CRE (113, 147). AP-1
is a pivotal trans activator of a number of genes recruited by
stress and inflammation. These include the genes for IL-1
and IL-2, CD40, CD30, TNF, and c-Jun itself. AP-1 also
participates in the transcriptional induction of proteases
and cell adhesion proteins, such as E-selectin, which are
important to inflammation (143, 260).
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
695
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
The bZIP transcription factor cAMP response element
binding protein (CREB) binds and trans activates genes
containing the cAMP response element (CRE). Numerous
divergent stimuli including mitogens, stresses and, of
course, agonists that elevate cAMP (via PKA), can activate
CREB trans activating function, a process that correlates
with phosphorylation of CREB at Ser133. In this context,
CREB serves as a point of signaling convergence, responsive
to multiple inputs in different physiological contexts. Thus
several protein kinases including the Rsks (333) and MK2
(285) are capable of phosphorylating CREB Ser133 in response to mitogen or stress. In addition, substantial evidence indicates that MSK1 is a physiologically relevant
stress- and mitogen-activated CREB Ser133 kinase (FIGURE
1). Regulation of CREB by PKA has been associated with
learning and memory (191). In contrast, in vivo, MSKs,
possibly through CREB regulation, may act redundantly as
negative regulators of innate immunity (6, 56) (section
IVC1).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
Regulation of AP-1 is complex and occurs at several levels. These include the direct phosphorylation/dephosphorylation of AP-1 components, the stabilization of
AP-1 components (notably c-Jun), and the recruitment
(typically via phosphorylation) of transcription factors
that induce c-jun or c-fos. These events can be activated
independently by several signaling pathways (FIGURE 2).
Phosphorylation of c-Jun, in resting cells, at a region
immediately upstream of the COOH-terminal of the
DNA binding domain (Thr231, Thr239, Ser243, Ser249)
inhibits DNA binding. This phosphorylation is catalyzed
in situ by either glycogen synthase kinase-3 (GSK3) or
casein kinase II (CKII), both of which are constitutively
active in resting cells. Of note, GSK3 is phosphorylated
and inactivated by a mechanism dependent on phosphatidylinositol-3-kinase (PI-3-K) (38). The same conditions
that activate AP-1 foster GSK3 inhibition. Thus, under
696
these circumstances, the c-Jun COOH-terminal phosphates are removed (17).
The trans activating activity of c-Jun and ATF2 correlates
strongly with phosphorylation within their NH2-terminal
trans activation domains (Ser63 and -73 for c-Jun, Thr69
and -71 for ATF2). The JNKs are the principal kinases
responsible for c-Jun Ser63/73 phosphorylation (59, 165,
225, 260). Thus all stress- and TNF-activated c-Jun
Ser63/73 kinase activity can be effectively removed either
by immunodepletion or genetic disruption of JNK (165,
300, 301) (FIGURE 2A). JNK also phosphorylates JunD, at
Ser90 and Ser100, a region of the JunD trans activation
domain similar to the phosphoacceptor domain of c-Jun
(138).
JNK activity also regulates c-Jun stability, albeit in a complex and incompletely understood manner (FIGS. 2B and
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
FIGURE 2. Regulation of transcription factors by MAPKs. A: key MAPK transcription factor substrates. Note
that the complex regulation of AP-1 involves direct phosphorylation (c-Jun) as well as phosphorylation of
elements that induce components (detailed in C). B: c-Jun phosphorylation by JNK can modulate c-Jun stability.
JNK and MEKK1 (independently or through JNK) can recruit the Itch E3 ligase to foster c-Jun ubiquitination.
One consequence of c-Jun destabilization is appropriate tolerization of TH2 cells. This can be compared with
JunB degradation, also mediated by MEKK1, which is important for suppressing IL-4 production and consequent TH2 polarization functions (FIGURE 4C). C: MAPKs phosphorylate and activate transcription factors
that upregulate AP-1 component genes. Note also that c-jun has an AP-1 element in its promoter and can be
induced by activated c-Jun/AP-1.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
4C). c-Jun is susceptible to Lys48-linked ubiquitination and
proteasomal degradation. Early studies (83, 204) indicated
that the binding of JNK to c-Jun (which promotes c-Jun
phosphorylation) targeted c-Jun for ubiquitination and
degradation (by an unknown mechanism), whereas dephosphorylation of c-Jun Ser63/73 rendered c-Jun resistant to
ubiquitination and degradation, prolonging its half life (FIGURE 2B).
JNKs and p38s phosphorylation of ATF2 at Thr69 and
Ser71 in the trans activation domain correlates with activation of ATF2 trans activating activity (102). Whether the
JNKs or p38s represent the dominant ATF2 kinases is cell
and stimulus dependent (45, 201).
AP-1 activation also involves the induction of genes encoding AP-1 components, a process that can be activated by the
ERKs, JNKs, and p38s (105, 143, 260) (FIGURE 3). Induction of c-fos, of the most rapid transcriptional events, requires a cis-acting element in the c-fos promoter, the serum
response element (SRE). A heterodimeric transcription factor containing two polypeptides, the serum response factor
(SRF) and a member of the ternary complex factor (TCF)
family, bind to the SRE and mediate c-fos induction (295).
FIGURE 3. Regulation of MAPKs by MEK-dependent and -independent phosphorylation. All three MAPK groups listed are regulated by
direct MEK phosphorylation; however, T cell p38␣ is also regulated
by Lck/Zap70-mediated Tyr phosphorylation, which triggers autophosphorylation at the activating phosphoacceptor sites.
The TCF family includes Elk-1 and Sap-1a (295). The Elk1
COOH terminus (Ser383, Ser389) is phosphorylated by the
JNKs and ERKs (but not p38). The corresponding residues
(Ser381 and Ser387) on Sap1a are phosphorylated by the
p38s (96, 131, 187, 322, 341). TCF phosphorylation enhances heteromerization with SRF and consequent trans
activation at the SRE (FIGURE 2).
As their name suggests, transcription factors of the myocyte
enhancer factor-2 (MEF2) subgroup of the MCM1-agamous and deficiens-SRF (MADS) box transcription factor
family regulate genes involved in myocyte differentiation.
However, some MEF2s, notably MEF2C, are expressed
more widely and regulate diverse transcriptional events. Of
the MEF2 family, only MEF2A and -C are MAPK substrates, a process that enhances MEF2A/C trans activating
activity (349). MEF2A is phosphorylated at Thr312 and
Thr319 by p38␣ (349). The p38s phosphorylate Thr293
and Thr300 of MEF2C (105). A MEF2C cis element is
present in the c-jun promoter. Indeed, MEF2A or -C, once
activated by p38s, can trans activate the c-Jun promoter.
Accordingly, p38 activation can contribute to the induction
of c-jun expression (105), which, in turn, contributes to the
activation of AP-1 (105, 349).
3. MAPK substrates have specific MAPK docking
sites which confer substrate for the JNKs and
p38s and which interact selectively with distinct
substrate binding motifs present on MAPKs
JNK-catalyzed phosphorylation of c-Jun is an excellent example of the mechanism of substrate recognition by members of the MAPK family. Despite being “proline-directed”
phosphorylating Ser/Thr residues only if followed immediately by proline-MAPK substrate specificity is directed in
large part by the presence of binding sites on MAPK substrates. These binding motifs are often quite distant from
the “proline-directed” phosphoacceptor sites and are typically specific for distinct MAPK subgroups. This enables the
selective interaction between MAPKs and their physiological substrates (15, 288, 289) (TABLES 2 and 3).
There are two broad classes of MAPK binding site present
on MAPK substrates. One, referred to as a D-domain, generally consists of a hydrophobic cluster of amino acid residues. These motifs are present in many transcription factor
substrates of MAPKs. For example, the JNKs, but not the
ERKs or p38s, bind c-Jun quite strongly through an interaction that requires a domain on c-Jun between residues 32
and 52, substantially distant from the phosphoacceptor
sites Ser63 and Ser73. This JNK docking site overlaps with
the so-called ␦-domain (AAs 30 –57), which has been linked
to the regulation of c-Jun oncogenicity and is deleted in
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
697
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Consistent with a role for JNK in destabilization of c-Jun,
JNKs can also indirectly contribute to ubiquitin (Ub)-dependent c-Jun degradation through phosphorylation and
activation of the HECT domain E3 ubiquitin ligase Itch.
Itch, so activated, then coordinates the E1 and Ubc7 (E2)dependent ubiquitination of c-Jun (92). The MAP3K
MEKK1 may regulate JNK-dependent Itch phosphorylation and, independently, may directly regulate Itch (73,
302).
TCF phosphorylation is central to regulation of this process.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
Table 2 MAPK and MAPK substrate/regulator docking sites
MAPK Substrate
Hydrophobic Domain
Polybasic Docking Site
c-Jun
JunD
JunB
MKK6
MNK1
25-AYGYSNPKILKQSMTLNLADP VGSLK-50
43-APPTSSM LKKDALTLSLADEGAAGLK-68
26-SLSLHDYKLLKPTLALNLADP YRGLK-51
4-SKGKKKNPGLKIP-16
399-CKSRLARRRALAQAG-413
Hydrophobic and polybasic MAPK docking sites on representative MAPK substrates. For the Jun family
members, only c-Jun and JunB can actually associate with JNKs. Only c-Jun and JunD, however, have
phosphoacceptor sites. Thus JunD must heterodimerize with JunB or c-Jun to serve as JNK substrate. The
site in c-Jun is associated with the ␦-domain, see text for details.
As with the c-Jun ␦-domain, ATF2 contains a hydrophobic
substrate binding motif (AAs 20 – 60) that lies well NH2
terminal to the phosphoacceptor sites (Thr69 and Thr71)
(102). Likewise, Elk-1 has a MAPK docking site, the D-domain (AAs 312–335), that also resides significantly NH2
terminal to the phosphoacceptor sites (Ser383, Ser389) and
is required for efficient ERK and JNK binding and phosphorylation. The D-domain is a common motif present on
many MAPK interactors (289, 322, 341).
A second class of MAPK binding domain is comprised of a
small stretch of basic amino acid residues. This domain is
prevalent among protein kinase substrates of MAPKs
(Rsks, MNKs, and MKs-2–5) as well as MAPK regulators
[notably the JNK-specific MAP2Ks MAPK kinase (MKK)-4
and some MKK7 isoforms, p38 MAP2Ks MKK3 and
MKK6; as well as MAPK phosphatases (TABLES 2 and 3)].
This polybasic domain binds to the so-called common
docking (CD domain) domain, an extracatalytic region
found in all MAPKs. CD motifs are rich in acidic amino acid
residues. The basic residues in these MAPK binding sites
likely interact electrostatically with the cognate MAPK CD
motifs, with differences among the basic MAPK docking
sites and MAPK CD domains likely conferring a high degree
of specificity among MAPKs for their substrates and activators (15, 288, 289).
E. Direct Activation of the ERKs, JNKs, and
p38s by Specific MAP2Ks
1. Activation of the ERKs by MAP2Ks-1 and -2
As with most MAPKs, the ERKs are activated by dual Tyr/
Thr phosphorylation within the activation loop (Thr185/
Tyr187 for ERK2 within the characteristic motif Thr-GluTyr). The existence of a MAPK “activator” was first demonstrated by fractionating cytosolic extracts of EGF-treated
cells on ion exchange columns and assaying the fractions for
an activity that could activate ERK1 and/or ERK2 in vitro.
Purification and molecular cloning of this activity revealed a
pair of dual-specificity (Thr/Tyr) protein kinases termed
variously MAPK-kinases-1 and -2 (MAPKKs-1/2) or
MAPK/ERK kinases-1 or -2 (MAP2Ks-1/2) (TABLE 1 and
FIGURE 3) that could phosphorylate the ERKs at the activating phosphoacceptor sites (229).
MEK1 can be inhibited by two relatively specific pharmacologic agents: PD98059 and U0126. PD98059 prevents
MEK1 phosphorylation by upstream kinases. U0126 does
not prevent phosphorylation of MEK1 by upstream activators, but instead restrains MEK1 in an inactive conformation, preventing its signaling to downstream elements.
Both of these compounds have been used extensively to
identify pathways in which the ERKs play a dominant
role (8, 9, 54).
2. JNK activation by MKK4 and MKK7
Table 3 Representative mammalian MAPK CD sites
MAPK
CD Site Sequence
ERK1
p38␣
JNK2
330-LEQYYDPTDEPVAE-343
308-FAQYHDPDDEPVAD-321
321-INVWYDPSEAEAPP-334
Representative MAPK CD motifs for ERK, p38, and JNK isoforms.
Note the relative predominance of acidic amino acids, especially for
ERK1 and p38␣.
698
Activating JNKs phosphorylation occurs at Thr183 and
Tyr185 (phosphoacceptor sites separated by a Pro residue)
and is catalyzed by two MAP2Ks. MAPK-kinase-4 (MKK4,
also called MAP2K4, SAPK/ERK kinase-1, SEK1; MEK4;
JNK kinase-1, JNKK1; and SAPK-kinase-1, SKK1, TABLE
1) can phosphorylate and activate all three JNK isoforms in
vivo and in vitro. MKK4 can also phosphorylate and activate p38 (60, 252) (FIGURE 3). MKK7 (also called
MAP2K7, MEK7, JNKK2, and SKK4, TABLE 1) bears significant homology to the JNK-specific Drosophila MAP2K
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
oncogenic v-Jun (not surprisingly, v-Jun does not bind JNK
and is not a JNK substrate) (47, 137, 138) (TABLES 2 and 3).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
hemipterous. MKK7 is expressed from several alternative
hnRNA splicing isoforms, each of which displays a strong
preference for JNK, activating all isoforms equally well.
MKK7 does not activate p38 at all (97, 182, 293, 329, 343)
(FIGURE 3).
F. A Divergent Array of Proinflammatory
and Stress-Activated MAP3K Families
Are Upstream of the ERKs, JNKs,
and p38s
1. General considerations
3. p38 activation by MKK3 and MKK6
4. MKK3/6-independent “noncanonical” activation
of p38
p38␣ can also undergo MKK-independent activation. Two
mechanisms have been documented. For the first, p38␣ directly interacts with the adapter protein TGF-␤-activated
kinase-1-binding protein-1 (TAB1; see sect. IIID4) leading
to autophosphorylation at the classical Thr180/Tyr182 activating phosphoacceptor sites (FIGURE 3). This interaction,
detected using overexpressed proteins, can be stimulated by
another adapter protein, TNF receptor-associated factor-6
(TRAF6; discussed below, see sect. IIID), and may couple
p38␣ to PAMP agonists (93). However, the overall physiological relevance of this mechanism is unclear. Indeed,
mouse embryo fibroblasts (MEFs in contrast to the situation described below for T lymphocytes), in which both
mkk3 and mkk6 are disrupted, show a profound defect in
all known mechanisms of p38␣ activation. The exceptions
are a small portion of the activation of p38␣ by hyperosmolarity, anisomycin, and peroxynitrite (142).
A second noncanonical p38␣ activation mechanism occurs
exclusively in T lymphocytes and involves direct Tyr phosphorylation at a site outside of the classical Thr180/Tyr182
activating phosphoacceptor sites. This Tyr phosphorylation
occurs at Tyr323 and is dependent on the nonreceptor Tyr
kinase Lck and its effector Tyr kinase, Zap70 (FIGURE 3).
Zap70 directly phosphorylates p38␣ at Tyr323. Tyr323
phosphorylation triggers p38␣ autophosphorylation at
Thr180/Tyr182. Tyr323 lies in a linker region between the
two lobes of the p38␣ polypeptide, a region implicated in
homodimerization. It is possible that phosphorylation of
this residue may promote trans phosphorylation or some
other mechanism of activation (250, 251). Lck and Zap70
are pivotal to T cell receptor (TCR) signaling. As discussed
in section IVC3, genetic studies document that this activation mechanism is essential to T cell function (135).
Consistent with the multiple different stimuli that recruit
stress-activated MAPK pathways, a dauntingly diverse
number of Ser/Thr kinases regulates stress-activated
MAPKs, and the enzymology of these MAP3Ks is accordingly complex.
2. The MEKK group
The mammalian MEKKs consist of MEKK-1 and -4. NF␬B-inducing kinase (NIK, not to be confused with the germinal center kinase family member Nck interacting kinase,
also abbreviated NIK, which is discussed below) is an additional MEKK. However, NIK is a specific activator of the
NF-␬B pathway (186) and will not be discussed here. The
MEKK catalytic domains are, of mammalian kinases, most
closely homologous to that of S. cerevisiae STE11, a
MAP3K recruited both by mating pheromone and hyperosmolarity (108). Outside of the kinase domains, however,
the MEKKs are structurally quite dissimilar. Mammalian
MEKKs can, in some instances, catalyze the activation of
multiple different MAP2Ks and interact with numerous putative regulatory proteins (16, 71, 95, 108, 169, 283, 335)
(TABLE 1).
A) MEKK1, A RELATIVELY SELECTIVE ACTIVATOR OF THE JNK PATHWAY. MEKK1 (MAP3K1) is a very large (⬃190 kDa) polypeptide consisting of a COOH-terminal kinase domain
[amino acids (AAs) 1221–1493] preceded by an extensive
NH2-terminal domain (AAs 1–1221). The NH2-terminal
domain includes two proline-rich segments (AAs 74 –149
and 233–291) with putative binding sites for proteins with
SH3 domains, a consensus binding site for 14 –3-3 proteins
(AAs 239 –243), two PH domains (AAs 439 – 455 and 643–
750), and two sites for cleavage by cysteine proteases of the
caspase family (Asp871, Asp874) (22, 77, 169, 335).
The NH2 terminus of MEKK1 also contains a plant homeodomain (PHD)/really interesting new gene (RING) do-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
699
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
The activating phosphoacceptor sites on the p38s are separated by a glycine residue (Thr180/Tyr182 for p38␣).
MKK3 (also called MAP2K3, MEK3, and SKK2) and
MKK6 (also called MAP2K6, MEK6, and SKK3, TABLE 1)
are both highly selective activators of the p38s and do not
activate JNK or ERK under all conditions tested (FIGURE 3).
There are very slight differences in the substrate specificity
of MKK3 and MKK6. MKK3 preferentially activates p38␣
and p38␤, while MKK6 can activate strongly all known
p38 isoforms (44, 45, 60, 227).
The stress-activated MAP2Ks, as with all MAP2Ks, are regulated by Ser/Thr phosphorylation. The regulatory phosphoacceptor sites lie within a conserved region the P-activation loop of subdomain VIII of the kinase domain. The
(human) MEK1 phosphoacceptor sites are Ser218 and Ser
222; those for MEK2 are Ser222 and Ser226. The human
MKK4 phosphorylation sites are Ser257 and Thr261, those
for MKK7 are Ser206 and Thr210, those for MKK3 are
Ser189 and Thr193, and those for MKK6 are Ser207 and
Thr211 (44, 60, 71, 182, 201, 227, 229, 252, 293, 329,
343).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
Thus MEKK1, in a manner dependent on the MEKK1
PHD/RING domain and phosphorylation of the MEKK1
polypeptide in its activation loop (Thr1381), can associate
with the E3 ligase Itch, resulting in activation of Itch. This
association occurs apparently independently of JNK (which
can also activate Itch, Ref. 92; see sect. IID2) and of any
discernable MEKK1-catalyzed phosphorylation of Itch
(73). This interaction is necessary for TCR induction of IL-4
and IL-6. Evidence suggests that in vivo MEKK1-dependent
ubiquitination and degradation of JunB impairs IL-4 production and consequent TH2 differentiation. Thus MEKK1
negatively regulates TH2 cytokine production (73). In contrast, Itch, downstream of MEKK1, and/or JNK may also
trigger Itch-dependent c-Jun ubiquitination in vivo, contributing to TH2 tolerance and reduced airway inflammation (302)
(FIGURE 4C). It should be noted that these studies of T-cell
polarization employed a MEKK1 knockout strategy that
deleted solely the MEKK1 catalytic domain, leaving the
remainder of MEKK1 intact. The resulting expression of
MEKK1 fragments could lead to off target effects.
MEKK1 also contains a SWI2/SNF2 and MuDr (SWIM)
domain (AAs 333–361, FIGURE 4). The function of this
domain is unclear. However, there is some indication that
FIGURE 4. Stress/inflammation-regulated MAP3Ks. A: complex regulation of MEKs by members of the
MEKK family. Some MEKKs may also regulate NF-␬B via phosphorylation of I␬B kinase (IKK). B: same as A,
except these are distinct MAP3Ks that are not considered members of the MEKK group. C: parallel functions
of MEKK1 in ubiquitination. MEKK1, through its RING domain, can apparently directly ubiquitinate ERK1
fostering its degradation. MEKK1 also binds the E3 ligase Itch through which it can foster ubiquitination and
degradation of JunB (as well as c-Jun as indicated in FIGURE 2B). This MEKK1/JNK/Itch-dependent
degradation of c-Jun and JunB is prominent in T lymphocytes and blunts TH2 cell functions. This contrasts with
MEKK1/Itch-dependent c-Jun ubiquitination, which is important to TH2 cell tolerogenesis (FIGURE 2C).
D: regulation of Tpl-2 by IKK and p105. Tpl-2 is restrained by the NF-␬B regulatory subunit p105. IKK
phosphorylates and triggers the ubiquitin-dependent degradation of p105. Tpl-2 is liberated and autophosphorylates, becoming active.
700
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
main with intrinsic E3 ubiquitin (Ub) ligase activity (AAs
438 – 486). This activity assembles Lys48-linked polyubiquitin chains exclusively and can promote autoubiquitination as well as the ubiquitination of ERK2, a reaction which
may attenuate Ras-dependent ERK activation and JNK activation (182). MEKK1 can also directly ubiquitinate and
degrade c-Jun, in vitro and, apparently, in situ, in a manner
dependent on its intact PHD/RING domain (MEKK1 disruption or mutagenesis of the PHD/RING domain Cys433
to Ala abolishes c-Jun ubiquitination) (331). This apparent
direct MEKK1-dependent c-Jun degradation is linked to
stress-induced apoptosis. Notably, MEKK1 ubiquitination
of c-Jun occurred independently of prior c-Jun Ser63/73
phosphorylation, in contrast to earlier studies (83, 204) in
which c-Jun ubiquitination/degradation correlated with cJun phosphorylation. However, as discussed below and in
section IID2, the MEKK1 PHD/RING domain also binds
the E3 Ub ligase Itch. Accordingly, it is somewhat unclear if
MEKK1 directly or indirectly ubiquitinates c-Jun.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
the SWIM domain may be involved also in E3 ligase activity. Thus MEKK1-related protein X (MEX) is an E3 ligase
structurally similar to the PHD/RING-SWIM domain region of MEKK1 (MEX does not have a kinase domain).
Structure function studies of MEX have revealed that the
SWIM domain is essential to MEX’s E3 ligase activity, suggesting that this motif in MEKK1 may play a similar role
(210).
B) MEKK-2 AND -3: MORE PROMISCUOUS ACTIVATORS OF THE JNKS,
P38S, AND ERKS.
MEKK-2 and -3 (MAP3K-2 and -3, respectively) are essentially a small MEKK subgroup sharing a
closer sequence similarity within their kinase domains
(⬎90% identity) than with that of MEKK1 (⬃65% identity). MEKK2 and MEKK3 are also considerably smaller
than MEKK1 (⬃70 and 71 kDa, respectively). In addition,
the noncatalytic portions of MEKK2 and -3 are quite similar. The MEKK2 and MEKK3 catalytic domains are
COOH terminal (AAs 362– 619 for MEKK2, AAs 368 –
626 for MEKK3). The NH2-terminal noncatalytic portions
of MEKK-2 and -3 contain no motifs indicative of function
or regulation (16, 71) (FIGURE 4A).
Upon overexpression, MEKK-2 and -3 each can activate the
ERK pathway (via MEK1) and the JNKs and p38s
(through, respectively, MKKs-4/7 and 3/6) (16, 55, 71, 72).
Indeed, genetic studies suggest that MEKK3 is a physiological MKK6-kinase (120). Biochemical and some genetic
studies have also shown that MEKK3 can, in response to
TNF, activate the NF-␬B pathway via direct activation of
the I␬B kinases (IKKs) (120) (FIGURE 4A, TABLE 1).
MEKK-2 and -3 have been implicated in Treg and TH17 cell
differentiation (29) (sect. IVC4), and MEKK3 is important
in cardiac development (340) (sect. IVB3).
The ⬃150kDa MEKK4 (MAP3K4, TABLE 1), like MEKK1, consists of
a COOH-terminal kinase domain (AAs 1337–1597) and an
extensive NH2-terminal regulatory region with numerous
potential protein interaction motifs. Among these are a putative binding site for SH3 domain-containing proteins
(AAs 27–38) (95, 283), a region required to bind proteins of
the growth arrest and DNA damage (GADD)-45 family
C) MEKK4: ACTIVATION OF JNK AND P38 PATHWAYS.
3. Other MAP3Ks
A) TAK1 RECRUITS JNK, P38, AND NF-␬B IN RESPONSE TO BOTH
PROINFLAMMATORY CYTOKINES AND TGF-␤ FAMILY CYTOKINES.
The ⬃60-kDa TGF-␤-activated kinase-1 (TAK1, MAP3K7)
consists of an NH2-terminal kinase domain (AAs 30 –294)
preceded by a short regulatory motif (AAs 1–22) and followed by a COOH-terminal extension (AAs 295–557), the
function of which is unclear. The short NH2-terminal motif
appears to serve an inhibitory role (336).
Two hybrid screening with TAK1 AAs 1–22 as bait identified TAK1 binding proteins (TAB)-1 and -2 were identified
in two hybrid screens that employed TAK1 AAs 1–22 as
bait. These two regulatory proteins along with a third TAB
protein (TAB3) are essential for coupling TAK1 to upstream signals (see section IIID) (34, 127, 134, 207, 266,
281, 304). Endogenous TAK1 is activated by TGF-␤,
IL-1, TNF, and PAMPs. In vitro and upon overexpression in situ, TAK1 can phosphorylate and activate
MKK-3, -4, and -6. TAK1 can also phosphorylate and
activate the IKKs, and there is considerable evidence that
TAK1 is a physiologically relevant IKK activator (127,
140, 201, 207, 304, 332, 336) (FIGURE 4B).
TAs-2 and -3 confer a unique mode of regulation on TAK1
wherein the enzyme can be activated by free and anchored
Lys63-linked polyubiquitin chains synthesized in a stimulus-dependent manner by members of the TNF receptorassociated factor (TRAF) family of E3 ubiquitin ligases
(140, 304, 332) (sect. IIID, FIGURE 6B). Genetic studies of
TAK1 suggest a prominent role for TAK1 in the recruitment
of NF-␬B by stimuli that recruit TRAF proteins (140, 287,
303, 304, 332). A physiologically relevant role for TAK1 in
JNK and p38 activation is likely as well, at least in certain
cell types (287, 303) (FIGURE 4B, sect. IVC1).
B) ACTIVATION OF BOTH THE JNKS AND THE P38S BY ASK1 AND ASK2.
Apoptosis signal-regulating kinases-1 and -2 (ASK1, also
called MAP3K5, MAPKKK5; ASK2 is also called
MAP3K6; TABLE 1) are a pair of very closely related ⬃150kDa polypeptides. The ASKs contain centrally located kinase domains (ASK1 AAs 677–936) as well as NH2-termi-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
701
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
MEKK1 can activate MKK4 and MKK7, and through
these, the JNKs. MKK4 is the preferred MEKK1 substrate,
and accordingly, MEKK1 is a relatively selective activator
of the JNKs (FIGURE 4). However, in some instances,
MEKK1 can also activate the ERKs and p38 (sect. IIID1,
TABLE 1) (338). MEKK1 can be activated, in cultured cells,
by TNF (10), microtubule poisons, oxidant stress (198),
and certain receptor Tyr kinase agonists (76). Of particular
note, MEKK1 is activated in a complex manner by the B
lymphocyte receptor protein CD40. Activation of MEKK1
by CD40 is key to CD40 recruitment of MAPKs (190) (sect.
IIID1).
(AAs 147–250) (284), and a putative PH domain (AAs 225–
398) (95, 283). The MEKK4 kinase domain shares ⬃55%
amino acid homology with that of MEKK-1, -2, and -3.
There is some uncertainty as to the precise substrate spectrum for MEKK4, with results suggesting that MEKK4 is
selective for MKK4 and the JNKs in vivo and in vitro (95).
In contrast, others have shown that MEKK4 can activate
MKK4 as well as MKK-3 and -6 (283). Still others have
reported that MEKK4 is selective for the p38 pathway in
vivo (35). The reason for these discrepancies is unclear.
MEKK4 may be an effector for stress pathways activated by
genotoxins and has been linked to TH1 cell maturation (see
sect. IVE3).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
nal (ASK1 AAs 1– 676) and COOH-terminal (ASK1 AAs
937–1375) extensions, both of which have been implicated
in binding polypeptides of the TNF receptor-associated factor (TRAF) family. The NH2-terminal extensions also bind
redox sensing enzyme thioredoxin. The ASKs can activate
MKK-4, -3, and -6 (125, 179, 211, 245, 309) (FIGURE 4B).
Endogenous ASK1/2 are activated by oxidant stress as well
as by TNF. Indeed, the activation of ASK1/2 by TNF is
dependent on TNF-induced generation of reactive oxygen
intermediates (ROIs) (100, 125, 179, 211, 245, 292).
ASK1/2 are likely to be an important effectors coupling
these agonists to the JNKs and p38s.
Despite their close structural similarities and the dependence of ASK2 on ASK1, ASK1 and ASK2 perform specific,
nonredundant functions in vivo where they regulate distinct
aspects of cell survival and inflammation in tumorigenesis
(126) (see sect. IVD7).
C) TPL-2, AN IMPORTANT PROINFLAMMATORY ERK PATHWAY ACTIVATOR AS WELL AS AN ACTIVATOR OF THE JNK AND P38 PATHWAYS.
Tumor progression locus-2 (Tpl-2), also called Cot and
MAP3K8 (TABLE 1), is a 58-kDa protein Ser/Thr kinase
with a central kinase domain (AAs 139 –394), similar to S.
cerevisiae STE11, a COOH-terminal regulatory domain
(AAs 395– 467) and an NH2-terminal domain, of unknown
function, which is truncated in some mRNA splicing isoforms to yield a 52-kDa form (25, 219). Tpl-2 was originally identified as an oncogene activated by insertional mutagenesis. Thus moloney leukemia virus proviral insertions
at Tpl-2, which always target the last intron of the gene,
elicit the enhanced expression of a COOH terminally truncated, constitutively active protein the expression of which
is enhanced, both through elevated transcription and
mRNA stabilization. It is likely, therefore, that the COOHterminal regulatory domain of Tpl-2 exerts a negative effect
on Tpl-2 activity (25).
When transiently overexpressed, Tpl-2 activates the JNKs
and ERKs with equal potency (25, 249), and Tpl-2 can
directly activate MEK1 and MKK4 in vitro and in transfected cells (249). However, ERK activation is likely to be
more physiologically relevant (69) (FIGURE 4B).
702
D) THE MIXED LINEAGE KINASES ARE POTENT AND SELECTIVE ACTIVATORS OF JNK AND P38 PATHWays.
The mixed lineage kinases
(MLKs) are a small family of protein Ser/Thr kinases with a
common structural configuration. An NH2-terminal kinase
domain is followed by one to two leucine zippers, a Cdc42/
Rac interaction and binding (CRIB) domain, and a COOHterminal proline-rich domain with several consensus SH3
binding motifs. The MLKs are clearly Ser/Thr specific. Despite this, as their names suggest, the kinase domains of the
MLKs contain structural features found in both Ser/Thr and
Tyr kinases. Four MLKs have been well described and are
clear activators of MAPK pathways: MLK1, MLK2 (also
called MKN28 cell-derived Ser/Thr kinase, MST, TABLE 1),
MLK3 (also called SH3 domain-containing proline-rich kinase, SPRK or protein Tyr kinase-1, PTK1), and dual leucine zipper kinase (DLK, also called MAPK upstream kinase, MUK, or zipper-containing protein kinase, ZPK; TABLE 1). In addition to the common features described above,
MLK2 and MLK3 contain SH3 domains NH2 terminal to
their kinase domains (68, 89, 90, 114).
MLK-1, -2, and -3 as well as DLK (MAP3K-9, -10, -11,
and -12, respectively) are potent activators of MKK4 and
especially MKK7, and through these, the JNKs. MLK3
can also activate the p38s via activation of MKK3 and
MKK6 (FIGURE 4B). The ERK pathway is not a strong
MLK target, although MEK1 can function as a MLK
substrate in vitro (90, 109, 110, 230, 291). MLKs may
function in innate immunity and metabolic regulation
(see sect. IV, C1 and E5).
E) TAO KINASES SHARE STRUCTURAL SIMILARITY TO BOTH STE20 AND
MAP3KS YET FUNCTION AS MAP3KS SELECTIVELY UPSTREAM OF THE
P38S.
The thousand-and-one (TAO) kinases (TAO1 and
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
ASK2 is more restricted in its expression than ASK1, with
ASK2 expressed primarily in the skin and gut epithelium
while ASK1 is ubiquitously expressed. Evidence indicates
that ASK-1 and -2 function as a heteromer. Indeed, disruption of ask1 leads to constitutive degradation of ASK2; thus
ASK2 cannot function biochemically in the absence of
ASK1. ASK2’s requirement for ASK1 does not depend on
ASK1’s kinase activity, and indeed, ASK2 can be stabilized
by kinase-dead ASK1. In contrast, ASK1 can function alone
(126, 282) (see also sect. IVD7).
Activation of Tpl-2 in macrophages by TNF and PAMPs
follows an unusual mechanism. In resting cells, the Tpl-2
58- and 52-kDa isoforms exist as a heteromer in complex
with the p105 NF-␬B1 subunit and a second protein, A20binding inhibitor of NF-␬B2 (ABIN2). Administration of
agonist (TNF, LPS) results in phosphorylation of p105 at
Ser931, Ser923, and Ser932, a reaction catalyzed by the
I␬B-kinase-␤ (IKK␤). A separate poorly understood kinase
phosphorylates the 58-kDa form of Tpl2 at Thr290. Phosphorylation of p105 results in ubiquitiation and proteasomal degradation of p105, dissociation of ABIN2, and
liberation of Tpl2 p52 and p58 heteromers, which autophosphorylate at Ser62 becoming active. ABIN2 stabilizes
inactive Tpl-2, but Thr290 and Ser62 phosphorylation of
Tpl-2, and consequent dissociation of ABIN2, while resulting in Tpl-2 activation, also renders Tpl-2 susceptible to
subsequent ubiquitination and proteasomal degradation,
thus providing for feedback regulation of this pathway (12,
218, 273) (FIGURE 4C). Genetic disruption studies reveal
cell-dependent Tpl-2 functions, notably in TRAF-dependent activation of ERK (69) (sect. IVC1).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
Despite their similarities to the Ste20/GCKs, the TAOs are
direct, specific activators of MKK3, and through MKK3,
the p38s (124) (FIGURE 4B). The ability of TAO1 to catalyze
directly the activation of MAP2Ks may be due to the kinase
domain homology with MLK2 in the substrate binding region. Of note, the remarkable in situ selectivity of the TAOs
for MKK3 may be due to the presence of a specific MKK3
docking site (AAs 315– 451 on TAO2) (32). Interestingly,
this region is rich in acidic amino acid residues, a property
shared wtih MAPK CD loops which bind to basic MAPK
interaction motifs on MAPK regulators and substrates (see
sect. IID3) (15, 288, 289).
III. SIGNALING COMPONENTS THAT
RECRUIT STRESS-ACTIVATED MAPK
CORE PATHWAYS
Sections I and II touched on the subject of mammalian
stress-activated MAP3K-MAP2K-MAPK core signaling
modules. Here we describe some of the known proximal
signaling components thought to couple these core modules
to upstream regulators.
A. Rho Family GTPases Can Activate the
JNKs and p38s
Members of the Ras superfamily of monomeric GTPases
are key proximal activators of signal transduction pathways. Ras itself is central to the activation, by mitogens, of
the ERK pathway. The regulation of Ras by receptor tyrosine kinases, and the consequent recruitment by Ras of
the Raf family of MAP3Ks have been extensively reviewed
(23, 192, 201, 203). The paradigm of Raf regulation points
to three crucial events governing MAP3K activation:
1) binding to an upstream activating protein and consequent membrane translocation, 2) phosphorylation, and
3) homoligomerization (236, 317, 350). These steps may be
important in the regulation of MAP3Ks recruited by stress
and inflammation.
At least some stress-activated MAPK core pathways are
putative targets of members of the Rho subfamily of the Ras
superfamily. Mammalian Rho family GTPases include the
Rho (RhoA-E), Rac (Rac1 and -2), and Cdc42 (Cdc42Hs,
G25K, Tc10, and Chp) groups. As with all Ras superfamily
members, Rho subfamily GTPases are active in the GTPbound state and inactive in the GDP-bound state (130).
Rac1 and Cdc42 are Ras effectors and may link JNK or p38
to agonists that recruit Ras, including some mitogens (130).
When overexpressed, constitutively active, GTPase-deficient (Val12) mutants of the Rac and Cdc42 groups, but not
the Rho group, are vigorous activators of the JNKs and
p38s (130). Consistent with these findings, expression of
the Dbl family Rho guanine nucleotide exchange factors
that recruit Rac and Cdc42, also results in strong JNK activation (130). The so-called Cdc42/Rac interaction and
binding (CRIB) domain, which interacts directly with the
effector loops of Rac1 and Cdc42Hs, is present in most, but
not all, Rac and Cdc42 targets (20). These include the
MAP3Ks MEKK4 and MLK-2 and -3. However, the precise
effectors that couple Rho GTPases to JNK and p38 remain
to be identified unambiguously through genetic studies.
The CRIB motifs of the MAP3Ks MLK2 and MLK3 enable
these MAP3Ks to interact directly with Cdc42Hs and Rac1
in a GTP-dependent manner in vitro, a property not demonstrated for MEKK4. However, although MLK2 and -3
may be Cdc42Hs effectors (below), it is unlikely that they
are direct Rac targets. In support of this, mutation of the
effector loop of Rac1 Phe37 to Ala has no effect on recruitment of the JNK pathway in situ. However, neither MLK3
nor MLK2 can interact with Phe37Ala-Rac1 in vivo (167,
290). As discussed below, Phe37Ala-Rac1 can bind the
scaffold protein plenty of SH3 domains (POSH), which may
serve to couple Rac1 to the JNKs (290) (FIGURE 5).
In contrast, MLK3 may be a Cdc42Hs effector. The leucine
zipper of MLK3 (see sect. IIF3, AAs 400 – 487) is critical for
homodimerization, which, in turn, is central to MLK3’s
ability to signal downstream to MKKs (89, 175). Of particular importance, coexpression of MLK3 with Val12Cdc42Hs significantly increases MLK3 to homodimerization (175), suggesting that GTP-Cdc42Hs binds and promotes MLK3 (or MLK2) homodimerization and activation
(FIGURE 5).
B. Several Scaffold Proteins Regulate
the JNKs
Insofar as mammalian cells express multiple different
MAP3Ks and additional upstream regulators of the JNKs
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
703
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
TAO2) are a pair of 1,001-amino acid Ser/Thr kinases.
Interestingly, the TAO kinase domains are structurally similar to S. cerevisiae STE20, a proximal kinase thought to be
important in the regulation of the Ste11p-Ste7p-Fus3p/
Kss1p yeast mating pheremone MAPK core signaling module (32, 108, 124). Not surprisingly, the TAO kinase domains are also significantly similar to mammalian Ste20s of
the germinal center kinase group (43% identity with GCK)
(see sect. IIIC) (32, 50, 108, 124, 199). However, there is
also appreciable identity with the MLK2 kinase domain
(33% overall), especially within the substrate binding motifs (124). Prostate-derived STE20-like kinase (PSK, TABLE
1) is a putative splicing isoform of TAO2. It contains an
extended COOH-terminal tail but is otherwise identical to
TAO2 expressed at elevated levels in prostate tumors (199).
The TAO kinase domains are NH2 terminal and are followed by very large (700 AA) COOH-terminal extensions,
the functions of which are unclear.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
nase (GCK) homolog, but not GCK-related, another
GCK homolog (see sect. IIIC) (63, 323). The binding of
HPK1 is probably indirect, via the binding of HPK1 to
MLK3. The ability of JIP1 to potentially link HPK1 to a
complete JNK-activating core signaling module corresponds well with the idea of JIP1 as a true scaffold protein. JIP1 is heavily expressed in the brain. Disruption of
jip1 indicates a role in stress responses, notably apoptosis in
response to excitotoxic stimuli (324). JIP1 has also been
linked to cortical neuron axonal development (48); JNKs
also have been linked to the promotion of neuronal development (sect. IVA1).
and p38s, to maintain signaling specificity, efficiency, and
integrity, it is imperative that individual MAP3K-MAP2KMAPK core signaling components be organized into groups
that are both subject to regulation by select activators and
able to recruit select effectors. This sequestration is mediated by scaffold proteins. These proteins collect specific
MAP3K-MAP2K-MAPK core modules into organized signaling oligomers.
1. JIPs: a family of scaffold proteins that may
regulate JNKs by nucleating core MLK-MKK7JNK signaling complexes
JNK interacting protein-1 (JIP1), a ⬃60-kDa polypeptide,
was the first discovered member of a family of mammalian
scaffold proteins that regulate JNK signaling. The JIP1
NH2-terminal domain (AAs 143–163) selectively binds the
JNKs. The JIP1 COOH-terminal extension includes an SH3
domain (AAs 491– 600). Between these is a proline-rich
segment (AAs 281– 448) that contains several motifs that
conform to the consensus for SH3 binding sites (63).
JIP1 interacts with kinases at all three levels of JNK core
signaling modules. Thus JIP1 binds the MAP3Ks MLK3
and DLK (but not MEKK1), the MAP2K, MKK7 (but not
MKK4), and JNK1. It remains to be established whether
this binding is at all sequential or regulated in any way. The
JIP1 SH3 domain is required for MLK3 binding, while a
segment of the JIP central region (AAs 283– 471) that overlaps with the proline-rich region is needed for MKK7
binding. JIP1 also interacts with hematopoietic progenitor kinase-1 (HPK1), a mammalian germinal center ki-
704
JIP3 is a large polypeptide (⬃144 kDa) expressed selectively
in brain and heart. JIP3 consists of a leucine zipper motif
(AAs 392– 427) as well as several coiled-coiled motifs. JIP3
can bind all three JNK isoforms, but, as with JIP1 and JIP2,
binding to JNK1 is strongest. JIP3 binds neither ERKs nor
p38s. The JNK binding region spans AAs 202–213, a region
conserved with a segment of the JIP1 JNK binding domain
(AAs 153–164). This conservation suggests a consensus
JNK binding sequence (Arg-X-X-Arg-Pro-Thr-Ser/Thr-LeuAsn-Val/Leu-Phe-Pro, where X is any amino acid) (128, 148)
notably similar to the basic MAPK docking sites, present in
many MAPK substrates, that bind to MAPK CD motifs
(section IID3, TABLES 2 and 3) (15, 288, 289). Recent genetic studies indicate a critical role for JIP3 in brain development, again a function also linked to the JNKs (149)
(sect. IVA1).
JIP4 is a fourth member of the JIP family. It is structurally
similar to JIP3. Interestingly, however, JIP4 appears not to
regulate JNK. Instead, biochemical studies indicate that
JIP4 regulates the p38s in a mechanism that requires MKK3
and MKK6 (150).
2. POSHs
Plenty of SH3 domains (POSH) 1–3 (also called SH3 domain containing ring finger, SH3RF 1–3) comprise a family
of ⬃90-kDa polypeptides. POSH proteins each contain
four SH3 domains (AAs 139 –190, 198 –254, 457–511, and
838 – 892 for POSH1), several clustered polyproline SH3
binding sites (AAs 368 – 405 for POSH1), and a RING do-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
FIGURE 5. Two routes by which the Rho family GTPases Rac and
Cdc42 regulate MLKs-2 and -3. Rac functions through the POSH
adapter/E3 ligase, and Cdc42 directly recruits MLKs through their
CRIB motifs.
JIP2 is a second JIP family member. It is quite similar structurally and functionally to JIP1, and, like JIP1 is almost
exclusively localized in brain. Accordingly, JIP2 can bind in
situ with MLK-2 and -3 (but not MEKK family MAP3Ks),
MKK7 (but not MEK1, MKK3, MKK4, or MKK6) and
JNKs 1–3 (but not p38 or ERK). Both JIP1 and JIP2 preferentially interact with JNK1 over JNK-2 or -3, and both
JIP1 and -2 can in overexpression studies potentiate JNK
activation by coexpressed MLK3. In addition, JIP1 and JIP2
(endogenous and ectopically expressed) can homo- and heterodimerize. However, JIP1 binds all JNK isoforms more
strongly than does JIP2 (345).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
main with E3 ubiquitin ligase activity (AAs 12–52 for
POSH1). Interestingly, POSH proteins contain only partial
CRIB motifs (localized to AAs 292–362 in POSH1), but can
still interact directly with Rac1 (but not Cdc42Hs, Ras, or
Rho). Ectopically overexpressed POSH1 stimulates robust
JNK and NF-␬B activation as well as apoptosis (144, 290,
326).
The function of the POSH E3 ligase activity is obscure.
POSHs can autoubiquitinate, a function which may serve to
suppress POSH-dependent apoptosis (144). In this capacity, POSHs can trigger JNK activation and JNK-dependent
apoptosis, especially in neuronal cells where POSH proteins
may function in the pathogenesis of Parkinson’s disease
(144, 326).
3. Actin binding protein-280 (ABP-280), a 280-kDa
putative scaffold protein that may regulate TNF
and lysophosphatidic acid activation of melanoma
cell JNKs
The 280-kDa actin binding protein ABP-280 (also called
filamin) crosslinks actin filaments into orthogonal arrays,
thereby contributing to the structure of the cortical actin
meshwork (277). ABP-280 may also be a scaffold protein
for cytokine regulation of JNK.
In situ and in vitro, ABP-280, through its COOH-terminal
region (AAs 2282–2454), interacts with MKK4. MEK1,
ERK1, and p38␣ do not interact with ABP-280. MKK4, in
turn, indirectly recruits MEKK1 and JNK to ABP-280, and
these polypeptides can be isolated with ABP-280 only if
MKK4 is present (188). MKK4 can itself act as a scaffold
protein (see below) (330). Thus the formation of MEKK1MKK4-JNK-ABP-280 complexes may enable the recruitment of JNK by selective upstream stimuli. ABP-280 loss of
function mutations indicate that ABP-280 is important for
the activation of MKK4 and the JNKs in intact melanoma
cells, in response to certain specific extracellular stimuli,
notably TNF (188).
4. The intrinsic scaffold properties of MKK4
MKK4, in addition to phosphorylating and activating JNK
and p38, functions as a scaffold protein. The scaffolding
properties of MKK4 enable it to form selective, dynamic
complexes with both an upstream activator (MEKK1) and a
substrate (JNK). Thus MEKK1 interacts with the NH2 terminus (AAs 1–77) of inactive, dephosphorylated MKK4.
Of note, because MEKK1 does not readily interact in vivo
or in vitro with MAP2K-1 or -2, MKK-3 or -6 or with
MKK7, none of these MAP2Ks is as good an in vitro substrate of MEKK1 as is MKK4 (330).
Upon the addition of ATP to the MKK4-MEKK1 complex,
in vitro, MEKK1 catalyzes the phosphorylation and activation of MKK4. This, in turn, fosters the dissociation of the
MKK4-MEKK1 complex (343). The phosphorylated, active form of MKK4 (but not inactive MKK4) then enters
into a second specific complex, with JNK, again mediated
by the MKK4 NH2-terminal MEKK1 binding domain. Importantly, the MEKK interaction motif also contains a polybasic MAPK docking site that interacts specifically with
the JNK CD domain (15, 288, 289). This second complex
dissociates upon phosphorylation and activation of JNK
(330).
C. The Germinal Center Kinases/Map4ks as
Regulators of JNK and p38 Core
Signaling Pathways
1. General considerations
The germinal center kinases (GCKs) are a family of Ser/Thr
protein kinases that possess NH2-terminal kinase domains
distantly related to that of S. cerevisiae Ste20p (TABLE 1). Of
note, however, the kinase domain of Ste20p (as well as
those of the more closely related mammalian PAKs) is
COOH terminal. Some members of the GCK family are
selective activators of the JNKs and p38s (50, 58). The
founding mammalian member of the GCK family is GCK
itself.
The kinase domains of the GCKs are followed by extensive
COOH-terminal regulatory domains (CTDs) (50). The
GCKs are effectively assembled according to sequence similarity of the catalytic domains into seven subfamilies that
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
705
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Effector loop mutants of Rac1 recruit POSH1, a process
that correlates tightly with the relative ability of these Rac1
mutants to activate the JNKs (FIGURE 5). Thus POSH1,
unlike MLK-2 and -3, can interact with Phe37Ala-Rac1, a
Rac1 effector mutant that does not provoke lammelopodium formation but does activate coexpressed JNK (167,
290). In contrast, POSH1 cannot interact with a second
Rac1 effector loop mutant, Tyr40Cys Rac1, which does not
activate the JNKs in situ (168, 303). Thus, although the
MLKs, which contain complete CRIB motifs, the POSHs,
but not the MLKs, are likely direct effectors for Rac1 activation of the JNKs, the MLKs may be effectors for
Cdc42Hs. Cdc42Hs does not interact with POSH proteins.
Still, the SH3 domains of POSH may function as scaffold
proteins, binding MLKs, that contain SH3 binding sites.
The proinflammatory adapter protein (sect. IIID) TNF receptor-associated factor-2 (TRAF2) can also interact with
ABP-280 through a process that requires the RING and Zn
finger domains of TRAF2. As with TNF activation of JNK,
activation of JNK by ectopically expressed TRAF2 is also
attenuated in melanoma cells bearing ABP-280 loss of function mutants (173). This finding fits with the idea that ABP280 assembles a TNF responsive signaling complex in these
melanoma cells.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
The CTDs of those GCKs that recruit the JNKs and p38s
are broadly similar, typically including at least two proline/
glutamic acid/serine/threonine (PEST) motifs and two or
more polyproline putative SH3 domain binding sites. The
CTDs of JNK/p38-activating GCKs also include a conserved citron homology domain (CNH) that resembles a
similar region in Citron, a conserved metazoan kinase involved in the control of the actin cytoskeleton (50, 58, 64,
82, 119, 139, 146, 154, 261, 279, 296, 344).
GCKR, GLK, NIK and, possibly, TNIK are all activated in
situ by TNF (64, 82, 224, 261, 344). Inasmuch as GCKs 1)
potently activate the JNKs/p38s and 2) stably associate with
core signaling module elements and their upstream activators, it is reasonable to propose that the GCKs coordinately
regulate MAPK core signaling modules in conjunction with
additional upstream elements.
2. The group I (GCK, GCKR, GLK, and HPK1) and IV
(NIK, TNIK, NRK, and MINK1) subfamilies of
GCKs/Map4ks each can activate JNK and p38
A) GCK.
Transient overexpression of GCK (Map4k2, TABLE
results in robust activation of JNK and p38. Two immediate downstream targets of GCK are MLK2 and -3. In
vitro, GCK can activate MLK3 and MEKK1 (27).
1)
B) GCKR.
The primary sequence of GCKR (Map4k5, also
called kinase homologous to STE20, KHS, TABLE 1) is
⬃60% identical to that of GCK, especially within the CNH
motif of the CTD (50, 58, 261, 296). GCKR is a highly
selective activator of the JNKs and, when ectopically expressed does not activate the ERKs, p38s or NF-␬B (261,
296).
TNF, CD40, and ultraviolet radiation all can activate endogenous GCKR (261, 262), and there is evidence that
GCKR is an important effector for TNF, at least in some cell
706
types. Thus GCKR antisense constructs, when expressed in
293 cells, strongly block TNF activation of the JNKs (261).
Interestingly, RNAi studies have also revealed a role for
GCKR in both canonical and noncanonical Wnt signaling,
at least in B lymphocytes (263). Wnt agonists bind to a class
of seven transmembrane receptors in the Frizzled family
and play an important role in cell fate decisions and oncogenesis (see also section IVB1). Canonical Wnt signaling
involves receptor mediated stabilization of ␤-catenin. This
stabilization requires signal-dependent inhibition of GSK3catalyzed phosphorylation of ␤-catenin and enables
␤-catenin to accumulate and migrate to the nucleus where it
dimerizes with TCF transcription factors to regulate gene
expression (99).
Noncanonical Wnt signaling is mediated by the Frizzledlinked adapter protein Disshevelled and triggers JNK activation. Mechanisms controlling this process are poorly understood. Disruption or silencing of gckr substantially impairs canonical and noncanonical Wnt3a activation of
JNK, GSK3-␤ phosphorylation/inhibition, and consequent
stabilization of ␤-catenin (263). This result suggests that
GCK has functions outside of JNK/MAPK regulation.
C) GLK. As with GCK and GCKR, GLK (Map4k3, TABLE 1)
is a potent and selective activator of the JNKs. Moreover,
the GLK CTD, especially within the CNH motif, is strikingly similar to those of GCK and GCKR. Endogenous
GLK is apparently activated in situ by TNF (64); however,
the physiological functions of endogenous GLK are unknown.
D) HPK1. While GCK, GCKR, and GLK share close structural
similarity, HPK1 (Map4k1, TABLE 1), a fourth group I
GCK, while sharing the broad structural features of JNK
and p38-activating GCKs, is structurally more distant from
GCK than is GCKR and GLK. The CTD of HPK1 includes
four consensus SH3 domain binding sites, the COOH terminal, most of which interacts with the SH3 domain of
MLK3 (sect. IIF3). Ectopically expressed HPK1 vigorously
activates the JNKs while activating p38 much more weakly.
No ERK activation is observed upon HPK1 overexpression
(50, 58, 119, 154).
The second and fourth SH3 binding motifs of HPK1 enable
HPK1 to bind to the closely related SH2/SH3 adapter proteins Crk and CrkL, an interaction that enables the EGFstimulated recruitment of HPK1 to the EGF receptor, at
least in transfected cells. The Crk(CrkL)-HPK1 interaction
may be physiologically relevant insofar as coexpression of
Crk or CrkL with HPK1 activates the kinase activity of
HPK1 and, in situ triggers synergistic JNK activation (176).
Although ectopically expressed HPK1 can interact with
MLK3 in situ, physiological effectors linking HPK1 to
MAPKs have not been identified genetically or through
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
within each also show marked similarities in their COOHterminal noncatalytic segments; moreover, each of these
subfamilies has a single homolog in Drosophila or Caenorhabditis elegans (50, 58). The mammalian members of the
group 1 GCK subfamily, i.e., GCK, GCKR(KHS), GLK,
and HPK1, have each been shown to act as upstream activators of JNK in some context. In addition, the group IV
GCK subfamily, NIK, TNIK, and NRK, also function as
upstream activators of JNK (TABLE 1) (50, 58, 64, 82, 119,
139, 146, 154, 261, 279, 296, 344). Numerous studies
indicate that those GCKs that recruit the JNKs/p38s do so
by recruiting MAP3Ks (27, 64, 82, 119, 139, 146, 154,
261, 279, 296, 344). Accordingly, members of the GCK
family have been designated MAPK-kinase-kinase-kinases
(Map4ks) by Human Genome Project (TABLE 1), in spite of
the fact that little clear evidence exists showing that GCKs
can activate MAP3Ks by phosphorylation (27).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
RNAi (154). HPK1 may also have several non-MAPK-related functions relevant to the negative regulation of TCR
signaling (sect. IVC1).
E) NIK. NIK (Map4k4, also called HPK1/GCK-like kinase,
HGK, TABLE 1) and its close relatives TNIK (below) and
Nrk cluster into the group IV GCK subfamily, a subfamily
distinct from the group I GCKs and most closely related to
the Drosophila kinase Misshapen. However, as with the
group I GCK subfamily, transient overexpression of NIK
strongly activates the JNKs (279, 344).
F) TNIK. TRAF2 and Nck-interacting kinase (TNIK) is
⬃90% identical, at the amino acid level within the kinase
domain (AAs 1–306), ⬃90% identical within the Leu-rich
and CNH motifs (AAs 1019 –1360), and ⬃53% within the
remainder of the CTD to NIK. Predictably, therefore, TNIK
overexpression results in robust and selective JNK activation. As also expected, TNIK can interact with Nck in situ,
although the functional significance of this interaction is
unclear. TNIK also interacts with coexpressed TRAF2.
That said, there is no evidence that TNIK is recruited by
stimuli that signal through TRAF2 (82).
D. Adapter Proteins That Couple to TNF
Receptor Family Receptors and
Receptors of the IL-1R and TLR Group
Are Crucial Regulators of the JNKs
and p38s
Upon binding agonists, transmembrane receptors of the tumor necrosis factor family (the TNFR family) and the IL-1
receptor/TLR family, as well as cytosolic receptors of the
nuclear binding/oligomerization domain-leucine-rich repeat (NOD-LRR) and retinoic acid-inducible gene (RIG)like-helicase (RLH) families provoke a suite of diverse inflammatory responses. As such, these receptors are crucial
to immune cell development, innate and acquired immunity, and the pathogenesis of a number of diseases such as
cancer and type 2 diabetes mellitus (see section IV). Thus
these receptor families are among the most important acti-
The TRAFs are critical proinflammatory signal transducing
adapter proteins. As such, most proinflammatory receptor
signaling pathways converge on one or more of the seven
known TRAF proteins which, in turn, recruit key downstream signaling components. Genetic studies indicate that
TNF signaling to MAPKs requires TRAF2, while IL-1 and
PRR signaling to MAPKs requires TRAF6. TRAF2 also
relays endoplasmic reticulum stress signals to the JNKs (7,
103, 222).
The TRAFs consist of two closely spaced COOH-terminal
TRAF domains (TRAF-N followed by TRAF-C), a central
zinc finger motif and, with the exception of TRAF1, an
NH2-terminal RING finger motif. The TRAF domains are
responsible for binding to a subset of TRAF effectors and
for binding upstream activator adapter proteins such as
TNF receptor associated death domain (TRADD) for the
TNF pathway, and myeloid differentiation factor-88
(MyD88) for the TLRs. The function of the Zn finger domains is unclear; however, the RING fingers possess E3
ubiquitin ligase activity and are important for the activation
of downstream effectors (7, 103, 213, 222).
A recent series of elegant studies has identified the a major
mechanism by which TRAF proteins signal. The RING domains of TRAFs enable autoubiquitination, generating
poly-Ub chains linked at Ub-Lys63. Of particular importance, TRAFs can also generate free Lys63-linked poly-Ub
chains which act as second messengers recruiting the TAB2TAK1 complex. This complex is required for efficient activation by IL-1 and PAMPs of NF-␬B and, at least in some circumstances, JNK and p38 (140, 222, 304, 332) (FIGURE 6B).
1. In B cells, the TNF receptor family member CD40
recruits TRAF2 and TRAF3 into a large complex
that activates MEKK1. In response to toll-like
receptor agonists, MEKK1 may also be activated
by partial proteolytic cleavage coupled to the
adapter protein ECSIT
CD40 is a receptor present principally on B cells, where it
acts to trigger B cell maturation and activation upon engagement by CD40 ligand (CD40L), a membrane-associated agonist present on activated T cells. CD40 is structurally similar to TNF receptors and recruits TRAF proteins
for signal transduction. Matsuzawa and colleagues (190)
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
707
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
NIK contains two consensus SH3 binding motifs (AAs
574 –579 and AAs 611– 616), both of which can interact
with the SH3 domains of the SH2/SH3 adapter protein Nck
(279). The significance of the interaction between NIK and
Nck is unclear. Overexpression studies suggest a link between NIK/Nck and the Eph Tyr kinases in the recruitment
of JNK (11). In addition, given that Nck is thought to be
involved in the regulation of the actin cytoskeleton, one
function of NIK may be to regulate cell motility and shape.
Indeed, Drosophila Misshapen is required for embryonic
dorsal closure (280). NIK can be activated by TNF (344).
Misshapen is an effector for Drosophila TNF receptor-associated factors (TRAFs; sect. IIID), suggesting a similar
role for NIK (178, 262, 346).
vators of the JNKs, p38s. Although widely divergent structurally, the basic mode of signaling for these receptor families is quite similar; these receptors do not possess intrinsic
enzymatic activity relevant to signaling. Instead, the receptors signal by recruiting intracellular adapter proteins (19,
40, 147, 151, 195). The repertoire of adapter proteins used
by these receptors, as well as the spatial and temporal association of these adapters and the downstream effectors they
target are complex, cell specific, and still incompletely understood.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
708
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
showed that CD40 engagement triggers the formation of a
large complex containing TRAF2, TRAF3, the ubiquitin
(Ub) conjugating enzyme Ubc13, the E3 Ub ligases cellular
inhibitor of apoptosis proteins cIAP1 and cIAP2, the I␬Bkinase (IKK) regulatory scaffolding subunit IKK␥, and
MEKK1. TRAF2, Ubc13, and IKK␥ are required for complexing with MEKK1 and subsequent CD40 activation of
ERK, JNK, and p38; however, MEKK1 activation does not
occur without CD40 stimulus-dependent dissociation of
the complex from CD40 itself, a process that requires prior
cIAP1/2-dependent degradation of TRAF3 (FIGURE 6C).
This complex mode of signaling may enable spatial and
temporal regulation of signal transduction. Indeed, CD40
recruits other TRAFs, notably TRAF6, for activation of
NF-␬B (88, 190).
2. GCKs may function with TRAFs
to activate MAP3Ks
As discussed in section IIIC2, GCK, an effector for TRAF6,
and an activator of MLK2 and -3, is recruited by PRRs. In
contrast, GCKR, GLK, NIK, and TNIK are putative TNF
effectors (64, 261, 344). Accumulating evidence indicates
also that PRRs and TNFRs link to GCK, GCKR and, possibly, NIK and TNIK through the TRAFs (sect. IIIC) (82,
262, 350, 351). Thus TNF activates GCKR in a TRAF2dependent manner and antisense GCKR constructs blunt
TNF activation of 293 cell JNK (261, 262).
GCKR can bind TRAF2 in situ while GCK interacts with
TRAF6. These interactions require the CNH domains of
both kinases (82, 262, 346). GCKR is activated upon coexpression with TRAF2, a process that requires the TRAF2
3. ASK1 is an effector for TRAFs-specifically, a
ASK1 is a TNF effector recruited by TRAF2: the
role of cellular redox in ASK1 regulation
ASK1 and the closely related ASK2 are MAP3Ks that function upstream of the JNKs (through activation of MKK4)
and the p38s (through MKK3 and MKK6). TNF activates
the MAP3K activity of ASK1 (125, 126, 282) (sect. IIF3).
Upon ectopic expression, ASK1 interacts in situ with recombinant TRAF2, -5, and -6, and TNF stimulates a transient interaction between endogenous TRAF2 and ASK1 in
situ. Recombinant TRAF2 can also activate coexpressed
ASK1 (179, 211).
The regulation of ASK1 by TRAF2 is redox dependent.
Thus the TNF-dependent association of endogenous ASK1
and TRAF2 is efficiently attenuated by free radical scavengers (179). ASK1 is also inhibited endogenously by the redox sensing oxidoreductase thioredoxin (Trx). However,
for this inhibition to occur, Trx must be in a reduced state.
Thus, upon treatment of cells with oxidant stresses (H2O2),
and consequent Trx oxidation, ASK1 dissociates from Trx
and binds TRAF2. The exchange of Trx for TRAF2 triggers
activation of ASK1 (245) (FIGURE 6D). TNF can elicit the
production of ROIs, and JNK activation by coexpressed
TRAF2 is partially reversed by free radical scavengers
(211). Inasmuch as expression of TRAF2, but not a mutant
TRAF2 construct missing the RING effector domain, triggers the production of ROI (179), the E3 ligase activity of
TRAF2 may contribute to ROI production (245).
4. TAK1 is a target for TGF-␤ through its association
with TAB1, and for PRRs, TNF and IL-1 through
its association with TAB2
TAK1 binding protein-1 (TAB1), TAB2, and TAB3 are regulatory subunits that bind and foster activation of TAK1
and its downstream effectors, the JNKs and p38s (34, 127,
134, 140, 266, 281). TAB1 and TAB2 are structurally quite
FIGURE 6. Regulation of proximal components that recruit MAP3Ks. A: regulation of TGF-␤-activated kinase-1 (TAK1) by TRAF6-dependent
formation of Lys63-linked free polyubiquitin chains. Engagement of the IL-1 receptor recruits TRAF6, activating its autoubiquitination and free
Lys63-linked polyubiquitination chain synthesis. Free K63-linked poly-Ub activates TAK1 by binding the regulatory subunit TAK-1 associated and
binding protein-2 (TAB2). B: CD40L/CD40 recruitment of MEKK1 in B cells. In resting cells, MEKK1 is part of a large complex that includes
TRAF3, IKK␥, the E3 ligases cIAP1 and -2, and the E2 Ub transfer protein Ubc13. Engagement of CD40 triggers Ubc13/cIAP1/2-dependent
ubiquitination of TRAF3 and consequent TRAF3 degradation. The remaining complex dissociates from CD40, and the now active MEKK1
phosphorylates substrates. C: regulation of ASK1 by reactive oxygen intermediates (ROIs). Resting cell ASK1 is associated with reduced
thioredoxin (Trx). Cytokine-induced ROI production oxidizes Trx leading to its dissociation from ASK1 and to the binding of TRAF2. This activates
ASK1. D: T cell receptor engagement activates MEKK4 via promoting an association with GADD45␤ or GADD45␥. This is key to subsequent
activation of p38 and to the propagation of TH1 cell functions.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
709
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Evolutionarily conserved signaling intermediate in Toll
pathways (ECSIT) is an adapter protein that interacts with
the TRAF domain of TRAF6 (but not TRAF2 or -5), and
with full-length MEKK1. This interaction apparently stimulates the LPS/TLR4-stimulated partial proteolytic cleavage
of MEKK1 into an 80-kDa constitutively active fragment,
thereby triggering JNK activation (158). The cell/tissue
types in which this reaction occurs are unclear, however.
Moreover, recent studies suggest several other pathways
coupling TLRs to JNK (sects. IIIC1 and IVC1). More recent
experiments have implicated ECSIT in the LPS-stimulated
production of reactive oxygen intermediates (319).
RING domain (261). In line with this, TRAF2 can also
foster the Lys63-linked ubiquitination of GCKR. This is in
sharp contrast to the mode of activation of GCK by TRAF6
which occurs independently of the RING domain (sect.
IIIC2).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
binoside (AraC), N-acetoxy-2-acetylaminofluorene, cisplatinum (CDDP), mitomycin-C, etc. However, the spectrum of genotoxic stresses capable of activating the JNKs,
as well as the upstream components that couple these
stresses to the JNKs, remain nebulous, a consequence, perhaps, of the nonspecific effects these toxins have on cells
(31, 153, 180, 216, 258).
There is evidence that, via TAB1, TAK1 couples the JNKs
and p38s to TGF-␤ superfamily receptors (201, 236). In
certain cell types, TGF-␤ and BMPs-2 and -4 can activate
the JNKs and p38s; and the MAP3K catalytic activity of
murine osteoblastic cell (MC3T3-E1) TAK1 is activated by
TGF-␤ and BMP4. Trimeric complexes of BMPR-I, the E3
Ub ligase XIAP and TAB1 can be isolated (337). XIAP also
interacts directly with the type-I receptors for BMPs-2 and
-4. Given that XIAP expression in Xenopus embryos reproduces the ventralizing effects of BMP, whereas introduction
of the TAB1 binding region of XIAP prevents the expression
of ventral mesoderm markers elicited by a constitutively
active mutant type-I BMP receptor, it is likely that the
XIAP-TAB1-TAK1 complex is physiologically relevant
(337). This complex may couple TAK1 to TGF-␤ family
receptors.
Saito and co-workers (284) showed that MEKK4 is activated by genotoxins that transcriptionally induce proteins
of the growth arrest and DNA damage-45 (GADD45) family. GADD genes are rapidly induced by DNA damage and
are thought to play a role in coordinating the cellular response to genotoxins. gadd45 was one of the first GADD
genes to be identified (79, 145).
TAK1 is also strongly recruited by proinflammatory cytokines and PAMPs. This process requires TRAF-catalyzed
formation of free Lys63-linked poly-Ub chains (140, 332).
Although CUE domains interact with poly-Ub, evidence
suggests that it is the Npl4-Zn finger motif of TAB2 that
binds poly-Ub. This binding engages the TAB2-TAK1 complex, thereby fostering TAK1 activation (162, 254). The
precise mechanism of TAK1 activation remains somewhat
obscure, but may involve induced proximity/aggregation
(140, 332).
E. Regulation of MEKK4 by DNA Damage
May Involve a Direct Interaction With
GADD45 Homologs
The majority of chemotherapeutic and radiation treatment
protocols that are used to treat cancer kill cells by producing irreversible DNA damage and consequent tumor cell
apoptosis. Unwanted side effects of these treatments include the collateral killing of normal cells both through
DNA damage and nonspecific redox stress. To enable the
improvement of conventional anticancer treatments, it is
important to understand how chemical and radiant genotoxins trigger cell death.
At the cellular level, genotoxic stresses trigger rapid cell
cycle arrest, and, to maintain the integrity of the genome,
either DNA and general cellular repair, or, if the damage is
too great, apoptosis. In certain instances, JNK and p38 have
been shown to be activated by genotoxins such as UV and ␥
radiation, methylmethane sulfonate (MMS), cytosine ara-
710
All three gadd45 homologs (GADD45␣, -␤, and -␥) can,
when ectopically overexpressed, interact with MEKK4 and
activate JNK and p38 in a manner reversed by dominant
inhibitory MEKK4, and purified GADD45 can activate
MEKK4 in vitro (284).
However, it must be noted that it is unclear whether or not,
in response to genotoxic stress, GADD45 proteins are induced to sufficient levels to activate MEKK4 at the times
when MEKK4 and JNK are activated. Thus neither endogenous GADD45␣, -␤, nor -␥ is present at detectable levels
when JNK is fully activated by the DNA damaging chemicals MMS (259, 308). Moreover, JNK is activated by MMS
and other genotoxins in gadd45␣⫺/⫺ cells. Lastly, ectopic
expression of either GADD45␣, -␤, or -␥ does not activate
coexpressed JNK (308). Lastly, there is little or no detectable JNK activation in fibroblasts (p53 positive) treated
with ionizing radiation sufficient to induce GADD45 expression (259).
Some clarification comes from genetic studies indicating
that the GADD45-MEKK4 interaction is relevant, but for
p38 activation selectively; and that GADD45-␤ and -␥ are
key to MEKK4 regulation. GADD45-␣ and MEKK4 are
also implicated in TCR-mediated p38 activation and T-cell
differentiation (35, 250, 251) (FIGURE 6E and sect. IVC3).
IV. BIOLOGICAL FUNCTIONS OF THE JNKS
AND P38 MAPKS
The pathway architectures and kinase/substrate relationships described above were in many cases first identified
using in vitro assays or by transient expression of cDNAs in
transformed cells in long-term culture. The results of such
experiments identify relationships that are possible, and
other kinds of studies are required to ascertain whether such
relationships actually occur between the elements when expressed endogenously in an intact cell. A well-characterized
cohort of chemical kinase inhibitors is available, and the
advent of RNAi-induced mRNA depletion has greatly facil-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
dissimilar. TAB1 is a pseudophosphatase with an NH2terminal domain of unknown function (AAs 26 –323) that
is strikingly similar to the catalytic subunit of protein phosphatase 2C. TAB2 and TAB3 contain NH2-terminal coupling of ubiquitin to ER degradation (CUE) domains (AAs
13–55 of TAB2) and COOH-terminal Npl4-Zn finger motifs (AAs 671– 695 of TAB2).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
A. JNKs and the Central Nervous System
Overexpression of JNK in primary neurons or PC12 cells in
culture will induce neurite extension, at least in part
through phosphorylation of microtubule-associated proteins (28), and activated JNK is required for axonegenesis
(212). JNK is activated after injury to central or peripheral
neurons. In the periphery, JNK inhibitors suppress cell
death; however, they also inhibit the ability of the surviving
neurons to regenerate their axons (206). The molecular details (i.e., JNK isoforms, signaling complexes, etc.) that underlie this duality of action remain to be uncovered. Nevertheless, a role for JNK inhibitors in stroke and a variety of
inflammatory and degenerative CNS pathologies continues
to be actively investigated.
1. The JNKs and brain development
Studies of single and combinatorial jnk isoform knockout
mice have revealed specific stress response and developmental roles for the different JNK isoforms expressed in the
brain.
The glutamate receptor agonist kainic acid rapidly induces
seizures resembling those of epilepsy. Kainate also causes
the widespread apoptosis hippocampal neurons, a process
similar to that which occurs in stroke and Alzheimer’s disease. In contrast to the ubiquitously expressed JNK1 and
JNK2, JNK3 is expressed selectively in brain, heart, and
testis. Disruption of jnk3 produces animals that are strikingly resistant to kainate seizures (339) and, more importantly, from kainate-induced hippocampal neuronal apoptosis (339). Hippocampal cell death is a major clinical
indicator in stroke and Alzheimer’s disease.
The establishment of a role for JNK3 in the brain’s response
to stress prompted the investigation of the role of the JNKs
in normal brain development and function. Analysis of
jnk1/jnk2, jnk2/jnk3, jnk1/jnk3 double-knockout mice revealed that JNK-1 and -2, acting together, play a pivotal
role in regional-specific apoptosis during early brain development (161, 240). In contrast, jnk1/jnk3 and jnk2/jnk3
mice are viable. Thus deletion of jnk1 or jnk2 alone is
insufficient to affect brain development; moreover, JNK3 is
unlikely to be a rate-limiting component in brain development (339).
The combined disruption of jnk1 and jnk2 is embryonically
lethal, resulting in an abnormally prominent hindbrain exencephaly. Normally, during the process of cephalic neurulation, the closure of the hindbrain neural tube is a key step,
and regulated cell death is important to hindbrain neural
tube closure. However, in an indication of reduced apoptotic cell degeneration, jnk1/jnk2 double knockouts exhibited
reduced pynknotic nuclei at the lateral edges of the converging hindbrain, compared with wild-type or single knockout
controls (161, 240). Thus appropriate developmental hindbrain apoptosis requires the collaborative function of JNK1
and JNK2.
In contrast to the regulated apoptosis promoted by JNK1/2
in the hindbrain, the jnk1/jnk2 double knockout increased
forebrain apoptosis [assayed as pynknotic nuclei or terminal deoxynucleotidyl transferase-mediated dUTP nick end
labeling (TUNEL) assays which serve as an indicator of the
internucleosomal cleavage characteristic of apoptotic cells],
suggesting instead that the JNKs reduce apoptosis in the
forebrain (161, 240). The increased forebrain apoptosis in
the jnk1/jnk2 double knockout mice arises from an excessive activation of caspase-3, as detected by increased immunoreactivity for the cleaved, 17-kDa active form of
caspase-3, but not its inactive 35-kDa precursor (161, 240).
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
711
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
itated such validation studies, when appropriate measures
are used to account for off-target effects. RNAi and chemical inhibitors are most conveniently applied to transformed
cell lines in permanent culture. Thus it must be recognized
that, in addition to whatever mutations were engendered in
vivo in the progression of those cells to the transformed
state, the wiring of their signaling pathways may have been
altered further in a manner that enables them to survive and
proliferate in the nonphysiological in vitro milieu. This caveat does not necessarily invalidate the physiological relevance of epistatic relationships identified in such cells, but
strictly speaking limits them to the cell type in which they
are demonstrated. Some additional demonstration is
needed to establish whether or not such a signaling event
occurs only in the context of the transformed state. A relationship so identified may occur universally in all cell types
in vivo, or more realistically, may occur only in a tissuespecific and/or developmentally restricted context. Replication of the effect using chemical inhibitors, RNAi-induced
mRNA depletion or antisense RNA in primary cells or in
living organisms, when technically feasible, provides very
strong support. More commonly, gene modification in a
model organism is taken as a gold standard. This approach
can however be confounded by several mechanisms, e.g.,
the presence of genes of overlapping or interacting function,
whose expression may be altered during development so as
to compensate for deletion of the target gene. Such a problem can be mitigated by conditional inactivation of the target gene postdevelopment; this is perhaps the closest mimic
of the effects of a druglike modifier of target action. Global
gene knockouts present a phenotype that is the sum of the
action of that gene product is all tissues, which may not
uncover the direct effects of the target in any one tissue.
Target actions in one cell type can have indirect effects on a
second tissue that differ from the action of the target itself in
that second tissue. The gene function in a specific cell type is
best revealed by conditional inactivation of the target in a
cell type-specific manner. In the examples that follow, the
application of all these approaches will be illustrated and
their limitations occasionally revealed.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
Thus JNK1 and JNK2 perform opposing functions with
regard to fore- and hindbrain developmental apoptosis.
In further support for a role for JNKs in brain development,
disruption of the gene encoding the neuron-specific JNK
scaffold JIP1 attenuates neuronal axon development (48)
and neuronal stress responses (324). Disruption of JIP3
strongly impairs development of the telencephalic commissure, a major connection between the left and right brain
hemispheres (149). Whether JIP3 also plays a role in the
developmental functions of JNKs revealed by the jnk1/jnk2
deletions is unclear.
2. JNKs and the suppression of autophagy
in the brain
Autophagy is a process of cellular self-digestion that is initiated typically in times of nutrient paucity. During au-
The initiation of autophagy is marked by the appearance of
forms of microtubule-associated light chain-3 (LC3) which
have been covalently modified with phosphatidylethanolamine to form LC3-II. LC3-II is typically detected as a
polypeptide migrating faster upon SDS-PAGE than its unmodified counterpart LC3-I. Inasmuch as LC3-II also binds
to burgeoning autophagosome membranes, LC3-II usually
can be observed as distinct punctae in the cytosols of autophagic cells. Autophagy also coincides with the proteasomal degradation of sequestome-1 (p62/SQSTM1) which
binds autophagic target proteins and LC3-II (160, 174,
342) (FIGURE 7).
Disruption of all three JNK isoforms in neurons results in
elevated LC3-II, and a loss of p62/SQSTM1. In addition,
the JNKTKO neurons survive in culture longer than wildtype counterparts. This autophagy requires the transcription factor FoxO, which induces expression of Bnip3. Bnip3
promotes autophagy by displacing the autophagic effector
Beclin-1 from inactive Bcl-XL complexes (334). This con-
FIGURE 7. JNKs in the central nervous system block autophagy, as mediated by several processes (association of Bnip3 with Bcl-XL, lipidation of LC3-I to form LC3-II and autophagosome formation, and degradation
of the p62 sequesosome protein). Liberation of these autophagic processes enables neuronal survival when all
three JNK isoforms are genetically disrupted.
712
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
A more recent study has examined the consequences of
neuron-specific JNK1–3 triple deletion. Because the combined disruption of jnk1 and jnk2 is embryonic lethal, specifically causing substantially impaired brain development,
Davis and colleagues (334) crossed jnk2⫺/⫺;jnk3⫺/⫺
mice, which are viable, with mice bearing a floxed allele of
jnk1 to generate jnk1LoxP/LoxP; jnk2⫺/⫺;jnk3⫺/⫺ mice.
Cerebellar granule neurons were isolated from these mice
and treated with adenoviral Cre recombinase to disrupt the
floxed jnk1 allele, thereby creating triple knockout JNKTKO
neurons. Remarkably, the JNKTKO neurons are viable, relying on derepression of autophagy for survival (334).
tophagy, cytosolic and organellar components are surrounded by a double membrane autophagosome which
fuses with the lysosome, thereby enabling digestion of the
autophagosome contents and release of the resulting raw
materials. Autophagy is also utilized by the innate immune
system to destroy cytosolic pathogens and has been implicated in the pathology of cancer, diabetes, and heart disease. Several reviews on autophagy have recently appeared
(160, 174, 342).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
trasts with results reported by Wei and co-workers (315,
316) showing that JNK promotes autophagy by triggering
the interaction of Bcl2 with Beclin-1. Interestingly, JNKTKO
neurons show elevated cyclin-dependent kinase activity,
which appears to be required for the FoxO-dependent induction of Bnip3 (334) (FIGURE 7).
Davis and colleagues (334) were also able to generate viable
mice in which all three JNK isoforms were selectively disrupted in the Purkinje cell layer. These neurons were significantly larger than their wild-type counterparts and, like the
JNKTKO cultured neurons, manifested elevated autophagosomes, indicative of enhanced autophagy. Thus the JNKs
act redundantly to suppress neuronal autophagy and survival in vivo (334).
1. JNKs may function with the Wnt pathways to
modulate heart development
As was noted above (see sect. IIIC2), the Wnts are a family
of secreted proteins that recruit receptors of the Frizzled
family to control key cell fate decisions. Noncanonical Wnt
signaling proceeds through the adapter protein Dissheveled
to JNK, possibly through the GCK family kinase GCKR (at
least in B cells) (263).
Wnt11 is required for the adoption of a cardiac cell fate in
Xenopus. Contemporaneously, Wnt11 triggers enhanced
phospho-JNK, and inhibition of JNK signaling with a dominant inhibitory JNK1 construct blocks cardiogenesis stimulated by Wnt11. It should be pointed out, however, that
disruption of the JNKs singly has no apparent effect on
cardiogenesis (41, 70, 217, 235). Moreover, double disruption of the JNKs in various combinations also does not seem
to affect early cardiogenesis, despite causing a substantial
impairment of brain development (161, 240) (sect. IVA1).
Still, all three JNK isoforms are expressed in the heart, and
it is conceivable that they all perform redundant functions
in cardiogenesis (functions that could be blunted by overexpression of dominant negative JNK). Disruption of all
three JNKs in the heart has not been reported.
2. p38s and cardiogenesis
The role(s) of p38s in cardiogenesis is less well studied.
There has been some evidence that p38s may function
downstream of BMP receptors, but the p38 isoforms and
signaling mechanisms remain to be elucidated unambiguously (235, 299). p38 activity has been implicated in the
differentiation of p19 cells into cardiomyocyte-like cells;
however, this mechanism, as with the studies of BMP signaling, has not been well characterized (53, 235, 299). Targeted disruption of cardiac p38␣, or, indeed, total body
disruption of each of the other three p38 isoforms, does not
apparently affect normal heart development (94, 235).
Gene disruption studies have demonstrated the importance
of stress-activated MAPK pathways to cardiovascular development. Thus disruption of mekk3 is lethal, with death
occurring at embryonic day 11. mekk3⫺/⫺ mice manifest a
striking impairment of blood vessel development and angiogenesis. Of particular note, this reduced blood vessel
formation occurs without a loss of expression of either vascular endothelial cell growth factor (VEGF) or VEGF receptors (340). Instead, disruption of mekk3 produces a pronounced, intrinsic defect in the development of embryonic
endothelial cells. In the placenta, this, consequently, impairs the development of embryonic, but not maternal
blood vessels (340). Given that MEKK3, via p38, activates
the transcription factor MEF2C, which is critical to cardiovascular development (340), it is reasonable to propose that
MEKK3 signaling to p38 contributes significantly to the
regulation of embryonic angiogenesis (340).
Having said this, it is important to note that disruption of
MEF mekk3 impairs not only p38 activation, but NF-␬B
activation as well (120). Accordingly, it is possible that the
effects of mekk3 disruption on heart development involve
attenuation of NF-␬B instead of, or in addition to, MAPKs.
4. JNKs and p38s in cardiovascular
disease-confounding results
Numerous studies using pharmacological agents (often of
dubious specificity) and ectopic overexpression of dominant negative/constitutively active MAPK signaling components have been performed to determine the functions of
JNKs and p38s in conditions such as ischemic reperfusion
disease, heart failure, and cardiac hypertrophy. These studies have produced confusing and often confounding results,
likely due to the approaches used and the pleiotropic functions of JNKs and p38s in various aspects of cardiac pathology. A recent excellent review on this subject was published
in this journal (235), and the reader is referred there for
additional information. The results supporting a salutary or
pathophysiological functions for JNKs and p38s will need
to be addressed with more specific genetic approaches.
A role for JNKs and p38s in inflammatory and vascular
diseases (atherosclerosis, transplant vasculopathy, etc.) is
anticipated but as yet unproven. Thus, for example, numerous elements implicated in atherogenesis (TLR-2, -4, and -6
and the adapter MyD88, see also sect. IVC1) are also required for JNK and p38 activation (19, 40, 46, 147, 151).
C. MAPK Pathways and Immune Function
1. MAPK pathways in innate immunity
and TLR responses
Innate immunity represents the first line of defense against
invading microbial pathogens. Much of the innate immune
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
713
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
B. JNKs and p38s in Cardiac Development
3. MEKK3 and cardiovascular development
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
response is triggered by engagement of PRRs. PRRs include
the membrane-associated TLRs as well as cytosolic nucleotide oligomerization domain-leucine-rich repeat (NODLRR) and RIG-like helicase (RLH) proteins. These receptors, upon engagement with agonists, recruit a variety of
adapter proteins, most notably the adapter proteins myeloid differentiation factor-88 (MyD88) and Toll/IL-1 receptor (TIR)-domain-containing adapter-inducing interferon-␤ (TRIF), culminating in recruitment of the E3 Ub ligase
TRAF6 (see sect. IIID). PRRs have been the subject of many
recent review articles (19, 147, 151, 195).
Activation of PRRs by microbial pathogens leads to a fulminant inflammatory response including cytokine production, leukocyte activation, fever, and, in extreme systemic
inflammation/infection, edema, shock and disseminated intravascular coagulation. Much of the inflammatory program implemented by PRRs involves three major signaling
mechanisms: NF-␬B, interferon response factor, and
MAPKs (ERK, JNK, and p38) (19, 147, 151, 195). This
discussion will be limited to the complex and cell specific
functions of MAPKs.
A) TPL-2. Genetic knockout studies of Tpl-2 reveal an important role for this enzyme in innate and acquired immunity,
principally in activation of the ERK pathway by PAMPs
and TNF. tpl-2⫺/⫺ mice produce low levels of TNF upon
systemic treatment with LPS and are strikingly resistant to
LPS-induced systemic toxicity and lethality. Examination
of tpl-2⫺/⫺ macrophages revealed that LPS stimulation
failed to activate MEK1 and ERK while JNK and p38 signaling remained intact. Disruption of tpl-2 also impaired
LPS-stimulated macrophage TNF release, as well as activation of ERK by TNF itself. The loss of TNF release by
LPS-stimulated tpl-2⫺/⫺ macrophages appears to be due
mainly to the loss of the stimulatory effect of Tpl2 on the
ERK-dependent transport of TNF mRNA from the nucleus
to the cytosol (69) (FIGURE 8).
Whereas macrophage ERK is targeted principally by Tpl-2,
the situation in MEFs is more complicated, indicating a
cell-specific quality to Tpl-2 function. Thus disruption of
tpl-2 in MEFs impairs LPS activation of ERK, JNK
(through, respectively, MEK1 and MKK4), and NF-␬B. The
latter involves the lack of Tpl-2 catalyzed phosphorylation
714
FIGURE 8. Summary of MAPK regulation by PAMPs in the innate
immune system. Recruitment of these MAPK pathways enables a
robust inflammatory response and, in the case of Tpl-2-dependent
signaling, is essential for endotoxin lethality in vivo.
of NF-␬B p65 at Ser276, an event that enhances NF-␬B
transactivation function (52).
Consistent with these findings are studies showing a role for
Tpl-2 in innate and acquired immunity. Thus disruption of
tpl-2 (or jnk2) strongly attenuates the development of a
Crohn’s-like inflammatory bowel pathology in mice expressing a tnf transgene in which the AU-rich element (sect.
IVC5) is deleted (resulting in constitutive TNF expression),
genetic evidence confirming that Tpl-2 not only regulates
TNF production, but is also downstream of TNF in vivo
(157) (FIGURE 8).
Tpl-2 also is critical to host defense, a finding consistent
with the central role of Tpl-2 in PAMP signaling, a critical
initial step in the host defense response. Thus deficiency of
tpl-2 impairs interferon (IFN)-␥ production and consequent
host defense against Toxoplasma gondii. Reconstitution of
rag2⫺/⫺ mice with Tpl-2-deficient T cells recapitulates this
defect in host defense, indicating a key role for Tpl-2 in T
cell-mediated host defense (313).
Upon antigen-induced maturation, CD4⫹ TH cells undergo
polarization to either TH1 or TH2 effector cells. TH1 cells
secrete IFN-␥ and promote a form of B cell Ig class switching that gives rise to Ig isotypes which provoke macrophage
phagocytosis. Accordingly, TH1 cells play a key role in cellmediated immunity (the process by which cells that are
infected with viruses and other intracellular microbes such
as mycobacteria are eliminated). In contrast, TH2 cells produce IL-4 and IL-5. IL-4 also triggers B cell class switching.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
PRRs are engaged both by PAMPs, molecular moieties present on invading pathogens, and by DAMPs, molecules produced endogenously in times of physiological stress.
DAMPs include oxidized LDL (oxLDL) in atherosclerosis
and crystalline uric acid (in gout). Specific PAMPs and
DAMPs recruit distinct groups of PRRs. Notably, LPS from
Gram-negative bacteria recruits TLR4, and bacterial peptidoglycans recruit TLR-2 and -6. TLR-2, -4, and -6 are also
recruited by oxLDL (19, 46, 147, 151, 195).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
However, in contrast to IFN-␥, IL-4 promotes the production of Ig isotypes that do not engender phagocytosis (IgE
or, in humans, IgG4). IL-5 activates eosinophils. Accordingly, TH2 cells contribute to immune responses that are
phagocytosis-independent, to those dependent on IgE, or to
those that employ eosinophils. Such responses include allergy and defenses against helminth parasites.
TH1 cells naturally express more Tpl-2 polypeptide, and
disruption of tpl-2, via impairment of ERK activation, biases towards an exaggerated TH2 response. Consistent with
this, CD4⫹ T cells from tpl2⫺/⫺ mice exhibited a poor
induction of T-bet and a failure to induce Stat4 proteinboth signs of weakened T cell-dependent ERK activation, in
particular, development of TH1 cells (314) (FIGURE 9A).
This fits with studies of the JNK knockouts (66, 239), thus
suggesting roles for Tpl-2 and ERK as well as JNK in the
differentiation of both subsets of polarized TH cells (FIGURE
9A and sect. IVC3). Interestingly, lack of MEKK1 also biases TCR-stimulated CD4⫹ T cells toward Th2 differentiation, probably as a result of deficient JNK1 and itch activation (73).
B) TAK1 AND MEKK3.
Several studies indicate that in addition
to Tpl-2, TAK1 and MEKK3 are also involved in the responses to PAMPs and that these responses are cell- and
stimulus-dependent. Thus TAK1 is required in T lymphocytes for TCR; IL-2, -7, and 15 recruitment of JNK (FIGURE
9B); and in fibroblasts and B-lymphocytes for PAMP and
cytokine activation of JNK (253, 257, 303). TAK1 is also
key to early hematopoietic stem cell survival and to the
prolonged in vivo survival of hepatocytes (287).
Similarly, in mouse embryonic fibroblasts (MEFs), disruption of mekk3 impairs IL-1 and LPS activation of JNK and
p38, but not ERK. In MEFs, MEKK3 is crucial to LPS and
IL-1 induction of IL-6 and to TLR-8-mediated activation of
MEF cell JNK (120, 226). In macrophage (myeloid tumor)
cell lines, however, silencing of mekk3 only modestly inhibits early-onset (⬍60 min) LPS stimulation of JNK, p38, and
ERK, while exerting a more profound effect at time points
later than 2 h, coincident with strongly impaired IL-6 and
granulocyte-macrophage colony stimulating factor (GMCSF) induction. Silencing of mekk3 has no effect on induction of tnf (155) (FIGURE 8).
Not all elements implicated in MAPK signaling
contribute positively to inflammatory responses. Thus
MSK1 and MSK2 (sect. IID1) appear to be negative regulators of TLR signaling. Thus downstream of p38 and
ERK1/2, MSK1/2 are needed to resolve inflammatory responses by restricting the LPS-stimulated production of
proinflammatory cytokines. This is achieved by MSK1/2dependent induction of the MAPK-inactivating phosphatase dual specificity protein phosphatase 1 (DUSP1) and by
induction of the anti inflammatory cytokine IL-10. Accordingly, mice doubly deficient in MSK1 and MSK2 are hypersensitive to LPS-mediated endotoxic shock and manifest
exaggerated inflammation in a model of contact eczema
(PMA induced). Thus MSK1 and MSK2 have a negativefeedback role in innate immunity (6) (FIGURE 8). Interestingly, in contrast, DUSP1 induction, in dendritic cells, is
apparently mediated instead by MK1/2 phosphorylation
and activation of Rsk2 and -3 (348) (see sect. IID1).
D) HPK1. HPK1, which is structurally similar to GCK (see
sect. IIIC2), also appears to exert a negative effect on both
innate and acquired immune functions, although it is not
clear if all of these processes are mediated by MAPKs. Thus
disruption of hpk1 renders dendritic cells superior to their
wild-type counterparts. They express higher levels of the T
cell costimulatory molecules CD80 and CD86 and produce
more proinflammatory cytokines (IL-12, IL-1␤, TNF and
IL-6). Consistent with this, hpk1⫺/⫺ dendritic cells can
more strongly stimulate T-cell proliferation than can
hpk1⫹/⫹ dendritic cells. Interestingly, disruption of dendritic cell hpk1 improves the clearance of Lewis lung carcinoma cells by xenograft studies. Exactly how HPK1 negatively regulates dendritic cell function is unclear (4). As
noted in section IIIC2, transfected HPK1 is a potent activator of JNK and p38; however, these MAPKs are implicated
in proinflammatory and immunostimulatory processes.
In lymphocytes, HPK1 also negatively regulates TCR signaling to AP-1. Thus HPK1 associates with and inactivates
the adapter protein SH2 domain-containing leukocyte protein of 76 kDa (SLP76). This negative regulation involves
HPK1-catalyzed phosphorylation of SLP76 Ser376, which
prevents subsequent Tyr phosphorylation of SLP76 and
TCR signaling. It is possible that HPK1 performs a similar
function in dendritic cells (62, 177).
2. The JNKs are key regulators of T cell maturation,
activation and protection from FasL apoptosis
As revealed in three key genetic studies, the JNKs play a
complex role in T-cell maturation and function. APCs express antigen in complex with major histocompatibility
complex proteins (MHC-I or II), which interacts with the
TCR on naive T cells. CD28, engaged in parallel with the
TCR initiates a parallel set of responses, the so-called costimulatory response, needed to elicit optimal T-cell activation culminating in the clonal proliferation of mature, naive
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
715
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
TH1/TH2 polarization and the consequent type of immune
response are influenced in part by the cytokine milieu present at the time of antigen presentation. Thus certain antigen-presenting cells (APCs; most notably macrophages and
dendritic cells) produce IL-12. IL-12 promotes polarization
along the TH1 trajectory (the type I cytokine immune response), whereas mature T cells (TH2 cells in particular,
which can also function as APCs) produce IL-4, leading to
additional differentiation along the TH2 path.
C) MSKS.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
716
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
FIGURE 9. MAPKs and lymphocyte functions. A: MAPKs generally favor TH1 T cell polarization. JNK
activation, Tpl-2-dependent ERK activation, and Lck/Zap70 or GADD45/MEKK4-dependent p38 activation all
have been shown through transgenic mouse studies to contribute to TH1 cell polarization. Somewhat less clear
is whether or not these MAPKs actively suppress TH2 polarization, or if genetic disruption merely produces a
relative excess of TH2 cells. B: T cell receptor (TCR) engagement recruits several parallel MAPK mechanisms.
These include TH1 polarization, through JNK2 activation, Lck/Zap70 or GADD45/MEKK4-dependent p38
activation, or Tpl-2-dependent ERK activation. TCR signaling to MEKK1, independently or through JNK, also
recruits Itch to suppress TH2 polarization. JNK1 itself has been directly linked to active suppression of TH2
polarization; whether JNK1 itself does this through the MEKK1/Itch mechanism is somewhat unclear, as the
JNK isoforms involved in this process have not been identified (see also FIGURE 4C). It is unclear if MEKK1dependent JNK activation is sufficient to mediate the JNK-dependent TH1 bias, or if other mechanisms are
involved. Notably, MEKK3 activation by TCRs contributes to IFN-␥ production and pathogen clearance, while
IL-2-, -7-, and -15-dependent JNK activation and responses require TAK1. C: MEKK3, along with MEKK2, also
suppresses in TH17 and Treg polarization. TCR recruitment of MEKK2/3-dependent ERK activation promotes
ERK-dependent phosphorylation of SMADs-2 and -3, which, in turn, prevents activating SMAD2/3 phosphorylation catalyzed by TGF-␤ receptors. TGF-␤ receptor signaling, through SMAD2/3 and the co-SMAD4, fosters
TH17 and Treg polarization.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
Although loss of either jnk1 or jnk2 shifts T-cell differentiation toward a TH2 phenotype, the two JNK isoforms promote differentiation toward TH1 by different mechanisms.
Thus the jnk2-deficient TH0 cell fails to produce IFN-␥,
indicating that TH1 differentiation is impaired. In contrast,
jnk1-deficient TH0 cells produce substantial IFN-␥ but nevertheless progress to a TH2 phenotype because of a marked
overexpression of IL-4, IL-5, IL-10, and IL-13-cytokines,
which promote TH2 differentiation. Thus JNK2 is required
for TH1 differentiation, and its absence enables progression
to TH2 as a default, whereas JNK1 actively represses TH2
differentiation by suppressing expression of the cytokines
necessary for this process, and thus deficiency of jnk1 disinhibits TH2 differentiation. The inhibition of TH2 differentiation by JNK1 is likely due to its negative regulation of
NFAT and its ability to activate the E3 ubiquitin ligase Itch,
which promotes JunB degradation (37, 66, 67, 73, 239,
278, 302) (FIGURE 9A).
Regarding upstream signaling components that recruit JNK
to regulate T-cell polarization, as noted in section II, D2
and F2, studies employing transgenic mice expressing a
truncated form of MEKK1 in which the kinase domain was
deleted (·KD) indicated that MEKK1 was necessary for
TCR recruitment of JNK. MEKK1 can, either directly or
through JNK, recruit the E3 Ub ligase Itch and regulate
JunB and c-Jun stability. MEKK1 kinase function and Itch
activity act to regulate negatively the production of IL-6 as
well as the TH2 cytokine IL-4 by promoting the ubiquitination and degradation of JunB. In turn, this suppresses TH2
polarization (73). It is unclear if the observed active suppression of TH2 polarization incurred by JNK1 is regulated
through the MEKK1/Itch mechanism, per se inasmuch as
the specific JNK isoform recruited by MEKK1 in this path-
way has not been identified. In contrast, MEKK1-dependent c-Jun degradation attenuates TH2 tolerance and airway inflammation (302).
Similar studies exploiting the ·KD MEKK1 mice also suggest a role for MEKK1 in B-cell function. ·KD mice show
defective germinal center formation, as well as CD40 activation of JNK and p38 and B-cell receptor-stimulated antibody production. MEKK1 deficiency also led to B-cell hyperproliferation after CD40 ligation (88). As with the T-cell
studies, however, these results need to be interpreted with
some caution inasmuch as the ·KD mice retain the MEKK1
noncatalytic region, and this may lead to off target effects.
The ablation of autoreactive cells occurs early in T- and
B-cell development. This process is controlled in part by
FasL-induced apoptosis. Disruption of mkk4 in mice attenuates the activation of ES cell JNK by anisomycin and heat
shock (208). However, disruption of mkk4 is lethal early in
embryonic development, likely due to an impairment of
hepatogenesis. mkk4⫺/⫺ embryos die just after the formation of the primitive vasculature, and contemporaneously
with the onset of hepatogenesis. MKK4 likely functions to
blunt apoptosis in the liver inasmuch as embryonic lethality
consequent to disruption of mkk4 is accompanied by substantial hepatocyte apoptosis (102, 228). Accordingly, to
enable examination of the role of MKK4 (and, by extension, the JNKs) in immune cell function, rag2-deficient blastocysts were microinjected with mkk4⫺/⫺ ES cells to produce chimeric mice with a MKK4-deficient immune system
against a wild-type background. The chimeric T cells were
remarkably hypersensitive to apoptosis induced by FasL.
This suggests that MKK4 and its substrates may function to
protect T cells from apoptosis stimulated by FasL (91, 209).
3. MEKK4, GADD45s, and alternative p38 pathway
in T-cell differentiation and responses
In contrast to the JNKs, activation of p38 in T cells is not
strictly dependent on costimulation, but is activated by
TCR, CD28, and cytokines, e.g., IL-12, in what appears to
be a cumulative manner (65). As discussed in section IIE4,
T-cell p38␣ can be activated by a MAP2K/MAP3K-independent mechanism involving Tyr323 phosphorylation,
followed by autophosphorylation at the activating phosphoacceptor motif Thr180/Tyr182. Tyr323 phosphorylation is catalyzed by the Tyr kinase Zap70 and can be
blocked by GADD45␣. Disruption of gadd45␣ leads to
constitutive elevation of p38 activity accompanied by a lupus-like autoimmune disease, suggesting perhaps a key role
for this T-cell specific p38 phosphorylation (250, 251).
The physiological relevance of the noncanonical p38 pathway was documented in a recent study wherein a genetic
knockin of a mutant p38␣ (Y323F) selectively prevented in
vivo activation of T-cell p38␣. This defect had no effect on
CD4⫹ TH1 differentiation, but instead weakened TH1 re-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
717
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
CD4⫹ T cells to CD4⫹ TH cells. This process requires, in
part, the production and subsequent action of certain cytokines such as IL-2. The levels of JNK1, JNK2, as well as
MKK-4 and -7 are all quite low in mature naive CD4⫹ T
cells, but increase greatly upon CD3/CD28 costimulation.
As implied by this behavior, JNK1/2 are largely dispensable
for activation of naive T cells to the TH state, and T cells
doubly deficient in jnk1 and jnk2 show robust proliferation
and unimpaired (and perhaps increased) production of
IL-2. The essential functions of JNK-1 and -2 are concerned
with the subsequent differentiation of the activated CD4⫹
T cells into the TH1 or TH2 phenotypes. Inactivation of the
genes encoding either JNK1 or JNK2 favors the differentiation of the activated T cells toward the TH2 phenotype (37,
66, 67, 239, 278) (FIGURE 9A). Similarly, treatment of patient-derived T-cell clones reactive with a myelin basic protein (MBP) peptide fragment (AAs 83–99) with peptide mutants that enable activation of JNK/p38 but not ERK provokes a TH1-biased response, whereas other MBP(83–99)
mutant peptides that enable ERK but not JNK/p38 activation induce TH2 biased differentiation (267).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
sponses, most notably the production of IFN-␥ (135)
URE 9B).
(FIG-
It would appear from these two sets of findings that the
different isoforms of GADD45 can differentially regulate
p38 activation through the Zap70 nonconventional (inhibited by GADD45␣), or the more conventional MEKK4
pathway (enhanced by GADD45␤ and ␥), and through this,
TH1 differentiation and function (FIGURE 9B).
MKK3 and its p38 targets also contribute to TH1 development by promoting the production of IL-12 (181). IL-12 is
produced by macrophages and dendritic cells as part of the
type I cytokine, proinflammatory immune response. As
mentioned above, this response helps to foster the differentiation and maturation of naive T cells into TH1 cells. Disruption of mkk3 results in mice that are viable and fertile.
However, these mice manifest severely impaired macrophage and dendritic cell IL-12 production. In addition,
IFN-␥ production and antigen-driven differentiation of naive T cells are reduced significantly. Accordingly, deletion of
mkk3 attenuates the type I cytokine immune response
(181). Pharmacological inhibition of cultured macrophage
cell p38 pathway blunts IL-12 gene transcription. This suggests that a significant component of p38’s effect on IL-12
expression is transcriptional.
4. MEKK3 and T cell responsiveness: MEKK2,
MEKK3, and the differentiation of TH17 and Treg
T cells
A recent study implicates MEKK3 in T-cell responsiveness. Conditional disruption of T lymphocyte mekk3 has
no effect on TH1/2 differentiation, but instead substantially reduces TCR recruitment of ERK, JNK, and p38
(consistent with MEKK3’s apparent in vitro biochemical
promiscuity, see sect. IIF2). In addition, under in vitro
nonpolarizing conditions (TH0-TCR engagement in the
absence of cytokines that lead to effector T cell differentiation), TCR-mediated induction of IFN-␥ production
(ordinarily a TH1 response in vivo) is also impaired upon
718
TH17 and Treg T cells represent additional subpopulations
of CD4⫹ T cells that function in regulating the severity of
inflammation and, perhaps, autoimmunity. TH17 cells are
thought to be largely proinflammatory. Their differentiation is promoted in part by IL-17 (which is also secreted
by TH17 cells) and TGF-␤. In contrast, Treg cells are
thought to dampen immune responses and suppress autoimmune and pathological inflammatory responses.
Treg differentiation is dependent on the transcription
factor Foxp3 and, as with TH17 cell differentiation, is
also fostered in part by TGF-␤.
Combined, whole body disruption of mekk2 and mekk3
is lethal; however, mekk2⫺/⫺ mice crossed with
mekk3Lck⫺Cre/⫺ mice enable T cell lineage-specific deletion of MEKK2 and -3. These mice manifest an excess
accumulation of both TH17 and Treg cells in lymphoid
tissues (spleen). This process is cell-intrinsic, as the excess
TH17 and Treg accumulation is maintained even if
knockout embryonic stem cells are used to reconstitute
rag2-deficient embryos (306).
Mechanistically, TGF-␤ in vitro can stimulate naive
CD4⫹CD62LhiCD44l°CD25⫺ T cells into Foxp3⫹ Treg
cells. Likewise, in vitro, TGF-␤ can be used to stimulate the
differentiation of naive CD4⫹ T cells into IL-17-secreting
TH17 cells. Both of these processes are substantially enhanced upon contemporaneous disruption of T cell mekk2
and mekk3, and pharmacological inhibition of TGF-␤ receptor signaling reduces Treg and TH17 differentiation in
vivo in mekk2⫺/⫺ mekk3Lck⫺Cre/⫺ mice. This suggests that
mekk2⫺/⫺ mekk3Lck⫺Cre/⫺ mice are hyperresponsive to
TGF-␤ (306) (FIGURE 9C).
Clues as to the mechanistic basis of how the combined
disruption of mekk2 and mekk3 affects T-cell polarization come from the observation that MEKK2 and
MEKK3 can indirectly regulate the intensity of TGF-␤
signaling, a critical component of both TH17 and Treg
differentiation. Thus disruption of mekk2 and mekk3
impairs TCR recruitment of ERK and p38. This, in turn,
reduces ERK (but not p38)-dependent, inhibitory phosphorylation of Similar to Mothers Against Decapentaplegic (SMAD)-2 at Thr220, Ser245, Ser250, and Ser255,
and SMAD3 at Thr179, Ser204, Ser208, and Ser215. The
SMADs are transcriptional regulators that are direct targets of the TGF-␤ receptor. Phosphorylation by the receptors (at Ser465 and Ser467 for SMAD2, or Ser423 and
Ser425 for SMAD3) activates triggers binding to
SMAD4, nuclear translocation and activation of TGF-␤
target genes. Inhibitory ERK phosphorylation of the
SMAD2/3 blocks phosphorylation of the SMAD proteins
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
In contrast to this, earlier studies indicate a more conventional role for GADD45s and MAP2K/MAP3K-dependent
p38 activation in T-cell function. As noted in section IIIE,
GADD45, especially GADD45␤ and -␥, may function to
interact with and activate MEKK4; however, the function
of this mechanism in the DNA damage response is unclear
(259, 284, 308). Instead, disruption of mekk4 reduces TCR
activation of CD4⫹ T-cell p38 (FIGURE 9B). Of note,
mekk4⫺/⫺ T cells do not produce IFN-␥, indicating that
TH1 function or differentiation is impaired. Further studies
showed that expression of GADD45␤ or -␥ promotes TH1
function (IFN-␥ production), but only in mekk4⫹/⫹ T
cells. This GADD45␤ or ␥-MEKK4 pathway appears to
integrate signals from the TCR to influence positively TH1
cell differentiation (35).
disruption of T cell mekk3. In vivo, this leads to attenuated clearance of Listeria monocytogenes-OVA infection
(306) (FIGURE 9B).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
by the TGF-␤ receptor, thereby reducing the potency of
TGF-␤ receptor signaling (189, 264). Thus attenuation
of this phosphorylation upon disruption of mekk2/3 results in enhanced TGF-␤ receptor signaling in T cells,
thereby increasing TH17 and Treg differentiation (306)
(FIGURE 9C).
Although MEKK2/3 can influence the level of both proinflammatory TH17 and anti-inflammatory Treg cells, it is
likely that the effect on TH17 cells is more dominant. Thus
disruption of mekk2/3 in T cells causes a more severe autoimmune response in the multiple sclerosis model experimental autoimmune encephalitis (306).
A robust and efficient innate immune response requires the
induction of proinflammatory genes, not just increased
transcription, but posttranscriptional stabilization of key
mRNAs. Accordingly, proinflammatory stimuli such as
bacterial LPS can stabilize mRNAs for IL-1 and TNF as well
as those encoding secondary cytokines such as IL-6, IL-8,
and genes that regulate the biosynthesis of inflammatory
lipids (cyclooxygenase-2, COX2). The posttranscriptional
induction of these mRNAs is more rapid than de novo transcription, and while some of these cytokine genes are also
induced at the transcriptional level, the posttranscriptional
induction mechanism appears dominant inasmuch as the
appearance of these mRNAs is insensitive to the transcriptional inhibitor actinomycin. This posttranscriptional
mRNA stabilization is largely dependent on p38s-␣ and -␤
and their target kinase MK2 (87, 112, 159, 233, 234, 265)
(sect. IID1), although JNK can also influence the stabilization of certain mRNAs.
We are only beginning to understand the mechanism by
which p38s target mRNAs for stabilization. mRNAs that
manifest a high rate of turnover include those encoding
TNF, IL-1, COX2, IFN-␥, IL-2, IL-3, IL-6, and (in humans) IL-8 frequently. These mRNAs contain in their 3=
untranslated regions (UTRs), so-called AU-rich elements
(AREs). AREs are distinguished by the presence of multiple copies of the sequence AUUUA. If transferred to
stable mRNAs, such as that for ␤-globin, AREs confer in
resting cells destabilization of the target mRNA. In addition, AREs enable signal-induced stabilization of target
mRNAs (39). Expression of MKK6 enhances the stability
of a chimeric mRNA consisting of the coding region for
␤-globin and a 161-nt 3= ARE from the IL-8 mRNA. This
reaction can be reversed by a kinase-inactive, dominant
inhibitory mutant of p38␣ (327). The simplest interpretation of these findings is that the ectopically expressed
MKK6 recruits p38 which, in turn, induces stabilization
of AU-rich mRNAs. Consistent with this, a constitutively
active mutant of MK2 could trigger ARE-dependent
Further support for a role for MK2 and, possibly, MK3 in
signal-induced posttranscriptional regulation of mRNA
comes from genetic knockout studies. Thus disruption of
mk2 severely impairs LPS induction of TNF, and
mk2⫺/⫺ mice show a substantial resistance to LPS-induced shock, due in large part to this loss of TNF production (87, 159). While this effect confers protection
against chronic, “sterile” inflammation, including collagen-induced arthritis, disruption of mk2 increases susceptibility to microbial infection, due to a significant reduction in host defense responses (e.g., to Listeria monocytogenes infection) (87, 159). This observation suggests
that some caution is warranted in targeting pivotal inflammation pathways to treat chronic inflammatory conditions such as arthritis, atherosclerosis, or type 2 diabetes, as this may produce unwanted attenuation of host
defense responses.
Surprisingly, disruption of mk3, which is structurally similar to MK2, is much less effective at reducing LPS induction
of TNF; and an mk2;mk3 double knockout produces a
modest further decrease in LPS induction of TNF beyond
that incurred upon disruption of mk2 alone. Of note, MK3
is, for reasons that remain unclear, considerably less enzymatically active (when assayed against Hsp27) than MK2,
perhaps accounting for MK3’s reduced impact on TNF production (87, 233, 234).
As a further demonstration of MK2’s dominant role in LPS
regulation of TNF production, disruption of mk5 is without
effect on LPS-stimulated stabilization of TNF mRNA. Indeed, mk5⫺/⫺ mice have no clear phenotype. Analysis of
endogenous MK5 activation reveals a differential substrate
specificity (despite the observation that recombinant MK-2,
-3, and -5 have overlapping substrate specificity in vitro)
(265).
MK2 promotes to ARE-mediated mRNA stabilization
through the phosphorylation of proteins that bind to AREs
and regulate the stability of the cognate mRNAs. Thus
MK2, in an agonist-stimulated, p38␣/␤-dependent manner,
phosphorylates at least three proteins linked to the regulation of mRNA stabilization: tristetraprolin (TTP, at Ser52
and 178), HuR, and heterogeneous nuclear ribonucleoprotein A0 (hnRNP-A0) (57, 112, 185, 238).
Of these, TTP is best understood. TTP binds AREs (notably
that of the TNF message) and fosters the deadenylation and
degradation of the cognate mRNAs. MK2 phosphorylation
blocks TTP association with AREs, perhaps due to phosphorylation-dependent binding of 14 –3-3 proteins to TTP.
Phosphorylation also triggers TTP destabilization and deg-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
719
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
5. Role of the JNKs and p38s in promoting cytokine
release via posttranscriptional regulation of
mRNAs encoding proinflammatory proteins
mRNA stabilization. This suggests that p38-activated
mRNA stabilization is mediated by MK2, a p38 substrate
(80, 237, 327).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
radation. Consistent with this, disruption of ttp leads to
pathophysiological increases in constitutive TNF levels (87,
112, 185) (FIGURE 10).
Interestingly, metabolic stress (hypoxia, nutrient withdrawal), in addition to autophagy, leads to the formation of
stress granules (SGs). SGs are regions of stalled mRNA
translation the formation of which coincides with p38 activation and TTP phosphorylation. By this process, TTP is
excluded from SGs, thereby preventing the degradation of
mRNAs present in the SGs (87) (FIGURE 10).
Although AREs can regulate mRNA stability, this regulation is frequently manifested differentially. Thus, for
example, whereas the mRNA for c-fos contains an ARE,
the c-fos mRNA remains labile upon stimulation with
agents that regulate known ARE binding proteins. Thus
it is plausible to suggest that additional cis elements confer signal-induced mRNA stabilization. IL-2 induction is
a characteristic feature of T-cell activation. IL-2 release
requires TCR (CD3) and CD28 costimulation (278).
While transcriptional induction of contributes strongly
to IL-2 production, the signal-induced stabilization of
IL-2 mRNA further increases the expression, by activated T cells, of IL-2. The JNKs regulate the stabilization
of the IL-2 message through a cis element that spans from
the 5=-UTR to the beginning of the coding region (30).
SB202190 selectively blocks p38␣ and -␤ and is structur-
D. Proinflammatory MAPK Pathways
and Cancer
The initial observation that JNKs (and later p38s and
ERKs) participated in the activation of AP-1, components
of which include the c-Jun protooncoprotein, immediately
suggested a role in oncogenesis. Indeed, early studies of
c-Jun phosphorylation identified c-Jun phosphorylation at
Ser63/73 as a key step in Ha-Ras transformation (270).
Subsequent work identifying JNKs as targets for proinflammatory cytokines, and as enzymes linked to apoptosis, suggested that the JNKs were likely to play a complex role in
FIGURE 10. p38␣/␤ and MK2-dependent mRNA stabilization, a key event in inflammatory cytokine production. LPS and other PAMPs can trigger the rapid production of proinflammatory cytokines such as TNF through
posttranscriptional stabilization of their mRNAs. Tristetraprolin (TTP) ordinarily binds to mRNAs that contain
AU-rich elements (AREs, AUUUA in the figure). This binding fosters the polyadenylation and degradation of the
target mRNAs. LPS activation of p38␣/␤ and consequent activation of MK2 leads to MK2-catalyzed TTP
phosphorylation and dissociation from AREs, thereby preventing mRNA degradation. hnRNP-A0 and HuR are
also ARE binding proteins and MK2 substrates. They associate with ARE-containing mRNAs in response to
PAMPs and may function to stabilize these mRNAs; however, their mechanisms of function are unclear.
720
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
The functions of HuR and hnRNP-A0 are less clear. HuR
may, in contrast to TTP, stabilize mRNAs by binding, in
a p38/MK2-dependent manner. hnRNP-A0 may function
in a similar manner inasmuch as its phosphorylation by
MK2 increases binding to the TNF message (57, 238)
(FIGURE 10).
ally similar to SB20358. At high concentrations, however, SB202190 also blocks JNK. Pretreatment of cells
either with SB202190 at levels sufficient to blunt JNK
activity, or with cyclosporin, which blocks CD3/CD28
activation of JNK (278) (and p38) prevents IL-2 mRNA
stabilization. Of note, the concentrations of SB202190
required to inhibit significantly IL-2 mRNA stabilization
were considerably higher than those necessary for complete inactivation of p38. Thus it is likely that the drug
was affecting JNK. In contrast, concentration of
SB202190 needed to reverse IL-2 message stabilization
corresponded closely with that required to prevent the
exclusively JNK-catalyzed phosphorylation of c-Jun in
situ. In further support for a role for the JNKs in mRNA
stabilization, expression of a constitutively active mutant
of the JNK-specific MKK7 or of MEKK1, but not a constitutively active mutant of the p38-specific MKK6, or
the ERK-specific Raf-1, triggered stabilization of the IL-2
mRNA. Likewise a dominant inhibitory MKK7 mutant
blocked MEKK1- or CD3/CD28-stimulated stabilization
of the IL-2 mRNA (30).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
the development of cancers. This has proven to be true, with
JNKs and their regulators implicated both in pro- and antioncogenic mechanisms.
1. JNKs and cell proliferation
Disruption of jnk1 modestly impairs the in vitro proliferation of MEFs. In contrast, disruption of jnk2 modestly enhances proliferation. Interestingly, contemporaneous disruption of jnk1 and jnk2 significantly impairs cell proliferation, more so than disruption of jnk1 alone. In these cells,
JNK1 accounts for most of the observed c-Jun phosphorylation activity (JNK3 is not detectable in MEFs) (294). The
more predominant role for JNK1 in proliferation is consistent with this.
2. JNKs and cellular apoptosis
The implication of the JNKs in neuronal apoptosis is of
obvious importance to brain development and function (see
sect. IVA); however, apoptosis is also key to tumorigenesis.
Thus disruption of jnk1 modestly suppresses ultraviolet-C
(UV-C)-induced MEF apoptosis, while disruption of jnk2 is
without apparent effect. However, contemporaneous disruption of jnk1 and jnk2 almost completely abolishes
UV-C-induced MEF apoptosis. The combinatorial effect
of the JNKs is apparently independent of p53 inasmuch
as p53 and p73 levels are unaffected by jnk1/2 disruption. Instead, JNK1 and JNK2 are together required for
UV-C-induced cytochrome c release and mitochondrial
apoptosis. Indeed, reintroduction of cytochrome c into
the cytosol restores apoptosis in jnk1⫺/⫺;jnk2⫺/⫺
MEFs (294) (FIGURE 11B).
JNK-mediated mitochondrial apoptosis involves recruitment and regulation of members of the breakpoint cluster
region (Bcl)-2 family. Bcl-2 itself is antiapoptotic, preventing cytochrome c release. The activity of Bcl-2 is blocked by
sequestration, mediated by the Bcl-2 homology region 3
(BH3) domain-only proteins Bax and Bak. The related BH3
proteins Bim and Bmf may also sequester Bcl-2, thereby
promoting apoptosis. These proteins can also bind and activate Bax and Bak (172) (FIGURE 11C).
Bim is itself sequestered and inhibited by binding to the
dynein light chain (DLC)-1. This binding requires a short
motif in Bim containing a consensus JNK phosphorylation
motif. Indeed, JNK phosphorylates Bim at Thr56 and Ser58
in this motif, disrupting the Bim-DLC1 interaction and free-
3. JNK1 and chemical-induced
hepatocellular carcinoma
Studies using knockout mice further support the notion that
JNK1 is integral to cell proliferation and tumorigenesis.
Thus partial hepatectomy is a well-established model for
cell proliferation and organ regeneration. Posthepatectomy
regeneration begins with mitosis of adult hepatocytes but
ultimately involves all cell types in the liver, with extensive
crosscommunication and intercellular signaling. Disruption
of jnk1 or treatment with a JNK inhibitor impairs liver
regeneration after partial hepatectomy. As to possible
mechanisms, RNAi studies using cultured hepatocellular
carcinoma cell lines indicate that JNK1 elevates the expression of c-Myc which, in turn, suppresses the induction of
the cell cycle inhibitory protein p21 (122) (FIGURE 11A).
Diethyl nitrosamine (DEN) is a hepatotoxin that engenders
hepatocyte death followed by compensatory regeneration.
Sustained administration at sublethal doses results in multiple cycles of hepatocyte death and regeneration, ultimately accompanied by the development of hepatocellular
carcinoma. DEN-induced liver cancer dramatically enhanced upon disruption of liver IKK␤, is strikingly attenuated (albeit not eliminated) upon disruption of jnk1, but not
jnk2. While this may be due in part to the elevation in JNK
activation that occurs consequent to disruption of ikk␤
(286), it is important to note, as discussed below, that neither JNK1 nor JNK2 has any hepatocyte-intrinsic effect on
hepatocyte proliferation. The global jnk1 knockouts attenuate the DEN-induced HCC by preventing the Kupffer
and/or stellate cells from providing key prooncogenic signals. As predicted by JNK depletion from hepatocarcinoma
cell lines, DEN-induced, jnk1⫺/⫺ liver cancers show reduced c-Myc levels and elevated p21. Suppression of p21 is
also required for optimal tumorigenesis. Thus JNK1 is
clearly key to tumorigenesis in this model (247, 286) (FIGURE 11A).
4. A more complex, combined and interactive role
for JNK1 and JNK2 in hepatocellular carcinoma
revealed by dual jnk1/jnk2 disruption
Although a global deficiency of jnk1 clearly suppresses
DEN-induced hepatocellular carcinoma, recent studies of
the combined role of JNK1 and JNK2 comparing hepatocyte-specific and general JNK deletion identifies a more
complex role for JNKs in both promoting and reducing
hepatocellular carcinoma. Compound disruption of both
jnk1 and jnk2 in hepatocytes was achieved by crossing the
hepatocyte specific JNK1 deficient albumin (Alb)-Crejnk1LoxP/LoxP mice with jnk2⫺/⫺ mice. In contrast to the
global disruption of JNK1 in all liver cell types (see sect.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
721
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
The MEF studies were confirmed and extended in studies of
the proliferation of hepatocellular carcinoma cells. Thus
disruption of jnk1 in these cells impaired proliferation and
tumorigenesis after xenotransplantation (121). This evidence supported a straightforward role for JNK1 in tumor
development.
ing Bim to promote Bax-dependent mitochondrial apoptosis (172) (FIGURE 11C).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
IVD2), double knockout of jnk1 and jnk2 in hepatocytes,
despite modestly slowing cell proliferation (BrDU incorporation), did not affect liver regeneration after partial hepatectomy, and did not inhibit DEN-induced hepatocarcino-
722
genesis, but actually enhanced DEN-induced tumor size,
without changing overall tumor incidence or number (51).
A global disruption of jnk1 and jnk2 in all liver cell types,
i.e., in hepatocytes and nonparenchymal cells such as, e.g.,
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
FIGURE 11. JNKs and tumorigenesis. A: JNKs promote hepatocellular carcinoma through excessive compensatory proliferation. DEN triggers necrotic hepatocyte death and associated secretion of IL-1␣. IL-1␣ stimulates the
JNK-dependent production of Kupffer cell IL-6. This IL-6 promotes the compensatory proliferation of hepatocytes,
contributing to tumorigenesis. Both IKK␤ and p38␣ suppress ROI production and consequent IL-1␣ production,
thereby attenuating DEN hepatocarcinogenesis. Note, therefore, how it is the Kupffer cells that, through JNK,
control hepatocarcinogenesis. B: JNK activation by ultraviolet light (UV-C) promotes mitochondrial cytochrome c
release and resulting mitochondrial/cytochrome c-dependent apoptosis. C: JNK-dependent mitochondrial apoptosis can involve Bcl-2-related proteins. Bcl-2 itself blocks apoptosis. This inhibition is, in turn, inhibited by Bim. Bim
inhibition of Bcl-2 is prevented by dynein light chain (DLC)-1 binding and sequestration of Bim. JNK phosphorylation
of Bim dissociates the Bim-DLC1 complex freeing Bim to inhibit Bcl-2 action. D: ASK1 and ASK2 promote dermal
epithelial tumorigenesis by promoting compensatory proliferation in a manner somewhat similar to JNK-dependent
hepatocellular carcinogenesis (see A). DMBA/PMA treatment leads to the production of reactive oxygen intermediates (ROI), which activate the ASKs. ASK2 then promotes apoptosis, a process that requires ASK1-dependent
ASK2 stabilization. However, ASK1 independently promotes the release of TNF and IL-6, which promote compensatory proliferation of epithelial cells.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
5. p38␣ antagonizes JNK-dependent
hepatocarcinogenesis
While JNK-dependent proliferation is an important contributor to hepatocarcinogenesis, p38␣ (as with IKK␤/NF␬B; Refs. 247, 286) antagonizes this effect. Thus, although
global p38␣ deletion is embryonic lethal due to placental
insufficiency (1) and impaired lung function (121), MEFs
and hematopoietic cells from p38␣⫺/⫺ mice show increased proliferation due to sustained JNK activation (121).
As indicated above, ikk␤ disruption fosters enhanced DENinduced hepatocarcinogenesis in part due to dysregulated
JNK activation (247, 248). However, ikk␤ disruption also
leads to increased ROI production. Interestingly, in the
DEN model, ROI production is also elevated upon disruption of p38␣, despite the presence of intact IKK␤. This
elevated ROI production leads to increased IL-1␣ release
which, as for the combined jnk1/jnk2 disruption, contributes to sustained JNK-dependent proliferation. Thus p38␣,
by suppressing ROI-dependent IL-1␣ release, blocks hepatocarcinogenesis (248) (FIGURE 11A). The link between ROI
and carcinogenesis is noteworthy given the role of ROIactivated ASK1/2 in cancer (discussed below).
6. JNK is required for Ras-dependent tumor
formation in lungs
As noted above, the JNKs, especially JNK1, via activation
of AP-1-dependent gene expression, are critical to cell proliferation. Compared with p53⫺/⫺ (Trp53) MEFs, Trp53
MEFs with an additional compound deletion of jnk1,jnk2
show reduced proliferation. Similar results are seen if the
Trp53, jnk1/2-depleted cells express an ectopic activated
K-Ras. Indeed, JNKs are required for Ras-induced suppression of contact growth inhibition and coincident loss of
E-cadherin (26).
This result was extended to in vivo analysis in studies of
mice bearing a lung-specific G12D Ras knock-in versus
KrasG12D mice with a conditional JNK1/2 deficiency
(jnk1LoxP/LoxP;jnk2⫺/⫺). In this instance, lung-specific
JNK1 disruption was achieved by nasal instillation of adenoviral Cre. Here, disruption of the JNKs caused a reduction in the number of K-Ras-G12D-driven hyperplastic
lung lesions and adenomas (26).
7. MKK4 and cancer-evidence that MKK4 can act
as both a tumor suppressor and tumor promoter
Studies of elements upstream of the JNKs and p38s are less
well developed; however, there is evidence that MKK4 has
a role in cancer. In this regard, it is important to remember
that MKK4 is one of two MKKs implicated in JNK activation and three in p38 activation (sect. IIE2) and that it,
MKK7, MKK3, and MKK6 may respond to different stimuli, in distinct cell types, to produce different responses.
Loss of function mutations in mkk4 have been identified in
⬃5% of tumors from several tissues, and a number of studies have highlighted the importance of MKK4 in suppressing prostate and ovarian tumor cell metastasis (305, 325).
On the other hand, ectopic expression of MKK4 in breast
cancer cell lines devoid of mkk4 promotes proliferation,
while RNAi silencing of mkk4 impairs proliferation of
mkk4-positive breast cancer cells (205, 325). Moreover,
skin-specific deletion of mkk4 renders mice resistant to
chemical carcinogen-induced tumorigenesis, a process
mediated by JNK-dependent induction of the EGF receptor (78).
8. Differential roles for ASK1 and ASK2 in apoptosis
and tumorigenesis
Apoptosis is a critical mechanism that prevents inappropriate cell proliferation and tumorigenesis. In contrast, inflammation can facilitate tumor cell invasion as well as proliferation of the tumor and plays a major role in metastasis.
ASK1 and ASK2 are MAP3Ks that recruit the JNKs and
p38s equivalently; however, ASK2 is unstable and only
functions biochemically as part of a heteromer with ASK1
(see sect. IIF3). Using knockout mice, Iriyama et al. (126)
showed that ASK2, in cooperation with ASK1, which is
required for ASK2 stability, was tumor suppressive, promoting apoptosis in gut epithelial cells. In contrast, ASK1
alone triggered the production, by inflammatory cells, of
cytokines that favored proinflammatory tumorigenesis
(126). Thus, as with the JNKs, ASK-1 and -2 perform complex functions in tumorigenesis.
ASK2, in contrast to the ubiquitously expressed ASK1, is
most prominently expressed in the skin and gut, notably
in the skin keratinocyte layer (126, 282). Disruption of
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
723
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Kupffer and Stellate cells, was achieved by crossing Mx1Cre-jnk1LoxP/LoxP mice with jnk2⫺/⫺ mice and administering poly-IC. As with the earlier, global deletion of jnk1,
disruption of both jnk1 and jnk2 in hepatocytes and nonparenchymal cells did suppress DEN-induced hepatocellular carcinoma, and the JNK-dependent proliferation involved downregulation of p21 and upregulation of c-Myc
(51). The most plausible explanation for the different
responses to hepatocyte-specific versus global deletion of
JNK1 on DEN-induced liver carcinogenesis is that the
absence of JNK1 from the nonhepatocyte cell population
diminishes their ability to provide to hepatocytes signals
necessary for carcinogenic transformation, such as the
protumorigenic cytokines IL-6 and TNF. The effects of
hepatocyte-selective jnk1 deficiency on posthepatectomy
regeneration and DEN carcinogenesis indicate that, in
contrast to MEFs, hepatocyte JNKs contribute little to
the regulation of hepatocyte proliferation (51, 121, 247,
248, 252) (FIGURE 11A).
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
ask2 does not affect ASK1 levels. Using a two-stage 7,12-dimethylbenz[a]anthracine (DMBA)/phorbol ester (PMA) skin tumorigenesis model, Iriyama et al. (126) observed that
disruption of ask2 dramatically increased the incidence
and frequency of the resulting dermal papillomas. ASK2
is activated by DMBA and can initiate tumorigenesis, but
more prominently promotes keratinocyte apoptosis in
response to DMBA. In this capacity, ASK2 fosters the
removal of keratinocytes irrevocably damaged by DMBA
treatment. As noted in section IIF3, both ASKs are activated by ROIs, and indeed, DMBA triggers ROI formation, which is the basis for consequent ASK2 activation
(126).
Having said this, the proinflammatory role of ASK1 may
actually suppress gut tumorigenesis in inflammatory colitis. Colitis triggered by microbial pathogens is linked
with an increased incidence of cancer. When ask1⫺/⫺
mice were treated with dextran sodium sulfate or Citrobacter rodentium to induce colitis, apoptosis of bone
marrow cells was reduced, and colitis was greater than
controls, as was subsequent colitis-induced tumorigenesis (106, 107).
E. Proinflammatory MAPK Signaling in
Metabolic Disease and Type 2 Diabetes
1. General considerations
The interplay between proinflammatory and hormonal signaling mechanisms is becoming more established. This is
especially true for type 2 diabetes, a common complication
of obesity and the increasingly sedentary lifestyle seen in
Western countries. It is evident that inflammatory signaling
pathways can often adversely influence insulin sensitivity,
glucose tolerance, and overall metabolic control, contributing to the development of insulin resistance and, eventually,
type 2 diabetes.
Thus obesity and insulin resistance, which typically progress to type 2 diabetes mellitus, are associated with a
chronic inflammation that includes abnormally elevated
levels of proinflammatory cytokines (notably, TNF, IL-1,
and IL-6). These abnormalities appear to originate in adi-
724
Signaling by free fatty acids (FFAs) as well as ER stress have
also been implicated in the pathophysiology of obesity and
insulin resistance. Both FFAs and ER stress can activate the
JNKs pathway; however, as is discussed below, mechanisms
coupling these stimuli to the JNKs are still nebulous. Despite
this, strong evidence has emerged that JNK, especially JNK1,
possibly in conjunction with other MAPK groups, can directly
contribute to insulin resistance (FIGURE 12A).
2. Cellular and biochemical effects
An appreciation for the role of MAPKs in the development of insulin resistance first came from studies of the
effects of inflammatory cytokines such as TNF-␣ on insulin resistance in 3T3-L1 adipocytes and rodent models
and on the proximal elements in the insulin signal transduction pathway, i.e., the insulin receptor and the insulin
receptor substrate (IRS; for review, see Ref. 321) proteins
(115, 116, 117). TNF-␣ was found to impair the ability
of IRS-1 to associate with the type 1 PI-3 kinase (141) as
well as with the insulin receptor (220), concomitant with
increased IRS-1 Ser/Thr phosphorylation. The recognition of the central role of the stress-activated MAPKs as
effectors of TNF led to an examination of the contribution of these kinases to the negative regulation of insulin
signaling (23, 111), with the demonstration that active
JNK associates with and phosphorylates IRS1 in cells at
several sites including Ser307; moreover, mutation of
IRS-1(Ser307) to Ala largely eliminated the ability of to
suppress insulin-stimulated IRS-1 Tyr phosphorylation
in CHO cells (2, 3). The importance of JNK in the pathogenesis of insulin resistance was strongly supported by
the finding that feeding mice a high-fat diet (HFD) results
in JNK activation in muscle, fat, and liver; moreover,
global knockout of JNK1 protected mice against obesity
when fed a HFD and strongly mitigated the consequent
insulin resistance through improved insulin signaling
(111). Several mechanisms have been proposed for the
ability of a HFD to activate JNK apart from the induction
of adipose tissue-derived inflammatory cytokines, including PKC activation, the induction of ER stress, or PRR
signaling (129, 156, 215, 221). The role of IRS-1 Ser/Thr
phosphorylation as a key mediator of insulin resistance
has received considerable attention; the initial observations were followed by a series of reports demonstrating
a consistent correlation between IRS-1 (Ser307; Ser312
in humans) phosphorylation and the occurrence of insu-
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
As expected from initial findings concerning ASK2, disruption of ask1 also attenuates in vivo stability of ASK2 (see
also sect. IIF3). In line with this, keratinocyte p38 activation
by DMBA is also reduced. Particularly surprising, however,
was the observation that disruption of ask1 (with consequent degradation of ASK2) or ask1, plus ask2 did not
elevate DMBA/PMA-induced tumorigenesis as did disruption of ask2 alone. The reason for this is that ASK1 is
required for the induction, in skin, by DMBA/TPA of proinflammatory cytokines, notably TNF and IL-6 (126) (FIGURE 11D).
pose tissue, which can produce substantial levels TNF. The
source of these cytokines may be adipose cells themselves.
Indeed, a common histological event in burgeoning insulin
resistance is the recruitment by adipose cells of circulating
myeloid cells. These cells accumulate around stressed or
dying adipocytes to produce so-called crownlike structures
which, in turn, produce cytokines and other inflammatory
mediators that can influence metabolic control in ways that
are still poorly understood.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
FIGURE 12. JNK1 and metabolic disease. Note how in most instances, JNK1 acts to inhibit functions that
themselves prevent insulin resistance. The notable exception is hepatic JNK1. A: cellular mechanisms of
JNK-dependent metabolic dysfunction. High-fat diet (HFD) triggers either ER stress or activation of PKC-␨/␪dependent recruitment of JNK1. Activated JNK1 phosphorylates the IRS proteins blunting their Tyr phosphorylation, thereby preventing recruitment of PI-3-kinase-dependent metabolic functions and impairing insulin
action. B: JNK1 exerts multiple tissue-specific effects to mediate HFD-induced obesity and insulin resistance.
See text for details.
lin resistance in cellular systems and in genetic or HFDinduced models of rodent obesity (2, 3, 84, 116 –118).
Recently however, a knock-in of an IRS-1 (Ser307Ala)
mutation in the mouse established that the elimination of
this phosphorylation site is accompanied by worsened
insulin resistance and impaired insulin signaling, estab-
lishing that Ser307 phosphorylation is, surprisingly, a
positive regulatory event (43). Nevertheless, IRS-1 is
phosphorylated at multiple Ser/Thr residues in states of
insulin resistance, and it remains likely that some combination of these phosphorylations does confer negative
regulation (201) (FIGURE 12A).
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
725
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
3. A prominent role for JNK1 as a negative
regulator of insulin sensitivity in obesity: whole
body knockout and bone marrow transplantation
studies versus tissue-specific knockouts
The effect of jnk1 disruption on weight gain coincides with
a general reduced adiposity in all fat depots plus a reduction
in adipocyte size. Interestingly, disruption of jnk1 results in
no difference in plasma triglyceride, cholesterol, or FFA
concentrations, indicating that lipid metabolism, food intake, and absorption are not affected by disruption of jnk1
(111, 215) (FIGURE 12B).
ob/ob mice develop spontaneous obesity and type 2 diabetes due to loss of function mutations in the leptin receptor.
Crossing ob/ob mice with jnk1⫺/⫺ mice to produce combination jnk1⫺/⫺/ob/ob significantly reduces the extent of
weight gain compared with ob/ob mice. Likewise, jnk1⫺/
⫺/ob/ob mice also manifest reduced hyperinsulinemia and
hyperglycemia. All of the metabolic effects of the jnk1 deletion coincide with a reduced constitutive phosphorylation
of IRS1 at Ser307. Thus JNK1, but not JNK2 (the reasons
for this difference are unknown), appears to play an important role in the development of obesity-induced insulin resistance, possibly through direct inhibition of IRS recruitment (111).
Exactly what activates JNK1 in response to HFD/obesity is
somewhat unclear. Two mechanisms are emerging as reasonable candidates. The first is elevated circulating triglycerides or FFAs acting as direct agonists for pattern recognition receptors. This is discussed below and in more detail in
section IVE5. The second mechanism involves FFAs triggering ER stress (FIGURE 12A).
ER stress is incurred by elevations in circulating lipids, and
obesity places stress on the protein synthetic machinery, of
which the ER is an integral part. ER stress is marked by
phosphorylation and activation of the protein kinase-R-like
kinase (PERK) and phosphorylation of its substrate translational initiation factor 2␣ (eIF2␣). Both of these phosphorylation events, as well as phosphorylation of the JNK
and c-Jun polypeptides at the activating sites (Ser63/73;
726
In the effort to better understand how active JNK promotes
whole body insulin resistance, a variety of tissue-specific
jnk1 and dual jnk1/jnk2 mouse knockouts have been evaluated. The phenotypes exhibited by these models point out
the widespread impact of JNK in metabolic regulation as
well as the complex adaptive and maladaptive mechanisms
that operate in the wake of selective interruption of such a
key regulator.
The most straightforward phenotype is exhibited by the
adipose tissue-specific deletion of JNK1 (AdT jnk1KO).
These mice exhibit essentially normal adipose development
and when placed on a HFD, gain weight similarly to their
jnk1-replete littermates. Nevertheles, the HFD-fed mice
with jnk1-deficient adipose tissue are essentially protected
from the development of whole body insulin resistance, as
exhibited by insulin tolerance test. This improvement is due
entirely to improved insulin action at the liver; muscle glucose uptake showed little change. HFD increases the expression of a variety of proinflammatory adipokines such as
TNF and IL-6; the jnk1-deficient adipose tissue exhibited a
marked reduction in IL-6 mRNA and IL-6 blood levels were
also significantly reduced, whereas expression of TNF and
resistin were unaltered by jnk1 deficiency. SOCS3, an inhibitor of insulin receptor and IRS-1 signaling, is strongly induced in liver by a HFD through IL-6, and the livers of the
HFD-fed AdT jnk1 knockout mice exhibited markedly reduced levels of SOCS3 and enhanced insulin receptor tyrosine phosphorylation compared with the HFD-fed, jnk1sufficient littermates. Thus deletion of jnk1 from adipose
tissue greatly enhances insulin responsiveness in liver by a
hormonal mechanism (241) (FIGURE 12B).
Obesity results in JNK activation in skeletal muscle, the
tissue that is the major site of glucose consumption in vivo.
As expected, JNK1 inactivation in skeletal muscle is accompanied by an improvement in insulin sensitivity in mice fed
a HFD; unexpectedly however, there is no difference in
weight gain or improvement in glucose tolerance, despite
the protection from muscle insulin resistance and the preserved muscle glucose uptake. jnk1-deficient muscle exhibits markedly reduced lipoprotein lipase, and circulating triglyceride levels are significantly increased in mice with muscle JNK1 deficiency. These mice develop hepatic steatosis
and evidence of inflammation in liver and adipose tissue;
hepatic glucose output and adipose tissue glucose utilization remain insulin resistant, presumably offsetting the beneficial response of skeletal muscle (243) (FIGURE 12B).
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Disruption of jnk1 also improves the balance in the levels of
hormones implicated in insulin sensitivity. Thus the ratio of
30-kDa adipocyte complement-related protein (ACRP30)
to adiponectin concentration (a measure of the endocrine
regulation of fatty acid oxidation) is higher in jnk1⫺/⫺
mice, while resistin (a hormone implicated in insulin resistance) levels are lower. In line with this, HFD-fed wild-type
mice developed mild hyperglycemia, while jnk1⫺/⫺ mice
had significantly lower blood glucose under identical dietary conditions. Moreover, the obese wild-type but not the
jnk1⫺/⫺ mice also developed hyperinsulinemia, and the
jnk1⫺/⫺ mice fared better in intraperitoneal insulin and
glucose tolerance tests (111) (FIGURE 12B).
sect. IID2) are strikingly elevated in hepatic tissue from
obese mice. Not surprisingly, in Fao hepatoma cells, this
phosphorylation coincides with decreased IRS1 levels and
Tyr phosphorylation as well as phosphorylation and activation of the Akt protein kinase, a major insulin target
downstream of PI-3-kinase (111, 215) (FIGURE 12A).
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
The importance of hepatic insulin resistance in obesity
would suggest that liver-specific inactivation of jnk1 would
have a substantial salutary effect. Moreover, expression of a
dominant inhibitory JNK1 in the liver of obese diabetic
mice does diminish hepatic glucose output, blood sugar,
and insulin resistance (268). Surprisingly however, hepatocyte-specific ablation of jnk1 results in liver steatosis, insulin resistance, and glucose intolerance (242). This discrepancy is unexplained; whether it indicates a role for hepatocyte JNK2 (silencing of jnk2 in adult mouse liver,
subsequent to HFD-induced steatosis, enhances hepatocyte
apoptosis; Ref. 268) or perhaps contributions of JNK1 acting in nonparenchymal cells (as occurs in the DEN-liver
cancer model; Ref. 51) to the genesis of hepatic insulin
resistance remains to be explored (FIGURE 12B).
It is notable that whereas the improved insulin sensitivity of
the HFD-fed mice with global disruption of jnk1 is accompanied by a resistance to weight gain, the improvement seen
in the HFD-fed adipose-specific jnk1 knockout occurs despite a weight gain comparable to the control HFD-fed
mice. The explanation for the ability of global jnk1 knockout to limit weight gain on a HFD appears due in part to the
impact of CNS jnk1 deficiency. A HFD is accompanied by
increased phosphorylation of hypothalamic JNK, and local
administration of a JNK inhibitor reduces food intake and
restores sensitivity of insulin-induced inhibition of food intake (61). Two groups (13, 244) reported the phenotype of
mice in which jnk1 was disrupted selectively in the nervous
system by crossing floxed-jnk1 mice with mice bearing a
Cre recombinase construct driven by the nestin promoter
(Nestin-Cre mice). jnk1 disruption in the CNS produced
mice which, when fed a standard chow diet exhibited a
Other abnormalities in pituitary-hypothalamic function are
evident in the mice with CNS jnk1 deficiency, e.g., somewhat reduced growth hormone and insulin-like growth factor I levels. However, the major factor in the protection
from obesity is likely the activation of the hypothalamicpituitary-thyroid axis (13, 244).
On the basis of the data described, inhibition of JNK activity is currently viewed as a reasonable strategy for the treatment of type 2 diabetes, where obesity and insulin resistance
are major factors. It will be of considerable interest to determine whether potent, selective JNK inhibitors, administered systemically, affect glucose, lipid, and energy metabolism in humans, and if so, by what mechanisms.
5. Upstream signaling elements coupling HFD and
obesity to JNK1
Although few studies have emerged to provide insight into
the molecular components that recruit JNK in metabolic
disease, it is becoming evident that innate immune signaling
mechanisms are key to this process.
Using knockout mice, Greenberg and colleagues (221) have
shown that disruption of the MAP3K tpl-2, while having no
effect on HFD-induced body weight gain or adipose depot
mass, substantially improves insulin sensitivity (hyperinsulinemic-euglycemic clamps) and reduces hepatic gluconeogenesis. This improvement coincides with reduced hepatic
steatosis and a substantial reduction in inflammatory markers (reduced adipose tissue crownlike structures, reduced
hepatic and macrophage cytokine production). This reduction in obesity-induced insulin resistance and inflammation upon tpl-2 disruption is accompanied by a reduction
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
727
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Obesity, whether genetic or induced by HFD, is accompanied by the infiltration of adipose tissues by cells of the
myeloid lineage, which become activated, produce inflammatory cytokines, and are thought to contribute thereby to
a low-grade systemic inflammatory state (156). Consequently, three reports have examined the contribution of
JNK1 in myeloid lineages to the insulin resistance of obesity, whether by engineering myeloid specific JNK1 knockout, by reconstituting irradiated wild-type recipients with
JNK1-deficient bone marrow or reciprocally, by giving WT
bone marrow to irradiated, globally JNK1-deficient recipients. Whereas Solinas et al. (271) reported that JNK1-deficient bone marrow protected irradiated recipients from
HFD-induced insulin resistance without altering adiposity,
two other groups failed to observe altered glycemic regulation and insulin sensitivity in response to HFD either with
myeloid-specific deletion of JNK1 (241) or restoration of
wild-type versus jnk1⫺/⫺ bone marrow to irradiated mice
(271, 299). Thus, although the data are somewhat conflicting, on balance, it is unlikely that myeloid JNK1 contributes
significantly to systemic insulin resistance in obesity (FIGURE 12B).
somewhat smaller body length and reduced lean body mass.
However, when placed on a HFD, the mice with CNS jnk1
deficiency exhibited considerably less weight gain, improved insulin sensitivity in most organs and, when assayed
by insulin tolerance test, somewhat lower fasting glucose.
Glucose tolerance, revealed by glucose tolerance tests, were
similar compared with HFD-fed jnk1⫹/⫹ controls (13,
244). The protection from adiposity resulting from HFD
reflected a slightly lesser food intake, but a markedly increased level of energy expenditure, due largely to an activation of the hypothalamic-pituitary-thyroid axis (FIGURE
12B). The mice with CNS jnk1 deficiency showed increased
levels of the circulating thyroid hormones T3, T4, and pituitary thyroid stimulating hormone; hypothalamic levels of
thyroid hormone releasing hormone (TRH) mRNA were
unaltered, but TRH receptor mRNA levels were considerably increased (13, 244). Belgardt et. al. (13) also showed
that treatment of the rat pituitary-derived cell line GH4C1
with the (admittedly nonspecific; Refs. 8, 9, 54) JNK inhibitor SP600125 gave a 60% increase in TRH receptor
mRNA, whereas inhibitors of ERK or PI3K were without
effect.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
in HFD-induced gonadal fat tissue JNK and ERK activation (221). These findings are consistent with the established role of Tpl-2 as an important MAPK regulator
recruited by TNF and pattern recognition receptors (FIGURE 12, A AND B).
V. CONCLUDING REMARKS
When we completed the initial version of this review in
2001, many of the potential wiring diagrams for stressactivated MAPK pathways had been dissected at the biochemical level, and there was a clear need to place these
pathways into a physiological context to understand how
these pathways functioned in normal and pathophysiological circumstances. It has now become apparent that stressregulated MAPK pathways are critical both to normal physiology and to many diseases for which inflammation plays
an important role.
Inflammation is emerging as a central theme not only in
diseases caused by microbial pathogens, but in many of the
chronic conditions that continue to plague western cultures.
Thus elements of TLR signaling have been shown to be
required for type 2 diabetes, cancer, autoimmune disorders,
and atherosclerosis.
For several of these chronic conditions, a complete picture is
missing as to how proximal elements couple to more distal
elements to elicit the established pathology. Type 2 diabetes
is perhaps the best characterized of these conditions in
terms of linking both proximal (TLR, MyD88) and distal
(JNK1) elements to the disease as it manifests itself in numerous tissues. We also know that Tpl-2-dependent ERK
activation contributes to pathology. However, it is still unclear how TLRs/MyD88 link to JNK1, or the receptor complexes to Tpl-2, in different tissues and how these pathways
integrate to bring about the disease phenotype.
For other chronic conditions, roles for MAPKs are not
clearly established, despite an unambiguous link to inflam-
728
Finally, there is a tendency for researchers in a field to
believe that their field holds the panacea for all ills. While it
is true that stress- and inflammatory pathway-activated
MAPKs have been linked to numerous pathologies, caution
is warranted when considering these MAPK pathways or,
indeed, any pathway, as a therapeutic target. As we have
noted, for example, while inhibition of p38/MK2 may prevent chronic inflammation (e.g., arthritis), it can also lead to
an enhanced susceptibility to infectious disease. With the
advent of improved mouse models, genomic, proteomic,
and metabolomic studies, we anticipate considerable progress in these areas.
ACKNOWLEDGMENTS
Address for reprint requests and other correspondence:
J. M. Kyriakis, Molecular Cardiology Research Institute,
Tufts Medical Center, 800 Washington St., Box 8486, Boston, MA 02111 (e-mail: [email protected]).
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared
by the authors.
REFERENCES
1. Adams RH, Porras A, Alonso G, Jones M, Vintersten K, Panelli S, Valladares A, Perez
L, Klein R, Nebreda AR. Essential role of p38␣ MAP kinase in placental but not
embryonic cardiovascular development. Mol Cell 6: 109 –116, 2000.
2. Aguirre V, Uchida T, Yenush L, Davis R, White MF. The Jun NH2-terminal kinase
promotes insulin resistance during association with insulin receptor substrate-1 and
phosphorylation of Ser(307). J Biol Chem 275: 9047–9054, 2000.
3. Aguirre V, Werner ED, Giraud J, Lee YG, Shoelson SE, White MF. Phosphorylation of
Ser307 in insulin receptor substrate-1 blocks interactions with the insulin receptor
and inhibits insulin action. J Biol Chem 277: 1531–1537, 2002.
4. Alzabin S, Bhardwaj N, Kiefer F, Sawasdikosol S, Burakoff S. Hematopoietic progenitor kinase 1 is a negative regulator of dendritic cell activation. J Immunol 182: 6187–
6194, 2009.
5. Anafi M, Kiefer F, Gish GD, Mbamalu G, Iscove NN, Pawson T. SH2/SH3 adaptor
proteins can link tyrosine kinases to a Ste20-related protein kinase, HPK1. J Biol Chem
272: 27804 –27811, 1997.
6. Anaieva O, Darragh J, Johansen C, Carr JM, McIlrath J, Park JM, Wingate A, Monk CE,
Toth R, Santos SG, Iverson L, Arthur JSC. The kinases MSK1 and MSK2 act as negative
regulators of Toll-like receptor signaling. Nat Immunol 9: 1028 –1036, 2008.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
It is likely, however, that other mechanisms may couple
metabolic stress signals to JNK and p38. Thus, in MEFs,
addition of FFAs activates JNK by a process that requires
MLK3. This process also requires PKC-␨ insofar as
pkc␨⫺/⫺ MEFs fail to display FFA activation of JNK or
MLK3. In white and brown adipose tissue, HFD results in
JNK and MLK3 activation. Disruption of mlk3 reduces
HFD-induced JNK activation in brown but not white adipose tissue, although it is unclear what cells or agonists are
recruiting JNK in this instance. Interestingly, disruption of
mlk3 also reduces accumulation of fat in brown but not
white adipose tissue. Given that MLK3 and MLK2 often
manifest overlapping roles, it is possible that these kinases
serve redundant functions that would not be apparent upon
disruption of mlk3 alone (18, 129).
matory pathways that recruit MAPKs. Thus, for example,
atherogenesis has been linked to dysfunctional recruitment
of TLR-2, -4, and -6, as well as MyD88. Moreover, engagement of these receptors activates JNK and p38. However,
we still do not know the required elements needed to link
the receptor complexes to MAPKs and trigger atherogenesis. JNK and its upstream elements have been linked both to
tumor promotion and suppression; nevertheless, the mechanisms by which these enzymes are activated in different
tumor contexts remain nebulous.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
7. Arch RH, Gedrich RW, Thompson CB. Tumor necrosis factor receptor-associated
factors (TRAFs): a family of adapter proteins that regulates life and death. Genes Dev
12: 2821–2830, 2000.
28. Chang L, Jones Y, Ellisman MH, Goldstein LS, Karin M. JNK1 is required for maintenance of neuronal microtubules and controls phosphorylation of microtubule-associated proteins. Dev Cell 4: 521–533, 2003.
8. Bain J, McLauchlan H, Elliott M, Cohen P. The specificities of protein kinase inhibitors:
an update. Biochem J 371: 199 –204, 2003.
29. Chang X, Liu F, Wang X, Lin A, Zhao H, Su B. The kinases MEKK2 and MEKK3
regulate transforming growth factor-␤-mediated helper T cell differentiation. Immunity 34: 201–212, 2011.
9. Bain J, Plater L, Elliott N, Shpiro N, Hastie CJ, McLauchlan H, Klevernic I, Arthur JS,
Alessi DR, Cohen P. The selectivity of protein kinase inhibitors: a further update.
Biochem J 408: 297–315, 2007.
10. Baud V, Liu Zg Bennett B, Suzuki N, Xia Y, Karin M. Signaling by proinflammatory
cytokines: oligomerization of TRAF2 and TRAF6 is sufficient for JNK and IKK activation and target gene induction via an amino terminal effector domain. Genes Dev
13:1297–1308, 1999.
11. Becker E, Huynh-Do U, Holland S, Pawson T, Daniel TO, Skolnik EY. Nck interacting
Ste20 kinase couples Eph receptors to c-Jun N-terminal kinase and integrin activation.
Mol Cell Biol 20: 1537–1545, 2000.
13. Belgardt BF, Mauer J, Wunderlich FT, Ernst MB, Pal M, Spohn G, Brönneke HS,
Brodesser S, Hampel B, Schauss AC, Brüning JC. Hypothalamic and pituitary c-Jun
N-terminal kinase 1 signaling coordinately regulates glucose metabolism. Proc Natl
Acad Sci USA 107: 6028 – 6033, 2010.
14. Ben-Levy R, Leighton IA, Doza YN, Attwood P, Morrice N, Marshall CJ, Cohen P.
Identification of novel phosphorylation sites required for activation of MAPKAP kinase-2. EMBO J 14: 5920 –5930, 1995.
15. Biondi RM, Nebreda AR. Signalling specificity of Ser/Thr protein kinases through
docking-site-mediated interactions. Biochem J 372: 1–13, 2003.
16. Blank JL, Gerwins P, Elliot EM, Sather S, Johnson GL. Molecular cloning of mitogen
activated protein/ERK kinase kinases (MEKK) 2 and 3. J Biol Chem 271: 5361–5368,
1996.
17. Boyle WJ, Smeal T, Defize LH, Angel P, Woodgett JR, Karin M, Hunter T. Activation
of protein kinase C decreases phosphorylation of c-Jun at sites that negatively regulate
its DNA binding activity. Cell 64: 573–584, 1991.
18. Brancho D, Ventura JJ, Jaeschke A, Doran B, Flavell RA, Davis RJ. Role of MLK3 in the
regulation of mitogen-activated protein kinase signaling cascades. Mol Cell Biol 25:
3670 –3681, 2005.
19. Brodsky IE, Medzhitov R. Targeting of immune signaling networks by bacterial pathogens. Nat Cell Biol 11: 521–526, 2009.
20. Burbelo PD, Drechsel D, Hall A. A conserved binding motif defines numerous candidate target proteins for both Cdc42 and Rac GTPases. J Biol Chem 270: 29071–29074,
1995.
21. Buxade M, Parra-Palau JL, Proud CG. The Mnks: MAP kinase-interacting kinases
(MAP kinase signal-integrating kinases). Front Biosci 13: 5359 –5373, 2008.
22. Cardone MH, Salveson GS, Widmann C, Johnson G, Frisch S. The regulation of
anoikis: MEKK-1 activation requires cleavage by caspases. Cell 90: 315–323, 1997.
23. Carlson CJ, Koterski S, Sciotti RJ, Braillard Poccard G, Rondinone CM. Potential role
of p38 in the downregulation of GLUT4 expression. Diabetes 52: 634 – 641, 2003.
24. Cavigelli M, Dolfi F, Claret FX, Karin M. Induction of c-fos expression through JNKmediated TCF/Elk-1 phosphorylation. EMBO J 14: 5957–5964, 1995.
25. Ceci JD, Patriotis CP, Tsatsanis C, Makris AM, Kovatch R, Swing DA, Jenkins NA,
Tsichlis PN, Copeland NG. Tpl-2 is an oncogenic kinase that is activated by carboxyterminal truncation. Genes Dev 11: 688 –700, 1997.
26. Cellurale C, Sabio G, Kennedy NJ, Das M, Barlow M, Sandy P, Jacks T, Davis RJ.
Requirement of c-Jun NH2-terminal kinase for Ras-initiated tumor formation. Mol Cell
Biol 31: 1565–1576, 2011.
27. Chadee DN, Yuasa T, Kyriakis JM. Direct activation of mitogen-activated protein
kinase kinase kinase MEKK1 by the Ste20p homologue GCK and the adapter protein
TRAF2. Mol Cell Biol 22: 737–749, 2002.
31. Chen YR, Meyer CF, Tan TH. Persistent activation of c-Jun N-terminal kinase-1
(JNK1) in ␥ radiation induced apoptosis. J Biol Chem 271: 631– 634, 1996.
32. Chen Z, Hutchinson M, Cobb MH. Isolation of the protein kinase TAO2 and identification of its mitogen-activated protein kinase/extracellular signal-regulated kinase
kinase binding domain. J Biol Chem 274: 28803–28807, 1999.
33. Cheng L, Karin M. Mammalian MAP kinase signaling cascades. Nature 410: 37– 40,
2001.
34. Cheung PC, Nebreda AR, Cohen P. TAB3, a new binding partner of the protein kinase
TAK1. Biochem J 378: 27–34, 2004.
35. Chi H, Lu B, Takekawa M, Davis RJ, Flavell RA. GADD45␤/GADD45␥ and MEKK4
comprise a genetic pathway mediating STAT4-independent IFN-␥ production in T
cells. EMBO J 23: 1576 –1586, 2004.
36. Chow CW, Rincón M, Cavanagh J, Dickens M, Davis RJ. Nuclear accumulation of
NFAT4 opposed by the JNK signal transduction pathway. Science 278: 1638 –1641,
1997.
37. Chow CW, Dong C, Flavell RA, Davis RJ. c-Jun NH2-terminal kinase inhibits targeting
of the protein phosphatase calcineurin to NFATc1. Mol Cell Biol 20: 5227–5234, 2000.
38. Cross DAE, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 378: 785–789,
1995.
39. Clark A, Dean J, Tudor C, Saklatvala J. Post-transcriptional gene regulation by MAP
kinases via AU-rich elements. Front Biosci 14: 847– 871, 2009.
40. Coban C, Ishii KJ, Akira S. Immune interventions of human diseases through toll-like
receptors. Adv Exp Med Biol 655: 63– 80, 2009.
41. Cohen ED, Tian Y, Morrisey EE. Wnt signaling: an essential regulator of cardiovascular
differentiation, morphogenesis and progenitor self-renewal. Development 135: 789 –
798, 2008.
42. Cohen P. Targeting protein kinases for the development of anti-inflammatory drugs.
Curr Opin Cell Biol 21: 317–325, 2009.
43. Copps KD, Hancer NJ, Opare-Ado L, Qiw Walsh C W, White MF. Irs1 serine 307
promotes insulin sensitivity in mice. Cell Metab 11: 84 –92, 2010.
44. Cuenda A, Alonso G, Morrice N, Jones M, Meier R, Cohen P, Nebreda A. Purification
and cDNA cloning of SAPKK3, the major activator of RK/p38 in stress- and cytokinestimulated monocytes and epithelial cells. EMBO J 15: 4156 – 4164, 1996.
45. Cuenda A, Cohen P, Buee-Scherrer V, Goedert M. Activation of stress-activated
protein kinase-3 (SAPK3) by cytokines and cellular stresses is mediated via SAPKK3
(MKK6); comparison of the specificities of SAPK3 and SAPK2 (RK/p38). EMBO J 16:
296 –305, 1997.
46. Curtiss LK, Tobias PS. Emerging role of Toll-like receptors in atherosclerosis. J Lipid
Res 50: S340 –S345, 2009.
47. Dai T, Rubie E, Franklin CC, Kraft A, Gillespie DA, Avruch J, Kyriakis JM, Woodgett JR.
Stress-activated protein kinases bind directly to the delta domain of c-Jun in resting
cells: implications for repression of c-Jun function. Oncogene 10: 849 – 855, 1995.
48. Dajas-Bailador F, Jones EV, Whitmarsh AJ. The JIP1 scaffold protein regulates axonal
development in cortical neurons. Curr Biol 18: 221–226, 2008.
49. Dalby KN, Morrice N, Caudwell FB, Avruch J, Cohen P. Identification of regulatory
phosphorylation sites in mitogen-activated protein kinase (MAPK)-activated protein
kinase-1a/p90rsk that are inducible by MAPK. J Biol Chem 273: 1496 –1505, 1998.
50. Dan I, Watanabe NM, Kusumi A. The Ste20 group kinases as regulators of MAP kinase
cascades. Trends Cell Biol 11: 220 –230, 2001.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
729
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
12. Beinke S, Robinson MJ, Hugunin M, Ley SC. Lipopolysaccharide activation of the
TPL-2/MAP2K/extracellular signal-regulated kinase mitogen-activated protein kinase
cascade is regulated by I␬B kinase-induced proteolysis of NF-␬B1 p105. Mol Cell Biol
24: 9658 –9667, 2004.
30. Chen CY, Del Gatto-Konczak F, Wu Z, Karin M. Stabilization of interleukin-2 mRNA
by the c-Jun NH2-terminal kinase pathway. Science 280: 1945–1949, 1998.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
51. Das M, Garlick DS, Greiner DL, Davis RJ. The role of JNK in the development of
hepatocellular carcinoma. Genes Dev 25: 634 – 645, 2011.
52. Das S, Cho J, Lambertz I, Kelliher MA, Eliopoulos AG, Du K, Tsichlis PN. Tpl2/Cot
signals activate ERK, JNK and NF-␬B in a cell-type and stimulus-specific manner. J Biol
Chem 280: 23748 –23757, 2005.
53. Davidson SM, Morange M. Hsp25 and the p38 MAPK pathway are involved in differentiation of cardiomyocytes. Dev Biol 218: 146 –160, 2000.
72. Ellinger-Ziegelbauer H, Kelly K, Siebenlist U. Cell cycle arrest and reversion of Rasinduced transformation by a conditionally activated form of mitogen-activated protein
kinase kinase kinase 3. Mol Cell Biol 19: 3857–3868, 1999.
73. Enzler T, Chang X, Facchinetti V, Melino G, Karin M, Su B, Gallagher E. MEKK1 binds
HECT E3 ligase itch by its amino-terminal RING motif to regulate Th2 cytokine gene
expression. J Immunol 183: 3831–3838, 2009.
54. Davies SP, Reddy H, Caivano M, Cohen P. Specificity and mechanism of action of
some commonly used protein kinase inhibitors. Biochem J 351: 95–105, 2000.
74. Erickson AK, Payne DM, Martino PA, Rossomando AJ, Shabanowitz J, Weber MJ,
Hunt DF, Sturgill TW. Identification by mass spectrometry of threonine 97 in bovine
myelin basic protein as a specific phosphorylation site for mitogen-activated protein
kinase. J Biol Chem 265: 19728 –19735, 1990.
55. Deacon K, Blank JL. MEK kinase 3 directly activates MKK6 and MKK7, specific activators of the p38 and c-Jun NH2-terminal kinases. J Biol Chem 274: 16604 –16610,
1999.
75. Eyers PA, Craxton M, Morrice N, Cohen P, Goedert M. Conversion of SB 203580insensitive MAP kinase family members to drug-sensitive forms by a single amino-acid
substitution. Chem Biol 5: 321–329, 1998.
56. Deak M, Clifton AD, Lucocq J, Alessi DR. Mitogen- and stress-activated protein
kinase-1 (MSK1) is directly activated by MAPK and SAPK2/p38, may mediate activation of CREB. EMBO J 17: 4426 – 4441, 1998.
76. Fanger GR, Johnson NL, Johnson GL. MEK kinases are regulated by EGF and selectively interact with Rac/Cdc42. EMBO J 16: 4961– 4972, 1997.
58. Delpire E. The mammalian family of sterile 20p-like protein kinases. Pflügers Arch 458:
953–967, 2009.
59. Dérijard B, Hibi M, Wu LH, Barrett T, Su B, Deng T, Karin M, Davis RJ. JNK1: a protein
kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun
transactivation domain. Cell 76: 1025–1037, 1994.
60. Dérijard B, Raingeaud J, Barrett T, Wu LH, Han J, Ulevitch RJ, Davis RJ. Independent
human MAP kinase signal transduction pathways defined by MEK and MKK isoforms.
Science 267: 682– 685, 1995.
61. De Souza CT, Araujo EP, Bordin S, Ashimine Zollner RL R, Boschero AC, Saad MJ,
Velloso LA. Consumption of a fat-rich diet activates a proinflammatory response and
induced insulin resistance in the hypothalamus. Endocrinology 146: 4192– 4199, 2005.
62. Di Bartolo V, Bontagne B, Salek M, Hjungwirth B, Carrette F, Fourtane J, Sol-Foulon
N, Michel F, Schwartz O, Lehmann WD, Acuto O. A novel pathway down-modulating
T cell activation involves HPK-1-dependent recruitment of 14 –3-3 proteins on SLP76. J Exp Med 204: 681– 691, 2007.
63. Dickens M, Rogers JS, Cavanagh J, Raitano A, Xia Z, Halpern JR, Greenberg ME,
Sawyers CL, Davis RJ. A cytoplasmic inhibitor of the JNK signal transduction pathway.
Science 277: 693– 696, 1997.
64. Diener K, Wang XS, Chen C, Meyer CF, Keesler G, Zukowski M, Tan TH, Yao Z.
Activation of the c-Jun N-terminal kinase pathway by a novel protein kinase related to
human germinal center kinase. Proc Natl Acad Sci USA 94: 9687–9692, 1997.
77. Fanger GR, Widmann C, Porter AC, Sather S, Johnson GL, Vaillancourt RR. 14 –3-3
proteins interact with specific MEK kinases. J Biol Chem 273: 3476 –3483, 1998.
78. Finegan KG, Tournier C. The mitogen-activated protein kinase kinase 4 has a prooncogenic role in skin cancer. Cancer Res 70: 5797–5806, 2010.
79. Fornace AJ Jr, Alamo I Jr, Hollander C. DNA damage-inducible transcripts in mammalian cells. Proc Natl Acad Sci USA 85: 8800 – 8804, 1988.
80. Freshney NW, Rawlinson L, Guesdon F, Jones E, Cowley S, Hsuan J, Saklatvala J.
Interleukin-1 activates a novel protein kinase cascade that results in the phosphorylation of Hsp27. Cell 78: 1039 –1049, 1994.
81. Francisco JN, Eric N. MEF2: a transcriptional target for signaling pathways controlling
skeletal muscle growth and differentiation. Curr Opin Cell Biol 11: 683– 688, 1999.
82. Fu CA, Shen M, Huang BCB, Lasaga J, Payan DG, Luo Y. TNIK, a novel member of the
terminal center kinase family that activates the c-Jun N-terminal kinase pathway and
regulates the cytoskeleton. J Biol Chem 274: 30729 –30737, 1999.
83. Fuchs SY, Dolan L, Davis RJ, Ronai Z. Phosphorylation-dependent targeting of c-Jun
ubiquitination by Jun N-kinase. Oncogene 13: 1531–1535, 1996.
84. Fujishiro M, Gotoh Y, Katagiri H, Sakoda H, Ogihara T, Anai M, Onishi Y, Ono H, Abe
M, Shojima N, Fukushima Y, Kikuchi M, Oka Y, Asano T. Three mitogen-activated
protein kinases inhibit insulin signaling by different mechanisms in 3T3–L1 adipocytes.
Mol Endocrinol 17: 487– 497, 2001.
85. Fukunaga R, Hunter T. MNK1, a new MAP kinase-activated protein kinase, isolated by
a novel expression screening method for identifying protein kinase substrates. EMBO
J 16: 1921–1933, 1997.
65. Dodeller F, Schulze-Koops H. The p38 mitogen-activated protein kinase signaling
cascade in CD4 T cells. Arthritis Res Ther 8: 205, 2006.
86. Furic L, Rong L, Larsson O, Koumakpayi Yoshida K IH, Brueschke A, Petroulakis E,
Robichaud N, Pollak M, Gaboury LA, Pandolfi PP, Saad F, Sonenberg N. eIF4E phosphorylation promotes tumorigenesis and is associated with prostate cancer progression. Proc Natl Acad Sci USA 107: 14134 –14139, 2010.
66. Dong C, Yang DD, Wysk M, Whitmarsh AJ, Davis RJ, Flavell RA. Defective T cell
differentiation in the absence of JNK1. Science 282: 2092–2095, 1998.
87. Gaestel M. MAPKAP kinases-MKs-two’s company, three’s a crowd. Nat Rev Mol Cell
Biol 7: 120 –130, 2006.
67. Dong C, Yang DD, Tournier C, Whitmarsh AJ, Xu J, Davis RJ, Flavell RA. JNK is
required for effector T-cell function but not for T cell activation. Nature 405: 91–94,
2000.
88. Gallagher E, Enzler T, Matsuzawa A, Anzelon-Mills A, Otero D, Holzer R, Janssen E,
Gao M, Karin M. Kinase MEKK1 is required for CD40-dependent activation of the
kinases JNK and p38, germinal center formation, B cell proliferation and antibody
production. Nat Immunol 8: 57– 63, 2007.
68. Dorow DS, Devereux L, Dietzsch E, De Kretser T. Identification of a new family of
human epithelial protein kinases containing two leucine/isoleucine-zipper domains.
Eur J Biochem 213: 701–710, 1993.
69. Dumitru CD, Ceci JD, Tsatsanis C, Kontoyiannis D, Stamatakis K, Lin JH, Patriotis C,
Jenkins NA, Copeland NG, Kollias G, Tsichlis PN. TNF-␣ induction by LPS is regulated
posttranscriptionally via a Tpl2/ERK-dependent pathway. Cell 103: 1071–1083, 2000.
70. Eisenberg C, Eisenberg L. WNT11 promotes cardiac tissue formation of early mesoderm. Dev Dyn 216: 45–58, 1999.
71. Ellinger-Ziegelbauer H, Brown K, Kelly K, Siebenlist U. Direct activation of the stressactivated protein kinase and extracellular signal-regulated protein kinase (ERK) pathways by an inducible mitogen-activated protein kinase/ERK kinase kinase 3 derivative.
J Biol Chem 272: 2668 –2674, 1997.
730
89. Gallo KA, Mark MR, Scadden DT, Wang Z, Gu Q, Godowski PJ. Identification and
characterization of SPRK, a novel src-homology 3 domain-containing proline-rich
kinase with serine/threonine kinase activity. J Biol Chem 269: 15092–15100, 1994.
90. Gallo KA, Johnson GL. Mixed-lineage kianse control of JNK and p38 pathways. Nat Rev
Mol Cell Biol 3: 663– 672, 2002.
91. Ganiatsas S, Kwee L, Fujiwara Y, Perkins A, Ikeda T, Labow MA, Zon LI. SEK1
deficiency reveals mitogen-activated protein kinase cascade crossregulation and leads
to abnormal hepatogenesis. Proc Natl Acad Sci USA 95: 6881– 6886, 1998.
92. Gao M, Labuda T, Xia Y, Gallagher E, Fang D, Liu YC, Karin M. Jun turnover is
controlled through JNK-dependent phosphorylation of the E3 ligase itch. Science 306:
271–275, 2004.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
57. Dean JL, Wait R, Mahtani KR, Sully G, Clark AR, Saklatvala J. The 3= untranslated
region of tumor necrosis factor alpha mRNA is a target of the mRNA-stabilizing factor
HuR. Mol Cell Biol 21: 721–730, 2001.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
93. Ge B, Gram H, Di Padova F, Huang B, New L, Ulevitch RJ, Luo Y, Han J. MAPKKindependent activation of p38␣ mediated by TAB1-dependent autophosphorylation
of p38␣. Science 295: 1291–1294, 2002.
94. Gerits N, Kostenko S, Moens U. In vivo functions of mitogen-activated protein kinases: conclusions from knock-in and knockout mice. Transgenic Res 16: 281–314,
2007.
95. Gerwins P, Blank JL, Johnson GL. Cloning of a novel mitogen-activated protein kinasekinase-kinase, MEKK4, that selectively regulates the c-Jun amino terminal kinase pathway. J Biol Chem 272: 8288 – 8295, 1997.
96. Gille H, Sharrocks AD, Shaw PE. Phosphorylation of transcription factor p62TCF by
MAP kinase stimulates ternary complex formation at c-fos promoter. Nature 358:
414 – 417, 1992.
by altering tristetraprolin expression, stability and binding to adenine/uridine-rich
element. Mol Cell Biol 26: 2399 –2407, 2006.
113. Hogan PG, Chen L, Nardone J, Rao A. Transcriptional regulation by calcium, calcineurin and NFAT. Genes Dev 17: 2205–2232, 2003.
114. Holtzman LB, Merritt SE, Fan G. Identification, molecular cloning and characterization
of dual leucine zipper bearing kinase. J Biol Chem 269: 30808 –30817, 1994.
115. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis
factor-␣: direct role in obesity-linked insulin resistance. Science 259: 87–91, 1993.
116. Hotamisligil GS, Murray DL, Spiegelman BM. Tumor necrosis factor ␣ inhibits signaling from the insulin receptor. Proc Natl Acad Sci USA 91: 4854 – 4858, 1994.
117. Hotamisligil GS, Budavari A, Murray D, Spiegelman BM. Reduced tyrosine kinase
activity of the insulin receptor in obesity-diabetes. Central role of tumor necrosis
factor ␣. J Clin Invest 94: 1543–1549, 1994.
98. Goedert M, Cuenda A, Craxton M, Jakes R, Cohen P. Activation of the novel stressactivated protein kinase SAPK4 by cytokines and cellular stresses is mediated by SKK3
(MKK6); comparison of its substrate specificity with that of other SAP kinases. EMBO
J 16: 3563–3571, 1997.
118. Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, Spiegelman BM. IRS-1mediated inhibition of insulin receptor tyrosine kinase activity in TNF-␣ and obesityinduced insulin resistance. Science 271: 665– 668, 1996.
99. Gordon MD, Nusse R. Wnt signaling: multiple pathways, multiple receptors, and
multiple transcription factors. J Biol Chem 281: 22429 –22433, 2006.
119. Hu MCT, Qiu WR, Wang X, Meyer CF, Tan TH. Human HPK1, a novel human
hematopoietic progenitor kinase that activates the JNK/SAPK kinase cascade. Genes
Dev 10: 2251–2264, 1996.
100. Gotoh Y, Cooper JA. Reactive oxygen species- and dimerization-induced activation of
apoptosis signal-regulating kinase 1 in tumor necrosis factor-␣ signal transduction. J
Biol Chem 273: 17477–17482, 1998.
120. Huang Q, Yang J, Lin Y, Walker C, Cheng J, Liu Zg, Su B. Differential regulation of
interleukin 1 receptor and Toll-like receptor signaling by MEKK3. Nat Immunol 5:
98 –103, 2004.
101. Gum RJ, McLaughlin MM, Kumar S, Wang Z, Bower MJ, Lee JC, Adams JL, Livi GP,
Goldsmith EJ, Young PR. Acquisition of sensitivity of stress-activated protein kinases
to the p38 inhibitor, SB 203580, by alteration of one or more amino acids within the
ATP binding pocket. J Biol Chem 273: 15605–15610, 1998.
121. Hui L, Bakiri L, Mairhorfer A, Schweifer N, Haslinger C, Kenner L, Komnenovic V,
Scheuch H, Beug H, Wagner EF. p38␣ suppresses normal and cancer cell proliferation
by antagonizing the JNK-c-Jun pathway. Nat Genet 39: 741–749, 2007.
102. Gupta S, Campbell D, Dérijard B, Davis RJ. Transcription factor ATF2 regulation by
the JNK signal transduction pathway. Science 267: 389 –393, 1995.
122. Hui L, Zatloukal K, Scheuch H, Stepniak E, Wagner EF. Proliferation of human HCC
cells and chemically induced mouse liver cancers requires JNK1-dependent p21
downregulation. J Clin Invest 118: 3943–3953, 2008.
103. Ha H, Han D, Choi Y. TRAF-mediated TNFR-family signaling. Curr Protocol Immunol
11: 9D, 2009.
104. Han J, Lee J-D, Bibbs L, Ulevitch RJ. A MAP kinase targeted by endotoxin and hyperosmolarity in mammalian cells. Science 265: 808 – 811, 1994.
105. Han J, Jiang Y, Li Z, Kravchenko VV, Ulevitch RJ. MEF2C participates in inflammatory
responses via p38-mediated activation. Nature 386. 563–566, 1997.
106. Hayakawa Y, Hirata Y, Nakagawa H, Sakamoto K, Hikiba Y, Otsuka M, Ijichi H,
Ikenoue T, Tateishi K, Akanuma M, Ogura K, Yoshida H, Ichijo H, Omata M, Maeda S.
Apoptosis signal-regulating kinase 1 regulates colitis and colitis-associated tumorigenesis by the innate immune responses. Gastroenterology 138: 1055–1067, 2009.
107. Hayakawa Y, Hirata Y, Nagakawa H, Sakamoto K, Hikiba Y, Kinoshita H, Nakata W,
Takahashi R, Tateishi K, Tada M, Akanuma M, Yoshida H, Takeda K, Ichijo H, Omata
M, Maeda S, Koike K. Apoptosis signal-regulating kinase 1 and cyclin D1 compose a
positive feedback loop contributing to tumor growth in gastric cancer. Proc Natl Acad
Sci USA 108: 780 –785, 2011.
108. Herskowitz I. MAP kinase pathways in yeast: for mating and more. Cell 80: 187–197,
1995.
109. Hirai Si Katoh M, Terada M, Kyriakis JM, Zon LI, Rana A, Avruch J, Ohno S. MST/
MLK2, a member of the mixed lineage kinase family, directly phosphorylates and
activates SEK1, an activator of c-Jun N-terminal kinase/stress-activated protein kinase. J Biol Chem 272: 15167–15173, 1997.
110. Hirai Si Noda K, Moriguchi T, Nishida E, Yamashita A, Deyama T, Fukuyama K, Ohno
S. Differential activation of two JNK activators, MKK7 and SEK1, by MKN28-derived
nonreceptor serine/threonine kinase/mixed lineage kinase 2. J Biol Chem 273: 7406 –
7412, 1998.
123. Huot J, Houle F, Marceau F, Landry J. Oxidative stress-induced actin reorganization
mediated by the p38 mitogen-activated protein kinase/heat shock protein 27 pathway
in vascular endothelial cells. Circ Res 80: 383–392, 1997.
124. Hutchinson M, Berman KS, Cobb MH. Isolation of TAO1, a protein kinase that
activates MEKs in stress-activated protein kinase cascades. J Biol Chem 273:
28625–28632.
125. Ichijo H, Nishida E, Irie K, ten Dijke P, Saitoh M, Moriguchi T, Takagi M, Matsumoto
K, Miyazono K, Gotoh Y. Induction of apoptosis by ASK1, a mammalian MAPKKK that
activates SAPK/JNK and p38 signaling pathways. Science 275: 90 –94, 1997.
126. Iriyama T, Takeda K, Nakamura H, Morimoto Y, Kuroiwa T, Mizukami J, Umeda T,
Noguchi T, Naguro I, Nishitoh H, Saegusa K, Tobiume K, Homma T, Shimada Y,
Tsuda H, Aiko S, Imoto Inazawa J I, Chida K, Kamei Y, Kozuma Taketani Y S, Matsuzawa A, Ichijo H. ASK1 and ASK2 differentially regulate the counteracting roles of
apoptosis and inflammation in tumorigenesis. EMBO J 28: 843– 853, 2009.
127. Ishitani T, Takaesu G, Ninomiya-Tsuji J, Shibuya H, Gaynor RB, Matsumoto K. Role of
TAB2-related protein TAB3 in IL-1 and TNF signaling. EMBO J 22: 6277– 6288, 2003.
128. Ito M, Yoshioka K, Akechi M, Yamashita S, Takamatsu N, Sugiyama K, Hibi M, Nakabeppu Y, Shiba T, Yamamoto KI. JSAP1, a novel Jun N-terminal protein kinase (JNK)binding protein that functions as a scaffold factor in the JNK signaling pathway. Mol Cell
Biol 19: 7539 –7548, 1999.
129. Jaeschke A, Davis RJ. Metabolic stress signaling mediated by mixed lineage kinases.
Mol Cell 27: 498 –508, 2007.
130. Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol 21:
247–269, 2005.
111. Hirosumi J, Tuncman G, Chang L, Görgün CZ, Uysal KT, Maeda K, Karin M, Hotamisligil G. A central role for JNK in obesity and insulin resistance. Nature 420: 334 –
337, 2002.
131. Janknecht R, Hunter T. Convergence of MAP kinase pathways on the ternary complex
factor Sap-1a. EMBO J 16: 1620 –1627, 1997.
112. Hitti E, Iakovleva T, Brook M, Deppenmeier S, Gruber AD, Radzioch D, Clark AR,
Blackshear PJ, Kotlyarov A, Gaestel M. Mitogen-activated protein kinase-activated
protein kinase 2 regulates tumor necrosis factor mRNA stability and translation mainly
132. Jiang Y, Chen Li Z C, Guo W, Gegner JA, Lin S, Han J. Characterization of the structure
and function of a new mitogen-activated protein kinase (p38␤). J Biol Chem 271:
17920 –17926, 1996.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
731
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
97. Glise B, Bourbon H, Noselli S. hemipterous encodes a novel Drosophila MAP kinase
kinase, required for epithelial cell sheet movement. Cell 83: 451– 461, 1995.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
133. Jiang Y, Gram H, Zhao M, New L, Gu J, Feng L, Di Padova F, Ulevitch RJ, Han J.
Characterization of the structure and function of the fourth member of the p38 group
mitogen-activated protein kinases, p38␦. J Biol Chem 272: 30122–30128, 1997.
155. Kim K, Duramad O, Qin XF, Su B. MEKK3 is essential for lipopolysaccharide-induced
interleukin-6 and granulocyte-macrophage colony-stimulating factor production in
macrophages. Immunology 120: 242–250, 2007.
134. Jin G, Klika A, Callahan M, Faga B, Danzig J, Jiang Z, Li X, Stark GR, Harrington J, Sherf
B. Identification of a human NF-␬B activating protein TAB3. Proc Natl Acad Sci USA
101: 2028 –2033, 2004.
156. Könner AC, Brüning JC. Toll-like receptor: linking inflammation to metabolism.
Trends Endocrinol 22: 16 –23, 2011.
135. Jirmanova L, Sarma DN, Jankovic D, Mittelstadt PR, Ashwell JD. Genetic disruption of
p38␣ Tyr323 phosphorylation prevents T-cell receptor-mediated p38␣ activation
and impairs interferon-␥ production. Blood 113: 2229 –2237, 2009.
136. Johnson GL. Defining MAPK interactomes. ACS Chem Biol 6: 18 –20, 2011.
157. Kontoyannis D, Boulougouris G, Manoloukos M, Armaka M, Apostolaki M, Pizzaro T,
Kotlyarov A, Foster I, Flavell R, Gaestel M, Tsichlis P, Cominelli F, Kollias G. Genetic
dissection of the cellular pathways and signaling mechanisms in modeled tumor necrosis factor-induced Crohn’s-like inflammatory bowel disease. J Exp Med 196: 1563–
1573, 2002.
137. Kallunki T, Su B, Tsigelny I, Sluss HK, Dérijard B, Moore G, Davis R, Karin M. JNK2
contains a specificity-determining region responsible for efficient c-Jun binding and
phosphorylation. Genes Dev 8: 2996 –3007, 1994.
158. Kopp E, Medzhitov R, Carothers J, Xiao C, Douglas I, Janeway C, Ghosh S. ECSIT is
an evolutionarily conserved intermediate in the Tol/IL-1 signal transduction pathway.
Genes Dev 13: 2059 –2071, 1999.
138. Kallunki T, Deng T, Hibi M, Karin M. c-Jun can recruit JNK to phosphorylate dimerization partners via specific docking interactions. Cell 87: 929 –939, 1996.
159. Kotlyarov A, Neininger C, Schubert R, Eckert C, Birchmeier HD, Volk M, Gaestel M.
MAPKAP kinase 2 is essential for LPS-induced TNF-␣ biosynthesis. Nat Cell Biol 1:
94 –97, 1999.
160. Kroemer G, Mariño G, Levine B. Autophagy and the integrated stress response. Mol
Cell 40: 280 –293, 2010.
140. Kanayama A, Seth RB, Sun L, Ea CK, Hong M, Shaito A, Chiu YH, Deng L, Chen ZJ.
TAB2 and TAB3 activate the NF-␬B pathway through binding to polyubiquitin chains.
Mol Cell 15: 535–548, 2004.
161. Kuan CY, Yang D, Samata Roy DR, Davis RJ, Rakic P, Flavell RA. The JNK1 and JNK2
protein kinases are required for regional specific apoptosis during early brain development. Neuron 22: 667– 676, 1999.
141. Kanety H, Feinstein R, Papa MZ, Karasik A. Tumor necrosis factor ␣-induced phosphorylation of insulin receptor substrate-1 (IRS-1). Possible mechanism for suppression of insulin-stimulated tyrosine phosphorylation of IRS-1. J Biol Chem 270: 23780 –
23784, 1995.
162. Kulathu Y, Akutsu M, Bremm A, Hofmann K, Komander D. Two sided ubiquitin
binding explains specificity of the TAB2 NZF domain. Nat Struct Mol Biol 16: 1328 –
1330, 2009.
142. Kang YJ, Seit-Nebi A, Davis RJ, Han J. Multiple activation mechanisms of p38 alpha
mitogen-activated protein kinase. J Biol Chem 281: 26225–26234, 2006.
163. Kyriakis JM, Avruch J. pp54 MAP2 kinase, a novel serine/threonine protein kinase
regulated by phosphorylation and stimulated by poly-L-lysine. J Biol Chem 265: 17355–
17363, 1990.
143. Karin M, Gallagher E. From JNK to pay dirt: jun kinases, their biochemistry, physiology
and clinical importance. IUBMB Life 57: 283–295, 2005.
144. Kärkkäinen S, van der Linden M, Renkema GH. POSH2 is a RING finger E3 ligase with
Rac1 binding activity through a partial CRIB domain. FEBS Lett 584: 3867–3872, 2010.
145. Kastan MB, Zhan Q, el-Diery WS, Carrier F, Jacks T, Walsh WV, Plunkett BS, Vogelstein B, Fornace AJ Jr. A mammalian cell cycle checkpoint pathway utilizing p53 and
GADD45 is defective in ataxia-telangiectasia. Cell 71: 587–597, 1992.
146. Katz P, Whalen G, Kehrl JH. Differential expression of a novel protein kinase in human
B lymphocytes: preferential localization in the germinal center. J Biol Chem 269:
16802–16809, 1994.
147. Kawai T, Akira S. The role of pattern-recognition receptors in innate immunity:
update on Toll-like receptors. Nat Immunol 11: 373–384, 2010.
148. Kelkar N, Gupta S, Dickens M, Davis RJ. Interaction of a mitogen-activated protein
kinase signaling module with the neuronal protein JIP3. Mol Cell Biol 20: 1030 –1043,
2000.
149. Kelkar N, Delmotte MH, Weston CR, Barrett T, Sheppard BJ, Flavell RA, Davis RJ.
Morphogenesis of the telencephalic commissure requires scaffold protein JNK-interacting protein 3 (JIP3). Proc Natl Acad Sci USA 100: 9843–9848, 2003.
150. Kelkar N, Standen CL, Davis RJ. Role of the JIP4 scaffold protein in the regulation of
mitogen-activated protein kinase signaling pathways. Mol Cell Biol 25: 2733–2743,
2005.
151. Kenny EF, O=Neill LA. Signalling adaptors used by Tol-like receptors: an update.
Cytokine 43: 343–350, 2008.
152. Keshet Y, Seger R. The MAP kinase signaling cascades: a system of hundreds of
components regulates a diverse array of physiological functions. Methods Mol Biol 661:
3–38, 2010.
153. Kharbanda S, Re R, Pandey P, Shafman TD, Feller S, Weichselbaum RR, Kufe DW.
Activation of the c-Abl tyrosine kinase in the stress response to DNA damaging
agents. Nature 376: 785–788, 1995.
154. Kiefer F, Tibbles LA, Anafi M, Janssen A, Zanke BW, Lassam N, Pawson T, Woodgett
JR, Iscove NR. HPK1, a hematopoietic protein kinase activating the SAPK/JNK pathway. EMBO J 15: 7013–7025, 1996.
732
164. Kyriakis JM, Brautigan DL, Ingebritsen TS, Avruch J. pp54 microtubule-associated
protein-2 kinase requires both tyrosine and serine/threonine phosphorylation for
activity. J Biol Chem 266: 10043–10046, 1991.
165. Kyriakis JM, Banerjee P, Nikolakaki E, Dai T, Rubie EA, Ahmad MF, Avruch J,
Woodgett JR. The stress-activated protein kinase subfamily of c-Jun kinases. Nature
369: 156 –160, 1994.
166. Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase pathways activated by stress and inflammation. Physiol Rev 81: 807– 869, 2001.
167. Lamarche N, Tapon N, Stowers L, Burbelo P, Aspenstrom P, Bridges T, Chant J, Hall
A. Rac and Cdc42 induce actin polymerization and G1 cell cycle progression independently of p65PAK and the JNK/SAPK MAP kinase cascade. Cell 87: 519 –529, 1996.
168. Lambert H, Charette SJ, Bernier AF, Guimond A, Landry J. HSP27 multimerization
mediated by phosphorylation-sensitive intermolecular interactions at the amino terminus. J Biol Chem 274: 9378 –9385, 1999.
169. Lange-Carter CA, Pleiman CM, Gardner AM, Blumer KJ, Johnson GL. A divergence in
the MAP kinase regulatory network defined by MEK kinase and Raf. Science 260:
315–319, 1993.
170. Lechner C, Zahalka MA, Giot JF, Moller NP, Ullrich A. ERK6, a mitogen-activated
protein kinase is involved in C2C12 myoblast differentiation. Proc Natl Acad Sci USA
93: 4355– 4359, 1996.
171. Lee JC, Laydon JT, McDonnell PC, Gallagher TF, Kumar S, Green D, McNulty D,
Blumenthal MJ, Heys JR, Landvatter SW, Strickler JE, McLaughlin MM, Siemens IR,
Fisher SM, Livi GP, White JR, Adams JL, Young P. A protein kinase involved in the
regulation of inflammatory cytokine biosynthesis. Nature 273: 739 –746, 1994.
172. Lei K, Davis RJ. JNK phosphorylation of Bim-related members of the Bcl2 family
induceds Bax-dependent apoptosis. Proc Natl Acad Sci USA 100: 2432–2437, 2003.
173. Leonardi A, Ellinger-Ziegelbauer H, Franzoso G, Brown K, Siebenlist U. Physical and
functional interaction of filamin (actin-binding protein-280) and tumor necrosis factor
receptor-associated factor 2. J Biol Chem 275: 271–278, 2000.
174. Levine B, Mizushima N, Virgin HW. Autophagy in immunity and inflammation. Nature
469: 323–335, 2011.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
139. Kanai-Azuma M, Kanai Y, Okamoto M, Hayashi Y, Yonekawa H, Yazaki K. Nrk: a
murine X-linked NIK (Nck-interacting kinase)-related kinase gene expressed in skeletal muscle. Mech Dev 89: 155–160, 1999.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
175. Leung IWL, Lassam N. Dimerization via tandem leucine zippers is essential for the
activation of the mitogen-activated protein kinase kinase kinase, MLK-3. J Biol Chem
273: 32408 –32415, 1998.
196. Minamino T, Yujiri T, Papst PJ, Chan ED, Johnson GL, Terada N. MEKK1 suppresses
oxidative stress-induced apoptosis of embryonic stem cell-derived cardiac myocytes.
Proc Natl Acad Sci USA 96: 15127–15132, 1999.
176. Ling P, Yao Z, Meyer CF, Wang XS, Oehrl W, Feller SM, Tan TH. Interaction of
hematopoietic progenitor kinase 1 with adapter proteins Crk and CrkL leads to
synergistic activation of c-Jun N-terminal kinase. Mol Cell Biol 19: 1359 –1368, 1999.
197. Minden A, Lin A, McMahon M, Lange-Carter C, Dérijard B, Davis RJ, Johnson GL,
Karin M. Differential activation of ERK and JNK mitogen-activated protein kinases by
Raf-1 and MEKK. Science 266: 1719 –1723, 1994.
177. Liou J, Kiefer F, Dant A, Hashimoto A, Cobb MH, Kurosaki T, Weiss A. HPK1 is
activated by lymphocyte antigen receptors and negatively regulates AP-1. Immunity
12: 399 – 408, 2000.
198. Minden A, Lin A, Claret FX, Abo A, Karin M. Selective activation of the JNK signaling
cascade and c-Jun transcriptional activity by the small GTPases Rac and Cdc42Hs. Cell
81: 1147–1157, 1995.
178. Liu H, Su YC, Becker E, Treisman J, Skolnik EY. A Drosophila TNF-receptor-associated factor bindds the Ste20 kinase Misshapen and activates Jun kinase. Curr Biol 9:
101–104, 1999.
199. Moore TM, Garg R, Johnson C, Coptcoat MJ, Ridley AJ, Morris JDH. PSK, a novel
STE20-like kinase derived from prostatic carcinoma that activates the c-Jun N-terminal kinase mitogen-activated protein kinase pathway and regulates actin cytoskeletal
organization. J Biol Chem 275: 4311– 4322, 2000.
179. Liu H, Nishitoh H, Ichijo H, Kyriakis JM. Activation of apoptosis signal-regulating
kinase-1 (ASK1) by TNF receptor-associated factor-2 requires prior dissociation of
the ASK1 inhibitor thioredoxin. Mol Cell Biol 20: 2198 –2208, 2000.
181. Lu HT, Yang DD, Wysk M, Gatti E, Mellman I, Davis RJ, Flavell RA. Defective IL-12
production in mitogen-activated protein (MAP) kinase kinase3 (mkk3)-deficient mice.
EMBO J 18: 1845–1857, 1999.
182. Lu X, Nemoto S, Lin A. Identification of c-Jun NH2-terminal protein kinase (JNK)activating kinase 2 as an activator of JNK but not p38. J Biol Chem 272: 24751–24754,
1997.
183. Lu Z, Xu S, Joazerio C, Cobb MH, Hunter T. The PHD domain of MEKK1acts as an
E.3 ubiquitin ligase and mediates ubiquitination and degradation of ERK1/2. Mol Cell 9:
945–956, 2002.
184. Macián F, López-Rodríguez C, Rao A. Partners in transcription: NFAT and AP-1.
Oncogene 20: 2476 –2489, 2001.
185. Mahtani KR, Brook M, Dean JL, Sully G, Saklatvala J, Clark AR. Mitogen-activated
protein kinase p38 controls the expression and posttranslational modification of tristetraprolin, a regulator of tumor necrosis factor alpha mRNA stability. Mol Cell Biol 21:
6461– 6469, 2001.
186. Malinin NL, Boldin MP, Kovalenko AV, Wallach D. MAP3K-related kinase involved in
NF-␬B induction by TNF, CD-95 and IL-1. Nature 385: 540 –544. 1997.
187. Marais R, Wynne J, Treisman R. The SRF accessory protein Elk-1 contains a growth
factor-regulated transcriptional activation domain. Cell 73: 381–393, 1993.
188. Marti A, Luo Z, Cunningham C, Ohta Y, Hartwig J, Stossel T, Kyriakis JM, Avruch J.
Actin-binding protein-280 binds the stress-activated protein kinase (SAPK) activator
SEK1 and is required for tumor necrosis factor-␣ activation of SAPK in mammalian
cells. J Biol Chem 272: 2620 –2628, 1997.
189. Massagué J, Seoane J, Wotton D. Smad transcription factors. Genes Dev 19: 2783–
2810, 2005.
190. Matsuzawa A, Tseng PH, Vallabhapurapu S, Luo JL, Zhang W, Wang H, Vignali DAA,
Gallagher E, Karin M. Essential cytoplasmic translocation of a cytokine receptorassembled signaling complex. Science 321: 663– 668, 2008.
191. Mayr B, Montminy M. Transcriptional regulation by the phosphorylation-dependent
factor CREB. Nat Rev Mol Cell Biol 2: 599 – 609.
192. McKay MM, Morrison DK. Integrating signals from RTKs to ERK/MAPK. Oncogene 26:
3113–3121, 2007.
193. McLaughlin MM, Kumar S, McDonnell PC, Van Horn S, Lee JC, Livi GP, Young PR.
Identification of mitogen-activated protein (MAP) kinase-activated protein kinase-3, a
novel substrate of CSBP p38 MAP kinase. J Biol Chem 271: 8488 – 8492, 1996.
194. Merten S, Craxton M, Goedert M. SAP kinase-3, a new member of the family of
mammalian stress-activated protein kinases. FEBS Lett 383: 273–276, 1996.
195. Meylan E, Tschopp J, Karin M. Intracellular pattern recognition receptors in the host
response. Nature 442: 39 – 44, 2006.
201. Morino K, Neschen S, Bilz S, Sono S, Tsirigotis D, Reznick RM, Moore I, Nagai Y,
Samuel V, Sebastian D, White M, Philbrick W, Shulman GI. Muscle-specific IRS-1 Ser
¡ Ala transgenic mice are protected from fat-induced insulin resistance in skeletal
muscle. Diabetes 57: 2644 –2651, 2008.
202. Morooka H, Bonventre JV, Pombo CM, Kyriakis JM, Force T. Ischemia and reperfusion enhance ATF-2 and c-Jun binding to cAMP response elements and to an AP-1
binding site from the c-Jun promoter. J Biol Chem 270: 30084 –30092, 1995.
203. Morrison DK, Davis RJ. Regulation of MAP kinase signaling modules by scaffold proteins in mammals. Annu Rev Cell Dev Biol 19: 91–118, 2003.
204. Musti AM, Treier M, Bohmann D. Reduced ubiquitin-dependent degradation of c-Jun
after phosphorylation by MAP kinases. Science 275: 400 – 402, 1997.
205. New L, Jiang Y, Zhao M, Liu K, Zhu W, Flood LJ, Kato Y, Parry GCN, Han J. PRAK, a
novel protein kinase regulated by the p38 MAP kinase. EMBO J 17: 3372–3384, 1998.
206. Newbern J, Taylor A, Robinson M, Lively MO, Milligan CE. c-Jun N-terminal kinase
signaling regulates events associated with both health and degeneration in motoneurons. Neuroscience 147: 680 – 692, 2007.
207. Ninomiya-Tsuji J, Kishioto K, Hiyama A, Inoue JI, Cao Z, andMatsumoto K. The kinase
TAK1 can activate the NIK-I␬B as well as the MAP kinase cascade in the IL-1 signalling
pathway. Nature 398: 252–256, 1999.
208. Nishina H, Fischer KD, Radvanyi L, Shahinian A, Razqallah H, Rubie EA, Bernstein A,
Mak TW, Woodgett JR, Penninger JM. Stress-signalling kinase Sek1 protects thymocytes from apoptosis mediated by CD95 and CD3. Nature 385: 350 –353, 1997.
209. Nishina H, Vaz A, Billia P, Nghiem M, Sasaki T, de la Pompa JL, Furlonger K, Paige C,
Hui C, Fischer KD, Kishimoto H, Iwatsubo T, Katada T, Woodgett JR, Penninger JM.
Defective liver formation and liver cell apoptosis in mice lacking the stress signaling
kinase SEK1/MKK4. Development 126: 505–516, 1999.
210. Nishito Y, Hasegawa M, Inohara N, Núñez G. MEX is a testis-specific E3 ubiquitin
ligase that promotes death receptor-induced apoptosis. Biochem J 396: 411– 417,
2006.
211. Nishitoh H, Saitoh M, Mochida Y, Takeda K, Nakano H, Rothe M, Miyazono K, Ichijo
H. ASK1 is essential for JNK/SAPK activation by TRAF2. Mol Cell 2: 389 –395, 1998.
212. Oliva AA Jr, Atkins CM, Copenagle L, Banker GA. Activated c-Jun N-terminal kinase
is requried for axon formation. J Neurosci 26: 9462–9470, 2006.
213. O’Neill LA, Dunne A, Edjeback M, Gray P, Jefferies C, Wietek C. Mal and MyD88:
adapter proteins involved in signal transduction by Toll-like receptors. J Endotoxin Res
9: 55–59, 2003.
214. Oyadomari S, Mori M. Roles of CHOP/GADD153 in endoplasmic reticulum stress.
Cell Death Differ 11: 389 –391, 2004.
215. Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Özdelen E, Tuncman G, Görgün
C, Glimcher LH, Hotamisligil G. Endoplasmic reticulum stress links obesity, insulin
action and type 2 diabetes. Science 306: 457– 461, 2004.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
733
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
180. Liu ZG, Baskaran R, Lea-Chou ET, Wood LD, Chen Y, Karin M, Wang JYJ. Three
distinct signalling responses by murine fibroblasts to genotoxic stress. Nature 384:
273–276, 1996.
200. Moriguchi T, Kuroyanagi N, Yamaguchi K, Gotoh Y, Irie K, Kano T, Shirakabe K, Muro
Y, Shibuya H, Matsumoto K, Nishida E, Hagiwara M. A novel kinase cascade mediated
by mitogen-activated protein kinase kinase 6 and MKK3. J Biol Chem 271: 13675–
13679, 1996.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
216. Pandey P, Raingeaud J, Kaneki M, Weichselbaum R, Davis RJ, Kufe D, Kharbanda S.
Activation of p38 mitogen-activated protein kinase by c-Abl-dependent and -independent mechanisms. J Biol Chem 271: 23755–23779, 1996.
217. Pandur P, Lasche M, Eisenberg LM, Kuhl M. Wnt-11 activation of a non-canonical Wnt
signalling pathway is required for cardiogenesis. Nature 418: 636 – 641, 2002.
218. Papoutsopoulou S, Symons A, Tharmalingham T, Belich MP, Kaiser F, Kioussis D,
O’Garra A, Tybulewicz V, Ley SC. ABIN-2 is required for optimal activation of Erk
MAP kinase in innate immune responses. Nat Immunol 7: 606 – 615, 2006.
219. Patriotis C, Makris A, Bear SE, Tsichlis PN. Tumor progression locus 2 (Tpl-2) encodes a protein kinase involved in the progression of rodent T-cell lymphomas and in
T-cell activation. Proc Natl Acad Sci USA 90: 2251–2255, 1993.
220. Paz K, Hemi R, LeRoith D, Karasik A, Elhanany E, Kanety H, Zick Y. A molecular basis
for insulin resistance. Elevated serine/threonine phosphorylation of IRS-1 and IRS-2
inhibits their binding to the juxtamembrane region of the insulin receptor and impairs
their ability to undergo insulin-induced tyrosine phosphorylation. J Biol Chem 272:
29911–29918, 1997.
222. Pineda G, Ea CK, Chen ZJ. Ubiquitination and TRAF signaling. Adv Exp Med Biol 597:
80 –92, 2007.
223. Pombo CM, Bonventre JV, Avruch J, Woodgett JR, Kyriakis JM, Force T. The stressactivated protein kinases are major c-Jun amino-terminal kinases activated by ischemia and reperfusion. J Biol Chem 269: 25546 –26551, 1994.
224. Pombo CM, Kehrl JH, Sánchez I, Katz P, Avruch J, Zon LI, Woodgett JR, Force T,
Kyriakis JM. Activation of the SAPK pathway by the human STE20 homologue germinal center kinase. Nature 377: 750 –754, 1995.
225. Pulverer BJ, Kyriakis JM, Avruch J, Nikolakaki E, Woodgett JR. Phosphorylation of
c-jun mediated by MAP kinases. Nature 353: 670 – 674, 1991.
226. Qin J, Yao J, Cui G, Xiao H, Kim TW, Fraczek J, Wightman P, Sato S, Akira S, Puel A,
Casanova JL, Su B, Li X. TLR8-mediated NF-␬B and JNK activation are TAK1-independent and MEKK3-dependent. J Biol Chem 281: 21013–21021, 2006.
227. Raingeaud J, Whitmarsh AJ, Barett T, Dérijard B, Davis RJ. MKK3- and MKK6-regulated gene expression is mediated by the p38 mitogen-activated protein kinase signal
transduction pathway. Mol Cell Biol 16: 1247–1255, 1996.
228. Rajasingh J, Bord E, Luedemann C, Asai J, Hamada H, Thorne T, Qin G, Goukassian D,
Zhu Y, Losordo DW, Kishore R. IL-10-induced TNF-␣ mRNA destabilization is mediated via IL-10 suppression of p38 MAP kinase activation and inhibition of HuR
expression. FASEB J 20: 2112–2114, 2006.
229. Raman M, Chen W, Cobb MH. Differential regulation and properties of MAPKs.
Oncogene 26: 3100 –3112, 2007.
230. Rana A, Gallo K, Godowski P, Hirai Si Ohno S, Zon LI, Kyriakis JM, Avruch J. The
mixed lineage protein kinase SPRK phosphorylates and activates the stress-activated
protein kinase activator, SEK1. J Biol Chem 271: 19025–19028, 1996.
231. Richards SA, Fu J, Romanelli A, Shimamura A, Blenis J. Ribosomal S6 kinase 1 (RSK1)
activation requires signals dependent on and independent of the MAP kinase ERK.
Curr Biol 9: 810 – 820, 1999.
232. Rincón M, Davis RJ. Regulation of the immune response by stress-activated protein
kinases. Immunol Rev 228: 212–224, 2009.
233. Ronkina N, Kotlyarov A, Dittrich-Breiholz O, Kracht M, Hitti E, Milarski K, Askew R,
Marusic S, Lin LL, Gaestel M, Telliez JB. The mitogen-activated protein kinase
(MAPK)-activated protein kinases MK2 and MK3 cooperate in stimulation of tumor
necrosis factor biosynthesis and stabilization of p38 MAPK. Mol Cell Biol 27: 170 –181,
2007.
234. Ronkina N, Menon MB, Schwermann J, Tiedje C, Hitti E, Kotlyarov A, Gaestel M.
MAPKAP kinases MK2 and MK3 in inflammation: complex regulation of TNF biosynthesis via expression and phosphorylation of tristetraprolin. Biochem Pharmacol 80:
1915–1920, 2010.
235. Rose BA, Force T, Wang Y. Mitogen-activated protein kinase signaling in the heart:
angels versus demons in a heart-breaking tale. Physiol Rev 90: 1507–1546, 2010.
734
237. Rouse J, Cohen P, Trigon S, Morange M, Alonso-Llamazares A, Zamanillo D, Hunt T,
Nebreda A. A novel kinase cascade triggered by stress and heat shock that stimulates
MAPKAP kinase-2 and phosphorylation of the small heat shock proteins. Cell 78:
1027–1037, 1994.
238. Rousseau S, Morrice N, Peggie M, Campbell DG, Gaestel M, Cohen P. Inhibition of
SAPK2a/p38 prevents hnRNP A0 phosphorylation by MAPKAP-K2 and its interaction
with cytokine mRNAs. EMBO J 21: 6505– 6514, 2002.
239. Sabapathy K, Hu Y, Kallunki T, Schreiber M, David JP, Jochum W, Wagner EF, Karin M.
JNK2 is required for efficient T-cell activation and apoptosis but not for normal
lymphocyte development. Curr Biol 9: 116 –125, 1999.
240. Sabapathy K, Jochum W, Hochedlinger K, Chang L, Karin M, Wagner EF. Defective
neural tube morphogenesis and altered apoptosis in the absence of both JNK1 and
JNK2. Mech Dev 89: 115–124, 1999.
241. Sabio G, Das M, Mora A, Zhang Z, Jun JY, Ko HJ, Barrett T, Kim JK, Davis RJ. A stress
signaling pathway in adipose tissue regulates hepatic insulin resistance. Science 322:
1539 –1543, 2008.
242. Sabio G, Cavanagh-Kyros J, Ko HJ, Jung DY, Gray S, Jun JY, Barrett T, Mora A, Kim JK,
Davis RJ. Prevention of steatosis by hepatic JNK1. Cell Metab 10: 491– 498, 2009.
243. Sabio G, Kennedy NJ, Cavanagh-Kyros J, Jung DY, Ko HJ, Ong H, Barrett T, Kim JK,
Davis RJ. Role of muscle c-Jun-NH2-terminal kinase 1 in obesity-induced insulin resistance. Mol Cell Biol 30: 106 –115, 2010.
244. Sabio G, Cavanagh-Kyros J, Barrett T, Jung DY, Ko HJ, Ong H, Morel C, Mora A, Reilly
J, Kim JK, Davis RJ. Role of the hypothalamic-pituitary-thyroid axis in metabolic regulation by JNK1. Genes Dev 24: 256 –264, 2010.
245. Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y, Kawabata M, Miyazono K, Ichijo H. Mammalian thioredoxin is a direct inhibitor of apoptosis signalregulating kinase (ASK) 1. EMBO J 17: 2596 –2606, 1998.
246. Saklatvala J. The p38 MAP kinase pathway as a therapeutic target in inflammatory
disease. Curr Opin Pharmocol 4: 372–377, 2004.
247. Sakurai T, Maeda S, Chang L, Karin M. Loss of hepatic NF-␬B activity enhances
chemical hepatocarcinogenesis through sustained c-Jun-N-terminal kinase activation.
Proc Natl Acad Sci USA 103: 10544 –10551, 2006.
248. Sakurai T, He G, Matsuzawa A, Yu GY, Maeda S, Hardiman G, Karin M. Hepatocyte
necrosis induced by oxidative stress and IL-1␣ release mediate carcinogen-induced
compensatory proliferation and liver tumorigenesis. Cancer Cell 14: 156 –165, 2008.
249. Salmerón A, Ahmad TB, Carlile GW, Pappin D, Narsimhan RP, Ley SC. Activation of
MEK-1 and SEK-1 by Tpl-2 proto oncoprotein, a novel MAP kinase kinase kinase.
EMBO J 15: 817– 826, 1996.
250. Salvador JM, Mittelstadt PR, Guszczynski T, Copeland TD, Yamaguchi H, Appella E,
Fornace AJ Jr, Ashwell JD. Alternative p38 activation pathway mediated by T cell
receptor-proximal tyrosine kinases. Nat Immunol 6: 390 –395, 2005.
251. Salvador JM, Mittelstadt PR, Belova GI, Fornace AJ Jr, Ashwell JD. The autoimmune
suppressor Gadd45␣ inhibits the T cell alternative p38 activation pathway. Nat Immunol 6: 396 – 402, 2005.
252. Sánchez I, Hughes RT, Mayer BJ, Yee K, Woodgett JR, Avruch J, Kyriakis JM, Zon LI.
Role of SAPK/ERK kinase-1 in the stress-activated pathway regulating transcription
factor c-Jun. Nature 372: 794 –798, 1994.
253. Sato S, Sanjo H, Takeda K, Ninomiya-Tsuji J, Yamamoto M, Kawai T, Matsumoto K,
Takeuchi O, Akira S. Essential function for the kinase TAK1 in innate and adaptive
immune responses. Nat Immunol 6: 1087–1095, 2005.
254. Sato Y, Yoshikawa A, Yamashita M, Yamagata A, Fukai S. Structural basis for specific
recognition of Lys 63-linked polyubiquitin chains by NZF domains of TAB2 and TAB3.
EMBO J 28: 3903–3909, 2009.
255. Schinkmann K, Blenis J. Cloning and characterization of a human STE-20-like protein
kinase with unusual cofactor requirements. J Biol Chem 272: 28695–28703, 1997.
256. Schreiber SL, Crabtree GR. The mechanism of action of cyclosporin A and FK506.
Immunol Today 13: 136 –142, 1992.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
221. Perfield JW, Lee Y, Shulman GI, Samuel VT, Jurczak MH, Ghang E, Xie C, Tsichlis PN,
Obin MS, Greenberg AS. Tumor progression locus 2 (TPL2) regulates obesity-associated inflammation and insulin resistance. Diabetes 60: 1168 –1176, 2011.
236. Roskoski R Jr. Raf protein-serine/threonine kinases: structure and regulation. Biochem
Biophys Res Commun 399: 313–317, 2010.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
257. Schuman J, Chen Y, Podd A, Yu M, Liu HH, Wen R, Chen ZJ, Wang D. A critical role
of TAK1 in B-cell receptor-mediated nuclear factor-␬B activation. Blood 113: 4566 –
4574, 2009.
258. Shafman TD, Saleem A, Kyriakis J, Weichselbaum R, Kharbanda S, Kufe DW. Defective induction of stress-activated protein kinase activity in ataxia-telangiectasia cells
exposed to ionizing radiation. Cancer Res 55: 3242–3245, 1995.
259. Shaulian E, Karin M. Stress-induced JNK activation is independent of gadd45 induction. J Biol Chem 274: 29595–29598, 1999.
277. Stossel TP, Hartwig JH, Janmey PA, Kwiatkowski DJ. Cell crawling two decades after
Abercrombie. Biochem Soc Symp 65: 267–280, 1999.
278. Su B, Jacinto E, Hibi M, Kallunki T, Karin M, Ben-Neriah Y. JNK is involved in signal
integration during costimulation of T lymphocytes. Cell 77: 727–736, 1994.
279. Su YC, Han J, Xu S, Cobb M, Skolnik EY. NIK is a new Ste20-related kinase that binds
NCK and MEKK1 and activates the SAPK/JNK cascade via a conserved regulatory
domain. EMBO J 16: 1279 –1290, 1997.
280. Su YC, Treisman JE, Skolnik EY. The Drosophila Ste20-related kinase misshapen is
required for embryonic dorsal closure and acts through a JNK MAPK module on an
evolutionalrily conserved signaling pathway. Genes Dev 12: 2371–2380, 1998.
261. Shi CS, Kehrl JH. Activation of stress-activated protein kinase/c-Jun N-terminal kinase,
but not NF-␬B, by the tumor necrosis factor (TNF) receptor 1 through a TNF
receptor-associated factor 2- and germinal center kinase related-dependent pathway.
J Biol Chem 272: 32102–32107, 1997.
281. Takaesu G, Kishida S, Hiyama A, Yamaguchi K, Shibuya H, Irie K, Ninomiya-Tsuji J,
Matsumoto K. TAB2, a novel adaptor protein, mediates activation of TAK1 MAPKKK
by linking TAK1 to TRAF6 in the IL-1 signal transduction pathway. Mol Cell 5: 649 –
650, 2000.
262. Shi CS, Leonardi A, Kyriakis J, Siebenlist U, Kehrl JH. TNF-mediated activation of the
stress-activated protein kinase pathway: TNF receptor-associated factor 2 recruits
and activates germinal center kinase related. J Immunol 163: 3279 –3285, 1999.
282. Takeda K, Shimozono R, Noguchi T, Umeda T, Morimoto Y, Naguro I, Tobiume K,
Saitoh M, Matsuzawa A, Ichijo H. Apoptosis signal-regulating kinase (ASK) 2 functions
as a mitogen-activated protein kinase kinase kinase in a heteromeric complex with
ASK1. J Biol Chem 282: 7522–7531, 2007.
263. Shi CS, Huang NN, Harrison K, Han SB, Kehrl JH. The mitogen-activated protein
kinase kinase kinase kinase GCKR positively regulates canonical and noncanonical Wnt
signaling in B lymphocytes. Mol Cell Biol 26: 6511– 6521, 2006.
264. Shi Y, Massagué J. Mechanisms of TGF-␤ signaling from the membrane to the nucleus.
Cell 113: 685–700, 2003.
265. Shi Y, Kotlyarov A, Laabeta K, Gruber AD, Butt E, Marcus K, Meyer HE, Friedrich A,
Volk HD, Gaestel M. Elimination of protein kinase MK5/PRAK activity by targeted
homologous recombination. Mol Cell Biol 23: 7732–7741, 2003.
266. Shibuya H, Yamaguchi K, Shirakabe K, Tonegawa A, Gotoh Y, Ueno N, Irie K, Nishida
E, Matsumoto K. TAB1: an activator of the TAK1 MAPKKK in TGF-␤ signal transduction. Science 272: 1179 –1182, 1996.
267. Singh RA, Zhang JZ. Differential activation of ERK, p38, and JNK required for Th1 and
Th2 deviation in myelin-reactive T cells induced by altered peptide ligand. J Immunol
173: 7299 –7307, 2004.
268. Singh R, Wang Y, Xiang Y, Tanaka KE, Gaarde WA, Czaja MJ. Differential effects of
JNK1 and JNK2 inhibition on murine steatohepatitis and insulin resistance. Hepatology
49: 87–96, 2009.
269. Sithanandam G, Latif F, Smola U, Bernal RA, Duh FM, Li H, Kuzmin I, Wixle V, Geil L,
Shresta S, Lloyd PA, Bader S, Sekido Y, Tartof KD, Kashuba VI, Zabarovsky ER, Dean
M, Klein G, Zbar B, Lerman MI, Minna JD, Rapp UR, Allikmets A. 3pK, a new mitogenactivated protein kinase-activated protein kinase located in the small cell lung cancer
tumor suppressor gene region. Mol Cell Biol 16: 868 – 876, 1996.
270. Smeal T, Binetruy B, Mercola D, Grover-Bardwick A, Heidecker G, Rapp UR, Karin
M. Oncogenic and transcriptional cooperation with Ha-Ras requires phosphorylation
of c-Jun on serines 63 and 73. Nature 354: 494 – 496, 1991.
271. Solinas G, Vilcu C, Neels JG, Bandyopadhyay GK, Luo JL, Naugler W, Grivennikov S,
Wynshaw-Boris A, Scadeng M, Olefsky JM, Karin M. JNK1 in hematopoietically derived cells contributes to diet-induced inflammation and insulin resistance without
affecting obesity. Cell Metab 6: 386 –397, 2007.
272. Sonenberg N, Gingras AC. The mRNA 5= cap-binding protein eIF-4E and control of
cell growth. Curr Opin Cell Biol 10: 268 –275, 1998.
273. Stafford MJ, Morrice NA, Peggie MW, Cohen P. Interleukin-1 stimulated activation of
the COT catalytic subunit through the phosphorylation of Thr290 and Ser62. FEBS
Lett 580: 4010 – 4014, 2006.
274. Stein B, Yang MX, Young DB, Janknecht R, Hunter T, Murray BW, Barbosa M. p38 –2,
a novel mitogen-activated protein kinase with distinct properties. J Biol Chem 272:
19509 –19517, 1997.
283. Takekawa M, Posas F, Saito H. A human homolog of the yeast Ssk2/Ssk22 MAP kinase
kinase kinases, MTK1, mediates stress-induced activation of the p38 and JNK pathways. EMBO J 16: 4973– 4982, 1997.
284. Takekawa M, Saito H. A family of stress-inducible GADD45-like proteins mediate
activation of the stress-responsive MTK1/MEKK4 MAPKKK. Cell 95: 521–530, 1998.
285. Tan Y, Rouse J, Zhang A, Cariaty S, Cohen P, Comb M. FGF and stress regulate CREB
and ATF1 via a pathway involving p38 MAP kinase and MAPKAP kinase-2. EMBO J 15:
4629 – 4642, 1996.
286. Tang G, Minemoto Y, Dibling B, Purcell NH, Li Z, Karin M, Lin A. Inhibition of JNK
activation through NF-␬B target genes. Nature 414: 313–317, 2001.
287. Tang M, Wei X, Guo Y, Breslin P, Zhang S, Zhang S, Wei W, Xia Z, Diaz M, Akira S,
Zhang J. TAK1 is required for the survival of hematopoietic cells and hepatocytes in
mice. J Exp Med 205: 1611–1619, 2008.
288. Tanoue T, Adachi M, Moriguchi T, Nishida E. A conserved docking motif in MAP
kinases common to substrates, activators and regulators. Nature Cell Biol 2: 110 –116,
2000.
289. Tanoue T, Nishida E. Molecular recognitions in the MAP kinase cascades. Cell Signal
15: 455– 462, 2003.
290. Tapon N, Nagata K, Lamarche N, Hall A. A new Rac target POSH is an SH3-containing
scaffold protein involved in the JNK and NF-␬B signalling pathways. EMBO J 17:
1395–1404, 1998.
291. Tibbles LA, Ing YL, Kiefer F, Chan J, Iscove N, Woodgett JR, Lassam NJ. MLK-3
activates the SAPK/JNK and p38/RK pathways via SEK1 and MKK3/6. EMBO J 15:
7026 –7035, 1996.
292. Tobiume K, Matsuzawa A, Takahashi T, Nishitoh H, Morita K, Takeda K, Minowa O,
Miyazono K, Noda T, Ichijo H. ASK1 is required for sustained activations of JNK/p38
MAP kinases and apoptosis. EMBO Rep 2: 222–228, 2001.
293. Tournier C, Whitmarsh AJ, Cavanagh J, Barrett T, Davis RJ. Mitogen-activated protein
kinase kinase 7 is an activator of the c-Jun NH2-terminal kinase. Proc Natl Acad Sci USA
94: 7337–7342, 1997.
294. Tournier C, Hess P, Yang DD, Xu J, Turner TK, Nimnual A, Bar-Sagi D, Jones SN,
Flavell RA, Davis RJ. Requirement of JNK for stress-induced activation of the cytochrome c-mediated death pathway. Science 288: 870 – 873, 2000.
295. Treisman R. Regulation of transcription by MAP kinase cascades. Curr Opin Cell Biol 8:
205–215, 1996.
275. Stokoe D, Campbell DG, Nakielny S, Hidaka H, Leevers S, Marshall C, Cohen P.
MAPAKP kinase-2: a novel protein kinase activated by mitogen-activated protein
kinase. EMBO J 11: 3985–3994, 1992.
296. Tung RM, Blenis JJ. A novel human SPS1/STE20 homologue, KHS, activates Jun N-terminal kinase. Oncogene 14: 653– 659, 1997.
276. Stokoe D, Engel K, Campbell DG, Cohen P, Gaestel M. Identification of MAPKAP
kinase 2 as a major enzyme responsible for the phosphorylation of the small mammalian heat shock proteins. FEBS Lett 313: 307–313, 1992.
297. Ueda T, Sasaki M, Elia AJ, Chio II, Hamada K, Fukunaga R, Mak TW. Combined
deficiency for MAP kinase interacting kinase 1 and 2 (Mnk1 and Mnk2) delays tumor
development. Proc Natl Acad Sci USA 107: 13984 –13990, 2010.
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
735
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
260. Shaulian E, Karin M. AP-1 as a regulator of cell life and death. Nat Cell Biol 4: E131–
E136, 2002.
JOHN M. KYRIAKIS AND JOSEPH AVRUCH
298. Vallerie SN, Furuhashi M, Fucho R, Hotamisligil G. A predominant role for parenchymal c-Jun amino terminal kinase (JNK) in the regulation of systemic insulin sensitivity.
PLOS One 3: e3151, 2008.
319. West PA, Brodsky IE, Rahner C, Woo DK, Erdjument-Bromage H, Tempst P, Walsh
MC, Choi Y, Shadel GS, Ghosh S. TLR signalling augments macrophage bactericidal
activity through mitochondrial ROS. Nature 471: 476 – 480, 2011.
299. Van Wijk B, Moorman AFM, van den Hoff MJB. Role of bone morphogenetic proteins
in cardiac differentiation. Cardiovasc Res 74: 244 –255, 2007.
320. Weston CR, Davis RJ. The JNK signal transduction pathway. Curr Opin Cell Biol 19,
142–149, 2007.
300. Ventura JJ, Kennedy NJ, Lamb JA, Flavell RA, Davis RJ. c-Jun NH2-terminal kinase is
essential for the regulation of AP-1 by tumor necrosis factor. Mol Cell Biol 23: 2871–
2882, 2003.
321. White MF. IRS proteins and the common path to diabetes. Am J Physiol Endocrinol
Metab 283: E413–E422, 2002.
301. Ventura JJ, Hübner A, Zhang C, Flavell RA, Shokat KM, Davis RJ. Chemical genetic
analysis of the time course of signal transduction by JNK. Mol Cell 21: 701–710, 2006.
302. Venuprasad K, Elly C, Gao M, Salek-Ardakani S, Harada Y, Luo JL, Yang C, Croft M,
Inoue K, Karin M, Liu YC. Convergence of itch-induced ubiquitination with MEKK1JNK signaling in Th2 tolerance and airway inflammation. J Clin Invest 116: 1117–1126,
2006.
304. Wang C, Deng L, Hong M, Akkaraju GR, Inoue J, Chen ZJ. TAK1 is a ubiquitindependent kinase of MKK and IKK. Nature 412: 346 –351, 2001.
305. Wang L, Pan Y, Dai JL. Evidence of MKK4 pro-oncogenic activity in breast and pancreatic tumors. Oncogene 23: 5978 –5985, 2004.
306. Wang S, Zhang F, Chen F, Liu D, Zheng Y, Zhang Y, Dong C, Su B. MEKK3 regulates
IFN-␥ production in T cells through the Rac1/2-dependent MAPK cascades. J Immunol
186: 5791–5800, 2011.
307. Wang X, Ron D. Stress-induced phosphorylation and activation of the transcription
factor CHOP (GADD153) by p38 MAP kinase. Science 272: 1347–1349, 1996.
308. Wang X, Gorospe M, Holbrook NJ gadd45 is not required for activation of c-Jun
N-terminal kinase or p38 during acute stress. J Biol Chem 274: 29599 –29602, 1999.
309. Wang XS, Diener K, Jannuzzi D, Trollinger D, Tan TH, Lichenstein H, Zukowski M,
Yao Z. Molecular cloning and characterization of a novel protein kinase with a catalytic
domain homologous to mitogen-activated protein kinase kinase kinase. J Biol Chem
271: 31607–31611, 1996.
323. Whitmarsh AJ, Cavanagh J, Tournier C, Yasuda J, Davis RJ. A mammalian scaffold
complex that selectively mediates MAP kinase activation. Science 281: 1671–1674,
1998.
324. Whitmarsh AJ, Kuan CY, Kennedy NJ, Kelkar N, Haydar TF, Mordes JP, Appel M,
Rossini AA, Jones SN, Flavell RA, Rakic P, Davis RJ. Requirement of the JIP1 scaffold
protein for stress-induced JNK activation. Genes Dev 15: 2421–2432, 2001.
325. Whitmarsh AJ, Davis RJ. Role of mitogen-activated protein kinase kinase 4 in cancer.
Oncogene 26: 3172–3184, 2007.
326. Wilhelm M, Kukekov NV, Xu Z, Greene LA. Identification of POSH2, a novel homologue of the c-Jun N-terminal kinase scaffold protein POSH. Dev Neurosci 29: 355–
362, 2007.
327. Winzen R, Kracht M, Ritter B, Wilhelm A, Chen CY, Shyu AB, Muller M, Gaestel M,
Resch K, Holtmann H. The p38 MAP kinase pathway signals for cytokine-induced
mRNA stabilization via MAP kinase-activated protein kinase 2 and an AU-rich regiontargeted mechanism. EMBO J 18: 4969 – 4980, 1999.
328. Witowski JA, Johnson GL. Ubiquitylation of MEKK1 inhibits its phosphorylation of
MKK1 and MKK4 and activation of the ERK1/2 and JNK pathways. J Biol Chem 278:
1403–1406, 2003.
329. Wu Z, Wu J, Jacinto E, Karin M. Molecular cloning and characterization of human
JNKK2, a novel Jun NH2-terminal kinase-specific kinase. Mol Cell Biol 17: 7407–7416,
1997.
330. Xia Y, Wu Z, Su B, Murray B, Karin M. JNKK1 organizes a MAP kinase module through
specific and sequential interactions with upstream and downstream components mediated by its amino terminal extension. Genes Dev 12: 3369 –3381, 1998.
310. Wang XS, Diener K, Manthey CL, Wang Sw Rosenzweig B, Bray J, Delaney J, Cole
CN, Chan-Hui PY, Mantlo N, Lichenstein HS, Zukowski M, Yao Z. Molecular cloning
and characterization of a novel p38 mitogen-activated protein kinase. J Biol Chem 272:
23688 –23674, 1997.
331. Xia Y, Wang J, Xu S, Johnson GL, Hunter T, Lu Z. MEKK1 mediates the ubiquitination
and degradation of c-Jun in response to osmotic stress. Mol Cell Biol 27: 510 –517,
2007.
311. Waskiewicz AJ, Flynn A, Proud CG, Cooper JA. Mitogen-activated protein kinases
activate the serine/threonine kinases Mnk1 and Mnk2. EMBO J 16: 1909 –1920, 1997.
332. Xia ZP, Sun L, Chen X, Pineda G, Jiang X, Adhikari A, Zeng W, Chen ZJ. Direct
activation of protein kinases by unanchored polyubiquitin chains. Nature 461: 114 –
119, 2009.
312. Waskiewicz AJ, Johnson JC, Penn B, Mahalingam M, Kimball SR, Cooper JA. Phosphorylation of the cap-binding protein eukaryotic translation initiation factor 4E by
protein kinase Mnk1 in vivo. Mol Cell Biol 19: 1871–1880, 1999.
333. Xing J, Ginty DD, Greenberg ME. Coupling of the RAS-MAPK pathway to gene
activation by RSK2, a growth factor-regulated CREB kinase. Science 273: 959 –963,
1996.
313. Watford WT, Hissong BD, Durant LO, Yamane H, Muul LM, KannoY, Tato CM,
Ramos HL, Berger AE, Meilke L, Pesu M, Solomon B, Frucht DM, Paul WE, Sher A,
Jankovic D, Tsichlis PN, O’Shea JJ. Tpl2 kinase regulates T cell interferon-gamma
production and host resistance to Toxoplasma gondii. J Exp Med 205: 2803–2812,
2008.
334. Xu P, Das M, Reilly J, Davis RJ. JNK regulates FoxO-dependent autophagy in neurons.
Genes Dev 25: 310 –322, 2011.
314. Watford WT, Wang CC, Tsatsanis C, Mielke LA, Eliopoulos AG, Daskalakis C, Charles
N, Odom S, Rivera J, O’Shea J, Tsichlis PN. Ablation of tumor progression locus 2
promotes a type 2 Th cell response in ovalbumin-immunized mice. J Immunol 184:
105–113, 2010.
315. Wei Y, Pattingre S, Sinha S, Bassik M, Levine B. JNK1-mediated phosphorylation of
Bcl-2 regulates starvation-induced autophagy. Mol Cell 30: 678 – 688, 2008.
316. Wei Y, Sinha S, Levine B. Dual role of JNK1-mediated phosphorylation of Bcl-2 in
autophagy and apoptosis regulation. Autophagy 4: 949 –951, 2008.
335. Xu S, Robbins DJ, Christerson LB, English JM, Vanderbilt C, Cobb MH. Cloning of rat
MEK kinase 1 cDNA reveals an endogenous membrane-associated 195-kDa protein
with a large regulatory domain. Proc Natl Acad Sci USA 93: 5291–5295, 1996.
336. Yamaguchi K, Shirakabi K, Shibuya H, Irie K, Oishi I, Ueno N, Taniguchi T, Nishida E,
Matsumoto K. Identification of a member of the MAPKKK family as a potential mediator of TGF-␤ signal transduction. Science 270: 2008 –2011, 1995.
337. Yamaguchi K, Nagai Si Ninomiya-Tsuji J, Nishita M, Tamai K, Irie K, Ueno N, Nishida
E, Shibuya H, Matsumoto K. XIAP, a cellular member of the inhibitor of apoptosis
protein family links the receptors to TAB1-TAK1 in the BMP signaling pathway. EMBO
J 18: 179 –187, 1999.
317. Wellbrock C, Karasarides M, Marais R. The RAF proteins take centre stage. Nat Rev
Mol Cell Biol 5: 875– 885, 2004.
338. Yan M, Dai T, Deak JC, Kyriakis JM, Zon LI, Woodgett JR, Templeton DJ. Activation of
stress-activated protein kinase by MEKK1 phosphorylation of its activator SEK1. Nature 372: 798 – 800, 1994.
318. Wendel HG, Silva RL, Malina A, Mills JR, Zhu H, Ueda T, Watanabe-Fukunaga R,
Fukunaga R, Teruya-Feldstein J, Pelletier J, Lowe SW. Dissecting eIF4E action in
tumorigenesis. Genes Dev 21: 3232–3237, 2007.
339. Yang DD, Kuan CY, Whitmarsh AJ, Roncon M, Zheng TS, Davis RJ, Rakic P, Flavell RA.
Absence of excitotoxicity-induced apoptosis in the hippocampus of mice lacking the
JNK3 gene. Nature 389: 865– 870, 1997.
736
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
303. Wan YY, Chi H, Xie M, Schneider MD, Flavell RA. The kinase TAK1 integrates antigen
and cytokine receptor signaling for T cell development, survival and function. Nat
Immunol 7: 851– 858, 2006.
322. Whitmarsh AJ, Shore P, Sharrocks AD, Davis RJ. Integration of MAP kinase signal
transduction pathways at the serum response element. Science 269: 403– 407, 1995.
MAMMALIAN MAPK SIGNAL TRANSDUCTION PATHWAYS
340. Yang J, Boerm M M, McCarthy M, Bucana C, Fidler IJ, Zhuang Y, Su B. Mekk3 is
essential for early embryonic cardiovascular development. Nat Genet 24: 309 –313,
2000.
346. Yuasa T, Ohno S, Kehrl JH, Kyriakis JM. Tumor necrosis factor signaling to stressactivated protein kinase (SAPK)/Jun NH2-terminal kinase (JNK) and p38. J Biol Chem
273: 22681–22692, 1998.
341. Yang SH, Whitmarsh AJ, Davis RJ, Sharrocks AD. Differential targeting of MAP kinases
to the ETS-domain transcription factor Elk-1. EMBO J 17: 1740 –1749, 1998.
347. Yujiri T, Sather S, Fanger GR, Johnson GL. Role of MEKK1 in cell survival and activation
of JNK and ERK pathways defined by targeted gene disruption. Science 282: 1911–
1914, 1998.
342. Yang Z, Klionsky DJ. Mammalian autophagy: core molecular machinery and signaling
regulation. Curr Opin Cell Biol 22: 124 –131, 2010.
343. Yao Z, Deiner K, Wang XS, Zukowski M, Matsumoto G, Zhou G, Mo R, Sasaki Nishina
H, Hui CC, Tan TH, Woodgett JR, Penninger JM. Activation of stress-activated protein kinases/c-Jun N-terminal kinases (SAPKs/JNKs) by a novel mitogen-activated
protein kinase kinase (MKK7). J Biol Chem 272: 32378 –32383, 1997.
344. Yao Z, Zhou G, Wang XS, Brown A, Diener K, Gan H, Tan TH. A novel human
STE20-related protein kinase, HGK, that specifically activates the c-Jun N-terminal
kinase signaling pathway. J Biol Chem 274: 2118 –2125, 1999.
345. Yasuda J, Whitmarsh AJ, Cavanagh J, Sharma M, Davis RJ. The JIP group of mitogenactivated protein kinase scaffold proteins. Mol Cell Biol 19: 7245–7254, 1999.
348. Zaru R, Ronkina N, Gaestel M, Arthur JS, Watts C. The MAPK-activated kinase Rsk
controls an acute toll-like receptor signaling response in dendritic cells and is activated
through two distinct pathways. Nat Immunol 8: 1227–1235, 2007.
349. Zhao M, New L, Kravchenko VV, Kato Y, Gram H, di Padova F, Olson EN, Ulevitch RJ,
Han J. Regulation of the MEF2 family of transcription factors by p38. Mol Cell Biol 19:
21–30, 1999.
350. Zhu J, Balan V, Bronisz A, Balan K, Sun H, Leicht DT, Luo Z, Qin J, Avruch J,
Tzivion G. Identification of Raf-1 S471 as a novel phosphorylation site critical for
Raf-1 and B-Raf kinase activities and for MEK binding. Mol Biol Cell 16: 4733– 4744,
2005.
Downloaded from http://physrev.physiology.org/ by 10.220.33.2 on June 17, 2017
Physiol Rev • VOL 92 • APRIL 2012 • www.prv.org
737