Content uploaded by William Douglas Figg
Author content
All content in this area was uploaded by William Douglas Figg on Mar 19, 2024
Content may be subject to copyright.
NEURODEGENERATIVE DISEASE
Inhibition of p38α MAPK restores neuronal p38γ MAPK
and ameliorates synaptic degeneration in a mouse
model of DLB/PD
Michiyo Iba
1
†, Changyoun Kim
1
†*, Somin Kwon
1
, Marcell Szabo
1
, Liam Horan-Portelance
1
,
Cody J. Peer
2
, William D. Figg
2
, Xylena Reed
3
, Jinhui Ding
4
, Seung-Jae Lee
5
, Robert A. Rissman
6
,
Mark R. Cookson
3
, Cassia Overk
6
, Wolf Wrasidlo
6
, Eliezer Masliah
1,7
*
Alterations in the p38 mitogen-activated protein kinases (MAPKs) play an important role in the pathogenesis of
dementia with Lewy bodies (DLB) and Parkinson’s disease (PD). Activation of the p38α MAPK isoform and mis-
localization of the p38γ MAPK isoform are associated with neuroinflammation and synaptic degeneration in DLB
and PD. Therefore, we hypothesized that p38α might be associated with neuronal p38γ distribution and syn-
aptic dysfunction in these diseases. To test this hypothesis, we treated in vitro cellular and in vivo mouse models
of DLB/PD with SKF-86002, a compound that attenuates inflammation by inhibiting p38α/β, and then investi-
gated the effects of this compound on p38γ and neurodegenerative pathology. We found that inhibition of p38α
reduced neuroinflammation and ameliorated synaptic, neurodegenerative, and motor behavioral deficits in
transgenic mice overexpressing human α-synuclein. Moreover, treatment with SKF-86002 promoted the redis-
tribution of p38γ to synapses and reduced the accumulation of α-synuclein in mice overexpressing human α-
synuclein. Supporting the potential value of targeting p38 in DLB/PD, we found that SKF-86002 promoted the
redistribution of p38γ in neurons differentiated from iPS cells derived from patients with familial PD (carrying
the A53T α-synuclein mutation) and healthy controls. Treatment with SKF-86002 ameliorated α-synuclein–
induced neurodegeneration in these neurons only when microglia were pretreated with this compound.
However, direct treatment of neurons with SKF-86002 did not affect α-synuclein–induced neurotoxicity, sug-
gesting that SKF-86002 treatment inhibits α-synuclein–induced neurotoxicity mediated by microglia. These
findings provide a mechanistic connection between p38α and p38γ as well as a rationale for targeting this
pathway in DLB/PD.
Copyright © 2023 The
Authors, some
rights reserved;
exclusive licensee
American Association
for the Advancement
of Science. No claim
to original U.S.
Government Works
INTRODUCTION
Lewy body diseases (LBDs) are a set of age-related neurodegenera-
tive disorders characterized by the abnormal accumulation and
spreading of α-synuclein (α-syn) in specific regions of the brain
(1–3). Affecting more than 1 million people in the United States
alone (4), LBDs are the second-most common cause of dementia
after Alzheimer’s disease (AD) and include dementia with Lewy
bodies (DLB), idiopathic Parkinson’s disease (PD), and PD demen-
tia (5,6). The precise mechanisms through which α-syn accumula-
tion and propagation lead to neurodegeneration are currently under
intense investigation and potentially include alterations in synaptic
vesicle function (7–9), lysosomal/autophagy impairment (10–14),
mitochondrial dysfunction (15–17), inflammation (18–22), and sig-
naling alterations (23–28).
Various studies have shown that alterations in p38 mitogen-ac-
tivated protein kinases (MAPKs) might be involved in the process of
neurodegeneration in AD and related dementias (29,30). Of the
four p38 MAPK isoforms, studies in AD have mostly focused on
p38α and p38β because they are widely expressed in the brain and
are closely linked to glial activation and the production of proin-
flammatory cytokines (29,31,32). Although much less is known
about p38γ in the central nervous system (CNS), we have recently
shown that p38γ is abundantly localized at the presynaptic termi-
nals of neurons in healthy control brains and plays a role in neuro-
plasticity (26). Moreover, p38γ is unique from the other isoforms in
its ability to bind to PDZ domains, which are critical for anchoring
receptor proteins at the synapse to the cytoskeleton (33). Many p38γ
substrates are synaptic proteins, including α-1 syntrophin, postsyn-
aptic density protein 95 (PSD-95), and tau, further supporting a
direct synaptic role for p38γ (34–37). In the brains of patients
with DLB, there is a notable shift in p38γ distribution, where
instead of localizing at the synapse, p38γ accumulates in the cell
body and colocalizes with α-syn deposits (22). This mislocalization
may contribute to the synaptic dysfunction in DLB, which is
thought to precede neurodegeneration in dementias (22).
Because of the importance of p38 MAPKs, their inhibitors, par-
ticularly those targeting p38α, have been extensively tested in AD
animal models (38) and subsequently in patients with AD (39–
41). These studies showed that p38α inhibitors (42) such as VX-
745 (neflamapimod) reduced inflammation and β-amyloid (Aβ)–
induced neurotoxicity. Moreover, VX-745 was tested in a phase 2
1
Laboratory of Neurogenetics, Molecular Neuropathology Section, National Insti-
tute on Aging, National Institutes of Health, Bethesda, MD 20892, USA.
2
Clinical
Pharmacology Program, National Cancer Institute,National Institutes of Health, Be-
thesda, MD 20892, USA.
3
Laboratory of Neurogenetics, Cell Biology and Gene Ex-
pression Section, National Institute on Aging, National Institutes of Health,
Bethesda, MD 20892, USA.
4
Laboratory of Neurogenetics, Computational Biology
Group, National Institute on Aging, National Institutes of Health, Bethesda, MD
20892, USA.
5
Department of Biomedical Sciences, Neuroscience Research Institute,
and Department of Medicine, Seoul National University College of Medicine, Seoul
03080, Republic of Korea.
6
Department of Neurosciences, University of California,
San Diego, La Jolla, CA 92093, USA.
7
Division of Neuroscience, National Institute on
Aging, National Institutes of Health, Bethesda, MD 20892, USA.
*Corresponding author. Email: changyoun.kim@nih.gov (C.K.); eliezer.masliah@
nih.gov (E.M.)
†These authors contributed equally to this work.
CORRECTION POSTED 25 MAY 2023
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 1 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
clinical trial in patients with mild to moderate AD (43) and in pa-
tients with DLB (NCT04001517) for safety and clinical outcomes
(24), warranting further investigation. These findings suggest that
the anti-inflammatory actions of p38α/β inhibitors could also ame-
liorate neurodegeneration in DLB. Therefore, in the current study,
we investigated whether p38α-related neuroinflammation is associ-
ated with neuronal mislocalization of p38γ and synaptic degenera-
tion in DLB/PD. To address this, we examined the effects of p38
kinase inhibition with a tool compound (SKF-86002) (44) in in
vitro and in vivo models of DLB/PD. This compound and its
analogs have been shown to inhibit p38α/β and block nitric oxide
production (21,45), tumor necrosis factor–α (TNF-α) and interleu-
kin-1 (IL-1) release (46–48), stress-induced apoptosis (49), chemo-
taxis of immune cells (50–52), synaptic long-term depression (53–
58), and axonal transport deficits in various animal models (55). In
this study, we showed that administration of SKF reduced neuroin-
flammation in a DLB/PD mouse model overexpressing human α-
syn. Furthermore, we validated that treatment with the p38 kinase
inhibitor could block neuroinflammation, reverse the mislocaliza-
tion of p38γ, and attenuate DLB/PD-like phenotypes such as α-
syn accumulation, neuronal loss, and motor deficits in DLB/PD
mice. Therefore, these results suggest the existence of a pathogenic
link between neuroinflammation, p38γ relocation, and synaptic de-
generation and provide a molecular explanation for the potential
beneficial effects of p38α modulators in DLB/PD.
RESULTS
SKF-86002 crosses the blood-brain barrier and decreases
p38 activity in α-syn transgenic mice
To test the hypothesis that attenuating neuroinflammation with a
p38 inhibitor might ameliorate the synaptic neurodegenerative pa-
thology in rodent and cell-based models of DLB/PD by reversing
the mislocalization of p38γ, we used SKF-86002 (hereafter referred
as SKF) as a tool compound. SKF was selected for this study because
it has been shown to block p38α [median inhibitory concentration
(IC
50
): 820 nM] (59) in addition to casein kinase 1 delta (60) and
displayed excellent properties for CNS penetration characterized by
a low molecular weight (297.4 Da), low polar surface area (PSA;
27.96), and logP (partition coefficient, 2.73); other more selective
p38 inhibitors such as BIRB 796, VX-745, p38 MAPK inhibitor
IV, SB 706504, and DBM 1285 dihydrochloride displayed relatively
higher PSA/logP values (table S1).
To validate blood-brain barrier (BBB) penetrance, C57BL6 mice
were given SKF (20 mg/kg) by intraperitoneal injection and eutha-
nized at 0, 0.5, 1, 6, and 24 hours after injection (Fig. 1A). We found
that SKF had excellent systemic absorption after an intraperitoneal
dose, with a plasma half-life of around 2.4 hours (Fig. 1A). Brain
exposure was about 40% of the plasma exposure, indicating excel-
lent brain uptake and the ability of SKF to efficiently cross the BBB.
Full pharmacokinetic results are presented in table S2 [note that sys-
temic volume of distribution (V
d
) and clearance (CL) cannot be cal-
culated in the brain because the drug was not directly delivered
there]. To confirm the effects of SKF in the CNS, wild-type (WT)
mice were treated with various concentrations (0 to 40 mg/kg) of
SKF. Immunoblot analysis showed a dose-dependent decrease of
phosphorylated activating transcription factor 2 (Atf2), a well-
known p38 substrate, in the cytosolic and particulate fractions
(Fig. 1, B and C). Likewise, phosphorylated p38 was decreased in
whole brain lysates (Fig. 1D). We next treated human α-syn–over-
expressing transgenic mice (α-syn tg) and nontransgenic (non-tg)
littermates with SKF (20 mg/kg) 5 days per week for 12 weeks. To
assess target engagement, we first incubated human recombinant
ATF2 with p38-containing non-tg mouse brain homogenate.
ATF2 phosphorylation was determined by immunoblot analysis
and served as a proxy for p38 kinase activity (Fig. 1E). Incubation
of ATF2 with brain homogenate induced phosphorylation of ATF2,
whereas it was decreased by SKF treatment and completely inhibit-
ed in the absence of adenosine triphosphate (ATP) (Fig. 1E). We
then compared ATF2 expression in the brains of non-tg and α-
syn tg mice (Fig. 1F). Total Atf2 was not altered by genotype and
treatment in the mice. However, phosphorylated Atf2/total Atf2
ratio was decreased in α-syn tg mice treated with SKF (P< 0.05;
Fig. 1, G and H). Moreover, phosphorylated p38 was increased in
the cytosolic fraction of the vehicle-treated α-syn tg mice and
reduced by SKF treatment in both fractions (Fig. 1). Consistent
with immunoblot analysis, immunohistochemical analysis illustrat-
ed the reduction of phosphorylated Atf2 immunoreactivity in the
nuclei of neuronal and nonneuronal cells in brains of SKF-admin-
istrated α-syn tg mice (Fig. 1K).
Inhibition of p38 by SKF attenuates cytokine-mediated
neuroinammation in α-syn tg mice
Next, we investigated the anti-inflammatory effects of p38α inhibi-
tion in α-syn tg mice. Immunohistochemical analysis demonstrated
that the percentage of p38α-positive cell area was significantly in-
creased in cells of the neocortex, hippocampus, and striatum of α-
syn tg compared with non-tg mice (P< 0.001 for all three brain
regions), and this increase was reversed by SKF treatment (P<
0.0001 for all three brain regions; Fig. 2, A to D). Analysis using
an antibody against glial fibrillary acidic protein (Gfap), an astro-
glial marker, showed that compared with the non-tg mice, the
vehicle-treated α-syn tg mice showed about a 30% increase in
Gfap immunoreactivity in the neocortex and hippocampus but
not in the striatum (Fig. 2, E to H). Treatment with SKF significantly
reduced astrogliosis in the neocortex and hippocampus of the α-syn
tg mice (P< 0.05; Fig. 2, E to H). Similarly, vehicle-treated α-syn tg
mice showed about a 25% increase in cells positive for the microglial
cell marker Iba-1 in the neocortex and hippocampus (P< 0.05 for
neocortex and P< 0.01 for hippocampus), but not in the striatum,
compared with the non-tg mice (Fig. 2, I to L). Treatment with SKF
reduced microgliosis in both brain regions in the α-syn tg mice (P<
0.05 for neocortex and P< 0.001 for hippocampus; Fig. 2, I to L). To
further ascertain the effects of SKF on inflammation by an indepen-
dent approach, concentrations of proinflammatory cytokines and
chemokines were determined by quantitative polymerase chain re-
action (PCR) (Fig. 2, M to Q). The expressions of Tnf,Il-6,Il-13,
Cxcl1, and Ccl2 were significantly increased in the brains of α-syn
tg mice compared with non-tg mice (P< 0.01, Tnf;P< 0.0001, Il-6;P
< 0.01, Il-13;P< 0.05, Cxcl1; and P< 0.001, Ccl2), and expression
was reduced by SKF treatment, except for Tnf and Cxcl1 (P< 0.01,
Il-6;P< 0.05, Il-13; and P< 0.01, Ccl2). To further confirm these
findings in an in vitro system, we treated non-tg primary microglia
with neuronal conditioned media in the presence or absence of SKF
(fig. S1A). Control LZCM [β-galactosidase (LacZ) over-expressing
conditioned media] and neuron-released α-syn–containing condi-
tioned media (αSCM) were obtained from β-galactosidase or
human α-syn overexpressing differentiated SH-S5Y5 neuronal
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 2 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
Fig. 1. SKF-86002 pharmacokinetics and target engagement in the CNS of mice. (A) Concentration of SKF in the blood and brain of the non-tg mouse was deter-
mined by pharmacokinetic analysis (n= 3 per group). (Bto D) Immunoblot analysis of non-tg mice brain homogenates administrated with indicated amounts of SKF is
shown. Cytosolic (B) and particulate (C) fractionswere probed with phosphor-Atf2 (p-Atf2), total Atf2, and β-actin. Whole brain lysates (D) wereprobed with phosphor-p38
(p-p38), total p38, and β-actin. Phosphor-Atf2 and phosphor-p38 band intensities were determined by densitometric quantification and normalized to total Atf2 and p38,
respectively. Data are means ± SEM. *P< 0.05, **P< 0.01, and ****P< 0.0001 (one-way ANOVA with Tukey’s multiple comparison post hoc test). n= 3 per group. (E)
Immunoblot analysis of p38 kinase assay is shown. Human recombinant ATF2 was incubated with non-tg mouse brain homogenate in the presence and absence of ATP
and SKF. The blot was probedwith phosphor-ATF2 and total ATF2. (Fto J) Immunoblot analysis of non-tg and α-syn tg mice administratedwith either vehicle (Veh) or SKF.
Whole brain lysates (F) and cytosolic/particulate fractions (G) were probed with phosphor-Atf2, total Atf2, phosphor-p38, and β-actin. Phosphor-Atf2 (H) and phosphor-
p38 (I and J) band intensities were determined by densitometric quantification. Phosphor-Atf2 and phosphor-p38 were normalized to total Atf2 and β-actin, respectively.
Data are means ± SEM. *P< 0.05 (two-way ANOVA with Tukey’s multiple comparison post hoc test). n= 4 per group. (K) Representative images from immunohistochemical
analysis for phosphor-Atf2 in the neocortices of injected mice are shown. Scale bar, 10 μm.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 3 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
Fig. 2. The p38 inhibitor SKF recties
elevated p38αand neuroinammation
in α-syn tg mice. Non-tg and α-syn tg
mice were injected with vehicle or SKF (20
mg/kg) five times weekly for 12 weeks. (A)
Representative images from immunohis-
tochemical stainings for p38αin the neo-
cortices, hippocampi, and striata of
injected mice are shown. Scale bars, 250
μm (low magnification) and 25 μm (high
magnification). (Bto D) Percentage of
p38α-positive cell area was analyzed in the
neocortices (Ctx) (B), hippocampi (HP) (C),
and striata (ST) (D) of the mice. Data are
means ± SEM. ***P< 0.001 and ****P<
0.0001 (two-way ANOVA with Tukey’s
multiple comparison post hoc test). n= 6
for non-tg + Veh, n= 7 for non-tg + SKF, n
= 5 for α-syn tg + Veh, and n= 6 for α-syn
tg + SKF. (E) Representative images from
immunohistochemical analysis for Gfap in
the neocortices, hippocampi, and striata
of injected mice are shown. Scale bars, 250
μm (low magnification) and 25 μm (high
magnification). (Fto H) Optical density of
Gfap immunoreactivity was analyzed in
the neocortices (F), hippocampi (G), and
striata (H) of the mice. Data are means ±
SEM. *P< 0.05, **P< 0.01, and ****P<
0.0001 (two-way ANOVA with Tukey’s
multiple comparison post hoc test). n= 6
for non-tg + Veh, n= 7 for non-tg + SKF, n
= 5 for α-syn tg + Veh, and n= 6 for α-syn
tg + SKF. (I) Representative images from
immunohistochemical stainings for Iba-1
in the neocortices, hippocampi, and
striata of injected mice are shown. Scale
bars, 250 μm (low magnification) and 25
μm (high magnification). (Jto L) Number
of Iba-1–positive cells was analyzed in the
neocortices (J), hippocampi (K), and striata
(L) of the mice. Data are means ± SEM. *P<
0.05 and **P< 0.01 (two-way ANOVA with
Tukey’s multiple comparison post hoc
test). n= 6 for non-tg + Veh, n= 7 for non-
tg + SKF, n= 5 for α-syn tg + Veh, and n= 6
for α-syn tg–SKF. (Mto Q) Quantitative
gene expression analysis from the brains
of mice injected with vehicle or SKF. Ex-
pressions of Tnf (M), Il-6 (N), Il-13 (O), Cxcl1
(P), and Ccl2 (Q) were analyzed by quan-
titative PCR. Data are means ± SEM. *P<
0.05, **P< 0.01, ***P< 0.001, and ****P<
0.0001 (two-way ANOVA with Tukey’s
multiple comparison post hoc test). n= 7
for non-tg + Veh, n= 6 for non-tg + SKF, n
= 5 for α-syn tg + Veh, and n= 6 for α-syn tg + SKF.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 4 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
cells, respectively (fig. S1A). The expressions of Il-1β and Il-6 were
increased by αSCM treatment in the microglia (P< 0.0001) and
were inhibited by SKF pretreatment (P< 0.0001; fig. S1, B and C).
Immunofluorescence analysis also showed an increase in Il-6 by
αSCM treatment in Iba-1–positive microglia, which was attenuated
by SKF pretreatment (P< 0.0001; fig. S1, D and E). Next, we com-
pared in vitro effects of SKF with those of other more selective p38
MAPK inhibitors at blocking cytokine expression in mouse primary
microglia (Fig. 3 and table S1). WT primary microglia were pretreat-
ed with p38 MAPK inhibitors for 30 min and exposed to neuron-
released αSCM for 2 hours (Fig. 3A). As shown before, the expres-
sions of Il-1β, Il-6, and Il-10 were significantly increased in micro-
glia exposed to αSCM (P< 0.0001; Fig. 3, B to D). SKF treatment
decreased about 50% of αSCM-induced cytokine gene expressions
in microglia. VX-745, SB 706504, and DBM 1285 were most effi-
cient at decreasing αSCM-induced cytokine expression (P<
0.0001). p38 MAPK inhibitor IV reduced about 50% of αSCM-
induced Il-1β expression (P< 0.0001) but did not affect Il-6 expres-
sion and induced further expression of Il-10 in αSCM-exposed mi-
croglia. Together, these results suggest that administered SKF can
reach the brain at high concentrations, thereby reducing microglial
neuroinflammation via p38α inhibition in α-syn tg mice.
p38αinhibition restored the correct localization of
neuronal p38γand ameliorated α-syn neurohistopathology
in α-syn tg mice
Next, we assessed whether inhibition of p38α affected the localiza-
tion of p38γ in neurons. p38γ is mislocalized in the brains of pa-
tients with DLB/PD and α-syn tg models (26).
Immunohistochemical analysis demonstrated that p38γ was
widely distributed in the neuropil in a punctate-like fashion in
both vehicle- and SKF-treated non-tg mice, but p38γ formed
Lewy body (LB)–like inclusions in the neuronal cell bodies of α-
syn tg mice (Fig. 4A). Neuropil-located p38γ was significantly
reduced in the brains of α-syn tg compared with non-tg mice (P
< 0.001 for all brain regions tested; Fig. 4, B to D), whereas the per-
centage of p38γ-positive cell area was increased (P< 0.0001 for neo-
cortex, P< 0.01 for hippocampus, and P< 0.001 for striatum; Fig. 4,
E to G). Administration of SKF restored a large portion of p38γ to
the neuropil with a concomitant decrease in neuronal perikaryon
p38γ accumulation in the α-syn tg mice (Fig. 4, A to G). In addition,
immunoblot analysis demonstrated the induction of p38γ in both
cytosolic and particulate fractions of the vehicle-treated α-syn tg
mice (P< 0.05), whereas SKF treatment restored p38γ in the partic-
ulate fraction of the α-syn tg mice (P< 0.01; Fig. 4, H to J).
Given that abnormal α-syn accumulation promoted p38γ mis-
localization, and treatment of SKF could restore it, we next explored
to what extent this compound might also attenuate the α-syn neuro-
histopathology. Immunohistochemical analysis revealed that α-syn
was distributed in the neuropils of non-tg mice, but it was accumu-
lated in both the neuropils and neuronal cell bodies of α-syn tg mice
brains (Fig. 5A). Treatment with SKF significantly reduced the ac-
cumulation of α-syn in the neuropil in the neocortex and hippo-
campus but not in the striatum (P< 0.0001 for neocortex and
hippocampus; Fig. 5, B to D). Moreover, SKF treatment significantly
reduced the number of α-syn–positive cells in those three brain
regions of α-syn tg mice compared with vehicle-treated α-syn tg
mice (P< 0.001 for neocortex, P< 0.0001 for hippocampus, and
P< 0.05 for striatum; Fig. 5, E to G). Immunoblot analysis also
showed an increase in monomeric and high–molecular weight olig-
omeric α-syn in the brains of vehicle-treated α-syn tg mice com-
pared with non-tg mice (P< 0.0001; Fig. 5, H to J).
Administration of SKF reduced α-syn monomer by 25%, whereas
oligomeric α-syn was not affected by SKF. Likewise, human α-syn
enzyme-linked immunosorbent (ELISA) assay demonstrated that
about 25% of human α-syn was reduced in the cytosolic fraction
of brain homogenates from α-syn tg mice treated with SKF com-
pared with vehicle-treated mouse brain homogenates (P< 0.01;
Fig. 5K). The particulate fraction was not altered by SKF (Fig. 5L).
To further investigate the effect of p38α inhibition on the rela-
tion of neuronal p38γ localization with α-syn, we performed double
immunolabeling analysis. Consistent with previous studies (22), in
non-tg mice, p38γ colocalized with synaptophysin, a synaptic
vesicle marker (Fig. 5, M and O). In contrast, the colocalization of
p38γ with synaptic vesicles was significantly reduced by around 40%
in the α-syn tg mice (P< 0.05), whereas it was rescued by SKF treat-
ment (P< 0.05; Fig. 5, M and O). In non-tg mice, only a small pro-
portion of p38γ was colocalized with α-syn at the neuropil (Fig. 5, N
and P). However, p38γ and α-syn were strongly colocalized as LB-
like inclusions in the neuronal cell bodies of α-syn tg mice (P<
Fig. 3. The eects of p38 inhibitors in α-syn–induced microglial cytokine gene expression. Primary microglia from WT mice were pretreated with control, SKF, p38
MAPK inhibitor IV (IV), VX-745 (VX), SB 706504 (SB), or DBM 1285 (DBM) for 30 min, followed by challenge with LZCM (control) or αSCM for 2 hours. Microglial cytokine
gene expression was determined by quantitative PCR. (A) Schematic illustration of experiment procedure. (Bto D) Expressions of Il-1β(B), Il-6 (C), and Il-10 (D) were
analyzed by quantitative PCR. Data are means ± SEM. ****P< 0.0001 (two-way ANOVA with Tukey’s multiple comparison post hoc test). n= 6 per group.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 5 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
0.0001; Fig. 5, N and P). Moreover, administration of SKF reduced
about 90% of p38γ and α-syn LB-like inclusions in the neuronal cell
bodies of α-syn tg mice (P< 0.001; Fig. 5, N and P). In conclusion,
these results support that inhibition of neuroinflammation by a
p38α inhibitor restores the mislocalization of p38γ and α-syn
from LB-like inclusions to the synapses in α-syn tg mice.
SKF ameliorates altered expression of genes that modulate
synaptic plasticity in α-syn tg mice
p38γ is primarily localized to nerve terminals and has been known
to play a role in synaptic plasticity (26); a recent study suggested that
the alteration of p38γ might be associated with synaptic degenera-
tion in DLB/PD (22). Given that p38α inhibition restored the local-
ization of p38γ from LB-like inclusions to the synapses and
ameliorated α-syn neuropathology in α-syn tg mice, we next inves-
tigated the expressions of synaptic plasticity–associated proteins
(figs. S2 and S3). Immunohistochemical analysis with an antibody
against synaptophysin showed about 35% loss of immunoreactivity
in the neuropils in the neocortices, hippocampi, and striata of α-syn
tg compared with the non-tg mice (fig. S2, A to D). The reduction
was rescued by treatment with SKF in all three brain regions of α-
syn tg mice (P< 0.05). The optical density of soluble N-ethylmalei-
mide–sensitive factor attachment protein 25 (SNAP25) was also
reduced by 35 to 40% in the neocortices and hippocampi of α-syn
tg compared with non-tg mice, whereas it was rescued by SKF ad-
ministration (P< 0.05; fig. S2, E to H). The optical density of
SNAP25 was not altered by genotype or SKF treatment in the stri-
atum. We next analyzed cysteine string protein (CSP) in the non-tg
and α-syn tg mouse brains (fig. S2, I to L). CSP has been known to
be associated with neurodegeneration and maintaining the confor-
mation of SNAP25, thereby facilitating its entry into the membrane-
fusing SNAP receptor complex (61). Compared with the vehicle-
treated non-tg mice, the SKF-treated non-tg mice showed increased
CSP immunostaining in the neocortex and hippocampus but not in
the striatum. Although the expression of CSP was not affected by
genotype, the SKF-treated α-syn tg mice displayed increased CSP
in the neocortex and hippocampus but not in the striatum. The ex-
pression of synaptobrevin 2, which has been known to be associated
with synaptic function (62), was not affected by genotype or SKF
treatment (fig. S2, M to P). In agreement with the image analysis,
immunoblot analysis for synaptophysin showed a trend toward a
reduction and a decrease in SNAP25 in the vehicle-treated α-syn
tg mice, and the alterations were partially rescued by SKF treatment
(fig. S3, A to C). Immunoblot analysis demonstrated that the expres-
sion of CSP and synaptobrevin 2 was similar across the four groups
(fig. S3, A, D, and E).
The C-terminal region of p38γ has been known to interact with
many PDZ domain–containing synaptic proteins, including α-1-
Fig. 4. SKF rescues the aberrant expression and localization of p38γin brains of α-syn tg mice. Non-tg and α-syn tg mice were injected with vehicle or SKF (20 mg/
kg) five times weekly for 12 weeks. (A) Representative images from immunohistochemical stainings for p38γin the neocortices, hippocampi, and striata of injected mice.
Scale bars, 250 μm (low magnification) and 25 μm (high magnification). (Bto D) Percentage of p38γ-positive neuropil areaanalyzed in the neocortices (B), hippocampi (C),
and striata (D) of injected mice. (Eto G) Percentage of p38γ-positive cell area analyzed in the neocortices (E), hippocampi (F), and striata (G) of injected mice. Data are
means ± SEM. *P< 0.05, **P< 0.01, ***P< 0.001, and ****P< 0.0001 (two-way ANOVA with Tukey’s multiple comparison post hoc test). n= 6 for non-tg + Veh, n= 7 for non-
tg + SKF, n= 5 for α-syn tg + Veh, and n= 6 for α-syn tg + SKF. (Hto J) Immunoblot analysis of injected mice. Cytosolic and particulate fractions were probed with total
p38γand β-actin (H). Cytosolic (I) and particulate (J) total p38γband intensity was determined by densitometric quantification and normalized to β-actin. Data are means
± SEM. *P< 0.05 and **P< 0.01 (two-way ANOVA with Tukey’s multiple comparison post hoc test). n= 4 per group.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 6 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
Fig. 5. The p38 inhibitor SKF recties abnormal accumulation of α-syn and mislocalization of p38 in brains of α-syn tg mice. Non-tg and α-syn tg mice were
injected with vehicle or SKF (20 mg/kg) five times weekly for 12 weeks. (A) Representative images from immunohistochemical stainings for α-syn in the neocortices,
hippocampi, and striata of injected mice are shown. Scale bars, 250 μm (low magnification) and 25 μm (high magnification). (Bto D) Percentage of α-syn–positive neuropil
area analyzed in the neocortices (B), hippocampi (C), and striata (D) of injected mice. (Eto G) Number of α-syn–positive cells analyzed in the neocortices (E), hippocampi
(F), and striata (G) of injected mice. Data are means ± SEM. *P< 0.05, **P< 0.01, ***P< 0.001, and ****P< 0.0001 (two-way ANOVA with Tukey’s multiple comparison post
hoc test). n= 6 for non-tg + Veh, n= 7 for non-tg + SKF, n= 5 for α-syn tg + Veh, and n= 6 for α-syn tg + SKF.(Hto J) Immunoblot analysis of injected mice. (H) Whole brain
lysates were probed with α-syn and β-actin. Band intensities of monomeric (I) and oligomeric (J) α-syn were determined by densitometric quantification and normalized
to β-actin. Data are means ± SEM. **P< 0.01 and ****P< 0.0001 (two-way ANOVA with Tukey’s multiple comparison post hoc test). n= 4 per group. (Kand L) Amounts of
human α-syn in cytosolic (K) and particulate (L) fractions were determined by ELISA. Data are means ± SEM. **P< 0.01 (unpaired two-tailed Student’sttest). n= 6 per
group. (Mand N) Representative images from double immunostaining analysis for p38γwith synaptophysin (M) and p38γwith α-syn (N) in the neocortices of injected
mice. Scale bars, 10 μm (low magnification) and 5 μm (high magnification). (Oand P) Percentages of p38γ/synaptophysin–(O) and p38γ/α-syn (P)–positive cells were
analyzed in the neocortices of the injected mice. Data are means ± SEM. *P< 0.05, ***P< 0.001, and ****P< 0.0001 (two-way ANOVA with Tukey’s multiple comparison
post hoc test). n= 4 per group.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 7 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
syntrophin (33). α-1-syntrophin is a necessary adaptor protein for
proper scaffolding and receptor function at the synapse, particularly
in neuromuscular junctions (63). Immunohistochemical analysis
showed that α-1-syntrophin is distributed in the neuronal cell
body and dendrites (fig. S4A). Consistent with our previous study
(22), the immunoreactivity against α-1-syntrophin in the neuronal
processes was reduced in α-syn tg mice compared with non-tg mice
(fig. S4, A to D). Administration of SKF rescued the reduction of
synaptic α-1-syntrophin in the α-syn tg mice (P< 0.01 for neocor-
tex, P< 0.001 for hippocampus, and P< 0.01 for striatum; fig. S4, A
to D). However, immunoblot analysis demonstrated that the total
amount of α-1-syntrophin was similar across the four groups in
the cytosolic and particulate fractions (fig. S4, E to G).
Our results support the notion that administration of SKF recti-
fies the mislocalization of p38γ from the cytoplasm (where it accu-
mulated with α-syn) to the synapses. These effects were
accompanied by a restoration of the expression of other synaptic
proteins, such as synaptophysin and SNAP25.
Administration of SKF ameliorates neurodegeneration and
behavioral decits in α-syn tg mice
Next, we investigated whether inhibition of p38α affects neurode-
generation and behavioral deficits in α-syn tg mice. The number
of NeuN-positive neurons was decreased by about 40% in the
deeper layers of the neocortices, hippocampal CA3, and striata of
α-syn tg compared with non-tg mice (P< 0.01 for neocortex and
hippocampus; Fig. 6, A to D). SKF treatment significantly amelio-
rated the loss of neuronal cells in the neocortices, hippocampi, and,
to a lesser extent, in the striata of α-syn tg mice (P< 0.05 for neo-
cortex and hippocampus; Fig. 6, A to D). The reduced optical
density of the dopaminergic marker tyrosine hydroxylase (TH) in
the striata of α-syn tg mice was rescued by SKF treatment (Fig. 6,
E and F). The number of TH-positive neurons in the substantia
nigra was not altered by genotype or SKF treatment (Fig. 6, E
and G).
Last, we analyzed whether administration of SKF could rescue
behavioral deficits of α-syn tg mice, such as motor impairment, hy-
peractivity, and cognition (64). The impairments of motor learning
and coordination were determined by rotarod (fig. S5, A to E).
Compared with non-tg mice, α-syn tg mice showed reduced
motor learning and coordination impairments (fig. S5, A to C).
Treatment with SKF showed a trend toward an amelioration of
motor behavior deficits in α-syn tg mice, whereas non-tg mice
were not affected by the treatment. The improvement in rotarod
performance between days 2 and 3 was lost in α-syn tg mice but
could be reestablished by SKF treatment (fig. S5, D and E). In the
open-field test, the vehicle-treated α-syn tg mice showed increased
activity compared with non-tg mice, which was not rescued by SKF
Fig. 6. The p38 inhibitor SKF reduces neurodegeneration in α-syn tg mice. Non-tg and α-syn tg mice were injected with vehicle or SKF (20 mg/kg) five times weekly
for 12 weeks. (A) Representative images from immunohistochemical stainings for NeuN in the neocortices, hippocampi, and striata of injected mice are shown. Scale bars,
250 μm (low magnification) and 25 μm (high magnification). (Bto D) Number of NeuN-positive cells was analyzed in the neocortices (B), hippocampi (C), and striata (D) of
injected mice. (E) Representative images from immunohistochemical stainings for TH in the striata and substantia nigrae of injected mice are shown. Scale bars, 25 μm
(high magnification). (Fand G) Optical density of TH in striatum (F) and number of TH-positive cells in substantia nigra (G) were analyzed in the injected mice. Data are
means ± SEM. *P< 0.05 and **P< 0.01 (two-way ANOVAwith Tukey’s multiple comparison post hoc test). n= 6 for non-tg + Veh, n= 7 for non-tg + SKF, n= 5 for α-syn tg +
Veh, and n= 6 for α-syn tg + SKF.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 8 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
treatment (fig. S5, F to H). In the fear conditioning test, the vehicle-
treated α-syn tg mice showed a reduced reaction after a shock stim-
ulus compared with non-tg mice; administration of SKF showed a
trend toward an improved reaction in α-syn tg mice (fig. S5, I to K).
Furthermore, linear regression analysis revealed that latency to fall
in the rotarod test for α-syn tg mice was associated with the amount
of α-syn accumulation, neuronal loss, and synaptophysin immuno-
reactivity, whereas SKF treatment reduced these correlations (fig.
S5, L to N). These findings support the notion that treatment
with SKF ameliorates neurodegeneration, improves motor learning,
and partially improves cognitive deficits in α-syn tg mice.
SKF ameliorates mislocalization of p38γand α-syn in
patient-derived neurons and reduces microglial
neurotoxicity
To validate the therapeutic effects of p38 inhibition described in α-
syn tg mice, we used forebrain neurons (iNeurons) differentiated
from induced pluripotent stem cells (iPSC) derived from healthy
donors and donors with the α-syn A53T mutation (Fig. 7, A to
H). The cultures used contained forebrain iNeurons and glial cells
(fig. S6, A and B). The cells were incubated with or without SKF for
1 or 18 hours. p38γ and α-syn were diffusely distributed in the neu-
ronal cell bodies and neuritic processes in iNeurons derived from
healthy donors (Fig. 7, A to D). In contrast, consistent with the ob-
servations in mouse models and postmortem brain tissues from pa-
tients with DLB/PD, p38γ and α-syn were concentrated in the
neuronal cell bodies and, to a lesser extent, in the neuritic processes
of A53T iNeurons. Low (1 μM) and high (50 μM) doses of SKF
treatment reversed p38γ accumulation in the cell body and synaptic
α-syn loss in both short-term (1 hour) and long-term (18 hours)
incubations. To further characterize neurohistopathology, we im-
munolabeled iNeurons with synaptophysin and Map2, a presynap-
tic terminal and dendritic marker, respectively (Fig. 7, E to H).
Synaptophysin immunoreactivity was not different at baseline
after a 1-hour incubation but increased after low- and high-dose
treatment with SKF for both control iNeurons and, to a greater
extent, A53T iNeurons. (Fig. 7, E and G). Synaptophysin immuno-
reactivity was reduced in A53T iNeurons compared with control
iNeurons incubated for 18 hours with low and high doses of SKF.
The reduced Map2 immunoreactivity in A53T iNeurons was re-
stored after low- and high-dose treatment with SKF (Fig. 7, F
and H).
Fig. 7. The p38 inhibitor SKF restores p38γand α-syn mislocalization and neurodegeneration. Healthy control and A53T-mutant patient–derived human iPSC
neurons and glia mixed cultures were treated with low (1 μM) or high (50 μM) concentrations of SKF. After 1 or 18 hours of incubation, the cells were analyzed by
double immunostaining analysis. (Aand B) Representative images from double immunostaining analysis of p38γand α-syn in SKF-treated cells after 1 hour (A) and
18 hours (B) of incubations. Arrows indicate neurites. Scale bars, 50 μm (low magnification) and 25 μm (high magnification). (Cand D) Pixel intensity of p38γin the
cell body (C) and percentage of α-syn–positive neurite area (D) were analyzed. ****P< 0.0001 (unpaired two-tailed Student’sttest). n= 8 per group. (Eand F) Repre-
sentative images from double immunostaining of synaptophysin and Map2 in SKF-treated cells after 1 hour (E) and 18 hours (F) of incubations. Scale bars, 50 μm (low
magnification) and 25 μm (high magnification). (Gand H) Percentages of synaptophysin-positive (G) and Map2-positive (H) area are analyzed. *P< 0.05 and **P< 0.01
(unpaired two-tailed Student’sttest). n= 8 per group.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 9 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
To further investigate whether the effects of p38α inhibition at
rescuing the synaptic and neurodegenerative pathology were indi-
rectly mediated by the anti-inflammatory effects on glial cells or
by direct effects on neurons, we performed two sets of in vitro ex-
periments (Fig. 8). We first treated WT mouse primary neurons
with conditioned media obtained from primary microglia (Fig. 8,
A and B). Microglial-conditioned media (MgCM) was obtained
from the cells after pretreatment with or without SKF and exposure
to either control LZCM or αSCM for 1 hour, and then the cells were
incubated with fresh media for an additional 6 hours. To assess neu-
rotoxicity, WT mouse primary neurons were treated with MgCM
for 18 hours. After an incubation, neuronal cell viability was deter-
mined by CyQUANT assay (Fig. 8B). Neuronal viability was de-
creased by about 40% after MgCM-αSCM treatment as compared
with MgCM control treatment (P< 0.01). Neurotoxicity of
MgCM-αSCM was completely rescued by pretreatment of microglia
with SKF (MgCM-SKF-αSCM; P< 0.001). We next treated mouse
primary neurons with MgCM-αSCM in the presence and absence of
SKF (Fig. 8, C and D). Treatment with MgCM-αSCM resulted in
about 50% neuronal loss, but it was not rescued by SKF treatment.
These findings support the notion that the neuroprotective effects
of p38α inhibition might be indirectly mediated by the anti-inflam-
matory effects on microglia and not by targeted (or off-target)
effects of SKF on neurons.
DISCUSSION
The present study showed that the p38α inhibitor SKF reduced neu-
roinflammation, rescued the abnormal colocalization of p38γ and
α-syn in the cytoplasm, and promoted the normal redistribution
of p38γ to the presynaptic terminals in the α-syn tg mouse model
of DLB/PD. This was accompanied by the amelioration of synaptic
and neuronal degeneration, which correlated with improvements in
motor learning and reduced cognitive deficits in the DLB/PD
mouse model. SKF also restored the expression of synaptic proteins
and redistribution of p38γ in A53T iNeurons. α-syn–mediated mi-
croglial neurotoxicity was rescued only when mouse microglia were
pretreated with SKF but not when mouse primary neurons were
treated with SKF. This suggests that SKF acts by inhibiting a neuro-
toxic microglial response rather than by acting directly on neurons.
p38 MAPK signaling is known to be a potent modulator of in-
flammation and apoptosis in neurodegenerative diseases and is
Fig. 8. The p38 inhibitor SKF reduces α-syn–mediated microglial neurotoxicity. (A) Illustrated is the experimental design to obtain MgCM and to test microglial
neurotoxicity. Non-tg mouse primary microglia were pretreated with either control or SKF for 30 min, followed by challenge with LZCM (control) or αSCM. After 1-hour
incubation, cells were incubated with fresh neurobasal media for additional 6 hours, and cultured media were collected. (B) Non-tg mouse primary neurons were treated
with MgCM-control, MgCM-SKF, MgCM-αSCM, or MgCM-SKF-αSCM for 18 hours. The neuronal viability was determined by CyQUANT assay. Data are means ± SEM. **P<
0.01 and ***P< 0.001 (two-way ANOVAwith Tukey’s multiple comparison post hoc test). n= 4 per group. (C) Non-tg mouse primary neurons were incubated with either
MgCM-control or MgCM-αSCM in the presence and absence of SKF. (D) Neuronal viability was determined by cyquant assay.Data are means ± SEM. ***P< 0.001 (two-way
ANOVA with Tukey’s multiple comparison post hoc test). n= 4 per group.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 10 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
considered an important therapeutic target for AD and DLB/PD
(39). Although all four isoforms are known to be expressed in the
brain, the p38γ isoform is highly expressed in the CNS, concentrates
in synapses, and has a key role in neuroplasticity (65). We have pre-
viously shown that both in patients with DLB/PD and in models of
synucleinopathy, there is a notable translocation of p38γ from the
synapses to the neuronal cell bodies, where it colocalizes with the α-
syn aggregates and that this might play a role in the mechanisms of
synaptotoxicity (22,26).
The SKF compound was initially thought to be a dual inhibitor
of cyclooxygenase and 5-lipoxygenase because of its potent inhibi-
tion on cytokine production (47,48,66). It was soon found to be a
p38α inhibitor based on its similar structure and activity to
SB203580, another pyridinyl imidazole that displays competitive
binding at the ATP-binding pocket of p38α (46,67). Much of our
knowledge of SKF binding and activity is based on that of SB203580,
which has similar binding affinity for both phosphorylated and de-
phosphorylated forms of p38 and does not affect phosphorylation
status (46,67–69). Furthermore, both SKF and SB203580 are
thought to be specific for the p38α/β isoforms, probably with no
effects on p38γ (46). As such, we propose that SKF or related p38
inhibitors may restore p38γ localization to the synapse and prevent
synaptic loss and neurodegeneration through both indirect and
direct cellular mechanisms. In the current study, we used SKF as
a tool compound because of its superior ability to cross the BBB
at high concentrations, and it displays a comparable anti-inflamma-
tory profile compared to other more selective p38 MAPK inhibitors
(Fig. 3 and table S1).
Neuroinflammation is a hallmark of neurodegenerative diseases,
where a plethora of stressful stimuli from protein aggregates, infec-
tion, and other cellular debris trigger the dual phosphorylation of
p38 at Thr
180
and Tyr
182
by MAPK kinase kinase (MKK) 3 or
MKK6 (30). One such stimulus might be α-syn aggregates that
propagate from cell to cell, including astroglia and microglia (2,
70–72), which activates the p38α inflammatory cascade in the
cells (71,73,74). Phosphorylated p38α in turn activates MAPK-ac-
tivated protein kinase 2, leading to enhanced productions of nitric
oxide and cytokines such as TNF-α, IL-1β, and IL-6 (75,76). The
mechanisms for the alterations of p38γ in DLB/PD are less well
known; however, the aggregated α-syn might bring p38γ from the
synapses to the cytoplasm, and the resulting inflammation stresses
and further impairs axonal transport and clearance of aggregated
proteins in neurons (fig. S7) (22). In support of this possibility,
we showed that the neuroprotective effects of SKF treatment were
accompanied by decreased astrogliosis and microgliosis and reduc-
tions of Tnf, Il-6, and Il-13 in the murine model of DLB/PD and Il-
1β and Il-6 in vitro. In addition, others have shown that treatment
with SKF or SB203580 diminishes the production of IL-1β, IL-6,
TNF-α, interferon-γ, and inducible nitric oxide synthase in
neurons (77,78) and glia (44,79–84).
Although our results support an indirect effect of SKF on
neurons in the DLB/PD mouse model, inhibition of p38α might
also have a direct neuroprotective effect on neurons in DLB/PD
because p38 inhibitors have been known to modulate neuronal sur-
vival by modulating apoptotic pathways involving cross-talk
between the p38 and p53 pathways (85). For example, SB203580
has been shown to reverse apoptotic indicators in toxin-induced
in vitro models of PD, decreasing the ratio of B cell lymphoma 2
(Bcl-2) to Bcl-2–associated X-protein 3 (Bax3) and the cleavage of
caspase 3 and poly(adenosine diphosphate–ribose) polymerase (44,
86–88). Alternatively, translocation of p38γ and reversal by SKF
could be mediated by cross-talk between p38α and p38γ. In this
regard, it has been shown that autoactivation of p38 can result in
transphosphorylation of Tyr
323
(89,90). Research in the context
of myocardial injury where the characteristic readout of autoactiva-
tion, a reduction in p38 dual phosphorylation when the ATP-
binding site is occupied by a catalytic site inhibitor, is commonly
observed when p38 is activated by cardiac stresses such as ischemia
(91,92). Therefore, enhanced neuronal survival might result from
both direct and indirect effects on inflammation by this class of p38
inhibitors, which, in turn, could potentially improve α-syn trans-
port and clearance, leading to redistribution of p38γ from the cyto-
plasm to more physiological compartments such as the synapses
(fig. S7).
In addition, p38α also has been known to modulate the balance
between apoptotic and autophagic cell death in response to stressful
stimuli (93,94). Therefore, if p38γ mislocalization was preceded by
synaptic loss in DLB/PD, then preventing synaptic loss may be suf-
ficient to indirectly restore p38γ and its substrates to the synapse.
Thus, the restoration of p38γ may prevent further synaptic dysfunc-
tion via proper modulation of key synaptic proteins, including α-1-
syntrophin and PSD-95 that are involved in synaptic plasticity. In
this regard, p38 MAPK is required for synaptic plasticity, namely,
a sustained decrease in synaptic strength termed long-term depres-
sion induced by N-methyl-D-aspartate glutamate receptors and
AMPA glutamate receptors (57,95,96). However, further investiga-
tion is required to determine whether p38γ mislocalization in DLB/
PD is linked to dysregulated receptor anchoring and activity.
Another p38γ substrate associated with synaptic dysfunction is
tau, a microtubule-associated protein (97). Nonphosphorylated tau
promotes microtubule stability, whereas hyperphosphorylation is
associated with neurofibrillary tangle formation in AD (98,99).
Tau has also been strongly implicated in LB dementias given prev-
alent overlapping pathology with α-syn and evidence to suggest that
each promotes the aggregation of the other (100). Tau phosphory-
lation by p38γ was shown to rescue Aβ-induced excitotoxicity in
AD animal models, whereas deletion of p38γ exacerbated it (97).
In primary neurons, on the other hand, blocking p38α activity
with SB203580 prevents Aβ-induced and IL-6–dependent tau hy-
perphosphorylation (78,101). Given that tau can be phosphorylated
by different p38 isoforms at different residues, the effects of SKF on
p38γ-induced tau phosphorylation and subsequent synaptic alter-
ations need to be closely considered in future studies (30).
Last, an improved class of p38α inhibitors, VX-745 (neflamapi-
mod), originally developed for the treatment of rheumatoid arthri-
tis, has been tested in patients with mild to moderate AD (41). In a
preliminary clinical trial, VX-745 showed some improvements in
memory and amyloid protein accumulation (41). However, a sub-
sequent multicenter phase 2, randomized, double-blind, placebo-
controlled trial (161 randomized to either VX-745 or matching
placebo for 24 weeks) did not show an effect on improving episodic
memory function in patients with mild AD, although there were
modest improvements of markers of synaptic dysfunction (102).
VX-745 has now been tested in 91 individuals with DLB in a
phase 2 clinical trial reporting improvements in cognition after 16
weeks (103). On the basis of our results, a possible explanation for
the improvements in cognition and synaptic markers in these clin-
ical trials might be related to effects on p38γ; however, development
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 11 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
of biomarkers for p38γ and more detailed studies will be needed to
investigate this possibility.
Our study has some limitations. We used only one representative
DLB/PD animal model with a relatively small sample size of animals
in this study. Therefore, the findings require further validation with
a larger number of animals and using various established DLB/PD
animal models. In addition, nonmotor symptoms in our DLB/PD
animal model, such as an anxiety-like hyperactivity, were not
rescued by SKF treatment, although the motor deficits were amelio-
rated by SKF. Therefore, combinatorial therapy may be necessary to
improve patients’nonmotor symptoms. Last, SKF has the potential
for multitarget effects; thus, future experiments with brain pene-
trant high-affinity p38 MAPK inhibitors are warranted.
Despite these limitations, our results showed that the anti-in-
flammatory effects of p38 inhibition promote synaptic improve-
ments, including p38γ redistribution to the synapses, providing a
mechanistic connection between p38α (which mediates neuroin-
flammation) and p38γ (which mediates synaptic function) in
DLB/PD pathogenesis. This cross-talk between the anti-inflamma-
tory effects of drugs targeting p38α and ameliorating synaptic pa-
thology via p38γ provides a rationale for targeting this signaling
pathway in DLB/PD.
MATERIALS AND METHODS
Study design
The aim of this study was to identify the mechanistic connection
between p38α and p38γ as therapeutic targets for DLB/PD. In
vivo experiments were performed on mice overexpressing human
WT α-syn and their non-tg littermate controls to evaluate the phar-
macokinetics, target engagement, and therapeutic efficacy of the
p38α inhibitor SKF-86002. Injections, dissections, and sample col-
lection were conducted in a nonblinded fashion, whereas data col-
lection and analysis were blinded. Representative images were
obtained from mouse brains in randomly chosen areas. Primary
mouse microglia and neurons were used for in vitro experiments
to determine the cellular target of SKF-86002. Human iPSC-
derived iNeurons and glial cells were used to evaluate the effect of
SKF-86002 on p38γ cellular localization and neurohistopathology.
Sample size was determined by availability and previous experience
with α-syn tg mice, and no outlier was excluded from the study.
Animal model and treatment
For this study, 2- to 3-month-old (cohort 1) or 5- to 7-month-old
(cohort 2) mice overexpressing human WT α-syn under the Thy-1
promoter [Tg(Thy1-SNCA)61Ema, α-syn tg] (https://informatics.
jax.org/allele/MGI:5435401) and their non-tg littermate controls
were used (n= 24 for cohort 1 and n= 36 for cohort 2). This line
was previously shown to develop extensive age-dependent patho-
logical accumulation of α-syn in the neocortex, hippocampus, and
basal ganglia accompanied by a loss of dopaminergic fibers in the
striatum and significant behavioral and motor deficits (104,105)
mimicking some aspects of DLB/PD pathology. We used cohort 1
for the main analysis, and cohort 2 was used for behavioral analysis.
Both non-tg and α-syn tg mice were treated with SKF-86002 (20
mg/kg; MedChemExpress, Monmouth Junction, NJ), which was
dissolved in 10% dimethyl sulfoxide (DMSO) (non-tg, n= 6; α-
syn tg, n= 5 for cohort 1; non-tg, n= 10; α-syn tg, n= 8 for
cohort 2) or vehicle control 10% DMSO (vehicle) (non-tg, n= 7;
α-syn tg, n= 6 for cohort 1; non-tg, n= 9; α-syn tg, n= 9 for
cohort 2) and administered five times per week for 12 weeks by in-
traperitoneal injection. After 3 months, mice were euthanized by
CO
2
, and brains were collected. The left hemispheres were stored
at −80°C until use for biochemical and RNA analysis. The right
hemispheres were stored in 4% paraformaldehyde (PFA), cut into
40-μm sagittal sections using a vibratome, and stored at −30°C in
cryoprotectant buffer [phosphate-buffered saline (PBS)/ethylene
glycol/glycerol, 4:3:3 ratio] until use for immunohistochemical
analysis. The animal study was reviewed and approved by the Insti-
tutional Animal Care and Use Committee of the National Institutes
of Health (463-LNG-2018, 463-LNG-2021, and 463-LNG-2024).
Human forebrain neuron dierentiation and treatment
iPSC lines SNCA WT (NN0004337, RUDCR, Piscataway, NJ) and
SNCA A53T (NN0004344, RUDCR) were grown on Matrigel
(Corning, Corning, NJ)–coated plates in Essential 8 medium
(Thermo Fisher Scientific, Waltham, MA) until 90% confluency.
Cells were then differentiated as previously described (106). iPS
cells were differentiated in N3 medium [50% Dulbecco’s modified
Eagle’s medium/F12 (Thermo Fisher Scientific)] with 0.5× Gluta-
MAX (Thermo Fisher Scientific), 50% Neurobasal (Thermo
Fisher Scientific) with 1× penicillin-streptomycin (Thermo Fisher
Scientific), 0.5× B-27 minus vitamin A (Thermo Fisher Scientific),
0.5× N2 supplement (Thermo Fisher Scientific), 0.5× minimum es-
sential medium nonessential amino acids (Thermo Fisher Scien-
tific), 0.055 mM 2-mercaptoethanol (Thermo Fisher Scientific),
and insulin (1 μg/ml; Millipore Sigma, Burlington, MA)] plus 1.5
μM dorsomorphin (Tocris Bioscience, Bristol, UK) and 10 μM
SB431542 (Stemgent, Boston, MA). Medium was replaced daily
for 11 days. On day 12, dorsomorphin and SB431542 were
removed, and cells continued to be fed each day with N3 basal
medium. Starting on day 16, N3 was supplemented with 0.05 μM
retinoic acid (Millipore Sigma), and medium was replaced daily
through day 20. Cells were then split 1:2 using Accutase cell disso-
ciation reagent (Thermo Fisher Scientific) and replated onto poly-L-
ornithine (Millipore Sigma)–, fibronectin (Thermo Fisher Scien-
tific)–, and laminin (Millipore Sigma)–coated plates on day 20. Dif-
ferentiating neurons were fed with N4 media [the same as N3 plus
0.05 μM retinoic acid, brain-derived neurotrophic factor (2 ng/ml;
R&D Systems, Minneapolis, MN), and glial cell line–derived neuro-
trophic factor (2 ng/ml; R&D Systems) with Y-27632 dihydrochlor-
ide (Thermo Fisher Scientific)]. Media were changed the following
day to N4 media without Y-27632 dihydrochloride. Neurons were
frozen in Synth-a-Freeze (Thermo Fisher Scientific) on day 24 and
thawed when ready for use. Induced forebrain neurons were thawed
and plated in a freshly coated poly-L-ornithine–, fibronectin-, and
laminin-coated six-well dish in N4 with Y-27632 dihydrochloride.
Media were replaced the following day with N4 without Y-27632
dihydrochloride and thereafter every other day. Neurons were
split using Accutase and plated on Matrigel-coated 12-mm poly-
D-lysine/laminin glass coverslips (Corning) at a density of 225,000
cells per well. Differentiated forebrain neurons were treated with
either PBS or SKF (1 or 50 μM). After short-term (1 hour) or
long-term (18 hours) incubations, the cells were washed with PBS
and fixed with 4% PFA for immunofluorescence microscopy
analysis.
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 12 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
Statistical analysis
Data acquisition and quantification were performed by experiment-
ers blinded to genotype except for the in vitro molecular analyses.
The number of independent experiments in in vitro assays or the
number of human participants and mouse subjects in in vivo exper-
iments is represented by n. All data are presented as group means ±
SEM. Statistical analyses were performed using GraphPad Prism 9
(San Diego, CA). Statistical analyses were conducted using unpaired
ttest for two groups or two-way analysis of variance (ANOVA) fol-
lowed by Tukey’s multiple comparison test for other experiments.
To test correlation between variables, linear regression analysis
and Pearson correlation “r”coefficient were used. No animal or
sample was excluded from the analysis.
Supplementary Materials
This PDF le includes:
Materials and Methods
Figs. S1 to S7
Tables S1 to S4
References (107–111)
Other Supplementary Material for this
manuscript includes the following:
Data file S1
MDAR Reproducibility Checklist
View/request a protocol for this paper from Bio-protocol.
REFERENCES AND NOTES
1. H. A. Lashuel, C. R. Overk, A. Oueslati, E. Masliah, The many faces of α-synuclein: From
structure and toxicity to therapeutic target. Nat. Rev. Neurosci. 14, 38–48 (2013).
2. W. S. Kim, K. Kagedal, G. M. Halliday, Alpha-synuclein biology in Lewy body diseases.
Alzheimers Res. Ther. 6, 73 (2014).
3. T. F. Outeiro, D. J. Koss, D. Erskine, L. Walker, M. Kurzawa-Akanbi, D. Burn, P. Donaghy,
C. Morris, J. P. Taylor, A. Thomas, J. Attems, I. McKeith, Dementia with Lewy bodies: An
update and outlook. Mol. Neurodegener. 14, 5 (2019).
4. N. N. I. o. Aging (2021), vol. 2021, https://www.nia.nih.gov/health/what-lewy-body-
dementia-causes-symptoms-and-treatments.
5. I. G. McKeith, B. F. Boeve, D. W. Dickson, G. Halliday, J. P. Taylor, D. Weintraub, D. Aarsland,
J. Galvin, J. Attems, C. G. Ballard, A. Bayston, T. G. Beach, F. Blanc, N. Bohnen, L. Bonanni,
J. Bras, P. Brundin, D. Burn, A. Chen-Plotkin, J. E. Duda, O. el-Agnaf, H. Feldman,
T. J. Ferman, D. ffytche, H. Fujishiro, D. Galasko, J. G. Goldman, S. N. Gomperts, N. R. Graff-
Radford, L. S. Honig, A. Iranzo, K. Kantarci, D. Kaufer, W. Kukull, V. M. Y. Lee, J. B. Leverenz,
S. Lewis, C. Lippa, A. Lunde, M. Masellis, E. Masliah, P. McLean, B. Mollenhauer,
T. J. Montine, E. Moreno, E. Mori, M. Murray, J. T. O’Brien, S. Orimo, R. B. Postuma,
S. Ramaswamy, O. A. Ross, D. P. Salmon, A. Singleton, A. Taylor, A. Thomas, P. Tiraboschi,
J. B. Toledo, J. Q. Trojanowski, D. Tsuang,Z. Walker, M. Yamada, K. Kosaka, Diagnosis and
management of dementia with Lewy bodies: Fourth consensus report of the DLB Con-
sortium. Neurology 89, 88–100 (2017).
6. N. A. Arnaoutoglou, J. T. O’Brien, B. R. Underwood, Dementia with Lewy bodies - from
scientific knowledge to clinical insights. Nat. Rev. Neurol. 15, 103–112 (2019).
7. M. Nguyen, Y. C. Wong, D. Ysselstein, A. Severino, D. Krainc, Synaptic, mitochondrial, and
lysosomal dysfunction in Parkinson’s disease. Trends Neurosci. 42, 140–149 (2019).
8. D. A. Scott, I. Tabarean, Y. Tang, A. Cartier, E. Masliah, S. Roy, A pathologic cascade leading
to synaptic dysfunction in alpha-synuclein-induced neurodegeneration. J. Neurosci. 30,
8083–8095 (2010).
9. D. Sulzer, R. H. Edwards, The physiological role of α-synuclein and its relationship to
Parkinson’s Disease. J. Neurochem. 150, 475–486 (2019).
10. E.-J. Bae, D.-K. Kim, C. Kim, M. Mante, A. Adame, E. Rockenstein, A. Ulusoy, M. Klinkenberg,
G. R. Jeong, J. R. Bae, C. Lee, H.-J. Lee, B.-D. Lee, D. A. Di Monte, E. Masliah, S.-J. Lee, LRRK2
kinase regulates α-synuclein propagation via RAB35 phosphorylation. Nat. Commun. 9,
3465 (2018).
11. L. Crews, B. Spencer, P. Desplats, C. Patrick, A. Paulino, E. Rockenstein, L. Hansen, A. Adame,
D. Galasko, E. Masliah, Selective molecular alterations in the autophagy pathway in
patients with Lewy body disease and in models of alpha-synucleinopathy. PLOS ONE 5,
e9313 (2010).
12. A. M. Cuervo, L. Stefanis, R. Fredenburg, P. T.Lansbury, D. Sulzer, Impaired degradation of
mutant alpha-synuclein by chaperone-mediated autophagy. Science 305,
1292–1295 (2004).
13. L. A. Cunningham, D. J. Moore, Endosomal sorting pathways in the pathogenesis of
Parkinson’s disease. Prog. Brain Res. 252, 271–306 (2020).
14. G. Minakaki, S. Menges, A. Kittel, E. Emmanouilidou, I. Schaeffner, K. Barkovits,
A. Bergmann, E. Rockenstein, A. Adame, F. Marxreiter, B. Mollenhauer, D. Galasko,
E. I. Buzás, U. Schlötzer-Schrehardt, K. Marcus, W. Xiang, D. C. Lie, K. Vekrellis, E. Masliah,
J. Winkler, J. Klucken, Autophagy inhibition promotes SNCA/alpha-synuclein release and
transfer via extracellular vesicles with a hybrid autophagosome-exosome-likephenotype.
Autophagy 14, 98–119 (2018).
15. J.-H. Park, J. D. Burgess, A. H. Faroqi, N. N. DeMeo, F. C. Fiesel, W. Springer, M. Delenclos,
P. J. McLean, Alpha-synuclein-induced mitochondrial dysfunction is mediated via a
sirtuin 3-dependent pathway. Mol. Neurodegener. 15, 5 (2020).
16. S. Smolders, C. Van Broeckhoven, Genetic perspective on the synergistic connection
between vesicular transport, lysosomal and mitochondrial pathways associated with
Parkinson’s disease pathogenesis. Acta Neuropathol. Commun. 8, 63 (2020).
17. G. M. Compagnoni, A. D. Fonzo, S. Corti, G. P. Comi, N. Bresolin, E. Masliah, The role of
mitochondria in neurodegenerative diseases: The lesson from Alzheimer’s disease and
Parkinson’s disease. Mol. Neurobiol. 57, 2959–2980 (2020).
18. I. C. Bras, M. Xylaki, T. F. Outeiro, Mechanisms of alpha-synuclein toxicity: An update and
outlook. Prog. Brain Res. 252, 91–129 (2020).
19. A. S. Harms, S. A. Ferreira, M. Romero-Ramos, Periphery and brain, innate and adaptive
immunity in Parkinson’s disease. Acta Neuropathol. 141, 527–545 (2021).
20. A. S. Harms, J. H. Kordower, A. Sette, C. S. Lindestam Arlehamn, D. Sulzer, R. H. Mach,
Inflammation in experimental models of α-synucleinopathies. Mov. Disord. 36,
37–49 (2021).
21. G. P. Williams, A. M. Schonhoff, A. Jurkuvenaite, N. J. Gallups, D. G. Standaert, A. S. Harms,
CD4 T cells mediate brain inflammation and neurodegeneration in a mouse model of
Parkinson’s disease. Brain 144, 2047–2059 (2021).
22. M. Iba, C. Kim, M. Sallin, S. Kwon, A. Verma, C. Overk, R. A. Rissman, R. Sen, J. M. Sen,
E. Masliah, Neuroinflammation is associated with infiltration of T cells in Lewy body
disease and α-synuclein transgenic models. J. Neuroinammation 17, 214 (2020).
23. D. J. Dues, D. J. Moore, LRRK2 and protein aggregation in Parkinson’s disease: Insights
from animal models. Front. Neurosci. 14, 719 (2020).
24. U. A. Germann, J. J. Alam, P38αMAPK signaling-A robust therapeutic target for Rab5-
mediated neurodegenerative disease. Int. J. Mol. Sci. 21, 5485 (2020).
25. T. Guttuso Jr., K. L. Andrzejewski, D. G. Lichter, J. K. Andersen, Targeting kinases in Par-
kinson’s disease: A mechanism shared by LRRK2, neurotrophins, exenatide, urate, niloti-
nib and lithium. J. Neurol. Sci. 402, 121–130 (2019).
26. M. Iba, C. Kim, J. Florio, M. Mante, A. Adame, E. Rockenstein, S. Kwon, R. Rissman,
E. Masliah, Role of alterations in protein kinase p38gamma in the pathogenesis of the
synaptic pathology in dementia with Lewy bodies and alpha-synuclein transgenic
models. Front. Neurosci. 14, 286 (2020).
27. S. K. Jha, N. K. Jha, R. Kar, R. K. Ambasta, P. Kumar, p38 MAPK and PI3K/AKT signalling
cascades in Parkinson’s disease. Int. J. Mol. Cell Med. 4, 67–86 (2015).
28. J. Obergasteiger, G. Frapporti, P. P. Pramstaller, A. A. Hicks, M. Volta, A new hypothesis for
Parkinson’s disease pathogenesis: GTPase-p38 MAPK signaling and autophagy as con-
vergence points of etiology and genomics. Mol. Neurodegener. 13, 40 (2018).
29. A. D. Bachstetter, L. J. Van Eldik, The p38 MAP kinase family as regulators of proinflam-
matory cytokine production in degenerative diseases of the CNS. Aging Dis. 1,
199–211 (2010).
30. S. A. Correa, K. L. Eales, The role of p38 MAPK and its substrates in neuronal plasticity and
neurodegenerative disease. J. Signal. Transduct. 2012, 649079 (2012).
31. B. Canovas, A. R. Nebreda, Diversity and versatility of p38 kinase signalling in health and
disease. Nat. Rev. Mol. Cell Biol. 22, 346–366 (2021).
32. C. Falcicchia, F. Tozzi, O. Arancio, D. M. Watterson,N. Origlia, Involvement of p38 MAPK in
synaptic function and dysfunction. Int. J. Mol. Sci. 21, 5624 (2020).
33. A. Escos, A. Risco, D. Alsina-Beauchamp, A. Cuenda, p38γand p38δmitogen activated
protein linases (MAPKs), new stars in the MAPK galaxy. Front. Cell Dev. Biol. 4, 31 (2016).
34. V. Buee-Scherrer, M. Goedert, Phosphorylation of microtubule-associated protein tau by
stress-activated protein kinases in intact cells. FEBS Lett. 515, 151–154 (2002).
35. A. Ittner, P. R. Asih, A. R. P. Tan, E. Prikas, J. Bertz, K. Stefanoska, Y. Lin, A. M. Volkerling,
Y. D. Ke, F.Delerue, L. M. Ittner, Reduction of advanced tau-mediated memory deficits by
the MAP kinase p38γ.Acta Neuropathol. 140, 279–294 (2020).
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 13 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
36. G. Sabio, J. S. C. Arthur, Y. Kuma, M. Peggie, J. Carr, V. Murray-Tait, F. Centeno, M. Goedert,
N. A. Morrice, A. Cuenda, p38γregulates the localisation of SAP97 in the cytoskeleton by
modulating its interaction with GKAP. EMBO J. 24, 1134–1145 (2005).
37. G. Sabio, S. Reuver, C. Feijoo, M. Hasegawa, G. M. Thomas, F. Centeno, S. Kuhlendahl,
S. Leal-Ortiz, M. Goedert, C. Garner, A. Cuenda, Stress- and mitogen-induced phos-
phorylation of the synapse-associated protein SAP90/PSD-95 by activation of SAPK3/
p38gamma and ERK1/ERK2. Biochem. J. 380, 19–30 (2004).
38. S. M. Roy, V. L. Grum-Tokars, J. P. Schavocky, F. Saeed, A. Staniszewski, A. F. Teich,
O. Arancio, A. D. Bachstetter, S. J. Webster, L. J. van Eldik, G. Minasov, W. F. Anderson,
J. C. Pelletier, D. M. Watterson, Targeting human central nervous system protein kinases:
An isoform selective p38αMAPK inhibitor that attenuates disease progression in Alz-
heimer’s disease mouse models. ACS Chem. Nerosci. 6, 666–680 (2015).
39. J. K. Lee, N. J. Kim, Recent advances in the inhibition of p38 MAPK as a potential strategy
for the treatment of Alzheimer’s disease. Molecules 22, 1287 (2017).
40. N. Maphis, S. Jiang, G. Xu, O. N. Kokiko-Cochran, S. M. Roy, L. J. van Eldik, D. M. Watterson,
B. T. Lamb, K. Bhaskar, Selective suppression of the αisoform of p38 MAPK rescues late-
stage tau pathology. Alzheimers Res. Ther. 8, 54 (2016).
41. P. Scheltens, N. Prins, A. Lammertsma, M. Yaqub, A. Gouw, A. M. Wink, H. M. Chu,
B. N. M. van Berckel, J. Alam, An exploratory clinical study of p38αkinase inhibition in
Alzheimer’s disease. Ann. Clin. Transl. Neurol. 5, 464–473 (2018).
42. J. P. Duffy, E. M. Harrington, F. G. Salituro, J. E. Cochran, J. Green, H. Gao, G. W. Bemis,
G. Evindar, V. P. Galullo, P. J. Ford, U. A. Germann, K. P. Wilson, S. F. Bellon, G. Chen,
P. Taslimi, P. Jones, C. Huang, S. Pazhanisamy, Y. M. Wang, M. A. Murcko, M. S. S. Su, The
discovery of VX-745: A novel and selective p38αkinase inhibitor. ACS Med. Chem. Lett. 2,
758–763 (2011).
43. D. Sadowsky, C. Kunzel, ’Usual and customary’practice versus the recommendations of
experts: Clinician noncompliance in the prevention of bacterial endocarditis. J. Am. Dent.
Assoc. 118, 175–180 (1989).
44. M. Gomez-Lazaro, M. F. Galindo, C. G. Concannon, M. F. Segura, F. J. Fernandez-Gomez,
N. Llecha, J. X. Comella, J. H. M. Prehn, J. Jordan, 6-Hydroxydopamine activates the mi-
tochondrial apoptosis pathway through p38 MAPK-mediated, p53-independent activa-
tion of Bax and PUMA. J. Neurochem. 104, 1599–1612 (2008).
45. B. K. Yoo, J. W. Choi, C. Y. Shin, S. J. Jeon, S. J. Park, J. H. Cheong, S. Y. Han, J. R. Ryu,
M. R. Song, K. H. Ko, Activation of p38 MAPK induced peroxynitrite generation in LPS plus
IFN-gamma-stimulated rat primary astrocytes via activation of iNOS and NADPH oxidase.
Neurochem. Int. 52, 1188–1197 (2008).
46. J. C. Lee, S. Kassis, S. Kumar, A. Badger, J. L. Adams, p38 mitogen-activated protein kinase
inhibitors—mechanisms and therapeutic potentials. Pharmacol. Ther. 82,
389–397 (1999).
47. B. J. Votta, D. R. Bertolini, Cytokine suppressive anti-inflammatory compounds inhibit
bone resorption in vitro. Bone 15, 533–538 (1994).
48. D. A. Hartman, S. J. Ochalski, R. P. Carlson, The effects of antiinflammatory and antiallergic
drugs on cytokine release after stimulation of human whole blood by lipopolysaccharide
and zymosan A. Inamm. Res. 44, 269–274 (1995).
49. S. C. Frasch, J. A. Nick, V. A. Fadok, D. L. Bratton, G. S. Worthen, P.M. Henson, p38 mitogen-
activated protein kinase-dependent and -independent intracellular signal transduction
pathways leading to apoptosis in human neutrophils. J. Biol. Chem. 273,
8389–8397 (1998).
50. S. Santen, A. Mihaescu, M. W. Laschke, M. D. Menger, Y. Wang, B. Jeppsson, H. Thorlacius,
p38 MAPK regulates ischemia-reperfusion-induced recruitment of leukocytes in the
colon. Surgery 145, 303–312 (2009).
51. K. Nakamura, V. M. Nemani, F. Azarbal, G. Skibinski, J. M. Levy, K. Egami, L. Munishkina,
J. Zhang, B. Gardner, J. Wakabayashi, H. Sesaki, Y. Cheng, S. Finkbeiner, R. L. Nussbaum,
E. Masliah, R. H. Edwards, Direct membrane association drives mitochondrial fission by
the Parkinson disease-associated protein alpha-synuclein. J. Biol. Chem. 286,
20710–20726 (2011).
52. O. Lindemann, D. Umlauf, S. Frank, S. Schimmelpfennig, J. Bertrand, T. Pap, P. J. Hanley,
A. Fabian, A. Dietrich, A. Schwab, TRPC6 regulates CXCR2-mediated chemotaxis of murine
neutrophils. J. Immunol. 190, 5496–5505 (2013).
53. V. Y. Bolshakov, L. Carboni, M. H. Cobb, S. A. Siegelbaum, F. Belardetti, Dual MAP kinase
pathways mediate opposing forms of long-term plasticity at CA3-CA1 synapses. Nat.
Neurosci. 3, 1107–1112 (2000).
54. S. M. Gallagher, C. A. Daly, M. F. Bear, K. M. Huber, Extracellular signal-regulated protein
kinase activation is required for metabotropic glutamate receptor-dependent long-term
depression in hippocampal area CA1. J. Neurosci. 24, 4859–4864 (2004).
55. K. L. Gibbs, B. Kalmar, E. R. Rhymes, A. D. Fellows, M. Ahmed, P. Whiting, C. H. Davies,
L. Greensmith, G. Schiavo, Inhibiting p38 MAPK alpha rescues axonal retrograde transport
defects in a mouse model of ALS. Cell Death Dis. 9, 596 (2018).
56. C. C. Huang, J. L. You, M. Y. Wu, K. S. Hsu, Rap1-induced p38 mitogen-activated protein
kinase activation facilitates AMPA receptor trafficking via the GDI.Rab5 complex. Potential
role in (S)-3,5-dihydroxyphenylglycene-induced long term depression. J. Biol. Chem. 279,
12286–12292 (2004).
57. P. R. Moult, S. A. Correa, G. L. Collingridge, S. M. Fitzjohn, Z. I. Bashir, Co-activation of p38
mitogen-activated protein kinase and protein tyrosine phosphatase underlies metabo-
tropic glutamate receptor-dependent long-term depression. J. Physiol. 586,
2499–2510 (2008).
58. T. M. Sanderson, E. L. Hogg, G. L. Collingridge, S. A. Correa, Hippocampal metabotropic
glutamate receptor long-term depression in health and disease: Focus on mitogen-acti-
vated protein kinase pathways. J. Neurochem. 139, 200–214 (2016).
59. S. Kluter, C. Grütter, T. Naqvi, M. Rabiller, J. R. Simard, V. Pawar, M. Getlik, D. Rauh, Dis-
placement assay for the detection of stabilizers of inactive kinase conformations. J. Med.
Chem. 53, 357–367 (2010).
60. T. Anastassiadis, S. W. Deacon, K. Devarajan, H. Ma, J. R. Peterson, Comprehensive assay of
kinase catalytic activity reveals featuresof kinase inhibitor selectivity. Nat. Biotechnol. 29,
1039–1045 (2011).
61. R. D. Burgoyne, A. Morgan, Cysteine string protein (CSP) and its role in preventing neu-
rodegeneration. Semin. Cell Dev. Biol. 40, 153–159 (2015).
62. C. Grefen, M. R. Blatt, SNAREs—molecular governors in signalling and development. Curr.
Opin. Plant Biol. 11, 600–609 (2008).
63. M. E. Adams, K. N. Anderson, S. C. Froehner, The α-syntrophin PH and PDZ domains
scaffold acetylcholine receptors, utrophin, and neuronal nitric oxide synthase at the
neuromuscular junction. J. Neurosci. 30, 11004–11010 (2010).
64. S. Roshanbin, A. Aniszewska, A. Gumucio, E. Masliah, A. Erlandsson, J. Bergström,
M. Ingelsson, S. Ekmark-Lewén, Age-related increase of alpha-synuclein oligomers is
associated with motor disturbances in L61 transgenic mice. Neurobiol. Aging 101,
207–220 (2021).
65. P. R. Asih, E. Prikas, K. Stefanoska, A. R. P. Tan, H. I. Ahel, A. Ittner, Functions of p38 MAP
kinases in the central nervous system. Front. Mol. Neurosci. 13, 570586 (2020).
66. S. Kumar, J. Boehm, J. C. Lee, p38 MAP kinases: Key signalling molecules as therapeutic
targets for inflammatory diseases. Nat. Rev. Drug Discov. 2, 717–726 (2003).
67. R. L. Thurmond, S. A. Wadsworth, P. H. Schafer, R. A. Zivin, J. J. Siekierka, Kinetics of small
molecule inhibitor binding to p38 kinase. Eur. J. Biochem. 268, 5747–5754 (2001).
68. D. Casper, M. Bukhtiyarova, E. B. Springman, A Biacore biosensor method for detailed
kinetic binding analysis of small molecule inhibitors of p38alpha mitogen-activated
protein kinase. Anal. Biochem. 325, 126–136 (2004).
69. Z. Wang, B. J. Canagarajah, J. C. Boehm, S. Kassisà, M. H. Cobb, P. R. Young, S. Abdel-
Meguid, J. L. Adams, E. J. Goldsmith, Structural basis of inhibitor selectivity in MAP
kinases. Structure 6, 1117–1128 (1998).
70. H.-J. Lee, J.-E. Suk, C. Patrick, E.-J. Bae, J.-H. Cho, S. Rho, D. Hwang, E. Masliah, S.-J. Lee,
Direct transfer of alpha-synuclein from neuron to astroglia causes inflammatory re-
sponses in synucleinopathies. J. Biol. Chem. 285, 9262–9272 (2010).
71. C. Bussi, J. M. Peralta Ramos, D. S. Arroyo, E. A. Gaviglio, J. I. Gallea, J. M. Wang, M. S. Celej,
P. Iribarren, Autophagy down regulates pro-inflammatory mediators in BV2 microglial
cells and rescues both LPS and alpha-synuclein induced neuronal cell death. Sci. Rep. 7,
43153 (2017).
72. C. Kim, A. Beilina, N. Smith, Y. Li, M. Kim, R. Kumaran, A. Kaganovich, A. Mamais, A. Adame,
M. Iba, S. Kwon, W.-J. Lee, S.-J. Shin, R. A. Rissman, S. You, S.-J. Lee, A. B. Singleton,
M. R. Cookson, E. Masliah, LRRK2 mediates microglial neurotoxicity via NFATc2 in rodent
models of synucleinopathies. Sci. Transl. Med. 12, eaay0399 (2020).
73. A. Klegeris, S. Pelech, B. I. Giasson, J. Maguire, H. Zhang, E. G. McGeer, P. L. McGeer, Alpha-
synuclein activates stress signaling protein kinases in THP-1 cells and microglia. Neurobiol.
Aging 29, 739–752 (2008).
74. E. H. Rannikko, S. S. Weber, P. J. Kahle, Exogenous α-synuclein induces toll-like receptor 4
dependent inflammatory responses in astrocytes. BMC Neurosci. 16, 57 (2015).
75. J. Zhang, B. Shen, A. Lin, Novel strategies for inhibition of the p38 MAPK pathway. Trends
Pharmacol. Sci. 28, 286–295 (2007).
76. R. K. Singh, M. Diwan, S. G. Dastidar, A. K. Najmi, Differential effect of p38 and MK2 kinase
inhibitors on the inflammatory and toxicity biomarkers in vitro. Hum. Exp. Toxicol. 37,
521–531 (2018).
77. C. S. Piao, J. B. Kim, P. L. Han, J. K. Lee, Administration of the p38MAPK inhibitor SB203580
affords brain protection with a wide therapeutic window against focal ischemic insult.
J. Neurosci. Res. 73, 537–544 (2003).
78. R. A. Quintanilla, D. I. Orellana, C. Gonzalez-Billault, R. B. Maccioni, Interleukin-6 induces
Alzheimer-type phosphorylation of tau protein by deregulating the cdk5/p35 pathway.
Exp. Cell Res. 295, 245–257 (2004).
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 14 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
79. R. Bhat, E. P. Crowe, A. Bitto, M. Moh, C. D. Katsetos, F. U. Garcia, F. B. Johnson,
J. Q. Trojanowski, C. Sell, C. Torres, Astrocyte senescence as a component of Alzheimer’s
disease. PLOS ONE 7, e45069 (2012).
80. A. Nazmi, K. Dutta, S. Das, A. Basu, Japanese encephalitis virus-infected macrophages
induce neuronal death. J. Neuroimmune Pharmacol. 6, 420–433 (2011).
81. A. Simi, M. Porsmyr-Palmertz, A. Hjerten, M. Ingelman-Sundberg, N. Tindberg, The neu-
roprotective agents chlomethiazole and SB203580 inhibit IL-1beta signalling but not its
biosynthesis in rat cortical glial cells. J. Neurochem. 83, 727–737 (2002).
82. T. Thomas, M. Timmer, K. Cesnulevicius, E. Hitti, A. Kotlyarov, M. Gaestel, MAPKAP kinase
2-deficiency prevents neurons from cell death by reducing neuroinflammation –rele-
vance in a mouse model of Parkinson’s disease. J. Neurochem. 105, 2039–2052 (2008).
83. J. Waak, S. S. Weber, A. Waldenmaier, K. Görner, M. Alunni-Fabbroni, H. Schell, D. Vogt-
Weisenhorn, T.-T. Pham, V. Reumers, V. Baekelandt, W. Wurst, P. J. Kahle, Regulation of
astrocyte inflammatory responsesby the Parkinson’s disease-associated gene DJ-1. FASEB
J. 23, 2478–2489 (2009).
84. M. Dewil, V. F. dela Cruz, L. Van Den Bosch, W. Robberecht, Inhibition of p38 mitogen
activated protein kinase activation and mutant SOD1(G93A)-induced motor neuron
death. Neurobiol. Dis. 26, 332–341 (2007).
85. S. Cardaci, G. Filomeni, G. Rotilio, M. R. Ciriolo, p38(MAPK)/p53 signalling axis mediates
neuronal apoptosis in response to tetrahydrobiopterin-induced oxidative stress and
glucose uptake inhibition: Implication for neurodegeneration. Biochem. J. 430,
439–451 (2010).
86. S. Ghatan, S. Larner, Y. Kinoshita, M. Hetman, L. Patel, Z. Xia, R. J. Youle, R. S. Morrison, p38
MAP kinase mediates bax translocation in nitric oxide-induced apoptosis in neurons.
J. Cell Biol. 150, 335–348 (2000).
87. C.-R. Wu, C.-W. Tsai, S.-W. Chang, C.-Y. Lin, L.-C. Huang, C.-W. Tsai, Carnosic acid protects
against 6-hydroxydopamine-induced neurotoxicity in in vivo and in vitro model of Par-
kinson’s disease: Involvement of antioxidative enzymes induction. Chem. Biol. Interact.
225, 40–46 (2015).
88. F. Wu, Z. Wang, J.-H. Gu, J.-B. Ge, Z.-Q. Liang, Z.-H. Qin, p38(MAPK)/p53-mediated Bax
induction contributes to neurons degeneration in rotenone-induced cellular and rat
models of Parkinson’s disease. Neurochem. Int. 63, 133–140 (2013).
89. J. M. Salvador, P. R. Mittelstadt, T. Guszczynski, T. D. Copeland, H. Yamaguchi, E. Appella,
A. J. Fornace Jr., J. D. Ashwell, Alternative p38 activation pathway mediated by T cell
receptor-proximal tyrosine kinases. Nat. Immunol. 6, 390–395 (2005).
90. M. S. Marber, J. D. Molkentin, T. Force, Developing small molecules to inhibit kinases
unkind to the heart: p38 MAPK as a case in point. Drug Discov. Today Dis. Mech. 7,
e123–e127 (2010).
91. N. Maulik, T. Yoshida, Y. L. Zu, M. Sato, A. Banerjee, D. K. Das, Ischemic preconditioning
triggers tyrosine kinase signaling: A potential role for MAPKAP kinase 2. Am. J. Physiol.
275, H1857–H1864 (1998).
92. M. S. Marber, B. Rose, Y. Wang, The p38 mitogen-activated protein kinase pathway—A
potential target for intervention in infarction, hypertrophy, and heart failure. J. Mol. Cell.
Cardiol. 51, 485–490 (2011).
93. H. Ichijo, E. Nishida, K. Irie, P. t. Dijke, M. Saitoh, T. Moriguchi, M. Takagi, K. Matsumoto,
K. Miyazono, Y. Gotoh, Induction of apoptosis by ASK1, a mammalian MAPKKK that ac-
tivates SAPK/JNK and p38 signaling pathways. Science 275, 90–94 (1997).
94. X. Sui, N. Kong, L. Ye, W. Han, J. Zhou, Q. Zhang, C. He, H. Pan, p38 and JNK MAPK
pathways control the balance of apoptosis and autophagy in response to chemothera-
peutic agents. Cancer Lett. 344, 174–179 (2014).
95. G. L. Collingridge, S. Peineau, J. G. Howland, Y. T. Wang, Long-term depression in the CNS.
Nat. Rev. Neurosci. 11, 459–473 (2010).
96. G. M. Thomas, R. L. Huganir, MAPK cascade signalling and synaptic plasticity. Nat. Rev.
Neurosci. 5, 173–183 (2004).
97. A. Ittner, S. W. Chua, J. Bertz, A. Volkerling, J. van der Hoven, A. Gladbach, M. Przybyla,
M. Bi, A. van Hummel, C. H. Stevens, S. Ippati, L. S. Suh, A. Macmillan, G. Sutherland,
J. J. Kril, A. P. G. Silva, J. P. Mackay, A. Poljak, F. Delerue, Y. D. Ke, L. M. Ittner, Site-specific
phosphorylation of tau inhibits amyloid-βtoxicity in Alzheimer’s mice. Science 354,
904–908 (2016).
98. L. Buee, T. Bussiere, V. Buee-Scherrer, A. Delacourte, P. R. Hof, Tau protein isoforms,
phosphorylation and role in neurodegenerative disorders. Brain Res. Brain Res. Rev. 33,
95–130 (2000).
99. E. Mandelkow, M. von Bergen, J. Biernat, E.-M. Mandelkow, Structural principles of tau and
the paired helical filaments of Alzheimer’s disease. Brain Pathol. 17, 83–90 (2007).
100. J. L. Robinson, E. B. Lee, S. X. Xie, L. Rennert, E. R. Suh, C. Bredenberg, C. Caswell, V. M. van
Deerlin, N. Yan, A. Yousef, H. I. Hurtig, A. Siderowf, M. Grossman, C. T. McMillan, B. Miller,
J. E. Duda, D. J. Irwin, D. Wolk, L. Elman, L. McCluskey, A. Chen-Plotkin, D. Weintraub,
S. E. Arnold, J. Brettschneider, V. M.-Y. Lee, J. Q. Trojanowski, Neurodegenerative disease
concomitant proteinopathies are prevalent, age-related and APOE4-associated. Brain
141, 2181–2193 (2018).
101. E. Giraldo, A. Lloret, T. Fuchsberger, J. Vina, Aβand tau toxicities in Alzheimer’s are linked
via oxidative stress-induced p38 activation: Protective role of vitamin E. Redox Biol. 2,
873–877 (2014).
102. N. D. Prins, J. E. Harrison, H.-M. Chu, K. Blackburn, J. J. Alam, P. Scheltens; for the REVERSE-
SD Study Investigators, A phase 2 double-blind placebo-controlled 24-week treatment
clinical study of the p38 alpha kinase inhibitor neflamapimodin mild Alzheimer ’s disease.
Alzheimers Res. Ther. 13, 106 (2021).
103. J. J. Alam et al., paper presented at the Clinical Trials on Alzheimer ’s Disease, Digital
conference, 4 to 7 November 2020.
104. M. F. Chesselet, F. Richter, C. Zhu, I. Magen, M. B. Watson, S. R. Subramaniam, A pro-
gressive mouse model of Parkinson’s disease: The Thy1-aSyn ("Line 61") mice. Neuro-
therapeutics 9, 297–314 (2012).
105. E. Rockenstein, M. Mallory, M. Hashimoto, D. Song, C. W. Shults, I. Lang, E. Masliah, Diff-
erential neuropathological alterations in transgenic mice expressing alpha-synuclein
from the platelet-derived growth factor and Thy-1 promoters. J. Neurosci. Res. 68,
568–578 (2002).
106. M. F. Burkhardt, F. J. Martinez, S. Wright, C. Ramos, D. Volfson, M. Mason, J. Garnes, V. Dang,
J. Lievers, U. Shoukat-Mumtaz, R. Martinez, H. Gai, R. Blake, E. Vaisberg, M. Grskovic,
C. Johnson, S. Irion, J. Bright, B. Cooper, L. Nguyen, I. Griswold-Prenner, A. Javaherian, A
cellular model for sporadic ALS using patient-derivedinduced pluripotent stem cells. Mol.
Cell. Neurosci. 56, 355–364 (2013).
107. S. Roshanbin, A. Aniszewska, A. Gumucio, E. Masliah, A. Erlandsson, J. Bergström,
M. Ingelsson, S. Ekmark-Lewén, Age-related increase of alpha-synuclein oligomers is
associated with motor disturbances in L61 transgenic mice. Neurobiol. Aging 101,
207–220 (2021).
108. M. Iba, C. Kim, M. Sallin, S. Kwon, A. Verma, C. Overk, R. A. Rissman, R. Sen, J. Misra Sen,
E. Masliah, Neuroinflammation is associated with infiltration of T cells in Lewy body
disease and α-synuclein transgenic models. J. Neuroinammation 17, 214 (2020).
109. M. Iba, C. Kim, J. Florio, M. Mante, A. Adame, E. Rockenstein, S. Kwon, R. Rissman,
E. Masliah, Role of alterations in protein kinase p38γin the pathogenesis of the synaptic
pathology in dementia with Lewy bodies and α-synuclein transgenic models. Front.
Neurosci. 14, 286 (2020).
110. C. Kim, A. Beilina, N. Smith, Y. Li, M. Kim, R. Kumaran, A. Kaganovich, A. Mamais, A. Adame,
M. Iba, S. Kwon, W.-J. Lee, S.-J. Shin, R. A. Rissman, S. You, S.-J. Lee, A. B. Singleton,
M. R. Cookson, E. Masliah, LRRK2 mediates microglial neurotoxicity via NFATc2 in rodent
models of synucleinopathies. Sci. Transl. Med. 12, eaay0399 (2020).
111. C. Kim, D.-H. Ho, J.-E. Suk, S. You, S. Michael, J. Kang, S. J. Lee, E. Masliah, D. Hwang, H.-
J. Lee, S.-J. Lee, Neuron-released oligomeric α-synuclein is an endogenous agonist of
TLR2 for paracrine activation of microglia. Nat. Commun. 4, 1562 (2013).
Acknowledgments: We thank the National Institute of Mental Health IRP Rodent Behavioral
Core for supporting mouse behavior tests. Funding: This research was supported entirely by
the Intramural Research Program of the National Institutes of Health, National Institute on
Aging grant (1ZIAAG000936) to E.M. Author contributions: The study was designed by M.I.,
C.K., and E.M. In vitro experiments were performed by C.K., S.K., M.S., and S.-J.L. In vivo
experiments were performed by M.I., L.H.-P., J.D., R.A.R., and M.R.C. Pharmacokinetic analysis
was performed by M.I., C.J.P., W.D.F., and W.W. iPSCexperiments were designed and performed
by C.K., X.R., and M.R.C. The manuscript was prepared by M.I., C.K., S.K., M.S., L.H-P., C.O., and E.M.
Competing interests: S.-J.L. is a founder and CEO of Neuramedy Co. Ltd. The other authors
declare that they have no competing interests. Data and materials availability: All data
associated with this study are present in the paper or the Supplementary Materials. The
datasets used and/or analyzed during the current study, including raw values, are presented in
data file S1. The α-syn tg mouse line (Tg(Thy1-SNCA)61Ema) can be obtained from R.A.R.
(University of California, San Diego) through a material transfer agreement.
Submitted 20 April 2022
Resubmitted 05 October 2022
Accepted 21 April 2023
Published 10 May 2023
10.1126/scitranslmed.abq6089
SCIENCE TRANSLATIONAL MEDICINE |RESEARCH ARTICLE
Iba et al.,Sci. Transl. Med. 15, eabq6089 (2023) 10 May 2023 15 of 15
Downloaded from https://www.science.org at National Institutes of Health on March 19, 2024
