2

0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
nature medicine advance online publication 1
Although the molecular events underlying the relationship between
obesity and insulin resistance remain uncertain
14
, numerous studies
have implicated an inflammatory link
57
. Obesity produces a state
of chronic, low-grade inflammation in liver and fat accompanied by
the local secretion of cytokines and chemokines that attenuate insu-
lin action. Knockout or pharmacological inhibition of inflammatory
pathways can disrupt the link between genetic- or diet-induced obes-
ity and insulin resistance, suggesting that local inflammation is a key
step in the generation of cellular resistance to important hormones
that regulate metabolism
711
.
The nuclear factor-B (NF-B) transcriptional program is activated
in obese fat and liver and, to a lesser extent, in muscle and seems to
have an important role in the development of insulin resistance
1215
.
NF-B activation is triggered by phosphorylation of the regulatory
protein IB. Four different IB kinases (IKKs) have been identified:
IKK-, IKK-, IKK- and TBK1. A role for IKK- and IKK- in
NF-B activation has been firmly established, but whether TBK1
and IKK- regulate the pathway remains uncertain
1618
. However,
the genes encoding both of these kinases contain B regulatory sites
in their promoter regions, and activation of the NF-B transcrip-
tional pathway induces their expression
19
, suggesting that they may
act downstream of this transcriptional pathway.
We recently reported that expression of both Tbk1 and Ikbke mRNA
and TBK1 and IKK- protein are increased during a high-fat diet (HFD)
in adipose tissue, and the expression of IKK- is increased in liver
15
.
Moreover, deletion of the Ikbke gene rendered mice partially resistant
to the HFD-dependent development of obesity, insulin resistance,
hepatic steatosis and inflammation, leading us to search for small-
molecule inhibitors of these kinases as a potential therapeutic option
to treat these conditions. We report here the identification of one
such compound, amlexanox, which had been previously developed
for the treatment of asthma, allergic rhinitis and aphthous ulcers but
has an unclear mechanism of action
20,21
. Administration of this selec-
tive TBK1 and IKK- inhibitor to obese mice produces reversible
weight loss and improved insulin sensitivity, reduced inflammation
and attenuated hepatic steatosis without affecting food intake. These
data suggest that IKK- and TBK1 are part of a counterinflammatory
process that sustains energy storage in the context of insulin resist-
ance
22
. Disruption of this process by amlexanox thus increases adap-
tive energy expenditure and restores insulin sensitivity. Because of the
apparent safety of this drug in patients, we propose that it undergo
study for the treatment of obesity, type 2 diabetes and nonalcoholic
fatty liver disease in patients.
RESULTS
Obesity increases IKK- and TBK1 activity through NF-kB
In a previous study
15
we reported that the mRNA levels of both
Ikbke and Tbk1 are elevated in white adipose tissue (WAT) from
1
Life Sciences Institute, University of Michigan, Ann Arbor, Michigan, USA.
2
Department of Internal Medicine, University of Michigan, Ann Arbor, Michigan, USA.
3
Department of Molecular and Integrative Physiology, University of Michigan, Ann Arbor, Michigan, USA.
4
Gene Expression Laboratory, Salk Institute for Biological
Sciences, La Jolla, California, USA.
5
Storr Liver Unit, Westmead Millennium Institute and University of Sydney, Westmead Hospital, Westmead, New South Wales,
Australia.
6
Howard Hughes Medical Institute, Salk Institute for Biological Sciences, La Jolla, California, USA.
7
Department of Medicine, University of California San
Diego, San Diego, California, USA.
8
Present addresses: Muscle Metabolism Discovery Performance Unit, GlaxoSmithKline, Research Triangle Park, North Carolina,
USA (S.-H.C.), Institute of Endocrinology, Diabetes and Metabolism, Rambam Health Care Campus, Haifa, Israel (N.M.W.) and Medical Oncology Division, Washington
University School of Medicine, Saint Louis, Missouri, USA (I.H.).
9
These authors contributed equally to this work. Correspondence should be addressed to
A.R.S. (saltiel@umich.edu).
Received 12 January 2012; accepted 26 December 2012; published online 10 February 2013; doi:10.1038/nm.3082
An inhibitor of the protein kinases TBK1 and IKK-
improves obesity-related metabolic dysfunctions in mice
Shannon M Reilly
1,9
, Shian-Huey Chiang
1,8,9
, Stuart J Decker
1
, Louise Chang
1
, Maeran Uhm
13
,
Martha J Larsen
1
, John R Rubin
1
, Jonathan Mowers
13
, Nicole M White
1,8
, Irit Hochberg
1,8
, Michael Downes
4
,
Ruth T Yu
4
, Christopher Liddle
4,5
, Ronald M Evans
4,6
, Dayoung Oh
7
, Pingping Li
7
, Jerrold M Olefsky
7
&
Alan R Saltiel
13
Emerging evidence suggests that inflammation provides a link between obesity and insulin resistance. The noncanonical IkB
kinases IKK- and TANK-binding kinase 1 (TBK1) are induced in liver and fat by NF-kB activation upon high-fat diet feeding
and in turn initiate a program of counterinflammation that preserves energy storage. Here we report that amlexanox, an approved
small-molecule therapeutic presently used in the clinic to treat aphthous ulcers and asthma, is an inhibitor of these kinases.
Treatment of obese mice with amlexanox elevates energy expenditure through increased thermogenesis, producing weight loss,
improved insulin sensitivity and decreased steatosis. Because of its record of safety in patients, amlexanox may be an interesting
candidate for clinical evaluation in the treatment of obesity and related disorders.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
2 advance online publication nature medicine
HFD-fed mice as compared to normal-chow diet (NCD) controls,
whereas Ikbke mRNA is elevated in liver. Immune complex assays
revealed that both IKK- and TBK1 kinase activity were elevated
in livers from HFD-fed mice (Fig. 1a). In the adipose tissue, we
also found higher kinase activity, even when normalized to the
elevated protein expression. We found this elevated expression
after 67 weeks of HFD feeding, directly correlating with the onset
of inflammatory macrophage infiltration in adipose tissue and liver
(data not shown).
We modeled inflammation in 3T3-L1 cells by treating differentiated
adipocytes with the inflammatory cytokine tumor necrosis factor
(TNF-)
23
. Chronic TNF- treatment increased the mRNA levels of
both kinases (Fig. 1b). These effects were prevented by pretreatment
of cells with the IKK- inhibitor compound VIII
24
. Furthermore,
TNF- produced a dose-dependent increase in IKK- protein expres-
sion in these cells, as well as downregulation of peroxisome prolifera-
tor activated receptor- (PPAR-). TNF- also produced increased
phosphorylation of TBK1 at Ser172, which is associated with activa-
tion (Fig. 1c).
To investigate whether inflammatory signals are responsible for the
induction of these kinases in vivo, we treated HFD-fed mice with the
PPAR- agonist rosiglitazone or the -3 fatty acid GPR120 agonists
eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Both
rosiglitazone and -3 fatty acids reduce the inflammation and insulin
resistance associated with obesity
25
. Diet supplemented with -3 fatty
acids markedly attenuated the HFD-induced elevation in the expres-
sion of IKK- mRNA and protein expression in both WAT and liver
(Fig. 1d,e). Treatment of obese mice with rosiglitazone also reduced
expression of Ikbke in adipocytes as well as in the stromal vascular
fraction of WAT (Fig. 1f). Because neither rosiglitazone nor -3 fatty
acid diet supplementation caused weight loss
25
, the elevated expres-
sion of these kinases from high-fat feeding is probably the result of
inflammation rather than a direct result of obesity.
Amlexanox is an inhibitor of IKK- and TBK1
Because deletion of the Ikbke gene in mice produced partial resistance
to the effects of a HFD
15
, we screened a library of 150,000 chemi-
cal compounds to search for inhibitors of IKK- using recombinant
baculoviral-expressed, purified full-length IKK- with myelin
basic protein (MBP) as a substrate and identified amlexanox as a
high-affinity inhibitor (Fig. 2a). This previously discovered drug of
unknown mechanism is used to treat asthma and allergic rhinitis in
Japan and aphthous ulcers in the United States
20,21
.
Dose-response experiments revealed that amlexanox blocked
IKK- activity with a half-maximal inhibitory concentration (IC
50
)
of approximately 12 M (Fig. 2b). Amlexanox also blocked the activ-
ity of TBK1 at approximately the same concentrations but had no
effect on IKK- or IKK- and at these concentrations did not block
any other kinases from a broad panel representing most kinase fami-
lies. TBK1 and IKK- share an overall 65% sequence similarity and
are 72% identical in the ATP-binding region. Inhibition of IKK- or
TBK1 (data not shown) by amlexanox was competitive for its substrate
ATP (Fig. 2c), indicating that it interacts with the enzymes in the
ATP-binding site. This is consistent with a model of the compound
docked in the presumed ATP-binding pocket of TBK1 (Fig. 2d) on
the basis of its published structure
26
.
Amlexanox increased the phosphorylation of TBK1 at Ser172
in 3T3-L1 adipocytes and blocked polyinosinic:polycytidylic acid
(poly I:C)-stimulated phosphorylation of interferon responsive
factor-3 (IRF3), a presumed substrate of IKK- and TBK1 (ref. 27
and Fig. 2e). Reduced IRF3 phosphorylation was also produced
by the previously identified IKK- and TBK1 inhibitor Cay10576
(cayman)
28
. Furthermore, cayman treatment of RAW264.7 macro-
phages stimulated with lipopolysaccharide (LPS) or poly I:C also
blocked the phosphorylation of IRF3 and stimulated the phosphory-
lation of TBK1 at Ser172 (Fig. 2f). The phosphorylation of TBK1 was
also increased in peritoneal macrophages derived from Ikbke knock-
out mice (data not shown), which was probably caused by blockade of
feedback inhibition of the pathway
29
, as evidenced by higher IKK-
phosphorylation in cayman-treated RAW264.7 cells.
Amlexanox produces reversible weight loss in obese mice
We fed C57BL/6 mice a diet of 45% fat and gavaged them daily with
a vehicle control or with 25 or 100 mg per kg body weight amlexanox
and monitored their body weight (Fig. 3a). Serum concentrations
of the drug exceed 5 M at these doses
30
. Treatment of mice with
c
TNF-
(ng ml
1
)
0 1 10 100
pTBK1
TBK1
IKK-
PPAR-
Akt
0
1
2
3
4
Liver WAT
*
#
I
K
K
-

a
c
t
i
v
i
t
y

(
A
U
)
0
1
2
3
Liver WAT
T
B
K
1

a
c
t
i
v
i
t
y

(
A
U
)
a
*
*
NCD
HFD
0
1
2
3
4
5
Ikbke Tbk1
R
e
l
a
t
i
v
e

g
e
n
e

e
x
p
r
e
s
s
i
o
n

No treatment
TNF-
TNF- + compound VIII
Compound VIII
b
*

0
1
2
3
4
5
Liver WAT
R
e
l
a
t
i
v
e

g
e
n
e

e
x
p
r
e
s
s
i
o
n
NCD
HFD
HFD + -3 FA
d

*
f
0
0.05
0.10
0.15
0.20
0.25
SVF Adipocytes
R
e
l
a
t
i
v
e

g
e
n
e
e
x
p
r
e
s
s
i
o
n
HFD
HFD + rosiglitazone
*
*
e
-3 FA
pTBK1
TBK1
IKK-
RalA
Figure 1 Induction of IKK- and TBK1 in obese mice is a result of increased inflammation. (a) TBK1 and IKK-
activity in the liver and WAT of NCD-fed or HFD-fed mice. n = 4 per group. *P < 0.05 for NCD compared to HFD,
#
P < 0.1 for NCD compared to HFD (Students t test). AU, arbitrary units. (b) Expression of Ikbke and Tbk1 in
differentiated 3T3-L1 cells without treatment or treated with TNF-, compound VIII or both TNF- and compound VIII. Results are representative of
multiple experiments. *P < 0.05 for no treatment compared to treatment with TNF- only,

P < 0.05 for treatment with TNF- compared to TNF-

plus compound VIII (analysis of variance (ANOVA) with Tukey-Kramer post hoc analysis). (c) IKK-, TBK1 and PPAR- protein expression and TBK1
phosphorylation at Ser172 (pTBK1) after TNF- treatment. Results were replicated in more than three experiments. Akt was used as a loading control.
(d) Ikbke expression in liver and WAT of NCD-fed or HFD-fed mice with or without -3 fatty acids (FA).

P < 0.05 for NCD compared to HFD,

*P < 0.05 for HFD compared to HFD plus -3 FA. n = 9 per group (ANOVA with Tukey-Kramer post hoc analysis). (e) IKK- protein expression and
TBK1 phosphorylation at Ser172 in HFD-fed mice with and without -3 FA. RalA was used as a loading control. (f) Ikbke expression in adipocytes and
stromal vascular fraction (SVF) of mice fed HFD or HFD plus rosiglitazone. *P < 0.05 for HFD compared to HFD plus rosiglitazone (Students t test).
n = 8 in the HFD group, n = 4 in the HFD plus rosiglitazone group. All data in the figure are shown as the mean s.e.m.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
nature medicine advance online publication 3
either dose of amlexanox prevented the weight gain produced by a
HFD; drug-treated mice maintained weights equivalent to those of
vehicle-treated mice on a control diet through 12 weeks. There was
no effect of the drug on food intake, either at the beginning or end
of the study (Fig. 3b).
We also evaluated amlexanox in mice with established obesity
(Fig. 3c and Supplementary Fig. 1). Treatment with amlexanox pro-
duced a 10-g weight loss in HFD-fed obese mice after only 4 weeks of
treatment with no further effect thereafter. Moreover, escalation of
the dose at this time point had no further effect on weight (data not
shown). Withdrawal of the drug after 8 weeks of treatment restored
weight gain, and the mice returned to their original weights after
68 weeks (Fig. 3c). Weight loss during the treatment phase was
accompanied by more than a 6-g reduction in overall mass of adi-
pose tissue, as well as an 80% decrease in fasting leptin concentration,
whereas the concentrations of serum triglycerides, free fatty acids and
cholesterol were unchanged (Fig. 3d and data not shown).
To determine whether the effects of amlexanox could be seen in
another model of obesity, we treated ob/ob mice
31,32
with 100 mg
per kg body weight of amlexanox or vehicle control and monitored
their body weight (Fig. 3e). Although the drug had no effect on food
intake, amlexanox produced a 7- to 8-g weight loss after 4 weeks of
treatment. Amlexanox also caused a significant decrease in adipose
tissue mass in these mice and an increase in circulating adiponectin
concentrations (Fig. 3f,g).
Because amlexanox had no effect on food intake, we reasoned that
the weight loss was the result of increased energy expenditure. We thus
used metabolic cages to monitor energy expenditure in diet-induced
obese (DIO) mice treated with or without amlexanox. Four-week
treatment of DIO mice with 25 mg per kg body weight amlexanox
resulted in significantly (P < 0.001) higher oxygen consumption as
compared to treatment with vehicle control, which is consistent with
an increase in energy expenditure (Fig. 3h). Exhaled carbon dioxide
was also significantly (P = 0.001) higher in the amlexanox-treated
mice, and their respiratory exchange ratio remained unchanged
compared to control-treated mice (Supplementary Fig. 2), although
amlexanox had no effect on NCD-fed mice, in which neither IKK-
nor TBK1 was induced. Together, these data suggest that amlexanox
might induce an increase in thermogenesis in obese mice. Amlexanox
treatment produced an approximate 1 C increase in body temp-
erature compared to HFD-fed vehicle-treated mice, restoring the
temperatures to those of NCD-fed mice (Fig. 3i).
Amlexanox improves insulin sensitivity in obese mice
To assess in more detail the metabolic impact of amlexanox treatment,
we treated DIO mice with 25 mg per kg body weight amlexanox or
vehicle control for 8 weeks either before or after obesity was estab-
lished and then assessed their metabolic parameters. Mice treated
with amlexanox concurrently with a HFD had improved glucose tol-
erance after obesity was established, with an approximate 3040%
reduction in the area under the curve for glucose (Fig. 4a). Treatment
of mice with amlexanox after established DIO reversed the elevations
in fasting serum insulin concentrations caused by the HFD, suggest-
ing improved insulin sensitivity (Fig. 4b). Insulin tolerance tests
revealed that amlexanox treatment significantly improved insulin
sensitivity in mice with established DIO, as indicated by a restoration
0
0.4
0.6
0.8
1.0
1.2
1.4
0.2 0.1 0 0.1 0.2 0.3 0.4
A
c
t
i
v
i
t
y

1

(
p
M
B
P

1
)
ATP
1
(M
1
)
Vehicle
Amlexanox (1 M)
Amlexanox (3 M)
Amlexanox (10 M)
a e
d
b
c
IKK-
TBK1
[
3
2
P
]
-
M
B
P
f
pTBK1
TBK1
pIRF3
IRF3
pIKK-
IKK-
3.23
3.08
2.88
LPS Poly I:C
IKK-
RalA
IKK-
pTBK1
TBK1
pIRF3
IRF3
IKK-
RalA
Poly I:C
Amlexanox (M)
Amlexanox (M)
O
O
O
OH
N NH
2
Cayman (M)
Cayman (M) 10 2 0 10 2 0 10 2 0

0.5 1.0 2.5 5.0 12.5 25 50


12
1 2 4 1 2 4
25 50 12 25 50
Thr156
Lys38
0.2
0
Figure 2 Amlexanox is a specific inhibitor of IKK-
and TBK1. (a) Diagram of the chemical structure
of amlexanox. (b) Dose-response relationship of
amlexanox inhibition of IKK- and TBK1 activity as
determined by MBP phosphorylation showing an IC
50

of approximately 12 M. Results were replicated in
more than three experiments. (c) Lineweaver-Burk
plot showing competition of amlexanox with ATP for
inhibition of IKK-. Results are representative of three
experiments. (d) Model of amlexanox binding to the ATP-binding site of TBK1. Top, surface model showing the binding of amlexanox in the active
site of TBK1; bottom, hydrogen bonding of amlexanox (cyan) in the active site of TBK1. (e) TBK1 phosphorylation (pTBK1) at Ser172 and IRF3
phosphorylation at Ser396 (pIRF3) in 3T3-L1 adipocytes treated with amlexanox or cayman and with or without poly I:C. Results were replicated in
more than three experiments. RalA was used as a loading control. (f) IRF3, IKK- (pIKK-) and TBK1 phosphorylation in RAW264.7 cells treated with
or without cayman, LPS and poly I:C. Results were replicated in multiple experiments. RalA was used as a loading control.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
4 advance online publication nature medicine
in insulin responsiveness to the levels of NCD-fed mice (Fig. 4c and
Supplementary Fig. 3a). Amlexanox did not affect insulin sensitivity
in NCD-fed mice. After 4 weeks of treatment, amlexanox produced
marked improvements in glucose tolerance (Fig. 4d) and insulin sen-
sitivity (Fig. 4e and Supplementary Fig. 3b) in ob/ob mice.
We used hyperinsulinemic-euglycemic clamp studies to investi-
gate the tissue-specific effects of amlexanox on insulin sensitivity.
Consistent with improved overall insulin sensitivity, the glucose infu-
sion rate was significantly higher in amlexanox-treated mice than in
vehicle-treated mice (Fig. 4f and Supplementary Fig. 3c). Insulin-
mediated suppression of hepatic glucose production was substantially
enhanced by amlexanox treatment (Fig. 4g). However, we found no
change in glucose turnover rate (Fig. 4h). Amlexanox treatment had
no effect on glucose handling in NCD-fed mice (Supplementary
Fig. 3dg). These results indicate that the liver is an important target
of amlexanox in mediating improved insulin sensitivity. Furthermore,
these results support the action of amlexanox through IKK- and
TBK1, both of which are induced in the liver by HFD. Amlexanox-
treated mice showed enhanced insulin-mediated suppression of
serum free fatty acids, which is indicative of increased suppression of
adipose tissue lipolysis (Fig. 4i). Moreover, although the basal phos-
phorylation of hormone-sensitive lipase (HSL) was higher in adipose
tissue from amlexanox-treated mice than vehicle-treated mice, the
dephosphorylation of HSL produced by insulin was enhanced after
amlexanox treatment compared to vehicle-treated mice on HFD
(Supplementary Fig. 3h). Phosphorylation of Akt in response to
insulin was also enhanced in amlexanox-treated mice compared to
vehicle-treated controls, indicating improved insulin sensitivity in
WAT (Supplementary Fig. 3i).
Amlexanox reverses hepatic steatosis in obese mice
We treated DIO mice with 25 mg per kg body weight amlexanox or
vehicle control for 8 weeks and examined their livers postmortem
(Fig. 5a). The hepatomegaly normally observed in HFD-fed mice
h
2000
2500
3000
3500
4000
4500
6 p.m. 6 a.m. 6 p.m. 6 a.m. 6 p.m. 6 a.m.
V
O
2

(
m
l

k
g

1

h

1
)

NCD vehicle NCD amlexanox
(25 mg per kg)
30
32
34
36
38
HFD amlexanox
(25 mg per kg)
HFD vehicle
NCD vehicle
C
o
r
e

t
e
m
p
e
r
a
t
u
r
e

(

C
)
i
2000
2500
3000
3500
4000
4500
6 p.m. 6 a.m. 6 p.m. 6 a.m. 6 p.m. 6 a.m.
V
O
2

(
m
l

k
g

1

h

1
)
HFD vehicle HFD amlexanox
(25 mg per kg)
0
20
40
60
80
Fat Lean Fluid
P
e
r
c
e
n
t
a
g
e

b
o
d
y

w
e
i
g
h
t
0
10
20
30
40
50
Body Fat Lean Fluid
W
e
i
g
h
t

(
g
)
d

*
NCD vehicle HFD vehicle
HFD amlexanox (25 mg per kg) NCD amlexanox (25 mg per kg)
0
10
20
30
40
B
o
d
y

f
a
t

(
g
)
Amlexanox
(100 mg per kg)
Vehicle
f
*
0
10
20
30
40
A
d
i
p
o
n
e
c
t
i
n

(

g

m
l

1
)
g
*
Amlexanox
(100 mg per kg)
Vehicle
30
35
40
45
50
55
60
0 2 4 6 8 10 12
B
o
d
y

w
e
i
g
h
t

(
g
)
Duration of daily gavage (weeks)
Vehicle
Amlexanox (100 mg per kg)
e
HFD amlexanox (25 mg per kg)
HFD vehicle
b
F
o
o
d

i
n
t
a
k
e

(
m
g

p
e
r

g
)
0
50
100
150
200
NCD vehicle
30
35
40
45
50
55
0 2 4 6 8 10 12 14 16
B
o
d
y

w
e
i
g
h
t

(
g
)
Duration of daily gavage (weeks)
HFD vehicle
HFD withdrawal
HFD amlexanox
c
HFD amlexanox
(100 mg per kg)
HFD vehicle
0
20
40
60
80
100
F
o
o
d

i
n
t
a
k
e

(
m
g

p
e
r

g
)
a
20
25
30
35
40
45
50
55
0 5 10 15 20 25
B
o
d
y

w
e
i
g
h
t

(
g
)
Duration of daily gavage (weeks)
NCD vehicle
HFD vehicle
HFD amlexanox
(25 mg per kg)
HFD amlexanox
(100 mg per kg)
Figure 3 Daily amlexanox
gavage both prevents
and reverses diet-
induced or genetic
obesity. (a) Body weight
of mice treated
preventatively with
25 mg per kg body weight
amlexanox, 100 mg per kg
amlexanox or vehicle control.
n = 5 per group. (b) Food intake after 1 week of treatment (left) or 6 weeks of preventative gavage (right) with vehicle in NCD-fed mice and HFD-mice or
with amlexanox in HFD-fed mice. n = 7 per group. (c) Body weight of mice treated with amlexanox or vehicle control and mice in which treatment was
withdrawn after eight weeks. n = 7 per group. (d) Total (left) and relative (right) fat and lean body mass of mice in each treatment group. n = 4 for the
NCD groups, n = 8 for the HFD groups. (e) Body weight of ob/ob mice gavaged with amlexanox or vehicle control. n = 8 per group. (f) Total body fat mass
of ob/ob mice gavaged with amlexanox or vehicle control. n = 6 per group. (g) Serum adiponectin concentrations in ob/ob mice gavaged with amlexanox
or vehicle control. n = 8 per group. (h) Oxygen consumption (VO
2
) of mice in each treatment group. n = 4 for the NCD groups, n = 8 for the HFD groups.
The HFD-fed amlexanox-treated mean values were significantly higher than the HFD-fed vehicle-treated mean values during all three light and dark
cycles; P < 0.05 (Students t test). (i) Core body temperature in NCD-fed mice gavaged with vehicle or amlexanox and in HFD-fed mice gavaged with
vehicle or amlexanox. *P < 0.05 for HFD-fed vehicle-treated controls compared to HFD-fed amlexanox-treated mice,

P < 0.05 for NCD-fed vehicle-

treated controls compared to HFD-fed vehicle-treated controls (ANOVA with Tukey-Kramer post hoc analysis). *P < 0.05 for ob/ob vehicle-treated
controls compared to ob/ob amlexanox-treated mice (Students t test). All data in the figure are shown as the mean s.e.m. or as a dot plot showing
each data point and the mean.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
nature medicine advance online publication S
was largely reversed by the drug, with a greater than 20% reduction
in liver weight (Fig. 5b). Moreover, triglyceride content in the liver
was more than 50% lower in the amlexanox-treated mice compared to
mice treated with vehicle control. Hepatic glycogen content, which is
elevated in HFD-fed mice, was reduced by amlexanox (Fig. 5c). Large
lipid droplets were apparent by H&E staining in the livers from vehicle-
treated control DIO mice but were largely disseminated by treatment
with amlexanox, which is consistent with the marked reductions in
triglycerides, glycogen and liver weight (Fig. 5d). These benefi-
cial reductions in hepatic lipids were reversed in livers from mice
that were taken off amlexanox treatment and continued on a HFD
(Fig. 5e). Steatosis was also resolved in ob/ob mice treated with the
drug (Supplementary Fig. 4ac).
To further investigate the hepatosteatotic phenotype, we measured
the expression of key metabolic and inflammatory genes in the livers
of NCD-fed, HFD-fed and amlexanox-treated HFD-fed (control)
mice. Mice treated with amlexanox had lower expression of certain
key lipogenic genes compared to control mice, including those encod-
ing fatty acid synthase (Fasn), malic enzyme 1 (Me1), acetyl CoA
carboxylase (Acaca) and stearyl CoA desaturase 1 (Scd1) (Fig. 5f).
Indicative of a higher lipid load, HFD was associated with higher
expression of Ppara, Pparg, Cd36, Fabp4 (encoding fatty acid bind-
ing protein 4) and Lpl (encoding lipoprotein lipase). The expression
of these genes was lower in amlexanox-treated and NCD-fed mice.
Gluconeogenic gene expression was modestly higher in HFD-fed
mice and was not lower with amlexanox treatment. Glycolytic genes,
the expression levels of which were higher in HFD-fed mice relative to
NCD-fed mice, were expressed at lower levels in amlexanox-treated
mice (Fig. 5g). In addition, we found lower levels of HFD-induced
expression of several inflammatory genes, including Tnfa, Cxcl10
and Ccl5 (also known as Rantes), which encode proinflammatory
cytokines (Fig. 5h), Ccl2 (also known as Mcp1) and Ccl3 (also
known as Mip-1), which encode chemotactic factors, and Emr1 and
Itgax (also known as F4/80 and Cd11c), which encode markers of
macrophage infiltration. Expression of Cybb, encoding cytochrome
b-245, -polypeptide, which is involved in the microbicidal oxidase
system of phagocytes, was lower in the drug-treated group. Genetic
measures of hepatosteatosis were similarly lower in ob/ob mice
(Supplementary Fig. 4d,e).
Amlexanox reduces adipose tissue inflammation in obese mice
Infiltration of inflammatory macrophages into WAT was markedly
lower in amlexanox-treated mice compared to vehicle-treated
controls, as assessed by H&E staining (Fig. 6a). This amlexanox-
dependent reduction in the appearance of macrophages in crown-
like structures
33
was accompanied by a marked attenuation in the
levels of mRNAs encoding key inflammatory genes in adipose tissue,
including Tnfa, Ccl2, Ccl3, Itgax and Emr1 (Fig. 6b). We also mea-
sured the concentration of cytokines in the serum of control and
amlexanox-treated mice. Although the drug had no effect on the circu-
lating concentrations of Mcp1 or Rantes, the concentrations of TNF-,
IL-1 and Mip-1 were lower (Fig. 6c). Serum concentrations of the
anti-inflammatory cytokine IL-10 (ref. 34) were higher in amlexanox-
treated mice compared to controls. We found higher expression of fat
0
50
100
150
0 30 60 90 120 150 180
P
e
r
c
e
n
t
a
g
e

o
f

b
a
s
a
l

b
l
o
o
d

g
l
u
c
o
s
e
Minutes after insulin injection
Vehicle
Amlexanox (100 mg per kg)
e
*
*
*
*
* *
0
100
200
300
400
500
0 30 60 90 120
B
l
o
o
d

g
l
u
c
o
s
e

(
m
g

d
l

1
)
Minutes after glucose gavage
NCD vehicle HFD amlexanox
(25 mg per kg)
a
* * *

#
# #
HFD vehicle
0
100
200
300
400
500
600
0 20 40 60 80 100 120
B
l
o
o
d

g
l
u
c
o
s
e

(
m
g

d
l

1
)
Minutes after glucose gavage
d
*
*
*
*
Vehicle
Amlexanox (100 mg per kg)
0
20
40
60
80
100
120
0 30 60 90 120
P
e
r
c
e
n
t
a
g
e

o
f

b
a
s
a
l

b
l
o
o
d

g
l
u
c
o
s
e
Minutes after insulin injection
NCD vehicle
NCD amlexanox (25 mg per kg)
HFD vehicle
HFD amlexanox (25 mg per kg)
c
*
* *
F
a
s
t
i
n
g

b
l
o
o
d

g
l
u
c
o
s
e

(
m
g

d
l

1
)
F
a
s
t
i
n
g

i
n
s
u
l
i
n

(
n
g

m
l

1
)
b

0
2
4
6
8
10
*
80
100
120
140
160
180
200
HFD amlexanox
(25 mg per kg)
HFD vehicle
NCD vehicle
f
0
1,000
2,000
3,000
4,000
A
U
C

G
I
G
0

1
2
0

m
i
n

(
m
g

p
e
r

k
g
)
Vehicle
Amlexanox
(50 mg per kg)
*
0
1
2
3
4
F
o
l
d

s
u
p
p
r
e
s
s
i
o
n
o
f

s
e
r
u
m

F
F
A
i
Vehicle
Amlexanox
(50 mg per kg)
*
h
0
10
20
30
40
50
G
l
u
c
o
s
e

t
u
r
n
o
v
e
r

r
a
t
e

(
m
g

p
e
r

k
g

p
e
r

m
i
n
)
Vehicle
Amlexanox
(50 mg per kg)
0
20
40
60
80
100
P
e
r
c
e
n
t
a
g
e
s
u
p
p
r
e
s
s
i
o
n

o
f

H
G
P
Vehicle
Amlexanox
(50 mg per kg)
*
g
0
5
10
15
20
25
Pre Post
H
G
P

(
m
g

p
e
r

k
g

p
e
r

m
i
n
)
Vehicle
Amlexanox
(50 mg per kg)
*
Figure 4 Amlexanox treatment improves insulin sensitivity and glucose tolerance. (a) Oral glucose tolerance test in the preventative treatment
group. n = 5 per group. (b) Fasting blood glucose and serum insulin concentrations in each treatment group. n = 8 per group. (c) Insulin tolerance
test in each treatment group. n = 8 per group. (d) Oral glucose tolerance test of ob/ob mice gavaged with amlexanox or vehicle control. n = 8 per
group. (e) Insulin tolerance test in ob/ob mice gavaged with amlexanox or vehicle control. n = 8 per group. (fi) Hyperinsulinemic-euglycemic clamp
in amlexanox-treated (50 mg per kg body weight) and vehicle-treated HFD-fed mice. n = 9 per group. (f) Area under the curve (AUC) for glucose
infusion rate during clamp (GIR
0120 min
). (g) Hepatic glucose production (HGP) before (pre) and after (post) the clamp (left) and percentage
suppression of hepatic glucose production (right). (h) Glucose turnover rate during the clamp. (i) Fold suppression of serum free fatty acid (FFA)
concentrations during the clamp. *P < 0.05 for HFD-fed vehicle-treated controls compared to HFD-fed amlexanox-treated mice or ob/ob vehicle-
treated controls compared to ob/ob amlexanox-treated mice,

P < 0.05 for NCD-fed vehicle-treated controls compared to HFD-fed vehicle-treated

controls (ANOVA with Tukey-Kramer post hoc analysis (ac) or Students t test (di)). All data in the figure are shown as the mean s.e.m. or as a dot
plot showing each data point and the mean.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
6 advance online publication nature medicine
cellenriched proteins such as Slc2a4 and Pparg, which are indicative
of improved insulin sensitivity, in WAT from amlexanox-treated mice
(Fig. 6b). The expression of uncoupling protein 1 (UCP1) was also
higher in the adipose tissue of amlexanox-treated DIO mice as
compared to HFD-fed vehicle-treated controls (Fig. 6d).
To evaluate whether amlexanox might control energy metabolism
by directly affecting inflammation, we examined the effects of the
drug on primary macrophages or the cultured RAW267.2 macro-
phage cell line. The addition of amlexanox did not affect macrophage
migration in a wound-healing assay performed in RAW267.2 cells
(Supplementary Fig. 5a). Bone marrowderived and peritoneal
macrophages from Ikbke knockout mice were slightly more responsive
to the effects of LPS in inducing the expression of various cytokines
(Supplementary Fig. 5bd). We also evaluated the effect of these
inhibitors on poly I:C or TNF-stimulated NF-B activation in
adipocytes and found a slight increase in the activation of NF-B after
pretreatment with the IKK- and TBK1 inhibitors (Supplementary
Fig. 5e). Moreover, bone marrow transplant from Ikbke-deficient mice
into wild-type mice on a HFD did not influence weight or insulin
sensitivity (data not shown). Together with the increase in TBK1
phosphorylation induced by the inhibitors described in Figure 2,
these data indicate that inhibition of IKK- and TBK1 does not
directly block the proinflammatory effects of cytokines or LPS on
inflammatory signaling or the NF-B pathway but, rather, may pre-
vent feedback inhibition that is a consequence of elevated expression
of TBK1 and IKK-. Thus, these noncanonical kinases do not act as
IB kinases and are not directly proinflammatory. However, IKK-
and TBK1 may attenuate the inflammatory response without elimi-
nating it. Thus, the induction of these counterinflammatory kinases
may contribute to the generation of low-grade, sustained inflamma-
tion in obesity by preventing its resolution.
Amlexanox promotes energy expenditure in adipose tissue
To discern whether amlexanox directly influences phosphorylation
in vivo, we first examined a number of proteins known to undergo
hyperphosphorylation in states of obesity. As previously described
35
,
phosphorylation of proteins in the mammalian target of rapamycin
complex1 (mTORC1) pathway such as S6K and S6 were elevated
in response to a HFD; this increase was largely attenuated in
amlexanox-treated DIO mice (Fig. 6d and Supplementary Fig. 6a,b).
HFD amlexanox HFD vehicle
d
NCD vehicle HFD withdrawal HFD amlexanox
e
HFD vehicle
a b
ND vehicle HFD vehicle HFD amlexanox
L
i
v
e
r

g
l
y
c
o
g
e
n
(

g

g
l
u
c
o
s
e

p
e
r

m
g
)

c
*
0
10
20
30
40
HFD amlexanox
(25 mg per kg)
HFD vehicle
*

L
i
v
e
r

t
r
i
g
l
y
c
e
r
i
d
e

(
m
g

p
e
r

g
)

*
L
i
v
e
r

m
a
s
s

(
g
)
0
1
2
3
0
100
200
300
400
HFD amlexanox (25 mg per kg) NCD vehicle
h
0
1
2
3
4
Tnfa Cxcl10 Ccl5 Emr1 Itgax Ccl2 Ccl3 Cybb
R
e
l
a
t
i
v
e

g
e
n
e

e
x
p
r
e
s
s
i
o
n
*

NCD vehicle HFD vehicle HFD amlexanox (25 mg per kg)

0
1
2
3
4
Fasn Me1 Acaca Scd1 Ppara Pparg Cd36
R
e
l
a
t
i
v
e

g
e
n
e

e
x
p
r
e
s
s
i
o
n
NCD vehicle HFD vehicle HFD amlexanox (25 mg per kg)
f g
0
1
2
3
4
G6pc Pck1 Pklr Gck
R
e
l
a
t
i
v
e

g
e
n
e

e
x
p
r
e
s
s
i
o
n

*
#
*
Fabp4 Lpl
NCD vehicle
HFD vehicle
HFD amlexanox
(25 mg per kg)
HFD vehicle
Figure 5 Amlexanox treatment
reverses hepatic steatosis.
(a) Macroscopic pictures
of the livers of mice in each
group indicated. Scale bar, 1 cm.
(b) Total liver weight and
triglyceride content of NCD-fed
mice or HFD-fed mice gavaged
with vehicle or amlexanox.
n = 6 per group. (c) Liver
glycogen content in HFD-fed mice
treated with amlexanox or vehicle. n = 5 per group. (d,e) Representative
images of H&E-stained livers in the groups indicated. Scale bars, 2 mm.
(fh) Expression of lipid metabolism genes (f), glucose metabolism genes (g)
and inflammatory genes and macrophage markers (h) in the livers of mice
in each treatment group. n = 6 per group. *P < 0.05 for HFD-fed vehicle-
treated controls compared to HFD-fed amlexanox-treated mice,

P < 0.05 for

NCD-fed vehicle-treated controls compared to HFD-fed vehicle-treated controls,
#
P < 0.1 for NCD-fed vehicle-treated controls compared to HFD-fed vehicle-
treated controls (ANOVA with Tukey-Kramer post hoc analysis (b) or
Students t test (c,fh)). All data in the figure are shown as the mean s.e.m. or
as a dot plot showing each data point and the mean.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
nature medicine advance online publication 7
Amlexanox also blocked induction of IKK- and TBK1 protein by a
HFD in adipose tissue. This finding was consistent with the fact that
overexpression of wild-type but not kinase-inactive TBK1 in cells
enhanced the stimulation of S6K phosphorylation by insulin, whereas
knockdown of Ikbke or Tbk1 in 3T3-L1 adipocytes lowered the amount
of insulin-stimulated S6 phosphorylation (Supplementary Fig. 6c). In
adipose tissue from HFD-fed mice, HSL phosphorylation was lower,
which is consistent with the well-established desensitization of the
-adrenergic pathway in this tissue during obesity
36
. Amlexanox
treatment prevented the reduction in HSL phosphorylation associ-
ated with a HFD (Fig. 6d and Supplementary Fig. 6a,b).
In vehicle-treated control mice on a HFD, brown adipose tissue
(BAT) accumulated large lipid droplets and acquired some of the
features of white fat. Brown fat reverted to its normal appearance in
amlexanox-treated mice (Supplementary Fig. 6d), with a corresponding
reversal of HFD-induced triglyceride accumulation (Supplementary
Fig. 6e). After withdrawal of amlexanox treatment, lipids again accu-
mulated in the BAT, reverting to the HFD phenotype of control obese
mice (Supplementary Fig. 6f). UCP1 protein expression was also
markedly elevated in the BAT of amlexanox-treated mice (Fig. 6e
and Supplementary Fig. 6g). Expression of the brown fatspecific
markers Prdm16, Ppargc1a, Ppargc1b, Cidea and Ucp1 was higher in
amlexanox-treated mice (Fig. 6f). There was also an increase in the
expression of cytochrome oxidases and Elovl3, which are involved in
lipid metabolism in the BAT. Amlexanox treatment reduced IKK-
and TBK1 protein expression in BAT (Fig. 6g and Supplementary
Fig. 6h). We also found reduced signaling through the mTORC1
pathway in BAT of amlexanox-treated mice as compared to controls.
Ex vivo treatment with amlexanox acutely stimulated lipid oxidation
and oxygen consumption in BAT explants from HFD-fed mice, support-
ing a direct effect of amlexanox on energy expenditure (Fig. 6h,i).
To evaluate the roles of IKK- and TBK1 in mediating the effects
of the drug, we treated Ikbke knockout mice with amlexanox and
monitored their weight loss and insulin sensitivity. The weight loss
produced by amlexanox was equivalent between wild-type and Ikbke
knockout mice, indicating that the upregulation of TBK1 in these
mice may compensate for IKK- inhibition (Supplementary Fig. 7a).
We also monitored insulin sensitivity by intermittent insulin toler-
ance tests over several weeks of treatment (Supplementary Fig. 7b).
Although Ikbke knockout mice were more insulin sensitive than their
wild-type littermates, the drug produced an additional improvement,
achieving the insulin sensitivity observed in wild-type mice treated
with amlexanox. Improvements in insulin sensitivity preceded weight
loss, indicating that improved insulin sensitivity may not be secondary
0
0.5
1.0
1.5
2.0
T
n
f
a
C
c
l3
C
c
l2
C
c
l5
C
x
c
l1
0
I
l1
0
I
t
g
a
x
E
m
r
1
S
lc
2
a
4
P
p
a
r
g
R
e
l
a
t
i
v
e

g
e
n
e

e
x
p
r
e
s
s
i
o
n
Vehicle
Amlexanox (100 mg per kg)
a b
*
*
*
*
*
0
100
200
300
400
500
600
I
l-
1
0
I
l-
1

M
c
p
1
M
ip
-
1

R
a
n
t
e
s
T
N
F
-

S
e
r
u
m

c
y
t
o
k
i
n
e

(
p
g

m
l

1
)
Vehicle
Amlexanox (100 mg per kg)
c
*
*
*
*
Vehicle Amlexanox
*
*
0
2
4
6
Fasted Fed
R
e
l
a
t
i
v
e
p
h
o
s
p
h
o
r
y
l
a
t
i
o
n
0
5
10
15
Fasted Fed
R
e
l
a
t
i
v
e
p
h
o
s
p
h
o
r
y
l
a
t
i
o
n
d
0
0.5
1.0
1.5
2.0
pTBK1
R
e
l
a
t
i
v
e
p
h
o
s
p
h
o
r
y
l
a
t
i
o
n
0
1
2
3
4
UCP1 IKK- TBK1
R
e
l
a
t
i
v
e

p
r
o
t
e
i
n

l
e
v
e
l
s
(
n
o
r
m
a
l
i
z
e
d

t
o

c
o
n
t
r
o
l
)
Vehicle
Amlexanox (25 mg per kg)
*
*
*
*
pS6
pHSL
* *
*
*
Vehicle
Amlexanox
0
50
100
150
200
WT DKO
O
C
R

(
p
m
o
l

m
i
n

1
p
e
r

w
e
l
l
)
*
j
0
20
40
60
80
100
L
i
p
i
d

o
x
i
d
a
t
i
o
n

r
a
t
e
(
c
.
p
.
m
.

m
g

1
h

1
)
Vehicle
Amlexanox
h
*
0
20
40
60
80
100
120
NCD HFD
O
C
R

(
n
m
o
l

m
i
n

1
g

1
)

Vehicle
Amlexanox
i
*
0
0.5
1.0
1.5
2.0
pTBK1 pS6
R
e
l
a
t
i
v
e
p
h
o
s
p
h
o
r
y
l
a
t
i
o
n
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
IKK- TBK1 S6
R
e
l
a
t
i
v
e

p
r
o
t
e
i
n

l
e
v
e
l
s
(
n
o
r
m
a
l
i
z
e
d

t
o

c
o
n
t
r
o
l
)
Vehicle
Amlexanox (25 mg per kg)
g
*
*
*
0
1
2
P
r
d
m
1
6
P
p
a
r
g
c
1
a
P
p
a
r
g
c
1
b
C
id
e
a
U
c
p
1
C
o
x
7
a
1
C
o
x
7
a
2
C
o
x
8
b
E
lv
o
l3
R
e
l
a
t
i
v
e

G
e
n
e
e
x
p
r
e
s
s
i
o
n

(
A
U
)
Vehicle
Amlexanox (25 mg per kg)
f
*
*
*
*
*
*
*
*
*
e
HFD vehicle HFD amlexanox
Figure 6 Amlexanox reduces
inflammation and increases energy
expenditure in adipose tissue.
(a) Representative images of
H&E-stained sections from WAT
of ob/ob mice treated with vehicle
control or amlexanox. Scale bars, 1 mm. (b) Expression of inflammatory genes and macrophage markers in WAT from ob/ob mice gavaged with vehicle
control or amlexanox. n = 6 per group. (c) Serum cytokine concentrations in ob/ob mice gavaged with vehicle control or amlexanox. n = 6 per group.
(d) Quantified amounts of protein and phosphorylation in WAT from mice treated with amlexanox or vehicle control (see also Supplementary Fig. 6b).
(e) BAT stained for UCP1 in HFD-fed vehicle-treated controls and amlexanox-treated mice. Scale bars, 200 m. (f) Expression of BAT-specific markers
in BAT. n = 8 per group. AU, arbitrary units. (g) Quantified amounts of protein and phosphorylation in BAT from mice treated with amlexanox or
vehicle control (see also Supplementary Fig. 6h). (h) Lipid oxidation rate in ex vivo BAT treated with amlexanox or vehicle control. n = 6 per group.
(i) Oxygen consumption rate (OCR) in ex vivo BAT from NCD-fed and HFD-fed mice treated with amlexanox or vehicle control. n = 5 per group. (j) Oxygen
consumption rate in wild-type (WT) and Ikbke and Tbk1 double-knockout (DKO) MEFs treated with amlexanox or vehicle control. n = 5 wells per group.
*P < 0.05 for vehicle-treated controls compared to amlexanox-treated mice (Students t test). All data in the figure are shown as the mean s.e.m.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
8 advance online publication nature medicine
to weight loss. Taken together with the data in Figure 4, which show
that the effects of amlexanox are comparable to those of other
insulin-sensitizing drugs such as the thiazolidinediones
37
, these
results suggest that amlexanox is a bona fide insulin sensitizer.
As Tbk1 knockout mice are not viable, we used mouse embryonic
fibroblasts (MEFs) to further investigate the importance of these
kinases in the specificity of amlexanox. Treatment of wild-type MEFs
with amlexanox produced a dose-dependent increase in the phospho-
rylation of TBK1 and the P65 NF-B subunit, presumably because
of feedback inhibition
27
. Both effects were absent in Ikbke and Tbk1
double-knockout MEFs (Supplementary Fig. 7ce). However,
MEFs with single knockout of either Ikbke or Tbk1 retained respon-
siveness to amlexanox. In Ikbke knockout MEFs, we found TBK1
phosphorylation and degradation of IB after amlexanox treat-
ment, whereas Tbk1 knockout MEFs showed robust phosphoryla-
tion of IKK- in response to amlexanox treatment. Amlexanox also
enhanced oxygen consumption in wild-type but not Ikbke and Tbk1
double-knockout MEFs, indicating that the energy expenditure-
promoting effect of amlexanox is mediated specifically through inhi-
bition of both IKK- and TBK1 (Fig. 6j). Taken together, these data
suggest that amlexanox elicits its effects though specific inhibition
of both IKK- and TBK1.
DISCUSSION
Compelling evidence has accrued over the past few years that both
genetic and dietary forms of rodent obesity are accompanied by the
generation of low-grade inflammation in key metabolic tissues, which
serves as an important link between obesity and sustained insulin
resistance
711
. Almost all rodent models of obesity-induced insulin
resistance are associated with chronic inflammation, and a strong cor-
relation between insulin resistance and inflammatory markers in stud-
ies of several patient groups has been shown
3841
. This inflammation
is generally low grade and is accompanied by macrophage switching
from a type 2 to a type 1 polarization state
34,42,43
, as well as the pres-
ence of additional inflammatory cells
4447
. Inhibition of inflammatory
signaling by knockout of key pathways in obese mice can disrupt the
link between obesity and insulin resistance
11,4855
. Likewise, differ-
ent anti-inflammatory pharmacological and genetic approaches can
also reduce insulin resistance in rodents and patients
5658
. Coculture
studies and neutralizing antibodies directed against different proin-
flammatory cytokines have demonstrated in vitro that inflammatory
signals derived from M1-polarized macrophages can desensitize cells
to insulin through different mechanisms
23
.
Although the molecular links between inflammation and insulin
resistance are not yet completely understood, it is clear that the NF-B
program has a key role
13
. Disruption of NF-B signaling by targeted
knockout of the canonical IB kinase IKK- gene or pharmacologi-
cal inhibition of this pathway can restore insulin sensitivity in obese
states. However, the roles of the noncanonical IKKs, TBK1 and IKK-
in this process are less well understood. Our previous studies indicated
that deletion of the Ikbke gene partially restored insulin sensitivity
and reduced weight in mice fed a HFD despite having no apparent
effect on NF-B signaling, although the relevant mechanism was not
defined
15
. Here we describe amlexanox, a newly identified chemical
inhibitor of IKK- and TBK1 that reproduces many of these effects.
Administration of this compound to mice reduced weight, insulin
resistance, inflammation and steatosis in three different obese mouse
models. These effects are apparent within 34 weeks after starting drug
therapy and are reversible. They are probably attributed to increased
triglyceride hydrolysis and oxidation, along with reduced lipogenesis
and glycogenesis. Signaling pathways that modulate these processes
are attenuated in vitro by the addition of amlexanox or another struc-
turally unrelated inhibitor to cultured adipocytes.
The precise mechanisms by which amlexanox produces these bene-
ficial effects in obese rodents have not yet been completely elucidated.
Although amlexanox is known to be a mast cell stabilizer of unknown
mechanism
20
, and depletion of mast cells may have beneficial meta-
bolic effects
59
, most of the in vivo and in vitro evidence points to a
role for the drug in increasing expenditure of energy while reducing
its storage in adipocytes and hepatocytes. Furthermore, the lack of
a phenotype in wild-type mice reconstituted with Ikbke knockout
bone marrow indicates that the role of IKK- in bone marrow-derived
cells such as mast cells and macrophages is less important than its
role in other cell types such as adipocytes and hepatocytes. Although
IKK- and TBK1 expression is elevated as part of the inflammatory
program downstream of NF-B, the kinase targets of the drug do not
seem to be direct participants in the increased inflammatory program.
In fact, the reduced inflammation observed in vivo with amlexanox
treatment may be an indirect effect of improved metabolic disease
or, perhaps, of elimination of a feedback pathway that maintains
inflammation at low levels such that inflammation is permitted to
resolve. Moreover, despite the fact that administration of amlexanox
to obese mice restores insulin sensitivity, these compounds are not
direct insulin sensitizers in vitro. Phosphorylation analyses indicate
that IKK- and TBK1 may co-opt signaling pathways that drive energy
conservation in the context of insulin resistance, ensuring that energy
expenditure is kept in check by producing catecholamine resistance
in adipose tissue, in the process reducing HSL phosphorylation and
UCP1 expression while also increasing mTOR signaling to drive
lipogenesis and glycogen synthesis. Thus, together these data suggest
that expression of TBK1 and IKK- by obesity-dependent inflam-
mation may conserve energy during the transition from an insulin-
dependent to an insulin-independent state of nutrient storage.
Moreover, it is probable that the improvement in insulin resist-
ance produced by treatment with amlexanox results from reducing
excessive energy storage while preventing homeostatic feedback loops
(such as mTOR) that block insulin receptor signaling
60
.
Amlexanox seems to be a relatively safe drug with a long history of
use in patients. As such, there might be an interesting opportunity for
repurposing this agent for diabetes and obesity. Future clinical studies
will be needed to explore this possibility and validate the hypothesis
that these noncanonical IB kinases contribute to counterinflamma-
tory actions that sustain metabolic disease.
METHODS
Methods and any associated references are available in the online
version of the paper.
Note: Supplementary information is available in the online version of the paper.
ACkNOWLEDgMENTS
We thank J. Hung and X. Peng for excellent technical assistance and members
of the Saltiel laboratory for helpful discussions. This work was supported
by US National Institutes of Health grants DK60597 and 60591 to A.R.S.,
F32DK09685101 to S.M.R., F30DK089687 to J.M. and R24DK090962 to A.R.S.,
J.M.O. and R.M.E. R.M.E., M.D. and R.T.Y. are supported by the Leona M. and
Harry B. Helmsley Charitable Trust. We thank S. Akira (Department of Host
Defense, Research Institute for Microbial Diseases, Osaka University, Osaka,
Japan) for providing MEFs. We also acknowledge support from the Michigan
Diabetes Research and Training Center (DK020572), Michigan Institute for
Clinical and Health Research (UL1-RR024986), Nathan Shock Center, Michigan
Metabolic Phenotyping Core and Michigan Nutritional Obesity Research
Center (DK089503).

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
arti cl e s
nature medicine advance online publication 9
AUTHOR CONTRIBUTIONS
A.R.S. and S.M.R. wrote the manuscript. S.M.R. created figures. S.M.R. produced
the data in Figures 1e, 3e,i, 4fi, 5c,fh and 6cj and in Supplementary Figures
1, 2ch, 5b,e,g,h and 6a,b. S.-H.C. produced the data in Figures 3ad,fh,j, 4ae,
5a,ce and 6a,b and in Supplementary Figures 2a,b, 3 and 5a,d,f. S.J.D. and M.J.L.
performed the screen identifying amlexanox as an IKK- inhibitor. L.C. and S.J.D.
produced the data in Figures 1b and 2b,c,f. M.U. produced the data in Figures 1a
and 2e and in Supplementary Figures 5c and 6ce. J.R.R. produced the images
in Figure 1d. J.M. produced the data in Figure 1c. N.M.W. produced the data in
Supplementary Figure 4ad. I.H. produced the data in Supplementary Figure 4e.
M.D., R.T.Y., C.L. and R.M.E. contributed data not shown, which was instrumental
in generating Figure 5fh. D.O. produced the data in Figure 1d and the samples
for Figure 1e. P.L. produced the data in Figure 1f under the guidance of J.M.O.
COMPETING FINANCIAL INTERESTS
The authors declare no competing financial interests.
Published online at http://www.nature.com/doifinder/10.1038/nm.3082.
Reprints and permissions information is available online at http://www.nature.com/
reprints/index.html.
1. Taniguchi, C.M., Emanuelli, B. & Kahn, C.R. Critical nodes in signalling pathways:
insights into insulin action. Nat. Rev. Mol. Cell Biol. 7, 8596 (2006).
2. Reaven, G.M. The insulin resistance syndrome: defnition and dietary approaches
to treatment. Annu. Rev. Nutr. 25, 391406 (2005).
3. Reaven, G.M. Why Syndrome X? From Harold Himsworth to the insulin resistance
syndrome. Cell Metab. 1, 914 (2005).
4. Stumvoll, M., Goldstein, B.J. & van Haeften, T.W. Type 2 diabetes: principles of
pathogenesis and therapy. Lancet 365, 13331346 (2005).
5. Biddinger, S.B. & Kahn, C.R. From mice to men: insights into the insulin resistance
syndromes. Annu. Rev. Physiol. 68, 123158 (2006).
6. Doria, A., Patti, M.E. & Kahn, C.R. The emerging genetic architecture of type 2
diabetes. Cell Metab. 8, 186200 (2008).
7. Olefsky, J.M. & Glass, C.K. Macrophages, infammation, and insulin resistance.
Annu. Rev. Physiol. 72, 219246 (2010).
8. Li, P. et al. Functional heterogeneity of CD11c-positive adipose tissue macrophages
in diet-induced obese mice. J. Biol. Chem. 285, 1533315345 (2010).
9. Lumeng, C.N. & Saltiel, A.R. Infammatory links between obesity and metabolic
disease. J. Clin. Invest. 121, 21112117 (2011).
10. Gregor, M.F. & Hotamisligil, G.S. Infammatory mechanisms in obesity. Annu. Rev.
Immunol. 29, 415445 (2011).
11. Hotamisligil, G.S. Endoplasmic reticulum stress and the infammatory basis of
metabolic disease. Cell 140, 900917 (2010).
12. Wunderlich, F.T. et al. Hepatic NF-B essential modulator defciency prevents
obesity-induced insulin resistance but synergizes with high-fat feeding in
tumorigenesis. Proc. Natl. Acad. Sci. USA 105, 12971302 (2008).
13. Arkan, M.C. et al. IKK- links infammation to obesity-induced insulin resistance.
Nat. Med. 11, 191198 (2005).
14. Goldfne, A.B. et al. The effects of salsalate on glycemic control in patients with
type 2 diabetes: a randomized trial. Ann. Intern. Med. 152, 346357 (2010).
15. Chiang, S.H. et al. The protein kinase IKK regulates energy balance in obese mice.
Cell 138, 961975 (2009).
16. Chau, T.L. et al. Are the IKKs and IKK-related kinases TBK1 and IKK-epsilon
similarly activated? Trends Biochem. Sci. 33, 171180 (2008).
17. Chariot, A. The NF-Bindependent functions of IKK subunits in immunity and
cancer. Trends Cell Biol. 19, 404413 (2009).
18. Kawai, T. & Akira, S. Signaling to NF-B by Toll-like receptors. Trends Mol. Med. 13,
460469 (2007).
19. Kravchenko, V.V., Mathison, J.C., Schwamborn, K., Mercurio, F. & Ulevitch, R.J.
IKKi/IKK plays a key role in integrating signals induced by pro-infammatory stimuli.
J. Biol. Chem. 278, 2661226619 (2003).
20. Makino, H., Saijo, T., Ashida, Y., Kuriki, H. & Maki, Y. Mechanism of action of an
antiallergic agent, amlexanox (AA-673), in inhibiting histamine release from mast
cells. Acceleration of cAMP generation and inhibition of phosphodiesterase.
Int. Arch. Allergy Appl. Immunol. 82, 6671 (1987).
21. Bell, J. Amlexanox for the treatment of recurrent aphthous ulcers. Clin. Drug
Investig. 25, 555566 (2005).
22. Saltiel, A.R. Insulin resistance in the defense against obesity. Cell Metab. 15,
798804 (2012).
23. Lumeng, C.N., Deyoung, S.M. & Saltiel, A.R. Macrophages block insulin action in
adipocytes by altering expression of signaling and glucose transport proteins. Am.
J. Physiol. Endocrinol. Metab. 292, E166E174 (2007).
24. Karin, M., Yamamoto, Y. & Wang, Q.M. The IKK NF-B system: a treasure trove for
drug development. Nat. Rev. Drug Discov. 3, 1726 (2004).
25. Oh, D.Y. et al. GPR120 is an omega-3 fatty acid receptor mediating potent anti-
infammatory and insulin-sensitizing effects. Cell 142, 687698 (2010).
26. Ma, X. et al. Molecular basis of Tank-binding kinase 1 activation by
transautophosphorylation. Proc. Natl. Acad. Sci. USA 109, 93789383 (2012).
27. Clark, K., Takeuchi, O., Akira, S. & Cohen, P. The TRAF-associated protein TANK
facilitates cross-talk within the IB kinase family during Toll-like receptor signaling.
Proc. Natl. Acad. Sci. USA 108, 1709317098 (2011).
28. Bamborough, P. et al. 5-(1H-Benzimidazol-1-yl)-3-alkoxy-2-thiophenecarbonitriles
as potent, selective, inhibitors of IKK-n kinase. Bioorg. Med. Chem. Lett. 16,
62366240 (2006).
29. Clark, K. et al. Novel cross-talk within the IKK family controls innate immunity.
Biochem. J. 434, 93104 (2011).
30. Torii, H. et al. Metabolic fate of amoxanox (AA-673), a new antiallergic agent, in
rats, mice, guinea-pigs and dogs. Jpn. Pharmacol. Ther. 13, 49334954 (1985).
31. Friedman, J.M. Modern science versus the stigma of obesity. Nat. Med. 10, 563569
(2004).
32. Friedman, J.M. & Halaas, J.L. Leptin and the regulation of body weight in mammals.
Nature 395, 763770 (1998).
33. Weisberg, S.P. et al. Obesity is associated with macrophage accumulation in adipose
tissue. J. Clin. Invest. 112, 17961808 (2003).
34. Lumeng, C.N., Bodzin, J.L. & Saltiel, A.R. Obesity induces a phenotypic switch in
adipose tissue macrophage polarization. J. Clin. Invest. 117, 175184 (2007).
35. Um, S.H. et al. Absence of S6K1 protects against age- and diet-induced obesity
while enhancing insulin sensitivity. Nature 431, 200205 (2004).
36. Seals, D.R. & Bell, C. Chronic sympathetic activation: consequence and cause of
age-associated obesity? Diabetes 53, 276284 (2004).
37. Olefsky, J.M. & Saltiel, A.R. PPAR and the treatment of insulin resistance. Trends
Endocrinol. Metab. 11, 362368 (2000).
38. Pradhan, A.D., Manson, J.E., Rifai, N., Buring, J.E. & Ridker, P.M. C-reactive
protein, interleukin 6, and risk of developing type 2 diabetes mellitus. J. Am. Med.
Assoc. 286, 327334 (2001).
39. Festa, A., DAgostino, R. Jr., Tracy, R.P. & Haffner, S.M. Elevated levels of acute-
phase proteins and plasminogen activator inhibitor-1 predict the development of type
2 diabetes: the insulin resistance atherosclerosis study. Diabetes 51, 11311137
(2002).
40. Blackburn, P. et al. Postprandial variations of plasma infammatory markers in
abdominally obese men. Obesity (Silver Spring) 14, 17471754 (2006).
41. Aron-Wisnewsky, J. et al. Human adipose tissue macrophages: m1 and m2 cell
surface markers in subcutaneous and omental depots and after weight loss. J. Clin.
Endocrinol. Metab. 94, 46194623 (2009).
42. Schenk, S., Saberi, M. & Olefsky, J.M. Insulin sensitivity: modulation by nutrients
and infammation. J. Clin. Invest. 118, 29923002 (2008).
43. Odegaard, J.I. & Chawla, A. Alternative macrophage activation and metabolism.
Annu. Rev. Pathol. 6, 275297 (2011).
44. Nishimura, S. et al. CD8
+
effector T cells contribute to macrophage recruitment
and adipose tissue infammation in obesity. Nat. Med. 15, 914920 (2009).
45. Winer, S. et al. Normalization of obesity-associated insulin resistance through
immunotherapy. Nat. Med. 15, 921929 (2009).
46. Feuerer, M. et al. Lean, but not obese, fat is enriched for a unique population of
regulatory T cells that affect metabolic parameters. Nat. Med. 15, 930939 (2009).
47. Winer, D.A. et al. B cells promote insulin resistance through modulation of T cells
and production of pathogenic IgG antibodies. Nat. Med. 17, 610617 (2011).
48. Nakamura, T. et al. Double-stranded RNAdependent protein kinase links pathogen
sensing with stress and metabolic homeostasis. Cell 140, 338348 (2010).
49. Summers, S.A. Sphingolipids and insulin resistance: the fve Ws. Curr. Opin. Lipidol. 21,
128135 (2010).
50. Vandanmagsar, B. et al. The NLRP3 infammasome instigates obesity-induced
infammation and insulin resistance. Nat. Med. 17, 179188 (2011).
51. Saberi, M. et al. Hematopoietic cell-specifc deletion of toll-like receptor 4
ameliorates hepatic and adipose tissue insulin resistance in high-fatfed mice. Cell
Metab. 10, 419429 (2009).
52. Wellen, K.E. et al. Coordinated regulation of nutrient and infammatory responses
by STAMP2 is essential for metabolic homeostasis. Cell 129, 537548 (2007).
53. Lesniewski, L.A. et al. Bone marrowspecifc Cap gene deletion protects against
high-fat dietinduced insulin resistance. Nat. Med. 13, 455462 (2007).
54. Holland, W.L. et al. Inhibition of ceramide synthesis ameliorates glucocorticoid-,
saturated-fat-, and obesity-induced insulin resistance. Cell Metab. 5, 167179
(2007).
55. Shi, H. et al. TLR4 links innate immunity and fatty acid-induced insulin resistance.
J. Clin. Invest. 116, 30153025 (2006).
56. Solomon, D.H. et al. Association between disease-modifying antirheumatic drugs
and diabetes risk in patients with rheumatoid arthritis and psoriasis. J. Am. Med.
Assoc. 305, 25252531 (2011).
57. Hundal, R.S. et al. Mechanism by which high-dose aspirin improves glucose
metabolism in type 2 diabetes. J. Clin. Invest. 109, 13211326 (2002).
58. Zhang, X. et al. Selective inactivation of c-Jun NH2-terminal kinase in adipose
tissue protects against diet-induced obesity and improves insulin sensitivity in both
liver and skeletal muscle in mice. Diabetes 60, 486495 (2011).
59. Liu, J. et al. Genetic defciency and pharmacological stabilization of mast cells reduce
diet-induced obesity and diabetes in mice. Nat. Med. 15, 940945 (2009).
60. Zick, Y. Ser/Thr phosphorylation of IRS proteins: a molecular basis for insulin
resistance. Sci. STKE 2005, pe4 (2005).

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
nature medicine doi:10.1038/nm.3082
ONLINE METHODS
Reagents. All chemicals were obtained from Sigma-Aldrich unless otherwise
stated. Antibodies to IKK- (2690), TBK1 (3013), pTBK1 (Ser172; 5483), IKK-
and IKK- (2684), pIKK- and pIKK- (Ser176; 2687), AKT (9272), pAKT
(Ser473; 9272), S6K (2708, diluted 1:250), pS6K (Thr389; 9205), S6 (2317), pS6
(Ser235 and Ser236; 2211), IRF3 (4302), pIRF3 (Ser396; 4947), HSL (4107),
pHSL (Ser563; 4139, or Ser660; 4126) and PPAR- (2443) were purchased
from Cell Signaling. RalA-specific antibody (610222) was obtained from BD
Bioscience. UCP1-specific antibody (UCP11-A, diluted 1:250 for WAT detec-
tion) was obtained from Alpha Diagnostics. All antibodies were diluted 1:1,000
unless otherwise noted. Enhanced chemiluminescence reagents were pur-
chased from NEN, Inc. EDTA-free protease inhibitor tablet was purchased from
Roche Diagnostics.
Animals and animal care. We fed wild-type male C57BL/6 mice a HFD consist-
ing of 45% of calories from fat (D12451 Research Diets Inc.) starting at 8 weeks
of age for 1224 weeks, and we maintained NCD-fed male C57BL/6 controls on
normal chow diet consisting of 4.5% fat (5002 Lab Diet). We fed the C57BL/6
mice diets containing -3 fatty acids as previously described
25
. Rosiglitazone
treatment was administered for 3 weeks by addition of the compound to the diet
in mice that had been on HFD for 16 weeks. Each mouse consumed on average
3.5 mg per kg body weight rosiglitazone per day. Amlexanox was administered
by daily oral gavage. For the prevention groups, amlexanox (25 mg per kg or
100 mg per kg body weight) administration was begun concurrently with HFD
feeding at 8 weeks of age. For the treatment groups, 25 mg per kg body weight
amlexanox treatment was begun at 20 weeks of age after 12 weeks of HFD.
To test the effect of amlexanox withdrawal, mice in the treatment group were
switched from amlexanox gavage to vehicle control after 8 weeks of amlexanox
treatment. We fed control and ob/ob mice a normal chow diet and gavaged them
with 100 mg per kg body weight amlexanox or vehicle control beginning at
10 weeks of age. Mice were housed in a specific pathogen-free facility with a
12-h light, 12-h dark cycle and given free access to food and water. All animal
use was in compliance with the Institute of Laboratory Animal Research Guide
for the Care and Use of Laboratory Animals and approved by the University
Committee on Use and Care of Animals at the University of Michigan and
University of California San Diego.
Food intake. The remaining weight of food provided was determined daily for
singly housed mice. Daily food consumption was calculated from a 3-d average.
Energy expenditure and respiratory quotient. C57BL/6 mice in the amlexanox
treatment group were placed in metabolic cages. The University of Michigan
Metabolic Phenotyping Core measured oxygen consumption (VO
2
), carbon
dioxide production (VCO
2
) and spontaneous motor activity during 3 consecu-
tive days using the Comprehensive Laboratory Monitoring System (Columbus
Instruments), an integrated open-circuit calorimeter equipped with an optical
beam activity monitoring system
53
. Values presented are normalized to body
weight. The respiratory quotient was calculated by dividing the carbon dioxide
production by the oxygen consumption. We used the mean values for the light
and dark cycles to analyze statistical significance.
Body composition. The University of Michigan Metabolic Phenotyping
Core used nuclear magnetic resonance analysis to quantify body fat, lean body
mass and fluid content in ob/ob mice and C57BL/6 mice in the amlexanox
treatment group.
Hyperinsulinemic-euglycemic clamp. C57BL/6 mice were placed on HFD
starting at 8 weeks of age, and starting at 16 weeks of age, the mice were
gavaged daily with 50 mg per kg body weight amlexanox or vehicle control.
After 4 weeks of treatment, the right jugular vein and carotid artery were
surgically catheterized, and mice were given 5 d to recover from the surgery.
During this recovery period, amlexanox was delivered in drinking water. On
the day of the clamp, a final oral gavage treatment was performed. After a 5- to
6-h fast, hyperinsulinemic clamp studies were performed on conscious mice
using the protocol adopted from the Vanderbilt Mouse Metabolic Phenotyping
Center by the University of Michigan Animal Phenotyping Core consisting of a
90-min equilibration period followed by a 120-min experimental period (t = 0
to t = 120 min). Insulin was infused at 4.0 mU per kg body weight per min.
This experiment was performed at the University of Michigan Metabolic
Phenotyping Core. We measured serum free fatty acid concentrations using
the colorimetric NEFA-HR(2) kit (Wako).
Core body temperature. We performed rectal temperature measurements using
a YSI 4600 Precision thermometer (YSI, Inc.).
Blood chemistry analysis. Blood glucose was measured by OneTouch Ultra
Glucometer. Plasma from mice fasted for 6 h was isolated from whole blood
collected into heparinized tubes. We used an insulin ELISA kit (Crystal Chem
Inc.) to measure serum insulin concentrations. Leptin and adiponectin con-
centrations were measured by ELISA kits purchased from Cayman Chem Inc.
We quantified serum cytokine concentrations using luminex technology in a
multianalyte panel plate purchased from Millipore. Additionally, we measured
TNF- concentrations using ELISA kits purchased from R&D Systems.
Glucose and insulin tolerance tests. For glucose tolerance tests, after a 6-h fast,
we gavaged mice with glucose at a dose of 1.5 g per kg body weight (C57BL/6
mice) or 1.2 g per kg body weight (ob/ob mice). For insulin tolerance tests,
mice were fasted for 3 h and then given an intraperitoneal injection of insulin
(1.2 units per kg body weight for C57BL/6 mice and 2.0 units per kg body
weight for ob/ob mice). We measured the basal blood glucose level and then at
15, 30, 45, 60, 90, 120 and 180 min from tail blood using the OneTouch Ultra
glucometer (Lifescan).
Tissue lipid content. We isolated liver lipids by ethanolic KOH saponification
and BAT lipids by chloroform extraction and then quantified triglyceride con-
centrations using the Triglyceride Reagent kit (Thermo Scientific).
Liver glycogen content. We digested liver tissue in a 30% potassium hydro-
xide solution and then precipitated glycogen using 70% ethanol. After three
washes (resuspend pellet in water, then add ethanol to 70% and centrifuge
down the pellet) to remove any traces of glucose, we digested glycogen by
addition of amyloglucosidase. Released glucose was quantified using a colori-
metric kit (Wako).
Stromal vascular fraction (SVF) and adipocyte isolation. We digested
excised WAT in PBS containing 1% BSA and 1 mg ml
1
type II collagenase for
30 min at 37 C with gentle agitation. The cell suspension was filtered through a
100-m filter and then centrifuged at 700g for 5 min to separate floating
adipocytes from the SVF pellet. We washed floating adipocytes twice with
PBS containing 1% BSA and collected the SVF pellets after each wash.
Western blot analysis. We homogenized tissues in lysis buffer (50 mM Tris,
pH 7.5, 5 mM EDTA, 250 mM sucrose, 1% NP-40, 2 mM dithiothreitol (DTT),
1 mM sodium vanadate, 100 mM NaF, 10 mM Na
4
P
2
O
7
and freshly added
protease inhibitor tablet) and then incubated them for 1 h at 4 C
15
. We cen-
trifuged crude lysates at 14,000g for 15 min twice and determined the protein
concentration using BioRad Protein Assay Reagent. Samples were diluted in SDS
sample buffer. Bound proteins were resolved by SDS-PAGE and transferred to
nitrocellulose membranes (Bio-Rad). Individual proteins were detected with
the specific antibodies and visualized on film using horseradish peroxidase
conjugated secondary antibodies (Bio-Rad) and Western Lightning Enhanced
Chemiluminescence (Perkin Elmer Life Sciences). Alternatively, infrared flu-
orescent secondary antibodies were used for detection and quantification of
specific protein on the Odyssey CLx imager (LI-COR).
Histochemistry. Tissues were fixed in formalin for 3 d. The University of
Michigan Cancer Center Research Histology Laboratory performed histology.
A 1:100 dilution of the UCP1-specific antibody (UCP11, Alpha Diagnostics) at
10 g ml
1
was used for immunohistochemical detection of UCP1 protein.
Gene expression analysis. We rinsed isolated mouse tissues in PBS and then
froze them in liquid nitrogen and stored them at 80 C until extraction.

2
0
1
3

N
a
t
u
r
e

A
m
e
r
i
c
a
,

I
n
c
.


A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.
nature medicine doi:10.1038/nm.3082
We extracted total RNA from liver, WAT and BAT tissues, as well as differentiated
3T3-L1 cells, using the RNeasy Lipid Tissue Kit (Qiagen) according to the
manufacturers instructions with the inclusion of a DNase digestion step. We
extracted total RNA from bone marrowderived macrophages and SVF cells
using the RNeasy Kit (Qiagen) with a DNase step. We used the Superscript
First-Strand Synthesis System for RT-PCR (Invitrogen) with random primers
for reverse transcription. Real-time PCR amplification of the complementary
DNA was performed on samples in triplicate with Power SYBR Green PCR
Master Mix (Applied Biosystems) using the Applied Biosystems 7900HT Fast
real-time PCR System. We chose Adrp (also known as Plin2) and GAPDH as
the internal controls for normalization after screening several candidate genes;
their expression was not substantially affected by experimental conditions.
The sequences of all primers used in this study are listed in Supplementary
Table 1. Data were analyzed using the 2
CT
method, and the control sample
value was normalized to 1. Other data were quantified using an internal stand-
ard curve. Statistical significance was determined using the unpaired hetero-
scedastic Students t test with one averaged sample value per mouse.
Lipid oxidation rate. We excised intrascapular BAT and placed it in DMEM
with 2% BSA with or without 50 M amlexanox then incubated it at 37 C
for 1 h, after which we changed the medium to DMEM with 2% BSA,
0.25 mM carnitine, 0.2 mM palmitic acid and trace levels of
3
H-palmitic acid
and incubated it for 1 more hour at 37 C and then collected the medium
and isolated the aqueous faction. Lipid oxidation was determined by the conver-
sion of
3
H-palmitic acid to
3
H
2
O.
IKK- and TBK1 in vitro kinase assays. We performed in vitro kinase assays by
incubating purified kinase (IKK- or TBK1) in kinase buffer containing 25 mM
Tris (pH 7.5), 10 mM MgCl2, 1 mM DTT and 10 M ATP for 30 min at 30 C in
the presence of 0.5 Ci -[
32
P]-ATP and 1 g MBP per sample as a substrate. We
stopped the kinase reaction by adding 4 SDS sample buffer and boiling for 5
min at 95 C. Supernatants were resolved by SDS-PAGE, transferred to nitrocel-
lulose and analyzed by autoradiography using a Typhoon 9410 phosphorimager
(GE Lifesciences). We quantified the bands using ImageQuant.
IKK- and TBK1 immunocomplex kinase assay. We collected liver and WAT
from C57BL/6 mice on NCD or HFD and homogenized the tissues using a Dounce
homogenizer with lysis buffer containing 50 mM Tris (pH 7.5), 150 mM NaCl,
2 mM EDTA, 5 mM NaF, 25 mM -glycerophosphate, 1 mM sodium orthovanad-
ate, 10% glycerol, 1% Triton X-100, 1 mM DTT and 1 mM PMSF in the presence
of protease inhibitors (Roche Diagnostics). We incubated tissue cell lysates for
1 h at 4 C and cleared them by spinning at 13,000 r.p.m. for 15 min at 4 C in a
table-top centrifuge. Each 1 mg of lysate was subjected to immunoprecipitation
using 5 l of undiluted rabbit polyclonal antibody to TBK1 or IKK- for 1.5 h at
4 C. We harvested immunocomplexes by incubation with ProtA beads (Roche
Diagnostics) for 2 h at 4 C. We washed immunoprecipitates once with lysis
buffer and three times with wash buffer containing 20 mM 4-(2-hydroxyethyl)-
1-piperazineethanesulfonic acid (HEPES) (pH 7.4), 50 mM NaCl, 20 mM
-glycerophosphate, 1 mM sodium orthovanadate, 5 mM NaF, 10 mM MgCl
2
and
1 mM DTT. We performed an in vitro kinase assay using the immunoprecipitated
kinases as described above. Relative amounts of MBP phosphorylation were
detected by autoradiograph and normalized to the amounts of IKK- or TBK1
kinase detected in the immunoprecipitate by immunoblotting.
Cell culture. 3T3-L1 fibroblasts (American Type Culture Collection) were
cultured in 10% neonatal calf serum in DMEM for proliferation and 10%
FBS in DMEM for differentiation. We grew the cells to confluence and then
induced differentiation 2 d later with 500 M 3-isobutyl-1-methylxanthine,
250 nM dexamethasone and insulin (1 g ml
1
) for 3 d followed by insulin-only
treatment for 2 d. We routinely used cells within 7 d after completion of the
differentiation process and only used cultures in which >90% of cells showed
adipocyte morphology. We serum starved 3T3-L1 adipocytes with 0.5% FBS
in DMEM before treatment. TNF- treatments (50 ng ml
1
, unless otherwise
noted) were performed during the 24 h before harvest and after pretreatment
with the IKK- inhibitor compound VIII (EMD Biosciences) for 1 h where
indicated. We pretreated 3T3-L1 adipocytes for 1 h with amlexanox at the
given concentrations and then treated them with 20 g ml
1
of poly I:C for
1 h. Alternatively, we treated 3T3-L1 adipocytes with 50 M forskolin for
15 min after a 30-min amlexanox pretreatment. We treated cells with or without
10 nM of insulin for 15 min. We determined glycerol release into the superna-
tant (DMEM without phenol red) using a colorimetric assay. We serum starved
RAW264.7 cells with 0.5% FBS in DMEM medium and pretreated them with
or without Cay10576 (Cayman Chemical). The cells were then treated with
LPS (0.5 g ml
1
) or poly I:C (50 g ml
1
) for 1 h. We harvested the cells for
total RNA and analyzed them by real-time PCR. We also resolved cell lysates
by SDS-PAGE and analyzed them by immunoblotting using the indicated
antibodies. Wild-type and Ikbke Tbk1 double-knockout and Ikbke or Tbk1
single-knockout MEFs were kindly provided by S. Akira. Cells were cultured
in DMEM with 10% FBS and serum starved in 0.5% FBS for 12 h before treat-
ment. TNF- treatment (50 ng ml
1
) was performed for 15 min at the end of
a 2-h pretreatment with amlexanox.
Oxygen consumption rate. The Michigan Metabolic Phenotyping Core and
Michigan Nutritional Obesity Research Center measured oxygen consumption
in cells and ex vivo tissue using an extracellular flux analyzer (model XF24-3,
Seahorse Bioscience).
Molecular modeling of amlexanox in the TBK1 ATP-binding site. We mod-
eled amlexanox into the known structure of the TBK1 catalytic domain (Protein
Data Bank ID codes 4EUT and 4EUU and ref. 26). The best fit of amlexanox in
the active site of TBK1 was obtained using AutoDock Vina.
Statistical analyses. Averaged values are presented as the mean s.e.m. When
comparing two groups, we determined statistical significance using Students
t test. When more than two groups were investigated, we first performed an
analysis of variance (ANOVA) to establish that not all groups were equal. After
a statistically significant ANOVA result, we performed between-group compari-
sons using the Tukey-Kramer post hoc analysis. ANOVA and Tukey-Kramer tests
were performed using Stata version 12.0.
Additional methods. Detailed methodology is described in the Supplementary
Methods.