mmc3

Report
Unsaturated Fatty Acids Stimulate Tumor Growth
through Stabilization of b-Catenin
Graphical Abstract
Highlights
d Unsaturated fatty acids (uFAs) inhibit b-catenin degradation
by inactivating FAF1
d b-catenin degradation can be independently inhibited by Wnt
signaling and uFAs
d uFAs promote growth of kidney cancers by stabilization of
b-catenin
Authors
Hyeonwoo Kim, Carlos Rodriguez-Navas,
Rahul K. Kollipara, ..., James Brugarolas,
Ralf Kittler, Jin Ye
Correspondence
jin.ye@utsouthwestern.edu
In Brief
Kim et al. demonstrate that excess
unsaturated fatty acids produced in
cancer cells stabilize b-catenin. This
mechanism, which is independent of
Wnt-mediated stabilization of b-catenin,
requires direct interaction of the fatty
acids with FAF1, a protein facilitating
degradation of b-catenin. This interaction
inactivates FAF1, thereby stabilizing
b-catenin.
Kim et al., 2015, Cell Reports 13, 495–503
October 20, 2015 ª2015 The Authors
http://dx.doi.org/10.1016/j.celrep.2015.09.010

Cell Reports
Report
through Stabilization of b-Catenin
Hyeonwoo Kim,1 Carlos Rodriguez-Navas,1 Rahul K. Kollipara,2 Payal Kapur,3,4 Ivan Pedrosa,5,6 James Brugarolas,3,7
Ralf Kittler,2 and Jin Ye1,*
1Department of Molecular Genetics
2Eugene McDermott Center for Human Growth and Development
3Kidney Cancer Program in Simmons Comprehensive Cancer Center
4Department of Pathology
5Advanced Imaging Research Center
6Department of Radiology
7Hematology-Oncology Division, Department of Internal Medicine
University of Texas Southwestern Medical Center, Dallas, TX 75390, USA
*Correspondence: jin.ye@utsouthwestern.edu
http://dx.doi.org/10.1016/j.celrep.2015.09.010
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
SUMMARY
Some cancer cells exhibit elevated levels of free fatty
acids (FAs) as well as high levels of b-catenin, a tran-
scriptional co-activator that promotes their growth.
Here, we link these two phenomena by showing
that unsaturated FAs inhibit degradation of b-cate-
nin. Unsaturated FAs bind to the UAS domain of
Fas-associated factor 1 (FAF1), a protein known to
bind b-catenin, accelerating its degradation. FA
binding disrupts the FAF1/b-catenin complex, pre-
venting proteasomal degradation of ubiquitinated
b-catenin. This mechanism for stabilization of b-cat-
enin differs from that of Wnt signaling, which blocks
ubiquitination of b-catenin. In clear cell renal cell car-
cinoma (ccRCC) cells, unsaturated FAs stimulated
cell proliferation through stabilization of b-catenin.
In tissues from biopsies of human ccRCC, elevated
levels of unsaturated FAs correlated with increased
levels of b-catenin. Thus, targeting FAF1 may be an
effective approach to treat cancers that exhibit
elevated FAs and b-catenin.
INTRODUCTION
Cancer cells alter their metabolism to provoke cell proliferation.
One metabolic alteration in cancers is the accumulation of free
fatty acids (FAs), which facilitate cell proliferation through a
mechanism that remains elusive (Nomura et al., 2010). To
pinpoint this mechanism, we studied FA-interacting proteins
that may link free FAs to oncogenic signaling pathways. We pre-
viously identiﬁed UAS domains, which contain $160 amino acid
residues, as the motifs that bind unsaturated, but not saturated,
FAs (Kim et al., 2013). This domain polymerizes upon its interac-
tion with unsaturated FAs (Kim et al., 2013). Mammalian cells ex-
press two homologous proteins that contain UAS domains (Kim
et al., 2013): Ubxd8, a protein maintaining cellular FA homeosta-
sis by stimulating degradation of Insig-1 (Ye and DeBose-Boyd,
2011), and Fas-associated factor 1 (FAF1), a protein that facili-
tates degradation of b-catenin (Zhang et al., 2011, 2012).
Ubxd8 senses the cellular content of unsaturated FAs to regu-
late degradation of Insig-1, a protein that inhibits transcription of
all genes required for synthesis of FAs (Kim and Ye, 2014; Ye and
DeBose-Boyd, 2011). Through their direct binding to the UAS
domain of Ubxd8, unsaturated FAs cause Ubxd8 to polymerize
and dissociate from Insig-1 so that ubiquitinated Insig-1 cannot
be delivered to proteasomes for degradation (Kim et al., 2013;
Lee et al., 2008, 2010). Like Ubxd8, unsaturated, but not satu-
rated, FAs trigger polymerization of FAF1 upon their interaction
with the UAS domain of the protein (Kim et al., 2013). The func-
tional signiﬁcance of the interaction between unsaturated FAs
and FAF1 remains unknown.
FAF1 has been reported to be required for degradation of
b-catenin (Zhang et al., 2011, 2012), a transcriptional co-acti-
vator that stimulates expression of genes driving cell prolifera-
tion (Anastas and Moon, 2013). In normal cells, the degradation
of b-catenin is regulated by Wnt signaling: b-catenin is constitu-
tively phosphorylated by the b-catenin destruction complex,
which marks b-catenin for ubiquitination followed by rapid pro-
teasomal degradation (Clevers and Nusse, 2012; Moon et al.,
2002); Wnt signaling inactivates the b-catenin destruction com-
plex, thereby inhibiting phosphorylation of b-catenin and conse-
quently ubiquitination and degradation of the protein (Clevers
and Nusse, 2012; Moon et al., 2002). Mutations that inactivate
proteins required for degradation of b-catenin lead to various
cancers as a result of aberrant accumulation of b-catenin
(Clevers, 2006). However, some cancer cells contain elevated
levels of b-catenin in the absence of these mutations (Barker
and Clevers, 2006). Based on our previous observations with
Ubxd8, we hypothesized that unsaturated FAs may bind to the
UAS domain of FAF1, leading to inactivation of FAF1 and conse-
quently stabilization of b-catenin.
In the current study, we report that unsaturated FAs indeed
inhibit degradation of b-catenin by inactivating FAF1. We
Cell Reports 13, 495–503, October 20, 2015 ª2015 The Authors 495

demonstrate the clinical significance of these findings by pro-
viding evidence that excess unsaturated FAs stabilize b-catenin
in clear cell renal cell carcinoma (ccRCC), which represents a
majority of kidney cancers (Li and Kaelin, 2011). These results
suggest that compounds blocking the interaction between
FAF1 and unsaturated FAs may be useful in treating cancers
whose proliferation is provoked by unsaturated FA-mediated
stabilization of b-catenin.
RESULTS
Unsaturated FAs Inhibit Degradation of b-Catenin
through Their Interaction with FAF1
We first used SRD-13A cells, a line of mutant CHO cells, to deter-
mine whether unsaturated FAs inhibit degradation of b-catenin.
These cells are auxotrophic for FAs, and consequently, their
content of FAs can be controlled easily by the amount of FAs
added into the culture medium (Rawson et al., 1999). We pre-
incubated the cells in FA-depleted medium and then supple-
mented the medium with various FAs. The effect of these FAs
on levels of b-catenin was determined by immunoblot analysis.
As a positive control, we treated the cells with Wnt3a to block
b-catenin degradation. b-catenin was barely detectable in cells
cultured in the absence of FAs (Figure 1A, lane 1). Addition of
palmitate (C16:0), a saturated FA, did not raise the amount of
b-catenin (Figure 1A, lane 2). However, oleate (C18:1) and other
unsaturated FAs markedly increased the levels of b-catenin (Fig-
ure 1A, lanes 3–8), reaching levels that were comparable to those
in cells treated with Wnt3a (Figure 1A, lane 9). In a pulse-chase
experiment, we demonstrated that oleate increased the amount
of b-catenin by inhibiting degradation of the protein (Figure 1B).
If unsaturated FAs stabilize b-catenin through inactivation of
FAF1, then knockdown of FAF1 should increase the amount of
b-catenin, regardless of the presence of unsaturated FAs. To
test this hypothesis, we transfected SRD-13A cells with a control
siRNA or siRNAs targeting FAF1. Oleate markedly raised the
amount of b-catenin in cells transfected with the control siRNA
(Figure 1C, lanes 1 and 2). Transfection of the cells with two
siRNAs targeting different regions of FAF1 reduced its mRNA
by 70%–80% (Figure S1A) and raised the amount of b-catenin
in cells cultured in the absence of FAs (Figure 1C, lanes 3 and
5). In the absence of FAF1, oleate did not further increase the
amount of b-catenin (Figure 1C, lanes 4 and 6).
Another way to examine the role of FAF1 in FA-regulated
degradation of FAF1 is to generate a mutant version of FAF1
that does not bind unsaturated FAs. Because the mutant
FAF1 may not be inactivated by unsaturated FAs, degradation
of b-catenin should not be inhibited by unsaturated FAs in cells
expressing the mutant FAF1. Our previous structural analysis of
the UAS domain of FAF1, which directly binds unsaturated FAs,
identified a surface patch that is highly enriched in positively
charged amino acid residues (Kim et al., 2013). We demon-
strated that these positively charged residues are conserved
in the UAS domain of Ubxd8 and that substitution of these res-
idues with glutamate abolished the interaction between Ubxd8
and unsaturated FAs (Kim et al., 2013). We thus expressed
β-catenin
Actin
1 2 3 4 5 6 7 8
Wnt3a
FA
Lane 9
16:0 18:1 20:518:2 18:3 20:4 22:6
A C
1 2 3 4 6
siRNA
Oleate
Lane
GFP FAF1(1) FAF1(2)
1 2 3 4 5 6 7
pTK-HSV-β-catenin
pCMV-Myc-FAF1
Oleate
Lane
WT Mutant
D
E
242
146
66
480
720
1048
kDa
β-catenin
Actin
β-catenin
Actin
FAF1
5
Time of chase(min)
Oleate
B
10
5
Relativeamountofβ-catenin
100
50
100 20
(%ofcontrol)
Figure 1. Unsaturated FAs Stabilize b-Cate-
nin through Inactivating FAF1
(A) SRD-13A cells were seeded at 3.5 3 105
/
60-mm dish on day 0. On day 1, cells were
depleted of FAs by incubation in medium A sup-
plemented with 5% delipidated fetal calf serum
(DFCS), 5 mg/ml cholesterol, and 1 mM sodium
mevalonate for 14 hr. On day 2, cells were
switched to medium A supplemented with 0.5%
DFCS, 5 mg/ml cholesterol, and 1 mM sodium
mevalonate in the absence or presence of the
indicated FAs (100 mM) or mouse Wnt3a (5 ng/ml).
Following incubation for 6 hr, cells were harvested.
Cell lysate was subjected to SDS-PAGE followed
by immunoblot analysis.
(B) SRD-13A cells seeded as described in (A) were
subjected to pulse-chase analysis as described in
Experimental Procedures. Results are reported as
means ± SE from three independent experiments.
(C)SRD-13A cells were seeded at1.5 3 105
/60-mm
dish on day 0. On day 1, cells were transfected with
20mM of the indicated siRNA.On days 3 and 4, cells
were depleted of FAs and then treated with oleate
as described in (A). Cell lysate was subjected to
SDS-PAGE followed by immunoblot analysis.
(D) Indicated purified proteins were incubated with
oleate added as stock solutions dissolved in
ethanol, subjected to BN-PAGE, and visualized
with Coomassie blue staining.
(E) SRD-13A cells were seeded as described in
(A). On day 1, cells were transfected with 1 mg/dish
pTK-HSV-b-catenin, 0.3 mg/dish pCMV-Myc-FAF1 (WT), or 0.6 mg/dish pCMV-Myc-FAF1 (mutant). Following incubation for 8 hr, cells were depleted of FAs and
then treated with or without oleate on day 2 as described in (A). Cell lysate was subjected to SDS-PAGE followed by immunoblot analysis.
496 Cell Reports 13, 495–503, October 20, 2015 ª2015 The Authors

and purified a mutant variant of the UAS domain of FAF1 (FAF1
325–491) with the corresponding amino acid substitutions
(K368E, R370E, R372E, K373E, K457E, and R458E). The inter-
action between oleate and the FAF1 UAS domain was detected
by oleate-induced polymerization of the protein as revealed by
blue native PAGE (BN-PAGE) (Kim et al., 2013; Lee et al.,
2010). Oleate induced the polymerization of the wild-type UAS
domain (Figure 1D, lanes 1–4), but not the mutant UAS domain
(Figure 1D, lanes 5–8). This observation indicates that the
mutant UAS domain of FAF1 does not interact with unsaturated
FAs. Next, we transfected SRD-13A cells with plasmids encod-
ing b-catenin and wild-type FAF1 or mutant FAF1 with the mu-
tations in the UAS domain described above. When b-catenin
was overexpressed by transfection, the protein was present
even in the absence of FAs and there was no further stabilization
by oleate (Figure 1E, lanes 2 and 3). Cotransfection of wild-type
FAF1 reduced the amount of b-catenin in FA-depleted cells, and
this reduction was blocked by oleate (Figure 1E, lanes 4 and 5).
This result suggests that FAF1 is a limiting factor for degradation
of b-catenin when it is overexpressed. Cotransfection with a
plasmid encoding mutant FAF1 reduced the amount of b-cate-
nin, but this reduction was no longer reversed by addition of
oleate (Figure 1E, lanes 6 and 7). Collectively, these findings
suggest that FAF1 accelerates b-catenin degradation and that
the interaction between unsaturated FAs and the UAS domain
of FAF1 is required for these FAs to inhibit degradation of
b-catenin.
The effect of unsaturated FAs on degradation of b-catenin is
not restricted to SRD-13A cells. Unsaturated, but not saturated
FAs, also stabilized the protein in HEK293 cells that are not
auxotrophic for FAs (Figure S1B). A difference between this
experiment and that performed in SRD-13A cells is that, in addi-
tion to incubating cells in FA-free medium, we treated HEK293
cells with A939572, an inhibitor of stearoyl-CoA desaturase-1
(SCD1), which catalyzes the rate-limiting step in biosynthesis
of unsaturated FAs (Paton and Ntambi, 2010), to deplete the cells
of unsaturated FAs.
Unsaturated FAs Stabilize b-Catenin through a
Mechanism Different from Wnt Signaling
Because HEK293 cells have been used to study Wnt-regulated
degradation of b-catenin (Li et al., 2012), we used these cells
to compare unsaturated FA versus Wnt-mediated stabilization
of b-catenin. We first determined the effect of unsaturated FAs
on phosphorylation of b-catenin. Both oleate and Wnt3a in-
creased the amount of b-catenin in HEK293 cells cultured in
FA-depleted medium (Figure 2A, panel 4). To analyze phosphor-
ylation of b-catenin, we treated the cells with a proteasome in-
hibitor MG132 to prevent the degradation of the phosphorylated
protein (Figure 2A, panel 2). Whereas Wnt3a inhibited phosphor-
ylation of b-catenin (Figure 2A, panel 1, lane 3), oleate increased
the amount of phosphorylated b-catenin (Figure 2A, panel 1,
lane 2). To determine the impact of these treatments on ubiquiti-
nation of b-catenin, we immunoprecipitated b-catenin from ly-
sates of cells treated with MG132 and measured ubiquitinated
b-catenin as detected by high-molecular-weight smears in
immunoblot analysis of the immunoprecipitates with anti-b-cat-
enin and anti-ubiquitin. Ubiquitinated b-catenin was observed in
untreated cells (Figure 2B, lane 1). In contrast to Wnt3a, which
reduced the level of ubiquitinated b-catenin (Figure 2B, lane 3),
oleate increased the amount of ubiquitinated b-catenin (Fig-
ure 2B, lane 2). The results of Wnt treatment are consistent
with previous observations that Wnt inhibits phosphorylation
and ubiquitination of b-catenin. In contrast, unsaturated FAs pre-
vent the degradation of ubiquitinated b-catenin.
Next, we investigated whether unsaturated FAs disrupted
the FAF1/b-catenin complex that is known to be required for
degradation of b-catenin (Zhang et al., 2012). For this purpose,
we treated HEK293 cells with MG132 to prevent proteasomal
degradation of b-catenin. We immunoprecipitated FAF1 and
IB : Ubiquitin
IB : β-catenin
β-catenin
(-) MG132
(+) MG132
A
β-catenin
FAF1
Input
Pellet
Sup.
FAF1
18:1 20:4
B
p-β-catenin
(S33/37/T41)
β-catenin
Actin
Actin
Wnt3a
CIP : β-catenin
β-catenin
FAF1
β-catenin
FAF1
C
1 2 3
Oleate
Lane
Wnt3a
1 2 3
Oleate
Lane
IP Antibody
1 2 3
FA
Lane 4
Ubiquitinated
β-catenin
Ubiquitinated
β-catenin
Panel
1
2
3
4
5
Panel
1
2
3
4
5
6
Figure 2. Unsaturated FAs Inhibit Degradation of b-Catenin at a Post-ubiquitination Step
HEK293 cells were seeded at 4.0 3 105
/60-mm dish on day 0. On day 2, cells were incubated in medium B supplemented with 10% DFCS and 1 mM A939572 for
16 hr. On day 3, cells were switched to serum-free medium B supplemented with 1 mM A939572 and treated with 100 mM oleate, 40 ng/ml human Wnt3a, or 10 mM
MG132 as indicated for 6 hr.
(A) Cell lysate was subjected to SDS-PAGE followed by immunoblot analysis.
(B) b-catenin was immunoprecipitated from lysate of the cells treated with MG132, and the immunoprecipitates were subjected to immunoblot analysis.
(C) Cytosolic fractions of the cells treated with MG132 were subjected to immunoprecipitation with a control antibody (C) or anti-FAF1. Aliquots of the cytosol
fraction before the immunoprecipitation (input) and the pellet and supernatant fractions of the immunoprecipitation were loaded at a ratio of 1:10:1 on SDS-PAGE
followed by immunoblot analysis.

determined the amount of b-catenin that was co-precipitated.
b-catenin was co-immunoprecipitated with FAF1 in FA-depleted
cells (Figure 2C, panel 4, lane 2), but not in cells treated with
oleate or arachidonate (C20:4), another unsaturated FA (Fig-
ure 2C, panel 4, lanes 3 and 4).
Stabilization of b-Catenin by Unsaturated FAs Is
Required for Proliferation of ccRCC Cells
ccRCC is characterized by excess lipid accumulation, owing to
increased synthesis of FAs and cholesterol (Drabkin and Gem-
mill, 2010; Li and Kaelin, 2011). The increased synthesis of unsat-
urated FAs appears to be important for proliferation of ccRCC
cells, as growth of mouse xenograft tumors was inhibited by
treatment with A939572, an inhibitor of SCD1 required for syn-
thesis of unsaturated FAs (von Roemeling et al., 2013). A recent
study also reported increased b-catenin levels as a predictor of
poor clinical outcomes in ccRCC patients (Krabbe et al., 2014).
We thus hypothesized that accumulation of excess unsaturated
FAs in ccRCC tumor cells may promote their proliferation
through stabilization of b-catenin.
To test this hypothesis, we studied the impact of unsaturated
FAs on b-catenin levels in the ccRCC cell line SW156. These cells
contain much more triglyceride, cholesterol, and free unsatu-
rated FAs than HEK293 cells (Figures S2A–S2C). Unlike SRD-
13A cells, in which b-catenin is not associated with membranes,
in SW156 cells, a significant fraction of b-catenin is membrane-
associated (Figure S2D). This pool of b-catenin is known to be
required for cell adhesion, but not for stimulation of cell prolifer-
ation, and it is not subjected to rapid degradation (Fagotto,
2013). Therefore, we assessed the effect of FAs on stabilization
of cytosolic b-catenin. As observed in SRD-13A cells, we found
that addition of unsaturated, but not saturated, FAs stabilized
cytosolic b-catenin in SW156 cells (Figure 3A). We then deter-
mined the effect of unsaturated FAs on expression of cyclin
D1, a target gene of b-catenin that drives cell proliferation (Shtut-
man et al., 1999; Tetsu and McCormick, 1999). We found that un-
saturated, but not saturated, FAs caused a marked increase in
cyclin D1 mRNA expression (Figure 3B). To confirm that the
increased expression of cyclin D1 was caused by elevated b-cat-
enin, we treated the cells with NCX4040, which inhibits the tran-
scriptional co-activating activity of b-catenin (Nath et al., 2003).
Treatment with NCX4040 completely abolished oleate-induced
expression of cyclin D1 mRNA (Figure 3C).
Next, we analyzed the effect of unsaturated FA-mediated sta-
bilization of b-catenin on cell proliferation. We incubated SW156
cells in FA-depleted medium in the absence or presence of the
SCD1 inhibitor A939572 and then treated the cells with various
concentrations of oleate. In the absence of exogenous oleate,
a small amount of b-catenin was detected in cells incubated in
the absence of A939572 (Figure 3D, lane 1). The protein disap-
peared upon treatment with A939572 (Figure 3D, lane 2). In par-
allel with the amount of b-catenin, the cells proliferated in the
absence, but not the presence, of A939572 (Figure 3E, oleate
at 0 mM). Addition of exogenous oleate increased the amount
of b-catenin in cells treated with or without A939572 (Figure 3D,
lanes 3–12). At 20 mM, the added oleate raised the amount of
b-catenin in cells treated with A939572 to the same level as
that in cells incubated in the absence of the inhibitor (Figure 3D,
lanes 11 and 12). In parallel with the amount of b-catenin, exog-
enous oleate increased the proliferation rate of the cells treated
with or without A939572 (Figure 3E). At 20 mM of added oleate,
cell proliferation rate was the same in the presence or absence
of A939572 (Figure 3E). To make a more-direct comparison,
we used mass spectroscopy to measure the amount of free un-
saturated FAs in cells subjected to all of these treatments. We
plotted the amount of b-catenin and the rate of cell growth
against the content of unsaturated FAs. The amount of b-catenin
(red) and the rate of cell growth (blue) were positively correlated
with the intracellular content of unsaturated FAs, and the linear
fitting of both data sets was almost superimposable (Figure 3F).
These results strongly suggest that unsaturated FA-mediated
stabilization of b-catenin is responsible for cell proliferation.
In addition to SW156 cells, we observed accumulation of lipids
including unsaturated FAs in 786-O cells, another line of ccRCC
cells (Figures S2A–S2C). In these cells, unsaturated, but not
saturated, FAs also stabilized b-catenin (Figure S2E) in a FAF1-
dependent manner, as b-catenin was stabilized regardless of
the presence of the FAs in the cells in which FAF1 was knocked
down by the transfected siRNA (Figure S2F). Similar to SW156
cells, the stabilization of b-catenin in 786-O cells was correlated
to cell proliferation (Figure S2G).
Excess Unsaturated FAs Increase Oncogenic Activity of
b-Catenin in Specimens of ccRCC Tumors
To further substantiate the clinical relevance of our findings,
we analyzed the correlation between unsaturated FAs and the
intracellular distribution of b-catenin in biopsies exercised
from patients with ccRCC. We determined the subcellular local-
ization of b-catenin in 25 tumors by immunohistochemistry.
Whereas most tumor cells showed b-catenin staining on plasma
membranes (Figure 4A), a few of them, for example tumor no.
22383, had intracellular staining of the protein (Figure 4B). The
strong intracellular staining of b-catenin (>60%) was only
observed in grade 4 ccRCC cells, which constitute the most
aggressive form of the tumors (Figure 4C, red bars; no. 21886,
no. 22383, no. 24152, and no. 21470). We then used mass spec-
troscopy to measure free unsaturated FAs in all of the 25 tumors
and found that only these four tumors accumulated higher levels
of the FAs in comparison to their benign controls (Figure 4C, blue
bars; no. 21886, no. 22383, no. 24152, and no. 21470). Other tu-
mors had no or low intracellular staining of b-catenin (<35%; Fig-
ure 4C, red bars), and most of them contained lower levels of
unsaturated FAs compared to their benign controls (Figure 4C,
blue bars). The probability of the correlation between increased
intracellular distribution of b-catenin and increased levels of un-
saturated FAs in tumor cells occurring by chance is 10À5
as
determined by Pearson analysis. These results suggest that, in
ccRCC, increased levels of unsaturated FAs stabilize intracel-
lular b-catenin.
We then performed gene expression analyses using RNA-seq
data from 532 human ccRCC samples generated by The Cancer
Genome Atlas project. Because SCD1 catalyzes the rate-limiting
step in the biosynthesis of unsaturated FAs, tumors with higher
expression of the gene are expected to produce more unsatu-
rated FAs. Thus, if unsaturated FAs stabilize b-catenin, tumors
with higher expression of SCD1 should also express higher

levels of b-catenin target genes. We identiﬁed 1,024 genes as the
top 5% of genes whose expression is positively correlated with
SCD1 expression (Table S1). When we analyzed these genes
for enrichment in functional annotation categories, we found
enrichment in biological processes that are relevant to cell prolif-
eration and cancer signaling (Figure S3). When we analyzed
transcription factor motifs in the promoter regions of these
genes, we found enrichment for a LEF1 motif, which is typical
β-catenin
Actin
A
D
β-catenin
Actin
18:0
0
100
200
300
5 10 2015
PercentageofCellGrowth
Oleate (μM)
0
A939572
B
C
0
2
4
6
8
10
16:0 18:0 18:1 18:2 18:3
Relativeamountof
1
3
2
Oleate :
4
NCX4040
RelativeamountofcyclinD1mRNA
1 2 3 4 5 6 7 8
FA
Lane 9
16:0 18:1 20:518:2 18:3 20:4 22:6
A939572
Oleate (μM)
Lane 1 2 3 4 5 6 7 8 9 10 11 12
20105210
E
F
Unsaturated Free FAs (ng/mg protein)
100
200
300
0 Relativeamountofβ-cateninprotein
100 200 3000
PercentageofCellGrowth
1
2
3
0
FA :
cyclinD1mRNA
R = 0.8661 R = 0.7787
2 2
P = 1.13e P = 1.45e
-5 -4
1
0
Figure 3. Excess Unsaturated FAs Promote Proliferation of SW156 through Stabilization of b-Catenin
(A) SW156 cells were seeded at 3.5 3 105
/60-mm dish on day 0. On day 1, cells were depleted of FAs by incubation in medium C supplemented with 10% DFCS
and 1 mM A939572 for 24 hr. On day 2, cells were switched to serum-free medium C supplemented with 1 mM A939572 and treated with 100 mM of the indicated
FA for 6 hr. Cytosolic fractions were subjected to SDS-PAGE followed by immunoblot analysis.
(B and C) SW156 cells were seeded on day 0 and depleted of FAs on day 1 as described in (A). On day 2, cells were switched to serum-free medium C sup-
plemented with 1 mM A939572 and treated with 100 mM of the indicated FA in the absence or presence of 10 mM NCX4040 for 9 hr. The amount of cyclin D1 mRNA
was determined by qRT-PCR with the value in untreated FA-depleted cells set at 1. Results are reported as means ± SE from three independent experiments.
(D) SW156 cells were seeded at 1.0 3 105
/60-mm dish on day 0. On day 1, cells were switched to medium C supplemented with 10% DFCS and the indicated
concentration of oleate in the absence or presence of 1 mM A939572. On day 3, after incubation for 60 hr, cytosolic fractions of the cells were subjected to
immunoblot analysis.
(E) SW156 cells were seeded at 800/well in a 96-well plate on day 0. On day 1, cell number was determined in some wells of the cells. The rest of the cells were
treated the same as described in (D). On day 3, after incubation for 60 hr, the cell number in each well was determined, and the percentage increase in the cell
number compared to that in day 1 was presented. Results are reported as means ± SE from three independent experiments.
(F) Densitometry quantiﬁcation of b-catenin protein in (D) (red) and relative cell growth observed in (E) (blue) were plotted against intracellular concentration of
unsaturated FAs measured by mass spectroscopy in all of the treatment conditions in (D) and (E).

for the gene-specific transcription factors that use b-catenin as a
co-activator (Arce et al., 2006; Schuijers et al., 2014). The LEF1
motif was the second most significantly enriched motif in pro-
moters for genes whose expression is positively correlated
with SCD1 expression (adjusted p = 4.3eÀ22). Out of the 1,939
human genes with this motif in their promoter regions, 126 genes
were found in the set of 1,024 genes whose expression was posi-
tively correlated with SCD1 expression (Figure 4D; Table S1).
This number is 2.74 times higher than expected from random
overlap between the two gene sets. These analyses suggest
that the potential target genes of b-catenin are overexpressed
in ccRCC cells that express high levels of SCD1 mRNA.
DISCUSSION
The current study supports a model shown in Figure 5, illus-
trating how unsaturated FAs and Wnt independently inhibit
proteasomal degradation of b-catenin. Previous studies have
demonstrated that b-catenin is constitutively phosphorylated
by the b-catenin destruction complex, which marks b-catenin
for ubiquitination by a b-Trcp-containing E3 ubiquitin ligase com-
plex (Clevers and Nusse, 2012; Moon et al., 2002). The ubiquiti-
nated b-catenin may then be targeted by FAF1 to proteasomes
for degradation (Figure 5, middle panel). Activation of Wnt
signaling recruits the destruction complex to plasma mem-
branes, thereby preventing phosphorylation and ubiquitination
of b-catenin and resulting in stabilization of the protein (Clevers
and Nusse, 2012; Moon et al., 2002; Figure 5, left panel). In
contrast to Wnt signaling, unsaturated FAs do not affect phos-
phorylation or ubiquitination of b-catenin. Instead, the FAs
disrupt the FAF1/b-catenin complex by triggering polymerization
of FAF1. Consequently, ubiquitinated b-catenin is not targeted to
proteasomes for degradation, thereby stabilizing b-catenin (Fig-
ure 5, right panel).
Exactly how FAF1 targets ubiquitinated b-catenin to protea-
somes for degradation remains unclear. A previous study re-
ported that recognition of ubiquitinated Insig-1 by proteasomes
required recruitment of p97 to Insig-1, a reaction mediated by
Ubxd8 that is a homolog of FAF1 (Ikeda et al., 2009). Inasmuch
as FAF1 also binds p97 (Ewens et al., 2014), the mechanism
B
2238320μm
A
20μm23780
C
RelativeamountofFreeUnsaturatedFAs
(Tumor/Benign)
20
40
60
80
100
0
2
1
3
4
0
RelativeIntracellularstainingofβ-catenin
22190
22139
22216
22505
21886
21913
21940
21972
24117
24288
24316
23769
23698
23569
23334
22881
22591
22497
23780
23729
22383
24152
21470
24329
18047
Grade 2 Grade 3 Grade 4
** **
*
**
1939 1024
126
Genes that contain
β-catenin binding sites
Genes whose expression
with SCD1 expressionin their promoters
is positively correlated
Ratio of enrichment : 2.74
Raw P value : 1.4e-24
Adjusted P value : 4.3e
-22
D
Figure 4. Increased Levels of Unsaturated FAs Are Correlated to Elevated Levels of Intracellular b-Catenin in ccRCC Patient Specimens
(A and B) Immunohistochemistry staining of b-catenin in the indicated ccRCC patient specimen was performed as described in Experimental Procedures.
(C) The relative amount of unsaturated FAs in tumors was measured through mass spectroscopy analysis, with the value in their benign controls set at 1 (blue
bars). The relative intracellular staining of b-catenin in ccRCC patient specimens was determined as described in Experimental Procedures (red bars). The
pathological grade of the tumors is indicated. The sample numbers for the tumors with higher levels of intracellular staining of b-catenin and unsaturated FAs are
highlighted in brown. The results are reported as means ± SE from three independent measurements. Paired Student’s t test was performed to determine the
statistical significance of the increase in the amount of unsaturated fatty acids in the highlighted tumors compared to their benign controls: *p = 0.02; **p < 0.01.
(D) Venn diagram displaying overlap in putative target genes of b-catenin and genes whose expression is positively correlated with SCD1 expression. Adjusted
p = 4.3 3 10À22
.

through which FAF1 facilitates degradation of ubiquitinated
b-catenin may be similar to that through which Ubxd8 stimulates
degradation of ubiquitinated Insig-1.
The strongest evidence indicating that unsaturated FAs inhibit
degradation of b-catenin through a mechanism different from
Wnt signaling comes from the observations that, in contrast
to activation of Wnt signaling, these FAs do not affect phosphor-
ylation or ubiquitination of b-catenin. This mechanism is different
from an earlier report showing the correlation between FA
synthesis and stabilization of b-catenin: in that study, FA syn-
thesis was shown to be required for palmitoylation of Wnt
(Fiorentino et al., 2008), a post-translational modiﬁcation of
Wnt critical for its signaling function. Instead of inhibiting ubiqui-
tination of b-catenin, unsaturated FAs prevent degradation of
ubiquitinated b-catenin. In the absence of proteasomal degrada-
tion, the polyubiquitin chains on b-catenin may be removed from
the protein by deubiquitinating enzymes. The presence of the
strong deubiquitination activity in mammalian cells may explain
why the effect of unsaturated FAs on stabilization of nonubiqui-
tinated b-catenin is more pronounced than that of ubiquitinated
b-catenin.
The current study establishes that accumulation of unsatu-
rated FAs can act as an oncogenic mechanism to increase
b-catenin levels, which is different from the well-established
mechanism caused by genetic inactivation of proteins in Wnt
signaling. We determined that accumulation of excess unsatu-
rated FAs was responsible for stabilization of b-catenin in
some ccRCC tumors. We observed that aberrant stabilization
of b-catenin in several grade 4 ccRCC tumors, the most-aggres-
sive form of the cancer, was correlated to their increased levels
of unsaturated FAs. Through bioinformatics analysis, we showed
that the potential target genes of b-catenin were overexpressed
in ccRCC cells that express high levels of SCD1 mRNA, which
encodes the enzyme catalyzing the rate-liming step in synthesis
of unsaturated FAs. These genes include Cyclin D1, met proto-
oncogene, and matrix metallopeptidase 14 that have been previ-
ously determined to be the target gene of b-catenin (Herbst et al.,
2014; Shtutman et al., 1999; Tetsu and McCormick, 1999).
Among these genes, we showed that Cyclin D1 was activated
by b-catenin in ccRCC cells. In addition to ccRCC, it will be inter-
esting to determine whether accumulation of unsaturated FAs is
also responsible for aberrant stabilization of b-catenin in other
cancers that do not contain genetic mutations affecting the
Wnt pathway.
Our ﬁnding provides mechanistic insights into the observa-
tions that accumulation of free FAs facilitates development and
progression of certain cancers (Nomura et al., 2010, 2011).
Similar to ccRCC, these cancer cells may acquire FAs through
enhancing their endogenous synthesis. They may also obtain
FAs from plasma, which may explain why obesity, a condition
associated with increased amount of free FAs in circulation, is
a risk factor for cancer development (Park et al., 2014).
The current study further demonstrates the critical roles
played by the UAS domain in transmitting signals elicited by un-
saturated FAs. We have shown previously that binding of unsat-
urated FAs to the UAS domain of Ubxd8 is crucial for feedback
inhibition of FA synthesis through inhibiting degradation of In-
sig-1 (Ye and DeBose-Boyd, 2011). In the current study, we
show that binding of unsaturated FAs to the UAS domain of
FAF1 is required for these FAs to inhibit degradation of b-catenin.
These results suggest that compounds blocking the interaction
between unsaturated FAs and the UAS domain of FAF1 should
destabilize b-catenin. It will be interesting to determine whether
the UAS domain of FAF1 could be targeted by drugs to treat can-
cers whose proliferation is dependent on unsaturated FA-medi-
ated stabilization of b-catenin.
β-catenin
FzdPlasma Membrane Fzd LRP
Wnt
Nucleus
LRPFzd
β-catenin
destruction
complex
P
Ub
Ub
Ub
Ub
Ub
Proteasomal
degradation
P
E3 ligase
Complex
β-catenin
destruction
complex
FAF1
E3 ligase
Complex
FzdPlasma Membrane Fzd LRP
β-catenin
destruction
complex
P
Ub
Ub
Ub
Ub
Ub
P
E3 ligase
Complex
UnsaturatedWnt
FAF1
FAF1FAF1FAF1FAF1FAF1FAF1
β-catenin
β-catenin
β-catenin
β-catenin β-catenin
β-catenin
β-catenin
β-catenin
β-catenin
Proteasome
Proteasome
Proteasome
Nucleus
FAs
DUB?
Figure 5. Model Illustrating that Unsaturated FAs and Wnt Stabilize b-Catenin via Different Mechanisms
In the absence of Wnt and unsaturated fatty acids, b-catenin is phosphorylated by the b-catenin destruction complex, a reaction marking the protein for
ubiquitination followed by rapid degradation by proteasomes. Activation of Wnt signaling stabilizes b-catenin by inhibiting phosphorylation and ubiquitination of
the protein through recruiting the destruction complex to plasma membranes. In contrast to Wnt, unsaturated FAs do not affect phosphorylation or ubiquitination
of b-catenin. Instead, the FAs trigger polymerization of FAF1, causing dissociation of the protein from b-catenin. As a result, b-catenin is stabilized as ubiq-
uitinated b-catenin is not delivered to proteasomes for degradation. In the absence of proteasomal degradation, polyubiquitin chains on b-catenin may be
cleaved off by reactions catalyzed by deubiquitinating enzymes (DUB), allowing b-catenin to activate its target genes in nucleus.

EXPERIMENTAL PROCEDURES
Measurement of Lipids
Free FAs from approximately 0.5 million cells or 5 mg of tumors were quantified
using gas chromatography-electron capture negative ionization-mass spec-
trometry as previously described (Morselli et al., 2014; Quehenberger et al.,
2011). Triglycerides and free cholesterol were isolated from lipid extracts
using Biotage isolute NH2 SPE cartridges according to the instruction of
the manufacturer. The amount of triglycerides was determined by the
amount of free FAs released from triglycerides following saponification as
previously described (Quehenberger et al., 2011). The amount of cholesterol
was determined by Infinity Cholesterol according to the instruction of the
manufacturer.
Immunohistochemistry
Immunohistochemistry staining for b-catenin in ccRCC tumors was performed
exactly as previously described (Krabbe et al., 2014). Intensity of intracellular
(Ii) and membrane staining (Im) of b-catenin as well as percentage of the
cells showing intracellular (Pi) and membrane staining (Pm) of the protein
was determined. Relative intracellular staining of b-catenin was calculated
as Ii 3 Pi/(Ii 3 Pi + Im 3 Pm) 3 100%.
SUPPLEMENTAL INFORMATION
Supplemental information includes Supplemental Experimental Procedures,
three figures, and one table and can be found with this article online at
http://dx.doi.org/10.1016/j.celrep.2015.09.010.
AUTHOR CONTRIBUTIONS
J.Y. and H.K. designed and orchestrated the entire study and wrote the manu-
script. H.K. and C.R.-N. performed the experiments. R.K.K. and R.K. per-
formed bioinformatics and statistical analysis. P.K. performed pathological
analysis. J.B. and I.P. were involved in design of the study.
ACKNOWLEDGMENTS
We thank Drs. Joseph Goldstein and Michael Brown for constant advices and
critical evaluations of our manuscript; Lisa Beatty, Ijeoma Dukes, Muleya Ka-
paale, and Hue Dao for help with tissue culture; Jeff Cormier for qRT-PCR; and
Saada Abdalla for technical assistance. This work was supported by grants
from the NIH (HL-20948 and CA-154475) and Welch Foundation (I-1832).
C.R.-N. is supported by Clayton Foundation for Research.
Received: January 15, 2015
Revised: August 17, 2015
Accepted: September 2, 2015
Published: October 8, 2015
REFERENCES
Anastas, J.N., and Moon, R.T. (2013). WNT signalling pathways as therapeutic
targets in cancer. Nat. Rev. Cancer 13, 11–26.
Arce, L., Yokoyama, N.N., and Waterman, M.L. (2006). Diversity of LEF/TCF
action in development and disease. Oncogene 25, 7492–7504.
Barker, N., and Clevers, H. (2006). Mining the Wnt pathway for cancer thera-
peutics. Nat. Rev. Drug Discov. 5, 997–1014.
Clevers, H. (2006). Wnt/b-catenin signaling in development and disease. Cell
127, 469–480.
Clevers, H., and Nusse, R. (2012). Wnt/b-catenin signaling and disease. Cell
149, 1192–1205.
Drabkin, H.A., and Gemmill, R.M. (2010). Obesity, cholesterol, and clear-cell
renal cell carcinoma (RCC). Adv. Cancer Res. 107, 39–56.
Ewens, C.A., Panico, S., Kloppsteck, P., McKeown, C., Ebong, I.-O., Robin-
son, C., Zhang, X., and Freemont, P.S. (2014). The p97-FAF1 protein complex
reveals a common mode of p97 adaptor binding. J. Biol. Chem. 289, 12077–
12084.
Fagotto, F. (2013). Looking beyond the Wnt pathway for the deep nature of
b-catenin. EMBO Rep. 14, 422–433.
Fiorentino, M., Zadra, G., Palescandolo, E., Fedele, G., Bailey, D., Fiore, C.,
Nguyen, P.L., Migita, T., Zamponi, R., Di Vizio, D., et al. (2008). Overexpression
of fatty acid synthase is associated with palmitoylation of Wnt1 and cyto-
plasmic stabilization of b-catenin in prostate cancer. Lab. Invest. 88, 1340–
1348.
Herbst, A., Jurinovic, V., Krebs, S., Thieme, S.E., Blum, H., Go¨ ke, B., and
Kolligs, F.T. (2014). Comprehensive analysis of b-catenin target genes in
colorectal carcinoma cell lines with deregulated Wnt/b-catenin signaling.
BMC Genomics 15, 74.
Ikeda, Y., Demartino, G.N., Brown, M.S., Lee, J.N., Goldstein, J.L., and Ye, J.
(2009). Regulated endoplasmic reticulum-associated degradation of a poly-
topic protein: p97 recruits proteasomes to Insig-1 before extraction from
membranes. J. Biol. Chem. 284, 34889–34900.
Kim, H., and Ye, J. (2014). Cellular responses to excess fatty acids: focus on
ubiquitin regulatory X domain-containing protein 8. Curr. Opin. Lipidol. 25,
118–124.
Kim, H., Zhang, H., Meng, D., Russell, G., Lee, J.N., and Ye, J. (2013). UAS
domain of Ubxd8 and FAF1 polymerizes upon interaction with long-chain un-
saturated fatty acids. J. Lipid Res. 54, 2144–2152.
Krabbe, L.M., Westerman, M.E., Bagrodia, A., Gayed, B.A., Darwish, O.M.,
Haddad, A.Q., Khalil, D., Kapur, P., Sagalowsky, A.I., Lotan, Y., et al. (2014).
Dysregulation of b-catenin is an independent predictor of oncologic outcomes
in patients with clear cell renal cell carcinoma. J. Urol. 191, 1671–1677.
Lee, J.N., Zhang, X., Feramisco, J.D., Gong, Y., and Ye, J. (2008). Unsaturated
fatty acids inhibit proteasomal degradation of Insig-1 at a postubiquitination
step. J. Biol. Chem. 283, 33772–33783.
Lee, J.N., Kim, H., Yao, H., Chen, Y., Weng, K., and Ye, J. (2010). Identification
of Ubxd8 protein as a sensor for unsaturated fatty acids and regulator of tri-
glyceride synthesis. Proc. Natl. Acad. Sci. USA 107, 21424–21429.
Li, L., and Kaelin, W.G., Jr. (2011). New insights into the biology of renal cell
carcinoma. Hematol. Oncol. Clin. North Am. 25, 667–686.
Li, V.S., Ng, S.S., Boersema, P.J., Low, T.Y., Karthaus, W.R., Gerlach, J.P.,
Mohammed, S., Heck, A.J., Maurice, M.M., Mahmoudi, T., and Clevers, H.
(2012). Wnt signaling through inhibition of b-catenin degradation in an intact
Axin1 complex. Cell 149, 1245–1256.
Moon, R.T., Bowerman, B., Boutros, M., and Perrimon, N. (2002). The promise
and perils of Wnt signaling through b-catenin. Science 296, 1644–1646.
Morselli, E., Fuente-Martin, E., Finan, B., Kim, M., Frank, A., Garcia-Caceres,
C., Navas, C.R., Gordillo, R., Neinast, M., Kalainayakan, S.P., et al. (2014). Hy-
pothalamic PGC-1a protects against high-fat diet exposure by regulating ERa.
Cell Rep. 9, 633–645.
Nath, N., Kashfi, K., Chen, J., and Rigas, B. (2003). Nitric oxide-donating
aspirin inhibits b-catenin/T cell factor (TCF) signaling in SW480 colon cancer
cells by disrupting the nuclear b-catenin-TCF association. Proc. Natl. Acad.
Sci. USA 100, 12584–12589.
Nomura, D.K., Long, J.Z., Niessen, S., Hoover, H.S., Ng, S.-W., and Cravatt,
B.F. (2010). Monoacylglycerol lipase regulates a fatty acid network that pro-
motes cancer pathogenesis. Cell 140, 49–61.
Nomura, D.K., Lombardi, D.P., Chang, J.W., Niessen, S., Ward, A.M., Long,
J.Z., Hoover, H.H., and Cravatt, B.F. (2011). Monoacylglycerol lipase exerts
dual control over endocannabinoid and fatty acid pathways to support pros-
tate cancer. Chem. Biol. 18, 846–856.
Park, J., Morley, T.S., Kim, M., Clegg, D.J., and Scherer, P.E. (2014). Obesity
and cancer–mechanisms underlying tumour progression and recurrence. Nat.
Rev. Endocrinol. 10, 455–465.
Paton, C.M., and Ntambi, J.M. (2010). Loss of stearoyl-CoA desaturase activ-
ity leads to free cholesterol synthesis through increased Xbp-1 splicing. Am. J.
Physiol. Endocrinol. Metab. 299, E1066–E1075.

Quehenberger, O., Armando, A.M., and Dennis, E.A. (2011). High sensitivity
quantitative lipidomics analysis of fatty acids in biological samples by
gas chromatography-mass spectrometry. Biochim. Biophys. Acta 1811,
648–656.
Rawson, R.B., DeBose-Boyd, R., Goldstein, J.L., and Brown, M.S. (1999).
Failure to cleave sterol regulatory element-binding proteins (SREBPs)
causes cholesterol auxotrophy in Chinese hamster ovary cells with genetic
absence of SREBP cleavage-activating protein. J. Biol. Chem. 274, 28549–
28556.
Schuijers, J., Mokry, M., Hatzis, P., Cuppen, E., and Clevers, H. (2014). Wnt-
induced transcriptional activation is exclusively mediated by TCF/LEF.
EMBO J. 33, 146–156.
Shtutman, M., Zhurinsky, J., Simcha, I., Albanese, C., D’Amico, M., Pestell, R.,
and Ben-Ze’ev, A. (1999). The cyclin D1 gene is a target of the b-catenin/LEF-1
pathway. Proc. Natl. Acad. Sci. USA 96, 5522–5527.
Tetsu, O., and McCormick, F. (1999). b-catenin regulates expression of cyclin
D1 in colon carcinoma cells. Nature 398, 422–426.
von Roemeling, C.A., Marlow, L.A., Wei, J.J., Cooper, S.J., Caulﬁeld, T.R., Wu,
K., Tan, W.W., Tun, H.W., and Copland, J.A. (2013). Stearoyl-CoA desaturase
1 is a novel molecular therapeutic target for clear cell renal cell carcinoma. Clin.
Cancer Res. 19, 2368–2380.
Ye, J., and DeBose-Boyd, R.A. (2011). Regulation of cholesterol and fatty acid
synthesis. Cold Spring Harb. Perspect. Biol. 3, a004754.
Zhang, L., Zhou, F., van Laar, T., Zhang, J., van Dam, H., and Ten Dijke, P.
(2011). Fas-associated factor 1 antagonizes Wnt signaling by promoting b-cat-
enin degradation. Mol. Biol. Cell 22, 1617–1624.
Zhang, L., Zhou, F., Li, Y., Drabsch, Y., Zhang, J., van Dam, H., and ten Dijke,
P. (2012). Fas-associated factor 1 is a scaffold protein that promotes b-trans-
ducin repeat-containing protein (b-TrCP)-mediated b-catenin ubiquitination
and degradation. J. Biol. Chem. 287, 30701–30710.

Cell Reports
Supplemental Information
through Stabilization of -Catenin
Hyeonwoo Kim, Carlos Rodriguez-Navas, Rahul K. Kollipara, Payal Kapur, Ivan
Pedrosa, James Brugarolas, Ralf Kittler, and Jin Ye

Supplemental Data
Figure S1. Unsaturated FAs stabilize β-catenin through inactivation of FAF1 (Related to
Figure 1).
(A) Relative amount of FAF1 mRNA in SRD-13A cells transfected with the indicated siRNA
shown in Figure 1C was determined by RT-QPCR, with the value in cells transfected with the
control siRNA set at 1. Results are reported as means ± S.E. from three independent
experiments.
(B) HEK-293 cells were seeded and treated as described in Figure 2A. Whole cell lysates were
subjected to immunoblot analysis with the indicated antibodies.

Figure S2. Lipid analysis and subcellular localization of β-catenin in ccRCC cells (Related
to Figure 3).
(A-C) On day 0, SW156 and 786-O cells were seeded at 3.5 × 105
per 60-mm dish, whereas
HEK-293 cells were seeded at 4.0 × 105
per 60-mm dish. On day 2, cells were harvested and the
amount of free FAs (A), triglycerides (B), and cholesterol (C) in the cells was determined as
described in Experimental procedures.
(D) SRD-13A cells were seeded and treated as described in Figure 1A. SW156 cells were
seeded and treated as described in Figure 3A. Cytosolic and membrane fractions were subjected
to immunoblot analysis with the indicated antibodies.
(E and G) 786-O cells were set up, treated and analyzed the same as SW156 cells described in
Figures 3A and 3E, respectively.
(F) 786-O cells were set up, transfected with the indicated siRNA, and analyzed the same as
SRD-13A cells described in Figure 1C.

Figure S3. Enrichment map of genes co-expressed with SCD1 in ccRCC (Related to Figure
4)
P values for enrichment between gene signature sets were calculated (hypergeometric test), and
used to construct a network of significantly enriched signatures (FDR cut-off 1%). The node
size corresponds to the number of genes in a signature set, the edge intensity correlates with the
significance of enrichment between two signature set.
Table S1. Genes whose expression is positively correlated with SCD1 expression (Related
to Figure 4).
The top 5% of genes whose expression is positively correlated with SCD1 expression was
identified. The potential target genes of β-catenin are labeled in yellow.

Supplemental Experimental Procedures
Materials
We obtained FAs from Nu-Chek Prep, Inc.; FA-free bovine serum albumin from Roche
Applied Science; Recombinant human or mouse Wnt3a from R & D systems; MG132 and Nonidet
P-40 alternative (NP-40) from Calbiochem; NCX4040 and a rabbit anti-actin from Sigma-Aldrich;
A939572 from Biovision; a mouse anti-HSV from Novagen; a mouse anti-β-catenin from BD
Biosciences; a rabbit anti-phospho-β-catenin (Ser33/37/Thr41) from Cell Signaling; a mouse anti-
ubiquitin from Santa Cruz Biotechnology; horseradish peroxidase-conjugated donkey anti-mouse
and anti-rabbit IgGs (affinity-purified) from Jackson ImmunoResearch Laboratories; and Ni-NTA
agarose from Qiagen. Hybridoma cells producing IgG-9E10, a mouse monoclonal antibody against
Myc-tag, was obtained from the American Type Culture Collection. A rabbit polyclonal antibody
against human FAF1 was generated by immunizing rabbits with full-length human FAF1.
Delipidated fetal calf serum (DFCS) was prepared from newborn calf serum by n-butyl alcohol and
isopropyl ether extraction as previously described (Hannah et al., 2001). All FAs added into culture
media were conjugated to bovine serum albumin as previously reported (Hannah et al., 2001).
ccRCC tumors were obtained from the Tissue Management Shared Resource, the Simmons Cancer
Center, University of Texas Southwestern Medical Center, which provides Institutional Review
Board-approved centralized tissue procurement services.
Plasmid constructs
pCMV-Myc-FAF1 encodes human FAF1 with five copies of the c-Myc epitope
(EQKLISEEDL) at its NH2-terminus under control of the CMV promoter; pTK-HSV-β-catenin
encodes full-length β-catenin preceded by two copies of the HSV epitope tag (QPELAPEDPED) at
the NH2-terminus under the control of the thymidine kinase (TK) promoter; and pAcHLT-

FAF1(325-419) was generated to produce the indicated fragment of FAF1 in sf9 cells through
recombinant baculovirus expression system as previously described (Kim et al., 2013).
Oligonucleotide site-directed mutagenesis was carried out with complementary primers using
QuickChange Site-Directed Mutagenesis kit from Stratagene. Open reading frames in all plasmids
were confirmed by DNA sequencing.
Cell culture
SRD-13A cells are a clone of mutant CHO cells deficient in Scap (Rawson et al., 1999).
They were maintained in medium A (1:1 mixture of Ham’s F-12 medium and DMEM, 100 units/ml
penicillin, 100 μg/ml streptomycin sulfate) supplemented with 5% (v/v) FCS, 5 μg/ml cholesterol, 1
mM sodium mevalonate, and 20 μM sodium oleate. HEK-293 cells were grown in medium B
(Dulbecco’s modified Eagle’s medium with 1.0 g/l glucose, 100 units∕ml penicillin, and 100 μg∕ml
streptomycin) supplemented with 10% FCS. SW156 and 786-O cells were grown in medium C
(Dulbecco’s modified Eagle’s medium with 4.5 g/l glucose, 100 units∕ml penicillin, and 100 μg∕ml
streptomycin) supplemented with 10% FCS. Except for SW156 and 786-O cells that were incubated
in 5% CO2, all cells were maintained at 37 °C in 8.8% CO2.
Immunoblot analysis
Immunoblot analyses in the current study were performed with IgG-9E10 (1 μg∕ml), a
rabbit anti–FAF1 (1 μg∕ml), a rabbit anti-actin (1∶2,000), a mouse anti-HSV (1:5,000), a mouse anti-
β-catenin (1:2,000), a mouse anti-ubiquitin (1:200), and a rabbit anti-phospho-β-catenin
(Ser33/37/Thr41) (1:1,000). Horseradish peroxidase-conjugated donkey anti-mouse and anti-rabbit
IgGs (0.2 μg∕ml) were used as the secondary antibody. Bound antibodies were visualized by
chemiluminescence using the SuperSignal substrate system (Pierce) according to the
manufacturer’s instructions.

RNA interference
Duplexes of siRNA were synthesized by Dharmacon Research. The two siRNA sequences
targeting hamster FAF1 in CHO cells are GAACGUGAAGCCAGAGAAAUU and
AGUCAUUAUUGGAGGUAAAUU, and the two siRNA targeting human FAF1 are
GGGCUUGGGAUCUGACAAAUU and GAACGUGAAGCCAGAGAAAUU. The control siRNA
targeting GFP was reported previously (Adams et al., 2004). Cells were transfected with siRNA (20
µM) using Lipofectamine RNAiMAX reagent (Invitrogen) as described by the manufacturer.
qRT-PCR
qRT-PCR was performed as previously described (Liang et al., 2002). Each measurement
was made in triplicate from cell extracts pooled from duplicate dishes. The relative amounts of
RNAs were calculated through the comparative cycle threshold method by using human 36B4 or
cyclophilin mRNA as the invariant control.
Transient transfection
SRD-13A cells were transfected with plasmids with X-tremeGENE HP DNA Transfection
Reagent (Roche Applied Science) according to the manufacturer's protocol. Total plasmid
concentration was adjusted to 2 µg/dish by the empty vector pcDNA3.1.
Pulse-chase analysis
SRD-13A cells cultured in 60-mm dishes were incubated in medium A with 5% DFCS, 5
μg/ml cholesterol, 1 mM sodium mevalonate for 14 h followed by incubation in medium A with
0.5% DFCS in the absence or presence of 100 µM oleate for 4 h. The cells were pulse-labeled with
150 mCi/ml [35
S] Protein Labeling Mix (Perkin-Elmer Life Sciences) in 2 ml of medium A
supplemented with 0.5% DFCS in the absence or presence of 100 µM oleate for 1 h, and chased for
various times in medium A supplemented with 0.5% DFCS, 0.2 mM unlabeled methionine, and 0.4

mM unlabeled cysteine in the absence or presence of 100 µM oleate. Following the chase, cells
pulled from 4 dishes were lysed in 0.3 ml of buffer A (25 mM Tris-HCl pH 7.2, 0.15 M NaCl, 5
µg/ml pepstatin, 10 µg/ml leupeptin, 2 µg/ml aprotinin, 2 µg/ml N-[N-(N-Acetyl-L-leucyl)-L-
leucyl]-L-norleucine, 1% NP-40), and β-catenin was immunoprecipitated with anti-β-catenin and
protein A/G beads as previously described (Sakai et al., 1997). Aliquots of the immunoprecipitates
were subjected to SDS/PAGE, transferred to Hybond C-Extra nitrocellulose filters, exposed to a
phosphorimaging plate at room temperature for 24 h, and scanned in a Molecular Dynamics Storm
820 phosphorimaging device. ImageJ was used for quantification of radiolabeled β-catenin.
Immunoprecipitation
Immunoprecipitation of β-catenin was carried out as described in the pulse-chase analysis.
For immunoprecipitation of FAF1, pooled cell pellets from five 100-mm dishes of the HEK-293
cells were suspended in 0.5 ml of buffer B (25 mM Tris-HCl pH 7.2, 0.15 M NaCl, 5 µg/ml
pepstatin, 10 µg/ml leupeptin, 2 µg/ml aprotinin, 2 µg/ml N-[N-(N-Acetyl-L-leucyl)-L-leucyl]-L-
norleucine, 10 µM MG132) to obtain cytosol fraction through centrifugation as previously
described (Sakai et al., 1996). FAF1 was immunoprecipitated with 15 µg anti-FAF1 and protein
A/G beads for 5 h at 4 °C with an established protocol (Sakai et al., 1997).
BN-PAGE analysis
Wild type and mutant UAS domain of FAF1 were expressed, purified, incubated with FAs,
and analyzed by BN-PAGE exactly as previously described (Kim et al., 2013).
Cell growth assay
Cell growth was assessed by CellTiter-Glo® Luminescent Cell Viability Assay (Promega)
according to the instruction of the manufacturer.
Bioinformatics analysis

To identify genes that are co-expressed with SCD1 in ccRCC tumors, we retrieved RNA-
Seq data of 532 tumor samples from TCGA (https://tcga-
data.nci.nih.gov/tcga/dataAccessMatrix.htm?mode=ApplyFilter&showMatrix=true&diseaseType=
KIRC&tumorNormal=TN&tumorNormal=T&tumorNormal=NT&platformType=3&platformType
=5&platformType=27&platformType=38&platformType=43). Using this data, spearman rank
ordered correlation was computed between SCD1 expression and expression of every gene in the
genome. The top 5% of genes with positive correlation with SCD1 expression were selected for
further analyses.
Transcription factors motifs enrichment analysis was performed on genes that are co-
expressed with SCD1 in ccRCC tumor samples using WebGestalt. WebGestalt computes the
significance of enrichment using hypergeometric test and adjusts for multiple hypothesis testing.
Also, we performed pathway enrichment analysis on this gene list using canonical pathway gene
signatures obtained from MsigDB database. We calculated hypergeometric p-values for pathway
signatures that were overrepresented in the SCD1 co-expressed genes and adjusted for multiple
hypotheses testing using Benjamini-Hochberg method. Significantly enriched pathways (FDR ≤
1%) were selected and similarly the hypergeometric test was used to calculate the enrichment p-
value between each pair of significantly enriched pathways. These p-values were used to plot the
edges in enrichment map to represent the strength of enrichment between gene sets.

Supplemental References
Adams, C.M., Reitz, J., De Brabander, J.K., Feramisco, J.D., Li, L., Brown, M.S., and
Goldstein, J.L. (2004). Cholesterol and 25-hydroxycholesterol inhibit activation of SREBPs by
different mechanisms, both involving SCAP and Insigs. J Biol Chem 279, 52772-52780.
Hannah, V.C., Ou, J., Luong, A., Goldstein, J.L., and Brown, M.S. (2001). Unsaturated fatty
acids down-regulate SREBP isoforms 1a and 1c by two mechanisms in HEK-293 Cells. J Biol
Chem 276, 4365-4372.
Kim, H., Zhang, H., Meng, D., Russell, G., Lee, J.N., and Ye, J. (2013). UAS domain of Ubxd8
and FAF1 polymerizes upon interaction with long-chain unsaturated fatty acids. J Lipid Res 54,
2144-2152.
Liang, G., Yang, J., Horton, J.D., Hammer, R.E., Goldstein, J.L., and Brown, M.S. (2002).
Diminished hepatic response to fasting/refeeding and liver X receptor agonists in mice with
selective deficiency of sterol regulatory element-binding protein-1c. J Biol Chem 277, 9520-
9528.
Rawson, R.B., DeBose-Boyd, R., Goldstein, J.L., and Brown, M.S. (1999). Failure to cleave
sterol regulatory element-binding proteins (SREBPs) causes cholesterol auxotrophy in Chinese
hamster ovary cells with genetic absence of SREBP cleavage-activating protein. J Biol Chem
274, 28549-28556.
Sakai, J., Duncan, E.A., Rawson, R.B., Hua, X., Brown, M.S., and Goldstein, J.L. (1996).
Sterol-regulated release of SREBP-2 from cell membranes requires two sequential cleavages,
one within a transmembrane segment. Cell 85, 1037-1046.
Sakai, J., Nohturfft, A., Cheng, D., Ho, Y.K., Brown, M.S., and Goldstein, J.L. (1997).
Identification of complexes between the COOH-terminal domains of sterol regulatory element-
binding proteins (SREBPs) and SREBP cleavage-activating protein. J Biol Chem 272, 20213-
20221.

mmc3

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (19)

Similar to mmc3

Similar to mmc3 (14)

mmc3