ONCOLOGY LETTERS 15: 4423-4426, 2018

Pro‑metastatic signaling of the trans fatty acid

elaidic acid is associated with lipid rafts
SHINGO KISHI1, RINA FUJIWARA‑TANI1, YI LUO2, ISAO KAWAHARA1, KEI GOTO1, KIYOMU FUJII1,
HITOSHI OHMORI1, CHIE NAKASHIMA1, TAKAMITSU SASAKI1 and HIROKI KUNIYASU1

1
Department of Molecular Pathology, Nara Medical University, Kashihara, Nara 634‑8521, Japan;
2
Jiangsu Province Key Laboratory of Neuroregeneration, Nantong University, Nantong, Jiangsu 226001, P.R. China

Received September 11, 2017; Accepted January 11, 2018

DOI: 10.3892/ol.2018.7817

Abstract. Trans fatty acids (TFAs) are risk factors for cardio- Min mice (6) and induces the proliferation of Ehrlich ascites
vascular disorders, and the cancer‑promoting effects of TFAs sarcoma cells (7). EA enhances the metastasis of CRC in
have been previously reported. The present study examined the mouse models (4) by increasing cancer cell stemness and
effects and signaling of elaidic acid (EA), a TFA, in colorectal epithelial‑mesenchymal transition (EMT) (5). EA also induces
cancer (CRC) cells. Oral intake of EA was found to increase chemoresistance in CRC cells by upregulating its stemness (8).
metastasis of HT29 human CRC cells. Results indicated that, in LCFAs bind to specific membrane‑bound receptors, namely,
the plasma membrane, EA was integrated into cholesterol rafts, G‑protein coupled receptor 40 (GPR40) and GPR120 (9). EA
which contain epidermal growth factor receptors (EGFR). EA also binds to GPR40 and 120, inducing the transactivation of
increased nanog and c‑myc, and decreased PGC‑1A through EGFR via c‑src (5). Lipid rafts represent a special localiza-
lipid raft‑associated EGFR signaling in HT29 cells. Depletion tion of epidermal growth factor receptors (EGFRs), altering
of cholesterol by methyl‑β‑cyclodextrin treatment abrogated its signaling pathways (10) and transactivation (11). Rao et al
the EA‑induced stemness and oxidative phosphorylation. have reported that EA can affect Fas/FasL‑induced apoptosis
Simvastatin treatment also abrogated EA‑enhanced tumor via lipid rafts (12). Lipid rafts may play an important role in
growth. These results indicate that EA enhances the stemness EGFR signaling in EA‑treated cells.
by activating EGFR in lipid rafts. In the present study, we examined the effect of EA on
pro‑metastatic EGFR signals that may be associated with
Introduction cholesterol rafts.

Colorectal cancer (CRC) is the third leading cause of Materials and methods
cancer‑related deaths in Japan (1), and its incidence has been
reported to be increasing because of the popularity of western Cell culture and reagents. HT29, a human colon cancer cell
lifestyle (2,3). We have previously reported the pro‑tumoral line, was purchased from Dainihon Pharmacy Co. (Tokyo,
effects of elaidic acid (EA), a major trans fatty acid Japan). The cells were routinely maintained in Dulbecco's
(TFA) (4,5). EA is a long‑chain fatty acid (LCFA) containing modified Eagle's medium (DMEM) supplemented with
18 carbon chains in a trans configuration, and is a structural 10% fetal bovine serum (both Sigma, St. Louis, MO, USA)
isomer of oleic acid (OA). EA promotes carcinogenesis in in 5% CO2 at 37˚C. Cell morphology was evaluated daily by
microscopic examination. Each cell line was routinely tested
for Mycoplasma contamination by genomic PCR. The viability
of each cell line was tested by trypan blue exclusion assay.
Correspondence to: Professor Hiroki Kuniyasu, Department EA (CAS no. 112‑79‑8), OA (CAS no. 112‑80‑1; both WAKO
of Molecular Pathology, Nara Medical University, 840 Shijo‑cho, Pure Chemicals, Osaka, Japan), and methyl‑β ‑cyclodextrin
Kashihara, Nara 634‑8521, Japan (MbCD; Sigma) were purchased from the mentioned suppliers.
E‑mail: [email protected]‑net.ne.jp Cell membrane cholesterol was depleted by incubating the
cells in a sterile‑filtered MbCD medium (RPMI‑1640 medium
Abbreviations: CRC, colorectal cancer; TFA, trans fatty acid;
supplemented with 10 mM MbCD and cholesterol‑free serum;
LCFA, long-chain fatty acid; EA, elaidic acid; OA, oleic acid;
MbCD, methyl‑β‑cyclodextrin; EGF, epidermal growth factor; GPR, Sigma).
G‑protein coupled receptor; EGF, epidermal growth factor; EGFR,
EGF receptor; ERK, extracellular signal‑regulated kinase; EMT, Assessment of cell growth. The cells (1x10 4 per well) were
epithelial‑mesenchymal transition; SIM, simvastatine seeded in a 12‑well dish. Cell growth was assessed using a
3‑(4,5‑dimethylthiazol‑2‑yl)‑2,5‑diphenyltetrazolium bromide
Key words: epidermal growth factor receptor, lipid raft, cholesterol (MTT) assay or by counting the number of cells with an
Autocytometer (Sysmex, Kobe, Japan) after 48 h, as described
previously (13).
4424 KISHI et al: Trans FATTY ACIDS ASSOCIATED WITH LIPID RAFTS

Animal models. BALB/cSlc‑nu/nu mice (males, 5 weeks old)

were purchased from Japan SLC (Shizuoka, Japan). The mice
were maintained according to the institutional guidelines
approved by the Committee for Animal Experimentation of
Nara Medical University (Kashihara, Japan), in accordance
with the current regulations and standards established by the
Ministry of Health, Labor, and Welfare. Each experimental
group comprised 5 animals.
For the subcutaneous tumor model, HT29 cancer cells
(1x107 cells) suspended in Hank's balanced salt solution (Sigma)
were inoculated into the scapular subcutaneous tissue of mice.
For analyzing the metastasis to the liver, lung, and peritoneum,
cancer cells (1x106 cells) were inoculated into the spleen, the
tail vein, or peritoneal cavity, respectively. EA or OA were
administrated through a gavage. The growth of subcutaneous Figure 1. Oral intake of EA enhanced the metastasis of CT26 cells. Metastasis
tumors and the metastatic status of the cells was assessed after of HT29 human CRC cells in nude mice treated with EA, OA (10 mg/mouse
4 weeks. For assessing growth of subcutaneous tumors, the in 30% ethanol), or vehicle (30% ethanol) by gavage administration. Tumors
were assessed 4 weeks after cancer cell inoculation by counting the number
diameters of the tumors were measured every week. For peri- of tumors at the liver surface. Error bars are the mean ± SD. *P<0.0001 and
toneal metastasis, the weight of mesenterium with tumors was #
P<0.05 vs. control. CRC, colorectal cancer; EA, elaidic acid; OA, oleic acid.
measured. For metastases to the liver and lung, the number
of tumors at the organ surface was counted. Simvastatin
(50 mg/kg/day; WAKO) was administrated through a gavage
for 10 days after the inoculation of cancer cells. Twelve fractions (1 ml each) were collected from the top to
the bottom of the gradient (F1‑F12). Immunoblotting analysis
Western blot analysis. Whole‑cell lysates were prepared as was performed to identify the fractions containing lipid rafts.
described previously (14). Cell fractions were extracted by Cholesterol levels were measured using the Amplex Red
processing the cells with the Cell Fractionation kit (Abcam, cholesterol assay kit (Thermo Fisher Scientific, Inc., Waltham,
Cambridge, MA, USA), according to the manufacturer's MA, USA).
instructions.
Proteins (20 µg) were separated from the cell lysates by Statistical analysis. Statistical analyses of experimental
sodium dodecyl sulfate‑polyacrylamide gel electrophoresis on data were performed using the Mann‑Whitney U test or
12.5% gels. They were then electrotransferred onto nitrocellu- the Kruskal‑Wallis test with Dunn's multiple comparison
lose membranes for immunoblotting analysis. The membranes (nonparametric ANOVA). Differences with a two‑sided
were incubated with primary antibodies, followed by incuba- P‑value <0.05 were considered to indicate a statistically
tion with peroxidase‑conjugated IgG antibodies (Medical & significant difference.
Biological Laboratories Co., Ltd, Nagoya, Japan). A tubulin
antibody (Oncogene Research Products, Cambridge, MA, USA) Results
was used to measure the amount of proteins loaded in each
lane. Immune complexes were visualized using the CSA system EA enhances metastasis. The HT29 cells were inoculated into
(DAKO, Carpinteria, CA, USA). Antibodies against EGFR the subcutaneous tissue, spleen (liver metastasis), tail vein
(Transduction Laboratories, Lexington, KY, USA), phosphory- (lung metastasis), and peritoneum of nude mice treated with
lated EGFR, phosphorylated ERK1/2 (pThr204), nanog (clone EA, OA, or control vehicle by gavage administration (Fig. 1).
1E6C4), c‑Myc (clone 9E10; Santa Cruz Biotechnology, Santa The EA‑treated mice showed higher growth of subcutaneous
Cruz, CA, USA), phosphorylated p38 (pThr180/pTyr182; Bioss tumors and metastases to liver, lungs, and peritoneum than
Inc., Woburn, MA, USA), peroxisome proliferator‑activated control mice. In contrast, only peritoneal metastasis was
receptor γ coactivator‑1α (PGC‑1A; Proteintech Group, Inc., increased in the OA‑treated mice, compared to the control
Rosemount, IL, USA) were used as primary antibodies. mice.

Extraction of raft fractions. Lipid rafts were isolated from EA activates EGFR in lipid rafts. We examined the integra-
5x107 cells. Briefly, cell pellets disrupted in 1 ml buffer A tion of EA into lipid rafts (Fig. 2A). We observed that EA was
[50 mM Tris‑HCl (pH 8.0), 10 mM MgCl 2, 0.15 M NaCl, preferentially integrated into cholesterol rafts, significantly
1% Triton X‑100, 5% glycerol, 50 mM PMSF, 1x protease increasing their cholesterol content (Fig. 2B). In contrast,
inhibitor cocktail (Sigma), and 0.03% β‑mercaptoethanol] were MbCD preferentially decreased the cholesterol content of
centrifuged for 5 min at 500 x g and 4˚C. The obtained super- EA‑induced rafts (Fig. 2A). EA integration into cholesterol
natant was added to 1 ml buffer A [50 mM Tris‑HCl (pH 8.0), rafts upregulated the accumulation and induced the phos-
10 mM MgCl2, 0.15 M NaCl, and 80% sucrose] to make up phorylation of EGFR (Fig. 2B). In contrast, MbCD suppressed
the final concentration of sucrose to 40%. A discontinuous EA‑induced phosphorylation of EGFR and ERK1/2
sucrose gradient was obtained by stratifying 7.5 ml buffer A (Fig. 2C, D). Furthermore, the levels of phosphorylated p38
with 38% sucrose and 2 ml buffer A with 15% sucrose. The were higher than the levels of phosphorylated ERK1/2 in
gradient was ultracentrifuged for 18 h at 100,000 x g and 4˚C. cholesterol‑raft depleted cells (Fig. 2D).
 ONCOLOGY LETTERS 15: 4423-4426, 2018 4425

Figure 2. Effect of EA on EGFR activation in lipid rafts. (A) Effect of EA on cholesterol content in total membrane or detergent‑resistant raft fractions.*P<0.001 vs. 0
group. (B) EGFR and phosphorylated EGFR levels in lipid rafts or membrane fractions in CT26 cells. (C) Effect of MbCD on EA‑induced EGFR phos-
phorylation. (D) Effect of MbCD on EA‑induced ERK1/2 and p38 phosphorylation in CT26 cells. Tubulin was used as an internal control. Each bar or point
represents mean ± SD of three independent experiments. EA, elaidic acid; EGFR, epidermal growth factor receptor; MbCD, methyl‑β‑cyclodextrin; pERK1/2,
phosphorylated ERK1/2; pp38, phosphorylated p38.

Figure 3. Effect of EA, with or without SIM, on stemness and energy metabolism. (A) Protein expression levels of nanog, c‑Myc, and PGC‑1A in HT29 cells
treated with EA or EA and MbCD. Tubulin was used as an internal control. (B) Effect of simvastatin on tumor growth of HT29 subcutaneous tumors in nude
mice treated with EA. SIM (20 mg/kg/day) was administrated with a gavage for 10 days after the inoculation. Error bars are the mean ± SD. EA, elaidic acid;
SIM, simvastatin; PGC‑1A, proliferator‑activated receptor γ coactivator‑1α.

Effect of EA on cancer stem cell‑proteins. Next, we examined Discussion

the role of lipid rafts on the stemness of cancer cells, which
is known to be enhanced by EA (5). EA treatment increased Results of the present study indicated that EA could strongly
nanog production in HT29 cells; however, it was abrogated enhance cancer metastasis by increasing their stemness via
by cholesterol depletion after MbCD treatment (Fig. 3A). EGFR activation in lipid rafts.
c‑Myc protein levels were also increased by EA treatment In our data, HT29 cells showed enhanced metastasis after
and abrogated by cholesterol depletion. In contrast, PGC‑1A EA gavage administration. EA activated EGFRs, as reported
levels were decreased by EA and abrogated by cholesterol in our previous report (5); however, it was important that
depletion. EA‑induced EGFR signals disappeared after the absorbance
Simvastatine, which has been reported to deplete lipid of cholesterol, which caused lipid raft depletion. Raft depletion
rafts in a cholangiocarcinoma mouse model (15), was also abrogated the EA‑induced expression of nanog and c‑myc.
administrated to the EA‑treated mice. In the subcutaneous These findings suggested that EA‑associated signal pathways
HT29 cell tumor model, lipid raft depletion inhibited the may prefer lipid rafts.
EA‑induced tumor growth (Fig. 3B). The tumor growth rate EA can be incorporated into plasma membranes composed
in the simvastatine and EA‑treated mice was lower than that of stearic or palmitic acid, to be subsequently converted into
in the EA‑treated mice and even than that in the control triglycerides (16), cholesterol esters, phospholipids, or other
mice. LCFAs such as OA and LA (17). Our results were consistent
4426 KISHI et al: Trans FATTY ACIDS ASSOCIATED WITH LIPID RAFTS

with the findings of these studies; we observed that EA could  4. Ohmori H, Fujii K, Kadochi Y, Mori S, Nishiguchi Y, Fujiwara R,
Kishi S, Sasaki T and Kuniyasu H: Elaidic acid, a trans‑fatty
easily integrate into the plasma membrane, especially choles- acid, enhances the metastasis of colorectal cancer cells.
terol rafts, of CRC cells. Pathobiology 84: 144‑151, 2017.
We found that lipid rafts played an important role in EA  5. Fujii K, Luo Y, Fujiwara‑Tani R, Kishi S, He S, Yang S, Sasaki T,
Ohmori H and Kuniyasu H: Pro‑metastatic intracellular signaling
signaling. Lipid rafts are microdomains within the lipid bilayer of the elaidic trans fatty acid. Int J Oncol 50: 85‑92, 2017.
of cellular membranes. Levels of non‑constitutive proteins in  6. Molin M, Berstad P, Benth JS, Alexander J, Paulsen JE and
lipid rafts fluctuate dramatically in different cancers, affecting Almendingen K: Effect of different degrees of hydrogenated
fish oil on intestinal carcinogenesis in Min/+ mice. Anticancer
various signaling pathways (18). We have previously reported that Res 33: 477‑483, 2013.
EA binds to GPR40 and GPR120, which signals SRC‑mediated  7. Awad AB: Trans fatty acids in tumor development and the host
crosstalk to EGFRs (5). Another previous study had shown that survival. J Natl Cancer Inst 67: 189‑192, 1981.
 8. Tanabe E, Kitayoshi M, Fujii K, Ohmori H, Luo Y, Kadochi Y,
the crosstalk between GPRs and EGFRs in the raft could activate Mori S, Fujiwara R, Nishiguchi Y, Sasaki T and Kuniyasu H:
phospholipases C, D, and A2, which could activate protein kinase Fatty acids inhibit anticancer effects of 5‑fluorouracil in mouse
C, ERK, and Akt (19). Lipid rafts are known to be thicker than cancer cell lines. Oncol Lett 14: 681‑686, 2017.
 9. Hara T, Kashihara D, Ichimura A, Kimura I, Tsujimoto G and
the surrounding membrane and the propensity of EGFR to be Hirasawa A: Role of free fatty acid receptors in the regulation
located in rafts is characterized by a thick lipid bilayer (35.0 Å) of energy metabolism. Biochim Biophys Acta 1841: 1292‑1300,
around the transmembrane segment of the receptor (20). 2014.
10. Zhang Z, Wang L, Du J, Li Y, Yang H, Li C, Li H and Hu H:
The raft‑associated EA signal upregulated nanog and Lipid raft localization of epidermal growth factor receptor alters
c‑Myc, while it downregulated PGC‑1A. c‑Myc is associated matrix metalloproteinase‑1 expression in SiHa cells via the
with cell proliferation and aerobic glycolysis, and PGC‑1A MAPK/ERK signaling pathway. Oncol Lett 12: 4991‑4998, 2016.
11. Overmiller AM, McGuinn KP, Roberts BJ, Cooper F,
is associated with stemness and oxidative phosphoryla- Brennan‑Crispi DM, Deguchi T, Peltonen S, Wahl JK III and
tion (21,22). The upregulation of nanog in the present data and Mahoney MG: c‑Src/Cav1‑dependent activation of the EGFR by
our previous data (5) indicated that EA increased the stemness Dsg2. Oncotarget 7: 37536‑37555, 2016.
12. Rao H, Ma LX, Xu TT, Li J, Deng ZY, Fan YW and Li HY: Lipid
of cancer cells; however, the cancer cells may have been of a rafts and Fas/FasL pathway may involve in elaidic acid‑induced
proliferative nature with glycolytic energy metabolism, which apoptosis of human umbilical vein endothelial cells. J Agric
is different from dormant stem cells that possess oxidative Food Chem 62: 798‑807, 2014.
13. Kuniyasu H, Yano S, Sasaki T, Sasahira T, Sone S and Ohmori H:
phosphorylation metabolism. The induction of proliferative Colon cancer cell‑derived high mobility group 1/amphoterin
cancer stem cells may be the most striking characteristic of induces growth inhibition and apoptosis in macrophages. Am J
EA to enhance cancer metastasis. Pathol 166: 751‑760, 2005.
14. Kuniyasu H, Oue N, Wakikawa A, Shigeishi H, Matsutani N,
We observed that cholesterol depletion inhibited Kuraoka K, Ito R, Yokozaki H and Yasui W: Expression of
EA‑enhanced tumor growth below the level of untreated receptors for advanced glycation end‑products (RAGE) is closely
tumors in the subcutaneous tumor mouse model. Importantly, associated with the invasive and metastatic activity of gastric
cancer. J Pathol 196: 163‑170, 2002.
lipid raft depletion may inhibit EGFR signals in the presence 15. Miller T, Yang F, Wise CE, Meng F, Priester S, Munshi MK,
or absence of EA. Statins have been reported to inhibit the Guerrier M, Dostal DE and Glaser SS: Simvastatin stimulates
growth of JAK2‑mutated cells by inhibiting lipid rafts (23) and apoptosis in cholangiocarcinoma by inhibition of Rac1 activity.
Dig Liver Dis 43: 395‑403, 2011.
inducing apoptosis by inhibiting Rac1 (15). Thus, lipid rafts act 16. Harvey KA, Walker CL, Xu Z, Whitley P and Siddiqui RA:
as a platform for collecting multiple stimulants, receptors, and Trans fatty acids: induction of a pro‑inflammatory phenotype in
signals to activate cancer progression (24). The mechanism endothelial cells. Lipids 47: 647‑657, 2012.
17. Vidgren HM, Louheranta AM, Agren JJ, Schwab US and
of integration of EA to cholesterol raft should be elucidated Uusitupa MI: Divergent incorporation of dietary trans fatty acids
in further examinations; however, our data suggests that lipid in different serum lipid fractions. Lipids 33: 955‑962, 1998.
rafts may be promising targets for anticancer treatment. 18. Staubach S and Hanisch FG: Lipid rafts: Signaling and
sorting platforms of cells and their roles in cancer. Expert Rev
Proteomics 8: 263‑277, 2011.
Acknowledgements 19. Rozengurt E: Mitogenic signaling pathways induced by
G protein‑coupled receptors. J Cell Physiol 213: 589‑602, 2007.
20. Postic G, Ghouzam Y, Etchebest C and Gelly JC: TMPL: A
The authors thank Ms. Tomomi Masutani for assistance database of experimental and theoretical transmembrane protein
with the preparation of this manuscript. The present study models positioned in the lipid bilayer. Database (Oxford) 2017,
was supported by MEXT KAKENHI (grant nos. 16H05164, 2017.
21. Sancho P, Burgos‑Ramos E, Tavera A, Bou Kheir T, Jagust P,
17K15648, 17K19923 and 16K19087). Schoenhals M, Barneda D, Sellers K, Campos‑Olivas R,
Graña O, et al: MYC/PGC‑1α balance determines the metabolic
References phenotype and plasticity of pancreatic cancer stem cells. Cell
Metab 22: 590‑605, 2015.
 1. Wakao F, Nishimoto H, Kataonoda K, Tsukuma H and 22. Scognamiglio R, Cabezas‑Wallscheid N, Thier MC, Altamura S,
Mikami H (eds.): Cancer Statistics in Japan, 2013. National Reyes A, Prendergast ÁM, Baumgärtner D, Carnevalli LS,
Cancer Res Inst, Tokyo, 2013. Atzberger A, Haas S, et al: Myc depletion induces a pluripotent
 2. Kimura Y, Kono S, Toyomura K, Nagano J, Mizoue T, Moore MA, dormant state mimicking diapause. Cell 164: 668‑680, 2016.
Mibu R, Tanaka M, Kakeji Y, Maehara Y, et al: Meat, fish and 23. Griner LN, McGraw KL, Johnson JO, List AF and Reuther GW:
fat intake in relation to subsite‑specific risk of colorectal cancer: JAK2‑V617F‑mediated signalling is dependent on lipid rafts
The Fukuoka colorectal cancer study. Cancer Sci 98: 590‑597, and statins inhibit JAK2‑V617F‑dependent cell growth. Br J
2007. Haematol 160: 177‑187, 2013.
 3. Mizoue T, Tanaka K, Tsuji I, Wakai K, Nagata C, Otani T, 24. Irwin ME, Mueller KL, Bohin N, Ge Y and Boerner JL: Lipid
Inoue M and Tsugane S; Research Group for the Development raft localization of EGFR alters the response of cancer cells to
and Evaluation of Cancer Prevention Strategies in Japan: Alcohol the EGFR tyrosine kinase inhibitor gefitinib. J Cell Physiol 226:
drinking and colorectal cancer risk: An evaluation based on a 2316‑2328, 2011.
systematic review of epidemiologic evidence among the Japanese
population. Jpn J Clin Oncol 36: 582‑597, 2006.