pISSN 1598-2998, eISSN 2005-9256

Cancer Res Treat. 2020;52(3):815-829 https://doi.org/10.4143/crt.2019.380

Original Article Open Access

G –Coupled Receptor 30 Mediates the Anticancer Effects Induced by Eicosapentaenoic Acid in Ovarian Cancer Cells

Yue Zhao, PhD1 Purpose Meng-Fei Zhao, BS1 While numerous epidemiological studies have indicated that omega-3 polyunsaturated fatty acids have anticancer properties in various cancers, the effects and mechanisms of eicos- Mei-Lin Yang, PhD2 apentaenoic acid (EPA) in ovarian cancer cell growth are poorly understood. Tian-Yu Wu, BS1 Cong-Jian Xu, PhD3 Materials and Methods Jing-Mei Wang, PhD4 ES2 ovarian clear cell carcinoma cells and SKOV3 adenocarcinoma cells were treated with Chao-Jun Li, PhD1 palmitic acid or EPA, followed by flow cytometry and cell counting to measure apoptosis and proliferation, respectively. A modified protein lipid overlay assay was used to further verify Xi Li, PhD5 whether EPA was a ligand of G protein–coupled receptor 30 (GPR30) in ES2 cells. The levels of apoptosis-related , phosphorylated AKT, and phosphorylated ERK1/2 were detec- ted to explore the underlying mechanism. Finally, inhibitory effect of EPA on tumor growth via GPR30 was determined in vitro and in vivo. *A list of author’s a!liations appears at the Results end of the paper. EPA suppressed ES2 ovarian clear cell carcinoma cells growth via GPR30, a novel EPA receptor, by inducing apoptosis. As a ligand of GPR30, EPA activated the GPR30-cAMP– protein kinase A signaling pathway. When GPR30 was suppressed by siRNA or its inhibitor G15, the antiproliferative action of EPA was impaired. Furthermore, EPA inhibited tumor + C +o r+re s+p o+n d+e n+c e +: Y +u e + Z h+a o+, P+h D+ + + + + + + growth by blocking the activation of AKT and ERK. In the mouse xenograft model, EPA + + + + + + + + + + + + + + + + + + + + decreased tumor volume and weight through GPR30 by blocking tumor cell proliferation. + S +ta t+e K+e y+ L +a b +o r a+t o +ry +o f +P h +a r +m a+c e +u t i+ca l+ + + + Biotechnology and Jiangsu Key Laboratory of + + + + + + + + + + + + + + + + + + + + Conclusion + M +o +le c +u l a+r M+ e +d i c+i n +e, M+ e+d i+ca l+ S c+h o +o l +o f + + + + + N +a n+j i n+g +U n +iv +er s+it y +, N +a n+j i n+g +21 0+0 9+3 , +C h +in +a + + These results confirm that EPA is a tumor suppressor in human ovarian clear cell carcinoma + T +e l :+ 8 6+-0 2+5 8+3 5 +9 6 +84 5+ + + + + + + + + + + + cells and functions through a novel fatty acid receptor, GPR30, indicating a mechanistic + F +a x +: 8 6+- 0 +2 5 +83 5+9 6+8 4 +5 + + + + + + + + + + + linkage between omega-3 fatty acids and cancers. + E +-m +a i +l: z +h y +u e +@ n +ju +.e d +u . c+n + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + C +o -+co r+r e +sp o+n d+e n+c e+: C +h a+o -+Ju n+ L +i, P+h D+ + + + + + S +ta t+e K+e y+ L +a b +o r a+t o +ry +o f +P h +a r +m a+c e +u t i+ca l+ + + + + B +io t+e c +h n +o l o+g y+ a n+d + Ji a+n g +s u + K +ey +L a +b o +ra t+o r +y o +f + Key words + + + + + + + + + + + + + + + + + + + + + M +o +le c +u l a+r M+ e +d i c+i n +e, M+ e+d i+ca l+ S c+h o +o l +o f + + + + Eicosapentaenoic acid, GPR30, Ovarian neoplasms, + N +a n+j i n+g +U n +iv +er s+it y +, N +a n+j i n+g +21 0+0 9+3 , +C h +in +a + + Cell proliferation + T +e l :+ 8 6+-0 2+5 8+3 5 +9 6 +28 9+ + + + + + + + + + + + + F +a x +: 8 6+- 0 +2 5 +83 5+9 6+2 8 +9 + + + + + + + + + + + + E +-m +a i l+: l i+c j @+n j+u . e+d u+. c +n + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + C +o -+co r+r e +sp o+n d+e n+c e+: X +i L+i , +P h +D + + + + + + + + B +io l+o g +y S+c i +en c+e +In s+ti t +u t e+s , +C h +o n +g q +in g+ M +e d+i c +a l + + U +n i+v e +rs i+ty ,+ Y u+z h+o n+g ,+ C h+i n +a + + + + + + + + + T +e l :+ 8 6+-0 2+3 6+3 6 +6 4 +84 0+ + + + + + + + + + + + + F +a x +: 8 6+- 0 +2 3 +63 6+6 4+8 4 +0 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + E +-m +a i l+: l i+x i @+ s h+m +u .+ed u+. c +n + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + R +e c +ei v +e d + J u+l y + 7 ,+ 2 0+1 9 + + + + + + + + + + + + A +c c +e p +te d+ M+ a +rc h+ 3 +, 2 0+2 0+ + + + + + + + + + + P +u b +li s +h e +d O+n +li n +e M+ a+r c h+ 5 +, 2 +02 0+ + + + + + + + + + + + + + + + + + + + + + + + + + + + * +Y u +e Z+h a+o ,+ M +en +g- F+e i + Z h+a o+, a +n d + M +e i -+L i +n Y +a n+g + + c +o n +tr i b+u t+e d + e q+u a+l l y+ t o+ t h+i s + w o+r k+. + + + + + + + + + + + + + + + + + + + + + + + + + +

│ https://www.e-crt.org │ Copyright ⓒ 2020 by the Korean Cancer Association 815 This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Cancer Res Treat. 2020;52(3):815-829

Introduction which GPR30 is highly expressed and correlated with a poor prognosis [8]. In this study, we found that EPA suppressed ES2 ovarian The study of the association between diet and human can- clear cell carcinoma (OCCC) cells growth through GPR30 by cer has drawn considerable attention in recent decades, espe- activating caspase-3 and inducing apoptosis. Furthermore, cially which and how dietary factors might influence cancer GPR30 was expressed in patients with various ovarian can- risk. The new concept that changing diet can alter the epige- cers, and 300 µM EPA suppressed human ovarian cancer cell netic state of genes, thereby increasing or decreasing cancer growth via GPR30 without influencing normal cells. All risk, has been put forth in recent years [1]. Epidemiological these findings open a new front in the battle against ovarian studies have indicated a connection between various fatty cancer. acids (FAs) and ovarian cancer risk [1]. It has been reported that polyunsaturated fatty acids (PUFAs) can generate sev- eral omega-3 lipid metabolites and inhibit omega-6 arachi- donic acid production, thus playing a beneficial role [2]. Materials and Methods Previous study has shown the benefit of the omega-3 FAs on breast cancer incidence [3], but little information exists on the effects and mechanisms of eicosapentaenoic acid (EPA) in 1. Materials ovarian cancer cell growth. Ovarian cancer is the most fatal gynecologic malignancy FAs, FA-free bovine serum albumin (BSA), and forskolin worldwide [4]. It is mostly detected at an advanced stage, were purchased from Sigma-Aldrich (St. Louis, MO). Fura- leading to a high mortality rate. The pathogenesis of ovarian 2/AM was purchased from Dojindo (Kumamoto, Japan). G1 cancer has long been investigated but is still poorly under- and G15 were purchased from TOCRIS Bioscience (Ellisville, stood. Obviously, it is urgent to optimize or develop new MI) and Merck (San Diego, CA), respectively. YM254890 was therapeutic strategies in the management of this disease. from Yama-nouchi Pharmaceutical Co., Ltd. (Ibaraki, Japan). EPA, one kind of omega-3 polyunsaturated FA, has been observed to be significantly lower in the plasma of patients 2. Cell lines and culture with any of several cancers than in the plasma of healthy individuals [4], implying its considerable role in cancer deve- ES2 and SKOV3 cell lines were obtained from the Ameri- lopment. Several studies have suggested that EPA inhibits can Type Culture Collection (ATCC). ES2 and SKOV3 cells ovarian cancer cell growth and suppresses tumor metastasis were cultured in McCoy’s 5A medium supplemented with [5]. 10% fetal bovine serum and antibiotics. Cells were cultured Compelling evidence shows that FAs could achieve their in Opti-MEM (Gibco, Grand Island, NY) without fetal bovine biological effect by governing the fluidity and configuration serum. FAs were first dissolved in pure ethanol and then of membrane receptors, cell signaling pathways, transcrip- freshly diluted to the culture medium containing 1% BSA. tion of genes, and inflammatory response [6]. In recent years, The ethanol and BSA were 0.05% (v/v) and 0.1% (v/v), res- many studies have described and characterized several G pectively. protein–coupled receptors (GPRs) for FAs in cancer cells, which have been reported to be critical components of the 3. Cell proliferation assay and cytotoxicity assay sensing apparatus in the human body and to have specifici- ties for FAs of differing saturation degrees and chain lengths. For the assay of ES2 and SKOV3 cell proliferation, 5!103 1 (GPR40) and GPR120, which both cells per well in Opti-MEM with different types and concen- have G!q/11 as their G! subunit, are receptors for long-chain trations of FAs were seeded in triplicate, and cell numbers unsaturated FAs. GPR30, another novel FA receptor, is a were monitored after 24, 48, and 72 hours using the Cell 7-transmembrane estrogen receptor of the G!s family pre- Counting Kit-8 (Dojindo). The cytotoxicity of various FAs at dominantly localized in the endoplasmic reticulum. Many different concentrations on ES2 and SKOV3 cells was meas- steroids and some synthetic estrogen-receptor ligands func- ured with the Cytotoxicity LDH Assay Kit-WST (Dojindo). tion by binding to GPR30 [7]. Once bound, the production of All experiments were performed in triplicate wells for each cyclic AMP (cAMP) was promoted, causing epidermal condition and repeated three times independently. growth factor cleavage and epidermal growth factor receptor transactivation [7]. GPR30 is expressed not only in normal tissues, including the uterus, ovaries, and mammary glands, but also in several cancers, including ovarian cancer, in

816 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

4. Cell apoptosis assay G15, ovarian cancer xenograft models were generated in old female nude mice at 4 to 6 weeks of age by injecting 1!107 A total of 0.5!106 cells were cultured into 6-well culture ES2 cells into subcutaneous fat. Animals were divided into plates and cocultured with or without 300 µM EPA for 48 four groups with 10 per group. Nude mice bearing ovarian hours. For cell apoptosis analysis, cells were determined with tumors were received dimethyl sulfoxide (DMSO) in combi- the Annexin V/PI Apoptosis Detection Kit (Dojindo) and nation with MeOH as a control, EPA in combination with analyzed by a FACSCalibur (BD Bioscience, San Jose, CA). DMSO, MeOH in combination with G15 or EPA in combina- tion with G15. Mice were given EPA (4 µg) once daily, while 5. Cytosolic calcium determination G15 (10 µg) once daily from day 5. Tumor volumes were cal- culated as 1/2!length!width2. Ki67 and GPR30 were stained Cells were trypsinized in 3.5-cm Petri dishes, centrifuged, to assay the inhibitory effects of EPA and GPR30 expression, and resuspended in phosphate buffered saline (PBS) with 3 respectively. mM Fura-2/AM. Cells were treated with the Ca2+ indicator Fura-2/AM (5 µM) to measure [Ca2+]i. After washed for three 9. Tissue microarray construction and immunohistochem- times to remove extracellular Fura-2/AM, the Fura-2 was ical analysis detected using Felix fluorescence data acquisition as previ- ously shown [9]. The dose of YM254890 was based on previ- Ovarian cancer patient (n=173) samples were collected ous findings [10,11]. from the Department of Pathology of Nanjing Drum Tower Hospital. The tissue microarray (TMA) was made as previ- 6. cAMP measurement ously reported [15]. The immunohistochemical stains were assessed by three separate observers who did not know Cells were seeded and treated with G15 or GPR30 small patient characteristics. The classification of GPR30 expression interfering RNA (siRNA). The dose of G15 was based on level was based on the staining intensity of GPR30. All pro- Wang et al. [12] and Bai et al. [13], who found that the 8 µM tocols for humans were reviewed and approved (kyy2015- dose would be sufficient to obtain antagonistic effects on 38). The detailed information of TMA was provided in S1-S3 GPER1/GPR30 without affinity for other estrogen receptors. Tables. Twenty-four hours later, the cells were seeded in Opti-MEM medium for 5 hours and then treated with BSA, palmitic 10. Quantitative real-time PCR acid, or EPA. Cells were washed with PBS twice and frozen and thawed three times after treatment. The cAMPs were Quantitative real-time PCR was performed using SYBR quantified with an Enzyme Immunoassay Kit (Cayman Select Master Mix (Vazyme, Nanjing, China) in an ABI 7300 Chemical, Ann Arbor, MI). sequence detector. The results were normalized to the 18S rRNA expression level in each sample. The primers 7. Lipid-protein overlay assay were from PrimerBank (http://pga.mgh.harvard.edu/pri- merbank/). The nitrocellulose membrane was spotted with EPA (1, 3, and 5 mM) and then blocked in buffer as previously reported 11. Antibodies and immunoblotting [14]. Next, the membrane was incubated with purified GFP- GPR30 protein (100, 200, and 400 ng/mL) overnight at 4°C, The antibodies used were cleaved caspase-3, AKT, p-AKT, then scanned by Typhoon FLA9500 (GE Healthcare, Upp- ERK, p-ERK, phosphorylated protein kinase A (p-PKA) sub- sala, Sweden) to detect GFP. strate (Cell Signaling Technology, Danvers, MA), actin (Sigma-Aldrich), and GPR30 (Santa Cruz Biotechnology, 8. Ovarian cancer xenograft models Santa Cruz, CA).

Ovarian cancer xenograft models were generated in old 12. RNA interference female nude mice at 4 to 6 weeks of age by injecting 1!107 ES2 cells treated with or without adenoviral GPR30 shRNA The siRNA against GPR30 was GCACCUGUGGCUGAC- into subcutaneous fat. Animals were divided into four GAAUUU. ES2 cells were transfected with RNA interference groups with 10 per group. For the experimental group, 10 µL (RNAi) oligonucleotides using Lipofectamine RNAiMAX EPA was added to 90 µL aqueous vehicle. For the control (Invitrogen, Carlsbad, CA). group, 10 µL ethanol was added to 90 µL aqueous vehicle. Mice were given EPA (4 µg) once daily. For the treatment of

VOLUME 52 NUMBER 3 JULY 2020 817 Cancer Res Treat. 2020;52(3):815-829

A ES2 SKOV3 150 150 Palmitic acid 140 140 Palmitoleic acid ) ) Oleic acid

% 130 % 130 ( (

Linoleic acid n 120 n 120

o o Arnchidnic acid i i t 110 t 110 a a Linolenic acid r r

e e EPA

f 100 f 100 i i l l

o 90 o 90 r r p p

l 80 l 80 l l e e

C 70 C 70 60 60 50 50 0 50 100 150 200 300 0 50 100 150 200 300 Concentration (µM) Concentration (µM)

B ES2 SKOV3

m 2.0 Control m 2.0 Control n n

0 Palmitic acid 0 Palmitic acid 5 5 4 4

EPA EPA t t

a 1.5 a 1.5

e e c c n n a a

b 1.0 b 1.0 r r o o s s b b a a

e 0.5 e 0.5 v v i i t t a a l l e e

R 0 R 0 200 300 400 200 300 400 Concentration (µM) Concentration (µM)

C ES2 SKOV3 8 6 Palmitic acid Palmitoleic acid Linoleic acid . .

o 6 o Linolenic acid N N

l l 4 Oleic acid l l

e e Control c c Arnchidnic acid e 4 e v v

i i EPA t t a a l l 2 e e

R 2 R

0 0 0 1 2 3 0 1 2 3 Time (day) Time (day) Fig. 1. Eicosapentaenoic acid (EPA) inhibits cell proliferation in ovarian cancer cells. (A) Dose-response curves of palmitic acid, palmitoleic acid, oleic acid, linoleic acid, arachidonic acid, linolenic acid, and EPA in ES2 cells and SKOV3 cells. (B) The toxicity of palmitic acid and EPA at different concentrations on ES2 and SKOV3 cells. (C) Cell proliferation of ES2 and SKOV3 cells in time course for various fatty acids in ES2 cells and SKOV3 cells were treated with 300 µM fatty acids. (Con- tinued to the next page)

13. Adenoviral expression vectors 14. Statistics

The GPR30 shRNA sequence was GCACCTGTGGCTGA- Data from at least three independent experiments were CGAATTT. Adenovirus was amplified and purified using analyzed and are expressed as means±standard deviation. Sartorius Adenovirus Purification kits (Sartorius, Germany). The nonpaired Student’s t test was used for these analyses. A difference was considered significant at p < 0.05, p < 0.01,

818 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

D E Control 4 20 6 Control

10 ) 2.45% 0.010% EPA l % ( e 3 l 15 v 10 l e e l

c

2 A 4 c

10 i 10 N t o R 1 t m p

10 o 5 e p 97.5% 0.021% v i t 0 A 2

10 a 0 1 2 3 4 l

10 10 10 10 10 0 e R I

P Control EPA EPA 104 0 6.86% 8.06% x a 2 xl n Control EPA a im m cl l- vi 3 B B u B c vi 10 P B ur Cleaved- S 102 caspase-3 Actin 101 78.7% 6.35% 100 100 101 102 103 104 AnnexinV

F G 2.0 30 e Control Control s

n EPA EPA a ) b r %

o 1.5 (

s l l

b 20 m e a

n c

f

0 o c

i 5 l 1.0 t 4 e

o t t v a p e l

o 10 e p

v 0.5 A i t a l e R 0 0 OEC ES2 SKOV3 OEC ES2 SKOV3

Fig. 1. (Continued from the previous page) (D) Apoptosis by EPA in ES2 cells. The cells were treated with or without 300 µM EPA for 48 hours. Apoptosis was detected by cleaved caspase-3. PI, propidium iodide. (E) The expression of proapoptotic genes and antiapoptotic genes in ES2 cells treated with or without 300 µM EPA for 48 hours was analyzed by quantitative reverse transcription polymerase chain reaction. (F) The toxicity of EPA on normal ovarian epithelial cells (OEC), ES2, and SKOV3 cells. (G) The apoptosis rate of normal OEC, ES2, and SKOV3 cells treated with EPA. Values are presented as mean±standard deviation from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

and p < 0.001. R was used for the correlation analysis of Results GPR30 expression level with overall survival of ovarian can- cers patients. Then packages “survival” and “survminer” were used for survival analysis using Kaplan-Meier method 1. EPA inhibits cell proliferation in ovarian cancer cells and generate the survival curve. FAs have an effect on cancer by altering cell membrane 15. Ethical statement synthesis and saturation and via cholesterol lipid hormones that affect the function and distribution of signaling macro- All animal procedures were carried out in accordance with molecules [6]. Thus, we speculated that various FA treat- the approval of the Animal Care (#CS20) and Use Committee ments might influence different aspects of cancer develop- at the Model Animal Research Center of Nanjing University ment in ovarian cancer cells. Ovarian carcinoma is divided in Nanjing. into four major histological types: clear cell, mucinous, serous, and endometrioid [16]. OCCCs makes up 25% of ovarian car-

VOLUME 52 NUMBER 3 JULY 2020 819 Cancer Res Treat. 2020;52(3):815-829

A B 4 Palmitic acid 4 Palmitic acid EPA EPA

x Control x Palmitic acid+Ym u u

l 3 BSA l 3 EPA+Ym f f n n i i

+ + a a C C 2 2 e e v v i i t t a a l l

e 1 e 1 R R

0 0 0 50 100 150 200 0 50 100 150 200 Time (sec) Time (sec)

C 2.0 Control BSA PAL ns .

o 1.5 EPA N

l l e c

e 1.0 v i t a l e

R 0.5

0 Vehicle Ym

D E EPA+vehicle 4 5 ns 2.0 EPA+vehicle

10 ) 2.28% 1.48% EPA+Ym % l

( 4

e

3 l l 10 v e e 1.5 l

c

3

2 A c

10 i t N o R

t 2

1 p

m 1.0

10 o e p

94.2% 2.02% v 1 i A 0 t

10 a 0 1 2 3 4 l

10 10 10 10 10 0 e 0.5 R I

P EPA EPA EPA+Ym +vehicle +Ym 104 0 2.76% 1.45% x a 2 xl n a im m cl l- vi 3 B B u B c vi 10 EPA+vehicle EPA+Ym P B ur Cleaved- S 102 caspase-3 101 Actin 93.6% 2.22% 100 100 101 102 103 104 Annexin V-FITC Fig. 2. G protein–coupled receptor 30 (GPR30) acts as a novel eicosapentaenoic acid (EPA) receptor in ES2 cells. (A, B) ES2 cells were loaded with 3 µM of the fluorescent calcium probe fura-2A/M. (A) Cells were stimulated in phosphate buffered saline alone (control), 300 µM palmitic acid bound to bovine serum albumin (BSA) (0.1%), 300 µM EPA bound to BSA (0.1%), or BSA alone. (B) ES2 cells were incubated with YM254890 (Gq protein inhibitor) at 37°C for 5 minutes before the addition of fatty acids. The arrows indicate the onset of stimulation. (C) Cell proliferation of ES2 cells treated with 0.1% BSA, 300 µM palmitic acid (PAL), or 300 µM EPA in the absence or presence of YM254890 for 24 hours. (D) Apoptosis of ES2 cells treated with 300 µM EPA in the absence or presence of YM254890. Apoptosis was detected by cleaved caspase-3. PI, propidium iodide. (E) The expression of proapoptotic genes and antiapoptotic genes in ES2 cells treated with 300 µM EPA in the absence or presence of YM254890 for 48 hours was analyzed by quantitative reverse transcription polymerase chain reaction (RT- PCR). (Continued to the next page)

820 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

F G 80 s 40 R P n G i

e e t l o 60 v 30 r e l p

n g o i m / s l

40 s 20 o e r m p p

x P e

20 e 10 M v i A t c a l e

0 R 0 l n L 3 4 0 2 2 0 9 0 ro li A A PA R 8 4 R 3 2 1 3 t o BS P E A R R A 1 1 1 R on sk F P P F R R R P C r F G G F P P P G Fo G G G

H I 0

3 0

L R 3 A P 80 EPA P G R P 0 G

3 e R l P v 60 e G l

n o i

s 40 s e r p x e 20 e v i t a l e 0 R SKOV3 ES2 CAOV3 A2780 Fig. 2. (Continued from the previous page) (F) Intracellular cAMP by EPA in ES2 cells. Cells were treated with 0.1% BSA, 300 !M PAL, or 300 µM EPA for 24 hours for cAMP measurement. (G) The relative expression of various GPRs in ES2 cells by RT-PCR. Data were normalized to the expression level of free fatty acid receptor 3 (FFAR3). (H) Interaction of GPR30 and EPA was detected by a protein lipid overlay assay. Increasing amounts of EPA were incubated with increasing concentrations of purified GFP-GPR30 harvested from ES2 cell lysates. (I) Relative expression of GPR30 was measured by quantitative RT- PCR in different ovarian cancer cell lines. Data were normalized to the expression level in SKOV3. Values are presented as mean±standard deviation from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant.

cinoma in Japan, and the lethality rate of adenocarcinoma is µM (Fig. 1B). Thus, 300 µM was an appropriate concentration also high [16]. Compared with other types of ovarian carci- for use in subsequent experiments. The pro-proliferation nomas, OCCCs have a poorer prognosis and are more fre- effect of palmitic acid and the inhibitory effect of EPA on ES2 quently platinum resistant [17]. Here, we selected the ES2 and SKOV3 cells were also time-dependent. Palmitic acid OCCC and SKOV3 adenocarcinoma cell lines for study. ES2 treatment increased cell growth in both ovarian cancer cell and SKOV3 were first treated with several FAs at different lines. However, the growth of ES2 cells was significantly doses for 24 hours (0-300 µM). We found that palmitic acid, decreased when treated with EPA compared to that of SKOV3 a saturated FA, obviously enhanced cell growth, while EPA cells (Fig. 1C), suggesting that EPA had a much more pro- inhibited ES2 and SKOV3 cell proliferation in a dose-depen- nounced effect in ES2 cells than in SKOV3 cells. dent manner (Fig. 1A), which is in accordance with the pub- As previously reported, n-3 PUFAs (docosahexaenoic acid lished anticancer effects of EPA [18]. However, EPA promo- or EPA) inhibit colorectal cancer cell proliferation by induc- ted cell proliferation in CAOV3 and A2780 cell lines, indicat- ing cell apoptosis [19]. Therefore, we detected cell apoptosis ing the antitumor effect of EPA may be cell line specific event caused by EPA treatment in ES2 cells. The flow cytometry (S4 Fig.). We also analyzed the toxicity of palmitic acid and results showed that apoptotic cells were approximately EPA on ES2 and SKOV3 cells by using a cytotoxicity assay. 14.4% in EPA-treated ES2 cells, more than that in control We found that cell toxicity affected ES2 cells at a 400 µM con- (0.03%). The protein expression analysis also confirmed that centration, while the effect on SKOV3 cells was minor at 400 EPA led to cleaved caspase-3 activation (Fig. 1D). mRNA

VOLUME 52 NUMBER 3 JULY 2020 821 Cancer Res Treat. 2020;52(3):815-829

A A 2.0 siNC l

e siGPC30 v e l

n 1.5

o ns i ns s s s R e P r p G 1.0

x f e o

e v i t 0.5 a l e R 0 B GPR30 GPR120 GPR40 4 e d a

r B g 3 Control n 3 o

i BSA s

s PAL e 2 . r

o EPA p x N

l e 2 l

0 e 3 c

1 R e P v i G t a l 1 0 e R C rs C C C C rs S e S M C E e G c G C c rs H an L an e c c c II I an e e c 0 p yp d ty t ie e e if siNC siGPR30 d ad ss ra r a g -g cl h- w n ig lo u h r d r e an C e th th O 4 Control O BSA PAL .

C o 3 EPA N

l

1.00 l e c

e 2 y t v i i l t i 0.75 a b l a e b R 1 o r p

0.50 l

a GPR30 grade v i 0

v 0 r

u 0.25 1 NC G15 S 2 3 Fig. 4. G protein–coupled receptor 30 (GPR30) is involved 0 in eicosapentaenoic acid (EPA)–induced cell antiprolifer- 0 10 20 30 40 50 Time (mo) ation and proapoptosis in ES2 cells. (A) The efficacy of GPR30 small interfering RNA (siRNA). (B) Cell prolifera- Fig. 3. G protein–coupled receptor 30 (GPR30) is expre- tion of ES2 cells treated with 0.1% bovine serum albumin ssed in patients with various ovarian cancers. The expres- (BSA), 300 µM palmitic acid (PAL), or 300 µM EPA with sion level of GPR30 in patients with various ovarian or without GPR30 siRNA treatment. (C) Cell proliferation cancers according to a tissue microarray and immunohis- of ES2 cells treated with 0.1% BSA, 300 µM PAL, or 300 µM tochemistry. (A) Immunohistochemical staining of GPR30 EPA with or without G15. (Continued to the next page) in ovarian tissues with various ovarian cancers. (B) The expression grade of GPR30 in patients with various ovar- ian cancers (n=173). HGSC, high-grade serous carcinoma; LGSC, low-grade serous carcinoma; MC, mucinous carci- noma, CCC, clear cell carcinoma; EC, endometrioid ade- expression analysis confirmed that EPA-induced proapop- nocarcinoma. (C) The correlation analysis of GPR30 expre- totic genes activation and antiapoptotic genes inhibition in ssion level with overall survival of patients (The grade ES2 cells (Fig. 1E). Importantly, the cytotoxicity and apopto- score 0 to 3 represents the intensity of GPR30 from lowest sis of normal ovarian epithelial cells were not affected by 300 to highest). µM EPA treatment (Fig. 1F and G). Together, these data

822 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

D E EPA+siNC 20 6 104 ) Control 8.86% 10% % ( EPA l

l l 3 15 e EPA+siNC v 10 e e c l EPA+siGPR30

c i 2 A 4 10 t 10 N o t R 1 p m o

10 p 5 e A 73.2% 7.9% v i

0 t 2

10 a 0 1 2 3 4 l

10 10 10 10 10 0 e R I

P EPA EPA EPA+siGPR30 +siNC +siGPR30 104 0 9.73% 6.89% x m a l2 xl in a i m c l- iv 3 30 B B u B c v 10 R P B ur C P S l N G 2 o si si 10 tr + + n A A A o P P P 101 C E E E 82.1% 1.25% Cleaved- 100 caspase-3 100 101 102 103 104 Actin AnnexinV Fig. 4. (Continued from the previous page) (D) Apoptosis of ES2 cells treated with 300 µM EPA and control RNA interference (RNAi) or 300 µM EPA and GPR30 RNAi for 48 hours. Apoptosis was detected by cleaved caspase-3. PI, propidium iodide. (E) The expression of proapoptotic genes and antiapoptotic genes in ES2 cells treated with 300 µM EPA or 300 µM EPA and GPR30 siRNA for 48 hours was analyzed by quantitative reverse transcription polymerase chain reaction. Values are pre- sented as mean±standard deviation from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

indicate that EPA induces ovarian cancer cell apoptosis. Similarly, we studied the effect of YM254890 on palmitic acid proliferation as well as the EPA antiproliferative effect 2. GPR30 acts as a novel EPA receptor in ES2 cells in ES2 cells. As shown in Fig. 2C, YM254890 decreased pal- mitic acid-induced proliferation by 15% (Fig. 2C). However, Previous findings suggest the receptors of long-chain FAs YM254890 did not affect EPA-induced antiproliferation or are GPR40 and GPR120 [20], and these GPR agonists could proapoptosis effects (Fig. 2C and D). Consistent with the data induce cell proliferation [21]. There are four subfamilies of ! above, the expression of proapoptotic genes, antiapoptotic subunits of G : G!s, G!i, G!q, and G!12. GPR40 and genes, and cleaved caspase-3 as influenced by EPA was not GPR120 couple to the !q subfamily of G proteins, which can changed by YM254890 (Fig. 2E). The results above suggest be inhibited by YM254890 [10], increasing the downstream that GPR40 or GPR120, whose subunit is G!q, is not involved 2+ effector Ca [22]. To determine whether GPR40 and GPR120 in the anticancer effect of EPA in ES2 cells. Unlike the G!q 2+ mediate the effects of palmitic acid and EPA in ES2 cells, we family, whose downstream effect is increasing [Ca ], G!s 2+ measured changes in [Ca ] in cells following palmitic acid increases but G!i decreases the level of cAMP [22]. To identify or EPA treatments in the absence and presence of the Gq which G! subunit functioned when ES2 cells were treated inhibitor YM254890, which can completely block the Gq with EPA, we first measured the production of cAMP in ES2 pathways [10]. The modest increase in Ca2+ that we detected cells treated with palmitic acid or EPA. The results showed could be attributed to G"#, which could have activated PLC" that EPA but not palmitic acid increased cAMP compared to on the cell membrane and finally cause Ca2+ release from the vehicle (Fig. 2F), suggesting that the EPA receptor is coupled endoplasmic reticulum. Without YM254890 treatment, to G!s protein. palmitic acid and EPA activated [Ca2+] increases of 4- and Based on the above results, we hypothesized that there

1.5-fold, respectively (Fig. 2A). Interestingly, with YM254890 exists some GPR whose G!s subunit plays a dominant role in treatment, the increase in [Ca2+] by EPA did not change, while the EPA effect in ES2 cells. Thus, we detected various GPR the increase in [Ca2+] by palmitic acid was inhibited (Fig. 2B). proteins expression levels in ES2 cells. Quantitative PCR

These results indicate that palmitic acid functions mainly analysis showed that GPR30, coupled to G!s protein [23], was through the Gq pathway, while EPA might have its effects significantly higher than other GPRs (Fig. 2G). We also through other G! pathways. detected and graded expression of GPR30 in patients with

VOLUME 52 NUMBER 3 JULY 2020 823 Cancer Res Treat. 2020;52(3):815-829

A siNC siGPR30 l l ro ro t A L t A L on S A A on S A A C B P EP C B P EP GPR30

p-AKT AKT p-ERK

ERK

Actin

B 5 Control 4 Control BSA BSA T PAL K PAL

K 4 R 3 A EPA E EPA / / T K K 3 R E A - - p p 2

e e v v 2 i i t t a a l l e e 1 R R 1

0 0 siNC siGPR30 siNC siGPR30

C D 60 Control 60 Control

n EPA n EPA i i e e t t o o r r p p 40 40 g g m m / / l l o o m m p p 20 20 P P M M A A c c 0 0 NC G15 siNC siGPR30 Fig. 5. Eicosapentaenoic acid (EPA) participates in the G protein–coupled receptor 30 (GPR30)–cAMP–protein kinase A sig- naling pathway in ES2 cells. (A) After 24 hours of serum starvation in Opti-MEM, cells were incubated for 24 hours in Opti- MEM and stimulated with phosphate buffered saline alone (control), 0.1% bovine serum albumin (BSA), 300 µM palmitic acid (PAL) bound to BSA (0.1%) or 300 µM EPA bound to BSA (0.1%) with or without GPR30 siRNA treatment. (B) p- AKT/AKT and p-ERK/ERK were analyzed by Western blotting and quantified by ImageJ. (C, D) ES2 cells were treated with or without 300 µM EPA in the absence and presence of G15 (C) or GPR30 siRNA (D) for 24 hours for cAMP measurement. (Continued to the next page)

various ovarian cancer, including high-grade serous carcino- ovarian cancers, highly in LGSC while lowly in MC (Fig. 3A mas, low-grade serous carcinomas (LGSC), mucinous carci- and B). Furthermore, the correlation analysis of GPR30 nomas (MC), clear cell carcinomas (CCC), and endometrioid expression level with overall survival of ovarian cancers adenocarcinoma. The immunohistochemistry results showed patients showed that patients with higher GPR30 expression that GPR30 was widely expressed in patients with various level (grade 3) showed shorter survival time, while patients

824 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

E F 5 30 R 0 P 3 G y i t

R i P s 4 5 + v G 1 1 1 in i ol si G G G ol t tr + + + + k c n A A A A A s a o P P P P P or 3

C E E E E E F A K P

e 2 v i t a l

e 1 R

p-PKA 0 substrate ol A 0 5 1 0 in tr P 3 1 G 3 ol n E PR G + PR k o + A G rs C iG PA EP i o +s E +s F A 1 EP +G A EP

Fig. 5. (Continued from the previous page) (E) Phospho-(Ser/Thr) substrate was detected by Western blotting in ES2 cells treated with EPA, EPA+GPR30 siRNA, EPA+G15, EPA+G1 (500 nM, the specific agonist of GPR30) or EPA+G1+GPR30 siRNA. (F) PKA activities were quantified by ImageJ, and the data are expressed as PKA activity (n=3). Values are presented as mean±standard deviation from three independent experiments. **p < 0.01, ***p < 0.001.

with lower GPR30 expression level (grade 0) showed longer proapoptotic effect of EPA on ES2 cells could not be reversed survival time (Fig. 3C, S1-S3 Table).To further verify whether when GPR119 was knocked down (S5 Fig.), indicating

EPA was a ligand of GPR30 in ES2 cells, we used a modified GPR119, another G!s coupled receptor, did not play an essen- protein lipid overlay assay [14]. As shown in Fig. 2H, the tial role in mediating the anticancer effect of EPA in ES2 cells. interaction between EPA and GPR30 existed and was dose- Altogether, the above results establish that GPR30 is essential dependent. Moreover, the relative expression level of GPR30 in mediating the effect of EPA in ES2 cells. was markedly lower in SKOV3 cells compared with ES2 cells, A2780 cells, and CAOV3 cells, indicating that GPR30 func- 4. EPA inhibits cell proliferation through the GPR30- tions differently in different cell lines (Fig. 2I). cAMP-PKA signaling pathway in ES2 cells

3. GPR30 mediates the EPA effect in ES2 cells The phosphoinositide 3-kinase (PI3K)/AKT signaling pathway is the most common pathway related to cancer pro- To determine whether GPR30 mediates the effect of EPA liferation [24]. In recent years, numerous reports have demon- in ES2 cells, we knocked down GPR30 using RNAi and strated that EPA is implicated in several signaling pathways, found that GPR30 expression was reduced by 80% without such as RAS/ERK/C/EBP" and nuclear factor #B, contribut- changes in GPR40 or GPR120 (Fig. 4A). The decrease in cell ing to tumor development and [25]. To evalu- number following EPA treatment was significantly reversed ate whether EPA affects cell proliferation through these in ES2 cells treated with GPR30-specific siRNA (Fig. 4B). Sim- cellular response pathways, we investigated phosphorylated ilarly, we used G15, the specific antagonist of GPR30, to test AKT and ERK1/2 in ES2 cells treated with or without EPA. the EPA effect in ES2 cells. As shown in Fig. 4C, the inhi- Interestingly, palmitic acid stimulated AKT and ERK1/2 bitory effect of EPA was again impaired. The siRNA and phosphorylation, while EPA inhibited their phosphorylation inhibitor results indicated that EPA-induced antiprolifera- after 24 hours of treatment (Fig. 5A). To further determine tion in ES2 cells is mediated through GPR30. whether GPR30 was involved in the activation of AKT and Furthermore, we knocked down GPR30 to examine its ERK1/2, we knocked down GPR30 as mentioned above. The effect on the EPA proapoptotic action in ES2 cells. The num- phosphorylation levels of AKT and ERK1/2 in response to ber of apoptotic cells dramatically declined when ES2 cells palmitic acid did not obviously change, and that in response were treated with GPR30 siRNA compared with negative- to EPA was dramatically increased when ES2 cells were control siRNA (Fig. 4D). Additionally, the expression of treated with GPR30 siRNA (Fig. 5A and B). These findings apoptotic genes as influenced by EPA was reversed when suggest that the development of signaling pathways via GPR30 was knocked down (Fig. 4D and E). However, the GPR30 includes PI3K-AKT and ERK.

VOLUME 52 NUMBER 3 JULY 2020 825 Cancer Res Treat. 2020;52(3):815-829

AB shLacZ shLacZ shGPR30 shGPR30 shLacZ+ethanol +ethanol +EPA +ethanol +EPA

shLacZ+EPA H&E shGPR30+ethanol

shGPR30+EPA Ki67 1 cm

GPR30

C D 150 shLacZ+ethanol 0.5 shLacZ+EPA ) 3 shGPR30+ethanol ) 0.4 m g

shGPR30+EPA (

m t (

100 h e g

i 0.3 ns m e u l w

o r v

o 0.2 r o 50 m u m T u

T 0.1

0 0 475 6 8 9 10 11 12 13 14 shLacZ shLacZ shGPR30 shGPR30 Days after injection +ethanol +EPA +ethanol +EPA

EF DMSO+MeOH DMSO+EPA G15+MeOH G15+EPA DMSO+MeOH H&E DMSO+EPA

G15+MeOH Ki67 G15+EPA 1 cm GPR30

Fig. 6. Eicosapentaenoic acid (EPA) blocks tumor growth via G protein–coupled receptor 30 (GPR30) in mouse xenografts. (A, B) Nude mice bearing ovarian tumors (ES2 cells) were received ethanol in combination with LacZ shRNA as a control, EPA in combination with LacZ shRNA, ethanol in combination with GPR30 shRNA or EPA in combination with GPR30 shRNA. (A) Xenograft tumors (scale bar=1 cm). (B) Ki67 and GPR30 expression (scale bar=50 µm). Tumor volume (C) and tumor weight (D) in (A). (E, F) Nude mice bearing ovarian tumors (ES2 cells) were received dimethyl sulfoxide (DMSO) in combination with MeOH as a control, EPA in combination with DMSO, MeOH in combination with G15 or EPA in combi- nation with G15. (E) Xenograft tumors (scale bar=1 cm). (F) Ki67 and GPR30 expression (scale bar=50 µm). (Continued to the next page)

The results in Figs. 2 and 4 show that EPA inhibits ES2 cell EPA via GPR30 in ES2 cells. Thus, we knocked down GPR30 growth through GPR30. Due to the inhibitory effect on can- in ES2 cells by GPR30 siRNA and G15 as described above. cer cell growth by G!s-cAMP-PKA signaling activation [26], The cAMP production promoted by EPA was attenuated fol- the G!s-cAMP-PKA signaling pathway may be affected by lowing downregulation of GPR30 (Fig. 5C and D). Addition-

826 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

G H 2,000 2.0 ) 3 ) m g

1,500 ( 1.5

m t (

h e g i m e u l

1,000 w 1.0

o r v

o r o m u m

500 T 0.5 u T

0 0 DMSO DMSO G15 G15 DMSO DMSO G15 G15 +MeOH +EPA +MeOH +EPA +MeOH +EPA +MeOH +EPA Fig. 6. (Continued from the previous page) Tumor volume (G) and tumor weight (H) in (E). Values are presented as mean±stan- dard deviation from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

ally, EPA increased PKA activity compared to the treatment proliferation significantly (Fig. 6E-G). The combination of vehicle, as detected by the phospho-(Ser/Thr) PKA sub- EPA and G15 was unable to inhibit tumor growth, as G15 strates. Notably, G1, the specific agonist of GPR30, also inhibited antitumor effect of EPA by impairing GPR30 activ- increased PKA activity. Moreover, PKA activity stimulated ity. These in vivo results also suggest that EPA inhibits tumor by EPA was suppressed by GPR30 knockdown and G15. growth via GPR30 in human ovarian clear cancer cells. However, PKA activity stimulated by EPA, G1, and siGPR30 did not change compared to its activity stimulated by EPA and siGPR30 (Fig. 5E and F). These results indicate that GPR30-cAMP-PKA signaling is involved in the EPA effect in Discussion ES2 cells.

5. EPA blocks tumor growth via GPR30 in mouse xeno- Extensive research implies that dysregulation of lipid grafts metabolism is correlated with ovarian cancer progression [27]. EPA, an n-3 polyunsaturated FA, has anticancer effects To confirm the above observations, we also conducted a in many cancer cells, such as colorectal cancer [28], breast mouse xenograft model to further verify the role of GPR30 cancer [3], pancreatic cancer [28], and ovarian cancer [5]. In in ES2 cells. In accordance with previous findings, EPA alone our study, EPA-induced apoptosis in ES2 OCCC cells follow- inhibited tumor growth significantly (Fig. 6A and E). To fur- ing induction of antiproliferation through GPR30, a novel ther confirm the function of GPR30, we knocked down EPA receptor. Additionally, EPA stimulated the activation GPR30 using adenovirus in a mouse xenograft model. As of caspase-3, blunted the activation of AKT and ERK1/2 and shown in Fig. 6A, EPA dramatically decreased tumor growth, functioned through the GPR30-cAMP-PKA signaling path- while EPA was unable to inhibit tumor growth when GPR30 way. was knocked down. hematoxylin and eosin (H&E) staining Classical free fatty acid receptors, such as GPR40, and and labeling for the cancer proliferation marker Ki67 also GPR120, might also mediate the function of EPA in ovarian showed that EPA lost its inhibitory function when ES2 cells cancer cells. Since G!q is the ! subunit of both GPR40 and were treated with GPR30 shRNA (Fig. 6B). The change in GPR120, whose activation leads to a rapid increase in Ca2+, tumor volume and weight supported the above results we detected the Ca2+ concentration after adding EPA, and an (Fig. 6C and D). Interestingly, we found that the deletion of approximately 1.5-fold increase was observed. Importantly, GPR30 in ES2 cells decreased tumor growth in mice, suggest- YM254890, a specific inhibitor of the G!q unit, did not inhibit ing that GPR30 is associated with tumor development. the increase in Ca2+ caused by EPA, suggesting that neither Therefore, we antagonized GPR30 activity by treatment with GPR40 nor GPR120 is the specific receptor of EPA. We found GPR30 antagonist G15 in the mouse xenograft model. a novel EPA receptor, GPR30, in ovarian cancer cells, con- Although the tumor weight had no significant change due firmed by a modified protein lipid assay [14], thus broaden- to individual variation (Fig. 6H), the tumor size and H&E ing the concept of cancer metabolism. GPR30, which was staining showed that EPA or G15 alone both inhibited cell once thought to be an , has been implicated

VOLUME 52 NUMBER 3 JULY 2020 827 Cancer Res Treat. 2020;52(3):815-829

in both rapid and transcriptional events in response to estro- the chemotherapy resistance of OCCC through GPR30. gen. Ligands of GPR30 are mainly steroids and some syn- Answering these questions could provide a theoretical basis thetic estrogen-receptor ligands, and the pro-proliferation for the clinical treatment of OCCC. effects of E2 in hormone-related tumors are well known. When we blocked GPR30 expression by shRNA in vivo, we Electronic Supplementary Material also blocked the pro-proliferation effects of E2 because of the lack of ER! and ER" in ES2 cells. Therefore, the volume and Supplementary materials are available at Cancer Research and weight of these tumors were significantly decreased, as Treatment website (https://www.e-crt.org). shown in Fig. 6D. Above all, we first proved that besides steroids, EPA is also a ligand for GPR30. Conflicts of Interest Oxidative stress has been reported to affect cancer cell development. For example, reactive oxygen species (ROS) Conflict of interest relevant to this article was not reported. participate in cancer cell progression and proliferation, cell apoptosis, and energy metabolism [29]. Previous reports Acknowledgments showed that EPA mainly causes ROS-induced apoptosis [28]. The cell death, which mainly occurs in the late apoptosis We thank Prof. Zhongren Ding from Shanghai Medical College phase, is due to the intracellular ROS-induced caspase-8 of Fudan University for kindly providing YM254890 from Yama- activation [30]. Other evidence has demonstrated that EPA nouchi Pharmaceutical Co., Ltd. (Ibaraki, Japan). This work was also induces autophagic cell death [28]. Here, we show that supported by the Chinese National Science Foundation (31601153 EPA increased the activation of caspase-3 and stimulated the to Y.Z., 81770861 and 31571401 to X. Li), the Nature Science Foun- expression of apoptotic genes. We found that EPA promoted dation of Jiangsu Province (BK20160619 to Y.Z.), the Social Devel- a less antiproliferative signal in ES2 cells but an insensitive opment Fund of Jiangsu Province (BE2017708 to C.L.), the Fun- response in SKOV3 cells treated with GPR30 siRNA. Consid- damental Research Funds for the Central Universities (14380269, ering that different types of ovarian cancer have distinctive 14380343, 14380469 to Y.Z.), and by China Postdoctoral Science morphologic and molecular-genetic features, GPR30 may be Foundation (2016M601778 to Y.Z.). Chong Qing Science and Tech- coupled to G proteins more tightly in ES2 cells than in SKOV3 nology Foundation (cstc2018jcyjAX0232 to X. Li), Chong Qing Edu- cells. Another reason may be that the enrichment of GPR30 cation Foundation (KJZD-K201800402 to X. Li). protein varies between ovarian cancer cells and that SKOV3 cells have a low expression level of GPR30 (Fig. 2I). Besides, Author Details EPA exists pro-proliferation effect on CAOV3 and A2780 cell lines, indicating the antitumor effect of EPA may be cell-line 1State Key Laboratory of Pharmaceutical Biotechnology and Jiangsu specific event. Therefore, further investigation about the Key Laboratory of Molecular Medicine, Medical School of Nanjing effect of EPA on other CCC cells should be required. University, Nanjing, 2Department of Obstetrics and Gynecology, Our observations suggest that in ES2 OCCC cells, EPA Xiamen Chang Gung Hospital, Xiamen, 3Obstetrics and Gynecology 4 functions through the GPR30-G!s-cAMP-PKA signaling path- Hospital, Fudan University, Shanghai, Department of Pathology, way. Originally, phosphorylation of ERK1/2 is increased by Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing G"#, which is dissociated from G!. Previous work also demon- University Medical School, Nanjing, 5Biology Science Institutes, strated that PKA stimulated by the ! subunit of GPR30 could Chongqing Medical University, Yuzhong, China decrease the phosphorylation of ERK1/2 [26]. We think that crosstalk between them may account for the decreased phos- phorylation level in our study. Further investigation will elu- cidate the specific functions of signaling pathways activated by EPA, including autophagy, metabolism, cell cycle. Overall, our study provides evidence that GPR30 is a novel receptor of EPA in ES2 cells that mediates its anticancer effects. EPA might inhibit proliferation by inducing apopto- sis. Additionally, EPA might influence the GPR30-ERK or GPR30-AKT signaling pathway. In our future work, we will examine the mechanism of the effects that EPA has on other backgrounds. In addition, because other researchers have provided evidence that EPA can improve the sensitivity to chemotherapy [2], we wondered whether EPA could improve

828 CANCER RESEARCH AND TREATMENT Yue Zhao, GPR30 Mediates Anticancer Effects of EPA

References

1. Sapienza C, Issa JP. Diet, nutrition, and cancer epigenetics. resistance to platinum-based chemotherapy. Cancer. 2000;88: Annu Rev Nutr. 2016;36:665-81. 2584-9. 2. Rose DP, Connolly JM. Omega-3 fatty acids as cancer chemo- 17. Landrum LM, Java J, Mathews CA, Lanneau GS Jr, Copeland preventive agents. Pharmacol Ther. 1999;83:217-44. LJ, Armstrong DK, et al. Prognostic factors for stage III epithe- 3. Velentzis LS, Keshtgar MR, Woodside JV, Leathem AJ, Tit- lial ovarian cancer treated with intraperitoneal chemotherapy: comb A, Perkins KA, et al. Significant changes in dietary a Gynecologic Oncology Group study. Gynecol Oncol. 2013; intake and supplement use after breast cancer diagnosis in a 130:12-8. UK multicentre study. Breast Cancer Res Treat. 2011;128:473- 18. Kwan HY, Fu X, Liu B, Chao X, Chan CL, Cao H, et al. Subcu- 82. taneous adipocytes promote melanoma cell growth by activat- 4. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA ing the Akt signaling pathway: role of palmitic acid. J Biol Cancer J Clin. 2010;60:277-300. Chem. 2014;289:30525-37. 5. Brinton EA, Mason RP. Prescription omega-3 fatty acid prod- 19. Sun H, Hu Y, Gu Z, Owens RT, Chen YQ, Edwards IJ. Omega- ucts containing highly purified eicosapentaenoic acid (EPA). 3 fatty acids induce apoptosis in human breast cancer cells and Lipids Health Dis. 2017;16:23. mouse mammary tissue through syndecan-1 inhibition of the 6. Santos CR, Schulze A. Lipid metabolism in cancer. FEBS J. MEK-Erk pathway. Carcinogenesis. 2011;32:1518-24. 2012;279:2610-23. 20. Kotarsky K, Nilsson NE, Flodgren E, Owman C, Olde B. A 7. Maggiolini M, Picard D. The unfolding stories of GPR30, a human activated by free fatty acids and new membrane-bound estrogen receptor. J Endocrinol. 2010; thiazolidinedione drugs. Biochem Biophys Res Commun. 204:105-14. 2003;301:406-10. 8. Heublein S, Mayr D, Friese K, Jarrin-Franco MC, Lenhard M, 21. Gu Z, Suburu J, Chen H, Chen YQ. Mechanisms of omega-3 Mayerhofer A, et al. The G-protein-coupled estrogen receptor polyunsaturated fatty acids in prostate cancer prevention. Bio- (GPER/GPR30) in ovarian granulosa cell tumors. Int J Mol Sci. med Res Int. 2013;2013:824563. 2014;15:15161-72. 22. Dorsam RT, Gutkind JS. G-protein-coupled receptors and can- 9. Mazzuca MQ, Mata KM, Li W, Rangan SS, Khalil RA. Estrogen cer. Nat Rev Cancer. 2007;7:79-94. receptor subtypes mediate distinct microvascular dilation and 23. Prossnitz ER, Oprea TI, Sklar LA, Arterburn JB. The ins and reduction in [Ca2+]I in mesenteric microvessels of female rat. outs of GPR30: a transmembrane estrogen receptor. J Steroid J Pharmacol Exp Ther. 2015;352:291-304. Biochem Mol Biol. 2008;109: 350-3. 10. Jin J, Mao Y, Thomas D, Kim S, Daniel JL, Kunapuli SP. RhoA 24. Shu YJ, Weng H, Ye YY, Hu YP, Bao RF, Cao Y, et al. SPOCK1 downstream of G(q) and G(12/13) pathways regulates pro- as a potential cancer prognostic marker promotes the prolif- tease-activated receptor-mediated dense granule release in eration and metastasis of gallbladder cancer cells by activating platelets. Biochem Pharmacol. 2009;77:835-44. the PI3K/AKT pathway. Mol Cancer. 2015;14:12. 11. Taniguchi M, Nagai K, Arao N, Kawasaki T, Saito T, Moritani 25. Brazil DP, Yang ZZ, Hemmings BA. Advances in protein Y, et al. YM-254890, a novel platelet aggregation inhibitor pro- kinase B signalling: AKTion on multiple fronts. Trends duced by Chromobacterium sp. QS3666. J Antibiot (Tokyo). Biochem Sci. 2004;29:233-42. 2003;56:358-63. 26. Castoria G, Migliaccio A, D'Amato L, Di Stasio R, Ciociola A, 12. Wang C, Lv X, Jiang C, Davis JS. The putative G-protein cou- Lombardi M, et al. Integrating signals between cAMP and pled estrogen receptor agonist G-1 suppresses proliferation of MAPK pathways in breast cancer. Front Biosci. 2008;13:1318- ovarian and breast cancer cells in a GPER-independent man- 27. ner. Am J Transl Res. 2012;4:390-402. 27. Nieman KM, Kenny HA, Penicka CV, Ladanyi A, Buell-Gut- 13. Bai LY, Weng JR, Hu JL, Wang D, Sargeant AM, Chiu CF. G15, brod R, Zillhardt MR, et al. Adipocytes promote ovarian can- a GPR30 antagonist, induces apoptosis and autophagy in cer metastasis and provide energy for rapid tumor growth. human oral squamous carcinoma cells. Chem Biol Interact. Nat Med. 2011;17:1498-503. 2013;206:375-84. 28. Fukui M, Kang KS, Okada K, Zhu BT. EPA, an omega-3 fatty 14. Pal D, Dasgupta S, Kundu R, Maitra S, Das G, Mukhopadhyay acid, induces apoptosis in human pancreatic cancer cells: role S, et al. Fetuin-A acts as an endogenous ligand of TLR4 to pro- of ROS accumulation, caspase-8 activation, and autophagy mote lipid-induced resistance. Nat Med. 2012;18:1279- induction. J Cell Biochem. 2013;114:192-203. 85. 29. Sabharwal SS, Schumacker PT. Mitochondrial ROS in cancer: 15. Hu L, Lau SH, Tzang CH, Wen JM, Wang W, Xie D, et al. initiators, amplifiers or an Achilles' heel? Nat Rev Cancer. Association of Vimentin overexpression and hepatocellular 2014;14:709-21. carcinoma metastasis. Oncogene. 2004;23:298-302. 30. Gajate C, Gonzalez-Camacho F, Mollinedo F. Lipid raft con- 16. Sugiyama T, Kamura T, Kigawa J, Terakawa N, Kikuchi Y, nection between extrinsic and intrinsic apoptotic pathways. Kita T, et al. Clinical characteristics of clear cell carcinoma of Biochem Biophys Res Commun. 2009;380:780-4. the ovary: a distinct histologic type with poor prognosis and

VOLUME 52 NUMBER 3 JULY 2020 829