CBR-2020-3046-Jin,1P

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 1

　B

　CANCER BIOTHERAPY AND RADIOPHARMACEUTICALS Volume 00, Number 00, 2020 ª Mary Ann Liebert, Inc. DOI: 10.1089/cbr.2019.3046 CBR-2019-3046-ver9-Jin_1P Type: research-article Original Article

　AU1 c

　MiR-524-5p Suppresses Migration, Invasion, and EMT Progression in Breast Cancer Cells Through Targeting FSTL1

　 AU2 c Taobo Jin, * Yun Zhang, * and Tianya Zhang

　Abstract

　 Background: The effects of miR-524-5p on breast cancer (BC) have not been investigated, though study showed that miR-524-5p has an anticancer function. Thus, this study investigated the effects of miR-524-5p on BC cells and its potential molecular mechanism. AU3 c Materials and Methods: The expression of miR-524-5p from the collected BC samples was determined. Cell counting kit-8 (CCK-8) assay was performed to examine the effect of miR-524-5p on BC cells viability. The target for miR-524-5p was predicted by bioinformatics and further verified by luciferase assay. Wound healing assay and transwell assay were performed to determine cell migration and invasion of BC cells. The expressions of Follistatin-like 1 (FSTL1) and related proteins in epithelial–mesenchymal transition (EMT) were detected by Western blotting and quantitative real-time polymerase chain reaction. AU4 c Results: MiR-524-5p was low-expressed in BC samples, and upregulation of miR-524-5p suppressed BC cell viability, migration, and invasion. FSTL1 was predicted as a target for miR-524-5p. In addition, overexpressed FSTL1 effectively abolished the effect of miR-524-5p on inhibiting FSTL1 expression, and reversed the inhibitory effects of miR-524-5p on the migration, invasion of BC cells as well as the effect of miR-524-5p on regulating the expressions of matrix metalloproteinase 2 ( MMP2), matrix metalloproteinase 9 (MMP9), E-cadherin, and N-cadherin. Conclusions: Our findings suggest that miR-524-5p targeting FSTL1 adversely affects the progression of migration, invasion, and EMT of BC cells, thus, miR-524-5p is possibly a target for BC treatment.

　Keywords: breast cancer, EMT, FSTL1, invasion, migration, miR-524-5p

　 Introduction

　 reast cancer (BC) is one of the major causes of cancer-

　related death among women worldwide. 1

　Surgery and

　chemotherapy are therapeutic strategies, which are widely

　AU5 c used for treating BC, and are frequently associated with ad- verse side-effects. 2,3

　However, for patients with advanced BC, the efficacies of surgery and chemotherapy are low due to distant metastasis of cancer. 4

　Distant metastasis is an im- portant factor contributing to BC development, 5–7

　and many BC patients have metastatic BC at the time of diagnosis, and their

　5-year

　survival

　rates

　are

　only

　* 27%. 8

　Similar

　to

　other cancers, the occurrence, development, and metastasis of BC are closely related to the abnormal expressions of multiple

　genes such as oncogenes, tumor-suppressor and metastasis- related genes. 9

　Although early detection and treatment strategies of BC have been greatly improved, BC cells have strong invasion and metastasis abilities; as a result, the morbidity, mortality, and survival rates of BC patients are still not satisfactory. 10–12

　Therefore, a novel approach to the prevention and treatment of BC is urgently needed.

　MicroRNAs (miRNAs) are noncoding RNAs with vital biological function in cells. 10

　Evidence showed that abnor- mal expressions of miRNAs are highly related to the oc- currence and development of various cancer types, and that miRNAs could function as typical oncogenes and tumor- suppressor genes. 13

　In addition, recent studies have proved that miRNAs can be used as biomarkers for early screening,

　 Department of Thyroid and Breast Surgery, Zhuji People’s Hospital, Zhuji City, China.

　*These authors contributed equally to this work.

　 Address correspondence to: Tianya Zhang; Department of Thyroid and Breast Surgery, Zhuji People’s Hospital; No. 9, Jianmin Road, Taozhu Street, Zhuji City 311800, Zhejiang Province, China

　E-mail: tianya_zhangty@163.com

　 1

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 2

　2 JIN ET AL.

　diagnosis, and treatment of various cancers. 14–16

　MiR-524- 5p was reported to suppress melanoma cell growth through targeting BRAF and ERK2 17

　and inhibiting glioma pro- gression by directly regulating Hes1 and Jagged1. 18

　Studies also showed that miR-524-5p has the ability to regulate cancer cell proliferation and epithelial–mesenchymal tran- sition (EMT) progression. 19

　However, there is a lack of study on the functions of miR-524-5p in BC.

　In this study, the expression of miR-524-5p in BC tissue samples was determined, and the function of miR-524-5p in growth, migration, invasion, and EMT of BC cells was explored. Furthermore, we determined the possible target for miR-524-5p. The purpose of the research was to deter- mine whether miR-524-5p is a target for BC treatment.

　Materials and Methods

　Cell culture Human BC SK-BR-3 and MDA-MB-453 cell lines (Amer- ican Type Culture Collection, Rockville, MD) 10,20,21

　were cultured in Dulbecco’s modified Eagle’s medium (DMEM;

　 T ABLE

　1.

　 C LINICOPATHOLOGICAL

　V ARIABLES

　AND

　E XPRESSION

　of

　MI R- 524-5 P

　miR-524-5p expression

　 Gbico, MA) containing 10% fetal bovine serum (FBS; Gbico) at 37 ° C with 5% CO 2

　in a humid environment.

　 Tissue samples BC lesions

　and

　corresponding

　noncancerous

　samples (Table 1) were collected from 20 BC patients who under- b T1 went surgical excision at Zhuji People’s Hospital between January 2017 and January 2019. BC tissues were stored in a refrigerator at - 80 ° C for later use.

　All

　patients

　did

　not

　re- ceive chemotherapy or radiotherapy before the surgery. The current research was approved by the Ethics

　Committee

　of Zhuji People’s Hospital. Informed consent was signed by all participants.

　 Transfection MiR-524-5p mimics (mimic), Follistatin-like 1 (FSTL1) overexpression vector (FSTL1), mimic control and negative control (NC) were obtained from GenePharma (Shanghai, China). The mimic and FSTL1 were diluted to 20 lm by RNase-free H 2 O (Beyotime, Shanghai, China) and stored at

　- 20 ° C in a refrigerator for later use. 1.0 · 10 6

　BC cells were seeded into each well of six-well plates in 2 mL complete

　medium and grown until they were 60%–70% confluent. Two micrometers mimic and FSTL1, which were diluted by

　100 lL DMEM, were added to 3 lL Lipofectamine 2000 b AU6 (Invitrogen, MA). The above two mixtures were mixed and incubated for 15 min at room temperature. Finally, the mixed

　 Variables Patient number (%)

　Negative N (%)

　Positive

　N (%) p

　liquid was added to the cells of each well, and 1.8 mL DMEM was then added to the cells and held for an additional 48 h. Mimic control and NC were conducted in parallel.

　Total 20 10 (50.0) 10 (50.0)

　Age (years) 52.3 – 9.7

　 51.8 – 9.9 0.910

　TNM stage 0.039

　I 6 (30.0) 1 (5.0) 5 (25.0)

　II 9 (45.0) 5 (25.0) 4 (20.0)

　III 5 (25.0) 4 (20.0) 1 (5.0)

　Grade 0.057

　Tumor size 0.211

　£ 2 cm 9 (45.0) 3 (15.0) 6 (30.0)

　> 2, £ 5 cm 9 (45.0) 5 (25.0) 4 (20.0)

　> 5 cm 2 (10.0) 2 (10.0) 0 (0)

　 Quantitative real-time polymerase chain reaction miRNAs were extracted from clinical BC samples and BC cells (SK-BR-3 and MDA-MB-453 cells) using a miRcute miRNA Isolation Kit (TianGEN, Beijing, China). In brief, for BC samples, a grinding rod was used to grind the samples in liquid nitrogen in a 1.5 mL centrifugal tube containing lysis buffer, while for BC cells the cells were collected into 1.5 mL centrifugal tube and added to lysis buffer. Two hundred mi- croliters chloroform was then added to the cells and shaken for 1 min. After resting the cells for 5 min at room tempera-

　ture, the cells were centrifuged for 20 min (13,400 g), the b AU7

　Lymph node

　0.370 miRNA solution was transferred into a new 1.5 mL tube, and added to ethanol and centrifuged for another 15 min (13,400 g). The sediments were miRNAs, and RNase-free H 2 O was used to dilute the miRNA sediments. Total RNAs were ex-

　ER status 0.370

　 The median was the cutting point, higher than the median was the high expression, lower than the median was the low expression.

　tracted from BC cells using TRIzol reagent (Invitrogen). In brief, the cells were lysed by TRIzol and collected into a new

　1.5 mL centrifugal tube; chloroform was added to the tube and centrifuged for 20 min (1400 g). The supernatant was collected, and added to an equal volume of isopropanol and centrifuged for 5 min (1400 g), and the miRNA sediments were diluted using RNase-free H 2 O. Then, PrimeScript RT kit (Takara, Dalian, China) was used to reverse transcribe miRNAs and mRNAs into cDNAs according to the instruc- tions. Finally, gene expression was determined by quantita- tive real-time polymerase chain reaction (q-PCR) assays using Verso 1-step RT-qPCR Kit (Thermo Scientific, MA) in ABI 7500 Fast Real-Time PCR System (Applied Biosystems, CA), and the condition of q-PCR was set at 95 ° C for 30 s, at

　1 or 2

　14 (70.0)

　5 (25.0)

　9 (45.0)

　3

　6 (30.0)

　5 (25.0)

　1 (5.0)

　status

　 Negative

　11 (55.0)

　4 (20.0)

　7 (35.0)

　Positive

　9 (45.0)

　6 (30.0)

　3 (15.0)

　Negative

　11 (55.0)

　7 (35.0)

　4 (20.0)

　Positive

　9 (45.0)

　3 (15.0)

　6 (30.0)

　PR status

　0.350

　Negative

　13 (65.0)

　8 (40.0)

　5 (25.0)

　 Positive

　7 (35.0)

　2 (10.0)

　5 (25.0)

　 HER-2 status

　 0.170

　Negative

　12 (60.0)

　4 (20.0)

　8 (40.0)

　 Positive

　8 (40.0)

　6 (30.0)

　2 (10.0)

　 Triple negative

　0.350 Yes

　7 (35.0)

　5 (25.0)

　2 (10.0)

　No

　13 (75.0)

　5 (25.0)

　8 (40.0)

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 3

　THE ROLE OF MI R-524-5 P

　IN BREAST CANCER 3

　60 ° C for 30 s, 45 cycles at 60 ° C for 30 s. 17

　U6 and b-actin

　served as internal references. All primer sequences are shown

　T2

　c

　 in

　Table

　2.

　mRNA

　and

　miRNA

　were

　quantified

　by

　2 - 66 CT

　method. 18

　 room temperature. The cell numbers from three random areas on each membrane were counted under a phase-contrast mi- croscope ( · 200) (Axio Lab.A1 pol; Leica). ImageJ software (Version 1.8.0) was used to analyze the images.

　 Cell counting kit-8 assays Cell counting kit-8 (CCK-8; Bioengineering Institute,

　Luciferase reporter assays The fragments of FSTL1-3 ¢ -UTR containing

　 wild-type b AU8

　Nanjing,

　China)

　was

　used

　to

　determine

　cell

　viability. and mutant binding sites for miR-524-5p were inserted into

　CCK-8 contains WST-8 [2-(2-methoxy-4-nitrophenyl)-3- (4-nitrophenyl)-5-(2,4disulfophenyl)-2H-tetrazolium, mono- sodium salt]. In CCK-8 measurement, the dye of WST-8 was reduced by dehydrogenase in cells to form a water-soluble orange-colored product (formazan). The amount of the produced formazan dye by cellular dehydrogenases is cor- related with the number of living cells. Therefore, the cell viability can be simply estimated by recording the optical density of formazan at 450 nm using a microplate reader. 22

　In brief, after the transfection, the cells were seeded into 96-well plates at 1.0 · 10 4

　cells in 100 lL complete medium. After growing for 24, 48, and 72 h, the cells were incubated by CCK-8 reagent (0.5 mg/mL) for 15 min. Finally, the ab- sorbance of each well was detected at 450 nm using a mi- croplate reader (Infinite M200 PRO; Tecan Austria GmbH, Austria).

　 Wound healing assay After the transfection was completed, the cells were placed into six-well plates at 3.5 · 10 5

　cells in complete medium and continued to culture until the cell confluence reached 95%. Then, a vertical wound in each well was created by a 20 lL pipette tip, and DMEM without FBS was added into each well. Images ( · 100) in each well were collected at 0 and 24 h under a phase-contrast microscope (Axio Lab.A1 pol; Leica, Solms, Germany). ImageJ software (Version 1.8.0) was used to analyze the images in this assay.

　 Transwell assays Transwell cell culture chambers were precoated by Ma- trigel (Corning Life Sciences, NY), and the transwell inserts were placed into a 24-well plate. The transfected cells were diluted into 2 · 10 5

　and pipetted into the upper chambers containing suspension solution with 0.2 mL EMEM without FBS, and the corresponding complete medium was added into the lower chamber. The cells were incubated for 24 h, the upper-side of the polycarbonate membrane was wiped, leav- ing the underside of the membrane containing invaded cells. Finally, the cells were stained by crystal violet for 15 min at

　pmirGLO luciferase vectors (GenePharama). After the b AU9

　FSTL1-3 ¢ -UTR WT (or MUT) were cotransfected with mimic by Lipofectamine 2000 for 48 h. The luciferase ac- tivity of the cells was determined by dual-luciferase reporter assay (Promega, CA) by GloMax fluorescence reader (Pro- mega). PsiCHECK-2 vector transfected into cells served as internal control.

　 Western blot assays Total proteins from the cells were isolated by RIPA lysis buffer (Beyotime), and the total protein concentrations were determined by a BCA assay kit (Pierce, MA). Finally, total protein (30 lg) was separated on 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels, elec- troblotted and transferred to NC membranes (Millipore, MA). Then, the membranes were blocked by 5% skimmed milk for 1 h at room temperature and incubated with antibodies [FSTL1 (1:1000, ab11805, 50 kD; Abcam, CA), matrix me- talloproteinase 9 (MMP9) (1:1000, ab73734, 78 kD; Abcam), matrix metalloproteinase 2 (MMP2) (1:1000, ab37150, 72 kD; Abcam), E-cadherin (1:1000, ab40772, 97 kD; Ab- cam), N-cadherin (1:1000, ab18203, 130 kD; Abcam), and b-actin (1:1000, 42 kD, ab8226; Abcam)] at 4 ° C overnight. Next day, HRP-conjugated secondary antibodies [goat anti- rabbit IgG secondary antibody (1:5000, L3012, Signalway Antibody, PA), goat antimouse IgG secondary antibody (1:5000, ab205719; Abcam), and donkey antigoat IgG sec- ondary antibody (1:5000, ab6885; Abcam)] were incubated

　with the membranes for 1 h at room temperature. Finally, b AU10 SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific) was used to incubate the membranes signal de-

　tection.

　Image

　Lab ™

　Software

　(version 3.0) was used for

　densitometric analysis and quantification of the Western blotting data (Bio-Rad Laboratories, Inc., Hercules, CA).

　 Statistical analysis Student’s t test and one-way analysis of variance (ANOVA) were used to analyze the data by SPSS software (version 18.0).

　Post hoc tests were performed by LSD and Dunnet’s. The data b AU11

　 T ABLE

　2.

　 Q UANTITATIVE

　R EAL -T IME

　P OLYMERASE

　C HAIN

　R EACTION

　P RIMERS

　Target gene Forward primers, 5¢-3¢ Reverse primers, 5¢-3¢

　FSTL1

　CCCAGTTGTTTGCTATCAGTCC

　TGTAGTTGCTGCCTTTAGAGAAC miR-524-5p

　AGTTGCTGCAGTGATTC

　GTGCATCGTCCGAGCT

　MMP2

　TACAGGATCATTGGCTACACACC

　GGTCACATCGCTCCAGACT

　MMP9

　TGTACCGCTATGGTTACACTCG

　GGCAGGGACAGTTGCTTCT

　E-cadherin

　CGAGAGCTACACGTTCACGG

　GGGTGTCGAGGGAAAAATAGG

　N-cadherin

　TGCGGTACAGTGTAACTGGG

　GAAACCGGGCTATCTGCTCG

　U6

　CTCGCTTCGGCAGCACA

　AACGCTTCACGAATTTGCGT

　b-actin

　CATGTACGTTGCTATCCAGGC

　CTCCTTAATGTCACGCACGAT

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 4

　4 JIN ET AL.

　were shown as means – standard deviation. All experiments were conducted in triplicate. Statistically significant result was shown as p < 0.05.

　Results

　MiR-524-5p was low-expressed in BC tissues A comparative analysis of miR-524-5p expression level in clinic primary BC and adjacent samples was conducted,

　and the result revealed that miR-524-5p expression was obviously lower in BC tissue

　compared

　with

　that

　in

　ad- jacent tissues ( p < 0.01) (Fig. 1A). Moreover, for clinical b F1 samples, we took the median as the cutoff point, higher

　than the median was regarded as high expression

　and lower than the median was regarded as low expression. As shown in Table 1, there was a significant negative corre- lation between TNM stage and the miR-524-5p expression in these patients.

　 FIG. 1. MiR-524-5p was low-expressed in BC tissues and inhibited BC cell viabilities. (A) miR-524-5p expression in primary BC lesions and adjacent tissues collected from 40 pairs was analyzed by q-PCR, U6 was used as housekeeping gene of miR-524-5p (**p < 0.01 vs. adjacent tissues). (B, C) The transfection efficiency of miR-524-5p mimic in BC cells was determined by q-PCR, U6 served as housekeeping gene of miR-524-5p. (D, E) Viability of BC cells was measured using CCK-8 assays after transfection. All experiments were conducted in triplicate (*p < 0.05, **p < 0.01, vs. control; # p < 0.05, ## p < 0.01, vs. mimic control). BC, breast cancer; CCK-8, cell counting kit-8; q-PCR, quantitative real-time polymerase

　chain reaction.

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 5

　THE ROLE OF MI R-524-5 P

　IN BREAST CANCER 5

　Overexpressed miR-524-5p inhibited viabilities of SK-BR-3 and MDA-MB-453 cells To determine the functions of miR-524-5p in BC cells, miR-524-5p mimic was transfected into BC cells, and sub- sequently, the changes in viabilities of SK-BR-3 and MDA- MB-453 cells were detected after 48 and 72 h. As shown in Figure 1B and C, the expression of miR-524-5p had no difference in control and mimic control groups, while it was upregulated in the two cells in mimic group ( p < 0.01), in- dicating that miR-524-5p mimic was successfully trans- fected into the two cells. We also found that cell viability in mimic control and control groups had no statistical differ- ence, while in mimic group the viability of the two cells was significantly reduced, compared with control and mimic

　control groups (48 h: p < 0.05; 72 h: SK-BR-3, p < 0.05; MDA-MB-453, p < 0.01, respectively) (Fig. 1D, E).

　Overexpressed miR-524-5p inhibited migration and invasion of SK-BR-3 and MDA-MB-453 cells Whether miR-524-5p regulated SK-BR-3 and MDA-MB- 453 cell migration and invasion was determined, as shown in

　F2 c Figure 2A and B, 24 h after scratches, no significant differ- ence was observed in the cell migration into the wounded (clear) area of the cell monolayers in control and mimic control groups. However, in the mimic group, relative mi- gration rate was significantly reduced compared with control and mimic control groups (SK-BR-3, p < 0.05;

　MDA-MB-

　453, p < 0.01, respectively). As for invasion ability, as shown

　in Figure 2C and D, similar to wound healing assays, in the mimic group, the relative invasion rate was significantly re- duced compared with control and mimic control groups (SK-BR-3, p < 0.01; MDA-MB-453, p < 0.01, respectively).

　Thus, the results showed that miR-524-5p could inhibit the

　migration and invasion of SK-BR-3 and MDA-MB-453 cells.

　 MiR-524-5p specifically targets FSTL1 TargetsScan predicted that the possible target for miR- 524-5p was FSTL1, as the target sequence at 1945–1951 base

　 was

　 in

　 pairs

　 with

　 FSTL1-3 ¢ -UTR

　 in

　 the study

　F3 c (Fig. 3A). To further confirm the prediction, luciferase re-

　porter assays were conducted. As shown in Figure 3B and C, luciferase activities decreased in the two cells cotransfected with wild-type FSTL1 and miR-524-5p mimic compared with NC group (SK-BR-3, p < 0.01; MDA-MB-453, p < 0.01,

　respectively).

　However,

　after

　cotransfection

　with FSTL1

　mutation (mut) and miR-524-5p mimic, no differences in the luciferase activities in the Control + FSTL1 and miR-524- 5p + FSTL1-mut groups were identified. Therefore, FSTL1 is targeted by miR-524-5p.

　 Overexpressed FSTL1 partly reversed the inhibitory effects of miR-524-5p on viabilities of SK-BR-3 and MDA-MB-453 cells We determined whether the inhibitory effects of miR- 524-5p on SK-BR-3 and MDA-MB-453 cell viabilities were

　F4 c mediated by downregulating FSTL1, as shown in Figure 4, overexpression of FSTL1 significantly increased the vi- abilities of the two cells, compared with control and NC groups ( p < 0.05), while miR-524-5p mimic greatly reduced the cell viability as compared with control and NC groups

　 (48 h, p < 0.05; 72 h, p < 0.01, respectively). However, in FSTL1 + mimic group, the inhibitory effect of miR-524-5p mimic was partly reversed by overexpressed FSTL1 com- pared with mimic group (48 h, p < 0.05; 72 h, p < 0.05 and p < 0.01, respectively).

　 Overexpressed FSTL1 partly reversed the inhibitory effect of miR-524-5p on migration and invasion of SK-BR-3 and MDA-MB-453 cells Whether the inhibitory effect of miR-524-5p on migration and invasion of SK-BR-3 and MDA-MB-453 cells was me- diated through downregulating FSTL1 was determined by conducting wound healing and transwell assays. As shown in

　Figure 5A–D, in FSTL1 group, 24 h after the wounding ex- b F5 periment, the relative migration rates were increased both b AU12 in the two cells compared with control and NC groups

　( p < 0.01). In mimic group, the relative migration rates were reduced in the two cells as compared with control and NC groups ( p < 0.01), while in FSTL1 + mimic group the inhibi- tory effect of miR-524-5p mimic on cell migration in the two cells was partly reversed by overexpression of FSTL1 com- pared with FSTL1 and mimic groups (SK-BR-3, p < 0.01; MDA-MB-453, p < 0.01 and p < 0.05, respectively). As shown in Figure 5E–H, the changes in cell invasion were similar to the results in wound healing assays. In the FSTL1

　group, the relative invasion rates in the two cells were sig- nificantly increased as compared with control and NC groups ( p < 0.01), while in mimic group the relative invasion rates of the two cells were reduced compared with control and NC groups ( p < 0.01). However, in FSTL1 + mimic group, the inhibitory effect of miR-524-5p mimic on cell invasion was partly reversed by overexpressed FSTL1 in the two cells as compared with FSTL1 and mimic groups ( p < 0.01). Thus, the results showed that overexpression of FSTL1 partly re- versed the inhibitory effect of miR-524-5p on migration and invasion of SK-BR-3 and MDA-MB-453.

　 Overexpressed FSTL1 partly reversed the inhibitory effect of miR-524-5p on the expression of FSTL1 The relation between miR-524-5p and FSTL1 was de- termined by Western blot and q-PCR assays, as

　shown

　in Figure 6A, B, and D, E, FSTL1 protein expression in SK- b F6 BR-3 and MDA-MB-453 cells was upregulated

　in

　FSTL1 group compared with control and NC groups ( p < 0.01); however, FSTL1 protein was downregulated in mimic group compared with control and NC groups ( p < 0.05 and

　p < 0.01,

　respectively).

　In

　FSTL1 + mimic

　group,

　the

　ex-

　pression of FSTL1 protein was lower than that in the FSTL1 group ( p < 0.05) but higher than that in the mimic group

　 ( p < 0.01). The mRNA expression level of FSTL1 in the two cells after transfection with FSTL1 and/or mimic was con- sistent with its protein expression (Fig. 6C, F). Thus, the results showed that miR-524-5p specifically targets FSTL1.

　 Overexpressed FSTL1 partly reversed the inhibitory effect of miR-524-5p on EMT progression We further detected the level of transcription

　and

　trans- lation of some key factors involved in EMT. As

　shown

　in Figure 7A–C and E–G, in FSTL1 group, the protein ex- b F7 pressions of MMP2, MMP9, and N-cadherin in SK-BR-3

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 6

　 FIG. 2. Overexpressed miR-524-5p inhibited migration and invasion of SK-BR-3 and MDA-MB-453 cells. (A, B) Cell migration was analyzed by wound healing assay in SK-BR-3 and MDA-MB-453 cells after transfection (magnifica- tion · 100). (C, D) Cell invasion was analyzed by transwell assay in SK-BR-3 and MDA-MB-453 cells after transfection (magnification · 200). All experiments were conducted in triplicate (*p < 0.05, **p < 0.01, vs. control; # p < 0.05, ## p < 0.001, vs. mimic control).

　 6

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 7

　THE ROLE OF MI R-524-5 P

　IN BREAST CANCER 7

　 FIG. 3. MR-524-5p specifically targets FSTL1. (A) FSTL1 mRNA 3 ¢ -UTR contains a binding site of miR-524-5p. (B, C) Luciferase assay validated that miR-524-5p targets FSTL1 in SK-BR-3 and MDA-MB-453 cells, luciferase from firefly was used as reporter gene, and luciferase from the kidney served as internal reference gene. All experiments were con- ducted in triplicate (**p < 0.01,

　vs. Control + FSTL1; ## p < 0.01, vs. miR-524-3p + FSTL1-mut). FSTL1,

　Follistatin-like 1; mut, mutation.

　 and MDA-MB-453 cells were upregulated ( p < 0.05), while the protein expression of E-cadherin was downregulated compared with control and NC groups ( p < 0.01). In mimic group, the protein expressions of MMP2, MMP9, and N-cadherin in the two cells were significantly down- regulated ( p < 0.01), whereas the protein expression of E-cadherin was upregulated compared with control and NC groups ( p < 0.01). In FSTL1 + mimic group, the effects of miR-524-5p on these proteins were reversed by FSTL1 in the two

　cells

　compared

　with

　FSTL1

　and

　mimic

　groups ( p < 0.05), and the mRNA expression levels of these factors in the two cells were consistent with the protein expressions (Fig. 7D, E and I, J). Thus, the results proved that FSTL1 partly reversed the inhibitory effect of miR-524-5p on EMT progression, further confirming that the regulatory effects of miR-524-5p on the cell viability, migration, and invasion of SK-BR-3 and MDA-MB-453 cells were associated with the inhibition of FSTL1.

　Discussion

　 Cancer metastasis is a major cause of death for cancer patients. 23

　Abnormal changes in cancer-related genes con- tribute greatly to the occurrence and development of cancer and cancer metastasis. 9,24

　Researchers proved the potential values of miRNAs in cancer screening, diagnosis, and treatment. 14–16

　miRNAs are a type of small noncoding RNAs with 18–25 nucleotides in length, and they are vital regulators in regulating 30%–60% human gene expressions in the biological processes such as cell proliferation, an- giogenesis, EMT, and apoptosis. 25

　MiR-524-5p was recently reported to have the ability of inhibiting cancer cell growth such as melanoma and glioma. 17–19

　However, there is a lack of report on the role of miR-524-5p in BC. Thus, this study determined the role of miR-524-5p in BC, and we found that miR-524-5p was low-expressed in BC tissues compared with adjacent tissues. Such a phenomenon in this study is

　FIG. 4. Overexpressed FSTL1 partly reversed the inhibitory effects of miR-524-5p on viabilities of SK-BR-3 and MDA- MB-453 cells. (A) SK-BR-3 cell viability was determined by CCK-8 assay after the transfection. (B) MDA-MB-453 cell viability was determined by CCK-8 assays after transfection. All experiments were conducted in triplicate (*p < 0.05,

　**p

　<

　0.01,

　vs.

　control;

　# p

　<

　0.05,

　## p

　<

　0.01,

　vs.

　NC;

　^p

　<

　0.05,

　vs.

　FSTL1;

　6 p

　<

　0.05,

　66 p

　<

　0.01,

　vs.

　mimic).

　NC,

　negative

　control.

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 8

　 FIG. 5. Overexpressed FSTL1 partly reversed the inhibitory effect of miR-524-5p on migration and invasion of SK-BR-3 and MDA-MB-453 cells. (A, D) Cell migration was analyzed by wound healing assay in SK-BR-3 and MDA-MB-453 cells after the transfection (magnification · 100). (B, C) Quantification of wound healing assay. (E, H) Cell invasion was analyzed by transwell assay in SK-BR-3 and MDA-MB-453 cells after the transfection (magnification · 200). (F, G) Quantification of transwell assay. All experiments were conducted in triplicate (**p < 0.01, vs. control; ## p < 0.01, vs. NC;

　^^p

　<

　0.01,

　vs.

　FSTL1;

　6 p

　<

　0.05,

　66 p

　<

　0.01,

　vs.

　mimic).

　8

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 9

　THE ROLE OF MI R-524-5 P

　IN BREAST CANCER 9

　FIG. 5.

　(Continued).

　 consistent with previous studies, 17–19

　indicating that miR- 524-5p plays an important role in the progression of BC; however, the underlying mechanisms still require further investigation. Subsequently, the effects of overexpressed miR-524-5p on biological abilities of BC progression were

　detected, and we found that miR-524-5p remarkably in- hibited the viability, migration, and invasion of BC cells, suggesting that miR-524-5p has a functional role in BC.

　miRNAs can affect the pathogenesis of various diseases through specifically targeting certain mRNA. 14

　In this study,

　 FIG. 6. Overexpressed FSTL1 partly reversed the inhibitory effect of miR-524-5p on the expression of FSTL1. (A, D) The FSTL1 protein expression was determined by Western blot assay in SK-BR-3 and MDA-MB-453 cells after the transfection. (B, E) Quantification of Western blot assay. b-actin was used as an internal control of proteins. (C, F) The mRNA expression of FSTL1 was detected by q-PCR assay in SK-BR-3 and MDA-MB-453 cells after the transfection. b-Actin was used as housekeeping gene of mRNA. All experiments were conducted in triplicate (*p < 0.05, **p < 0.01, vs. control;

　# p

　<

　0.05,

　## p

　<

　0.01,

　vs.

　NC;

　^p

　<

　0.05,

　vs.

　FSTL1;

　66 p

　<

　0.01,

　vs.

　mimic).

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 10

　10 JIN

　ET AL.

　FIG. 7. Overexpressed FSTL1 partly reversed the inhibitory effect of miR-524-5p on EMT progression. (A, F) The

　protein expressions of MMP2, MMP9, E-cadherin, and N-cadherin were detected by Western blot assay in SK-BR-3 and MDA-MB-453 cells after the transfection. b-Actin was used as an internal control of proteins. (B, C, G, H) Quantification of Western blot assay. (D, E, I, J) The mRNA expressions of MMP2, MMP9, E-cadherin, and N-cadherin were determined by q-PCR assay in SK-BR-3 and MDA-MB-453 cells after the transfection. b-Actin was used as housekeeping gene of mRNA. All experiments were conducted in triplicate (*p < 0.05, **p < 0.01, vs. control; # p < 0.05, ## p < 0.01, vs. NC;

　^p

　<

　0.05,

　 ^^p

　<

　0.01,

　 vs.

　 FSTL1;

　 6 p

　<

　0.05,

　 66 p

　<

　0.01,

　 vs.

　 mimic).

　 EMT,

　 epithelial–mesenchymal

　 transition;

　 MMP2,

　matrix metalloproteinase 2; MMP9, matrix metalloproteinase 9.

　to further detect the antitumor effect mechanism of miR- 524-5p, the result from bioinformatics prediction showed that FSTL1 was a potential effector of miR-524-5p, which was validated by luciferase assay results. FSTL1 is involved in a series of physiological and pathological processes such as osteoarthritis, reactive arthritis, and ulcerative, 26

　and it regulates biological processes, including cell proliferation, immune response, cell apoptosis, and metabolism. 26–29

　This study showed that miR-524-5p targeted FSTL1 and had an inhibitory effect on BC cells. In addition, it is reported that

　FSTL1 is directly regulated by miR-198 to inhibit cell mi- gration. 30

　MiR-32-5p also regulates FSTL1 to inhibit cell survival in Mycobacterium. 31

　To further confirm the specific mechanism of the effect of miR-524-5p/FSTL1 signal on BC, CCK-8, wound healing, transwell, Western blot, and q-PCR assays were performed. The results showed that FSTL1 partially reversed the inhibitory effect of miR-524- 5p on the viability, migration, and invasion ability of BC cells. Moreover, miR-524-5p could downregulate the ex- pression of FSTL1 at protein and mRNA levels, suggesting

　CBR-2019-3046-ver9-Jin_1P.3d

　03/19/20 9:07pm Page 11

　THE ROLE OF MI R-524-5 P

　IN BREAST CANCER 11

　FIG. 7.

　(Continued).

　that miR-524-5p could reduce the transcription and trans- lation of FSTL1 and exert the inhibitory effect on BC cell viability, migration, and invasion.

　EMT plays a vital role in the metastasis of cancer cells. 32–36

　During EMT, epithelial cells gradually transform into interstitial cells, causing cell migration and inva- sion. 37,38

　EMT process can be characterized by decreased expressions of adhesion molecules (e.g., E-cadherin) and increased expression of N-cadherin. 39,40

　MMP2 and MMP9 are EMT markers that could be increased significantly during EMT process. 41,42

　Therefore, changes in EMT-related gene expressions serve as important markers in indicating the degree of metastasis in tumor cells. In the present research, our results showed that miR-524-5p/FSTL1 could signifi- cantly regulate migration and invasion of BC cells. Fur- thermore, we detected expressions of relative key factors at protein and mRNA levels, and found that miR-524-5p downregulated MMP2, MMP9, and N-cadherin expressions,

　 upregulated the expression of E-cadherin at protein and mRNA levels. However, FSTL1 produced the opposite effect to miR-524-5p. The data also revealed that FSTL1 could partly reverse the function of miR-524-5p in the expressions of above factors, which further proved that FSTL1 was tar- geted and inhibited by miR-524-5p.The findings proved that miR-524-5p/FSTL1 has the ability of regulating the pro- gression of EMT in BC cells.

　This study found that miR-524-5p attenuated the pro- gression of migration, invasion, and EMT in BC ce...