Formononetin relieves the facilitating effect of lncRNA AFAP1-AS1-miR-195/miR-545 axis on progression and chemo-resistance of triple-negative breast cancer

This investigation attempted to discern whether formononetin restrained progression of triple-negative breast cancer (TNBC) by blocking lncRNA AFAP1-AS1-miR-195/miR-545 axis. We prepared TNBC cell lines (i.e. MDA-MB-231 and BT-549) and normal human mammary epithelial cell line (i.e. MCF-10A) in advance, and the TNBC cell lines were, respectively, transfected by pcDNA3.1-lncRNA AFAP1-AS1, si-lncRNA AFAP1-AS1, pcDNA6.2/GW/EmGFP-miR-545 or pcDNA6.2/GW/EmGFP-miR-195. Resistance of TNBC cells in response to 5-Fu, adriamycin, paclitaxel and cisplatin was evaluated through MTT assay, while potentials of TNBC cells in proliferation, migration and invasion were assessed via CCK8 assay and Transwell assay. Consequently, silencing of lncRNA AFAP1-AS1 impaired chemo-resistance, proliferation, migration and invasion of TNBC cells (P<0.05), and over-expression of miR-195 and miR-545, which were sponged and down-regulated by lncRNA AFAP1-AS1 (P<0.05), significantly reversed the promoting effect of pcDNA3.1-lncRNA AFAP1-AS1 on proliferation, migration, invasion and chemo-resistance of TNBC cells (P<0.05). Furthermore, CDK4 and Raf-1, essential biomarkers of TNBC progression, were, respectively, subjected to target and down-regulation of miR-545 and miR-195 (P<0.05), and they were promoted by pcDNA3.1-lncRNA AFAP1-AS1 at protein and mRNA levels (P<0.05). Additionally, formononetin significantly decreased expressions of lncRNA AFAP1-AS1, CDK4 and Raf-1, while raised miR-195 and miR-545 expressions in TNBC cells (P<0.05), and exposure to it dramatically contained malignant behaviors of TNBC cells (P<0.05). In conclusion, formononetin alleviated TNBC malignancy by suppressing lncRNA AFAP1-AS1-miR-195/miR-545 axis, suggesting that molecular targets combined with traditional Chinese medicine could yield significant clinical benefits in TNBC.


INTRODUCTION
Triple-negative breast cancer (TNBC), responsible for around 15% of global breast cancer (BC) cases, is histopathologically featured by shortages of estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor (HER)-2 [1]. Given its insensitivity responding to HER2-targeted therapy and endocrine therapy, TNBC was principally tackled by chemotherapies founded upon anthracycline and taxane [2], whose clinical efficacy, nonetheless, turned less encouraging than desired owing to development of drug resistance [3,4]. As a consequence, profound comprehension of drug-resistance is indispensable to perfect strategies for TNBC treatment.
In addition, the prominent role of traditional Chinese medicines (TCMs) in suppressing tumorigenesis has also been increasingly recognized at home and abroad. For instance, formononetin, an isoflavonoid isolated from astragalus membranaceus and spatholobus suberectus, was found to impair capabilities of BC cells in proliferating, migrating and invading via blockade of PI3K/Akt signaling [18,19]. Synergy of formononetin with metformin or everolimus also pronouncedly antagonized growth of BC cells by depressing ERK1/2 signaling [20] and mTOR signaling [21]. Notably, exposure to formononetin could significantly alter miRNA profiling in human umbilical vein endothelial cells (HUVECs), such as elevating expressions of miR-375 and miR-200b [22], both of which were crucial protectors against BC progression [23,24]. Despite these discoveries, lncRNAs, which were likely to act upon miRNAs through classical competing endogenous (ce) RNA manner [25], were barely explored regarding their implication in formononetin-involved BC inhibition, let alone lncRNA/miRNA axes.
To bridge this gap, this investigation was designed to unveil lncRNAs (e.g. lncRNA AFAP1-AS1) and associated miRNA networks that were involved in the protective impact of formononetin against TNBC development, which might be conducive to clinical treatment of TNBC.

DISCUSSION
Early recurrence, swift progression and poor prognosis constitute major obstacles to successful treatment of TNBC [38,39], so in-depth understanding of TNBC etiology is required, and formulating treatment strategies that work for TNBC has become a necessity.

CONCLUSIONS
Collectively, formononetin exerted anti-TNBC function by reducing the influence of lncRNA AFAP1-AS1 on miR-545-3p/CDK4 axis and miR-195/Raf-1 axis, which were associated with TNBC exacerbation and chemoresistance ( Figure 8). There were, however, a couple of deficiencies in the experimental design. For AGING one thing, although tumor growth in TNBC-bearing mice models, was inhibited by formononetin at the concentration of 80 mg/kg (Supplementary Table 4), along with decreased lncRNA AFAP1-AS1 level and increased miR-545-3p/miR-195 level in the tumor tissues ( Supplementary Figure 8), joint effects of formononetin, si-lncRNA AFAP1-AS1 and miR-545-3p/miR-195 mimic on tumor growth in the mice models were not studied. For another, considering that singletarget therapy led to smaller objective response rates than multiple-target therapy in terms of treating solid tumors [84,85], combined application of molecular targets and formononetin might be viable for TNBC treatment, but this point was not clinically supported. . Inhibitory rate (%) of chemo-drugs on growth of TNBC cells was assessed based on the formula of (1-A490 drug group /A490 control group ) × 100%, and half maximal inhibitory concentration (IC50) values were calculated. The experiments were conducted with ≥ 3 replicates.

Cell treatment by formononetin
MDA-MB-231 and BT-549 cells adjusted to the density of 5×10 4 /ml were seeded into 96-well culture plates, and they were starved in serum-free medium for 24 h. Afterwards, the TNBC cells were exposed to 10 μmol/L, 20 μmol/L, 40 μmol/L, 80 μmol/L and 160 μmol/L formononetin (batch number: 111703-200603, China National Institute for Food and Drug Control), respectively, for 24 h. The experiments were repeated for ≥ 3 times.

Real-time quantitative PCR (RT-PCR)
BC tissues frozen within liquid nitrogen, as well as BC cell lines, were lysed after addition of 1ml TRIzol reagent (Invitrogen, USA), through which total RNAs were isolated. Concentration and purity of the RNAs were assessed using an ultraviolet (UV) spectrophotometer (model: NanoDropND-1000, NanoDrop Technologies, USA), and RNA samples whose A260/A280 ratio lied between 1.8 and 2.1 were reserved. Reverse transcription of the RNAs was implemented following procedures described in PrimeScript TM RT Master Mix kit (Takara, Japan) or miScript II RT kit (Qiagen, Germany), and the obtained cDNAs were amplified by employing real-time PCR kit (Takara, Japan) or miScript SYBR® Green PCR kit (Qiagen, Germany). Primers for genes were ordered in Supplementary Table 5, and their relative expression was normalized by means of 2 -ΔΔCt method [86]. These experiments were repeated for at least 3 times.

Western blotting
After denaturation at 105° C for 5 min, total protein extracted from BC tissues and cell lines was separated by electrophoresis, successively experiencing 1) 80 V for 2~3 h and 2) 100 V for 90 min. With usage of electrophoretic transfer apparatus (model: Mini Trans-Blot, Bio-Rad, USA), proteins on the gel were transferred onto polyvinylidene fluoride (PVDF) membrane through AGING wet method. Afterwards, the membrane was placed within 10 ml blocking buffer (i.e. 2% skim milk) for 1 h, and protein samples were incubated by primary antibodies (rabbit-anti-human, Abcam, USA) against CDK4 (1: 2000, Catalog No: ab108357), Raf-1 (1: 2000, Catalog No: ab137435) and GAPDH (1: 10000, Catalog No: ab181602) at 4° C for overnight. Then the products were incubated by goat anti-rabbit IgG H&L labelled by horseradish peroxidase (HRP) (1:5000, Catalog No: ab205718, Abcam, USA) at room temperature for 2 h. Development of protein samples was carried out by adopting chemiluminescence (ECL) (Pierce, USA), and gray values of protein bands were determined through utilization of Image-Pro Plus software (Media Cybernetics, USA). The experiments were carried out for at least 3 times.

CCK-8 assay
MDA-MB-231 and BT-549 cells were seeded into 96well plates at the density of 3000 cells per well. After overnight culture, 10 μl CCK-8 reagent (Dojindo, Japan) was supplemented gently into each well at the time point of 0 h. After cultivation at 37° C for 24 h, 48 h, 72 h and 96 h, absorbance (A) of TNBC cells in each well was monitored at 450 nm on the microplate reader (Bio-Rad, USA). These experiments were performed for at least 3 times.

Transwell assay
Cell migration MDA-MB-231 and BT-549 cells at the concentration of 1×10 5 /ml were paved onto the upper Transwell chamber (Corning Costar, USA), and 600 μl DMEM medium that contained 10% FBS was poured into the lower transwell chamber (Corning Costar, USA). After routine culture for 24 h, the TNBC cells were stained by 0.1% crystal violet (Solarbio Life Sciences, China), thereafter photographs were taken under optical microscope (Olympus, USA). The experiments were undertaken with ≥ 3 replicates.

Cell invasion
Procedures of cell invasion assay were mostly consistent with those of cell migration assay, except that Matrigel diluted by DMEM (ratio: 1/6) was added into the upper Transwell chamber (Corning Costar, USA), after which suspension of MDA-MB-231/BT-549 cells and DMEM medium were supplemented.

Statistical analyses
Data analyses in this investigation were fulfilled using SPSS ver.20 software (SPSS Inc. Chicago, IL, USA). Among them, quantitative data [mean ± standard deviation (SD)] were processed by student's t-test or analysis of variance (ANOVA), and categorical data (n) were analyzed via chi-square test. Differences were statistically significant when two-sided P value was less than 0.05.

Ethics approval and consent to participate
This investigation has obtained approvals from Longhua Hospital Affiliated to Shanghai University of TCM and AGING the ethics committee of Longhua Hospital Affiliated to Shanghai University of TCM in advance.

Availability of data and materials
The data used to support findings of this study are available from the corresponding author upon reasonable requests.

CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.

FUNDING
This study is supported by the National Natural Science Fund for youth (Grant No. 81704074).

Editorial note
& This corresponding author has a verified history of publications using a personal email address for correspondence.  Please browse Full Text version to see the data of Supplementary Table 1 Collection of triple negative breast cancer (TNBC) specimens: Tumor tissues and adjacent normal tissues were collected from a total of 249 BC patients, who has been histopathologically confirmed as TNBC (n=94) and non-TNBC (n=155), recruited from Breast Department of Longhua Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, from December of 2011 to November of 2016Participants in this program all met following criteria: 1) they were females with primary BC; 2) they underwent surgeries; 3) they have signed informed consents; and 4) their tissues were frozen within < 30 min since being excised. Moreover, the applicants were excluded if: 1) their TNBC subtype could not be verified; 2) their tissues were unavailable from surgery; 3) they participated in this program without consents from their family; and 4) their cancer tissues were not stored timely. This program was approved by Longhua Hospital Affiliated to Shanghai University of Traditional Chinese Medicine and the ethics committee of Longhua Hospital Affiliated to Shanghai University of Traditional Chinese Medicine. High: highly-expressed lncRNA AFAP1-AS1; Low: lowly-expressed lncRNA AFAP1-AS1; OR: odds ratio; CI: confidence interval; *: statistical significance.

Supplementary Tables
AGING Supplementary : Establishment of triple negative breast cancer (TNBC)-bearing nude mice models: A total of 40 SPF-grade BALB/c female mice, aged 6.5 weeks, were provided by experimental animal center of Longhua Hospital Affiliated to Shanghai University of Traditional Chinese Medicine. After the mice were acclimate to surroundings at 25° C for 1week, they were injected by 0.1 ml MDA-MB-231 cell suspension (around 1 × 10 6 cells) subcutaneously in the chest. On the 2 nd day, there existed a node in the injection site of each mouse, and TNBC-bearing mice models were established successfully. Then the mice models were divided into control group (n=10), formononetin (20 mg/kg) group (n=10), formononetin (40 mg/kg) group (n=10) and formononetin (80 mg/kg) group (n=10). Formononetin was intra-peritoneally injected into each mouse model once per day, for continuously 21 days. Body weight and tumor weight (W) of the mice models were recorded, and tumor growth inhibition rate (%), which was equivalent to (1-W formononetin /W control ) × 100%, was calculated. What's more, tumors of the mice models were excised to determine levels of lncRNA AFAP1-AS1-1, miR-545 and miR-195 with PCR. *: P<0.05 when compared with control group.