Exosomal lncRNA OIP5-AS1 confers cells trastuzumab resistance in breast cancer through miR-381-3p/HMGB3 axis CURRENT STATUS: UNDER REVIEW

Background Long noncoding RNA OPA-interacting protein 5 antisense transcript 1 (OIP5-AS1) was confirmed to involve in the malignancy of breast cancer. However, whether exosomal OIP5-AS1 implicated in trastuzumab resistance remains unclear. Methods Cell viability, migration and apoptosis were analyzed by cell counting kit-8, colony formation, transwell assay or flow cytometry. The expression of OIP5-AS1 and microRNA (miR)-381-3p was detected using quantitative real-time polymerase chain reaction. Exosomes were isolated by ultracentrifugation and qualified by nanoparticle tracking analysis (NTA) software. Western blot was used to detect the levels of tumor susceptibility gene 101 (TSG101), CD81, CD63 or high-mobility group protein B3 (HMGB3). The interaction between miR-381-3p and OIP5-AS1 or HMGB3 was confirmed by dual-luciferase reporter assay and pull-down assay. In vivo experiments were conducted using murine xenograft models. Results OIP5-AS1 was elevated in trastuzumab-resistant breast cancer cells, and OIP5-AS1 knockdown rescued trastuzumab sensitivity. Extracellular OIP5-AS1 was packaged into exosomes, which were secreted by trastuzumab-resistant cells, and could be absorbed by trastuzumab-sensitive cells in breast cancer. Importantly, intercellular transfer of OIP5-AS1 via exosomes enhanced trastuzumab resistance in vitro. OIP5-AS1 was a sponge of miR-381-3p; besides, miR-381-3p targeted HMGB3. Murine xenograft analysis showed exosomal OIP5-AS1 induced trastuzumab resistance in vivo. Exosomal OIP5-AS1 was dysregulated in the serum of breast cancer patients, and might be a promising diagnostic biomarker in trastuzumab resistance. Conclusion Intercellular transfer of OIP5-AS1 by exosomes enhanced trastuzumab resistance in breast cancer via miR-381-3p/HMGB3


Background
Breast cancer is the most commonly diagnosed female malignancy on a global scale, which is a serious threat to the health of women worldwide [1]. A cancer cell has a variety of receptors on its surface, which can be bound by chemicals to result in changes within the cancer cell. HER2 is a member of the epidermal growth factor receptor (EGFR) family that occurs in about one-third of all breast cancers [2]; importantly, HER2 amplification cause aggressive cancer phenotype and poor clinical outcome [3,4]. Trastuzumab, a humanized antibody of HER2, is thought to be a successful strategy to block HER2 receptors when there is overexpression, thereby blocking growth of the cancer and prolonging the overall survival of HER2 + breast cancer patients in adjuvant and metastatic settings [5,6]. Nevertheless, the response rate to trastuzumab-containing therapies still needs to be improved due to the development of acquired trastuzumab resistance [7]. Therefore, a better understanding on the mechanism of trastuzumab resistance is essential for developing new approaches to overcome trastuzumab resistance in breast cancer.
Recently, it has been documented that cells can communicate with each other via micro vesicles [8]. Exosomes are one of extracellular vesicles, contain lipids, protein, noncoding or coding RNAs, can be secreted by numerous cell types, like cancer cells, and can be absorbed by other cells to transfer and exchange the cargo [9,10], thus, exosomes are considered as potential modes of intercellular communication [11]. Recently, emerging evidence has revealed that exosomes potentially impact therapeutic response of the recipient cells via the transfer of proteins and lncRNAs [12]. Long noncoding RNA OPAinteracting protein 5 antisense transcript 1 (OIP5-AS1), mapped on chromosome 15q15.1, is a novel identified and promising tumor-associated lncRNA. It has been demonstrated that OIP5-AS1 plays various roles in multiple cancers, and contributes to the deteriorate of 4 malignant tumors [13]. In breast cancer, OIP5-AS1 was found significantly up-regulated, and functioned as an oncogene via regulating cell malignant phenotypes [14]. Whereas, the role of OIP5-AS1 in trastuzumab resistance in breast cancer still needs to be explored.
Thus, we attempted to elaborate the functions of OIP5-AS1 in trastuzumab resistance in breast cancer, explored whether the exosome-transmitted OIP5-AS1 conferred drug resistance to recipient cells, as well as the potential molecular mechanism underlying OIP5-AS1 effects on breast cancer.

Patients and specimens
Blood samples were collected from 57 breast cancer patients diagnosed by histopathological examination at The Central Hospital Of Enshi Tujia And Miao Autonomous Prefecture. All blood samples were centrifuged at 3000 g for 10 min after collecting for 1 h, and the supernatant serum was collected using RNase-free tubes and stored at -80℃ until used. All patients only received trastuzumab-based neo-adjuvant chemotherapy and were classified into trastuzumab-resistant (Non-response, N=30) and tamoxifen-sensitive (Response, N=27) depending on the sensitivity to trastuzumab.

Cell culture
Human breast cancer cell lines SKBR3 and BT474 were purchased from Shanghai Academy of life Science (Shanghai, China) and cultured in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) harboring with 15% fetal bovine serum (FBS) and 0.1 IU/mL insulin at 37℃ with 5% CO 2 . Trastuzumab resistant breast cancer cells, named SKBR3-TR and BT474-TR were established by continuously exposing parental cells to increasing concentration of trastuzumab (Sigma, St. Louis, MO, USA) for more than 6 months until cells displayed resistance to trastuzumab. Trastuzumab resistant cells were maintained in the same 5 media supplemented with 3 μg/mL trastuzumab.

Cell viability assay
Parental or resistant cells (5000 cells/well) were seeded in 96-well plates overnight.
Following transfection or trastuzumab treatment (0, 0.3125, 0.625, 1.25, 2.5, 5, or 10 μg/mL) for additional 48 h, cells in 96-well plates were incubated with 10 μL cell counting kit-8 solution (Sigma) for 4 h at 37℃. The optical density (OD) at 450 nm was determined by a microplate reader and the half maximal inhibitory concentration (IC50) value of trastuzumab was assessed on the basis of the relative survival curve.

Quantitative real-time polymerase chain reaction (qRT-PCR)
Whole-RNA extracts from parental or resistant cells were prepared using TRIzol reagent

Colony formation assay
Transfected parental or resistant cells (5000/well) suspended in RPMI-1640 medium with 0.5 μg/mL trastuzumab were seeded in 6-well plates. After 21 days of cultures at 37℃ with 5% CO 2 , cell colonies were fixed with methanol and stained with 0.1% crystal violet.
Finally, the number of visible colonies (≥50 cells) was counted.

Cells migration assay
The migratory capacity of cells was performed by a 24-well transwell chamber (8 μm; Corning Costar, Cambridge, MA). Transfected cells suspended in serum-free RPMI-1640 medium were filled the top chambers. Then 500 μL RPMI-1640 medium mixed with 10% FBS was added into the lower chambers. 24 h later, cells on the lower face of the membranes were fixed and stained. Finally, migrated cells in five random fields were counted with a microscope.
In brief, after transfection with the designed vector for 48 h, cells were interacted with 5 μL FITC annexin V and 10 μL PI. Finally, the apoptotic rate was measured by a flow cytometer.

Exosome (exo) isolation
Exosomes were isolated from serum samples or cells using ultracentrifuge method. Cell culture fluid from exosome-depleted medium or serum was centrifuged at 3000 g for 30 7 min at 4℃ to remove cell fragments. Then the resulting supernatant was further centrifuged at 100,000 g for 70 min at 4℃, and filtered using 0.22 μm filtration.
Subsequently, pelleted exosomes was washed with PBS, and centrifuged at 100,000 g for 70 min again. Finally, purified exosomes were resuspended in PBS for the detection of the size and quality of exosomes using nanoparticle tracking analysis (NTA) software or for functional assays. Exosomes pellets were interacted with Trizol reagent to isolate RNA, and were lysed with RIPA lysis buffer used for protein detection. For blocking of exosome release, parental or resistant breast cancer cells were treated with GW4869 (10 μM) or Vehicle (as control) for 48 h. For exosome co-cultures, exosomes (50 μg/ml) were incubated with parental SKBR3 and BT474 cells (5 × 10 5 ) in a 6-well plate with 10% FBS exosome-depleted culture medium for 48 h.

Transmission electron microscopy (TEM)
Purified exosomes were dropped on the carbon-coated copper grid and allowed to absorb for 5 min at 37℃, and then stained with 2% phosphotungstic acid solution for 2 min, followed by washing with PBS for 3 times. After air-dried, the grid was visualized using a transmission electron microscope (TEM) (JEOL, Akishima, Japan).

Western blot
Proteins were extracted from cells or exosomes using RIPA lysis buffer (Beyotime, Beijing, China) and quantified determined using a bicinchoninic acid (BAC) Protein Assay Kit

Pull-down assay
Biotin (Bio)-miR-381-3p and Bio-NC synthesized by Genepharma Company were transfected into SKBR3 and BT474 for 48 h. Then, cells were lysed, and the lysates were incubated with M-280 streptavidin magnetic beads (Invitrogen). After elution, the beadbound RNA complex was purified and subjected for qRT-PCR analysis.

Xenograft experiments in vivo
Female BALB/c mice (Five-week-old) from Jinan Pengyue Animal Center (Jinan, China) were randomly divided into four groups (N = 5 each). Each group was injected with BT474 cells (1 × 10 6 ) in the flank region of mice. When tumors grew to 100 mm 3 , groups 2, 3, and 4 were intratumorally injected with trastuzumab (3mg/kg) every 2 days, and the negative control of group 1 was injected with PBS; besides, isolated exosomes (10 μg) from BT474-TR cells loaded with si-OIP5-AS1 lentiviral vector (si-OIP5-AS1#1) or si-NC lentiviral vector (si-NC) were injected into the center of tumor of groups 4 and 3 every two days, respectively. Tumor volume was calculated every 4 days. At day 32, all mice were sacrificed and tumor masses were weighed and harvested for further molecular analysis.

Statistical analysis
Numerical results from three independent experiments were manifested as the mean ± standard deviation (SD). The statistical difference between each group was analyzed by Student's t test or one-way analysis of variance (ANOVA) with GraphPad Prism 7 software.
Receiver operating characteristic (ROC) curves were plotted to analyze the diagnostic value of exosomal OIP5-AS1. P values less than 0.05 was considered as statistically significant.

OIP5-AS1 is up-regulated in trastuzumab-resistant breast cancer cells
To explore the role of OIP5-AS1 in trastuzumab resistance, trastuzumab-resistant breast cancer cells, named SKBR3-TR and BT474-TR, were established using the parental SKBR3 and BT474 cells to continuously expose with increasing concentration of trastuzumab for more than 6 months. Then cell viability was detected in parental and resistant cells.
Results showed the viability of parental SKBR3 and BT474 cells was significantly inhibited by trastuzumab in 0.3125-10 μg/mL compared with SKBR3-TR and BT474-TR cells, and the IC 50 values of SKBR3-TR and BT474-TR cells to trastuzumab were markedly higher than that in SKBR3 and BT474 cells (Fig. 1A, B); thus, SKBR3-TR and BT474-TR cells were defined as resistance. Afterwards, the level of OIP5-AS1 was analyzed in breast cancer cells, we found OIP5-AS1 was markedly higher in SKBR3-TR and BT474-TR cells than that in corresponding parental SKBR3 and BT474 cells (Fig. 1C, D). These data indicated that OIP5-AS1 increased might be related to trastuzumab resistance in breast cancer.

OIP5-AS1 knockdown restores trastuzumab sensitivity in trastuzumab resistant breast cancer cells
To investigate the detailed functions of OIP5-AS1 in trastuzumab resistance, we knocked down OIP5-AS1 by transfecting with constructed si-OIP5-AS1 plasmid into SKBR3-TR and BT474-TR cells. As shown in Fig 2A, si-OIP5-AS1 of three forms notably reduced OIP5-AS1 expression compared with the si-NC group, and then sh-OIP5-AS1#1 was selected for subsequent analysis. By contrast with si-NC group, OIP5-AS1 down-regulation significantly reduced the IC 50 values of SKBR3-TR and BT474-TR cells to trastuzumab (Fig. 2B); also, colony formation analysis demonstrated that OIP5-AS1 knockdown combined with trastuzumab markedly decreased the number of colonies formed of SKBR3-TR and BT474-TR cells (Fig. 2C, D). Meanwhile, transwell assay displayed the number of migrated and invaded SKBR3-TR and BT474-TR cells was suppressed by OIP5-AS1 knockdown (Fig. 2E,   F). Besides that, OIP5-AS1 depletion elevated cell apoptosis in SKBR3-TR and BT474-TR cells (Fig. 2G, H). Taken together, OIP5-AS1 knockdown impeded breast cancer cells to trastuzumab resistance.

Exosomal OIP5-AS1 derived from trastuzumab resistant cells can be absorbed by trastuzumab-sensitive cells in breast cancer
Exosomes can be actively secreted by most cell types, and exosome-contained lncRNAs can be transmitted and exchanged into culture medium. In order to explore the impact of exosome transfer on trastuzumab resistance in breast cancer, we first isolated exosomes from the conditioned medium supernatant of SKBR3-TR and BT474-TR cells. The vesicles displayed a round shape with bilayered membranes, and the diameter from 40 nm to 250 nm under a TEM (Fig. 3A). Then NTA analysis further confirmed that the predominant size of the vesicles was 100 nm (Fig. 3B); also, as exhibited by western blot analysis, the exosomal markers CD63, CD81 and TSG101 was detectable in exosomes but not in cell lysates (Fig. 3C). Additionally, the expression change of OIP5-AS1 in the culture medium of SKBR3-TR and BT474-TR cells was detected after treatment with RNase. Results exhibited that OIP5-AS1 expression in culture medium was little affected upon RNase treatment, but 11 was greatly reduced when treated with RNase and Triton X-100 simultaneously (Fig. 3D), indicating that OIP5-AS1 presented in exosomes. The treatment of SKBR3-TR and BT474-TR cells with GW4869, an inhibitor of the secretion of exosomes from cells, led to a reduction in the quantity of exosomes (Fig. 3E) as well as OIP5-AS1 expression (Fig. 3F).
All these results indicated that extracellular OIP5-AS1 was packaged into exosomes, which were also secreted by SKBR3-TR and BT474-TR cells.
Next, exosomes from the culture supernatant of SKBR3 and BT474 cells were isolated, and qRT-PCR analysis revealed OIP5-AS1 expression in SKBR3-TR and BT474-TR secreted exosomes was significantly higher than that in exosomes from parental cells (SKBR3 and BT474) (Fig. 3G). Importantly, SKBR3 and BT474 cells co-cultured with exosomes derived from SKBR3-TR and BT474-TR cells were treated with GW4869 or Vehicle, and we found the levels of OIP5-AS1 was increased by exosomes incubation, but was decreased by GW4869 treatment (Fig. 3H), revealing OIP5-AS1 could be transmitted into trastuzumabsensitive cells via exosome transfer.

OIP5-AS1 is a sponge of miR-381-3p
To explore the underlying mechanism of OIP5-AS1, miRNA targets were searched through starBase program, and miR-381-3p was found that might be a target of OIP5-AS1 (Fig. 5A).
To confirm this prediction, first, SKBR3 and BT474 cells were transfected with miR-NC or miR-381-3p, and miR-381-3p expression was significantly elevated by miR-381-3p transfection compared with miR-NC (Fig. 5B). Then a dual luciferase reporter assay was conducted and results showed that the luciferase activity of OIP5-AS1 wt of both forms was decreased in SKBR3 and BT474 cells treated with miR-381-3p; however, there was no significant difference in luciferase activity of on the two forms OIP5-AS1 mut (Fig. 5C, D).

HMGB is a target of miR-381-3p
According to the prediction of starBase program, miR-381-3p was found to have the binding sites on HMGB (Fig. 6A). The results of dual luciferase reporter assay displayed that miR-381-3p overexpression reduced the luciferase activity of HMGB wt of both forms but not mutant reporter vector in SKBR3 and BT474 cells (Fig. 6B, C), indicating that miR-381-3p directly targeted to HMGB. Meanwhile, western blot analysis exhibited that miR-381-3p restoration suppressed HMGB expression, while this inhibition was reversed by OIP5-AS1 overexpression in SKBR3 and BT474 cells (Fig. 6D). Therefore, miR-381-3p targetedly suppressed HMGB expression and OIP5-AS1 positively regulated HMGB via miR-381-3p. Interestingly, HMGB was increased in SKBR3-TR and BT474-TR cells compared with their parental SKBR3 and BT474 cells (Fig. 6E); thus, HMGB was also linked to trastuzumab resistance in breast cancer.

Serum exosomal OIP5-AS1 level is associated with trastuzumab resistant in breast cancer patients
We further attempted to analyze the expression level of exosomal OIP5-AS1 in 57 serum samples from breast cancer patients receiving trastuzumab treatment. Results implied that the serum exosomal OIP5-AS1 remarkably higher in patients who did not respond to treatment than in those who responded to trastuzumab (Fig. 8A). Afterwards, the diagnostic potential of exosomal OIP5-AS1 in serum was calculated. As revealed by ROC analysis, an area under curve (AUC) of 0.764 with a diagnostic sensitivity and specificity reaching 59.26% and 93.33% respectively (95% CI = 0.6326-0.8958) was observed (Fig.   8B). Additionally, patients were classified into a low and high exosomal OIP5-AS1 expression groups according to the cut-offs (1.355) established by ROC, and we found the proportion of patients not responding to trastuzumab was greatly higher in the high exosomal OIP5-AS1 expressing group than in the low expressing group (Fig. 8C). These data indicated that exosomal OIP5-AS1 was also dysregulated in the serum of breast cancer patients and might be a promising diagnostic biomarker for trastuzumab resistance in patients with breast cancer.

Discussion
Molecular targeted therapies are one of the major and revolutionized modalities of medical treatment, which interfere with specific molecules to block cancer growth, survival, and progression [15]. In breast cancer, some molecular targeted therapies, including trastuzumab and lapatinib, have been approved and demonstrated remarkable clinical success in the treatment of the cancer. They direct against HER2 and bevacizumab, as well as vascular endothelial growth factor (VEGF) [16,17]. LncRNAs are a class of RNAs with lengths exceeding 200 nucleotides and have been widely revealed to implicate in various physical and pathological processes, such as cell differentiation, proliferation, apoptosis and metabolism, thereby affecting development and progression of cancers [18][19][20][21]. They are considered as candidate therapeutic targets.
OIP5-AS1 is a functional RNA, and has been identified to sever as an oncogene to participate in the evolution of tumorigenesis and drug resistance in many types of cancer.
For example, OIP5-AS1 accelerated the progression of gastric cancer by regulating HMGA2 through miR-367-3p [22]. OIP5-AS1 interacted with ROCK1 to promote cell carcinogenesis in cervical cancer via absorbing miR-143-3p. OIP5-AS1 induced cisplatin resistance in osteosarcoma through regulating the LPAATβ/PI3K/AKT/mTOR signaling pathway via a mechanism involving miR-340-5p [23]. Importantly, OIP5-AS1 was confirmed to act as an oncogene via regulating cell malignant phenotypes in breast cancer [14]. Thus, we thought OIP5-AS1 might also involve in the regulation of drug resistance, and then confirmed that knockdown of OIP5-AS1 restored trastuzumab sensitivity in trastuzumab resistant breast cancer cells.
Exosomes are one of extracellular vesicles, and have been reported to involve in the modulation of chemoresistance of the recipient cells via the transfer of lncRNAs in diverse cancers [24,25]. In breast cancer, the effects of exosomal lncRNA in drug resistance were also demonstrated. For instance, exosome-mediated transfer of AFAP1-AS1 induced trastuzumab resistance in breast cancer via the interaction with AUF1 and the activation of ERBB2 translation [26]. Intercellular transfer of lncRNA H19 promoted doxorubicin resistance in breast cancer [27]. Thus, whether exosomal OIP5-AS1 implicated in trastuzumab resistance in breast cancer was investigated. Results exhibited that extracellular OIP5-AS1 was packaged into exosomes, which were secreted by trastuzumab resistant cells, and could be taken up by trastuzumab sensitive cells via exosomes transfer, thereby disseminated trastuzumab resistance to recipient cells in vitro as well as inhibited cytotoxicity induced by trastuzumab in tumor growth in vivo in breast cancer.
Additionally, exosomal OIP5-AS1 was also dysregulated in the serum of breast cancer patients, and might act as a promising diagnostic biomarker for trastuzumab resistance in breast cancer patients. Therefore, exosomal OIP5-AS1 may be useful for the treatment of breast cancer and the prediction of trastuzumab resistance. The potential underlying molecular mechanism of the action of OIP5-AS1 was further probed. In this study, we found OIP5-AS1 directly bound to miR-381-3p and subsequently served as a miRNA sponge to up-regulate the expression of the miR-381-3p target gene HMGB3. MiR-381-3p is a wellrecognized anti-tumor miRNA, which overexpression was confirmed to suppress cell proliferation, cell cycle progression and migration in breast cancer [28]. HMGB3 belongs to HMGB family, and plays significant effects on DNA repair, recombination, transcription and replication [29]. Besides that, it has been demonstrated to HMGB3 silence could inhibit cell growth and progression in breast cancer [30,31]. Thus, we considered that exosomal OIP5-AS1 may confer trastuzumab resistance in breast cancer cells via the regulation of miR-381-3p/HMGB3 axis.

Conclusion
This study demonstrated that exosome-mediated transfer of OIP5-AS1 partially induced trastuzumab resistance in breast cancer through miR-381-3p/HMGB3 axis, indicating a therapeutic strategy for the trastuzumab resistance in patients with breast cancer.