Cancer-Derived Exosomal miR-651 as a Diagnostic Marker Restrains Cisplatin Resistance and Directly Targets ATG3 for Cervical Cancer

Objective Cancer-derived exosomes can facilitate drug resistance in cervical cancer. However, the mechanisms remain elusive. Herein, we observed the roles of exosomal miR-651 in cisplatin resistance of cervical cancer. Methods Circulating miR-651 was detected in cervical cancer and healthy individuals. The diagnostic efficacy was determined. When transfected with miR-651 mimics, cisplatin resistance, apoptosis, and proliferation were assessed. The cancer-derived exosomes were separated and identified. We observed the uptake of PKH67-labeled exosomes by HeLa/S cells. After coculture with exosomes secreted by HeLa/S or HeLa/DDP cells, malignant behaviors were examined in HeLa/S cells. The interactions between ATG3 and miR-651 were validated by dual luciferase report. Biological behaviors were investigated for HeLa/S cells cocultured with exosomes secreted by miR-651 mimic-transfected HeLa/DDP cells. Results Downregulated circulating miR-651 was found in cancer subjects than healthy individuals. It possessed high sensitivity and accuracy in diagnosing cervical cancer (AUC = 0.9050). Lower miR-651 expression was confirmed in HeLa/DDP than HeLa/S cells. Forced miR-651 lessened cisplatin resistance and proliferation and elevated apoptosis in HeLa cells. ATG3 was a direct target of miR-651. The exosomes isolated from HeLa cells were rich in CD63, CD9, and CD81 proteins, thereby identifying the isolated exosomes. Exosomes secreted by HeLa/DDP cells can be absorbed by HeLa/S cells. When being cocultured with exosomes secreted by HeLa/DDP cells, malignant behaviors of HeLa/S cells were enhanced, which were ameliorated by miR-651 mimic exosomes. Conclusion Our findings showed that cancer-derived exosomal miR-651 restrained cisplatin resistance and directly targeted ATG3, indicating that exosomal miR-651 could be a therapeutic agent.


Introduction
Cervical cancer is a major killer that seriously threatens women's health [1]. Although the application of cervical cancer vaccine can effectively reduce the incidence of cervical cancer, mortality as well as morbidity of this malignancy still ranks fourth among women worldwide [2]. The current treatment methods for cervical cancer are comprehensive such as surgery, chemotherapy, radiotherapy, and immunotherapy [3]. Nevertheless, not all subjects can achieve satisfactory outcomes, especially for subjects with advanced stages [4][5][6]. For locally advanced individuals, the current treatment strategy is neoadjuvant chemotherapy followed by surgery for chemotherapy-sensitive patients, while insensitive patients are directly converted to radiotherapy [7]. If there are indicators that can effectively predict the sensitivity of patients to chemotherapy, this part of the patients who are not sensitive to chemotherapy can save the time and expense of neoadjuvant chemotherapy and directly transfer to radiotherapy to obtain better outcomes [8]. Therefore, continuing to study the pathogenesis of cervical cancer, looking for specific therapeutic targets and markers for predicting chemotherapy sensitivity is still extremely critical for cervical cancer therapy.
Exosomes are extracellular vesicles with a diameter of 50-140 nm, which are formed by cells through regulation processes such as endocytosis, fusion, and efflux [9][10][11]. Exosomes carry signals between cells by transmitting small biologically active molecules substances like microRNAs (miRNAs) [12]. Exosomes can be separated from various body fluids and cell culture supernatants, and their function and composition are determined by their cell sources [13]. Exosomes are abundant in the microenvironment of malignant tumors, which are involved in the processes of invasion, metastases, angiogenesis, and chemotherapy resistance of malignant tumors [14,15]. miRNA is an endogenous, noncoding single-stranded small RNA, with 19-23 nucleotides in length [16]. Mature miRNAs have the functions of inhibiting target mRNA expression after transcription [17]. Various miRNA markers have been identified for cervical cancer. Exosomal miRNAs have been promising players for cervical cancer [18]. Previous research has highlighted the impacts of miR-651 on non-small cell lung cancer [19] and nasopharyngeal carcinoma [20]. Unfortunately, it remains uncharted concerning miR-651 on cervical cancer. Here, our study identified circulating miR-651 as a diagnostic marker upon this malignancy. Moreover, forced exosomal miR-651 could restrain cisplatin resistance and directly target ATG3 for cervical cancer cells, suggesting the action of miR-651 on cervical cancer progress.

Patients and Specimens.
From January 2018 to January 2019, 30 newly diagnosed cervical cancer patients were recruited in the Daping Hospital, Army Medical University. All patients were confirmed histopathologically. Meanwhile, 30 healthy age-matched individuals were selected. This study excluded cervical cancer subjects who had undergone surgery, radiotherapy, chemotherapy, etc. Furthermore, subjects with a history of severe organ disease or other systemic tumors were also excluded. All participants provided written informed consent. This study met the requirements of the Ethics Committee of Daping Hospital, Army Medical University (2018002). 5 mL whole blood samples were harvested from each subject by using EDTA anticoagulation tube. Following centrifugation and separation, samples were collected in an EP tube and stored at -80°C for later use.

Transfection.
HeLa cells were seeded on a 24-well plate (4 × 10 4 cells/well). On the second day, the cells were transfected with miR-651 mimics (Genepharma, Shanghai, China), miR-651 inhibitors (Genepharma, Shanghai, China), or negative control (NC; Genepharma, Shanghai, China). 50 μL serum-free medium Opti-MEM was used to dilute 1.25 μL miR-651 mimics, miR-651 inhibitors, or NC at a concentration of 20 μmol/L and incubated for 5 min at room temperature. Similarly, 1 μL Lipofectamine™ 2000 was diluted with 50 μL serum-free medium Opti-MEM and incubated at room temperature for 5 min. The diluted miR-651 mimics, miR-651 inhibitors, or NC were mixed with Lipofectamine™ 2000 and incubated for 20 s, which were then added to 24 plates. After culturing for 5 h, the mixed culture medium was removed and fresh culture medium was replaced, and then culturing was continued for 24 h. . HeLa cells were inoculated into 96-well plates (8 × 10 3 cells/well). After continuing to incubate for 12 h, the DMEM medium was discarded, and 200 μL DMEM medium was added containing 0, 50, 100, 200, 400, and 800 μg/mL DDP. After 48 h, 20 μL CCK-8 solution (Dojindo, Japan) was added to each well. After culturing for 2 h, the culture medium was discarded. The absorbance at 450 nm was measured with a multifunctional microplate reader. The half inhibitory concentrations (IC50) of DDP were determined on HeLa cells.

Cell Counting
2.6. Flow Cytometry. The cells were trypsinized into a single cell suspension and washed with PBS. The cells were inoculated into a 6-well plate (2:0 × 10 5 /well). 5 μL Annexin V-FITC and 5 μL PI (Sigma, USA) were added, separately. After mixing thoroughly, the cells were incubated for 10 min at room temperature in the dark. The cells were tested on the flow cytometer (BD, Germany). 2.11. Uptake of Exosomes. PKH67 is a membrane labeling dye that can bind to the lipid membrane of exosomes and emit green fluorescence, which can be used to identify the presence of exosomes. 100 μL exosomes were incubated with 1 μL PKH-26 dye, which were then incubated by 1 mL Diluent C, 200 μL 1% BSA/PBS, and 3 μL PKH-26 dye for 20 min. Under centrifugation at 100 000 g for 70 min, the PKH67-labeled exosome pellet was obtained. The HeLa/S cells were cultured in a 12-well plate. When they reached 70% confluence, they were replaced with a fresh medium containing PKH67-labeled exosomes and incubated for 24 h. After washing with PBS, the cells were fixed with paraformaldehyde (Sigma, USA) for 20 min. DAPI was used for staining the nucleus. The uptake of exosomes by HeLa/S cells was investigated under a confocal fluorescence microscope (Leica, Germany).

Western
Blot for ATG3, LC3II/I, and p62. Tissues and cells were added with protein lysis buffer containing 100× protease and phosphatase inhibitors and repeatedly pipetted. After centrifugation, the supernatant was transferred to a new EP tube. BCA (Sigma, USA) measured the total protein concentration. After boiling and denaturing, the sample was loaded on 8% SDS-PAGE electrophoresis. The proteins were transferred to the PVDF membrane (Millipore, USA). The 5% skimmed milk powder was used for sealing membranes at room temperature for 2 h. After washing 3 times with TBST, membranes were incubated with primary antibodies containing ATG3 (1/500; ab233562; Abcam, USA), LC3I/II (1/1000; ab128025; Abcam, USA), p62 (1/1000; ab91526; Abcam, USA), and GAPDH (1/10000; ab181602; Abcam, USA) at 4°C overnight. Following washing with TBST,    5 Disease Markers secondary antibody (1/10000; ab7090; Abcam, USA) incubation was presented for 2 h at room temperature. The bands were exposed by ECL chemiluminescence. GAPDH was utilized as an internal reference.
2.14. Statistical Analysis. The measurement data are expressed by the mean ± standard deviation. SPSS 18.0 software (SPSS Inc., USA) was utilized for statistical analysis.
Comparisons between groups were performed through Student's t test or one-way ANOVA. Receiver operating characteristic curves (ROCs) were drawn, and area under the curve (AUC) was calculated for assessing the diagnostic potential of circulating miR-651 in cervical cancer. ATG3 expression was analyzed using the UALCAN database (http://ualcan .path.uab.edu/analysis.html). Pearson's analysis was presented between miR-651 and ATG3 in 30 pairs of cervical cancer and healthy individuals. p value < 0.05 indicated statistical significance.

Circulating miR-651 as a Prognostic Marker for Cervical
Cancer. This study recruited 30 cervical cancer patients and 30 healthy individuals. Circulating miR-651 expression was detected via RT-qPCR. Data suggested that lowered miR-651 expression was found in cervical cancer plasma than normal specimens (p < 0:0001; Figure 1(a)). We assessed  the diagnostic potential of circulating miR-651 in cervical cancer. In Figure 1(b), the data showed that circulating miR-651 displayed a highly sensitive and accurate capacity for diagnosing cervical cancer (AUC = 0:9050, p < 0:0001).
The above findings were indicative of circulating miR-651 as an underlying diagnostic marker of cervical cancer.

Forced miR-651 Exerts Inhibitory Action on Cisplatin Resistance and Proliferation and Motivates Apoptosis in
Cervical Cancer Cells. Then, this study assessed whether forced miR-651 ameliorated cisplatin resistance of cervical cancer cells. HeLa/S as well as HeLa/DDP cells were separately transfected with miR-651 mimics. By RT-qPCR, miR-651 expression was determined. As expected, forced miR-651 expression was confirmed in HeLa/S and HeLa/DDP (both p < 0:0001) cell lines (Figure 2 and 2(l)). The above data suggested that forced miR-651 restrained proliferative capacity of cervical cancer cells.

Sensitive Cervical Cancer Cells Absorb Exosomes Secreted by Cisplatin-Resistant Cancer Cells.
This study observed the functions of exosomes on cisplatin resistance in cervical cancer. We isolated exosomes in the culture supernatant of HeLa/DDP cells through high-speed centrifugation. To evaluate whether exosomes were successfully isolated, western blot was utilized for examining exosome surface biomarkers including CD63, CD9, and CD81 in white precipitate samples. In Figure 3(a), white precipitate samples exhibited CD63, CD9, and CD81 expression, which was indicative of the success isolation of exosomes. We then investigated whether HeLa/S cells possessed the functions of absorbing exosomes secreted by cisplatin-resistant cancer cells. Following coculturing cisplatin-resistant exosomes and HeLa/S cells, this study found that PKH67-labeled green fluorescence exhibited a uniform distribution in HeLa/S cellular cytoplasm (Figure 3(b)). These findings confirmed that exosomes secreted by cisplatin-resistant cancer cells can be absorbed by sensitive cervical cancer cells.    Figures 4(b) and 4(c)).
Meanwhile, compared to exosomes from HeLa/S cells, the increase in colony formation capacities of HeLa/S cocultured with exosomes from HeLa/DDP cells was confirmed (p < 0:01). Hence, DDP-resistant exosomes elevated proliferation of colony formation. As shown in flow cytometry, fol-lowing coculture with exosomes from HeLa/DDP cells, apoptosis of HeLa/S cells was distinctly restrained (p < 0:05; Figures 4(d) and 4(e)). In comparison to coculture of exosomes from HeLa/S cells, there was a marked decrease in apoptosis of HeLa/S cells cocultured by exosomes from HeLa/DDP cells (p < 0:05). The above data suggested that cisplatin-resistant exosomes lowered apoptotic levels of cervical cancer cells.

Discussion
This study identified circulating miR-651 was downregulated in cervical cancer, which could be utilized as a marker for diagnosing this malignancy. Cancer-derived exosomal miR-651 may ameliorate cisplatin resistance and malignant progress of cervical cancer, highlighting exosomal miR-651 as a therapeutic agent against cervical cancer.
The miRNAs carried by exosomes are highly conservative and stable, which have been considered diagnostic markers or therapeutic agents in cervical cancer [22]. Zheng et al. confirmed exosomal let-7d-3p and miR-30d-5p as markers for diagnosing cervical cancer [23]. Konishi et al. confirmed that exosomal miR-22 could become a promising drug delivery system concerning cervical cancer radiotherapies [24]. Ma et al. proposed a circulating miRNAsignature containing miR-146a-5p, miR-151a-3p, miR-2110, and miR-21-5p upon cervical cancer diagnosis [25]. This study confirmed the downregulation of circulating miR-651 in cervical cancer. The AUC was 0.9050, suggesting that circulating miR-651 could be a sensitive and accurate diagnostic marker for cervical cancer. Its diagnostic value will be validated in a cohort with a larger sample size. Intriguingly, lower miR-21-5p expression was confirmed in HeLa/DDP than HeLa/S cells and forced its expression reduced IC50 values of DDP, indicating that miR-21-5p downregulation was in relation to cisplatin resistance. Previously, Shi   14 Disease Markers chemosensitivity of this malignancy through transketolase [27]. These findings highlighted the implications of miR-NAs on drug resistance. Furthermore, our data showed that forced miR-651 induced apoptosis and weakened proliferation of HeLa cells, indicating miR-651 as a tumor suppressor gene. In this study, ultracentrifugation was used to successfully isolate the exosomes of HeLa/S and HeLa/DDP cells. HeLa/S cells may absorb the exosomes secreted by HeLa/DDP cells, consistently with previous research [21]. We found that HeLa/DDP cell-derived exosomes can promote the proliferation and inhibit their apoptosis in HeLa/S cells, indicating that the exosomes secreted by HeLa/DDP cells may induce malignant transformation. Meanwhile, exosomes secreted from HeLa/DDP cells induced cisplatin resistance of HeLa/S cells. A previous study demonstrated that miR-106a/b from cisplatin resistant liver cancer cells may facilitate cisplatin resistance in cervical cancer cells [28]. Here, we found that exosomes secreted from miR-651 mimic-transfected HeLa/DDP cells weakened proliferation and lowered apoptotic levels for HeLa/S cells. ATG3 was a direct target of miR-651 in cervical cancer. MiR-651 carried by exosomes lessened ATG3 expression of HeLa/S cells. Chen et al. developed an autophagy-related model that can be predictive of cervical cancer subjects' outcomes [29]. Suppression of autophagy may weaken the progress of this malignancy [30]. Thus, cancer-derived exosomal miR-651 could restrain malignant behaviors of cervical cancer cells through ATG3. Previously, targeting autophagy could ameliorate metastasis and chemoresistance for various malignancies [31][32][33]. Furthermore, exosomes secreted from miR-651 mimicstransfected HeLa/DDP cells restrained the autophagy of HeLa/S cells. These data indicated that exosomal miR-651 ameliorated drug resistance of cervical cancer by lessening autophagy. However, several limitations of our study should be pointed out. Firstly, the clinical significance of exosomal miR-651 will be verified in a larger cervical cancer cohort. Second, the role of exosomal miR-651 on cervical cancer progression should be validated in vivo.

Conclusion
Collectively, this study showed that cancer-derived exosomal miR-651 may restrain cisplatin resistance and progression and directly targeted ATG3 in cervical cancer. Hence, exosomal miR-651 could be a therapeutic agent against cervical cancer.

Data Availability
The datasets analyzed during the current study are available from the corresponding author on reasonable request.