Exosome-transmitted LINC00960 and LINC02470 promote the epithelial-mesenchymal transition and aggressiveness of bladder cancer cells CURRENT STATUS: POSTED

Background Exosomes are essential for several tumor progression-related processes, including epithelial-mesenchymal transition (EMT). Long noncoding RNAs (lncRNAs) comprise a major group of exosomal components and regulate the neoplastic development of several cancer types; however, the progressive roles of exosomal lncRNAs in bladder cancer have rarely been addressed. In this study, we identified two potential aggressiveness-promoting exosomal lncRNAs, LINC00960 and LINC02470; we found that these lncRNAs potently induced EMT during bladder cancer progression. Methods Low-grade bladder cancer cells (TSGH-8301) were treated with conditioned media or exosomes derived from high-grade bladder cancer cells (T24 or J82), and the aggressiveness-promoting effects were evaluated. Cell viability, cell migratory/invasive activities and clonogenicity were compared to assess the response to these intercellular transmissions. Exosome-transmitted lncRNA candidates were screened with bioinformatic pipelines, and their expression levels were validated in bladder cancer cells and exosomes. Two novel lncRNAs, LINC00960 and LINC02470, were selected, and their roles and regulatory mechanisms in inducing the aggressiveness of bladder cancer cells were investigated. group. In summary, either T24 siLINC00960-Exos or T24 siLINC02470-Exos attenuated the T24-Exos-induced EMT process and its related regulatory mechanism in recipient TSGH-8301 cells. These results suggest that both LINC00960 and LINC02470 play comprehensive roles in EMT regulation and thereby promote the aggressiveness of recipient TSGH-8301 cells. Our findings indicate that exosome-transmitted LINC00960 and LINC02470 derived from high-grade bladder cancer cells promote the aggressive behavior of low-grade bladder cancer cells through intercellular communication and promote EMT by promoting β-catenin signaling, Notch signaling and Smad2/3 signaling. Both lincRNAs may serve as potential liquid biomarkers for the prognostic surveillance of bladder cancer patients in the future.

In this study, two potent candidate lncRNAs, LINC00960 and LINC02470, were screened out based on a series of selective criteria. After validation of their intracellular and exosomal expression levels, their biological effects and molecular regulation were evaluated, and the results indicated that exosome-transmitted LINC00960 and LINC02470 from high-grade bladder cancer cells aggravates low-grade bladder cancer cell malignant behaviors and promoted the EMT process.

Materials And Methods Cell lines
Four human bladder cancer cell lines (low-grade, TSGH-8301 and TSGH-9202, and high-grade, T24 and J82; originally acquired from ATCC or Bioresource Collection and Research Center) were used in this study. All cells were incubated in RPMI 1640 medium containing 10% fetal bovine serum, 1 µg/ml penicillin and 1 µg/ml streptomycin (Life Sciences, Palo Alto, CA, USA) at 37 °C in a 5% CO 2 humidified incubator.

Isolation of exosomes from conditioned media
Exosomes were isolated by differential centrifugation of conditioned media collected from TSGH-8301, T24 and J82 cells. Cells were grown in medium containing 10% exosome-depleted FBS (SBI System Biosciences, Palo Alto, CA, USA). After removing cells and other debris by centrifugation at 3000 × g for 30 minutes, the supernatant was subsequently centrifuged at 10000 × g for 1 hour to remove shedding vesicles and other large vesicles. Finally, the supernatant was recentrifuged at 120000 × g for 3 hours at 4 °C. The exosome pellets were resuspended in PBS and stored at 4 °C before experimental analyses.

Nanoparticle tracking analysis
The number and size of exosomes were directly tracked using the NanoSight NS 300 system (NanoSight Technology, Malvern, UK). Exosomes were resuspended in PBS at a concentration of 5 µg/ml and further diluted 100-fold to achieve a concentration between 20 and 100 objects per frame. Samples were manually injected into the sample chamber at ambient temperature. Each sample was detected in triplicate with a 488-nm laser and a high-sensitivity scientific complementary metal-oxide semiconductor camera at a camera setting of 13 with an acquisition time of 60 seconds and a detection threshold setting of 7. The detection threshold was similar in all the samples and was applied using NTA 3.0 analytical software.

Transmission electron microscopy
For conventional transmission electron microscopy, the exosome pellet was placed in a droplet of mixed buffer (1:1 of 2.5% glutaraldehyde (in 0.1 M sodium cacodylate, pH 7.4) and 4% paraformaldehyde (in 1 × PBS)) and fixed overnight at 4 °C. Samples were rinsed in PBS buffer (3 times, 10 minutes each) and further fixed in 1% osmium tetroxide (in double distilled water) for 50 minutes at room temperature. The samples were then embedded in 10% gelatin, fixed in glutaraldehyde at 4 °C, and cut into tiny blocks (< 1 mm 3 ). The samples were dehydrated with an alcohol gradient (70%, 90%, 95%, and 100%) for 10 minutes at each step. Pure alcohol was then exchanged with propylene oxide, and specimens were embedded in increasing concentrations (25%, 50%, 75% and 100%) of Quetol-812 epoxy resin mixed with propylene oxide for a minimum of 2 hours per step. Samples were embedded in pure, fresh Quetol-812 epoxy resin and polymerized at 70 °C for 24 hours. Ultrathin sections (300 nm) were cut using a Leica UC6 ultramicrotome. After staining with uranyl acetate for 10 minutes and lead citrate for 5 minutes at room temperature, exosome morphology was observed by transmission electron microscopy (Hitachi, HT7700, Japan) under operation at 120 kV and adjusted to the appropriate zoom.
TSGH-8301 cells were pretreated with Hoechst DNA counterstain (Sigma-Aldrich, San Jose, CA, USA) for 24 hours. Subsequently, cells were treated with CD9-FITC-labeled exosomes in serum-free medium. Treated cells were photographed for their entirety using the phase-contrast status at low power assessment every 30 minutes, and spontaneously fluorescent images were photographed with the Hoechst DNA channel (360 nm/460 nm excitation/emission) and FITC channel (400 nm/455 nm excitation/emission) using a Lionheart FX Automated Microscope. The dynamic uptake of exosomes was recorded using a BioTek Lionheart FX microscope (BioTek, Winooski, VT, USA). All experiments were performed in triplicate.

Cell viability assay
The viability of the bladder cancer cells was determined using the 3-[4, 5-dimethylthiazol-2-yl]-2, 5diphenyl-tetrazolium bromide (MTT, Sigma-Aldrich) assay. Cells were seeded in 96-well plates at 5000 cells/well and cultured overnight. After treatment with the indicated concentrations of conditioned media or exosomes for 48 hours, cells were incubated with 0.1 mg/ml MTT for 3 hours, and formazan was dissolved in dimethyl sulfoxide (Sigma-Aldrich) at room temperature for 10 minutes. The absorbance at 560 nm was measured with a spectrophotometer (Bio-Rad Inc, Hercules, CA, USA). All experiments were performed in triplicate.

Cell migration and invasion assay
In wound healing assays, 1 × 10 5 cells were seeded in 6-well plates and incubated to 90% confluence before transfection. After treatment with the indicated concentrations of conditioned media or exosomes for 24 hours, cells were scraped with a sterile 200 µl pipette tip to generate a clear line in the wells at time 0. The migrated cells were observed with a phase-contrast microscope every 8 hours (Leica DMI4000B, Bucks, UK), and the wound width at the designated time was measured with ImageJ software. All experiments were performed in triplicate.
Transwell migration assays were performed using 8 µm Transwell chambers (Corning, Steuben County, NY, USA) with 1 × 10 4 cells for each exosome treatment. Transwell invasion assays were evaluated with the same chambers-coated with 1 mg/ml Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) with 2 × 10 4 cells for each exosome treatment. The migration and invasion chambers were incubated in a humidified 5% CO 2 incubator at 37 °C for 24 hours. Cells were then fixed with 4% paraformaldehyde, and the inner surface of the upper chambers was wiped with cotton swabs to remove unmigrated or uninvaded cells. After washing, the chambers were stained with crystal violet (Sigma-Aldrich) for 15 minutes, and the Transwell membranes were torn and kept in slides. Five random fields of each treatment were photographed at 100 × magnification, and the crystal violetstained area was calculated using ImageJ software. Each condition was plated in triplicate.

Colony formation assays
Exosome effects on clonogenicity were evaluated with a colony-formation assay. In brief, 0.5 ml of 0.5% agarose in complete medium was used as the bottom agar in a 12-well plate, and 2 × 10 4 cells were mixed with 0.3% agarose in complete medium at 48 hours after treatment with the indicated concentrations of exosomes. Cells were maintained in a humidified 5% CO 2 incubator at 37 °C for 21 days with fresh medium replacement every three days. Cell colonies were stained with crystal violet (Sigma-Aldrich) for 1 minute and destained with tap water for 15 minutes. Colonies were counted using ImageJ software for each well, and triplicate repeats were performed for each condition.

Western blotting
Total protein was extracted from cultured cells and exosomes using RIPA lysis and extraction buffer

RT-qPCR assay
Total RNA was extracted from cultured cells and exosomes using TOOLSmart RNA Extractor (BIOTOOLS, Taiwan (R.O.C.)). The concentration of total RNA was evaluated using a NanoDrop spectrophotometer (Thermo Fisher Scientific). Total RNA was reverse transcribed into cDNA using the ToolsQuant II Fast RT Kit with Oligo (dT) primer (BIOTOOLS) in a 20 µL reaction system consisting of 1000 ng template, 2 µL of 10 × RT Reaction Premix with Oligo (dT) primer, 1.5 µL of ToolsQuant II Fast RT and RNase-free ddH 2 O. The mixture was centrifuged briefly and incubated with reverse transcription at 42 °C for 15 minutes, followed by enzyme inactivation at 85 °C for 5 minutes. ConMed revealed a slower growth rate than that of the PBS control, which implied that each ConMed contained debris and metabolites and decreased the cell growth rate. However, each ConMed markedly increased the growth rate in a dose-dependent manner (Fig. 1A). Moreover, compared to the PBS control, T24-ConMed also increased the migratory ability in the wound healing assay ( The expression of epithelial-mesenchymal transition-associated molecules was promoted after exosome treatment Since T24-Exos and J82-Exos exerted higher inductive effects on cell motility than cell viability, which indicated that T24-Exos and J82-Exos participated in some signaling pathways that obviously induced cell migration and moderately increased cell proliferation. Therefore, the EMT process seemed to be a potential candidate for further analyses. TSGH-8301 cells were separately treated with PBS control, TSGH-8301-Exos (40 µg/ml), T24-Exos (40 µg/ml), and J82-Exos (40 µg/ml) for 3 days, and treated cells were harvested for western blotting. In Fig. 4, EMT transcription factors (EMT-TF) were first compared, and Snail, Slug and Zeb2 were significantly upregulated after T24-Exos or J82-Exos treatment, while Twist was significantly upregulated after T24-Exos treatment compared to PBS control treatment. Subsequently, markers of epithelial-like or mesenchymal-like properties were compared, and the epithelial marker E-cadherin was significantly reduced after T24-Exos or J82-Exos treatment, although it was also slightly reduced after TSGH-8301-Exos treatment. Conversely, the mesenchymal marker N-cadherin and the interstitial indicator vimentin were increased after T24-Exos or J82-Exos treatment. Moreover, matrix metallopeptidases were analyzed to infer the invasive potential, and MMP2 and MMP9 were significantly increased after T24-Exos or J82-Exos treatment. In short, T24-Exos or J82-Exos induced EMT in recipient TSGH-8301 cells by upregulating most EMT-TFs and resulted in a decrease in E-cadherin but an increase in N-cadherin, vimentin, MMP2 and MMP9.
These results implicated that there were some broad-spectrum regulators in the exosomes derived from high-grade bladder cancer cells, which led to comprehensive induction of the EMT process in recipient TSGH-8301 cells.

Screening and validation of exosomal lncRNAs
LncRNAs comprise a considerable proportion of exosomal components, and several lncRNAs have been reported to contribute to multiple oncogenic steps, including tumor formation, progression, and/or metastatic processes in many cancer types; hence, this study was designed to evaluate the potential of exosome-transmitted lncRNAs to induce aggressiveness by promoting the EMT process.
Moreover, three major upstream signaling cascades of the EMT process described in Fig. 4 were also analyzed. In Fig. 8, T24 siLINC00960-Exos or T24 siLINC02470-Exos significantly reduced β-catenin and TCF4 expression (β-catenin signaling), Notch1, Notch4 and HES1 expression (Notch signaling), and Smad2/3 expression and their phosphorylation/activation (Smad2/3 signaling) when compared to that in the T24-Exos-treated group. In summary, either T24 siLINC00960-Exos or T24 siLINC02470-Exos attenuated the T24-Exos-induced EMT process and its related regulatory mechanism in recipient TSGH-8301 cells. These results suggest that both LINC00960 and LINC02470 play comprehensive roles in EMT regulation and thereby promote the aggressiveness of recipient TSGH-8301 cells.

Discussion
Bladder cancer is usually multifocal with high recurrence and metastasis, and the EMT process has been comprehensively involved in bladder cancer progression [24,25]. Nonmuscle invasive bladder cancer (NMIBC) accounts for more than 70% of newly diagnosed bladder cancers, and more than half of NMIBCs ultimately develop recurrence; approximately 15% of NMIBCs progress to muscle-invasive and/or metastatic diseases, which are highly connected with the EMT process [26,27]. The multifocality of bladder cancer has been hypothesized to originate from intraluminal monoclonal expansion (clonal) [28,29] or spontaneous transformation of multiple cells by virulent environmental agents (field effect) [30]. It has been characterized by early genetic instability or aberrantly epigenetic methylation [31][32][33]. Our findings implicated that exosomes function as potential environmental agents as the "field effect" to promote bladder cancer progression, which might lead to higher severity of multifocal tumors or more malignant behaviors of NMIBC cells.
The key regulators of the EMT process are E-cadherin, N-cadherin, vimentin, Snail, Slug, Twist, and Zeb-2, and most of them have been correlated with bladder cancer progression via either genetic or epigenetic regulation [34]. Currently, little is known about the association between the epigenetic regulation of exosome-transmitted lncRNAs in the EMT process and bladder cancer progression. In this study, we found that exosomes derived from high-grade bladder cancer cells indeed promoted the viability, motility, and clonogenicity of low-grade bladder cancer cells via enhancement of the EMT process. Furthermore, we identified two novel exosomal lncRNAs, LINC00960 and LINC02470, that both play pivotal roles in the EMT process and promote the aggressiveness of bladder cancer cells (Fig. 9A).
Exosomes are key elements that can facilitate intercellular communication and modulate tumor cells by influencing major cellular processes such as apoptosis, cell differentiation, angiogenesis and metastasis [35]. Bladder cancer cells have been reported to undergo EMT transformation after exposure to muscle-invasive bladder cancer exosomes [36]. Exosomes have also been reported to enhance cancer progression and recurrence in hepatocellular carcinoma via the MAPK/ERK signaling pathway [37]. Our results showed that low-grade bladder cancer cells (TSGH-8301) treated with highgrade bladder cancer cell-derived exosomes (T24-Exos or J82-Exos) had an increase in mesenchymal proteins (N-cadherin and vimentin) and a decrease in E-cadherin. T24-Exos or J82-Exos also significantly promoted the migratory and invasive abilities of recipient TSGH-8301 cells. The morphology of the recipient cells gradually changed into spindle-shaped mesenchymal-like cells after T24-Exos or J82-Exos treatment (Supplementary Fig. 1). Comparatively, TSGH-8301-Exos treatment did not change their behaviors and morphology in an autocrine manner. This suggests that exosomes derived from cells of different aggressiveness lead to different levels of aggressiveness-promoting effects in recipient bladder cancer cells.
The three most representative EMT upstream pathways, β-catenin/TCF signaling, Notch signaling and Smad2/3 signaling, were assessed to determine their involvement in exosome-induced EMT. Aberrant activation of β-catenin/TCF signaling is involved in a number of tumors, most notably colorectal carcinomas. TCF4 has been reported to transactivate Snail, Slug and Zeb1 and promote the EMT process [38,39]. Notch1 signaling increases the DNA binding ability of NF-κB and thereby induces the expression of MMP9, which remodels the extracellular matrix and facilitates the extravasation of several cancer cells. Notch also stabilizes cytoplasmic β-catenin and activates other pathways, such as ERK and NF-κB, which induce the expression of Snail, Slug and LEF-1 transcription factors [40][41][42].
Smad complexes bind to regulatory elements and induce the transcription of key genes associated with EMT. Expression of activated Smad2 promotes mesenchymal spindle tumor cell invasion, which also regulates the expression of Snail, Slug and Twist to suppress the expression of E-cadherin [43][44][45]. Our results showed that β-catenin expression can be induced by treatment with each type of exosome, but its downstream transcription factors, especially TCF4, were significantly induced by T24-Exo treatment alone. The expression levels of Notch1, Notch4, Smad2 or Smad3 were all significantly upregulated by T24-Exos or J82-Exos treatment. T24-Exos or J82-Exos also induced the expression of the EMT-transcription factors Slug, Snail, Twist and Zeb2, which ultimately suppressed E-cadherin expression but promoted N-cadherin, vimentin, MMP2 and MMP9 expression in bladder cancer cells.
Several lncRNAs are involved in the EMT process and the degree of malignancy of tumors [46]. The main advantage of lncRNAs that make them suitable as cancer diagnostic and prognostic biomarkers is their high stability while circulating in body fluids, especially when they are encapsulated in exosomes or apoptotic bodies [47]. Several lncRNAs are associated with bladder cancer initiation and progression. LINC00958 has a regulatory role in bladder cancer progression, as its knockdown decreases cell viability, migration, and invasion [48]. LINC00355 is upregulated in bladder cancer samples and contributes to apoptosis inhibition, cell proliferation, and migration [49]. The urothelial carcinoma-associated (UCA1) lncRNA is also involved in cell proliferation, migration, invasiveness, and drug resistance of bladder cancer cells [50]. LINC00152 is highly expressed in bladder cancer and confers its carcinogenic effects by activating the Wnt/β-catenin signaling pathway [51]. Cooperatively, exosomes and their contained lncRNAs are regarded as components of cell signal transmission that modulate endogenous cellular microenvironments because exosomes are able to relocate functional lncRNAs between cells [52]. In this study, we identified two novel lncRNAs, LINC00960 and LINC02470, that were highly expressed in high-grade bladder cancer cells and their cell-derived exosomes. To our knowledge, this is the first study to address the function of LINC00960 and LINC02470, not only in intracellular regulation of malignancy traits but also in intercellular communication, illustrating their driving force in influencing the behaviors of bladder cancer cells. Our data demonstrated that siLINC00960-Exos or siLINC02470-Exos derived from T24 cells significantly inhibited proliferation, migration, invasion, and colony formation in recipient cells compared with their parental T24 exosomes. In addition, T24 siLINC00960-Exos treatment significantly reduced the migratory capability of recipient cells compared to that with siLINC02470-Exos treatment, which implied that LINC00960 and LINC02470 regulates the EMT process via different routes. Moreover, we found that T24 siLINC00960-Exos or siLINC02470-Exos treatment obviously reduced β-catenin signaling, Notch signaling and Smad2/3 signaling; notably, the major EMT transcription factors, Snail and Slug, were significantly reduced, which led to the downregulation of N-cadherin and vimentin [53].
In summary, we demonstrated that high-grade bladder cancer cell-derived exosomes promoted the malignant traits of low-grade bladder cancer cells. T24-Exos or J82-Exos-transmitted LINC00960 and LINC02470 induced the aggressiveness of bladder cancer cells via autocrine or paracrine effects. In contrast, T24 siLINC00960-Exos or siLINC02470-Exos attenuated these aggressive-promoting effects and inhibited the EMT process. Thus, T24-Exos or J82-Exos-transmitted LINC00960 and LINC02470 could be important aggressiveness-promoting factors in bladder cancer progression (Fig. 9B). And it also implicated that intercellular epigenetic regulations play critical roles during bladder cancer progression.

Conclusions
Our findings indicate that exosome-transmitted LINC00960 and LINC02470 derived from high-grade bladder cancer cells promote the aggressive behavior of low-grade bladder cancer cells through intercellular communication and promote EMT by promoting β-catenin signaling, Notch signaling and Smad2/3 signaling. Both lincRNAs may serve as potential liquid biomarkers for the prognostic surveillance of bladder cancer patients in the future.

Competing interests
The authors declare no competing interests.         Exosomes derived from LINC00960-knockdown or LINC02470-knockdown T24 cell conferred