In Vitro Cell.Dev.Biol.—Animal https://doi.org/10.1007/s11626-017-0217-3

Laminar shear stress inhibits high glucose-induced migration and invasion in human bladder cancer cells Yu-Hsiang Lee 1,2

&

Chien-Hsuan Yeh 1

Received: 16 August 2017 / Accepted: 17 November 2017 / Editor: Tetsuji Okamoto # The Society for In Vitro Biology 2017

Abstract High glucose has been known to play a pathogenic role in the development and progression of bladder cancer in diabetics, whereas the leading cause of death in such patients is mainly attributed to hyperglycemia-enhanced metastasis. In addition to the impact of glucose, cancer cells may be affected by laminar shear stress (LSS) generated from interstitial, blood, and/or lymphatic fluid flows during metastasis. Although the effect of flow-induced mechanical force on cancer pathophysiology has been extensively investigated, very little is understood regarding the cells that are simultaneously stimulated by LSS and hyperglycemia. To address this issue, the influence of LSS on bladder cancer cell motility in a hyperglycemic environment was examined. Based on the results of cell movement and protein expression analyses, we found that both cell migration and invasion were up- and downregulated by 25 mM glucose and 12 dynes/cm2 LSS, respectively. Furthermore, the motility of the cells with simultaneous hyperglycemic and LSS stimulations was significantly reduced compared with that of the cells stimulated by high glucose alone (P < 0.05), demonstrating that the LSS rather than hyperglycemia played the dominant role in regulation of cell motility. These results implied that LSS with an intensity ≥ 12 dynes/cm2 may serve as a feasible tool to reduce bladder cancer motility in diabetics.

* Yu-Hsiang Lee [email protected]

1

Department of Biomedical Sciences and Engineering, National Central University, No. 300, Jhongda Rd, 32001 Taoyuan City, Taiwan, Republic of China

2

Department of Chemical and Materials Engineering, National Central University, Taoyuan City, Taiwan, Republic of China

Keywords Glucose . Laminar shear stress . Bladder cancer . Migration . Invasion

Introduction Diabetes mellitus has been recognized as an increasingly prevalent metabolic disease detected not only in the total population but also in young adults worldwide (NCD Risk Factor Collaboration 2016). In addition to the known cardiovascular complications due to hyperglycemia, epidemiologic evidences suggest that people with diabetes (predominantly type 2) are at a significantly higher risk for the development of solid organ malignancies such as liver (Wang et al. 2012a), pancreatic (Ben et al. 2011), and endometrial (Zhang et al. 2013) cancers that seriously threaten human health. The association between diabetes and cancer incidence could be interpreted, because many risk factors are shared by the two diseases such as aging, obesity, physical inactivity, and/or some medications used to treat hyperglycemia including insulin, the insulin analog glargine, and thiazolidinedione as reported in the previous studies (Hemkens et al. 2009; Turner et al. 2014). Bladder cancer is also highly associated with diabetes (Fang et al. 2013). According to statistics from the World Health Organization, bladder cancer ranks as the ninth most frequently diagnosed cancer with a higher incidence rate in men and is the 13th leading cause of cancer death worldwide (Antoni et al. 2017). The occurrence of bladder cancer appears to be multifactorial; both exogenous environmental factors such as cigarette smoking and endogenous molecular factors play a crucial role in cancer initiation and/or development (Chu et al. 2013; McBeth et al. 2015). Moreover, it has been demonstrated that diabetes mellitus (predominantly type 2) and/or certain diabetes

LEE AND YEH

treatments (e.g., thiazolidinedione) may increase the incidence as well as mortality of bladder cancer (Zhu et al. 2013). Such diabetes-induced neoplasticity is potentially due to hyperinsulinemia, hyperglycemia, and/or chronic inflammation (Giovannucci et al. 2010); however, the plausible biological links between the two diseases remain unclear. Metastasis, which involves cellular migration and invasion, is one of the largest obstacles for anticancer treatment and is the main cause of bladder cancer death in the clinic. Among the various pathogenic factors of diabetes, hyperglycemia may promote carcinogenesis by activating signaling pathways related to insulin and insulin-like growth factor 1 (IGF-1) receptors (Ryu et al. 2014), which may enhance multiple cancer phenotypes including proliferation, antiapoptosis activity, invasion, and metastasis. A recent study further revealed that hyperglycemia may promote cancerous metastasis by inducing the epithelial-mesenchymal transition (EMT) via enhanced oxidative stress in the surrounding tissues (Iwatsuki et al. 2010). These observations imply that reduced blood glucose levels would be desirable for efforts to retard/stop cancer development and/or progression. However, increasing evidence has shown that many glucose-lowering therapies and/or medicines such as thiazolidinedione (Turner et al. 2014), pioglitazone (Aschenbrenner 2017), and ciglitazone (Plissonnier et al. 2011) may conversely increase the risk of bladder cancer. Thus, the development of new strategies to reduce cancer cell motility in the presence of high-glucose is certainly needed for bladder cancer therapy in diabetics. The cellular mechanical microenvironment has been known to play a critical role in cellular physiology and/ or function (Nerem 2006). As reported in a previous study, cancer cells may experience shear stress induced by interstitial flow and/or blood/lymphatic flow within tumor tissues, and such mechanical forces may modulate cancer cell growth and development (Swartz and Lund 2012). Although the effectiveness of fluid shear stress on cancer pathophysiology has been widely investigated (Chang et al. 2008; Qazi et al. 2011), very little information is available regarding cancer metastasis when the cells are simultaneously stimulated with mechanical force and hyperglycemia. In this study, we aimed to investigate whether/how laminar shear stress (LSS) influenced bladder cancer cell migration and invasion in hyperglycemic environments to understand the interactions between diabetic bladder cancer metastasis and mechanical stimulation in vitro. We first examined bladder cancer cell motility under various glucose and/or LSS treatments, followed by the detection of the protein expression levels of migration- and invasion-related genes to identify a possible mechanism.

Materials and Methods Cell culture and reagents The bladder cancer cells used in this study were human bladder female transitional carcinoma cells (BFTC-905 cells) purchased from the Bioresource Collection and Research Center (Hsinchu, Taiwan ROC) and were maintained by using RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 1.5 g/L sodium bicarbonate, 15% fetal bovine serum (FBS), and 1% penicillin/ streptomycin at 37°C with 5% CO2. In this study, the glucose medium was prepared using D-(+) glucose (Sigma, St. Louis, MO) at concentrations of 5 and 25 mM to imitate physiologically relevant glucose concentrations in normal and diabetic subjects, respectively (Hien et al. 2016). All of the reagents were used as received.

LSS setup Slides with cultured cells were mounted in a parallel-plate flow chamber as described in our previous study (Lee et al. 2016) where the BFTC-905 cells were continuously scoured with serum-free medium (SFM; RPMI-1640 + 1% penicillin/streptomycin) with and without glucose throughout the time course. In this study, LSS (τ) with an intensity of 12 dynes/cm2 was used in all shearing experiments to simulate the level of LSS at the tumor periphery (Jain et al. 2007). This level of LSS was obtained by setting the appropriate volumetric flow rate of the medium (Q; mL/min) according to the Navier–Stokes equation: τ¼

6μQ bh2

ð1Þ

where μ is the viscosity of the SFM with and without glucose (~ 0.01 dynes-s/cm2) and b and h represent the width and height, respectively, of the space above the cell monolayer. Cells in the shearing experiments were stimulated by LSS and glucose (5 or 25 mM) simultaneously.

Cellular viability and growth after stimulation with glucose and/or LSS To assess the effect of LSS on cells in the presence or absence of glucose, a total of 1.2 × 107 BFTC-905 cells were aliquoted onto six sterilized glass slides 24 h prior to the experiment. After washed twice with PBS, the cells on two slides were separately treated with 5 and 25 mM glucose for 4 h, whereas the cells on the other three slides were treated with 12 dynes/cm2 LSS using glucose-free, 5 mM glucose-, or 25 mM glucose-containing SFM for 4 h. The slide with neither glucose nor LSS was employed as the control. The cell viability of each group was examined at 0 (immediately) and 24 h post stimulation, whereas the cell growth was monitored continuously for 6 d using hemocytometry with trypan blue exclusion method.

SHEAR STRESS INHIBITS GLUCOSE-INDUCED MOTILITY

Apoptosis assay Cellular nuclear counterstaining was performed using Hoechst 33258 (Sigma, 1:1000) as reported elsewhere (Wu et al. 2007), and the images were obtained using fluorescence microscopy. Furthermore, cell apoptosis was quantitatively measured by the DNA fragmentation assay using the Cell Death Detection ELISAPLUS Kit (Roche, Indianapolis, IN) according to the manufacturer’s instructions. The absorbance of anti-DNA-peroxidase (POD) molecules associated with cytoplasmic histone-associated DNA fragments induced from the apoptotic cells was detected by spectrofluorometry at a wavelength of 405 nm. The absorbance values were analyzed after normalization to the background signal. Cell migration and invasion assay 2.4 × 107 BFTC-905 cells were aliquoted onto 12 sterilized glass slides 24 h prior to the experiment. After washed twice with PBS, the cells were treated with 0 (static) or 12 dynes/cm2 LSS using glucose-free, 5 mM glucose-, or 25 mM glucosecontaining SFM for 1 or 4 h as indicated in Table 1. Afterward, 1 × 105 cells in 200 μL of SFM for each group were placed into a transwell insert (Corning, NY), and FBS was used as the chemoattractant to induce cell migration according to the manufacturer’s instructions. After incubation at 37°C for 24 h, the cells that migrated through the membrane were stained with Giemsa or calcein-AM and were photographed using optical and fluorescent microscopy, respectively. In addition, the levels of fluorescence expressed by the cells were detected using a spectrofluorometer with excitation and emission wavelengths of 485 and 520 nm, respectively, and were quantitatively represented by relative fluorescence units (RFUs). A standard curve of cell number (N) vs. RFUs was established prior to the experiment (N = 23.183 × RFUs + 2.206; R2 = 0.9975). In Table 1. Experimental design for the cell migration and invasion assays

the cell invasion assay, the membranes of the transwell inserts were coated with basement membrane extract (BME) before use, and the incubation time for cell movement was set as 48 h. The cell motility in each group was quantitatively defined by the RM value, which is the ratio of the migrated cell number to the original cell number placed into the transwell insert (i.e., 1 × 105). Western blot Protein expression levels of protein kinase B (PKB or AKT; Cell signaling, Danvers, MA), phospho-AKT (Ser473) (p-AKT; Cell Signaling), caveolin-1 (CAV-1; Cell Signaling), phospho-CAV-1 (Tyr14) (p-CAV-1; Cell Signaling), membrane type-1 matrix metalloproteinase (MT1-MMP; Cell Signaling), and glyceraldehyde 3phosphate dehydrogenase (GAPDH; Cell Signaling) were detected using Western blot as described in the previous study (Lee et al. 2016). Briefly, 30 μg of whole protein extract from each sample was resolved by sodium dodecyl sulfate polyacrylamide gels (SDS-PAGE) and then transferred to 0.45-μm polyvinylidene fluoride membranes. The membrane was then treated with blocking buffer (5% w/v skim dry milk in PBS-Tween-20 (0.05% v/v; PBS-T)) for 90 min at room temperature. The primary antibodies directed against the aforementioned proteins were first diluted 1:1000 in blocking buffer and were then incubated with the membranes at 4°C for 12 h. Next, the membranes were washed three times with PBS-T and incubated with HRP-conjugated goat anti-mouse/ rabbit IgG secondary antibodies for 90 min at room temperature. After washed with PBS-T, the membranes were treated with chemiluminescent substrates, and the level of each molecular band was analyzed using the ChemiDoc-IT 810 imaging system (UVP, Upland, CA). In this study, variations in the expression levels of AKT, CAV-1, and MT1-MMP in each

Group

Treatment*

Remark

1 2 3 4 5 6 7

Incubation in glucose-free SFM 1-h incubation in 5 mM glucose SFM 1-h incubation in 25 mM glucose SFM 4-h incubation in 5 mM glucose SFM 4-h incubation in 25 mM glucose SFM 1-h LSS† using glucose-free SFM 1-h LSS† using 5 mM glucose SFM 1-h LSS† using 25 mM glucose SFM 4-h LSS† using glucose-free SFM 4-h LSS† using 5 mM glucose SFM 4-h LSS† using 25 mM glucose SFM 72 h recovery‡ after 4 h LSS using glucose-free SFM

Blank control Static setting

8 9 10 11 12 *

Various treatments were conducted before the transwell-mediated migration/invasion assay

†

LSS was performed with an intensity of 12 dynes/cm2 in all of the shearing experiments

‡

Dynamic setting

Cells were maintained in a 37°C incubator for 72 h; they were cultivated with normal growth medium in the first 48 h and then with SFM in the last 24 h

LEE AND YEH

Cytotoxicity of LSS to BFTC-905 cells The toxicity of LSS on BFTC-905 cells in the presence or absence of glucose was first evaluated through morphological observation and examinations of cellular viability and proliferation after treatment. As shown in Fig. 1-I, the cells were able to maintain normal morphology after 4 h of exposure to ≤ 25 mM glucose under static conditions (Fig. 1-I; B and C), whereas the cells were distorted after exposure to 12 dynes/cm2 LSS for 4 h no matter

whether glucose was present or not (Fig. 1-I; D–F). Nonetheless, the cells in all groups exhibited > 90% viability within 24 h post treatment (Fig. 1-II) and showed a similar specific growth rate of 0.674 ± 0.025 d−1 (P = NS) within 6 d, which was comparable to the group with neither LSS nor glucose (0.682 ± 0.037 d−1; Fig. 1-III). These results indicated that the synergistic effect of 12 dynes/cm2 LSS and 25 mM glucose is nontoxic to BFTC-905 cells. To assess the effect of morphological change/distortion on cell fate, BFTC-905 cell apoptosis immediately after treatment with LSS in the presence or absence of glucose was examined. Figure 2-I exhibits photomicrographic images of Hoechststained cells under various treatments in which intense blue staining represents apoptotic DNA fragments in the nuclei. The levels of cellular apoptosis decreased and increased after treatment with glucose (Fig. 2-I; A–C) and LSS (Fig. 2-I, A vs. D), respectively. On the other hand, according to the results from detection of the absorbance of cytoplasmic histoneassociated DNA fragments as shown in Fig. 2-II, the cellular apoptotic levels lightly decreased to 7 (P = NS) and 12% (P = NS) after treatment with 5 or 25 mM glucose for 4 h under static conditions. However, the apoptosis rate observed in the shearing group was approximately 1.1-fold (P = NS) higher

Figure 1. Effect of glucose and/or LSS on BFTC-905 cells. (I) Micrographic images of BFTC-905 cells after shearing with 0 (A–C) or 12 (D–F) dynes/cm2 LSS for 4 h using SFM in the presence of 0 (A and D), 5 (B and E), or 25 (C and F) mM glucose. Arrows denote the flow direction. All images were photographed by optical microscopy at ×200 magnification. Scale bar = 50 μm. (II) Viabilities of BFTC-905 cells under various treatments. The cell viabilities of the above six settings

were measured after 0- (immediately) and 24-h incubation at 37°C using hemocytometry with trypan blue exclusion method. Values are the mean ± SD (n = 3). *Glu denotes glucose. (III) Growth kinetic curves of BFTC-905 cells in the above six groups. After treatment with glucose and/or LSS, the growth kinetic curve of each group was established through cell counting every 24 h for 6 d. Values are the mean ± SD (n = 3).

group were quantitatively analyzed using the blot intensity values of p-AKT/AKT, p-CAV-1/CAV-1, and MT1-MMP, respectively, after normalization to GAPDH. Statistical analysis All data were obtained from triplicate experiments and are presented as the mean ± standard deviation (SD). Statistical analyses were conducted using the MedCalc software in which comparisons for one condition between two groups were performed by Student’s t test followed by Dunnett’s post hoc test at a significance level of P < 0.05 throughout the study.

Results and Discussion

SHEAR STRESS INHIBITS GLUCOSE-INDUCED MOTILITY

Figure 2. Effect of glucose and/or LSS on BFTC-905 cell apoptosis. (I) Photomicrographic images of Hoechst 33258-stained BFTC-905 nuclei after shearing with 0 (static; A–C) or 12 (D–F) dynes/cm2 LSS using normal (A and D) or glucose-containing (B, C, E, and F) SFM for 4 h. All images were photographed using an inverted fluorescence microscope at ×40 magnification. Arrows denote the flow direction. Scale bar = 20 μm. (II) Quantitative analysis of the cell apoptosis. Both

static and sheared BFTC-905 cells were treated with medium containing 0, 5, or 25 mM glucose for 4 h. The cellular apoptosis in each group was determined based on the absorbance of cytoplasmic histone-associated DNA fragments, which was detected using a spectrofluorometer set at a wavelength of 405 nm. Each absorbance value was presented after normalization to the background signal. Values are the mean ± SD (n = 3).

than that of the group in static, demonstrating that neither glucose nor LSS alone or in combination was able to induce marked apoptosis as compared with the static culture without glucose (Fig. 2-I; A). These results further confirmed that the synergistic effect of 12 dynes/cm2 LSS and 5 mM/25 mM glucose is nontoxic to BFTC-905 cells.

Fig. 3-IV) higher than that without static recovery, indicating that the detrimental effect of LSS on cell migration was reversible. Furthermore, such mechanical force may be able to inhibit the cell motility induced by hyperglycemia. Our data showed that the migrated cell number of 25 mM glucose-treated BFTC905 cells dramatically decreased if they were simultaneously sheared with 12 dynes/cm2 LSS for 4 h (Fig. 3-I and II, E vs. K), showing a 2-fold reduction of the RM value compared with that cultured with 25 mM glucose alone (P < 0.05). Similar results were observed in cell invasion experiments where the invasive capability of the BFTC-905 cells was upand downregulated by treatment with glucose and LSS, respectively, in a dose- and time-dependent manner. The LSS may have constantly inhibited cellular invasiveness even when the cells were stimulated with both hyperglycemia and LSS simultaneously as shown in Fig. 4-I/-II, and such reduced invasiveness of cells due to LSS was reversible after the cells were statically maintained at 37°C for 3 d as illustrated in Fig. 4-III and IV. Based on the RFU analyses (Fig. 4-IV), the RM for the group pretreated with 25 mM glucose for 4 h was 2- and 1.4fold (P < 0.05 for each) higher than the values obtained from the cells with 0 and 5 mM glucose pre-treatment, respectively, while it conversely showed a 1.4-fold reduction (P < 0.05) compared with the control when the cells were simultaneously sheared with 12 dynes/cm2 LSS throughout the time course.

Effect of glucose and/or LSS on cell migration and invasion Figure 3 exhibits colorimetric (Fig. 3-I) and fluorometric (Fig. 3-II) analyses of cell migration for statically cultured and 12 dynes/cm2 LSS-sheared BFTC-905 cells using 0-, 5-, or 25mM glucose SFM for 1 or 4 h. In the static settings, our data showed that the number of migrated cells increased in a dose(Fig. 3-I and II, A–C; D vs. E) and exposure time- (Fig. 3-I and II, B vs. D; C vs. E) dependent manner and that the RM value (cell migration capability) was significantly enhanced by 1.8fold (P < 0.05) when the cells were treated with the 5-fold higher concentration of glucose (5 vs. 25 mM) for 4 h (Fig. 3-III). In the shearing groups, the number of migrated BFTC905 cells diminished along with the time of LSS treatment (Fig. 3-I and II, A vs. F vs. I), and the RM value dramatically decreased to 35 and 48% (P < 0.05) when the cells were sheared with 12 dynes/cm2 LSS without glucose for 1 and 4 h, respectively (Fig. 3-III). In addition, the RM value for cells incubated for72 h at 37°C after 4-h LSS (Fig. 3-III) was 2-fold (P < 0.05,

LEE AND YEH

Figure 3. Analyses of BFTC-905 cell migration capability after various treatments. (I and II) Micrographic images of Giemsa- (I) and calceinAM-stained (II) BFTC-905 cells treated with (i) neither glucose nor LSS (blank control (A)), (ii) 5 mM (B and D), or 25 mM glucose (C and E) for 1 (B and C) and 4 (D and E) h under static conditions, (iii) 12 dynes/cm2 LSS using glucose-free SFM for 1 (F) and 4 (I) h, or (iv) 12 dynes/cm2 LSS using 5 mM (G and J) or 25 mM glucose (H and K) SFM for 1 (G and H) and 4 (J and K) h before transwell-mediated migration examination. Images in (I) and (II) were photographed at ×200 magnification using optical and fluorescence microscopy, respectively. Scale bars in (I) and (II) represent 200 and 100 μm, respectively. (III) Micrographic images of Giemsa- (A) and calcein-AM-stained (B) BFTC905 cells that were statically maintained in a 37°C incubator for 4 h after

shearing with 12 dynes/cm2 LSS by glucose-free SFM prior to the transwell-mediated migration assay. Images in (A) and (B) were photographed at ×200 magnification using optical and fluorescence microscopy, respectively. Scale bars = 200 and 100 μm in images (A) and (B), respectively. (IV) Quantitative analyses of fluorescence levels of migrated BFTC-905 cells shown in (I–III) using spectrofluorometry with excitation and emission wavelengths of 485 and 520 nm, respectively. Values are the mean ± SD (n = 3). *P < 0.05. †P < 0.05 compared with the blank control. aP < 0.05 compared with the static group incubated with 5 mM glucose for 4 h. bP < 0.05 compared with the static group incubated with 25 mM glucose for 4 h. ‡LSS was performed using 5 or 25 mM glucose SFM as indicated by the color/ pattern of the bar.

Through the cell migration (Fig. 3) and invasion (Fig. 4) analyses, we demonstrated that bladder cancer cell motility can be up- and downregulated by hyperglycemia and LSS (≥ 12 dynes/cm2), respectively, showing a pro- and anti-metastasis effect for each stimulus as reported in the previous studies (Qazi et al. 2011; Duan et al. 2014). However, what is of interest is that the hyperglycemia-enhanced movement can be effectively inhibited when the cells were concurrently treated with LSS, and the resultant RM value was even lower than that obtained from the control (neither glucose nor LSS treatment). Such outcomes implied that LSS may be able to disturb glucose metabolism (e.g., Warburg effect) in the bladder cancer cells and thus play a dominant role in cell motility.

Effect of glucose and/or LSS on protein expression levels of AKT/p-AKT, CAV-1/p-CAV-1, and MT1-MMP Figure 5 shows the Western blot results of protein expression analyses of AKT ± phosphorylation, CAV-1 ± phosphorylation, and MT1-MMP, which are highly associated with tumor motility and progression (Shiomi and Okada 2003; Urra et al. 2012; Wang et al. 2012b) under various conditions. In the cells treated with 25 mM glucose for 4 h, the expression levels of pAKT/AKT and p-CAV-1/CAV-1 were significantly enhanced by 3- (Fig. 5-II; P < 0.05) and 3.2-fold (Fig. 5-III; P < 0.05), respectively, whereas that of MT1-MMPT decreased 30% (Fig. 5-IV; P < 0.05) compared with the expression observed in the control. On the other hand, all three genes were downregulated by LSS and exhibited significant reductions of 30, 32, and 40% (P < 0.05 for each) for p-AKT/AKT, p-CAV-1/

SHEAR STRESS INHIBITS GLUCOSE-INDUCED MOTILITY

Figure 4. Analyses of BFTC-905 cell invasion capability after various treatments. (I and II) Micrographic images of Giemsa- (I) and calceinAM-stained (II) BFTC-905 cells that were treated with (i) neither glucose nor LSS (blank control; A), (ii) 5 (B and D) or 25 mM (C and E) glucose for 1 (B and C) and 4 (D and E) h under static conditions, (iii) 12 dynes/ cm2 LSS using glucose-free SFM for 1 (F) and 4 (I) h, or (iv) 12 dynes/ cm2 LSS using 5 mM (G and J) or 25 mM glucose (H and K) SFM for 1 (G and H) and 4 (J and K) h before transwell-mediated invasion examination. Images in both (I) and (II) were photographed at ×200 magnification using optical and fluorescence microscopy, respectively. Scale bars in (I) and (II) represent 200 and 100 μm, respectively. (III) Micrographic images of Giemsa- (A) and calcein-AM-stained (B) BFTC905 cells that were statically maintained in a 37°C incubator for 4 h after shearing with 12 dynes/cm2 LSS by glucose-free SFM prior to the

transwell-mediated invasion assay. Images in (A) and (B) were photographed at ×200 magnification using optical and fluorescence microscopy, respectively. Scale bars = 200 and 100 μm in images (A) and (B), respectively. (IV) Quantitative analyses of fluorescence levels of migrated BFTC-905 cells shown in (I–III) using spectrofluorometry with excitation and emission wavelengths of 485 and 520 nm, respectively. Values are the mean ± SD (n = 3). *P < 0.05. †P < 0.05 compared with the blank control. aP < 0.05 compared with the static group incubated with 5 mM glucose for 1 h. bP < 0.05 compared with the static group incubated with 25 mM glucose for 1 h. cP < 0.05 compared with the static group incubated with 5 mM glucose for 4 h. d P < 0.05 compared with the static group incubated with 25 mM glucose for 4 h. ‡LSS was performed using 5 or 25 mM glucose SFM as indicated by the color/pattern of the bar.

CAV-1, and MT1-MMP, respectively, after exposure to 12 dynes/cm2 LSS for 4 h in the absence of glucose. In the group stimulated with both 25 mM glucose and LSS, all three genes were dramatically reduced after a 4-h treatment and exhibited 2.2-, 4.1-, and 1.8-fold (P < 0.05 for each) lower expression levels for p-AKT/AKT, p-CAV-1/CAV-1, and MT1-MMP, respectively, compared to the group with 25 mM glucose alone. Based on the analyses of p-AKT/AKT and MT1-MMP, our data showed that high glucose was able to enhance cell migration but seemed to suppress invasiveness, whereas both cell activities can be arrested by LSS. These results were consistent with those of previous studies (McLennan et al. 2000; Qazi et al. 2011; Duan et al. 2014). However, one may note that the group with hyperglycemia exhibited a remarkably

increased invasive cell number compared with the cells without glucose as shown in Fig. 4 (I/II, A vs. E). Such a conflicting outcome could have occurred because MT1-MMP is just one of many factors, but not the only and sufficient factor, to induce cell invasion. To clarify the mechanism of hyperglycemia-enhanced cell invasiveness for BFTC-905 cells, other types of invasion-related proteins instead of MT1-MMP such as collagenase (e.g., MMP-1, MMP-8, and MMP-13), stromelysin (e.g., MMP-3 and MMP-10), matrilysin (e.g., MMP-7 and MMP-26), and gelatinase (e.g., MMP-2 and MMP-9) should be examined, and these efforts are currently in progress. Through the analysis of protein expression of CAV-1, the mechanical stress recipient on the cell membrane which may

LEE AND YEH

Figure 5. Analyses of motility-related protein expression levels ofBFTC-905 cells under various treatments. (I) Western blots of pAKT, AKT, p-CAV-1,CAV-1, MT1-MMP, and GAPDH proteins extracted from the statically cultured and 12 dynes/cm2 LSS-sheared BFTC-905 cells in the presence or absence of glucose for 1 or 4 h. The condition of each lane is indicated beneath the blotting image. The static culture setting with neither glucose nor LSS was employed as the blank control. (II–IV) Quantitative analyses of the protein expression levels of

p-AKT/AKT (II), p-CAV-1/CAV-1 (III), and MT1-MMP (IV). Each bar represents the level of relative optical intensity of the protein after normalization to GAPDH. Values are the mean ± SD (n = 3). *P < 0.05. † P < 0.05 compared with the blank control (solid black). aP < 0.05 compared with the group sheared with 5 mM glucose SFM for 1 h. b P < 0.05 compared with the group sheared with 25 mM glucose SFM for 1 h.

regulate cell migratability through the PI3K/AKT/mTOR pathway (Yang et al. 2016), we found that the level of pCAV-1/CAV-1 in the cells with synergistic impact of 25 mM glucose and 12 dynes/cm2 LSS was promptly reduced within 1 h as expressed by the cells with LSS alone, implicating that the LSS was the dominant regulator (reducer) of BFTC-905 cell migration among the dual stimuli as evidenced in Fig. 3 (I/II, E vs. K). Combined with a decreased level of MT1MMP, the cells with both hyperglycemia and LSS exhibited a tremendously reduced invasiveness as manifested in Fig. 4 (I/II, E vs. K). Taken together, these results indicated that LSS may be able to serve as a feasible tool to reduce bladder cancer cell motility in diabetics.

protective role in arresting cancer metastasis in diabetics. Moreover, since enhanced motility is one of the contributing factors to disease progression in many types of cancer, the efforts presented in this study may provide new insight into cancer treatment.

Conclusion In this study, we demonstrated that the high glucose-induced migration and invasion of bladder cancer cells can be effectively inhibited when the cells were concurrently sheared with 12 dynes/cm2 LSS for ≥ 1 h, indicating that LSS may play a

Funding information This work was financially supported by the Ministry of Science and Technology, R.O.C. (MOST 106-2221-E-008060; Y.-H. Lee).

References Antoni S, Ferlay J, Soerjomataram I, Znaor A, Jemal A, Bray F (2017) Bladder cancer incidence and mortality: a global overview and recent trends. Eur Urol 71:96–108 Aschenbrenner DS (2017) Updated review confirms potential risk of bladder cancer with pioglitazone. Am J Nurs 117:25 Ben Q, Xu M, Ning X, Liu J, Hong S, Huang W, Zhang H, Li Z (2011) Diabetes mellitus and risk of pancreatic cancer: a meta-analysis of cohort studies. Eur J Cancer 47:1928–1937 Chang SF, Chang CA, Lee DY, Lee PL, Yeh YM, Yeh CR, Cheng CK, Chien S, Chiu JJ (2008) Tumor cell cycle arrest induced by shear

SHEAR STRESS INHIBITS GLUCOSE-INDUCED MOTILITY stress: roles of integrins and Smad. Proc Natl Acad Sci U S A 105: 3927–3932 Chu H, Wang M, Zhang Z (2013) Bladder cancer epidemiology and genetic susceptibility. J Biomed Res 27:170–178 Duan W, Shen X, Lei J, Xu Q, Yu Y, Li R, Wu E, Ma Q (2014) Hyperglycemia, a neglected factor during cancer progression. Biomed Res Int 2014:461917 Fang H, Yao B, Yan Y, Xu H, Liu Y, Tang H, Zhou J, Cao L, Wang W, Zhang J, Zhao L, Chen X, Zhang F, Zhao Y (2013) Diabetes mellitus increases the risk of bladder cancer: an updated metaanalysis of observational studies. Diabetes Technol Ther 15:914– 922 Giovannucci E, Harlan DM, Archer MC, Bergenstal RM, Gapstur SM, Habel LA, Pollak M, Regensteiner JG, Yee D (2010) Diabetes and cancer: a consensus report. Diabetes Care 33:1674–1685 Hemkens LG, Grouven U, Bender R, Günster C, Gutschmidt S, Selke GW, Sawicki PT (2009) Risk of malignancies in patients with diabetes treated with human insulin or insulin analogues: a cohort study. Diabetologia 52:1732–1744 HienTT, TurczyńskaKM, DahanD, EkmanM, GrossiM, SjögrenJ, NilssonJ, BraunT, BoettgerT, Garcia-VazE, StenkulaK, Swärd 1, Gomez MF, Albinsson S (2016) Elevated glucose levels promote contractile and cytoskeletal gene expression in vascular smooth muscle via rho/protein kinase C and actin polymerization. J Biol Chem291: 3552–3568 Iwatsuki M, Mimori K, Yokobori T, Ishi H, Beppu T, Nakamori S, Baba H, Mori M (2010) Epithelial-mesenchymal transition in cancer development and its clinical significance. Cancer Sci 101:293–299 Jain RK, Tong RT, Munn LL (2007) Effect of vascular normalization by antiangiogenic therapy on interstitial hypertension, peritumor edema, and lymphatic metastasis: insights from a mathematical model. Cancer Res 67:2729–2735 Lee YH, Lee CC, Huang CH, Ho FM (2016) Laminar shear stress promotes nicotine-induced inflammation and hemostatic expression in human endothelial cells. Cell Mol Bioeng 9:466–477 McBeth L, Grabnar M, Selman S, Hinds TD Jr (2015) Involvement of the androgen and glucocorticoid receptors in bladder cancer. Int J Endocrinol 2015:384860 McLennan SV, Martell SY, Yue DK (2000) High glucose concentration inhibits the expression of membrane type metalloproteinase by mesangial cells: possible role in mesangium accumulation. Diabetologia 43:642–648 NCD Risk Factor Collaboration (NCD-RisC) (2016) Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet 387:1513–1530 Nerem RM (2006) Tissue engineering: the hope, the hype, and the future. Tissue Eng 12:1143–1150

Plissonnier ML, Fauconnet S, Bittard H, Lascombe I (2011) The antidiabetic drug ciglitazone induces high grade bladder cancer cells apoptosis through the up-regulation of TRAIL. PLoS One 6:e28354 Qazi H, Shi ZD, Tarbell JM (2011) Fluid shear stress regulates the invasive potential of glioma cells via modulation of migratory activity and matrix metalloproteinase expression. PLoS One 6:e20348 Ryu TY, Park J, Scherer PE (2014) Hyperglycemia as a risk factor for cancer progression. Diabetes Metab J 38:330–336 Shiomi T, Okada Y (2003) MT1-MMP and MMP-7 in invasion and metastasis of human cancers. Cancer Metastasis Rev 22:145–152 Swartz MA, Lund AW (2012) Lymphatic and interstitial flow in the tumour microenvironment: linking mechanobiology with immunity. Nat Rev Cancer 12:210–219 Turner RM, Kwok CS, Chen-Turner C, Maduakor CA, Singh S, Loke YK (2014) Thiazolidinediones and associated risk of bladder cancer: a systematic review and meta-analysis. Br J Clin Pharmacol 78: 258–273 Urra H, Torres VA, Ortiz RJ, Lobos L, Díaz MI, Díaz N, Härtel S, Leyton L, Quest AF (2012) Caveolin-1-enhancedmotility and focal adhesion turnover require tyrosine-14 but not accumulation to the rear in metastatic cancer cells. PLoS One 7:e33085 Wang C, Wang X, Gong G, Ben Q, Qiu W, Chen Y, Li G, Wang L (2012a) Increased risk of hepatocellular carcinoma in patients with diabetes mellitus: a systematic review and meta-analysis of cohort studies. Int J Cancer 130:1639–1648 Wang Z, Yang J, Fisher T, Xiao H, Jiang Y, Yang C (2012b) Akt activation is responsible for enhanced migratory and invasive behavior of arsenic-transformed human bronchial epithelial cells. Environ Health Perspect 120:92–97 Wu CC, Li YS, Haga JH, Kaunas R, Chiu JJ, Su FC, Usami S, Chien S (2007) Directional shear flow and rho activation prevent the endothelial cell apoptosis induced by micropatterned anisotropic geometry. Proc Natl Acad Sci U S A 104:1254–1259 Yang H, Guan L, Li S, Jiang Y, Xiong N, Li L, Wu C, Zeng H, Liu Y (2016) Mechanosensitive caveolin-1 activation-induced PI3K/Akt/ mTOR signaling pathway promotes breast cancer motility, invadopodia formation and metastasis in vivo. Oncotarget 7: 16227–16247 Zhang ZH, Su PY, Hao JH, Sun YH (2013) The role of preexisting diabetes mellitus on incidence and mortality of endometrial cancer: a meta-analysis of prospective cohort studies. Int J Gynecol Cancer 23:294–303 Zhu Z, Zhang X, Shen Z, Zhong S, Wang X, Lu Y, Xu C (2013) Diabetes mellitus and risk of bladder cancer: a meta-analysis of cohort studies. PLoS One 8:e56662