The increased expression of Piezo 1 and Piezo 2 ion channels in human and mouse bladder carcinoma

Material and methods. The detection of developmental expression was performed on P0–P90 days in bladder tissue of Balb/c strain mice. Mice were divided into bladder cancer (n = 40) and control groups (n = 10). Bladder cancer in mice was created by using N-butyl-N-(4-hydroxybutyl)nitrosamine (BBN). In the study, 60 human subjects were included, whose normal tissues were used as controls. After the histopathological evaluation, the expression of Piezo1/2 genes was examined by reverse transcription polymerase chain reaction (RT-PCR) and immunohistochemistry in tumor and normal tissues.


Introduction
Bladder cancer is the 4 th most common cancer type in men. 1 Nearly 90% of patients with bladder cancer have transitional cell carcinoma of the bladder (TCC).Bladder cancers with a new diagnosis are usually -in 75% of cases -non-muscle invasive. 2According to several studies, there are many etiological factors that contribute to the disease, such as male gender, tobacco use, chemical agents, diet, exposure to cyclophosphamide, ionizing radiation, phenacetin-containing analgesics, and genetic factors. 3,4ome of the channel types localized in the plasma membrane and intracellular organelles are involved in carcinogenesis and contribute to the basic phenotypes, like metabolic reprogramming, migration, unlimited proliferative potential, apoptosis resistance, induction of angiogenesis, and invasion of the cancer cells.][7] Mechanotransduction is the conversion of a mechanical stimulus into chemical/biological activity.This is a basic physiological process of mammalian cells and it affects many critical processes, such as pain, embryonic development and vascular tone regulation. 8Mechanosensitive ion channels (MSCs) are associated with the physiological functions of some biological systems, e.g., the musculoskeletal system, neuromuscular system and/or respiratory system. 91][12] They have no sequence similarity with any other channel. 13Both of these Piezo channels are cation--selective mechanical channels and they have a selectivity sequence of preference for ions Ca 2+ > K + > Na + > Mg 2+ . 14,15iezo1 and Piezo2 are homologous and they were both cloned in 2010. 8,13Piezo proteins form a homotetramer of about 1.2 MDa and have been highlighted as a central pore module. 13,16Expression profiles of messenger RNA (mRNA) and proteins for Piezo1 and Piezo2 have been detected in numerous MSC tissues. 10The expression of Piezo1 has been shown in endothelial cells, red blood cells, cochlear hair cells, kidney, skin, lung, and bladder tissue. 11,17,18Piezo1 also plays a significant physiological role in the hematopoietic and cardiovascular system. 11,12,19,20It has been documented for its responsibility for cell volume regulation of red blood cells.][23] Homozygous mutant Piezo1 mice are embryonically lethal. 11,19However, blood cell-specific Piezo1 knockout mice display increased erythrocyte size and osmotic fragility. 24iezo2 has been shown to be expressed in the dorsal root and trigeminal ganglia, lung, bladder, and Merkel cells. 17he suggestion that Piezo1 and Piezo2 channels regulate mammalian development, physiology and carcinogenesis has yet to be clarified.Therefore, in this study, we aimed to evaluate Piezo1 and Piezo2 ion channel protein expressions in postnatal period (P0-P90) in mice; we also compared the expressions in human and mouse tumor and non-tumor bladder tissues.

Patients
We conducted a prospective controlled study with patients that were admitted with painless gross hematuria and low urinary tract symptoms (LUTS).The study was approved by the Fırat University (Elazığ, Turkey) Local Ethics Committee and written informed consent was obtained from each participant before the study was initiated.Bladder tissue samples were obtained from 60 patients (14 females and 46 males).Transurethral resection was performed in the presence of bladder tumor and a biopsy was taken if the suspicion was determined.Routine pathological diagnoses were performed according to hematoxylin and eosin (H&E) staining.During the pathological analysis, the samples were categorized according to the WHO/International Society of Urological Pathology (ISUP) 1998 classification. 25Patients in whom a pathological examination revealed TCC constituted the patient group and those with a benign pathology examination constituted the control.After a histopathological evaluation for urothelial carcinoma, the expression of Piezo1 and Piezo2 gene was analyzed by quantitative real-time polymerase chain reaction (qRT-PCR).Immunohistochemistry of the tumor and normal tissues of patients was performed to determine whether there was a difference between the expression of Piezo1 and Piezo2 channels in tumor and normal tissues.

Animals
The ethical Animal Research Committee of Fırat University approved the study protocol.Balb/c-type mice were housed in standard plastic cages.Animals were maintained in light rooms for 12 h and in dark rooms for 12 h.All animals were allowed free access to water and food.Bladder tissue samples of 2 mice for each day: P0, P7, P14, P21, P28, P36, and P90 were used in each postnatal period for determining the developmental expression.Mice were arranged as 2 groups: the bladder cancer group (n = 40) and the control group (n = 10).N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN) (0.05% in the drinking water) was applied to induce bladder cancer in mice for 3 months.Carcinoma formation was expected for 4 months following this process.After the detection of pathologic carcinoma by a pathologist with H&E staining in 2 mice 4 months later, all animals were decapitated after 3% pentobarbital abdominal anesthesia.Bladder tissues were stored in liquid nitrogen at -80°C for qRT-PCR and in 10% neutral buffered formalin for immunohistochemical analysis.

RNA extraction
Total RNA was isolated from human and mouse bladder tissue using Trizol (Life Technologies, San Francisco, USA).Bladder tissue was homogenized using a homogenizer (Next Advance, Inc., Averill Park, USA).Total RNA was dissolved in 100 µL of diethyl pyrocarbonate (DEPC)--treated H 2 O, aliquoted and stored at -80°C.A Qubit ® 2.0 Fluorometer (Invitrogen by Life Technologies TM , Zug, Austria) was used to determine RNA concentration and purity.

Real-time polymerase chain reaction
High Capacity cDNA Reverse Transcription Synthesis Kits (Applied Biosystems, Inc., Foster City, USA) were used for the conversion of total RNA to complementary DNA (cDNA), using 1 µg of total RNA.Complementary DNA isolation was performed in a gradient thermal cycler (Biometra, Göttingen, Germany) with a profile of 25°C for 10 min, 37°C for 120 min, 85°C for 5 min, and 4°C for 60 min.
All of the samples were run in batch.As a negative control, samples were run simultaneously with no reverse transcriptase enzyme (RT negative) or RNA (no template controls) to control for RNA and genomic DNA contamination, respectively.ABI Prism 7500 Fast Real Time PCR Instrument (Applied Biosystems) was used for the reverse transcription polymerase chain reaction (RT-PCR) analysis using Piezo1 (Hs00207230-m1 and Mm01241549-m1) and Piezo2 (Hs00401026-m1 and Mm01265861-m1).All the results were standardized according to the levels of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Hs02758991-g1 and Mm99999915-g1).

Immunohistochemistry staining
Human bladder sections (4-5 μm thick) were prepared from paraffin embedded blocks for immunohistochemical staining.Peptide blocking and immunohistochemical studies were performed with the indirect immunoperoxidase staining technique on paraffin-embedded material.

H&E -hematoxylin and eosin; BBN -N-butyl-N-(4-hydroxybutyl)nitrosamine.
Non-specific staining was blocked using a 20 min incubation with a serum-blocking solution (Immuno Cruz Staining System; Santa Cruz Biotechnology, Dallas, USA).The sections were incubated with a Piezo1 (NBP1-78537; Novus Biologicals, Littleton, USA) and Piezo2 (NBP1-78538; Novus Biologicals) antibody diluted 1:200 for 60 min in a humid chamber at room temperature.The sections were then incubated with streptavidin peroxidase (TS-125-HR; Lab Vision Corp., Fremont, USA) for 30 min after they were washed 3 times with phosphate buffered saline (PBS).Drops of 3-amino-9-ethylcarbazole (AEC) substrate + AEC chromogen (AEC Substrate, TA-015-HAS; AEC Chromogen, TA-002-HAC; Lab Vision Corp.) solution were applied, and the tissues were washed with PBS after labeling was checked by light microscopy.Mayer's hematoxylin was used for the counterstaining of sections, rinsed in PBS and distilled water, and then mounted with a closure solution (Large Volume Vision Mount, TA-125-UG; Lab Vision Corp.).After the slides were analyzed by the BX50 optical microscopy (Olympus Corp., Tokyo, Japan), they were photographed.The evaluation of immunohistochemical staining was performed semi-quantitatively, based on the spreading and levels of the staining.The levels for immunostaining in bladder tissue were detected by subjective visual scoring of the brown stain.Then, they were compared with normal tissue.The scoring levels were as follows: weak staining intensity (nearly equal to normal tissue = 0), moderate staining intensity (1) and strong staining intensity (2).

Statistical analysis
Statistical Package for the Social Sciences (SPSS) software v. 22 for Windows (IBM, Armonk, USA) was used for the analysis of data for each experiment.If the data satisfied the suppositions of normality and homogeneity of parametric analyses (one-way independent measure ANOVA and t-tests), variance was performed along with Tukey's honest significant difference (HSD) posthoc test, if it was appropriate.A p-value of <0.05 was accepted as statistically significant.All data was presented as mean ± standard deviation (SD) or standard error of the mean (SEM).

Molecular findings
A total of 60 patients with urothelial carcinoma were included in the analysis for staging.The majority (63.3%) of the urothelial tumors were in the T1 stage.All patients in the study group were smokers.Patient characteristics are shown in Table 1.
In the BBN-treated group, 9 animals died during the procedure.A histopathologic evaluation of all the bladders was used to determine BBN-induced carcinogenesis and revealed benign proliferative changes in 32.2% (10/31) and urothelial carcinoma in 67.7% (21/31) of the animals treated with BBN (Fig. 1).In situ, carcinoma in 66.6% (14/21) samples and adenocarcinoma in 33.3% (7/21) samples of urothelial carcinomas was also determined.As expected, this contrasted significantly with the control group, in which no evidence of bladder cancer was detected (Fig. 1).
We first performed the gene expression analysis by RT-PCR in tumor samples.Piezo1 expression significantly increased on day P14, P21, P28, P36, and P90 in comparison with day P1, whereas Piezo2 expression significantly increased on day P7, P21 and P28, and decreased on day P90 compared to day P1 (Fig. 2A).Piezo1 was detected to be upregulated in 53.3% (32/60), downregulated in 13.33% (8/60) and not changed in 30.7% (20/60) of the human bladder tissue samples.Piezo2 was detected to be upregulated in 60% (36/60), downregulated in 20% (12/60) and not changed in 20% (12/60) of the human bladder tissue samples.The expression levels of Piezo1 and Piezo2 calculated in all the human bladder cancer group tissues was significantly increased compared to normal tissue (t-test; p = 0.00 and p = 0.00, respectively) (Fig. 2B).On the other hand, the expression levels of Piezo1 and Piezo2 in the BBN-induced bladder cancer group was significantly higher than in the normal control group (t-test; p < 0.00 and p = 0.04, respectively) (Fig. 2C).
The detection rate of Piezo1 was 100% in bladder carcinoma specimens.The expression of Piezo1 in bladder carcinoma was significantly associated with tumor size and stage, but it had no association with other clinicopathological features.The expression of Piezo1 in bladder carcinoma patients with ≥3 tumor size was significantly higher than in bladder carcinoma patients with <3 tumor size (p = 0.00).Piezo1 expression was significantly upregulated in ≥pT2 (muscle invasive bladder cancer with pathological findings) as compared to pT0 (no evidence of residual carcinoma in the cystectomy specimens after an initial cancer diagnosis in the biopsy or transurethral resection specimens) and pT1 (high-risk non-muscle invasive bladder cancer with pathological findings) patients (pT0 vs ≥pT2, p = 0.006; pT1 vs ≥pT2, p = 0.000, respectively).Piezo2 expression was increased in ≥pT2 patients as compared to pT0 patients (pT0 vs ≥pT2, p = 0.003).pT0 vs pT1 and pT1 vs ≥pT2 were not significantly different (p > 0.05).A higher expression level of Piezo2 in high-grade tumors was detected (p = 0.003).The data also demonstrated that the expression level of Piezo2 had no significant association with tumor grade.Piezo1 and Piezo2 expression was not significantly different in bladder cancer specimens with lymph node metastasis as compared to specimens without lymph node metastasis; there were also no differences with regard to age and gender (p > 0.05).

Immunohistochemical findings
To examine the localization of Piezo1 and Piezo2 in the human bladder, the immunohistochemical analysis was performed.Immunohistochemistry revealed that Piezo1 immunoreactivity was mainly observed in the urothelial layer with less immunoreactivity in stromal tissues (Fig. 3A,3B).In contrast, little Piezo2 expression was detected in urothelial tissues (Fig. 3C,3D).Piezo1 and Piezo2 were mainly localized in the plasma membrane and the cytoplasm near the nucleus.

Discussion
In the present study, we investigated pathological changes in Piezo1 and Piezo2 in the bladder cancers of humans and mice.We obtained the following novel findings: -mRNA expressions of Piezo1 and Piezo2 genes in cancerous bladder tissues were increased as compared to control bladder tissues; -Piezo1 expression was correlated with tumor stage, grade and size; -Piezo2 expression increased significantly only in ≥pT2 patients as compared to pT0.Piezo1 and Piezo2 proteins take part in mechanosensory processes and they are stretch-activated cation channels.Almost every cell can sense the mechanical signals.
In particular, stretch-activated cation channels in cells have a basic role in sensing these mechanical signals. 8,26,27oste et al. found that the mRNA expression of Piezo1 is the highest in the bladder, skin and lungs of adult mice. 10he loss or absence of Piezo1 plays an important role in increased cell migration and metastases of lung cancer.
A decrease of Piezo1 in gastric tumors has been related to reduced cell migration. 14,17,28The authors also described the extrusion of living cells at colon surfaces.This suggests that the loss of Piezo1 in colon tissue could lead to polyp formation. 17All of these reports point at Piezo proteins as critical for cell migration, elongation and proliferation.In the present study, we observed similar results; mRNA expression was high in bladder cancer tissues, but higher expression levels could lead to cell migration.To support our results, the knockdown of Piezo1 in gastric epithelium cancer cells was shown to reduce cell migration.The authors suggested that Piezo1 overexpression promotes the invasion and metastasis of gastric cancer. 17Increases in the Piezo ion channel expression can cause dysfunction and leakage at these ion channels.This process may also result in proliferative changes in bladder cancer tissue. 29ccording to a recent study, haploinsufficiency of Piezo1 leads to endothelial abnormality, indicating the importance of this gene for vascular structure. 19Angiogenesis is a finely tuned process, which is the result of the equilibrium between pro-and anti-angiogenic factors.In solid tumor angiogenesis, the balance is highly in favor of the production of new, but poorly functioning blood vessels, which are initially intended to provide growing tumors with nutrients and oxygen. 30Piezo1 ion channels may thus affect different hallmarks of cancer, especially tumor angiogenesis.
Yang et al. reported that Piezo2 knockdown decreased angiogenesis and vascular hyperpermeability, and inhibited endothelial cell proliferation, migration and tube formation.Piezo2 knockdown in endothelial cells resulted in the suppression of cell proliferation, migration and invasion of glioma tumor cells, which regulate glioma angiogenesis via Ca 2+ /Wnt11/β-catenin signaling. 26These studies provide evidence for Piezo2 as a critical regulator of tumor angiogenesis and vascular permeability.The present study showed that Piezo2 expression was increased in bladder tumor samples compared to human and mouse control tissues.Inspired by these findings, we speculated that Piezo2 may contribute to the carcinogenesis of bladder cancer by regulating tumor angiogenesis.However, the function of Piezo2 in bladder tumor angiogenesis has not been elucidated.It is not known whether the changes in Piezo channel expression in bladder cancer tissue occur due to the basic steps of cancer progression or secondary to other variations.For this reason, we aimed to investigate postnatal expressions of Piezo channels in this study.Expression variation on certain days supports the notion that these channels may play a role in bladder development.

Conclusions
According to our study, Piezo proteins are thought to be involved in cancer and they regulate mechanical transmission.Increased expression of Piezo1 and Piezo2 in bladder cancer indicates that these channels may be therapeutic targets.The use of blockers for these channels has been determined to inhibit the cell proliferation.Further research will reveal whether Piezo1 and Piezo2 are causally involved in the carcinogenesis and progression of bladder cancer, which represents a potential therapeutic target or a prognostic factor.

Fig. 1 .
Fig. 1.Representative H&E-stained images of urothelial cross-sections from BBN-induced bladder cancer in mice Note the urothelial cancers in the BBN-induced mice: A -note the normal-appearing urothelium in the control group of mice; B -note the high-grade carcinoma-in-situ lesion in BBN-treated mice; C -note the high-grade, papillary bladder tumor in BBN-treated mice; magnification: ×400; D -fresh tumor samples with adenocarcinomas in mice.

Fig. 2 .
Fig. 2. Quantitative RT-PCR analysis of Piezo1 and Piezo2 in bladder tissue in mice and humans A -results are expressed as mean ±SD for 14 mice (2 mice for each day) on day P0, P7, P14, P21, P28, P36, and P90 for determining the developmental expression; Bresults are expressed as mean ±SD for 10 mice from the control group and 31 BBN-treated animals; C -results are expressed as mean ±SD for cancerous and healthy tissue of 60 patients; error bars represent SD significantly different (p < 0.05) from control or P0 ; * shows the statistical significance.RT-PCR -real-time polymerase chain reaction; BBN -N-butyl-N-(4-hydroxybutyl)nitrosamine.

Fig. 3 .
Fig. 3. Immunohistochemical staining for Piezo1 and Piezo2 in human bladder tissue Representative bladder sections stained for Piezo1 (A -control; B -cancer) and Piezo2 (C -control; D -cancer) in carcinomas and control tissue in human bladder samples.Piezo1 and Piezo2 staining was detected especially in the cytoplasm and membrane urothelial cells.Note the marked upregulation of both Piezo1 and Piezo2 in human bladder cancer.Scale bar: 20 µm; magnification ×200.

Table 1 .
Distribution of variables in bladder cancer patients