This study was approval by the Wake Forest University Health Sciences Institutional Review Board. A total of 50 urine samples were collected from 10 healthy male individuals ranging from 25 to 57 years of age and divided into four age groups (20, 30, 40, 50 years of age). A total of 117 USCs clones were isolated, expanded, and characterized as previously described (Zhang et al., 2008; Bodin et al., 2010; Wu et al., 2011). Briefly, USCs were grown in culture media composed of keratinocyte serum-free medium (KSFM) and embryonic fibroblast medium (EFM) mixed at a ratio of 1:1 (Zhang et al., 2008). Only wells in 24 plates that contained single cell were scored and used for further experimentation.

Telomerase activity levels were measured with the Telo TAGGG Telomerase PCR ELISA plus kit (Roche Applied Science, Mannheim, Germany), according to the manufacturer’s recommendations. HEK 293 cells were used as a positive control, and human BMSC and smooth muscle cells (SMC) were used as negative controls. Briefly, 2 × 10⁵ cells at passage two were collected after being trypsinized and washed with cold PBS. Telomerase added the telomeric repeats (TTAGGG) in the kit to the 3’ end of the biotin-labeled synthetic P1-TS primer. These elongated products, and the internal standard (IS) included in the same reaction vessel, were amplified by PCR using the primers P1-TS and the anchor-primer P2. The PCR products were divided into two aliquots, denatured, and hybridized separately to digoxigenin (DIG) labeled detection probes specific for the telomeric repeats and for IS (P3-Std). The resulting products were immobilized via the biotin label onto a streptavidin-coated microplate. Immobilized amplicons were detected with an antibody against digoxigenin that is conjugated to horseradish peroxidase (Anti-DIG-HRP) and the sensitive peroxidase substrate TMB. Absorbance values were measured as the A_450nm reading against a blank (reference wavelength A_690nm) by using a spectrophotometer. Relative telomerase activity (RTA) within different samples in an experiment were obtained using the following formula (Kim et al., 1994; Kim and Wu, 1997):RTA = [(AS–AS, O)/AS, IS]/[(ATS, 8–ATS8, 0)/ATS8, IS]] × 100.

AS: absorbance of sample; AS, 0: absorbance of heat-treated sample; AS, IS: absorbance of Internal standard (IS) of the sample; ATS8: absorbance of control template (TS8); ATS8, 0: absorbance of lysis buffer; ATS8, IS: absorbance of Internal standard (IS) of the control template (TS8). The kit included the IS and TS8.

To further determine the influence of time on telomerase activity, the RTA of two pairs of USCs (age group 20–50 years), telomerase activity positive (TA⁺) and telomerase activity negative (TA⁻), after every five passages or the end passage were measured by the above protocol (Kim et al., 1994).

The USCs in passage three were seeded in 96-well plates at a density of 2,500 cells/well. The culture medium was replaced every second day. Cell proliferation was determined on days 1, 3, 5, and 7 using an MTS cell proliferation assay kit (CellTiter 96^® AQueous One Solution Cell Proliferation Assay, Promega) according to the manufacturer’s instructions. Briefly, the MTS reagent was incubated with the cells in the dark for 1 h at 37°C. Following incubation, the absorbance was measured at 490 nm using a spectrophotometer (Molecular Devices Inc., Sunnyvale, CA, United States ). Triplicate measurements were conducted for each time point. The population doubling (PD) and doubling time (DT) calculations were determined based on single USCs at p0 up to the maximum passage (p16) for each clone. The USCs were trypsinized when they reached 70–80% confluence, and the cells were counted manually using a hemocytometer. The PD and DT were calculated using the following formula (Bharadwaj et al., 2011; Bharadwaj et al., 2013):

PD = ln (N_f/N_i)/ln (2) and DT = C_t/PD

N_f is the final number of cells, N_i is the initial number of cells, and C_t is the culture time.

Flow cytometry analysis for the USCs involved staining the USCs with specific labeled anti-human antibodies: CD25-PE, CD31-FITC, CD34-FITC, CD44-FITC, CD45-FITC, CD73-PE, CD90-FITC, CD105-PerCP-CY5.5, CD117-PE, CD140b-PE, and CD146-PE (BD Pharmingen™). Briefly, USCs (p4) were trypsinized, and 5.0 × 10⁵ cells were washed with pre-chilled PBS containing 1% bovine serum albumin (BSA). The fluorescence conjugated antibodies listed above were incubated with USCs on ice for 30 min in the dark. IgG1-PE, IgG1-FITC, IgG2b-FITC, and IgG1-PerCP-Cy^TM5.5-conjugated isotype control antibodies were used to determine background fluorescence. Cells were washed twice with wash buffer, passed through a 70 µm filter, and analyzed using FACSCalibur™ flow cytometry (BD Biosciences, Franklin Lakes, NJ, United States).

To determine the differentiation capacity difference between USCs^TA+ and USCs^TA−, cells were subjected to the following induction described below, and their changes in morphology and/or histochemical staining for specific components were recorded.

Smooth muscle cell induction: Three pairs of USCs^TA+ and USCs^TA− (p3) from different age groups were plated at 2,000 cells/cm² in smooth muscle differentiation media containing equal amounts of DMEM (high glucose) and EFM with 10% fetal bovine serum (FBS) and 2.5 ng/mL TGF-β1 and 5 ng/mL PDGF-BB (R&D Systems, Minneapolis, MN, United States). Cell morphology was recorded before and after growth factor additions for up to 14 days.

Uroepithelial induction: Three pairs of USCs^TA+ and USCs ^TA− (p3) were plated at 3,000 cells/cm² in a medium containing equal amounts of KSFM and EFM with 2% FBS and 30 ng/mL EGF (R&D Systems, Minneapolis, MN, United States ) mixed with the conditional medium of urothelial cells (UC, 1:1) for 14 days.

Osteogenic induction: USCs (p3) were seeded at a density of 4,000 cells/cm² and cultured in serum containing DMEM low-glucose medium with 100 nM dexamethasone, 10 mM β-glycerophosphate and 50 mM ascorbic acid-2-phosphate (Wako Chemicals, Richmond, VA, United States ) for 28 days. The induced cells were harvested after 28 days and fixed in 95% ethanol before histochemical staining. For detection of calcium secreted by the osteogenic-differentiated cells, Alizarin Red S staining was conducted. Briefly, the fixed cells were incubated with 0.5% Alizarin Red S dye (pH 4.1) to sufficiently cover the cell layer for 3–5 mins. Excess dye was removed with distilled water before photo documentation.

Proteins of SMC and UC-induced USCs^TA+ and USCs ^TA− and were extracted using RIPA lysis buffer with a proteinase inhibitor cocktail (Complete mini; Roche Applied Sciences). Protein samples (15–30 µg) were analyzed on a 6–12% SDS-PAGE, and after electrophoresis the proteins were transferred to PVDF membranes (Millipore, Billerica, MA, United States ). The membranes were blocked with skimmed milk and incubated with the primary antibody at an appropriate dilution overnight at 4°C. After secondary antibody incubation, hybridization was detected using the Western Lightning Chemiluminescence reagent (PE, Waltham, MA, United States).

To assess the barrier function of urothelial-differentiated USCs^TA+ and USCs^TA−, non-induced USCs^TA+ and USCs^TA−, SMCs, and ureter UCs, these cells were cultured on 0.4 µm Transwell inserts (353,090, Becton Dickinson) placed in 6-well dishes as previously reported (Cross et al., 2005) with minor modifications. Briefly, the inserts were coated with collagen IV (3 µg/cm²) and air-dried in a laminar hood. Cells (1 × 10⁵/cm²) were plated in 1.5 mL of 1 mg/mL tracer-containing medium (FITC-dextran, 4 kDa, Sigma, FD4) in the insert (top chamber) and 3 mL tracer-free medium in the bottom well. Phenol-free medium was used to avoid interference from the indicator in the assay. A day before the tracer was added, the media were supplemented with 2 mM CaCl₂ solution. After 3 h, 100 µL media aliquots were removed for fluorescence measurements (excitation at 490 nm and emission at 520 nm). Results were plotted as a percentage relative to the “no cell” control.

To determine the chromosomal stability of cultured USCs, the karyotypes of early (p4) and late passage (TA⁺ clone at p12, TA⁻ clone at p8) cells were measured (Ge et al., 2011). Cultured cells were trypsinized with a 0.25% Trypsin-EDTA solution, resuspended in hypotonic solution (0.075 M KCl) and then fixed with methanol/acetic acid solution in a 3:1 proportion. The metaphase spread on glass slides was digested by trypsin and then stained with Giemsa stain to generate G bands along each chromosome. Standard cytogenetic analysis was performed under microscopy. Chromosomal image capture and karyotyping were performed using CytoVision^®, version 3.7.

Agar assays are often used to distinguish tumor cells from non-transformed or normal cells because normal cells cannot undergo anchorage-independent growth or thrive on an agar substrate. To evaluate whether both USCs^TA+ and USCs^TA− induce tumorigenicity, USCs were tested on agar gels. HeLa cells and SMC were used as positive and negative controls, respectively. Briefly, 0.35% upper agar layer and 4% base agar layer were prepared in 35 mm tissue-treated dishes. Cells were seeded at the top of the upper agar at a density of 5,000 cells/well. Culture medium was changed twice a week. Cell morphology, proliferation rate, anchorage-independent growth, and cell colony formation were observed under a phase contrast microscope. After culturing for 2 weeks, all the samples were stained with 1 mL of 0.05% nitroblue tetrazolium (NBT) prepared in PBS and sterilized (0.2 micron). This was incubated overnight at 37°C. The cells that took up NBT and showed a violet color were determined to be live cells.

Experiments using nude mice were approved by the Wake Forest University Health Sciences Institutional Animal Care and Use Committee. To further determine the non-tumorigenicity of USCs^TA+ in vivo, two USCs^TA+ (p5) were implanted in the kidney subcapsular region of NSG (NOD.Cg-Prkdcscid-Il2rgtm1Wjl/SzJ) mice. H9 (human embryonic stem cell line) was used as a positive control. A total of 12 six-week-old female NSG mice (Jackson Labs, Bar Harbor) were used and divided into two groups: USC^TA+ (3 mice/clone) and H9 groups (6 mice/group). In total, 2 × 10⁶ cells in 20 µl of PBS were injected into the subcapsular region of the right kidney. After 8 weeks, the mice were sacrificed and bilateral kidneys were fixed in 4% paraformaldehyde, dehydrated, and embedded in paraffin. For evaluation of general graft histology and teratoma formation, routine hematoxylin and eosin (H&E) staining was performed.

A total of 80/117 USC clones (p2) from 10 individuals showed detectable TA (70.2% were USCs^TA+). There were no significant differences in the positive rate for TA in the 20–40 years age group: 20 s: 30/39 (76.9%), 30s: 14/19 (73.7%), and 40s: 16/22 (72.7%) represented USC clones, respectively, but the RTA notably decreased to 16/27 (59.2%) in the 50 years age group, as shown in Figure 1 and Table 1. TA could not be detected in human BMSCs at p2.

To determine if the TA in human USCs decreased as the passage progressed, we measured the TA of two pairs of USCs^TA+ and USCs^TA− at early passage (p2) and late passages (p7, p11 or the last passage for USCs^TA+) from the 20s and 50s groups. The strength of RTA of USC^TA+ clones decreased gradually with passage (Figure 2), which seems to make a formerly TA+ clone equivalent to a TA⁻clone. The RTA remained at undetectable levels through all the passages of USC^TA− clones.

The USC^TA+ and USCs^TA− paired clones from a single donor were tested in parallel to prevent other factors from affecting the results of the comparison. The USC^TA+ clone A35 which was isolated from a 27-year-old male donor, could be passed more than the USCs^TA− clone A42 which was isolated from the same donor, and consistently maintained the original “rice-grain like” morphology until they reached cell senescence at p16. Similarly, the USC^TA− clones steadily displayed the similar cell morphologic appearances and finally displayed a larger, flattened, and the typical “fried egg” morphology of quiescent cells at p9 (Figure 3).

According to the cell growth curve data (Figure 4), the USC^TA+ clones grew more rapidly than the USC^TA− clones. Consequently, the PD of USC^TA+ clones increased significantly compared to that of the USC^TA− clones, regardless of the individual’s age (p < 0.001). The mean DTs of USC^TA+ clones were significantly shorter than those of the USC^TA− clones (Table 2; Figure 5) (p < 0.001).

However, the relative TA of USCs was not completely coordinated with the cell proliferation capacity (PD, DT, and passage) (Table 2).

To identify the cell surface markers of USC clones, six pairs of USCs^TA+ and USCs^TA− were subjected to flow cytometry analysis. All the USC clones showed strong positive MSCs markers including CD44, CD90, CD73, CD105, and CD146 and were negative for hematopoietic stem cell markers including CD25, CD31, CD34, CD45, and CD117 (Figure 6; Table 3). However, there was no significant difference in the CD105 expression between USCs^TA+ and USCs^TA−.

The USC^TA+ and USCs^TA− paired clones from a single donor were tested in parallel to prevent other factors from affecting the results of the comparison. To test the difference in differentiation capability between USCs^TA+ and USCs^TA−, both USCs^TA+ and USCs^TA− (p3) were induced to osteogenic, myogenic, and uroepithelial differentiation (Table 4). Both the USCs^TA+ clone A35 that was isolated from a 27 year-old male donor, and the USCs^TA− clone A42 that was isolated from the same donor (p3) differentiated into the smooth muscle lineage and the urothelial lineage (Figures 7A,B). Urothelial-differentiated cells developed a cobblestone-like morphology (Figure 7Ai) and expressed the urothelial-specific proteins uroplakin-Ia and uroplakin-III and the generic epithelial cell markers CK7, CK13, and AE1/AE3 (Figure 7Aii). Furthermore, the expression of these proteins was significantly higher in the UC-induced USCs^TA+ clones [A35 and J29 (from another male aged 55 years)] than in the UC-induced USCs^TA− clones (A42 and J18 [from the same male aged 55 years as clone J29]) and uninduced USC clones. In an assay of cellular barrier function, the UC-induced USCs^TA+ clone-A35 showed increased expression of specific tight junction protein markers (E-cadherin) compared to the UC-induced USCs^TA−clone-A42 and uninduced USC clones (Figure 7Ai). In barrier function assays, both the urothelial-differentiated USCs (USC^TA+ 41.0% ± 1.7% and USC^TA− 44.8% ± 0.4%) showed significant reduction in leakage of fluorescent tracer through the insert in vitro (p < 0.01 and p < 0.001, respectively), which means lower permeability and higher tight junction property, compared to the non-differentiated USCs (USC^TA+ 63.1% ± 4.3%, USC^TA− 62.2% ± 1.1%) (Figure 7Aiii) at day 3; and similar results received at day 7 (p < 0.01), the leakage percentage of urothelial-differentiated USCs (USC^TA+-A35 35.5% ± 1.2% and USC^TA−-A42 38.9% ± 4.3%), less than the non-differentiated USCs (USC^TA+-A35 54.8% ± 2.6%, USC^TA−-A42 59.6% ± 0.5%). However, there was only a slight difference in leakage protection at day 3 between the UC-induced USC^TA+-A35 and USC^TA−-A42 clones (p < 0.05). In addition, no significant differences between them, although reduce in leakage, were noted in UC-induced USC^TA+-A35 clone on day 7.

Both myogenic differentiated USCs^TA+ and USCs^TA− became elongated and spindle-shaped (Figure 7Bi) and expressed smoothelin, a smooth muscle-specific protein marker, and calponin (Figure 7Bii). Moreover, the expression of smoothelin and calponin of SMC- differentiated USCs^TA+-A35 were higher than the SMC-differentiated USCs^TA−-A42 confirmed by western blotting (Figure 7Bii). Finally, both USCs^TA+-A35 and USCs^TA−-A42 were also induced to differentiate into the osteogenic lineage using our previous protocols. USCs^TA+-A35 could be induced to osteocytes but not USCs^TA−-A42, as confirmed by Alizarin S Red staining (Figure 7C). Moreover, both USCs^TA+-A35 and USCs^TA−-A42 were difficult to differentiate into adipocytes, as evidenced by Oil Red-O staining (data not shown).

To investigate the potential susceptibility of USCs^TA+ to malignant transformation, cells were tested via cytogenetic analysis, agar culture in vitro, and teratoma formation in vivo, and the results were compared to those of USCs^TA− and controls.

Both TA+ and TA- USC clones in the early (p4) and late passage (USCs^TA+-A35, J29 clone in p12 and USCs^TA−-A42, J18 clone at p8) displayed a normal karyotype of 1 X and 1 Y chromosome, as expected for a male donor, and a normal diploid (2n = 44) complement of autosomes and a pair of sex chromosomes (Figure 8A; Table 5). No multiploidy or obvious chromosomal rearrangements in metaphase were detected by Giemsa bandings at p4 or late passage of both USC clones.

USC^TA+ clones-A35, J29 remained the same size on day 14 as that on day 1, after being cultured in soft agar, like colonies of SMCs. However, HeLa cancer cells formed large colonies on day 14 (Figure 8Bi). These single cells (SMC and USC) or colonies (HeLa cancer cells) were viable, which was confirmed by NBT staining (Figure 8Bii). Furthermore, no teratoma was formed when all USCs^TA+-A35 were implanted in the subcapsular region of the kidney of NSG mice after 8 weeks. All human ES cell lines (H9) formed derivatives of the three embryonic germ layers (Figure 8C).