SEARCH

SEARCH BY CITATION

Keywords:

  • CXCR4;
  • detection;
  • bladder cancer;
  • chemokine;
  • fluorescent imaging

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We previously reported that the expression of CXC chemokine receptor-4 (CXCR4) was upregulated in invasive bladder cancers and that the small peptide T140 was a highly sensitive antagonist for CXCR4. In this study, we identified that CXCR4 expression was induced in high-grade superficial bladder tumors, including carcinoma in situ and invasive bladder tumors. To visualize the bladder cancer cells using urinary sediments from the patients and chemically induced mouse bladder cancer model, a novel fluorescent CXCR4 antagonist TY14003 was developed, that is a T140 derivative. TY14003 could label bladder cancer cell lines expressing CXCR4, whereas negative-control fluorescent peptides did not label them. When labeling urinary sediments from patients with invasive bladder cancer, positive-stained cells were identified in all patients with bladder cancer and positive urine cytology but not in controls. Although white blood cells in urine were also labeled with TY14003, they could be easily discriminated from urothelial cells by their shape and size. Finally, intravesical instillation of TY14003 into mouse bladder, using N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN)-induced bladder cancer model, demonstrated that fluorescent signals were detected in the focal areas of bladder of all mice examined at 12 weeks of BBN drinking by confocal microscopy and fluorescent endoscopy. On the contrary, all the normal bladders were found to be negative for TY14003 staining. In conclusion, these results indicate that TY14003 is a promising diagnostic tool to visualize small or flat high-grade superficial bladder cancer.

Bladder cancer is the 5th most common malignancy in the Western society, and >60,000 new cases of bladder cancer are diagnosed each year in the United States, accounting for approximately 13,000 deaths annually.1 More than 70% of bladder cancers present as moderate- to well-differentiated non–muscle-invasive papillary cancer and are treated with endoscopic transurethral resection. However, 50% of patients suffer intravesical recurrence within 2 years, and 5–25% progress to muscle-invasive cancer after repeated recurrence. In particular, high-grade bladder cancer tends to invade muscle layer and metastasize. Fifty-three percent of patients with high-grade non–muscle invasive bladder cancer progress and 36% of them eventually undergo total cystectomy.2 Finding small or flat malignant urothelium, including carcinoma in situ (CIS), is essential for early detection and treatment of high-grade bladder cancer, but it is sometimes difficult to diagnose lesions by conventional “white light” cystoscopy. Therefore, it is highly desirable to establish new detection methods for high-grade non–muscle invasive bladder cancer.

For this purpose, several new optical detection methods have been developed. One approach uses a fluorescent cystoscope to visualize a precursor protoporphyrin IX after intravesical 5-aminolevulinic acid administration. This is the most common photosensitive agent today that has a high affinity for cancer cells and enhances the visual contrast between benign and malignant cells. However, because of its low specificity (33–71%), it is not easy to distinguish between malignant and inflammatory lesions.3 A second detection method uses hexaminolevulinate fluorescence cystoscopy. However, this method has a higher false-positive rate (39% vs. 31%) compared with white light cystoscopy in inflammatory lesions.4 Optical diagnosis using narrow band imaging (NBI) flexible cystoscopy is more sensitive than the white light cystoscope, and in one study, 12.6% of recurrent cancers were identified only by NBI. However, slightly higher false-positive rates were reported compared with white light cystoscope (18.2% vs. 14.2%, respectively).5 Thus, a new molecular probe to improve the optical detection of superficial bladder cancer is still desirable.

Chemokines represent another class of biomarkers and are a superfamily of small, secreted proteins (8–10 kDa) that function with their specific, transmembrane G-protein-coupled receptors in immune and inflammatory reactions, including angiogenesis, collagen production, B-cell lymphopoiesis and bone marrow myelopoiesis. Until now, 18 chemokine receptors (CCR1 to CCR10, CXCR1 to CXCR6, XCR1, and CX3CR1) have been identified. CXCR4 is the only chemokine receptor whose mRNA expression level was upregulated in bladder cancer cell lines and in muscle-invasive bladder cancer tissue samples (pT2-pT4) through screening the expression levels of all 18 chemokine receptors in normal urothelium and bladder cancer.6 CXCR4 was also reported to be highly expressed in many kinds of malignant diseases,7 including chronic lymphatic B-cell leukemia, pancreatic cancer, breast cancer, small-cell lung cancer, malignant melanoma, colorectal cancer and glioblastoma.8, 9 A 14-mer peptide T140, which was initially identified as an anti-HIV agent, specifically binds to CXCR4. In this study, we developed and evaluated a fluorescent probe of T140 analog to visualize high-grade bladder cancer.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Fluorescent CXCR4 antagonistic peptides

We previously reported that T140 is a highly specific antagonist of CXCR4. T140 consists of 14 amino acid residues (H-Arg-Arg-Nal-Cys-Tyr-Arg-Lys-D-Lys-Pro-Tyr-Arg-Cit-Cys-Arg-OH) and one disulfide bridge between Cys4 and Cys13. Fluorescent antagonist TY14003 was designed based on the biostable analog of T140.10 (Ac-Arg-Arg-Nal-Cys-Tyr-Cit-Arg-D-Lys*-Pro-Tyr-Arg-Cit-Cys-Arg-NH2; D-Lys* indicates the carboxyfluorescein-labeled D-Lys). Two negative control fluorescent peptides were synthesized: TR14009 consisting the stereoisomeric amino acids sequence (Ac-D-Arg-D-Arg-D-Nal-D-Cys-D-Tyr-D-Cit-D-Arg-Lys*-D-Pro-D-Tyr-D-Arg-D-Cit-D-Cys-D-Arg-NH2) and TR14010 has a completely different sequence but preserves the disulfide bond (Ac-Cit-Tyr-Arg-Cys-Arg-Arg-Cit-D-Lys*-Pro-Arg-Tyr-Arg-Cys-Nal-NH2). These peptides were synthesized as described previously10 and were labeled with carboxyfluorescein at epsilon-amino group of Lys (or D-Lys for TR14010). CXCR4 antagonistic activities of these T140 analogs were evaluated calculating the inhibition of [125I]SDF-1 binding to CXCR4 transfectants of Chinese hamster ovary cells, as described before.10 The bioactivity (IC50 values) for TY14003, TR14009 and TR14010 were 0.011, 1.227 and 2.030 μM, respectively. Peptides were diluted with phosphate-buffered saline (PBS) to the final concentration of 100 nM for cultured cells. For bladder instillation, the reagents were diluted either to 200 nM for confocal microscopy or to 2 μM for fluorescent endoscopy.

Cell culture

TCCSUP, KU7 and T24 human bladder cancer cell lines were used. The cells were cultured on tissue culture plates in RPMI 1640 supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin in 5% CO2 at 37°C. Cells for each experiment were harvested at 60–70% confluency. Mouse fibroblast cells were primarily cultured from peritoneal connective tissue of a C57BL/6 mouse, as described previously with some modification.11 Three millimeter square connective tissue was washed with PBS and spread on the bottom of a 6-cm culture dish for 15–30 min. After the tissue is almost dried to increase attachment to the bottom of the dish, the growth medium was gently filled and cultured up to 2–4 weeks. When the area is occupied with fibroblasts, cells are subcultured on a new dish. The cells within second passage are used for labeling by TY14003.

Human bladder cancer specimen and urine samples

Bladder cancer samples were obtained by transurethral cold-cup biopsy. They were analyzed by western blot (n = 5), immunohistochemistry (n = 41) or immediately used in imaging using the fluorescent T140 analog TY14003 (n = 6). Two nonmalignant urothelial samples were taken from a patient with a contracted bladder due to spinal disease and a patient with solitary urothelial cancer in the prostatic urethra. Voided urine samples from patients with bladder cancer (n = 9) and chronic cystitis due to neurogenic voiding dysfunction (n = 3) were collected. In addition, urine samples from patients who finished intravesical Bacillus Calmette-Guérin (BCG) therapy were also collected (n = 2). Ten milliliters of fresh urine was subsequently centrifuged by 3,000 rpm for 5 min. The supernatant was discarded, and the sedimentation was gently mixed with 20 μL of 100 nM TY14003. Five minutes later, a droplet was mounted on a glass slide and observed by confocal microscopy. Informed consent on written documents was obtained from each patient before examination. TY14003 positive- and negative-stained urothelial cells were counted in patients with bladder cancer (n = 3, class5 cytology), within 10 fields of view (×200).

Mouse bladder cancer model

A total of 50 C57BL/6 mice were housed in a plastic cage with wood chips for bedding in a specific pathogen-free room kept at 24 ± 2°C and 40–70% humidity with a 13/11-hr light/dark cycle. Six- to eight-week-old mice were given drinking water with 0.025% N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN) (Tokyo Kasei Kogyo, Japan), whereas control mice were given water alone. The BBN solution was prepared freshly twice a week and administered in drinking bottles. Histopathologic analysis of bladders from mice fed with BBN for 4, 8, 12, 16, 20, and 24 weeks was performed. The mice were euthanized with CO2, and the bladders were harvested. Negative-control mice were sacrificed at 16–20 weeks of age. For light microscopy, bladders were fixed in formalin and embedded in paraffin. For routine histology, 5-μm sections were cut and stained with hematoxylin–eosin in accordance with standard procedures. The histopathologic diagnosis of the bladder cancer was given by a pathologist (H.K.)

Western blot analysis

Total cellular protein extracts (20 μg) from cultured cells, mouse bladder tissues and human bladder urothelium were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis using a 12% acrylamide gel. Proteins were transferred to a polyvinylidene fluoride membrane. The membrane was blocked for 60 min with 5% dry milk and incubated overnight at 4°C with antibodies: anti-CXCR4 rabbit polyclonal antibody diluted 1/1,000 and anti–beta-actin mouse monoclonal antibody diluted 1/20,000 (Abcam, Cambridge, UK). The membranes were washed 3 times, and incubated at room temperature with the appropriate horseradish peroxidase-conjugated secondary antibodies for 1 hr. After washing the membranes, stained bands were detected using an ECL-advanced system (GE Healthcare, UK)

Immunohistochemistry for CXCR4 in mouse and human bladder cancer

Nine and 4 mouse bladders after 8 and 14 weeks of BBN were inflated with and fixed in 10% formalin, respectively. Five mice without BBN were also used as negative controls. Human bladder cancer specimens obtained by transurethral cold-cup biopsy were also fixed in 10% formalin overnight. The tissues were embedded in paraffin and were stained with hematoxylin–eosin in accordance with standard procedures. To assess the expression of the chemokine receptor CXCR4, the formalin-fixed and paraffin-embedded bladder was stained with a CXCR4 rabbit polyclonal antibody (Abcam, Cambridge, UK), as described previously.12 After deparaffinization, endogenous peroxidase activity was blocked by 0.3% H2O2 in methyl alcohol for 30 min. The glass slides were washed in PBS (6 times, 5 min each) and mounted with 1% goat normal serum in PBS for 30 min. Subsequently, primary antibody was applied at a dilution of 1:800 overnight at 4°C. They were incubated with biotinylated goat anti-rabbit serum (second antibody) diluted to 1:300 in PBS for 40 min, followed by washes in PBS (6 times, 5 min). Avidin–biotin–peroxidase complex (ABC) (ABC-Elite, Vector Laboratories, Burlingame, CA) at a dilution of 1:100 in bovine serum albumin was applied for 50 min. After washing in PBS (6 times, 5 min), coloring reaction was performed with diaminobenzidine (DAB), and nuclei were counterstained with hematoxylin. The specificity of the staining was confirmed by using the corresponding rabbit IgG as a negative control for the primary antibody. Staining intensity was judged on the basis of cytoplasmic stain in the majority of the mucosa for each section by the 2 independent observers. Faint stain in normal urothelium is defined as (−), and the stain often demonstrated in cancer with submucosal invasion is defined as (2+). Intermediate stain is defined as (1+) stain. To quantify the CXCR4-stain, densitometric scans were also performed with ImageJ software (National Institutes of Health; http://rsbweb.nih.gov/ij/). Images were saved as gray TIFF files for analysis. The signal density of CXCR4 in the cytoplasm of urothelial cells was measured with ImageJ by the gray value (black: 0; white: 255). As the background signals, those in submucosal tissue were also quantified by the gray value. The gray values were measured for the randomly selected fields. Mean gray values were used as a measure of the selected fields. The ratio of signal density in epithelial cytoplasm compared with that in the submucosa is defined as “fold signal density,” which is calculated as follows: mean gray value in submucosa/mean gray value in epithelial cytoplasm (Supporting Information Fig. S3). In each group, the densities were compared with those in another group by post hoc test Protected Least Significant Difference (PLSD) using commercially available software.

T140 labeling in cultured cells

Cultured cells were rinsed with PBS and then incubated with 100 nM of TY14003 for 5 min at room temperature, rinsed with PBS, and observed by Biozero confocal fluorescent microscopy (Biozero; KEYENCE, Japan). Experiments were repeated >3 times using different mice in each group.

T140 fluorescent imaging in mouse bladder cancer

Mouse bladders with 12 weeks of BBN intake were instilled with TY14003. For microscopic observation, they were euthanized 5–10 min later, and their bladders were harvested and sectioned in the mid-sagittal plane. The bladders were rinsed with PBS twice, placed on a plastic dish and observed using Biozero confocal laser microscopy (Supporting Information Fig. S1). For endoscopic observation, the anterior bladder walls were cut and opened under anesthesia 5–10 min after the instillation. Bladder mucosas were washed twice and observed either by endoscopy with white light or fluorescent light (Olympus Optical, Tokyo, Japan).

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

High expression of CXCR4 in human and mouse high-grade bladder cancers

CXCR4 is highly expressed in human invasive bladder cancers.6 Consequently, we assessed the induction kinetics of the CXCR4 biomarker in a mouse BBN bladder carcinogenesis model. BBN is a well-known carcinogen and induces high-grade and invasive bladder cancer in mice (Fig. 1a). Bladder CIS developed in 2 of 9 mice (22%), 7 of 8 mice (86%) and all 12 mice (100%) pathologically after 8, 12 and 16 weeks of 0.025% BBN drinking, respectively. Four of 12 mice (33%) mice had invasive bladder cancers (more than pT1) after 16 weeks of BBN (Fig. 1a). Immunohistochemistry demonstrated that staining of CXCR4 in the bladder was detected in one third of mice examined after 8 weeks and in all mice after 14 weeks of BBN drinking, whereas faint staining was detected in control mice. The signal density of CXCR4 also began to increase from the early stage of carcinogenesis (Fig. 1b). Furthermore, the upregulation of CXCR4 expression was confirmed by Western blot analysis (Fig. 1c). These data indicate that the expression of CXCR4 was induced in the early stages of carcinogenesis of high-grade bladder cancer.

thumbnail image

Figure 1. (a) Mouse bladder cancer development in a mouse BBN bladder cancer model. (b) Immunohistochemistry of CXCR4 in mouse bladder cancer caused by BBN drinking for 8 and 14 weeks. CXCR4 is expressed almost in accord with the stage of mouse bladder carcinogenesis. CXCR4 staining in bladder tumors rated as (1+) and (2+) were 33% and 100% after 8 and 14 weeks drinking of BBN, respectively. The signal density of CXCR4 also increased from the early stage of carcinogenesis. Fold signal density is the ratio of signal density in epithelial cytoplasm compared with that in the submucosa. Statistical analysis: post hoc test (Fisher's PLSD). NCT: negative control using corresponding rabbit IgG. Normal bladder is used in the example of (−) stain. (c) Representative data of Western blot analysis in mouse bladder cancer. CXCR4 is expressed in all the mice pTis bladder cancer caused by 0.025% BBN for 8 weeks.

Download figure to PowerPoint

We assessed the expression of CXCR4 in human bladder CIS or high-grade superficial bladder cancers by western blot analysis and immunohistochemistry. Western blot analysis demonstrated that CXCR4 was strongly expressed in all human high-grade bladder cancers and bladder cancer cell lines examined. On the other hand, the biomarker was not detected in normal bladder urothelial cells (Fig. 2a). Immunohistochemistry using anti-CXCR4 antibody demonstrated that (+1) to (+2) staining was observed in almost all of the high-grade bladder cancer (n = 6) and CIS (n = 16), in comparison with nonmalignant urothelium (n = 6) and low-grade pTa cancer (n = 13) in which 17% and 23% showed only (1+) staining for CXCR4, respectively. The signal densities of CXCR4 were significantly increased in CIS and high-grade bladder cancer (Fig. 2b).

thumbnail image

Figure 2. (a) Western blot analysis for CXCR4 of human bladder cancer specimens and bladder cancer cell lines. All the cell lines expressed CXCR4 and high-grade superficial bladder cancer strongly expressed CXCR4. Nonmalignant urothelium did not show obvious expression of CXCR4. (b) Immunohistochemistry of human bladder cancer specimens also demonstrated weak and moderate stain for CXCR4 in CIS and high-grade bladder cancer, respectively. Nonmalignant urothelium and low-grade cancer showed faint stain for CXCR4. Immunohistologic stains were rated as (−), (1+) and (2+). The signal densities of CXCR4 were significantly increased in CIS and high-grade bladder cancer. Statistical analysis: post hoc test (Fisher's PLSD).

Download figure to PowerPoint

Ex vivo fluorescent labeling of human bladder cancer cells with TY14003

Because T140 is a peptide that binds specifically to CXCR4, our expectation was that bladder cancer cell lines could be labeled by TY14003. To confirm this notion, TCCSUP cells were labeled with TY14003 without formalin fixation, and a preferential cell membrane labeling pattern was observed (Fig. 3a). Similar staining was also observed in KU7 and T24 human bladder cancer cell lines (data not shown), although no staining was seen in mouse fibroblast cells (Fig. 3b). Preincubation of 100 ng/mL of SDF1, a specific ligand for CXCR4, caused internalization of CXCR4 leading to the decreased signals of TY14003 in TCCSUP cells (Fig. 3c). Two negative-control peptides, TR14009 and TR14010, failed to stain TCCSUP cells (Supporting Information Figs. S2a and S2b).

thumbnail image

Figure 3. TY14003 is a specific CXCR4 antagonist. Confocal laser microscopy with phase-contrast images and fluorescent images of the same fields of cells are shown in the left and right panels, respectively. (a) Fluorescent imaging of TCCSUP cells using TY14003 (×400). (b) Mouse fibroblast cells cultured from peritoneal connective tissue is not labeled by TY14003 (×400). (c) TY14003 cannot bind to internalized CXCR4, 5 min after administration of SDF-1 (×400) (d) Microscopic images of human urine sedimentation using TY14003 with a representative view of urothelial cells from patients with invasive bladder cancer. The urine cytology was class 4 (×200). (e) Urine sedimentation after sixth weekly intravesical BCG therapy. Urothelial cells are not labeled by TY14003 (white arrow), although white blood cells are labeled (dark arrow head). The urine sedimentation turned out to be class 2 (×200) (f) Ex vivo imaging of human bladder cancer using TY14003 (×50). Bladder cancer specimen was obtained by cold-cup biopsy.

Download figure to PowerPoint

thumbnail image

Figure 4. Representative microscopic view (a) and macroscopic view (b) of cut surfaces of mouse bladders. 0w: normal bladder, 12w: carcinoma in situ (CIS) caused by BBN. (c) After 12-week BBN drinking, multiple CIS lesions were observed by confocal laser microscopy with fluorescent images. Confocal laser microscopy of mouse bladder mucosa with phase-contrast image (upper column, ×100) and fluorescent image using TY14003 (lower column, ×100 and ×400). (d) Mouse bladders were directly observed under anesthesia by endoscope with white light images and fluorescent images. TY14003 instilled into bladder was illuminated through the bladder wall (right panel). (e) After the anterior bladder walls were cut and opened, bladder mucosa was washed twice and observed by white light (upper panels) and fluorescent light (lower panels). Fluorescent stain for TY14003 was observed in mice after 12-weeks BBN drinking (circle).

Download figure to PowerPoint

To test the utility of this biomarker in the clinical setting, urinary sediments and biopsy specimens from patients with bladder cancer were analyzed to assess whether TY14003 labeled bladder cancer cells under these conditions. As shown in Figure 3d, positive-stained cells were identified in patients with bladder cancers and positive urine cytology, but not in patients without bladder cancers and positive urine cytology. In patients with bladder cancer, the average ratio of positive TY14003 staining in urothelial cells was 78.8% (standard deviation, 9.4%). TY14003 displayed minimal fluorescence binding to nonmalignant urothelial cells in patients with previous intravesical BCG therapy and class 2 cytology (Fig. 3e) or in sediments from patients with chronic urinary infection (n = 4, data not shown). As anticipated, white blood cells in urine sediments were also labeled with TY14003. Microscopically, these inflammatory cells were easily discriminated from urothelial cells based on size and shape criteria. Cold-cup biopsy specimens of bladder cancers were labeled by TY14003 in all 6 patients examined (Fig. 3f), whereas the negative control, TR14009, failed to stain bladder cancer cells (Supporting Information Fig. S2c).

In vivo fluorescent labeling of mouse BBN-induced bladder cancers with intravesical instillation of TY14003

TY14003 was administered intravesically into mice with BBN-induced bladder cancer. In our experiments, high-grade superficial bladder cancers, including CIS, were observed after 12 weeks of BBN drinking (Fig. 1a; Figs. 4a and 4b). Fluorescent signals were detected in focal areas of bladder of all mice at 12 weeks of BBN drinking by intravesical instillation of TY14003 using confocal microscopy, although bladder cancers could not be identified both by direct vision and by phase-contrast microscopy because of flat tumors (Fig. 4c). All normal bladders were negative for staining using TY14003 (n = 3). Negative-control peptides, TR14009 and TR14010, failed to show obvious signals in both normal mouse bladder (Supporting Information Figs. S2d and S2e) or after 16 weeks of BBN drinking (Supporting Information Figs. S2f and S2g). Next, in vivo observation of the mouse bladder was performed by endoscope with fluorescent light. Focal fluorescent stains for TY14003 could also be observed by endoscopy in the mouse bladders 12 weeks after BBN drinking (n = 3; Fig. 4e).

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

This study confirms that CXCR4 is a potential molecular marker for high-grade bladder cancer and is expressed early in the mouse BBN-induced carcinogenesis model and in human bladder cancer clinical samples. BBN is one of the most suitable carcinogens for investigating invasive bladder cancer in mouse models because BBN specifically and effectively causes invasive bladder cancer in mice. Under our experimental conditions, CIS was microscopically visible after 8 weeks, and submucosal invasion could be seen after 16 weeks of BBN. Interestingly, CXCR4 was initially observed at 8 weeks of BBN before development of invasive bladder cancers. One report confirmed that CXCR4 is strongly expressed in high-grade and muscle-invasive bladder cancer,6 although there has been no study analyzing the CXCR4 expression especially in CIS or high-grade non–muscle-invasive bladder cancer. Our immunohistochemical analysis demonstrated that all the CIS and high-grade non–muscle-invasive bladder cancer expressed CXCR4, although low-grade non–muscle-invasive bladder cancer only had weak expression of CXCR4, consistent with the previous reports. Because normal urothelium expresses only faint CXCR4, these expression profiles indicate that CXCR4 is a potential new molecular target for the detection of high-grade non–muscle-invasive bladder tumors, including CIS.

Many genetic and epigenetic alterations have been identified in bladder cancers through comprehensive analysis, and some molecules demonstrated increased expression in non–muscle-invasive bladder cancers. Some molecular markers, such as NMP22, UroVysion/FISH, ImmunoCyt, seem to have some utility as complementary or replacement of cytology. However, none of them have been proved sensitive and specific enough to replace cystoscopy.13 Desirable conditions for imaging cancers using detection markers are—first, molecules localized in the nucleus/cytoplasm are not suitable for detection marker. Second, they should stain only bladder tumors and demonstrate good contrast against normal mucosa. Third, the costs for processing molecular probes should not be expensive. Fourth, the test should not be influenced by its own technical difficulty. The 14-mer peptide T140 was developed from structure–activity relationship studies of 2 antibacterial and antiviral peptides; 17-mer peptide tachyplesins and 18-mer peptide polyphemusins isolated from the Japanese horseshoe crab (Tachypleus tridentatus) and the American horseshoe crab (Limulus polyphemus), respectively. T140 possesses potent anti-HIV activity by binding specifically to CXCR4. The fluorescent-tagged T140 derivative, TY14003, has some desirable features for the detection markers of bladder cancer. First, TY14003 is a specific ligand for the transmembrane receptor CXCR4. Second, TY14003 is differentially expressed in benign and malignant cells because of its high solubility and good clearance in vivo.10 Finally, TY14003 should be available at low cost for clinical use because this peptide can be manufactured easily compared with antibodies.

In this study, we developed a new molecular probe, TY14003, from T140 and evaluated its efficacy to detect bladder cancer by 2 methods: direct application on human urinary exfoliated cells and intravesical administration into the mouse bladder. First, TY14003 is also an effective tool for microscopic detection of cancer cells in urine sedimentation. Even faint fluorescent signal in a single cancer cell can be detected in microscopic fluorescent observation, although the other molecular markers secreted in urine would be undetectable in such a small amount. Fluorescence microscopy using TY14003 can easily discriminate cancer cells from white blood cells by their shape. Although urine cytology is still a standard method for detection of bladder cancer, its sensitivity and specificity largely depend on the skill of the pathologists. On the other hand, fluorescent detection using TY14003 seems easier and requires less skill compared with conventional cytology. This method is based on the biological signal of cancer cells, although conventional cytology is based on the shape, nucleus, and nuclear stain of the cell. To verify the sensitivity and specificity, fluorescent cancer detection in urine needs to be further analyzed with a large number of cohorts. Second, we instilled TY14003 into bladder using a mouse model to evaluate the possibility of in vivo fluorescent observation of bladder cancer in human. One advantage of intravesical instillation of TY14003 is that it is safer than intravenous or intraperitoneal administration. When the drug is instilled into the bladder only a small amount is absorbed systemically. Although we did not observe lethal reaction in a mouse model, minor and long-term adverse events after the instillation should be evaluated in toxicity studies. Another advantage is that intravesical instillation can directly and effectively label the urothelial cancers on bladder mucosa compared with intravenous administration. To realize the clinical utility, prospective studies need to be conducted to verify the safety of intravesical instillation and efficacy to detect high-grade and non–muscle-invasive bladder cancer.

There may be a limitation in detecting low-grade bladder cancer by TY14003, even though it is a promising molecular probe to detect high-grade non–muscle-invasive bladder cancers. We have not evaluated the ability of TY14003 to detect low-grade bladder cancer. Although high-grade bladder cancers express CXCR4, low-grade bladder cancers express CXCR4 weakly. This implies that there is a limitation of detecting low-grade bladder cancer using TY14003. There are now high-speed and high-resolution instrumentations for scanning the fluorescence-stained samples using image analysis techniques: the NanoZoomer HT (Hamamatsu Photonics, Japan), BD FocalPoint™ GS Imaging System (Becton, Dickinson and Company, NJ), Naunce System (Cambridge Research & Instrumentation, Inc., MA), etc. Using such a system, TY14003 might possibly detect low-grade bladder cancer cells. On the other hand, we should also make effort to find and develop a good molecular marker for low-intermediate cancer.

In conclusion, these results indicate that TY14003 may be a useful diagnostic molecular reagent for detecting bladder cancer. It may be particularly useful for detecting CIS and high-grade bladder cancer at an early stage.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We thank Mr. Satoshi Takekoshi (Olympus Optical Co. Ltd, Tokyo, Japan) for the technical assistance.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_25145_sm_supfig1.tif412KSupporting Figure 1
IJC_25145_sm_supfig2-A-F.tif2652KSupporting Figure 2 A–F
IJC_25145_sm_supfig2-G.tif583KSupporting Figure 2G
IJC_25145_sm_supfig3.tif429KSupporting Figure 3
IJC_25145_sm_supfig4.tif781KSupporting Figure 4

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.