SEARCH

SEARCH BY CITATION

Keywords:

  • P-cadherin;
  • bladder cancer;
  • migration

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

OBJECTIVE

To identify changes associated with P-cadherin expression in bladder cancer and evaluate the potential role of such events in determining the clinical outcome and cell behaviour, as the function of P-cadherin in normal epithelium is unknown, as is its potential role in neoplastic progression in different cancers.

MATERIALS AND METHODS

In all, 536 bladder tumour specimens

from 408 patients were assembled in seven tissue microarrays. Paraffin sections from each array were processed for immunohistochemistry to assess the expression of P-cadherin. The expression of P-cadherin was forced using lipofectin, followed by an assessment of migration and invasion potential using standard in vitro assays.

RESULTS

The absence of P-cadherin staining was associated with muscle-invasive disease, grade 3 (P < 0.001) and nodal disease (P = 0.009). Similar results were obtained when considering cytoplasmic and unrestricted localization of P-cadherin (P < 0.001), except for nodal involvement. The group with cytoplasmic location of P-cadherin showed a shorter cancer-specific survival than the group with membrane location of P-cadherin (P = 0.03). Forced expression of P-cadherin in EJ and UM-UC-3 cells, that constitutively lack P-cadherin expression, resulted in modulation of catenin expression and enhanced migration of EJ and UM-UC-3/P-cadherin transfectants (>200%).

CONCLUSIONS

These results showed that loss of expression, cytoplasmic relocation or unrestricted tissue location of P-cadherin was associated with a poor clinical outcome and prognosis in bladder cancer. From the in vitro work it is evident that P-cadherin plays a role in regulating the migration potential of bladder carcinoma cells.


Abbreviations
TMA

tissue microarray

IHC

immunohistochemistry

DMEM

Dulbecco’s Modified Eagle’s Medium.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Cadherins have been well described as transmembrane glycoprotein molecules involved in calcium-dependent cell-cell adhesion [1]. The classic cadherins, E-, P-, N- and R-cadherin, have an extracellular N-terminal domain, a transmembrane and a cytoplasmic carboxy-terminal domain. The cytoplasmic domain of the classic cadherins binds the catenin family members, where associations are necessary to maintain the cell-cell adhesion function of the cadherin complex. Consistent with this concept, numerous studies reported the altered expression status of individual catenins linked to prognosis in different cancers [2–6].

Classic cadherin profiles have been widely studied in various human cancers, showing co-expression of classic cadherins in normal mucosa, often with a change in cadherin expression profiles accompanying disease progression. The loss or reduced expression of E-cadherin expression is a common event in tumorigenesis, resulting in increased migratory and invasive capacity, with E-cadherin playing an invasive suppressor role [7,8]. By contrast, novel expression of N-cadherin has been recorded in different tumours [9–11] and has been shown to promote the invasive potential of epithelial cells in vitro[12–14]. Although there have been few studies of R-cadherin expression in human tumours [15,16] it is proposed that R-cadherin has a tumour-suppressor function in gastrointestinal tumours [16], and has been shown to promote the migratory behaviour of epithelial cells in vitro[17].

Although over-expression of P-cadherin has been reported to promote the migratory and invasive behaviour of breast and pancreatic carcinoma cell lines [18,19], reduced P-cadherin expression has been reported to be associated with primary cultures from progressive melanomas [20], with restoration of P-cadherin expression in melanoma cell lines resulting in reduced invasive potential [21]. Loss or reduced expression of P-cadherin in tumour tissue has been reported to be associated with different clinical variables of poor prognosis in breast [22], oral squamous cell carcinoma [23] and melanocytic skin tumours [24]. By contrast, alternative studies reported elevated P-cadherin expression linked to tumour aggressiveness in endometrial cancer [25], melanocytic [26] and breast carcinomas [27]. Although P-cadherin may have a different role in tumour progression in different organs, there is no clearly identified role for P-cadherin as a clinical indicator of progression or outcome, in a specific cancer.

We previously reported an association between decreased E-cadherin expression and later-stage bladder tumours with a poor prognosis [28]. In addition, we identified novel expression of N-cadherin in human bladder tumours [9] linked to invasive behaviour in bladder carcinoma cells [14]. In the present study, we investigated the expression profile of P-cadherin at different stages of urothelial neoplastic progression using tissue microarrays (TMAs), and assessed the functional significance of forced expression of P-cadherin in bladder carcinoma cell lines lacking expression of this cadherin family member.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

We used tissue samples from patients diagnosed with bladder cancer at the Lahey Clinic Medical Center between 1990 and 2005 under a protocol approved by the insitutional review board. Formalin-fixed, paraffin-embedded tumour tissue from patients was retrieved from the archives of the Pathology Department. The TNM classification system was used.

In all, 536 tumours (269 superficial, 238 invasive and 29 Tx) were assembled in seven TMAs. All tumours were TCC and none of the patients in this study had neoadjuvant therapy. TMAs were designed with replicas for each tumour sample and each control. Controls included normal tissue from prostate, testis, tonsil, liver, cerebellum, kidney, lung and bladder. Strategic placement of control cores in each TMA enabled definitive orientation during the scoring of the TMAs. Each TMA consisted of 400 tissue cores (four cores per specimen) and control tissue cores for immunohistochemistry (IHC) validation and orientation. Several 4 µm sections were cut and stored at 4 °C in the presence of a desiccant before IHC staining. Individual TMA sections were deparaffinized and antigen retrieved in citrate buffer (pH 6.0; Dako Corporation, Carpinteria, CA, USA) for 20 min. The antibodies were diluted with Dako antibody diluent solution. For IHC staining we used an automated stainer (Autostainer Plus, Dako) using a high-sensitivity polymer-based detection system (EnVisionTM, Dako). A negative control was included using a nonspecific mouse antibody solution (Dako) substituting for the primary antibody.

Mouse monoclonal antibody to P-cadherin (Transduction Laboratories, Lexington, KY, USA) was diluted at 1:100 (2.5 µg/mL) for use in IHC. The negative control reagent (Dako) is a cocktail of non-immune mouse IgG and IgM, and was obtained pre-diluted. The specificity of the P-cadherin antibody in IHC was determined using xenograft sections derived from bladder cell lines in which the expression profile of P-cadherin had previously been identified.

The tissue sections were scored semiquantitatively, assessing staining intensity and location, i.e. membrane or cytoplasmic. A staining intensity scale of 0–3 was applied. In normal bladder tissue cores the staining intensity was recorded as 3 where P-cadherin was localized to the basal cell layer. Staining was recorded as 0 only in the presence of an identifiable basal cell layer defined by the presence of a basement membrane and underlying stroma. When P-cadherin staining was throughout the tumour tissue, staining was recorded as unrestricted. In tissue sections that had heterogeneous staining throughout the section or between cores from the same tumour, the worst-case scenario score was assigned to that sample when >5% of the tumour cells had this phenotype; 92% concordance between scores from different cores of the same tumour was recorded. In samples where scoring differed between cores the worst-case scenario score for expression level and location was recorded for the analysis. Specimens that had no staining or faint staining in <5% cells were scored as negative. Each TMA was scored independently by I.C.S. and M.L. Where discordant results were obtained, both individuals re-reviewed the stained cores to obtain a consensus. Expression of P-cadherin compared with alternative proteins of the cadherin complex was assessed using previously published results from the same TMAs [28].

Groups were compared using the Pearson chi-square test. Correlation between markers was determined by Spearman’s correlation coefficient. Survival analysis was restricted to the cystectomy group where follow-up data were available. For univariable overall and cancer-specific survival analysis we used the product-limit procedure (Kaplan-Meier method), with the surgery date as the entry date. The log-rank (Cox-Mantel) test was used to compare survival curves for different categories of each variable, and a Cox proportional hazards regression model was used for multivariable analysis.

The human bladder carcinoma cell lines EJ and UM-UC-3 were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum and penicillin/streptomycin. Cells were grown on glass slides, washed with PBS and fixed in 3.7% formaldehyde for 15 min at room temperature. Cells were then rinsed in three changes of PBS and permeabilized in 0.5% Triton X-100 in PBS. After three washes in PBS, cells were incubated with P-cadherin antibody (1:200 Cell Signalling Technology, Beverly, MA, USA) using an automated immunohistochemical processor (Model 320; Ventana Medical Systems, Tucson, AZ).

Mouse monoclonal antibodies to cytokeratin (Dako) and P-cadherin (Transduction Labs) were diluted 1:25 (0.75 µg/mL) and 1:100 (2.5 µg/mL), respectively, for use in IHC. The following antibodies were used in Western blot analysis: P-cadherin α-catenin, β-catenin, plakoglobin (1:200; 1:500; 1:1000, 1:2500, respectively, Cell Signalling Technology), α-tubulin (1:500 dilution, Sigma, St Louis, MO, USA).

For transfection, EJ and UM-UC-3 cells were plated at 3 × 105 cells per 60 mm dish 24 h before lipofection. The puromycin-resistance gene alone or the P-cadherin construct (Origene, Rockville, MD, USA) plus the puromycin-resistance gene were mixed with lipofectin reagent (Gibco BRL, Gaithersburg, MD, USA) and incubated with cells at 37 °C overnight in the absence of serum. After incubation, cells were washed and maintained in standard medium. At 48 h after transfection, cells were split into puromycin-containing medium (2 µg/mL) to select successfully transfected colonies. Individual surviving colonies screened for drug selection were ring-cloned 2 weeks later. Each colony was taken through three rounds of limiting dilution for clonal selection.

For Western blot analysis, subconfluent dishes of cells were washed in PBS, followed by lysis in hot sample buffer (2 × 0.08 m Tris, pH 6.8; 0.07 m SDS, 10% glycerol, 0.001% bromophenol blue and 1 mm CaCl2) and sheared through a 26-G needle. Lysates were then assayed for protein concentration using the BCA method (Pierce, Rockford, IL, USA). After determining the protein content, β-mercaptoethanol (1%) was added to each sample. Samples were boiled for 5 min and protein was loaded into lanes of a 7.5% polyacrylamide gel. Proteins were transferred overnight onto nitrocellulose. Membranes were blocked in 10% milk in TBS with 0.05% Tween and placed on primary antibody overnight at 4 °C. Blots were washed in TBS-Tween, three times for 15 min each wash and secondary antibody linked to horseradish peroxidase was incubated with the blots for 60 min at room temperature. Blots were then washed in TBS-Tween as previously described and developed with an enhanced chemiluminescence kit (Amersham, Arlington Heights, IL, USA).

For immunoprecipitation, subconfluent dishes of cell lines were lysed in 1 mL of PBS TDS (PBS, pH 7.4, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 2 mmol/L phenylmethylsulphonyl fluoride, and 100 µg/mL aprotinin). Lysis was carried out on ice for 20 min followed by clarification in a microfuge at 4 °C. Supernatants were removed and a 10-µL aliquot was taken for protein estimation using the BCA protein assay system (Pierce). For immunoprecipitation, protein concentrations were standardized between samples using 400 µg protein from each preparation. Lysates were incubated overnight at 4 °C with β-catenin (Transduction Lab) sandwiched with a goat antimouse anti-Ig whole molecule (Sigma). Protein A-Sepharose beads were added the next morning and incubated with the mix for 90 min at room temperature. Immune complexes were washed three times in PBS TDS followed by three washes in 0.1% PBS. To disrupt protein complexes, 35 µL of reducing sample buffer was added to each sample, boiled for 5 min and loaded onto 7.5% acrylamide gels. Western blot analysis was then performed using P-cadherin as the primary antibody.

In vitro migration and invasion assays were carried out using modified Boyden chambers consisting of Transwell (8 µm pore size; Corning Costar Corp., Cambridge, MA, USA) membrane filter inserts in 24-well tissue culture plates. For invasion assays, the upper surfaces of the membranes were coated with Matrigel (Collaborative Biomedical Products, Bedford, MA, USA) and placed into 24-well tissue culture plates containing 600 µL of NIH/3T3 conditioned media. For migration assays, no Matrigel was used and the chemoattractant consisted of 600 µL fibronectin at 10 µg/mL. Cells (1 × 105) suspended in serum-free DMEM were added to each Transwell chamber and allowed to migrate/invade toward the underside of the chamber for 4–16 h (depending on the cell line) at 37 °C. Cells that passed through the membranes were fixed in methanol, stained with crystal violet, and counted under a light microscope.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The construction of seven TMAs included 536 TCC samples from 408 patients. The TNM and histological grade of the samples is shown in Table 1. In this analysis the superficial tumour group included pTa, pTis and pT1 lesions, with the invasive group represented by T2–T4 tumours.

Table 1.  The histological and clinical variables of arrayed TCCs of the bladder
Variable (n)SamplesPatientsCystectomy
  • *

    Ta and Tis not included in the analysis. All Tis were grade 3.

No.536408215
n (%):
Pathological stage   
 pTa195 (36.4)145 (35.5)  2 (0.9)
 pTis 16 (3) 12 (2.9)  11 (5.1)
 pT1 58 (10.8) 35 (8.6) 24 (11.2)
 pT2-T4238 (44.4)203 (49.8)176 (81.9)
 pTx 29 (5.4) 13 (3.2)  2 (0.9)
Grade   
 I/II185 (34.5)138 (33.8) 16 (7.4)
 III351 (65.5)270 (66.2)199 (92.6)
Stage/grade   
 Ta/G1-G2154 (28.7) 118 (28.9)  1 (0.45)
 TaG3 41 (7.7) 27 (6.6)  1 (0.45)
 T1G2  8 (1.5)  5 (1.2)  3 (1.4)
 T1G3 50 (9.3) 30 (7.4) 21 (9.8)
 T2-T4/G2 18 (3.4) 13 (3.2) 12 (5.6)
 T2-T4/G3220 (41)190 (46.6)164 (76.3)
 TxG2  5 (0.9)  2 (0.5)  0
 TxG3 24 (4.5)  11 (2.7)  2 (0.9)
N/M stage   
 N+ (211/194/174) 67 (31.8) 61 (31.4) 49 (28.2)
 M+ (274/224/192) 16 (5.8)  11 (4.9)  7 (3.6)
Vascular invasion*   
 yes (312/243/202) 141 (45.2)126 (51.9)103 (51)
Male 308 (75.5) 
Mean (sd):   
  age at cystectomy  67.6 (10.1) 
 follow-up for survivors, years   4.34 (3.21) 

There was membranous P-cadherin staining in the basal cell layer of normal bladder mucosa (Fig. 1A), with no P-cadherin staining in 34.9% and 51.7% of superficial and invasive tumours, respectively (P < 0.001, Table 2; Fig. 1B). When considering cytoplasmic and unrestricted locations of P-cadherin (Fig. 1C,D, respectively), the results were similar. There was no nuclear P-cadherin. Invasive tumours had cytoplasmic location and unrestricted P-cadherin staining in 88.6% and 95.6%, compared with 65.1% and 73.7% in the superficial group, respectively (P < 0.001). High-grade TCCs (grade 3) were significantly linked to the same P-cadherin characteristics (P < 0.001, Table 2). Nodal disease was correlated with the absence of P-cadherin expression (47.9% node-negative vs 67.9% node-positive, P = 0.009), but not with re-localization or unrestricted staining (Table 2). There was no correlation between P-cadherin expression and lymphovascular invasion. In the univariable survival analysis, the group with cytoplasmic P-cadherin had a shorter cancer-specific survival than the group with membranous P-cadherin. The 5-year survival rate for the cytoplasmic group was 55%, while no deaths were recorded in the nine patients with P-cadherin in the membrane (P = 0.03; Fig. 2A). There was no significant cancer-specific survival difference between P-cadherin expression levels (Fig. 2B) and the restricted vs unrestricted location of P-cadherin (Fig. 2C). Although altered P-cadherin location correlated with factors known to be associated with patient outcome, it was not an independent predictor of survival in multivariable analysis.

image

Figure 1. IHC staining patterns of P-cadherin in bladder tissue; (A) normal bladder mucosa showing basal location of membranous P-cadherin; (B) low-grade superficial papillary bladder tumour showing an absence of P-cadherin even in the basal cells abutting the basement membrane; (C) cytoplasmic location of P-cadherin throughout a high-grade tumour; (D) membrane-localized P-cadherin throughout the bladder tumour. Original ×400.

Download figure to PowerPoint

Table 2.  Associations between P-cadherin staining patterns and clinicopathological factors in patients with bladder cancer
Factor% (N), P*
No stainingCytoplasmicUnrestricted
  • *

    Two-tailed chi-square test.

Tumour stage   
 Superficial34.9 (94), <0.00165.1 (114), <0.00173.7 (129), <0.001
 Invasive51.7 (123)88.6 (101)95.6 (109)
Grade   
 I/II31.4 (58), <0.00159.1 (75), <0.00173.2 (93), <0.001
 III48.7 (171)86 (154)90.5 (162)
Vascular invasion   
 No47.4 (81), 0.1685.6 (77), 0.3893.3 (84), 0.96
 Yes55.3 (78)90.3 (56)93.5 (58)
Node status   
 Negative47.9 (67), 0.00990.5 (67), 0.5789.2 (66), 0.71
 Positive67.2 (45)86.4 (19)86.4 (19)
image

Figure 2. Kaplan-Meier curves recording cancer-specific survival among patients with bladder cancer after cystectomy. P-cadherin expression (A) membrane or cytoplasmic location; (B) high or low expression levels; (C) restricted or unrestricted location in epithelial tissue. ( ) represents the number of patients in each group.

Download figure to PowerPoint

Comparative analysis of P-cadherin results with the expression of other members of the cadherin complex, as reported previously by this group using the same TMAs [28], showed a significant correlation between the absence of expression, cytoplasmic and unrestricted location of P-cadherin with low expression and cytoplasmic location of the cadherin complex components plakoglobin, E-cadherin, β-catenin, and p120ctn. Here, coincident E-cadherin and P-cadherin cytoplasmic location had the strongest correlation (R = 0.285, P < 0.001). For mesenchymal markers, the novel expression of vimentin correlated with cytoplasmic and unrestricted location of P-cadherin (R = 0.23, P < 0.001 and R = 0.16, P = 0.05, respectively).

The EJ and UM-UC-3 bladder carcinoma cells were co-transfected with the P-cadherin and puromycin-resistance gene. After selecting puromycin-resistant colonies the total cell lysates were screened in Western blot analysis to identify P-cadherin-expressing cells. Selected transfectants were taken through three rounds of limiting dilution and further screened to identify clones that expressed P-cadherin at high levels (Fig. 3A). Clones expressing P-cadherin had a more flattened morphology in both EJ and UM-UC-3 recipients. Three P-cadherin-expressing clones of EJ and UM-UC-3 were selected and maintained with three puromycin control clones, and used in experiments throughout the study. Immunoprecipitation of β-catenin from each of these clones was used to show co-precipitation of P-cadherin within a cadherin/catenin complex (Fig. 3B). Assessment of cadherin/catenin expression in the presence of forced P-cadherin expression showed an increase in α- and β-catenin, a decrease in plakoglobin (Fig. 3C,D) and no change in the levels of N-cadherin or p120ctn. Immunocytochemical staining of transfectants showed the presence of P-cadherin throughout the cytoplasm of EJ and UM-UC-3 transfectants with limited membranous staining (Fig. 4A–D).

image

Figure 3. Detection of P-cadherin protein in Western blot analysis using protein-standardized total cell lysates from transfectants (A,C,D) or after immunoprecipitation of β-catenin (B). In panels A–D, Lane 1, RT4 bladder carcinoma cell line; lane 2, EJ/puro 2; lane 3, EJ/puro 6; lane 4, EJ/puro 8; lane 5, EJ/P-cad 1; lane 6, EJ/P-cad 9; lane 7, EJ/P-cad 10; lane 8, UM-UC-3/puro 7; lane 9, UM-UC-3/puro 8; lane 10, UM-UC-3/puro 20; lane 11, UM-UC-3/P-cad 3; lane 12, UM-UC-3/P-cad 5; lane 13, UM-UC-3/P-cad 16. In panels C and D, catenin expression is recorded (α, β, γ and p120ctn) in total cell lysates from transfectants.

Download figure to PowerPoint

image

Figure 4. Immunocytochemistry showing P-cadherin expression in EJ (A,B) and UM-UC-3 (C,D) transfectants. Panel A, EJ/puro 2; panel B, EJ/P-cad 1; panel C, UM-UC-3/puro 7; panel D, UM-UC-3/P-cad 3. Original ×200.

Download figure to PowerPoint

Screening of EJ and UM-UC-3/P-cadherin transfectants and control lines in in vitro migration assays showed a significant increase in migratory capacity in P-cadherin-expressing clones. In UM-UC-3/P-cadherin clones, migration was increased by almost three times over control cell lines (Fig. 5A). Similarly, in EJ transfectants expressing P-cadherin, migration was doubled over control lines (Fig. 5A). Using the same cell pool in parallel in vitro invasion assays, there was no significant change in invasive potential between the P-cadherin-expressing and control cell lines in repeated assays (data not shown).

image

Figure 5. A, In vitro migration results showing increased migration in representative clones of both EJ and UM-UC-3/P-cadherin transfectants compared to puromycin control cells. B, in vitro migration results in EJ/P-cad 1, EJ/P-cad 9, EJ/puro 2 and EJ/puro 6 after knockdown of P-cadherin using siRNA.

Download figure to PowerPoint

To show a direct link between P-cadherin expression and increased migration we used siRNA to ‘knockdown’ P-cadherin in EJ transfectants. Knockdown of P-cadherin was confirmed in Western blot analysis and a sample of the same cells was used in in vitro migration assays. Knockdown of P-cadherin expression in EJ transfectants resulted in reduced migratory potential to 36–49% of that of the control clones represented by the same cells transfected with a scrambled sequence (Fig. 5B). EJ puromycin control transfectants showed no significant migratory change in the same experiment.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Aberrant expression of the classic cadherins has been shown to be involved in several human cancers, where the roles of E-cadherin and N-cadherin in tumorigenesis have been extensively studied. By contrast, little is known of the role of P-cadherin in normal tissue or whether changes in P-cadherin expression represent primary events in neoplastic progression. In the present study we assessed the expression and location of P-cadherin in a large series of well-characterized bladder tumours, to investigate the potential prognostic value of changing events linked to different clinical variables. In addition, we identified a role for P-cadherin in modulating the migratory behaviour of bladder carcinoma cells.

P-cadherin is localized to the basal cells in normal bladder mucosa, possibly linked to the proliferative potential or the differentiation status of this urothelial cell compartment. The apparent involvement of P-cadherin in bladder cell motility might indicate that P-cadherin has a role in regulating the migration of basal cells to the intermediate cell layer in normal bladder mucosa. The restricted expansion of P-cadherin-expressing cells in low-grade bladder tumours might indicate a proliferating basal cell compartment resulting in unrestricted P-cadherin expression in late-stage bladder tumours representative of a basal cell origin. In urothelial neoplastic progression the aberrant expression of P-cadherin, including loss of expression, cytoplasmic and unrestricted location, was found to be a poor prognostic indicator linked to increasing grade and stage. However, only the absence of P-cadherin was significantly correlated with lymph node involvement but not lymphovascular invasion. Interestingly, reduced levels of P-cadherin were previously reported to be associated with invasive bladder lesions in a study identifying differential gene expression patterns in superficial and invasive bladder tumours [29]. In the present study there was a significant differential in cancer-specific survival between patients where P-cadherin was localized to the cytoplasm or membrane. The 5-year survival rate was 55% for the cytoplasmic group, while there were no deaths in the membrane group within the period of observation. However, the latter group included only nine patients.

We previously published the expression profile of different members of the cadherin complex, including E-cadherin, β, γ and p120ctn linked to prognosis using the same TMAs [28,30]. Integration of these data showed a significant correlation between aberrant expression of P-cadherin and reduced expression, cytoplasmic location of the aforementioned cadherin/catenin components, where coincident E- and P-cadherin cytoplasmic location showed the strongest correlation. Reduced expression of E-cadherin within this tumour group has been shown to be associated with the invasive phenotype, with no difference in survival [28]. Previous reports involving immunostaining of P-cadherin in breast and melanocytic tumours showed clinical significance linked to a poor prognosis associated with both increased [26,27] and decreased [22,24] expression of P-cadherin in the same tissue. Hence, there is no clear picture as to the role of P-cadherin in the neoplastic process. It is possible that P-cadherin has no primary role in neoplasia, accounting for the lack of concordance between similar studies from the same tissue. However, notably P-cadherin knockout mice are viable, unlike their E-cadherin counterparts, and have a phenotype of precocious mammary gland development leading to hyperplasia and dysplasia [31]. Such events suggest a role for P-cadherin in neoplastic progression.

In a prognostic clinical scenario it might be necessary to partner P-cadherin status with additional biomarkers to attain any significant clinical correlates. The scoring of TMAs is semiquantitative, and given the restricted location of P-cadherin in normal bladder mucosa, caution is required in the scoring of tissue cores. For instance, the plane of the section can determine the staining observed, i.e. a section through the tip of a well-differentiated papillary frond will show no staining when indeed the lesion might retain a basal staining pattern. We hoped that the representation of four cores for each sample would resolve this issue, but in the present study we only recorded cores with no staining if a basement membrane was present. In addition, the use of a worst-case scenario score does not embrace the representation of phenotypes occurring in <5% of carcinoma cells, although they might be a significant determinant in disease progression.

We introduced P-cadherin into bladder carcinoma cell lines that lack E-cadherin and are representative of late-stage tumours, where novel expression of N-cadherin in these cells has been shown to play a primary role in establishing the invasive phenotype [14]. Immunocytochemistry showed a predominant cytoplasmic location of P-cadherin in transfectants where immunoprecipitation showed a continued association with β-catenin. Forced P-cadherin expression significantly enhanced the migratory potential of these cells but did not change their invasive capabilities.

In the present study we showed that the loss of expression, the cytoplasmic relocation or unrestricted tissue location of P-cadherin was associated with a poor clinical outcome and prognosis in bladder cancer. In addition, it is clear from our in vitro work that P-cadherin can modulate the migration potential of bladder carcinoma cells.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

This work was supported by a grant from the R. E. Wise Research and Education Institute. B. Silva Neto M.D. was sponsored by CAPES/Brazil.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES