Introduction
Carcinoma of the bladder is the most frequent tumor in the urological tract, in which transitional cell carcinoma of bladder (BTCC) accounts for over 90 percent. With the characteristics of easy recurrence, some BTCCs advance to greater malignancy or metastasize to other organs when they recur, making them difficult to be cured.
Phosphorylation and dephosphorylation are the major mechanisms that regulate the functional activities of diverse proteins and correlate with many cellular processes, such as transcriptional regulation, apoptosis, the cellular cycle, protein degradation, and message transferring [
1,
2]. As a superfamily of dephosphorylation, protein tyrosine phosphatase (PTP) is defined by the signature C (X) 5R active site motif, and is encompassed by a large group of enzymes that play a key role in the regulation of protein dephosphorylation. In recent years, several members of the PTP family have been found and proven to have a close correlation with the growth of many tumors.
The recently reported phosphatase of regenerating liver (PRL) is a new member of the PTP family. Until now, PRL has been found to consist of PRL-1, PRL-2 and PRL-3, which are closely related with each other. PRLs are highly homologous in amino acids, at 89% between PRL-1 and PRL-2, 78% between PRL-1 and PRL-3, and 79% between PRL-2 and PRL-3 in humans. Among all PTPs, only the PRL protein is characterized by containing a CaaX box at its carboxyl termini where they are posttranslationally farnesylated, while others do not have it [
3].
Up-regulation of PRL genes appears to be associated with cell growth, differentiation, motility, oncogenic transformation, tumor invasion and metastasis. PRL-1 was the first identified immediate-early gene that could be induced in the rat regenerating liver [
4]. Zeng and colleagues demonstrated that the highly expressed PRL-3 in ovary cells of Chinese hamster manifested enhanced motility, invasive activity and metastatic tumor formation in nude mice [
5]. PRL-2 was up-regulated in prostate cancer [
6]. Saha and colleagues found that PRL-3 was expressed in a higher level in human metastatic tissues derived from colon cancer compared to the primary cancers or normal colorectal epithelium [
7]. Previous experiments identified the abnormal expression of PRL in gastric, colorectal, ovarian and breast cancers in human beings and other tumor cell lines [
7-
10]. The above-mentioned researches indicate the potentiality of PRL as a novel tumor biomarker which may contribute to the early diagnosis of metastasis. Therefore, the measurement of PRL has a great value in clinical application. The present study is aimed at examining the changes of mRNA and protein expression of PRL-1 and PRL-3 and certifying the relationship between BTCC and pathological changes of PRLs.
Materials and methods
Patients and samples
Immunohistochemical staining was done on 30 cases of BTCC and 15 cases of normal bladder mucosa which all were treated in theSecond Aaffiliated Hospital of Zhengzhou University from 2006 to 2007. The patients’ age ranged from 43 to 87, with a mean age of 65.77 years. Among 30 cases of BTCC, there were 21 males and 9 females. The samples were fixed in 10% neutral-buffered formalin, routinely processed, embedded in paraffin, cut into 4-μm thick serial sections and mounted on plus slides. In the same period, the mRNA expression was detected in fresh BTCC tissues of 30 other patients (20 males and 10 females) and fresh normal bladder mucosa of 10 cases that were treated in the same hospital. The age ranged from 15 to 80, with a mean age of 59.73 years. The samples were cut and immediately stored at -80°C until processing. All these samples were confirmed by pathology. According to World Health Organization (WHO) pathological grade, the BTCC cases were classified into the grade I (hyper-differentiation), II (medium-differentiation) and III (hypo-differentiation) respectively, whereas superficial invasion (Tis-T1) and deep invasion (T2-T4) were staged according to the TNM-Union International Contre Le Cancer classification system.
Immunohistochemistry
Paraffin-embedded sections (4-μm thick) were deparaffinated, rehydrated and repaired by microwaving in citrate buffer (10 mol/L) for 5 min. Then the slides were placed in 3% hydrogen peroxide solution (H2O2) to inhibit endogenous peroxidase activity. After that the slides were blocked with 1% goat serum albumin for 30 min and subsequently incubated with PRL-1 or PRL-3 monoclonal antibody (Institute of Molecular and Cell Biology in Singapore, 1∶70 dilutions) overnight at 4°C. The next day, the reaction product was visualized by incubation with a secondary antibody, streptavidin-peroxidase (SP) and diamino-benzidine (DAB) for 2 min at room temperature. After each step, the slides were rinsed gently with phosphate-buffered saline (PBS) thrice. The counterstaining was performed with hematoxylin. For the negative control, the primary antibody was substituted with PBS. We used pancreatic carcinoma and rectal carcinoma as the positive controls.
The staining in the cytoplasm and the cytoblast was evaluated. The scores for PRL-1 and PRL-3 staining was graded as follows: no staining or staining observed in no more than 30% of tumor cells was scored 0; faint/barely perceptible staining or detected in 30%-60% of tumor cells was scored as 1; yellow or deeper yellow staining or ≥60% of tumor cells was given a score 2. According to the cross products of the staining and the quantity of the tumor cells, a score of 2 was considered positive. The immunostaining was evaluated independently by three oncologic pathologists without any knowledge of the clinical data.
Semi-quantitative reverse transcription polymerase chain reaction (RT-PCR) analyses
RNA isolation
Total RNA was isolated from fresh BTCC tissues and normal bladder mucosa tissues over ice following the Trizol reagent directions (Invitrogen, USA), and the total RNA was stored in non-RNAase water. RNA results were quantified by absorbance (A) at 260 and 280 nm.
RT-PCR
Total RNA was reversely transcribed and polymerase chain reaction was done following the TaKaRa PrimeScriptTM RT-PCR kit introduction (TaKaRa Biotechnology, Dalian, China), and the positive control β-actin (20 µmol/L) was amplified in the same RT-PCR tube. The PCR reaction consisted of 35 cycles of 94°C for 30 s; 49°C (PRL-1) or 52°C (PRL-3) for 30 s; and 72°C for 1 min. The sequences of specific primers were as follows: PRL-1 forward 5'-TACTGGATTGAAGAATTGCTGCT-3', and reverse 5'-ACAACAACCAGGTTCTTCACGA-3'; PRL-3 forward 5'-GCACCTTCATTGAGGACCTGA-3', and reverse 5'-CCTGAGCTACATAACGCAGCA-3'; β-actin forward 5'-CAAGGCCAACCGCGAGAAGATG-3', and reverse 5'-GTCCAGGGCGACGTAGCACAGC-3'.
The PCR products were separated by electrophoresis in 2% agarose gel (80 mV, 30 min) and visualized by staining with ethidium bromide. The product of PRL-1 was a 490 bp fluorescent light visualized using the ultraviolet facility (Stratagene, USA), and there was a 473 bp fluorescent light for the PRL-3 product. β-actin was visualized as a 330 bp fluorescent light.
Statistical analysis
All data were analyzed by Statistical Program for Social Sciences 11.0 statistical package. A Fisher’s exact test or chi-square test (c2-test) of probabilities was performed to assess the associations between PRL-1 and PRL-3 positive expression rate and their respective clinicopathological characteristics. The correlation analysis between PRL-1 and PRL-3 was studied by Spearman correlation. A two-tailed P value of less than 0.05 was considered statistically significant.
Results
Expression of PRL-1 in BTCC and normal bladder epithelia
Immunohistochemistry revealed that the expression level of PRL-1 protein was higher in BTCC than in normal bladder epithelia. Cytoplasm and cytoblast staining was identified by the presence of the yellow or dark yellow granular immunoreaction products (Fig. 1). PRL-1 was positively expressed in 22 of 30 (73.3%) cases of BTCC, and in 6 of 15 (40.0%) of normal bladder epithelia (Table 1), with the difference being significant between them (c2=4.727, df=1, P=0.030<0.05). Compared with BTCC patients with superficial invasion (3 of 9, 33.3%, Fisher’s Exact Test, P=0.003<0.05), the positive expression rate of PRL-1 protein was significantly increased in those with deep invasion (19 of 21, 90.5%), which indicated the potential association of PRL-1 expression with the invasion and metastasis of BTCC. Our study also detected an increased rate of positive staining of PRL-1 which paralleled the pathological grade from grades I to III (Table 1), although there was statistical difference only between grade I and grade III (Fisher’s Exact Test, P=0.032<0.05). The present results indicated that PRL-1 might play an important role in carcinogenesis of BTCC.
Expression of PRL-3 in BTCC and normal bladder epithelia
The immunohistochemical staining of PRL-3 revealed the same distribution (cytoplasm and cytoblast) as that of PRL-1 (Fig. 2). The proportion of positive staining of PRL-3 was significantly increased in BTCC (56.7%), compared with that in normal epithelia (20.0%) (c2=5.455, df=1, P=0.020<0.05, Table 1). The positive rate of 71.4% in the deep invasive group was significantly higher than the 22.2% in the superficial invasion group (Fisher’s Exact Test, P=0.020<0.05, Table 1). There was no significant difference among pathological grades (Table 1).
Expression of PRL-1 and PRL-3 mRNA in BTCC and bladder normal epithelia
The expression of PRL-1 mRNA was detected for a 490 bp fluorescent light in UV, and 473 bp for PRL-3 (Fig. 3). In all of the 30 cases of BTCC, we found 17 positive cases of PRL-1 mRNA (56.7%) and 10 cases of PRL-3 mRNA (33.3%). In 10 cases of normal urothelia, there was one case positive for PRL-1 mRNA and no positive cases of PRL-3 mRNA, which revealed a statistically significant difference between BTCC and normal urothelia in both PRL-1 and PRL-3 mRNA expression (Fisher’s Exact Test, all P<0.05, Table 2). There was a significant increase of PRL-1 mRNA in the deep invasive group, compared with the superficial invasion group (Fisher’s Exact Test, P=0.025<0.05) (Fig. 3 and Table 2). There was no change in PRL-3 mRNA (Table 2).
Expression of PRL-1 and PRL-3 in different gender, age, and recurrence groups in BTCC
To detect the relationships of PRL-1 and PRL-3 with different genders, ages, and the recurrence groups, we analyzed the data of PRL-1 and PRL-3 protein and their mRNA expression and did not find any significant differences among the different groups not only for PRL-1 but also for PRL-3 (all P>0.05). These results indicated that there was no relationship between the expression of PRL-1 and PRL-3 and gender, age or recurrence (Tables 1 and 2).
Analyses of the correlation between PRL-1 and PRL-3
The statistical analyses indicated no significant difference in the positive expression rate of protein and mRNA between PRL-1 and PRL-3 in BTCC. However, a positive correlation was found between PRL-1 and PRL-3 in the expression of protein and mRNA (Tables 3 and 4).
Discussion
As a subgroup of the protein phosphatase superfamily, PTPs play a fundamental role in regulating diverse proteins, therefore participating in every aspect of cellular physiologic and pathogenic processes such as transcriptional regulation, apoptosis, cellular cycle, protein degradation, and message transferring [
11]. PRLs represent a novel class of the PTP family and consist of 3 members (PRL-1, 2, 3) which are highly homologous in their amino acids. Previous research has proved that PRLs facilitate cellular proliferation, activation, invasion and metastasis, therefore inducing the formation of metastasis [
5]. PRL-1 is the first one to be identified as an immediate early gene in the liver regeneration of rat [
4]. Thereafter, the other two members of PRLs, PRL-2 and PRL-3, were found to be expressed in normal tissue and many kinds of tumors. PRL-1 and PRL-2 proteins distribute in a widespread manner in human normal tissues. The PRL-1 expression level is more variable and shows less intensity than the level of PRL-2 in the same tissue or cell types [
11]. Previous studies demonstrated the preferential expression of PRL-3 in the normal tissues of heart and skeletal muscle [
12]. Overexpression of PRL-1 and PRL-2 is associated with the transformation of mouse fibroblasts and hamster pancreatic epithelial cells in culture and facilitates tumor growth in nude mice, suggesting that both PRL-1 and PRL-2 participate in tumorigenesis [
13]. Similarly, PRL-3 has been proven to enhance the growth of embryonic kidney fibroblasts in humans [
12]. Expressed in many tumors, with particularly high expression in the metastasis of human colorectal cancer, PRL-3 is supposed to play an important role in the invasion and metastasis of malignant tumors, which indicates a new role for molecular biomarkers. Nowadays, more and more studies have reported on the abnormal expression of PRLs in various human tumors, such as colorectal tumor, gastric tumor, ovarian cancer, and hepatic cancer [
14-
17], and proved their close correlations with metastasis and invasion of tumors. However, up to now, there has been no report about the expression of PRLs in BTCC except for research on the protein expression of PRL-1 and PRL-2 in normal urological epithelia. By using RT-PCR and immunohistochemistry (streptavidin-perosidase, SP), we measured the mRNA and protein levels of PRL-1 and PRL-3 in BTCC and normal bladder mucosa. This study firstly revealed the up-regulated proteins and mRNA levels of PRL-1 and PRL-3 in BTCC compared with the normal urothelia. Zeng and colleagues have proved the enhanced ability for invasion and metastasis by transferring PRL-1 and PRL-3 to Chinese hamster ovaries [
5]. In accordance with that, our finding also demonstrates the overexpressed protein levels of PRL-1 and PRL-3 in deep invasion compared with superficial invasion of BTCC, which indicates the contribution of PRL-1 and PRL-3 to the growth and development of BTCC. To our surprise, we also found that PRL-1 was expressed in normal bladder epithelia in a low or moderate level, which was similar to other literature reported. However, there was some difference from these literatures, which reported PRL-1 being expressed at weak to moderate levels in the urothelia, and with the strongest expression occurring in the intermediate and basal cell layers. Our study exhibited that the PRL-1 protein was expressed strongly in all urothelia. The difference probably came from the different antibodies. However, our findings may indicate that the expression of PRL-1 protein must be related with the cells’ terminal differentiation.
The importance of PRL-3 in cancer diagnosis and prognosis has been manifested by its over-expression in the metastasis of colorectal cancer compared with no-metastatic tumor or normal colorectal epithelia [
4,
7]. Liver metastasis from other cancers such as the pancreas, esophagus, or stomach cancer did not express high levels of PRL-3, suggesting the specificity for colorectal cancer [
18]. The over-expression of PRL-3 in colorectal cancer metastasis was associated with gene amplification in a small subset of cases [
7]. Normal Chinese hamster ovary cells transfected with PRL-3 showed increased mobility and invasiveness, and these effects were markedly reduced when a catalytically inactive mutant was used [
5]. Furthermore, ectopic cells expressing PRL-3 were able to form experimental metastasis in mice and this effect was dependent on the catalytic activity of the phosphatase [
19,
20]. These results indicated that PRL-3 could be an important factor contributing to the invasive, metastatic properties of cancer cells. To discuss the probable relation of PRLs with invasion and metastasis, we approached the relationship between the deep invasive group and the superficial invasion group. Our data reveals the up-regulated expression of PRL-1 and PRL-3 in the deep invasive group compared with that in the superficial invasion group, indicating the close implication in response to invasion and metastasis of BTCC. The present study also proves that there are changes in the PRL-1 protein which parallel the degree of pathological differentiation, suggesting that the PRL-1 protein might be involved in the malignant change of BTCC.
Comparison of the expression of protein and mRNA of PRL-1 with that of PRL-3 in BTCC indicates a clearly positive correlation between PRL-1 and PRL-3 in both protein and mRNA, although there was no significant difference between their expressions. Recently, many studies showed that inhibiting the expression of PRLs could obviously inhibit the growth, invasion and metastasis of tumors by the technology of RNA interference or using the inhibitors of PRLs.
To sum up, PRL-1 and PRL-3 might play an important role in BTCC. The enhanced expression of PRL-1 and PRL-3 parallels the pathological differentiation and invasive ability of tumor cells, indicating a new biomarker and a potential new treatment target for BTCC.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(178KB)
