Virchows Arch (2014) 464:717–724 DOI 10.1007/s00428-014-1578-6

ORIGINAL ARTICLE

Loss of nuclear prothymosin-α expression is associated with disease progression in human superficial bladder cancer Yuh-Shyan Tsai & Yeong-Chin Jou & Chun-Liang Tung & Chang-Te Lin & Cheng-Huang Shen & Syue-Yi Chen & Hsin-Tzu Tsai & Chen-Li Lai & Chao-Liang Wu & Tzong-Shin Tzai

Received: 6 August 2013 / Revised: 22 January 2014 / Accepted: 28 March 2014 / Published online: 15 April 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract In this paper, we report a study on the clinical relevance of prothymosin-α expression and its correlation with intratumoral Foxp3 + and CD8 + lymphocytes (Foxp3+TIL and CD8+TIL) in bladder cancer patients. We used immunohistochemical staining for prothymosin-α, Foxp3, and CD8 on 101 tumor specimens harvested by endoscopic resection. The results were correlated with clinicopathological variables and clinical outcome in bladder cancer patients, particularly in 73 patients with superficial disease, using the log-rank test and Cox proportional hazard model. Overall, of the tumors, 30 % were negative, 34 % showed nuclear, and 37 % showed cytoplasmic prothymosin-α expression. Foxp3+TILs were detected in 11 % of patients Both Yeong-Chin Jou and Yuh-Shyan Tsai contributed equally to this work. Electronic supplementary material The online version of this article (doi:10.1007/s00428-014-1578-6) contains supplementary material, which is available to authorized users. Y.
(nonnuclear vs. nuclear, p=0.096). Patients with a history of urothelial carcinoma have a higher frequency of nonnuclear prothymosin-α expression than those without (p=0.016, chisquare test). By univariate and multivariate analyses of cases with superficial disease, grade and stage were identified as independent predictors for recurrence-free survival (p=0.016 and 0.016, respectively). Higher stage and nonnuclear prothymosin-α expression independently predict shorter progression-free survival (p=0.006 and 0.043, respectively). The presence of Foxp3+TILs was significantly associated with disease progression by univariate analysis (p=0.022), but not by multivariate analysis (p=0.147). In vitro assays showed that J82 cells which express ectopically nuclear prothymosin-α exhibit higher growth rate and secrete less TGF-β1 than those with cytoplasmic expression or control cells. Altogether, prothymosin-α expression is a determinant of disease progression in superficial bladder cancer. Foxp3+TILs tend to be found more often in bladder cancer with nonnuclear prothymosin-α expression. Future study is required to unravel their interaction. Keywords Bladder neoplasms . Thymosin . Foxp3 . Lymphocyte . Prognosis

Introduction Bladder cancer is the fourth most common cancer in men and the ninth most common cancer in women in the USA [1]. Approximately 70–80 % of bladder cancers are superficial at initial diagnosis, and these are associated with activation of the receptor tyrosine kinase (RTK)-Ras pathway. Most superficial tumors (about 70 %) recur and about 15 % progress to muscle-invasive disease, despite adjuvant intravesical chemotherapy or immunotherapy [2]. In contrast, 20–30 % of bladder cancers are muscle invasive, advanced, or metastatic and

718

harbor structural and functional defects in the tumor suppressors p53 and/or the retinoblastoma protein (RB) [2]. Molecular pathways involved in these cancers, which include RAS, fibroblast growth factor receptor (FGFR), p53, epidermal growth factor receptor (EGFR) family, vascular endothelial growth factor (VEGF), epithelial-to-mesenchymal transition, and cyclooxygenase-II (COXII), are the subject of intense investigation, as they play a pivotal role in tumor recurrence and progression of bladder cancer [2]. To date, a biomarker for early detection of bladder cancer and preoperative differential diagnosis has not yet been identified [3–5]. Despite advances in the understanding of the tumor microenvironment, relatively little is known about tumor immunology, which contrasts with the abundance of knowledge regarding tumor cell biology. As bladder cancer is the only solid tumor, other than melanoma, for which immunotherapy confers a therapeutic benefit [6, 7], it is worthwhile to investigate the significance of immunologically important molecules in bladder cancer, which might interact with those factors or pathways that are critical in urothelial tumorigenesis. Prothymosin-α (PTMA) is a small acid protein that plays an important role in several biological functions, such as cell proliferation [8], apoptosis [9], and immunomodulation. In prior studies published by other researchers, a predominantly punctuated nuclear distribution of PTMA in T24 bladder cancer cells and other cell types in vitro under confocal microscopy was reported [10, 11], suggesting that PTMA is involved in transcription and is associated with RNA synthesis and/or processing [10]. An immunohistochemical study showed nuclear localization of PTMA in both malignant and normal gastric epithelial cells [12], as well as in the nuclei of human prostate cancer cells [11]. Cytoplasmic PTMA inhibits apoptosome formation, while alpha-(trichloromethyl)-4pyridineethanol (PETCM), an activator of caspase-3, relieves PTMA and induces the inhibition of apoptosome formation [9]. PTMA binds to and is internalized by peripheral blood mononuclear cells (PBMCs) and induces nonspecific as well as tumor-specific responses in cytotoxic T lymphocytes (CTLs) in a mouse model [13, 14]. Interestingly, a PTMA immunoreactive carboxyl-terminal decapeptide TKKQKTDEDD, containing the nuclear localization signal (NLS), is able to stimulate lymphocytes and induce dendritic cell maturation, which is thought to be important for immunomodulation [15]. Our previous studies have shown that plasma thymosin-α1 level is a potential marker to differentiate between renal tumors (renal pelvis tumor vs. renal cell carcinoma) [16], and urinary PTMA level might be a potential detection marker for bladder cancer [17]. In a study of a small cohort of 91 patients with urothelial carcinoma of upper urinary tract, we demonstrated that patients with cytoplasmic PTMA-expressing tumors are more likely to recur after primary nephroureterectomy than those with nuclear PTMA

Virchows Arch (2014) 464:717–724

expression [18]. Nonetheless, it is unclear whether PTMA expression influences intratumoral immunity. Thus, the aim of this study is to explore the prognostic value of PTMA expression in bladder cancer and its association with intratumoral CD8+ or Foxp3+ lymphocytes (CD8+TIL or Foxp3+TIL).

Methods and materials Patient population and study samples The study was undertaken with the approval and institutional oversight of the Institutional Review Board for the Protection of Human Subjects at both Chia-Yi Christian Hospital (IRB101014) and the National Cheng Kung University Hospital (ER-95-49). Formalin-fixed paraffin-embedded tumor specimens from 101 patients with bladder cancer receiving transurethral resection of bladder tumor (TURBT) were retrospectively retrieved from the archives of the National Cheng Kung University Hospital (Tainan, Taiwan; n=67) and Chia-Yi Christian Hospital (Chia-Yi, Taiwan; n=34). All patients who had had TURBT and had been followed up for at least 1 year were included in the study. Tumors were staged according to the 2007 TNM classification and graded using the 2004 WHO classification. All patients received the same follow-up regularly, according to the modified NCCN Clinical Practice Guidelines in Oncology. All patients with a superficial tumor received regular follow-up after intravesical therapy, including cystoscopy and surveillance by imaging of the upper urinary tract. The regimens for intravesical therapy included either 40 mg epirubicin, 30 mg mitomycin-C weekly for 8 weeks, or 81 mg Bacillus Calmette-Guérin (BCG) vaccine for 6 weeks. Once the patient was diagnosed with muscle-invasive disease, either radical or partial cystectomy was suggested, and surgery was followed by adjuvant systemic chemotherapy with or without radiotherapy. Immunohistochemistry (IHC) Serial 5-μm sections were cut for either hematoxylin and eosin (H & E) staining or IHC. Sections were stained for PTMA, CD8, or Foxp3 immunoreactivity according to previou sly rep or ted m eth od s [ 1 9, 2 0] . B r i e f l y, a f t e r deparaffinization and rehydration, tissue sections were autoclaved in citrate buffer (pH 6.0) at 120 °C for 15 min, treated with 3 % H2O2 in methanol for 25 min, and then treated with skimmed milk in phosphate buffered saline for 30 min. Nonspecific background staining was minimized by preincubating the array with 0.3 % bovine serum albumin (BSA) in 0.1 M Tris-buffered saline, pH 8.0, for 1 h. Slides were incubated either with primary anti-Foxp3 antibody (clone 236/E7; ABCAM, Cambridge, MA, USA; dilution,

Virchows Arch (2014) 464:717–724

1:200) for 45 min, anti-CD8 monoclonal antibody (DAKO, Glostrup, Denmark; dilution, 1:200) for 1 h, or anti-PTMA monoclonal antibody (2F11, Alexis Biochemicals, Lausen, Switzerland; dilution, 1:4,800) for 2 h at room temperature, respectively. The immunoreactivity was visualized with a BioGenex IHC kit (BioGenex Laboratories Inc., San Ramon, CA), and the slides were then counterstained with hematoxylin for microscopic examination. Samples were analyzed blindly by one pathologist (Tung CL). The evaluation methods were similar to or modified from previous studies [18, 19, 21]. Tumors with less than 5 % staining cells were considered negative for PTMA expression. The others were classified as nuclear expression (positive staining in the cell nuclei with or without cytoplasmic staining) or cytoplasmic expression (positive staining in the cytoplasm without any nuclear staining) [18]. The average number of CD8+TILs was manually determined in ten random 0.0328mm2 digital images captured under high power (×320). All counts were repeated three times by the same pathologist, and the average of the repeat counts was used for statistical analysis [19]. As for Foxp3+TILs, any tumor specimens exhibiting more than five immunoreactive lymphocytes within one high power field were thought to represent the presence of Foxp3+TILs [21]. Cells, viral transfection, cell proliferation, and TGF-β1 ELISA assays A human bladder cancer cell line (J82, ATCC® Number: HTB-1™) was cultured in DME medium supplemented with 10 % fetal bovine serum (FBS), 2 mM L-glutamine, and 50 μg/ml gentamicin. Once cell growth reached near 80– 90 % confluence, cells were subcultured or trypsinized for subsequent experiments. The full-fragment and NLS-deleted human PTMA gene were cloned into a lentiviral vector (pWPXL-enhancer-WTPTMA and pWPXL-enhancer-dNLSPTMA). After co-transfection with pMD2.G vector into 293 cells, the supernatants containing the lentiviruses were collected and stored for future experiments. Thereafter, J82 cells were infected with lentivirus carrying the indicated gene, and then three J82 transfectants were collected by cell sorting. After growing on coverslips, each J82 transfectant was individually fixed with formaldehyde fixative solution for PTMA immunohistochemical staining. Also, cell pellets were collected for confirmation of protein expression using UREA-PAGE assay [22]. For the in vitro cell proliferation assay, cells were seeded in a 96-well plate at 5,000 cells per well and cultivated overnight in 2 % FBS-containing medium. When control wells were near 80–90 % confluence, cell viability was determined with an MTT assay. In brief, 3-[4,5-dimethylthiazol-2yl]-2,5-diphenyltetrazolium bromide (Sigma, St. Louis, MO, USA) was added to the cells to a final concentration of 0.05 % and incubated 4 h at 37 °C. The media were then collected for

719

TGF-β1 ELISA assay. To the cells, 150 μl of dimethyl sulfoxide (DMSO) was added and the plates were incubated for 3 min. The optical density (O.D.) of each well was determined using a microplate reader at a wavelength of 570 nm. The percentage of cell viability is calculated by the O.D. value of treated cells normalized with the O.D. value of control cells. The growth index was calculated by the formula: (%48 h or %72 h / %24 h). The measurement of the TGF-β1 protein was determined by antibody-capture bioassay with a TGF-β1 kit (R & D Systems, MN, USA) according to the manufacturer’s protocols. Statistical analysis Correlation of PTMA expression (nuclear and cytoplasmic) with clinicopathological data was examined. Clinical outcome of all patients or patients with superficial disease included tumor recurrence in the urinary bladder after complete TURBT and tumor progression into and beyond muscle layers. Recurrence-free survival (RFS) and progression-free survival (PFS) were calculated from complete TURBT to the date of the first documented tumor recurrence in the urinary bladder or tumor progression into muscle-invasive disease or more, respectively. Statistical analysis was performed using Statistical Package for Social Sciences, version 12.0, software (SPSS). The correlation between PTMA localization, the number of CD8+TILs, and the presence of Foxp3+TILs and clinicopathological factors was analyzed with the chi-square test. Correlations of PTMA expression pattern and other clinicopathological variables with disease recurrence or progression were analyzed with Kaplan-Meier plots, the logrank test, and the multivariate Cox regression model. All p values reported are two sided and considered significant if p<0.05.

Results Clinicopathological correlates of PTMA expression in human bladder cancer Our case series comprised of 65 male and 36 female patients (mean age 68 years). Forty-two patients had a previous history of urothelial carcinoma of the urinary bladder or upper urinary tract and 22 patients were on regular hemodialysis for endstage renal disease (ESRD). In 71 cases, multiple tumors had been found and four cases were associated with carcinoma in situ (CIS). As NLS-intact PTMA enters into the nucleus to exert its main biological functions, such as RNA processing or transcription, we considered three expression patterns: negative, cytoplasmic only, and nuclear (including nuclear only and

720

Virchows Arch (2014) 464:717–724

nuclear with cytoplasmic PTMA expression). PTMA expression patterns are illustrated in Fig. 1a–d. Of all bladder tumors, 30 (30 %) were negative for PTMA, 10 (10 %) showed nuclear expression, 37 (37 %) showed cytoplasmic, and 24 (24 %) showed mixed nuclear and cytoplasmic expression. Table 1 shows the correlation of PTMA expression, according to these patterns, with clinicopathological parameters. Of the six clinical factors (age, gender, history of urothelial carcinoma, ESRD on dialysis, tumor multiplicity, and associated CIS), only a history of urothelial cell carcinoma was significantly associated with the PTMA expression pattern (negative vs. cytoplasmic vs. nuclear, p=0.027; nonnuclear vs. nuclear, p=0.016, chi-square test), indicating that recurrent bladder tumors more often are PTMA negative or show cytoplasmic expression. Neither tumor grade nor tumor stage correlated with the PTMA expression pattern (all p values >0.05, chisquare test) (Table 1). CD8+ and Foxp3+ lymphocytes were detected in the tumor tissue (Fig. 1e, f). The median number of CD8+TILs detected was 4.4 (25 and 75 % percentile, 0.6 and 15.8, respectively). Foxp3+TILs were detected in 11 patients. PTMA expression patterns did not correlate with the numbers of CD8+TILs or the presence of Foxp3+TILs (p values >0.05, Table 1). Analysis of the findings on the 73 superficial bladder tumors showed similar results (Supplementary Table 1). Fig. 1 Immunohistochemical staining for prothymosin-α expression and intratumoral Foxp3+ and CD8+ lymphocytes. a Negative prothymosin-α expression; b cytoplasmic expression of prothymosin-α protein; c nuclear expression of prothymosin-α protein; d both nuclear and cytoplasmic expressions of prothymosin-α protein; e intratumoral CD8+ lymphocytes; f intratumoral Foxp3+ lymphocytes (arrows) (×320)

Prognostic significance of PTMA, Foxp3+TILs, and CD8+TILs in superficial bladder cancer Since the standard therapy for muscle-invasive tumors is not TURBT, we further analyzed the prognostic value of PTMA expression pattern, the presence of Foxp3 + TILs and CD8+TILs in the 73 superficial tumors. Of these, 43 (47 %) received post-TURBT intravesical therapy (27 epirubicin, 4 BCG, and 3 mitomycin-C). During follow-up (median, 48 months), 36 patients (49 %) recurred, 21(29 %) progressed, and 18 (25 %) died. By urivariate analysis, tumor stage was significantly associated with RFS [HR, 2.39; 95 % confidence interval (CI), 1.31–4.37; p=0.005], as well as tumor grade (HR, 5.26; 95 % CI, 1.64–16.7; p=0.005). Neither intravesical therapy nor PTMA expression nor the presence of Foxp3+TILs or CD8+TILs was significantly predictive for tumor recurrence (all p values, >0.05). Multivariate analysis showed that both tumor stage and grade are independently prognostic for tumor recurrence (HR, 2.13; 95 % CI, 1.15–3.98 for tumor stage and HR, 4.35; 95 % CI, 1.32–4.3 for tumor grade, respectively) (Table 2). By univariate analysis, tumor stage, PTMA expression patterns, the presence of Foxp3+TILs and intravesical therapy

a

b

c

d

e

f

Virchows Arch (2014) 464:717–724

721

Table 1 Clinicopathological correlates of PTMA expression in 101 human bladder urothelial tumors Characteristics

Total Age, mean ± SD (years) Gender Male Female History of UC No Yes ESRD on dialysis No Yes Multiplicity Single Multiple Associated CIS Yes No Tumor grade Low High Tumor stage Ta T1 T2 T3 at least Average CD8+TILs Median 25, 75 % percentile Foxp3+TILs Negative Positive

No. of patients (%)

PTMA localization

pa

pb

0.288

0.159

Negative

C

N

101 68.0±11.6

30 70.4±7.8

37 68.3±13.4

34 65.7±12.0

65 (64.3) 36 (35.6)

20 10

21 16

24 10

0.455

0.351

59 (58.4) 42 (41.6)

16 14

17 20

26 8

0.027

0.016

79 (79.2) 22 (21.8)

23 7

26 11

30 4

0.181

0.125

30 (29.7) 71 (70.3)

7 23

9 28

14 20

0.198

0.072

4 97

0 30

3 34

1 33

0.222

0.708

11 (10.9) 90 (89.1)

3 27

2 35

6 28

0.250

0.170

37 (36.6) 36 (35.6) 25 (24.8) 3 (3.0)

10 10 8 2

15 12 9 1

12 14 8 0

0.787

0.567

4.4 0.6, 15.8

6.5 1.3, 17.0

2.8 1.3, 17.0

5.7 0.6, 18.1

0.953

0.894

90 (89.1) 11 (10.9)

27 3

30 7

33 1

0.096

0.093

UC urothelial carcinoma, CIS carcinoma in situ, TIL tumor-infiltrating lymphocytes a

One hundred one patients with bladder tumors, analyzed with chi-square test or Fisher’s exact test

b

One hundred one patients with nonnuclear PTMA vs. nuclear expression, analyzed with chi-square test or Fisher’s exact test

were significantly associated with PFS (Table 2). Applying the rule of “ten outcomes per variable,” we randomly chose two variables, significant in the univariate analysis, for multivariate analysis. The results showed that only tumor stage and PTMA expression are independent predictors for disease progression [HR (95 % CI), 3.97 (1.48–10.6) and 3.26 (1.04– 10.2); p values, 0.006 and 0.043, respectively). Considering that intravesical therapy might have had an effect on outcome, all the calculations were re-run after stratification of patients according to whether or not they had received intravesical therapy. The results showed a similar effect (Fig. 2).

TGF-β1 secretion in PTMA-transfected J82 cells To provide experimental evidence in vitro for the biological significance of cytoplasmic PTMA expression, we established three J82-derived cell lines, which ectopically express wildtype PTMA, NLS-deleted PTMA, or the empty vector (J82WTPTMA-GFP, J82-dNLSPTMA-GFP, and J82-GFP). Using immunohistochemical staining (Fig. 3a) and UREA-PAGE assays (Fig. 3b), we confirmed the nuclear and cytoplasmic expression patterns. J82-WTPTMA-GFP cells exhibited a higher growth rate than the other two cell lines (Fig. 3c). In

722

Virchows Arch (2014) 464:717–724

Table 2 Univariate and multivariate analyses of variables associated with survival in 73 superficial bladder cancer patients Variablesa

Univariate analysis Gender History of UC ESRD Multiplicity Intravesical therapy Grade Stage PTMA expression pattern CD8+TILs Foxp3+TILs Multivariate analysisb Grade Stage PTMA expression pattern

RFS

PFS

HR (95 % CI)

p

HR (95 % CI)

p

0.66 (0.36–1.21) 1.56 (0.87–2.78) 0.83 (0.41–1.67) 1.42 (0.74–2.69) 0.89 (0.49–1.59) 5.26 (1.64–16.7) 2.39 (1.31–4.37) 1.05 (0.58–1.91) 1.35 (0.90–2.02) 1.60 (0.71–3.60)

0.179 0.137 0.596 0.291 0.686 0.005 0.005 0.878 0.151 0.262

1.29 (0.54–3.06) 1.75 (0.74–4.13) 0.84 (0.31–2.30) 4.08 (0.95–17.5) 0.36 (0.13–0.97) 2.30 (0.81–5.12) 2.93 (1.16–7.42) 3.48 (1.17–10.4) 1.95 (0.08–4.74) 3.27 (1.19–8.98)

0.563 0.201 0.733 0.059 0.044 0.133 0.023 0.025 0.142 0.022

4.35 (1.32–14.3) 2.13 (1.15–3.96)

0.016 0.016

– 3.97 (1.48–10.6)

– 0.006

0.94 (0.50–1.78)

0.847

3.26 (1.04–10.2)

0.043

HR hazard ratio, 95 % CI 95 % confidence interval, TIL tumor-infiltrating lymphocytes a

Variables include gender (male vs. female), previous history of urothelial carcinoma tumor site (no vs. yes), ESRD on dialysis (no vs. yes), morphology (papillary vs. nonpapillary), multiplicity (single vs. multiple), intravesical therapy (no vs. yes), tumor grade (low vs. high), pathological staging (Ta vs. T1), and PTMA expression pattern [nonnuclear (negative and cytoplasmic) vs. nuclear], and CD8+ TILs (0–8 vs. >8)

b

Cox proportional hazard ratio

contrast, J82-WTPTMA-GFP cells secreted less TGF-β1 than the other two cell lines (Fig. 3d).

Discussion PTMA protein can be detected in human blood, urine, and cancer cells [16, 23, 24] and activates lymphocytes and dendritic cells [15, 25]. It is a small, acid, nuclear polypeptide which was found recently to play an important role in the pathogenesis of pulmonary emphysema and polycystic kidney disease through

Fig. 2 Proportion of disease progression-free survival of 73 patients with superficial bladder cancer according to the prothymosin-α expression pattern (a) or the presence of Foxp3+TILs (b). PFS progression-free survival, Foxp3+TIL intratumoral Foxp3+ lymphocytes

inhibition of the association of nuclear factor-κB and histone deacetylases with histones [26, 27]. The nuclear form of PTMA protein contains a C-terminal nuclear localization signal (NLS) and is released in a nonvesicular manner in neuronal cells under necrosis-induced stress. The cytoplasmic form of PTMA is deficient of the NLS [28]. During early apoptosis, nuclear PTMA is cleaved primarily at D99 by activated capsase-3 [9, 29, 30]. The cleaved PTMA polypeptide lacks the NLS and may relocate to the cytoplasm or cell membrane [9]. Skopeliti et al. reported that C-terminal PTMA peptides containing the NLS as well as full-length PTMA protein enhance cell proliferation of

Virchows Arch (2014) 464:717–724

723

Fig. 3 Effect of ectopic prothymosin-α expression on cell proliferation and TGF-β1 secretion in J82 cells. Confirmation of ectopic expression of wild-type PTMA or NLS-deleted PTMA using a immunohistochemical staining or b urea-PAGE assay. c In vitro cell proliferation assay showed that J82-WTPTMA-GFP cells exhibit a higher growth rate than the other two cell lines, J82-dNLSPTMAGFP and J82-GFP. d J82WTPTMA-GFP cells secreted less TGF-β1 than the other two cell lines. GFP green fluorescence protein, PTMA prothymosin-α, WT wild type, dNLS deleted nuclear leading signal, tRNA transfer ribonucleic acid, O.D. optical density. *p<0.05; **p<0.01; ***p<0.001

peripheral blood mononuclear cells (PBMC) and concluded that its immunologically active site is located at the C-terminus [31]. Several studies have reported that the tumor microenvironment of bladder cancer is immunosuppressive [32]. Loskog et al. reported that patients with bladder cancer have an immunosuppressive regulatory profile, including the presence of intratumoral regulatory T lymphocytes, TGF-β1, IL-4, and unresponsive peripheral blood T cells [33]. Foxp3 is a transcription factor required for inducing immunosuppressive capacity in regulatory T lymphocytes, which is associated with TGF-β1 signaling. We found that superficial bladder cancer patients of which the tumor showed loss of nuclear PTMA expression (including negative or cytoplasmic only cases) had a shorter progression-free survival than those with nuclear PTMA expression. We hypothesize that nonnuclear PTMA expression reflects the loss of the C-terminal immunologically active site and failure to elicit an antitumor immune response, resulting in worse outcome. Our in vitro study showed that nuclear PTMA-expressing J82 cells secrete less TGF-β1 than those with cytoplasmic expression only, even though they showed a higher growth rate. We conclude that in superficial bladder cancer, loss of nuclear PTMA expression is associated with progressive disease, which might be related to an insufficient immune response in which TGF-β1 signaling might play a role. These findings warrant further investigations of the role of PTMA and TGF-β1 signaling in the tumor microenvironment.

Our study has some limitations. Firstly, the small number of patients and subject events did not allow simultaneous calculation of all significant variables, as the study was statistically underpowered. Therefore, a role of PTMA in urothelial cell carcinogenesis should be validated on larger numbers of patients and samples and with longer follow-up. Secondly, in immunohistochemical studies, the antibody used or applied cutoff values can have a significant impact on the results of statistical analysis. Thirdly, the tissue specimens were obtained through TURBT, in which the number of intratumoral lymphocytes might have been underestimated and the incidence of the associated CIS might be underreported.

Conclusion Our results show that nuclear PTMA expression is significantly more often lost in bladder cancer patients with a positive history of urothelial carcinoma. No associations were found between PTMA expression and other clinicopathological parameters. Loss of nuclear PTMA expression is an independent prognostic factor for disease progression in superficial bladder cancer, which might be through loss of an antitumor immune response in which TGF-β1 might play a role. The PTMA expression pattern was neither associated with the presence of CD8+TILs nor of Foxp3+TILs, although Foxp3+TILs tended

724

to be found more often in bladder cancer with loss of nuclear PTMA expression. Acknowledgments This study was supported by the National Science Council of Taiwan (Grants NSC 97-2314-B-006-051-MY3 and 98-2314-B006-051-MY2) and Chia-Yi Christian Hospital (Grants 97-07 and 98-07).

Virchows Arch (2014) 464:717–724

16.

17.

Conflict of interest We declare that we have no conflict of interest. 18.

References 19. 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Global cancer statistics. CA Cancer J Clin 61:69–90 2. Wu XR (2005) Urothelial tumorigenesis: a tale of divergent pathways. Nat Rev Cancer 5:713–725 3. Roupret M, Zigeuner R, Palou J, Boehle A, Kaasinen E, Sylvester R et al (2011) European guidelines for the diagnosis and management of upper urinary tract urothelial cell carcinomas: 2011 update. Eur Urol 59:584–594 4. Stenzl A, Cowan NC, De Santis M, Kuczyk MA, Merseburger AS, Ribal MJ et al (2011) Treatment of muscle-invasive and metastatic bladder cancer: update of the EAU guidelines. Eur Urol 59:1009–1018 5. Babjuk M, Oosterlinck W, Sylvester R, Kaasinen E, Bohle A, PalouRedorta J et al (2011) EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder, the 2011 update. Eur Urol 59: 997–1008 6. Liakou CI, Narayanan S, Ng Tang D, Logothetis CJ, Sharma P (2007) Focus on TILs: prognostic significance of tumor infiltrating lymphocytes in human bladder cancer. Cancer Immun 7:10 7. Simpson AJ, Caballero OL, Jungbluth A, Chen YT, Old LJ (2005) Cancer/testis antigens, gametogenesis and cancer. Nat Rev Cancer 5: 615–625 8. Wu CL, Shiau AL, Lin CS (1997) Prothymosin alpha promotes cell proliferation in NIH3T3 cells. Life Sci 61:2091–2101 9. Jiang X, Kim HE, Shu H, Zhao Y, Zhang H, Kofron J et al (2003) Distinctive roles of PHAP proteins and prothymosin-alpha in a death regulatory pathway. Science 299:223–226 10. Vareli K, Frangou-Lazaridis M, van der Kraan I, Tsolas O, van Driel R (2000) Nuclear distribution of prothymosin alpha and parathymosin: evidence that prothymosin alpha is associated with RNA synthesis processing and parathymosin with early DNA replication. Exp Cell Res 257:152–161 11. Klimentzou P, Drougou A, Fehrenbacher B, Schaller M, Voelter W, Barbatis C et al (2008) Immunocytological and preliminary immunohistochemical studies of prothymosin alpha, a human cancerassociated polypeptide, with a well-characterized polyclonal antibody. J Histochem Cytochem 56:1023–1031 12. Wang M, Pan JY, Song GR, Chen HB, An LJ, Qu SX (2007) Altered expression of estrogen receptor alpha and beta in advanced gastric adenocarcinoma: correlation with prothymosin alpha and clinicopathological parameters. Eur J Surg Oncol 33:195–201 13. Baxevanis CN, Gritzapis AD, Spanakos G, Tsitsilonis OE, Papamichail M (1995) Induction of tumor-specific T lymphocyte responses in vivo by prothymosin alpha. Cancer Immunol Immunother 40:410–418 14. Cordero OJ, Sarandeses C, Nogueira M (1995) Prothymosin alpha receptors on lymphocytes. J Interferon Cytokine Res 15:731–737 15. Skopeliti M, Iconomidou VA, Derhovanessian E, Pawelec G, Voelter W, Kalbacher H et al (2009) Prothymosin alpha immunoactive carboxyl-terminal peptide TKKQKTDEDD stimulates lymphocyte reactions, induces dendritic cell maturation and adopts a beta-sheet

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32. 33.

conformation in a sequence-specific manner. Mol Immunol 46:784– 792 Jou YC, Tsai YS, Hsieh HY, Chen SY, Tsai HT, Chen KJ et al (2013) Plasma thymosin-alpha1 level as a potential biomarker in urothelial and renal cell carcinoma. Urol Oncol 31(8):1806–1811. doi:10.1016/ j.urolonc.2012.03.011 Tzai TS, Tsai YS, Shiau AL, Wu CL, Shieh GS, Tsai HT (2006) Urine prothymosin-alpha as novel tumor marker for detection and follow-up of bladder cancer. Urology 67:294–299 Jou YC, Tung CL, Tsai YS, Shen CH, Syue-Yi C, Shiau AL et al (2009) Prognostic relevance of prothymosin-alpha expression in human upper urinary tract transitional cell carcinoma. Urology 74: 951–957 Lin CT, Tung CL, Tsai YS, Shen CH, Jou YC, Yu MT et al (2012) Prognostic relevance of preoperative circulating CD8-positive lymphocytes in the urinary bladder recurrence of urothelial carcinoma. Urol Oncol 30:680–687 Tsai YS, Jou YC, Lee GF, Chen YC, Shiau AL, Tsai HT et al (2009) Aberrant prothymosin-alpha expression in human bladder cancer. Urology 73:188–192 Winerdal ME, Marits P, Winerdal M, Hasan M, Rosenblatt R, Tolf A et al (2011) Foxp3 and survival in urinary bladder cancer. BJU Int 108:1672–1678 Lukashev DE, Chichkova NV, Vartapetian AB (1999) Multiple tRNA attachment sites in prothymosin alpha. FEBS Lett 451:118– 124 Panneerselvam C, Haritos AA, Caldarella J, Horecker BL (1987) Prothymosin alpha in human blood. Proc Natl Acad Sci U S A 84: 4465–4469 Dominguez F, Magdalena C, Cancio E, Roson E, Paredes J, Loidi L et al (1993) Tissue concentrations of prothymosin alpha: a novel proliferation index of primary breast cancer. Eur J Cancer 29A: 893–897 Cordero OJ, Sarandeses CS, Lopez JL, Cancio E, Regueiro BJ, Nogueira M (1991) Prothymosin alpha enhances interleukin 2 receptor expression in normal human T-lymphocytes. Int J Immunopharmacol 13:1059–1065 Su BH, Tseng YL, Shieh GS, Chen YC, Shiang YC, Wu P et al (2013) Prothymosin alpha overexpression contributes to the development of pulmonary emphysema. Nat Commun 4:1906 Li KJ, Shiau AL, Chiou YY, Yo YT, Wu CL (2005) Transgenic overexpression of prothymosin alpha induces development of polycystic kidney disease. Kidney Int 67:1710–1722 Matsunaga H, Ueda H (2010) Stress-induced non-vesicular release of prothymosin-alpha initiated by an interaction with S100A13, and its blockade by caspase-3 cleavage. Cell Death Differ 17:1760–1772 Enkemann SA, Wang RH, Trumbore MW, Berger SL (2000) Functional discontinuities in prothymosin alpha caused by caspase cleavage in apoptotic cells. J Cell Physiol 182:256–268 Evstafieva AG, Belov GA, Kalkum M, Chichkova NV, Bogdanov AA, Agol VI et al (2000) Prothymosin alpha fragmentation in apoptosis. FEBS Lett 467:150–154 Skopeliti M, Voutsas IF, Klimentzou P, Tsiatas ML, Beck A, Bamias A et al (2006) The immunologically active site of prothymosin alpha is located at the carboxy-terminus of the polypeptide. Evaluation of its in vitro effects in cancer patients. Cancer Immunol Immunother 55:1247–1257 Olsson CA, Rao CN, Menzoian JO, Byrd WE (1972) Immunologic unreactivity in bladder cancer patients. J Urol 107:607–609 Loskog A, Ninalga C, Paul-Wetterberg G, de la Torre M, Malmstrom PU, Totterman TH (2007) Human bladder carcinoma is dominated by T-regulatory cells and Th1 inhibitory cytokines. J Urol 177:353–358