Cancer Research AACR Membership Protein Translation and Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vimalachandran, D.
Right arrow Articles by Costello, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vimalachandran, D.
Right arrow Articles by Costello, E.
[Cancer Research 65, 3218-3225, April 15, 2005]
© 2005 American Association for Cancer Research

Cell and Tumor Biology

High Nuclear S100A6 (Calcyclin) Is Significantly Associated with Poor Survival in Pancreatic Cancer Patients

Dale Vimalachandran1, William Greenhalf1, Christopher Thompson1, Jutta Lüttges4, Wendy Prime2, Fiona Campbell3, Andrew Dodson3, Richard Watson1, Tatjana Crnogorac-Jurcevic5, Nicholas Lemoine5, John Neoptolemos1 and Eithne Costello1

1 Division of Surgery and Oncology and 2 Cancer Tissue Bank Research Centre, Department of Pathology, University of Liverpool; 3 Department of Pathology, Royal Liverpool University Hospital, Liverpool, United Kingdom; 4 Department of Pathology, University of Schleswig-Holstein, Kiel, Germany; and 5 Cancer Research UK Clinical Centre, Queen Mary's School of Medicine at Barts and The London, London, United Kingdom

Requests for reprints: Eithne Costello, Division of Surgery and Oncology, University of Liverpool, Royal University Hospital, Daubly Street, Liverpool, Merseyside, United Kingdom. Phone: 44-151-706-4178; E-mail: ecostell{at}liv.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Recent studies have reported elevated levels of S100A6 in pancreatic ductal adenocarcinoma cells. Here, we describe a detailed analysis of S100A6 expression in benign (n = 32), malignant (n = 60), and premalignant pancreatic ductal cells [96 pancreatic intraepithelial neoplasias (PanIN) from 46 patients]. S100A6 staining was more intense in malignant cells than in benign cells (P = 0.0001). In malignant cells, staining was higher in the nucleus than in the cytoplasm (P = 0.003). Univariate analysis revealed a significant decrease in survival time for patients with high levels of nuclear (P = 0.01) but not cytoplasmic (P = 0.20) S100A6. No evidence was found for an association between nuclear S100A6 expression and other variables, including gender, age at surgery, tumor size or grade, nodal metastases, resection margin, vascular invasion, perineural invasion, p53 or Smad4 levels (both linked to survival in previous studies), or the p65 subunit of nuclear factor-{kappa}B (a potential regulator of S100A6). Although nodal metastases and resection margin involvement were also associated with poor survival (P = 0.06 in both cases), multivariate analysis suggests that nuclear S100A6 is a significant independent indicator of survival (P = 0.003). Whereas PanIN 1a lesions showed a general absence of S100A6 staining, there was a progressive increase in the proportion of positively stained PanINs with increasing PanIN grade. In particular, we observed an increase in the frequency and intensity of nuclear staining. Our results suggest that up-regulation of S100A6 is an early event in pancreatic cancer development and that elevated levels of nuclear S100A6 may affect clinical outcome.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Pancreatic ductal adenocarcinoma is a serious health problem. Worldwide, an estimated 213,000 people died of the disease in the year 2000, and in the Western world particularly, it is a leading cause of cancer death (1, 2). Although resection and adjuvant chemotherapy offers a significant survival advantage (3), only up to 9% of patients undergo potentially curative surgery (4) and most patients are dead from the disease within 6 to 8 months of diagnosis (1, 2). The molecular characterization of pancreatic cancer is relatively advanced with genetic alterations documented for several specific genes, including K-ras, p53, p16INK4a, and Smad4 (for review, see ref. 5). Furthermore, the classification of precancerous ductal lesions into a standard system (6) has resulted in the development of a pathologic progression model akin to the colorectal adenoma-carcinoma model. Nevertheless, additional progress is dependent on elaborating, still further, the biological nature of this disease.

Both gene expression profiling experiments (7, 8) and a proteomic study (9) of pancreatic cancer have led to the observation that S100A6 is overexpressed in pancreatic cancer cells. Our studies (8, 9) revealed pancreatic tumor cells to have both a greater frequency and a higher intensity of S100A6 expression compared with normal pancreatic ductal epithelial cells. S100A6 is a low molecular weight calcium binding protein, the crystal structures of which, in the calcium-free and calcium-bound states, are consistent with it being a calcium sensor (10). It belongs to a family of proteins of the EF-hand type known as the S100 protein family. Members of this family can interact with effector proteins, thereby regulating enzyme activities, dynamics of cytoskeleton constituents, cell growth and differentiation, and calcium homeostasis (for review, see ref. 11). Whereas the role of S100A6 has yet to be clearly defined, it has been implicated in several cellular processes. It was first identified as a protein whose expression increased in human fibroblasts in response to growth factor stimulation (12). More recently, it has been shown that expression of antisense S100A6 inhibited proliferation in fibroblasts (13). Its expression has also been shown to increase in the renal collecting duct in response to vasopressin, suggesting a role in transepithelial ion transport (14). It has been implicated in the calcium-dependent secretion of insulin from pancreatic ß cells (15) and in the ubiquitination pathway regulating the proteolytic degradation of ß-catenin (16).

Despite the seemingly disparate roles proposed for S100A6, one consistent feature is its up-regulation in a variety of tumors (12, 17–19). In this context, it has been linked to metastasis (17, 18, 20). In a study of the metastatic ability of cutaneous melanoma cell lines in nude mice, elevated S100A6 correlated with high metastatic ability (20). Recently, two independent groups found S100A6 expression to be particularly high at the invasive margins of colorectal adenocarcinoma, suggesting a role in tumor progression and invasion (17, 18). The regulation of S100A6 gene expression is poorly understood, although roles for the transcription factors activator protein-1, upstream transcription factor-1, and nuclear factor-{kappa}B (NF-{kappa}B) have been proposed (21–23).

In this study, we have undertaken a detailed analysis of S100A6 cytoplasmic and nuclear expression in benign, malignant, and premalignant pancreatic ductal cells. We correlated S100A6 overexpression with molecular and clinicopathologic variables, including outcome. In addition, because RelA, the p65 subunit of NF-{kappa}B, has been reported to be constitutively activated in pancreatic cancer (24), we also sought to determine whether S100A6 overexpression was associated with nuclear RelA.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell lines and Western blotting. The human pancreatic adenocarcinoma cell line Panc-1 was obtained from the American Type Culture Collection (Rockville, MD) and maintained in RPMI 1640 supplemented with 10% FCS, 2 mmol/L L-glutamine, 2,500 IU/mL penicillin, and 5 µg/mL streptomycin (all from Sigma, Poole, United Kingdom) at 37°C in a humidified atmosphere of 5% CO2. For Western analysis, Panc-1 cells were collected and resuspended in buffer containing 50 mmol/L HEPES (pH 7.5), 150 mmol/L NaCl, 10% glycerol, 0.1% Triton X-100, and 0.2 mmol/L CaCl2 supplemented with Complete protease inhibitors (Roche Diagnostics, Indianapolis, IN) and sonicated on ice. Protein quantification was by the Bradford method, and specified quantities of protein were electrophoretically separated by SDS-PAGE on 10% to 20% Tris-HCl gradient gels (Bio-Rad Laboratories Ltd.,Hemel Hempstead, Hertfordshire, United Kingdom). Separated proteins were electrophoretically transferred to Hybond nitrocellulose membranes (Amersham Biosciences, Chalfont St. Giles, Buckinghamshire, United Kingdom), and membranes were blocked in 5% milk in PBS for 30 minutes before application of a rabbit anti-S100A6 polyclonal antibody (DAKOCytomation, Cambridgeshire, United Kingdom). Membranes were washed thrice in TBS containing 0.1% Tween 20 for 10 minutes followed by incubation with a horseradish peroxidase–conjugated goat anti-rabbit secondary antibody (DAKO). After extensive washing in TBS containing 0.1% Tween 20, protein-antibody complexes conjugated with peroxidase were visualized with Western Lighting chemiluminescence reagent (Perkin-Elmer Life Sciences, Boston, MA).

Generation and purification of recombinant S100A6. A recombinant S100A6 protein containing a 6x His COOH-terminal tag was generated as follows: reverse transcription PCR was done on RNA isolated from Panc-1 cells using the primers 5'-ATGGCATGCCCCTGGATCAG-3' and 5'-TCAGCCCTTGAGGGCTTCAT-3'. The resulting amplicon was cloned into a pBAD-TOPO vector (Invitrogen, Carlsbad, CA) and transformed into Escherichia coli. Optimal expression of the His-tagged S100A6 in E. coli was achieved following incubation of cultures with 0.02% (w/v) L(+)-arabinose (Sigma) for 4 hours. For extraction of protein, bacteria were harvested, resuspended in buffer containing 20 mmol/L phosphate, 0.5 mol/L NaCl, and 20 mmol/L imidazole (pH 7.5) along with Complete EDTA-free protease inhibitors, and sonicated on ice for four periods of 20 seconds. Bacterial supernatants were loaded onto HisTrap HP nickel columns (Amersham Biosciences), and contaminating proteins were removed by extensive washing of the nickel matrix with binding buffer containing 100 mmol/L imidazole. S100A6-His was eluted in the presence of 300 mmol/L imidazole, which was subsequently removed from protein preparations using HiTrap desalting columns (Amersham Biosciences). The purity of each S100A6-His preparation was confirmed using SDS-PAGE on 10% to 20% Tris/tricine gels (Bio-Rad) and colloidal Coomassie blue G-250 staining as described previously (9).

Pancreatic cancer tissue microarray analysis. A rabbit anti-human S100A6 antibody, raised against purified recombinant human S100A6 monomer expressed in E. coli, and a murine anti-human p53 monoclonal antibody (clone DO-7) were obtained from DAKOCytomation (Cambridgeshire, United Kingdom). Anti-RelA, a peptide antibody raised against the nuclear localization signal of p65, was obtained from Roche (formerly Boehringer-Mannheim). A monoclonal antibody raised against Smad4 (clone B8) was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). These were used to perform immunohistochemical analysis on a pancreatic cancer tissue microarray generated at the Cancer Tissue Bank Research Centre, University of Liverpool (Liverpool, United Kingdom) using the method described by Kononen et al. (25). The pancreatic cancer tissue microarray contained matched duplicate nonmalignant and malignant specimens from 60 patients treated at the Royal Liverpool University Hospital (Liverpool, United Kingdom) between 1994 and 2003. Nonmalignant specimens from 32 of the 60 patients contained sufficient benign ductal cells for evaluation. In addition, a separate microarray containing 22 benign ducts, 35 pancreatic intraepithelial neoplasia (PanIN) 1a, 26 PanIN 1b, 22 PanIN 2, and 13 PanIN 3 lesions was also analyzed.

Immunohistochemistry was done as described previously (9). Briefly, microarray sections 5 µm in thickness were deparaffinized in xylene and then rehydrated through alcohol to distilled water. Antigen retrieval, as recommended by the suppliers of the S100A6 antibody, was done by pressure cooking the slides in 10 mmol/L EDTA (pH 7.4) for 3 minutes. Immunohistochemical staining was done using an automatic staining system (Autostainer, DAKOCytomation). Separate slides were incubated for 40 minutes with anti-S100A6 antibody (0.25 µg/mL), anti-Smad4 antibody (2.0 µg/mL), anti-p53 antibody (3.35 µg/mL), and anti-RelA (10 µg/mL). Slides were then rinsed in TBS-Tween 20, and the antibody localization was visualized by incubating sections with a horseradish peroxidase–conjugated labeled polymer for 30 minutes followed by diaminobenzidine (DAKOCytomation) for 10 minutes. Slides were counterstained with hematoxylin and dehydrated with 100% ethanol and xylene, and coverslips mounted with distrene, plasticizer, and xylene mountant (VWR International Ltd.,Poole, United Kingdom). To achieve anti-S100A6 antibody neutralization, the rabbit anti-S100A6 polyclonal antibody was mixed with an excess of the purified recombinant S100A6-His protein and incubated overnight at 4°C. This, instead of the primary antibody, was then used as a negative control.

Scoring of the pancreatic cancer tissue microarray sections was done by a specialist histopathologist (F.C.). The information recorded included the subcellular location of S100A6 staining (nuclear and cytoplasmic), intensity of staining [grade 0 (negative), 1 (weak), 2 (moderate), and 3 (strong)], and percentage of cells demonstrating positive immunoreactivity (0 for no staining, 1 point for <20%, 2 points for 20-50%, and 3 points for >50% of cells). The total score for each compartment was obtained as the product of intensity and extent (percentage of cells stained) of staining. Negative or weakly positive cases had a score of 0 to 3, moderately positive cases had a score of 4 to 6, and strongly positive cases had a score of >6. S100A6 expression was than divided into two groups, low expression (negative or weakly stained) and high expression (strong or moderately stained). Assessment of the staining of p53 and Smad4 was done in a similar fashion. For p53, a score of ≥4 was defined as representing p53 overexpression (cutoff obtained by correlating p53 protein expression with known mutation status from tissue and juice analysis in a subset of patients; data not shown). For Smad4, negative cases were defined as the complete absence of expression as described previously (26). The PanIN microarray was scored in a similar fashion by a separate specialist pathologist (J.L.). Subcellular compartments were scored individually, and the staining intensity was graded as negative, weak, or moderate. Due to the small size of PanIN lesions, the proportion of cells staining was not recorded.

Statistical analyses. Clinicopathologic variables were extracted from histopathologic reports and included patients' age at surgery, gender, tumor size, tumor grade, lymph node status, and resection margin status. Reporting of vascular or perineural invasion was not recorded systematically in all reports. To obtain associations between S100A6 expression and clinicopathologic variables, data were cross-tabulated and Fisher's two-sided exact test or {chi}2 test was done as appropriate. Immunohistochemical scores were compared using the Wilcoxon rank-sum test. To evaluate the effect of S100A6 expression on patient survival, life tables were constructed from survival data and Kaplan-Meier curves were plotted. Overall survival was measured from date of initial surgery to date of death, counting death from any cause as the end point or the last date of information as the end point if no event was documented. For Kaplan-Meier curves, comparisons between groups were done using the log-rank test. A Cox proportional hazards regression model was created to assess the prognostic value of S100A6 expression in multivariate analyses. All statistical analyses were done using Statview version 5.01 (SAS Institute, Inc., Cary, NC). Significance was set at P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Clinicopathologic data. Patient and tumor characteristics are summarized in Table 1 (column 1). Univariate survival analysis was done on all pathologic and molecular variables. Of these, only the presence of nodal metastases (log-rank {chi}21 = 3.4; P = 0.06) and positive resection margins (log-rank {chi}21 = 3.45; P = 0.06) were associated with decreasing patient survival.


View this table:
[in this window]
[in a new window]

  		Table 1. Nuclear S100A6 expression and association with clinicopathologic variables

 
Specificity of S100A6 antibody staining. The anti-S100A6/calcyclin polyclonal antibody (DAKOCytomation) was assessed to determine its suitability for analysis of S100A6 expression in pancreatic tissues. Western blot analysis of a whole cell lysate from Panc-1 pancreatic ductal adenocarcinoma cells (Fig. 1A) revealed the detection of a prominent protein band corresponding to the reported size of S100A6 on SDS-PAGE gels (27). We reported previously that Western analysis on a pancreas tumor protein extract separated in two dimensions revealed a single spot, correctly positioned for S100A6 (9). To establish that the staining of sections detected in our immunohistochemical experiments was specific to the anti-S100A6 antibody, a recombinant S100A6 protein containing a His tag was generated and used to compete-out S100A6-specific antibodies before immunoassays. The bacterially produced, His-tagged S100A6 protein was purified to homogeneity (Fig. 1B) and, before immunostaining, was mixed with the anti-S100A6 polyclonal antibody or, as a control, sodium phosphate buffer. Immunostaining with preabsorbed antibody effectively prevented staining (Fig. 1C).



View larger version (49K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Specificity of S100A6 specific antibody binding. A, Western analysis of a pancreatic cancer cell line (Panc-1) using polyclonal S100A6 antibody reveals a single protein band at 6 kDa. B, purity of His-tagged S100A6 recombinant protein confirmed by Coomassie blue G-250 staining following two-dimensional PAGE separation. C, incubation of pancreatic cancer tissue sections with His-tagged S100A6 recombinant protein and antibody mixture (i) results in abolition of immunostaining when compared with the control (ii; antibody and PBS).

 
Pattern of S100A6 staining in pancreatic ductal adenocarcinoma. The pancreatic samples contained in the pancreatic cancer tissue microarray enabled the analysis of benign ducts from 32 patients and malignant samples from 60 patients. The overall intensity of S100A6 staining was higher in malignant ducts than in benign ducts. The median score in the nuclear compartment of malignant specimens was 6 (interquartile range, 4-9) compared with 2 (interquartile range, 1-3) in benign ducts (P < 0.0001, Wilcoxon rank-sum test). Similarly, the median cytoplasmic score in malignant ducts was 3 (interquartile range, 3-6) compared with 1.5 (interquartile range, 0-1.5) in benign ducts (P < 0.0001). The difference between nuclear and cytoplasmic staining in malignant ducts was significant (P = 0.003, Wilcoxon rank-sum test).

For analysis of malignant specimens, we categorized the immunostaining pattern in malignant specimens into two categories, low (which included cases with a score of 0-3) and high (which included moderately positive cases with a score of 4-6 and strongly positive cases with a score of >6). Figure 2 shows representative examples of each type of staining. Forty-seven of 60 (78%) patients exhibited strong or moderate immunoreactivity in the nuclear compartment, whereas only 26 of 60 (43%) patients exhibited strong or moderate immunoreactivity in the cytoplasmic compartment. Weak staining was also occasionally observed in stromal cells and inflammatory cells.



View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Immunohistochemical staining demonstrating S100A6 expression in malignant ductal epithelium. A, S100A6-negative expression; B, weak expression; C, moderate expression; D, strong expression, where the most intense staining was observed in the nuclear compartment. Some additional staining was also noted in stromal cells and inflammatory cells. Magnification, x40.

 
High nuclear S100A6 expression is associated with short survival. Only four patients were still alive at the time of the study. We examined the relationship between both nuclear and cytoplasmic S100A6 expression and survival. The median survival of patients with tumors showing low nuclear S100A6 staining was 28.6 months [95% confidence interval (95% CI), 15.2-42.9 months]. By contrast, the median survival of patients with tumors containing high nuclear S100A6 levels was significantly shorter at 9.7 months (95% CI, 6.5-11.9 months; log-rank {chi}21 = 5.75; P = 0.01; Fig. 3A). This association, however, did not exist for cytoplasmic expression of S100A6. The median survival of patients with tumors shows low cytoplasmic S100A6 staining of 13.1 months (95% CI, 6.5-25.2 months) versus 9.7 months (95% CI, 6.9-14.1 months) for patients with high cytoplasmic expression (log-rank {chi}21 = 1.64; P = 0.2; Fig. 3B).



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Kaplan-Meier survival curves for 60 patients with pancreatic cancer. A, survival curves for patients according to their nuclear S100A6 expression. (i) Nuclear S100A6 low expression (weak or negative immunostaining) and (ii) nuclear S100A6 high expression (strong or moderate immunostaining). Test statistics: log-rank test (1 df, P = 0.01). B, survival curves for patients according to their cytoplasmic S100A6 expression. (i) Cytoplasmic S100A6 low expression (weak or negative immunostaining) and (ii) nuclear S100A6 high expression (strong or moderate immunostaining). Test statistics: log-rank test (1 df, P = 0.2).

 
Correlation with clinicopathologic and molecular variables. Our survival analysis showed a clear link between nuclear S100A6 expression and survival. To address whether there were confounding variables, we examined the relationship between nuclear S100A6 expression and other recognized pathologic prognostic factors (tumor size, grade, lymph node metastases, resection margin status, vascular invasion, and perineural invasion) and molecular factors (p53 and Smad4). Additional variables were also evaluated (Table 1).

For the variables of gender, age at surgery, tumor grade, tumor size, nodal status, and resection margin status, data were available for all 60 patients. There was no significant association with these variables and nuclear S100A6 expression (Table 1). Data for vascular or perineural invasion were available for only 46 and 44 patients, respectively. There was no evidence for an association between nuclear S100A6 expression and any of these variables (Table 1). High cytoplasmic S100A6 expression, however, was observed in 18 of 29 (62%) of patients with vascular invasion compared with 3 of 17 (18%) of patients without vascular invasion (P = 0.005, Fisher's exact test).

Immunohistochemical analysis of tumor samples from all 60 patients (Table 1) in the study revealed overexpression of the p53 protein in 33% (20 of 60). Of these, 75% (15 of 20 patients) had high nuclear S100A6. A similar proportion of the patients with low or undetectable p53 (32 of 40, 80%) also had high nuclear S100A6. Thus, there was no evidence for a link between overexpression of p53 and high nuclear S100A6. Smad4 expression (Table 1) was determined in 53 patients; 7 patients were excluded from the microarray analysis due to insufficient specimen material. Consistent with previous work (28), we found that 47% (25 of 53) of patients lacked Smad4 protein expression. Of these 25 patients, 17 (68%) had high nuclear S100A6 expression. Of the 28 Smad4-expressing patients, 24 (86%) had high nuclear S100A6 expression. S100A6 nuclear expression was therefore observed in a relatively similar proportion of Smad4-positive and Smad4-negative patients.

Multivariate survival analysis. A multivariate Cox proportional hazards model using those variables associated with survival in our study (nodal metastases, positive resection margins, and nuclear S100A6 expression) revealed only S100A6 expression to be a significant independent prognostic indicator (log-rank {chi}21 = 8.79; P = 0.003; Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2. Univariate and multivariate analyses of significant risk factors in 60 patients with pancreatic cancer

 
Nuclear RelA is not strongly associated with high S100A6 expression. To determine whether there was an association between high S100A6 expression and activated RelA, we carried out immunohistochemistry using an antibody that recognizes an epitope of RelA, which overlaps its nuclear localization signal. Immunohistochemistry was done on the pancreatic cancer tissue microarray; however, only 40 of 60 malignant cases and 27 of 32 benign cases were available for analysis. RelA (Fig. 4A) was detected in the cytoplasm of benign ducts from 26 of 27 (96%) patients. However, nuclear RelA staining was observed in a minority only (7 of 27, 26%) of these benign ducts. Cytoplasmic but not nuclear staining of acinar cells was also detected with this antibody (data not shown).



View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Immunohistochemical staining using anti-RelA peptide antibody, which is raised against the nuclear localization signal of p 65. A, benign ductal epithelium demonstrating weak cytoplasmic staining and negative nuclear staining; B, malignant ductal epithelium demonstrating positive immunostaining, particularly in the nuclei of cells (arrowheads). Magnification, x40.

 
By contrast, nuclear RelA staining (Fig. 4B) was detected in 57% (23 of 40) of cancer samples. Thus, nuclear RelA was present in a significantly higher proportion of tumor samples than benign ducts (P = 0.01). The staining in many cancer samples was reciprocal (i.e., RelA was detected either in the cytoplasm or in the nucleus; Fig. 4B).

Of the 40 malignant cases assessed for RelA, corresponding nuclear S100A6 expression data were available for 38 cases. We examined whether patients with nuclear RelA had increased S100A6 expression compared with patients lacking nuclear RelA (Table 3). Of 21 patients with detectable nuclear RelA, 16 (76%) expressed high nuclear S100A6. Of 17 patients who had undetectable nuclear RelA, 12 (70%) had high nuclear S100A6 expression. Similarly, there was no significant association between cytoplasmic S100A6 and RelA. Thus, the presence of nuclear RelA was not apparently associated with S100A6 expression. Of note, however, a slightly higher proportion of patients with nuclear RelA also had nuclear S100A6 (Table 3). In addition, the median score for the intensity of nuclear S100A6 staining in patients with nuclear RelA was 6 (interquartile range, 1.75-9) compared with 4 (interquartile range, 0-6) in patients lacking nuclear RelA, although this was not statistically significant (P = 0.19). The median score for cytoplasmic S100A6 staining was the same (3) for each group.


View this table:
[in this window]
[in a new window]

  		Table 3. Association between nuclear S100A6 expression and nuclear RelA (p65) expression

 
S100A6 expression in PanIN lesions. Given our observation that nuclear S100A6 was associated with poor survival, we wished to establish whether overexpression of S100A6 was an early or late event in the development of pancreatic cancer. This was examined by carrying out immunohistochemical analysis on a separate microarray containing 22 benign ducts, 35 PanIN 1a, 26 PanIN 1b, 22 PanIN 2, and 13 PanIN 3 lesions. Examples of immunostained PanIN lesions are shown in Fig. 5. The percentage of each type of PanIN lesion exhibiting S100A6 expression (including cytoplasmic and/or nuclear S100A6) is presented in Fig. 6A. We observed S100A6 expression in a small percentage only of PanIN 1a lesions. By contrast, 62% of PanIN 1b lesions were positively stained for S100A6 and this increased with increasing PanIN grade (Fig. 6A). Nuclear staining was more variable than cytoplasmic staining and reached higher levels. When nuclear staining was categorized into negative, weak, or moderate and the percentage of each type of PanIN lesion displaying these types of nuclear staining was plotted, it was clear that the frequency and the intensity of nuclear staining increased with increasing PanIN grade (P < 0.0001; {chi}2 test; Fig. 6B).



View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Immunodetection of S100A6 in PanIN lesions. PanIN 1a (A) with undetectable S100A6, PanIN 1b (B), PanIN 2 (C), and PanIN 3 (D) with detectable S100A6 expression.

 


View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. A, S100A6 immunoreactivity in PanIN lesions. Lesions were scored for S100A6 immunoreactivity (including cytoplasmic and nuclear staining), and the percentage of S100A6 positive lesions in each distinct PanIN group was plotted (*, P < 0.05, {chi}2 test). B, nuclear staining of PanIN lesions. Nuclear S100A6 staining was graded as weak, negative, or moderate. For each type of PanIN, the percentage of lesions with negative, weak, or moderately stained nuclei was plotted (*, P < 0.05, {chi}2 test).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study, we analyzed the expression of S100A6 in benign and premalignant pancreatic ductal cells as well as in pancreatic cancer cells. Our study revealed the majority of pancreatic cancer cases to have high levels of nuclear S100A6. Furthermore, those patients with high levels of nuclear S100A6 had a significantly reduced survival time compared with patients who have low or undetectable nuclear S100A6. That high nuclear S100A6 levels correlate with poor survival in pancreatic cancer patients was not anticipated, as this relationship has not, to our knowledge, been shown for any other cancer type. Other S100 family members, by contrast, have been linked to poor outcome in certain cancers. Gongoll et al. (29) showed that S100A4 expression in colorectal cancer tumors was accompanied by a significantly decreased patient survival time compared with S100A4-negative tumors. Likewise, psoriasin (S100A7) expression in estrogen receptor–negative invasive breast cancer was associated with a poor clinical outcome (30).

It is unclear why high nuclear S100A6 levels in pancreatic tumor cells are associated with poor survival. The expression of the protein has been shown to increase in response to growth factor stimulation and it may be necessary for proliferation (12, 13). Thus, S100A6 may be a marker of cell growth in pancreatic cancer cells or may be promoting the growth of pancreatic cancer cells. Alternatively, S100A6 overexpression has been linked to metastasis in two tumor types (17, 18, 20). If S100A6 overexpression was promoting metastasis in pancreatic cancer, then this could account for its association with poor prognosis. Indeed, in this study, we observed an association between cytoplasmic S100A6 expression and vascular invasion. Due to the aggressive nature of pancreatic cancer, almost 90% of patients are unsuitable for resection of their cancer on presentation (4). As a consequence, tissue specimens for studies such as this one are extremely scarce. The number of patients analyzed in the present study is small, and although we interrogated our data set in an attempt to link nuclear S100A6 with variables that might clarify its relationship to poor survival, no obvious link to such a variable emerged. It is possible, however, that, with a larger patient set, relationships between S100A6 and other patient variables may emerge that may help explain why nuclear S100A6 is associated with poor survival.

Given that the transcription factor NF-{kappa}B is constitutively activated in pancreatic cancer (24) and that this factor (in particular, NF-{kappa}B-containing p65/RelA) has also been proposed as a potential regulator of S100A6 gene expression (23), we considered it important to examine whether S100A6 expression was a downstream event of NF-{kappa}B activation and thus a surrogate marker for this. Although the incidence of nuclear RelA (activated RelA) was significantly higher in tumor samples (P = 0.01) than in nonmalignant ductal cells, a similar proportion of the nuclear-positive and nuclear-negative RelA patients had high S100A6 expression. Thus, a direct link between nuclear RelA and S100A6 expression was not observed. It may be that nuclear RelA does not necessarily reflect constitutive activation of NF-{kappa}B in these samples and that a more direct measure of NF-{kappa}B activation is required. Alternatively, NF-{kappa}B family members, other than RelA, may be involved. It is also possible that the levels of S100A6 protein are controlled by mechanisms, either transcriptional or post-transcriptional that do not involve NF-{kappa}B.

Having observed the poor survival of patients with high nuclear S100A6 expression, we subsequently analyzed precursor lesions of pancreatic cancer. The genetic progression of PanIN lesions is well documented, with the rate of mutations increasing through the higher grades. Based on the study of such mutations, the major step in progression to malignancy occurs between grades 1 and 2/3 (31). We found that cytoplasmic S100A6 expression increases sharply between PanIN 1a and 1b. However, nuclear S100A6 levels increase between PanIN 1b and 2 and that expression continues to increase through to PanIN 3. Thus, the expression of S100A6 in PanIN lesions would seem to parallel the general rate of genetic mutations in these lesions.

In summary, we have shown that nuclear S100A6 expression is associated with poor survival in pancreatic cancer patients. This finding was not clearly attributable to any other investigated patient variable. It is impossible to say whether high nuclear S100A6 is an underlying cause of tumor aggressiveness or whether it is simply a marker of it. However, what is clear is that S100A6 overexpression occurs early in the development of pancreatic cancer and may contribute to the etiology of the disease. Importantly, it is the detection of early lesions at an asymptomatic stage that may allow curative resection of pancreatic cancer and offer a greater chance of cure. In this context, S100A6 may have a future role as a biomarker for pancreatic cancer. Additionally, however, the ability to distinguish patients based on prognosis is important and is likely to have increasingly significant implications for the treatment of pancreatic cancer patients in the future. S100A6 may represent a useful prognostic marker, although a larger study will be required to determine if this is the case.


   Acknowledgments
 
Grant support: Cancer Research UK, Royal College of Surgeons of England, and European Union (QLG1-CT-2002-01196).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Kathrin Richter for excellent technical assistance.

Received 12/ 2/04. Revised 2/10/05. Accepted 2/14/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000. The global picture. Eur J Cancer 2001;37:4–66.[CrossRef]
  2. Jemal A, Murray T, Samuels A, et al. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5–26.[Abstract/Free Full Text]
  3. Neoptolemos JP, Stocken DD, Friess H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 2004;350:1200–10.[Abstract/Free Full Text]
  4. Sener S, Frengen A, Menck HR, Winchester DP. Pancreatic cancer: a report of treatment and survival trends for 100,313 patients diagnosed from 1985-1995, using the National Cancer Database. J Am Coll Surg 1999;189:1–7.[CrossRef][Medline]
  5. Bardeesy N, DePinho RA. Pancreatic cancer biology and genetics. Nat Rev Cancer 2002;2:897–909.[CrossRef][Medline]
  6. Hruban RH, Adsay NV, Albores-Saavedra J, et al. Pancreatic intraepithelial neoplasia: a new nomenclature and classification system for pancreatic duct lesions. Am J Surg Pathol 2001;25:579–86.[CrossRef][Medline]
  7. Logsdon CD, Simeone DM, Binkley C, et al. Molecular profiling of pancreatic adenocarcinoma and chronic pancreatitis identifies multiple genes differentially regulated in pancreatic cancer. Cancer Res 2003;63:2649–57.[Abstract/Free Full Text]
  8. Crnogorac-Jurcevic T, Missiaglia E, Blaveri E, et al. Molecular alterations in pancreatic carcinoma: expression profiling shows that dysregulated expression of S100 genes is highly prevalent. J Pathol 2003;201:63–74.[CrossRef][Medline]
  9. Shekouh A, Thompson CC, Prime W, et al. Development and application of laser capture microdissection combined with two-dimensional electrophoresis for the discovery of differentially regulated proteins in pancreatic ductal adenocarcinoma. Proteomics 2003;3:1988–2001.[CrossRef][Medline]
  10. Otterbein LR, Kordowska J, Witte-Hoffman C, et al. Crystal structures of S100A6 in the Ca(2+) free and Ca(2+) bound states: the calcium sensor mechanism of S100 proteins revealed at atomic resolution. Structure (Camb) 2002;10:557–67.
  11. Donato R. Intracellular and extracellular roles of S100 proteins. Microsc Res Tech 2003;60:540–51.[CrossRef][Medline]
  12. Calabretta B, Kaezmarek L, Mars W, et al. Cell cycle specific genes differentially expressed in human leukaemias. Proc Natl Acad Sci U S A 1985;13:4463–7.
  13. Breen EC, Tang K. Calcyclin (S100A6) regulates pulmonary fibroblast proliferation, morphology, and cytoskeletal organisation in vitro. J Cell Biochem 2003;88:848–54.[CrossRef][Medline]
  14. Courtis-Coutry N, Le Meollic C, Boulkroun S, et al. Calcyclin is an early vasopressin-induced gene in the renal collecting duct. Role in the long term regulation of ion transport. J Biol Chem 2002;277:25728–34.[Abstract/Free Full Text]
  15. Okazaki K, Niki I, Iino S, et al. A role of calcyclin, a Ca(2+) binding protein, on the Ca(2+) dependant insulin release from the pancreatic ß cell. J Biol Chem 1994;269:6149–52.[Abstract/Free Full Text]
  16. Filipek A, Jastrzebska B, Nowotny M, Kuznicki J. CacyBP/SIP, a calcyclin and Siah-1 interacting protein, binds EF-hand proteins of the S100 family. J Biol Chem 2002;277:28848–52.[Abstract/Free Full Text]
  17. Stulik J, Osterreicher J, Koupilova K, et al. Differential expression of the Ca2+ binding S100A6 protein in normal, preneoplastic and neoplastic colon mucosa. Eur J Cancer 2000;36:1050–9.
  18. Komatsu K, Kobune-Fujiwara Y, Andoh A, Ishiguro S, et al. Increased expression of S100A6 at the invading fronts of the primary lesion and liver metastasis in patients with colorectal adenocarcinoma. Br J Cancer 2000;83:769–74.[CrossRef][Medline]
  19. Hsieh HL, Schafer BW, Sasaki S, Heizmann CW. Expression analysis of S100 proteins and RAGE in human tumours using tissue microarrays. Biochem Biophys Res Commun 2003;307:375–81.[CrossRef][Medline]
  20. Weterman MA, Stoopen GM, van Muijen GN, et al. Expression of calcylin in human melanoma cell lines correlates with metastatic behaviour in nude mice. Cancer Res 1992;52:1291–6.[Abstract/Free Full Text]
  21. Bottini F, Mazzocco K, Abbondi T, Tonini G. Identification of an AP-1-like sequence in the promoter region of calcylin, a S100 like gene. Enhancement of binding activity during retinoic acid induced neuroblastoma cell differentiation. Neurosci Lett 1994;181:35–8.[Medline]
  22. Lesniak W, Jezierska A, Kuznicki J. Upstream stimulatory factor is involved in the regulation of the human calcyclin (S100A6) gene. Biochim Biophys Acta 2000;1517:73–81.[Medline]
  23. Joo HJ, Kim JW, Lee Y, et al. Involvement of NF-{kappa}B in the regulation of S100A6 gene expression in human hepatoblastoma cell line HepG2. Biochem Biophys Res Commun 2003;307:274–80.[CrossRef][Medline]
  24. Wang W, Abbruzzese JL, Evans DB, et al. The nuclear factor-{kappa}B RelA transcription factor is constitutively activated in human pancreatic adenocarcinoma cells. Clin Cancer Res 1999;5:119–27.[Abstract/Free Full Text]
  25. Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high throughput molecular profiling of tumour specimens. Nat Med 1998;4:844–47.[CrossRef][Medline]
  26. Wilentz RE, Su GH, Dai JL, et al. Immunohistochemical labeling for dpc4 mirrors genetic status in pancreatic adenocarcinomas: a new marker of DPC4 inactivation. Am J Pathol 2000;156:37–43.[Abstract/Free Full Text]
  27. Komatsu K, Murata K, Kameyama M, et al. Expression of S100A6 and S100A4 in matched samples of human colorectal mucosa, primary colorectal adenocarcinomas and liver metastases. Oncology 2002;63:192–200.[CrossRef][Medline]
  28. Tascilar M, Skinner HG, Rosty C, et al. The SMAD4 protein and prognosis of pancreatic ductal adenocarcinoma. Clin Cancer Res 2001;7:4115–21.[Abstract/Free Full Text]
  29. Gongoll S, Peters G, Mengel M, et al. Prognostic significance of calcium binding protein S100A4 in colorectal cancer. Gastroenterology 2002;123:1478–84.[CrossRef][Medline]
  30. Emberley ED, Niu Y, Kliiewer EV, et al. Psoriasin (S100A7) expression is associated with poor outcome in oestrogen receptor negative invasive breast cancer. Clin Cancer Res 2003;9:2627–31.[Abstract/Free Full Text]
  31. Luttges J, Kloppel G. Precancerous conditions of pancreatic carcinoma. J Hepatobiliary Pancreat Surg 2000;7:568–74.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
S. Hu, M. Arellano, P. Boontheung, J. Wang, H. Zhou, J. Jiang, D. Elashoff, R. Wei, J. A. Loo, and D. T. Wong
Salivary Proteomics for Oral Cancer Biomarker Discovery
Clin. Cancer Res., October 1, 2008; 14(19): 6246 - 6252.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. ProteomicsHome page
L. M. Orre, M. Pernemalm, J. Lengqvist, R. Lewensohn, and J. Lehtio
Up-regulation, Modification, and Translocation of S100A6 Induced by Exposure to Ionizing Radiation Revealed by Proteomics Profiling
Mol. Cell. Proteomics, December 1, 2007; 6(12): 2122 - 2131.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
H. J. Whiteman, M. E. Weeks, S. E. Dowen, S. Barry, J. F. Timms, N. R. Lemoine, and T. Crnogorac-Jurcevic
The Role of S100P in the Invasion of Pancreatic Cancer Cells Is Mediated through Cytoskeletal Changes and Regulation of Cathepsin D
Cancer Res., September 15, 2007; 67(18): 8633 - 8642.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
K. Ohuchida, K. Mizumoto, J. Yu, H. Yamaguchi, H. Konomi, E. Nagai, K. Yamaguchi, M. Tsuneyoshi, and M. Tanaka
S100A6 Is Increased in a Stepwise Manner during Pancreatic Carcinogenesis: Clinical Value of Expression Analysis in 98 Pancreatic Juice Samples
Cancer Epidemiol. Biomarkers Prev., April 1, 2007; 16(4): 649 - 654.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
C C Thompson, F J Ashcroft, S Patel, G Saraga, D Vimalachandran, W Prime, F Campbell, A Dodson, R E Jenkins, N R Lemoine, et al.
Pancreatic cancer cells overexpress gelsolin family-capping proteins, which contribute to their cell motility
Gut, January 1, 2007; 56(1): 95 - 106.
[Abstract] [Full Text] [PDF]

Home page
 Int ImmunolHome page
R. Frances, J. R. Tumang, and T. L. Rothstein
Extreme skewing of annexin II and S100A6 expression identified by proteomic analysis of peritoneal B-1 cells
Int. Immunol., January 1, 2007; 19(1): 59 - 65.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
E. Giovannetti, V. Mey, S. Nannizzi, G. Pasqualetti, M. Del Tacca, and R. Danesi
Pharmacogenetics of anticancer drug sensitivity in pancreatic cancer.
Mol. Cancer Ther., June 1, 2006; 5(6): 1387 - 1395.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vimalachandran, D.
Right arrow Articles by Costello, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vimalachandran, D.
Right arrow Articles by Costello, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online