Expression of neutral endopeptidase (NEP/CD10) on pancreatic tumor cell lines, pancreatitis and pancreatic tumor tissues



Neutral endopeptidase (NEP/CD10) is a cell surface zinc metalloprotease cleaving peptide bounds on the amino terminus of hydrophobic amino acids and inactivating multiple physiologically active peptides. Loss or decrease in NEP/CD10 expression have been reported in many types of malignancies, but the role of NEP/CD10 in pancreatic carcinoma has not yet been identified. Using real-time RT-PCR, flow cytometry as well as immunohistochemistry, NEP/CD10 expression was quantified in both pancreatic carcinoma cell lines and in tumor specimens obtained from patients with primary pancreatic carcinomas. Three out of 8 pancreatic carcinoma cell lines exhibit heterogeneous NEP/CD10 expression levels: PATU-8988T expressed the highest NEP/CD10 levels, whereas HUP-T4 and HUP-T3 cells showed a moderate to low NEP/CD10 expression. NEP/CD10 immunoreactivity was found in 6 of 24 pancreatic ductal adenocarcinomas, but also in 3 of 6 tissues of patients with chronic pancreatitis. NEP/CD10 expression in pancreatic tumor lesions and cell lines was not associated with tumor grading and staging. Treatment of PATU-8988T cells with the histone deacetylase inhibitors sodium butyrate and valproic acid induced an increase of NEP/CD10 expression. This was accompanied by a reduced cell proliferation rate of PATU-8988T cells, which was increased by the addition of the enzyme activity inhibitors phosphoramidon and thiorphan. Thus, NEP/CD10 is differentially expressed in pancreatic tumors and might be involved in the proliferative activity of pancreatic cancer cells. However, further studies are needed to provide more detailed information of the role of NEP/CD10 under physiological and pathophysiological conditions of the pancreas. © 2007 Wiley-Liss, Inc.

Cancer of the exocrine pancreas accounts for 2–3% of all cancers, but is the fourth most frequent cause of cancer deaths.1 Pancreatic cancer is more common among males than females with the peak incidence occurring at age 60. The 5 years survival rate of patients with pancreatic cancer is less than 5%. Although the etiology of this disease remains unclear, cigarette smoking and alcohol abuse have been related with an increased incidence of pancreatic cancer.

Neutral endopeptidase (NEP/CD10, EC is a cell-surface metallopeptidase that is normally expressed by numerous tissues including prostate, kidney, breast, intestine, endometrium, adrenal glands and lung. This enzyme disintegrates peptide bonds on the amino terminus of hydrophobic amino acids and inactivates a variety of physiologically active peptides like atrial natriuretic factor, substance P, bradykinin, oxytocin, Leu- and Met-enkephalins, neurotensin, bombesin, calcitonin gene-related peptide, endothelin-1 and bombesin-like peptides.2, 3, 4, 5, 6, 7, 8, 9

NEP/CD10 reduces the local concentration of peptides available for receptor binding and signal transduction. The biological function of NEP/CD10 appears to be organ specific. In the lung, it modulates the effect of tachykinins, such as substance P that mediate inflammation.10 Recent studies showed a regulating function of NEP/CD10 in the kidney and vascular epithelium, and has a direct impact on the level of circulating atrial natriuretic factor.11 In the endometrium, it regulates endothelin-1 which causes vasoconstriction of endometrial arterioles during specific phases of the ovulatory cycle.12 NEP/CD10 has also been implicated in controlling cellular proliferation by hydrolyzing bombesin-like peptides, which are potent mitogens for fibroblasts and bronchial epithelial cells.13, 14 Decrease or loss of NEP/CD10 expression have been reported in many types of malignancies including renal cancer,15 invasive bladder cancer,16 poorly differentiated stomach cancer,17 small cell and nonsmall cell lung cancer,18 endometrial cancer19 and prostate cancer.20 Therefore, it is speculated that reduced NEP/CD10 promotes peptide-mediated proliferation by allowing an accumulation of higher peptide concentrations at the cell-surface which may facilitate the development or progression of neoplasia.21

Since only limited information exists about the NEP/CD10 expression in physiological and pathophysiological conditions of the pancreas,22, 23, 24 the main purpose of our study is the quantification of NEP/CD10 mRNA and protein in pancreatitis and in pancreatic cancer lesions of different staging, histotypes and grading. In addition to the constitutive NEP/CD10 expression, the regulation of NEP/CD10 by cytokines and modifiers of the histone acetylation as well as their effects on proliferation was investigated in pancreatic carcinoma cell lines.


BA, butyric acid; BeA, betulinic acid; EGF, epidermal growth factor; FACS, fluorescence-activated cell sorting; IFN, interferon; NEP/CD10, neutral endopeptidase; PE, phycoerythrin; PMA, phorbol myristate acid; RA, retinoic acid; TGF, transforming growth factor; TNF-α, tumor necrosis factor-α; VPA, valproic acid.

Material and methods


The cytokines tumor necrosis factor (TNF)-α, interferon (IFN)-β, interleukin (IL)-6 and transforming growth factor (TGF)-β1 were purchased from Strathmann Biotec AG (Hannover, Germany), phorbol myristate acid (PMA) and epidermal growth factor (EGF) from CALBIOCHEM (Darmstadt, Germany), sodium butyrate (BA) from MERCK (Darmstadt, Germany), betulinic acid (BeA) from Carl Roth GmbH & Co. (Karlsruhe, Germany), the phycoerythrin-(PE)-conjugated anti-CD10 monoclonal antibody (mAb) used for FACS analysis (H10a IgG1) from BD Biosciences (Heidelberg, Germany) and the mouse anti-CD10 mAb (clone 56C6) from Novocastra Laboratories (Newcastle, UK). Retinoic acid (RA), valproic acid (VPA) and phosphoramidon were obtained from Sigma-Aldrich (Steinheim, Germany), whereas Thiorphan was purchased from Fluka Chemie Gmbh (Steinheim, Germany).

Pancreatic cancer cell lines and culture conditions

The 8 pancreatic ductal adenocarcinoma cell lines used in this study PATU-8988T, PATU-8988S, PATU-8902, MIA PaCa-2, HUP-T3, HUP-T4, CAPAN-1 and CAPAN-2 were either obtained from the Deutsche Sammlung für Mikroorganismen und Zelllinien (DSMZ; Braunschweig, Germany) or the American Type Tissue Culture Collection (ATCC). Cells were cultured in a humidified atmosphere containing 5% CO2 at 37°C. Their characteristics and growth conditions are listed in Table I. All media were supplemented with penicillin–streptomycin 10,000 U/10,000 μg/ml (BIOCHROM AG, Berlin, Germany). Cells were passaged every 4–7 days using trypsin-EDTA, Invitrogen (Karlsruhe, Germany) and 1 × 105 cells were seeded onto 25 cm2 tissue culture flasks grown until confluency, and then processed for mRNA or protein quantification, respectively.

Table I. Characteristics and Biology of Pancreatic Carcinoma Cell Lines used in Our Study
Cell line/originDoubling time and morphologyCulture mediaOriginMetastasisDifferentiation status
PATU-8988T/DSMZAbout 22 hr, adherent cells growing in monolayersDMEM/F12 + 5% FCS + 5% horse serum + 2 mM L-glutamineLiver metastasis of a primary pancreatic adenocarcinomaNoWell
PATU-8988S/DSMZAbout 40–60 hr, adherent epitheloid cells growing in monolayersDMEM/F12 + 5% FCS + 5% horse serum + 2 mM L-glutamineLiver metastasis of a primary pancreatic adenocarcinoma (sister of PATU-T)Yes, in particular lungPoor
PATU-8902/DSMZAbout 25–40 hr, epithelial adherent cells growing in monolayersDMEM/F12 + 5% FCS + 5% horse serum + 2 mM L-glutaminePrimary ductal pancreatic adenocarcinoma (grade II)NoPoor
HUP-T3/DSMZAbout 38 hr, epitheloid cells growing adherent in monolayers and clustersMEM + 10% FCS + 1% nonessential aa + 1% Na-pyruvatePancreatic carcinoma from malignant ascitesNoWell
HUP-T4/DSMZAbout 38 hr, epitheloid cells growing adherent in monolayersMEM + 10% FCS + 1% nonessential aa + 1% Na-pyruvatePapillotubular pancreatic carcinoma from malignant ascitesNoPoor
CAPAN-1/DSMZAbout 50–100 hr, adherent fibroblastic semi-confluent monolayerRPMI 1640 + 15% FCS + 2 mM L-glutamineLiver metastasis of a pancreatic ductal adenocarcinomaNoWell
CAPAN-2/DSMZAbout 50–70 hr, adherent, epithelial-like cells growing in monolayerRPMI 1640 + 15% FCS + 2 mM L-glutaminePancreatic adenocarcinomaYesWell
MIA PaCa-2/ATCCAbout 40 hr, adherent cells growing in monolayersDMEM/F12 + 10% FCS + 2.5% horse serum + 1% Na-pyruvate + 4 mM L-glutaminePancreatic carcinomaYesWell

Cell culture treatment

Cells were grown in complete medium until 75% confluency, followed by a 24 hr serum free interval before stimulation for 24, 48, 72 hr with the following reagents: 0 (untreated); 50 ng/ml EGF, 10 ng/ml PMA, 1 ng/ml TGF-β1, 100 U/ml TNF-α, 100 U/ml IL-6, 100 U/ml IFN-β, 2.5–7.5 μg/ml BeA dissolved in DMSO, 1–4 mM BA, 2 mM VPA and 0.5–1.5 μM RA was carried out. The medium was changed every day, and after the appropriate stimulation time, cells were photographed and prepared for analysis of gene and protein expression.

RNA isolation and cDNA synthesis

Cells were lysed in 1 ml of TRIzol™ reagent (WKS, Frankfurt, Germany) and total cellular RNA was isolated from cells according to the manufacturer's instructions developed by Chomczynski and Sacchi.25 After DNase I treatment, the RNA integrity was confirmed by denaturing agarose gel electrophoresis, and the concentration of total cellular RNA was quantified by determination of optical density at 260 nm (OD 260). The first strand of DNA was synthesized according to the manufacturer's instructions using 500 ng total RNA, 50 pmol of random primer (Roche Diagnostics, Mannheim, Germany) and 0.5 μl (200 U/μl) Superscript™ II-RT (Invitrogen). The resulting complementary DNA (cDNA) was subjected either to conventional PCR amplification or quantitative RT-PCR.

CD10 standard preparation

For the construction of standard RNA, a composite primer was synthesized. Primer 1 (5′-GATTT AGGTG ACACT ATAGA ATACC TCCGA GAAAA GGTGG ACAA-3′) obtained a 5′-SP6 RNA polymerase binding site (underlined) followed by a NEP/CD10 specific sequence. Upon gel purification of the PCR product obtained with primer 1 and 2 (5′-TGAGTCCACCAGTCAACGAC-3′) using the QIA quick Gel Extraction Kit (Qiagen, Hilden, Germany), in vitro transcription from the SP6 promoter (Roche Diagnostics) was performed. The copy RNA was quantified at 260 nm and used as standard in the quantitative RT-PCR reaction.

Quantitative RT-PCR

For quantitation, 1 μl of the reverse transcriptase reaction mixture was added to 25 μl reaction mixture of the SYBR Green Quantitect kit (Qiagen) and 0.5 μM each of the primer 3 (5′-CCTCCGAGAAAAGGTGGACAA-3′) and primer 2. A negative control containing the complete Master Mix without DNA template was included. Samples of both standard cDNA and sample cDNA, respectively, were run in triplicates on the Rotor-Gene 2000 (Corbett Research, Sydney). Initial denaturation at 95°C for 15 min was followed by 40 cycles with denaturation at 95°C for 15 sec, annealing at 60°C for 30 sec and extension at 72°C for 30 sec. The fluorescence intensity of the double-strand specific SYBR Green, reflecting the amount of formed PCR product, was read after each extension step at 72°C. For verification of the PCR products melting curves were generated. Absolute RNA amounts were determined with the Rotor-Gene software version 4.6 in quantitation mode. Ten microliters of each PCR product were visualized after agarose gel electrophoresis and ethidium bromide staining.

Flow cytometry

Fluorescence-activated cell sorting (FACS) was performed to quantify the CD10 surface expression levels. Briefly, cells were incubated with the PE-conjugated anti-human mouse CD10 mAb IgG for 15 min, fixed with 2% paraformaldehyde for 10 min followed by 2 washings with PBS. FACS data were acquired on a FACS calibur (Beckton Dickinson, Heidelberg, Germany), 10,000 events were recorded and analyzed using the CELL Quest software (Becton Dickinson). Fluorescence data were expressed as relative mean fluorescence intensity (MFI) in percentages. The IgG isotype control Ab (Dianova- Immunotech, Hamburg, Germany) served as negative control.


Immunohistochemical analysis using the mouse anti-CD10 mAb (clone 56C6, Novocastra Laboratories, Newcastle, UK) was performed on collected tissues from 30 patients who underwent surgery for diagnostic or therapeutic purpose at the Martin-Luther University Halle-Wittenberg between 2003 and 2004. Data on histology, tumor size, nodal status and tumor stage were extracted from the pathological report at primary diagnosis (Table II). The local committee of medical ethics approved the use of human tissues and all patients gave their written consent.

Table II. Characteristics of the Pancreatic Tumor Lesions Analyzed
No. of lesions analyzedSexAgeGradingpTpNCD10 expression
  1. Lesions no. 1–24 represent samples of ductal adenocarcinoma, whereas lesions no. 25–30 represent samples obtained from patients with chronic pancreatitis. NEP/CD10 staining of lesions was performed as described in Material and Methods using an anti-CD10 mAb. +, positive for CD10 staining. −, negative for CD10 staining.

25M56Not applicable+
26M42Not applicable+
27W73Not applicable
28W60Not applicable
29M66Not applicable
30W53Not applicable+

For antigen retrieval, slides were treated in sodium citrate buffer with pH 6 in a microwave for 4 × 5 min. Slides were incubated with the primary antibody diluted 1:25 in PBS for 30 min at 37°C. After washing the slides in PBS, a streptavidin-biotin system with horse raddish peroxidase was applied according to a standard protocol employing antibody dilutions provided by Zytomed (Berlin, Germany). For color development aminoethyl-carbazol was used. After color development was stopped, slides were stained with haemalaun and cover slipped using glycerine/gelatine. Immunohistochemical staining was evaluated by a pathologist (MK). Tissues with >25% stained cells were classified as CD10 positive.

Cell proliferation assay

Analysis of cell proliferation was performed on PATU-8988T, HUP-T3 and HUP-T4 cell lines using a MTT assay [3-(4,5-Dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (SIGMA-ALDRICH CHEMIE)] according to the instructions of the manufacturer. 4 × 103 cells/well were seeded onto 96-wells plates, allowed to attach for 24 hr before cells were treated for 24, 48, 72 hr with 2 mM BA, 2 mM VPA with and w/o 10 μM phosphoramidon or 5 μM thiorphan respectively at each time point 20 μl MTT (5 mg/ml) was added to each well and the plates were incubated at 37°C for 4 hr to allow the conversion of MTT to formazan by the mitochondrial dehydrogenase. Formazan quantification was assessed measuring the optical density at 570 nm using 96-well multiscanner auto reader (TECAN Austria GmbH). The effect of each formulation was expressed as a percentage, when comparing treated cells with control cells incubated only with the culture medium.

Statistical analysis

All experiments were reproduced at least 3 times with different passages of the cell lines. The data of NEP/CD10 RNA and protein expression levels are presented as the mean or median, and statistical analyses were performed using t-test. p < 0.05 was considered statistically significant.


Heterogeneous NEP/CD10 mRNA and protein expression in cultured human pancreatic tumor cell lines

The NEP/CD10 mRNA expression of 8 pancreatic tumor cell lines was quantified in the presence of NEP/CD10 standard mRNA using real-time RT-PCR. The different pancreatic tumor cell lines analyzed expressed heterogeneous NEP/CD10 mRNA levels and could be classified into 4 groups (Fig. 1): cells with high NEP/CD10 expression levels (PATU-8988T), medium NEP/CD10 expression levels (HUP-T4), low NEP/CD10 expression levels (HUP-T3) and cells lacking NEP/CD10 mRNA expression (CAPAN-1, PATU-8988S, PATU-8902, CAPAN-2, MIA PaCa-2).

Figure 1.

(a) The expression of NEP/CD10 on human pancreatic tumor cell lines. NEP/CD10 mRNA expression of pancreatic tumor cell lines was quantified in the presence of NEP/CD10 standards using real-time RT-PCR. Pancreatic tumor cell lines analyzed expressed heterogeneous NEP/CD10 mRNA levels. (b) Cell-surface expression of NEP/CD10 on human pancreatic tumor cell lines. Cells were stained with the PE-labeled NEP/CD10-specific mouse mAb or PE-labeled isotype-matched mouse IgG. FACS analysis was performed on human pancreatic tumor cell lines. Pancreatic tumor cell lines analyzed expressed heterogeneous NEP/CD10 protein levels.

In addition, the 8 different pancreatic tumor cell lines were tested for NEP/CD10 protein expression by flow cytometry using an anti-NEP/CD10 specific mAb. As shown in Figure 1b, the levels of NEP/CD10 protein expression were comparable to that of the NEP/CD10 mRNA in the pancreatic tumor cell lines analyzed. Again, the highest NEP/CD10 protein expression was detected in PATU-8988T cells, whereas in HUP-T4 and HUP-T3 cells exhibit moderate to low NEP/CD10 surface protein expression, respectively. As expected, the other 5 pancreatic carcinoma cell lines were not stained by the anti-NEP/CD10 specific mAb (CAPAN-1, PATU-8988S, PATU-8902, CAPAN-2, MIA PaCa-2).

NEP/CD10 expression of pancreatitis and carcinoma tissues

To determine whether there exists a correlation between NEP/CD10 expression and inflammation and/or tumor development, the frequency of NEP/CD10 mRNA expression in pancreatitis and pancreatic tumor lesions was analyzed in vivo using quantitative RT-PCR. Interestingly, there exists no significant difference in NEP/CD10 mRNA expression levels in pancreatic carcinoma compared to that of pancreatitis tissues (data not shown). Regarding protein expression immunoreactivity for NEP/CD10 was detectable in a subset of pancreatic carcinomas exhibiting a membranous staining pattern in tumor cells (Fig. 2). A positive staining was found in 6 of 24 (25%) pancreatic ductal adenocarcinomas. However, the heterogeneous NEP/CD10 expression pattern was not associated with tumor grading, staging and metastasis formation (Table II). Three of 6 chronic pancreatitis demonstrated a focal staining of residual ducts. In addition, in most cases NEP/CD10 immunoreactivity was found perineural within neutrophils or in adjacent duodenal mucosa which served as internal positive control.

Figure 2.

Immunohistochemical detection of NEP/CD10 expression in pancreatic carcinomas as well as in chronic pancreatitis. Perineural (arrow indicating nerve) infiltrating ductal adenocarcinoma with membranous expression of NEP/CD10 (a). Ductal adenocarcinoma negative for NEP/CD10 (*), normal duodenal mucosa served as internal positive control (b). Chronic pancreatitis with focal expression of NEP/CD10 in residual ducts and intravascular neutrophils (arrow) (c, d). [Color figure can be viewed in the online issue, which is available at]

Regulation of NEP/CD10 expression on pancreatic carcinoma cell lines

The influence of various cytokines (TGF-β1, TNF-α, IL-6, IFN-β) as well as of different mediators regulating cell proliferation, differentiation histone acetylation and inducing apoptosis (EGF, PMA, BA, RA, VPA and BeA) on the NEP/CD10 protein expression of pancreatic carcinoma cell lines was assessed by flow cytometry. The NEP/CD10 protein expression of PATU-8988T was slightly upregulated after stimulation with EGF and TNF-α, while in HUP-T3, HUP-T4 EGF and PMA induced NEP/CD10 protein expression. In contrast, TGF-β1, IL-6 and IFN-β did not affect the NEP/CD10 protein expression in these cells. Furthermore, EGF, TNF-α and PMA did neither affect NEP/CD10 mRNA nor protein expression in PATU-8988S and MIA PaCa-2 cell lines (data not shown). Treatment of PATU-8988T, HUP-T3 and HUP-T4 cell lines with different concentrations of BeA and RA did not change the NEP/CD10 surface expression levels, whereas a dose-dependent upregulation of NEP/CD10 protein expression with a maximum 4-fold increase was detected upon treatment with 4 mM BA for 3 days (p = 0.04) in PATU-8988T cells. In addition, VPA treatment resulted in an 2.7-fold increase of NEP/CD10 protein expression (p = 0.019; Fig. 3).

Figure 3.

Effects on NEP/CD10 expression of PATU-8988T cells exposed to varying doses of BA, RA, BeA and VPA for 3 days. After 72 hr treatment NEP/CD10 protein expression was determined using flow cytometry. As shown BA and VPA up-regulated the NEP/CD10 protein expression (p ≤ 0.05), whereas RA and BeA had no effect.

Correlation of functional NEP/CD10 expression with proliferation

Since both histone deacetylation inhibitors BA and VPA may affect cellular proliferation, PATU-8988T, HUP-T3 and HUP-T4 cells were treated with BA and VPA in the presence or absence of phosphoramidon and thiorphan both inhibitors of the enzymatic activity of NEP/CD10 and cell growth was investigated using the MTT assay. Treatment of PATU-T with 2 mM BA or 2 mM VPA resulted in a significant (p ≤ 0.01) reduced cell growth after 72 hr (Figs. 4a and 5). Furthermore, addition of phosporamidon or thiorphan increased the proliferation rate of VPA od BA treated cells. In HUP-T4 cells VPA, but not BA significantly decreased the cell growth (Fig. 4b). The VPA effect on the proliferation in HUPT4 as well as in HUP-T3 cells (data not shown) was much lower than in the high NEP/CD10-expressing PATU-8988T cells.

Figure 4.

(a) Effects of BA and VPA on the proliferation rate. PATU-8988T cells were exposed to BA and VPA with and w/o the NEP/CD10 inhibitors phosphoramidon and thiorphan. The proliferation rate of PATU-T cells was determined after 72 hr using MTT assay. (b) Effects of BA and VPA on the proliferation rate. HUP-T4 cells were exposed to BA and VPA with and w/o the NEP/CD10 inhibitors phosphoramidon and thiorphan. The proliferation rate of HUP-T4 cells was determined after 72 hr using MTT assay.

Figure 5.

Antiproliferative effects of both BA and VPA after treatment of PATU-8988T cells for 3 days. The pictures made by light microscopy show the decreasing numbers of cells after treatment with BA or VPA. [Color figure can be viewed in the online issue, which is available at]


This study investigated the expression and regulation of neutral endopeptidase (NEP/CD10) in 8 cell lines of pancreatic ductal adenocarcinoma as well as the expression of NEP/CD10 in tissue samples of patients with pancreatitis and pancreatic tumors. The results demonstrated heterogeneous NEP/CD10 expression levels in the different pancreatic carcinoma cells lines analyzed. Although the PATU-8988T and PATU-8988S cell lines derived from the same original tumor,26 they exhibit a different cell morphology, differentiation status, gene expression pattern and in vivo growth. PATU-8988T cells expressed high levels of NEP/CD10 mRNA and protein, whereas PATU-8988S expressed neither NEP/CD10 mRNA nor protein.26 HUP-T4 cells expressed moderate mRNA and protein levels of NEP/CD10, and HUP-T3 cells showed low NEP/CD10 mRNA and protein expression.27 Lack of NEP/CD10 expression was detected in the cell lines CAPAN-1, CAPAN-2, MIAPaCa-2 and PATU-8902. Although the different cell lines analyzed are of distinct differentiation status and also exhibit a distinct in vivo growth (Table I), the heterogeneous NEP/CD10 expression pattern observed appears not to be correlated with their biological behaviour.28, 29, 30, 31

In addition, quantification of NEP/CD10 mRNA amounts in tissues of pancreatic carcinomas vs. pancreatitis did not show significant differences. This might be due to the heterogeneity of the tissues contaminated by various normal cells. Expression of NEP/CD10 protein was detectable in a subset of ductal pancreatic adenocarcinomas as well as tissue samples of pancreatitis. NEP/CD10 was expressed in a membranous staining pattern in tumor cells, but was not associated with tumor grading or staging. However, further studies with a larger number of samples might give insights into the prognostic role of NEP/CD10 expression in these diseases.

The heterogeneous mRNA expression of NEP/CD10 in different tumors suggest its transcriptional regulation. The NEP/CD10 mRNA expression can be regulated by 2 alternative promoters, which control its tissue- and developmental stage-specific gene expression.21, 32 Both type 1 and 2 NEP/CD10 regulatory regions are characterized by the presence of multiple transcription initiation sites and the absence of a TATA box and consensus initiator elements. The major type 2 promoter has functionally important transcription factor binding sites, one of which is identical to CCAAT-binding transcription factor/nuclear transcription factor Y (CBF/NFY)33 which might mediate tissue-specific expression of NEP/CD10 by alternatively splicing. Decreased NEP/CD10 expression in cancer can be caused by hypermethylation of the NEP/CD10 promoter as shown in the human prostata cancer cell line PC3. Treatment of this cell line with the demethylating agent 5-aza-2′-deoxycytidine results in an increased NEP/CD10 transcription.34 However, exposure of PATU-8988T cells to 5-aza-2′-deoxycytidine did not change the mRNA expression levels of NEP/CD10 (data not shown).

Since it has been demonstrated that various substances modulated NEP/CD10 expression in cultured leukemia cells,35 the influcence of growth factors, cytokines and differentiation/acetylation modulators on the NEP/CD10 expression was determined in PATU-8988T, HUP-T3 and HUP-T4 cells. The expression of NEP/CD10 protein was slightly upregulated after stimulation with EGF which is in contrast to renal vascular smooth muscle cells where NEP/CD10 mRNA expression was reduced after stimulation with EGF.36 The cytokine TGF-β did not influence the NEP/CD10 expression in PATU-8988T, HUP-T3 and HUP-T4 cells, whereas in human endometrial stromal cells an approximately 60% reduced NEP/CD10 expression was detectable upon 3 days of TGF-β treatment.37 These results are again controversial to renal vascular smooth muscle cells in which TGF-β was found to upregulate NEP/CD10 expression.36 TNF-α did not influence the NEP/CD10 protein expression in PATU-8988T, HUP-T3 and HUP-T4 cells, but a TNF-α-mediated increase of NEP/CD10 expression was found on human bronchial epithelial cell.38 The PMA-induced down-regulation of the NEP/CD10 mRNA level was equivalent to results on the acute lymphoblastic leukemia cell line REH. In contrast, PMA could induce both NEP/CD10 mRNA and protein expression in HUP-T4.35 IL-6 had no effect on NEP/CD10 protein expression in PATU-8988T, HUP-T3 and HUP-T4 cells which is in line with results from a pre-B ALL cell line, where IL-6 did not stimulate the expression of NEP/CD10 mRNA.39

In addition, the effect of BeA, RA, BA and VPA, which can induce growth arrest, differentiation and/or inhibition of histone deacetylation, was tested on pancreatic carcinoma cells. An upregulation of both NEP/CD10 mRNA and protein expression was found by 2 mM BA treatment of PATU-8988T, HUP-T3 and HUP-T4 cells. Concentrations of 3 mM BA and higher induced an approximately 40% growth arrest of these cells, a result compatible with other investigations in pancreatic tumor cell lines.40, 41, 42, 43, 44, 45, 46 Expression profiling of BA-treated lung carcinoma cells resulted in changes of genes related to cytokine signaling and cancer metastasis.47 The differential regulation of metastasis-associated genes by BA provides an explanation for the anti-invasive properties of butyrate.

Both VPA and BA are members of the class of histone deacetylase inhibitors which are discussed as one of the most promising class of new anti-cancer drugs.48 VPA induced an 2.7-fold higher NEP/CD10 expression in PATU-8988T, HUP-T3 and HUP-T4 cells, whereas the effect of BA was 2-fold higher than the action of VPA in PATU-8988T. Inhibitors of histone deacetylase induce hyperacetylation in chromatin usually resulting in activation of certain genes. They induce terminal cell differentiation and/or apoptosis in cancer cells.49 Both BA and VPA were able to induce NEP/CD10 expression, which was accompanied by inhibition of the proliferation of PATU-8988T and HUP-T4 cells. Addition of the NEP/CD10 enzyme activity inhibitors, phosphoramidon and thiorphan, separately to BA or VPA resulted in a normalization of the proliferation rate suggesting the involvement of enzyme activity of NEP/CD10. In prostate carcinoma, NEP/CD10 can regulate cell migration via mechanisms both dependent and independent of its catalytic function.50 Furthermore, catalytically active NEP/CD10 inhibits neuropeptide-mediated activation of the insulin growth factor-1 and the resulting downstream activation of PI-3 kinase and Akt/PKB kinase.51 Recently, it has been reported that NEP/CD10 and the PTEN tumor suppressor associate directly through electrostatic interactions.52

RA that induces growth arrest and differentiation of tumor cells via RAR and RXR did not affect the NEP/CD10 mRNA and protein expression. In comparison with other cell lines, RA induced down-regulation of NEP/CD10 on SK-N-BE(2) cells, but had no effect on BE(2)-M17 cells.44 RA exerts differential effects on the growth and apoptosis of a pair of metastatic and non-metastatic clones derived from the MDA-MB-435 breast cancer cell line.45 BeA induced a slight upregulation of NEP/CD10 protein expression and strongly inhibited the growth of PATU-8988T cells. BeA, a natural component isolated from Birch trees, effectively induces apoptosis in neuroectodermal and epithelial tumor cells and exerts little toxicity in animal trials. Furthermore, BeA induced growth inhibition in neoplastic cell lines and did not affect non-neoplastic cell lines and proliferating normal lymphocytes.46, 53

In summary, we could show a heterogeneous NEP/CD10 mRNA and protein expression pattern in tumor lesions of pancreatic carcinoma patients and pancreatic carcinoma cell lines. Studies of characterizing the natural substrates of the protease, the NEP/CD10 associated differential gene expression and its function in pancreatic carcinomas are currently in progress to further elucidate the complex role of the cell surface ectoprotease for tumor cell growth, motility and proliferation.


Many thanks to Kathrin Hammje, Christl Sauer and Anja Winkler for excellent technical support. In addition, we thank for the excellent secretarial help of Claudia Stoerr and Liis Pellenen.