Effects of Chemotherapeutic Agents on Expression of Somatostatin Receptors in Pancreatic Tumor Cells

Cell Lines and Tissue Culture

All cell lines were purchased from the American Type Culture Collection (Rockville, MD); BxPC-3, Panc-1, ASPC-1, and Capan-1 are pancreatic adenocarcinoma cell lines. Cells were cultured in RPMI 1640 medium (Seromed, Berlin, Germany) supplemented with 10% fetal calf serum, 4 mmol/L l-glutamine, and antibiotics in 5% CO₂/95% O₂ air at 37°C. The fetal calf serum was heat inactivated for 30 min at 56°C.

Cells were grown in 162-cm² culture flasks (Costar, Cambridge, MA) for 2–3 d. At that point the cells were subconfluent, and the chemotherapeutic agents were added in their respective concentration and cells were grown for another 4 d. Cells were gently harvested with a solution of 0.05% trypsin and 0.02% ethylenediaminetetraacetic acid (GIBCO Laboratories, Grand Island, NY) and counted in a cell counter (Sysmex Toa, Tokyo, Japan).

Synthesis and Radiolabeling of DOTA-LAN

DOTA-LAN was synthesized in a 3-step reaction using the commercially available LAN (Somatuline [Ipsen, Maidenhead, U.K.]; d-Nal-Cys-Tyr-d-Trp-Lys-Val-Cys-Thr-NH₂) and DOTA as starting materials. The terminal amino group of LAN was blocked using t-dibutyl dicarbonate to produce ε-Boc LAN. Eighty milligrams DOTA, 74 mg N-hydroxysuccinimide, and 80 mg [ε-Boc-Lys⁵]LAN were dissolved in 15 mL water and 30 mL N,N-dimethylformamide. One hundred milligrams N,N′-dicyclohexylcarbodiimide were added, and the solution was stirred at room temperature for 16 h. The Boc-protecting group was cleaved, and the product [DOTA-d-Nal¹]LAN was purified by high-performance liquid chromatography as described (19).

An aliquot of this DOTA-LAN solution was incubated with ¹¹¹InCl₃ (Nordion International, Ottawa, Ontario, Canada; no carrier added, >1,850 MBq/mL in 0.05 mol/L HCl). A reaction ratio of approximately 20 MBq ¹¹¹In per nanomole of conjugate was used, and the complexing reaction was allowed to proceed for 30 min at 100°C. An aliquot was then removed, and the purity was assessed by instant thin-layer chromatography. The radiolabeling efficiency was typically >97% and was critically dependent on the chemical purity (metal cations) of the ¹¹¹InCl₃. The radiolabeled conjugate was then formulated in 0.15 mol/L NaCl, 0.05 mol/L NH₄OAc, 0.2 mol/L ascorbic acid, 0.1% human serum albumin, and 4 mmol/L diethylenetriaminepentacetic acid before sterilization by filtration through a 0.22-μm-pore membrane filter.

Chemotherapeutic Agents

All chemotherapeutic agents used in these experiments were approved substances for treatment of tumor patients. For each series of experiments an untreated control group and 2 groups treated with the chemotherapeutic agent in various concentrations were used. All chemotherapeutic substances except gemcitabine were purchased from Sigma Chemie GmbH (Deisenhofen, Germany). Gemcitabine was obtained from Eli Lilly GmbH (Vienna, Austria). The drugs were stored frozen as 100 × stock solutions in dimethyl sulfoxide.

Effects of Chemotherapeutic Agents on Cell Growth and Cell Cycle

To evaluate the effects of the chemotherapeutic agents on cell growth and survival, we determined the reduction in cell number after treatment with the chemotherapeutic agents, the percentage of cells in the different phases of the cell cycle before and after treatment, and the percentage of apoptotic cells before and after treatment with the chemotherapeutic agents. Cells of the control group were treated with medium alone. Cells were seeded at an initial cell density of 10⁵ cells per milliliter in 162-cm² tissue culture flasks (Costar) and incubated for 72 h under tissue culture conditions. For all experiments the cells were kept in RPMI 1640 medium supplemented with 10% fetal calf serum and 4 mmol/L l-glutamine, and cell numbers in controls and treated cultures were counted in triplicate in a microcell counter (Sysmex Toa).

For staining of the DNA and cell cycle analysis, 4 × 10⁶ cells were washed in phosphate-buffered saline (PBS; Sigma, St. Louis, MO), resuspended in 100 μL PBS, and added to 4 mL 70% ethanol (precooled at −20°C) under continuous mixing. The cells were fixed for 30 min on ice, washed again with PBS, resuspended in 200 μL 0.1% Triton X-100 and PBS supplemented with 2 μg/mL ribonuclease A (type XII-A; Sigma) and 50 μg/mL propidium iodide, and incubated overnight at room temperature. Cells were analyzed using an Epics XL-MLC (Coulter, Miami, FL) flow cytometer (10,000 cells), and the resulting histograms were processed using MulticycleAV software (Phoenix Flow Systems, San Diego, CA). Data are presented as percentage of cells (mean ± SD; n = 3) in G1/0, G2M, or S phase. Apoptotic cells were calculated as the percentage of cells appearing as a subG1 (subG1/0 cell DNA content) peak in cell cycle histograms.

Northern Blot Analysis of SSTRs

Plasmids containing the receptor complementary DNAs (cDNAs) for the hSSTRs (hSSTR1–hSSTR4) were kindly provided by Prof. Dr. G. Bell (Howard Hughes Medical Institute, Chicago, IL). The plasmid carrying hSSTR5 was kindly provided by Dr. A.M. O’Carroll (National Institutes of Health, Bethesda, MD). The precision of each receptor clone was confirmed by agarose gel analysis of restriction fragments and a complete sequence analysis. Isolation of plasmid DNA was performed with the Qiagen plasmid purification kit (Qiagen, Hilden, Germany).

Tumor cells, previously snap frozen in liquid nitrogen, were homogenized in Trizol reagent (Molecular Research Center Inc., Cincinnati, OH) using a glass homogenizer. Total RNA from the homogenate was extracted according to the manufacturer’s protocol: a modification of the RNA extraction method described by Chomczynski et al. in 1987 (20). Northern transfer of approximately 20 μg total RNA was performed using a Nytran membrane (Schleicher & Schuell, Vienna, Austria) by capillary blotting overnight. RNA was fixed to the membranes by ultraviolet crosslinking. Specific probes for Northern hybridization were generated by restriction cutting of the plasmids carrying the probes with the appropriate restriction enzymes. After separation with an agarose gel, the probes were purified with the Qiaex gel purification kit (Qiagen) and labeled using the Redivue random prime labeling kit (Amersham, Buckinghamshire, U.K.) and ³²P-deoxycytidine 5′-triphosphate (Amersham). Hybridization was performed as described (20,21). Briefly, the membranes were prehybridized at 42°C in a hybridization solution containing 50% formamide, 5 × Denhardt’s solution, 5 × saline sodium citrate (SSC), 0.2% sodium dodecyl sulfate (SDS), and 100 μg/mL salmon sperm DNA. After 4 h of prehybridization, the labeled probe was added in fresh hybridization buffer and hybridization was performed overnight. Thereafter, blots were rinsed twice at room temperature with 2 × SSC buffer containing 0.1% SDS, then rinsed twice at 42°C with 0.2 × SSC buffer containing 0.1% SDS, and finally used to expose an x-ray film (Hyperfilm; Amersham).

Radioligand Binding of ¹¹¹In-DOTA-LAN In Vitro

To evaluate ligand binding to SSTRs expressed on various tumor cells, direct binding experiments were performed (22). All experiments were performed at least in triplicate and consisted of 2 incubation series: total binding and nonspecific binding. Specific binding was expressed as the difference of total and nonspecific binding. Before cells were used in binding experiments, they were washed twice in assay buffer (0.15 mol/L NaCl, 20 mmol/L Hepes-KOH, pH 7.4) containing 0.1% bovine serum albumin. In saturation studies, 2 × 10⁵ cells were incubated with increasing concentrations of ¹¹¹In-DOTA-LAN (0.1–30 nmol/L) in the absence (total binding) or the presence (nonspecific binding) of 0.01–50 nmol/L unlabeled DOTA-LAN under equilibrium conditions (90 min at room temperature). The final assay volume was 0.2 mL. Displacement studies were performed by incubating 2 × 10⁵ cells with 0.05 nmol/L ¹¹¹In-DOTA-LAN in the absence (total binding) or the presence (nonspecific binding) of increasing concentrations (0.01–1,000 nmol/L) of unlabeled DOTA-LAN. The reactions were terminated by addition of 1 mL ice-cold 20 mmol/L Hepes-KOH buffer containing 0.1% bovine serum albumin followed by rapid centrifugation (4,000 rpm, 10 min, 4°C) and 2 buffer washes. Radioactivity associated with the pellet was counted in a γ-counter for 1 min. The saturation binding data were evaluated according to Scatchard (23).

Statistical Analysis

Statistic calculations were done with the Texasoft program (Kwikstat, Cedar Hill, TX) using the Mann–Whitney test with a threshold for significance at P < 0.05.

Expression of SSTR1–SSTR5 in Pancreatic Adenocarcinoma Cell Lines

The expression of SSTR subtypes in BxPC-3, Panc-1, ASPC-1, and Capan-1 cells was tested using subtype-specific cDNA clones and Northern blot analysis. Figure 1A shows the relative mean signal intensities of densitometric measurements of the resulting hybridizations. SSTR3 is predominantly expressed in the cell lines investigated here, with low levels of SSTR4 and SSTR5 and no significant expression of SSTR1 and SSTR2 subtypes.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

(A) Relative intensities of chemoluminescence detection of SSTR1–SSTR5 after hybridization with respective cDNA clones for 4 pancreatic cancer cell lines (mean ± SEM; 3 experiments). Intensities obtained for SSTR1 and SSTR2 are not statistically different from SSTR-negative COS7 cells (not shown). (B) Results of binding experiments using DOTA-LAN determined as specific binding (femtomoles per million cells), number of sites per cell (× 10⁴), and K_d (nmol/L) for high- and low-affinity sites (1 and 2), respectively, from 3 experiments (mean ± SD; determinations done in triplicate). Number of binding sites per cell is not statistically significant for BxPC-3, Panc-1, and ASPC-1 cells.

Binding of DOTA-LAN to Pancreatic Adenocarcinoma Cell Lines

To study the binding properties of radiolabeled DOTA-LAN with established human pancreatic tumor cell lines, saturation studies were performed with intact tumor cells. All cell lines bound ¹¹¹In-DOTA-LAN in a specific manner. For cell lines BxPC-3, Panc-1, and ASPC-1 the binding could be subdivided into 2 binding classes—namely, high-affinity and low-affinity binding sites (dissociation constants K_d1 = 0.1–5 nmol/L vs. K_d2 = 7.5 −28 nmol/L). The Capan-1 cell line expressed only 1 class of binding sites. Figure 1B shows the maximal binding capacity (B_max), the total number of binding sites per cell, and the binding affinities (K_d) for the various tumor cell lines and receptors. Because BxPC-3, Panc-1, and ASPC-1 cells exhibited similar binding characteristics, most of the following experiments were performed with BxPC-3 cells.

Effects of Chemotherapeutics on Growth of Pancreatic Cancer Lines

The tumor cell lines were treated with appropriate concentrations of the chemotherapeutic agent in tissue culture flasks, and cell numbers of medium controls and treated cultures were counted after exposure to 0.05% trypsin and 0.02% ethylenediaminetetraacetic acid. The result of these measurements is shown in Figure 2A for BxPC-3 cells. Exposure of the cells to 1.0 or 2.0 μg/mL gemcitabine, 2.5 or 5 μg/mL cisplatin, 1.5 or 3 μg/mL camptothecin, 0.1–0.2 μg/mL mitomycin C, and 1.0 or 2.0 μg/mL doxorubicin resulted in reductions in cell numbers ranging from 6% for cisplatin to 75% for camptothecin. For Panc-1 cells, the reduction is 30.2% ± 3.1% and 28.8% ± 1.6% for 1.0 and 2.0 μg/mL gemcitabine, respectively. When Panc-1 cells were treated with 5 μg/mL cisplatin, the cell number reduction was 6% ± 5.25% and 10 μg/mL cisplatin gave a reduction of 4.7% ± 2.3%. ASPC-1 cells, when treated with 65 μg/mL 5-fluorouracil, revealed a reduction of the cell number of 49.7% ± 2.5% and, after treatment with 130 μg/mL 5-fluorouracil, a reduction of 47.6% ± 2%. The reduction of the cell number of Capan-1 cells was 18.6% ± 1.4% when cells were treated with 65 μg/mL 5-fluorouracil and 26.7% ± 0.9% when treated with 130 μg/mL 5-fluorouracil.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

(A) Inhibitory effects of respective chemotherapeutic agents on BxPC-3 cells (% inhibition compared with medium controls; n = 3; mean ± SEM) and percentage of subG1 cells in same cultures (n = 3; mean ± SEM). (B) Alterations in cell cycle distributions in BxPC-3 cells in response to chemotherapeutic agents. Percentage of cells in G1/0 phase, G2M phase, and S phase were calculated from histograms of propidium iodide–stained cells (n = 3; mean ± SEM). con = control; gemca = gemcitabine (1 and 2 μg/mL); cisPt = cisplatin (2.5 and 5 μg/mL); cpt = camptothecin (1.5 and 3 μg/mL); mito = mitomycin C (0.1 and 0.2 μg/mL); doxo = doxorubicin (1 and 2 μg/mL).

Induction of Apoptosis in Response to Chemotherapeutic Agents

The human pancreatic adenocarcinoma tumor cell lines BxPC-3, Panc-1, ASPC-1, and Capan-1 showed a percentage of 1.7%–10.35% of apoptotic cells in medium controls (Fig. 2A). Apoptotic cells were detected as a fraction of subG1 cells on cell cycle histograms. All chemotherapeutics induced the appearance of apoptotic cells, with gemcitabine showing highest activity, followed by mitomycin C, camptothecin, doxorubicin, and cisplatin.

Again, similar results were found for Panc-1cells, with approximately 50% apoptotic cells for gemcitabine pretreatment and approximately 5% apoptotic cells for cisplatin. When ASPC-1 and Capan-1 cells were treated with 5-fluorouracil, the percentage of apoptotic cells was also increased, ranging from 35% to −57% apoptotic cells (data not shown).

Cell Cycle Distribution

Cell cycle distribution was measured by analysis of histograms obtained with propidium iodide–stained fixed cells in flow cytometry (Fig. 2B for BxPC-3 cells). Under these experimental conditions, gemcitabine, cisplatin, and camptothecin induced an accumulation of cells in S phase and at the higher concentration in G2M phase, whereas mitomycin C arrested cells primarily in G1/0 phase and treatment with doxorubicin in G2M phase. Alterations in S-phase distribution are significantly different from controls for all treatments except 1 μg/mL doxorubicin and in G2M-phase distribution for all treatments except 2.5 μg/mL cisplatin, 0.2 μg/mL mitomycin C, and 1 μg/mL doxorubicin.

Similar results were found for Panc-1 cells, with accumulation in S or G2M phase for gemcitabine treatment and S phase for cisplatin. 5-Fluorouracil induced high S-phase accumulation of ASPC-1 and Capan-1 cells (data not shown).

Effects of Chemotherapeutics on Binding of ¹¹¹In-DOTA-LAN to Pancreatic Cancer Cells

After treating BxPC-3 cells with gemcitabine, the high-affinity binding sites were diminished and the low-affinity binding sites were reduced dose dependently (Fig. 3). In contrast, cisplatin in concentrations of 2.5 and 5 μg/mL increased the expression of the high-affinity and low-affinity receptors of pancreatic tumor BxPC-3 cells significantly. After treatment with camptothecin in concentrations of 1.5 and 3 μg/mL, the high-affinity binding sites of the BxPC-3 cells were diminished, with increased expression of low-affinity sites for 1.5 μg/mL camptothecin. Treatment with mitomycin C (0.1 and 0.2 μg/mL) as well as doxorubicin resulted in disappearance of high-affinity sites and increased expression of low-affinity sites. The differences in expression of high-affinity sites are statistically significant for all treatments (B_max and number of sites) and for all binding tests except those involving doxorubicin. 5-Fluorouracil reduced the amount of high-affinity DOTA-LAN binding in ASPC-1 cells, without induction of low-affinity binding sites, and binding of the tracer was not affected in Capan-1 cells. Regression analysis revealed a weak correlation of the number of low-affinity binding sites to the percentage of apoptotic cells when results for gemcitabine were excluded (r² = 0.62) because gemcitabine shows a different behavior regarding cytotoxicity.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Effects of chemotherapeutic agents on binding of DOTA-LAN and number of binding sites per cell for high- and low-affinity sites (1 and 2), respectively, in BxPC-3 cells (n = 3; mean ± SEM). gemca = gemcitabine (1 and 2 μg/mL); cisPt = cisplatin (2.5 and 5 μg/mL); cpt = camptothecin (1.5 and 3 μg/mL); mito = mitomycin C (0.1 and 0.2 μg/mL); doxo = doxorubicin (1 and 2 μg/mL).

To check for the reversibility of the suppression of high-affinity binding sites, gemcitabine was used in the concentration of 1 μg/mL on BxPC-3, Panc-1, and ASPC-1 cells for 4 d. Thereafter, the cells were washed with medium and incubated for another 4 d in tissue culture (Fig. 4). These experiments are referred to as gemcitabine reexpression experiments. These experiments show that treatment with gemcitabine not only is reversible but also is followed by significant overexpression of high- and low-affinity binding sites in the case of BxPC-3 and Panc-1 cells and extremely high overexpression of low-affinity sites in ASPC-1 cells within 4 d after treatment.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Reexpression of SSTR in pancreatic cancer cells after pretreatment with gemcitabine at 1 μg/mL (gemca1) and further incubation for 4 d (reexp; n = 3; mean ± SEM). Low-affinity binding of ASPC-1 cell reexpression exceeds scale used here (636.5 ± 11.5 fmol/10⁶ cells). con = control.