Panc-1 and ASPC-1 human pancreatic cancer cells (HPCCs) were obtained from the American Type Culture Collection (Manassas, VA, USA). All cells were cultured in DMEM (Hyclone; GE Healthcare Life Sciences, Logan, UT, USA) with 10% fetal bovine serum (FBS) (Hyclone; GE Healthcare Life Sciences), 2 Mm L-glutamine, 100 U/ml penicillin and 100 mg/ml streptomycin, at 5% CO₂ and 37°C. All cell culture reagents were purchased from Gibco; Thermo Fisher Scientific, Inc. (Waltham, MA, USA). The Panc-1 and ASPC-1 cells (1×10⁴) were exposed to gemcitabine at serially increased concentrations, with an initial concentration of 10 nM, to generate gemcitabine-resistant cells. Following 2 weeks of adaptation, the concentration was doubled. The cells were adapted to a final gemcitabine concentration of 640 nM, termed gemcitabine-resistant (GR)-Panc-1 cells and GR-ASPC-1 cells.

The total RNA expression levels of hENT1 and RRM1 were determined using a previously described protocol (6,8). Briefly, the total RNA of cell was extracted by TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.) according to the protocol. Triplicates of each gene and each specimen were used, with GAPDH as an internal standard. The single-strand cDNA for PCR template was synthesized from 10 µg of total RNA by Illumina TotalPrep RNA amplification kit (Ambion; Thermo Fisher Scientific, Inc.). StepOne™ Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) was used in the RT-PCR assay. The RT-PCR was performed with a total reaction volume of 20 µl, including 10 µl Power SYBR Green PCR Master mix (Applied Biosystems; Thermo Fisher Scientific, Inc.), 5 pmol of forward and reverse primer respectively and 2 µl of cDNA. Threshold cycle(Ct) was observed in the amplification with 35 cycles of 1 min at 95°C, 1 min at 58°C, and 1 min at 72°C. And the relative of mRNA expression levels was calculated by the relative quantitation method 2^−∆∆Cq (15). The fold change to control sample of each samples was calculated. Relative gene quantification was performed using StepOne™ software 2.1 (Applied Biosystems; Thermo Fisher Scientific, Inc.). The following primers were used: GAPDH, forward 5′-CGG AGT CAA CGG ATT TGG TCG TAT-3′ and reverse 5′-AGC CTT CTC CAT GGT GGT GAA GAC-3′; hENT1, forward 5′-AGC AGG CAA AGA GGA ATC TGG AGT-3′ and reverse 5′-GAA GGC AAA GGC AGC CAT GAA GAA-3′; RRM1, forward 5′-CAT CCA CAT TGC TGA GCC TA-3′ and reverse 5′-GAT TAG CCG CTG GTC TTG TC-3′.

For each sample, 5×10⁶ cells were lysed for 30 min in lysis buffer (Beyotime Institute of Biotechnology, Beijing, China) on ice, and the debris was centrifuged at 12,000 × g for 12 min at 4°C. A BCA assay (Beyotime Institute of Biotechnology) was used to determine the protein concentration, following which 30 µg proteins were separated using 8–15% SDS-PAGE and blotted onto a PVDF membrane. The membrane was blocked using blocking solution of 5% nonfat dry milk in PBST (0.05% Tween20 in PBS). The membrane was then incubated with primary antibodies for 2 h at 37°C at a 1:1,000 dilution, following which the PVDF membrane was washed three times for 10 min in PBST. The membrane was then incubated with secondary antibodies for 1 h at 37°C at a 1:10,000 dilution, following which the PVDF membrane was washed three times for 10 min in PBST. All antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA), the details of antibodies were followed: hENT1 (sc-48489, polyclonal, goats anti-human), GAPDH (sc-293335, monoclonal, mouse anti-human), RRM1 (sc-22786, monoclonal, rabbit anti-human), Twist1 (sc-134136, polyclonal, mouse anti-human), Slug (sc-166902, monoclonal, mouse anti-human), E-cadherin (sc-33743, polyclonal, rabbit anti-human), α-SMA (sc-53142, monoclonal, mouse anti-human), NICD (sc-74276, monoclonal, mouse anti-human), Gli1 (sc-20687, polyclonal, rabbit anti-human), PARP (sc-27034, polyclonal, goats anti-human), Capase3 (sc-1224, polyclonal, goats anti-human), goats IgG (sc-2419), mouse IgG (sc-516176), rabbit IgG (sc-2794).

A CCK8 was used to assess cell viability. The cells (10⁴/well) were plated on a 96-well plate and attached overnight, the cells were treated with the drugs for 24 h, following which the medium was removed and the cells were washed three times with PBS. Subsequently, 90 µl DMEM and 10 µl CCK8 were added to each well, and incubated for 1.5 h at 37°C. A microplate reader was used to measure the optical density values at 450 nm.

To the adequately suspended treated cells, 0.4% (w/v) Trypan blue solution was added, for which the volume ratio of cell suspension to Trypan blue solution was 9:1. The cells were counted under a microscope, with dead cells failing to exclude the dye, and the death rate was calculated as follows: Total death rate=(number of dead cells/total cells)x100%.

Transient transfection of cells with small interfering (si)RNAs and plasmids were performed using Lipofectamine 2000 (Invitrogen Life Technologies; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. At 36 h post-transfection, the drugs were added to the cells, which were collected following a 24 h period for western blot analysis. The sequences of (si)RNAs were as follows: hENT1, 5′-AUGACAUUGUUGAAGAUGGCA-3′), Twist1, 5′-AAACAUUUGUUUUAAGGAGAA-3′); Slug, 5′-ACUAAUGGGGCUUUCUGAGCC-3′); NICD, (5′-UAAAGAGAGAAUAUCGUAGUC-3′); Gli1, (5′-UUUCAUACACAGAUUCAGGCU-3′).

A BioCoat Matrigel invasion chamber system (Corning Incorporated, NY, USA) was used to determine cell invasion. In a Transwell plate, the lower chamber was filled with culture medium without cells, and the upper chamber was filled with cell suspension (1×10⁵) and medium containing 10% FBS. The Transwell plate was incubated at 37°C for 24 h. The cells adherent to upper chamber surface were removed, and the cells adherent to lower chamber surface were stained with 4% paraformaldehyde for 15 min, rinsed with water and dried. Crystal violet was extracted with 50% ethanol containing 0.1 M sodium citrate, and the absorbance was measured at 600 nm.

Following treatment, the cells were incubated with Annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI; Beyotime Institute of Biotechnology) at room temperature for 15 min. A FACScan flow cytometer was used for analysis of the apoptotic ratio.

A total of 30 male 6-week-old BALB/c nude mice were purchased from the Institute of Zoology, Chinese Academy of Sciences (Beijing, China). All the mice were fed in the specific-pathogen free environment, the plastic cage was sealed with an air filter, animal isolators, air laminator, and air laminar flow chamber were equipped, and the feeding environment with temperature 24–28°C, relative humidity 50~60%, ventilation required 10 to 15 times per hour, natural circadian light. and the mice were given the sterilized food, and the water with bacitracin (4 g/l) and neomycin (4 g/l). All animal experiments were performed according to the guidelines of the Institutional Animal Care and Use Committee of the Institute of Zoology, Chinese Academy of Sciences. Each BALB/c nude mouse was subcutaneously inoculated with 5×10⁶ Panc-1 or GR-Panc-1 cells in their right and left hind footpads. At 2 days post-inoculation, the mice inoculated with Panc-1 cells were administered with 10 mg/kg gemcitabine three times every day via intraperitoneal injection. The mice inoculated with GR-Panc-1 cells were administered with 10 mg/kg gemcitabine or co-administered with 100 mg/kg sclareolide and 10 mg/kg gemcitabine three times every day via intraperitoneal injection, with the sclareolide injected 2 h prior to the gemcitabine. The sclareolide and gemcitabine were dissolved in saline. The mice were sacrificed by anesthesia with pentobarbital at 2, 3 and 4 weeks after the cell inoculation, and the tumor volumes were measured.

The tumors were immersed in 4% paraformaldehyde for 24 h, and were then dehydrated in 30% sucrose solution, following which the tissues were paraffin-embedded for sectioning (10 µm). The sections were treated using an in situ Cell Death Detection kit (Roche Diagnostics, Indianapolis, IN, USA) according to the manufacturer's protocol.

The data are represented as the mean ± standard deviation from triplicate experiments. All the data were processed by SPSS version 19.0 (IBM SPSS, Armonk, NY, USA). Two-way analysis of variance was used to analyze the variance of different groups. P<0.05 was considered to indicate a statistically significant difference.

The GR-Panc-1 and GR-ASPC-1 cells were induced with increasing concentrations of gemcitabine from 10 nm. The GR-HPCCs (GR-Panc-1 and GR-ASPC-1 cells) and HPCCs (Panc-1 and ASPC-1 cells) were treated with different concentrations of gemcitabine. No significant alterations in cell viability or cell death ratio were observed in the GR-HPCCs following exposure to increasing gemcitabine concentrations. For the HPCCs, the cell viability decreased and the cell death ratio increased with increased gemcitabine concentrations (Fig. 1A). The GR-HPCCs were treated with different concentrations of gemcitabine combined with 10 µM sclareolid. At a gemcitabine concentration of 0.5 µM, the cell viability and cell death ratio were significantly altered, compared with the cells treated with gemcitabine alone. The optimal concentration of gemcitabine was 1 µM (Fig. 1B). Similarly, GR-HPCCs were treated with different concentrations of sclareolide combined with 1 µM gemcitabine. At a sclareolide concentration of 5 µM, the cell viability and cell death ratio were significantly altered, compared with the cells treated with gemcitabine alone, with an optimal sclareolide concentration of 10 µM (Fig. 1C). These results indicated that 10 µM sclareolide enhanced gemcitabine-induced GR-HPCC death.

Several studies have reported that hENT1 and RRM1 are important in the treatment of pancreatic cancer with gemcitabine (5,6,8). HENT1 is a transporter of gemcitabine, and RRM1 is a target of gemcitabine. Higher expression levels of hENT1 and lower expression levels of RRM1 may assist in the treatment of pancreatic cancer with gemcitabine. The mRNA and protein expression levels of hENT1 were lower in the GR-HPCCs, compared with the HPCCs, and the mRNA and protein expression levels of RRM1 were higher in the GR-HPCCs, compared with the HPCCs. This suggested that hENT1 and RRM1 were important in gemcitabine resistance (Fig. 2A and B). The HPCCs and GR-HPCCs were treated with gemcitabine combined with sclareolide, and the results showed that sclareolide upregulated the mRNA and protein expression levels of hENT1 and downregulated the mRNA and protein expression levels of RRM1 (Fig. 1A and B). Knocking down hENT1 by transfection of the GR-Panc-1 cells with siRNA resulted in a reduction in the cell death ratio induced by gemcitabine combined with sclareolide, compared with the cells transfected with control siRNA (Fig. 2C). The overexpression of RRM1 also inhibited the cell death ratio induced by gemcitabine combined with sclareolide in the GR-Panc-1 cells (Fig. 2D). These results showed that sclareolide enhanced gemcitabine-induced GR-HPCC death through targeting hENT1 and RRM1.

Previous studies have reported that EMT is in the active state in gemcitabine-resistant pancreatic cancer. In order to confirm whether sclareolide can affect the EMT in GR-HPCCs, the epithelial cell marker, E-cadherin, and mesenchymal cell marker, α-smooth muscle actin (SMA) were detected in the HPCCs and GR-HPCCs, and the expression levels of TWIST1 and Slug were examined. The HPCCs and GR-HPCCs were treated with gemcitabine alone or combined with sclareolide, the results showed that the expression level of E-cadherin was lower in the GR-HPCCs, compared with the HPCCs, and the expression of α-SMA was higher, indicating that EMT was in the active state in the GR-HPCCs. In addition, the expression levels of TWIST1 and Slug were higher in the GR-HPCCs, compared with the HPCCs, and all alterations were inhibited by sclareolide (Fig. 3A). Several studies have reported that TWIST1 and Slug are mediators of EMT as transcriptional factors, and the expression levels of TWIST1 and Slug have been found to be higher in several cancer EMT phenotypes (16,17). In the present study, on detecting the GR-HPCC EMT phenotype through knocking down TWIST1 and Slug by siRNA in the GR-Panc-1 cells, the expression of α-SMA was inhibited and the expression of E-cadherin was enhanced. In addition, the expression level of hENT1 was inhibited and that of RRM1 was enhanced (Fig. 3B and C). Additionally, the cell death ratio was increased in GR-Panc-1 cells induced by gemcitabine following the knock down of TWIST1 and Slug (Fig. 3D and E). The invasive ability of the GR-HPCCs was more marked, compared with that of the HPCCs, and sclareolide suppressed the invasive ability of the GR-HPCCs (Fig. 3F). These results indicated that sclareolide suppressed the GR-HPCC EMT phenotype and resensitized GR-HPCCs to gemcitabine through the TWIST1 and Slug pathway.

Several studies have reported that NICD and Gli1 can mediate TWIST1 and Slug (18–20). In order to fully understand the upstream mediators of TWIST1 and SLUG, the expression levels of NICD and Gli1 were detected in HPCCs and GR-HPCCs, which were treated with gemcitabine alone or combined with sclareolide. The results showed that the expression levels of NICD and Gli1 in the GR-HPCCs were higher, compared with those in the HPCCs, and sclareolide inhibited the expression of NICD and Gli1 (Fig. 4A). In the GR-Panc-1 cells, the expression levels of TWIST1, Slug and RRM1 were inhibited, and the expression of hENT1 was enhanced. Treatment with gemcitabine combined with sclareolide induced an increase in the GR-HPCC death ratio when NICD and Gli1 were knocked down by siRNA (Fig. 4B-E). These results suggested that sclareolide resensitized the GR-HPCCs to gemcitabine through the NICD/Gli1 pathway.

In order to clarify the mechanism of sclareolide-stimulated gemcitabine-induced cell death, the apoptosis of GR-HPCCs treated with gemcitabine alone or with sclareolide was detected using Annexin V-FITC and PI. Additionally, the markers of apoptosis, poly (ADP-ribose) polymerase (PARP) and Capase-3 were detected. When the GR-HPCCs were treated with gemcitabine and sclareolide, the apoptotic ratio was higher, and the expression levels of PARP and Capase-3 were higher, compared with the GR-HPCCs treated with gemcitabine alone (Fig. 5A and B). These results showed that the mechanism of sclareolide-stimulated gemcitabine-induced cell death involved stimulating apoptosis. In order to confirm the effect in vivo, nude mice were used to establish a pancreatic tumor xenograft model with Panc-1 or GR-Panc-1 cells to examine the effect of co-treatment with gemcitabine and sclareolide. Through intraperitoneal injection, gemcitabine and sclareolide were administered three times each day, with the sclareolide administered 2 h prior to gemcitabine administration. As expected, following administration with gemcitabine, the volume of tumors was lower in the nude mice xenografted with Panc-1 cells, compared with the nude mice xenografted with GR-Panc-1 cells. In the nude mice xenografted with GR-Panc-1 cells, the volume of tumors was lower following the co-administration of gemcitabine and sclareolide, compared with that following administration of gemcitabine alone (Fig. 5C). The results of the TUNEL assay showed that co-administrating gemcitabine and sclareolide induced apoptosis and inhibited tumor growth in the xenografted GR-Panc-1 model (Fig. 5D).