AMD070 trihydrochloride (C₂₁H₃₀Cl₃N₅; CID 11256587) was purchased from Med Chem Express (Cosmo Bio Co., Ltd.), and dimethyl sulfoxide (DMSO) was obtained from Sigma-Aldrich (Merck KGaA). AMD070 solution (326.90 mM) was prepared in DMSO, stored in small aliquots at −20°C, and then thawed and diluted in cell culture medium as required. CXCL12 was purchased from R&D Systems.

The human pancreatic duct epithelial (HPDE) cell line H6c7 was purchased from Kerafast. Human skin fibroblasts (FBs; cat. no. T0904) were purchased from Applied Biological Materials. The human PaCa cell lines AsPC-1, BxPC-3, Capan2, MIA PaCa-2, PANC-1, and SW1990 were purchased from the American Type Culture Collection (ATCC). The H6c7 cell line was maintained in keratinocyte serum-free medium (Gibco/Thermo Fisher Scientific, Inc.). The AsPC-1, BxPC-3, and Capan2 cell lines were maintained in RPMI-1640 medium (Sigma Aldrich; Merck KGaA). The MIA PaCa-2, PANC-1, and SW1990 cell lines and FBs were maintained in Dulbecco's modified Eagle's medium (DMEM; Sigma Aldrich; Merck KGaA). Both RPMI-1640 medium and DMEM were supplemented with 10% fetal bovine serum (FBS; Gibco/Thermo Fisher Scientific, Inc.). All media were supplemented with 10 mg/ml streptomycin, 10,000 U/ml penicillin, and 25 µg amphotericin B (Gibco/Thermo Fisher Scientific, Inc.). All cell lines were cultured at 37°C in a humidified incubator with 5% CO₂.

Radiation-resistant cancer cell lines have been previously established from nasopharyngeal, esophageal, breast, and lung cancers (33–37). Here, we established radiation-resistant PaCa cell lines by referencing these methods. PaCa cell lines (AsPC-1, BxPC-3, MIA PaCa-2, and SW1990) were seeded in 100-mm dishes and cultured. Upon reaching 50% confluence, the cells were irradiated with 2 Gy radiation and incubated until reaching 90% confluence, after which the cells were passaged. With each passage, the irradiation process was repeated until the total radiation dose reached at least 60 Gy. The radiation resistance of the cell lines was assessed by colony formation assay.

Total mRNA from normal and radiation-resistant MIA PaCa-2 cells was isolated using the RNeasy Plus Mini kit (Qiagen, Inc.) according to the manufacturer's instructions. The mRNA microarray experiments were performed by Hokkaido System Science Co., Ltd. Transcripts amplified from total mRNA were hybridized to a SurePrint G3 Human 8×60K v3 array (Agilent Technologies, Inc.) according to the manufacturer's protocol. The results were analyzed using Agilent Genomic Workbench Software (Agilent Technologies, Inc.).

Irradiating a small number of cells with a high radiation dose will kill all cells. Therefore, the number of radiation-resistant PaCa cells to be seeded (200, 400, 1,000, 2,000, or 40,000) depended on the dose (0, 2, 4, 6, or 10 Gy). The cells were seeded in 60-mm dishes, cultured overnight, irradiated with each radiation dose, and cultured at 37°C for 14 days. To assess the efficacy of AMD070, a CXCR4 antagonist, PaCa cells were seeded in 60-mm dishes and treated with 2.5 µM AMD070 for 72 h. The treated cells were seeded in 60-mm dishes, cultured overnight, irradiated with 2 Gy radiation, and cultured at 37°C for 14 days. Cells were fixed and stained using a Diff-Quick cell staining kit (Dade Behring), and colonies were counted under five different fields. A colony was defined as a group of at least 50 cells.

Pancreatic tissues were analyzed from 92 patients who underwent surgery at Nagoya City University Hospital (Nagoya, Japan) between January 2006 and December 2016. The mean age of the patients was 67.6 years (range 32–85 years), and the male-female ratio was 62 males and 30 females. All pancreatic tissues were obtained from patients or their relatives who provided informed consent. This study was conducted upon the approval of the institutional review board established by Nagoya City University (approval no. 60–18-0025, date of approval; May 6, 2018).

Specimens were fixed in 10% formalin and then embedded in paraffin. Specimens were sectioned into 3-µm-thick slices, and the sections were deparaffinized, subjected to autoclave treatment in 10 mM sodium citrate buffer for 10 min at 120°C, and cooled to room temperature. Next, the sections were treated with 0.3% H₂O₂ in methanol for 30 min, blocked with Block Ace (Megmilk Snow Brand; KAC Co., Ltd.) for 10 min, and incubated with anti-CXCR4 antibody (1:250; Proteintech Group; cat. no. 60042-1-Ig) overnight at 4°C, followed by EnVision+ System- HRP Labelled Polymer anti-mouse (DAKO/Agilent Technologies; cat. no. K4001) for 45 min at room temperature. The peroxidase reaction was visualized by incubating the sections with the Liquid DAB+ Substrate Chromogen System (DAKO/Agilent Technologies; cat. no. K3467), followed by hematoxylin counterstaining. Immunohistochemical staining was evaluated as follows. The intensity of CXCR4 immunostaining was graded semi-quantitatively on a 4-point scale (−, +, ++, +++) by three independent observers. Of the 92 patients, 8 had stage 1 disease, 4 had stage 2 disease, 37 had stage 3 disease, 41 had stage 4 disease, and 2 had an unknown disease stage. However, analyses were performed in a blinded manner, and the observer was not aware of the patient's stage and outcome. The concordance rate was greater than 90%. Differences in opinion were resolved by consensus with a fourth evaluator. The cases were classified into a high expression group and a weak expression group according to the intensity of immunostaining in cancer cells, in which an immunostaining score of ++ or +++ was defined as high expression.

PaCa cells (5×10⁴) were seeded in glass chamber slides and cultured overnight. The cells were fixed using 4% paraformaldehyde for 20 min at room temperature. Next, the cells were permeabilized with 0.1% Triton-X for 3 min and incubated with blocking buffer [3% bovine serum albumin in phosphate-buffered saline (FUJIFULM Wako Pure Chemical Corp.)] for 1 h at room temperature. The cells were incubated with anti-CXCR4 antibody (1:200; Abcam; cat. no. ab124824) overnight at 4°C, followed by Alexa Fluor 488 goat anti-rabbit IgG secondary antibody (1:1,000; Abcam; cat. no. ab6939) for 1 h at room temperature. The nuclei were visualized by DAPI staining at room temperature for 10 min. Images of the stained slides were captured using a BZ-X710 fluorescence microscope (Keyence Corporation) at ×200 magnification.

Total RNA was isolated from HPDE and PaCa cells using an RNeasy Plus Mini kit (Qiagen GmbH), according to the manufacturer's protocols, and quantified using a NanoDrop 1000 (Thermo Fisher Scientific, Inc.). Total RNA (1 µg) was reverse transcribed using Super Script III First-Strand Synthesis Super Mix for RT-qPCR (Invitrogen/Thermo Fisher Scientific, Inc.) following the manufacturer's protocols. RT-qPCR was performed using TaqMan Fast Advanced Master Mix and TaqMan Gene Expression Assays for CXCR4 (Hs00607978_s1) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Hs99999905_m1) on a 7900HT Fast Real-Time PCR System (all from Applied Biosystems/Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used: initial denaturation at 95°C for 20 sec, followed by 40 cycles at 95°C for 1 sec and 60°C for 20 sec. The expression level of CXCR4 was reported relative to that of GAPDH in each sample, using the relative standard curve method (38).

Proteins were extracted from cells using radioimmunoprecipitation lysis buffer containing Protease Inhibitor Single Use Cocktail and Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific, Inc.). The protein concentrations were measured using a Pierce BCA protein assay kit (Thermo Fisher Scientific, Inc.). Protein extracts (20 or 30 µg) were denatured at 90°C for 5 min and separated on 10% Mini-PROTEAN TGX Precast gels (Bio-Rad Laboratories). The protein bands were transferred to nitrocellulose membranes and blocked in iBind Flex Solution (iBind Flex Buffer, iBind Flex Additive, and distilled water; Thermo Fisher Scientific, Inc.) for 20 min at room temperature. The primary and secondary antibody reactions were performed for 3 h at room temperature using the iBind Flex Western System (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The membranes were incubated with anti-CXCR4 (1:1,000; Proteintech Group; cat. no. 60042-1-Ig) or anti-GAPDH (1:1,000; Santa Cruz Biotechnology; cat. no. SC-47724) primary antibodies, followed by horseradish peroxidase-conjugated goat anti-mouse polyclonal secondary antibodies (1:2,000; DAKO/Agilent Technologies; cat. no. P0447). The protein-antibody complexes were visualized using SuperSignal West Femto Chemiluminescent Substrate or Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific, Inc.). The immunoreactive protein bands were detected using an Amersham Imager 600 (Cytiva), and the densities of the detected bands were calculated using ImageJ software 1.52v (National Institutes of Health).

CXCR4 small interfering RNA (siRNA; s15412: CCUGUUUCCGUGAAGAAAA) and nontargeting negative control siRNA (Silencer Select Negative Control No. 1; cat. no. 4390843: sequence not provided) were predesigned siRNAs purchased from Thermo Fisher Scientific, Inc. PaCa cells were seeded at 2.5×10⁵ cells/well in 6-well plates, cultured overnight, and then transfected with siRNA. According to the manufacturer's instructions, siRNAs and Lipofectamine RNAiMAX (Invitrogen/Thermo Fisher Scientific, Inc.) were mixed with Opti-MEM (Invitrogen/Thermo Fisher Scientific, Inc.) and incubated for 5 min at room temperature. The siRNA-lipid complex was diluted in DMEM to achieve a final siRNA concentration of 10 nM. Cells were incubated for 48 h in a 5% CO₂ incubator at 37°C.

In vitro invasion assays were performed using Corning BioCoat Matrigel Invasion Chambers (Corning, Inc.) according to the manufacturer's protocol. Normal and radiation-resistant PaCa cells (1×10⁵) were seeded in the upper chamber, which contained DMEM without FBS. The chemoattractant used in the lower chamber was 10% FBS in DMEM. In addition, AMD070 (1 µM) and CXCL12 (100 ng/ml) were added to the lower chamber, or the cells were cocultured with FBs. After incubation for 24 h, the upper surface of the upper chambers was wiped with a cotton swab, and the invading cells were fixed and stained using a Diff-Quick cell staining kit (Dade Behring). The number of cells in nine random microscopic fields (×200 magnification) was counted.

Differences between two samples were analyzed using unpaired t-tests. Multiple group comparisons were performed by one-way analysis of variance with the post-hoc Bonferroni test for subsequent comparisons of individual groups. Comparisons of groups with two independent variables were performed using two-way analysis of variance. Comparisons of patient stage were performed using Fisher's exact test. Survival curves based on CXCR4 expression were generated using the Kaplan-Meier method and were compared using log-rank tests. Results with a P-value <0.05 were considered statistically significant. The data from experiments performed in at least triplicate are expressed as means ± standard deviations.

Resected tissue specimens from patients with PaCa were subjected to CXCR4 immunostaining (Fig. 1A). The patients were divided into low and high CXCR4 expression groups according to the intensity of CXCR4 immunostaining, and survival curves were generated. Of the 29 patients in the high expression group, 2 had stage 1 disease, none had stage 2 disease, 9 had stage 3 disease, 17 had stage 4 disease, and 1 had an unknown disease stage. Of the 63 patients in the low expression group, 6 had stage 1 disease, 4 had stage 2 disease, 28 had stage 3 disease, 24 had stage 4 disease, and 1 had an unknown disease stage. There were no differences in staging between the high and low expression groups (P=0.233). Overall survival (OS) was significantly worse in the high expression group (P=0.0068; Fig. 1B).

Expression of CXCR4 in H6c7 HPDE cells, which are derived from the near normal pancreatic duct epithelium, and in PaCa cell lines (AsPC-1, BxPC-3, Capan2, MIA PaCa-2, PANC-1, and SW1990) was evaluated by RT-qPCR and western blotting. Both CXCR4 mRNA (Fig. 2A) and protein (Fig. 2B and 2C) levels were significantly higher in PaCa cell lines than in HPDE cells (P<0.05).

We succeeded in establishing radiation resistance in two PaCa cell lines, MIA PaCa-2 and SW1990. MIA PaCa-2 and SW1990 cells were irradiated with 120 and 60 Gy, respectively. The decrease in colonization after exposure to high doses of radiation was significantly attenuated in the radiation-resistant cells compared with that in their normal counterparts (P<0.05; Fig. 3A-D).

To investigate comprehensive differences in cDNA expression between normal and radiation-resistant MIA PaCa-2 cells, we used a cDNA microarray containing 62,976 probe sets. Of these probes, 2,397 had higher expression (cut-off value, 2-fold) and 2,154 had lower expression (cut-off value, 0.5-fold) in radiation-resistant cells compared with that in normal MIA PaCa-2 cells. Among stem cell markers of PaCa and chemokine receptors, CXCR4 showed the highest expression level (Table I).

Immunofluorescence staining of CXCR4 in normal and radiation-resistant PaCa cell lines confirmed the higher expression of CXCR4 in radiation-resistant PaCa cell lines (Fig. 4A). RT-qPCR and western blotting further confirmed that CXCR4 mRNA (Fig. 4B) and protein (Fig. 4C-F) levels were significantly increased in radiation-resistant PaCa cell lines compared with those in normal PaCa cell lines (P<0.05).

RT-qPCR was performed to evaluate changes in the expression of CXCR4 mRNA in PaCa cell lines transfected with CXCR4 siRNA. Transfection with CXCR4 siRNA significantly downregulated CXCR4 in PaCa cell lines compared with that in control cells and cells transfected with negative control siRNA (P<0.05; Fig. S1).

There were no significant differences in cell invasion ability between the negative control group and the CXCR4-knockdown group. Addition of CXCL12 enhanced the invasion ability of PaCa cell lines, and this effect was suppressed by CXCR4 knockdown (Fig. S2).

Cell invasion ability was greater in radiation-resistant PaCa cell lines (MIA PaCa-2 and SW1990) than in normal PaCa cell lines. The addition of CXCL12 enhanced the invasion ability of PaCa cell lines, and this effect was suppressed by AMD070 (AMD), which has been reported to act as a CXCR4 antagonist. Similarly, coculture with FBs enhanced the invasion ability of PaCa cell lines, and this increase was suppressed by AMD070 (Fig. 5A-D).

Irradiation (2 Gy) significantly suppressed the colonization of both normal and radiation-resistant MIA PaCa-2 cells (P<0.05). AMD070 treatment significantly suppressed the colonization of radiation-resistant MIA PaCa-2 cells after irradiation (2 Gy) compared with cells treated with irradiation alone (P<0.05; Fig. 6A and B).