Gemcitabine Enhances Kras-MEK–Induced Matrix Metalloproteinase-10 Expression Via Histone Acetylation in Gemcitabine-Resistant Pancreatic Tumor-initiating Cells

Abstract

Objectives

Advanced pancreatic ductal adenocarcinoma exhibits resistance to systemic chemotherapy, leading to a poor prognosis. We previously isolated KMC07, a human pancreatic tumor-initiating cell line, from a patient who had developed resistance to gemcitabine chemotherapy. To enhance the anticancer effects of gemcitabine, we investigated the molecular mechanism underlying the resistance of KMC07 cells to this drug.

Methods

KMC07 cells were treated with gemcitabine, and subsequently, gene expression and functional analyses were performed using microarray technology, quantitative polymerase chain reaction, immunoblotting, immunohistochemistry, chromatin immunoprecipitation, and cell transplantation into nude mice.

Results

KMC07 cells, but not BxPC-3, PANC-1, MIA PaCa-2, or AsPC-1 cells, expressed matrix metalloproteinase-10 mRNA, and its expression level was increased by gemcitabine treatment. KMC07 cells were found to harbor a constitutively active Kras mutation, and a MEK inhibitor suppressed matrix metalloproteinase-10 mRNA expression. Gemcitabine enhanced histone H3 acetylation at the matrix metalloproteinase-10 promoter, and a histone acetyltransferase inhibitor reduced the gemcitabine-enhanced matrix metalloproteinase-10 mRNA expression. Gemcitabine induced the expression of matrix metalloproteinase-10 protein in KMC07-derived pancreatic tumors in vivo.

Conclusions

We demonstrated the constitutive activation of the Kras-MEK-matrix metalloproteinase-10 signaling pathway in KMC07 cells, and this pathway was further enhanced by gemcitabine through histone acetylation. Our findings may provide novel insights into gemcitabine-based treatment strategies for gemcitabine-resistant pancreatic ductal adenocarcinoma.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is currently the fourth leading cause of cancer-related mortality. Gemcitabine, a nucleoside analogue, is the established standard treatment for advanced PDAC. The FOLFIRINOX treatment protocol has shown improved overall survival compared to gemcitabine alone but requires carefully selected patients with good performance status and without several contraindications. Therefore, gemcitabine remains a key drug for treatment, and understanding the molecular mechanisms related to gemcitabine resistance is crucial for improving treatment outcomes and increasing overall survival.

A constitutively active (CA) mutation in Kras is found in almost 95% of PDAC cases. In preclinical models, the Kras CA mutation induces the formation of premalignant lesions in the pancreas and, in combination with a p53 mutation or chronic inflammation, can induce PDAC. Because the Kras CA mutation plays a central role in the development of PDAC, small molecule inhibitors targeting Kras signaling molecules have been developed. However, recent clinical application has highlighted that many of these inhibitors activate resistance-causing feedback mechanisms and raise concerns regarding their clinical toxicity.

Extracellular matrix-related proteins are involved in the proliferation and metastasis of PDAC. Matrix metalloproteinase-10 (MMP-10) expression is induced in pancreatic cancer, and its inhibition results in the suppression of metastasis and invasion. MMP-10 expression is regulated by Kras signaling in lung cancer stem cells and by histone acetylation in vascular endothelial cells. However, the molecular mechanisms involving MMP-10 that underlie gemcitabine-resistant PDAC remain unclear. We previously isolated gemcitabine-resistant pancreatic tumor-initiating cell lines, designated as KMC cell lines, from advanced PDAC patients who were resistant to gemcitabine chemotherapy. In this study, we examined the molecular mechanisms of gemcitabine resistance using one of these cell lines, KMC07.

Materials and Methods

KMC07, Stromal, and Pancreatic Cancer Cell Lines

The human pancreatic tumor-initiating cell line KMC07 was isolated from a patient with gemcitabine-resistant PDAC. The mouse stromal cell line PA6 was maintained in $\alpha$-minimum essential medium containing 10% fetal calf serum. KMC07 cells were co-cultured with PA6 cells in serum-free Stem medium containing 0.1 $\mu$M 2-mercaptoethanol, 50 U/mL of penicillin, and 50 $\mu$g/mL of streptomycin at 37°C in a humidified atmosphere containing 5% CO2. The human pancreatic cancer cell lines BxPC-3, PANC-1, MIA PaCa-2, and AsPC-1 were obtained from the American Type Culture Collection.

Engraftment of KMC07 Cells

Approximately 8-week-old male nude mice were used in this study. All mice were housed and used under approved protocols in accordance with the Kobe University guidelines for the care and use of laboratory animals. KMC07 cells co-cultured with PA6 cells were harvested, dissociated, and suspended in 1 mL of phosphate-buffered saline (PBS). Nonspecific antibody binding was blocked using purified rat anti-mouse CD16/CD32 monoclonal antibodies and a human FcR blocking reagent for 15 minutes on ice. KMC07 cells, after blocking, were separated from mouse PA6 cells expressing mouse PDGFR$\beta$ by MACS Separator using a biotin-conjugated anti-mouse PDGFR$\beta$ monoclonal antibody and anti-biotin Microbeads according to the manufacturers’ protocols.

The purity of the separated KMC07 cell population ranged from 95% to 98%, as evaluated by FACS analyses. After a small left abdominal flank incision was made, 200 $\mu$L of the separated KMC07 cells (1 x 10$^6$ KMC07 cells in 200 $\mu$L of serum-free Stem medium) was injected into a region of the pancreas tail using a 1 mL disposable syringe attached to a 30-gauge needle. The abdominal wound was closed with an Auto-clip. After 8 weeks, the mice were euthanized under anesthesia. Tissue samples were fixed in 10% phosphate-buffered formalin overnight and embedded in paraffin. For gemcitabine treatment experiments, 8 weeks after injection of KMC07 cells into subcutaneous fat or a region of the pancreas tail, 120 mg/kg of gemcitabine or saline (for controls) was injected intraperitoneally twice a week for 4 weeks. The mice were then euthanized under anesthesia, and tissue samples were fixed in 10% phosphate-buffered formalin overnight and embedded in paraffin.

Immunohistochemistry

Immunohistochemical staining was performed as previously described. Immunostaining to detect MMP-10 expression in human tissues was performed on 3-$\mu$m sections from formalin-fixed, paraffin-embedded tissues placed on coated glass slides and dried at room temperature overnight. Sections were dewaxed in xylene and rehydrated according to standard procedures. For antigen retrieval, the tissue sections were boiled in 1x Target Retrieval Solution pH 9.0 for 40 minutes. The samples were cooled at room temperature for 20 minutes and rinsed with distilled water three times, followed by peroxidase block with 3% H2O2 in methanol for 5 minutes. After rinsing twice with distilled water, the samples were immersed in Tris-buffered saline-Tween20 (TBST) (25 mM Tris–HCl [pH 7.4], 75 mM NaCl, and 0.1% Tween20) for 5 minutes and then incubated with a primary antibody (a rabbit anti-human MMP-10 polyclonal antibody diluted 1:250 in Dako Real antibody diluent) at 4 °C overnight. The samples were rinsed three times with TBST. Primary antibody detection was performed with rabbit horseradish peroxidase polymer probe at room temperature for 30 minutes, followed by two rinses with TBST according to the manufacturer’s instructions. The signal was developed with diaminobenzidine for 10 minutes. The samples were rinsed with distilled water three times, counterstained with hematoxylin for 1 minute, and dehydrated in alcohol solution and xylene. Parallel sections were stained with hematoxylin and eosin for identification of cancerous and normal tissues.

Microarrays

KMC07 cells co-cultured with PA6 cells in a 10-cm Corning culture dish were incubated with 2 $\mu$M gemcitabine or PBS. After 48 hours, RNAs were extracted using an RNAeasy mini kit and subjected to microarray analyses as previously described.

Reverse Transcription-Polymerase Chain Reaction and Quantitative Real Time-Polymerase Chain Reaction Assays

Reverse transcription-polymerase chain reaction (RT-PCR) was performed as previously described. Quantitative real-time (qRT)-PCR was performed with a QuantiTect SYBR Green RT-PCR kit using the Applied Biosystems 7500 Real-Time PCR System according to the manufacturers’ protocols. Quantification of MMP10 mRNA expression was determined by the relative standard curve method with human $\beta$-actin as an endogenous control. The amount of MMP10 promoter genomic DNA immunoprecipitated in chromatin immunoprecipitation (ChIP) assays was quantified using PCR. The primer sequences used were: human-$\beta$-actin-F/104: 5′-AGCCTCGCCTTTGCCGATCC-3′, human-$\beta$-actin-R/104: 5′-TTGCACATGCCGGAGCCGTT-3′, human-MMP-10-F/119: 5′-AGTTTGGCTCATGCCTACCC-3′, human-MMP-10-R/119: 5′-TCATGAGCAGCAACGAGGAA-3′, human-MMP-2-F/222: 5′-CCGCCTTTAACTGGAGCAAA-3′, human-MMP-2-R/222: 5′-TTTGGTTCTCCAGCTTCAGG-3′, human-MMP-10 promoter-F/149: 5′-ACCAAGCTTGTCA-GCTCTCTTT-3′, human-MMP-10 promoter-R/149: 5′-CAGCCTACATC-AGTATTTTCCTTCA-3′.

Kras Mutation Analyses

Kras and Nras mutations (codons: 12, 13, 59, 61, 117, 146) were analyzed by the PCR-reverse sequence specific oligonucleotide (PCR-rSSO) method as previously described. Briefly, genomic DNA was obtained from xenograft tumor tissue samples. Tissue samples were fixed in 10% phosphate-buffered formalin overnight and embedded in paraffin. Ten-micron sections were prepared using formalin-fixed, paraffin-embedded tissues placed on coated glass slides and dried at room temperature overnight. The samples were sent to a research institution for analysis. Genomic DNA was isolated from the samples, and 50 ng of template DNA was amplified by PCR using a biotin-labeled primer. The PCR products and fluorescent beads (oligonucleotide probes complementary to wild and mutant genes were bound to the beads) were hybridized and labeled with streptavidin-phycoerythrin. The products were subjected to PCR-rSSO assays, and collected data were analyzed using specific software.

Immunoblotting

KMC07 cells co-cultured with PA6 cells in a 6-well Costar cell culture plate were treated with 0.125 $\mu$M gemcitabine for 48 hours. Cells were washed twice with 3 mL of ice-cold PBS and resuspended with 200 $\mu$L of Laemmli sample loading buffer (2% SDS, 20 mM dithiothreitol, 10% glycerol, 62.5 mM Tris-HCl [pH 6.8], and 0.002% bromophenol blue). The samples were sonicated for 10 seconds and boiled for 5 minutes. The samples were subjected to SDS-PAGE, along with molecular weight markers, using e-PAGEL 5%–20% gel. Proteins were transferred from the gel to a nitrocellulose membrane for 7 minutes using a specific gel transfer system according to the manufacturers’ protocols.

The blotted membrane was rinsed with PBS for 5 minutes and blocked with 5% skim milk in TBST at room temperature for 1 hour. The membrane was washed three times with 25 mL of TBST for 5 minutes and incubated with a primary antibody (a rabbit anti-human MMP-10 polyclonal antibody diluted 1:100 in 5% bovine serum albumin [BSA] in TBST) at 4 °C overnight. The membrane was rinsed three times with TBST for 5 minutes and incubated with a secondary antibody (an horseradish peroxidase [HRP]-conjugated goat anti-rabbit IgG [H + L] diluted 1:20,000 in TBST with 5% BSA) and StrepTactin-HRP conjugate diluted 1:20,000 in TBST with 5% BSA at room temperature for 1 hour.

The membrane was rinsed five times with TBST for 5 minutes and incubated with a chemiluminescent substrate according to the manufacturer’s protocol. The chemiluminescent signals were captured using a CCD camera-based imager and analyzed using specific software. To detect human CoxIV (3E11) protein as an endogenous control, the blotted membrane was incubated with an HRP-conjugated rabbit anti-human CoxIV monoclonal antibody diluted 1:1000 in TBST with 5% BSA at room temperature for 1 hour.

ChIP Analyses

Chromatin immunoprecipitation (ChIP) assays were performed using a specific kit according to the manufacturer’s protocol. Briefly, KMC07 cells were co-cultured with PA6 cells in Stem medium in a 10-cm Corning culture dish and treated with 0.125 $\mu$M gemcitabine or PBS for 48 hours. A 10% formalin solution was added to each dish, and cells were incubated at 37°C for 10 minutes. Cells were washed twice with ice-cold PBS containing a protease inhibitor cocktail, harvested using a cell scraper into a 1.5-mL conical tube, and pelleted by centrifugation at 2000g at 4°C for 4 minutes. Cells (approximately 1 x 10$^6$ cells) were resuspended in SDS Lysis Buffer containing a protease inhibitor cocktail and incubated for 10 minutes on ice.

The lysate was sonicated six times using an ultrasonic disrupter at 20 Watts for 30 seconds on ice (with 5-minute intervals). The samples were centrifuged at 21,500g at 4°C for 10 minutes, and the supernatants were transferred into a 2-mL conical tube and diluted with ChIP Dilution Buffer. A 200-$\mu$L aliquot of the diluted sample was transferred to another tube as an “input sample.” The remaining samples were incubated with Salmon Sperm DNA/Protein A Agarose-50% Slurry at 4°C for 30 minutes with rotation and centrifuged at 1000g at 4°C for 1 minute. The supernatant was transferred to a conical tube and incubated with an anti–acetyl-Histone H3 polyclonal antibody at 4°C overnight with constant rotation. Sixty microliters of Salmon Sperm DNA/Protein A Agarose-50% Slurry was added to the conical tube and incubated at 4°C for 1 hour with constant rotation.

After centrifugation at 1000g at 4°C for 1 minute, the supernatant was removed, and the Protein A agarose was sequentially washed with Low Salt Immune Complex Wash Buffer, High Salt Immune Complex Wash Buffer, LiCl Immune Complex Wash Buffer, and TE Buffer at 4°C. The Protein A agarose was incubated with 250 $\mu$L of elution buffer (1% SDS and 0.1 M NaHCO3) for 15 minutes and centrifuged at 1000g for 1 minute at room temperature. The supernatant was transferred to another tube, and this elution procedure was repeated. The combined supernatant (total 500 $\mu$L), “elution sample,” and “input sample” (total 200 $\mu$L) were mixed with 5 M NaCl (final concentration 0.2 M) and incubated at 65°C for 4 hours.

After incubation, the samples were mixed with reaction buffer (final concentration: 36 mM Tris-HCl [pH 6.5], 18 mM ethylenediaminetetraacetic acid, 0.18 M NaCl, and 0.36 g/L Proteinase K) and incubated further at 45°C for 1 hour. The samples were treated with an equal volume of phenol/chloroform/isoamyl alcohol mixture, pH 7.9, and precipitated with ethanol and linear polyacrylamide. The precipitates were washed with 70% ethanol and dissolved in 20 $\mu$L of DNase-free water, followed by qRT-PCR using human MMP-10 promoter primers.

Statistical Analyses

Results for continuous variables were expressed as the mean ± standard error (SE). Statistically significant differences were determined by the Student’s t-test. Significance was defined as a p-value less than 0.05. This study was performed according to Institutional Review Board-approved guidelines at Kobe Medical Center and Kobe University School of Health Sciences and approved by the Ethics Committees of these institutions.

RESULTS

MMP-10 mRNA Expression Was Upregulated by Gemcitabine in KMC07 Cells

We previously isolated and characterized seven pancreatic tumor-initiating cell lines from seven independent PDAC patients who exhibited resistance to gemcitabine treatment. The KMC07 patient received gemcitabine treatment for three years. During the first two years, his liver metastases disappeared, and his pancreatic tumor size decreased (partial response). However, after this initial response, the liver metastases slowly recurred, accompanied by peritoneal dissemination in the final year of treatment.

The KMC07 cell line was derived from this patient, suggesting that these cells might have acquired or inherently possessed gemcitabine resistance. Engraftment of KMC07 cells into the pancreatic tail or subcutaneous fat of nude mice resulted in tumor formation, along with liver and lung metastases. To identify genes in KMC07 cells that are upregulated in response to gemcitabine, KMC07 cells were treated with or without gemcitabine and then subjected to microarray analysis. We found that MMP-10 exhibited the highest induction on the array, followed by synaptotagmin XVI and transmembrane protein 40.

In this study, we characterized the interaction between MMP-10 mRNA upregulation and gemcitabine resistance. We examined MMP-10 mRNA expression levels in KMC07 cells and various cultured pancreatic cancer cell lines. KMC07 cells, but none of the other cultured cell lines, expressed MMP-10 mRNA, and its expression was upregulated upon gemcitabine treatment. KMC07 cells did not express MMP-2 mRNA, a gene known to be expressed by PANC-1 cells. Gemcitabine upregulated MMP-10 mRNA expression in KMC07 cells in a dose-dependent manner. We also confirmed that gemcitabine upregulated MMP-10 protein expression in KMC07 cells.

Requirement of Histone Acetylation for Gemcitabine-Dependent MMP-10 mRNA Expression

We investigated whether histone acetylation was involved in the gemcitabine-enhanced MMP-10 expression in KMC07 cells. Gemcitabine-enhanced MMP-10 mRNA expression was reduced by C646, a histone acetyltransferase inhibitor. Trichostatin A (TSA), a histone deacetylase (HDAC) inhibitor, enhanced MMP-10 mRNA expression, while C646 reduced TSA-enhanced MMP-10 mRNA expression. TSA-enhanced MMP-10 mRNA expression was also reduced by U0126, a MEK inhibitor. Chromatin immunoprecipitation (ChIP) assays revealed that gemcitabine enhanced the histone H3 acetylation level at the MMP-10 promoter. These results demonstrate that histone acetylation is required for the gemcitabine-induced upregulation of MMP-10 mRNA expression.

Induction of MMP-10 Protein Expression in KMC07-Derived Tumor by Gemcitabine In Vivo

Gemcitabine reduced the growth rate of KMC07-derived subcutaneous tumors in nude mice. To test whether gemcitabine upregulated MMP-10 expression in KMC07-derived tumors, KMC07 cells were injected into the pancreatic tail region of nude mice, and gemcitabine was subsequently administered intraperitoneally. The growth rate of KMC07-derived orthotopic pancreatic tumors was also reduced in the presence of gemcitabine. Histological analysis showed that gemcitabine enhanced the invasion of KMC07-derived tumor cells into normal pancreas tissues and partially upregulated MMP-10 expression. In contrast, in the absence of gemcitabine, KMC07-derived tumor cells grew expansively within the fibrous capsule. These results suggest that gemcitabine upregulates MMP-10 protein expression in KMC07-derived orthotopic pancreatic tumors.

DISCUSSION

This study provides the first demonstration that the gemcitabine-enhanced Kras$^{G12V}$-MEK-MMP-10 signaling pathway is activated in gemcitabine-resistant PDAC both in vivo and in vitro. This suggests that the acquisition of gemcitabine resistance may activate the Kras$^{G12V}$ signaling pathway, which is typically inactivated in naive PDAC.

The KMC07 cell line characterized in this study is one of seven KMC cell lines that we have previously reported. The other six KMC cell lines also expressed MMP-10 mRNA at levels similar to that of the KMC07 cell line in the absence of gemcitabine. MMP-10 has been shown to be required for the invasion and metastasis of pancreatic cancer. MMP-10, in cooperation with MMP-1, cleaves collagen type 1, leading to the reorganization of the extracellular matrix. MMP-1 is also required for Kras CA mutant-regulated invasion of human pancreatic cancer cells via ERK2. Taken together, these results suggest that KMC07 cells reorganize the extracellular matrix through Kras$^{G12V}$-induced activation of MMP-10, followed by invasion and metastasis.

MIA PaCa-2, PANC-1, and AsPC-1 cells, but not BxPC-3 cells, express the Kras CA mutant. However, these cell lines did not express MMP-10 mRNA. If the presence of the Kras CA mutation were sufficient to induce MMP-10 mRNA expression, these cultured cell lines would express MMP-10 mRNA. Therefore, our results indicate that a Kras CA mutation is necessary but not sufficient for MMP-10 induction and that other unknown factors might be required for Kras CA mutant-MEK-dependent induction of MMP-10.

A low level of histone acetylation in pancreatic cancer is a predictor of poor survival, and HDAC inhibitors are currently being evaluated in clinical trials. In contrast, we showed that gemcitabine enhanced histone H3 acetylation levels at the MMP-10 promoter, and this enhancement was reduced by a histone acetyltransferase inhibitor. Moreover, TSA enhanced MMP-10 mRNA expression in KMC07 cells, consistent with a previous report that HDAC-7 repressed MMP-10 gene transcription by associating with myocyte enhancer factor-2. Another group recently showed that gemcitabine treatment increased histone acetylation levels at the promoter of a pro-oncogenic molecule, miR-21, in cultured pancreatic cancer cell lines. Taken together, these results suggest that the effects of histone acetylation on PDAC may be context-dependent. The mechanism by which gemcitabine increases histone H3 acetylation levels at the MMP-10 promoter remains unknown. However, it is possible that gemcitabine blocks DNA replication and leads to a DNA damage response that includes acetylation of histones in a context-dependent manner.

We showed that gemcitabine-enhanced MMP-10 expression was completely inhibited by a MEK inhibitor, U0126. A Notch inhibitor and a TGF-$\beta$ receptor antagonist partially reduced gemcitabine-enhanced MMP-10 mRNA expression. Therefore, the Kras$^{G12V}$-MEK signaling pathway is necessary for MMP-10 expression, and Notch and TGF-$\beta$ signals might function as enhancers for MMP-10 expression. To improve the anticancer effects of gemcitabine on PDAC, the molecular mechanisms of gemcitabine resistance are under intense investigation. NEO2734 Our results demonstrate a unique gemcitabine-resistance mechanism using novel pancreatic tumor-initiating cells and suggest that not only MMP-10 but also other cancer-related genes downstream of Kras$^{G12V}$ signaling pathways may be activated when PDAC becomes resistant to gemcitabine. It is possible that a combination of gemcitabine and MEK inhibitors, as well as MMP-10 inhibitors, might synergistically improve the overall survival of gemcitabine-resistant PDAC patients.