nab-Paclitaxel potentiates gemcitabine activity by reducing cytidine deaminase levels in a mouse model of pancreatic cancer

Abstract

Nanoparticle albumin-bound (nab)-paclitaxel, an albumin-stabilized paclitaxel formulation, demonstrates clinical activity when administered in combination with gemcitabine in patients with metastatic pancreatic ductal adenocarcinoma (PDA). The limited availability of patient tissue and exquisite sensitivity of xenografts to chemotherapeutics have limited our ability to address the mechanistic basis of this treatment regimen. Here, we used a mouse model of PDA to show that the coadministration of nab-paclitaxel and gemcitabine uniquely demonstrates evidence of tumor regression. Combination treatment increases intratumoral gemcitabine levels attributable to a marked decrease in the primary gemcitabine metabolizing enzyme, cytidine deaminase. Correspondingly, paclitaxel reduced the levels of cytidine deaminase protein in cultured cells through reactive oxygen species-mediated degradation, resulting in the increased stabilization of gemcitabine. Our findings support the concept that suboptimal intratumoral concentrations of gemcitabine represent a crucial mechanism of therapeutic resistance in PDA and highlight the advantages of genetically engineered mouse models in preclinical therapeutic trials.

Significance: This study provides mechanistic insight into the clinical cooperation observed between gemcitabine and nab-paclitaxel in the treatment of pancreatic cancer.

Figure 1. nab-paclitaxel slows tumor growth, improves survival, and decreases metastasis

A) The percentage of mice that survived for 8 days, exhibited at least one metastasis, or developed ascites was quantified. Analysis of terminal blood draws were used to measure white blood cell count, neutrophil/granulocyte count, platelet count, and hemoblogin. Normal ranges for healthy littermate non-tumor bearing mice as well as untreated KPC mice are listed. NA = not applicable, ND = not determined. B) Liver metastasis score was quantified by factoring the number and size of metastases throughout the liver. Please see materials and methods for additional information. (n≥9) C) Waterfall plot of tumor response of individual tumors from each cohort. nab-paclitaxel monotherapy is significantly better than vehicle, but not gemcitabine (p=0.006 and p=0.120, respectively).

Figure 2. nab-paclitaxel targets the tumor epithelial cells

A) 8 KPC cell lines were exposed to a dose range of paclitaxel, nab-paclitaxel, docetaxel, or gemcitabine for 3 days to determine the GI₅₀ of each agent. Data is representative of four independent experiments. B) 3 KPC cell lines were exposed to sub-GI₅₀ levels of agents. Cells were pre-treated with DMSO or 10μM paclitaxel for 24 hours and/or treated with 30nM gemcitabine for 2 days. Data is representative of two independent experiments. The dotted lines represent predicted additive effect of combination therapy. Proliferation (C) and apoptosis (D) in tumors was measured via quantitative immunohistochemistry for Ki67 and cleaved caspase 3, respectively. (n=8) E) 10-20 high powered fields per tumor were quantified by performing co-immunofluorescence for cleaved caspase 3 and E-cadherin. (n≥9)

Figure 3. nab-paclitaxel promotes elevated intratumoral gemcitabine levels

A) The dFdC:dFdU ratio in bulk tumor was quantified in mice 2 hours after the last dose of gemcitabine. (n≥12) B) Intratumoral levels of dFdCTP were measured in duplicate samples from mice in each cohort 2 hours after the last dose of gemcitabine. (n≥12) C) Intratumoral levels of paclitaxel were measured in samples from mice in each cohort 4 hours after the last dose of nab-paclitaxel. (n≥7) D) 2 KPC cell lines were pre-treated with 10μM paclitaxel or DMSO for 36 hours or 10μM THU (cytidine deaminase inhibitor) for 30 minutes as a positive control and incubated with 1μM gemcitabine for 2 hours. dFdCTP levels were then measured. Data is representative of three independent experiments.

Figure 4. nab-paclitaxel and paclitaxel destabilizes cytidine deaminase protein

A) 40μg of bulk tumor cell lysates were immunoblotted for indicated proteins. B) Immunohistochemistry for cytidine deaminase reveals reduced protein levels in tumour epithelial cells. Scale bar = 50 μm. (n=8) C) RNA isolated from 5 KPC cell lines treated for 36 hours with 10μM paclitaxel was subjected to qRT-PCR, revealing no alterations in mRNA levels compared to controls. RQ values were generated using actin as an endogenous control. D) Protein lysates were generated from the same 5 KPC cell lines treated for 36 hours with DMSO or 10μM paclitaxel and immunoblotted for indicated proteins. Data is representative of four independent experiments. E) Protein lysates were generated from KPC cells pre-treated for 36 hours with DMSO or 10μM paclitaxel followed by 10μM MG132 for 0, 3, 10, or 30 minutes.

Figure 5

Paclitaxel inactivates cytidine deaminase through induction of reactive oxygen species (ROS). KPC cells were pretreated with 10μM paclitaxel and/or 5mM N-acetylcysteine (NAC) for 4 hours and A) incubated with CM-H2DCFDA to assess intracellular ROS via flow cytometry (n=3) or B) assessed for intracellular redox state via GSH-Glo. (n=3) C) Protein lysates were generated from KPC cells treated for 36 hours with 10μM paclitaxel and/or 5mM NAC and immunoblotted for indicated proteins. D) KPC cells were pretreated with 10μM paclitaxel and/or 5mM NAC for 36 hours and incubated with 1μM gemcitabine for 1 hour. Intracellular dFdCTP was measured (n=3).