Servicios personalizados
Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Citado por SciELO 
Citado por Google 
Similares en
SciELO 
Similares en Google 
Permalink
Highlights
Recurrence and chemoresistance in pancreatic cancer are largely related to cancer stem cells. The aim of this study was to determinate the effect of different chemotherapy protocols used in the clinic on tumor stem cell markers in vitro. The results show that gemcitabine monotherapy and oxaliplatin monotherapy increase these markers while the combination of 5-fluorouracil and irinotecan (FOLFIRI) decreases them. The expression of the different markers also changes depending on the treatment used.
Introduction
The current treatment of pancreatic cancer, the 9th most frequent and 4th deadliest type of cancer, remains ineffective. Although surgical resection can be applied when the tumor is well defined (20% of cases)1, most patients with pancreatic cancer are treated with chemotherapy. Pancreatic cancer has a highly aggressive nature and, in most cases, it is diagnosed in advanced stages of the disease (III and IV). In stage IV, even with chemotherapeutic treatment, usually based on Gemcitabine-Abraxane (GEM-ABRAX) or 5-Fluorouracil (5-FU)-Irinotecan (CPT-11)- Oxaliplatin (OXA) also named FOLFIRINOX, the average survival is around 10 months. In localized resectable stages, the 5-year relapse rate is approximately 60% in patients treated with adjuvant chemotherapy regimens such as FOLFIRINOX and 80% in those exclusively treated with GMZ2-5.
One of the main causes for the failure of chemotherapeutic treatments in pancreatic adenocarcinoma (PAD) is the development of drug resistance. Accordingly, a limited cytotoxic diffusion within the tumor has been described, seemingly due the specific nature of the extracellular matrix of PAD, which shows high levels of hyaluronic acid and fibronectin that originate a hypoxic environment6. In addition, the presence of cancer stem cells (CSCs) in these tumors is considered a key factor for drug resistance. In fact, CSCs have been demonstrated to show higher levels of different proteins involved in carcinogenesis compared to non-cancer stem cells. These proteins include poly-ADP ribose polymerase (PARP), which is implicated in DNA repair; P-glycoprotein, BCRP and MRP, which participate in drug detoxification; or NFkB and STAT3, which are involved in anti-apoptotic mechanisms7. Moreover, CSCs could be responsible for the poor prognosis of metastatic pancreatic cancer and high relapse rates despite adjuvant treatment. Pancreatic CSCs are positive for CD133, CD24, CD44 and EpCAM8. Besides, multiple factors related to cell differentiation and stemness (e.g., Nanog, SOX2, Oct4) have been proposed to explain the high recurrence and aggressiveness of pancreatic cancer cells9.
In this context, the main objective of this article was to analyze the cytotoxicity and antitumor capacity of different chemotherapeutic regimens on pancreatic CSCs in vitro, as well as to determine which regimens have the highest cytotoxicity and antitumor effects on certain subpopulations of CSCs.
Methods
Cell culture
The pancreatic adenocarcinoma cell line PANC-1, obtained from Center of Scientific Instrumentation of Granada University, was grown in Dulbecco's Modified Eagle's Medium (DMEM), supplemented with 10% fetal bovine serum (FBS) and ATB (antibiotic, streptomycin + amphotericin B) at 1%. Cell line was maintained at 37°C in an atmosphere containing 5% CO2.
In vitro proliferation assays
PANC-1 Cells were seeded in 24-well plates (8000 cells/well) and incubated overnight. Then, the different treatments, 5-FU, OXA, CPT-11, GMZ (Sigma Aldrich-Merck, Darmastadt, Germany) and Abraxane (nab-paclitaxel) (Celgene, New Jersey, EEUU) were dissolved in DMEM and were administered without FBS and ATB. In the same way, there were control wells. In brief, after 72h, plates were fixed with 10% acid trichloroacetic (TCA) for 20 minutes and washed with distilled water for 3 times. Once dried, they have been stained with sulforhodamine B, which binds to the basic proteins of membrane of the cells that remain glued to the wells of the plate, those that were live in the trial. They have been stained for 20 min, after which they have been washed with 1% acetic acid three times. They have been left dry. Once dry, tris-base has been added to the wells for 20 min and then they obtained 100 μL that have been added to a 96-well plate to read with a spectrophotometer and the Ascent software at 492nm. Absorbance values and inhibitory concentration 50 (IC50) values were calculated.
RT-qPCR assay
To determine modulation of CSCs markers expression, a Real-Time PCR was carried out. RNA (1 μg) (RNeasy Mini Kit, Qiagen, MD, USA) was reversed transcribed with M-MLV reverse transcriptase (Sigma, Italy). RNA was transcribed to cDNA by retrotranscription assay and a qPCR was carried out using the primers for pancreatic stem cells markers CD24, CD44, CD133, SOX2 and Oct4 (Sigma Aldrich-Merck, Darmastadt, Germany). In addition, GADPH was used as an endogenous control. The PCR cycling program was: 50°C (2min), 95°C (2 min), 40 cycles of denaturation at 95°C (30 s), annealing at 56°C (30 s), and extension at 72°C (40 s), followed by a melting curve analysis (range 56-95°C) with increments of 0.5°C/ 5 s to assess the primer specificity. The target transcripts were independently normalized to GADPH (housekeeping gene), and the RNA of the T0 cells was used as the calibration control. The results were expressed on a logarithmic scale as fold changes (FCs), with the 2^(-ΔΔCt) method.
Results and discussion
Pancreatic CSCs are defined by simultaneous positivity for, at least, three stem cell markers among which are CD24, a small cell surface mucin; CD44, a surface receptor for hyaluronic acid and fibronectin and CD133 or prominin, a tyrosine phosphorylase that is located in lipid rafts8. Previous studies showed great resistance of CSCs from different tumor types to chemotherapeutic agents in monotherapy such as GMZ, Paclitaxel, 5-FU, CPT-11 and OXA. In fact, GMZ selected pancreatic CSCs10, colorectal CSCs were resistant to OXA and CPT-1111,12, and gastric and ovarian CSCs were resistant to 5-FU and Paclitaxel, respectively13,14. To our knowledge, this is the first study analyzing the modulation of stemness markers by RT-qPCR in pancreatic cancer cells after exposure to different drug concentrations (near to IC50) (Table 1) both for monotherapy (GMZ, ABRAX, 5-FU, OXA and CPT-11) and combined therapy (GMZ-ABRAX, FOLFOX and FOLFIRI).
Interestingly, we demonstrated a higher percentage of CSCs in monotherapy regimens that were especially higher with the use of OXA. In addition, GMZ in monotherapy significantly increased several stemness markers (Figure 1.B). These results suggested that GMZ in adjuvant monotherapy could favor the presence of CSCs and tumor recurrence. By contrast, FOLFOX and FOLFIRI regimes were associated with a lower percentage of CSCs (Figure 1.C). FOLFIRINOX was not tested because of its high cytotoxicity. The association of GMZ and ABRAX did not significantly affect the proportion of CSCs.
In addition, an analysis of the RT-qPCR results showed that GMZ primarily selected positive CSCs for CD133, CD24, SOX2 and Oct4; OXA for CD44 and CPT-11 for SOX2. In addition, 5-FU reduced CSC marker expression whereas ABRAX did not seemingly select any of them. Regarding the combined regimens, GMZ- ABRAX selected pancreatic positive CSCs for CD133, CD24 and SOX2; FOLFOX showed a high tendency to select CD133 positive cells while FOLFIRI significantly decreased positive CSCs for CD133 and CD24 (Figure 1.C). These results suggest that the chemotherapeutic regimens showed a predilection for different CSC subpopulations with different stemness markers. This differential sensitivity could be used to develop personalized treatments after histological studies or liquid biopsy. Indeed, CD44 CSCs have been reported to show resistance to GMZ treatment8. However, our results showed that only OXA selected CD44 positive CSCs while GMZ increased the presence of CD24, CD133, Oct4 and SOX2 positive CSCs. Furthermore, a significant progression of pancreatic cancer has been shown in patients using GMZ-ABRAX treatment, probably by selecting for CD133-positive CSCs8. In this case, our results supported this hypothesis since GMZ-ABRAX selected cells positive for CD133, CD44 and SOX2.
Figure 1. Modulation of CSCs markers in PANC-1 cells after exposure to monotherapy and combined therapy. A, Graphical representation of the determination of CSCs markers and their proportion in different treatments. B and C, qPCR analysis to determine the relative fold change of CD24, CD44, CD133, SOX2 and Oct4 mRNA expression in PANC-1 cells after exposure to monotherapy using 5-FU (100 µM), OXA (20 µM), CPT-11 (20 µM), GMZ (7 µM), and ABRAX (0.011 µM), and combined therapy using GMZ-ABRAX (20 + 0.011 µM), 5-FU-OXA (100 + 20 µM) and 5-FU-CPT11 (100 + 20 µM). Untreated samples (control) were established as calibrator. Data represents the mean value ± S.D. of triplicate experiences. * p<0.05.
Table 1. Ic50 value of different chemotherapies agents tested in PANC-1.
| Chemotherapy | Ic50 (µM) |
|---|---|
| Abraxane | 0.0096 ± 0.001 |
| Gemcitabine | 6.78 ± 0.42 |
| 5-Fluorouracile | 97.27 ± 1.62 |
| Oxaliplatin | 18.32 ± 1.05 |
| CPT-11 | 16.42 ± 0.98 |
Based on our in vitro results, the combination of the two monotherapies that did not select CSCs or cells with stemness markers (i.e. 5-FU and ABRAX) could show significant benefits in the treatment of pancreatic cancer. In addition, modulation of CSCs by the combined therapies tested in culture cells was observed. Accordingly, we could suggest conducting randomized controlled trials based on the combination of GMZ-ABRAX and FOLFIRI applied alternately (e.g. 2:1) in metastatic stages of pancreatic cancer. This would imply the use of treatment cycles focused on CSCs rather than merely the tumor mass, as is the case with GMZ-ABRAX. At tumor progression to GMZ-ABRAX, FOLFIRI could be a good chemotherapeutic regime because it reduces the CSCs population and stemness markers that are upregulated by GMZ-ABRAX regime. With regard to adjuvant therapy, conducting clinical trials to demonstrate the superiority -or non-inferiority- of FOLFIRI over FOLFIRINOX would also be clarifying since pancreatic CSCs seem to be more sensitive to the first regimen. According, to this results the use of GMZ in monotherapy as adjuvant treatment in PAD could be counterproductive due to selection of several stemness markers and further studies should be developed to confirm these results.
In conclusion, pancreatic CSCs and different cell subpopulations with stemness have a differential sensitivity to chemotherapeutic regimens. Specifically, pancreatic CSCs showed the greatest sensitivity to FOLFIRI compared to the rest of regimens analyzed in this study. These results could contribute to the development of personalized treatments and to delay tumor recurrence and progression in pancreatic cancer. In fact, this is the first in vitro study that tests the differential cytotoxicity of chemotherapeutic agents against different populations of pancreatic CSCs.
