Establishment of the Clone of Gastrointestinal Stromal Tumor Cells with the Signs of Multiple Drug Resistance and Assessment of Its Properties

Despite a long existing point of view about the low efficacy of chemotherapy for patients with gastrointestinal stromal tumors (GISTs) [1, 2], recent data indicates GISTs sensitivity to the certain types of chemotherapeutic drugs. For example, it was shown rather good perspectives of some chemotherapeutic drugs to manage the patients with metastatic and inoperable forms of GIST [3, 4]. The results obtained by our research group also demonstrate that some GIST cell lines are sensitive to the topoisomerase type II inhibitors in vitro and in vivo [5, 6]. Of note, the chemotherapeutic drugs indicated above were found to be effective against both types of GIST cell lines known to be sensitive and resistant to the targeted therapy by imatinib. Besides, we also found the imatinib to induce sensitization of GIST tumor cells to chemotherapeutic drugs indicated above [7]. One of the potential mechanisms of this phenomenon might be due to ability of imatinib to inhibit the processes of DNA homologous recombination and, as a consequence, attenuate the ability of tumor cells to repair DNA damage, induced by doxorubicin and etoposide [8]. The results of this experimental data correlate with the findings of some clinical investigations, having shown the efficacy of imatinib application in combination with small doses of doxorubicin in patients with GIST [9]. It is noteworthy, that the efficacy of this combination was demonstrated also in the group of patients, in which traditional KIT/PDGFRA mutations were not determined.

The aforementioned testifies to good perspectives of applying some chemotherapeutic drugs to treat patients with GISTs. But the fact of a steady development of malignant neoplasm chemoresistance following the beginning of chemotherapy in cancer patients is quite evident. The time period for the tumor to remain sensitive to chemotherapy depends on many factors and may vary in each individual case. Consequently, an individual assessment of tumor sensitivity to chemotherapeutic drugs and selection of individual appropriate schemes of chemotherapy for cancer patients, including those with GIST, seems to be very promising.

One of the criteria used in searching for novel perspective compounds possessing anti-tumor activity is related to their ability to induce death of the tumor cells resistant to the known chemotherapeutic drugs. Due to a long existing erroneous, to our opinion, point of view of the low efficacy of chemotherapy to treat GIST patients, there are no currently available GIST cell lines that possess phenotypical features of multiple drug resistance (MDR).

The aim of the study was to obtain a tumor cell clone of gastrointestinal stromal tumors possessing resistance to chemotherapeutic drugs of different mode of action, and to assess its sensitivity to various groups of such chemotherapeutic drugs.

Materials and Methods. GIST T-1 cell line sensitive to the targeted imatinib was chosen as an object of investigation [10]. To obtain a clone of GIST cells with MDR features, tumor cells were cultured with paclitaxel for 8 months. The initial dose of the chemotherapeutic drug in the cell culture was 0.4 nmol, the final one amounted to 24 nmol. The cells were cultured under standard conditions (37°C, 5% CO₂) in the culture RPMI-1640 medium supplemented with L-glutamine, penicillin and streptomycin (PanEco, Russia), as well as 10% fetal calf serum (HyClone, USA). Every 2 months a comparative evaluation of sensitivity of GIST cells and their clone to different groups of chemotherapeutic drugs: doxorubicin, vinblastine, paclitaxel, imatinib (Sigma-Aldrich, USA), etoposide, hydroxyurea (Calbiochem, USA), and cisplatin was performed. For this purpose, tumor cells were inoculated into the wells of a flat-bottom 96-well cell culture plate (Corning, USA) and cultured in triplicates for 48–72 h at 37°C in a humidified atmosphere containing 5% CO₂. Cytotoxicity of the above mentioned compounds was estimated colorimetrically by using the CellTiter 96^® AQueous MTS Reagent Powder (Promega, USA) and Multiscan FC plate analyzer (Thermo Scientific, USA) at 492 nm wavelength. Proliferative capacity of GIST cells was assessed in the presence of paclitaxel by on-line measurement of the cell index using iCELLigence RTCA cell analyzer (ACEA Biosciences, USA) in accordance with the manufacture’s protocol.

The expression level of the proteins, being the markers of DNA damage, apoptosis and chemoresistance, was determined by immunoblotting and appropriate monoclonal antibodies.

The results were statistically analyzed using Student t-test in Microsoft Excel 2007 program.

Results and Discussion. After 8 months of culturing GIST T-1 cells in the presence of gradually growing concentrations of paclitaxel (29^th passage), the signs of resistance to the chemotherapeutic drug indicated above were found — the obtained clone of
GIST tumor cells was designated as a T-1-29R.

The results of on-line assessment of proliferative activity in the presence of paclitaxel proved the chemoresistance of T-1-29R cells. For example, proliferative capacity of “maternal” and “daughter” cell clones in the absence of this drug (control) was found to be similar (Figure 1, (a)). When T-1-29R were cultured in the presence of paclitaxel in the doses of 1 and 10 nmol (Figure 1 (b), (c), respectively), their proliferation rate was significantly higher when compared to “maternal” GIST T-1 clone. Culturing of “maternal” and “daughter” GIST cells in the presence of high doses of paclitaxel (100 nmol) was fatal to both types of tumor cells (Figure 1 (d)).

---

Figure 1. Assessment of proliferative capacity of GIST T-1 and T-1-29R cells in the absence of the paclitaxel ((a) control) and its various concentrations: (b) 1 nmol; (c) 10 nmol; (d) 100 nmol. The tumor cell growth curves in vitro were obtained using iCELLigence device

---

Data obtained by using the photocolorimetric MTT-based assay also showed the higher viability of T-1-29R cells when compared to “maternal” GIST T-1 tumor cells in the presence of the paclitaxel, that indicated the acquired resistance to paclitaxel in GIST T-1-29R cells (Figure 2).

---

Figure 2. Cytotoxic activity of paclitaxel against GIST T-1 and T-1-29R cells. Cytotoxicity was determined by MTT-based assay. Statistically significant difference of values between the examined cell lines: * p<0.01; ** p<0.001

---

Apart from the resistance to paclitaxel in GIST T-1-29R cells, the resistance was also noted to topoisomerase type II inhibitors, e.g. doxorubicin (Figure 3).

---

Figure 3. Cytotoxic activity of doxorubicin against GIST T-1 and T-1-29R cells. Cytotoxicity was determined by MTT-based assay. Statistically significant difference of values between the examined cell lines; * p<0.001

---

Based on the investigations we determined the effective concentrations of chemotherapeutic agents inducing 50% inhibition of the survival (IC₅₀) of T-1 and T-1-29R tumor cells. Comparative values of IC₅₀ for the chemotherapeutic agents presented in the Table indicate, that GIST clone T-1-29R cells possesses cross-resistance to type II topoisomerase inhibitors, and also to some other chemotherapeutic agents affecting dynamic state of the spindle microtubules (taxanes).

---

Cytotoxic properties of chemotherapeutic agents against gastrointestinal stromal tumor cells

---

Since apoptosis is believed to be the main mechanism of tumor cells death after the exposure to abovementioned chemotherapeutic agents, an ability of paclitaxel to induce apoptosis of “maternal” (T-1) and “daughter” (T-1-29R) GIST cells was performed. Incubation of “maternal” GIST (T-1) cells with paclitaxel resulted in a significant and dose-dependent increased levels of expression of the cleaved forms of caspase-3 and poly (ADP-ribose) polymerase (PARP), known as a common apoptosis markers (Figure 4 (a)). As we expected, no increase of apoptotic markers indicated above was found in T-1-29R cells after paclitaxel exposure (Figure 4 (b)).

---

Figure 4. Paclitaxel (PTX) induces apoptosis of GIST T-1 cells (a), but not of T-1-29R cells (b). Cleaved forms of PARP and caspas-3 (cl. Casp-3) were used as apoptosis markers, an increased expression of pH2AX and pPLK-1 were used as the markers of M-phase. Paclitaxel concentrations are shown in nmol. Actin expression level was used as a loading control

---

Similarly, incubation of GIST T-1 cells (but not T-1-29R) with the topoisomerase type II inhibitors induced a dose-dependent increase of expression levels of the apoptotic markers (Figure 5 (a), (b), respectively). Of note, the level of histone H2A expression, phosphorylated by serin residue in position 139 (Ser139) (pH2AX), which is a well-known marker of double-strand DNA breaks, increased significantly in GIST T-1 cells after the exposure to chemotherapeutic agents indicated above (Figure 5 (a)). At the same time, only a minimal increase of pH2AX expression was detected for T-1-29R cells after the exposure to etoposide (Figure 5 (b)). The effect of doxorubicin was also less efficient when compared to “maternal” GIST T-1 cells.

---

Figure 5. Doxorubicin (Dox) and etoposide (Eto) induce apoptosis in GIST T-1 cells (a), but not in T-1-29R (b). Cleaved forms of PARP and caspas-3 (cl. Casp-3) were used as apoptosis markers, increased expression of pH2AX (Ser139) was used as a marker of DNA double-strand breaks. Concentrations of etoposide are shown in μmol, doxorubicin in μg/ml

---

The decreased expression of apoptotic markers and double-strand breaks in T-1-29R cells after the exposure to topoisomerase type II inhibitors might be due to the numerous factors, such as activation of DNA repair processes after the exposure to genotoxic drugs indicated above, intensification of drug elimination from the tumor cells and, therefore, less time of their action on DNA, etc. Taking into account the time period used to assess the expression level of pH2AX in our experimental model, the first scenario seems unlikable. A significant increase of P-glycoprotein (MDR-1) expression responsible for elimination of cytostatics from the cells and causing, therefore, the development of MDR indicates the high probability for intensification of drug elimination from T-1-29R tumor cells (Figure 6). Yet, no changes have been noted in the expression levels of some other transporter proteins, responsible for MDR development, e.g. ABCG2 (the basal expression of this proteins was high).

---

Figure 6. Dynamics of expression of chemoresistance markers (MDR-1, MRP-1, and ABCG2) in GIST T-1 cells during their culture in the presence of paclitaxel: 1, 2, 3 — 2, 5, 8 months of culture, respectively

---

Conclusion. The conducted study enabled us to obtain a clone of GIST T-1 tumor cell line, possessing resistance to the various groups of chemotherapeutic drugs, i.e. influencing the dynamic state of the spindle microtubules (paclitaxel), topoisomerase type II inhibitors (doxorubicin and etoposide). The obtained clone of GIST cells with the features of multiple drug resistance can be further used to screen already existing and newly-synthesized chemical compounds for their cytotoxic and anti-tumor activities against GIST cells. These findings also demonstrate the sensitivity of GIST cells to type II topoisomerase inhibitors and taxanes, which confirms the necessity of revising the existing opinion on GISTs chemoresistance.

Study Funding. The work was supported by the Russian Scientific Fund (grant No.14-15-00342).

Conflicts of Interest. The authors have no conflicts of interest to declare.