GSK-3β Can Regulate the Sensitivity of MIA-PaCa-2 Pancreatic and MCF-7 Breast Cancer Cells to Chemotherapeutic Drugs, Targeted Therapeutics and Nutraceuticals

Glycogen synthase kinase-3 (GSK-3) is a regulator of signaling pathways. KRas is frequently mutated in pancreatic cancers. The growth of certain pancreatic cancers is KRas-dependent and can be suppressed by GSK-3 inhibitors, documenting a link between KRas and GSK-3. To further elucidate the roles of GSK-3β in drug-resistance, we transfected KRas-dependent MIA-PaCa-2 pancreatic cells with wild-type (WT) and kinase-dead (KD) forms of GSK-3β. Transfection of MIA-PaCa-2 cells with WT-GSK-3β increased their resistance to various chemotherapeutic drugs and certain small molecule inhibitors. Transfection of cells with KD-GSK-3β often increased therapeutic sensitivity. An exception was observed with cells transfected with WT-GSK-3β and sensitivity to the BCL2/BCLXL ABT737 inhibitor. WT-GSK-3β reduced glycolytic capacity of the cells but did not affect the basal glycolysis and mitochondrial respiration. KD-GSK-3β decreased both basal glycolysis and glycolytic capacity and reduced mitochondrial respiration in MIA-PaCa-2 cells. As a comparison, the effects of GSK-3 on MCF-7 breast cancer cells, which have mutant PIK3CA, were examined. KD-GSK-3β increased the resistance of MCF-7 cells to chemotherapeutic drugs and certain signal transduction inhibitors. Thus, altering the levels of GSK-3β can have dramatic effects on sensitivity to drugs and signal transduction inhibitors which may be influenced by the background of the tumor.


Introduction
Glycogen synthase kinase-3 (GSK-3) is a family of kinases consisting of GSK-3α and GSK-3β. The GSK-3 family members are highly conserved and expressed in many different types of cells and tissues [1][2][3][4]. They function as kinases and phosphorylate many
In the following studies, we investigated the effects of GSK-3β expression on the sensitivity of the MIA-PaCa-2 pancreatic cancer and the MCF-7 breast cancer cell lines to various chemotherapeutic drugs, signal transduction inhibitors, and nutraceuticals. The MIA-PaCa-2 PDAC cell line is a good model to examine the effects of GSK-3β on PDAC drug sensitivity as it contains mutations at KRAS and TP53 [48], two of the most frequently mutated genes in PDAC. Expression of ectopic WT-GSK-3β often increased the resistance of MIA-PaCa-2 cells to various chemotherapeutic drugs including: 5FU, paclitaxel, cisplatin, docetaxel, irinotecan, doxorubicin, daunorubicin, and mitoxantrone in comparison to cells transfected with either KD-GSK-3β or the empty vector pLXSN. In contrast, cells transfected with WT-GSK-3β were more sensitivity to an inhibitor that targets BCL2 and BCLXL than cells transfected with KD-GSK-3β or pLXSN. WT-GSK-3β also had increased metabolic and glycolytic activities in comparison to cells transfected with KD-GSK-3β. We also examined the effects of introduction WT-GSK-3β and KD-GSK-3β on the sensitivity of MCF-7 breast cancer cells to chemotherapeutic drugs, signal transduction inhibitors and a nutraceutical. In contrast to the results observed in MIA-PaCa-2 PDAC cells, where WT-GSK-3β acted as a tumor promoter and KD-GSK-3β functioned as a tumor suppressor, in MCF-7 breast cancer cells, KD-GSK-3β functioned as a tumor promoter and WT-GSK-3β functioned as a tumor suppressor. These results document the complexity of GSK-3β in regulation of therapeutic sensitivity which is likely dependent of the presence of different mutations in various cell types.

Cell Culture and Chemotherapeutic Drugs, Signal Transduction Inhibitors and Nutraceuticals
MIA-PaCa-2 PDAC cells (ATCC CRM-CRL-1420) were obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). The cells were recovered from a 65-year old Caucasian male PDAC patient [48]. MIA-PaCa-2 cells were cultured as described [49]. MCF-7 breast cancer cells (ATCC ® HTB-22 TM ) were obtained from the ATCC. They were derived from a metastatic site pleural effusion of a breast carcinoma from a 69-year old female [50]. MCF-7 cells were cultured as described [37]. Chemotherapeutic drugs, signal transduction inhibitors and nutraceuticals were obtained from either Sigma-Aldrich (Saint Louis, MO, USA) or Selleck Chemicals (Houston, TX, USA).

Cell Proliferation Assays in the Presence of Chemotherapeutic Drugs, Signal Transduction Inhibitors, and Nutraceuticals
MIA-PaCa-2 + WT-GSK-3β, MIA-PaCa-2 + KD-GSK-3β, MIA-PaCa-2 + pLXSN, MCF-7 + WT-GSK-3β, MCF-7 + KD-GSK-3β, and MCF-7 + pLXSN cells were seeded in 96-well cell culture plates (BD Biosciences, San Jose, CA) at a density of 5000 cells/well in 100 µL of phenol red free RPMI-1640 containing 1% FBS as described [37,49]. The treatment medium was prepared by performing 10 two-fold serial dilutions to create a range of 11 concentrations of the different drugs, signal transduction inhibitors, and nutraceuticals. After 72 h of treatment (four days after seeding), the amount of 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) (Sigma-Aldrich) reduction in each well was quantified as described [37,49]. The absorbance at 570 nM was determined with a FL600 microplate fluorescence reader (Bio-Tek Instruments; Winooski, VT, USA) as described [37,49]. The mean and corresponding standard deviation of normalized adjusted absorbance was calculated from three replicate wells for each drug concentration. The inhibitory concentration of 50% (IC 50 ) is defined in this context as the concentration of the drug that causes MIA-PaCa-2 or MCF-7 cells to proliferate at a rate that is half as rapid as cells incubated in the absence of the drug.
The drug concentrations used in human therapy are usually much higher than those used to treat tissue culture cells. This is because drug delivery to humans has many more factors which restrict the effects of the drugs in various human organs (e.g., liver) than in tissue culture cells in vitro. None of the drugs/nutraceuticals in our studies exceeded the drug concentrations that are used clinically. Thus, our studies provide a model for the effects of these drugs in vitro.

Colony-Formation Assays
MIA-PaCa-2 + WT-GSK-3β, MIA-PaCa-2 + KD-GSK-3β, and MIA-PaCa-2 + pLXSN cells were collected and seeded in 6-well cell culture plates at a density of 500 cells/well in 2 mL of DMEM + 5% FBS for each well (three replicate wells for each condition) as described [37,49]. Next, 24 h after seeding, plates were then treated with different concentrations of 5FU, gemcitabine, doxorubicin, tideglusib, metformin, or berberine in 2 mL of DMEM + 5% FBS for each well and incubated for three weeks at 37 • C as described [37,49]. At the end of the three-week treatment period, fixed cells were incubated in Giemsa stain (Sigma) for 5 min at room temperature as described [37,49]. Colonies consisted of at least 50 cells and the number of colonies present in each well was counted. The mean number of colonies and corresponding standard deviation was calculated from three replicate wells for each condition. The colony formation abilities were determined three times for each cell type and each treatment condition. Statistical significance was calculated using the GraphPad QuickCalcs software (San Diego, CA, USA) using an unpaired t test with a 95% confidence interval.

Real-Time Cell Metabolic Analysis
Mitochondrial activity was measured by performing mitochondrial stress tests and glycolysis stress tests with the Seahorse instrument (Agilent, Santa Clara, CA, USA) as described [53]. Briefly, exponentially growing cells in tissue culture flasks were washed with phosphate buffered saline and then treated with 1X trypsin (Life Technologies) for 5 min. The cells were then briefly centrifuged, and the cell numbers were determined on an automatic cell counter after staining with trypan blue. 100,000 cells of each cell type in a volume of 200 microliters of standard tissue culture medium was then added to 5 wells for each cell type on a Seahorse 24 well plate. The cells were allowed to adhere to the plate for 1 h at 37 • C. Then, the 24 well plate was placed in the Seahorse instrument and the various agents were added at the indicated time periods on the graphs (e.g., glucose, oligomycin, 2-deoxyglucose, BAM15, rotenone, and antimycin A). After the Seahorse experiments were performed, the actual protein concentrations in each well were determined and standardized. Briefly, the cells were lysed using the RIPA buffer, and total protein content of each well was determined using the Pierce BCA Protein Assay Kit (Cat# 23227, ThermoFisher Scientific, (Waltham, MA, USA). Then the oxygen consumption rate (OCR) or the extracellular acidification rate (ECAR) value of each well was divided by the total protein concentration of that well. The statistical significance was determined by the Mann-Whitney test with Graph Pad software (San Diego, CA, USA).

Results
To determine the effects of GSK-3β on the sensitivity of PDAC cancer cells to chemotherapy, targeted therapy and nutraceuticals, MIA-PaCa-2 PDAC cells were transfected with wild-type (WT), kinase dead (KD) forms of GSK-3β [7,51]. As controls for these experiments, we also examined the effects of pLXSN, which is an empty vector encoding NeoR [52]. The effects of various chemotherapeutic drugs on the drug sensitivity of control empty vector pLXSN and untransfected cells (parental lines) were examined in some cases. The sensitivities of the empty vector pLXSN and untransfected cells were similar.

Effects of GSK-3β on Sensitivities of PDAC Cells to Chemotherapeutic Drugs Used to Treat PDAC Patients
The effects of WT-GSK-3β, KD-GSK-3β, and pLXSN on the sensitivity to MIA-PaCa-2 cells to chemotherapeutic drugs were used to treat PDAC are presented in Figures 2 and 3, and Table 1. MIA-PaCa-2 cells were used in the following study as they represent an in vitro model for pancreatic cancer. MIA-PaCa-2 cells have an activating mutation in KRAS and a gain of function mutation at TP53, as well as some other mutations important in pancreatic cancer cells. They are also estrogen-receptor (ER) positive and metastatic [48]. These characteristics are often present in pancreatic cancer. Other pancreatic cancer cell lines lack some of these properties.
Introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in an increase of the IC 50 values, i.e., decrease of sensitivity, of the cells to all the tested chemotherapeutic drugs used to treat PDAC patients, as compared to the control. The greatest decrease of sensitivity was observed with docetaxel and oxaliplatin.
In turn, introduction of KD-GSK-3β into these cells resulted in about 2-fold decreases of the IC 50 values, i.e., increase of sensitivity, of the cells to almost all the tested chemotherapeutics. The exception was oxaliplatin to which the sensitivity increased over 13 times.
Comparison of the effects of transfection of the cells with WT-GSK-3β and KD-GSK-3β, the most dramatic decreases in sensitivities to chemotherapeutic drugs were observed after introduction of the WT-GSK-3β for oxaliplatin (almost 37-fold) and docetaxel (7.5-fold).
For the rest of the tested chemotherapeutics, the sensitivities decreased about 2-4-fold.
3.2. Effects of WT-GSK-3β and KD-GSK-3β on the Sensitivities of MIA-PaCa-2 Cells to Chemotherapeutic Drugs Used to Treat Patients with Other Types of Cancer Next we examined the effects of introduction of WT-and KD-GSK-3β on the chemosensitivities of other drugs frequently used to treat other types of cancer patients as this may provide additional information important for determining the effects of GSK-3β on chemotherapeutic drug-resistance. Results of these experiments are presented in Figure 4 and Table 1. As in the case of chemotherapeutics used for pancreatic cancer treatment, introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in an increase of the IC 50 values, i.e., decrease of sensitivity, of the cells to all the tested compounds. Table 1. Effects of WT-GSK-3β, KD-GSK-3β, and pLXSN empty vector on sensitivity of MIA-PaCa-2 pancreatic cancer cells to chemotherapeutic drugs, signal transduction inhibitors, and nutraceuticals. The concentrations presented in the Table 1 represent the inhibitory concentration 50 (IC 50 ) values (determined as previously described [37,49] for tested substances).    Next we examined the effects of introduction of WT-and KD-GSK-3β on the chemosensitivities of other drugs frequently used to treat other types of cancer patients as this may provide additional information important for determining the effects of GSK-3β on chemotherapeutic drug-resistance. Results of these experiments are presented in Figure 4 and Table 1. As in the case of chemotherapeutics used for pancreatic cancer treatment, introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in an increase of the IC50 values, i.e., decrease of sensitivity, of the cells to all the tested compounds.  These experiments were repeated 4 times and similar results were obtained. Statistical analyses were performed via the Student's T test on the means and standard deviations of various treatment groups. *** p < 0.0001, ** p < 0.005, and * p < 0.05.

Effects of WT-GSK-3β and KD-GSK-3β on Sensitivities to GSK-3β Inhibitors
As stated previously, GSK-3 plays various roles in cancer, including tumor promoter and tumor suppressor activities. Inhibition of GSK-3 activity has been proposed for the treatment of PDAC [54,55]. Thus, the effects of four structurally diverse GSK-3 inhibitors [3] were examined: SB415286, tideglusib, 6-bromoindirubin-30-oxime (BIO), and CHIR99021. Results of these experiments are presented in Figure 5 and summarized in Table 1.
KD-GSK-3β reduced the resistance of MIA-PaCa-2 cells to the GSK-3 inhibitors 8-fold to SB415286 and about 2-fold for the remaining three compounds, compared with the cells transfected with the empty vector.
Direct comparison of the effects of the introduction of WT-GSK-3β and KD-GSK-3β into MIA-PaCa-2 cells revealed augmented resistance to tideglusib over 70-fold and to SB415286 20-fold. Thus, the results indicated an inverse correlation between WT-GSK-3β and vulnerability of MIA-PaCa-2 cells to chemotherapy and GSK-3 inhibitors.
KD-GSK-3β reduced the resistance of MIA-PaCa-2 cells to the GSK-3 inhibitors 8-fold to SB415286 and about 2-fold for the remaining three compounds, compared with the cells transfected with the empty vector.
Introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in increases of the IC 50 values (decrease of sensitivity) to all the tested inhibitors, as compared to MIA-PaCa-2 cells transfected with the empty vector. In this context, it was not unexpected that the introduction of kinase-dead GSK-3β into these cells increased their sensitivities to the tested inhibitors of diverse pathways.
Direct comparison of the effects of the introduction of WT-GSK-3β and KD-GSK-3β to MIA-PaCa-2 cells revealed that increased expression of WT-GSK-3β augmented the resistance of the cells to EGFR pathway inhibitor ARRY-543 50-fold. The increases in resistance to remaining pathways inhibitors were lower but still significant.
Introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in increases of the IC50 values (decrease of sensitivity) to all the tested inhibitors, as compared to MIA-PaCa-2 cells transfected with the empty vector. In this context, it was not unexpected that the introduction of kinase-dead GSK-3β into these cells increased their sensitivities to the tested inhibitors of diverse pathways.
Direct comparison of the effects of the introduction of WT-GSK-3β and KD-GSK-3β to MIA-PaCa-2 cells revealed that increased expression of WT-GSK-3β augmented the resistance of the cells to EGFR pathway inhibitor ARRY-543 50-fold. The increases in resistance to remaining pathways inhibitors were lower but still significant.

Effects of WT-GSK-3β and KD-GSK-3β on Sensitivity to the BCL2/BCLXL ABT-737 Inhibitor
BCL2 and BCLXL play critical roles in apoptosis and cancer development and they can be regulated by GSK-3 [3]. The IC50 of the BCL2/BCLXL ABT-737 inhibitor in MIA-PaCa-2 + pLXSN cells was approximately 350 nM ( Figure 7). Introduction of WT-GSK-3β into these cells reduced the IC50 to approximately 7 nM, i.e., 50-fold lower than that observed in MIA-PaCa-2 + pLXSN cells. Upon introduction of KD-GSK-3β, the IC50 for ABT-737 was approximately 350 nM, the same as in MIA-PaCa-2 + pLXSN cells. The IC50 for ABT-737 was approximately 50-fold lower in MIA-PaCa-2 + WT-GSK-3β than in MIA-PaCa-2 + KD-GSK-3β cells (Figure 7). Commonly used drugs such as metformin and chloroquine were originally developed to treat diseases such as type II diabetes and malaria, respectively. Recently, these

Effects of WT-GSK-3β and KD-GSK-3β on Drugs Used to Treat Diabetes, Malaria, and the Nutraceutical Berberine
Commonly used drugs such as metformin and chloroquine were originally developed to treat diseases such as type II diabetes and malaria, respectively. Recently, these drugs have been shown to have anti-cancer properties [3,11,12,49].
Cells 2021, 10, x 13 of 27 the effects of the introduction of WT-GSK-3β and KD-GSK-3β to MIA-PaCa-2 cells revealed that increased expression of WT-GSK-3β elevated the resistance to berberine over 3-fold ( Figure 8B, Table 1). The IC50 of chloroquine in MIA-PaCa-2 + pLXSN cells was similar as that observed in MIA-PaCa-2 + KD-GSK-3β cells, and it was approximately 22-fold lower than in MIA-PaCa-2 + WT-GSK-3β ( Figure 8C, Table 1), which means that introduction of WT-GSK-3β increased resistance of the cells to this anti-malarial drug.
Thus, WT-GSK-3β increased the resistance of MIA-PaCa-2 to some drugs used to treat diabetes and malaria.

Effects of WT-GSK-3β and KD-GSK-3β on the Sensitivities to Drugs Used to Suppress Cancer Progression and Metastasis
Galectin-1 has been implicated in the metastasis of many cancers, including PDAC [65]. OTX008 is a galectin-1 inhibitor [66]. Introduction of WT-GSK-3β into MIA-PaCa-2 cells slightly increased the IC50 for this compound, as compared to control cells. Introduction of KD-GSK-3β had a more pronounced effect-it decreased the IC50 almost 60-fold. Thus, in comparison to KD-GSK-3β-expressing cells, introduction of WT-GSK-3β resulted in nearly 70-fold decrease in sensitivity of MIA-PaCa-2 cells to OTX008 ( Figure 9A, Table  1). Berberine is used in traditional medicine to treat diabetes and other diseases [49]. Berberine can inhibit cell growth and induce apoptosis in many cells. It can also induce TP53. Berberine and metformin share some similar effects [49].
Introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in a berberine IC 50 1.7-fold higher than that observed in MIA-PaCa-2 + pLXSN cells. In turn, upon the introduction of KD-GSK-3β, the IC 50 for berberine was about 2-fold reduced. Direct comparison of the effects of the introduction of WT-GSK-3β and KD-GSK-3β to MIA-PaCa-2 cells revealed that increased expression of WT-GSK-3β elevated the resistance to berberine over 3-fold ( Figure 8B, Table 1).
The IC 50 of chloroquine in MIA-PaCa-2 + pLXSN cells was similar as that observed in MIA-PaCa-2 + KD-GSK-3β cells, and it was approximately 22-fold lower than in MIA-PaCa-2 + WT-GSK-3β ( Figure 8C, Table 1), which means that introduction of WT-GSK-3β increased resistance of the cells to this anti-malarial drug.
Thus, WT-GSK-3β increased the resistance of MIA-PaCa-2 to some drugs used to treat diabetes and malaria.

Effects of WT-GSK-3β and KD-GSK-3β on the Sensitivities to Drugs Used to Suppress Cancer Progression and Metastasis
Galectin-1 has been implicated in the metastasis of many cancers, including PDAC [65]. OTX008 is a galectin-1 inhibitor [66]. Introduction of WT-GSK-3β into MIA-PaCa-2 cells slightly increased the IC 50 for this compound, as compared to control cells. Introduction of KD-GSK-3β had a more pronounced effect-it decreased the IC 50 almost 60-fold. Thus, in comparison to KD-GSK-3β-expressing cells, introduction of WT-GSK-3β resulted in nearly 70-fold decrease in sensitivity of MIA-PaCa-2 cells to OTX008 ( Figure 9A, Table 1).

Effects of WT-GSK-3β and KD-GSK-3β on the Sensitivities to Drugs Used to Suppress Cancer Progression and Metastasis
Galectin-1 has been implicated in the metastasis of many cancers, including PDAC [65]. OTX008 is a galectin-1 inhibitor [66]. Introduction of WT-GSK-3β into MIA-PaCa-2 cells slightly increased the IC50 for this compound, as compared to control cells. Introduction of KD-GSK-3β had a more pronounced effect-it decreased the IC50 almost 60-fold. Thus, in comparison to KD-GSK-3β-expressing cells, introduction of WT-GSK-3β resulted in nearly 70-fold decrease in sensitivity of MIA-PaCa-2 cells to OTX008 ( Figure 9A, Table  1).  Serpine-1, also known as plasminogen activator inhibitor-1 (PAI-1), is a gene implicated in the progression of many cancers [67]. It is regulated by TP53 in many cells. We have previously shown that serpine-1 is regulated by TP53/miR-34a in MIA-PaCa-2 cells as well as primary PDAC patient samples [68]. Tiplaxtinin is a serpine-1 inhibitor [69]. The IC 50 of tiplaxtinin in MIA-PaCa-2 + pLXSN cells was approximately 20 nM ( Figure 8C), which is similar to MIA-PaCa-2 + KD-GSK-3β cells. Introduction of WT-GSK-3β into MIA-PaCa-2 cells resulted in a 10-fold increase of this value ( Figure 9B, Table 1).
The hedgehog (Hh) pathway has been implicated in the progression of PDAC and other cancer types [70]. GSK-3 interacts with certain components of the Hh pathway. Vismodegib (Erivedge) inhibits the Hh pathway as it is an antagonist of the smoothened receptor (SMO), which is a key regulator of the pathway. Its effects in combination with gemcitabine and nab-paclitaxel have examined in phase 2 studies with PDAC patients [71,72].
Isoliquiritin is a natural product derived from licorice ( Figure 10B). It is a flavonoid and has broad effects including anti-oxidant, anti-inflammatory, and anti-cancer properties [75]. It can activate TP53 in lung cancer cells [76]. Isoliquiritin has been shown to inhibit the in vitro invasiveness of PDAC cells [77]. The IC 50 of isoliquiritin in MIA-PaCa-2 + pLXSN cells was approximately 600 nM, it was increased about 5-times in cells overexpressing WT-GSK-3β ( Figure 10B, Table 1). In contrast, the gilteritinib IC 50 in MIA-PaCa-2 + KD-GSK-3β cells was approximately 28 nM.
Introduction of KD-GSK-3β to MIA-PaCa-2 cells resulted in decreases of IC 50 values (increases in sensitivities) of the cells to all the tested chemotherapeutics.
The highest increase of sensitivity was observed for isoliquiritin (over 21-fold) and doxorubicin (~6-fold).
In summary, comparing the effects of transfection of the cells with WT-GSK-3β to KD-GSK-3β, there was over 100-fold decrease of WT-GSK-3β-transfected MIA-PaCa-2 cells sensitivity to isoliquiritin, about 9-fold to doxorubicin, and much lower to other tested drugs. other cancer types [70]. GSK-3 interacts with certain components of the Hh pathway. Vismodegib (Erivedge) inhibits the Hh pathway as it is an antagonist of the smoothened receptor (SMO), which is a key regulator of the pathway. Its effects in combination with gemcitabine and nab-paclitaxel have examined in phase 2 studies with PDAC patients [71,72].

Effects of WT-GSK-3β and KD-GSK-3β on Colony Formation in the Presence of Chemotherapeutic Drugs
To determine whether the changes in chemotherapeutic drug sensitivity observed by MTT analysis in MIA-PaCa2 + WT-GSK-3β, MIA-PaCa-2 + KD-GSK-3β, and MIA-PaCa-2 + pLXSN cells were also observed in larger scale cultures, the effects of three chemotherapeutic drugs were examined using colony formation analysis ( Figure 11). We chose to examine the effects of 5FU, gemcitabine, and doxorubicin on colony formation as they are both used in therapy of PDAC patients, and doxorubicin is a commonly prescribed chemotherapeutic drug to treat various cancers including breast and leukemia patients. The data for each cell line and each drug treatment were normalized to the untreated control samples and compared. When MIA-PaCa-2 + WT-GSK-3β were plated in 5FU ( Figure 11A), gemcitabine ( Figure 11B), or doxorubicin ( Figure 11C) more colonies were observed when MIA-PaCa-2 + pLXSN cells or MIA-PaCa-2 + WT-GSK-3β were plated under the same conditions than with MIA-PaCa-2 +KD-GSK-3β cells. Thus, WT-GSK-3β and KD-GSK-3β elicited positive and negative effects, respectively, on the sensitivity of MIA-PaCa-2 cells to chemotherapeutic drugs as determined by both MTT analysis and colony formation.

Effects of WT-GSK-3β and KD-GSK-3β on Colony Formation in Presence of a GSK-3 inhibitor, the AMPK Activator Metformin, and the Nutraceutical Berberine
The abilities of MIA-PaCa-2 + WT-GSK-3β, MIA-PaCa-2 + KD-GSK-3β, and MIA-PaCa-2 + pLXSN to form colonies in the presences of a GSK-3 inhibitor, the type II diabetes drug metformin, and the nutraceutical berberine were also determined ( Figure 12, Panels A, B, and C). We examined the effects of these three compounds on colony formation. Tideglusib is a GSK-3 inhibitor, which has been used in clinical studies; metformin is a common type-II diabetes drug; and berberine is a nutraceutical used in traditional medicine for various ailments. In general, more colonies were observed in MIA-PaCa-2 + WT-GSK-3β cells than either MIA-PaCa-2 + KD-GSK-3β or MIA-PaCa-2 + pLXSN cells, and less colonies were observed at higher drug concentrations. Thus, the MTT and colony formation assays yielded similar results indicating that expression of WT-GSK-3β enhanced resistance to a GSK-3 inhibitor, the type-II diabetes drug metformin, and the nutraceutical berberine.
both used in therapy of PDAC patients, and doxorubicin is a commonly prescribed chemotherapeutic drug to treat various cancers including breast and leukemia patients. The data for each cell line and each drug treatment were normalized to the untreated control samples and compared. When MIA-PaCa-2 + WT-GSK-3β were plated in 5FU ( Figure  11A), gemcitabine ( Figure 11B), or doxorubicin ( Figure 11C) more colonies were observed when MIA-PaCa-2 + pLXSN cells or MIA-PaCa-2 + WT-GSK-3β were plated under the same conditions than with MIA-PaCa-2 +KD-GSK-3β cells. Thus, WT-GSK-3β and KD-GSK-3β elicited positive and negative effects, respectively, on the sensitivity of MIA-PaCa-2 cells to chemotherapeutic drugs as determined by both MTT analysis and colony formation. In each condition, the cells were plated in 3 wells of a 6 well plate. The colony formation abilities were determined three times for each cell type and each treatment condition and similar results were observed. *** p < 0.0001, ** p < 0.005, and * p < 0.05.

Effects of WT-GSK-3β and KD-GSK-3β on Colony Formation in Presence of a GSK-3 inhibitor, the AMPK Activator Metformin, and the Nutraceutical Berberine
The abilities of MIA-PaCa-2 + WT-GSK-3β, MIA-PaCa-2 + KD-GSK-3β, and MIA-PaCa-2 + pLXSN to form colonies in the presences of a GSK-3 inhibitor, the type II diabetes drug metformin, and the nutraceutical berberine were also determined ( Figure 12, Panels A, B, and C). We examined the effects of these three compounds on colony formation. Tideglusib is a GSK-3 inhibitor, which has been used in clinical studies; metformin is a common type-II diabetes drug; and berberine is a nutraceutical used in traditional medicine for various ailments. In general, more colonies were observed in MIA-PaCa-2 + WT-GSK-3β cells than either MIA-PaCa-2 + KD-GSK-3β or MIA-PaCa-2 + pLXSN cells, and less colonies were observed at higher drug concentrations. Thus, the MTT and colony formation assays yielded similar results indicating that expression of WT-GSK-3β enhanced resistance to a GSK-3 inhibitor, the type-II diabetes drug metformin, and the nutraceutical berberine. In each condition, the cells were plated in 3 wells of a 6 well plate. The colony formation abilities were determined three times for each cell type and each treatment condition and similar results were observed. *** p < 0.0001, ** p < 0.005, and * p < 0.05.

Effects of Introduction of WT-GSK-3β and KD-GSK-3β on Metabolic Activity in MIA-PaCa-2 Cells
Cancer cells require a large amount of adenosine triphosphate (ATP) to grow rapidly. ATP is generated by glycolysis and mitochondrial oxidative phosphorylation. To determine the effects of GSK-3β on mitochondrial activity and metabolism, glycolysis and mitochondrial stress tests were performed on the various cells on the Seahorse instrument. The Seahorse instrument measures mitochondrial oxidative phosphorylation on the basis of the oxygen consumption rate (OCR), by performing real-time and live cell analysis. The instrument can also measure glycolysis by analyzing the extracellular acidification rate (ECAR). The effects of WT-GSK-3β, KD-GSK-3β on respiratory capacity were determined on MIA-PaCa-2 + pLXSN, MIA-PaCa-2 + WT-GSK-3β, and MIA-PaCa-2 + KD-GSK-3β cells.
GSK-3 has been shown to be a mitochondria oxidative metabolism regulator in stud- In each condition, the cells were plated in 3 wells of a 6 well plate. The colony formation abilities were determined three times for each cell type and each treatment condition and similar results were observed. *** p < 0.0001, ** p < 0.005, and * p < 0.05.

Effects of Introduction of WT-GSK-3β and KD-GSK-3β on Metabolic Activity in MIA-PaCa-2 Cells
Cancer cells require a large amount of adenosine triphosphate (ATP) to grow rapidly. ATP is generated by glycolysis and mitochondrial oxidative phosphorylation. To determine the effects of GSK-3β on mitochondrial activity and metabolism, glycolysis and mitochondrial stress tests were performed on the various cells on the Seahorse instrument. The Seahorse instrument measures mitochondrial oxidative phosphorylation on the basis of the oxygen consumption rate (OCR), by performing real-time and live cell analysis. The instrument can also measure glycolysis by analyzing the extracellular acidification rate GSK-3 has been shown to be a mitochondria oxidative metabolism regulator in studies with B cells obtained from GSK-3 α and β knock-out mice [78], and Mv1Lu lung epithelial cells [79]. Studies have revealed that reduction of GSK-3 activity decreased cellular O 2 consumption rate and it has been suggested that this may be a result of inhibition of respiratory complex IV activity in the absence of active GSK-3 [79].
On the other hand, it has been also shown that GSK-3 can down-regulate mitochondrial respiration by inhibition of pyruvate dehydrogenase and oxidative phosphorylation, by inhibiting respiratory chain complex I [80]. However, the effects of GSK-3β on PDAC mitochondrial activity are not well elucidated. The results presented here demonstrated that in MIA-PaCa-2 + pLXSN and MIA-PaCa-2 + WT-GSK-3β cells, all parameters of mitochondrial respiration were practically identical (Figures 13 and 14) and differences between these cells were statistically insignificant. However, there was a significant difference between these cells and MIA-PaCa-2 + KD-GSK-3β cells.  (Figures 13 and 14) and differences between these cells were statistically insignificant. However, there was a significant difference between these cells and MIA-PaCa-2 + KD-GSK-3β cells.  The basal mitochondrial respiration was significantly lower in MIA-PaCa-2 + KD-GSK-3β cells than in MIA-PaCa-2 + WT-GSK-3β and MIA-PaCa-2 + pLXSN. Transfection of the cells with KD-GSK-3β reduced their maximal respiratory and respiratory capacity levels as compared to cells transfected with pLXSN or GSK-3β (MIA-PaCa-2 + pLXSN or MIA-PaCa-2 + WT-GSK-3β cells) (Figures 13 and 14). Furthermore, MIA-PaCa-2 + KD-GSK-3β exhibited not only lower levels of mitochondrial oxidation but also many-fold reduced glycolytic activity compared to MIA-PaCa-2 + WT-GSK-3β or MIA-PaCa-2 + pLXSN cells (Figures 14 and 15). The reduction of all glycolytic parameters (basal glycolysis, glycolytic capacity and the reserve) in MIA-PaCa-2 + KD-GSK-3β cells, presumably  (Figures 13 and 14) and differences between these cells were statistically insignificant. However, there was a significant difference between these cells and MIA-PaCa-2 + KD-GSK-3β cells.  The basal mitochondrial respiration was significantly lower in MIA-PaCa-2 + KD-GSK-3β cells than in MIA-PaCa-2 + WT-GSK-3β and MIA-PaCa-2 + pLXSN. Transfection of the cells with KD-GSK-3β reduced their maximal respiratory and respiratory capacity levels as compared to cells transfected with pLXSN or GSK-3β (MIA-PaCa-2 + pLXSN or MIA-PaCa-2 + WT-GSK-3β cells) (Figures 13 and 14). Furthermore, MIA-PaCa-2 + KD-GSK-3β exhibited not only lower levels of mitochondrial oxidation but also many-fold reduced glycolytic activity compared to MIA-PaCa-2 + WT-GSK-3β or MIA-PaCa-2 + pLXSN cells (Figures 14 and 15). The reduction of all glycolytic parameters (basal glycolysis, glycolytic capacity and the reserve) in MIA-PaCa-2 + KD-GSK-3β cells, presumably  (Figures 13 and 14). Furthermore, MIA-PaCa-2 + KD-GSK-3β exhibited not only lower levels of mitochondrial oxidation but also many-fold reduced glycolytic activity compared to MIA-PaCa-2 + WT-GSK-3β or MIA-PaCa-2 + pLXSN cells (Figures 14 and 15). The reduction of all glycolytic parameters (basal glycolysis, glycolytic capacity and the reserve) in MIA-PaCa-2 + KD-GSK-3β cells, presumably reflects the lower levels of glycolytic enzymes-a result of weaker stimulation of glycolysis by NF-κB which transcriptional activity is known to be regulated in GSK-3β-dependent manner (Figure 15).  Inhibition of GSK-3 activity can decrease the metabolic properties of the cells reducing both glycolysis and mitochondrial respiration. An overview of the effects of GSK-3 on metabolic properties and the development of PDAC is presented in Figure 16.

Effects of Introduction of WT-GSK-3β, KD-GSK-3β, and pLXSN on Therapeutic Sensitivity of MCF-7 Breast Cancer Cells
To ascertain whether GSK-3β may play different roles in various cancer types, we examined the effects of WT-GSK-3β, KD-GSK-3β, and pLXSN on the therapeutic sensitivity of MCF-7 breast cancer cells. Previously, we determined that introduction of KD-GSK-3β increased the resistance of MCF-7 cells to doxorubicin and tamoxifen, drugs which are used to treat ER+ breast cancers [37].
We examined the effects of WT-GSK-3β and pLXSN on the sensitivity of MCF-7 breast cancer cells to the chemotherapeutic drug docetaxel, and the GSK-3 inhibitors Figure 15. Effects of presence of WT-GSK-3β, KD-GSK-3β and pLXSN on glycolysis. Glycolysis for STAT, glycolytic capacity, and glycolytic reserve for STAT were measured by the Seahorse instrument. STAT is an abbreviation for statistics used in study which was the Mann-Whitney test.
In contrast to downregulation of GSK-3 (MIA-PaCa-2 + KD-GSK-3β cells), the overexpression of WT-GSK-3β had practically no effect on metabolic parameters of MIA-PaCa-2 cells except glycolytic capacity which was lower in these cells than in MIA-PaCa-2 + pLXSN cells.
Inhibition of GSK-3 activity can decrease the metabolic properties of the cells reducing both glycolysis and mitochondrial respiration. An overview of the effects of GSK-3 on metabolic properties and the development of PDAC is presented in Figure 16.  Inhibition of GSK-3 activity can decrease the metabolic properties of the cells reducing both glycolysis and mitochondrial respiration. An overview of the effects of GSK-3 on metabolic properties and the development of PDAC is presented in Figure 16. To ascertain whether GSK-3β may play different roles in various cancer types, we examined the effects of WT-GSK-3β, KD-GSK-3β, and pLXSN on the therapeutic sensitivity of MCF-7 breast cancer cells. Previously, we determined that introduction of KD-GSK- Figure 16. Interactions between the GSK-3β and glycolysis, metabolism, respiratory capacity, and drug sensitivity.

Effects of Introduction of WT-GSK-3β, KD-GSK-3β, and pLXSN on Therapeutic Sensitivity of MCF-7 Breast Cancer Cells
To ascertain whether GSK-3β may play different roles in various cancer types, we examined the effects of WT-GSK-3β, KD-GSK-3β, and pLXSN on the therapeutic sensitivity of MCF-7 breast cancer cells. Previously, we determined that introduction of KD-GSK-3β increased the resistance of MCF-7 cells to doxorubicin and tamoxifen, drugs which are used to treat ER+ breast cancers [37].

Drug, Signal
3.12. Effects of the Introduction of WT-GSK-3β, KD-GSK-3β, and pLXSN on Sensitivity of MCF-7 Breast Cancer Cells to the Type-II Diabetes Drug Metformin and the Nutraceutical Berberine The effects of the type II diabetes drug metformin and the nutraceutical berberine were examined on the MCF-7 breast cancer cell line ( Figure 18, Table 2). Suppression of GSK-3 has been observed to increase AMPK activity and autophagy in some cells [81]. Introduction of WT-GSK-3β into MCF-7 cells did not change the IC 50 to metformin but it did increase the sensitivity to berberine 2.4-fold in comparison to MCF-7 + pLXSN cells ( Table 2). GSK-3 has been observed to increase AMPK activity and autophagy in some cells [81]. Introduction of WT-GSK-3β into MCF-7 cells did not change the IC50 to metformin but it did increase the sensitivity to berberine 2.4-fold in comparison to MCF-7 + pLXSN cells ( Table 2). We next examined the effects of KD-GSK-3β and pLXSN, on the sensitivity of MCF-7 breast cancer cells to metformin and berberine. Introduction of KD-GSK-3β into MCF-7 cells decreased the IC50 to metformin 5-fold ( Figure 18, Table 2). In contrast, introduction of KD-GSK-3β increased the IC50 to tideglusib 1.5-fold in comparison to MCF-7 + pLXSN cells ( Figure 18, Table 2).
In contrast, the addition of a suboptimal concentration of the GSK-3 inhibitor SB415286 had more moderate effects on IC 50 concentration of GSK-3 inhibitor tideglusib in both MCF-7 + pLXSN (1.6×↓) and MCF-7 + KD-GSK-3β (2.4×↑) cells but it did increase the tideglusib IC 50 in MCF-7 + WT-GSK-3β cells 19.4-fold indicating that GSK-3 was playing a tumor suppressor role in these cells, and suppression of its activity increased therapeutic resistance ( Figure 20, Table 3).
As an alternative approach to examine the roles that GSK-3β may play in regulation of chemosensitivity, we investigated the effects of combination of the GSK-3 inhibitor tideglusib with low concentrations of docetaxel, SB415286, metformin, and berberine on MCF-7 + pLXSN, MCF-7 + WT-GSK-3β, and MCF-7 + KD-GSK-3β cells (Figures 19-22 and Table  3). Addition of a low dose of docetaxel reduced the IC50 for tideglusib 77-, 4.5-, and 500fold in MCF-7 + pLXSN, MCF-7 + WT-GSK-3β, and MCF-7 + KD-GSK-3β cells, respectively ( Figure 18 and Table 3).  In contrast, the addition of a suboptimal concentration of the GSK-3 inhibitor SB415286 had more moderate effects on IC50 concentration of GSK-3 inhibitor tideglusib in both MCF-7 + pLXSN (1.6×↓) and MCF-7 + KD-GSK-3β (2.4×↑) cells but it did increase the tideglusib IC50 in MCF-7 + WT-GSK-3β cells 19.4-fold indicating that GSK-3 was playing a tumor suppressor role in these cells, and suppression of its activity increased therapeutic resistance ( Figure 20, Table 3). The addition of a suboptimal concentration of metformin did reduce the IC50 concentration of the GSK-3 inhibitor tideglusib in both MCF-7 + pLXSN and MCF-7 + KD-GSK-3β cells, 5.1-and 10-fold, respectively, but increased the IC50 of tideglusib in MCF-7 + WT-GSK-3β cells 5.6-fold, indicating that GSK-3 was playing a tumor suppressor role in these cells, and suppression of its activity increased therapeutic resistance ( Figure 21, Table 3).  The addition of a suboptimal concentration of metformin did reduce the IC 50 concentration of the GSK-3 inhibitor tideglusib in both MCF-7 + pLXSN and MCF-7 + KD-GSK-3β cells, 5.1-and 10-fold, respectively, but increased the IC 50 of tideglusib in MCF-7 + WT-GSK-3β cells 5.6-fold, indicating that GSK-3 was playing a tumor suppressor role in these cells, and suppression of its activity increased therapeutic resistance ( Figure 21, Table 3). In contrast, the addition of a suboptimal concentration of the GSK-3 inhibitor SB415286 had more moderate effects on IC50 concentration of GSK-3 inhibitor tideglusib in both MCF-7 + pLXSN (1.6×↓) and MCF-7 + KD-GSK-3β (2.4×↑) cells but it did increase the tideglusib IC50 in MCF-7 + WT-GSK-3β cells 19.4-fold indicating that GSK-3 was playing a tumor suppressor role in these cells, and suppression of its activity increased therapeutic resistance ( Figure 20, Table 3). The addition of a suboptimal concentration of metformin did reduce the IC50 concentration of the GSK-3 inhibitor tideglusib in both MCF-7 + pLXSN and MCF-7 + KD-GSK-3β cells, 5.1-and 10-fold, respectively, but increased the IC50 of tideglusib in MCF-7 + WT-GSK-3β cells 5.6-fold, indicating that GSK-3 was playing a tumor suppressor role in these cells, and suppression of its activity increased therapeutic resistance ( Figure 21, Table 3). Addition of a suboptimal concentration of berberine reduced the IC50 concentration of tideglusib in both MCF-7 + pLXSN and MCF-7 + KD-GSK-3β cells, 2.3-and 333-fold, respectively, but increased the IC50 in MCF-7 + WT-GSK-3β cells 111-fold indicating that GSK-3 was playing a tumor suppressor role in these cells, and berberine could not func- Addition of a suboptimal concentration of berberine reduced the IC 50 concentration of tideglusib in both MCF-7 + pLXSN and MCF-7 + KD-GSK-3β cells, 2.3-and 333-fold, respectively, but increased the IC 50 in MCF-7 + WT-GSK-3β cells 111-fold indicating that GSK-3 was playing a tumor suppressor role in these cells, and berberine could not function to decrease cell growth in the presence of WT-GSK-3β expression ( Figure 22, Table 3). Indeed, in MCF-7 + KD-GSK-3β cells, berberine was able to significantly inhibit cell growth at suboptimal concentrations. Thus, GSK-3 could play key roles in sensitivity of breast cancer cells to drugs, signal transduction inhibitors and nutraceuticals.  Table 3. Effects of pLXSN, WT-GSK-3β, and KD-GSK-3β on sensitivity of MCF-7 breast cancer cells to treatment with the GSK-3 inhibitor tideglusib in combination with drugs, signal transduction inhibitors, and a nutraceutical as determined by MTT analysis (as described previous and [49]).

Discussion
MIA-PaCa-2 cells have an activating mutation in the KRAS gene and a mutant TP53 gene that encodes a gain of function (GOF) activity. Recently, regulatory loops have been observed in cells with mutant TP53 and mutant KRAS genes, which result in elevated KRAS activity [47]. GSK-3β is a downstream signaling protein important in KRas-dependent growth and survival in mutant KRas-dependent cells such as MIA-PaCa-2 cells [47]. Thus, increased GSK-3β activity upon introduction of the WT-GSK-3β plasmid into MIA-PaCa-2 should make the cells more resistant to most drugs and signal transduction inhibitors. Suppression of GSK-3 activity with the KD-GSK-3β plasmid could decrease KRasdependent proliferation.
Previously, we observed that introducing KD-GSK-3β into MCF-7 breast cancer cells increased their resistance to the chemotherapeutic drug doxorubicin and the hormonal based drug tamoxifen in comparison to MCF-7 cells that inherited WT-GSK-3β [37]. MCF-7 cells have WT KRAS and TP53 and mutant PIK3CA genes. The presence of certain mutations in some cells may explain the ability of WT-GSK-3β to act like a tumor suppressor in some cells (e.g., MCF-7) but also act like a tumor promoter in other cells (e.g., MIA-PaCa-2). We have recently summarized the tumor promoter and tumor suppressor roles of GSK-3 [3,4].
In our previous studies with MCF-7 breast cancer cells, we compared the levels of  Table 3. Effects of pLXSN, WT-GSK-3β, and KD-GSK-3β on sensitivity of MCF-7 breast cancer cells to treatment with the GSK-3 inhibitor tideglusib in combination with drugs, signal transduction inhibitors, and a nutraceutical as determined by MTT analysis (as described previous and [49]).

Discussion
MIA-PaCa-2 cells have an activating mutation in the KRAS gene and a mutant TP53 gene that encodes a gain of function (GOF) activity. Recently, regulatory loops have been observed in cells with mutant TP53 and mutant KRAS genes, which result in elevated KRAS activity [47]. GSK-3β is a downstream signaling protein important in KRasdependent growth and survival in mutant KRas-dependent cells such as MIA-PaCa-2 cells [47]. Thus, increased GSK-3β activity upon introduction of the WT-GSK-3β plasmid into MIA-PaCa-2 should make the cells more resistant to most drugs and signal transduction inhibitors. Suppression of GSK-3 activity with the KD-GSK-3β plasmid could decrease KRas-dependent proliferation.
Previously, we observed that introducing KD-GSK-3β into MCF-7 breast cancer cells increased their resistance to the chemotherapeutic drug doxorubicin and the hormonal based drug tamoxifen in comparison to MCF-7 cells that inherited WT-GSK-3β [37]. MCF-7 cells have WT KRAS and TP53 and mutant PIK3CA genes. The presence of certain mutations in some cells may explain the ability of WT-GSK-3β to act like a tumor suppressor in some cells (e.g., MCF-7) but also act like a tumor promoter in other cells (e.g., MIA-PaCa-2). We have recently summarized the tumor promoter and tumor suppressor roles of GSK-3 [3,4].
In our previous studies with MCF-7 breast cancer cells, we compared the levels of GSK-3β protein and the extent of S9-phosphorylated GSK-3β protein, which is an indicator of its activity, by western blot analysis [37]. GSK-3β was dephosphorylated MCF-7 and MCF-7 + WT-GSK-3β cells upon treatment with doxorubicin indicating activation of GSK-3β. In contrast, GSK-3β was not activated in the MCF-7 +KD-GSK-3β cells upon doxorubicin treatment, and the cells were in fact more resistant to doxorubicin treatment [37].
In this study, we examined the effects of WT-GSK-3β and KD-GSK-3β on the sensitivity of MIA-PaCa-2 cells pancreatic cancer cells and MCF-7 breast cancer cells to a panel of chemotherapeutic drugs, signal transduction inhibitors, and nutraceuticals. Introducing WT-GSK-3β increased the IC 50 s of MIA-PaCa-2 pancreatic cancer cells to many drugs commonly used to treat PDAC, while WT-GSK-3 increased the sensitivity of MCF-7 breast cancer cells to certain drugs, signal transduction inhibitors, and nutraceuticals. MIA-PaCa-2 transfected with the pLXSN empty vector often displayed an intermediate sensitivity in comparison to cells transfected with either WT-GSK-3β or KD-GSK-3β. Thus, WT-GSK-3β was promoting resistance to these chemotherapeutic drugs and serving a tumor promoter role in MIA-PaCa-2 cells but a tumor suppressor in MCF-7 breast cancer cells. In contrast, KD-GSK-3β was promoting sensitivity (decreased the IC 50 s) to these drugs in MIA-PaCa-2 cells and serving a tumor suppressor role. KD-GSK-3β served a tumor promotor role (increased the IC 50 s) in MCF-7 cells.
Suppression of either mutant KRas or MEK1 also decreased the downstream effects of mutant KRas signaling. MIA-PaCa-2 cells with introduced WT-GSK-3β were more resistant to the PD0325901 MEK inhibitor than cells with the KD-GSK-3β or pLXSN. Thus, augmenting the level of WT-GSK-3β could increase the resistance of cells which contain mutant KRAS to MEK1 inhibitors. Treatment with MEK inhibitors have been shown to increase autophagy in pancreatic cancer with mutant KRAS genes. Suppression of autophagy was observed to synergize with MEK inhibitors [82]. Downstream of MEK is ERK. ERK can prime substrates for GSK-3β [83]. Some of substrates GSK-3β phosphorylates may alter the proliferation of the cells. Thus, WT-GSK-3 can alter the sensitivity to MEK inhibitors in PDAC cells with mutant KRAS. In contrast, we did not observe a significant difference in sensitivity to MEK inhibitors in MCF-7 breast cancer cells which have WT-KRAS upon treatment with MEK inhibitors by themselves [37]. However, treatment with MEK inhibitors did relieve the doxorubicin-and 4HT-resistance of the MCF-7 + KD-GSK-3β cells.
MIA-PaCa-2 cells with introduced WT-GSK-3β were also more sensitive to the BCL2/BCLXL inhibitor ABT-737 than cells transfected with KD-GSK-3β or pLXSN. Some of the targets of GSK-3 are BCL2-family members [84]. Thus, elevated expression of WT-GSK-3β made the cells more sensitive to the induction of apoptosis induced by ABT-737.
Similar observations have been observed in hematopoietic cells with a combination of small molecule inhibitors that target BCL2/BCLXL and PI3K/AKT signaling pathways [85]. Interactions between GSK-3 and the pro-apoptotic Bim molecule have been shown to increase the pro-apoptotic effects of BCL2/BCLXL inhibitors in human myeloid leukemia cells which were also treated with PI3K inhibitors.
Our results point to the effects that both WT-GSK-3β and KD-GSK-3β can have on sensitivity of certain pancreatic and breast cancer cells to chemotherapy and targeted therapy. These results are important as chemotherapeutic drugs can alter the activity of GSK-3β [3,86]. Therefore, without knowing the downstream consequences, suppression or activation of GSK-3β may change the sensitivity to targeted therapeutics. Clearly the role of GSK-3β in sensitivity to various drugs and signal transduction inhibitors should be further examined.
GSK-3 is an established therapeutic target and many compounds (e.g., lithium-chlo ride, tideglusib, and others) have been shown to suppress GSK-3 activity. Treatment with various GSK-3 inhibitors could influence the sensitivity to various drugs used to treat cancer patients.
The inhibition of GSK-3 activity can also render PDAC cells more sensitive to chemotherapeutic drugs, signal transduction inhibitors, and nutraceuticals. In addition, we have previously observed that treatment of PDAC cells with low doses of metformin increases the sensitivity to multiple chemotherapeutic drugs and signal transduction inhibitors [87].
Predicting and determining which cancers will be sensitive to GSK-3 inhibition is very complicated. ER-negative breast cancers often have mutations in TP53 and other tumor suppressor and oncogenes such as KRAS which could influence GSK-3 beta expression and sensitivity to GSK-3 inhibitors. Often ER-negative breast cancers, especially triple-negative breast cancers (TNBC), are very drug resistant and may have EMT. GSK-3 plays critical roles in EMT due to interactions with the Wnt/β-catenin pathway. Inhibiting GSK-3 expression in certain TNBC decreased resistance to therapeutic drugs [88]. NF-κB is overexpressed and KRAS is mutant in certain TNBC (e.g., MDA-MB-231 cells), which may result in their sensitivity to GSK-3 inhibitors [89]. GSK-3 inhibitors have been proposed to regulate the cancer stem cell properties in TNBC [90]. Thus, further elucidation of the role of GSK-3 and its effective targeting may increase breast as well as pancreatic cancer therapy.