Abstract The established role of bitter melon juice (BMJ), a natural product, in activating master metabolic regulator adenosine monophosphate-activated protein kinase in pancreatic cancer (PanC) cells served as a basis for pursuing deeper investigation into the underlying metabolic alterations leading to BMJ efficacy in PanC. We investigated the comparative metabolic profiles of PanC cells with differential KRAS mutational status on BMJ exposure. Specifically, we employed nuclear magnetic resonance (NMR) metabolomics and in vivo imaging platforms to understand the relevance of altered metabolism in PanC management by BMJ. Multinuclear NMR metabolomics was performed, as a function of time, post-BMJ treatment followed by partial least square discriminant analysis assessments on the quantitative metabolic data sets to visualize the treatment group clustering; altered glucose uptake, lactate export and energy state were identified as the key components responsible for cell death induction. We next employed PANC1 xenograft model for assessing in vivo BMJ efficacy against PanC. Positron emission tomography ([18FDG]-PET) and magnetic resonance imaging on PANC1 tumor-bearing animals reiterated the in vitro results, with BMJ-associated significant changes in tumor volumes, tumor cellularity and glucose uptake. Additional studies in BMJ-treated PanC cells and xenografts displayed a strong decrease in the expression of glucose and lactate transporters GLUT1 and MCT4, respectively, supporting their role in metabolic changes by BMJ. Collectively, these results highlight BMJ-induced modification in PanC metabolomics phenotype and establish primarily lactate efflux and glucose metabolism, specifically GLUT1 and MCT4 transporters, as the potential metabolic targets underlying BMJ efficacy in PanC. Introduction Pancreatic cancer (PanC) is currently ranked as the fourth leading cause of cancer-related deaths in the USA and is estimated to be the second leading cause of such fatalities by the year 2020 (1). Given its fairly inaccessible anatomical location, routine health examinations are not helpful for early detection of the pancreatic disease, as such patients are usually diagnosed in later stages of the malignancy. Additionally, there has been also a marked increase in PanC patient population resistant to conventional chemotherapeutic agents; PanC chemoresistance has been attributed to the dense stromal environment and broad heterogeneity of mutations. The increasing repertoire of genetic modifications further confers heightened proliferative ability and capacity to PanC tumors for survival under unfavorable conditions (2). Notably, PanCs also possess a very intricately designed metabolic profile favoring excess aerobic glycolysis (Warburg effect) in addition to altered glutamine metabolism, contributing to tumor cell proliferation and PanC progression (3). Resulting amplified production in lactate and enhanced lactate export from the cancerous cell results in acidosis leading to enhanced tumorigenic invasiveness (characterized by increase in tumor cell migration and metastasis) (4). Warburg effect is regulated via multiple pathways/factors, especially adenosine monophosphate (AMP) activated protein kinase (AMPK) that functions as a metabolic checkpoint by modulating and regulating cellular response to energy availability (5). Under stressful conditions, AMPK is phosphorylated and activated in response to an elevated AMP/adenosine triphosphate (ATP) ratio, shifting the cellular metabolism to an oxidative phosphorylation phenotype causing proliferation arrest (6). Cancer cells typically experience a loss of AMPK activity, which contributes to their glycolytic phenotype (7). Recent studies provide an intriguing insight connecting AMPK activity loss and poor prognosis with increased desmoplasia in pancreatic ductal adenocarcinoma (PDAC), highlighting the importance of AMPK phosphorylation in inhibition of PanC cell migration and invasion potential (8,9). Established tumors are reported to possess a downregulated expression or a lack of functionally active AMPK and its targets (10,11). Numerous studies have confirmed the loss of AMPK activation as a frequent event in various cancers, especially in PanC promotion and progression (12,13). Tumor metabolism adapts for increased survival by causing altered sensing, nutrient uptake and utilization and efficient efflux of toxic byproducts (14). Normal non-cancerous cells typically utilize nutrients via multiple nutrient-sensing pathways and increased sensitivity in response to minute changes in cellular nutrient levels, correlating with their abundance where the energy requirements are met by increasing mitochondrial oxidative phosphorylation and activating autophagy. However, cancer cell survival overrides the normal cell machinery by inducing a dysbalance between biosynthetic and catabolic pathways that allows for rapid and uninhibited proliferation in unfavorable conditions of low nutrient and oxygen levels (14). Altogether, the studies suggest that targeting an aberrant cancer cell metabolism in PanC could be an effective strategy to manage this deadly malignancy. Furthermore, oncogenic KRAS present in ≥90% PDACs is documented as the driver of glucose metabolism in PanC triggering key metabolic alterations downstream of its activation (15). Together, accumulation of these pathway modifications for elevated nutrient requirements confers metabolic plasticity to PanC tumor cells, thus enabling PanC to accustom with these rapid metabolic changes. Due to the severity of PanC and its related complex metabolic profile, a variety of targeted therapies/therapeutic agents have proven unsuccessful or deemed ineffective in the clinic (16). To overcome that, stronger chemotherapeutic agents are employed for increasing patient life span by a few months but not without numerous associated side effects. As a result, PanC patients undergoing chemotherapy display a very poor and dismal quality of life (17,18). Accordingly, in recent times, there has been a heightened interest in screening and recognizing the potency of natural nontoxic dietary/non-dietary products as anticancer agents and their underlying mechanisms, particularly in PanC (19). Bitter melon juice (BMJ), derived from the fruits of bitter melon (Momordica charantia, family: Cucurbitaceae), is one such dietary product gaining importance in mainstream therapeutics after having been exploited for its potential in alternative medicine for decades (20). The fruit is a rich source of minerals, vitamins, glycosides, saponins, alkaloids, fixed oils and mainly bitter and non-bitter cucurbitane-type triterpene glycosides and their aglycones, proteins and steroids (20). Importantly, few short-term/uncontrolled clinical studies have indicated the efficacy of BMJ in type II diabetes, where its administration improved glucose tolerance and reduced blood glucose levels in the patients (21,22). With regards to its anticancer potential, different preparations of bitter melon including BMJ have been reported to possess significant anticancer preclinical efficacy against a wide variety of cancer types; e.g. skin, breast, prostate, and colon (19,20). Notably, previous studies in our lab have identified AMPK activation-mediated apoptotic cell death as a major mechanism in BMJ efficacy against PanC in cell culture and xenograft models (23). In another study, we reported that BMJ also facilitates the drug sensitivity in gemcitabine-resistant PanC cells and has the potential to target PanC cancer stem cells (24,25). The present study is a continuation of our research efforts in delineating the mechanisms associated with anti-PanC efficacy of BMJ. Specifically, we employed state-of the-art nuclear magnetic resonance (NMR) metabolomics and in vivo imaging platforms to assess whether BMJ health benefits associated with modulating glucose metabolism could also be preferentially involved in modulation of PanC cell metabolome and then establishing the PanC molecular metabolomic targets involved in BMJ efficacy. Materials and methods Cell lines and reagents Human PanC PANC1 and BxPC3 cell lines were obtained from american type culture collection (ATCC) during the past 4–6 years, and aliquots were frozen. Both cell lines were tested and authenticated using DNA profiling for polymorphic short tandem repeat markers at the University of Colorado Molecular Biology Core Facility by us most recently in December 2018. For studies, both cell lines were grown under standard culture conditions (37°C, 95% humidified air and 5% CO2). PANC1 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, ATCC) with high glucose, whereas BxPC3 cells were grown in RPMI 1640 (1X) media (Life Technologies); both media were supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. Lyophilized BMJ preparation and characterization was done as reported by us earlier (23). [1-13C] glucose (# CLM-420-0) and all deuterated solvents for NMR experiments were from Cambridge Isotopes (Cambridge, MA). Cell-staining reagents were Trypan blue dye (Life Technologies) and crystal violet dye (Sigma–Aldrich). AMPK inhibitor compound C (# 171260) was from Millipore Sigma. Primary antibodies used in immunohistochemistry (IHC) analysis were: Ki67 (#ab16667), CD31 (#ab28364) and lactate dehydrogenase (LDH) (#ab47010) from Abcam; cleaved-caspase 3 (#9661, Cell Signaling Technologies); and pAMPK (#PA5-37821, Invitrogen). Biotin-conjugated secondary antibodies were: anti-rabbit (#31820, Invitrogen) and anti-mouse (#E0433, Dako, Glostrup, Denmark). The diaminobenzidene kit (#SK4100) was from Vector Laboratories. Primary antibodies for immunofluorescence were: GLUT1 (#ab652, Abcam) and MCT4 (#sc376140, Santa Cruz Biotechnology). The secondary antibodies for immunofluorescence were: goat anti-mouse (#A11032) Alexa flour 594 conjugated and goat anti-rabbit (#A11008) Alexa flour 488 conjugated (Life Technologies). For immunofluorescence, standard protocols were carried out as described by us previously (26). Cell extraction and quantitative NMR metabolomics analysis For NMR spectroscopy on endogenous metabolites, PANC1 (~1.3 × 106 cells) and BxPC3 (~2 × 106 cells) per 150 mm3 culture dishes were incubated with 5 mM [1-13C] glucose for 4–72 h with/without BMJ (2% v/v) treatments and then extracted with 12% perchloric acid; both water soluble and lipid extracts were then subjected to high resolution 1H-,13C- and 31P NMR. Spectra were obtained on a Bruker 400 MHz Avance III spectrometer as described previously (27,28). Cell culture media samples were also collected (last 4 h in the presence of [1-13C] glucose) and analyzed by 1H/13C-NMR for glucose uptake/lactate export studies. Experiments were performed partly at the Animal Imaging Shared Resource (AISR, University of Colorado Cancer Center). Tumor xenograft study and tissue processing All animal experiments were performed according to the Institutional Animal Care and Use Committee (IACUC)-approved animal protocol University of Colorado Denver-Anschutz Medical Campus (UCD-AMC). Female athymic nude [Crl:NU(NCr)-Foxn1nu] mice (n = 40) were purchased from Charles River Laboratories and housed at the animal facility (UCD-AMC) for a week for acclimatization and fed AIN-76A diet (Envigo). Animals aged ~6 weeks were injected with PANC1 cells (2 × 106) suspended in 50 µl of serum-free medium (DMEM), mixed with 50 µl of Matrigel (1:1) subcutaneous into the dorsal right flanks of each mouse. One group of the animals was initiated on BMJ treatment of 200 mg/kg, 5 days a week, in 100 µl water, 24 h post-cell injection (prevention approach: BMJ-1). The remaining tumor-bearing animals were followed for 2.5–3 weeks until the tumors grew to a size of ~100 mm3 and then randomized into control (untreated) and late BMJ (intervention approach: BMJ-2) fed groups. Here, the animals were continued on BMJ treatment of 200 mg/Kg for ~7 weeks. Altogether, there were 12–14 mice per each group. Tumor volumes and body weights of the animals were recorded biweekly. The tumor volume was measured using a digital caliper and calculated using the formula 0.5236 L1 (L2)2, where L1 is the long axis and L2 is the short axis of the tumor. At study completion, the animals were euthanized by CO2 asphyxiation followed by exsanguination, tumors were excised and a part of the tumor was either flash frozen or fixed in 10% phosphate-buffered formalin for histopathological analyses. Standard lab protocols (as described previously) were used for IHC analysis (29). Positron emission tomography ([18FDG-PET) and magnetic resonance imaging on tumor-bearing animals All in vivo animal imaging studies were performed at the Colorado Animal Imaging Shared Resource (AISR). Magnetic resonance imaging (MRI) was employed for non-invasively assessing tumor volumes and cell density in the tumor-bearing animals. Bruker 4.7 Tesla/ 16-cm MRI/MRS PharmaScan (Bruker Medical, Billerica, MA) with a mouse volume transmitter/receiver radiofrequency (RF) coil (36 mm diameter) was used for all MRI studies. Briefly, the mouse was anesthetized with 2% isoflurane/oxygen and inserted into the 4.7 Tesla MRI scanner. After obtaining a tripilot localizer scan, proton-density (PD) RARE MRI scans were obtained for anatomical assessment of the tumor volume, followed by diffusion-weighted (DW) MRI scans to calculate apparent diffusion coefficients (ADC) as a marker of tumor cell density. All MRI acquisition and image analysis (tumor volumes in mm3 and ADC in s/mm2) were performed using a proprietary Bruker ParaVision v4.0 software. All MRI protocols and MRI/DWI image analysis have been reported by our team previously (29–32). For FDG-PET studies, four representative animals from each group (control, BMJ-1 and BMJ-2) were fasted for 4 h and blood glucose levels were monitored prior to injection of 250 µCi of FDG (purchased through PetNet, Denver, CO) as described previously by our team (33–35). After 60 mins of awake uptake, the animals were anesthetized with 2% isoflurane and a 10 min PET scan was acquired using Siemens Inveon μPET scanner and Inveon Acquisition Workplace software (IAW v1.5). All PET scans were acquired in a double-sampling mode to improve spatial resolution (1.2 mm). Regions of interest (ROI) were manually drawn around the tumors on scan slides and total radioactivity of the ROI determined (in kBq/ml) using Inveon Research Workplace software (IRW v2.0). The standardized uptake values (SUVs) were calculated as tissue activity (kBq/ml)/(corrected injected dose [kBq], where the corrected dose is calculated as C = C0 × e(−0.006317 × t) (18F decay constant of 0.006317). Statistical analysis and image acquisition Statistical analyses excluding principal component and multivariate analysis (discussed later) were performed using Sigma Stat software (version 3.5, Jandel Scientific). Quantitative data are presented as mean ± standard error of the mean. Statistical significance of difference between control and treatment groups was determined through one-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons. All multivariate analysis on quantitative NMR metabolomics data sets was performed using expanded MetaboAnalyst software (University of Edmonton, Edmonton, Alberta, Canada). P < 0.05 was considered significant. For immunofluorescence in cells, image acquisition was done on Olympus FV1000 confocal microscope at Advanced Light Microscopy Core of UCD-AMC and analysis was done using FV-Viewer software (Olympus). For immunofluorescence in tissues, A1-HD confocal microscope from Nikon was and NIS-Elements confocal microscope imaging software (Nikon) was used for data analysis. Results Differential effects of short-term and long-term BMJ exposure on cell viability and colony formation of human PanC cell lines We have previously observed that exposure of PanC cell lines with lower doses of lyophilized BMJ preparation (≤2% v/v) till 72 h decreases total cell numbers without inducing significant cell death, whereas higher doses of BMJ (≥ 3% v/v) significantly decrease cell viability. To investigate whether the exposure with lower BMJ doses could eventually lead to PanC cell death, we performed viability assays after long-term exposure with BMJ. Specifically, cellular viability was tracked as a function of time from 72 h onward until 144 h of BMJ exposure. During the course of the assay, spent media was aspirated and fresh media with/without BMJ or dimethyl sulfoxide (DMSO) was added to the wells after every 72 h until study completion. Trypan blue exclusion assay was performed for determining effect on cell growth and viability of PANC1 and BxPC3 cells (Figure 1A). Results indicated that BMJ (2% v/v) exposure from 72 to 144 h caused a significant decrease in the total cell count throughout the course of the study for both cell lines: ~95% decrease in total cell count for PANC1 cells and ~80% decrease in total cell count of BxPC3 cells post-144 h of BMJ treatment. Additionally, cell death induction increased significantly after BMJ exposures were given beyond 72 h. Specifically, although BMJ-induced cell death was minimal by 72 h (~10% for PANC1 and ~15% for BxPC3 cells), it significantly increased to ~50% cell death for PANC1 and ~55% cell death for BxPC3 cells after 144 h of BMJ exposure. Figure 1. Open in new tabDownload slide Long-term BMJ exposure decreases cell viability and clonogenic potential of human PanC cells. (A) Columns represent the changes in total cell number and % death of PanC cells PANC1 (top) and BxPC3 (bottom) with BMJ (2% v/v) treatments after 72, 96, 120 and 144 h of exposure. (B) Decrease in clonogenic potential of PANC1 and BxPC3 PanC cells with single exposure (2% v/v) of BMJ (top) and multiple exposures (every 72 h) of BMJ (bottom) over a course of 6 days. The pictures signify the changes in C (DMSO control) and BMJ-treated PanC cell colony numbers represented by crystal-violet-stained colonies; experiment was performed in six-well plates. ***P ≤ 0.001. Figure 1. Open in new tabDownload slide Long-term BMJ exposure decreases cell viability and clonogenic potential of human PanC cells. (A) Columns represent the changes in total cell number and % death of PanC cells PANC1 (top) and BxPC3 (bottom) with BMJ (2% v/v) treatments after 72, 96, 120 and 144 h of exposure. (B) Decrease in clonogenic potential of PANC1 and BxPC3 PanC cells with single exposure (2% v/v) of BMJ (top) and multiple exposures (every 72 h) of BMJ (bottom) over a course of 6 days. The pictures signify the changes in C (DMSO control) and BMJ-treated PanC cell colony numbers represented by crystal-violet-stained colonies; experiment was performed in six-well plates. ***P ≤ 0.001. Concurrently, the effect of long-term BMJ exposure (either as single or as multiple dosing) on colony formation of PANC1 and BxPC3 cells was also determined. Briefly, PanC cells seeded for colony formation assay were treated with/without BMJ (2%, v/v). One set of experiment involved single BMJ exposure at study initiation (after cell seeding) and no fresh BMJ dosing was done during media change/replenishment every 72 h. Second set of experiment involved BMJ exposure at study initiation (after cell seeding), followed by fresh BMJ dosing during media change/replenishment every 72 h. On study completion (day 6 after first BMJ dosing), crystal-violet-stained colonies (≥50 cells) were counted for both experimental conditions. Single BMJ exposure of PANC1 and BxPC3 cells caused ~87% and ~73% decrease, respectively, in total number of colonies compared with control wells. On the other hand, multiple fresh BMJ exposures during the course of colony formation were more potent and exhibited ~99% and ~95% decrease in PANC1 and BxPC3 colonies, respectively (Figure 1B). BMJ differentially modulates the metabolic profile of PanC cells based upon the KRAS mutational status KRAS mutation has been defined as a key player in ≥90% of PDACs, majorly contributing to their growth and progression. Previously published studies from our group have also established the role of BMJ in targeting and activating the master metabolic regulator AMPK causing apoptotic cell death in PanC (23). Subsequently, here, we investigated the comparative metabolic profiles of PanC cells with differential KRAS mutational status (PANC1; mutated KRAS and BxPC3; wild-type KRAS) on BMJ exposure. NMR-spectroscopy-based metabolites assessment was performed in both cell lines as a function of time following treatments with or without BMJ. The partial least square discriminant analysis (PLS-DA) was performed on the quantitative metabolic data sets (each metabolite identified and quantified as micromole per gram cells) to visualize the treatment group clustering (the top panels, Figure 2A) for both cell lines. The middle panels represent the putative metabolic biomarkers that are responsible for group clustering between untreated and BMJ-treated cells and thereby helps enhance the estimation accuracy by identifying a subset of important predictors (Figure 2B). Of all the metabolites screened via1H, 13C and 31P NMR spectroscopy, a few key components of interest (glucose and lactate metabolic pathway components), highlighted from PLS-DA, were further examined and plotted to investigate their modified levels with BMJ treatment (Figure 2C). 13C- lactate was an interesting component that exhibited a significant buildup within the cells with BMJ treatment for both PANC1 and BxPC3 cells, correlating with observed biological effect by 72 h of BMJ treatment. This effect was accompanied by a significant decrease in lactate export with BMJ exposure for PANC1 cells, whereas no changes were observed in BxPC3 cells. Furthermore, increased intracellular glucose levels, along with a significant decrease in ATP/adenosine diphosphate (ADP) ratios strongly pointed to cellular energy restriction with no new ATP generation by the cells with BMJ treatments; there was a complete dependence on the available intracellular glucose, which was quickly being consumed as observed by increased intracellular glycolysis and lactate buildup within the cells. More importantly, there was increased lactate export in PANC1 cells with 4 h of BMJ treatment (data not shown), which during long-term exposures seemed to induce a metabolic phenotype with decreased lactate efflux out of the cell. Increased lactate buildup (acidosis) within cell could be a possible mechanism leading to cell death after long-term exposures with BMJ doses. Interestingly, more significant and robust results were observed for PANC1 cells in response to BMJ treatments as opposed to BxPC3 cells where BMJ exhibited less significant metabolite changes compared with the untreated control cells, indicating that PanC cells harboring a mutated KRAS were more susceptible to metabolic changes induced my BMJ. Figure 2. Open in new tabDownload slide BMJ alters the PanC cell metabolome. Multivariate analysis—PLS-DA for group clustering (A) and metabolic biomarker identification (B) conducted on PANC1 (left panel) and BxPC3 (right panel) cells after 72 h treatments with/without BMJ (2%, v/v) addition. The graphs were generated using MetaboAnalyst software. (C) depicts the endogenous metabolite comparative analysis between control (DMSO treated) and BMJ-treated PanC cells PANC1 (left) and BxPC3 (right) generated from 1H, 13C and 31P NMR spectroscopy. Endogenous metabolite concentrations are presented as nmol/g cell wet weight. ***P ≤ 0.001, **P ≤ 0.01 and *P ≤ 0.05. Figure 2. Open in new tabDownload slide BMJ alters the PanC cell metabolome. Multivariate analysis—PLS-DA for group clustering (A) and metabolic biomarker identification (B) conducted on PANC1 (left panel) and BxPC3 (right panel) cells after 72 h treatments with/without BMJ (2%, v/v) addition. The graphs were generated using MetaboAnalyst software. (C) depicts the endogenous metabolite comparative analysis between control (DMSO treated) and BMJ-treated PanC cells PANC1 (left) and BxPC3 (right) generated from 1H, 13C and 31P NMR spectroscopy. Endogenous metabolite concentrations are presented as nmol/g cell wet weight. ***P ≤ 0.001, **P ≤ 0.01 and *P ≤ 0.05. BMJ exposure modifies glucose and lactate transporter expression in PanC cells Results from 1H, 13C and 31P NMR spectroscopy were validated in PanC cells PANC1 and BxPC3 by immunofluorescence analysis of glucose and lactate transporter expression status after treatment with BMJ. The glucose and lactate transporters glucose transporter type 1 (GLUT-1) and monocarboxylate transporter 4 (MCT4), respectively, are established to be upregulated and contribute to the aggressiveness of PanCs (36,37). Immunofluorescence of GLUT1 and MCT4 transporters using confocal microscopy over a time course of 4, 12 and 72 h revealed the difference in expression between BMJ-treated versus control samples. Results indicated a decrease in GLUT1 expression of PANC1 cells at 4, 12 and 72 h after BMJ exposure compared with controls; however, the decrease, although still present, was less pronounced by 72 h in BxPC3 cells (Figure 3A). There was a sharp decrease in MCT4 expression as a function time with BMJ exposure—the expression showed a trend toward decrease at 4 h, the expression decreased further by 12 h and the molecule was barely present by 72 h of BMJ exposure (for both PANC1 and BxPC3 cells, Figure 3B). Notably, quantification of the immunofluorescence data indicated that BMJ did not impact the subcellular localization (cytoplasmic versus membrane levels) but instead caused an overall decrease in the expression of the GLUT1 and MCT4 transporters (Supplementary Figures 1 and 2, available at Carcinogenesis Online). Figure 3. Open in new tabDownload slide BMJ treatment downregulates glucose and lactate transporter expression in PanC cells. PanC cells were treated with BMJ (2% v/v) and probed as a function of time for (A) glucose transporter GLUT1 (green) and (B) lactate transporter MCT4 (red) in PANC1 and BxPC3 cells at 4, 12 and 72 h time points in the presence or absence of BMJ. The transporter expression was analyzed by immunofluorescence staining. 4’,6-diamidino-2-phenylindole is represented by blue-stained nuclei. All images were captured at ×1000 magnification. Figure 3. Open in new tabDownload slide BMJ treatment downregulates glucose and lactate transporter expression in PanC cells. PanC cells were treated with BMJ (2% v/v) and probed as a function of time for (A) glucose transporter GLUT1 (green) and (B) lactate transporter MCT4 (red) in PANC1 and BxPC3 cells at 4, 12 and 72 h time points in the presence or absence of BMJ. The transporter expression was analyzed by immunofluorescence staining. 4’,6-diamidino-2-phenylindole is represented by blue-stained nuclei. All images were captured at ×1000 magnification. Next, to determine whether the observed decrease in GLUT1 and MCT4 expression by BMJ was mediated via its effect on AMPK activation, we performed the above in vitro studies in the presence of AMPK inhibitor, compound C. Western blot analysis (Supplementary Figure 3, available at Carcinogenesis Online) confirmed our previous observation that BMJ causes AMPK activation and that BMJ-mediated apoptosis [as indicated by increased expression of cleaved poly (ADP ribose) polymerase] was compromised when BMJ-mediated AMPK activation was inhibited by compound C. Notably, even though BMJ-induced apoptosis is reversed by addition of AMPK inhibitor, inhibition of AMPK activation did not reverse BMJ-mediated decrease in the levels of glucose and lactate transporters (Supplementary Figures 4–6, available at Carcinogenesis Online). In fact, addition of compound C (alone) was found to decrease the membrane expression of both transporters—GLUT1 and MCT4 (Supplementary Figures 4–6, available at Carcinogenesis Online); instead these transporters were found to be concentrated in the cytoplasm by compound-C exposure (immunofluorescence-puncta formation in cytoplasm). On the other hand, treatment of the PanC cells with a combination of BMJ + compound C caused an overall decrease in the expression of both the transporters and limited their presence to the cytoplasm (Supplementary Figures 4–6, available at Carcinogenesis Online). Taken together, the results indicated that BMJ effect on AMPK activation is not an upstream event but was possibly a downstream event to BMJ-mediated decrease in the levels of GLUT1 and MCT4. FDG-PET and MRI demonstrate BMJ efficacy in PANC1 flank tumor xenograft model To further investigate the relevance of our in vitro outcomes with regards to in vivo scenario, we employed PANC1 xenograft mouse model for corroborating the in vitro findings. PANC1 tumor model was specifically chosen based on the enhanced metabolic modifications compared with BxPC3 cell line in response to BMJ treatment (as inferred from our in vitro NMR metabolomics data). Athymic nu/nu mice were randomized into three groups: control, BMJ-1 (prevention approach, BMJ: 200 mg/kg in 100 µl water dosing initiated 24 h post-flank cell injection) and BMJ-2 (intervention approach, BMJ: 200 mg/kg in 100 µl water dosing initiated in established tumors of size ~100 mm3). Four representative mice from each cohort underwent MRI and FDG-PET scans at: (i) baseline determination prior to animal randomization, (ii) day 0—when the tumors grew to ~100 mm3, (iii) Cycle 1 at day 6 and day 7 post-randomization and BMJ-2 initiation and (iv) Cycle 2 at study end (Figure 4A). Figure 4. Open in new tabDownload slide DW-MRI and FDG-PET scans depicting changes in PANC1 tumor-bearing animals dosed with BMJ. (A) Schematic following the experimental/imaging regimen of untreated control versus BMJ-dosed PANC1 xenografts in selected mice. The study involved chasing animals over a period of 73 days from three different treatment cohorts: control (n = 4): untreated; prevention approach (n = 4): BMJ-1 dosing initiated 24 h post-PANC1 cell injection; intervention approach (n = 4): BMJ-2 dosing initiated when tumors were established and grew to a size of ~100 mm3. FDG-PET scans and MRI imaging was performed first to set baselines (data not shown—at day 0, before randomizing animals into control and BMJ-2 groups), next at Cycle 1 (early time point of MRI at day 6 and FDG scans at day 7) and Cycle 2 involving final set of scans (MRI at day 72 and FDG scans at day 73). (B) Left panel shows representative proton density MRI scans of animals from each treatment cohort (from Cycle 1 and Cycle 2). Right panel shows quantitative imaging end points—ADCs derived from DW-MRI of untreated and BMJ-(1 and 2) treated animals. (C) Left panel displays the representative images of [18F] FDG uptake in PANC1 xenografts of Control, BMJ-1 and BMJ-2 groups at study end (Cycle 2—day 73). Right panel quantitates the changes in the SUVs between untreated and BMJ-(1 and 2) treated animals from Cycle 1 and Cycle 2 of imaging. Tumors in MRI images are pointed out by arrows (white). Tumors in FDG-PET images are pointed out by manually drawn boundaries and arrows (white). Figure 4. Open in new tabDownload slide DW-MRI and FDG-PET scans depicting changes in PANC1 tumor-bearing animals dosed with BMJ. (A) Schematic following the experimental/imaging regimen of untreated control versus BMJ-dosed PANC1 xenografts in selected mice. The study involved chasing animals over a period of 73 days from three different treatment cohorts: control (n = 4): untreated; prevention approach (n = 4): BMJ-1 dosing initiated 24 h post-PANC1 cell injection; intervention approach (n = 4): BMJ-2 dosing initiated when tumors were established and grew to a size of ~100 mm3. FDG-PET scans and MRI imaging was performed first to set baselines (data not shown—at day 0, before randomizing animals into control and BMJ-2 groups), next at Cycle 1 (early time point of MRI at day 6 and FDG scans at day 7) and Cycle 2 involving final set of scans (MRI at day 72 and FDG scans at day 73). (B) Left panel shows representative proton density MRI scans of animals from each treatment cohort (from Cycle 1 and Cycle 2). Right panel shows quantitative imaging end points—ADCs derived from DW-MRI of untreated and BMJ-(1 and 2) treated animals. (C) Left panel displays the representative images of [18F] FDG uptake in PANC1 xenografts of Control, BMJ-1 and BMJ-2 groups at study end (Cycle 2—day 73). Right panel quantitates the changes in the SUVs between untreated and BMJ-(1 and 2) treated animals from Cycle 1 and Cycle 2 of imaging. Tumors in MRI images are pointed out by arrows (white). Tumors in FDG-PET images are pointed out by manually drawn boundaries and arrows (white). Animals from each treatment cohort were followed in a time-course study using anatomical PD MRI (Cycle 1 and 2) (Figure 4B, left panel). Functional imaging end points included (early time point of day 6/7 post-randomization and at study end): tumor cellularity by DWI and tumor metabolic activity by FDG-PET. A marked change was seen in imaging parameters within the BMJ treatment groups. ADC, correlating with the extent of tissue cellularity and intact cell membranes, was shown to increase with BMJ treatment; highest ADC for BMJ-1 was observed at early and late imaging endpoints, indicating a decreased cellular density/cellularity in BMJ group. ADC values for control mice (1.05 ± 0.33, ×103 mm2/s) in Cycle 1 increased with BMJ treatment in BMJ-1 (1.56 ± 0.03, ×103 mm2/s) and BMJ-2 (1.48 ± 0.04, ×103 mm2/s) groups. Although the ADCs for controls in Cycle 2 dropped (0.92 ± 0.11, ×103 mm2/s) compared with Cycle 1 with accompanying increases in tumor volumes, BMJ treatment continued to maintain higher ADCs in treated mice for both BMJ-1 (1.15 ± 0.08, ×103 mm2/s) and BMJ-2 (1.12 ± 0.25, ×103 mm2/s) groups (Figure 4B, right panel). As FDG uptake and accumulation assessment by PET marks the carbohydrate metabolism rate, reflecting the cellular metabolic activity, BMJ treatment decreased FDG uptake in the tumors as presented by the low SUVs and represented in the images (Figure 4C). The SUVs recorded for Cycle 1 were higher in controls (2.1 ± 0.11) compared with BMJ-1- (0.77 ± 0.08) and BMJ-2- (1.42 ± 0.25) treated animals. For Cycle 2, the recorded SUVs for controls were much higher than Cycle 1 (5.2 ± 1.04), whereas BMJ-administered animals displayed decreased SUVs in BMJ-1 (2.57 ± 0.35) and BMJ-2 (4.27 ± 2.15) groups compared with controls as depicted in the images and bars (Figure 4C). Comparative anti-PanC tumor efficacy of preventive and intervention dosing approach with BMJ Longitudinal assessment of tumor volumes throughout the course of the study indicated that tumor volumes of BMJ groups in both prevention and intervention protocols were overall decreased relative to the controls. Additionally, BMJ feeding did not show any observable toxicity in terms of non-significant changes in body weights (data not shown) compared with controls. At study end, it was observed that BMJ administration via the prevention approach caused a significant (~89%, P ≤ 0.001) decrease, whereas the intervention approach caused an 80% (P ≤ 0.001) decrease in the tumor volumes compared with untreated controls (Figure 5A). There was, however, no significant difference in the tumor volumes between the two BMJ groups at study end. Interestingly, it was appreciable that in the intervention group where the BMJ-2 dosing was initiated in established tumors of ~100 mm3 volume, the study end tumors after BMJ dosing remained arrested in this size range only as compared with untreated control tumors, which had increased to a size of ~450 mm3 at study completion. Figure 5. Open in new tabDownload slide BMJ treatment decreases tumor volumes in PANC1 xenografts and exerts efficacy via inhibiting cell proliferation, inducing apoptosis and decreasing the microvessel density. (A) Tumor volumes were plotted as a function of time for control (untreated), BMJ-1 (prevention approach) and BMJ-2 (intervention approach), followed by a difference in tumor volumes at the end of study (day 73—right panel). (B) Representative images from IHC analyses of Ki67, cleaved (C)-caspase3 and CD31 with a quantitative representation of the staining from each treatment group. All images were captured at ×400 magnification. Data shown were a mean of ~10–12 animals per treatment group. ***P ≤ 0.001 and *P ≤ 0.05. Figure 5. Open in new tabDownload slide BMJ treatment decreases tumor volumes in PANC1 xenografts and exerts efficacy via inhibiting cell proliferation, inducing apoptosis and decreasing the microvessel density. (A) Tumor volumes were plotted as a function of time for control (untreated), BMJ-1 (prevention approach) and BMJ-2 (intervention approach), followed by a difference in tumor volumes at the end of study (day 73—right panel). (B) Representative images from IHC analyses of Ki67, cleaved (C)-caspase3 and CD31 with a quantitative representation of the staining from each treatment group. All images were captured at ×400 magnification. Data shown were a mean of ~10–12 animals per treatment group. ***P ≤ 0.001 and *P ≤ 0.05. Next, IHC analysis of efficacy markers such as Ki67 (tumor cell proliferation), cleaved-caspase 3 (apoptosis) and CD31 (microvessel density for angiogenesis) in control and BMJ-fed tumors was performed (Figure 5B). Results indicated that irrespective of the treatment approach applied (prevention or intervention), BMJ administration caused a significant decrease in tumor cell proliferation as evidenced by less percentage of positively stained Ki67 cells in BMJ-1 (~12%) and BMJ-2 (~18%) groups compared with untreated controls (~38%) (Figure 5B). Interestingly, a marked increase in apoptosis induction was also observed in the tumor tissues from the BMJ-fed mice in the prevention approach, where scores for cleaved-caspase 3 drastically increased in BMJ-1 animals (3.5) compared with controls (<1; Figure 5B). However, not much effect on cleaved-caspase 3 expression was observed in the tumors from intervention group, which displayed a slight increase in staining relative to controls, although not significant (Figure 5B). Regarding microvessel density of the xenografts, both prevention and treatment approaches showed a strong and significant decrease in CD31 immunoreactivity scores compared with controls indicating that angiogenesis process was being targeted by BMJ irrespective of when the drug treatments were initiated (Figure 5B). Together, these results indicated that even though both prevention (BMJ-1) and intervention (BMJ-2) approaches displayed significant efficacy in inhibiting PanC tumor growth, the mechanisms underlying the potential benefits could slightly differ between the approaches as indicated by significant apoptosis induction in prevention approach (unlike when the treatment is more intervention focused). In addition, antiangiogenic effect is more pronounced in prevention approach versus the intervention approach, indicating that the stage of tumorigenesis determines the mechanism via which BMJ mediates its anticancer effects. BMJ targets and modulates PanC cell metabolome in vivo Next, we assessed BMJ effect on key metabolic molecules/enzymes in the above tumor tissues. Since NMR studies indicated toward modulation of lactate levels and targeting of its extracellular efflux as one of the significant changes induced by BMJ in the PanC cells, we determined the expression level of the enzyme LDH, which is involved in the conversion of pyruvate to lactate. Importantly, increased tumor LDH levels have been previously associated with PanC progression and poor prognostic outcome (38). Notably, IHC staining for LDH expression in tumors (Figure 6A) demonstrated a significant decrease in LDH levels in both prevention- (BMJ-1) and intervention- (BMJ-2) treated groups (≤1.5) compared with untreated controls, thereby supporting the in vitro findings. Consistent with our previously published study showing that BMJ activates AMPK in PanC cells in culture and MiaPaCa2 mouse xenograft models (23), we also found that BMJ feeding results in a significant increase (>3-fold) in pAMPK tumor levels in both BMJ-dosed prevention and intervention groups (Figure 6A). Figure 6. Open in new tabDownload slide BMJ dosing modulates tumor LDH and pAMPK levels and also decreases glucose and lactate transporter expression in vivo. (A) PANC1 xenografts were probed for LDH and pAMPKTh172 levels by IHC analysis and the positive brown-staining was quantified. Images were acquired at ×400 magnification. (B) Representative images from each treatment cohort depicting GLUT1 (green) and MCT4 (red) expression in PANC1 xenograft tissue of control, BMJ-1 and BMJ-2 groups by immunofluorescence staining. Images were acquired at ×200 with digitally magnified insets. ***P ≤ 0.001. (C) Determination of cellular fluorescence from fluorescent images depicted in ‘B’ using Image J software. Bars indicate corrected total cell fluorescence (±standard deviation) obtained for expression of GLUT1 and MCT4 levels from three different sites. For immunofluorescence in tissues, A1-HD confocal microscope from Nikon and NIS-Elements confocal microscope imaging software (Nikon) were used for data capturing. Quantification of immunofluorescence was done using Image J software. Figure 6. Open in new tabDownload slide BMJ dosing modulates tumor LDH and pAMPK levels and also decreases glucose and lactate transporter expression in vivo. (A) PANC1 xenografts were probed for LDH and pAMPKTh172 levels by IHC analysis and the positive brown-staining was quantified. Images were acquired at ×400 magnification. (B) Representative images from each treatment cohort depicting GLUT1 (green) and MCT4 (red) expression in PANC1 xenograft tissue of control, BMJ-1 and BMJ-2 groups by immunofluorescence staining. Images were acquired at ×200 with digitally magnified insets. ***P ≤ 0.001. (C) Determination of cellular fluorescence from fluorescent images depicted in ‘B’ using Image J software. Bars indicate corrected total cell fluorescence (±standard deviation) obtained for expression of GLUT1 and MCT4 levels from three different sites. For immunofluorescence in tissues, A1-HD confocal microscope from Nikon and NIS-Elements confocal microscope imaging software (Nikon) were used for data capturing. Quantification of immunofluorescence was done using Image J software. Immunofluorescence imaging showed a significant decrease in GLUT1 expression in the tumor tissues from both BMJ-dosed prevention and intervention groups compared with untreated controls (Figure 6B-C, left panel). MCT4 expression was also significantly decreased (the decrease was relatively more significant than the decrease in GLUT 1 expression), where tumor tissues from both BMJ-dosed prevention and intervention groups showed a marked reduction in this transporter protein levels compared with untreated controls (Figure 6B and C, right panel). Of note, even in the tissues, BMJ only caused a decrease in total levels of the transporters and did not impact the subcellular localization (cytoplasmic versus membrane levels) of the GLUT1 and MCT4 transporters (Supplementary Figure 2, available at Carcinogenesis Online). Since an upregulated MCT4 expression is imperative in controlling intracellular pH and lactate-based metabolism (specifically lactate shuttling) to aid in tumor growth and survival under stressful conditions (39), our findings suggest that BMJ effectively targets and modulates PanC cell metabolism in vivo, with lactate being a major molecule of interest. Discussion PanC cells like other tumor cells undergo cellular reprograming to meet their bioenergetic and biosynthetic demands, with glycolytic shift emerging as the primary metabolic hallmark in the process of carcinogenesis (40). The resulting malignant metabolic phenotype is programed to convert glucose into lactate, even in the presence of sufficient oxygen levels, which drives the pathological requirements of cancerous cells, enabling rapid tumor growth and progression as opposed to normal cells (40). Over two-thirds of PanC patients deal with an impaired glucose tolerance as noted by improved glucose levels post-surgical resection of PanC, suggesting a strong correlation between PanC and altered glucose metabolism (41). With regards to the generated lactate, its homeostasis in cells (both normal and cancerous) occurs via facilitative and proton-linked lactate transporters (MCTs); these transporters are responsible for flushing the excess levels of lactate outside the plasma membrane and inhibiting continued glycolysis, thereby limiting the toxic buildup of lactate and intracellular acidification (42,43). Cells with an increasingly glycolytic phenotype use MCTs for exporting high levels of intracellular LDH-generated lactate outside the cell, thereby establishing MCTs as the key regulators of intracellular pH and lactate, where increased lactate export correlates with highly aggressive PDACs (42). For maintaining a higher rate of glycolysis, a characteristic feature of PDAC, the cells are required to secrete more lactate, since increased intracellular lactate accumulation causes inhibition of glycolysis by negative regulation of LDH activity (44). Together, both glycolysis and lactate pathways are recognized as important therapeutic targets for PanC management. Bitter melon is a widely consumed vegetable in Asia and Africa and its various forms, such as aqueous extract, alkaline chloroform extract, pulp, aqueous extract powder and whole plant extract, are shown to possess antidiabetic and hypoglycemic activity in cell culture and animal and clinical studies (20). Notably, diabetes is recognized as an important risk factor for pancreatic cancer; almost half the PanC patient population are diabetic at diagnosis (45). Our recent studies have reported an increased AMPK phosphorylation and activation with BMJ exposure in PanC cells in culture and mouse tumor xenografts (23). This formed the basis to study BMJ-induced changes in PanC cell metabolome; specifically, we utilized 1H, 13C and 31P NMR-spectroscopy-associated metabolic profiling coupled with multivariate statistical analysis for an in-depth analysis of PanC cell metabolome after BMJ treatments. Notably, BMJ treatment for 72 h efficiently targeted the metabolic pathways, particularly glycolysis and lactate pathways in PanC cell lines. Its exposures induced a significant decrease in glucose uptake, lactate export and ATP/ADP ratios, thus, signifying restricted uptake of glucose in the energy-deprived PanC cells. Our results suggest a ‘metabolic switch’ during BMJ exposure from early to late time points. Briefly, there was an increase in lactate export in PANC1 cells (data not shown) during first 4 h of BMJ exposure (associated with cell growth inhibition), followed by increase in lactate buildup inside the PanC cells by 72 h of BMJ treatment—a plausible cause of intracellular acidification leading to cell death induction at later time points. These biological effects were found to be mediated by a decrease in expression of the glucose and lactate transporters GLUT1 and MCT4 by BMJ both in vitro and in vivo (46). The effect on these transporters by BMJ is a highly significant finding relevant to alteration of PanC metabolome in the presence and absence of functional KRAS status. This is because oncogenic KRAS has been shown (particularly reported in cancer cells harboring mutated/oncogenic KRAS) to promote glucose uptake by increased GLUT1 expression resulting in higher glycolytic activity, ATP generation and lactate synthesis leading to poor tumor prognosis and cancer progression. This could explain why BMJ shows significant modulation of the metabolites in the mutated KRAS cell line (PANC1) compared with its mild effect in wild-type KRAS harboring BxPC3 PanC cell line (47). Furthermore, mutant KRAS has been also shown to be associated with enhanced expression of hexokinases 1 and 2, phosphofructokinase-1 and most importantly LDH, the genes encoding rate limiting enzymes of glycolytic pathway, which results in elevated glycolytic flux (44). Importantly, we observed that LDH, which is responsible for catalyzing the conversion of pyruvate to lactate, was also decreased in BMJ-exposed PANC1 xenograft tumors, indicating that LDH level modulation by BMJ could be another factor contributing toward its regulation of aberrant PanC cell metabolism. This postulation is further supported by increased pAMPK activity in PANC1 xenografts following BMJ treatment, thereby corroborating our earlier findings (23). The use of imaging platforms for tumor cellularity and metabolome assessments further indicated that BMJ-treated PanC tumors were associated with (i) higher ADC values as calculated from DW-MRI (indicating decrease in tissue cellularity) and (ii) decreased FDG SUVs as detailed from PET scans (indicating decrease in metabolic vitality). These results are highly significant from a clinical stand point since both low ADC and high SUV values are clinically established radiological biomarkers for tumor aggressiveness and poor survival (48,49). The in vivo imaging end-point outcomes in the present study with BMJ (almost similar in prevention and intervention approaches) provide a strong translational evidence of antitumor efficacy of BMJ irrespective of tumor stage. Collectively, our findings employing the NMR-metabolomics approach and in vivo metabolic imaging multiplatforms provide an in-depth account of underlying mechanism of BMJ efficacy associated with modulating PanC cell metabolism and identified primarily lactate efflux and glucose metabolism, specifically GLUT1 and MCT4 transporters, as the potential metabolic targets underlying BMJ efficacy in PanC. Funding This work was supported by the National Cancer Institute R01 grant CA195708. NMR and imaging studies were performed at the Animal Imaging Shared Resources, University of Colorado Cancer Center (P30CA046934, National Cancer Institute). Imaging experiments were performed in the University of Colorado Denver, Anschutz Medical Campus Advanced Light Microscopy Core. Conflict of Interest Statement: The authors declare that there are no conflicts to disclose. Abbreviations Abbreviations ADC apparent diffusion coefficient ADP adenosine diphosphate AMP adenosine monophosphate AMPK AMP-activated protein kinase ATP adenosine triphosphate BMJ bitter melon juice DMEM Dulbecco’s modified Eagle’s medium DMSO dimethyl sulfoxide IHC immunohistochemistry LDH lactate dehydrogenase NMR nuclear magnetic resonance PanC pancreatic cancer PDAC pancreatic ductal adenocarcinoma PLS-DA partial least square discriminant analysis SUV standardized uptake values References 1. Siegel , R.L. et al. ( 2018 ) Cancer statistics, 2018 . CA Cancer J. Clin. , 68 , 7 – 30 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Adamska , A. et al. ( 2017 ) Pancreatic ductal adenocarcinoma: current and evolving therapies . Int. J. Mol. Sci. , 18, 1338 . WorldCat 3. Hosein , A.N. et al. ( 2018 ) Pancreatic cancer metabolism: molecular mechanisms and clinical applications . Curr. Oncol. Rep. , 20 , 56 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Cairns , R.A. et al. ( 2011 ) Regulation of cancer cell metabolism . Nat. Rev. Cancer , 11 , 85 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Hardie , D.G . ( 2014 ) AMP-activated protein kinase: maintaining energy homeostasis at the cellular and whole-body levels . Annu. Rev. Nutr. , 34 , 31 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Shackelford , D.B. et al. ( 2009 ) The LKB1-AMPK pathway: metabolism and growth control in tumour suppression . Nat. Rev. Cancer , 9 , 563 – 575 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Zadra , G. et al. ( 2015 ) Dissecting the dual role of AMPK in cancer: from experimental to human studies . Mol. Cancer Res. , 13 , 1059 – 1072 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Feig , C. et al. ( 2012 ) The pancreas cancer microenvironment . Clin. Cancer Res. , 18 , 4266 – 4276 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Xie , D. et al. ( 2015 ) Pancreatic cancer stromal biology and therapy . Genes Dis. , 2 , 133 – 143 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Hawley , S.A. et al. ( 2003 ) Complexes between the LKB1 tumor suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream kinases in the AMP-activated protein kinase cascade . J. Biol. , 2 , 28 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Zheng , B. et al. ( 2009 ) Oncogenic B-RAF negatively regulates the tumor suppressor LKB1 to promote melanoma cell proliferation . Mol. Cell , 33 , 237 – 247 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Duan , W. et al. ( 2017 ) Desmoplasia suppression by metformin-mediated AMPK activation inhibits pancreatic cancer progression . Cancer Lett. , 385 , 225 – 233 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Chen , K. et al. ( 2017 ) Loss of AMPK activation promotes the invasion and metastasis of pancreatic cancer through an HSF1-dependent pathway . Mol. Oncol. , 11 , 1475 – 1492 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Perera , R.M. et al. ( 2015 ) Pancreatic cancer metabolism: breaking it down to build it back up . Cancer Discov. , 5 , 1247 – 1261 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Camelo , F. et al. ( 2018 ) The intricate metabolism of pancreatic cancers . Adv. Exp. Med. Biol. , 1063 , 73 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat 16. McMillin , D.W. et al. ( 2013 ) The role of tumour-stromal interactions in modifying drug response: challenges and opportunities . Nat. Rev. Drug Discov. , 12 , 217 – 228 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Von Hoff , D.D. et al. ( 2013 ) Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine . N. Engl. J. Med. , 369 , 1691 – 1703 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Faris , J.E. et al. ( 2013 ) FOLFIRINOX in locally advanced pancreatic cancer: the Massachusetts General Hospital Cancer Center experience . Oncologist , 18 , 543 – 548 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Dhar , D. et al. ( 2018 ) Mechanisms and drug targets for pancreatic cancer chemoprevention . Curr. Med. Chem. , 25 , 2545 – 2565 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Raina , K. et al. ( 2016 ) Promise of bitter melon (Momordica charantia) bioactives in cancer prevention and therapy . Semin. Cancer Biol. , 40–41 , 116 – 129 . Google Scholar Crossref Search ADS PubMed WorldCat 21. Tsai , C.H. et al. ( 2012 ) Wild bitter gourd improves metabolic syndrome: a preliminary dietary supplementation trial . Nutr. J. , 11 , 4 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Fuangchan , A. et al. ( 2011 ) Hypoglycemic effect of bitter melon compared with metformin in newly diagnosed type 2 diabetes patients . J. Ethnopharmacol. , 134 , 422 – 428 . Google Scholar Crossref Search ADS PubMed WorldCat 23. Kaur , M. et al. ( 2013 ) Bitter melon juice activates cellular energy sensor AMP-activated protein kinase causing apoptotic death of human pancreatic carcinoma cells . Carcinogenesis , 34 , 1585 – 1592 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Somasagara , R.R. et al. ( 2015 ) Bitter melon juice targets molecular mechanisms underlying gemcitabine resistance in pancreatic cancer cells . Int. J. Oncol. , 46 , 1849 – 1857 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Dhar , D. et al. ( 2018 ) Bitter melon juice exerts its efficacy against pancreatic cancer via targeting both bulk and cancer stem cells . Mol. Carcinog. , 57 , 1166 – 1180 . Google Scholar Crossref Search ADS PubMed WorldCat 26. Kumar , S. et al. ( 2014 ) Silibinin strongly inhibits the growth kinetics of colon cancer stem cell-enriched spheroids by modulating interleukin 4/6-mediated survival signals . Oncotarget , 5 , 4972 – 4989 . Google Scholar PubMed WorldCat 27. Klawitter , J. et al. ( 2009 ) Time-dependent effects of imatinib in human leukaemia cells: a kinetic NMR-profiling study . Br. J. Cancer , 100 , 923 – 931 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Serkova , N.J. et al. ( 2009 ) Metabolomics of cancer . Methods Mol. Biol. , 520 , 273 – 295 . Google Scholar Crossref Search ADS PubMed WorldCat 29. Raina , K. et al. ( 2013 ) Inositol hexaphosphate inhibits tumor growth, vascularity, and metabolism in TRAMP mice: a multiparametric magnetic resonance study . Cancer Prev. Res. (Phila). , 6 , 40 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Deep , G. et al. ( 2017 ) Silibinin inhibits hypoxia-induced HIF-1α-mediated signaling, angiogenesis and lipogenesis in prostate cancer cells: in vitro evidence and in vivo functional imaging and metabolomics . Mol. Carcinog. , 56 , 833 – 848 . Google Scholar Crossref Search ADS PubMed WorldCat 31. Serkova , N.J. et al. ( 2016 ) Metabolic imaging to assess treatment response to cytotoxic and cytostatic agents . Front. Oncol. , 6 , 152 . Google Scholar Crossref Search ADS PubMed WorldCat 32. Frey , L. et al. ( 2014 ) ADC mapping and T1-weighted signal changes on post-injury MRI predict seizure susceptibility after experimental traumatic brain injury . Neurol. Res. , 36 , 26 – 37 . Google Scholar Crossref Search ADS PubMed WorldCat 33. Schlaepfer , I.R. et al. ( 2015 ) Inhibition of lipid oxidation increases glucose metabolism and enhances 2-deoxy-2-[(18)F]Fluoro-D-glucose uptake in prostate cancer mouse xenografts . Mol. Imaging Biol. , 17 , 529 – 538 . Google Scholar Crossref Search ADS PubMed WorldCat 34. Shukla , S.K. et al. ( 2017 ) MUC1 and HIF-1alpha signaling crosstalk induces anabolic glucose metabolism to impart gemcitabine resistance to pancreatic cancer . Cancer Cell , 32 , 71–87.e7 . Google Scholar PubMed WorldCat 35. Tentler , J.J. et al. ( 2010 ) Assessment of the in vivo antitumor effects of ENMD-2076, a novel multitargeted kinase inhibitor, against primary and cell line-derived human colorectal cancer xenograft models . Clin. Cancer Res. , 16 , 2989 – 2998 . Google Scholar Crossref Search ADS PubMed WorldCat 36. Basturk , O. et al. ( 2011 ) GLUT-1 expression in pancreatic neoplasia: implications in pathogenesis, diagnosis, and prognosis . Pancreas , 40 , 187 – 192 . Google Scholar Crossref Search ADS PubMed WorldCat 37. Baek , G. et al. ( 2014 ) MCT4 defines a glycolytic subtype of pancreatic cancer with poor prognosis and unique metabolic dependencies . Cell Rep. , 9 , 2233 – 2249 . Google Scholar Crossref Search ADS PubMed WorldCat 38. Xiao , Y. et al. ( 2017 ) Prognostic relevance of lactate dehydrogenase in advanced pancreatic ductal adenocarcinoma patients . BMC Cancer , 17 , 25 . Google Scholar Crossref Search ADS PubMed WorldCat 39. Parks , S.K. et al. ( 2013 ) Disrupting proton dynamics and energy metabolism for cancer therapy . Nat. Rev. Cancer , 13 , 611 – 623 . Google Scholar Crossref Search ADS PubMed WorldCat 40. Tennant , D.A. et al. ( 2010 ) Targeting metabolic transformation for cancer therapy . Nat. Rev. Cancer , 10 , 267 – 277 . Google Scholar Crossref Search ADS PubMed WorldCat 41. Pannala , R. et al. ( 2009 ) New-onset diabetes: a potential clue to the early diagnosis of pancreatic cancer . Lancet. Oncol. , 10 , 88 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat 42. Halestrap , A.P . ( 2012 ) The monocarboxylate transporter family–structure and functional characterization . IUBMB Life , 64 , 1 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 43. Dimmer , K.S. et al. ( 2000 ) The low-affinity monocarboxylate transporter MCT4 is adapted to the export of lactate in highly glycolytic cells . Biochem. J. , 350(Pt 1) , 219 – 227 . Google Scholar Crossref Search ADS PubMed WorldCat 44. Ying , H. et al. ( 2012 ) Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism . Cell , 149 , 656 – 670 . Google Scholar Crossref Search ADS PubMed WorldCat 45. Ben , Q. et al. ( 2011 ) Diabetes mellitus and risk of pancreatic cancer: A meta-analysis of cohort studies . Eur. J. Cancer , 47 , 1928 – 1937 . Google Scholar Crossref Search ADS PubMed WorldCat 46. McCarty , M.F. et al. ( 2010 ) Manipulating tumor acidification as a cancer treatment strategy . Altern. Med. Rev. , 15 , 264 – 272 . Google Scholar PubMed WorldCat 47. Yun , J. et al. ( 2009 ) Glucose deprivation contributes to the development of KRAS pathway mutations in tumor cells . Science , 325 , 1555 – 1559 . Google Scholar Crossref Search ADS PubMed WorldCat 48. Jafar , M.M. et al. ( 2016 ) Diffusion-weighted magnetic resonance imaging in cancer: Reported apparent diffusion coefficients, in-vitro and in-vivo reproducibility . World J. Radiol. , 8 , 21 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat 49. Wilson , J.M. et al. ( 2017 ) Correlation of 18F-fluorodeoxyglucose positron emission tomography parameters with patterns of disease progression in locally advanced pancreatic cancer after definitive chemoradiotherapy . Clin. Oncol. (R. Coll. Radiol.) , 29 , 370 – 377 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For Permissions, please email: email@example.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Carcinogenesis – Oxford University Press
Published: Sep 18, 2019
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera