Gemcitabine is the first-line chemotherapy for pancreatic cancer. To overcome the often-acquired gemcitabine resistance, other drugs are used in combination with gemcitabine. It is well-known that cancer cells reprogram cellular metabolism, coupled with the up-regulation of selective nutrient transporters to feed into the altered metabolic pathways. Our previous studies have demonstrated that the amino acid transporter SLC6A14 is markedly up-regulated in pancreatic cancer and that it is a viable therapeutic target. α-Methyltryptophan (α-MT) is a blocker of SLC6A14 and is effective against pancreatic cancer in vitro and in vivo. In the present study, we tested the hypothesis that α-MT could synergize with gemcitabine in the treatment of pancreatic cancer. We investigated the effects of combination of α-MT and gemcitabine on proliferation, migration, and apoptosis in a human pancreatic cancer cell line, and examined the underlying mechanisms using 1H-NMR-based metabolomic analysis. These studies examined the intracellular metabolite profile and the extracellular metabolite profile separately. Combination of α-MT with gemcitabine elicited marked changes in a wide variety of metabolic pathways, particularly amino acid metabolism with notable alterations in pathways involving tryptophan, branched-chain amino acids, ketone bodies, and membrane phospholipids. The metabolomic profiles of untreated control cells and cells treated with gemcitabine or α-MT were distinctly separable, and the combination regimen showed a certain extent of overlap with the individual α-MT and gemcitabine groups. This represents the first study detailing the metabolomic basis of the anticancer efficacy of gemcitabine, α-MT and their combination.

Introduction

Pancreatic cancer is the fourth leading cause of cancer-related death worldwide [1]. It is a deadly malignancy characterized by a few early symptoms before reaching its advanced stage [2]. According to recent statistics, 458 918 new cases of pancreatic cancer were found, along with 432 242 new deaths in 2018 worldwide [3]. Currently, surgery is still the most effective therapy for pancreatic cancer, especially at the early stage [4]. Unfortunately, due to the rapid progression and late diagnosis, about half of the patients with this cancer would have already reached the metastatic stage at the time of initial diagnosis, and thus lose the opportunity for radical surgery [5]. As such, the management of patients with advanced pancreatic cancer involves mostly palliative therapy. Gemcitabine is a pyrimidine nucleoside analog that is commonly used in the treatment of solid cancers such as the breast, lung, and pancreatic cancers. It is the first-line treatment for advanced pancreatic cancer [6]. Other therapeutics are often used jointly with gemcitabine to overcome the frequently acquired gemcitabine resistance, but none of these combination strategies has produced marked improvements in the clinical outcome. Therefore, the identification of new agents that would synergize with gemcitabine in blocking the growth and proliferation of pancreatic cancer cells is sorely needed. In the present study, we focused on one of the well-documented cancer cell-specific metabolic alterations as a potential drug target for exploitation as a combination therapy with gemcitabine in pancreatic cancer.

Altered cellular metabolism has long been recognized as a distinctive feature of cancer [7]. For example, cancer cells have an increased demand for amino acids because of their rapid proliferation rate. Both normal cells and cancer cells could synthesize non-essential amino acids; however, the synthesis rate cannot meet the increased demand in cancer cells. Therefore, selective amino acid transporters that import exogenous amino acids are up-regulated in cancer cells [8,9]. Most amino acid transporters have narrow substrate specificity; in contrast, SLC6A14 (ATB0,+) has a broad substrate specificity with the ability to transport neutral as well as cationic amino acids [10]. Most published reports on amino acid transport in cancer cells have focused on three transporters, namely SLC1A5/ASCT2 [11], SLC7A5/LAT1 [12], and SLC7A11/xCT [13], but these transporters do not have the ability to transport as many amino acids as SLC6A14 does. Furthermore, SLC1A5, SLC7A5, and SLC7A11 are obligatory exchangers; in contrast, SLC6A14 is a unidirectional transporter. The expression of SLC6A14 in healthy tissues is fairly restricted [14], but highly up-regulated in many solid tumors, including cervical [15], colorectal [16], breast [17], and pancreatic cancers [18]. Therefore, SLC6A14 is an actionable drug target for cancer therapy. We have shown that α-methyltryptophan (α-MT) is a blocker of SLC6A14 transport function [19]. Blockade of SLC6A14 with α-MT inhibits the growth and proliferation of breast [17] and pancreatic cancer cells [18] in vitro and in vivo. This represents a novel molecular target for cancer treatment because, to date, none of the chemotherapeutics currently in clinical use targets a nutrient transporter. Since gemcitabine and α-MT target different metabolic pathways, we initiated the present study to determine whether blockade of SLC6A14 transport function with α-MT would synergize with gemcitabine to elicit a more efficacious therapeutic effect in pancreatic cancer.

Here we investigated the anticancer effects and the underlying molecular mechanisms of gemcitabine and α-MT when used either individually or in combination by assessing not only the traditional cell-biological effects but also metabolomic profiles. Monitoring the changes in the levels of various metabolites inside the cells as well as in the extracellular milieu using the metabolomics approach could provide useful information on the rates and directions of metabolic pathways within the cells [20,21]. Nuclear magnetic resonance (NMR)-based metabolomics enables qualitative and quantitative analysis of most metabolites in cells, and hence provides a broad and unbiased view of the differences in metabolic pathways in cells between normal and pathological conditions [22]. NMR-based metabolomics have been widely used to study the antitumor mechanisms of drugs on malignant cells [23–26]. For example, Lv et al. [27] were able to examine the anticancer mechanism of metformin in HepG2 using the NMR-based metabolomics approach. 1H-NMR has become a highly valuable method to directly analyze cellular metabolic changes in response to drug exposure. In the present study, we determined the effect of combining α-MT with gemcitabine on proliferation and migration of human pancreatic cancer cells in vitro and used the 1H-NMR, combined with multivariate statistical methods, to evaluate the metabolic responses in the cancer cells in response to these two drugs, either individually or in combination.

Materials and methods

Reagents

α-Methyltryptophan (α-MT) and gemcitabine were purchased from Shenzhen Qimeike Biotechnology Co., Ltd. The human pancreatic cancer cell line, PANC-1, was obtained from the Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China). RPMI 1640 medium, 0.25% trypsin-EDTA, and penicillin/streptomycin were purchased from GE Healthcare Life Sciences (Marlborough, MA, U.S.A.). Fetal bovine serum (FBS) was purchased from Life Technologies Co., Ltd (Paisley, U.K.). Methylthiazolyldiphenyl-tetrazolium bromide (MTT) was purchased from Aladdin Biochemical Co., Ltd. (Shanghai, China). Hoechst 33342 and Giemsa Stain were purchased from Beijing Solarbio Science & Technology Co., Ltd (Beijing, China). All other chemicals and reagents were of analytical grade.

Cell culture

PANC-1 cells were cultured in RPMI medium 1640 (Hyclone) with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Cells were cultured in a humidified atmosphere with 5% CO2 at 37°C (Thermo Fisher Scientific, U.S.A.).

Cell viability assay

MTT assay was performed to assess the anticancer effect of the drugs against PANC-1 cells. In brief, 100 µl PANC-1 cell suspension was seeded at a cell density of 1 × 103 cells per well in a 96-well plate. After 12 h, the cells were exposed to different concentrations of gemcitabine (0.01, 0.02, 0.04, 0.08, 0.1, 0.2, 0.4 µM), and α-MT (0.5, 1, 2, 2.5, 3, 3.5, 4, 4.5 mM) for 48 h. The control group was treated with the same volume of saline for 48 h. After the treatments, 20 µl of MTT (5 mg/ml) was added to each well and incubated for 4 h, then the medium was removed carefully, following which 150 µl DMSO was added to solubilize the formazan crystalline product. The absorbance of the resultant solutions was determined at 490 nm on a microplate reader (Multiskan Mk3, Thermo Labsystems, Finland). Cell viability was calculated using the untreated cells as control and empty well with the same procedure as blank. Each experiment was repeated three times.

The combination index (CI) was further calculated using the isobologram equation [28,29]: CI = a/A + b/B. In this equation, a is the IC50 of α-MT in combo treatment, and b is the IC50 of gemcitabine in combo treatment; A is the IC50 of α-MT, and B is the IC50 of gemcitabine, when used individually by themselves. CI < 0.9, indicates a synergistic effect; CI > 1.1, indicates an antagonistic effect; 0.9 ≤ CI ≤ 1.1, indicates an additive effect.

Hoechst staining analysis

PANC-1 cells were seeded into six-well plates with a density of 1.0 × 105 cells/well. After 12 h allowing for adherence, the cells were treated with PBS, gemcitabine (0.1 µM), α-MT (2.0 mM), and gemcitabine + α-MT (combo), respectively. After 48 h treatment, the cells were stained with 0.5 ml Hoechst 33342 (2 µg/ml) for 10 min in the dark and then washed with PBS three times. The fluorescent images of nuclei were observed and recorded by fluorescence microscopy (IX73, Olympus, Tokyo, Japan). The experiment was repeated three times.

Wound healing assay

PANC-1 cells were seeded into six-well plates. When the confluence of the cells reached ∼90%, the cells were scratch-wounded with a sterile 200 µl pipette tip and washed three times to remove cell debris. After that, the cells were incubated with gemcitabine (0.2 µM) with or without α-MT (2.5 mM). The cells were allowed to grow for another 24 h after which the wound area was photographed with microscope (Nikon, Japan). Scratched areas in each well were measured. For each group, the cell scratch healing rate was calculated as a percent wound closure relative to baseline. This experiment was done in triplicate.

Clonogenic assay

Cells were seeded in six-well plates at a low density (5000 cells/well). After adherence, cells were treated with PBS, gemcitabine (0.1 µM), α-MT (2 mM) and the combination of gemcitabine (0.1 µM) and α-MT (2 mM). If any plate reached up to 60%–70% confluence, the treatment was stopped and the cells were fixed using ice-cold 100% methanol. After 20 min, methanol was removed, and 2 ml of Giemsa Stain was added. The cells were kept in room temperature for 2 h. Then, the cells were washed with PBS three times, and photographed. Following that, the stain was dissolved with 1% sodium dodecyl sulfate in 0.2 M NaOH, and the absorbance at 630 nm was recorded.

Extraction of extracellular metabolites

The PANC-1 cells at a density of 1 × 106 cells/dish were cultured in conditioned media (control, gemcitabine, α-MT, and gemcitabine + α-MT) for 48 h. Metabolites in the culture medium were extracted as described in our previously published protocol [30]. Briefly, 1 ml of the culture medium from each dish was transferred to a 15 ml centrifuge tube for extracellular metabolite extraction. Then, 2 ml ice-cold methanol and 2 ml ice-cold chloroform were added into the medium. The mixture was centrifuged at 8000 rpm for 20 min at 4°C. The supernatant was lyophilized and stored at −80°C for metabolite analysis.

Extraction of intracellular metabolites

The PANC-1 cells at a density of 1 × 106 cells/dish were cultured in conditioned media (control, gemcitabine, α-MT, and gemcitabine + α-MT) for 48 h. The cells were then washed with ice-cold PBS. The cells were trypsinized and collected by centrifugation. The methanol-chloroform-water was used for the extraction of metabolites from the cells as described previously [31]. Briefly, cell pellets were resuspended in 300 µl ice-cold methanol and 150 µl ice-cold chloroform and sonicated on ice for 30 min. After that, 225 µl ice-cold chloroform and 225 µl water were added to form an emulsion, which was vortexed thoroughly and kept on ice for 15 min. The mixture was centrifugated at 12 000 rpm for 20 min at 4°C. The supernatant was collected, lyophilized, and stored at −80°C for metabolite analysis.

1H-NMR-based Metabolomic analysis

The dried metabolite extracts were dissolved in 500 µl D2O containing 0.1 mM TSP as a chemical shift reference. The solution was gently mixed and centrifuged at 12 000 rpm for 15 min to remove the sediments. Then the samples were transferred into 5 mm NMR tubes. The NMR data of cell extracts and medium extracts were collected using a Bruker AVANCE III 600.13 MHz spectrometer (Bruker BioSpin, Rheinstetten, Germany) equipped with a triple resonance probe. The 1H-NMR spectra were collected at 298 K with 64 K data points, 6 s relaxation delay, 2.65 s per scan acquisition time and a spectral width of 12 000 Hz, using a one-dimensional ZGPR pulse sequence. Specifically, 256 scans were set for extracellular samples, and 1024 scans were for intracellular samples.

Data reduction and multivariate data analysis

The NMR spectrum preprocessing and spectrum identification were conducted as previously described [30,32,33] using the human metabolome database. The 1H-NMR spectra were Fourier transformed, automatically phased, and referenced to the methyl peak of lactate (CH3, 1.33 ppm) using the Bruker Topspin 2.1 software (Bruker BioSpin, Germany). The ‘icoshift’ procedure was used to align NMR spectra in MATLAB (R2012b, The Mathworks Inc., Natick, MA, U.S.A.) [34]. The spectral regions approximately from 0.7 to 9.5 ppm excluding residual water signal (4.4–5.1 ppm for extracellular samples, 4.6–5.1 ppm for intracellular samples) were subdivided and integrated to binning data with a size of 0.01 ppm for qualitative analysis and 0.0015 ppm for quantitative analysis. The qualitative data was normalized to the total sum of the spectral intensity. The concentration of metabolites was calculated according to its peak area by reference to the TSP concentration, expressed as relative units (r.u.). The final data was calculated and presented as the compound amount in 106 cells.

Multivariate Data Analysis was performed using SIMCA 12.0 software (Umetrics, Umeå, Sweden). Partial least squares-discriminate analysis (PLS-DA) was used to examine the metabolic differences among the four groups. The parameters R2 and Q2 represent the variance and the predictive capability of the model, respectively. The corresponding variable importance in the projection (VIP) value was performed to identify the specific metabolites in distinguishing metabolic profiles [35].

Statistical analysis

The SPSS software (version 16.0) was used for the evaluation of the significant differences between groups. Data were analyzed using ANOVA and reported as the means ± SD. Significance was determined using ANOVA, with P < 0.05 considered as a statistically significant difference.

Results

Synergistic effect of gemcitabine and α-MT against pancreatic cancer cells

We first used MTT assay to determine the anticancer effect of gemcitabine and α-MT in pancreatic cancer cells (PANC-1). As shown in Figure 1A,B, both gemcitabine and α-MT were able to inhibit PANC-1 cell proliferation in a dose-dependent manner. α-MT as a single drug exhibited significant growth inhibitory effect with an IC50 value of 4.2 mM. In our previous studies, we have found that the 2.5 mM of α-MT showed a significant inhibitory effect on SLC6A14-positive tumor cells, and no noticeable effect on normal cells [19]. Here, we have confirmed these previous findings. We chose 2 mM α-MT to test its synergistic effect with gemcitabine against pancreatic cancer cells. The results showed that α-MT could significantly enhance the anticancer effect of gemcitabine (Figure 1B), and the IC50 for gemcitabine decreased from 0.4 µM to 0.1 µM, indicating a strong synergistic effect of the Combo with α-MT and gemcitabine against PANC-1 cells. We then calculated the combination index [28]. The CI value of the Combo treatment is 0.73, indicating a synergistic effect between α-MT and gemcitabine. We then used Hoechst staining to study the anticancer effect of the Combo (0.1 µM gemcitabine and 2 mM α-MT) (Figure 1C). In the Combo group, the number of apoptotic nuclei was significantly increased compared with gemcitabine alone, α-MT alone, and the control group. The scratch-wound migration assay showed that the cell migration was dramatically inhibited when exposed to the Combo (Figure 1D,E), and the clonogenic assay confirmed the enhanced anti-proliferation ability of the Combo (Figure 1F,G). Collectively, these results suggested the α-MT was able to significantly enhance the therapeutic effect of gemcitabine against PANC-1 cells.

α-MT enhanced the therapeutic effect of gemcitabine (Gem) in pancreatic cancer cells.

Figure 1.
α-MT enhanced the therapeutic effect of gemcitabine (Gem) in pancreatic cancer cells.

MTT assay was used to test the anticancer effects of (A) α-MT, and (B) Gem with or without α-MT on PANC-1 cells. (C) Hoechst staining was used to monitor apoptosis in PANC-1 cells after treatment with control, Gem (0.1 µM), α-MT (2 mM), and the Combo (Gem, 0.1 µM; α-MT, 2 mM). Red arrow, Scale bar = 500 µm. (D) The photographed wound area after various treatments. (E) Quantitative analysis of wound healing assay. (F) Clonogenic assay after various treatments. (G) Quantitative analysis of clonogenic assay by dissolving the stain and reading at 630 nm. The different letters in panels E & G indicate statistically significant differences (P < 0.05) among the values in the bars.

Figure 1.
α-MT enhanced the therapeutic effect of gemcitabine (Gem) in pancreatic cancer cells.

MTT assay was used to test the anticancer effects of (A) α-MT, and (B) Gem with or without α-MT on PANC-1 cells. (C) Hoechst staining was used to monitor apoptosis in PANC-1 cells after treatment with control, Gem (0.1 µM), α-MT (2 mM), and the Combo (Gem, 0.1 µM; α-MT, 2 mM). Red arrow, Scale bar = 500 µm. (D) The photographed wound area after various treatments. (E) Quantitative analysis of wound healing assay. (F) Clonogenic assay after various treatments. (G) Quantitative analysis of clonogenic assay by dissolving the stain and reading at 630 nm. The different letters in panels E & G indicate statistically significant differences (P < 0.05) among the values in the bars.

Intracellular and extracellular metabolic profiles in PANC-1 cells

We then used 1H-NMR to analyze the metabolite profiles of PANC-1 cells after different treatments to uncover the underlying mechanisms of the increased anticancer effect by the Combo. The intracellular and extracellular metabolites were analyzed separately, and the typical 1H-NMR spectra are shown in Supplementary Figures S1, S2, respectively. A total 37 metabolites were identified from 1H-NMR-based metabolic profile, including tryptophan-related metabolites (tryptophan, Trp; kynurenine, Kyn), amino acid metabolites (pyroglutamate, Pglu; glutamate, Glu; glutamine, Gln; N-acetylaspartate, NAA; tyrosine, Tyr; histidine, His; phenylalanine, Phe; methionine, Met; arginine, Arg; valine, Val; isoleucine, Ile; leucine, Leu; alanine, Ala; glutathione, GSH), energy-related metabolites (lactate, Lac; glucose, Glc; UDP-glucose, UDP-Glc; succinate, Suc; pyruvate, Pyr; creatine, Cre; nicotinamide adenine dinucleotide, NAD+; adenosine triphosphate, ATP; guanosine triphosphate, GTP), membrane-related phospholipid metabolism (ethanolamine, ETA; choline, Cho; phosphocholine, PC; glyceryl phosphocholine, GPC), ketone bodies (3-hydroxybutyrate, 3-HB), osmolytes (myo-inositol, Myo), and other metabolic products (butyrate, But; acetate, Ace; formate, For; methylamine, MA; dimethylamine, DMA; carnosine, Car). The spectral resonances of the metabolites were assigned according to the 600 MHz Reference Library of the Chenomx NMR suite 7.0 (Chenomx Inc., Edmonton, Canada), which was also confirmed by a public NMR database (Human Metabolome Database; www.hmdb.ca).

Alterations in metabolomic profiles

To explore the metabolomic changes in PANC-1 cells induced by Combo treatment in comparison with the individual treatments (α-MT or gemcitabine), metabolomic data extracted from the NMR spectra were then subjected to partial least-squares discriminate analysis (PLS-DA). As shown in Figure 2, the results indicated distinct metabolomic profiles in response to different treatments. The resulting PLS-DA score plots of both the intracellular metabolites (Figure 2A) and the extracellular metabolites (Figure 2B) clearly showed that the metabolomic profiles for the Control, gemcitabine, and α-MT groups were distinct and separable, and the Combo group showed a certain extent of overlap with α-MT and gemcitabine groups. Furthermore, gemcitabine and Combo groups were put together in a PLS-DA plot to asess the specific impact α-MT on the metabolomic profile of the gemcitabine group. A significant discrimination was observed between the control and the Combo groups (Figure 2C). The contribution of these variables was ranked by the VIP scores generated from the PLS-DA model. Twenty-one intracellular metabolites, namely ETA, Met, Ace, Lac, Trp, PC, Glu, Glc, Leu, Myo, Pyr, Cho, Iso, Pglu, butyrate, Val, 3-HB, His, Ala, Arg, and Phe, were responsible for the metabolomic difference between the gemcitabine and Combo groups. The results for the extracellular metabolites also showed that the plots of the gemcitabine and Combo groups were clearly separated (Figure 2E). The VIP scores suggested that 15 metabolites, namely OPC, ETA, GSH, Myo, Trp, GPC, Cho, Leu, Cre, Glc, Glu, Arg, Val, NAA, Ala, Ile and butyrate, contributed to the metabolomic differences between the two groups. These results suggest that addition of α-MT to gemcitabine elicited a considerable effect on the metabolomic phenotype of the treatment with gemcitabine alone, thus offering a glimpse of why the Combo is more efficacious as an anticancer regimen than α-MT or gemcitabine alone.

NMR-based PANC-1 cell metabolomic analysis.

Figure 2.
NMR-based PANC-1 cell metabolomic analysis.

PLS-DA score plot was used to identify the variations in the intracellular metabolomic profile (A) and in the extracellular metabolomic profile (B) among the four groups. PLS-DA score plots specifically for the gemcitabine (Gem) and Combo groups in terms of intracellular metabolomic profiles (C) and extracellular metabolomic profiles (E). VIP score for intracellular metabolites (D) and extracellular metabolites (F).

Figure 2.
NMR-based PANC-1 cell metabolomic analysis.

PLS-DA score plot was used to identify the variations in the intracellular metabolomic profile (A) and in the extracellular metabolomic profile (B) among the four groups. PLS-DA score plots specifically for the gemcitabine (Gem) and Combo groups in terms of intracellular metabolomic profiles (C) and extracellular metabolomic profiles (E). VIP score for intracellular metabolites (D) and extracellular metabolites (F).

Analysis of intracellular and extracellular metabolites

To further confirm the enhancing effect of α-MT on the antitumor efficacy of gemcitabine in pancreatic cancer cells, the levels of specific metabolites from1H-NMR data were quantified and compared. The detailed variations in intracellular and extracellular metabolites related to tryptophan metabolism are shown in Figure 3. The assigned spectra for tryptophan overlap with that of α-MT, and the amount of ‘tryptophan' actually represents the sum of tryptophan and α-MT. Therefore, after treatment, the extracellular amount of (tryptophan + α-MT) in the α-MT group was over 30-fold higher than that in control group, and the intracellular amount of (tryptophan + α-MT) in the α-MT group was ∼5-fold higher than that in control group (Figure 3A). While the intracellular kynurenine was the metabolic product only from tryptophan, and not from α-MT, intracellular kynurenine levels could be used to assess the effect of α-MT on the metabolic pattern of PANC-1 cells when used with gemcitabine as a Combo. As shown in Figure 3B, the intracellular level of kynurenine after treatment with α-MT alone or as a Combo with gemcitabine was significantly decreased. In contrast, treatment with gemcitabine alone had no effect on intracellular kynurenine levels. These data show that the observed changes in intracellular levels of kynurenine as completely due to α-MT. Kynurenine production is one of the most important metabolic pathways of tryptophan, which is responsible for ∼90% of tryptophan catabolism. In this pathway, indoleamine-2,3-dioxygenase (IDO), an intracellular heme-containing enzyme, is responsible for the conversion of tryptophan to kynurenine. The decreased intracellular kynurenine after α-MT treatment, either alone or in combination with gemcitabine, could possibly due to two distinct mechanisms. Firstly, α-MT is possibly an inhibitor of IDO, thus reducing the generation of kynurenine from tryptophan. Secondly, α-MT reduces the entry of extracellular tryptophan into cells via the blockade of SLC6A14. As kynurenine levels inside the cells remained unaltered with gemcitabine alone, the data suggest that gemcitabine has no effect on IDO. The influence of α-MT by itself or in combination with gemcitabine on SLC6A14-mediated tryptophan entry into cells is difficult to dissect because of the inability of NMR spectra to differentiate between tryptophan and α-MT. Interestingly, the intracellular levels of tryptophan plus α-MT were lower in the Combo group than in α-MT group (Figure 3C). This suggests that gemcitabine reduces the entry of either tryptophan or α-MT or both into cells, but the underlying mechanisms remained unknown.

Quantitative analysis of metabolites related to tryptophan metabolism in PANC-1 cells in response to various treatments (Control, no treatment; gemcitabine, Gem).

Figure 3.
Quantitative analysis of metabolites related to tryptophan metabolism in PANC-1 cells in response to various treatments (Control, no treatment; gemcitabine, Gem).

(A) Extracellular levels of tryptophan+α-MT; (B) Intracellular levels of kynurenine; (C) Intracellular levels of tryptophan+α-MT. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

Figure 3.
Quantitative analysis of metabolites related to tryptophan metabolism in PANC-1 cells in response to various treatments (Control, no treatment; gemcitabine, Gem).

(A) Extracellular levels of tryptophan+α-MT; (B) Intracellular levels of kynurenine; (C) Intracellular levels of tryptophan+α-MT. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

α-MT is a known blocker of ATB0,+/SLC6A14; therefore, α-MT is expected to block the entry of amino acids into cells via this transporter. As a result, it is logical to expect significant increases in the extracellular levels of specific amino acids that serve as substrates for the transporter. In addition, it is also expected to elicit significant changes in intracellular levels of specific amino acids and other related metabolites. To evaluate the validity of these predictions, we analyzed the extracellular and intracellular amino acids and certain selective related metabolites (Figure 4; closed bars represent extracellular data whereas open bars represent intracellular data). α-MT, when used alone, increased the levels of several amino acids in the extracellular medium. The observed increases in neutral amino acids and cationic amino acids were expected based on the known substrate specificity of SLC6A14, which is blocked by α-MT. What is interesting is the increase in the extracellular levels of glutamate, pyroglutamate, and N-acetylaspartate. The changes in these amino acids could not be explained solely based on blockade of SLC6A14. Changes in intracellular metabolic pathways induced by α-MT, either directly or indirectly, ought to be responsible for these changes. As expected, the levels of most of the amino acids inside the cells were lower in the α-MT group. Interestingly, the intracellular levels of glutathione were also lower in the α-MT group. This is important for the antitumor effect of this drug because glutathione is a potent antioxidant and therefore a decrease in this metabolite might decrease the survivability of cancer cells. Quite unexpectedly, treatment with gemcitabine alone also increased the extracellular levels of most amino acids almost in a manner similar to that of the α-MT group. It is possible that gemcitabine influences the expression of selective amino acid transporters in cancer cells. What is significant is that the synergism between α-MT and gemcitabine in the Combo group is clearly evident from the intracellular levels of most amino acids.

Quantitative analysis of amino acids and related metabolites in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Figure 4.
Quantitative analysis of amino acids and related metabolites in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Histogram with closed bars represents extracellular metabolites: (A) isoleucine, (B) leucine, (C) valine, (G) tyrosine, (H) arginine, (I) phenylalanine, (J) glutamate, (K) glutamine, (L) histidine, and (M) pyroglutamate; Histogram open bars represents intracellular metabolites: (D) isoleucine, (E) leucine, (F) valine, (N) tyrosine, (O) arginine, (P) phenylalanine, (Q) glutamate, (R) glutamine, (S) glutathione, (T) N-acetylaspartate. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

Figure 4.
Quantitative analysis of amino acids and related metabolites in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Histogram with closed bars represents extracellular metabolites: (A) isoleucine, (B) leucine, (C) valine, (G) tyrosine, (H) arginine, (I) phenylalanine, (J) glutamate, (K) glutamine, (L) histidine, and (M) pyroglutamate; Histogram open bars represents intracellular metabolites: (D) isoleucine, (E) leucine, (F) valine, (N) tyrosine, (O) arginine, (P) phenylalanine, (Q) glutamate, (R) glutamine, (S) glutathione, (T) N-acetylaspartate. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

With regard to energy-related metabolic pathways (Figure 5), α-MT and gemcitabine decreased the intracellular levels of glucose, lactate, alanine, UDP-glucose, fumarate, creatine, NAD+, ATP, and GTP. The extracellular levels of glucose, alanine and pyruvate increased in the α-MT and gemcitabine groups. There was synergism between α-MT and gemcitabine as evident in the Combo group in terms of extracellular levels of glucose, alanine and pyruvate. The same was true for the intracellular levels of most metabolites examined.

Quantitative analysis of metabolites related to energy metabolism in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Figure 5.
Quantitative analysis of metabolites related to energy metabolism in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Histogram with closed bars represents extracellular metabolites: (A) glucose, (B) lactate, (C) alanine, and (D) pyruvate; Histogram with open bars represents intracellular metabolites: (E) glucose, (F) lactate, (G) alanine, (H) UDP-glucose, (I) fumarate, (J) creatine, (K) NAD+, (L) ATP, and (M) GTP. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

Figure 5.
Quantitative analysis of metabolites related to energy metabolism in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Histogram with closed bars represents extracellular metabolites: (A) glucose, (B) lactate, (C) alanine, and (D) pyruvate; Histogram with open bars represents intracellular metabolites: (E) glucose, (F) lactate, (G) alanine, (H) UDP-glucose, (I) fumarate, (J) creatine, (K) NAD+, (L) ATP, and (M) GTP. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

In the metabolites related to membrane-associated phospholipid metabolism (Figure 6), α-MT and gemcitabine decreased the intracellular concentrations of ethanolamine, GPC, PC, and choline, whereas these drugs increased the extracellular concentrations of ethanolamine, GPC, PC, and choline significantly compared with the control group. The Combo treatment increased extracellular levels of PC and decreased intracellular levels of ethanolamine, GPC, and PC compared with gemcitabine alone or α-MT alone.

Quantitative analysis of metabolites related to membrane-associated phospholipids in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Figure 6.
Quantitative analysis of metabolites related to membrane-associated phospholipids in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Histogram with closed bars represents extracellular metabolites: (A) ethanolamine, (B) GPC, (C) phosphocholine, and (D) choline; Histogram with open bars represents intracellular metabolites: (E) ethanolamine, (F) GPC, (G) phosphocholine, and (H) choline. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

Figure 6.
Quantitative analysis of metabolites related to membrane-associated phospholipids in PANC-1 cells after various treatments (Control, no treatment; gemcitabine, Gem).

Histogram with closed bars represents extracellular metabolites: (A) ethanolamine, (B) GPC, (C) phosphocholine, and (D) choline; Histogram with open bars represents intracellular metabolites: (E) ethanolamine, (F) GPC, (G) phosphocholine, and (H) choline. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

As for other metabolites, compared with the control group, individual α-MT and gemcitabine treatment groups increased the extracellular levels of 3-hydroxybutyrate (Figure 7A), formate (Figure 7B), acetate (Figure 7E), and butyrate (Figure 7F), but decreased the intracellular levels of myo-inositol (Figure 7G), methylamine (Figure 7H), acetate (Figure 7K), and butyrate (Figure 7L); the Combo group displayed a synergistic effect. The gemcitabine group and the Combo group showed a significant effect on the intracellular levels of carnosine (Figure 7J) and extracellular levels of methanol (Figure 7D), but the α-MT group did not show any effect on these metabolites, suggesting the sole participation of gemcitabine in these events. With regard to intracellular dimethylamine (Figure 7I), only the Combo group showed a significant decrease. There were no changes in the extracellular levels of ethanol in any of the treatment groups (Figure 7C).

Quantitative analysis of other metabolites in PANC-1 cells after various treatments (Con, no treatment; gemcitabine, Gem).

Figure 7.
Quantitative analysis of other metabolites in PANC-1 cells after various treatments (Con, no treatment; gemcitabine, Gem).

Closed bars represents extracellular metabolites: (A) 3-hydroxybutyrate, (B) formate, (C) ethanol, (D) methanol, (E) acetate, and (F) butyrate. Open bars represents intracellular metabolites: (G) myo-inositol, (H) methylamine, (I) dimethylamine, (J) carnosine, (K) acetate, and (L) butyrate. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

Figure 7.
Quantitative analysis of other metabolites in PANC-1 cells after various treatments (Con, no treatment; gemcitabine, Gem).

Closed bars represents extracellular metabolites: (A) 3-hydroxybutyrate, (B) formate, (C) ethanol, (D) methanol, (E) acetate, and (F) butyrate. Open bars represents intracellular metabolites: (G) myo-inositol, (H) methylamine, (I) dimethylamine, (J) carnosine, (K) acetate, and (L) butyrate. The different letters above the bars indicate statistically significant differences (P < 0.05) among the values in the bars.

Taken collectively, the data provide evidence for a significant influence of the combination treatment of α-MT plus gemcitabine on multiple metabolic pathways deduced from the observed alterations in the intracellular levels of various metabolites (Figure 8). Compared with the other groups, the Combo treatment significantly decreased the uptake of most amino acids and nutrients; many related intracellular metabolites were also affected. In addition, specific intracellular metabolic pathways were inhibited, such as IDO-mediated conversion of tryptophan to kynurenine. The level of lactate, which plays a critical role in cancer progression and metastasis, was reduced in the Combo group, indicating that glycolysis as well as glutaminolysis, two of the pathways that lead to the generation of lactate in tumor cells, are suppressed. Other energy-rich metabolites, including fumarate, creatine, NAD+, ATP, and GTP, were also decreased in the Combo group, further providing evidence of decreased metabolic activity and cell viability.

Summary of the changes in prominent metabolites in PANC-1 cells in response to the combined treatment of α-MT and gemcitabine.

Figure 8.
Summary of the changes in prominent metabolites in PANC-1 cells in response to the combined treatment of α-MT and gemcitabine.

The changes in both intracellular and extracellular compartments are indicated. The pathways were deduced based on the KEGG database and the small molecule pathway database (SMPDB). The arrows denote significant differences (increase or decrease) in metabolite levels in PANC-1 cells exposed to the Combo treatment compared with treatment with gemcitabine alone.

Figure 8.
Summary of the changes in prominent metabolites in PANC-1 cells in response to the combined treatment of α-MT and gemcitabine.

The changes in both intracellular and extracellular compartments are indicated. The pathways were deduced based on the KEGG database and the small molecule pathway database (SMPDB). The arrows denote significant differences (increase or decrease) in metabolite levels in PANC-1 cells exposed to the Combo treatment compared with treatment with gemcitabine alone.

Discussion

Gemcitabine is widely used in the treatment of pancreatic cancer. Within the cancer cells, gemcitabine gets converted into difluoro-dCTP, which inhibits DNA synthesis, thereby inhibiting cell proliferation. Even though it is the first-line therapy for pancreatic cancer, the development of resistance in patients treated with the drug has become a significant clinical problem. Therefore, additional chemotherapeutics are commonly combined with gemcitabine to increase the therapeutic efficacy and also to overcome the resistance to gemcitabine. Nonetheless, none of these drug regimens represents a cure for pancreatic cancer; all they do at the best is to increase the lifespan of the patients by a few years. Therefore, we are in desperate need for new, hitherto unexplored but logically based, molecular targets to make a significant difference in the way we manage this disease. Nutrient transporters offer a new avenue in this regard [36,37]. Cancer cells up-regulate the glucose transporter GLUT1 (SLC2A1) [36], but this transporter is not suitable as a drug target because its function is obligatory for several normal tissues, especially the brain. This is evident from the serious clinical consequences that result from loss-of-function mutations in the transporter. In contrast, amino acid transporters offer an attractive alternative. There are more than three dozen transporters for amino acids; these transporters are expressed differentially in different tissues and also have overlapping substrate selectivity. Among these, SLC6A14 is unique. This transporter is not essential for normal tissues, and Slc6a14-null mice are normal and fertile with no readily observable deleterious consequences [38]. As this transporter is up-regulated markedly in pancreatic cancer and blockade of its transport function decreases the growth and proliferation of pancreatic cancer cells in vitro and in vivo, this transporter offers a logical drug target for pancreatic cancer treatment. We have already identified α-MT as a blocker of this transporter [19] and have also shown that it is able to reduce pancreatic cancer growth [18]. Therefore, we initiated the present study to evaluate the potential of this compound as a candidate for combination therapy with gemcitabine. As the molecular targets for the two compounds are totally different, we thought that the two drugs could elicit a synergistic effect in the treatment of pancreatic cancer. The results of the present study provide evidence for the validity of this approach and offer a proof-of-principle for the use of the two drugs in combination. Addition of α-MT to the gemcitabine regimen potentiates the efficacy of the latter in various parameters relevant to cancer therapy, namely cell proliferation, migration, and cell death. We then explored the molecular mechanisms underlying the synergism between the two drugs by using the metabolomic approach. This forms the primary focus of the current study. We used NMR-based metabolomics in a human pancreatic cancer cell line to understand what α-MT and gemcitabine do individually and in combination to elicit their anticancer effects. The results of these studies indicate that the enhanced antitumor efficacy of the combination of the two drugs is based on altrations in metabolic pathways invoving cellular energy, amino acids, phospholipids, ketone bodies, and short-chain fatty acids.

The most important metabolic pathway that could be directly relevant to the therapeutic efficacy of the combination regimen is related to the conversion of tryptophan into kynurenine. There is a clearcut distinction between α-MT and gemcitabine in the involvement of this pathway. The conversion of tryptophan to kynurenine is reduced only by α-MT; gemcitabine has no or little effect on this pathway. Indoleamine-2,3-dioxygenase (IDO) is responsible for this conversion. The importance of this pathway in cancer therapy lies in its ability to reverse the strategies of immune evasion orchestrated by the tumor cells [39]. IDO1 is up-regulated in antigen-presenting cells in the tumor microenvironment, which enhances the metabolism of tryptophan into kynurenine. When these cells present the tumor-specific antigens to T cells to elicit an antitumor immune response, tryptophan gets depleted in the immediate vicinity (the so-called immune cell synapse) because of the up-regulated IDO1 in antigen-presenting cells, which reduces the availability of tryptophan to T cells and consequently suppress their expansion and hence their antitumor efficacy [39]. This suggests that inhibition of IDO1 would reverse this phenomenon and potentiate the antitumor efficacy of the immune system. Indeed, several small molecule inhibitors of IDO are in various stages of development as potential drugs for cancer therapy [40]. One such small molecule is 1-methyltryptophan. Given the structural similarity between 1-methyltryptophan and α-MT, it is likely that α-MT also functions as an inhibitor of IDO1, thus explaining why kynurenine levels were lower in pancreatic cancer cells in response to α-MT exposure. It is known that IDO1 is up-regulated not only in antigen-presenting cells in the tumor microenvironment but also in tumor cells themselves [39]. It is important to note that gemcitabine does not affect this pathway; only α-MT does. Therefore, gemcitabine functions solely in the chemotherapy angle. α-MT could add the immunotherapy angle to synergize with gemcitabine in pancreatic cancer treatment.

The present project was initiated based on the function of α-MT as a blocker of the amino acid transporter SLC6A14. Therefore, the potential of this drug to prevent amino acid entry into tumor cells should not be ignored. Induction of amino acid starvation in tumor cells is a logical way to reduce their proliferation and growth. SLC6A14 transports all known essential amino acids; it also transports glutamine, a unique amino acid involved in various metabolic pathways that are obligatory for tumor growth. Glutamine is a nitrogen source for the synthesis of pyrimidines and purines, which are indispensable for tumor cell growth [41,42]. Moreover, glutaminolysis, one of the tumor-specific metabolic pathways, involves the entry of glutamine into the TCA cycle [41,42]. Our studies have shown that the intracellular concentration of glutamine in the Combo was significantly lower than that in the gemcitabine group or the α-MT group. This is mirrored by the higher levels of glutamine in the extracellular medium in the Combo group than in the gemcitabine or α-MT group. α-MT blocks the glutamine transporter SLC6A14, and the decreased entry of glutamine in the Combo group would decrease DNA/RNA synthesis, reduce glutaminolysis, and suppress mTOR signaling, thus offering a broad mechanistic approach for effective cancer therapy.

Branched-chain amino acids (valine, isoleucine and leucine) are essential amino acids. As with glutamine, the levels of these three amino acids were also reduced inside the cancer cells in response to the Combo treatment. In addition to the specific metabolic pathways involving these amino acids, leucine is the most potent activator of mTOR. Liu et al. [43] have shown that leucine supplementation enhances tumor growth in a murine model of pancreatic cancer. The decreased levels of leucine in cancer cells exposed to the Combo treatment suggest suppressed mTOR signaling, another mechanism for the antitumor efficacy of the Combo regimen.

The intracellular level of glutathione in the Combo group was significantly lower than that in the gemcitabine group or in the α-MT group. Glutathione is a potent intracellular antioxidant [44]. Glutathione levels are elevated in pancreatic cancer, and glutathione depletion induces apoptosis in pancreatic cancer cells [45]. As the intracellular levels of glutathione are reduced in pancreatic cancer cells in response to the Combo treatment, this could also contribute to the superior antitumor efficacy of the Combo. SLC7A11 is a key determinant of intracellular levels of glutathione because of its ability to mediate the influx of cystine into cells in exchange for glutamate. The intracellular levels of glutamate in cancer cells are lower in the Combo group. Interestingly, glutamate is not a substrate for SLC6A14 but glutamine is. Blockade of SLC6A14 by α-MT decreases cellular levels of glutamate indirectly by reducing glutamine levels, thereby suppressing the function of SLC7A11.

Pancreatic cells utilize metabolic reprogramming to meet their energy demand for survival. The resulting malignant metabolic phenotype (Warburg Effect) [36,37] is characterized by enhanced glycolysis with conversion of glucose to lactate. The metabolic reprogramming from oxidative phosphorylation to glycolysis satisfies the energy requirements of cancer cells [46]. Dhar et al. [47] have employed the NMR-based metabolomics approach to reveal that altered glucose uptake, lactate export, and energy state were the key components in pancreatic cancer. In our study, compared with the control group, both α-MT and gemcitabine increased extracellular concentrations of glucose and pyruvate, and decreased intracellular concentrations of UDP-glucose, glucose, and ATP, indicating the suppressed energy-related nutrient uptake and decreased intracellular energy transformation. The intracellular level of lactate was also decreased, which might be due to the decreased uptake of glucose and suppressed cell viability. The extracellular level of lactate decreased in the α-MT group and in the Combo group, but not in the gemcitabine group. These results collectively suggest that the Combo treatment could compromise glycolysis by decreasing glucose uptake. Other energy-related metabolites, including fumarate, creatine, NAD+, GTP, and alanine, were also analyzed in this study. The intracellular levels of these metabolites after the Combo treatment were lower than in the monotherapy groups. In summary, our findings indicate that the Combo therapy could also inhibit ‘metabolic reprogramming' and energy metabolism in tumor cells, which was not observed in gemcitabine monotherapy.

The synthesis of cell membranes involves the metabolism of ethanolamine, glyceryl phosphocholine (GPC), phosphocholine (PC), and choline [48]. In our study, the concentrations of ethanolamine, GPC, PC, and choline were increased outside of the cells but decreased inside the cells in both α-MT and gemcitabine groups. The Combo treatment showed an addidive effect on these parameters. Myo-inositol serves as an osmolyte in cancer cells. Myo-inositol could regulate cancer cell proliferation and migration through PI3K/AKT signaling [49]. Our studies show that α-MT and gemcitabine decrease the intracellular content of myo-inositol, while the Combo therapy shows the additive effect. A decrease in intracellular levels of myo-inositol might impact on cellular osmolality, which could cause cell shrinking and death.

3-Hydroxybutyrate, the principal ketone body, serves as an alternative energy source in tumor cells. In the present study, 3-hydroxybutyrate was only detected outside the cell, and the level of 3-hydroxybutyrate in the Combo group was higher than in the monotherapy groups. Studies have shown that short-chain fatty acids could reduce proliferation and induce apoptosis in cancer cells [50]. The intracellular concentrations of butyrate and acetate in the Combo group were significantly lower than in the monotherapy groups. Formate supplies one-carbon group in de novo synthesis of nucleotides. Methylamine and dimethylamine are involved in formate metabolism. Our studies show that the Combo therapy decreased the extracellular level of formate and increased the intracellular level of methylamine and dimethylamine compared with the two monotherapies.

In conclusion, α-MT and gemcitabine co-treatment were found to significantly enhance anticancer activity by inhibiting cell migration, proliferation and apoptotic cell death. α-MT sensitizes the pancreatic cancer cells to gemcitabine by targeting several biochemical pathways associated with energy production, amino acids, membrane phospholipits, short-chain fatty acids and formate. These studies provide a rationale basis for future evaluation of the therapeutic efficacy of a combination therapy with gemcitabine and α-MT for pancreatic cancer.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was financially supported by the National Natural Science Foundation of China [81803443, 81903551], the Natural Science Foundation of Zhejiang Province (LQ19H300001), the Wenzhou Science and Technology Bureau [ZY2019007, Y20180180, Y20180208, Y20190177], Zhejiang Pharmaceutical Association [2019ZYY39, 2018ZYY44, 2018ZYY15], and the Special Project for Significant New Drug Research and Development in the Major National Science and Technology Projects of China (2020ZX09201002).

Author Contributions

A.C., H.Z., and Z.C. contributed equally to this work. A.C., H.Z., and Z.C. performed most of the experiments and data analysis. X.L. and C.L. contributed to the cell culture work. Q.Y., V.G., R.C., and L.K. conceived the study. A.C. and L.K. wrote the manuscript, and Y.B. and V.G. edited and finalized the manuscript.

Abbreviations

     
  • ATP

    adenosine triphosphate

  •  
  • CI

    combination index

  •  
  • GPC

    glyceryl phosphocholine

  •  
  • GTP

    guanosine triphosphate

  •  
  • IDO

    indoleamine-2,3-dioxygenase

  •  
  • MT

    Methyltryptophan

  •  
  • MTT

    Methylthiazolyldiphenyl-tetrazolium bromide

  •  
  • NAA

    N-acetylaspartate

  •  
  • NAD

    nicotinamide adenine dinucleotide

  •  
  • NMR

    nuclear magnetic resonance

  •  
  • PC

    phosphocholine

  •  
  • PLS-DA

    Partial least squares-discriminate analysis

  •  
  • VIP

    variable importance in the projection

References

References
1
Cordelier
,
P.
(
2018
)
Keep quiet and stay in line! Smart polymers to keep an eye on pancreatic tumors
.
Mol. Ther.
26
,
940
941
2
Kamisawa
,
T.
,
Wood
,
L.D.
,
Itoi
,
T.
and
Takaori
,
K.
(
2016
)
Pancreatic cancer
.
Lancet
388
,
73
85
3
Rawla
,
P.
,
Sunkara
,
T.
and
Gaduputi
,
V.
(
2019
)
Epidemiology of pancreatic cancer: global trends, etiology and risk factors
.
World J. Oncol.
10
,
10
27
4
Dumont
,
R.
,
Puleo
,
F.
,
Collignon
,
J.
,
Meurisse
,
N.
,
Chavez
,
M.
,
Seidel
,
L.
et al (
2017
)
A single center experience in resectable pancreatic ductal adenocarcinoma: the limitations of the surgery-first approach. Critical review of the literature and proposals for practice update
.
Acta Gastroenterol. Belg.
80
,
451
461
PMID:
[PubMed]
5
Ying
,
J.E.
,
Zhu
,
L.M.
and
Liu
,
B.X.
(
2012
)
Developments in metastatic pancreatic cancer: is gemcitabine still the standard?
World J. Gastroenterol.
18
,
736
745
6
Colucci
,
G.
,
Labianca
,
R.
,
Di Costanzo
,
F.
,
Gebbia
,
V.
,
Carteni
,
G.
,
Massidda
,
B.
et al (
2010
)
Randomized phase III trial of gemcitabine plus cisplatin compared with single-agent gemcitabine as first-line treatment of patients with advanced pancreatic cancer: the GIP-1 study
.
J. Clin. Oncol.
28
,
1645
1651
7
Ward
,
P.S.
and
Thompson
,
C.B.
(
2012
)
Metabolic reprogramming: a cancer hallmark even Warburg did not anticipate
.
Cancer Cell
21
,
297
308
8
Bhutia
,
Y.D.
,
Babu
,
E.
,
Ramachandran
,
S.
and
Ganapathy
,
V.
(
2015
)
Amino acid transporters in cancer and their relevance to ‘glutamine addiction’: novel targets for the design of a new class of anticancer drugs
.
Cancer Res.
75
,
782
788
9
Bhutia
,
Y.D.
and
Ganapathy
,
V.
(
2016
)
Glutamine transporters in mammalian cells and their functions in physiology and cancer
.
Biochim. Biophys. Acta
1863
,
2531
2539
10
Sikder
,
M.O.F.
,
Yang
,
S.
,
Ganapathy
,
V.
and
Bhutia
,
Y.D.
(
2017
)
The Na+/Cl-coupled, broad-specific, amino acid transporter SLC6A14 (ATB0,+): emerging roles in multiple diseases and therapeutic potential for treatment and diagnosis
.
AAPS J.
20
,
12
11
Liu
,
Y.
,
Zhao
,
T.
,
Li
,
Z.
,
Wang
,
L.
,
Yuan
,
S.
and
Sun
,
L.
(
2018
)
The role of ASCT2 in cancer: a review
.
Eur. J. Pharmacol.
837
,
81
87
12
Kandasamy
,
P.
,
Gyimesi
,
G.
,
Kanai
,
Y.
and
Hediger
,
M.A.
(
2018
)
Amino acid transporters revisited: new views in health and disease
.
Trends Biochem. Sci.
43
,
752
789
13
Koppula
,
P.
,
Zhang
,
Y.
,
Zhuang
,
L.
and
Gan
,
B.
(
2018
)
Amino acid transporter SLC7A11/xCT at the crossroads of regulating redox homeostasis and nutrient dependency of cancer
.
Cancer Commun. (Lond)
38
,
12
14
Bhutia
,
Y.D.
,
Babu
,
E.
,
Prasad
,
P.D.
and
Ganapathy
,
V.
(
2014
)
The amino acid transporter SLC6A14 in cancer and its potential use in chemotherapy
.
Asian J. Pharm. Sci.
9
,
293
303
15
Gupta
,
N.
,
Prasad
,
P.D.
,
Ghamande
,
S.
,
Moore-Martin
,
P.
,
Herdman
,
A.V.
,
Martindale
,
R.G.
et al (
2006
)
Up-regulation of the amino acid transporter ATB0,+ (SLC6A14) in carcinoma of the cervix
.
Gynecol. Oncol.
100
,
8
13
16
Gupta
,
N.
,
Miyauchi
,
S.
,
Martindale
,
R.G.
,
Herdman
,
A.V.
,
Podolsky
,
R.
,
Miyake
,
K.
et al (
2005
)
Upregulation of the amino acid transporter ATB0,+ (SLC6A14) in colorectal cancer and metastasis in humans
.
Biochim. Biophys. Acta
1741
,
215
223
17
Karunakaran
,
S.
,
Ramachandran
,
S.
,
Coothankandaswamy
,
V.
,
Elangovan
,
S.
,
Babu
,
E.
,
Periyasamy-Thandavan
,
S.
et al (
2011
)
SLC6A14 (ATB0,+) protein, a highly concentrative and broad specific amino acid transporter, is a novel and effective drug target for treatment of estrogen receptor-positive breast cancer
.
J. Biol. Chem.
286
,
31830
8
18
Coothankandaswamy
,
V.
,
Cao
,
S.
,
Xu
,
Y.
,
Prasad
,
P.D.
,
Singh
,
P.K.
,
Reynolds
,
C.P.
et al (
2016
)
Amino acid transporter SLC6A14 is a novel and effective drug target for pancreatic cancer
.
Br. J. Pharmacol.
173
,
3292
3306
19
Karunakaran
,
S.
,
Umapathy
,
N.S.
,
Thangaraju
,
M.
,
Hatanaka
,
T.
,
Itagaki
,
S.
,
Munn
,
D.H.
et al (
2008
)
Interaction of tryptophan derivatives with SLC6A14 (ATB0,+) reveals the potential of the transporter as a drug target for cancer chemotherapy
.
Biochem. J.
414
,
343
355
20
Lin
,
C.Y.
,
Wu
,
H.
,
Tjeerdema
,
R.S.
and
Viant
,
M.R.
(
2007
)
Evaluation of metabolite extraction strategies from tissue samples using NMR metabolomics
.
Metabolomics
3
,
55
67
21
Bradley
,
S.A.
,
Ouyang
,
A.
,
Purdie
,
J.
,
Smitka
,
T.A.
,
Wang
,
T.
and
Kaerner
,
A.
(
2010
)
Fermentanomics: monitoring mammalian cell cultures with NMR spectroscopy
.
J Am. Chem. Soc.
132
,
9531
9533
22
Zhang
,
A.
,
Sun
,
H.
,
Xu
,
H.
,
Qiu
,
S.
and
Wang
,
X.
(
2013
)
Cell metabolomics
.
OMICS
17
,
495
501
23
D'Alessandro
,
A.
and
Zolla
,
L.
(
2013
)
Proteomics and metabolomics in cancer drug development
.
Expert Rev. Proteomics
10
,
473
488
24
Halama
,
A.
(
2014
)
Metabolomics in cell culture—a strategy to study crucial metabolic pathways in cancer development and the response to treatment
.
Arch. Biochem. Biophys.
564
,
100
109
25
D'Alessandro
,
G.
,
Quaglio
,
D.
,
Monaco
,
L.
,
Lauro
,
C.
,
Ghirga
,
F.
,
Ingallina
,
C.
et al (
2019
)
1H-NMR metabolomics reveals the Glabrescione B exacerbation of glycolytic metabolism beside the cell growth inhibitory effect in glioma
.
Cell Commun. Signal.
17
,
108
26
Gao
,
L.
,
Wang
,
K.X.
,
Zhang
,
N.N.
,
Li
,
J.Q.
,
Qin
,
X.M.
and
Wang
,
X.L.
(
2018
)
1H nuclear magnetic resonance based metabolomics approach reveals the metabolic mechanism of (-)-5-hydroxy-equol against hepatocellular carcinoma cells in vitro
.
J. Proteome Res.
17
,
1833
1843
27
Lv
,
Y.
,
Tian
,
N.
,
Wang
,
J.
,
Yang
,
M.
and
Kong
,
L.
(
2018
)
Metabolic switching in the hypoglycemic and antitumor effects of metformin on high glucose induced HepG2 cells
.
J. Pharm. Biomed. Anal.
156
,
153
162
28
Chou
,
T.C.
(
2010
)
Drug combination studies and their synergy quantification using the Chou-Talalay method
.
Cancer Res.
70
,
440
446
29
Yao
,
Q.
,
Gutierrez
,
D.C.
,
Hoang
,
N.H.
,
Kim
,
D.
,
Wang
,
R.
,
Hobbs
,
C.
et al (
2017
)
Efficient codelivery of paclitaxel and curcumin by novel bottlebrush copolymer-based micelles
.
Mol. Pharm.
14
,
2378
2389
30
Wang
,
D.
,
Zhao
,
L.
,
Zheng
,
H.
,
Dong
,
M.
,
Pan
,
L.
,
Zhang
,
X.
et al (
2018
)
Time-dependent lactate production and amino acid utilization in cultured astrocytes under high glucose exposure
.
Mol. Neurobiol.
55
,
1112
1122
31
Teng
,
Q.
,
Huang
,
W.
,
Collette
,
T.W.
,
Ekman
,
D.R.
and
Tan
,
C.
(
2008
)
A direct cell quenching method for cell-culture based metabolomics
.
Metabolomics
5
,
199
208
32
Zhao
,
L.
,
Dong
,
M.
,
Ren
,
M.
,
Li
,
C.
,
Zheng
,
H.
and
Gao
,
H.
(
2018
)
Metabolomic analysis identifies lactate as an important pathogenic factor in diabetes-associated cognitive decline rats
.
Mol. Cell Proteomics
17
,
2335
2346
33
Zhao
,
L.
,
Dong
,
M.
,
Wang
,
D.
,
Ren
,
M.
,
Zheng
,
Y.
,
Zheng
,
H.
et al (
2018
)
Characteristic metabolic alterations identified in primary neurons under high glucose exposure
.
Front. Cell Neurosci.
12
,
207
34
Savorani
,
F.
,
Tomasi
,
G.
and
Engelsen
,
S.B.
(
2010
)
Icoshift: a versatile tool for the rapid alignment of 1D NMR spectra
.
J. Magn. Reson.
202
,
190
202
35
Dromms
,
R.A.
and
Styczynski
,
M.P.
(
2012
)
Systematic applications of metabolomics in metabolic engineering
.
Metabolites
2
,
1090
1122
36
Ganapathy
,
V.
,
Thangaraju
,
M.
and
Prasad
,
P.D.
(
2009
)
Nutrient transporters in cancer: relevance to Warburg hypothesis and beyond
.
Pharmacol. Ther.
121
,
29
40
37
Brown
,
T.P.
and
Ganapathy
,
V.
(
2020
)
Lactate/GPR81 signaling and proton motive force in cancer: role in angiogenesis, immune escape, nutrition, and Warburg phenomenon
.
Pharmacol. Ther.
206
,
107451
38
Babu
,
E.
,
Bhutia
,
Y.D.
,
Ramachandran
,
S.
,
Gnanaprakasam
,
J.P.
,
Prasad
,
P.D.
,
Thangaraju
,
M.
et al (
2015
)
Deletion of the amino acid transporter Slc6a14 suppresses tumour growth in spontaneous mouse models of breast cancer
.
Biochem. J.
469
,
17
23
39
Munn
,
D.H.
and
Mellor
,
A.L.
(
2016
)
IDO in the tumor microenvironment: inflammation, counter-regulation, and tolerance
.
Trends Immunol.
37
,
193
207
40
Ricciuti
,
B.
,
Leonardi
,
G.C.
,
Puccetti
,
P.
,
Fallarino
,
F.
,
Bianconi
,
V.
,
Sahebkar
,
A.
et al (
2019
)
Targeting indoleamine-2,3-dioxygenase in cancer: scientific rationale and clinical evidence
.
Pharmacol. Ther.
196
,
105
116
41
Daye
,
D.
and
Wellen
,
K.E.
(
2012
)
Metabolomic reprogramming in cancer: unrevaling the role of glutamine in tumorigenesis
.
Semin. Cell Dev. Biol.
23
,
362
369
42
Son
,
J.
,
Lyssiotis
,
C.A.
,
Ying
,
H.
,
Wang
,
X.
,
Hua
,
S.
,
Ligorio
,
M.
et al (
2013
)
Glutamine supports pancreatic cancer growth through a KRAS-regulated metabolic pathway
.
Nature
496
,
101
105
43
Liu
,
K.A.
,
Lashinger
,
L.M.
,
Rasmussen
,
A.J.
and
Hursting
,
S.D.
(
2014
)
Leucine supplementation differentially enhances pancreatic cancer growth in lean and overweight mice
.
Cancer Metab.
2
,
6
44
Kou
,
L.
,
Jiang
,
X.
,
Huang
,
H.
,
Lin
,
X.
,
Zhang
,
Y.
,
Yao
,
Q.
et al (
2020
)
The role of transporters in cancer redox homeostasis and cross-talk with nanomedicines
.
Asian J. Pharm. Sci.
15
,
145
157
45
Schnelldorfer
,
T.
,
Gansauge
,
S.
,
Gansauge
,
F.
,
Schlosser
,
S.
,
Beger
,
H.G.
and
Nussler
,
A.K.
(
2000
)
Glutathione depletion causes cell growth inhibition and enhanced apoptosis in pancreatic cancer cells
.
Cancer
89
,
1440
1447
46
Tennant
,
D.A.
,
Duran
,
R.V.
and
Gottlieb
,
E.
(
2010
)
Targeting metabolic transformation for cancer therapy
.
Nat. Rev. Cancer
10
,
267
277
47
Dhar
,
D.
,
Raina
,
K.
,
Kant
,
R.
,
Wempe
,
M.F.
,
Serkova
,
N.J.
,
Agarwal
,
C.
et al (
2019
)
Bitter melon juice-intake modulates glucose metabolism and lactate efflux in tumors in its efficacy against pancreatic cancer
.
Carcinogenesis
40
,
1164
1176
48
Glunde
,
K.
,
Bhujwalla
,
Z.M.
and
Ronen
,
S.M.
(
2011
)
Choline metabolism in malignant transformation
.
Nat. Rev. Cancer
11
,
835
848
49
Vucenik
,
I.
and
Shamsuddin
,
A.M.
(
2003
)
Cancer inhibition by inositol hexaphosphate (IP6) and inositol: from laboratory to clinic
.
J. Nutr.
133
,
3778S
3784S
50
Bhutia
,
Y.D.
,
Babu
,
E.
,
Ramachandran
,
S.
,
Yang
,
S.
,
Thangaraju
,
M.
and
Ganapathy
,
V.
(
2016
)
SLC transporters as a novel class of tumour suppressors: identity, function and molecular mechanisms
.
Biochem. J.
473
,
1113
1124

Author notes

*

These authors contributed equally to this work.

Supplementary data