Fig 1.
A schematic presentation of simultaneous prevention of lactate and ROS/SOX generations by using small-molecule compounds to reduce the activity of cancer-specific 6-phosphofructo-1-kinase (PFK1).
Dysregulated metabolic processes in cancer lead to the formation of numerous glycolytic intermediates, including cytosolic NADH. As abundant NADH is formed at the level of glyceraldehyde-3-phosphate, the cellular redox system becomes unbalanced. Excess NADH must be reoxidized immediately to avoid damaging disruption through lactate and superoxide formation. While NADH is mainly reoxidized in the cytosol by reducing pyruvate to lactate, part of the NADH enters the mitochondrial compartment through the aspartate-malate shuttle. Mitochondrial NADH is oxidized by donating electrons to the electron transport chain (ETC) in oxidative phosphorylation (OXPHOS). If there is an excess of electrons, some can incompletely reduce the oxygen with a single electron and form superoxide. Small molecule cmpds of 6-phosphofructo-1-kinase (PFK1) reduce the activity of the enzyme’s highly active, cancer-specific form to the level of normal PFK1 enzymes. At the same time, the metabolism in the cancer cells is reduced, and the formation of abundant NADH is suppressed. Restoring the redox balance eliminates the need for NADH reoxidation, and prevents lactate and superoxide formation.
Fig 2.
Preliminary screening of selected compounds for suppressing lactate and superoxide formation in Jurkat cells.
First, all 33 commercially available compounds selected by screening were examined to determine whether they suppress the formation of lactate and mitochondrial superoxide (SOX) by Jurkat cells. After 36 hours of incubation, statistically significant differences were observed between untreated (vehicle) and treated cells exposed to 50 µM of each inhibitor. In the Jurkat cells, four cmpds were found to suppress lactate (P* < 0.005), compared to the vehicle. However, the lowest level of significant difference in lactate suppression between treated and untreated cells was observed by cmpd No. 9, (P** < 0.001). By measuring SOX suppression, cmpd No. 30 proved to be the most successful (P** < 0.001), while cmpds No. 3 and 9 showed slightly higher values (P* < 0.005), compared to the vehicle. All compounds that showed a more potent inhibitory effect on a tumorigenic cell are marked in red.
Table 1.
A list of small-molecule inhibitors reducing human PFK1 iso-enzyme activities.
Fig 3.
Half maximal inhibitory concentrations (IC50) of selected compounds.
In the graphs, the red horizontal lines present the intersections among the concentrations of modified sfPFK-M enzymes, while horizontal blue lines are from native nPFK-M enzymes. The vertically dashed line shows the half-maximal PFK-M activities of enzymes treated by different concentrations of selected compounds. Selected compounds were evaluated for their ability to reduce purified human PFK1 enzyme activities in vitro. Cmpds No. 3 and 9, which preferentially bind to the ATP-binding site of PFK-M, inhibited both the native (85 kDa) enzyme and the shorter (47 kDa) fragment. Still, more significant inhibition was observed in modified, cancer-specific shorter forms. Reaching IC50 relative activity at cmpd No. 3, 15 µM and cmpd No. 9, 17 µM). In contrast, more significant differences were observed between the inhibitions of native and modified PFK-L enzymes. At lower concentrations, almost no deactivation of the native (85 kDa) activity was observed. However, more effective inhibition of the shorter (70 kDa) fragment of modified PFK-L was observed with cmpd No. 29, 30 (IC50 8 µM) and cmpd No. 31, 32 (IC50 9 µM). The data represents three independent measurements and are given as mean ±(SD) mean (n = 3).
Fig 4.
Computational models of catalytic binding sites of two PFK1 iso-enzymes as a docking target for small-molecule cmpd.
(Above) shows a computer model of the amino acid residues of the PFK-P/PFK-M ATP binding sites with cmpd No. 9 as the ligand. (Below) shows a modified PFK-L model in which arginine was replaced by threonine and cmpd No. 30 as the ligand. A supercomputer used both models to virtual screen a ZINC Drugs NOW database.
Fig 5.
Dose-dependent reduction of lactate formation by Jurkat cells.
Dose-dependent effects of cmpds No. 9 and 30 on lactate suppression were observed by the tumorigenic Jurkat cells. 5A. Simultaneously, no significant cytostatic (cells counted after 72 hours of incubation). 5B, and no cytotoxic effects (percentage of dead cells) after 72 hours of incubation). 5C was observed under dose-dependent conditions. Although initially (up to 24 hours), no reduced inhibitory effect of both compounds can be observed, the steadily increasing lactate formation in later incubation phases indicated an instability of the compounds. The data represents three independent measurements and are given mean ± SD (n = 3).
Fig 6.
(a) Lactate suppression by sequential re-insertion of cmpd at low concentration in Jurkat cells.
Lactate excretion in tumorigenic Jurkat cells remained effectively reduced when the cmpd No. 9 and 30) were added periodically every 24 hours at low concentrations (10, 15, and 20 µM). At 72 hours of incubation, the values of statistically significant differences between the vehicle and the cells treated with 10 µM of No. 9 and No. 30 were P < 0.0005, respectively. Data represents three independent measurements expressed as mean ±SD (n = 3). (b) Superoxide (SOX) and (ROS) suppression by sequential re-insertion of cmpds at low concentrations in Jurkat cells. Suppressed ROS and SOX formation was observed in Jurkat cells when cmpds No. 9 or 30 were sequentially added to the medium at low concentrations (5, 10, and 15 µM). Although some suppression of SOX formation was observed at 5 µM of both camps, significant differences were observed between the vehicle and cells treated with both cmpds (10 and 15 µM). The dose-dependent effects of the different concentrations of cmpds No. 9 and 30 were also examined regarding the suppression of ROS formation. However, only SOX formation in Jurkat cells appeared more potently suppressed. However, at 72 hours of inoculation, the cmpds tested for suppression reactive oxygen species (ROS) proved less efficient in reaching the P values of < 0.1, compared to the vehicle. Under identical growth conditions, the statistically significant values of SOX detected by 10 µM concentrations were P < 0.005 for No. 9 and P < 0.005 for No. 30, compared to the vehicle. The data are representative of three independent measurements and are given as mean ±SD (n = 3).
Fig 7.
The influence of selected PFK1 compounds on glycolytic and respiratory parameters in Jurkat cells.
Respiratory fluxes, maximal respiration rates, glycolytic rates, and glycolytic capacities were measured in Jurkat cells in the presence and absence of selected cmpd. Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR), measured in the first 20 minutes of the experiment, were significantly higher in the untreated cells (vehicle) than in the treated cells. For cmpd No. 9, the statistical significances between untreated and treated cells were P** < 0.01 for respiration and P*** < 0.005 for glycolytic rate, while for cmpd No. 30, significances of P** < 0.01 for respiration and P*** < 0.005 for glycolytic rate, were determined. After 20 minutes, oligomycin A and FCCP were added, which made it possible to evaluate the maximal respiration and glycolytic capacities. In Jurkat cells treated with cmpd No. 9, the highest values of < P* 0,05 were obtained for maximal respiration and P*** value <0.005 for glycolytic capacity, compared to the vehicle while in the cells treated with cmpd No. 30, the values for maximal respiration were P** < 0.01 and for glycolytic capacity P*** < 0.005, compared to the vehicle. Data represents three independent measurements expressed as mean ±SD (n = 3).
Fig 8.
In the co-culture of tumorigenic Jurkat cells and activated T-cells, apoptosis is enhanced in the presence of PFK1 cmpd.
It is assumed that extracellular lactate in tumors triggers an acidosis that prevents the normal functioning of immune cells. Since PFK1 cmpds are thought to suppress lactate formation, acidosis may not be formed, so fully functional T cells could defeat cancer cells by apoptosis. Accordingly, increased late apoptosis was expected in a co-culture of activated T cells and tumorigenic Jurkat cells treated with PFK1 cmpds. In fact, after 18 hours of co-culture incubation with untreated cells (vehicle), only 18% (Q2-R2) of the cells were detected at the late apoptosis stage using a flow cytometer. However, in the cells treated with 10 µM cmpd No. 9 under identical environmental conditions, significantly higher proportions of cells were observed in the late apoptosis stage (Q2-R2). No. 9 (46.83%) or 10 µM of cmpd No. 30 (44.63%).
Fig 9.
Schematic presentation of lactate and superoxide functions in cancer development.
The schematic shows the involvement of lactate and superoxide in the development of solid tumors with specific cancer characteristics. While extracellularly elevated lactate levels trigger acidosis in the tumor matrix, which leads to immune response, contributes to chronic inflammation, stimulates angiogenesis, and promotes metastasis formation, superoxide is considered a precursor of hydroxyl radicals, the most potent endogenous mutagen. The involvement of lactate (yellow) and superoxide (blue) in carcinogenesis is illustrated in the scheme previously published in “Hallmarks of Cancer: The Next Generation” [1].