Targeting cancer lactate metabolism with synergistic combinations of synthetic catalysts and monocarboxylate transporter inhibitors

Synthetic anticancer catalysts offer potential for low-dose therapy and the targeting of biochemical pathways in novel ways. Chiral organo-osmium complexes, for example, can catalyse the asymmetric transfer hydrogenation of pyruvate, a key substrate for energy generation, in cells. However, small-molecule synthetic catalysts are readily poisoned and there is a need to optimise their activity before this occurs, or to avoid this occurring. We show that the activity of the synthetic organometallic redox catalyst [Os(p-cymene)(TsDPEN)] (1), which can reduce pyruvate to un-natural d-lactate in MCF7 breast cancer cells using formate as a hydride source, is significantly increased in combination with the monocarboxylate transporter (MCT) inhibitor AZD3965. AZD3965, a drug currently in clinical trials, also significantly lowers the intracellular level of glutathione and increases mitochondrial metabolism. These synergistic mechanisms of reductive stress induced by 1, blockade of lactate efflux, and oxidative stress induced by AZD3965 provide a strategy for low-dose combination therapy with novel mechanisms of action. Graphical abstract Supplementary Information The online version contains supplementary material available at 10.1007/s00775-023-01994-3.


Biological studies
Cell maintenance. All cell lines used in this study were maintained in pyruvate-free Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% foetal calf serum, 1% penicillin/streptomycin, and glutamine. All cells were grown as adherent monolayers in a humidified 5% CO2 environment (37°C) and passaged using trypsin/EDTA (0.25%) upon reaching 90% confluence.
Antiproliferative activity. Briefly, 5 × 10 3 cells were seeded per well (150 µL) in 96-well plates and incubated for 48 h. Separately, stock solutions (typically 100 µM) of test compounds were prepared in culture medium containing dimethylsulfoxide to aid solubility of the osmium catalysts (dimethylsulfoxide working concentrations did not exceed 0.5%) and cells were exposed to six concentrations of test compound (typically 0.1-100 µM, 200 µL) for 24 h. Stock solutions were analysed by ICP-OES to determine the exact concentration of test compound in solution. After 24 h exposure, cells were washed with PBS (200 µL) and incubated for a further 72 h in fresh culture medium (recovery time). Cell viability was determined using the SRB assay. Absorbance measurements were determined using a Thermo Fisher Multiskan FC microplate reader (absorbance at 490 nm). IC50 concentrations (viable treated cells compared to untreated controls) were determined as duplicates of triplicates in two independent sets of experiments and standard deviations were calculated. Significance testing was carried out using a two-tailed t-test assuming unequal sample variances. Antiproliferative activity experiments were also carried out in the presence of AZD3965 (0, 0.1, 1 or 10 µM) or L-BSO (5 µM) which was administered to cells independently, but within 5 min of administering the metal catalyst.
Cellular accumulation of Os. Briefly, 1 × 10 6 cells were seeded in P100 culture dishes (10 mL) and incubated for 24 h. Media was then removed, and cells were exposed to either: (i) culture medium only, (ii) 10 µM (1 × IC50) compound 1, (iii) 10 µM (1 × IC50) compound (R,R)-1 + 2 mM formate; in culture medium for 24 h. No recovery time was allowed. Cells were harvested using trypsin/EDTA, counted and cell pellets collected for subsequent analysis by ICP-MS. Cell pellets were digested overnight using 200 µL concentrated (72%) trace metal grade nitric acid (80°C) after which time solutions were diluted using doubly-distilled water supplemented with ascorbic acid (100 mg L -1 ) and thiourea (10 mM) to achieve a final working acid concentration of 3.6% v/v. 189 Os quantification was carried out using an Agilent 7900 Series ICP-MS in He-gas collision mode. Experiments were carried out in triplicate with triplicate instrumental replicates of each sample (triplicate of triplicates). Significance testing was carried out using a two-tailed t-test assuming unequal sample variances.

Modulation of antiproliferative activity (in-cell catalysis).
Antiproliferative activities were determined as described above with the following modifications: a fixed (equipotent) concentration of osmium 1 catalyst was used (5.5 µM, 0.5 × IC50), with concentration predetermined by ICP-OES prior to administration to cells. Sodium formate (0.5, 1.0 or 2.0 mM) was co-administered independently, but within 5 min of addition of catalyst 1. Cell viability experiments were also carried out using sodium acetate (0-2 mM) in place of sodium formate.
Catalytic reduction of intracellular pyruvate to D-lactate. The catalytic generation of Dlactate was determined as previously described. 17 Briefly, 30 × 10 6 MCF7 human breast cancer cells were seeded in T75 cell culture flasks and incubated for 24 h. After this time, cells were exposed to 11 µM (1 × IC50) of catalyst (R,R)-1 or (S,S)-1 in combination with 2 mM sodium formate and/or 1 µM AZD3965; in culture medium for 24 h without recovery time. All possible negative control experiments (osmium catalyst only, sodium formate only, AZD3965 only, sodium formate and AZD3965 only) were also established. Cells were harvested using trypsin/EDTA and cell pellets of equal cell count (40 × 10 6 cells) were prepared. The D-lactate assay kit (Cayman Chemical) was used to measure intracellular lactate according to the manufactures' instructions. Fluorescence readings were determined using a Promega GloMax Multi+ microplate reader (λex 530-540 nm, λem 585-595 nm). Experiments were carried out in triplicate and standard deviations were reported. Significance testing was carried out using a two-tailed t-test assuming unequal sample variances.
Glutathione quantitation. Briefly, 1 × 10 6 MCF7 human breast cancer cells were seeded in P100 culture dishes (10 mL) and incubated for 48 h. Medium was then removed, and cells were exposed to culture medium containing 11 µM (1 × IC50) compound (R,R)-1 either in the presence or absence of AZD3965 (1 µM) and/or sodium formate (2 mM), in culture medium for 24 h without recovery time. All possible negative control experiments (osmium catalyst only, sodium formate and AZD3965 without osmium catalyst) were also established. After this time, cells were washed with PBS (5 mL) and harvested using trypsin/EDTA to obtain cell pellets which were washed with ice cold PBS and lysed by high speed vortexing and centrifugation in combination with freeze/thaw cycling. Glutathione quantitation was achieved using the Glutathione Colorimetric Detection Kit (Invitrogen, purchased from Thermo Fisher Scientific) following the manufacturer's instructions. Experiments were carried out in triplicate and standard deviations are reported. Determined glutathione levels were normalised to cell density, which in turn were converted to molar concentrations by assuming a constant cell volume (1.760 ´ 10 -12 L). 39 Significance testing was carried out using a two-tailed t-test assuming unequal sample variances.
This experiment was also carried out with the following modification: cells were treated with 11 µM (1 × IC50) compound (R,R)-1 in the presence of 5 µM L-BSO for 24 h without recovery time. Cellular glutathione was quantified as described above.     Table S5. Antiproliferative activity modulation (normalized cell survival, %) of MCF7 breast cancer cells treated with catalyst (R,R)-1 in combination with MCT-1 inhibitor AZD3965 (0-10 µM). Significance testing (two-tailed t-test assuming unequal sample variances) are reported compared to the absence of co-factor administration. 24 h exposure time, 72 h recovery time. Cell viability was determined using the sulforhodamine B assay. * p < 0.05, ** p < 0.01.     Table S8. Antiproliferative activity modulation (IC50, µM) of MCF7 breast cancer cells treated with catalyst (R,R)-1 in the presence or absence of L-BSO (5 µM). Significance testing (twotailed t-test assuming unequal sample variances) are reported compared to the absence of cofactor administration. 24 h exposure time, 72 h recovery time. Cell viability was determined using the sulforhodamine B assay.