Disruption of the Physical Interaction Between Carbonic Anhydrase IX and the Monocarboxylate Transporter 4 Impacts Lactate Transport in Breast Cancer Cells

Abstract

It has been previously established that breast cancer cells exhibit high expression of the monocarboxylate (lactate) transporters (MCT1 and/or MCT4) and carbonic anhydrase IX (CAIX) and form a functional metabolon for proton-coupled lactate export, thereby stabilizing intracellular pH. CD147 is the MCT accessory protein that facilitates the creation of the MCT/CAIX complex. This study describes how the small molecule Beta-Galactose 2C (BGal2C) blocks the physical and functional interaction between CAIX and either MCT1 or MCT4 in Xenopus oocytes, which reduces the rate of proton and lactate flux with an IC₅₀ of ~90 nM. This value is similar to the K_i for inhibition of CAIX activity. Furthermore, it is shown that BGal2C blocks hypoxia-induced lactate transport in MDA-MB-231 and MCF-7 breast cancer cells, both of which express CAIX. As in oocytes, BGal2C interferes with the physical interaction between CAIX and MCTs in both cell types. Finally, X-ray crystallographic studies highlight unique interactions between BGal2C and a CAIX-mimic that are not observed within the CAII active site and which may underlie the strong specificity of BGal2C for CAIX. These studies demonstrate the utility of a novel sulfonamide in interfering with elevated proton and lactate flux, a hallmark of many solid tumors.

Keywords: BGal2C; MCT1/4; breast cancer; carbonic anhydrase IX; drug development; protein–drug interaction.

Figure 1

BGal2C inhibits the CAIX-induced increase in proton flux in Xenopus oocytes. (A) Chemical structure of BGal2C. (B) Example of an original recording of the intracellular H⁺ concentration in MCT4 + CAIX-expressing Xenopus oocytes. (C) Rate of change in intracellular H⁺ concentration (Δ[H⁺]/Δt) during application of 3 and 10 mM lactate in oocytes expressing MCT4 (lighter shaded bars) or MCT4 + CAIX (darker shaded bars). Gray bars represent control cells; green bars represent cells incubated with 10 μM BGal2C for 1 h. * p ≤ 0.05, n.s. not significant; ANOVA; Mean + S.E.M. (D) Original recording of intracellular H⁺ concentration in a MCT4 + CAIX-expressing oocyte. (E) Relative MCT4 transport activity is plotted against the BGal2C concentration. The regression curve (red line) was created using a Hill1 fit. The green dotted line indicates relative MCT4 transport in oocytes expressing MCT4 without CAIX. The blue data point shows the relative MCT4 transport activity in MCT4 + CAIV-expressing oocytes in the presence of 10 μM BGal2C. * p ≤ 0.05, ** p ≤ 0.01; paired t-test; mean ± S.E.M.

Figure 2

BGal2C inhibits lactate transport in hypoxic breast cancer cells. (A) Fluorescent signals for mTFP (460–500 nm) ((A₁), blue) and Venus (520–550 nm) ((A₂), yellow) from MDA-MB-231 cells, transfected with the lactate-sensitive FRET nanosensor Laconic. (B) Original recordings of the relative change in intracellular lactate concentration during application of 3 and 10 mM lactate in hypoxic MDA-MB-231 cells before (black trace) and after incubation with 10 µM BGal2C (green trace). The red lines indicate the rate of change in intracellular lactate concentration (lactate transport), as shown in figure (C). (C) Rate of change in intracellular lactate concentration during application and removal of lactate in hypoxic MDA-MB-231 cells in the absence (gray bars) or presence (green bars) of 10 μM BGal2C. (D) Rate of change in intracellular lactate concentration during application and removal of lactate in hypoxic MCF-7 cells in the absence or presence of 10 μM BGal2C. (E) Rate of change in intracellular lactate concentration during application and removal of lactate in hypoxic MCF-7 cells, in which CAIX was knocked down with siRNA, in the absence or presence of 10 μM BGal2C. ** p ≤ 0.01, *** p ≤ 0.001, n.s. not significant; paired t-test; mean ± S.E.M.; n = number of cells/number of batches.

Figure 3

BGal2C disrupts the transport metabolon formed between MCTs and CAIX in MDA-MB-231 cells. In situ proximity ligation assay (PLA) for MCT1 + CAIX (A,B) and MCT4 + CAIX (C,D) in hypoxic MDA-MB-231 cells. Primary antibodies (α) included those for MCT1, MCT4 (as appropriate) and CAIX. Cells were pre-incubated without (A,C) or with 10 μM BGal2C for one hour (B,D). For controls, the PLA was performed without primary antibodies (E). Panels (A₁–E₁) show the PLA signals (red), nuclei staining (blue) and actin staining (green). Panels (A₂–E₂) show exclusively the PLA signals. (F) Antibody staining against CAIX (green) without (F₁) or with 10 μM BGal2C (F₂). (G) Quantification of the PLA signals, as illustrated in Panels (A–E). *** p ≤ 0.001; ANOVA; mean + S.E.M.; n = number of pictures/number of batches.

Figure 4

BGal2C disrupts the transport metabolon formed between MCT1 and CAIX in MCF-7 cells. In situ proximity ligation assay (PLA) for MCT1 + CAIX (A,B) in MCF-7 cells. Primary antibodies (α) included those for MCT1 and CAIX. Cells were pre-incubated without (A) or with 10 μM BGal2C for 1 h (B). For control, the PLA was performed without primary antibodies (C). Panels (A₁–C₁) show the PLA signals (red), nuclei staining (blue) and actin staining (green). Panels (A₂–C₂) show exclusively the PLA signals. (D) Antibody staining against CAIX (green) in the absence (D₁) and presence of BGal2C (D₂). (E) Quantification of the PLA signals as PLA signals per nucleus in MCF-7 cells, as described in Panels (A–C). *** p ≤ 0.001; ANOVA; mean + S.E.M.; n = number of pictures/number of batches.

Figure 5

X-ray crystal structure of BGal2C bound to CAII and CAIX. The images depict a surface view of CAII (A) (PDB: 7RRE) and CAIX-mimic (B) (PDB: 7RRF), overlaid onto a cartoon backbone with the inhibitor (yellow) bound. The hydrophobic pocket (orange) and hydrophilic pocket (purple) are shaded, as are the two zinc atoms (magenta spheres). BGal2C binds through a zinc-bound solvent (red sphere) within the active site of hCAII but directly to the zinc within the active site of hCAIX-mimic in addition to on the surface. The boxes show a zoomed-in view with H94, H96 and H119 depicted as sticks. The blue mesh is the density carved to 1.5 Å with a signal to noise of 1.0.

Figure 6

Model of the CAIX-CD147 binding interface. (A) Ig1 domain of CD147 (tan) and catalytic domain of CAIX (green) (taken from (12). (B) Interaction between CAIX His64 and CD147 Glu73. Note the distance between CD147-Glu73 and CAIX-His200 in the “in” confirmation is within hydrogen-bonding range. (C) The structure of BGal2C is superposed onto the CAIX-CD147 structure. Positioning of BGal2C is based on the data shown in Figure 5. Note that amino acids 65–70 of CD147 sterically clash with the sugar motif of BGal2C.

Figure 7

Model of the mode of action of BGal2C. (A) In hypoxic breast cancer cells, glycolysis is the prime energy source, which leads to vast production of lactate and protons. Both ions are rapidly removed from the cell by the transport metabolon formed between MCT and CAIX. In this complex, CAIX, which is directly bound to the Ig1 domain of the MCT1/4 chaperone CD147, serves as a proton antenna for the transporter, which rapidly exchanges H⁺ between the transporter pore and surrounding protonatable residues (blue circles) at the extracellular site of the plasma membrane. Thereby, CAIX counteracts the formation of proton microdomains around the transporter pore and drives the efflux of lactate and protons from the cell. (B) Binding of BGal2C to CAIX disrupts the direct interaction between CAIX and CD147. The disruption of the transport metabolon inhibits MCT transport activity, which leads to intracellular accumulation of lactate and protons, which in turn could result in a decrease in glycolytic activity and ultimately a reduction in cell proliferation.