Supplementary Fig. 4: Binding and transport activities for CST_ZM and hCST.

a, CMP binding to CST_ZM using GFP-fusions as estimated by a shift in ΔT_m after addition of 1 mM CMP to WT (open bars), and to mutants of residues not interacting with crystallographic waters (N106A and S267A) in the cavity, but are present in close proximity to the water network (filled bars). Residues W209 and N206 are located in the CMP binding pocket of CST_ZM, but are not conserved in hCST. Non-specific binding was estimated with addition of UMP (red bars). b, The GFP-TS assay was used to determine binding of CMP to purified GFP-fusions of CST_ZM (black squares) and hCST (cyan squares). Binding affinities (K_d) were calculated from data points recorded over a range of CMP concentrations, and these were fitted by non-linear regression using data from 3 independent experiments. As has been observed in other transporters (Shukla, S. et al., J Biol Chem. 292, 7066-7076, 2017), the absolute apparent binding affinities in detergent, although consistently different, were significantly weaker than IC₅₀ and K_M values determined in membranes. c, as in b for the CST_ZM Ser79Ala and hCST Ser95Ala mutants shown to have enhanced CMP binding in (Fig. 2b, c). The binding affinities were estimated using the solubilized membranes. d, as in b for CMP-Sia; see Methods). e, Time-dependent uptake of [³H]-CMP by CST_ZM (open circles). Non-specific uptake was estimated with empty liposomes (filled squares). f, Time-dependent uptake of [³H]-CMP by hCST (open circles). Non-specific uptake was estimated with empty liposomes (filled squares). In all experiments errors bars, s.e.m.; n = 3.