Application Notes

2 Results MST relies on the analysis of a fluorescent target and its thermophoresis before and after binding its ligand. A ligand that exhibits fluorescence would interfere with fluorescence monitoring, and would require a different experimental setup. Therefore, when setting up label-free MST experiments, a pre-test is conducted to ensure that the ligand of interest does not fluoresce in the relevant range. As expected, maltose did not show any fluorescence, since it doesn't have aromatic structures (Figure 2). The assay buffer MST-T also showed no signal. Figure 2: LabelFree pre-test showing capillaries containing: (1)-(3) MBP at 125 nM, 250 nM, 500 nM, respectively; (4) 1 mM maltose; (5) MST-P buffer; (6) water. All MBP concentrations show a good signal between 2,500 and 25 000 fluorescence counts (1)-(3). Maltose does not show any fluorescence (4), making it a suitable ligand for LabelFree experiments. The buffer also doesn't fluoresce in the relevant range (5). An MST experiment was then set up, using MBP at a constant final concentration of 250 nM. Maltose was titrated, with the highest concentration in the experiment being 1 mM. The resulting binding curve showed a large signal amplitude, yielding high-quality binding data with a K d of 3.3 µM (Figure 3). This is in good agreement with the literature, where an affinity of 1.2-2.3 µM for this interaction has been reported [5, 6]. Figure 3: Binding curve of maltose titrated against MBP. Thermophoretic traces are shown in the inlet. Both graphs display merged data from three independent experiments. The large amplitude of the binding signal of more than 100 F norm units can be explained by the change in conformation that MBP undergoes upon maltose binding [5]. MBP consists of two globular domains that close to bind a maltose molecule. This conformational change was confirmed in recent Surface Acoustic Wave experiments [6]. A conformational change leads to a re-orientation of water molecules in a protein's hydration shell, which is one of the three factors contributing to thermophoretic behavior, besides surface area and charge. A conformational change upon binding can therefore result in a pronounced thermophoretic effect, making MST very sensitive to conformational changes. While this results in a large amplitude in this example, it is by no means necessary for MST. All interactions, with or without binding-induced conformational changes, will have an effect on thermophoresis through surface area, charge, or hydration shell, making them detectable by MST. As a negative control, N,N',N''-triacetylchitotriose, also known as (GlcNAc) 3 , was titrated against MBP. The experiment clearly showed no interaction (Figure 4). Figure 4: Comparison of the dose-response curves of maltose (red) and (GlcNAc) 3 (orange) titrated against MBP. Clearly, (GlcNAc) 3 shows no interaction with MBP. The graph displays merged data from three independent experiments each. The possibility of viscosity artefacts was analyzed using another negative control system consisting of maltose and lysozyme. Maltose was titrated over a wide range starting at a very high concentration (1:3 dilution series, starting at 500 mM). No effect of maltose concentration on lysozyme thermophoresis was observed (Figure 5). This showed that viscosity did not affect the binding assay at concentrations relevant to these

• Cover