Fig 1.
Comparison of surface architecture and signal generation of SPR versus FM.
(A) In SPR, the incoming laser light excites surface plasmon polaritons at the gold liquid interface. The momentum of the incident mode needed for the excitation is dependent on the refractive index in the vicinity of the surface (in the evanescent field of the polariton). Hence, there exists an intensity minimum at a certain angle of reflection for a given wavelength, which can be detected by various interrogation schemes (angular, wavelength and intensity). Upon binding of analytes to the surface, the refractive index changes and therefore the momentum needed for the excitation of the polaritons. This leads to a change in the angle at which the minimum of intensity is recorded (angle interrogation). In an SPR experiment, the change in angle of the intensity minimum is recorded over time to create a sensogram. (B) In FM, coherent incoming laser light is coupled to the waveguide on the sensor surface at the incoupling grating. Then, the guided wave propagates in the waveguide and illuminates via the evanescent field the mologram (a coherent assembly of recognition elements called ridges and grooves). Part of the evanescent field is coherently diffracted into the focal spot by constructive interference (red shading). The intensity of this spot is quadratically proportional to the amount of bound mass of the analyte difference between ridges and grooves. Random scattering from non-specific binding at ridges and grooves destructively interferes (the phase of scattered light is shifted by half a wavelength between the center of the red and blue line) and only linearly contributes to the intensity of the focal spot (blue shading). Then, at the end of the sensor, an outcoupling beam is measured in order to translate the measured intensity into a diffraction efficiency of the mologram and from this into a coherent mass density. In a FM experiment, the change of intensity (diffraction efficiency) is recorded over time to create a sensogram. In SPR, referencing is carried out by measuring two sensors in separate channels (mm apart) independently and subtracting the reference from the signal digitally. Contrary, in FM referencing is intrinsic (due to alternating ridges and grooves, only a 200 nm apart) and is carried out by the wave nature of light. Only the difference is measured by the intensity in the focal spot. This self-referencing on the submicron scale over 1000 of signal and reference regions per spot and direct differential detection by diffraction is what renders FM much more robust than refractometric sensors (e.g., SPR). Figure was created with BioRender.
Fig 2.
Backfilling the FM sensor surface with click chemistry.
Left: The time trace sensogram collected while backfilling of FM sensor with TCF4(7-30) peptide and its mutant. The median trace out of 18 molograms is shown. The coherent mass density increase is displayed on the y-axis. The x-axis shows the time dimension. Right: Scheme of backfilling process on the sensor surface. Initial surface chemistry; MTz in ridges and Tz in grooves (Step 1). In step 2, the non-binding mutant peptide was injected and primarily reacted with Tz in grooves. In step 3, since all Tz is depleted in grooves, the peptide only immobilized in ridges. In step 4, the binding peptide was only immobilized in the ridges until the signal decreased back to baseline. Step 5 depicts fully backfilled sensor surface. Figure was created with BioRender.
Fig 3.
FM and SPR measurements of β-catenin/TCF4 interactions and β-catenin competition with a free peptide.
Normalized sigmoidal dose response (y-axis) of β-catenin (x-axis) response versus concentration monitored in buffer (A) and cell lysate (B). Normalized sigmoidal dose response (y-axis) of TCF4(7-51) (x-axis) response in solution with β-catenin versus concentration monitored in buffer (C) and cell lysate (D). For FM experiments, a representative sensor containing 18 molograms, the normalized equilibrium response (Req) for each injection of protein or peptide is plotted from 3 individual experiments. For SPR experiments, the normalized equilibrium response (Req) for each injection of protein or peptide from 3 experiments on the same sensor containing 3 different immobilization densities (5, 25, 50 RU, which under the assumption 1 pg/mm2 = 1 RU, corresponds to 2, 10, 20 fmol/mm2) is shown. The 95% confidence interval band is shown as dotted line for each experiment. The normalized sigmoidal dose response curve obtained from SPR experiments in buffer are shown in all plots for comparison to FM data. The sigmoidal dose response curves were fitted with the non-linear regression model in Graphpad Prism 10.4.1 (log(inhibitor) vs. normalized response – Variable slope).
Fig 4.
Exemplary time traces of beta-catenin titration over TCF4(7-30) peptide surface and corresponding competition experiments with a free TCF4(7-51) peptide collected with FM.
(A) Dose response of beta-catenin (up to 1000 nM) to determine binding affinity constant; (B) Competition of beta-catenin by a free peptide TCF(7-51) titrated in dose response (up to 1000 nM) at 250 nM beta-catenin.
Fig 5.
Overview of binding affinities (KD) and half maximum inhibitory concentrations (IC50) obtained in FM measurements performed in buffer.
The experiments were in triplicates on 3 different sensors. The boxplot analysis represents the indivudual binding affinities (A, KD) and half maximum inhibitory concentrations (B, IC50) calculated from 18 indivudual molograms with software provided by Lino Biotec. The box contains 50% of the values and the median value is represented as line, while the whiskers contain the upper and lower 25% of all molograms of an experiment. (C) Median KD and IC50 values calculated from 18 molograms as well as the avarage values calculated from three experiments with standart deviation.
Fig 6.
Overview of binding affinities (KD) and half maximum inhibitory concentrations (IC50) obtained in FM measurements performed in cell lysate (1 Mio cells/mL).
The experiments were in triplicates on 3 different sensors. The boxplot analysis represents the indivudual binding affinities (A, KD) and half maximum inhibitory concentrations (B, IC50) calculated from 18 indivudual molograms with software provided by Lino Biotec. The box contains 50% of the values and the median value is represented as line, while the whiskers contain the upper and lower 25% of all molograms of an experiment. (C) Median KD and IC50 values calculated from 18 molograms as well as the avarage values calculated from three experiments with standart deviation.