Probing the existence of non-thermal Terahertz radiation induced changes of the protein solution structure

During the last decades discussions were taking place on the existence of global, non-thermal structural changes in biological macromolecules induced by Terahertz (THz) radiation. Despite numerous studies, a clear experimental proof of this effect for biological particles in solution is still missing. We developed a setup combining THz-irradiation with small angle X-ray scattering (SAXS), which is a sensitive method for detecting the expected structural changes. We investigated in detail protein systems with different shape morphologies (bovine serum albumin, microtubules), which have been proposed to be susceptible to THz-radiation, under variable parameters (THz wavelength, THz power densities up to 6.8 mW/cm2, protein concentrations). None of the studied systems and conditions revealed structural changes detectable by SAXS suggesting that the expected non-thermal THz-induced effects do not lead to alterations of the overall structures, which are revealed by scattering from dissolved macromolecules. This leaves us with the conclusion that, if such effects are present, these are either local or outside of the spectrum and power range covered by the present study.


SI-
: Extracted parameters for SAXS data collection of a BSA solution (c = 5.8 mg/mL) exposed to Φ = 6.5 mW/cm 2  In another attempt to explore if THz-radiation can induce any structural changes in BSA that can be detected by SAXS, lyophilized BSA powder was exposed to THz radiation for ~24 hours using THz source II. With this approach, we follow the study of Cherkasova et al. 1 , who irradiated BSA powder with radiation of 3.6 THz and 10 mW and reported changes in the UV absorption and circular dichroism spectra, and related these to conformational changes of the protein. Our THz setup II covers in particular 3.6 THz, the frequency used in that study.
Two aliquots of the same batch of lyophilized BSA powder were filled into standard polypropylene tubes. One of these BSA powders was exposed at ambient temperature for ~24 h to THz radiation of source II within its tube, which is transparent to THz 2 . The other specimen was stored similarly but non-exposed to THz. Afterwards, both samples were shipped to the beamline, where the powders were dissolved in HEPES buffer immediately before the data collection. Standard batch mode measurements were performed within an in-vacuum quartz capillary using the robotic P12 sample changer with continuous flow 3 . Before and after the protein solutions, the buffer solution was measured. In addition, a third, freshly prepared BSA sample was measured for comparison. For all three samples, several SAXS data collections were performed within a time range of ~150 min. For each collection, 40 frames of 100 ms exposure time were taken. Similar to the previous spectroscopic study, SAXS patterns were taken at different time points after dissolving the powders.
The resulting SAXS curves -similar to those from the combined THz-SAXS measurements -do not reveal any significant structural changes (SI- Fig. 4a). While a weak but systematic decrease of the radius of gyration (SI- Fig. 4b) and an increase of the monomer fraction (SI- Fig. 4c) can be detected for both samples, which we attribute to a dissociation of dimers after mixing, the difference between exposed and non-exposed samples are neglectable (SI- Fig. 4d). The difference to the third reference BSA sample is effectively larger than those between the samples exposed and non-exposed to THz.
It should be noted that due the lower scattering background of the sample changer setup, SAXS curves up to larger scattering angles can be determined. As these can be all reasonably well fitted as a monomer-dimer mixture, also significant THz-induced changes in the wide-angle scattering part of the curves can be excluded. There have been also no effects due to storing both samples at ambient temperature for the time of THz-exposure. As the previously mentioned study actually reported on changes in the solution structure of BSA after THz exposure of the powder 1 , our results reveal that any such effects appear to be highly localized and not affecting the protein structure on any length scale accessible by SAXS in our experiment. While a full cylinder of radius R = 16.5 nm can describe the position of the first minimum for the batch II data, it fails to match the second minima. A better description is achieved for the hollow cylinder model of inner radius R i = 8.4 nm and outer radius R o = 13.2 nm. Note that the wall thickness t = R o -R i = 4.8 nm is similar to the value reported in previous studies 4,5 . The smearing of the experimental data is due to the size polydispersity of MT, which was also reported previously. Because of this, we limited our analysis to a qualitative description of the SAXS profiles, which is feasible for understanding the structural features of the p(r) function.
For the SAXS profiles of two touching (hollow) cylinders, the position of the minima is unaffected but an additional modulation of the curves can be seen at smaller angles (s < 0.2 nm -1 ). This contribution is significant when looking on the corresponding p(r) functions. For both types of single cylinders, a clear maximum can be seen for r = 20 -30 nm, whose actual position depends on the filling (full vs hollow), and is similar to the cylinder diameter. The weak modulations at larger distances are attributed to artefacts due to the indirect Fourier transform from the modelled curves and are not further considered.
For the pair-distance distribution of the two touching cylinders (cylinder dimers), however, the maximum is much broader and shifted to larger distances, reflecting the effective dimer cross-section.
The contribution of the cylinder diameter has nearly disappeared and is only implied by a weak shoulder in case of the full cylinder diameter.
Comparing the p(r) functions from the experimental data with the modelled curves suggests, that for both MT batches, at least two types of MT species are present in solution, MT monomers and MT dimers, whose composition changes between the batches.

SI-Figure 6: Computed SAXS profiles from high-resolution crystallographic structures for different
THz-exposure. a) Hen egg white lysozyme crystals exposed to 12 mW @ 0.4 THz (Ref. 6), b) Bovine Trypsin crystals exposed to 1 mW @ 0.5 THz (Ref. 7). SAXS curves were computed using CRYSOL and shifted for clarity. c) Relative changes ΔR g /R g upon THz-exposure for lysozyme (Lys) and different trypsin (Try) crystals. Reference data refer to crystal samples not exposed to THz-radiation.
In order to compare the experimental SAXS curves for proteins in solutions from the present study with the results from previous X-ray crystallography experiments of protein crystals exposed to THzradiation 6, 7 , theoretical small angle scattering profiles of the crystallographic structures were computed using the ATSAS program CRYSOL. Fig. 6a) shows the computed SAXS profiles of lysozyme crystals alternatingly irradiated with THzradiation in the non-exposed state (THz off) and when exposed to 12 mW@ 0.4 THz (THz on) as well as of a reference crystal that was not exposed at all.

SI-
While a detailed analysis of the crystallographic data indicates slight changes in the electron density of a single helix upon THz-radiation 6 , the theoretical SAXS curves do not show any indication of a change of protein structure. The same conclusion is drawn for crystal data from trypsin exposed to 1 mW@ 0.5 THz (Fig. 6b), for which an increase of the anisotropy of atomic displacements for neighbouring residues was reported 7 .
In terms of the relative change of the radius of gyration for both proteins (Fig. 6c), alternating THzexposure results in more changes than for the reference samples, however, the effect of these changes is less than 0.1%. This is even weaker than the effect observed for the experimental SAXS curves of BSA. Despite the close packing in the crystal, and the absence of large changes, an influence of proteins in solution may be possible. §SI-5 THz spectra from the microfluidic cell The microfluidic cell was designed for combined THz-SAXS measurements 2 . In combination with THz source II, it is possible to determine THz spectra from aqueous solutions. In particular, it allows determining the THz transmission. SI- Fig. 7 shows the time-domain THz signal from source II (without cell), from the empty cell and from the cell filled with buffer and BSA solution. When passing the cell, the top-to-bottom intensity drops from ~ 200 nA to ~ 90 nA, which is close to the expected transmission of ~ 50% of the two polystyrene windows of a total thickness of 3 mm. When filling the cell with an aqueous solution (either buffer or BSA solution), the transmitted intensity significantly drops (< 3 nA), revealing strong THz absorption of the solution. Thus, it is assured that the power of THz laser is mainly deposited in the sample volume.
This strong difference of THz absorption was also utilized for the alignment of the THz cell with respect to the X-ray beam. To ensure that the sample was completely illuminated, the transmitted signal was recorded. Positioning the cell with filled channel and minimizing the transmitted signal showed that the sample was properly exposed to the THz beam and thus the maximum power deposited.