• bioprotection
• fast dynamics
• glucose
• levoglucosan
• lysozyme
• mean square displacement
• neutron scattering
• QENS

Lyophilization is a dehydration process widely used to long-term storage of biological samples such as proteins.
However during this process proteins can undergo irreversible degradation. This one can be avoided with inclusion, in the protein formulation, of stabilizers such as polyalcohols or sugars.
Understanding fast dynamics (at timescales in the ns-ps range) of this kind of systems provides important information on how to avoid processes that are responsible for destabilization and degradation of biomolecules and biopharmaceuticals formulations [1, 2].
This thesis presents studies carried out using elastic incoherent neutron scattering, supported by dielectric spectroscopy. By combining neutron and dielectric spectroscopy data, we are able to follow molecular dynamics over an extremely broad frequency range.
Neutron scattering experiments were performed by means of the high resolution backscattering spectrometer SPHERES atMLZ (HeinzMaier-Leibnitz Zentrum,Munich,Germany) and the thermal neutron backscattering spectrometer IN13 at ILL (Institut Laue-Langevin, Grenoble, France).
The sample analysed on SPHERES spectrometer is sorbitol, a simple glass forming hydrogen bonded system, which is often used as excipient in formulation of pharmaceuticals.
The main aim of this experiment was to investigate the response of the caged molecule dynamics at short times from nanoseconds to picoseconds, over a broad temperature range, spanning from the deep glassy state to the liquid state, crossing the glass transition temperature (Tg).
An additional feature of such neutron scattering spectrometers is that is possible to register both energy and momentum (Q) exchange of beam neutrons with atoms of the sample: in this way information on both time and space scale of the atomic motions can be acquired. Dynamics has been investigated in terms of mean square atomic displacement (MSD) as a function of temperature. In harmonic solids, at temperature where quantistic effects can be neglected, the MSD usually shows a linear temperature behaviour with temperature.
This happens also in our case but with some deviations. In fact, MSD showed the presence of at least three crossover temperatures, at which the slope of temperature dependence markedly changes. Interestingly, the lowest two temperatures have been found matching the secondary and primary glass transition temperature of sorbitol, i.e. the temperature where localized β- and structural cooperative α- relaxation, respectively, attains a very long time (100 s). In other terms, an evidence has been found that the vitrification of the slow dynamics has a consequence on the fast motions of our systems. Such a relation has been explained in literature within a theoretical framework where slow and fast dynamics in glass-forming systems are coupled [3–5]. Finally, the highest temperature transition found by neutron scattering reveals to be just a result of the relaxation entering in the spectrometer frequency window, as recently explained in literature[6]. At this temperature nothing particular happens to the sample, it is the effect that part of the motions of the systems are faster than the resolution time of the spectrometer: in fact, such motions cannot elastically contribute to the neutron scattering. This phenomenon is usually named in literature “dynamic transition”. Taking advantage of the facilities of SPHERES instrument, a preliminary study has been carried out, aiming to reveal such a motions, analysing the quasi-elastic response, i.e. due to atoms exchanging small amount of energy with neutron. A promising good matching with information coming from dielectric spectroscopy has been found.
On the IN13 spectrometer we have studied more complex systems, such as mixtures of lysozyme with two different sugars in dry and hydrated conditions. Mixtures have been obtained by freeze drying an aqueous solution of lysozyme and sugar at 1:1 weight fraction.
Sugars considered were in particular: levoglucosan (LG) and glucose (G). While lysozyme was a common hydrogenated sample, both sugars were fully deuterated, as well as the water used for the solutions and the hydration was deuterium oxide. Lyophilysed samples so obtained were kept in controlled atmosphere in order to avoid contamination with humidity. This kind of preparation provides samples particularly suitable to single out the protein dynamics in different environment without relevant contributions from the other component of the mixture. In fact, due to the very strong scattering cross section that hydrogen have with respect to deuterium and the other atoms, the incoherent response is dominated by the hydrogen atoms. Therefore the high contrast returns the response of just the protein atoms in different environments. In the same way as done at SPHERES, we consider MSD as a function of temperature and we compare neutron scattering results with dielectric spectroscopy performed on the same samples. Again, looking at theMSD of lysozyme atoms, some crossover temperatures have been found, matching the transitions involving slow dynamics. In some cases also the above mentioned “dynamic transition” was found at high temperature. The most striking result of this study was that lysozyme embedded in levoglucosan shows significantly smaller MSD than glucose at all the temperatures, both below and above the glass transition temperatures. This can explain the better bioprotective properties of LG with respect to G reported in literature [7]. In fact, a direct relation between MSD and stability of proteins has been recently proved by means of experiments and numerical simulations [8].
Summarizing, this thesis work provides some original results that can be useful for further developments and applications.
Among them, we can mention:
1) the strong relation between the amplitude of fast atomic motions (in the timescales from nano- to pico-seconds) and slow dynamics in glass-forming systems
2) the possibility to test, by studying fast dynamics, how much effective can be a glassy matrix in bioprotection.