Proteins are essential for life because they are able to perform and manage biological functions in such a complex way that much care is needed for the comprehension of these mechanisms at different levels. Although each protein accomplishes a specific task, we also need to be able to describe their common properties and characteristics. Proteins, consisting of complex linear chains of amino acid residues, can be considered as polypeptides. The amino acid residues, depending on their position in the chain, interact with each other and can form collapsed structures such as
-strands. Indeed, proteins are characterized by a stable three-dimensional structure, called the native state, indicating the precise way of folding of that protein, and ultimately, the biological activity developed by the protein itself [1
]. Although a protein with 100 residues needs from milliseconds to seconds in order to fold, the number of possible configurations is in the order of 10
, which corresponds to a folding time in the order of about 10
]. This inconsistency, known as Levinthal’s paradox [4
], was, and actually is, the subject of numerous studies that try to address this important issue [5
]. The main consideration concerns the existence of different interactions between amino acids with a different weight in term of strength and priority. These selective interactions, together with random thermal motions, induce fast conformational changes with a progressively lower potential energy that allow the formation of the native structure of proteins. Indeed, not every configuration will be explored but only those that lead to the most stable configuration. This mechanism is well illustrated by a funnel-shaped energy landscape [1
]. When proteins are heated beyond the so-called denaturation threshold, they unfold irreversibly. Above the irreversible denaturation threshold, proteins lose their three-dimensional structure becoming a linear chain of amino acids and do not function any more. Furthermore, denatured proteins manifest heavy configurational changes, loss of solubility and aggregation processes. The unfolding process is strongly dependent on the temperature and on the time that proteins remain (permanence time) at that temperature. Under certain thermodynamic conditions (e.g., pressure and concentration), it can be reversed allowing proteins to re-fold in the right way [10
]. Hence, upon heating above a certain temperature in the range 40–60
C, the protein native state undergoes rapid conformational changes before it unfolds completely. These structural changes characterize the so-called intermediate state (see Figure 1
) populated by a broad range of structures resulting from competing mechanisms, such as hydrophilic and hydrophobic interactions [13
], and large-scale motion [14
]. This competition depends on the temperature, and recently the value of about 280 K has been identified as the onset temperature of the hydrophobic effect [15
The intermediate state is highly unstable [18
] and sometimes the folding pathway of proteins is altered: proteins do not fold correctly and assume a wrong
three-dimensional structure (misfolding). Actually, there is much interest in determining the cause of misfolding because it is one of the origins of many degenerative illnesses, such as Parkinson’s and Alzheimer’s diseases [19
]. Comprehension at a microscopic level of the different mechanisms involved in the folding/unfolding process has indeed a primary role in understanding the causes and eventually finding the remedy to stop or slow down the advance of the wrong reactions/interactions. Finally, the so-called protein folding problem concerns the understanding of the physical processes that drive the polypeptide chain towards the lowest free energy state [8
] which is determined by the balance between enthalpic and entropic costs involving both the protein and its hydrating solvent. Water, thanks to its abilities in developing hydrogen bond networks, is essentially the unique solvent able to control the structure, stability, dynamics, and thus the function of biomolecules. More profoundly, water molecules interacting with a protein may have both a structural and a dynamical character. Water molecules that are located within the protein internal cavities constitute part of the protein structure. Moreover, water molecules that cover the protein surface and make part of the so-called first hydration layer, allow protein residues to have the right dynamical flexibility that determine its specific biological activity [21
]. In this context, the study of the properties and the influence of hydration water with respect to the proteins structure and dynamics, is ultimately fundamental to having a clear picture of the mechanisms involved in the folding/unfolding processes.
Since the water monolayer covering the protein surface is essentially a bidimensional hydrogen-bonded network connecting the different water clusters and the hydrophilic moieties of the protein surface [22
], the protein dynamics, and thus its function, is slave to water dynamics [24
]. Hence, the magnitude of the self-diffusion of hydration water strictly influences the large-amplitude motion of protein residues that is needed for the corresponding functionality [27
Actually, the comprehension of the various kinds of interactions occurring between the different protein moieties and the solvent is an intriguing open question [29
]. It is fundamental to analyze the interplay between hydrophilic and hydrophobic interactions that moves the thermodynamic properties of the system toward the corresponding equilibrium state. This interplay is strongly influenced by the temperature favoring the formation of hydrogen-bonded structures on cooling. Note that the secondary structure of proteins such as
-sheets builds up by means of hydrophilic interactions and indeed hydrogen bonds [30
], except for various membrane proteins for which the additional interaction with the lipid bilayer must be taken into account [31
]. In order to focus such issues, proper experiments must be planned. It is well accepted that internal water favors the first steps of protein folding but at the end is partially thrown out from the inner core through cooperative hydrophobic interactions mediated by the corresponding hydrophobic moieties of proteins [33
When water solvates amphiphilic molecules, the interplay between hydrophilic and hydrophobic interactions provokes a decrease in entropy with the subsequent build-up of well defined geometric structures such as ellipsoids, cylinders, layers, bilayers, etc. [35
]. Thermodynamical variables, chemical composition, and the high directionality of hydrogen bonds determine the final geometry and the three-dimensional structural arrangement [36
Recent theoretical advances allowed the prediction of protein structure starting from the knowledge of the amino acid sequence, thanks to the inclusion of water-mediated interactions for the study of the folding process [39
]. One of the most promising approaches for the interpretation of the protein folding problem is based on a statistical approach that describes the protein’s potential surface as a rugged funnel-like landscape [40
]. This latter approach is known as the free energy landscape theory, and by means of the same statistical approach used for the description of disordered systems, polymers, and phase transitions of finite systems [41
], it considers that the folding process proceeds towards the formation of an ensemble of well defined structures finally “precipitating” in the protein native state [40
reports a schematic energy landscape with only two wells with different heights and widths corresponding to the high-temperature conformation of an almost linear chain of amino acids (right side) and to the stable native configuration with the corresponding three-dimensional structure (left side) [45
]. It is worth noting that our results, discussed in the next section, show that hydrogen bonds are effective and “incisive" only for the lowest well and here water molecules can develop a soft tetrahedral network surrounding the protein surface. On the contrary, at high temperature, all water molecules are essentially free and hydrogen bonds are not able to form between either protein residues or water molecules.
Herein, we review our recent results obtained by means of Nuclear Magnetic Resonance (NMR) and Fourier Transform InfraRed (FTIR) spectroscopies concerning the investigation of the mechanisms that intervene in and determine the evolution of the folding/unfolding process of hydrated lysozyme. Lysozyme is a small globular protein of about 15 kDa with antibacterial properties and was discovered by Fleming in 1922 during a search for medical antibiotics. Being ubiquitous among living organisms, lysozyme is probably the most studied enzyme in many different scientific fields including physics, biology, and medicine both theoretically and experimentally [11
We make use of the mentioned experimental probes in order to obtain new microscopic insights for a better elucidation of those processes and interactions that determine the evolution of the protein folding/unfolding process. In fact, even when working on two different energy/time scales, both techniques provide microscopic details for water and/or the individual functional groups. We report our findings obtained by exploring the Energy Landscape from the thermal region of configurational stability, i.e., that of the protein native state, up to that of the irreversible denaturation where the protein reverts back to its unfolded state, a linear chain of amino acids. By means of different kinds of experimental investigations we figure out the role of water in mediating the relevant interactions as well as the importance of hydrophobic moieties for a correct folding pathway.
Overall, the aim of this review is to bring together the knowledge acquired to give a deeper comprehension of the investigated processes. This is done by considering existing interpretations and by suggesting future perspectives.
3. Materials and Methods
We studied hydrated hen egg white lysozyme with hydration level corresponding to the first hydration shell. In such a way we were able to focus on the most important interactions exerted between the protein and the water molecules. Samples were prepared by hydrating dried protein powders isopiestically at 5 C in a closed chamber with 100% relative humidity until the hydration levels of h = 0.3 and 0.37 were achieved. In our experiments we used two different techniques that enabled us to obtain details underlying the folding/unfolding process of proteins. FTIR measurements were performed at atmospheric pressure using a PerkinElmer Spectrum GX Fourier transform spectrometer with an attenuated total reflection geometry. Spectra were recorded with a resolution of 4 cm by performing 250 repetitive scans. The temperature scan was started at 180 K; we acquired the spectrum each 10 K up to 350 K by maintaining a stability of 0.1 K.
NMR experiments were performed by using a Bruker AVANCE 700-MHz NMR spectrometer. We acquired
H NMR spectra of hydrated lysozyme with h = 0.3 as a function of temperature in diverse heating–cooling cycles with an accuracy of ±0.1 K controlled by using a cold nitrogen flow and a heating element. The complete investigated temperature range was 298–366 K, with 1 K steps for those thermal cycles that invert before T
345 K and 2 K steps for the complete cycles that invert after the irreversible denaturation. The chemical shift of ethylene glycol was used for temperature calibration. Since the samples were essentially hydrated powders, in order to enhance the spectral resolution, we used the experimental technique known as High-Resolution Magic Angle Spinning. This is a very powerful NMR experimental technique to study semi-solid systems such as low-hydrated protein powders. It takes advantage of spinning the sample holder at high frequency (starting from 4000 rpm for biological materials) while it is tilted at about 54.74
with respect to the direction of the static magnetic field. This last condition ensures that the Hamiltonian terms, including the dipolar coupling which is responsible for broadening NMR peaks, can be neglected and the NMR signal is more resolved [97
]. The possibility to study intact tissues or bio-systems without the need for any chemical treatment has aroused great interest towards this technique [98
]. Finally in our experiments, hydrated protein powders were placed in a 50
L rotor and spun at 4000 Hz to enhance the field homogeneity. We used 64 k points in the time domain, 128 scans and a minimum relaxation delay of 2 s. All spectra were processed (line broadening, Fourier transform, phase correction, and baseline adjustment) by using the standard routines of the Bruker software Xwinnmr version 3.5. Further details on sample and experimental methods can be found elsewhere [11
In this work, we review our recent results obtained on hydrated lysozyme by means of FTIR and NMR experimental techniques. Our findings indicate that the hydrogen bonds formed between water molecules and both the carbonyl oxygen (C=O) and the secondary amide hydrogen (N–H) of protein or peptide trigger the low temperature (≈225 K) dynamic transition, below which the protein is not active. At 350 K, above the lysozyme denaturation threshold, the signal from antiparallel -sheets is enhanced. The onset of this contribution marks the helix-to-sheet transformation and the subsequent processes associated with protein aggregation. The thermal behavior of the FWHM of the Amide II modes highlights the thermal regions where the energetic transfer with the O–H bending mode of water is favored, that is just when lysozyme is in its native state up to the reversible region of the unfolding process. Water molecules begin to loose their ability to develop stable hydrogen bonds for T > T* ≈ 320 K: their lifetime decreases on increasing the temperature.
Our NMR experiments shed light on the different energetic evolutions of both hydrophilic and hydrophobic moieties of hydrated lysozyme during the processes of folding and unfolding. The magnetization of the CH group shows contrasting behavior with respect to that of both the CH and NH groups, evidencing an enhanced mobility for the former. This is caused, on the one hand, by the burial effect that hydrogen bonding produces mainly on the more hydrophobic methyl groups, and on the other hand, by the stronger hydrogen bonds formed by water with the secondary amine groups.
Future studies aiming to rationalize the energetic behaviors of similar processes, such as those involving the formation of amyloid fibrils, may take advantage of these approaches and corresponding results. Furthermore, our findings may help understand the role of water in the protein function of organisms such as hyperthermophiles and psycrophiles living in extreme temperatures [102
], as well as its mediation in protein-sugar systems [103