Effects of sound energy on proteins and their complexes

Mechanical energy in the form of ultrasound and protein complexes intuitively have been considered as two distinct unrelated topics. However, in the past few years, increasingly more attention has been paid to the ability of ultrasound to induce chemical modifications on protein molecules that further change protein–protein interaction and protein self‐assembling behavior. Despite efforts to decipher the exact structure and the behavior‐modifying effects of ultrasound on proteins, our current understanding of these aspects remains limited. The limitation arises from the complexity of both phenomena. Ultrasound produces multiple chemical, mechanical, and thermal effects in aqueous media. Proteins are dynamic molecules with diverse complexation mechanisms. This review provides an exhaustive analysis of the progress made in better understanding the role of ultrasound in protein complexation. It describes in detail how ultrasound affects an aqueous environment and the impact of each effect separately and when combined with the protein structure and fold, the protein–protein interaction, and finally the protein self‐assembly. It specifically focuses on modifying role of ultrasound in amyloid self‐assembly, where the latter is associated with multiple neurodegenerative disorders.

Frequencies between 20 and 100 kHz are mainly utilized in the food industry, while US frequencies above 1 MHz are applied for medical purposes and industrial imaging.Powerful ultrasound can modify the medium in which it propagates [6].Generally, US induces physical effects, which include thermal, mechanical, and chemical effects on a liquid.The waves generated by US propagate in a medium, creating alternating pressure cycles, namely, rarefactions and compressions (Fig. 1A).These cycles, in turn, create correspondingly an alternating negative and positive pressure in the medium, which ruptures the fluid to form cavitation bubbles [7].After several cycles of rarefaction and compression, cavitation bubbles resonate with the propagating waves, consequently contracting and expanding.At a certain point, however, the external pressure force cannot restrain the cavity bubble, and when local pressure is restored, the bubble reaches its critical size and collapses, creating intense microturbulence, resulting in a huge increase in pressure and temperature; it was shown that US-induced cavitation bubble collapse can produce very high localized temperatures up to 5000 K and pressures of more than 1000 bar [8,9].
With an increase in ultrasound frequency, the duration of the rarefaction cycle decreases, indicating that generally, lower frequency US creates better cavitation because the cycles of compressions and rarefactions are longer.At higher frequencies, rarefaction cycles do not have the time needed for cavitation bubbles to grow to a critical size [17,18].In this regard, depending on the frequency, US regimes conditionally can be divided into either cavitating (< 1-2 MHz) or noncavitating types (> 1-2 MHz); however, it is worth mentioning that these modes may change; they also depend on the acoustic energy density [19].
Cavitation is not limited to purely physical effects, such as pressure and temperature; it is also affected by macroturbulence that occurs in the medium responsible for mechanical impact, manifested in microjet formation [20] and acoustic streaming determined as a steady flow in fluid driven by the absorption of the high-amplitude acoustic oscillations.In addition, it induces intensive liquid mixing and consequently, strong shear forces are generated in the adjacent fluid [21].Strong shear forces may lead to the mechanical activation of a particular chemical reaction or to mechanical bond cleavage [22].It was found that the shear forces created by cavitation bubbles collapsing are the strongest at low frequencies  and fall significantly as the frequency increases to 1 MHz [22][23][24].
In addition to the mechanical and physical effects of US, irradiation of aqueous solutions with US may initiate the formation of free radicals (especially Á HO) as a cavitation-driven result of water sonolysis [25][26][27].
The hydroxyl radical has a high oxidizing potential and can directly oxidize organic substrates.The rate of the formation and accumulation of free radicals in a solution during ultrasound treatment depends on the dose of acoustic energy, which, in turn, determines the intensity of the cavitation process; an increase in the intensity of ultrasound usually leads to an increase in the number of active cavitation bubbles, which also accelerates the formation of hydroxyl radicals [28].In addition, sonication may produce a significant amount of hydrogen peroxide [29].With 20 kHz of US, the H 2 O 2 concentration linearly increases with the irradiation time within an hour and with the US acoustic intensity, it almost does not depend on pH.The rate of radical output in sonochemical reactions increases with frequency to a maximum in the medium and high-frequency range (about 300-800 kHz) and then decreases at higher frequencies (> 1 MHz) [23].
Ashokkumar et al. [30] found that the radical action has a peak at 358 kHz, whereas maxima from 582 to 863 kHz were reported by Mason et al. [19,31].This is due to the large radius of the bubble before collapsing at lower frequencies [32], leading to a larger volume of voids, which the surrounding liquid will tend to replace while collapsing.In sum, this set of factors, created by collapsing cavitation bubbles, is strong enough to induce significant damage to many molecules and materials.
Since the discovery of non-audible sound waves in 1794, ultrasound technology has become a conceptual tool that paved the way for key healthcare applications, ranging from diagnostics and disease treatment to organic sonochemistry [33] and ultrasoundresponsive innovative materials [34,35].Its biological, chemical, and physical effects have been extensively studied [36].
Sonochemistry and sonochemical treatments [37,38] are also actively used in protein physics and chemistry.US is used to activate and inactivate enzymes, depending on the conditions [39].Its activity mainly depends on the amino acid composition and protein conformation [40].Recently, US has been proposed for Alzheimer's and Parkinson's disease treatment, specifically for enhancing blood-brain barrier permeability (the cavitating regime) and for amyloidogenic plaque ablation (the non-cavitating regime).It was also found that US can be utilized to disaggregate amyloid plaques when using scanning US, enabling removal of amyloid-β [41,42].US has been widely used for protein extraction [43], consequently influencing and changing its functional properties [44,45].However, despite a constant interest in the intersection between the US field and the protein complex field, there is still scattered information about the effects of ultrasound on protein molecules and their complexes.
Importantly, the primary factors influencing protein structure during sonication are the intensity of the US, frequency, and exposure time (direct sonication Fig. 1.Physicochemical effects of ultrasound on proteins: (A) main US-derived factors affecting protein structure upon US irradiation and a mechanistic representation of protein monomers' interaction and the structural changes imposed by cavitation.The image of the acoustic streaming effect was adapted from [10] Copyright (2015) with permission from Elsevier; (B) effect of US-formed radicals on amino acids and the primary protein structure [11,12]; (C) effect of US on the protein secondary structure (C1), adapted from [13], Copyright 2023 the Authors, published by PNAS, 2023, and on the tertiary structure (C2), adapted from [14], Copyright 2023 the Authors; (D) effect of US on protein fibrillar self-assembly; (E) effect on the disassembly of protein complexes.the right-hand panel shows the effects of US on proteins at different organization levels.Separate images were created with BioRender.com,Blender 2.93, and Autodesk Fusion 360.Protein structures of C2 were prepared using CHIMERAX 1.7 [15,16].parameters).Besides these parameters, US can also impact other protein properties, such as solubility, the viscosity of the solutions, the emulsifying properties, conductivity, and enzymatic activity [46][47][48][49].When conducting sonication, several essential factors come into play, including the type of sonicator (probe or bath), its size and shape, its position in the solution, the presence of other molecules, and the composition of the solution [50].
This review focuses solely on the direct impact of US, considering factors such as the frequency, sonication power, and related parameters (e.g., energy) on the structural composition of proteins, including their primary (Fig. 1B), secondary, and tertiary (Fig. 1C), as well as quaternary structures.In addition, it focuses on the proteins' ability to self-assemble (Fig. 1D) into protein complexes as well as their disassembly (Fig. 1E).In addition, this review will discuss how such structural modifications are reflected in changes in the intrinsic protein characteristics [51].However, we will not discuss the biological effects [52] of US treatment, its interactions with living organisms, or applications in food protein chemistry/industry.Furthermore, we will discuss the US effects on amyloidogenic protein self-assembly and amyloid disassembly as well as on individual proteins rather than on 'protein isolates', 'protein hydrolysates', or 'protein concentrates'.The impact of US treatment on such systems can be found in the literature [53][54][55][56][57][58][59][60][61][62][63][64][65][66][67].

Effects of mechanical energy generated by US on the protein structure
Depending on the US parameters, different proteins can exhibit various behaviors.Table 1 briefly summarizes the impact of ultrasound conditions (the frequency and power parameters) on selected studied proteins.This includes effects on the primary, secondary, and tertiary structures and the self-assembly (the formation of supramolecular complexes) or the disassembly of protein complexes, with special emphasis on fibrillar assemblies.
Based on an overview of the literature, it can be concluded that for sonochemistry applications [91], as well as scientific research, there is a general preference for the use of low to medium frequency ultrasound, up to 1-2 MHz, due to the diverse effects produced within this range.The impact of ultrasound on protein structure has been explored across a broad frequency range from 20 kHz to 1 MHz frequencies (see Table 1).In particular, low-frequency ultrasound up to 50 kHz is appealing due to its cavitation effect and high shear rates.Mid-range ultrasound at around 500 kHz is advantageous for free radical generation and a medium shear regime.Ultrasound above 1 MHz is essentially devoid of these effects.It is worth noting that sonochemical activity at high frequencies requires more power, since more energy is needed to accelerate particle interactions.Therefore, operating frequencies in the 20-50 kHz range have been traditionally favored by chemists.However, frequencies in the 100 kHz to 1 MHz range have recently received increased interest owing to their advantages in generating free radicals and in developing affordable equipment capable of producing sound waves within these frequencies at high operating powers [23].
In general, it can be concluded that changes in the structure of proteins occur at all levels of their structural organization.Ultrasound can affect all types of bonds (covalent, ionic, hydrogen, disulfide, or hydrophobic) comprising the protein structure.Such USimposed changes modify the intra/inter-molecular interactions that underlie further structural modifications of proteins [7].It was found that proteins can undergo partial unfolding under the action of ultrasound, especially at low intensity.This process is accompanied by changes in the secondary structure of proteins, manifested in a change in the helix/sheet/loop composition.Such changes can cause rupture of many non-covalent bonds as well as changes in the tertiary structure.The prolonged action of high-intensity ultrasound can break covalent peptide bonds, leading to the formation of shorter peptide fragments and even individual amino acids [7].An increase in exposure time also leads to changes in aromatic or sulfurcontaining amino acids due to an increase in the concentration of US-formed radicals in the environment surrounding the protein molecule, ultimately forming disulfide (S-S) cross-links.There are many reports on the formation of protein aggregates or supramolecular protein complexes that are enabled either with or without US-imposed changes in the primary, secondary, and tertiary structures [7].In addition, if pre-formed protein complexes (in particular, fibrillar complexes) are already present in the solution, their destruction occurs because of US exposure.

Effects on amino acids and the primary protein structures
Highly reactive free radicals, particularly hydroxyl radicals ( Á OH), have a profound effect on small molecules [92][93][94][95][96] and, in particular, on amino acids and the primary structure of proteins.Such changes were investigated in detail by using UV spectroscopy,   polyacrylamide gels, chromatography, or by using mass spectrometry.When reactive oxygen species (ROS) interact with protein molecules, the abstraction of a hydrogen atom from an amino acid residue can take place, forming radicals centered on a carbon atom.Consequently, these radicals then react with O 2 to form peroxyl radicals and if formed on alpha carbon, then such compounds usually decompose due to instability, forming new amino derivatives, which subsequently undergo spontaneous hydrolysis to form two peptide fragments.When peroxyl radicals form on amino acid side chains, they can rearrange into alcohols or aldehydes through reaction pathways that currently are not fully understood.Exposing tryptophan to hydroxyl radicals leads to the opening of the indole ring and the formation of kynurenine.Histidine reacts to generate 2-oxohistidine, asparagine, and aspartate.Free radicals reacting with tyrosyl residues produce tyrosyl phenoxy radicals.The predominant fate of these structures is to undergo selfdimerization, resulting in the formation of di-tyrosine cross-links.In contrast, phenylalanine oxidation generates ortho-(2-hydroxyphenylalanine), meta-tyrosine (3hydroxyphenylalanine), and tyrosine [12] (see Fig. 1A,  B).If the action of the radicals is high enough, an oxidative cross-linking of the proteins may occur [97].
By using an EPR-spin trapping technique, P. Riesz and his team detected the formation of superoxide anion radicals ( Á O À 2 ) in an argon-saturated aqueous solution induced by 50 kHz ultrasound.The existence of hydroxyl radicals Á HO and hydroperoxyl radicals Á HOO was confirmed in an oxygen-saturated solution.
However, no traces of hydroperoxyl radicals Á HOO were found in the argon-saturated solution.These data suggest that in an oxygen-free environment, superoxide anion radicals Á O À 2 are formed directly as a consequence of water sonolysis that occurred at a sufficiently high temperature resulting from cavitation bubble collapse [98].
In another study performed using UV spectroscopy, Castellano et al. [68] found that sonication (at 1 MHz) of a myoglobin-rich solution changes the absorption spectra of the 280 nm band.This observation suggests that the aromatic rings of aromatic amino acids are modified by sonication.
According to Tao et al. [71], treatment of myofibrillar protein with 20 kHz US leads to an increase in number of free amines and to a slight increase in the free sulfhydryl groups, manifested by the ultrasoundinduced partial unfolding of the protein.
Note that free sulfhydryl groups might undergo further oxidation.Chen, Kong et al. [72] suggested that under oxidative stress caused by radicals, the cysteine residues located in the tail of the myosin heavy chain could be prone to forming intermolecular disulfide bonds, which induce cross-linking between molecules.
Wan et al. [99] demonstrated that the action of radicals plays a crucial role in trypsin performance.A key sign of trypsin molecular structure damage by ultrasound was the presence of at least 10 types of molecular fragments of trypsin in the sample solution.
In a recent study [13], we explored the effects of US on lysozyme protein by using 20 kHz ultrasound.We observed negligible effects on the protein's primary structure, suggesting that the number of radicals produced under these conditions did not induce any significant structural or behavioral changes in the protein monomers.A similar observation was reported on ovalbumin by Gedanken et al. [80].
In a study conducted by Ashokkumar et al. [70], the impact of both low-frequency (20 kHz) and mediumfrequency (414 kHz) ultrasound on the amino acid content and the secondary structural integrity of dairy proteins (α-s1/α-s2/β/κ-Caseins, BSA, α/β-lactoglobulins, lactoferrin) was explored.It was found that both treatment conditions did not influence the protein's primary structure, even under prolonged and extreme processing conditions (6 h at 355 kHz).Further studies [77] clarified that low-frequency (20 kHz) ultrasonication of α/βlactoglobulins and their mixtures increased the reactive thiol content in β-lactoglobulin.
In another study by Wang, Cao et al. [88], the effect of ultrasound (20 kHz) on actomyosin was examined.Compared to untreated samples, the reactive sulfhydryl (SH) content of US-treated samples significantly increased at 100 and 150 W for 20 min.These observations indicate that several processes may occur during the sonication: (a) exposure of buried SH groups as a result of stretching and the subsequent unfolding of proteins and (b) cleavage of disulfide bonds, reaching a maximum of 150 W. At 200 W, the observed significant decrease in the total and reactive SH content of the protein occurred as a result of oxidation, as well as due to the formation of new S-S bonds or disulfide exchange.
Overall, these studies confirmed the significant role of free radicals formed during ultrasonic irradiation of protein-containing aqueous solutions at different frequencies and amplitudes.It has been shown that aromatic amino acid residues are generally the first to undergo radical attack, followed by their oxidation [100].Other amino acids are also susceptible to various forms of reactive oxygen species.Research works show that one of the main routes of ROS-mediated oxidation of amino acid residues involves Á OH-mediated abstraction of a hydrogen atom from the α-carbon of amino acids, as well as from the carbons of the protein polypeptide backbone and the aliphatic side chains of hydrophobic amino acids, which are considered the primary and most vulnerable reactive sites of attack by radicals.Subsequently, the mechanism underlying ROS-mediated protein oxidation involves the rapid conversion of a carbon-centered radical into a peroxyl radical, followed by further transformation into an alkyl peroxide by reacting with the protonated form of the superoxide radical.
The oxidative effect of Á HOO radicals is associated with the formation of alkoxy radicals, followed by their transformation into hydroxy derivatives [68].The effect of free radicals on protein and their subsequent modification is more pronounced and facilitated even when the protein is characterized by a low structural order and/or has a low molecular weight that is small enough.
The action of US also includes the oxidation/reduction of sulfhydryl groups that either form or disrupt disulfide, which may have the same action on protein supramolecular aggregates.However, note that radical action on proteins is very specific and strongly depends on the sonication parameters, and often, due to the specificity of US treatment, no significant changes in the primary protein structure may occur.

Effects of US on the secondary and tertiary structures: changes in the protein folding state
The secondary structure of a protein, defined by αhelix, β-sheet, β-turn, and random coils (RC), is largely affected by sonication.The secondary structure of a protein is determined by correlating the proportions of the α/β/RC content and their ratios.The main methods for studying changes in the secondary structure are Fourier transform infrared spectroscopy (FTIR) and circular dichroism (CD) spectroscopy.
The influence of ultrasound on the tertiary structure consists of changes in the spatial structure of the protein, entailing corresponding changes in the functional properties, which have been intensively studied [44,45].The effects of US on the tertiary structure also result in changes in protein hydrophobicity and in the number of free sulfhydryl groups found on the protein surface.Changes in the tertiary structure of the proteins are usually monitored by zeta-potential, fluorescence spectroscopy (intrinsic fluorescence of aromatic amino acids or while binding with fluorescent markers), and the free sulfhydryl content in proteins determined using Ellman's reagent.The surface hydrophobicity is determined using various fluorescent probes such as ANS (8-Amino-1-naphthalene sulfonic acid), which is used in protein folding studies [101,102].
Castellano et al. [68] studied the effect of highfrequency US (1 MHz) on six proteins: α-dominant, cytochrome c (cyt c), lysozyme (lys), myoglobin, bovine serum albumin (BSA) and β-dominant, trypsinogen, and α-chymotrypsinogen A. They showed, using CD and FTIR spectroscopy analysis, that under 1 MHz US conditions, with varying treatment times, the secondary structure of cyt c undergoes changes that predominantly lead to an increase in the random coil content, along with a decrease in α-helixes.The secondary structure of lysozyme was significantly changed with treatment time variation, and it was accompanied by changes in all secondary structural units.In contrast, such ultrasound treatment seems to induce small changes in the secondary structure of BSA, and it has a negligible effect on the secondary structure of myoglobin.Castellano et al. also found that such US treatment has very weak effects on the secondary structure of trypsinogen and α-chymotrypsinogen A. Therefore, by considering proteins that have α-helix as the main structural motif, it was found that variations in the secondary structure are smaller in proteins with a higher percentage of the main structure.Hence, the higher order of a protein structure may serve as a 'screen' for ultrasound waves [68], that is, β-sheets generally appeared more stable than α-helices and sound waves may 'reflect' from them.
Sonication of lysozyme at 20 kHz, which our group previously reported [13], showed that the content of random coils and α-helices decreased with increasing power and delivered energy, whereas there was a concomitant increase in the content of aggregated β-sheets in protein monomers.At maximum delivered energy values (1000 J), thermal denaturation/hydrolysis of protein monomers was observed.
Tao et al. [71] found that 20 kHz US sonication of myofibrillar proteins led to a slight increase in the αhelical content.They also proposed that the oxidationinduced aggregated β-sheet motifs could be destroyed due to the cavitation effect and that the protein conformation would be partially restored by increasing the content of α-helix.Using intrinsic protein fluorescence originating from aromatic amino acids, they also found that US slightly changes the tertiary structure of a protein, which was manifested by the fluorescence quenching of the characteristic peak of tyrosine, tryptophan, and phenylalanine, confirming that the dislocation events are due to a more polar microenvironment.In another work [72] performed in slightly different conditions of protein preparation, Liu et al. found that similar US treatment leads to a gradual decrease in α-helixes and β-turns, along with a parallel increase in the number of β-sheets and random coils.Similar effects on the secondary structure were shown earlier by Meiering et al. [69] for α-rich proteins: lys, BSA, and myoglobin in HEPES buffer at pH 7.8.Interestingly, the same treatment parameters (20 kHz of US) did not affect the structure of β-rich proteins such as SOD, Tm0979, and Hisactophilin.Therefore, these results are consistent with reports by Castellano et al. [68] and indicate that β-rich proteins are less sensitive to US due to the 'shielding' of US pressure waves.
Huang et al. [84] demonstrated that sonication of soybean protein isolate consisting of 7S (β-conglycinin) and 11S (glycinin) globulins (β-rich proteins) results in no significant difference in α-helix content before and after US treatment, whereas a slight decrease in βsheet content and an increase in random coil content were detected.Thus, US could improve the unfolding of the molecular structure, destroying the hydrophobic interactions.
Sonication of Aβ(1-40) [73] at 50 kHz in perdeuterated 2,2,2-trifluoroethanol in which the peptide is present in the solution as a stable monomer with α-helical secondary structural motifs revealed that these motifs were not significantly affected by sonication.
Ashokkumar et al. [70] showed that low-frequency (20 kHz) ultrasound treatment of hydrolyzed sonicated skim milk proteins (lactoglobulins, caseins, see Table 1) increased β-sheet content with a predominance of antiparallel structures accompanied by a protein surface hydrophobicity increase [76,77], causing unfolding and partial denaturation of native protein folding.
After treatment with 20 kHz of US, silk fibroin (SF) protein formed β-sheets upon increasing the sonication time.Moreover, the molar weight of silk fibroin protein was significantly reduced during the sonication as a result of disrupting the heavy chain, whereas the light chain fraction remained unaffected [78].In their studies, Kashirskii et al. [79] formed films from fibroin solutions and reported that after treating protein solutions with ultrasound, the films contained a larger fraction of β-structures and a smaller fraction of αhelices compared with films obtained without ultrasonic treatment.The occurrence of conformational changes in the protein was further confirmed by X-ray diffraction patterns of silk films.The X-ray diffraction patterns of the silk films also confirm the occurrence of conformational changes in fibroin upon ultrasonic treatment.
Yu et al. [85], also reported that sonication of amylase, papain, and pepsin at 40 kHz resulted in changes in the secondary structure.They emphasized that more intense ultrasonic treatment leads to destabilization of van der Waals interactions and the disruption of the hydrogen bonds in polypeptide chains, which ultimately leads to modification of the secondary structure of the entire protein [7].In addition, due to strong shears and microflows resulting from the collapse of bubbles during US irradiation, disturbances in the tertiary structure are observed, followed by partial denaturation of enzymes [103].
Xu et al. [87] have studied the effect of US on ovomucin at different exposure times.They found that during the first 30 min, there was increase in α-helix content, which drops at the 40-min time point.The same trend was observed for RC.The percentage of βturn increased with the treatment time; it reached a maximum after 20 min of sonication and then dropped; however, it was always higher than the untreated sample.The percentage of β-sheet decreased and had no appreciable dependence on the treatment time.They also found that the protein molecule tended to unfold during treatment and that its surface hydrophobicity increased.
Cao et al. [88] demonstrated that actomyosin displayed a significant decrease in α-helix, an increase in β-sheet content with increasing sonication power, and at high US power, β-turns may completely disappear.Such changes also lead to significant surface hydrophobicity changes in the protein, with a maximum at 150 W, followed by a decrease at 200 W due to denaturation of the protein.
Wang et al. [86] studied how variations of low intensity (5-30 min) and low frequency (18-29 kHz) ultrasound affected the cellulase structure and the enzymatic activity.They found that US at 5-20 W with the frequency of 18-26 kHz (5-10 min of treatment) led to a reduced percentage of α-helices and βsheets but that it increased the content of β-turn and RC.In contrast, the maximum treatment time (30 min) with the highest frequency (29 kHz) and US power (50 W) led to an increase in the α-helix and βturn content, whereas the percentage of RC was reduced.Surprisingly, no significant action on β-sheet content was observed.It was also found that US may have destroyed the hydrophobic interactions in protein molecules, causing more hydrophobic groups and regions inside the molecules to be exposed.These effects eventually reduced the enzymatic activity of the protein.Researchers have used proteins that have good solubility in aqueous solutions in all the previously mentioned cases.
In contrast, De Leo et al. [89] focused on the US treatment of the reaction center protein from Rhodobacter sphaeroides, a membrane protein with low solubility in aqueous environments.They studied ultrasound-induced protein denaturation and also assessed the role of surfactants in US-driven protein perturbations in different media.Three different models were used: (a) LDAO micelles, (b) a liposomal system with three different lipid compositions mimicking the native membrane, and finally (c) a surfactantfree buffer as the control.They found that in all systems, under the US treatment, membrane proteins denature and lose their activity due to significant disturbances in their secondary structure.
Herrmann et al. [104] showed that if certain proteins are genetically modified with long polypeptide chains having an increased charge, then such systems become sensitive to US.They then applied this method in two aspects: (a) to develop derivatives of GFP whose fluorescence can be 'switched off' by US and (b) for UScontrolled regulation of catalytic activity using trypsin in combination with a protein inhibitor (Fig. 2A).
Liang et al. [105] used US treatment to transform and stabilize conformation around the active site of HRP.First, they ultrasonicated the protein and transferred the active center of the enzyme from a closed to an open state.In the second step, a synthetic MOF was used to quickly 'lock in' the enzyme's new conformation.This approach ensured that the enzyme not only retained but also enhanced its enzymatic activity.They also observed that without the MOF, the enzyme rapidly denatured after ultrasonication.Subsequent MD simulations demonstrated that the enhanced activity of the HRP enzyme is due to a specific open-site conformation that promotes more efficient substrate binding to the enzyme (see Fig. 2B,C).Spectroscopic analysis of proteins also corroborates earlier conclusions indicating that the degree of the influence of therapeutic ultrasound on the protein structure depends on a number of parameters strictly related to their structural organization.To modify or even destroy a protein using ultrasound, it is crucial to consider its structural organization, size, and molecular weight [68].
Overall, ultrasonication causes significant changes in the secondary structure of almost all proteins, causing changes in the tertiary structure, which sometimes can even lead to protein denaturation.It is also worth noting that the effects of protein unfolding and structural rearrangement depend on many factors and sometimes they can even work in opposition in the same protein.However, gentle US treatment can be used for protein activation to open specific binding sites that might be further used in different applications.

Effects of US on the structure of protein complexes: Supra-molecular assembly/ aggregation and disassembly using US energy
Structural changes occurring in the primary, secondary, and tertiary structures of proteins, accompanied by cross-linking, additional hydrophobic, and other intermolecular interactions, may lead to the formation of large protein complexes via protein aggregation or self-assembly.US can also lead to the decomposition of the pre-formed protein structures via cavitation and other physical effects.The assembly/disassembly of protein complexes is usually accompanied by changes in their solubility, phase composition, and solution turbidity [66].The effects of US on such complexes have been extensively studied by TEM, SEM, AFM techniques, optical and fluorescent microscopies, and their modifications.The size of the complexes is usually determined by DLS.The atomic structure of the ensembles is usually studied using XRD, SAXS, WAXS, and electron diffraction.
Castellano et al. [68] observed that 1 MHz US treatment of trypsinogen for 30 min, along with αchymotrypsinogen A, produced aggregates of different sizes, which is indicative of protein self-assembly.
Sonication of lys monomers with 20 kHz US caused protein to be converted into particle-like supramolecular assemblies whose size increases with the delivered energy and may reach 1100 nm [13].Generally, exposing monomeric lysozyme protein to ultrasound leads to the formation of two types of aggregates: nonamyloidogenic assemblies and other complexes with a negligible percentage of β-sheets that deviate from the classical amyloid state.When aggregated lysozyme nanofibrils were exposed to sound energy, the formation of smaller fragments contaminated with a small percentage of amorphous aggregates was observed, consistent with previous literature reports.The length of the fragments decreased proportionally as the energy applied increased.Notably, this direct correlation between fibril length shortening and an increase in acoustic energy halts when reaching a threshold of 30-60 nm, beyond which the fibrils resist further fragmentation [13] (Fig. 3A-D).
According to Tao et al. [71], treating myofibrillar protein fibrils with 20 kHz US leads to a decrease in particle size, which has been measured by DLS; US also breaks the protein fibrils.Additionally, it was found that the zeta potential of protein decreased during US action, confirming the change in the polar environment of the protein and indicating that the reduced positive charge favors the formation of protein aggregates.Furthermore, aggregation of positively charged MP might reveal the inner polar groups on the surface, thus enhancing the overall net charge presence.Oxidation of SH groups may also be attributed to protein aggregation.Similar effects were observed by Chen, Kong et al. [72]: when the US power was increased to 450 W, the filamentous structure of the protein was partially destroyed, and when the power was increased even more up to 600 W, there was a tendency of the protein to aggregate due to significant protein unfolding with the subsequent exposure of hydrophobic groups on the surface of the protein.
According to Huang et al. [84], US also acts as a disaggregating force with respect to the 7S (β-conglycinin) and 11S (glycinin) globulins contained in soybean isolate.The same was reported for ovomucin [87] and actomyosin [88].
Filippov et al. [73] studied the aggregation of Aβ(1-40) at 50 kHz with US treatment vs. sonication time and found that this protein does not form any aggregates in a non-sonicated sample.However, when treated, it starts to form aggregates.For treatment times up to 5 min, the percentage of aggregated structures reaches 40%, whereas longer treatment times exceeding 7 min result in a decrease in the apparent percentage of protein aggregates to approximately 25%.Moreover, it was found that after US treatment, aggregate particles do not tend to form β-sheet structures typical of amyloid fibrils.A comparative analysis of fibrils formed during the standard seeding reaction and when formed during ultrasonic treatment showed that the latter fibrils were thinner (about a quarter of the diameter) and did not have longitudinal periodicity.
The effect of US on the critical fibril size was also shown earlier; Goto et al. [74] revealed that repetitive US pulses of 1 min, followed by an incubation period of 9 min, led to β2-microglobulin fibrillation after a lag time of several hours, followed by the breakdown of the pre-formed fibrils.This was further confirmed [75], and US can induce a monomeric solution of β2microglobulin to form amyloid fibrils.However, further treatment with US can break down pre-formed fibrils into shorter fibrils.MD simulations explained this phenomenon later [107].When an amyloid fibril exceeds a certain critical length, its hydrophobic residues can serve as a nucleus for bubble formation, leading to bubble collapse on the fibril surface.Conversely, if the amyloid fibril is not long enough, these hydrophobic residues are unable to initiate bubble formation, and the amyloid fibrils remain undisturbed.This phenomenon explains why after ultrasonication amyloid fibrils tend to have a consistent length.
Additionally, Goto et al. [42] showed that ultrasonication is an effective tool for accelerating nucleation, leading to fibril formation.They also found that there is a critical concentration of the Aβ, after which the ultrasonic pulses accelerate spontaneous fibrillation (above 1 μM); however, ultrasonic pulses accelerate the depolymerization of fibrils into monomers at 1 μM.
Therefore, it was confirmed that US has two contrasting effects on amyloid fibrillation; first, it promotes the formation of nuclei in an oversaturated solution of monomeric proteins, thereby accelerating spontaneous nucleation; second, it leads to the fragmentation of fibrils.Combining these two effects results in a balanced interaction between fibril formation and disintegration.Consequently, the broken fibrils produced accumulate in homogeneous and short amyloid structures.Moreover, it was demonstrated [108] that it is possible to directly observe fibrils of minimal size and of low polydispersity under certain conditions, which is achieved by the free energy minimum under competition between ultrasonication-induced fibril production and breakdown [75].
In investigating the effect of US on silk fibroin (SF) protein, Kaplan et al. [78] found that this protein formed stable gels after US treatment at 20 kHz.The formation of gel structures can be explained by interprotein disulfide bond formation from cysteine residue oxidation due to the presence of free radicals that further cross-link the proteins, resulting in increased βsheet formation and gelation.However, after the treatment time was increased, the gel was converted into solution due to the collapse of established disulfide bonds, hydrophobic inter-and intra-chain interactions, and bonds forming β-sheets.
Upon US treatment at 17-20 kHz, transthyretin may form stable amyloid-like fibers [82] accompanied by quaternary structure disintegration; the number of tetrameric and dimeric forms of the protein decreased.Kuwata et al. showed that mPrP  can form two types of aggregates: stiff fibrils and shorter fibrils accompanied by disordered aggregates, depending on the intensity of ultrasonic energy [83].Generally, the results indicate that amyloid fibrillation is similar to the crystallization of solutes from a supersaturated solution [109].
Ashokkumar et al. [76] treated β-LG with 20 kHz US, which led to the formation of β-lg amyloid crystals rather than amyloid fibrils.Importantly, they showed that an increase in sonication time (0-60 min) and input power (4-24 WÁcm À3 ) increased the mean crystal length.They also proposed a scheme and a mechanism of ultrasonic protein aggregation in terms of amyloid crystal formation by low-frequency ultrasound.First, the localized shear stress and hightemperature cause protein denaturation.This leads to exposing the hydrophobic regions that enable unfolded proteins to accumulate at the cavitation bubble interface.Owing to protein-protein interactions that take place at the interface, oligomers are formed.Furthermore, oligomers can detach from the interface and accelerate the growth of protein crystals.During adiabatic bubble collapse, which accumulated at the interface, proteins can collide with sufficient energy during the implosion process for catalytic events to surpass the required energy barriers and condense into ordered aggregates.
Zhu et al. [106] studied the morphological characteristics of collagen upon US treatment (see Fig. 3E-G).After exposure to ultrasound above 13.94WÁcm À2 for 10 min in 1 mM acetic acid, the d-period structure of the collagen fiber was fuzzy, demonstrating that the ultrasonic activity can induce apparent changes in the structure of the collagen fibers.Chang et al. [110] also reported that cavitation of ultrasound affected the cross-linking and stability of lysylpyridinoline and hydroxylysylpyridinoline in collagen fibrils, which may induce their structural instability.Although ultrasound can damage collagen fibers via acoustic cavitation, the repetitive gap/overlap pattern in collagen remains stable after ultrasound exposure due to the hierarchical organization of the collagen tissue.
US protects surfactant-free reaction center membrane proteins from Rhodobacter sphaeroides by tending to form large aggregates during US treatment, probably resulting in a structure that protects them from further denaturation [89].
Overall, the structure-modifying effect on preformed protein fibril break-up is due to the shock waves produced during the implosion of cavitation bubbles.This event, coupled with the emission of thermal energy, predominantly affects the structure of the protein monomers or the mechanical force that contributes to the fragmentation of the fibrillar structures.On the other hand, often proteins undergo selfassembly and, depending on the US parameters, they can control the type, size, and number of these protein complexes.However, these aggregate structures and their polymorphism are still under discussion.

Brief remarks about the effects of US on protein structure and the assembly/disassembly mechanisms revealed by MD simulations
Experimental data on fibril disruption by US cavitation bubbles were confirmed by molecular dynamic simulation [107,111].Under negative pressure, a bubble typically forms at the hydrophobic residues of a fibril within the transmembrane region and most β-strands retain their secondary structures.However, as pressure turns positive, during the compression process, the cavitation bubbles collapse, directing the surrounding water molecules toward the hydrophilic residues outside the transmembrane region, resulting in further fibril breakup.Smaller amyloids require extended US treatment for breakup since fewer hydrophobic areas initiate bubble creation (Fig. 4A-D).
The effect of US on the critical fibril size was determined by Goto et al. [74,75] and the effect on amyloid fibrils, followed by their breakage into shorter fibrils, was further confirmed and explained by the MD simulations mentioned earlier [107].
Deriu et al. [112] utilized MD to explore the structural changes induced by stable ultrasonic cavitation on S-shaped Aβ1-42 amyloid fibrils; this form is considered as more stable than the U-shaped Aβ1-42 counterpart.The research indicated that US treatment might influence the folding dynamics and kinetics of the S-shaped aggregates, showing a significant relationship with fibril polymorphism.Thus, one of the assumptions explaining the destabilization of amyloid may be (a) the influence of amino acid residues that do not participate in forming the key secondary structures of protein fragments, or (b) the influence of weakly structured amyloid regions, or both, which act as sources of instability for the entire fibril, which, in turn, leads to the formation of nanofractures that can propagate throughout the structure until the molecular assembly is partially or completely unfolded (see Fig. 4E,F).Furthermore, in our recent work about the formation of amyloid fibrils with US, by using MD, we found that at a low amplitude of 20 kHz frequency sonication, the fold modification effect of US is imposed mostly on the helical motifs of lysozyme protein [14].
Overall, the MD method is a powerful tool for studying and predicting the behavior of proteins under US action.It may determine what happens to proteins and their complexes under US action (controlled simulated conditions).MD simulations help to confirm experimental data and expand molecular-scale information on structural transitions in the proteins.However, due to the complexity of the effects that US exerts on aqueous media and the dynamic nature of protein complexes, MD's ability to simulate the full spectrum of US effects on proteins, and concomitantly to decipher the consequent changes in protein fold, are limited.These limitations arise from technological difficulties mediating between direct US effects on protein species and between US-derived changes in solvent in which proteins are dispersed.

Brief remarks on the combined effects of US on protein structure and their assembly/ disassembly
Since directly deciphering complex US effects, including shear, streaming, cavitation, temperature, and pressure (see Fig. 1A), as well as the impact of the airliquid interface on the proteins' structural characteristics, is a non-trivial task, researchers have tried to better understand and study the contribution of each factor separately.
In this regard, Silva et al. [113], in attempting to unravel the effect of an extremely high-pressure regime on protein fibrils, showed that the formed fibrils of transthyretin and α-synuclein are capable of reversible disassembly in a series of compression-decompression cycles.This phenomenon is consistent with the MD simulation results discussed earlier and it indicates the presence of defects in the fibrils.This explains why amyloid fibrils are exposed to high hydrostatic pressure.Winter et al. [114] showed that under high hydrostatic pressure (1500 bar), insulin forms amyloid structures with a unique circular morphology.However, they also pointed out that increased hydrostatic pressure either retards or disables folding kinetics compared with standard pressure, or it may induce alternative non-native conformations in proteins or lead to the dissociation of once-formed protein aggregates, depending on the polypeptide.
Jungbauer et al. [115] attempted to create experimental conditions that simulate the effect of cavitation, along with the effect of the air/liquid interface, in order to determine the mechanism underlying protein aggregation during cavitation.In their study, they tested nine functional proteins (α-lactalbumin, two antibodies, fibroblast growth factor 2, GCSF, GFP, hemoglobin, HSA, and lys).
The choice of such a set of proteins was mainly due to their natural tendency to aggregate, sometimes also spontaneously.Jungbauer et al. compared the results of the behavior of proteins during cavitation with the results of protein aggregation during foaming through a micro-hole.Both results were compared with protein aggregation under the influence of shear stress only in the absence of separation at the interface.
Importantly, they showed that among all the proteins tested, GCSF, HSA, and hemoglobin were the proteins most sensitive to aggregation in both investigated systems, having an even more pronounced aggregation effect at foaming than with cavitation.The remaining six proteins under the experimental conditions did not show any tendency to aggregate; therefore, the behavior of proteins during foaming is similar to the effect of cavitation in terms of the interaction of proteins with the air-liquid interface.It was also found that in the absence of phase separation at the interface under the action of only shear rates up to 10 8 s À1 , no signs of aggregation of any of the proteins were observed.Thus, it was concluded that the protein aggregation associated with cavitation is caused by interaction at air/liquid interfaces and not by the action of radicals, as reported in some sources.
In addition, it was reported in the same study that the closer the pH of solutions of specific proteins is to their isoelectric point, the greater their tendency to aggregate when exposed to cavitation.However, in experiments when surfactants were added to protein solutions, it was found that protein aggregation could be almost completely suppressed.
In addition, the effect of the protein concentration in solution on their aggregation was investigated; it was found that an increase in the concentration of proteins can also suppress aggregation during cavitation, which is explained by the rapid saturation of the surface of the vapor cavities with protein.
In their study, Jungbauer et al. cite Thomas and Dunnill work [116], who also found no evidence of shear-induced aggregation of catalase and urease, even at rates up to 10 6 s À1 .Horse cytochrome C [117] and antibodies [118] also showed no signs of aggregation at shear rates of 2 × 10 5 s À1 and 2.5 × 10 5 s À1 , respectively.In contrast, BSA, β2-microglobulin, and GCSF were prone to aggregation induced by extensional flow [119].Thus, the high shear rates associated with cavitation were insufficient for the aggregation of mediumsized proteins.However, the increase in the air/liquid interface, associated with the growth of bubbles in the cavity, causes protein aggregation.However, debate continues as to whether elongation forces, in addition to shear rate, can unfold proteins.
In studying the action of US (100-500 W for 1-20 min) on trypsin, Wan et al. [99] established that the protein desaturates under these conditions due to partially exposing the molecule at the water-air interface, confirming the significant role of the cavitationdriven effect following protein assembly/disassembly.
Our recent study [13] aimed to determine whether the changes in the secondary structure content and the formation of protein complexes originate from the mechanical perturbation (which ideally is aimed to mimic the streaming effects of US) or from the temperature gradient (which may stem from cavitation bubble collapse) created by the US.To this end, we compared the contribution of each effect separately in the absence of US.We observed that mechanical mixing did not cause any noticeable alterations in the protein monomers' structure.In contrast, when subjected to heat, protein monomers underwent thermal denaturation.Both mechanical mixing and heat, applied separately or simultaneously, showed no significant structural modifications in the fibrillar assemblies.These results indicate the high sensitivity of soluble monomeric protein molecules to temperature, which induces the protein to adopt a fibrillar structure, enriched with β-sheet structures.
Despite that the thermal effect on proteins has been extensively studied and is well known to gain a deeper understanding of the unfolding and the aggregation mechanism of insulin at different ionic strengths, Smirnovas and Winter [120] carried out a thermodynamic study of its aggregation under various salt shielding conditions.Since the temperature rises in the absence of salts, insulin begins to destabilize and unfold.A further increase in temperature contributes to the domination of random coil-rich structures in solution, which subsequently contributes to the formation of compact nuclei followed by their growth into fibrils.
In the presence of 0.1 M NaCl and increasing the temperature to 55-60 °C, a partial unfolding of the protein occurs with the formation of oligomeric nonfibrillar aggregates or oligomers.A further increase in temperature leads to the reorganization of aggregates and the formation of more compact nuclei, which triggers their elongation.Subsequently, the maturation and formation of more compact nuclei from amorphous oligomers occurs, and consequently, their number increases, which in turn, leads to the formation of more complex morphological aggregates such as floccules or tangled networks.
The presence of the hot spots, formed upon cavitation bubble collapse and so-called microjets, has traditionally been associated with an overall temperature increase in the reaction cells, which, in turn, causes the irreversible denaturation of protein.However, several studies have demonstrated that by controlling the heating rates, it is possible to manipulate the assembly/ disassembly process in proteins.For example, the denaturation of collagen fibers takes place at a temperature range between 53 and 63 °C [106].The denaturation process involves the breakage of the hydrogen bonds, loosening the fibrillar structure, and contraction of the collagen molecules.Thus, operating at the lower level of this temperature range (at 53 °C and below) can induce a structural re-arrangement without causing complete protein denaturation.Lysozyme protein is in a micro-compartmentalized state and undergoes fibrillation at 65 °C [121], similar to insulin.Park et al. [122] demonstrated the ability of insulin to form amyloid fibrils above 100 °C.The morphology of insulin amyloid fibrils underwent sharp changes with increasing temperature.
Overall, in our view, the initial protein fold is the key factor defining further responsivity of the protein structure and behavior specifically regarding the effect of US.Thus, most of the stable structured proteins, especially those proteins containing a high percentage of β-sheet conformations, would be less sensitive to shear forces and more sensitive to thermal fluctuations, whereas unstructured proteins (with a dominant random coil secondary structure) are expected to exhibit a high sensitivity to the acting shear forces [123].

Enzymatic activity
In the previous parts, we summarized that during ultrasound treatment, the structure of proteins can significantly change.These changes, in turn, can affect the functional properties of proteins, including, in particular, their enzymatic activity.Most of the US effects occur primarily as a result of the collapse of cavitation bubbles.The conditions created during this process (discussed previously) can accelerate the transport of substrates and reaction products to and from enzymes, improve mass transfer in enzyme reactor systems, and thus increase the efficiency of enzymatic reactions.However, high-velocity flow, which can be accompanied by the formation of ROS and heat, can also potentially negatively affect the stability of the biocatalysts.Therefore, the activation and inactivation of enzymes depend on either suitable or inappropriate conditions, respectively, when using ultrasound [39].
Recently, it was demonstrated that under mild US conditions, enzyme activity might be enhanced.In this respect, Avivi and Gedanken reported that α-amylase microspheres synthesized using high-intensity US exhibited excellent enzymatic activity compared with native α-amylase, which is considered as crucial enzyme for industrial applications possessing high thermal stability.They characterized it and conducted catalytic experiments of the sonochemically produced microspheres [124].The enzymatic activity of the amylase microspheres was 27% higher than that of the native protein after a short reaction time (3 min).In comparison, over a longer reaction time (1 h), it reached 56% of the activity of the native protein [124].
Another study by Zhao et al. [125] showed an intensification of the polymerization reaction with lipase.Lipases are well-known enzymes that can catalyze esterification and transesterification reactions in aqueous and organic solvents, ionic liquids, or in solventfree conditions.They carried out a reaction kinetics study using different concentrations of lipase to determine the role of ultrasonication during the reaction [125].The data obtained indicate that the response is enhanced by ultrasound and that the enhancement effect is more significant when a lower concentration of lipase is used.A study of the kinetics of the reaction to produce poly (ethylene glutarate) found that ultrasound appreciably intensifies condensation polymerization.When the conventional water bath treatment was replaced by sonication followed by 6 h in a vacuum bath, the reaction time was reduced from 24 to 7 h, which had a positive effect on the enzymes, resulting in a higher degree of polymerization and a better monomer conversion rate in a shorter time [125].
Overall, enhancing or inhibiting the activity of a specific enzyme will depend on the US operating system, including the enzyme's nature, the intensity and duration of the ultrasound treatment, and the specific experimental conditions.

Effect of US on protein fibrillation kinetics
Ultrasonication is employed in amyloid studies due to its ability to induce protein fibril formation.In our recently reported study [13], we thoroughly investigated the impact of ultrasound on both monomeric and fibrillar lysozyme.Importantly, we found that low-amplitude ultrasound treatment of monomeric lysozyme modified the protein fold and further accelerated its aggregation.In contrast, higher amplitude resulted in detrimental effects and decreased the rate of fibril formation (Fig. 5A).Aggressive US parameters were reported to initiate the formation of amorphous aggregates that do not serve as a nucleation point.Exposing fibrillar species to US leads to fragmentation into smaller species, accelerating the growth phase of amyloid fibrils, with no lag phase observed (Fig. 5B).The changes in the chemical kinetics of amyloid growth owing to US exposure have been examined using the ThT assay (a standard assay for studying the progress and kinetics of protein amyloid aggregation), and the aggregation rates in response to US exposure have been further analyzed by assessing the number of fibrils' ends.The growth phase linearly depends on the number of free ends of ultrasonically obtained fragments (Fig. 5C).
Goto et al. revealed that ultrasonication has the potential to initiate spontaneous fibril formation from a β2-microglobulin monomer solution in the absence of seeds.The kinetics measured, based on light scattering and ThT fluorescence, were in agreement with each other.The results indicated a slight variation in the lag time in ultrasonication-induced fibril formation and the concentration-dependent lag time; the higher the concentration, the shorter the lag time.Interestingly, ThT fluorescence was more than 2-fold stronger in the sonication-induced fibrils than in the standard fibrils.These results suggest that the fibrils have more ThTbinding sites, namely, more β-sheet content [74].
In another work, Goto et al. described how ultrasonication substantially enhances the fibril formation of β2-microglobulin, particularly in dilute monomer solutions, leading to the generation of short, dispersed fibrils.To illustrate the phase diagram of the aggregation reaction of amyloidogenic proteins, they introduced a half-time heat map concept.Their findings demonstrated that the energy landscape of an aggregation reaction during US treatment significantly changes due to the impact of cavitation caused by US in comparison with shaking [126] (see Fig. 5D-F).
A recent study by Pathak et al. [76], using lowfrequency US, showed β-LG amyloid crystal formation.The ThT fluorescence assay exhibited a lag time, followed by increased fluorescence intensity after 45 min, and the increase of ThT fluorescence after this period was in a good correlation with the average fibril length increase.This observation suggests a buildup of β-stacking due to the ultrasonic treatment.
Hu and Li [127] reported that ultrasound (20 kHz) accelerates fibrillation by promoting protein denaturation and unfolding.They also showed that the surface hydrophobicity of β-LG increases immediately after ultrasonication, followed by an increase in ThT binding fluorescence intensity, indicating a rise in particle size.So et al. [128] hypothesized that the principal limitation to fibril formation was the substantial free-energy barrier, which ultrasound effectively diminishes.Notably, even under neutral pH conditions, the appearance of fibrils was observed after a lag time of just 1.5 h, compared with the 2-day lag time when shaking alone (mimicking the mechanical forces) was applied.
Nakajima et al. [129] developed an optimized sonoreactor to analyze amyloid fibrils and found that transient cavitation was crucial to accelerate primary nucleation.They found that homogeneous nucleation requires a higher free energy barrier than does heterogeneous nucleation caused by cavitation bubbles.By controlling the amplitude and frequency of US treatment, they showed that β2-microglobulin forms fibrils and that the lag time is reduced from 25 to 8 h due to the acceleration effect caused by sonication.
Overall, the ability to control protein aggregation through ultrasound offers promising applications in biotechnology and medicine.The effects of ultrasound on protein aggregation are complex and can vary, depending on the exact conditions used (such as the frequency and the power of the ultrasound, the exposure duration, and the characteristics of the protein itself).On the one hand, US can impede the overall protein aggregation rate by breaking and denaturing part of the monomeric species.On the other hand, ultrasound can also accelerate protein aggregation; sometimes the shock waves from the collapsing bubbles can induce structural changes to the proteins, making them more likely to interact and aggregate.It is important to note that although these mechanisms have been proposed to explain the effects of ultrasound on protein aggregation, they are not yet fully understood.

Conclusions and future perspectives
Thus, ultrasonic treatment affects the structure of proteins; however, these effects are highly dependent on many parameters directly related to ultrasound: the structural order of proteins and the environment in which the treatment is performed.These ultrasound effects are manifested by changes in the protein's structural hierarchical organization.We believe that further research in this area should address questions related to the direct (in situ) effect of ultrasound on protein molecules.The changes caused by ultrasound ultimately lead to a change in the functions of proteins; this is manifested in the effect on their enzymatic activity, as well as in changes in their aggregation properties.In addition, experiments that simulate the behavior of a protein, as well as the effects of selfassembly or disassembly under conditions simulating cavitation, for example, in a laboratory-on-a-chip, are of particular interest.Since molecular dynamics methods have proven themselves well in modeling the behavior of proteins during cavitation and have confirmed several experimental effects, it would be interesting to expand such studies by further specifying the behavior of proteins during sonication and by considering the rapidly developing machine learning methods and the further accumulation of experimental data, which will allow a more accurate prediction of these effects.

3015FEBS
Letters 597 (2023) 3013-3037 ª 2023 The Authors.FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.

3016FEBS
Letters 597 (2023) 3013-3037 ª 2023 The Authors.FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.

Fig. 2 .
Fig. 2. Left panel (A): scheme of intra/intermolecular US-induced protein unfolding, reproduced from [104], published by John Wiley & Sons in 2021.Right panel (B, C): MD simulations on the conformational change of HRP after US treatment and its activity after MOF immobilization: (B) surface representations of HRP prior to and after US treatment; N is denoted in dark blue, C is denoted in light blue, and Fe is denoted in red; (C) the distance of F68 to F142 and F179 in the HRP structure during US pretreatment.Reprinted with permission from [105] Copyright 2022 American Chemical Society.

Fig. 4 .
Fig. 4. (A) Amyloid fibril model (PDB: 2BEG) from experimental conformation, (B) one of the conformations for the non-equilibrium MD simulations obtained after equilibration, (C) periodic pressure change as a function of time.(D) MD simulations of Aβ fibril breakup induced by cavitation, adapted with permission from [107] Copyright 2014 American Chemical Society.Snapshots reporting (E) PDB: 2MXU and (F) PDB: 5OQV structures at different times during the process that models ultrasonication, adapted with permission from [112] Copyright (2014) with permission from Elsevier.

Fig. 5 .
Fig. 5. Top panel (A-C): analysis of the chemical kinetics of lysozyme aggregation for (A) untreated and US-treated monomers and (B) the kinetics of aggregation initiated by fragments obtained under different US conditions, (C) the linear dependency of the fibrillation rate of lysozyme on the number of free ends obtained by US, adapted from [13], Copyright 2023 the Authors, published by PNAS.Bottom panel (D-F): (D) half-time (t half value) heat maps of the aggregation reactions under ultrasonication, (E) schematic illustration of fibril formation and fragmentation under ultrasonication compared to shaking, (F) the relationship between the seed concentration and the t half value under three different agitations: quiescence, shaking, and ultrasonication, adapted with permission from [126] Copyright 2021 American Chemical Society.

Table 1 .
Effect of US treatment parameters on the structure and assembly properties of various proteins.
FEBS Letters 597 (2023) 3013-3037 ª 2023 The Authors.FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.
FEBS Letters 597 (2023) 3013-3037 ª 2023 The Authors.FEBS Letters published by John Wiley & Sons Ltd on behalf of Federation of European Biochemical Societies.