Biomolecular Crowding Arising from Small Molecules, Molecular Constraints, Surface Packing, and Nano-Confinement

Abstract

The effect of macromolecular crowding on the structure, dynamics, and reactivity of biomolecules is well-established and the relevant research has been extensively reviewed. Herein, we focus our discussion on crowding effects arising from small co-solvent molecules and densely packed surface conditions. In addition, we highlight recent efforts that capitalize on the excluded volume effect for various tailored biochemical and biophysical applications. Specifically, we discuss how a targeted increase in local mass density can be exploited to gain insight into the folding dynamics of the protein of interest and how confinement via reverse micelles can be used to study a range of biophysical questions, from protein hydration dynamics to amyloid formation.

A large number of macromolecules and molecular complexes exist inside a living cell, leading to a very crowded environment.¹ Due to volume exclusion, the accessible free space and available water molecules become limited for all cellular components, which, in principle, can have a profound effect on the physical and chemical properties of the biomolecules confined inside the cell. To characterize the excluded volume effect on various biophysical and biochemical processes, in vitro studies have employed a variety of macromolecules, such as polymers (e.g., dextran and Ficoll) and large proteins (e.g., lysozyme and bovine serum albumin), to tune the degree of crowding of the system of interest. As expected, these studies revealed that macromolecular crowding can indeed affect, sometimes significantly, the thermodynamic and/or kinetic properties of the biological molecule(s) in question. Interested readers can find several up-to-date reviews on this very active research area.^2–5 What is surprising, however, is that a number of recent studies have indicated that the excluded volume effect can also arise from unexpected sources, such as trimethylamine N-oxide (TMAO) and trifluoroethanol (TFE). Findings from these studies suggest that molecular crowding arising from molecules that are a fraction of the size of macromolecular crowders may be more common than previously assumed and should be considered when interpreting experimental results. Below, we will first discuss this phenomenon in the context of biomolecular folding and then highlight recent advances in using specifically engineered molecular-level crowding and/or confinement to achieve various goals in biophysical chemistry.

Nano-Crowding Arising from Solvents

The phrase “nano-crowding” was first coined by Thirumalai, Straub, and coworkers.⁶ In an effort to understand the stabilizing mechanism of a commonly encountered osmolyte, trimethylamine N-oxide (TMAO), they conducted all-atom molecular dynamics (MD) simulations on a series of peptides in a 1 M TMAO solution. For short dipeptides, which were used as models of the unfolded state, they found that TMAO molecules preferentially form hydrogen bonds (H-bonds) with peptide backbone units and basic sidechains, thus limiting their conformational degrees of freedom. For longer polypeptides that are capable of forming intramolecular H-bonds, they found that addition of TMAO increased the population of α-helical conformations as the co-solvent molecules are depleted from the backbone. As shown (Figure 1), simulations of the Aβ_16–22 peptide in various concentrations of TMAO yielded similar results, as the peptide formed a more compact structure with increasing concentrations of co-solvent. These findings thus prompted the authors to conclude that TMAO can act as a nano-crowding agent by entropically destabilizing a peptide’s unfolded state while stabilizing the folded state via the excluded volume effect. A recent simulation study of Shea and coworkers⁷ on how osmolytes affect the structural changes and aggregation propensities of the amyloid protein tau also supports this mechanism.

Figure 1.

The probability distribution of the radius of gyration, P(R_g), of Aβ_16–22 in varying TMAO concentrations, as indicated. As the concentration of TMAO increased, the R_g decreased indicating that the peptide became more compact. Reprinted with permission from ref. 6. Copyright 2011 American Chemical Society.

While the proposed crowding role of TMAO is corroborated by its ability to promote compact protein structures,^8,9 a more direct experimental validation of this phenomenon came from a recent study by Ma et al.¹⁰ Specifically, they used two-dimensional infrared (2D IR) spectroscopy and the nitrile stretching vibration of p-cyanophenylalanine (Phe_CN) as a site-specific IR reporter¹¹ to directly assess the excluded volume effect of TMAO. Their experimental design was based on the notion that any entropic effect arising from volume exclusion of the co-solvent molecules will lead to an increase in the degree of conformational rigidity of the peptide and protein systems in question, thereby resulting in a longer spectral diffusion time of the IR reporter.¹² The latter is a measure of the dynamics that lead to fluctuations in the vibrational frequency of the IR probe. As indicated (Figure 2), their 2D IR results showed that in the presence of TMAO, the spectral diffusion kinetics of the nitrile probe became slower, as evidenced by an increase in the offset of the respective kinetic traces. Since this kinetic offset results from protein or peptide conformational dynamics that contribute to the nitrile frequency fluctuations but are too slow to be resolved in these experiments, their finding thus corroborates the idea that TMAO can influence protein conformational stability and dynamics via the excluded volume effect.

Figure 2.

Spectral diffusion dynamics of the nitrile stretching vibration of Phe_CN in HP35-P76 and HP35-P58 obtained under different solvent conditions, as measured by the center line slope (CLS) of the corresponding 2D IR spectrum as a function of the waiting time (T). Addition of TMAO leads to an increase in the vertical offset of the kinetics, indicating an increase in the percentage of conformational states that interconvert on a timescale that is slower than the time window of the experiment. Reprinted with permission from ref. 10. Copyright 2014 Proceedings of the National Academy of Science.

Another piece of experimental evidence supporting this conclusion came from a recent study of Nesbitt and coworkers.¹³ Using single-molecule fluorescence resonance energy transfer (smFRET) spectroscopy, they examined the folding/unfolding kinetics of an 8 base pair DNA duplex and found that TMAO stabilized the folded structure by both increasing the folding rate and decreasing the unfolding rate. However, further van’t Hoff and Eyring analyses led them to conclude that the TMAO-induced stabilization of the folded state largely resulted from an increased entropic cost of the unfolding process. The authors suggested that such an entropic mechanism could be due to TMAO acting as a nano-crowder. Their conclusion was in line with an earlier computational study¹⁴ on RNA hairpin formation in the presence of TMAO, which showed that if the interactions between TMAO and the RNA are repulsive, then this osmolyte is effectively acting as a small crowding agent that stabilizes the folded state via the excluded volume effect.

Such a crowding effect is not unique to TMAO, as other osmolyte molecules have also been shown or proposed to exhibit similar actions. For example, the statistical study of Graziano¹⁵ on a model protein suggested that the ability of trehalose to counteract the denaturation effect of urea is due to the increased cost of cavity creation in trehalose/urea solutions, which outweighs the destabilizing interactions of the protein with urea molecules. This essentially suggests that trehalose protects the folded state from unfolding via the excluded volume effect. Additionally, a recent study by Jain et al.¹⁶ hypothesized that the ability of the osmolyte glycine betaine (GB) to counteract the destabilizing effect of chemical denaturants and temperature changes on ferricytochrome c and ferrocytochrome c is partially the result of zwitterionic GB being excluded from the protein surface in much the same way as TMAO. Moreover, using MD simulations, Zhou and coworkers¹⁷ observed that in the presence of guanidinium chloride (GdmCl), urea actually induced a crowding effect on two proteins, hen egg-white lysozyme and protein L, causing them to collapse. In this case, GdmCl, which has a stronger electrostatic interaction with the protein, caused urea to be depleted from the protein surface. However, the displaced urea molecules accumulated near the protein, producing a “local crowding” effect. Taken together, these studies suggest that when a co-solvent molecule is preferentially depleted from the surface of a biomolecule, a measurable excluded volume effect may arise.

Motivated by these previous studies and also the fact that trifluoroethanol (TFE), a co-solvent commonly used to enhance protein secondary structure formation, can form clusters at certain volume percentages,¹⁸ Culik et al.¹⁹ hypothesized that TFE could act as a nano-crowder to increase the rate of protein folding. To test this hypothesis, they studied the effect of various percentages of TFE in water on the folding rate of an intrinsically disordered peptide, the phosphorylated kinase inducible domain (pKID) peptide. As shown (Figure 3), upon increasing the TFE percentage from 5% to 15%, there was an increase in the relaxation rate and hence, the folding rate of the peptide, which remained unchanged up to 30% TFE. Further increase of the TFE percentage led to a decrease in the relaxation rate. Since the cluster formation of TFE was previously shown to reach its maximum around 30% TFE, these results thus support the idea that TFE can affect the structure and dynamics of polypeptides via volume exclusion.

Figure 3.

Conformational relaxation rate constant, k_R, of pKID versus temperature, T, for various TFE percentages, as indicated. The solid lines are to guide the eye. Reprinted with permission from ref. 19. Copyright 2014 American Chemical Society.

We hope these abovementioned examples provide compelling evidence that small molecules can act like macromolecules to crowd biopolymers and the resultant excluded volume effect can be significant. In addition, they demonstrate the need for examining other commonly used co-solvents in biophysics and biochemistry for this effect.

Nano-Crowding Arising from Molecular Constraints

The degree of crowding of a solution can be tuned by the addition of an appropriate amount of a co-solvent or co-solute. However, this approach would not be useful in applications where only a specific region of the biomolecule of interest needs to be crowded. One promising and practical approach of site-specifically increasing the local mass density and hence, the degree of crowding, is to insert a specific construct to the desired molecular location. Indeed, several studies have taken advantage of recent developments in peptide chemistry and protein design and used molecular cross-linkers,²⁰ such as azobenzene and m-xylene, to manipulate molecular crowding in a site-specific manner. For example, in order to mimic the crowded environment within which α-helix folding takes place in proteins, Hamm and coworkers²¹ employed a photo-switchable azobenzene moiety to simultaneously serve as a photo-trigger of the folding kinetics of an alanine-rich peptide and a structural constraint to reduce its conformational flexibility. They found, akin to the excluded volume effect, that this constraint decreased the entropic cost of folding by limiting the conformational motions of the unfolded peptide. In addition, they found that the azobenzene cross-linker stabilized kinetic traps along the folding pathway that were not present in the unconstrained system and that escape from these traps was the rate-limiting step to complete α-helix formation. Moreover, by altering the bulkiness of the sidechains in the peptide near the azobenzene cross-linker, Paoli et al.²² found that peptides with bulkier sidechains folded slower than the alanine-rich peptide. This was attributed to an increased steric hindrance of the cross-linker on the rearrangement of these sidechains.²³

In a similar study, Markiewicz et al.²⁴ used a molecular cross-linker to help assess the effect of internal friction on protein folding dynamics (Figure 4). While it is well known that internal friction arising from adjacent structural elements, due to interactions and/or sterics, can decrease the folding rate as the polypeptide chain becomes increasingly compact throughout the folding process, a direct measurement of its effect has been difficult. Using the mini-protein Trp-cage as an example, the authors showed that it is possible to employ an “inert” cross-linker to strategically increase the local mass density and hence allow a direct estimate, via folding kinetics measurements, of the magnitude of internal friction arising from such a mimic of protein secondary structural elements. Specifically, they placed an m-xylene cross-linker in a congested region of the α-helix of Trp-cage and examined the effect of this moiety on the folding thermodynamics and kinetics of this mini-protein. While the m-xylene cross-linker did not induce any appreciable change in the thermal stability of Trp-cage, it did slow down both the folding and unfolding rates in comparison to those of the unconstrained protein. These rate decreases are consistent with the notion that, in this case, the m-xylene cross-linker acts as a local crowder to increase the frictional drag or roughness of the underlying folding energy landscape. Moreover, using these kinetic rates, an equation for the roughness of barrier-crossing,²⁵ and transition state theory, the authors estimated the increase in the internal friction due to the cross-linker to be between 0.4–1.0 k_BT.

Figure 4.

Cartoon illustration of the free energy landscape of the unconstrained (pink) and constrained (blue) Trp-cage peptide, showing that the cross-linker increases the internal friction of the folding process.

While these examples demonstrate the possibility of using a cross-linker to evaluate the effect of localized crowding on protein folding dynamics, care must be taken when utilizing this approach, as incorporation of an external structural element into a protein can lead to changes in other structural and/or physical properties of the protein²⁶ besides the intended change in local mass density. Despite this shortcoming, we believe that this strategy, when implemented appropriately, offers a convenient means to tune the excluded volume in the region of interest, thus allowing assessment of its effect in a site-specific manner.

Nano-Crowding via Molecular Packing

Many biophysical and/or biochemical techniques require coating of a substrate surface with specific biomolecules. For example, DNA microarrays rely on functional DNA molecules tethered to a solid support to monitor gene expression.²⁷ To achieve high throughput and sensitivity, it is common to use densely-packed surfaces in these types of applications. However, crowding effects incurred from densely-attached molecular moieties can significantly alter their structure,²⁸ structural transition,²⁹ functionality,³⁰ and reactivity. For example, the study of Peterson et al.³¹ showed that the DNA hybridization rate and efficiency to bind DNA probes immobilized on gold surfaces depended strongly on the probe density. In the lowest probe densities studied, 100% hybridization was possible, whereas only ~10% efficiency was obtained when the probe density reached 12.0 × 10¹² molecules/cm². Similarly, the study of Majumdar and coworkers³² on the effect of salt concentration on DNA immobilization and hybridization found that the hybridization density was the largest for 200–400 mM salt solutions since the probe density decreased at lower salt concentrations yet surface crowding caused repulsive interactions at higher salt concentrations. In another study of surface immobilized DNA molecules, Scoles and coworkers^33,34 found that the ability of a restriction enzyme to cut the attached DNA in half is highly dependent on the packing density of the DNA with a threshold density above which the reaction was impeded. By analyzing the experimental data using a semiquantitative model, they concluded that the restriction enzyme must have entered the DNA matrix from the side as access from the top was hindered. However, the two-dimensional diffusion of the enzyme became highly restricted in high DNA densities due to crowding, resulting in incomplete or negligible cleavage of the matrix. As shown by Saha et al,³⁵ the density of surface packing can also have a significant effect on the enzymatic activity of proteins. By studying the reactivity of trypsin protease immobilized on copper sulfide nanoparticles, the authors found that at monolayer surface coverage (~2.0 mg/m²), 98% of the enzymatic activity was retained, but under more densely packed conditions (e.g., ~14.0 mg/m²) the enzyme lost 23% of its catalytic activity. They ascribed this loss of activity to a change in the tertiary structure of the attached protein molecules caused by steric crowding effects on the surface.

It is clear that packing density can have a profound influence on the physical and/or chemical properties of biological molecules that are adsorbed on or attached to the surface of an engineered substrate, and the resulting effect can be quantified in many cases. However, for naturally-occurring biological surfaces such as cell membranes, where a large number of different molecules co-exist, the effects of crowding are much less explored and understood. In one computational study, Phillips and coworkers³⁶ found that a crowding-induced entropic tension in the membrane can change the gating energies of bacterial mechanosensitive membrane channels by more than 2 k_BT. Other studies^37–39 have shown that protein crowding can cause membrane buckling and deformation. Because cell membranes play a critical role in many cellular processes, it is expected that more studies relating crowding to biological membrane activity will be carried out in the future.

Nano-Confinement via Inclusion

Another mechanism by which the excluded volume effect on a biophysical system can be studied is through confinement. By restricting the volume of a molecule of interest through a confining boundary, an excluded volume effect is produced, similar to the way in which solvent molecules and molecular constraints can crowd a molecule of interest. While there are many strategies to experimentally and computationally confine a molecule, including liposomes,^40,41 sol-gel matrices,^42,43 nanofluidics,^44,45 among others,^46–52 herein we only focus on reverse micelle (RM) encapsulation, a widely used technique in the field of biophysical chemistry. A RM consists of self-aggregating surfactants that form a spherical shell in an organic solvent, with the nonpolar tails of the surfactants oriented outwards into the solvent and the polar head groups of the surfactants sequestering a nanopool of water inside the RM. While many surfactants, such as anionic sodium dodecyl sulfate (SDS), cationic hexadecyltrimethylammonium bromide (CTAB), and neutral Triton X-100 (TX-100), are capable of forming RMs,⁵³ anionic sodium bis(2-ethylhexyl) sulfosuccinate (AOT) is the most commonly used surfactant in biophysical studies. By increasing the water loading parameter (w₀), defined as the ratio of the concentration of water to surfactant, the size of the nanopool of water, and thus the RM, also increases. Therefore, one can control the number of water molecules inside a RM, providing an easy control over the degree of hydration of the encapsulated guest molecules. As the activity of confined water differs from that of bulk water,⁵⁴ RMs of relatively low w₀ values have been used extensively to study the chemical, physical, or photophysical properties of a wide variety of molecular systems. Because of space limitation, below we only discuss studies that utilize RMs to study how hydration and confinement mediate the structure, dynamics, and function of biomolecules. For readers interested in other applications and aspects of RMs, we direct their attention to several recent reviews.^55–58

The work of Wand and coworkers^59,60 demonstrated that AOT RMs in low-viscosity solvents are ideally suited to study protein hydration dynamics using NMR spectroscopy. This is because the confinement of both the water and the protein inside the RM slows down the residence time of the water on the protein and removes any ambiguity between long-range contributions from bulk water molecules and those from the hydration water molecules, i.e., the water molecules in the first and second layer around the protein, thereby allowing high-resolution NMR techniques to be used. For example, using both ¹⁵N- and ¹³C-resolved NOESY and ROESY experiments, they initially studied the protein ubiquitin and determined the location of hydration waters on the surface of the protein and the effective correlation time between the water and the protein at each location, effectively mapping the hydration dynamics of the protein surface (Figure 5). Interestingly, the areas of restricted hydration dynamics were found to be areas of ubiquitin that are important to its binding interactions with other proteins. After this initial application, Wand and coworkers have extended and optimized this RM encapsulation strategy for many other proteins, including cytochrome c and flavodoxin, and have used this method to investigate ligand binding, among other processes.⁶¹

Figure 5.

Ubiquitin hydration surface and comparison to its protein-protein interaction surface, shown from two perspectives. Average σNOE/σROE ratios from −0.5 (blue) to 0 (red) are mapped to the surface, showing the regions of slower hydration dynamics (blue) and faster hydration dynamics (red). The yellow circle shows the region of ubiquitin binding interactions. Reprinted with permission from ref. 59. Copyright 2011 American Chemical Society.

In a different application, Mukherjee et al.^62,63 utilized AOT RMs to investigate the effect of hydration and confinement on the helix-coil equilibrium of alanine-based peptides. They found that as w₀ decreased (i.e., from 20 to 6), the helicity of the peptide increased, indicating that dehydration of the peptide backbone preferentially stabilized helical structure. However, further decreasing w₀ to 4 resulted in a decrease in helicity, indicating that the excluded volume effect becomes dominant under this confinement condition, forcing the peptide to adopt an ensemble of conformations that are shorter than the α-helical structure. In addition, their results demonstrated that the degree of dehydration of the peptide backbone can be quantified via IR spectroscopy, as the hydrated and dehydrated backbone units show distinguishable amide I bands.^64,65 The latter finding suggests that it is possible to use this approach (i.e., the systematic variation of w₀) and isotope-editing⁶⁶ to determine the mechanism of backbone dehydration in a site-specific manner.

The MD simulations of Straub and coworkers⁶⁷ provided further atomistic details of the abovementioned peptide system in spherically restricted and unrestricted RMs of w₀ = 6. They found that α-helical structure was stable under both conditions, consistent with the experimental observation. In addition, they found that in restricted RMs, the amount of water molecules within 4 Å of each residue varied between 8 and 23; however, most residues had about 12 water molecules nearby while only the N- and C-terminal residues were close to the AOT tail groups. Conversely, in unrestricted RMs, they found that the degree of hydration was much smaller with about 6 water molecules per residue and that the peptide interacted more with the AOT tails (Figure 6). The latter indicates that there is more “free water” in the unrestricted RM, which may increase the entropy of the system. Similar results of peptide interactions with the water/AOT interface were observed by Garcia and coworkers.⁶⁸ Interestingly, using Phe_CN as a site-specific probe, Mukherjee et al.⁶⁹ further showed that when confined in AOT RMs of w₀ = 6, a membrane-binding and amphipathic peptide mastoparan × (MPx) was found to be located in the interfacial region of RMs. Additionally, their results suggested that the hydrophobic face of the amphipathic MPx α-helix was actually directed towards the polar core of the RM such that the positively-charged, hydrophilic side of the helix could face the negatively-charged AOT head groups.

Figure 6.

MD Snapshots of an alanine-based peptide in a spherically restrained (left, pink) and a spherically unrestrained (right, green) RM, showing the interaction between the peptide, the surrounding water molecules (blue), and the AOT tail groups (gray). Reprinted with permission from ref. 66. Copyright 2014 American Chemical Society.

For proteins, the stabilizing or destabilizing effect arising from RM confinement may be more complicated. For example, Chowdhury and coworkers⁷⁰ found that confining myoglobin (MB) in relatively small AOT RMs (w₀ = 4 to 12) led to a decrease in its secondary structure with a trend that increasing w₀ induced a greater destabilization. Their results suggest that effects beyond simple confinement are at play. In addition, they found that while MB unfolded cooperatively in buffer, it exhibited noncooperative unfolding transitions in RMs. They attributed those behaviors to two factors: (1) the size of the protein was comparable to the size of the confining cage, i.e., the RM, and (2) the polarity of the RM head groups had a destabilizing effect on the protein.

Further experimental evidence that the RM polarity can affect the conformation of a confined protein came from the work of Pal and coworkers,⁷¹ who studied the electron transfer yield from riboflavin binding protein (RBP) to the bound cofactor riboflavin (Rf) in both anionic AOT and cationic CTAB RMs. They found that, in similar size water pools, the electron transfer was observed to occur in CTAB RMs but not in AOT RMs. They attributed this difference to the polarity of the confining surface; the anionic AOT interacted strongly with the negatively-charged protein, which in turn affected the Rf binding site such that Rf bound to RBP differently, destroying the electron transfer pathway. Meanwhile, the cationic CTAB did not disrupt the protein conformation and thus, the binding, allowing the electron transfer to occur similarly to the process observed in buffer. Using molecular dynamics simulations, Garcia and coworkers⁷² also showed that the stability of the miniprotein Trp-cage confined in fullerene balls depended on the polarity of the confining wall. Their results indicated that under nonpolar confinement, the folded state of the protein was stabilized while polar confinement destabilized the folded state because of interactions with the walls of the confining cage. Additionally, Li et al.,⁷³ through Gō-model simulations, observed more conformational switching of adenylate kinase from the active state to the inactive state when the attractive interactions between the confining cage and the protein were increased. Taken together, the effect of the confinement polarity appears to be both sequence- and structure-specific to the protein encapsulated.

It has been predicted that dehydration plays an important role in the formation and growth of amyloid fibrils.⁷⁴ To test this hypothesis, Mukherjee et al.⁷⁵ employed AOT RMs of various sizes to systematically tune the degree of hydration of two model amyloid peptides, Aβ_16–22 and Sup35_7–13, and used infrared spectroscopy to follow their aggregation kinetics under those confined conditions. Their results indeed showed that in a more dehydrated environment (e.g., w₀ = 6) these peptides aggregated much faster than in a less dehydrated condition (e.g., w₀ = 20). Similar results were also obtained in a molecular simulation study.⁷⁶ Additionally, Axelsen and coworkers⁷⁷ found that monomeric Aβ_1–40 formed extended β-strands in AOT RM using FTIR spectroscopy, suggesting that this conformation could “seed” the formation of amyloid fibrils.

The examples discussed above provide compelling evidence that small RMs are versatile nano-confining cages of biological molecules and afford a wide range of applications. However, such confinement does not only limit the free space accessible by the encapsulated molecules, but can also affect their properties via other factors, including the water density and activity, specific and non-specific interactions with the confining cage, and the ability to exchange contents between RM cavities. Thus, in practice, potential complications arising from these non-excluded volume factors need to be considered when designing confinement experiments using RMs.

Summary and Outlook

Since the cellular environment is highly crowded due to the co-existence of various macromolecules and their complexes, the study of how macromolecular crowding affects the structure, dynamics, and function of biological molecules is currently a very active research area. As discussed above, however, several recent studies indicate that even small molecules, such as TMAO and TFE, can also act as effective crowding agents to influence, for example, the conformational dynamics and stability of proteins. Thus, these studies highlight the need to examine the potential crowding ability of other commonly encountered small co-solvent molecules in biochemical and biophysical studies. For instance, considering that TMAO is one of the many small organic molecules that nature uses to cope with various stresses, it is possible that other osmolytes, such as glycine, taurine, and dimethylsulfoniopropionate, could exert a similar entropic effect as TMAO to stabilize the folded state of proteins. Hence, it would be particularly worthwhile to verify this point in future studies. In addition, as highlighted above, a specifically engineered crowded environment can be created, for example, by increasing the local mass density or through confinement and the resultant crowding effect can be investigated to provide insight into various biochemical and biophysical questions. In particular, we believe that the strategy of using a structural motif to crowd or constrain a specific volume element in the biomolecule of interest has the potential to be useful in the following applications: (1) to help assess the magnitude of internal friction encountered in protein folding dynamics, (2) to trap a high free energy state,^78,79 and (3) to modulate the on- and off-rates of ligand binding by exposing binding sites in proteins using a photolyzable constraint. Similarly, the confining strategy via RM inclusion could be extended to, for example, (1) study the excluded volume effect on the structure and dynamics of intrinsically disordered proteins or other metastable proteins,⁸⁰ (2) study the structure and folding dynamics of peptides and proteins that develop secondary and/or tertiary interactions at a low degree of hydration such as the late embryogenesis abundant (LEA) proteins, which fold into α-helices upon desiccation,⁸¹ (3) expand the time window of observation in laser-induced temperature-jump experiments⁸² using an organic solvent (e.g, isooctane) that has a smaller thermal conductivity than water to slow down the rate of heat transfer from the water pool in the RMs to the surroundings, and (4) investigate the water molecules inside of the cytosol of E. coli cells using infrared spectroscopy by removing bulk water contributions through RM encapsulation.

QUOTES.

What is surprising, however, is that several recent studies have indicated that the excluded volume effect can also arise from unexpected sources, such as TMAO and TFE.

One promising and practical approach of site-specifically increasing the local mass density and hence, the degree of crowding, is to insert a specific construct to the desired molecular location.

Biographies

Mary Rose Hilaire is a Ph.D. candidate in the graduate group of Chemistry at the University of Pennsylvania and a NSF Graduate Research Fellow. She graduated with a B.S. in Chemistry from Temple University in 2012. Her research interests include using non-natural amino acids to probe protein folding and protein-ligand binding events using fluorescence and infrared spectroscopies.

Rachel M. Abaskharon is a Ph.D. candidate in the Chemistry Department at the University of Pennsylvania and a NSF Graduate Research Fellow. She graduated from The College of New Jersey in 2012 with a B.S. in Chemistry and a B.A. in Mathematics. Her current research interests lie in using photochemical triggers to initiate protein folding reactions and gain new insights into the complexities of the protein folding free energy landscape.

Feng Gai is the Edmund J. and Louise W. Kahn Endowed Term Professor of Chemistry at the University of Pennsylvania. His current research interests focus on protein folding, the protein structure-dynamic-function relationship, as well as the development of new spectroscopic probes and methods. http://gaigroup.chem.upenn.edu/