Friday, November 6, 2009

How do spores survive?

How do bacterial spores survive in a dormant state for years, perhaps in the face of high temperatures or toxic substances? It has been long suspected that the state of water in the cell compartments plays a role. Bertil Halle and his coworkers at Lund have now looked that the state of water in Bacillus subtilis spores using deuterium and oxygen-17 spin relaxation, and they find that the water is not glassy, contrary to some earlier suggestions (E. P. Sunde et al., PNAS 10.1073/pnas.0908712106; paper here). However, the water permeability of the inner membrane is unusually low, providing a barrier to the transport of toxic substances. And some of the key enzymes in the core of the spore seem to be in a relatively dehydrated state, their rotational mobility severely reduced, which might be expected to reduce the tendency of the denatured proteins to aggregate – in other words, the changes in hydration may not provide stabilization against heat-denaturation in itself, but will avoid this becoming an irreversible process.

While most studies of hydration forces between surfaces have tended to focus on hydrophobic surfaces, the nature of the interaction between hydrophilic surfaces is also controversial. It is repulsive, but the reason for this remains debated. There is some suggestion that several mechanisms might act at different length scales – for example, genuine ‘hydration’ effects due to water orientation at moderate separations (between about 0.4 and 0.8 nm), and undulation effects at larger separations. Max Berkowitz at UNC has studied this phenomenon previously using simulations of lipid bilayers (Lu & Berkowitz, J. Chem. Phys. 124, 101101 (2006) and Mol. Phys. 104, 3607 (2006)), but now he and Changsun Eun return to the problem with more accurate simulations in which the lipid head groups are allowed to be mobile (Eun & Berkowitz, J. Phys. Chem. B 113, 13222-13228; 2009 – paper here). They do indeed find three regimes. At short range (<1 nm), the repulsion is dominated by steric van der Waals interactions between the lipid headgroups. They focus mainly on the intermediate-range (1-1.6 nm) interaction, which they argue is due to the free-energy cost of removing waters hydrating the head groups. William Jorgensen has continued his examination of water in protein binding sites, an earlier instance of which was mentioned in the previous post. With Julien Michel and Julian Tirado-Rives, he reports a MD method for determining how water molecules will be situated in binding sites with or without the ligand (J. Phys. Chem. B 113, 13337-13346; 2009 – paper here). The accuracy of the method is shown by comparison with five cases where the crystal structures are known.

Another extension of earlier work: Nicolas Giovambattista, Peter Rossky and Pablo Debenedetti look at how temperature affects the behaviour of water confined between hydrophobic, hydrophilic and heterogeneous nanoscale plates (J. Phys. Chem. B 113, 13723-13734; 2009 - paper here) – the system they considered earlier in Phys. Rev. E 73, 041604 (2006), J. Phys. Chem. C 11, 1323 (2007) and PNAS 105, 2274 (2008). Cooling enables the water to approach the hydrophobic plates more closely, consistent with the expected suppression of the vapour phase. It also blurs the differences in water density between hydrophobic and hydrophilic regions of a heterogeneous surface. This would be consistent with invasion of hydrophobic cavities by water in cold denaturation.

Somewhat related is a study by Ateeque Malani at the Indian Institute of Science in Bangalore and coworkers on the differences in water structure when confined between pairs of two types of hydrophilic surface: hydroxylated silca and mica (J. Phys. Chem. B 113, 13825-13839; 2009 – paper here). While an oscillatory solvation force and a bulk-like H-bond network near the interface are found for silica (these are simulations), the network is disrupted near mica, where there are potassium ions at the surface which are themselves hydrated.

Water diffusion on the surfaces of lipid vesicles has been studied by Ravinath Kausik and Songi Han at UCSB using Overhauser dynamic nuclear polarization of hydrogen-1 NMR (JACS ASAP; paper here). They find diffusion coefficients about half those of bulk water; the key result here is a demonstrating of the feasibility of the technique for obtaining this sort of information. And James Skinner and colleagues at Wisconsin use MD and IR spectroscopy to study water inside reverse micelles (P. A. Pieniazek et al., J. Phys. Chem. B ASAP; paper here). They say that the distance from the surfactant headgroups over which the water becomes bulk-like increases with decreasing micelle size (increasing curvature), eventually becoming larger than the micelle radius. In the smallest micelle (containing 52 water molecules), the water seems to be near-glassy, with very slow rotational relaxation.

Vincent Craig at ANU, now working with Christine Henry, has extended his long-standing studies of the effects of solutes on bubble coalescence. They have looked at the effect on this phenomenon of osmolytes: sucrose and other sugars, and urea (Langmuir 25, 11406-11412; 2009 – paper here). Urea seems to have little effect, but sucrose and other sugars show an inhibiting influence on coalescence. This suggests that, contrary to what one might have been tempted to infer from previous studies on electrolytes, the inhibitory effect does not stem from solute charge. They speculate that concentration gradients close to the bubble-water interface may instead be responsible.

The influence of urea and another osmolyte, trimethylamine-N-oxide on the structure of water and hen egg-white lysozyme are studied using FTIR by Janusz Stangret and colleagues at the Gdansk University of Technology in Poland (A. Panuszko et al., J. Phys. Chem. B ASAP; paper here). Water structure is barely affected by urea, they say, but more strongly perturbed by TMAO, forming stronger and more ‘ordered’ H-bonds. They monitor the protein via the amide I band and suggest that the changes seen there are consistent with changes in water structure, resembling in the case of TMAO changes that are evident on dehydration. This all seems to be presented within the framework of osmolytes exerting indirect effects via their influence on water structure – but I guess one would want to know precisely how the osmolytes interact with the protein itself.

Haiping Fang at Shanghai and colleagues have continued their investigation of water transport through nanochannels. They show how symmetry-breaking of water orientation in a one-dimensional H-bonded chain threading through a carbon nanotube can give rise to spontaneous unidirectional net flux in the absence of any external pressure gradient (R. Wan et al., Phys. Chem. Chem. Phys. 11, 9898-9902; 2009 – paper here doi:10.1039/b907926m). And Haiping also has a paper in PNAS (10.1073/pnas.0902676106; paper here) reporting simulations in which the presence of a single-electron charge in one arm of a Y-shaped carbon nanotube junction can, by flipping the orientation of a water molecule in a single-file chain within the nanotube, reorient the dipoles of the water chains in the other two branches, thus multiplying the single-electron signal. With a suitable arrangement of junctions, it can be multiplied more than twofold.

On a similar topic, Padmanabhan Balaram and colleagues at the Indian Institute of Science use MD simulations to look at the structure of one-dimensional water chains inside the hydrophobic core of a tubular synthetic protein (U. S. Raghavender et al., JACS 131, 15130-15132; 2009 – paper here). They find two distinct states in different peptides: one in which the water molecules are disordered over two possible positions in the chain, the other in which the molecules are perfectly ordered along a sixfold screw axis.

There’s a very neat demonstration of how water in binding sites can be engineered to improve function in a paper on a catalytic antibody by Ian Wilson at Scripps and colleagues (E. W. Debler et al., PNAS 10.1073/pnas.0902700106; paper here). They find that an oriented water molecule in the hydrophobic pocket of the antibody 13G5, which catalyses the cleavage of benzisoxazoles, stabilizies the developing charge on the leaving group. And in a single-residue Glu-to-Ala mutant, a hydrogen-bonded complex involving four water molecules is restructured in a way that enhances still further the rate of the proton transfer involved in the process. A key role for water in another catalytic antibody is reported by Orlando Acevedo at Auburn University in Alabama (J. Phys. Chem. B ASAP; paper here). He looks at the antibody 4B2, which catalyses both a Kemp elimination and an allylic isomerization of an unsatuated ketone. For the former, water molecules in the active site help stabilize the transition state during proton abstraction, while in the latter case the water takes an active role as a proton source. Acevedo suggests that water might be usefully engaged in other designed catalysts to perform this function as a proton donor.

A paper on the behaviour of water and proteins confined in nanoporous (c. 5 nm) silica by Eduardo Reátegui and Alptekin Aksan at Minnesota (J. Phys. Chem. B 113, 13048-13060; 2009 – paper here) contains rather more information than I can easily digest yet. They use FTIR to characterize what is happening to the water, which is of course a slightly blunt tool on its own – certainly, the claim to see liquid-liquid transitions analogous to the putative HDL-LDL transition seems a big one to make on these grounds alone. Changes in the encapsulated proteins, monitored by amide IR bands, seem to mirror those seen in water OH bands, but it’s again not too clear what these actually correspond to in terms of structure or function.

Sotiris Xantheas and Greg Voth, working with Francesc Paesani, have developed an ab initio force field for water that, in MD simulations, provides a good fit for the experimental IR spectra probing H-bond dynamics (J. Phys. Chem. B ASAP; paper here).

The hydrodynamics of water at surfaces has been a controversial topic, and one with some important practical implications. It seems clear that the common no-slip assumption for fluids at solid interfaces doesn’t necessarily hold at the nanoscale. Roland Netz and his coworkers have investigated this for hydrophilic and hydrophobic surfaces using MD simulations (C. Sendner et al., Langmuir 25, 10768-10781; 2009 – paper here). They find something like an inverse square dependence of slip length on contact angle for hydrophobic surfaces, but slip lengths of typically only a few nm for realistic contact angles. This is little affected by dissolved gas at the interface, and the viscosity of the interfacial water is only a few times higher than that of the bulk, with molecular motions being purely diffusive. In contrast, on hydrophilic surface water molecules may become trapped, there is no slip, and the interfacial water viscosity may be enhanced significantly. In the same vein (and consistent with these results), Bharat Bushan and colleagues report measurement of the hydrodynamic forces acting on a glass sphere glued to an AFM tip as it approaches a mica surface (A. Maali et al., Langmuir 25, 12002-12005; 2009 – paper here). They say that the measurements are consistent with a no-slip assumption at both glass and mica surfaces.

Xavier Tadeo and colleagues at the Centro de Investigación Cooperativa bioGUNE in Derio, Spain, have looked at how the anions of the Hofmeister series affect protein stability, using as their test case the IGg binding domain of protein L from Streptoccocal magnus (ProtL) (Biophys. J. 97, 2595-2603; 2009 – paper here). They look at changes in thermostability of a lysine-to-glutamine mutant in the presence sodium salts of sulfate, phosphate, fluoride, nitrate, perchlorate and thiocyanate, and say that the results are consistent with stabilization of the native state by an increase in solution surface tension due to the anions. I’ve not seen the full paper, but I do wonder whether such bulk effects on surface tension can be a reliable guide to what is going on here, without knowing how the ions are partitioned at the protein-solvent interface.

More on Hofmeister effects: Xin Wen and colleagues at CSU at Los Angeles look at the effects of monovalent salts on the activity of the antifreeze protein DAFP-1 from the beetle Dendroides canadensis (S. Wang et al., J. Phys. Chem. B ASAP; paper here). Specifically, they use DSC to monitor how the difference in melting and freezing point of water due to the antifreeze protein is altered by the salts: salting-out seems, as might be expected, to enhance the adsorption of DAFP-1 on the ice surface, thereby boosting its activity.

Henry Ashbaugh at Tulane University has a nice paper on how different sequences of hydrophobic and hydrophilic monomers in a heteropolymer will affect its conformation in aqueous solution (J. Phys. Chem. B 113, 14043-14046; 2009 – paper here). Intermediate segregation of monomer types favours a collapsed conformation, while more strictly alternating monomers favours a random coil.

James Beattie and his coworkers in France and Australia have another paper making the case that the interface of water with air or oil is basic rather than acidic, due to specific adsorption of hydroxide (P. Creux et al., J. Phys. Chem. B ASAP; paper here). This claim is made on the basis of measurements of the zeta potential. I suspect the debate will continue.

There may not be another post from me here for a couple of months, owing to an imminent new arrival in the family. No doubt this means I’ll miss some interesting papers in the interim. But do feel free to send me or tell me of interesting ones ( Hope to be back up and running after Christmas – have a good one.

Tuesday, September 29, 2009

Water in drug design... and why green tea keeps you young?

Can the displacement of water from binding sites be used as a tool for drug design? John Ladbury has written a fair amount on this question in the past (e.g. Chem. Biol. 3, 973; 1996), but it remains hard to deduce any general principles. In principle both the enthalpy and the entropy of ligand binding can be enhanced by displacing water molecules, but the balance is subtle and not obviously generic. Julien Michel, Julian Tirado-Rives and William Jorgensen at Yale have now looked at what can be learnt from some specific cases, namely ligands designed to bind to three proteins: scytalone dehydratase, p38-aMAP kinase and EGFR kinase (JACS ja906058w – paper here). They find for a series of ligands that in general the binding affinity correlates with ease of water displacement, but also that it can be important to ensure that it can be important to compensate for the free-energy increase of water displacement by building in additional favourable interactions between the binding site and the displacing group – that is, it’s not necessarily enough to expel the water; one has to put something amenable in its place. My reading of this paper is that there are no wholly reliable short-cuts or rules of thumb – the details matter.

Nicolas Giovambattista, Pablo Debenedetti and Peter Rossky continue their investigations of how surface topography affects hydrophobicity (PNAS 106, 15181; 2009 – paper here). They have used simulations to look at the hydration of a silica surface on which they vary the surface polarity and topography (atomic- to micrometre-scale roughness), and find that the two can couple in such a way as make a somewhat polar surface more hydrophobic than an apolar one. This suggests that surface patterning could be a powerful tool for altering hydrophobicity.

Feng Guo and Joel Friedman at the Albert Einstein College of Medicine have probed Hofmeister effects by looking at the effect of adding anions and cations to a solution of gadolinium (III) ions, both free and coordinated to organic and biological molecules (JACS 131, 11010; 2009 – paper here). The idea is that changes in the OH stretching mode of waters in the Gd hydration shell offer a sign of what is happening elsewhere in the solution. It’s not entirely clear how the former changes relate to the latter, but Guo and Friedman suggest two possibilities. For example, the tight hydration clusters of high-charge-density cations might both alter water structure quite generally, and might sequester water molecules in a way that frees up anions to interact with the Gd hydration shell. In the first of these cases, the interpretation is predicated on a notion of inhomogeneous bulk water structure composed of a mixture of high- and low-density water clusters, based on Wilse Robinson’s model. Frankly, I’m not persuaded that this is a persuasive way to interpret the findings – the two-state water model is of course controversial. But the overall conclusion – that Hofmeister ordering of salts can be best explained by disruption of hydrogen bonding in hydration-shell water rather than bulk water – has a ring of truth to it.

But the Hofmeister plot thickens further. Yanjie Zhang and Paul Cremer at Texas A&M have also been investigating these effects, and they find that for lysozyme, aggregation in salt solution (followed by looking at cloudiness of the solution and eventual phase separation) seems to follow two different Hofmeister series depending on salt concentration (PNAS 106, 15249; 2009 – paper here). At low concentration, the usual series for anions (which dominate aggregation effects in this case) is reversed, and is correlated with size and hydration thermodynamics of the ions. At high concentrations, there is a normal Hofmeister series correlated with the ionic polarizability.

Daisuke Matsuoka and Masayoshi Nakasako at Keio University in Japan have conducted a massive study of the probability distributions of water molecules hydrating proteins, using nearly 18,000 crystal structures from the Protein Data Bank (J. Phys. Chem. B 113, 11274; 2009 – paper here). They find that the angular distributions are narrowest in the direction of N-H and O-H bonds, and that pairs of polar atoms are often arranged to satisfy the tetrahedral H-bonding geometry, suggesting that – if one may put it like this – the protein’s secondary structure is arranged to ‘suit’ the solvent rather than vice versa.

Is the hydrophobic core of a protein rigid or fluid? In reality neither extreme seems terribly likely, but Liliya Vugmeyster at the University of Alaska-Anchorage and colleagues have probed its dynamics using deuteron solid-state NMR (JACS ja902977u – paper here). Specifically, they look at the chicken villin headpiece subdomain protein HP36 labelled with deuterons at either of the two methyl groups of a leucine residue. They find that there is restricted diffusion along an arc, and larger jumps between rotameric conformers.

Eduard Schreiner and colleagues at Bochum have used ab initio MD simulations to look at the reaction kinetics and mechanisms of peptide bond formation and hydrolysis among activated amino acids (alpha-amino acid N-carboxyanhydrides, NCAs). These are used in the synthesis of large polypeptides, and may also have prebiotic relevance. The team compares the processes in water under ambient conditions and in hot pressurized water, such as might be found at hydrothermal vents (JACS ja9032742 – paper here). The hydrolysis mechanisms are different in the two cases, but in both the barrier to hydrolysis is greater than that to peptide bond formation, showing that peptide production is feasible in either case.

Ali Eftekhari-Bafrooei and Eric Borguet at Temple University use IR pump-probe spectroscopy to look at how the vibrational dynamics of O-H stretching of interfacial water groups is altered by surface charge at the interface with silica (JACS 131, 12035; 2009 – paper here). They say that, while the vibrational dynamics are retarded at the neutral surface (at low pH), surface charging at higher pH causes polarization of the water molecules that leads to more bulk-like behaviour.

Changes in water dynamics are also the subject of a paper by Matías Pomata in the National Commission for Atomic Energy of Argentina and colleagues, who consider this issue for carbohydrate (fructose) solutions (J. Phys. Chem. B jp904019c – paper here). They say that for sugar concentrations above about 45%, the sugar molecules form a percolating H-bonded network encompassing patches of solvent, resulting in a slowing of translational, rotational and H-bonding dynamics (especially for water-solute) that is comparable to what is seen for the hydration layers of proteins (e.g. Pizzitutti et al., J. Phys. Chem. B 111, 7584; 2007).

Most of the work on the pseudo-glass transition of proteins around 200 K has been based on globular proteins. How do extended structural proteins such as elastin and collagen compare? That question is probed by Catalin Gainaru and colleagues at TU Dortmund (J. Phys. Chem. B jp9065899 – paper here). Elastin exhibits something like a glass transition at temperatures only slightly below physiological, but not collagen. Yet the dielectric relaxation appears to be the same for both molecules between about 140 and 220 K, showing that relaxation is Arrhenius-like (thermally activated) over the entire temperature range: there seems in this case nothing ‘anomalous’ about the dynamics in this regime.

Some time ago Haiping Fang at the Shanghai Institute of Applied Physics sent me a preprint describing MD simulations of the wetting properties of water films on polar, neutral surfaces. This has now been published (C. Wang et al., Phys. Rev. Lett. 103, 137801 (2009) – paper here). The paper looks at the behaviour of a water monolayer on a surface designed to resemble a semiconductor such as InSb(110). On some such surfaces (including metals), a monolayer at low temperatures is found to adopt a 2D ice-like structure with no dangling bonds, making it somewhat hydrophobic. But one would expect thermal fluctuations to create some dangling OH bonds at room temperature, increasing the hydrophilicity. That does happen here, but surprisingly the monolayer nevertheless remains significantly hydrophobic even at room temperature, so that a water droplet does not wet it.

On the same theme of interfacial water at solid surfaces, Andrei Sommer at Ulm and colleagues have continued their investigations of thin water films on diamond (Crystal Growth and Design 9, 3852; 2009 – paper here). They have studied the nature of water fimls on hydrogenated and non-hydrogenated nanocrystalline diamond at room temperature with atomic force acoustic microscopy. For the hydrogenated (hydrophobic) surfaces, water monolayers seem to be securely anchored and crystalline, promoting lubrication. The authors suggest that similar water layers might exist at the surfaces of biological fibrous tissues such as elastin, promoting easy sliding – an idea proposed in 1971 by Albert Szent-Györgyi (Perspect. Biol. Med. 14, 239; 1971).

Andrei and colleagues also found that red laser light can influence the structure of the water films, from which they think they may be able to develop new skin treatments. This is what Andrei has just sent to me:
“From what we learned about the interaction of red light with interfacial water layers on hydrophobic and hydrophilic surfaces, we designed a model of physiological aging and applied it in a real skin rejuvenation study. The good news is that the paper dealing with the model and its results (Facial Rejuvenation in the Triangle of ROS, Crystal Growth & Design, October 7 issue) produced an extraordinary impact in Google (Key words: green tea, light, wrinkles.)

“The bad news is that the journalists who wrote about the work overlooked the point in the study, whose implications are more interesting: we put forward the first physicochemical explanation of the shortening of the telomeres, and thereby of cellular aging. In a nutshell: we postulated that the shortening of telomeres is due to the coincidence of mechanical pulling during cell division and the build-up of a glue-like interfacial water in the space between the nuclear matrix and telomeres. The latter is modulated by an increase in interfacial pH, triggered by the emergence of reactive oxygen species (ROS) in the cell. The increase of intracellular ROS is induced by oxidative stress, which can have two different causes: external (e.g., UV radiation, air pollution), or internal (e.g., replicative stress). In both scenarios the principal actors involved in the production of ROS are mitochondria in the cell. This part of the process is known and comprehensively described in the literature.

“The exciting thing in all this is that, according to our research, it is possible to liquidify the glue-like interfacial water layers (to reduce the viscosity) by shining moderately intense red light on them. (Albert Szent Györgyi would like to read this, because it relates interfacial water with the the biological clock underlying the limited division potential of somatic cells, which is the length of the telomeres.)

“In our study we used to rejuvenate the skin a combination of topically applied green tea and red light (670 nm). We just discovered a recent paper (R. Chan et al., Brit. J. Nutr., August 12, 2009) in which the authors report that green tea is instrumental in preventing the shortening of the telomeres, and they extrapolate that drinking green tea might extend our life by 5 years.

“Interestingly the cells of turtles - the methuselahs in our world - present no or only very limited shortening of their telomeres. Their shells provide them a perfect protection from destructive environmental impacts, including UV, and air pollution (i.e. volcanic ashes on a primitive Earth). In addition, their oxygen turnover rate is very low.”

This is fascinating stuff, and I must explore it some more. In the meantime, I think I will put the kettle on.

Tuesday, August 25, 2009

Nanobubbles: birth, motion, and consequences

The role of bridging bubbles in the so-called ‘long-range hydrophobic force’ seems now fairly well established. In a study of this effect, Viveca Wallqvist and colleagues argue that in cases where this is the identified mechanism of attraction, it would be preferable to call it a ‘capillary force’ rather than a ‘hydrophobic interaction’. Their study looks at the effect of surface roughness on such forces between hydrophobic surfaces (V. Wallqvist et al., Langmuir 25, 9197 (2009) – paper here). They find that the range and magnitude of the force can vary significantly at different points on nanostructured surfaces due to local variations in contact angle. A high density of nanoscale crevices leads to accumulation of air bubbles that coalesce and weaken the capillary attraction.

William Ducker offers an explanation of the low contact angle and the unusual stability of these nanobubbles, in terms of a thin film of surface-active contaminant at the air-water interface (Langmuir 25, 8907 (2009) – paper here). This, he says, will both decrease the surface tension (and thus the contact angle) and hinder gas diffusion out of the bubble. It’s an alternative to Michael Brenner and Detlef Lohse’s suggestion of a dynamic stabilization of the nanobubbles (Phys. Rev. Lett. 101, 214505 (2008)).

Yi Zhang and colleagues at the Shanghai Institute of Applied Sciences suggest that a precursor to these nanobubbles may be a multilayer (bilayer or trilayer) of adsorbed gas at the hydrophobic interface (L. Zhang et al., Langmuir 25, 8860 (2009) – paper here). They have imaged such bi- and trilayer islands of gas, of up to micron-sized lateral dimensions, at the surface of HOPG, and have watched them evolve into nanobubbles, sometimes under the influence of the AFM tip used for imaging.

And in the same vein, Bharat Bhushan and coworkers look at how the mobility of nanobubbles at hydrophobic surfaces is affected by surface heterogeneity (Y. Wang et al., Langmuir 25, 9328 (2009) – paper here). They find that surfaces partly and totally covered with polystyrene films, which form small islands or can become indented at the nanoscale by the presence of bubbles, tend to have relatively immobile nanobubbles compared with bare, smooth hydrophobic surfaces. Bubble immobility reduces the frictional drag force between such surfaces in motion, and so is sometimes desired in mechanical contexts.

How do you measure hydrophobicity? Macroscopically, that is of course done using contact angles. But what are the microscopic signatures? This question is examined by Shekhar Garde and colleagues at RPI using an extensive range of simulation studies of water at various surfaces ranging from very hydrophobic to very hydrophilic (R. Godawat et al., PNAS advance online publication - paper here). They say that water density is a poor measure of hydrophobicity, but that both the probability of cavity formation and the free energy of binding of hydrophobic solutes to the surface correlates much better with the macroscopic wetting properties. This paper adds weight to the notion that it is in the dynamic rather than the structural characteristics of water that the true nature of hydrophobicity is located.

Gerhard Hummer and colleagues have developed a rather comprehensive model of gated proton pumping in cytochrome c oxidase (Y. C. Kim et al., PNAS advance online publication - paper here). This reveals show the electrostatic interaction between the proton loading site and the electron source for reduction of oxygen at the heme site is central to the pumping efficiency, and also how gating is accomplished.

Water in protein cavities is hard to detect with diffraction methods if it is disordered. Robert Goldbeck at UC Santa Cruz, Raymond Esquerra at San Franscisco State University and their colleagues have shown that non-specific hydration of the cavities of myoglobin mutants can be detected optically via its perturbing effect on the optical spectrum of the pentcoordinate heme group (R. Goldbeck et al., JACS ASAP ja903409j - paper here).

How do membrane proteins compensate, within the hydrocarbon core of a membrane, for loss of hydrophobic interactions? One possibility is that they enhance packing efficiency and thus van der Waals interactions between the hydrophobic residues. Or they might have stronger hydrogen-bonding interactions between hydrophilic regions. But James Bowie and colleagues at UCLA report structural and thermodynamic arguments for why neither plays a strong role (JACS 131, 10846 (2009) – paper here). Rather, they suspect that the reduced entropy cost of folding in membrane proteins might be responsible for their stability.

There is a small clutch of papers on the structure of the air-water interface and other species located there. Yi Qin Gao and colleagues at Texas A&M use vibrational sum frequency spectroscopy and MD simulations to investigate orientational ordering of water molecules, and suggest that this occurs to any significant degree only in the first two layers at the surface (Y. Fan et al., J. Phys. Chem. B ASAP jp900117t - paper here). Joyce Noah-Vanhoucke and Phillip Geissler argue that the preferential segregation of some ions at the air-water interface is due primarily to the way the ions induce deformations of the interfacial geometry, causing electrostatic fluctuations that are not accounted for in the conventional picture (PNAS advance online publication - paper here).

In water confined to nanoscale dimensions (as in the crowded environment of a cell), hydration effects can be quite different from those of the bulk. Margaret Cheung and colleagues at Houston provide an illustration of this by using MD to look at the conformations of hexane in nanoscale water droplets (D. Homouz et al., J. Phys. Chem. B ASAP jp907318d - paper here). They find that the hexane molecules are situated at the droplet surface, where disruption of the H-bonding favours the all-trans conformation.

Meanwhile, Michael Fayer and colleagues at Stanford use ultrafast IR spectroscopy to look at water dynamics close to the neutral and ionic surfaces of reverse micelles (E. E. Fenn et al., PNAS advance online publication - paper here). They find that the orientational relaxations times are rather similar in both cases, both being significantly slower than in the bulk, and conclude that it is the mere presence of an interface, rather than its chemical nature, which exerts the dominant effect.

Nanoscopic water films on metals and other simple surfaces are known often to adopt ordered structures in the first one or two monolayers that may or may not be like bulk ice. On Pt(111), for example, it seems to form a monolayer that is flat rather than having the puckering expected of a ‘slice of ice’. Now Greg Kimmel, Bruce Kay and colleagues find that water on graphene (supported on Pt(111) also forms a flat film, here two monolayers thick (G. A. Kimmel et al., JACS ASAP ja904708f - paper here). This structure has been predicted previously for confined bilayers between hydrophobic walls, but it seems that confinement is not needed to induce it. The bilayer has no dangling bonds or lone pairs on either face, and so one might anticipate that it will itself be somewhat hydrophobic, as indeed Greg Kimmel and others found previously for the flat monolayer on Pt(111) (G. A. Kimmel et al., Phys. Rev. Lett. 95, 166102 (2005).).

Jürgen Köfinger and Christoph Dellago have used MD calculations to probe the dynamics and dielectric response of single-file water chains in narrow pores (Phys. Rev. Lett. 103, 080601; paper here). This supplies a baseline for using dielectric spectroscopy to investigate the properties of such highly confined water, for example enabling the diffusion of defects in the H-bonded chain to be studied.

Kafui Tay and Anne Boutin at the Université Paris-Sud XI have studied the dynamics of hydrated electrons using MD, and say that their diffusion is dictated by fluctuations in the H-bonded network: in the temperature region where the diffusion is Arrhenius-like, the activation energy is determined by H-bond breaking (J. Phys. Chem. B ASAP jp810538f - paper here).

Lars Pettersson, Anders Nilsson and their colleagues at Stanford, Stockholm and in Japan have published a controversial paper claiming to see inhomogeneities in water structure on length scales of around 1 nm (C. Huang et al., PNAS advance online publication - paper here). They say that they see these using SAXS, and argue that the density contrast is due to the coexistence of two water structures: one tetrahedral, the other with distorted H-bonds, related respectively to low- and high-density liquid water. This is, needless to say, a revival of the very old two-state picture of water structure, which in various forms goes right back to Roentgen. It will be disputed, no doubt, but demonstrates again how remarkably tenacious this two-state notion is.

Monday, July 27, 2009

How proteins loosen up

What happens during protein denaturation? One emerging view is that at least some forms of denaturation (such as pressure-induced) involve penetration of water into the hydrophobic interior. But that picture is challenged in a paper by Santosh Kumar Jha and Jayant Udgaonkar at the Tara Institute of Fundamental Research in Bangalore, at least for the case of denaturant-induced (GdnHCl) unfolding (PNAS 10.1073/pnas.0905744106; paper here). They have used UV circular dichroism measurements on the small plant protein monellin to show that here unfolding seems to involve a dry molten-globule intermediate, reached from the native state in a rather sharp configurational transition.

Also on this topic, Paul Cremer and colleagues at Texas A&M have used FTIR and thermodynamic measurements to probe the notion that urea denatures via direct hydrogen-bonding to the protein surface (L. B. Sagle et al., JACS 131, 9304 (2009); paper here). They challenge that idea, saying that hydrogen-bonding of urea seems to actually promote hydrophobic collapse of a polyamide (PNIPAM, here used as a protein analogue).

Ahmed Zewail and his colleagues have looked at how solvent motion couples to the unfolding of a model protein (melittin) in the presence of a denaturant (trifluoroethanol) (C. M. Othon et al., PNAS 10.1073/pnas.0905967106; paper here). Using time-resolved fluorescence spectroscopy, they see an abrupt change in solvent dynamics at a critical TFE concentration associated with a change in protein structure, in which the tetramers dissociate into loosely bound monomers owing to penetration of TFE into the hydrophobic core. The dissociation of the monomers happens in a distinct second step.

Sapna Sarupria and Shekhar Garde at RPI have studied the compressibility and fluctuations of hydration shells of hydrophobic solutes and proteins using MD simulations (Phys. Rev. Lett. 103, 037803; 2009 – paper here). They say that the compressibility is non-monotonic as a function of solute size, and that it is greater near hydrophobic solutes, relative to the bulk. The latter implies that hydrophobic interactions get weaker as pressure is increased, which may be important for pressure-induced denaturation. This pressure sensitivity is also dependent on the curvature of the solute, being greater for low-curvature surfaces. That might have a role in the pressure dissociation of multi-subunit proteins.

Incidentally, I am preparing a feature article on denaturation for Chemistry World, and would welcome any papers that might be relevant to this.

Padmanabhan Balaram and colleagues at the Indian Institute of Science in Bangalore report a very nice crystal structure of a model peptide containing a hydrophobic channel containing a linear water wire of nine molecules (U. S. Raghavender et al., JACS 10.1021/ja9038906; paper here). Looks like a good model system for studying such structures.

There is a clutch of papers on hydration dynamics of proteins. A painstaking study of femtosecond dynamics in the hydration network of apomyoglobin by Dongping Zhong and colleagues at Ohio State has revealed two distinct classes of water-network relaxation (L. Zhang et al., JACS 10.1021/ja902918p; paper here). One comes from collective hydrogen-bond rearrangements in the water shell, while the other is considerably slower and results from coupled water-protein motions. They are also able to follow changes in these motions in the transition from the native to the molten-globule state. This looks like the kind of careful study that is needed to really figure out what the intimate coupling of protein and water motions entails.

M. Vogel at the Technical University of Darmstadt has used MD simulations to look at how the dynamics of the hydration shells of peptide analogues of structural proteins (elastic and collagen) change with temperature (J. Phys. Chem. B 113, 9386 (2009); paper here). He finds that there is a change from diffusive motion at higher temperatures to jump-like motion on cooling, corresponding to a weak fragile-to-strong crossover of the water dynamics.

And Giorgio Schirò and colleagues at the University of Rome III use dielectric spectroscopy to study the dynamics of myoglobin confined in porous silica at low hydration levels, with only one or two layers of water around the protein (G. Schirò et al., J. Phys. Chem. B 113, 9606 (2009); paper here). They find that confinement has a big effect relative to hydrated myoglobin powder, suppressing the cooperativity of the water motions and the strong coupling to the protein dynamics. All the same, there still seems to be some slaving of protein relaxation in the porous medium to one mode of solvent relaxation.

Dor Ben-Amotz and colleagues at Purdue have seen the spectroscopic signature of dangling OH bonds in the hydration shells of small dissolved nonpolar molecules, similar to those seen at macroscopic water-oil interfaces (P. N. Perera et al., PNAS 10.1073/pnas.0903675106; paper here). And Pier Luigi Silvestrelli at the University of Padova offers further evidence, from first-principles calculations, that hydrophobic groups (here the methyl of methanol) don’t immobilize water molecules, iceberg-like, in the immediate hydration shell, but rather, merely slow down many surrounding molecules (J. Phys. Chem. B 10.1021/jp9044447; paper here).

It’s occasionally and justifiably said that too little attention has been given to polysaccharide hydration, in contrast to proteins. As a result, we know relatively little about glycoprotein hydration. Claudio Margulis and colleagues at Iowa State attempt to redress that imbalance somewhat with a paper looking at the hydration shells of a diverse range of carbohydrates (S. K. Ramadugu et al., J. Phys. Chem. B 10.1021/jp904981v; paper here). I’m not sure I can easily summarize the results, but there looks to be a lot of valuable information here, for example in terms of how water structure and dynamics are affected by branching, size and type of linkage in the polysaccharides.

The behaviour of water in carbon nanotubes continues to intrigue as a model for hydrophobic protein pores. MD simulations by Biswaroop Mukherjee of the Indian Institute of Science and coworkers suggest that jump reorientation of water molecules inside narrow nanotubes involves a switch of which of the two hydrogens are H-bonded to a neighbour, in contrast to such jumps in the bulk which involve H-bonding to a different neighbour (B. Mukherjee et al., J. Phys. Chem. B 10.1021/jp904099f; paper here).

Francesco Mallamace, Gene Stanley and their collaborators have a paper in Nature Physics that describes the appearance of a fractional Stokes-Einstein relation (which provides information about viscosity, connecting the self-diffusion coefficient to temperature and relaxation time) below 290 K in water confined within silica nanopores (2 nm diameter) (Nature Physics 10.1038/nphys1328; paper here). They suggest that this switch marks a crossover to a water structure that is locally more similar to LDA ice – a point at which the proportions of HDA-like and LDA-like configurations starts to change rapidly. It’s an intriguing idea, and there is apparently some indication (I can say no more yet) that the dynamical crossover might be a more general phenomenon for liquids. Some, though, will reasonably wonder whether water within 2-nm silica pores can really be considered representative of the bulk.

Perhaps relevant in this respect, Patrick Huber at Saarland University in Germany and colleagues have studied the dynamics of capillary rise in silica pores 3-5-5 nm across (S. Gruener et al., Phys. Rev. E 79, 067301 (2009); paper here). They find that they can account for the timescales of pore filling according to macroscopic hydrodynamics, so long as they assume the presence of a ‘sticky preadsorbed boundary layer of about two monolayers of water molecules’. In other words, there is dynamical partitioning of the water ‘filling’ into two components.

Masakazu Matsumoto at Nagoya University offers a new picture of the density maximum of water cooled towards freezing (Phys. Rev. Lett. 103, 017801 (2009); paper here). He challenges the common, somewhat arm-waving idea that the decrease in density below 4 C is due to a dominance of LDA-like local configurations. Rather, he says, a proper description of the associated structural changes needs to be broken down in more detail: the anomaly seems to stem from a combination of the change in average hydrogen bond length as a function of temperature (which is monotonic) and the contraction of the HB network due to bond-angle distortion. I’m imagining (it is not made explicit) that this differs from a decline in H-bond-breaking caused by a switch from HDA-like to LDA-like.

Yizhak Marcus at the Hebrew University of Jerusalem uses standard partial molar volumes to calculate the hydration numbers of a range of univalent and divalent ions under ambient conditions (J. Phys. Chem. B 10.1021/jp9027244; paper here). And László Pusztai at the Hungarian Academy of Sciences and colleagues use simulations and neutron/X-ray diffraction to find the hydration numbers of CsCl over a range of concentrations (V. Mile et al., J. Phys. Chem. B 10.1021/jp900092g; paper here). But the numbers don’t agree: Marcus calculates hydration numbers of 1.5-2.5 (depending on the definition) for Cs and 1.4-2.0 for Cl (at 25 C), whereas Pusztai et al. find respective figures of 8-6.5 (for increasing salt concentration) and 5-7. I’m not clear why the numbers are so different. Thanks to Jan Engberts for pointing me to these two papers.

How do specific ions bind to protein surfaces? The answer to this question promises to shed light on the much debated Hofmeister effects on protein aggregation, but there is still no consensus. One common rule of thumb talks of the law of ‘matching water affinities’, whereby cations and anions form ion pairs if they are more or less matched in size. Berk Hess and Nico van der Vegt present simulations which suggest that this simple relationship breaks down for alkali metal cations binding to carboxylate groups on protein surfaces (PNAS 10.1073/pnas.0902904106 – paper here). They argue that the picture is more complicated than such as simple physico-chemical law can express, involving the nature of the hydrated ion complex and the possibility of water-bridged interactions.

The mechanism of fast proton transport in water has been long debated, with the traditional view of Grotthus-like proton hopping along water chains now refined to a picture that tends to invoke intermediates of either the Eigen or Zundel ions (H3O+.3H2O or H5O2+). Rather similar considerations have been applied to the transport on hydroxide ions – are they just a mirror image of proton transport, or do they involve other ionic species such as H3O2-? Andrei Tokmakoff at MIT and his coworkers now present femtosecond pump-probe IR spectroscopic results that they say points to the significant involvement of a Zundel-like transition state in proton transfer in hydroxide solutions, in which a proton is delocalized between a hydroxide ion and a water molecule (S. T. Roberts et al., PNAS 10.1073/pnas.0901571106 – paper here).

Shuxun Cui at Southwest Jiaotong University in Chengdu has published a paper in which he speculates about the prebiotic implications of his recent single-molecule force spectroscopy work (some with Herman Gaub) on the structures of DNA in water and non-aqueous media (see for example JACS 128, 6636 (2006) and JACS 129, 14710 (2007)). He argues that double-stranded DNA can be seen as an adaptation to an aqueous environment (S. Cui, IUBMB Life 61, 860 (2009) – paper here). I have the strong sense that this, rather than Lawrence Henderson’s ‘fitness of the environment’, is the right way round to be examining this question.

Monday, June 22, 2009

Small molecules and protein folding

The role of small molecules – denaturants and osmolytes – in protein folding is much in need of a good review article (or have I missed one?). Julio Fernández and colleagues have used single-molecule force spectroscopy to look at how the osmolyte glycerol interacts with ubiquitin as the protein is mechanically unfolded (S. Garcia-Manyes et al., PNAS 10.1073/pnas.09020106 – not yet online). Glycerol stabilizes the protein against unfolding, and apparently also promotes hydrophobic collapse of the unfolded conformation. They think that while glycerol stabilizes the folded state via direct interaction with the protein, ethanol seems to exert a weaker stabilizing effect via an indirect interaction involving the disruption of ‘water structure’. The promotion of hydrophobic collapse in the presence of glycerol (which is not seen for ethanol) seems to be a separate effect, perhaps due to the enhanced destabilization of exposed hydrophobic surface due to the polar surface area of glycerol.

The electronic state of water molecules confined in a close-packed rodlike micelle lattice is significantly different from that in the gas and bulk liquid phases, according to Jan-Erik Rubensson and colleagues at Uppsala University (J. Gråsjö et al., J. Phys. Chem. B 113, 8201; 2009 – paper here). They have probed this question using soft X-ray absorption and emission, and say that the water molecules among micelles are stabilized relative to the bulk, perhaps because of interaction with the chloride counterions in solution.

Pablo Debenedetti and coworkers have also studied nanoconfined water, here in a slit-like space between two hydrophilic silica surfaces using MD simulations (S. R.-V. Castrillón et al., J. Phys. Chem. B 10.1021/jp9025392 – paper here). They find rotational slowing within 0.5 nm of the surfaces, and translational slowing within 1 nm.

The difference in hydrophobic interactions in acidic solutions relative to salt solutions is investigated by Greg Voth and colleagues (H. Chen et al., J. Phys. Chem. B 113, 7291; 2009 – paper here). They say that in acid (HCl) solution they see interactions between the hydrophobe surface and the hydrated protons, owing to the amphiphilic character of the latter. This could explain why hydrated protons are anomalous in the Hofmeister series, promoting solubilization of nonpolar solutes despite having a similar radius to salting-out cations such as potassium and ammonium.

And on matters Hofmeister, Bernd Rode and colleagues at the University of Innsbruck have carried out quantum simulations of the hydration of beryllium ions, and find that the tetrahedral first hydration shell has very slow exchange dynamics (S. S. Azam et al., J. Phys. Chem. B 10.1021/jp903536k – paper here). They refer to this as a strong ‘structure-forming’ behaviour – I can see what they mean, but does it invite confusion with the already confused issue of ‘structure-making’?

The dynamic Stokes shift – the slower decay of a frequency-shifted fluorescent probe molecule – close to protein surfaces relative to the bulk solution has been attributed in the past to much slower water motions in the hydration shell. But Bertil Halle and Lennart Nilsson question this interpretation in a new paper (J. Phys. Chem. B 113, 8210; 2009 – paper here). They say that the slower decay can be understood by a solvent polarization effect, and does not probe hydration dynamics at all.

Zoran Arsov at the Josef Stefan Institute in Ljubljana and colleagues report the weakening of hydrogen bonds in water confined between lipid bilayers, using a form of FTIR (Z. Arsov et al., ChemPhysChem 10.1002/cphc.200900185 – paper here). The water films separating bilayers in the lamellar phases (phospholipids DMPC and POPE) studied here are very thin – 2 and 0.6 nm respectively. So a disruption of bulk structure is presumably to be expected. They suggest that this perturbation may contribute to the attractive hydration force between the bilayers.

There has been a long debate, going back to Faraday and Tyndall, on whether ice has a liquid-like layer on its surface and if so, what this looks like. Xiao-Yang Zhu and colleagues at Minnesota have investigated this with interfacial force microscopy (M. P. Goertz et al., Langmuir 10.1021/la9001994 – paper here). They see a liquid-like layer tens of nanometres thick, but suggest that it is in fact viscoelastic.

Tuesday, June 16, 2009

More catching up

Changes in my circumstances have delayed this one, and of course the longer I delay, the worse it gets. I bring things somewhat up to date here, but there’s still more to come.

First, an ad: there is an RSC Faraday Discussion on ‘Wetting Dynamics of Hydrophobic and Structured Surfaces’ in Richmond, Virginia on 12-14 April 2010. Given the list of organizers and invited speakers, it is sure to be very good. Details are here.

More on water flow inside carbon nanotubes, which is attracting increasing interest because of the possibilities for water purification and desalination. John Thomas and Alan McGaughey at Carnegie Mellon find in MD simulations that the water structure changes significantly for tubes of 0.83 to 1.39 nm – from single chains to stacked pentagons and hexagons and finally to bulk-like (Phys. Rev. Lett. 102, 184502; 2009 – paper here). This seems to significantly affect the (pressure-driven) flow velocity in a non-monotonic way, particularly when the liquid has a layer-like profile.

Thomas Angel and his coworkers have looked at the roles of water molecules in photosensitive rhodopsin-like G protein-coupled receptors (T. E. Angel et al., PNAS 10.1073/pnas.0903545106 – paper here). They find that waters associated with highly conserved residues seem to be crucial to function, in particular providing the plasticity needed to transmit a signal from the retinal binding pocket to the intracellular surface.

Sow-Hsin Chen and coworkers find, using QENS, that lysozyme remains flexible (‘soft’) at low temperatures (210-240 K) when moderate pressure (around 1 kbar) is applied (X.-q. Chu et al., J. Phys. Chem. B 10.1021/jp900557w – paper here). Surprisingly, the dynamics under these conditions are actually faster than those under ambient conditions, and reflect those of the hydration water.

And speaking of low-temperature environments, David Wharton and Craig Marshall have outlined some of the survival strategies of Antarctic organisms in a nice brief review in J. Biol. 8, 39; 2009 – paper here. And Todd Sformo at the University of Alaska at Fairbanks has told me about a very interesting paper reporting an Arctic gnat that simultaneously uses freeze tolerance and freeze avoidance in different parts of its body – these strategies are usually mutually exclusive (T. Sformo et al., J. Compar. Physiol. B: Biochem. System. Envir. Physiol. 10.1007/s00360-009-0369-x; 2009 - paper here. (You see, this is the kind of nice stuff I fear I’m missing all the time…)

Huib Bakker’s group has used THz and femtosecond IR spectroscopy to study proton hydration (K. J. Tielrooij et al., Phys. Rev. Lett. 102, 198303; 2009 – paper here). They find that protons induce a drop in dielectric constant corresponding to an effect on 19 water molecules per proton. Four of these are involved in direct solvation, being irrotationally bound to the proton, but the others are perturbed by becoming implicated in proton motion.

Staying with proton transport, Greg Voth and colleagues have used the MS-EVB method to look at proton transfer in human carbonic anhydrase II (C. M. Maupin et al., JACS 10.1021/ja8091938 – paper here). The proton transfer here, between a zinc-bound OH group and the His64 residue, is the rate-limiting step, and involves a water cluster in the active site. There are some insights here into the ways proteins may use hydrophobic interfaces to control and facilitate proton transport. And Ana-Nicoleta Bondar at the University of California at Irvine and colleagues have studied how protons achieve long-distance transport in bacteriorhodopsin from the acceptor residue Asp85 to the extracellular proton release group (P. Phatak et al., JACS 10.1021/ja809767v – paper here). Bound water molecules in the active site are again implicated.

Prashanth Athri and W. David Wilson at Georgia State University show how interfacial water can help the DNA-binding agent DB921 to bind in the minor groove despite an imperfect geometric match (JACS 10.1021/ja809249h – paper here). These results might offer clues to exploiting water mediation in designing DNA-binding molecules.

More on urea and denaturation: Frank Gabel at the Institut de Biologie Structurale in Grenoble and colleagues have used SANS and SAXS to study the binding of urea to denatured ubiquitin (JACS 10.1021/ja9013248 – paper here). They find that acid-induced denaturation recruits about 20 urea molecules from solution to bind to the protein, supporting the view that these direct interactions between protein and denaturant are the cause of denaturation.

Some studies of water at lipid membranes. M. D. Fayer and colleagues at Stanford look at the hydration of AOT reverse micelles, compared to the lamellar phase, using ultrafast IR spectroscopy, and conclude that short-range, direct interactions with the head groups, rather than more general nanoconfinement effects, seen to be responsible for the orientiation retardation of water molecules (D. E. Moilanen et al., JACS 131, 8318; 2009 – paper here). And Zhancheng Zhang and Max Berkowitz at UNC have looked at the slowing of water orientational relaxation in the hydration layer of phospholipids bilayers using MD (J. Phys. Chem. B 113, 7676; 2009 – paper here). Berkowitz and Changsun Eun have also looked (via MD) at the hydration of lipid headgroups attached to two parallel graphene plates, as a model for interactions between bilayers (J. Phys. Chem. B 10.1021/jp901747s – paper here). They find a repulsive interaction between the plates that has three regimes, dependent on the plate separation. At small distances (0.75-1 nm) the repulsion is steric. At intermediate distances (1-1.6 nm) it results from dehydration of the head groups, and at large separations (1.7-2.4 nm) – well, I must be missing something in my rapid reading here, but all I can glean is that this is water-mediated too.

Alenka Luzar and colleagues have compared MD simulations of the hydration of monosodium glutamate with the recent neutron data from Sylvia McLain et al. (J. Phys. Chem. B 110, 21251; 2006). They find that the simulations could not reproduce the reduction in water-water correlations seen experimentally, pointing to some of the shortcomings of the classical potentials used (C. D. Daub et al., J. Phys. Chem. B. 113, 7687; 2009 – paper here). And Janusz Stangret and colleagues have characterized the hydration of carboxylate ions using FTIR spectroscopy (E. Gojlo et al., J. Phys. Chem. B 10.1021/jp811346x – paper here). They find that two water molecules induce symmetry-breaking of the carboxylate group, providing non-equivalent proton donors to the oxygen atoms.

Jacob Petrich and colleagues at Iowa State describe a new method for probing the dynamics of proteins – specifically, measuring the solvation correlation function – by monitoring the fluorescence from two coumarins with different lifetimes (S. Bose et al., J. Phys. Chem. B 10.1021/jp9004345 – paper here).

Benoît Roux at Chicago and coworkers consider the ‘topological control hypothesis’ for selective ion binding to proteins, which postulates that selectivity is controlled primarily by the number of ligands coordinating the ion – which can in turn be predicted from the average coordination structure in bulk water – and not from their chemical nature (H. Yu et al., J. Phys. Chem. B 10.1021/jp901233v – paper here). They find, perhaps not surprisingly, that this hypothesis has some serious limitations in predicting binding free energies.

Kelly Gaffney and coworkers at Stanford use ultrafast IR spectroscopy to look at hydrogen-bond dynamics in sodium perchlorate solution (S. Park et al., J. Phys. Chem. B 10.1021/jp9016739 – paper here). They find that the dynamics support an orientational jump model in which the making and breaking of H-bonds is the predominant control on reorientation times. MD simulations also indicate that the anion hydration shells have two distinct shells, and that the molecules in the inner shell donate one H-bond each to the perchlorate ion.

Nicolas Giovambattista, Peter rossky and Pablo Debenedetti have been trying to map out the phase behaviour of water confined between hydropholic, hydrophobic and heterogeneous plates at various temperatures and pressures (Phys. Rev. E 73, 041604; 2006 and J. Phys. Chem. C 11, 1323; 2007). They have now extended this work by looking at the effects of varying the T and P simultaneously between 220-300 K and –0.2 to 0.2 GPa (J. Phys. Chem. B 10.1021/jp9018266 – paper here). It’s hard to summarize all the information in this rich paper, but one general conclusion is that the plates become effectively less hydrophobic (the vapour phase is suppressed) as the temperature drops. An underlying motive for this work is to understand the pressure- and cold-denaturation of proteins and how this is tied up with invasion of hydrophobic cavities by water.

Wednesday, May 6, 2009

Catching up

I have been on travel and caught up with numerous other things for several weeks, which not only means I have been unable to post but also that I’ve doubtless missed some interesting things over the recent period. If so, apologies for that – and feel free to let me know! But note also that my email address at Nature won’t be active for much longer, as I am very shortly leaving my role there as a writer – I’m no longer able to keep that going alongside the other demands on my time. I will no doubt keep writing occasionally for Nature nonetheless, but will be henceforth reachable at

To business, and in no particular order from the stack of papers in my pile…

Martina Havenith and her coworkers at Bochum have continued their highly revealing work on hydration structures using terahertz spectroscopy. In a new paper (B. Born et al., JACS 131, 3752 (2009) – paper here) they look at the hydration networks around various small peptides at low hydration levels, and find that the collective motions of peptide + solvent that seem to characterize protein hydration shells disappear below a minimum number of hydration waters, which appears to be well below that required for monolayer coverage.

This issue of solvent-protein dynamical correlations is also studied by Nigel Scrutton and colleagues at the University of Manchester (D.J. Heyes et al., Angew. Chem. Int. Ed. 48, 3850; 2009 – paper here). They find that proton tunnelling in protochlorophyllide oxireductase, a light-driven enzyme involved in chlorophyll biosynthesis, requires protein motions that appear to be slaved to the solvent, whereas hydride transfer in this species does not. It seems significant that proton, but not hydride, transfer occurs only close to the protein’s glass transition temperature, when the protein-solvent coupling is likely to be most acute.

In my recent ChemPhysChem paper (here), I referred to some recent work on denaturants by Jeremy England and coworkers, inadvertently describing it as though it came out of Vijay Pande’s lab (J. L. England, V. S. Pande & G. Haran, JACS 130, 11854; 2008). Actually I gather it was done mostly at the lab of Gilad Haran at the Weizmann Institute in Israel. Apologies for that oversight. Gilad has told me of his recent work on protein collapse using single-molecule FRET (G. Ziv & G. Haran, JACS 131, 2942; 2009 – paper here; and G. Ziv et al., Phys. Chem. Chem. Phys. 11, 83; 2009 – paper here). The former paper argues that the action of denaturants operates primarily by modulating the collapse of the denatured state, rather than acting on the transition from molten globule to folded state.

A recent special issue of J. Phys. Chem. B on aqueous solutions and interfaces (113(13)) has given me a lot to catch up with. Sotiris Xantheas and Greg Voth give a nice overview here. Among the interesting papers therein, Greg profiles the case for the amphiphilic nature of the hydrated proton at the interface with hydrophobic media (S. Iuchi et al., J. Phys. Chem. B 113, 4017; 2009 – paper here); Janamejaya Chowdhary and Branka Ladanyi use MD to investigate the dynamics of hydrogen-bond making and breaking (113, 4045; 2009 – paper here), and Yves Rezus and Huib Bakker use femtosecond IR spectroscopy to look at how various amphiliphilic small molecules alter water structure in the bulk (113, 4038; 2009 – paper here). Rezus and Bakker find that trimethylamine-N-oxide seems to be special in the latter regard, increasing the rate of reorientation of mobile water, which might be interpreted as a sign that it increases the number of defects in the H-bonded network.

On much the same topic, Greg Voth and colleagues have used MD to look at the interactions of hydrophobic, nonpolar solvents in aqueous salt (NaCl) and acid (HCl) solution (H. Chen et al., J. Phys. Chem. B ASAP; paper here). They find unusual solvated-proton structures in the acid which are bound to the hydrophobes, again supporting the idea that the hydrated protons act as amphiphiles. This might explain why protons seem anomalous in the Hofmeister series. And Michael Brindza and Robert Walker at the University of Maryland use SHG to look at solvation mechanisms of small molecules (p-nitroanisole, indoline) at the interface of polar solids and various liquids, offering a broader context for understanding what water in particular does in such cases (JACS 131, 6207; 2009 – paper here). Alenka Luzar and her coworkers have used MD to look at the solvation of monosodium glutamate, making a comparison to experimental data on this system (C. D. Daub et al., J. Phys. Chem. B ASAP; paper here). On the whole the agreement is good with the neutron-diffraction data, but there are some important differences – for example, the simulations couldn’t reproduce the experimentally observed reduction in water-water correlations – that presumably point to inadequacies of using classical potentials. (Incidentally, my spellchecker tirelessly insists that ‘solvation’ should be ‘salvation’ – forgive me if I don’t fail to catch them all…)

Alenka’s comparison with experiment here relies heavily on Alan Soper’s neutron data and its analysis using empirical potential structure refinement, which was conducted for those data by Alan with Sylvia McLain and Anthony Watts. Sylvia has sent me a paper in which she, with Soper and Watts alongside Jeremy Smith and Isabella Daidone, use this same technique to deduce the nature of hydration and interaction between various amino acid dimmers (S. E. McLain et al., Angew. Chem. Int. Ed. 47, 9059; 2008 – paper here). This is extremely interesting, as it challenges the conventional view that the main driving force for association is the interaction of hydrophobic regions. In contrast, this study finds that it is the charged sites on the peptides that dominate the association, and that interaction decreases as hydrophobicity increases. Could the same apply in protein folding itself?

Sason Shaik and his collaborators have now published the full analysis of the role on internal waters in the action of cytochrome P450 StaP (Y. Wang et al., JACS ASAP; 2009 – paper here), the preliminary account of which I discussed in my CPC paper.

Anders Nilsson, Lars Pettersson and their coworkers have now published the work that I referred to in a Nature article last year, in which they challenge the view that X-ray and neutron diffraction data uniquely support the conventional tetrahedral-coordinate picture of water structure (K. T. Wikfeldt et al., J. Phys. Chem. B 113, 6246; 2009 – paper here). Needless to say, this remains highly controversial stuff.

Pedro de Pablo and colleagues at Madrid have reported some very striking results on the desiccation of viruses. They say that drying of two different viruses causes the ejection of DNA and, in one case, collapse of the remaining capsid owing to capillary forces of the water menisci inside (C. Carrasco et al., PNAS 106, 5475; 2009 – paper here).

A couple more papers on nanoconfined water, in hydrophobic but less explicitly biological environments. Sow-Hsin Chen and colleagues say that supercooled water, which shows a density minimum under confinement in hydrophilic mesoporous materials, has none such in a hydrophobic material (Y. Zhang et al., J. Phys. Chem. B ASAP: paper here). And Gene Stanley and colleagues look at the H-bond dynamics of TIP5P water in a hydrophobic nanopore slit (S. Han et al., Phys. Rev. E 79, 041202; 2009 – paper here). They say that the H-bonds are shorter-lived in this case than in the bulk, and the relaxation time is smaller, but the general qualitative behaviour is much the same (e.g. the temperature-dependence of the average H-bond lifetime, and non-exponential lifetime distributions).

Ahmed Zewail and Ding-Shyue Yang recently reported ultrafast electron crystallography of water at low temperatures (c. 150 K) at the surface of graphite (PNAS 106, 4122; 2009 – paper here). It’s not strictly relevant to water in biology, perhaps, but an issue that I’ve started to follow increasingly and which touches on water’s general ability to order at interfaces. The results imply that, perhaps contrary to expectation (especially when compared with hydrophobic H-terminated silicon), this hydrophobic surface doesn’t disrupt a highly ordered interfacial structure. It seems this may be because the graphite surface is stepped, which apparently allows it to template a cubic-ice structure.

I recently wrote a column for Nature Materials (8, 250; 2009 – see here) on hydrophobicity at larger scales than the molecular, and in particular on the discussions of wetting by water of nano- and microstructured surfaces via Wenzel or Cassie states. This is relevant to biology insofar as it bears on the question of wettability of, e.g. insect legs and lotus leaves. There is a nice recent simulation paper on the topic by Takahiro Koishi at the University of Fukui and colleagues (PNAS doi:10.1073/pnas.0902027106 – not yet online), which describes the different conditions (of surface topology and so forth) under which Wenzel and Cassie wetting exist, and the possibility of their coexistence. And Abraham Marmur has flagged up his paper from last year (Langmuir 24, 7573; 2008 – paper here) in which he too looks at general geometric considerations that might promote such high-contact-angle states.

Here’s an intriguing thing that had escaped my notice until now: it seems that voltages may be generated in carbon nanotubes along the tube axis when water flows through them (see S. Ghosh et al., Science 299, 1042; 2003 – paper here). That’s suggestive from the perspective that views nanotubes as simple analogues of hydrophobic protein channels. Already it seems that this idea has been explored for power conversion (Y. C. Zhao et al., Adv. Mater. 20, 1772; 2008). Now Quanzi Yuan and Ya-Pu Zhao of the Institute of Mechanics in Beijing offer an explanation for the phenomenon in terms of the alignment of water molecules in the hydrogen-bonded chain within the nanotube (JACS ASAP; paper here).

More on this general topic from Bo Liu at the Graduate University of the Chinese Academy of Sciences in Beijing and coworkers, who describe simulations of a model system in which end-functionalized short carbon nanotubes are embedded in a phospholipid bilayer as mimics of aquaporin (B. Liu et al., Nano Lett. 9, 1386; 2009 – paper here). Two charges are placed near the tube midpoints, to simulate the NPA region of aquaporin where water selectivity and proton gating is thought to happen. But proton conduction couldn’t be studied explicitly in these classical MD simulations. That aside, the device seems to work more or less as planned, in theory. But can we make it?

Well, that’s not the end of it, but my desk looks a whole lot better.

Friday, March 13, 2009

There’s more to life than sequence

I have been meaning for some time to write about an interesting paper in JACS by Naoki Sugimoto’s group in Kobe. It found its way into an article that I wrote this week for Nature’s online news. So I’ve decided to simply post this article here – it’s not all strictly relevant to water in biology, but hopefully is interesting stuff anyway. This is the version before editing, which has more detail.

Shape might be one of the key factors in the function of mysterious ‘non-coding’ DNA.

Everyone knows what DNA looks like. Its double helix decorates countless articles on genetics, has been celebrated in sculpture, and was even engraved on the Golden Record, our message to the cosmos on board the Voyager spacecraft.

The entwined strands, whose form was deduced in 1953 by James Watson and Francis Crick, are admired as much for their beauty as for the light they shed on the mechanism of inheritance: the complementarity between juxtaposed chemical building blocks on the two strands, held together by weak ‘hydrogen’ bonds like a zipper, immediately suggested to Crick and Watson how information encoded in the sequence of blocks could be transmitted to a new strand assembled on the template of an existing one.

With the structure of DNA ‘solved’, genetics switched its focus to the sequence of the four constituent units (called nucleotide bases). By using biotechnological methods to deduce this sequence, they claimed to be ‘reading the book of life’, with the implication that all the information needed to build an organism was held within this abstract linear code.

But beauty has a tendency to inhibit critical thinking. There is now increasing evidence that the molecular structure of DNA is not a delightfully ordered epiphenomenon of its function as a digital data bank but a crucial – and mutable – aspect of the way genomes work. A new study in Science [1] underlines that notion by showing that the precise shape of some genomic DNA has been determined by evolution. In other words, genetics is not simply about sequence, but about structure too.

The standard view – indeed, part of biology’s ‘central dogma’ – is that in its sequence of the four fundamental building blocks (called nucleotide bases) DNA encodes corresponding sequences of amino-acid units that are strung together to make a protein enzyme, with the protein’s compact folded shape (and thus its function) being uniquely determined by that sequence.

This is basically true enough. Yet as the human genome was unpicked nucleotide base by base, it became clear that most of the DNA doesn’t ‘code for’ proteins at all. Fully 98 percent of the human genome is non-coding. So what does it do?

We don’t really know, except to say that it’s clearly not all ‘junk’, as was once suspected – the detritus of evolution, like obsolete files clogging up a computer. Much of the non-coding DNA evidently has a role in cell function, since mutations (changes in nucleotide sequence) in some of these regions have observable (phenotypic) consequences for the organism. We don’t know, however, how the former leads to the latter.

This is the question that Elliott Margulies of the National Institutes of Health in Bethesda, Maryland, Tom Tullius of Boston University, and their coworkers set out to investigate. According to the standard picture, the function of non-coding regions, whatever it is, should be determined by their sequence. Indeed, one way of identifying important non-coding regions is to look for ones that are sensitive to sequence, with the implication that the sequence has been finely tuned by evolution.

But Margulies and colleagues wondered if the shape of non-coding DNA might also be important. As they point out, DNA isn’t simply a uniform double helix: it can be bent or kinked, and may have a helical pitch of varying width, for example. These differences depend on the sequence, but not in any straightforward manner. Two near-identical sequences can adopt quite different shapes, or two very different sequences can have a similar shape.

The researchers used a chemical method to deduce the relationship between sequence and shape. They then searched for shape similarities between analogous non-coding regions in the genomes of 36 different species. Such similarity implies that the shapes have been selected and preserved by evolution – in other words, that shape, rather than sequence per se, is what is important. They found twice as many evolutionarily constrained (and thus functionally important) parts of the non-coding genome than were evident from trans-species correspondences using only sequence data.

So in these non-coding regions, at least, sequence appears to be important only insofar as it specifies a certain molecular shape and not because if its intrinsic information content – a different sequence with the same shape might do just as well.

That doesn’t answer why shape matters to DNA. But it suggests that we are wrong to imagine that the double helix is the beginning and end of the story.

There are plenty of other good reasons to suspect that is true. For example, DNA can adopt structures quite different from Watson and Crick’s helix, called the B-form. It can, under particular conditions of saltiness or temperature, switch to at least two other double-helical structures, called the A and Z forms. It may also from triple- and quadruple-stranded variants, linked by different types of hydrogen-bonding matches between nucleotides. One such is called Hoogsteen base-pairing.

Biochemist Naoki Sugimoto and colleagues at Konan University in Kobe, Japan, have recently shown that, when DNA in solution is surrounded by large polymer molecules, mimicking the crowded conditions of a real cell, Watson-Crick base pairing seems to be less stable than it is in pure, dilute solution, while Hoogsteen base-pairing, which favours the formation of triple and quadruple helices, becomes more stable [2-4].

The researchers think that this is linked to the way water molecules surround the DNA in a ‘hydration shell’. Hoogsteen pairing demands less water in this shell, and so is promoted when molecular crowding makes water scarce.

Changes to the hydration shell, for example induced by ions, may alter DNA shape in a sequence-dependent manner, perhaps being responsible for the sequence-structure relationships studied by Margulies and his colleagues. After all, says Tullius, the method they use to probe structure is a measure of “the local exposure of the surface of DNA to the solvent.”

The importance of DNA’s water sheath on its structure and function is also revealed in work that uses small synthetic molecules as drugs that bind to DNA and alter its behaviour, perhaps switching certain genes on or off. It is conventionally assumed that these molecules must fit snugly into the screw-like groove of the double helix. But some small molecules seem able to bind and show useful therapeutic activity even without such a fit, apparently because they can exploit water molecules in the hydration shell as ‘bridges’ to the DNA itself [5]. So here there is a subtle and irreducible interplay between sequence, shape and ‘environment’.

Then there are mechanical effects too. Some proteins bend and deform DNA significantly when they dock, making the molecule’s stiffness (and its dependence on sequence) a central factor in that process. And the shape and mechanics of DNA can influence gene function at larger scales. For example, the packaging of DNA and associated proteins into a compact form, called chromatin, in cells can affect whether particular genes are active or not. Special ‘chromatin-remodelling’ enzymes are needed to manipulate its structure and enable processes such as gene expression of DNA repair.

None of this is yet well understood. But it feels reminiscent of the way early work on protein structure in the 1930s and 40s grasped for dimly sensed principles before an understanding of the factors governing shape and function transformed our view of life’s molecular machinery. Are studies like these, then, a hint at some forthcoming insight that will reveal gene sequence to be just one element in the logic of life?


1. Parker, S. C. J. et al., Science Express doi:10.1126/science.1169050 (2009). Paper here.
2. Miyoshi, D., Karimata, H. & Sugimoto, N. J. Am. Chem. Soc. 128, 7957-7963 (2006). Paper here.
3. Nakano, S. et al., J. Am. Chem. Soc. 126, 14330-14331 (2004). Paper here.
4. Miyoshi, D. et al., J. Am. Chem. Soc. doi:10.1021/ja805972a (2009). Paper here.
5. Nguyen, B., Neidle, S. & Wilson, W. D. Acc. Chem. Res. 42, 11-21 (2009). Paper here.

Tuesday, March 3, 2009

Making sense of solvent slaving

In my previous post I mentioned work by Pablo Debenedetti on ‘toy models’ of water. The places to look are: Buldyrev et al., PNAS 104, 20177 (2007) (here) for the solvation thermodynamics of ‘spherical’ water; and Patel et al., Biophys. J. 93, 4116 (2007) (here) and J. Chem. Phys. 128, 175102 (2008) (here) for water-explicit lattice models of proteins.

And in discussing recent work on the mechanism of urea-induced protein denaturation, I neglected to mention Bruce Berne’s PNAS paper from late last year with Ruhong Zhou, Dave Thirumalai and Lan Hua (105, 16928; paper here). That paper on MD simulations for lysozyme anticipated the more recent work showing that denaturation seems to be caused by direct urea-protein interactions: the urea displaced water from the first hydration shell and penetrates into the hydrophobic core to give a ‘dry globule’.

The notion that protein dynamics are ‘slaved’ to those of the hydration shell has been floating around for some time now. Hans Frauenfelder and colleagues have now brought considerable focus to the idea (PNAS doi:10.1073/pnas.0900336106; paper here) with dynamical measurements using dielectric spectroscopy, Mossbauer and neutron scattering. They find that large-scale protein motions follow the fluctuations of the solvent and are dependent on solvent viscosity. There are two classes of fluctuation in the solvent, alpha and beta, with different timescales. It seems that the former are ‘structural’ in nature and control protein shape; the latter are those to which the protein’s internal motions are slaved.

Jianxing Song at the National University of Singapore, who I met in Hangzhou, has sent me three papers on the intriguing solubilisation of ‘water-insoluble’ proteins in pure water. He and his coworkers found this effect in 2006 for a range of diverse proteins (M. Li et al., Protein Science 15, 1835 (2006) – paper here; M. Li et al., Biophys. J. 91, 4201 (2006) – paper here). They attributed it to the tendency of the ‘insoluble’ proteins to form partially folded states with many exposed hydrophobic residues, such that only a very low ionic strength is sufficient to screen out repulsive interactions and cause aggregation. In pure water, however, those electrostatic interactions remain sufficiently strong to suppress aggregation and precipitation. Jianxing has now provided an overview of this work, expanding on the importance of pH for this effect, in FEBS Letters (doi:10.1016/j.febslet.2009.02.022; paper here).

Roberto Righini at the University of Florence and colleagues have used IR spectroscopy to identify and quantify the various aqueous species that solvate the polar heads of phospholipids in bilayers (V. V. Volkov et al., J. Phys. Chem. B doi:10.1021/jp806650c; paper here). And Davide Donadio and coworkers in California have shown how electronic charge fluctuations show up in the IR spectra of water close to nonpolar surfaces (here graphite) (D. Donadio et al., J. Phys. Chem. B doi:10.1021/jp807709z; paper here). Still in that neck of the woods, Chuan-Shan Tian and Ron Shen have used sum-frequency-IR spectroscopy to sort out the nature of hydrogen-bonding at the air-water interface (JACS 131, 2791 (2009) – paper here). And Heather Allen and colleagues at Ohio State University have used this and other spectroscopic techniques to study hydration structure of the air-water interface for various divalent nitrates (M. Xu et al., J. Phys. Chem. B doi:10.1021/jp806565a; paper here).

Haiping Fang in Shanghai continues his exploration of how water and solutes can be manipulated within the confinement of carbon nanotubes. He and colleagues now show, using MD simulations, how a single charge outside a nanotube can be used to move a hydrated peptide inside it, regardless of whether the peptide itself is charged (P. Xiu et al., JACS 131, 2840 (2009) – paper here). This is due to the dipole-orientational ordering of the water molecules caused by confinement and interaction with the external charge.

There’s more, but later.