Folding and Denaturation
Protein folding is driven by the aqueous environment, particularly the hydrophobic interactions,
due to the unfavorable entropy decrease (mostly translational
[686a, 1577]) forming a large surface
area of non-polar groups with water. Consider a water
molecule next to a surface to which it cannot hydrogen bond.
The incompatibility of this surface with the low-density
water that forms over such a surface 
encourages the surface minimization that drives the proteins'
tertiary structure formation (for example, see ). An alternative view of this process is that the protein conformation tends to minimize the disruption of the water matrix . Such hydrophobic collapse is
necessarily accompanied and guided by (secondary) structural
hydrogen-bond formation between favorable peptide linkages
in parallel with their desolvation .a A driving force for this, in crowded intracellular
environments, is the release of water to be available
for the hydration of other solutes and maximizing its entropy [686b]. The folding route is controlled
by the desolvation barriers 
and aided and directed by water-mediated contacts zipping up neighboring residues . Similar
factors help organize proteins involved in quaternary and
equilibrium cluster formation, where each water-mediated interaction
has been estimated to contribute an average of 4.4 kJ mol-1 to protein-protein interface stabilization .
Water is thus intimately involved in guiding protein folding
and needs to be involved in protein structural prediction
The importance of subtle hydration forces is shown in the α-helix to β-sheet
conformational transition that accompanies the racemic self-assembly
of polylysine .
Opposite is shown schematic potential energy funnel for the folding of proteins without sufficient water present. it highlights the many barriers to the preferred minimum energy structure on the folding pathway. There are numerous local minima that might trap the protein in an inactive three-dimensional molecular conformation.The top rim represents the high energy of the unfolded protein with folding lowering the energy towards a minimum energy structure that is at the bottom of the funnel. It should be noted that these funnels represent three-dimensional landscapes, whereas the actual energy landscapes are multidimensional.
When a protein is fully hydrated, the potential energy landscape is seen to be considerably smoothed.a Under conditions of sufficient hydration, this allows proteins to attain their active minimum-energy conformation in a straightforward and rapid manner. The potential energy barriers are lowered and smoothed due to the ease with which water molecules can lubricate the movement of the amino acid backbone and side groups by the rapid formation and exchange of hydrogen bonds. Similar effects may be seen on the activity of enzymes with hydration  although complete hydration is unnecessary for some activity to be evident. Under physiological conditions inside cells the water is known to be more ordered (see intracellular water). Such water promotes both the folding rate and stability of the protein  even further.
Although indicated as such in the cartoons, there is not one 'minimum' structure but a collection of substates with small energetic differences. Jumps between these substates, eased by hydration, allows and determines the flexibility that the protein needs for its biological actions.
As the amide I stretch vibration (~1680 cm-1) is similar to
liquid water's bend vibration (v2, ~1645 cm-1), transfer
of energy from water hydrogen-bonded to protein asparagine
and glutamine groups is facilitated .
This explains the increased structural instability of
proteins containing greater numbers of surface asparagine
and glutamine residues and, in particular, is of relevance
to the α-helix - β-sheet
structural instability in prions .b
Although the native state of a protein resides
at a minimum on the potential energy surface, there is no
reason to suppose that this structure is the global minimum
free energy structure as its folding route is a guided, rather
than random, process. It is clear that other structures with lower minima exist, such as those often irreversibly produced on denaturation using intermolecular interactions .
Compatible solutes (osmolytes, for example, betaine), that stabilize the surface low-density water and
increase the surface tension, will also stabilize the protein's
structure (see also the Hofmeister
effect and the solubility of
Many proteins are glycosylated with increased solubility. The role of the carbohydrate groups has been debated for many years. It now appears that this increased solubility is mainly as the low intermolecular interaction between surface glycans reduces the tendency for aggregation (and crystallization) rather than the glycan groups increasing the interactions with water . Rather unexpectedly, deglycosylated proteins appear to have stronger interaction with water (by weight) and more extensive water binding (by molecule) than their glycoproteins ; perhaps because some carbohydrate hydroxyl groups replace several of the polypeptide surface water interactions.
[Back to Top ]
Proteins may form crystals when precipitated slowly from an aqueous solution (e.g. of ammonium sulfate). Slow precipitation is required to produce small numbers of larger crystals rather than very large numbers of small crystals. Crystals of undenatured proteins for structural analysis are best formed with water molecules retained within the crystal lattice. Crystallisation of native proteins appears to have a three-step mechanism involving nucleation, in which mesoscopic metastable protein clusters of dense liquid serve as precursors to the ordered crystal nuclei followed by crystal growth . This processseems to involve an aqueous biphasic separation and fits nicely with the two-state structuring in liquid water, where the crystallisation takes place within the dense phase.
Protein denaturation involves a change in the protein structure (generally an unfolding) with the loss of activity. Water is critical, not only for the correct folding of proteins
but also for the maintenance of this structure. Heat denaturation
and loss of biological activity has been linked to the breakup
of the 2-D-spanning water network (see above) around the protein
 (due to increasing hydrogen bond breakage with temperature),
which otherwise acts restrictively on protein vibrational
The free energy change on folding or unfolding is due to the
combined effects of both protein folding/unfolding and hydration
changes. These compensate to such a large extent that the
free energy of stability of a typical protein is only 40-90
kJ mol-1 (equivalent to very few hydrogen bonds),
whereas the enthalpy change (and temperature times the entropy
change) may be greater than ±500 kJ mol-1 different. There are both enthalpic and entropic contributions
to this free energy that change with temperature and so give
rise to heat denaturation and, in some cases, cold denaturation.
Protein unfolding at higher temperatures (heat denaturation) is easily understood but
the widespread existence of protein unfolding at low temperatures
is surprising, particularly as it is unexpectedly accompanied
by a decrease in entropy . Heat denaturation is endothermic (on heating) but cold denaturation is exothermic (on cooling) .
The free energy on going from the native
(N) state to the denatured (D) state is given by
The overall free energy change (
depends on the combined effects of the exposure of the
interior polar and non-polar groups and their interaction
with water together with the consequential changes in
the water-water interactions on
The graph is meant to be indicative only. Denaturation
is only allowed when is negative; its rate is then dependent circumstances
and may be fast or immeasurably slow. On heat denaturation and are generally both positive but on cold denaturation they are both negative.
The midpoint temperatures of both heat and cold denaturation
may be determined from peaks in the temperature dependence
of the heat capacity, where additional heat is being absorbed
by the intermediate structures.
The enthalpy of transfer of polar groups from the protein
interior into water is positive at low temperatures and negative
at higher temperatures .
This is due to the polar groups creating their own ordered
water, which generates a negative enthalpy change due to the
increased molecular interactions. Balanced against this is
the positive enthalpy change as the pre-existing water structure
and the polar interactions within the protein both have to
be broken. As water naturally has more structure at lower
temperatures, the breakdown of the water structure makes a
greater positive contribution to the overall enthalpy at lower
In contrast, the enthalpy of transfer of non-polar groups
from the protein interior into water is negative below about
25 °C and positive above .
At lower temperatures, non-polar groups enhance pre-existing
order such as the clathrate-related ES structure , discussed elsewhere, generating enthalpy
but this effect is lost with increasing temperature, as any
pre-existing order is also lost. At higher temperatures, the
creation of these clathrate structures requires an enthalpic
input. Thus, there is an overall positive enthalpy of unfolding
at higher temperatures. An equivalent but alternative way
of describing this process is that at lower temperatures the
clathrate-type structure optimizes multiple van der Waals
molecular interactions whereas at higher temperatures such
favorable structuring is no longer available.
At ambient temperature, the entropies of hydration of both
non-polar and polar groups are negative 
indicating that both create order in the aqueous environment.
However these entropies differ with respect to how they change
with increasing temperature. The entropy of hydration of non-polar
groups increases through zero with increasing temperature,
indicating that they are less able to order the water at higher
temperatures and may, indeed, contribute to its disorder by
interfering with the extent of the hydrogen-bonded network
and allowing an easier molecular rotation of water. Also,
there is an entropy gain from the greater freedom of the non-polar
groups when the protein is unfolded. In contrast, the entropy
of hydration of polar groups decreases, becoming more negative
with increasing temperature, as they are able to create ordered
hydration shells even from the more disordered water that
exists at higher temperatures. A consequence is that the water is more ordered around hydrophilic groups, compared with just water, as the temperature is raised and that this hydrophilic hydration has negative heat capacity .
Overall, protein stability depends on the balance between
these enthalpic and entropic changes. For globular proteins,
the ΔG of unfolding has a maximum
10-30 °C, decreasing both colder and hotter through zero
with the thermodynamic consequences of both cold and heat
denaturation. The hydration of the internal non-polar groups
is mainly responsible for cold denaturation as their energy
of hydration is greatest when cold. Thus, it is the increased
natural structuring of water at lower temperatures that causes
cold destabilization of proteins in solution (that is, the entropic cost of denaturation, due to the structuring
of the water molecules around the exposed groups, is reduced) .
An equivalent alternative view is that the hydrophobic interactions increase as the temperature is raised from a low value, such that the extended polypeptide chain, present at very low temperatures, folds up to produce an active globular protein so releasing water molecules to the bulk environment.
Heat denaturation is primarily due to the increased entropic
effects of the non-polar residues (that is, the increased
entropy gain of the unfolded chain is not much reduced by
the small amount of entropy loss caused to the solute). Although
both processes have been reported to lead to irreversible
changes, which often occur cooperatively, cold inactivation
in supercooled water is usually likely to be reversible and
it is any ice crystal formation that leads to observed irreversible
effects. Interestingly, proteins from thermophilic organisms
tend to have higher amounts of non-polar residues and lower
amounts of polar residues when compared to comparable proteins
from mesophilic organisms .
This is related to decreased bound water around thermophilic enzymes in crystals and solution  as part of their strategy for stability
(other factors being increased salt bridges and main chain
hydrogen bonds) .
Protein stability has been directly tied to the equilibrium
structuring of water between low-density and higher density
forms [210, 416, 1481]
(see also). This provides
an equivalent but alternative way of looking at the above
analysis. Opposite shows a representation of the pressure-temperature (P/T) phase diagram for proteins showing heat-, cold- and pressure-denaturation . The diagram is also representative of the solubility of other polymers, such as starch, the aggregation of non-polar solutes  and is related to the pressure-temperature relationship of the thermal expansion of water .
Denaturation may be effectively treated as increased
solubility of the unfolded form in a manner similar to that
given in the treatment of the anomalous
solubility behavior of non-polar gases. Thus, protein aggregates and amyloid fibrils (such as are found in prion and Alzeimer's diseases) may be dissolved on cooling or under high pressure .
The effect of
pH on denaturation (for example, low pH causing easier heat
may be understood by recognizing that extremes of pH cause
an increase in higher density clustering which may be partially
reversed by the presence of non-ionic
There is a change in volume with denaturation ()
varying with the protein concerned but typically from negative
(that is, overall volume of water and protein is smaller
on denaturation) at low temperatures to slightly positive
at high temperatures. This is due to the released nonpolar
residues producing less-dense water (for example, ES)
at low temperatures but less able to do this at high temperature
whereas the released polar groups cause a greater increase
in density at low temperatures, due to their destruction of
the low-density water, than at high temperatures, where there
is less destroyable low-density water. A small pressure increase
may stabilize the protein against both cold and heat denaturation.
At low temperatures, a small pressure increase reduces the
size of the enthalpic contribution of non-polar group hydration
due to the reduced aqueous structuring. At high temperatures
the enthalpic cost of hydrating these non-polar groups is
increased when under a small pressure increase, and this may help the increased thermal stability of proteins seen in crowded environments . Under higher
pressure, proteins take up water into empty cavities . This
penetrating water eases the process of denaturation by destabilizing
the internal links. The negative volume change at higher pressures, due to cavity filling,
helps shift the enthalpy change in favor of denaturation across
the temperature range but has been particularly noted at low temperatures, in line with the Figure above (for further discussion see ).
Elastin is an important protein with properties governed by its interactions with water. It consists of a high proline, high glycine hydrophobic chain (for example, typical section PGVGV) that cannot form regular α-helices or β-sheets but does form an extended structure probably containing β-spiral sections with most of the peptide links hydrogen bonded to water . On warming the structure undergoes a structural collapse due to the difficulty in maintaining the low-density clathrate structures around its hydrophobic groups.
[Back to Top ]
a Correct folding is also aided by the molecular chaperones (chaperonins) , where the chaperonin increases the density and hydration of the water surrounding the misfolded protein to drive correct re-folding . [Back]
bPrion fibril formation has also been explained as being due to the increase of water trimers and hydrophobicity of the Mn-linked (as opposed to the normal Cu-linked) protein in solution. . Water release has an important effect on the rate of initial fibril formation, due to the slow rate with which it is expelled from the hydrophilic amide side chains . [Back]