Conformation unfolded proteins

The biologiccJ function of a protein or peptide is often intimately dependent upon the conformation(s) that the molecule can adopt. In contrast to most synthetic polymers where the individual molecules can adopt very different conformations, a protein usually exists in a single native state. These native states are found rmder conditions typically found in Uving cells (aqueous solvents near neutred pH at 20-40°C). Proteins can be unfolded (or denatured) using high-temperature, acidic or basic pH or certain non-aqueous solvents. However, this unfolding is often reversible cind so proteins can be folded back to their native structure in the laboratory. 

We will discuss three different approaches to engineer a more thermostable protein than wild-type T4 lysozyme, namely (1) reducing the difference in entropy between folded and unfolded protein, which in practice means reducing the number of conformations in the unfolded state, (2) stabilizing tbe a helices, and (3) increasing the number of bydropbobic interactions in tbe interior core. 

Another interesting application area of PHMD simulations is to investigate electrostatic interactions in the unfolded states of proteins. A traditional view that unfolded proteins adopt random conformational states that are devoid of electrostatic and hydrophobic interactions, are recently challenged by experimental data [20, 69], REX-CPHMD folding simulations of the 35 residue C-terminal subdomain of the villin headpiece domain revealed a significant deviation from the standard pKa values for several titratable residues. Additional simulations, in which a charged group is neutralized confirmed the existence of specific electrostatic interactions in the unfolded states (JK and CLB, manuscript in preparation). 

The conformational plasticity supported by mobile regions within native proteins, partially denatured protein states such as molten globules, and natively unfolded proteins underlies many of the conformational (protein misfolding) diseases (Carrell and Lomas, 1997 Dobson et al., 2001). Many of these diseases involve amyloid fibril formation, as in amyloidosis from mutant human lysozymes, neurodegenerative diseases such as Parkinson s and Alzheimer s due to the hbrillogenic propensities of a -synuclein and tau, and the prion encephalopathies such as scrapie, BSE, and new variant Creutzfeldt-Jacob disease (CJD) where amyloid fibril formation is triggered by exposure to the amyloid form of the prion protein. In addition, aggregation of serine protease inhibitors such as a j-antitrypsin is responsible for diseases such as emphysema and cirrhosis. 

Fluorescence correlation spectroscopy (FCS) measures rates of diffusion, chemical reaction, and other dynamic processes of fluorescent molecules. These rates are deduced from measurements of fluorescence fluctuations that arise as molecules with specific fluorescence properties enter or leave an open sample volume by diffusion, by undergoing a chemical reaction, or by other transport or reaction processes. Studies of unfolded proteins benefit from the fact that FCS can provide information about rates of protein conformational change both by a direct readout from conformation-dependent fluorescence changes and by changes in diffusion coefficient. 

IS POLYPROLINE II A MAJOR BACKBONE CONFORMATION IN UNFOLDED PROTEINS ... 

Tanford (1968) reviewed early studies of protein denaturation and concluded that high concentrations of Gdm-HCl and, in some cases, urea are capable of unfolding proteins that lack disulfide cross-links to random coils. This conclusion was largely based on intrinsic viscosity data, but optical rotation and optical rotatory dispersion (ORD) [reviewed by Urnes and Doty (1961) ] were also cited as providing supporting evidence. By these same lines of evidence, heat- and acid-unfolded proteins were held to be less completely unfolded, with some residual secondary and tertiary structure. As noted in Section II, a polypeptide chain can behave hydrodynamically as random coil and yet possess local order. Similarly, the optical rotation and ORD criteria used for a random coil by Tanford and others are not capable of excluding local order in largely unfolded polypeptides and proteins. The ability to measure the ORD, and especially the CD spectra, of unfolded polypeptides and proteins in the far UV provides much more incisive information about the conformation of proteins, folded and unfolded. The CD spectra of many unfolded proteins have been reported, but there have been few systematic studies. 

What is the predominant secondary structure in unfolded proteins In thermally unfolded proteins, there appears to be no predominant secondary structure. Individual residue conformations are distributed over the Pn, a, and f regions that constitute energy minima for single residues in aqueous solution (Section II,B). Still, since there is increasing evidence that the Pn conformation is at the global minimum, Pn conformers must be the most abundant in this ensemble. Short Pn- and o -helices and /1-strands will be present, but will rarely exceed two or three residues in length. 

It is common today to consider a peptide or an unfolded protein to be random coil —i.e., a blend of conformations—if their CD spectra resemble the traditionally accepted random coil CD spectrum (Blanco and Serrano, 1995 Dyson et al., 1992 Jimenez et al., 1993 Luisi et al., 1999 Munoz et al., 1995 Najbar et al., 2000 Viguera etal., 1996). As we argue here, this CD spectrum does not correspond to random coil at all. Theoretical calculated spectra representing blends of different conformations have been published by Krimm s group (Ronish and Krimm, 1972, 1974). While the results are uncertain because of the unknown compositional coefficients for each conformation in the blend, these spectra are clearly unrelated to that of Pn and many unfolded peptides and proteins. We need to reconsider what kind of CD spectrum is an appropriate representative for a true random coil. 

---