[Prev][Next][Index][Thread]
WSN: free energy of folding, contact energies (fwd)
Sender: bruce_bush@merck.com (Bruce Bush)
Subject: WSN: free energy of folding, contact energies
Rafael Szeinfeld writes:
> I'm doing some montecarlo simulations on the HP protein folding
> model of Ken Dill. Until the moment I'm just writing a program that
> reproduces his model. ... if we are assigning a unit[y]
> of energy to the hydrophobic contact (H-H), what would be the energy of a
> polar contact (P-P) and hydrophobic-polar (H-P) ?
> *E-MAIL : szeinfel@snfma1.if.usp.br | Sao Paulo - SP - Brazil *
His original question was: what is the experimental free energy of transfer
of amino acids between solvents?
Shawn Huston correctly points out that contact energy and free energy of
transfer are very different, though related, questions.
Also, the "energy of a [polar|nonpolar] contact"
within an enzyme is *not* necessarily the best number to use in
a folding simulation. The papers of Dill et al probably discuss their
choice of P-P and H-P and H-H parameters. If you're trying to reproduce
their work you should certainly start with their parameters.
Hundreds of papers have been published on the subjects. Here are just a few.
There are books by Ben-Naim (_Hydrophobic Interactions_, 1980) and others.
a) Free energy of transfer or of vaporization (solubility)
-- McAuliffe, C (1966) J Phys Chem 70, 1267-75
measured water solubilities of hydrocarbons (alkanes).
In principle these give free energies of transfer to water
from the pure (neat) hydrocarbon. From these data, ...
-- ... Reynolds, J; Gilbert, D; and Tanford, C. (1974)
ProcNatAcadSci(USA) 71,2925-7
concluded that each CH2 group adds 884 cal/mole to trnsfer free energy.
-- Cabani, S. et al (1981) J. Solution Chemistry 10*8) 563-595
tabulate hundreds of vapor - aqueous solution thermodynamic
properties, mostly enthalpies of transfer but also free energies.
From these they derive "group contributions" for some of the
functional groups that appear in many of the test compounds.
-- Wolfenden, R. et al, Biochemistry (1981) 20, 849-55
"Affinities of amino acid _side chains_ for solvent water"
measured vapor-aqueous solution equilibria for the side chains
only. Ala: +1.94 kcal/mole; Val 1.99; Leu +2.28; Ser -5.06; Tyr -6.11
-- Creighton, T, _Proteins_ (p.142 of the 1984 edition) summarizes work
of Wolfenden and others on both "hydrophilic energies" (vapor to
water equilibria) and "hydrophobic energies" (organic solvent to
water equilibria).
-- Ben-Naim, A. in several papers (1978 and after) disputes the use of
standard state used by Wolfenden and others. He and Y. Marcus
tabulate free energies by their analysis in "Solvation thermodynamics
of nonionic solvents" (1984), J. Chem. Phys 81(14), 2016-27).
b) Pairing energy versus transfer energy
Some authors claim that these are very different - for example that two
small nonpolar solutes don't necessarily make a stronger pair in water than
they would in vacuum. See Wood, R. and Thompson,P.T. (1990) PNAS 87, 946.
Other authors disagree. Some of the disagreements are discussed in a
review paper by Bloekzijl, W. and Engberts, J.B.F.N. (1993)
Angew. Chem. Int. Ed. Engl. 32, 1545-79 with the remarkable title
"Hydrophobic effects. Opinions and facts." The 371 references in this
article are a good starting point to the literature!
c) Contribution to folding stability
-- Starting with Kauzmann,W. (Adv. Protein Chem. 14,1), most authors
have assigned the driving force of protein folding to a hydrophobic
interaction, or to the formation of a hydrophobic core. But Privalov,P
and Gill,S (1988), Adv. Prot. Chem. 39, 191,argued the opposite:
that hydrophobic interactions are actually destabilizing
-- a classic reference is Tanford, C (1963?), J. Am Chem Soc 84, 4240-7
"Contribution of hydrophobic interactions to the stability of
the globular conformation of proteins". He assigns to ( buried )
side chains a contribution to the stability. For example,
Ile 2970 cal/mole; Pro 2600; Thr 440 ; Trp 3000 ; Lys 1500 (!)
-- Before choosing numbers you must think carefully whether there *is such a
thing* as one set of additive contributions to folding free energy.
Would you expect overall stability to be just a sum of contributions?
Ask the question the other way: if you measure the stability change
from deleting each side chain, can you add up these values (plus the
strength of mainchain hydrogen bonds) to get total folding energy?
In short, this is much more than a programming exercise! I hope you can
discuss the literature with your colleagues, and then decide what's
worth calculating. Good luck!
Bruce_bush@merck.com Merck Research Labs, Rahway NJ 07065 USA