QUESTIONS: alpha-helix "signals" in proteins

Simon Brocklehurst Bioc smb18 at
Wed Jul 6 15:49:17 EST 1994

kenp at banyo (Kenneth Prehoda) writes:

>Simon Brocklehurst (Bioc) (smb18 at wrote:
>: Ken Prehoda <kenp at> writes:

>Let's look at what you are saying in a little more detail.  I 
>assume you are saying there is some state which has a free
>energy lower than our observed "native" state.  However, this is
>really irrelevant.

   Maybe we're talking at cross purposes, 'cos I should have 
thought that this is extremely relevant to the kinetic/thermodynamic 

>Between the native and unfolded states
>thermodynamic control still holds (i.e. it is path _independent_).
>I don't see how you can argue that the folding pathway is
>important without having multiple native states.  Where are
>these other states???

    They're kinetically inaccessible.  That is, if you don't have 
transition states of sufficiently low energy to allow the protein to
pass through them, then you won't get to a particular energy minimum 
(native state).

I don't see how you can know that there aren't other low energy states 
on the free energy surface if the protein is being "directed" to 
fold along a particular pathway.

    So the question is, how is the protein directed to get to
a particular transition state and thence to the native, isn't it?  

>My impression is that i,i+4 side chains have steric interactions. This
>is from discussions with members of Baldwin's group - I can't refer
>to any paper's, although they're supposedly out there.  You can
>easily envision this with a little molecular modelling, though.

   Well in alpha helices i,i+4 side-chains are close in 
space, and you do observe a limited number of time-averaged side-chain 
conformations in such positions (I think Mike Sternberg did some analysis
of side-chain dihedral angles a while back - in J. Mol. Biol. ?).  
But these 'restrictions' do not cause a big problem for helix formation, 
because there are plenty of conformations available that are 

>: (stuff deleted)

>: >What are your arguments against hydrogen bonding "directing" folding?  

>:   Here goes:

>:   1) There is no thermodynamic (!!) advantage for main-chain polar groups
>:      to make intramolecular hydrogen bonds rather than to make hydrogen
>:      bonds with solvent -- is there?

>How do you know this?  You are assuming amide-amide hydrogen bonds
>are equal in strength to amide-water hydrogen bonds.  Additionally,
>you are neglecting any cooperativy considerations that may be
>present.  For example, and entropic advantage for hydrogen
>bonds is that once one h-bond is formed adjacent hydrogen bonds
>no longer must pay that entropic cost.  An possible enthalpic
>advantage is that once one h-bond is formed, adjacent amides
>become more polarized resulting in a more favorable enthalpy.

   Do you know that whether any/all of these suggestions are correct?

>:   2) Unfortunately, for many proteins whose folding pathways have
>:      been studied so far, it has not been possible to tell if hydrophobic 
>:      collapse occurs before or after the formation of stable, 
>:      long-lived hydrogen bonds.  In one case (the protein interleukin
>:      1-beta) , though, it was possible to tell: the interpretation is 
>:      difficult, but it seems that hydrophobic collapse occurs _before_ 
>:      the formation of long-lived, stable hydrogen bonds.

>So? what does this have to do with the thermodynamic stability of
>the protein?  The pathway does not determine the stability, but
>instead, it is the end states. I hope you aren't arguing against

     Didn't understand the second to last sentence, so at the risk
of repeating myself:

    I am arguing that the end state you see depends on the pathway
you are allowed to go down.  The important point on the pathway
being, of course, the transition state.  So from where I stand,
the idea that hydrophobic collapse happens before hydrogen
bond formation is extremely relevant: the early stages of folding being, 
all about pushing the protein towards the transition state.
 |  ,_ o     Simon M. Brocklehurst,
 | /  //\,   Oxford Centre for Molecular Sciences,
 |   \>> |   Department of Biochemistry, University of Oxford,
 |    \\,    Oxford, UK.
 |           E-mail: smb at

More information about the Proteins mailing list