The Impact of Backbone Hydrogen Bonds

By Jason Socrates Bardi

In 1973, Christian Anfinsen shook the biology world with a simple but powerful experiment involving protein folding. He unfolded a protein called RNase A by placing it in a solution of 8 M urea. He then washed away the urea, and the catalytic activity of the enzyme returned, showing that the stable three-dimensional structure of a chain of amino acids is determined by the sequence of those amino acids.

But how exactly do the amino acids dictate the structure? For the last three decades, chemists and biologists have attempted to answer this question by dissecting out all the attractive and repulsive forces between their amino acids, their metal ions or other cofactors, and the surrounding solution.

Scientists, for instance, have done experiments in which they mutate these side chains and measure the effect of these mutations on such factors as the protein's thermal stability or kinetics of folding. However, the amino acid side chains are not the only parts of a protein that dictate how it will fold. One of the most common features of a protein structure is "backbone" hydrogen bonds—electrostatic attractions that arise because one part of the protein backbone that contains a partially positively charged nitrogen comes in close contact with another part of the backbone that contains a partially negatively charged oxygen.

Many scientists have thought that these backbone "H-bonds" contribute significantly to the folding and stability of a protein structure, and this idea has been supported by numerous computational models and indirect measurements.

Yet, few studies measured experimentally how these backbone hydrogen bonds influence the folding and stability of a protein. Part of the problem is that mucking with an amino acid's backbone atoms is not easy. The biggest barrier is the chemistry—making modified amino acid starting materials is complicated and they are not commercially available. Until recently, nobody has ever replaced all the backbone hydrogen bonds in a single protein to experimentally analyze their distinct contributions to a protein's folding.

Now, Jeffery Kelly and his colleagues in the Department of Chemistry and The Skaggs Institute for Chemical Biology at The Scripps Research Institute have done just that. In a recent issue of the journal Nature, Kelly and his colleagues describe the energetic contributions of all the individual backbone hydrogen bonds in one protein by making a series of mutants of that protein in which the H-bonds are all knocked out one by one.

"This study represents a significant step towards trying to understand the energetic contributions of hydrogen bonds in a protein in solution," says Kelly, the report's lead author. Kelly, who is the Lita Annenberg Hazen Professor of Chemistry and vice president of academic affairs at Scripps Research, adds that this is only one of two systematic analyses of the contributions of H-bonds to the folding and stability of a protein. The other was done on a helix-rich protein in Philip Dawson's laboratory at The Scripps Research Institute. Dawson is a coauthor on the Kelly paper.

Not All Hydrogen Bonds Are Created Equal

Kelly's graduate student Songpon Deechongkit found a way to optimize the available chemistries to make hydroxy acids and generate modified proteins in which the backbone atoms of particular amino acid residues known to participate in hydrogen bonding are swapped for an atom that is no longer capable of hydrogen bonding.

The protein they chose to study was the PIN WW domain, a small, 34-amino acid protein that lends itself to this sort of analysis because it can tolerate amino acid replacements at nearly every position in its chain and because it can be synthesized by chemical methods. The PIN WW domain also has been well characterized structurally and thermodynamically. It folds reliably into a beta sheet—one of the most common types of protein folds in which sections of the protein chain line up next to each other—and it is known to make 11 backbone hydrogen bonds within this beta sheet structure.

Deechongkit made mutant forms of PIN WW domain that contained "ester" residues, which were chemically identical to the amino acids linked by amides except that they contained an oxygen atom where the backbone nitrogen atom would normally go. The loss of the nitrogen atom eliminated the hydrogen bonds donor.

In their paper, Kelly, Deechongkit, Dawson, Martin Gruebele of the University of Illinois, Urbana, and their colleagues report results of thermodynamic and kinetic experiments conducted to determine which of the hydrogen bonds were most important for the folding.

They found that of the 11 hydrogen bonds in the WW domain, only five of these are really important for the protein's stability. Mutating them had the most severe effect on the stability of the protein. As one might predict, these five hydrogen bonds are buried in the interior of the protein.

The microenvironment of a hydrogen bond is important in determining the strength of this electrostatic interaction, explains Kelly, and inside the confines of the interior of a protein, the environment is much less polar than at the surface of the protein. At the surface, the electrostatic interference of the surrounding polar water molecules reduces the strength of the hydrogen bonds.

The significance of this, says Kelly, is that if chemists and biologists are designing proteins and are concerned about how substituting amino acids might affect the stability of their protein, they probably need only worry about those hydrogen bonds that are in the interior. "Not all hydrogen bonds are created equal," he says.

Additionally, the paper reveals how perturbations to a different set of hydrogen bonds affect the rate of folding, rather than the stability.

Now Kelly and his colleagues are examining the effect of putting nonpolar atoms like carbon in place of the amides in the WW domain amino acid backbone. Replacing positively charged nitrogen with negatively charged oxygen, as they did in their recent study, not only eliminates favorable (+/-) hydrogen bonds. It also introduces an additional source of electronegativity, which may be contributing to the instability of the mutant molecule through unfavorable (-/-) interactions. To gauge how significant these interactions are, they are now conducting similar experiments replacing the nitrogens with neutral carbon.

The article, "Context-dependent contributions of backbone hydrogen bonding to beta-sheet folding energetics" by Songpon Deechongkit, Houbi Nguyen, Evan T. Powers, Philip E. Dawson, Martin Gruebele, and Jeffery W. Kelly appears in the July 1, 2004 issue of the journal Nature. See

This work was made possible through funds provided by the National Institutes of Health, The Skaggs Institute for Research, the Lita Annenberg Hazen Foundation, the National Science Foundation, the Alfred P. Sloan Foundation, and a Norton B. Gilula Fellowship.

Send comments to:







"Not all hydrogen bonds are created equal."

— Jeffery Kelly