Skip to main content
Biology LibreTexts

H Bonds and Backbone Interactions

H Bonds and Backbone Interactions: A Quantum Mechanical Perspective

A more complete description of the energetics underlying protein structure and stability would involve quantum mechanical effects. These have not been studied in detail since quantum mechnanical calculations are difficult to perform on large molecules. Recent work has led an understanding of a new source of stability of proteins based on overlap of proximal molecular orbitals that stabilizes a protein through electron delocalization. Two common examples of electron delocalization that increase stability are resonance and hyperconjugation. Resonance involves delocalization of electrons through overlap of adjacent atomic p orbitals to form low energy molecular bonding orbitals. You may remember hyperconjugation being used to explain the relative stability of tertiary carbocations compared to secondary or primary ones. The typical explanation involves electron donation and pushing from the adjacent methyl groups which effectively decreases the charge on the carbocation. A more correct explanation involves overlap of molecular orbitals, specifically of a C-H s orbital from one of the C-H bonds of the methyl group with the partially filled p orbital on the carbon atom containing the charge as seen in the figure below.

Figure: Stabilization of Carbocations by Hyperconjugation


It should also not be surprising that orbital overlap accounts for hydrogen bonds interactions (in contrast to the more simplistic explanation of the attractions between the partial charges on hydrogen bond donors and acceptors). Bartlett et al (2010) have recently described two types of orbital overlaps involving electron delocalization that stabilize protein structures. Molecular orbital description show that two lone pairs on the oxygen (hydrogen bond donor) in amide bonds are different. One pair, ns (Figure a below) is in a nonbonding orbital with 60% s character, while the other, np (figure b below), has approximately 100% p character. Figure c below shows that a hydrogen bond between the carbonyl oxygen of the ith amino acid and the amide H of the ith+4 amino acid can be described in quantum mechanical terms as the overlap (i.e. delocalization) of the ns nonbonding orbital on oxygen and the s* antibonding N-H orbital. What about the np electrons on the carbonyl O? They found evidence for a new kind of stabilizing influence in proteins as the np orbital overlaps with the an antibonding p* orbital on the ith+1 carbonyl group (figure d below) when the two amino acids are proximal as parts of secondary structures, including alpha and 310 helices as well as in twisted b sheets.

Figure: Orbital involvement in H-bonds and n to pi* interactions in alpha helices


(a) s-rich lone pair of an amide oxygen. (b) p-rich lone pair of an amide oxygen. (c) ns to σ*: hydrogen bond in an a-helix. (d) np to p*: n to p* interaction in an a-helix. Reprinted by permission from Macmillan Publishers Ltd: Nature Chemical Biology. Bartlett, G. et al. n to pi* interactions in protein. 6, pg 615 (2010)

To test which phi-psi angles would allow close enough approach for np to p* interactions, the investigators performed quantum mechanical computations on possible conformations of AcAla4NHMe for which np to p* where possible (see figure A below). These calculations showed these interaction were possible for values of d ≤ 3.2 Å and 99° ≤ θ ≤ 119° (figure a below). Distances less than 3.2 angstroms, van der Waals surfaces of the C and O on adjacent carbonyls overlap, suggesting that the interaction is quantum mechanical and not classically mediated. Next they calculated d andq values for on high resolution x-ray structures of known proteins. Red shading in the Ramachandran plot in figure c below phi/psi regions in actual proteins that meet the required geometry for np to p* overlap. Hence these geometry are abundant in proteins, especially in secondary structures described above.

Figure: Ramachandran plots of n to pi* interactions.



(a) Definitions of dihedral angles φ (C′i–1–Ni–Cα i–C′i) and ψ (Ni–Cαi–C′i–Ni+1), distance (d) and planar angle θ. Criteria for an n to p* interaction in the crystallographic analyses: d ≤ 3.2 Å; 99° ≤ θ ≤ 119°. (b) Computational data showing the energy. Reprinted by permission from Macmillan Publishers Ltd: Nature Chemical Biology. Bartlett, G. et al. n to pi* interactions in protein. 6, pg 615 (2010)

Results from thermophilic organism: What kinds of modifications are made to the sequence of a protein as the temperature that the organism thrives increases? A recent study by Szilagyi and Zavodszky (Structure, 8, pg 493, 2000) studied 93 structures of 25 proteins, 29 from organisms that live at elevated temperatures (thermophiles, >50oC for optimal growth ) and 64 at nominal temperatures (mesophiles). Here are their results:

  • the number of H-bonds and secondary structure elements do not correlate with temperature, but the number of salt bridges do.
  • in hyperthermophiiles (>80oC for optimal growth organisms) that thrive at very high temperatures (100oC), few internal cavities were found.
  • in those that thrive at intermediate high temperatures (45-80oC) the surface had more polar residues.
  • generally there was an increase in weaker ion pairs (increased distance between the charged side chains) in the hot group, but increases in strong and weak ion-ion bonds in the very hot group.

Kashefi and Lovley recently reported the identification of a bacteria obtained from a hydrothermal vent in the northeastern Pacific ocean. In a laboratory setting, the strain grew in water at a temperature of 121oC under high pressure. These are the same conditions used in autoclaves to produce sterile samples. Cell doubling took place under these conditions in 24 hours. The authors suggest that this strain would be useful to determine molecules and their properties necessary for such high temperature growth.

Using a computational program called Rossetta Design (PNAS, 97, 10383 (2003)), Korkegian et al determined mutations in buried side chains of the homodimer cytosine deaminase. Buried residues are presumably are important in the stability of a protein for mutagenesis that would increase the melting temperature (Tm) of the protein. In the program, a target sequence was "threaded" onto the sequence of the template protein (the wild type protein) and changes made to side chains in the random sequence. Energies were calculated and those changes resulting in lower energies were saved. Target residues (88) within 4 angstroms of the active site and the dimer interface were fixed to those in the wild-type template in order to minimize alterations in the catalytic activity of the enzyme, cytosine deaminase, that they chose to study. Remember, the goal of the study was not to increase the catalytic activity of the enzyme, but rather increase its themostability. The rest (65) were changed and energies calculated. 49% of the amino acids subjected to random change produced no change in amino acid compared to the template (wild-type) side chain. 16 changes on the surface were ignored. Two sets of changes were observed, one involving amino acids packed between an alpha helix and beta strands, and the other set between two alpha-helices. These later mutants, when prepared in the lab using recombinant DNA technology, were soluble at high protein concentrations, and could be studied. Three different mutants (A23L, I140L, V108I) were made which increased the TM by about 2 degrees. However, a triple mutation had TM values 10 degrees higher than the wild-type protein and a 30-fold longer T1/2 at 50 degrees C. When the triple mutant was introduced into bacteria, the bacteria grew better at higher temperatures. Crystal structures of both the wild-type and triple mutants shown essentially an identical fold, with about 70 A2 of additional surface area buried in the mutant protein.

Beeby et al. analyzed sequence and structural data from P. aerophilum (archea) and Thermus thermpilus (thermophilic bacteria) and found that disulfide bonds stabilized proteins from these species. Cytoplasmic protein from eukaryotes don't have disulfides due to the presence of reducing agents (such as glutathione) in the cell. In those thermophiles with disulfides in proteins, a novel protein, protein disulfide oxidoreductase, was found, which catalyzes the formation of sulfide bonds. Finally, Berezovsky and Shakhnovich have also analyzed proteins from hyperthermophilic archea and bacteria and compared them to analogous proteins from mesophilic bacteria. They found two types of stabilizations of hyperthermophilic proteins, depending on the evolutionary history of the organism. Proteins from cells that originally evolved in high temperature conditions (archea) were very compact (maximizing van der Waals interactions, had a high number of contacts per residue, and a high percentage of hydrophobic residues), but did not use specific structural stabilizing interactions (like electrostatic in salt bridges). In contrast, proteins from cells the originally evolved under mesophilic conditions, but later adapted to hyperthermophilic conditions had proteins that evolved specific sequences features that stabilized electrostatic interactions (more charged residues, salt bridges, .

Answers to Protein Chapter Preclass Discussion Questions