In a statistical study of amino acid placement in alpha helices in proteins, Richardson and Richardson observed that Asn frequently occurs in the so-called N-cap position, the amino acid appearing just prior to the first residue possessing the phi, psi angles accompanying the helical conformation. Examination of a model of an alpha helix demonstrates that the Asn side chain is capable of forming hydrogen bonding interactions with the unpaired N-H groups of the mainchain amides at the N-terminus of the helix (at position N2 and N3). To make these interactions the backbone psi dihedral angle of the N-cap residue must rotate away from a typical helical conformation (psi = -47 degrees) to psi = 145 degrees. Therefore, these interactions would not be accessible to a residue fixed in the helical conformation, explaining the relative scarcity of Asn residues appearing at position N1.
Glutamine, despite having a similar sidechain functionality, does not appear at the N-cap position with any greater frequency that would be statistically expected. Inspection of a model with Gln at postion 1 shows that the larger side chain makes it difficult to orient Gln correctly with respect to the unpaired N-H groups. There is great difficulty associated with placing the sidechain carboxamide carbonyl oxygen within H-bonding geometry of these groups, and it requires an unfavorable eclipsing conformation of the Gln sidechain. Given the enthalpically unfavorable conformation as well as the greater loss of entropy associated with fixing the geometry of the Gln sidechain, it is clear this residue is not as desirable at the N-cap position as Asn.
Proline often occurs at position N1 in the helix, but not position N2. The advantages of proline at N1 are that it restricts the phi dihedral of the N1 residue to something near correct helix conformation and that the amide nitrogen of proline is alkylated, so it does not offer an unpaired N-H group at the N-terminus to add to helix dipole. These advantages would also hold true at position N2, but at this position the additional steric bulk associated with the alkyl substituion at the mainchain nitrogen is problematic. It invades the vdW radius of the mainchain nitrogen of residue N1. The vdW conflict is not a problem at N1 since the corresponding mainchain nitrogen of the N-cap residue can swing out of the way without affecting the structure of the helix.