Alanine scannin
g muta
genesis, double mutant cycles, andX-ray crystallo
graphy were used tocharacterize the interface between the anti-hen e
gg white lysozyme(HEL) antibody D1.3 and HEL. Twelveout of the 13 non
glycine contact residues on HEL, as determined by thehi
gh-resolution crystal structureof the D1.3-HEL complex, were individually truncated to alanine.Only four positions showed a
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
G(
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
Gmutant -
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
Gwild-type) of
greater than 1.0 kcal/mol,with HEL residue Gln121 provin
g the most criticalfor bindin
g (
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
G = 2.9 kcal/mol). These residuesform a conti
guous patch at the periphery of the epitopereco
gnized by D1.3. To understand how potentially disruptivemutations in the anti
gen are accommodatedin the D1.3-HEL interface, we determined the crystal structure to 1.5&Arin
g; resolution of the complex betweenD1.3 and HEL mutant Asp18
![](/ima<font color=)
ges/entities/rarr.
gif"> Ala. This mutation results in a
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
![](/ima<font color=)
ges/
gifchars/Delta.
gif" BORDER=0 >
G of only 0.3 kcal/mol, despite theloss of a hydro
gen bond and seven van der Waals contacts to the Asp18side chain. The crystal structurereveals that three additional water molecules are stably incorporatedin the anti
gen-antibody interface atthe site of the mutation. These waters help fill the cavitycreated by the mutation and form part of arearran
ged solvent network linkin
g the two proteins. To furtherdissect the ener
getics of specific interactionsin the D1.3-HEL interface, double mutant cycles were carried out tomeasure the couplin
g of 14 aminoacid pairs, 10 of which are in direct contact in the crystal structure.The hi
ghest couplin
g ener
gies, 2.7and 2.0 kcal/mol, were measured between HEL residue Gln121 and D1.3residues V
LTrp92 and V
LTyr32, respectively. The interaction between Gln121 andV
LTrp92 consists of three van der Waalscontacts,while the interaction of Gln121 with V
LTyr32 ismediated by a hydro
gen bond. Surprisin
gly, however,most cycles between interface residues in direct contact in the crystalstructure showed no si
gnificantcouplin
g. In particular, a number of hydro
gen-bonded residue pairswere found to make no net contributionto complex stabilization. We attribute these results toaccessibility of the mutation sites to water, suchthat the mutated residues exchan
ge their interaction with each other tointeract with water. This impliesthat the stren
gth of the protein-protein hydro
gen bonds in theseparticular cases is comparable to that ofthe protein-water hydro
gen bonds they replace. Thus, the simplefact that two residues are in directcontact in a protein-protein interface cannot be taken as evidencethat there necessarily exists a productiveinteraction between them. Rather, the majority of such contactsmay be ener
getically neutral, as in theD1.3-HEL complex.