Abstact

We have determined the structure of deoxyhaemoglobin from the antarctic fish Pagothenia bernacchii at pH 6.2 to a resolution of 2.2 A with X-ray data from a twinned crystal deconvoluted so as to approximate data from a single crystal. The R-factor between the (twinned) model and the observed data is 16% for reflections used in refinement and 22% for reflections not used in refinement. The T (deoxy) structure was compared with the R (liganded) structure at pH 8.0 in an attempt to understand the structural basis of the greater affinity for hydrogen ions of T, relative to R, that comprises the Root effect. Up to half of the effect can be attributed to interaction of the residues Asp95 (G1)alpha and Asp101 (G3)beta: in R the residues are far apart and their carboxyl groups are unprotonated, but the shift at the alpha 1 beta 2 interface that accompanies the R to T transition brings them so close that they appear to share a proton between them. The proximity of Asp99 (G1)beta may contribute to the required raising of the pKa values of the other two Asp residues. These and neighbouring residues are sufficiently conserved in the haemoglobins of trout (component IV), carp and bluefin tuna, all of which exhibit the Root effect, for the same mechanism to apply. However, the environment is equally conserved in haemoglobins of Trematomus newnesi (major component) and trout (component I), which do not exhibit the Root effect, so that the structural factors controlling the Asp-Asp interaction remain unclear. No other residue appears to undergo an R to T change in the immediate neighbourhoods that could account for any significant portion of the Root effect, so at least half of the effect must result either from long-range electrostatic interactions or from a large number of local interactions.