T. Niermann et K. Kirschner, THE PREDICTED SECONDARY STRUCTURE OF THE G-TYPE GLUTAMINE AMIDOTRANSFERASE IS COMPATIBLE WITH TIM-BARREL TOPOLOGY, Protein engineering, 8(6), 1995, pp. 535-542
Glutamine amidotransferase (GAT) subunits or domains catalyze an impor
tant partial reaction in many complex biosynthetic reactions. The stru
cture of one member of the F-type GATs is known, but the structure of
the unrelated G-type is still unknown, Because many protein sequences
are available for anthranilate synthase component II (product of the t
rpG gene), we have predicted its average secondary structure by a join
t prediction method [Niermann and Kirschner (1991a) Protein Engng, 4,
359-370], The predicted eight beta-strands and seven alpha-helices fol
low an 8-fold cyclic repetition of a beta-strand-loop-alpha-helix-loop
module with helix alpha(7) missing. This pattern of secondary structu
re suggests that the G-type GAT domain has an 8-fold beta alpha-barrel
topology, as found first in triose phosphate isomerase (TIM-barrel).
This model is supported by the location of known catalytically essenti
al residues in loops between beta-strands and alpha-helices. Evidence
from published sequencing and mutational studies on selected members o
f the GAT superfamily (carbamoyl phosphate, imidazoleglycerol phosphat
e, GMP and CTP synthases) support both the secondary structure predict
ion and the TIM-barrel topology.