A P‐loop‐like motif in a widespread ATP pyrophosphatase domain: Implications for the evolution of sequence motifs and enzyme activity

Abstract A conserved amino acid sequence motif was identified in four distinct groups of enzymes that catalyze the hydrolysis of the α–β phosphate bond of ATP, namely GMP synthetases, argininosuccinate synthetases, asparagine synthetases, and ATP sulfurylases. The motif is also present in Rhodobacter capsulata AdgA, Escherichia coli NtrL, and Bacillus subtilis OutB, for which no enzymatic activities are currently known. The observed pattern of amino acid residue conservation and predicted secondary structures suggest that this motif may be a modified version of the P‐loop of nucleotide binding domains, and that it is likely to be involved in phosphate binding. We call it PP‐motif, since it appears to be a part of a previously uncharacterized ATP p yropho p hatase domain. ATP sulfurylases, NtrL, and OutB consist of this domain alone. In other proteins, the pyrophosphatase domain is associated with amidotransferase domains (type I or type II), a putative citrulline‐aspartate ligase domain or a nitrilase/amidase domain. Unexpectedly, statistically significant overall sequence similarity was found between ATP sulfurylase and 3′‐phosphoadenosine 5′‐phosphosulfate (PAPS) reductase, another protein of the sulfate activation pathway. The PP‐motif is strongly modified in PAPS reductases, but they share with ATP sulfurylases another conserved motif which might be involved in sulfate binding. We propose that PAPS reductases may have evolved from ATP sulfurylases; the evolution of the new enzymatic function appears to be accompanied by a switch of the strongest functional constraint from the PP‐motif to the putative sulfate‐binding motif. © 1994 Wiley‐Liss, Inc.