Senescence marker protein-30 (SMP-30) is a candidate enzyme that can function as a catalytic bioscavenger of organophosphorus (OP) nerve agents. We purified SMP-30 from mouse (Mo) liver and compared its hydrolytic activity towards various esters, lactones, and G-type nerve agents with that of human paraoxonase1 (Hu PON1) and squid diisopropylfluorophosphatase (DFPase). All three enzymes contain one or two metal ions in their active sites and fold into six-bladed β-propeller structures. While Hu PON1 hydrolyzed a variety of lactones, the only lactone that was a substrate for Mo SMP-30 was d-(+)-gluconic acid δ-lactone. Squid DFPase was much more efficient at hydrolyzing DFP and G-type nerve agents as compared to Mo SMP-30 or Hu PON1. The K(m) values for DFP were in the following order: Mo SMP-30>Hu PON1>squid DFPase, suggesting that the efficiency of DFP hydrolysis may be related to its binding in the active sites of these enzymes. Thus, homology modeling and docking were used to simulate the binding of DFP and selected δ-lactones in the active sites of Hu SMP-30, Hu PON1, and squid DFPase. Results from molecular modeling studies suggest that differences in metal-ligand coordinations, the hydrophobicity of the binding pockets, and limited space in the binding pocket due to the presence of a loop, are responsible for substrate specificities of these enzymes.