CATALYSIS BY A NEW SIALIDASE, DEAMINONEURAMINIC ACID RESIDUE-CLEAVINGENZYME (KDNASE SM), INITIALLY FORMS A LESS STABLE ALPHA-ANOMER OF 3-DEOXY-D-GLYCERO-D-GALACTO-NONULOSONIC ACID AND IS STRONGLY INHIBITED BYTHE TRANSITION-STATE ANALOG, -DIDEHYDRO-D-GLYCERO-D-GALACTO-2-NONULOPYRANOSONIC ACID, BUT NOT BY 2-DEOXY-2,3-DIDEHYDRO-N-ACETYLNEURAMINIC ACID

Deaminoneuraminic acid residue-cleaving enzyme (KDNase Sm) is a new si alidase that has been induced and purified from Sphingobacterium multi vorum. Catalysis by this new sialidase has been studied by enzyme kine tics and H-1 NMR spectroscopy. V-max/K-m values determined for synthet ic and natural substrates of KDNase Sm reveal that 4-methylumbellifery l-KDN (KDN alpha 2MeUmb, V-max/K-m = 0.033 min(-1)) is the best substr ate for this sialidase, presumably because of its good leaving group p roperties. The transition state analogue, ehydro-2,3-dideoxy-D-galacto -D-glycero-nonulosonic acid, is a strong competitive inhibitor of KDNa se Sm (K-i = 7.7 mu M versus K-m = 42 mu M for KDN alpha 2MeUmb). 2-De oxy-2,3-didehydro-N-acetylneuraminic acid and 2-deoxy-2,3-didehydro-N- glycolylneuraminic acid are known to be strong competitive inhibitors for bacterial sialidases such as Arthrobacter ureafaciens sialidase; h owever, KDNase Sm activity is not significantly inhibited by these com pounds. This observation suggests that the hydroxyl group at C-5 is im portant for recognition of the inhibitor by the enzyme. Reversible add ition of water molecule (or hydroxide ion) to the reactive sialosyl ca tion, presumably formed at the catalytic site of KDNase Sm, eventually gives rise to two different adducts, the alpha- and beta-anomers of f ree 3-deoxy-D-glycero-D-galacto-nonulosonic acid. H-1 NMR spectroscopi c studies clearly demonstrate that the thermodynamically less stable a lpha-form is preferentially formed as the first product of the cleavag e reaction and that isomerization rapidly follows, leading to an equil ibrium mixture of the two isomers, the beta-isomer being the major spe cies at equilibrium. Therefore, we propose that KDNase Sm catalysis pr oceeds via a mechanism common to the known exosialidases, but the reco gnition of the substituent at C-5 by the enzyme differs.