AdoHcy-hydrolase

SMM synthesis is mediated by the enzyme methionine S-methyltransferase (MMT) through the essentially irreversible, AdoMet-mediated methylation of methionine.48"5 Both MMT and SMM are unique to plants 48,50 The opposite reaction, in which SMM is used to methylate homocysteine to yield two molecules of methionine, is catalyzed by the enzyme homocysteine S-methyltransferase (HMT).48 Unlike MMT, HMTs also occur in bacteria, yeast, and mammals, enabling them to catabolize SMM of plant origin, and providing an alternative to the methionine synthase reaction as a means to methylate homocysteine. Plant MMT and HMT reactions, together with those catalyzed by AdoMet synthetase and AdoHcy hydrolase, constitute the SMM cycle (Fig. 2.4).4... 

The loss of a methyl group from AdoMet in each of the reactions yields S-ad-enosylhomocysteine (AdoHcy) and this is subsequently hydrolysed to adenosine and Hey by AdoHcy-hydrolase. Hey sits at a metabolic branch point and can be remethylated to methionine by way of two reactions. One is the 5-methyltetrahydrofo-late dependent reaction catalysed by methionine synthase, which itself is reductively methylated by cobalamin (vitamin B12) and AdoMet, requiring methionine synthase reductase. 5-Methyltetrahydrofolate is generated from 5,10-methylenetetrahydrofo-late (MTHF) by MTHF reductase. The second remethylation reaction is catalysed by betaine methyltransferase, which is restricted to the liver, kidney and brain, while methionine synthase is widely distributed. 

The laboratory measurement of AdoMet and AdoHcy is not routinely performed, but plays an increasing role in studies of pathogenesis of elevated Hey as well as the differential diagnosis of hypermethioninaemia due to deficiency of MAT, glycine methyltransferase or AdoHcy hydrolase. 

AdoHcy hydrolase is inactivated by a variety of other nucleosides, including the cyclopentenyl analog of adenosine, neplanocin A (Borchardt et al., 1984). Complete inactivation is observed with 0.5 equivalents of neplanocin A per active site, yielding 0.5 equivalents each of NADH and adenine (Wolfson et al., 1986). The mechanism of the reaction was proposed to parallel that suggested for 2 -dAdo, although the elimination of adenine would require a cis-elimination, in contrast to the trans-elimination possible with a 2 -deoxy-3 -ketonucleoside. Subsequent studies concurred that NADH was formed in the reaction but demonstrated that the inactivation is reversible by addition of NAD, suggesting that inactivation is simply due to reduction of the coenzyme (Matuszewska and Borchardt, 1987). 

In eukaryotes, AdoHcy is metabolized through a single metabolic pathway catalyzed by AdoHcy hydrolase, an enzyne that was first characterized by de la Haba and Cantoni (1957)". Recently AdoHcy... 

The reaction catalyzed by AdoHcy hydrolase is reversible and the equilibrium is strongly in the direction of synthesis d. Physiologically, the reaction proceeds in the hydrolytic direction because the products of the reaction, adenosine (Ado) and L-homocysteine (Hey), can be efficiently removed by a number of different enzymatic systems (Fig. 2). 

As shown in Fig. 2, an unexpected catalysis by AdoHcy hydrolase was observed by us, namely the synthesis of S-inosylhomocysteine. 

Little is known as yet about the comparative biochemistry of AdoHcy hydrolase. We have evidence that in beef liver there are two isozymes with very similar isoelectric properties. It is of interest that the purified enzyme from beef liver, molecular weight of 240,000, is composed of four identical subunits, whereas the lupin... 

Eloranta examined the distribution of the activity of AdoHcy hydrolase in different rat tissues. He found that the activity of the hydrolase is 1,000 times greater than that of the AdoMet synthetase. In keeping with this difference, it is generally found that the concentrations of AdoHcy in the cells is 1/4 to 1/10 that of AdoMet. 

From a chemical standpoint, the reaction catalyzed by AdoHcy hydrolase is quite unusual. In the synthetic direction, the reaction involves the replacement of the 5 -OH group of adenosine with a thiol nucleophile. The 5 -OH is a poor cleaving group and, since the reaction does not require ATP, a mechanism for the cleavage of the hydroxyl group is not immediately apparent. In the direction of hydrolysis, the situation is not any easier, since the reaction involves the hydrolysis of a thioether. 

Palmer and Abeles investigated this interesting problem and found that the enzyme contains one tightly bound NAD per subunit, a result that we have confirmed in our laboratory. The discovery that AdoHcy hydrolase contains tightly bound NAD led Palmer and Abeles to propose the complex mechanisms shown in Fig. 3. The mechanism proposed by Palmer and Abeles involves the oxidation of the 3 -hydroxyl group of AdoHcy by enzyme bound NAD. The result... 

With regard to the other product of AdoHcy hydrolase, adenosine, the interest in this compound has been greatly stimulated by the discovery in 1972, that an inborne error of metabolism, adenosine deaminase deficiency, is associated with severe immuno-deficiency involving both B and T lymphocytes. Intracellularly, adenosine can be formed only from AMP by 5 -nucleotidase and from AdoHcy by AdoHcy hydrolase probably the hydrolysis of AdoHcy is the predominant pathway. It can be removed from the cell by release into the extracellular fluid, by deamination to inosine and by reutilization to AMP. 

Although AdoHcy hydrolase is not found in bacteria, AdoHcy is cleaved irreversibly to adenine and S-rlbosyl-L-homocystelne by AdoHcy nucleosidase ". This enzyme has been partially purified from E, aoliy and found to catalyze also the hydrolysis of 5 -methy1-thioadenosine to adenine and methylthiorlbose. The enzyme is inactive towards AdoMet and a variety of other purine and pyrimidine nucleosides. The specific activity of AdoHcy nucleosidase is 1000 x greater than that of AdoHcy hydrolase emphasizing the need to remove AdoHcy, a potent inhibitor of reactions which utilize AdoMet as a substrate. 

P. Chiang et al. screened a large number of compounds for their ability to inhibit AdoHcy hydrolase and identified among them a compound, 3-deazaadenosine (DZA), that is a very potent inhibitor of the enzyme. The inhibition is due to the fact that 3-DZA, like adenosine binds very tightly to the enzyme, but unlike adenosine is not a substrate for adenosine deaminase or adenosine kinase. ... 

The last two enz)nne reactions are responsible mainly for the metabolism of adenosine. The modification of the N-atom at the 3-position is highly specific, because tubercidin and formycin, respectively the 7- and 9-deazaadenosine analogs of adenosine, are neither inhibitors nor substrates for AdoHcy hydrolase. 

It should be emphasized that we do not yet know whether the effects seen are due to the inhibition of AdoHcy hydrolase, and a resultant increase in AdoHcy and/or to utilization of 3-DZA as a substrate and formation of 3-DZA-Hcy, and/or, finally to the metabolic immobilization of Hey residues in these purinylhomocysteine compounds. We expect to have an answer to these interesting questions in the near future. 

It can, therefore, be suggested as a general working hypothesis that the efficacy of 3-DZA and other AdoHcy hydrolase inhibitors and/ or AdoHcy analogs as antiviral agents relates to their ability to modulate or inhibit the reactions that lead to the synthesis of a fully methylated 5 -cap structure in viral mRNAs. Considerable experimental evidence supporting this hypothesis has been accumulated, but the critical demonstration of its validity is not yet available. 

Several compounds were examined for concentration-dependent inactivation of AdoHcy hydrolase from beef liver (Table 1), and candidates were selected for evaluation of time-dependent inactivation (Table 2). The 5 -a-fluoro thioethers 10a, 10b, and 10c were quite potent inactivators. However, it was observed that lOa-c were unstable in the buffer test solutions, and other compounds were formed (HPLC) soon after the samples were dissolved. H NMR peaks at 6 9.8 and -11.7 (DMSO-de) were present after 10a was allowed to stand in aqueous solution. The first peak was stable upon addition of D2O, but the second peak at 5 11.7 rapidly disappeared upon D2O exchange. This is consistent with chemical hydrolysis of the 5 -a-fluoro thioethers 10 to give the epimoic 4 -carbaldehyde 28, hydroxy enol etho 29, and aldehyde dihydrate 30 products (see Scheme 4). 

Table 1. ( alitative inhibition of bovine liver AdoHcy hydrolase... 

The possibility that a component of this "adenosine 5 -aldehyde" mixture is the actual inhibitor was supported by the following observations (1) the chemically stable 7b was not an inhibitor of AdoHcy hydrolase (2) Ado and Hey were rapidly converted into AdoHcy by AdoHcy hydrolase, but 7b and Hey gave no detected AdoHcy (3) the chemically stable sulfones 18 and 20 derived firom 10b were not inhibitors (4) the uridine analogues of 10b [5 (R/S)-fluoro-5 -5-(4-methoxyphenyl)-5 -thiouridine] w e not inhibitors. 

Scheme 4 Possible inhibitors of AdoHcy hydrolase derived from adenosine S -aldehyde...

Slight differences in inhibitory concentrations and rates of inactivation of AdoHcy hydrolase were observed between diastereomer-enriched samples of the 4-chlorophenylthio compounds 10c (Table 2) and larger differences relative to the 4-methoxyphenylthio analogues 10b were found. Current studies indicate that different rates and/or mechanisms of chemical hydrolysis of the a-fluoro thioethers can account for these differences. 

It is noteworthy that qualitative differences exist between the 25( ) and 26(Z) 5 -chloromethylene isomers with respect to inhibition of AdoHcy hydrolase. The authentic Z-isomer 26 is a time-dependent inactivator with potency comparable to that of its 5 -fluoromethylene analogue 15 (Table 2). In contrast, the -isomer 25 showed little activity in our preliminary screen (Table 1) and was reported to be a reversible inhibitor in another study (in which the Z-configuration had been tentatively assigned). ... 

---