Differential interaction of porphyrins used in photoradiation therapy with ferrochelatase

Ferrochelatase from Rhodopseudomonas sphaeroides: substrate specificity and role of sulfhydryl and arginyl residues

Purified ferrochelatase (protoheme ferrolyase; EC 4.99.1.1) from the bacterium Rhodopseudomonas sphaeroides was examined to determine the roles of cationic and sulfhydryl residues in substrate binding. Reaction of the enzyme sulfhydryl residues with N-ethylmaleimide or monobromobimane resulted in a rapid loss of enzyme activity. Ferrous iron, but not porphyrin substrate, had a protective effect against inactivation by these two reagents. Quantitation with 3H-labeled N-ethylmaleimide revealed that inactivation required one to two sulfhydryl groups to be modified. Modification of arginyl residues with either 2,3-butanedione or camphorquinone 10-sulfonate resulted in a loss of ferrochelatase activity. A kinetic analysis of the modified enzyme showed that the Km for ferrous iron was not altered but that the Km for the porphyrin substrate was increased. These data suggested that arginyl residues may be involved in porphyrin binding, possibly via charge pair interactions between the arginyl residue and the anionic porphyrin propionate side chain. Modification of lysyl residues had no effect on enzyme activity. We also examined the ability of bacterial ferrochelatase to use various 2,4-disubstituted porphyrins as substrates. We found that 2,4-bis-acetal- and 2,4-disulfonate deuteroporphyrins were effective substrates for the purified bacterial enzyme and that N-methylprotoporphyrin was an effective inhibitor of the enzyme. Our data for the ferrochelatase of R. sphaeroides are compared with previously published data for the eucaryotic enzyme.

Interactions with glutathione S-transferases of porphyrins used in photodynamic therapy and naturally occurring porphyrins

Several naturally occurring porphyrins and porphyrins used in photodynamic therapy inhibit glutathione S-transferase isoenzymes either purified from rat liver or lung or in cytosol from normal and from cancerous (Morris 7288C hepatoma) liver. Although differences occur in the type and amount of transferases in normal and cancerous liver and in the liver of rats bearing an extrahepatic tumour, these enzymes are potential binding sites for porphyrins. Porphyrin structure is an important factor in determining the affinity of binding, as shown by the relative inhibitory effectiveness. Of the dicarboxylic porphyrins in the mixture used clinically, OO'-diacetylhaematoporphyrin and monohydroxyethylmonovinyldeuteroporphyrin are more effective inhibitors than haematoporphyrin and protoporphyrin IX. Of the naturally occurring porphyrins the order of effectiveness is protoporphyrin IX (dicarboxylic) greater than coproporphyrin (tetracarboxylic) greater than uroporphyrin (octacarboxylic) and type I greater than type III isomers of both uroporphyrin and coproporphyrin, and the synthetic tetra-meso-phenylporphinetetrasulphonate is a better inhibitor (apparent Ki = 250 nM) than coproporphyrin, which contains a comparable number of negative charges. In addition, iron-porphyrin chelates are more effective inhibitors of the transferases, with 25-fold decrease in Ki value, than the free porphyrins. These results indicate that one means whereby porphyrins accumulate in tissues is the occupation of intracellular binding sites, such as the transferases. Since porphyrins inhibit the activity of these important detoxifying enzymes, there will be metabolic consequences to the cell.

Spectroscopic examination of the active site of bovine ferrochelatase

Spectrofluorometric techniques have been employed to examine the active site of the terminal enzyme of the heme biosynthetic pathway, ferrochelatase (protoheme ferrolyase, EC 4.99.1.1). The fluorescence of both endogenous tryptophan and exogenous 2-(4-maleimidylanilino)naphthalene-6-sulfonic acid (MIANS) has been examined. The fluorescence emission of the enzyme's active site bound MIANS is at 428 nm while the enzyme tryptophan(s) yielded a single fluorescence emission maximum at 347 nm. These values are characteristic of a polar environment for tryptophan and a relatively nonpolar environment for the MIANS. The dynamic fluorescence quenching constants for acrylamide of MIANS and tryptophan are 3.00 M-1 and 1.85 M-1, respectively. Quenching constants for KI of both fluorescent centers were approximately 1 M-1. These data suggest that both fluorophores are poorly accessible to the external anionic contact quencher but that an unchanged quencher, while larger, is still better able to penetrate the enzyme's active site. The extrapolated anisotropies (r0) for ferrochelatase-bound MIANS and tryptophan are 0.198 and 0.307. The dissociation constant (KD) determined by fluorescence anisotropy of protoporphyrin was 1.5 microM with the calculated number of porphyrin binding sites as 1.0 per 40000 daltons. A model is presented for the active site of ferrochelatase based upon the data presented here and previously. This model proposes that the active site is a hydrophobic pocket similar in nature to the heme binding crevices found in many hemoproteins.