Substrate oxidation by the heme edge of fungal peroxidases. Reaction of Coprinus macrorhizus peroxidase with hydrazines and sodium azide

The peroxidase from Coprinus macrorhizus is inactivated by phenylhydrazine or sodium azide in the presence of H2O2. Inactivation by phenylhydrazine results in formation of the delta-meso-phenyl and 8-hydroxymethyl derivatives of the prosthetic heme group and covalent binding of the phenyl moiety to the protein but not in the detectable formation of Fe-phenyl- or N-phenylheme adducts. Alkylhydrazines are catalytically oxidized but do not inactivate the enzyme. Catalytic oxidation of sodium azide produces the azidyl radical and results in its addition to the delta-meso position of the prosthetic heme group. Comparison of the heme adducts obtained with C. macrorhizus peroxidase with those generated by horseradish peroxidase shows that the regiochemistry of the addition reactions is the same in both cases. The results suggest that substrates interact primarily or exclusively with the heme edge rather than the ferryl oxygen of C. macrorhizus peroxidase and indicate that the interaction occurs with the same sector of the heme edge as in horseradish peroxidase. The active-site topologies of this pair of plant and fungal peroxidases thus appear to be similar, although the observation that alkylhydrazines add to the heme edge of horseradish but not C. macrorhizus peroxidase clearly shows that there are significant differences in the two active sites.     

Inactivation of lignin peroxidase by phenylhydrazine and sodium azide

Lignin peroxidase (LiP) is rapidly inactivated in a concentration-dependent manner by H2O2 and either phenylhydrazine or sodium azide. Full inactivation of isozyme 2b (H8) requires approximately 50 eq of phenylhydrazine or 80 eq of sodium azide. Anaerobic incubation of isozyme 2b with [14C]phenylhydrazine and H2O2 results in 77% loss of catalytic activity and covalent binding of 0.45 mol radiolabel/mol of enzyme. Comparable but not identical results are obtained with an isozyme mixture. A lag period is observed before the peroxidative activity can be measured when an aliquot of an incubation with sodium azide is diluted into the mixture used to assay residual catalytic activity. This lag is associated with reversible accumulation of a catalytically inert species with a Compound III-like spectrum. No meso-phenyl, iron-phenyl, or N-phenyl adducts are formed with phenylhydrazine but a low yield of what appears to be delta-meso-azidoheme is obtained with sodium azide. LiP is thus less susceptible to meso heme additions and more susceptible to oxidative heme degradation than horseradish peroxidase. The data suggest that the active of LiP resembles the closed structure of horseradish peroxidase more than it does the open structure of the globins, catalase, chloroperoxidase, or cytochrome P450.     

The dual oxygenase and peroxidase activities of porphobilinogen oxygenase and horseradish peroxidase: a study using the reaction with phenylhydrazine.

Porphobilinogen oxygenase and horseradish peroxidase show dual oxygenase and peroxidase activities. By treating porphobilinogen oxygenase with phenylhydrazine in the presence of H2O2 both activities were inhibited. When horseradish peroxidase was treated in the same manner only the peroxidase activity was lost while its oxygenase activity toward porphobilinogen remained unchanged. The phenylhydrazine treatment alkylated the prosthetic heme group of porphobilinogen oxygenase and N-phenylheme as well as N-phenylprotoporphyrin IX were isolated from the treated hemoprotein. In horseradish peroxidase the modified heme was mainly 8-hydroxymethylheme. The apoproteins of the alkylated enzymes were isolated and recombined with hemin IX. The oxygenase and peroxidase activities of porphobilinogen oxygenase were entirely recovered in the reconstituted enzyme, while the reconstituted horseradish peroxidase regained 75% of its peroxidase activity.     

Inactivation of catalase by phenylhydrazine. Formation of a stable aryl-iron heme complex

Catalase promotes the H2O2-dependent oxidation of phenylhydrazine to benzene but simultaneously is subject to a pseudo-first order inactivation process. Each inactivation event is subtended by catalytic turnover of three molecules of phenylhydrazine and 52 molecules of H2O2. The dimethyl ester of N-phenylprotoporphyrin IX is extracted with acidic methanol from the inactivated enzyme, but the prosthetic heme with a phenyl sigma-bonded to the iron atom is obtained by gentle extraction with 2-butanone. The absolute chirality of N-ethylprotoporphyrin IX isolated from catalase inactivated with ethylhydrazine confirms that the prosthetic heme has the same chiral orientation in the active site as it does in hemoglobin. The known inactivation of methemoglobin by phenylhydrazine is shown to depend on H2O2 but not oxygen. The results demonstrate that the H2O2-dependent oxidation of phenylhydrazine by catalase and other hemoproteins results in sigma-coordination of a phenyl residue to the prosthetic heme iron. This process may play a role not only in phenylhydrazine-mediated erythrocyte lysis but also in the activation of guanylate cyclase.     

Structure-mechanism relationships in hemoproteins. Oxygenations catalyzed by chloroperoxidase and horseradish peroxidase

Chloroperoxidase and H2O2 oxidize styrene to styrene oxide and phenylacetaldehyde but not benzaldehyde. The epoxide oxygen is shown by studies with H2(18)O2 to derive quantitatively from the peroxide. The epoxidation of trans-[1-2H]styrene by chloroperoxidase proceeds without detectable loss of stereochemistry, as does the epoxidation of styrene by rat liver cytochrome P-450, although much more phenylacetaldehyde is produced by chloroperoxidase than cytochrome P-450. Chloroperoxidase and cytochrome P-450 thus oxidize styrene by closely related oxygen-transfer mechanisms. Horseradish peroxidase does not oxidize styrene but does oxidize 2,4,6-trimethylphenol to 2,6-dimethyl-4-hydroxymethylphenol. The new hydroxyl group is partially labeled in incubations with H2(18)O but not H2(18)O2. The hydroxyl group thus appears to be introduced by addition of oxygen to the benzylic radical and water to the quinone methide intermediate but not by a cytochrome P-450-like oxene transfer mechanism. The results support the thesis that substrates primarily or exclusively react with the heme edge of horseradish peroxidase but are able to react with the ferryl oxygen of chloroperoxidase.     

The catalytic site of manganese peroxidase. Regiospecific addition of sodium azide and alkylhydrazines to the heme group.

Manganese peroxidase (MnP), which normally oxidizes Mn2+ to Mn3+, is rapidly and completely inactivated in an H2O2-dependent reaction by 2 equivalents of sodium azide. The inactivation is paralleled by formation of the azidyl radical and high yield conversion of the prosthetic heme into a meso-azido adduct. The meso-azido enzyme is oxidized by H2O2 to a Compound II-like species with the Soret band red-shifted 2 nm relative to that of native Compound II. The time-dependent decrease in this Compound II-like spectrum (t1/2 = 2.3 h) indicates that the delta-meso azido heme is more rapidly degraded by H2O2 than the prosthetic heme of control enzyme (t1/2 = 4.8 h). MnP is also inactivated by phenyl-, methyl-, and ethylhydrazine. The phenylhydrazine reaction is too rapid for kinetic analysis, but KI = 402 microM and kinact = 0.22/min for the slower inactivation by methylhydrazine. Reaction with phenylhydrazine at pH 4.5 does not yield iron-phenyl, N-phenyl, or meso-phenyl heme adducts. Ethylhydrazine inactivates the enzyme both at pH 4.5 and 7.0, but only detectably produces delta-meso-ethyl-heme at pH 7.0. Reconstitution of apo-MnP with hemin or delta-meso-ethylheme yields enzyme with, respectively, 50 and 5% of the native activity. The delta-meso-alkyl group thus suppresses most of the catalytic activity of the enzyme even though a Compound II-like species is still formed with H2O2. Finally, Co2+ inhibits the enzyme competitively with respect to Mn2+ but does not inhibit its inactivation by azide or the alkylhydrazines. The results argue that substrates interact with the heme edge in the vicinity of the delta-meso-carbon. They also suggest that Mn2+ and Co2+ bind to a common site close to the delta-meso-carbon without blocking the approach of small molecules to the heme edge. An active site model is proposed that accommodates these results.     

Mechanism-based inactivation of horseradish peroxidase by sodium azide. Formation of meso-azidoprotoporphyrin IX

Catalytic turnover of sodium azide by horseradish peroxidase, which produces the azidyl radical, results in inactivation of the enzyme with KI = 1.47 mM and kinact = 0.69 min-1. Inactivation of 80% of the enzyme requires approximately 60 equiv each of NaN3 and H2O2. The enzyme is completely inactivated by higher concentrations of these two agents. meso-Azidoheme as well as some residual heme are obtained when the prosthetic group of the partially inactivated enzyme is isolated and characterized. Reconstitution of horseradish peroxidase with meso-azidoheme yields an enzyme without detectable catalytic activity even though reconstitution with heme itself gives fully active enzyme. The finding that catalytically generated nitrogen radicals add to the meso carbon of heme shows that biological meso additions are not restricted to carbon radicals. The analogous addition of oxygen radicals may trigger the normal and/or pathological degradation of heme.     

Cardiolipin activates cytochrome c peroxidase activity since it facilitates H2O2 access to heme

In this work, the effect of liposomes consisting of tetraoleyl cardiolipin and dioleyl phosphatidylcholine (1 : 1, mol/mol) on the rate of three more reactions of Cyt c heme with H2O2 was studied: (i) Cyt c (Fe2+) oxidation to Cyt c (Fe3+), (ii) Fe...S(Met80) bond breaking, and (iii) heme porphyrin ring decomposition. It was revealed that the rates of all those reactions increased greatly in the presence of liposomes containing cardiolipin and not of those consisting of only phosphatidylcholine, and approximately to the same extent as peroxidase activity. These data suggest that cardiolipin activates specifically Cyt c peroxidase activity not only because it promotes Fe...S(Met80) bond breaking but also facilitates H2O2 penetration to the reaction center.     

Topology of the chloroperoxidase active site: regiospecificity of heme modification by phenylhydrazine and sodium azide.

Chloroperoxidase (CLP) from Caldariomyces fumago is rapidly and irreversibly inactivated by phenylhydrazine and H2O2 but not by H2O2 alone. Inactivation is characterized by a phenylhydrazine-to-CLP partition ratio of approximately 15, formation of trans-azobenzene, and formation of a sigma-bonded phenyl-iron heme complex with a characteristic absorption maximum of 472 nm. Anaerobic extraction of the heme complex from the protein, followed by exposure to dioxygen under acidic conditions, shifts the phenyl group from the iron to the porphyrin nitrogens and yields the four possible N-phenylprotoporphyrin IX regioisomers. Oxidation of the iron-phenyl complex within the intact protein by ferricyanide or high peroxide concentrations results in protein-directed phenyl migration to give exclusively the N-phenylprotoporphyrin IX regioisomers with the phenyl group on pyrrole rings A and C. CLP also catalyzes the H2O2-dependent oxidation of azide to the azidyl radical and is inactivated by azide in the presence of H2O2. Inactivation of CLP by azide and H2O2 results in loss of heme Soret absorbance and formation of delta-meso-azidoheme. These results suggest a topological model for the CLP active site and indicate that the tertiary structure of the enzyme permits substrates to interact with both the delta-meso heme edge and catalytic ferryl (FeIV = O) species, in agreement with the fact that CLP catalyzes both H2O2-dependent peroxidation and monooxygenation reactions.     

Guanylate cyclase from bovine lung. Evidence that enzyme activation by phenylhydrazine is mediated by iron-phenyl hemoprotein complexes

The mechanism of activation of soluble guanylate cyclase purified from bovine lung by phenylhydrazine is reported. Heme-deficient and heme-containing forms of guanylate cyclase were studied. Heme-deficient enzyme was activated 10-fold by NO but was not activated by phenylhydrazine. Catalase or methemoglobin enabled phenylhydrazine to activate guanylate cyclase 10-fold and enhanced activation by NO to over 100-fold. Heme-containing enzyme was activated only 3-fold by phenylhydrazine but over 100-fold by NO. Added hemoproteins enhanced enzyme activation by phenylhydrazine to 12-fold without enhancing activation by NO. Reducing or anaerobic conditions inhibited, whereas oxidants enhanced enzyme activation by phenylhydrazine plus catalase, and KCN had no effect. In contrast, enzyme activation by NO and NaN3 was inhibited by oxidants or KCN. NaN3 required native catalase, whereas phenylhydrazine also utilized heat-denatured catalase for enzyme activation. Thus, the mechanism of guanylate cyclase activation by phenylhydrazine differed from that by NO or NaN3. Guanylate cyclase activation by phenylhydrazine resulted from an O2-dependent reaction between phenylhydrazine and hemoproteins to generate stable iron-phenyl hemoprotein complexes. These complexes activated guanylate cyclase in the absence of O2, but lost activity after acidification, basification, or heating. Gel filtration of prereacted mixtures of [U-14C]phenylhydrazine plus hemoproteins resulted in co-chromatography of radioactivity, protein, and guanylate cyclase stimulating activity, and yielded a phenyl-hemoprotein binding stoichiometry of four under specified conditions (one phenyl/heme). [14C]Phenyl bound to heme-containing but not heme-deficient guanylate cyclase and binding correlated with enzyme activation. Moreover, reactions between enzyme and iron-[14C] phenyl hemoprotein complexes resulted in the exchange or transfer of iron-phenyl heme to guanylate cyclase and this correlated with enzyme activation.     

Structural characterization of lactoperoxidase in the heme environment by proton NMR spectroscopy

The heme environmental structures of lactoperoxidase (LP) have been studied by the use of hyperfine-shifted proton NMR and optical absorption spectra. The NMR spectra of the enzyme in native and cyanide forms in H2O indicated that the fifth ligand of the heme iron is the histidyl imidazole with an anionic character and that the sixth coordination site is possibly vacant. These structural characteristics are quite similar to those of horseradish peroxidase (HRP), suggesting that these may be prerequisite to peroxidase activity. The pH dependences of the spectra of LP in cyanide and azide forms showed the presence of two ionizable groups with pK values of 6 and 7.4 in the heme vicinity, which is consistent with the kinetic results. The group with pK = 7.4 is associated with azide binding to LP in a slow NMR exchange limit, which is in contrast to the fast entry of azide to HRP.     

Magnetic resonance spectral characterization of the heme active site of Coprinus cinereus peroxidase

Examination of the peroxidase isolated from the inkcap Basidiomycete Coprinus cinereus shows that the 42,000-dalton enzyme contains a protoheme IX prosthetic group. Reactivity assays and the electronic absorption spectra of native Coprinus peroxidase and several of its ligand complexes indicate that this enzyme has characteristics similar to those reported for horseradish peroxidase. In this paper, we characterize the H2O2-oxidized forms of Coprinus peroxidase compounds I, II, and III by electronic absorption and magnetic resonance spectroscopies. Electron paramagnetic resonance (EPR) and nuclear magnetic resonance (NMR) studies of this Coprinus peroxidase indicate the presence of high-spin Fe(III) in the native protein and a number of differences between the heme site of Coprinus peroxidase and horseradish peroxidase. Carbon-13 (of the ferrous CO adduct) and nitrogen-15 (of the cyanide complex) NMR studies together with proton NMR studies of the native and cyanide-complexed Coprinus peroxidase are consistent with coordination of a proximal histidine ligand. The EPR spectrum of the ferrous NO complex is also reported. Protein reconstitution with deuterated hemin has facilitated the assignment of the heme methyl resonances in the proton NMR spectrum.     

Oxygen exchange between the Fe(IV) = O heme and bulk water for the A2 isozyme of horseradish peroxidase

Resonance Raman spectra were observed for compound II of horseradish peroxidase A2, and the Fe(IV) = O stretching Raman line was identified at 775 cm-1. This Raman line shifted to 741 cm-1 upon a change of solvent from H2(16)O to H2(18)O, indicating occurrence of the oxygen exchange between the Fe(IV) = O heme and bulk water. The oxygen exchange took place only at the acidic side of the heme-linked ionization with pKa = 6.9.     

Scavenging of H2O2 and production of oxygen by horseradish peroxidase

Peroxidases catalyze many reactions, the most common being the utilization of H2O2 to oxidize numerous substrates (peroxidative mode). Peroxidases have also been proposed to produce H2O2 via utilization of NAD(P)H, thus providing oxidant either for the first step of lignification or for the "oxidative burst" associated with plant-pathogen interactions. The current study with horseradish peroxidase characterizes a third type of peroxidase activity that mimics the action of catalase; molecular oxygen is produced at the expense of H2O2 in the absence of other reactants. The oxygen production and H2O2-scavenging activities had temperature coefficients, Q10, of nearly 3 and 2, which is consistent with enzymatic reactions. Both activities were inhibited by autoclaving the enzyme and both activities had fairly broad pH optima in the neutral-to-alkaline region. The apparent Km values for the oxygen production and H2O2-scavenging reactions were near 1.0 mM H2O2. Irreversible inactivation of horseradish peroxidase by exposure to high concentrations of H2O2 coincided with the formation of an absorbance peak at 670 nm. Addition of superoxide dismutase (SOD) to reaction mixtures accelerated the reaction, suggesting that superoxide intermediates were involved. It appears that horseradish peroxidase is capable of using H2O2 both as an oxidant and as a reductant. A model is proposed and the relevance of the mechanism in plant-bacterial systems is discussed.     

Spectral characterization of diarylpropane oxygenase, a novel peroxide-dependent, lignin-degrading heme enzyme

Diarylpropane oxygenase, an H2O2-dependent lignin-degrading enzyme from the basidiomycete fungus Phanerochaete chrysosporium, catalyzes the oxygenation of various lignin model compounds with incorporation of a single atom of dioxygen (O2). Diarylpropane oxygenase is also capable of oxidizing some alcohols to aldehydes and/or ketones. This enzyme (Mr = 41,000) contains a single iron protoporphyrin IX prosthetic group. Previous studies revealed that the Soret maximum of the ferrous-CO complex of diarylpropane oxygenase is at approximately 420 nm, as in ferrous-CO myoglobin (Mb), and not like the approximately 450 nm absorption of the CO complex of the ubiquitous heme monooxygenase, cytochrome P-450. This spectral difference between two functionally similar heme enzymes is of interest. To elucidate the structural requirements for heme iron-based oxygenase reactions, we have compared the electronic absorption, EPR, and resonance Raman (RR) spectral properties of diarylpropane oxygenase with those of other heme proteins and enzymes of known axial ligation. The absorption spectra of native (ferric), cyano, and ferrous diarylpropane oxygenase closely resemble those of the analogous myoglobin complexes. The EPR g values of native diarylpropane oxygenase, 5.83 and 1.99, also agree well with those of aquometMb. The RR spectra of ferric diarylpropane oxygenase have their spin- and oxidation-state marker bands at frequencies analogous to those of aquometMb and indicate a high-spin, hexacoordinate ferric iron. The RR spectra of ferrous diarylpropane oxygenase have frequencies analogous to those of deoxy-Mb that suggest a high-spin, pentacoordinate Fe(II) in the reduced form. The RR spectra of both ferric and ferrous diarylpropane oxygenase are less similar to those of horseradish peroxidase, catalase, or cytochrome c peroxidase and are clearly distinct from those of P-450. These observations suggest that the fifth ligand to the heme iron of diarylpropane oxygenase is a neutral histidine and that the iron environment must resemble that of the oxygen transport protein, myoglobin, rather than that of the peroxidases, catalase, or P-450. Given the functional similarity between diarylpropane oxygenase and P-450, this work implies that the mechanism of oxygen insertion for the two systems is different.     

Resonance Raman spectroscopic evidence for heme iron-hydroxide ligation in peroxidase alkaline forms

Horseradish peroxidase will convert from a five-coordinate high-spin heme at neutral pH to a six-coordinate low-spin heme at alkaline pH. Though alkaline forms of other heme proteins such as hemoglobin and myoglobin are known to contain a heme-ligated hydroxide, alkaline horseradish peroxidase has been considered not to contain a ligated hydroxide. Several alternatives have been proposed which would be stronger field ligands than a hydroxide ion. In this report we provide resonance Raman evidence, using Soret excitation, that alkaline horseradish peroxidase does in fact contain a heme iron-ligated hydroxyl group. The band was located for isoenzymes C and A-1 by its sensitivity to 18O substitution and confirmed with 54Fe, 57Fe, and 2H. An isoenzyme of turnip peroxidase was investigated and found to also contain a ligated hydroxide at alkaline pH. The observed peroxidase Fe(III)-OH frequencies are 15-25 cm-1 higher than the corresponding frequencies of alkaline methemoglobin and metmyoglobin and correlate with changes in spin-state distribution. This is explained in the context of hydrogen bonding to a distal histidine which results in increased ligand field strength facilitating the formation of low-spin hemes. It has been demonstrated that the ferryl/ferric redox potential of horseradish peroxidase is markedly lowered at alkaline pH (Hayashi, Y., and Yamazaki, I. (1979) J. Biol. Chem. 254, 9101-9106). These observations are rationalized in terms of oxidation of a ligated ferric hydroxyl group facilitated through base catalysis by a distal histidine.     

Expression of a synthetic gene for horseradish peroxidase C in Escherichia coli and folding and activation of the recombinant enzyme with Ca2+ and heme

A synthetic gene encoding horseradish peroxidase isoenzyme C (HRP C) has been synthesized and expressed in Escherichia coli. The nonglycosylated recombinant enzyme (HRP C*) was produced in inclusion bodies in an insoluble inactive form containing only traces of heme. HRP C* was solubilized and conditions under which it folded to give active enzyme were determined. Folding was shown to be critically dependent upon the concentrations of urea, Ca2+, and heme and on oxidation by oxidized glutathione. Purification of active HRP C* from the folding mixture gave a peroxidase, with about half the activity of HRP C. Glycosylation is thus not essential for correct folding and activity. The C-terminal and N-terminal extensions to HRP identified previously in cloned cDNA sequences are also not required for correct folding. However, Ca2+ appears to play a key role in folding to give the active enzyme. The overall yield of purified active enzyme was 2-3%, but this could be increased by reprocessing material that precipitated during folding.     

Porphobilinogen oxygenase. Purification and evidence of its hemoprotein structure

Porphobilinogen oxygenase oxidizes porphobilinogen to 2-hydroxy-5-oxo-porphobilinogen. This enzyme isolated from wheat germ has been purified to homogeneity, as judged by polyacrylamide gel electrophoresis under both nondenaturing and denaturing conditions. The molecular weight of the enzyme formed from two identical (or very similar) polypeptide chains is 70,000. It has a pI of 9.0 indicating its cationic nature. The pure enzyme contains 1 mol of high-spin heme and 2 mol of non-heme iron. It requires both of these as well as molecular O2 and a reducing agent for catalytic activity. Although the enzyme has many characteristics of a peroxidase, hydrogen peroxide cannot substitute for oxygen and dithionite for catalysis. The catalytic reaction is not affected by catalase, superoxide dismutase, or by hydroxyl radical scavengers. A comparison between porphobilinogen oxygenase and a commercial preparation of horseradish peroxidase was made. The latter also catalyzes aerobic porphobilinogen oxidation, with dithionite as electron donor. Here the oxidation of porphobilinogen is inhibited by superoxide dismutase and was not affected by catalase.     

Analysis and computer simulation of aerobic oxidation of reduced nicotinamide adenine dinucleotide catalyzed by horseradish peroxidase.

Under suitable experimental conditions the aerobic oxidation of NADH catalyzed by horseradish peroxidase occurred in four characteristic phases: initial burst, induction phase, steady state, and termination. A trace amount of H2O2 present in the NADH solution brought about initial burst in the formation of oxyperoxidase. About 2 mol of oxyperoxidase was formed per mol of H2O2. When a considerable amount of the ferric enzyme still remained, the initial burst was followed by an induction phase. In this phase the rate of oxyperoxidase formation from the ferric enzyme increased with the decrease of the ferric enzyme and an approximately exponential increase of oxyperoxidase was observed. A rapid oxidation of NADH suddenly began at the end of the induction phase and the oxidation continued at a relatively constant rate. In the steady state, oxygen was consumed and H2O2 accumulated. A drastic terminating reaction suddenly set in when the oxygen concentration decreased under a certain level. During the reaction, H2O2 disappeared accompanying an accelerated oxidation of NADH and the enzyme returned to the ferric form after a transient increase of peroxidase compound II. Time courses of NADH oxidation, O2 consumption, H2O2 accumulation, and formation of enzyme intermediates could be simulated with an electronic computer using 11 elementary reactions and 9 rate equations. The results were also discussed in relation to the mechanism for oscillatory responses of the reaction that appeared in an open system with a continuous supply of oxygen.     

Structure and catalytic mechanism of horseradish peroxidase. Regiospecific meso alkylation of the prosthetic heme group by alkylhydrazines

Horseradish peroxidase is inactivated in a time-, H2O2-, and concentration-dependent manner by phenylethyl-, ethyl-, and methylhydrazine. The pseudo- first order kinetic constants for these inactivation reactions at pH 7 are: phenylethyl (KI = 115 microM, kinact = 1.5 min-1, partition ratio = 11), ethyl (KI = 145 microM, kinact = 0.08 min-1, partition ratio = 32), and methyl (KI = 3000 microM, kinact = 0.12 min-1, partition ratio = 80). At pH 5, the constants for the phenylethyl reaction change to KI = 1540 microM and kinact = 0.86 min-1. A transient absorbance at approximately 830 nm, suggestive of an isoporphyrin intermediate, is seen during these reactions. The prosthetic heme is converted by each of the three alkylhydrazines into the corresponding delta-meso-alkylated heme. Complete inactivation of the enzymes by methyl-, ethyl-, and phenylethylhydrazine is associated with alkylation of 60-70, 70, and 90%, respectively, of the prosthetic heme groups. The absence of N-alkylation and the high specificity for the delta-meso position, even with agents as small as methylhydrazine, strengthen the proposal that electron abstraction is mediated by the heme edge rather than the ferryl oxygen of horseradish peroxidase.