Mechanistic insights into C–H activation from radical clock chemistry: oxidation of substituted methylcyclopropanes catalyzed by soluble methane monooxygenase from Methylosinus trichosporium OB3b

https://doi.org/10.1016/S0167-4838(00)00199-0Get rights and content

Abstract

The soluble methane monooxygenase (MMO) system isolated from Methylosinus trichosporium OB3b catalyzes the adventitious oxidation of alkyl substituted methylcyclopropanes. If the chemical mechanism of C–H activation by MMO involves formation of a radical or carbocation intermediate at the methyl C–H of these ‘radical clock’ substrates, then cyclopropyl ring opened alcohols may appear in the product mixture due to rearrangement of the intermediate. The lifetime of radical intermediates can be determined from known rearrangement rate constants, kr. Rearrangement was observed during the oxidation of 1,1,2,2-tetramethylcyclopropane (kr=1.7–17.5×108 s−1, 30°C) but not for cis- or trans-1,2-dimethylcyclopropane (kr=1.2–6.4×108 s−1, 30°C) or the very fast radical clock, trans-2-phenylmethylcyclopropane (kr=3.4×1011 s−1, 30°C). The results show that the occurrence of rearranged products fails to correlate with either the chemical nature of the C–H bond being broken, which is very similar for all of the methylcyclopropanes studied here, or the magnitude of the radical kr value. This study suggests that the steric properties of the substrate play an important role in determining the outcome of the reaction. Substrates with bulky substituents near the C–H bond that is attacked appear to yield intermediates with sufficient lifetimes to rearrange. In contrast, substrates with less steric bulk are postulated to be able to approach the reactive oxygen species in the MMO active site more closely so that intermediates are either rapidly quenched or undergo subsequent interaction with the dinuclear iron cluster of MMO that prevents rearrangement.

Introduction

The methane monooxygenase (MMO) system of methanotrophic bacteria catalyzes one of the most challenging reactions in chemistry, the functionalization of the inert C–H bond in methane at ambient temperature. The soluble form of MMO has been isolated from several sources, but the enzymes from Methylosinus trichosporium OB3b [1], [2] and Methylococcus capsulatus (Bath) [3], [4] have been most thoroughly studied. Each consists of three protein components: a hydroxylase (MMOH), a reductase, and a regulatory component termed ‘B’. The hydroxylation reaction is carried out at the active site of MMOH, which contains a bis-μ-hydroxo bridged diiron cluster [5], [6]. Due to the similarities in oxidation substrates and products, MMO is considered to be the non-heme analog of cytochrome P450 monooxygenase (P450) [7], [8].

In recent years, the mechanism of catalysis by MMO has been the focus of extensive studies [2], [4]. The reaction of MMO consists of two major processes: oxygen activation and substrate oxidation (Fig. 1) [9]. The product of the first process is an enzyme intermediate termed compound Q (Q) that contains a powerful oxidant associated with the active site diiron cluster. Q reacts with the alkane substrate in the second process to yield the alcohol enzyme–product complex termed compound T from which the product is released as the enzyme returns to its resting state. Much less is currently known about the chemical mechanism of the Q reaction with alkanes than about the oxygen activation process. From spectroscopic studies, Q has been shown to contain two Fe(IV) atoms [13], and to be electronically equivalent to the heme Fe(IV) π-cation radical intermediate postulated to be the reactive oxidizing species of P450 [8], [9], [14], [15]. Accordingly, mechanistic studies of the substrate activation steps of MMO have suggested an oxygen rebound mechanism similar to the molecular mechanism proposed for P450 [14]. In this type of mechanism, the hydrocarbon oxidation process starts with the hydrogen atom abstraction from the hydrocarbon substrate by the oxidizing species to yield an intermediate substrate radical (Fig. 2). The process is completed by the rebound of the hydroxyl group from the iron to the substrate radical. This radical based mechanism is supported in the case of MMO by the observation of a large deuterium kinetic isotope effect (KIE) for the reaction of Q with deuterated methane [16], the reported release of radical intermediates from the active site [17], and the partial inversion of stereochemistry in the MMO catalyzed oxidation of chiral 1-[1H,2H,3H]ethane [18]. However, these studies have also revealed the complexity of the MMO catalyzed reaction in aspects that cannot be fully explained by the simple P450-like mechanism. For example, the primary deuterium KIE for the methane reaction with Q (50–100) is much larger than the classical limit, suggesting that a tunneling or similar complex process must be invoked. Also, the observation of incomplete racemization of the product from chiral ethane requires that the rebound reaction occurs on a time scale that is probably too fast for physical translocation of molecules in the active site.

Another frequently used tool in the detection of radical intermediates in enzyme catalyzed reactions is the so-called radical clock mechanistic probe. These substrates are capable of producing diagnostic products by undergoing fast reorganization if a substrate radical is formed during reaction [19]. Unfortunately, various clock oxidation experiments with MMO have produced conflicting results, providing evidence both for and against the involvement of substrate radical intermediates [20], [21], [22], [23], [24]. The earliest studies using cyclopropylbenzene with M. capsulatus MMO [20] and the more definitive substrate, 1,1-dimethylcyclopropane, with M. trichosporium MMO [21] indicated the formation of substrate radicals and/or cations during the reaction, supporting a P450-like reaction in which a discrete intermediate is generated. However, more recent studies using M. capsulatus MMO [22] and radical clock substrates such as 1,2-dimethylcyclopropane and aryl substituted methylcyclopropanes failed to evoke any evidence for a radical intermediate during oxidation. As a result, new concerted models for C–H activation by MMO were invoked to account for the results.

One possible source of misleading results from radical clock substrates is the distortion in the probe structure which would alter the rearrangement rate constant kr. This was addressed using another type of probe, methylcubane, which has a constrained cubane structure that cannot be distorted as it is bound in the active site [23], [24], [25]. Nevertheless, the cubane structure can rearrange at a characteristic rate if a radical is formed on the methyl side chain. Once again, significantly different results were initially reported for MMO isolated from different bacteria. The M. capsulatus MMO catalyzed oxidation yielded only the methyl hydroxylation product, which was inconsistent with formation of an intermediate radical with a reasonable lifetime for the oxygen rebound mechanism [25]. In contrast, the M. trichosporium MMO catalyzed methylcubane oxidation yielded all possible oxidation products, but the major product was tentatively identified from mass spectrum assignments as a rearranged species originating from a cubylcarbinyl radical intermediate [23]. The results of the latter study are consistent with a P450-like mechanism, but the fact that the distribution of products is not the same as expected for a radical based reaction in solution led to the proposal that steric interactions of the substrates in the enzyme active site affect the outcome of the reaction. In recently published studies, the M. capsulatus MMO experiments were repeated and all of the products identified in the M. trichosporium study were, in fact, detected [24]. Additionally, an authentic standard was prepared which allowed the majority product to be identified as a rearrangement based on an intermediate cation rather than a radical. The source and chemical nature of the cationic species are not known.

In the past, it has proven beneficial to conduct structural and mechanistic studies using MMO from both M. trichosporium and M. capsulatus because the enzymes provide independent views of what is probably the same underlying chemical mechanism. Accordingly, we have reinvestigated and extended the study of the oxidation of the cyclopropane family of radical clocks using the M. trichosporium MMO. In contrast to the results reported for recent studies of M. capsulatus MMO, the current study shows that at least some of these substrates react through radical or cationic species. However, the results also demonstrate the important role that steric properties of the substrates play on the rate and products of the reaction.

Section snippets

Enzyme preparation

The components of MMO were purified to homogeneity from M. trichosporium OB3b. Details of the growth of the bacterial strain and protein purification procedures were published previously [10], [26]. The specific activity of MMOH preparations used for the experiments were in the range of 600–1000 nmol/min/mg, assayed at 23°C.

Chemicals

trans-2-Phenylmethylcyclopropane and the corresponding authentic materials for conventional and rearranged hydroxylation products were generously provided by M. Newcomb,

Results

Here we report the oxidation of four different alkyl substituted methylcyclopropanes catalyzed by the reconstituted MMO system from M. trichosporium OB3b. When a radical is generated on the methyl carbon of these substrates, the subsequent reactions can yield either ring opened, rearranged product alcohol (rP) or the conventional unrearranged product cyclopropyl alcohol (uP) (Fig. 3). The kr values have been determined for these substrates in solution [27], [28]. Provided that this rate is

The origin of the inconsistency in radical clock studies of MMO

In this study, we have investigated the oxidation of a series of methylcyclopropane radical clock substrates catalyzed by MMO from M. trichosporium OB3b. The study was designed to allow comparison of the reactions of structurally similar probes that have a range of kr values and also to allow comparison with analogous studies using the MMO from M. capsulatus (Bath). Among the radical clock substrates investigated here, the cis- and trans-1,2-dimethylcyclopropanes and trans

Conclusion

The oxidation reaction catalyzed by M. trichosporium MMO yields rearranged products for sterically hindered methylcyclopropane probes, but not for others. The range of results reported here provides evidence for both the formation of intermediates during reaction and a complex process that cannot be accounted for completely by the original (radical based) oxygen rebound mechanism. It suggests that substrates approach the activated oxygen center differently depending on their size and bulk, and

Acknowledgments

This work was supported by National Institutes of Health Grant GM40466.

References (46)

  • J.D. Lipscomb

    Biochemistry of the soluble methane monooxygenase

    Annu. Rev. Microbiol.

    (1994)
  • B.J. Wallar et al.

    Dioxygen activation by enzymes containing binuclear non-heme iron clusters

    Chem. Rev.

    (1996)
  • A.L. Feig et al.

    Reactions of non-heme iron(II) centers with dioxygen in biology and chemistry

    Chem. Rev.

    (1994)
  • P.R. Ortiz de Montellano, Cytochrome P450: Structure, Mechanism, and Biochemistry, Plenum Press, New York,...
  • J.D. Lipscomb et al.

    MMO – P450 in wolf’s clothing

    J. Biol. Inorg. Chem.

    (1998)
  • S.K. Lee et al.

    Oxygen activation catalyzed by methane monooxygenase hydroxylase component: Proton delivery during the O–O bond cleavage steps

    Biochemistry

    (1999)
  • S.-K. Lee et al.

    A transient intermediate of the methane monooxygenase catalytic cycle containing a FeIVFeIV cluster

    J. Am. Chem. Soc.

    (1993)
  • T.J. McMurry, J.T. Groves, in: P.R. Ortiz de Montellano (Ed.), Cytochrome P-450: Structure, Mechanism, and...
  • L. Shu et al.

    An Fe(IV)2O2 diamond core structure for the key intermediate Q of methane monooxygenase

    Science

    (1997)
  • J.C. Nesheim et al.

    Large isotope effects in methane oxidation catalyzed by methane monooxygenase: Evidence for C–H bond cleavage in a reaction cycle intermediate

    Biochemistry

    (1996)
  • P.C. Wilkins et al.

    Biological methane activation involves the intermediacy of carbon-centered radicals

    Eur. J. Biochem.

    (1992)
  • N.D. Priestley et al.

    Cryptic stereospecificity of methane monooxygenase

    J. Am. Chem. Soc.

    (1992)
  • D. Griller et al.

    Free radical clocks

    Acc. Chem. Res.

    (1980)
  • Cited by (30)

    • Oxygen-18 tracer studies of enzyme reactions with radical/cation diagnostic probes

      2005, Biochemical and Biophysical Research Communications
    • Substrate radical intermediates in soluble methane monooxygenase

      2005, Biochemical and Biophysical Research Communications
      Citation Excerpt :

      The failure to observe DMPO- or POBN-trapped radicals from substrate turnover reactions using the complete M. trichosporium OB3b MMO system could have many origins, the most obvious of which is that substrate radicals may not have been made. However, some of the substrates investigated here, such as ethane and 1,1,2,2-tetramethylcyclopropane, are thought to make radical intermediates based on the formation of rearranged products [14,23], yet do not produce any spin adducts. These observations make the alternative possibility that radical intermediates formed, but do not escape from the active site to contact the radical-trap more attractive.

    View all citing articles on Scopus
    View full text