Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Oct 1.
Published in final edited form as: Curr Opin Biotechnol. 2017 Jul 13;47:102–111. doi: 10.1016/j.copbio.2017.06.005

Exploiting and Engineering Hemoproteins for Abiological Carbene and Nitrene Transfer Reactions

Oliver F Brandenberg 1, Rudi Fasan 2,*, Frances H Arnold 1,*
PMCID: PMC5617781  NIHMSID: NIHMS889974  PMID: 28711855

Abstract

The surge in reports of heme-dependent proteins as catalysts for abiotic, synthetically valuable carbene and nitrene transfer reactions dramatically illustrates the evolvability of the protein world and our nascent ability to exploit that for new enzyme chemistry. We highlight the latest additions to the hemoprotein-catalyzed reaction repertoire (including carbene Si–H and C–H insertions, Doyle-Kirmse reactions, aldehyde olefinations, azide-to-aldehyde conversions, and intermolecular nitrene C–H insertion) and show how different hemoprotein scaffolds offer varied reactivity and selectivity. Preparative-scale syntheses of pharmaceutically relevant compounds accomplished with these new catalysts are beginning to demonstrate their biotechnological relevance. Insights into the determinants of enzyme lifetime and product yield are providing generalizable cues for engineering heme-dependent proteins to further broaden the scope and utility of these non-natural activities.

Graphical Abstract

graphic file with name nihms889974u1.jpg

Introduction

The rapidly growing implementation of biocatalysis for production of chemicals is driven by the ability of enzymes to operate under mild, aqueous reaction conditions with high rates and selectivity. The result can be a significant reduction in industrial waste and cost of production compared to stoichiometric syntheses and other catalytic processes. The importance of directed evolution and other protein engineering tools for tuning enzyme activity, selectivity or stability to meet industrial process requirements has been thoroughly established [1,2]. However, a significant hurdle is presented by the fact that a number of valuable chemical transformations, including many catalyzed by synthetic catalysts, do not have biocatalytic counterparts. The necessary enzymes simply do not exist. One effective approach to creating new enzymes has been to engineer existing proteins to exhibit new, synthetically useful reactivity, mainly by exploiting their metallo- and other cofactors for new chemistry and improving those activities by directed evolution [3–5]. The resulting fully genetically-encoded catalysts could, at least in principle, be incorporated into in vivo biosynthetic pathways to access new products or provide alternative routes to existing products. Repurposing existing proteins for new chemistry, a trick used by evolution for eons, has proven particularly fruitful with heme-dependent proteins. In this review we cover the most recent developments as well as some of the key limitations in this growing area.

In a seminal paper, Coelho and coworkers demonstrated that cytochrome P450BM3 and other heme-containing proteins can catalyze the cyclopropanation of styrene using ethyl diazoacetate (EDA) as a carbene precursor [6]. Their work was inspired by the well-established capacity of ‘biomimetic’ metalloporphyrin complexes to catalyze carbene and nitrene transfer reactions, providing ample options for the introduction of C–C, C–N and other bonds into organic compounds. Coelho et al. further demonstrated that the activity and stereoselectivity of cytochrome P450BM3 as a cyclopropanation biocatalyst could be improved by mutating the protein. The evolvability of P450BM3 for this abiotic activity was underscored in subsequent reports showing that replacement of the conserved P450 axial cysteine residue with serine [7] or histidine [8] significantly boosts cyclopropanation activity. Furthermore, P450 variants catalyzed carbene N–H insertion to make new C–N bonds [9]. The new activities were not limited to carbene transfer reactions: P450 variants were shown to catalyze intramolecular nitrene C–H insertion reactions [10,11], and ‘P411s’ with an axial Cys-to-Ser mutation were shown to be competent catalysts for sulfimidation [12], aziridination [13], and intramolecular C–H amination [14,15] reactions. In each case, the observed activities were found to exceed those of previously-reported transition metal complexes.

Fasan and coworkers showed that these novel activities are not restricted to cytochrome P450s: the small, well-expressed, oxygen-transport hemoprotein myoglobin can become a powerful and selective cyclopropanation [16], carbene N–H and S–H insertion [17,18], and intramolecular C–H amination catalyst [19]. These contributions, including the underlying concepts of chemical intuition and enzyme promiscuity guiding the discovery and enhancement of these non-biological activities, have been reviewed in detail [5,20,21].

In this review, we focus on three recent, concurrent developments. First, we discuss how the earlier studies paved the way to the discovery of yet more non-natural carbene and nitrene transfer reactions mediated by engineered hemoproteins and how the cornucopia of existing hemoproteins is being leveraged to discover new, unprecedented reactivities. Second, we show how the field is transitioning from proof-of-concept studies towards implementation, with several studies demonstrating preparative-scale production of pharmaceutically relevant compounds. Third, we illustrate how the growing body of literature, including recent reports of mechanism-based enzyme inactivation, provide stimulus for further protein engineering to address what controls (and limits) these non-natural activities of hemoproteins.

New carbene transfer reactions catalyzed by engineered heme proteins

A compelling demonstration of the power of hemoprotein engineering to create a new enzyme is the recent report of biocatalytic carbon-silicon bond formation [22••], an activity not known in the biological world (Figure 1). Current rhodium, iridium and copper complexes developed to promote asymmetric carbene insertions into Si–H bonds of silanes all have limited catalytic efficiencies (<100 total turnovers (TTN)). Testing a variety of proteins for carbene insertion into the Si-H bond of phenyldimethylsilane, Kan et al. discovered that a variety of hemoproteins produce the organosilicon product with 10 to 50 TTN, although with low enantio-induction [22••]. A highly thermostable cytochrome c from the marine bacterium Rhodothermus marinus (Rma), however, catalyzed the transformation with excellent enantioselectivity (97% ee). Subsequent directed evolution by mutagenesis of active-site residues yielded an Rma cytochrome c triple variant (V75T,M100D,M103E), which provided 70% isolated product yield in preparative-scale (0.1 mmol) bacterial whole-cell reactions with 98% ee and 3,400 TTN. The enzyme exceeded the activity of the best-reported synthetic catalysts by up to 80-fold, while accepting a range of silane substrates and showing a high chemoselectivity for Si–H insertion over cyclopropanation or N–H insertion, a feature not exhibited by small-molecule transition metal catalysts.

Figure 1. Carbene transfer reactions catalyzed by engineered hemoproteins.

Figure 1

Open in a new tab

Carbene transfer by hemoproteins requires reduction of the heme iron from the Fe(III) to the catalytically active Fe(II) state. Subsequent encounter with a diazo compound results in release of N2 (not shown) and formation of a putative iron carbenoid reactive intermediate. Iron carbenoid formation in the heme pocket has been exploited for a range of carbene transfer reactions. The iron heme is returned to the catalytically active Fe(II) state (not shown) following carbene transfer. While most reactions were achieved with the native iron heme cofactor, C–H insertion activity (shown in lighter gray) required replacement of the iron heme with an Ir, Mn or Co porphyrin cofactor.

Mechanistic investigations of myoglobin-catalyzed S–H insertion reactions [18] provided a basis for developing catalysts that promote the [2,3]-sigmatropic rearrangement of allylic and propargylic sulfides in the presence of α-diazoester reagents [23•]. While this Doyle-Kirmse reaction provides a powerful C–C bond forming strategy for the synthesis of compounds bearing a new stereogenic center, synthetic catalysts exhibit only modest catalytic efficiency and low enantioselectivity (<10–50% ee). In contrast, the myoglobin triple mutant Mb(L29S,H64V,V68F) showed high product conversions (57–99%), high total turnovers (1,000–8,800 TTN), and up to 71% ee across a range of substrates. The scope of the reaction could be extended to conversion of propargylic sulfides to give allenes, which are valuable synthetic intermediates. A representative [2,3]-sigmatropic rearrangement product (ethyl 2-(phenylthio)pent-4-enoate) was isolated in 84% yield.

In another study, Tyagi et al. engineered myoglobins for aldehyde olefination reactions [24], providing a convenient and mild alternative to the widely used Wittig reaction. Screening of Mb active-site variants identified Mb(F43V,V68F), which converts a broad range of aryl aldehydes and alkyl α-diazo acetates into the corresponding olefin products with high catalytic efficiency (1,000–4,900 TTN) and very good to excellent trans selectivity (91–99% de). The TTNs reported for the Mb variant are 10- to 100-fold higher than those for similar transformations catalyzed by organometallic complexes working in organic solvents and at elevated temperature. Mb product conversion was found to not exceed 50%, however, due to inhibition by the phosphine (or arsine) oxide side product generated during the reaction.

Focusing on a similar reaction, Hauer and coworkers mined the E. coli proteome in search of enzymes capable of catalyzing the olefination of benzaldehyde in the presence of EDA [25]. The most promising biocatalyst they identified was YfeX, a heme-containing protein of unknown function. Under optimized conditions, this hemoprotein showed 92% selectivity for the trans olefin product, albeit at low product conversions (<5%) and with only 15–20 turnovers. Interestingly, YfeX-catalyzed benzaldehyde olefination was reported to take place in the absence of triphenylphosphine/arsine, suggesting that this transformation may involve a different mechanism than that proposed for myoglobin-catalyzed aldehyde olefination [24].

Progress in terms of substrate scope and synthetic utility is also being made with enzymatic cyclopropanation. Gober et al. recently reported cyclopropanation of dihydroalanine (DHA) residues contained in thiopeptides, a class of ribosomally synthesized peptides with a range of bioactivities, to yield 1-amino-2-cyclopropane carboxylic acid moieties [26]. Two P450 enzymes involved in monooxygenation reactions within the bacterial thiopeptide biogenesis pathway were engineered to harbor mutations previously found to be beneficial for promoting P450-mediated cyclopropanation [27•]. The resulting P450 variants were able to catalyze the cyclopropanation of DHA moieties in peptide model substrates and in complex linear and cyclic thiopeptide cores, offering new opportunities for thiopeptide functionalization. In another recent report, Tinoco et al. introduced a two-compartment, chemobiocatalytic strategy for realizing olefin cyclopropanation reactions in the presence of 2-diazo-1,1,1-trifluoroethane (CF3CHN2) as the carbene source. This transformation yields trifluoromethyl-substituted cyclopropanes, which are valuable building blocks for medicinal chemistry [28•]. In this system, gaseous CF3CHN2 was generated ex situ and subsequently utilized by whole-cell catalysts expressing engineered Mb cyclopropanation variants. This approach was applied to a range of vinylarene substrates to deliver the corresponding trifluoromethyl cyclopropanes in high yields and with excellent diastereo- and enantioselectivity.

A detailed account of artificial metalloenzymes containing non-native cofactors is beyond the scope of this short review, and this subject has been reviewed recently [4,29]. However, we will briefly describe how substitution of the heme cofactor with non- native metalloporphyrins is emerging as an alternative strategy for engineering biocatalysts for carbene and nitrene transfer reactions. Brustad and coworkers evolved variants of P450BM3 that are capable of incorporating iron-deuteroporphyrin IX, demonstrating the feasibility of engineering cytochrome P450 variants with orthogonal cofactor specificity [30]. Bordeaux et al. demonstrated that Mb variants incorporating a non-natural Mn- or Co-porphyrin IX cofactor catalyze intramolecular C–H amination reactions with sulfonyl azides, with reactivities comparable to their iron-containing counterpart [19]. This work was recently expanded by Sreenilayam et al., who reported the recombinant expression of Mb variants incorporating various first, second and third row transition metal cofactors (Mn, Co, Ru, Rh, Ir) [31]. The resulting Mb variants were catalytically active in cyclopropanation, N-H and S-H insertion reactions albeit with generally reduced efficiency compared to the iron-containing counterparts. Notably, Mn- and Co-substituted Mb variants showed activity for an intermolecular carbene C–H insertion reaction with phthalan and EDA, a reaction not supported by the iron- containing myoglobins and previously achieved only using an Ir-containing biocatalyst (vide infra).

By replacing the heme in myoglobin and the thermostable P450 CYP119 with Ir(Me)- mesoporphyrin IX, Hartwig and coworkers expanded the reactivity of these hemoproteins to carbene C–H insertion [32••,33]. Most notably, Ir(Me)-PIX-substituted variants of CYP119 catalyzed intramolecular carbene C–H insertion of various ethyl 2- diazo-2-(2-methoxyphenyl)acetate derivatives with 120–600 TTN and 76–98% ee, and up to 35,000 TTN under very low catalyst loadings (0.00125 mol%). The CYP119- derived biocatalysts were also found to be active for the intermolecular carbene C–H insertion reaction with phthalan and EDA. Ir-substituted CYP119 variants were also shown to be active catalysts for the cyclopropanation of olefins currently not accepted by Fe-containing enzymes, including various internal, aliphatic and electron-deficient alkenes [34]. These artificial enzymes have the shortcomings of requiring a cofactor containing rare and expensive metal (Ir) and specialized in vitro reconstitution or expression protocols for their assembly. This may limit the range of hemoprotein scaffolds that can be used and potentially complicates any effort to couple these enzymes to other activities in vivo. Whether these limitations can be overcome or whether engineered hemoprotein variants containing the native iron heme cofactor can access such challenging reactions as carbene C–H insertion remains to be shown.

Nitrene transfer reactions catalyzed by engineered hemoproteins

Among the new, non-natural activities discovered for hemoproteins are a variety of nitrene transfer reactions previously reported using transition metal complexes and, in one early case, a cytochrome P450 [35] (Figure 2). Recent studies established that cytochrome P450s and other heme-dependent enzymes (horseradish peroxidase, catalase, chloroperoxidase) and proteins (myoglobin, hemoglobin) can activate arylsulfonyl azide substrates and catalyze their conversion into sultam products via an intramolecular C–H amination reaction [11,14,15,19]. Subsequent studies expanded the scope of P450-catalyzed intramolecular C–H aminations in the context of carbonazidates, enabling the synthesis of protected forms of 1,2-aminoalcohols, which are valuable synthetic intermediates [10]. Other studies revealed that P450 variants from the serine-ligated cytochrome ‘P411’ lineage in combination with tosyl azide reagents can catalyze intermolecular nitrene addition to sulfides (sulfimidation) and alkenes (aziridination) [12,13].

Figure 2. Nitrene transfer reactions catalyzed by engineered hemoproteins.

Figure 2

Open in a new tab

Enzymatic nitrene transfer requires reduction of the iron heme from the Fe(III) to the catalytically active Fe(II) state. Encounter with azide substrates results in release of N2 (not shown) and formation of a putative iron nitrenoid reactive intermediate. Iron nitrenoid formation by hemoproteins has been exploited in a range of biocatalytic nitrene transfer reactions. The iron heme is returned to the catalytically active Fe(II) state (not shown) following nitrene transfer.

Additional synthetically useful nitrene-mediated transformations have appeared in the last two years. Prier et al. leveraged the high sulfimidation reactivity of P411 enzymes to catalyze the conversion of phenylallylsulfides into allylic amines [36•]. This reaction involves a spontaneous [2,3]-sigmatropic rearrangement of the sulfimidation product generated by the P411-catalyzed reaction between the allyl sulfide and tosyl azide. Carrying out these transformations in whole cells, the authors discovered that the intracellular environment results in loss of the thiophenol group, effectively leading to the formation of protected allylic amines. Starting from an initial P411 variant catalyzing this reaction with only moderate activity (<5% conversion), a significantly more efficient biocatalyst (P-5) was obtained after several rounds of directed evolution (79% conversion; 2,000 TTN). Importantly, the evolved enzyme also featured high enantioselectivity in both the sulfimidation (88–98% ee) and the sulfimidation/sigmatropic rearrangement reactions (68–82% ee).

Giovani et al. have shown that engineered myoglobins can catalyze the oxidative deamination of benzylic azides to yield aldehydes [37•], providing a biocatalytic alternative for aldehyde synthesis by alcohol oxidation with chromium- or manganese-based reagents. Myoglobin variant Mb(H64V,V68A) converted a broad range of primary azides to the corresponding aldehydes with up to 89% yield, exhibiting between 100- and 1000-fold higher catalytic activity than previously reported synthetic catalysts. In an extension of this study, investigation of a broader panel of hemoproteins revealed that engineered P450s catalyze the same reaction with high efficiency and selectivity [38]. A P450BM3 variant previously determined to be active in C–H amination reactions enabled the nearly quantitative conversion of benzyl azide to benzaldehyde at a catalyst loading of only 0.01 mol% (>10,000 TTN). This enhanced reactivity enabled the conversion of secondary azides to ketones (49–76%), a transformation that proceeds to only <2% conversion in the presence of Mb(H64V,V68A).

Interestingly, Carrera and coworkers recently reported the enzymatic transformation of benzyl azide to benzonitrile using a Rieske non-heme iron oxygenase (toluene dioxygenase) [39]. While not involving a heme enzyme, this reaction was proposed to involve catalytic intermediates that are reminiscent of those implicated in hemoprotein-catalyzed nitrene C–H insertion [11] and azide oxidation [37•].

Another noteworthy contribution relates to the recent application of engineered P450s for the synthesis of imidazolidin-4-ones via an intramolecular oxidative C–H amination [40]. While this reaction does not involve a nitrene-mediated process per se, it represents a previously undocumented enzymatic transformation useful for the creation of new C–N bonds, which further highlights the functional plasticity of cytochrome P450 enzymes.

Finally, Prier et al. recently demonstrated P450-mediated intermolecular amination of benzylic C–H bonds [41•]. The authors discovered that a P411 variant evolved for the imidation of allylic sulfides [36•] displays activity for the amination of 4-ethylanisole with tosyl azide. Subsequent rounds of directed evolution delivered variant P411CHA, providing up to 1300 TTN and 99% ee for benzylic C–H amination of a range of alkylarene substrates. Engineering the enzyme scaffold endows the native iron heme cofactor, by itself an inactive C–H amination catalyst, with activity for a challenging non-natural reaction.

Preparative-scale synthesis of cyclopropane pharmaceutical intermediates

Efforts to utilize these new biocatalysts for the preparation of high-value compounds are already starting to appear. In a first example, Wang et al. demonstrated the utility of an engineered P450-based carbene transfer catalyst, P450BM3 HStar, for producing the chiral precursor to the antidepressant levomilnacipran in a single biocatalytic step [8]. Applying P450BM3 HStar in whole-cell reactions, the cyclopropane core of levomilnacipran was obtained in 86% isolated yield, 96% de and 92% ee (Figure 3A). Another recent study used myoglobin-based cyclopropanation catalysts for preparative-scale synthesis: applying a structure-reactivity guided approach, Bajaj et al. engineered a panel of myoglobin variants to access trans-(1R,2R)-configured aryl-substituted cyclopropanes [42•], complementing the stereoselectivity of a previously developed trans-(1S,2S)-selective Mb(H64V,V68A) cyclopropanation catalyst [16]. Using these biocatalysts in whole-cell transformations yielded gram-scale amounts (up to 4.7 g) of the chiral cyclopropane cores of four drugs (tranylcypromine, tasimelteon, an investigational TRPV1 inhibitor, and ticagrelor) (Figure 3B–E). The desired trans-(1S,2S)- or trans-(1R,2R)-configured cyclopropane intermediates could be obtained in high yield (75–91%) and significantly higher diastereo- and enantioselectivity (98–99.9% de, 96–99.9% ee) than was possible with asymmetric cyclopropanation protocols adopted in the process-scale syntheses of these compounds. In the case of ticagrelor, the biocatalytic reaction offers a significantly shorter route (2 vs. 4–5 steps) to preparation of its chiral cyclopropane core compared to previously reported methods. In a parallel study, Hernandez and colleagues also reported the stereoselective synthesis of the ticagrelor cyclopropane (Figure 3B) using an engineered variant of group II truncated hemoglobin from Bacillus subtilis. Mutation of active-site residues in this protein led to an improved variant (Y25L,T45A,Q49A), which produced the desired enantiomer in 79% yield, 96% de and 99% ee in 0.4 mmol-scale reactions [43]. The above studies highlight the biotechnological relevance of these new enzymes and underscore the advantages of testing a variety of protein scaffolds to find the best starting points for achieving desirable transformations and selectivities.

Figure 3. Preparative-scale synthesis of cyclopropane-containing drug precursors.

Figure 3

Open in a new tab

Preparative-scale cyclopropanation reactions reported to date using engineered hemoproteins, providing access to various pharmaceutical intermediates with excellent yields, diastereo- and enantioselectivities.

Current limitations

P450 oxygenases, myoglobins, and other hemoproteins have evolved dedicated mechanisms to limit self-inactivation through oxidative damage by reactive oxygen species in P450s or oxidation to metmyoglobin. However, they have not had a chance to evolve comparable protection from the reactive intermediates produced in the active site during carbene and nitrene transfer reactions. This issue was recently addressed by Renata et al., who showed that a proficient P450 cyclopropanation variant, P450BM3 HStar, undergoes mechanism-based inactivation whereby the reactive carbene species generated in the active site is transferred to nucleophilic groups on the porphyrin ring and enzyme side chains [44•] (Figure 4A). These alkylation side reactions severely limit catalyst lifetime. Substitution of affected amino acid residues with non-nucleophilic residues resulted in a variant with ~2-fold improved activity (TTN). Nevertheless, the variant is still alkylated at various sites and has a short catalyst lifetime, highlighting significant room for further improvement.

Figure 4. Limitations to carbene and nitrene transfer efficiency.

Figure 4

Open in a new tab

(A) Carbene transfer biocatalysts suffer from alkylation of the porphyrin ring or nucleophilic protein residues by the reactive carbenoid intermediate, ultimately resulting in enzyme inactivation. Replacing nucleophilic residues, as shown by Renata et al. [44•] or otherwise promoting productive transfer to the desired substrate could ameliorate the problem and deliver improved cyclopropanation enzymes. (B) Nitrene transfer catalysts are affected by overreduction/protonation of the heme nitrene intermediate, resulting in unproductive consumption of azide substrate. For intramolecular C–H amination, mutation of a residue involved in proton transfer within the native P450 catalytic cycle (T268) results in higher product formation [11].

Nitrene transfer catalysts suffer from a different problem. All hemoprotein-catalyzed nitrene transfer reactions reported to date are accompanied by unproductive consumption of the azide substrate to give a sulfonamide (or carbamate) byproduct [10–14,19,36•,41•]. This unproductive pathway has been ascribed to over-reduction and protonation of the putative heme-nitrene intermediate (Figure 4B) [11]. Consistent with this, mutation of a highly conserved residue (T268) implicated in proton transfer to the hydroxoferryl intermediate as part of the native P450 catalytic cycle was found to improve the activity (TTN) of these enzymes as C–H amination catalysts [11,14]. The extent of this side reaction also depends on the nature of the substrate [11] and optimization of the enzyme active site for a target substrate was shown to improve desired product formation, possibly by facilitating productive interception of the nitrene [12]. More recently, heme substitution with an Ir-porphyrin provided another strategy to obtain P450-based catalysts capable of catalyzing the intramolecular C–H amination of arylsulfonyl azides with reduced propensity for azide reduction [45]. The TTN supported by these artificial enzymes, however, was comparable to those of previously reported hemoprotein-based catalysts, which suggests that other inactivating mechanisms may be at play. Another, currently unexplored strategy to overcome the problem associated with over-reduction of the iron-nitrene intermediate could be the engineering of P450 variants with altered (i.e., slower) rates of electron transfer [11].

Outlook - future opportunities for hemoprotein engineering

The preceding sections highlighted how diverse hemoprotein scaffolds can be adapted for various carbene and nitrene transfer reactions, with different scaffolds frequently showing altered reactivity or selectivity for a given transformation. Future efforts are anticipated to continue to take advantage of the wealth of natural hemoprotein scaffolds contained in genomic and structural databases (with >3900 hemoprotein structures currently deposited in the PDB) [46,47] to unveil catalysts with novel or superior activity or selectivity for abiological transformations. The ever-decreasing costs of DNA synthesis render it increasingly economical to obtain synthetic (hemoprotein) genes for rapid testing and engineering [48]. The nascent activities discovered in this vast collection can then be improved and tuned by directed evolution.

The studies reviewed here have begun to provide important clues and guidance for fine-tuning carbene and nitrene transfer reactivity in hemoproteins by protein engineering. Even non-enzyme hemoproteins like myoglobin or cytochrome c can be turned into proficient enzymes by “carving out” active sites through mutation of a few selected residues surrounding the heme pocket [16,22••,43]. In the context of P450s, Gober and Brustad showed that the T268A mutation, previously found to be beneficial for boosting cyclopropanation activity in P450BM3 [6], can be transferred to other cytochrome P450s to improve cyclopropanation activity [27•]. McIntosh et al. showed that the thermostable P450 CYP119 can accommodate substitution of its axial Cys residue with His, a mutation that also enhances cyclopropanation activity in P450BM3, by undergoing substantial structural rearrangement [49]. Intriguingly, soluble CYP119 enzymes with diverse spectroscopic characteristics were obtained for all 19 possible axial Cys substitutions, providing a small library of diverse potential starting points for further engineering.

Further guidance in hemoprotein engineering may come from computational approaches. Using model systems, recent DFT studies have investigated the nature of the putative iron carbenoid intermediate, shedding light on factors governing its formation and reactivity [50–52]. In particular, work by Sharon et al. on carbene N–H insertion [52] indicated that altering the basicity of the enzyme active site may allow access to aliphatic amines, potentially extending the substrate scope from the aromatic amines reported to date [9,17]. This proposal has not yet been tested experimentally.

To date, none of these new catalysts have been utilized in enzymatic cascades or metabolic engineering to create new biosynthetic pathways. With most carbene and nitrene transfer enzymes showing similar or higher turnovers in whole-cell contexts compared to purified enzyme, integrating them into metabolic pathways to produce non-natural chemicals is certainly tempting. However, successful realization may require further efforts to engineer the enzymes to exhibit lifetimes and kinetic properties (e.g., Km) that are more compatible with synthesis in living cells. A proof-of-concept for interfacing microbial metabolism with a non-natural reaction was recently published by Wallace and coworkers, who demonstrated styrene production from glucose in E. coli with subsequent cyclopropanation using a synthetic cyclopropanation catalyst [53•].

In conclusion, the field has come a long way since the initial report of hemoprotein-catalyzed olefin cyclopropanation just four years ago. The studies reviewed here provide ample inspiration for future protein engineering to discover yet more novel reactivities and to enhance the catalytic efficiencies of the current, nascent enzymes. Growing genomic databases and decreasing costs of DNA synthesis render it increasingly simple to sample natural hemoproteins for desired reactivities. Thus, propelled by the lack of biocatalytic alternatives for many synthetically useful reactions, we anticipate many more members to join the family of hemoproteins catalyzing non-natural reactions.

Highlights.

  • Engineered hemoproteins expand the biocatalytic repertoire

  • Different hemoprotein scaffolds offer diverse activities and selectivities

  • New enzymes enable preparative-scale syntheses of pharmaceutical intermediates

  • Insights into factors limiting enzyme lifetime offer new options for engineering

Acknowledgments

O.F.B. acknowledges support from the Deutsche Forschungsgemeinschaft (BR 5238/1-1); R.F. acknowledges support from the National Institute of Health (GM098628) and National Science Foundation grant CHE-1609550; F.H.A. acknowledges support from the National Science Foundation, Office of Chemical, Bioengineering, Environmental and Transport Systems SusChEM Initiative (grant CBET-1403077) and the Division of Molecular and Cellular Biosciences (grant MCB-1513007), the Defense Advanced Research Projects Agency Biological Robustness in Complex Settings Contract HR0011–15-C-0093, the Caltech Innovation Initiative (CI2) Program, and the Jacobs Institute for Molecular Engineering for Medicine at Caltech. The authors thank Dr. Jennifer Kan, Dr. Stephan Hammer, Dr. Xiongyi Huang, Kelly Zhang and Anders Knight for helpful discussions and comments on the manuscript.

Footnotes

Conflict of interest statement

The authors declare the following competing financial interest(s): Caltech and FHA have a financial ownership interest in Provivi, Inc., the company that sponsored a part of this research through the National Science Foundation STTR Program. FHA and Caltech may benefit financially from this interest if the company is successful in making product(s) that is/are related to this research. The terms of this arrangement have been reviewed and approved by Caltech in accordance with its conflict of interest policies.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Denard CA, Ren H, Zhao H. Improving and repurposing biocatalysts via directed evolution. Curr Opin Chem Biol. 2015;25:55–64. doi: 10.1016/j.cbpa.2014.12.036. [DOI] [PubMed] [Google Scholar]
  • 2.Otte KB, Hauer B. Enzyme engineering in the context of novel pathways and products. Curr Opin Biotechnol. 2015;35:16–22. doi: 10.1016/j.copbio.2014.12.011. [DOI] [PubMed] [Google Scholar]
  • 3.Prier CK, Arnold FH. Chemomimetic biocatalysis: exploiting the synthetic potential of cofactor-dependent enzymes to create new catalysts. J Am Chem Soc. 2015;137:13992–14006. doi: 10.1021/jacs.5b09348. [DOI] [PubMed] [Google Scholar]
  • 4.Hyster TK, Ward TR. Genetic Optimization of Metalloenzymes: Enhancing Enzymes for Non-Natural Reactions. Angew Chem Int Ed Engl. 2016 doi: 10.1002/anie.201508816. [DOI] [PubMed]
  • 5.Renata H, Wang ZJ, Arnold FH. Expanding the enzyme universe: accessing non-natural reactions by mechanism-guided directed evolution. Angew Chem Int Ed Engl. 2015;54:3351–3367. doi: 10.1002/anie.201409470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Coelho PS, Brustad EM, Kannan A, Arnold FH. Olefin cyclopropanation via carbene transfer catalyzed by engineered cytochrome P450 enzymes. Science. 2013;339:307–310. doi: 10.1126/science.1231434. [DOI] [PubMed] [Google Scholar]
  • 7.Coelho PS, Wang ZJ, Ener ME, Baril SA, Kannan A, Arnold FH, Brustad EM. A serine-substituted P450 catalyzes highly efficient carbene transfer to olefins in vivo. Nat Chem Biol. 2013;9:485–487. doi: 10.1038/nchembio.1278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wang ZJ, Renata H, Peck NE, Farwell CC, Coelho PS, Arnold FH. Improved cyclopropanation activity of histidine-ligated cytochrome P450 enables the enantioselective formal synthesis of levomilnacipran. Angew Chem Int Ed Engl. 2014;53:6810–6813. doi: 10.1002/anie.201402809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Wang ZJ, Peck NE, Renata H, Arnold FH. Cytochrome P450-Catalyzed Insertion of Carbenoids into N-H Bonds. Chem Sci. 2014;5:598–601. doi: 10.1039/C3SC52535J. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Singh R, Kolev JN, Sutera PA, Fasan R. Enzymatic C(sp)-H Amination: P450-Catalyzed Conversion of Carbonazidates into Oxazolidinones. ACS Catal. 2015;5:1685–1691. doi: 10.1021/cs5018612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Singh R, Bordeaux M, Fasan R. P450-catalyzed intramolecular C-H amination with arylsulfonyl azide substrates. ACS Catal. 2014;4:546–552. doi: 10.1021/cs400893n. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Farwell CC, McIntosh JA, Hyster TK, Wang ZJ, Arnold FH. Enantioselective imidation of sulfides via enzyme-catalyzed intermolecular nitrogen-atom transfer. J Am Chem Soc. 2014;136:8766–8771. doi: 10.1021/ja503593n. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Farwell CC, Zhang RK, McIntosh JA, Hyster TK, Arnold FH. Enantioselective Enzyme-Catalyzed Aziridination Enabled by Active-Site Evolution of a Cytochrome P450. ACS Cent Sci. 2015;1:89–93. doi: 10.1021/acscentsci.5b00056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.McIntosh JA, Coelho PS, Farwell CC, Wang ZJ, Lewis JC, Brown TR, Arnold FH. Enantioselective intramolecular C-H amination catalyzed by engineered cytochrome P450 enzymes in vitro and in vivo. Angew Chem Int Ed Engl. 2013;52:9309–9312. doi: 10.1002/anie.201304401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Hyster TK, Farwell CC, Buller AR, McIntosh JA, Arnold FH. Enzyme-controlled nitrogen-atom transfer enables regiodivergent C-H amination. J Am Chem Soc. 2014;136:15505–15508. doi: 10.1021/ja509308v. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bordeaux M, Tyagi V, Fasan R. Highly diastereoselective and enantioselective olefin cyclopropanation using engineered myoglobin-based catalysts. Angew Chem Int Ed Engl. 2015;54:1744–1748. doi: 10.1002/anie.201409928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Sreenilayam G, Fasan R. Myoglobin-catalyzed intermolecular carbene N-H insertion with arylamine substrates. Chem Commun (Camb) 2015;51:1532–1534. doi: 10.1039/c4cc08753d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Tyagi V, Bonn RB, Fasan R. Intermolecular carbene S-H insertion catalysed by engineered myoglobin-based catalysts. Chem Sci. 2015;6:2488–2494. doi: 10.1039/c5sc00080g. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bordeaux M, Singh R, Fasan R. Intramolecular C(sp(3))H amination of arylsulfonyl azides with engineered and artificial myoglobin-based catalysts. Bioorg Med Chem. 2014;22:5697–5704. doi: 10.1016/j.bmc.2014.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Gober JG, Brustad EM. Non-natural carbenoid and nitrenoid insertion reactions catalyzed by heme proteins. Curr Opin Chem Biol. 2016;35:124–132. doi: 10.1016/j.cbpa.2016.09.004. [DOI] [PubMed] [Google Scholar]
  • 21.Arnold FH. The nature of chemical innovation: new enzymes by evolution. Q Rev Biophys. 2015;48:404–410. doi: 10.1017/S003358351500013X. [DOI] [PubMed] [Google Scholar]
  • 22••.Kan SB, Lewis RD, Chen K, Arnold FH. Directed evolution of cytochrome c for carbon-silicon bond formation: Bringing silicon to life. Science. 2016;354:1048–1051. doi: 10.1126/science.aah6219. First biocatalytic carbon-silicon bond formation, catalyzed by engineered variants of cytochrome c from Rhodothermus marinus. The engineered catalyst shows excellent chemo- and enantioselectivities and exceeds the turnover of existing synthetic catalysts for carbon-silicon bond formation by 80 fold. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23•.Tyagi V, Sreenilayam G, Bajaj P, Tinoco A, Fasan R. Biocatalytic Synthesis of Allylic and Allenyl Sulfides through a Myoglobin-Catalyzed Doyle-Kirmse Reaction. Angew Chem Int Ed Engl. 2016;55:13562–13566. doi: 10.1002/anie.201607278. Myoglobin variants catalyze [2,3]-sigmatropic rearrangements of allylic and propargylic sulfides in the presence of α-diazoester reagents. This transformation was applied to a range of sulfide substrates with high conversion (up to 99%) and up to 71% enantioselectivity. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Tyagi V, Fasan R. Myoglobin-Catalyzed Olefination of Aldehydes. Angew Chem Int Ed Engl. 2016;55:2512–2516. doi: 10.1002/anie.201508817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Weissenborn MJL, Borlinghaus Sebastian A, Kuhn Niels, Kummer Miriam, Rami Stefanie, Plietker Fabian, Hauer Bernd, Bernhard Enzyme-Catalyzed Carbonyl Olefination by the E. coli Protein YfeX in the Absence of Phosphines. ChemCatChem. 2017;8:1636–1640. [Google Scholar]
  • 26.Gober JG, Ghodge SV, Bogart JW, Wever WJ, Watkins RR, Brustad EM, Bowers AA. P450-Mediated Non-natural Cyclopropanation of Dehydroalanine-Containing Thiopeptides. ACS Chem Biol. 2017 doi: 10.1021/acschembio.7b00358. [DOI] [PubMed] [Google Scholar]
  • 27•.Gober JG, Rydeen AE, Gibson-O’Grady EJ, Leuthaeuser JB, Fetrow JS, Brustad EM. Mutating a Highly Conserved Residue in Diverse Cytochrome P450s Facilitates Diastereoselective Olefin Cyclopropanation. Chembiochem. 2016;17:394–397. doi: 10.1002/cbic.201500624. Interesting report showing the transferability of the known activating T268A mutation across different P450 enzymes to yield improved cyclopropanation catalysts with diverse stereo- and enantioselectivities. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28•.Tinoco A, Steck V, Tyagi V, Fasan R. Highly Diastereo- and Enantioselective Synthesis of Trifluoromethyl-Substituted Cyclopropanes via Myoglobin-Catalyzed Transfer of Trifluoromethylcarbene. J Am Chem Soc. 2017;139:5293–5296. doi: 10.1021/jacs.7b00768. The authors developed a two-compartment chemobiocatalytic strategy for coupling ex situ generation of gaseous CF3CHN2 with myoglobin-catalyzed cyclopropanation to afford chiral trifluoromethyl-substituted cyclopropanes. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Upp DM, Lewis JC. Selective C-H bond functionalization using repurposed or artificial metalloenzymes. Curr Opin Chem Biol. 2017;37:48–55. doi: 10.1016/j.cbpa.2016.12.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Reynolds EW, McHenry MW, Cannac F, Gober JG, Snow CD, Brustad EM. An Evolved Orthogonal Enzyme/Cofactor Pair. J Am Chem Soc. 2016;138:12451–12458. doi: 10.1021/jacs.6b05847. [DOI] [PubMed] [Google Scholar]
  • 31.Sreenilayam G, Moore EJ, Steck V, Fasan R. Metal Substitution Modulates the Reactivity and Extends the Reaction Scope of Myoglobin Carbene Transfer Catalysts. Advanced Synthesis & Catalysis. 2017 doi: 10.1002/adsc.201700202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32••.Dydio P, Key HM, Nazarenko A, Rha JY, Seyedkazemi V, Clark DS, Hartwig JF. An artificial metalloenzyme with the kinetics of native enzymes. Science. 2016;354:102–106. doi: 10.1126/science.aah4427. Replacement of the heme cofactor with an Ir-porphyrin in P450 CYP119 delivers a biocatalyst capable of catalyzing intra- and intermolecular carbene C-H insertions with up to 35,000 turnovers. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Key HM, Dydio P, Clark DS, Hartwig JF. Abiological catalysis by artificial haem proteins containing noble metals in place of iron. Nature. 2016;534:534–537. doi: 10.1038/nature17968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Key HM, Dydio P, Liu Z, Rha JY, Nazarenko A, Seyedkazemi V, Clark DS, Hartwig JF. Beyond Iron: Iridium-Containing P450 Enzymes for Selective Cyclopropanations of Structurally Diverse Alkenes. ACS Cent Sci. 2017;3:302–308. doi: 10.1021/acscentsci.6b00391. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Svastits E, Dawson J, Breslow R, Gellman S. Functionalized nitrogen atom transfer catalyzed by cytochrome P-450. Journal of the American Chemical Society. 1985;107:6427–6428. [Google Scholar]
  • 36•.Prier CK, Hyster TK, Farwell CC, Huang A, Arnold FH. Asymmetric Enzymatic Synthesis of Allylic Amines: A Sigmatropic Rearrangement Strategy. Angew Chem Int Ed Engl. 2016;55:4711–4715. doi: 10.1002/anie.201601056. A substrate walk approach was applied to engineer cytochrome P411 variants catalyzing sulfimidation/sigmatropic rearrangement reactions to access allylic amines. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37•.Giovani S, Singh R, Fasan R. Efficient conversion of primary azides to aldehydes catalyzed by active site variants of myoglobin. Chem Sci. 2016;7:234–239. doi: 10.1039/c5sc02857d. Myoglobin variants catalyze oxidative deamination of primary azides to give aldehydes, providing a biocatalytic alternative to conventional aldehyde synthesis protocols based on alcohol oxidation with stoichiometric oxidizing reagents. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Giovani S, Alwaseem H, Fasan R. Aldehyde and Ketone Synthesis by P450-Catalyzed Oxidative Deamination of Alkyl Azides. ChemCatChem. 2016;8:2609–2613. doi: 10.1002/cctc.201600487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Vila MA, Pazos M, Iglesias C, Veiga N, Seoane G, Carrera I. Toluene Dioxygenase-Catalysed Oxidation of Benzyl Azide to Benzonitrile: Mechanistic Insights for an Unprecedented Enzymatic Transformation. Chembiochem. 2016;17:291–295. doi: 10.1002/cbic.201500653. [DOI] [PubMed] [Google Scholar]
  • 40.Ren X, O’Hanlon J, Morris M, Robertson J, Wong L. Synthesis of Imidazolidin-4-ones via a Cytochrome P450-Catalyzed Intramolecular C–H Amination. ACS Catalzsis. 2016;6:6833–6837. [Google Scholar]
  • 41•.Prier CK, Zhang RK, Buller AR, Brinkmann-Chen S, Arnold FH. Enantioselective, intermolecular benzylic C–H amination catalysed by an engineered iron-haem enzyme. Nature Chemistry. 2017 doi: 10.1038/nchem.2783. Directed evolution of a ‘P411’ variant delivered an enzyme catalyzing amination of benzylic C-H bonds with up to 1300 TTN and high enantioselectivity. While C-H amination typically requires precious metals in small-molecule catalysts, this work showcases new iron heme reactivity achieved by engineering the protein. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42•.Bajaj P, Sreenilayam G, Tyagi V, Fasan R. Gram-Scale Synthesis of Chiral Cyclopropane-Containing Drugs and Drug Precursors with Engineered Myoglobin Catalysts Featuring Complementary Stereoselectivity. Angew Chem Int Ed Engl. 2016;55:16110–16114. doi: 10.1002/anie.201608680. Myoglobin variants were engineered to deliver trans-(1R,2R) cyclopropanes, complementing existing trans-(1S,2S)-selective variants. The stereodivergent myoglobins were used to synthesize four drug precursors at gram scale with high diastereo- and enantioselectivity. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Hernandez KE, Renata H, Lewis RD, Kan SBJ, Zhang C, Forte J, Rozzell D, McIntosh JA, Arnold FH. Highly Stereoselective Biocatalytic Synthesis of Key Cyclopropane Intermediate to Ticagrelor. ACS Catalysis. 2016;6:7810–7813. doi: 10.1021/acscatal.6b02550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44•.Renata H, Lewis RD, Sweredoski MJ, Moradian A, Hess S, Wang ZJ, Arnold FH. Identification of Mechanism-Based Inactivation in P450-Catalyzed Cyclopropanation Facilitates Engineering of Improved Enzymes. J Am Chem Soc. 2016;138:12527–12533. doi: 10.1021/jacs.6b06823. A mass-spectrometric approach was use to reveal mechanism-based inactivation processes affecting P450-based cyclopropanation catalysts. Site-directed mutagenesis of two nucleophilic residues prone to alkylation resulted in a P450 variant with >2-fold higher activity. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Dydio P, Key HM, Hayashi H, Clark DS, Hartwig JF. Chemoselective, Enzymatic C-H Bond Amination Catalyzed by a Cytochrome P450 Containing an Ir(Me)-PIX Cofactor. J Am Chem Soc. 2017 doi: 10.1021/jacs.6b11410. [DOI] [PubMed] [Google Scholar]
  • 46.Smith LJ, Kahraman A, Thornton JM. Heme proteins--diversity in structural characteristics, function, and folding. Proteins. 2010;78:2349–2368. doi: 10.1002/prot.22747. [DOI] [PubMed] [Google Scholar]
  • 47.Poulos TL. Heme enzyme structure and function. Chem Rev. 2014;114:3919–3962. doi: 10.1021/cr400415k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Currin A, Swainston N, Day PJ, Kell DB. Synthetic biology for the directed evolution of protein biocatalysts: navigating sequence space intelligently. Chem Soc Rev. 2015;44:1172–1239. doi: 10.1039/c4cs00351a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.McIntosh JA, Heel T, Buller AR, Chio L, Arnold FH. Structural Adaptability Facilitates Histidine Heme Ligation in a Cytochrome P450. J Am Chem Soc. 2015;137:13861–13865. doi: 10.1021/jacs.5b07107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Khade RL, Fan W, Ling Y, Yang L, Oldfield E, Zhang Y. Iron porphyrin carbenes as catalytic intermediates: structures, Mossbauer and NMR spectroscopic properties, and bonding. Angew Chem Int Ed Engl. 2014;53:7574–7578. doi: 10.1002/anie.201402472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Khade RL, Zhang Y. Catalytic and Biocatalytic Iron Porphyrin Carbene Formation: Effects of Binding Mode, Carbene Substituent, Porphyrin Substituent, and Protein Axial Ligand. J Am Chem Soc. 2015;137:7560–7563. doi: 10.1021/jacs.5b03437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Sharon DA, Mallick D, Wang B, Shaik S. Computation Sheds Insight into Iron Porphyrin Carbenes’ Electronic Structure, Formation, and N-H Insertion Reactivity. J Am Chem Soc. 2016;138:9597–9610. doi: 10.1021/jacs.6b04636. [DOI] [PubMed] [Google Scholar]
  • 53•.Wallace S, Balskus EP. Interfacing microbial styrene production with a biocompatible cyclopropanation reaction. Angew Chem Int Ed Engl. 2015;54:7106–7109. doi: 10.1002/anie.201502185. Proof-of-concept study showing that bacterial production of styrene from glucose can interface with styrene cyclopropanation using a synthetic catalyst. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES