Making and Breaking Carbon-Carbon Bonds in Class C Radical SAM Methyltransferases

Abstract

Radical S-adenosylmethionine (SAM) enzymes utilize a [4Fe-4S]¹⁺ cluster and S-(5′-adenosyl)-L-methionine, (SAM), to generate a highly reactive radical and catalyze what is arguably the most diverse set of chemical reactions for any known enzyme family. At the heart of radical SAM catalysis is a highly reactive 5′-deoxyadenosyl radical intermediate (5′-dAdo●) generated through reductive cleavage of SAM or nucleophilic attack of the unique iron of the [4Fe-4S]⁺ cluster on the 5′ C atom of SAM. Spectroscopic studies reveal the 5′-dAdo● is transiently captured in an Fe-C bond (Ω species). In the presence of substrate, homolytic scission of this metal-carbon bond regenerates the 5′-dAdo● for catalytic hydrogen atom abstraction. While reminiscent of the adenosylcobalamin mechanism, radical SAM enzymes appear to encompass greater catalytic diversity. In this review we discuss recent developments for radical SAM enzymes involved in unique chemical rearrangements, specifically regarding class C radical SAM methyltransferases. Illuminating this class of radical SAM enzymes is especially significant as many enzymes have been shown to play critical roles in pathogenesis and the synthesis of novel antimicrobial compounds.

Graphical Abstract

All class C radical S-adenosylmethionine (SAM) methyltransferases (RSMTs) utilize two molecules of SAM (SAM₁ & SAM₂) in order to catalyze the methyl transfer reactions to Sp²-hybridized carbon centers.

INTRODUCTION

Arguably the most catalytically diverse enzyme superfamily found in Nature, the radical S-(5′-adenosyl)-L-methionine (SAM) enzymes utilize an iron-sulfur cluster (formally [4Fe-4S]¹⁺) and SAM to initiate radical reactions throughout all domains of life.(1, 2) Due to their substrate diversity, radical SAM enzymes have limited sequence homology and initially a minimal sequence motif, focused on the cluster binding site (CX₃CX₂C), was used to identify these enzymes.(1) Early biochemical characterization had shown that reduction of the [4Fe-4S]²⁺ cluster to the catalytically-active [4Fe-4S]⁺ state in vitro could be accomplished by using sodium dithionite or other strong reducing agents such as titanium(III)citrate.(3) However, later work revealed that artifacts arose when dithionite is used as a reductant in some systems(4) and therefore the preferred method is to use a physiological electron donor, such as a NADPH and/or a flavodoxin reducing system (5, 6). At the center of the radical SAM mechanism is a unique Fe site within the [4Fe-4S] cluster that is chelated by the α-amino and α-carboxylate groups of SAM(7). Reductive cleavage of SAM by the [4Fe-4S]¹⁺ cluster generates a 5′-deoxyadenosyl radical intermediate (5′-dAdo●). (1) The proposed mechanism for generation of this universal radical intermediate involves electron transfer from a [4Fe-4S]¹⁺ cluster to the SAM sulfonium ion, promoting S-[5′-C] bond cleavage and formation of the 5′- dAdo●(1) radical intermediate.(8) Recently, using advanced spectroscopic techniques, Horitani et al. demonstrated the 5′- dAdo● will add to the [4Fe-4S]²⁺ and generate an organometallic intermediate, termed “Ω” (Figure 1), where the unique iron of the [4Fe-4S] cluster is covalently bound to 5′-dAdo through an Fe-C bond (9). While the precise role in a universal mechanism remains unknown, this organometallic intermediate has since been identified in several radical SAM enzymes and may be common to all.(10, 11) Similar to adenosylcobalamin, Ω intermediate formation safely stores the radical until homolytic scission of the Fe-C bond regenerates the 5′- dAdo●, once substrate is properly positioned. The 5′-dAdo● then initiates catalysis through hydrogen atom abstraction from the substrate. Mechanistically speaking, this is where the common structure/function properties of radical SAM enzymes end, and the radical rearrangements become extremely diverse. In fact, radical SAM enzymes are often divided according to the type of chemistry they initiate and/or additional structural motifs they possess. At present, Radical SAM enzyme classes have expanded to include glycyl radical enzyme activating enzymes,(12, 13) enzymes catalyzing sulfur insertion,(14–18) mutases,(19, 20) enzymes involved in metallocofactor biosynthesis,(21) enzymes involved in complex dehydrogenation reactions,(22) and methyltransfer reactions that may or may not involve additional chemical rearrangements.(23)

Figure 1. Formation of the 5′-deoxyadenosyl radical in the radical SAM protein superfamily and generation of the catalytically competent methylene radical.

(A) Recent work has shown that the 5′-dAdo• is formed through an “Ω” intermediate, involving a Fe-C bond formed upon reductive cleavage of SAM.(80) Specifically, Ω may be formed directly by concerted reductive cleavage and nucleophilic attack of the unique iron on the 5′-carbon, or through an intermediate where SAM is reductively cleaved followed by recombination of the 5′-dAdo• with the [4Fe-4S]²⁺. (B) Generation of a methylene radical in the class C RSMTs. The 5′-dAdo• generated at the catalytic [4Fe-4S] cluster abstracts a hydrogen atom from the methyl group of a second molecule of SAM bound at distinct site. In CpdH SAM2 functions as part of a hydrogen atom relay,(41) while class C RSMTs have evolved to exploit a methylene radical to catalyze difficult methylation reactions.

Radical SAM methyltransferases (RSMTs) are further subdivided into four classes based upon protein architecture, cofactor/cosubstrate requirement, and predicted catalytic mechanism.(23–25) Class A methyltransferases methylate sp²-hybridized carbon centers of RNA bases using two conserved cysteine residues within the active site. Class B enzymes utilize a cobalamin cofactor to facilitate the methylation of both sp²-hybridized and sp³-hybridized centers.(26) Class C enzymes have high sequence homology with the enzyme anaerobic coproporphyrinogen III oxidase (initially named HemN, now CpdH)(27) and methylate sp²-hybridized centers but are catalytically diverse in their substrates (Figure 2). These reactions include biosynthesis of several natural products, as well as critical steps in anaerobic metabolism.(28–31) Lastly, Class D methyltransferases use methylenetetrahydrofolate to methylate sp²-hybridized centers.(23) In this review we focus on the class C RSMTs and identify some underlying mechanistic themes running through this important class of metalloenzymes.

Figure 2. Representative Class C radical SAM methyltransferases (RSMTs).

Class C RSMT are differentiated, as shown in the bottom panel, by their radical SAM domain and CpdH domain. NosN/NocN, YtkT, Jaw5, and TbtI are involved in carbon-carbon bond forming reactions while ChuW/HutW and CpdH are implicated in carbon-carbon bond breaking.

Anaerobic heme biosynthesis and implications for the class C RSMT mechanism.

Identified as an oxygen-independent coproporphyrinogen III oxidase, CpdH is essential to anaerobic heme biosynthesis, catalyzing the oxidative decarboxylation of two propionate groups of coproporphyrinogen III to protoporphyrinogen IX.(32) CpdH was one of the first radical SAM enzymes to be characterized biochemically when the requirement for SAM precursors, L-Methionine, and ATP was recognized in cell free assays.(33–36) CpdH was also one of the first radical SAM enzymes to be structurally characterized, revealing the (βα)₆ TIM barrel, a structural fold fairly conserved in the core of radical SAM enzymes.(1) Although controversial at first, a mechanistically important observation was the presence of two SAM binding sites (Figure 3) in the crystallographic model of CpdH.(37) The first molecule of SAM (SAM1) coordinates the unique iron atom of the catalytic [4Fe-4S] cluster, as expected. As can be seen in Figure 3 the second molecule (SAM2) binds adjacent to the first SAM molecule and interacts with several residues that were later found to be highly conserved across the class C RSMT family.(38) Mutagenesis of the amino acid residues binding SAM2 supports an essential function in catalysis.(38)

Figure 3. Cartoon representation of the CpdH model (PDB ID 1OLT) highlighting the relative spatial orientation of the [4Fe-4S] cluster, two SAM molecules, as well as the catalytic TIM-barrel (green) and C-terminal (magenta) domain.(81).

The [4Fe-4S] cluster is shown as spheres with the cysteine ligand and both SAM molecules (SAM1 and SAM2) shown in stick format. Iron, sulfur, carbon, nitrogen, and oxygen are colored orange, cyan, tan, blue, and red, respectively.

Based on the crystal structure and subsequent EPR studies of reduced [4Fe-4S]¹⁺ CpdH interacting with SAM and substrate coproporphyrinogen III, a catalytic mechanism was proposed(38, 39). The first step of the mechanism was based on a substrate-derived signal using coproporphyrinogen III that was regio-specifically labelled (¹⁵N or ²H) and revealed an allylic radical with the majority of spin density on the β-carbon of the propionate side chain (Figure 4, Step II).(38, 39) Layer et al. demonstrated early on that conversion of coproporphyrinogen III to protoporphyrin IX required two molecules of SAM and the intermediate porphyrin (initially termed “harderoporphyrinogen”, now monovinyl, monopropionyl deuteroporphyrinogen) could exit and re-enter the active site.(40) CpdH has subsequently been re-named to CpdH to more accurately depict the function and additional changes that have been observed in our understanding of heme biosynthetic pathways.(27) This intermediate is shown in Figure 4, step III. In order to properly position the next propionate group for catalysis, the porphyrin must rotate 90°, relative to the previous orientation. If the monovinyl, monopropionyl deuteroporphyrinogen intermediate (Figure 4, Step III) is not rotated, then an adduct could form (Figure 4, Step IV). However, initially there was no evidence for this adduct and it was suggested that the second SAM molecule (SAM2) was simply repositioned along with the intermediate to facilitate a second round of reductive cleavage by the catalytic [4Fe-4S] cluster. Regardless of the precise mechanism, one thing that could be agreed upon is that following the first decarboxylation/oxidation, the products of the reductive cleavage of SAM1 must exit the active site and a new molecule of SAM, (originally thought to be SAM2) must bind to the catalytic [4Fe-4S] cluster for the next round of catalysis. However, because a rational conclusion was to think that catalysis was limited to only the SAM molecule bound to the catalytic cluster (where reductive cleavage and formation of the 5′-dAdo● occurs), the role of the second SAM remained controversial.

Figure 4. Proposed mechanism of Adduct formation in CpdH (formally termed HemN).

Substrate (I) and two molecules of SAM (SAM1 and SAM2) are bound in the active site of CpdH. SAM1 coordinates the [4Fe-4S] cluster and is reductively cleaved to generate the 5′-dAdo radical (5′-dAdo•) and l-methionine. The 5′-dAdo• then abstracts a hydrogen atom from the methyl group of the SAM2 to generate a transient methylene radical. This radical is responsible for abstracting a hydrogen from the substrate to form a transient substrate radical (II).(41) Previous investigations have shown that radical rearrangement leads to the monovinyltripropionic porphyrin intermediate (III) which must be re-oriented within the active site,(40) a step that may involve exiting the active site altogether. If the intermediate is not reoriented then another round of radical generation will result in the covalent adduct (IV).

The controversy was eventually addressed when evidence for a mechanistic connection between all class C RSMT enzymes was revealed through a detailed investigation of CpdH by Ji et al. in 2019.(41) These investigators provided data that directly addressed two mechanistic points. First, the work provided evidence that during turnover, SAM2 functions in a “hydrogen atom relay” and is not cleaved, supporting a significantly different role for SAM2. In the modified mechanism, a third molecule of SAM (SAM3) enters the active site along with the monovinyl, monopropionyl deuteroporphyrinogen intermediate, specifically replacing the cleavage products of SAM1 (methionine and 5′-deoxyadenosine) at the catalytic [4Fe-4S] cluster. Second, they provided evidence for an unproductive shunt by isolating and identifying the SAM2-porphyrin adduct shown in Figure 4 (Step IV). Turnover experiments using isotopically-labelled SAM unequivocally demonstrated that the adduct was the result a SAM1-derived 5′-dAdo● abstracting a hydrogen atom from the methyl group of SAM2 and the subsequent methylene radical adding to an improperly positioned intermediate.(41) Specifically, if the monovinyl, monopropionyl deuteroporphyrinogen intermediate had not been repositioned, the SAM2 methylene radical attacks the double bond of the vinyl group, forming the adduct.

In addition to the mechanistic connection between CpdH and the class C RSMTs, Ji et al. also advanced the CpdH mechanism by proposing a new role for SAM2 in a hydrogen atom relay. This work correctly differentiated the roles of the two SAM binding sites in the oxidative decarboxylation reaction and identified the SAM2 binding site as the site for decarboxylation of both propionates. In light of their work, it stands to reason that the CpdH active site adopts two different conformations depending on whether coproporphyrinogen III or the monovinyl, monopropionyl deuteroporphyrinogen intermediate is bound. Since proper control of the radical requires a distinct electrostatic environment, problems will arise if another molecule of SAM (SAM3) binds to the catalytic [4Fe-4S] cluster prior to intermediate reorientation. In this case, reductive cleavage of SAM3 leads to hydrogen atom relay malfunction, and covalent adduct formation (Figure 4, IV). In this sense, the CpdH covalent adduct is an evolutionary whisper of the mechanistic role of two SAM molecules in the class C RSMTs because, as will be discussed, methylene radical formation is a mechanistic feature that differentiates these enzymes from other RSMTs. Members in this subclass are found across diverse bacterial phyla and share sequence homology with CpdH.(42) To date, the majority of enzymes(30, 31, 43) appear to be involved in the biosynthesis of complex secondary metabolites.(29, 44–47) The following sections will discuss representative reactions of class C RSMT and the enzymes that catalyze them.

Cyclopropanation: YtkT, and Jaw5.

Two radical SAM enzymes have now been identified in the cyclopropanation of natural products. YtkT was identified from a cluster of genes involved in the biosynthesis of Yatakemycin (YTM) and the enzyme C10Q is involved in the synthesis of a similar compound termed CC-1065.(44, 48) Yatakemycin is a DNA alkylating agent with broad activity against bacteria, fungal pathogens and a tumor cell line (49) (Figure 5). Deletion of the ytkT gene resulted in accumulation of a metabolic product similar to YTM which lacked the cyclopropane ring. This indicated that YtkT was responsible for providing the methyl group essential for spirocyclopropane ring formation.(44) YtkT Purification and reconstitution under experimental conditions typical for RS enzymes (anaerobic, strong reductant, SAM, etc.) resulted in conversion of the intermediate metabolite to YTM via methylation of a sp²-hybridized carbon atom. (44) The proposed mechanism of YtkT involved an enzymatic methyl-transfer from SAM to a double bond followed by transfer of a proton from the methyl group coupled to ring closing (44); however, research on these enzymes remains sparse. In fact, only a single paper has appeared on the mechanism of the cyclopropanation catalyzed by C10P and C10Q in CC-1065 biosynthesis.(48)

Figure 5. The general mechanism of the class C radical S-Adenosyl-L-methionine thiazole methyl transferase, YtkT.

The reaction has been shown to require two molecules of SAM and the red circle highlights the methyl group that is added by YtkT. Figure adapted from (46). Evidence for formation of a methylene radical and radical addition to a sp² hybridized carbon center was confirmed, for the first time, through the detection of a SAM-substrate adduct.(48)

Jaw5 is another enzyme that catalyzes cyclopropanation during the synthesis of the antifungal agent, jawsamycin (50). Based on sequence similarity to CpdH, and assignment of other open reading frames in this polyketide biosynthetic pathway, Jaw5 was proposed to be responsible for the installation of the cyclopropane modification Figure 6. Two mechanisms for Jaw5 have been presented,(23, 51) both require two molecules of SAM and a methylene radical intermediate. In the first mechanism, the methylene radical adds to the ene-one of the substrate, yielding a substrate radical that attacks the bridging methylene carbon of SAM to generate the cyclopropane product. An unexplained facet of this mechanism it that the radical cation of SAH will require additional reduction at a later step. In an alternative mechanism, the reduction step occurs sooner and the methylene radical is reduced to an ylide that attacks the ene-one followed by carbanion formation and subsequence production of cyclopropane and SAH. Both of these mechanisms were reviewed by Bauerle et al.(23) but an alternative mechanism, that has not been investigated further, is the possibility that Jaw5 functions with other proteins in the operon as part of a larger protein complex during the stepwise synthesis of jawsamycin.

Figure 6. General mechanism proposed for a novel class C RSMT, termed “cyclopropanase”,(47) that catalyzes the addition of cyclopropyl groups in the biosynthesis of the antifungal agent jawsamycin.(82).

Two mechanisms have been proposed, the first involving a methylene radical form on the second SAM molecule, which undergoes radical addition to the α,β-unsaturated polyketide chain. In a second mechanism the SAM methylene radical is transformed to its SAM ylide by acquiring an electron before adding to substrate with SAH being the leaving group during formation of the cyclopropyl moiety.

Thiopeptide Biosynthesis: TbtI and NosN/NocN.

Antibiotic biosynthesis represents another natural product pathway that often requires a class C RSMT, and with the rising incidence of antibiotic resistance (52) thiopeptide antibiotics may provide a viable solution. The ribosomally synthesized and posttraslationally modified peptide (RiPP) thiopeptide, thiomuracin, was shown to be dependent upon the gene tbtI, an annotated radical SAM enzyme.(43, 53) Through in vitro reconstitution and enzymatic timing, Mahanta et al. demonstrated that TbtI methylates the thiazole natural product (31, 43) (Figure 7). A more detailed investigation from this same group was one of the first investigations to address how two molecules of SAM, initially seen in the CpdH crystal structure, could be involved in the class C RSMT mechanism. Using deuterated SAM (D₃ at the methyl position only), Zhang et al later demonstrated that during TbtI catalysis, the 5′-dAdo product contained a single deuterium atom, providing evidence that SAM1 generates a 5′-dAdo● that abstracts a hydrogen atom from the methyl group of SAM2 during catalysis, producing a methylene radical as an intermediate (30). A mass increase of +2 in the TbtA hexazole core confirmed that the two deuterium atoms were incorporated into the substrate, with further assays suggesting that the third proton was contributed from solvent (30).

Figure 7. Methylation reaction catalyzed by the enzyme TbtI in the biosynthesis of thiomuracin, a ribosomally synthesized peptide antibiotic.

The figure has been adapted from Zhang et al.(30) and highlights the hydrogen atom transfer events. Investigation of TbtI also produced a SAM adduct consistent with addition of a SAM-derived methylene radical to a carbon-carbon double bond.

Synthesis of the antibiotic thiopeptide nosiheptide (NOS), and similarly nocathiacin, involve two class C RSMTs, NosN and NocN, respectively.(28, 54, 55) Initial cloning, sequencing, and characterization indicated that the indolic acid side chain ring is synthesized by genes nosL and nosN (54). Overall, the NOS biosynthetic pathway involves a complex set of posttranslational modifications to a short (13 amino acid) peptide (NosM).(28) A homologous set of genes are involved in the biosynthesis of nocathiacin (specifically nocI, nocN, and nocK, the latter is a gene fusion and contains both NosK and NosJ homologs).(56) Nocathiacin’s core macrocycle is initiated by thiazole formation, a cyclodehydration of the thiol side chains of six cysteine residues and is catalyzed by NosF, NosG, as well as NosH. Only one cysteine (Cys8) is left intact for acylation with 3-methyl-2-indolic acid (MIA).(57)

NosL catalyzes the conversion of tryptophan to MIA(58) and has been shown to tolerate fluorinated L-tryptophan substrates.(59) Initially, Yu et al proposed that NosN transferred a methyl group onto the indole side ring followed by hydroxylation and formation of the ester linkage.(60) Because of this activity, NosN was proposed to belong to the class C radical SAM methyltransferase family, although NosN was later shown to be multi-functional.(29) A bottleneck in understanding the NosN mechanism was the availability of substrate. A significant advance was made when it was shown that MIA or 3,4-dimethyl-2-indolic acid (DMIA) could be transferred to a substrate analog, N-acetylcysteamine (SNAC).(45) It was also suggested that the mechanism of NosN might be different, as 5′-methylthioadenosine (MTA) formation was detected during formation of 3,4-dimethylindolic acid (DMIA)-SNAC.(45) However, a subsequent investigation using MIA-SNAC as a substrate for NosN did not detect any MTA production.(29) LaMattina et al. also identified the 4-methylene-3-methlindolic acid-SNAC (MMIA-SNAC) intermediate and used isotopically labelled SAM to demonstrate that SAM, and not MTA, is the source of the methyl group on MMIA-SNAC. Specifically, when [methyl-²H₃]-SAM was used in their assay, the MMIA-SNAC generated showed a mass shift of +2, while the 5’-dAdo produced had a mass shift of +1.(29) These findings support the common theme among class C RSMT catalyzed reactions that a SAM-derived 5′-dAdo● abstracts the hydrogen atom from the methyl group of another molecule of SAM to generate a methylene radical that adds to a double bond on the substrate. In the case of NosN, the methylation is followed by ester linkage formation.(29, 57, 61) Regardless, the reported chemistry supports the utilization of two molecules of SAM, a unifying feature of all class C RSMTs. Generally speaking, the operon responsible for thiomuracin synthesis phylogenetically clusters within a group of enzymes responsible for thiazole C-methylation and Trp C-methylation, suggesting that TbtI and NosN have a more distant relationship to other CpdH-like or class C RSMT members (43).

Anaerobic heme degradation.

Biochemical characterization of CpdH (Initial named HemN), coupled with the minimal sequence motif used to identify radical SAM enzymes, lead to the mis-annotation of a number of CpdH-like genes as “anaerobic coproporphyrinogen III oxidase”.(62) Since that time, three functional classes have emerged from the collection of proteins initially annotated as CpdH.(62) This includes genuine CpdH enzymes,(40) heme chaperone proteins (HemW),(63) and anaerobic heme degrading enzymes (ChuW).(64) ChuW was identified in the hemolytic pathogen E. coli O157:H7(65, 66) as part of a heme uptake and degradation operon. This operon is expressed when iron levels are low under the control of the ferric uptake regulator (Fur). Low iron levels induce the “Coli heme utilization” (Chu) operon and ChuW is expressed, along with the proteins ChuX and ChuY, from a single promoter. All three genes are part of a larger heme uptake operon that is common to several enteric pathogens, including V. cholerae.(64, 67–69) Deletion of the entire operon, in several organisms, has been shown to impair the ability of the organism to utilize heme as the sole iron source.(70–72) Subsequent work has demonstrated that these operons contain enzymes involved in degradation of heme under both aerobic and anaerobic conditions.(65) In E. coli O157:H7, ChuW was shown to catalyze the anaerobic degradation of heme.(64) Similar to what has been proposed for other class C RSMTs, the reaction was proposed to proceed through formation of a methylene radical, addition of that methylene radical to a double bond followed by beta-scission of the porphyrin ring. The net result is the release of iron and production of a reactive tetrapyrrole termed “anaerobilin”.(64, 69) The structure of anaerobilin is shown in Figure 8, intermediate VI. Evidence that anaerobilin is toxic comes from characterization of a ChuY deletion strain.(73) ChuY has been shown to be an NADPH-dependent anaerobilin reductase.(68) The ability of the ChuY knockout strain to successfully infect human cells was greatly impaired.(73) This observation is consistent with the role of ChuY in reducing the reactivity of anaerobilin, similar to the function of biliverdin reductase in the aerobic catabolism of heme.

Figure 8.

Anaerobic heme degradation mechanism proposed for the enzyme HutW in Vibrio cholera.

In the current mechanism proposed for ChuW, it has been shown that the iron atom is not required for catalysis.(69) This is consistent with mechanisms proposed for other class C RSMT enzymes and underscores an exceptional level of control that the peptide environment imposes on the radical species being generated. A central tenant in the mechanism of class C RSMTs is the requirement for two molecules of SAM. One SAM molecule is responsible for generation of the 5′-dAdo● that abstracts a hydrogen atom from a second SAM molecule to generate a methylene radical. For ChuW, the methylene radical adds to a double bond at the meso carbon atom of the porphyrin ring.(69) A more general theme for all class C RSMTs is that the methylene radical is adds to a sp²-hybridized carbon center. For ChuW, and presumably HutW, protonation of the porphyrin ring facilitates a β-scission reaction and accounts for the incorporation of a single, non-exchangeable proton in the anaerobilin product.(64) Similar to Jaw5, further reduction, hydride formation and protonation may all be required to quench the radical, depending on the precise mechanism and the exact product structure.(64)

Although the ChuW and HutW proteins are phylogenetically distinct(74) when their protein sequences are compared, the two proteins likely perform similar chemistry when considered in the context of their heme uptake operons. Specifically, E. coli O157:H7 encodes chuW, chuX, and chuY behind a single promoter.(66) Similarly, V. cholerae, encodes hutW, hutX, and hutZ behind a single promoter.(67) A significant curiosity has arisen regarding the function of the third gene in each of these operons (ChuY in E. coli and HutZ in V. cholerae). ChuY utilizes NADPH to reduce the ChuW turnover product. In contrast, HutZ has a notably different structure and reported function when compared to ChuY. ChuY contains a Rossman fold and distinct NADPH binding site while HutZ has a split barrel fold and no recognizable site for nucleotide binding. It stands to reason that if the “W” genes are class C RSMTs responsible for the anaerobic opening of the porphyrin ring and iron liberation, then both organisms must deal with the resulting toxic tetrapyrrole product (Figure 8, Intermediate VI). ChuY has been shown to perform this function in E. coli O157:H7,(68) but there is no homologous enzyme in V. cholerae. This enigma was recently addressed when it was discovered that recombinantly-produced HutW could catalyze opening of the porphyrin ring as well as reduction of the tetrapyrrole product.(75) An interesting aspect of the investigation was the observation that NADPH could be used as the electron source without any intermediary electron transfer proteins such as a flavodoxin/ferredoxin or the (flavodoxin/ferredoxin):NADP+ oxidoreductase.(75) How this electron transfer occurs is unclear but direct electron transfer from NADPH to heme has been observed. Specifically for the “class X” cytochrome P450s. These P450 enzymes, such as P450nor, were the first P450s discovered to directly utilize NADPH.(76–78) Based on this literature, electron bifurcation mechanisms, and studies focused on the transient species observed during NAD(P)H oxidation it is tempting to speculate on a new mechanism for the class C RSMT HutW (Figure 8). The mechanism shown in Figure 8 also addresses at least one issue that was unresolved for the proposed ChuW mechanism. Specifically, in order to open the porphyrin ring, at least two electrons and two protons are required. One electron is required for reductive cleavage of SAM (radical generation) and the other for reduction of a radical intermediate on the tetrapyrrole. Given that the substrate heme most likely contains a ferrous iron atom in vivo, it was convenient to suggest that one electron could come from the ferrous heme. It has since been demonstrated that metal-free porphyrins can be used as substrates for ChuW and HutW, thus eliminating a possible role for the metal ion in the reaction.(69, 75)

Despite the differences in the electron transfer mechanisms described above, there are mechanistic similarities between ChuW and HutW, an acidic residue has been proposed to donate the first proton to the tetrapyrrole intermediate in both enzymes. A second proton and electron are required to quench the radical intermediate that is generated. In this regard, the mechanism proposed for HutW (Figure 8) addresses this issue. Specifically, a radical ion is formed following reduction of the [4Fe-4S] cluster. The reduced (formally 1+) cluster reductively cleaves SAM1 to generate a 5′-dA radical (5′-dA•) that extracts a hydrogen atom from SAM2 to generate a methylene radical. The methylene radical adds to the double bond of the porphyrin ring and protonation by the enzyme sets a radical rearrangement into motion that results in breaking of the porphyrin ring. Consistent with what has been proposed in the literature for a multistep hydride transfer,(79) a powerful one-electron oxidant, such as the radical intermediate shown in Figure 8, will complete the multistep electron transfer process and oxidation of the first molecule of NADPH (Figure 8, conversion of molecule V to molecule VI). Two additional hydride transfers will further reduce the tetrapyrrole as shown in Figure 8. The product (Figure 8, molecule 8) has been identified by mass spectroscopy and fragmentation but it is important to note that it has not been isolated in quantities that are suitable for NMR or small molecule crystallography.(75)

The proposed mechanism and observations reported by Brimberry et al. raise significant questions regarding the electron transfer reaction and the specific sequence of events. In particular, does the [4Fe-4S] cluster participate in the reduction of the tetrapyrrole or do conformational changes facilitate the direct transfer of a hydride to the bridging β- and δ-carbon atoms? Evidence presented for P450nor would suggest that the latter is true.(78) In this case, an exciting proposal is one whereby conformation-dependent electron transfer pathways exist in HutW. In this case, the first electron transfer pathway involves transfer of a single electron to the [4Fe-4S]²⁺ cluster while the second electron transfer pathway directs electrons and a proton to the bridging β- and δ-carbon atoms of the tetrapyrrole intermediate (Figure 8, conversion of molecule V to molecule VI).

Conclusions

Given the mechanistic diversity that has been reported for the class C RSMTs and the critical role that these enzymes play in the biosynthesis of novel therapeutic compounds, a complete understanding the structure/function relationships will have tremendous potential for biotechnological innovation. In addition, the unique combination of control exerted on the organic radical(s) require for catalysis, combined with what seems to be an ever-increasing repertoire of inorganic catalytic diversity associated with the [4Fe-4S] cluster, will also have much broader implications.

Highlights.

• Class C radical SAM enzymes

• Carbon-carbon bond breaking and making with a [4Fe-4S] cluster

• Evolution of a universal mechanism