- Email: [email protected]
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases
Function of genetically encoded pyrrolysine in corrinoiddependent methylamine methyltransferases Joseph A Krzycki Methanogenesis from trimethylamine, dimethylamine or monomethylamine is initiated by a series of corrinoiddependent methyltransferases. The non-homologous genes encoding the full-length methyltransferases each possess an in-frame UAG (amber) codon that does not terminate translation. The amber codon is decoded by a dedicated tRNA, and corresponds to the novel amino acid pyrrolysine in one of the methyltransferases, indicating pyrrolysine to be the 22nd genetically encoded amino acid. Pyrrolysine has the structure of lysine with the eN in amide linkage with a pyrroline ring. The reactivity of the electrophilic imine bond is the basis for the proposed function of pyrrolysine in activating and optimally orienting methylamine for methyl transfer to the cobalt ion of a cognate corrinoid protein. This reaction is essential for methane formation from methylamines, and may underlie the retention of pyrrolysine in the genetic code of methanogens. Addresses Department of Microbiology, OSU Biochemistry Program, Ohio State University, Columbus, Ohio 43210, USA e-mail: [email protected]
Current Opinion in Chemical Biology 2004, 8:484–491 This review comes from a themed section on Mechanisms Edited by Hung-Wen Liu and Ja`nos Re´tey Available online 9th September 2004 1367-5931/$ – see front matter # 2004 Elsevier Ltd. All rights reserved. DOI 10.1016/j.cbpa.2004.08.012
Abbreviations DMA dimethylamine meTHF methyl-tetrahydrofolate MMA monomethylamine TMA trimethylamine
Introduction Twenty genetically encoded, often called canonical, amino acids comprise the vast majority of residues in virtually all proteins. Over 300 non-canonical residues have also been identified [1] that are not as universally distributed, are usually present in low abundance, and generally are post-translational modifications of one of the canonical amino acids. Selenocysteine and pyrrolysine are exceptional, as these two non-canonical residues are introduced into protein during translation under the direction of what are typically stop codons. Current Opinion in Chemical Biology 2004, 8:484–491
Selenocysteine is synthesized on tRNASec that decodes UGA in the proper context as selenocysteine [2,3]. Pyrrolysine is found in a methanogen methyltransferase as the residue corresponding to an in-frame UAG in the encoding gene [4,5], and a dedicated tRNAPyl decodes UAG [6]. The genetic encoding of selenocysteine and pyrrolysine has led to their respective consideration as the 21st and 22nd amino acids [3,7]. This short review surveys the available data supporting the genetic encoding of pyrrolysine, then focuses on the proposed function of pyrrolysine in the methanogenic methylamine methyltransferases.
In-frame UAG codons in methylamine methyltransferase genes The discovery of pyrrolysine was rooted in the study of the biochemistry of methane formation by Methanosarcina species from monomethylamine (MMA), dimethylamine (DMA) and trimethylamine (TMA). Methanogenesis from these substrates is initiated by three methyltransferases that specifically methylate their cognate corrinoid proteins with one of these methylamines [8–10]. Figure 1 lists the methyltransferase nomenclature. All three methylated corrinoid proteins are substrates for MtbA, a methylcobamide:coenzyme M methyltransferase [11,12] effecting demethylation of corrinoid using zinc to form the nucleophilic coenzyme M thiolate [13,14]. The reduction of methyl-coenzyme M to methane is the major energy-conserving step of methanogenesis [15]. The M. barkeri mtmB1 gene encoding MtmB, the characterized MMA methyltransferase, provided the first harbinger of pyrrolysine. MtmB is a homotrimer with subunits of 50 kDa, but the reading frame of mtmB1 apparently terminates at a mid-frame UAG codon [5]. The subsequent sequencing of the mttB and mtbB genes [16], respectively encoding the circa 50 kDa TMA and DMA methyltransferases [8,9], also revealed in-frame amber codons that do not terminate translation. Data from Methanosarcina [17,18] and Methanococcoides (GenBank NZ_AADH00000000) genomes has thus far shown that all methanogen methylamine methyltransferase genes share the common trait of an in-frame amber codon. The C- and N-terminal sequences of DMA and MMA methyltransferases confirmed that UAG is not the stop codon in the encoding genes [16,19]. Analysis of transcripts [16] revealed no UAG modification, and since little or no UAG-terminated mtmB gene product is detectable in cells [19], normal function of UAG as a stop codon www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 485
Figure 1
appeared to be circumvented during translation to produce full-length MtmB. CH3 Co Bzm
HR
O
N
H3C
S
N
*MttB (TMA) *MtbB (DMA) *MtmB (MMA)
MttC (TMA) MtbC (DMA) MtmC (MMA)
Pyrrolysine encoded by the mtmB1 amber codon
O S
Edman degradation and tandem mass spectrometry of tryptic fragments of MtmB confirmed that the in-frame UAG of mtmB1 is translated at the ribosome but both methods revealed lysine at the UAG position [19]. The excellent work of our collaborators, Michael Chan and his laboratory, resolved this conundrum. The crystal structure of MtmB revealed lysine, but with eN in amide linkage with (4R,5R)-4 substituted-pyrroline-5-carboxylate [4], demonstrating an early proposal that UAG could encode a specialized residue [5]. Further analysis of derivatized pyrrolysine in MtmB crystals has led to the recent assignment of the C-4 substituent as a methyl group [20]. Very recently, our laboratory in collaboration with those of Kari B Green-Church and Jon Amster obtained the first detection of pyrrolysine by a technique other than crystallography. Tandem mass spectrometry revealed the UAG-encoded residue has the exact mass predicted for the structure of pyrrolysine shown in Figure 1 (R Pitsch et al., unpublished data).
O
MtbA O
O S
CH3R
HS
Co Bzm
O
H N N Current Opinion in Chemical Biology
Enzymatic components involved in Coenzyme M methylation with methylamines. (a) Schematic of the corrinoid dependent demethylation of methylamines. Specificity for a particular methylamine by a methyltransferase or corrinoid protein is indicated in parenthesis. Asterisks mark those proteins whose genes contain in-frame amber codons. The corrin ring is represented by the square. Histidine coordination to cobalt is suggested by homology modeling of MtmC (see Figure 3). R = -N(CH3)2, -N(CH3)H, or –NH2. The nature of the base is unknown, but 5-OH benzimidazole (Bzm) predominates in this methanogen [35].
Figure 2
(a)
2
(R)
N O–
O–
O
1
5 (R)
O
(R)
O (R)
NH2
(R)
(R)
O
NH2
O–
O
(R)
+
N 3
4
(R)
NH
NH2
O NH2 OH
H
(S)
NH2 (S)
HO HO
(b)
PylS PylBCD Lys + tRNAPyl Lys-tRNAPyl LysRS1/LysRS2
Pyl-tRNAPyl
O Pyl
O
(c)
PylBCD
PylS Pyl
Pyl-tRNAPyl
Current Opinion in Chemical Biology
Schemes for formation of pyrrolysyl-tRNAPyl. (a) A speculative route of pyrrolysine (pyl) formation suggested by the similarities of PylB, PylC, and PylD to enzymes of known functions [6]. This general scheme would be feasible for (b) modification of lys-tRNAPyl to pyl-tRNAPyl, as well as (c) formation of free pyrrolysine for direct charging of tRNAPyl. www.sciencedirect.com
Current Opinion in Chemical Biology 2004, 8:484–491
486 Mechanisms
The UAG-decoding tRNAPyl
However, a dedicated tRNA for pyrrolysine, as with all other genetically encoded amino acids, is indicative of pre-translational synthesis of pyrrolysyl-tRNAPyl. Indirect and direct methods to form pyrrolysyl-tRNAPyl are possible (Figure 2). Slow synthesis of lysyl-tRNAPyl by PylS [6] has been reported, as well as by a complex of both class I and class II lysyl-tRNA synthetases, termed LysRS1 and LysRS2, respectively [23]. However, we have shown that mutants lacking most of the LysRS1 gene still aminoacylate tRNAPyl. (A Mahapatra et al., unpublished data). On the other hand, properly folded PylS does not use lysine as a substrate [23]. This problem may be resolved by our recent work that indicates pyrrolysyl-tRNApyl is formed directly by PylS. L-pyrrolysine has been synthesized in the laboratory of Michael Chan [20], and PylS can charge tRNAPyl with the synthetic amino acid (Blight et al., unpublished data). This exciting result indicates that pyrrolysine may be synthesized as a cytoplasmic amino acid in the same manner as the canonical 20 amino acids, in contrast to selenocysteine, which is synthesized on tRNA [3].
Further support for the genetic encoding of pyrrolysine comes from the pylTSBCD genes found near the mtmB1 gene in Methanosarcina spp. The pylT gene encodes tRNAPyl (or tRNACUA), which has the CUA anticodon to decode UAG as a sense codon, while PylS has the catalytic residues of the active site common to class II aminoacyl-tRNA synthetases. PylB, PylC and PylD are related to proteins whose traits are suggestive of potential roles in the biosynthesis of pyrrolysine [6]. Deletion of the tRNAPyl gene from Methanosarcina acetivorans completely obliterates the ability to make active methylamine methyltransferases, but has no effect on other methyltransferases, consistent with a unique role for pyrrolysine in methylamine biochemistry (A Mahapatra and J. Krzycki, unpublished data). The tRNA has an unusually small D loop, and further has the potential to form a six (rather than the typical five) base-pair anticodon stem [6]. Structure probing recently confirmed the elongated anticodon stem in tRNAPyl transcripts; however, this unusual structure does not prevent aminoacyltRNAPyl binding by translation factor EF-Tu [21]. A stem-loop with conserved elements just 30 of the UAG codon in mttB and mtmB genes may be involved in circumventing UAG function as a stop codon during decoding as pyrrolysine [22].
MtmC, substrate of the pyrrolysyl-protein MtmB The metabolically relevant product of MtmB activity is methyl-Co(III)-MtmC, which displays the electronic spectra of a single bound hexacoordinate methyl-Co(III) 5-hydroxybenzimidazolylcobamide cofactor [24]. MtmC cycles between methyl-Co(III)- and Co(I)-MtmC during catalysis. MMA:CoM methyl transfer can be initiated
It is formally possible that pyrrolysine is made following lysine incorporation into the nascent polypeptide. Figure 3
(a)
(b)
H106
E259 Q333 Y335
E229 O202
Current Opinion in Chemical Biology
Tertiary structures of MtmB and its substrate, MtmC. (a) Theoretical MtmC (GenBank AF013713) structure made by the Swiss-Model server [36] using the cobalamin binding domain of MetH (PDB ID code, 1bmtA) as template. The root mean square deviation between the Ca atoms of the MtmC model and the template is 0.4 A˚. The corrinoid is placed in the MtmC model by superimposing the MetH and MtmC structures. Coils that deviate significantly from the template are in red. (b) MtmB showing location of pyrrolysine (O202) and surrounding residues proposed to participate in methyl group transfer from MMA to MtmC. Current Opinion in Chemical Biology 2004, 8:484–491
www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 487
with methyl-Co(III)-MtmC, but hydroxy-Co(III)-MtmC requires reduction, presumably to Co(I)-MtmC since lowpotential reducing agents are required [10]. MethylCo(III)-MtmC and MtbA can catalyze continuous methyliodide:coenzyme M methyl transfer, consistent with generation of Co(I)-MtmC by methyl cation transfer to coenzyme M, followed by nucleophilic attack by Co(I) on methyliodide [10]. MtbB, the DMA methyltransferase, as well as MtaB, the methanol methyltransferase, will methylate free cob(I)alamin with their respective substrates [8,25]. Taken together, these results indicate that methanogen methylotrophic corrinoid proteins such as MtmC are methylated in the Co(I) state, forming a hexacoordinate methyl-Co(III) species whose demethylation regenerates Co (I) corrinoid, in keeping with the behavior of other corrinoid-dependent methyltransferases [26]. MtmC, as well as other small methanogenic corrinoid proteins, are close homologs of the cobalamin-binding module of MetH [5,16,25,27]. Homology modeling using the known structure of the MetH cobalamin-binding module [28] as template produces a MtmC model with little deviation from the Ca backbone in the regions of
MetH contacting cobalamin (Figure 3). This yields insight into the function of MtmC, as the well-studied MetH binds cobalamin with the benzimidazole base bound to the protein, and a histidinyl residue serving as a ligand to cobalt. The MtmC model displays conservation of histidine (H106) as cobalt ligand, as well as conserved placement of the seryl (S159) and aspartyl (D104) residues that in MetH are proposed to participate in protonating and deprotonating the histidinyl ligand as hexacoordinate methyl-Co(III) or tetracoordinate Co(I) are formed during catalysis. This is suggested as a means by which conformation changes occur for cobalamin to interact with the three methyltransferase modules of MetH [29,30]. MtmC must interact with the different active sites of MtbA and MtmB during catalysis, and the ligands to cobalt may play a similar role.
Pyrrolysine in MtmB mediated methylation of MtmC The crystal structure of MtmB revealed a dimer of trimers with the subunit fold that of a modified ab TIM barrel. The C-terminus of the central cluster of b sheets forms a negatively charged surface cleft that faces out from the core of hexamers towards the solvent. Pyrrolysine is found
Figure 4
Co
B H
Co CH3
B H
Y335
N H
CH3
H
E229 H
N E229 H
Q333
N
Q333
H
N
H
Y335 H
H
E259 O
E259
O
Nε
Nε + CH3NH3+ – NH3
H
Co
B N H E229
Y335
H H
H E259 O
E229 H
H H
Q333
N
– CH3-MtmC Nε
Y335 N
Q333
N
CH3
B H
H O
E259 Nε Current Opinion in Chemical Biology
Postulated role of pyrrolysine during MtmB-catalysed methylation of the corrinoid cofactor of MtmC. The scheme shown is a variation on the mechanism originally proposed by Hao et al. [4]. www.sciencedirect.com
Current Opinion in Chemical Biology 2004, 8:484–491
488 Mechanisms
near the center of this cleft, with sufficient surrounding space to allow close approach of a corrin ring to the UAGencoded residue [4].
an in-line nucleophilic attack by tetracoordinate Co(I)MtmC. Although it was initially proposed that Y335 and Q333 participate in binding methylamine, they could also serve to H-bond with the ammonia released from the pyrroline ring, enhancing the equilibrium between 2aminopyrrolysine and pyrrolysine in the direction of product removal from the enzyme. This model of pyrrolysine function has the strong appeal of applicability to pyrrolysine function in the MMA, DMA and TMA methyltransferases. This cannot be said of alternative models involving pyrrolysine, such as imine bond hydrolysis and formation of an external imine bond between pyrrolysine and methylamine, which is feasible with MMA, but not TMA.
Evidence for reactivity of the C-2 atom of the pyrrolysine ring came from two crystal forms of MtmB obtained from solutions containing NaCl or (NH4)2SO4. In the NaCl crystal form, the pyrroline was unmodified at the C-2 position. The second crystal form was a mixture of two pyrrolysine states. One resembles pyrrolysine in the NaCl crystal form, while the second was modeled with the pyrrolysine ring rotated 908 and with an amine adduct at the C-2 position. Residues within H-bonding distance of pyrrolysine were repositioned in the presence of 2aminopyrrolysine. In addition, two nearby residues were observed at partial occupancy to H-bond with ammonia.
While lacking sequence similarity, MtmB is a structural homolog of the cobalamin dependent methyl-tetrahydrofolate (meTHF) and homocysteine methyltransferase modules of MetH [31], as well as AcsE-MeTr, a meTHF:corrinoid methyltransferase involved in acetylCoA synthesis [32]. The residues for binding substrates methylating or demethylating corrinoid in these methyltransferases are at the C-terminal portion of the inner aspect of the ab TIM barrel of each protein, as are pyrrolysine and surrounding residues in MtmB.
The reactivity of the C-2 carbon and the changes in the residues surrounding pyrrolysine led Hao et al. to propose a model for MtmB methylation of Co(I) corrinoid cofactor with MMA [4]. A variation on this proposal is shown in Figure 4. An unidentified base deprotonates the MMA ion attracted to the negatively charged cleft, otherwise the enzyme must utilize the small fraction of unionized MMA present at neutral pH. Y335 and Q333 are proposed to assist in the initial MtmB binding of MMA and positioning such that electrophilic attack of the C-2 atom on the lone electron pair of the nitrogen of unionized MMA occurs. The observed reorientation of the pyrroline ring would result as nearby residues move to H-bond with the new C-2 substituent, orienting the methyl group for
The location of pyrrolysine is particularly striking when compared with the MetH meTHF methyltransferase domain. The methylamine group of the proposed 2methylamino-pyrrolysine intermediate and the methyl group of meTHF are within 3 A˚ of each other when
Figure 5
(a)
(b)
Current Opinion in Chemical Biology
Structural homology of MetH MeTHF methyltransferase module and MtmB. (a) Structure of MeTHF methyltransferase module (PDB ID code 1Q8J, residues 300 to 559) with bound MeTHF, surrounding residues are those implicated in binding the substrate. (b) MeTHF methyltransferase domain superimposed with MtmB from (NH4)2SO4 crystal form (PDB ID code 1L2Q). The two proteins were aligned using the DALI program [37], with 230 Ca atoms aligned with a root mean squares difference of 3.1 A˚ . Only conserved core elements are shown in dark green, (MtmB) or light blue-green (MetH MeTHF domain). Arrows point to the C-2 amine of 2-aminopyrrolysine (red), or the methyl group of MeTHF (black). Current Opinion in Chemical Biology 2004, 8:484–491
www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 489
the TIM barrel of MtmB is optimally superimposed onto that of the meTHF domain (Figure 5) using a structural alignment program. This apparent conservation of geometry of methyl groups transferred to homologous corrinoid proteins further supports the notion that pyrrolysine acts to optimally position a methylammonium species for methyl transfer to corrinoid.
Acknowledgements Work in the author’s laboratory was supported from funds from NSF (MCB-9808914), DOE (DEFG02-92ER20042) and NIH (GM061796). The author thanks our collaborators over the years with special acknowledgement to our recent collaborations with William Metcalf and Kari Green-Church, and to David Grahame for useful discussions. I especially wish to thank Bing Hao and Michael Chan, with whom my laboratory was privileged to share in the discovery of pyrrolysine.
References and recommended reading Conclusions Recent discoveries indicate the genetic encoding of the 22nd amino acid may be much different than the 21st. However, like selenocysteine, pyrrolysine appears to have been recruited/retained in the code because of its unique properties. Pyrrolysine is the first example of a genetically encoded electrophilic residue, a property previously found only in protein prosthetic groups or posttranslationally modified residues [33]. Although the proposed function of pyrrolysine fits current observations, very much remains to be demonstrated. For example, the hypothetical catalytic intermediate of 2 methylaminopyrrolysine remains unobserved, and even the necessity of pyrrolysine for catalysis by MtmB has not yet been directly tested. However, the on-going development of genetic systems for Methanosarcina spp. [34] is now making site-directed mutagenesis feasible in a host decoding UAG as pyrrolysine.
Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest 1.
Garavelli JS: The RESID Database of Protein Modifications as a resource and annotation tool. Proteomics 2004, 4:1527-1533.
2.
Stadtman TC: Selenocysteine. Annu Rev Biochem 1996, 65:83-100.
3.
Commans S, Bo¨ ck A: Selenocysteine inserting tRNAs: an overview. FEMS Microbiol Rev 1999, 23:335-351.
4.
Hao B, Gong W, Ferguson TK, James CM, Krzycki JA, Chan MK: A new UAG-encoded residue in the structure of a methanogen methyltransferase. Science 2002, 296:1462-1466. This paper is the first description of the UAG-encoded residue of MtmB as pyrrolysine. The structure of MtmB was solved for the two different crystal forms originating from supernatants containing NaCl or (NH4)2SO4. By estimating the occupancies of two different conformations of the UAGencoded residue, models for both the structure and function of pyrrolysine were proposed. 5.
Given pyrrolysine’s potential to form amine or imine derivatives, it might be present in other enzymes with amine substrates. If so, it should prove of wider distribution than the small number of microbes now known to possess the pyl genes. Nonetheless, even with the present restricted distribution to species of Methanosarcina, Desulfitobacterium and Methanococcoides, pyrrolysine now serves as an important illustration that the natural genetic code is sufficiently flexible to allow relatively idiosyncratic incorporation of a highly specialized residue. Biological methane formation from an entire class of substrates appears to be based upon this premise.
Update Peptide sequences made from the DMA and MMA methyltransferase genes before and after their in-frame amber codons have previously been obtained, are consistent with UAG translation as pyrrolysine in both proteins [16,19]. In contrast, evidence of translation of the UAG codon in the TMA methyltransferase gene has been based largely on the perceived requirement for UAG readthrough to produce the full-length TMA methyltransferase [9,16]. Recently, in a mass spectrometric survey of proteins in Methanococcoides burtonii, peptides from a TMA methyltransferase were identified [38]. Although a pyrrolysyl-containing peptide was not observed, peptides were identified from both the N- and C-terminal portions of the protein, consistent with translation of UAG in this methyltransferase as pyrrolysine. www.sciencedirect.com
Burke SA, Lo SL, Krzycki JA: Clustered genes encoding the methyltransferases of methanogenesis from monomethylamine. J Bacteriol 1998, 180:3432-3440.
6.
Srinivasan G, James CM, Krzycki JA: Pyrrolysine encoded by UAG in Archaea: charging of a UAG-decoding specialized tRNA. Science 2002, 296:1459-1462. The first description of tRNACUA (tRNAPyl). The presence of tRNACUA is important supporting evidence that pyrrolysine is encoded by the mtmB1 amber codon, rather than being a post-translational modification. The tRNA gene was identified using a computer program (tRNAScan) developed by the group of Sean Eddy. Contigs of the partially sequenced M. barkeri Fusaro genome were scanned, with preference given to those that encode methylamine catabolism proteins, as microbial genes often cluster with others for a particular biochemical pathway. In addition, this paper described the other pyl genes, including pylS, which has recently been shown to be a pyrrolysyl-tRNA synthetase.
7. Atkins JF, Gesteland R: The 22nd amino acid. Science 2002, 296:1409-1411. This perspective published with the pyrrolysine and pyl gene description papers relates pyrrolysine to the larger picture of the genetic code and recoding events such as selenocysteine. 8.
Ferguson DJ Jr, Gorlatova N, Grahame DA, Krzycki JA: Reconstitution of dimethylamine:coenzyme M methyl transfer with a discrete corrinoid protein and two methyltransferases purified from Methanosarcina barkeri. J Biol Chem 2000, 275:29053-29060.
9.
Ferguson DJ Jr, Krzycki JA: Reconstitution of trimethylaminedependent coenzyme M methylation with the trimethylamine corrinoid protein and the isozymes of methyltransferase II from Methanosarcina barkeri. J Bacteriol 1997, 179:846-852.
10. Burke SA, Krzycki JA: Reconstitution of monomethylamine:coenzyme M methyl transfer with a corrinoid protein and two methyltransferases purified from Methanosarcina barkeri. J Biol Chem 1997, 272:16570-16577. 11. Ferguson DJ Jr, Krzycki JA, Grahame DA: Specific roles of methylcobamide:coenzyme M methyltransferase isozymes in metabolism of methanol and methylamines in Methanosarcina barkeri. J Biol Chem 1996, 271:5189-5194. Current Opinion in Chemical Biology 2004, 8:484–491
490 Mechanisms
12. Burke SA, Krzycki JA: Involvement of the ‘‘A’’ isozyme of methyltransferase II and the 29-kilodalton corrinoid protein in methanogenesis from monomethylamine. J Bacteriol 1995, 177:4410-4416. 13. Gencic S, LeClerc GM, Gorlatova N, Peariso K, Penner-Hahn JE, Grahame DA: Zinc-thiolate intermediate in catalysis of methyl group transfer in Methanosarcina barkeri. Biochemistry 2001, 40:13068-13078. 14. Kruer M, Haumann M, Meyer-Klaucke W, Thauer RK, Dau H: The role of zinc in the methylation of the coenzyme M thiol group in methanol:coenzyme M methyltransferase from Methanosarcina barkeri. Eur J Biochem 2002, 269:2117-2123. This article and [13] address zinc ligation to enzyme and coenzyme M in MtbA and MtaA, two methylcobamide:CoM methyltransferases involved respectively with the methylamine and methanol methyltransferases. 15. Thauer RK: Biochemistry of methanogenesis: a tribute to Marjory Stephenson. 1998 Marjory Stephenson Prize Lecture. Microbiology 1998, 144:2377-2406. 16. Paul L, Ferguson DJ, Krzycki JA: The trimethylamine methyltransferase gene and multiple dimethylamine methyltransferase genes of Methanosarcina barkeri contain in-frame and read-through amber codons. J Bacteriol 2000, 182:2520-2529. 17. Galagan JE, Nusbaum C, Roy A, Endrizzi MG, Macdonald P, FitzHugh W, Calvo S, Engels R, Smirnov S, Atnoor D et al.: The genome of M. acetivorans reveals extensive metabolic and physiological diversity. Genome Res 2002, 12:532-542. See annotation to [18]. 18. Deppenmeier U, Johann A, Hartsch T, Merkl R, Schmitz RA, Martinez-Arias R, Henne A, Wiezer A, Ba¨ umer S, Jacobi C et al.: The Genome of Methanosarcina mazei: Evidence for Lateral Gene Transfer Between Bacteria and Archaea. J Mol Microbiol Biotechnol 2002, 4:453-461. This article and [17] demonstrate conclusively that all methylamine methyltransferase genes (present in multiple copies) in these genomes possess in-frame amber codons. Prior to the availability of these complete genomes, it remained possible that other methylamine methyltransferase genes lacking amber codons were responsible for production of full-length methylamine methyltransferases, rather than UAG translation as pyrrolysine. 19. James CM, Ferguson TK, Leykam JF, Krzycki JA: The amber codon in the gene encoding the monomethylamine methyltransferase isolated from Methanosarcina barkeri is translated as a sense codon. J Biol Chem 2001, 276:34252-34258. 20. Hao B, Gang Z, Kang PT, Soares JA, Ferguson TK, Gallucci J, Krzycki JA, Chan MK: Reactivity and chemical synthesis of L-pyrrolysine - the 22nd genetically-encoded amino acid. Chem Biol 2004, in press. 21. Theobald-Dietrich A, Frugier M, Giege R, Rudinger-Thirion J: Atypical archaeal tRNA pyrrolysine transcript behaves towards EF-TU as a typical elongator tRNA. Nucleic Acids Res 2004, 32:1091-1096. Structure probing reported in this paper confirmed the proposed atypical secondary structure of tRNAPyl with a six base-pair anticodon stem. This recalls the elongated acceptor stem of tRNASec, and suggests unusual tRNA structure may prove a common theme of tRNAs for genetically encoded non-canonical amino acids. Further, by changing the anticodon of tRNAPyl to achieve slow lysylation with yeast lysyl-tRNA synthetase, the authors demonstrate that lysyl-tRNAPyl is recognized by E. coli translation factor EF-Tu. This is unexpected, as selenocysteinyl-tRNASec is not recognized by EF-Tu, rather a dedicated selenocysteine translation factor exists. However, this result does not discount the possibility that an archaeal translation factor specific for pyrrolysyl-tRNAPyl exists in the Methanosarcina species. 22. Namy O, Rousset JP, Napthine S, Brierley I: Reprogrammed genetic decoding in cellular gene expression. Mol Cell 2004, 13:157-168. Notes the presence of stem-loops with conserved elements downstream of the UAG codon in mtmB and mttB. 23. Polycarpo C, Ambrogelly A, Ruan B, Tumbula-Hansen D, Ataide SF, Ishitani R, Yokoyama S, Nureki O, Ibba M, Soll D: Activation of the pyrrolysine suppressor tRNA requires Current Opinion in Chemical Biology 2004, 8:484–491
formation of a ternary complex with class I and class II lysyl-tRNA synthetases. Mol Cell 2003, 12:287-294. Multiple clones producing recombinant PylS were found not to possess lysyl-tRNA synthetase activity, discounting earlier indications that the enzyme was capable of this reaction (see [6]). Furthermore, the authors demonstrate that LysRS1 and LysRS2 can together slowly aminoacylate tRNAPyl transcripts with lysine. Recent work by our group demonstrates PylS is a pyrrolysyl-tRNA synthetase, indicating that two competing reactions involving tRNAPyl may exist in vivo, only one of which would directly act to produce pyrrolysyl-tRNAPyl. 24. Kremer JD, Cao X, Krzycki J: Isolation of two novel corrinoid proteins from acetate-grown Methanosarcina barkeri. J Bacteriol 1993, 175:4824-4833. 25. Sauer K, Harms U, Thauer RK: Methanol:coenzyme M methyltransferase from Methanosarcina barkeri. Purification, properties and encoding genes of the corrinoid protein MT1. Eur J Biochem 1997, 243:670-677. 26. Banerjee R, Ragsdale SW: The many faces of vitamin B12: catalysis by cobalamin-dependent enzymes. Annu Rev Biochem 2003, 72:209-247. 27. Paul L, Krzycki JA: Sequence and transcript analysis of a novel Methanosarcina barkeri methyltransferase II homolog and its associated corrinoid protein homologous to methionine synthase. J Bacteriol 1996, 178:6599-6607. 28. Drennan CL, Huang S, Drummond JT, Matthews RG, Ludwig ML: How a protein binds B12: A 3.0 A˚ X-ray structure of B12-binding domains of methionine synthase. Science 1994, 266:1669-1674. 29. Bandarian V, Pattridge KA, Lennon BW, Huddler DP, Matthews RG, Ludwig ML: Domain alternation switches B(12)-dependent methionine synthase to the activation conformation. Nat Struct Biol 2002, 9:53-56. The structure of the MetH cobalamin binding domain binding a methyltransferase domain (the activation domain) reveals the displacement of the corrinoid from the histidine ligand, suggesting a mechanism for control of conformation as the cobalamin module moves from methyltransferase to methyltransferase module in this multi-domain protein. 30. Dorweiler JS, Finke RG, Matthews RG: Cobalamin-dependent methionine synthase: probing the role of the axial base in catalysis of methyl transfer between methyltetrahydrofolate and exogenous cob(I)alamin or cob(I)inamide. Biochemistry 2003, 42:14653-14662. Demonstration that the ligand triad (D, S and H) does not greatly contribute to catalysis, but may play a role in conformational changes controlling cobalamin interaction with different methyltransferase domains. 31. Evans JC, Huddler DP, Hilgers MT, Romanchuk G, Matthews RG, Ludwig ML: Structures of the N-terminal modules imply large domain motions during catalysis by methionine synthase. Proc Natl Acad Sci USA 2004, 101:3729-3736. The structures of the homocysteine and meTHF methyltransferase were determined, showing for the first time how a corrinoid-dependent methyltransferase might bind and orient substrates acting as methyl donors and acceptors to the central cobalt ion. Like the MtmB and AcsE-Mtr proteins, the MetH methyltransferases adopt the ubiquitous TIM barrel fold, suggesting that may be a near-common feature of the majority of corrinoid dependent methyltransferases. 32. Doukov T, Seravalli J, Stezowski JJ, Ragsdale SW: Crystal structure of a methyltetrahydrofolate- and corrinoiddependent methyltransferase. Structure Fold Des 2000, 8:817-830. 33. Retey J: Discovery and role of methylidene imidazolone, a highly electrophilic prosthetic group. Biochim Biophys Acta 2003, 1647:179-184. One of the few other electrophilic groups found in proteins is reviewed in detail. 34. Meuer J, Kuettner HC, Zhang JK, Hedderich R, Metcalf WW: Genetic analysis of the archaeon Methanosarcina barkeri Fusaro reveals a central role for Ech hydrogenase and ferredoxin in methanogenesis and carbon fixation. Proc Natl Acad Sci USA 2002, 99:5632-5637. One of the first uses of genetics to investigate the biochemistry of methanogenesis, it is representative of the ongoing pioneering development of this essential group of techniques from the Metcalf laboratory. www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 491
35. Pol A, van der Drift C, Vogels GD: Corrinoids from Methanosarcina barkeri: structure of the alpha-ligand. Biochem Biophys Res Commun 1982, 108:731-737. 36. Schwede T, Kopp J, Guex N, Peitsch MC: SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res 2003, 31:3381-3385. 37. Holm L, Sander C: Dali: a network tool for protein structure comparison. Trends Biochem Sci 1995, 20:478-480. 38. Amber Goodchild A, Saunders NF, Ertan H, Raftery M, Guilhaus M, Curmi PMG, Cavicchioli R: A proteomic determination of cold
www.sciencedirect.com
adaptation in the Antarctic archaeon Methanococcoides burtonii. Mol Microbiol 2004, 53:309-321.
Now in press The work referred to as (Blight et al., unpublished data) is now in press: 39. Blight SK, Larue RC, Mahapatra A, Longstaff DG, Chang E, Zhao G, Kang PT, Green-Church KB, Chan MK, Krzycki JA: Direct charging of tRNACUA with pyrrolysine in vitro and in vivo. Nature 2004, in press.
Current Opinion in Chemical Biology 2004, 8:484–491
Current Opinion in Chemical Biology 2004, 8:484–491 This review comes from a themed section on Mechanisms Edited by Hung-Wen Liu and Ja`nos Re´tey Available online 9th September 2004 1367-5931/$ – see front matter # 2004 Elsevier Ltd. All rights reserved. DOI 10.1016/j.cbpa.2004.08.012
Abbreviations DMA dimethylamine meTHF methyl-tetrahydrofolate MMA monomethylamine TMA trimethylamine
Introduction Twenty genetically encoded, often called canonical, amino acids comprise the vast majority of residues in virtually all proteins. Over 300 non-canonical residues have also been identified [1] that are not as universally distributed, are usually present in low abundance, and generally are post-translational modifications of one of the canonical amino acids. Selenocysteine and pyrrolysine are exceptional, as these two non-canonical residues are introduced into protein during translation under the direction of what are typically stop codons. Current Opinion in Chemical Biology 2004, 8:484–491
Selenocysteine is synthesized on tRNASec that decodes UGA in the proper context as selenocysteine [2,3]. Pyrrolysine is found in a methanogen methyltransferase as the residue corresponding to an in-frame UAG in the encoding gene [4,5], and a dedicated tRNAPyl decodes UAG [6]. The genetic encoding of selenocysteine and pyrrolysine has led to their respective consideration as the 21st and 22nd amino acids [3,7]. This short review surveys the available data supporting the genetic encoding of pyrrolysine, then focuses on the proposed function of pyrrolysine in the methanogenic methylamine methyltransferases.
In-frame UAG codons in methylamine methyltransferase genes The discovery of pyrrolysine was rooted in the study of the biochemistry of methane formation by Methanosarcina species from monomethylamine (MMA), dimethylamine (DMA) and trimethylamine (TMA). Methanogenesis from these substrates is initiated by three methyltransferases that specifically methylate their cognate corrinoid proteins with one of these methylamines [8–10]. Figure 1 lists the methyltransferase nomenclature. All three methylated corrinoid proteins are substrates for MtbA, a methylcobamide:coenzyme M methyltransferase [11,12] effecting demethylation of corrinoid using zinc to form the nucleophilic coenzyme M thiolate [13,14]. The reduction of methyl-coenzyme M to methane is the major energy-conserving step of methanogenesis [15]. The M. barkeri mtmB1 gene encoding MtmB, the characterized MMA methyltransferase, provided the first harbinger of pyrrolysine. MtmB is a homotrimer with subunits of 50 kDa, but the reading frame of mtmB1 apparently terminates at a mid-frame UAG codon [5]. The subsequent sequencing of the mttB and mtbB genes [16], respectively encoding the circa 50 kDa TMA and DMA methyltransferases [8,9], also revealed in-frame amber codons that do not terminate translation. Data from Methanosarcina [17,18] and Methanococcoides (GenBank NZ_AADH00000000) genomes has thus far shown that all methanogen methylamine methyltransferase genes share the common trait of an in-frame amber codon. The C- and N-terminal sequences of DMA and MMA methyltransferases confirmed that UAG is not the stop codon in the encoding genes [16,19]. Analysis of transcripts [16] revealed no UAG modification, and since little or no UAG-terminated mtmB gene product is detectable in cells [19], normal function of UAG as a stop codon www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 485
Figure 1
appeared to be circumvented during translation to produce full-length MtmB. CH3 Co Bzm
HR
O
N
H3C
S
N
*MttB (TMA) *MtbB (DMA) *MtmB (MMA)
MttC (TMA) MtbC (DMA) MtmC (MMA)
Pyrrolysine encoded by the mtmB1 amber codon
O S
Edman degradation and tandem mass spectrometry of tryptic fragments of MtmB confirmed that the in-frame UAG of mtmB1 is translated at the ribosome but both methods revealed lysine at the UAG position [19]. The excellent work of our collaborators, Michael Chan and his laboratory, resolved this conundrum. The crystal structure of MtmB revealed lysine, but with eN in amide linkage with (4R,5R)-4 substituted-pyrroline-5-carboxylate [4], demonstrating an early proposal that UAG could encode a specialized residue [5]. Further analysis of derivatized pyrrolysine in MtmB crystals has led to the recent assignment of the C-4 substituent as a methyl group [20]. Very recently, our laboratory in collaboration with those of Kari B Green-Church and Jon Amster obtained the first detection of pyrrolysine by a technique other than crystallography. Tandem mass spectrometry revealed the UAG-encoded residue has the exact mass predicted for the structure of pyrrolysine shown in Figure 1 (R Pitsch et al., unpublished data).
O
MtbA O
O S
CH3R
HS
Co Bzm
O
H N N Current Opinion in Chemical Biology
Enzymatic components involved in Coenzyme M methylation with methylamines. (a) Schematic of the corrinoid dependent demethylation of methylamines. Specificity for a particular methylamine by a methyltransferase or corrinoid protein is indicated in parenthesis. Asterisks mark those proteins whose genes contain in-frame amber codons. The corrin ring is represented by the square. Histidine coordination to cobalt is suggested by homology modeling of MtmC (see Figure 3). R = -N(CH3)2, -N(CH3)H, or –NH2. The nature of the base is unknown, but 5-OH benzimidazole (Bzm) predominates in this methanogen [35].
Figure 2
(a)
2
(R)
N O–
O–
O
1
5 (R)
O
(R)
O (R)
NH2
(R)
(R)
O
NH2
O–
O
(R)
+
N 3
4
(R)
NH
NH2
O NH2 OH
H
(S)
NH2 (S)
HO HO
(b)
PylS PylBCD Lys + tRNAPyl Lys-tRNAPyl LysRS1/LysRS2
Pyl-tRNAPyl
O Pyl
O
(c)
PylBCD
PylS Pyl
Pyl-tRNAPyl
Current Opinion in Chemical Biology
Schemes for formation of pyrrolysyl-tRNAPyl. (a) A speculative route of pyrrolysine (pyl) formation suggested by the similarities of PylB, PylC, and PylD to enzymes of known functions [6]. This general scheme would be feasible for (b) modification of lys-tRNAPyl to pyl-tRNAPyl, as well as (c) formation of free pyrrolysine for direct charging of tRNAPyl. www.sciencedirect.com
Current Opinion in Chemical Biology 2004, 8:484–491
486 Mechanisms
The UAG-decoding tRNAPyl
However, a dedicated tRNA for pyrrolysine, as with all other genetically encoded amino acids, is indicative of pre-translational synthesis of pyrrolysyl-tRNAPyl. Indirect and direct methods to form pyrrolysyl-tRNAPyl are possible (Figure 2). Slow synthesis of lysyl-tRNAPyl by PylS [6] has been reported, as well as by a complex of both class I and class II lysyl-tRNA synthetases, termed LysRS1 and LysRS2, respectively [23]. However, we have shown that mutants lacking most of the LysRS1 gene still aminoacylate tRNAPyl. (A Mahapatra et al., unpublished data). On the other hand, properly folded PylS does not use lysine as a substrate [23]. This problem may be resolved by our recent work that indicates pyrrolysyl-tRNApyl is formed directly by PylS. L-pyrrolysine has been synthesized in the laboratory of Michael Chan [20], and PylS can charge tRNAPyl with the synthetic amino acid (Blight et al., unpublished data). This exciting result indicates that pyrrolysine may be synthesized as a cytoplasmic amino acid in the same manner as the canonical 20 amino acids, in contrast to selenocysteine, which is synthesized on tRNA [3].
Further support for the genetic encoding of pyrrolysine comes from the pylTSBCD genes found near the mtmB1 gene in Methanosarcina spp. The pylT gene encodes tRNAPyl (or tRNACUA), which has the CUA anticodon to decode UAG as a sense codon, while PylS has the catalytic residues of the active site common to class II aminoacyl-tRNA synthetases. PylB, PylC and PylD are related to proteins whose traits are suggestive of potential roles in the biosynthesis of pyrrolysine [6]. Deletion of the tRNAPyl gene from Methanosarcina acetivorans completely obliterates the ability to make active methylamine methyltransferases, but has no effect on other methyltransferases, consistent with a unique role for pyrrolysine in methylamine biochemistry (A Mahapatra and J. Krzycki, unpublished data). The tRNA has an unusually small D loop, and further has the potential to form a six (rather than the typical five) base-pair anticodon stem [6]. Structure probing recently confirmed the elongated anticodon stem in tRNAPyl transcripts; however, this unusual structure does not prevent aminoacyltRNAPyl binding by translation factor EF-Tu [21]. A stem-loop with conserved elements just 30 of the UAG codon in mttB and mtmB genes may be involved in circumventing UAG function as a stop codon during decoding as pyrrolysine [22].
MtmC, substrate of the pyrrolysyl-protein MtmB The metabolically relevant product of MtmB activity is methyl-Co(III)-MtmC, which displays the electronic spectra of a single bound hexacoordinate methyl-Co(III) 5-hydroxybenzimidazolylcobamide cofactor [24]. MtmC cycles between methyl-Co(III)- and Co(I)-MtmC during catalysis. MMA:CoM methyl transfer can be initiated
It is formally possible that pyrrolysine is made following lysine incorporation into the nascent polypeptide. Figure 3
(a)
(b)
H106
E259 Q333 Y335
E229 O202
Current Opinion in Chemical Biology
Tertiary structures of MtmB and its substrate, MtmC. (a) Theoretical MtmC (GenBank AF013713) structure made by the Swiss-Model server [36] using the cobalamin binding domain of MetH (PDB ID code, 1bmtA) as template. The root mean square deviation between the Ca atoms of the MtmC model and the template is 0.4 A˚. The corrinoid is placed in the MtmC model by superimposing the MetH and MtmC structures. Coils that deviate significantly from the template are in red. (b) MtmB showing location of pyrrolysine (O202) and surrounding residues proposed to participate in methyl group transfer from MMA to MtmC. Current Opinion in Chemical Biology 2004, 8:484–491
www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 487
with methyl-Co(III)-MtmC, but hydroxy-Co(III)-MtmC requires reduction, presumably to Co(I)-MtmC since lowpotential reducing agents are required [10]. MethylCo(III)-MtmC and MtbA can catalyze continuous methyliodide:coenzyme M methyl transfer, consistent with generation of Co(I)-MtmC by methyl cation transfer to coenzyme M, followed by nucleophilic attack by Co(I) on methyliodide [10]. MtbB, the DMA methyltransferase, as well as MtaB, the methanol methyltransferase, will methylate free cob(I)alamin with their respective substrates [8,25]. Taken together, these results indicate that methanogen methylotrophic corrinoid proteins such as MtmC are methylated in the Co(I) state, forming a hexacoordinate methyl-Co(III) species whose demethylation regenerates Co (I) corrinoid, in keeping with the behavior of other corrinoid-dependent methyltransferases [26]. MtmC, as well as other small methanogenic corrinoid proteins, are close homologs of the cobalamin-binding module of MetH [5,16,25,27]. Homology modeling using the known structure of the MetH cobalamin-binding module [28] as template produces a MtmC model with little deviation from the Ca backbone in the regions of
MetH contacting cobalamin (Figure 3). This yields insight into the function of MtmC, as the well-studied MetH binds cobalamin with the benzimidazole base bound to the protein, and a histidinyl residue serving as a ligand to cobalt. The MtmC model displays conservation of histidine (H106) as cobalt ligand, as well as conserved placement of the seryl (S159) and aspartyl (D104) residues that in MetH are proposed to participate in protonating and deprotonating the histidinyl ligand as hexacoordinate methyl-Co(III) or tetracoordinate Co(I) are formed during catalysis. This is suggested as a means by which conformation changes occur for cobalamin to interact with the three methyltransferase modules of MetH [29,30]. MtmC must interact with the different active sites of MtbA and MtmB during catalysis, and the ligands to cobalt may play a similar role.
Pyrrolysine in MtmB mediated methylation of MtmC The crystal structure of MtmB revealed a dimer of trimers with the subunit fold that of a modified ab TIM barrel. The C-terminus of the central cluster of b sheets forms a negatively charged surface cleft that faces out from the core of hexamers towards the solvent. Pyrrolysine is found
Figure 4
Co
B H
Co CH3
B H
Y335
N H
CH3
H
E229 H
N E229 H
Q333
N
Q333
H
N
H
Y335 H
H
E259 O
E259
O
Nε
Nε + CH3NH3+ – NH3
H
Co
B N H E229
Y335
H H
H E259 O
E229 H
H H
Q333
N
– CH3-MtmC Nε
Y335 N
Q333
N
CH3
B H
H O
E259 Nε Current Opinion in Chemical Biology
Postulated role of pyrrolysine during MtmB-catalysed methylation of the corrinoid cofactor of MtmC. The scheme shown is a variation on the mechanism originally proposed by Hao et al. [4]. www.sciencedirect.com
Current Opinion in Chemical Biology 2004, 8:484–491
488 Mechanisms
near the center of this cleft, with sufficient surrounding space to allow close approach of a corrin ring to the UAGencoded residue [4].
an in-line nucleophilic attack by tetracoordinate Co(I)MtmC. Although it was initially proposed that Y335 and Q333 participate in binding methylamine, they could also serve to H-bond with the ammonia released from the pyrroline ring, enhancing the equilibrium between 2aminopyrrolysine and pyrrolysine in the direction of product removal from the enzyme. This model of pyrrolysine function has the strong appeal of applicability to pyrrolysine function in the MMA, DMA and TMA methyltransferases. This cannot be said of alternative models involving pyrrolysine, such as imine bond hydrolysis and formation of an external imine bond between pyrrolysine and methylamine, which is feasible with MMA, but not TMA.
Evidence for reactivity of the C-2 atom of the pyrrolysine ring came from two crystal forms of MtmB obtained from solutions containing NaCl or (NH4)2SO4. In the NaCl crystal form, the pyrroline was unmodified at the C-2 position. The second crystal form was a mixture of two pyrrolysine states. One resembles pyrrolysine in the NaCl crystal form, while the second was modeled with the pyrrolysine ring rotated 908 and with an amine adduct at the C-2 position. Residues within H-bonding distance of pyrrolysine were repositioned in the presence of 2aminopyrrolysine. In addition, two nearby residues were observed at partial occupancy to H-bond with ammonia.
While lacking sequence similarity, MtmB is a structural homolog of the cobalamin dependent methyl-tetrahydrofolate (meTHF) and homocysteine methyltransferase modules of MetH [31], as well as AcsE-MeTr, a meTHF:corrinoid methyltransferase involved in acetylCoA synthesis [32]. The residues for binding substrates methylating or demethylating corrinoid in these methyltransferases are at the C-terminal portion of the inner aspect of the ab TIM barrel of each protein, as are pyrrolysine and surrounding residues in MtmB.
The reactivity of the C-2 carbon and the changes in the residues surrounding pyrrolysine led Hao et al. to propose a model for MtmB methylation of Co(I) corrinoid cofactor with MMA [4]. A variation on this proposal is shown in Figure 4. An unidentified base deprotonates the MMA ion attracted to the negatively charged cleft, otherwise the enzyme must utilize the small fraction of unionized MMA present at neutral pH. Y335 and Q333 are proposed to assist in the initial MtmB binding of MMA and positioning such that electrophilic attack of the C-2 atom on the lone electron pair of the nitrogen of unionized MMA occurs. The observed reorientation of the pyrroline ring would result as nearby residues move to H-bond with the new C-2 substituent, orienting the methyl group for
The location of pyrrolysine is particularly striking when compared with the MetH meTHF methyltransferase domain. The methylamine group of the proposed 2methylamino-pyrrolysine intermediate and the methyl group of meTHF are within 3 A˚ of each other when
Figure 5
(a)
(b)
Current Opinion in Chemical Biology
Structural homology of MetH MeTHF methyltransferase module and MtmB. (a) Structure of MeTHF methyltransferase module (PDB ID code 1Q8J, residues 300 to 559) with bound MeTHF, surrounding residues are those implicated in binding the substrate. (b) MeTHF methyltransferase domain superimposed with MtmB from (NH4)2SO4 crystal form (PDB ID code 1L2Q). The two proteins were aligned using the DALI program [37], with 230 Ca atoms aligned with a root mean squares difference of 3.1 A˚ . Only conserved core elements are shown in dark green, (MtmB) or light blue-green (MetH MeTHF domain). Arrows point to the C-2 amine of 2-aminopyrrolysine (red), or the methyl group of MeTHF (black). Current Opinion in Chemical Biology 2004, 8:484–491
www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 489
the TIM barrel of MtmB is optimally superimposed onto that of the meTHF domain (Figure 5) using a structural alignment program. This apparent conservation of geometry of methyl groups transferred to homologous corrinoid proteins further supports the notion that pyrrolysine acts to optimally position a methylammonium species for methyl transfer to corrinoid.
Acknowledgements Work in the author’s laboratory was supported from funds from NSF (MCB-9808914), DOE (DEFG02-92ER20042) and NIH (GM061796). The author thanks our collaborators over the years with special acknowledgement to our recent collaborations with William Metcalf and Kari Green-Church, and to David Grahame for useful discussions. I especially wish to thank Bing Hao and Michael Chan, with whom my laboratory was privileged to share in the discovery of pyrrolysine.
References and recommended reading Conclusions Recent discoveries indicate the genetic encoding of the 22nd amino acid may be much different than the 21st. However, like selenocysteine, pyrrolysine appears to have been recruited/retained in the code because of its unique properties. Pyrrolysine is the first example of a genetically encoded electrophilic residue, a property previously found only in protein prosthetic groups or posttranslationally modified residues [33]. Although the proposed function of pyrrolysine fits current observations, very much remains to be demonstrated. For example, the hypothetical catalytic intermediate of 2 methylaminopyrrolysine remains unobserved, and even the necessity of pyrrolysine for catalysis by MtmB has not yet been directly tested. However, the on-going development of genetic systems for Methanosarcina spp. [34] is now making site-directed mutagenesis feasible in a host decoding UAG as pyrrolysine.
Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest 1.
Garavelli JS: The RESID Database of Protein Modifications as a resource and annotation tool. Proteomics 2004, 4:1527-1533.
2.
Stadtman TC: Selenocysteine. Annu Rev Biochem 1996, 65:83-100.
3.
Commans S, Bo¨ ck A: Selenocysteine inserting tRNAs: an overview. FEMS Microbiol Rev 1999, 23:335-351.
4.
Hao B, Gong W, Ferguson TK, James CM, Krzycki JA, Chan MK: A new UAG-encoded residue in the structure of a methanogen methyltransferase. Science 2002, 296:1462-1466. This paper is the first description of the UAG-encoded residue of MtmB as pyrrolysine. The structure of MtmB was solved for the two different crystal forms originating from supernatants containing NaCl or (NH4)2SO4. By estimating the occupancies of two different conformations of the UAGencoded residue, models for both the structure and function of pyrrolysine were proposed. 5.
Given pyrrolysine’s potential to form amine or imine derivatives, it might be present in other enzymes with amine substrates. If so, it should prove of wider distribution than the small number of microbes now known to possess the pyl genes. Nonetheless, even with the present restricted distribution to species of Methanosarcina, Desulfitobacterium and Methanococcoides, pyrrolysine now serves as an important illustration that the natural genetic code is sufficiently flexible to allow relatively idiosyncratic incorporation of a highly specialized residue. Biological methane formation from an entire class of substrates appears to be based upon this premise.
Update Peptide sequences made from the DMA and MMA methyltransferase genes before and after their in-frame amber codons have previously been obtained, are consistent with UAG translation as pyrrolysine in both proteins [16,19]. In contrast, evidence of translation of the UAG codon in the TMA methyltransferase gene has been based largely on the perceived requirement for UAG readthrough to produce the full-length TMA methyltransferase [9,16]. Recently, in a mass spectrometric survey of proteins in Methanococcoides burtonii, peptides from a TMA methyltransferase were identified [38]. Although a pyrrolysyl-containing peptide was not observed, peptides were identified from both the N- and C-terminal portions of the protein, consistent with translation of UAG in this methyltransferase as pyrrolysine. www.sciencedirect.com
Burke SA, Lo SL, Krzycki JA: Clustered genes encoding the methyltransferases of methanogenesis from monomethylamine. J Bacteriol 1998, 180:3432-3440.
6.
Srinivasan G, James CM, Krzycki JA: Pyrrolysine encoded by UAG in Archaea: charging of a UAG-decoding specialized tRNA. Science 2002, 296:1459-1462. The first description of tRNACUA (tRNAPyl). The presence of tRNACUA is important supporting evidence that pyrrolysine is encoded by the mtmB1 amber codon, rather than being a post-translational modification. The tRNA gene was identified using a computer program (tRNAScan) developed by the group of Sean Eddy. Contigs of the partially sequenced M. barkeri Fusaro genome were scanned, with preference given to those that encode methylamine catabolism proteins, as microbial genes often cluster with others for a particular biochemical pathway. In addition, this paper described the other pyl genes, including pylS, which has recently been shown to be a pyrrolysyl-tRNA synthetase.
7. Atkins JF, Gesteland R: The 22nd amino acid. Science 2002, 296:1409-1411. This perspective published with the pyrrolysine and pyl gene description papers relates pyrrolysine to the larger picture of the genetic code and recoding events such as selenocysteine. 8.
Ferguson DJ Jr, Gorlatova N, Grahame DA, Krzycki JA: Reconstitution of dimethylamine:coenzyme M methyl transfer with a discrete corrinoid protein and two methyltransferases purified from Methanosarcina barkeri. J Biol Chem 2000, 275:29053-29060.
9.
Ferguson DJ Jr, Krzycki JA: Reconstitution of trimethylaminedependent coenzyme M methylation with the trimethylamine corrinoid protein and the isozymes of methyltransferase II from Methanosarcina barkeri. J Bacteriol 1997, 179:846-852.
10. Burke SA, Krzycki JA: Reconstitution of monomethylamine:coenzyme M methyl transfer with a corrinoid protein and two methyltransferases purified from Methanosarcina barkeri. J Biol Chem 1997, 272:16570-16577. 11. Ferguson DJ Jr, Krzycki JA, Grahame DA: Specific roles of methylcobamide:coenzyme M methyltransferase isozymes in metabolism of methanol and methylamines in Methanosarcina barkeri. J Biol Chem 1996, 271:5189-5194. Current Opinion in Chemical Biology 2004, 8:484–491
490 Mechanisms
12. Burke SA, Krzycki JA: Involvement of the ‘‘A’’ isozyme of methyltransferase II and the 29-kilodalton corrinoid protein in methanogenesis from monomethylamine. J Bacteriol 1995, 177:4410-4416. 13. Gencic S, LeClerc GM, Gorlatova N, Peariso K, Penner-Hahn JE, Grahame DA: Zinc-thiolate intermediate in catalysis of methyl group transfer in Methanosarcina barkeri. Biochemistry 2001, 40:13068-13078. 14. Kruer M, Haumann M, Meyer-Klaucke W, Thauer RK, Dau H: The role of zinc in the methylation of the coenzyme M thiol group in methanol:coenzyme M methyltransferase from Methanosarcina barkeri. Eur J Biochem 2002, 269:2117-2123. This article and [13] address zinc ligation to enzyme and coenzyme M in MtbA and MtaA, two methylcobamide:CoM methyltransferases involved respectively with the methylamine and methanol methyltransferases. 15. Thauer RK: Biochemistry of methanogenesis: a tribute to Marjory Stephenson. 1998 Marjory Stephenson Prize Lecture. Microbiology 1998, 144:2377-2406. 16. Paul L, Ferguson DJ, Krzycki JA: The trimethylamine methyltransferase gene and multiple dimethylamine methyltransferase genes of Methanosarcina barkeri contain in-frame and read-through amber codons. J Bacteriol 2000, 182:2520-2529. 17. Galagan JE, Nusbaum C, Roy A, Endrizzi MG, Macdonald P, FitzHugh W, Calvo S, Engels R, Smirnov S, Atnoor D et al.: The genome of M. acetivorans reveals extensive metabolic and physiological diversity. Genome Res 2002, 12:532-542. See annotation to [18]. 18. Deppenmeier U, Johann A, Hartsch T, Merkl R, Schmitz RA, Martinez-Arias R, Henne A, Wiezer A, Ba¨ umer S, Jacobi C et al.: The Genome of Methanosarcina mazei: Evidence for Lateral Gene Transfer Between Bacteria and Archaea. J Mol Microbiol Biotechnol 2002, 4:453-461. This article and [17] demonstrate conclusively that all methylamine methyltransferase genes (present in multiple copies) in these genomes possess in-frame amber codons. Prior to the availability of these complete genomes, it remained possible that other methylamine methyltransferase genes lacking amber codons were responsible for production of full-length methylamine methyltransferases, rather than UAG translation as pyrrolysine. 19. James CM, Ferguson TK, Leykam JF, Krzycki JA: The amber codon in the gene encoding the monomethylamine methyltransferase isolated from Methanosarcina barkeri is translated as a sense codon. J Biol Chem 2001, 276:34252-34258. 20. Hao B, Gang Z, Kang PT, Soares JA, Ferguson TK, Gallucci J, Krzycki JA, Chan MK: Reactivity and chemical synthesis of L-pyrrolysine - the 22nd genetically-encoded amino acid. Chem Biol 2004, in press. 21. Theobald-Dietrich A, Frugier M, Giege R, Rudinger-Thirion J: Atypical archaeal tRNA pyrrolysine transcript behaves towards EF-TU as a typical elongator tRNA. Nucleic Acids Res 2004, 32:1091-1096. Structure probing reported in this paper confirmed the proposed atypical secondary structure of tRNAPyl with a six base-pair anticodon stem. This recalls the elongated acceptor stem of tRNASec, and suggests unusual tRNA structure may prove a common theme of tRNAs for genetically encoded non-canonical amino acids. Further, by changing the anticodon of tRNAPyl to achieve slow lysylation with yeast lysyl-tRNA synthetase, the authors demonstrate that lysyl-tRNAPyl is recognized by E. coli translation factor EF-Tu. This is unexpected, as selenocysteinyl-tRNASec is not recognized by EF-Tu, rather a dedicated selenocysteine translation factor exists. However, this result does not discount the possibility that an archaeal translation factor specific for pyrrolysyl-tRNAPyl exists in the Methanosarcina species. 22. Namy O, Rousset JP, Napthine S, Brierley I: Reprogrammed genetic decoding in cellular gene expression. Mol Cell 2004, 13:157-168. Notes the presence of stem-loops with conserved elements downstream of the UAG codon in mtmB and mttB. 23. Polycarpo C, Ambrogelly A, Ruan B, Tumbula-Hansen D, Ataide SF, Ishitani R, Yokoyama S, Nureki O, Ibba M, Soll D: Activation of the pyrrolysine suppressor tRNA requires Current Opinion in Chemical Biology 2004, 8:484–491
formation of a ternary complex with class I and class II lysyl-tRNA synthetases. Mol Cell 2003, 12:287-294. Multiple clones producing recombinant PylS were found not to possess lysyl-tRNA synthetase activity, discounting earlier indications that the enzyme was capable of this reaction (see [6]). Furthermore, the authors demonstrate that LysRS1 and LysRS2 can together slowly aminoacylate tRNAPyl transcripts with lysine. Recent work by our group demonstrates PylS is a pyrrolysyl-tRNA synthetase, indicating that two competing reactions involving tRNAPyl may exist in vivo, only one of which would directly act to produce pyrrolysyl-tRNAPyl. 24. Kremer JD, Cao X, Krzycki J: Isolation of two novel corrinoid proteins from acetate-grown Methanosarcina barkeri. J Bacteriol 1993, 175:4824-4833. 25. Sauer K, Harms U, Thauer RK: Methanol:coenzyme M methyltransferase from Methanosarcina barkeri. Purification, properties and encoding genes of the corrinoid protein MT1. Eur J Biochem 1997, 243:670-677. 26. Banerjee R, Ragsdale SW: The many faces of vitamin B12: catalysis by cobalamin-dependent enzymes. Annu Rev Biochem 2003, 72:209-247. 27. Paul L, Krzycki JA: Sequence and transcript analysis of a novel Methanosarcina barkeri methyltransferase II homolog and its associated corrinoid protein homologous to methionine synthase. J Bacteriol 1996, 178:6599-6607. 28. Drennan CL, Huang S, Drummond JT, Matthews RG, Ludwig ML: How a protein binds B12: A 3.0 A˚ X-ray structure of B12-binding domains of methionine synthase. Science 1994, 266:1669-1674. 29. Bandarian V, Pattridge KA, Lennon BW, Huddler DP, Matthews RG, Ludwig ML: Domain alternation switches B(12)-dependent methionine synthase to the activation conformation. Nat Struct Biol 2002, 9:53-56. The structure of the MetH cobalamin binding domain binding a methyltransferase domain (the activation domain) reveals the displacement of the corrinoid from the histidine ligand, suggesting a mechanism for control of conformation as the cobalamin module moves from methyltransferase to methyltransferase module in this multi-domain protein. 30. Dorweiler JS, Finke RG, Matthews RG: Cobalamin-dependent methionine synthase: probing the role of the axial base in catalysis of methyl transfer between methyltetrahydrofolate and exogenous cob(I)alamin or cob(I)inamide. Biochemistry 2003, 42:14653-14662. Demonstration that the ligand triad (D, S and H) does not greatly contribute to catalysis, but may play a role in conformational changes controlling cobalamin interaction with different methyltransferase domains. 31. Evans JC, Huddler DP, Hilgers MT, Romanchuk G, Matthews RG, Ludwig ML: Structures of the N-terminal modules imply large domain motions during catalysis by methionine synthase. Proc Natl Acad Sci USA 2004, 101:3729-3736. The structures of the homocysteine and meTHF methyltransferase were determined, showing for the first time how a corrinoid-dependent methyltransferase might bind and orient substrates acting as methyl donors and acceptors to the central cobalt ion. Like the MtmB and AcsE-Mtr proteins, the MetH methyltransferases adopt the ubiquitous TIM barrel fold, suggesting that may be a near-common feature of the majority of corrinoid dependent methyltransferases. 32. Doukov T, Seravalli J, Stezowski JJ, Ragsdale SW: Crystal structure of a methyltetrahydrofolate- and corrinoiddependent methyltransferase. Structure Fold Des 2000, 8:817-830. 33. Retey J: Discovery and role of methylidene imidazolone, a highly electrophilic prosthetic group. Biochim Biophys Acta 2003, 1647:179-184. One of the few other electrophilic groups found in proteins is reviewed in detail. 34. Meuer J, Kuettner HC, Zhang JK, Hedderich R, Metcalf WW: Genetic analysis of the archaeon Methanosarcina barkeri Fusaro reveals a central role for Ech hydrogenase and ferredoxin in methanogenesis and carbon fixation. Proc Natl Acad Sci USA 2002, 99:5632-5637. One of the first uses of genetics to investigate the biochemistry of methanogenesis, it is representative of the ongoing pioneering development of this essential group of techniques from the Metcalf laboratory. www.sciencedirect.com
Function of genetically encoded pyrrolysine in corrinoid-dependent methylamine methyltransferases Krzycki 491
35. Pol A, van der Drift C, Vogels GD: Corrinoids from Methanosarcina barkeri: structure of the alpha-ligand. Biochem Biophys Res Commun 1982, 108:731-737. 36. Schwede T, Kopp J, Guex N, Peitsch MC: SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res 2003, 31:3381-3385. 37. Holm L, Sander C: Dali: a network tool for protein structure comparison. Trends Biochem Sci 1995, 20:478-480. 38. Amber Goodchild A, Saunders NF, Ertan H, Raftery M, Guilhaus M, Curmi PMG, Cavicchioli R: A proteomic determination of cold
www.sciencedirect.com
adaptation in the Antarctic archaeon Methanococcoides burtonii. Mol Microbiol 2004, 53:309-321.
Now in press The work referred to as (Blight et al., unpublished data) is now in press: 39. Blight SK, Larue RC, Mahapatra A, Longstaff DG, Chang E, Zhao G, Kang PT, Green-Church KB, Chan MK, Krzycki JA: Direct charging of tRNACUA with pyrrolysine in vitro and in vivo. Nature 2004, in press.
Current Opinion in Chemical Biology 2004, 8:484–491