Substrate Recognition Mechanism of the Putative ... - ACS Publications

Subscriber access provided by NEW YORK UNIV

Article

Substrate Recognition Mechanism of the Putative Yeast Carnosine N-methyltransferase Xiwen Liu, Jialiang Wu, Yujie Sun, and Wei Xie ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.7b00328 • Publication Date (Web): 27 Jun 2017 Downloaded from http://pubs.acs.org on July 3, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Substrate Recognition Mechanism of the Putative Yeast Carnosine N-methyltransferase Xiwen Liu1, Jialiang Wu1, Yujie Sun1, Wei Xie1, *

1

Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory for

Biocontrol, School of Life Sciences, The Sun Yat-Sen University, 135 W. Xingang Rd., Guangzhou 510275, People's Republic of China.

*: Correspondence Wei Xie: [email protected]. Tel.: 862039332943; Fax: 862039332847

Running title: Substrate Recognition Mechanism of N-methyltransferase

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

ABSTRACT Anserine (β-alanyl-N(Pi) -methyl-L-histidine) is a natural metabolite present in skeletal muscle and the central nervous system of vertebrates, and plays important physiological roles in living organisms. The production of anserine is catalyzed by carnosine N-methyltransferases, which transfer a methyl group to carnosine (β-alanyl-L-histidine). However, the structural basis of the substrate recognition for the enzymes is unknown. We present the crystal structure of the putative carnosine N-methyltransferase from yeast named YNL092W in complex with SAH, solved by the single-wavelength anomalous dispersion (SAD) method. The protein contains a typical Rossmann domain and a characteristic N-terminal helical domain. At the cofactor-binding site, SAH forms extensive interaction network with the enzyme. The individual contribution of each residue to ligand affinity and enzyme activity was assessed by ITC and methyltransferase assays after mutagenesis of the key residues. Additionally, docking studies and activity assays were conducted in order to identify the binding site for carnosine and a plausible complex model was proposed. Furthermore, we discovered that two disulfide bridges might be functionally important to the enzyme. By comparison to structure- and sequence-similar methyltransferases, we deduce that the enzyme most likely acts on a protein substrate. Our structural analyses shed light on the catalytic mechanism and substrate recognition by YNL092W. KEYWORDS:

Crystal

structure,

enzyme

catalysis,

molecular

N-methyltransferase, Rossmann domain.

2


docking,

carnosine

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ABBREVIATION SAH, S-adenosyl-L-homocysteine; SAM, S-adenosyl-L-methionine; ITC, isothermal titration calorimetry; HNMT, histamine N-methyltransferase; ROS, reactive oxygen species; MTase, methyltransferase; SEC-SLS, size exclusion chromatography-Static Light Scattering; RMSD, root mean square deviation.

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

INTRODUCTION Anserine (β-alanyl-N(Pi)-methyl-L-histidine) and carnosine (β-alanyl-L-histidine, the precursor of anserine) are naturally occurring dipeptides consisting of beta-alanine and histidine. Both molecules are neutral molecules with a positive and negative end, and therefore behave as zwitterion. Present in skeletal muscle and the central nervous system of vertebrates in large quantities,1-3 they have a number of properties that proved to be important. They were originally considered to be the buffers to neutralize the acidity caused by lactic acid in the muscle tissue due to their abundance and close-to-neutral pKas.4 Alternatively, the dipeptides may scavenge reactive oxygen species (ROS) as well as unsaturated aldehydes generated from the peroxidation of fatty acids during oxidative stresses. Additionally, carnosine acts as an anti-glycating agent, as it slows down the formation of advanced glycation end-products (AGEs), which play a pathogenic role in the development and progression in many degenerative diseases such as Alzheimer's.5 The formation of anserine requires the methyltransfer step of a methyl group to the histidine moiety of carnosine, catalyzed by carnosine N-methyltransferases (Fig.1a). Although carnosine has been discovered for a long time, very little is known about the responsible enzyme. Drozak et al. identified chicken carnosine N-methyltransferase as a histamine N-methyltransferase-like (HNMT-like) protein.6 However, HNMT-like proteins were not found in available mammalian genomes, even though carnosine N-methyltransferase activity is apparently present in many mammalians species.6 This paradox led to further investigations and the discovery of the UPF0586 protein as the anserine producer. The UPF0586 protein was expressed herterologously in COS-7 cells, and the purified enzyme has been demonstrated to display methyltransfer activity as confirmed by 4


Page 4 of 29

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


both chromatographic and mass spectrometry analyses.7 Following this work, Drozak et al. also identified the genes encoding carnosine N-methyltransferase in yeast, rat, and human, and expressed them in COS-7 cells. Methyltransfer activity assays were carried out in vitro to verify the enzymes’ activities.7 The carnosine N-methyltransferases from different sources all contain an N-2227 domain, and a few conserved residues within the domain were thought to be involved in the binding of SAM. Moreover, recombinant UPF0586 proteins from the rat and human sources expressed in HeLa cells are localized in both cytosol and the nucleus, in line with previous studies conducted by Uhlen et al.8 YNL092W (another name for the yeast ortholog of UPF0586) has been shown to display moderate methyltransfer activity toward carnosine. Later, Szczepińska et al. showed that YNL092W is a protein MTase via in-vitro MTase activity assays.9 When the purified recombinant protein was incubated with [3H]-SAM, a protein band at ~55 kDa occurred on the tritium screen. This methylated product corresponded to the molecular weight of YNL092W, suggesting that the enzyme was capable of methylating itself. However, no further investigation was carried out beyond this point. Although the enzymes responsible for the formation of anserine have been identified and characterized biochemically, their catalytic mechanism is poorly understood. Importantly, the physiological substrates of YNL092W are unknown. In this study, we solved the 2.2-Å structure of YNL092W in complex with the ligand SAH using the single-wavelength anomalous dispersion (SAD) method, the first structure of carnosine N-methyltransferases. We investigated the interactions between the enzyme and the SAH cofactor by ITC and methyltransfer assays, as well as the possible binding mode of carnosine. These studies provide insights into the substrate recognition mechanism by YNL092W, and help us to understand the physiological role of this enzyme. 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

RESULTS The overall structure of YNL092W. The YNL092W protein construct used for crystallization contains 406 residues, with six extra amino acids GPHMAS being appended at the N-terminus after the removal of the affinity tag. Each asymmetric unit contains two monomers, and they form a dimer. Chain A is visible from N8-W219, M225-P324, I332-N395, while chain B is visible from Q9-N395, free of internal disorders. The refined model contains a total of 763 amino acids, 212 water molecules and a Na+ ion. In addition, each chain also contains a SAH molecule. The model is of good geometry, with no residues falling in the outlier region of the Ramachandran plot. The refinement statistics are summarized in Table 1. Each monomer is composed of an N-terminal helical domain (αN) and a basic Rossmann domain at the C terminus. The αN domain is composed of four helices. α4 is the longest helix in the αN domain while α1-3s are relatively short. Because the presence of several kinks in the connecting regions between the α1-3 helices, they are not coaxial. α4 runs almost antiparallel to the imperfect helix consisting of α1-3. The basic Rossmann domain comprises a β-sheet of seven strands, which are sandwiched by four helices on each side (Fig. 1b). The topology of the domain is illustrated in Fig. 1c. The 7 strands adopt a β↓β↓β↓β↓β↓β↑β↓ orientation, with the C-terminal strand β11 antiparallel to all the other strands. The two monomers in the asymmetric unit form a stable dimer with a cradle shape, consistent with SEC-SLS analyses (Size exclusion chromatography-static light scattering) (data not shown). Each chain forms numerous interactions with the other subunit including hydrogen bonds, salt bridges and hydrophobic interactions and buries 5759.4 Å2 surface area. The β10-β11 loop (C369-Y385) forms a 6


Page 6 of 29

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


β-turn. Along with its counterpart from the other subunit, they form part of the interface and seal the bottom of the cradle. The cradle is rather deep, with its depth as large as ~ 40 Å. The two monomers resemble each other, and the superposition of the two chains yielded an RMSD less than 0.6 Å for 371 Cα atoms. The SAH-binding mode. At the substrate-binding pocket of each subunit, the map shows clear density of one molecule of SAH, which forms extensive network with the surrounding residues. Specifically, the adenine ring stacks onto the side chain of F168 and also forms an “edge-to-face” interaction with F276 (Fig. 2a). Meanwhile, the base forms three specific hydrogen bonds with the side chains of Y286, S234, and T279 respectively, while the N1 nitrogen forms another with the main chain of F235. In addition, the ribose donates two hydrogen bonds to the terminal carboxylate group of E167 through the 2’ and 3’ hydroxyls. Moreover, the amino acid moiety adds eight more contacts. In particular, the nitrogen of the methionyl moiety donates two hydrogen bonds to the backbone carbonyl oxygens of G146 and N274 respectively. Dissimilar to the two non-specific interactions with the nitrogen, all the contacts on the terminal carboxylate group are specific hydrogen bonds or salt bridges from side chains of specific residues. Particularly, the carboxylate makes two salt bridge interactions with R111, while receiving a total of four hydrogen bonds from N274, Y371, and Q108 respectively. Moreover, there are three well-positioned water molecules that are involved in hydrogen bonding contacts with the SAH ligand. Because the aforementioned interactions are mostly with conserved residues (except for S234 and N274), we therefore made the mutations on the key residues (Q108A, S234A, E167A, R111A, N274A, D278A/T279A, Y371A and Y286A) to test their contribution to the affinity of SAM using ITC. All 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the mutants except for Y371A were well expressed and behaved well on as assessed by gel-filtration chromatography, suggesting no significant impairment to the folding process induced by the mutations. Among these mutants, N274A also had substantial reduction in terms of expression level. The ITC results showed that the affinity of the WT enzyme to SAM is rather tight, with a KD ~ 0.34 µM (Fig. 2b). The kM of the enzyme to SAM reported by Drozak et al. was 2 µM.8 The discrepancy may be due to the different parameters measured and different assays employed. The different expression hosts of the recombinant UPF0586 (mammalian cells vs. E. coli cells) might play an additional role. Interestingly, we discovered that it appears that the enzyme dimer binds only one SAM molecule, because the stoichiometry value n is consistently less than 0.4. Additionally, SAH binds to the enzyme with an even slightly lower stoichiometry (data not show). These observations implied that the enzyme catalyzes through the “half-of-the-sites reactivity” mechanism, in which only one active site within the dimer enzyme is functional, but this theory warrants more investigation. In contrast, all the mutants except for Y286A barely displayed any affinity to SAM, as suggested by the small binding energies released as well as the irregular shape of the titration curves (Fig. 2b). Y286A displayed ~10-fold lower affinity to SAM, as this residue forms a long-distance hydrogen bond (3.45 Å) with the N6 atom of adenine. We further conducted the methyltransfer activity assays using carnosine as the substrate, and discovered that the activity results showed a similar trend to those of ITC. Except for N274A and S234A (the two mutations on non-conserved residues), which retained ~2/3 and 1/3 activities of that of WT, the rest of the mutants only exhibited 0.84-10% activities (15-min time point measurements) (Fig. 2c).

8


Page 8 of 29

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Possible binding mode of carnosine. Although we tried cocrystallization or soaking in the presence of carnosine or the tripeptide GGH at a concentration as high as 50 mM, we did not observe the ligand density in the resulting crystal structures. This may be attributed to the poor affinity for the ligand (KM for carnosine is ~21.5 mM).7 The substrate-binding regions (the Rossmann fold) of most methyltransferases are conserved, but their substrate-binding sites are highly variable due to diversified substrates.10 In order to gain understanding on the binding mode of the carnosine, we employed the program Autodock to dock the carnosine ligand into its binding pocket. The identification of the binding site was performed with Autodock vina (ver. 1.5.6) and the protein dimer with the SAH cofactor was used as the receptor, due to the possibility of carnosine’s binding to the dimer interface. The grid box was set with all three dimensions ~25 Å and the potential binding sites generated by the program were further selected according to the two following criteria: 1) the carnosine molecule should be close enough to the SAH molecule present in the crystal structure, and the N atom of carnosine should be near the predicted methyl group in order to be methylated; 2) the carnosine molecule should be ideally stabilized by specific interactions with the enzyme. According to these two criteria, we discovered several potential carnosine-binding sites and we selected the first three models with top scores for experimental validation. The distances between the N atom of carnosine and the S atom to which the methyl group is attached are 3.7, 6.5 and 3.9 Å in these models respectively, and the carnosine molecules in each model makes at least one hydrogen bond interaction with the enzyme (Figs. 3a-c). To decide which mode is closer to the genuine mode, we made the Y385A, H309A, L307F, D278A, S104A and S169A mutations, and tested the methyltransfer activities of these variants. All the mutants behaved normally on a gel-filtration column. Activity 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

assays indicated that while the S169A mutation greatly stimulated enzymatic activity, the Q108A and D278A mutants had only residual activities left (the D278A/T279A mutant was similar to the D278A single mutant, Fig. 3d). On the other hand, the activities of S104A and L307F were nearly unchanged, compared to that of the WT enzyme. The high activity of L307F and relative high activity of H309A (> 40% WT activity) argue against the first candidate model. Although L307 makes a hydrogen bond with carnosine using its main chain, a mutation to the bulkier phenylalanine is very likely to disrupt the binding of carnosine (Fig. 3a). In the second candidate model, among the mutants of the three key residues, Q108A was the only one to show low activity, which contradicts this model as well (Fig. 3b). The loss of activity by the Q108A mutation is probably due to the double roles of Q108, which also forms a hydrogen bond with SAH. On the other hand, all the mutations Y385A, D278A and D278AT279A led to substantial or even the largest loss in enzymatic activities, supporting the interactions presented by this model (Fig. 3c). In this model, carnosine is located close to the dimer interface, where it is held by hydrogen bonding interactions from Q108, Y385, D278 and F275. Judging from our modeling and mutagenesis studies, the third model may be closest to reality. The disulfide bonds. Two intra-subunit disulfide bridges, C123-C369 and C231-C238 are found in the finished structure. These two disulfide bridges connect α5-β11 and β5-α9 respectively (Fig. 4a). Among these four cysteines, only C123 is conserved. On the other hand, C231 is located on β5, one of the strands forming the core Rossmann fold. Since carnosine has been reported to display antioxidant properties, we wonder whether the disulfide bond is involved in catalysis or structure-stability maintenance. We therefore created the C123A and C238A mutants, and tested if the disruption of the bridges affects activity of the enzyme. Interestingly, C123A displays strongly enhanced activity than 10


Page 10 of 29

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


WT while C238A significantly reduced its activity (Fig. 4b). Coincidentally, both the disulfide bonds connect the outermost β-strand with nearby substructures. Close inspection indicated that the disruption of the two disulfide bonds may impact the binding of SAM/SAH. Particularly, the C238 residue is in the vicinity of the adenine ring of SAH in the crystal structure. The breakage of this disulfide bond may loosen the interactions with the β5-α9 loop, and may in turn affect the positioning of the S234F235 dipeptide, which forms two hydrogen bonds with the adenine ring. In contrast, C123 is located near the carboxylate tail of SAM/SAH, and the mutation may affect the local structure containing Y371. Y371 forms a hydrogen bond with the carboxylate oxygen. Although both mutations could disturb the binding of the SAM cofactor indirectly as we described above, it is difficult to explain their distinct effects on activity, especially the considerable boost by the C123A mutation. Structure comparison to homologs. Dali search for structurally similar proteins11 resulted in similarity with several proteins, and the top three are the PH0226 protein from Pyrococcus horikoshii OT3 (PDB ID: 1VE3), the putative SAM-dependent methyltransferase MMP1179 from Methanococcus maripaludis (PDB ID: 3D1C), Phosphoethanolamine N-methyl transferase from Plasmodium knowlesi (PDB ID: 4IV8), and the methyltransferase BVU_3255 from Bacteroides vulgatus (PDB ID: 3T7R). All these proteins are annotated as a SAM-binding proteins/enzymes, but were deposited in the PDB database without related publications. Among these proteins, the closest structural homolog PH0226 can be aligned to YNL092W with an RMSD of 2.03 Å for 156 Cα atoms, suggesting relative large variations between the two structures. Additionally, MMP1179 was not well aligned to our protein and thus removed from comparison. On the other hand, we found that the 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

human histamine N-methyltransferase (HNMT, PDB ID: 2AOT), which methylates histamine in the process of histamine metabolism, is structurally similar to YNL092W. Hence it was added to our list for comparison although it was not a hit from the Dali search. By overlaying these structures, we found that while the core structure of the Rossmann fold (the seven-strand sheet with flanking helices) is preserved among these proteins, and large variations were mainly found at the regions outside the core domain, such as insertions between the strands or additions at the N-termini (Fig. 5a). The most evident example is YNL092W, which possesses the long characteristic N-terminal helices. These additions may be involved in protein-substrate interactions and expand substrate recognition capability for more diverse substrates. Besides the conservation of the Rossmann fold, the binding modes of the SAM/SAH cofactors are conserved as well. Although the structures of the mammalian homologs of YNL092W (mouse or human) are not available, sequence alignment indicated that the three structures are highly similar. The N-terminal extensions (~60-residue, in human and ~100-residue in chicken) are very likely to be unstructured as predicted by the Xtalpred server.12 These regions are rich in glycine, alanine as well as polar residues. Despite the variable αN domains, the rest parts of these proteins share sequence identities as high as 65.3% and form the conserved SAM-binding domain (Fig. 5b). In addition, the key residues involved in substrate recognition are highly conserved, indicating that eukaryotic YNL092Ws share a similar catalytic mechanism. DISCUSSION. We present for the first time the complete methyltransferase structure containing the novel N-2227 fold, which serves as a SAM-binding domain. Most small-molecule methyltransferases only have additions at the N-termini, typically formed by two α-helices.10 12


Page 12 of 29

Page 13 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


YNL092W fits this description, with the four-helix αN domain at the N-terminus. Additionally, YNL092W displays moderate activity toward carnosine and is thus considered as a carnosine N-methyltransferase. Interestingly, the HMNT protein, which transfers a methyl group to the imidazole ring of histamine, aligned well with YNL092W structurally, although HNMT (-like) proteins are not apparently related to the YNL092W enzyme in sequence. Therefore, reptilian-avian HNMT-like, vertebrate HNMT and yeast YNL092W not only share clear structural similarities, but also functional similarities as they are all capable of catalyzing the N-methylation of the imidazole ring from small molecules.6 However, the true physiological functions of YNL092W remain mysterious. Despite the ubiquitous presence of UPF0586 in eukaryotic and even certain prokaryotic organisms,1 many of these species lack the ability to synthesize carnosine. Moreover, the absence of carnosine in yeasts also argues against its function as the carnosine N-methyltransferase. Therefore Drozak et al. predicted that the substrate for UPF0586 might be a protein, which could be methylated on a histidine residue at its C-terminus.7 Through activity assays, Szczepińska et al. demonstrated that YNL092W was indeed a protein methyltransferase capable of automethylation.9 These findings add to the interests of the problem. In order to understand the structure-activity relationship of YNL092W, we determined its crystal structure bound with SAH. Surface analyses of the enzyme reveal a deep groove at the dimer interface, and the SAM-binding sites flank the groove. From a structural point of view, large substrates like proteins could be well accommodated in the groove. However, the identification of the authentic substrates is needed to reveal the true function of this protein. METHODS Cloning, expression and protein purification. The full-length gene encoding (GenBank accession 13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

no. NP_014307) was amplified by PCR from the cDNA of the Saccharomyces cerevisiae using the primers

5’-AATAGATTCATATGGACGAGAATGAATTTGAT-3’ and

5’-TTGCACTTAAGCTT

TCATTATGATTCATTGGTGGGGTT-3’. After double digestion by the restriction enzymes NdeI and HindⅢ, the PCR product was inserted into a modified pET-28a (+) vector. The mutations were created through Quikchange PCRs based on the WT enzyme. The expression plasmids were transformed into Escherichia coli strain BL21 (DE3) cells for overexpression of the target protein. The transformed cells were cultured overnight in Luria-Bertani broth containing 30 mg L-1 kanamycin at 37°C. A 2-L fresh culture medium was inoculated with 20 mL of the overnight culture. When A600 reached 0.8 at 37°C, the expression of YNL092W was induced by 0.2 mM isopropyl β-D-thiogalactopyranoside (IPTG) and was kept shaking overnight at 25°C. The Escherichia coli cells were then pelleted by centrifugation at 4, 000 rpm for 20 min and resuspended in pre-chilled nickel-nitrilotriacetic acid (Ni-NTA) buffer A containing 20 mM Tris-HCl pH 8.0, 250 mM NaCl, 10 mM imidazole, 1 mM β-mercaptoethanol and 1 mM phenylmethylsulfonyl fluoride (PMSF). The resuspended cells were disrupted by ultrasonication and the supernatant was obtained by centrifugation at 14, 000 rpm for 1 h at 4°C. The supernatant was then applied onto Ni-NTA affinity resin (Qiagen) pre-equilibrated with Ni-NTA buffer A. The target protein was eluted with Ni-NTA buffer B containing 20 mM Tris-HCl (pH 8.0), 250 mM NaCl, 250 mM imidazole, 1 mM β-mercaptoethanol and 1 mM PMSF. The fractions were pooled and dialyzed in a buffer consisting of 20 mM Tris-HCl (pH 8.0), 50 mM NaCl, and 1 mM DTT. Then the samples were loaded to a Hitrap Q HP column (GE Healthcare), and the protein of interests was collected. The (His)6-tag at the N-terminus was cleaved off by treating with the PreScission protease overnight in the presence of 14


Page 15 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


5 mM β-mercaptoethanol, and the reaction mix was subsequently applied onto a Histrap column (GE Healthcare) to remove uncut protein. The unbound portion was pooled and dialyzed to a buffer containing 20 mM Tris-HCl (pH 8.0), 200 mM NaCl and 1 mM DTT for two hours. The protein was further concentrated to 2.5 mg ml-1, then flash frozen in liquid nitrogen and stored at -80°C. For proteins used in activity assays, the wild type and mutants proteins were dialyzed against a buffer containing 20 mM HEPES (pH = 7.5), 10 mM KCl, 1 mM DTT, 1 mM EDTA, and 1 mM MgCl2 for two hours. The protein was further concentrated to 0.5 mg ml-1, then flash frozen in liquid nitrogen and stored at -80°C. Crystallization and structure determination. The cocrystals with SAH were obtained using the sitting-drop vapor diffusion method. YNL092W and SAH were mixed in a molar ratio 1:50 to allow the formation of a complex and the final concentration of YNL092W was 2.5 mg ml-1. The complex was then mixed with equal volume of the reservoir solution at 25°C. Crystals were obtained in a condition of 30-35% PEG 400, 0.1 M NaOAc pH5.5, and 0.1 M NaCl. The fully-grown crystals were soaked for 1-3 min in a cryoprotective solution containing all the components of the reservoir solution supplemented 20% glycerol (v/v). For phase determination, 0.5 M NaI was added to the cryoprotective solution, where the crystals were soaked for 15 min. The soaked crystals were mounted on nylon loops and flash frozen in liquid nitrogen. The native data were collected from frozen crystals at -173 °C on ADSC CCD detectors using Beamline 17U1 (BL17U1)13 at the Shanghai Synchrotron Radiation Facility (SSRF, Shanghai, P.R. China), using a wavelength 0.9785 Å and diffracted to 2.2 Å. The iodide-soaked data were collected from frozen crystals at -173°C on a Pilatus 6M CCD detector using Beamline 19U1 (BL19U1) using a wavelength 1.4333 Å, and diffracted to 2.3 Å. The data were 15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

processed with the program HKL3000 14 and the space groups of the native crystals or NaI-derivatives belonged to P212121. The structure was solved by single-wavelength anomalous dispersion using the program Crank2.15 The mean figure of merit after the substructure determination was 0.22. The phase was further improved by density modification, increasing the mean figure of merit to 0.58. The automated model-building by Buccaneer finished with 790 residues built into the model.16 There were two molecules in the asymmetric unit, and the model was further built manually according to the electron density map with COOT.17 The high-resolution native data (2.2 Å) were merged in during the refinement process. Multiple cycles of refinement alternating with model rebuilding was carried out by PHENIX. Refine.18 The final R-factor was 20.4% (Rfree = 24.3%) (Table 2). The final model geometry was analyzed by MolProbity.19 The Ramachandran plot of the final model has 99.2%, 0.80% and 0% of the residues in the most favorable, generously allowed and disallowed region. The structural figures were produced with PyMOL (www.pymol.org). The topology of the was generated by the program Pro-origami.20 The docking models were generated by Autodock vina. 21 Isothermal titration calorimetry (ITC) assays. ITC experiments for the binding of SAM to YNL092W were performed at 25 °C using the PEAQ iTC titration calorimeter (MicroCal). 40-µl SAM (150-200 µM) was dissolved in the buffer containing 20 mM Tris-HCl, pH 8.0, and 150 mM NaCl, and titrated against 30-40 µM wild-type or mutant YNL092W present in the same buffer. The first injection (0.5-µl) was followed by 19 injections of 2-µl drops. The background values were subtracted from the experimental readings before data analyses. The MicroCal ORIGIN software was used to determine the site-binding models that produced good fits. Individual peaks from titrations were integrated and displayed in a Wiseman plot. The first drop was omitted from the calculation. 16


Page 16 of 29

Page 17 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Modeling studies on the binding mode of carnosine. The dimer structure (PDB 5X62) was chosen as the receptor due to the possibility of carnosine binding to the dimer interface. The SAH ligand was not removed prior to running the program. Protein preparation and subsequent molecular docking were performed using AutoDock Vina 1.5.6 with the application’s default settings. The carnosine coordinates were obtained from PubChem. The protein and carnosine structures were saved in the .pdbqt (Protein Data Bank, Partial Charge) format after their parameters were set up. The search was performed in a rectangular box with dimensions 24 Å × 22 Å × 26 Å. The ligand and the receptor molecules were kept rigid during the experiments. The pairs with the lowest binding energy were considered to have the best docking conformations. Activity assays. Assay of carnosine N-methyltransferase activity was similar to that by Drozak et al.7 The standard incubation mixture (55 µl) contained 50 mM HEPES pH 7.5, 10 mM KCl, 1 mM EDTA, 1 mM MgCl2, 1 mM DTT, 20 mM carnosine, and 1 µM [1H +3H]-SAM (PerkinElmer, 82.7 Ci/mmol) with the ratio of 1H-SAM to 3H-SAM is 8:1. 70 nM YNL092W WT or mutants were added to start the reaction, which were carried out at 37 °C for designated periods. The reaction was stopped by the addition of 50 µl of the reaction mixture to 100 µl of ice-cold 10% (w/v) HClO4. The samples were diluted with 60 µl of H2O and centrifuged at 13,000 g for 10 min. 30-µl 3 M K2CO3 was added to the supernatant to neutralize the HClO4, and the precipitate was removed by centrifugation at 13,000 g for 10 min. The supernatant was diluted 5 times with 20 mM HEPES pH 7.5, 1 ml of which was applied to Dowex 50W-X4 columns (1 ml, Na+ form) pre-equilibrated with 10 ml buffer of 20 mM HEPES pH 7.5. The columns were washed with 10 ml of 20 mM HEPES pH 7.5, to remove minor radioactive contaminants. Anserine was eluted with a buffer containing 10 ml of 20 mM 17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

HEPES pH 7.5, 0.5 M NaCl. To elute the extra [1H +3H]-SAM, the columns were washed with 8 ml of 1 M NH3⋅H2O. 1.5ml eluate was mixed with 9 ml of scintillation fluid (PerkinElmer), and the incorporated radioactivity was analyzed with a Beckman LS6000 IC liquid scintillation counter. Accession code The atomic coordinates and structure factors have been deposited in the Protein Data Bank with the accession code 5X62. Acknowledgements We thank staff members from BL17U1 and BL19U1 beamlines (National Center for Protein Sciences Shanghai (NCPSS)) at Shanghai Synchrotron Radiation Facility, for assistance during data collection. Author contributions WX conceived and designed research; XL, JW and YS performed research; WX analyzed data and wrote the paper. All authors reviewed the results and approved the final version of the manuscript. Funding This work was supported by the Fundamental Research Funds for the Central Universities 16lgjc76, the Science and Technology Program of Guangzhou 201504010025 and 201605030012. Competing financial interests The authors declare no competing financial interests

18


Page 18 of 29

Page 19 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1. Data collection and refinement statistics. Crystals

Native

NaI derivative

Data collection

SSRF BL17U1

SSRF BL19U1

Wavelength

0.9785 Å

1.4333 Å

Space group

P212121

P212121

a, b, c (Å)

71.5, 72.9, 175.2

71.4, 73.0, 175.3

α, β, γ (°)

90, 90, 90

Resolution (Å) b

50-2.20 (2.28-2.20)

90, 90, 90 a

50-2.30 (2.38-2.30)

0.118 (1.052)

0.1 (0.677)

Rpim

0.048 (0.415)

0.029 (0.198)

I/σ(I)

16.5 (1.9)

23.36 (2.42)

Completeness (%)

99.9 (100.0)

99.9 (99.9)

Redundancy

7.2 (7.4)

12.9 (12.4)

Rmerge

Refinement Resolution (Å)

38.45-2.20 (2.25-2.20)

No. reflections

46985

Rworkc /Rfreed

0.204/0.243

No. atoms Protein

5872

Ligand

52 (SAH)/1(Na+)

Water

212

B-factors (Å 2) Protein

47.7

Ligand

36.3 (SAH)/39.3(Na+)

Water

46.8

R.m.s deviations Bond lengths (Å)

0.007

Bond angles (º)

0.91

Ramachandran favored (%)

99.21

Allowed

0.79

Outliers (%)

0.0

a

Values in parentheses are for the highest-resolution shell. bRmerge =Σ |(I - < I > )|/σ(I), where I is the observed intensity. cRwork = Σhkl ||Fo| -

|Fc||/ Σhkl |Fo|, calculated from working data set. dRfree is calculated from 5.0% of data randomly chosen and not included in refinement.

19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure legends Figure 1. The overall structure of the YNL092W-SAH complex. (a) A scheme showing the methyltransfer reaction catalyzed by YNL092W. The methyl group is colored red. (b) The structure shown in the ribbon rendition in two orthogonal views. The αN and the Rossmann domains are colored purple and green respectively for one of the subunits, and gray for the other. The N- and C-termini are indicated and the two SAH molecules are shown as spheres. The two disulfide bonds are shown as the ball-and-stick model and circled. (b) The topology organization of an YNL092W monomer, where the secondary elements are colored-coded as cold colors for the N-terminal domain and hot colors for the N-terminal domain. Figure 2. The interactions and recognition pattern of SAH with the enzyme. (a) The simulated-annealing omit Fo-Fc map is contoured at 3σ and the hydrogen bonds formed by SAH and the active site residues are shown by the red dashed lines (distance < 3.5 Å). (b) ITC measurements of the binding for WT and mutants to SAM. The dissociation constant (KD) and the binding stoichiometry values (N) are shown below the fitted curves. (c) The activity assays for the WT and mutants. Two time points were taken: 15- and 30-min. The WT activities at these two time points were normalized to 100% (the normalized 100% at 15-min corresponds to 8338.5 cpm, while the normalized 100% at 30-min corresponds to 14471.5 cpm), and the activities of all other mutants were represented in percentages relative to those of WT at their time points respectively. Values are the Means ± range of at least two separate experiments. Figure 3. The proposed binding mode of carnosine to YNL092W. (a)-(c) The top three binding models of carnosine to YNL092W, generated by Autodock. The ligands are depicted in the ball-and-stick model. The coloring scheme is the same as in Fig. 1a. The distances between the N 20


Page 20 of 29

Page 21 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(from carnosine) and S atoms (from SAH) are indicated by the blue dashed lines and the values are shown above the lines. SAH and carnosine are colored in magenta and yellow respectively. Units are in angstroms. (d) The activity assays for the WT and mutants, as performed in Fig. 2c. Figure 4. The influence of the disulfide bonds. (a) The locations of the C231-C238, and the C123-C369 disulfide bonds. The β9-α5 loop is colored orange. The SAH ligand is depicted in the ball-and-stick model and colored magenta. The hydrogen bonds formed by SAH and surrounding residues are shown by the red dashed lines (distance < 3.5 Å). (b) The activity assays on the C123A and C238A mutants. Figure 5. Structure comparison to homologs. (a) The structural comparison of the SAH-bound YNL092W (color scheme as in Fig. 1a, PDB ID: 5X62) with PH0226 protein from Pyrococcus horikoshii OT3 (yellow, PDB ID: 1VE3), Phosphoethanolamine N-methyl transferase from Plasmodium knowlesi (red, PDB ID: 4IV8), Bacteroides vulgatus methyltransferase BVU_3255 (cyan, PDB ID: 3T7R, no ligand bound), and human HNMT (orange, PDB ID: 2AOT). The lower right panel shows the close-up view of the superposition of the SAH/SAM ligands depicted in ball-and-stick models. (b) Multiple sequence alignment of YNL092W and the homologs. The secondary structure elements of YNL092W are labeled above the sequences. The red dots indicate residues responsible for SAH binding and the blue dots indicate residues responsible for carnosine binding in Model III.

21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References (1) (2)

(3) (4) (5) (6)

(7)

(8)

(9)

(10)

(11) (12)

(13)

(14)

(15) (16) (17)

Crush, K. G. (1970) Carnosine and related substances in animal tissues. Comp. Biochem. Physiol. 34, 3-30. Mannion, A. F., Jakeman, P. M., Dunnett, M., Harris, R. C., and Willan, P. L. (1992) Carnosine and anserine concentrations in the quadriceps femoris muscle of healthy humans. Eur. J. Appl. Physiol. Occup. Physiol.64, 47-50. Boldyrev, A. A., Aldini, G., and Derave, W. (2013) Physiology and pathophysiology of carnosine. Physiol. Rev. 93, 1803-1845. Boldyrev, A. A. (2007) Carnosine and oxidative stress in cells and tissues. pp. 37–38, Nova Scientific Publisher, New York. Szwergold, B. S. (2005) Carnosine and anserine act as effective transglycating agents in decomposition of aldose-derived Schiff bases. Biochem. Biophys. Res. Commun. 336, 36-41. Drozak, J., Chrobok, L., Poleszak, O., Jagielski, A. K., and Derlacz, R. (2013) Molecular identification of carnosine N-methyltransferase as chicken histamine N-methyltransferase-like protein (hnmt-like). PloS one 8, e64805. Drozak, J., Piecuch, M., Poleszak, O., Kozlowski, P., Chrobok, L., Baelde, H. J., and de Heer, E. (2015) UPF0586 Protein C9orf41 Homolog Is Anserine-producing Methyltransferase. J. Biol. Chem. 290, 17190-17205. Uhlen, M., Oksvold, P., Fagerberg, L., Lundberg, E., Jonasson, K., Forsberg, M., Zwahlen, M., Kampf, C., Wester, K., Hober, S., Wernerus, H., Bjorling, L., and Ponten, F. (2010) Towards a knowledge-based Human Protein Atlas, . Nat. Biotechnol.28, 1248-1250. Szczepinska, T., Kutner, J., Kopczynski, M., Pawlowski, K., Dziembowski, A., Kudlicki, A., Ginalski, K., and Rowicka, M. (2014) Probabilistic approach to predicting substrate specificity of methyltransferases, PLoS Comput. Biol.10, e1003514. Martin, J. L., and McMillan, F. M. (2002) SAM (dependent) I AM: the S-adenosylmethionine-dependent methyltransferase fold. Curr. Opin. Struct. Biol. 12, 783-793. Holm, L., and Sander, C. (1995) 3-D lookup: fast protein structure database searches at 90% reliability. Proceedings. Proc. Int. Conf. Intell. Syst. Mol. Biol., 5th. 3, 179-187. Slabinski, L., Jaroszewski, L., Rychlewski, L., Wilson, I. A., Lesley, S. A., and Godzik, A. (2007) XtalPred: a web server for prediction of protein crystallizability. Bioinformatics. 23, 3403-3405. Wang, Q. S., Yu, F., Huang, S., Sun, B., Zhang, K. H., Liu, K., Wang, Z. J., Xu, C. Y., Wang, S. S., Yang, L. F., Pan, Q. Y., Li, L., Zhou, H., Cui, Y., Xu, Q., Earnest, T., and He, J. H. (2015) The macromolecular crystallography beamline of SSRF, Nucl. Sci. Tech. 26, 12-17. Minor, W., Cymborowski, M., Otwinowski, Z., and Chruszcz, M. (2006) HKL-3000: the integration of data reduction and structure solution--from diffraction images to an initial model in minutes. Acta Crystallogr. D Biol. Crystallogr.62, 859-866. Skubak, P., and Pannu, N. S. (2013) Automatic protein structure solution from weak X-ray data. Nat. Commun. 4, 2777. Cowtan, K. (2006) The Buccaneer software for automated model building. 1. Tracing protein chains. Acta Crystallogr. D Biol. Crystallogr. 62, 1002-1011. Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and development of 22


Page 22 of 29

Page 23 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(18)

(19)

(20) (21)

Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486-501. Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart, P. H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213-221. Chen, V.B., Arendall,W. B. 3rd., Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G. J., Murray, L.W., Richardson, J.S., Richardson, D. C. (2010) MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D Biol. Crystallogr. 66, 12-21. Stivala, A., Wybrow, M., Wirth, A., Whisstock, J. C., and Stuckey, P. J. (2011) Automatic generation of protein structure cartoons with Pro-origami. Bioinformatics.27, 3315-3316. Sandeep, G., Nagasree, K. P., Hanisha, M., and Kumar, M. M. (2011) AUDocker LE: A GUI for virtual screening with AUTODOCK Vina. BMC Res. Notes. 4, 445.

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. The overall structure of the YNL092W-SAH complex. (a) A scheme showing the methyltransfer reaction catalyzed by YNL092W. The methyl group is colored red. (b) The structure shown in the ribbon rendition in two orthogonal views. The αN and the Rossmann domains are colored purple and green respectively for one of the subunits, and gray for the other. The N- and C-termini are indicated and the two SAH molecules are shown as spheres. The two disulfide bonds are shown as the ball-and-stick model and circled. (b) The topology organization of an YNL092W monomer, where the secondary elements are coloredcoded as cold colors for the N-terminal domain and hot colors for the N-terminal domain. 127x190mm (300 x 300 DPI)


Page 24 of 29

Page 25 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. The interactions and recognition pattern of SAH with the enzyme. (a) The simulated-annealing omit Fo-Fc map is contoured at 3σ and the hydrogen bonds formed by SAH and the active site residues are shown by the red dashed lines (distance < 3.5 Å). (b) ITC measurements of the binding for WT and mutants to SAM. The dissociation constant (KD) and the binding stoichiometry values (N) are shown below the fitted curves. (c) The activity assays for the WT and mutants. Two time points were taken: 15- and 30-min. The WT activities at these two time points were normalized to 100% (the normalized 100% at 15-min corresponds to 8338.5 cpm, while the normalized 100% at 30-min corresponds to 14471.5 cpm), and the activities of all other mutants were represented in percentages relative to those of WT. Values are the Means ± range of at least two separate experiments. 140x135mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. The proposed binding mode of carnosine to YNL092W. (a)-(c) The top three binding models of carnosine to YNL092W, generated by Autodock. The ligands are depicted in the ball-and-stick model. The coloring scheme is the same as in Fig. 1a. The distances between the N (from carnosine) and S atoms (from SAH) are indicated by the blue dashed lines and the values are above the lines. SAH and carnosine are colored in magenta and yellow respectively. Units are in angstroms. (d) The activity assays for the WT and mutants, as performed in Fig. 2c. 95x79mm (300 x 300 DPI)


Page 26 of 29

Page 27 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. The influence of the disulfide bonds. (a) The locations of the C231-C238, and the C123-C369 disulfide bonds. The β9-α5 loop is colored orange. The SAH ligand is depicted in the ball-and-stick model and colored magenta. The hydrogen bonds formed by SAH and surrounding residues are shown by the red dashed lines (distance < 3.5 Å). (b) The activity assays on the C123A and C238A mutants. 93x32mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. Structure comparison to homologs. (a) The structural comparison of the SAH-bound YNL092W (color scheme as in Fig. 1a, PDB ID: 5X62) with PH0226 protein from Pyrococcus horikoshii OT3 (yellow, PDB ID: 1VE3), Phosphoethanolamine N-methyl transferase from Plasmodium knowlesi (red, PDB ID: 4IV8), Bacteroides vulgatus methyltransferase BVU_3255 (cyan, PDB ID: 3T7R, no ligand bound), and human HNMT (orange, PDB ID: 2AOT). The lower right panel shows the close-up view of the superposition of the SAH/SAM ligands depicted in ball-and-stick models. (b) Multiple sequence alignment of YNL092W and the homologs. The secondary structure elements of YNL092W are labeled above the sequences. The red dots indicate residues responsible for SAH binding and the blue dots indicate residues responsible for carnosine binding in Model III. 127x135mm (300 x 300 DPI)


Page 28 of 29

Page 29 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphic abstract 68x43mm (300 x 300 DPI)


Substrate Recognition Mechanism of the Putative ... - ACS Publications

Recommend Documents