Communication pubs.acs.org/JACS
A Glycopeptidyl-Glutamate Epimerase for Bacterial Peptidoglycan Biosynthesis Ruoyin Feng,† Yasuharu Satoh,‡ Yasushi Ogasawara,‡ Tohru Yoshimura,§ and Tohru Dairi*,‡ †
Graduate School of Chemical Sciences and Engineering, Hokkaido University, N13 & W8, Kita-ku, Sapporo, Hokkaido 060-8628, Japan ‡ Graduate School of Engineering, Hokkaido University, N13 & W8, Kita-ku, Sapporo, Hokkaido 060-8628, Japan § Graduate School of Bioagricultural Sciences, Nagoya University, Furou-chou, Chikusa-ku, Nagoya, Aichi 464-8601, Japan S Supporting Information *
revealed that Xanthomonas oryzae has no orthologues of Glu racemase or D-Glu aminotransferase, even though it possesses the other Mur orthologues that possibly participate in peptidoglycan biosynthesis, suggesting that X. oryzae biosynthesizes D-Glu via an alternative enzyme(s). Here we investigated the mechanism by shotgun cloning with a D-Glu auxotrophic Escherichia coli mutant and in vitro studies with recombinant enzymes. To search for the gene(s) for D-Glu biosynthesis, we carried out a shotgun cloning experiment with D-Glu auxotrophic E. coli WM335 and genomic DNA of X. oryzae MAFF311018 as the donor DNA. Consequently, we obtained dozens of colonies on LB plates without D-Glu. The transformants were randomly selected, and the sequences of the inserted DNAs were determined. All of the inserts had a common region containing four genes, XOO_1318 to XOO_1321 (Figure S1a).6 Subcloning experiments showed that two genes, XOO_1319 and XOO_1320, which are annotated as a function of unknown protein and MurD, respectively, were essential and that neither XOO_1319 nor XOO_1320 complemented the D-Glu auxotrophy (Figure S1b). Since XOO_1320 has similarity to MurD (Table S1) and possibly ligates D-Glu to UDP-MurNAc-L-Ala (1), XOO_1319 was suggested to be a Glu racemase. To examine this possibility, we performed in vitro experiments with recombinant XOO_1319 (Figure S2a) and L-Glu as a substrate. Chiral analysis was performed by Marfey’s method using 1-fluoro-2,4dinitrophenyl-5-L-alanine amide (L-FDAA).7 However, the enzyme showed no racemase activity even with reaction mixtures containing pyridoxal-5′-phosphate (PLP), which is an essential cofactor for typical amino acid racemases (Figure S3). We were also unable to detect racemase activity with both XOO_1319 and XOO_1320 (Figure S3). We then investigated whether XOO_1320 had the predicted UDP-MurNAc-L-Ala-D-Glu synthetase activity. Recombinant XOO_1320 (Figure S2b) was incubated with enzymatically prepared 1 (Figures S4−S6)8 and D-Glu in the presence of ATP and Mg2+. However, extremely small amounts of the product were detected only when large amounts of recombinant XOO_1320 (50 μg mL−1, 0.96 μM) were used with incubation for 2 h (Figures 1b and S7a), in contrast to MurD recombinant
ABSTRACT: D-Glutamate (Glu) supplied by Glu racemases or D-amino acid transaminase is utilized for peptidoglycan biosynthesis in microorganisms. Comparative genomics has shown that some microorganisms, including Xanthomonas oryzae, perhaps have no orthologues of these genes. We performed shotgun cloning experiments with a D-Glu auxotrophic Escherichia coli mutant as the host and X. oryzae as the DNA donor. We obtained complementary genes, XOO_1319 and XOO_1320, which are annotated as a hypothetical protein and MurD (UDP-MurNAc-L-Ala-D-Glu synthetase), respectively. By detailed in vitro analysis, we revealed that XOO_1320 is an enzyme to ligate L-Glu to UDPMurNAc-L-Ala, providing the first example of MurD utilizing L-Glu, and that XOO_1319 is a novel enzyme catalyzing epimerization of the terminal L-Glu of the product in the presence of ATP and Mg2+. We investigated the occurrence of XOO_1319 orthologues and found that it exists in some categories of microorganisms, including pathogenic ones.
A
lmost all eubacteria possess peptidoglycan to maintain cell integrity. Its biosynthetic pathway and mechanism have been well-studied because the biosynthetic enzymes are primary targets of chemotherapeutic reagents for which very useful antibiotics such as β-lactams, glycopeptides, and fosfomycin have been successfully developed.1−3 UDP-MurNAc-pentapeptide, a unit of peptidoglycan, is synthesized by nine enzymes. Briefly, UDP-N-acetylglucosamine (UDP-GlcNAc) is formed from fructose-6-phospate by GlmS, GlmM, and GlmU. Then UDP-GlcNAc is converted to UDP-MurNAc by MurA and MurB, followed by successive addition of L-Ala, D-Glu, mesodiaminopimelate (or L -Lys), and D-Ala- D -Ala, by four structurally similar enzymes, MurC to MurF (Figure 1a). During this process, D-Glu is usually supplied by Glu racemases (MurI). In some microorganisms, D-amino acid aminotransferase supplies D-Glu by transamination from D-Ala to αketoglutarate.4 To the best of our knowledge, the D-Glu formation mechanism is limited to these two enzymes in microorganisms. We have been searching for an alternative biosynthetic pathway in the primary metabolism and found the futalosine pathway for menaquinone biosynthesis.5 In that study, we © XXXX American Chemical Society
Received: February 6, 2017 Published: March 15, 2017 A
DOI: 10.1021/jacs.7b01221 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society
together, we hypothesized that XOO_1320 and XOO_1319 are a UDP-MurNAc-L-Ala-L-Glu synthetase and an L-Glu epimerase of the product (UDP-MurNAc-L-Ala-L-Glu epimerase), respectively. To confirm this hypothesis, recombinant XOO_1320 was incubated with 1 and L-Glu under the same conditions as described above. By LC−MS analysis, we detected large amounts of a new product (100 times that produced with DGlu; Figures 1b and S7b) that had the same mass spectrum as 2 but eluted at a different time (Figures 1b and S10a). The chirality of the Glu moiety of the product was then analyzed by the same method as described above. As shown in Figure 2, the reaction product included only L-Glu, clearly showing that XOO_1320 was a MurD utilizing L-Glu as the main substrate.
Figure 1. Peptidoglycan biosynthesis by MurD/MurI or XOO_1319/ XOO_1320. (a) Typical and newly identified peptidoglycan biosynthetic pathways, catalyzed by MurD/MurI and XOO_1319/ XOO_1320, are shown. (b) UDP-MurNAc-L-Ala-L/D-Glu synthetase activities. E. coli MurD (top), XOO_1320 (middle), and a mixture of XOO_1320 and XOO_1319 (bottom) were incubated with UDPMurNAc-L-Ala (1) and D-Glu (left column) or L-Glu (right column). HPLC traces of the reaction products are shown. (c) UDP-MurNAc-LAla-L-Glu epimerase activity of recombinant XOO_1319. Boiled (top) or active enzyme (second) was used for the in vitro assay. HPLC traces of the reaction products are shown. Enzymatically prepared UDP-MurNAc-L-Ala-D-Glu (2) (third) and UDP-MurNAc-L-Ala-L-Glu (3) (bottom) were also analyzed.
Figure 2. Chiral analysis of the terminal Glu residue of the reaction products. L-FDAA derivatives of Glu released by acid hydrolysis of the reaction product formed from UDP-MurNAc-L-Ala (1) and L-Glu by recombinant XOO_1320 (top) and from UDP-MurNAc-L-Ala-L-Glu (3) by recombinant XOO_1319 (second) were analyzed by HPLC with UV detection at 340 nm. The standards D-Glu (third) and L-Glu (bottom) were also analyzed.
enzyme of E. coli, which completely converted 1 to UDPMurNAc-L-Ala-D-Glu (2) within a 2 h incubation (Figure 1b). As mentioned above, both XOO_1319 and XOO_1320 were shown to be essential by the genetic complementation experiment. We therefore incubated both recombinant enzymes with 1 and L-Glu under the same conditions as described above. By LC−MS analysis, we surprisingly and unexpectedly detected large amounts of a new product that had the same mass spectrum and elution time as 2 formed by E. coli MurD with 1 and D-Glu as the substrates (Figures 1b and S8). In contrast, no products were formed when D-Glu was used as the substrate (Figure 1b). We therefore examined the chirality of the terminal Glu moiety of the product by a modified Marfey’s method.9 The product purified with HPLC was hydrolyzed under acidic conditions, and the released amino acid was reacted with 1-fluoro-2,4-dinitrophenyl-5-L-leucine amide (LFDLA) and analyzed by LC−MS. As shown in Figure S9, the reaction product contained only D-Glu. Taking these results
We next examined whether XOO_1319 had UDP-MurNAcepimerase activity. Recombinant XOO_1319 was incubated with enzymatically synthesized and purified UDPMurNAc-L-Ala-L-Glu (3) (Figures S11 and S12) under various conditions, and the chirality of the terminal Glu of the product was examined. Consequently, we found that the terminal L-Glu of the substrate was converted into D-Glu in the presence of Mg2+ and ATP (Figures 1c, 2, S10b, and S13), demonstrating that XOO_1319 is a novel type of glycopeptidyl-glutamate epimerase. Then the formation of AMP or ADP after the reaction was investigated to understand the substrate activation mechanism. As shown in Figure S14, we detected only AMP, suggesting that the substrate was activated by adenylation. We L-Ala-L-Glu
B
DOI: 10.1021/jacs.7b01221 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Journal of the American Chemical Society
■
ACKNOWLEDGMENTS We are grateful to Professor Walter Messer of the Max-Planck Institute for Molecular Genetics for providing E. coli WM335. We also thank Dr. Eri Fukushi of the GCMS and NMR Laboratory, Graduate School of Agriculture, Hokkaido University, for measuring NMR spectra. This work was supported by Grants-in-Aid for Scientific Research on Innovative Areas from MEXT, Japan (JSPS KAKENHI Grant 16H06452) and Grants-in-Aid for Scientific Research from JSPS (15H03110) to T.D.
also examined whether XOO_1319 catalyzed the reverse reaction with 2 as the substrate. As shown in Figure S15, however, no epimerase activity was detected, showing that only 3 can be activated by the enzyme with ATP. Most orthologues of XOO_1319 and XOO_1320 exist next to each other in the genome. XOO_1320 has relatively high identities (30−89% amino acid identity; Table S2) to orthologues found in microorganisms possessing XOO_1319 but shows low identity to other MurD enzymes, including that of E. coli (less than 30% amino acid identity; Table S1). Since the crystal structure of E. coli MurD has been solved and the catalytic mechanism is well-established,10−13 the crystal structure of XOO_1320 and comparison with that of E. coli MurD would enable us to determine how XOO_1320 recognizes the isomer. XOO_1319 is composed of 451 amino acids and has no putative conserved domains or cofactor binding domains. We are therefore unable to estimate the exact reaction mechanism of the enzyme at this stage. The reaction might start with adenylation of the α-carboxyl group, followed by epimerization at Cα and hydrolysis to give the final product (Figure S16). This reaction mechanism is consistent with the observation that the reaction is irreversible. A crystallographic structure might provide clues even in this case. We investigated the occurrence of XOO_1319 orthologues among bacteria (Table S2 and Figure S17) and found that it exists in Gammaproteobacteria including the genera Stenotrophomonas, Dyella, Frateuria, Rhodanobacter, Pseudoxanthomonas, and Lysobacter besides Xanthomonas and Xylella. Several rare actinobacteria such as the genera Micromonospora, Actinoplanes, Verrucosispora, and Salinispora also possess orthologues. Moreover, we detected orthologues in a few strains belonging to Alphaproteobacteria, including the genera Devosia, Pelagibacterium, and Parvularcula. Orthologues with low similarity are found in a plant pathogenic fungus (Phytophthora sojae), cyanobacteria (Anabaena sp. and Cylindrospermum stagnale), and an amoeba (Acanthamoeba castellanii). Of these, recombinant enzymes of orthologues found in rare actinobacteria, a Micromonospora strain and a Salinispora strain, and in Stenotrophomonas maltophilia known to cause nosocomial infections were prepared and confirmed to have the same activities as XOO_1319 and XOO_1320 (Figures S18−22). On the basis of the results, we propose that XOO_1319 and XOO_1320 should be designated as UDP-N-acetylmuramic acid (UDP-MurNAc)-L-Ala-L-Glu epimerase (MurL) and UDPMurNAc-L-Ala-L-Glu synthetase (MurD2), respectively.
■
Communication
■
REFERENCES
(1) van Heijenoort, J. Nat. Prod. Rep. 2001, 18, 503. (2) El Zoeiby, A.; Sanschagrin, F.; Levesque, R. C. Mol. Microbiol. 2003, 47, 1. (3) Moraes, G. L.; Gomes, G. C.; Monteiro de Sousa, P. R.; Alves, C. N.; Govender, T.; Kruger, H. G.; Maguire, G. E. M.; Lamichhane, G.; Lameira, J. Tuberculosis 2015, 95, 95. (4) Thorne, C. B.; Gòmez, C. G.; Housewright, R. D. J. Bacteriol. 1955, 69, 357. (5) Hiratsuka, T.; Furihata, K.; Ishikawa, J.; Yamashita, H.; Itoh, N.; Seto, H.; Dairi, T. Science 2008, 321, 1670. (6) Ochiai, H.; Inoue, Y.; Takeya, M.; Sasaki, A.; Kaku, H. JARQ 2005, 39, 275. (7) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591. (8) Raymond, J. B.; Price, N. P.; Pavelka, M. S., Jr FEMS Microbiol. Lett. 2003, 229, 83. (9) Fujii, K.; Ikai, Y.; Oka, H.; Suzuki, M.; Harada, K. Anal. Chem. 1997, 69, 5146. (10) Bertrand, J. A.; Auger, G.; Fanchon, E.; Martin, L.; Blanot, D.; van Heijenoort, J.; Dideberg, O. EMBO J. 1997, 16, 3416. (11) Bertrand, J. A.; Auger, G.; Martin, L.; Fanchon, E.; Blanot, D.; Le Beller, D.; van Heijenoort, J.; Dideberg, O. J. Mol. Biol. 1999, 289, 579. (12) Saio, T.; Ogura, K.; Kumeta, H.; Kobashigawa, Y.; Shimizu, K.; Yokochi, M.; Kodama, K.; Yamaguchi, H.; Tsujishita, H.; Inagaki, F. Sci. Rep. 2015, 5, 16685. (13) Šink, R.; Kotnik, M.; Zega, A.; Barreteau, H.; Gobec, S.; Blanot, D.; Dessen, A.; Contreras-Martel, C. PLoS One 2016, 11, e0152075.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b01221. Experimental details and additional data (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*
[email protected] ORCID
Yasuharu Satoh: 0000-0001-6671-7758 Tohru Dairi: 0000-0002-3406-7970 Notes
The authors declare no competing financial interest. C
DOI: 10.1021/jacs.7b01221 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
