CHEMBIOCHEM COMMUNICATIONS DOI: 10.1002/cbic.201300602
Rapid Determination of the Amino Acid Configuration of Xenotetrapeptide Carsten Kegler, Friederike I. Nollmann, Tilman Ahrendt, Florian Fleischhacker, Edna Bode, and Helge B. Bode*[a] An E. coli strain with deletions in five transaminases (DaspC DilvE DtyrB DavtA DybfQ) was constructed to be unable to degrade several amino acids. This strain was used as an expression host for the analysis of the amino acid configuration of nonribosomally synthesized peptides, including the novel peptide “xenotetrapeptide” from Xenorhabdus nematophila, by using a combination of labeling experiments and mass spectrometry. Additionally, the number of d-amino acids in the produced peptide was assigned following simple cultivation of the expression strain in D2O.
Introduction Natural-product research has been the source of several important drugs over the last 60 years; natural products or derivatives thereof are major therapeutics, particularly as anti-infectives.[1–3] As drug discovery in general is very costly and little revenue can be expected from this kind of research (especially in the anti-infectives field), pharmaceutical companies have reduced their efforts in this area in favor of developing treatments for chronic or “more serious” diseases like cancer. The decline in industrial natural-product research has not been followed in academic research, which has started to revitalize the field by using modern methods of bioinformatics,[4] analytical chemistry,[5–7] and molecular biology.[8–10] The combination of these approaches allows rapid identification and access to novel natural products, including those that are not produced under standard conditions. Additionally, compounds that are naturally produced in only minute amounts (insufficient for isolation, structure elucidation, and bioactivity testing) can nowadays been identified quite easily. Especially important is the more recent adoption of mass spectrometry (MS) as a routine and highly sensitive tool to increase the number of natural products that can be identified in a single organism.[6, 7] However, the advantage of a high sensitivity might prove disadvantageous, as natural products identified by MS methods might be difficult to isolate because of their low production levels. Fortunately, progress in organic synthesis allows alternative access
to these compounds, if they can be synthesized with reasonable effort. However, in order to facilitate their synthesis, the absolute configurations of the desired natural products must be known. Peptide natural products can consist of l- and d-amino acids: epimerization (E)[11] or dual condensation/epimerization (C/E)[12] domains in nonribosomal peptide synthetases (NRPS) convert incorporated l-amino acids into d-amino acids of the final product.[13] Additionally, it has been shown that d-amino acids can be directly incorporated by NRPSs.[14] The recent identification of the proteusin family of natural products has shown that d-amino acids can also be found in ribosomally derived peptides using a so far unknown mechanism catalyzed by enzymes belonging to the radical S-adenosylmethionine (rSAM) family.[15] The presence of d-amino acids makes peptides more stable against proteases, and can result in a different conformation of the peptide chain.[16] Usually the amino acid configuration is determined by peptide hydrolysis or from the NMR-determined 3D structure, however, both methods require the isolation of the desired peptide in sufficient amounts (often difficult, as already stated). For these compounds a real need for the determination (and confirmation) of the amino acid configuration exists; stable isotope labeling (with loss of label due to epimerization) can be used to detect the configuration based on MS methods, as we have shown previously for peptides produced in Photorhabdus luminescens and Xenorhabdus nematophila.[17] However, the incorporation of [2-2H]-labeled amino acids allowing the rapid determination of epimerization (E) or rSAM activity is only possible in mutants with blocked amino transaminase (TA) activity, as otherwise the label is rapidly lost due to general TA activity.[17] Deuterium labels have also been used extensively to elucidate other biochemical pathways because of their simple detection by MS.[18, 19] Here we describe an Escherichia coli strain with deletions in five TAs, and its use as expression host for the determination of the peptide configurations of the known GameXPeptide A (1)[17] and a novel peptide from X. nematophila, “xenotetrapeptide” (2). Such a strain is especially useful, as heterolo-
[a] Dr. C. Kegler,+ F. I. Nollmann,+ T. Ahrendt, F. Fleischhacker, E. Bode, Prof. Dr. H. B. Bode Merck Stiftungsprofessur fr Molekulare Biotechnologie Fachbereich Biowissenschaften, Goethe Universitt Frankfurt Max-von-Laue-Strasse 9, 60438 Frankfurt am Main (Germany) E-mail: [email protected] [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cbic.201300602.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 826 – 828
826
CHEMBIOCHEM COMMUNICATIONS
www.chembiochem.org
gous expression of unknown biosynthesis gene clusters is now a widespread method in natural-product research.[20–22]
Results and Discussion As the major TAs in E. coli involved in amino acid metabolism are AspC, IlvE and TyrB,[23] the corresponding genes were deleted in E. coli DH10B. In order to test this strain for structure elucidation of natural products, the gene from P. luminescens encoding GxpS (the NRPS that produces GameXPeptide A, 1)[17] and an uncharacterized gene (XNC1_2022) encoding an NPRS from X. nematophila ATCC19061 were transformed into this strain. XNC1_2022 was shown to be involved in the production of a compound with m/z 411.3 [M+H] + and 433.3 [M+Na] + by the loss of production by an insertion mutant. Analysis of XNC1_2022 in detail revealed four modules, including two dual condensation/epimerization (C/E) domains (see Figure S1 in the Supporting Information) predicted to be involved in amino acid epimerization of valine-1 and 3, thereby resulting in the predicted structure: cyclo(vLvV). The predicted structure is in agreement with the formula C21H38N4O4 as determined by HR-ESIMS for X. nematophila and E. coli expressing XNC1_2022, and was confirmed by labeling experiments in 15N and 13C medium, thus allowing simple determination of the number of nitrogen and carbon atoms (Figure S2).[17] Thus the compound was named xenotetrapeptide (2), and the corresponding NRPS was renamed xenotetrapeptide synthetase (XtpS; Scheme 1). The correct configuration of 1 as cyclo(vLflL) was determined by using the addition of fully deuterated Leu, Val, and Phe in an E. coli triple mutant (deletions of aspC, tyrB, and ilvE) that expressed gxpS (Figure S3); the results were similar to those described in P. luminescens naturally producing 1.[17] For 2 the presence of l-Leu was shown by the expected 10 Da shift resulting from the incorporation of 2H10-Leu (data not shown). However, upon addition of 2H8-valine, the resulting mixture of isotopomers with mass shifts of 21 and 22 Da indicated either partial epimerization or (more likely) valine-specific TA activity from additional TAs (Figure 1 A). Thus TA-encoding
Scheme 1. Domain organization of XtpS involved in the biosynthesis of xenotetrapeptide 2 with bound biosynthesis intermediates. Domains, A: adenylation, T: thiolation, C/E: dual condensation/epimerization, C: condensation, TE: thioesterase.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. HPLC/MS data of xenotetrapeptide (2) from cultivation of E. coli wild-type (WT) or TA-deficient strains (Dtriple and Dpenta) expressing xtpS in A) LB medium, and B) M9 minimal medium with l-[2H10]leucine or l-[2H8]valine.
genes avtA and ybfQ[24, 25] were also deleted (E. coli Dpenta); this showed predominantly the expected mass shift of 22 Da upon expression of xtpS with addition of 2H8-valine, in agreement with the presence of one l-(2H8 label) and two d-valine moieties (2H7 label) in 1 (Figure 1 B). In order to test the predicted position of the d- and l-valines, 2 was synthesized by solid phase peptide synthesis; this showed retention time and fragmentation pattern identical to those of the natural compound (Figure S4). As some deuterated amino acids are either not available or very expensive, we tested whether the labeling approach can be reversed by growing E. coli and Dpenta in 100 % D2O with nondeuterated medium components: amino acid epimerization should result in a mass shift of +1 Da, so, for 1 and 2, mass shifts of 3 and 2 Da would be observed because of the presence of three and two d-amino acids in the final compounds (Figure 2). This was indeed the case. Although the results were not perfect (probably due to residual TA activity prior to amino acid incorporation catalyzed by the NRPS: see small peaks resulting from isotopomers with one deuterium less than expected), the significant difference between wildtype and Dpenta strains clearly shows that the method works (Figure 2). We observed slightly slower E. coli growth in 100 % D2O, but production of both peptides was sufficient for MS analysis. The supernatant from 100 % D2O cultures was also ChemBioChem 2014, 15, 826 – 828
827
CHEMBIOCHEM COMMUNICATIONS
www.chembiochem.org the determination of the absolute configuration of any peptide.
Acknowledgements The authors are grateful to Anne Sydow for help with the construction of E. coli mutants. This work was supported by the BMBF within the Biokatalyse 2021 program and partially by a European Research Council starting grant under grant agreement no. 311477. Keywords: absolute configuration · d-amino acids · deuterium exchange · mass spectrometry · peptides · structure elucidation
Figure 2. HPLC/MS data of E. coli wild-type (WT) or Dpenta growing in H2O or D2O and expressing either A) gxpS or B) xtpS.
analyzed by GC/MS for additional amino acids (the expected +1 Da mass shift allows rapid identification of amino acids for which TA activity is responsible). No TA activity was observed for Ile, Leu, Val, Tyr, or Glu (Figure S5, Figure S6). However, Glu was consumed completely in the wild-type strain, possibly because of its role as an amino donor in TA reactions. A decrease in TA activity was observed for Ala, Phe, Gly, Met, and His; no TA activity was observed for Lys under the conditions tested (Figure S5, Figure S6). Asp, Pro, Ser, Thr, and Trp were not detectable after two days, so could not be analyzed. Thus in order to additionally target (at least) Ala, Phe, Gly, Met, and His for stereochemical analysis in peptides, further TAs (i.e., YfdZ, ArgD, AstC, GlyA, HisC, SerC, and YbdL) need to be deleted; these also act on amino acids and are highly conserved across enterobacteria. Although this is currently underway by our group, the analysis of Dpenta already shows the simple use and potential of such multiple TA mutants for stereochemical analysis of peptides. Moreover, a—still hypothetical—Ddoceda strain with deletions in all known amino-acid-specific TAs in E. coli[24] might require additional biosynthesis intermediates (e.g., several different 2-keto acids) and thus might be more difficult and more expensive to grow. In summary, we have shown that E. coli with deletions in TAencoding genes can be used for determination of the absolute configurations of several amino acids in NRPS-derived peptides; it will also be useful for ribosomally made peptides in similar experiments. In particular, growth in D2O with nondeuterated amino acids as peptide precursors allows rapid identification of the number of epimerized amino acids, and can easily be combined with additional labeling experiments for 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[1] G. M. Cragg, D. J. Newman, Biochim. Biophys. Acta Gen. Subj. 2013, 1830, 3670 – 3695. [2] D. J. Newman, G. M. Cragg, J. Nat. Prod. 2012, 75, 311 – 335. [3] G. M. Cragg, P. G. Grothaus, D. J. Newman, Chem. Rev. 2009, 109, 3012 – 3043. [4] M. H. Medema, K. Blin, P. Cimermancic, V. de Jager, P. Zakrzewski, M. A. Fischbach, T. Weber, E. Takano, R. Breitling, Nucleic Acids Res. 2011, 39, W339 – W346. [5] R. D. Kersten, Y.-L. Yang, Y. Xu, P. Cimermancic, S.-J. Nam, W. Fenical, M. A. Fischbach, B. S. Moore, P. C. Dorrestein, Nat. Chem. Biol. 2011, 7, 794 – 802. [6] J. D. Watrous, P. C. Dorrestein, Nat. Rev. Microbiol. 2011, 9, 683 – 694. [7] S. W. Fuchs, C. C. Sachs, C. Kegler, F. I. Nollmann, M. Karas, H. B. Bode, Anal. Chem. 2012, 84, 6948 – 6955. [8] J. Fu, X. Bian, S. Hu, H. Wang, F. Huang, P. M. Seibert, A. Plaza, L. Xia, R. Mller, A. F. Stewart, Y. Zhang, Nat. Biotechnol. 2012, 30, 440 – 446. [9] S. Kushnir, U. Sundermann, S. Yahiaoui, A. Brockmeyer, P. Janning, F. Schulz, Angew. Chem. Int. Ed. 2012, 51, 10664 – 10669; Angew. Chem. 2012, 124, 10820 – 10825. [10] J. H. Kim, Z. Feng, J. D. Bauer, D. Kallifidas, P. Y. Calle, S. F. Brady, Biopolymers 2010, 93, 833 – 844. [11] T. Stachelhaus, C. T. Walsh, Biochemistry 2000, 39, 5775 – 5787. [12] C. J. Balibar, F. H. Vaillancourt, C. T. Walsh, Chem. Biol. 2005, 12, 1189 – 1200. [13] S. A. Sieber, M. A. Marahiel, Chem. Rev. 2005, 105, 715 – 738. [14] K. Hoffmann, E. Schneider-Scherzer, H. Kleinkauf, R. Zocher, J. Biol. Chem. 1994, 269, 12710 – 12714. [15] M. F. Freeman, C. Gurgui, M. J. Helf, B. I. Morinaka, A. R. Uria, N. J. Oldham, H.-G. Sahl, S. Matsunaga, J. Piel, Science 2012, 338, 387 – 390. [16] P. M. Fischer, Curr. Protein Pept. Sci. 2003, 4, 339 – 356. [17] H. B. Bode, D. Reimer, S. W. Fuchs, F. Kirchner, C. Dauth, C. Kegler, W. Lorenzen, A. O. Brachmann, P. Grn, Chem. Eur. J. 2012, 18, 2342 – 2348. [18] N. L. Brock, S. R. Ravella, S. Schulz, J. S. Dickschat, Angew. Chem. Int. Ed. 2013, 52, 2100 – 2104; Angew. Chem. 2013, 125, 2154 – 2158. [19] C. A. Citron, N. L. Brock, P. Rabe, J. S. Dickschat, Angew. Chem. Int. Ed. 2012, 51, 4053 – 4057; Angew. Chem. 2012, 124, 4129 – 4133. [20] J. P. Gomez-Escribano, M. J. Bibb, J. Ind. Microbiol. Biotechnol. 2014, 41, 425 – 431. [21] C. D. Deane, D. A. Mitchell, J. Ind. Microbiol. Biotechnol. 2014, 41, 315 – 331. [22] J. D. Hegemann, M. Zimmermann, S. Zhu, D. Klug, M. A. Marahiel, Biopolymers 2013, 100, 527 – 542. [23] D. H. Gelfand, R. A. Steinberg, J. Bacteriol. 1977, 130, 429 – 440. [24] S. H. Kim, B. L. Schneider, L. Reitzer, J. Bacteriol. 2010, 192, 5304 – 5311. [25] H. Yoneyama, H. Hori, S.-J. Lim, M. Murata, K. Ando, E. Isogai, R. Katsumata, Biosci. Biotechnol. Biochem. 2011, 75, 930 – 938. Received: September 18, 2013 Revised: January 15, 2014 Published online on March 11, 2014
ChemBioChem 2014, 15, 826 – 828
828
Rapid Determination of the Amino Acid Configuration of Xenotetrapeptide Carsten Kegler, Friederike I. Nollmann, Tilman Ahrendt, Florian Fleischhacker, Edna Bode, and Helge B. Bode*[a] An E. coli strain with deletions in five transaminases (DaspC DilvE DtyrB DavtA DybfQ) was constructed to be unable to degrade several amino acids. This strain was used as an expression host for the analysis of the amino acid configuration of nonribosomally synthesized peptides, including the novel peptide “xenotetrapeptide” from Xenorhabdus nematophila, by using a combination of labeling experiments and mass spectrometry. Additionally, the number of d-amino acids in the produced peptide was assigned following simple cultivation of the expression strain in D2O.
Introduction Natural-product research has been the source of several important drugs over the last 60 years; natural products or derivatives thereof are major therapeutics, particularly as anti-infectives.[1–3] As drug discovery in general is very costly and little revenue can be expected from this kind of research (especially in the anti-infectives field), pharmaceutical companies have reduced their efforts in this area in favor of developing treatments for chronic or “more serious” diseases like cancer. The decline in industrial natural-product research has not been followed in academic research, which has started to revitalize the field by using modern methods of bioinformatics,[4] analytical chemistry,[5–7] and molecular biology.[8–10] The combination of these approaches allows rapid identification and access to novel natural products, including those that are not produced under standard conditions. Additionally, compounds that are naturally produced in only minute amounts (insufficient for isolation, structure elucidation, and bioactivity testing) can nowadays been identified quite easily. Especially important is the more recent adoption of mass spectrometry (MS) as a routine and highly sensitive tool to increase the number of natural products that can be identified in a single organism.[6, 7] However, the advantage of a high sensitivity might prove disadvantageous, as natural products identified by MS methods might be difficult to isolate because of their low production levels. Fortunately, progress in organic synthesis allows alternative access
to these compounds, if they can be synthesized with reasonable effort. However, in order to facilitate their synthesis, the absolute configurations of the desired natural products must be known. Peptide natural products can consist of l- and d-amino acids: epimerization (E)[11] or dual condensation/epimerization (C/E)[12] domains in nonribosomal peptide synthetases (NRPS) convert incorporated l-amino acids into d-amino acids of the final product.[13] Additionally, it has been shown that d-amino acids can be directly incorporated by NRPSs.[14] The recent identification of the proteusin family of natural products has shown that d-amino acids can also be found in ribosomally derived peptides using a so far unknown mechanism catalyzed by enzymes belonging to the radical S-adenosylmethionine (rSAM) family.[15] The presence of d-amino acids makes peptides more stable against proteases, and can result in a different conformation of the peptide chain.[16] Usually the amino acid configuration is determined by peptide hydrolysis or from the NMR-determined 3D structure, however, both methods require the isolation of the desired peptide in sufficient amounts (often difficult, as already stated). For these compounds a real need for the determination (and confirmation) of the amino acid configuration exists; stable isotope labeling (with loss of label due to epimerization) can be used to detect the configuration based on MS methods, as we have shown previously for peptides produced in Photorhabdus luminescens and Xenorhabdus nematophila.[17] However, the incorporation of [2-2H]-labeled amino acids allowing the rapid determination of epimerization (E) or rSAM activity is only possible in mutants with blocked amino transaminase (TA) activity, as otherwise the label is rapidly lost due to general TA activity.[17] Deuterium labels have also been used extensively to elucidate other biochemical pathways because of their simple detection by MS.[18, 19] Here we describe an Escherichia coli strain with deletions in five TAs, and its use as expression host for the determination of the peptide configurations of the known GameXPeptide A (1)[17] and a novel peptide from X. nematophila, “xenotetrapeptide” (2). Such a strain is especially useful, as heterolo-
[a] Dr. C. Kegler,+ F. I. Nollmann,+ T. Ahrendt, F. Fleischhacker, E. Bode, Prof. Dr. H. B. Bode Merck Stiftungsprofessur fr Molekulare Biotechnologie Fachbereich Biowissenschaften, Goethe Universitt Frankfurt Max-von-Laue-Strasse 9, 60438 Frankfurt am Main (Germany) E-mail: [email protected] [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cbic.201300602.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 826 – 828
826
CHEMBIOCHEM COMMUNICATIONS
www.chembiochem.org
gous expression of unknown biosynthesis gene clusters is now a widespread method in natural-product research.[20–22]
Results and Discussion As the major TAs in E. coli involved in amino acid metabolism are AspC, IlvE and TyrB,[23] the corresponding genes were deleted in E. coli DH10B. In order to test this strain for structure elucidation of natural products, the gene from P. luminescens encoding GxpS (the NRPS that produces GameXPeptide A, 1)[17] and an uncharacterized gene (XNC1_2022) encoding an NPRS from X. nematophila ATCC19061 were transformed into this strain. XNC1_2022 was shown to be involved in the production of a compound with m/z 411.3 [M+H] + and 433.3 [M+Na] + by the loss of production by an insertion mutant. Analysis of XNC1_2022 in detail revealed four modules, including two dual condensation/epimerization (C/E) domains (see Figure S1 in the Supporting Information) predicted to be involved in amino acid epimerization of valine-1 and 3, thereby resulting in the predicted structure: cyclo(vLvV). The predicted structure is in agreement with the formula C21H38N4O4 as determined by HR-ESIMS for X. nematophila and E. coli expressing XNC1_2022, and was confirmed by labeling experiments in 15N and 13C medium, thus allowing simple determination of the number of nitrogen and carbon atoms (Figure S2).[17] Thus the compound was named xenotetrapeptide (2), and the corresponding NRPS was renamed xenotetrapeptide synthetase (XtpS; Scheme 1). The correct configuration of 1 as cyclo(vLflL) was determined by using the addition of fully deuterated Leu, Val, and Phe in an E. coli triple mutant (deletions of aspC, tyrB, and ilvE) that expressed gxpS (Figure S3); the results were similar to those described in P. luminescens naturally producing 1.[17] For 2 the presence of l-Leu was shown by the expected 10 Da shift resulting from the incorporation of 2H10-Leu (data not shown). However, upon addition of 2H8-valine, the resulting mixture of isotopomers with mass shifts of 21 and 22 Da indicated either partial epimerization or (more likely) valine-specific TA activity from additional TAs (Figure 1 A). Thus TA-encoding
Scheme 1. Domain organization of XtpS involved in the biosynthesis of xenotetrapeptide 2 with bound biosynthesis intermediates. Domains, A: adenylation, T: thiolation, C/E: dual condensation/epimerization, C: condensation, TE: thioesterase.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. HPLC/MS data of xenotetrapeptide (2) from cultivation of E. coli wild-type (WT) or TA-deficient strains (Dtriple and Dpenta) expressing xtpS in A) LB medium, and B) M9 minimal medium with l-[2H10]leucine or l-[2H8]valine.
genes avtA and ybfQ[24, 25] were also deleted (E. coli Dpenta); this showed predominantly the expected mass shift of 22 Da upon expression of xtpS with addition of 2H8-valine, in agreement with the presence of one l-(2H8 label) and two d-valine moieties (2H7 label) in 1 (Figure 1 B). In order to test the predicted position of the d- and l-valines, 2 was synthesized by solid phase peptide synthesis; this showed retention time and fragmentation pattern identical to those of the natural compound (Figure S4). As some deuterated amino acids are either not available or very expensive, we tested whether the labeling approach can be reversed by growing E. coli and Dpenta in 100 % D2O with nondeuterated medium components: amino acid epimerization should result in a mass shift of +1 Da, so, for 1 and 2, mass shifts of 3 and 2 Da would be observed because of the presence of three and two d-amino acids in the final compounds (Figure 2). This was indeed the case. Although the results were not perfect (probably due to residual TA activity prior to amino acid incorporation catalyzed by the NRPS: see small peaks resulting from isotopomers with one deuterium less than expected), the significant difference between wildtype and Dpenta strains clearly shows that the method works (Figure 2). We observed slightly slower E. coli growth in 100 % D2O, but production of both peptides was sufficient for MS analysis. The supernatant from 100 % D2O cultures was also ChemBioChem 2014, 15, 826 – 828
827
CHEMBIOCHEM COMMUNICATIONS
www.chembiochem.org the determination of the absolute configuration of any peptide.
Acknowledgements The authors are grateful to Anne Sydow for help with the construction of E. coli mutants. This work was supported by the BMBF within the Biokatalyse 2021 program and partially by a European Research Council starting grant under grant agreement no. 311477. Keywords: absolute configuration · d-amino acids · deuterium exchange · mass spectrometry · peptides · structure elucidation
Figure 2. HPLC/MS data of E. coli wild-type (WT) or Dpenta growing in H2O or D2O and expressing either A) gxpS or B) xtpS.
analyzed by GC/MS for additional amino acids (the expected +1 Da mass shift allows rapid identification of amino acids for which TA activity is responsible). No TA activity was observed for Ile, Leu, Val, Tyr, or Glu (Figure S5, Figure S6). However, Glu was consumed completely in the wild-type strain, possibly because of its role as an amino donor in TA reactions. A decrease in TA activity was observed for Ala, Phe, Gly, Met, and His; no TA activity was observed for Lys under the conditions tested (Figure S5, Figure S6). Asp, Pro, Ser, Thr, and Trp were not detectable after two days, so could not be analyzed. Thus in order to additionally target (at least) Ala, Phe, Gly, Met, and His for stereochemical analysis in peptides, further TAs (i.e., YfdZ, ArgD, AstC, GlyA, HisC, SerC, and YbdL) need to be deleted; these also act on amino acids and are highly conserved across enterobacteria. Although this is currently underway by our group, the analysis of Dpenta already shows the simple use and potential of such multiple TA mutants for stereochemical analysis of peptides. Moreover, a—still hypothetical—Ddoceda strain with deletions in all known amino-acid-specific TAs in E. coli[24] might require additional biosynthesis intermediates (e.g., several different 2-keto acids) and thus might be more difficult and more expensive to grow. In summary, we have shown that E. coli with deletions in TAencoding genes can be used for determination of the absolute configurations of several amino acids in NRPS-derived peptides; it will also be useful for ribosomally made peptides in similar experiments. In particular, growth in D2O with nondeuterated amino acids as peptide precursors allows rapid identification of the number of epimerized amino acids, and can easily be combined with additional labeling experiments for 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[1] G. M. Cragg, D. J. Newman, Biochim. Biophys. Acta Gen. Subj. 2013, 1830, 3670 – 3695. [2] D. J. Newman, G. M. Cragg, J. Nat. Prod. 2012, 75, 311 – 335. [3] G. M. Cragg, P. G. Grothaus, D. J. Newman, Chem. Rev. 2009, 109, 3012 – 3043. [4] M. H. Medema, K. Blin, P. Cimermancic, V. de Jager, P. Zakrzewski, M. A. Fischbach, T. Weber, E. Takano, R. Breitling, Nucleic Acids Res. 2011, 39, W339 – W346. [5] R. D. Kersten, Y.-L. Yang, Y. Xu, P. Cimermancic, S.-J. Nam, W. Fenical, M. A. Fischbach, B. S. Moore, P. C. Dorrestein, Nat. Chem. Biol. 2011, 7, 794 – 802. [6] J. D. Watrous, P. C. Dorrestein, Nat. Rev. Microbiol. 2011, 9, 683 – 694. [7] S. W. Fuchs, C. C. Sachs, C. Kegler, F. I. Nollmann, M. Karas, H. B. Bode, Anal. Chem. 2012, 84, 6948 – 6955. [8] J. Fu, X. Bian, S. Hu, H. Wang, F. Huang, P. M. Seibert, A. Plaza, L. Xia, R. Mller, A. F. Stewart, Y. Zhang, Nat. Biotechnol. 2012, 30, 440 – 446. [9] S. Kushnir, U. Sundermann, S. Yahiaoui, A. Brockmeyer, P. Janning, F. Schulz, Angew. Chem. Int. Ed. 2012, 51, 10664 – 10669; Angew. Chem. 2012, 124, 10820 – 10825. [10] J. H. Kim, Z. Feng, J. D. Bauer, D. Kallifidas, P. Y. Calle, S. F. Brady, Biopolymers 2010, 93, 833 – 844. [11] T. Stachelhaus, C. T. Walsh, Biochemistry 2000, 39, 5775 – 5787. [12] C. J. Balibar, F. H. Vaillancourt, C. T. Walsh, Chem. Biol. 2005, 12, 1189 – 1200. [13] S. A. Sieber, M. A. Marahiel, Chem. Rev. 2005, 105, 715 – 738. [14] K. Hoffmann, E. Schneider-Scherzer, H. Kleinkauf, R. Zocher, J. Biol. Chem. 1994, 269, 12710 – 12714. [15] M. F. Freeman, C. Gurgui, M. J. Helf, B. I. Morinaka, A. R. Uria, N. J. Oldham, H.-G. Sahl, S. Matsunaga, J. Piel, Science 2012, 338, 387 – 390. [16] P. M. Fischer, Curr. Protein Pept. Sci. 2003, 4, 339 – 356. [17] H. B. Bode, D. Reimer, S. W. Fuchs, F. Kirchner, C. Dauth, C. Kegler, W. Lorenzen, A. O. Brachmann, P. Grn, Chem. Eur. J. 2012, 18, 2342 – 2348. [18] N. L. Brock, S. R. Ravella, S. Schulz, J. S. Dickschat, Angew. Chem. Int. Ed. 2013, 52, 2100 – 2104; Angew. Chem. 2013, 125, 2154 – 2158. [19] C. A. Citron, N. L. Brock, P. Rabe, J. S. Dickschat, Angew. Chem. Int. Ed. 2012, 51, 4053 – 4057; Angew. Chem. 2012, 124, 4129 – 4133. [20] J. P. Gomez-Escribano, M. J. Bibb, J. Ind. Microbiol. Biotechnol. 2014, 41, 425 – 431. [21] C. D. Deane, D. A. Mitchell, J. Ind. Microbiol. Biotechnol. 2014, 41, 315 – 331. [22] J. D. Hegemann, M. Zimmermann, S. Zhu, D. Klug, M. A. Marahiel, Biopolymers 2013, 100, 527 – 542. [23] D. H. Gelfand, R. A. Steinberg, J. Bacteriol. 1977, 130, 429 – 440. [24] S. H. Kim, B. L. Schneider, L. Reitzer, J. Bacteriol. 2010, 192, 5304 – 5311. [25] H. Yoneyama, H. Hori, S.-J. Lim, M. Murata, K. Ando, E. Isogai, R. Katsumata, Biosci. Biotechnol. Biochem. 2011, 75, 930 – 938. Received: September 18, 2013 Revised: January 15, 2014 Published online on March 11, 2014
ChemBioChem 2014, 15, 826 – 828
828
