In situ detection of the intermediates in the biosynthesis

of surfactin, a lipoheptapeptide from Bacillus subtilis

OKB 105, by whole-cell cell matrix-assisted laser
desorption/ionization time-of-ight mass spectrometry in
combination with mutant analysis
Joachim Vater
*
, Christopher Wilde and Henny Kell
Institut fu r Chemie, Arbeitsgruppe Biochemie und Molekulare Biologie, Technische Universita t Berlin, Franklinstr. 29, D-10587 Berlin, Germany
Received 9 November 2008; Revised 26 January 2009; Accepted 12 March 2009
An innovative technique to investigate the intermediates involved in the biosynthesis of the
lipoheptapeptide surfactin from Bacillus subtilis OKB105 combining whole-cell matrix-assisted
laser desorption/ionization time-of-ight mass spectrometry (MALDI-TOFMS) with targeted gener-
ation of knock-out mutants was demonstrated. This method allows efcient, sensitive detection of
biosynthetic intermediates in a minimum of time directly at the outer surface of microbial cells
picked from agar plates or in surface extracts prepared thereof. Biosynthesis of surfactin is encoded
by the srf-operon which is organized into four open reading frames which have been attributed to
three multifunctional NRPS enzymes (SrfA-C) and a thioesterase/acyltransferase enzyme SrfD. For
the wild-type strain OKB 105 only the end product surfactin was found mass spectrometrically. For
the detection of lipopeptide intermediates three plasmid- and transposon-insertion mutants were
generated interrupting the surfactin assembly line at dened positions. Strain LAB 327 was mutated
in the spacer region between enzymes SrfA and B. Here only SrfA was active with the lipotripeptide
b-OH-acyl-L-Glu-L-Leu-D-Leu as the end product. Mutant OKB 120 bears a transposon mutation in
SrfB between the rst and second amino acid activating modules SrfB1 and SrfB2. It showed all
intermediates from the lipodi- until to the lipotetrapeptide b-OH-acyl-L-Glu-L-Leu-D-Leu-L-Val. In
LAB 223 SrfC was knocked out by a transposon mutation. It produced the lipohexapeptide b-OH-
acyl-L-Glu-L-Leu-D-Leu-L-Val-L-Asp-D-Leu. Our work highlights the applicability and the poten-
tial of whole-cell MALDI-TOFMS as an innovative efcient tool for the analysis of intermediate steps
of biosynthetic pathways. Copyright # 2009 John Wiley & Sons, Ltd.
Surfactin is a cyclic lipoheptapeptide that is produced by
various bacillus strains.
111
It appears as a complex of
numerous isoforms varying both in their peptide moiety and
b-hydroxy fatty acid side chain.
59
Surfactin is distinguished
by superior surface-, interface- and membrane-active proper-
ties
8,9
and has attained high interest for biomedical and
biotechnological applications.
8,1217
Biosynthesis of surfactin is encoded by the srf-operon
18,19
with a size of 27 kb which is organized into four open reading
frames
10,11,19,20
which have been attributed to three multi-
functional NRPS enzymes (SrfA-C)
21,22
and an external
thioesterase/acyltransferase enzyme SrfD.
23
Figure 1 shows
a diagram depicting the organization of the surfactin
biosynthetic system. Surfactin synthetase forms surfactin
in an assembly line process by the interaction of multiple 4
0
-
phosphopantetheine cofactors,
24
one at each amino acid
activating module. Substrate amino acids and lipopeptide
intermediates remain thioester-linked to the carriers
throughout the entire biosynthetic process. SrfA and SrfB
each comprise three amino acid activating modules. SrfA
activates glutamic acid and two leucines, while SrfB
thioesteries valine, aspartic acid and leucine. Surfactin
formation is initiated by transfer of the b-hydroxy fatty acid
component from coenzyme A to the L-glutamic acid starter
amino acid residing at the reaction center of the rst module
of SrfA in the thioester-activated form.
23
This process is
mediated by the acyl transferase enzyme SrfD.
23
From the
genetic organization of the srf-operon and the modular
architecture of surfactin synthetase derived thereof the
assembly line of surfactin biosynthesis can be modelled (see
Fig. 1) and the following sequence of lipopeptide intermedi-
ates can be predicted: SrfA should elongate the growing
lipopeptide chain until to the lipotripeptide b-OH-acyl-L-
Glu-L-Leu-D-Leu which is transferred to SrfB for extension
to the lipohexapeptide b-OH-acyl-L-Glu-L-Leu-D-Leu-L-
Val-L-Asp-D-Leu. Surfactin formation is terminated by
SrfC, a one-module enzyme, which contributes the C-
terminal leucine residue. Finally the assembled linear
RAPID COMMUNICATIONS IN MASS SPECTROMETRY
Rapid Commun. Mass Spectrom. 2009; 23: 14931498
Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.4031
*Correspondence to: J. Vater, Institut fu r Chemie, Arbeitsgruppe
Biochemie und Molekulare Biologie, Technische Universitat
Berlin, Franklinstr. 29, D-10587 Berlin, Germany.
E-mail: Vater@chem.tu-berlin.de
Copyright # 2009 John Wiley & Sons, Ltd.
lipoheptapeptide is converted into its cyclic form by lactone
formation and released into the medium by the action of the
internal thioesterase domain of SrfC. This sequence of events
has not yet been experimentally veried in detail. In general,
scarce information is available for the intermediates in
assembly line processes, such as the elongation reactions in
the biosynthesis of nonribosomal peptides and polyketides.
Isolation and analysis of thioester-bound intermediates often
is a complicated, multi-step, time-consuming task which
needs highly sensitive techniques, such as radiochemical or
advanced mass spectrometric methodology, for their detec-
tion. In this paper we present an innovative procedure
detecting the lipopeptide products of B. subtilis mutants
showing knock-outs at different loci of surfactin synthetase
(see Fig. 1) by whole-cell matrix-assisted laser desorption/
ionization time-of-ight mass spectrometry (MALDI-TOFMS)
which allows the visualization of such intermediates in situ at
the outer surface of the investigated mutant cells efciently
with reasonable specicity and sensitivity in a minimum of
time.
EXPERIMENTAL
Microorganisms and growth conditions
The surfactin producer strain Bacillus subtilis OKB105
25
and
nonproducing mutant derivatives
18
OKB 120, LAB 223 and
LAB 327 were grown on agar plates in the Landy medium
26
containing 1.5 (w/v) agar-agar for 48 h at 28C and stored at
room temperature prior to mass spectrometric analysis.
Whole-cell MALDI-TOFMS
MALDI-TOF mass spectra were recorded using a Bruker
Reex MALDI-TOF instrument equipped with a 337 nm
nitrogen laser for desorption and ionization, as described
previously.
7,10,2729
a-Cyano-4-hydroxycinnamic acid (CCA)
was used as matrix. Whole-cell MALDI-TOFMS analyses of
surfactin and its intermediate lipopeptides were performed
with intact cells picked from agar plates, spotted onto the
target and covered with matrix medium, a saturated solution
of CCAin 40%aqueous acetonitrile/0.1%triuoroacetic acid
(TFA). Alternatively, 2 mL portions of surface extracts were
Figure 1. Assembly line for the biosynthesis of surfactin. SrfA, SrfB and SrfC indicate the amino acid activating subunits of
surfactin synthetase which are encoded by the srf-operon. SrfA and SrfB each comprise three amino acid activating modules
labeled as modules 13 (SrfA1-3) and modules 46 (SrfB1-3), respectively. Srf C is a one-module enzyme (module 7). A, C and E
designate the adenylation, condensation and epimerization domains in a module. PCP (peptidyl carrier protein) functions as the
thiolation domain bearing the 4
0
-phosphopantetheine carrier. Along the assembly substrate amino acids and lipopeptide
intermediates are covalently attached to the active sulfhydryl group of the cofactor in a thioester bond. TE is the C-terminal
thioesterase domain of SrfC catalyzing the cyclization and release of surfactin by macrolactone formation. FA, fatty acid. Arrows
indicate the mutation loci for mutants LAB327, OKB120 and LAB223.
Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 14931498
DOI: 10.1002/rcm
1494 J. Vater, C. Wilde and H. Kell
mixed with the same volume of matrix solution. Surface
extracts of B. subtilis OKB 105 and mutants strains were
prepared by extraction of cells picked from agar plates with
30 mL 70% acetonitrile/0.1% TFA.
RESULTS AND DISCUSSION
In nonribosomal peptide biosynthesis catalyzed by NRPS
multi-enzyme systems amino acid substrates and (lipo)pep-
tide intermediates are covalently tethered to the 4
0
-phospho-
pantetheine carriers at the thiolation sites of the T-domains in
the amino acid activating modules.
24
Generally, in the
biosynthetic process intermediates are rapidly chanelled into
the end product. Therefore, their level within the living cell
and in the growth medium is usually so low that highly
sensitive mass spectrometric methodology is needed for their
detection. With this in mind MALDI-TOFMS is the method of
choice to investigate complex samples, such as whole cells as
well as crude cell and culture ltrate extracts.
In this study we investigated the biosynthesis of the
lipoheptapeptide surfactin fromB. subtilis OKB 105 and used
whole-cell MALDI-TOFMS as an innovative technique for
the detectionof surfactinandbiosynthetic intermediates in situ.
In the last decade this innovative method has efciently been
applied for mass spectrometric ngerprinting
30,31
and meta-
bolic proling of microorganisms.
7,10,2729,3234
The mass
spectrum in Fig. 2(A) of a surface extract taken from whole
cells of the wild-type strain B. subtilis OKB105 shows that in
this case only the endproduct surfactincouldbe detectedat the
outer surface of this organism. Similar results were obtained
when an aliquot of a cell suspension was spotted onto the
target and covered with matrix solution. Surfactin appeared as
a complex of various isoforms exhibiting b-hydroxy fatty acid
components with chain lengths of 13, 14 and 15 carbon atoms.
The species with mass numbers of m/z 1030.8; 1044.8 and
1058.8 were attributed to the sodiumadducts of C13, C14 and
C15 variants of surfactin, while those at m/z 1046.8; 1060.8
and 1074.8 represent the corresponding potassium adducts.
Biosynthetic intermediates were not found for the wild-type
strain. Presumably, the component enzymes of surfactin
synthetase SrfA-C are tightly coupled within the cell in a
supramolecular complex channelling lipopeptide intermedi-
ates rapidly towards surfactin. Under such conditions they
were not released into the cell medium and could therefore
not be detected by mass spectrometry.
The classical approach to analyze the intermediate steps of
nonribosomal biosynthesis usually is a complicated time-
consuming multi-step task which requires (a) preparation of
a cell-free extract of the producer organism; (b) isolation and
purication of the component enzymes of a (lipo)peptide
synthetase; and (c) building up the growing (lipo)peptide
chain by combination of these proteins in series and pro-
viding them with the appropriate substrate mixtures. At this
stage the growing (lipo)peptide chain can be monitored by
incorporation of radioactively labelled amino acid substrates,
as demonstrated for bioactive peptides produced by B. brevis,
such as gramicidin S
3537
and tyrocidine,
38
for example.
However, for the ultimate identication of the intermediate
products, particularly when they are processed by internal
modication reactions, such as glycosylation or cyclization,
the following tasks were required: (d) release of the inter-
mediates from the reaction centers either by alkaline hydro-
lysis, oxidative cleavage with performic acid or early
termination steps, such as internal cyclization reactions; (e)
separation and purication of the intermediates by chroma-
tographic procedures; and, nally, (f) their identication and
structural characterization by amino acid analysis, Edman
degradation and particularly by sensitive mass spectrometric
techniques.
Great progress has been achieved in the investigation of
intermediates in such complex processes, as the biosynthesis
of nonribosomal peptides and polyketides, by the develop-
ment of advanced mass spectrometric methodology of high
resolution and accuracy, such as electrospray ionization
Fourier transform mass spectrometry.
39
By this technique
detailed biosynthetic information on substrate loading and
specicity, active site mapping, the intermediates in the
elongation process, timing of tailoring functions and
characterization of modules that deviate from the colinearity
rules of NRPS and PKS has been achieved.
4044
This method
in common with several other mass spectrometer instrument
types is able to visualize the isotope distribution of large
peptides in complex mixtures, such as partially digested
proteins and whole domains with molecular masses greater
than 5 kDa which can be analyzed with high mass accuracy;
however, this highly sensitive and informative procedure
also needs about ten steps to properly map the active sites of
NRPS and PKS.
39
The rst step after preparation of a cell-free
extract is to obtain the investigated protein with a purity of
higher than 8090% for digestion by cyanogen bromide or
proteases, such as GluC, LysC or trypsin to obtain large
fragments which have to be separated by high-performance
liquid chromatography (HPLC methodology) and analyzed
by mass spectrometric analysis.
Whole-cell MALDI-TOFMS applied in this paper is much
simpler andfaster andaffords sensitive detectionof biosynthetic
intermediates by a strategy of generating knock-out mutants
interrupting the assembly line for surfactin production at
denite positions. This technique allowed the monitoring of
surfactin intermediates in situ directly at the outer surface of
mutant cells with reasonable specicity and excellent sensi-
tivity in a minimum of time. The method could allow more
denite assignment of structure if a MALDI-TOF-TOF
instrument is available, affording tandem mass spectro-
metric (MS/MS) capability which would assist interpret-
ation. We used three plasmid- and transposon-insertion
mutants,
18
LAB327, LAB223 and OKB120, which have been
characterized in detail in a previous study.
45
By mutation at
different loci the surfactin assembly line is interrupted at
denite positions (see Fig. 1) thus enabling the release of
intermediate products which can be predicted from the
modular architecture of surfactin synthetase. Strain LAB327
is mutated in the spacer region between enzymes SrfAand B.
It contains SrfA in functional form, while SrfB and C are
missing. Therefore, the lipotripeptide b-OH-acyl-L-Glu-L-
Leu-D-Leu is expected as the end product. Mutant OKB 120
bears a transposon mutation in SrfB between the rst and
second amino acid activating modules SrfB1 and SrfB2. In
this case formation of the lipotetrapeptide b -OH-acyl-L-Glu-
L-Leu-D-Leu-L-Val should be accomplished. LAB 223 shows
Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 14931498
DOI: 10.1002/rcm
Surfactin biosynthesis by whole-cell MALDI-TOFMS 1495
a transposon mutation in SrfC. In this strain SrfA and B are
expressed, but it is decient in SrfC. This mutant should be
able to produce the lipohexapeptide b-OH-acyl-L-Glu-L-
Leu-D-Leu-L-Val-L-Asp-D-Leu. To detect these intermedi-
ates mutant strains were grown on Landy agar plates for
3 days at 288C. Cell material was picked from agar plates,
directly spotted on the target and embedded in matrix
solution. An alternative procedure which yielded cleaner
samples showing better resolved mass spectra with higher
peak intensities was to prepare surface extracts of these
organisms by extraction of cell material with 70% aceto-
nitrile/0.1% TFA. In fact, the alkali-cationized C14- and
Figure 2. MALDI-TOF mass spectra of surface extracts prepared from intact whole cells of B. subtilis OKB105 and surfactin
knock-out mutants showing the production of surfactins by the wild-type strain (A) and the formation of C13-C15 lipopeptide
intermediates by mutants LAB327 (B), OKB120 (CE), and LAB223 (F). For preparation of surface extracts cell material was
picked fromagar plates and extracted with 70%acetonitrile/0.1%TFA(v/v). C13C15 isoforms of surfactin were detected between
m/z 10001100. Biosynthetic intermediates were observed in the mass range of m/z 5001000.
Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 14931498
DOI: 10.1002/rcm
1496 J. Vater, C. Wilde and H. Kell
C15-variants of the intermediates of surfactin formation were
found as predicted from the modular architecture of
surfactin synthetase. Their MALDI-TOF mass spectra are
shown in Figs. 2(B)2(F). The mass data are compiled in
Table 1. In the case of LAB327 the lipotripeptide appeared
(Fig. 2(B)). OKB120 showed all intermediates fromthe lipodi-
until to the lipotetrapeptide (Figs. 2(C)2(E)), while for
LAB223 the lipohexapeptide was found (Fig. 2(F)).
Obviously, similar to surfactin production by the wild-type
strain, in LAB223 the lipopeptide chain is rapidly growing
until to the lipohexapeptide as the end product of SrfB
without release of upstream intermediates.
The great advantage of whole-cell MALDI-TOFMS is that
the intermediates of NRPS and PKS can be detected directly
during cell growth. There is no need to prepare cell-free
systems to isolate the producer enzymes and to purify the
intermediates. However, this attractive technique needs
modication when (A) the attachment of the intermediates
which are bound to the reaction centers covalently is very
stable with the consequence that their level in the cellular
medium would be extremely low; (B) the intermediates are
not released by the cell; and (C) when they are not adsorbed
to the outer surface of the cell wall. To detect them in cases
(A) and (B) extracts of disintegrated cells have to be
prepared. If necessary the intermediates have to be released
fromthe producer enzymes by mild transthiolation reactions
using thiol reagents, such as dithioerythritol, cysteamine or
cysteine, followed by mass spectrometric detection. In case
(C) a search for intermediates has to be performed in extracts
of the culture ltrate.
In this study surfactin biosynthesis was taken as a
prominent, well-characterized biosynthetic system to
demonstrate the applicability and the potential of whole-
cell MALDI-TOFMS as an innovative efcient technique for
the analysis of the intermediate steps of biosynthetic
pathways with reasonable specicity and in a minimum of
time. Our results highlight the utility of combined genetic
and in situ mass spectrometric strategies to gain insight into
the molecular details of the biosynthesis of natural
compounds in general.
Acknowledgements
We thank Professor P. Zuber for providing us with the
B. subtilis knock-out mutants.
REFERENCES
1. Kakinuma A, Hori M, Isono M, Tamura G, Arima K. Agric.
Biol. Chem. 1969; 33: 971.
2. Kakinuma A, Sugino H, Isono M, Tamura G, Arima K. Agric.
Biol. Chem. 1969; 33: 973.
3. Kakinuma A, Hori M, Sugino H, Yoshida I, Isono M, Tamura
G, Arima K. Agric. Biol. Chem. 1969; 33: 1523.
4. Kakinuma A, Ouchida A, Shima T, Sugino H, Isono M,
Tamura G, Arima K. Agric. Biol. Chem. 1969; 33: 1669.
5. Kowall M, Vater J, Kluge B, Stein T, Franke P, Ziessow D.
J. Colloid Interface Sci. 1998; 204: 1.
6. Bonmatin JM, Laprevote O, Peypoux F. Comb. Chem. High
Throughput Screen. 2003; 6: 541.
7. Vater J, Gao X, Hitzeroth G, Wilde C, Franke P. Comb. Chem.
High Throughput Screen. 2003; 6: 557.
8. Peypoux F, Bonmatin JM, Wallach J. Appl. Microbiol. Biotech-
nol. 1999; 51: 553.
9. Stein T. Mol. Microbiol. 2005; 56: 845.
10. Hofemeister J, Conrad B, Adler B, Hofemeister B, Feesche J,
Kucheryava N, Steinborn G, Franke P, Grammel N,
Zwintscher A, Leenders F, Hitzeroth G, Vater J. Mol. Gen.
Genom. 2004; 272: 363.
11. Koumoutsi A, Chen X-H, Henne A, Liesegang H, Hitzeroth
G, Franke P, Vater J, Borriss R. J. Bacteriol. 2004; 186: 1084.
12. Vollenbroich D, Pauli G, O

zel M, Vater J. Appl. Environ.

Microbiol. 1997; 63: 44.
13. Vollenbroich D, O

zel M, Vater J, Kamp RM, Pauli G. Biolo-

gicals 1997; 25: 289.
14. Desai JD, Banat I. Microbiol. Mol. Biol. Rev. 1997; 61: 47.
15. Cameotra SS, Makkar RS. Curr. Opin. Microbiol. 2004; 7:
262.
Table 1. Detection of intermediates in the biosynthesis of surfactin by MALDI-TOFMS at whole cells and in surface extracts of
B. subtilis OKB105 and srf knock-out mutant strains
Strain Locus of mutation Intermediate or end product Mass number m/z
OKB105 C13-surfactin [MNa]

1030.8
C13-surfactin [MK]

1046.8
C14-surfactin [MNa]

1044.8
C14-surfactin [MK]

1060.8
C15-surfactin [MNa]

1058.8
C15-surfactin [MK]1074.8
LAB327 between SrfA and SrfB C14-b-OH-ELL [MK]

638.6
C15-b-OH-ELL [MK]

652.4
OKB120 in SrfB between module 4 and module 5 C14-b-OH-EL [MNa]

509.4
C14-b-OH-EL [MK]

525.6
C15-b-OH-EL [MNa]

523.3
C15-b-OH-EL [MK]

539.4
C14-b-OH-ELL [MK]

638.6
C15-b-OH-ELL [MNa]

636.4
C15-b-OH-ELL [MK]

652.4
C13-b-OH-ELLV [MNa]

707.2
C15-b-OH-ELLV [MNa]

735.5
C15-b-OH-ELLV [MK]

751.5
LAB223 in SrfC C14-b-OH-ELLVDL [MK]

965.6
C15-b-OH-ELLVDL [MNa]

963.6
C15-b-OH-ELLVDL [MK]

979.6
Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 14931498
DOI: 10.1002/rcm
Surfactin biosynthesis by whole-cell MALDI-TOFMS 1497
16. Rodrigues L, Banat IM, Teixeira J, Oliveira R. J. Antimicrob.
Chemother. 2006; 57: 609.
17. Singh A, van Damme JD, Ward OP. Biotechnol. Adv. 2007; 25:
99.
18. Nakano MM, Magnuson R, Myers A, Curry J, Grossman AD,
Zuber P. J. Bacteriol. 1991; 173: 1770.
19. Cosmina P, Rodriguez F, de Ferra F, Grandi G, Perego M,
Venema G, van Sinderen D. Mol. Microbiol. 1993; 8: 821.
20. Van Sinderen D, Galli G, Cosmina P, de Ferra F, Withoff S,
Venema G, Grandi G. Mol. Microbiol. 1993; 8: 833.
21. Ulrich C, Kluge B, Palacz Z, Vater J. Biochemistry 1991; 30:
6503.
22. Menkhaus M, Ulrich C, Kluge B, Vater J, Vollenbroich D,
Kamp RM. J. Biol. Chem. 1993; 268: 7678.
23. Steller S, Sokoll A, Wilde C, Bernhard F, Franke P, Vater J.
Biochemistry 2004; 43: 11331.
24. Stein T, Vater J, Kruft V, Otto A, Wittmann-Liebold B, Franke
P, Panico M, McDowell R, Morris HR. J. Biol. Chem. 1996; 271:
15428.
25. Nakano MM, Marahiel MA, Zuber P. J. Bacteriol. 1988; 170:
5662.
26. Landy M, Warren GH, Rosenman SB, Colio LG. Proc. Soc.
Exp. Biol. Med. 1948; 67: 539.
27. Leenders F, Stein TH, Kablitz B, Franke P, Vater J. Rapid
Commun. Mass Spectrom. 1999; 13: 943.
28. Vater J, Kablitz B, Wilde C, Mehta N, Cameotra SS. Appl.
Environ. Microbiol. 2002; 68: 6210.
29. Hitzeroth G, Vater J, Franke P, Gebhardt K, Fiedler H-P.
Rapid Commun. Mass Spectrom. 2005; 19: 2935.
30. Van Baar BLM. FEMS Microbiol. Rev. 2000; 24: 193.
31. Lay JO Jr. Mass Spectrom. Rev. 2001; 20: 172.
32. Erhard M, von Do hren H, Jungblut P. Nat. Biotechnol. 1997;
15: 906.
33. Welker M, Fastner J, Erhard M, von Do hren H. Environ.
Toxicol. 2002; 17: 367.
34. Stein T. Rapid Commun. Mass Spectrom. 2008; 22: 1146.
35. Gevers W, Kleinkauf H, Lipmann F. Proc. Natl. Acad. Sci.
USA 1969; 63: 1335.
36. Froshov O, Zimmer TL, Laland SG. FEBS Lett. 1070; 7:
68.
37. Gadow A, Vater J, Schlumbohm W, Palacz Z, Salnikow J,
Kleinkauf H. Eur. J. Biochem. 1983; 132: 229.
38. Laland SG, Zimmer TL. Essays Biochem. 1973; 9: 31.
39. Dorrestein PC, Kelleher NL. Nat. Prod. Rep. 2006; 23: 893.
40. Shaw-Reid CA, Kelleher NL, Losey HC, Gehring AM, Berg
C, Walsh CT. Chem. Biol. 1999; 6: 385.
41. Mazur MT, Walsh CT, Kelleher NL. Biochemistry 2003; 42:
13393.
42. McLoughlin SM, Kelleher NL. J. Am. Chem. Soc. 2004; 126:
13265.
43. Miller LM, Mazur MT, McLoughlin SM, Kelleher NL. Protein
Sci. 2005; 14: 2702.
44. Dorrestein PC, Blackhall J, Straight PD, Fischbach MA,
Garneau-Tsodikova S, Edwards DJ, McLoughlin SM, Lin
M, Gerwick WH, Kolter R, Walsh CT, Kelleher NL. Biochem-
istry 2006; 45: 1537.
45. Vollenbroich D, Mehta N, Zuber P, Vater J, Kamp RM.
J. Bacteriol. 1994; 176: 395.
Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 14931498
DOI: 10.1002/rcm
1498 J. Vater, C. Wilde and H. Kell