Solid supported chemical syntheses of both components of the lantibiotic lacticin 3147

Article abstract of DOI:10.1021/ja206017p

Lantibiotics are antimicrobial peptides produced by bacteria. Some are employed for food preservation, whereas others have therapeutic potential due to their activity against organisms resistant to current antibiotics. They are ribosomally synthesized and

Full text of DOI:10.1021/ja206017p

COMMUNICATION
pubs.acs.org/JACS
Solid Supported Chemical Syntheses of Both Components of the
Lantibiotic Lacticin 3147
Wei Liu, Alice S. H. Chan, Hongqiang Liu, Stephen A. Cochrane, and John C. Vederas*
Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2
S Supporting Information
b
ABSTRACT: Lantibiotics are antimicrobial peptides pro-
duced by bacteria. Some are employed for food preserva-
tion, whereas others have therapeutic potential due to their
activity against organisms resistant to current antibiotics.
They are ribosomally synthesized and posttranslationally
modified by dehydration of serine and threonine residues
followed by attack of thiols of cysteines to form monosulfide
lanthionine and methyllanthionine rings, respectively. Chemi-
cal synthesis of peptide analogues is a powerful method to
verify stereochemistry and access structureÀactivity rela-
tionships. However, solid supported synthesis of lantibiotics
has been difficult due to problems in generating lanthionines
and methyllanthionines with orthogonal protection and
good stereochemical control. We report the solid-phase
syntheses of both peptides of a two-component lantibiotic,
lacticin 3147. Both successive and interlocking ring systems
were synthesized on-resin, thereby providing a general
methodology for this family of natural products.
Figure 1. (A) Nisin A. (B) Lacticin 3147 A1 (1) and A2 (2) peptides.
The blue rectangles indicate the pentapeptide fragments that were
synthesized in solution phase (vide infra).
of the structural genes of lantibiotics enables the rapid genera-
tion of analogues for structureÀactivity relationship (SAR)
studies.11À14 An alternative approach to analogue generation,
which has provided fascinating insights into enzyme mechanism,
has relied on reconstitution of the biosynthetic enzymes and in
vitro mutasynthesis.15,16
acteriocins are antimicrobial peptides that are ribosomally
synthesized and exported by bacteria to eliminate competing
B
microorganisms.1,2 Some are unmodified peptides (typically
25 to 60 residues), but the others, such as the lantibiotics
(lanthionine-containing antibiotics), undergo extensive post-
translational modifications such as dehydration of serine and
threonine residues and monosulfide ring formation via Michael
addition of cysteine thiols on the resulting dehydroamino acid
moieties.3 This can result in formation of lanthionine (Lan) or
methyllanthionine (MeLan) residues that can be either non-
overlapping (successive) or interlocking on the peptide chain.
The dehydroalanine units derived from serine can also be
reduced to D-alanine residues in the peptide.4 Nisin A, first
discovered in 1928, has been widely used in preservation of dairy
products for almost 50 years without significant development of
resistance.5 It contains both successive rings (A, B, C) and
interlocking ones (D, E) (Figure 1A). Its potent broad spectrum
activity (often low nanomolar) against Gram-positive bacteria is
due to binding of the peptidoglycan precursor lipid II and
membrane pore formation.6,7 However, it is generally not
suitable for meat preservation or as a therapeutic agent because
it is unstable at neutral or basic pH and reacts readily with
nucleophiles such as water or thiols.8 Efforts to find more stable
compounds led to the isolation of two component lantibiotics
such as lacticin 3147 (Figure 1B)9 and the alkaline stable
haloduracins.10 As these peptides are gene-encoded, mutation
Nevertheless, there are limitations to bioengineering. The
structures of the modified precursor peptides may be incompa-
tible with the enzymatic machinery for posttranslational mod-
ification. As a result, production of the target peptide may be
reduced or even abolished. Even if the targeted compound is
formed in substantial quantity in bacterial hosts, lack of activity
can make detection and isolation of the product difficult. In
principle, total chemical synthesis of lantibiotics offers an alter-
native approach that can allow greater structural variation for
SAR investigations and production in sufficient quantities for
clinical studies. In the 1980s, Shiba and co-workers completed
the first solution-phase synthesis of a natural lantibiotic, nisin A.17
This work was a monumental achievement, but chemical synth-
esis of lantibiotics has proven difficult, and a variety of elegant
approaches have not yet led to total syntheses of such com-
pounds.18À21 Major obstacles include generation of protected
Lan and MeLan with good stereochemical purity bearing masking
Received: June 28, 2011
Published: August 17, 2011
r
2011 American Chemical Society
14216
dx.doi.org/10.1021/ja206017p J. Am. Chem. Soc. 2011, 133, 14216–14219
|

Journal of the American Chemical Society
COMMUNICATION
groups that are compatible with standard peptide synthesis and
that can be selectively removed to allow macrocyclic ring
formation. Our group has reported the solid-supported peptide
syntheses (SPPS) and activity of analogues of lacticin 3147 A2
peptide wherein the sulfur is replaced with oxygen22 or two
carbons23 and also where MeLan bridges are replaced with Lan
residues.24 The only natural lantibiotic to be synthesized using
SPPS is lactocin S, a single component antimicrobial that has no
MeLan bridges or interlocking rings in its structure.25 In the
present study, we describe facile solid supported syntheses of
both the A1 and A2 peptides (Figure 1B, compounds 1 and 2,
respectively) of lacticin 3147, a two-component lantibiotic.
These synergistic peptides and their analogues have considerable
therapeutic potential as they are stable at physiological pH.
Lacticin 3147 is active against Mycobacterium tuberculosis26 and
a variety of pathogens resistant to current antibiotics, such as
methicillin-resistant Staphylococcus aureus (MRSA) and vanco-
mycin-resistant enterococci (VRE).27 It has been incorporated
into a teat seal product for prevention and treatment of mastitis
in cows28 and is being investigated for use in oral hygiene
products.29
paration of orthogonally protected lanthionines, but we chose
aziridine ring-opening22 by cysteine derivatives as a rapid method
that provided very good stereochemical control (Scheme 1, see
Supporting Information for details). This approach is effective
for rapid generation of stereochemically pure Lan and MeLan
derivatives bearing distal protecting groups such as allyl and alloc,
as well as p-nitrobenzyloxycarbonyl (p-NZ) and p-nitrobenzyl
(p-NB) that can be selectively removed on-resin by palladium or
by tin chloride reduction, respectively. This allows construction
of interlocking rings (see below). For the synthesis of the lacticin
3147 A1 peptides, the aziridine opening afforded the necessary
derivatives of Lan 3b and MeLan 3a and 4. The absolute stereo-
chemistries of these compounds as well as of the Lan and MeLan
residues in 1 and 2 were confirmed by deprotection/hydrolysis,
conversion to volatile derivatives (dimethyl esters with bis-N-
pentafluoropropanoyl groups) and chiral gas chromatographyÀ
mass spectrometric (GCÀMS) analyses.30 Other isomers of
Lan and MeLan were prepared as standards. After the absolute
configurations of the two peptides were confirmed to be as
initially proposed,9 we commenced the synthesis of lacticin 3147
A2 (2).
Previous studies directed to chemical synthesis of lantibiotics
mentioned above offered a number of approaches to the pre-
The protected MeLan 3a was first loaded onto 2-chlorotrityl
polystyrene resin with the low substitution of 0.15 mmol gÀ1 to
3
give 5, which is crucial to avoid intermolecular coupling during
on-resin ring formation (Scheme 2). Then, alanine and protected
arginine were introduced using standard Fmoc-SPPS methodol-
ogy to give the linear tripeptide 6. The allyl and alloc groups of
the MeLan residue in 6 were removed by Pd(PPh3)4 and PhSiH3,
followed by removal of the Fmoc with 20% piperidine in DMF.
Compound 7 was then obtained by cyclization using PyBOP/
HOBt. Subsequently, rings B and A were constructed in a similar
fashion to give tricyclic product 8. An additional ten residues
from isoleucine to proline (residues 15 to 6) were then installed.
The N-terminal pentapeptide 10 was synthesized in solution due
to its heavily modified nature. Coupling of 10 to 9 on-resin was
followed by acidic cleavage from the resin with concomitant
removal of all acid labile protecting groups. The liberation of the
N-terminus as a primary enamine results in spontaneous hydro-
lysis to the desired R-keto amide to provide the lacticin 3147 A2
peptide 2.
Scheme 1. Synthesis of Protected Lan and
MeLan Derivatives 3a, 3b and 4a
a Reagents and conditions: (a) allyl chloride, NaHCO3, quantitative. (b)
MsCl, Et3N, 0 °C. (c) 3:3:1 TFA/DCM/MeOH, 0 °C. (d) DNs-Cl,
Na2CO3, 61% over three steps. (e) Fmoc-Cys-OH, BF3 Et2O, 0 °C,
3
The synthesis of the A1 peptide 1 began with the loading
40%. (f) Thioglycolic acid, DIPEA. (g) Alloc chloride, DIPEA, 44%. (h)
p-NBOH, p-TsOH. (i) TrtCl, Et3N, 60% over two steps. (j) MsCl, Et3N,
reflux. (k) 3:3:1 TFA/DCM/MeOH, 0 °C, p-NZCl, Na2CO3, 60%. (l)
of Fmoc-Lys(Boc)-OH onto Wang resin (0.15 mmol gÀ1),
3
followed by standard coupling of orthogonally protected
Fmoc-Cys-OH, BF3 Et2O, 46%.
MeLan 3a to give 11 (Scheme 3). The next six residues were
3
Scheme 2. Solid Phase Synthesis of Lacticin 3147 A2 (2)
14217
dx.doi.org/10.1021/ja206017p |J. Am. Chem. Soc. 2011, 133, 14216–14219

Journal of the American Chemical Society
COMMUNICATION
Scheme 3. Solid Phase Synthesis of Lacticin 3147 A1 (1)a
a The resin-bound peptides were protected on side chains at asterisk sites. The free peptide was not side-chain-protected.
then coupled on-resin sequentially, including the orthogonally
protected MeLan 4 to give linear peptide 12. After deprotection
of allyl and alloc groups on the first of the MeLan residues using
Pd(PPh3)4 and PhSiH3, ring D in 13 was synthesized through
on-resin amide formation in the presence of PyBOP and HOBt.
The following residue (Leu) was then installed. Next, the p-NZ
and p-NB groups in the second MeLan moiety were removed
simultaneously using SnCl2. Subsequent Fmoc-deprotection and
cyclization furnished the interlocking rings C and D in 14. SPPS
synthesis was continued with initial attachment of protected
Lan 3b. This was followed by linking 9 more residues needed for
ring B and then Pd deprotection of the distal Lan allyl and
alloc groups as before. Cyclization of the large ring B proceeded
smoothly, and attachment of three more residues gave the
tricyclic peptide 15. The N-terminal pentapeptide 16 containing
ring A was synthesized in solution because of its extensive and
densely packed modifications (see Supporting Information).
Coupling between fragments 15 and 16 on-resin was followed
by acidic cleavage with concomitant removal of all acid labile
protecting groups to yield the A1 peptide 1 of lacticin 3147.
After purification by reverse-phase HPLC, analysis of syn-
thetic A1 1 and A2 2 by MALDI-TOF MS showed peaks at
3305.3 and 2847.4 Da (M + H)+, respectively, which are identical
to those of the natural products. Further MS/MS studies
confirmed the correct sequence. The overall yields of A1 and
A2 were 1.4% and 1.0%, respectively. Each synthesis encom-
passed more than fifty steps, and therefore the yield of each
coupling and deprotection is greater than 95%. Each synthetic
peptide was tested in conjunction with its synergistic partner of
natural lacticin 3147 against the Gram-positive indicator organ-
ism Lactococcus lactis subspecies cremoris HP. As expected,
synthetic 1 and 2 showed biological activity indistinguishable
from the corresponding natural peptide isolated from the pro-
ducer organism, L. lactis subspecies lactis DPC 3147 (Figure 2).
In each case, both 1 and 2 show inherent antimicrobial activity,
but there is a strong synergism as indicated by the enhanced
inhibition zone in the area between the application of two
peptides. Our earlier studies with oxa- and bis des-methyl
analogues of 2 had shown that the antimicrobial activity of
this two peptide system is exquisitely sensitive to small alterations
in lanthionine ring geometry. For example, replacement of sulfur
by oxygen in 2 gives an analogue that retains independent
Figure 2. Spot-on-lawn tests for antimicrobial activity showing zone of
growth inhibition against L. lactis subspecies cremoris HP. The activities
of synthetic lacticin 3147 A1 and synthetic 3147 A2 were compared to
the corresponding natural peptide, shown in (i) and (ii) respectively. LC
results are shown in (iii).
antimicrobial activity against Gram-positive bacteria but lacks
the synergistic activity with the natural A1 peptide 1. Interest-
ingly, the analogue of 2 wherein MeLan bridges are replaced with
Lan (bis-desmethyl) shows the converse effect: it has no inherent
activity but displays synergistic activity with 1. These effects as
well as potent activity of the parent system that we have now
synthesized can be explained through dual modes of action,
namely, inhibition of peptidoglycan biosynthesis through bind-
ing of lipid II by each individual component and subsequent pore
formation by the ternary complex consisting of lipid II and both
peptides.31
The present synthesis of both components of a two-peptide
lantibiotic on solid support using standard Fmoc methodology
serves as a template for production of Lan and MeLan containing
peptides having either sequential or interlocking rings. This rapid
14218
dx.doi.org/10.1021/ja206017p |J. Am. Chem. Soc. 2011, 133, 14216–14219

Journal of the American Chemical Society
COMMUNICATION
and semiautomated methodology will facilitate the development
of high-throughput SAR studies of lantibiotics in cases that
are not readily accessible by enzymatic transformation either in
bacteria or in vitro.
(14) Field, D.; Hill, C.; Cotter, P. D.; Ross, R. P. Mol. Microbiol.
2010, 78, 1077.
(15) Levengood, M. R.; Knerr, P. J.; Oman, T. J.; van der Donk,
W. A. J. Am. Chem. Soc. 2009, 131, 12024.
(16) Goto, Y.; Li, B.; Claesen, J.; Shi, Y.; Bibb, M. J.; van der Donk,
W. A. PLoS Biol. 2010, 8, e1000339.
(17) Fukase, K.; Kitazawa, M.; Sano, A.; Shimbo, K.; Horimoto, K.;
Fujita, H.; Kubo, A.; Wakamiya, T.; Shiba, T. Bull. Chem. Soc. Jpn. 1992,
65, 2227.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental details, NMR
b
and GCÀMS data. This material is available free of charge via
(18) Paul, M.; van der Donk, W. A. Mini-Rev. Org. Chem. 2005,
2, 343.
the Internet at http://pubs.acs.org.
(19) Zhu, X. M.; Schmidt, R. R. Eur. J. Org. Chem. 2003, 20, 4069.
(20) Bregant, S.; Tabor, A. B. J. Org. Chem. 2005, 70, 2430.
(21) Narayan, R. S.; VanNieuwenhze, M. S. Org. Lett. 2005, 7, 2655.
(22) Liu, H.; Pattabiraman, V. R.; Vederas, J. C. Org. Lett. 2009,
11, 5574.
’ AUTHOR INFORMATION
Corresponding Author
john.vederas@ualberta.ca
(23) Pattabiraman, V. R.; Stymiest, J. L.; Derksen, D. J.; Martin, N. I.;
Vederas, J. C. Org. Lett. 2007, 9, 699.
(24) Pattabiraman, V. R.; McKinnie, S. M. K.; Vederas, J. C. Angew.
Chem., Int. Ed. 2008, 47, 9472.
(25) Ross, A. C.; Liu, H.; Pattabiraman, V. R.; Vederas, J. C. J. Am.
Chem. Soc. 2010, 132, 462.
(26) Carroll, J.; Draper, L. A.; O’Connor, P. M.; Coffey, A.; Hill, C.;
Ross, R. P.; Cotter, P. D.; O’Mahon, J. Int. J. Antimicrob. Agents 2010,
36, 132.
’ ACKNOWLEDGMENT
This work was supported by the Natural Sciences and
Engineering Research Council of Canada (NSERC), Alberta
Innovates Health Solutions, and the Canada Research Chair in
Bioorganic and Medicinal Chemistry. We thank Mark Miskolzie,
Randy Whittal, Jing Zheng and Don Morgan for the assistance
with spectral analyses.
(27) Galvin, M.; Hill, C.; Ross, R. P. Lett. Appl. Microbiol. 1999,
28, 355.
(28) Ryan, M. P.; Flynn, J.; Hill, C.; Ross, R. P.; Meaney, W. J. Dairy
Sci. 1999, 82, 2625.
(29) Lawton, E. M.; Ross, R. P.; Hill, C.; Cotter, P. D. Mini-Rev. Med.
Chem. 2007, 7, 1236.
(30) Kusters, E.; Allgaier, H.; Jung, G.; Bayer, E. Chromatographia
1984, 18, 287.
(31) Wiedemann, I.; B€ottinger, T.; Bonelli, R. R.; Wiese, A.; Hagge,
S. O.; Gutsmann, T.; Seydel, U.; Deegan, L.; Hill, C.; Ross, P.; Sahl, H.-G.
Mol. Microbiol. 2006, 61, 285.
’ ABBREVIATIONS
Abu-S-Ala, β-methyllanthionine; Ala-S-Ala, lanthionine; Boc, bu-
tyloxycarbonyl; Dhb, dehydrobutyric acid; DIPEA, N,N-diiso-
propylethylamine; DNs-Cl, 2,4-dinitrobenzenesulfonyl chloride;
Fmoc, fluorenylmethyloxycarbonyl; HOBt, 1-hydroxybenzotria-
zole; NMM, N-methylmorpholine; NMP, 1-methylpyrrolidinone;
Pmc, 2,2,5,7,8-pentamethylchroman-6-sulfonyl; p-NB, p-nitroben-
zylp-NZ, p-nitrobenzyloxycarbonyl; PyBOP, (benzotriazol-1-yloxy)-
tripyrrolidinophosphonium hexafluorophosphate; SPPS, solid phase
peptide synthesis; TFA, trifluoroacetic acid; TFFH, tetramethyl-
fluoroformamidinium hexafluorophosphate; Trt, trityl
’ REFERENCES
(1) Nes, I. F.; Yoon, S.; Diep, D. B. Food Sci. Biotechnol. 2007,
16, 675.
(2) Drider, D.; Rebuffat, S. Prokaryotic Antimicrobial Peptides: From
Genes to Applications; Springer Verlag: New York, 2011.
(3) Willey, J. M.; van der Donk, W. A. Annu. Rev. Microbiol. 2007,
61, 477.
(4) Cotter, P. D.; O’Connor, P. M.; Draper, L. A.; Lawton, E. M.;
Deegan, L. H.; Hill, C.; Ross, R. P. Proc. Natl. Acad. Sci. U.S.A. 2005,
102, 18584.
(5) Chatterjee, C.; Paul, M.; Xie, L.; van der Donk, W. A. Chem. Rev.
2005, 105, 633.
(6) Breukink, E.; Wiedemann, I.; van Kraaij, C.; Kuipers, O. P.; Sahl,
H.; de Kruijff, B. Science 1999, 286, 2361.
(7) Hasper, H. E.; de Kruijff, B.; Breukink, E. Biochemistry 2004,
43, 11567.
(8) Schneider, N.; Werkmeister, K.; Pischetsrieder, M. Food Chem.
2011, 127, 847.
(9) Martin, N. I.; Sprules, T.; Carpenter, M. R.; Cotter, P. D.;
Hill, C.; Ross, R. P.; Vederas, J. C. Biochemistry 2004, 43, 3049.
(10) Cooper, E.; McClerren, A. L.; Chary, A.; van der Donk, W. A.
Chem. Biol. 2008, 15, 1035.
(11) Ross, A. C.; Vederas, J. C. J. Antibiot. 2011, 64, 27.
(12) Moll, G. N.; Kuipers, A.; Rink, R. Antonie van Leeuwenhoek
2010, 97, 319.
(13) Suda, S.; Westerbeek, A.; O’Connor, P. M.; Ross, R. P.; Hill, C.;
Cotter, P. D. Chem. Biol. 2010, 17, 1151.
14219
dx.doi.org/10.1021/ja206017p |J. Am. Chem. Soc. 2011, 133, 14216–14219