Synthesis and biological activity of oxa-lacticin A2, a lantibiotic analogue with sulfur replaced by oxygen

Article abstract of DOI:10.1021/ol9025205

[Chemical Equation Presented] An oxidatively stable analogue 3 of laetiein 3147 A2 (2), wherein the sulfur atoms are replaced with oxygens, was synthesized using solution phase peptide synthesis and sequential on-resln cycllzations. Biological evaluation

Full text of DOI:10.1021/ol9025205

ORGANIC
LETTERS
2009
Vol. 11, No. 24
5574-5577
Synthesis and Biological Activity of
Oxa-Lacticin A2, a Lantibiotic Analogue
with Sulfur Replaced by Oxygen
Hongqiang Liu, Vijaya R. Pattabiraman, and John C. Vederas*
Department of Chemistry, UniVersity of Alberta, Edmonton, Alberta, Canada T6G 2G2
john.Vederas@ualberta.ca
Received August 23, 2009
ABSTRACT
An oxidatively stable analogue 3 of lacticin 3147 A2 (2), wherein the sulfur atoms are replaced with oxygens, was synthesized using solution
phase peptide synthesis and sequential on-resin cyclizations. Biological evaluation suggests that oxa-lacticin A2 (3) retains independent
antimicrobial activity against Gram-positive bacteria but lacks the synergistic activity with natural lacticin A1 that is characteristic of the
native lacticin A2 peptide.
Lantibiotics1 (e.g., nisin A, lacticin 3147) are bacterially
produced antimicrobial peptides (bacteriocins) that preserve
food and have potential as drugs due to their potent activity
against a wide range of Gram-positive bacteria, including
methicillin-resistant Staphylococcus aureus (MRSA) and
vancomycin-resistant Enterococci (VRE). Lantibiotics occur
in dairy products as well as other foods and are often
consumed as part of a human diet. They contain the unusual
amino acids, meso-lanthionine and (2S,3S,6R)-3-methyl-
lanthionine, as a result of post-translational modification. The
two-component lantibiotic, lacticin 3147, consists of peptides
A1 (1) and A2 (2) that display not only individual antimi-
crobial activity but also synergistic activity in nanomolar
concentration (Figure 1).2 It is postulated that the individual
activity results from direct binding of lacticin A1 or A2
peptides to the peptidoglycan biosynthesis prescursor, lipid
II, thereby inhibiting the formation of the bacterial cell wall.3
The synergistic activity is attributed to pore formation in the
cell membrane facilitated by a trimeric complex between A1-
lipid II and A2 peptide.4 The dual functionality of these
peptides makes lacticin 3147 several orders of magnitude
more potent than most traditional antibiotics such as
penicillin.2a The lack of toxicity of many lantibiotics to
mammals and their ability to avoid development of signifi-
cant resistance in pathogenic bacteria make them attractive
targets for development as antibiotics. However, the sulfur
atoms in lanthionine rings are vulnerable to oxidation to
sulfoxides, which renders many lantibiotics inactive.2a,5 To
address this problem and to examine structure-activity
(2) (a) Martin, N. I.; Sprules, T.; Carpenter, M. R.; Cotter, P. D.; Hill,
C.; Ross, R. P.; Vederas, J. C. Biochemistry 2004, 43, 3049–3056. (b)
Galvin, M.; Hill, C.; Ross, R. P. Lett. Appl. Microbiol. 1999, 28, 355. (c)
Lawton, E. M.; Ross, R. P.; Hill, C.; Cotter, P. D. Mini-ReV. Med. Chem.
2007, 7, 1236–1247.
(3) (a) Hasper, H. E.; Kramer, N. E.; Smith, J. L.; Hillman, J. D.;
Zachariah, C.; Kuipers, O. P.; de Kruijff, B.; Breukink, E. Science 2006,
313, 1636–1637. (b) Breukink, E.; de Kruijff, B. Nat. ReV. Drug DiscoVery
2006, 5, 321–332.
(1) For selected reviews see: (a) Chatterjee, C.; Paul, M.; Xie, L.; van
der Donk, W. A. Chem. ReV. 2005, 105, 633–684. (b) Cotter, P. D.; Hill,
C.; Ross, R. P. Curr. Protein Pept. Sci. 2005, 6, 61–75. (c) Willey, J. M.;
van der Donk, W. A. Annu. ReV. Microbiol. 2007, 61, 447–501. (d)
McAuliffe, O.; Ross, R. P.; Hill, C. FEMS Microbiol. ReV. 2001, 25, 285–
308.
(4) Wiedemann, I.; Bottiger, 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, 285–296.
(5) Wilson-Stanford, S.; Kalli, A.; Hakansson, K.; Kastrantas, J.;
Orugunty, R. S.; Smith, L. Appl. EnViron. Microbiol. 2009, 75, 1381–1387
.
10.1021/ol9025205  2009 American Chemical Society
Published on Web 11/18/2009

relationships, a number of research groups have examined
the chemical synthesis of lantibiotic analogues and frag-
ments.6 Only one total synthesis of a natural lantibiotic (nisin
A) has been reported using solution phase methods.7 Two
solid-supported syntheses of analogues of lantibiotics have
been recently described by us.6a,8
Figure 2. Retrosynthetic analysis of 3, oxa-lacticin 3147 A2.
ously reported by our group.10 The tert-butyl groups were
selectively removed under acidic conditions to give 5a and
5b, respectively (Scheme 1).
Figure 1. Lacticin 3147 component A1 (1) and A2 (2) and oxygen
analogue of A2 (3).
We now report the solid phase synthesis and biological
evaluation of oxa-lacticin A2 (3), an analogue of A2 (2)
wherein all sulfur atoms in the lanthionine and methyl-
lanthionine rings are replaced with oxygen atoms. Although
sulfur and oxygen have different single bond lengths to
carbon (1.80 vs 1.43 Å), the more acute bond angle of
C-S-C compared to C-O-C bonds (101.4° vs 111.4°)
compensates to make the overall length of meso-lanthionine
and its oxygen analogue, meso-3-(oxa)diaminopimelic acid
(meso-oxa-DAP), very similar (ca. 6% difference).9
Our synthetic strategy involved solid phase peptide
synthesis, followed by fragment coupling between a tricyclic
peptide on-resin and an N-terminal pentapeptide (1-5)
prepared in solution (Figure 2). The on-resin formation of
the “oxa-lanthionine” rings could be done via an intramo-
lecular cyclization using orthogonally protected meso-oxa-
DAP (or 3-methyl-oxa-DAP) precursors.8
Scheme 1
.
Synthesis of oxa-DAP and 3-Methyl-oxa-DAP
Precursors for SPPS
To initiate Fmoc solid-phase peptide synthesis (SPPS), the
protected (2R,3S,6S)-3-methyl-oxa-DAP derivative 5b was
attached to Wang resin with a low substitution of 0.1 mmol/
g. The low loading is crucial to avoid interstrand cross-
linking during on-resin cyclization of the deprotected oxa-
DAP (or 3-methyl-oxa-DAP) residue.11 Standard Fmoc SPPS
furnished the linear sequence of ring C (7) using PyBOP as
the coupling reagent. The selective removal of the allyl group
with Pd(PPh3)4 and PhSiH3 followed by Fmoc deprotection
with 20% piperidine afforded the precursor for on-resin
cyclization. The cyclization proceeded readily using PyBOP/
HOBt in 2 h (Scheme 2). A small sample of the resin was
treated with TFA/H2O/TIPS (95:2.5:2.5) to cleave the peptide
from it.
Analysis of this peptide by mass spectrometry demon-
strated that ring C of 8 was formed without any observable
dimer. The pNZ group was then removed under reductive
conditions with SnCl2 (3 × 45 min) in the presence of a
low concentration of acid (1.6 mM HCl/dioxane).12 Comple-
tion of the reaction was ascertained by MALDI-TOF MS
The tetra-orthogonally protected meso-oxa-DAP (4a) and
its 3-methyl derivative (4b) ((2R,3S,6S)-isomer) were pre-
pared using an aziridine ring-opening methodology previ-
(6) (a) Pattabiraman, V. R.; Stymiest, J. L.; Derksen, D. J.; Martin, N. I.;
Vederas, J. C. Org. Lett. 2007, 9, 699–702. (b) Ghalit, N.; Reichwein, J. F.;
Hilbers, H. W.; Breukink, E.; Rijkers, D. T. S.; Liskamp, R. M. J.
Chembiochem 2007, 8, 1540–1554. (c) Cobb, S. L.; Vederas, J. C. Org.
Biomol. Chem. 2007, 5, 1031–1038. (d) Ghalit, N.; Rijkers, D. T. S.;
Liskamp, R. M. J. J. Mol. Catal. A-Chem. 2006, 254, 68–77. (e) Paul, M.;
van der Donk, W. A. Mini-ReV. Org. Chem. 2005, 2, 23–37. (f) Narayan,
R. S.; VanNieuwenhze, M. S. Org. Lett. 2005, 7, 2655–2658. (g) Matteucci,
M.; Bhalay, G.; Bradley, M. Tetrahedron Lett. 2004, 45, 1399–1401.
(7) Fukase, K.; Kitazawa, M.; Sano, A.; Shimbo, K.; Horimoto, S.;
Fujita, H.; Kubo, A.; Wakamiya, T.; Shiba, T. Bull. Chem. Soc. Jpn. 1992,
65, 2227–2240.
(8) (a) Pattabiraman, V. R.; McKinnie, S. M. K.; Vederas, J. C. Angew.
Chem., Int. Ed. 2008, 47, 9472–9475. (b) Bergant, S.; Tabor, A. B. J. Org.
(10) Liu, H.; Pattabiraman, V. R.; Vederas, J. C. Org. Lett. 2007, 9,
4211–4214.
Chem. 2005, 70, 2430–2438
.
(9) (a) Glowka, M. L.; Parthasarathy, R. Acta Cryst. C 1986, C42, 620–
623. (b) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,
A. G.; Taylor, R. J. Chem. Soc., Perkin Trans 2 1987, S1–S19.
(11) Interstrand cross-linking and dimer formation is observed when
higher loading resin is used. We believe the pNZ urethane adjacent to the
activated carboxyl suppresses epimerization during macrolactamization.12
Org. Lett., Vol. 11, No. 24, 2009
5575

Scheme 2. Solid Phase Synthesis of Residues (6-29) of Oxa-lacticin 3147 A2
analysis after treatment of a small portion of the resin with
TFA to liberate the peptide.
between a proline and dehydroamino acid was sluggish and
required 7 h for completion. To avoid aggregation of the
long hydrophobic peptide chain, N-methylpyrrolidinone
(NMP) was used as the solvent for the coupling (Scheme
3).13
The final deprotection and cleavage of oxa-lacticin (3)
from the resin was effected with (97.5:2.5) TFA/TIPS.
Purification by reverse-phase HPLC and analysis by MALDI-
TOF MS showed a clean peak at 2799.2 Da (M + H)
corresponding to the mass of the desired oxa-lacticin A2.
Detailed MS/MS studies confirmed the correct sequence and
connection of the ether rings in 3. The overall yield was
0.3% for ca. 53 steps (22 deprotections and couplings, 3
macrocyclizations, 3 pNZ removals, and 3 allyl deprotec-
tions). This corresponded to just under 90% per step, but as
the attachment of 12 proceeded in low yield, it is likely that
each deprotection/coupling was >95%.
Fmoc-SPPS was continued to introduce the required amino
acids for the formation of ring B. After removal of the allyl
and Fmoc groups as before, the cyclization was allowed to
proceed on-resin with PyBOP/HOBt for 2.5 h. Cleavage of
a small portion of the peptide as before followed by MALDI-
TOF mass spectrometry showed a major peak corresponding
to the bicyclic product 9. After removal of the pNZ group
with SnCl2, the linear precursor to ring A was prepared by
SPPS in a manner similar to the formation of rings C and
B. Removal of the allyl and Fmoc protecting groups and
on-resin cyclization gave the tricylic compound 10. Mass
spectrometric analysis of a small fraction of the peptide
cleaved from resin confirmed the generation of tricyclic 10
as the major product. More importantly, as before, no
dimerized or oligomerized peptides were observed. After
removal of the pNZ group, residues 6-15 were introduced
using standard Fmoc SPPS to furnish 11.
Preliminary biological evaluation of oxa-lacticin A2 (3)
was done in conjunction with natural lacticin A1 (1) against
a panel of Gram-positive bacteria.2a Oxa-lacticin A2 (3)
exhibited only independent antibacterial activity (L. lactis
subsp. cremoris HP and Leuconostoc mesenteroides Y105)
Completion of the synthesis was done by on-resin coupling
of the pentapeptide 12, which was obtained by a method
previously reported by us.6 This difficult fragment coupling
(12) Isidro-Llobet, A.; Guasch-Camell, J.; Alvarez, M.; Albericio, F.
Eur. J. Org. Chem. 2005, 14, 3031–3039.
(13) Synthesis Notes 3.5. NoVaBiochem Catalog; EMD Biosciences: San
Diego, CA, 2006-2007.
5576
Org. Lett., Vol. 11, No. 24, 2009

the B/C ring portion of the lantibiotic analogue.1c,3,4 This
would result in inhibition of cell wall biosynthesis and the
observed antimicrobial activity for the lone peptide. The
methyl groups on carbons adjacent to sulfur in 2 are essential
for the independent activity,8a and this may also be true for
3.
Scheme 3
.
Fragment Coupling of Residues (1-5) to Complete
the Synthesis of Oxa-lacticin 3147 A2
Figure 3. Spot-on-lawn tests for antimicrobial activity showing zone
of growth inhibition against L. lactis subsp. cremoris HP: (a) the
synergistic and inherent activity of natural A1 (1) and A2 (2) [top];
the inherent independent activity of natural A1 (1) and oxa-A2 (3)
[bottom]; (b) the activity of A1(1) and oxa-A2 (3) at differing
distances (no synergism).
or no biological activity [five other Gram-positive bacteria]
(Figure 3a). When compared to natural A2, oxa-A2 showed
a reduced spectrum of activity. A serial dilution assay of
oxa-A2 (3) and natural A2 (2) indicated that the inherent
antimicrobial activity of the oxygen analogue was ap-
proximately 20 times less than natural lacticin A2. Interest-
ingly, oxa-A2 (3) showed no synergistic antimicrobial
activity with natural A1 (1) against any of the organisms
tested (Figure 3b). In contrast, our previous work with
bis(desmethyl) lacticin A2,8a which has the two ꢀ-methyl-
lanthionine bridges replaced by lanthionines, demonstrated
that the desmethyl analogue bearing sulfurs has very little
independent activity but shows a strong synergistic effect
with natural A1 (1).
These results support the hypothesis that the two-peptide
lantibiotic lacticin 3147 has at least two modes of action.8a
In the natural system, synergistic activity results from pore
formation in the cell membrane via the interaction of the
A2 peptide with the lacticin A1-lipid II complex.4 The
distance between carbons 2 and 6 in lanthionine is about
0.7 Å longer than the corresponding distance in oxa-DAP.
Thus, the sulfur-containing rings in natural 2 will be slightly
larger than those in the unnatural oxygen derivative 3. The
loss of synergistic activity for oxa analogue 3 presumably
results from poor recognition of the A1-lipid II complex.
This may be due to an additive effect of the three modifica-
tion sites or perhaps to differences in the size of sulfur vs
oxygen (van der Waals radius 1.80 vs 1.52 Å) or their
electrostatic properties (Pauling scale electronegativity 2.58
vs 3.44). However, the small conformational changes may
be tolerated, for the direct binding to lipid II, possibly by
In conclusion, multiple on-resin cyclizations using trior-
thogonally protected meso-oxa-DAP and 3-methyl-oxa-DAP
strategy together with fragment coupling were successfully
utilized for the chemical synthesis of oxa-lacticin A2. These
unusual oxygen bridged bisamino acids could serve as
oxidatively stable conformational constraints in medically
relevant peptides. The biological evaluation of oxa-lacticin
A2 supported the proposal of two modes of action for the
two component peptides, lacticin 3147 A1 and A2.
Acknowledgment. We thank Randy Whittal, Jing Zheng,
Lisa Nicolai, and Avena Ross (University of Alberta) for
help with mass spectrometry studies and assistance with
biological testing. This work was supported by the Natural
Sciences and Engineering Research Council of Canada
(NSERC), the Canada Research Chair in Biooganic and
Medicinal Chemistry, and the Alberta Heritage Foundation
for Medical Research (AHFMR) (studentship to H.L.).
Supporting Information Available: Experimental pro-
cedures and spectral data for all compounds synthesized,
HPLC trace, and MS/MS analysis as well as biological
testing assay. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL9025205
Org. Lett., Vol. 11, No. 24, 2009
5577