Biomimetic synthesis of lantibiotics

Article abstract of DOI:10.1002/(SICI)1521-3765(20000417)6:8<1455::AID-CHEM1455>3.0.CO;2-M

The lantibiotics are a class of highly posttranslationally modified small peptide antibiotics containing numerous lanthionine and dehydroamino acid residues. We have prepared peptides containing multiple dehydroamino acids and cysteine residues in order t

Full text of DOI:10.1002/(SICI)1521-3765(20000417)6:8<1455::AID-CHEM1455>3.0.CO;2-M

FULL PAPER
Biomimetic Synthesis of Lantibiotics
Sarah Burrage,[a] Tony Raynham,[b] Glyn Williams,[b] Jonathan W. Essex,[a] Carl Allen,[a]
Marianne Cardno,[a] Vinay Swali,[a] and Mark Bradley*[a]
Abstract: The lantibiotics are a class of
highly posttranslationally modified
small peptide antibiotics containing nu-
merous lanthionine and dehydroamino
acid residues. We have prepared pepti-
des containing multiple dehydroamino
acids and cysteine residues in order to
probe the biomimetic synthesis of the
lantibiotics from their precursor pepti-
des. A novel synthetic methodology was
developed to allow the synthesis of
multiple dehydroamino acid containing
peptides. Cyclisations were rapid, quan-
titative and regiospecific. Remarkably
the peptide sequences alone appear to
contain sufficient information to direct a
series of stereo- and regiospecific ring
closures. Thus both the two linear pep-
tides for the B and E-rings closed
stereoselectively. In the case of the
A-ring precursor peptide which con-
tained two dehydroamino acids, cyclisa-
tion was again totally regioselective,
although not totally stereoselective.
Keywords: biomimetic synthesis
´
lantibiotics ´ peptides ´ solid-phase
synthesis
tional modifications.[9] Their direct ribosomal origin was first
indicated by the inhibition of lantibiotic production by
inhibitors of both RNA and protein synthesis. Ingram,
inspired by the structures of the lantibiotics, proposed that
certain serine and threonine residues were initially dehydrat-
ed to dehydroalanines and dehydroaminobutyrates, respec-
tively. He further postulated that a series of subsequent
stereospecific Michael additions yielded meso-(2S,6R)-lan-
Introduction
The lantibiotics are a family of polycyclic peptides produced
by a broad range of bacteria, including Bacillus, Lactococcus,
Streptomyces and Staphylococcus.[1] The first lantibiotic to be
discovered was nisin in 1928,[2] with its structure being
determined some 40 years later.[3] This was followed by the
structural assignment of subtilin, a related lantibiotic.[4] Since
this time many more lantibiotics have been discovered,
isolated and characterised (Figure 1). These bacteriocins
typically contain unusually high proportions of nonproteino-
genic amino acids often accounting for one third of the total
residues. Thus nisin and subtilin[5, 6] each contain eight
unnatural amino acids, including two dehydroalanines, one
dehydroaminobutyrate and five lanthionine residues. De-
tailed NMR studies on some of the lantibiotics have been
reported,[7] while elegant total or partial synthesis of some
members of this family has been achieved.[8]
thionines and (2S,3S,6R)-3-methyllanthionines (Figure 2),[10]
a
prediction that was confirmed some 20 years later when the
primary transcripts of the structural genes were determined.[9]
The dehydration of serine/threonine residues is thought to
occur immediately after the primary transcript is released
from the ribosome; indeed the existence of a multiply
dehydrated precursor has been confirmed by ion-spray mass
spectrometry[11] and amino acid analysis in the case of Pep5. It
is proposed that nucleophilic addition of the cysteine thiol
onto the a,b-unsaturated amino acid produces an ªenol(ate)º
intermediate which is selectively protonated affording the
observed (2S,6R) stereochemistry.
Biosynthetically, the lantibiotics are ribosomally derived
peptides which have been subjected to extensive posttransla-
[a] Prof. M. Bradley, Dr. S. Burrage, Dr. J. W. Essex, C. Allen,
Dr. M. Cardno, V. Swali
Lanthionine formationÐregio- and stereospecificity: Since
the mature forms of the lantibiotics contain dehydroamino
acids and multiple lanthionines, the rationality of both regio-
and stereospecificities of cyclisation is raised. In addition, the
relative stereochemical orientation of the meso-lanthionine
varies within the lantibiotic family. The regio- and stereo-
selectivities observed may be the product of enzyme catalysis,
or could simply result from the natural conformation of the
peptide. The work reported here was aimed at unravelling this
process and involved the synthesis of peptides containing
Department of Chemistry, Southampton University
Highfield, Southampton, SO17 1BJ (UK)
E-mail: mb14@soton.ac.uk
[b] Dr. T. Raynham,[‡] Dr. G. Williams
Roche Discovery Welwyn
Broadwater Road, Welwyn Garden City, Herts, AL7 3AY (UK)
[ ] Current address:
‡
Cambridge Discovery Chemistry Ltd
(Formerly Cambridge Combinatorial Ltd)
The Merrifield Centre, 12 Rosemary Lane, Cambridge CB13LQ (UK)
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1455 $ 17.50+.50/0
1455

M. Bradley et al.
FULL PAPER
Leu
Ala
Gly
Gln
Dhb
S
Dha
S
B
D
Glu
Leu
Abu
Ala
Gln
Phe Leu
Asn
C
A
Lys
Lys
COOH
Ile
Ala
Ala
Dha
Leu Abu
Trp Lys
Val
Ala
H2N
Abu
Ala
Ala Abu
E
Pro Gly
S
S
S
subtilin
Leu
Ala
Met
Gly
S
Dha
Leu
D
S
Gly
Ile
Ala
Abu
Lys
His
A
C
B
Abu
Ala
His
Lys
Val Dha COOH
Ala
Ser
Ile
Asn Met
Ala
Ile
H2N
Dhb
Lys Abu
Ala Abu
Ala
Ala
E
Pro
Gly
S
S
S
nisin
Phe
S
Lys
Ala
Leu
Ala
S
Ile Ala
Ala Lys
Dhb Gly
H2N
Ala
Gly
Abu
Pro
Ala
Tyr
NH
Ala
S
epidermin
Phe
Asn Ala
S
S
Ala
S
Lys
O
COOH
Ala
Val
Abu
Ala
Gly
Arg
Ala
Pro Ala Ile
Gly
Ala
Gln
Ala
Gln
Dhb
Lys
Arg
Ala Dhb
Leu Phe
Leu
Lys
Ala
Lys
Gly
O
Val
Asn
Lys
S
Pep5
Lys
Figure 1. Some typical lantibiotics.
HS
O
O
O
O
H
N
H
N
OH
O
HS
HS
H2N
N
OH
N
H
dehydration
N
H
N
H
O
O
(1)
O
O
O
O
N
H
N
H
N
H
N
H
?
O
O
O
O
S
Michael
addition
O
O
H
N
H
N
H2N
N
OH
N
H
S
N
H
N
H
S
O
protonation
subtilin B-ring (4)
N
H
N
H
N
H
N
H
O
O
O
HS
O
O
O
O
H
N
H
N
Figure 2. Ingramꢁs proposal for the formation of the unusual amino acids.
OH
N
H
N
H
N
H
O
O
O
NH
(2)
CONH2
H2N
H2N
?
NH2
O
S
single and multiple dehydroalanine and cysteine residues for
biomimetic cyclisation studies. Evaluation of the cyclic
products would give valuable information concerning the
actual biosynthetic route to the lanthionine residues.
O
O
N
H
N
H
N
OH
N
H
N
H
NH
H
O
O
O
CONH2
NH2
O
The B- and E-rings of subtilin and nisin (Figure 3) were
selected as the first targets for cyclisation studies for a number
of reasons. Firstly, regiochemistry was not an issue because the
precursor peptides contain only one dehydroamino acid.
Secondly, the central residues of the B-ring (Pro ± Gly) are
known to promote b-turns possibly favouring cyclisation. The
B-ring has also been previously studied[12] and would allow us
to determine if solid-phase based cyclisations would mimic
those in solution, an important future issue with regards to the
synthesis of these complex precursor peptides. The E-ring
peptide would allow us to determine if the turn element was
essential for cyclisation, as well as giving a competing lysine
residue of interest in a number of lantibiotics which contain
the lysinoalanine motif.
subtilin E-ring (5)
Figure 3. Proposed biomimetic B- and E-ring closures.
Having developed the synthetic methodology to make
these relatively simple mono-dehydroamino acid containing
peptides, the methodology could then be extended to prepare
the A-ring precursor (Figure 4), containing two dehydroami-
no acids and a single thiol. This would allow monitoring of
both the regio- and stereoselectivity of cyclisation, and could
result in the formation of up to four possible cyclic products
(two regioisomers each containing either d,l or l,l-lanthio-
nines).
1456
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1456 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Lantibiotics
1455±1466
was prepared in 14% yield from Boc-Leu-Ser-Ala-OMe. In
this case the serine residue was used without side chain
protection throughout the synthesis without mishap. The
dehydroamino acid containing peptides 7 and 8 were then
available for coupling to the solid-phase linked peptides 9 and
10, which were synthesised with standard solid-phase Fmoc
chemistry,[23, 24] to give the full length linear protected peptides
11 and 12. Following deprotection, peptides 1, 2 and 13 were
ready for solution and solid-phase cyclisation studies.
NH2
O
OH
O
H
N
O
O
H
N
H
N
OH
H2N
N
H
N
H
N
H
(3)
O
O
O
O
SH
A-ring precursor
biomimetic cyclisation
lantibiotic A-ring ? (6)
Figure 4. Proposed biomimetic A-ring cyclisation.
Synthetic routes to dehydroamino acid containing peptides:
The chemistry of dehydroamino acids is well documented and
has been extensively reviewed.[13] Methods of synthesis often
rely on the activation of serine or threonine residues for
example with dichloroacetyl chloride or tosyl chloride fol-
lowed by elimination with triethylamine or DABCO, respec-
tively.[14, 15] They also include phosphonium activation of
serine and elimination (Ph3P/DEAD)[16] or EDC activation
followed by CuCl-mediated elimination.[17] On the solid phase
fewer methods have been reported. Of principal importance
are the coupling of preformed dehydroamino acid N-carbox-
yanhydrides (NCAꢁs),[18] the solid-phase Hofmann elimina-
tion of quaternary amines[19] and the attachment of peptides
by a cysteine side chain followed by oxidation to the sulfone
and elimination.[20] The stepwise incorporation of dehydroa-
mino acids residues into peptides by either a solid-phase or
solution route is also often difficult. Dehydroamino acids
couple very poorly[13, 21] and are inherently reactive towards
nucleophiles (including amines). Protected peptides contain-
ing dehydroamino acid residues are always prone to decom-
position, although they appear much more robust when
unprotected, while N-terminal dehydroamino acids decom-
pose to give the keto amide and ammonia.[13] Clearly, if such
residues are to be incorporated into peptides they should be
introduced as masked residues which can be converted to the
a,b-unsaturated amino acids as late as possible in the
synthetic scheme, or inserted within small peptide fragments
such that the exposed termini couple efficiently.
Trt
S
O
H
N
O
O
Wang
linker
Boc-NH
H N
O
N
+
OH
2
N
H
N
H
N
H
(9)
O
O
O
O
(7)
i)
R2
S
O
H
N
O
O
H
N
R1-NH
O-R3
N
N
H
N
H
N
H
O
O
O
O
(11) R1 = Boc, R2 = Trt,
3 = Wang linker-resin
(13) R1 = Boc, R2 = H,
3 = Wang linker-resin
(1) R1 = H, R2 = H, R3 = H
O
ii)
iv)
R
iii)
R
S
O
H
N
O
H
N
R1-NH
O R3
N
N
N
H
N
H
O
H
O
O
O
subtilin B-ring
(4a) R1 = Boc,
iii)
R
3 = Wang linker-resin
(4) R1 = H, R3 = H
Scheme 1. Subtilin B-ring biomimetic synthesis. i) PyBrop, DMAP, DI-
PEA, CH2Cl2; ii) 2% TFA, 2% TIS, CH2Cl2; iii) 95% TFA, CH2Cl2;
iv) pH 8.0, 50mm NEt3 ´ AcOH or 5% NMM, DMF for (13).
Trt
S
O
O
O
O
H
N
H
N
HMPB
O
Boc-NH
CO2H
+
H N
2
N
linker
N
N
H
H
H
O
O
(10)
O
(8)
CONH2
i)
NH-Boc
R2
S
O
O
H
H
H
N
O R4
R1-NH
N
N
N
N
H
N
H
O
H
O
O
O
CONH2
NH-R3
(12) R1 = Boc, R2 = Trt, R3 = Boc,
Results and Discussion
R4 = HMPB linker-resin
ii)
iii)
(2) R1 = H, R2 = H, R3 = H, R4 = H
Synthesis of dehydroamino acid containing peptides: The
dehydroamino acid and thiol containing peptides synthesised
for this study were:
H-Leu-Dha-Pro-Gly-Cys-Val-Gly-OR
H-Leu-Dha-Ala-Asn-Cys-Lys-Ile-OH
H-Lys-Dha-Glu-Dha-Leu-Cys-Ala-OH
S
O
O
O
H
H
N
N
OH
N
N
N
(1)
(2)
(3)
H
H
H
NH
O
O
O
H2N
CONH2
NH2
O
(5)
subtilin E-ring
Scheme 2. Subtilin E-ring biomimetic synthesis. i) DIC, HOBt, CH2Cl2/
DMF (9:1); ii) 50% TFA, 5%TIS, CH2Cl2; iii) pH 8.0, 50mm NEt3 ´ AcOH.
(R ˆ H or linker-resin) (1) ˆ B-ring precursor
(2) ˆ E-ring precursor
(3) ˆ A-ring precursor
Peptides 1 and 2 were prepared by a fragment coupling
procedure as outlined in Schemes 1 and 2. The crucial
dehydroamino acid containing peptide Boc-Leu-Dha-Pro-
OH (7) was synthesised in 23% yield from the tripeptide Boc-
Leu-Ser(Bzl)-Pro-OMe,[22] by hydrogenation, followed by
dehydration (CuCl/EDC)[15] and ester hydrolysis.[22] Using a
similar approach the tripeptide Boc-Leu-Dha-Ala-OH (8)
Peptide 3 (see Figure 4) containing two dehydroamino
acids and one cysteine residue was prepared by a novel
method, developed for the incorporation of multiple dehy-
droamino acids into peptides.[25] This was achieved by the
synthesis of peptides containing S-methylcysteine and the
pyrolytic elimination of the S-methylcysteine sulfoxides.[25]
The use of S-methylcysteine obviated the need for side chain
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1457 $ 17.50+.50/0
1457

M. Bradley et al.
FULL PAPER
protection, and if desired, Fmoc/tBu orthogonality could be
maintained throughout the synthesis. It was envisaged that
oxidation and elimination would be performed immediately
prior to global deprotection. As an initial trial study the
tripeptide Boc-Leu-Cys(Me)-Pro-OMe (14) was assembled in
solution with standard Boc chemistry in 34% overall yield.
Addition of peptide 14 to a solution of cold sodium
metaperiodate in dioxane (1.1 equiv) gave the diastereomeric
sulfoxides 15 in excellent yield and purity. Pyrolytic elimi-
nation was found to be solvent dependent; the best results
were obtained by refluxing the sulfoxides 15 overnight in
dioxane. Purification by column chromatography led to
isolation of the protected tripeptide 16 in 88% yield
(Scheme 3). Elimination also occurred on treatment of the
Boc protected sulfoxide tripeptide 15 with DBU in MeOH,
whilst simultaneous elimination and saponification was ac-
complished in high yield by 1m NaOH.
which was deprotected to give the desired peptide 3. It was
important that oxidation, elimination, deprotection and
coupling was accomplished in a single day due to the observed
lability of the protected dehydoamino acid containing pep-
tides.
Biomimetic cyclisations
Synthesis of the B-ring (Scheme 1): The linear peptide 1 was
analysed by HPLC following cleavage from the solid phase
and was shown to be essentially pure (>95%) by HPLC
thereby giving confidence for the subsequent solid-phase and
solution based cyclisations. Thus the trityl group was removed
from 11 by repeated washes with 2% TFA/2% TIS/96%
CH2Cl2 to give 13. Quantification of the thiol in peptide 13
was attempted by the Ellman test.[26] However, this assay gave
a much lower substitution than expected (0.05 mmol per g cf.
0.25mmol per g), and lead to the development of a new test
which was designed to monitor the consumption of the
dehydroamino acid during cyclisation. This was based on the
very well known addition of thiols onto dehydroamino acids.
Thus the peptidyl resin 13 was treated with a large excess of b-
mercaptoethylamine allowing the rapid addition of the thiol
to the a,b-unsaturated amino acid. (A small portion of the
resin was treated with TFA and the addition confirmed by ES-
MS and HPLC). A quantitative ninhydrin assay was then
carried out allowing solid-phase monitoring of the cyclisa-
tion.[27] Cyclisation was carried out with 5% NMM in DMF
and monitored as described above. The reaction was complete
within 1.5 h furnishing the cyclic peptide 4a. This was cleaved
and purified by RP-HPLC to give 4 which was fully
characterised by NMR (see below).
O
O
H
H
N
N
i)
BocNH
N
BocNH
Me
N
O
MeS
O
S
O
O
OMe
OMe
(15)
O
iii)
(14)
ii)
O
O
H
N
H
N
iii)
BocNH
N
BocNH
N
O
O
(16)
O
O
OH
OMe
(7)
Scheme 3. Thioether approach to dehydroamino acids. i) NaIO4 (aq),
dioxane (99%); ii) dioxane, reflux (88%) or DBU, MeOH (96%); iii) 1m
NaOH, MeOH (95%).
This methodology was extended to the synthesis of peptide
3 (Scheme 4) which contains two dehydroamino acids. Thus
the desired precursor peptide Boc-Lys(Boc)-Cys(Me)-
Glu(OtBu)-Cys(Me)-Leu-OMe (17) was obtained by frag-
ment condensation of Boc-Lys(Boc)-Cys(Me)-OH (18) with
the tripeptide H-Glu(OtBu)-Cys(Me)-Leu-OMe (19) medi-
ated by EDC and HOBt in an overall yield of 36% from the
starting amino acids. Peptide 17 was oxidised to give 20 which
was treated with 1m NaOH in MeOH to facilitate two
eliminations and ester hydrolysis in under 2 h to give 21.
Subsequent coupling of 21 to H-Cys(Trt)-Ala-OtBu gave 22
The cyclisation of the B-ring in solution was undertaken to
confirm the results above. The synthesis was identical to that
above and peptide 9 was cleaved and deprotected from the
resin prior to cyclisation to give 1. Cyclisation to give 4 was
carried out in 50mm triethylammonium acetate (pH 8) and
was monitored by RP-HPLC by using a C8 column (C18
columns did not separate the product and starting peptide).
The cyclisation was found to proceed cleanly with complete
consumption of starting material in under 10 min as deter-
mined by UV/Vis analysis of the cyclisation reaction and
HPLC analysis, following removal of aliquots and quenching
with acid. A single product,
identical to that observed in
the solid-phase cyclisation was
observed by RP-HPLC. The
cyclisation was also performed
in deuterated buffer and incor-
NHBoc
O
NHBoc
O
OtBu
OtBu
O
O
ii)
O
O
H
H
H
H
N
N
OMe
N
N
OMe
BocNH
i)
N
N
BocNH
N
N
H
SMe
H
H
SMe
H
O
O
(17)
O
O
O
O
SMe
SMe
O
(20)
O
poration of
a single nonex-
Boc-Lys(Boc)-Cys(Me)-OH (18)
+ H-Glu(OtBu)-Cys(Me)-Leu-OMe (19)
iii)
changeable deuteron was ob-
served by ES-MS confirming
that cyclisation had occurred
with deuterium inclusion at the
newly generated alpha centre.
NHBoc
NHBoc
O
O
O
OtBu
OtBu
O
N
iv)
BocNH
O
O
O
Me
H
H
N
H
N
H
H
v)
N
OtBu
N
OH
(3)
BocNH
N
N
N
H
N
H
N
H
H
H
O
O
O
O
O
O
(21)
O
S-Trt
(22)
Synthesis
(Scheme 2): The E-ring cyclisa-
tion was similarly investigated.
of
the
E-ring
Scheme 4. Synthesis of a pentapeptide containing two dehydroalanine residues. i) EDC, HOBt, NEM, CH2Cl2,
75%; ii) NaIO4 (aq), dioxane (99%); iii) a) DBU, MeOH (86%), 1m NaOH, MeOH (92%) or b) 1m NaOH,
MeOH (93%); iv) H-Cys(Trt)-Ala-OtBu (21), EDC, HOBt (72%); v) 50% TFA, 2% TIS, CH2Cl2.
1458
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1458 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Lantibiotics
1455±1466
The linear peptide 2 was prepared in an analogous manner to
the B-ring precursor. Cyclisation gave a single product 5
which was fully characterised by NMR studies (see below)
and was determined to be the desired lanthionine cyclic
peptide, clearly demonstrating that the turn element in the
B-ring is unnecessary for cyclisation to take place. Again
cyclisation was rapid and experiments in deuterated buffer
showed the expected incorporation of a single deuterium
atom into the cyclic product.
istry was determined as meso from these studies in accordance
with the literature.[12]
The E-ring: 1-D NMR analysis verified the presence of only
one compound and further 2-D NMR analysis enabled a full
assignment of the cyclic product to be obtained. The loss of
the olefinic protons, the splitting of the NH resonance
(dehydroamino acid NHꢁs are singlets) and the gain of a
CHa proton were all observed as expected. In contrast to the
B-ring, only a single conformer was observed. A detailed
examination of the ring was undertaken by COSY, TOCSY
and ROESY NMR spectroscopy, allowing the sequential
assignment of the amino acids (also confirmed by Edman
sequencing). The stereochemistry of the lanthionine residue
was examined by using the modified distance geometry
program DIANA[28] and the NOE constraints obtained above
measured from the spectrum. The best structure gave the
expected meso-lanthionine, although there was not a sub-
stantial difference between the two target functions (4.9
versus 3.8). The pattern of coupling constants (3JNH±CHa) was
also analysed but again no concrete conclusions could be
made with regard to the lanthionine stereochemistry. Model-
ling studies (MacroModel 5.0[29]) on the enolate suggested
that protonation from the surrounding solvent would be most
likely to give the d-amino acid with protonation taking place on
the less hindered face opposite to that of the attacking sulfur.
Synthesis of the A-ring (Figure 4): The linear A-ring pre-
cursor peptide 3 was produced in good yield and purity (86%
by RP-HPLC) (Scheme 4). Considering the problems of
stability, the peptide was cyclised crude. The reaction was
again found to be complete within 10 min but this time
produced two products in a 3:1 ratio (RP-HPLC). These were
separated by RP-HPLC and their structures determined by
NMR, as detailed below, to be the stereo- rather than
regioisomers of the natural A-ring.
NMR and modelling studies
The B-ring: The samples of cyclic peptide produced both on
the solid phase and in solution were identical by RP-HPLC
and NMR. The presence of the Lan-Hb at d ˆ 2.96, 2.86 and
2.60 could be clearly detected (Figure 5). The appearances of
resonances corresponding to the amide NHs and the loss of
the olefinic resonances as well as a full spectral assignment
confirmed that cyclisation had occurred. Two ring conformers
were observed which interconverted on the NMR timescale,
but which coalesced when recorded at 310 K. NOE studies
were performed on the B-ring sample. Most importantly an
NOE was observed between the Pro-Ha and the Lan1-Ha
indicating that the proline amide bond was cis configured as
expected. Modelling indicated that the distance between
these protons was about 2.4 Š. The lanthionine stereochem-
The A-rings: A series of 1-D and 2-D NMR experiments
showed that both of the new compounds were cyclic. Assign-
ment was facilitated by TOCSY experiments and confirmed
that each was a single diastereoisomer. The existence of a
cyclic peptide in both cases was confirmed by a number of
features in the 1-D NMR spectra, notably one set of
resonances corresponding to Dha-Hb protons and Dha-NH
were missing and there was the expected resonance for the
Lan-NH. There were also signals corresponding to the
Figure 5. 500 MHz NMR spectrum ([D6]DMSO, 310 K) of the B-ring 4 of subtilin.
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1459 $ 17.50+.50/0
1459

M. Bradley et al.
FULL PAPER
both Glu-Ha and Glu-NH.
These NOEs again can only
arise from a Dha at position 4.
A strong NOE was observed
between Dha-Hb and Leu-NH,
confirming that dehydroalanine
was at position 4 in the peptide.
In summary, the presence of
several NOEs including those
between Lan2-NH and Lys1-Ha
and Dha4-Hb' and Leu5-NH al-
lowed the regiochemical assign-
ment of the major product of
cyclisation.
Regiochemistry of the minor
product
The minor product was also
analysed by a NOESY experi-
ment. The NOE observed be-
tween the Lan-NH and Leu-Ha
Figure 6. 500 MHz NMR spectrum ([D6]DMSO, 303 K) of the major product formed by cyclisation of peptide
H-Lys1-Lan2-Glu3-Dha4-Leu5-Lan6-Ala7-OH (3)ÐA-ring of subtilin.
again facilitated the full assign-
presence of only a single dehydroalanine residue and the
characteristic ABX resonances observed for the Lan-Hb were
found just below d ˆ 3.0, clearly displaying the presence of
four protons (Figure 6).
ment of the lanthionine protons. A larger number of NOEs
were observed for this peptide than for the major isomer,
including one between the Lan2-Hb and Lan6-NH which
indicated that a cyclic peptide had been formed. Investigation
of the NOEs associated with Lan2-NH showed a strong
correlation with Lys-Ha and a medium correlation with Lys-
Hb as a result of cyclisation at position 2. The regiochemical
assignment was confirmed by the NOEs which arose from the
dehydroalanine residue. Again strong NOEs were observed
between Dha4-NH and both Glu-Ha and Glu-NH, indicating
that cyclisation had occurred at position 2. Strong correlations
were found between the Leu5-NH and both the Dha4-Hb
verifying that Dha was at position 4 in the peptide. The
evidence described above supports the conclusion that
cyclisation had occurred at position 2. The A-ring cyclisation
was found to occur regiospecifically in the absence of
enzymes. The results imply that the peptide conformation is
sufficient to induce Michael addition only at position 2 since
no cyclisation onto position 4 was detected. Interestingly,
whilst the cyclisation appears to be regioselective in the
absence of catalysis both possible stereoisomers at position 2
were obtained. However, there was some degree of stereo-
specificity since the isomers were produced in a 3:1 ratio.
Regiochemical assignment of the major product
Having confirmed cyclisation had occurred the regio- and
stereoselectivity of the process was then established. The
observation of an NOE between Lan6-NH and Leu5-Ha
allowed the identification of the resonances associated with
Lan6 (Figure 7). In conjunction with the TOCSY experiments
it was therefore possible to distinguish between the two
lanthionine halves and fully assign the resonances.
NH2
O
OH
O
O
O
H
H
N
H
N
N
OH
H2N
N
N
N
H
H
H
O
O
O
O
H'
H
S
Figure 7. The major product H-Lys1-Lan2-Glu3-Dha4-Leu5-Lan6-Ala7-OH
formed as a result of cyclisation of peptide 3 showing the regiochemistry of
the isomer formed as deduced from the depicted NOEs.
Molecular modellingÐE-ring
Cyclisation could have resulted in the formation of
lanthionines at either residue 2 or 4. A series of NOEs were
recorded which indicated that Michael addition had occurred
at position 2. Considering first the NOEs associated with
Lan2, there was a strong correlation between Lan2-NH and
Lys1-Ha and a medium NOE observed between Lan2-NH and
Lys-Hb. A medium NOE was observed between Lan2-Hb and
the Glu3-NH supporting the hypothesis that cyclisation had
occurred at position 2. The NOEs related to the dehydroa-
lanine residue further validated the regiochemical assign-
ment. Strong correlations were found between Dha-NH and
Unfortunately it was not possible to assign the stereochem-
istry of the newly generated alpha centre using the recorded
NOEs directly. The correlations observed were very similar
for both peptides and hence an unambiguous assignment
could not be made.
In order to determine the absolute stereochemistry at
residue 2, molecular modelling studies were undertaken. Both
d,l-lanthionine and l,l-lanthionine were modelled and it was
hoped that assignment would be possible by comparison of
the experimentally observed NOEs with the calculated
interproton distances. The A-ring was constructed with both
1460
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1460 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Lantibiotics
1455±1466
a d and an l centre at residue 2. The lowest energy
conformation of the rings was obtained by a two step
procedure involving potential energy minimisation followed
by molecular dynamics simulated annealing. The latter
process constituted heating the molecules to a very high
temperature (600 K) and then slow cooling over 1000 ps, thus
furnishing them with sufficient energy to overcome local
maxima and resulting in a conformation situated at an energy
minimum. This was carried out using MacroModel 5.0[29] and
was repeated three times to ensure that a true low energy
conformer had been found. The OPLS force field was used
throughout this work together with the GB/SA solvent model
for water. For the l isomer, each anneal resulted in the
generation of the same structure indicating a definite energy
minimum that presumably is the global energy minimum.
However, from the three anneals for the d isomer three
different structures were obtained. The conformations result-
ing from the second (II) and third (III) anneal were both
utilised in the subsequent simulations since both may
represent the lowest energy conformation. A molecular
dynamics simulation was then performed at 300 K so as to
reproduce the molecular environment present in the NMR
experiments. The simulation was run for 5000 ps with distance
measurements recorded every 1 ps. From this data set the
useful interatomic distances were selected, that is those where
the proton ± proton separation differed by >0.5 Š between
the two isomers. These values were averaged to produce an
effective interproton distance in Angstroms using Equa-
tion (1).
NOE for the minor compound. However, assignment of the
weak NOE to the l isomer, since it falls around the upper
distance range, cannot be made from this information. It
follows that conclusions regarding which isomer is the natural
one (d isomer) can only be drawn from examples where an
NOE was observed for each compound. In the case of the
Lan2-Hb and Glu-NH there was a significant difference in the
interatomic distance (4.1 Š in the l isomer cf. 3.5/3.34 Š in
the d isomer) and a difference in the NOEs observed
(medium for the major product and strong for the minor
one). However, an assignment of the major product as the l
isomer would be incorrect since neither interproton distance
falls within the boundary for a strong NOE. The NOEs also
observed between the Glu-Ha and Glu-NH are weak in the
major product and strong in the minor compound. The
proton ± proton separation was found by modelling to be
2.2 Š for the l isomer and 2.9/2.9 Š for the d isomer. This
implies that the major product has d stereochemistry. Further
evidence to support this can be gained from the Leu-Ha and
Lan-NH example. Here the interproton distance is 3.6 Š for
the l isomer and 3.0/2.8 Š for the d isomer and a strong and a
medium NOE observed in the major and minor products,
respectively, again suggesting that the major product is the d
isomer. Fewer NOEs were also observed for the major
product further supporting the tentative assignment of the
major product of cyclisation as the d isomer. The simulated
annealing studies of the d isomer failed to produce a single
lowest energy conformer. If the molecule does not have a
single preferred lowest energy conformer and is constantly
changing conformation then the number of NOEs observed
would diminish. However, this cannot be used as proof of the
stereochemistry at residue 2. Therefore, whilst the data
obtained suggests that a d centre is present at residue 2 in
the major product of cyclisation, it is not conclusive and
cannot be used to generate a definite assignment of absolute
stereochemistry.
6
1/6
rav ˆ hrij
i
(1)
rav is the final average simulation distance for atom pairs i, j,
rij is the instantaneous value of this distance and indicates that
an average is calculated over the course of the molecular
dynamics trajectory (Table 1).[30]
Table 1. Average interproton distances from molecular modelling studies. Only
the results which gave a predicted difference of over 0.5 Š between the d and l
forms and where a difference in NOEs is observed are listed.[a]
Conclusion
l Isomer d Isomer (II) d Isomer (III) Major Minor
Three biomimetic cyclisations have been studied based on the
biosynthesis of the A-, B- and E-rings of subtilin. The required
dehydroamino acid and cysteine containing peptides were
synthesised by using both solid-phase and solution method-
ology. All cyclisations were found to be rapid and quantita-
tive. In the case of the B- and E-rings single stereoisomers
were produced. Although the stereochemistry was not
determined beyond doubt the lanthionines in both of these
cases are probably meso in accord with natures precedent.
Numerous desulfurisation trials with Raney nickel were
unsuccessful with the small amounts of compound available
to us, nor was Edman sequencing in comparison with modified
nisin conclusive. In the case of the A-ring two products were
formed, in a 3:1 ratio. These were found to be cyclic with
Michael addition having occurred regioselectively but not
stereoselectively onto Dha2, resulting in the isolation of the
natural regioisomer of the subtilin A-ring. No addition onto
the other dehydroamino acid could be detected. Unfortu-
nately the assignment of absolute stereochemistry at this point
product product
Lan-Hb Glu-NH 4.1
Glu-NH Glu-Ha 2.2
Glu-NH Dha-NH 2.4
Glu-Ha Glu-Hg 2.9
Glu-Hg Dha-NH 5.2
Dha-NH Leu-NH 2.4
Leu-Ha Lan-NH 3.6
3.5
2.9
1.8
3.9
2.7
1.7
3.0
3.4
2.9
1.9
3.9
2.6
1.7
2.8
m
w
±
m
w
±
s
s
m
±
±
m
m
s
[a] NOEs are given as strong (s), medium (m) or weak (w).
The intensity of an NOE is affected not only by proton
separation but also by a number of other factors, including the
timescale of the relative motions of the atoms involved.
Therefore the assignment of strong, medium or weak to an
observed NOE contact cannot simply be made by comparison
with the above distance criteria. For example, the molecular
modelling indicated that the separation between Glu-Hg and
Dha-NH is 5.2 Š in the l isomer and 2.7/2.6 Š in the d isomer.
A weak NOE was observed for the major product and no
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1461 $ 17.50+.50/0
1461

M. Bradley et al.
FULL PAPER
temperature for 16 h. The reaction mixture was then heated to 40 8C for
2.5 h. The reaction mixture was cooled, filtered and concentrated in vacuo.
The resulting oil was dissolved in EtOAc (100 mL) and washed with sat.
NaHCO3 (100 mL), 10% citric acid (100 mL), sat. NaHCO3 (100 mL),
water (100 mL) and brine (100 mL). The organic phases were dried
(MgSO4) and concentrated in vacuo to give an orange oil. Purification by
column chromatography on silica gel (eluting with 1:1 EtOAc/hexane)
yielded the title compound as a colourless foam (2.1 g, 76% yield). Rf ˆ
0.27 (EtOAc/hexane 1:1); IR: nÄmax ˆ 3286 (m), 2957 (m), 2369 (w), 1745 (m,
was not possible either by NOE studies, by desulfurisation or
comparison with a d,l-lanthionine residue obtained from
nisin. Molecular modelling however in conjunction with the
NOE work indicated that the stereochemistry of the major
product corresponded to the natural d,l-lanthionine. In
peptides containing both cysteine and lysine residues the
cysteine residue was not unexpectedly the only residue to
cyclise. It is fascinating to begin to understand how these
linear peptides contain sufficient information to direct both
regio- and stereospecific ring closures. To further these aims
linear peptides containing the precursors to the combined
ABC rings are now being prepared.
ˆ
ˆ
ˆ
ester, C O), 1712 (s, urethane, C O), 1638 (s, amide, C O), 1514 (m), 1438
1
(m), 1366 (w), 1248 (w), 1212 (s), 1171 (s), 1041 (m), 1016 cm (m);
1H NMR (300 MHz, CDCl3): d ˆ 0.92 (2d, 6H, J ˆ 6, 6 Hz; Leu-6Hd), 1.45
(s, 9H; C(CH3)3), 1.65 ± 1.69 (m, 2H; Leu-2Hb), 1.96 ± 2.02 (m, 4H; Pro-
2Hb‡Pro-2Hg), 2.15 (s, 3H; Cys-CH3), 2.15 ± 2.18 (m, 1H; Leu-Hg), 2.72
(ABX, 1H, J ˆ 14, 7 Hz; Cys-Hb), 2.94 (ABX, 1H, J ˆ 14, 6 Hz; Cys-Hb),
3.73 ± 3.79 (s‡m, 4H; CO2CH3‡Pro-Hd), 4.10 ± 4.14 (m, 1H; Pro-Hd), 4.49
(dd, 1H, J ˆ 7, 6 Hz; Cys-Ha), 4.93 ± 4.99 (m‡dd, 2H, J ˆ 6, 8 Hz; Leu-
Ha‡Pro-Ha), 6.82 (d, 1H, J ˆ 9 Hz; CONH), 6.99 (d, 1H, J ˆ 8 Hz;
CONH); 13C NMR (75.5 MHz, CDCl3): d ˆ 16.5 (Cys-CH3), 23.1 (Leu-Cd),
24.8 (Leu-Cg), 25.0 (Pro-Cg), 28.4 (C(CH3)3), 29.2 (Pro-Cb), 36.7 (Cys-Cb),
41.6 (Leu-Cb), 47.3 (Pro-Cd), 49.8 (Ca), 52.4 (Ca), 52.3 (Ca), 59.1 (CO2CH3),
80.2 (C(CH3)3), 155.7 (OCONH), 169.5 (CONH), 172.3 (CONH), 172.6
Experimental Section
General information: NMR spectra were recorded on Bruker AC300 or
Bruker DRX 500 spectrometers. ESI mass spectra were recorded by using a
VG platform quadrupole electrospray ionisation mass spectrometer. High
resolution accurate mass measurements were carried out at 10000
resolution by using mixtures of polyethylene glycols and/or polyethylene
glycol methyl ethers as mass calibrants for FAB. Infrared spectra were
recorded on a BioRad Golden Gate FTS 135. All samples were run as neat
solids or oils. UV/Vis spectra were measured on a Hewlett Packard 8452A
diode array spectrophotometer. RP-HPLC was carried out with a Hewlett
Packard HP1100 Chemstation. 1) column: C18 ODS, i.d. 150 mm  3 mm
(0.5 mLmin 1), gradient: 0.1%TFA/H2O to 0.04% TFA/MeCN over
20 min; 2) column: C8 Techsphere, i.d. 250 mm  4.6 mm (1 mLmin 1),
gradient: 0.1%TFA/H2O to 0.04% TFA/MeCN over 20 min; 3) column:
C18 ODS, i.d. 250 mm  10 mm (2.5 mLmin 1), gradient : 0.1%TFA/H2O
to 0.04% TFA/MeCN over 45 min.
‡
‡
(CO2CH3); ES-MS: m/z: 460.3 [M‡H] ; HR-MS: [M‡H] : C21H38N3O6S
calcd 460.2481, found 460.2492; HPLC 1) (l220): 18.0 min.
Boc-Leu-Cys((O)Me)-Pro-OMe (15): Sodium metaperiodate (1.05 g,
1.1 equiv) was dissolved in water (20 mL) and cooled to 0 8C. Boc-Leu-
Cys(Me)-Pro-OMe (2.1 g, 4.6 mmol, 1.0 equiv) was dissolved in dioxane
(40 mL) and added dropwise to the oxidant. Stirring was continued for 1 h.
The reaction mixture was concentrated to remove organic solvent and the
product extracted with CH2Cl2 (2 Â 100 mL). The combined organic phases
were washed with water (2 Â 150 mL), brine (150 mL) and dried (MgSO4).
Concentration of the organic phases in vacuo yielded both diastereoisom-
ers of the title compound as a colourless glass (2.15 g, 99% yield). Rf ˆ 0.38
ˆ
(MeOH/EtOAc 1:9); IR: nÄmax ˆ 3281 (w), 2958 (w), 1743 (m, ester, C O),
ˆ
ˆ
1705 (s, urethane, C O), 1642 (s, amide, C O), 1515 (m), 1437 (m), 1365
Abbreviations: DABCO: 1,4-diazabicyclo[2.2.2]octane, DBU: 1,8-diazabi-
cyclo[5.4.0]undec-7-ene, DCC: dicyclohexyl carbodiimide, DEAD: diethyl
azodicarboxylate, DIC: diisopropyl carbodiimide, DMAP: 4-dimethylami-
nopyridine, DMF: N,N-dimethyl formamide, EDC: dimethylaminopropy-
lethyl carbodiimide hydrochloride, HOBt: 1-hydroxybenzotriazole, NEM:
N-ethyl morpholine, NMM: N-methyl morpholine, TIS: triisopropyl silane;
resin: HMPB: 4-hydroxymethyl-3-methoxyphenoxybutyric acid.
1
(w), 1247 (w), 1165 (s), 1042 (m), 1017 cm
(m); HPLC (l220):
12.3‡12.5 min; (gradient 1); 1H NMR (300 MHz, CDCl3): d ˆ 0.89 ± 0.94
(m, 6H; Leu-6Hd), 1.45 (s, 11H; C(CH3)3‡Leu-2Hb), 1.64 ± 1.67 (m, 1H;
Leu-Hg), 1.96 ± 2.03 (m, 3H; Pro-Hb‡Pro-2Hg), 2.20 ± 2.24 (m, 1H; Pro-
Hb), 2.65‡2.75 (2s, 3H; Cys-(O)CH3), 2.96 ± 3.11 (m, 2H; Cys-Hb), 3.75 ±
3.82 (s‡m, 5H; CO2CH3‡Pro-2Hd), 4.08 ± 4.10 (m, 1H; Ha), 4.48 ± 4.52 (m,
1H; Ha), 5.01 ± 5.08 (m, 2H; Ha‡CONH), 7.36 (bd, 1H, J ˆ 8 Hz; CONH);
13C NMR (75.5 MHz, CDCl3): d ˆ 21.8 (Leu-Cg), 23.2‡24.7 (Leu-Cd), 24.9
(Pro-Cg), 28.3 (C(CH3)3), 29.1 (Pro-Cb), 39.3‡39.4 (Cys-(O)CH3), 41.3
(Leu-Cb), 46.3 (Pro-Cd), 52.5 (Ca), 53.1 (Ca) 53.4 (Ca), 57.9 (CO2CH3),
59.1‡59.2 (Cys-Cb), 80.0 (C(CH3)3), 155.8 (OCONH), 168.2 (CONH),
General resin handling procedures (all reactions were performed man-
ually): Manipulations using resin were carried out with 10 mL of solvent
per gram of resin.
Solid-phase peptide synthesis:[23, 24]
‡
i) Couplings: The resin was swollen in a minimum amount of CH2Cl2 for
30 min. N-Fmoc amino acid (2 equiv) and HOBt (2 equiv) were dissolved
in CH2Cl2 with a few drops of DMF and stirred at room temperature for
10 min. DIC (2.2 equiv) was added and the mixture stirred for a further
10 min before addition to the resin. The resin was agitated at room
temperature for 2 h to effect coupling. The resin was washed with DMF
(3 Â ), CH2Cl2 (3 Â ), MeOH (3 Â ) and Et2O (2 Â ). The resin was dried
under vacuum for 30 min. For ester formation DMAP (0.3 equiv) was
included in the coupling step.
172.3 (CONH), 172.8 (CO2CH3); ES-MS: m/z: 498.4 [M‡Na] ; HR-MS
‡
[M‡H] : C21H38N3O7S calcd 476.2430, found 476.2445; HPLC 1) (l220): 12.2
and 12.4 min.
Boc-Leu-Dha-Pro-OMe (16): Boc-Leu-Cys((O)Me)-Pro-OMe (1.45 g,
3.1 mmol, 1.0 equiv) was dissolved in MeOH (15 mL), DBU (1.11 g,
1.10 mL, 2 equiv) added and the reaction mixture stirred at room temper-
ature for 1.5 h. The reaction mixture was evaporated to dryness. The
resulting glass was purified by column chromatography on silica gel
(eluting with EtOAc/hexane 1:1) yielding the title compound as
a
ii) Fmoc removal: The resin was treated with 20% piperidine in DMF with
sequential treatments of 5 and 15 min. The resin was then filtered and
washed with DMF (3 Â ), CH2Cl2 (3 Â ), MeOH (3 Â ) and Et2O (2 Â ) and
dried under vacuum.
colourless glass (1.22 g, 96% yield). Rf ˆ 0.58 (EtOAc); IR: nÄmax ˆ 3285
ˆ
ˆ
(w), 1957 (w), 2873 (w), 1744 (m, ester, C O), 1679 (m, urethane, C O),
ˆ
ˆ
1625 (s, amide, C O), 1507 (m, alkene, C C), 1365 (w), 1281 (w), 1163 (s),
1046 (w), 1021 cm (w); 1H NMR (300 MHz, [D6]DMSO): d ˆ 0.86 (2d,
1
iii) Cleavage: The resin (100 mg) was swollen in a minimum amount of
CH2Cl2 (0.3 mL). TFA (9.5 mL)/TIS (0.2 mL) was added and the resin
agitated at room temperature for 90 min. The resin was removed by
filtration through a glass wool plug, washed with TFA, the filtrate
concentrated to ca. 1 mL and Et2O (25 mL) added. The resulting
precipitate was collected by centrifugation and washed with Et2O (4 Â
25 mL).
6H, J ˆ 6, 6 Hz; Leu-6Hd), 1.45 ± 1.52 (s‡m, 11H; C(CH3)3‡Leu-2Hb),
1.60 ± 1.64 (m, 1H; Leu-Hg), 1.95 ± 2.01(m, 3H; Pro-Hb‡Pro-2Hg), 2.20 ±
2.23 (m, 1H; Pro-Hb), 3.35 (s, 3H; CO2CH3), 3.52 ± 3.65 (m, 2H; Pro-2Hd),
4.02 (d, 1H, J ˆ 9 Hz; Ha), 4.29 (dd, 1H, J ˆ 5, 3 Hz; Ha), 4.80 (s, 1H; Dha-
Hb), 5.47 (s, 1H; Dha-Hb), 7.04 (d, 1H, J ˆ 8 Hz; Leu-CONH), 9.77 (s, 1H;
Dha-CONH); 13C NMR (75.5 MHz, [D6]DMSO): d ˆ 21.3‡22.9 (Leu-Cd),
24.2 (Leu-Cg), 24.7 (Pro-Cg), 28.1 (C(CH3)3), 28.8 (Pro-Cb), 40.2 (Leu-Cb),
48.9 (Pro-Cd), 51.8 (Ca), 52.7 (Ca), 58.4 (CO2CH3), 78.2 (C(CH3)3), 102.6
(Dha-Cb), 137.7 (Dha-Ca), 155.5 (OCONH), 164.9 (CONH), 171.9
Boc-Leu-Cys(Me)-Pro-OMe (14): TFA ´ H-Cys(Me)-Pro-OMe (2.0 g,
6 mmol, 1.0 equiv), Boc-Leu-OH (1.4 g, 1.0 equiv), HOBt (0.92 g,
1.0 equiv) and NMM (3.0 g, 3.3 mL, 5 equiv) were dissolved in CH2Cl2
and the mixture cooled on an ice bath whilst DCC (1.36 g, 1.1 equiv) was
added portionwise. Stirring was continued on ice for 1 h and at room
‡
‡
(CONH), 172.3 (CO2CH3); ES-MS: m/z: 412.4 [M‡H] , 434.4 [M‡Na] ,
‡
‡
450.3 [M‡K] ; HR-MS [M‡H] : C20H34N3O6 calcd 412.2448, found
412.2463; HPLC 1) (l220): 14.6 min.
1462
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1462 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Lantibiotics
1455±1466
Boc-Leu-Dha-Pro-OMe (16): Boc-Leu-Cys((O)Me)-Pro-OMe (250 mg,
0.52 mmol) was dissolved in dioxane (5 mL) and the reaction mixture
was refluxed. After the solution was heated under reflux for 18 hours,
concentration of the reaction mixture yielded an orange oil which was
purified by column chromatography on silica gel (eluting with EtOAc/
hexane 1:1) giving the title compound as a colourless glass (210 mg, 88%
yield).
treated with 95% TFA as desribed above. The title compound was
precipitated with Et2O and isolated as a white powder (6 mg, 75% yield);
RP-HPLC (3): 9.4 min; data as below.
H-Leu-Lan-Pro-Gly-Lan-Val-Gly-OH (4) B-ring: H-Leu-Dha-Pro-Gly-
Cys-Val-Gly-OH (10 mg, 0.016 mmol) was dissolved in water (9 mL) and
triethylammonium acetate buffer (5mm, pH 8, 1 mL) added. Aliquots were
removed from the reaction every 30 s for 10 min and then every minute for
a further 5 min. Analysis by RP-HPLC 2) (C8 Techsphere column) showed
the reaction to be complete after 10 min. The reaction mixture was
lyophilised yielding the pure material (10 mg, 100% yield); 1H NMR
(500 MHz, [D6]DMSO): d ˆ 0.68 ± 0.85 (m, 12H, Leu-6Hd‡Val-6Hg),
1.25 ± 1.35 (m, 2H, Leu-2Hb), 1.65 ± 1.90 (m, 1H, Leu-Hg), 1.71 ± 1.76 (m,
1H, Pro-Hg), 1.95 ± 2.04 (m, 3H, Pro-2Hb‡Val-Hb), 2.23 ± 2.27 (m, 1H, Pro-
Hg), 2.60 (ddd, 2H, J ˆ 16, 6, 5 Hz; Lan-2Hb), 2.86 (dd, 1H, J ˆ 16, 9 Hz;
Lan-Hb), 2.98 (dd, 1H, J ˆ 16, 9 Hz; Lan-Hb), 3.33 ± 3.48 (m, 2H; Pro-2Hd),
3.65 ± 3.75 (m, 4H; 2Gly-2Ha), 4.10 (2d, 2H, J ˆ 8, 7 Hz; Leu-Ha‡Val-Ha),
4.55 ± 4.58 (m, 1H; Lan-Ha), 4.64 ± 4.68 (m, 1H; Lan-Ha), 4.90 (dd, 1H, J ˆ
8, 4 Hz; Pro-Ha), 7.58 (s, 1H, J ˆ 7 Hz, Lan-NH), 7.86 (d, 1H, J ˆ 8 Hz, Val-
NH), 8.23 (t, 1H, J ˆ 6 Hz; Gly-NH), 8.77 (t, 1H, J ˆ 7 Hz; Gly-NH), 8.98
(d, 1H, J ˆ 8 Hz; Lan-NH); 13C NMR (125 MHz, [D6]DMSO): d ˆ
18.5‡19.3 (Leu-Cd), 22.3‡22.5 (Val-Cg), 23.2 (Pro-Cg), 24.9 (Pro-Cb), 31.4
(Leu-Cg), 32.7 (Val-Cb), 35.8 (Leu-Cb), 40.9 (2Gly-Ca), 42.1 (Lan-Cb), 45.0
(Lan-Cb), 52.4 (Pro-Cd), 52.8 (Ca), 53.8 (Ca), 60.3 (Ca), 61.4 (Ca), 64.4 (Ca),
171.2 (CONH), 171.4 (CONH), 172.0 (CONH), 172.8 (CONH), 174.0
Boc-Leu-Dha-Pro-OH (7):[12] Boc-Leu-Dha-Pro-OMe (750 mg, 1.8 mmol)
was dissolved in MeOH (15 mL) and cooled on an ice bath whilst 1m NaOH
(10 mL) was added dropwise. The reaction was allowed to warm to room
temperature over 1.5 h. The reaction mixture was concentrated to ca.
15 mL, acidified to pH 4 with 2m KHSO4 and the product extracted into
CH2Cl2 (2 Â 30 mL). The combined organics were washed with water
(50 mL), brine (50 mL), dried (MgSO4) and concentrated in vacuo yielding
the desired material as a colourless glass (680 mg, 95% yield). Rf ˆ 0.26
ˆ
(EtOAc); IR: nÄmax ˆ 3283 (w), 2959 (w), 2933 (w), 1674 (s, urethane, C O),
ˆ
ˆ
1643 (s, amide, C O), 1619 (s, acid, C O), 1515 (m), 1448 (m), 1367 (w),
1283 (w), 1162 (s), 1047 cm (w); 1H NMR (300 MHz, [D6]DMSO): d ˆ
1
0.87 (2d, 6H, J ˆ 6, 6 Hz; Leu-6Hd), 1.45 ± 1.51 (s‡m, 11H; C(CH3)3)‡Leu-
2Hb), 1.78 ± 1.89 (m, 4H; Pro-2Hg‡Pro-Hb‡Leu-CHg), 2.20 (m, 1H; Pro-
Hb), 3.32 ± 3.45 (m, 2H; Pro-2Hd), 4.04 (bdd, 1H, J ˆ 9, 5 Hz; Ha), 4.23 (dd,
1H, J ˆ 5, 6 Hz; Ha), 4.80 (s, 1H; Dha-Hb), 5.48 (s, 1H; Dha-Hb), 7.03 (d,
1H, J ˆ 8 Hz; Leu-CONH), 9.65 (s, 1H; Dha-CONH); 13C NMR
(75.5 MHz, [D6]DMSO): d ˆ 21.4‡23.1 (Leu-Cd), 24.5 (Leu-Cg), 28.2
(C(CH3)3), 29.0 (Pro-Cg), 33.4 (Pro-Cb), 40.4 (Leu-Cb), 49.0 (Pro-Cd), 52.8
(Ca), 58.6 (Ca), 102.4 (Dha-Cb), 138.0 (Dha-Ca), 155.5 (OCONH), 164.8
(CONH), 171.9 (CONH), 173.2 (CO2H); ES-MS: m/z: 342.2 [(M
‡
(CONH), 174.3 (CONH), 175.8 (CO2H); ES-MS: m/z: 614.2 [M‡H] ,
‡
‡
636.2 [M‡Na] ; HR-MS [M‡H] : C26H44N7O8S calcd 614.2972, found
614.3126; HPLC 1) (l220): 9.5 min.
‡
‡
tBu)‡H] , 398.3 [M‡H] .
H-Leu-[2-2H]Lan-Pro-Gly-Lan-Val-Gly-OH (4) B-ring: The above reac-
tion was repeated with a solution of H-Leu-Dha-Pro-Gly-Cys-Val-Gly-OH
(0.1 mg) in D2O (900 mL). Deuterated phosphate buffer (100 mL, 5 mm,
pD 8) was added and the reaction progress assessed (by UV) every 30 s for
10 min. The sample was then lyophilised, redissolved in water and
lyophilisation repeated twice and the sample analysed by mass spectrom-
Boc-Leu-Dha-Pro-Gly-Cys(Trt)-Val-Gly-Wang linker-resin (11): H-Gly-
Cys(Trt)-Val-Gly-Wang-resin (9) (100 mg, 0.024 mmol, 1.0 equiv) was
swollen in CH2Cl2 for 30 min. Boc-Leu-Dha-Pro-OH (7) (40 mg,
0.096 mmol, 4 equiv), PyBroP (50 mg, 4.4 equiv) and DMAP (7 mg,
2.4 equiv) were dissolved in a minimum amount of CH2Cl2 (ca. 1 mL)
and DIPEA (12 mg, 17 mL, 4 equiv) added. This solution was added to the
resin and the reaction mixture shaken at room temperature for 1.5 h. A
ninhydrin test showed that the coupling was incomplete, therefore the
coupling was repeated using Boc-Leu-Dha-Pro-OH (20 mg, 2 equiv),
PyBroP (25 mg, 2.2 equiv), DMAP (3.5 mg, 1.2 equiv) and DIPEA (6 mg,
8 mL, 2 equiv). After the mixture had been shaken for 30 min at room
temperature, a negative ninhydrin test was obtained. The resin was washed
with DMF (3 Â 5 mL), CH2Cl2 (3 Â 5 mL), MeOH (3 Â 5 mL), Et2O (2 Â
5 mL) and dried under high vacuum for 2 h.
‡
etry. ES-MS: m/z: 615.3 [M‡H] .
Boc-Leu-Dha-Ala-OH (8): Boc-Leu-Ser-OH (3 g, 9.4 mmol) was dissolved
in freshly distilled THF (20 mL) together with Et3N (0.95 g, 9.4 mmol) and
cooled to 15 oC for 15 min, under N2. Isobutylchloroformate (1.28 g,
9.4 mmol) was added and the mixture stirred at 15 oC for 30 min under
N2. To this mixture, HCl ´ H-Ala-OMe (1.45 g, 14.1 mmol) in water (18 mL)
and Et3N (1.42 g, 14.1 mmol) were added rapidly in one portion. The
mixture was vigorously stirred at room temperature for 3 h. The reaction
mixture was concentrated to approx. 20 mL and the product extracted with
Et2O (3 Â 20 mL). The combined organic extracts were washed with water
(50 mL), brine (50 mL), dried (MgSO4), filtered and the solvent removed in
vacuo to give the crude tripeptide. Purification by column chromatography
(silica gel, hexane/EtOAc 80:20) gave the title compound as a white
crystalline solid (1.65 g, 44% yield). Rf ˆ 0.1 (EtOAc/hexane 20:80); m.p.:
TFA ´ H-Leu-Dha-Pro-Gly-Cys-Val-Gly-OH (1): Boc-Leu-Dha-Pro-Gly-
Cys(Trt)-Val-Gly-HMPB-resin (11) (100 mg) was treated with 95% TFA/
CH2Cl2 as described above. The resulting peptide was obtained as a white
powder (18 mg, 75% yield). 1H NMR (360 MHz, D2O): d ˆ 0.99 (2d, 12H,
J ˆ 8, 7 Hz; Leu-6Hd‡Val-6Hg), 1.75 ± 1.83 (m‡d, 3H, J ˆ 7 Hz; Pro-
2Hg‡Leu-Hg), 2.00 ± 2.12 (m, 3H, Leu-2Hb‡Pro-Hb), 2.18 (dq, 1H, J ˆ 7,
7 Hz; Pro-Hb), 2.44 (dd, 1H, J ˆ 8, 7 Hz; Val-Hb), 2.96 (t, 2H, J ˆ 6 Hz; Cys-
2Hb), 3.70 ± 3.74 (m, 1H, Pro-Hd), 3.85 ± 3.88 (m, 1H, Pro-Hd), 4.02‡4.04
(2s, 4H, 2Gly-2Ha), 4.13 (t, 1H, J ˆ 8 Hz; Leu-Ha), 4.24 (d, 1H, J ˆ 7 Hz;
Val-Ha), 4.52 (t, 1H, J ˆ 7 Hz; Pro-Ha), 4.60 (dd, 1H, J ˆ 6, 5 Hz; Cys-Ha),
5.38 (d, 1H, J ˆ 2 Hz; Dha-Hb), 5.52 (d, 1H, J ˆ 2 Hz; Dha-Hb); 13C NMR
(90 MHz, D2O): d ˆ 18.5‡19.3 (Val-Cg), 21.8‡22.7 (Leu-Cd), 24.8 (Pro-Cg),
25.7 (Leu-Cg), 26.2 (Val-Cb), 30.5 (Pro-Cb), 31.1 (Cys-Cb), 40.8 (Leu-Cb),
42.1‡43.5 (2Gly-Ca), 51.5 (Pro-Cd), 52.7 (Ca), 56.8 (Ca), 60.5 (Ca), 62.0
(Ca), 110.1 (Dha-Cb), 136.1 (Dha-Ca), 167.8 (CONH), 169.8 (CONH), 172.4
(CONH), 172.8 (CONH), 174.0 (CONH), 174.5 (CONH), 175.8 (CO2H);
123 ± 125 oC; IR: nÄmax ˆ 3425 (s, broad, urethane, amide, NH), 1744 (s, ester,
1
C O), 1654, 1508 cm (s, urethane, amide, C O); 1H NMR (300 MHz,
[D6]DMSO): d ˆ 0.97‡1.00 (2d, 6H, J ˆ 6.5, 6.5 Hz; CH(CH3)2), 1.44 (d,
3H, J ˆ 7 Hz; CHCH3), 1.49 (s, 9H; C(CH3)3), 1.55 ± 1.62 (brm, 2H;
CH2CH(CH3)2), 1.69 ± 1.83 (brm, 1H; CH2CH(CH3)2), 3.76 (s, 3H;
CO2CH3), 3.83 (dd, 2H, J ˆ 6 Hz; CH2OH), 4.08 ± 4.18 (m, 1H; CHa),
4.43 ± 4.53 (m, 2H; 2CHa); 13C NMR (75 MHz, DEPT analysis,
[D6]DMSO): d ˆ 17.5 (CHCH3), 21.7‡23.4 (CH(CH3)2), 25.8 (CH(CH3)2),
28.6 (C(CH3)3), 41.8 (CH2(CH3)2), 49.5, 52.8, 54.7, 56.3 (CO2CH3)‡(3CHa),
63.0 (CH2OH), 80.7 (C(CH3)3), 158.1 (OCONH), 171.8, 174.4, 175.7
ˆ
ˆ
‡
‡
(2CONH)‡(CO2CH3); ES-MS: m/z: 829 [2M‡Na] , 426 [M‡Na] ; HR-
‡
‡
‡
‡
ES-MS: m/z: 614.3 [M‡H] , 636.4 [M‡Na] ; HR-MS [M‡H] :
C26H44N7O8S calcd 614.2972, found 614.2992; HPLC 1) (l220): 7.6 min.
MS [M‡H] C18H34O7N3 calcd 404.2397, found 404.2412.
CuCl (0.08 g, 0.77 mmol) and EDC (0.48 g, 2.5 mmol) was added at room
temperature to a stirred solution of Boc-Leu-Ser-Ala-OMe (1 g, 2.5 mmol)
in dry MeCN (50 mL). The reaction was stirred for 3 h following which an
equal volume of EtOAc was added and the precipitated urea removed by
filtration. The organic product was washed with water (50 mL), brine
(50 mL), dried over MgSO4 and the solvent removed in vacuo. The title
compound was obtained as a pale yellow oil after column chromatography
(silica gel, hexane/EtOAc 7:3; 0.54 g, 57% yield). Rf ˆ 0.18 (hexane/EtOAc
Boc-Leu-Lan-Pro-Gly-Lan-Val-Gly-Wang linker-resin (4a): Boc-Leu-
Dha-Pro-Gly-Cys(Trt)-Val-Gly-Wang-resin (95 mg, 0.023 mmol, 1.0 equiv)
was shaken in 2%TFA/2%TIS/96%CH2Cl2 (3 mL) for 15 min at room
temperature, filtered and the process repeated four more times. After five
15 min exposures the resin was washed with CH2Cl2 (2 Â 5 mL), DMF (2 Â
5 mL), CH2Cl2 (2 Â 5 mL), MeOH (2 Â 5 mL), Et2O (2 Â 5 mL) and dried.
The resin (95 mg, 0.023 mmol, 1.0 equiv) was swollen in DMF (1 mL).
5%NMM in DMF (5 mL) was added and the mixture shaken at room
temperature for 3 h. The resin was washed with DMF (3 Â 5 mL), CH2Cl2
(3 Â 5 mL), MeOH (3 Â 5 mL), Et2O (2 Â 5 mL) and dried. Boc-Leu-Lan-
ˆ
7:3); IR: nÄmax ˆ 3449 (s, urethane, amide, NH), 1736 (s, ester, C O), 1695 (s,
1
1
ˆ
ˆ
urethane, C O), 1663 cm (urethane, amide, C O); H NMR (300 MHz,
CDCl3): d ˆ 0.93‡0.95 (2d, 6H, J ˆ 6.5, 6.5 Hz; CH(CH3)2), 1.44 (s, 9H;
C(CH3)3), 1.37 ± 1.56 (m, 4H; CHCH2CH(CH3)2‡CHCH3), 1.57 ± 1.69
Pro-Gly-Lan-Val-Gly-Wang-resin (50 mg, 0.25 mmol per
g resin) was
Chem. Eur. J. 2000, 6, No. 8 ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1463 $ 17.50+.50/0
1463

M. Bradley et al.
FULL PAPER
(brm, 2H; CHCH2CH(CH3)2), 3.78 (s, 3H; CO2CH3), 4.06 ± 4.23 (brm, 1H;
CHa), 4.61 (dq, 1H, J ˆ 7.5, 7.5 Hz, CHa), 4.98 (d, 1H, J ˆ 7.5 Hz,NH),
(250 mL) and brine (250 mL). The organics were dried (MgSO4), filtered
and the filtrate concentrated in vacuo to give a pale yellow foam.
Purification by column chromatography on silica gel (eluting with EtOAc/
hexane 1:1) produced the desired compound as a white solid (3.0 g, 75%
yield). Rf ˆ 0.16 (EtOAc/hexane 1:1); IR: nÄmax ˆ 3271 (m), 2973 (w), 2932
ˆ
5.35‡6.47 (2s, 2H; C CH2), 6.97 (brd, 1H, J ˆ 9 Hz, NH), 8.61 (brs, 1H;
NH); 13C NMR (75 MHz, DEPT analysis, CDCl3): d ˆ 18.23, 21.85, 23.14,
24.93 (CHCH3)‡(CH(CH3)2)‡(CH(CH3)2), 28.40 (C(CH3)3), 41.59
(CH2(CH3)2), 48.76, 52.82, 54.12 (CO2CH3)‡(2CHa), 80.38 (C(CH3)3),
ˆ
ˆ
ˆ
(w), 1714 (m, ester, C O), 1689 (m, urethane, C O), 1634 (s, amide, C O),
1515 (m), 1391 (w), 1365 (w), 1245 (m), 1154 (s); m.p.: 132 ± 135 8C;
1H NMR (400 MHz, [D6]DMSO): d ˆ 0.85 (2d, 6H, J ˆ 6, 6 Hz; Leu-6Hd),
1.25 ± 1.30 (m, 2H; Lys-2Hb), 1.40 (s, 27H; 3C(CH3)3), 1.45 ± 1.55 (m, 7H;
Leu-Hg‡Leu-2Hb‡Lys-2Hg‡Lys-2Hd), 1.70 ± 1.73 (m, 1H; Glu-Hb), 1.80 ±
1.85 (m, 1H; Glu-Hb), 2.05 (2s, 6H; 2Cys-CH3), 2.15 (t, 2H, J ˆ 8 Hz; Glu-
2Hg), 2.60 (ABX, 2H, J ˆ 14, 7 Hz; 2Cys-Hb), 2.75 (ABX, 2H, J ˆ 14, 6 Hz;
2Cys-Hb), 2.90 ± 2.98 (m, 2H; Lys-2He), 3.60 (s, 3H; CO2CH3), 3.90 ± 3.96
(m, 1H; Lys-Ha), 4.25 ± 4.30 (m, 2H; Leu-Ha‡Glu-Ha), 4.50 (ABX, 2H,
J ˆ 7, 6 Hz; 2Cys-Ha), 6.70 (brt, 1H, J ˆ 7 Hz; OCONH), 6.90 (d, 1H, J ˆ
7 Hz; Lys-CONH), 7.90 (d, 1H, J ˆ 6 Hz; Cys-CONH), 8.05 (d, 1H, J ˆ
7 Hz; Cys-CONH), 8.20 (d, 1H, J ˆ 8 Hz; Glu-CONH), 8.40 (d, 1H, J ˆ
7 Hz; Leu-CONH); 13C NMR (75.5 MHz, [D6]DMSO): d ˆ 15.8 (Cys-
CH3), 15.9 (Cys-CH3), 21.6‡21.8 (Leu-Cd), 23.1 (Lys-Cg), 24.8 (Leu-Cg),
28.2‡28.6‡28.7 (3C(CH3)3), 29.1 (Lys-Cb), 29.8 (Glu-Cb), 30.5 (Lys-Cd),
31.4 (Glu-Cg), 38.8 (Cys-Cb), 40.9 (Cys-Cb), 48.0 (Lys-Ce), 50.6 (Ca), 51.8
(Ca), 52.3 (Ca), 52.6 (Ca), 68.3 (CO2CH3), 77.7‡78.3‡80.8 (3C(CH3)3),
155.7 (2OCONH), 169.4 (CONH), 170.1 (CONH), 170.7 (CONH), 171.4
ˆ
ˆ
102.77 (C CH2), 133.80 (C CH2), 155.78 (OCONH), 163.42, 171.96, 173.37
(2CONH)‡(CO2CH3); ES-MS: m/z: 1178 [3M‡Na] , 793 [2M‡Na] , 408
‡
‡
‡
‡
[M‡Na] ; HR-MS [M‡H] C18H32O6N3 calcd 386.2291, found 386.2322.
Boc-Leu-Dha-l-Ala-OMe (0.39 g, 1.02 mmol) was dissolved in THF
(2 mL), 2.5m NaOH (50 mL) was added and the mixture stirred at room
temperature for 2 h. The aqueous mixture was washed with EtOAc
(100 mL) and subsequently acidified to pH 1 ± 2 (2m HCl). The precipitate
was extracted with EtOAc (2 Â 100 mL) and the combined organic extracts
washed with water (100 mL), brine (100 mL), dried (MgSO4) and the
solvent removed in vacuo. The title compound was obtained as a colourless
oil after purification by column chromatography (silica gel, hexane/EtOAc/
AcOH 50:50:2; 0.31 g, 82% yield). Rf ˆ 0.26 (hexane/EtOAc/AcOH
ˆ
50:50:2); IR: nÄmax ˆ 3329 (m, urethane, amide, NH), 1719 (s, acid, C O),
1
ˆ
ˆ
1656 (s, urethane, C O), 1630 (s, amide, C O), 1508 cm (s, urethane,
1
ˆ
amide, C O); H NMR (300 MHz, CDCl3): d ˆ 0.89‡0.92 (2d, 6H, J ˆ 6.5,
6.5 Hz, CH(CH3)2), 1.40 (s, 9H; C(CH3)3), 1.41 (d, 3H, J ˆ 7 Hz; CHCH3),
1.50 ± 1.70 (2m, 3H; CH2CH(CH3)2), 3.96 ± 4.12 (brm, 1H; CHa), 4.41 (dq,
‡
(CONH), 171.7 (CO2CH3), 173.0 (CO2tBu); ES-MS: m/z: 893.4 [M‡H] ,
ˆ
1H, J ˆ 7, 7.5 Hz; CHa), 5.46‡6.29 (2s, 2H; C CH2), 5.76 (brd, 1H, J ˆ
‡
915.4 [M‡Na] .
9 Hz; NH), 7.29 (d, 1H, J ˆ 7 Hz; NH), 8.69 (s, 1H; NH); 13C NMR
(75 MHz, CDCl3): d ˆ 17.8 (CHCH3), 21.7‡23.2 (CH(CH3)2), 24.9
(CH2CH(CH3)2), 28.4 (C(CH3)3), 41.8 (CH2CH(CH3)2), 48.9‡53.9
Boc-Lys(Boc)-Cys((O)Me)-Glu-(OtBu)-Cys((O)Me)-Leu-OMe (20): So-
dium metaperiodate (265 mg, 2.2 equiv) was dissolved in water (10 mL)
and cooled on an ice bath. Boc-Lys(Boc)-Cys(Me)-Glu-(OtBu)-Cys(Me)-
Leu-OMe (500 mg, 0.56 mmol, 1.0 equiv) was dissolved in dioxane (20 mL)
and added dropwise to the oxidant. The reaction was stirred on ice for 1 h
and at 40 8C for 4 h. The reaction mixture was concentrated to ca. 10 mL,
water (20 mL) added and the product extracted into CH2Cl2 (2 Â 30 mL).
The combined organic phases were washed with water (50 mL), brine
(50 mL), dried (MgSO4) and concentrated in vacuo to give a colourless
glass (510 mg, 99% yield). Rf ˆ 0.53 (10% MeOH/CH2Cl2); IR: nÄmax ˆ 3286
ˆ
ˆ
(2CHa), 80.3 (C(CH3)3), 104.2 (C CH2), 133.8 (C CH2), 156.3 (OCONH),
‡
163.9, 172.6, 175.8 (2CONH)‡(CO2H); ES-MS: m/z: 765 [2M‡Na] , 394
‡
‡
‡
[M‡Na] , 372 [M‡H] ; HR-MS [M‡H] C17H30O6N3 calcd 372.2135,
found 372.2110.
Boc-Leu-Dha-Ala-Asn-Cys(Trt)-Lys(Boc)-Ile-HMPB-linker-resin
(12):
Boc-Leu-Dha-Ala-OH (8) was coupled to the H-Asn-Cys(Trt)-Lys(Boc)-
Ile-HMPB-resin (10) (0.2 g, 0.038 mmol) as described above and the
reaction followed by the ninhydrin test until negative. The peptide resin
was then filtered, washed with DMF (1 Â 15 mL), CH2Cl2 (4 Â 5 mL) and
MeOH (2 Â 5 mL) and dried under vacuum. MALDI-TOF (DHB matrix,
ˆ
ˆ
(m), 2962 (w), 2933 (w), 1733 (m, ester, C O), 1684 (m, urethane, C O),
ˆ
1638 (s, amide, C O), 1514 (s), 1453 (w): 1366 (w), 1248 (w), 1158 (s),
1
1
‡
1018 cm (w); H NMR (400 MHz, [D6]DMSO): d ˆ 0.90 (d‡m, 6H, J ˆ
7 Hz; Leu-6Hd), 1.17 ± 1.28 (m, 2H, Lys-2Hg), 1.42 (3s, 27H; 3C(CH3)3),
1.52 ± 1.64 (m, 7H; Leu-Hg‡Leu-2Hb‡Lys-2Hb‡Lys-2Hd), 1.84 ± 1.93 (m,
2H; Glu-2Hb), 2.22 ± 2.27 (m, 2H; Glu-2Hg), 2.61 (2s, 6H; 2Cys-(O)CH3),
2.94 (dd, 2H, J ˆ 8, 7 Hz; Lys-2He), 3.14 ± 3.35 (m, 4H; 2Cys-2Hb),
3.62‡3.67 (2s, 3H; CO2CH3), 3.98 ± 4.03 (m, 1H; Lys-Ha), 4.33 ± 4.46 (m,
2H; Leu-Ha‡Glu-Ha), 4.71 ± 4.78 (m, 2H; 2Cys-Ha), 6.75 (brs, 1H;
OCONH), 6.90 ± 6.96(m, 1H; Lys-CONH), 8.21 ± 8.27 (m, 3H; 3CONH);
13C NMR (75.5 MHz, CDCl3): d ˆ 21.3‡21.4 (Leu-Cd, 22.8 (Lys-Cg), 24.2
(Leu-Cg), 26.9 (Lys-Cb), 27.8‡28.3‡28.4 (3C(CH3)3), 29.3 (Glu-Cb), 30.9
(Lys-Cd), 31.4 (Glu-Cg), 46.1‡47.6 (2Cys-(O)CH3), 48.0 (Lys-Ce), 50.6 (Ca),
52.2 (Ca), 52.4 (Ca), 54.5 (Ca), 55.0 (CO2CH3), 55.4‡55.5 (2Cys-Cb), 55.9
(Ca), 77.4‡78.3‡79.7 (3C(CH3)3), 155.7 (2OCONH), 169.4 (CONH), 169.9
(CONH), 170.0 (CONH), 170.9 (CONH), 171.7 (CO2CH3), 172.6 (CO2-
TFA cleavage in situ):m/z: 731.8 [M‡H] (H-Leu-Dha-Ala-Asn-Cys-Lys-
Ile-OH).
H-Leu-Dha-Ala-Asn-Cys-Lys-Ile-OH
(2):
Boc-Leu-Dha-Ala-Asn-
Cys(Trt)-Lys(Boc)-Ile-HMPB-resin (12) (0.21 g, 0.038 mmol) was shaken
in 50% TFA/CH2Cl2 containing 5% TIS v/v for 10 min. The resin was
filtered through a glass wool plug, the solvent was removed in vacuo and
the residue redissolved in the minimum amount of methanol (0.5 mL). The
peptide was precipitated into cold ether (10 mL) and isolated by
centrifugation. The supernatant was discarded and the pellet resuspended
in ether (10 mL), again pelleted by centrifugation and dried (N2) (20.5 mg,
‡
‡
73% yield). ES-MS: m/z: 730.7 [M‡H] , 366.1 [M ‡ 2H] ; HR-MS
‡
[M‡H] C31H56O9N9S calcd 730.3922, found 730.3888.
H-Leu-Lan-Ala-Asn-Lan-Lys-Ile-OH (E-ring) (5): H-Leu-Dha-Ala-Asn-
Cys-Lys-Ile-OH (1 mg, 0.0014 mol) was cyclised as previously described.
RP-HPLC 1): One major product observed, cyclic 7-mer (E-ring); ES-MS:
‡
‡
‡
tBu); ES-MS: m/z: 925.4 [M‡H] , 947.4 [M‡Na] ; HR-MS: [M‡H]
C40H73N6O14S2 calcd 925.4626 found 925.4709.
‡
‡
m/z: 730.4 [M‡H] , 366.1 [M ‡ 2H]2‡ (cyclic 7-mer); HR-MS [M‡H]
Boc-Lys(Boc)-Dha-Glu(OtBu)-Dha-Leu-OH (21): Boc-Lys(Boc)-Cys-
((O)Me)-Glu(OtBu)-Cys((O)Me)-Leu-OMe (475 mg, 0.51 mmol,
C31H56O9N9S calcd 730.3922, found 730.3932; 1H NMR (500 MHz,
H2O‡one drop D2O): d ˆ (residue, NH, aH, bH, others, JNH,CHa), Leu1
4.13, 1.73, 1.73 gH: 1.66 dCH3: 0.95, 0.95; Lan2 9.49, 5.15, 3.00, 3.00, J ˆ
8.5 Hz; Ala3 8.60, 4.25, 1.46, J <2 Hz; Asn4 8.68, 5.06, 2.98, 2.62 gNH2: 7.61,
6.93, J ˆ 9.3 Hz; Lan5 7.50, 4.11, 3.62, 2.83, J ˆ 5.3 Hz; Lys6 8.49, 4.42 1.88,
1.88 gCH2: 1.45, 1.45 dCH2: 1.82, 1.82, eCH2: 3.04, 3.04, J ˆ 7.5 Hz; Ile7 8.13,
4.27 1.94 gCH2: 1.45, 1.25, gCH3: 0.90, dCH3: 0.87, J ˆ 8.1 Hz.
1.0 equiv) was dissolved in MeOH (22.5 mL) and cooled on an ice-salt
bath. 1m NaOH (103 mg, 2.5 mL, 5 equiv) was added dropwise and the
reaction allowed to warm to room temperature over 1.5 h. The reaction
mixture was concentrated in vacuo to ca. 3 mL and acidified to pH 4 with
2m KHSO4. The product was extracted into CH2Cl2 (3 Â 40 mL) and
washed with water (100 mL), brine (100 mL), dried (MgSO4) and
concentrated in vacuo to give a colourless foam (370 mg, 93% yield).
Rf ˆ 0.29 (MeOH/EtOAc 1:9); IR: nÄmax ˆ 3302 (m), 2962 (w), 2934 (w),
H-Leu-Lan-Ala-Asn-Lan-Lys-Ile-OHÐcyclisation in deuterated buffer:
‡
ES-MS: m/z: 731.4 [M‡H] (deuterated cyclic 7-mer).
ˆ
ˆ
ˆ
Boc-Lys(Boc)-Cys(Me)-Glu-(OtBu)-Cys(Me)-Leu-OMe (17): Boc-Lys-
(Boc)-Cys(Me)-OH (2.3 g, 1.1 equiv), HOBt (882 mg, 1.2 equiv), EDC
(1.03 g, 1.2 equiv) and NEM (1.04 g, 1.1 mL, 2 equiv) were dissolved in
CH2Cl2 (150 mL). H-Glu(OtBu)-Cys(Me)-Leu-OMe (2.0 g, 4.5 mmol,
1.0 equiv) was dissolved in CH2Cl2 (20 mL) and added to the reaction
mixture. The reaction was stirred at room temperature overnight. The
reaction mixture was diluted with CH2Cl2 (150 mL) and washed with sat.
NaHCO3 (250 mL), 5% citric acid (250 mL), sat. NaHCO3 (250 mL), water
1681 (s, urethane, C O), 1652 (s, amide, C O), 1630 (s, acid, C O), 1504
(s), 1392 (w), 1367 (w), 1249 (w), 1161 (s), 1120 (w), 1020 cm 1 (w); 1H NMR
(400 MHz, [D6]DMSO): d ˆ 0.91 (2d, 6H, J ˆ 6, 6 Hz; Leu-6Hd), 1.17 ± 1.26
(m, 2H; Lys-2Hg), 1.42 (brs, 27H; 3C(CH3)3), 1.52 ± 1.64 (m, 7H; Leu-
Hg‡Leu-2Hb‡Lys-2Hb‡Lys-2Hd), 1.92 ± 2.03 (2m, 2H; Glu-2Hb),
2.21 ± 2.28 (m, 2H; Glu-2Hg), 2.88 ± 2.93 (m, 2H; Lys-2He), 3.98 ± 4.02 (m,
1H; Lys-Ha), 4.34 ± 4.37 (m, 1H; Leu-Ha), 4.42 ± 4.45 (m, 1H; Glu-Ha), 5.61
(2s, 2H; Dha-Hb), 6.20 (brs, 2H; Dha-Hb), 6.83 (brt, 1H, J ˆ 6 Hz;
1464
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1464 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8

Lantibiotics
1455±1466
‡
OCONH), 7.31 (brd, 1H, J ˆ 7 Hz; Lys-CONH), 8.45 (brd, 1H, J ˆ 7 Hz;
Glu-CONH), 8.62 (brd, 1H, J ˆ 8 Hz; Leu-CONH), 9.05‡9.10 (2s, 2H;
2Dha-CONH); 13C NMR (75.5 MHz, [D6]DMSO): d ˆ 21.3 (Leu-Cd), 23.0
(Lys-Cg), 24.6 (Leu-Cg), 26.5 (Lys-Cb), 27.9‡28.3‡28.4 (3C(CH3)3), 29.3
(Glu-Cb), 30.9 (Lys-Cd), 31.7 (Glu-Cg), 40.6 (Leu-Cb), 48.7 (Lys-Ce), 51.1
(Ca), 53.8 (Ca), 55.5 (Ca), 77.5‡78.6‡79.9 (3C(CH3)3), 103.5 (2Dha-Cb),
134.6 (2Dha-Ca), 155.7 (OCONH), 163.8 (CONH), 164.3 (CONH), 170.5
(CONH), 171.7 (CONH), 171.9 (CO2tBu), 173.8 (CO2H); ES-MS: m/z:
(CONH), 171.7 (CONH), 173.1 (2CO2tBu); ES-MS: m/z: 1255.6 [M‡H] ,
‡
1277.6 [M‡Na] .
H-Lys-Dha-Glu-Dha-Leu-Cys-Ala-OH
(3):
Boc-Lys(Boc)-Dha-Glu-
(OtBu)-Dha-Leu-Cys(Trt)-Ala-OtBu (22) (250 mg, 0.2 mmol) was dis-
solved in CH2Cl2 (9.6 mL) and TFA (10 mL) added to give a bright yellow
solution. TIS (0.4 mL) was added dropwise, removing the yellow colour.
The reaction was stirred at room temperature for 1 h. The reaction mixture
was concentrated to ca. 1 mL and Et2O (40 mL) added. The resulting
precipitate was collected by centrifugation, washed with Et2O (3 Â 40 mL)
and dried (135 mg, 96% yield). Rf ˆ 0.25 (CH2Cl2/MeOH/AcOH/H2O
40:18:3:2); 1H NMR (360 MHz, [D6]DMSO): d ˆ 0.94 (2d, 6H, J ˆ 6, 6 Hz;
Leu-6Hd), 1.22 (d, 3H, J ˆ 8 Hz; Ala-3Hb), 1.33 ± 1.75 (m, 9H; Leu-
Hg‡Leu-2Hb‡Lys-2Hb‡Lys-2Hg‡Lys-2Hd), 1.98 ± 2.05 (m, 2H; Glu-2Hb),
2.40 ± 2.65 (m, 2H; Glu-2Hg), 2.83 (m, 3H; Lys-2He‡Cys-Hb), 2.90 (ABX,
1H, J ˆ 14, 6 Hz; Cys-Hb), 3.55 ± 3.59 (m, 1H; Ha), 3.96 (s, 1H; Ha), 4.28 (t,
1H, J ˆ 6 Hz; Ha), 4.34 (dd, 1H, J ˆ 8, 4 Hz; Ha), 4.47 (dd, 1H, J ˆ 6, 4 Hz;
Ha), 5.67 (s, 1H; Dha-Hb), 5.83 (s, 1H; Dha-Hb), 6.02 (s, 1H; Dha-Hb), 6.28
‡
‡
‡
683.3 [(M Boc)‡H] , 783.3 [M‡H] ; HR-MS: [M‡H] C37H63N6O12
calcd 783.4504, found 783.4559.
Boc-Lys(Boc)-Dha-Glu(OtBu)-Dha-Leu-Cys(Trt)-Ala-OtBu (22): Boc-
Lys(Boc)-Dha-Glu(OtBu)-Dha-Leu-OH (21) (370 mg, 0.47 mmol,
1.0 equiv) and H-Cys(Trt)-Ala-OtBu (230 mg, 1.0 equiv) were dissolved
in CH2Cl2 (15 mL) and cooled on an ice-salt bath under an inert
atmosphere. HOBt (80 mg, 1.1 equiv) was added followed by EDC
(110 mg, 1.2 equiv). The reaction was allowed to warm to room temper-
ature over 3 h. The reaction mixture was diluted with CH2Cl2 (50 mL) and
washed with sat. NaHCO3 (50 mL), 10% citric acid (50 mL), sat. NaHCO3
(50 mL), water (50 mL), brine (50 mL), dried (MgSO4), and concentrated
in vacuo yielding a yellow foam. Purification by column chromatography
on silica gel (MeOH/CH2Cl2 3:97) followed by crystallisation with EtOAc/
hexane gave the title compound as a colourless foam (423 mg, 72% yield).
‡
‡
(s, 1H; Dha-Hb); ES-MS: m/z: 701.2 [M‡H] ; HR-MS [M‡H] :
C29H48N8O10S calcd 701.3214, found 701.3480; HPLC 1) (l220): 8.6 min.
A-ring cyclisation of (3): H-Lys-Dha-Glu-Dha-Leu-Cys-Ala-OH (25 mg,
0.036 mmol) was dissolved in H2O (22.5 mL) and pH 8.0 triethylammo-
nium acetate buffer (50mm, 2.5 mL) added. Small aliquots (100 mL) were
removed every 30 s for 10 min then at 20 and 30 min, quenched with 2%
TFA in water (20 mL). The reaction was complete within 10 min and was
lyophilised. The peptides were purified by HPLC 3) (13 mg, 52% (major),
4 mg, 16% (minor)).
1
Rf ˆ 0.22 (EtOAc/hexane 1:2); H NMR (360 MHz, [D6]DMSO): d ˆ 0.95
(2d, 6H, J ˆ 7, 7 Hz; Leu-6Hd), 1.30 (d, 3H, J ˆ 7 Hz; Ala-3Hb), 1.34 ± 1.45
(m, 38H; 4C(CH3)3‡Lys-2Hb), 1.60 ± 1.75 (m, 7H; Leu-Hg‡Leu-2Hb‡Lys-
2Hg‡Lys-2Hd), 1.98 ± 2.10 (m, 2H; Glu-2Hb), 2.30 ± 2.35 (m, 2H; Glu-2Hg),
2.42 (ABX, 1H, J ˆ 14, 8 Hz; Cys-Hb), 2.53 (ABX, 1H, J ˆ 14, 7 Hz; Cys-
Hb), 3.02 ± 3.09 (m, 2H; Lys-2He), 4.00 ± 4.04 (m, 1H; Lys-Ha), 4.14 (dq,
1H, J ˆ 7, 7 Hz; Ala-Ha), 4.45 ± 4.49 (dd‡m, 3H, J ˆ 8, 7 Hz; Cys-Ha‡Leu-
Ha‡Glu-Ha), 5.68 (s, 1H; Dha-Hb), 5.73 (s, 1H; Dha-Hb), 6.20 (s, 1H; Dha-
Hb), 6.30 (s, 1H; Dha-Hb), 6.80 (brt, 1H, J ˆ 7 Hz; Lyse-OCONH), 7.32 ±
7.45 (m, 15H; Trt-H), 8.15 (d, 1H, J ˆ 9 Hz; CONH), 8.17 (d, 1H, J ˆ 7 Hz;
CONH), 8.53 (d, 1H, J ˆ 8 Hz; CONH), 8.68 (d, 1H, J ˆ 7 Hz; CONH),
9.70 (d, 1H, J ˆ 7 Hz; CONH), 9.18 (s, 1H; Dha-CONH), 9.25 (s, 1H; Dha-
CONH); 13C NMR (75.5 MHz, CDCl3): d ˆ 18.1 (Ala-Cb), 21.9‡22.5 (Leu-
Cd), 23.0 (Lys-Cb), 24.8 (Leu-Cg), 26.5 (Lys-Cg), 27.3 (Glu-Cb), 27.9
(C(CH3)3), 28.1 (C(CH3)3), 28.3 (C(CH3)3), 28.5 (C(CH3)3), 29.7 (Lys-Cd),
31.8 (Glu-Cg), 33.3 (Cys-Cb), 41.1 (Leu-Cb), 49.0 (Lys-Ce), 52.5 (Ca), 54.8
(Ca), 55.2 (Ca), 7.6 (Ca), 57.7 (Ca), 67.1 (C-Ph3), 79.2 (C(CH3)3), 80.4
(C(CH3)3), 81.4 (C(CH3)3), 81.8 (C(CH3)3), 106.3‡107.3 (Dha-Cb), 126.8
(para-C(Trt)), 128.0 (ortho-C(Trt)), 129.6 (meta C-(Trt)), 134.0 (Dha-Ca),
134.7 (Dha-Ca), 144.5 (ipso-C (Trt)), 156.2 (OCONH), 156.3 (OCONH),
163.8 (CONH), 164.4 (CONH), 169.1 (CONH), 170.4 (CONH), 171.5
Major isomer H-Lys-Lan-Glu-Dha-Leu-Lan-Ala-OH (A-ring) (6a):
1H NMR (500 MHz, [D6]DMSO): d ˆ 0.84‡0.88 (2d, J ˆ 6, 6 Hz; Leu-
6Hd), 1.23 (d, J ˆ 8 Hz; Ala-3Hb), 1.30 ± 1.39 (m, Lys-2Hg), 1.48 ± 1.53 (m,
Lys-2Hd), 1.60 ± 1.64 (m, Leu-2Hb‡Leu-Hg), 1.67 ± 1.70 (m, Lys-2Hb‡Glu-
Hb), 2.07 ± 2.11 (m, Glu-Hb), 2.18 ± 2.29 (m, Glu-2Hg), 2.71 ± 2.74 (m, Lys-
2He), 2.80 (dd, J ˆ 15, 6 Hz; Lan-Hb), 2.87 (d, J ˆ 10 Hz; Lan-Hb), 2.98 (dd,
J ˆ 10, 6 Hz; Lan-Hb), 3.03 (dd, J ˆ 15, 5 Hz; Lan-Hb), 3.79 ± 3.83 (m, Lys-
Ha), 4.20 (dq; J ˆ 8, 7 Hz; Ala-Ha), 4.22 ± 4.28 (m, Leu-Ha), 4.30 ± 4.41 (m,
Lan-Ha‡Glu-Ha), 4.48 ± 4.50 (m, Lan-Ha), 5.50 (s, Dha-Hb), 6.03 (s, Dha-
‡
Hb), 7.58 ± 7.80 (brs, Lys-NH3 ), 7.88 (d, J ˆ 7 Hz; Lan2-NH), 8.05 (d, J ˆ
‡
8 Hz; Ala-NH), 8.05 ± 8.20 (brs, Lys-NH3 ), 8.51 (s, Dha-NH), 8.74 (d, J ˆ
6 Hz; Lan-NH), 8.77 (d, J ˆ 7 Hz; Leu-NH), 8.50 (d, J ˆ 8 Hz; Glu-NH);
‡
‡
ES-MS: m/z: 700.8 [M‡H] , 723.3 [M‡Na] ; HPLC 1) (l220): 11.0 min.
Minor product H-Lys-Lan-Glu-Dha-Leu-Lan-Ala-OH (A-ring isomer)
(6b): 1H NMR (500 MHz, [D6]DMSO): d ˆ 0.85‡0.89 (2d, J ˆ 6, 6 Hz;
Leu-6Hd), 1.23 (d, J ˆ 8 Hz; Ala-3Hb), 1.33 ± 1.38 (m, Lys-2Hg), 1.50 ± 1.65
Table 2. NOEs found for the major product of cyclisation.[a]
Lys
Lan1
Glu
Dha
Leu
Lan2
Ala
a
b
N
a
b
N
a
b
g
N
b
b'
N
a
b
N
a
b
N
a
b
Lys
a
b
s
s
m
s
s
Lan1
N
a
b
m
m
s
s
m
s
s
s
s
m
w
m
m
Glu
N
a
b
m
w
m
s
m
m
w
m
m
m
s
m
m
g
w
Dha
Leu
N
b
w
w
w
w
s
w
m
s
w
w
w
w
w
w
m
w
w
s
b'
N
a
b
s
m
w
w
s
w
w
m
s
w
m
s
m
s
s
s
N
a
b
w
m
m
s
m
w
w
Lan2
Ala
m
m
m
w
s
w
N
a
b
w
w
s
m
s
w
m
[a] NOEs are given as strong (s), medium (m) or weak (w).
Chem. Eur. J. 2000, 6, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1465 $ 17.50+.50/0
1465

M. Bradley et al.
FULL PAPER
Table 3. NOEs observed for the minor product of cyclisation.[a]
Lys
Lan
Glu
Dha
Leu
Lan
Ala
a
b
g
N
a
b
N
a
b
g
N
b
b'
N
w
a
b
N
a
b
N
a
b
Lys
Lan
Glu
a
b
s
m
s
m
w
s
m
s
g
N
a
b
m
w
m
m
s
m
s
s
m
m
m
w
w
m
m
w
s
s
w
w
N
a
b
s
s
m
s
w
m
s
m
w
m
s
m
m
g
Dha
Leu
Lan
Ala
N
b
w
w
w
w
w
w
w
s
m
m
s
w
w
m
w
w
m
w
w
s
b'
N
a
b
s
m
w
w
s
w
w
m
m
w
m
s
m
N
a
b
w
m
w
m
m
s
m
s
m
m
m
m
s
s
N
a
b
m
w
s
s
s
w
s
[a] NOEs are given as strong (s), medium (m) or weak (w).
(m, Lys-2Hd‡Leu-Hg‡Leu-2Hb‡Lys-2Hb), 1.84 ± 1.89 (m, Glu-Hb), 2.15 ±
2.33 (m, Glu-Hb‡Glu-2Hg), 2.68 ± 2.80 (m, Lys-2He‡Lan1-Hb‡Lan-Hb),
2.98 (dd, J ˆ 14, 6 Hz; Lan-Hb), 3.08 (dd, J ˆ 15, 5 Hz; Lan-Hb), 3.81 ± 3.84
(m, Lys-Ha), 4.12 ± 4.19 (m, Glu-Ha‡Ala-Ha), 4.39 ± 4.44 (m, Leu-Ha),
4.49 ± 4.54 (m, Lan-Ha‡Lan-Ha), 5.52 (s, Dha-Hb), 6.00 (s, Dha-Hb), 7.58 ±
[8] a) A. Polinsky, M. G. Cooney, A. Toy-Palmer, G. Osapay, M. Good-
man, J. Med. Chem. 1992, 35, 4185 ± 4194; b) T. Wakamiya, K. Shimbo,
A. Sano, K. Fukase, T. Shiba, Bull. Chem. Soc. Jpn. 1983, 56, 2044 ±
2049; c) K. Fukase, T. Wakamiya, T. Shiba, Bull. Chem. Soc. Jpn. 1986,
59, 2505 ± 2508.
‡
7.80 (brs, Lys-NH3 ), 7.98 (d, J ˆ 7 Hz; Lan-NH), 8.12 ± 8.20 (brs‡d, J ˆ
[9] N. Schnell, K.-D Entian, U. Schneider, F. Gotz, H. Zähner, R. Kellner,
G. Jung, Nature 1988, 333, 276 ± 278.
‡
8 Hz; Lys-NH3 ‡Ala-NH), 8.44 (d, J ˆ 7 Hz; Leu-NH), 8.51 (d, J ˆ 8 Hz;
Glu-NH), 8.68 (s, Dha-NH), 8.65 (d, J ˆ 6 Hz; Lan-NH); ES-MS: m/z:
[10] L. Ingram, Biochim. Biophys. Acta. 1970, 224, 263 ± 265.
[11] H.-P. Weil, A. G. Beck-Sickinger, J. Metzger, S. Stevanovic, G. Jung,
M. Josten, H.-G Sahl, Eur. J. Biochem. 1990, 194, 217 ± 233.
[12] P. L. Toogood, Tetrahedron Lett. 1993, 34, 7833 ± 7836.
[13] U. Schmidt, A. Lieberknecht, J. Wild, Synthesis 1988, 3, 159 ± 172.
[14] K. W. Li, J. Wu, W. N. Xing, J. A. Simon, J. Am. Chem. Soc. 1996, 118,
7237 ± 7238.
‡
‡
700.8 [M‡H] , 723.3 [M‡Na] ; HPLC 1) (lmax): 11.3 min.
Acknowledgments
[15] K. Goodall, A. Parsons, Tetrahedron Lett. 1995, 36, 3259 ± 3260.
[16] R. J. Cherney, L. Wang, J. Org. Chem. 1996, 61, 2544 ± 2546.
[17] M. J. Miller, J. Org. Chem. 1980, 45, 3131 ± 3132.
[18] C. G. Shin, S. Honda, K. Morooka, Y. Yonezawa, Bull. Chem. Soc. Jpn.
1993, 66, 1844 ± 1846.
M.B. and J.W.E. would like to thank the Royal Society for their generous
support the EPSRC for studentships to S.B., M.C. and C.A., Aplin and
Barrett for supplying nisin as a reference compound, the referees for very
helpful suggestions and David Turner for the NMR studies.
[19] C. Blettner, M. Bradley, Tetrahedron Lett. 1994, 35, 467 ± 470.
[20] M. Yamada, T. Miyajima, H. Horikawa, Tetrahedron Lett. 1998, 39,
289 ± 292.
[21] C.-G. Shin, Y. Yonezawa, T. Yamada, J. Yoshimura, Bull. Chem. Soc.
Jpn. 1982, 55, 2147 ± 2152.
[22] M. Bodanszky, A. Bodanszky, The Practice of Peptide Synthesis, 2nd
ed., Springer, Berlin, 1994.
[23] G. B. Fields, R. L. Noble, Int. J. Peptide Protein Res. 1990, 35, 161 ± 214.
[24] E. Atherton, R. C. Sheppard, Solid Phase SynthesisÐA Practical
Approach, IRL Press, 1989.
[25] S. Burrage, T. Raynham, M. Bradley, Tetrahedron Lett. 1998, 39,
2831 ± 2834.
[26] G. L. Ellman, Arch. Biochem. Biophys. 1959, 70, 82 ± 89.
[27] V. K. Sarin, S. B. H. Kent, J. P. Tam, R. B. Merrifield, Anal. Biochem.
1981, 117, 147 ± 157.
[28] P. Güntert, W. Braun, K. Wüthrich, J. Mol. Biol. 1991, 217, 517 ± 530.
[29] F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M. Lipton,
C. Caufield, G. Chang, T. Hendrickson, W. C. Still, J. Comput. Chem.
1990, 11, 440 ± 467.
[30] G. J. Pernia, J. D. Kilburn, J. W. Essex, R. J. Mortishire-Smith, M.
Rowley, J. Am. Chem. Soc. 1996, 118, 10220 ± 10227.
[1] a) J. N. Hansen, Annu. Rev. Microbiol. 1993, 47, 535 ± 564; b) G. Jung,
Angew. Chem. 1991, 103, 1067 ± 1084; Angew. Chem. Int. Ed. Engl.
1991, 30, 1051 ± 1068; c) H.-G. Sahl, R. W. Jack, G. Bierbaum, Eur. J.
Biochem. 1995, 230, 827 ± 853; d) R. N. H. Konings, C. W. Hilbers
(Eds.), Antonie van Leeuwenhoek Int. J. Gen. Mol. Microbiology,
1996, 69, 85 ± 202.
[2] L. A. Rogers, J. Bacteriol. 1928, 321 ± 325.
[3] G. W. Buchman, S. Banerjee, J. N. Hansen, J. Biol. Chem. 1988, 263,
16260 ± 16266.
[4] S. Banerjee, J. N. Hansen, J. Biol. Chem. 1988, 263, 9508 ± 9514.
[5] H. Allgaier, G. Jung, R. G. Werner, U. Schneider, H. Zähner, Angew.
Chem. 1985, 97, 1052 ± 1054; Angew. Chem. Int. Ed. Engl. 1985, 24,
1051 ± 1053.
[6] H.-G. Sahl, M. Grossgarten, W. R. Widger, W. A. Cramer, H. Brandis,
Antimicrob. Agents Chemother. 1985, 27, 836 ± 840.
[7] a) D. E. Palmer, D. F. Mierke, C. Pattaroni, M. Goodman, Biopoly-
mers 1989, 28, 397 ± 408; b) L.-Y. Lian, W. C. Chan, S. D. Morley,
G. C. K. Roberts, B. W. Bycroft, D. Jackson, Biochem. J. 1992, 283,
413 ± 420; c) H. W. van den Hooven, H. S. Rollema, R. J. Siezen, C. W.
Hilbers, O. P. Kuipers, Biochemistry 1997, 36, 14137 ± 14145;
d) F. J. M. van de Ven, G. Jung, Antonie van Leeuwenhoek Int. J.
Gen. Mol. Microbiol. 1996, 69, 99 ± 107.
Received: June 22, 1999
Revised version: October 8, 1999 [F1868]
1466
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0608-1466 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 8