Concise total synthesis of the thiazolyl peptide antibiotic GE2270 A

Article abstract of DOI:10.1002/chem.200701823

The potent antibiotic thiazolylpeptide GE2270 A was synthesized starting from N-tert-butyloxycarbonyl protected valine in a longest linear sequence of 20 steps and with an overall yield of 4.8 %. Key strategy was the assembly of the 2,3,6-trisubstituted p

Full text of DOI:10.1002/chem.200701823

DOI: 10.1002/chem.200701823
Concise Total Synthesis of the Thiazolyl Peptide Antibiotic GE2270 A
Oscar Delgado,[a, b] H. Martin Müller,[a, c] and Thorsten Bach*[a]
Dedicated to Professor Wolfgang A. Herrmann on the occasion of his 60th birthday
2322
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

FULL PAPER
Abstract: The potent antibioticthiazo-
lylpeptide GE2270 A was synthesized
starting from N-tert-butyloxycarbonyl
protected valine in a longest linear se-
quence of 20 steps and with an overall
yield of 4.8%. Key strategy was the as-
sembly of the 2,3,6-trisubstituted pyri-
dine core by consecutive cross-coupling
reactions starting from 2,6-dibromo-3-
iodopyridine. The complete Southern
fragment was installed by Negishi
cross-coupling of 3-zincated 2,6-dibro-
mopyridine at the terminal 2-iodothia-
zole of a trithiazole (87%). The sub-
stituent at C-6 representing the North-
ern part of the molecule was intro-
duced in form of the truncated tert-
ducted as an intramolecular Stille
cross-coupling occurring at C-2 of the
pyridine core and providing the desired
product in 75% yield. The required
stannane was obtained by amide bond
formation (87%) between a complex
dithiazole fragment representing the
Eastern part of GE2270 A and a 3,6-
disubstituted 2-bromopyridine. Final
steps included attachment of a serine-
proline amide dipeptide to the North-
ern part of the molecule (65%), forma-
tion of the oxazoline ring and silyl
ether deprotection (55% overall).
butyl
2-bromothiazole-4-carboxylate
after metalation to a zincreagent by
another Negishi cross-coupling (48%).
Decisive step of the whole sequence
was the macrocyclization to a 29-mem-
bered macrolactam, which was con-
Keywords:
cross-coupling
antibiotics
macrocycles
·
·
·
natural products · total synthesis
Introduction
The GE2270 antibioticGE2270 A is the longest known
and best studied member of this compound class. It was iso-
lated by Selva et al. from Planobispora rosea ATCC 53773
and its structure was reported in 1991.[5] Later several relat-
ed compounds were reported, which were found in the same
microorganism and which were named GE2270 factors E,
D1, D2, C1, C2a, C2b, B1, B2, and T.[6] They differ from
GE2270 in the substitution or in the binding pattern (single
vs double bond) at the marked positions (a, Figure 1). In
addition there are other structurally related thiopeptides,[7]
which also bear a central triply substituted pyridine frag-
ment, for example, the amythiamicins,[8] thiostreptone,[9] or
micrococcin P.[10]
The increasing number of resistances observed in bacterial
infections poses a significant clinical problem and becomes a
growing threat for many patients suffering from these dis-
eases. The search for antibiotics with new modes of action is
of utmost importance. Unless new antibiotic drugs are
found in the near future, mankind may soon be disarmed in
its fight against infectious diseases.[1] In this context, the bac-
terial elongation factor EF-Tu[2] has been established as a
validated drug target representing an ubiquitous enzyme es-
sential for bacterial protein biosynthesis. It is significantly
different from the human elongation factor eEF-1a guaran-
teeing desirable target specificity. Currently, four structurally
different compound classes are known to inhibit EF-Tu effi-
ciently, the prototypical examples being kirromycin, enacy-
cloxin IIa, pulvomycin, and GE2270 A.[3] It has been shown
that the binding sites of pulvomycin and GE2270 A are
identical, whereas kirromycin and enacycloxin IIa possess
different modes of action. GE2270 A and pulvomycin
hinder the formation of the ternary complex between EF-
Tu, guanosin triphosphate (GTP) and aminoacyl tRNA by
binding to the domain 2 of the enzyme.[4]
[a] Dr. O. Delgado,+ Dr. H. M. Müller,+ Prof. Dr. T. Bach
Lehrstuhl für Organische Chemie 1
Technische Universität München, Lichtenbergstrasse 4
85747 Garching (Germany)
Fax : (+49)89-289-13315
E-mail: thorsten.bach@ch.tum.de
[b] Dr. O. Delgado+
Figure 1. Structure of GE2270 A (1) and locations (a) of structure var-
iation in other GE2270 thiazolyl peptides.
Current address: Janssen-Cilag
Medicinal Chemistry Department, Jarama 75
45007 Toledo (Spain)
[c] Dr. H. M. Müller+
The constitution and configuration of GE2270 A was elu-
cidated by a combination of analytical and synthetic meth-
ods. In the initial assignment the thiazole units D and E
were mistakenly interchanged.[5] The correction of this as-
signment was accompanied by proof for the absolute config-
uration of the valine-, asparagine-, serine- and proline-de-
Current address: Department of Chemistry and Biochemistry
University of Colorado at Boulder, 215 UCB
Boulder, CO 80309-0215 (USA)
[+] Both authors contributed equally to the project.
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2323

T. Bach et al.
rived moieties in the Northern and Southern part of the
molecule.[11] The elusive configurational assignment for the
phenylserine fragment in the Eastern section of GE2270 A
(ring C) was initially based on crystallographic data of a 1:1
molar complex between Escherichia coli EF Tu-GDP (gua-
nine nucleotide diphosphate) and GE2270 A (2.35 Š resolu-
tion).[4a] The assignment was certainly biased by the fact that
most, if not all, phenylserines in natural products derived
from bacteria had been shown to be threo-(2S,3R)-config-
ured.[12] In the context of preliminary studies towards
GE2270 A we prepared the two tetrathiazoles 2 [threo-
(2S,3R)] and 3 [erythro-(2R,3R)] in enantiomerically and
diastereomerically pure form[13] with the intention to com-
pare their spectral data with the data of a degradation prod-
uct of GE2270 A (Figure 2) earlier reported by Tavecchia
et al.[11b]
diketone, as implemented by Ciufolini et al. in their synthe-
sis of the Bycroft–Gowland structure of micrococcin P.[22]
Kelly et al. employed a N-pivaloyl protected 2-amino-3-
bromo-6-ethoxypyridine as a starting material for the syn-
thesis of micrococcinic acid.[23] Substitution reactions at the
pyridine were conducted by cross-coupling at the individual
positions after appropriate activation. The Shin group has
reported the synthesis of an advanced linear precursor of a
diastereoisomer of GE2270 A, following a strategy that
relies on the attachment of the thiazole rings successively to
an appropriately functionalized pyridine core making exten-
sive use of classical condensation reactions.[24,25]
Our own syntheticstrategy to GE2270 A was based on a
regioselective cross-coupling[26] at the central pyridine frag-
ment (Figure 3).[27] Contrary to the above-mentioned cross-
Figure 2. Structure of synthetically prepared diastereomeric tetrathiazoles
2 and 3 and of the degradation product ent-3 obtained from GE2270 A.
Figure 3. Synthetic plan of three consecutive cross-coupling reactions at a
central pyridine core I with idealized coupling partners 4–6 indicated
(PG=protecting group).
Surprisingly, threo-compound 2 was not identical to the
degradation product. Instead, compound 3 proved to be the
enantiomer of the degradation product, to which structure
ent-3 could be unambiguously assigned. As a consequence,
the erythro-(S,S)-configuration was also assigned to the ste-
reogeniecncters at the phenylserine-derived part of
GE2270 A. The structural assignment was confirmed in 2006
when the structure of the ternary complex of T. thermophi-
lus EF-Tu, guanylyl imino diphosphate (GDPNP) and
GE2270 A was solved at 1.6 Š resolution.[4b,14]
coupling strategy,[23] we planned to address the individual
halogenated positions by a judicious choice of reagents,
rather than to activate them successively. The propensity of
2,3,6-trihalogenated pyridine (I) for a regioselective metala-
tion or a regioselective halogen–metal exchange was de-
duced from reactions[28] occurring on 2,3-dihalopyridines[29]
or on 2,5-dihalopyridines.[29a,30] By this means, position C-3
of the central pyridine was to be initially addressed and con-
verted into a nucleophile, which in turn should be cross-cou-
pled (first cross-coupling) with an electrophilic building
block 6. The second cross-coupling was expected for steric
reasons to occur at position C-6 of the pyridine[31] replacing
substituent X6 by an appropriate nucleophile 4 (TBDPS=
The most notable synthetic strategy successfully imple-
mented in the preparation of thiopeptides[15] is the biomim-
[16]
eticapproahc,
according to which the central pyridine
core is built by a hetero-Diels–Alder reaction. This strategy
has been beautifully applied by Moody et al. to the synthesis
of amythiamicin D[17] and by Nicolaou et al. to the synthesis
of thiostrepton.[18] In 2006, Nicolaou, Chen et al. used this
strategy successfully to establish a synthesis of GE2270 A[19]
and of another GE2270 factor, GE2270 T. Additional pyri-
dine ring construction methods, which have been elegantly
applied to the synthesis of thiopeptides, include the Bohl-
mann–Rahtz reaction,[20] as used by Moody et al. in the syn-
thesis of promothiocin A,[21] and the condensation of a 1,5-
ꢀ
tert-butyldiphenylsilyl). Eventually, the C C bond formation
at C-2 was to be conducted as a Stille cross-coupling[32] in
line with our previous experience[33] that sterically encum-
bered heterocyclic ring positions with limited reactivity are
best addressed by Stille or Suzuki cross-coupling reactions.
For this cross-coupling we had originally projected the N-
2324
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
tert-butyloxycarbonyl (Boc) protected Eastern fragment 5c
(TBS=tert-butyldimethylsilyl), which was in its deprotection
scheme orthogonal to the other protecting groups. An alter-
native substrate with a base-labile protecting group was 9-
fluorenylmethyloxycarbonyl (Fmoc) derivative 5a. A macro-
lactamization occurring between the N-terminus of the gly-
cine and the thiazole carboxylic acid D of the Southern frag-
ment were planned to conclude the synthesis of the central
skeleton of GE2270 A. Oxazoline ring closure and deprotec-
tion should in two final steps establish the target molecule.
It turned out that the retrosyntheticroute as depicted in
Figure 3 required modifications as the synthesis progessed.
Introduction of the complete Northern fragment was impos-
sible and a truncated analogue was used instead. More im-
portantly, the third cross-coupling was conducted after the
amide bond formation, which connects the Eastern fragment
via the glycine to thiazole carboxylic acid D (Figure 3) in
the Southern part. This modification, employing an intramo-
lecular Stille cross-coupling for the macrocyclization, im-
proved the synthetic efficiency by a factor of five. Details of
our synthetic endeavor are given in this full account.[34]
exchange was conducted with butyl lithium at ꢀ788C in
Et2O and after transmetalation with ZnCl2 (0.5m solution in
THF) the Negishi cross-coupling[37] was performed using
5 mol% of [PdCl2ACHTRE(UNG PPh3)2] as the catalyst. The yield of dibro-
mide 15 was high if based on recovered starting material
(Scheme 2). The fact that the cross-coupling remained in-
complete with the pyridine being the limiting reagent was
unsatisfactory, however. Yields were variable and depended
on the metalation conditions. In addition, access to pure
2,3,6-tribromopyridine (14) according to the reported proce-
dure[38] was tedious, because other bromopyridines were also
formed and difficult to remove. We therefore decided to re-
investigate the reaction leading to ester 15 in an attempt to
optimize the cross-coupling conditions.
Scheme 2. Synthesis of dithiazole 15 by metalation/Negishi cross-cou-
pling.
Results and Discussion
Preparation of thiazoles, pyridines, and cross-coupling stud-
ies at C-3: A series of thiazole-4-carboxylates was required
either for implementation into the Northern fragment 4 or
into the Southern fragment 6. The readily available ethyl 2-
aminothiazole-4-carboxylate (8),[35] which was prepared
from pyruvate 7, served as valuable intermediate which
could after diazotation be converted into iodide 9 and bro-
mide 12[36] (Scheme 1). The free iodothiazolecarboxylic acid
It was shown that both the bromine–lithium (BuLi, Et2O,
ꢀ788C) and the bromine–magnesium (iPrMgBr, THF, 08C)
exchange proceed completely. If the transmetalation to zinc
(ZnCl2 or ZnBr2 in THF) was conducted after complete bro-
mine–metal exchange, however, the Negishi cross-coupling
did not go to completion under a variety of conditions nei-
ther with thiazolyl bromide 12 nor with iodide 9. Direct
cross-coupling of the magnesium compound (Kumada cross-
coupling) was not feasible, either. Since we felt that an in-
sufficient transmetalation may be responsible for the failure
of the cross-coupling we attempted to prepare the zinc spe-
cies by direct zincation.[39] Expectedly, the zincation (Zn,
TMSCl, BrCH2CH2Br) occurred with high regioselectivity at
the 3-position of tribromide 14 but it could not be driven to
completion even under forcing conditions (DMA, 1508C).
In order to facilitate zincation we prepared 2,6-dibromo-3-
iodopyridine (18, Scheme 3). The best regioselectivitity in its
preparation was achieved if the bromination (NBS=N-bro-
mosuccinimide) was conducted starting from commercially
available 3-aminopyridine (16). Although we were not able
to completely reproduce the yield (92%) reported for this
transformation[40] the method delivered reasonable amounts
of regioisomerically pure 2,6-dibromopyridine 17. The iodo-
de-amination[41] was straightforward and gave the desired
2,6-dibromo-3-iodopyridine (18). The higher yielding
method for its synthesis started from 2,6-dichloropyridine
(19) but suffered from an incomplete regioselectivity in the
deprotonation/iodination step.[42] The 3-iodo product 20 was
after re-crystallization still contaminated with the 4-iodopyr-
idine (regioisomericratio rr=92:8). The bromo-de-chlorina-
tion[43] proceeded smoothly.
Scheme 1. Synthesis of different 2-halothiazole-4-carboxylates.
(10) was obtained by saponification and could easily be con-
verted into the corresponding tert-butyl ester 11 by an acid-
catalyzed tert-butylation. In a similar fashion the tert-butyl
2-bromothiazole-4-carboxylate (13) was prepared from ethyl
ester 12.
In the synthesis of tetrathiazoles 2 and 3 we had already
conducted a regioselective metalation cross-coupling se-
quence starting from tribromide 14.[13] The halogen–metal
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2325

T. Bach et al.
tedious procedure by washing an ethyl acetate solution of 23
with an aqueous AgNO3 solution. After a single chromato-
graphicpurification step the ester 23 was sufficiently pure to
undergo a high yielding hydrogenolysis to the free acid,
which was converted into the primary amide 24 via the
mixed anhydride generated with isobutyl chloroformate
(IBC) and N-methylmorpholine (NMM). Hydrogen and
Pearlmanꢁs catalyst [Pd(OH)2/C] proved in our hands to be
superior reagents for the hydrogenolysis compared with
other typically used reagent combinations, for example, H2
[Pd], HCOOH [Pd/C], or cyclohexene [Pd/C]. In the amide
formation step, an excess of methyl amine and prolonged re-
action times (>1 h) should be avoided to suppress forma-
tion of the undesired diamide, which resulted from a reac-
tion at the methyl ester. After Bocdeprotection the enantio-
mericpurity ( >95% ee) of the resulting amino acid 25 was
assessed by the Mosher method.[49]
Scheme 3. Two approaches for the synthesis of the trihalogenated pyri-
dine 18.
We were pleased to find that the direct zincation of
iodide 18 was feasible at room temperature upon addition
of a THF solution of 18 to activated zinc[39] in DMA. The re-
action was complete after 30 min as indicated by GC control
of the hydrolysis products. The reaction could be conducted
in THF/DMA mixtures up to a ratio of 2.5:1, which proved
to be beneficial for the subsequent cross-coupling step.
Indeed, Negishi cross-coupling with bromide 12 gave re-
producibly a yield of 60% if either [PdCl
A
(PPh3)2] or [Pd2-
(2-furyl)phos-
phane] was used as the catalyst at 458C. As the conversion
was not complete we attempted to further improve the reac-
tion by replacing the bromide 12 with the more reactive
ethyl 2-iodothiazole-4-carboxylate (9). In this reaction, [Pd2-
ACHTREUNG
CHTREUNG
72%. Formation of the ligated Pd0 species was indicated in
the former case by a color change of the solution from red
to grey. Since preliminary studies suggested that not only
the 2-iodothiazole-4-carboxylates but also the corresponding
secondary amides underwent the required cross-coupling re-
actions, we embarked on the synthesis of the complete
Southern fragment with an iodide at the 2-position of thia-
zole ring F (compound 6b, Figure 3).
Scheme 4. Synthesis of the aspartate-derived thiazole 24.
Attempts to obtain the methyl ester 24 or related esters
by reaction of a thioamide derived from carboxylic acid 21
with an appropriate b-bromo-a-ketobutanoate were met
with limited success. Best yields were in the range of 35–
40%. Contrary to that, the same thioamide reacted nicely
and cleanly (80–90%) with the corresponding a-pyruvates
(i.e., b-bromo-a-ketopropanoates) indicating that the steri-
cally more hindered secondary bromide is by far more diffi-
cult to substitute than a primary bromide. With these results
in mind we decided to assemble the valine-derived thiazole
31, also bearing a substituent at C-5, by a Hantzsch synthe-
sis[50] with ethyl a-bromopyruvate and subsequent alkyla-
tion[51] at C-5 rather than by introducing the methoxymethyl
group via the bromide in the thiazole forming step
(Scheme 5). Indeed, Boc-protected thiazole 26 is literature
known and was prepared from N-Boc-protected valine in
four steps and 77% overall yield by the method earlier de-
scribed by Meyers.[52] Attempted alkylation of thiazole 26
proceeded smoothly using three equivalents of LDA in the
deprotonation step. It turned out, however, in a Mosher
analysis conducted at a later stage of the synthesis that
minor racemization had occurred during the alkylation
(89% ee). This observation forced us to switch protecting
groups and to convert the N-Bocprotected amine 26 into its
Synthesis of the Southern fragment: The trithiazole 6 invites
amide bond formation reactions to link the individual thia-
zole fragments (rings D, E, F in Figure 1). Retrosynthetical-
ly, the two more complex thiazoles resulting from this con-
sideration contain stereogenic centers derived from aspar-
tate (ring E) and from valine (ring D). Amino and carboxyl
termini require orthogonal protecting group strategies. We
relied on the Gabriel thiazole synthesis[44] for the synthesis
of thiazole 24 (Scheme 4).[45] Starting from N-Boc-protected
monobenzylated amino acid 21, dipeptide 22 was generated
employing O-(benzotriazol-1-yl)-N,N,N’,N’-tetramethyluro-
nium hexafluorophosphate (HBTU) in the presence of 1-hy-
droxybenzotriazole (HOBt) as the peptide coupling reagent.
[46]
After oxidation with 2-iodoxybenzoicaicd (IBX)
and
treatment of the resulting ketone with the Lawesson re-
agent[47] the desired ring closure occurred uneventfully. The
Lawesson reagent, however, led to significant sulfur impuri-
ties, which severely hampered the subsequent hydrogenolyt-
iclceavage of the benzylester
23.[48] While the impurities
could be removed by repeated (up to ten times) chromato-
graphic purification we found that we could circumvent this
2326
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
N-trityl (Tr) protected analogue 27. The bulkier protecting
group shields the amine proton effectively and allowed de-
protonation at C-5 with 1.1 equivalents of LDA. Methoxy-
methylation with methoxymethyl iodide (MOM-I) delivered
product 28 in very good yield. The use of MOM-I was man-
datory as methoxymethyl chloride (MOM-Cl) was too weak
an electrophile to provide a significant conversion.
Scheme 6. Assembly of the Southern fragment 6b.
in a diastereomericratio of 74:26 and 75:25, respectively, for
product 6b. The result is surprising in light of the fact that
we found the saponification of methyl ester 24—in agree-
ment with literature precedence[17]—to proceed without any
racemization. Presumably, the low solubility and as a conse-
quence the heterogeneous saponification conditions are re-
sponsible for the undesired loss of stereochemical integrity.
The saponification of ester 24 proceeds in a homogenous so-
lution. An electronic factor exerted by the thiazolylacyl
group is less likely as subsequent saponification reactions of
the ethyl ester at thiazole D in products derived from 6b
proceeded racemization-free. Fortunately, the trithiazole for-
mation of D-E-F (24, 29 then 10) proceeded not only devoid
of any racemization but delivered product 6b in even higher
yield than the other sequence. Intermediate 33 was Boc-de-
protected with AcCl in MeOH and subsequently connected
with the free acid 10 to provide the desired Southern frag-
ment 6b.
Scheme 5. Synthesis of the valine-derived thiazole 30.
While the ethyl ester 28 was used for further synthetic
elaboration, the synthesis of benzyl ester 30 is also shown in
Scheme 5 because complete Mosher analysis was conducted
with this compound (>95% ee). Since benzyl ester 30 was
obtained from ethyl ester 28 the enantiomericpurity of this
and any other precursor can be safely assumed. Deprotec-
tion of the trityl group (28 ! 29, 30 ! 31) was straightfor-
ward with TFA (trifluoroacetic acid) in CH2Cl2 as the sol-
vent and proceeded quantitatively. The possibility of further
functionalization at C-5 was briefly evaluated because this
option could be useful if different thiazole rings were to be
introduced into ring D for structure–activity relationship
studies. Indeed, iodination and stannylation reactions at C-5
of 27 were easily accomplished opening up different venues
for further syntheticmanipulations.
Linkage of the Southern and the Northern fragment to the
pyridine core: Given the high density of functional groups
in iodothiazole 6b as compared to the model compound 9, it
was a pleasurable surprise to note that the previously devel-
oped cross-coupling conditions could be transferred to the
reaction of zincated iodide 18 and the Southern fragment
6b. Essentially no modifications were required to assemble
the desired 3-substituted 2,6-dibromopyridine 34 (Figure 4).
Zincated iodide 18 underwent a clean Negishi cross-coupling
with iodide 6b at 458C in a solvent mixture of THF/DMA
Having the three building blocks 10, 24/25, and 29 in hand
we attempted their assembly to trithiazole 6b. N-(3-Dime-
thylaminopropyl)-N’-ethylcarbodiimide
hydrochloride
(EDC)[53] turned out to be an excellent activating reagent to
mediate the required amide bond formation in high yields.
While both connection pathways, that is, F-E-D (10, 25 then
29) or D-E-F (24, 29 then 10) were feasible (Scheme 6) it
turned out that the latter route was more reliable with
regard to the stereochemical integrity of the asparagine-de-
rived thiazole fragment.
using [Pd2ACHTER(UNG dba)3]/tfp (6 mol%) as the catalyst. The yield was
high (87%) and the product 34 was obtained as a single dia-
stereoisomer.
After the straightforward amide-bond formation between
10 and 25 to dithiazole 32 the subsequent saponification led
to a racemization at the stereogenic center in a-position of
ring E as evident from the fact that the corresponding tri-
thiazole was obtained as a mixture of two diastereoisomers
in a ratio of 85:15. Attempts to suppress the racemization
by applying other methods to the saponification reaction
were not successful. Both Me3SnOH (C2H4Cl2, 808C)[54] and
KOSiMe3 (THF/CH2Cl2, RT)[55] led to racemization resulting
The next step according to our initial synthetic strategy
(Figure 3) was to link the complete Northern fragment to
the position C-6 of dibromopyridine 34. Tripeptide 41
(Scheme 7) was targeted as an obvious precursor for zinc re-
agent 4. The synthesis of this building block from carboxylic
acid 33, serine methyl ester hydrochloride (36)[56] and pro-
line amide (40)[57] was straightforward although some time
was spent optimizing the reaction conditions for the peptide
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2327

T. Bach et al.
Boc-dipeptide 43 was readily converted into diketopipera-
zine 44 upon hydrogenolysis (H2, Pd/C in EtOH, 82%
yield). We later found a solution to circumvent the cycliza-
tion reaction making use of the corresponding ammonium
salt obtained after Bocdeprotection (see below).
Figure 4. Structure of the key intermediate 2,6-dibromopyridine 34.
coupling. Eventually, it was found that O-[(ethoxycarbonyl)-
cyanomethylenamino]-N,N,N’,N’-tetramethyluronium tetra-
fluoroborate (TOTU)[58] was best suited to link thiazole car-
boxylicacid 35 and serine ester 36 to dipeptide 37 and that
a reagent combination of O-(benzotriazol-1-yl)-N,N,N’,N’-
tetramethyluronium hexafluorophosphate (TBTU) and 1-
hydroxybenzotriazole (HOBt) was the best choice[59] to es-
tablish the amide bond between carboxylic acid 39 and pro-
line amide (40) generating the desired tripeptide 41. The in-
termediary steps, silyl protection (37 ! 38, DMAP=N,N-di-
methylaminopyridine) and saponification (38 ! 39), pro-
ceeded smoothly in high yields.
Scheme 8. Diketopiperazine formation upon deprotection of dipeptide
43.
The Negishi cross-coupling reactions of appropriately
metalated Northern fragments to dibromide 34 turned out
to be the most frustrating and most time consuming part of
the whole total synthesis project. Metalation was facile
using a direct reductive zincation method[39] or a bromine–
lithium exchange reaction with subsequent transmetalation
to zinc. The success of the metalation was indicated by com-
plete hydro-de-halogenation after aqueous work-up. A host
of different conditions for the Negishi cross-coupling was
screened. In no instance, however, was a cross-coupling re-
action to the electrophile 34 observed. Model electrophiles
(e.g. 2-bromopyridine or dibromopyridine 15) also failed to
react in most cases. The only exception was the successful
cross-coupling of zincated 41 with 2-bromopyridine in THF
employing [PdCl2ACHTRE(UNG PPh3)2] as the catalyst (63% yield). The
success of this reaction led us to speculate that the high
number of acidic protons in the substrate 34 and in the re-
agent may be responsible for the failure of the reaction. It
was reasoned that an explanation for the exclusive isolation
in all cases of the protodebrominated Northern fragment is
that the relatively acidic amide protons quenched the orga-
nozinc intermediate before entering the Pd-catalytic cycle.
Aiming to reduce the number of protons in the zinc reagent
we prepared several analogues of compound 41, which were
metalated and subjected to Negishi cross-coupling condi-
tions with dibromide 34. A selection of these compounds
45–48 is shown in Figure 5, none of which, however, under-
went the desired reaction. The modification of the catalyst
(catalysts, that were employed, include among others [Pd2-
Scheme 7. Synthesis of the Northern fragment 41.
Attempts to prepare tripeptides such as 41 by performing
the coupling in reversed order were hampered by the unde-
sired diketopiperazine[60] formation of the dipeptides from
serine and proline amide. As an example the dipeptide 43,
which was easily obtained from protected serine 42 and pro-
line amide 40,[59] could not be coupled under conventional
conditions to thiazole carboxylic acid 35. Upon attempted
generation of the free amine, rapid cyclization to product 44
was observed prior to any peptide coupling. Even after pro-
longed stirring of the reaction mixture employed for depro-
tection the undesired diketopiperazine 44 could be isolated,
albeit in low yield (Scheme 8). The situation was even worse
if the deprotection was conducted under non-acidic condi-
tions. The N-benzyloxycarbonyl protected analogue of N-
A
[Pd
(dba)3]/P(Cy)3,
[Pd
(PPh3)4],
[PdCl2-
A
ACHTREUNG
ther we settled on the use of a zincated thiazole derived
from ethyl 4-bromo-2-thiazole carboxylate (12, Scheme 1).
The reagent had been previously employed in the cross-cou-
pling with dibromide 15[13] and it was therefore not surpris-
2328
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
Figure 5. Various Northern fragments 45–48 tried in cross-coupling reac-
tions.
Scheme 9. Synthesis of 2-bromopyridine 50 by regioselective Negishi
cross-coupling.
ing that model reactions with 2,6-dibromopyridine were
equally successful using [Pd
A
phine oxide was used in further reactions and the triphenyl-
phosphine oxide was chromatographically removed at a
later stage of the synthesis (see below).
ture of DMA and THF as the solvent. The reaction with di-
bromide 34 was unfortunately far away from being as clean
and as reproducible as the test reaction. If a large excess of
the zincreagent was used (20 equiv) the desired ocupling
product could be obtained together with 50% of unreacted
starting material. A separation of the product and the start-
ing material by column chromatography was impossible. An
important breakthrough was achieved, when the order of re-
agent addition was inverted. Previously, the organozinccom-
pound, which was obtained by reductive metalation, had
been added to a solution of the catalyst and dibromide 34 in
an appropriate solvent. Now, the organozinccompound and
the catalyst were stirred in a separate flask and a solution of
dibromide 34 was slowly added. By this means, a more re-
producible procedure was established, which allowed for fur-
ther optimization. In the first place, the ethyl ester 12 was
replaced by tert-butyl ester 13 as starting material, because
the former compound did not provide the required protect-
ing group orthogonality. Secondly, a catalyst screening was
Assembly of the Eastern fragment and macrolactamization:
With 2-bromopyridine 50 in hand we turned our interest to
the remaining substitution reaction at the pyridine core
(Scheme 10). The required Eastern fragment for cross-cou-
pling at C-2 of 50 was prepared in close analogy to a
method previously described.[61] Syntheticfocus was put on
the Fmoc-protected fragment 5a instead of the Boc-frag-
ment 5c (Figure 3) because the introduction of the tert-butyl
ester via thiazole G in 50 (Scheme 9) required the removal
of the glycine N-protecting group under basic rather under
acidic conditions. The known O-protected aminoalcohol 51
was coupled with Fmoc-protected glycine using bromotri-
(pyrrolidino)phosphonium hexafluorophosphate (PyBrop)
as the coupling reagent. Earlier research in our group had
established that regioselective cross-coupling reactions at
2,4-dibromothiazole (53) can be conducted as either Negishi
conducted, from which [PdCl2ACHTERU(GN PPh3)2] and [PdCl2ACHTREU(GN dppf)]
emerged as best suited, whereas many other palladium cata-
lyst failed to provide any cross-coupling reaction or resulted
in rapid double substitution at C-6 and C-2. Catalyst load-
ings needed to be high to guarantee full conversion. Thirdly,
the progress of the reaction was monitored with time. It was
found that with 30 mol% of [PdCl2ACHTRE(UNG PPh3)2] as the catalyst
and with 8 equiv of zincreagent 49 (Scheme 9) the reaction
proceeded cleanly to the desired cross-coupling product 50
with a conversion of about 70% after three hours. After
3.5 h side products became visible with the doubly coupled
(C-6 and C-2) being the most severe. As a consequence, it
was recommendable to stop the reaction after three hours.
The crude product yield was around 60% but triphenylphos-
phine oxide, which was one major impurity, could not be re-
moved by flash chromatography. An HPLC separation (re-
versed phase: ODS-A, MeCN/H2O 70:30, 15 mLminꢀ1) was
required to obtain material of sufficient analytical purity.
For practical purposes, the material obtained after flash
chromatography containing about 15–20% triphenylphos-
Scheme 10. Synthesis of the Eastern fragment 5b.
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2329

T. Bach et al.
or Stille cross-coupling reactions.[62] Given the propensity of
the Fmocgroup for basiclceavage we opted for a Stille
cross-coupling in this case. To this end, bromide 52 was con-
verted into the corresponding stannane by a Pd-catalyzed
stannyl-de-bromination with hexamethylditin.[63] The latter
intermediate was not isolated but directly coupled with di-
bromide 53 to yield the desired dithiazole[64] 54. The stan-
nyl-de-bromination reaction was repeated under similar con-
ditions on this 4-bromodithiazole yielding the Eastern frag-
ment 5a, which could be deprotected to yield the free
amine 5b.
amino acid generated the desired macrocycle 56 albeit in
low yield (20–30% over three steps). Other peptide cou-
pling reagents were also employed, leading to no conversion
(PyBrop) or even lower yield (10%, HATU, HOAt). Our
observations support the findings of Nicolaou, Chen et al.
that the macrolactamization to the basic GE2270 skeleton
was a low-yielding process.[19] Indeed, the macrolactamiza-
tion in their synthesis of GE2270 A, which was conducted
with pentafluorophenyl diphenylphosphinate (FDPP) as the
coupling agent at the very same junction, had provided a
yield of 30%. More significantly, they had found that mac-
rolactamization reactions at the other possible amide bonds
were even lower in yield or did not occur at all.
The apparent difficulties with the macrolactamization
prompted us to explore a different perhaps even more ele-
gant approach to the assembly of compound 56. We fancied
that the initial formation of the amide bond between the
glycine part of 5b and the thiazole carboxylic acid at ring D
was feasible in an intermolecular fashion and the subsequent
macrocyclization would possibly occur in a subsequent step
by a Stille cross-coupling. It was decided to use the same set
of reactions (DPPA-mediated peptide coupling and [Pd-
We were pleased to find that the final Stille cross-coupling
of dithiazole 5a at 2-bromopyridine 50 did not require much
optimization. The use of standard conditions generally em-
ployed for the Stille coupling of 4-stannylated thiazoles (tol-
uene, [PdACHTREUNG(PPh3)4], 908C, 12–18 h) led to the predominant
formation of the desired pyridine 55 in 52% yield
(Scheme 11). It was found that in order to get full conver-
sion, at least 1.6 equivalents of the eastern subunit 5a were
necessary. The use of a more polar solvent (dioxane) or ad-
ditives (CuI) also led to the formation of the desired prod-
uct but in lower yields. The deprotection of the Fmoc group
in product 55 was not trivial. Standard conditions led to
moderate yields (50–70%) and in some runs epimerization
at an unidentified stereocenter occurred. However, the final
key macrolactamization could be attempted once the corre-
sponding carboxylic acid was released with LiOH. To our
delight, the use of diphenylphosphoryl azide (DPPA) and
Hünigꢁs base in a diluted solution of the corresponding
AHCTRE(UNG PPh3)4]-catalyzed Stille coupling) but in the reverse order.
The stability of the 4-stannylated thiazole was crucial, since
it would have to remain inert towards a range of reaction
conditions. Luckily, the Fmoc protection offered the advant-
age of being easily cleaved in the presence of an amine
base. In the event, treatment of bisthiazole 5a with piperi-
dine in DMF followed by concentration of the reaction mix-
ture in vacuo yielded the desired amine 5b, ready to be
trapped in the subsequent peptide-bond formation. Saponi-
fation of ester 50 to the free carboxylic acid 57 was straight-
forward and proceeded epimerization-free (Scheme 12). As
in the intramolecular case, the reactant chosen for the
amide-bond formation was DPPA and the macrocycle pre-
cursor 58 was obtained in 87% yield (two steps from the
ethyl ester 50). Amazingly, the yield in the intermolecular
reaction exceeded that obtained in the macrolactamization
attempts by a factor of almost three. The stage was set for
the decisive step. This reaction would combine in a single
syntheticoperation two of the presumably most challenging
tasks of the complete synthesis: the formation of the fully
substituted pyridine core and the formation of the 29-mem-
bered macrocycle. To our delight, the conditions employed
in the previous Stille coupling (that led to the acyclic precur-
sor 55) could be successfully adapted to the macrocycliza-
tion reaction. In order to avoid the formation of any poly-
mers the substrate and the catalyst were stirred for three
days in hot toluene (908C) under high dilution conditions
(c=0.001m). The yield observed (75%) was excellent,
taking into account the complexity of the process. Interest-
ingly, the only product that could be isolated from this step
was the macrocycle 56, even in cases in which the starting
material was contaminated with phosphine oxides or with
isomers arising from the previous cross-coupling steps. The
intramolecular Stille cross-coupling depicted in Scheme 12 is
with regard to molecular size (MW =1.54 kDa) and function-
Scheme 11. Macrocyclization by amide-bond formation.
2330
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
tected in this event, the TBS group was removed to produce
the monoprotected dipeptide 59, which was further N-de-
protected with TFA to produce the ammonium salt 60
(Scheme 13).
Upon mixing this freshly generated ammonium salt with
Scheme 13. Final steps of the total synthesis of GE2270 A (1).
Scheme 12. Macrocyclization by Stille cross-coupling.
the coupling reagent TOTU and the free carboxylic acid in
DMF and adding in immediate succession an excess of base,
we could efficiently generate the desired amide 61 in very
short reaction times. With this procedure, which reveals the
serine amine in the presence of the coupling reagent and
the acid, we exploit the higher reaction rate of the peptide
coupling step versus the diketopiperazine formation.[67] For
the oxazoline formation we made use of N,N-(diethylami-
no)sulfur trifluoride (DAST), a reagent that has become
very popular[68] for the formation of different heterocycles in
total synthesis. It offers several advantages, for example, the
convenient handling of the reagent and the high reactivity
(which allows the reaction to be carried out at low tempera-
tures). With the limited amounts of substrate we had in
hand there was very little room for reaction optimization.
The first experiments were conducted with a previously pre-
pared 0.1m stock solution. However, the result from these
tests was the almost complete recovery of the starting hy-
droxyamide 61. This behavior was initially attributed to the
common difficulties associated with the reactions that are
carried in such a small scale (ca. 1 mmol). Some unsuccessful
tests with Burgess reagent led us back to the more selective
DAST. To our delight, when a large excess of DAST (20–
30 equiv) was added to a solution of 61 in CH2Cl2 at ꢀ788C
total conversion to the desired oxazoline was observed.
With the oxazoline in hand, only the deprotection of the
phenylserine alcohol remained to be done. The experience
we had accumulated during the project in the handling of
al group density one of the most complex reactions of this
type ever conducted.[65] It appears, as if the thiazole rings B-
F exerted a positive template effect for the cyclization and
rather helped than hurt the desired reaction.[66] Nonetheless,
the reaction is testimony to the immense synthetic potential
of cross-coupling methodology, in general, and of the Stille
cross-coupling, specifically.
Conclusion of the synthesis and analytical data: With the
macrocycle 56 in hand, only one amide bond remained to be
formed. The hydrolysis of the tert-butyl ester proceeded
smoothly upon addition of TFA in CH2Cl2, without affecting
the remaining silyl ether. It was anticipated that the key
issue in the final peptide coupling would not be the choice
of coupling reagent but the handling of the dipeptide 43
(Scheme 8) and analogues thereof. It is well known that un-
protected serine-proline fragments react in an intramolecu-
lar fashion to form diketopiperazines (Scheme 8).[60] One
possible detour would be to build up the remaining frag-
ment stepwise attaching first the serine to the free thiazole
carboxylic acid and then the proline amide.[19] We reasoned,
however, that the diketopiperazine formation could be
halted by having a Bocprotecting group at the serine nitro-
gen, since the acidic conditions employed for the deprotec-
tion would generate the corresponding ammonium salt,
which in turn cyclizes only slowly to the diketopiperazine
(see above). As there was no need for serine to be O-pro-
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2331

T. Bach et al.
Table 1. 13C NMR data of natural and syntheticGE2270 A in compari-
son.
the TBS protecting group (both in the introduction and re-
moval of silyl ethers at that position in intermediates in the
erythro and threo series) indicated that a basic fluoride
source (such as TBAF) would be suitable to affect the de-
sired desilylation. This transformation could be also carried
out in acidic or neutral media although requiring longer re-
action times. In the event, treatment of a solution of the
silyl ether in THF with TBAF (2.5 equiv) yielded after one
hour a sample containing GE2270 A (1). Unfortunately, our
target molecule was accompanied by considerable amounts
of tetrabutylammonium impurities derived from the desily-
lating reagent. This problem that has been often faced by
others in the deprotection leading to very polar com-
pounds[69] and could in our hands only be circumvented by
the use of HPLC. While this additional purification step
lowered the combined yield in the final two steps to 55% it
did deliver material of satisfying analytical purity.
Ring Assignment
Selva
Tavecchia
et al.[b][11b]
This
et al.[a][5b]
work[c]
E:
D:
D:
C5-CH3
11.8
11.7
17.7
18.2
11.9
17.8
18.4
CH
CH
A
E
G
G
Pro: C-4
E: NHCH3
Pro: C-3
24.1
25.7
29.6
33.9
24.0
25.6
29.5
33.8
24.1
25.7
29.6
33.9
D:
E:
CH
(CH3)2
CH2CONHCH3 37.6
37.4
37.4
Gly: CH2
Pro: C-5
41.1
46.9
41.0
46.8
41.0
46.9
E:
D:
CH
CH
CH
48.1
55.3
58.1
47.9
55.2
58.0
47.9
55.2
58.0
(NHGly)
D:
CH2OCH3
58.5
58.4
58.5
Pro: C-2
60.0
59.9
60.0
D:
H:
H:
CH2OCH3
C-4
C-5
67.2
67.9
69.3
67.1
67.7
69.2
67.3
67.8
69.3
The identity of the syntheticmaterial and the natural
product was based on the comparison of NMR data. The
presence of many isolated protons and longer sequences of
carbon atoms without any attached protons make extensive
PhCHOH
C-5
C-5
C-5
C-5*
73.8
73.6
73.7
C:
A:
B:
F:
116.3
118.5
122.9
126.5
126.5
127.5
127.6
127.8
128.5
139.4
141.0
141.6
141.6
142.0
143.5
144.7
146.7
149.4
150.1
150.3
153.3
159.2
160.2
160.4
161.1
161.3
164.7
165.7
167.7
168.1
168.3
169.4
169.7
170.9
173.5
116.1
118.4
122.9
126.5
126.6
126.6
127.4
127.4
127.7
139.3
140.8
141.1
141.5
141.9
143.5
144.5
146.7
149.2
149.9
150.1
153.2
159.1
160.0
160.2
161.0
161.1
164.5
165.4
167.6
168.0
168.2
169.3
169.6
170.2
173.4
116.3
118.5
123.0
126.6
126.9
127.5
127.6
127.8
128.6
139.3
140.7
141.2
141.7
141.9
143.6
144.6
146.7
149.3
150.0
150.2
153.2
159.2
160.1
160.3
161.0
161.2
164.6
165.4
167.6
168.1
168.3
169.4
169.6
171.0
173.5
1
delay times (10 s) a requirement to obtain H and 13C NMR
spectra of reasonable quality. As an example the single pro-
tons at H-5 of thiazole rings B and C would deliver a
1H NMR integration of only 0.5 if normal delay times of 1 s
were used. In Table 1 the 13C NMR data are collected for
our final product as compared to the data previously report-
ed by Selva et al.[5b] and by Tavecchia et al.[11b] for the natu-
ral product. Assignments are given for the pyridine ring A,
for the thiazole rings B–G, for the oxazoline ring H, and for
the amino acid derived fragments proline (Pro), phenylser-
ine (Phe) and glycine (Gly). Assignments which can be mu-
tually interchanged are marked. The fit of the 13C NMR
data is excellent with the 1H NMR data also perfectly
matching the reported values.
Phe: C-2, C-6*
Phe: C-4
A:
C-3
Phe: C-3, C-5
G:
E:
D:
A:
C-5
C-5
C-5
C-4
Phe: C-1
E:
D:
G:
C:
F:
A:
A:
B:
H:
D:
B:
F:
E:
F:
D:
H:
G:
E:
C-4
C-4
C-4
C-4
C-4
C-2#
C-6#
C-4
C-2
CO
C-2
CO
CO
C-2
C-2
CO
CO
CO
Conclusion
In summary, a concise and convergent synthesis of the thia-
zolylpeptide GE2270 A was achieved, which proceeds with
an overall yield of 4.8% along the longest linear sequence
starting from N-Bocprotected valine (20 steps). The single
most remarkable reaction step is the facile macrocyclic ring
closure to a 29-membered ring by an intramolecular Stille
cross-coupling reaction (75% yield). The step completes the
assembly of elaborated fragments to a central pyridine core
employing regioselective cross-coupling and metalation re-
actions. The strategy allows a facile variation of all sites in
the macrocycle and may therefore be valuable for the syn-
thesis of GE2270 A analogues. In addition, the modular ap-
proach opens a venue to replace the thiazoles by other
hetero- and homocyclic rings and it should allow an increase
of the notoriously low hydrophilicity of the compound. Pre-
vious experiments in this area[70] were limited to the North-
ern fragment, because it was the only part of the molecule,
which could be semisynthetically addressed. The concise
Gly: CO
E:
C:
CONHCH3
C-2
Pro: CONH2
[a] [D6]DMSO, 125 MHz, 338C. [b] [D6]DMSO, 125 MHz, 358C.
[c] [D6]DMSO, 150 MHz, 258C.
synthesis we have developed should therefore be a good
starting point for evaluating the chemical genetics of the
GE2270 family of antibiotics.
2332
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
erate (15.0 mL, 17.0 mg, 122 mmol). The reaction mixture was stirred for
16 h at room temperature. The reaction was quenched by the slow addi-
tion of saturated aq. NaHCO3 (50 mL). After 20 min the aqueous layer
was extracted with EtOAc (2”100 mL), and the combined organic ex-
tracts were dried (Na2SO4) and concentrated in vacuo. The residue was
purified by flash chromatography (P/Et2O 4:1) to yield the title com-
pound 11 (223 mg, 0.72 mmol, 86%) as a colorless solid. M.p. 888C; Rf =
0.61 (P/Et2O 3:1); 1H NMR (360 MHz, CDCl3): d=8.00 (s, 1H; H-5),
Experimental Section
General: All reactions involving water-sensitive chemicals were carried
out in flame-dried glassware with magneticstirring under argon. Tetrahy-
drofuran (THF), diethyl ether (Et2O) and 1,2-dimethoxyethane (DME)
were distilled from sodium immediately prior to use. Dichloromethane,
triethylamine, pyridine and diisopropylethylamine were distilled from cal-
cium hydride. All other chemicals were either commercially available or
prepared according to the cited references. TLC: Merck glass sheets
(0.25 mm silica gel 60, F254), eluent given in brackets. Detection was per-
formed by UV or coloration with cerium ammonium molybdate (CAM)
(CH3)3]; 13C NMR (90 MHz, CDCl3): d=159.1, 151.0,
1.60 ppm [s, 9H; CACHTREUNG
132.5, 100.9, 82.8, 28.3 ppm; IR (KBr): n˜ =3126 (w), 2969 (w), 1709 (s),
1496 (m), 1393 (s), 1365 (m), 1314 (m), 1250 (s), 1160 (m), 1088 (w), 979
(m), 869 (w), 844 (w), 797 (w), 749 (m) cmꢀ1; HRMS (EI): m/z: calcd for
C8H10NO2SI: 310.9477 [M+], found 310.9480.
or KMnO4. H and 13C NMR spectra were recorded at ambient tempera-
1
ture unless otherwise indicated. Chemical shifts are reported relative to
tetramethylsilane as internal standard. Apparent multiplets which occur
as a result of the accidental equality of coupling constants of magnetical-
ly nonequivalent protons are marked as virtual (virt.). Flash chromatog-
raphy was performed on silica gel 60 (Merck, 230–400 mesh) (ca. 50 g for
1 g of material to be separated) with the indicated eluent. Common sol-
vents for chromatography [pentane (P), ethyl acetate (EtOAc), diethyl
ether (Et2O), dichloromethane (CH2Cl2)] were distilled prior to use. The
preparation and analytical data of compounds 5b, 20, 30, 32, 37–39, 41–
44, 55 are reported in the Supporting Information. 13C NMR spectra of
all new compounds are also presented in the Supporting Information.
2-Bromothiazole-4-carboxylic acid ethyl ester (12):[36] To a stirred solu-
tion of 8 (7.98 g, 31.5 mmol), CuSO4·5H2O (15.7 g, 63.1 mmol) and NaBr
(13.0 g, 126 mmol) in 50% H2SO4 (90 mL) at ꢀ108C was added a pre-
cooled solution of NaNO2 (2.61 mg, 37.8 mmol) in H2O (25 mL) drop-
wise. The mixture was stirred for 1 h at ꢀ58C and 2 h at room tempera-
ture and then water (120 mL) was added. The aqueous layer was extract-
ed with CH2Cl2 (5”80 mL) and the combined organic extracts were
washed with 5% aq. NaHCO3 (80 mL) and saturated aq. NaCl (40 mL)
and dried (Na2SO4). The solvent was removed in vacuo and the residue
was purified by flash chromatography (P/Et2O 3:1) to yield the desired
ethyl ester 12 (3.35 g, 14.2 mmol, 45%) as a colorless solid. M.p. 708C;
Rf =0.28 (P/Et2O 3:1) ; 1H NMR (360 MHz, CDCl3): d=8.11 (s, 1H; H-
5), 4.42 (q, 3J=7.1 Hz, 2H; CH2CH3), 1.40 ppm (t, 3J=7.1 Hz, 3H;
CH2CH3); 13C NMR (90 MHz, CDCl3): d=160.3, 147.5, 136.9, 130.9, 62.0,
14.4 ppm.
2-Iodothiazole-4-carboxylic acid ethyl ester (9)
Synthesis of 2-aminothiazole-4-carboxylic acid ethyl ester: A mixture of
thiourea (1.78 g, 23.4 mmol) and ethyl 3-bromopyruvate (7) (2.94 mL,
4.56 mmol, 21.0 mmol) was heated to 1008C for 1 h. After cooling down
to room temperature the crude product was washed with acetone
(20 mL) and dried in vacuo to give 2-aminothiazole-4-carboxylic acid
ethyl ester hydrobromide (8) (4.02 g, 15.9 mmol, 76%). To obtain the 2-
aminothiazole-4-carboxylic acid ethyl ester the hydrobromide salt 8 was
dissolved in CH2Cl2 and washed with saturated aq. NaHCO3, dried
(Na2SO4) and the solvent removed in vacuo.
2-Bromothiazole-4-carboxylic acid tert-butyl ester (13)
Saponification: Ethyl ester 12 (3.25 g, 13.8 mmol) was dissolved in
MeOH (40 mL) and H2O (10 mL). Potassium carbonate (19.0 g,
138 mmol) was added and the stirring was continued for 3 h at room tem-
perature. The reaction mixture was partitioned between H2O (300 mL)
and CH2Cl2 (200 mL). The aqueous layer was acidified to pH ꢁ1 with 3n
HCl solution. The aqueous layer was extracted with CH2Cl2/THF (10:1,
3”250 mL). The combined organic layers were dried (Na2SO4) and con-
centrated to yield the carboxylic acid 35 (2.88 g, 13.8 mmol, 99%) as a
colorless solid.
2-Iodothiazole-4-carboxylic acid ethyl ester (9): To a stirred solution of 2-
aminothiazole-4-carboxylic acid ethyl ester (344 mg, 2.00 mmol) in THF
(4 mL) at 08C was added consecutively diiodomethane (0.8 mL, 2.66 g,
10.0 mmol) and slowly tert-butylnitrite (1.00 mL, 0.87 g, 8.40 mmol). The
reaction mixture was allowed to reach room temperature and the stirring
was continued for 2.5 h. The mixture was concentrated in vacuo and the
residue was dissolved in EtOAc(150 mL), washed with saturated aq.
NaHCO3 (50 mL) and saturated brine (50 mL), dried (Na2SO4) and the
solvents were removed in vacuo. The crude product was purified by flash
chromatography (P/Et2O 3:1) to yield the desired compound 9 (426 mg,
1.50 mmol, 75%) as a colorless solid. M.p. 828C; Rf =0.31 (P/Et2O 3:1);
1H NMR (360 MHz, CDCl3): d=8.13 (s, 1H; H-5), 4.43 (q, 3J=7.1 Hz,
Esterification: To a stirred solution of 2-bromothiazole-4-carboxylic acid
(35) (2.88 g, 13.8 mmol) in CH2Cl2 (80 mL) and THF (48 mL) was added
tert-butyl 2,2,2-trichloroacetimidate (6.06 g, 27.7 mmol) and boron tri-
fluoride diethyletherate (250 mL, 288 mg, 2.00 mol). The reaction mixture
was stirred for 16 h at room temperature. The reaction was quenched by
the slow addition of saturated aq. NaHCO3 (150 mL). After 20 min the
aqueous layer was extracted with EtOAc (2”250 mL), and the combined
organicextracts were dried (Na 2SO4) and concentrated in vacuo. The res-
idue was purified by flash chromatography (P/Et2O 4:1) to yield the title
3
2H; CH2CH3), 1.41 ppm (t, J=7.1 Hz, 3H; CH2CH3); 13C NMR (CDCl3,
90 MHz): d=160.0, 149.6, 133.4, 101.3, 62.0, 14.5 ppm; IR (KBr): n˜ =
3128 (w), 2979 (w), 2360 (s), 2341 (m), 1708 (s), 1500 (m), 1404 (m), 1304
(m), 1234 (s), 1021 (m), 988 (m), 752 (w) cmꢀ1; HRMS (EI): m/z: calcd
for C6H6NO2SI: 282.9164 [M+], found 282.9163.
compound 13 (2.70 mg, 10.2 mmol, 74%) as
1008C; Rf =0.52 (P/Et2O 3:1); 1H NMR (360 MHz, CDCl3): d=7.98 (s,
1H; H-5), 1.56 ppm [s, 9H; C
(CH3)3]; 13C NMR (90 MHz, CDCl3): d=
a colorless solid. M.p.
AHCTREUNG
159.1, 148.5, 136.4, 130.0, 82.7, 28.0 ppm; HRMS (EI): m/z: calcd for
C8H10BrNO2S: 262.9616 [M+], found 262.9621.
2-Iodothiazole-4-carboxylic acid (10): 2-Iodothiazole-4-carboxylic acid
ethyl ester (9) (2.04 g, 7.20 mmol) was dissolved in MeOH (20 mL) and
H2O (5 mL). Potassium carbonate (9.90 g, 72.0 mmol) was added and the
stirring was continued for 3 h at room temperature. The reaction mixture
was partitioned between H2O (150 mL) and CH2Cl2 (100 mL). The aque-
ous layer was acidified to pH ꢁ1 with 3n HCl solution. The aqueous
layer was extracted with CH2Cl2/THF (10:1, 3”100 mL). The combined
3-Amino-2,6-dibromopyridine (17):[40] To a stirred suspension of 3-amino-
pyridine (16) (133 mg, 1.40 mmol) in CCl4 (7.5 mL) was added N-bromo-
succinimide (503 mg, 2.8 mmol). The reaction mixture was stirred in
darkness for 2 h at room temperature. Silica gel, CH2Cl2 (15 mL), THF
(15 mL) and triethylamine (3 mL) were added to the mixture and con-
centrated in vacuo. The residue was purified by flash chromatography (P/
EtOAc4:1 and 1% triethylamine) to give the title ocmpound
17
organiclayers were dried (Na SO4) and concentrated to yield the carbox-
2
(164 mg, 0.65 mmol, 46%) as a colorless solid. The yield was varying be-
ylicacid 10 (1.82 g, 7.10 mmol, 99%) as a colorless solid. M.p. 1808C (de-
composition); H NMR (360 MHz, CDCl3, CD3OD): d=8.05 ppm (s, 1H;
1
tween 20 and 46%. Polymerisation occurred as a side reaction. Rf =0.29
3
(P/EtOAc5:1); 1H NMR (360 MHz, CDCl3): d=7.20 (d, J=8.2 Hz, 1H;
H-5); 13C NMR (90 MHz, CDCl3, CD3OD): d=161.3, 149.6, 133.6,
101.7 ppm; IR (KBr): n˜ =3095 (s), 1672 (s), 1484 (w), 1433 (m), 1319 (w),
1228 (m), 1112 (w), 985 (s), 739 (w) cmꢀ1; HRMS (EI): m/z: calcd for
C4H2NO2SI: 254.8851 [M+], found 254.8852.
H-4), 6.90 (d, 3J=8.2 Hz, 1H; H-5), 4.15 ppm (brs, 2H; NH2); 13C NMR
(CDCl3, 90 MHz): d=141.3, 127.7, 127.4, 126.6, 124.4 ppm.
2,6-Dibromo-3-iodopyridine (18)
2-Iodothiazole-4-carboxylic acid tert-butyl ester (11): To a stirred solution
of 10 in CH2Cl2 (5 mL) and THF (3 mL) was added tert-butyl 2,2,2-tri-
chloroacetimidate (424 mg, 1.94 mmol) and boron trifluoride diethyleth-
Synthesis from 17: Compound 17 (916 mg, 3.64 mmol) was suspended in
6n H2SO4 (6 mL). A solution of sodium nitrite (376 mg, 5.50 mmol) in
H2O (1.5 mL) was added dropwise at 08C. After 20 min a solution of po-
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2333

T. Bach et al.
tassium iodide (1.33 g, 8 mmol) in H2O (3 mL) was added and the reac-
tion mixture was allowed to reach room temperature. The stirring was
continued for 20 min and potassium carbonate was added until a pH 10
was reached. The aqueous layer was extracted with EtOAc (2”150 mL)
and the combined organic extracts were dried (Na2SO4) and concentrated
in vacuo. The residue was purified with flash chromatography (P/Et2O
40:1) to give a brown solid. A second purification by flash chromatogra-
phy (P/Et2O 40:1) yielded the desired compound 18 (750 mg, 2.06 mmol,
57%) as a colorless solid. M.p. 848C; Rf =0.55 (P/Et2O 20:1); 1H NMR
(360 MHz, CDCl3): d=7.89 (d, 3J=8.1 Hz, 1H; H-4), 7.18 ppm (d, 3J=
8.1 Hz, 1H; H-5); 13C NMR (90 MHz, CDCl3): d=149.9, 147.5, 140.1,
128.1, 98.2 ppm; IR (KBr): n˜ =3094 (m), 2360 (m), 1518 (s), 1390 (s),
1309 (s), 1100 (m), 995 (s), 824 (m), 764 (m), 646 (w) cmꢀ1; HRMS (EI):
m/z: calcd for C5H2NBr2I: 360.7599 [M+], found 360.7598.
(KBr): n˜ =3345 (w), 2976 (w), 1715 (s), 1498 (w), 1333 (w), 1224 (w),
1167 (m) cmꢀ1; HRMS (EI) calcd for C21H26N2O6S: 434.1511 [M+], found
434.1514.
2-(1-tert-Butoxycarbonylamino-2-methylcarbamoylethyl)-5-methylthia-
zole-4-carboxylic acid methyl ester (24)
Hydrogenolysis: Pd(OH)2 on carbon (200 mg, 7 mol%) was added to a
solution of thiazole 23 (829 mg, 1.91 mmol) in MeOH (20 mL). The at-
mosphere in the reaction vessel was changed to hydrogen (1 atm) and
stirred at 608C for 16 h. After cooling to room temperature, the reaction
mixture was filtered through Celite and washed with MeOH (20 mL).
The combined organic layer was concentrated in vacuo to yield the crude
acid as a colorless solid.
Reaction with methylamine: To a stirred solution of the above carboxylic
acid in THF (15 mL) at ꢀ258C was added N-methylmorpholine (210 mL,
193 mg, 1.91 mmol) followed by isobutyl chloroformiate (210 mL, 193 mg,
1.91 mmol). After 10 min aq. methylamine (40%; 178 mL, 2.30 mmol)
was added and the stirring was continued for 1 h while the mixture was
allowed to warm to room temperature. The mixture was concentrated in
vacuo and the crude product was dissolved in EtOAc (300 mL). The or-
ganiclayer was washed with saturated aq. NH 4Cl (100 mL), saturated
brine (100 mL), dried (Na2SO4) and concentrated in vacuo. Purification
by flash chromatography (EtOAC) yielded the title compound 24
ACHTREUNG(S,S)-2-(3-Benzyloxycarbonyl-2-tert-butoxycarbonylaminopropionylami-
no)-3-hydroxybutyric acid methylester (22): To a stirred solution of (S)-
N-a-(tert-butyloxycarbonyl)-aspartic acid b-benzyl ester (21) (8.88 g,
27.5 mmol) in DMF (120 mL) at ꢀ258C was added a solution of the (S)-
threonine methyl ester hydrochloride (H-Thr-OMe) (5.60 g, 33.0 mmol)
in DMF (14 mL), HBTU (12.5 g, 33.0 mmol) and HOBt (5.18 g,
33.0 mmol), followed by the slow addition of triethylamine (14 mL,
10.2 g, 0.10 mol) over 2 h. The stirring was continued for 16 h while the
mixture was allowed to warm to room temperature. The solvent was re-
moved in vacuo and the residue dissolved in EtOAc (500 mL). The re-
sulting solution was washed with saturated aq. NH4Cl (250 mL), saturat-
ed aq. NaHCO3 (250 mL), saturated brine (250 mL) and dried (Na2SO4).
After the solvent was removed the crude product was purified by flash
chromatography (P/EtOAc 1:1) to yield the dipeptide 22 (10.1 g,
23.1 mmol, 84%) as a white foam. M.p. 488C; Rf =0.20 (P/EtOAc1:1);
(567 mg, 1.59 mmol, 83%) as a colorless solid. Rf =0.33 (EtOAc); [a]D20
=
1
ꢀ35.9 (c=0.10 in MeCN), H NMR (360 MHz, CDCl3): d=6.62–6.60 (m,
1H; NH), 6.19 (brs, 1H; NH), 5.18–5.13 (m, 1H; CH), 3.84 (s, 3H;
CO2CH3), 3.08–3.04 (m, 1H; CHHCO2NHCH3), 2.80 (dd, 2J=15.1, 3J=
4.9 Hz, 1H; CHHCO2NHCH3), 2.66–2.64 (m, 6H, CH3; NHCH3),
(CH3)3]; 13C NMR (90 MHz, CDCl3): d=170.9, 169.5,
1.40 ppm [s, 9H; CACHTREUNG
162.9, 155.4, 145.3, 140.9, 80.2, 52.0, 50.1, 39.2, 28.4, 26.2, 13.2 ppm; IR
(KBr): n˜ =3334 (s), 2352 (m), 1704 (s), 1673 (s), 1649 (s), 1547 (w), 1519
(s), 1328 (m), 1223 (m), 1162 (m), 1055 (m), 852 (w) cmꢀ1; HRMS (EI):
m/z: calcd for C15H23N3O5S: 357.1358 [M+], found 357.1359.
1
= ꢀ15.0 (c=1.00 in CH3CN); H NMR (360 MHz, CDCl3): d=7.40–7.30
3
(m, 5H; Aryl-H), 7.20 (d, J=8.9 Hz, 1H; NH), 5.70 (brs, 1H; NH), 5.16
(d, 2J=12.3 Hz, 1H; PhCHH), 5.12 (d, 2J=12.3 Hz, 1H; PhCHH), 4.57–
4.54 (m, 2H; CHCO2CH3, CHNHBoc), 4.32–4.30 (m, 1H; CHCH3OH),
3.74 (s, 3H; CO2CH3), 3.06 (dd, 2J=17.1, 3J=4.9 Hz, 1H; CHHCO2Bn),
(S)-2-[2-Methyl-1-(tritylamino)propyl]-thiazole-4-carboxylic acid ethyl
ester (27)
2.78 (dd, 2J=17.1, 3J=5.8 Hz, 1H; CHHCO2Bn), 1.45 [s, 9H; C
ACHTRE(UNG CH3)3],
Boc deprotection: Trifluoroacetic acid (4 mL) was added at room temper-
ature to a stirred solution of thiazole 26 (1 g, 3.05 mmol) in CH2Cl2
(20 mL). After 90 min the reaction mixture was concentrated in vacuo
and then azeotroped with toluene (3”10 mL) to give the corresponding
ammonium salt as a white solid. This material was used without further
purification.
1.19 ppm (d, 3J=6.4 Hz, 3H; CHCH3OH); 13C NMR (90 MHz, CDCl3):
d=171.8, 171.3, 171.1, 155.6, 135.5, 128.7, 128.5, 128.4, 80.8, 68.3, 67.1,
57.6, 52.7, 51.0, 36.3, 28.4, 20.0 ppm; IR (KBr): n˜ =3354 (m), 2978 (w),
1738 (s), 1715 (s), 1673 (s), 1537 (s), 1519 (s), 1504 (s), 1455 (w), 1367
(m), 1164 (s), 1023 (w), 860 (w), 736 (w), 698 (w) cmꢀ1; HRMS (EI): m/z:
calcd for C19H26N2O7: 394.1740 [MꢀC2H4O+], found 394.1742.
Tritylation: The above ammonium salt was dissolved in DMF (10 mL)
and then trityl chloride (850 mg, 3.05 mmol) and triethylamine (1.10 mL,
799 mg, 7.89 mmol) were sequentially added. The reaction mixture was
stirred for 16 h at room temperature and concentrated in vacuo. The re-
sulting residue was partitioned between EtOAc(200 mL) and saturated
aq. NaHCO3 (100 mL). The organiclayer was dried (Na 2SO4) and the
solvent removed in vacuo. Purification by flash chromatography (P/Et2O
4:1 and 0.5% triethylamine) yielded the title compound 27 (1.41 g,
2.99 mmol, 98%) as a white foam. M.p. 528C; Rf =0.40 (P/Et2O 5:1);
[a]2D0 =ꢀ70.5 (c=1.00 in MeCN); 1H NMR (360 MHz, CDCl3): d=7.89
(s, 1H; H-5), 7.43–7.39 (m, 6H; Aryl-H), 7.21–7.11 (m, 9H; Aryl-H),
(S)-2-(2-Benzyloxycarbonyl-1-tert-butoxycarbonylaminoethyl)-5-methyl-
thiazole-4-carboxylic acid methyl ester (23)
Oxidation to the amidoketone: IBX (14.2 g, 50.7 mmol) was added at
room temperature to
a stirred solution of dipeptide 22 (10.3 g,
23.5 mmol) in MeCN (150 mL). The reaction mixture was heated to
reflux for 3.5 h. After cooling to room temperature, the reaction was fil-
tered filtered through Celite and washed with EtOAc(200 mL). The sol-
vents were removed in vacuo.
Thiazole synthesis: The crude amidoketone was dissolved in THF
(240 mL) and Lawessonꢁs reagent (14.3 g, 35.4 mmol) was added. The re-
action mixture was heated to reflux for 5 h and then cooled to room tem-
perature. The mixture was diluted with EtOAc(500 mL) and washed
with water (200 mL), saturated aq. NH4Cl (200 mL), saturated aq.
NaHCO3 (200 mL) and saturated brine (200 mL). The organiclayer was
dried (Na2SO4) and the solvents removed in vacuo. The crude product
was purified by flash chromatography (P/EtOAc 3:1) to give a yellow
solid. This residue was dissolved in EtOAc(400 mL) and washed with
5% aq. AgNO3 (4”100 mL) and dried (Na2SO4). The mixture was con-
centrated in vacuo and purified by flash chromatography (P/EtOAc 3:1)
to yield the desired thiazole 23 (4.59 g, 10.6 mmol, 45%) as a white foam.
3
3
4.43–4.33 (m, 2H; CH2CH3), 4.17 (dd, J=6.5, J=4.1 Hz, 1H; CH), 2.87
(d, 3J=6.5 Hz, 1H; NH), 1.71–1.67 [m, 1H; CH(CH3)2], 1.38 (t, 3J=
7.1 Hz, 3H; CH2CH3), 0.86 [d, 3J=6.8 Hz, 3H; CH
(CH3)(CH3)],
0.76 ppm [d, 3J=6.8 Hz, 3H; CH (CH3)]; 13C NMR (90 MHz,
(CH3)
AHCTREUNG
A
ACHTREUNG
A
ACHTREUNG
CDCl3): d=176.7, 161.8, 145.9, 129.2, 128.0, 127.9, 127.5, 126.7, 72.2, 61.2,
60.4, 34.7, 20.2, 16.9, 14.5 ppm; IR (KBr): n˜ =3101 (m), 2959 (m), 1732
(s), 1491 (w), 1445 (w), 1205 (s), 1111 (w), 1027 (w), 902 (w), 832 (w),
750 (w), 711 (m) cmꢀ1
; HRMS (EI): m/z: calcd for C26H23N2O2S:
427.1480 [M+], found 427.1473.
(S)-5-Methoxymethyl-2-[2-methyl-1-(tritylamino)propyl]-thiazole-4-car-
boxylic acid ethyl ester (28): Diisopropylamine (0.30 mL, 215 mg,
2.12 mmol) was added at ꢀ788C to a stirred solution of nBuLi (0.85 mL,
2.13 mmol, 2.50m in hexane) in THF (20 mL). After 10 min a precooled
(ꢀ788C) solution of thiazole 27 (912 mg, 1.94 mmol) in THF (9 mL) was
added rapidly by syringe. 10 min later iodomethoxymethane (0.50 mL,
1.02 g, 5.90 mmol) was added and the stirring was continued at ꢀ788C
for 5 min. The reaction mixture was quenched by addition of saturated
Rf =0.49 (P/EtOAc2:1);
[
a]2D0 =ꢀ29.2 (c=1.00 in MeCN); 1H NMR
(360 MHz, CDCl3): d=7.40–7.25 (m, 5H; Aryl-H), 5.96–5.93 (m, 1H;
NH), 5.31–5.29 (m, 1H; CH), 5.09 (s, 2H; PhCH2), 3.90 (s, 3H;
CO2CH3), 3.35–3.30 (m, 1H; CHHCO2Bn), 3.03 (dd, 2J=16.8, 3J=
5.4 Hz, 1H; CHHCO2Bn), 2.71 (s, 3H; CH3), 1.45 ppm [s, 9H; CACHTRE(UNG CH3)3];
13C NMR (90 MHz, CDCl3): d=171.0, 168.1, 162.9, 155.1, 145.5, 141.0,
135.5, 128.7, 128.5, 128.3, 80.6, 66.8, 55.7, 49.4, 38.6, 28.4, 13.3 ppm; IR
2334
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
aq. NH4Cl (10 mL). The mixture was diluted with EtOAc(250 mL) and
washed with saturated aq. NaHCO3 (100 mL) and saturated brine
(100 mL). The organiclayer was dried (Na 2SO4) and concentrated in
vacuo. The residue was purified by flash chromatography (P/Et2O=4:1
and 0.5% triethylamine) to yield the title compound 28 (739 mg,
CH2Cl2 (250 mL), washed with saturated aq. NaHCO3 (75 mL), dried
(Na2SO4) and concentrated in vacuo to yield the crude amine. The mate-
rial was used without further purification.
Peptide coupling: The above amine and compound 10 (328 mg,
1.29 mmol) were dissolved in DMF (18 mL). The solution was cooled to
ꢀ108C and HOBt hydrate (555 mg, 3.90 mmol) was added. This was
stirred for 20 min at ꢀ108C before EDC (295 mg, 1.54 mmol) was added.
The mixture was allowed to warm to room temperature and stirred for
16 h. The solvent was removed in vacuo and the residue dissolved in
EtOAc(350 mL) and washed with 10% citricacid (100 mL), saturated
aq. NaHCO3 (100 mL), saturated brine (50 mL), dried (Na2SO4) and con-
centrated in vacuo. Purification by flash chromatography (EtOAc) pro-
vided the tristhiazole 6b (942 mg, 1.28 mmol, 99%) as a white foam. Rf =
0.44 (EtOAc); [a]2D0 =ꢀ43.2 (c=1.00 in CHCl3); 1H NMR (500 MHz,
CDCl3): d=8.90 (d, 3J=8.9 Hz, 1H; NH), 8.52 (d, 3J=8.6 Hz, 1H;
NHCHiPr), 8.13 (s, 1H; CHF), 6.69 (brq, 3J=4.6 Hz, 1H; E: NHCH3),
1.44 mmol, 74%) as white foam. Rf =0.30 (P/Et2O 4:1); [a]D20
=
ꢀ80.5(c=1.00 in MeCN); 1H NMR (360 MHz, CDCl3): d=7.43–7.39 (m,
6H; Aryl-H), 7.21–7.11 (m, 9H; Aryl-H), 4.94 (d, 2J=14.8 Hz, 1H;
CH2OCH3), 4.87 (d, 2J=14.8 Hz, 1H; CH2OCH3), 4.46–4.30 (m, 2H;
CH2CH3), 4.12 (dd, 3J=6.7, 3J=3.9 Hz, 1H; CH), 3.46 (s, 3H;
3
CH2OCH3), 2.76 (d, J=6.7 Hz; NH), 1.60–1.49 [m, 1H; CH
A
(t, 3J=7.0 Hz, 3H; CH2CH3), 0.86 [d, 3J=6.9 Hz, 3H; CH
0.75 ppm [d, 3J=6.9 Hz, 3H; CH
(CH3)
A
ACHTREUNG
A
ACHTREUNG
CDCl3): d=174.1, 162.7, 147.9, 146.0, 139.0, 129.1, 127.9, 126.6, 72.3, 68.8,
61.1, 60.4, 59.2, 34.6, 20.1, 16.8, 14.6 ppm; IR (KBr): n˜ =2960 (m), 2932
(w), 1707 (s), 1487 (m), 1446 (w), 1320 (m), 1212 (s), 1087 (m), 1066 (w),
902 (w), 765 (w), 708 (m) cmꢀ1; HRMS (EI): m/z: calcd for C28H27N2O3S:
471.1742 [MꢀC3H7+], found 471.1745.
3
3
5.85–5.78 (m, 1H; E: CHCH2), 5.26 (dd, J=8.6, J=5.5 Hz, 1H; CHiPr),
4.94 (s, 2H; CH2OCH3), 4.43 (q, 3J=7.0 Hz, 2H; CH2CH3), 3.51 (s, 3H;
CH2OCH3), 3.20 (dd, 2J=14.7, 3J=4.4 Hz, 1H; CHHCONHCH3), 2.99
(dd, 2J=14.7, 3J=6.2 Hz, 1H; CHHCONHCH3), 2.74 (s, 3H; CH3), 2.66
Bisthiazole 33
Saponification of methyl ester 24: Aq. lithium hydroxide solution (1m;
5.00 mL, 5.00 mmol) was added to a solution of methyl ester 24 (515 mg,
1.44 mmol) in methanol (25 mL) at 08C. The reaction mixture was stirred
at room temperature for 16 h, concentrated in vacuo and the crude solid
(d, 3J=4.6 Hz, 3H; NHCH3), 2.47–2.37 [m, 1H; CH
7.0 Hz, 3H; CH2CH3), 1.01 [d, 3J=6.8 Hz, 3H; CH
0.98 ppm [d, 3J=6.8 Hz, 3H; CH (CH3)]; 13C NMR (90 MHz,
(CH3)
ACHTREUNG
A
ACHTREUNG
A
ACHTREUNG
CDCl3): d=170.9, 168.5, 166.1, 162.2, 162.2, 159.3, 151.7, 148.3, 142.3,
141.9, 139.4, 130.4, 100.7, 68.8, 61.7, 59.4, 56.3, 48.6, 39.1, 34.2, 26.4, 19.2,
18.1, 14.5, 12.6 ppm; IR (KBr): n˜ =3372 (m), 2963 (w), 2929 (w), 1711
(m), 1667 (s), 1653 (s), 1541 (s), 1488 (m), 1404 (w), 1370 (w), 1331 (w),
1211 (m), 1096 (w), 981 (w), 732 (w) cmꢀ1; HRMS (EI): m/z: calcd for
C25H31O6N6IS3: 734.0506 [M+], found 734.0500.
was partitioned between EtOAc(20 mL) and H O (75 mL). The aqueous
2
layer was acidified with aq. HCl (2n) to pH ꢁ3 and extracted with
EtOAc(3”75 mL). The combined organicextracts were dried (Na
2SO4)
and concentrated in vacuo to yield the carboxylic acid. The material was
used without further purification.
Trityl deprotection: To
a stirred solution of thiazole 28 (739 mg,
2,6-Dibromopyridine 34
1.44 mmol) in CH2Cl2 (22.5 mL) at room temperature was added tri-
fluoroacetic acid (4.5 mL). After 20 min the reaction mixture was diluted
with CHCl3 (20 mL) and concentrated in vacuo. The residue was dis-
solved in CHCl3 (150 mL) and washed with saturated aq. NaHCO3
(50 mL). The aqueous layer was extracted with CHCl3 (2”30 mL). The
combined organic layers were dried (Na2SO4) and concentrated in vacuo
to yield the crude amine 29. This material was used without further pu-
rification.
Zincation: DMA (1.9 mL) and 1,2-dibromoethane (33.0 mL, 72.0 mg,
383 mmol L) were added to a flame dried flask charged with zinc dust
(227 mg, 3.47 mmol). The zincsuspension was shortly heated with a heat
gun until evolution of ethylene occurred and then allowed to reach 258C.
This procedure was repeated three times. TMSCl (112 mL, 96.0 mg,
0.88 mmol) was added neat and the reaction mixture was stirred for
5 min. 2,6-Dibromo-3-iodopyridine (18) (596 mg, 1.64 mmol) dissolved in
5 mL THF was added. The stirring was continued for 30 min at 258C,
then the zincdust was allowed to settle (30 min).
Peptide coupling: The above carboxylic acid and amine 29 were dissolved
in DMF (20 mL). The solution was cooled to ꢀ108C and HOBt hydrate
(621 mg, 4.30 mmol) was added. This was stirred for 20 min at ꢀ108C
before EDC (330 mg, 1.70 mmol) was added. The mixture was allowed to
warm to room temperature and stirred for 16 h. The solvent was removed
in vacuo and the residue dissolved in EtOAc (350 mL) and washed with
10% citric acid (100 mL), saturated aq. NaHCO3 (100 mL), saturated
brine (50 mL), dried (Na2SO4) and concentrated in vacuo. Purification by
flash chromatography (EtOAc) provided the bisthiazole 33 (788 mg,
1.32 mmol, 92%) as a white foam. Rf =0.54 (EtAOc); [a]2D0 =ꢀ57.1 (c=
Negishi cross-coupling: The supernatant liquid containing the zincated di-
bromo pyridine was added to a solution of trithiazole 6b (424 mg,
0.58 mmol), [PdACHTRE(UNG dba)2] (21.0 mg, 36.5 mmol, 6 mol%), TFP (17.0 mg,
73.0 mmol, 12 mol%) in 4.2 mL THF. The reaction mixture was stirred
for 12 h at 458C. After quenching with saturated aq. NH4Cl (50 mL), the
aqueous layer was extracted with EtOAc (3”50 mL). The combined or-
ganiclayers were dried (Na 2SO4) and concentrated in vacuo. The crude
product was purified by flash chromatography (EtOAc) to yield the de-
sired product 34 (426 mg, 0.51 mmol, 87%) as a pale yellow solid. M.p.
86–898C; Rf =0.39 (EtOAc); [a]2D0 =ꢀ30.6 (c=1.00 in CHCl3); 1H NMR
1
3
1.00 in CHCl3), H NMR (360 MHz, CDCl3): d=8.37 (d, J=8.6 Hz, 1H;
NH), 6.65 (d, 3J=8.5 Hz, 1H; NHBoc), 6.46 (brs, 1H; NHCH3), 5.27–
5.21 (m, 2H; CHiPr, CHNHBoc), 4.92 (s, 2H; CH2OCH3), 4.40 (q, 3J=
7.1 Hz, 2H; CH2CH3), 3.49 (s, 3H; CH2OCH3), 3.11 (dd, 2J=14.6, 3J=
3.1 Hz, 1H; CHHCONHCH3), 2.81 (dd, 2J=14.6, 3J=4.6 Hz, 1H;
CHHCONHCH3), 2.74 (s, 3H; CH3), 2.58 (d, 3J=4.4 Hz, 3H; NHCH3),
3
3
(500 MHz, CDCl3): d=9.42 (d, J=8.8 Hz, 1H; NH), 8.50 (d, J=8.2 Hz,
2H; NHCHiPr, CHpy), 8.36 (s, 1H; CHF), 7.59 (d, J=8.2 Hz, 1H; CHpy),
3
6.66 (d, 3J=4.5 Hz, 1H; NHCH3), 5.83–5.79 (m, 1H; CHCH2), 5.26 (dd,
3J=8.2, 3J=5.4 Hz, 1H; CHiPr), 4.94 (s, 2H; CH2OCH3), 4.42 (q, 3J=
7.0 Hz, 2H; CH2CH3), 3.51 (s, 3H; CH2OCH3), 3.30 (dd, 2J=14.8, 3J=
4.2 Hz, 1H; CHHCONHCH3), 2.94 (dd, 2J=14.8, 3J=5.2 Hz, 1H;
2.43–2.33 [m, 1H; CH
3H; CH2CH3), 0.98 [d, 3J=6.8 Hz, 3H; CH
3J=6.8 Hz, 3H; CH (CH3)]; 13C NMR (90 MHz, CDCl3): d=171.2,
(CH3)
ACHTREUNG(CH3)2], 1.46 [s, 9H; CACHTRENUG
A
ACHTREUNG
A
N
CHHCONHCH3), 2.73 (s, 3H; CH3), 2.62 (d, 3J=4.5 Hz, 3H; NHCH3),
3
168.7, 168.3, 162.3, 162.2, 155.5, 148.3, 142.0, 142.0, 139.4, 80.2, 68.8, 61.7,
59.4, 56.2, 50.1, 39.2, 34.2, 28.5, 26.2, 19.3, 18.1, 14.5, 12.7 ppm; IR (KBr):
n˜ =3353 (s), 2976 (m), 2933 (m), 1715 (s), 1681 (s), 1651 (s), 1556 (m),
1504 (m), 1371 (m), 1334 (w), 1210 (w), 1164 (m), 1103 (w), 1026 (w), 863
(w), 736 (m) cmꢀ1; HRMS (ESI): m/z: calcd for C26H40O7N5S2: 598.2364
[M+H+], found 598.2362.
2.44–2.39 [m, 1H; CH
3J=6.8 Hz, 3H; CH
(CH3)
(CH3)
(CH3)]; 13C NMR (90 MHz, CDCl3): d=171.3, 168.6, 166.5, 162.2,
(CH3)2], 1.42 (t, J=7.0 Hz, 3H; CH2CH3), 1.00 [d,
G
(CH3)], 0.98 ppm [d, 3J=6.8 Hz, 3H; CH-
ACHTREUNG
A
ACHTREUNG
162.2, 162.0, 160.6, 149.8, 148.4, 142.3, 142.1, 141.4, 141.3, 139.2, 139.0,
130.3, 127.8, 126.3, 68.8, 61.8, 59.4, 56.2, 48.7, 38.9, 34.3, 26.3, 19.2, 18.1,
14.5, 12.7 ppm; IR (KBr): n˜ =3366 (m), 2969 (m), 2929 (w), 1711 (m),
1667 (s), 1535 (s), 1485 (m), 1406 (w), 1370 (w), 1331 (m), 1267 (w), 1208
(m), 1110 (m), 1071 (w), 735 (m) cmꢀ1; HRMS (EI): m/z: calcd for
C30H32O6N7Br2S3: 841.0021 [M+], found 840.9987.
Tristhiazole 6b
Boc deprotection of bisthiazole 33: Acetyl chloride (0.90 mL, 1.00 g,
12.7 mmol) was added dropwise to EtOH (9.00 mL) at 08C. Next, bis-
thiazole 33 (768 mg, 1.28 mmol) was added and the reaction mixture was
allowed to warm to room temperature and the stirring was continued for
16 h. The solvent was removed in vacuo and the residue was dissolved in
2-Bromopyridine 50
Zincation of tert-butyl ester 13: DMA (1 mL) and 1,2-dibromoethane
(10.0 mL, 22.0 mg, 116 mmol) were added to a flame dried flask charged
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2335

T. Bach et al.
with zincdust (66.0 mg, 1.00 mmol). The zincsuspension was shortly
heated with a heat gun until evolution of ethylene occurred and then al-
lowed to reach 258C. This procedure was repeated three times. Trime-
thylsilyl chloride (TMSCl) (27.0 mL, 23.0 mg, 213 mmol) was added neat
and the reaction mixture was stirred for 5 min. tert-Butyl ester 13
(85.0 mg, 0.32 mmol) dissolved in 0.5 mL DMA was added. The stirring
was continued for 30 min at 258C, then the zincdust was allowed to
settle (30 min).
EtOAc7:3) gave white solid that was dissolved in toluene (6 mL). After
the addition of 2,4-dibromothiazole (53) (180 mg, 0.74 mmol) and [Pd-
AHCTRE(UNG PPh3)4] (52.0 mg, 45 mmol) the resulting mixture was stirred for 10 h at
908C. Saturated aq. NaHCO3 solution (20 mL) was added, the biphasic
mixture was extracted with EtOAc (3”20 mL), the combined organic ex-
tracts were dried (MgSO4) and concentrated in vacuo. Purification by
flash chromatography (P/EtOAc 7:3) yielded bithiazole 54 (248 mg,
0.32 mmol, 80%) as a glassy solid. Rf =0.32 (P/EtOAc7:3); [ a]2D0 =ꢀ3.90
(c=1.00 in CHCl3); 1H NMR (360 MHz, CDCl3): d=7.87 (s, 1H), 7.78
(d. 3J=7.5 Hz, 2H), 7.58 (brs, 2H), 7.44–7.38 (m, 2H), 7.35–7.23 (m,
7H), 7.14 (s, 1H), 7.00 (d, 3J=8.4 Hz, 1H), 5.54 (dd, 3J=8.4 Hz, 5.0 Hz,
1H), 5.34 (brs, 1H), 5.22 (d, 3J=5.0 Hz, 1H), 4.48–4.37 (m, 1H), 4.23
(brs, 1H), 4.00–3.91 (m, 1H), 3.84 (dd, 3J=16.6 Hz, 5.5 Hz, 1H), 0.87 (s,
9H), 0.00 (s, 3H), ꢀ0.17 ppm (s, 3H); 13C NMR (90 MHz, CDCl3): d=
168.1, 167.1, 163.2, 156.4, 147.4, 143.6, 141.3, 139.7, 128.2, 127.8, 127.1,
126.6, 125.9, 125.0, 120.0, 117.4, 117.0, 76.1, 67.4, 60.3, 57.6, 47.0, 44.6,
25.7, 18.0, ꢀ4.8, ꢀ5.4 ppm; IR (neat): n=3308 (s), 2926 (s), 2854 (s),
1715 (s), 1674, 1506 (s), 1450 (s), 1250 (s), 1099 (s), 1071 (m), 838 (m),
778 (m) cmꢀ1; MS (LC/ESI): m/z (%): 777/775 (40/30) [M+H+], 506
(92), 492 (100).
Negishi cross-coupling: The supernatant liquid containing the zinc organ-
yl 49 was transferred to a flask containing [PdCl2ACHTRE(UNG PPh3)2] (8.30 mg,
12.0 mmol, 30 mol%). To this mixture a solution of pyridine 34 (33.7 mg,
40.0 mmol) in DMA (0.5 mL) was added dropwise over 10 min at 458C.
The stirring was continued at 458C for 3.5 h. The reaction mixture was
partitioned between saturated aq. NH4Cl (10 mL) and EtOAc(3”
25 mL). The combined extracts were dried (Na2SO4) and concentrated in
vacuo. Purification by flash chromatography (EtOAc) afforded a yellow
foam (22 mg) containing the desired pyridine 50 (18.2 mg, 19.2 mmol,
48%, colorless film) together with triphenylphosphine oxide (3.80 mg).
Separation of the oxidated ligand from compound 50 was performed by
reverse phase HPLC (RP, ODS-A, MeCN/H2O 70:30, 15 mLminꢀ1). Rf =
0.47 (EtOAc); [a]2D0 =ꢀ23.5 (c=1.00 in CHCl3); 1H NMR (360 MHz,
CDCl3): d=9.34 (d, 3J=8.9 Hz, 1H; NH), 8.73 (d, 3J=8.2 Hz, 1H;
CHpy), 8.51 (d, 3J=8.6 Hz, 1H; NHCHiPr), 8.39 (d, 3J=8.2 Hz, 1H;
CHpy), 8.37 (s, 1H; CHF), 8.20 (s, 1H; CHA), 6.69 (brq, 3J=4.8 Hz, 1H;
NHCH3), 5.88–5.82 (m, 1H; CHCH2), 5.27 (dd, 3J=8.6, 3J=5.3 Hz, 1H;
CHiPr), 4.94 (s, 2H; CH2OCH3), 4.43 (q, 3J=7.0 Hz, 2H; CH2CH3), 3.51
(s, 3H; CH2OCH3), 3.27 (dd, 2J=14.8, 3J=4.2 Hz, 1H; CHHCONHCH3),
3.01 (dd, 2J=14.8, 3J=5.8 Hz, 1H; CHHCONHCH3), 2.74 (s, 3H; CH3),
AHCTRE{UGN (S,S)-[2-(tert-Butyldimethylsilanyloxy)-2-phenyl-1-(4-trimethylstannanyl-
[2,4’]bithiazolyl-2’-yl)-ethylcarbamoyl]-methyl}-carbamic acid 9H-fluo-
ren-9-ylmethyl ester (5a): To a stirred solution of bithiazole 54 (119 mg,
153 mmol) in degassed dioxane (4.5 mL) was added hexamethylditin
(157 mL, 0.76 mmol) followed by [PdCl2ACHTRE(UNG PPh3)2] (23.4 mg, 20 mol%). The
resulting dark brown solution was stirred for 6 h at 908C, allowed to
reach room temperature and concentrated under reduced pressure. Pu-
rification by flash chromatography (P/EtOAc 8:2!1:1) afforded bisthia-
zole 5a (116 mg, 135 mmol, 88%) as a glassy solid. Rf =0.34 (P/EtOAc
7:3); 1H NMR (360 MHz, CDCl3): d=7.87 (s, 1H; C: H-5), 7.78 (d, 3J=
7.5 Hz, 2H), 7.58 (brs, 2H), 7.44–7.38 (m, 2H), 7.35–7.23 (m, 7H), 7.14
(s, 1H), 7.00 (d, 3J=8.4 Hz, 1H), 5.54 (dd, 3J=8.4, 3J=5.0 Hz, 1H; CH-
Gly), 5.34 (brs, 1H), 5.22 (d, 3J=5.0 Hz, 1H; CHOTBS), 4.48–4.37 (m,
1H), 4.23 (brs, 1H), 4.00–3.91 (m, 1H), 3.84 (dd, 3J=16.6, 3J=5.5 Hz,
2.66 (d, 3J=4.8 Hz, 3H; NHCH3), 2.45–2.35 [m, 1H; CH
9H; C
(CH3)3], 1.42 (t, 3J=7.0 Hz, 3H; CH2CH3), 1.01 [d, 3J=7.3 Hz,
3H; CH(CH3) (CH3)(CH3)];
(CH3)], 0.99 ppm [d, 3J=7.3 Hz, 3H; CH
ACHTRE(UGN CH3)2], 1.63 [s,
ACHTREUNG
A
G
A
ACHTREUNG
13C NMR (90 MHz, CDCl3): d=171.2, 168.5, 166.8, 166.4, 162.4, 162.3,
162.2, 160.6, 160.4, 151.4, 150.2, 149.8, 148.5, 142.4, 142.0, 140.8, 139.6,
139.4, 131.7, 129.9, 126.4, 119.4, 82.6, 68.9, 61.8, 59.5, 56.2, 48.6, 38.9, 34.4,
28.3, 26.4, 19.2, 18.1, 14.6, 12.7 ppm; IR (KBr): n˜ =3389 (s), 2965 (w),
2924 (w), 1713 (m), 1666 (s), 1537 (m), 1482 (m), 1369 (w), 1337 (m),
1267 (w), 1254 (w), 1158 (w), 1100 (w), 1020 (w) cmꢀ1; HRMS (ESI):
m/z: calcd for C38H44O8N8BrS4: 947.1343 [M+H+], found 947.1337.
1H), 0.84 [s, 9H; OSi
G
G
ACHTRE(UGN CH3)3], 0.01 [s,
3H; Si(CH3)
A
G
A
ACHTREUNG
Carboxylic acid 57: Aq. lithium hydroxide solution (1m; 1 mL,
1.00 mmol) was added to a solution of the ethyl ester 50 (92.0 mg,
97.9 mmol) in tBuOH (2 mL) and THF (1 mL) at room temperature. The
reaction mixture was stirred at room temperature for 1 h, concentrated in
vacuo and the crude solid was dissolved in H2O (10 mL). The aqueous
layer was acidified with aq. HCl (2n) to pH ꢁ3 and extracted with
CHCl3 (2”25 mL) and EtOAc(2”25 mL). The ocmbined organicex-
tracts were dried (Na2SO4) and concentrated in vacuo to yield the car-
boxylicacid 57 (92.0 mg, 97.9 mmol, 100%.) as a colorless solid. [a]2D0 =+
30.3 (c=0.50 in CH3OH); 1H NMR (360 MHz, CD3OD, 378C): d=8.72
ACHTREUNG{(S,S)-[1-(4-Bromothiazol-2-yl)-2-(tert-butyldimethylsilanyloxy)-2-phenyl-
ethylcarbamoyl]-methyl}-carbamic acid 9H-fluoren-9-ylmethyl ester (52):
PyBrOP (472 mg, 1.02 mmol) and diisopropylamine (442 mL, 2.55 mmol)
was added at 08C to a stirred solution of amine 51 (350 mg, 0.85 mmol)
and Fmoc-Gly-OH (303 mg, 1.02 mmol) in CH2Cl2 (15 mL). After 4 h at
08C, the reaction mixture was quenched by the addition of saturated aq.
NaHCO3 solution (20 mL). The biphasicmixture was extratced with
CH2Cl2 (3”20 mL), the combined organic extracts were dried (MgSO4)
and concentrated in vacuo. Purification by flash chromatography (P/
EtOAc7:3) yielded 52 (583 mg, 0.84 mmol, 99%) as a glassy solid. Rf =
3
(d, J=8.2 Hz, 1H; CHpy), 8.42 (s, 1H; CHF), 8.35 (s. 1H; CHA), 8.22 (d,
3J=8.2 Hz, 1H; CHpy), 5.82 (virt. t, 3J=5.8 Hz, 1H; CHCH2), 5.13 (d,
3J=6.7 Hz, 1H; CHiPr), 4.93 (s, 2H; CH2OCH3), 3.44 (s, 3H;
CH2OCH3), 3.22–3.09 (m, 2H; CH2CONHCH3), 2.71–2.69 (m, 6H;
0.25 (P/EtOAc7:3);
[
a]2D0 =+0.70 (c=0.90 in CHCl3); 1H NMR
3
(360 MHz, CDCl3): d=7.77 (d, J=7.5 Hz, 2H), 7.63–7.56 (m, 2H), 7.44–
7.37 (m, 2H), 7.35–7.21 (m, 7H), 7.10 (s, 1H), 6.90 (d, 3J=8.6 Hz, 1H),
5.49 (dd, 3J=8.8 Hz, 6.4 Hz, 1H), 5.43 (brs, 1H), 5.12 (d, 3J=6.4 Hz,
NHCH3, CH3), 2.48–2.34 [m, 1H; CH
[d, 3J=6.7 Hz, 3H; CH (CH3)], 0.97 ppm [d, 3J=6.7 Hz, 3H; CH-
(CH3)
(CH3)
(CH3)]; 13C NMR (90 MHz, CD3OD, 378C): d=173.1, 170.5, 168.1,
ACHRTUG(NE CH3)2], 1.62 [s, 9H; CCAHRTE(UGN CH3)3], 1.01
A
ACHTREUNG
3
1H), 4.40 (d, J=7.5 Hz, 2H), 4.26–4.17 (m, 1H), 3.88–3.78 (m, 1H), 3.73
A
ACHTREUNG
(dd, 3J=17.0 Hz, 5.5 Hz, 1H), 0.81 (s, 9H), ꢀ0.09 (s, 3H), ꢀ0.21 ppm (s,
3H); 13C NMR (90 MHz, CDCl3): d=168.4, 168.1, 156.3, 143.7, 141.2,
139.8, 128.2, 127.7, 127.1, 127.0, 126.7, 125.0, 124.5, 119.7, 117.0, 76.5,
67.2, 60.3, 57.3, 47.0, 44.3, 25.5, 17.9, ꢀ4.9, ꢀ5.5 ppm; IR (neat): n˜ =3308
(m), 3064 (w), 2953 (s), 2928 (m), 2856 (m), 1737 (s), 1712 (s), 1258 (m),
1100 (m), 838 (m) cmꢀ1; HRMS (ESI): m/z: calcd for C34H38N3O4SSiBr:
692.1608 [M+H+], found 692.1607; MS (ESI): m/z (%): 694/692 (100/90)
[M+H+], 415 (42).
168.0, 164.5, 164.0, 163.7, 162.5, 161.8, 152.4, 150.9, 150.5, 149.2, 143.9,
142.8, 142.1, 141.6, 140.5, 132.8, 131.7, 128.0, 120.0, 83.6, 69.4, 59.4, 57.6,
50.0, 39.9, 34.5, 28.5, 26.5, 19.9, 18.7, 12.7 ppm; IR (KBr): n˜ =3324 (s),
3109 (m), 2967 (m), 2923 (m), 1717 (s), 1673 (s), 1651 (s), 1536 (s), 1489
(m), 1418 (w), 1347 (w), 1253 (m), 1229 (m), 1160 (s), 1097 (m), 1017
(m), 993 (w), 935 (w) cmꢀ1
;
HRMS (ESI): m/z: calcd for
C36H38O8N8BrS4: 917.0873 [MꢀH+], found 917.0869.
Organostannane 58: Carboxylicaicd
57 (21.0 mg, 23.0 mmol) and the
A
amine 5b (16.0 mg, 25.0 mmol) were dissolved in DMF (2 mL) and
cooled to 08C. To this mixture diisopropylethylamine (14.5 mL, 11 mg,
85 mmol) and DPPA (9.00 mL, 11.5 mg, 41.8 mmol) were consecutively
added. The reaction mixture was allowed to reach room temperature
over 16 h and then partitioned between CH2Cl2 (10 mL) and saturated
aq. NH4Cl (15 mL). The aqueous layer was extracted with CH2Cl2 (2”
25 mL), the combined organic extracts were dried (Na2SO4) and concen-
oxy)-2-phenylethylcarbamoyl]-methyl}-carbamic acid 9H-fluoren-9-yl-
methyl ester (54): Hexamethylditin (336 mL, 1.60 mmol) was added to a
stirred solution of thiazole 52 (280 mg, 0.40 mmol) and [PdACHTRE(UNG PPh3)4]
(46.2 mg, 0.04 mmol, 10 mol%) in toluene (5 mL). The reaction mixture
was stirred for 2 h at 858C, allowed to reach room temperature and then
concentrated to ca. 1 mL. Purification by flash chromatography (P/
2336
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
trated in vacuo. Purification by flash chromatography (P/EtOAc 1:3)
(35.0 mg, 0.12 mmol, 63%) as a colorless foam. Rf =0.21 (EtOAc/MeOH
1
gave the bromopyridine 58 (30.8 mg, 20.0 mmol, 87%) as a colorless oil.
9:1); [a]2D0 =ꢀ40.4 (c=1.00 in CHCl3); H NMR (360 MHz, CD3OD): d=
a]2D0 =ꢀ16.4 (c=1.00 in CHCl3); 1H NMR
4.56–4.46 (m, 2H; CHCH2O, CHCONH2), 3.85–3.67 (m, 4H; CH2OH,
NCH2), 2.28–2.18 (m, 1H; CH2CHH), 2.05–1.96 (m, 3H; Pro: CH2CHH),
Rf =0.35 (P/EtOAc1:3);
[
3
3
(360 MHz, CDCl3): d=9.61 (d, J=8.8 Hz, 1H; NH), 8.75 (d, J=8.1 Hz,
1H; CHpy), 8.40 (d, 3J=8.1 Hz, 1H; CHpy), 8.34 (s, 1H; CHF), 8.21 (s,
1H; CHA), 8.11 (d, 3J=9.2 Hz, 1H; NHCHiPr), 7.95 (t, 3J=5.6 Hz, 1H;
NH), 7.86 (s, 1H; CHC), 7.36 (s, 1H; CHB), 7.24–7.16 (m, 5H; CHph),
7.12 (d, 3J=8.5 Hz, 1H; NH), 6.59 (brq, 3J=4.6 Hz, 1H; NHCH3), 5.80–
5.73 (m, 1H; CHCH2), 5.48 (dd, 3J=8.5, 3J=5.4 Hz, 1H; CHCHOTBS),
5.28 (dd, 3J=9.2, 3J=5.6 Hz, 1H; CHiPr), 5.19 (d, 3J=5.4 Hz, 1H;
CHOTBS), 5.06 (s, 2H; CH2OCH3), 4.22 (dd, 2J=16.6, 3J=6.4 Hz, 1H;
CHH), 4.00 (dd, 2J=16.6, 3J=4.8 Hz, 1H; CHH), 3.48 (s, 3H;
CH2OCH3), 3.18 (dd, 2J=15.7, 3J=4.4 Hz, 1H; CHHCONHCH3), 2.94
(dd, 2J=15.7, 3J=4.7 Hz, 1H; CHHCONHCH3), 2.75 (s, 3H; CH3), 2.65
1.44 ppm [s, 9H;
(CH3)3]; 13C NMR (90 MHz, CD3OD): d=177.1,
CACHTREUNG
172.6, 157.8, 80.8, 63.4, 61.6, 55.4, 47.0, 30.7, 28.7, 25.7 ppm; IR (neat):
n˜ =3333 (s), 2979 (s), 1651 (s), 1519 (m), 1454 (m), 1367 (m), 1266 (m),
1167 (m), 1062 (w), 736 (m) cmꢀ1
; HRMS (EI): m/z: calcd for
C13H23N3O5: 301.1637 [M+], found 301.1626.
Hydroxyamide 61
Boc deprotection of tert-butyl ester 56: Trifluoroacetic acid (0.4 mL) was
added at room temperature to a stirred solution of the macrocyle 56
(13.5 mg, 10.4 mmol) in CH2Cl2 (2 mL). After 3 h the reaction mixture
was concentrated in vacuo and then azetroped with toluene (3”5 mL) to
give the corresponding ammonium salt as a colorless solid. This material
was used in the subsequent peptide coupling without further purification.
(d, 3J=4.6 Hz, 3H; NHCH3), 2.37–2.26 [m, 1H; CH
(CH3)
(CH3)3], 0.96 [d, 3J=6.9 Hz, 3H; CH (CH3)], 0.96 [d, 3J=6.9 Hz,
3H; CH(CH3)(CH3)], 0.80 [s, 9H; Si(CH3)2C(CH3)3], 0.37 [s, 9H; Sn-
(CH3)3], ꢀ0.05 [s, 3H; Si(CH3)(CH3)], ꢀ0.22 ppm [s, 3H; Si(CH3)-
ACHTREUNG
(CH3)],; 13C NMR (90 MHz, CDCl3): d=171.1, 168.6, 168.3, 167.0, 166.8,
ACHTRE(UNG CH3)2], 1.64 [s, 9H;
C
N
G
ACHTREUNG
A
G
A
ACHTREUNG
Deprotection of dipeptide 59: Trifluoroacetic acid (0.2 mL) was added at
room temperature to a stirred solution of the dipeptide 59 (17.0 mg,
56.4 mmol) in CH2Cl2 (2 mL). After 2 h the reaction mixture was concen-
trated in vacuo and then azetroped with toluene (3”5 mL) to give the
corresponding ammonium salt 60 as a colorless solid. This material was
used in the subsequent peptide coupling without further purification.
A
G
A
ACHTREUNG
166.7, 163.3, 162.5, 162.4, 162.2, 161.3, 160.8, 160.4, 151.4, 150.3, 149.9,
149.4, 144.7, 142.2, 142.2, 141.1, 140.8, 139.9, 139.6, 131.7, 129.9, 128.4,
128.3, 126.7, 126.5, 126.3, 119.5, 116.1, 82.5, 76.4, 68.3, 59.3, 58.0, 55.6,
48.7, 43.0, 38.5, 34.3, 28.3, 26.3, 25.8, 19.3, 18.2, 18.0, 12.8, ꢀ4.7, ꢀ5.2,
ꢀ8.7 ppm; IR (KBr): n˜ =3382 (s), 2960 (w), 2926 (m), 2853 (w), 1666 (s),
1651 (s), 1538 (m), 1504 (m), 1368 (w), 1252 (w), 1159 (w), 1101 (w),
1071 (w), 1017 (w), 838 (w), 777 (w) cmꢀ1; HRMS (ESI): m/z: calcd for
C61H76O9N12BrS6SiSn: 1539.2153 [M+H+], found 1539.2104.
Peptide coupling: To a mixture of TOTU (5.90 mg, 18.0 mmol) and the
ammonium salt 60 was added a solution of the above carboxylic acid in
DMF (1.2 mL) and immediately after diisopropylethylamine (20.0 mL,
15.0 mg, 117 mmol). After 2 h the mixture was warmed to room tempera-
ture and the stirring was continued for 3 h. The mixture was partitioned
between pH 5.5 buffer (3 mL) and CHCl3 (5”10 mL). The combined or-
ganicextracts were dried (Na 2SO4) and concentrated in vacuo. Purifica-
tion by flash chromatography (CH2Cl2/MeOH 92:8) yielded the b-hy-
droxy amide 61 (9.60 mg, 6.80 mmol, 65%) as a colorless oil. The desired
compound 61 was accompanied by an impurity derived from the coupling
reagent. Further purification of the alcohol 61 by reverse phase HPLC
was not successful. Rf =0.12 (CH2Cl2/MeOH 9:1); [a]2D0 =+122.0 (c=0.25
in MeOH); 1H NMR (600 MHz, CD3OD): d=9.06 (d, 3J=8.5 Hz, 1H;
NH), 8.98 (d, 3J=8.1 Hz, 1H; NH), 8.91 (d, 3J=7.8 Hz, 1H; NH), 8.80
Macrocycle 56
Stille macrocylisation: To a stirred solution of bromopyridine 58 (24.0 mg,
15.6 mmol) in degassed toluene (15.5 mL) was added [PdACHTRE(UNG PPh3)4]
(4.00 mg, 3.50 mmol, 22 mol%) and the resulting solution was heated to
858C for 35 h. The reaction mixture was concentrated in vacuo to 0.5 mL
and purified by flash chromatography (CH2Cl2/MeOH 98:2) to yield the
macrolide 56 (15.2 mg, 11.7 mmol, 75%) as a pale yellow solid together
with triphenylphoshine oxide. Rf =0.34 (P/EtOAc1:3); [ a]2D0 =+46.6 (c=
1
3
0.55 in CHCl3); H NMR (360 MHz, CDCl3): d=8.89 (d, J=9.2 Hz, 1H;
NHCHCH2), 8.73 (d, 3J=7.8 Hz, 1H; NHCHiPr), 8.37 (d, 3J=8.2 Hz,
1H; CHpy), 8.35 (s, 1H; CHF), 8.19 (s, 1H; CHA), 8.14 (s, 1H; CHB), 8.10
(d, 3J=8.1 Hz, 1H; CHpy), 7.76 (dd, J=9.4 Hz, J=3.4 Hz, 1H; CH2NH),
7.23 (s, 1H; CHC), 7.18–7.14 (m, 3H; CHph), 7.04–6.98 (m, 2H; CHph),
6.74–6.68 (brq, 3J=4.8 Hz, 1H; NHCH3), 6.67 (d, 3J=7.0 Hz, 1H; NH),
5.42 (dd, 3J=7.0, 3J=4.1 Hz, 1H; CHCHOTBS), 5.39–5.33 (m, 1H;
CHCH2), 5.18 (dd, 3J=7.8, 3J=4.4 Hz, 1H; CHiPr), 5.14–5.05 (m, 3H;
CH2OCH3, CHOTBS), 4.87 (dd, 2J=17.2, 3J=9.4 Hz, 1H; CHHNH),
3.79 (dd, 2J=17.2, 3J=3.4 Hz, 1H; CHHNH), 3.50 (s, 3H; CH2OCH3),
2.64 (s, 3H; CH3), 2.64–2.60 (m, 1H; CHHCONHCH3), 2.59 (d, 3J=
3
3
(brq, J=4.2 Hz, 1H; NHCH3), 8.71 (d, J=7.6 Hz, 1H; NHCHiPr), 8.50
(s, 1H; CHF), 8.41 (s, 1H; CHA), 8.40 (d, J=8.1 Hz, 1H; CHpy), 8.33 (d,
3
3J=8.1 Hz, 1H; CHpy), 8.25 (s, 1H; CHB), 7.64 (brs, 1H; NH), 7.50 (d,
3J=7.3 Hz, 2H; CHph), 7.47 (s, 1H; CHC), 7.42 (virt. t, 3J=7.3 Hz, 2H;
CHph), 7.38 (d, 3J=7.1 Hz, 1H; CHph), 5.48–5.44 (m, 1H; CHCH2OH),
5.32–5.30 (m, 1H; CHCH2), 5.17 (dd, 3J=7.6, 3J=4.8 Hz, 1H; CHiPr),
4.99–4.92 (m, 3H; CHHOH, CH2OCH3), 4.59–4.54 (m, 1H; CHHOH),
4.39–4.30 (m, 1H; CHHNH), 4.07–4.04 (m, 1H; CHCONH2), 3.99–3.89
(m, 2H; NCH2), 3.76 (dd, 2J=16.8, 3J=3.7 Hz, 1H; CHHNH), 3.46 (s,
3H; CH2OCH3), 2.76–2.72 (m, 1H; CHHCONHCH3), 2.65 (s, 3H; CH3),
4.8 Hz, 3H; NHCH3), 2.32–2.18 [m, 1H; CH
(CH3)3], 1.02–0.83 [m, 16H; CH(CH3)2,
CHHCONHCH3], 0.11 [s, 3H; Si(CH3)
(CH3)], ꢀ0.17 ppm [s, 3H; Si-
(CH3)
(CH3)]; 13C NMR (90 MHz, CDCl3): d=169.5, 168.7, 168.6, 168.0,
N
2.62 (d, 3J=4.2 Hz, 3H; NHCH3), 2.36–2.26 [m, 1H; CH
2.18 (m, 1H; CH2CHH), 2.12–1.99 (m, 3H; CH2CHH), 1.37–1.34 (m,
ACHTRE(UGN CH3)2)], 2.26–
A
G
A
ACHTREUNG
3
1H; CHHCONHCH3), 0.98 [d, J=6.8 Hz, 3H; CH
3J=6.7 Hz, 3H; CH
(CH3)(CH3)], 0.64 [s, 9H; Si(CH3)2C
[s, 3H; Si(CH3) (CH3)
(CH3)], ꢀ0.37 ppm [s, 3H; Si
(CH3)]; 13C NMR
ACHRTUNEG(CH3)ACHTRENUG
A
ACHTREUNG
A
G
E
ACHTREUNG
A
ACHTREUNG
G
G
G
ACHTREUNG
167.9, 166.0, 165.1, 162.6, 162.1, 161.2, 160.6, 160.1, 154.4, 150.9, 150.4,
150.1, 148.3, 145.1, 142.4, 140.6, 140.4, 140.3, 138.5, 132.2, 129.5, 128.7,
128.4, 127.7, 126.3, 125.2, 123.2, 118.8, 115.4, 82.4, 76.3, 68.2, 59.3, 56.1,
53.5, 48.2, 41.3, 38.2, 34.6, 28.4, 26.4, 25.9, 18.5, 18.4, 18.1, 12.4, ꢀ4.3,
ꢀ5.2 ppm; IR (film): n˜ =3350 (m), 2929 (s), 2855 (m), 1727 (s), 1666 (s),
1555 (s), 1493 (s), 1454 (s), 1367 (m), 1254 (m), 1162 (m), 1103 (m), 701
(90 MHz, CD3OD): d=177.2, 173.8, 172.1, 171.2, 169.7, 169.4, 166.7,
166.6, 164.1, 164.0, 163.4, 162.8, 162.7, 161.8, 155.4, 152.0, 151.6, 150.5,
149.5, 146.5, 143.3, 143.3, 142.4, 142.2, 142.0, 141.7, 129.6, 129.1, 128.6,
128.4, 127.7, 127.3, 124.2, 119.9, 117.0, 78.5, 69.0, 63.2, 61.7, 60.4, 59.4,
57.3, 54.6, 49.7, 46.2, 41.9, 39.1, 35.8, 30.9, 26.8, 26.0, 25.9, 18.7, 18.7, 18.6,
12.5, ꢀ4.6, ꢀ5.4, ppm; IR (neat): n˜ =3382 (s), 2960 (m), 2926 (m), 1667
(s), 1651 (s), 1538 (m), 1504 (m), 1368 (w), 1252 (m), 1159 (w), 1101 (m),
838 (w), 778 (w) cmꢀ1; HRMS (ESI): m/z: calcd for C63H72N14O11S6SiK:
1459.3236 [M+K+], found 1459.3265.
(s) cmꢀ1
[M+H+], found 1295.3241.
ACHTREUNG(S,S)-[2-(2-Carbamoylpyrrolidin-1-yl)-1-hydroxymethyl-2-oxoethyl]-car-
; HRMS (ESI): m/z: calcd for C58H67N12O9S6Si: 1295.3242
bamic acid tert-butyl ester (59): Tetrabutylammonium fluoride (TBAF)
(1m in THF, 220 mL, 0.22 mmol) was added at 08C to a stirred solution of
TBS-ether 43 (77.0 mg, 0.19 mmol) in THF (3 mL). After 30 min the re-
action mixture was allowed to reach room temperature and the stirring
was continued for 1 h. The reaction mixture was partitioned between
CHCl3 (10 mL) and saturated aq. NH4Cl (3 mL) and the aqueous layer
was extracted with CHCl3 (3”10 mL). The combined organic extracts
were dried (Na2SO4), concentrated in vacuo and purified by flash chro-
matography (EtOAc/MeOH 95:5) to yield the desired dipetide 59
GE2270 A (1)
Oxazoline formation: DAST (9.70 mL, 11.8 mg, 73.0 mmol) was added at
ꢀ788C to a stirred solution of b-hydroxy amide 61 (4.00 mg, 2.80 mmol)
in CH2Cl2 (0.8 mL). After 1 h, anhydrous potassium carbonate (18.0 mg,
0.13 mmol) was added and the resulting white slurry was allowed to
reach room temperature. The reaction was poured into saturated aq.
NaHCO3 (5 mL) and the biphasicmixture was extratced with CH 2Cl2
(5”10 mL). The combined organic extracts were dried (Na2SO4) and con-
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2337

T. Bach et al.
centrated in vacuo to afford crude oxazoline (5 mg) as a colorless oil.
TBS deprotection. The crude oxazoline was dissolved in THF (1.5 mL) at
08C and TBAF (1m in THF, 9 mL, 9 mmol) was added. After 2 h the reac-
tion mixture was concentrated and purified by reverse phase HPLC (RP,
ODS-A, MeCN/H2O 20:80 ! 100:0, 30 min, 15 mLminꢀ1) to afford
GE2270 A (1) (2.1 mg, 1.6 mmol, 57%) as a colorless solid. HPLC purifi-
cation was necessary to free the natural product from the TBAF derived
side product. Rf =0.11 (CH2Cl2/MeOH=9:1); [a]2D0 =+71.4 (c=0.15 in
CHCl3); 1H NMR (750 MHz, [D6]DMSO, 258C): d=9.02 [d, 3J=7.8 Hz,
1H; NHCHCHOH], 8.71–8.66 (m, 2H; 2”NH), 8.61 (s, 1H; CHF), 8.55
(s, 1H; CHA), 8.46–8.42 (m, 1H; CH2NH), 8.43 (d, 3J=8.1 Hz, 1H;
CHpy), 8.30 (s, 1H; CHB), 8.28 (d, 3J=8.1 Hz, 1H; CHpy), 7.40–7.36 (m,
2H; CONHH, NHCH3), 7.36 (s, 1H; CHC), 7.33–7.31 (m, 2H; CHph),
7.29 (virt. t, 3J=7.5 Hz, 2H; CHph), 7.26–7.22 (m, 1H; CHph), 6.96 (s, 1H;
CONHH), 6.03 (d, 3J=4.5 Hz, 1H; OH), 5.30 (virt. td, 3J=8.6, 3J=
3.7 Hz, 1H; CHCH2), 5.26–5.22 (m, 2H; OCH2CH, CHCHOH), 5.20
(dd, 3J=8.1, 3J=4.8 Hz, 1H; CHCHiPr), 5.02–4.96 (m, 3H; CH2OCH3,
CHOH), 4.81 (virt t, 2/3J=7.6 Hz, 1H; OCHHCH), 4.57 (dd, 2/3J=9.7 Hz,
2/3J=7.9 Hz, 1H; OCHHCH), 4.27 (dd, 2J=17.0, 3J=8.2 Hz, 1H;
CHHNH), 4.24 (dd, 3J=8.7, 3J=3.6 Hz, 1H; CHCONH2), 3.99–3.95 (m,
1H; NCHH), 3.86–3.81 (m, 1H; NCHH), 3.79 (dd, 2J=17.0, 3J=3.8 Hz,
1H; CHHNH), 3.39 (s, 3H; CH2OCH3), 2.72 (dd, 2J=16.1, 3J=3.7 Hz,
1H; CHHCONHCH3), 2.59 (s, 3H; CH3), 2.48 (d, 3J=4.6 Hz, 3H;
[6] E. Selva, P. Ferrari, M. Kurz, P. Tavecchia, L. Colombo, S. Stella, E.
Restelli, B. P. Goldstein, F. Ripamonti, M. Denaro, J. Antibiot. 1995,
48, 1039–1042.
[7] Review: M. C. Bagley, J. W. Dale, E. A. Merritt, X. Xiong, Chem.
Rev. 2005, 105, 685–714.
[8] a) K. Shimanaka, N. Kinoshita, H. Iinuma, M. Hamada, T. Takeuchi,
J. Antibiot. 1994, 47, 668–674; b) K. Shimanaka, Y. Takahashi, H.
Iinuma, H. Naganawa, T. Takeuchi, J. Antibiot. 1994, 47, 1145–1152;
c) K. Shimanaka, Y. Takahashi, H. Iinuma, H. Naganawa, T. Takeu-
chi, J. Antibiot. 1994, 47, 1153–1159.
[9] a) J. F. Pagano, M. J. Weinstein, H. A. Stout, R. Donovick, Antibiot.
Ann. 1955–56, 554–559; b) J. Vandeputte, J. D. Dutcher, Antibiot.
Ann. 1955–56, 560–561; c) B. A. Steinberg, W. P. Jambor, L. O.
Suydam, Antibiot. Ann. 1955–56, 562–565.
[10] a) J. Walker, A. Olesker, L. Valente, R. Rabanal, G. Lukacs, J.
Chem. Soc. Chem. Commun. 1977, 706–708; b) B. W. Bycroft, M. S.
Gowland, J. Chem. Soc. Chem. Commun. 1978, 256–258.
[11] a) P. Tavecchia, P. Gentili, M. Kurz, C. Sottani, R. Bonfichi, S. Lo-
ciuro, E. Selva, J. Antibiot. 1994, 47, 1564–1567; b) P. Tavecchia, P.
Gentili, M. Kurz, C. Sottani, R. Bonfichi, E. Selva, S. Lociuro, E.
Restelli, R. Ciabatti, Tetrahedron 1995, 51, 4867–4890.
[12] Examples include: a) chloramphenicol: J. D. Dunitz, J. Am. Chem.
Soc. 1952, 74, 995–999; b) biphenomycin A: F. K. Brown, J. C.
Hempel, J. S. Dixon, S. Amato, L. Mueller, P. W. Jeffs, J. Am. Chem.
Soc. 1989, 111, 7328–7333; c) lysobactin: A. A. Tymiak, T. J. McCor-
mick, S. E. Unger, J. Org. Chem. 1989, 54, 1149–1157.
NHCH3), 2.21–2.10 (m, 2H; CH
CH2CHH), 1.36–1.28 (m, 1H; CHHCONHCH3), 0.88 [d, J=6.8 Hz, 3H;
CH(CH3) (CH3)(CH3)];
(CH3)], 0.85 ppm [d, 3J=6.8 Hz, 3H; CH
ACHTRE(UGN CH3)2, CH2CHH), 1.99–1.86 (m, 3H;
3
A
G
A
ACHTREUNG
13C NMR (150 MHz, [D6]DMSO, 258C): d=173.5, 171.0, 169.6, 169.4,
168.3, 168.1, 167.6, 165.4, 164.6, 161.2, 161.0, 160.3, 160.1, 159.2, 153.2,
150.2, 150.0, 149.3, 146.7, 144.6, 143.6, 141.9, 141.7, 141.2, 140.7, 139.3,
128.6, 127.8, 127.6, 127.5, 126.9, 126.6, 123.0, 118.5, 116.3, 73.7, 69.3, 67.8,
67.3, 60.0, 58.5, 58.0, 55.2, 47.9, 46.9, 41.0, 37.4, 33.9, 29.6, 25.7, 24.1, 18.4,
17.8, 11.9 ppm; IR (neat): n˜ = 3351 (bs), 2960 (m), 2942 (s), 1659 (s),
1443 (m), 1225 (w), 1159 (w), 1094 (m), 750 (m) cmꢀ1; HRMS (ESI):
m/z: calcd for C56H56N15O10S6: 1290.2659 [M+H+], found 1290.2664.
[13] G. Heckmann, T. Bach, Angew. Chem. 2005, 117, 1223–1226;
Angew. Chem. Int. Ed. 2005, 44, 1199–1201.
[14] In this structure, the proline amide in the Northern part of
GE2270 A was erroneously shown as being (R)-configured.
[15] Recent review on the synthesis of thiopeptides: R. A. Hughes, C. J.
Moody, Angew. Chem. 2007, 119, 8076–8101.
[16] a) U. Mocek, A. R. Knaggs, R. Tsuchiya, T. Nguyen, J. M. Beale,
H. G. Floss, J. Am. Chem. Soc. 1993, 115, 7557–7568; b) U. Mocek,
Z. Zeng, D. OꢁHagan, P. Zhou, L. D. G. Fan, J. M. Beale, H. G.
Floss, J. Am. Chem. Soc. 1993, 115, 7992–8001.
[17] a) R. A. Hughes, S. P. Thompson, L. Alcaraz, C. J. Moody, Chem.
Commun. 2004, 946–948; b) R. A. Hughes, S. P. Thompson, L. Al-
caraz, C. J. Moody, J. Am. Chem. Soc. 2005, 127, 15644–15651.
[18] a) K. C. Nicolaou, B. S. Safina, C. Funke, M. Zak, F. J. Zecri, Angew.
Chem. 2002, 114, 2017–2020; Angew. Chem. Int. Ed. 2002, 41, 1937–
1940; b) K. C. Nicolaou, M. Nevalainen, B. S. Safina, M. Zak, S.
Bulat, Angew. Chem. 2002, 114, 2021–2025; Angew. Chem. Int. Ed.
2002, 41, 1941–1945; c) K. C. Nicolaou, M. Nevalainen, M. Zak, S.
Bulat, M. Bella, B. S. Safina, Angew. Chem. 2003, 115, 3540–3546;
Angew. Chem. Int. Ed. 2003, 42, 3418–3424; d) K. C. Nicolaou, B. S.
Safina, M. Zak, A. A. Estrada, S. H. Lee, Angew. Chem. 2004, 116,
5197–5202; Angew. Chem. Int. Ed. 2004, 43, 5087–5092; e) K. C.
Nicolaou, M. Zak, B. S. Safina, S. H. Lee, A. A. Estrada, Angew.
Chem. 2004, 116, 5202–5207; Angew. Chem. Int. Ed. 2004, 43, 5092–
5097; f) K. C. Nicolaou, B. S. Safina, M. Zak, S. H. Lee, M. Nevalai-
nen, M. Bella, A. A. Estrada, C. Funke, F. J. ZØcri, S. Bulat, J. Am.
Chem. Soc. 2005, 127, 11159–11175; g) K. C. Nicolaou, M. Zak, B. S.
Safina, A. A. Estrada, S. H. Lee, M. Nevalainen, J. Am. Chem. Soc.
2005, 127, 11176–11183.
[19] K. C. Nicolaou, B. Zou, D. H. Dethe, D. B. Li, Chen, D. Y.-K. Chen,
Angew. Chem. 2006, 118, 7950–7956; Angew. Chem. Int. Ed. 2006,
45, 7786–7792.
[20] a) M. C. Bagley, X. Xiong, Org. Lett. 2004, 6, 3401–3404; b) M. C.
Bagley, C. Glover, D. Chevis, Synlett 2005, 649–651.
[21] a) C. J. Moody, M. C. Bagley, J. Chem. Soc. Chem. Commun. 1998,
2049–2050; b) M. C. Bagley, K. E. Bashford, C. L. Hesketh, C. J.
Moody, J. Am. Chem. Soc. 2000, 122, 3301–3313.
Acknowledgements
This project was supported by the Deutsche Forschungsgemeinschaft (Ba
1372-9), by the Alexander von Humboldt Foundation (Research Scholar-
ship to O.D.), by the Universität Bayern e.V. (Predoctoral Scholarship to
H.M.M.), and by the Fonds der Chemischen Industrie. We thank Wacker-
Chemie (Munich), DSM Fine Chemicals (Linz) und Umicore (Hanau)
for the donation of chemicals. The help of Dipl.-Chem. Jochen Klages in
recording the NMR spectra is gratefully acknowledged. We thank Danie-
la Kampen for initial synthetic experiments conducted in her Diplom
thesis.
[1] Review: F. von Nussbaum, M. Brands, B. Hinzen, S. Weigand, D.
Häbich, Angew. Chem. 2006, 118, 5194–5254; Angew. Chem. Int.
Ed. 2006, 45, 5072–5129.
[2] Reviews: a) R. Hilgenfeld, J. R. Mesters, T. Hogg, The Ribosome:
Structure, Function, Antibiotics and Cellular Interactions (Eds.: R. A.
Garrett, S. R. Douthwaite, A. Liljas, A. T. Matheson, P. B. Moore,
H. F. Noller), ASM Press Washington DC, 2000, pp. 347–357;
b) I. M. Krab, A. Parmeggiani, Prog. Nucl. Acid Res. Mol. Biol.
2002, 71, 513–551.
[3] A. Parmeggiani, P. Nissen, FEBS Lett. 2006, 580, 4576–4581.
[4] a) S. E. Heffron, F. Jurnak, Biochemistry 2000, 39, 37–45; b) A. Par-
meggiani, I. M. Krab, S. Okamura, R. C. Nielsen, J. Nyborg, P.
Nissen, Biochemistry 2006, 45, 6846–6857.
[22] a) M. A. Ciufolini, Y.-C. Shen, J. Org. Chem. 1997, 62, 3804–3805;
b) M. A. Ciufolini, Y.-C. Shen, Org. Lett. 1999, 1, 1843–1846.
[23] T. R. Kelly, C. Jagoe, Z. Gu, Tetrahedron Lett. 1991, 32, 4263–4266.
[24] a) K. Okumura, H. Saito, C.-g. Shin, K. Umemura, J. Yoshimura,
Bull. Chem. Soc. Jpn. 1998, 71, 1863–1870; b) K. Okumura, T.
[5] a) E. Selva, G. Beretta, N. Montanini, G. S. Saddler, L. Gastaldo, P.
Ferrari, R. Lorenzetti, P. Landini, F. Ripamonti, B. P. Goldstein, M.
Berti, L. Montanaro, M. Denaro, J. Antibiot. 1991, 44, 693–701; b) J.
Kettenring, L. Colombo, P. Ferrari, P. Tavecchia, M. Nebuloni, K.
Vekey, G. G. Gallo, E. Selva, J. Antibiot. 1991, 44, 702–715.
2338
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2322 – 2339

A Thiazolyl Peptide Antibiotic
FULL PAPER
Suzuki, C.-g. Shin, Heterocycles 2000, 53, 765–770; c) T. Suzuki, A.
Nagasaki, K. Okumura, C.-g. Shin, Heterocycles 2001, 55, 835–840.
[25] For a related sequence, see also: a) K. Okumura, M. Shigekuni, Y.
Nakamura, C.-g. Shin, Chem. Lett. 1996, 1025–1026; b) C.-g. Shin,
K. Okumura, M. Shigekuni, Y. Nakamura, Chem. Lett. 1998, 139–
140; c) K. Okumura, A. Ito, D. Yoshioka, C.-g. Shin, Heterocycles
1998, 48, 1319–1324; d) K. Okumura, Y. Nakamura, C.-g. Shin, Bull.
Chem. Soc. Jpn. 1999, 72, 1561–1569; e) K. Okumura, T. Suzuki, Y.
Nakamura, C.-g. Shin, Bull Chem. Soc. Jpn. 1999, 72, 2483–2490.
[26] For a comprehensive review on regioselective cross-coupling reac-
tions on heterocycles, see: S. Schrçter, C. Stock, T. Bach, Tetrahe-
dron 2005, 61, 2245–2267.
[27] For cross-coupling reactions on heterocycles in general, see: a) J. J.
Li, G. W. Gribble, Palladium in Heterocyclic Chemistry, Pergamon
Press, Oxford, 2000; b) V. N. Kalinin, Synthesis, 1992, 413–432.
[28] Review: M. Schlosser, Angew. Chem. 2005, 117, 380–398; Angew.
Chem. Int. Ed. 2005, 44, 376–393.
[29] a) F. Trecourt, G. Breton, V. Bonnet, F. Mongin, F. Marsais, G. QuØ-
guiner, Tetrahedron 2000, 56, 1349–1360; b) T. Takahashi, Y. Li, P.
Stepnicka, M. Kitamura, Y. Liu, K. Nakajima, M. Kotora, J. Am.
Chem. Soc. 2002, 124, 576–582.
[30] A. Bouillon, J.-C. Lancelot, V. Collot, P. B. Bovy, S. Rault, Tetrahe-
dron 2002, 58, 2885–2890.
[31] W. Yang, Y. Wang, J. R. Corte, Org. Lett. 2003, 5, 3131–3134.
[32] a) D. Milstein, J. K. Stille, J. Am. Chem. Soc. 1979, 101, 4992–4998;
b) J. K. Stille, Angew. Chem. 1986, 98, 504–519; Angew. Chem. Int.
Ed. Engl. 1986, 25, 508–524; c) V. Farina, V. Krishnamurthy, W. J.
Scott, Org. React. 1997, 50, 1–652.
[33] a) T. Bach, L. Krüger, Eur. J. Org. Chem. 1999, 2045–2057; b) T.
Bach, S. Heuser, Tetrahedron Lett. 2000, 41, 1707–1710; c) T. Bach,
S. Heuser, Angew. Chem. 2001, 113, 3283–3284; Angew. Chem. Int.
Ed. 2001, 40, 3184–3185; d) T. Bach, S. Heuser, Chem. Eur. J. 2002,
8, 5585–5592; e) T. Bach, M. Bartels, Synthesis 2003, 925–939; f) S.
Schrçter, T. Bach, Synlett 2005, 1957–1959.
[47] B. S. Pedersen, S. Scheibye, N. H. Nilsson, S.-O. Lawesson, Bull. Soc.
Chim. Belg. 1978, 87, 223–228.
[48] Similar problems have been reported in ref. [17b].
[49] J. A. Dale, D. L. Dull, H. S. Mosher, J. Org. Chem. 1969, 34, 2543–
2549. We used both enantiomericforms of a-methoxy-a-trifluoro-
methylphenylacetyl chloride so that both diastereomeric products
were at hand for analytical comparison (see Supporting Informa-
tion).
[50] A. Hantzsch, Ann. Chem. 1889, 250, 257–273.
[51] S. Deng, J. Taunton, Org. Lett. 2005, 7, 299–301.
[52] E. Aguilar, A. I. Meyers, Tetrahedron Lett. 1994, 35, 2473–2476.
[53] a) J. C. Sheehan, P. A. Cruickshank, G. L. Boshart, J. Org. Chem.
1961, 26, 2525–2528; b) S. V. Downing, E. Aguilar, A. I. Meyers, J.
Org. Chem. 1999, 64, 826–831.
[54] K. C. Nicolaou, A. A. Estrada, M. Zak, S. H. Lee, B. S. Safina,
Angew. Chem. 2005, 117, 1402–1406; Angew. Chem. Int. Ed. 2005,
44, 1378–1382.
[55] E. D. Laganis, B. L. Chenard, Tetrahedron Lett. 1984, 25, 5831–5834.
[56] S. Guttmann, R. A. Boissonnas, Helv. Chim. Acta 1958, 41, 1852–
1867.
[57] M. T. Rispens, O. J. Gelling, A. H. M. de Vries, A. Meetsma, F. van
Bolhuis, B. Feringa, Tetrahedron 1996, 52, 3521–3546.
[58] A. Peyman, K.-H. Budt, J. Spanig, D. Ruppert, Angew. Chem. 1993,
105, 1852–1854; Angew. Chem. Int. Ed. Engl. 1993, 32, 1720–1722
and references therein.
[59] A. Avenoza, J. H. Busto, J. M. Pergrina, F. Rodrꢂguez, J. Org. Chem.
2002, 67, 4241–4249 and references therein.
[60] Precedence for this cyclization reaction does, however, exist; see for
example: M. M. Lenman, S. L. Ingham, D. Gani, Chem. Commun.
1996, 85–87.
[61] O. Delgado, G. Heckmann, H. M. Müller, T. Bach, J. Org. Chem.
2006, 71, 4599–4608.
[62] T. Bach, S. Heuser, J. Org. Chem. 2002, 67, 5789–5795.
[63] H. Azizian, C. Eaborn, A. Pidcock, J. Organomet. Chem. 1981, 215,
49–58.
[64] For a review on the synthesis of polyazoles related to natural prod-
ucts, see: E. Riego, D. Hernꢃndez, F. Albericio, M. ꢄlvarez, Synthe-
sis 2005, 1907–1922.
[65] For a review on the use of Pd-catalyzed cross-coupling reactions in
natural product synthesis, see: K. C. Nicolaou, P. G. Bulger, D.
Sarlah, Angew. Chem. 2005, 117, 4516–4563; Angew. Chem. Int. Ed.
2005, 44, 4442–4489.
[66] For a similar template effect, see for example: H. Guthmann, M.
Kçnemann, T. Bach, Eur. J. Org. Chem. 2007, 632–638.
[67] One reviewer suggested that the more rapid formation of the pep-
tide bond vs the diketopiperazine is due to the presence of a free
hydroxy group in intermediate 60. It was suggested that first the
ester of this hydroxy group is coupled to the acid followed by an
O,N acyl shift under basic conditions.
[68] a) P. Lafargue, P. Guenot, J.-P. Lellouche, Heterocycles 1995, 41,
947–958; b) A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno, D. R. Wil-
liams, Org. Lett. 2000, 2, 1165–1168.
[34] H. M. Müller, O. Delgado, T. Bach, Angew. Chem. 2007, 119, 4855–
4858; Angew. Chem. Int. Ed. 2007, 46, 4771–4774.
[35] A. T. Ung, S. G. Pyne, Tetrahedron: Asymmetry 1998, 9, 1395–1407.
[36] T. R. Kelly, F. R. Lang, J. Org. Chem. 1996, 61, 4623–4633.
[37] a) E.-i. Negishi, A. O. King, N. Okukado, J. Org. Chem. 1977, 42,
1821–1823; b) E.-i. Negishi, Acc. Chem. Res. 1982, 15, 340–348.
[38] H. J. den Hertog, E. Fahrenhorst, Recl. Trav. Chim. Pays-Bas 1948,
67, 380.
[39] A. S. B. Prasad, T. M. Stevenson, J. R. Citineni, V. Nyzam, P. Kno-
chel, Tetrahedron 1997, 53, 7237–7254.
[40] V. Canibano, J. F. Rodriguez, M. Santos, M. A. Sanz-Tejedor, M. C.
Carreno, G. Gonzalez, J. L. Garcia-Ruano, Synthesis 2001, 2175–
2179.
[41] J. T. Kuethe, A. Wong, I. W. Davies, J. Org. Chem. 2004, 69, 7752–
7754.
[42] E. Marzi, A. Bigi, M. Schlosser, Eur. J. Org. Chem. 2001, 1371–
1376.
[43] G. S. Hanan, U. S. Schubert, D. Volkmer, E. Riviere, J. M. Lehn, N.
Kyritsakas, J. Fischer, Can. J. Chem. 1997, 75, 169–182.
[44] S. Gabriel, Chem. Ber. 1910, 43, 1283–1287.
[45] Analogous to: L. Somogyi, G. Haberhauer, J. Rebek, Jr., Tetrahe-
dron 2001, 57, 1699–1708.
[69] A solution for this problem has been recently suggested: Y. Kabura-
gi, Y. Kishi, Org. Lett. 2007, 9, 723–726.
[70] J. Clough, S. Chen, E. M. Gordon, C. Hackbarth, S. Lam, J. Trias,
R. J. White, G. Candiani, S. Donadio, G. Romanꢅ, R. Ciabatti, J. W.
Jacobs, Bioorg. Med. Chem. Lett. 2003, 13, 3409–3414.
[46] a) C. Hartmann, V. Meyer, Chem. Ber. 1893, 26, 1727–1732; b) M.
Frigerio, M. Santagostino, S. Sputore, J. Org. Chem. 1999, 64, 4537–
4538; c) J. D. More, N. S. Finney, Org. Lett. 2002, 4, 3001–3003.
Received: November 21, 2007
Published online: February 12, 2008
Chem. Eur. J. 2008, 14, 2322 – 2339
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2339