Final elucidation of the absolute configuration of the signal metabolite hormaomycin

Article abstract of DOI:10.1002/chem.200400406

The complete absolute configuration of hormaomycin 1a has been established by HPLC and HPLC/MS experiments with appropriately derivatized 4-propylprolines, (2S,4S)-6 and (2R,4R)-6, as well as 4-(Z)-propenylprolines. cis-5 and trans-5, and also feeding exp

Full text of DOI:10.1002/chem.200400406

FULL PAPER
Final Elucidation of the Absolute Configuration of the Signal Metabolite
Hormaomycin
Boris D. Zlatopolskiy,[a] Karin Loscha,[a] Petra Alvermann,[a] Sergei I. Kozhushkov,[a]
Sergej V. Nikolaev,[b] Axel Zeeck,*[a] and Armin de Meijere*[a]
Abstract: The complete absolute con-
figuration of hormaomycin 1a has been
established by HPLC and HPLC/MS
experiments with appropriately deriva-
tized 4-propylprolines, (2S,4S)-6 and
(2R,4R)-6, as well as 4-(Z)-propenyl-
prolines, cis-5 and trans-5, and also
feeding experiments with enantiomeri-
cally pure samples of the deuterium-la-
beled
3-(2’-nitrocyclopropyl)alanine,
were prepared for the first time and al-
lowed one to unequivocally assign the
hitherto unknown absolute configura-
tions of the last four stereocenters in
hormaomycin 1a. As a bonus, some
new information about the biosynthesis
of this molecule has also been gath-
ered.
(2S)-3,3-[D2]15 and (2S)-2,2’-[D2]15,
and 4-(Z)-propenylproline 2’,4-[D2]-
(2S,4R)-5. The latter five amino acids
Keywords: isotopic labeling · natu-
ral products · peptide lactone ·
structure elucidation
Introduction
acid residues have never been found in any natural product
before. The absolute configuration of the first four above-
mentioned structural elements of 1a and later, partially, the
configurations of the (3-Ncp)Ala residues were clarified.[1,3]
However, the absolute configuration of the two stereocen-
ters in the (4-Pe)Pro and at the a-carbons of both (3-
Ncp)Ala residues remained undetermined. Before the total
synthesis of 1a could be initiated,[4] a full elucidation of the
structure of 1a (Figure 1) had to be performed and this is
presented here.
Hormaomycin (1a), an unusual peptide lactone, which was
first isolated in 1989 from Streptomyces griseoflavus[1] and
found to have an interesting and surprisingly broad spec-
trum of biological activities,[2] shows structural features re-
markable even for the wide range of structurally flexible
secondary metabolites from microorganisms. The initial
structure investigation[1] disclosed that besides one residue
of the proteinogenic (S)-isoleucine [Ile], 1a contains two
units of 3-(2S,3R)-methylphenylalanine [(b-Me)Phe], one of
(R)-allo-threonine [a-Thr] as well as two moieties of 3-
(1’R,2’R)-(trans-2’-nitrocyclopropyl)alanine
[(3-Ncp)Ala]
and one of 4-(Z)-propenylproline [(4-Pe)Pro]. The side
chain of 1a is terminated with the residue of 5-chloro-1-hy-
droxypyrrole-2-carboxylic acid [Chpca]. The latter three
[a] Dr. B. D. Zlatopolskiy, K. Loscha, Dr. P. Alvermann,
Dr. S. I. Kozhushkov, Prof. Dr. A. Zeeck, Prof. Dr. A. de Meijere
Institut für Organische und Biomolekulare Chemie
der Georg-August-Universität Göttingen
Tammannstrasse 2, 37077 Göttingen (Germany)
Fax : (+49)551-39475
E-mail: azeeck@gwdg.de
armin.demeijere@chemie.uni-goettingen.de
[b] Dr. S. V. Nikolaev
Chemistry Department, St. Petersburg State University
Universitetski Prospekt 26, Petrodvorets
198504 St. Petersburg (Russia)
Figure 1. The absolute configuration of hormaomycin 1a. Asterisks (*)
mark the stereocenters for which the absolute configuration was un-
known before.
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
4708
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200400406
Chem. Eur. J. 2004, 10, 4708 – 4717

4708 – 4717
Results and Discussion
To begin with the determination of the configuration of the
(4-Pe)Pro moiety in hormaomycin 1a, a fast and simple
access to 4-propenylproline [H-(4-Pe)Pro-OH], even as a
mixture of stereoisomers, was desirable. In view of the lack
of any reasonably effective preparative routes to H-(4-
Pe)Pro-OH,[5] the procedure of Moniot et al.[6] was chosen
as a starting point. Tosylate 3, prepared from 4-hydroxypro-
line (2) according to the published procedure,[6] was success-
fully alkylated with an excess of lithium di-(Z)-propenylcup-
rate[7] in THF/Et2O (1:2) to give a mixture of N-Boc-protect-
ed cis- (cis-4) and trans- (trans-4) 4-propenylprolines (ratio
1:2.2 according to the 13C NMR spectrum), which was insep-
arable by conventional methods, in 81% yield. The isomeri-
zation of the double bond was insignificant and did not
exceed 4–10% according to the 13C NMR spectrum (deter-
mined by comparison of the signal intensities of the methyl
groups: d=13.1 ppm for the (Z)-isomer and 17.8 ppm for
the (E)-isomer). A portion of this mixture was separated by
preparative HPLC to give the pure individual diastereo-
mers.[8]
Scheme 1. Synthesis of cis- (cis-5) and trans- (trans-5) 4-(Z)-propenylpro-
lines and NOE effects for cis-4 and trans-4. a) lithium di-(Z)-propenyl-
cuprate, Et2O/THF 2:1, ꢀ40!208C, 16 h; b) TFA, 208C, 20 min; c) 2m
HCl/EtOAc, then four crystallizations from CH2Cl2/Et2O.
The relative configuration of each stereoisomer was un-
hormaomycin 1a. An MS/MS technique was used to deter-
mine the appropriate fraction. These experiments showed
that 1a definitely contains a cis-(4-Pe)Pro moiety.[12]
ambiguously
established
by
NOESY experiments
1
(Scheme 1). Although the H NMR spectra of these substan-
ces showed the presence of two rotamers, the signals of 2-H,
3-Ha, 3-Hb and 4-H were cleanly separated from each other.
For the trans-isomer, strong correlations between 2-H (d=
4.40 ppm) and 3-Ha (d=1.82, 2.01 ppm), as well as between
4-H (d=3.18–3.33 ppm) and 3-Hb (d=2.18, 2.41 ppm), while
no correlations between either 2-H and 3-Hb, or 4-H and 3-
Ha, or between 2-H and 4-H, were observed. On the contra-
ry, for the cis-isomer the direct correlation between 2-H
(d=4.22, 4.28 ppm) and 4-H (d=2.99–3.16 ppm) as well as
strong correlations between 2-H and 3-Ha (d=2.38,
2.44 ppm), and 4-H and 3-Ha could be seen, while no corre-
lation between 2-H and 3-Hb (d=1.75, 1.88 ppm) and only a
weak correlation between 4-H and 3-Hb were observed.
Both isomers were deprotected with trifluoroacetic acid,
and the appropriate amino acids—cis-5 and trans-5—were
obtained. Whereas trans-4 was deprotected smoothly to give
the corresponding hydrotrifluoroacetate in virtually quanti-
tative yield, the desired amino acid cis-5·HCl was obtained
after several recrystallizations of the hydrochloride in only
26% yield.
Once the relative configuration of the (4-Pe)Pro residue
in hormaomycin 1a was established, the absolute configura-
tion of this fragment had to be determined. Several experi-
ments failed, because it was impossible to isolate this amino
acid (or its derivatives) from the total hydrolysate of the
natural depsipeptide due to the instability of 4-(Z)-prope-
nylproline in the presence of strong acids routinely applied
for the total hydrolysis of peptides. At the same time, it had
been known for some time that the catalytic hydrogenation
of 1a with palladium or platinum catalysts transforms the
(4-Pe)Pro moiety into the 4-propylproline residue (and also
reduces other functionalities of the molecule).[13] Therefore,
both enantiomers of cis-4-propylproline, which were neces-
sary for the determination of the absolute configuration,
were prepared (Scheme 2).
One sample of each amino acid was derivatized with the
(S)-FDVA [Na-2,4-dinitro-5-fluorophenyl-(S)-valine amide]
reagent[9] and the enantiomeric purity of each isomer was
determined according to the method of Marfey.[10] This ex-
periment showed ee 69% for the trans-isomer, and almost
full racemization for the cis-isomer. These data were in
good agreement with the data on the enantiomeric purities
of the isomeric 4-phenylprolines prepared by Moniot et al.
according to this procedure.[6] A second sample of each ste-
reoisomer of 4-(Z)-propenylproline was transformed into a
dabsyl (4-dimethylaminoazobenzene-4’-sulfonyl) deriva-
tive,[11] and these were compared by HPLC with the sample
of DABS-(4-Pe)Pro-OH obtained after the HPLC separa-
tion from the DABS-derivatized total hydrolysate of natural
Scheme 2. Synthesis of both enantiomers of cis-4-propylproline (6). a) H2,
Pd/C, EtOAc, 208C, 15 h; b) 2m HCl/EtOAc, 208C, 1 h; c) LiBEt3H,
THF, ꢀ788C, 1 h; d) Et3SiH, BF3·Et2O, CH2Cl2, ꢀ788C, 2.5 h; e) 6m HCl,
858C, 2h.
Chem. Eur. J. 2004, 10, 4708 – 4717
www.chemeurj.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4709

A. de Meijere et al.
FULL PAPER
Towards that end, the double bond of the enantiomerical-
ly pure N-Boc protected 4-(Z)-propenylproline (2S,4R)-4[14]
was hydrogenated over 10% Pd/C in ethyl acetate, and then
the N-Boc protection was removed with 2m HCl in EtOAc
to give the known (2S,4S)-6[15] as the hydrochloride in 79%
yield. The enantiomer (2R,4R)-6 was obtained after the de-
protection of (2R,4R)-10. The latter was prepared from the
cis-allyl derivative (2R,4R)-8 which in turn was obtained by
alkylation with allyl bromide of the enolate of the N-Boc-
protected tert-butyl (R)-pyroglutamate (R)-7[16] in the same
way as it was described for the (2S,4S)-enantiomer.[17] Hy-
drogenation of the double bond in (2R,4R)-8 over 10% Pd/
C, followed by two-step reduction of the lactam moiety of
the intermediate N,O-protected 4-propylpyroglutamic acid
(2R,4R)-9 according to the protocol of Ezquerra et al. com-
pleted this synthesis.[18] Both enantiomers of cis-4-propylpro-
line were transformed to (S)-FDVA derivatives, which were
compared with the correspondingly derivatized hydrolysate
of hydrogenated hormaomycin 1a using an HPLC/MS tech-
nique according to the advanced Marfey method.[19] These
If D2O was used for the quenching of the enolate, a re-
versed selectivity was observed. The mixture of epimers was
separated, the trans-isomer epi-[D]11 was deprotonated
again, and the same mixture of epimers was obtained after
quenching of the enolate. The required isomer was separat-
ed and combined with the first portion to give the deuterat-
ed product [D]11 (90% deuterium content according to
1H NMR spectrum) in 67% overall yield.[21] This intermedi-
ate was transformed into the N-Boc protected doubly deute-
rium-labeled (2S,4R)-4-(Z)-propenylproline as it was previ-
ously described for the nondeuterated compound
(Scheme 4).[14,22] In this case an ylide, obtained from (1,1-di-
deuterio)ethyltriphenylphosphonium bromide was used on
the Wittig olefination step.[23] The cleavage of Boc group
with TFA gave the necessary product only in 36% yield.
Therefore, several other conditions for the Boc-deprotection
were tried (Scheme 4).[24] To increase the yield, the p-nitro-
benzyl ester [D2]12 (which was prepared by esterification of
(2S,4R)-[D2]4[25] with p-nitrobenzyl bromide and K2CO3 in
MeCN in 93% yield) was N-Boc deprotected with 5m HCl
in Et2O to give the corresponding amino ester as a hydro-
chloride in 82% yield. The latter was hydrolyzed with equi-
molar quantity of 1m NaOH to give, after careful acidifica-
tion, the dideuterated amino acid (2S,4R)-[D2]5 in almost
quantitative yield as a mixture with NaCl. This mixture was
directly used for the feeding experiment.
experiments showed that hormaomycin 1a contains
(2S,4R)-(4-Pe)Pro residue.
a
This conclusion was in accord with the results of the feed-
ing experiments with doubly deuterated (2S,4R)-2’,4-[D2]5.
The positions for the deuterium labeling were chosen to es-
tablish the possible biotransformation of the deuterated
amino acid into its stereoisomers: (R)-cis or (S)-trans. If epi-
merization at the chiral centers of this amino acid in the
course of the biosynthesis of hormaomycin 1a had taken
place, it would have caused the loss of the deuterium label
at C-4, and the peptide lactone labeled only at C-2’ of the
(4-Pe)Pro fragment would have been obtained in the feed-
ing experiments. In addition, the second deuterium label at
The feeding of (2S,4R)-2’,4-[D2]5 led to the correspond-
ingly doubly deuterated 1b without any loss of deuterium
label. These experiments also cleanly demonstrated that the
biosynthesis of H-(4-Pe)Pro-OH in cell takes place before
the assembly of the peptide chain of 1a.
In the course of previous investigations to elucidate the
configurations of the (3-Ncp)Ala moieties in 1a it became
1
C-2’ led to a visible change in the H NMR spectrum of 1a
in a relatively peak-free region (between 5.3 and 6.0 ppm)
as well as avoided any scrambling of the deuterium label
during the construction of the double bond.
The nitrile 11[14] was treated with a small excess of LDA
in THF at ꢀ788C, and the intermediate enolate was
quenched with the saturated solution of Na2SO4 in D2O at
the same temperature (to obtain more of the desired cis-
isomer according to Takano et al.).[20] After this the separa-
ble mixture of [D]11 and its epimer epi-[D]11 was obtained
in 2:1 ratio (Scheme 3).
Scheme 4. The final stages in the preparation of the doubly deuterium-la-
beled amino acid (2S,4R)-[D2]5. a) TFA (neat), 208C, 20 min, then 15 lyo-
philisations from water; b) 5n HCl/Et2O, 20 8C, 4 h; c) BF3·Et2O, CH2Cl2,
208C, 1 h; d) TMSI, CH2Cl2, 08C, 30 min; e) TFA/Me2S/CH2Cl2, 08C, 1 h;
f) SnCl4, CH2Cl2, 2 80C, 2h; g) 180 8C, 30 min; h) p-O2N-PhCH2Br,
K2CO3, MeCN, 708C, 3 h; i) 5n HCl/Et2O, 20 8C, 3 h; j) 1n NaOH, 08C,
30 min, then 1n HCl.
Scheme 3. Introduction of the deuterium label at C-4 of the nitrile 11.
a) LiHMDS, THF, ꢀ788C, 90 min; b) Na2SO4/D2O.
4710
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 4708 – 4717

Hormaomycin
4708 – 4717
known that both (3-Ncp)Ala residues have the same
(1’R,2’R) configuration in the nitrocyclopropyl ring and are
epimers with respect to the configuration at the a-carbons,[3]
yet it remained unclear which epimer was incorporated in
the ring and which was attached in the side chain. To clarify
this situation, feeding experiments with enantiomerically
pure deuterium-labeled H-(3-Ncp)Ala-OH had to be carried
out. The enantioselective alkylation of the Oppolzer sultam-
derived glycine equivalent 13[26] with the known 1,1-dideu-
terio-(2’-nitrocyclopropyl)methyl iodide 1,1-[D2]16[27] gave a
mixture of enantiomerically pure diastereomers 14 in 15%
yield (Scheme 5). The subsequent two-step hydrolysis of the
latter under acidic and then under basic conditions gave the
corresponding mixture of diastereomers of the 3,3-dideuter-
io-(2’-nitrocyclopropyl)alanine (2S)-3,3-[D2]15.[28] The corre-
spondingly deuterated hormaomycin 1c was indeed ob-
tained in the feeding experiments with (2S)-3,3-[D2]15. Ac-
cording to its ESI-MS spectrum, the molecular mass of this
2
product was four units higher than that of 1a. The H, 13C
1
and H NMR spectra confirmed that the 3-[D2]-(3-Ncp)Ala
moiety had been incorporated both in the ring and in the
side chain of 1a with the same probability. This indicates
firstly that H-(3-Ncp)Ala-OH is completely biosynthesized
before the assembly of the amino acids takes place and sec-
ondly that the cells of the hormaomycin-producing strain
comprise an epimerase which can convert a solely available
(2S)-isomer of H-(3-Ncp)Ala-OH to the (2R)-isomer neces-
sary to complete the biosynthesis of 1a. In order to prove
this implication, feeding experiments with one of the enan-
tiomerically pure H-(3-Ncp)Ala-OH, labeled at C-2, had to
be carried out.
Scheme 5. Synthesis of deuterated (2S,1’RS,2’RS)-3-(trans-2’-nitrocyclo-
propyl)alanines 3,3-[D2]15 and 2,2’-[D2]15 as well as related deuterium
exchange experiments. Reagents and conditions: a) nBuLi, THF, ꢀ708C,
1 h; b) 1,1-[D2]16, HMPT, THF, ꢀ78 !208C, 48 h; c) 0.5m HCl, THF,
208C, 48 h; d) LiOH, THF/H2O, 20 8C, 48 h; e) Na2CO3 (cat.), MeOD/
CDCl3/D2O, 50 8C, 24 h (twice); f) 16, NaH, CD3CN/DMF, ꢀ70!208C,
50 min; g) 6m HCl, MeOH, 658C, 10 min; h) Amberlite IRA-120 in H+
form; i) NaH, CD3CN/DMF, ꢀ25!208C, 50 min.
Recently, the diastereoselective alkylation of the Belo-
kon-type complex (2S)-17[29] with racemic trans-(2’-nitrocy-
clopropyl)methyl iodide 16 was applied to prepare all four
possible isomers of H-(3-Ncp)Ala-OH in good yield.[30] Fol-
lowing this protocol, the deuterated compound (2S)-[D2]17,
which was prepared by a hydrogen/deuterium isotope ex-
change in excellent yield (94%) and with a very high degree
of deuteration (> 97% of the incorporation of two deuteri-
um atoms per molecule),[31] was deprotonated with sodium
hydride. The enolate thus generated in turn was alkylated
with the racemic iodide 16 in a mixture of [D3]acetonitrile[32]
and dimethylformamide to give a 2-deuterated (2S)-(2’-ni-
trocyclopropyl)alanine precursor in 70% yield. The MS and
NMR spectra of this product showed that a hydrogen/deute-
rium exchange had also taken place in the 2’’-position of the
cyclopropyl ring, in other words, the product was the precur-
sor to (2S)-2,2’-dideuterio-2-(2’-nitrocyclopropyl)alanine
(2S)-2,2’-[D2]15. This amino acid was obtained with 88%
deuterium label in the 2- and 60% in the 2’-position by hy-
drolysis of (2S)-[D2]18 and ion exchange chromatography in
68% yield.
of the (3-Ncp)Ala moiety of 18.[34] Under the same condi-
tions, nitrocyclopropane (20) and the O-trityl protected
(1’S,2’S)-(2’-nitrocyclopropyl)methanol 19[3] gave deuterated
analogues in 25 and 72% yield. In the case of 19, the
1H NMR spectrum of the crude product showed that partial
(about 5–7%) epimerization had taken place.
The successful feeding experiment with (2S)-2,2’-[D2]15
gave the deuterium labeled hormaomycin 1d which, accord-
1
ing to its H NMR data (see Figure 2), was enriched with
deuterium at three positions, namely at the 2’-position of
the (3-Ncp)Ala I moiety in the lactone ring of 1a as well as
at the C-2and C-2 ’ positions of the (3-Ncp)Ala II residue in
the side chain of 1a.[35] This observation rigorously proves
that the (2R)-(3-Ncp)Ala moiety is placed in the ring and
the (2S)-residue is positioned in the side chain of 1a.
It is also noteworthy that the degree of deuteration at C-2
position of (3-Ncp)Ala II (40%) was significantly lower
than the 88% in the fed sample of (2S)-2,2’-[D2]15 as meas-
ured by the respective decrease of the intensity of the 2-H
1
This observation appeared to markedly contradict earlier
results by Seebach et al. on the possibility of deprotonating
and alkylating of nitrocyclopropanes.[33] Therefore, the un-
deuterated Ni complex (2S,2’S,1’’R,2’’R)-18[24] in a CD3CN/
DMF mixture was treated with sodium hydride to give, after
quenching with dilute D2SO4 in D2O, 55 and 65%, respec-
tively, incorporation of deuterium in the 2’- and 2’’-positions
signal of (3-Ncp)Ala II moiety in the H NMR spectrum of
1d compared with the appropriate signal in the 1H NMR
spectrum of the nondeuterated hormaomycin 1a. By con-
trast, the degree of deuteration at C-2’ of both (3-Ncp)Ala
moieties of 1d (60%) was as high as for (2S)-2,2’-[D2]15.
This corroborates that the deuterium at C-2of (2 S)-2,2’-
[D2]15 is partly exchanged before the hormaomycin mole-
Chem. Eur. J. 2004, 10, 4708 – 4717
www.chemeurj.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4711

A. de Meijere et al.
FULL PAPER
Figure 2. Comparison of 1H NMR (600 MHz, CDCl3) spectra of hormaomycin 1a and deuterium-labeled hormaomycins 1b, 1c, 1d. The arrows indicate
signals, the intensity of which has decreased due to deuteration.
cule is assembled at the respective synthetase or synthetases.
Normally, an epimerase is part of the corresponding module
of the multienzyme complex. In the case of this amino acid
two differently located epimerases possibly compete.
In summary, the hitherto unknown absolute configura-
tions of four stereocenters in hormaomycin 1a have been es-
tablished by appropriate HPLC and HPLC/MS as well as
feeding experiments with several enantiomerically pure deu-
terium-labeled 3-(2’-nitrocyclopropyl)alanine and cis-4-(Z)-
propenylproline specimen, all prepared for the first time. As
a benefit, some new information about the biosynthesis of
this molecule has also been gathered.
in D2O (d=67.19 ppm). IR spectra: Bruker IFS 66 (FT-IR) spectrometer
as KBr pellets or oils between KBr plates. MS: EI-MS: Finnigan MAT
95, 70 eV, high resolution EI-MS spectra with perfluorokerosene as refer-
ence substance; DCI-MS: Finnigan MAT 95, 200 eV, reactant gas NH3;
ESI-MS: Finnigan LCQ. HPLC-MS: pump: Flux Instruments Rheos
4000; degasser: Flux Instruments ERC 3415a; detector: Linear UVIS-
205; data system: Flux Instruments Janeiro; ESI: Finnigan LCQ, positive
and negative mode; data system: Finnigan LCQ Xcalibur; column: Crom
Superspher 100 RP-18 endcapped (4 mm, 2”100 mm); HPLC conditions:
eluent A: H2O (0.05% HCOOH), eluent B: 90% MeCN (0.05%
HCOOH), 30!70%
B . HPLC:
for 30 min, flow rate: 300 mLminꢀ1
pump: Kontron 322 system, detector: Kontron DAD 440, mixer: Kontron
HPLC 360, data system: Kontron Kromasystem 200, columns: Knauer
Nucleosil-100 C18 (analytical, 5 mm, 3 mm”250 mm), Knauer Nucleosil-
100 C18 (preparative, 5 mm, 8 mm”250 mm). HPLC conditions: A, iso-
cratic, 55% MeCN in H2O (0.1% TFA); B, isocratic 55% MeOH in H2O
for 20 min, then 55!75% MeOH for 8 min, then 75!80% MeOH for
10 min; C, isocratic, 78% acetonitrile in H2O (0.05% HCOOH). Optical
rotations: Perkin-Elmer 241 digital polarimeter, 1 dm3 cell. M.p.: Büchi
510 capillary melting point apparatus, uncorrected values. TLC: Macher-
ey-Nagel precoated sheets, 0.25 mm Sil G/UV254. Column chromatogra-
phy: Merck silica gel, grade 60, 230–400 mesh and Baker silica gel, 40–
140 mesh. Starting materials: Anhydrous diethyl ether and THF were ob-
tained by distillation from sodium/benzophenone, CH2Cl2 and DMF from
molecular sieves 4 Š. Compounds 3,[6] (2S,4R)-4,[14] (R)-13,[26] 16,[27,30] 1,1-
[D2]16,[27] 17,[29] 19[3] were prepared as described elsewhere, and (2R,4R)-
8 as it was previously described for the (2S,4S)-isomer.[15b] All other
chemicals were used as commercially available (Merck, Acros, BASF,
Bayer, Aldrich, Fluka, Hoechst, Degussa AG). Organic extracts were
dried over anhydrous MgSO4. Microbiology: Shaking incubator: Braun
Experimental Section
General remarks: Chemistry: 1H NMR spectra: Bruker AM 250
(250 MHz), AMX 300 (300 MHz) or Varian Unity 300 (300 MHz), Inova
500 (500 MHz), Inova 600 (600 MHz). Proton chemical shifts are report-
ed in ppm relative to residual peaks of deuterated solvents. Higher order
NMR spectra were approximately interpreted as first-order spectra, if
possible. 13C NMR spectra [additional DEPT (Distortionless Enhance-
ment by Polarization transfer) or APT (Attached Proton Test)]: Bruker
AM 250 (62.9 MHz), AMX 300 (75.5 MHz) or Varian Unity 300
(75.5 MHz), Inova 500 (125.7 MHz), Inova 600 (150.9 MHz) instruments.
13C chemical shifts are reported relative to peak of solvent or to dioxane
4712
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 4708 – 4717

Hormaomycin
4708 – 4717
BS4. Fermenter: Braun Biostat M. media (per liter of distilled water):
M2Ca: malt extract (10 g), glucose (4 g), yeast extract (4 g), agar-agar
(20 g), CaCO3 (0.5 g), pH 7.0; M6: d-mannitol (20 g), soybean flour
(20 g), meat extract (20 g), NaCl (2 g), l-valine (0.3 g), ZnSO4·6 H2O
(0.5 g), pH 7.3; M10: d-mannitol (50 g), l-asparagine (3 g), K2HPO4
(1 g), NaCl (25 g), MgSO4·7H2O (50 mg), CaCl2·2H2O (50 mg),
CH3COONa (420 mg), meso-inositol (100 mg), vitamins solution (1 mL):
thiamine hydrochloride (1 g), calcium d-pantothenate (1.2g), flavine
mononucleotide (1 g), nicotinic acid (2.3 g), pyridoxine hydrochloride
(12g), p-aminobenzoic acid (200 mg), vitamin B12 (100 mg), folic acid
(10 mg), biotin (6 mg), trace elements solution (10 mL): CaCl2·2H2O
(8 g), MnCl2·2H2O (5 g), ZnCl2 (50 mg), CuCl2·2H2O (50 mg),
FeCl2·6H2O (50 mg). In each case the quantities of compounds for the
preparation of 1 L of media are given. For the dilution demineralized
water was used. After adjusting the pH with 0.5m NaOH or 0.5m HCl,
all media were sterilized at 1218C for 30 min. The vitamins solution was
added to M10 after sterile filtration and autoclavation. All ingredients
were purchased from Merck, Gibco, Difco, Sigma, Riedel de Haen and
Henselwerk GmbH.
residue was filtered through a pad of silica gel (6 cm) with MeOH/CHCl3
1:10 to give, after evaporation of solvents, 4-(Z)-propenylproline
(155 mg, 81%) as an inseparable mixture of diastereomers with a ratio of
2.2:1 (according to the 13C NMR spectrum). An aliquot (25 mg) of this
mixture was separated by HPLC (conditions A, flow rate 3 mLminꢀ1) to
give the trans-4 isomer (13.8 mg, tR =6.84 min) and the cis-4 isomer
(5.4 mg, tR =6.35 min) in pure form.
trans-4: M.p. 41–458C; 1H NMR (500 MHz, CDCl3): d=1.42, 1.48 [2”s,
9H, C(CH3)3], 1.66 (d, J=6.0 Hz, 3H; 3’-H), 1.82(ddd, J=11.9, 11.9,
9.1 Hz, 0.7H, 3-Ha), 2.01 (ddd, J=11.9, 11.9, 10.3 Hz, 0.3H, 3-Ha), 2.11–
2.24 (m, 0.3H; 3-Hb), 2.41 (dd, J=12.6, 6.2 Hz, 0.7H; 3-Hb), 2.97 (dd, J=
10.0 Hz, 0.7H, 5-Ha), 3.05 (dd, J=9.5 Hz, 0.3H, 5-Ha), 3.18–3.33 (m, 1H,
4-H), 3.58 (dd, J=8.5, 8.5 Hz, 0.7H, 5-Hb), 3.76 (dd, J=9.3, 9.3 Hz, 0.3H,
5-Hb), 4.31 (d, J=9.5 Hz, 0.3H, 2-H), 4.40 (d, J=8.5 Hz, 0.7H, 2-H), 5.25
(dd, J=9.8, 9.8 Hz, 1H, 1’-H), 5.52(dq, J=9.8, 7.0 Hz 1H, 2’-H), 6.40–
7.50 (br, 1H, CO2H); 13C NMR (125.7 MHz, CDCl3): d=13.2( +, CH3),
28.2, 28.3 [+, C(CH3)3], 34.2, 35.2 (+, C-4), 35.1, 37.0 (ꢀ, C-3), 51.4, 51.9
(ꢀ, C-5), 58.9, 59.2( +, C-2), 80.3, 81.3 [Cquat, C(CH3)3], 126.7, 126.9 (+,
C-2’), 129.2, 129.6 (+, C-1’), 153.8, 156.0 (Cquat, NCO2), 175.4, 178.4
(Cquat, C-1); IR (KBr): n˜ =3023, 2981, 2880, 1722, 1650, 1404, 1227,
Total hydrolysis of hormaomycin 1a and hydrogenated hormaomycin:
The respective peptide was hydrolyzed with 6m HCl/AcOH (1:1, 0.5 mL
per 1 mg of peptide) in sealed tubes at 1058C within 24 h. The solvent
was then removed under reduced pressure, and the residue was used for
further derivatization.
1159 cmꢀ1 MS (EI, 70 eV): m/z (%): 255 (1) [M +], 210 (8) [M +
;
ꢀCHO2], 182()2 [ M +ꢀC4H9O], 154 (100) [M +ꢀC5H9O2], 110 (70)
[C7H12N+], 87 (8), 57 (80) [C4H9+]; HRMS (EI): calcd for C13H21NO4:
255.1471; found 255.1471.
cis-4: M.p. 65–688C; 1H NMR (500 MHz, CDCl3): d=1.41, 1.45 [2”s,
9H, C(CH3)3], 1.65 (d, J=6.5 Hz, 3H; 3’-H), 1.75 (ddd, J=12.5, 10.3,
10.3 Hz, 0.5H, 3-Hb), 1.88 (ddd, J=12.0, 10.0, 10.0 Hz, 0.5H, 3-Hb), 2.38
(dddd, J=13.0, 7.0, 7.0, 7.0 Hz, 0.5H; 3-Ha), 2.44 (dddd, J=12.5, 6.5, 6.5,
6.5 Hz, 0.5H; 3-Ha), 2.99–3.16 (m, 2H, 4-H, 5-Ha), 3.67 (dd, J=9.3,
7.3 Hz, 0.5H, 5-Hb), 3.69–3.78 (m, 0.5H, 5-Hb), 4.22 (dd, J=8.3, 8.3 Hz,
0.5H, 2-H), 4.28 (dd, J=8.3, 8.3 Hz, 0.5H, 2-H), 5.23 (ddq, J=10.8, 10.8,
1.8 Hz, 1H, 1’-H), 5.52(dq, J=10.8, 6.5 Hz, 1H, 2’-H), 6.82–7.57 (br, 1H,
CO2H); 13C NMR (125.7 MHz, CDCl3): d=13.2( +, CH3), 28.2, 28.4 [+,
C(CH3)3], 35.8, 36.0 (+, C-4), 35.9, 37.4 (ꢀ, C-3), 51.5, 52.1 (ꢀ, C-5),
59.3, 59.6 (+, C-2), 80.4, 81.0 [Cquat, C(CH3)3], 126.6, 126.9 (+, C-2’),
129.2, 129.3 (+, C-1’), 153.6, 155.6 (Cquat, NCO2), 176.3, 178.6 (Cquat, C-1);
IR (KBr): n˜ =3020, 2975, 2943, 1736, 1633, 1441, 1252, 1168 cmꢀ1; MS
(EI, 70 eV): m/z (%): 255 (1) [M +], 210 (8) [M +ꢀCHO2], 182(2) [ M +
ꢀC4H9O], 154 (100) [M +ꢀC5H9O2], 110 (70) [C7H12N+], 87 (8), 57 (80)
[C4H9+]; HRMS (EI): calcd for C13H21NO4: 255.1471; found 255.1471.
Derivatization of the amino acids 5 or the hydrolysate of 1a with DABS-
Cl (4-dimethylaminoazobenzene-4’-sulfonyl chloride): 20 drops of a fresh-
ly prepared solution of DABS-Cl (4 nmolmLꢀ1) was added to the sample
(ca. 0.1 mg) of cis-5 or trans-5, or the hydrolysate of 1a dissolved in five
drops of 50mm NaHCO3 (pH 8.1), and it was heated at 708C for 10 min.
After this, the reaction mixture was diluted with 50mm phosphate buffer
(1 mL; pH 7.0)/ethanol (1:1) and was directly used for the HPLC experi-
ments.
Derivatization of the amino acids 6 or the hydrolysate of hydrogenated
1a with (S)-FDVA: The pH of a solution of 0.4 mg of the respective
sample in H2O (0.5 mL) was adjusted to 9 with 1m NaHCO3 and a 1%
solution of (S)-FDVA in acetone (0.1 mL) was added. The reaction mix-
ture was stirred at 408C for 1 h, and then the pH was adjusted to 6–7
with 0.1m HCl. After this, the reaction mixture was diluted with MeCN
(1 mL) and was directly used for HPLC/MS experiments.
(2S,4R)-4-Tosyloxy-N-Boc-proline (3):[6] M.p. 138–1398C; Rf =0.19
trans-4-(Z)-Propenylproline hydrotrifluoroacetate (trans-5·TFA): Com-
pound trans-4 (13.8 mg, 0.054 mmol) was deprotected with TFA (1 mL)
at 208C for 20 min. The resulting solution was concentrated under re-
duced pressure to give the hydrotrifluoroacetate trans-5·TFA (13.9 mg,
96%) as a hydroscopic light pink solid. 1H NMR (250 MHz, CD3OD):
d=1.72(dd, J=7.0, 1.8 Hz, 3H, 3’-H), 2.11–2.29 (m, 1H, 3-Ha), 2.35–2.53
(m, 1H, 4-H), 3.02(dd J=9.5, 11 Hz, 1H, 3-Hb), 3.60 (dd, J=7.5,
11.1 Hz, 1H, 5-Hb), 4.55 (dd, J=4.7, 9.6 Hz, 1H, 2-H), 5.35 (ddq, J=9.8,
9.8, 1.8 Hz, 1H, 1’-H), 5.74 (ddq, J=9.8, 1.0, 7.0 Hz, 1H, 2’-H); the resid-
ual peak of CD2HOD superimposed the signal of 5-Ha; 13C NMR
(62.9 MHz, CD3OD): d=13.5 (+, C-3’), 36.2( ꢀ, C-3), 36.2( +, C-4), 51.9
(ꢀ, C-5), 60.7 (+, C-2), 117.9 (q, J=289 Hz, CF3), 128.9 (+, C-2’), 129.7
(+, C-1’), 162.1 (q, J=37.4 Hz, CF3CO), 171.9 (Cquat, C-1); IR (film): n˜ =
3250–1950, 1679, 1631, 1410, 1203, 1137 cmꢀ1; MS (EI, 70 eV): m/z (%):
155 (3) [M +], 110 (100) [M +ꢀCHO2], 87 (16), 69 (43) [CF3+], 51 (16),
45 (45) [CHO2+], 41 (15) [C3H5+]; HRMS (EI): calcd for C8H13NO2:
155.0946·113.9928; found 155.0946.
[hexane/EtOAc 1:2(1.5% AcOH)];
[
a]=ꢀ62.7 (c=0.33, CHCl3);
1H NMR (250 MHz, CDCl3): d=1.39, 1.44 [2”s, 9H, C(CH3)3], 2.14–2.67
(m, 2H, 3-HA, 5-HA), 2.46 (s, 3H, CH3), 3.45–3.81 (m, 2H, 3-HB, 5-HB),
4.39, 4.42(2”t,
J=8 Hz, 1H, 2-H), 5.02 (m, 1H, 4-H), 7.37 (d, J=
8.25 Hz, 2H, 3’-H), 7.79 (d, J=8.25 Hz, 2H, 2’-H); 13C NMR (62.9 MHz,
CDCl3): d=21.6 (+, CH3), 28.0, 28.2 [+, C(CH3)3], 35.3, 37.0 (ꢀ, C-3),
51.8, 52.3 (ꢀ, C-5), 51.2( +, C-2), 78.2, 78.6 [Cquat, C(CH3)3], 81.2, 81.8
(+, C-4), 127.7 (+, C-3’), 130.1 (+, C-2’), 133.1, 133.2(C quat, C-4’), 145.3
(Cquat, C-1’), 153.2, 155.0 (Cquat, NCO2), 175.2, 177.5 (Cquat, C-1); IR
(KBr): n˜ = 3062, 2985, 2956, 1754, 1635, 1433, 1171 cmꢀ1; elemental anal-
ysis calcd (%) for C17H23NO7S (385.4): C 52.98, H 6.01, N 3.63; found C
52.67, H 6.19, N 3.55.
cis- and trans-N-Boc-4-(Z)-Propenylprolines (cis-4 and trans-4): A solu-
tion of cis-1-bromopropene (0.3 mL, 5.01 mmol) in diethyl ether (3 mL)
was added within 5 min with stirring at ꢀ108C under argon (Ar) to a Li
chip (0.087 g, 12.53 mmol) in anhydrous diethyl ether (7 mL). The mix-
ture was stirred at the same temperature for an additional 1 h and was
then added to a suspension of CuBr·Me2S (0.529 g, 2.57 mmol) in diethyl
ether (15 mL) at ꢀ45 to ꢀ408C (internal temperature) within 10 min.
After 1 h, a solution of 3 (0.29 g, 0.75 mmol) in THF (10.5 mL) was
added within 15 min at the same temperature to the resulting almost
clear yellowish-green solution of lithium di-(Z)-propenylcuprate, and the
reaction mixture was allowed to warm up to 208C over a period of 16 h.
The dark-colored reaction mixture was cooled in an ice-water bath and
treated with saturated NH4Cl (5–10 mL). Sodium hydroxide (10%) was
added to adjust the pH to 10, and the organic layer was separated. The
aqueous phase was washed with Et2O (3”10 mL), acidified with 4n
H2SO4 to pH 2–3 and then extracted with ethyl acetate (2”20 mL). The
combined organic layers were washed with water (2”10 mL), then with
brine (2”5 mL), dried and concentrated under reduced pressure. The
cis-4-(Z)-Propenylproline hydrochloride (cis-5·HCl): Compound cis-4
(10.2mg, 0.040 mmol) was deprotected with TFA (1 mL) at 20 8C for
20 min. The resulting solution was concentrated under reduced pressure.
The residue was taken up with 2m HCl/EtOAc (3”2mL), and the solu-
tion then was concentrated under reduced pressure. The residue was re-
crystallized four times from CH2Cl2/Et2O to give, after prolonged drying
at 0.01 Torr, hydrochloride cis-5·HCl (2.0 mg, 26%) as a hydroscopic
brownish semisolid. 1H NMR (250 MHz, D2O): d=1.47 (dd, J=6.9,
1.9 Hz, 3H, 3’-H), 1.61–1.81 (m, 1H, 3-Hb), 2.45 (ddd, J=13.0, 7.1,
7.1 Hz, 1H, 3-Ha), 2.90 (dd J=10.1, 10.1 Hz, 1H, 5-Ha), 3.24 (ddddd, J=
9.0, 9.0, 9.0, 9.0, 9.0 Hz, 1H, 4-H), 3.31–3.42(m, 1-H, 5-H b), 4.25 (dd, J=
8.9 Hz, 1H, 2-H), 5.11 (ddq, J=8.8, 8.8, 1.9 Hz, 1H, 1’-H), 5.44–5.58 (m,
1H, 2’-H); 13C NMR (62.9 MHz, D2O): d=13.1 (+, C-3’), 35.6 (ꢀ, C-3),
36.5 (+, C-4), 50.8 (ꢀ, C-5), 60.5 (+, C-2), 127.7 (+, C-2’), 129.6 (+, C-
Chem. Eur. J. 2004, 10, 4708 – 4717
www.chemeurj.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4713

A. de Meijere et al.
FULL PAPER
1’), 172.7 (Cquat, C-1); MS (ESI): positive mode: m/z (%): 178 (6) [M +
+Na], 156 (100) [M ++H].
(2R,4R)-4-Propylproline hydrochloride [(2R,4R)-6·HCl]: Compound
(2R,4R)-10 (70 mg, 0.22 mmol) was heated with 6m HCl (2mL) at 85 8C
for 2h. The reaction mixture was concentrated under reduced pressure.
The traces of water were removed by codistillation with toluene (2”
10 mL), and the residue was recrystallized three times from CH2Cl2/Et2O
to give, after prolonged drying at 0.01 Torr, (2R,4R)ꢀ6·HCl (34 mg,
79%) as an extremely hygroscopic colorless solid. [a]2D0 =7.7 (c=0.88,
Determination of the relative configuration of the (4-Pe)Pro residue in 1:
The total hydrolysate of 1, derivatized with DABS-Cl, was separated by
HPLC (conditions B, flow rate 2.5 mLminꢀ1), and then ESI/MS spectra
were measured for the obtained fractions. The peak with tR =32.8 min
corresponded, according to its ESI/MS spectrum, to the (4-Pe)Pro deri-
vate. It was compared by HPLC with the derivates of cis- and trans-pro-
1
CHCl3); H NMR (250 MHz, D2O): d=0.83 (m, 3H, 3’-H), 1.19–1.54 (m,
4H, 1’-H, 2’-H), 1.70 (dddd, J=12.8, 12.8, 10.0, 10.0 Hz, 1H; 4-H), 2.26–
2.49 (m, 1H, 3-Ha), 2.50–2.68 (m, 1H, 3-Hb), 2.95 (dd, J=10.9, 10.9 Hz,
1H, 5-Ha), 3.50 (dd, J=10.9, 8.3 Hz, 1H, 5-Hb), 4.35 (dd, J=8.9, 8.9 Hz,
1H, 2-H); 1H NMR (250 MHz, CD3OD): d=0.99 (t, J=7 Hz, 3H, 3’-H),
1.30–1.62(m, 4H, 1 ’-H, 2’-H), 1.74 (dddd, J=11.7, 11.7, 10.2, 10.2 Hz,
1H; 4-H), 2.33–2.68 (m, 1H, 3-Ha), 2.65 (ddd, J=12.2, 6.7, 6.7 Hz, 1H, 3-
Hb), 2.98 (dd, J=10.2, 10.2 Hz, 1H, 5-Ha), 3.53 (dd, J=11.7, 7.8 Hz, 1H,
5-Hb), 4.37 (dd, J=8.6, 8.6 Hz, 1H, 2-H); 13C NMR (62.9 MHz, D2O):
d=14.1 (+, C-3’), 21.5 (ꢀ, C-2’), 34.6 (ꢀ, C-1’), 35.3 (ꢀ, C-3), 38.8 (+,
C-4), 51.6 (ꢀ, C-5), 60.6 (+, C-2), 173.0 (Cquat, C-1); IR (KBr): n˜ =3432,
2960, 2931, 2873, 1736, 1389, 1252 cmꢀ1; MS (EI, 70 eV): m/z (%): 157
(1) [M +], 112(100) [ M +ꢀHCO2], 87 (4), 70 (5) [C4H8N+], 69 (11), 43
(15) [C3H7+], 41 (18) [C3H5+]; HRMS (EI): calcd for C8H15NO2:
157.1103; found 157.1103.
penylproline. Conditions B, flow rate 0.8 mLminꢀ1
: cis-isomer: tR =
32.57 min, mixed sample: 1 peak; trans-isomer: tR =33.37 min, mixed
sample: 2peaks; conditions C, flow rate 0.8 mLmin ꢀ1: cis-isomer: tR =
9.71 min, mixed sample:
sample: 2peaks.
1 peak, trans-isomer: tR =10.04 min, mixed
(2S,4S)-Propylproline hydrochloride [(2S,4S)-(6)·HCl]: (2S,4R)-4 (20 mg,
0.078 mmol) was hydrogenated over 10% Pd/C (5 mg) in EtOAC (2mL)
under ambient pressure of hydrogen at 208C for 15 h. The reaction mix-
ture was filtered, the solvent was removed under reduced pressure, and
the residue was treated with 2n HCl in EtOAc at 208C for 1 h. The sol-
vent was removed and (2S,4S)-6·HCl was obtained as an extremely hy-
groscopic colorless solid (12mg, 79%). [ a]2D0 =ꢀ7.8 (c=1.0, CHCl3).
tert-Butyl (2R,4R)-N-Boc-4-propylpyroglutaminate [(2R,4R)-9]: Com-
pound (2R,4R)-8 (220 mg, 0.68 mmol) was hydrogenated over 10% Pd/C
(90 mg) in EtOAc (10 mL) under ambient pressure of hydrogen at 208C
for 15 h. The reaction mixture was filtered, the solvent was removed
under reduced pressure, and the crude product was purified by column
chromatography (Rf =0.46, hexane/EtOAc 4:1) to give (2R,4R)-9 as a
colorless oil (198 mg, 90%). 1H NMR (250 MHz, CDCl3): d=0.89 (t, J=
7.1 Hz, 3H, 3’-H), 1.28–1.50 (m, 3H, 1’-Ha, 2’-H), 1.46 [s, 9H, C(CH3)3],
1.48 [s, 9H, C(CH3)3], 1.52–1.67 (m, 1H, 3-Ha), 1.72–1.91 (m, 1H, 1’-Hb),
2.42–2.61 (m, 2H, 3-Hb, 4-H), 4.36 (dd, J=8.8, 6.3 Hz, 1H, 2-H);
13C NMR (62.9 MHz, CDCl3): d=13.6 (+, C-3’), 20.1 (ꢀ, C-2’), 27.7 (ꢀ,
C-1’), 27.7 [+, 2 ”C(CH3)3], 33.0 (ꢀ, C-3), 42.2 (+, C-4), 57.9 (+, C-2),
81.8 [Cquat, C(CH3)3], 83.0 [Cquat, C(CH3)3], 149.3 (Cquat, NCO2), 170.4
(Cquat, C-1), 175.4 (Cquat, C-5); IR (film): n˜ =2978, 2934, 2874, 1792, 1744,
1719, 1316, 1158 cmꢀ1; MS (CI): m/z (%): 362(8) [ M ++2NH4ꢀH], 345
(88) [M ++NH4], 328 (6) [M ++H].
Determination of the absolute configuration of the (4-Pe)Pro residue in
1a: Hormaomycin 1a (5 mg) was hydrogenated over 10% Pd on charcoal
(30 mg) in methanol (3 mL) under ambient pressure of hydrogen for
30 min. The reaction mixture was filtered and concentrated under re-
duced pressure to give the crude material, an aliquot of which (ca.
0.5 mg) was further hydrolyzed, and after the derivatization with (S)-
FDVA was used for HPLC/MS experiments. Detection: UV: 340 nm;
ESI-MS (positive) m/z: 437.7–438.7; ESI-MS (negative) m/z: 435.7–436.7;
(S)-FDVA-(2S,4S)-6: tR =14.77 min, mixed sample: 1 peak; (S)-FDVA-
(2R,4R)-6: tR =17.98 min, mixed sample: 2peaks.
(2S,4S)-
and
(2S,4R)-N-Boc-O-TBDMS-4-deuterio-4-cyanoprolinol,
[D]11 and epi-[D]11: An ice-cold solution of LDA, prepared from nBuLi
(15 mL of 2.44m solution in hexane, 36.60 mmol) and diisopropylamine
(5.9 mL, 41.85 mmol) in THF (30 mL), for 40 min was added dropwise to
a solution of the nitrile 11 (9.5 g, 27.90 mmol) in THF (50 mL) at ꢀ788C.
Stirring was continued for an additional 90 min, and then the reaction
was quenched by addition of a saturated solution of Na2SO4 in D2O
(10 mL) under vigorous stirring at the same temperature. The mixture
was allowed to warm to 208C, and it was then concentrated under re-
duced pressure, the residue was taken up with Et2O (70 mL), filtered
through Celiteꢁ, concentrated again and separated by column chromato-
graphy (EtOAc/hexane 1:3.7, cis-isomer: Rf =0.38, trans-isomer: Rf =
0.54) to give [D]11 (4.21 g) and a mixed fraction enriched with epi-[D]11
(3.62g, 10.63 mmol), which was dissolved in THF (20 mL), again treated
with a solution of LDA, prepared from nBuLi (5.7 mL, 13.91 mmol) and
diisopropylamine (2.23 mL, 15.91 mmol) in THF (15 mL), and quenched
with a saturated solution of Na2SO4 in D2O (3 mL) as it was described
above. Separation of the mixture by column chromatography gave a
second crop of the labeled cis-nitrile [D]11 (2.40 g). It was combined
with the first fraction and finally purified by column chromatography as
described above to give [D]11 (6.39 g, 67%, 90% D according to its
1H NMR spectrum) as a colorless oil, which gradually solidified to a col-
orless solid.
[D]11: m.p. 53–558C; [a]D20 =ꢀ25.9 (c=1.1, CHCl3); 1H NMR (250 MHz,
CDCl3): d=0.01 [s, 6H, Si(CH3)2], 0.88 [s, 9H, SiC(CH3)3], 1.41 [s, 9H,
C(CH3)3], 2.27–2.45 (m, 2H, 3-H), 2.92 (dd, J=8.3, 8.3 Hz, 0.1H, 4-H),
3.35 (d, J=11.3 Hz, 1H, 5-Ha), 3.53–4.05 (m, 4H, 1-H, 2-H, 5-Hb);
13C NMR (62.9 MHz, CDCl3): d=ꢀ5.6 [+, Si(CH3)2], 18.0 (Cquat, SiC),
25.7 [+, SiC(CH3)3], 26.2 (t, J=21.7 Hz, C-4), 28.2 [+,C(CH3)3], 30.9,
32.1 (ꢀ, C-3), 49.5, 49.9 (ꢀ, C-5), 57.7 (+, C-2), 62.1, 63.3 (ꢀ, C-1), 80.0
[Cquat, C(CH3)3], 119.9 (Cquat, CN), 153.3 (Cquat, NCO2); IR (film): n˜ =
2952, 2895, 2858, 2241, 1700, 1471, 1405, 1258, 1177, 1101 cmꢀ1; MS (CI):
m/z (%): 359 (1) [M ++NH4], 342(100) [ M ++H]; elemental analysis
(%) for C17H31DN2O3Si (341.5): C 59.78, H 9.74, N 8.20; found C 59.54,
H 9.60, N 8.03.
tert-Butyl (2R,4R)-N-Boc-4-propylprolinate [(2R,4R)-10]: A 1m solution
of lithium triethylborohydride in THF (0.8 mL) was added under an at-
mosphere of nitrogen to a solution of (2R,4R)-9 (198 mg, 0.61 mmol) in
THF (5 mL) at ꢀ788C. After 1 h the reaction was quenched with saturat-
ed aqueous NaHCO3 (2mL), and the mixture warmed to 0 8C. Three
drops of 30% H2O2 were added, and the mixture was stirred at 08C.
After 20 min, the organic solvent was removed under reduced pressure,
and the aqueous layer was extracted with CH2Cl2 (3”10 mL). The com-
bined organic layers were dried, filtered and concentrated. The residue
was dried at 0.01 Torr for 3 h, then dissolved in anhydrous CH2Cl2 (5 mL)
containing triethylsilane (0.1 mL, 0.62mmol), and the mixture was
cooled to ꢀ788C under an atmosphere of nitrogen. Boron trifluoride
etherate (0.1 g, 0.71 mmol) was then added to the reaction mixture for
10 min. After 30 min, triethylsilane (0.1 mL, 0.62mmol) and boron tri-
fluoride etherate (100 mg, 0.71 mmol) were added, and the reaction mix-
ture was stirred for 2h at the same temperature. Saturated aqueous
NaHCO3 (2mL) was added, and the reaction mixture was extracted with
CH2Cl2 (3”10 mL), the extracts were dried, concentrated under reduced
pressure, and the crude product was purified by column chromatography
(Rf =0.56, EtOAc/hexane 1:1) to give (2R,4R)-10 (150 mg, 79%) as a col-
orless oil. [a]2D0 =75.4 (c=0.98, CHCl3) [lit. for the (2S,4S)-isomer: [15b];
1
[a]2D0 =ꢀ52( c=1.1, CHCl3)]; H NMR (250 MHz, CDCl3): d=0.89 (t, J=
6.9 Hz, 3H, 3’-H), 1.22–1.38 (m, 5H, 3-Ha, 1’-H, 2’-H), 1.42[s, 9H,
C(CH3)3], 1.42, 1.43 [2”s, 9H, C(CH3)3], 1.97–2.21 (m, 1H, 4-H), 2.37
(ddd, J=8.1, 8.1, 8.1 Hz, 0.5H, 3-Hb), 2.42 (ddd, J=7.8, 7.8, 7.8 Hz, 0.5H,
3-Hb), 2.94 (dd, J=10.1, 10.1 Hz, 1H, 5-Ha), 3.62(dd, J=10.1, 7.3 Hz,
0.3H, 5-Hb), 3.74 (dd, J=10.1, 7.3 Hz, 0.7H, 5-Hb), 4.06 (dd, J=8.1,
8.1 Hz, 0.7H, 2-H), 4.11 (dd, J=8.1, 8.1 Hz, 0.3H, 2-H); 13C NMR
(62.9 MHz, CDCl3): d=14.1 (+, C-3’), 21.3 (ꢀ, C-2’), 27.9, 28.0 [+,
C(CH3)3], 28.3, 28.4 [+, C(CH3)3], 35.0, 36.2( ꢀ, C-1’), 37.3 (ꢀ, C-3),
37.6, 38.3 (+, C-4), 52.1, 52.4 (ꢀ, C-5), 59.8 (+, C-2), 79.4, 79.6 [Cquat
,
The mixed fractions enriched with the trans-epimer were purified by
column chromatography as described before to give epi-[D]11 (0.56 g,
C(CH3)3], 80.7 [Cquat, C(CH3)3], 153.8 (Cquat, NCO2), 172.4 (Cquat, C-1); IR
(film): n˜ =2980, 2931, 2871, 1717, 1703, 1398, 1367, 1174, 1152 cmꢀ1; MS
(CI): m/z (%): 331 (12) [M ++NH4], 314 (100) [M ++H].
5.9%, 82% D according to ESI-MS spectrum) as a colorless oil. [a]D20
=
ꢀ60.2( c=0.48, CHCl3); 1H NMR (250 MHz, CDCl3): d=0.01 [s, 6H,
4714
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 4708 – 4717

Hormaomycin
4708 – 4717
Si(CH3)2], 0.85 [s, 9H, SiC(CH3)3], 1.44 [s, 9H, C(CH3)3], 2.15–2.46 (m,
2H, 3-H), 3.24–3.47 (m, 0.2H, 4-H), 3.41–3.80 (m, 3.5H, 0.5”1-H, 2-H,
5-H), 3.87–4.07 (m, 1.5H, 1.5”1-H); 13C NMR (62.9 MHz, CDCl3): d=
ꢀ5.7 [+, Si(CH3)2], 18.0 (Cquat, SiC), 25.7 [+, SiC(CH3)3], 26.5 (t, J=
15.7 Hz, C-4), 28.2 [+, C(CH3)3], 32.4, 33.2 (ꢀ, C-3), 49.4, 49.5 (ꢀ, C-5),
57.5 (+, C-2), 63.5, 64.0 (ꢀ, C-1), 80.0 [Cquat, C(CH3)3], 120.0 (Cquat, CN),
153.2(C quat, NCO2); IR (film): n˜ =2955, 2930, 2885, 2858, 2245, 1702,
1473, 1393, 1257, 1177, 1121 cmꢀ1; MS (ESI): positive mode: m/z (%):
364 (48) [M ++Na].
with Et2O (10”10 mL). The pH of the water fraction was carefully ad-
justed to 6.5–7.5 with 1m HCl (ca. 2.40 mL), and water was removed
under reduced pressure to give, after prolonged drying at 0.01 Torr, a
mixture of (2S,4R)-[D2]5 and NaCl (0.655 g, 0.292 g of NaCl, 98%) as a
colorless solid. This product had the same spectral characteristics as the
one described above, and it was used for the feeding experiment without
any further manipulations.
Alkylation of (R)-13 with rac-trans-1-(iododideuteriomethyl)-2-nitrocy-
clopropane,(11,-[D 2]16)—Preparation of 14: n-BuLi (4.6 mL of 2.58n
solution in hexane, 11.90 mmol) within 30 min was added at ꢀ708C to a
solution of (R)-13 (2.43 g, 10.61 mmol) in anhydrous THF (600 mL). The
reaction mixture was stirred at the same temperature for 30 min, and a
solution of the iodide 1,1-[D2]16 (2.43 g, 10.63 mmol) in a mixture of
HMPT (15 mL) and THF (40 mL) was then added within 45 min. Stirring
was continued at the same temperature for an additional 24 h, and then
the mixture was allowed to warm to 208C over a period of 48 h, the re-
action was quenched with half-saturated sodium chloride (1 L) and the
mixture was extracted with CH2Cl2 (3”200 mL). The combined organic
layers were washed with water (2”100 mL), dried and concentrated
under reduced pressure. The residue was recrystallized from Et2O/light
petroleum to give 14 (850 mg, 15%) as a colorless solid. M.p. 2048C;
1H NMR (250 MHz, CDCl3): mixture of diastereomers: d=0.92(s, 3H,
CH3), 1.06 (s, 3H, CH3), 0.97–1.15 (m, 1H, 3’’-H), 1.26–1.46 (m, 2H),
1.63, 1.73 (ddd, J=5.7, 3.5, 9.4 Hz, 1H; 3’’-H), 1.85–2.15 (m, 6H), 3.36 (s,
2H, 6-H), 3.88 (dd, J=5.9, 6.4 Hz, 1H, 3-H), 3.97, 4.13 (ddd, J=3.3, 3.5,
6.8 Hz, 1H, 2’’-H), 4.76, 4.78 (s, 1H, 2’-H), 7.04–7.07, 7.12–7.16, 7.25–7.50
and 7.65–7.69 (m, 10H, Ph-H); 13C NMR (62.9 MHz, CDCl3): d=17.7,
18.8 (ꢀ, C-3’’), 19.8 (+, CH3), 20.6 (+, CH3), 23.0 (+, C-1’’), 26.4 (ꢀ, C-
9), 32.6 (ꢀ, C-8), 38.3 (ꢀ, C-2), 44.3 (+, C-1), 47.7 (Cquat, C-10*), 48.5
(Cquat, C-7*), 52.9 (ꢀ, C-6), 58.7, 59.9 (+, C-2’’), 64.7, 64.9 (+, C-2’*),
65.1, 65.2( +, C-3*), 127.4, 127.7, 128.0, 128.6, 128.8, 128.9 and 130.60 (+,
Ph-C), 139.0 (Cquat, Ph-C), 171.4 (Cquat, C=N and C=O); the signal of the
CD2 carbon could not be detected because of its low intensity; IR (KBr):
n˜ =3084, 3061, 2983, 2960, 2914, 1689, 1628, 1577, 1539, 1445, 1364, 1333,
1166 cmꢀ1; MS (EI, 70 eV): m/z (%): 537 (7) [M +], 491 (9) [M +ꢀNO2],
295 (100) [C18H15D2N2O2+]; HRMS (EI): calcd for C29H31D2N3O5S:
537.2266; found 537.2266.
(2S,4R)-4,2’-Dideuterio-4-(Z)-propenylproline,(2 S,4R)-[D2]5: The N-
protected acid (2S,4R)-[D2]4 (0.155 g, 0.602mmol) was dissolved in tri-
fluoroacetic acid (2mL) at 20 8C. After 20 min, all volatiles were re-
moved under reduced pressure, and the yellow oily residue was taken up
with bidistilled water (5 mL). The solution was then concentrated under
reduced pressure (this operation was repeated 15 times). The residue was
recrystallized three times from MeOH/Et2O to give (2S,4R)-[D2]5
(34 mg, 36%) as
a
colorless solid. [a]2D0 =ꢀ4.2( c=0.31, MeOH);
1H NMR (250 MHz, D2O): mixture of isotopomers: d=1.63 (s, 3H, 3’-
H), 1.75 (dd, J=12.5, 8.3 Hz, 1H, 3-Ha), 2.54 (dd, J=8.3, 12.5 Hz, 1H, 3-
Hb), 3.03 (d, J=11.5 Hz, 1H, 5-Ha), 3.36 (m, 0.15H, 4-H), 3.48 (d, J=
11.5 Hz, 1H, 5-Hb), 4.16 (dd, J=8.3, 8.3 Hz, 1H, 2-H), 5.27 (m, 1H, 1’-
H), 5.52(m, 0.15H, 2 ’-H); 13C NMR (62.9 MHz, D2O): d=12.3, 12.4 (+,
CH3), 35.5 (ꢀ, C-3), 35.8, (t, J=21.6 Hz, C-4), 36.1 (+, C-4), 50.0 (ꢀ, C-
5), 61.2( +, C-2), 127.3, 127.4 (+, C-1’), 127.8 (t, J=23.0 Hz, C-2’), 128.5
(+, C-2’), 174.2(C quat, C-1); IR (KBr): n˜ =3101, 3008, 2969, 2915, 2856,
2361, 1626, 1388, 1315 cmꢀ1; MS (EI, 70 eV): m/z (%): 157 (1) [M +], 112
(100) [M +ꢀCHO2], 87 (16), 54 (7), 41 (36) [C3H5+]; MS (ESI): positive
mode: m/z (%)=202 (16) [M +ꢀH+2Na], 180 (6) [M ++Na]; negative
mode: m/z: 156 (100) [MꢀꢀH]; HRMS (EI): calcd for [C8H11D2NO2]:
157.1072; found 157.1072.
p-Nitrobenzyl
(2S,4R)-N-Boc-4,2’-dideuterio-4-(Z)-propenylprolinate,
[D2]12: A suspension of K2CO3 (0.467 g, 3.379 mmol) in a solution of
(2S,4R)-[D2]4 (0.87 g, 3.381 mmol) and p-nitrobenzyl bromide (0.767 g,
3.550 mmol) in MeCN (5 mL) was stirred in a sealed tube at 708C for
3 h. The mixture was then cooled, diluted with Et2O (10 mL), filtered,
concentrated under reduced pressure, and the residue was purified by
column chromatography (Rf =0.43, EtOAc/hexane 1:3) to give [D2]12
(1.23 g, 93%) as
a
colorless glass. [a]2D0 =ꢀ37.0 (c=0.86, CHCl3);
(2S,1’RS,2’RS)-3,3-Dideuterio-3-(trans-2’-nitrocyclopropyl)alanine hydro-
chloride,(2 S)-3,3-[D2]15]: 0.5m HCl (21 mL) was added with stirring to a
solution of compound 14 (850 mg, 1.58 mmol) in THF (21 mL), and stir-
ring was continued for 48 h. Then the reaction mixture was taken up with
Et2O (40 mL), the layers were separated and the aqueous phase was ad-
ditionally extracted with Et2O (3”20 mL). The organic phases were dis-
carded, and the aqueous fraction was concentrated under reduced pres-
sure. The residue was dissolved in THF/H2O (1:1, 42mL), and
LiOH·H2O (265 mg, 6.32 mmol) was added. After 48 h, H2O (32mL) was
added, and the mixture was extracted with CH2Cl2 (3”20 mL). The pH
of the aqueous fraction was adjusted to 5 with 1m HCl, and it was con-
centrated to give (2S)-3,3-[D2]15 as a mixture with LiCl (540 mg, contain-
ing max. 280 mg of (2S)-3,3-[D2]15), which was directly used for the feed-
ing experiment. 1H NMR (250 MHz, D2O): d=1.13–1.28 (m, 1H, 3’-H),
1.71–1.84 (m, 1H, 3’-H), 1.92–2.06 (m, 1H, 1’-H), 3.71 (s, 1H, 2-H), 4.18–
4.26 (m, 1H, 2’-H); 13C NMR (62.9 MHz, D2O): d=19.1 (ꢀ, C-3’), 24.0
(+, C-1’), 33.0 (quint, J=25 Hz, C-3), 54.2 (+, C-2), 61.6 (+, C-2’), 173.0
(Cquat, C-1); IR (film): n˜ =3200–2700, 2530, 1992, 1729, 1544, 1484, 1371,
1H NMR (250 MHz, CDCl3): mixture of isotopomers: d=1.34, 1.44 [2s,
9H, C(CH3)3], 1.63 (s, 3H, 3’-H), 1.63–1.80 (m, 1H, 3-Ha), 2.37–2.51 (m,
1H, 3-Hb), 3.06 (d, J=10.5 Hz, 1H, 5-Ha), 3.67, 3.78 (2d, J=10.5 Hz,
1H, 5-Hb), 4.30, 4.37 (2dd, J=8.8, 8.8 Hz, 1H, 2-H), 5.19 (d, J=13.8 Hz,
1H, Bzl-Ha), 5.20–5.26 (m, 1H, 1’-H), 5.35 (d, J=13.8 Hz, 1H, Bzl-Hb),
5.49–5.59 (m, 0.2H, 2’-H), 7.52(d, J=8.5 Hz, 2H, Ar-H), 8.20 (dd, J=
8.5, 8.5 Hz, 2H, Ar-H); 13C NMR (62.9 MHz, CDCl3): d=12.9, 13.0 (+,
CH3), 28.0, 28.2 [+, C(CH3)3], 35.1 (t, J=23.3 Hz, C-4), 36.3 (+, C-4),
36.4, 37.3 (ꢀ, C-3), 51.3, 51.7 (ꢀ, C-5), 58.9, 59.0 (+, C-2), 64.9 (ꢀ, C-
Bzl), 79.9 [Cquat, C(CH3)3], 123.5, 123.6 (+, C-Ar), 126.7, 126.8 (+, C-2’),
128.0, 128.3 (+, C-Ar), 128.9, 129.0 (+, C-1’), 142.7, 143.1 (Cquat, C-Ar),
147.4, 147.6 (Cquat, C-Ar), 153.2, 154.0 (Cquat, NCO2), 172.3, 172.5 (Cquat
,
C-1); MS (EI, 70 eV): m/z (%): 392/391 (1/0.4) [M +], 291/290 (12/5) [M +
ꢀC5H9O2], 247/246 (4/2), 212/211 (50/20) [M +ꢀC8H6NO4], 156 (100)
[C8H10D2NO2+], 155 (35) [C8H11DNO2+], 112(80) [C 7H10D2N+], 111
(28) [C7H11DN+], 57 (92) [C4H9+]; HRMS (EI): calcd for C20H24D2N2O6:
392.1916; found 392.1916; elemental analysis (%) for C20H25DN2O6
(391.4): calcd for C 61.37, H 6.95, N 7.16; found C 61.41, H 6.66, N 7.08.
1210 cmꢀ1
.
(2S,4R)-4,2’-Dideuterio-4-(Z)-propenylproline,(2 S,4R)-[D2]5: A 5m HCl
solution in Et2O (20 mL) was added to a solution of the N-protected
amino ester [D2]12 (1.108 g, 2.823 mmol) in Et2O (5 mL) and stirring was
continued in the dark for 2h. The mixture was then filtered, and a fresh
5m HCl solution in Et2O (10 mL) was added. After 1 h, the mixture was
diluted with hexane (10 mL), and the formed precipitate was separated,
washed with Et2O and dried to give p-nitrobenzyl (2S,4R)-4,2’-dideuter-
io-4-(Z)-propenylprolinate hydrochloride (0.775 g, 83%). MS (ESI): posi-
tive mode: m/z (%): 293/292 (100/35) [M ++H].
2-Dideuterated nickel complex,(2 S)-2’,2’-[D2]17: Unlabelled (S)-17
(6.00 g, 12.043 mmol) was taken up with a mixture of CH3OD (10 mL),
D2O (7 mL) and CDCl3 (5 mL), and Na2CO3 (200 mg, 2.410 mmol) was
added. The mixture was stirred at 508C for 24 h in a sealed flask, concen-
trated under reduced pressure, CH3OD (10 mL), D2O (7 mL) and CDCl3
(5 mL) were added again, and the resulting biphasic reaction mixture was
stirred at 508C for an additional 24 h. The reaction mixture was then
cooled, diluted with CHCl3 (30 mL), the organic phase was separated,
washed with H2O (2”30 mL), dried and concentrated to give after re-
crystallization from CHCl3/Et2O/hexane (2S)-2’,2’-[D2]17 (5.66 g, 94%;
D2 and D contents > 97% and < 3% respectively, as determined by
MS) as a dark red solid. M.p. 210–2138C; 1H NMR (250 MHz, CDCl3):
d=1.97–2.19 (m, 2H, 4-H), 2.28–2.46 (m, 1H, 3-Ha), 2.46–2.63 (m, 1H, 3-
Hb), 3.21–3.41 (m, 1H, 5-Ha), 3.45 (dd, J=10.9, 5.6 Hz, 1H, 5-Hb), 3.59–
3.72(m, 1H, 2-H), 3.68 (d, J=25.0 Hz, 1H, Bzl-Ha), 4.46 (d, J=25 Hz,
A 1m NaOH solution (4.89 mL) was added dropwise to an ice-cold solu-
tion of p-nitrobenzyl (2S,4R)-4,2’-dideuterio-4-(Z)-propenylprolinate hy-
drochloride (0.775 g, 2.357 mmol) in MeOH (5 mL) within 10 min. Stir-
ring was continued at the same temperature for an additional 30 min,
after that the mixture was diluted with water (20 mL), methanol was re-
moved under reduced pressure, and the reaction mixture was extracted
Chem. Eur. J. 2004, 10, 4708 – 4717
www.chemeurj.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4715

A. de Meijere et al.
FULL PAPER
1H, Bzl-Hb), 6.63–6.72(m, 1H, Ar-H), 6.74–6.81 (m, 1H, Ar-H), 6.88–
7.01 (m, 1H, Ar-H), 7.03–7.12(m, 1H, Ar-H), 7.14–7.24 (m, 1H, Ar-H),
7.24–7.33 (m, 1H, Ar-H), 7.35–7.46 (m, 2H, Ar-H), 7.44–7.58 (m, 3H,
Ar-H), 8.04 (d, J=8 Hz, 2H, Ar-H), 8.26 (d, J=8.5 Hz, 1H, Ar-H);
13C NMR (62.9 MHz, CDCl3): d=23.4 (ꢀ, C-4), 30.3 (ꢀ, C-3), 57.2( ꢀ,
C-5), 62.8 (ꢀ, C-Bzl), 69.5 (+, C-2), 120.4, 123.9, 125.3, 125.9, 128.5,
128.7, 129.0, 129.2, 129.4, 131.4, 131.8 and 132.8 (+, Ar-C), 124.8, 133.0,
134.2, 142.2 (Cquat, Ar-C), 171.3 (Cquat, C=N), 176.9 (Cquat, C=O), 181.0
Deuteration of 19: Compound 19 (300 mg, 0.84 mmol) was treated with
NaH (40 mg, 60% suspension in mineral oil, 1.0 mmol) in CD3CN/DMF
(2:1, 1.5 mL) according to GP 1, and the crude product was purified by
column chromatography (Rf =0.41, EtOAc/hexane 1:14) to give the deu-
terated product 2’-[D]19 (215 mg, 72%; D content >99%, as determined
by MS) as a colorless solid. [a]D20 =63.4 (c=0.61, CHCl3) [lit. for the non-
deuterated 19:[3] [a]D20 =66.2( c=1.00, CHCl3)]; 1H NMR (250 MHz,
CDCl3): d=1.31 (dd, J=6.0, 7.5 Hz, 1H, 3’-Ha), 1.84 (dd, J=5.8, 10.5 Hz,
1H, 3’-Hb), 2.17–2.31 (m, 1H, 1’-H), 3.08 (dd, J=5.6, 10.1 Hz, 1H, 1-H),
3.30 (dd, J=4.5, 10.3 Hz, 1H), 7.22–7.43 (m, 15H, Ar-H); 13C NMR
(62.9 MHz, CDCl3): d=15.3 (ꢀ, C-3’), 25.2 (+, C-1’), 57.3 (t, J=27.7 Hz,
C-2’), 61.6 (ꢀ, C-1), 86.7 (Cquat, CPh3), 127.1, 127.9, 128.4 (+, Ar-C),
ꢀ
(Cquat, CO N); the signal of the CD2 carbon could not be detected be-
cause of its low intensity; IR (KBr): n˜ =2973, 2869, 1673, 1642, 1589,
1442, 1336, 1264, 1176 cmꢀ1; MS (EI, 70 eV): m/z (%): 499 (40) [M +],
455 (100) [M +ꢀCO2], 363 (8), 267 (10), 265 (17), 217 (41), 160 (41), 91
(78) [C7H7+]; HRMS (EI): calcd for C27H23D2N3NiO3: 499.1375; found
499.1375; elemental analysis calcd (%) for C27H23D2N3NiO3 (500.2): C
64.83, H 5.44, N 8.40; found C 65.03, H 5.77, N 8.32.
ꢀ
143.4 (Cquat, Ar C); IR (KBr): n˜ =3111, 3089, 3062, 3029, 2969, 2942,
2893, 2871, 1524, 1355, 1218, 1115, 1033 cmꢀ1; MS (EI, 70 eV): m/z (%):
360 (26) [M +], 283 (12) [M +ꢀC6H5], 243 (100) [CPh3+], 165 (44)
[CPh2+]; HRMS (EI): calcd for C23H20DNO3: 360.1584; found 360.1584;
elemental analysis calcd (%) for C23H20DNO3 (360.4): C 76.65, H 6.15, N
3.89; found C 76.47, H 6.06, N 3.72.
Alkylation of 2-dideuterated nickel complex,(2 S)-2’,2’-[D2]17 with rac-
trans-1-(iodomethyl)-2-nitrocyclopropane 16—Preparation of (2S,2’S)-
2’,2’’-[D2]18: A suspension of (S)-2’,2’-[D2]17 (4.60 g, 9.20 mmol) in a
mixture of DMF (4.0 mL) and CD3CN (8 mL) was degassed with three
freeze-pump-thaw cycles at ꢀ708C. NaH (504 mg, 60% in oil,
12.60 mmol) and iodide 16 (2.12 g, 9.34 mmol) were then added. The
cooling bath was removed, and the reaction mixture was vigorously stir-
red for 40 min. The reaction flask was then immersed in an ice/water
bath and first D2O (2mL), then D 2SO4 (1.26 g, 12.60 mmol) in D2O
(5 mL) were carefully added. After this, water (70 mL) was added, and
the separated dark-red oil was triturated to give a red-brown solid
(5.70 g) which was further purified by column chromatography (Rf =0.37,
CHCl3/acetone 6:1) and then by crystallization from CHCl3/MeOH to
give (2S,2’S)-2’,2’’-[D2]18 (3.87 g, 70%) as a dark-red solid. According to
its 1H NMR spectrum, the deuterium content was 85% at C-2’ and 60%
Deuteration of 20: Nitrocyclopropane (20) (255 mg, 2.93 mmol) was
treated with NaH (152mg, 60% suspension in mineral oil, 3.80 mmol) in
CD3CN/DMF (2:1, 2.4 mL) according to GP 1. The reaction was
quenched with 5% acetic acid in D2O (3 mL), and the mixture was ex-
tracted with Et2O (3”10 mL). The combined organic layers were washed
with water (3”10 mL), brine (2”10 mL), dried, filtrated through a short
(5 cm) column with alumina and concentrated under atmospheric pres-
sure at a bath temperature <458C. The residue was subjected to bulb-to-
bulb distillation under reduced pressure (125 Torr) and the fraction col-
lected at a bath temperature >508C was 1-[D]20 (72mg, 90% purity,
25%; D content >97%, as determined by 1H NMR) as a colorless
liquid. 1H NMR (250 MHz, CDCl3): d=1.12–1.22 (m, 2H), 1.62 (dd, J=
5.3, 5.3 Hz, 2H); 13C NMR (62.9 MHz, CDCl3): d=11.4 (ꢀ), 53.8 (t, J=
28.3 Hz); MS (EI, 70 eV): m/z (%): 46 (38) [NO2+], 42(100) [C 3H4D+].
at C-2’’, respectively. M.p.
>
2608C; [a]2D0 =1708 (c=0.05, MeOH);
1H NMR (250 MHz, CDCl3): mixture of diastereo- and isotopomers: d=
0.46–0.58 (m, 0.5H, 3’’-Ha), 0.85–0.96 (m, 0.5H, 3’’-Ha), 1.05–1.28 (m,
0.5H, 3’-Ha), 1.45 (dd, J=8.4, 14.4 Hz, 0.5H, 3’-Ha), 1.73–1.84 (m, 0.5H,
3’’-Hb), 1.84–1.96 (m, 0.5H, 3’’-Hb), 1.97–2.32 (m, 2.5H, 4-H, 0.5”3’-Hb),
2.38–2.58 (m, 2H, 3-Ha, 1’’-H), 2.58–2.83 (m, 1.5H, 3-Ha, 0.5”3’-Hb),
3.33–3.41 (m, 0.15H, 2’-H), 3.40–3.73 (m, 3H, 2-H, 5-H), 3.56 (dd, J=2.3,
12.8 Hz, 1H, Bzl-Ha), 3.92(ddd, J=3.4, 3.4, 7.0 Hz, 0.30H, 2’’-H, 2’’-H),
4.00 (dd, J=14.6, 3.9 Hz, 0.13H, 2’’-H), 4.44 (d, J=12.8 Hz, 1H, Bzl-Hb),
(2S,1’RS,2’RS)-2,2’-dideuterio-3-(2’-nitrocyclopropyl)alanine,(2
S)-2,2’-
[D2]15: 6m HCl (32mL) was added to a suspension of finely ground
(2S,2’S)-2’,2’’-[D2]18 (3.87 g, 6.47 mmol) in refluxing methanol (16 mL).
The mixture was gently heated under reflux for 10 min. The resulting
green syrup was then concentrated, the residue was taken up with ice-
cold water (50 mL), the precipitate was filtered off, and washed with cold
water (20 mL). The filtrate was combined with the washings, its pH was
carefully adjusted to 6 with aqueous ammonia, and then it was extracted
with CHCl3 (3”50 mL). Preswollen Amberlite IRA-120 in the H+ form
(90 mL) was added to the water fraction, and the mixture was stirred for
15 h. The ion-exchange resin was filtered off, washed with water until the
eluent reached pH 5 and treated with 10% aqueous NH3 (2”150 mL) for
2h. The combined filtrates were concentrated under reduced pressure to
give a colorless solid, which was dissolved in hot water (10 mL). The so-
lution was filtered and diluted with EtOH (20 mL). The formed precipi-
tate was filtered off to give (2S)-2,2’-[D2]15 (770 mg, 68%; 60% of
(2S,1’S,2’R)-isomer according to its 13C NMR spectrum) as a colorless
ꢀ
ꢀ
6.56–6.73 (m, 2H, Ar H), 6.88 (t, J=8 Hz, 1H, Ar H), 7.08–7.24 (m,
ꢀ
ꢀ
ꢀ
2H, Ar H), 7.29–7.59 (m, 5H, Ar H), 7.59–7.65 (m, 1H, Ar H), 8.07
ꢀ
ꢀ
(dt, J=7.8, 1.5 Hz, 2H, Ar H), 8.13 (dd, J=3.3, 8.5 Hz, 1H, Ar H);
13C NMR (62.9 MHz, CDCl3): d=17.7, 17.8, 18.2, 18.4 (ꢀ, C-3’’), 21.5,
21.57, 21.61, 21.7 (+, C-1’’), 23.8 (ꢀ, C-4), 30.6 (ꢀ, C-3), 36.1, 36.4 (ꢀ, C-
3’), 57.2( ꢀ, C-5), 58.5 (+, C-2), 63.1, 63.2 (ꢀ, C-Bzl), 68.3 (t, J=19.3 Hz,
C-2ꢂ), 69.97, 70.03 (+, C-2’’), 120.6, 123.5, 123.6, 127.08, 127.10, 127.15,
127.19, 128.70, 128.72, 128.8, 129.0, 129.3, 129.8, 130.0, 131.3, 132.3 and
ꢀ
ꢀ
133.0 (+, Ar C), 125.9, 133.2, 133.4, 133.4, 142.3 (Cquat, Ar C), 170.9,
ꢀ
171.0 (Cquat, C=N), 178.1, 178.2(C quat, C=O), 180.4 (Cquat, CO N); IR
(KBr): n˜ =3071, 2972, 2922, 2866, 1667, 1624, 1589, 1537, 1436, 1365,
1
1339, 1257, 1166 cmꢀ1; MS (EI, 70 eV): m/z (%): 598 (7) [M +], 553 (15)
solid. The H-spectrum showed 88% of deuterium at C-2and 58% at C-
[M + CO2], 439 (9), 348 (7), 217 (22), 160 (100), 91 (23) [C7H7+]; HRMS
2’, respectively. According to the ESI-MS spectrum, the D2 and D con-
ꢀ
tents were 52and 42% respectively. M.p. 193–195 8C (decomp.); [a]D20
=
(EI): calcd for C31H29DN4NiO5: 597.1632; found 597.1632; elemental
analysis calcd (%) for C31H29DN4NiO5 (598.3): C 62.23, H 5.22, N 9.36;
found C 61.94, H 5.12, N 9.22.
3.3 (c=0.3, H2O); 1H NMR (250 MHz, D2O): mixture of diastereo- and
isotopomers: d=1.24–1.39 (m, 1H, 3’-H), 1.73–2.25 (m, 4H, 3-H, 1’-H, 3’-
H), 3.80 (dd, J=5.9, 5.9 Hz, 1H, 2-H), 4.18–4.26 (m, 1H, 2’-H); 13C NMR
(62.9 MHz, D2O): d=20.4, 20.5, 20.7, 20.9 (ꢀ, C-3’), 24.3, 24.4, 24.6, 24.7
(+, C-1’), 33.7, 33.8 (ꢀ, C-3), 55.9 (t, J=23.4 Hz, C-2), 61.6, 61.7 (+, C-
2’), 176.0, 176.1 (Cquat, C-1); IR (film): n˜ =3418, 3034, 2102, 1598, 1530,
1440, 1394, 1365 cmꢀ1; MS (ESI): positive mode: m/z (%): 221/220 (40/
30) [M ++2NaꢀH]; negative mode: m/z (%): 175/174 (3/3) [MꢀH+].
Deuterium exchange experiments—Deuteration of (2S,2’S,1’’R,2’’R)-18,
19 and 20 (GP 1): NaH (0.6 mmol, 60% suspension in mineral oil) was
added at ꢀ258C to a stirred solution or suspension of the respective nitro
compound (0.3 mmol) in anhydrous CD3CN/DMF 2:1 (1.5 mL). The mix-
ture was allowed to warm to 208C, and stirring was continued for an ad-
ditional 50 min. The reaction was quenched with 0.5m D2SO4 in D2O
(1.2mL), and the mixture was diluted with water (15 mL). The crude
products were separated by filtration or extraction with Et2O and puri-
fied by recrystallization or distillation.
Feeding experiments—General procedure (GP 2): The Streptomyces gri-
seoflavus strain W-384 was obtained from Prof. H. Wolf (Stuttgart). The
M2Ca agar in a culture tube was inoculated with the spores of the strain
W-384 and the tube incubated at 288C until the sporulation was complet-
ed (grey colored aerial mycelia, orange colored agar, incubation up to
three weeks). The tubes were stored at 48C.
Deuteration
of
(2S,2’S,1’’R,2’’R)-18:
The
nickel
complex
(2S,2’S,1’’R,2’’R)-18 (250 mg, 0.42 mmol) was treated with NaH (38 mg,
60% suspension in mineral oil, 0.95 mmol) in CD3CN/DMF (2:1, 1.5 mL)
according to GP 1, and the crude product was crystallized from CHCl3/
Et2O to give the deuterated compound (181 mg, 71%; 55% and 65% in-
corporation of the deuterium label at the 2’- and 2’’- positions of the (3-
Aliquots of this material (2”2mL, not older than 2months) were used
to inoculate medium M6 (2”50 mL) in two 250 mL flasks, and the flasks
were incubated in a shaking incubator at 278C and 120 rpm for 31 h. The
seed culture was transferred into medium M10 [900 mL; just before inoc-
1
Ncp)Ala fragment, respectively, according to the H NMR spectrum).
4716
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 4708 – 4717

Hormaomycin
4708 – 4717
ulation the vitamins solution (1 mL) and five drops of olive oil (to pre-
vent foaming) were added] in a 1 L fermenter.
ed (2S)-4-(Z)-propenylprolines starting from (2S)-pyroglutamic acid
see: E. Melotto, Dissertation, Universität Göttingen (Germany),
1999.
The production culture was incubated at 278C, with a stirring rate of
700 rpm and an air flow at a rate of 1.6 vvm over 20 h. pH 7 was main-
tained automatically by addition of 2m citric acid. The temperature was
then decreased to 208C. After 24 h, the solution of the respective syn-
thetic precursor substance in distilled, sterilized water (50 mL; the pH of
this solution was adjusted to 7.0 with 0.5m NaOH or 0.5m HCl) was
added over a period of 10 h. After 66 h, the production culture was har-
vested by filtration. The separated mycelium was lyophilized, homoge-
nized and extracted with ethyl acetate (2”250 mL) by applying sonifica-
tion for 15 min. The collected organic extracts were filtered and concen-
trated under reduced pressure. The resulting crude material was purified
twice by column chromatography (acetone/hexane 2:3, Rf =0.33 and
CHCl3/CH3OH 9:1, Rf =0.63). The thus prepared deuterated hormaomy-
cins had purities of about 90% or higher, as determined by HPLC analy-
ses.
[6] J. K. Thottathil, J. L. Moniot, Tetrahedron Lett. 1986, 27, 151–154.
Although known, compound 3 was not even characterized by a melt-
ing point.
[7] Noticeable excess of the alkylating agent should be used because of
its low reactivity and instability at the reaction temperature.
[8] Later it was shown that the appropriate trans- and cis-prolinoles can
be separated by conventional column chromatography. See ref. [14].
[9] H. Brückner, C. Keller-Hoehl, Chromatographia 1990, 30, 621–629.
[10] P. Marfey, Carlsberg Res. Commun. 1984, 49, 591–596.
[11] R. Knecht, J.-Y. Chang, Anal. Chem. 1986, 58, 2375–2379.
[12] P. Alvermann, Dissertation, Universität Göttingen (Germany), 2001.
[13] E. Rößner, Dissertation, Universität Göttingen (Germany), 1989.
[14] B. Zlatopolskiy, H.-P. Kroll, E. Melotto, A. de Meijere, Eur. J. Org.
Chem. 2004, to be submitted.
[15] a) A. M. P. Koskinen, H. Rapoport, J. Org. Chem. 1989, 54, 1859–
1866; b) C. M. Moody, D. Young, J. Chem. Soc. Perkin Trans. 1 1997,
3519–3530.
[16] This compound was prepared as described for its (S)-enantiomer: E.
Morera, F. Pinnen, G. Lucente, Org. Lett. 2002, 4, 1139–1142.
[17] M. Steger, D. W. Young, Tetrahedron 1999, 55, 7935–7956.
[18] J. Ezquerra, A. Escribano, A. Rubio, M. J. Remuiꢃꢄn, J. J. Vaquero,
Tetrahedron Lett. 1995, 36, 6149–6152.
[19] Compare: a) K. Fujii, Y. Ikai, T. Mayumi, H. Oka, M. Suzuki, K.
Harada, Anal. Chem. 1997, 69, 3346–3352; b) Y. Suzuki, M. Ojika,
Y. Sakagami, K. Kaida, R. Fudou, T. Kameyama, J. Antibiot. 2001,
54, 2 2 –2 8.
[20] S. Takano, W. Uchida, S. Hatakeyama, K. Ogasawara, Chem. Lett.
1982, 733–736.
[21] For small scale runs (0.5–1.5 mmol) yields were ca. 10% higher.
[22] See Supporting Information for details.
[23] H. Heimgartner, H. Schmid, H. J. Hansen, Helv. Chim. Acta 1972,
55, 1385–1401.
Feeding experiment with (2S,4R)-2’,4-[D2]5: It was carried out according
to GP 2with the mixture of (2 S,4R)-2’,4-[D2]5 and NaCl (430 mg, max.
240 mg of (2S,4R)-2’,4-[D2]5, 1.53 mmol) to give the deuterated hormao-
mycin 1b (10.5 mg). 1H NMR (600 MHz, CDCl3) (see Figure 2): de-
creased: 5.65 [0.25H, 2’-H, (4-Pe)Pro], 3.29 [0.15H, 4-H, (4-Pe)Pro]; MS
(ESI): positive mode: m/z (%): 1154 [M ++Na]; negative mode: m/z
(%): 1130 [MꢀH+] (undeuterated hormaomycin 1a: MS (ESI): positive
mode: m/z (%): 1152[
[MꢀH+]).
M
++Na]; negative mode: m/z (%): 1128
Feeding experiment with (2S)-3,3-[D2]15: It was carried out according to
GP 2with a mixture of (2 S)-3,3-[D2]15 and LiCl (396 mg, max. 197 mg of
(2S)-3,3-[D2]15, 1.11 mmol) to give the deuterated hormaomycin 1c
(8.1 mg). 1H NMR (500 MHz, CDCl3) (see Figure 2): decreased: ꢀ0.07,
0.48 [2H, 3-H, (3-Ncp)Ala I], 1.60, 1.81 [2H, 3-H, (3-Ncp)Ala II]; D
NMR (76.7 MHz, CDCl3): 0.34, 1.69 (1:1); MS (ESI): positive mode: m/z
(%): 1156 [M ++Na]; negative mode: m/z (%): 1132[ M +ꢀH].
Feeding experiment with (2S)-2,2’-[D2]15: It was carried out according to
GP 2with (2 S)-2,2’-[D2]15 (425 mg, 2.41 mmol) to give the deuterated hor-
maomycin 1d (10.0 mg). 1H NMR (600 MHz, CDCl3) (see Figure 2): de-
creased: 5.14 [0.5H, 2-H, (3-Ncp)Ala II], 4.05 [0.5H, 2’-H, (3-Ncp)Ala II],
2.94 [0.4H, 2’-H, (3-Ncp)Ala I]; MS (ESI): positive mode: m/z (%): 1155/
1154/1153 [M ++Na]; negative mode: m/z (%): 1131/1130/1129 [M +ꢀH].
[24] T. W. Green, P. G. M. Wuts, Protecting Groups in Organic Synthesis,
3rd ed., Wiley, New York, 1999.
[25] The nondeuterated (2S,4R)-12 was also prepared; see Supporting In-
formation for details.
[26] H. Josien, A. Martin, G. Chassaing, Tetrahedron Lett. 1991, 32,
6547–6550.
[27] M. Brandl, S. I. Kozhushkov, K. Loscha, O. V. Kokoreva, D. S. Yufit,
J. A. K. Howard, A. de Meijere, Synlett 2000, 1741–1744.
[28] The mixture of 13 and LiCl obtained after the hydrolysis was direct-
ly used for the feeding experiment. No attempts for its separation
were made.
[29] Y. N. Belokonꢂ, V. I. Tararov, V. I. Maleev, T. F. Savelꢂeva, M. G.
Ryzhov, Tetrahedron: Asymmetry 1998, 9, 4249–4252.
[30] O. V. Larionov, T. F. Savelꢂeva, K. A. Kochetkov, N. S. Ikonnikov,
S. I. Kozhushkov, D. S. Yufit, J. A. K. Horward, V. N. Khrustalev,
Y. N. Belokon, A. de Meijere, Eur. J. Org. Chem. 2003, 869–877.
[31] Y. Elemes, U. Ragnarsson, J. Chem. Soc. Perkin Trans. 1 1996, 537–
540. To the best of our knowledge, the preparation of the deuterated
17 has not been published before.
Acknowledgments
This work was supported by the Deutsche Forschnungsgemeinschaft
(SFB-416, Projects A3 and B5) and the Fonds der Chemischen Industrie.
The authors are indebted to Dr. H. Frauendorf (Göttingen) for perform-
ing the HPLC/MS experiments, to Mrs. M. Klingebiel and Mr. H.-P.
Kroll (Göttingen) for excellent technical assistance, to Mr. O. V. Lario-
nov (Göttingen) for a generous gift of nitrocyclopropane and to Dr. B.
Knieriem (Göttingen) for his careful proofreading of the final manu-
script.
[1] a) N. Andres, H. Wolf, H. Zähner, E. Rössner, A. Zeeck, W. A.
König, V. Sinnwell, Helv. Chim. Acta 1989, 72, 426–437; b) E. Röss-
ner, A. Zeeck, W. A. König, Angew. Chem. 1990, 102, 84–85;
Angew. Chem. Int. Ed. Engl. 1990, 29, 64–65.
[2] a) H. Zähner, N. Andres, A. Zeeck, E. Rössner, H. Wolf, W. A.
König, V. Sinnwell, A. Fredenhagen, Eur. Pat. Appl. 1990
EP 385936 AI 19900905; b) N. Andres, H. Wolf, H. Zähner, Z. Na-
turforsch. 1990, 45c, 851–855; c) H. Wolf, N. Andres, DECHEMA
Monogr. 1993, 129, 53–61.
[32] When non-deuterated acetonitrile was used in this reaction, a loss of
ca. 50% of the deuterium label occurred.
[33] Y. Kai, P. Knochel, S. Kwiatkowski, J. D. Dunitz, J. F. M. Oth, D.
Seebach, H.-O. Kalinowski, Helv. Chim. Acta 1982, 65, 137–161.
These authors, upon deprotonation of nitrocyclopropane as well as
2-methylnitrocyclopropane with different bases and subsequent
quenching of the resulting nitronates with a variety of electrophiles,
including D2O, obtained only products of their dimerization and/or
oxidative dimerization.
[3] a) J. Zindel, A. de Meijere, J. Org. Chem. 1995, 60, 2968–2973; b) J.
Zindel, A. Zeeck, W. A. König, A. de Meijere, Tetrahedron Lett.
1993, 34, 1917–1920.
[34] An analogous experiment with (2S,2’S,1’’R,2’’S)-18 was unsuccessful
because of solubility problems.
1
[35] For the full assignment of the H- and 13C- NMR spectra of 1a, see:
[4] B. Zlatopolskiy, A. de Meijere, Chem. Eur. J. 2004, 10, in press; see
also: B. Zlatopolskiy, Dissertation, Universität Göttingen (Germa-
ny), 2003.
P. Henne, Dissertation, Universität Göttingen, 1994; B. Geers, Dis-
sertation, Universität Göttingen (Germany), 1998.
[5] a) For the synthesis of a mixture of all possible isomers of 4-propen-
yl- and allylprolines see: K. Osugi, Yakugaku Zasshi 1958, 78, 1338–
1342; b) For the synthesis of the epimeric mixture of N-Boc-protect-
Received: April 26, 2004
Published online: August 20, 2004
Chem. Eur. J. 2004, 10, 4708 – 4717
www.chemeurj.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4717