Biosynthetic precursors of the lipase inhibitor lipstatin

Article abstract of DOI:10.1021/jo016285v

Three putative intermediates in the biosynthesis of the lipase inhibitor lipstatin were synthesized in stable isotope-labeled form and were added to fermentation cultures of Streptomyces toxytricini. Biosynthetic lipstatin was isolated and analyzed by NMR spectroscopy. [3,10,11,12-²H]-(3S,5Z,8Z)-3-hydroxytetradeca-5,8-dienoic acid (9) was shown to serve as a direct biosynthetic precursor of lipstatin. [7,8^-2H₂]Hexylmalonate (11) was also incorporated into lipstatin, albeit at a relatively low rate. The leucine moiety of [¹³C-formyl,¹⁵N]-N-formylleucine (10) was diverted to lipstatin under loss of the ¹³C-labeled formyl residue.

Full text of DOI:10.1021/jo016285v

J . Org. Chem. 2002, 67, 2257-2262
2257
Biosyn th etic P r ecu r sor s of th e Lip a se In h ibitor Lip sta tin
Christoph A. Schuhr,† Wolfgang Eisenreich,† Markus Goese,† Peter Stohler,‡
Wolfgang Weber,‡ Ernst Kupfer,‡ and Adelbert Bacher*,†
Lehrstuhl fu¨r Organische Chemie und Biochemie, Technische Universita¨t Mu¨nchen,
Lichtenbergstr. 4, D-85747 Garching, Federal Republic of Germany, and F. Hoffmann-LaRoche AG,
Pharma Research Basel, Discovery Technologies, CH-4070 Basel, Switzerland
adelbert.bacher@ch.tum.de
Received November 15, 2001
Three putative intermediates in the biosynthesis of the lipase inhibitor lipstatin were synthesized
in stable isotope-labeled form and were added to fermentation cultures of Streptomyces toxytricini.
Biosynthetic lipstatin was isolated and analyzed by NMR spectroscopy. [3,10,11,12-2H]-(3S,5Z,8Z)-
3-hydroxytetradeca-5,8-dienoic acid (9) was shown to serve as a direct biosynthetic precursor of
lipstatin. [7,8-2H2]Hexylmalonate (11) was also incorporated into lipstatin, albeit at a relatively
low rate. The leucine moiety of [13C-formyl,15N]-N-formylleucine (10) was diverted to lipstatin under
loss of the 13C-labeled formyl residue.
Ch a r t 1
In tr od u ction
The lipophilic â-lactone lipstatin is a product of Strep-
tomyces toxytricini.1,2 The compound irreversibly inhibits
pancreatic lipase by covalent modification of a catalytic
serine residue.3 The tetrahydro derivative of lipstatin
(Orlistat, Xenical) is used for the management of severe
obesity (Chart 1).
We have shown previously that the â-lactone moiety
of lipstatin is biosynthesized by condensation of two fatty
acid residues with chain lengths of 8 and 14 carbon
atoms, respectively, which have been proposed to be
obtained by incomplete â oxidation of long-chain fatty
acids.4,5 Former studies had also shown that [10,11,12-
2H]-(5Z,8Z)-tetradeca-5,8-dienoic acid is incorporated into
lipstatin by growing cultures of S. toxytricini.6 3-Hydroxy-
tetradeca-5,8-dienoic acid or a derivative thereof had been
proposed as a further downstream intermediate in the
biosynthetic pathway of lipstatin.6
To accurately determine the committed precursors, we
synthesized isotope-labeled (3S,5Z,8Z)-3-hydroxytetradeca-
5,8-dienoic acid (9), N-formylleucine (10), and hexyl
malonate (11). The compounds were proffered to cultures
of S. toxytricini in separate experiments. Labeling pat-
terns of isolated lipstatin were determined by NMR
spectroscopy.
Resu lts
[3,10,11,12-2H]-(3S,5Z,8Z)-3-Hydroxytetradeca-5,8-di-
enoic acid (9) was prepared as described in the Experi-
mental Section in an overall yield of 11%. Briefly, the
dipropyl ester of 3-ketoglutarate was converted into
[3-2H]-3-hydroxyglutaric acid dipropyl ester (2) by cata-
lytic deuteration. [3-2H]-(3S)-3-Hydroxyglutaric acid mono-
propyl ester (3) was obtained by reaction of 2 with pig
liver esterase. Compound 3 was converted into the 3-tert-
butyldiphenylsiloxy derivative 4. Subsequently, 4 was
converted via 5 to [3-2H]-(3S)-5-oxo-3-tert-butyldiphenyl-
siloxypentanoic acid propyl ester (6). Wittig reaction of
6 and [5,6,7-2H]-(3Z)-triphenylnon-3-enylphosphonium
bromide (7) afforded the tert-butyldiphenylsiloxy deriva-
tive 8, which was finally deprotected to yield 9 (Scheme
1).
* To whom correspondence should be addressed. Tel: +49-89-289-
13360. Fax: +49-89-289-13363.
† Technische Universita¨t Mu¨nchen.
‡ F. Hoffmann-LaRoche AG.
(1) Weibel, E. K.; Hadvary, P.; Hochuli, E.; Kupfer, E.; Lengsfeld,
H. J . Antibiot. 1987, 40, 1081-1085.
(2) Hochuli, E.; Kupfer, E.; Maurer, R.; Meister, W.; Mercadal, Y.;
Schmidt, K. J . Antibiot. 1987, 40, 1086-1091.
(3) Lu¨thi-Peng, Q.; Ma¨rki, H.-P.; Hadvary, P. FEBS Lett. 1992, 299,
111-115.
(4) Eisenreich, W.; Kupfer, E.; Weber, W.; Bacher, A. J . Biol. Chem.
1997, 272, 867-874.
(5) Goese, M.; Eisenreich, W.; Kupfer, E.; Weber, W.; Bacher, A. J .
Biol. Chem. 2000, 275, 21192-21196.
(6) Goese, M.; Eisenreich, W.; Kupfer, E.; Stohler, P.; Weber, W.;
Leuenberger, H. G.; Bacher, A. J . Org. Chem. 2001, 66, 4673-4678.
10.1021/jo016285v CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/12/2002

2258 J . Org. Chem., Vol. 67, No. 7, 2002
Schuhr et al.
Sch em e 1
the known 1H NMR signal assignments of lipstatin4
2
(Figure 2A), the H NMR signals at 5.0 and 2.0 ppm can
be unequivocally attributed to H-5 and H-12, respectively.
The approximate deuterium abundance of position 5 was
1.0% as determined by NMR spectroscopy using the
signal of the unlabeled chloroform solvent as reference.
Intact incorporation of 9 into lipstatin should result
in deuterium enrichments of positions 5, 12, and 13/14.
The observed integral ratio of these signals was 1.0/0.7/
3.2, whereas the corresponding ratio of the 2H NMR
signals of positions 3, 10, and 11/12 of compound 9 was
1.0/0.7/1.8 (Table 1). Therefore, it appears likely that the
2
enhanced intensity of the H signals of lipstatin at 1.2/
2
1.3 ppm reflects not only H label at positions 13/14 but
additionally at positions 2′, 3′, and 4′.
F igu r e 1. 2H NMR spectrum of [3,10,11,12-2H](3S,5Z,8Z)-3-
hydroxytetradeca-5,8-dienoic acid (9).
Earlier studies had shown that octanoate can serve as
the C8 building block in lipstatin biosynthesis.4 In analogy
with other biosynthetic Claisen condensations,7 it ap-
pears possible that octanoate is activated by carboxyl-
ation to the assumed precursor hexyl malonate. To test
this hypothesis, we prepared [7,8-2H2]hexyl malonate
(11), which was applied to a culture of S. toxytricini with
a large excess of linoleic acid.
Ta ble 1. 2H NMR Da ta of
[3,10,11,12-2H](3S,5Z,8Z)-3-Hyd r oxytetr a d eca -5,8-d ien oic
Acid (9) a n d Lip sta tin fr om th e F eed in g Exp er im en t w ith
[3,10,11,12-2H]-9
compd
position δ 2H (ppm) integral
[3,10,11,12-2H]-(3S,5Z,8Z)-3-
hydroxytetradeca-5,8-dienoic
acid (9)
3
4.1
1.0
The 2H NMR spectrum of isolated lipstatin showed two
signals with equal intensities at 1.2 and 0.8 ppm reflect-
10
11/12
5
12
13/14
2′, 3′, 4′
2.0
1.3/1.2
5.0
2.0
1.3/1.2
0.7
1.8
1.0
0.7
3.2
2
ing H enrichments at positions 5′ and 6′ (Figure 2C).
lipstatin
The approximate deuterium abundance of positions 5′
and 6′ was 0.2% as determined by NMR using the solvent
signal as internal reference.
Finally, we studied the origin of the formyl leucine
moiety since it was still unknown whether the formyl
residue is attached to leucine before or after the ami-
noacylation of the lipophilic C22 precursor of lipstatin.
[13C-formyl,15N]-N-formylleucine (10) was synthesized as
described in the Experimental Section and was proffered
to a fermentation culture of S. toxytricini together with
unlabeled (S)-leucine at a ratio of about 1:9. Lipstatin
was isolated and analyzed by 1H NMR spectroscopy. The
amide 1H signal showed satellites which could be un-
The 2H NMR spectrum of 9 (Figure 1) showed four
2
signals. Relative intensities of the H signals are sum-
1
marized in Table 1. On the basis of the H NMR signal
assignments of 9 (see the Experimental Section), the
2
observed H NMR signals were attributed to positions 3
and 10-12. The relative intensities of the signals of
positions 10-12 reflected the isotopomer composition of
the precursor 7.6
[3,10,11,12-2H]-9 was dissolved in a 5-fold excess of
linoleic acid and was proffered to a fermentation culture
of S. toxytricini over a period of 30 min. Lipstatin was
1
equivocally attributed to H15N coupling by 15N decou-
pling experiments (Figure 3). Their intensities correspond
1
2
isolated and was analyzed by H and H NMR spectro-
2
scopy. H NMR signals were observed at chemical shifts
(7) Weissman, K. J .; Timoney, M.; Bycroft, M.; Grice, P.; Hanefeld,
of 5.0, 2.0, 1.3, and 1.2 ppm (Figure 2B). On the basis of
U.; Staunton, J .; Leadlay, P. L. Biochemistry 1997, 36, 13849-13855.

Biosynthetic Precursors of the Lipase Inhibitor Lipstatin
J . Org. Chem., Vol. 67, No. 7, 2002 2259
2
F igu r e 2. 1H NMR spectrum of lipstatin isolated after feeding of 9 (A), the corresponding H NMR spectrum (B), and 2H NMR
spectrum of lipstatin isolated after feeding of 11 (C).
hydrogen due to its coupling with the 13C formyl carbon
atom would have been expected (J ≈ 4 Hz).
The CI mass spectrum signal of [M - CO2 + H+]+ (m/z
448) showed an isotope peak at m/z 449 that was about
2 percentage points more intensive than the correspond-
ing signal of an unlabeled reference, whereas the signal
at m/z 450 was not enhanced (data not shown). These
data show that the 15N label but no 13C label was
transferred to lipstatin.
Discu ssion
The transfer of deuterium from the 3 position of
proffered 9 shows that the â-carbon atom of 9 did not
pass through an oxidation/reduction cycle, which would
have resulted in the loss of the label. In conjunction with
the data from our earlier incorporation studies,4-6 we can
F igu r e 3. 1H NMR signal of the amide proton of lipstatin
isolated after feeding of 10. Satellites in A are due to coupling
with 15N (1J HN ) 91 Hz). The integrals of the satellites
correspond to 2.6% 15N abundance, which is significantly above
normal levels of 0.37%. The further splitting (J ) 8.6 Hz),
which can also be observed for the central signal if not
truncated, is the result of 1H1H coupling. B was obtained under
1H{15N}-decoupling conditions.
rule out a structural modification of the hydroxydienyl
side chain of 9 during the biochemical transformation
into lipstatin.
2
The enhancement of the H signal intensity at 1.2/1.3
ppm was also observed after feeding of the dehydroxy
derivative of 9.6 It therefore appears likely that an
incorporation of 9 into the hexyl side chain of lipstatin
takes place, after its conversion to the thioester of
octanoic acid and hexyl malonate.
From the data, it can be concluded that about one-third
of the overall amount of precursor incorporated into
lipstatin is incorporated via the hexyl side chain.
The relatively low incorporation of 11 leaves room for
several different interpretations. Hexyl malonate could
be the actual biosynthetic precursor but is less efficiently
to a 15N abundance of 2.6%, which is significantly above
the natural abundance level of 0.37%.
1
The intensity of 13C satellites of the formyl H signal
was determined as 1.1%, reflecting the natural abun-
dance of 13C. In contrast, if 10 had been incorporated
intact, one would expect an intensity of 3.3%. Addition-
ally, a further splitting of the signal of the formamide

2260 J . Org. Chem., Vol. 67, No. 7, 2002
Schuhr et al.
to a pH of 2 by the addition of 1 M hydrochloric acid and
extracted with ethyl acetate. Evaporation of the solvent and
converted to the required thioester form as compared to
octanoate. Alternatively, the proffered hexyl malonate
could be enzymatically decarboxylated prior to utilization
for the biosynthesis of lipstatin. The incorporation data
are not sufficient to conclusively prove the proposed
hypothesis of R-carboxylation of octanoate.
distillation afforded 73 mmol of 3 (13.8 g, 94%, ee 87%, [R]25
D
+1.88° (c ) 0.1 g/mL in acetone)).
3: 1H NMR δ (ppm) 0.9 (t, J ) 7 Hz, 3H, 3′), 1.5-1.6 (m,
2H, 2′), 2.5/2.6 (2s, 4H, 2/4), 4.0 (t, J ) 7 Hz, 2H, 1′), 4.4 (m,
0.25H, 3); 13C NMR δ (ppm) 10.2 (3′), 21.8 (2′), 40.4/40.5 (2/4),
64.6 (3), 66.5 (1′), 171.9 (1), 176.1 (5).
We found that only 15N label but no 13C label is
incorporated into lipstatin from compound 10. Obviously,
incorporation can only take place after hydrolysis of the
formamide motif. The data also show that the assumed
formate moiety liberated by hydrolysis of the proffered
15N13C-labeled formyl leucine is not incorporated into
lipstatin to a significant extent.
All reported data are consistent with the biosynthetic
scheme shown in Figure 4. A fraction of the thioester of
3-hydroxytetradeca-5,8-dienoic acid is obviously con-
verted to hexyl malonic thioester via an octanoic acid
derivative. This hexyl malonic thioester undergoes a
condensation in conjunction with decarboxylation afford-
ing a C22-moiety with the main fraction of the hydroxy-
acyl thioester. Reduction of the keto group, ring closure
to the â-lactone, and introduction of the leucine residue
followed by its formylation afford lipstatin.
[3-2H ]-(3S)-3-ter t-Bu t yld ip h en ylsiloxyglu t a r ic Acid
Mon op r op yl Ester (4). tert-Butyldiphenylsilyl chloride (17.3
g, 62.9 mmol) was added dropwise to a solution containing 5
g (26.3 mmol) of 3 and 6.44 g (94.6 mmol) of imidazole in 50
mL of dry CH2Cl2 at 3 °C under a nitrogen atmosphere. The
solution was incubated at room temperature with shaking
overnight. An aqueous solution of 5% sodium carbonate (160
mL) was added. After several days, CH2Cl2 was evaporated,
and the remaining aqueous phase was extracted with diethyl
ether. It was adjusted to pH 2 and extracted again. The organic
phase was evaporated, affording a slightly yellow syrup (9.1
g, 21.3 mmol, 81%).
4: 1H NMR δ (ppm) 0.9 (t, J ) 7 Hz, 3H, 3′), 1.0 (s, 9H,
tBu-Me), 1.5-1.6 (m, 2H, 2′), 2.5-2.7 (br, 4H, 2/4), 3.9 (m, 2H,
1′), 4.5 (m, 0.25H, 3), 7.3-7.7 (m, 10H, Phe); 13C NMR δ (ppm)
10.4 (3′), 19.2 (tBu-C), 21.8 (2′), 26.8 (tBu-Me), 41.4/41.6 (2/4),
66.1 (1′), 67.0 (3), 127.7/129.8/133.2/135.9 (Phe), 170.8 (1), 176.9
(5).
[3-2H]-(3S)-5-Hyd r oxy-3-ter t-bu tyld ip h en ylsiloxyp en -
ta n oic Acid P r op yl Ester (5). A solution of borane-THF
complex in THF (11 mL, 1 M) was added to a solution of 4
(3.62 g, 8.4 mmol) in 13 mL of dry THF at -18 °C and under
an inert atmosphere. The mixture was allowed to warm to 0
°C. After 7 h, one volume of ice-water and sodium carbonate
solution were added slowly. The mixture was extracted with
diethyl ether. The organic phase was dried over sodium sulfate.
The solvent was evaporated under reduced pressure. The
residue was applied to a column of silica gel 60 (48 cm × 3
cm) that was developed with hexane/ethyl acetate (2:1) yielding
3.1 g (7.4 mmol, 88%) of 5.
5: 1H NMR δ (ppm) 0.9 (t, J ) 7 Hz, 3H, 3′), 1.0 (s, 9H,
tBu-Me), 1.5-1.6 (m, 2H, 2′), 1.7-1.8 (m, 2H, 4), 2.5 (s, 2H,
2), 3.6 (m, 2H, 5), 3.9 (t, J ) 7 Hz, 2H, 1′), 4.3 (m, 0.25H, 3),
7.3-7.7 (m, 10H, Phe); 13C NMR δ (ppm) 10.3 (3′), 19.2 (tBu-
C), 21.7 (2′), 26.8 (tBu-Me), 38.9 (4), 41.9 (2), 59.1 (5), 66.0/
68.6 (1′/3), 127.6/129.7/133.5/135.8 (Phe), 171.2 (1).
[3-2H]-(3S)-5-Oxo-3-ter t-b u t yld ip h en ylsiloxyp en t a n o-
ic Acid P r op yl Ester (6). A 3.65 g (8.8 mmol) portion of 5
was added to 3.75 g (17.4 mmol) of pyridinium chlorochromate
in 30 mL of dry CH2Cl2. The solution was incubated for 3 h
with shaking under a dry atmosphere. The mixture was
extracted with ether. The organic phase was concentrated
under reduced pressure and applied to a column of silica gel
(43 cm × 3 cm), which was developed with hexane/ethyl acetate
(2:1), yielding 2.34 g (5.7 mmol, 64%) of 6.
6: 1H NMR δ (ppm) 0.9 (t, J ) 7 Hz, 3H, 3′), 1.0 (s, 9H,
tBu-Me), 1.5-1.7 (q, J ) 7 Hz, 2H, 2′), 2.5-2.8 (br, 4H, 2/4),
3.9 (t, J ) 7 Hz, 2H, 1′), 4.6 (m, 0.25H, 3), 7.3-7.8 (m, 10H,
Phe), 9.6 (s, 1H, 5); 13C NMR δ (ppm) 10.3 (3′), 19.1 (tBu-C),
21.8 (2′), 26.8 (tBu-Me), 41.9 (2), 50.3 (4), 65.9/66.1 (1′/3), 127.7/
129.9/133.2/135.7 (Phe), 170.5 (1), 200.7 (5)
[3,10,11,12-2H ]-(3S,5Z,8Z)-3-ter t-Bu t yld ip h en ylsiloxy-
tetr a d eca -5,8-d ien oic Acid P r op yl Ester (8). The phos-
phonium salt of labeled bromononene (7) (4.31 g, 9.2 mmol)
was added to 9.0 mL of a 1 M solution of sodium bis-silyl amide
at -17 °C in a dry THF/toluene solution (5:1) under nitrogen.
The orange-colored solution was allowed to warm to room
temperature and was then cooled to -90 °C. A 3.1 g portion
of 6 (7.5 mmol) in 5 mL of THF was added dropwise, and the
mixture was shaken overnight with warming to room tem-
perature. A saturated aqueous solution of ammonium chloride
was added, and the mixture was extracted with diethyl ether.
The extract was concentrated under reduced pressure. The
residue was applied to a column of silica gel (37 cm × 2.5 cm),
which was developed with pentane/diethyl ether (97:3), yield-
ing pure 8 (2.42 g, 4.6 mmol, 62%).
Exp er im en ta l Section
Ma ter ia ls. Commercially available reagents were used
without further purification. Pig liver esterase (EC 3.1.1.1) was
purchased from Fluka (Buchs, Switzerland). The phosphonium
salt 7 was synthesized as published and is a complex mixture
of isotopomers.6 The reactions were done in dry solvents at
room temperature unless stated otherwise.
Syn th esis of 2H-La beled In ter m ed ia tes. (3S,5Z,8Z)3-
Hyd r oxytetr a d eca -5,8-d ien oic Acid (9). The synthesis of
isotope labeled 9 was achieved by methods modified from ref
8.
[3-2H]-3-Hydr oxyglu tar ic Acid Dipr opyl Ester (2). 3-Ox-
oglutaric acid (1) (21 g, 144 mmol) was dissolved in 50 mL of
dry propanol (40 g, 0.66 mol) containing 8.75 g (0.24 mmol) of
HCl. The mixture was incubated with shaking at 35 °C for 10
min and was then kept at room temperature overnight. The
solution was extracted with diethyl ether. The organic phase
was washed with aqueous sodium carbonate, dried over
sodium sulfate, and evaporated to dryness under reduced
pressure, yielding 30.9 g (134 mmol, 93%) of the liquid dipropyl
ester of 1.
Without further purification, the ester was dissolved in 160
mL of propanol containing 1% (v/v) of triethylamine. Catalytic
deuteration was performed with D2 at atmospheric pressure
over Raney nickel (7.2 g) as catalyst. The catalyst was removed
by filtration after 18 h, and the solution was evaporated under
reduced pressure. Vacuum distillation of the residue (bp 100-
112 °C, ∼0.1 mbar) afforded 2 (118 mmol, 27.5 g, 88%). The
deuterium abundance at position 3 was about 75%, as the
catalyst was produced in nondeuterated solvents.
2: 1H NMR δ (ppm) 0.9 (t, J ) 8 Hz, 6H, 3′/3′′), 1.6-1.7 (m,
4H, 2′/2′′), 2.6 (s, 4H, 2/4), 3.5 (s, 1H, OH), 4.1 (t, J ) 7 Hz,
4H, 1′/1′′), 4.5 (m, 0.25H, 3); 13C NMR δ (ppm) 10.2 (3′, 3′′),
21.8 (2′, 2′′), 40.6 (2, 4), 64.6 (3), 66.2 (1′, 1′′), 171.7 (1, 5).
[3-2H]-(3S)-3-Hyd r oxyglu ta r ic Acid Mon op r op yl Ester
(3). Pig liver esterase (3620 U) in 270 mL of 30 mM phosphate
buffer (pH ) 7) was added to 18 g (77.5 mmol) of 2. The
mixture was incubated with stirring at 4 °C. The pH was
continually adjusted to 7.0 by the addition of 1 M NaOH. The
reaction came to an end after the consumption of 1 equiv of
the base (44 h). The solution was brought to pH 9 and
extracted with diethyl ether. The aqueous phase was adjusted
(8) Baader, E.; Bartmann, W.; Beck, G.; Bergmann, A.; Fehlhaber,
H.-W.; J endralla, H.; Kesseler, K.; Saric, R.; Schu¨ssler, H.; Teetz, V.;
Weber, M.; Wess, G. Tetrahedron Lett. 1988, 29, 2563-2566.

Biosynthetic Precursors of the Lipase Inhibitor Lipstatin
J . Org. Chem., Vol. 67, No. 7, 2002 2261
F igu r e 4. Hypothetical mechanism of lipstatin biosynthesis. Isotope label transferred from [3,10,11,12-2H](3S,5Z,8Z)-3-hydroxy-
tetradeca-5,8-dienoic acid (9) is indicated by *. Isotope label transferred from [7,8-2H2]hexyl malonate (11) is indicated by O, and
isotope labeled transferred from [13C-formyl, 15N]-N-formylleucine (10) is indicated by 9.
8: 1H NMR δ (ppm) 0.9 (t, J ) 7 Hz, 6H, 14/3′), 1.0 (s, 9H,
tBu-Me), 1.2-1.4 (br, 4.5H, 11-13), 1.5-1.7 (br, 2H, 2′), 2.0
(br, 1.5H, 10), 2.2 (d, J ) 7 Hz, 2H, 4), 2.5 (s, 2H, 2), 2.6 (br,
2H, 7), 3.8-4.0 (m, 2H, 1′), 4.2-4.3 (m, 0.25H, 3), 5.2-5.4 (m,
4H, 5/6/8/9), 7.3-7.8 (m, 10H, Phe); 13C NMR δ (ppm) 10.3
(3′), 14.0 (14), 19.2 (tBu-C), 21.8 (2′), 22.5/29.2/31.4 (11/12/13),
25.6 (7), 26.9 (tBu-Me), 27.1 (10), 34.9 (4), 41.5 (2), 65.9 (1′),
70.3 (3), 124.5/127.3/127.5/129.6/130.4/130.8/133.8/135.9 (5/6/

2262 J . Org. Chem., Vol. 67, No. 7, 2002
Schuhr et al.
8/9/Phe), 171.5 (1); GC tret ) 28.4 min; MS main peaks for
unlabeled 8 were m/z ) 463 ([M - tBu]+) and m/z ) 199 ([Ph2-
SiOH]+). 3J coupling constants of the olefinic protons were 10.8
Hz. This is in the typical range for the (Z)-configuration and
very similar to the value observed in lipstatin (11.0 Hz).
[3,10,11,12-2H ]-(3S,5Z,8Z)-3-H yd r oxyt et r a d eca -5,8-d i-
en oic Acid (9). A solution containing 1.60 g (3.05 mmol) of 8
in 15 mL of dry THF was added to a tetrabutylammonium
fluoride solution in THF (1 M, 10.7 mL) at 0 °C. The mixture
was incubated with shaking for 2 days at room temperature
and was then added to the same volume of water. The mixture
was extracted with diethyl ether, and the solvent was evapo-
rated. The residue was applied to a column of silica gel (20
cm × 2.5 cm), which was developed with pentane/diethyl ether
(18:5), yielding 0.48 g (1.68 mmol, 55%) of the propyl ester of
9.
h. KOH (14 M)/ethanol (1:2) was added, and the mixture was
refluxed for 2 h. The mixture was brought to a pH of 1 under
cooling and was then extracted with diethyl ether. The solution
was evaporated under reduced pressure. The resulting solid
was recrystallized from ethyl acetate affording 11 as white
crystals (5.1 g, 26.5 mmol) with an overall yield of 85%.
11: 1H NMR δ (ppm) 0.9 (m, 2H, 8), 1.2-1.3 (m, 7H, 4-7),
1.8 (m, 2H, 3), 3.2 (t, J ) 7 Hz, 1H, 2); 13C NMR δ (ppm) 14.2
(8), 23.5, 28.4, 30.0, 32.7, 53.1 (2), 173.3 (1, 1′); 2H NMR δ (ppm)
0.9 (s, 1H, 8), 1.3 (s, 1H, 7).
F er m en ta tion a n d F eed in g Exp er im en ts. S. toxytricini
was grown in 8 L of medium in a 14 L Chemap fermentor
(Ma¨nnedorf, Switzerland) according to published procedures.9
In experiment A, 0.2 g of 9 diluted in 1.1 g linoleic acid was
added in seven aliquots at intervals of 5 min.
In experiment B, 1.1 g of 10 in 20 mL of water was added
continuously within 20 h.
In experiment C, 1.62 g of 11 in 100 mL of NaOH (2%) was
added continuously within 10 h.
Fermentation was stopped by heating to 70 °C for 10 min.
Isola tion a n d P u r ifica tion of Lip sta tin . Lipstatin was
isolated and purified as described previously.4
The propyl ester was dissolved in 5 mL of 10 M KOH/ethanol
(1:4) and refluxed at 95 °C. A 15 mL portion of water was
added, and the solution was extracted with diethyl ether. HCl
(2 N) was added to adjust the pH to 2. The aqueous solution
was again extracted with diethyl ether. The organic phase was
concentrated under reduced pressure and afforded 9 (0.36 g,
1.48 mmol, 88%) as a clear oil.
An a lytica l Meth od s. NMR spectra were obtained using
Bruker AVANCE DRX 500 or AM 360 spectrometers. 1H NMR
spectra were measured without decoupling or with 15N-
9: 1H NMR δ (ppm) 0.9 (t, J ) 7 Hz, 3H, 14), 1.2-1.4 (m,
4.5H, 11-13), 2.0 (t, J ) 7 Hz, 1.5H, 10), 2.2-2.4 (m, 2H, 4),
2.4-2.6 (m, 2H, 2), 2.8 (t, J ) 7 Hz, 2H, 7), 4.1 (m, 0.25H, 3),
5.3-5.6 (m, 4H, 5/6/8/9); 13C NMR δ (ppm) 14.1 (14), 22.6/29.3/
31.5 (11/12/13), 25.8 (7), 27.3 (10), 34.3 (4), 40.5 (2), 67.9 (3),
1
decoupling, 13C NMR spectra with H-decoupling, and 2H NMR
spectra with 1H-decoupling. Chemical shifts were referenced
to internal tetramethylsilane.
2
124.2/127.1/130.8/131.9 (5/6/8/9), 177.8 (1); H NMR δ (ppm)
Optical rotation was measured using a Perkin-Elmer 241
MC polarimeter.
1.25/1.32 (2s, 1.40H, 11/12), 2.0 (s, 0.47H, 10), 4.1 (s, 0.75H,
3) (Figure 1); [R]25 ) +1.9° (c ) 0.084 g/mL, acetone).
D
A GC/MS system consisting of a GC 17A with a MS QP 5000
from Shimadzu was used to characterize synthetic products.
A DB5 column (length: 30 m) was used with a helium flow of
0.8 mL/min. Samples were injected at a temperature of 180
°C. After 3 min, the column was heated to 290 °C at a rate of
5 °C/min.
Chemical ionization mass spectra were measured with
ammonia as reactand gas and a ionizing energy of 230 eV.
TLC was performed on precoated silica gel 60 F254 sheets
from Merck. Column chromatography was done with silica gel
60 from Fluka.
Syn th esis of [13C-F or m yl,15N]-N-F or m ylleu cin e (10).
Racemic [15N]leucine (2.0 g, 15.1 mmol) was added to 20 g (0.43
mol, 15.1 mL) of [13C]formic acid. The mixture was cooled with
ice, and 5.3 mL of acetic anhydride was added dropwise. After
incubation overnight, 7 mL of water was added dropwise. The
solution was concentrated under reduced pressure. The pre-
cipitate was purified by incubation in water with shaking
overnight affording 1.83 g (11.4 mmol, 75%) of pure 10.
10: 1H{13C,15N} NMR δ (ppm) 1.0 (d, J ) 6 Hz, 6H, 5/5′),
1.6 (m, 1H, 4), 1.8 (m, 2H, 3), 4.8 (m, 1H, 2), 5.9 (d, J ) 8 Hz,
1H, NH), 8.2 (s, 1H, CHO); 13C{15N} NMR δ (ppm) 20.6 (5′),
22.3 (5), 24.4 (4), 40.6 (3), 52.6 (2), 163.6 (formyl), 179.5 (1)
Syn th esis of [7,8-2H2]-Hexylm a lon a te (11). 6-Bromohex-
1-ene (37.4 mmol, 5 mL) was added to a solution of diethyl
malonate (37.4 mmol, 5.68 mL) and 37.4 mmol of NaOEt in
20 mL of dry ethanol. The mixture was refluxed for 16 h and
was then extracted with diethyl ether. The solvent was
evaporated to yield crude 7-hexenylmalonic acid diethyl ester
(8.4 g, 34.7 mmol), which was deuterated under atmospheric
pressure with 0.6 g of Rh/C as catalyst in methanol within 3
Ack n ow led gm en t. This work was supported by the
Deutsche Forschungsgemeinschaft and the Fonds der
Chemischen Industrie.
J O016285V
(9) Bacher, A.; Stohler, P.; Weber, W. European Patent Application
EP 803 576, 29 Oct 1997.