Originally published In Press as doi:10.1074/jbc.M003094200 on May 4, 2000
J. Biol. Chem., Vol. 275, Issue 28, 21192-21196, July 14, 2000
Biosynthetic Origin of Hydrogen Atoms in the Lipase Inhibitor
Lipstatin*
Markus
Goese
,
Wolfgang
Eisenreich,
Ernst
Kupfer§,
Wolfgang
Weber§, and
Adelbert
Bacher¶
From the Lehrstuhl für Organische Chemie und
Biochemie, Technische Universität München,
Lichtenbergstrasse 4, D-85747 Garching, Germany and the
§ Department of Biotechnology, Hoffmann-La Roche AG, Pharma
Preclinical Research, CH-4070 Basel, Switzerland
Received for publication, April 12, 2000
 |
ABSTRACT |
The lipase inhibitor lipstatin is biosynthesized
in Streptomyces toxytricini via condensation of a
C14 precursor and a C8 precursor, which are
both obtained from fatty acid catabolism. To study the mechanism of
this reaction in more detail, S. toxytricini was grown in
medium containing a mixture of
U-13C,U-2H-lipids and unlabeled sunflower oil
or in a medium containing 70% D2O. Lipstatin was isolated
and analyzed by 1H,2H, and 13C NMR
spectroscopy. Hydrogen atoms at C-2, C-3, and C-4 of lipstatin were
found to be derived from solvent protons. The formation of the
lipstatin precursor 3-hydroxy-
5,8-tetradecadienoyl-CoA
by 
oxidation of linoleic acid explains the incorporation of solvent
hydrogen into the 4 position of lipstatin. The hydrogen in position 3 of lipstatin is most probably introduced from solvent by
proton/deuterium exchange of a redox cofactor involved in the reduction
of the keto group in the branched chain keto acid arising by a
decarboxylative condensation. The incorporation of solvent hydrogen at
position 2 can be explained by epimerization of a chiral intermediate
at C-2 and C-3. Epimerization may involve a dehydration-rehydration mechanism.
| |
INTRODUCTION |
Lipstatin is an inhibitor of pancreatic lipase that is produced by
Streptomyces toxytricini (1, 2). The tetrahydro derivative of lipstatin (Orlistat, Xenical®) is used for treatment of
severe obesity. The lipophilic -lactone irreversibly inactivates
lipase by covalent modification of the serine residue of its catalytic
triad (3).
Earlier in vivo incorporation studies using universally
13C-labeled lipids indicated that the -lactone moiety of
lipstatin is biosynthesized from a C14 and a C8
moiety, which are both obtained by partial catabolism of fatty acids
(Fig. 1) (4). To analyze this biosynthetic transformation in more
detail, we decided to study the fate of precursor hydrogen atoms by
in vivo stable isotope incorporation experiments.
| |
EXPERIMENTAL PROCEDURES |
Materials--
A U-13C,U-2H-lipid
mixture was obtained by acetone extraction of algal biomass
(Scenedesmus obliquus) grown with
13CO2 in D2O (H. Oschkinat, Freie
Universität Berlin, Germany). The acetone solution was evaporated
to dryness. The residue was used without purification.
Fermentation--
Fermentation experiments with shaking cultures
of S. toxytricini were conducted as described earlier (4).
In the first experiment, the medium (50 ml) was supplemented with
2.7 g of sunflower oil and 0.5 g of
U-13C,U-2H-lipid from S. obliquus.
In the second experiment, the culture medium contained a mixture of
D2O and H2O (7:3, v/v).
Lipstatin, isolated as described earlier (4), was 93% pure as judged
by high performance liquid chromatography.
NMR Spectroscopy--
NMR measurements were performed in
CDCl3 at 17 °C using a four-channel Bruker DRX 500 spectrometer operating at 500.13 MHz for 1H experiments,
125.76 MHz for 13C experiments, and 76.77 MHz for
2H NMR experiments. The spectrometer was equipped with a
lock switch unit for 2H decoupling experiments using the
lock channel. One-dimensional 1H and 13C NMR
experiments and two-dimensional
13C1H-correlated spectroscopy experiments
(CH-COSY) were performed with standard Bruker software (XWINNMR).
Simultaneous 1H and 2H decoupling of
13C was achieved with a WALTZ 16-pulse sequence during
relaxation (1H) and acquisition
(1H,2H). Prior to Fourier transformation, the
free induction decay was multiplied with a Gaussian function.
1H and 13C signal assignments of lipstatin have
been reported earlier (2, 4).
| |
RESULTS |
The -lactone moiety of lipstatin is biosynthesized from two
long-chain fatty acids (Fig. 1) (4). To
map the origin of individual hydrogen atoms in the biosynthetic
reaction sequence, S. toxytricini was grown in a medium
containing U-13C,U-2H-lipids (obtained from an
algal culture grown with 13CO2 in
D2O) and sunflower oil at a ratio of 1:5.4 (w/w). A second, complementary fermentation experiment was performed with unlabeled sunflower oil in a medium containing 70% D2O.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1.
Retrobiosynthetic illustration of lipstatin
biosynthesis (4). 1, lipstatin;
2, 3-hydroxy-5,8-tetradecadienoyl-CoA;
3, octanoyl-CoA; 4, linolyl-CoA.
|
|
1H and 2H NMR spectra of lipstatin isolated
from these cultures are shown in Fig. 2.
In the lipstatin sample obtained from the experiment with
double-labeled U-13C,U-2H-lipid, no deuterium
was observed by 2H NMR spectroscopy at position 2 of the
-lactone moiety, and little, if any deuterium was observed at
position 3 (Fig. 2D). Of the two diastereotopic hydrogen
atoms at C-4, the one resonating at lower field (H-4) was apparently
devoid of deuterium label. Moreover, deuterium was absent at the 
position of the leucine moiety (H-2'') and in the formyl moiety.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 2.
Parts of the 1H and
2H NMR spectra of lipstatin isolated from S. toxytricini. A, 1H NMR of
natural abundance lipstatin. B and C,
1H, respectively. 2H NMR of lipstatin obtained
from a culture grown in 70% D2O. D,
13C-decoupled 2H NMR of lipstatin obtained from
a culture supplied with a mixture of
U-13C,U-2H-algal lipid and unlabeled sunflower
oil (1:5.4, w/w).
|
|
The distribution of deuterium observed in the experiment with unlabeled
sunflower oil in D2O was complementary to that observed in
the previous experiment. Specifically, deuterium was detected in
positions 2, 3, and 4 (downfield shifted signal) of the -lactone moiety (Fig. 2C). Deuterium was also detected in the position of the leucine moiety (H-2'') and in the formyl moiety. The
partial replacement of 1H by 2H in these
positions is also reflected in the 1H NMR spectrum. A
comparison between the 1H NMR spectra of unlabeled
lipstatin (Fig. 2A) and the lipstatin sample from the
experiment with 70% D2O (Fig. 2B) shows that
the signal intensities of the hydrogen atoms 2, 3, and 4 (downfield shifted signal) are significantly reduced.
Additional information on the distribution of 1H and
2H in these lipstatin samples was gleaned from
1H,2H-decoupled 13C NMR spectra.
Relevant signals from the experiment with unlabeled lipid in 70%
D2O are shown in Fig. 3. The
13C signals of C-2' to C-6 are complex multiplets
comprising up to eight lines. In a spectrum obtained with
1H decoupling but without 2H decoupling, the
signal components shifted by more than 150 ppb were substantially
broadened or absent (Fig. 3, bottom).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 3.
13C NMR signals of lipstatin
isolated from S. toxytricini grown in medium
containing 70% D2O. Upper row,
1H,2H-decoupled signals; lower row,
1H-decoupled signals. Capital letters denote
isotopomers shown in Fig. 5. For C-4, -, -, and
 -2H shifts are indicated by arrows.
|
|
The signal multiplicity reflects the presence of multiple isotopomers
carrying 0 to 3 deuterium atoms at C-2, C-3, and/or C-4. Heavy isotope
shifts for substitution with one deuterium atom of carbon atoms in the
respective , , and positions were found in the ranges of
426 to 198 ppb, 120 to 52 ppb, respectively, 47 to 18 ppb.
The chemical shift increments are additive in isotopomers with multiple
deuterium substitutions. Direct evidence on the origin of the signal
multiplets in Fig. 3 was obtained by a 13C1H
COSY experiment recorded under 2H decoupling with high
resolution in the 13C dimension. As an example, the four
13C signals at high field in the complex signal pattern of
C-4 correlated with one of the H-4 1H signals (H-4* at 1.97 ppm) (Fig. 4). The four 13C
signals at lower field correlated with both signals of the
diastereotopic H-4 atoms (H-4* and H-4). On the basis of these data,
all 13C signals in Fig. 3 can be unequivocally attributed
to individual isotopomers, designated A-H (Fig.
5).
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 4.
Parts of the 1H,
2H-decoupled 13C1H-COSY spectrum
showing the C-4 signals of lipstatin isolated from S. toxytricini grown in medium containing 70%
D2O. Corresponding signals in the one-dimensional
1H NMR and in the one-dimensional
1H,2H-decoupled 13C NMR spectra are
shown as projections.
|
|
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 5.
Isotopomers of the
-lactone moiety of lipstatin isolated from S. toxytricini grown in medium containing 70%
D2O. The relative abundance of each isotopomer
estimated from 1H,2H-decoupled 13C
NMR spectra is displayed. 2H upfield shifts are indicated
by negative values in italics.
|
|
The fraction of each respective isotopomer in the biosynthetic mixture
can be determined from the signal integrals in the 1H,2H-decoupled 13C NMR spectra.
The fractional composition of the isotopomer mixture (Fig. 5) is
overdetermined by the experimental data because 2H
substitution at C-2, C-3, and C-4 is reflected by isotope shifts of
each of the carbon atoms 2, 3, and 4. However, it should be noted that
the relaxation rates of 13C atoms are substantially
modulated by directly bound 2H atoms. The respective signal
components were therefore not included in the statistical evaluation of
the label distribution.
An average deuterium content of 1.6 deuterium atoms per -lactone
moiety of lipstatin is obtained from the 13C-derived data
in Fig. 5 and Table I. Mass spectrometric
analysis of the derivatized -lactone moiety (Fig.
6) indicates an average deuterium content
of 1.8 deuterium atoms, in close agreement with the 13C NMR
data. The average 2H enrichment at the
2H-labeled positions is therefore approximately 55%. This
value is close to the D2O content of the culture
medium.
View this table:
[in this window]
[in a new window]
|
Table I
2H enrichments of lipstatin isolated from S. toxytricini grown
in medium containing 70% D2O
Calculations are made from signal integrals in the 1H NMR,
2H NMR, and 1H, 2H-decoupled 13C NMR
spectra.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 6.
Mass spectra of
2-hexyl-3,5-bis(trimethylsilyloxy)-(7Z,10Z)-hexadecadienoic
acid pyrrolidine amide (structure shown as inset)
obtained from lipstatin by pyrrolidine aminolysis followed by
methanolysis and silylation with trimethylsilylchloride
(TMS). Top, from natural
abundance lipstatin; bottom, from lipstatin isolated from
S. toxytricini grown in medium containing 70%
D2O.
|
|
The same type of analysis can be applied to the signals of the
formyl-leucine moiety of the biosynthetic lipstatin. The presence of isotopomers K through N (Fig. 7) at
relatively high abundance can be deduced from the 13C
signals shown in Fig. 8. These data
reflect the presence of 47% 2H at the -carbon of
leucine and 48% 2H in the formyl group. Minor amounts of
deuterium (13% relative 2H enrichment) were also detected
by 2H NMR analysis in the 5'' and 6'' position of the
leucine moiety.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 7.
Isotopomers of the formyl-leucine moiety of
lipstatin isolated from S. toxytricini grown in medium
containing 70% D2O including their relative
abundance. For details see Fig. 5.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 8.
13C NMR signals of lipstatin
(formyl-leucine moiety) isolated from S. toxytricini
grown in medium containing 70% D2O.
Top, 1H,2H-decoupled signals;
bottom, 1H-decoupled signals. K, L,
M, and N indicate the isotopomers shown in Fig.
7.
|
|
Because of extensive 13C13C coupling, the
13C NMR spectrum of the lipstatin sample from the
experiment with U-13C,U-2H-lipid is rather
complex. Nevertheless, these data can be interpreted using the heavy
isotope shift values described above and the
13C13C coupling constants reported earlier (4).
The signal of C-2 shows 13C13C coupling to C-1,
C-1', and C-2' (Fig. 9). By comparison
with the singlet representing the lipstatin fraction derived from the unlabeled sunflower oil, the multiplet is shifted to higher field by
223 ppb. This shift is due to the combined heavy isotope shifts of
multiple 13C and 2H neighbor atoms (two -
and two -2H upfield shifts). Signal components shifted
to even higher field by 365 ppb (-2H shift) would have
resulted if deuterium had been carried over to position 2 from the
double-labeled lipid precursor. The absence of these hypothetical
signal components (indicated by arrows below the spectrum in
Fig. 9) confirms that the lipstatin fraction biosynthesized from
double-labeled lipid is devoid of deuterium in position 2.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 9.
Labeling pattern of lipstatin from the
experiment with U-13C,U-2H-lipid (top)
and 1H, 2H-decoupled 13C NMR
signals of C-2 and C-4. Multiple 13C labeling is
indicated by bold lines. Isotope shift values (the sum of
2H and 13C shifts) are indicated by
arrows, respectively, shown in italic letters for
selected positions. 13C13C couplings are
indicated, and coupling partners are given in italic
letters. *, indicates 13C NMR signals of lipstatin
isotopomers with natural 13C abdundance. Arrows
below the spectra indicate the absent signals of hypothetical
lipstatin species (see text).
|
|
The 13C signal of C-4 of lipstatin from the experiment with
the double-labeled lipid is a pseudotriplet arising from the coupling of C-4 to two adjacent 13C atoms in position 3 and 5. The
pseudotriplet is shifted to higher field by 561 ppb (comprising
13C isotope shifts and one -, one -, and two
-2H shifts). The presence of a fractional amount of
deuterium at C-3 or of a second deuterium atom at C-4 in this sample
would have been conducive to additional signal components shifted to higher field by 130 (-2H shift) or 360 ppb
(-2H shift), respectively. Again, the absence of these
hypothetical signal components (Fig. 9) documents that no detectable
amount of deuterium was carried over into position 3 and one of the H-4 positions of lipstatin via the C14 moiety derived from the
double-labeled lipid precursor.
| |
DISCUSSION |
An analysis of lipstatin derived from a fermentation experiment
with deuterated water shows that the hydrogen atoms at positions C-2
and C-3 and one hydrogen atom at C-4 of lipstatin have been biosynthetically introduced from solvent water. The exchange with the
solvent goes to apparent equilibrium as shown by the experiment with U-13C,U-2H-lipid as precursor.
3-Hydroxy-5,8-tetradecadienoyl-CoA (compound
2, Fig. 10) obtained by oxidation of linolyl-CoA (compound 4) has been proposed to
serve as the committed precursor for the formation of the -lactone
moiety of lipstatin by Claisen condensation with an octanoyl derivative
(4). The proposed hydration of the putative intermediate 5 is conducive to the introduction of solvent hydrogen into position 2 of
the molecule.
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 10.
Hypothetical mechanism of lipstatin
biosynthesis in S. toxytricini. Protons
introduced from the solvent during fermentation are indicated by
circles.
|
|
The mechanistic and stereochemical details of the proposed condensation
are still unknown. A hypothetical mechanism among several possibilities
is shown in Fig. 10. By analogy with the stereochemistry of
erythromycin polyketide synthase (5, 6), it appears plausible that the
C8 precursor could be carboxylated yielding a
S-hexylmalonyl intermediate (compound 6). A decarboxylative condensation could then proceed by inversion of this
chiral center yielding intermediate 7 with a 2R
configuration. As in fatty acid biosynthesis (7), the reduction of the
position 3 keto group of intermediate 7 could yield
intermediate 8 with a 2R,3R
configuration. It would then be necessary to invert the chiral centers
2 and 3 of intermediate 8 to obtain the
2S,3S configuration of lipstatin. In accordance with the biochemical pathways of unsaturated fatty acids, these epimerizations could proceed by a sequence of dehydration and rehydration steps, yielding compound 10 (8, 9). The rehydration of compound 9 would be conducive to the
incorporation of solvent hydrogen at C-2 of lipstatin, which is not
easily explained otherwise.
The presence of solvent hydrogen at position 3 of lipstatin suggests
that a reducing agent involved in the generation of reducing equivalents for the reduction of the position 3 keto group of intermediate 7 in Fig. 10 is subject to hydrogen exchange with solvent water. Such an exchange process would be similar to
solvent hydrogen exchange of NADPH observed in the context of fatty
acid biosynthesis in Escherichia coli (10) and could proceed
at the level of a reduced flavocoenzyme. In line with this hypothesis,
we have found that S. toxytricini grown in
D2O-enriched culture medium forms
[5,5-2H2]proline by reduction of glutamate,
apparently via a deuterated reducing agent (data not shown).
We have also found that the -hydrogen of the leucine moiety of
lipstatin is derived from solvent hydrogen. On the other hand, only a
relatively small amount of solvent hydrogen is found in the methyl
groups of the leucine moiety. The biosynthesis of leucine from two
molecules of pyruvate and one molecule of acetyl-CoA involves the
introduction of one solvent hydrogen atom into each of the leucine
methyl groups because of the transformation of phosphoenol pyruvate to
pyruvate (11-13). Additional solvent hydrogen can be introduced into
pyruvate by spontaneous proton exchange because of the relatively high
acidity of the pyruvate methyl group. Nevertheless, the 2H
enrichments of the formyl-leucine methyl groups in lipstatin from the
labeling experiments reported in this paper were low, thus suggesting
that only a minor fraction of leucine incorporated into lipstatin has
been biosynthesized de novo, whereas the bulk of lipstatin
has been derived from unlabeled leucine present in the complex culture
medium. This proposition is well in line with earlier incorporation
experiments with 13C-labeled lipids (4). Moreover, the
labeling pattern of the leucine moiety in lipstatin suggests that the
preformed leucine in the culture medium is subject to extensive
transamination involving proton exchange of the pyridoxal phosphate
intermediate with the solvent.
The hydrogen atom of the formyl group in lipstatin has also been
subject to equilibration with solvent hydrogen. Hydrogen exchange could
have occurred at the level of methenyl tetrahydrofolate (14).1
The hypothetical biosynthetic pathway in Fig. 10 is in line with all of
the observed labeling patterns. The stereochemical features of
lipstatin biosynthesis require additional studies.
| |
ACKNOWLEDGEMENTS |
We thank G. Oesterhelt for providing the mass
spectra. The expert help of A. Werner and F. Wendling with the
preparation of the manuscript is gratefully acknowledged.
| |
FOOTNOTES |
*
This work was supported by a grant from the Deutsche
Forschungsgemeinschaft (SFB 369) and by the Fonds der Chemischen
Industrie.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
¶
To whom correspondence should be addressed. Tel.:
+49-89-289-13360; Fax: +49-89-289-13363; E-mail:
adelbert.bacher@ch.tum.de.
Published, JBC Papers in Press, May 4, 2000, DOI 10.1074/jbc.M003094200
1
D. Arigoni, personal communication.
| |
REFERENCES |
| 1.
|
Weibel, E. K.,
Hadvary, P.,
Hochuli, E.,
Kupfer, E.,
and Lengsfeld, H.
(1987)
J. Antibiot. (Tokyo)
40,
1081-1085
|
| 2.
|
Hochuli, E.,
Kupfer, E.,
Maurer, R.,
Meister, W.,
Mercadal, Y.,
and Schmidt, K.
(1987)
J. Antibiot. (Tokyo)
40,
1086-1091
|
| 3.
|
Lüthi-Peng, Q.,
Märki, H.-P.,
and Hadvary, P.
(1992)
FEBS Lett.
299,
111-115
|
| 4.
|
Eisenreich, W.,
Kupfer, E.,
Weber, W.,
and Bacher, A.
(1997)
J. Biol. Chem.
272,
867-874
|
| 5.
|
Cane, D. E.,
Liang, T.-C.,
Taylor, P. B.,
Chang, C.,
and Yang, C.-C.
(1986)
J. Am. Chem. Soc.
108,
4957-4964
|
| 6.
|
Weissman, K. J.,
Tomoney, M.,
Bycroft, M.,
Grice, P.,
Hamerfeld, U.,
Staunton, J.,
and Leadley, P. F.
(1997)
Biochemistry
36,
13849-13855
|
| 7.
|
Alberts, A. W.,
Majerus, P. W.,
Talamo, B.,
and Vagelos, P. R.
(1964)
Biochemistry
3,
1563-1571
|
| 8.
|
Smeland, T. E.,
Cuebas, D.,
and Schulz, H.
(1991)
J. Biol. Chem.
266,
23904-23908
|
| 9.
|
Jin, S. J.,
Hoppel, C. L.,
and Tserng, K. Y.
(1992)
J. Biol. Chem.
267,
119-125
|
| 10.
|
Saito, K.,
Kawaguchi, A.,
Okuda, S.,
Seyama, Y.,
and Yamakawa, T.
(1980)
Biochim. Biophys. Acta
618,
202-213
|
| 11.
|
Robinson, J. L.,
and Rose, I. A.
(1972)
J. Biol. Chem.
247,
1096-1105
|
| 12.
|
Rose, I. A.,
and Kuo, D. J.
(1989)
Biochemistry
28,
9579-9585
|
| 13.
|
Goese, M.,
Kammhuber, K.,
Bacher, A.,
Zenk, M. H.,
and Eisenreich, W.
(1999)
Eur. J. Biochem.
263,
447-454
|
| 14.
|
Poe, M.,
and Benkovic, S. J.
(1980)
Biochemistry
19,
4576-4582
|
Copyright © 2000 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Copyright © 2000 by the American Society for Biochemistry and Molecular Biology.
TOP
ABSTRACT
INTRODUCTION
