Nonactin biosynthesis: The initial committed step is the condensation of acetate (malonate) and succinate

Article abstract of DOI:10.1021/ja016965f

Nonactin is a macrotetrolide antibiotic produced by Streptomyces griseus subsp. griseus ETH A7796 that has shown activity against the P170-glycoprotein efflux pump associated with multiple drug resistant cancer cells. Nonactin is a polyketide, albeit a highly atypical one. The structure is composed of two units of each of the enantiomers of nonactic acid, arranged in a macrocycle, so that the molecule has S4 symmetry and is achiral. The monomer units, (+)- and (-)-nonactic acid, are derived from acetate, succinate, and propionate, although the exact details of the assembly process are quite unclear. We have used feeding experiments with a series of multiple stable isotope labeled precursors to elucidate the details of the first committed step of nonactic acid biosynthesis. We have found that the ¹³C label from 3-ketoadipate is incorporated specifically into both nonactic acid and its homologue, homononactic acid. The data conclusively show that the first committed step of nonactin biosynthesis is the coupling of a succinate derivative with either acetate or malonate. The differentiation into either nonactate or homononactate occurs after the initial condensation; the homologues are not derived from use of a different "starter unit" by the nonactate polyketide synthase. The first step of nonactin biosynthesis involves achiral intermediates; differentiation between the known enantiocomplementary biosynthesis pathways to form each enantiomer of the precursor monomer units likely occurs after the initial condensation reaction.

Full text of DOI:10.1021/ja016965f

Published on Web 03/01/2002
Nonactin Biosynthesis: The Initial Committed Step Is the
Condensation of Acetate (Malonate) and Succinate
Michael E. Nelson and Nigel D. Priestley*
Contribution from the Department of Chemistry, The UniVersity of Montana,
Missoula, Montana 59812-1656
Received August 29, 2001
Abstract: Nonactin is a macrotetrolide antibiotic produced by Streptomyces griseus subsp. griseus ETH
A7796 that has shown activity against the P170-glycoprotein efflux pump associated with multiple drug
resistant cancer cells. Nonactin is a polyketide, albeit a highly atypical one. The structure is composed of
two units of each of the enantiomers of nonactic acid, arranged in a macrocycle, so that the molecule has
S4 symmetry and is achiral. The monomer units, (+)- and (-)-nonactic acid, are derived from acetate,
succinate, and propionate, although the exact details of the assembly process are quite unclear. We have
used feeding experiments with a series of multiple stable isotope labeled precursors to elucidate the details
of the first committed step of nonactic acid biosynthesis. We have found that the 13C label from 3-ketoadipate
is incorporated specifically into both nonactic acid and its homologue, homononactic acid. The data
conclusively show that the first committed step of nonactin biosynthesis is the coupling of a succinate
derivative with either acetate or malonate. The differentiation into either nonactate or homononactate occurs
after the initial condensation; the homologues are not derived from use of a different “starter unit” by the
nonactate polyketide synthase. The first step of nonactin biosynthesis involves achiral intermediates;
differentiation between the known enantiocomplementary biosynthesis pathways to form each enantiomer
of the precursor monomer units likely occurs after the initial condensation reaction.
Introduction
Streptomyces griseus subsp. griseus ETH A7796 (DSM40695)
produces a group of macrocyclic, ionophore antibiotics known
as the macrotetrolides.1-8 The parent compound, nonactin
(Figure 1, 1), has been shown to possess antitumor activity both
against mammalian cell lines in vitro and against Crocker
Sarcoma 180 in studies in mice.1 Nonactin has also been shown
to be an effective inhibitor of the 170-kDa-P-glycoprotein-
mediated efflux of 4-O′-tetrahydropyranyldoxorubicin in mul-
tidrug resistant erythroleukemia K562 cells at subtoxic concen-
trations.9 The macrotetrolide antibiotics also act as ionophores
being effective against Gram positive bacteria, mycobacteria,
and fungi.
The macrotetrolide homologues (Figure 1) are derived from
the sequential substitution of ethyl groups for methyl groups
on the macrocyclic backbone. The naturally occurring homo-
(1) Bennett, R. E.; Brindle, S. A.; Giuffre, N. A.; Jackson, P. W.; Kowald, J.;
Pansy, F. E.; Perlman, D.; Trejo, W. H. Antimicrob. Agents Chemother.
1961, 169-172.
(2) Corbaz, R.; Ettinger, L.; Gaumann, E.; Keller-Schlierlein, W.; Kradolfer,
F.; Kyburz, E.; Neipp, L.; Prelog, V.; Zahner, H. HelV. Chim. Acta 1955,
38, 1445.
Figure 1. Structures of the naturally occurring macrotetrolides and the
monomer precursor acids.
(3) Dobler, M. HelV. Chim. Acta 1972, 55, 1371-1384.
(4) Dutcher, J. D. Antimicrob. Agents Chemother. 1961, 173-177.
(5) Gerlach, H.; Hutter, R.; Keller-Schlierlein, W.; Seibl, J.; Zahner, H. HelV.
Chim. Acta 1967, 50, 1782-1793.
logues exhibit a wide range of potency.1,2 The minimum
inhibitory concentration of nonactin against Staphylococcus
aureus and Mycobacterium boVis is over an order of magnitude
greater than that of dinactin; the differences in activity follow
the differences in ion-binding affinity. The association constant
for the K+ complex of dinactin is 7-fold greater than that of
nonactin.10
(6) Keller-Schierlein, W.; Gerlach, H. Fortschr. Chem. Org. Naturst. 1968,
26, 161-189.
(7) Menshikov, G. P.; Rubinsthein, M. M. J. Gen. Chem. USSR 1956, 26, 2267.
(8) Meyers, E.; Pansy, F. E.; Perlman, D.; Smith, D. A.; Weisenborn, F. L. J.
Antibiot. 1965, 18, 128.
(9) Borrel, M. N.; Pereira, E.; Fiallo, M.; Garnier-Suillerot, A. Eur. J. Biochem.
1994, 223, 125-133.
9
2894 VOL. 124, NO. 12, 2002 J. AM. CHEM. SOC.
10.1021/ja016965f CCC: $22.00 © 2002 American Chemical Society

Nonactin Biosynthesis
A R T I C L E S
Scheme 1. Late Steps of the Nonactate Biosynthesis Pathway
Figure 2. Biosynthesis origin of the atoms in the nonactate monomer.
The generic macrotetrolide structure is a cyclic tetraester
composed of monomers of either nonactate 6, homononactate
7, or bishomononactate 8 (Figure 1). For example, the 32-
membered macrocyclic ring of nonactin is composed of two
units of (+)-nonactate and two units of (-)-nonactate, joined
alternately (+)-(-)-(+)-(-) around the macrocycle so that
the whole structure has S4 symmetry and is, therefore, achiral.
Resolving the details of macrotetrolide biosynthesis is a
significant task as any credible hypothesis has to account for
the production of both enantiomers of the precursor monomer
units (6-8). Further, the incorporation of a succinate into the
precursor backbone is an unusual twist on conventional polyketide
biosynthesis.
Scheme 2. Early Steps of the Nonactate Biosynthesis Pathway
Showing the Two Alternate Hypotheses of Robinson; Initial
Coupling of a C2 Unit (Acetate or Malonate) with Either Acetate or
Succinate
Initial biosynthesis studies utilized 14C-labeled compounds
and established that acetate, propionate, and succinate were the
primary metabolic precursors of the macrotetrolides.11-13 The
early work was confirmed and greatly extended by Robinson
and co-workers who used extensive feeding studies with both
stable and radiolabeled compounds (Figure 2).14-19 Robinson
postulated a pathway for nonactate biosynthesis that was based
upon polyketide biosynthesis and demonstrated the highly
unusual nature of the pathway. The late steps of nonactate
biosynthesis were confirmed by Robinson and Spavold who
synthesized the acyclic, late intermediates 10 and 11 (Scheme
1). The compounds were enantioselectively incorporated into
nonactin; 10 was converted into (+)-nonactate, and 11 was
converted into (-)-nonactate. The data showed that the late steps
of nonactate biosynthesis occur via a pair of enantiocomple-
mentary pathways.19 To this date, however, the details of the
conversion of the primary metabolites acetate, propionate, and
succinate into 10 and 11 remain unconfirmed. Robinson and
co-workers had proposed a pathway in which the coupling of
an acetate (or malonate) with an acetate to generate acetoacetate
was the first committed step (Scheme 2). The acetoacetate was
then coupled to a succinate whereupon a rather unusual
rearrangement reaction involving a decarboxylation step affords
the diketoheptanoate 9. It was originally thought that differentia-
tion into the pair of enantiocomplementary pathways would
occur quite late in the overall pathway; extension with meth-
ylmalonate, following typical polyketide chemistry, would
generate the achiral intermediate 13 (Scheme 2). The enantio-
complementary pathways were believed to branch at this point,
dependent upon the specificity of ketoreductase enzymes. The
reductive steps, together with an elimination of water, would
generate the confirmed intermediates 10 and 11. The pathway
was consistent with all the data derived from direct incorporation
of labeled acetate, propionate, and succinate. One difficulty,
noted by Robinson, was in the interpretation of data obtained
in a feeding study with ethyl [1,2,3,4-13C4]acetoacetate.16,18
While observing no direct, intact incorporation of label into the
expected positions, labeling of atoms derived from propionate
and methylmalonate was observed. Robinson postulated that the
acetoacetate was first reduced to butyrate. The action of a
coenzyme B12-dependent mutase would convert the butyrate into
isobutyrate; subsequent oxidation would generate methylma-
lonate, thereby explaining the observed incorporation of label
into positions of nonactate formally derived from propionate
and methylmalonate. The indirect result, based upon a lack of
(10) Izatt, R. M.; Bradshaw, J. S.; Nielsen, S. A.; Lamb, J. D.; Christensen, J.
J.; Sen, D. Chem. ReV. 1985, 85, 271-339.
(11) Pape, H. Arch. Microbiol. 1972, 82, 254-264.
(12) Stahl, P.; Pape, H. Arch. Microbiol. 1972, 85, 239-248.
(13) Stahl, P. O. Untersuchungen zur Biosynthese der Macrotetrolide bei
Streptomyces griseus; Eberhard-Karls-Universitat: Tubingen, 1975.
(14) Ashworth, D. M.; Robinson, J. A.; Turner, D. L. J. Chem. Soc., Chem.
Commun. 1982, 491-493.
(15) Ashworth, D. M.; Robinson, J. A. J. Chem. Soc., Chem. Commun. 1983,
1327-1329.
(16) Ashworth, D. M.; Robinson, J. A.; Turner, D. L. J. Chem. Soc., Perkin
Trans. 1 1988, 1719-1727.
(17) Ashworth, D. M.; Clark, C. A.; Robinson, J. A. J. Chem. Soc., Perkin Trans.
1 1989, 1461-1467.
(18) Clark, C. A.; Robinson, J. A. J. Chem. Soc., Chem. Commun. 1985, 1568-
1569.
(19) Spavold, Z. M.; Robinson, J. A. J. Chem. Soc., Chem. Commun. 1988,
4-6.
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2895

A R T I C L E S
Nelson and Priestley
Scheme 3. Synthesis of a 13C-Labeled, Thioester-Activated
Acetoacetate Derivative
Scheme 4. Synthesis of a 13C/2H-Labeled, Thioester-Activated
Hydroxybutyrate Derivative
corresponding enolate was generated at low temperatures and
quenched with an imidazolide preformed from [1-13C1]acetic
acid and 1,1-carbonyldiimidazole. The conditions used represent
an optimized method in our hands; many variants of the coupling
were evaluated subject to the constraints on reagent stoichiom-
etry imposed by using two 13C-labeled fragments. Currently,
we obtain consistent yields between 50 and 75%. The ketone
15 was protected using standard conditions as a dioxolane
derivative 16, whereupon the free acid was unmasked by
hydrogenolysis. Standard coupling to N-caprylcysteamine gave
17. Deprotection of the ketone gave the target 18 compound in
30% overall yield.
direct incorporation, argued against Robinson’s first biosynthesis
hypothesis. Robinson suggested that if the lack of direct
incorporation of acetoacetate were not due to the specificity of
a nonactate polyketide synthase then an alternative pathway for
the initial stages of nonactate biosynthesis might be operative.
The latter pathway, as proposed by Robinson, was based upon
the coupling of an acetate (or malonate) with succinate as the
first committed step (Scheme 2). Intermediate 12, formed by
such a condensation, in one of a number of possible pathways,
would then be acylated with either acetate (to eventually form
nonactate, 6) or propionate (to eventually form homononactate,
7). After acylation, transposition of the thioester would allow
for spontaneous decarboxylation generating 9, a central inter-
mediate in both of Robinson’s hypotheses. The evidence for
the latter pathway, however, was based upon indirect evidence
- the lack of direct incorporation of label from ethyl acetoac-
etate. In this current report, we describe the synthesis and use
of dual isotope-labeled, thioester-activated analogues of ac-
etoacetate, hydroxybutyrate, and 3-ketoadipate (12). The incor-
poration of label, or lack thereof, from the compounds into
nonactate 6 and homononactate 7 conclusively shows that
Robinson’s second hypothesis is operative in S. griseus. The
coupling of acetate (or malonate) with succinate is the first
committed step of nonactate biosynthesis.
Synthesis of a [3-2H1,3-13C1]Hydroxybutyrate Thioester.
It was conceivable that nonactate biosynthesis could be initiated
with a thioester-activated hydroxybutyrate. We sought to
synthesize [3-13C1,3-2H1]hydroxybutyrate activated as its N-
caprylcysteamine thioester 22 to investigate the possibility
(Scheme 4). The dual label pattern was chosen to determine if
any observed 13C-label incorporation was due to oxidation of
22 to acetoacetate. Such an oxidation would separate the 13C
2
and H labels. Generation of the appropriately labeled benzyl
acetoacetate was accomplished using lithium enolate chemistry.
Reduction of benzyl [3-13C1]acetoacetate with NaBD4 readily
generated the racemic hydroxybutyrate, 19. Protection of the
alcohol as the tert-butyldimethylsilyl ether (20), replacement
of the benzyl ester with the thioester (21), and liberation of the
target compound (22) proceeded relatively uneventfully to give
the labeled target in approximately 37% overall yield.
Synthesis of a [2,3-13C2]-3-Ketoadipate Thioester. The
3-ketoadipate derivative 28 (Scheme 5) was chosen as a target
as the compound is the product of a first biosynthesis step
coupling of acetate (or malonate) with succinate, that is, the
first intermediate on the alternative pathway proposed by
Robinson. The route to 28 required an unsymmetrically 13C-
labeled succinate derivative. Labeled succinate derivatives are
available but usually have a symmetrical disposition of the 13C-
labels. The labeling in the succinate in our work would remain
unscrambled by judicious choice of protecting groups for the
carboxyl groups. The enolate derived from benzyl [1-13C1]-
acetate was quenched with unlabeled tert-butyl bromoacetate
to afford the succinate diester 23. The benzyl group was
removed by hydrogenolysis, and the free carboxyl group (24)
was activated as an imidazolide. The electrophile so generated
was used to quench an enolate generated from benzyl [2-13C1]-
acetate. In this manner, the contiguous, dual 13C labels were
Results and Discussion
Synthesis of a [2,3-13C2]Acetoacetate Thioester. We sought
to synthesize a dual 13C-labeled acetoacetate thioester to confirm
Robinson’s results from the ethyl acetoacetate feeding experi-
ment. The thioester-activated target was chosen as it has been
observed that incorporation into a polyketide of a putative,
thioester-activated intermediate can be achieved even when the
corresponding ester or acid failed to show incorporation.
The synthesis of the labeled acetoacetate was based heavily
upon chemistry we had previously used for the synthesis of [5,6-
13C2]-4,6-diketoheptanoate.20 Benzyl [2-13C1]acetate was pre-
pared from commercially available sodium [2-13C1]acetate by
reaction with benzyl bromide in DMF (Scheme 3). The
(20) Nelson, M. E.; Derrer, A.; Priestley, N. D. 1999, unpublished work.
9
2896 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002

Nonactin Biosynthesis
A R T I C L E S
Scheme 5. Synthesis of a 13C-Labeled, Thioester-Activated
Scheme 6. Interpretation of Observed 13C-Label Incorporationa
3-Ketoadipate Derivative
a The dual label is incorporated intact into nonactin and monactin and is
observed, after degradation and derivatization, in the methyl ester derivatives
of nonactate 29 and homononactate 30.
the compounds fed were either achiral (18 and 28) or racemic
(22), and we logically assume that no discrimination between
enantiocomplementary biosynthesis pathways occurs.
Feeding Results. In the case of the thioesters of acetoacetate
18 and hydroxybutyrate 22, no dual 13C-13C label incorporation,
introduced into 25. Protection of the ketone 25 as a dioxane
proved surprisingly problematic. Reaction rates were quite slow
with a number of 1,3- and 1,2-diols; attempts to force the
reaction caused extensive decomposition presumably through
loss of the tert-butyl ester under the acidic conditions. Using
bis-silyl ether diol derivatives with TMSOTf as catalyst also
proved ineffective. Eventually, reaction at reflux for 20 h with
neopentyl glycol using PPTS as catalyst proved to be the most
efficient procedure. Replacement of the benzyl ester with the
thioester was accomplished as described for our other syntheses
(20 to 21). Removal of the dioxane and tert-butyl ester protecting
groups was accomplished by reflux with pTsOH in acetone to
afford the target molecule in 4% overall yield.
2
in the former case, or H-13C dual label incorporation, in the
latter case, was observed in either nonactate or homononactate.
No significant single 13C enrichments were observed in either
case. The amount of compound fed was lower than that used
by Robinson. The indirect incorporation of label from ethyl
[1,2,3,4-13C4]acetoacetate observed by Robinson was obtained
in part because of the relatively large sample fed and the
presumed slow, continuous release of acetoacetate from the ethyl
ester by nonspecific hydrolysis.
In the case of the thioester of [2,3-13C2]-3-ketoadipate 28,
however, the results were strikingly different. In the 13C NMR
spectra of both methyl nonactate and methyl homononactate
clear doublets arising from 13C-13C coupling were observed at
positions assigned to C6 and C7 (Figure 3). The 13C-13C dual
label of the precursor is incorporated intact. The position of
label incorporation is entirely consistent with Robinson’s
alternate biosynthesis hypothesis, but unlike the previous results
of the indirect 13C incorporation from acetoacetate, the current
data support direct incorporation of the label. The data support
the notion that the first committed step of nonactate biosynthesis
is the coupling of an acetate (or malonate) unit with succinate
to generate 3-ketoadipate (12; Scheme 2). The anomalous
incorporation derived from ethyl acetoacetate observed by
Robinson, and the lack of incorporation from the acetoacetate
and hydroxybutyrate thioesters, while being negative evidence,
also support the latter pathway. Because label from 3-ketoadipate
is found in both nonactate and homononactate, it is evident that
the differentiation of the pathways to form the homologues
occurs after the first committed step. Naturally, acylation of
the 3-ketoadipate with acetyl-CoA would lead to nonactate;
acylation with propionyl-CoA would lead to homononactate.
The synthesis of nonactate and homononactate is not controlled,
therefore, by a choice of either acetate or propionate as a “starter
unit” by the nonactate polyketide synthase. The divergence of
the biosynthesis pathways to form both enantiomers of nonactate
most likely occurs after the first committed step as the
intermediate 12 is achiral. The result, therefore, has great
General Feeding Study Protocol and Analysis Procedures.
The labeled compounds were administered in ethanol solution
to fermentative cultures of S. griseus at the onset of mac-
rotetrolide production. Care was taken to ensure that the sample
of S. griseus used was capable of the synthesis of mac-
rotetrolides. Any strain that produced less than 250 mg L-1 of
macrotetrolides in control culture was considered to be com-
promised, and a fresh set of spore suspensions was prepared
from our original standard culture. In this manner, yields of
macrotetrolides of 1000-3000 mg L-1 could be consistently
achieved. At 48 h following administration of the labeled
compound, the macrotetrolides were recovered by acetone
extraction of the culture cell paste.16,17 A preliminary fraction-
ation of the extract to obtain a macrotetrolide sample was
achieved by chromatography on silica gel. Screening of the
preliminary mixture for 13C-label incorporation was done by
13C NMR. The macrotetrolide fraction, being mainly a mixture
of nonactin and monactin, was subjected to methanolysis to
afford a mixture of methyl nonactate 29 and methyl homonon-
actate 30 (Scheme 6); the esters were readily separable by
chromatography on silica gel.16,17 Analysis by 13C NMR was
carried out on the cleaner, less complicated, methyl ester
samples. The procedures, of course, generate mixtures of the
enantiomers of 29 and 30. Further analysis of the incorporation
of label into the individual enantiomers was not carried out as
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2897

A R T I C L E S
Nelson and Priestley
nonactate occurs after the initial step potentially by acylation
of the â-ketothioester product. Partition between the enantio-
complementary pathways most likely occurs after the initial
committed step.
Experimental Section
General. Chemical Synthesis. Solvents were obtained from Fisher
Scientific. All other chemicals, unless noted otherwise, were obtained
from the Aldrich Chemical Co. (Milwaukee, WI). The solvents CH2-
Cl2 (over CaH2), diethyl ether (over Na/K-benzophenone), and THF
(over Na/K-benzophenone) were dried and distilled prior to use. MgSO4
refers exclusively to anhydrous MgSO4. Solutions were concentrated
by evaporation in vacuo. All synthesis procedures, unless noted
otherwise, were carried out under a slight positive pressure of dry argon
gas. Column chromatography was performed using Merck Silica Gel
60. NMR spectra were acquired at 400 MHz (1H) and 100 MHz (13C),
were referenced to the residual solvent, and are reported as chemical
shift (δ/ppm), splitting pattern, coupling constant (J/Hz), and intensity.
General. Strains. Streptomyces griseus subsp. griseus ETH A7796
was obtained from the Deutsche Sammlung von Mikroorganismen und
Zellkulturen GmbH (DSM 40695). S. griseus was grown in 2 × TSB
liquid medium25 or on R2YE solid medium.25 For macrotetrolide
production, S. griseus was propagated by previously reported methods.26
Macrotetrolides were isolated as described previously.26 In each of the
feeding experiments, 100 mg of a 1:1 mixture of labeled compound
and its unlabeled counterpart were added to a 250 mL shake flask
fermentative culture of S. griseus.
Benzyl Acetate (14). Benzyl bromide (5.9 mL, 50 mmol) was added
via syringe to a suspension of sodium acetate (2.0 g, 24 mmol) in DMF
(50 mL). The stirred mixture was heated at 96 °C for 3 days. Water
(200 mL) and Et2O (300 mL) were added to the cooled mixture, mixed,
and the organic layer was recovered. The water fraction was further
extracted with Et2O (2 × 300 mL). The pooled Et2O fractions were
dried and concentrated, and the product was purified by chromatography
on silica gel, eluting with EtOAc-hexanes (9:1), to give 14 (3.5 g,
94.8%), which was shown to be identical to authentic material.
Benzyl [2-13C1]Acetate. This was prepared according to the method
for the unlabeled material from benzyl bromide and sodium [2-13C1]-
1
acetate (4.86 g, 89.1%). H NMR (400 MHz, CDCl3): δ 2.08 (3H, d,
J1H-13C ) 134.5 Hz), 5.10 (2H, s), 7.30 (5H, m). 13C NMR (100 MHz,
CDCl3): δ 21.1, 66.5, 128.5, 128.8, 136.3, 171.0 (d, J13C-13C ) 59.4
Hz).
Benzyl 3-Oxobutanoate (15). Acetic acid (0.95 mL, 16.4 mmol)
was added to a stirred suspension of CDI (2.7 g, 16.6 mmol) in dry
THF (16 mL). Gas was evolved, and the CDI was dissolved over 10
min at room temperature to generate a solution of the imidazolide.
n-BuLi (35.3 mmol) in hexanes (25 mL) was added dropwise to a
solution of diisopropylamine (5.4 mL, 38.4 mmol) in dry THF (30 mL)
and stirred in a dry ice-acetone bath until the temperature reached
-60 °C. Benzyl acetate (4.8 mL, 33.2 mmol) was added via syringe,
maintaining the temperature below -20 °C, and then stirred until the
reaction temperature fell to -60 °C. The acetylimidazole solution was
transferred directly into the reaction vessel, again maintaining the
temperature below -20 °C. The reaction was stirred for 1 min,
Figure 3. Expansions of the 13C NMR spectra of methyl nonactate 29 and
methyl homononactate 30 derived from macrotetrolides isolated from a
culture to which 28 had been added. A: C-6 of 29 (0.26 ( 0.02% specific
incorporation; 1.4 ( 0.1% absolute incorporation). B: C-7 of 29 (0.30 (
0.02% specific; 1.6 ( 0.1% absolute incorporation). C: C-5 of 29 (peak
with no enrichment shown for comparison). D: C-6 of 30 (0.51 ( 0.04%
specific; 1.4 ( 0.1% absolute incorporation). E: C-7 of 30 (0.43 ( 0.04%
specific; 1.2 ( 0.1% absolute incorporation). F: C-5 of 30 (peak with no
enrichment shown for comparison).
(21) Smith, W. C.; Xiang, L.; Shen, B. Antimicrob. Agents Chemother. 2000,
44, 1809-1817.
(22) Walczak, R. J.; Woo, A. J.; Strohl, W. R.; Priestley, N. D. FEMS Lett.
2000, 183, 171-175.
implication for hypotheses which relate the deduced functions
of gene products of the nonactin biosynthesis gene cluster to
chemical conversions occurring in the biosynthesis pathway.21-24
In conclusion, feeding experiments with three 13C-labeled,
thioester-activated potential precursors have been used to
demonstrate that the first committed step of macrotetrolide
biosynthesis is the coupling of a succinate unit with an acetate
(or malonate). Differentiation to form the homologues of
(23) Woo, A. J.; Strohl, W. R.; Priestley, N. D. Antimicrob. Agents Chemother.
1999, 43, 1662-1668.
(24) Kwon, H.-J.; Smith, W. C.; Xiang, L.; Shen, B. J. Am. Chem. Soc. 2001,
123, 3385-3386.
(25) Kieser, T.; Bibb, M. J.; Buttner, M. J.; Chater, K. F.; Hopwood, D. A.
Practical Streptomyces Genetics; The John Innes Foundation: Norwich,
2000.
(26) Priestley, N. D. Studies on Enzymes InVolVed in the Primary and Secondary
Metabolism of Antibiotic Producing Streptomyces; Southampton Univer-
sity: Southampton, 1991.
9
2898 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002

Nonactin Biosynthesis
A R T I C L E S
quenched by addition of water (100 mL), acidified to pH 4 with
concentrated HCl, and then partitioned between water and CHCl3 (200
mL). The aqueous portion was washed with CHCl3 (2 × 200 mL), and
the pooled organic portions were dried with Na2SO4, filtered, and
concentrated. The product was recovered by chromatography on silica
gel, eluting with EtOAc-hexanes (15:85), as an oil (2.1 g, 66.4%),
which was shown to be identical to authentic material.
(3H, t, J1H-1H ) 7.1 Hz), 1.26 (8H, m), 1.46 (3H, dd, J1H-13C ) 4.5,
2.6 Hz), 1.59 (2H, m), 2.14 (2H, t, J1H-1H ) 7.8 Hz), 2.88 (2H, dd,
J1H-13C ) 115.8, 5.2 Hz), 3.05 (2H, t, J1H-1H ) 7.1 Hz), 3.45 (2H, dt,
J1H-1H ) 6.5, 5.8 Hz), 3.98 (4H, d, J1H-13C ) 3.2 Hz), 5.84 (1H, br s).
13C NMR (100 MHz, CDCl3): δ 14.1, 22.6, 24.6 (dt, J13C-13C ) 22.9,
6.1 Hz), 25.6, 28.9, 29.0, 29.2, 31.7, 36.7, 39.4, 52.6, 64.8, 107.5
(enriched, d, J13C-13C ) 42.7 Hz), 174.4, 195.7 (d, J13C-13C ) 45.8 Hz).
S-[2-(Octanoylamino)ethyl] 3-Oxobutanethioate (18). The ketal
17 (1.02 g, 3.1 mmol) and pTsOH (0.20 g, 1.1 mmol) were dissolved
in acetone (30 mL), and the stirred mixture was heated at reflux for 18
h. The mixture was concentrated, resuspended in Et2O-CHCl3 (1:1,
200 mL), and washed with 5% (w/v) aqueous NaHCO3 solution and
then brine. The organic phase was dried over anhydrous Na2SO4,
filtered, and concentrated. The target compound was obtained by
chromatography on silica gel, eluting with EtOAc-hexanes (3:2), as a
white solid (0.68 g, 76.2%). IR (powder-ATR): 3283, 1714, 1684, 1627,
1561 cm-1. UV (acetonitrile): 236 (ꢀ ) 3800 dm3 mol-1 cm-1), 281
nm (2300). 1H NMR (400 MHz, CDCl3): δ 0.85 (3H, t, J1H-1H ) 6.8
Hz), 1.25 (8H, m), 1.58 (2H, m), 1.92 (1H, enol, s), 2.12 (0.67H, enol,
t, J1H-1H ) 7.8 Hz), 2.13 (1.33H, keto, t, J1H-1H ) 7.9 Hz), 2.24 (2H,
keto, s), 3.06 (0.67, enol, t, J1H-1H ) 6.0 Hz), 3.07 (1.33H, keto, t,
J1H-1H ) 6.2 Hz), 3.45 (1.33H, keto, q, J1H-1H ) 6 Hz), 3.45 (0.67H,
enol, q, J1H-1H ) 5.8 Hz), 3.68 (1.5H, keto, s), 5.44 (0.3H, enol, s),
5.94 (0.66H, keto, br s), 5.99 (0.33H, enol, br s), 12.57 (0.5H, enol-H,
s). 13C NMR (100 MHz, CDCl3): δ 14.0, 21.0, 22.6, 25.6, 25.6, 27.7,
28.95, 29.2, 29.25, 30.4, 31.6, 36.6, 36.7, 38.9, 39.7, 58.0, 99.85, 173.4,
173.5, 173.95, 192.2, 194.3, 199.7. ESI-HRMS calcd for C14H26NO3S
[M + H]+, 288.1633; found, 288.1617. Anal. Calcd for C14H26NO3S:
C, 58.5; H, 8.77; N, 4.87; S, 11.16. Found: C, 58.40; H, 8.74; N, 4.82;
S, 11.16.
Benzyl [3-13C1]-3-Oxobutanoate. This was prepared as was the
1
unlabeled material from [1-13C1]acetic acid (2.6 g, 82.1%). H NMR
(400 MHz, CDCl3): δ 2.24 (3H, d, J1H-13C ) 6.4 Hz), 3.49 (2H, d,
J1H-13C ) 6.4 Hz), 5.1 (2H, s), 7.35 (5H, m). 13C NMR (100 MHz,
CDCl3): δ 30.15 (d, J13C-13C ) 42.7 Hz), 50.0 (d, J13C-13C ) 36.7
Hz), 67.1, 128.35, 128.5, 128.6, 135.2, 166.9 (enol, enriched), 175.9,
200.4 (keto, enriched).
Benzyl (2-Methyl-1,3-dioxolan-2-yl)acetate (16). Ketoester 15 (0.94
g, 4.9 mmol), ethylene glycol (0.86 mL, 15.4 mmol), and PPTS (0.32
g, 1.3 mmol) were dissolved in benzene (50 mL), and the mixture was
heated at reflux for 11 h with azeotropic removal of water by means
of a Dean-Stark trap. The mixture was concentrated and suspended
in Et2O (100 mL) and washed first with aqueous saturated NaHCO3
solution (100 mL) and then brine (100 mL). The organic phase was
dried, filtered, and concentrated; the dioxolane product was then
recovered by chromatography on silica gel, eluting with EtOAc-
hexanes (1:4), as an oil (1.02 g, 88.2%). IR (liquid film): 1736 cm-1
.
UV (acetonitrile): 257 (ꢀ ) 190 dm3 mol-1 cm-1). 1H NMR (400 MHz,
CDCl3): δ 1.49 (3H, s), 2.71 (2H, s), 3.92 (4H, m), 5.13 (2H, s), 7.35
(5H, m). 13C NMR (100 MHz, CDCl3): δ 24.5, 44.1, 64,7, 66.3, 107.5,
128.10, 128.2, 128.4, 135.4, 169.2. ESI-HRMS calcd for C13H16O4-
Na [M + Na]+, 259.0946; found, 259.0957. Anal. Calcd for C13H16O4:
C, 66.09; H, 6.83. Found: C, 65.86; H, 6.87.
S-[2-(Octanoylamino)ethyl] [2,3-13C2]-3-Oxobutanethioate. This
was prepared as described for the unlabeled material (0.65 g, 94.6%).
1H NMR (400 MHz, CDCl3): δ 0.85 (3H, t, J1H-1H ) 6.8 Hz), 1.25
(8H, m), 1.58 (2H, m), 1.92 (0.4H, enol, dd, J1H-13C ) 6.2, 3.8 Hz),
2.14 (2H, t, J1H-1H ) 7.3 Hz), 2.25 (2.6H, keto, dd, J1H-13C ) 6.0, 1.5
Hz), 3.07 (2H, t, J1H-1H ) 6.6 Hz), 3.44 (2H, q, J1H-1H ) 6.4 Hz),
3.67 (1.9H, keto, dd, J1H-13C ) 131.1, 6.2 Hz), 5.44 (0.1H, enol, ddd,
J1H-13C ) 167.5, 4.4 Hz, J1H-1H ) 0.7 Hz), 5.91 (1H, b), 12.57 (0.1H,
enol-H, t, J1H-13C ) 4.6 Hz). 13C NMR (100 MHz, CDCl3): δ 14.03,
21.02 (enol, dt, J13C-13C ) 23.3, 6.9 Hz), 22.6, 25.6, 25.65, 27.7, 29.0,
29.2, 29.3, 30.4 (keto, dt, J13C-13C ) 19.8, 14.12 Hz), 31.6, 36.6, 36.7,
38.9, 39.7, 58.0 (keto, enriched, d, J13C-13C ) 36.6 Hz), 99.8 (enol,
enriched, d, J13C-13C ) 69.8 Hz), 173.4, 173.5, 173.9 (enol, enriched,
d, J13C-13C ) 69.4 Hz), 192.2 (enol, dt, J13C-13C ) 22.9, 2.3 Hz), 194.3
(keto, t, J13C-13C ) 30.9 Hz), 199.7 (keto, enriched, d, J13C-13C ) 36.6
Hz).
Benzyl 3-Hydroxybutyrate (19). Acetoacetate 15 (2.1 g, 11.0 mmol)
was dissolved in THF (110 mL) and water (22 mL) and placed in an
ice bath. NaBH4 (0.13 g, 3.3 mmol) was added and stirred for 30 min
at 0 °C. More NaBH4 (0.04 g, 1.1 mmol) was added every 20 min
until no starting material was present (40 min total). The reaction was
diluted with water (100 mL) and partitioned between water and Et2O
(3 × 200 mL). The pooled Et2O portions were dried with MgSO4,
filtered, and concentrated. The product was recovered by chromatog-
raphy on silica gel, eluting in EtOAc-hexanes (1:4), as an oil (1.63 g,
76.1%), which was shown to be identical to authentic material.
Benzyl [3-13C1,3-2H1]-3-Hydroxybutanoate. This was prepared as
described for the unlabeled material from benzyl [3-13C1]-3-oxobu-
tanoate (2.26 g, 85.7%). 1H NMR (400 MHz, CDCl3): δ 1.21 (3H, d,
J1H-13C ) 4.4 Hz), 2.5 (2H, m), 2.82 (1H, s), 5.14 (2H, s), 7.35 (5H,
m). 13C NMR (100 MHz, CDCl3): δ 22.3 (d, J13C-13C ) 39.7 Hz),
42.7 (d, J13C-13C ) 36.6 Hz), 63.9 (enriched, t, J13C-13C ) 22.8 Hz),
67.5, 128.25, 128.4, 128.6, 135.55, 172.7.
Benzyl [2-13C1]-([2-13C1]-2-Methyl-1,3-dioxolan-2-yl)acetate. This
was prepared as described for the unlabeled material (0.86 g, 81.1%).
1H NMR (400 MHz, CDCl3): δ 1.49 (3H, dd, J1H-13C ) 4.5, 2.6 Hz),
2.72 (3H, dd, J1H-13C ) 130.0, 5.8 Hz), 3.94 (4H, m), 5.14 (2H, s),
7.36 (5H, m). 13C NMR (100 MHz, CDCl3): δ 24.5 (dt, J13C-13C
)
24.4, 6.1 Hz), 44.1 (enriched, d, J13C-13C ) 44.3 Hz), 64.7, 66.3, 107.5
(enriched, d, J13C-13C ) 42.7 Hz), 128.1, 128.2, 128.4, 135.4, 169.3 (t,
J13C-13C ) 27.5 Hz).
S-[2-(Octanoylamino)ethyl] (2-Methyl-1,3-dioxolan-2-yl)ethaneth-
ioate (17). Pd on carbon catalyst (10%, 0.10 g) was added to a stirred
solution of 16 (1.02 g, 4.3 mmol) in THF (22 mL) under an atmosphere
of Ar. The Ar was replaced by H2 (balloon), and the suspension was
stirred overnight at room temperature. Celite was added, and the
suspension was filtered. The Celite was washed with CHCl3 (100 mL).
The combined organic fractions were concentrated and then dissolved
in dry CH2Cl2 (34 mL). N-Caprylcysteamine (1.89 g, 9.3 mmol), DCC
(2.27 g, 11.0 mmol), and DMAP (0.13 g, 1.1 mmol) were added, and
the suspension was stirred at room temperature for 8 h. Oxalic acid
(0.5 g, 5.6 mmol) was added, and the mixture was concentrated,
resuspended in Et2O, and the solid matter was removed by filtration.
The target thioester was recovered by chromatography on silica gel,
eluting with EtOAc-hexanes (3:2), as an oil (1.02 g, 71.5%). IR (liquid
film): 3293, 1691, 1646, 1546 cm-1. UV (acetonitrile): 235 nm (ꢀ )
1
2383 dm3 mol-1 cm-1). H NMR (400 MHz, CDCl3): δ 0.84 (3H, t,
J1H-1H ) 6.4 Hz), 1.25 (8H, m), 1.44 (3H, s), 1.57 (2H, m), 2.11 (2H,
t, J1H-1H ) 8.0 Hz), 2.88 (2H, s), 3.01 (2H, t, J1H-1H ) 6.4 Hz), 3.41
(2H, dt, J1H-1H ) 6.0, 6.4 Hz), 3.95 (4H, s), 5.84 (1H, br s). 13C NMR
(100 MHz, CDCl3): δ 14.0, 22.5, 24.6, 25.6, 28.9, 28.9, 29.2, 31.6,
36.6, 39.3, 52.5, 64.7, 107.4, 173.3, 195.5. ESI-HRMS calcd for
C16H29NO4SNa [M + Na]+, 354.1715; found, 354.1716. Anal. Calcd
for C16H29NO4SNa: C, 58.8; H, 7.9; N, 4.2; S, 9.7. Found: C, 58.08;
H, 8.90; N, 4.44; S, 9.44.
S-[2-(Octanoylamino)ethyl] [2-13C1]-([2-13C1]-2-Methyl-1,3-diox-
olan-2-yl)ethanethioate. This was prepared as described for the
unlabeled material from benzyl [2-13C1]-([2-13C1]-2-methyl-1,3-diox-
olan-2-yl)acetate (0.8 g, 69.3%). 1H NMR (400 MHz, CDCl3): δ 0.86
Benzyl 3-{[tert-Butyl(dimethyl)silyl]oxy}butanoate (20). Alcohol
19 (1.63 g, 8.4 mmol), tert-butyldimethylsilyl chloride (2.37 g, 15.7
mmol), and imidazole (1.89 g, 27.7 mmol) were dissolved in dry DMF
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2899

A R T I C L E S
Nelson and Priestley
(4 mL) and stirred for 18 h at room temperature. The mixture was
partitioned between water (100 mL) and CHCl3 (3 × 100 mL). The
pooled CHCl3 portions were dried with MgSO4, filtered, and concen-
trated. The product was recovered by chromatography on silica gel,
eluting with EtOAc-hexanes (5:95), as an oil (1.91 g, 73.8%). IR
(liquid film): 1738 cm-1. UV (acetonitrile): 258 nm (ꢀ ) 180 dm3
IR (powder-ATR): 3298, 1684, 1638 cm-1. UV (acetonitrile): 233
nm (ꢀ ) 4700 dm3 mol-1 cm-1). 1H NMR (400 MHz, CDCl3): δ 0.84
(3H, t, J1H-1H ) 6.8 Hz), 1.2 (3H, t, J1H-1H ) 6.4 Hz), 1.24 (8H, m),
1.57 (2H, m), 2.12 (2H, t, J1H-1H ) 7.5 Hz), 2.69 (2H, m), 3.01 (2H,
dt, J1H-1H ) 5.9, 1.5 Hz), 3.02 (1H, br s), 3.42 (2H, dq, J1H-1H ) 6.2,
1.8 Hz), 4.22 (1H, m), 6.01 (1H, br s). 13C NMR ( 100 MHz, CDCl3):
δ 14.0, 22.5, 22.65, 25.6, 28.8, 28.9, 29.2, 31.3, 36.6, 38.95, 52.45,
64.9, 173.6, 199.3. ESI-HRMS calcd for C14H27NO3SNa [M + Na]+,
312.1609; found, 312.1590. Anal. Calcd for C14H27NO3S: C, 58.1; H,
9.4; N, 4.84; S, 11.08. Found: C, 58.18; H, 9.42; N, 4.73; S, 11.02.
S-[2-(Octanoylamino)ethyl] [3-13C1,3-2H1]-3-Hydroxybutanethio-
ate. This was prepared as described for unlabeled material from S-[2-
(octanoylamino)ethyl] [3-13C1,3-2H1]-3-{tert-butyldimethylsilyloxy}-
butanoate (1.45 g, 81.6%). 1H NMR (400 MHz, CDCl3): δ 0.85 (3H,
t, J1H-1H ) 6.8 Hz), 1.2 (3H, d, J1H-13C ) 4.5 Hz), 1.25 (8H, m), 1.57
(2H, m), 2.13 (2H, t, J1H-1H ) 7.5 Hz), 2.69 (2H, m), 2.92 (1H, s),
3.01 (2H, dt, J1H-1H ) 6.0, 1.6 Hz), 3.42 (2H, dq, J1H-1H ) 6.2, 2.1
Hz), 5.94 (1H, s). 13C NMR (100 MHz, CDCl3): δ 14.0, 22.5 (d,
J13C-13C ) 38.9 Hz), 22.6, 25.6, 28.8, 28.95, 29.2, 31.7, 36.65, 39.0,
52.3 (d, J13C-13C ) 35.9 Hz), 64.6 (enriched, t, J13C-13C ) 22.1 Hz),
173.6, 199.3.
1
mol-1 cm-1). H NMR (400 MHz, CDCl3): δ 0.02 (3H, s), 0.05 (3H,
s), 0.85 (9H, s), 1.19 (3H, d, J1H-1H ) 6.2 Hz), 2.42 (1H, dd, J1H-1H
)
9.38, 5.18 Hz), 2.53 (1H, dd, J1H-1H ) 6.8, 7.8 Hz), 4.3 (1H, m), 5.10
(2H, d, J1H-1H ) 5.5 Hz), 7.35 (5H, m). 13C NMR (100 MHz, CDCl3):
δ -5.0, -4.5, 17.9, 23.9, 25.7, 44.85, 65.8, 66.15, 128.2, 128.2, 128.5,
135.9, 171.45. ESI-HRMS: calcd for C17H28O3NaSi [M + Na]+,
331.1705; found, 331.1710. Anal. Calcd for C17H28O3Si: C, 66.19; H,
9.15. Found: C, 66.29; H, 9.27.
Benzyl [3-13C1,3-2H1]-3-tert-Butyldimethylsilyloxybutanoate. This
was prepared as described for unlabeled material (3.59 g, 100.0%). 1H
NMR (400 MHz, CDCl3): δ 0.02 (3H, s), 0.05 (3H, s), 0.85 (9H, s),
1.18 (3H, d, J13C-1H ) 4.3 Hz), 2.41 (1H, dd, J13C-1H ) 3.8, J1H-1H
)
14.5 Hz), 2.52 (1H, dd, J13C-1H ) 6.2, J1H-1H ) 14.7 Hz), 5.10 (2H, d,
J1H-1H ) 5.4 Hz), 7.34 (5H, m). 13C NMR (100 MHz, CDCl3): δ -5.0,
-4.5, 17.9, 23.7 (d, J13C-13C ) 38.9 Hz), 25.7, 44.7 (d, J13C-13C ) 38.9
Hz), 65.4 (enriched, t, J13C-13C ) 22.1 Hz), 66.1, 128.1, 128.2, 128.5,
135.9, 171.4.
1-Benzyl 4-tert-Butyl Succinate (23). A clean, dry three-neck flask
equipped with an addition funnel was charged with dry THF (60 mL)
and diisopropylamine (4.8 mL, 29.2 mmol) and cooled in a dry ice-
acetone bath. n-Butyllithium (29.2 mmol) in hexanes (11.4 mL) was
added via syringe and stirred until the temperature dropped below -70
°C. Benzyl acetate (4.2 mL, 29.1 mmol) in dry THF (30 mL) was added
dropwise to the reaction flask, maintaining the temperature below -70
°C. When complete, the flask was transferred to a liquid N2-EtOH
slush bath and stirred until the temperature dropped below -100 °C,
while the addition funnel was charged with tert-butyl 2-bromoacetate
(7 mL, 43.4 mmol) in THF (30 mL). The bromide solution was added
dropwise to the reaction, maintaining the temperature below -100 °C,
then stirred for 30 min at -100 °C. The reaction was quenched by
addition of water (100 mL) and allowed to warm to room temperature.
The mixture was partitioned between water (200 mL) and Et2O (200
mL). The organic portion was dried with MgSO4, filtered, and
concentrated. The product was recovered by chromatography on silica
gel, eluting with EtOAc-hexanes (5:95), as an oil (5.2 g, 67.6%). IR
(liquid film): 1727 cm-1. UV (acetonitrile): 258 nm (ꢀ ) 180 dm3
S-[2-(Octanoylamino)ethyl] 3-{[tert-Butyl(dimethyl)silyl]oxy}bu-
tanoate (21). Butanoate 20 (1.91 g, 6.2 mmol) was dissolved in dry
THF (25 mL). Pd-C catalyst (10%, 0.19 g) was added and stirred
under a H2 atmosphere (balloon) for 2 h. The reaction was filtered
through Celite, and the Celite was washed with CHCl3 (150 mL). The
mixture was concentrated and dissolved in dry CH2Cl2 (40 mL) together
with N-caprylcysteamine (2.54 g, 12.5 mmol), DCC (2.47 g, 12 mmol),
and DMAP (0.28 g, 2.3 mmol). The mixture was stirred at room
temperature for 18 h. The suspension that formed was filtered, and the
solid was washed with CH2Cl2 (100 mL). The filtrate was concentrated,
resuspended in CH2Cl2 (20 mL), and filtered. The product was recovered
by chromatography on silica gel, eluting with EtOAc-hexanes (35:
65), as an oil (1.63 g, 62.6%). IR (neat): 3281, 1692, 1651 cm-1. UV
1
(acetonitrile): 233 nm (ꢀ ) 1000 dm3 mol-1 cm-1). H NMR (400
MHz, CDCl3): δ 0.01 (3H, s), 0.04 (3H, s), 0.84 (9H, s), 0.86 (3H, t,
J1H-1H ) 6.6 Hz), 1.17 (3H, d, J1H-1H ) 6.1 Hz), 1.26 (8H, m), 1.59
(2H, m), 2.13 (2H, t, J1H-1H ) 7.4 Hz), 2.57 (1H, dd, J1H-1H ) 14.2,
4.9 Hz), 2.74 (1H, dd, J1H-1H ) 14.2, 10.6 Hz), 3.0 (2H, m), 3.42 (2H,
m), 4.27 (1H, m), 5.80 (1H, s). 13C NMR (100 MHz, CDCl3): δ -5.1,
-4.5, 14.0, 17.95, 22.6, 23.8, 25.6, 25.7, 28.7, 29.0, 29.2, 31.6, 36.7,
39.4, 53.8, 65.9, 173.25, 198.1. ESI-HRMS calcd for C20H42NO3SSi
[M + H]+, 404.2655; found, 404.2671. Anal. Calcd for C20H41NO3-
SSi: C, 59.5; H, 10.24; N, 3.47; S, 7.94. Found: C, 59.38; H, 10.36;
N, 3.49; S, 7.88.
S-[2-(Octanoylamino)ethyl] [3-13C1,3-2H1]-3-{tert-Butyldimethyl-
silyloxy}butanoate. This was prepared as described for unlabeled
material from benzyl [3-13C1,3-2H1]-3-tert-butyldimethylsilyloxybu-
tanoate (2.58 g, 53.1%). 1H NMR (400 MHz, CDCl3): δ 0.02 (3H, s),
0.04 (3H, s), 0.85 (9H, s), 0.86 (3H, t, J1H-1H ) 6.8 Hz), 1.17 (3H, d,
J1H-13C ) 4.2 Hz), 1.26 (8H, m), 1.59 (2H, m), 1.61 (1H, s), 2.13 (2H,
t, J1H-1H ) 7.2 Hz), 2.57 (1H, dd, J13C-1H ) 3.7, J1H-1H ) 14.1 Hz),
2.74 (1H, dd, J13C-1H ) 6.0, J1H-1H ) 14.1 Hz), 3.0 (2H, m), 3.42 (2H,
m), 5.77 (1H, s). 13C NMR (100 MHz, CDCl3): δ -5.05, -4.5, 14.1,
18.0, 22.6, 23.7 (d, J13C-13C ) 39.7 Hz), 25.7, 25.7, 31.7, 36.75, 39.4,
53.7 (d, J13C-13C ) 38.2 Hz), 65.6 (enriched, t, J13C-13C ) 21.3 Hz),
173.3, 198.1.
1
mol-1 cm-1). H NMR (400 MHz, CDCl3): δ 1.42 (9H, s), 2.59 (4H,
m), 5.12 (2H, s), 7.34 (5H, m). 13C NMR (100 MHz, CDCl3): δ 28.0,
29.4, 30.3, 66.4, 80.7, 128.2, 128.5, 135.8, 171.4, 172.25. ESI-HRMS
calcd for C15H20O4Na [M + Na]+, 287.1259; found, 287.1260. Anal.
Calcd for C15H20O4: C, 68.16; H, 7.63. Found: C, 67.80; H, 7.65.
1-Benzyl 4-tert-Butyl [1-13C1]Succinate. This was prepared as
described for the unlabeled material from benzyl [1-13C1]acetate (4.74
g, 58.6%). 1H NMR (400 MHz, CDCl3): δ 1.41 (9H, s), 2.58 (4H, m),
5.12 (2H, d, J1H-13C ) 3.1 Hz), 7.34 (5H, m). 13C NMR (100 MHz,
CDCl3): δ 28.0, 29.3 (d, J13C-13C ) 58.4 Hz), 30.3 (d, J13C-13C ) 1.5
Hz), 66.4 (d, J13C-13C ) 2.5 Hz), 80.7, 128.2, 128.5, 138.8 (d, J13C-13C
) 2.3 Hz), 171.4, 172.3 (enriched).
tert-Butyl Succinate (24). Succinate 23 (5.2 g, 19.7 mmol) was
dissolved in dry THF (40 mL). Pd-C catalyst (10%, 0.52 g) was added
and stirred under a H2 atmosphere (balloon) for 3 h. The reaction
mixture was filtered through Celite, and the Celite was washed with
CHCl3 (200 mL). The filtrate was concentrated to give the product as
an oil which crystallized upon refrigeration (2.87 g, 84.5%). The product
was identical to authentic material.
tert-Butyl [1-13C1]Succinate. This was prepared as described for
S-[2-(Octanoylamino)ethyl] 3-Hydroxybutanethioate (22). Bu-
tanoate 21 (1.63 g, 3.9 mmol) was dissolved in THF (13 mL), water
(13 mL), and AcOH (42 mL), and stirred at room temperature for 48
h. The mixture was partitioned between water (100 mL) and CHCl3
(200 mL). The organic phase was recovered, dried with MgSO4, filtered,
and concentrated. The product was recovered by chromatography,
eluting with EtOAc-hexanes (6:4), as a white solid (0.93 g, 82.8%).
1
the unlabeled material (2.51 g, 81.1%). H NMR (400 MHz, CDCl3):
δ 1.42 (9H, s), 2.52 (2H, m), 2.61 (2H, m), 10.35 (1H, br s). 13C NMR
(100 MHz, CDCl3): δ 28.0, 29.1 (d, J13C-13C ) 56.5 Hz), 30.05 (d,
J13C-13C ) 2.3 Hz), 81.0, 171.4, 178.6 (enriched).
1-Benzyl 6-tert-Butyl 3-Oxohexanedioate (25). tert-Butyl succinate
(3.45 g, 20.1 mmol) was dissolved in dry THF (20 mL). Recrystallized
9
2900 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002

Nonactin Biosynthesis
A R T I C L E S
CDI (3.25 g, 20.1 mmol) was added, and the reaction was stirred at
room temperature, forming a solution of the imidazolide in 10 min. A
clean, dry, three-neck flask equipped with a thermometer and addition
funnel was charged with dry THF (70 mL) and diisopropylamine (9.4
mL, 57.2 mmol) and was placed in a dry ice-acetone bath to cool.
n-Butyllithium (56.2 mmol) in hexanes (21.6 mL) was added slowly
and then stirred until the temperature dropped to below -70 °C. Benzyl
acetate (5.2 mL, 36.0 mmol) in THF (36 mL) was added dropwise
from the addition funnel, maintaining the temperature below -70 °C.
The reaction was stirred for 5 min while the imidazolide solution was
transferred into the addition funnel via cannula. The imidazolide solution
was added dropwise, maintaining the temperature below -70 °C, and
then stirred for 5 min before the reaction was quenched by addition of
water (100 mL), and allowed to warm to room temperature. The mixture
was partitioned between water (200 mL) and Et2O (200 mL). The Et2O
portion was collected, dried with Na2SO4, filtered, and concentrated.
The product was recovered by chromatography, eluting with EtOAc-
) 43.2 Hz), 128.1, 128.1, 128.5, 135.8, 169.1 (d, J13C-13C ) 1.0 Hz),
172.8 (d, J13C-13C
) 4.0 Hz).
tert-Butyl 3-[5,5-Dimethyl-2-(2-{[2-(octanoylamino)ethyl]sulfa-
nyl}-2-oxoethyl)-1,3-dioxan-2-yl]propanoate (27). Compound 26
(2.26 g, 5.77 mmol) was dissolved in dry THF (10 mL). Pd-C catalyst
(10%, 0.23 g) was added and stirred at room temperature under a H2
atmosphere (balloon) for 3 h. The suspension was filtered through
Celite, the Celite was washed with CHCl3 (100 mL), and the filtrate
was concentrated. The residue was dissolved in dry CH2Cl2 (20 mL).
N-Caprylcysteamine (1.78 g, 8.8 mmol), DCC (1.80 g, 8.8 mmol), and
DMAP (0.21 g, 1.7 mmol) were added, and the mixture was stirred at
room temperature for 18 h. The suspension was diluted with CHCl3
(50 mL), water (30 mL), and aqueous saturated NaHCO3 (30 mL) and
stirred. Iodine, to remove the excess thiol, was added in small portions
to the stirred mixture until a red-yellow color persisted; aqueous
saturated Na2S2O3 solution was then added until the color just
disappeared. The layers were separated, and the organic portions were
dried with MgSO4, filtered, and concentrated. The product was
recovered by chromatography on silica gel, eluting with EtOAc-
hexanes (2:3), as an oil (1.4 g, 51.6%). IR (liquid film): 3298, 1727,
1682, 1650 cm-1. UV (acetonitrile): 236 nm (ꢀ ) 4000 dm3 mol-1
cm-1). 1H NMR (400 MHz, CDCl3): δ 0.83 (3H, s), 0.84 (3H, t, J1H-1H
) 6.8 Hz), 1.05 (3H, s), 1.25 (8H, m), 1.40 (9H, s), 1.57 (2H, m), 2.12
(4H, m), 2.40 (2H, t, J1H-1H ) 7.4 Hz), 3.01 (2H, t, J1H-1H ) 7.4 Hz),
hexanes (1:9), as an oil (3.83 g, 62.3%). IR (liquid film): 1720 cm-1
.
UV (acetonitrile): 251 nm (ꢀ ) 410 dm3 mol-1 cm-1). 1H NMR (400
MHz, CDCl3): δ 1.41 (9H, s), 2.50 (2H, t, J1H-1H ) 6.6 Hz), 2.77
(2H, t, J1H-1H ) 6.2 Hz), 3.53 (2H, s), 5.16 (2H, s), 7.34 (5H, m). 13
C
NMR (100 MHz, CDCl3): δ 28.0, 29.1, 49.2, 37.55, 67.1, 80.75, 128.3,
128.4, 128.55, 135.2, 166.8, 171.5, 200.9. ESI-HRMS calcd for
C17H22O5Na [M + Na]+, 329.1365; found, 329.1352. Anal. Calcd for
C17H22O5: C, 66.65; H, 7.24. Found: C, 66.48; H, 7.28.
3.06 (2H, s), 3.40 (2H, q, J1H-1H ) 5.8 Hz), 3.42 (2H, d, J1H-1H
)
1-Benzyl 6-tert-Butyl [2,3-13C2]-3-Oxohexanedioate. This was
10.7 Hz, 2H), 3.60 (d, J1H-1H ) 12.0 Hz), 6.15 (1H, t, J14N-1H ) 5.2
Hz). 13C NMR (100 MHz, CDCl3): δ 14.0, 22.4, 22.55, 22.9, 25.6,
28.0, 29.0, 29.2, 29.6, 31.65, 32.4, 36.5, 39.1, 46.2, 70.5, 80.1, 97.8,
172.95, 173.5, 192.2. ESI-HRMS calcd for C25H45NO6SNa [M + Na]+,
510.2865; found, 510.2881. Anal. Calcd for C25H45NO6S: C, 61.57;
H, 9.3; N, 2.87; S, 6.58. Found: C, 61.75; H, 9.33; N, 2.99; S, 6.65.
S-[2-(Octanoylamino)ethyl] [2,3-13C2]-3-[2-(2-tert-Butyloxy-2-oxo-
ethyl)-5,5-dimethyl-1,3-dioxan-2-yl]propanethioate. This was pre-
pared as described for the unlabeled material (0.66 g, 78.6%). 1H NMR
(400 MHz, CDCl3): δ 0.83 (3H, s), 0.84 (3H, t, J1H-1H ) 6.8 Hz),
1.05 (3H, s), 1.25 (8H, m), 1.40 (9H, s), 1.57 (2H, m), 2.12 (4H, m),
2.39 (2H, dt, J1H-1H ) 7.3 Hz, J1H-13C ) 3.4 Hz), 3.02 (2H, t, J1H-1H
) 5.8 Hz), 3.02 (2H, dd, J1H-13C ) 130.0, 5.5 Hz), 3.40 (2H, q, J1H-1H
) 6.3 Hz), 3.41 (2H, dd, J1H-1H ) 12.0, J1H-13C )5.8 Hz), 3.60 (2H,
dd, J1H-13C ) 1.6 Hz, J1H-1H ) 11.8), 6.16 (1H, t, J14N-1H ) 5.5). 13C
NMR (100 MHz, CDCl3): δ 14.0, 22.4, 22.55, 22.9, 25.6, 28.0, 29.0,
29.2 (d, J13C-13C ) 2.7 Hz), 29.55 (d, J13C-13C ) 2.3 Hz), 31.6, 32.3
(dd, J13C-13C ) 47.7, 3.4 Hz), 36.5, 39.05, 47.3 (enriched, d, J13C-13C
) 42.4 Hz), 70.51, 80.1, 98.9 (enriched, d, J13C-13C ) 42.3 Hz), 174.1
(d, J13C-13C ) 3.8 Hz), 174.6, 196.3.
6-{[2-(Octanoylamino)ethyl]sulfanyl}-4,6-dioxohexanoic acid (28).
Thioester 27 (1.42 g, 2.98 mmol) was dissolved in acetone (30 mL).
pTsOH (0.19 g, 1.0 mmol) was added, and the reaction was stirred at
reflux for 24 h. The cooled mixture was partitioned between water (50
mL) and EtOAc (3 × 100 mL). The pooled EtOAc fractions were dried
over MgSO4, filtered, and concentrated. The product was recovered
by chromatography on silica gel, eluting with EtOAc-hexanes-AcOH
(40:58:2). The column eluate was concentrated down to a minimal
volume, mostly comprised of AcOH, and the product was then
precipitated out by addition of water (50 mL). The suspension was
refrigerated, filtered, washed with 0.1 M HCl, and then dried under
vacuum to give a white solid (0.28 g, 26.7%). IR (powder-ATR): 3283,
1719, 1689, 1632, 1556 cm-1. UV (acetonitrile): 236 (ꢀ ) 3200 dm3
mol-1 cm-1), 281 nm (79). 1H NMR (400 MHz, CD3OD): δ 0.89 (3H,
prepared as described for the unlabeled material from tert-butyl [1-13C1]-
1
succinate and benzyl [2-13C1]acetate (1.32 g, 63.0%). H NMR (400
MHz, CDCl3): δ 1.41 (9H, s), 2.50 (2H, m), 2.77 (2H, m), 3.53 (2H,
dd, J1H-13C ) 130.5, 6.3 Hz), 5.16 (2H, s), 7.34 (5H, m). 13C NMR
(100 MHz, CDCl3): δ 28.0, 29.1 (dd, J13C-13C ) 2.1, 0.96 Hz), 37.55
(dd, J13C-13C ) 41.2, 14.11 Hz), 49.2 (keto, enriched, d, J13C-13C
38.5 Hz), 67.1 (d, J13C-13C ) 0.95 Hz), 80.8, 89.2 (enol, enriched, d,
J13C-13C ) 72.4 Hz), 128.3, 128.4, 128.6, 135.2, 166.9 (d, J13C-13C
)
)
2.3 Hz), 171.5 (d, J13C-13C ) 2.5 Hz), 177.2 (enol, enriched, d, J13C-13C
) 73.1 Hz), 201.0 (keto, enriched, d, J13C-13C ) 38.5 Hz).
tert-Butyl 3-{2-[2-(Benzyloxy)-2-oxoethyl]-5,5-dimethyl-1,3-di-
oxan-2-yl}propanoate (26). Hexanedioate 25 (1.11 g, 3.6 mmol),
neopentyl glycol (4.49 g, 43.1 mmol), and PPTS (1.45 g, 5.8 mmol)
were dissolved in benzene (30 mL), and the mixture was heated at
reflux for 20 h with azeotropic removal of water by means of a Dean-
Stark trap. The mixture was concentrated and resuspended in Et2O (100
mL) and washed first with aqueous saturated NaHCO3 solution (100
mL) and then brine (100 mL). The organic phase was dried, filtered,
and concentrated; the dioxane product was recovered by chromatog-
raphy on silica gel, eluting with EtOAc-hexanes (1:9), as an oil (0.71
g, 50.4%). IR (liquid film): 1723 cm-1. UV (acetonitrile): 257 nm (ꢀ
) 190 dm3 mol-1 cm-1). 1H NMR (400 MHz, CDCl3): δ 0.81 (3H, s),
1.05 (3H, s), 1.41 (9H, s), 2.17 (2H, t, J1H-1H ) 8.1 Hz), 2.41 (2H, t,
J1H-1H ) 8.1 Hz), 2.85 (2H, s), 3.41 (2H, d, J1H-1H ) 12.0 Hz), 3.60
(2H, d, J1H-1H ) 11.7 Hz), 5.12 (2H, s), 7.34 (5H, m). 13C NMR (100
MHz, CDCl3): δ 22.3, 22.9, 28.0, 29.4, 29.5, 32.2, 38.0, 66.4, 70.4,
79.8, 97.9, 128.1, 128.4, 135.7, 169.05, 172.7. ESI-HRMS calcd for
C22H32O6Na [M + Na]+, 415.2097; found, 415.2098. Anal. Calcd for
C22H32O6: C, 67.32; H, 8.22. Found: C, 67.53; H, 8.33.
tert-Butyl [2,3-13C2]-3-{2-[2-(Benzyloxy)-2-oxoethyl]-5,5-dimethyl-
1,3-dioxan-2-yl}propanoate. This was prepared as described for the
unlabeled material from 1-benzyl 6-tert-butyl [2,3-13C2]-3-oxohex-
anedioate (0.69 g, 40.8%; 53.3% based on recovered starting material).
1H NMR (400 MHz, CDCl3): δ 0.82 (3H, s), 1.05 (3H, s), 1.41 (9H,
s), 2.18 (2H, m), 2.41 (2H, m), 2.86 (2H, dd, J1H-13C ) 130.1, 5.7 Hz),
3.42 (2H, dd, J1H-13C ) 5.7 Hz, J1H-1H ) 11.8), 3.61 (2H, dd, J1H-13C
) 2 Hz, J1H-1H ) 11.8), 5.12 (2H, s), 7.34 (5H, m). 13C NMR (100
MHz, CDCl3): δ 22.4, 22.9, 28.0, 29.4, 29.6 (d, J13C-13C ) 2.5 Hz),
32.2 (dd, J13C-13C ) 47.7, 3.8 Hz), 38.05 (enriched, d, J13C-13C ) 43.2
Hz), 66.5 (d, J13C-13C ) 1.0 Hz), 70.4, 79.9, 97.9 (enriched, d, J13C-13C
t, J1H-1H ) 6.8 Hz), 1.30 (8H, m), 1.58 (2H, m), 2.15 (2H, t, J1H-1H
)
7.4 Hz), 2.53 (2H, t, J1H-1H ) 6.5 Hz), 2.83 (2H, t, J1H-1H ) 6.5 Hz),
3.05 (2H, t, J1H-1H ) 6.5 Hz), 3.34 (2H, q, J1H-1H ) 6.5 Hz). 13C NMR
(100 MHz, CDCl3): δ 14.1, 23.3, 26.6, 28.2, 29.5, 29.8, 29.9, 32.6,
36.7, 38.2, 39.4, 175.7, 176.2, 193.0, 202.4. ESI-HRMS calcd for
C16H28NO5S [M + H]+, 346.1688; found, 346.1698. Anal. Calcd for
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2901

A R T I C L E S
Nelson and Priestley
C16H27NO5S: C, 55.63; H, 7.88; N, 4.05; S, 9.28. Found: C, 55.58;
H, 7.88; N, 3.96; S, 9.13.
enriched, dt, J13C-13C ) 70.2, 25.2 Hz, J13C-2H ) 20.0 Hz), 175.7 (enol,
enriched, d, J13C-13C ) 70.2 Hz), 175.7, 176.2, 193.0 (d, J13C-13C
2.3 Hz), 202.4 (keto, enriched, d, J13C-13C ) 37.4 Hz).
)
6-{[2-(Octanoylamino)ethyl]sulfanyl}-[4,5-13C2]-4,6-dioxohex-
anoic acid. This was prepared as described for the unlabeled material
1
(0.14 g, 30.3%). H NMR (400 MHz, CD3OD): δ 0.89 (3H, t, J1H-1H
Acknowledgment. We acknowledge the financial support of
the NIH (CA77347) and NSF (MCB-0111054). M.E.N. thanks
the AFPE for a graduate fellowship. We also thank one of the
referees for insightful and helpful comments.
) 6.8 Hz), 1.30 (8H, m), 1.57 (2H, m), 2.15 (2H, t, J1H-1H ) 7.4 Hz),
2.53 (2H, m), 2.83 (2H, m), 3.04 (2H, t, J1H-1H ) 6.5 Hz), 3.34 (2H,
t, J1H-1H ) 6.5 Hz). 13C NMR (100 MHz, CDCl3): δ 14.1, 23.35, 26.6,
28.2, 29.5, 29.8, 29.9, 32.6, 36.7, 38.2 (dd, J13C-13C ) 41.2, 13.7 Hz),
39.4, 57.1 (keto, enriched, sept, J13C-13C ) 20.0 Hz), 99.4 (enol,
JA016965F
9
2902 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002