Efficient syntheses of 13C-labelled erythromycin biosynthetic intermediates. 2: (2S,3S,4S,5R,6R,7R)-3,6,7-trihydroxy-2,4,6-trimethyl[1- 13C]nonan-5-olide and S-2-acetylaminoethyl (2R,3S,4S,5R,6S,7R)-3,5,6, 7-tetrahydroxy-2,4,6-trimethyl [1-13C]nonanethioate

Article abstract of DOI:10.1002/jlcr.1506

The ¹³C-labelled putative erythromycin biosynthetic intermediates, ((2S,3S,4S,5R,6R,7R)-3,6,7-trihydroxy-2,4,6-trimethyl[1- ¹³C]nonan-5-olide and S-2-acetylaminoethyl (2R,3S,4S,5R,6S,7R)-3,5,6, 7-tetrahydroxy-2,4,6-trimethyl[1-¹³C]nonanethioate), which would be useful for the investigation of the chain elongation mechanism in erythromycin biosynthesis, were efficiently synthesized via aldol condensation of aldehyde derived from (2S,3R,4R,5R)-tert-butyldimethylsilyloxy-5-3,4-O-isopropylidene-2, 4-dimethylheptanol, which was obtained in our previous work on erythromycin A synthesis, and sodium [1-¹³C]propionate (after conversion to ester). Copyright

Full text of DOI:10.1002/jlcr.1506

Research Article
Received 3 October 2007,
Revised 22 January 2008,
Accepted 2 February 2008
Published online 20 March 2008 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1506
13
Efficient syntheses of C-labelled
erythromycin biosynthetic intermediates.
2: (2S,3S,4S,5R,6R,7R)-3,6,7-trihydroxy-2,4,
6-trimethyl[1-13C]nonan-5-olide and
S-2-acetylaminoethyl (2R,3S,4S,5R,6S,7R)
-3,5,6,7-tetrahydroxy-2,4,6-trimethyl
[1-13C]nonanethioate
Ã
Katsumi Iida,a Masahiro Kajiwara,a Mineo Fukui{,b Tadashi Nakatay,b and
Takeshi Oishizb
The 13C-labelled putative erythromycin biosynthetic intermediates, ((2S,3S,4S,5R,6R,7R)-3,6,7-trihydroxy-2,4,6-tri-
methyl[1-13C]nonan-5-olide and S-2-acetylaminoethyl (2R,3S,4S,5R,6S,7R)-3,5,6,7-tetrahydroxy-2,4,6-trimethyl[1-13C]nona-
nethioate), which would be useful for the investigation of the chain elongation mechanism in erythromycin biosynthesis,
were efficiently synthesized via aldol condensation of aldehyde derived from (2S,3R,4R,5R)-tert-butyldimethylsilyloxy-5-3,4-
O-isopropylidene-2,4-dimethylheptanol, which was obtained in our previous work on erythromycin A synthesis, and sodium
[1-13C]propionate (after conversion to ester).
Keywords: 13C-labelling synthesis; erythromycin biosynthetic intermediate; sodium [1-13C]propionate
Introduction
Results and discussion
Strategy for 13C-labelling of synthesis of the putative
erythromycin biosynthetic intermediates
Erythromycin A, a 14-membered macrolide antibiotic pro-
duced by Saccharopolyspora erythraea, is widely used in
clinical medicine. Cane et al. obtained much information
about the chain elongation process in the erythromycin
biosynthetic pathways from feeding experiments with 2H-,
13C- and/or 18O-labelled compounds in S. erythraea.1–5 In
addition, groups led by Leadlay and Katz have employed a
genetic approach to study the biosynthesis of 6-deoxyery-
thronolide B, the first biosynthetic macrolide intermediate of
erythromycin A, in S. erythraea.6–9 We are interested in the
biosynthetic pathways to 6-deoxyerythronolide B, especially
the chain elongation mechanism in erythromycin biosynthesis.
In this connection, we have reported an efficient 13C-labelling
synthesis of the triketide S-2-acetylaminoethyl (2R,3R,4R,5R)-
3,5-diacetoxy-2,4-dimethyl-4-([13C]methoxy)heptanethioate,
which we have proposed to be an erythromycin biosynthetic
intermediate.10
Although syntheses of biosynthetic intermediates of erythro-
mycin have been reported,11 the routes are complex and
unsuitable for efficient 13C-labelling. Therefore, we planned to
develop a more efficient synthetic strategy to obtain the 13C-
labelled compounds by utilizing optically active synthetic
{Deceased.
yPresent address: Department of Chemistry, Faculty of Science, Tokyo University
of Science, 1-3 Kagurazaka, Shinjuku-ku, Tokyo 162-8601, Japan.
zRetired.
aDepartment of Medicinal Chemistry, Meiji Pharmaceutical University, 2-522-1
Noshio, Kiyose-shi, Tokyo 204-8588, Japan
bRIKEN (The Institute of Physical and Chemical Research), 2-1 Hirosawa, Wako-
shi, Saitama 351-0198, Japan
Here, we describe the efficient 13C-labelling synthesis of
two more polyketides, which are putative erythromycin
biosynthetic intermediates, as a continuation of our previous
work.10
*Correspondence to: Katsumi Iida, Department of Medicinal Chemistry, Meiji
Pharmaceutical University, 2-522-1 Noshio, Kiyose-shi, Tokyo 204-8588, Japan.
E-mail: iida@my-pharm.ac.jp
J. Label Compd. Radiopharm 2008, 51 218–225
Copyright r 2008 John Wiley & Sons, Ltd.

K. Iida et al.
intermediates obtained in our previous work.12,13 We have predominantly affords the 3,4-syn-product according to the
already described an efficient 13C-labelling synthesis of the Felkin–Anh model,15–17 we chose the propionic acid ester, which
putative erythromycin biosynthetic intermediate, S-2-acetylami- should predominantly afford the 2,3-anti-product on diaster-
noethyl (2R,3R,4R,5R)-3,5-diacetoxy-2,4-dimethyl-4-([13C]meth- eoselective aldol condensation with 6 having C-2-b-Me, in order
oxy)heptanethioate.10
to obtain the product having the same absolute configuration
As we have previously obtained several aldehydes such as as C-8 of 1. As Heathcock’s ester 7 has been reported to
3,12,13 whose absolute configurations are identical to the C- predominantly afford the 2,3-anti-product on diastereoselective
10–C-13 asymmetric carbon atoms of erythromycin A (1), here aldol condensation with aldehyde,18 the diastereoselective aldol
we attempted to synthesize 2a, which is equivalent to the C- condensation of aldehydes 6 and 7 was carried out in the
7–C-21 segment of 1 (inside the dotted square in Scheme 1), as presence of LDA at ꢀ1001C for 30 min. This reaction predomi-
a
13C-labelled putative erythromycin biosynthetic intermediate. nantly afforded 8a in a 54% yield from 6 with the 8a/8b ratio of
The basic strategy for synthesis of 2a was to employ 9:1, as expected.19 Thioester 9a was obtained from 8a via
diastereoselective aldol condensation of an aldehyde such as silylation with TMSOTf, DIBAL-H reduction, oxidation with
3 with sodium [1-13C]propionate (4) (after conversion to the ruthenium (IV) oxide, and thioesterification with 2-acetylami-
ester).
noethanethiol in a 30% overall yield. The desilylation of 9a with
As shown in Scheme 2,14 we prepared aldehyde 6 from ester hydrogen fluoride/pyridine did not afford 2a (unlabelled), but
5 obtained in our previous work12 via silylation with trimethyl- generated the d-lactone 10a (unlabelled)20 in an 80% yield,
silyl trifluoromethanesulfonate (TMSOTf), diisobutylaluminum analogously to the result in our previous paper.10 That is, the
hydride (DIBAL-H) reduction, and pyridium dichromate oxida- chair form of 10a (unlabelled) shown in Scheme 2 might be the
tion in a 66% overall yield. As diastereoselective aldol lowest-energy chair conformation, as the C-2–Me and C-
condensation of an aldehyde branched at C-2 with ester 5–C(OH)(Me)CH(OH)Et moieties can adopt equatorial positions.
Scheme 1. Strategy for synthesis of 13C-labelled putative erythromycin biosynthetic intermediate 2a from 3 and 4.
Scheme 2. Synthesis of 2a (unlabelled) from 5 and 7.
J. Label Compd. Radiopharm 2008, 51 218–225
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

K. Iida et al.
Thus, 10a (unlabelled) might be generated from 9a through the atmosphere at room temperature for 30 min gave ester 14 in a
attack of the 5-hydroxyl oxygen on the carbonyl carbon with 79% yield from 4. Oxidation of alcohol 13 with pyridinium
elimination of 2-acetylaminoethanethioxyl after desilylation. dichromate (PDC) in the presence of zeolite (MS-4A)23 gave
Therefore, as the synthesis of 2a (unlabelled) appeared to be aldehyde 12 at room temperature for 1.5 h in a 92% yield. The
difficult, we considered the synthesis of 10a as an alternative. aldol condensation of 12 and 14 in the presence of LDA under
Moreover, as shown in Scheme 3, we attempted to synthesize an argon atmosphere at ꢀ781C for 30 min gave a mixture of
the tetraol and thioester 2b, even though 2b has the opposite products. The mixture could be separated by silica gel
absolute configuration at C-8 of 1. We thought that the d- chromatography to afford a product (6% yield), a mixture
lactone might not be generated from 2b after desilylation, (67% yield) of two inseparable products in equal quantity, based
owing to 1,3-diaxial interaction between C-2-Me and C-4-Me. on the 1H NMR spectrum, and another product (18% yield).
Thus, 2a, which can be derived from 10a, and 2b should be These results indicate that all four products were diastereomers
obtainable from esters 11a and 11b, respectively. Further, 11a (11a, 11b, 11c, and 11d) with regard to the newly generated
and 11b should be obtainable by aldol condensation of asymmetric carbon atoms at C-2 and C-3.
aldehyde 12 having C-2-b-Me derived from alcohol 13 (obtained The absolute configurations at C-2 and C-3 of these four
in our previous work12), with benzyl ester 14 derived from products were determined by the consideration of both the
sodium [1-13C]propionate (4).
Zimmerman–Traxler and the Felkin–Anh models for this aldol
condensation and the analysis of the 1H NMR spectra of the four
products (unlabelled), which were separately synthesized from
12 and 14 (unlabelled), and their derivatives. Firstly, 14 treated
with LDA might be transformed to (E)- and (Z)-enolates of 14 in
similar amounts. On the basis of the Zimmerman–Traxler model,
the aldol condensation with (E)-enolate predominantly affords
the 2,3-anti-product, whereas that with (Z)-enolate predomi-
nantly affords the 2,3-syn-product. Therefore, the aldol con-
densation with 12 having C-2-b-Me was examined. In the case of
the (E)-enolate of 14, the major product was the 2,3-anti-3,4-syn-
product 11a (2S,3S) and the minor product was the 2,3-anti-3,4-
anti-product 11d (2R,3R), as predicted by the Felkin–Anh model.
In the case of the (Z)-enolate of 14, the major product was the
2,3-syn-3,4-syn-product 11b (2R,3S) and the minor product was
Synthesis of the putative erythromycin biosynthetic inter-
mediates
As shown in Scheme 4, [1-13C]propionyl chloride was prepared
from sodium [1-13C]propionate (4) treated with phthaloyl
chloride under an argon atmosphere at 1501C for 1.5 h on the
basis of the methods described in a previous paper.4 Although
esterification of benzyl alcohol with 4 was carried out by using
benzyl alcohol and 4 (unlabelled) in the presence of n-
butyllithium under an argon atmosphere at ꢀ781C for 40 min
in the Evans method,21 ester 14 (unlabelled)22 was obtained in
only a 62% yield from 4 (unlabelled). However, esterification of
benzyl alcohol with 4 in the presence of Et3N under an argon
Scheme 3. Strategy for syntheses of 13C-labelled putative erythromycin biosynthetic intermediates 2b and 10a from 4 and 13.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 218–225

K. Iida et al.
Scheme 4
Scheme 5
the 2,3-syn-3,4-anti-product 11c (2S,3R), as predicted. Our spectrum of the mixture of two inseparable unlabelled products,
unpublished 1H NMR data for similar compounds indicated that obtained from aldol condensation of 12 and 14 (unlabelled), led
the 3,4-anti-form showed J2,3 = 2–4 Hz and the 3,4-syn-form us to postulate that these products were 11a (unlabelled) and
showed J2,3 = 9–10 Hz, irrespective of the absolute configuration 11b (unlabelled), i.e. the 3,4-syn-form. Thus, we designated this
at C-2. The values of J2,3 = 8.8, 8.9, 9.0, and 9.6 Hz in the 1H NMR mixture as 11a,b (unlabelled), and we similarly designated the
J. Label Compd. Radiopharm 2008, 51 218–225
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

K. Iida et al.
mixture of two inseparable products obtained from the aldol VALORA-III FT-IR spectrometer. MS spectra were obtained with a
condensation of 12 and 14 as 11a,b. On the other hand, the JEOL JMS-DX-302 spectrometer.
values of J2,3 = 3.4, 3.5 Hz and J2,3 = 1.7, 4.9 Hz in the 1H NMR
spectra of the two minor unlabelled products led us to postulate Benzyl [1-13C]propionate (14)
that these products were 11c (unlabelled) and 11d (unlabelled),
Sodium [1-13C]propionate (4) (1 g, 10.4 mmol) and phthaloyl
i.e. the 3,4-anti-form. Further, as shown in Scheme 5, Swern
chloride (2 ml, 13.9 mmol) were mixed and heated at 1501C for
1.5 h under an argon atmosphere. The resulting [1-13C]propionyl
chloride was distilled into dry CH2Cl2 (2 ml). To a solution of
benzyl alcohol (1.3 ml, 12.6 mmol) and Et3N (7 ml, 50.2 mmol) in
dry CH2Cl2 (2 ml) was added dropwise the above solution of
[1-13C]propionyl chloride in dry CH2Cl2 under an argon atmo-
sphere at 01C, and the whole was stirred for 30 min at room
temperature. The reaction mixture was diluted with Et2O, and
washed with 10% HCl, sat. NaHCO3 aq., and brine, dried over
anhydrous MgSO4 and evaporated. Chromatography of the
crude product on silica gel with Et2O/hexane (2:1) gave 14
oxidation24 of the 3-hydroxy ester 11a,b (unlabelled) with
DMSO, trifluoroacetic anhydride, and Et3N under an argon
atmosphere at ꢀ781C for 30 min gave the 3-oxo ester 18a,b in
an 81% yield. Zinc borohydride (Zn(BH4)2) reduction of 3-oxo
ester affords the 2,3-syn-product,25 and its application (argon
atmosphere, 01C, 2.5 h) to 18a,b gave the 2,3-syn-3-hydroxy
esters 11b (unlabelled), which was identified on the basis of
J
which was identified on the basis of J2,3 = 3.4, 3.5 Hz, in 35 and
2,3 = 8.8, 8.9 Hz in its 1H NMR spectrum, and 11c (unlabelled),
1
45% yields, respectively. By comparison of the H NMR spectra
of the two minor unlabelled products, which were generated
3
(1.35 g, 79%), 1H NMR (CDCl3) d: 1.17 (dt, 3H, J1H13C = 5.5 Hz,
1
from 12 and 14 (unlabelled), with the H NMR spectrum of 11c
J = 7.6 Hz, 3-H3), 2.39 (quintet, 2H, J = 7.6 Hz, 2-H2), 5.12 (d, 2H,
3J1H13C = 3.1 Hz, phenyl-CH2), 7.35 (m, 5H, phenyl-H5); 13C NMR
(CDCl3) d: 174.3 (1-13C); FAB-MS (glycerol) m/z: 166 (MH1).
(unlabelled), which was synthesized via Swern oxidation and
Zn(BH4)2 reduction of 11a,b (unlabelled), these two minor
unlabelled products were identified as 11c (unlabelled) and 11d
(unlabelled). Moreover, as shown in Scheme 4, the hydrogeno-
lysis of 11a,b on palladium–carbon at room temperature for 2 h
afforded two acids that were separable by silica gel chromato-
(2R,3R,4R,5R)-5-tert-Butyldimethylsilyloxy-3,4-O-isopropyli-
dene-2,4-dimethylheptanal (12)
1
graphy, 15a and 15b, in a 45% yield. By comparison of the H
To a solution of (2S,3R,4R,5R)-tert-butyldimethylsilyloxy-5-3,4-O-
isopropylidene-2,4-dimethylheptanol (13) (4.17 g, 12.0 mmol) in
dry CH2Cl2 (20 ml) was added zeolite (MS-4A, 9.2 g) at room
temperature. To this suspension was added PDC (9.2 g,
24.5 mmol) at 01C, and the whole was stirred for 1.5 h at room
temperature. The reaction mixture was diluted with Et2O and
filtered through Florisil and Celite. Repeated elution with Et2O
gave 12 (3.83 g, 92%), 1H NMR (CDCl3) d: 0.07 (s, 3H,
isopropylidene-CH3), 0.10 (s, 3H, isopropylidene-CH3), 0.89 (s,
9H, SiC(CH3)3), 0.98 (t, 3H, J = 7.6 Hz, 7-H3), 1.16 (s, 3H, 4-CH3),
1.24 (d, 3H, J = 7.3 Hz, 2-CH3), 1.37 (s, 3H, SiCH3), 1.40 (s, 3H,
SiCH3), 1.48 (m, 1H, 6-H), 1.79 (m, 1H, 6-H), 2.73 (m, 1H, 2-H), 3.59
(t, 1H, J = 5.6 Hz, 5-H), 4.35 (d, 1H, J = 5.9 Hz, 3-H), 9.64 (d, 1H,
J = 2.5 Hz, 1-H).
NMR spectra of the two acids 15a (unlabelled) and 15b
(unlabelled) prepared from 11a,b (unlabelled) by hydrogeno-
lysis with that of 15b (unlabelled), which was prepared from
hydrogenolysis of 11b (unlabelled) synthesized via Swern
oxidation and zinc borohydride reduction of 11a,b (unlabelled),
the absolute configurations of the two acids could be identified.
Consequently, we had established the absolute configuration of
all four products generated from the aldol condensation of 12
and 14. As shown in Scheme 4, the product obtained in a 6%
yield was identified as 11c, that obtained in a 67% yield was
identified as a mixture of 11a and 11b in equal quantity, and
that obtained in an 18% yield was identified as 11d.
Deprotection and cyclization of 15a in a 48% HF/CH3CN (1:9) at
room temperature for 6 h directly gave the desired d-lactone 10a
in a 50% yield. Thioesterification4,26 of 15b with 2-acetylami-
noethanethiol in DMF in the presence of diphenylphosphonyl
azide and Et3N at room temperature for 19h gave the thioester
16b in a 93% yield. Finally, deprotection of 16b in 48% HF/CH3CN
(1:9) at room temperature for 8 h gave the diol 17b (14% yield)
and the tetraol 2b (73% yield), and repeated reaction of
recovered 17b gave further 2b in a 61% yield. Finally, the
desired 2b was obtained from 15b in a 76% overall yield. Thus,
we had obtained the desired 13C-labelled putative erythromycin
biosynthetic intermediates 10a and 2b.
Benzyl (2S,3S,4S,5R,6R,7R)- and benzyl (2R,3S,4S,5R,6R,7R)-
7-tert-butyldimethylsilyloxy-3-hydroxy-5,6-O-isopropylidene-
2,4,6-trimethyl[1-13C]nonanoate (11a,b), (2S,3R,4S,5R,6R,7R)
(11c) and (2R,3R,4S,5R,6R,7R) (11d)
To a solution of diisopropylamine (1.3 ml, 9.28 mmol) in dry THF
(12 ml) was added dropwise n-butyllithium (1.4 M in hexane,
5.9 ml, 8.26 mmol) at ꢀ781C under an argon atmosphere, and
the mixture was stirred for 5 min at 01C. To this solution was
added dropwise a solution of 14 (1.35 g, 8.18 mmol) in dry THF
(5 ml) at ꢀ781C, and the solution was stirred for 30 min. To this
solution was added dropwise
a solution of 12 (3.83 g,
Experimental
11.1 mmol) in dry THF (5 ml) at ꢀ781C, and the whole was
stirred for 30 min at this temperature. The reaction was
quenched with sat. NH4Cl aq. and extracted with Et2O. The
combined extract was washed with brine, dried over anhydrous
MgSO4, and evaporated. Chromatography of the crude product
on silica gel with Et2O/hexane (1:4) gave 11c (259 mg, 6%) and
11a,b (2.80 g, 67%), and further elution with Et2O:hexane (1:2)
Materials and instruments
Sodium [1-13C]propionate (4) (99 atom% 13C) was purchased
from Cambridge Isotope Laboratories. (2S,3R,4R,5R)-tert-Butyldi-
methylsilyloxy-5-3,4-O-isopropylidene-2,4-dimethylheptanol (2)
(495% e.e.) synthesized in our previous work12 was used. All
other chemicals were of analytical grade and commercially
available. 1H NMR and 13C NMR spectra were recorded on a JEOL
GX-400, GX-500, or GSX-400 (1H: 400 or 500 MHz and 13C:
100 MHz) spectrometer. IR spectra were recorded on a Jasco
1
gave 11d (739 mg, 18%), data of 11a,b; H NMR (CDCl3) d: 0.08
(s, 3H, isopropylidene-CH3), 0.08 (s, 3H, isopropylidene-CH3), 0.90
(s, 9H, SiC(CH3)3), 0.94 (t, 3H, J = 7.5 Hz, 9-H3), 1.01 (d, 3H,
J = 7.0 Hz, 4-CH3), 1.30 (dd, 3H, 3J1H13C = 5.0 Hz, J = 6.9 Hz, 2-CH3),
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 218–225

K. Iida et al.
1.34 (s, 3H, 6-CH3), 1.39 (s, 3H, SiCH3), 1.58 (s, 3H, SiCH3), 1.66, (m, CH3), 1.28 (d, 3H J = 6.4 Hz, 2-CH3), 1.36 (s, 3H, SiCH3), 1.41 (s, 3H,
1H, 8-H), 1.80 (m, 1H, 4-H), 2.68 (m, 1H, 2-H), 3.54 (dd, 1H, J = 4.6, SiCH3), 1.43, (m, 1H, 8-H), 1.78 (m, 1H, 8-H), 2.01 (br, 1H, 4-H),
6.4 Hz, 7-H), 3.82 (d, 1H, J = 9.2 Hz, 3-H), 4.04 (d, 1H, J = 3.4 Hz, 5- 2.57 (br, 1H, 2-H), 3.56 (br, 1H, 7-H), 3.92 (br, 1H, 3-H), 4.03 (br,
H), 5.09 (m, 2H, phenyl-CH2), 7.35 (m, 5H, phenyl-H5); 13C NMR 1H, 5-H).
(CDCl3) d: 174.6 (1-13C); FAB-MS (glycerol) m/z: 510 (MH1), data
of 11a,b (unlabelled); 1H NMR (CDCl3) d: 0.08 (s, 3H, (2S,3S,4S,5R,6R,7R)-3,6,7-Trihydroxy-2,4,6-trimethyl[1-13C]-
nonan-5-olide (10a)
isopropylidene-CH3), 0.09 (s, 3H, isopropylidene-CH3), 0.90 (s,
9H, SiC(CH3)3), 0.95 (t, 3H, J = 7.6 Hz, 9-H3), 1.01 (d, 3H, J = 6.8 Hz,
4-CH3), 1.02 (d, 3H, J = 7.2 Hz, 4-CH3), 1.10 (s, 3H, 6-CH3), 1.30 (d,
3H, J = 6.8 Hz, 2-CH3), 1.33 (s, 3H, SiCH3), 1.39 (s, 3H, SiCH3), 1.48
(m, 1H, 8-H), 1.67 (m, 1H, 8-H), 1.79 (m, 1H, 4-H), 1.97 (m, 1H, 4-
H), 2.67 (dq, 1H, J = 6.8, 9.0 Hz, 2-H), 2.83 (dq, 1H, J = 6.8, 8.8 Hz,
2-H), 3.54 (t, 1H, J = 5.5 Hz, 7-H), 3.59 (t, 1H, J = 5.5 Hz, 7-H), 3.82
(dd, 1H, J = 2.1, 8.9 Hz, 3-H), 3.96 (dd, 1H, J = 1.6, 9.6 Hz, 3-H), 4.04
(d, 1H, J = 3.4 Hz, 5-H), 4.05 (d, 1H, J = 3.9 Hz, 5-H), 5.10 (m, 2H,
phenyl-CH2), 7.34 (m, 5H, phenyl-H5), data of 11c (unlabelled); 1H
NMR (CDCl3) d: 0.07 (s, 3H, isopropylidene-CH3), 0.10 (s, 3H,
isopropylidene-CH3), 0.90 (s, 9H, SiC(CH3)3), 0.96 (t, 3H, J = 7.4 Hz,
9-H3), 0.96 (d, 3H, J = 6.8 Hz, 4-CH3), 1.19 (s, 3H, 6-CH3), 1.20 (d,
3H, J = 8.1 Hz, 2-CH3), 1.34 (s, 3H, SiCH3), 1.41 (s, 3H, SiCH3), 1.70
(m, 1H, 8-H), 1.84 (m, 1H, 4-H), 2.74 (dq, 1H, J = 3.4, 7.3 Hz, 2-H),
3.57 (t, 1H, J = 5.4 Hz, 7-H), 3.89 (dd, 1H, J = 3.5, 8.5 Hz, 3-H), 4.39
(s, 1H, 5-H), 5.08 (m, 2H, phenyl-CH2), 7.35 (m, 5H, phenyl-H5),
data of 11d (unlabelled); 1H NMR (CDCl3) d: 0.07 (s, 6H,
2 ꢁ isopropylidene-CH3), 0.90 (s, 9H, SiC(CH3)3), 0.98 (t, 3H,
J = 7.5 Hz, 9-H3), 1.10 (d, 3H, J = 9.8 Hz, 4-CH3), 1.10 (s, 3H, 6-CH3),
1.33 (s, 3H, SiCH3), 1.34 (d, 3H, J = 11.2 Hz, 2-CH3), 1.41 (s, 3H,
SiCH3), 1.76 (m, 1H, 8-H), 2.02 (m, 1H, 4-H), 2.67 (dq, 1H, J = 1.7,
7.3 Hz, 2-H), 3.55 (t, 1H, J = 5.4 Hz, 7-H), 3.60 (d, 1H, J = 4.9 Hz, 3-
H), 3.90 (d, 1H, J = 4.2 Hz, 5-H), 5.10 (m, 2H, phenyl-CH2), 7.35 (m,
5H, phenyl-H5).
A solution of 15a (389 mg, 0.927 mmol) in 48% HF/CH3CN (1:9,
9.5 ml) was stirred for 6 h at room temperature, then added
dropwise to a suspension of NaHCO3 in CH2Cl2, and the whole
was stirred for 30 min at room temperature. To this suspension
was added anhydrous MgSO4 and the whole was stirred for
30 min at room temperature. The suspension was filtered and
the filtrate was evaporated. Chromatography of the crude
product on silica gel with AcOEt/hexane (2:1) gave 10a
1
(114 mg, 50%), H NMR (CDCl3) d: 1.08 (t, 3H, J = 7.3 Hz, 9-H3),
1.13 (s, 3H, 6-CH3), 1.18 (d, 3H, J = 7.3 Hz, 4-CH3), 1.33 (dd, 3H,
3J1H13C = 5.0 Hz, J = 7.2 Hz, 2-CH3), 1.40 (m, 1H, 8-H), 1.72 (m, 1H,
2
8-H), 2.22 (m, 1H, 4-H), 2.72 (ddt, 1H, J1H13C = 15.2 Hz, J = 3.7,
7.6 Hz, 2-H), 3.40 (dt, 1H, J = 1.8, 11.0 Hz, 7-H), 3.88 (dt, 1H,
J = 3.7, 11.9 Hz, 3-H), 4.97 (d, 1H, J = 2.8 Hz, 5-H); 13C NMR
(CDCl3) d: 174.1 (1-13C); IR (CHCl3) cmꢀ1: 3447, 2973, 2938, 1685,
1457, 1382, 1331, 1160, 1115, 979; FAB-MS (glycerol) m/z: 248
(MH1).
S-2-Acetylaminoethyl (2R,3S,4S,5R,6R,7R)-7-tert-butyldi-
methylsilyloxy-3-hydroxy-5,6-O-isopropylidene-2,4,6-
trimethyl[1-13C]nonanethioate (16b)
To a solution of 15b (588 mg, 1.40 mmol) and 2-acetylami-
noethanethiol (1.53 g, 12.8 mmol) in DMF (800 ml) was added
diphenylphosphonyl azide (900 ml, 4.19 mmol) under an argon
atmosphere. To this solution was added Et3N (1.2 ml, 8.61 mmol)
at 01C, and the whole was stirred for 19 h at room temperature.
The reaction mixture was quenched with sat. NH4Cl aq. and
extracted with Et2O. The combined extract was washed with
10% Na2CO3 aq. and brine, dried over anhydrous MgSO4, and
evaporated. Chromatography of the crude product on silica gel
with AcOEt/hexane (2:1) gave 16b (861 mg, 93%), 1H NMR
(CDCl3) d: 0.09 (s, 3H, isopropylidene-CH3), 0.10 (s, 3H,
isopropylidene-CH3), 0.91 (s, 9H, SiC(CH3)3), 0.97 (t, 3H,
J = 7.6 Hz, 9-H3), 1.06 (d, 3H, J = 7.0 Hz, 4-CH3), 1.17 (s, 3H, 6-
(2S,3S,4S,5R,6R,7R)-7-tert-Butyldimethylsilyloxy-3-hydroxy-
5,6-O-isopropylidene-2,4,6-trimethyl[1-13C]nonanoic acid
(15a) and (2R,3S,4S,5R,6R,7R) (15b)
To a solution of 11a,b (4.88 g, 9.57 mmol) in MeOH (50 ml) was
added 10% palladium–carbon (500 mg), and the mixture was
stirred for 2 h at room temperature under a hydrogen atmo-
sphere. The reaction mixture was filtered through a Celite pad
and evaporated. Chromatography of the crude product on silica
gel with AcOEt/hexane (1:1) gave 15b (1.8 g, 45%) and further
elution with CHCl3:MeOH (10:1) gave 15a (1.8 g, 45%), data of
3
CH3), 1.32 (dd, 3H, J1H13C = 5.4 Hz, J = 6.7 Hz, 2-CH3), 1.34 (s, 3H,
1
15a; H NMR (CDCl3) d: 0.07 (s, 3H, isopropylidene-CH3), 0.09 (s,
3H, isopropylidene-CH3), 0.89 (s, 9H, SiC(CH3)3), 0.98 (t, 3H,
J = 7.5 Hz, 9-H3), 1.03 (d, 3H, J = 6.7 Hz, 4-CH3), 1.21 (s, 3H, 6-CH3),
SiCH3), 1.41 (s, 3H, SiCH3), 1.41, (m, 1H, 8-H), 1.73 (m, 1H, 8-H),
1.87 (m, 1H, 4-H), 1.97 (s, 3H, COCH3), 2.87 (m, 1H, 2-H), 2.95 (m,
1H, SCH), 3.06 (m, 1H, SCH), 3.42 (m, 2H, SCH2CH2), 3.56 (dt, 1H,
J = 4.9, 5.8 Hz, 7-H), 3.88 (d, 1H, J = 7.6 Hz, 3-H), 4.04 (d, 1H,
J = 3.4 Hz, 5-H), 5.75 (br, 1H, NH); 13C NMR (CDCl3) d: 202.3
(1-13C); FAB-MS (glycerol) m/z: 521 (MH1).
3
1.25 (dd, 3H, J1H13C = 4.9 Hz, J = 7.0 Hz, 2-CH3), 1.35 (s, 3H,
SiCH3), 1.43 (s, 3H, SiCH3), 1.78, (m, 1H, 8-H), 2.02 (br, 1H, 4-H),
2.56 (br, 1H, 2-H), 3.50 (br, 1H, 7-H), 3.90 (br, 1H, 3-H), 4.12 (br,
1H, 5-H); 13C NMR (CDCl3) d: 177.5 (1-13C); FAB-MS (glycerol) m/z:
420 (MH1), data of 15b; 1H NMR (CDCl3) d: 0.08 (s, 3H,
isopropylidene-CH3), 0.09 (s, 3H, isopropylidene-CH3), 0.89 (s, 9H,
SiC(CH3)3), 0.97 (t, 3H, J = 7.6 Hz, 9-H3), 1.05 (d, 3H, J = 6.7 Hz, 4-
S-2-Acetylaminoethyl (2R,3S,4S,5R,6R,7R)-3,7-dihydroxy-5,6-
O-isopropylidene-2,4,6-trimethyl[1-13C]nonanethioate (16b)
and S-2-acetylaminoethyl (2R,3S,4S,5R,6R,7R)-3,5,6,7-tetra-
hydroxy-2,4,6-trimethyl[1-13C]nonanethioate (7b)
3
CH3), 1.20 (s, 3H, 6-CH3), 1.28 (dd, 3H, J1H13C = 4.9 Hz, J = 7.0 Hz,
2-CH3), 1.36 (s, 3H, SiCH3), 1.43 (s, 3H, SiCH3), 1.43, (m, 1H, 8-H),
1.75 (m, 1H, 8-H), 2.01 (m, 1H, 4-H), 2.71 (q, 1H, J = 7.3 Hz, 2-H),
3.57 (t, 1H, J = 5.5 Hz, 7-H), 3.94 (m, 1H, 3-H), 4.08 (d, 1H,
J = 2.4 Hz, 5-H); 13C NMR (CDCl3) d: 177.5 (1-13C); FAB-MS
(glycerol) m/z: 420 (MH1), data of 15b (unlabelled); 1H NMR
(CDCl3) d: 0.08 (s, 3H, isopropylidene-CH3), 0.09 (s, 3H,
isopropylidene-CH3), 0.88 (s, 9H, SiC(CH3)3), 0.98 (t, 3H,
J = 7.6 Hz, 9-H3), 1.02 (d, 3H, J = 6.4 Hz, 4-CH3), 1.20 (s, 3H, 6-
A solution of 16b (441 mg, 0.847 mmol) in 48% HF/CH3CN (1:9,
10 ml) was stirred for 8 h at room temperature, then added
dropwise to a suspension of NaHCO3 in CH2Cl2, and the whole
was stirred for 30 min at room temperature. To this suspension
was added anhydrous MgSO4 and the whole was stirred for
30 min at room temperature. The suspension was filtered and
the filtrate was evaporated. Chromatography of the crude
J. Label Compd. Radiopharm 2008, 51 218–225
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

K. Iida et al.
product on silica gel with CHCl3/MeOH (30:1) gave 17b (47 mg,
11b (unlabelled) was transformed to 15b (unlabelled) via the
1
14%), and further elution with CHCl3:MeOH (7:1) gave 2b above-mentioned procedure, data of 15b (unlabelled); H NMR
(227 mg, 73%). 17b (279 mg, 0.688 mmol) was transformed to 2b (CDCl3) d: 0.08 (s, 3H, isopropylidene-CH3), 0.09 (s, 3H,
(155 mg, 61%) for 7 h by the same procedure, data of 2b; 1H isopropylidene-CH3), 0.88 (s, 9H, SiC(CH3)3), 0.98 (t, 3H,
NMR (CDCl3) d: 1.06 (t, 3H, J = 7.3 Hz, 9-H3), 1.06 (s, 3H, 6-CH3), J = 7.6 Hz, 9-H3), 1.02 (d, 3H, J = 6.4 Hz, 4-CH3), 1.20 (s, 3H, 6-
3
1.12 (d, 3H, J = 7.0 Hz, 4-CH3), 1.31 (dd, 3H, J1H13C = 5.5 Hz, CH3), 1.28 (d, 3H, J = 6.4 Hz, 2-CH3), 1.36 (s, 3H, SiCH3), 1.41 (s, 3H,
J = 6.7 Hz, 2-CH3), 1.44 (m, 1H, 8-H), 1.59 (m, 1H, 8-H), 1.83 (m, 1H, SiCH3), 1.43, (m, 1H, 8-H), 1.78 (m, 1H, 8-H), 2.01 (br, 1H, 4-H),
4-H), 1.97 (s, 3H, COCH3), 2.90 (m, 1H, 2-H), 3.02 (q, 2H, SCH2), 2.57 (br, 1H, 2-H), 3.56 (br, 1H, 7-H), 3.92 (br, 1H, 3-H), 4.03 (br,
3.34 (dd, 1H, J = 1.8, 10.7 Hz, 7-H), 3.44 (m, 2H, SCH2CH2), 4.04 (d, 1H, 5-H).
1H, J = 1.8 Hz, 5-H), 5.88 (br, 1H, NH); 13C NMR (CDCl3) d: 202.9
(1-13C); IR (CHCl3) cmꢀ1: 3447, 3007, 2978, 2937, 2879, 1656, Conclusion
1529, 1456, 1375, 1093, 1045, 974, 960, 924; FAB-MS (glycerol)
We efficiently synthesized 13C-labelled putative erythromycin
biosynthetic intermediates, 10a and 2b, via aldol condensation
J = 7.3 Hz, 9-H3), 1.08 (d, 3H, J = 7.0 Hz, 4-CH3), 1.14 (s, 3H, 6-CH3),
m/z: 367 (MH1), data of 17b; 1H NMR (CDCl3) d: 1.03 (t, 3H,
of aldehyde derived from the optically active synthetic inter-
mediate 13 (obtained in our previous work for erythromycin A
synthesis) and benzyl ester 14 derived from sodium [1-13C]pro-
pionate (4) for investigations on the biosynthesis of erythromycin.
1.32 (s, 6H, 2 ꢁ isopropylidene-CH3), 1.35 (br, 3H, 2-CH3), 1.67 (br,
2H, 8-H2), 2.90 (m, 1H, 2-H), 3.02 (m, 2H, SCH2), 3.41 (m, 1H, 7-H),
3.44 (m, 2H, SCH2CH2), 3.99 (d, 1H, J = 5.2 Hz, 5-H), 4.22 (m, 1H, 3-
H), 5.84 (br, 1H, NH).
11b (unlabelled) and 11c (unlabelled) from 11a,b (unla-
belled) via 18a,b, and 15b (unlabelled) from 11b (unla-
belled)
To a solution of DMSO (180 ml, 2.54 mmol) in dry CH2Cl2 (4 ml)
was added trifluoroacetic anhydride (240 ml, 1.70 mmol) at
ꢀ781C under an argon atmosphere, and the mixture was stirred
Supplementary electronic material for this paper is available in
Wiley Interscience at http://www.interscience.wiley.com/jpages/
1099–1034/suppmat/
for 15 min. To this solution was added a solution of 11a,b
(unlabelled) (440 mg, 0.865 mmol) in dry CH2Cl2 (2 ml) at ꢀ781C,
and the whole was stirred for 20 min. To this solution was added
Et3N (540 ml, 3.87 mmol), and the whole was stirred for 30 min.
The reaction was quenched with sat. NH4Cl aq. and extracted
with Et2O. The combined extract was washed with brine, dried
over anhydrous MgSO4, and evaporated. Chromatography of
the crude product on silica gel with Et2O/hexane (1:20) gave
benzyl (2RS,4S,5R,6R,7R)-7-tert-butyldimethylsilyloxy-5,6-O-iso-
propylidene-3-oxo-2,4,6-trimethylnonanoate (18a,b) (354 mg,
81%). To a solution of 18a,b (135 mg, 0.266 mmol) in dry Et2O
(3.3 ml) was added Zn(BH4)2 (0.2 M in Et2O, 3.3 ml, 0.66 mmol) at
01C under an argon atmosphere, and the mixture was stirred for
2.5 h. The reaction was quenched with water (170 ml) at 01C, and
the whole was stirred for 15 min. To this suspension was added
sat. NH4Cl aq. (170 ml) at room temperature, and the whole was
stirred for 30 min. This solution was dried over anhydrous
MgSO4 with stirring for 30 min and filtered, and the filtrate was
evaporated. Chromatography of the crude product on silica gel
with Et2O/hexane (1:10) gave 11c (unlabelled) (61 mg, 45%) and
11b (unlabelled) (47 mg, 35%), data of 11b (unlabelled); 1H NMR
(CDCl3) d: 0.08 (s, 3H, isopropylidene-CH3), 0.08 (s, 3H,
isopropylidene-CH3), 0.89 (s, 9H, SiC(CH3)3), 0.95 (t, 3H,
J = 7.6 Hz, 9-H3), 1.01 (d, 3H, J = 6.8 Hz, 4-CH3), 1.09 (s, 3H, 6-
CH3), 1.30 (d, 3H, J = 6.8 Hz, 2-CH3), 1.33 (s, 3H, SiCH3), 1.39 (s, 3H,
SiCH3), 1.65 (m, 2H, 8-H2), 1.79 (m, 1H, 4-H), 2.68 (dq, 1H, J = 6.8,
8.8 Hz, 2-H), 3.54 (dd, 1H, J = 4.6, 6.1 Hz, 7-H), 3.82 (dd, 1H, J = 1.8,
8.9 Hz, 3-H), 4.04 (d, 1H, J = 3.2 Hz, 5-H), 5.10 (m, 2H, phenyl-CH2),
7.36 (m, 5H, phenyl-H5), data of 11c (unlabelled); 1H NMR (CDCl3)
d: 0.07 (s, 3H, isopropylidene-CH3), 0.10 (s, 3H, isopropylidene-
CH3), 0.90 (s, 9H, SiC(CH3)3), 0.96 (t, 3H, J = 7.4 Hz, 9-H3), 0.96 (d,
3H, J = 6.8 Hz, 4-CH3), 1.19 (s, 3H, 6-CH3), 1.20 (d, 3H, J = 8.1 Hz, 2-
CH3), 1.34 (s, 3H, SiCH3), 1.41 (s, 3H, SiCH3), 1.70 (m, 2H, 8-H2),
1.84 (m, 1H, 4-H), 2.74 (dq, 1H, J = 3.4, 7.3 Hz, 2-H), 3.57 (t, 1H,
J = 5.4 Hz, 7-H), 3.89 (dd, 1H, J = 3.5, 8.5 Hz, 3-H), 4.39 (s, 1H, 5-H),
5.08 (m, 2H, phenyl-CH2), 7.35 (m, 5H, phenyl-H5).
References
[1] D. E. Cane, H. Hasler, T.-C. Liang, J. Am. Chem. Soc. 1981, 103,
5960–5962.
[2] D. E. Cane, H. Hasler, P. B. Taylor, T.-C. Liang, Tetrahedron 1983, 39,
3449–3455.
[3] D. E. Cane, T.-C. Liang, P. B. Taylor, C. Chang, C.-C. Yang, J. Am.
Chem. Soc. 1986, 108, 4957–4964.
[4] D. E. Cane, C.-C. Yang, J. Am. Chem. Soc. 1987, 109, 1255–1257.
[5] D. E. Cane, P. C. Prabhakaran, W. Tan, W. R. Ott, Tetrahedron Lett.
1991, 32, 5457–5460.
[6] J. Cortes, S. F. Haydock, G. A. Roberts, D. J. Bevitt, P. F. Leadlay,
Nature 1990, 348, 176–178. DOI: 10.1038/348176a0.
[7] J. S. Tuan, J. M. Weber, M. J. Staver, J. O. Leung, S. Donadio, L. Katz,
Gene 1990, 90, 21–29.
[8] D. J. Bevitt, J. Cortes, S. F. Haydock, P. F. Leadlay, Eur. J. Biochem.
1992, 204, 39–49.
[9] S. Donadio, M. J. Staver, J. B. McAlpine, S. J. Swanson, L. Katz,
Science 1991, 252, 675–679.
[10] See the accompanying paper.
[11] S. Masamune, P. A. McCarthy, Total synthesis of macrolides. In
Macrolide Antibiotics-Chemistry, Biology, and Practice, Chapter 4
(Ed.: S. Omura), Academic Press, New York, 1984, pp. 127–198.
[12] T. Nakata, M. Fukui, T. Oishi, Tetrahedron Lett. 1988, 29, 2219–2222.
[13] T. Nakata, M. Fukui, T. Oishi, Tetrahedron Lett. 1988, 29, 2223–2226.
[14] The experiments shown in Scheme 2 are described in the
supplemental data.
[15] D. A. Evans, J. V. Nelson, T. R. Taber, In Topics in Stereochemistry,
Vol. 13 (Eds.: N. L. Allinger, E. L. Eliel, S. H. Wilen), Wiley, New York,
1982, pp. 1–115.
[16] C. H. Heathcock, Comprehensive Carbanion Chemistry. In Studies
in Organic Chemistry, Vol. 5, Part B, Chapter 4 (Eds.: E. Buncel,
T. Durst), Elsevier, Amsterdam, 1984, pp. 177–237.
[17] C. H. Heathcock, In Asymmetric Synthesis, Vol. 3 (Ed.: J. D.
Morrison), Academic Press, New York, 1984, pp. 111–212.
[18] M. C. Pirrung, C. H. Heathcock, J. Org. Chem. 1980, 45, 1727–1728.
[19] 1H NMR (CDCl3) data of 8a; d: 0.14 (s, 9H, Si(CH3)3), 0.15 (s, 9H,
Si(CH3)3), 0.17 (s, 9H, Si(CH3)3), 0.91 (d, 3H, J = 7.1 Hz, 4-CH3), 0.92 (t,
3H, J = 7.6 Hz, 9-H3), 1.23 (s, 3H, 6-CH3), 1.31 (d, 3H, J = 7.1 Hz, 2-
CH3), 1.44 (m, 1H, 8-H), 1.56 (m, 1H, 8-H), 2.07 (m, 1H, 4-H), 2.18 (s,
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 218–225

K. Iida et al.
6H, Ph(CH3)2), 2.66 (d, 1H, J = 10.3 Hz, 3-OH), 2.85 (dq, 1H, J = 7.1,
J = 7.2 Hz, 2-CH3), 1.40 (m, 1H, 8-H), 1.72 (m, 1H, 8-H), 2.22 (m, 1H,
4-H), 2.72 (dt, 1H, J = 3.7, 7.6 Hz, 2-H), 3.40 (dt, 1H, J = 1.8, 11.0 Hz,
7-H), 3.88 (dt, 1H, J = 3.7, 11.9 Hz, 3-H), 4.97 (d, 1H, J = 2.8 Hz, 5-H).
[21] D. A. Evans, J. Bartroli, T. L. Shih, J. Am. Chem. Soc. 1981, 103,
2127–2129.
[22] F. J. Leeper, S. E. Shaw, P. Satish, Can. J. Chem. 1994, 72, 131–141.
[23] J. Herscovici, K. Antonakis, Chem. Comm. 1980, 561–562.
[24] S. L. Huang, K. Omura, D. Swern, Synthesis 1978, 297–298.
[25] N. Nakata, T. Oishi, Tetrahedron Lett. 1980, 21, 1641–1644.
[26] S. Yamada, Y. Yokoyama, T. Shioiri, J. Org. Chem. 1974, 39,
3302–3303.
9.5 Hz, 2-H), 3.49 (dd, 1H, J = 2.2, 9.5 Hz, 7-H), 3.80 (d, 1H, J = 3.2 Hz,
5-H), 3.93 (d, 1H, J = 10.0 Hz, 3-H), 7.04 (s, 3H, phenyl-H3), that of
8b; d: 0.11 (s, 9H, Si(CH3)3), 0.12 (s, 9H, Si(CH3)3), 0.16 (s, 9H,
Si(CH3)3), 0.85 (t, 3H, J = 7.3 Hz, 9-H3), 0.94 (d, 3H, J = 6.8 Hz, 4-CH3),
1.20 (s, 3H, 6-CH3), 1.48 (m, 2H, 8-H2), 1.54 (d, 3H, J = 7.3 Hz, 2-CH3),
1.89 (m, 1H, 4-H), 2.15 (s, 6H, Ph(CH3)2), 3.00 (d, 1H, J = 10.3 Hz, 3-
OH), 3.09 (dq, 1H, J = 7.3, 9.0 Hz, 2-H), 3.33 (dd, 1H, J = 3.2, 9.3 Hz, 3-
H), 3.44 (t, 1H, J = 6.3 Hz, 7-H), 4.15 (d, 1H, J = 3.2 Hz, 5-H), 7.04 (s,
3H, phenyl-H3).
[20] 1H NMR (CDCl3) data of 10a (unlabelled); d: 1.08 (t, 3H, J = 7.3 Hz, 9-
H3), 1.13 (s, 3H, 6-CH3), 1.18 (d, 3H, J = 7.3 Hz, 4-CH3), 1.33 (d, 3H,
J. Label Compd. Radiopharm 2008, 51 218–225
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org