An Expedient Synthesis of LAF389, a Bengamide B Analogue

Article abstract of DOI:10.1021/op0341162

An optimized, convergent, safe synthesis of LAF389 (9), an anticancer agent analogous to bengamide B, is described. Starting from α-D- glucoheptonic (D-glycero-D-gulo-heptonic acid) γ-lactone (10), the lactone 15 was constructed in five steps. Major impro

Full text of DOI:10.1021/op0341162

Organic Process Research & Development 2003, 7, 856−865
An Expedient Synthesis of LAF389, a Bengamide B Analogue
David D. Xu, Liladhar Waykole,* John V. Calienni, Lech Ciszewski, George T. Lee, Wenming Liu, Joanna Szewczyk,
Kevin Vargas, Kapa Prasad, Oljan Repicˇ, and Thomas J. Blacklock
Process R & D, Chemical and Analytical DeVelopment, NoVartis Institute for Biomedical Research, One Health Plaza,
East HanoVer, New Jersey 07936, U.S.A.
Abstract:
intermediate 8. It is a much shorter synthesis of bengamide
compared to other published ones. It utilizes commercially
available natural products 1 and 10 which contain all of
stereogenic centers of bengamide B. This synthesis was used
for the preparation of initial batches of the drug substance
to support the preclinical studies.
An optimized, convergent, safe synthesis of LAF389 (9), an anti-
cancer agent analogous to bengamide B, is described. Starting
from r-D-glucoheptonic (D-glycero-D-gulo-heptonic acid) γ-lac-
tone (10), the lactone 15 was constructed in five steps. Major
improvements were made in the preparation of the aldehyde
precursor 14 and subsequent olefination to yield 15 via a
modified Julia protocol. This olefination was significantly
improved by using TMSCl as an additive. The second fragment
of the drug substance, E-caprolactam 7, was obtained in two
one-pot operations from (5R)-5-hydroxy-L-lysine (1). Finally,
opening 15 with 7 using sodium 2-ethyl hexanoate (Na-EH) gave
8, which on deprotection yielded LAF389.
Although short and convergent, this synthesis was not
feasible for large-scale preparation of the drug substance.
The pitfalls in the procedure to prepare 15 included conver-
sion of the R-D-glucoheptonic-γ-lactone to aldehyde 14. It
involved protection and selective deprotection of the hy-
droxyl groups followed by periodate oxidation. The solution
of product 14 was found to be acid- as well as base sensitive
and handling during the conversion to 15 was difficult. The
olefination step itself needed major improvements. The Takai
conditions5 using chromium dichloride and corresponding
gem-diiodide6 were not suitable for scale-up work. Also, the
conversion of lysine 1 to lactam 7 was a tedious multistep
operation and needed to be simplified, and coupling of 7
with 15 must be improved to avoid epimerization at C2 of
lactone 15.
Introduction
LAF389 is an analogue of bengamide B, a member of
the bengamides isolated from Jaspidae sponges.1 Like its
natural congeners, the synthetic bengamide is a general
inhibitor of cell proliferation2 and, hence, was developed as
an anti-cancer agent. Several syntheses of bengamides A,
B, and E have been reported.3 However, all are either too
long or contain unacceptable reagents and operations and
are thus unsuitable for large-scale preparation of LAF389.
To support the ongoing campaign, a short and convergent
synthesis of bengamides B and E has been devised by us
earlier.4 This synthesis of LAF389 involved the coupling of
a polyhydroxylated lactone fragment 15 with a substituted
2-aminocaprolactam 7 (Scheme 1) to construct the key
* Author for correspondence. E-mail: liladhar.waykole@pharma.novartis.com.
(1) (a) Quinoa, E.; Adamczeski, M.; Crews, P.; Bakus G. J. J. Org. Chem.
1986, 51, 4494. (b) Adamczeski, M.; Quinoa, E.; Crews, P. J. Am. Chem.
Soc. 1989, 111, 647. (c) Adamczeski, M.; Quinoa, E.; Crews, P. J. Org.
Chem. 1990, 55, 240. (d) D’Auria, M. V.; Giannini, C.; Minale, L.;
Zampella, A.; Debitus, C.; Frostin, M. J. Nat. Prod. 1997, 60, 814.
(2) For studies on the effect of Bengamides on cell proliferation, see (a) Kinder,
F. R.; Bair, K. W.; Bontempo, J.; Crews, P.; Czuchta, A. M.; Nemzek, R.;
Thale, Z.; Vattay, A.; Versace, R. W.; Weltchek, S.; Wood, A.; Zabludoff,
S. D.; Phillips, P. E. Proc. Am. Assoc. Cancer Res. 2000, 41, 600. (b)
Phillips, P. E.; Bair, K. W.; Bontempo, J.; Crews, P.; Czuchta, A. M.; Kinder,
F. R.; Vattay A.; Versace, R. W.; Wang, B.; Wang, J.; Wood, A. Proc.
Am. Assoc. Cancer Res. 2000, 41, 59.
(3) Bengamide synthesis papers: (a) Broka, C. A.; Ehrler, J. Tetrahedron Lett.
1991, 32, 5907. (b) Mukai, C.; Moharram, S. M.; Kataoka, O.; Hanaoka,
M. J. Chem. Soc., Perkin Trans. 1 1993, 2849. (c) Marshall, J. A.; Luke,
G. P. J. Org. Chem. 1993, 58, 6229. (d) Gurjar, M. K.; Srinivas, N. R.
Tetrahedron Lett. 1991, 32, 3409. (e) Boeckman, R. K., Jr.; Clark, T. J.;
Shook, B. C. Org. Lett. 2002, 4(12), 2109.
Results and Discussion
After brief review of different options,3 we concluded that
further optimization of synthesis shown in Scheme 1 was a
better option. For the sake of discussion this is divided into
three parts: (a) preparation of the lactone 15, (b) preparation
of ꢀ-caprolactam 7, and (c) coupling of 15 with 7 and
conversion to the drug substance.
Preparation of Lactone 15. In the preparation of 15, the
bis-acetonide protection of D-glycero-D-gulo-heptonic acid
γ-lactone 10 was initially carried out in acetone with iodine
as the catalyst.7 This procedure suffered from a long reaction
time, inconsistent regioselectivity, and poor quality of the
isolated product. Therefore, the reaction conditions were
modified, and a much milder and more selective procedure
(4) (a) Kinder, F. R.; Wattanasin, S.; Versace, R. W.; Bair, K. W.; Bontempo,
J.; Green, M. A.; Lu, Y. J.; Marepalli, H. R.; Phillips, P. E.; Roche, D.;
Tran, L. D.; Wang, R.; Waykole, L.; Xu, D. D.; Zabludoff, S. J. Org. Chem.
2001, 66, 2118. (b) Liu, W.; Szewczyk, J. M.; Waykole, L.; Repicˇ, O.;
Blacklock, T. J. Tetrahedron Lett. 2002, 43, 1373.
(5) Okazoe, T.; Takai, K.; Utmoto, K. J. Am. Chem. Soc. 1987, 109, 951.
(6) Friedrich, E. C.; Falling, S. N.; Lyons, D. E. Synth. Commun. 1975, 33.
(7) Kartha, K. P. R. Tetrahedron Lett. 1986, 27, 3415.
856
•
Vol. 7, No. 6, 2003 / Organic Process Research & Development
10.1021/op0341162 CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/23/2003

Scheme 1. Discovery synthesis of LAF389
Scheme 2. D-Glycero-D-gulo-heptonic acid γ-lactone 10 to bis-acetonide 11
was developed, as shown in Scheme 2. We found that
formation of the bis-acetonide of 10 proceeded much more
rapidly in acetone in the presence of 2,2-dimethoxypropane
(DMP) and concentrated sulfuric acid as the catalyst. The
reaction could be run at 0 °C, ensuring consistent regio-
selectivity of about 5:1 (3,5 and 6,7 vs 2,3 and 6,7). The
byproduct, the tris-acetonide, was not observed under these
reaction conditions.
Vol. 7, No. 6, 2003 / Organic Process Research & Development
•
857

Scheme 3. O-Methylation of the hydroxyl group of bis-acetonide 11
Scheme 4. Selective removal of the terminal acetonide group of bis-acetonide 12
Table 1. Optimization of selective acetonide cleavage of
Bis-acetonide 12
In the conversion of 11 to 12, we found that these two
compounds are very sensitive to both acid and base. As a
result, we were forced to find nonhomogeneous conditions
for methylation. Several methylating agents such as dimethyl
carbonate, dimethyl methylphosphonate, (trimethylsilyl)-
diazomethane, dimethyl sulfite, and methyl trifluoromethane-
sulfonate in combination with bases such as K2CO3, KHCO3,
Cs2CO3, and K3PO4 in different solvents were investigated
(Scheme 3). The combination of potassium phosphate (3
equiv, screened through 14 mesh) and dimethyl sulfate (3
equiv) in toluene at room temperature gave the best results.
Product 12 was isolated in 50% yield and in high purity by
filtration of the heterogeneous reaction mixture, followed by
precipitation with heptane. Since the reaction is heteroge-
neous, the particle size of the solid base was found to be
critical for achieving consistent reactivity. Sieving of the
commercial potassium phosphate through screen (mesh 14)
was found to be necessary for large-scale preparation.
Our next objective was to achieve a selective removal of
the terminal acetonide group of 12. The earlier procedure
used 1:1 acetic acid and water to affect the hydrolysis of
the terminal acetonide.8 The reaction was very selective for
removing only the terminal acetonide group. However, the
method necessitated a high-vacuum distillation to remove
water and acetic acid at ambient temperature, and this
resulted in overdeprotection. To avoid this problem, several
solvents and catalysts were screened. It was found that the
6,7-acetonide can be selectively removed with a controlled
amount of water (5 equiv) in propionic acid and a catalytic
amount of p-toluenesulfonic acid in isopropyl acetate (Scheme
4). More importantly, under these reaction conditions,
product 13 was precipitated out directly from the reaction
mixture, thus avoiding vacuum distillation and extractive
workup. The different acids and workup conditions tried are
summarized in Table 1. Initially, the reaction was difficult
to reproduce with reaction time varying significantly from
run to run. It was later found that it was the acid impurity
(from hydrolysis or partial hydrolysis of dimethyl sulfate)
from the previous step that caused the problem. A catalytic
amount of p-toluenesulfonic acid was added to ensure a
acid
entry
(with 5 equiv H2O)
conditions
yield (%)
1
2
3
acetic acid
18 h RT, filtered,
i-PrOAc wash
69
81
72
isobutyric acid
butyric acid
64 h RT, 5 h 50 °C,
cold i-PrOAc wash
19 h RT, 5 h 45 °C,
∼75% conversion,
cold i-PrOAc wash
19 h RT, 5 h 45 °C,
cold i-PrOAc
4
5
6
propionic acid
propionic acid
76
67
83
5 h 50 °C, 60 h RT,
cold i-PrOAc
propionic acid/i-PrOAc, 2 h 40 °C, 0.5 h 0 °C,
1:2, p-TsOH (cat.) cold i-PrOAc
consistent reaction time (Table 1, entry 6). With this
modification, the hydrolysis step was found to be reproduc-
ible and complete within 2-3 h, offering the product in 83%
yield.
Conversion of the diol 13 to aldehyde 14 using sodium
periodate (aqueous) in acetonitrile was found to be prob-
lematic for scale-up. The aldehyde intermediate is thermally
labile, highly sensitive to both acidic and basic conditions
and readily prone to hydration. The reaction conditions were
modified (Scheme 5) using a minimum amount of water to
avoid extraction of the product from water. After completion
of the reaction, the reaction mixture was diluted with
isopropyl acetate and water was absorbed onto magnesium
sulfate. The crude product (mostly in the hydrate form) was
dried and dehydrated by azeotropic distillations of isopropyl
acetate. The desired product 14 was precipitated from the
concentrate in 70% yield.
The olefination of aldehyde 14 to obtain the lactone
fragment 15 was one of the most difficult and low-yielding
step in this synthesis. The Takai procedure5 using a large
excess of chromium dichloride and the expensive and
thermally unstable 1,1-diiodo-2,2-dimethylpropane6 was very
demanding, and the reaction outcome very much depended
upon the quality of the CrCl2. Good results could only be
obtained with a freshly opened bottle of anhydrous CrCl2.
The sensitivity of 14 and 15 to basic solution was found to
(8) Shing, T. K. M.; Zhou, Z. H.; Mak, C. W. J. Chem. Soc., Perkin Trans. 1
1992, 1907.
858
•
Vol. 7, No. 6, 2003 / Organic Process Research & Development

Scheme 5. Oxidative cleavage of diol 13 to aldehyde 14
Table 2. Effect of the additive (with 1.0 equiv of 15)
18
(equiv)
base
(equiv)
additive
(equiv)
addition mode temp
(time)
yield selectivity
(trans/cis)
entry
solvent
1
2
3
4
5
2.0
1.2
1.2
2.0
1.2
n-BuLi (1.0)
THF/MeCN
toluene
TMSCl 2.0
18 f 14 0 °C; 50 °C
50% (3:1)
50% (1:1)
48% (6:1)
50% (6:1)
41% (10:1)
(1 h)
n-BuLi (1.0)
HN(TMS)2 (1.2)
n-BuLi (1.0)
HN(TMS)2 (1.2)
n-BuLi (1.8)
HN(TMS)2 (2.0)
n-BuLi (1.2)
HN(TMS)2 (1.2)
TMSCl 2.0
14 f 18 -75 °C; 50 °C
(1 h)
14 f 18 -75 °C; 50 °C
(1 h)
14 f 18 -75 °C; 50 °C
(1 h)
14 f 18 -75 °C; 50 °C
(1 h)
THF/MeCN
THF/MeCN
THF/MeCN
BF3‚OEt2 1.0
BF3‚OEt2 2.0
BF3‚OEt2 0.12
Scheme 6. Formation of olefin 15 via Julia olefination
be a major hurdle. Therefore, Julia olefination9 was selected
as an option as the reaction conditions are relatively mild.
Olefination, initially, under these conditions gave poor yield
(<10%) again due to the sensitivity of 14 to base. A
breakthrough was realized in a serendipitous manner. It was
found that the reaction actually required an elevated tem-
perature to be completed. When we attempted to remove
the solvent on a rotavap at about 40 °C from a “failed”
reaction in which no desired product was detected (TLC),
the desired product had formed! Since in all the previous
“failed” attempts (in which product yields were <10%), the
aldehyde 14 was always completely consumed, we believed
that the rate-limiting step of Julia olefination was the
elimination step. After the above finding, a systematic
screening of additives and reaction conditions were carried
out. Selected results are listed in Table 2. To summarize,
we found that both TMSCl and BF3 etherate were excellent
additives showing significant improvement in yield. The
sulfone anion was shown to be stable only at low temper-
atures (<-20 °C) and had to be generated at a low
temperature (-55 to -78 °C). The sequence of addition also
showed significant impact on the outcome of the reaction,
with the best results being obtained when TMSCl was mixed
with the sulfone anion in THF at low temperatures (e.g., -70
°C), and this mixture was added to a precooled solution of
aldehyde in MeCN at 0 °C (Scheme 6). The choice of solvent
was limited by the poor solubility of the aldehyde in less
polar solvents. Acetonitrile was a good compromise with
respect to solubility, selectivity, and product yield.
It is postulated that TMSCl plays a dual role by (1)
attenuating the basicity of the reaction mixture and (2)
stabilizing the intermediate (Scheme 7). In the presence of
TMSCl, neither the starting aldehyde nor the reaction product
is exposed to either the sulfone anion or the alkoxide adduct,
and therefore, the decomposition of these compounds through
the elimination pathway is minimized. The cyclized inter-
mediate after the sulfone anion addition had been only
speculated.9 We were able to isolate an intermediate such
as 19 (R ) diphenylmethyl) as a relatively stable white
solid.10 The NMR data is in agreement with structure 19 (R
) diphenylmethyl); however, no effort was made to establish
exact stereochemistry. Julia and co-workers have studied the
relationship between stereochemistry of formation of olefins
(E/Z) from syn/anti heterocyclic â-hydroxysulfone.9 We
further found that the cyclic intermediate, 19 (R ) diphenyl-
methyl), can be selectiVely converted to exclusively trans-
or a mixture of cis/trans olefination products (Scheme 8).
When compound 19 (R ) diphenylmethyl) was subjected
to acidic conditions (e.g., MeSO3H), the trans product was
obtained exclusiVely. On the other hand, when 19 (R )
diphenylmethyl) was treated with base (TBAF, anhydrous
(9) (a) Baudin, J. B.; Hareau, G.; Julia, S. A.; Ruel, O. Tetrahedron Lett. 1991,
32, 1175-1178. (b) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley,
A. Synlett 1998, 26-28. (c) Julia, M.; Badet, B. Bull. Soc. Chim. Fr. 1975,
1363-1366. (d) Baudin, J. B.; Hareau, G.; Julia, S. A.; Ruel, O. Bull. Soc.
Chim. Fr. 1993, 130, 856-878. (e) Blakemore, P. R. J. Chem. Soc., Perkin
Trans. 1 2002, 2563.
Vol. 7, No. 6, 2003 / Organic Process Research & Development
•
859

Scheme 7. Postulated role of TMSCl in the formation of olefin 15
Scheme 8. Selective olefination of “trapped” intermediate
as 19 (R ) diphenylmethyl)
We are investigating this observation currently to make this
method a choice for selective trans/cis olefination. If proved
fruitful, the results will be published later. It was found that
only the trans product crystallized from tert-butyl methyl
ether, simplifying further the product isolation and purifica-
tion. The desired trans isomer was simply crystallized from
the extract aqueous workup. In the current case, a trans/cis
ratio of 4/1 was routinely obtained by simply heating the
reaction mixture.
Preparation of E-Caprolactam. Our next objective was
to improve and simplify the preparation of 7. The previous
process suffered from high dilution (low throughput),
expensive and hazardous reagents (HOBT, EDCI), and low
efficiency (chromatographic separations, high-vacuum distil-
lation). The removal of a large quantity of DMF and the
extraction of fairly water soluble intermediate 4 into meth-
ylene chloride were unacceptable for scale-up. To overcome
these problems, an alternative approach was (Scheme 9)
designed. The idea was to attenuate the reactivity of the
carboxylic acid so that the intermolecular-competing amide
bond formation is suppressed, and the reaction can be run
at higher concentration. The (5R)-5-hydroxy-L-lysine (1) was
easily converted to methyl ester by treatment with TMSCl
in methanol. The critical lactam formation was effected by
a unique cooperative acid-conjugate base catalysis resulting
from the base triethylamine and HCl (from TMSCl). Isolation
of 4 proved to be problematic because of water solubility
and contamination with Et3N-HCl. Therefore, 4 was con-
verted in situ to 6, which could be easily isolated and
purified. A solvent exchange was necessary for a complete
esterification. The major portion of methanol was removed
by distillation, and the residual methanol was azeotropically
distilled with EtOAc. The esterification of 4 with 6 proceeded
smoothly in the presence of a stoichiometric amount of
DMAP. A substoichiometric amount of DMAP resulted in
an incomplete reaction. Other amine bases such as triethyl-
amine, catalytic DMAP, and Na-2-ethylhexanoate11 gave
unsatisfactory results.
reagent grade), the trans/cis product ratio was 1:1. We could
not obtain exclusive cis-olefination product for this substrate
probably because of the bulkiness of the substituent. In the
absence of exact diastereomeric composition of the bicyclic
intermediate and the absolute stereochemistry of the central
carbon having four heteroatoms attached, these results could
not be unambiguously explained. Since formation of trans/
cis olefin will depend not only on the stereochemistry of
substituents but also on which bond cleavage occurs first.
(10) Compound 19 (R ) diphenylmethyl): To the solution of 2-(2,2-dimethyl-
propylsulfonyl)benzothiazole 18 (32.4 g, 0.12 mol) in 400 mL of anhydrous
THF at -78 °C was added a solution of LiN(TMS)2 (130 mL, 0.13 mol).
Reaction mixture was stirred for 1 h at -78 °C and treated dropwise with
TMS-Cl (16.6 mL, 0.13 mol). To the resulting suspension at -78 °C was
added a solution of diphenylacetaldehyde (17.8 mL, 0.1 mol) in 300 mL of
THF over the period of 50 min. The mixture was stirred at -78 °C for 1
h, at 0 °C for 30 min, at RT for 30 min, and at 50 °C for 1 h. The mixture
was quenched with water (300 mL) at 10 °C and concentrated under reduced
pressure (35 °C, 50 mbar) to an aqueous oily phase. Heptane (150 mL)
was added and stirred at 40 °C for 20 min. The crude solid 19 was filtered
and washed with water (30 mL). Filtrate phases were separated, and the
crude trans olefin was in the organic phase (heptane). The crude material
19 was purified by crystallization from MeOH to obtain (18.6 g, 0.034 mol)
as a white crystalline solid. 1H NMR (DMSO): δ 0.07 (s, 9H), 1.13 (s,
9H), 4.15 (s, 1H), 5.02 (d, J ) 10.7 Hz, 1H), 5.43 (d, 10.7 Hz, 1H), 8.49-
7.36 (m, 14H). 13C NMR (CDCl3): δ 0.00, 27.66, 34.13, 57.36, 72.70,
73.48, 122.59, 123.73, 125.97, 126.20, 126.97, 127.13, 127.61, 127.73,
128.12, 128.51, 134.98, 140.67, 141.93, 150.48, 171.06. MS m/z: 538.0
[MH+], 560.2 [M + Na+].
860
•
Vol. 7, No. 6, 2003 / Organic Process Research & Development

Scheme 9. Preparation of protected E-caprolactam 6 from amino acid (1)
Scheme 10. Removal of the tert-butyloxycarbonyl
protecting group from E-caprolactam 6
of 7 suggested the possibility of coordinated acid-base
catalysis for the amide formation. Sodium 2-ethyl hexanoate
was found to be an ideal base in promoting the amide
formation and easy isolation of the product.14 In the presence
of 2 equiv of Na-EH, the coupling of 15 and 7 proceeded
smoothly at room temperature without epimerization (Scheme
11). The product was obtained in 85-92% yield upon
dilution of the reaction mixture with water and heptane. Two
equivalents of the base were necessary for this coupling
reaction. A smaller amount of the base led to isomeric
acetonide-migrated byproduct formation.
For the conversion of 8 to 9, many of the known acid-
catalyzed acetonide deprotection methods were evaluated.
Use of p-TsOH/MeOH, PPTS/MeOH, and HCl/MeOH/
EtOAc resulted in decomposition. Amberlyst-15/THF/
water and TFA/THF/water gave 9 in 80-90% yield, ac-
companied by 5-10% of acetonide migration byproduct.
Aqueous HCl (1 N) and THF (2:1) mixture gave good re-
sults. The reaction proceeded at room temperature and gave
9 with very little byproduct formation (<1% by HPLC,
Scheme 12).
The discovery method used trifluoroacetic acid for the
deprotection of the amino group. The resulting product was
isolated as its free base via basification and chromatography.
This method was very difficult to reproduce on a large scale
because of (a) the difficulty in complete removal of TFA,
(b) the difficulty of extraction of 7 from aqueous solution,
and (c) chromatography. Attempts to achieve the deprotection
using HCl (g) in MeOH/EtOAc (1:1) and HCl (g) in EtOH/
EtOAc (1:6) gave lower yield because of solubility of the
product in solvents. We found that 6 can be cleanly
deprotected under anhydrous conditions in EtOAc using HCl
(Scheme 10). The deprotected product was precipitated
directly from the reaction as its HCl salt.
Coupling 15 with 7 Followed by Deprotection To
Obtain 10. Because of low reactivity between 15 and 7, the
coupling was initially realized under forced high concentra-
tion conditions. These conditions led to the formation of up
to 5% of the epimer12 (at the C-2 position). The removal of
the epimer was achieved by HPLC purification. Further,
despite the high cost of the starting material and the difficulty
in its preparation, 7 was used in excess (>2 equiv) in this
coupling reaction, making the transformation very inefficient.
Early attempts to couple the lactam hydrochloride 7 with
lactone 15 in the presence of inexpensive base sodium 2-ethyl
hexanoate (Na-EH, 1.5 equiv) gave the product 8 in 70-
85% yield along with 5-30% of unknown byproducts. Use
of 2-hydroxypyridine13 (0.4 equiv) to facilitate opening of
lactone and Na-EH (1.5 equiv) as a base gave the product 8
in 90% yield without byproducts. Similarly, use of the
combination of triethylamine (0.5 equiv) and Na-EH (1.5
equiv) gave 8 in 70% yield. Our experience in the formation
Initially, a two-stage crystallization was designed for
achieving the desired purity and the crystal form. MTBE
proved to be the best solvent for this purpose. Although the
process removed most of the impurities, the product was a
solvate with MTBE. A recrystallization from ethanol and
water to furnish the preferred polymorph E and subsequent
exposure to moisture produced the hemihydrate of 10 as the
final drug substance. In a later process, it was found that
removal of the acetonide of 8 could be achieved using 0.5
N HCl in ethanol at ambient temperature. After concentration
of the reaction mixture to 66% of the original volume, the
desired polymorph E of the drug substance 10 precipitated
in high yield and good purity.
Thus, a general acid-base-catalyzed amide formation
(opening of a lactone with an amine) was realized inter-
molecularly between lactone 15 and ꢀ-caprolactam 7. The
mild conditions circumvented the problems of decomposition
and epimerization encountered with the initial procedure.
Also, a simple and much milder method (aq HCl-EtOH) for
the deprotection of the acetonide group in 8 was developed.
(11) Fitt, J.; Prasad, K.; Repicˇ, O.; Blacklock, T. J. Tetrahedron Lett. 1998, 39,
6991.
(12) NMR data (NOE experiments) indicated epimerization at CHOMe position
next to the amide bond.
(13) (a) Swain, C. G.; Brown, J. F., Jr. J. Am. Chem. Soc. 1952, 74, 2538. (b)
Kim, S. Org. Prep. Proced. Int. 1988, 20, 145.
(14) Liu, W.; Xu, D. D.; Repicˇ, O.; Blacklock, T. J. Tetrahedron Lett. 2001, 42,
2439.
Vol. 7, No. 6, 2003 / Organic Process Research & Development
•
861

Scheme 11. Opening of lactone 15 with E-caprolactam 7
Scheme 12. Conversion of 8 to 9
These conditions lead to direct crystallization of the desired
polymorph of drug substance.
precoated plates with a 250 µm layer thickness. Plates were
visualized with either 5% ethanolic phosphomolybdic acid
solution and heat or 254 UV light depending upon substrate.
Specific analysis information and sample preparation is
available upon request.
Materials. The starting material (5R)-5-hydroxy-L-lysine
dihydrochloride monohydrate (1) was purchased from ven-
dor. All other reagents and solvents were purchased from
various commercial sources and used without further puri-
fication. All of the isolated intermediates were analyzed as
indicated above and used directly in subsequent steps. All
reactions were run in an inert nitrogen atmosphere unless
otherwise indicated.
3,5:6,7-Bis-O-(1-methylethylidene)-D-glycero-D-gulo-
heptonic Acid γ-Lactone (11). To a stirred slurry of
D-glycero-D-gulo-heptonic acid γ-lactone 10 (260.0 g, 1.25
mol) in acetone (1870 mL) at 25 °C was added 2,2-
dimethoxypropane (390 g, 461 mL, 3.75 mol). The slurry
was cooled to 0 ( 5 °C, sulfuric acid (28.2 g, 15.5 mL,
0.287 mol) was added, and the mixture was stirred for 8 h.
The solution was quenched with 8% sodium bicarbonate
(1157 mL, gas evolution). The mixture was stirred for 30
min at 10-15 °C (pH ) 8-8.5) and then stored in a
refrigerator for 8 h. The acetone and methanol were dis-
tilled under reduced pressure (50 mmHg, 25 ( 5 °C). Ethyl
acetate was added (2000 mL), and the mixture was stirred
at 25 ( 5 °C for 15 min. The phases were separated, and
the organic layer was washed with dilute aqueous sodium
chloride solution (966 mL) and saturated sodium chloride
solution (380 mL). The organic phase was concentrated at
Summary
In summary, the synthesis of LAF389 has been optimized
and the process has been successfully scaled up in the pilot
plant. Major improvements were made to O-methylation
(K3PO4/Me2SO4) and regioselective hydrolysis of the bis-
acetonide 12. Modified Julia olefination using TMSCl (and
BF3-etherate) offered much-improved reproducibility. These
conditions were thoroughly optimized to obtain the desired
trans isomer via direct crystallization. A multistep, one-pot
process using general acid-conjugate base-catalyzed cycliza-
tion of (5R)-5-hydroxy-L-lysine offered the seven-membered
lactam very efficiently. Also, a mild general acid-base-
catalyzed amide formation (opening of a lactone with an
amine) was realized.
Experimental Section
1
General. H and 13C NMR spectra were measured on a
300 mHz spectrometer in either CDCl3 or DMSO-d6. The
chemical purities were determined by HPLC qualitative and
quantitative analysis using a C18 steel column, 5 µm particle
size (4.6 mm × 250 mm). UV detection and mobile phase
were varied depending upon substrate. Optical purities were
determined by chiral HPLC (6) using a Chiracel-OD-R steel
column (4.6 mm × 250 mm) (the mobile phase was varied
depending upon substrate) or by Capillary Zone Electro-
phoresis analysis (ꢀ-caprolactam 7 and bengamide B ana-
logue 9). In-process control TLC was run using silica gel
862
•
Vol. 7, No. 6, 2003 / Organic Process Research & Development

ambient pressure to a residual volume of 500 mL. Heptane
(1500 mL) was added at 65 ( 5 °C. The slurry was cooled
to 25 °C, held for 30 min, cooled to 0 ( 5 °C, and held for
1 h. The solids were filtered and washed with heptane/ethyl
acetate (3:1) containing 30 ppm of the antistatic agent
Stadis 450 (2 × 100 mL). The product was dried at 45 °C
to give 11 as a white crystalline solid (207.9 g, 0.72 mol,
285.1 [M + Na+]. Anal. Calcd for C11H18O7: C, 50.38; H,
6.92. Found: C, 50.42; H, 6.95. [R]25 -97.63° (c 1.01,
D
MeOH).
5-O-Methyl-2,4-O-(1-methylethylidene-L-glucoronic Acid
γ-Lactone (14). In a 5-L, 4-necked round-bottomed flask
equipped with a mechanical stirrer and a thermocouple were
placed 2-O-methyl-3,5-O-(1-methylethylidene)-D-glycero-D-
gulo-heptonic acid γ-lactone (13, 50 g, 190.65 mmol), water
(175 mL), and acetonitrile (350 mL). The mixture was stirred
at 25 °C until a clear solution was obtained (10 min) and
then cooled to 5 °C. Sodium periodate (44.9 g, 209.7 mmol)
was added in portions at such a rate as to maintain a
temperature of 5-10 °C. The reaction mixture was vigor-
ously stirred at 5-10 °C for 1 h, then acetonitrile (300 mL)
and isopropyl acetate (900 mL) were added. The mixture
was cooled to 8 °C, and anhydrous magnesium sulfate (56.25
g) was added portionwise at or below 15 °C. The suspension
was cooled back to 8 °C, and sodium bicarbonate (16.25 g)
was added, followed immediately by the portionwise addition
of magnesium sulfate (168.8 g) at or below 20 °C. The
reaction mixture was warmed to 25 °C and stirred for 1 h.
Magnesium sulfate (87.5 g) was added, and the suspension
was stirred for 1 h. The solids were filtered and washed with
a mixture of acetonitrile (200 mL) and isopropyl acetate (600
mL). The filtrate was transferred to a 5-L, 4-necked round-
bottomed flask, and the solvent was distilled at 30-35 °C
(100-75 Torr) to a final volume of 450 mL. Isopropyl
acetate (1200 mL) was added, and the solvent was distilled
at ambient pressure (81-89 °C) to a volume of 800 mL.
(Caution: The material is unstable. Exposure to temperatures
above 70 °C must not exceed 1 h.) The mixture was cooled
to 25 °C, and the product precipitated. The solids were
filtered, washed with isopropyl acetate (105 mL), and dried
(25 °C, 50 mbar) to afford 5-O-methyl-2,4-O-(1-methyleth-
ylidene)-L-glucoronic acid γ-lactone (14, 35.1 g, 80%) as a
white solid, mp 152-154 °C. 1H NMR (CDCl3, 300 MHz):
δ 9.52 (s, 1H), 4.88-4.84 (m, 2H), 4.67 (dd, 1H, J ) 2.3,
2.1 Hz), 4.51 (d, 1H, J ) 4.0 Hz), 3.41 (s, 3H), 1.49 (s,
3H), 1.37 9s, 3H). 13C NMR (75 MHz, CDCl3): δ 199.2,
137.2, 98.7, 78.3, 73.4, 68.7, 67.3, 58.1, 29.2, 19.6. MS m/z:
231 (M + 1), 213, 203, 181, 173, 163. Anal. Calcd for
C10H14O6: C, 52.17; H, 6.13. Found: C, 52.01; H, 5.96.
8-Methyl-2-O-methyl-3,5-O-(1-methylethylidene)-6,7,8,9-
tetradeoxy-D-gulo-6-nonenonic Acid (6E)-γ-Lactone (15).
A stirred solution of 2-(2,2-dimethylpropylsulfonyl) benzo-
thiazole 18 (168.5 g, 0.63 mol) in anhydrous THF (2000
mL) was treated dropwise at -55 °C with 1 M lithium bis-
(trimethylsilyl)amide solution in THF (677.6 mL, 0.68 mol).
The solution was cooled to -78 °C and stirred for 1.5 h and
then treated dropwise with chlorotrimethylsilane (86 mL,
0.68 mol). The resulting suspension (-78 °C) was transferred
to a cold (0 to -15 °C) solution of 5-O-methyl-2,4-O-(1-
methylethylidene)-L-glucoronic acid γ-lactone 14 (120.0 g,
0.52 mol) and dry NaHCO3 (1.96 g, 0.029 mol) in anhydrous
acetonitrile (1960 mL) over 20 min. The resulting mixture
was warmed to 0 °C over 20 min, stirred for 1 h, warmed to
22 °C over 30 min, stirred for 1 h, warmed to 50 °C over 30
min, and stirred for 1 h. The mixture was cooled to 10 °C,
1
57.7%), mp 156 °C. H NMR (300 MHz, CDCl3): δ ppm
1.35 (s, 3H), 1.43-1.40 (d, 6H), 1.50 (s, 3H), 3.83-3.86
(m, 1H), 3.90-3.95 (m, 1H), 4.07-4.13 (m, 1H), 4.29-
4.35 (m, 1H), 4.48-4.53 (m, 1H), 4.61-4.63 (m, 1H). 13C
NMR (300 MHz, CDCl3): δ ppm 19.95, 25.24, 27.30, 29.24,
67.30, 68.46, 69.08, 69.84, 71.99, 73.42,99.20, 110.05,
175.16.
2-O-Methyl-3,5:6,7-bis-O-(1-methylethylidene)-D-
glycero-D-gulo-heptonic Acid γ-Lactone (12). To a slurry
of 3,5:6,7-bis-O-(1-methylethylidene)-D-glycero-D-gulo-hep-
tonic acid γ-lactone 11 (200 g, 0.694 mol) in toluene (1400
mL) was added delumped tripotassium phosphate (441.6 g,
2.08 mol) at 20-25 °C. Dimethyl sulfate (262.3 g, 2.08 mol)
was charged, and the slurry was stirred at 20-25 °C for 4
h. The solids were filtered, and the filter cake was rinsed
with toluene (2 × 240 mL). Heptane (1900 mL) was added
to the filtrate, and the product precipitated. The mixture was
stirred at 20-25 °C for 15 min, cooled to 0 ( 5 °C, and
held for 1 h. The solids were filtered and slurried in heptane/
toluene (400 mL, 1:1) at -5 ( 5 °C for 15 min to reduce
the dimethyl sulfate content. The solids were again filtered
and slurried in heptane/toluene (400 mL, 1:1) containing 30
ppm Stadis 450. The solids were filtered and dried at 45 °C
to give 12 as a white crystalline solid (108 g, 0.36 mol,
48.6%), mp 117 °C. 1H NMR (300 MHz, DMSO-d6): δ ppm
1.06 (s, 6H), 1.26 (s, 3H), 1.42 (s, 3H), 3.39 (s, 3H), 3.77-
3.81 (m, 1H), 3.90-3.94 (m, 1H), 3.99-4.03 (m, 1H), 4.07-
4.13 (m, 1H), 4.29-4.30 (d, 1H), 4.48-4.49 (d, 1H), 4.80,
(s, 1H). 13C NMR (300 MHz, CDCl3): δ ppm 19.78, 25.22,
27.34, 29.33, 59.6, 67.32, 67.95, 68.94, 69.92, 73.43, 76.98,
98.97, 110, 172.9. MS m/z: 287 [M - CH3]. Anal. Calcd
for C14H22O7: C, 55.62; H, 7.33; O, 37.04. Found: C, 53.44;
H, 6.93; O, 39.17.
2-O-Methyl-3,5-O-(1-methylethylidene)-D-glycero-D-
gulo-heptonic Acid γ-Lactone (13). To a stirred solution
of 2-O-Methyl-3,5:6,7-bis-O-(1-methylethylidene)-D-glycero-
D-guloheptonic acid γ-lactone 12 (98 g, 0.324 mol) in
isopropyl acetate (490 mL), propionic acid (243 g, 245 mL,
3.284 mol), and water (29.2 g) was added a solution of
p-toluenesulfonic acid (0.308 g, 0.0016 mol) dissolved in
water (5.6 mL), and the resulting solution was heated at 40
°C for 2 h. The resulting slurry was cooled to 4 °C, held for
1 h, filtered, washed with cold isopropyl acetate (2 × 225
mL), and dried at 50 °C to afford 13 as a white, crystalline
solid (71 g, 0.27 mol, 83%), mp 120 °C. 1H NMR (DMSO-
d6): δ ppm 1.24 (s, 3H), 1.42 (s, 3H), 3.15-3.37 (m, 1H),
3.39 (s, 3H), 3.48-3.56 (m, 2H), 3.93 (dd, J ) 1.6, 9.1 Hz,
1H), 4.36 (m, 1H), 4.41 (m, 1H), 4.48 (d, J ) 3.8 Hz, 1H),
4.78 (dd, J ) 2.1, 3.8 Hz, 1H), 4.96 (d, J ) 5.8 Hz, 1H).
13C NMR (DMSO-d6): δ ppm 19.69, 29.45, 57.90, 62.67,
67.61, 67.72, 69.11, 69.33, 78.80, 98.01, 173.86. MS m/z:
Vol. 7, No. 6, 2003 / Organic Process Research & Development
•
863

quenched with water (1.5 L), and concentrated under reduced
pressure (35 °C, 50 mbar) to a paste. MTBE (4.0 L) was
added, and the phases were separated. The organic phase
was washed with water (0.8 L) and concentrated at atmo-
spheric pressure to a volume of 640 mL. The concentrate
was cooled to 22 °C over 1 h, stirred for 30 min, cooled to
0 °C, and stirred for 30 min. The solids were filtered and
washed with cold (0 °C) MTBE (200 mL) to afford 15 as a
white crystalline solid (65.9 g, 0.23 mol, 44.5%), mp 179-
(3S,6R)-3-Aminohexahydro-6-cyclohexanecarboxy-2H-
azepin-2-one Hydrochloride (7). 6 (300 g, 0.85 mol) was
added in portions to a 2 N solution of HCl in ethyl acetate
(3.0 L) at 15 °C. The mixture was stirred at 23 °C for 6 h.
The resulting suspension was filtered and washed with ethyl
acetate (1200 mL). The solids were collected and dried at
40 °C (50 mbar) to give 7 as a white, crystalline solid (240.0
1
g, 98%), mp 285 °C (dec). H NMR (D2O, 300 MHz): δ
4.91 (s, 1 H), 4.25 (m, 1 H), 3.44 (s, 2 H), 2.34 (m, 1 H),
2.13 (m, 1 H), 1.72-2.00 (m, 5 H), 1.42-1.68 (m, 3 H),
1.00-1.39 (m, 5 H). 13C NMR (D2O, 75 MHz): δ 177.9,
173.1, 67.1, 52.2, 42.9, 42.0, 30.4, 28.6, 28.3, 25.1, 24.8,
24.7, 22.6. MS m/z: 255.2, 238.2. Anal. Calcd for
C13H23ClN2O3: C, 53.69; H, 7.99; N, 9.64. Found: C, 53.67;
1
181 °C. H NMR (CDCl3): δ ppm 1.04 (s, 9H), 1.49, 1.54
(2s, 6H), 3.36 (s, 3H), 4.00 (dd, J ) 4.1 Hz, 1H), 4.10 (d, J
) 3.9 Hz, 1H), 4.46 (dd, J ) 7.3, 2.4 Hz 1H), 4.72 (dd, J )
3.9, 2.1 Hz, 1H), 5.59 (dd, J ) 15.8, 7.5 Hz, 1H), 5.84 (dd,
J ) 15.8, 0.6 Hz, 1H). 13C NMR (CDCl3): δ ppm 19.28,
29.24 (t-bu), 33.24, 59.18, 67.57, 70.34, 71.90, 79.07, 98.72,
H, 7.91; N, 9.56; [R]24 +72.5° (c 1.0, H2O).
D
+
120.07, 147.07, 172.97. MS m/z: 301.8 [M + NH4 ]. Anal.
(6S)-6-[2-[(4R,5S,6R)-6-[(1E)-3,3-Dimethyl-1-butenyl]-
5-hydroxy-2,2-dimethyl-[1,3]-dioxan-4-yl]-(2R)-2-meth-
oxyacetylamino]hexahydro-7-oxo-(3R)-1H-azepin-3-ylcy-
clohexanecarboxylic Acid Ester (8). A flask was charged
with 7 (112.49 g, 0.39 mol), sodium 2-ethylhexanoate
(116.34 g, 0.70 mol), 15 (100.0 g, 0.35 mol), and tetrahydro-
furan (1.75 L). The solution was stirred at 23 °C for 20 h.
Water (350 mL) was added, followed by a slow addition of
heptane (3.5 L) at 23 °C. The resulting suspension was stirred
at room temperature for 3 h, cooled to 2 °C, and stirred for
an additional 2 h. The mixture was filtered and washed with
water (200 mL) and heptane (400 mL). The solids were
collected and dried at 40 °C (50 mbar) to give 8 as a white
Calcd for C15H24O5: C, 63.36; H, 8.51. Found: C, 63.27;
H, 8.42. [R]25 -148.12° (c 1.04, MeOH).
D
[(3R-cis)-6-(1,1-Dimethylethoxy)carbonylamino-
hexahydro-7-oxo-1H-azepin-3-yl]cyclohexanecarboxylic
Acid Ester (6). A stirred solution of (5R)-5-hydroxy-L-lysine
dihydrochloride monohydrate 1 (126.6 g, 0.50 mol) in
methanol (1500 mL) at 25 °C was treated dropwise with
chlorotrimethylsilane (135.8 g, 159 mL, 1.25 mol). The
solution was heated to 63 °C, stirred for 3 h, cooled to 50
°C, and treated dropwise with triethylamine (253 g, 349 mL,
2.5 mol). A precipitate immediately formed. The suspension
was stirred at 50 °C for 3 h, cooled to 25 °C, and treated in
one portion with di-tert-butyl dicarbonate (131 g, 0.6 mol).
A solution resulted within a few minutes, and it was stirred
for 1 h and concentrated under reduced pressure (50 °C, 100
mbar) to a thick paste. Ethyl acetate (2 L) was added, and
the concentration was repeated. Fresh ethyl acetate (2 L) was
charged, and the methanol content was monitored by GC
analysis. The charge/concentrate/charge sequence was re-
peated until the methanol content was e0.05 mg/mL. To
the suspension was added 4-(dimethylamino)-pyridine (79.4
g, 0.65 mol). The suspension was heated to 50 °C, treated
dropwise with cyclohexanecarbonyl chloride (88 g, 0.6 mol),
and stirred for 3 h. The mixture was quenched with water
(1 L), and the phases were separated. The organic phase was
washed with water, azeotropically dried (1 atm), and
concentrated to a volume of 800 mL. Heptane (1.4 L) was
added, the resulting suspension was stirred for 1 h at 4 °C,
and the solids were filtered, washed with 500 mL of a 2:1
mixture of heptane/ethyl acetate, and dried at 50 °C to afford
6 as a white crystalline solid (146 g, 0.41 mol, 82.5%), mp
220 °C. 1H NMR (CDCl3): δ ppm 1.17-1.38 (m, 5H), 1.43
(s, 9H), 1.59-1.67 (m, 1H), 1.68-1.82 (m, 3H), 1.85-2.05
(m, 4H), 2.08-2.13 (m, 1H), 2.25-2.35 (m, 1H), 3.39-
3.55 (m, 2H), 4.29 (dd, J ) 9.8, 5.8 Hz, 1H), 4.88 (d, J )
2.6 Hz, 1H), 5.90 (d, J ) 5.6 Hz, 1H), 6.29 (t, J ) 5.9 Hz,
1H). 13C NMR (CDCl3): δ ppm 25.72, 25.74, 26.02, 27.17,
28.75 (t-bu), 29.32, 29.36, 32.42, 43.54, 43.67, 53.27, 67.27,
80.00, 155.54, 175.46, 175.64. MS m/z: 377.1 [M + Na+].
Anal. Calcd for C18H30N2O5: C, 61.00; H, 8.53; N, 7.90.
Found: C, 61.21; H, 8.53; N, 7.84. [R]25D +79.84° (c 1.02,
MeOH).
1
solid (166.55 g, 88%), mp 209-211 °C. H NMR (CDCl3,
300 MHz): δ 7.50 (d, 1 H, J ) 6.4 Hz), 6.49 (t, 1 H, J )
6.2 Hz), 5.69 (d, 1 H, J ) 15.8 Hz), 5.43 (dd, 1 H, J )
15.8, 6.7 Hz), 4.83 (d, 1 H, J ) 3.6 Hz), 4.52 (m, 1 H), 4.20
(d, 1 H, J ) 6.7 Hz), 3.99 (d, 1 H, J ) 7.3 Hz), 3.83 (d, 1
H, J ) 7.4 Hz), 3.40-3.50 (m, 3 H), 3.39 (s, 3 H), 2.79 (br
s, 1 H), 2.21 (m, 1 H), 1.78-2.05 (m, 5 H), 1.50-1.70 (m,
4 H), 1.34 (s, 6 H), 1.10-1.30 (m, 5 H), 0.93 (s, 9 H). 13C
NMR (CDCl3, 75 MHz): δ 175.0, 174.5, 169.4, 145.1, 121.4,
99.4, 80.3, 74.3, 73.0, 66.7, 65.6, 59.0, 51.5, 43.0, 32.9, 31.8,
29.4, 29.2, 28.8, 28.7, 25.7, 25.5, 25.2, 25.1, 18.9. MS m/z:
538.9, 481.2. Anal. Calcd for C28H46N2O8: C, 62.43; H, 8.61;
N, 5.20. Found: C, 62.51; H, 8.58; N, 5.13. [R]24 +57.7°
D
(c 1.0, MeOH).
Hexahydro-7-oxo-(6S)-6-[(2R,3R,4S,5R,6E)-3,4,5-trihy-
droxy-2-methoxy-8,8-dimethyl-1-oxo-6-nonenylamino]-
(3R)-1H-azepin-3-ylcyclohexanecarboxylic Acid Ester (10).
To a solution of tetrahydrofuran (1.80 L) and 1 N hydro-
chloric acid (3.60 L) at 22 °C was added 8 (180.0 g, 0.335
mol). The solution was stirred at 22 °C for 4 h, cooled to 5
°C, and neutralized to pH 7.0 with 5.0 N aqueous sodium
hydroxide solution. Sodium chloride (1260 g) was added,
and the mixture was warmed to 22 °C. Ethyl acetate (1.80
L) was added, the mixture was shaken, and the phases were
separated. The organic phase was distilled under vacuum,
and 3.14 L of distillate was collected. Toluene (0.27 L) was
added, and the solution was again distilled under vacuum
until 0.30 L of distillate was collected. tert-Butyl methyl ether
(0.72 L) was charged at 50 °C, and the solution was stirred
at 50 °C for 0.5 h, cooled further to 22 °C, and stirred for
864
•
Vol. 7, No. 6, 2003 / Organic Process Research & Development

10 h. The resulting suspension was then cooled to 2 °C and
stirred for 2 h. The solids were filtered and washed with
tert-butyl methyl ether (0.16 L) and dried at 40 °C (50 mbar)
to give crude 9 as a white solid (160.8 g). The crude solids
(160.3 g) and ethanol (0.20 L) were charged to a 1-L flask,
and the suspension was heated to 60 °C. Water (0.60 L) was
slowly added. The resulting mixture was stirred at 60 °C
for 15 min, cooled slowly to 22 °C, and stirred for 6 h. The
suspension was cooled to 0 °C and stirred for 2 h, filtered
and washed with of water (0.20 L), and dried at 40 °C (50
mbar). The solids were exposed to air for 24 h to give 10
(133.6 g, 79%) as a white solid, mp 148-149 °C. 1H NMR
(CDCl3, 300 MHz): δ 8.00 (d, 1 H, J ) 6.4 Hz), 6.32 (t, 1
H, J ) 6.2 Hz), 5.83 (d, 1 H, J ) 14.9 Hz), 5.42 (dd, 1 H,
J ) 15.7, 7.3 Hz), 4.95 (m, 1 H), 4.57 (m, 1 H), 4.36 (d, 1
H, J ) 2.5 Hz), 4.24 (m, 1 H), 3.85 (s, 2 H), 3.60 (m, 2 H),
3.52 (s, 3 H), 3.45 (m, 2 H), 3.30 (s, 1 H), 2.35 (m, 1 H),
1.60-2.20 (m, 9 H), 1.15-1.50 (m, 5 H), 1.01 (s, 9 H). 13
C
NMR (CDCl3, 75 MHz): δ 175.6, 174.7, 172.4, 146.0, 123.6,
81.6, 74.8, 73.0, 72.9, 67.1, 60.2, 52.0, 43.6, 33.4, 32.3, 29.8,
29.4, 29.3, 26.1, 26.0, 25.8, 25.7. MS m/z: 499.0, 481.1,
445.3, 355.2. Anal. Calcd for C25H42N2O8: C, 60.22; H, 8.49;
N, 5.62. Found: C, 59.94; H, 8.37; N, 5.54. [R]24 +98.5°
D
(c 1.0, MeOH).
Acknowledgment
We gratefully acknowledge the valuable discussions with
Dr. Peter Giannousis and Mr. Joseph McKenna. We wish to
thank Dr. Frederick Kinder, Jr. for lead finding and valuable
scientific discussions throughout the development.
Received for review August 12, 2003.
OP0341162
Vol. 7, No. 6, 2003 / Organic Process Research & Development
•
865