A new and efficient synthesis of the HMG-CoA reductase inhibitor pitavastatin

Article abstract of DOI:10.1002/hlca.200790106

A new synthetic method for the preparation of pravastatin is described. The approach circumvents various synthetic problems associated with the buildup of the 3,5-dihydroxy-C₇ acid side chain of HMG-CoA reductase inhibitors (statins). The use of the C₆-amide derivative 5 instead of ester derivatives in the coupling reaction with carboxaldehyde 8 (Scheme 3) prevents undesired side reactions, such as eliminations and retro-aldol reactions. The method provides synthetic statins, such as pitavastatin, in >99% ee and exceptionally high overall yield. The enantiomerically pure starting material, (3S)-3-{[(tert-butyl)dimethylsilyl]oxy}-5-oxo-5-{[(1S)-1-phenylethyl]amino} pentanoic acid (3c), is prepared by an improved procedure from 3-{[(tert-butyl)dimethylsilyl]oxy}glutaric anhydride (1) and (1S)-1-phenylethylamine (2c; Scheme 1).

Full text of DOI:10.1002/hlca.200790106

Helvetica Chimica Acta – Vol. 90 (2007)
1069
A New and Efficient Synthesis of the HMG-CoA Reductase Inhibitor
Pitavastatin
by Murat Acemoglu*, Andre Brodbeck, AngelGarcia , Dominique Grimler, Marc Hassel, Bernhard Riss,
and Robert Schreiber
Novartis Pharma AG, Chemical & Analytical Development, Process Research & Development,
CH-4002 Basel
(phone: þ 41613245929; fax: þ 41613249536; e-mail:murat.acemoglu@novartis.com)
Dedicated to Professor Dr. Conrad Hans Eugster
A new synthetic method for the preparation of pitavastatin is described. The approach circumvents
various synthetic problems associated with the buildup of the 3,5-dihydroxy-C7 acid side chain of HMG-
CoA reductase inhibitors (statins). The use of the C6-amide derivative 5 instead of ester derivatives in the
coupling reaction with carboxaldehyde 8 (Scheme 3) prevents undesired side reactions, such as
eliminations and retro-aldol reactions. The method provides synthetic statins, such as pitavastatin, in
> 99% ee and exceptionally high overall yield. The enantiomerically pure starting material, (3S)-3-
{[(tert-butyl)dimethylsilyl]oxy}-5-oxo-5-{[(1S)-1-phenylethyl]amino}pentanoic acid (3c), is prepared by
an improved procedure from 3-{[(tert-butyl)dimethylsilyl]oxy}glutaric anhydride (1) and (1S)-1-phenyl-
ethylamine (2c; Scheme 1).
Introduction. – The low-density lipoprotein (LDL) cholesterol-lowering drugs
(statins) act by inhibition of the 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA)
reductase activity, resulting in the inhibition of cholesterol biosynthesis. Their
physiological activity is attributed to the presence of the 3,5-dihydroxyheptanoic acid
substructure, which resembles the HMG moiety of HMG-CoA. (R)-Configuration at
C(3) and ꢀsynꢁ-configuration of the OH groups were shown to be essential for obtaining
high affinities [1]. Since the discovery of the HMG-CoA reductase inhibitory activity of
the fungal metabolites compactin [2][3] and mevinolin [4] (¼ monacolin K, [5]), a
series of analogous compounds have been synthesized and introduced into the market.
Fig. 1 shows a selection of compounds marketed as HMG-CoA reductase inhibitors for
the treatment of hypercholesterolemia.
In recent years, a large number of enzymatic and microbiological methods have
been reported for the synthesis of the enantiomerically pure 3,5-dihydroxyheptanoic
acid side chain of statins [6]. Nevertheless, there are also synthetic approaches for the
construction of this moiety utilizing enantiomerically pure compounds from the chiral
pool as starting materials, and applying highly diastereoselective reactions. One of
these approaches is based on desymmetrization of the prochiral 3-{[(tert-butyl)dime-
thylsilyl]oxy}pentanedioic anhydride (1) with various, chiral nucleophiles. In their
pioneering work, Heathcock and co-workers have reported two methods for the
diastereoselective ring-opening reaction of anhydride 1 (Scheme 1):with (1 R)-1-
phenylethanol (2a) as nucleophile in the presence of Et3N and N,N-dimethylpyridine-4-
ꢂ 2007 Verlag Helvetica Chimica Acta AG, Zürich

1070
Helvetica Chimica Acta – Vol. 90 (2007)
Fig. 1. A selection ofmarketed HMG-CoA reductase inhibitors
amine (DMAP) at ꢀ 308, a ratio of 8 :1 was obtained for 3a/epi-3a in 70% yield [7].
The ratio of diastereoisomers was improved to 15 :1 by using only DMAP as base at
ꢀ 408. With (1R)-1-(1’-naphthyl)ethanol as nucleophile in the presence of DMAP in
CH2Cl2 as solvent, a mixture of epimers 3b/epi-3b in a ratio of up to 50 :1 was obtained
in 92% yield [8] (Scheme 1). These compounds have been demonstrated to be useful
intermediates for the synthesis of stereoisomeric HMG-CoA reductase inhibitors in

Helvetica Chimica Acta – Vol. 90 (2007)
1071
Scheme 1
Fig. 2. b-Keto phosphonate intermediates used in the published synthesis [9] (compound 4) and in this
work (compound 5)
multigram quantities [7][8]. However, for large scale synthesis, these methods turned
out to have several disadvantages:the enantiomerically pure alcohols were both
expensive and not readily available in bulk quantities. In addition, retro-aldol reactions
and b-elimination of the silyloxy group were observed as side reactions during
functionalization of these intermediates, e.g., during formation of phosphonic acid ester
derivatives, such as compound 4 (Fig. 2) [8]. From a practical point of view, these
methods remained to be expensive and very sensitive to reaction conditions.
Diastereoselective opening of anhydride 1 with enantiomerically pure (1S)-1-phenyl-
ethylamine (2c) was introduced by Karanewsky and co-workers [9] as a practical
alternative (Scheme 1). The corresponding product 3c was isolated as enantiomerically
pure, crystalline diastereoisomer in 72% yield, although the differentiation of the
enantiotopic CO groups in the ring-opening step was moderate, and the epimers 3c/epi-
3c were formed in a ratio of 79 :21. Since both (1R)- and (1S)-enantiomers of 1-
phenylethylamine are commercially available on large scale at reasonable prices, the
method of Karanewsky and co-workers [9] represents an interesting approach towards
an industrial-scale synthesis of the enantiomerically pure side chains of statins. Among
other methods reported for the asymmetric synthesis of the side-chain moieties of
statins were the application of a highly diastereoselective hetero-Diels – Alder reaction
[10], of diastereoselective aldol reactions with (ꢀ)-(1S)-2-hydroxy-1,2,2-triphenylethyl
acetate ((S)-HYTRA) [11], and of the Blaise reaction [12]. Finally, the synthesis of
fluorinated analogues have recently been reported by Ramachandran and co-workers
[13].

1072
Helvetica Chimica Acta – Vol. 90 (2007)
Pitavastatin is a powerful statin comprising a 3,5-dihydroxyhept-6-enoic acid
substructure instead of the 3,5-dihydroxyheptanoic acid moiety [14] (Fig. 1). The
double bond of the heptenoic acid substructure has (E)-configuration. Various
syntheses for pitavastatin and its lactone NK-1041) have been published:utilization of
Taberꢁs alcohol as chiral auxiliary [15], use of enantiomerically pure epichlorohydrin as
starting material [16], and resolution [17][18]. The different epimers of NK-104, the
corresponding lactone of pitavastatin1), were also synthesized by asymmetric aldol
reaction with (S)-HYTRA as chiral auxiliary [17]. In this report, a new, short, and
efficient synthesis of pitavastatin and NK-104 will be described.
Results and Discussion. – The method of Karanewsky and co-workers [9] for the
preparation of the enantiomerically pure compound 3c was the starting point of our
investigations. In addition to (1S)-1-phenylethylamine (2c), which had been used by
Karanewsky and co-workers in the presence of Et3N, we investigated the use of (1S)-1-
(1-naphthyl)ethylamine (2d) and (1S)-1-(2-naphthyl)ethylamine (2e) as chiral amines
i
for the diastereoselective ring opening of anhydride 1 in the presence of Pr2EtN:a
mixture of epimers in a ratio of 79 :21 was formed with 2d as nucleophile. The major
epimer was isolated by trituration in Et2O in 73% yield and with a diastereoisomer ratio
(d.r.) of 97.5 :2.5. With 2e as nucleophile, a mixture of epimers in a ratio of 68 :32 was
obtained. Again, the major epimer was isolated by trituration in 57% yield and
99 :1 d.r. In comparison, reaction of 2c with anhydride 1, under the same conditions,
afforded compound 3c as major epimer in 72% yield and > 99 :1 d.r. The epimers 3c/
epi-3c were formed in a ratio of 82 :18 in this case. In conclusion, both 2d and 2e gave
lower yields and diastereoselectivities as compared to 2c. Based on these results, the
reactions of the chiral amines 2d and 2e with anhydride 1 were not further investigated,
and the absolute configurations of the major diastereoisomers isolated from these
reactions were not determined.
The observed low diastereoselectivity of the ring-opening reaction of anhydride 1
with (1S)-1-phenylethylamine (2c) under the conditions similar or identical to those
used by Karanewsky and co-workers might partially be attributed to the use of an
achiral tertiary amine as base. The tertiary amine is supposed to promote the addition
of the nucleophile by cleaving the anhydride and forming an acylammonium
intermediate, which is then attacked by the nucleophile. In the case of (1S)-1-
phenylethylamine (2c) as nucleophile, we questioned the necessity for a tertiary amine
i
as catalyst:replacement of Pr2EtN by a second equivalent of (1S)-1-phenylethylamine
under the same conditions improved the diastereoselectivity, and a ratio of 86 :14 was
observed for 3c/epi-3c. Finally, by additional optimization of reaction conditions such as
solvent change and crystallization, a final ratio of 93 :7 was achieved for the formation
of 3c/epi-3c, and the desired, crystalline diastereoisomer 3c was isolated in 87.6% yield
(with regard to the expensive anhydride 1) and in > 99.8 :0.2 d.r.
1
)
There seems to be some confusion in the literature with regard to the synonyms NK-104 and
pitavastatin:In [15] and [16], the lactone derived from pitavastatin is named NK-104. In [17], NK-
104 is used as synonym for the monocalcium salt of the acid. In this report, NK-104 will be used as
synonym for the lactone and pitavastatin as synonym for the monocalcium salt.

Helvetica Chimica Acta – Vol. 90 (2007)
1073
Having developed an efficient method for the preparation of 3c, we turned our
attention to the further functionalization and application of this chiral intermediate.
According to [9], intermediate 3c is converted to b-keto phosphonate 4 (Fig. 2) by a
five-step sequence of reactions, involving cleavage of the amide by nitrosation with
dinitrogen tetraoxide and White rearrangement of the resulting N-nitrosoamide,
followed by treatment with excess dimethyl (lithiomethyl)phosphonate and esterifi-
cation with diazomethane. In this strategy, the elimination of the (tert-butyl)dimethyl-
silyloxy group was prevented by using the salt form of glutaric monomethyl ester for
the reaction with dimethyl (lithiomethyl)phosphonate. However, the elimination
remained to be a potential problem for the subsequent Horner – Wadsworth – Emmons
condensations with the b-keto phosphonate 4 (Fig. 2). In our design, we hypothesized
that the tedious hydrolysis of the N-[(1S)-1-phenylethyl]amide function of 3c was not
necessary and that the amide function would rather be advantageous to prevent
elimination during both preparation and subsequent Horner – Wadsworth – Emmons
condensation of the corresponding b-keto phosphonate 5 (Fig. 2). In the presence of
the amide function, excess base should lead to enolization of the amide instead of the
elimination. This hypothesis proved to be correct and has lead to a very short and highly
efficient synthesis of pitavastatin.
Activation of the carboxylic acid function of 3c with isobutyl carbonochloridate and
in situ treatment of the formed mixed anhydride with N-methoxymethanamine gave
the Weinreb intermediate 6 in 96% yield (Scheme 2). Reaction of this compound with a
fourfold excess of dimethyl (lithiomethyl)phosphonate in THFat ꢀ 788 afforded the b-
keto phosphonate 5 in 95% yield. Similarly, phosphonate 7 was obtained in 95% yield
from the reaction of 6 with an excess of diethyl (lithiomethyl)phosphonate. Because of
the bulk availability and low price of methylphosphonic acid dimethyl ester, the b-keto
phosphonate 5 was used for the following steps.
Scheme 2

1074
Helvetica Chimica Acta – Vol. 90 (2007)
Scheme 3
The Horner – Wadsworth – Emmons reaction of the b-keto phosphonate 5 with
aldehyde 8 (Scheme 3) was investigated in some detail:different bases (NaH, KOH,
i
K2CO3, Cs2CO3) in different solvents (THF, MeCN, PrOH, EtOH, MeOH, and
mixtures with H2O) were screened for the best yield and conversion. With NaH as base

Helvetica Chimica Acta – Vol. 90 (2007)
1075
in THF as solvent, no reasonable conversion was observed. With K2CO3 as base and
EtOH as solvent, the condensation was a very clean reaction with peak-to-peak
conversion to 9 without formation of any by-products. However, the reaction rate
decreased substantially with increasing reaction time, and an excess of 5 – 10 mol-% of
the phosphonate 5 had to be added to achieve > 95% conversion of the aldehyde. It is
noteworthy to mention that no elimination occurred under these conditions, regardless
of the very long reaction time at 408. With Cs2CO3 instead of K2CO3, the reaction could
be accelerated substantially, reducing the reaction time from 48 h to 16 h. However, the
formation of a by-product (3 – 5 area-% by HPLC) was observed in this case. With
KOH as base in EtOH, the reaction was much faster (completion after 2 – 3 h) but by-
product formation was observed also in this case (2 – 3 area-% by HPLC). Based on
these results, the Horner – Wadsworth – Emmons condensation was performed in EtOH
as solvent with K2CO3 as base.
After workup, the crude 9 (quant. yield) was used for the next step without
t
purification. The BuMe2Si protecting group was cleaved with either H3PO4 in MeCN
or HCl in EtOH. The conversion of 9 to 10 with HCl in EtOH was faster and cleaner.
After crystallization from toluene/hexane, pure 10 was isolated in 78% yield over two
steps with regard to 5 and in 86% yield with regard to aldehyde 8.
The stereoselective reduction of hydroxy ketone derivative 10 to the ꢀsynꢁ-diol 11
((3R,5S)) was accomplished with Et2BOMe/NaBH4 according to the method described
by Prasad and co-workers [19]. We investigated the effect of the amount of Et2BOMe
on the formation of the ꢀsynꢁ/ꢀantiꢁ diastereoisomers (Table):even in the presence of
only 0.5 equiv. of Et2BOMe, a ratio of 97.4 :2.6 was observed. Obviously, the reduction
of the hydroxy ketone/Et2BOMe complex was much faster than the reduction of the
free hydroxy ketone moiety. In all cases, the reaction proceeded very smoothly as a
peak-to-peak conversion according to HPLC analysis. The crude product 11 was
recrystallized from tBuOMe to obtain crystalline 11 in 99.9 :0.1 d.r. and 86% yield as a
t
1:1 solvate with BuOMe.
Table. Dependency ofthe ꢀsynꢁ-Selectivity on the Amount ofDiethylborinic Acid Methyl Ester
(Et2BOMe) for the Reduction of 10
Mol-equiv. of 10
Mol-equiv. of Et2BOMe
ꢀsynꢁ/ꢀantiꢁ (crude) [area-% by HPLC]
1.00
1.00
1.00
1.12
1.00
0.50
99.57:0.43
98.70 :1.30
97.40 :2.60
The hydrolysis of compound 11 with NaOH in EtOH/H2O was a clean reaction
under mild conditions. Participation of OHꢀC(5) in a neighboring-group effect can be
assumed to facilitate the hydrolysis in this case although no lactone intermediate was
observed during the hydrolysis. After hydrolysis, the sodium salt was not isolated but
was converted to NK-104 by protonation with HCl and subsequent lacton formation in
t
toluene. Crystallization from BuOMe/hexane afforded pure NK-104. Alternatively,
the calcium salt pitavastatin was obtained upon treatment of the sodium salt with
CaCl2. In both cases, the by-product of hydrolysis, (1S)-1-phenylethylamine, had to be
removed from the solution to obtain pure products. Pitavastatin precipitated in high

1076
Helvetica Chimica Acta – Vol. 90 (2007)
purity and was isolated as a hydrate, comprising 10.6% (w/w) of H2O. According to
electrophoresis on a chiral column, the product had > 99.9% e.e. The (3S,5R)-
enantiomer (ꢀantiꢁ) of pitavastatin was not detectable in the product at a detection limit
of 0.05%. In the HPLC, MS, NMR, and IR spectra, pitavastatin was identical to the
drug substance of Livaloꢃ, a drug marketed by Kowa Pharmaceuticals. The overall yield
of pitavastatin was 55% starting from 3c and 67% with regard to the aldehyde 8.
In summary, a new, short, and highly efficient method for the synthesis of statins was
devised. The new strategy afforded synthetic statins, such as pitavastatin, in > 99.9% ee
and in exceptionally high overall yield.
ExperimentalPart
General. Reagents and solvents were obtained from commercial sources and were used as received.
All reactions were carried out under N2 unless otherwise stated. Temp. were internally measured unless
otherwise stated. Column chromatography (CC):silica gel ( E. Merck, grade 60, particle size 0.040 –
0.063 mm, 230 – 400 mesh ASTM). Capillary zone electrophoresis (CE): Agilent-3D CE apparatus,
fused silica capillary at 208; UV detection (248 nm). Melting points: Leitz-Kofler hot-stage apparatus;
uncorrected. Specific optical rotations: Perkin-Elmer 341 polarimeter. IR Spectra: Bruker IFS66 or
Bruker IFS88 instrument; in cmꢀ1. NMR Spectra: Bruker DPX400, DMX-500, or DRX-500
spectrometer; d in ppm, J in Hz. Mass spectra: Fisons-VG Quattro-II (ESI), Finnigan TSQ-7000
(ApCI), or Finnigan 8430 (EI) instrument; in m/z.
(3S)-3-{[(tert-Butyl)dimethylsilyl]oxy}-5-oxo-5-{[(1S)-1-phenylethyl]amino}pentanoic Acid (3c).
t
The clear soln. of 1 (30 g, 0.123 mol), in BuOMe (210 ml), and heptane (120 ml) was cooled to ꢀ 788
to obtain a slurry. A soln. of (ꢀ)-(1S)-1-phenylethylamine (2c; 31.3 g, 0.258 mol; 99.6% e.e.) in heptane
(120 ml) was added slowly under mechanical stirring over 60 – 90 min at ꢀ 78 to ꢀ 758. The clear viscous
soln. was stirred for 2 h at ꢀ 78 to ꢀ 758 and then allowed to warm up to 20 – 258. H2O (100 ml) was
added followed by the addition of 20% aq. H3PO4 soln. (100 ml) until a pH of 2.5 – 3.5 was reached,
whilst keeping the temp. at 30 – 358. The mixture was then heated to reflux and stirred for 30 min under
reflux. Finally, the mixture was cooled to 0 – 58 and the product isolated by filtration. The filter cake was
washed with heptane/H2O 1:1 (80 ml) followed by 20% EtOH/H2O (100 ml). The product was dried at
50 – 608/10 – 20 mbar:39.4 g (87.7%) of 3c. White crystals. M.p. 170 – 1728. [a]2D0 ¼ ꢀ70.8 (c ¼ 1.10,
MeOH) ([9]:[ a]2D0 ¼ ꢀ69.2 (c ¼ 1.12, MeOH)). HPLC: > 99.8% of 3c and < 0.2% of epi-3c ((3R)).
Spectroscopic data:identical to those published in [9]. Anal. calc. for C 19H31NO4Si:C 62.43, H 8.55, N
3.83, Si 7.68; found:C 62.32, H 8.55, N 3.74, Si 7.73.
3-{[(tert-Butyl)dimethylsilyl]oxy}-5-{[(1S)-(naphthalen-1-yl)ethyl]amino}-5-oxopentanoic Acid (3d
or epi-3d, i.e., major epimer). A soln. of 1 (1.428 g, 5.84 mmol) in toluene (25 ml) was cooled to ꢀ 788,
and iPr2EtN (0.755 g, 5.84 mmol) was added, followed by (ꢀ)-(1S)-1-(1-naphthyl)ethylamine (2d;
1.082 g, 6.32 mmol). The mixture was mechanically stirred for 4.5 h at ꢀ 788, then allowed to warm up to
258, and stirred for an additional hour at 258. The reaction was quenched by pouring onto a 5% aq.
KHSO4 soln. (35 ml), and the product was extracted with AcOEt/THF 2 :1 (30 ml). The org. layer was
washed with 5% aq. KHSO4 soln. (20 ml) and brine (20 ml) and dried (MgSO4), the solvent evaporated
at 558, and the crude solid epimer mixture analyzed by HPLC:epimer ratio 79 :21. The mixture was
triturated in Et2O (10 ml), the formed suspension stirred for 3.5 h at 0 – 58, and the precipitate isolated by
filtration and dried for 16 h at 458 under reduced pressure:1.776 g (73%) of 3d or epi-3d (major epimer).
White, crystalline solid. HPLC:97.5% of major epimer and 2.5% of minor epimer. M.p. 169.5 – 172 8.
[a]2D0 ¼ ꢀ8.54 (c ¼ 1.0, CHCl3). IR:3334, 2950, 2928, 2856, 1696, 1618, 1552, 1299, 1258, 1209, 1152, 1096,
843, 831, 777. 1H-NMR ((D6)DMSO, 400 MHz):0.061 ( s, 3 H); 0.076 (s, 3 H); 0.85 (s, 9 H); 1.49 (d, J ¼ 7,
3 H); 2.25 – 2.31 (m, 2 H); 2.38 – 2.48 (m, 2 H); 4.45 – 4.55 (m, 1 H); 5.67 – 5.77 (m, 1 H); 7.47 – 7.60 (m,
4 H); 7.82 (d, J ¼ 8, 1 H); 7.94 (d, J ¼ 7.3, 1 H); 8.1 (d, J ¼ 7.9, 1 H); 12.22 (s, 1 H). 13C-NMR ((D6)DMSO,
125 MHz): ꢀ 4.5; ꢀ 4.3; 18.2; 22.0; 26.2 (3C); 43.0; 44.0; 44.3; 67.4; 122.8; 123.5; 125.9; 126.0; 126.6;

Helvetica Chimica Acta – Vol. 90 (2007)
1077
127.7; 129.1; 130.8; 133.8; 140.7; 169.0; 172.7. ESI-MS:414 ([ M ꢀ H]ꢀ). Anal. calc. for C23H33NO4Si:C
66.47, H 8.00, N 3.37, Si 6.76; found:C 66.09, H 8.06, N 3.22, Si 6.96.
3-{[(tert-Butyl)dimethylsilyl]oxy}-5-{[(1S)-(naphthalen-2-yl)ethyl]amino}-5-oxopentanoic Acid (3e
or epi-3e, i.e., major epimer). As described for 3d or epi-3d, from (1S)-1-(2-naphthyl)ethylamine (2e)
and 1:crude epimer mixture, ratio 68 :32. The major isomer 3e or epi-3e was isolated in 57% yield and
99.1% purity, comprising 0.9% of the minor isomer (by HPLC). M.p. 146 – 1488. [a]2D0 ¼ ꢀ58.36 (c ¼ 1.0,
CHCl3). IR:3329, 2927, 2855, 1700, 1623, 1559, 1305, 1257, 1209, 1154, 1098, 831, 814, 779, 748. 1H-NMR
((D6)DMSO, 400 MHz):0.057 ( s, 6 H); 0.844 (s, 9 H); 1.43 (d, J ¼ 7.1, 3 H); 2.28 – 2.38 (m, 2 H); 2.39 –
2.48 (m, 2 H); 4.45 – 4.54 (m, 1 H); 5.04 – 5.14 (m, 1 H); 7.45 – 7.53 (m, 3 H); 7.77 (s, 1 H), 7.85 – 7.93 (m,
3 H); 8.46 (d, J ¼ 8.1, 1 H); 12.25 (s, 1 H). 13C-NMR ((D6)DMSO, 125 MHz): ꢀ 4.55; ꢀ 4.36; 18.2; 22.7;
26.2 (3 C); 43.0; 44.1; 48.3; 67.4; 124.3; 125.3; 126.0; 126.5; 127.9; 128.0; 128.3; 132.5; 133.3; 142.8; 169.2;
172.7. ESI-MS:414 ([ M ꢀ H]ꢀ). Anal. calc. for C23H33NO4Si:C 66.47, H 8.00, N 3.37, Si 6.76; found:C
66.24, H 8.09, N 3.21, Si 6.74.
{(4R)-4-{[(tert-Butyl)dimethylsilyl]oxy}-2,6-dioxo-6-{[(1S)-1-phenylethyl]amino}hexyl}phosphonic
Acid Dimethyl Ester (5). A soln. of methylphosphonic acid dimethyl ester (140.83 g, 1.101 mol) in THF
(390 ml) was cooled to ꢀ 788, and 1.6m BuLi in hexane (374.4 g, 881 mmol) was added under
mechanical stirring within 3 h. After stirring for additional 60 min at ꢀ 788, a soln. of 6 (see below; 90 g,
220.2 mmol) in THF (360 ml) was added while maintaining the temp. at ꢀ 788. Stirring was continued for
2.5 h at ꢀ 788, and the reaction was quenched by slow addition of a soln. of AcOH (66.1 g, 1.100 mol) in
t
THF (46 ml) at ꢀ 788. The mixture was then allowed to warm up to r.t. and poured onto BuOMe
(1300 ml) and brine (1300 ml). The biphasic mixture was stirred for 15 min. Then the H2O layer was
t
extracted again with BuOMe (1300 ml), the combined org. phase washed with H2O (4 ꢁ 1300 ml) and
dried (MgSO4 (75 g)), and the solvent evaporated. The viscous oil (107.2 g) solidified by storage in the
refrigerator: 5 (104.0 g, 95%). [a]2D0 ¼ ꢀ37.7 (c ¼ 1.0, CHCl3). IR (film):3299, 2955, 2929, 2855, 1717,
1651, 1542, 1376, 1256, 1186, 1035, 838, 813, 779, 701. 1H-NMR ((D6)DMSO, 500 MHz):0.02 ( s, 3 H); 0.03
(s, 3 H); 0.82 (s, 9 H); 1.315 (d, J ¼ 7, 3 H); 2.21 (dd, J ¼ 14, 4.5, 1 H); 2.295 (dd, J ¼ 14, 7.9, 1 H); 2.70 –
2.85 (m, 2 H); 3.21 – 3.31 (m, 2 H); 3.64 (d, J(H,P) ¼ 11.2, 6 H); 4.40 – 4.50 (m, 1 H); 4.85 – 4.94 (m, 1 H);
7.15 – 7.25 (m, 1 H); 7.25 – 7.35 (m, 4 H); 8.33 (d, J ¼ 8.0, 1 H). 13C-NMR (CDCl3, 125 MHz): ꢀ 4.87;
ꢀ 4.82; 17.98; 22.22; 25.85 (3 C); 42.32 (d, J(C,P) ¼ 128); 43.58; 48.81; 49.74; 53.05 (d, J(C,P) ¼ 6.5); 53.16
(d, J(C,P) ¼ 6.5); 65.98; 126.30 (2 C); 127.41; 128.76 (2 C); 143.66; 169.32; 200.22 (d, J(C,P) ¼ 6.5). ESI-
MS:494 ([ M þ Na]þ). Anal. calc. for C22H38NO6PSi:C 56.03, H 8.12, N 2.97, P 6.57, Si 5.96; found:C
56.32, H 8.29, N 2.74, P 6.37, Si 6.09.
(3S)-3-{[(tert-Butyl)dimethylsilyl]oxy}-N1-methoxy-N1-methyl-N5-[(1S)-1-phenylethyl]pentanedi-
amide (6). To 3c (150 g, 410.3 mmol) and CH2Cl2 (1400 ml), 4-methylmorpholine (84.70 g, 820.7 mmol)
was added under mechanical stirring at r.t. The clear soln. was cooled to ꢀ 208, and isobutyl
carbonochloridate (57.19 g, ca. 98% (w/w) purity, 410.3 mmol) was added to the mixture at ꢀ 15 to
ꢀ 208. The mixture was stirred for 15 min at ꢀ 15 to ꢀ 208 and then treated with N-methoxymethan-
amine hydrochloride (40.43 g, 410.3 mmol). Stirring was continued for an additional hour at ꢀ 15 to
ꢀ 208, and the mixture was allowed to warm to r.t. After stirring for additional 4 h at r.t., the reaction was
quenched by addition of H2O (1400 ml). The biphasic mixture was stirred. Then the H2O layer was
extracted with CH2Cl2 (2 ꢁ 1000 ml), the combined org. phase washed with brine (1400 ml), dried
(MgSO4 (40 g)), and concentrated. The crude, wet product 6 (177.2 g) recrystallized from hexanes
(670 ml), and the crystalline product was dried at 408 under reduced pressure:160.23 g (96%) of 6.
HPLC:purity of 100 area-%. M.p. 69.6 – 70.1 8. [a]2D0 ¼ ꢀ27 (c ¼ 1.0, CHCl3). IR:3304, 2933, 1666, 1630,
1
1541, 1493, 1454, 1380, 1257, 1107, 1083, 940, 831, 779, 698. H-NMR ((D6)DMSO, 400 MHz):0.033 ( s,
3 H); 0.059 (s, 3 H); 0.842 (s, 9 H); 1.353 (d, J ¼ 7, 3 H); 2.25 – 2.45 (m, 3 H); 2.64 – 2.74 (m, 1 H); 3.08 (s,
3 H); 3.64 (s, 3 H); 4.49 – 4.59 (m, 1 H); 4.89-4.98 (m, 1 H); 7.18 – 7.27 (m, 1 H); 7.28 – 7.35 (m, 4 H); 8.33
(d, J ¼ 8.3, 1 H). 13C-NMR ((D6)DMSO, 100 MHz): ꢀ 4.09; ꢀ 4.05; 18.58; 23.31; 26.61 (3 C); 32.38;
44.74; 48.55 (2 C); 61.87; 67.69; 126.78 (2 C); 127.39; 129.04 (2 C); 145.72; 169.54; 171.67. ESI-MS:431
([M þ Na]þ). Anal. calc. for C21H36N2O4Si:C 61.73, H 8.88, N 6.86, Si 6.87; found:C 61.90, H 9.04, N 6.66,
Si 6.84.
{(4R)-4-{[(tert-Butyl)dimethylsilyl]oxy}-2,6-dioxo-6-{[(1S)-1-phenylethyl]amino}hexyl}phosphonic
Acid Diethyl Ester (7) As described for 5, from 6 and methylphosphonic acid diethyl ester: 7. Viscous oil.

1078
Helvetica Chimica Acta – Vol. 90 (2007)
[a]2D0 ¼ ꢀ40.1 (c ¼ 1.0, CHCl3). IR (film):3297, 2956, 2930, 2856, 1717, 1652, 1542, 1374, 1254, 1163, 1026,
966, 837, 779, 700. 1H-NMR ((D6)DMSO, 400 MHz):0.033 ( s, 3 H); 0.044 (s, 3 H); 0.83 (s, 9 H); 1.22 (td,
J(H,H) ¼ 7.1, J(H,P) ¼ 1.0, 6 H); 1.32 (d, J ¼ 7.1, 3 H); 2.22 (dd, J ¼ 14.2, 4.6, 1 H); 2.31 (dd, J ¼ 14.2, 7.8,
1 H); 2.78 (dd, J ¼ 17.1, 6.1, 2 H); 3.21 (d, J(H,P) ¼ 22, 2 H); 3.98 – 4.08 (m, 4 H); 4.45 – 4.52 (m, 1 H);
4.87 – 4.96 (m, 1 H); 7.20 – 7.27 (m, 1 H); 7.26 – 7.36 (m, 4 H); 8.32 (d, J ¼ 8.1, 1 H). 13C-NMR (D6)DMSO,
125 MHz): ꢀ 4.48 (2 C); 16.52; 16.57; 18.13; 22.92; 26.19 (3 C); 42.7 (d, J(C,P) ¼ 124); 43.87; 48.09;
51.30; 62.15; 62.20; 66.04; 126.33 (2 C); 126.95; 128.62 (2 C); 145.24; 169.05; 200.95 (d, J(C,P) ¼ 5.9).
ESI-MS:522 ([ M þ Na]þ). Anal. calc. for C24H42NO6PSi:C 57.69, H 8.47, N 2.80, P 6.20, Si 5.62; found:C
56.76, H 8.48, N 2.51, P 6.16, Si 5.50.
(3R,6E)-3-{[(tert-Butyl)dimethylsilyl]oxy}-7-[2-cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]-5-
oxo-[(1S)-1-phenylethyl]hept-6-enamide (9). A soln. of 5 (424.92 g, 901 mmol) in EtOH (950 ml) was
cooled to 0 – 58, and fine powdered anh. K2CO3 (119.8 g, ca. 850 mmol) was added, followed by aldehyde
8 (250 g, 858.1 mmol). The mixture was diluted with EtOH (950 ml) and allowed to warm to r.t. After
mechanical stirring for additional 30 min at r.t., the mixture was heated to 408 and stirred for 24 h at 408.
As HPLC revealed the presence of more than 10% of 8 in the mixture, 5 (21.17 g, 44.9 mmol) and K2CO3
(6 g, 43 mmol) were added, and the mixture was stirred for additional 24 h at 408 and 3 h at 458. The
reaction was quenched by addition of 5% aq. citric acid soln. (1800 ml), and the mixture was extracted
t
with BuOMe (1 ꢁ 3500 ml, 2 ꢁ 2500 ml). The org. layer was washed with H2O (3500 ml) and brine
(3500 ml) and dried (MgSO4 (70 g)), and the solvent was evaporated at 408:crude 9 (660.1 g, quant.) as
highly viscous oil, used directly for the next step without purification. An anal. sample was purified by CC
(silica gel, hexane/AcOEt). [a]2D0 ¼ ꢀ28.8 (c ¼ 1.0, CHCl3). IR:3300 (br.), 2954, 2928, 2840, 1647 (br.),
1
1605, 1540, 1513, 1489, 1253, 1223, 1094, 1066, 837, 778, 763, 699. H-NMR ((D6)DMSO, 400 MHz): ꢀ
0.07 (s, 3 H); 0.05 (s, 3 H); 0.77 (s, 9 H); 1.04 – 1.10 (m, 2 H); 1.22 – 1.32 (m, 2 H); 1.33 (d, J ¼ 7.1,
3 H); 2.21 (dd, J ¼ 14.2, 4.9, 1 H); 2.30 – 2.43 (m, 2 H); 2.60 – 2.73 (m, 2 H); 4.47 – 4.57 (m, 1 H); 4.87 –
4.97 (m, 1 H); 6.25 (d, J ¼ 16.6, 1 H); 7.18 – 7.24 (m, 1 H); 7.25 – 7.42 (m, 9 H); 7.45 (t, J ¼ 7.3, 1 H); 7.60 (d,
J ¼ 16.6, 1 H); 7.72 (t, J ¼ 7.0, 1 H); 7.91 (d, J ¼ 8.6, 1 H); 8.33 (d, J ¼ 8.1, 1 H). 13C-NMR (D6)DMSO,
125 MHz): ꢀ 4.58; ꢀ 4.40; 11.09; 11.20; 16.30; 18.10; 22.87; 26.13 (3 C); 44.06; 48.11; 48.21; 66.78; 115.94
(d, J(C,F) ¼ 21, 2 C); 125.69; 126.33 (2 C); 126.41; 126.62; 126.95; 127.78; 128.59 (2 C); 128.93; 130.40;
132.29; 132.35; 132.53; 135.11; 139.69; 145.25; 145.9; 147.03; 160.07; 162.34 (d, J(C,F) ¼ 246); 169.05;
197.66. ESI-MS:659 ([ M þ Na]þ). Anal. calc. for C39H45FN2O3Si:C 73.55, H 7.12, N 4.40, F 2.98, Si 4.41;
found:C 73.32, H 7.16, N 4.21, F 3.03, Si 4.30.
(3R,6E)-7-[2-Cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]-3-hydroxy-5-oxo-[(1S)-1-phenylethyl]-
hept-6-enamide (10). A soln. of crude 9 (660.1 g, theor. 858.1 mmol 9) in EtOH (2400 ml) was cooled to
0 – 48, and 2m aq. HCl (657.1 g of soln., 1276 mmol) was added dropwise at 0 – 48. The mixture was
allowed to warm to 258 and stirred for additional 4 h, until an in-process control (HPLC) indicated the
completion of the reaction. The mixture was poured onto 2% aq. NaHCO3 soln. (7400 ml) and extracted
with AcOEt (5000 ml). The org. layer was washed with brine (1800 ml) and the combined aq. phase
extracted with AcOEt (2000 ml). The combined org. phase was dried (MgSO4) and the solvent
evaporated:561.2 g of a honey-like crude product. The crude product was dissolved in toluene (860 ml)
and hexane (1075 ml) was added. The mixture was stirred at 608 for 2 h and at 508 for 1 h. Then, a
second portion of hexane (1075 ml) was added at 508, and the suspension was allowed to cool down to r.t.
Stirring was continued overnight at r.t. and for additional 3 h at 08. The product was isolated by filtration
and the filter cake washed with cold hexane (500 ml) and dried in vacuo at 408:390.9 g of 10, i.e., 385.43 g
after correction to 100% pure 10 (86% from 8 and 78% from 5 over 2 steps). Yellow powder. HPLC:
t
purity of 98.6 area-%. M.p. 145 – 1498 (crystallization from BuOMe/hexane gave crystals of m.p. 150 –
1518). [a]2D0 ¼ ꢀ26.9 (c ¼ 1.0, CHCl3). IR:3470 (br.), 3344, 1693, 1631, 1603, 1549, 1513, 1488, 1409, 1344,
1
1219, 1055, 1030, 995, 768, 697. H-NMR ((D6)DMSO, 400 MHz):1.02 – 1.13 ( m, 2 H); 1.22 – 1.32 (m,
2 H); 1.33 (d, J ¼ 7, 3 H); 2.21 (dd, J ¼ 14.2, 5.6, 1 H); 2.28 (dd, J ¼ 14.2, 7.4, 1 H); 2.37 – 2.48 (m, 1 H);
2.54 – 2.66 (m, 2 H); 4.22 – 4.32 (m, 1 H); 4.87 (d, J ¼ 5.3, 1 H); 4.86 – 5.00 (m, 1 H); 6.29 (d, J ¼ 16.6,
1 H); 7.16 – 7.26 (m, 1 H); 7.26 – 7.42 (m, 9 H); 7.44 (t, J ¼ 7, 1 H); 7.55 (d, J ¼ 16.6, 1 H); 7.71 (t, J ¼ 7, 1 H ) ;
7.91 (d, J ¼ 8.2, 1 H); 8.26 (d, J ¼ 8.2, 1 H). 13C-NMR ((D6)DMSO, 125 MHz):10.72; 10.77; 15.90; 22.46;
43.2; 47.64; 47.77; 64.69; 115.50 (d, J(C,F) ¼ 21.3, 2 C); 125.205; 125.89 (2 C); 125.93; 126.13; 126.49;
127.32; 128.14 (2 C); 128.46; 129.90; 131.91 (d, J(C,F) ¼ 7.4, 2 C); 132.04 (d, J(C,F) ¼ 3.2); 134.45; 138.79;

Helvetica Chimica Acta – Vol. 90 (2007)
1079
144.72; 145.41; 146.55; 159.72; 161.88 (d, J(C,F) ¼ 245.5); 169.28; 197.87. ESI-MS:545 ([ M þ Na]þ), 523
([M þ H]þ). Anal. calc. for C33H31FN2O3:C 75.84, H 5.98, N 5.36, F 3.64; found:C 75.74, H 6.13, N 5.39,
F 3.63.
(3R,5S,6E)-7-[2-Cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]-3,5-dihydroxy-[(1S)-1-phenylethyl]-
hept-6-enamide (11). A suspension of NaBH4 (30.26 g, ca. 96% (w/w), 768 mmol) in dry THF (1877 ml)
was cooled to ꢀ 788. A soln. of diethylborinic acid methyl ester (54.85 g, 548.4 mmol) in THF (54.85 g)
was added within 15 min at ꢀ 788, and the mixture was mechanically stirred for additional 5 min at ꢀ 788.
A soln. of 10 (290.7 g, 556 mmol) in dry THF (488 ml) and MeOH (581 ml) was added slowly within 2.5 h
to the mixture, maintaining the reaction temp. at ꢀ 788. After stirring for an additional hour at ꢀ 788, the
mixture was poured onto an ice-cold soln. of NaHCO3 (92.15 g) in H2O (4610 ml). Isopropyl acetate
(7346 ml) was added, and the biphasic mixture was stirred, until two clear phases were formed. The aq.
layer was extracted again with isopropyl acetate (2 ꢁ 3150 ml), the combined org. phase washed with
brine (2 ꢁ 3150 ml), and the solvent evaporated at 408. The obtained solid yellow foam was dissolved in
isopropyl acetate (630 ml), and the soln. was warmed to 508. A 35% aq. H2O2 soln. (174.02 g,
corresponding to 60.907 g of H2O2, 1790 mmol) was added within 45 min at 508, and the mixture was
stirred for an additional hour at 508. The reaction was quenched by addition of brine (3980 ml) at 45 – 508,
and the mixture was diluted with isopropyl acetate (2390 ml). The biphasic mixture was stirred for 20 min
at 45 – 508. The org. layer was treated with an aq. Na2SO3 soln. (118 g in 2850 ml of H2O) at 45 – 508 and
the mixture stirred for 5 min. Then the org. layer was washed with aq., half-sat. NaCl soln. (2 ꢁ 2390 ml)
and dried (MgSO4). The solvent was evaporated, the obtained crude, yellow, solid foam (349 g) dissolved
in tBuOMe (706 ml), and the soln. cooled to 08. After stirring for 30 min at 08, the formed suspension was
warmed to 308 and stirred for 15 min at 308. The suspension was allowed to cool to 258 and stirred
overnight at 258. Finally, the suspension was stirred for 2 h at 08 and for 3 h at ꢀ 208. The product was
isolated by filtration, and the filter cake was washed with heptane (2 ꢁ 75 ml) and dried at 408 under
t
reduced pressure:274.8 g (80.6%) of 11 as a 1:1 solvate with BuOMe. HPLC:purity of 98.9 area-%.
M.p. 59 – 758. The crystalline product comprised 99.89% of the desired ꢀsynꢁ-(3R,5S)-diastereoisomer
and 0.11% of its anti-(3R,5R)-epimer according to HPLC. Additional 18.08 g (5.3%) of 11 were obtained
from the mother liquor by CC (silica gel) and subsequent crystallization. Total yield:292.88 g (86%) of
11 · tBuOMe. [a]D20 ¼ ꢀ27 (c ¼ 1.0, CHCl3). IR:3479, 3296, 2977, 1642, 1557, 1513, 1490, 1411, 1365, 1215,
1
t
1158, 1116, 1068, 763, 698. H-NMR ((D6)DMSO, 500 MHz):0.96 – 1.06 ( m, 2 H); 1.105 (s, BuOMe);
1.15 – 1.26 (m, 3 H); 1.32 (d, J ¼ 7, 3 H); 1.42 – 1.49 (m, 1 H); 2.17 (dd, J ¼ 14.2, 5, 1 H); 2.22 (dd, J ¼ 14.2,
7.5, 1 H); 2.49 – 2.56 (m, 1 H); 3.32 (s, tBuOMe); 3.77 – 3.85 (m, 1 H); 4.10 – 4.17 (m, 1 H); 4.69 (d, J ¼ 4.7,
1 H); 4.83 (d, J ¼ 4.4, 1 H); 4.90 – 4.98 (m, 1 H); 5.62 (dd, J ¼ 16.2, 6, 1 H); 6.47 (dd, J ¼ 16.2, 1.1, 1 H);
7.15 – 7.20 (m, 1 H); 7.23 – 7.35 (m, 9 H); 7.38 (ddd, J ¼ 8.2, 6.9, 1.2, 1 H); 7.63 (ddd, J ¼ 8.2, 6.9, 1.3, 1 H);
7.86 (d, J ¼ 8.4, 1 H); 8.23 (d, J ¼ 8.1, 1 H). 13C-NMR ((D6)DMSO, 125 MHz):11.13; 11.28; 15.86; 23.0;
27.26 (tBuOMe); 43.93; 44.95; 48.05; 49.16 (tBuOMe); 65.75; 69.22; 72.50 (tBuOMe); 115.71 (d, J(C,F) ¼
21, 2 C); 123.90; 126.00; 126.05; 126.11; 126.36 (2 C); 126.95; 128.60 (2 C); 128.81; 129.26; 129.89; 132.26
(d, J(C,F) ¼ 7.7); 132.52 (d, J(C,F) ¼ 7.9); 133.40 (d, J(C,F) ¼ 2.8); 142.37; 144.08; 145.18; 146.36; 160.94;
162.04 (d, J(C,F) ¼ 244); 170.26. Anal. calc. for C38H45FN2O4:C 74.48, H 7.40, N 4.57, F 3.10; found:C
74.47, H 7.30, N 4.64, F 3.13.
(3R,5S,6E)-7-[2-Cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]-3,5-dihydroxy-hept-6-enoic Acid
Calcium Salt 1:2) (¼ Pitavastatin). To a soln. of 11 (4.0 g, 6.53 mmol) in EtOH (40 ml), H2O (40 ml)
and NaOH powder (2.64 g, 66 mmol) were added, and the mixture was heated at 50 – 558 for 26 h, until
an in-process control (HPLC) indicated complete hydrolysis. Then, 1m HCl (59 ml, 59 mmol) was added
within 15 min. The solvent was distilled under reduced pressure, and the residue was dissolved in H2O
(80 ml). The H2O soln. was extracted with tBuOMe (3 ꢁ 80 ml), and the org. phase was removed. The aq.
phase was evaporated and the residue redissolved in H2O (176 ml). Then, 1m HCl (6.53 ml, 6.53 mmol)
was added to precipitate the acid, followed by the addition of AcOEt (176 ml). The mixture was stirred
for 15 min to obtain 2 clear phases. The org. phase was washed with H2O (90 ml). Charcoal (0.5 g) was
added to the org. phase, and the mixture was stirred at 30 – 358 for 2 h. Filter-aid (Cellflock; 1.0 g) was
added, and stirring was continued for additional 30 min. Filtration and evaporation of the solvent at 30 –
358 gave the crude (3R,5S,6E)-7-[2-cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]-3,5-dihydroxyhept-6-
enoic acid as a white solid.

1080
Helvetica Chimica Acta – Vol. 90 (2007)
The crude acid (2.55 g, 6.05 mmol) was suspended in H2O (40.5 ml), and NaOH (0.260 g, 6.5 mmol)
was added to obtain a clear soln. of the corresponding sodium salt. A CaCl2 soln. (0.399 g, 3.49 mmol) in
H2O (2 ml) was added to the sodium salt soln. The immediately formed suspension was stirred for 4 h at
20 – 258 and for 2 h at 15 – 178. The product was isolated by filtration and the filter cake washed with cold
H2O and dried in vacuo at 20 – 258:2.95 g of pitavastatin comprising 10.6% ( w/w) H2O as a white,
crystalline powder, corresponding to 2.637 g (92%) of pure, anh. pitavastatin. [a]2D0 ¼ þ22.9 (c ¼ 1,
MeCN/H2O 1:1) ([17]:[ a]2D0 ¼ þ23.1 (c ¼ 1, MeCN/H2O 1:1)). Column electrophoresis: absence of the
(3S,5R)-enantiomer (detection limit 0.05%). HPLC, MS, NMR, and IR:identical to an authentic sample
of pitavastatin, the drug substance of Levaloꢃ. Anal. calc. for C50H46CaF2N2O8:Ca 4.55%; found:Ca
4.51% for anh. pitavastatin.
(4R,6S)-6-{(1E)-2-[2-Cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]ethenyl}tetrahydro-4-hydroxy-
2H-pyran-2-one (NK-104). As described for pitavastatin, 1.16 g of 11 were converted to (3R,5S,6E)-7-[2-
cyclopropyl-4-(4-fluorophenyl)quinolin-3-yl]-3,5-dihydroxy-hept-6-enoic acid. The crude free acid was
suspended in toluene (25 ml), and the suspension was heated to reflux for 1 h in a H2O separator. The
solvent was evaporated and the crude product purified by FC (silica gel, tBuOMe); 568 mg (74%) of pure
t
NK-104. White solid. The product was crystallized from BuOMe/hexane. M.p. 137 – 1398. [a]2D0 ¼ þ9.8
(c ¼ 1, CHCl3) ([14]:m.p. 136 – 139 8; [a]D20 ¼ þ9.0 (c ¼ 1.0, CHCl3); [16]:m.p. 138 – 139 8; [a]2D0 ¼ þ8.84
(c ¼ 0.92, CHCl3)). IR and NMR:in accord with NK-104. HR-MS:404.16561 ([ M þ H]þ; calc.
404.16565). Anal. calc. for C25H22FNO3:C 74.43, H 5.50, N 3.47, F 4.71; found:C 74.20, H 5.57, N 3.29, F
4.63.
REFERENCES
[1] C. E. Nakamura, R. H. Abeles, Biochemistry 1985, 24, 1364; Y. Jahng, Drug Future 1995, 20, 387.
[2] A. Endo, M. Kuroda, Y. Tsujita, J. Antibiot. 1976, 29, 1346; A. Endo, M. Kuroda, K. Tanzawa, FEBS
Lett. 1976, 72, 323; A. Endo, Y. Tsujita, M. Kuroda, K. Tanzawa, Eur. J. Biochem. 1977, 77, 31; A.
Endo, J. Med. Chem. 1985, 28, 401.
[3] A. G. Brown, T. C. Smale, T. J. King, R. Hasenkamp, R. H. Thompson, J. Chem. Soc., Perkin Trans. 1
1976, 1165.
[4] A. W. Alberts, J. Chen, G. Kuron, V. Hunt, J. Huff, G. Hoffman, J. Rothrock, M. Lopez, H. Joshua, E.
Harris, A. Patchett, R. Monaghan, S. Currie, E. Stapley, G. Albers-Schonberg, O. Hensens,
J. Hirschfield, K. Hoogstein, J. Liesch, J. Springer, J. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 3957.
[5] A. Endo, J. Antibiot. 1979, 32, 852; A. Endo, J. Antibiot. 1980, 33, 334.
[6] M. Mueller, Angew. Chem., Int. Ed. 2005, 44, 362; R. Ozegowski, A. Kunath, H. Schick, Liebigs Ann.
Chem. 1993, 805.
[7] C. H. Heathcock, C. R. Hadley, T. Rosen, P. D. Theisen, S. J. Hecker, J. Med. Chem. 1987, 30, 1858;
S. J. Hecker, C. H. Heathcock, J. Am. Chem. Soc. 1986, 108, 4586; T. Rosen, C. H. Heathcock, J. Am.
Chem. Soc. 1985, 107, 3731.
[8] P. D. Theisen, C. H. Heathcock, J. Org. Chem. 1988, 53, 2374.
[9] D. S. Karanewsky, M. F. Malley, J. Z. Gougoutas, J. Org. Chem. 1991, 56, 3744; Y. Kawakami, J.
Hiratake, Y. Yamamoto, J. Oda, J. Chem Soc., Chem. Commun. 1984, 779.
[10] P. M. Wovkulich, P. C. Tang, N. K. Chadha, A. D. Batcho, J. C. Barrish, M. R. Uskokovich, J. Am.
Chem. Soc. 1989, 111, 2596; S. J. Danishefsky, B. Simoneau, J. Am. Chem. Soc. 1989, 111, 2599.
[11] D. V. Patel, R. J. Schmid, E. M. Gordon, J. Org. Chem. 1992, 57, 7143.
[12] H. Shin, B. S. Choi, K. K. Lee, H. Choi, J. H. Chang, K. W. Lee, Synthesis 2004, 16, 2629.
[13] P. V. Ramachandran, K. J. Padiya, V. Rauniyar, M. V. R. Reddy, H. C. Brown, J. Fluorine Chem.
2004, 125, 615.
[14] H. Iwasaki, M. Suzuki, H. Yanagawa, Y. Fujikawa, M. Kitahara, Abstract Paper, ꢀ111th Annual
Congress of Pharm. Soc. Japan, April, Tokyoꢁ, 1991, 29V, 3-042; Abstract Paper, ꢀXIth International
Symposium on Drugs Affecting Lipid Metabolism, Florenceꢁ, May, 1992.
[15] T. Minami, T. Hiyama, Tetrahedron Lett. 1992, 33, 7525; G. B. Reddy, T. Minami, T. Hanamoto, T.
Hiyama, J. Org. Chem. 1991, 56, 5752.

Helvetica Chimica Acta – Vol. 90 (2007)
1081
[16] S. Takano, T. Kamikubo, T. Sugihara, M. Suzuki, K. Ogasawara, Tetrahedron: Asymmetry 1993, 4, 201.
[17] M. Suzuki, Y. Yanagawa, H. Iwasaki, H. Kanda, K. Yanagihara, H. Matsumoto, Y. Ohara, Y. Yazaki,
R. Sakoda, Bioorg. Med. Chem. Lett. 1999, 9, 2977.
[18] N. Miyachi, Y. Yanagawa, H. Iwasaki, Y. Ohara, T. Hiyama, Tetrahedron Lett. 1993, 34, 8267.
[19] K. Prasad, K.-M. Chen, O. Repic, G. E. Hardtmann, Tetrahedron: Asymmetry 1990, 1, 307; K.-M.
Chen, G. E. Hardtmann, K. Prasad, O. Repic, Tetrahedron Lett. 1987, 28, 155.
Received January 26, 2007