Synthesis of the chiral side chain of statins - Lactone versus lactol pathway

Article abstract of DOI:10.1002/ejoc.200600849

4-O-Protected (4R,6S)-4-hydroxy-6-(hydroxymethyl)tetrahydropyran-2-ones (3) derived from enantiomerically pure (3R,5S)-3-hydroxy-5,6-(isopropylidenedioxy) hexanoates (2) are frequently considered as pivotal intermediates for the synthesis of pharmacologic

Full text of DOI:10.1002/ejoc.200600849

FULL PAPER
DOI: 10.1002/ejoc.200600849
Synthesis of the Chiral Side Chain of Statins – Lactone versus Lactol
Pathway[‡]
Vitali I. Tararov,*[a] Natalia Andrushko,[a] Vasyl Andrushko,[a] Gerd König,[b]
Anke Spannenberg,[a] and Armin Börner*[a,c]
Keywords: Drug design / Lactones / Alcohols / Synthetic methods
4-O-Protected
(4R,6S)-4-hydroxy-6-(hydroxymethyl)tetra- key intermediate on the way to atorvastatin. Phosphonate 9b
hydropyran-2-ones (3) derived from enantiomerically pure
(3R,5S)-3-hydroxy-5,6-(isopropylidenedioxy)hexanoates (2)
are frequently considered as pivotal intermediates for the
and aldehyde 4b are nearest intermediates to fluvastatin and
rosuvastatin lactones. Unfortunately, Wittig–Horner-reaction
of phosphonate 9b under basic conditions was unsuccessful.
synthesis of pharmacologically important statins. Remarka- Finally, it was not possible to oxidize hydroxy tetrahydropy-
bly, up to now no proof for this assumption can be derived
from the literature. Our study revealed that only silyl-type
ranone 3b to related aldehyde 4b. Apparently the main
reason for this unexpected behaviour of these compounds
protecting groups can be successfully employed for 3-O-pro- consists in the acidity of the hydrogen atoms in the 3-position
tection of initial ethyl (3R,5S)-3-hydroxy-5,6-(isopropylidene- of the tetrahydropyranone-2-one ring which facilitates the
dioxy)hexanoate (1). After cyclization, hydroxy lactone 3b fast elimination of tBuPh2SiOH and further decomposition of
was transformed into tosylate 5b and successively into iodide
6b. The latter was converted into phosphonate 9b. All stereo-
chemical assignments and the diastereomerical purity of the
intermediates were confirmed by an X-ray structural analysis
of tosylate 5b. Surprisingly, neither tosylate 5b nor iodide 6b
could be converted into corresponding nitrile 7b which is the
the pyranone core. In contrast, the desired transformations
could be performed with related lactol 13 in hand. Thus, a
successful alternative for the preparation of the side chain of
statins was discovered.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
synthetic approach depicted in Scheme 1 is the choice of
the protecting group (PG). It must be stable under acidic
conditions since lactonization of 2, which gives rise to 3, is
known to be assisted by acids.[4] Attempts to introduce a
benzyl protecting group failed. Only silyl protection was
proven to be successful. It can be easily achieved by using
commercially available tBuMe2SiCl or tBuPh2SiCl.
Introduction
Statins like atorvastatin, fluvastatin or rosuvastatin
(Scheme 1) are inhibitors of 3-hydroxy-3-methylglutaryl co-
enzyme A reductase (HMG-CoA reductase) and became
the standard of care for treatment of hypercholesterolemia
because of its efficacy, safety and long-term benefits.[1] For
instance, atorvastatin calcium was the first totally synthetic
HMG-CoA-reductase inhibitor developed and marketed as
a single enantiomer. Currently, it ranks as one of the top
ten drugs sold in the world. Different pathways exist for the
preparation of these compounds.[2]
In a preceding publication we reported the efficient prep-
aration of optically pure ethyl (3R,5S)-3-hydroxy-5,6-(iso-
propylidenedioxy)hexanoate (1) on a multi gram scale.[3]
The potential use of this compound lies in the preparation
of precursors of several statins. One crucial problem in the
Some intermediate compounds represented in Scheme 1
can be found in relevant databases. Thus, cyclization of lin-
ear compound 2a as a ca. 1:1 mixture of C-3 epimers into
a diastereomeric mixture of lactones 3a (60% yield) was
for the first time described in ref.[4] Diastereomerically pure
crystalline tosylate 5a was isolated in an overall 28% yield
by column chromatography. Later on, the synthesis of com-
pound 2a with a diastereoselectivity of 60% was achieved
in a four step synthesis (ca. 25% yield).[5] Interestingly, the
Me- and tBu-esters parent to 2a failed to give lactone 3a.[5]
Approaches detailed in ref.[4,5] utilize (3S)-3,4-(isopropylid-
enedioxy)butanal as the initial compound which can be pre-
pared from (S)-malic acid.[6] Recently, a five step chemo–
enzymatic procedure for the preparation of 3a starting from
phloroglucitol (cis-1,3,5-cyclohexantriol) with an overall
13% yield was reported.[7] Iodide 6a was synthesized inde-
pendently with 50% de by iodolactonization of the pro-
tected enantioenriched (76% ee[8] and 88% ee[9]) (R)-3-hy-
[‡] Part II of a series of manuscripts dedicated to the synthesis of
statins. For part 1, see ref.[3]
[a] Leibniz-Institut für Katalyse an der Universität Rostock e.V.,
A.-Einstein-Str. 29a, 18059 Rostock, Germany
[b] Ratiopharm GmbH,
Graf-Arco-Str. 3, 89070 Ulm, Germany
[c] Institut für Chemie der Universität Rostock,
A.-Einstein-Str. 3a, 18059 Rostock, Germany
Fax: +49-0381-1281-5202
E-mail: armin.boerner@catalysis.de
Eur. J. Org. Chem. 2006, 5543–5550
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5543

V. I. Tararov, A. Börner et al.
FULL PAPER
Scheme 1. Possible pathways to the most important statin lactons.
droxy-5-hexenoic acid. The latter was prepared by a tions, the transformation of diastereomerically pure 2a into
chemo–enzymatic approach. It is noteworthy to mention
that the chloro analogue of iodide 6a could be synthesized
in a six step sequence (34% overall yield) starting from (S)-
β-trichloromethyl-β-propiolactone.[10] Protected hydroxy
lactone 3b was synthesized by a 15-step sequence starting
from -glucose and transformed via tosylate 5b into iodide
6b with a 17% overall yield.[11] Alternatively, 6b was pre-
pared with 60% de by iodolactonization of protected en-
antioenriched (R)-3-hydroxy-5-hexenoic acid (76% ee)[7]
which was prepared on the basis of a chemo–enzymatic ap-
proach.[12]
While much synthetic work has been focused on the
preparation of lactones 3, 5 and 6, we were not able to find
in the chemical and patent literature the requisite subse-
quent transformations to targeted statins.[13] In order to
clarify this matter, we devised a synthetic program with the
aim to test whether the lactone methodology depicted in
Scheme 1 could be employed for the synthesis of statins.
lactone 3a with a reasonable yield. In our hands, this pro-
cedure gave only 22–35% of raw semi-crystalline lactone 3a.
Recrystallization from hexane according to the recommen-
dations cited above afforded pure lactone 3a in only ca.
10% yield. Careful control and modification, respectively,
of the reaction conditions (preheated bath, concentration
and time) did not improve the yield.
To find the reason for these negative results we envisaged
a detailed study. Thus, samples of 2a were heated in 80%
AcOH at 100 °C in a preheated bath for 0.5 h, 1 h and 3 h.
After subsequent immediate cooling, the volatile substances
were evaporated in high vacuum at ambient temperature,
and the residues were then analyzed by NMR spectroscopy.
The following conclusions could be derived from these ex-
periments: (1) the isopropylidene group is cleaved first, (2)
the tBuMe2Si group is not stable under these conditions
since significant desilylation takes place even after 30 min,
(3) the cyclization reaction does not reach completion. Even
after 3 h, the signals of the EtO group were still present in
the NMR spectrum; meanwhile, decomposition under the
formation of unsaturated compounds started. At 60 °C and
at 25 °C in 80% AcOH, we also observed desilylation; how-
ever, in general, the isopropylidene group was removed
faster. Mixtures of AcOH/H2O have also been recom-
mended for the deprotection of tBuMe2Si ethers.[14] How-
ever, on the basis of our experiments and literature data,
we can conclude that it is impossible to achieve high or even
moderate yields of lactone 3a under these conditions.
One of the reasons for this failure might be comparable
substitution rates of the ethoxy group by the primary hy-
droxy group and with the desilylation reaction proceeding
Results and Discussions
Testing of the Lactone Pathway
Protection of syn-β-hydroxy ester 1 with the use of tBu-
Me2SiCl or tBuPh2SiCl and imidazole proceeded smoothly
and afforded protected compounds 2a,b after chromato-
graphic purification in high yield. Cyclization of compound
2a (as ca. 1:1 mixture of C-3 epimers) in an 80% AcOH
water solution at 100 °C was reported to give after
recrystallization 60% of a mixture of diastereoisomers of
3a.[4] Later on, 51% yield of lactone 3a after recrystalli-
zation was reported starting from 2a with a 60% de.[5] Un- in parallel.[15] We tried to remove the ethoxy group first
fortunately, we were unable to achieve, under these condi- and to then achieve lactonization of acid 10 as depicted
5544
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5543–5550

Synthesis of the Chiral Side Chain of Statins – Lactone versus Lactol Pathway
FULL PAPER
in Scheme 2. Hydrolysis of ester 2a proceeded under mild firmed by X-ray crystallography (Figure 1). The expected
conditions to afford acid 10 in a yield of 93%. The progress trans orientation of the substituents in the lactone ring has
of lactonization of acid 10 could be followed by TLC. It is been established. Thus, the stereochemistry of alcohol 1 has
interesting to note that acid 10 is surprisingly stable in boil- been unambiguously proven.
ing toluene even after 9 h. Treatment of this acid with 2.5%
(w/w) of p-TsOH in benzene at room temp. overnight re-
sulted in the full conversion and gave, after aqueous work
up, a solid material which after recrystallization afforded
35% of lactone 3a contaminated with ca. 10% of unidenti-
fied material. From the mother liquors, a less polar solid
compound with reasonable purity (ca. 90%) could be iso-
lated. The structure of this dominant compound was as-
signed on the basis of its NMR spectra. Observed chemical
shifts and integrals fit well with the structure of bis-silylated
lactone 11. In the more polar solvent Et2O under similar
conditions, transformation of 10 was significantly slower.
Complete conversion of acid 10 required 3 d. After aqueous
work up and chromatography, 20% of lactone 11 was iso-
lated. Lactone 3a (24% of crude product) contained ca.
30% of hitherto unidentified compounds.
Figure 1. Crystal structure of tosylate 5b. The thermal ellipsoids
correspond to 30% probability.
We found that tosylate 5b could be prepared from
alcohol 1 without chromatographic separation of interme-
diates 2b and 3b. Final purification could be achieved by
recrystallization of 5b from a hexane/AcOEt mixture.
Unfortunately, it was not possible to achieve substitution
of the tosyl group by cyanide under a variety of conditions.
Thus, heating of 5b with an excess of NaCN in DMSO
(80 °C, 14 h) resulted in the formation of tBuPh2SiOH,
which was isolated after aqueous work up in 91% yield.
Desilylation was also observed at room temperature in
Scheme 2. Synthesis of lactone 3a.
Next, we tried to improve the yield of the lactonization DMSO in the presence of an excess of NaCN. After 14 h,
of ester 2a by using another acidic catalyst. We found that 63% of tBuPh2SiOH was produced. Employment of alter-
heating 2a in a 90% aqueous dioxane solution in the pres- native solvents, like THF or Me2CO, as well as the usage
ence of pTsOHxPy (10%w/w) at 100 °C for 30 min gave, of an excess of Me2C(OH)CN in THF in the presence of
after aqueous work up, column chromatography and BuLi or tBuOLi under anaerobic conditions was likewise
recrystallization from hexane, an analytically pure sample unsuccessful. A literature search revealed at least two other
of lactone 3a in 47% yield. Interestingly, the properties of precedents where substitution of a pTsO group by cyanide
lactone 3a prepared by this methodology differed strikingly failed.[17] Thus, the problem was overcome by utilization of
with the values reported: compare m.p. 96–98 °C with re- pClC6H4SO2 and pO2NC6H4SO2 as leaving groups and by
ported [ref.[5b] 72–74 °C;[16] [α]2D4 = +0.33 (c 1 or 10, CDCl3); performing the reaction in DMSO at room temperature.[17a]
[α]2D0 = –7.5 (c 1, CDCl3)[5b]]. Surprisingly, for tosylate 5a Unfortunately, for our purpose these methods did not work.
we observed a stronger coincidence: compare m.p. 107– Tosylate 2a bearing the tBuMe2Si protecting group also
109 °C with reported m.p. 106–108 °C[4,5b] and [α]D22 = +8.2 underwent decomposition in the presence of an excess of
(c 1, CDCl3) with reported [α]D20 = +5 (c 0.82, CDCl3).[5b] NaCN in DMSO at room temperature.
Although we managed to obtain lactone 3a in moderate
yield, this quantity was not sufficient to test the other reac- tosylate 5b were likewise unsuccessful. Decomposition was
tions outlined in Scheme 1. observed when HP(O)(OEt)2 was employed in the presence
Attempts to introduce phosphorus-containing groups via
In contrast to HO protection of the lactone with the tBu- of a base, by refluxing with P(OMe)3 or P(OEt)3 and by
Me2Si group, the related tBuPh2Si protecting group was re- heating with PPh3 at 160 °C.
vealed to be much more stable under acidic conditions.
Nevertheless nucleophilic substitution of the pTsO group
Thus, heating of ester 2b in 80% AcOH (100 °C, 1 h) af- in compound 5b is possible. For example, it can be easily
forded lactone 3b in 81% yield after chromatographic puri- transformed into iodide 6b.[11] In our preparation, iodide 6b
fication. Also, pTsOHxPy (10%w/w) in 90% dioxane/water had a melting point equal to the reported value but its spe-
(100 °C, 1 h) mediated the cyclization of 2b to lactone 3b cific rotation differed significantly from the reported one:
with 98% yield. Although tosylate 5b was reported earlier, compare [α]2D3 = –9.5 (c 1, Me2CO) and reported [α]2D4
it was not characterized.[11] Now its structure could be con- –0.89 (c 1.08, Me2CO).[11]
=
Eur. J. Org. Chem. 2006, 5543–5550
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5545

V. I. Tararov, A. Börner et al.
FULL PAPER
Unfortunately, this transformation also did not open up after purification by column chromatography. At tempera-
access to the preparation of desired cyanide 7b. Thus, reac- tures below 10 °C, no conversion was observed. Homogeni-
tion of 6b with an excess of NaCN in DMSO at room tem- zation of the reaction mixture by exchange of CH2Cl2 with
perature resulted in quantitative desilylation. Reaction with acetonitrile gave a mixture of aldehydes and corresponding
1
neat PPh3 afforded a complex mixture. Interestingly, we acid (detected by H NMR spectroscopy).
found that iodide 6b could be transformed into phos-
Alternatively, the oxidation of alcohol 3b was tested in
phonate 9b [Px = P(O)(OEt)2] by refluxing in P(OEt)3 for the two-phase system CH2Cl2/H2O with the addition of
4 h with a 59% yield after isolation by column chromatog- 2 equiv. of iodine in the presence of a catalytic amount of
raphy. No product was formed in boiling P(OMe)3.
TEMPO (10 mol-%) and 3 equiv. of a 10% aqueous solu-
As a model reaction for the synthesis of statins, we tested tion of NaHCO3.[24] Usually, iodine acts as a superior
the reaction of phosphonate 9b with benzaldehyde. Unfor- chemoselective reoxidant of TEMPO compared to NaOCl.
tunately, in the presence of bases (DABCO in MeCN heated The reaction mixture was stirred overnight at 20 °C, fol-
at 82 °C or BuLi in THF cooled to –78 °C) this reaction lowed by work up with 10% aq. HCl and 10% aq.
resulted in decomposition. The only product isolated after Na2S2O3. After investigation of the reaction mixture, only
aqueous work up was tBuPh2SiOH.
traces of desired aldehyde 4b together with the correspond-
Finally, we tried to oxidize hydroxy lactone 3b into alde- ing acid were observed.
hyde 4b. Swern oxidation[18] gave a complex mixture of
products (6 different CHO signals were observed in the final
product). Pfitzner–Moffat[19] oxidation and usage of CrVI
Testing of the Lactol Pathway
reagents, for example PCC,[20] PDC,[21] CrO3·2Py,[22] only
resulted in the decomposition of the starting material.
As an alternative for the preparation of the function-
In further trials we employed 2,2,6,6-tetramethylpiperi- alized C7 side chain, we investigated whether an intermedi-
din-1-oxyl (TEMPO) in the presence of NaClO as an oxi- ate lactole would be more suitable. The sequence starting
dant. Usually, the TEMPO oxidation is mild and efficient with ethyl (3R,5S)-3-hydroxy-5,6-(isopropylidenedioxy)hex-
and has been demonstrated on a variety of primary anoate (2b) is represented in Scheme 3. First, reduction of
alcohols. In most cases the products were shown to be enan- ester 2b gave aldehyde 12 in good yield. Upon subsequent
tiomerically pure, which demonstrates the absence of any cleavage of the acetal, lactol 13 was obtained. Esterification
appreciable racemization during the formation of alde- of the primary hydroxy group with 4-chlorotoluenesulfonyl
hydes.[23] The oxidation of hydroxy lactone 3b was tested in chloride afforded sulfonate 14, which in turn was trans-
a two-phase CH2Cl2/H2O mixture, efficient stirring, with formed with NaCN into nitrile 15. Raney nickel catalyzed
1.1 equiv. of 1.26  aqueous NaClO in the presence of hydrogenation of the latter gave rise to desired amine 16. In
0.1 mol-% of TEMPO at 15 °C. After only a 3 min reaction order to show the usefulness of this material for the Paal–
time, a complex mixture of aldehydes, impurities and start- Knorr reaction, the amine was condensed with the appro-
ing compound 3b were observed. The stirring of the reac- priate functionalized diketone to give atorvastatin lactole
tion mixture overnight at 20 °C under the same conditions 17 in 71% yield. Oxidation and desilylation in order to ob-
also resulted in the formation of a complex mixture of alde- tain the free acid of atorvastatin can be conveniently per-
hydes and gave the corresponding acid in only 7% yield formed as recently suggested.[25]
Scheme 3. Successful synthesis of atovastatin by the lactol pathway.
5546
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5543–5550

Synthesis of the Chiral Side Chain of Statins – Lactone versus Lactol Pathway
FULL PAPER
7.7 Hz, 1 H, 6-CHaHb), 3.60 (dd, J = 7.7 and 6.1 Hz, 1 H, 6-
CHaHb), 3.84–3.96 (m, 2 H, OCH2CH3), 4.11–4.22 (m, 1 H, 3-CH),
4.52–4.62 (m, 1 H, 5-CH), 7.15–7.25 (m, 6 H, ArH), 7.72–7.83 (m,
4 H, ArH) ppm.1 3 C NMR (100 MHz, C6 D6 ): δ = 14.78
(OCH2CH3), 20.12 (CMe3), 26.60 (Me), 27.70 (Me), 27.78 (CMe3),
41.28 (4-CH2), 42.50 (2-CH2), 60.75 (OCH2CH3), 69.31 (3-CH),
70.24 (6-CH2), 73.22 (5-CH), 109.50 (CMe2), 128.55 (CH), 128.60
(CH), 130.61 (CH), 130, 64 (CH), 134.81 (C), 134.93 (C), 136.89
Conclusions
Although lactones 3, 5 and 6 were announced in the lit-
erature several times to be useful intermediates for the syn-
thesis of statins, our study clearly shows that they are not
stable under the conditions required for producing pivotal
precursors. Apparently, the main reason is associated with
the acidity of the hydrogen atoms in the 3-position of the
lactone ring which cause fast elimination of silanols and the
formation of unsaturated compounds.
As a more successful alternative, the employment of the
related lactol was shown. Relevant intermediates are stable
under the conditions of the incorporation of an exocyclic
amine group. Even under the conditions of the Paal–Knorr
reaction, the protected lactol was not affected. Thus, by this
methodology, atorvastatin lactol was obtained in good
overall yield which can be converted by known methods
into atorvastatin.[25]
1
(CH), 171.53 (COO) ppm. H NMR (400 MHz, CDCl3): δ = 1.04
(s, 9 H, CMe3), 1.19 (t, J = 7 Hz, 3 H, OCH2CH3), 1.25 (s, 3 H,
Me), 1.26 (s, 3 H, Me), 1.65–1.75 (m, 1 H, 4-CHaHb), 1.76–1.86
(m, 1 H, 4-CHaHb), 2.55 (dd, J = 15.2 and 6.3 Hz, 1 H, 2-CHaHb),
2.59 (dd, J = 15.2 and 6.2 Hz, 1 H, 2-CHaHb), 3.26 (dd, J = 8 and
8 Hz, 1 H, 6-CHaHb), 3.71 (dd, J = 8 and 5.9 Hz, 1 H, 6-CHaHb),
3.96–4.09 (m, 2 H, OCH2CH3), 4.13–4.23 (m, 1 H, 3-CH), 4.26–
4.37 (m, 1 H, 5-CH), 7.33–7.47 (m, 6 H, ArH), 7.61–7.73 (m, 4 H,
ArH) ppm. 13C NMR (100 MHz, CDCl3): δ = 14.43 (OCH2CH3),
19.57 (CMe3), 26.03 (Me), 27.11 (Me), 27.23 (CMe3), 40.54 (4-
CH2), 42.07 (2-CH2), 60.59 (OCH2CH3), 68.23 (3-CH), 69.71 (6-
CH2), 72.66 (5-CH), 108.92 (CMe2), 127.89 (CH), 127.97 (CH),
130.04 (CH), 130,10 (CH), 133.99 (C), 134.03 (C), 136.18 (CH),
171.54 (COO) ppm.
Experimental Section
(4R,6S)-4-tert-Butyldimethylsilyloxy-6-(hydroxymethyl)tetrahydro-
pyran-2-one (3a): A mixture of compound 2a (1.7 g, 4.91 mmol),
pTsOHxH2O (0.17 g), dioxane (6.8 mL) and water (0.68 mL) was
heated at 100 °C (preheated bath) for 30 min with stirring. The re-
sulted solution was cooled with water and diluted with AcOEt. The
mixture was washed successively with brine, saturated NaHCO3
solution and brine. It was then dried with Na2SO4. The solution
was concentrated and the residue was chromatographed on a SiO2
column (AcOEt/hexane 2:1). Fractions containing lactone 3a were
collected, and the solvents were evaporated. The residue was recrys-
tallized from hexane to afford lactone 3a (0.60 g, 47.0%). M.p. 96–
98 °C (Ref.[5b] 72–74 °C). [α]2D4 = +0.33 (c 1 or 10, CDCl3) {ref.[5b]
[α]2D0 = –7.5 (c 1, CDCl3)}. 1H NMR (400 MHz, CDCl3): δ = 0.064
(s, 3 H, SiMe), 0.071 (s, 3 H, SiMe), 0.87 (s, 9 H, CMe3), 1.73–1.82
(m, 1 H, 5-CHaHb), 1.86–1.97 (m, 1 H, 5-CHaHb), 2.30 (br. s, 1 H,
OH), 2.55–2.61 (m, 2 H, 3-CH2), 3.64 (dd, J = 12.5 and 4.8 Hz, 1
H, HOCHaHb), 3.89 (dd, J = 12.5 and 4.8 Hz, 1 H, HOCHaHb),
4.32–4.39 (m, 1 H, 4-CH), 4.73–4.82 (m, 1 H, 6-CH) ppm. 13C
NMR (100 MHz, CDCl3): δ = –4.61 (SiMe), –4.57 (SiMe), 18.25
(CMe3), 25.98 (CMe3), 32.21 (5-CH2), 39.50 (3-CH2), 63.78 (4-
CH), 64.98 (HOCH2), 77.15 (6-CH),170.5 (COO) ppm. C12H24O4Si
(260.402): calcd. C 55.35, H 9.29; found C 55.44, H 9.90.
General: Commercial reagents were used without additional purifi-
cation. NMR spectra were recorded with a Bruker ARX 400 spec-
trometer. For the description of the NMR spectra of lactones py-
ran-2-one, numeration of carbon atoms was used. Chemical shifts
1
are given relative to TMS as internal standard for H NMR and
relative to the residual CDCl3 peak for 13C NMR (δ
=
77.36 ppm).[26] Note that in some instances the aromatic carbon
atoms are not listed as part of the 13C NMR spectroscopic data.
This is due to a poor signal-to-noise ratio and extensive overlap-
ping of peaks. Optical rotations were measured with a “gyromat-
HP” instrument (Fa. Dr. Kernchen).
Ethyl (3R,5S)-3-Hydroxy-5,6-(isopropylidenedioxy)hexanoate (1):
Preparation of this compound by hydrogenation of ethyl (5S)-5,6-
(isopropylidenedioxy)-3-oxohexanoate has been described in ref.[3]
Ethyl
(3R,5S)-3-tert-Butyldimethylsilyloxy-5,6-(isopropylidenedi-
oxy)hexanoate (2a): To a stirred solution of alcohol 1 (2.26 g,
9.73 mmol) in DMF (4.5 mL) was added imidazole (1.5 g,
22.0 mmol) followed by tBuMe2SiCl (1.7 g, 11.3 mmol). The mix-
ture was stirred at 50 °C for 2 h. After cooling, the mixture was
diluted with water, and the product was extracted with AcOEt. The
extract was washed thoroughly with brine, dried with Na2SO4 and
the solvents evaporated. Product 2a (2.77 g, 82.2%) was isolated by
column chromatography on SiO2 with hexane/AcOEt (9:1).
(4R,6S)-4-tert-Butyldiphenylsilyloxy-6-(hydroxymethyl)tetrahydro-
pyran-2-one (3b): Method A: A mixture of compound 2b (7.63 g,
16.6 mmol) and AcOH (80%, 17 mL) was stirred at 100 °C (pre-
Ethyl (3R,5S)-3-tert-Butyldiphenylsilyloxy-5,6-(isopropylidenedi- heated oil bath) for 1 h. After cooling with cold water, the mixture
oxy)hexanoate (2b): A solution of alcohol 1 (4.0 g, 17.2 mmol) and
imidazole (2.5 g, 36.70 mmol) in DMF (8 mL) was cooled with
water, (10–15 °C) and tBuPh2SiCl (5.2 mL, 5.5 g, 20 mmol) was
added with stirring. The reaction mixture was stirred at room temp.
overnight. The mixture was diluted with water and AcOEt with
stirring. The organic layer was separated. The water phase was ex-
tracted additionally with AcOEt. The combined organic extracts
were washed with brine, dried with Na2SO4 and the solvents evapo-
rated. The residue was chromatographed on SiO2 (35ϫ8 cm col-
umn, hexane/AcOEt 9:1) to give 2b (7.63 g, 94.1%) as a very vis-
was diluted with water, and the product was extracted with AcOEt.
The extracts were washed thoroughly with saturated NaHCO3 and
brine, dried with Na2SO4 and the solvents evaporated. The residue
was subjected to column chromatography (AcOEt/hexane, 2:1) to
give lactone 3b (5.17 g, 81.1%) as a viscous oil. It was not possible
to remove traces of AcOEt even in high vacuum. Method B: A
mixture of 2b (1.23 g, 2.6 mmol), pTsOHxPy (0.12 g), dioxane
(5 mL) and water (0.5 mL) was stirred at 100 °C (preheated oil
bath) for 1 h. After cooling with water, the mixture was diluted with
AcOEt and washed successively with brine, saturated NaHCO3 and
cous colourless oil. 1H NMR (400 MHz, C6D6): δ = 0.92 (t, J = brine. The solution was dried with Na2SO4, and the solvents evapo-
7.1 Hz, 3 H, OCH2CH3), 1.15 (s, 9 H, CMe3), 1.23 (s, 3 H, Me), rated. The residue was dried in high vacuum to give lactone 3b
1.24 (s, 3 H, Me), 1.63–1.74 (m, 1 H, 4-CHaHb), 1.77–1.87 (m, 1 (0.98 g, 98%) containing only traces of dioxane and AcOEt. The
H, 4-CHaHb), 2.60 (dd, J = 15.3 and 5.9 Hz, 1 H, 2-CHaHb), 2.66
(dd, J = 15.3 and 6.5 Hz, 1 H, 2-CHaHb), 3.15 (dd, J = 7.7 and
sample did not require additional chromatographic purification. 1H
NMR (400 MHz, CDCl3): δ = 1.07 (s, 9 H, CMe3), 1.69–1.82 (m,
Eur. J. Org. Chem. 2006, 5543–5550
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5547

V. I. Tararov, A. Börner et al.
FULL PAPER
2 H, 5-CH2), 2.42 (dd, J = 17.7 and 4.1 Hz, 1 H, 3-CHaHb), 2.49–
2.62 (m, 2 H, 3-CHaHb+OH), 3.60 (dd, J = 12.3 and 4.8 Hz, 1 H,
CCDC-622927 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
HOCHaHb), 3.87 (dd, J = 12.3 and 3.0 Hz, 1 H, HOCHaHb), 4.31– Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
4.39 (m, 1 H, 4-CH), 4.86–4.94 (m, 1 H, 6-CH), 7.34–7.49 (m, 6 data_request/cif.
H, ArH), 7.59–7.66 (m, 4 H, ArH) ppm. 13C NMR (100 MHz,
(4R,6S)-4-tert-Butyldiphenylsilyloxy-6-(iodomethyl)tetrahydro-
CDCl3): δ = 19.39 (CMe3), 27.16 (CMe3), 31.67 (5-CH2), 39.09 (3-
pyran-2-one (6b): Tosylate 5b (3.44 g, 6.39 mmol) and NaI (9.6 g,
64 mmol) were refluxed in acetone (30 mL) overnight. Acetone was
CH2), 64.72 (4-CH), 64.88 (HOCH2), 77.29 (6-CH), 128.20 (CH),
130.41 (CH), 133.25 (C), 133.38 (C), 135.87 (CH), 135.91 (CH),
evaporated and the residue was diluted with water. The product
170.45 (COO) ppm.
was extracted with AcOEt. The combined extracts were washed
(4R,6S)-4-tert-Butyldimethylsilyloxy-6-(p-tosyloxymethyl)tetra-
hydropyran-2-one (5a): To a cold (ice-water bath) and stirred solu-
tion of lactone 3a (0.27 g, 1.04 mmol) in pyridine (1 mL) was added
pTsCl (0.23 g, 1.2 mmol) in one portion. After 1 h the bath was
removed, and the mixture was stirred overnight at ambient tem-
perature. The mixture was diluted with Et2O and washed success-
ively with brine, HCl (5%), saturated NaHCO3 and brine. The or-
ganic layer was dried with Na2SO4 and evaporated to give a solid
material. The solid was triturated under hexane/Et2O. The mixture
was stored overnight in a refrigerator. The solid material was fil-
tered and dried to give tosylate 5a (0.31 g, 72.1 %). M.p. 107–
109 °C (ref.[4,5b] 106–108 °C). [α]2D2 = +8.2 (c 1, CDCl3) {ref.[5b] [α]
with brine, dried with Na2SO4 and the solvents evaporated. The
yellow–brown residue was subjected to column chromatography
(SiO2, hexane/AcOEt 2:1). The isolated material was triturated un-
der hexane to cause crystallization and was left in a refrigerator
overnight. The solid product was filtered, washed with cold hexane
and dried to give the colourless iodide 6b (2.51 g, 79.5%). M.p. 77–
79 °C (ref.[11] 78–79 °C). [α]2D3 = –9.5 (c 1, Me2CO) {ref.[11] [α]2D4
=
–0.89 (c 1.08, Me2CO)}. 1H NMR (400 MHz, CDCl3): δ = 1.07 (s,
9 H), 1.60 (ddd, J = 14, 11.5 and 2.2 Hz, 1 H, 5-CHaHb), 2.05
(ddd, J = 14, 6 and 3.4 Hz, 1 H, 5-CHaHb), 2.42 (dd, J = 17.7 and
4.1 Hz, 1 H, 3-CHaHb), 2.60 (ddd, J = 17.7, 2.4 and 2.4 Hz, 1 H,
3-CHaHb), 3.35 (dd, J = 10.7 and 6.1 Hz, 1 H, ICHaHb), 3.38 (dd,
10.7 and 5.1 Hz, 1 H, ICHaHb), 4.28–4.36 (m, 1 H, 4-CH), 4.69–
4.78 (m, 1 H, 6-CH), 7.36–7.50 (m, 6 H, ArH), 7.58–7.67 (m, 4 H,
20
1
= +5 (c 0.82, CDCl3)}. H NMR (400 MHz, CDCl3): δ = 0.047
D
(s, 3 H, SiMe), 0.054 (s, 3 H, SiMe), 0.84 (s, 9 H, CMe3), 1.78–1.96
(m, 2 H, 5-CH2), 2.44 (s, 3 H, MeAr), 2.51–2.56 (m, 2 H, 3-CH2), ArH) ppm. 13C NMR (100 MHz, CDCl3): δ = 8.76 (ICH2), 19.41
4.14 (dd, J = 11 and 3.9 Hz, 1 H, OCHaHb), 4.19 (dd, J = 11 and
3.6 Hz, 1 H, OCHaHb), 4.31–4.37 (m, 1 H, 4-CH), 4.80–4.88 (m, 1
H, 6-CH), 7.35 (d, J = 8.1, 2 H, ArH), 7.78 (d, J = 8.1 Hz, 2 H,
ArH) ppm. 13C NMR (100 MHz, CDCl3): δ = –4.67 (SiMe), –4.63
(SiMe), 18.19 (CMe3), 21.98 (MeAr), 25.92 (CMe3), 32.34 (5-CH2),
39.33 (3-CH2), 63.48 (4-CH), 70.80 (OCH2), 73.29 (6-CH), 128.31
(CH), 130.31 (CH), 132.62 (C), 145.58 (C), 169.1 (COO) ppm.
(CMe3), 27.18 (CMe3), 36.25 (5-CH2), 38.82 (3-CH2), 64.52 (4-
CH), 74.64 (6-CH), 128.27 (CH), 130.48 (CH), 133.08 (C), 133.21
(C), 135.90 (CH), 169.37 (COO) ppm. C22H27IO3Si (494.438):
calcd. C 53.44, H 5.50, I 26.01; found C 53.94, H 5.22, I 25.67.
(4R,6S)-4-tert-Butyldiphenylsilyloxy-6-(diethoxyphosphonomethyl)-
tetrahydropyran-2-one (9b): Iodide 6b (1.0 g, 2.02 mmol) was re-
fluxed in P(OEt)3 (2.5 mL) for 4 h. After this period no starting
compound could be detected by TLC. The mixture was subjected
to column chromatography on SiO2 in AcOEt. The crude product
(1.02 g) was purified additionally by column chromatography to
afford 9b (0.6 g, 58.9%) as a colourless viscous material contami-
(4R,6S)-4-tert-Butyldiphenylsilyloxy-6-(p-tosyloxymethyl)tetra-
hydropyran-2-one (5b): The compound was prepared as described
for the preparation of 5a starting with lactone 3b (3.82 g,
9.93 mmol), pTsCl (2.5 g, 13.1 mmol) and pyridine (10 mL). Aque-
ous work up and recrystallization of the raw product from hexane/
AcOEt (2:1) afforded tosylate 5b (3.81 g, 71.2%). An additional
crop of 5b (0.39 g, 7.3%) was isolated from the mother liquors.
1
nated with AcOEt. H NMR (400 MHz, CDCl3): δ = 1.07 (s, 9 H,
CMe3), 1.32 (t, J = 7.1 Hz, 3 H, OCH2CH3), 1.33 (t, J = 7.1 Hz, 3
H, OCH2CH3), 1.61–1.72 (m, 1 H) and 2.03–2.63 (m, 5 H, CH2
groups of lactone moiety), 4.05–4.21 (m, OCH2CH3), 4.24–4.32 (m,
1
M.p. 124–126 °C. [α]2D4 = +11.7 (c 1, CDCl3). H NMR (400 MHz,
CDCl3): δ = 1.04 (s, 9 H, CMe3), 1.66–1.84 (m, 2 H, 5-CH2), 2.36 1 H, 4-CH), 5.13–5.26 (m, 1 H, 6-CH), 7.35–7.49 (m, 6 H, ArH),
(dd, J = 17.6 and 4.0 Hz, 1 H, 3-CHaHb), 2.44 (s, 3 H, CH3), 2.54 7.59–7.67 (m, 4 H, ArH) ppm. 13C NMR (100 MHz, CDCl3): δ =
(dt, J = 17.6 and 2.2 Hz, 1 H, 3-CHaHb), 4.12 (dd, J = 10.9 and
4.0 Hz, 1 H, O-CHaHb), 4.18 (dd, J = 10.9 and 3.8 Hz, 1 H, O- 33.02 (d, JP,C = 140, PCH2) 36.78 (d, JP,C = 4.8, 5-CH2), 39.0 (3-
CHaHb), 4.33 (s, 1 H, 4-CH), 4.94–5.03 (m, 1 H, 6-CH), 7.30–7.50 CH2), 62.23 (d, JP,C = 1.9, POCH2), 62.30 (d, JP,C = 1.9, POCH2),
16.65 (OCH2Me), 16.70 (OCH2Me), 19.37 (CMe3), 27. 11 (CMe3),
(m, 8 H, ArH), 7.56–7.63 (m, 4 H, ArH), 7.78 (d, J = 8.5 Hz, 2 H, 64.56 (4-CH), 71.97 (6-CH), 128.18 (CH), 130.36 (CH), 130.38
ArH) ppm. 13C NMR (100 MHz, CDCl3): δ = 19.37 (MeAr), 21.97 (CH), 133.15 (C), 133.27 (C), 135.87 (CH), 135.92 (CH), 169.69
(CH3), 27.14 (CMe3), 31.87 (5-CH2), 38.94 (3-CH2), 64.43 (4-CH), (COO) ppm. 31P NMR (162 MHz, CDCl3): δ = 25.85ppm.
70.72 (HOCH2), 73.39 (6-CH), 128.27 (CH), 128.29 (CH), 130.29
(3R,5S)-3-tert-Butyldimethylsilyloxy-5,6-(isopropylidenedioxy)hexa-
noic Acid (10): A mixture of compound 2a (1.84 g, 5.31 mmol),
aqueous NaOH (2 , 5.5 mL, 11 mmol) and EtOH (5 mL) was
(CH), 130.52 (CH), 132.60 (C), 132.99 (C), 133.11 (C), 135.81
(CH), 135.87 (CH), 145.55 (C), 168.95 (COO) ppm. C29H34O6SSi
(538.727): calcd. C 64.65, H 6.36, S 5.95; found C 64.78, H 6.38, S
stirred at ambient temperature overnight. It was diluted with water
5.84.
and extracted with CHCl3. The water solution was mixed with
X-ray Crystallographic Study of Tosylate 5b: Data were collected
with a STOE-IPDS-diffractometer with the use of graphite-mono-
chromated Mo-Kα radiation. The structure was solved by direct
methods (SHELXS-86)[27] and refined by full-matrix least-squares
techniques against F2 (SHELXL-93).[28] XP (BRUKER AXS) was
used for structure representation. Space group P21; monoclinic; a
Et2O, cooled in an ice bath and acidified with aqueous NaHSO4
(2 , 6 mL) with vigorous stirring. The organic layer was separated
and the water layer was additionally extracted with AcOEt. Com-
bined extracts were washed with brine, dried with Na2SO4 and the
solvents evaporated. The residue was dried in high vacuum to give
acid 9 (1.58 g, 93.4 %) containing traces of AcOEt. 1H NMR
= 9.625(2), b = 10.660(2), c = 13.965(3) Å; β = 93.17(3)°; V = (400 MHz, CDCl3): δ = 0.036 (s, 3 H, SiMe), 0.045 (s, 3 H, SiMe),
1430.7(5) Å3; Z = 2; ρcalcd. = 1.251 gcm–3; 7677 reflections mea-
sured; 4441 were independent of symmetry and 3865 were observed
[I Ͼ2σ(I)]; R1 = 0.030; wR2 (all data) = 0.062, 334 parameters.
0.84 (s, 9 H, CMe3), 1.32 (s, 3 H, Me), 1.37 (s, 3 H, Me), 1.72 (ddd,
J = 14, 6.3 and 5.2 Hz, 1 H, 4-CHaHb), 1.87 (ddd, J = 14, 7.8 and
4.5 Hz, 1 H, 4-CHaHb), 2.53 (dd, J = 15.4 and 7 Hz, 1 H, 2-
5548
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5543–5550

Synthesis of the Chiral Side Chain of Statins – Lactone versus Lactol Pathway
FULL PAPER
CHaHb), 2.59 (dd, J = 15.4 and 5.7 Hz, 1 H, 4-CHaHb), 3.50 (dd,
J = 7.9 and 7.4 Hz, 1 H, 6-CHaHb), 4.04 (dd, J = 7.9 and 5.9 Hz,
zation from hexane. M.p. 94–95 °C [Ref.[29] 97–98 °C]. [α]2D2 = –21.2
(c 4.03, CHCl3) {ref.[29] [α]2D4 = –11.2 (c 4.03, CHCl3)} {ref.[30]
1
1 H, 6-CHaHb), 4.13–4.25 (m, 1 H, 3-CH), 4.22–4.32 (m, 1 H, 5- [α]2D4 = –11.3 (c 0.195, CHCl3)}. H NMR (400 MHz, C6D6): δ =
CH), 11.1 (br. s, 1 H, COOH) ppm. 13C NMR (100 MHz, CDCl3):
1.13 (s, 9 H, CMe3), 1.15–1.25 (m, 1 H, H of CH2), 1.29–1.38 (m,
δ = –4.64 (SiMe), –4.34 (SiMe), 18.21 (CMe3), 26.01 (CMe3), 27.22 1 H, H of CH2), 1.39–1.48 (m, 1 H, H of CH2), 1.87–1.97 (m, 1 H,
(Me), 41.14 (4-CH2), 42.10 (6-CH2), 67.04 (3-CH), 70.0 (6-CH2), H of CH2), 2.18 (br. s, 1 H, OH), 3.35 (s, 3 H, OMe), 3.39 (m, 1
72.60 (5-CH), 109.19 (CMe2), 177.6 (COO) ppm.
H, OCHaHb), 3.52–3.62 (m, 1 H, OCHaHb), 4.11–4.22 (m, 2 H, 2-
CH, 4-CH), 5.02 (dd, J = 9.5 and 2.2 Hz, 1 H, 6-CH), 7.13–7.25
(m, 6 H, ArH), 7.62–7.70 (m, 4 H, ArH) ppm. 13C NMR
(100 MHz, C6D6): δ = (C6D6): 20.01 (CMe3), 27.81 (CMe3), 35.02
(CH2), 39.75 (CH2), 56.52 (OMe), 66.44 (OCH2), 68.00 (CH), 72.26
(CH), 100.6 (OCHO) ppm, aromatic C are omitted. C23H32O4Si
(400.583): calcd. C 68.96, H 8.05; found C 69.43, H 8.10.
Cyclization of Acid 10: A mixture of acid 10 (0.6 g, 1.88 mmol),
pTsOHxH2O (15 mg) and benzene (3 mL) was stirred overnight at
ambient temperature. An excess of saturated NaHCO3 was added
followed by AcOEt. The organic layer was separated and washed
with brine and dried with Na2SO4. The solvent was evaporated to
dryness leaving a solid material which was dissolved in boiling hex-
ane. The solution was cooled and kept in a refrigerator. The pre-
cipitate was filtered, washed with hexane and dried to yield lactone
3a (0.17 g, 34.7 %) having ca. 90 % purity. Mother liquors were
evaporated and the residue was fractionated by column chromatog-
raphy on SiO2 with hexane/acetone (4:1). First fraction was col-
lected. Evaporation of the solvents gave a solid material (0.8 g,
11.9 %). Main compound (ca. 90 % content) was identified as
(4R,6S)-4-tert-butyldimethylsilyloxy-6-(tert-butyldimethylsilyloxy-
methyl)tetrahydropyran-2-one (11) on the basis of NMR spectra
(2R,4R,6R)-4-tert-Butyldiphenylsilyloxy-2-cyanomethyl-6-methoxy-
tetrahydropyran (14): A solution of hydroxylactol 8 (8.3 g,
0.0207 mol) in pyridine (15 mL) was cooled with ice water, and
pClC6H4SO2Cl (5.51 g, 0.0261 mol) was added with stirring. After
1 h, the bath was removed and the mixture was stirred overnight
at room temp. Water (5 mL) was added, and the mixture was stirred
for an additional 1 h in order to destroy the excess sulfonyl chlo-
ride. The mixture was diluted with water and the product was ex-
tracted with AcOEt. The organic extract was washed successively
with brine, ca. 2 N aqueous HCl (until the washing solution be-
came acidic) and brine, dried with Na2SO4 and evaporated to give
14 (11.5 g, 96.4%) as a slightly yellow thick oil. This was dissolved
in DMSO (30 mL). After addition of NaCN (4.1 g, 0.083 mol), the
mixture was stirred at room temp. for 4 d. The mixture was diluted
with water and was stirred for an additional 1 h to dissolve all of
the inorganic material. The nonsoluble residue was filtered off,
washed with water and dried in vacuo to give 15 as a brownish
solid (8.13 g, 95.8%). For purification it was dissolved in a mixture
of toluene/AcOEt (10:1) and filtered through a 2–3 cm layer of SiO2
to remove coloured impurities. SiO2 was washed with the same mix-
ture and the combined washings were evaporated. The residue was
recrystallized from hexane/EtOH. After keeping in a refrigerator
overnight, the crystals were filtered off, washed with hexane and
dried in air to give cyanolactole 15 (6.70 g, 78.9% based on starting
1
which are given below. H NMR (400 MHz, CDCl3): δ = 0.071 (s,
3 H, SiMe), 0.073 (s, 3 H, SiMe), 0.078 (s, 3 H, SiMe), 0.078 (s, 3
H, SiMe), 0.88 (s, 9 H, CMe3), 0.89 (s, 9 H, CMe3), 1.83–1.96 (m,
2 H, 5-CH2), 2.51–2.63 (m, 2 H, 3-CH2), 3.74 (dd, J = 11 and
3.5 Hz, 1 H, OCHaHb), 3.83 (dd, J = 11 and 4.7 Hz, 1 H, OCH-
aHb), 4.34–4.40 (m, 1 H, 4-CH), 4.65–4.73 (m, 1 H, 6-CH) ppm.
13C NMR (100 MHz, CDCl3): δ = –5.10 (SiMe), –5.04 (SiMe),
–4.60 (SiMe), –4.56 (SiMe), 18.27 (CMe3), 18.56 (CMe3), 25.99
(CMe3), 26.13 (CMe3), 32.99 (5CH2), 39.74 (3-CH2), 63.84 (4-CH),
65.23 (OCH2), 76.53 (6-CH), 170.38 (COO) ppm.
(3R,5S)-3-tert-Butyldiphenylsilyloxy-5,6-(isopropylidenedioxy)hexa-
nal (12): A solution of ester 2b (3.34 g, 0.0071 mol) in Et2O (15 mL)
was cooled to –78 °C and DIBALH in toluene (1.5 , 5.1 mL,
0.0077 mol) was added within 5 min. The reaction mixture was kept
under the same conditions for an additional 10 min. MeOH
(10 mL) was then added. The cooling bath was removed and the
mixture was stirred at room temp. for 2 h. The precipitate formed
was filtered off and washed with Et2O. Combined washings were
evaporated and the product (thick oil) was isolated by column
chromatography on SiO2 (hexane/AcOEt 9:1) to give 2.34 g
(77.2%) of aldehyde 12. 13C NMR (100 MHz, C6D6): δ = 20.04
(CMe3), 26.57 (Me), 27.66 (Me), 27.75 (CMe3), 41.45 (4-CH2),
50.81 (2-CH2), 67.94 (3-CH), 70.22 (6-CH2), 72.98 (5-CH), 109.61
(CMe2), aromatic C are omitted, 200.60 (CHO) ppm.
hydroxylactole 8) as colourless crystals. M.p. 129–30 °C. [α]2D2
=
1
–23.0 (c 1, EtOH). H NMR (400 MHz, C6D6): δ = 0.80–0.90 (m,
1 H, H of CH2), 1.11 (s, 9 H, CMe3), 1.24–1.38 (m, 2 H, H, H of
CH2), 1.70 (dd, J = 16.6 and 5.9 Hz, 1 H, CHaHbCN), 1.77 (dd, J
= 16.6 and 6.5 Hz, 1 H, CHaHbCN), 1.81–1.89 (m, 1 H, H of CH2),
3.34 (s, 3 H, OMe), 3.94–4.04 (m, 2 H, 2-CH, 4-CH), 4.88 (dd, J
= 9.4 and 2.1 Hz, 1 H, 6-CH), 7.12–7.27 (m, 6 H, ArH), 7.58–7.67
(m, 4 H, ArH) ppm. 13C NMR (100 MHz, C6D6): δ = 19.96
(CMe3), 24.46 (CH2CN), 27.77 (CMe3), 38.02 (CH2), 39.12 (CH2),
56.56 (OMe), 67.15 (CH), 67.46 (CH), 100.59 (OCHO), 117.56
(CN) ppm, aromatic C are omitted. C24H31NO3Si (409.593): calcd.
C 70.38, H 7.63, N 3.42; found C 70.79, H 7.42, N 3.36.
(2R,4R,6S)-4-tert-Butyldiphenylsilyloxy-2-hydroxymethyl-6-meth-
oxytetrahydropyran (13): Aldehyde 12 (2.12 g, 0.00497 mol) was
dissolved in MeOH (5 mL) and HC(OMe)3 (2 mL) were added fol-
lowed by pTsOHxPy (0.2 g). The resultant solution was placed in
a hot bath (70–80 °C) and refluxed for 1 h. After cooling, water
and a saturated aqueous solution of NaHCO3 were added and the
product was extracted with AcOEt. Combined extracts were dried
with Na2SO4 and the solvents evaporated to dryness. The residue
was triturated under hexane to cause crystallization. The mixture
was kept in a refrigerator overnight. The solid product was filtered
off and dried to give 13 (1.0 g, 50.2%). An additional amount of
13 could be isolated from the mother liquors. These were evapo-
rated, and the residue was dissolved in MeOH (5 mL). pTsOH
(0.05 g) was added, and the solution was left at room temp. over-
night. The product (0.27 g, 13.5%) was isolated after aqueous work
up as given above. Analytical sample was prepared by recrystalli-
(2R,4R,6R)-2-(2-Aminoethyl)-4-tert-butyldiphenylsilyloxy-6-meth-
oxytetrahydropyran (16): A 30-mL autoclave was charged with cya-
nolactol 15 (0.52 g, 0.00127 mol), Raney-Ni (0.25 g of wet catalyst
washed 3 times with MeOH prior to hydrogenation), MeOH (8 mL)
and methanolic NH3 solution (7 , 2 mL). Hydrogenation was car-
ried out at 50 bar initial H2 pressure and room temp. After 5 h,
consumption of H2 ceased. The catalyst was decanted and washed
with MeOH. The methanol washings were evaporated. The residue
was dissolved in MeOH, and the solution was filtered through a
small pad of Celite to remove particles of the catalyst. Celite was
washed with MeOH. The clear solution was evaporated and dried
in vacuo to afford aminolactol as a colourless thick oil in quantita-
1
tive yield. H NMR (400 MHz, CDCl3): δ = 1.089 (s, 9 H, CMe3),
1.2–1.9 (complex multiplets, 8 H, four CH2), 3.51 (s, 3 H, OMe),
Eur. J. Org. Chem. 2006, 5543–5550
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5549

V. I. Tararov, A. Börner et al.
FULL PAPER
[9] S. J. Maddrell, N. J. Turner, A. Kerridge, A. J. Willetts, J.
Crosby, Tetrahedron Lett. 1996, 37, 6001–6004.
[10] M. Shimizu, K. Ishii, T. Fujisawa, Chem. Lett. 1997, 765–766.
[11] E. Baader, W. Bartmann, G. Beck, A. Bergman, H.-W.
Fehlhaber, H. Jendralla, K. Kesseler, R. Saric, H. Schüssler, V.
Teetz, M. Weber, G. Wess, Tetrahedron Lett. 1988, 29, 2563–
2566.
4.01–4.15 (m, 1 H, CH), 4.22–4.32 (m, 1 H, CH), 4.83 (dd, J = 9.6
and 1.9 Hz, 1 H, OCHO), 7.33–7.47 (m, 6 H, ArH), 7.59–7.68 (m,
4 H, ArH) ppm. 13C NMR (100 MHz, CDCl3): δ = 19.52 (CMe3),
37.31 (CMe3), 38.89 (CH2), 38.96 (CH2), 56.44 (OMe), 67.35 (CH),
69.53 (CH), 99.94 (OCHO) ppm, aromatic C are omitted.
Atorvastatin Lactol 17: A mixture of aminolactol 16 (0.48 g,
1.6 mmol), the relevant diketone (0.5 g, 0.00120 mol), pivalic acid
(0.1 g, 0.979 mmol) and solvent (n-heptane-THF/MePh, 100:50:60,
5 mL) were heated at reflux for 30 h under a slow flow of Ar. After
cooling, the mixture was diluted with AcOEt and washed success-
ively with saturated NaHCO3 solution and brine, dried with
Na2SO4 and the solvents evaporated. Flash chromatography of the
solid residue on SiO2 (hexane/AcOEt 5:1) afforded pyrrole 17
(0.66 g, 71.6%) as a yellowish solid. M.p. 169–171 °C. [α]2D2 = –24.4
[12] The enantiomer of iodide 6b was synthesized by iodolactoniz-
ation with 70% de and 70% ee by iodolactonization of pro-
tected enantioenriched (S)-3-hydroxy-5-hexenoic acid (71% ee)
prepared by a chemo–enzymatic sequence; see ref.[11]
[13] Probably the only related example is a model radical reaction
of iodide 6a with CH2=CHCH2SnBu3 or PhCH=CHCH2-
SnBu3 (ref.[8]), which is rather difficult to consider to be syn-
thetically valuable for the preparation of statins.
[14] E. J. Corey, A. Venkateswarlu, J. Am. Chem. Soc. J. Am. Chem.
Soc 1972, 94, 6190–6191. T. W. Greene, P. G. M. Wuts, Protec-
tive groups in Organic Synthesis, Wiley, New York, 1991, p. 80;
P. J. Kocienski, “Protecting groups” in The Foundation of Or-
ganic Chemistry Series (Eds.: D. Enders, R. Noyori, B. M.
Trost), Thieme, Stuttgart, 1994, p. 33.
1
(c 1, CHCl3). H NMR (400 MHz, C6D6): δ = (only characteristic
signals): 1.11 (s, 9 H, CMe3), 1.74 (d, J = 7.1 Hz, 3 H, CHMea),
1.75 (d, J = 7.1 Hz, 3 H, CHMeb), 3.33 (s, 3 H, OMe), 3.71 (sept.,
J = 7.1 Hz, 1 H, CHMe2), 3.83–3.98 (m, 2 H, CH2N), 4.91 (dd, J
= 9.6 and 1.9 Hz, 1 H, OCHO) ppm. 13C NMR (100 MHz, C6D6):
[15] Unexpected from this point of view is the fact that the methyl
ester parent to 2a afforded lactone 3a in 80% AcOH heated at
100 °C with a yield of 11% (ref.[5]).
δ
= 19.94 (CMe3), 22.53 (CHMea), 22.69 (CHMeb), 27.36
(CHMe2), 27.77 (CMe3), 38.56 (CH2), 38.60 (CH2), 39.54 (CH2),
42.15 (CH2), 56.17 (OMe), 68.00 (CH), 68.47 (CH), 100.33
(OCHO) ppm, aromatic C are omitted. C50H55FN2O4Si (795.067):
calcd. C 75.53, H 6.97, N 3.52; found C 75.62, H 6.93, N 3.10.
[16] M.p. 74–75 °C was reported for ca. 1:1 mixture of 4-C epimers
of lactone 3a (ref.[4]).
[17] a) U. Sunay, B. Fraser-Reid, Tetrahedron Lett. 1986, 27, 5335–
5338; b) P. L. Brower, D. E. Butler, C. F. Deering, T. V. Le, A.
Millar, T. N. Nanninga, B. D. Roth, Tetrahedron Lett. 1992, 33,
2279–2282.
[18] Handbook of Reagents for Organic Synthesis Oxidising and Re-
ducing Agents, (Eds.: S. D. Burke, R. L. Danheiser), John
Wiley & Sons, Weinheim, 1999, p. 154.
[19] K. E. Pfitzner, J. G. Moffatt, J. Am. Chem. Soc. 1965, 87, 5670.
[20] Handbook of Reagents for Organic Synthesis Oxidising and Re-
ducing Agents, (Eds.: S. D. Burke, R. L. Danheiser), John
Wiley & Sons, Weinheim, 1999, p. 323.
Acknowledgments
The authors are grateful to Ratiopharm GmbH (Ulm) for financial
support of this work. We appreciate skilled technical assistance by
Mrs. K. Mevius.
[21] Handbook of Reagents for Organic Synthesis Oxidising and Re-
ducing Agents, (Eds.: S. D. Burke, R. L. Danheiser), John
Wiley & Sons, Weinheim, 1999, p. 330.
[22] R. Ratliffe, R. Rodehorst, J. Org. Chem. 1970, 35, 4000–4002.
[23] P. L. Anelli, F. Montanari, S. Quici, Org. Synth. Coll. Vol. 1993,
8, 367; P. L. Anelli, F. Montanari, S. Quici, Org. Synth. 1990,
69, 212.
[1] E. S. Istvan, J. Deisenhofer, Science 2001, 292, 1160–1164.
[2] Contemporary Drug Synthesis (Eds.: J.-J. Li, D. S. Johnson,
D. R. Sliskovic, B. D. Roth),Wiley-Interscience, 2004, pp. 113–
123.
[3] V. I. Tararov, G. König, A. Börner, Adv. Synth. Catal., DOI:
10.1002/adsc.2006291.
[4] T. Rosen, M. J. Tashner, C. H. Heathcock, J. Org. Chem. 1984,
[24] R. A. Miller, R. S. Hoerrner, Org. Lett. 2003, 5, 285–287.
[25] Tararov, A. Korostylev, A. Börner, G. König, P. BobálЈ, J. Fran-
tisek, G. König, N. Jeker, K. Denike (ratiopharm GmbH), EP
06005510.0 (17.03.2006).
49, 3994–4003.
[5] a) J. Beecher, I. Brackenridge, S. M. Roberts, J. Tang, A. J. Wil-
letts, J. Chem. Soc., Perkin Trans. 1 1995, 1641–1643; b) J.
Tang, I. Brackenridge, S. M. Roberts, J. Beecher, A. J. Willetts,
Tetrahedron 1995, 51, 13217–13238.
[26] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997,
62, 7512–7515.
[6] See ref.[3] and literature cited therein.
[27] G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467.
[28] G. M. Sheldrick, SHELXL-93, University of Göttingen, Ger-
many, 1993.
[29] Y.-L. Yang, J. R. Falk, Tetrahedron Lett. 1982, 23, 4305–4308.
[30] A. P. Kozikowski, C.-S. Li, J. Org. Chem. 1985, 50, 778–785.
Received: June 19, 2006
[7] S. R. Ghorpade, U. R. Kalkote, S. P. Chavan, S. R. Bhide, T.
Ravindranatan, V. G. Puranic, J. Org. Chem. 2001, 66, 6803–
6806.
[8] F. Bennett, D. W. Knight, G. Fenton, Tetrahedron Lett. 1988,
29, 4865–4868; a) F. Bennett, D. W. Knight, G. Fenton, J.
Chem. Soc., Perkin Trans. 1 1991, 133–140.
Published Online: November 6, 2006
5550
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 5543–5550