Synthesis and highly stereoselective hydrogenation of the statin precursor ethyl (5S)-5,6-isopropylidenedioxy-3-oxohexanoate

Article abstract of DOI:10.1002/adsc.200600291

A search for the large-scale preparation of (5S)-5,6-(isopropylidenedioxy)- 3-oxohexanoates (2) - a key intermediate in the synthesis of pharmacologially important statins - starting from (S)-malic acid is described. The synthesis of the required initial compound methyl (3S)-3,4-(isopropylidenedioxy)butanoate (1) by Moriwake's reduction of dimethyl (S)-malate (3) has been improved. Direct 2-C chain elongation of ester 1 using the lithium enolate of tert-butyl acetate has been shown to be successful at a 3- to 5-fold excess of the enolate. Unfortunately, the product, tert-butyl (5S)-5,6-(isopropylidenedioxy)-3- oxohexanoate (2a) is unstable during distillation. Ethyl (5S)-5,6- (isopropylidenedioxy)-3-oxohexanoate (2b) was prepared alternatively on a multigram scale from (3S)-3,4-(isopropylidenedioxy)butanoic acid (7) by activation with N,N′-carbonyldiimidazole and subsequent reaction with Mg(OOCCH₂COOEt)₂. A convenient pathway for the in situ preparation of the latter is also described. Ethyl ester (2b) can be advantageously purified by distillation. The stereochemistry of the catalytic hydrogenation of β-keto ester (2b) to ethyl (55)-5,6-(isopropylidenedioxy)- 3-hydrohyhexanoate (syn-6 and anti-6) has been studied using a number of homogeneous achiral and chiral Rh(I) and Ru(II) complexes with phosphine ligands. A comparison of Rh(I) and Ru(II) catalysts with (S)- and (R)-BINAP as chiral ligands revealed opposite activity in dependence on the polarity of the solvent. No influence of the chiral backbone of substrate 2b on the enantioselectivity was noted. A ratio of syn-6/anti-6 = 2.3 was observed with an achiral (Ph₃P)₃RuCl₂ catalyst. Ru[(R)-Tol-BINAP]Cl₂ neutralized with one equivalent of AcONa afforded the most efficient catalytic system for the production of optically pure syn-(5S)-5,6-isopropylidenedioxy-3-hydroxyhexanoate (syn-6) at a preparative substrate/catalyst ratio of 1000:1.

Full text of DOI:10.1002/adsc.200600291

FULL PAPERS
DOI: 10.1002/adsc.200600291
Synthesis and Highly Stereoselective Hydrogenation of the Statin
Precursor Ethyl (5S)-5,6-Isopropylidenedioxy-3-oxohexanoate[1]
Vitali I. Tararov,a,* Gerd Kçnig,b and Armin Bçrnera,c,*
a
Leibniz-Institut für Katalyse an der Universität Rostock e.V., A.-Einstein-Str. 29a, 18059 Rostock, Germany
Ratiopharm GmbH, Graf-Arco-Str. 3, 89070 Ulm, Germany
Institut für Chemie der Universität Rostock, A.-Einstein-Str. 3a, 18059 Rostock, Germany
b
c
Fax : (+49)-0381-1281-5000; e-mail: armin.boerner@catalysis.de
Received: June 19, 2006; Accepted: September 29, 2006; Published on-line: November 28, 2006
Abstract: A search for the large-scale preparation of distillation. The stereochemistry of the catalytic hy-
(5S)-5,6-(isopropylidenedioxy)-3-oxohexanoates (2) drogenation of b-keto ester (2b) to ethyl (5S)-5,6-
– a key intermediate in the synthesis of pharmacolo- (isopropylidenedioxy)-3-hydrohyhexanoate
(syn-6
gially important statins – starting from (S)-malic acid and anti-6) has been studied using a number of ho-
is described. The synthesis of the required initial mogeneous achiral and chiral Rh(I) and Ru(II) com-
compound methyl (3S)-3,4-(isopropylidenedioxy)bu- plexes with phosphine ligands. A comparison of
tanoate (1) by Moriwakeꢀs reduction of dimethyl (S)- Rh(I) and Ru(II) catalysts with (S)- and (R)-BINAP
malate (3) has been improved. Direct 2-C chain as chiral ligands revealed opposite activity in de-
elongation of ester 1 using the lithium enolate of pendence on the polarity of the solvent. No influence
tert-butyl acetate has been shown to be successful at of the chiral backbone of substrate 2b on the enan-
a 3- to 5-fold excess of the enolate. Unfortunately, tioselectivity was noted. A ratio of syn-6/anti-6=2.3
the product, tert-butyl (5S)-5,6-(isopropylidene- was observed with an achiral (Ph3P)3RuCl2 catalyst.
dioxy)-3-oxohexanoate (2a) is unstable during distil- Ru[(R)-Tol-BINAP]Cl2 neutralized with one equiva-
lation. Ethyl (5S)-5,6-(isopropylidenedioxy)-3-oxo- lent of AcONa afforded the most efficient catalytic
hexanoate (2b) was prepared alternatively on a mul- system for the production of optically pure syn-(5S)-
tigram scale from (3S)-3,4-(isopropylidenedioxy)bu- 5,6-isopropylidenedioxy-3-hydroxyhexanoate (syn-6)
tanoic acid (7) by activation with N,N’-carbonyldiimi- at a preparative substrate/catalyst ratio of 1000:1.
dazole
and
subsequent
reaction
with
Mg(OOCCH2COOEt)2. A convenient pathway for
the in situ preparation of the latter is also described. Keywords: asymmetric catalysis; hydrogenation;
Ethyl ester (2b) can be advantageously purified by b-keto esters; rhodium; stereoselectivity
Introduction
droxy lactones C were formulated as possible building
blocks[4,5] as soon as the free hydroxy group could be
Inhibitors of the enzyme 3-hydroxy-3-methyl-glutaryl transformed into a variety of reactive leaving groups
coenzyme reductase (HMG-CoA reductase) commer- by functional group transformations and subsequent
cialized under the general trade name statins have coupling with appropriate building blocks.
become the standard of care for treatment of hyper-
A straightforward avenue to lactones C consists in
cholesterolemia due to the efficacy, safety and long- the cyclization of the protected hydroxy ester D.[4,6]
term benefits.[2] Atorvastatin calcium was the first to- The latter was prepared in low to reasonable diaste-
tally synthetic HMG-CoA-reductase inhibitor devel- reoselectivity with diastereomers that differ only in
oped and marketed as a single enantiomer. Currently, the configuration at C-3 (0% de,[4] 60% de,[6] 84%
it ranks at the top of the drugs best sold in the world. de[7]).
Biotechnologically produced or fully synthetic sta-
Our strategy is likewise shown in Scheme 1. It in-
tins[3] used in medical practice comply with general cludes two-carbon atom elongation of the acetonide 1
structures A and B (Scheme 1), where R is a hydro- to afford 3-keto esters 2 followed by catalytic hydro-
phobic heterocyclic or a carbocyclic moiety. Stereo- genation of the prochiral keto group to afford hy-
genic centers at C-3 and C-5 atoms must have the in- droxy esters D. To the best of our knowledge the cat-
dicated configurations. Structures A and B are equiv- alytic diastereoselective hydrogenation using molecu-
alent for retrosynthetic considerations. Protected hy- lar hydrogen of compounds 2 has never been studied
Adv. Synth. Catal. 2006, 348, 2633 – 2644
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2633

Vitali I. Tararov et al.
FULL PAPERS
Scheme 1. Retrosynthetic pathway to the enantiopure side chain of statins.
before. This transformation could open an economi-
Moriwakeꢀs reduction (BH3·SMe2, THF, cat.
cally and ecologically benign avenue to a series of sta- NaBH4)[9] allows the regioselective transformation of
tins. Besides the preparation of optically pure hydroxy one ester group of diester 3. Originally this transfor-
esters D, we were also interested to elaborate a mation was carried out by slow addition of a small
simple and efficient approach to 3-keto esters 2 which excess of the Me2S·borane complex to a THF solution
may be easily scaled up for industrial applications.
of the diester 3 followed by the addition of 5 mol%
of NaBH4 in one portion. At this point we were faced
by some difficulties. In one trial, addition of NaBH4
resulted in a gentle reaction with slow gas evolution.
In another trial, after the addition of NaBH4 a very
vigorous and uncontrolled reaction occurred with part
of the reaction mixture being splashed out. We no-
ticed also that not the whole amount of NaBH4 was
dissolved in the reaction mixture. This gave us the
Results and Discussions
Preparation of the Substrate for Hydrogenation
Preparation of the Intermediate Acetonide 1
The acetonide 1 is commercially available at a rather chance to reduce the amount of NaBH4 to 1 mol%.
high price. Alternatively, it can be prepared from (S)- We have found also that the reduction of diester 3
malic acid by a known sequence (Scheme 2). The in- could be achieved under controlled conditions when 1
termediate dimethyl (S)-malate (3) is also commer- mol% of NaBH4 was added to the solution of 3 in
cially available but for our purpose we improved the THF followed by addition of BH3·SMe2. In this case
protocol of the literature[8] by adding dimethyl ortho- the rate of the reaction (gas evolution) could be con-
formate during the esterification with methanol. Due trolled by the rate of the addition of the Me2S·borane
to this modification full conversion of (S)-malic acid complex. After the work-up the intermediate crude
was achieved and dimethyl ester 3 could be obtained product was transformed without additional chroma-
in one step with 93% yield.
tographic purification directly into the acetonide 1
with a yield of 74% after distillation.
2-C Elongation of Acetonide 1
An elongation of acetonide 1 by a 2-carbon unit af-
fording keto ester 2a can be achieved according to
Scheme 3 using an excess of the lithium enolate of
tert-butyl acetate. The reagent could be generated
with LDA. This approach was broadly used for the 2-
C elongation of 3-hydroxy esters in the preparation of
several 5-hydroxy-3-keto esters related to statins.[10] It
is also possible to elongate under these conditions
esters which lack a 3-hydroxy group,[11] but this is less
exploited. For the successful application of this mixed
ester condensation in the literature the use of a 3–4
Scheme 2. Preparation of the initial compound 1. a) MeOH,
HC
ACHTREUNG(OMe)3, cat. HCl, room temperature, 93%; b)
BH3·SMe2, THF, cat. NaBH4 (1 mol%), room temperature;
c) Me2C
74%.
ACHTREUNG(OMe)2, Me2CO, cat. p-TsOH, room temperature,
2634
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2633 – 2644

Statin Precursor Ethyl (5S)-5,6-Isopropylidenedioxy-3-oxohexanoate
FULL PAPERS
Scheme 3. Attempts for the synthesis of b-keto esters 2a,b by means of lithium enolates (other conditions: THF, ꢁ788C!
room temperature).
fold excess of LiCH2COO-t-Bu was recommended. been observed. Obviously the exclusive formation of
Interestingly, we were unable to find any reason for the lithium enolate of 2a in the presence of an excess
this in literature.
In order to clarify this matter, we undertook a de- and the formation of the tertiary alcohol 4.
tailed study of the reaction of acetonide 1 with The data presented in Table 1 were obtained using
of LiCH2COO-t-Bu prevents the second alkylation
LiCH2COO-t-Bu. We found that this reaction is com- an LDA solution prepared in situ from (i-Pr)2NH and
plicated by the formation of the bis-alkylated side- commercial BuLi (2.5M solution in hexane) for the
product 4. It was isolated by chromatography and generation of LiCH2COO-t-Bu. A commercial LDA
characterized by NMR. Our study revealed further- solution (ca. 2M solution in THF-heptane-EtPh)
more that the ratio 2a/4 is strongly dependent on the could be used as well without significant differences
ratio Li/1 (where Li corresponds to LiCH2COO-t- in the 2a/4 ratio. Unfortunately on the preparative
Bu). The data obtained are collected in Table 1. The scale it was not possible to purify keto ester 2a by dis-
1
ratios of products 2a:4 were determined by H NMR. tillation. Significant decomposition with tar formation
was observed and the yield of the pure material was
low (22%). The thermal instability of 2a is probably
Table 1. The influence of Li/1[a] ratio on the product ratio
due to the presence of the t-Bu ester group.
2a/4.
Attempts to synthesize the corresponding ethyl
ester 2b from acetonide 1 by application of the same
methodology using LiCH2COOEt resulted in a com-
plex mixture of unidentified products, probably due
to the easy oligomerization or polycondensation
which are hindered in the case of the t-Bu ester. Al-
though this mixture could be distilled in vacuum we
were not able to isolate pure 2b.
Run
Ratio Li/1
Ratio 2a/4
1
2
3
4
1:1
2:1
3:1
4:1
2.5:1[b]
1:1
3.5:1
8:1
[a]
Li corresponds to the portion of LiCH2COO-t-Bu.
Formation of unsaturated products was observed.
[b]
Finally attempts of transesterification of 2a to 2b
under acidic conditions were not successful since 2b is
At a ratio of Li/1=1:1 (run 1) the keto ester 4 was prone to rearrange to (2-furyl)acetate[12] in acidic
an important side-product. But in this trial not all ini- medium.
tial ester 1 was consumed and some additional unsa-
turated compounds of unidentified structure were ob-
served in the NMR spectrum. When the ratio of Li/1 2-C Elongation via Aldehyde 5
was ꢀ2 (runs 2–4) then complete conversion of ester
1 was achieved. Only two products could be observed Acetonide 1 could be reduced selectively to the alde-
in the NMR spectrum derived from run 3. An in- hyde 5 using diisobutylaluminium hydride (DIBALH)
crease of the Li/1 ratio resulted in an improvement of at ꢁ788C according to
a
published protocol
the yield of keto ester 2a (compare runs 3 and 4). In (Scheme 4).[13] The reduction proceeded cleanly and
run 4 additionally ca. 10% of MeCOCOO-t-Bu was gave the desired aldehyde 5 with reproducible yields
identified in the raw reaction mixture. In the best of 71–75% after distillation. The aldehyde 5 can be
runs 78–89% yield of keto ester 2a could be achieved elongated by insertion of the carbene generated from
using ratios of Li/1 in the range of 3:1 to 4:1 (runs 3 N2CHCOOEt catalyzed by SnCl2.[14] In general, this
and 4). The formation of the tertiary alcohol 4 is ki- reaction worked well, but we were faced by some pu-
netically controlled since by employment of an excess rification problems. Neither chromatography on SiO2
of LiCH2COO-t-Bu no formation of keto ester 2a has nor successive distillation gave the pure product 2b.
Adv. Synth. Catal. 2006, 348, 2633 – 2644
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2635

Vitali I. Tararov et al.
FULL PAPERS
Butyric acid derivative 7 could be easily prepared
from ester 1 by saponification and subsequent neu-
tralization. A known protocol utilized a 4-fold excess
of 2M LiOH water solution in THF.[17] We found that
complete hydrolysis can be achieved employing an
excess of 2M aqueous NaOH solution. The chemical
yield of the desired acid 7 was within a range 71–
78%. The acid could be successfully elongated by ap-
plying a general protocol developed for the synthesis
of b-keto esters.[18] Thus, imidazolide 8 was prepared
in situ and subsequently the solid complex Mg-
Scheme 4. Synthetic pathways via aldehyde 5. a) DIBALH,
Et2O, ꢁ78 C, 75%; b) N2CHCOOEt, CH2Cl2, cat. SnCl2,
room temperature; c) BrCH2COOEt, benzene, Zn, reflux,
71%; d) BrCH2COOEt, THF, Sn, room temperature, 41%.
AHCTRE(UNG MEMA)2 was added directly to the resultant solu-
tion. After work-up the raw keto ester 2b was isolated
in 84–94% yield. In contrast to the preparation of the
tert-butyl ester 2a the former could be distilled in
vacuum without decomposition and could be isolated
in 90% yield. The main advantage of this approach is
In another attempt the addition reaction of carban- that low temperatures or hazardous reagents are not
ions to aldehyde 5 affording alcohol D was studied. required.
Having aldehyde 5 in hand we achieved a hitherto un-
Originally the complex Mg
ACHTRE(UNG MEMA)2 was prepared
declared Reformatskii reaction. Thus, the reaction of by reaction of MgAHCTR(EUNG OEt)2 with a stoichiometric
aldehyde 5 with BrCH2COOEt promoted by activated amount of EtOCOCH2COOH (MEMAH)[19] in THF
Zn dust in boiling benzene[15] afforded a mixture of and subsequent evaporation and lingering drying in
alcohols 6 in 71% yield. No diastereoselection was vacuum with occasional crushing of the resultant
observed as found earlier for the addition of amorphous solid. In order to avoid this tedious proce-
LiCH2COOEt (generated from AcOEt and LDA).[4] dure and to find a more convenient way for triggering
The reaction promoted by activated Sn (generated the reaction of 8 with MEMAH, we tested different
from SnCl2 and LiAlH4)[16] in THF at room tempera- derivatives of this acid and additives. Main results are
ture gave a 41% yield of 6 with a diastereomeric compared in Table 2. C-Acylation of MEMA and its
ratio of 2:1 in favour of the undesired anti-isomer. K salt did not take place in the presence of simple ni-
The ratio of isomers 6 was established on the basis of trogen bases (runs 1–4). K and Li salts were not acy-
13C NMR spectroscopy (anti-isomer, d=73.5; syn- lated (runs 3 and 5). It was not even possible to ach-
isomer, d=74.6) and related to the reported data[7] ieve acylation of the Li dianion (runs 6 and 7). In con-
for the corresponding methyl ester (anti-isomer, d= trast, Zn and Al salts proved to be suitable for the
73.2; syn-isomer, d=74.4).
preparation of 2b (runs 8–10).
With Al salts some difficulties were faced during
the isolation of the product. With Zn salts the best
yield was observed when the symmetrical salt,
(MEMA)2Zn, was formed (run 9, compare with run
2-C Elongation via Acid 7
As an alternative a two-step sequence has been envis- 8). Although the yield of the product is moderate the
aged for the 2-C elongation of ester 1 (Scheme 5).
advantage consists in the in situ preparation of the
salt which does not demand any additional manipula-
tions. The usage of the unsymmetrical salt (MEM-
A)MgOAc, with the intention to reduce the quantity
of MEMA, resulted in a significant drop of the yield
(compare runs 12 and 11). Finally we found that the
most convenient way for the in situ preparation of
(MEMA)2Mg is heating a THF solution of MEMAH
with a slight excess of commercially available Mg
powder under reflux for 4 h. With this protocol keto
ester 2b was obtained after distillation in 82% yield
(run 13). The compound had satisfactory elemental
analysis and NMR spectra according to the data pre-
viously reported.[12] Unexpectedly, a striking contrast
was observed in the specific rotation: compare our
value [a]2D3: 6.6 (c 1.43, CHCl3) and reported [a]2D3:
ꢁ0.9 (c 1.43, CHCl3).[12] The reason for this strong de-
Scheme 5. Large scale preparation of substrate 2b for the
diastereoselective hydrogenation. a) 2 N NaOH, room tem-
perature and neutralization with 2 N NaHSO4, 78%; b) CDI
(N,N’-carbonyldiimidazole), THF, room temperature; c) Mg-
ACHTREUNG
2636
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2633 – 2644

Statin Precursor Ethyl (5S)-5,6-Isopropylidenedioxy-3-oxohexanoate
FULL PAPERS
Table 2. Comparison of procedures for C-acylation of MEMAH derivatives by imidazolide 8.[a]
Run Derivative[b]
Additive Yield of raw Procedure
2b [%][c]
1
2
3
4
5
6
7
8
MEMAH
MEMAH
MEMAK
MEMAK
MEMALi
A
G
G
ACHTREUNG
Et3N
pyridine traces
0
traces
5-fold molar excess of Et3N
5-fold molar excess of pyridine
Et3N
0
0
0
0
45
71
61
71
in situ addition of 1 mol equiv. of BuLi
in situ addition of 2 mol equivs. of BuLi
in situ addition of 2 mol equivs. of LDA
in situ addition of 1 mol equiv. of Et2Zn
in situ addition of 0.5 mol equivs. of Et2Zn
in situ addition of 0.5 mol equivs. of Me2AlCl
prepared by the reaction of stoichiometric amounts of Mg
MEMAH and AcOH with subsequent evaporation and drying in vacuum
prepared by the reaction of stoichiometric amounts of Mg(OEt)2,
9
10
11
ACHTREUNG
E
ACHTRE(UNG OEt)2,
12
13
R
84-94
82[d]
ACHTREUNG
MEMAH with subsequent evaporation and drying in vacuum
prepared in situ by refluxing of MEMAH with Mg in THF
C
[a]
[b]
[c]
General conditions: room temperature, THF, 16–20 h.
MEMAH=acid form, MEMA=monoanion, MEMA-H= dianion of enol form.
Yields of keto ester 2b were calculated on the raw product which was characterized as pure compound due to NMR (1H
and 13C) spectra.
[d]
After distillation (see Experimental Section).
viation is unclear. Nevertheless, in the present work stereoselectivity [runs 1–5; dppb=1,4-bis(diphenyl-
we observed the same sign and magnitude of rotation phosphino)butane]. A change of the solvent resulted
for structurally similar keto esters 2a and 2b, [a]2D3: 6.8 mainly in the change of the productivity while the
(c 1, EtOH) and [a]2D3: 7.8 (c 1, EtOH), correspond- diastereoselectivity was not affected. The Rh catalyst
ingly.
is more active in aprotic solvents than in MeOH. A
quite opposite tendency was noted for the achiral
complex (Ph3P)3RuCl2 (runs 7 and 8). It was more
active in MeOH than in aprotic solvents. In order to
find the desired product alcohol in the reaction mix-
Stereoselective Hydrogenations of Keto Ester 2b
In order to elucidate principal conditions for the ste- ture it was necessary to neutralize this complex with
reoselective hydrogenation of b-keto ester 2b to the Et3N (runs 9–11). The complex (Ph3P)3RuCl2·Et3N
desired syn-diol 6 (Scheme 6) first we studied the per- displayed the same solvent feasibility. In MeOH (run
formance of homogeneous achiral and chiral Rh and 9) reasonable diastereoselectivity was observed. From
Ru catalysts.
these experiments based on achiral catalysts we could
The syn/anti ratio was determined on the basis of expect that chiral Ru complexes must exhibit a rea-
NMR data. The hydrogenation was performed under sonable degree of diastereoselection in comparison
mild conditions (50 bar initial H2 pressure, room tem- with Rh complexes.
perature). The results of the hydrogenation of sub-
In Table 4 relevant examples of the hydrogenation
strate 2b with several homogeneous achiral catalysts of 2b with chiral Rh(I) catalysts are detailed. As ob-
are listed in Table 3. Hydrogenation of 2b catalyzed served with achiral analogues, chiral Rh(I) complexes
with RhACHTREUNG(dppb) catalysts did not give significant dia- gave higher activities in aprotic solvents (runs 1 and
2). The Rh catalyst with (S)-BINAP as ancillary
ligand induced reasonable syn-selectivity while with
(R)-BINAP as ligand a completely opposite result
was noted. Remarkably no influence of the chiral
backbone of the substrate on the syn-anti selectivity
was observed. A minor influence was detected with
pairs of (S,S)-(R,R)-Et-DUPHOS and corresponding
Me-DUPHOS ligands (runs 4 and 5, 6 and 7). Good
selectivity was observed also with catalysts based on
TolBINAP (run 8) and (R,S)-JOSIPHOS (run 9).
Scheme 6. Hydrogenation of keto ester 2b under homogene-
ously catalyzed conditions.
Adv. Synth. Catal. 2006, 348, 2633 – 2644
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2637

Vitali I. Tararov et al.
FULL PAPERS
Table 3. Table 3Hydrogenation of keto ester 2b with achiral homogeneous precatalysts.[a]
Run No
Precatalyst
Solvent
Time [min][b]
Conversion [%]
Ratio syn-/anti-6
1
2
3
4
5
6
7
8
[Rh
[Rh
[Rh
[Rh
[Rh
A
H
ACHTREUNG
A
C
MeOH
THF
Et2O
200
120
120
50
100
100
100
100
100
100
<10
<10
100
44
1.1
1.1
1.3
1.1
AcOEt
CH2Cl2
MeOH
CH2Cl2
THF
MeOH
CH2Cl2
THF
30
0.7
420
300
300
420
1200
1200
complex mixture
n.d.
n.d.
2.3
1.3
nd
9
10
11
ACHTREUNG
A
<10
[a]
Conditions: 2.5 mmol of substrate, 0.01 mmol of a precatalyst, 10 mL of solvent, room temperature, 50 bar initial H2 pres-
sure.
Indicated time corresponds to the end of H2 consumption.
The complex was prepared by treating commercial (Ph3P)3RuCl2 with an excess of Et3N in CH2Cl2 and subsequent evapo-
ration of the volatiles in vacuum (the real composition was not studied).
[b]
[c]
Table 4. Hydrogenation of keto ester 2b with chiral Rh precatalysts.[a]
Run
Precatalyst
Solvent
Time [min]
Conversion[%]
Ratio syn/anti
1
2
3
4
5
6
7
8
9
{Rh[(S)-BINAP]
{Rh[(S)-BINAP]
{Rh[(R)-BINAP]ACHTREUNG
C
MeOH
AcOEt
AcOEt
AcOEt
AcOEt
AcOEt
AcOEt
AcOEt
AcOEt
1000
120
120
70
100
100
100
100
100
100
100
100
100
3.3
3.1 (76/24)
0.34 (25/75)
2.1 (68/32)
0.7 (41/59)
2.3 (70/30)
0.78 (44/56)
0.34
{Rh
{Rh
{Rh
{Rh
C
ACHTREUNG
E
70
G
120
150
80
ACHTREUNG
Rh[(R)-Tol-BINAP]
{Rh
ACHTREUNG
G
A
50
2.3
[a]
Conditions:2.5 mmol of substrate, 0.01 mmol of a precatalyst, 10 mL of solvent, room temperature, 50 bar initial H2 pres-
sure.
In parallel we studied the catalytic performance of are characterized by d=ca. 110.[22] On the basis of
Ru(II) catalysts based on the diphosphine ligands these data we assumed that a thermodynamic equilib-
BINAP and Tol-BINAP, respectively. These catalysts rium is established between syn-6 and the rearranged
were introduced into chemical practice as highly product
9
during the hydrogenation reaction
enantioselective catalysts for the hydrogenation of b- (Scheme 7). This assumption was confirmed by a stan-
keto esters.[20] But significant improvement was neces- dard O-silylation procedure[4] of the mixture of alco-
sary to apply these catalytic systems for hydrogena- hols syn-6 and 9 followed by chromatographic separa-
tion of keto ester 2b considered herein. Results are tion of products 10 and 11. Compound 11 is less polar
detailed in Table 5.
than 10 and it was eluted first. Both isomers 10 and
First, we found that application of commercially 11 had the same element analysis and displayed the
available or prepared in-house[21] Ru[(R)-BINAP]Cl2 same peak pattern in the mass-spectrum (m/z=331
afforded a complex mixture of products although the [M+ꢁMe] and m/z=231 [M+ꢁ t-BuMe2Si]). The
conversion of keto ester 2b was complete (run 1). In chemical shift of the tertiary isopropylidene carbon
comparison with the related Rh(I) complex (Table 4 atom of compound 10 was found to be d=109 and for
runs 1 and 2) the latter catalyst did not work in compound 11 d=99.1. It is remarkable that 1,3-dioxo-
AcOEt (run 2). Along with the desired alcohol syn-6 lane 10 as well as 1,3-dioxane 11 showed only one set
(d=109.9)
a side product characterized in the of signals in the NMR spectrum. This gives clear evi-
13C NMR spectrum at d=99.4 was formed in MeOH. dence that the hydrogenation reaction produced alco-
It is well established that the chemical shift of the ter- hols syn-6 and 9 with excellent diastereomeric purity.
tiary isopropylidene carbon atom d=ca. 99, corre-
The same rearrangement was also observed when
sponds to a 6-membered ring while 5-membered rings Ru[(R)-Tol-BINAP]Cl2 was used as a catalyst. It
2638
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2633 – 2644

Statin Precursor Ethyl (5S)-5,6-Isopropylidenedioxy-3-oxohexanoate
Table 5. Hydrogenation of keto ester 2b with chiral Ru catalysts.[a]
FULL PAPERS
Run
Precatalyst
Solvent
Time [min]
Conversion [%]
Ratio syn/anti
1
2
3
4
5
6
7
8
Ru[(R)-BINAP]Cl2
Ru[(R)-BINAP]Cl2
MeOH
AcOEt
MeOH
MeOH
MeOH
MeOH
MeOH
EtOH
i-PrOH
96% aqueous EtOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
80
100
<10
100
100
0
100
100
100
0
syn-6 and 9
nd
>99% syn
98.2% syn
940
100
200
1310
380
120
300
340
340
140
750
1000
20 h
20 h
100
130
Ru[(R)-BINAP]Cl2 +0.5Et3N
Ru[(R)-BINAP]Cl2 +1.0Et3N
Ru[(R)-BINAP]Cl2 +2.0Et3N
Ru[(S)-BINAP]Cl2 +1.0Et3N
Ru[(R)-BINAP]Cl2·Et3N
Ru[(R)-BINAP]Cl2·Et3N
Ru[(R)-BINAP]Cl2·Et3N
Ru[(R)-BINAP]Cl2·Et3N
Ru[(R)-BINAP]Cl2 +1.0AcONa
Ru[(R)-BINAP]Cl2 +1.5AcONa
Ru[(R)-BINAP]Cl2 +2.0AcONa
2.3% syn
>99% syn
>99% syn
9
10
11
12
13
14
15
16
17
0
100
100
0
0
0
>99% syn
>98.1% syn
Ru[(R)-BINAP]
A
Ru((R)-BINAP)
A
Ru[(R)-TolBINAP]Cl2·Et3N
Ru[(R)-TolBINAP]Cl2·AcONa
100
100
>99% syn
>99% syn
[a]
Conditions:2.5 mmol of substrate, 0.01 mmol of a precatalyst, 10 mL of solvent, room temperature, 50 bar initial H2 pres-
sure.
Partial neutralization of Ru[(R)-BINAP]Cl2 could
also be achieved by treating the complex with an
excess of Et3N in CH2Cl2 followed by evaporation
and drying in vacuum. The resulted solid, designated
as Ru[(R)-BINAP]Cl2·Et3N {possible composition
might be Ru2Cl4[((R)-BINAP)]2·Et3N[23]}, was tested
in the hydrogenation of keto ester 2b. Run 7 illus-
trates that the activity of this complex is comparable
to the activity of the catalytic system arising after
neutralization in solution (compare with runs 3 and
4). The activity of the complex is strongly dependent
Scheme 7. Transacetalization reaction in acidic medium.
upon the polarity of the alcohol and follows the
order: MeOH>EtOH@i-PrOH (compare runs 7–9).
Surprisingly no activity was observed in 96% aqueous
became obvious that these complexes have acidic EtOH (run 10). We noticed that all previously studied
properties and it is necessary to neutralize them catalysts were poorly soluble in the reaction mixture.
before use. For this reason, we tested several nitrogen To overcome this problem we studied the influence of
bases (Et3N, BnNH2, imidazole, piperazine). These AcONa on the catalytic performance of Ru[(R)-BI-
bases had an equal influence on the activity of the NAP]Cl2 (runs 11–13). It was interesting to see that
catalysts. In Table 5 only examples with Et3N are AcONa acted in the same way as Et3N, e.g., lowering
listed (runs 3–5). The addition of Et3N inhibited com- the activity with the increase in the AcONa:Ru ratio
pletely the formation of the transacetalization product and complete inhibition of the hydrogenation at a
9. An increase in the concentration of Et3N affected ratio of 2 (runs 11–13). In the literature RuACHTREUNG
the activity of the catalyst. At an Ru/Et3N ratio equal
ACHTREUNG
to 2 complete inhibition of hydrogenation took place. tion of b-keto esters.[20a] Surprisingly, in our investiga-
It is interesting to note that imidazole and piperazine tions this complex prepared according to a known
possessing two basic centers behaved similarly to protocol[24] was completely ineffective in MeOH and
Et3N. The (R)-configuration of the BINAP ligand in- AcOEt (runs 14 and 15).
duced exclusively (within the analytical error) the
The parent Ru[(R)-TolBINAP]Cl2·Et3N complex
syn-configuration in the product 6 while (S)-BINAP exhibited a similar activity and syn-selectivity (run 16,
(run 6) had the opposite effect and anti-6 alcohol was compare with run 7). The solubility of this precatalyst
the predominant product. No significant influence of in MeOH was better than the analogous BINAP com-
the chiral backbone of the substrate has been ob- plex. Finally, we found that addition of 1 mol equiv.
served.
of AcONa during the preparation (see above) gave a
Adv. Synth. Catal. 2006, 348, 2633 – 2644
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2639

Vitali I. Tararov et al.
FULL PAPERS
solid mixture of Ru[(R)-Tol-BINAP]Cl2·AcONa (the Dimethyl (S)-Malate (3)
real composition was not determined) which per-
formed well (run 17).
Acetyl chloride (10 mL) was added to MeOH (100 mL).
After 30 min (S)-malic acid (50 g, 0.373 mol) followed by
Some complications were observed during our at-
tempts to increase the substrate to catalyst ratio.
Thus, additional products of unidentified structure
were observed when the hydrogenation was carried in
MeOH. Best results were obtained with Ru[(R)-Tol-
BINAP]Cl2·AcONa as a catalyst in EtOH. At 508C
and 100 bar H2 initial pressure it was possible to run
the hydrogenation procedure at a substrate to catalyst
ratio of 1000:1 with complete conversion within 4 h.
The product 2b was obtained with at least 99% de
and ca. 98% chemical purity.
HCACHTRE(UNG OMe)3 (80 mL, 77.6 g, 0.73 mol) were added to this so-
lution. The resultant solution was stirred at room tempera-
ture overnight. The volatiles were evaporated and the resi-
due was distilled in vacuum to afford 3; yield: 56.2 g
(92.9%); bp 66–688C/0.03 mbar. NMR spectra were in ac-
cordance to those recorded with a commercial sample.
Methyl (3S)-3,4-(Isopropylidenedioxy)butanoate (1)
To a solution of dimethyl (S)-malate (3) (113.4 g, 0.70 mol)
in THF (300 mL) NaBH4 (0.28 g, 0.0074 mol) was added in
one portion followed by slow addition of BH3·Me2S complex
(68 mL, 54.5 g, 0.72 mol) at room temperature with stirring.
During the addition the evolution of gaseous products took
place. The rate of gas evolution could be controlled by the
rate of the addition of the BH3·Me2S complex. When the ad-
dition was finished the reaction mixture was kept at ambient
temperature for 3 h. Then MeOH (285 mL) was added and
the resultant solution was left overnight at room tempera-
ture. The volatiles were evaporated and the viscous residue
was dried for 6 h in high vacuum. The residue was mixed
Conclusions
We have developed a convenient method for the
preparation of ethyl (5S)-5,6-isopropylidenedioxy-3-
oxohexanoate, a key intermediate in the synthesis of
pharmaceutically statins. Its catalytic hydrogenation
has been studied for the first time using a broad array
of achiral and chiral Rh(I) and Ru(II) complexes with
diphosphines as ancillary ligands. Interestingly, the in-
fluence of the chiral side chain in the b-keto ester on
the stereochemical course of the hydrogenation is
rather small. The activity of Rh(I) and Ru(II) com-
plexes with BINAP as ligand was opposite in depen-
dence on the polarity of the solvent. Moreover, the
ligand of a certain chirality induces opposite chirality
in the newly formed centre when coordinated to Rh
or Ru. The same effect was found earlier with a-acy-
lamino acrylic acids as substrates.[23] No diastereose-
lection was observed with Rh or Ru catalysts based
on BINAP. We found that Ru[(R)-Tol-BINAP]-
with acetone (300 mL), Me2CACHTRE(UNG OMe)2 (96.3 mL, 81.6 g, 0.78
mol) and p-TsOH·H2O (4 g, 0.021 mol). The reaction mix-
ture was left overnight at room temperature with stirring. It
was neutralized with solid Na2CO3 (4 g). The mixture was
stirred for additional 1 h, filtered off and evaporated. The
residue was distilled in vacuum (bp 748C/6 mbar) affording
compound 1; yield: 90.6 g (74.4%). NMR spectra were in
accordance to those obtained with a commercial sample.
General Procedure for Ester Condensation of
Acetonide 1 with MeCOO-t-Bu
All manipulations before work-up were done under Ar. A
2.5M hexane solution of BuLi (4.6”n mL, 11.5”n mmol)
was added to a solution of (i-Pr)2NH (1.7”n mL, 12.0”n
Cl2·AcONa is a highly efficient catalysts for the pro- mmol) in 10 mL of THF at 08C with stirring. After 5–10 min
the resulted solution was cooled down to ꢁ708C (EtOH/
duction of syn-(5S)-5,6-isopropylidenedioxy-3-hydroxy-
CO2 bath) and CH3COO-t-Bu (1.6”n mL, 11.9”n mmol)
hexanoate at a preparative substrate/catalyst ratio of
was added. The mixture was stirred at this temperature for
1000:1. In a subsequent publication we will describe
the application of syn-2b for the synthesis of statins.[25]
additional 30 min. Then neat acetonide 1 (2.0 g, 11.5 mmol)
was added dropwise (5–10 min). When addition was com-
plete the mixture was removed from the bath and stirred at
ambient environment for 1 h. The reaction was quenched by
addition of saturated NH4Cl solution (5”n mL) and the
product was extracted with AcOEt. The combined extracts
Experimental Section
were washed with brine and dried over Na2SO4. The solvent
was evaporated and the residue was analyzed by NMR. The
General Methods
MeOH and AcOEt (packed under N2) for hydrogenation
were purchased from Aldrich. CH2Cl2 was distilled over
CaH2, THF and Et2O were distilled over Na-Ph2CO under
Ar. Other commercial reagents were used without addition-
al purification. NMR spectra were recorded with a Bruker
ARX 400 spectrometer in CDCl3. Chemical shifts (d, in
results of runs 1–4 are presented in Table 1. Analytical sam-
ples of 2a and 4 were isolated from run 2 by column chro-
matography on SiO2 (AcOEt-hexane, 1:4), compound 4
being the less polar compound.
(5S)-tert-Butyl 5,6-(isopropylidenedioxy)-3-oxohexanoate
(2a): Colourless liquid; bp 102–1038C/0.06 mbar. During dis-
tillation significant decomposition was observed. An analyti-
cal sample was obtained by column chromatography on
SiO2 (hexane-AcOEt, 9:1). [a]2D3: 6.8 (c 1, EtOH). The opti-
cal purity was confirmed by GC on a 25 m Lipodex column
(658C) with a racemic sample as reference.
1
ppm) are given for H relative to TMS as internal standard
and for 13C relative to the residual CDCl3 peak (77.36 ppm).
Spin-spin coupling constants (J) are given in Hz. The optical
rotation was measured on a “gyromat-HP” instrument (Fa.
Dr. Kernchen).
2640
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2633 – 2644

Statin Precursor Ethyl (5S)-5,6-Isopropylidenedioxy-3-oxohexanoate
FULL PAPERS
About 10% of the relevant enol form was present in
CDCl3. Keto form: H NMR: d=1.35 (s, 3H, Me), 1.41 (s,
vacuum distillation did not give complete purification. Char-
acterization of pure keto ester 2b is given below.
1
3H, Me), 1.47 (s, 9H, CMe3), 2.74 (dd, 1H, J=17.1 and 7.0
Hz, 4-CHaHb), 2.99 (dd, 1H, J=17.1 and 5.9 Hz, 4-CHaHb),
3.38 (d, 1H, J=15.5 Hz, 2-CHaHb), 3.42 (d, 1H, J=15.5 Hz,
2-CHaHb), 3.57 (dd, 1H, J=8.3 and 6.6 Hz, 6-CHaHb), 4.19
(dd, 1H, J=8.3 and 6.1 Hz, 6-CHaHb), 4.43–4.51 (m, 1H, 5-
CH); 13C NMR: d=25.72 (Me), 27.15 (Me), 28.24 (CMe3),
47.38 (4-CH2), 51.29 (2-CH2), 69.60 (6-CH2), 71.73 (5-CH),
82.47 (CMe3), 109.24 (OCMe2O), 166.39 (COO), 201.46 (C=
Reformatski Reaction of Aldehyde 5
Zn promotion: Zinc dust was activated by washing with
dilute HCl solution (conc. HCl-H2O, 9:1). The resultant
powder was washed successively with water (till neutral
washings), acetone, Et2O and finally dried under high
vacuum. A solution of aldehyde 5 (1.0 g, 6.94 mmol) and
BrCH2COOEt (0.8 mL, 1.21 g, 7.23 mmol) in benzene
(5 mL) was added to a stirred and vigorously refluxed mix-
ture of benzene (5 mL), activated Zn dust (0.55 g,
8.41 mmol) and several crystals of I2 at such a rate as to
maintain the boiling of benzene. After the addition was
complete the resultant mixture was refluxed for additional
10 min. After cooling saturated NH4Cl solution and water
were added. The organic layer was separated, washed with
brine, dried over Na2SO4 and evaporated to afford a trans-
parent liquid; yield: 1.15 g (71.4%). Analysis of the NMR
spectra revealed a ca. 1:1 ratio of syn-6 and anti-6 alcohols.
Characterization of these products is given below.
1
O). Enol form: H NMR: d=1.36 (s, 3H, Me), 1.43 (s, 3H,
Me), 1.49 (s, 9H, CMe3), 2.36 (dd, 1H, J=14.2 and 6.4 Hz,
4-CHaHb), 2.51 (dd, 1H, J=14.2 and 6.6 Hz, 4-CHaHb),
3.66 (dd, 1H, J=8.3 and 6.5 Hz, 6-CHaHb), 4.10 (dd, 1H,
J=8.3 and 5.9 Hz, 6-CHaHb), 4.34–4.42 (m, 1H, 5-CH), 4.96
(s, 1H, C=CH), 12.26 (s, 1H, OH); 13C NMR: d=25.88
(Me), 27.24 (Me), 28.56 (CMe3), 39.99 (4-CH2), 69.44 (6-
CH2), 73.23 (5-CH), 81.32 (CMe3), 92.57 (C=CH), 109.51
(OCMe2O), 172.68 (COO), 173.63 (HOC=); anal. calcd. for
C13H22O5: C 60.45, H 8.58; found: C 60.35, H 8.64.
Di-tert-butyl
3-hydroxy-3-(2,3-isopropylidenedioxypro-
pyl)glutarate (4): Very viscous colourless oil contaminated
with AcOEt which was not possible to remove in vacuum.
1H NMR: d=1.35 (s, 3H, Me), 1.38 (s, 3H, Me), 1.46 (s, 9H,
CMe3), 1.47 (s, 9H, CMe3), 1.87 (dd, 1H, J=14.5 and 4.4
Hz, 4-CHaHb), 2.02 (dd, 1H, J=14.5 and 7.9 Hz, 4-CHaHb),
2.64 (s, 2H, 2-CH2), 2.65 (s, 2H, 3’-CH2), 3.51 (dd, 1H, J=
8.1 and 8.1 Hz, 6-CHaHb), 4.10 (dd, 1H, J=8.1 and 5.9 Hz,
6-CHaHb), 4.32–4.40 (m, 1H, 5-CH), position of OH is not
indicated; 13C NMR (CDCl3): d=26.14 ((Me), 27.09 (Me),
28.31 (CMe3), 42.80 (4-CH2), 43.82 (2-CH2), 45.42 (3’-CH2),
70.38 (6-CH2), 71.22 (3-C), 72.51 (5-CH), 81.23 (CMe3),
81.37 (CMe3), 108.91 (CMe2), 171.32 (COO), 171.35 (COO).
Sn promotion.[16] Commercial SnCl2 was dried under
vacuum at 1208C for 4 h. To a solution of SnCl2 (1.37 g,
7.23 mmol) in THF (5 mL) was added by portions LiALH
(0.14 g, 3.69 mmol) at 08C with stirring. When the addition
was complete the mixture was stirred for additional 10 min
at room temperature. To this mixture a solution of aldehyde
5
(1 g, 6.94 mmol) and BrCH2COOEt (0.8 mL, 1.21 g,
7.23 mmol) in THF (5 mL) was added with stirring. The re-
sultant mixture was stirred at room temperature for 4 h and
quenched by the addition of concentrated NH4Cl solution
followed by water. The product was extracted with AcOEt.
The thick precipitate caused difficulties during extraction.
The combined extracts were washed with brine, dried over
Na2SO4 and evaporated affording a liquid; yield: 0.66 g
(41%). Analysis of NMR spectra revealed a ca. 1:2 ratio of
syn-6 and anti-6 alcohol. Characterization of these products
is given below.
(3S)-3,4-(Isopropylidenedioxy)butanal (5)[13]
All manipulations before work-up were done under Ar. A
solution of acetonide 1 (10 g, 0.057 mol) in Et2O (100 mL)
was cooled down to ꢁ708C and a 1.5M solution of
DIBALH in toluene (8.4 mL, 0.0126 mol) was added over a
period of 5 min with stirring. The resultant clear solution
was stirred at this temperature for additional 10 min, then it
was quenched with MeOH (5 mL). The cooling bath was re-
moved and the solution was stirred at ambient conditions
for 2–3 h. The precipitate formed was filtered off and
washed with Et2O. The filtrate was concentrated and the
residue was distilled in vacuum to afford aldehyde 5; yield:
6.48 g (78.3%); bp 73–748C/15 mbar. NMR spectra were
similar to those reported in the literature.
(3S)-3,4-(Isopropylidenedioxy)butanoic acid (7)
Methyl (3S)-3,4-(isopropylidenedioxy)butanoate (1) (50 g,
0.287 mol) was added to an ice-cooled aqueous 2M NaOH
solution (287 mL, 0.574 mol) with stirring . The bath was re-
moved and the resultant mixture was stirred at room tem-
perature for 2 h. The mixture was extracted with CH2Cl2
(3”50 mL) and the organic extracts were discarded. The
water layer was mixed with Et2O (100 mL) and cooled in an
ice bath. To this mixture an aqueous NaHSO4 solution (2 N,
300 mL) was added. The mixture was vigorously stirred for
15 min. The organic layer was separated and the water layer
was additionally extracted with AcOEt (2”100 mL). The
combined organic extracts were dried over Na2SO4 and
evaporated. The residue was dried under high vacuum to
afford the liquid acid 7; yield: 36 g (78.3%). According to
its NMR spectra the product 7 had ca. 95% purity. It was
Reaction of Aldehyde 5 with N2CH2COOEt[14]
To
a
mixture of SnCl2 (0.13 g, 0.67 mmol) and
N2CH2COOEt (0.8 g, 7.0 mmol) in CH2Cl2 (15 mL) the alde-
hyde 5 (1.0 g, 6.94 mmol) was slowly added. When the addi-
tion was completed the reaction mixture was stirred at room
temperature for 2 h and concentrated under vacuum. The
residue was diluted with AcOEt. The resultant mixture was
filtered through a pad of SiO2. The SiO2 was washed with
AcOEt. The volatiles were evaporated and the residue was
purified by column chromatography on SiO2 (hexane-
AcOEt, 4:1) affording keto ester 2b; yield: 0.76 g (47.6%),
contaminated with side products of undetermined structure.
1
used further without additional purification. H NMR: d=
1.37 (3H, s, CH3), 1.43 (3H, s, CH3), 2.58 (1H, dd, J=16.2
and 6.7 Hz, 2-CHaHb), 2.75 (1H, dd, J=16.2 and 6.7 Hz, 2-
CHaHb), 3.68 (1H, dd, J=8.5 and 6 Hz, 4-CHaHb), 4.17
(dd, J=8.5 and 6 Hz, 4-CHaHb), 4.45–4.53 (1H, m, 3-CH),
11 (1H, br. s, COOH); 13C NMR: d=25.8 (CH3), 27.2
Adv. Synth. Catal. 2006, 348, 2633 – 2644
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2641

Vitali I. Tararov et al.
FULL PAPERS
(CH3), 39.2 (CH2), 69.4 (CH2), 72.1 (CH), 109.9 (C), 176.7
(COO).
(s, 1H, C=CH), 12.11 (s, 1H, OH); 13C NMR: d=14.45
(CH2CH3), 25.79 (Me), 27.17 (Me), 39.89 (4-CH2), 60.35
(CH2CH3), 69.30 (6-CH2), 73.12 (5-CH), 91.16 (C=CH),
109.50 (CMe2), 172.70 (COO), 174.33 (HOC=); anal. calcd.
for C11H18O5: C 57.38, H 7.88; found: C 56.85, H 7.87.
General Procedure for Reaction of Imidazolide 8
with MEMAH Derivatives
The acid 7 (0.9 g, 5.62 mmol) was reacted with N,N’-carbon-
yldiimidazole (CDI) (1 g, 6.17 mmol) in THF (5 mL) at
room temperature for 2 h with stirring. To the resultant solu-
tion the corresponding MEMAH derivative was added.
Solids were added directly and solutions (in 5 mL THF)
were introduced by syringe. The resultant reaction mixture
was stirred at room temperature overnight, evaporated and
the residue dissolved in AcOEt. The organic layer was sepa-
rated and washed successively with aqueous NaHSO4 solu-
tion (2M), saturated aqueous NaHCO3 solution and brine,
dried over Na2SO4 and evaporated. The residue was ana-
lyzed by NMR. Other conditions and results are presented
in Table 2.
General Procedure for Hydrogenation of b-Keto
Ester 2b (Catalysts Screening)
A detailed procedure is given in ref.[26] The screening was
performed in a 25-mL autoclave charged with b-keto ester
2b (0.58 g, 2.52 mmol), a precatalyst (0.01 mmol) and a sol-
vent (10 mL). The reaction was monitored by the pressure
decrease. When the reaction was completed the solvent was
evaporated and the residue was analyzed by NMR. The data
are listed in Table 4 and Table 5.
Ethyl (3R,5S)-5,6-(isopropylidenedioxy)-3-hydroxyhexa-
noate (syn-6): Prepared by hydrogenation with Ru[(R)-BI-
NAP]Cl2 catalyst with the addition of 1.0 mol equiv. of Et3N
(Table 5, run 4). The compound was not additionally puri-
1
fied. H NMR: d=1.28 (t, 3H, J=7.1 Hz, OCH2CH3), 1.36
Preparative Synthesis of Ethyl (5S)-5,6-
(Isopropylidenedioxy)-3-oxohexanoate (2b)
(s, 3H, Me), 1.42 (s, 3H, Me), 1.70–1.84 (m, 2H, 4-CH2),
2.49 (dd, 1H, J=16.0 and 4.8 Hz, 2-CHaHb), 2.55 (dd, 1H,
J=16.0 and 7.8 Hz, 2-CHaHb), 3.58 (dd, 1H, J=8.1 and 7.3
Hz, 6-CHaHb), 4.10 (dd, 1H, J=8.1 and 5.9 Hz, 6-CHaHb),
4.17 (q, 2H, J=7.1 Hz, OCH2CH3), 4.19–4.27 (m, 1H, 3-
CH), 4.27–4.35 (m, 1H, 5-CH), position of OH is not indi-
cated; 13C NMR: d=14.38 (OCH2CH3), 25.92 (Me), 25.09
(Me), 39.95 (4-CH2), 41.77 (2-CH2), 60.84 (OCH2CH3),
67.00 (3-CH), 69.72 (6-CH2), 74.63 (5-CH), 109.43 (CMe2),
172.39 (COO).
A mixture of monoethyl malonate (49.0 g, 0.371 mol) and
commercial Mg powder with particle size <0.1 mm (6.0 g,
0.248 mol) in THF (200 mL) was refluxed with stirring for
4 h to produce in situ Mg(OOCCH2COOEt)2. In another
flask, to a solution of the acid 7 (25.8 g, 0.161 mol) in THF
(100 mL) was added by portions solid CDI (28.8 g, 0.177
mol). The addition was accompanied by gas evolution. After
the addition was completed (5–10 min) the mixture was stir-
red at room temperature for 2 h to produce in situ imidazo-
lide 8. The resulted solution was mixed with the solution of
the Mg complex. The residual Mg powder was rinsed with
THF (50 mL) and the washings were added to the reaction
mixture. The reaction mixture was left with stirring over-
night at room temperature. The mixture was evaporated,
the residue was dissolved in AcOEt (200 mL) and acidified
with aqueous NaHSO4 solution (2M, 430 mL) with vigorous
stirring. The organic layer was separated, washed successive-
ly with aqueous NaHSO4 solution (2 N, 2”200 mL), saturat-
ed aqueous NaHCO3 solution (3”200 mL), dried over
Na2SO4 and evaporated. The residue was distilled in vacuum
(bp 90–938C/0.07 mbar) affording product 2b; yield: 30.5 g
(82.3%); bp 97–1008C/0.067 mbar; [a]2D3: 6.6 (c 1.43, CHCl3),
7.8 (c 1 in EtOH), {Ref.[12] [a]D23: ꢁ0.9 (c 1.43, CHCl3)}.
Ethyl (3S,5S)-5,6-(isopropylidenedioxy)-3-hydroxyhexa-
noate (anti-6): Prepared by the hydrogenation with Ru[(S)-
BINAP]Cl2 catalyst with the addition of 1.0 mol equiv. of
Et3N (Table 5, run 2). The compound was not additionally
1
purified. H NMR: d=1.28 (t, 3H, J=7.1 Hz, OCH2CH3),
1.34 (s, 3H, Me), 1.41 (s, 3H, Me), 1.70–1.75 (m, 2H, 4-
CH2), 2.47 (dd, 1H, J=16.2 and 8.5 Hz, 2-CHaHb), 2.55 (dd,
1H, J=16.2 and 3.9 Hz, 2-CHaHb), 3.57 (dd, 1H, J=8.1 and
7.5 Hz, 6-CHaHb), 4.10 (dd, 1H, J=8.1 and 5.9 Hz, 6-
CHaHb), 4.17 (q, 2H, J=7.1 Hz, OCH2CH3), 4.19–4.28 (m,
1H, 3-CH), 4.28–4.37 (m, 1H, 5-CH) position of OH is not
indicated; 13C NMR: d=14.37 (OCH2CH3), 25.90 (Me),
27.14 (Me), 40.95 (4-CH2), 41.97 (2-CH2), 60.90
(OCH2CH3), 65.69 (3-CH), 69.83 (6-CH2), 73.50 (5-CH),
108.96 (CMe2), 172.74 (COO).
About 10% of the relevant enol-form was present in
CDCl3. Keto-form: 1H NMR: d=1.29 (t, 3H, J=7.1 Hz,
OCH2CH3), 1.35 (s, 3H, Me), 1.40 (s, 3H, Me), 2.75 (dd,
1H, J=17.0 and 6.7 Hz, 4-CHaHb), 2.99 (dd, 1H, J=17.0
and 6.3 Hz, 4-CHaHb), 3.47 (d, 1H, J=15.9 Hz, 2-CHaHb),
3.51 (d, 1H, J=15.9 Hz, 2-CHaHb), 3.57 (dd, 1H, J=8.4
and 6.6 Hz, 6-CHaHb), 4.15–4.24 (m, 3H, 6-CHaHb+
OCH2CH3), 4.43–4.51 (m, 1H, 5-CH); 13C NMR: d=14.31
(CH2CH3), 25.65 (Me), 27.07 (Me), 47.36 (4-CH2), 49.95 (2-
CH2), 61.66 (CH2CH3), 69.48 (6-CH2), 71.67 (5-CH), 109.22
Characterization of Products Obtained by the
Hydrogenation under Ru[(R)-BINAP]Cl2 Catalysis
without the Addition of a Base (Table 5, run 1)
The hydrogenation product was mixed several times with n-
hexane and then evaporated. The residue was dried in high
vacuum and reacted with t-BuMe2SiCl under the same con-
ditions as given in ref.[4] The resultant product was purified
by chromatography on SiO2 (hexane-AcOEt. 9:1). Com-
pound 11 was eluted first. The raw compounds 10 and 11
were additionally purified under the same conditions.
1
(CMe2), 167.10 (COO), 200.97 (C=O). Enol-form: H NMR:
d=1.29 (t, 3H, J=7.2 Hz, OCH2CH3), 1.36 (s, 3H, Me),
1.42 (s, 3H, Me), 2.40 (dd, 1H, J=14.3 and 6.4 Hz, 4-
CHaHb), 2.53 (dd, 1H, J=14.3 and 6.9 Hz, 4-CHaHb), 3.67
(dd, 1H, J=8.3 and 6.3 Hz, 6-CHaHb), 4.10 (dd, 1H, J=8.3
and 5.9 Hz, 6-CHaHb), 4.15–4.24 (m, 2H, OCH2CH3), 5.06
Ethyl (3R,5S)-3-tert-butyldimethylsilyloxy-5,6-(isopropyli-
denedioxy)hexanoate (10): Colourless viscous liquid; [a]2D5:
1
ꢁ22.8 (c 1, EtOH); H NMR: d=0.015 (s, 3H, MeSi), 0.032
(s, 3H, MeSi), 0.83 (s, 9H, CMe3), 1.23 (t, 3H, J=7.1 Hz,
2642
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2633 – 2644

Statin Precursor Ethyl (5S)-5,6-Isopropylidenedioxy-3-oxohexanoate
OCH2CH3), 1.31, (s, 3H, Me), 1.35 (s, 3H, Me), 1.69 (ddd,
FULL PAPERS
Acknowledgements
1H, J=13.9, 5.8 and 5.8 Hz, 4-CHaHb), 1.85 (ddd, 1H, J=
13.9, 7.2 and 7.2 Hz, 4-CHaHb), 2.49 (d, 2H, J=6.5 Hz, 2-
CH2), 3.48 (dd, 1H, J=7.9 and 7.9 Hz, 6-CHaHb), 4.03 (dd,
1H, J=7.9 and 5.9 Hz, 6-CHaHb), 4.04–4.14 (m, 2H,
OCH2CH3), 4.14–4.21 (m, 1H, 3-CH), 4.21–4.31 (m, 1H, 5-
CH); 13C NMR: d=ꢁ4.62 (MeSi), ꢁ4.32 (MeSi), 14.48
(OCH2CH3), 18.20 (CMe3), 26.01 (CMe3), 26.09 (Me), 27.25
(Me), 41.27 (4-CH2), 42.46 (2-CH2), 60.63 (OCH2CH3),
67.23 (3-CH), 70.09 (6-CH2), 72.72 (5-CH), 108.91 (CMe2),
171.79 (COO); MS: m/z=331 [M+ꢁMe], 231 [M+ꢁ(t-
BuMe2Si)]; anal. calcd. for C17H34O5Si: C 58.92, H 9.89;
found: C 59.95, H 10.03.
The authors are grateful to Ratiopharm GmbH (Ulm) for fi-
nancial support of this work. We appreciate skilled technical
assistance by Mrs. K. Mevius.
References
[1] Part I of a series of manuscripts dedicated to the syn-
thesis of statins. For part II, see ref.[25]
[2] Contemporary Drug Synthesis (Eds.: J.-J. Li, D. S. John-
son, D. R. Sliskovic, B. D. Roth), Wiley-Interscience,
New York, 2004, pp. 113–123.
[3] E. S. Istvan, J. Deisenhofer, Science 2001, 292, 1160–
1164.
[4] T. Rosen, M. J. Tashner, C. H. Heathcock, J. Org.
Chem. 1984, 49, 3994–4003.
[5] 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.
[6] J. Beecher, I. Brackenridge, S. M. Roberts, J. Tang,
A. J. Willetts, J. Chem. Soc., Perkin Trans. 1 1995,
1641–1643; J. Tang, I. Brackenridge, S. M. Roberts, J.
Beecher, A. J. Willetts, Tetrahedron 1995 51, 13217–
13238.
[7] U. Maler, R. M. Devant, M. Brown, Chem. Ber. 1988,
121, 2035–2044.
[8] K. Mori, T. Takigawa, T. Matsuo, Tetrahedron 1979, 35,
933–940.
Ethyl (3R,5S)-6-tert-butyldimethylsilyloxy-3,5-(isopropyli-
denedioxy)hyhexanoate (11): Colourless viscous liquid;
1
[a]2D3: ꢁ4.7 (c 1, EtOH); H NMR: d=0.012 (s, 3H, MeSi),
0.016 (s, 3H, MeSi), 0.85 (s, 9H, CMe3), 1.14 (ddd, 1H, J=
12.7, 11.7 and 11.7 Hz, 4-CHaHb), 1.21 (t, 3H, J=7.1 Hz,
OCH2CH3), 1.32, (s, 3H, Me), 1.41 (s, 3H, Me), 1.63 (ddd,
1H, J=12.7, 2.5 and 2.5 Hz, 4-CHaHb), 2.35 (dd, 1H, J=
15.3 and 5.8 Hz, 2-CHaHb), 2.49 (dd, 1H, J=15.3 and 7.2
Hz, 2-CHaHb), 3.44 (dd, 1H, J=10.3 and 5.9 Hz, 6-CHaHb),
3.62 (dd, 1H, J=10.3 and 5.2 Hz, 6-CHaHb), 3.85–3.94 (m,
1H, 3-CH), 4.04–4.17 (m, 2H, OCH2CH3), 4.24–4.33 (m,
1H, 5-CH). 13C NMR: d=ꢁ5.02 (MeSi), ꢁ4.92 (MeSi), 14.48
(OCH2CH3), 18.59 (CMe3), 19.96 (Me), 26.15 (CMe3), 30.17
(Me), 33.61 (4-CH2), 41.90 (2-CH2), 60.66 (OCH2CH3),
66.11 (3-CH), 67.08 (6-CH2), 69.96 (5-CH), 98.95 (CMe2),
171.15 (COO); MS: m/z=331 [M+ꢁMe], 231 [M+ꢁ(t-
BuMe2Si)]; anal. calcd. for C17H34O5Si: C 58.92, H 9.89;
found: C 59.76, H 10.01.
Diastereoselective Hydrogenation of b-Keto Ester 2b
to Hydroxy Ester syn-6 in a Preparative Scale
[9] S. Saito, T. Hasegawa, M. Inaba, R. Nishida, T. Fujii, S.
Nomizu, T. Moriwake, Chem. Lett. 1984, 1389–1392; S.
Saito, T. Ishikawa, A. Kuroda, K. Koga, T. Moriwake,
Tetrahedron 1992, 48, 4067–4086.
Catalyst preparation: A stirred mixure of (R)-Tol-BINAP
(200 mg,
0.295 mmol),
[Ru
A
(73.6 mg,
[10] J. E. Lynch, R. P. Volante, R. V. Wattley, I. Shinkai, Tet-
rahedron Lett. 1987, 28, 1385–1388; G. Beck, H. Jen-
dralla, K. Kesseler, Synthesis 1995, 1014–1018, J.-L.
Scheffler, V. Bette, A. Mortreux, G. Nowogrocki, J.-F.
Carpentier, Tetrahedron Lett. 2002, 43, 2679–2682; K.
Prasad, K.-M. Chen, O. Repic, G. E. Hardtmenn, Tetra-
hedron: Asymmetry 1990, 1, 307–310; G. Wess, K.
Kesseler, E. Baader, W. Bartmann, G. Beck, A. Berg-
mann, H. Jendralla, K. Bock, G. Holzstein, H. Kleine,
M. Schnierer, Tetrahedron Lett. 1990, 31, 2545–2548;
L. Shao, H. Kawano, M. Saburi, Y. Ushida, Tetrahedron
1993, 49, 1997–2010; M. M. Claffey, C. J. Hayes, C. H.
Heathcock, J. Org. Chem. 1999, 64, 8267–8274; P.-F.
Deschnaux, T. Kallimopoulos, H. Stoeckli-Evans, A.
Jacot-Guillarmod, Helv. Chim. Acta 1989, 72, 731–737;
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.
0.147 mmol), AcONa (24.2 mg, 0.295 mmol) and DMF
(2 mL) was kept at 1008C (in a preheated oil bath) for
15 min under Ar. The volatiles were evaporated and the res-
idue was dried in high vacuum at 508C for 1 h. The solid
material was used as a catalyst for the hydrogenation with-
out further purification and characterization.
Hydrogenation: The catalyst (9.4 mg, ca. 0.01 mmol) and
b-keto ester 2b (2.4 g, 10.4 mmol) were placed in a 50-mL
autoclave. The autoclave was sealed and air was removed by
three vacuum-Ar cycles. Absolute oxygen-free EtOH
(20 mL) was added and the mixture was stirred under
100 bar initial H2 pressure at 508C. The consumption of H2
was followed by the decrease of the pressure. After ca. 4 h
the consumption of H2 ceased. The autoclave was opened,
the mixture was evaporated and the residue dried in high
vacuum. The conversion and the yield of syn-6 were quanti-
tative. NMR spectra were identical to the data reported
above. The product was contaminated with ca. 2% of impur-
ities of unknown structure.
[11] E. C. Taylor, D. J. Dumas, J. Org. Chem. 1980, 45,
2485–2489; J. D. Winkler, P. M. Hershberger, J. Am.
Chem. Soc. 1989, 111, 4654; S. Ohta, A. Shimabayashi,
S. Hatano, M. Okamoto, Synthesis 1983, 715–716; J. D.
Winkler, E. M. Doherty, Tetrahedron Lett. 1998, 39,
2253–2256.
[12] A. P. Kozikowski, C.-S. Li, J. Org. Chem. 1985, 50, 778–
785.
Adv. Synth. Catal. 2006, 348, 2633 – 2644
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2643

Vitali I. Tararov et al.
FULL PAPERS
[13] K. Hori, N. Hikage, A. Inagaki, S. Mori, K. Nomura, E.
Yoshi, J. Org. Chem. 1992, 57, 2888–2902.
Ohkuma, M. Widhalm, M. Kitamura, H. Takaya, S.
Akutagawa, N. Sayo, T. Saito, T. Taketomi, H. Kumo-
bayashi, J. Am. Chem. Soc. 1989, 111, 9134–9135.
[21] M. Kitamura, M. Tokunaga, M. Ohkuma, R. Noyori,
Tetrahedron Lett. 1991, 32, 4163.
[22] J. G. Buchanan, A. R. Edger, D. I. Rawson, P. Shahidi,
R. H. Wightman, Carbohydr. Res. 1982, 100, 75–86;
T. B. Grindley, C. J. P. Cote, C. Wickramage, Carbohydr.
Res. 1985, 140, 215–238.
[23] T. Ikariya, Y. Ishii, H. Kawano, T. Akai, M. Saburi, S.
Yoshikawa, S. Akutagawa, J. Chem. Soc., Chem.
Commun. 1985, 922–924.
[24] H. Takaya, T. Ohta, S. Inoue, M. Tokunaga, M. Kita-
mura, R. Noyori, Org. Synth. 1993, 72, 74–85.
[25] V. I. Tararov, N. Andrushko, V. Andrushko, G. Kçnig,
A Spannenberg, A. Bçrner, Eur. J. Org. Chem. 2006, in
press; DOI: 10.1002/ejoc.200600849.
[14] T. Yakura, A. Ueki, T. Kitamura, K. Tanaka, M.
Nameki, M. Ikeda, Tetrahedron 1999, 55, 7461–7470.
[15] Org. Synth. Col. Vol. 3, 408–410; J. W. Frankenfeld,
J. J. Werner, J. Org. Chem. 1969, 34, 3689–3691.
[16] T. Harada, T. Mukayama, Chem. Lett. 1982, 161–164.
[17] F. Tavares, J. P. Lawson, A. I. Meyers, J. Am. Chem.
Soc. 1996, 118, 3303–3304.
[18] D. W. Brooks, L. D.-L. Lu, S. Masamune, Angew.
Chem. 1979, 91, 76–77.
[19] MEMAH is commercially available. Otherwise it can
be prepared from the less expensive K salt: Org. Synth.
1957, 37, 34–36.
[20] a) M. Kitamura, T. Ohkuma, S. Inoe, N. Sayo, H. Ku-
mobayashi, S. Akutagawa, T. Ohta, H. Takaya, R.
Noyori, J. Am. Chem. Soc. 1988, 110, 629–631; b) R.
Noyori, T. Ohkuma, M. Kitamura, H. Takaya, N. Sayo,
H. Kumobayashi, S. Akutagawa, J. Am. Chem. Soc.
1987, 109, 5856–5858; c) R. Noyori, T. Ikeda, T.
[26] V. I. Tararov, R. Kadyrov, T. H. Riermeier, A. Bçrner,
Synthesis 2002, 375–380.
2644
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2633 – 2644