Development of a practical synthesis of novel, pyrrole-based HMG-CoA reductase inhibitors

Article abstract of DOI:10.1016/j.tet.2007.06.005

This paper describes the development of an efficient and scalable second generation synthesis of novel, pyrrole-based HMG-CoA reductase inhibitors. Compound 1 was identified as part of a discovery program aimed at finding improved treatments for hypercholesterolemia. Herein, we describe an efficient synthesis of its highly functionalized pyrrole core followed by attachment of the 3,5-dihydroxyhexanoic acid side chain via ylide olefination chemistry.

Full text of DOI:10.1016/j.tet.2007.06.005

Tetrahedron 63 (2007) 8124–8134
Development of a practical synthesis of novel, pyrrole-based
HMG-CoA reductase inhibitors
*
Jeffrey A. Pfefferkorn, Daniel M. Bowles, William Kissel, David C. Boyles, Chulho Choi,
Scott D. Larsen, Yuntao Song, Kuai-Lin Sun, Steven R. Miller and Bharat K. Trivedi
Pfizer Global Research and Development, 2800 Plymouth Road, Ann Arbor, MI 48105, USA
Received 3 April 2007; revised 30 May 2007; accepted 2 June 2007
Available online 9 June 2007
Abstract—This paper describes the development of an efficient and scalable second generation synthesis of novel, pyrrole-based HMG-CoA
reductase inhibitors. Compound 1 was identified as part of a discovery program aimed at finding improved treatments for hypercholesterol-
emia. Herein, we describe an efficient synthesis of its highly functionalized pyrrole core followed by attachment of the 3,5-dihydroxyhexanoic
acid side chain via ylide olefination chemistry.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
O
OH OH
N
1
CO2H
N
H
2
5
Coronary heart disease (CHD) is a leading cause of death
worldwide.1 Epidemiological evidence indicates that there
is a strong link between hypercholesterolemia and the risk
of CHD.2 As a consequence, significant efforts have been
undertaken to lower the cholesterol levels in hypercholester-
olemic patients to mitigate this risk factor. Currently, the
most successful and widely utilized method for treating
hypercholesterolemia is the use of HMG-CoA reductase
inhibitors, which block the rate limiting step of cholesterol
synthesis.3 As a class, HMG-CoA reductase inhibitors
have proven to be well tolerated and remarkably effective.
However, as revisions to National Cholesterol Education
Program (NCEP-ATP-III) cholesterol treatment guidelines
call for increasingly aggressive low density lipoprotein
(LDL) lowering in at-risk patients (for example, LDL
<70 mg/dL for highest risk patients)4 there is continued
need for new HMG-CoA reductase inhibitors with improved
efficacy relative to those currently available.
4
3
MeO
F
1
O
OH OH
CO2H
N
H
N
F
2: atorvastatin
Figure 1. Structure of HMG-CoA reductase inhibitor 1 and atorvastatin (2).
necessitated the development of a new synthetic strategy.
An early medicinal chemistry synthesis (Scheme 1) of this
template enabled access to key analogs for structure–activity
studies; however, the multi-gram preparation of compounds
of interest for further pre-clinical evaluation was hampered
by the limited scalability of this first generation route. As
a consequence, a second generation synthesis of 1 and
related analogs was developed and is described herein.
As part of a discovery program aimed at identifying novel
HMG-CoA reductase inhibitors, we investigated a series of
substituted pyrroles represented by inhibitor 1 (Fig. 1).5
Structurally, compound 1 and related analogs represent a
regioisomeric variation of the atorvastatin (LipitorÒ, 2)
template, but they offer a distinct pharmacological profile.5
From a synthetic viewpoint, transposition of the 3,5-
dihydroxyheptanoic acid side chain from the 1-position
(atorvastatin)6 to the 2-position (1) of the central pyrrole
2. First generation synthesis
Scheme 1 highlights the original medicinal chemistry route
to compound 17, a representative analog that was utilized
* Corresponding author. E-mail: jeffrey.a.pfefferkorn@pfizer.com
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.06.005

J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
8125
R
N
Br
NO2
a
b
F
EtO2C
F
F
NO2
d
c
3
5
CHO
nBu
N
F
F
e
F
7
8: R =H
9: R = iPr
F
F
4
6
f
O
O
OR
O
N
N
N
CO2Me
O
CHO
F
CHO
RO2C
PhHN
PhHN
h
i
OTBS
CO2Me
13
Ph3P
F
j
F
F
F
F
g
12
10: R = Et
11: R = H
14: R = TBS
15: R = H
k
O
OH OH
O
OH OH
N
CO2Me
N
CO2Na
PhHN
F
PhHN
F
l,m
16
17
F
F
Scheme 1. First generation synthetic route illustrated for HMG-CoA reductase inhibitor 17. Reagents and conditions: (a) AgNO2, Et2O, 0 ꢀC, 12 h, 36%; (b)
nBuNH2, C6H6, 80 ꢀC, 2 h, 100%; (c) AcOH, 25 ꢀC, 12 h, 82%; (d) CNCH2CO2Et, DBU, THF, 25 ꢀC, 12 h, 38%; (e) KOH (powdered), iPrI, DMSO, 25 ꢀC,
1.5 h, 62%; (f) POCl3, DMF, dichloroethane, 80 ꢀC, 3 h, 45%; (g) NaOH, MeOH, 60 ꢀC, 2 h, 100%; (h) (i) SOCl2, 80 ꢀC; (ii) PhNH2, Et3N, THF, 25 ꢀC, 12 h,
57%; (i) 13, toluene, 110 ꢀC, 48 h, 66%; (j) HF (48% in H2O), MeCN, 0 ꢀC, 2 h, 99%; (k) Et2B(OMe), NaBH4, AcOH (cat), THF, ꢁ78 ꢀC, 2 h, 58%; (l) 10%
Pd/C, H2, EtOH/THF (1:1), 25 ꢀC, 3 h, 100%; (m) NaOH, EtOH, 1 h, 100%.
for much of our exploratory synthetic studies.5 This first
generation route relied upon a Barton–Zard synthesis for
construction of the central pyrrole heterocycle.7 Initially,
4-fluorobenzyl bromide (3) was treated with silver nitrite8
to afford 1-fluoro-4-nitromethyl-benzene (5). In parallel,
benzaldehyde (4) was converted to imine 6 by treatment
with n-butylamine. Subsequent reaction of 5 and 6 in glacial
acetic acid provided 1-fluoro-4-(1-nitro-2-phenyl-vinyl)-
benzene (7). To complete the pyrrole synthesis, intermediate
7 was reacted with ethyl isocyanatoacetate in the presence of
DBU to generate pyrrole 8, regioselectively. N-Alkylation of
8 was then accomplished using iso-propyl iodide and
powdered KOH in DMSO to provide pyrrole 9, which was
subjected to formylation under Vilsmeier–Haack conditions
to generate pyrrole carboxaldehyde 10. The ethyl ester of
10 was saponified with aqueous NaOH, converted to the
corresponding acid chloride via treatment with SOCl2, and
reacted with aniline to afford anilide 12. Installation of the
3,5-dihydroxyhexanoic acid side chain commenced with
a Wittig olefination reaction between pyrrole carboxalde-
hyde 12 and stabilized phosphonium ylide 13 (prepared stereo-
selectively in eight steps from diethyl 3-hydroxyglutarate
according to the method of Konoike and Araki9) to provide
olefin 14 in 66% yield. Any unreacted 12 could be recycled
via a tedious chromatography. The TBS-protected side chain
hydroxyl of 14 was then liberated by treatment with aqueous
HF to provide b-hydroxy ketone 15, which was subjected to
a stereoselective syn-reduction with Et2B(OMe) and NaBH4
to produce syn-diol 16.10 The synthesis was completed by
hydrogenation of the side chain olefin over 10% Pd/C fol-
lowed by saponification of the terminal ester to provide
compound 17 as its sodium salt.
While this original route proved valuable for structure–activ-
ity studies, it posed a number of challenges for the multi-
gram scale-up synthesis of selected compounds of interest.
These limitations included: (a) a length of 21 steps (includ-
ing the required preparation of side chain reagent 13); (b) an
overall yield of <1%; (c) the use of potentially hazardous
high energy intermediates 5 and 7; (d) the use of several ex-
pensive reagents including silver nitrate and ethyl isocya-
noacetate; and (e) the need for at least 10 chromatographic
purifications.
3. Second generation synthesis
In light of the limited scalability of the original route, we
sought to develop an alternative synthesis as outlined retro-
synthetically in Scheme 2. Key elements of this second gen-
eration route included: (a) reversing the reacting partners in
the Wittig olefination (i.e., intermediate 19 as the ylide)
thereby enabling the use of aldehyde 20 (available in one
step from commercially available material) as a side chain
precursor; (b) utilization of an efficient pyrrole synthesis
previously reported by Gupton and co-workers where

8126
J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
a vinylogous amide (i.e., 21) or vinylogous iminium salt
equivalent is reacted with an a-amino acid ester (i.e., 22), re-
sulting in regioselective pyrrole formation.11 Of most impor-
tance, this latter approach was anticipated to be both scalable
and amenable to the construction of pyrroles (such as 19),
which contain a high degree of non-symmetric substitution
(i.e., R2sF).
contrast, acid mediated conditions (Table 1, entries 3 and
4) resulted in good conversion to the desired pyrrole 27 in
a regioselective manner. Comparison of entries 3 and 4 re-
vealed that a 2-fold excess of a-amino acid ester 22 (entry
3) afforded slightly improved yields relative to a stoichiomet-
ric amount (entry 4).
F
O
F
a
O
O
OH OH
N
CO2H
Cl
R1HN
R2
24
23
2
F
O
b
F
NMe2
25
O
O
O
c
OEt
N
CO2tBu
Br
OEt
R1HN
N
O
26
H
O
22
18
Scheme 3. Synthesis of pyrrole precursors 25 and 22. Reagents and condi-
tions: (a) AlCl3, C6H6, 50 ꢀC, 8 h, 97%; (b) N,N-dimethylformamide
R2
F
i
dimethyl acetal, toluene, 100 ꢀC, 15 h, 100%; (c) PrNH2, MTBE, 5 ꢀC,
16 h, 95%.
Olefination Reaction
O
N
X
Table 1. Optimization of pyrrole synthesis using amino acid ester 22
R1HN
R2
O
O
O
+
CO2tBu
F
N
OHC
O
EtO
OEt
+
N
H
20
F
19
O
NMe2
25
22
27
F
Entry Ratio of 22/25 Additive Solvent Temp (ꢀC) Time (h) Yield (%)
F
O
1
2
2:1
2:1
Cs2CO3 NMP 150
NaOEt EtOH 100
16
12
33
36
+
N
CO2Et
H
R2
NMe2
3
4
2:1
1:1
—
—
AcOH 120
AcOH 120
2
2
85
72
21
22
Scheme 2. Second generation retrosynthetic disconnections.
Given that our final target, 1, contained an amide moiety at
the site of the current ester in pyrrole 27, we next sought
to investigate whether or not the pyrrole forming reaction
shown in Table 1 could tolerate an a-amino amide in place
of a-amino acid ester 22. Such an adaptation would elimi-
nate the need for conversion of the ester to the amide at a later
stage in the synthesis. To evaluate this possibility, model a-
amino amide 30 was prepared as outlined in Scheme 4 where
chloroacetyl chloride (28) was reacted with benzyl amine in
the presence of triethylamine, and the resulting product 29
was then treated with iso-propyl amine to provide a-amino
amide 30. Gupton and co-workers had reported that, in
many cases, similar pyrrole forming reactions were more
facile if the vinylogous amide component was first converted
into a synthetically equivalent (yet more reactive) b-chloro-
enal. Fearing that the desired transformation would also be
somewhat sluggish in our system, vinylogous amide 25
was also converted to b-chloroenal 31 as outlined in Scheme
4.11 With a-amino amide 30, vinylogous amide 25, and b-
chloroenal 31 in hand, we investigated whether or not the
3.1. Pyrrole construction
Preliminary efforts to develop this second generation route
focused on constructing the central pyrrole using the method
of Gupton and co-workers as described. The requisite reac-
tion components were synthesized as outlined in Scheme 3.
Friedel–Crafts arylation between 4-fluoro-phenyl acetyl
chloride (23) and benzene in the presence of AlCl3 provided
2-(4-fluoro-phenyl)-1-phenylethanone (24), which was then
treated with N,N-dimethylformamide dimethyl acetal at ele-
vated temperature to provide vinylogous amide 25.11 Sepa-
rately, a-amino acid ester 22 was prepared by reaction of
ethyl bromoacetate (26) with iso-propyl amine. Next, a vari-
ety of conditions were examined for the reaction of 22 and
25 to form pyrrole 27 as outlined in Table 1. According to
Gupton report, the anticipated pyrrole synthesis could occur
under both base and acid mediated conditions.11 In our case,
representative base mediated conditions (Table 1, entries 1
and 2) afforded low yields and long reaction times. By

J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
8127
pyrrole formation would work with the amide motif already
in place as highlighted in Table 2. Attempted reaction of
vinylogous amide 25 with 30 under acidic conditions (Table
2, entry 1) failed to afford any product. Limited formation of
the desired product 32 was noted when 30 and the more
reactive b-chloroenal 31 were reacted under neutral condi-
tions in either DMF (8% yield, entry 2) or NMP (24% yield,
entry 3).
10, which was saponified with aqueous NaOH to afford carb-
oxylic acid 11. This intermediate was converted to the
corresponding acid chloride and was reacted with aniline
to afford anilide 12. Finally, the carboxaldehyde of 12 was
reduced to the corresponding alcohol 33 using NaBH4.
O
N
N
EtO2C
CHO
R
a
H
N
a
Cl
Cl
Cl
O
O
F
F
F
F
28
29
10: R = OEt
11: R = OH
12: R = NHPh
9
b
c
b
H
O
N
d
N
H
N
PhHN
F
OH
O
F
30
F
O
Cl
F
c
33
CHO
NMe2
Scheme 5. Synthesis of alcohol 33. (a) POCl3, DMF, dichloroethane, 80 ꢀC,
3 h, 45%; (b) NaOH, MeOH, 60 ꢀC, 2 h, 100%; (c) (i) SOCl2, 80 ꢀC; (ii)
PhNH2, Et3N, THF, 25 ꢀC, 12 h, 57%; (d) NaBH4, MeOH, 0 ꢀC, 1 h, 47%.
31
25
Scheme 4. Synthesis of pyrrole precursors 30 and 31. Reagents and condi-
i
tions: (a)BnNH2, Et3N, CH2Cl2, 0 ꢀC, 12 h, 74%; (b) PrNH2, THF/Et2O,
25 ꢀC, 48 h, 32%; (c) (i) POCl3, CH2Cl2, 45 ꢀC; (ii) H2O/THF, 25 ꢀC,
48 h, 99% (two steps).
To prepare for installation of the side chain via Wittig
olefination, the alcohol of model pyrrole 33 needed to be
transformed into an appropriate ylide precursor as outlined
in Table 3. Preparation of both a phosphonium salt and
a phosphonate ester was investigated. Initially, we sought
to convert alcohol 33 into bromide 34 that is to be used as
a common precursor for the preparation of both phosphonium
salt 35 and phosphonate ester 36. However, after several
attempts (Table 2, entries 1–3) at this transformation failed
(presumably due to the high reactivity of bromide 34), we in-
stead sought direct conversion of alcohol 33 into 35 and 36
as illustrated by entries 4 and 5, respectively. Specifically,
alcohol 33 was directly and quantitatively converted to
phosphonium bromide 35 via treatment with Ph3P$HBr at
50 ꢀC.12 Alternatively, alcohol 33 was converted to phos-
phonate ester 36 by treatment with P(OEt)3, NaBr, and
BF3$OEt2 at 25 ꢀC.13 This latter transformation presumably
preceded though a transient alkyl bromide formed by the
action of NaBr on 33.
Attempts to improve the conversion by addition of a base
(entry 4) resulted in no product formation. Additional mod-
ification (not shown) to improve the conversion of this reac-
tion was also unsuccessful and as such this approach was
abandoned in favor of the previous ester strategy illustrated
in Table 1.
Table 2. Efforts toward synthesis of amide pyrrole 32 using amino amide 30
F
O
25
O
N
NMe2
BnHN
NHBn
N
H
or
+
O
F
Cl
30
31
32
F
CHO
With the ylide precursors 35 and 36 in hand, attention was
focused on optimization of the downstream olefination reac-
tion. As shown in Table 4, the requisite aldehyde coupling
partner 20 was readily prepared by Swern oxidation of the
commercially available alcohol 37.14 Avariety of conditions
were evaluated to prepare olefin 38 starting from either phos-
phonium salt 35 or phosphonate ester 36. Initial experimen-
tation with 35 (Table 4, entries 1–3) revealed that use of
either nBuLi or NaHMDS at low temperature afforded the
desired olefin product (1:2 cis/trans mixture) in relatively
low yields 31–46%. The use of KOtBu did not afford the de-
sired product. Closer examination of the reaction products in
entries 1 and 2 revealed that in addition to the desired prod-
uct 38 the other major product from these reaction was
a methyl pyrrole of general structure 42 (Scheme 6). This
Entry Substrate Equivalents Additive Solvent Temp Time Yield
of 30
(ꢀC) (h)
(%)
0
1
25
2.0
—
AcOH 125
2
2
3
4
31
31
31
1.0
2.0
2.0
—
—
DMF
NMP
150
100
100
12
1
2
8
24
0
Cs2CO3 NMP
3.2. Side chain installation
The next stage of the synthesis required functionalization of
the pyrrole core in anticipation of installation of the 3,5-
dihydroxyhexanoic acid side chain (Scheme 5). Vilsmeier–
Haack formylation of 9 provided pyrrole carboxaldehyde

8128
J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
Table 3. Preparation of phosphonium salt 35 and phosphonate ester 36 as ylide precursors
O
O
O
N
N
N
R1HN
R2
OH
F
R1HN
R2
Br
F
R1HN
R2
R3
F
35:R3 = [PPh3]+Br-
36:R3 = P(O)(OEt)2
33
34
Entry
R1
R2
Product
Reagent
Solvent
Temp (ꢀC)
Time (h)
Yield (%)
1
2
3
4
5
Ph
Ph
Ph
Ph
Bn
F
F
F
F
H
34
34
34
35
36
CBr4, Ph3P
PBr3, pyridine
Ph3PBr2
Ph3PHBr
P(OEt)3, NaBr, BF3$OEt2
CH2Cl2
CH2Cl2
MeCN
CH2Cl2
CH2Cl2
25
0
50
50
25
12
12
12
2
0
0
0
100
99
12
Table 4. Summary of Wittig and Horner–Wadsworth–Emmons strategies for side chain installation
O
H
O
O
O
N
N
CO2tBu
R1HN
R3
O
O
R1HN
R2
+
Base
THF
CO2tBu
H
X
37: X = CH2OH
20: X = CHO
38
[Ox]
R2
F
F
35: R3 = [PPh3]+Br-
36: R3 = P(O)(OEt)2
Entry
R1
R2
R3
Base
Equivalents of base
Temp (ꢀC)
Time (h)
Yield (%)
1
2
3
Ph
Ph
Ph
F
F
F
[PPh3]+Brꢁ
[PPh3]+Brꢁ
[PPh3]+Brꢁ
nBuLi
NaHMDS
1.2
1.2
1.2
ꢁ78/25
ꢁ78/25
0
2
2
2
46
31
0
KOtBu
4
5
4-(OMe)Bn
4-(OMe)Bna
H
H
[PPh3]+Brꢁ
[PPh3]+Brꢁ
NaHMDS
NaHMDS
1.4
1.4
ꢁ78/25
ꢁ78/25
2
2
25
75
6
7
8
9
Bn
Bn
Bn
Bn
H
H
H
H
P(O)(OEt)2
P(O)(OEt)2
P(O)(OEt)2
P(O)(OEt)2
NaHMDS
LDA
LDA
1.2
1.2
2.4
2.5
0
0
0
25
2
2
2
8
0
0
0
0
NaH
a
Azeotropic drying of phosphonium salt 35 prior to use in order to reduce residual water content.
Major By-Product
H
O
N
O
O
O
O
N
+
N
CO2tBu
R1HN
R2
PPh3Br-
R1HN
R1HN
R2
Base, 20
H
+
R2
F
37
42
F
F
35
-OH
H2O
O
O
O
OH
PPh3
O-
PPh3
-
N
N
N
CH2
-[Ph3PO]
R1HN
R2
R1HN
R2
R1HN
R2
F
F
F
39
40
41
Scheme 6. Proposed mechanism for formation of by-product 42.

J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
8129
type of by-product was observed regardless of the base em-
ployed in the olefination reaction. The formation of by-prod-
uct 42 was presumed to be a result of hydrolytic cleavage of
the carbon–phosphorous bond of phosphonium salt 35. Such
a hydrolytic cleavage is precedented as outlined mechanisti-
cally in Scheme 6.15 Hydroxide (generated from residual
water and base) undergoes nucleophilic addition to phospho-
nium salt 35 generating pentavalent phosphorous intermedi-
ate 39, which undergoes deprotonation and loss of Ph3PO to
generate carbon-centered anion 41, which is protonated to
form the observed by-product 42. The residual water en-
abling this side reaction was likely to be present in phospho-
nium salt 35 from the preceding step (i.e., phosphonium salt
formation). After some experimentation, it was found that
conversion to the desired olefin could be markedly improved
by first removing all residual water from phosphonium salt
35 via azeotropic drying with IPA.16 This improvement is
illustrated in Table 4 by comparison of entry 4 (25% yield)
versus entry 5 (75% yield). This modification was adopted
for all subsequent Wittig olefination reactions.
We also investigated the construction of the requisite olefin
linkage via a Horner–Wadsworth–Emmons reaction using
phosphonate ester 36 (Table 4). This strategy offers improved
atom economy and purification advantages relative to the
Wittig approach. Disappointingly, reaction of phosphonate
36 with several bases at various stoichiometries (Table 4,
entries 6–9) followed by the addition of aldehyde 20 did
not result in the formation of the desired olefination product
38. In most cases unreacted phosphonate ester 36 was re-
covered. The relative success of the former Wittig approach
and contrasting failure of the Horner–Wadsworth–Emmons
F
F
O
O
F
a
b
O
OEt
+
N
H
Cl
O
NMe2
23
24
25
22
c
O
O
N
N
N
CHO
F
EtO2C
N
H
R
R
g
d
MeO
F
F
43: R = OEt
27
46: R = OH
e
f
h
47: R = [PPh3]+Br-
44: R = OH
45: R = NH(4-OMe)Bn
i
O
H
O
O
N
CO2tBu
N
H
H
MeO
48
F
j,k
O
OH OH
N
CO2tBu
N
H
MeO
49
F
l
O
OH OH
CO2-Na+
N
N
H
MeO
1
F
Scheme 7. Complete second generation synthesis of HMG-CoA reductase inhibitor 1. Reagents and conditions: (a) AlCl3, C6H6, 50 ꢀC, 8 h, 97%; (b) N,N-di-
methylformamide dimethyl acetal, toluene, 100 ꢀC, 15 h, 100%; (c) AcOH, 125 ꢀC, 2.5 h, 81%; (d) POCl3, DMF, dichloroethane, 80 ꢀC, 18 h, 98%; (e) NaOH,
MeOH, 25 ꢀC, 24 h, 83%; (f) (i) SOCl2, 75 ꢀC, 2 h; (ii) (4-OMe)BnNH2, Et3N, CH2Cl2, 0 ꢀC, 2 h, 55%; (g) LiAl(OtBu)3H, THF, 0 ꢀC, 0.5 h, 41%; (h)
Ph3P$HBr, CH2Cl2, 50 ꢀC, 100%; (i) NaHMDS, 20 (see Table 4), THF, ꢁ78 ꢀC, 75%; (j) 10% Pd/C, H2, MeOH, 25 ꢀC, 3 h; (k) 1 N HCl, MeOH, 25 ꢀC,
3 h, 77% over two steps; (l) NaOH, MeOH, 25 ꢀC, 48 h, 93%.

8130
J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
strategy were attributed to pKa differences between the respec-
tive ylide precursors 35 and 36. In the case of phosphonium
salt 35, the methylene protons (adjacent to the phosphorous)
are predicted to be more acidic (estimated pKa¼17) than the
N–H of the amide (estimated pKa¼19). By contrast, the meth-
ylene protons of phosphonate 36 (estimated pKa¼27) are
expected to be less acidic than the N–H the amide (pKa¼19)
suggesting that, in this case, initial deprotonation occurred
at the amide functionality. In an effort to overcome this prob-
lem, excess base (Table 4, entries 8 and 9) was employed
to attempt a double deprotonation reaction of phosphonate
ester 36; however, despite this modification only recovered
starting material was isolated, and as such the synthesis
was executed using the previous Wittig olefination strategy.
atmosphere unless otherwise noted. 1H and 13C NMR spec-
tra were recorded on a Varian 400 MHz Nuclear Magnetic
Resonance Spectrometer. NMR spectra were recorded in
CDCl3, CD3OD, or DMSO-d6 and chemical shifts are
reported relative to the residual solvent peak. The following
abbreviations were used to assign spectra: s¼singlet, d¼
doublet, t¼triplet, q¼quartet, s¼septet, m¼multiplet, br
s¼broad singlet. Mass spectral analysis was conducted on
a Waters Micromass ZQ instrument. Melting points were
determined on an Electrothermal Melting Point Apparatus
and are uncorrected.
Complete experimental procedures and spectral data for the
original synthesis of this template (Scheme 1, 5–17) will be
published separately along with structure–activity studies
and complete biological data.
3.3. Complete second generation synthesis
With reliable methods for construction of the pyrrole core
and attachment of the 3,5-dihydroxyhexanoic acid side
chain in place, the second generation synthesis of this new
class of HMG-CoA reductase inhibitors could be completed.
Scheme 7 illustrates the optimized route in a cohesive se-
quence for the synthesis of inhibitor 1. Pyrrole 27 was con-
structed from vinylogous amide 25 and amino acid ester 22
as previously described. Elaboration of 27 via Vilsmeier–
Haack reaction followed by saponification and amidation
provided amide 45 in good yield. Subsequent reduction of
45 to the corresponding alcohol 46 and treatment with
Ph3P$HBr afforded phosphonium salt 47. Azeotropic drying
of 47 followed by Wittig olefination provided compound 48
as an inconsequential mixture of cis/trans isomers. Finally,
hydrogenation of the olefin over Pd/C, cleavage of the ace-
tonide protecting group (HCl/MeOH), and saponification
afforded the final product 1 as its sodium salt.
5.2. 2-(4-Fluoro-phenyl)-1-phenyl-ethanone (24)
To a solution of benzene (182 mL) at 0 ꢀC was added AlCl3
(46.4 g, 348 mmol). A portion of 4-fluoro-phenyl acetyl
chloride (50.0 g, 290 mmol) was then added drop-wise
over 0.5 h maintaining the reaction temperature below
5 ꢀC. Once the addition was complete, the reaction was
allowed to warm to 25 ꢀC and then heated to 50 ꢀC for 8 h.
Subsequently, the reaction mixture was cooled to 25 ꢀC
and poured onto ice (400 g). To the resulting suspension
was added 10% HCl (50 mL) with stirring. The organic layer
was separated and washed with 10% HCl, saturated
NaHCO3 solution, and brine. The organic layer was then
dried (Na2SO4) and concentrated to afford a solid that was
triturated twice with hexane (200 mL) and then dried at
35 ꢀC under vacuum to afford 24 (59.90 g, 97%): 1H NMR
(CDCl3) d 7.97 (d, J¼6.0 Hz, 2H), 7.53 (t, J¼7.6 Hz, 1H),
7.43 (t, J¼6.0 Hz, 2H), 7.22–7.17 (m, 2H), 7.00–6.97 (t, J¼
8.8 Hz, 2H), 4.23 (s, 2H); MS (APCI+): m/z 215.0 (M+H).
Characterization data were consistent with previous litera-
ture reports of 24.17 Additional quantities of 24 were subse-
quently purchased from Precursor Chemicals, Inc.
4. Conclusion
In conclusion, the second generation route offered several
advantages over the original synthesis (Scheme 1) including
reducing the overall synthetic length from 21 to 13 steps and
increasing the overall yield from <1 to 8%. The number of
chromatographic separations was also reduced significantly
improving material throughput. Additionally, the second
generation route eliminated the use of two potentially haz-
ardous high energy intermediates and a variety of expensive
reagents. Overall, this new route enabled the preparation of
multi-gram quantities of 1 to support pre-clinical evaluation
of this novel HMG-CoA reductase inhibitor.
5.3. 3-Dimethylamino-1-(4-fluoro-phenyl)-2-phenyl-
propenone (25)
To a solution of 24 (56.90 g, 266 mmol) in toluene (400 mL)
was added N,N-dimethylformamide dimethyl acetal
(141 mL, 1.06 mol) and the reaction was heated to reflux
for 16 h. After cooling to 25 ꢀC, the solvent was removed un-
der reduced pressure to afford an orange solid that was re-
crystallized from toluene (175 mL). The solid was isolated
by filtration and washed with hexane (60 mL) to afford 25
1
(57.1 g, 80%) as a brown solid: H NMR (DMSO-d6) d
5. Experimental
7.33–7.28 (m, 5H), 7.15 (s, 1H), 7.14–7.01 (m, 4H), 3.31 (s,
6H); 13C NMR (CDCl3) d 195.1, 154.3, 141.8, 133.7, 133.6,
129.5, 128.8, 127.9, 114.9, 114.7, 111.1, 52.3, 42.7; MS
(APCI+): m/z 270.1 (M+H). Mp 110–111 ꢀC.
5.1. General procedures
All reagents and solvents were used as received from com-
mercial sources unless otherwise noted. Specifically, alcohol
37 was purchased from Kanaka Corporation and converted
to aldehyde 20 as previously described.13 Raw material 2-
(4-fluoro-phenyl)-1-phenyl-ethanone (24) was initially pre-
pared as described below and subsequently purchased in
multi-kilogram quantities from Precursor Chemicals, Inc.
All experiments were conducted under an inert nitrogen
5.4. Isopropylamino-acetic acid ethyl ester (22)
A 5-L, four-necked, round-bottomed flask equipped with
a mechanical stirrer, temperature probe, addition funnel,
and nitrogen bubbler was charged with iso-propyl amine
(2.22 kg, 37.2 mol) and methyl tert-butyl ether (2.5 L).
The resulting solution was cooled to 5 ꢀC and treated with

J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
8131
ethyl bromoacetate (2.70 kg, 16.2 mol) by addition funnel at
a sufficient rate to maintain the pot temperature less than
30 ꢀC (w1 h). The reaction mixture was stirred at 22 ꢀC for
16 h, and iso-propyl amine hydrobromide was removed by
filtration though Celite. The filtrates were then concentrated
under reduced pressure to produce 22 (2.27 kg, 95%) as
solvent was removed under reduced pressure and THF/water
(1:1, 100 mL) was added in one portion and the resulting re-
action mixture was stirred at 25 ꢀC for 48 h. Subsequently,
THF was removed under reduced pressure and CH2Cl2
(500 mL) was added. The organic layer was separated, dried
(Na2SO4), and concentrated to afford 31 (54.5 g, 99%) as
a light yellow solid that was a ca. 2:1 mixture of E/Z isomers.
This mixture could be utilized without additional purifica-
tion; however, an analytical sample of the E-isomer (identi-
fied by comparison to Ref. 11b) was isolated by silica gel
chromatography (10–20% EtOAc/hexane) and character-
ized: 1H NMR (CDCl3) d 9.62 (s), 7.55–7.50 (m, 5H),
7.24–7.17 (m, 2H), 6.93–6.85 (m, 2H); MS (APCI+): m/z
261.0 (M+H).
1
a yellow oil: H NMR (CDCl3) d 4.14 (q, J¼7.2 Hz, 2H),
3.56 (s, 2H), 2.74 (sept, J¼6.0 Hz, 1H), 1.59 (s, 1H), 1.23
(t, J¼7.2 Hz, 3H), 1.01 (d, J¼6.4 Hz, 6H). Characterization
data were consistent with previous literature reports of 22.18
5.5. 4-(4-Fluoro-phenyl)-1-isopropyl-3-phenyl-1H-
pyrrole-2-carboxylic acid ethyl ester (27)
To a 5-L round-bottomed flask equipped with mechanical
stirring were charged 25 (868.5 g, 3.23 mol) and glacial ace-
tic acid (1.3 L). The resulting solution was treated with 22
(953.3 g, 6.57 mol, 2.04 equiv) and the mixture was heated
to 120 ꢀC for 6 h, until full consumption of 25 was confirmed
by HPLC analysis. The reaction mixture was cooled to 85 ꢀC
and was diluted with water (325 mL) in one portion. The
mixture was allowed to cool to 20 ꢀC with stirring, and
was held at this temperature for 5 h. The resulting solid
was collected by filtration, washed with water (3ꢂ1 L),
and dried at 60 ꢀC under vacuum for 60 h to afford 27
5.8. Diethyl-(5-(benzylcarbamoyl)-3-(4-fluoro-phenyl)-
1-isopropyl-4-phenyl-1H-pyrrol-2-yl)methylphospho-
nate (36, R1=Bn, R2=F, R3=P(O)(OEt)2)
To a solution of 33 (1.14 g, 2.56 mmol) in CH2Cl2 (100 mL)
at 25 ꢀC were added triethyl phosphate (0.43 g, 2.56 mmol),
NaBr (0.26 g, 2.56 mmol), and BF3$OEt2 (0.36 g,
2.56 mmol), and the reaction mixture was stirred at 25 ꢀC
for 16 h. Subsequently, a saturated NaHCO3 solution
(20 mL) was added and the organic layer was separated,
washed with brine, dried (Na2SO4), and concentrated to af-
ford 36 (1.43 g, 99%), which was used without additional
1
(868.8 g, 77%) as a pale tan solid: H NMR (CDCl3) d
7.24–7.10 (m, 6H), 7.01–6.97 (m, 2H), 6.86 (t, J¼8.7 Hz,
2H), 5.45 (sept, J¼6.6 Hz, 1H), 4.02 (q, J¼7.0 Hz, 2H),
1.50 (d, J¼6.6 Hz, 6H), 0.90 (t, J¼7.2 Hz, 3H); 13C NMR
(CDCl3) d 162.7, 162.1, 136.3, 131.1, 130.8, 129.8, 129.7,
127.6, 126.6, 123.7, 120.8, 120.4, 115.2, 59.9, 49.0, 24.1,
13.7; MS(APCI+): m/z 352.3 (M+H). Anal. Calcd for
C22H22F1N1O2: C, 75.19; H, 6.31; N, 3.99. Found: C,
75.17; H, 6.40; N, 3.89. Mp 104–105 ꢀC.
1
purification: H NMR (CDCl3) d 7.14–7.10 (m, 8H), 7.00–
6.95 (m, 2H), 6.87–6.79 (m, 4H), 5.50 (br s, 1H), 4.84
(sept, J¼7.2 Hz, 1H), 4.25 (d, J¼5.6 Hz, 2H), 4.09–3.94
(m, 4H), 3.14 (d, J¼20.8 Hz, 2H), 1.64 (d, J¼7.2 Hz, 6H),
1.33–1.20 (m, 6H); MS (APCI+): m/z 563.1 (M+H).
5.9. 3,4-Bis(4-fluoro-phenyl)-5-formyl-1-isopropyl-N-
phenyl-1H-pyrrole-2-carboxamide (12)
5.6. N-Benzyl-2-(isopropylamino)acetamide (30)
Neat 9 (3.00 g, 8.12 mmol) was treated with thionyl chloride
(30 mL) and the resulting mixture was heated to 70 ꢀC for 1 h
and then cooled to 25 ꢀC. The excess thionyl chloride was re-
moved under reduced pressure and CH2Cl2 (50 mL) was
added. The reaction mixture was cooled to 0 ꢀC and was sub-
sequently treated with aniline (0.91 g, 9.75 mmol) and tri-
ethylamine (1.23 g, 12.2 mmol). The reaction mixture was
stirred for 12 h at 25 ꢀC after which time a saturated NaHCO3
solution was added. The organic layer was separated, dried
(Na2SO4), and concentrated to afford a crude oil that was pu-
rified by silica gel chromatography (10% EtOAc/hexanes) to
provide 12 (3.61 g, 34%). 1H NMR (DMSO-d6) d 10.63 (s,
1H), 9.36 (s, 1H), 7.43–7.41 (m, 2H), 7.35–7.11 (m, 6H),
7.07–6.96 (m, 5H), 5.20 (sept, J¼6.2 Hz, 1H), 1.51 (d,
J¼6.2 Hz, 6H); MS (APCI+): m/z 445.1 (M+H).
To a solution of chloroacetyl chloride (2.84 g, 25.1 mmol) in
CH2Cl2 at 0 ꢀC was added benzyl amine (2.69 g, 25.1 mmol)
followed by triethylamine (7.62 g, 75.3 mmol). The reaction
mixture was stirred at 0 ꢀC for 0.5 h and then warmed to
25 ꢀC for 12 h. Subsequently, a saturated solution of
NaHCO3 was added to the reaction mixture and the organic
layer was separated, washed with brine, dried (Na2SO4),
and concentrated to afford 29 (3.42 g, 18.6 mmol), which
was taken up in THF/Et2O (1:2, 100 mL). To this solution
at 25 ꢀC was added iso-propyl amine (2.20 g, 37.2 mmol)
and the resulting reaction mixture was stirred at 25 ꢀC for
48 h. The precipitate that developed was removed by filtra-
tion and the filtrate concentrated to an oil that was purified
by silica gel chromatography (5–10% MeOH/CH2Cl2) to af-
ford 30 (1.21 g, 32%): 1H NMR (DMSO-d6) d 8.22 (br s, 1H),
7.27–7.17 (m, 5H), 4.24 (d, J¼6.0 Hz, 2H), 3.07 (s, 2H), 2.61
(sept, J¼6.4 Hz, 1H), 2.21 (br s, 1H), 0.90 (d, J¼6.4 Hz, 6H);
13C NMR (CDCl3) d 172.2, 138.6, 128.8, 127.8, 127.5, 50.3,
49.9, 43.1, 23.0; MS (APCI+): m/z 207.2 (M+H).
5.10. 3,4-Bis(4-fluoro-phenyl)-5-(hydroxymethyl)-1-
isopropyl-N-phenyl-1H-pyrrole-2-carboxamide (33)
To a solution of 12 (1.23 g, 2.77 mmol) in MeOH (40 mL) at
0 ꢀC was added NaBH4 (0.16 g, 4.15 mmol) in one portion
and the reaction mixture was stirred at 0 ꢀC for 0.5 h and
then at 25 ꢀC for an additional 0.5 h. Subsequently, the reac-
tion solvent was removed under reduced pressure and then
CH2Cl2 (100 mL) and saturated NaHCO3 (50 mL) were
added. The organic layer was separated, dried (Na2SO4),
and concentrated to afford a crude oil that was purified by
5.7. (E)-3-Chloro-2-(4-fluoro-phenyl)-3-phenylacryl-
aldehyde (31)
To a solution of 25 (57.1 g, 212 mmol) in CH2Cl2 (250 mL)
at 25 ꢀC was added POCl3 (65.1 g, 424 mmol). The reaction
mixture was heated to 45 ꢀC for 5 h. After cooling, the

8132
J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
silica gel chromatography (20% EtOAc/hexanes) to provide
33 (0.58 g, 47%): 1H NMR (DMSO-d6) d 10.0 (s, 1H), 7.41–
7.39 (m, 2H), 7.22–7.18 (m, 2H), 7.12–6.91 (m, 9H), 5.12 (t,
J¼4.4 Hz, 1H), 4.75 (sept, J¼7.2 Hz, 1H), 4.33 (d,
J¼4.4 Hz, 1H), 1.55 (d, J¼7.2 Hz, 6H); MS (APCI+): m/z
447.1 (M+H).
stirred at room temperature for 24 h after which time the
reaction mixture was concentrated to dryness under reduced
pressure. The residual solid was dissolved in water (11 L).
The aqueous layer was then extracted with two portions of
MTBE. The combined organic layers were extracted with
0.5 M aqueous potassium hydroxide. The three aqueous
layers were combined, washed with MTBE, acidified to
pH 1 with 6 M aqueous hydrochloric acid, and extracted
with ethyl acetate. The combined organic layers were then
washed with 6 L of brine, dried (Na2SO4), and concentrated
under vacuum to provide 44 (498 g, 83%) as a light yellow
5.11. [3-(4-Fluoro-phenyl)-1-isopropyl-5-(phenylcarb-
amoyl)-4-(4-fluoro-phenyl)-1H-pyrrol-2-ylmethyl]-tri-
phenyl-phosphonium bromide (35)
1
To a solution of 33 (0.21 g, 0.47 mmol) in CH2Cl2 (20 mL)
was added triphenylphosphine hydrobromide (0.16 g,
0.47 mmol). The reaction mixture was heated to 50 ꢀC for
2.5 h after which time all starting material was consumed
as determined by TLC. The reaction solvent was removed un-
der reduced pressure and the resulting yellow foam was dried
under high vacuum for 12 h to provide 35 (0.36 g, 100%) in
sufficient purity for use in the next step. 1H NMR (DMSO-d6)
d 9.98 (s, 1H), 7.83–7.81 (m, 2H), 7.65–7.47 (m, 6H), 7.35–
7.33 (m, 4H), 7.26–7.16 (m, 8H), 7.05–6.92 (m, 4H), 6.82–
6.77 (m, 2H), 6.52–6.49 (t, J¼6.0 Hz, 2H), 5.13 (d,
J¼12.4 Hz, 2H), 4.31 (sept, J¼6.0 Hz, 1H), 1.01 (d,
J¼6.0 Hz, 6H); MS (APCI+): m/z 691.2 (M+H).
solid: H NMR (CDCl3) 9.37 (s, 1H), 7.21–7.16 (m, 5H),
7.13–6.99 (m, 4H), 5.37 (sept, J¼7.2 Hz, 1H), 1.54 (d,
J¼7.2 Hz, 6H); 13C NMR (DMSO-d6) d 181.1, 164.8, 163.9,
160.7, 137.3, 133.7, 131.6, 130.5, 128.9, 128.8, 128.5, 127.5,
126.2, 115.7, 51.3, 21.9; MS (APCI+): m/z 351.9 (M+H).
Anal. Calcd for C21H18F1N1O3 C, 71.78; H, 5.16; N, 3.99.
Found: C, 71.16; H, 5.03; N, 3.95. Mp 219–220 ꢀC.
5.14. 4-(4-Fluoro-phenyl)-5-formyl-1-isopropyl-3-
phenyl-1H-pyrrole-2-carboxylic acid 4-methoxy-
benzylamide (45)
Thionyl chloride (100 mL) was added in one portion to 44
(15.0 g, 42.7 mmol), and the reaction mixture was heated
to 75 ꢀC for 2 h after which time it was cooled to 25 ꢀC
and excess thionyl chloride was removed under reduced
pressure. Dichloromethane (250 mL) was added to the crude
acid chloride and the solution was cooled to 0 ꢀC, and then 4-
methoxybenzyl amine (6.44 g, 47.0 mmol) and triethyl-
amine (8.93 mL, 64.0 mmol) were added sequentially. The
reaction mixture was stirred at 0 ꢀC for an additional 2 h.
Saturated NaHCO3 solution was added and organic layer
was separated, dried (Na2SO4), and concentrated. The prod-
uct was purified by silica gel chromatography (10–20%
EtOAc/hexane) to afford 45 (11.14 g, 55%): 1H NMR
(CDCl3) d 9.44 (s, 1H), 7.17–7.14 (m, 3H), 7.05–7.00 (m,
2H), 6.98 (d, J¼7.6 Hz, 2H), 6.91 (t, J¼8.8 Hz, 2H), 6.71
(d, J¼8.8 Hz, 2H), 6.65 (d, J¼8.8 Hz, 2H), 5.57 (t, J¼
4.8 Hz, 1H), 5.43 (sept, J¼7.2 Hz, 1H), 4.23 (d, J¼5.6 Hz,
2H), 3.72 (s, 3H), 1.61 (d, J¼7.2 Hz, 6H); 13C NMR
(CDCl3) d 181.3, 162.6, 159.2, 132.9, 132.8, 130.1, 129.3,
128.8, 128.7, 128.6, 128.4, 128.3, 127.3, 124.8, 115.4,
114.2, 76.9, 55.4, 51.5, 43.9, 21.7; MS (APCI+): m/z 471.3
(M+H). Mp 169–170 ꢀC.
5.12. 4-(4-Fluoro-phenyl)-5-formyl-1-isopropyl-3-phen-
yl-1H-pyrrole-2-carboxylic acid ethyl ester (43)
A 5-L, four-necked, round-bottomed flask with mechanical
stirrer and reflux condenser was charged with phosphorus
oxychloride (306 g, 1.99 mol). The reactor was cooled
between ꢁ35 and ꢁ40 ꢀC, and a solution of N,N-dimethyl-
formamide (154 mL, 1.99 mol) in 475 mL of 1,2-dichloro-
ethane was slowly added via addition funnel maintaining
the temperature of solution between ꢁ35 and ꢁ40 ꢀC. The
reaction mixture was then allowed to slowly warm to
10 ꢀC and stirred for 1 h. A solution of 27 (350 g, 1.0 mol)
in 1,2-dichloroethane (1.8 L) was then added via addition
funnel maintaining a reaction temperature of 10 ꢀC. The
reaction mixture was then heated to 80 ꢀC for 18 h. The re-
action mixture was then cooled to 25 ꢀC, diluted with hep-
tane/ethyl acetate (1:1, 3.5 L), and poured over ice (3.5 L).
The layers were separated, the aqueous layer (8.8 L) was ex-
tracted with two portions of 1:1 ethyl acetate/heptane (4 L).
The combined organic layers were washed with a saturated
NaHCO3 solution and brine, dried (Na2SO4), and concen-
trated under vacuum to give 43 (371 g, 98%) as a light brown
5.15. 4-(4-Fluoro-phenyl)-5-hydroxymethyl-1-isopro-
pyl-3-phenyl-1H-pyrrole-2-carboxylic acid 4-methoxy-
benzylamide (46)
1
solid that was utilized without additional purification: H
NMR (CDCl3) d 9.52 (s, 1H), 7.21–7.17 (m, 3H), 7.12–
6.93 (m, 6H), 5.54 (sept, J¼6.9 Hz, 1H), 4.13 (q, J¼
7.0 Hz, 2H), 1.69 (d, J¼6.9 Hz, 6H), 0.79 (t, J¼7.2 Hz,
3H); 13C NMR (DMSO-d6) d 181.7, 163.2, 161.2, 133.2,
132.9, 132.8, 130.2, 129.6, 129.2, 128.2, 128.0, 127.1, 115.3,
115.1, 61.8, 51.3, 21.7, 13.6; MS (APCI+): m/z 380.1 (M+H).
Anal. Calcd for C23H12F1N1O3: C, 72.81; H, 5.84; N, 3.69.
Found: C, 72.80; H, 5.76; N, 3.65. Mp 88–90 ꢀC.
To a solution of 45 (11.1 g, 23.7 mmol) in THF (250 mL) at
0 ꢀC was added 1.0 M lithium tri-tert-butoxyalumino-
hydride (28.4 mL, 28.4 mmol). The reaction mixture was
stirred for 30 min at 0 ꢀC after which time TLC analysis in-
dicated the reaction was complete. The solvent was removed
under reduced pressure, and the residue was charged with
ethyl acetate (500 mL) and saturated NaHCO3 solution
(150 mL). The organic layer was separated, dried
(Na2SO4), and concentrated. The resulting oil was purified
by silica gel chromatography (35% EtOAc/hexane) to afford
5.13. 4-(4-Fluoro-phenyl)-5-formyl-1-isopropyl-3-
phenyl-1H-pyrrole-2-carboxylic acid (44)
1
To a solution of 43 (650 g, 1.71 mol) in THF (1.7 L) was
slowly added 6 M aqueous sodium hydroxide (1.13 L,
6.8 mol) followed by MeOH (3.8 L). The solution was
46 (4.64 g, 41%): H NMR (CDCl3) d 7.14–7.10 (m, 3H),
7.00–6.96 (m, 4H), 6.85 (t, J¼8.8 Hz, 2H), 6.72 (d,
J¼8.8 Hz, 2H), 6.65 (d, J¼8.8 Hz, 2H), 5.45 (t, J¼5.2 Hz,

J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
8133
1
1H), 4.97 (sept, J¼7.2 Hz, 1H), 4.57 (d, J¼4.8 Hz, 2H), 4.18
(d, J¼5.2 Hz, 2H), 3.72 (s, 3H), 1.66 (d, J¼7.2 Hz, 6H); 13C
NMR (CDCl3) d 163.7, 159.0, 134.4, 132.1, 130.7, 129.5,
129.2, 128.5, 128.4, 126.7, 125.9, 124.9, 124.0, 115.0,
114.0, 76.9, 62.3, 55.4, 20.0, 43.8, 22.8; MS (APCI+): m/z
473.2 (M+H). Mp 191–192 ꢀC.
hexane) to provide 49 (3.74 g, 77%): H NMR (CDCl3)
d 7.10–7.08 (m, 3H), 6.98–6.91 (m, 4H), 6.83 (t, J¼
8.8 Hz, 2H), 6.72 (d, J¼8.8 Hz, 2H), 6.64 (d, J¼8.8 Hz,
2H), 5.41 (t, J¼5.2 Hz, 1H), 4.78 (sept, J¼7.2 Hz, 1H), 4.17
(d, J¼5.6 Hz, 2H), 4.11–4.05 (m, 1H), 3.72 (s, 3H), 2.81–
2.75 (m, 1H), 2.70–2.59 (m, 1H), 2.28–2.27 (m, 2H), 1.63
(d, J¼7.2 Hz, 6H), 1.61–1.21 (m, 4H), 1.41 (s, 9H); 13C
NMR (CDCl3) d 172.3, 163.8, 160.3, 159.0, 135.0, 133.3,
132.2, 131.8, 130.4, 129.7, 129.2, 128.3, 126.5, 123.9, 120.8,
115.1, 114.9, 114.0, 81.3, 71.5, 69.4, 55.4, 49.2, 43.8, 42.5,
41.8, 38.3, 28.3, 22.8, 21.3; MS (APCI+): m/z 659.1 (M+H).
Anal. Calcd for C39H47F1N2O6: C, 71.10; H, 7.19; N, 4.25.
Found: C, 71.03; H, 7.29; N, 4.07. Mp 141–142 ꢀC.
5.16. [3-(4-Fluoro-phenyl)-1-isopropyl-5-(4-methoxy-
benzylcarbamoyl)-4-phenyl-1H-pyrrol-2-ylmethyl]-
triphenyl-phosphonium bromide (47)
To a solution of 46 (4.64 g, 9.82 mmol) in CH2Cl2 (100 mL)
was added triphenylphosphine hydrobromide (3.37 g,
9.82 mmol) in one portion. The reaction mixture was heated
to 50 ꢀC for 2.5 h after which time all starting material was
determined to be consumed by TLC. The reaction solvent
was removed under reduced pressure and the resulting yel-
low solid was then taken up in iso-propanol (150 mL) in
a flask equipped with a reflux head. Dry iso-propanol was
back-filled as necessary to maintain a consistent reaction
volume for 5 h. The resulting slurry was cooled to 25 ꢀC
and the excess solvent was removed under reduced pressure
5.18. (3R,5R)-7-[3-(4-Fluoro-phenyl)-1-isopropyl-5-(4-
methoxy-benzylcarbamoyl)-4-phenyl-1H-pyrrol-2-yl]-
3,5-dihydroxy-heptanoic acid sodium salt (1)
To a solution of 49 (3.39 g, 5.15 mmol) in MeOH (100 mL)
was added 1.03 N NaOH (5.11 mL, 5.25 mmol) and the re-
action mixture was stirred at 25 ꢀC for 48 h after which
time the reaction solvent was removed under reduced pres-
sure. The resulting solid was then azeotroped with toluene
(3ꢂ100 mL) and triturated with diethyl ether to provide
a light yellow solid that was dried under vacuum at 60 ꢀC
1
to afford 47 (7.82 g, 100%): H NMR (CDCl3) 7.70 (t, J¼
6.6 Hz, 3H), 7.61–7.48 (m, 8H), 7.45–7.42 (m, 2H), 7.24–
7.19 (m, 5H), 7.11–7.03 (m, 3H), 6.86 (dd, J¼8.0, 1.2 Hz,
5H), 6.70–6.67 (m, 3H), 6.61 (d, J¼6.4 Hz, 2H), 6.29 (br
s, 1H), 5.92 (br s, 1H), 4.84 (br s, 1H), 4.15 (d, J¼4.0 Hz,
2H), 3.70 (s, 3H), 1.21 (br s, 6H); 13C NMR (CDCl3) d
163.4, 163.0, 160.5, 159.0, 135.3, 134.5, 134.2, 133.3, 132.0,
130.5, 130.1, 129.7, 129.3, 129.0, 128.4, 126.8, 126.6, 125.3,
125.2, 119.0, 118.9, 118.1, 117.3, 115.7, 115.0, 114.0, 55.4,
51.6, 43.9, 22.8, 22.4; MS (APCI+): m/z 717.3 (M+).
1
to afford 1 (2.99 g, 93%): H NMR (DMSO-d6) d 8.27 (t,
J¼6.0 Hz, 1H), 7.40 (s, 1H), 7.06–6.89 (m, 9H), 6.81 (d, J¼
8.4 Hz, 2H), 6.65 (d, J¼8.8 Hz, 2H), 4.74 (s, 1H), 4.47 (sept,
J¼7.2 Hz, 1H), 4.06 (d, J¼5.6 Hz, 2H), 3.69–3.65 (m, 1H),
3.64 (s, 3H), 3.55–3.52 (m, 1H), 2.65–2.59 (m, 1H), 2.45–
2.42 (m, 1H), 1.95 (dd, J¼15.2, 3.6 Hz, 1H), 1.74 (dd,
J¼15.2, 8.4 Hz, 1H), 1.54–1.17 (m, 4H), 1.43 (d, J¼7.2 Hz,
6H). 13C NMR (DMSO-d6) d 176.6, 164.9, 159.9, 158.6,
135.4, 132.8, 132.6, 132.5, 131.3, 130.1, 129.2, 128.3, 125.8,
122.9, 119.2, 115.5, 115.3, 114.0, 68.8, 67.4, 55.7, 44.7,
44.5, 44.3, 42.8, 37.6, 23.1, 21.7; MS (APCI+): m/z 603.6
(M+H). Anal. Calcd for C35H38F1N2O6Na1$0.5H2O: C,
66.34; H, 6.20; N, 4.42. Found: C, 66.22; H, 6.23; N, 4.20.
Mp 210–211 ꢀC.
5.17. (3R,5R)-7-[3-(4-Fluoro-phenyl)-1-isopropyl-5-(4-
methoxy-benzylcarbamoyl)-4-phenyl-1H-pyrrol-2-yl]-
3,5-dihydroxy-heptanoic acid tert-butyl ester (49)
To a solution of 47 (7.82 g, 9.80 mmol) in THF (200 mL) at
ꢁ78 ꢀC was added 1.0 M NaHMDS in THF (13.7 mL,
13.7 mmol). An orange color was noted as the base was
added. The reaction mixture was stirred below ꢁ70 ꢀC for
5 min after which time a solution of 2014 (2.79 g,
10.8 mmol) in THF (10 mL) was slowly added. After the ad-
dition, the reaction mixture was stirred below ꢁ70 ꢀC for
30 min and then allowed to warm to 25 ꢀC over 1.5 h. The
reaction was quenched by drop-wise addition of saturated
NH4Cl. Ethyl acetate (250 mL) was added and organic layer
was separated, washed with water, dried (Na2SO4), and con-
centrated. The crude product was filtered through a pad of
silica gel (15–20% EtOAc/hexane) to remove Ph3PO and
provide 48 (5.11 g, 7.33 mmol) as an inseparable mixture
of cis/trans olefin isomers, which was taken up in MeOH
(200 mL). To this solution was added 10% Pd/C (500 mg).
The reaction vessel was then evacuated and filled with hy-
drogen gas (50 psi) for 3 h. The reaction mixture was purged
with nitrogen and then filtered through a pad of Celite. The
filtrates were charged with 1 N HCl (10 mL) and the mixture
was stirred for 3 h at 25 ꢀC. Subsequently, the organic sol-
vents were removed under reduced pressure and then ethyl
acetate (200 mL) and saturated NaHCO3 (100 mL) were
added. The organic layer was separated, washed with brine,
dried (Na2SO4), and concentrated. The crude product was
purified by silica gel chromatography (30–70% EtOAc/
References and notes
1. Grundy, S. M. J. Intern. Med. 1997, 241, 295–306.
2. (a) Knopp, R. H. N. Engl. J. Med. 1999, 341, 498–511; (b)
Castelli, W. P. Am. J. Med. 1984, 76, 4–12.
3. For reviews, see: (a) McKenney, J. M. Clin. Cardiol. 2003, 26
(Suppl. III), 32–38; (b) Speidal, K. M.; Hilleman, D. E. Expert
Opin. Pharmacother. 2006, 7, 1291–1304.
4. (a) Grundy, S. M.; Cleeman, J. I.; Merz, N. B.; Brewer, B.;
Clark, L. T.; Hunninghake, D. B.; Pasternak, R. C.; Smith,
S. C.; Stone, N. J. Circulation 2004, 110, 227–239; For addi-
tional discussion, see: (b) O’Keefe, J. H.; Cordain, L.; Harris,
W. H.; Moe, R. M.; Vogel, R. J. Am. Coll. Cardiol. 2004, 43,
2142–2146.
5. For an account of the structure–activity studies and in vivo
pharmacology of this series of HMG-CoA reductase inhibitors,
see: Pfefferkorn, J. A.; Song, Y.; Sun, K.-L.; Miller, S. R.;
Trivedi, B. K.; Choi, C.; Sorenson, R. J.; Bratton, L. D.;
Unangst, P. C.; Larsen, S. D.; Poel, T.-J.; Cheng, X.-M.; Lee,
C.; Erasga, N.; Auerbach, B.; Askew, V.; Dillon, L.;
Hanselman, J. C.; Lin, Z.; Lu, G.; Robertson, A.; Olsen, K.;
Mertz, T.; Sekerke, C.; Pavlovsky, A.; Harris, M. S.;

8134
J. A. Pfefferkorn et al. / Tetrahedron 63 (2007) 8124–8134
Bainbridge, G.; Caspers, N.; Chen, H.; Eberstadt, M. Bioorg.
52, 6879–6892; (d) Gupton, J. T.; Krolikowski, D. A.; Yu,
R. H.; Vu, P.; Sikorski, J. A.; Dahl, M. L.; Jones, C. R.
J. Org. Chem. 1992, 57, 5480–5483; (e) Gupton, J. T.;
Krolikowski, D. A.; Yu, R. H.; Sikorski, J. A.; Riesinger,
S. A. J. Org. Chem. 1990, 55, 4735–4740.
Med. Chem. Lett., in press. doi:10.1016/j.bmcl.2007.05.097
6. For a review on atorvastatin synthesis, see: Li, J.-J.; Johnson,
D. S.; Sliskovic, D. R.; Roth, B. D. Contemporary Drug
Synthesis; John Wiley and Sons: New York, NY, 2004; pp
113–124.
12. Hamanaka, N.; Kosuge, S.; Iguchi, S. Synlett 1990, 139–140.
13. (a) Burkhouse, D.; Zimmer, H. Synthesis 1984, 330–332; (b)
Natchev, I. Synthesis 1987, 1079–1084.
7. Barr, L.; Easton, C. J.; Lee, K.; Lincoln, S. F. Org. Biomol.
Chem. 2005, 3, 2990–2993.
8. Barton, D. H. R.; Zard, S. C. J. Chem. Soc., Chem. Commun.
1985, 1098–1100.
9. Konoike, T.; Araki, Y. J. Org. Chem. 1994, 59, 7849–7854.
10. Chen, K.-M.; Hardtmann, G. E.; Prasad, K.; Repic, O.; Shapiro,
M. J. Tetrahedron Lett. 1987, 28, 155–158.
11. (a) Gupton, J.; Krumpe, K.; Burnham, B.; Dwornik, K.;
Petrich, S.; Du, K.; Bruce, M.; Vu, P.; Vargas, M.; Keertikar,
K.; Hosein, K.; Jones, C.; Sikorski, J. Tetrahedron 1998, 54,
5075–5088; (b) Gupton, J. T.; Krumpe, K. E.; Burnham,
B. S.; Webb, T. M.; Shuford, J. S.; Sikorski, J. A.
Tetrahedron 1999, 55, 14515–14522; (c) Gupton, J. T.;
Petrich, S. A.; Smith, L. L.; Bruce, M. A.; Vu, P.; Du, K. X.;
Dueno, E.; Jones, C. R.; Sikorski, J. A. Tetrahedron 1996,
14. Radl, S. Synth. Commun. 2003, 33, 2275–2283.
15. Zanger, M.; VanderWerf, C. A.; McEwen, W. J. Am. Chem.
Soc. 1959, 81, 3806–3807.
16. Karl Fisher water analysis of various phosphonium salts syn-
thesized according to the method described in Table 3 indicated
that, as expected, they contained up to 1 equiv of water; how-
ever, azeotropic drying with IPA reduced the water content to
<0.5%. Attempts to remove this residual water using simply
high vacuum and elevated temperature were not effective.
17. Kim, S.-H.; Rieke, R. D. J. Org. Chem. 2000, 65, 2322–2330.
18. Cavill, J. L.; Elliott, R. L.; Evans, G.; Jones, I. L.; Platts, J. A.;
Ruda, A. M.; Tomkinson, C. O. Tetrahedron 2006, 62,
410–421.