Design and synthesis of novel, conformationally restricted HMG-CoA reductase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.05.097

Using structure-based design, a novel series of conformationally restricted, pyrrole-based inhibitors of HMG-CoA reductase were discovered. Leading analogs demonstrated potent inhibition of cholesterol synthesis in both in vitro and in vivo models and may

Full text of DOI:10.1016/j.bmcl.2007.05.097

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4531–4537
Design and synthesis of novel, conformationally restricted
HMG-CoA reductase inhibitors
Jeffrey A. Pfefferkorn,* Chulho Choi, Yuntao Song, Bharat K. Trivedi, Scott D. Larsen,
Valerie Askew, Lisa Dillon, Jeffrey C. Hanselman, Zhiwu Lin, Gina Lu,
Andrew Robertson, Catherine Sekerke, Bruce Auerbach, Alexander Pavlovsky,
Melissa S. Harris, Graeme Bainbridge and Nicole Caspers
Pfizer Global Research & Development, Michigan Laboratories, 2800 Plymouth Road, Ann Arbor, MI 48105, USA
Received 25 April 2007; revised 24 May 2007; accepted 30 May 2007
Available online 6 June 2007
Abstract—Using structure-based design, a novel series of conformationally restricted, pyrrole-based inhibitors of HMG-CoA reduc-
tase were discovered. Leading analogs demonstrated potent inhibition of cholesterol synthesis in both in vitro and in vivo models
and may be useful for the treatment of hypercholesterolemia and related lipid disorders.
Ó 2007 Elsevier Ltd. All rights reserved.
Despite advances in diagnosis and treatment, coronary
O
OH OH
heart disease (CHD) remains a leading cause of death
worldwide.1 HMG-CoA reductase inhibitors, which
block the rate limiting step of cholesterol synthesis, rep-
resent the current standard of care for patients with risk
factors for CHD.2,3 As a class, these drugs are well tol-
erated and remarkably effective. However, as new revi-
sions to National Cholesterol Education Program
(NCEP) cholesterol treatment guidelines call for increas-
ingly aggressive LDL-C lowering in at-risk patients (for
example, LDL-C < 70 mg/dl for highest risk patients)4
there is continued need for novel HMG-CoA reductase
inhibitors with increased efficacies.
CO2H
PhHN
N
1: atorvastatin
F
O
OH OH
N
CO2H
PhHN
As described in the preceding paper, during a program
to discover new HMG-CoA reductase inhibitors, we
identified a series of substituted N-iso-propyl pyrroles
(2, Fig. 1) which represent a regioisomeric variation of
the atorvastatin (1) template and offer a unique pharma-
cological profile and excellent potency against HMG-
CoA reductase (2: IC50 = 1.8 nM).5 Structural biology
studies with inhibitor 2 (Fig. 2) revealed that it bound
to the active site of HMG-CoA reductase consistent
with the binding mode of other known statins.6,7 Inter-
2
F
F
Figure 1. Structure of HMG-CoA reductase inhibitors 1 and 2.
estingly, examination of the bound conformation of 2
˚
revealed the close proximity (1.9 A) of the amide NH–
hydrogen and the ortho-hydrogen of the adjacent phenyl
A-ring (see Fig. 2). This proximity suggested the oppor-
tunity for constructing conformationally restricted ana-
logs of general structure 3 (Fig. 3).8 It is precedented
that reducing the number of rotatable bonds in a ligand
can, under appropriate circumstances, lead to an in-
crease in the free energy of binding due to a reduction
Keywords: HMG-CoA reductase inhibitor; Statin; Hypercholes-
terolemia.
*
Corresponding author. Tel.: +1 860 686 3421; fax: +1 860 715
4608; e-mail: jeffrey.a.pfefferkorn@pfizer.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.05.097

4532
J. A. Pfefferkorn et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4531–4537
Br O2N
a
F
F
F
c
NO2
4
5
N
CHO
Bu
R1
b
R1
R1
10: R1 = Br
11: R1 = Me
8: R1 = Br
6: R1 = Br
7: R1 = Me
9: R1 = Me
d
H
N
˚
N
Figure 2. X-ray crystal structure (2.1 A resolution) of inhibitor 2
bound to the active site of HMG-CoA reductase illustrating the
potential for introduction of a conformational restriction.
H
CO2Et
R1
H
CO2Et
R1
e
OH OH
O
n
14: R1 = Br
15: R1 = Me
12: R1 = Br
F
F
R1
N
CO2H
13: R1 = Me
N
Opportunity for
Conformational
Restriction (n = 0,1)
f
3
N
N
F
OHC
CO2Et
R1
OHC
CO2H
R1
g
Figure 3. General structure of conformationally restricted HMG-CoA
reductase inhibitors.
in the entropic cost of binding.9 Moreover, the success
of such strategies is typically enhanced when the bound
conformation of the ligand is established. To determine
if such a conformation restriction would be beneficial in
the current series, a set of prototype analogs were pre-
pared and evaluated as described herein.
F
F
16: R1 = Br
17: R1 = Me
18: R1 = Br
19: R1 = Me
h
O
N
Ph
N
OHC
OHC
C(O)Cl
R1
NH
R1
Synthesis of prototype inhibitors. Compounds for the
current studies were prepared as outlined in Schemes
1–4. As illustrated in Scheme 1, a Barton–Zard synthe-
sis10 was employed to construct key pyrrole intermedi-
ates 22 and 23.11 Initially, 4-fluorobenzyl bromide (4)
was reacted with AgNO2 to provide compound 5. In
parallel, 2-bromobenzaldehyde (6) was condensed with
n-butylamine to generate imine 8. Reaction of 5 and 8
in AcOH at ambient temperature provided nitroalkene
10. Subsequent reaction of nitroalkene 10 with ethyl iso-
cyanoacetate in the presence of DBU resulted in forma-
tion of pyrrole 12. N-Alkylation of pyrrole 12 was
accomplished with i-propyl iodide and KOH to give
14 which was formylated under Vilsmeier–Haack condi-
tions to afford pyrrole carboxaldehyde 16. Saponifica-
tion of 16 followed by conversion to the corresponding
acid chloride and reaction with aniline gave anilide 22
as a key intermediate for subsequent analog synthesis.
Separately, intermediate 23 was prepared from o-tolual-
dehyde (7) via a similar protocol.
i
F
F
22: R1 = Br
23: R1 = Me
20: R1 = Br
21: R1 = Me
Scheme 1. Reagents and conditions: (a) AgNO2, Et2O, 0 ! 25 °C,
24 h, 68%; (b) n-BuNH2, toluene, 120 °C, 2 h, 99%; (c) AcOH, 25 °C,
14 h, 78–92%; (d) DBU, ethyl isocyanoacetate, THF, 25 °C, 8 h, 33%;
(e) KOH, i-PrI, DMSO, 25 °C, 2 h, 51–61%; (f) POCl3, DMF,
dichloroethane, 80 °C, 18 h, 59–84%; (g) NaOH, MeOH, 60 °C, 3 h,
83–85%; (h) SOCl2, 70 °C, 1 h; (i) PhNH2, Et3N, CH2Cl2, 0 ! 25 °C,
12 h, 18–51% (two steps).
Pd(OAc)2, (9,9-dimethyl-4,5-bis(diphenylphosphino))-
xanthane, and Cs2CO3 at elevated temperature to affect
a Buchwald aryl-amidation reaction that generated
lactam 24.12
Installation of the 3,5-dihydroxyhexanoic acid side-
chain began with reduction of the pyrrole carboxalde-
hyde 24 to the corresponding alcohol 25. Treatment of
With pyrrole 22 in hand, analogs such as 28 which con-
tained a 6-membered fused ring system were prepared as
outlined in Scheme 2. Initially, 22 was treated with

J. A. Pfefferkorn et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4531–4537
4533
O
O
NH
O
O
N
N
N
N
Ph
Ph
Ph
Br
CHO
CHO
CHO
H
NH
Br
NH
Ph
N
a
a
F
F
F
F
22
N
23
N
24
N
29
b
b
O
O
O
O
N
Ph
H
Ph
c,d, e
c
P+Ph3Br-
OH
F
N
P+Ph3Br-
N
Ph
Ph
N
N
F
F
31
26
25
30
O
O
O
O
O
O
O
f
d
OHC
OtBu
OHC
OtBu
N
O
O
O
O
O
O
O
N
Ph
N
Ot
Bu
OtBu
Ph
N
27
e,f,g
32
F
F
g,h, i
O
OH OH
O
OH OH
O
O
N
N
Ph
O-Na+
O-Na+
N
Ph
N
F
F
33
28
Scheme 2. Reagents and conditions: (a) Pd(OAc)2, ((9,9-dimethyl-4,5-
bis(diphenylphosphino))xanthane, Cs2CO3, toluene, 100 °C, 16 h,
37%; (b) LiAl(O-t-Bu)3H, THF, 0 °C, 0.5 h, 49%; (c) Ph3PÆHBr,
CH2Cl2, 50 °C, 2 h, 100%; (d) n-BuLi, THF, ꢀ78 ! 25 °C, 2 h, 54%;
(e) HCl, MeOH, 25 °C, 12 h, 81%; (f) Pd–C, H2, EtOH, 25 °C, 6 h,
74%; (g) NaOH, MeOH, 25 °C, 8 h, 95%.
Scheme 3. Reagents and conditions: (a) AIBN, NBS, CCl4, 80 °C, 2 h,
15%; (b) NaH, THF, 0 ! 25 °C, 3 h, 84%; (c) POCl3, DMF,
dichloroethane, 80 °C, 18 h, 80%; (d) LiAl(O-t-Bu)3H, THF, 0 °C,
0.5 h, 67%; (e) Ph3PÆHBr, CH2Cl2, 50 °C, 2 h, 100%; (f) n-BuLi, THF,
ꢀ78 ! 25 °C, 2 h, 65%; (g) HCl, MeOH, 25 °C, 12 h; (h) Pd–C, H2,
EtOH, 25 °C, 6 h, 69% (two steps); (i) NaOH, MeOH, 25 °C, 8 h, 54%.
25 with Ph3PÆHBr resulted in quantitative conversion to
phosphonium salt 26.13 Deprotonation of 26 with
n-BuLi afforded a ylide that was reacted with t-butyl-2-
((4R,6S)-6-formyl-2,2-dimethyl-1,3-dioxan-4-yl)acetate14
to provide olefin 27 as an inconsequential mixture of
geometric isomers. The acetonide protecting group of
27 was removed by treatment with HCl/MeOH, the
side-chain olefin was hydrogenated over 10% Pd/C,
and the terminal ester was saponified with aqueous
NaOH to provide 28 as its carboxylate sodium salt.
the ortho-methyl position with NBS/AIBN provided
benzyl bromide 29. Unexpectedly, this radical reaction
resulted in concomitant deformylation of the pyrrole
ring. Various attempts to suppress this deformylation
reaction were unsuccessful; consequently, intermediate
29 was progressed forward in the synthesis. To effect a
nucleophilic ring closing reaction, 29 was treated with
NaH to provide compound 30 in good yield. Vilsme-
ier–Haack formylation of 30 then re-installed the
carboxaldehyde motif which was subsequently reduced
to the corresponding alcohol with LiAl(O-t-Bu)3H. This
alcohol was then treated with Ph3PÆHBr to afford
phosphonium salt 31. Deprotonation of 31 with n-BuLi
Analogs such as 33 which contained a 7-membered
fused ring were prepared from pyrrole intermediate 23
as outlined in Scheme 3. First, radical bromination of

4534
J. A. Pfefferkorn et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4531–4537
Intermediate 32 was then subjected to the same depro-
OH
tection, hydrogenation, and saponification sequence
described above to provide compound 33 as its carbox-
ylate sodium salt.
NO2
NO2
NO2
a
b
A third type of analog, bearing a 7-membered ring
fusion but lacking the phenyl A-ring, was prepared as
outlined in Scheme 4. Henry reaction between com-
pound 2 and 4-pentenal afforded b-nitro alcohol 34 that,
in a separate step, underwent elimination upon treat-
ment with trifluoroacetic anhydride and triethylamine
to nitro alkene 35.15,16 Reaction of 35 with ethyl iso-
cyanoacetate in the presence of DBU resulted in forma-
tion of a pyrrole which was subsequently alkylated with
i-propyl iodide to provide 36. The terminal olefin of 36
was oxidatively cleaved to an intermediate aldehyde that
was subjected to reductive amination conditions
[PhNH2, Na(OAc)3H] to afford secondary amine 37.
Treatment of amine 37 with trimethyl aluminum at ele-
vated temperature effected efficient conversion to lactam
38.17 Vilsmeier–Haack formylation, reduction, and reac-
tion with Ph3PHBr converted 38 into triphenyl phos-
phonium salt 39 which was then engaged in a Wittig
olefination reaction with t-butyl 2-((4R,6S)-6-formyl-
2,2-dimethyl-1,3-dioxan-4-yl)acetate14 to produce olefin
40 as a cis/trans mixture. Intermediate 40 was then sub-
jected to the same deprotection, hydrogenation, and
saponification sequence described above to provide
compound 41 as its carboxylate sodium salt.
F
2
F
34
F
35
c,d
N
N
e, f
EtO2C
EtO2C
PhHN
37
36
F
F
g
O
N
O
N
Ph
N
h,i, j
Ph
P+Ph3Br-
N
38
F
39
F
O
O
O
OHC
OtBu
The methods outlined in Schemes 1–4 were subsequently
utilized to prepare a series of analogs as highlighted in
Tables 1 and 2.
k
O
O
O
O
N
Ph
N
OtBu
Structure–activity studies. All new analogs were initially
evaluated in
a microsomal HMG-CoA reductase
assay.18 Analogs were also evaluated for their ability
to inhibit cholesterol synthesis in primary rat hepatocyte
cells.18 Additionally, compounds with promising in vitro
activities were then evaluated in an acute in vivo efficacy
model.18
40
F
l,m,n
O
OH OH
O
As shown in Table 1, several SAR trends were noted.
First, comparison of reference compound 2 (IC50
N
Ph
O-Na+
N
=
1.8 nM) to its 6- and 7-membered conformationally con-
strained counterparts 28 (IC50 = 16.7 nM) and 33
(IC50 = 9.6 nM), respectively, revealed that introduction
of the conformational restriction resulted in a 5- to
10-fold loss in enzyme inhibition potency. There was lit-
tle difference in enzyme potency between the 6- and 7-
membered conformational restrictions (i.e., 28 vs 33).
41
F
Scheme 4. Reagents and conditions: (a) CH2@CH(CH2)2CHO, KF,
i-PrOH, 25 °C, 16 h, 59%; (b) trifluoroacetic anhydride, Et3N, CH2Cl2,
ꢀ10 °C, 0.5 h, 84%; (c) DBU, ethyl isocyanoacetate, THF, 25 °C, 18 h,
92%; (d) KOH, i-PrI, DMSO, 25 °C, 2 h, 75%; (e) NaIO4, OsO4 (cat.)
THF:H2O, 0 ! 25 °C, 18 h, 52%; (f) PhNH2, Na(OAc)3BH, AcOH,
dichloroethane, 25 °C, 18 h, 79%; (g) Me3Al, Et3N, CH2Cl2:toluene,
60 °C, 16 h, 94%; (h) POCl3, DMF (cat.) dichloroethane, 80 °C, 18 h,
97%; (i) LiAl(O-t-Bu)3H, THF, 0 °C, 0.5 h, 56%; (j) Ph3PÆHBr,
CH2Cl2, 50 °C, 2 h, 100%; (k) NHMDS, THF, ꢀ78 ! 25 °C, 1 h,
85%; (l) HCl, MeOH, 25 °C, 2 h, 84%; (m) 10% Pd–C, H2, MeOH,
25 °C, 6 h, 72%; (n) NaOH, MeOH, 25 °C, 4 h, 96%.
Unexpectedly, whereas reference compound 2 was equi-
potent in the enzyme (IC50 = 1.8 nM) and cellular assays
(IC50 = 1.0 nM) both conformationally constrained pro-
totype compounds (28 and 33) were substantially less ac-
tive (10- to 100-fold) in the hepatocyte cellular assay
relative to the enzyme assay as illustrated by the data
in Table 1.
followed by addition of t-butyl 2-((4R,6S)-6-formyl-2,2-
dimethyl-1,3-dioxan-4-yl)acetate14 resulted in the for-
mation of olefin 32 as a mixture of geometric isomers.
To better understand the SAR of this conformationally
constrained series, a selected set of substituted anilides
(42–44), benzyl amides (45–48), and a secondary amide

J. A. Pfefferkorn et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4531–4537
4535
Table 1. Inhibitory activity of analogs 28, 33, and 42 – 49 against HMG-CoA reductase and hepatocyte cholesterol synthesis18
O
OH OH
O
R1
N
O-Na+
N
n
F
R1
n
HMG-CoA IC50 (nM)
Hepatocyte inhibition CS IC50 (nM)
Preparation method (scheme)
Simvastatin
2
—
—
—
—
0
49
1.8
4.0
1.0
—
—
2
28
33
42
43
44
45
46
47
48
49
Ph
Ph
16.7
9.6
1250
94
1
3
3-F Ph
3-Cl Ph
4-F Ph
Bn
0
1.9
2.9
246
100
NT
400
68
2
0
2
0
16.4
13.9
3.0
2
0
2
2-F Bn
3-F Bn
4-F Bn
H
0
2
0
37.4
4.7
NT
1000
8.8
2
0
2
0
11.7
2
Table 2. Inhibitory activity of analogs 41 and 50 against HMG-CoA reductase and hepatocyte cholesterol synthesis18
O
OH OH
O
R1
N
O-Na+
N
F
R1
HMG-CoA IC50 (nM)
Hepatocyte inhibition CS IC50 (nM)
Preparation method (scheme)
41
50
Ph
Bn
0.3
3.5
5.9
0.8
4
4
(49) were synthesized and evaluated as illustrated in Ta-
ble 1. Notably, installation of a 3-F (42) or 3-Cl (43)
substituent on the anilide ring afforded improved activ-
ity against HMG-CoA reductase relative to the unsub-
stituted case (i.e., 28). In the benzyl amide series,
installation of a 2-F (46) or a 4-F (48) substituent affor-
ded the best activity. However, in both the anilide ana-
logs (42–44) and benzyl amide analogs (45–48) the
substantial disconnect between enzyme inhibitory
potency and cellular activity persisted. By contrast, sec-
ondary amide 49, where R1 = H, was equipotent in both
in cellular assay.19 Notably, these three analogs each
lacked one phenyl ring relative to the other analogs.
As a result they had reduced molecular weight and lipo-
philicity as illustrated by a comparison of 49 (MW = 480,
cLogD (pH 7.4) = ꢀ0.50, HMG-CoA IC50 = 11.7 nM,
hepatocyte IC50 = 8.8 nM) versus 28 (MW = 556,
cLogD (pH 7.4) = 1.00, HMG-CoA IC50 = 16.7 nM,
hepatocyte IC50 = 1250 nM). One possible explanation
for the observed differences in cellular activities is the ac-
tive transport of selected inhibitors into hepatocyte cells.
For example, other known HMG-CoA reductase inhib-
itors have been reported to undergo active transport
into hepatocyte cells via the Organic Anion Transport-
ing Peptide (OATP) family of transporters.20 It is con-
ceivable that smaller, less lipophilic molecules such as
49 might be better substrates for transport as compared
to larger, more lipophilic, more rigid analogs such as 28
thus accounting for the improved cellular activity of the
former versus the latter compound.
enzyme (IC50 = 11.7 nM)
(IC50 = 8.8 nM).
and
cellular
assays
Analogs (41 and 50, Table 2) which contained a 7-mem-
bered fused ring but lacked the phenyl A-ring were also
evaluated. Interestingly both 41 (R1 = Ph) and 50
(R2 = Bn) exhibited good potency against HMG-CoA,
and they maintained comparable activity in the cellular
hepatocyte assay.
In order to further characterize the potential of this ser-
ies of inhibitors, several representative analogs were
evaluated in an in vivo efficacy model to measure their
ability to acutely inhibit cholesterol synthesis.18 In this
As described above, compounds 41, 49, and 50 exhibited
the best correlation between enzyme and cellular activity
whereas the other analogs were substantially less active

4536
J. A. Pfefferkorn et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4531–4537
model, mice were initially dosed with drug (1 mg/kg)
and then, after 0.5 h, given an intraperitoneal injection
of 14C sodium acetate. After 4.5 h, a whole blood sample
was obtained and analyzed for 14C cholesterol levels
which were compared to untreated control animals to
determine percent inhibition of acute cholesterol synthe-
sis. As illustrated in Table 3, reference compound simva-
statin (ꢀ45% @ 1 mg/kg) and benchmark compound 2
(ꢀ57% @ 1 mg/kg) exhibited robust acute inhibition of
cholesterol synthesis in this model. Three representative
conformationally constrained analogs (28, 41, and 45)
were then evaluated. Among these compounds, 41 was
the most active however, its efficacy was unfortunately
inferior to that of the reference compound 2 and
simvastatin.
Structural biology. Finally, to better evaluate whether
our original hypothesis (that one could lock these inhib-
itors into their active conformation through a ring
fusion) was accurate, we undertook structural biology
studies on a representative analog. Figure 4 illustrates
an overlay of the crystal structures of compounds 2
and 41 bound independently to the active site of
HMG-CoA reductase.21 As shown, the 7-membered
fused ring of 41 allows the amide motif to reasonably
approximate the amide conformation of compound 2
thereby maintaining a key hydrogen bond to Ser-565.
Minor distortion of the anilide ring was noted. Figure
5 illustrates an overlay of the crystal structures of
compounds 2 and 28 bound independently to the active
Figure 5. Overlay of X-ray crystal structures of inhibitors 2 [blue]
˚
(2.1 A resolution) and 28 [purple] (2.0 A resolution) bound to the
˚
active site of HMG-CoA reductase.
site of HMG-CoA reductase.21 As expected and in con-
trast to the case described above, the 6-membered fused
ring of inhibitor 28 substantially distorts the amide tra-
jectory compromising the hydrogen bond to Ser-565.
Additionally the anilide and phenyl A-rings also experi-
ence some distortion. Such changes likely contribute to
the reduced potency of 28 (IC50 = 16.7 nM) relative to
41 (IC50 = 0.3 nM) or 2 (IC50 = 1.8 nM).
Table 3. Activity of representative analogs in acute in vivo efficacy
model18
In conclusion, we have described the structure-based
design, synthesis, and biological evaluation of a series
of conformational constrained HMG-CoA reductase
inhibitors. Selected compounds exhibited good in vitro
potencies and demonstrated inhibition of cholesterol
synthesis in an acute efficacy model; however, in general
these conformationally restricted analogs were less
efficacious than their unconstrained counterparts.
HMG-CoA
IC50 (nM)
Hepatocyte
Inhibition CS
IC50 (nM)
Mouse acute
inhibition CS
(1 mg/kg) (%)
Simvastatin
2
49
1.8
4.0
1.0
ꢀ45
ꢀ57
ꢀ33
ꢀ14
ꢀ39
28
45
41
16.7
13.9
0.3
1250
400
5.9
Acknowledgments
We acknowledge the technical contributions made by
Fang Sun, Loola Al-Kassim, Susan Holly, Cindy Spes-
sard, and Siradanahalli Guru. We would also like to
thank Barry Finzel for his assistance with the X-ray
structures.
References and notes
1. Grundy, S. M. J. Intern. Med. 1997, 241, 295.
2. Castelli, W. P. Am. J. Med. 1984, 76, 4.
3. For review, see: Duriez, P. Expert Opin. Pharmacother.
2001, 2, 1777.
4. 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.
5. 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,
Figure 4. Overlay of X-ray crystal structures of inhibitors 2 [blue]
˚
(2.1 A resolution) and 41 [orange] (2.0 A resolution) bound to the
˚
active site of HMG-CoA reductase.

J. A. Pfefferkorn et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4531–4537
4537
K.; Mertz, T.; Sekerke, C.; Pavlovsky, A.; Harris, M. S.;
Bainbridge, G.; Caspers, N.; Chen, H.; Eberstadt, M.
Bioorg. Med. Chem. Lett. 2007, in press. doi:10.1016/
j.bmcl.2007.05.097.
Kissel, W.; Bolyles, D. C.; Choi, C.; Larsen, S. D.; Song,
Y.; Sun, K.-L.; Miller, S.; Trivedi, B. K. Tetrahedron, in
press.
12. Yang, B. H.; Buchwald, S. L. Org. Lett. 1999, 1, 35.
13. Hamanaka, N.; Kosuge, S.; Iguchi, S. Synlett 1990, 139.
14. Radl, S. Syn. Comm. 2003, 33, 2275.
15. Kambe, S.; Yasuda, H. Bull. Chem. Soc. Jpn. 1968, 41,
1441.
16. Denmark, S. E.; Marcin, L. R. J. Org. Chem. 1993, 58,
3850.
17. (a) Gale, P. A.; Camiolo, S.; Tizzard, G. J.; Chapman, C.
P.; Light, M. E.; Coles, S. J.; Hursthouse, M. B. J. Org.
6. For a review of the X-ray structures of other inhibitors
bound to HMG-CoA reductase, see: Istvan, E. Atheros-
cler. Suppl. 2003, 4, 3.
7. Crystallization protocol and PDB coordinates for the
X-ray structure in Figure 2 can be found in Ref. 5.
8. Additional evidence that such a conformational restriction
might be successful was provided by the fact that the
following N-methyl benzyl amide analog of compound 2
also demonstrated good potency against HMG-CoA
reductase suggesting that N-alkylation necessary for instal-
lation of the conformational restriction would be tolerated:
Chem. 2001, 66, 7849; (b) Schurer, S. C.; Blechert, S. Sylett
1999, 1879.
¨
18. In vitro IC50 values are reported as arithmetic mean for
n P 2 independent measurements unless otherwise noted.
For detailed assay protocols, see: Bratton, L. D.;
Auerbach, B.; Choi, C.; Dillon, L.; Hanselman, J. C.;
Larsen, S. D.; Lu, G.; Olsen, K.; Pfefferkorn, J. A.;
Robertson, A.; Sekerke, C.; Trivedi, B. K.; Unangst, P.
C. Bioorg. Med. Chem. 2007, in press. doi:10.1016/
j.bmc.2007.05.031.
O
OH OH
N
CO2H
N
Me
HMG-CoA IC50 = 3.0 nM
Hepatocyte IC50 = 0.4 nM
19. As described in Ref. 5, in addition to in vitro potency and
in vivo efficacy, an additional area of interest for HMG-
CoA reductase inhibitors is selectivity for hepatocyte cells
versus other cell types as this is thought to confer reduced
risk of statin-induced mylagia. Given that analogs 41, 49,
and 50 exhibited good hepatocyte activity, we also
evaluated them for selectivity in a myocyte cell line (Ref.
5). Unfortunately, none of these compounds exhibited
hepatoselectivity.
F
9. (a)For a review of conformation restriction, see: The
Practice of Medicinal Chemistry; Hart, P. A., Rich, D. H.,
Wermuth, C. G., Eds.; Academic Press: New York, 1996;
p 393; For selected examples, see: (b) Bunch, L.; Liljefors,
T.; Greenwood, J. R.; Frydenvang, K.; Brauner-Osborne,
H.; Krogsgaard-Larsen, P.; Madsen, U. J. Org. Chem.
2003, 68, 1489; (c) Graham, J. Curr. Top. Med. Chem.
2002, 2, 903.
20. Hsiang, B.; Zhu, Y.; Wang, Z.; Wu, Y.; Sasseville, V.;
Yang, W.-P.; Kirchgessner, T. G. J. Biol. Chem. 1999, 274,
37161.
21. Coordinates for the X-ray structures in Figures 4 and 5
have been deposited at the PDB under filenames 2Q6B
and 2Q6C, respectively.
10. Barton, D. H. R.; Zard, S. C. Chem. Commun. 1985, 1098.
11. For additional experimental details on the synthesis of
related compounds, see: Pfefferkorn, J. A.; Bowles, D. M.;