(3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(methyl(1-methyl-1H-1,2, 4-triazol-5-yl)amino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-enoic acid (BMS-644950): A rationally designed orally efficacious 3-hydroxy-3- methylglutaryl coenzyme-A reductase inhibitor wi

Article abstract of DOI:10.1021/jm800001n

3-Hydroxy-3-methylglutaryl coenzyme-A reductase (HMGR) inhibitors, more commonly known as statins, represent the gold standard in treating hypercholesterolemia. Although statins are regarded as generally safe, they are known to cause myopathy and, in rare

Full text of DOI:10.1021/jm800001n

2722
J. Med. Chem. 2008, 51, 2722–2733
(3R,5S,E)-7-(4-(4-Fluorophenyl)-6-isopropyl-2-(methyl(1-methyl-1H-1,2,4-triazol-5-yl)amino)pyri-
midin-5-yl)-3,5-dihydroxyhept-6-enoic Acid (BMS-644950): A Rationally Designed Orally Effica-
cious 3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor with Reduced Myotoxicity
Potential
Saleem Ahmad,* Cort S. Madsen,† Philip D. Stein, Evan Janovitz, Christine Huang, Khehyong Ngu, Sharon Bisaha,
Lawrence J. Kennedy, Bang-Chi Chen, Rulin Zhao, Doree Sitkoff, Hossain Monshizadegan, Xiaohong Yin, Carol S. Ryan,
Rongan Zhang, Mary Giancarli, Eileen Bird, Ming Chang, Xing Chen, Robert Setters, Debra Search, Shaobin Zhuang,
Van Nguyen-Tran, Carolyn A. Cuff, Thomas Harrity, Celia J. Darienzo, Tong Li, Richard A. Reeves, Michael A. Blanar,
Joel C. Barrish, Robert Zahler, and Jeffrey A. Robl
Bristol-Myers Squibb Research & DeVelopment, P.O. Box 4000, Princeton, New Jersey 08543
ReceiVed January 2, 2008
3-Hydroxy-3-methylglutaryl coenzyme-A reductase (HMGR) inhibitors, more commonly known as statins,
represent the gold standard in treating hypercholesterolemia. Although statins are regarded as generally
safe, they are known to cause myopathy and, in rare cases, rhabdomyolysis. Statin-dependent effects on
plasma lipids are mediated through the inhibition of HMGR in the hepatocyte, whereas evidence suggests
that myotoxicity is due to inhibition of HMGR within the myocyte. Thus, an inhibitor with increased selectivity
for hepatocytes could potentially result in an improved therapeutic window. Implementation of a strategy
that focused on in vitro potency, compound polarity, cell selectivity, and oral absorption, followed by extensive
efficacy and safety modeling in guinea pig and rat, resulted in the identification of compound 1b (BMS-
644950). Using this discovery pathway, we compared 1b to other marketed statins to demonstrate its
outstanding efficacy and safety profile. With the potential to generate an excellent therapeutic window, 1b
was advanced into clinical development.
Introduction
cular events.14 Current guidelines from the National Cholesterol
Education Program emphasize the need for aggressive lipid
lowering in patients with multiple risk factors, such as previous
heart attack, hypertension, or diabetes mellitus. For these high-
risk patients, the optimal goal for LDL-C is <100 mg/dL, as
compared to the previous goal of <130 mg/dL. Under these
guidelines it is expected that achieving good treatment-to-goal
success rates will require statins with greater LDL-C-reducing
capacity, similar to atorvastatin and rosuvastatin.15 Thus, the
rationale for statin-based therapies, and perhaps even more so
superstatin-based therapies, has only increased with accumulat-
ing clinical data.
Although statins are regarded as generally safe, skeletal
muscle-related toxicity remains an associated adverse effect.
Important for the clinical diagnosis of skeletal muscle toxicity
is the biochemical marker creatine kinase (CK). Creatine kinase
generates adenosine triphosphate via phosphorylation of ad-
enosine diphosphate and is primarily found in skeletal muscle
and the myocardium. Plasma levels of CK rise after muscle
cell membrane damage and subsequent leakage into the systemic
circulation. A review of the data from multiple clinical trials
involving statin therapy found that myalgia (muscle pain) may
affect an estimated 1% to 5% of patients.16 Although the
definition varies among studies, myopathy has traditionally been
defined as CK levels greater than 10 times the upper limit of
normal accompanied by symptoms (e.g., generalized myalgia,
fatigue, or muscle weakness).17 Myopathy is estimated to occur
in approximately 0.1% of patients who receive statin mono-
therapy.18 If myopathy is not recognized and therapy is
continued, it can progress further to life-threatening rhabdomy-
olysis.19
The risk for coronary heart disease is increased in individuals
with elevated concentrations of plasma low-density lipoprotein-
cholesterol (LDL-Ca).1 Inhibition of HMGR, the rate-limiting
enzyme in cholesterol biosynthesis, has proven to be one of
the most effective approaches for lowering plasma LDL-C and
reducing cardiovascular event rates.2 As part of a compensatory
mechanism to sterol depletion in the liver, inhibition of HMGR
ultimately leads to the increased production of LDL receptors
and subsequent clearance of LDL-C from systemic circulation.3
Although this compensatory mechanism is highly complex and
not fully understood, it has been suggested that sterol depletion
caused by HMGR inhibition activates the sterol response
element binding protein (SREBP), a transcription factor that
promotes expression of the LDL-receptor gene.4
HMGR inhibitors (statins) represent the gold standard in
treating hypercholesterolemia and mixed dyslipidemia. The first-
generation statins such as simvastatin,5 pravastatin,6 lovastatin,7
and fluvastatin8 show a modest ability to lower lipids in man
(LDL-C reduction 20–40%).9 Often referred to as “superstatins”,
the second-generation statins, such as atorvastatin,10 cerivastatin
(marketing discontinued),11 and rosuvastatin,12 elicit greater
reductions in LDL-C (40-60%).13 Several landmark clinical
trials have firmly established the effectiveness of statins in
lowering LDL-C and reducing overall mortality and cardiovas-
* To whom correspondence should be addressed. Phone: 609-252-6955.
Fax: 609-252-6804. E-mail: saleem.ahmad@bms.com.
†
Equal contributing author.
a Abbreviations used: LDL-C, low-density lipoprotein-cholesterol; HMGR,
3-hydroxy-3-methylglutaryl coenzyme-A reductase; CK, creatine kinase;
OATP, organic anion transporter polypeptide; TEMPO, 2,2,6,6-tetrameth-
ylpiperidine 1-oxyl; TC, total cholesterol; ALT, alanine aminotranferase;
AST, aspartate aminotransferase; P-gp, P-glycoprotein.
Rhabdomyolysis is a clinical syndrome that results from
severe and widespread necrosis of skeletal muscle with subse-
10.1021/jm800001n CCC: $40.75
2008 American Chemical Society
Published on Web 04/15/2008

3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9 2723
Figure 1. Structures of five major marketed statins and cerivastatin (which was withdrawn from the market in 2001). Simvastatin is administered
as the corresponding lactone form of the dihydroxy acid.
Figure 2. (Left) X-ray crystal structure of rosuvastatin (cyan) bound between two monomers of HMGR (green and orange).26 The dotted black
lines represent a putative hydrogen bond between one oxygen of the ligand’s methanesulfonyl group and the Ser565 side chain (2.7 Å), and a
favorable electrostatic interaction between the other oxygen of the methanesulfonyl group and Arg568 (3.8 Å to closest atom). Additional side
chains having favorable contacts with the ligand’s N-(methyl)methanesulfonyl group are shown in stick representation; Ser852 does not contact the
ligand but is depicted for clarity. (Right) Compound 1b (magenta) is shown modeled into the HMGR active site.27 Favorable electrostatic interactions
between the denoted nitrogen and Ser565 side chain (3.6 Å) as well as to Arg568 (4.2 Å to closest atom) are represented by dotted black lines.
Additional side chains having favorable contacts with the N-(methyl)triazole ring are shown in stick representation.
quent release of intracellular elements leading to acute renal
failure. Cases of statin-induced rhabdomyolysis are rare; yet,
because fatalities have occurred, rhabdomyolysis represents an
important concern for physicians prescribing this class of drugs.
Despite extensive research, the exact mechanism(s) of statin-
induced myotoxicity remains elusive.20,21 However, data suggest
that the toxicological effects of statins on the myocyte are the
direct result of HMGR inhibition and subsequent reduction of
downstream metabolites, and are not due to the lowering of
plasma cholesterol levels.22 Although statins have been utilized
successfully for nearly two decades, the withdrawal of ceriv-
astatin, due to its unacceptable myotoxicity profile, has led to
a greater appreciation that all statins are not equal with regard
to potential for serious adverse events. The ability to differentiate
statins at the preclinical stage is critical to successfully identify-
ing new and safer drug candidates within this class.
As part of the scientific rationale for the development of a
superstatin with an improved muscle safety profile, the focus
of our program involved implementing a compound design
strategy based on (1) minimizing peripheral exposure and (2)
reducing the capacity for muscle cell penetrance by altering
compound lipophilicity. While hydrophilic statins (e.g., prav-
astatin and rosuvastatin) probably enter the hepatocytes (but not
myocytes) via an active transport mechanism involving organic
anion transporter polypeptide (OATP), hydrophobic statins (e.g.,
cerivastatin) enter hepatocytes via passive diffusion.23,24 A main
focus of our program was based on the hypothesis that, unlike
lipophilic compounds, more polar molecules would have
diminished cell permeability in myocytes resulting in improved
cell selectivity.25 To aid in this effort, a thorough examination
of the X-ray crystal structures of various enzyme-bound statins
was carried out in order to identify polar regions within the
ligand binding site of HMGR that could tolerate or favor polar
groups on a ligand. Of particular interest was the crystal structure
of rosuvastatin, where a pendant methanesulfonyl group oc-
cupies a binding site adjacent to the side chains of serine565
and arginine568 (Figure 2). Molecular modeling revealed that
a small (five- or six-membered ring) heterocycle with optimally
placed heteroatoms would likely capitalize on potential binding
interactions similar to those observed with rosuvastatin. The
nature of the heterocycle along with its substituents could then
be used to modulate hydrophilicity and selectivity of the
molecules.
In order to expeditiously test the hypothesis, we designed an
approach that would introduce a variety of heterocycles at a
late stage of the synthesis, thereby avoiding a lengthy multistep

2724 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9
Ahmad et al.
Scheme 1a
a Reagents: (a) DIBAL/methylene chloride, 61%; (b) TEMPO/buffered bleach/EtOAc, 100% (unpurified); (c) sulfone 6/LiHMDS/THF, 99%; (d)
1-methyl-1H-1,2,4-triazol-5-amine/LiN(TMS)2/THF or DMF; (e) LiN(TMS)2/MeI or MeSO2Cl/THF or DMF; (f) HCl/THF; (g) NaOH/THF-MeOH
(35–70% from 5).
route to these highly complex molecules. This effort, coupled
with extensive in vitro and in vivo safety characterization,
resulted in the identification of compound 1b (BMS-644950)
as a novel, potent, and efficacious HMGR inhibitor with a
potentially superior safety profile.
compounds 1b and 1c. In a similar fashion, other amino
heterocycles could be utilized in this sequence to generate
compounds 8–14.
Results and Discussion
In Vitro Studies/SARs. The primary screens utilized have
been described in detail elsewhere and references therein.32 Rat
HMGR enzyme derived from liver microsomes was utilized to
directly measure enzyme inhibition. Freshly isolated hepatocytes
from rat and rat L6 myocytes grown in culture were utilized to
measure inhibition of cholesterol synthesis within the cell (Table
1). The ratio of the L6 myocyte IC50 to rat hepatocyte IC50 (cell
selectivity) was used as an indicator of myotoxic potential. A
variety of marketed statins were assayed in these in vitro screens
to enable a comparative evaluation versus our program com-
pounds. As anticipated, the more lipophilic statins (e.g., ceriv-
astatin) exhibited the lowest cell selectivity, whereas the more
polar statins (e.g., pravastatin and rosuvastatin) were identified
as being highly selective in these cell-based models. To measure
polarity, a high-throughput reversed-phase HPLC-based log P
assay33 was used for routine screening. The assay was validated
by comparing the data obtained for known statins with the data
from traditional octanol–water partitioning.34 Figure 3 displays
a plot of the two sets of log P values showing a good correlation
(R2 ) 0.98).
Chemistry
A wide variety of novel chemotypes were developed during
the course of this program to explore the effect of structural
modification and polarity on in vitro potency, selectivity, and
in vivo efficacy. The approach depicted in Scheme 1 involved
the utilization of the advanced intermediate 5, which was readily
converted to the various heterocyclic analogues to facilitate rapid
SAR development. Intermediate 5 was prepared in three steps
from the previously reported pyrimidine 2.28 Diisobutylalumi-
num hydride reduction of 2 afforded alcohol 3, which was
oxidized to the corresponding aldehyde 4 using TEMPO
(2,2,6,6-tetramethylpiperidine 1-oxyl) and buffered bleach.
Aldehyde4wasconvertedtointermediate5viaJulia/Kocienski29,30
olefination using the previously described sulfone 6.31 Treatment
of 5 with the anion of 1-methyl-1H-1,2,4-triazol-5-amine
(generated via deprotonation with LiN(TMS)2) afforded 7a in
73% yield. The reaction could be carried out by treating
1-methyl-1H-1,2,4-triazol-5-amine with LiN(TMS)2 in THF or
DMF prior to the addition of the sulfone 5, or alternatively by
addition of LiN(TMS)2 to a solution of 5 and 1-methyl-1H-
1,2,4-triazol-5-amine. Intermediate 7a could be converted
directly to the final compound 1a by deprotection of the side
chain via sequential treatment with HCl and NaOH. Alterna-
tively 7a could be converted to 7b (R ) Me) or 7c (R ) MeSO2)
by treatment with LiN(TMS)2 and methyl iodide or methane-
sulfonyl chloride prior to deprotection of the side chain to access
Most of the novel compounds met our in vitro potency
criteria, showing good enzyme-inhibitory and comparable
hepatocyte activities (Table 1). The in vitro activity did not vary
significantly by modifying the heterocyclic rings. This finding
could be related to the permissive nature of the larger binding
site where this group resides. In contrast, the nature of the
heterocycles along with their substituents extensively modulated

3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9 2725
Table 1. In Vitro Rat Enzyme, Rat Hepatocyte, Rat L6-Myocyte, and in Vivo Rat Data: Effect of Partitioning Coefficient (log P) on Cell Selectivity
enzyme
IC50 (nM)
hepatocytes
IC50 (nM)
myocytes
IC50 (nM)
cell selectivity
ratio
rat ED50 or %
a
b
b
compd
8ae
9a
9b
9c
10a
10b
10c
11a
12a
12b
12c
13a
13b
13c
14a
1a
HPLClog Pc
inhib at 1 mg/kgd
6.9 ( 0.3
2.3 ( 0.5
7.0 ( 1.8
33.0 ( 13
2.4 ( 0.9
2.5 ( 0.1
2.2
12.4 ( 0.7
1.8 ( 0.1
2.5 ( 0.0
3.7 ( 0.1
0.9 ( 0.4
0.5
4.2
0.9 ( 0.0
3.0
2.9 ( 0.1
1.4 ( 0.4
6.2 ( 1.7
3.1 ( 0.4
4.3 ( 1.3
9.8 ( 2.8
31.6 ( 4.4
7.7
1.4 ( 0.5
9
7.1 ( 1.9
132 ( 1 4
41
>1000
55 ( 4
135 ( 29
6277 ( 3176
4.9
1034 ( 342
63 ( 3
0.9
94
4.6
4.0
2.4
3.5
1.8
2.2
3.14
1.79
3.7
2.3
3.18
1.33
1.3
2.0
1.84
-0.1
1.12
20%
43%
ND
64%
60%
86%
61%
54%
62%
67%
23%
50%
0.067 mg/kg
42%
55%
ND
24%
5.5 ( 1.7
1.3 ( 0.2
2.5 ( 1.3
5.4 ( 1.9
1.9
3.7 ( 1.9
1.2 ( 0.2
7.7
1.5 ( 0.5
1.4 ( 0.1
1.0 ( 0.3
13.7 ( 1.2
3.7 ( 0.2
7.5 ( 1.9
3.95 ( 0.8
2.5 ( 0.8
0.6 ( 0.1
6.2 ( 1.3
2.3 ( 0.4
29 ( 4
>181.8
42
54
1162
2.6
279
53
>10 000
2289
>1299
1526
702
3138
>730
751
983 ( 55
3138 ( 832
>10 000
2778 ( 61
>10 000
995 ( 53
78 ( 22
65 ( 17
27 ( 0.8
1.7 ( 0.9
1519 ( 514
1c
1b
>1333
1.7
252
31
108
4.4
0.7
52
2.16 (0.3)
3.76 (1.2)
2.4 (-0.37)
4.84 (1.81)
4.49 (1.66)
2.2 (-0.67)
0.11 mg/kg
0.26 mg/kg
0.35 mg/kg
0.40 mg/kg
0.09 mg/kg
0.68 mg/kg
atorvastatin
rosuvastatin
simvastatin
cerivastatin
pravastatin
a HMGR activity was measured using a previously described method.35 Rat hepatic microsomes were used as the source of the enzyme, and activity was
determined by measuring the conversion of 14C-HMG-CoA to 14C-mevalonic acid. IC50 values were generated from 10-point dose-response curves (n g
2). b Cellular synthesis of cholesterol was measured as incorporation of 14C-acetate into cholesterol in rat hepatocytes and rat L6 myocytes (a rat skeletal
muscle cell line) using an adaptation from a previously described method.36 Effects are expressed as mean IC50 values (concentration of drug producing 50%
inhibition of cholesterol synthesis) based on 6-point dose-response curves derived from independent experiments (n g 2). c Log P was measured for
hydroxy acid forms by the HPLC method;33 values in parentheses are derived from octanol-buffer measurements reported in the literature34 (except for
compound 1b, which was determined in-house). d In vivo inhibition of hepatic cholesterol synthesis in rats as determined from a 4- to 6-point dose-response
curve (n ) 5 animals/dose group). ND ) not determined. e Series a (R ) H), b (R ) Me), c (R ) SO2Me).
polarity thus attenuates permeability in myocytes but not in
hepatocytes, leading to the observed selectivity ratios. Overall,
compounds with HPLC log P values > 3.0 (e.g., compound
10b, selectivity ) 54, HPLC log P ) 3.14) did not meet our in
vitro cell selectivity target of >100. Nonetheless, evaluation in
these in vitro screens provided numerous compounds, including
1b, with excellent enzymatic potency and a desirable cell
selectivity profile, which were suitable for evaluation in
subsequent in vivo efficacy and safety studies.
In Vivo Efficacy in the Acute Rat Model. Multiple
compounds met our criteria for in vitro potency and cell
selectivity. Thus, a medium-throughput in vivo rat assay (as
Figure 3. Correlation of log P values as measured by octanol–water
described in Experimental Section) was used to further dif-
partitioning and HPLC methods (log P values: octanol–water (pH 7.4
ferentiate program compounds. Acute inhibition of cholesterol
buffer)/HPLC (pH 7.0); pravastatin –0.67/2.2, rosuvastatin –0.37/2.4,
synthesis, as indicated by the inhibition of 14C acetate incor-
atorvastatin 1.2/3.76, simvastatin sodium 1.81/4.84, cerivastatin 1.66/
poration into hepatic sterols, was measured after oral adminis-
tration of a single dose of program compounds (Table 1). This
model provided a measure of in vivo activity as well as an
assessment of oral absorption. Not unexpectedly, significant
increases in compound polarity often led to reduced activity in
this assay (most likely due to decreased intestinal permeability
leading to poor oral absorption). This acute pharmacodynamic
assay served to effectively triage compounds for further testing
in more extensive chronic efficacy and safety in vivo models.
In general, compounds that displayed >40% inhibition at e1
mg/kg oral dose in the acute rat model were advanced into the
chronic guinea pig efficacy and safety model (Table 2).
4.49).
the polarity of compounds, providing a diverse set of analogues
with a broad range of log P values. In general, high cellular
selectivity ratios correlated well with decreasing log P values.
This correlation was observed for both the reference statins and
program compounds. The noted difference in activities between
cell types as related to compound polarity has been attributed
to the presence of endogenous transporters (e.g., OATPs)23,24
in hepatocytes, which allow for active transport of these
compounds into the cell. In contrast, the myocyte is thought to
be devoid of the related transporter mechanisms. Increasing

2726 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9
Table 2. In Vivo Guinea Pig Efficacy and Safety Data
Ahmad et al.
myotoxic dose in
rat ED50 or %
HPLC
guinea pig
guinea pigc
compd
10a
10b
12a
9c
13a
13b
1b
atorvastatin
rosuvastatin
simvastatin
pravastatin
cerivastatin
inhib at 1 mg/kga
log Pa
ED50 (mg/kg)b
(safety windowd)
60%
86%
62%
64%
2.2
3.14
2.3
1.8
1.3
90
26
>100 mg/kg (>1)
<10 mg/kg (<0.4)
>90 mg/kge (NA)
>100 mg/kg (NA)
>100 mg/kg (NA)
100 mg/kg (NA)
>100 mg/kg (>3.5)
100 mg/kg (2.0)
100 mg/kg (1.9)
none (>1.5)
>90 mg/kg (30%)
>100 (31%)
50%
>100 (28%)
0.067 mg/kg
0.11 mg/kg
0.26 mg/kg
0.35 mg/kg
0.40 mg/kg
0.68 mg/kg
0.09 mg/kg
2.0
>100 (27%)
2.16
3.76
2.4
4.84
2.2
28
55
54
81
>200 (31%)
1.2
none (NA)
4.49
0.6 mg/kg (0.25)
a Data reproduced from Table 1 for comparison. b Reduction in plasma total cholesterol (TC) levels after treatment with various doses of program compounds
(typically 10, 30, and 100 mg/kg) for 10 days. ED50 values represent the dose (extrapolated from the dose-response curve) in which plasma TC was lowered
by 50% (relative to mean value of vehicle control treatment group) at the end of the study. Values in parentheses represent the percent cholesterol lowering
at the highest dose tested. Dosing of reference statins: cerivastatin 0.1, 0.3, 0.6, 1.2 mg/kg; pravastatin 50, 100, 150, 200 mg/kg; simvastatin 30, 60, 90, 125
mg/kg; rosuvastatin 10, 25, 50, 75, 100 mg/kg; atorvastatin 25, 50, 100, 150, 200, 250 mg/kg, n ) 5 per dose group (except atorvastatin and rosuvastatin
n ) 8). c The lowest dose in which at least one individual animal within the treatment group exhibited a plasma CK value >2.5× the mean plasma CK value
of the vehicle control group or exhibited histopathologic evidence of myofiber degeneration. d The ED50 value for TC lowering divided by the highest dose
in which no animals exhibited a plasma CK value >2.5× the mean plasma CK value of the vehicle control group. e Highest dose ) 90 mg/kg. NA ) not
applicable.
Guinea Pig Model for Efficacy and Myotoxicity. In order
to best define the therapeutic window for new compounds, we
conducted both efficacy and safety studies in a single animal
model. We developed a 10-day efficacy and safety guinea pig
model to evaluate program compounds (Table 2).32 The selection
of this model was based on the ability of statins to reduce
cholesterol levels in guinea pig,37,38 an in-house observation
regarding the susceptibility of this species to statin-induced
myopathy, and similarity of lipid profiles between guinea pig
and human.39 The utility of this model to predict efficacy and
myotoxicity was demonstrated by using marketed statins with
well-established clinical profiles (cerivastatin, rosuvastatin,
atorvastatin, etc.).32
In general, highly polar compounds with log P values e 2.0,
though more selective in vitro (e.g., compounds 9c, 13a, and
13b), were found to exhibit modest in vivo reduction of plasma
cholesterol in guinea pigs (ED50 > 100 mg/kg). In contrast, more
lipophilic compounds such as 10b (HPLC log P ) 3.14) and
cerivastatin exhibited good efficacy and significantly reduced
plasma TC levels. While the diminished in vivo efficacy of the
more polar statins may be rationalized by reduced oral absorp-
tion as a result of attenuated intestinal permeability (caco
permeability <15 nm/s for 13b vs 156 nm/s for 10b), the robust
pharmacodynamic response for 13b in the acute rat model
indicates that poor oral absorption was not an issue. Thus,
differences in metabolism and clearance rates, perhaps species
related, must also be considered. Additionally, the pharmaco-
dynamic response in the rat model (inhibition of hepatic
cholesterol synthesis) is relatively acute, whereas the lowering
of plasma cholesterol in the guinea pig, as mediated by increased
LDL-C uptake by the liver, represents a more chronic response.
With regard to safety, as predicted by the in vitro cell selectivity
data, the more lipophilic compounds were found to be myotoxic
in guinea pigs (as indicated by elevated plasma CK levels and
histopathology) at doses e the ED50 for total cholesterol (TC)
reduction. The in vivo myotoxicity potential of compounds, as
detected by significant elevations of CK in guinea pigs, could
generally be predicted by hepatocyte/myocyte selectivity ratios
of <50–60. While there were some exceptions, compounds with
log P values of ∼2.0 to 2.8 generally displayed good hepatocyte
selectivity as well as good in vivo efficacy and safety. This value
may correspond to an optimal balance between cellular selectiv-
ity and intestinal permeability. The top lead triazole compound
1b, with an HPLC log P value of 2.16, lowered TC in guinea
pig with an ED50 value of 28 mg/kg in the absence of plasma
CK elevation.
The guinea pig data obtained for 1b was compared to data
obtained for multiple reference statins.32 In a direct assessment
(Figure 4), the difference in safety windows of 1b (ED50 ) 28
mg/kg) and cerivastatin (ED50 ) 1.2 mg/kg) was quite obvious.
For 1b, there were no incidences of CK elevation (>2.5 times
vehicle mean) up to the 100 mg/kg dose (∼3.5× the ED50).40
Thus, the myotoxic dose for 1b is defined in this model as being
>3.5× the ED50 (Table 2). In contrast, treatment with ceriv-
astatin at 1.2 mg/kg led to 50% TC lowering with 3 out of 5
animals exhibiting a >72-fold increase in plasma CK levels.
Analysis by this method indicated that 1b has a greater safety
window as compared to cerivastatin (0.25×), rosuvastatin
(1.9×), and atorvastatin (2.0×). Studies conducted with simv-
astatin (safety window >1.5×) were complicated due to poor
solubility of the drug and nonlinear exposure of the compound.
Thus, the safety windows for 1b and simvastatin can not be
justly compared. Pravastatin, which exhibitd no myotoxicity at
doses up to 200 mg/kg, was only moderately efficacious (31%
TC lowering) in this model and thus does not represent a good
comparator.
In addition to myotoxicity, on-target hepatotoxicity represents
an important potential adverse effect of statin therapy.41 Thus,
plasma levels of alanine aminotranferase (ALT) and aspartate
aminotransferase (AST) were monitored during our guinea pig
studies. All reference statins (with the exception of pravastatin)
caused elevations of ALT and AST (>2.5× control) at doses
equal to or just below their respective ED50 values for TC
lowering (data not shown). As a more definitive means for
assessing hepatotoxicity, histopathological analyses were per-
formed for the reference statins and 1b (Table 3). In general,
lesion severity coincided with liver enzyme elevation. For
atorvastatin, rosuvastatin, and 1b, minimal to moderate hepa-
tocellular degeneration was observed at doses near the ED50
for TC lowering. Less severe and/or a lower incidence of
hepatotoxicity was noted for the other three statins tested. Of
note, the histopathologic characteristics of hepatoxicity induced
by tested statins were similar, consistent with a common
mechanism of action. To support this conclusion, additional in

3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9 2727
Figure 4. Hypolipidemic effect (top) and corresponding plasma CK levels (bottom) in guinea pigs treated with 1b or cerivastatin (1.2 mg/kg).
Combined data from two different studies: study 1, n ) 5 at 0, 10, 30, and 100 mg/kg; study 2, n ) 8 at 0, 10, 25, 50, 75, and 100 mg/kg. TC and
CK levels (diamonds) and group means (bars) from individual animals. Data are presented as means ( SD. P values < 0.05 are considered
significant. aSignificant variance between individual CK values within this treatment group precluded meaningful data assessment based on P
value.
Table 3. Comparative Analysis of 1b with Reference Statins for Hepatotoxicity As Determined by Liver Histopathologya
maximum cholesterol
statin
lowering or ED50 (mg/kg)
histopathology of liver (lesions incidences)
1b
28
minimal to moderate hepatocellular degeneration at 25 mg/kg (4/8);
more severe at higher doses
atorvastatin
rosuvastatin
50
54
minimal to moderate hepatocellular degeneration at 50 mg/kg (2/5);
more severe at higher doses
minimal to moderate hepatocellular degeneration at 50 mg/kg (4/8);
more severe at higher doses
simvastatin
pravastatin
81
ca.31%
normal at all doses
minimal hepatocellular degeneration at the highest tested dose of
200 mg/kg (1/5)
cerivastatin
1.2
normal at all doses
a Data reported are from multiple independent studies.

2728 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9
Ahmad et al.
Table 4. Myotoxicity of 1b and Reference Statins in Weanling Rats As
Compared to Their ED50 Values for the Acute Inhibition of Hepatic
Cholesterol Synthesis in Adult Rats
the acute ED50 (0.11 mg/kg) for acute cholesterol synthesis
inhibition. For comparison, the myotoxic doses for reference
statins in weanling rats were 100 mg/kg each for atorvastatin
and rosuvastatin, 50 mg/kg for simvastatin, and <1 mg/kg for
cerivastatin.
acute adult
rat ED50
myotoxic dose
safety
statin
in weanling ratsa,b
windowb
atorvastatin
rosuvastatin
simvastatin
pravastatin
cerivastatin
1b
0.26 mg/kg
0.35 mg/kg
0.40 mg/kg
0.68 mg/kg
0.09 mg/kg
0.11 mg/kg
100 mg/kg
100 mg/kg
50 mg/kg
>500 mg/kg
<1 mg/kg
>50 mg/kg
384
286
125
>735
<11
>455
Summary
In Vitro Data. Compared to the reference statins (Tables 1,
5), compound 1b (IC50 ) 1.4 nM) was among the most potent
in the rat enzyme assay. The mean IC50 values for 1b for
inhibition of cholesterol synthesis in freshly isolated rat hepa-
tocytes and L6 myocytes were 3.95 and 995 nM, respectively.
Compound 1b was also determined to be similarly potent in
human primary hepatocytes, with a calculated IC50 value of 1.7
nM. In rat hepatocytes, the potency of 1b was similar to that of
cerivastatin (2.3 nM), atorvastatin (2.5 nM), and simvastatin
(6.2 nM), but lower than rosuvastatin (0.6 nM), and several-
fold higher than pravastatin (29 nM). With the exception of
pravastatin (IC50 ) 1519 nM), 1b was the least potent of the
reference statins in the L6 myocytes. Compound 1b exhibited
a high L6 myocyte/hepatocyte IC50 selectivity ratio (252×),
which was superior to both the polar statins rosuvastatin (108×)
and pravastatin (52×) as well as the less polar statins simvastatin
(4×) and atorvastatin (31×).46 In agreement with its increased
propensity for myotoxicity in humans, cerivastatin, the least
polar statin of the set, exhibited the lowest selectivity ratio
(0.7×).
a Weanling rats (n ) 5/treatment group) were dosed orally once daily
for 10 days; cerivastatin 1, 1.5, 2, 2.5, 3 mg/kg; pravastatin 300, 400, 500
mg/kg; simvastatin 10, 25, 50, 75 mg/kg; rosuvastatin 50, 100, 150, 200
mg/kg; atorvastatin 50, 100, 125, 150, 200 mg/kg; and compound 1b 10,
20, 30, 40, 50 mg/kg. b Determination of the myotoxic dose and calculation
of the safety window were as described for guinea pig in Table 2.
vivo studies were conducted in which mevalonate (a downstream
product of HMGR) was coadministered with either atorvastatin
(150 mg/kg) or 1b (125 mg/kg) for 10 days (data not shown).
For both compounds, mevalonate supplementation resulted in
normal plasma ALT/AST levels as well as normal liver
histology. Similar to conclusions derived from prior studies in
rabbit,42 it was concluded here that the hepatotoxic effects
observed in guinea pig for 1b and atorvastatin were target based.
Weanling Rat Modeling of Myotoxicity. In order to further
characterize the myotoxic potential of test statins, compound
1b, along with the five reference statins, was evaluated in a
previously described weanling rat model.43 Weanling rats (∼3
weeks old), perhaps due to their rapid growth, appear to be more
sensitive to HMGR inhibition in the muscle, and myopathy
occurs at relatively lower doses of statins as compared to the
adult rat.44 However, the myopathic signs (e.g., muscle weakness
and morbidity/mortality that correlate with increases in plasma
CK levels) are not age-dependent. In addition, while the
lipophilic statins (e.g., simvastatin and lovastatin) have been
reported to induce significant myopathy in this model, the
hydrophilic pravastatin does not cause myopathy.43
Animals were orally dosed once daily for 10 days. On the
final day of study animals were sacrificed and assessed for
myotoxicity based on plasma CK and skeletal muscle histopa-
thology (Table 4). Due to a strong hepatic compensatory
response, chronic statin treatment in rats has a relatively minimal
effect on plasma TC levels.45 Thus, the ED50 values for
inhibition of hepatic cholesterol synthesis derived from the acute
rat model (Table 1) were used as surrogate efficacy markers
for making comparisons to the respective myotoxic findings and
for estimating safety margins. Clearly, there are limitations to
this approach; such comparisons do not account for potential
differences between adult and weanling animals, nor do they
account for acute versus chronic pharmacology. Nonetheless,
this approach provides a reasonable means for relating toxicity
with pharmacodynamics. The calculated safety windows for the
reference statins were as follows: cerivastatin (most toxic) >>>
simvastatin > rosuvastatin > atorvastatin >>> pravastatin
(least toxic). Similar to the guinea pig, the weanling rat was
increasingly susceptible to myotoxicity when exposed to the
more lipophilic compounds. Compound 1b, with an acute rat
ED50 value of 0.11 mg/kg, was dosed to weanling rats at 10,
20, 30, 40, and 50 mg/kg, corresponding to approximate
multiples of 100-, 200-, 300-, 400-, and 500-fold the acute ED50
value, respectively. None of the animals dosed with 1b showed
signs of myopathy based on plasma CK levels, which were
comparable to the vehicle-treated group. Therefore, the myotoxic
dose for 1b is >50 mg/kg, which corresponds to >455-fold
In Vivo Data. In the guinea pig model (Tables 2, 5), 1b
displayed an ED50 ) 28 mg/kg (superior to atorvastatin and
rosuvastatin) with maximum efficacy in the superstatin range
(∼60% total cholesterol lowering) with no CK elevation
incidences (>2.5× vehicle mean) at doses up to 3.5× the ED50
for TC lowering (better than atorvastatin). In the weanling rat
model (Tables 4, 5), 1b did not result in a CK increase (>2.5×
vehicle mean) at doses >455× the ED50 for acute inhibition of
cholesterol synthesis. 1b induces hepatotoxicity in the guinea
pig in a manner that is similar to other marketed statins with
respect to both frequency and pathology, and the toxicity can
be completely ameliorated with the coadministration of
mevalonate.
Pharmacokinetic Data. Given that the liver is the target
organ for statin-mediated LDL-C lowering, and because in-
creased systemic exposure could lead to enhanced myotoxicity,
adequate oral absorption combined with efficient liver extraction
represent desirable pharmacokinetic properties for this class of
compounds. The oral bioavailability (AUCPO/AUCIV) of 1b was
determined in the guinea pig (41%), rat (2.8%), dog (11%), and
monkey (ca. 1%). Oral absorption (AUCPO/AUCIPT) was
determined in the guinea pig (64%) and rat (20%). These results
indicate moderate to highly efficient liver extraction in our
preclinical models (35% and 86% in guinea pig and rat,
respectively). A comparative analysis of drug levels in blood
(portal vein, hepatic vein, and inguinal vein), liver, and bile
samples obtained from an orally dosed monkey revealed 98%
hepatic extraction. High concentrations of drug in liver and bile
suggest that oral absorption was adequate in monkey. The
animal pharmacokinetic data for 1b compare favorably with the
human gut absorption and liver extraction data reported for
atorvastatin (30% and >70%, respectively) and rosuvastatin
(20% and 90%, respectively).47
Additional Profiling. Compound 1b is neither a substrate
for nor an inhibitor of cytochrome P450 enzymes (CYP1A2,
2C8, 2C9, 2C19, 2D6, and 3A4) and thus displays a low
potential for drug-drug interactions. In addition, 1b is not an

3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9 2729
Table 5. Summary of in Vitro and in Vivo Profiling Data for 1b As Compared to Reference Statins
1b
atorvastatin rosuvastatin simvastatin pravastatin cerivastatin
3.8 2.4 4.8 2.2 4.5
31.6
log P (HPLC)
2.2
in vitro data
rat enzyme IC50 (nM)
1.4
6.2
3.1
4.3
9.8
rat hepatocyte/rat myocyte IC50 (nM)
cell selectivity
guinea pig data ED50 (mg/kg)
3.95/995 2.5/78
0.6/65
108
54
6.2/27
4.4
81
29/1519
52
NA (31%) 1.2
2.3/1.7
0.7
252
28
31
50
myotoxic dose (CK > 2.5×)
none
none
100 mg/kg 100 mg/kg none
none
none
NA
none
0.68
>500
>735×
0.3 mg/kg
myotoxic dose (CK > 10×)
myotoxic dose (CK > 2.5×) relative to ED50
CK > 10000 U/L
none
none
1.9×
none
0.35
100
none
>1.5×
none
0.40
0.6 mg/kg
0.25×
yes
0.09
<1
<11×
>3.5× 2.0×
none
0.11
>50
none
0.26
100
rat data
acute ED50 (mg/kg)
myotoxic dose (CK > 2.5) in WR (mg/kg)
myotoxic dose (CK > 2.5) in WR (mg/kg) as related to ED50 >455× 384×
50
286×
125×
and were characterized by NMR analysis. Samples prepared for in
vivo studies were of g98% purity.
inducer of CYP3A4 and is neither a P-gp substrate nor an
inhibitor. There was no evidence of significant receptor binding
or enzyme inhibition in a broad receptor screening panel. Ames
and in vitro micronucleus assays were negative, suggesting a
low potential for mutagenicity. Furthermore, 1b displayed no
significant in vitro or in vivo cardiovascular liabilities.
Clinical Data. Compound 1b was dosed to humans in a
placebo-controlled, ascending single-dose study to evaluate the
safety, pharmacokinetics, and pharmacodynamics in healthy
subjects.48 Single oral doses of 1b in the range 0.25 to 120 mg
were safe and well tolerated. Indicative of its potential as a statin
with an excellent efficacy and safety (myotoxicity) profile,
compound 1b produced dose-dependent reductions in plasma
mevalonic acid with minimal systemic exposure. As an example,
at the 10 mg dose (N ) 6), compound 1b generated a geometric
mean Cmax of 4.72 (33% CV) ng/mL and a mean percent change
((SD) from baseline in plasma mevalonic acid AUC(0–24 h) of
-48.19 ((8.16). In comparison, rosuvastatin, when dosed (10
mg each morning) to healthy adult volunteers (n ) 21) for 14
days in a open-label clinical trial, generated a geometric mean
Cmax of 4.58 ng/mL and a mean percent change from baseline
in plasma mevalonic acid AUC(0–24 h) of -29.9.49
(4-(4-Fluorophenyl)-6-isopropyl-2-(methylsulfonyl)pyrimidin-5-
yl)methanol (3). A solution of DIBAL in methylene chloride (1
M, 25 mL, 25 mmol) was added over 5 min to a stirred solution of
ethyl 4-(4-fluorophenyl)-6-isopropyl-2-(methylsulfonyl)pyrimidine-
5-carboxylate28 (3.7 g, 10.1 mmol) in toluene (100 mL) at -78
°C. The reaction mixture was stirred at -78 °C for 30 min followed
by a slow addition of saturated ammonium chloride. The mixture
was stirred at RT for 30 min, and the organic layer was dried
(MgSO4) and concentrated. The crude product was subjected to
flash chromatography (silica gel/hexanes-ethyl acetate 80:20 to
20:80) to give the title compound 3 as a white solid (2.0 g, 61%):
1H NMR (400 MHz, CDCl3) δ 7.81 (2 H, dd, J ) 5.3, 8.6 Hz),
7.16 (2 H, t, J ) 8.6 Hz), 4.72 (2 H, s), 3.64 (1 H, m), 3.39 (3 H,
s), 2.69 (1 H, br s), 1.36 (6 H, d, J ) 6.7 Hz); MS (ESI) m/z 325
(M + H)+; analytical HPLC (system A) tR ) 2.89 min, (system
B) tR ) 2.29 min.
4-(4-Fluorophenyl)-6-isopropyl-2-(methylsulfonyl)pyrimidine-5-
carbaldehyde (4). To a solution of compound 3 (2.0 g, 6.2 mmol)
in ethyl acetate (20 mL) were added potassium bromide (73 mg,
0.61 mmol) and TEMPO free radical (10 mg, 0.064 mmol), and
the reaction mixture was cooled to 0 °C. A solution of buffered
commercial bleach (0.824 M, adjusted to pH 9.4 with solid sodium
bicarbonate, 15 mL) was then added over 0.5 h while maintaining
the reaction temperature between 5 and -5 °C. The mixture was
then washed sequentially with water, 5% sodium thiosulfate, 1 N
NaOH, and brine. The organic phase was dried (MgSO4) and
concentrated, and the crude aldehyde 4 thus obtained (1.98 g, 100%,
unpurified) was used as such in the next step. 1H NMR (400 MHz,
CDCl3) δ 10.05 (1 H, s), 7.69 (2 H, dd, J ) 5.2, 8.8 Hz), 7.26 (2
H, t, J ) 8.5 Hz), 3.88 (1 H, m), 3.44 (3 H, s), 1.37 (6 H, d, J )
6.8 Hz); LC/MS and LC showed a mixture of the aldehyde and
the corresponding methanol-hemiacetal; MS (ESI) m/z 323 (M +
H)+ and 355 (M + MeOH + H)+; analytical HPLC (system A) tR
) 2.36, 3.46 min.
In conclusion, compound 1b met or exceeded our initial
criteria for clinical development by generating an excellent
profile both in vitro and in the guinea pig and rat models. These
data, combined with results from additional toxicity studies and
early clinical trials, provide evidence that compound 1b has the
potential to be a highly efficacious statin with a reduced
propensity for myotoxicity, leading to an overall excellent
risk-benefit profile.
Experimental Section
General Procedures. All reagents and solvents purchased from
commercial resources were used without further purification. Silica
gel 60 (Merck) was used for flash chromatography, and silica gel
60 F254 (Merck) plates were used for thin-layer chromatography
(TLC). TLC spots were examined under UV light at 254 nm.
Analytical HPLC analyses were performed under the following
conditions. System A: Phenomenex Prime S5 C18 4.6 × 50 mm
column/water-MeOH-H3PO4 90:10:0.2 to 10:90:0.2 gradient over
4 min at 4 mL/min with 1 min hold at the end of the gradient.
System B (used with LC/MS): Phenomenex S5 C18 4.6 × 30 mm
column/water-MeOH-TFA 90:10:0.1 to 10:90:0.1 gradient over
2 min at 5 mL/min with 1 min hold at the end of the gradient.
System C (used with LC/MS): YMC S5 C18 4.6 × 50 mm column/
water-MeOH-TFA 90:10:0.1 to 10:90:0.1 gradient over 4 min at
4 mL/min with 1 min hold at the end of the gradient. System D:
YMC S5 C18 4.6 × 50 mm column/water-MeOH-H3PO4 90:
10:0.2 to 10:90:0.2 gradient over 4 min at 4 mL/min with 1 min
hold at the end of the gradient. NMR data were obtained using a
Bruker or JEOL 400 or 500 MHz spectrometer. All final compounds
were of g95% purity by the LC/MS and analytical HPLC systems
tert-Butyl 2-((4R,6S)-6-((E)-2-(4-(4-fluorophenyl)-6-isopropyl-2-
(methylsulfonyl)pyrimidin-5-yl)vinyl)-2,2-dimethyl-1,3-dioxan-4-
yl)acetate (5). A solution of lithium bis(trimethysilyl)amide (1 M
in THF, 667 mL, 0.667 mol) was added with stirring to a solution
of the aldehyde 4 from above (118 g, 0.365 mol) and compound
6
31 (165 g, 0.365 mol) in THF (1.5 L) at -60 °C over 15 min. The
mixture was allowed to come to RT, washed with saturated sodium
bicarbonate solution, dried (MgSO4), and concentrated. The crude
product was flash chromatographed using a short column of silica
gel (hexane-ethyl acetate 100:0 to 50:50 gradient) to give the title
compound 5 as a white solid (200 g, 99.6%): 1H NMR (400 MHz,
CDCl3) δ 7.70 (2 H, dd, J ) 8.8, 5.3 Hz), 7.11 (2 H, t, J ) 8.6
Hz), 6.61 (1 H, dd, J ) 16.1, 1.5 Hz), 5.60 (1 H, dd, J ) 16.1, 4.5
Hz), 4.29 (1 H, m), 4.46 (1 H, m), 3.46 (1 H, m), 3.40 (3 H, s),
2.45 (1 H, dd, J ) 15.1, 7.1 Hz), 2.30 (1 H, dd, J ) 15.4, 6.3 Hz),
2.3 Hz), 1.40-1.55 (17 H, m, 3 singlets overlapping a multiplet),
1.32 (6 H, d, J ) 6.8 Hz); MS (ESI) m/z 549 (M + H)+; analytical
HPLC purity (system C) 88%, tR ) 3.58 min.

2730 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9
Ahmad et al.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(methyl(1-
methyl-1H-1,2,4-triazol-5-yl)amino)pyrimidin-5-yl)-3,5-dihydroxy-
hept-6-enoate (1b). A solution of lithium bis(trimethylsilyl)amide
(1 M in THF, 146 mL, 146 mmol) was added dropwise at -60 °C
to a stirred solution of 1-methyl-1H-1,2,4-triazol-5-amine50 (14.3
g, 145.9 mmol) and compound 5 (33 g, 60.2 mmol) in 300 mL of
DMF. The reaction mixture was stirred at -60 °C for 30 min
followed by the addition of methyl iodide (68 g, 479 mmol). The
mixture was allowed to come to RT, stirred for 30 min, and
partitioned between ethyl acetate and saturated sodium bicarbonate
solution. The organic layer was dried (MgSO4) and concentrated,
and the residue was subjected to flash chromatography (silica gel/
hexanes-ethyl acetate 100:0 to 50:50 gradient) to afford tert-butyl
2-((4R,6S)-6-((E)-2-(4-(4-fluorophenyl)-6-isopropyl-2-(methyl(1-
methyl-1H-1,2,4-triazol-5-yl)amino)pyrimidin-5-yl)vinyl)-2,2-di-
methyl-1,3-dioxan-4-yl)acetate (7b) as a pale gummy solid (25 g,
72%): 1H NMR (400 MHz, CDCl3) δ 7.86 (1 H, s), 7.56 (2 H, dd,
J ) 9.07, 5.54 Hz), 7.05 (2 H, t, J ) 8.81 Hz), 6.47 (1 H, dd, J )
16.12, 1.01 Hz), 5.40 (1 H, dd, J ) 16.37, 5.79 Hz), 4.40 (1 H, dd,
J ) 10.58, 5.54 Hz), 3.67 (3 H, s), 4.27 (1 H, m), 3.29 (1 H, m),
3.58 (3 H, s), 2.44 (1 H, dd, J ) 15.11, 6.55 Hz), 2.29 (1 H, dd,
J ) 15.36, 6.30 Hz), 1.48 (3 H, s), 1.50 (2 H, m), 1.46 (9 H, s),
1.39 (3 H, s), 1.14 (6 H, t, J ) 6.55 Hz); MS (ESI) m/z 581 (M +
H)+; analytical HPLC (system D) tR ) 4.59 min.
Compound 7b was dissolved in THF (75 mL) and treated with
6 N HCl (22 mL). After 15 min, the reaction mixture was made
basic by the addition of 6 N NaOH (30 mL) followed by the
addition of methanol (5 mL). The mixture was stirred at RT for 15
min and concentrated, and the residue was subjected to reversed-
phase chromatography (C18/water-MeOH 100:0 to 50:50 gradient)
to afford the title compound as a white solid (11.1 g, 36% yield
from the sulfone 5): 1H NMR (500 MHz, CD3OD) δ 7.91 (1 H, s),
7.61 (2 H, dd, J ) 8.3, 5.5 Hz), 7.14 (2 H, t, J ) 8.8 Hz), 6.54 (1
H, d, J ) 16.5 Hz), 5.48 (1 H, dd, J ) 16.2, 6.3 Hz), 4.26-4.39
(1 H, m), 3.83-3.97 (1 H, m), 3.67 (3 H, s), 3.53 (3 H, s),
3.34-3.49 (1 H, m), 2.17-2.37 (2 H, m), 1.56-1.70 (1 H, m),
1.39-1.54 (1 H, m), 1.16 (3 H, s), 1.15 (3 H, s); MS (ESI) m/z
525 (M + H)+. Anal. (C24H28FN6NaO4 ·1.59H2O) C, H, N.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(1-methyl-
1H-1,2,4-triazol-5-ylamino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-
enoate (1a). This compound was prepared in a manner similar to
that described for the synthesis of compound 1b except that the
step involving alkylation with MeI was omitted. 1H NMR (500
MHz, CD3OD) δ 7.84 (s, 1 H), 7.61 (dd, J ) 8.6, 5.5 Hz, 2 H),
7.14 (t, J ) 8.9 Hz, 2 H), 6.54 (d, J ) 15.9 Hz, 1 H), 5.48 (dd, J
) 16.2, 6.4 Hz, 1 H), 4.32 (q, J ) 6.7 Hz, 1 H), 3.87-3.95 (m, 1
H), 3.38-3.48 (m, 1 H), 3.76 (s, 3 H), 2.23 (dd, 1 H), 2.31 (dd, 1
H), 1.58-1.68 (m, 1 H), 1.43-1.51 (m, 1 H), 1.20 (d, J ) 6.7 Hz,
6 H); MS (ESI) m/z 471 (M + H)+; analytical HPLC (system B)
tR ) 1.17 min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(N-(1-
methyl-1H-1,2,4-triazol-5-yl)methylsulfonamido)pyrimidin-5-yl)-
3,5-dihydroxyhept-6-enoate (1c). This compound was prepared in
a manner similar to that described for the synthesis of compound
1b except that MeI was substituted with methanesulfonyl chloride.
1H NMR (400 MHz, CD3OD) δ 8.04 (1 H, s), 7.67 (2 H, dd, J )
8.3, 5.5 Hz), 7.16 (2 H, t, J ) 9.0 Hz), 6.60 (1 H, d, J ) 17.0 Hz),
5.60 (1 H, dd, J ) 16.2, 6.0 Hz), 4.40-4.25 (1 H, m), 3.90 (4 H,
br s), 3.75 (3 H, s), 3.55-3.47 (1 H, m), 2.34 (1 H, dd, J ) 15.5,
4.9 Hz), 2.25 (1 H, dd, J ) 15.5, 7.9 Hz), 1.70-1.64 (1 H, m),
1.58-1.50 (1 H, m), 1.22 (3 H, d, J ) 1.58 Hz), 1.20 (3 H, d, J )
1.58 Hz); MS (ESI) m/z 549 (M + H)+; analytical HPLC (system
B) tR ) 1.38 min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(5-phenyl-
1,2,4-thiadiazol-3-ylamino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-
enoate (8a). This compound was prepared from compound 5 and
5-phenyl-1,2,4-thiadiazol-3-amine as described for compound 1a. 1H
NMR (400 MHz, CD3OD) δ 8.04 (2 H, dd, J ) 8.1, 1.5 Hz), 7.77 (2
H, dd, J ) 9.1, 5.5 Hz), 7.53-7.60 (3 H, m), 7.18 (2 H, t, J ) 8.8
Hz), 6.62 (1 H, dd, J ) 16.1, 1.0 Hz), 5.54 (1 H, dd, J ) 16.1, 6.0
Hz), 4.36 (1 H, q, J ) 6.0 Hz), 3.91-3.98 (1 H, m), 3.43-3.57 (1 H,
m), 2.34 (1 H, dd, J ) 19.6, 4.5 Hz), 2.25 (1 H, dd, J ) 15.1, 7.6 Hz),
1.61-1.71 (1 H, m), 1.47-1.55 (1 H, m), 1.32 (6 H, d, J ) 6.5 Hz);
MS (ESI) m/z 550 (M + H)+; analytical HPLC (system B) tR ) 1.71
min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(isoxazol-
3-ylamino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-enoate (9a). This
compound was prepared from compound 5 and isoxazol-3-amine as
described for compound 1a. 1H NMR (400 MHz, CD3OD) δ 8.46 (1
H, d, J ) 1.5 Hz), 7.67 (2 H, dd, J ) 9.1, 5.5 Hz), 7.26 (1 H, d, J )
1.5 Hz), 7.17 (2 H, t, J ) 8.8 Hz), 6.56 (1 H, dd, J ) 16.1, 1.0 Hz),
5.49 (1 H, dd, J ) 16.1, 6.5 Hz), 4.33 (1 H, q, J ) 6.5 Hz), 3.79-4.12
(1 H, m), 3.34-3.65 (1 H, m), 2.32 (3 H, dd, J ) 15.1, 4.5 Hz), 2.24
(1 H, dd, J ) 15.1, 8.1 Hz), 1.57-1.70 (1 H, m), 1.40-1.53 (1 H,
m), 1.29 (6 H, d, J ) 6.5 Hz); MS (ESI) m/z 457 (M + H)+; analytical
HPLC (system A) tR ) 3.33 min, (system B) tR ) 1.52 min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(isoxazol-
3-yl(methyl)amino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-enoate (9b).
This compound was prepared from compound 5 and isoxazol-3-amine
as described for compound 1b. 1H NMR (400 MHz, D2O) δ 8.30 (1
H, d, J ) 2.0 Hz), 7.43 (2 H, dd, J ) 8.6, 5.5 Hz), 7.09 (2 H, t, J )
8.8 Hz), 6.99 (1 H, d, J ) 2.0 Hz), 6.51 (1 H, d, J ) 15.6 Hz), 5.34
(1 H, dd, J ) 16.1, 7.1 Hz), 4.22 (1 H, q, J ) 6.5 Hz), 3.58-4.07 (1
H, m), 3.47 (3 H, s), 3.24-3.38 (1 H, m), 2.05-2.28 (2 H, m),
1.42-1.64 (1 H, m), 1.24-1.47 (1 H, m), 1.14 (3 H, d, J ) 6.5 Hz),
1.15 (3 H, d, J ) 7.1 Hz); MS (ESI) m/z 471 (M + H)+; analytical
HPLC (system A) tR ) 3.80 min, (system B) tR ) 1.88 min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(N-(isox-
azol-3-yl)methylsulfonamido)pyrimidin-5-yl)-3,5-dihydroxyhept-6-
enoate (9c). This compound was prepared from compound 5 and
isoxazol-3-amine as described for compound 1c. 1H NMR (400
MHz, CD3OD) δ 8.67 (1 H, d, J ) 1.5 Hz), 7.57 (2 H, dd, J ) 8.8,
5.3 Hz), 7.03-7.11 (2 H, m), 6.68 (1 H, d, J ) 2.0 Hz), 6.52 (1 H,
d, J ) 16.1 Hz), 5.52 (1 H, dd, J ) 16.1, 6.0 Hz), 4.07-4.33 (1 H,
m), 3.75-3.90 (1 H, m), 3.60 (3 H, s), 3.23-3.46 (1 H, m),
1.93-2.36 (2 H, m), 1.48-1.67 (1 H, m), 1.38-1.47 (1 H, m),
1.12 (6 H, d, J ) 7.1 Hz); MS (ESI) m/z 535 (M + H)+; analytical
HPLC (system A) tR ) 3.11 min, (system B) tR ) 0.95 min. Anal.
(C24H26FN4NaO7S·2.1H2O) C, H, N, F, Na, S.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(1-methyl-
1H-pyrazol-5-ylamino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-
enoate (10a). This compound was prepared from compound 5 and
1
1-methyl-1H-pyrazol-5-amine as described for compound 1a. H
NMR (400 MHz, CD3OD) δ 7.62 (2 H, dd, J ) 9.1, 5.5 Hz), 7.38
(1 H, d, J ) 2.0 Hz), 7.15 (2 H, t, J ) 8.8 Hz), 6.54 (1 H, dd, J )
16.1, 1.5 Hz), 6.39 (1 H, d, J ) 2.0 Hz), 5.46 (1 H, dd, J ) 16.1,
6.0 Hz), 4.31 (1 H, q, J ) 6.4 Hz), 3.81-4.13 (1 H, m), 3.76 (3 H,
s), 3.34-3.54 (1 H, m), 2.31 (1 H, dd, J ) 15.1, 5.0 Hz), 2.22 (1
H, dd, J ) 15.1, 8.1 Hz), 1.57-1.71 (1 H, m), 1.32-1.56 (1 H,
m), 1.23 (6 H, d, J ) 6.5 Hz); MS (ESI) m/z 470 (M + H)+;
analytical HPLC (system A) tR ) 3.21 min, (system B) tR ) 1.44
min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(methyl(1-
methyl-1H-pyrazol-5-yl)amino)pyrimidin-5-yl)-3,5-dihydroxyhept-
6-enoate (10b). This compound was prepared from compound 5
and 1-methyl-1H-pyrazol-5-amine as described for compound 1b.
1H NMR (400 MHz, CD3OD) δ 7.59 (2 H, dd, J ) 8.6, 5.5 Hz),
7.47 (1 H, d, J ) 2.0 Hz), 7.12 (2 H, t, J ) 8.8 Hz), 6.52 (1 H, d,
J ) 16.1 Hz), 6.16 (1 H, d, J ) 2.0 Hz), 5.44 (1 H, dd, J ) 16.1,
6.6 Hz), 4.30 (1 H, q, J ) 6.4 Hz), 3.87-3.93 (1 H, m), 3.63 (3 H,
s), 3.48 (3 H, s), 3.34-3.43 (1 H, m), 2.18-2.34 (2 H, m),
1.57-1.67 (1 H, m), 1.42-1.51 (1 H, m), 1.14 (6 H, d, J ) 6.6
Hz); MS (ESI) m/z 484 (M + H)+; analytical HPLC (system A) tR
) 3.65 min, (system B) tR ) 1.67 min. Anal. (C25H29FN5NaO4 ·
1.0H2O) C, H, N, F, Na.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(N-(1-
methyl-1H-pyrazol-5-yl)methylsulfonamido)pyrimidin-5-yl)-3,5-di-
hydroxyhept-6-enoate (10c). This compound was prepared from
compound 5 and 1-methyl-1H-pyrazol-5-amine as described for
compound 1c. 1H NMR (400 MHz, CD3OD) δ 7.66 (2 H, dd, J )
8.7, 5.4 Hz), 7.53 (1 H, d, J ) 2.0 Hz), 7.16 (2 H, t, J ) 8.8 Hz),
6.59 (1 H, d, J ) 16.2 Hz), 6.36 (1 H, d, J ) 2.0 Hz), 5.60 (1 H,

3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9 2731
dd, J ) 16.0, 6.6 Hz), 4.35 (1 H, q, J ) 6.4 Hz), 3.85-3.93 (1 H,
m), 3.71 (3 H, s), 3.31 (3 H, s), 3.51-3.44 (1 H, m), 2.32 (1 H, dd,
J ) 15.0, 5.0 Hz), 2.25 (1 H, dd, J ) 15.0, 8.0 Hz), 1.63-1.70 (1
H, m), 1.47-1.55 (1 H, m), 1.19 (6 H, d, J ) 6.7 Hz); MS (ESI)
m/z 548 (M + H)+; analytical HPLC (system B) tR ) 1.42 min.
Sodium (3R,5S,E)-7-(2-(3-tert-butyl-1-methyl-1H-pyrazol-5-
ylamino)-4-(4-fluorophenyl)-6-isopropylpyrimidin-5-yl)-3,5-dihy-
droxyhept-6-enoate (11a). This compound was prepared from
compound 5 and 3-tert-butyl-1-methyl-1H-pyrazol-5-amine as
described for compound 1a. 1H NMR (400 MHz, CD3OD) δ 7.63
(2 H, dd, J ) 8.8, 5.3 Hz), 7.15 (2 H, t, J ) 8.6 Hz), 6.54 (1 H, d,
J ) 15.6 Hz), 6.31 (1 H, s), 5.46 (1 H, dd, J ) 16.1, 6.5 Hz), 4.32
(1 H, q, J ) 6.5 Hz), 3.83-4.00 (1 H, m), 3.71 (3 H, s), 3.36-3.51
(1 H, m), 2.31 (1 H, dd, J ) 15.1, 4.5 Hz), 2.23 (1 H, dd, J )
15.6, 7.6 Hz), 1.54-1.73 (1 H, m), 1.40-1.54 (1 H, m), 1.29 (9
H, s), 1.24 (6 H, d, J ) 6.5 Hz); MS (ESI) m/z 526 (M + H)+;
analytical HPLC (system C) tR ) 2.92 min.
7.15 (2 H, t, J ) 8.81 Hz), 6.56 (1 H, dd, J ) 16.11, 1.51 Hz),
5.51 (1 H, dd, J ) 16.11, 6.04 Hz), 4.26-4.37 (1 H, m), 3.86-3.94
(1 H, m), 3.84 (3 H, s), 3.63 (3 H, s), 3.39-3.50 (1 H, m),
2.15-2.36 (2 H, m), 1.56-1.69 (1 H, m), 1.41-1.53 (1 H, m),
1.17 (6 H, d, J ) 7.05 Hz); MS (ESI) m/z 486 (M + H)+; analytical
HPLC (system A) tR ) 2.84 min, (system B) tR ) 1.45 min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(N-(1-
methyl-1H-tetrazol-5-yl)methylsulfonamido)pyrimidin-5-yl)-3,5-di-
hydroxyhept-6-enoate (13c). This compound was prepared from
compound 5 and 1-methyl-1H-tetrazol-5-amine as described for
compound 1c. 1H NMR (400 MHz, CD3OD) δ 7.77 (2 H, dd, J )
9.1, 5.5 Hz), 7.30 (2 H, t, J ) 8.8 Hz), 6.73 (1 H, d, J ) 16.1 Hz),
5.74 (1 H, dd, J ) 16.1, 6.0 Hz), 4.48 (1 H, q, J ) 6.5 Hz), 4.24
(3 H, s), 3.97-4.12 (1 H, m), 3.94 (3 H, s), 3.52-3.68 (1 H, m),
2.44 (1 H, dd, J ) 15.1, 5.0 Hz), 2.36 (1 H, dd, J ) 15.6, 8.1 Hz),
1.70-1.86 (1 H, m), 1.49-1.69 (1 H, m), 1.30 (3 H, d, J ) 6.5
Hz), 1.29 (3 H, d, J ) 6.5 Hz); MS (ESI) m/z 550 (M + H)+;
analytical HPLC (system B) tR ) 1.39 min.
Sodium (3R,5S,E)-7-(2-(3-carboxy-1-methyl-1H-pyrazol-5-ylami-
no)-4-(4-fluorophenyl)-6-isopropylpyrimidin-5-yl)-3,5-dihydroxy-
hept-6-enoate (14a). This compound was prepared as described for
compound 1 (via ethyl 5-amino-1-methyl-1H-pyrazole-3-carboxy-
late51). 1H NMR (400 MHz, CD3OD) δ 7.61 (2 H, dd, J ) 8.8, 5.3
Hz), 7.13 (2 H, t, J ) 8.8 Hz), 6.65 (1 H, s), 6.53 (1 H, d, J ) 16.1
Hz), 5.45 (1 H, dd, J ) 16.1, 6.5 Hz), 4.31 (1 H, q, J ) 6.0 Hz),
3.85-4.00 (1 H, m), 3.77 (3 H, s), 3.37-3.46 (1 H, m), 2.31 (1 H,
dd, J ) 15.1, 4.5 Hz), 2.23 (1 H, dd, J ) 15.1, 7.1 Hz), 1.58-1.68
(1 H, m), 1.42-1.51 (1 H, m), 1.22 (6 H, d, J ) 6.5 Hz); MS
(ESI) m/z 514 (M + H)+; analytical HPLC (system B) tR ) 1.35
min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(1-methyl-
1H-pyrazol-3-ylamino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-
enoate (12a). This compound was prepared from compound 5 and
1
1-methyl-1H-pyrazol-3-amine as described for compound 1a. H
NMR (400 MHz, CD3OD) δ 7.64 (2 H, dd, J ) 8.6, 5.5 Hz), 7.43
(1 H, d, J ) 2.5 Hz), 7.16 (2 H, t, J ) 8.8 Hz), 6.80 (1 H, d, J )
2.5 Hz), 6.53 (1 H, dd, J ) 16.1, 1.0 Hz), 5.45 (1 H, dd, J ) 15.9,
6.3 Hz), 4.31 (1 H, q, J ) 6.5 Hz), 3.82-4.04 (1 H, m), 3.78 (3 H,
s), 3.37-3.52 (1 H, m), 2.32 (1 H, dd, J ) 15.6, 5.0 Hz), 2.23 (1
H, dd, J ) 15.1, 7.6 Hz), 1.53-1.74 (1 H, m), 1.35-1.55 (1 H,
m), 1.28 (6 H, d, J ) 7.1 Hz); MS (ESI) m/z 470 (M + H)+;
analytical HPLC (system C) tR ) 2.57 min, (system B) tR ) 1.46
min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(methyl(1-
methyl-1H-pyrazol-3-yl)amino)pyrimidin-5-yl)-3,5-dihydroxyhept-
6-enoate (12b). This compound was prepared from compound 5
and 1-methyl-1H-pyrazol-3-amine as described for compound 1b.
1H NMR (400 MHz, CD3OD) δ 7.65 (2 H, dd, J ) 8.8, 5.5 Hz),
6.35 (1 H, d, J ) 2.2 Hz), 7.15 (2 H, t, J ) 8.9 Hz), 6.75 (1 H, d,
J ) 2.2 Hz), 6.53 (1 H, d, J ) 16.0 Hz), 5.43 (1 H, dd, J ) 16.0,
6.0 Hz), 4.33 (1 H, q, J ) 6.0 Hz), 3.95-3.87 (1 H, m), 3.83 (3 H,
s), 3.64 (3H, s), 3.35-3.49 (1 H, m), 2.33 (1 H, dd, J ) 16.0, 5.0
Hz), 2.25 (1 H, dd, J ) 16.0, 7.5 Hz), 1.56-1.68 (1 H, m),
1.36-1.48 (1 H, m), 1.26 (6 H, d, J ) 6.7); MS (ESI) m/z 484 (M
+ H)+; analytical HPLC (system B) tR ) 1.36 min.
Acute Cholesterol Synthesis Assay in Rat. Inhibition of acute
cholesterol synthesis in rats was determined essentially as previously
described.52 Cholesterol synthesis, as measured by incorporation
of 14C-acetate into hepatic sterols, was measured during a 4 h period
spanning the peak of the mid-dark cycle in which HMG-CoA
reductase levels are highest. Male Sprague–Dawley rats (Charles
River) between 180 and 220 g were housed in a reverse light cycled
room and were fed Purina rat chow (#5001) ad libitum for 7–10
days prior to use. Compounds, suspended in 0.5% carboxymeth-
ylcellulose (Sigma) in water, were dosed by oral gavage 2.5 h prior
to the peak of the mid-dark cycle. Rats were injected intraperito-
neally with (1-14C)-sodium acetate (1–3 mCi/mmol) 37.5 uCi/100
g body weight 30 min after receiving the oral dose. After 4 h, each
group of 5 rats was sacrificed in a CO2 chamber followed by
immediate cervical dislocation and exsanguination by severing of
the abdominal aorta. Excess blood was blotted, and a 1.0 g sample
was removed from the left lobe of the liver and frozen at -20 °C
in a borosilicate glass tube for subsequent analysis.
Lipids were extracted from livers by a modification of a
previously described method.53 Liver samples were mixed with 1.5
mL of 60% KOH in H2O and 3 mL of methanol and heated at 85
°C in a water bath for 2 h, with mixing every half-hour. After
cooling, 0.02 µCi (44 000 dpm) [1R, 2R (n)-3H] cholesterol (1 µCi/
µL, 39.0 Ci/mmol, Amersham Pharmacia Biotech) diluted with
absolute ethanol from the stock was added to each sample as an
internal control to monitor extraction efficiency. All samples, except
1 blank used for a control, were extracted with 7 mL of petroleum
ether and centrifuged at 500g for 5 min at room temperature. The
organic phase (4 mL) was collected into a scintillation vial. The
aqueous phase was re-extracted with 3 mL of petroleum ether, and
the organic phase (3 mL) was collected and pooled with the organic
phase collected with the first extraction. Extracts were dried under
nitrogen and resuspended in 0.5 mL of chloroform-methanol (2:
1). Samples were then mixed with 10 mL of Opti-fluor scintillation
fluid (Packard BioScience) and counted for both 3H and 14C using
a Beckman LS3801 counter. This extraction procedure resulted in
50–90% recovery of the added 3H-cholesterol internal standard.
Percent inhibition of cholesterol synthesis was then determined by
comparing average 14C value from drug-treated animal groups with
vehicle groups. Percent inhibition of cholesterol synthesis was
plotted relative to the drug dose (mg/kg of animal), and an ED50
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(N-(1-
methyl-1H-pyrazol-3-yl)methylsulfonamido)pyrimidin-5-yl)-3,5-di-
hydroxyhept-6-enoate (12c). This compound was prepared from
compound 5 and 1-methyl-1H-pyrazol-3-amine as described for
1
compound 1c. H NMR (400 MHz, CD3OD) δ 7.58-7.65 (3 H,
m), 7.14 (2 H, t, J ) 8.8 Hz), 6.60 (1 H, d, J ) 16.0 Hz), 6.39 (1
H, d, J ) 2.0 Hz), 5.55 (1 H, dd, J ) 16.0, 6.0 Hz), 4.34 (1 H, q,
J ) 6.0 Hz), 3.90 – 3.98 (1 H, m), 3.88 (3 H, s), 3.65 (3H, s),
3.43-3.50 (1 H, m), 2.32 (1 H, dd, J ) 16.0, 5.0 Hz), 2.24 (1 H,
dd, J ) 16.0, 7.5 Hz), 1.57-1.66 (1 H, m), 1.46-1.54 (1 H, m),
1.21 (6 H, d, J ) 6.6 Hz); MS (ESI) m/z 548 (M + H)+; analytical
HPLC (system B) tR ) 1.30 min.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(1-methyl-
1H-tetrazol-5-ylamino)pyrimidin-5-yl)-3,5-dihydroxyhept-6-
enoate (13a). This compound was prepared from compound 5 and
1
1-methyl-1H-tetrazol-5-amine as described for compound 1a. H
NMR (400 MHz, CD3OD) δ 7.63 (2 H, dd, J ) 8.6, 5.5 Hz), 7.15
(2 H, t, J ) 8.8 Hz), 6.55 (1 H, dd, J ) 16.1, 1.0 Hz), 5.50 (1 H,
dd, J ) 16.1, 6.0 Hz), 4.32 (1 H, q, J ) 6.5 Hz), 3.94 (3 H, s),
3.85-3.93 (1 H, m), 3.37-3.51 (1 H, m), 2.31 (1 H, dd, J ) 15.1,
5.0 Hz), 2.23 (1 H, dd, J ) 15.1, 7.6 Hz), 1.55-1.68 (1 H, m),
1.34-1.52 (1 H, m), 1.21 (6 H, d, J ) 6.5 Hz); MS (ESI) m/z 472
(M + H)+; analytical HPLC (system B) tR ) 1.27 min. Anal.
(C21H26FN7NaO4 ·1.60H2O) C, H, N, F, Na.
Sodium (3R,5S,E)-7-(4-(4-fluorophenyl)-6-isopropyl-2-(methyl(1-
methyl-1H-tetrazol-5-yl)amino)pyrimidin-5-yl)-3,5-dihydroxyhept-
6-enoate (13b). This compound was prepared from compound 5
and 1-methyl-1H-tetrazol-5-amine as described for compound 1b.
1H NMR (400 MHz, CD3OD) δ 7.61 (2 H, dd, J ) 9.06, 5.54 Hz),

2732 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9
Ahmad et al.
heterozygous familial hypercholesterolemia. Am. J. Cardiol. 2003, 92,
1287–1293.
(16) Thompson, P. D.; Clarkson, P.; Karas, R. H. Statin-associated
value (level of drug required to suppress cholesterol synthesis in
vivo by 50%) was calculated by best-fit analysis.
Weanling Rat Model of Myotoxicity. Female Sprague–Dawley
rats (3 weeks, Charles River) were housed in a reverse light cycled
room with 12 h dark/12 h light and fed Purina rat chow ad libitum.
After 5 days of reverse light cycling, animals were weighed and
dosed daily 2 h prior to mid-dark point of the light cycle with
compounds suspended in 0.5% carboxymethylcellulose in water
administered by oral gavage for 10 days. Animals were sacrificed
by CO2 asphyxiation 1 h after the last dose, blood was collected
from the vena cava in 0.05% EDTA, and animals were perfused
with 0.9% saline. Samples of skeletal muscle from the diaphragm,
quadriceps, femoris, trapezius, and triceps brachii were collected
in 10% neutral buffered formalin. Paraffin-embedded tissue sections
(5 µm) were prepared, stained with hematoxylin and eosin, and
subjected to histopathologic evaluation. Plasma levels of CK activity
were measured using an autoanalyzer (Hitachi 912, Roche Diag-
nostics). A myotoxic signal was defined as dose-dependent CK
elevations 2.5-fold above the mean value for the vehicle group or
histopathologic evidence of myofiber necrosis in multiple segments
of muscle tissue.
myopathy. JAMA 2003, 289, 1681–1690.
(17) Ballantyne, C. M.; Corsini, A.; Davidson, M. H.; Holdaas, H.;
Jacobson, T. A.; Leitersdorf, E.; Marz, W.; Reckless, J. P.; Stein, E. A.
Risk for myopathy with statin therapy in high-risk patients. Arch.
Intern. Med. 2003, 163, 553–564.
(18) Hamilton-Craig, I. Statin-associated myopathy. Med. J. Aust. 2001,
175, 486–489.
(19) Omar, M. A.; Wilson, J. P.; Cox, T. S. Rhabdomyolysis and HMG-
CoA reductase inhibitors. Ann Pharmacother 2001, 35, 1096–1107.
(20) Arora, R.; Liebo, M.; Maldonado, F. Statin-induced myopathy: the
two faces of Janus. J. CardioVasc. Pharmacol. Ther. 2006, 11, 105–
112.
(21) Chapman, M. J.; Carrie, A. Mechanisms of statin-induced myopathy:
a role for the ubiquitin-proteasome pathway. Arterioscler. Thromb.
Vasc. Biol. 2005, 25, 2441–2444.
(22) Flint, O. P.; Masters, B. A.; Gregg, R. E.; Durham, S. K. HMG CoA
reductase inhibitor-induced myotoxicity: pravastatin and lovastatin
inhibit the geranylgeranylation of low-molecular-weight proteins in
neonatal rat muscle cell culture. Toxicol. Appl. Pharmacol. 1997, 145,
99–110.
(23) Nezasa, K.; Higaki, K.; Takeuchi, M.; Nakano, M.; Koike, M. Uptake
of rosuvastatin by isolated rat hepatocytes: comparison with pravas-
tatin. Xenobiotica 2003, 33, 379–388.
Acknowledgment. We greatly appreciate the support of
Bristol-Myers Squibb Department of Discovery Analytical
Sciences.
(24) Schachter, M. Chemical, pharmacokinetic and pharmacodynamic
properties of statins: an update. Fundam. Clin. Pharmacol. 2005, 19,
117–125.
(25) Recently, researchers from Pfizer have also reported on their efforts
to find efficacious and less myotoxic statins. (a) Pfefferkorn, J. A.;
Choi, C.; Larsen, S. D.; Auerbach, B.; Hutchings, R.; Park, W.; Askew,
V.; Dillon, L.; Hanselman, J. C.; Lin, Z.; Lu, G. H.; Robertson, A.;
Sekerke, C.; Harris, M. S.; Pavlovsky, A.; Bainbridge, G.; Caspers,
N.; Kowala, M.; Tait, B. D. Substituted pyrazoles as hepatoselective
HMG-CoA reductase inhibitors: discovery of (3R,5R)-7-[2-(4-fluoro-
phenyl)-4-isopropyl-5-(4-methyl-benzylcarbamoyl)-2H-pyrazol-3-yl]-
3,5-dihydroxyheptanoic acid (PF-3052334) as a candidate for the
treatment of hypercholesterolemia. J. Med. Chem. 2008, 51, 31–45.
(b) Pfefferkorn, J. A.; Song, Y.; Sun, K.-L.; Miller, S. R.; Trivedi,
B. K.; Choi, C.; Sorenson, R. J.; Bratton, L. D.; Unangst, P. U.; Larsen,
S. D.; et al. Design and synthesis of hepatoselective, pyrrole-based
HMG-CoA reductase inhibitors. Bioorg. Med. Chem. Lett. 2007, 17,
4538–4544.
(26) Istvan, E. S.; Deisenhofer, J. Structural mechanism for statin inhibition
of HMG-CoA reductase. Science 2001, 292 (5519), 1160–1164.
(27) Compound 1b was built without the dihydroxy acid tail from the
rosuvastatin ligand crystal structure as a starting point and was
quantum-mechanically minimized with DFT-B3LYP(6-31G*) using
Jaguar (Maestro v. 80110, Schrodinger LLC, 2007). The resulting
structure was attached to the dihydroxy acid tail from rosuvastatin,
and the compound was docked rigidly into the rosuvastatin
HMGR structure using Glide (Maestro v. 80110, Schrodinger LLC,
2007) with vdW radii of nonpolar receptor atoms scaled by 0.9,
all other parameters default, to produce the structure shown in
Figure 2.
References
(1) Third Report of the National Cholesterol Education Program (NCEP)
Expert Panel on Detection, Evaluation, and Treatment of High Blood
Cholesterol in Adults (Adult Treatment Panel III) final report.
Circulation 2002, 106, 3143–3421.
(2) Ross, S. D.; Allen, I. E.; Connelly, J. E.; Korenblat, B. M.; Smith,
M. E.; Bishop, D.; Luo, D. Clinical outcomes in statin treatment trials:
a meta-analysis. Arch. Intern. Med. 1999, 159, 1793–1802.
(3) Brown, M. S.; Goldstein, J. L. A receptor-mediated pathway for
cholesterol homeostasis. Science 1986, 232 (4746), 34–47.
(4) Horton, J. D.; Goldstein, J. L.; Brown, M. S. SREBPs: activators of
the complete program of cholesterol and fatty acid synthesis in the
liver. J. Clin. InVest. 2002, 109, 1125–1131.
(5) Willard, A. K.; Smith, R. L.; Hoffman, W. F. 6(R)-[2-(8-Acyloxy-2-
methyl-6-methyl (or hydrogen)-polyhydro-1-naphthyl)ethyl]-4(R)-hy-
droxy-3,4,5,6-tetrahydro-2H-pyran-2-ones, the hydroxy acid forms of
these pyranones, salts and esters thereof, and a pharmaceutical
antihypercholesterolemic composition containing them Eur. Pat. Appl.
, EP 33538, 1981.
(6) Tsujita, Y.; Kuroda, M.; Shimada, Y.; Tanzawa, K.; Arai, M.; Kaneko,
I.; Tanaka, M.; Masuda, H.; Tarumi, C.; Watanabe, Y.; et al. CS-514,
a competitive inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A
reductase: tissue-selective inhibition of sterol synthesis and hypolipi-
demic effect on various animal species. Biochim. Biophys. Acta 1986,
877, 50–60.
(28) Watanabe, M.; Koike, H.; Ishiba, T.; Okada, T.; Seo, S.; Hirai, K.
Synthesis and biological activity of methanesulfonamide pyrimidine-
and N-methanesulfonyl pyrrole-substituted 3,5-dihydroxy-6-hep-
tenoates, a novel series of HMG-CoA reductase inhibitors. Bioorg.
Med. Chem. 1997, 5, 437–444.
(29) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A. A
stereoselective synthesis of trans-1,2-disubstituted alkenes based on
the condensation of aldehydes with metalated 1-phenyl-1H-tetrazol-
5-yl sulfones. Synlett 1998, 1, 26–28.
(7) Endo, A.; Monacolin, K. A new hypocholesterolemic agent produced
by a Monascus species. J. Antibiot. (Tokyo) 1979, 32, 852–854.
(8) Kathawala, F. G. Analogs of mevalolactone and derivatives thereof
and their use as pharmaceuticals. PCT Int. Appl. , WO 8402131, 1984.
(9) Jones, P.; Kafonek, S.; Laurora, I.; Hunninghake, D. Comparative dose
efficacy study of atorvastatin versus simvastatin, pravastatin, lovastatin,
and fluvastatin in patients with hypercholesterolemia (the CURVES
study). Am. J. Cardiol. 1998, 81, 582–587.
(10) Roth, B. D. Preparation of trans-6-[(carbamoylpyrrolyl)alkyl)]-4-
hydroxypyranones as hypocholesterolemics. US Patent 4,681,893,
1987.
(30) Julia, M.; Paris, J. M. Syntheses with the help of sulfones V. General
method of synthesis of double bonds. Tetrahedron Lett. 1973, 49,
4833–4836.
(11) Angerbauer, R.; Fey, P.; Huebsch, W.; Philipps, T.; Bischoff, H.;
Petzinna, D.; Schmidt, D. Preparation of 7-[4-(4-fluorophenyl)-2,6-
diisopropyl-5-methoxymethylpyrid-3-yl]-3,5-dihydroxy-6-hep-
tenoate isomers as HMG-CoA reductase inhibitors. Ger. Offen, DE
4040026, 1992.
(12) Hirai, K.; Ishiba, T.; Koike, H.; Watanabe, M. Pyrimidine derivatives
as HMG-CoA reductase inhibitors. Eur. Pat. Appl. , EP 521471, 1993.
(13) Blasetto, J. W.; Stein, E. A.; Brown, W. V.; Chitra, R.; Raza, A.
Efficacy of rosuvastatin compared with other statins at selected starting
doses in hypercholesterolemic patients and in special population
groups. Am. J. Cardiol 2003, 91 (5A), 3C–10C; discussion 10C.
(14) Brousseau, M. E. Statins, super-statins and cholesterol absorption
inhibitors. IDrugs 2003, 6, 458–463.
(31) Brodfuehrer, P. R.; Sattelberg, T. R.; Kant, J.;Qian, X. Process for
preparing chiral diol sulfones and dihydroxy acid HMG CoA reductase
inhibitors. PCT Int. Appl. , WO 2002098854.
(32) Madsen, C. S.; Janovitz, E.; Cuff, C. C.; Zhang, R.; Nguyen-Tran,
N.; Ryan, C. S.; Yin, X.; Monshizadegan, H.; Chang, M.; D’Arienzio,
C.; Scheer, S.; Setters, R.; Search, D.; Chen, X.; Zhuang, S.; Harrity,
T.; Apedo, A.; Huang, C.; Kowala, M.; Blanar, M. A.; Sun, C.-Q.;
Robl, J. A.; Stein, P. D. The guinea pig as a preclinical model for
demonstrating the efficacy and safety of statins. J. Pharmacol. Exp.
Ther. 2008, 324, 576–586.
(33) Zhao, Y.; Jona, J.; Chow, D. T.; Rong, H.; Semin, D.; Xia, X.; Zanon,
R.; Spancake, C.; Maliski, E. High-throughput logP measurement using
parallel liquid chromatography/ultraviolet/mass spectrometry and
sample-pooling. Rapid Commun. Mass. Spectrom. 2002, 16, 1548–
1555.
(15) Stein, E. A.; Strutt, K.; Southworth, H.; Diggle, P. J.; Miller, E.
Comparison of rosuvastatin versus atorvastatin in patients with

3-Hydroxy-3-methylglutaryl Coenzyme-A Reductase Inhibitor
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 9 2733
(34) Davidson, M. H. Rosuvastatin: a highly efficacious statin for the
treatment of dyslipidaemia. Expert Opin. InVestig. Drugs 2002, 11,
125–141.
(35) Edwards, P. A.; Lemongello, D.; Fogelman, A. M. Improved methods
for the solubilization and assay of hepatic 3-hydroxy-3-methylglutaryl
coenzyme A reductase. J. Lipid Res. 1979, 20, 40–46.
(46) The correlation of in vitro cell selectivity and hydrophilicity described
above has certain limitations. The fact that the slightly more lipophilic
simvastatin (log P ) 4.8) has a better safety profile in the clinic than
cerivastatin (log P ) 4.5) suggests that factors beyond lipophilicity
are likely to play critical roles for statin-induced myotoxicity. Such
factors may include relative potencies, differences in metabolism, and
differences in peripheral exposure (Evans, M. Rees, A. Effects of
HMG-CoA reductase inhibitors on skeletal muscle: Are all statins the
same? Drug Safety 2002, 25, 649–663. ). Simvastatin is dosed as a
lactone prodrug and is reported to have higher levels of liver extraction
in rat (Nezasa, K.-I. Higaki, K. Matsumura, T. Inazawa, K. Hasegawa,
H. Nakano, M. Koike, M. Liver-specific distribution of rosuvastatin
in rats: Comparison with pravastatin and simvastatin. Drug Metab.
Dispos. 2002, 30, 1158–1163. ) and human (Igel, M. Sudhop, T.
Bergmann, K. V. Metabolism and drug interactions of 3-hydroxy-3-
methylglutaryl coenzyme A-reductase inhibitors (statins). Eur. J. Clin.
Pharmacol. 2001, 57, 357–364. ) compared to other marketed statins,
thus minimizing peripheral exposure. In addition to hepatic clearance,
drug metabolism may also play an important role in statin-induced
myopathy. Atorvastatin is dosed as a relatively lipophilic parent drug
that is metabolized by cytochrome P450 3A4 into equally active more
polar hydroxy metabolites (Lea, A. P. Mctavish, D. Atorvastatin a
review of its pharmacology and therapeutic potential in the manage-
ment of hyperlipidemias. Drugs 1997, 53, 828–847). It has been
hypothesized that the more polar, active metabolites, which are less
likely to penetrate the myocyte cell membrane, are responsible for
the relatively low incidence of myopathy observed in the clinic
(Wierzbicki, A. S. Atorvastatin. Exp. Opin. Pharmacother. 2001, 2,
819–830).
(47) Igel, M.; Sudhop, T.; von Bergmann, K. Pharmacology of 3-hydroxy-
3-methylglutaryl-coenzyme A reductase inhibitors (statins), including
rosuvastatin and pitavastatin. J. Clin. Pharmacol. 2002, 42, 835–845.
(48) Li, T.; Reeves, R. Placebo-controlled, ascending single-dose study to
evaluate the safety, pharmacokinetics, and pharmacodynamics of BMS-
644950 in healthy subjects. Internal Report MB106001; Department
of Clinical Discovery, Bristol-Myers Squibb, Research & Develop-
ment.
(49) Martin, P. D.; Mitchell, P. D.; Schneck, D. W. Pharmacodynamic
effects and pharmacokinetics of a new HMG-CoA reductase inhibitor,
rosuvastatin, after morning or evening administration in healthy
volunteers. Br. J. Clin. Pharmacol. 2002, 54, 472–477.
(36) Capuzzi, D. M.; Margolis, S. Metabolic studies in isolated rat liver
cells. I. Lipid synthesis. Lipids 1971, 6, 601–608.
(37) Conde, K.; Pineda, G.; Newton, R. S.; Fernandez, M. L. Hypocho-
lesterolemic effects of 3-hydroxy-3-methylglutaryl coenzyme A
(HMG-CoA) reductase inhibitors in the guinea pig: atorvastatin versus
simvastatin. Biochem. Pharmacol. 1999, 58, 1209–1219.
(38) Suzuki, H.; Aoki, T.; Tamaki, T.; Sato, F.; Kitahara, M.; Saito, Y.
Hypolipidemic effect of NK-104, a potent HMG-CoA reductase
inhibitor, in guinea pigs. Atherosclerosis 1999, 146, 259–270.
(39) Guinea pig, similar to human up-regulates LDL receptor in response
to statin treatment: Conde, K.; Vergara-Jimenez, M.; Krause, B. R.;
Newton, R. S.; Fernandez, M. L. Hypocholesterolemic actions of
atorvastatin are associated with alterations on hepatic cholesterol
metabolism and lipoprotein composition in the guinea pig. J. Lipid
Res. 1996, 37, 2372–2382.
(40) Measurement of CK levels from a second dose-response study of 1b
(n ) 8 animals/group) in guinea pigs revealed one incident (>2.5×)
within the lowest dose group (10 mg/kg) with a value of 1104 U/L.
This value also represented the highest plasma CK value for any of
the 1b-treated animals. With the exception of this isolated incident,
no other instances (>2.5×) were observed at any dose up to 100 mg/
kg. The mean CK levels for the 0, 10, 25, 50, 75, and 100 mg/kg
doses were 366.6, 456.8, 420.8, 350.8, 429.0, and 432.3, respectively.
None of the values were determined to be statistically significant (p
< 0.05). The clear lack of a dose-dependent increase in CK strongly
suggests that the isolated incident identified within the 10 mg/kg dose
group was an anomaly and most likely not the result of drug treatment.
Unusually high CK levels were also observed for one animal in the
control group for the atorvastatin study. It is possible that elevated
CK level in this particular animal resulted from trauma, possibly due
to fighting or excessive struggling during dosing. Histopathologic
analyses of muscle tissues obtained from the 1b-treated animals
revealed that there were no myopathic lesions for any group.
(41) Armitage, J. The safety of statins in clinical practice. Lancet 2007,
370, 1781–1790.
(50) Kroger, C.; Schoknecht, G.; Beyer, H. 1,2,4-Triazoles. VI. Chem. Ber.
1964, 97 (2), 396–404.
(51) Vanotti, F.; Villa, M. Synthesis of novel derivatives of 1H-imida-
zol[1,2-b]pyrazole as potential CNS-agents. J. Heterocycl. Chem. 1994,
31, 737–743.
(42) Kornbrust, D. J.; MacDonald, J. S.; Peter, C. P.; Duchai, D. M.; Stubbs,
R. J.; Germershausen, J. I.; Alberts, A. W. Toxicity of the HMG-
coenzyme A reductase inhibitor, lovastatin, to rabbits. J. Pharmacol.
Exp. Ther. 1989, 248, 498–505.
(52) Alberts, A. W.; Chen, J.; Kuron, G.; Hunt, V.; Huff, J.; Hoffman, C.;
Rothrock, J.; Lopez, M.; Joshua, H.; Harris, E.; Patchett, A.;
Monaghan, R.; Currie, S.; Stapley, E.; Albers-Schonberg, G.; Hensens,
O.; Hirshfield, J.; Hoogsteen, K.; Liesch, J.; Springer, J. Mevinolin: a
highly potent competitive inhibitor of hydroxymethylglutaryl-
coenzyme A reductase and a cholesterol-lowering agent. Proc. Natl.
Acad. Sci. U.S.A. 1980, 77, 3957–3961.
(53) Dugan, R. E.; Slakey, L. L.; Briedis, A. V.; Porter, J. W. Factors
affecting the diurnal variation in the level of 3-hydroxy-3-methylglu-
taryl coenzyme A reductase and cholesterol-synthesizing activity in
rat liver. Arch. Biochem. Biophys. 1972, 152, 21–27.
(43) Reijneveld, J. C.; Koot, R. W.; Bredman, J. J.; Joles, J. A.; Bar, P. R.
Differential effects of 3-hydroxy-3-methylglutaryl-coenzyme A re-
ductase inhibitors on the development of myopathy in young rats.
Pediatr. Res. 1996, 39, 1028–1035.
(44) Smith, P. F.; Eydelloth, R. S.; Grossman, S. J.; Stubbs, R. J.; Schwartz,
M. S.; Germershausen, J. I.; Vyas, K. P.; Kari, P. H.; MacDonald,
J. S. HMG-CoA reductase inhibitor-induced myopathy in the rat:
cyclosporine A interaction and mechanism studies. J. Pharmacol. Exp.
Ther. 1991, 257, 1225–1235.
(45) Rebuffat, P.; Belloni, A. S.; Cavallini, L.; Mazzocchi, G.; Meneghelli,
V.; Nussdorfer, G. G. Effect of mevinolin on rat hepatocytes: a
morphometric study. Exp. Pathol. 1988, 35 (3), 133–139.
JM800001N