Large Dimeric Ligands with Favorable Pharmacokinetic Properties and Peroxisome Proliferator-Activated Receptor Agonist Activity in Vitro and in Vivo

Article abstract of DOI:10.1021/jm0309046

Two potent nonselective, but PPARα-preferring, PPAR agonists 5 and 6 were designed and synthesized in high yields. The concept of dimeric ligands in transcription factors was investigated by synthesizing and testing the corresponding dimers 7, 8a, and 8b in PPAR transactivation assays. The three dimeric ligands all showed agonist activity on all three PPAR receptor subtypes, but with different profiles compared to the monomers 5 and 6. Despite breaking all the "rule of five" criteria, the dimers had excellent oral bioavailability and pharmacokinetic properties, resulting in good in vivo efficacy in db/db mice. X-ray crystal structure and modeling experiments suggested that the dimers interacted with the AF-2 helix as well as with amino acid residues in the lipophilic pocket close to the receptor surface.

Full text of DOI:10.1021/jm0309046

J . Med. Chem. 2003, 46, 4883-4894
4883
La r ge Dim er ic Liga n d s w ith F a vor a ble P h a r m a cok in etic P r op er ties a n d
P er oxisom e P r olifer a tor -Activa ted Recep tor Agon ist Activity in Vitr o a n d in
Vivo
Per Sauerberg,* Paul S. Bury, J ohn P. Mogensen, Heinz-J osef Deussen, Ingrid Pettersson, J an Fleckner,
J an Nehlin, Klaus S. Frederiksen, Tatjana Albrektsen, Nanni Din, L. Anders Svensson, Lars Ynddal,
Erik M. Wulff, and Lone J eppesen
Per Sauerberg, Novo Nordisk A/ S, Novo Nordisk Park, 2760 Måløv, Denmark
Received May 22, 2003
Two potent nonselective, but PPARR-preferring, PPAR agonists 5 and 6 were designed and
synthesized in high yields. The concept of dimeric ligands in transcription factors was
investigated by synthesizing and testing the corresponding dimers 7, 8a , and 8b in PPAR
transactivation assays. The three dimeric ligands all showed agonist activity on all three PPAR
receptor subtypes, but with different profiles compared to the monomers 5 and 6. Despite
breaking all the “rule of five” criteria, the dimers had excellent oral bioavailability and
pharmacokinetic properties, resulting in good in vivo efficacy in db/db mice. X-ray crystal
structure and modeling experiments suggested that the dimers interacted with the AF-2 helix
as well as with amino acid residues in the lipophilic pocket close to the receptor surface.
In tr od u ction
sensitivity. To achieve this biological response several
dual acting PPARR,γ agonists (e.g. ragaglitazar/NNC61-
0029,7 tesaglitazar/AZ242,8 and LY4656089) have been
investigated in pre-clinical and clinical trials. Further-
more, triple PPARR,γ,δ agonists have recently attracted
interest, as this profile might further improve the
clinical effects on the diabetic dyslipidemic parameters5
and also lower the PPARγ-mediated side effects (e.g.
oedema). The latter might be achieved because a 3-10
times lower PPARγ dose was needed in rats and
monkeys to obtain the same insulin-sensitizing effect
when cotreated with a PPARδ agonist.10,11 We have
recently published on the identification of a group of
triple PPARR,γ,δ agonists12 (4) using the dual PPARR,γ
activator13 (1) as a structural template (Figure 1). The
goal of the present work was to extend the structural
knowledge of this group of PPARR,γ,δ agonists and to
investigate if the concept of dimeric ligands could be
applied to design PPAR agonists (Figure 1).
Type 2 diabetes is a polygenic and progressive meta-
bolic disorder characterized by insulin resistance, hy-
perglycaemia, hypertriglyceridaemia, and low plasma
HDL-cholesterol. Untreated type 2 diabetes leads to
several chronic diseases such as retinopathy, nephropa-
thy, neuropathy, and cardiovascular diseases such as
atherosclerosis, the latter leading to increased mortal-
ity.1 The use of peroxisome proliferator-activated recep-
tor γ (PPARγ) activators, e.g. rosiglitazone and piogli-
tazone (glitazones), in the treatment of type 2 diabetes
has been established over the last couple of years due
to the abilities of these compounds to lower blood glucose
and insulin levels and improve insulin sensitivity.2,3
Similarly, PPARR activators, e.g. fenofibrate and clofi-
brate (fibrates), have been used clinically for more than
two decades for their ability to lower plasma triglycer-
ides and moderately raise HDL-cholesterol.4 Recently,
very impressive increases in plasma HDL-cholesterol
and decreases in plasma triglycerides in obese Rhesus
monkeys treated with the PPARδ selective agonist
GW501516 have resulted in growing therapeutic inter-
est in the third PPAR receptor.5
PPARs belong to the superfamily of nuclear hormone
receptors including steroid receptor, thyroid receptor,
retinoid receptor, and others. The PPAR receptors are
ligand-dependent transcription factors, which upon
ligand binding heterodimerize with the retinoid X
receptor (RXR), recruit a coactivator, and modulate the
transcription of the target genes after binding to specific
peroxisome proliferator response elements.6
The present PPAR treatment of type 2 diabetes is still
inadequate. Neither the fibrates nor the glitazones
simultaneously lower triglycerides and increase HDL-c
as well as lower blood glucose and improve insulin
Thus, the present paper describes the design and
synthesis of novel monomeric and dimeric ligands with
full efficacy on all three PPAR receptors. Rat pharma-
cokinetic characterizations as well as in vivo evaluations
in db/db mice show that these rather large ligands have
surprisingly good in vivo properties making them pos-
sible drug candidates.
Ch em istr y
The biphenyl analogue 5 was made via a method
analogous to the earlier published procedure for com-
pounds 3 and 4.12 Commercially available 4-acetylbi-
phenyl was subjected to a Horner-Emmons reaction to
give the E-butenoate isomer 9 as the only isolated
product (Scheme 1). Diisobutyl aluminum hydride
(DIBAL-H) reduction of the ester gave the desired
alcohol 10 without reduction of the double bond. Alky-
lation of ethyl (S)-2-ethoxy-3-(4-hydroxyphenyl)pro-
panoate14 with 10 under Mitsunobu conditions, followed
* Corresponding author. Telephone:
+4544663939. e-mail: psa@novonordisk.com.
+4544434858. Fax:
10.1021/jm0309046 CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/10/2003

4884 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23
Sauerberg et al.
F igu r e 1. Graphical evolution of the project. The dual PPARR,γ agonist 1 was structurally modified to the triple PPARR,γ,δ
agonist 4.12 Further modification gave the PPARR-preferring triple PPARR,γ,δ activators 5 and 6. Applying the dimeric ligand
design concept gave the triple PPARR,γ,δ activators 7 and 8a .
Sch em e 1a
a
(a) (i) Na, EtOH; (ii) (EtO)2POCH2COOEt, room temperature. (b) 1 M DIBAL-H in toluene, THF, -70 °C. (c) Ethyl (S)-2-ethoxy-3-
(4-hydroxyphenyl)propanoate, P(Bu)3, ADDP, THF, 0 °C to room temperature. (d) (i) 1 N NaOH, EtOH, room temperature; (ii) 1 N HCl.
Sch em e 2a
a
(a) CH(OEt)3, ZnI2, gradual heating with distillation (1-3 h). (b) 2 N H2SO4, heating with distillation. (c) (i) (EtO)2POCH2COOEt,
NaOEt, toluene 1 h at room temperature; (ii) 2 N H2SO4. (d) DIBAL-H, toluene, -78°C; (ii) 4 N HCl. (e) CH3SO2Cl, DMF, 65-70 °C. (f)
Isopropyl (S)-2-ethoxy-3-(4-hydroxyphenyl)propanoate, K2CO3, KI, acetone, reflux. (g) 1 N NaOH, EtOH, 70 °C.
by hydrolysis of the ester to the carboxylic acid, gave
the desired product 5 in 52% overall yield from 4-acetyl-
biphenyl. As previously reported,13 coupling under Mit-
sunobu conditions and basic hydrolysis did not lead to
racemization of the product.
The phenylacetylene analogue 6 was made in a
slightly different way. The starting phenylacetylene
aldehyde 12 was made according to a published Organic
Synthesis procedure15 (Scheme 2). The aldehyde 12 was
then subjected to a similar Horner-Emmons reaction
to give the R,â-unsaturated ester 13.16 DIBAL-H reduc-
tion of the ester gave the alcohol17 as described above
for compound 5. However, the obtained alcohol was not
subjected to a Mitsunobu reaction, as this tended to give
lower yields and a difficult workup. Instead, the alcohol
was converted to the allylic chloride 14 in high yield
via the iminium salt generated in situ from methane-
sulfonyl chloride in DMF.18 Alkylation of the phenol

Dimeric PPAR Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23 4885
Sch em e 3a
a
(a) (i) Na, EtOH; (ii) (EtO)2POCH2COOEt. (b) 1 M DIBAL-H in toluene, THF. (c) (i) Pd(PPh3)4, 2 M Na2CO3, DME, room temperature;
(ii) 4-acetylphenylboronic acid, 65 °C to room temperature. (d) Imidazole, TBDMSCl, CH2Cl2, room temperature. (e) Ethyl (S)-2-ethoxy-
3-(4-hydroxy-phenyl)propanoate, P(Bu)3, ADDP, THF, 0 °C to room temperature. (f) 1.1 M N(Bu)4F, THF, room temperature. (g) 1 N
NaOH, EtOH.
derivative with 14 in acetone under standard conditions
gave the desired ester in high yield. Alkaline hydrolysis
of the ester gave the target molecule 6 in 40% yield over
the seven steps.
commercially available reagents 1,3- and 1,4-diiodoben-
zene and 2-penten-4-yn-1-ol, as shown in Scheme 4. The
palladium and copper(I) assisted reactions gave the
dialcohols 21a and 21b in high yields. These dialcohols,
21a and b, could, contrary to the dialcohol described
above, be reacted under Mitsunobu conditions to give
the esters of the target dimers. Standard hydrolysis
gave 8a and 8b.
The strategy for synthesizing the dimer 7 was origi-
nally very similar to the synthesis of compound 5. The
intention was to make the dialcohol (unprotected 19)
and react this with ethyl (S)-2-ethoxy-3-(4-hydroxy-
phenyl)propanoate to directly give the diester of 7. This
strategy failed, however, as the dialcohol turned out to
be unstable. The alternative and successful method is
outlined in Scheme 3. 4-Iodoacetophenone was reacted
with triethylphosphonoacetate followed by a DIBAL-H
reduction to give the intermediate 16. Again, the Hor-
ner-Emmons reaction gave 15 exclusively as the E iso-
mer, which was confirmed by NOE NMR experiments.
The iodophenyl alcohol 16 was coupled with 4-acetyl-
phenylboronic acid using Suzuki conditions to give 17.
As the dialcohol was unstable, the hydroxyl functional-
ity in 17 was protected as the tert-butyldimethylsilyl
ether (18) before the ketone was subjected to Horner-
Emmons condensation and reduction to give the mono-
protected dialcohol 19. Alkylation of ethyl (S)-2-ethoxy-
3-(4-hydroxy-phenyl)propanoate under Mitsunobu con-
ditions with the free alcohol of 19 gave 20. After fluoride
de-protection and repeated Mitsunobu coupling of the
second alcohol group, the ester of the desired dimer was
obtained. Alkaline hydrolysis of the esters gave the
target molecule 7 in moderate overall yield.
Resu lts a n d Discu ssion
We have recently published on our design of triple
PPARR,γ,δ activators starting from a dual PPARR,γ
agonist (Figure 1, compounds 1-4).12 The aim of the
present work was to extend the structural knowledge
of this type of compounds and to investigate if the
dimeric approach could be applied to PPAR ligands. As
in our previous research, the in vitro transactivation
assays with the ligand binding domains (LBDs) of each
of the three human PPAR (hPPAR) receptor subtypes
were used as the primary screening tools in that
effort. Further, to avoid misinterpretations of the
animal data due to the known possible species differ-
ences with PPARR ligands, the compounds were also
tested in the rat PPARR (rPPARR) transactivation
assay. To compare the efficacy of compounds from test
to test WY 14643, rosiglitazone and carbacyclin were
used as reference agonists in the hPPARR, hPPARγ, and
hPPARδ assays, respectively. The importance of high
and sustained plasma exposure for good in vivo PPAR
efficacy13 prompted us to subject interesting compounds
The dimeric compounds 8a and 8b were easily ob-
tained using cross-coupling conditions between the

4886 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23
Sauerberg et al.
Sch em e 4a
a
(a) (i) CuI, Pd(PPh3)4, NH(iPr)2, room temperature; (ii) 2-penten-4-yl-1-ol, 60°C. (b) isopropyl (S)-2-ethoxy-3-(4-hydroxy-phenyl)pro-
panoate, P(Bu)3, ADDP, THF, room temperature. (c) (i) 1 N NaOH, EtOH, 80 °C; (ii) 1 N HCl.
to pharmacokinetic evaluation early in the characteriza-
tion process.
designing dimeric ligands. Dimeric or bivalent ligands
have primarily been used in the design of G-protein-
coupled receptor antagonists and uptake inhibitors, but
also a few agonists have been reported. Improved
receptor subtype selectivity and/or improved potency
have been reported for some of these dimeric ligands
over their monomeric leads (for a review of this area
see S. Halazy20). Dimeric ligands are usually defined
as molecules containing two recognition units linked
covalently through either a common group or through
a spacer. We have earlier reported on the successful
design of potent and receptor subtype selective musca-
rinic agonists employing the dimeric ligand design
through a common group,21 and this approach was
consequently pursued. Several papers have demon-
strated that hydrogen bond interactions between the
AF-2 helix in the PPAR receptor protein and the
carboxylic acid/thiazolidinedione moiety in the ligand
are essential for agonist activity.22,23 Dimeric ligands
coupled through the lipophilic end of 5 and 6 giving the
bivalent molecules 7 and 8 were therefore designed
(Figure 1). Interestingly, all three compounds, 7, 8a , and
8b, were full agonists on the three PPAR receptor
subtypes, but they had different potency ratios com-
pared to 5 and 6 (Table 1). For instance, 7 and 8a were
more potent on PPARγ than on PPARR and PPARδ,
whereas 5 and 6 were most potent on PPARR. Further,
all the test compounds (5-8) had higher potency for the
hPPARR than for the rPPARR. None of the compounds
had transactivation activity on the heterodimeric part-
ner of PPAR, the hRXRR receptor (data not shown). This
demonstrates for the first time that the dimeric ap-
proach can be utilized to change the receptor subtype
selectivity of PPAR ligands in vitro.
The synthesized dimeric agonists were, however,
quite large with molecular weights above 500 g/mol,
mLogP above 4.15, cLogP above 5, the number of
rotatable bonds above 20, and polar surface areas (PSA)
above 120 Å2 (Table 4), all characteristics traditionally
indicating poor pharmacokinetic properties.24,25 Surpris-
ingly, all three dimeric ligands, but especially 7 and 8a ,
exhibited excellent pharmacokinetic properties with
extremely high maximal plasma concentrations (Cmax),
very high oral bioavailabilies (Fpo), very low plasma
clearances (CL), very low volume of distributions (Vss),
and the desired long plasma half-lives (T1/2(po), Table 2).
The reason for these excellent pharmacokinetic proper-
ties was not investigated further, but we speculate that
The triple PPARR,γ,δ agonist 4, Figure 1, (R: 77%,
EC50 ) 1.1 µM; γ: 101%, 0.3 µM; δ: 265%, 0.5 µM)12
was used as the structural lead. Replacement of one of
the biphenyl groups in 4 with a methyl group gave
compound 5, which retained full triple PPARR,γ,δ
agonist activity, although with lower PPARγ and PPARδ
potencies (Table 1). However, the rat pharmacokinetic
properties of 5 were excellent, with high plasma expo-
sure, high oral bioavailability, low clearance, and the
resulting long half-life after oral administration (Table
2). These properties resulted in a potent and efficacious
lowering of plasma triglycerides, blood glucose, and
plasma insulin levels in male db/db mice treated orally
once a day for 7 days (Table 3). The animals were dosed
for a further 2 days, whereupon an oral glucose test
(OGTT) was performed. The reduction of the blood
glucose area under curve (AUCglu) was considered to be
a more direct measure of improved insulin sensitivity.
Compound 5 also reduced the AUCglu potently and more
efficaciously than rosiglitazone and pioglitazone (Table
3). The db/db mice were considered an in vivo model
for primarily PPARγ activation since PPARR treatment
only had marginal effects (Table 3) and in the closely
related ob/ob mice model the antihyperglycemic activity
of PPARγ agonists correlated with their in vitro PPARγ
activity.19 Despite lower in vitro PPARγ potency, the
treatment effects of 5 were more pronounced than with
the two PPARγ reference compounds rosiglitazone and
pioglitazone (as well as with the PPARR compounds WY
14643 and fenofibrate). These improved effects could be
due to the good pharmacokinetic properties, the triple
PPARR,γ,δ activation or a combination of the two.
Further, replacing one of the phenyl groups in 5 with
an acetylene group gave compound 6, which was 10
times more potent on PPARR than 5, whereas the
potency and efficacy of PPARγ and PPARδ were basi-
cally unchanged (Table 1). The pharmacokinetic proper-
ties of 6 were comparable to those of 5, although the
plasma half-life was considerably shorter (Table 2). The
effects of 6 in db/db mice were also slightly less potent
and efficacious than obtained with 5, but the effects
were still more efficacious than seen with both rosigli-
tazone and pioglitazone (Table 3).
Structural information on the binding pockets of all
the PPAR receptor subtypes, all showing large binding
cavities, prompted us to investigate the possibility of

Dimeric PPAR Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23 4887
Ta ble 1. In Vitro hPPARR, rPPARR, hPPARγ, and hPPARδ Transactivation of Test and Reference Compoundsa
a
Compounds were tested in at least three separate experiments in five concentrations ranging from 0.01 to 30 µM. EC50’s were not
calculated (NC) for compounds producing transactivation lower than 25% at 30 µM. bFold activation relative to maximum activation
obtained with WY14643 (approximately 20-fold for human and approximately 75-fold for rat corresponded to 100%), with crosiglitazone
(approximately 120-fold corresponded to 100%) and with dcarbacyclin (approximately 250-fold corresponded to 100%).
the high oral bioavailability may be due to active
transporters (e.g. the Oatp in rats26) and/or enterohe-
patic recirculation, as earlier reported for PPAR ligands.13
The impressive in vivo effects in db/db mice confirmed
the tested dimeric agonists as being possible drug
candidates (Table 3 and Figures 2a and 2b).
In an attempt to explain the altered pharmacological
profiles of the dimeric ligands compared to their mono-
meric leads, a crystal of LBD-hPPARγ receptor was
soaked in a solution containing 7. The X-ray structure
showed that one-half of the ligand, the half closer to the
AF-2 helix, had well-defined electron densities, whereas

4888 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23
Sauerberg et al.
Ta ble 2. Single Dose Rat Pharmacokinetics after iv and po Administration
single dose rat pharmacokinetic
compound
5g
6,a r g
7
8a
8b
Cmax po,a ng/mL
AUC po,b (ng × min)/mL
F po,c %
CL,d mL/min/kg
Vss,e L/kg
T1/2 po,f min
9249
4570
23220
27780
7214
6974138
3732353
587145
6647520
5096888
873922
100
89
58
100
63
83
0.49
0.54
0.04
0.10
0.23
2.10
0.16
0.14
0.08
0.04
0.16
0.38
213
122
>400
>360
402
rosiglitazone
4420
182
Rats were given either a single dose iv (1.1 mg/kg) (n ) 8) or a single dose po (2.2 mg/kg) (n ) 8) of each of the test compounds. At each
of the time points (5, 15, 30, 60, 90, 120, 240, and 360 min), one animal was sacrificed and blood samples were analyzed for compound
plasma concentration. aMaximum plasma concentration after oral dosing. bEstimated area under the plasma concentration-time curve
c
e
f
g
after oral dosing. Oral bioavailability. dClearance. Volume of distribution during steady state. Oral half-life. Administered 1.74 mg/kg
iv and 3.42 mg/kg po.
Ta ble 3. In Vivo Effects in Male db/db Mice after Oral Treatment for 7-9 Days
in vivo effects in db/db mice after 7-9 days oral treatment
BG,a ED50
mg/kg
,
BG, % max
reduction
TG,b
ED50, mg/kg
TG, % max
reduction
insulin,
ED50, mg/kg
insulin, % max
reduction
AUCglu,c
ED50, mg/kg
AUCglu, % max
reduction
compound
5
0.2 ( 0.8
0.4 ( 0.7
0.2 ( 0.9
0.4 ( 0.6
0.9 ( 0.6
2.2 ( 1.0
>300
53 (2
54 ( 4
40 ( 7
55 ( 1
54 ( 10
36 ( 8
10 ( 3
35 ( 7
53 ( 3
0.3 ( 0.6
0.4 ( 0.6
0.3 ( 0.7
0.2 ( 1.1
0.1 ( 0.8
3.0 ( 0.5
>300
65 ( 1
57 ( 3
60 ( 3
56 ( 3
46 ( 4
32 ( 2
20 ( 6
58 ( 2
21 ( 5
0.2 ( 0.9
1.0 ( 1.5
0.2 ( 0.7
0.4 ( 0.6
0.6 ( 1.1
15 ( 0.9
191 ( 22
0.1 ( 1.3
15 ( 1.0
95 ( 1
92 ( 1
91 ( 2
90 ( 3
86 ( 2
45 ( 16
32 ( 16
43 ( 9
54 ( 10
0.3 ( 0.6
1.2 ( 0.7
0.4 ( 0.6
0.9 ( 0.6
0.5 ( 0.6
4.7 ( 1.0
>300
50 ( 5
45 ( 3
53 ( 5
52 ( 2
60 ( 6
13 ( 12
0 ( 6
6,a r g
7
8a
8b
WY 14643
fenofibrate
rosiglitazone
pioglitazone
0.9 ( 0.6
2.6 ( 0.2
0.1 ( 1.0
27 ( 0.6
0.8 ( 0.6
12 ( 0.8
16 ( 9
39 ( 8
Male db/db mice (n ) 6) were treated once a day by oral gavage for 9 days. Compounds 5, 6, 7, and 8a were tested at the doses 0.3, 1,
3, and 10 mg/kg/day, 8b at 0.1, 0.3, 1, and 3 mg/kg/day, WY 14643 at 1, 3, 10, and 30 mg/kg/day, fenofibrate at 10, 30, 100, and 300
mg/kg/day, rosiglitazone at 0.2, 0.6, 2, and 6 mg/kg/day, and pioglitazone at 3, 10, 30, and 100 mg/kg/day. ED50 values were calculated via
nonlinear regression using GraphPad PRISM 3.02 and are expressed as mean ( SEM. % max reduction is the maximum achieved reduction
a
b
relative to vehicle treated control group,( SEM. Nonfasting blood glucose after 7 days treatment. Nonfasting triglycerides after 7
days treatment. c Area under blood glucose time curve after OGTT on the 9th day of treatment.
Ta ble 4. Calculated Physical/Chemical Properties of Test and Reference Compounds
compound
MWa g/mol
mLogP
cLogPb
HBdonorsc
HBaccepd
rotatable bondse
PSA,f Å2
5
6
7
8a
8b
416.5
350.4
678.8
622.7
622.7
360.8
357.4
4.50
3.87
4.89
4.81
4.81
4.27
1.91
6.66
4.53
9.30
6.91
6.91
5.89
3.02
1
1
2
2
2
0
1
4
4
8
8
8
4
6
13
12
25
24
24
11
8
64.14
64.09
128.28
128.2
128.2
51.98
77.79
fenofibrate
rosiglitazone
a
b
d
Molecular weight. Calculated by Sybyl 6.6. c Number of hydrogen bond donors. Number of hydrogen bond acceptors. e Number of
rotatable bonds of nonterminal heavy (i.e., non-hydrogen) atoms. f Polar surface area, calculated by SAVOL 3.7.
derivatives13 and made the earlier reported hydrogen
bond interactions.22,23 Docking experiments of 7 into
the published structure of the heterodimer of the
hRXRR:hPPARγ LBDs respectively bound with 9-cis-
retinoic acid and GI262570, and coactivator peptides27
were performed to investigate possible interactions of
the flexible end of the ligand. To validate the predicted
binding mode, GI262570 was first docked into the
binding pocket. FlexX gave solutions with CScore equal
to five and 1.2 as the root-mean-square (rms) values to
the crystallized ligand.27 Docking of 7 gave solutions
with high CScore values and a binding mode which for
one part of the ligand occupied the same pocket as
GI262570 and for the other end of 7 a part of the binding
pocket reported to be occupied by the partial agonist
GW007228 (Figure 3). In the predicted binding mode a
hydrogen bond interaction between the carboxylic acid
group in 7 and Q273 was observed. As the docking was
performed with a rigid protein which is not a correct
description of ligand binding, interactions between the
carboxylic acid group and other amino acids in this part
F igu r e 2. Dose-related reduction of the nonfasted blood
glucose (a) and of the nonfasted plasma insulin (b) in male
db/db mice (n ) 6) treated for 7 days with 6, 7, 8a , and 8b
orally once a day. Values are expressed as mean ( SEM. *
represents P < 0.05, ** P < 0.01 using one-way ANOVA and
Dunetts multiple comparison test.
the opposite half could not be unambiguously positioned,
indicating high flexibility (Figure 3). The well-defined
part of 7 superimposed well with previously published
(S)-2-ethoxy-3-(4-substituted-phenyl)propionic
acid

Dimeric PPAR Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23 4889
is shown in Figure 5. In this alignment it can be seen
that there are differences in the amino acid sequences
between the three receptors which might affect the
interactions between the receptors and the ligand and
give rise to a change in the R/γ/δ profile for the dimeric
compounds. Due to the flexibility of the receptors, a
situation in which the carboxylic acid group introduces
different shapes for the different receptors cannot be
rejected. The difference in amino acid sequences and the
possibility of the binding pocket to adopt different
shapes for the hPPARR, hPPARγ, and hPPARδ recep-
tors are possible explanations of the change in receptor
subtype profile going from the monomeric to the dimeric
ligands.
In conclusion, the PPAR agonists 5 and 6 as well as
their dimeric analogues 7, 8a , and 8b were designed
and synthesized in good overall yields. In vitro PPARR,
γ, and δ transactivation data showed the designed
ligands to be potent PPAR agonists exhibiting different
PPAR receptor subtype profiles. X-ray crystal structure
and modeling experiments suggested that part of the
receptor subtype profile could be due to interactions
with amino acids residing close to the receptor surface.
Rat pharmacokinetic evaluations revealed surprisingly
favorable oral absorption and plasma half-lives of these
quite large bivalent ligands. The in vivo db/db mice
efficacy data confirmed that the evaluated dimeric
PPAR ligands might be suitable drug candidates, fur-
ther suggesting that the principle of dimeric ligands
could be applied more broadly than seen until now.29
F igu r e 3. Crystal structure of the ligand binding domain of
the hPPARγ receptor (green) soaked with the dimeric ligand
7 (light and dark magenta) and superimposed with the crystal
structure of the partial PPARγ agonist GW0072 (light blue).28
The light magenta part of 7 depicts the less well-defined part
of the crystal structure.
Exp er im en ta l Section
Ch em istr y. Melting points were determined on a Bu¨chi
capillary melting point apparatus and are uncorrected. 1H
NMR spectra were recorded at either 200 MHz on a Bruker
Advance DPX 200 instrument, at 300 MHz on a Bruker
Advance DRX 300 instrument, or at 400 MHz on a Bruker
Advance DRX 400 instrument, and mass spectra were recorded
on a Finnigan 5100 mass spectrometer. Capillary electro-
phoresis (CE) was run on an Agilent Technologies capillary
electrophoresis instrument. Capillary: Agilent part no. 61600-
62232, 72 cm, 50 µm i.d., bubble cell. Column chromatography
was performed on silica gel 60 (70-230 mesh, ASTM, Merck).
Elemental analyses were performed by the Novo Nordisk
Microanalytical Laboratory, Denmark, and were within (0.4%
of the calculated values.
Eth yl (E)-3-Bip h en yl-4-ylbu t-2-en oa te, 9. Sodium (6.0 g,
255 mmol) was added to ethanol 700 mL) at 20 °C and the
mixture stirred until the metal had fully reacted. Triethyl
phosphonoacetate (57.12 g, 255 mmol) was added, the mixture
stirred for 5 min, then 4-acetylbiphenyl (25.0 g, 127 mmol)
added to the stirred solution. The mixture was stirred at room
temperature for 24 h, the resulting suspension filtered, and
the crystals washed with ethanol (100 mL) to give the title
compound as white crystals in 26.6 g (79%) yield. 1H NMR
(CDCl3) δ 1.32 (3H, t, J ) 7 Hz), 2.62 (3H, d, J ) 1.5 Hz), 4.21
(2H, q, J ) 7 Hz), 6.20 (1H, d, J ) 1.5 Hz), 7.31-7.65 (9H, m).
(E)-3-Bip h en yl-4-ylbu t-2-en -1-ol, 10. A 1 M solution of
DIBAL-H in toluene (255 mL, 255 mmol) was added dropwise
at -70 °C over 1 h to a stirred solution of 9 (15.0 g, 56.3 mmol)
in dry THF (200 mL), and the mixture stirred for 90 min.
Saturated aqueous ammonium chloride was carefully added
to quench the reaction, and the resulting mixture was ex-
tracted with ethyl acetate (250 mL). The organic phases were
washed with brine, dried (Na2SO4), and evaporated. The
residue was crystallized from ethanol to give the title com-
pound as colorless crystals in 11.3 g (90%) yield. 1H NMR
(CDCl3) δ 1.40 (1H, br s), 2.12 (3H, d, J ) 1.5 Hz), 4.40 (2H,
d, J ) 6 Hz), 6.05 (1H, dt, J ) 5 Hz, 1.5 Hz), 7.35-7.70
(9H, m).
F igu r e 4. Superimposition of the ligand binding domain of
the hPPARγ receptor (green) crystallized with GI262570 (cyan)
in the hRXRR:hPPARγ:SRC-1 receptor complex27 and the four
most diverse FlexX docking solutions of the dimeric ligand 7
(magenta). The amino acids closest to the carboxylic acid in 7
(the one not interacting with the AF-2 helix) are shown as ball-
and-stick. Hydrogen bond interaction was observed between
the carboxylic acid in 7 and the NH2 group in the glutamine
Q273.
of the binding pocket cannot be ruled out. The amino
acids with hydrophilic side chains within 15 Å from the
carboxylic acid group in 7 are shown in ball-and-stick
in Figure 4, and interactions between the carboxylic acid
group in 7 and one of these amino acids could be
possible. The alignment of the amino acids in the
hPPARR, hPPARγ, and hPPARδ receptors in this pocket

4890 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23
Sauerberg et al.
F igu r e 5. Alignment of the amino acids close to the receptor surface in the hPPARR, hPPARγ, and hPPARδ receptors.
Eth yl (E)-(S)-3-[4-(3-Bip h en yl-4-ylbu t-2-en yloxy)p h e-
n yl]-2-eth oxyp r op a n oa te. To a solution of 10 (4.0 g, 17.8
mmol) and ethyl (S)-2-ethoxy-3-(4-hydroxyphenyl)propanoate14
(4.2 g, 17.8 mmol) in dry THF (500 mL) were added a solution
of tributylphosphine (7.2 g, 35.7 mmol) and azodicarboxylic
dipiperidide (9.0 g, 35.7 mmol) in dry THF (150 mL) at 0 °C.
The reaction mixture was allowed to warm to room temper-
ature and stirred for 24 h. The resulting mixture was evapo-
rated in vacuo and the residue purified by column chroma-
tography using ethyl acetate:n-heptane (1:5) as eluent to give
the title compound as an oil in 7.0 g (89%) yield. 1H NMR
(CDCl3) δ 1.17 (3H, t, J ) 7 Hz), 1.22 (3H, t, J ) 7 Hz), 2.17
(3H, d, J ) 2 Hz), 2.96 (2H, d, J ) 7 Hz), 3.29-3.37 (1H, m),
3.54-3.61 (1H, m), 3.97 (1H, t, J ) 7 Hz), 4.17 (2H, q, J )
7.5), 4.70 (2H, d, J ) 6 Hz), 6.11 (1H, m), 6.88 (2H, d, J ) 8
Hz), 7.17 (2H, d, J ) 8 Hz), 7.25-7.63 (9H, m).
(E)-(S)-3-[4-(3-Bip h en yl-4-ylb u t -2-en yloxy)p h en yl]-2-
eth oxyp r op a n oic Acid , 5. Sodium hydroxide (1 M, 33 mL,
33 mmol) was added to a solution of ethyl (E)-(S)-3-[4-(3-
biphenyl-4-ylbut-2-enyloxy)phenyl]-2-ethoxypropanoate (5.0 g,
11.0 mmol) in ethanol (60 mL) and the mixture stirred at 20
°C for 5 h. The reaction was acidified with 1 N hydrochloric
acid (31 mL), and the resulting precipitate was collected by
filtration and dried to give the title compound a white solid in
3.7 g (82%) yield. ee > 93% (CE). 1H NMR (CDCl3) δ 1.19 (3H,
t, J ) 7 Hz), 2.63 (3H, d, J ) 7 Hz), 2.93 (1H, dd, J ) 14 Hz,
7 Hz), 3.10 (1H, dd, J ) 14 Hz, 5 Hz), 3.40-3.65 (2H, m), 4.10
(2H, q, J ) 7 Hz), 4.72 (2H, d, J ) 7 Hz), 6.10 (1H, dt, J ) 5
Hz, 1 Hz), 6.90 (2H, d, J ) 8 Hz), 7.20 (2H, d, J ) 8 Hz), 7.35-
7.60 (9H, m). Mp 118-120 °C. Anal. (C27H28O4) C, H, N.
(E)-5-Ch lor op en t-3-en -1-yn ylben zen e, 14. To a solution
of (E)-5-phenylbut-2-en-4-yn-1-ol17 (53.0 g, 335 mmol) in DMF
(150 mL) was added methanesulfonyl chloride (42.2 g, 368
mmol) dropwise with stirring. The temperature rose to 60-
70 °C, and the reaction mixture was kept at that temperature
for 1 h after complete addition. The reaction mixture was
allowed to cool to ambient temperature followed by the
addition of water (150 mL). The mixture was then extracted
with toluene (2 × 100 mL), and the combined organic phases
were washed with water (2 × 100 mL). The solvent was
removed in vacuo to give 57.1 g (92%) of 14 as a colorless oil,
which was used without purification in the next step. 1H NMR
(acetone-d6) δ 4.23 (2H, dd, J ) 7.1 Hz, 1.0 Hz), 6.10 (1H, dt,
J ) 16.0 Hz, 1.0 Hz), 6.31 (1H, dt, J ) 7.1 Hz, 16.0 Hz), 7.32-
7.39 (3H, m), 7.41-7.48 (2H, m). 13C NMR 100 MHz (acetone-
d6) δ 45.43, 88.00, 92.47, 115.02, 124.19, 129.71, 129.88, 132.79,
139.71.
Isop r op yl (S)-(E)-2-Eth oxy-3-[4-(5-p h en ylp en t-2-en -4-
yn yloxy)p h en yl]p r op a n oa te. A mixture of isopropyl (S)-2-
ethoxy-3-(4-hydroxyphenyl)propanoate (47.14 g, 186.8 mmol),
14 (30.0 g, 169.8 mmol), potassium carbonate (35.21 g, 254.7
mmol), and potassium iodide (2.82 g, 17.0 mmol) in acetone
(100 mL) was refluxed for 4-5 h. The solid was filtered off
and the acetone removed in vacuo to yield an oil. The oil was
dissolved in tert-butylmethyl ether (MTBE) (100 mL) and
washed with 1 M NaOH (3 × 50 mL) and saturated NaCl
solution (50 mL). The organic phase was then dried over Na2-
SO4 and the solvent removed in vacuo after filtration to give
59.99 g (90%) of the title compound as a colorless oil, which
was used without purification in the next step. 1H NMR
(CDCl3) δ 1.15 (3H, d, J ) 7 Hz), 1.16 (3H, t, J ) 7 Hz), 1.22
(3H, d, J ) 7 Hz), 2.94 (2H, d, J ) 6.6 Hz), 3.35 (1H, m), 3.59
(1H, m), 3.94 (1H, t, J ) 6.6 Hz), 4.59 (2H, dd, J ) 5.3 Hz, 1.8
Hz), 4.98-5.07 (1H, m), 6.05 (1H, dt, J ) 15.8 Hz, 1.8 Hz),
6.36 (1H, dt, J ) 15.8 Hz, 5.3 Hz), 6.82 (2H, d, J ) 9 Hz), 7.16
(2H, d, J ) 9 Hz), 7.28-7.30 (3H, m), 7.41-7.43 (2H, m). 13C
NMR 100 MHz (CDCl3) δ 15.51, 22.23, 38.83, 66.48, 68.05,
68.75, 80.81, 87.55, 91.01, 112.66, 114.92, 123.54, 128.73,
130.12, 130.92, 131.93, 138.17, 157.50, 172.52.
(S)-(E)-2-E t h oxy-3-[4-(5-p h en ylp en t -2-en -4-yn yloxy)-
p h en yl]p r op a n ic Acid , 6. A mixture of isopropyl (S)-(E)-2-
ethoxy-3-[4-(5-phenylpent-2-en-4-ynyloxy)phenyl]propanoate
(39.25 g, 100.0 mmol), aqueous 1M NaOH (150 mL), and
ethanol (96%, 130 mL) was heated for 1-2 h to 70 °C with
stirring. Most of the ethanol was then distilled off followed by
the addition of saturated NaCl solution (150 mL). The basic
water phase was washed with MTBE (2 × 50 mL) at 50-55
°C. MTBE (50 mL) was added to the aqueous phase followed
by addition of concentrated hydrochloric acid until pH 2. The
organic phase was separated and the water phase extracted
once more with MTBE (50 mL). The combined organic phases
were dried over Na2SO4 and filtered, and the solvent remove
in vacuo to give 33.99 g (97%) of 6 as an oil, which crystallized
on standing. An analytical sample of 6 could be obtained by
crystallization as the benzylamine salt in MTBE followed by
1
liberation of the free acid. H NMR (acetone-d6) δ 1.10 (3H, t,
J ) 7 Hz), 2.90 (1H, dd, J ) 14 Hz, 8 Hz), 3.01 (1H, dd, J )
14 Hz, 4 Hz), 3.38 (1H, dq, J ) 7 Hz, 9 Hz), 3.63 (1H, dq, J )
7 Hz, 9 Hz), 4.04 (1H, dd, J ) 8 Hz, 4 Hz), 4.69 (2H, dd, J )
5.3 Hz, 1.8 Hz), 6.14 (1H, dt, J ) 16 Hz, 1.8 Hz), 6.43 (1H, dt,
J ) 16 Hz, 5.3 Hz), 6.90 (2H, d, J ) 8.5 Hz), 7.21 (2H, d, J )
8.5 Hz), 7.36-7.40 (3H, m), 7.43-7.47 (2H, m). 13C NMR 100
MHz (acetone-d6) δ 15.41, 38.77, 66.37, 68.09, 80.53, 88.03,
90.89, 112.38, 115.19, 124.01, 129.30, 129.39, 130.96, 131.32,
132.20, 139.66, 158.04, 173.38.
(S)-(E)-2-E t h oxy-3-[4-(5-p h en ylp en t -2-en -4-yn yloxy)-
p h en yl]p r op a n oic Acid L-Ar gin in e, 6-L-Ar gin in e Sa lt. (S)-
(E)-2-Ethoxy-3-[4-(5-phenylpent-2-en-4-ynyloxy)phenyl]pro-
panoic acid, 6, (35.04 g, 0.1 mol) was dissolved in 2-propanol
(260 mL). The solution was heated to reflux and a hot (∼70
°C) solution of L-arginine (17.42 g, 0.1 mol) dissolved in water
(350 mL) was added dropwise with stirring. The solution was
allowed to cool slowly to room temperature (the solution
became opaque at ∼65 °C and seeding might be necessary at
this temperature in order to ensure a proper crystallization).
The crystals formed were filtered off, washed with 2-propanol
(2 × 100 mL), and dried at 30-45 °C in vacuo to give 45.1 g
(86%) of the title compound. Mp 195 °C (DSC). ee > 96% (CE).
ESI-MS: 351 (MH+). 1H NMR (DMSO-d6) δ 0.99 (3H, t, J ) 7
Hz), 1.50-1.81 (4H, m), 2.66 (1H, dd, J ) 14 Hz, 9 Hz), 2.83
(1H, dd, J ) 14 Hz, 4 Hz), 2.95-3.63 (10H, m), 4.65 (2H, d, J
) 5.3 Hz), 6.14 (1H, d, J ) 16 Hz), 6.41 (1H, dt, J ) 16 Hz, 5.3
Hz), 6.84 (2H, d, J ) 8.5 Hz), 7.14 (2H, d, J ) 8.5 Hz), 7.36-
7.41 (3H, m), 7.42-7.48 (2H, m), 7.90-8.30 (4H, br.s). 13C NMR
100 MHz (DMSO-d6) δ 15.55, 24.88, 28.61, 38.78, 40.50, 53.76,
64.24, 67.35, 82.53, 87.86, 90.55, 111.78, 114.44, 122.68,
129.06, 130.41, 131.61, 132.53, 139.75, 156.41, 158.10, 172.52,
176.88. Anal. (C22H21O4.C6H14N4O2) C, H, N.
Eth yl (E)-3-(4-Iodoph en yl)bu t-2-en oate, 15. Sodium (5.52
g, 0.24 mol) was dissolved in ethanol (200 mL). A solution of
triethyl phosphonoacetate (62.7 g, 0.28 mol) in ethanol (100
mL) was slowly added, the mixture stirred for 20 min, and a
solution of 4-iodoacetophenone (49.21 g, 0.20 mol) in hot
ethanol (200 mL) added. The reaction mixture was stirred at
80 °C for 66 h. The mixture was cooled and the ethanol
evaporated. To the residue were added 1 N HCl (400 mL) and
ethyl acetate (400 mL). The aqueous layer was separated and
further extracted with ethyl acetate (2 × 200 mL). The
combined organic phases were washed with brine, dried
(MgSO4), filtered, and evaporated. The product was purified
by column chromatography using heptane:ethyl ether (39:1)
1
as eluent to give 58.2 g (92%) of the title compound. H NMR
(CDCl3) δ 1.31 (3H, t, J ) 7 Hz), 2.53 (3H, d, J ) 1.1 Hz), 4.21

Dimeric PPAR Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23 4891
(2H, q, J ) 7 Hz), 6.11 (1H, q, J ) 1.1 Hz), 7.19 (2H, d, J ) 8
Hz), 7.69 (2H, d, J ) 8 Hz).
THF (25 mL). The mixture was stirred for 30 min at -70 °C
followed by 2 h at room temperature. Ethanol (1 mL) was
carefully added, followed by 1N HCl (50 mL) and the resulting
mixture was extracted with ethyl acetate (2 × 50 mL). The
combined organic extracts were washed with brine, dried
(MgSO4), and evaporated to give 1.02 g (99%) of the title
(E)-3-(4-Iod op h en yl)bu t-2-en -1-ol, 16. Under an atmo-
sphere of nitrogen, 15 (10.1 g, 32.0 mmol) was dissolved in
dry THF (300 mL). The solution was cooled to -15 °C, and a
1 M solution of DIBAL-H in toluene (96.0 mL, 96.0 mmol) was
added slowly. The mixture was slowly warmed to room
temperature and stirred for 1 h. Methanol (50 mL) was
carefully added, followed by 1 N HCl (500 mL), and the
resulting mixture was extracted with ethyl acetate (3 × 500
mL). The combined organic extracts were washed with brine,
dried (MgSO4), and evaporated to give 8.8 g (100%) of the title
compound as a white crystalline solid. 1H NMR (DMSO) δ 1.38
(1H, t, J ) 5.3 Hz), 2.04 (3H, d, J ) 1.1 Hz), 4.35 (2H, t, J )
5.6 Hz), 5.93-6.00 (1H, m), 7.14 (2H, d, J ) 8.5 Hz), 7.65 (2H,
d, J ) 8.5 Hz).
1
compound. H NMR (CDCl3) δ 0.13 (6H, s), 0.96 (9H, s), 1.57
(1H, s), 2.07 (3H, s), 2.13 (3H, s), 4.37-4.46 (4H, m), 5.98 (1H,
dt, J ) 5 Hz, 1 Hz), 6.06 (1H, dt, J ) 5 Hz, 1 Hz), 7.46-7.52
(4H, m), 7.53-7.61 (4H, m).
Eth yl (E,E)-(S)-3-{4-[3-(4′-{4-[ter t-Bu tyldim eth ylsilyloxy]-
bu t -2-en -2-yl}b ip h en yl-4-yl)bu t -2-en -1-yloxy]p h en yl}-2-
eth oxyp r op a n oa te, 20. Under an atmosphere of nitrogen,
azodicarboxylic dipiperidide (0.91 g, 3.62 mmol) was added at
0-5 °C to a stirred solution of tributylphosphine (0.89 mL,
3.62 mmol), ethyl (S)-2-ethoxy-3-(4-hydroxyphenyl)propanoate
(0.60 g, 2.53 mmol) and 19 (1.02 g, 2.41 mmol) in dry THF (15
mL). The mixture was warmed to room temperature and
stirred for 18 h. The resulting mixture was diluted with water
and ethyl acetate, the aqueous layer collected and further
extracted with ethyl acetate. The organic layers were com-
bined, washed with brine, dried (MgSO4), and evaporated. The
crude product was then purified by column chromatography
using heptane:ethyl acetate (4:1) as eluent to give 1.18 g (76%)
of the title compound. 1H NMR (CDCl3) δ 0.13 (6H, s), 0.93
(9H, s), 1.18 (3H, t, J ) 7 Hz), 1.23 (3H, t, J ) 7 Hz), 2.07 (3H,
s), 2.18 (3H, s), 2.95 (2H, d, J ) 7 Hz), 3.31-3.42 (1H, m),
3.55-3.67 (1H, m), 3.98 (1H, t, J ) 7 Hz), 4.17 (2H, q, J ) 7
Hz), 4.42 (2H, d, J ) 6 Hz), 4.73 (2H, d, J ) 6 Hz), 5.95 (1H,
t, J ) 5 Hz), 6.12 (1H, t, J ) 5 Hz), 6.88 (2H, d, J ) 8 Hz),
7.18 (2H, d, J ) 8 Hz), 7.45-7.60 (8H, m).
Eth yl (E,E)-(S)-3-{4-[3-(4′-{4-Hyd r oxybu t-2-en -2-yl }-
b ip h e n yl-4-yl)b u t -2-e n -1-yloxy]p h e n yl}-2-e t h oxyp r o-
p a n oa te. A solution of 20 (1.18 g, 1.84 mmol) in dry THF was
cooled on ice, and a 1.1 M solution of tetrabutylammonium
fluoride in THF (1.93 mL, 2.12 mmol) was slowly added. The
reaction mixture was stirred at room temperature for 3 h. The
mixture was diluted with water and ethyl acetate, and the
aqueous layer was collected and further extracted with ethyl
acetate. The organic layers were combined, washed with brine,
dried (MgSO4), and evaporated to give 0.94 g (99%) of the title
compound. 1H NMR (CDCl3) δ 1.18 (3H, t, J ) 7 Hz), 1.22 (3H,
t, J ) 7 Hz), 2.12 (3H, s), 2.18 (3H, s), 2.96 (2H, d, J ) 7 Hz),
3.30-3.42 (1H, m), 3.53-3.67 (1H, m), 3.98 (1H, t, J ) 7 Hz),
4.17 (2H, q, J ) 7 Hz), 4.40 (2H, d, J ) 7 Hz), 4.74 (2H, d, J
) 7 Hz), 6.04 (1H, dt, J ) 5 Hz, 1 Hz), 6.12 (1H, dt, J ) 5 Hz,
1 Hz), 6.88 (2H, d, J ) 8 Hz), 7.18 (2H, d, J ) 8 Hz), 7.45-
7.62 (8H, m).
E t h yl (E,E)-(S,S)-3-{4-[3-(4′-{4-(4-(2-E t h oxy-2-et h oxy-
ca r b on yl-et h yl)p h en oxy)b u t -2-en -2-yl }b ip h en yl-4-yl)-
bu t-2-en -1-yloxy]p h en yl}-2-eth oxyp r op a n oa te. Under an
atmosphere of nitrogen, azodicarboxylic dipiperidide (0.50 g,
1.89 mmol) was added at 0-5 °C to a stirred solution of
tributylphosphine (0.37 mL, 1.89 mmol), ethyl (S)-2-ethoxy-
3-(4-hydroxyphenyl)propanoate (0.32 g, 1.32 mmol), and ethyl
(E,E)-(S)-3-{4-[3-(4′-{4-hydroxybut-2-en-2-yl }biphenyl-4-yl)-
but-2-en-1-yloxy]phenyl}-2-ethoxypropanoate (0.65 g, 1.26 mmol)
in dry THF (15 mL). The mixture was warmed to room
temperature and stirred for 18 h. The resulting mixture was
diluted with water and ethyl acetate, and the aqueous layer
was collected and further extracted with ethyl acetate. The
organic layers were combined, washed with brine, dried
(MgSO4), and evaporated to give 580 mg (63%) of the title
compound. 1H NMR (CDCl3) δ 1.17 (6H, t, J ) 7 Hz), 1.22 (6H,
t, J ) 7 Hz), 2.16 (6H, s), 2.97 (4H, d, J ) 7 Hz), 3.27-3.43
(2H, m), 3.52-3.69 (2H, m), 3.98 (2H, t, J ) 7 Hz), 4.17 (4H,
q, J ) 7 Hz), 4.73 (4H, d, J ) 7 Hz), 6.12 (2H, t, J ) 6 Hz),
6.88 (4H, d, J ) 8 Hz), 7.18 (4H, d, J ) 8 Hz), 7.43-7.63 (8H,
m).
(E)-1-[4′-(4-Hyd r oxy-2-bu t-2-en -2-yl)bip h en yl-4-yl]eth a -
n on e, 17. Tetrakis(triphenylphoshine)-palladium (0.46 g, 0.4
mmol, 4 mol %) was added, under nitrogen, to a stirred
solution of 16 (2.74 g, 10.0 mmol) in DME (100 mL), and the
solution stirred at room temperature for 10 min. Aqueous 2M
sodium carbonate (30 mL, 60 mmol) was then added, and the
mixture stirred for 10 min. Then 4-acetylphenylboronic acid
(3.28 g, 20.0 mmol) was added, and the reaction mixture was
heated to 65 °C for 18 h and at room temperature for another
3 days. The reaction mixture was diluted with 1N HCl (200
mL) and the products extracted into ethyl acetate (2 × 200
mL). The combined organic extracts were washed with brine,
dried (MgSO4), and evaporated to give the crude product.
Purification by column chromatography using heptane:ethyl
acetate (3:2) as eluent gave 2.0 g (75%) of the title compound
1
as a off-white solid. H NMR (CDCl3) δ 2.12 (3H, s), 2.64 (3H,
s), 4.41 (2H, d, J ) 6.8 Hz), 6.07 (1H, m), 7.54 (2H, d, J ) 8.5
Hz), 7.61 (2H, d, J ) 8.5 Hz), 7.71 (2H, d, J ) 8.5 Hz), 8.03
(2H, d, J ) 8.5 Hz).
(E)-1-{4′-[4-(ter t-Bu tyld im eth ylsilyloxy)bu t-2-en -2-yl]-
bip h en yl-4-yl}eth a n on e, 18. To a suspension of 17 (1.1 g,
4.13 mmol) in methylene chloride (40 mL) were added under
an atmosphere of nitrogen imidazole (0.42 g, 6.20 mmol) and
tert-butyldimethylsilyl chloride (0.78 g, 5.15 mmol). The
mixture was stirred at room temperature for 18 h. Methylene
chloride (15 mL) was added and the reaction mixture was
washed with water, sodium hydrogencarbonate solution and
brine. The organic phase was dried (MgSO4), filtered and
concentrated in vacuo. The residue was purified by column
chromatography using heptane:ethyl acetate (4:1) as eluent
to give 1.36 g (87%) of the title compound as a white crystalline
solid. Mp 100-106 °C. 1H NMR (CDCl3) δ 0.13 (6H, s), 0.97
(9H, s), 2.10 (3H, s), 2.65 (3H, s), 4.45 (2H, d, J ) 7 Hz), 5.98
(1H, dt, J ) 5 Hz, 1 Hz), 7.51 (2H, d, J ) 8 Hz), 7.60 (2H, d,
J ) 8 Hz) 7.69 (2H, d, J ) 8 Hz), 8.02 (2H, d, J ) 8 Hz).
Eth yl (E,E)-3-(4′-{4-[ter t-Bu tyld im eth ylsilyloxy]bu t-2-
en -2-yl}bip h en yl-4-yl)b u t -2-en oa t e. Sodium (0.42 g, 18.0
mmol) was added to ethanol (50 mL) at 20 °C and the reaction
mixture was stirred until the metal had fully reacted. Triethyl
phosphonoacetate (2.4 mL, 12.0 mmol) was added, and the
mixture was stirred for 5 min. Then 18 (1.14 g, 3.0 mmol) was
added to the stirred solution and the reaction was stirred at
room temperature for 24 h. To the reaction mixture was added
water and the product extracted with ethyl acetate (2 × 300
mL). The combined organic phases were washed with brine,
dried (MgSO4), filtered and concentrated in vacuo. The residue
was purified by column chromatography using heptane:ethyl
acetate (4:1) as eluent to give 1.13 g (81%) of the title
1
compound. H NMR (CDCl3) δ 0.12 (6H, s), 0.92 (9H, s), 1.32
(3H, t, J ) 7 Hz), 2.08 (3H, s), 2.62 (3H, s), 4.22 (2H, q, J ) 7
Hz), 4.42 (2H, d, J ) 7 Hz), 5.97 (1H, dt, J ) 5 Hz, 1 Hz), 6.20
(1H, d, J ) 1 Hz), 7.43-7.63 (8H, m).
(E,E)-3-(4′-{4-[ter t-Bu t yld im et h ylsilyloxy]b u t -2-en -2-
yl}biph en yl-4-yl)bu t-2-en -1-ol, 19. A 1M solution of DIBAL-H
in toluene (7.3 mL, 7.3 mmol) was, under an atmosphere of
nitrogen, added dropwise at -70 °C over 20 min to a stirred
solution of ethyl (E, E)-3-(4′-{4-[tert-butyldimethylsilyloxy]but-
2-en-2-yl}biphenyl-4-yl)-but-2-enoate (1.13 g, 2.43 mmol) in dry
(E,E)-(S,S)-3-{4-[3-(4′-{4-(4-(E t h oxy-2-ca r b oxyet h yl)-
p h en oxy)bu t-2-en -2-yl }bip h en yl-4-yl)-bu t-2-en -1-yloxy]-
p h en yl}-2-eth oxyp r op a n oic Acid , 7. To a solution of ethyl
(E,E)-(S,S)-3-{4-[3-(4′-{4-(4-(2-ethoxy-2-ethoxycarbonylethyl)-
phenoxy)-but-2-en-2-yl}biphenyl-4-yl)but-2-en-1-yloxy]phenyl}-

4892 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23
Sauerberg et al.
2-ethoxypropanoate (367 mg, 0.5 mmol) in ethanol (10 mL)
was added 1N sodium hydroxide (2 mL). The reaction mixture
was stirred at room temperature for 18 h and at 60 °C for 1 h.
The resulting mixture was diluted with water and ethyl
acetate, and the aqueous layer was collected and further
extracted with ethyl acetate (3 × 10 mL). The organic layers
were combined, washed with brine, dried (MgSO4) and evapo-
rated to give 180 mg (53%) of the title compound. ee > 99%
(CE). 1H NMR (CDCl3 + 1 dr. DMSO) δ 1.15 (6H, t, J ) 7 Hz),
2.90-30.80 (4H, m), 3.30-3.42 (2H, m), 3.60-3.71 (2H, m),
3.95 (2H, dd, J ) 5 Hz, 5 Hz), 4.73 (4H, d, J ) 7 Hz), 6.11
(2H, t, J ) 5 Hz), 6.88 (4H, d, J ) 8 Hz), 7.21 (4H, d, J ) 8
Hz), 7.51 (4H, d, J ) 8 Hz), 7.57 (4H, d, J ) 8 Hz). Mp 159-
162.5 °C. Anal. (C42H46O8) C, H, N.
(E,E)-5-[4-(5-Hyd r oxyp en t-3-en -1-yn yl)p h en yl]p en t-2-
en -4-yn -1-ol, 21a . To a solution of 1,4-diiodobenzene (10.57
g, 32.0 mmol) in diisopropylamine (120 mL) under a nitrogen
atmosphere were added copper(I) iodide (640 mg, 3.3 mmol)
and tetrakis(triphenylphosphine)palladium (640 mg, 0.55
mmol). After the mixture had stirred for 1 h, a solution of
2-penten-4-yn-1-ol (10.5 g, 128.1 mmol) in diisopropylamine
(60 mL) was added. After stirring under nitrogen at 60 °C for
2 h, the reaction mixture was cooled to room temperature and
filtered through Celite. The filtrate was washed with ethyl
acetate (50 mL) and evaporated to dryness. The product was
purified by column chromatography using methylene chloride:
THF (80:1) as eluent to give the crude product. The crude
product was washed with ether (2 × 100 mL) giving the title
compound in 8.35 g (100%) yield. 1H NMR (DMSO) δ 4.10 (4H,
m), 5.01 (2H, t, J ) 7 Hz), 5.95 (2H, dt, J ) 15 Hz, 2 Hz), 6.48
(2H, dt, J ) 15 Hz, 5 Hz), 7.44 (4H, s).
reaction mixture was concentrated in vacuo, and water and 1
N hydrochloric acid to pH 1 were added. The product was
extracted with ether (3 × 300 mL), and the combined organic
phases were dried (MgSO4), filtered, and concentrated in vacuo
to give the title compound as a crystalline product in 8.0 g
1
(62%) yield. ee > 92% (CE). H NMR (CDCl3) δ 1.19 (6H, t, J
) 7 Hz), 2.90-3.14 (4H, m), 3.42-3.66 (4H, m), 4.06 (2H, m),
4.62 (4H, dd, J ) 7 Hz, 2 Hz), 6.07 (2H, dt, J ) 15 Hz, 2 Hz),
6.39 (2H, dt, J ) 15 Hz, 5 Hz), 6.84 (4H, d, J ) 8 Hz), 7.15
(4H, d, J ) 8 Hz), 7.35 (4H, s). Mp 156-158 °C. Anal.
(C38H38O8) C, H, N.
(E,E)-(S,S)-3-{4-[5-(3-{5-[4-(2-Ca r b oxy-2-et h oxyet h yl)-
p h en oxy]p en t-3-en -1-yn yl}p h en yl)p en t-2-en -4-yn yloxy]-
p h en yl}-2-eth oxyp r op a n oic Acid , 8b. The title compound
was synthesized as described for compound 8a above in 6.52
1
g (99%) yield. ee > 96% (CE). H NMR (CDCl3) δ 1.18 (6H, t,
J ) 7 Hz), 2.88-3.15 (4H, m), 3.33-3.70 (4H, m), 4.05 (2H,
m), 4.62 (4H, dd, J ) 7 Hz, 2 Hz), 6.05 (2H, dt, J ) 15 Hz, 2
Hz), 6.38 (2H, dt, J ) 15 Hz, 5 Hz), 6.85 (4H, d, J ) 8 Hz),
7.17 (4H, d, J ) 8 Hz), 7.23-7.40 (3H, m), 7.50 (1H, s). Mp
76-79 °C. Anal. (C38H38O8) C, H, N.
Ch ir a l An a lysis. The CE analyses were performed on a
HP3DCE capillary electrophoresis instrument (Agilent, Wald-
born, Germany) equipped with an auto sampler, a capillary
cartridge, a high-voltage power supply, a diode array detector,
electrodes, and a hydrostatic injection system. The electro-
phoretic data system was the HP Chemstation software, and
the data were collected with a frequency of 10 Hz. The CE
separations were carried out with untreated fused-silica
capillaries from Agilent with the following dimensions: 80.5
cm total length with 72.0 cm effective length, 50 µm inner
diameter, and extended light path with an inner diameter of
150 µm at the detector window. The electrolyte was prepared
by dissolving 3.0% (w/v) sulfobuthyl ether-â-cyclodextrin (Ad-
vasep 4, Cydex, Inc., Overland Park, KS) and 0.50% (w/v)
dimethyl-â-cyclodextrin (Agilent, Waldborn, Germany) both in
50 mM borate buffer pH 9.3 (Agilent) followed by filtering
through a 0.45 µm polypropylene filter. To this solution was
added 5% (v/v) acetonitrile to give the final electrolyte. The
electrophoresis was carried out in normal polarity mode. The
electrophoretic conditions were as follows: voltage, 21 kV;
current, 50 µA; capillary temperature controlled at 25 °C;
injection was 50 mbar for 4.0 s; detection, UV at 205 nm with
reference of 380 nm. The sample concentration was 0.05 mg/
mL in 1/5 acetonitrile/5 mM borate buffer pH 9.3. The capillary
was conditioned with 0.1 N NaOH for 20 min daily and flushed
with 0.1 N NaOH (3 min), water (2 min), and electrolyte
(3 min) between each run.
(E,E)-5-[3-(5-Hyd r oxyp en t-3-en -1-yn yl)p h en yl]p en t-2-
en -4-yn -1-ol, 21b. The title compound was synthesized in 6.25
g (86.5%) yield as described above for 21a , using 1,3-diiodo-
1
benzene. H NMR (DMSO) δ 4.09 (4H, m), 5.02 (2H, t, J ) 7
Hz), 5.95 (2H, dt, J ) 15 Hz, 2 Hz), 6.38 (2H, dt, J ) 15 Hz,
5 Hz), 7.35-7.50 (4H, m).
Isopr opyl (E,E)-(S,S)-2-Eth oxy-3-{4-[5-(4-{5-[4-(2-eth oxy-
2-isop r op oxyca r bon yleth yl)p h en oxy]p en t-3-en -1-yn yl}-
p h en yl)p en t-2-en -4-yn yloxy]p h en yl}p r op a n oa te. Under
an atmosphere of nitrogen, azodicarboxylic dipiperidide (21.4
g, 84.8 mmol) was added at room temperature to a stirred
solution of tributylphosphine (17.1 g, 84.8 mmol), isopropyl (S)-
2-ethoxy-3-(4-hydroxyphenyl)propanoate (21.4 g, 84.8 mmol),
and 21a (7.6 g, 30.0 mmol) in dry THF (300 mL). After 1 h
water (100 mL) was added and the mixture was extracted with
methylene chloride (3 × 100 mL). The combined extracts were
dried and concentrated in vacuo, and the crude product was
purified by column chromatography using methylene chloride:
THF (20:1) as eluent to give 14.75 g (70%) of the title
compound. 1H NMR (CDCL3) δ 1.18 (6H, t, J ) 7 Hz), 1.23
(12H, t, J ) 7 Hz), 2.95 (4H, d, J ) 7 Hz), 3.30-3.43 (2H, m),
3.55-3.67 (2H, m), 3.95 (2H, t, J ) 7 Hz), 4.63 (4H, dd, J ) 7
Hz, 2 Hz), 5.05 (2H, m), 6.07 (2H, dt, J ) 15 Hz, 2 Hz), 6.39
(2H, dt, J ) 15 Hz, 5 Hz), 6.85 (4H, d, J ) 8 Hz), 7.17 (4H, d,
J ) 8 Hz), 7.27 (4H, s), 7.37 (4H, s).
Isopr opyl (E,E)-(S,S)-2-Eth oxy-3-{4-[5-(3-{5-[4-(2-eth oxy-
2-isop r op oxyca r bon yleth yl)p h en oxy]p en t-3-en -1-yn yl}-
p h en yl)p en t-2-en -4-yn yloxy]p h en yl}p r op a n oa te. The title
compound was synthesized as described above in 7.57 g (40%)
yield using 21b. 1H NMR (CDCl3) δ 1.17 (6H, t, J ) 7 Hz),
1.23 (12H, d, J ) 7 Hz), 2.95 (4H, d, J ) 7 Hz), 3.30-3.41
(2H, m), 3.55-3.66 (2H, m), 3.95 (2H, t, J ) 7 Hz), 4.62 (4H,
dd, J ) 2 Hz, 7 Hz), 5.00-5.10 (2H, m), 6.05 (2H, dt, J ) 15
Hz, 2 Hz), 6.40 (2H, dt, J ) 15 Hz, 5 Hz), 6.84 (4H, d, J ) 8
Hz), 7.18 (4H, d, J ) 8 Hz), 7.22-7.38 (3H, m), 7.50 (1H, s).
Mod elin g. The docking experiments were performed with
FlexX version 1.10.030-33 with assignment of formal charges
and DrugScore34 as docking method. Consensus scoring was
applied on the obtained solutions.35 FlexX was used with the
Sybyl6.8 interface.36
In Vitr o Tr a n sa ctiva tion . The PPAR transactivation
assay has been described previously.13 Briefly, the ligand
binding domains of the three human PPAR receptor subtypes
as well as that of rat PPARR (amino acids 167-469 (end)) were
fused to the DNA binding domain (amino acids 1-147) of the
yeast transcription factor Gal4. HEK293 cells were transiently
transfected with an expression vector for the respective PPAR
chimera along with a reporter construct containing five copies
of the Gal4 DNA binding site driving expression of a luciferase
reporter gene. All compounds were dissolved in DMSO and
diluted 1:1000 upon addition to the cells. Compounds were
tested in five concentrations ranging from 0.01 to 30 µM. Cells
were treated with compound for 24 h followed by luciferase
assay. EC50 values were calculated via nonlinear regression
using GraphPad PRISM 3.02 (GraphPad Software, San Diego,
CA). The results were expressed as means ( SD.
(E,E)-(S,S)-3-{4-[5-(4-{5-[4-(2-Ca r b oxy-2-et h oxyet h yl)-
p h en oxy]p en t-3-en -1-yn yl}p h en yl)p en t-2-en -4-yn yloxy]-
p h en yl}-2-eth oxyp r op a n oic Acid , 8a . To a solution of
isopropyl (E, E)-(S,S)-2-ethoxy-3-{4-[5-(4-{5-[4-(2-ethoxy-2-iso-
propoxycarbonylethyl)phenoxy]pent-3-en-1-ynyl}phenyl)pent-
2-en-4-ynyloxy]phenyl}propanoate (14.75 g, 20.9 mmol) in
ethanol (590 mL) was added 1 N sodium hydroxide (85 mL).
After the reaction mixture was stirred at 80 °C for 1 h, the
P h a r m a cok in etics. The procedure has previously been
described in detail.13 Briefly, the compounds were dosed po
and iv to male SD rats. The compounds were dissolved in 5%
ethanol, 10% HPCD, and phosphate buffer; pH 7.5-8.0. Blood

Dimeric PPAR Agonists
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23 4893
(3) Boyle, P. J .; King, A. B.; Olansky, L.; Marchetti, A.; Lau, H.;
Magar, R.; Martin, J . Effects of Pioglitazone and Rosiglitazone
on the Blood Lipid Levels and Glycemic Control in Patients with
Type 2 Diabetes Melitus: A Retrospective Review of Randomly
Selected Medical Records. Clin. Ther. 2002, 24, 3, 378-396.
(4) Staels, B.; Dallongeville, J .; Auwerx, J .; Schoonjans, K.; Leit-
ersdorf, E.; Fruchart, J .-C. Mechanism of action of fibrates on
lipid and lipoprotein metabolism. Circulation 1998, 98, 2088-
2093.
(5) Oliver, W. R.; Shenk, J . L.; Snaith, M. R.; Russell, C. S.; Plunket,
K. D.; Bodkin, N. L.; Lewis, M. C.; Winegar, D. A.; Sznaidman,
M. L.; Lambert, M. H.; Xu, H. E.; Sternbach, D. D.; Kliever, S.
A.; Hansen, B. C.; Willson, T. M. A selective peroxisome
proliferator-activated receptor δ agonist promotes reverse cho-
lesterol transport. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 5306-
5311.
(6) Kliewer, S. A.; Umesono, K.; Noonan, D. J .; Heyman, R. A.;
Evans, R. M. Convergence of 9-cis retinoic acid and peroxisome
proliferator signaling pathways through heterodimer formation
of their receptors. Nature 1992, 358, 771-774.
(7) Strand, J .; Sandeman, D.; Skovsted, B.; Edsberg, B. Mu¨ller, P.
Ragaglitazar (NNC 61-0029, (-) DRF 2725): The efficacy and
Safety of the Novel Dual PPARR and PPARγ Agonist in Patients
with Type 2 Diabetes. Diabetes 2002, 51, suppl. 2, A144.
(8) Cronet, P.; Petersen, J . F. W.; Folmer, R.; Blomberg, N.; Sjo¨blom,
K.; Karlsson, U.; Lindstedt, E.-L.; Bamberg, K.; Structure of the
PPARR and -γ ligand binding domain in complex with AZ 242;
Ligand selectivity and Agonist Activation in the PPAR family.
Structure 2001, 9, 699-706.
(9) Etgen, G. J .; Oldham, B. A.; J ohnson, W. T.; Broderick, C. L.;
Montrose, C. R.; Brozinick, J . T.; Misener, E. A.; Bean, J . S.;
Bensch, W. R.; Brooks, D. A.; Shuker, A. J .; Rito, C. A.;
McCarthy, J . R.; Ardecky, R. A.; Tyhonas, J . S.; Dana, S. L.;
Bilakovics, J . M.; Paterniti, J . R., J r.; Ogilvie, K. M.; Liu, S.;
Kauffman, R. F. A Tailored Therapy for the Metabolic Syndrome,
The Dual Peroxisome Proliferator-Activated Receptor-R/γ Ago-
nist LY465608 Ameliorates Insulin Resistance and Diabetes
Hyperglycemia While Improving Cardiovascular Risk Factors
in Preclinical Models. Diabetes 2002, 51, 1083-1087.
(10) Lewis, M. C.; Winegar, D. A.; Bodkin, N. L.; Hansen, B. C.;
Oliver, W. R. Improved Insulin Sensitivity and Dyslipidemia
Following Oral Administration of Combination PPARγ/R and
PPARδ Agonist in Obese Rhesus Monkeys. Diabetes 2002, 51,
suppl. 2, A140.
(11) Lewis, M. C.; Wilson, J . G.; Franklin, M. C.; Brown, J . G.; Strole,
C. A.; Oliver, W. R.; Winegar, D. A. Co-Administration of PPARγ,
R and δ Agonist, improves Efficacy Relative to PPARγ Agonist
Alone in ZDF rats. Diabetes 2002, 51, suppl. 2, A140.
(12) Mogensen, J . P.; J eppesen, L.; Bury, P. S.; Pettersson, I.;
Fleckner, J .; Nehlin, J .; Frederiksen, K. S.; Albrektsen, T.; Din,
N.; Mortensen, S. B.; Svensson, L. A.; Wassermann, K.; Wulff,
E. M.; Ynddal, L.;Sauerberg, P. Design and synthesis of novel
PPAR R/γ/δ triple activators using a known PPARR/γ dual
activator as structural template. Bioorg. Med. Chem. Lett. 2003,
13, 257-260.
samples were collected in EDTA tubes. Each data point
represents one animal.
Plasma samples were analyzed by high turbulence liquid
chromatography (HTLC) combined with tandem mass spec-
trometry (MS/MS).
In Vivo Mod el. The model has earlier been published in
detail.13 Briefly, 14 weeks old C57BL/KsBom-db/db male mice
(n ) 6 per dose) were dosed by gavage, with a suspension of
compound in 0.2% CMC + 0.4% Tween-80 in saline for 10 days.
After 7 days of treatment, nonfasted blood samples were drawn
and analyzed for full blood glucose and plasma triglycerides.
On day 9 of treatment an oral glucose tolerance test (OGTT)
was performed on overnight fasted animals. Blood samples
from the tail vein were drawn before and at 30, 60, and 120
min after the glucose dosing (3 g/kg).
Statistics: ED50 values were calculated via nonlinear re-
gression using GraphPad PRISM 3.02 (GraphPad Software,
San Diego, CA). The results were expressed as means ( SEM.
Differences between two groups were evaluated by one-way
ANOVA and Dunett’s multiple comparison test: * P < 0.05,
** P < 0.01. P values less than 0.05 were considered signifi-
cant.
Percent reduction was calculated using the equation: ((Cv
- Ct)/Cv) × 100, where Cv was the plasma concentration in
the vehicle treated group, Ct the plasma concentration in the
compound treated group.
Ma cr om olecu la r Cr ysta llogr a p h y. Ligand binding do-
main (LBD, amino acids C165 - Stop) PPARγ was expressed,
purified, and crystallized according to Ebdrup et al. (2003).14
The crystal space group and cell parameters obtained are
found in Table X+1, Supporting Information. A crystal was
transferred to a solution containing 48% polyethylene glycol
4000, 0.15 M Tris-HCl pH 8.0, and concentrated 7 and was
left soaking for 7 days. The crystal was thereafter flash-frozen
in liquid nitrogen and mounted on the goniostat in a nitrogen
gas-stream at 100 K. Crystallographic data were collected at
beamline I711, the MAX-laboratory, Sweden,37 using a marC-
CD system, and the data set was evaluated by the XDS
program package.38 The structure was subsequently refined
by, first, the CNX program system39 and, in later stages, by
Refmac40 of the CCP4 program system.41 Input model was the
PPARγ coordinates generated by Sauerberg et al. (2002),13
PDB code 1KNU. Introduction of 7 and corrections to the
model according to electron density maps were made with use
of the Quanta program.42 The program Xplo2D43 was used for
creation of ligand Parameter and Topology files used by the
CNX program. According to refinement statistics and electron
density maps, the occupancy of the ligand in the crystal was
around 0.75. For data collection, refinement, and model
statistics, see Supporting Information Table X+1. The coor-
dinates of the 7/PPARγ structure have been deposited in the
Brookhaven Protein Data Bank.
(13) Sauerberg, P.; Pettersson, I.; J eppesen, L.; Bury, P. S.; Mogensen,
J . P.; Wassermann, K.; Brand, C. L.; Sturis, J .; Wo¨ldike, H. F.;
Fleckner, J .; Andersen, A.-S. T.; Mortensen, S. B.; Svensson, L.
A.; Rasmussen, H. B.; Lehmann, S. V.; Polivka, Z.; Sindelar, K.;
Panajotova, V.; Ynddal, L.; Wulff, E. M. Novel tricyclic-R-
alkoxyphenylpropionic Acids: Dual PPARR/γ Agonists with
Hypolipidemic and Antidiabetic Activity. J . Med. Chem. 2002,
45, 789-804.
Ack n ow led gm en t . The technical assistance from
M. A. Zundel, B. Rosenberg, A. Bergholdt, V. Weil, B.
Bentzen, S. J . Mozer, A. Ryager. F. G. Gundertofte, R.
Burgdorf, N. Steffensen, L. Priskorn, H. Bach, A.
Heerwagen, A. Zeneca, K. M. Klausen, C. Christensen,
K. Pedersen, O. Larsson, S. von Eyben, P. S. Klifforth,
and A. Ravn is highly appreciated.
(14) Ebdrup, S.; Pettersson, I.; Rasmussen, H. B.; Deussen, H.-J .;
J ensen, A. F.; Mortensen, S. B.; Fleckner, J .; Pridal, L.; Nygaard,
L.; Sauerberg, P. Synthesis, and Biological and Structural
Characterization of the Dual Acting PPARR/γ Agonist Ragagli-
tazar. J . Med. Chem. 2003, 46, 1306-1317.
(15) Allen, C. F. H.; Edens, C. O., J r. Phenylpropargylaldehyde.
Organic Syntheses; Wiley: New York, 1955; Collect. Vol. III, pp
731-733.
(16) Krause, N. Synthesis of Allenes by 1,6-Addition of Organocu-
prates to Polarized Enynes. Chem. Ber. 1990, 123, 2173-2180.
(17) Takeuchi, R.; Tanabe, K.; Tanaka, S. Stereodivergent Synthesis
Su p p or tin g In for m a tion Ava ila ble: Table of interactions
between 7 and the amino acids of the LBD-PPARγ receptor
protein (Table X). Table of data collection, refinement and
model statistics (Table X+1). This material is available free
of charge via the Internet at http://pubs.acs.org.
of (E)- and (Z)-2-Alken-4-yn-1-ols from 2-Propynoic Acid:
A
Practical Route via 2-Alken-4-ynoates. J . Org. Chem. 2000, 65,
1558-1561.
(18) Taylor, E. C. Iminium Salts in Organic Chemistry. In Advances
in Organic Chemistry: Methods and Results, Ed.; J ohn Wiley
& Sons: New York. 1979.
(19) Willson, T. M.; Cobb, J . E.; Cowan, D. J .; Wiethe, R. W.; Correa,
I. D.; Prakash, S. R.; Beck, K. D.; Moore, L. B.; Kliever, S. A.;
Lehmann, J . M. The Structure-Activity Relationship between
Peroxisome Proliferator-Activated Receptor γ Agonism and the
Antihyperglycemic Activity of Thiazolidinediones. J . Med. Chem.
1996, 39, 665-668.
Refer en ces
(1) Porte, D., J r.; Schwartz, M. W. Diabetes complications: Why is
Glucose Potentially Toxic? Science 1996, 27, 699-700.
(2) Chilcott, J .; Tappenden, P.; J ones, M. L.; Wright, J . P. A
systematic Review of the Clinical Effictiveness of Pioglitazone
in the treatment of type 2 Diabetes Mellitus. Clin. Ther. 2001,
23, 1792-1823.
(20) Halazy, S. G-protein coupled receptor bivalent ligands and drug
design. Exp. Opin. Ther. Pat. 1999, 9, 431-446.

4894 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 23
Sauerberg et al.
(21) Christopoulos, A.; Grant, M. K. O.; Ayoubzadeh, N.; Kim, O. N.;
Sauerberg, P.; J eppesen, L.; El-Fakahany, E. E. Synthesis and
Pharmacological Evaluation of Dimeric Muscarinic Acetylcholine
Receptor Agonists. J . Pharmacol. Exp. Ther. 2001, 298, 1260-
1268.
(22) Nolte, R. T.; Wisely, G. B.; Westin, S.; Cobb, J . E.; Lambert, M.
H.; Kurokawa, R.; Rosenfeld, M. G.; Willson, T. M.; Glass, C.
K.; and Milburn, M. V. Ligand binding and co-activator assembly
of the peroxisome proliferator-activated receptor-gamma. Nature
1998, 395, 137-143.
(23) Uppenberg, J .; Svensson, C.; J aki, M.; Bertilsson, G.; J endeberg,
L.; Berkenstam, A. Crystal Structure of the Ligand Binding
Domain of the Human Nuclear Receptor PPARγ. J . Biol. Chem.
1998, 273, 31108-31112.
(24) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J .
Experimental and computational approaches to estimate solubil-
ity and permeability in drug discovery and development settings.
Adv. Drug Delivery Rev. 1997, 23, 3-25.
(25) Veber, D. F.; J ohnson, S. R.; Cheng, H.-Y.; Smith, B. R.; Ward,
K. W.; Kopple, K. D. Molecular Properties That Influence the
Oral Bioavailably of Drug candidates. J . Med. Chem. 2002, 45,
2615-2623.
(26) Faber, K. N.; Mu¨ller, M.; J ansen, P. L. M. Drug transport
proteins in the liver. Adv. Drug Delivery Rev. 2003, 55, 107-
124.
(27) Gampe, R. T.; Montana, V. G.; Lambert, M. H.; Miller, A. B.;
Bledsoe, R. K.; Milburn, M. V.; Kliever, S. A.; Willson, T. M.;
Xu, H. E. Asymmetry in the PARγ/RXRR Crystal Structure
Reveals the Molecular Basis of Heterodimerization among
Nuclear Receptors. Mol. Cell 2000, 5, 545-555.
(31) Kramer, B.; Rarey, M.; Lengauer, L. Evaluation of the FlexX
incremental construction algorithm for protein-ligand docking.
Proteins: Struct. Funct. Genet. 1999, 37, 228-241.
(32) Rarey, M.; Wefing, S.; Lengauer, T. Placement of medium-sized
molecular fragments into active sites of proteins. J . Comput.-
Aided Mol. Design 1996, 10, 41-54.
(33) Rarey, M.; Kramer, B.; Lengauer, T. Multiple automatic base
selection: Protein-ligand docking based on incremental con-
struction without manual intervention. J . Comput.-Aided Mol.
Des. 1997, 11, 369-384.
(34) Sotriffer, C. A.; Gohlke, H.; Klebe, G. Docking into Knowledge-
Based Potential Fields: A Comparative Evaluation of DrugScore.
J . Med. Chem. 2002, 45, 1967-1970.
(35) Clark, R. D.; Strizhev, A.; Leonard, J . M.; Blake, J . F.; Matthew,
J . B. Concensus scoring for ligand/protein interactions. J . Mol.
Graph. Mod. 2002, 20, 281-295.
(36) SYBYL 6.8 Tripos Inc., 1699 South Hanley Rd., St. Louis, MO
63144.
(37) Cerenius, Y.; Ståhl, K.; Svensson, L. A.; Ursby, T.; Oskarsson,
Å.; Albertsson, J .; Liljas, A. The crystallography beamline I711
at MAX II. J . Synchrotron Rat. 2000, 7, 203-208.
(38) Kabsch, W. Automatic processing of rotation diffraction data
from crystals of initially unknown symmetry and cell constants.
J . Appl. Crystallogr. 1993, 26, 795-800.
(39) Bru¨nger, A. T.; Krukowski, A.; Erickson, J . W. Slow-cooling
protocols for crystallographic refinement by simulated annealing.
Acta Crystallogr. Sect. A 1990, 46, 585-593.
(28) Oberfield, J . L.; Collins, J . L.; Holmes, C. P.; Goreham, D. M.;
Cooper, J . P.; Cobb, J . E.; Lenhard, J . M.; Hull-Ryde, E. A.; Mohr,
C. P.; Blanchard, S. G.; Parks, D. J .; Moore, L. B.; Lehmann, J .
M.; Plunket, K.; Miller, A. B.; Milburn, M. V.; Kliewer, S. A.;
Willson, T. M. A peroxisome proliferator-activated receptor γ
ligand inhibits adipocyte differentiation. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 6102-6106.
(40) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J . Refinement of
macromolecular structures by the maximum- likelihood method.
Acta Crystallogr. Sect. D 1997, 53, 240-255.
(41) Bailey. S. The ccp4 suite - programs for protein crystallography.
Acta Crystallogr. Sect. D 1994, 50, 760-763.
(42) Oldfield, T. J .; Hubbard, R. E. Analysis of c-alpha geometry in
protein structures. Proteins 1994, 18, 324-337.
(43) Kleywegt, G. J .; J ones, T. A. Databases in protein Crystal-
lography. Acta Crystallogr. Sect. D 1998, 54, 1119-1131.
(29) Several other dimeric ligands have subsequently shown potent
PPAR activity. Sauerberg et al. Unpublished.
(30) Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. A Fast Docking
Method using an Incremental Construction Algorithm J . Mol.
Biol. 1996, 261, 470-489.
J M0309046