Identification and characterization of 4-aryl-3,4-dihydropyrimidin-2(1H)-ones as inhibitors of the fatty acid transporter FATP4

Article abstract of DOI:10.1016/j.bmcl.2006.03.102

Several potent, cell permeable 4-aryl-dihydropyrimidinones have been identified as inhibitors of FATP4. Lipophilic ester substituents at the 5-position and substitution at the para-position (optimal groups being -NO₂ and CF₃) of the 4-aryl group led to active compounds. In two cases racemates were resolved and the S enantiomers shown to have higher potencies.

Full text of DOI:10.1016/j.bmcl.2006.03.102

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3504–3509
Identification and characterization of 4-aryl-3,4-dihydropyrimidin-
2(1H)-ones as inhibitors of the fatty acid transporter FATP4
Christopher Blackburn,a,* Bing Guan,a James Brown,a Courtney Cullis,a
Stephen M. Condon,a Tracy J. Jenkins,a Stephane Peluso,a Yingchun Ye,a
Ruth E. Gimeno,b Sandhya Punreddy,b Ying Sun,b Hui Wu,b Brian Hubbard,b
Virendar Kaushik,b Peter Tummino,b Praveen Sanchetti,b Dong Yu Sun,b
Tom Daniels,b Effie Tozzo,b Suresh K. Balanic and Prakash Ramana
aMedicinal Chemistry, Millennium Pharmaceuticals, Inc., 40 Landsdowne St. Cambridge, MA 02139, USA
bMetabolic Disease Biology, Millennium Pharmaceuticals, Inc., 40 Landsdowne St. Cambridge, MA 02139, USA
cDMPK, Millennium Pharmaceuticals, Inc., 40 Landsdowne St. Cambridge, MA 02139, USA
Received 30 January 2006; revised 28 March 2006; accepted 29 March 2006
Available online 27 April 2006
Abstract—Several potent, cell permeable 4-aryl-dihydropyrimidinones have been identified as inhibitors of FATP4. Lipophilic ester
substituents at the 5-position and substitution at the para-position (optimal groups being –NO2 and CF3) of the 4-aryl group led to
active compounds. In two cases racemates were resolved and the S enantiomers shown to have higher potencies.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Blocking the absorption of fats (triglycerides) by admin-
istration of an anti-absorptive agent is of interest for the
treatment of obesity.1 Ingested dietary triglycerides are
hydrolyzed by gastric and pancreatic lipases, and the
resulting fatty acids are taken up by enterocytes lining
the small intestine where they are re-esterified to triglyc-
erides and then transported into the blood. The lipase
inhibitor orlistat (XenicalTM) blocks fat absorption by
inhibiting the hydrolysis of dietary fat to fatty acids2
with administration leading to a concomitant decrease
in body weight and improvement of blood lipid profiles.
A family of proteins, termed fatty acid transport pro-
teins (FATPs), that mediate the uptake of fatty acids
into cells has been described.3,4 Previous studies5–8 pro-
vided evidence that fatty acid transport protein 4
(FATP4) mediates the transport of fatty acids from
the gut into enterocytes both in vitro and in vivo. We
therefore reasoned that inhibitors of FATP4 might be
expected to have benefits similar to orlistat. Since
FATP4 inhibition would result in the accumulation of
free fatty acids rather than triglycerides, we would also
expect a different, possibly improved, side-effect profile
compared to orlistat. The FATP family of proteins is
most closely related in sequence to the ATP-utilizing
acyl-CoA-synthetase enzymes.9–12 While much pub-
lished work on FATPs has primarily focused on net fat-
ty acid transport activity, recent data from these
laboratories13 and published in the literature9–12 show
that FATP4 and its closest homolog FATP1 have
acyl-CoA-synthetase activity. A prominent model of fat-
ty acid transport proposes that the acyl-CoA-synthetase
activity of FATPs mediates fatty acid uptake by forma-
tion of fatty acyl-CoA derivatives thereby creating a
concentration gradient that drives uptake.9 Moreover,
human FATP4-mediated uptake of labeled fatty acids
can be competed by long-chain fatty acids (>C-12),
but not by short-chain carboxylic acids, methyl esters
of fatty acids, heterocyclic isosteres of carboxylic acids
with long-chain substituents,14 or other lipophilic mole-
cules. These data are consistent with viewing FATP4 as
a long-chain acyl-CoA-synthetase.
Initially, we sought FATP4 inhibitors using an assay
conducted under comparable conditions to those used
for target validation.6 Thus, a cell-based assay was run
in an HTS format measuring inhibition of the uptake
of a fluorescent long-chain fatty acid, 12-BODIPY-lauric
Keywords: FATP4 inhibitor; Fatty acid; Biginelli condensation.
*
Corresponding author. Tel.: +1 617 761 6811; fax: +1 617 444
1483; e-mail: chris.blackburn@mpi.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.03.102

C. Blackburn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3504–3509
3505
Table 1. Inhibition of FATP4 mediated uptake of 12-BODIPY-lauric
acid in HEK 293 cells by dihydropyrimidinonesa,b
acid, into HEK 293 cells stably transfected with human
FATP4.15 In order to identify non-specific inhibitors,
cell-based counter-screens were run measuring inhibi-
tion of uptake of the same substrate into mouse FATP5
and human FATP2. Among the hits from the HTS that
showed selectivity over FATP2 and FATP5, the dihy-
dropyrimidinone 1a (Fig. 1) was found to have an
IC50 value of 1.2 lM. Herein we describe optimization
of this compound series leading to analog 1p.
NO2
O
R1
HN
O
O
N
H
CH3
Dihydropyrimidinones 1 can readily be prepared by the
Biginelli three-component condensation reaction be-
tween a urea, an aldehyde, and a b-ketoester.16–18 In or-
der to probe the SAR at the 5-position alkoxy group
without the need to prepare individual b-ketoesters we
sought a synthetic route to carboxylic acid 2 (Scheme 1).
1-2
Compound
R1
H
FATP4 IC50 (lM)
2
>30
1.2
1a
1b
1c
1d
1e
1f
CH2CH2SEt
CH2CH2SOEt
CH2CH2SO2Et
CH2CH2OEt
n-Pentyl
>30
>30
>30
2.0
Condensation between urea, 4-nitrobenzaldehyde, and
2-cyanoethylacetoacetate in the presence of boron triflu-
oride19 afforded 3 which, unlike other esters,20 could be
hydrolyzed by treatment with base under mild condi-
tions to give 2. Esterification of 2 using alcohols or phe-
nols in the presence of EDC afforded several examples
of series 1 (Table 1). To probe the effect of the oxidation
Et
Me
9.5
>30
1.5
1g
1h
1i
Allyl
trans-2-Butenyl
0.65
0.64
0.26
0.09
0.70
0.18
1.2
1j
trans-2-Pentenyl
trans-2-Hexenyl
trans-2-Heptenyl
cis-2-Pentenyl
2-Cyclohexenyl
Cyclobutyl
1k
1l
1m
1n
1o
1p
S-1p
R-1p
1q
1r
Cyclopentyl
Cyclopentyl
0.25
0.20
25.0
0.50
1.2
Cyclopentyl
Cyclohexyl
Cycloheptyl
1s
trans-4-Et-cyclohexyl
cis-4-Et-cyclohexyl
Ph
0.15
2.4
1t
1u
1v
>30
>30
CH2Ph
a For assay conditions see Ref. 15.
b All compounds are racemates unless indicated otherwise.
Figure 1. High-throughput screening hit 1a and optimized FATP4
inhibitor 1p.
state of the sulfur atom in 1a, we effected oxidations to
the sulfoxide 1b and sulfone 1c as shown in Scheme 2.
Investigations of the SAR at the para-position of the 4-
aryl group were confined to derivatives with the optimal
cyclopentyl ester (vide infra) at the 5-position. Reaction
of cyclopentanol with diketene afforded ketoester 4a
which was subjected to the BF3 catalyzed19 Biginelli
condensation with urea and the appropriate aldehyde
to give 5a–o. A representative amide, 6, was prepared
Scheme 1. Reagents and conditions: (a) urea, 4-nitrobenzaldehyde,
BF3ÆOEt2, (1.3 equiv), CuI (0.1 equiv.), HOAc (0.1 equiv), THF,
reflux, 18 h, 88%; (b) NaOH, Me2CO, H2O, 0–25 ꢁC, 95%; (c)
R1OH, EDC, DMAP, DMF, 50 ꢁC, 16 h, 25–55%.
Scheme 2. Reagents and conditions: (a) m-CPBA, CH2Cl2, 25 ꢁC, 8 h,
68%; (b) i—m-CPBA, CH2Cl2, reflux, 18 h; ii—crystallization
(acetone/hexane), 34%.

3506
C. Blackburn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3504–3509
Scheme 3. Reagents and conditions: (a) cyclopentanol, DIEA, CH2Cl2, 0–25 ꢁC, 95%; (b) cyclopentylamine, THF, 25 ꢁC, 30%; (c) urea, or N-alkyl-
urea, ArCHO, BF3ÆOEt2, (1.3 equiv), CuI (0.1 equiv), HOAc (0.1 equiv), THF, reflux, 18 h, 50–70%.
Figure 2.
similarly from ketoamide 4b, and 1-substituted deriva-
tives, 7, were synthesized from N-substituted ureas
(Scheme 3).
with increasing chain length (series 1h–l). In the case
of the 2-pentenyl esters the cis and trans isomers have
comparable activities (compare 1j with 1m). Alicyclic
groups were also investigated and, in general, potency
enhancements were observed relative to their open chain
analogs. Within this series, potency improved in the or-
der cycloheptyl (1r) = cyclobutyl (1o) < cyclohexyl
(1q) < cyclopentyl (1p). Substituted cyclohexyl deriva-
tives were also found to be active, and, for compounds
derived from 4-ethylcyclohexanol, the trans isomer 1s
was found to be some 20-fold more active than the cis
isomer 1t. In contrast, aryl (1u) and benzyl (1v) esters
were found to be inactive.
Dihydropyrimidinones 8 with heteroaryl and alkyl
groups at the 4-position shown in Figure 2 were pre-
pared similarly from the appropriate aldehyde. A repre-
sentative dihydrothiopyrimidinone 9 was obtained in
65% yield by condensing thiourea with 4-nitrobenzalde-
hyde and b-ketoester 4a in ethanolic HCl.21
In the whole cell uptake assay measuring the inhibition
of uptake of the fluorophore 12-BODIPY-lauric acid,15
carboxylic acid 2 and amide 6 proved to be inactive. In
contrast, several derivatives with lipophilic ester groups
at the 5-position exhibit significant activity (Table 1).
With regard to substituent effects in the 4-aryl ring
system, HTS data indicated that ortho- and meta-substi-
tuted compounds are considerably less potent than para
derivatives.22 When the optimal cyclopentyl ester group
was retained, we found several compounds with approx-
imately 1 lM activity (Table 2) where the para-substitu-
ents were OCF3 (5j), halogen (5d and 5e), and CF3 (5k)
but no significant potency enhancements were observed
compared to the nitro compound 1p. Replacement of
the substituted phenyl group with heteroaromatics did
not result in any significant activity, compounds 8a–e
(Fig. 2) all having IC50 values > 30 lM. Likewise, alkyl
The requirement for
a
lipophilic substituent is
further illustrated by comparison of thioether 1a
(IC50 = 1.2 lM) with the more polar analogs, sulfoxide
1b, sulfone 1c, and ether 1d all of which were found to
be inactive compounds. The n-pentyl analog 1e exhibits
comparable potency to 1a but shorter chains, as in ethyl
ester 1f and methyl ester 1g, confer little activity. Unsat-
urated aliphatic substituents were next investigated and
it was found that allylic esters show enhanced potencies
compared to saturated analogs. Potency also improved

C. Blackburn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3504–3509
3507
cell OD column23–25 and their orientations tentatively
assigned by comparing their relative retention times with
those of closely related compounds whose absolute
configurations have been established.24,25 On this basis,
we found that S(+)-1p was more potent than R(À)-1p
(Table 1) with a similar trend for the enantiomeric pair
5k (Table 2).
Table 2. Inhibition of FATP4 mediated uptake of 12-BODIPY-lauric
acid in HEK 293 cells by dihydropyrimidinonesa,b
R2
O
HN
O
In order to verify that these lead compounds inhibit
the uptake of native fatty acids as well as BODIPY-
labeled analogs, an assay measuring uptake of radiola-
beled lauric acid was configured.26 All compounds
tested (n ꢀ 20) showed similar potency and stereo-
preference in this assay compared to the BODIPY-
lauric acid assay. Finally, none of the compounds
listed in Tables 1 and 2 were found to have any activ-
ity in an uptake assay for mouse FATP5 or human
FATP2 (family members with ꢀ40% homology to
FATP4).
O
N
H
5
CH3
Compound
R2
FATP4 IC50 (lM)
5a
5b
5c
H
Me
>30
4.5
8.0
1.0
1.0
1.4
26
F
Cl
5d
5e
Br
CN
5f
5g
5h
5i
OH
OMe
O-i-Pr
OCF3
CF3
CF3
CF3
NH2
NMe2
SMe
5.0
12.7
1.1
1.0
0.6
>30
>30
16
Several members of the dihydropyrimidinone series de-
scribed above were found to be orally bioavailable in
DIO mice.27 In an efficacy experiment28 mice were
placed on a high fat diet, administered racemic 1p
(100 mpk), and then given access to food. Pharmaco-
kinetic analyses showed plasma concentrations of
1p > 3 lM at 0.5 h (the Tmax value) and 0.4 lM after
6 h. More importantly, high concentrations (>17-fold
the IC50) of 1p were detected both within the tissues
of the small intestine and in the lumen (Table 3).29
Despite these distributions, dosing of 1p had no signif-
icant effects on the total lipid content of either the lu-
men or the tissue of different parts of the small
intestine. In contrast, orlistat significantly decreased
tissue lipid in the duodenum and increased luminal
lipid in the ileum. These data demonstrate that dosing
of 1p does not affect fat absorption in vivo. Possible
reasons for the observed lack of efficacy include assay
conditions, potency and mechanism of action of the
inhibitors.30
5j
5k
S-5k
R-5k
5l
5m
5n
5o
8.0
>30
SO2Me
a For assay conditions see Ref. 15.
b All compounds are racemates unless indicated otherwise.
substitution at the 4-position (e.g., derivative 8f) and N-
1 substitution (derivatives 7a–d) led to inactive com-
pounds. Replacement of an oxygen atom by a sulfur
atom in the pyrimidinone ring resulted in a loss in
potency; for example, compound 9, analogous to 1p,
has an IC50 value of 3.4 lM.
The cyclopentyl esters 1p and 5k are among the most
potent analogs found in our SAR investigations. These
racemic compounds were separated into their compo-
nent enantiomers by chiral chromatography on a chiral-
In summary, our study represents the first report of spe-
cific small molecule inhibitors of a member of the FATP
family of proteins.
Table 3. The distribution and effect on lipid levelsa in the small intestine of dihydropyrimidinone 1p and orlistat dosed at 100 mpkb
Duodenum
Jejunum
26.7
Ileum
11.0
5.5
1p in lumen (lM)
1p in tissue (lM)
Lipid in lumen (%)
9.3
4.3
4.9
Vehicle
Orlistat
1p
69.7 ( 2.8)
69.6 ( 9.7)
91.0 ( 24.8)
32.1 ( 2.0)
38.2 ( 1.8)
23.6 ( 8.4)
6.6 ( 1.6)
30.6 ( 3.5)*
4.8 ( 1.4)
Lipid in tissue (%)
Vehicle
Orlistat
1p
9.7 ( 0.1)
5.9 ( 0.4)*
7.9 ( 1.4)
13.9 ( 1.7)
9.9 ( 1.4)
11.6 ( 1.0)
10.9 ( 1.6)
11.0 ( 1.7)
11.9 ( 1.0)
a Dried luminal content and lyophilized tissue homogenates were extracted successively with H2O–chloroform–methanol (1:3.3:4) and chloroform–
methanol (2:1) for 1 h. The extracts were dried, weighed and assigned as total lipid expressed as percentage of dried luminal material or lyophilized
tissue weight.
b Mean SEM.
* p < 0.01.

3508
C. Blackburn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3504–3509
20. The corresponding methyl ester afforded acid 2 in only
Acknowledgments
40% yield after prolonged refluxing in methanolic sodium
hydroxide. Mild hydrolysis of related esters has been
reported previously Nagarathnam, D.; Wetzel, J. M.;
Miao, S. W.; Marzabadi, M. R.; Chiu, G.; Wong, W. C.;
Hong, X.; Fang, J.; Forray, C.; Branchek, T. A.; Heydorn,
W. E.; Chang, R. S. L.; Broten, T.; Schorn, T. W.;
Gluchowski, C. J. Med. Chem. 1998, 41, 5320.
We thank Izumi Takagi and Nanda Gulavita for analyt-
ical chemistry support.
References and notes
21. Sherif, S. M.; Youseff, M. M.; Mobarak, K. M.; Abdel-
Fattah, A. M. Tetrahedron 1993, 49, 9561.
22. For example, the compounds shown below, which are
analogous to potent compounds described in the text,
showed no significant inhibition at 20 lM in the HTS.
1. Thomson, A. B. R.; De Pover, A.; Keelan, M.; Jarocka-
Cyrta, C.; Clandinin, M. Y. Methods Enzymol. 1997,
286, 3.
2. Ballinger, A.; Peikin, S. R. Eur. J. Pharmacol. 2002, 440,
109.
3. Schaffer, J. E.; Lodish, H. Cell 1994, 79, 427.
4. Hirsch, D.; Stahl, A.; Lodish, H. F. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 8625.
5. (a) Abumrad, N.; Coburn, C.; Ibrahimi, A. Biochim.
Biophys. Acta 1999, 1441, 4; (b) Hatch, G. M. J. Lipid Res.
2002, 43, 1380; (c) Schaffer, J. E. Am. J. Physiol.
Endocrinol. Metab. 2002, 282, E239.
6. Stahl, A.; Hirsch, D. J.; Gimeno, R. E.; Punreddy, S.;
Ge, P.; Watson, N.; Patel, S.; Kotler, M.; Raimondi,
A.; Tartaglia, L. A.; Lodish, H. F. Mol. Cells 1999, 4,
299.
7. (a) Gimeno, R. E.; Hirsch, D. J.; Punreddy, S.; Sun, Y.;
Ortegon, A. M.; Wu, H.; Daniels, T.; Stricker-Krongrad,
A.; Lodish, H. F.; Stahl, A. J. Biol. Chem. 2003, 278,
49512; (b) Schaffer, J. E.; Lodish Cell 1994, 79, 427.
8. Stuhlsatz-Krouper, S. M.; Bennett, N. E.; Schaffer, J. E.
J. Biol. Chem. 1998, 273, 28642.
23. Chiral preparative chromatography was conducted on a
Chiralcel OD column (2· 25 cm) eluting with isopropa-
nol–hexane (1:4) at a flow rate of 9 mL/min with UV
detection at 270 nm. R- and S-Designations were
assigned by comparison of the retention times and
the sign of the optical rotation of the individual
enantiomers with those published in the literature for
closely related compounds.22 Chromatography of race-
9. Coe, N. R.; Smith, A. J.; Frohnert, B. I.; Watkins, P. A.;
Bernlohr J. Biol. Chem. 1999, 274, 36300.
10. Hall, A. M.; Smith, A. J.; Bernlohr, D. A. J. Biol. Chem.
2003, 278, 43008.
mate 1p gave:
S isomer: tR = 12.81 min, [a]D +56.0
(MeOH); R isomer: tR = 13.45 min, [a]D À54.5 (MeOH).
Chromatography of racemate 5k gave:
tR = 10.94 min, [a]D +37.2 (MeOH);
S
isomer:
isomer:
R
11. Abumrad, N.; Harmon, C.; Ibrahimi, A. J. Lipid Res.
1998, 39, 2309.
tR = 13.45 min, [a]D À36.9 (MeOH).
24. Kappe, C. O.; Uray, G.; Roschger, P.; Lindner, W.;
Kratky, C.; Keller, W. Tetrahedron 1992, 48, 5473.
25. Krenn, W.; Verdino, P.; Uray, G.; Faber, K.; Kappe, C.
O. Chirality 1999, 11, 659.
12. Herrmann, T.; Buchkremer, F.; Gosch, I.; Hall, A. M.;
Bernlohr, D. A.; Stremmel, W. Gene 2001, 270, 31.
13. Hubbard, B. et al. manuscript in preparation. His-tagged
FATP4 was purified by chromatography on a nickel
column and shown to have 3-fold enhanced catalytic
activity as compared to crude membrane preparations.
Experiments were conducted with an homologous series of
long-chain fatty acids and the formation of the CoA
adduct monitored by LC-MS. Optimal activity was
observed at pH 8 in the presence of Mg2+ and Km values
were determined as follows: lauric acid, 21 lM; CoA,
29 lM; ATP, 470 lM.
26. HEK 293 cells stably transfected with human FATP4 were
incubated in assay buffer (Hank’s balanced salt solution,
10 mM Hepes, and 0.1% fatty acid free BSA, pH 7.5)
containing 50 nM [1-14C] lauric acid (specific activity
ꢀ50 mCi/mmol; American Radiochemicals, St. Louis,
MO) and 10 lM lauric acid for 1 h at ambient temperature.
Cells were then washed twice in assay buffer and lysed by
incubating in 50 mM NaOH for 15 min. Cell-associated
radioactivity was measured in a MicroBeta scintillation
counter. Assays were performed in triplicate with at least
two independent determinations for each data point. IC50
values were determined using the XL-fit software with
vector-transfected cells as 100% inhibition control.
27. For example, plasma exposure data for compound 1p and
two analogs in DIO in mice after 30 mg/kg oral dose are as
follows:
14. Several heterocyclic compounds including tetrazoles and
triazoles with long alkyl chains were found to have no
effect on the uptake of labeled fatty acids in the whole cell
assay.
15. HEK 293 cells stably transfected with human FATP4
were incubated for 1 h with 3.5 lM, 12-BODIPY-lauric
acid (Molecular Probes D-3823) in the presence of
1 mM taurocholate in 20 mM Hepes at pH 7.5 in the
presence of variable concentrations of inhibitor. Follow-
ing washing, intracellular fluorescence was measured on
a plate reader. Assays were performed in triplicate with
at least two independent determinations for each data
point. IC50 values were determined using the XL-fit
software with vector-transfected cells as 100% inhibition
control.
16. Kappe, C. O. Acc. Chem. Res. 2000, 33, 879.
17. Kappe, C. O.; Stadler, A. Org. React. 2004, 63, 1.
18. Kappe, C. O. J. Org. Chem. 1997, 62, 7201.
19. Hu, E. H.; Sidler, D. R.; Dolling, U. H. J. Org. Chem.
1998, 63, 3454.
1p
1a
5k
Plasma AUC (nM h)
Plasma T1/2 (h)
Plasma Cmax (nM)
4650
3.9
1320
6525
4.9
710
10,700
5.3
1300
28. C57BL/6 male mice were placed on a 45% high fat diet for
1 week. After an overnight fast, FATP4 inhibitor 1p
(racemic) or the lipase inhibitor orlistat (XenicalTM), or
vehicle was administered by oral gavage. Mice were then
given access to food immediately after dosing. After 6 h,
blood, stomach, small intestine (duodenum, jejunum, and

C. Blackburn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3504–3509
3509
ileum), other tissues and their contents were collected and
analyzed for lipids and 1p.
30. Analysis of FATP4 knock-out animals showed that 50%
inhibition of fatty acid uptake by enterocytes does not
result in a detectable defect in fat absorption in vivo,
suggesting a large excess capacity for fatty acid uptake in
the small intestine.7 This excess capacity will make it
more difficult to detect inhibition in vivo. Furthermore,
while compound 1p was able to inhibit fatty acid uptake
in a transfected cell system, we have not yet assessed its
ability to inhibit uptake in primary enterocytes in vitro
using conditions that mimic the environment of the small
intestine. Depending on the mechanism of inhibition, the
potency of 1p may be significantly altered in the
intestinal environment.
29. In order for FATP4 inhibitors to distribute to the site of
action (enterocytes) and to function at the intracellular
active site, cell permeable compounds are required, (several
compounds with attenuated permeabilities such as quater-
nary ammonium derivatives and high molecular weight
symmetrical dimers were also prepared but none showed
any FATP4 inhibitory activity; unpublished results). Sub-
sequent distribution into the blood compartment may,
however, be detrimental to efficacy; nonetheless, the resid-
ual high concentrations of 1p in intestinal tissues were
thought to be sufficient to demonstrate proof of concept.