Discovery of microsomal triglyceride transfer protein (MTP) inhibitors with potential for decreased active metabolite load compared to dirlotapide

Article abstract of DOI:10.1016/j.bmcl.2011.05.099

Analogues related to dirlotapide (1), a gut-selective inhibitor of microsomal triglyceride transfer protein (MTP) were prepared with the goal of further reducing the potential for unwanted liver MTP inhibition and associated side-effects. Compounds were designed to decrease active metabolite load: reducing MTP activity of likely human metabolites and increasing metabolite clearance to reduce exposure. Introduction of 4′-alkyl and 4′-alkoxy substituents afforded compounds exhibiting improved therapeutic index in rats with respect to liver triglyceride accumulation and enzyme elevation. Likely human metabolites of select compounds were prepared and characterized for their potential to inhibit MTP in vivo. Based on preclinical efficacy and safety data and its potential for producing short-lived, weakly active metabolites, compound 13 (PF-02575799) advanced into phase 1 clinical studies.

Full text of DOI:10.1016/j.bmcl.2011.05.099

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4150–4154
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of microsomal triglyceride transfer protein (MTP) inhibitors
with potential for decreased active metabolite load compared to dirlotapide
⇑
Ralph P. Robinson , Jeremy A. Bartlett, Peter Bertinato, Andrew J. Bessire, Judith Cosgrove, Patrick M. Foley,
Tara B. Manion, Martha L. Minich, Brenda Ramos, Matthew R. Reese, Theodore J. Schmahai,
Andrew G. Swick 1, David A. Tess, Alfin Vaz, Angela Wolford
Pfizer Global Research and Development, Groton Laboratories, Pfizer Inc., Groton, CT 06340, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Analogues related to dirlotapide (1), a gut-selective inhibitor of microsomal triglyceride transfer protein
(MTP) were prepared with the goal of further reducing the potential for unwanted liver MTP inhibition
and associated side-effects. Compounds were designed to decrease active metabolite load: reducing
MTP activity of likely human metabolites and increasing metabolite clearance to reduce exposure. Intro-
duction of 40-alkyl and 40-alkoxy substituents afforded compounds exhibiting improved therapeutic
index in rats with respect to liver triglyceride accumulation and enzyme elevation. Likely human metab-
olites of select compounds were prepared and characterized for their potential to inhibit MTP in vivo.
Based on preclinical efficacy and safety data and its potential for producing short-lived, weakly active
metabolites, compound 13 (PF-02575799) advanced into phase 1 clinical studies.
Received 18 April 2011
Accepted 25 May 2011
Available online 2 June 2011
Keywords:
MTP
Obesity
Dirlotapide
Metabolite
Soft drug
Ó 2011 Elsevier Ltd. All rights reserved.
Dirlotapide (1) is a potent gut-selective inhibitor of microsomal
triglyceride transfer protein (MTP).1–4 As such it inhibits fat
absorption by blocking the assembly of triglyceride-rich chylomi-
crons within enterocytes. Dirlotapide has demonstrated robust
weight loss efficacy in animals. For example, in a three month
efficacy study in dogs, a mean weight loss of nearly 20% was
achieved.1 The profound weight loss efficacy of dirlotapide is
attributed not only to reduced fat absorption, but also to an ano-
rectic effect, which is believed to result from the observed release
of anorectic peptides (e.g., PYY and GLP-1) from fat-filled cells in
the intestine. The compound is approved by the Food and Drug
Administration for the treatment of obesity in dogs, marketed
under the brand Slentrol.
Although dirlotapide also demonstrated significant weight loss
efficacy in human clinical trials, serum transaminase elevations
limited the dose and the efficacy that could be safely achieved.5
Similar enzyme elevations have been reported for structurally di-
verse MTP inhibitors targeted to the liver for the treatment of ath-
erosclerosis and hyperlipidemia. This effect is reversible and is
believed to be mechanism-related, a consequence of triglyceride
(TG) accumulation within the liver due to the inhibition of liver
MTP.6 Evidently, dirlotapide is not gut-selective in humans, even
though only a small fraction of an orally administered dose is
absorbed.
In humans, dirlotapide gives rise to metabolites, specifically the
N-dealkylation products 2 and 3, which accumulate and circulate
at concentrations 2–7 fold higher than parent (unpublished re-
sults). These compounds also circulate as major metabolites in rats
and dogs.7 Since 2 and 3 retain activity against MTP (unbound
potencies ꢀ30-fold less than parent, Table 1), it was postulated
that these contribute to the overall ‘MTP inhibitory load’ on the
liver and to the associated serum liver enzyme elevations. That
circulating active metabolites may contribute to the undesired
side-effects of dirlotapide suggested strategies that, alone or in
combination, could be applied in the design of new MTP inhibitors
with improved gut selectivity: 1) designing compounds expected
to give rise to metabolites with reduced activity against MTP,
and 2) designing compounds expected to produce metabolites
with increased clearance that do not accumulate. This letter
describes the successful effort to identify compounds having the
potential for increased gut selectivity based on these strategies.
CF3
CF3
Me O
N
Me
N
Me O
N
H
N
O
N
H
O
N
H
R
O
O
N
H
N
H
1 (dirlotapide)
2: R = Me
3
: R = H
⇑
Corresponding author. Tel.: +1 860 441 4923; fax: +1 860 686 6010.
E-mail address: ralph.p.robinson@pfizer.com (R.P. Robinson).
Current address: University of North Carolina Chapel Hill Nutrition Research
1
Institute, Kannapolis, NC 28081.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.05.099

R. P. Robinson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4150–4154
4151
Table 1
R
F
Me
N
O
O
N
H
O
N
H
N
4-14
R
ApoB IC50,
apparent
nM SD (n)
ApoB fraction
unbound
ApoB IC50
unbound pM
,
Dose
mg/kg
D
Food
D
Body Wt%
D
Liver TG (%)
a
a
Intake (%)
versus veh.
versus veh.
versus veh.a
1
2
3
4
5
6
7
8
9
10
N/A
N/A
N/A
CF3
H
Me
Et
n-Pr
i-Pr
t-Bu
2.7 5.2 (154)
1.8 0.56 (3)
1.5 (1)
0.74 (2)
2.8 (1)
2.2 (1)
0.55 (2)
0.44 (1)
0.60 (2)
0.0042
0.21
0.23
11.3
378
345
1.0
10
10
10
10
ꢁ11 to ꢁ50
ꢁ40 (ꢁ43)
ꢁ37 (ꢁ43)
ꢁ43* (ꢁ11)
ꢁ8 to ꢁ117
70–216
ꢁ106 (ꢁ115)
ꢁ108 (ꢁ115)
ꢁ96* (ꢁ8)
147 (112)
87 (112)
88* (70*)
0.0014
0.0043
9.5
10
ꢁ38 (ꢁ50*)
ꢁ75* (ꢁ117*)
43 (216*)
0.0011
0.00063
0.7
0.4
10
ꢁ38* (ꢁ11)
ꢁ60* (ꢁ46)
ꢁ12 (ꢁ5)
ꢁ91* (ꢁ8)
16 (70*)
1.2 3.8 (86)
10
ꢁ123* (ꢁ97*)
ꢁ32 (ꢁ25)
ꢁ67* (ꢁ10)
ꢁ96* (ꢁ32)
ꢁ89* (ꢁ75*)
ꢁ107* (75*)
ꢁ55 (ꢁ117)
unch. (120*)
unch. (38)
unch. (22)
36 (102)
1b
b
3 (MED)
10b
30b
100b
10
ꢁ36* (ꢁ6)
ꢁ33* (ꢁ8)
ꢁ41* (ꢁ37*)
ꢁ47* (ꢁ35*)
ꢁ27 (ꢁ50)
102 (136*)
119* (152*)
11
12
13
OMe
OEt
O-i-Pr
3.0 (1)
2.8 (1)
0.77 0.29 (15)
0.0038
2.0
10
3 (MED)
10
30
10
ꢁ37* (ꢁ38*)
ꢁ23*
ꢁ56 (ꢁ95*)
ꢁ65
16 (112*)
unch.
ꢁ30* (ꢁ28*)
ꢁ45*
ꢁ82 (ꢁ88*)
ꢁ152*
unch. (111*)
15
14
O-t-Bu
0.83 (2)
ꢁ40* (ꢁ38*)
ꢁ108* (ꢁ95*)
72* (112*)
*
Significantly different from vehicle (P 6 0.05).
a
In parentheses:% for compound 1, run as positive control.
25% SDD formulation; compound 1 (comparator) 95% SDD.
b
The dirlotapide metabolites 2 and 3 presumably accumulate
due to a lack of ‘soft sites’ for further oxidative metabolism. Thus,
our primary chemistry strategy was to incorporate an additional
site of metabolism on the left hand portion of 4. Even if N-dealky-
lation should occur first, the resulting metabolites would still be
susceptible to further transformation to more polar, less active
and more readily eliminated secondary metabolites. Modification
of the 40-trifluoromethyl group appeared particularly attractive
with the reasonable expectation that lipophilic replacements for
this group (e.g., alkyl, alkoxy) would retain MTP potency while
polar groups (such as those produced by oxidative metabolism of
40-alkyl and 40-alkoxy) would show reduced activity against the
target.
For a primary screen, we used the well-established cell-based
assay measuring the ability of compounds to inhibit apoB secretion
from human hepG2 cells.8 For select compounds, the extent of pro-
tein binding under the conditions of the apoB assay was measured
by equilibrium dialysis9, allowing free MTP potencies to be com-
pared. To evaluate efficacy and the potential for gut selectivity, po-
tent in vitro compounds were tested in rats to measure anorectic
activity and liver TG accumulation. Thus, rats placed on a high
fat (45%) diet were given daily oral doses of compound, adminis-
tered in a self-emulsifying drug delivery system (SEDDS) vehicle
for three days.10 During this period, food intake was measured
and afterwards, liver TG accumulation was determined and used
to rank order compounds for the potential to produce liver enzyme
elevations. In this model, even sub-efficacious doses of 1 produced
statistically significant increases in liver TG.
Although there is no evidence to suggest structure-specific tox-
icity associated with dirlotapide, we chose as our chemical starting
point the aminoquinoline analogue 4 that had been prepared as
part of the original research effort leading to 1. This compound dis-
plays very high potency against MTP and had the advantage of not
falling within the scope of corporate ‘structural alerts’ developed as
a guide to avoid possible class-related toxicity attributed to certain
functional groups (in the case of dirlotapide, an aniline-like core).
Like 1, compound 4 showed no apparent therapeutic index be-
tween efficacy and elevation of liver TG in the rat model.
First we prepared a series of analogues of 4 (Scheme 1)11 in
which the 40-trifluoromethyl group is replaced by hydrogen (5)
or small alkyl groups (e.g., 6 to 10, Table 1). Simple replacement
of the 40-trifluoromethyl group with hydrogen or alkyl provided
compounds that retained potent MTP activity, and in the case of
6, 9 and 10, robust in vivo efficacy on par with or greater than 1
and 4. Free potencies appeared to increase with alkyl chain size
as suggested by the in vitro data for 6, 9 and 10. These three com-
pounds showed potential for gut selectivity judging from reduced
TG elevation relative to 1 and 4 at 10 mg/kg.
Next, we prepared compounds incorporating 40-alkoxy groups
(Scheme 1). With the exception of O-t-Bu (14), these offered the
possibility that oxidative O-dealkylation would rapidly produce
the corresponding phenol, which in turn would be susceptible to
clearance via phase II metabolism. Based on the decreased
in vivo efficacy of 11 relative to 1 (e.g., 27% reduction in food take
versus 50% decrease for 1 at 10 mg/kg), 40-OMe did not appear to
be optimal. On the other hand, very promising compounds were
identified from the 40-O-i-Pr series, for example, 13, which showed
CF3
F
O
Me
N
O
N
H
O
N
H
N
4

4152
R. P. Robinson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4150–4154
R
R
CO2H
CO2Bn
27
CO2H
CO2H
O
a-c
H2N
N
d, e
N
H
N
N
26
j
F
h, i
NH2 f, g
NHMe
Me
N
4-14
F
F
HCl.H2N
O
Scheme 1. Reagents and conditions: (a) carbonyl diimidazole/EtOAc, then BnOH, 23 °C; (b) ureaꢂH2O2/CH2Cl2/phthalic anhydride, 23 °C; (c) p-TsCl/CH2Cl2, then NH4Cl/Et3N,
23 °C; (d) 26/EDC/DMAP/CH2Cl2, 23 °C; (e) LiOHꢂH2O/MeOH/THF/H2O, 23 °C; (f) EtOCHO/reflux; (g) LiAlH4/THF, 0–23 °C; (h) N-BOC-
L-phenylglycine/PyBrOP/i-Pr2NEt/CH2Cl2,
23 °C; (i) HCl/dioxane, 0–23 °C; (j) EDC/Et3N/DMAP/CH2Cl2, 23 °C.
Table 2
Dose mg/kg (po)
D
Food Intake (%) versus veh.*
D
Body Wt% versus veh.*
D
Liver TG (%) versus veh.*
D
ALT % versus veh.*
AUC ng h/mL
Dirlotapide (1)
3
ꢁ11
ꢁ120
479*
685*
770*
122
37
75
331
10 (MED)
30
Compound 10
3 (MED)
10
ꢁ36*
ꢁ38*
ꢁ269*
ꢁ262*
1962*
997*
ꢁ25*
ꢁ51*
ꢁ48*
ꢁ85
43
168
143
237
12
42
65
ꢁ255*
ꢁ220*
219*
212*
30
Compound 13
10 (MED)
30
100
ꢁ21*
ꢁ25*
ꢁ43*
ꢁ82
113*
184*
303*
10
51
47
170
ꢁ148*
ꢁ178*
191
629*
*
Significantly different from vehicle (P 6 0.05).
efficacy equivalent to 1 and 4 and did not elicit appreciable liver TG
elevations at 10 mg/kg.
basis, the TI’s of compounds 10 and 13 with respect to efficacy
and ALT are ꢀ10-fold (30/3 and 100/10). Using systemic exposure
data (AUC’s), the TI’s decrease to >5-fold (65/12) for 10 and ꢀ3 fold
(170/51) for 13.
Based on their promising in vitro and in vivo profiles, com-
pounds 10 and 13 were advanced to further define their efficacy
and potential for improved gut-selectivity relative to 1. Because
the SEDDS formulation used for in vivo screening is not an accept-
able formulation for human clinical trials, the compounds were
formulated as 25% spray-dried dispersions (SDD) prior to carrying
out definitive preclinical efficacy and safety studies.12 Preliminary
assessment of SDD-formulated 10 in the routine three-day rat
model (Table 1) revealed good efficacy (minimally effective dose,
MED = 3 mg/kg) and evidence of gut selectivity as previously ob-
served using SEDDS. Next, SDD-formulated 10 and 13 were tested
alongside 1 (95% SDD) in a seven-day rat model in which changes
in food intake, body weight, liver TG, as well as serum alanine
transaminase (ALT) were measured. In this definitive study, dose-
effect relationships were examined (1 and 10 dosed at 3, 10, and
30 mg/kg q.d., 13 dosed at 1, 3, 10, 30 and 100 mg/kg q.d.) and sys-
temic exposures were measured (Table 2).
In agreement with the results in Table 1, compound 1 showed
no evidence of gut selectivity in the definitive study, either on
the basis of liver TG or serum ALT elevations. With respect to effi-
cacy and ALT, the therapeutic index (TI) of 1 was ꢀ1 based on dose
and exposure. Although its minimally efficacious dose (MED) in the
seven day study remained unchanged (3 mg/kg), compound 10
elicited statistically significant effects on liver TG at a lower dose
than before (10 mg/kg) while compound 13 produced appreciable
TG effects at its MED (10 mg/kg). Importantly, however, 10 showed
no increase in ALT at any dose, while statistically significant in-
crease of ALT was observed only at 100 mg/kg with 13. On a dose
Metabolite identification studies for 10 and 13 in vitro (rat, dog,
and human microsomes) and for 10 in vivo (rat and dog) confirmed
that, as expected, both compounds undergo extensive CYP-medi-
ated oxidation (hydroxylation or O-dealkylation) at the 40-position
accompanied by N-dealkylation (demethylation and debenzyla-
tion) of the terminal L
-phenylglycinamide.13 Importantly, for both
compounds, the metabolite profile observed in rats was quite sim-
ilar to that observed in human liver microsomes. Thus, provided
the metabolites were shown not accumulate in vivo or were, in
general, less active than parent drug, it would be reasonable to
hypothesize that the improved TI observed in rats was due to re-
duced metabolite load compared to 1 and that the improved TI
should translate to humans.
We prepared a number of metabolites of 10 and 13 as standards
for PK studies and for in vitro characterization (Table 3, Schemes 2
and 3): 15, 16, 21, 22, 23–formed from parent by 40-oxidation; 19,
20, 24, 25–formed from parent by N-dealkylation, and 17,
18–formed from 10 by a combined 40-oxidation and N-dealkyla-
tion.14 With the exception of the N-demethylated metabolites 20
and 25 (which themselves are further metabolized at the 40-posi-
tion), all those profiled exhibited at least ꢀ30-fold loss of free po-
tency with respect to parent. As expected, the two metabolites
formed by reaction at two sites (17 and 18) were less potent than
the corresponding single-site product (compare 17 to 15 and 18 to
16). Generally speaking, however, except for the carboxylic acids
18 and 23, the loss of potency associated with metabolism at the

R. P. Robinson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4150–4154
4153
Table 3
R1
R2
O
N
O
N
R3
H
O
N
H
N
Cmpd
R1
R2
R3
ApoB IC50
apparent nM SD (n)
,
ApoB fraction
unbound
ApoB IC50
unbound pM
,
Fold decrease free
potency relative
to parent
10 and metabolites
10
15
16
17
18
19
20
t-Bu
Me
Me
Me
H
H
Me
H
4-F-Bn
4-F-Bn
4-F-Bn
4-F-Bn
4-F-Bn
H
1.2 3.8 (86)
1.1 0.35 (10)
7.5 2.2 (9)
4.3 0.52 (3)
12 2.4 (3)
0.56 0.15 (12)
1.2 (1)
0.00063
0.023
0.76
25
66
58
253
21
1
CMe2CH2OHa
CMe2CO2Ha
CMe2CH2OH
CMe2CO2H
t-Bua
33
87
76
333
28
4
0.0088
0.0135
0.0211
b
b
0.038
0.0023
t-Bu
4-F-Bn
2.8
13 and metabolites
13
21
22
23
24
25
O-i-Pr
OH
OCH(Me)CH2OH
OCH(Me)CO2H
O-i-Pr
Me
Me
Me
Me
Me
H
4-F-Bn
4-F-Bn
4-F-Bn
4-F-Bn
H
0.77 0.29 (15)
11 (1)
9.4 (1)
22 (1)
0.63 0.14 (4)
1.21 0.92 (3)
0.0038
0.0095
0.013
2.9
1
105
122
484
101
7.4
36
42
167
35
3
0.022
0.160
O-i-Pr
4-F-Bn
0.0061b
a
Major circulating metabolite (>3% parent response) in rat and dog in vivo.
Predicted using a computational model developed specifically for HepG2 apoB protein binding.
b
OTBDMS
CO2H
OTBDMS
a-c
d, e
B
CO2H
O
O
Br
F
f
CO2K
F
O
R
26
N
H2N
N
N
H
g
O
H2N
N
h
R
N
HCl.H2N
OTBDMS
O
F
O
O
R
N
N
H
j
i
O
16
15
(R = Me)
(R = Me)
N
N
H
18 (R = H)
17 (R = H)
Scheme 2. Synthesis of 40-metabolites of 10. Reagents and conditions: (a) LiAlH4/Et2O, ꢁ78 °C, warm to 23 °C; (b) TBDMSCl/imidazole/CH2Cl2, 23 °C; (c) n-BuLi/THF, ꢁ78 °C,
then 2-isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, warm to 23 °C; (d) methyl 2-iodobenzoate/Pd(PPh3)4/Cs2CO3/DME, 100 °C; (e) LiOHꢂH2O/MeOH/THF/H2O, 23 °C;
(f) KOH/i-PrOH/H2O, 85 °C; (g) EDC/THF, 23 °C; (h) EDC/DMAP/THF, 23 °C; (i) TBAF/AcOH/THF, 23 °C; (j) Jones reagent/acetone, 23 °C.
40-position was not as dramatic as was hoped for, being in order of
magnitude seen for metabolism of 1 to give 2 and 3. Therefore, it
OH
F
Me
N
O
was important to demonstrate a reduced capacity of metabolites
of 10 and 13 to accumulate in vivo relative to dirlotapide metabo-
lites 2 and 3.
a
O
N
H
11
O
N
H
N
Indeed, following single dose oral administration of 10 (25%
SDD) to rats (30 mg/kg) and beagle dogs (2.5 mg/kg), circulating
exposures (AUC) of the major metabolites 15, 16, and 19 were
20% or less than that of parent in both species as determined using
synthetic standards. The half-lives of these three metabolites in
rats and dogs were approximately equivalent to parent due to for-
mation rate limited clearance. Systemic exposure to parent drug
was very low for 10 and 13 in both species (bioavailabilites <3%).
We did not perform a parallel study with 13 to measure metabolite
levels in vivo, making the reasonable assumption based on the sim-
b
O
CO2R
O
F
Me
O
c
22
23
N
N
(R = Me)
H
O
N
N
d
H
(R = t-Bu)
Scheme 3. Synthesis of 40-metabolites of 13. Reagents and conditions: (a) BBr3/
CH2Cl2, ꢁ78 °C warm to 23 °C; (b) NaH/methyl or t-butyl 2-bromopropionate/DMF,
23 °C (c) NaBH4/MeOH/THF, 55 °C; (d) TFA/CH2Cl2, 23 °C.

4154
R. P. Robinson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4150–4154
4. For a review on MTP inhibitors, see: Williams, S. J.; Best, J. D. Expert Opin. Ther.
Pat. 2003, 13, 479.
5. https://www.slentrol.com/DisplayIn.aspx (March, 2011).
ilarity of structures that metabolites of 13 would behave similarly
to those of 10 in vivo.
Based on their potential for an improved therapeutic index over
dirlotapide (1) as demonstrated in preclinical models and the pre-
dicted reduced systemic exposure to MTP-active species (parent
drug and metabolites), compounds 10 and 13 progressed to regu-
latory toxicology studies to support further development. Showing
a clear safety advantage over 10 in regulatory safety studies, com-
pound 13 (PF-02575799) advanced into human phase 1 clinical
studies for treatment of obesity.
6. (a) Joy, T. R.; Hegele, R. A. Nat. Clin. Prac. Cardiovasc. Med. 2008, 5, 506; (b)
Burnett, J. R.; Watts, G. F. Expert Opin. Ther. Targets 2007, 11, 181.
7. Compound 3 does not circulate as a major metabolite in dog. Dog has metabolic
processes to limit circulating active metabolites.
8. Bertinato, P.; Couturier, M. A.; Hamanaka, E. S.; Ewing, M. D.; Robinson, R. P., Jr.;
Tickner, D. L. Preparation of substituted quinolines as MTP/Apo-B secretion
inhibitors for treating obesity and associated conditions. WO 2005080373,
2005.
9. Maurer, T. S.; DeBartolo, D. B.; Tess, D. A.; Scott, D. O. Drug Met. Disp. 2005, 33,
175.
10. Male Sprague–Dawley rats (ꢀ300 g; n = 5 per group), acclimatized over seven
days to a high fat diet (45% kcal from fat, Research Diets D01050602 M),
received three consecutive daily doses of vehicle (SEDDS:H2O 10:90) or
compound in the same vehicle by standard oral gavage at 1.5 h prior to the
dark cycle. The SEDDS vehicle was made up of MiglyolÒ 812 (10% by volume),
triacetin (30%), polysorbate 80 (20%), and CapmulÒ MCM (30%).
11. Ref. 8 provides detailed experimental procedures for the preparation of 10 as
well as spectroscopic data for compounds 4, 6–14, 19, 20 and 25.
12. Spray dried dispersions were prepared by spray drying a solution containing
test compound and polymer (hydroxypropylmethylcellulose acetate
succinate). For details, see Ref. 8
Acknowledgements
Thanks to David Griffith for review of the manuscript and very
helpful comments.
References and notes
1. Li, J.; Bronk, B. S.; Dirlam, J. P.; Blize, A. E.; Bertinato, P.; Jaynes, B. H.; Hickman,
A.; Miskell, C.; Pillai, U. A.; Tibbitts, J. S.; Haven, M. L.; Kolosko, N. L.; Barry, C. J.;
Manion, T. B. Bioorg. Med. Chem. Lett. 2007, 17, 1996.
2. Wren, J. A.; Gossellin, J.; Sunderland, S. J. J. Vet. Pharmacol. Ther. 2007, 30, 11.
3. For other efforts to discover gut-selective MTP inhibitors, see: (a) Vu, C. B.;
Milne, J. C.; Carney, D. P.; Song, J.; Choy, W.; Lambert, P. D.; Gagne, D. J.; Hirsch,
M.; Cote, A.; Davis, M.; Lainez, E.; Meade, N.; Normington, K.; Jirousek, M. R.;
Perni, R. B. Bioorg. Med. Chem. Lett. 2009, 19, 1416; (b) Mera, Y.; Odani, N.;
Kawai, T.; Hata, T.; Suzuki, M.; Hagiwara, A.; Katsushima, T.; Kakutani, M. J.
Pharmacol. Exp. Ther. 2011, 336, 321.
13. Metabolite identification was not performed for 13 in vivo. For both 10 and 13,
hydroxylation of the biphenyl group also occurred to afford phenolic
metabolites, with and without accompanying 40-oxidation and N-
dealkylation. These were not profiled.
14. N-dealkylated metabolites 19, 20, 24 and 25 were prepared in a manner similar
to that in Scheme 1 starting with the amide coupling of N-BOC-L-phenylglycine
to methylamine hydrochloride (to afford 19 and 24) or 4-fluorobenzylamine
(to afford 20 and 25).