Synthesis of a new inhibitor of breast cancer resistance protein with significantly improved pharmacokinetic profiles

Article abstract of DOI:10.1016/j.bmcl.2015.11.077

The design, synthesis, in vitro inhibitory potency, and pharmacokinetic (PK) profiles of Ko143 analogs are described. Compared to commonly used Ko143, the new breast cancer resistance protein (BCRP) inhibitor (compound A) showed the same potency and a sig

Full text of DOI:10.1016/j.bmcl.2015.11.077

Bioorganic & Medicinal Chemistry Letters 26 (2016) 551–555
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of a new inhibitor of breast cancer resistance protein with
significantly improved pharmacokinetic profiles
b
c
a
a
a
Yuexian Li a, , Jiyeon Woo , Jessica Chmielecki , Cindy Q. Xia , Mingxiang Liao , Bei-Ching Chuang ,
⇑
Johnny J. Yang a, Miao Y. Guan a, Mihaela Plesescu a, Shimoga R. Prakash a
a Department of Drug Metabolism and Pharmacokinetics, Takeda Pharmaceuticals International Co., 35 Landsdowne Street, Cambridge, MA 02139, USA
b Boston University, 881 Commonwealth Avenue, Boston, MA 02215, USA
c Northeastern University, 360 Huntington Avenue, Boston, MA 02115, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The design, synthesis, in vitro inhibitory potency, and pharmacokinetic (PK) profiles of Ko143 analogs are
described. Compared to commonly used Ko143, the new breast cancer resistance protein (BCRP) inhibitor
(compound A) showed the same potency and a significantly improved PK profile in rats (lower clearance
[1.54 L/h/kg] and higher bioavailability [123%]). Ko143 on the other hand suffers from poor bioavailabil-
ity. Compared to Ko143, compound A would be a useful probe for delineating the role of BCRP during
in vivo studies in animals.
Received 16 September 2015
Revised 17 November 2015
Accepted 20 November 2015
Available online 22 November 2015
Keywords:
BCRP
Ó 2015 Elsevier Ltd. All rights reserved.
Inhibitors
Ko134
Ko143
PK profiles
Compound A
Resistance to anticancer agents is a major cause of chemother-
apeutic failure in cancer.1 The 75 kDa human breast cancer resis-
tance protein (BCRP) is a polytopic plasma membrane transport
protein that has been detected in many drug-resistant cell lines,
solid tumors, and hematological malignancies. BCRP is one of the
most important transporters involved in multidrug resistance
and drug–drug interactions (DDIs).1–3 Ko143 is a novel fumitre-
morgin C analog and a more potent and specific inhibitor than
other known inhibitors of BCRP such as Novobiocin, XR9576,
GF120918, Gefitinib, and Imatinib (Fig. 1).4,5 More importantly,
Ko143 is nontoxic at effective in vitro and in vivo concentrations,
which makes it one of the most promising compounds for the
development of modulators of BCRP-mediated efflux.4 In our
efforts to investigate the applications of BCRP inhibitors in the
development of new anticancer agents, we developed two syn-
thetic routes for the synthesis of Ko143.6 Unfortunately, Ko143
exhibits high clearance (65.9 L/h/kg) and low bioavailability (3%)
in rats (Table 1) perhaps due to hydrolysis of the ester moiety.
According to FDA guidance, both P-gp (P-glycoprotein) and BCRP
are expressed in the gastrointestinal tract, liver, and kidney, and
have a role in limiting oral bioavailability. Therefore, all investiga-
tional drugs should be evaluated in vitro to determine whether
they are potential substrates of P-gp or BCRP.7 For drugs that are
highly permeable and highly soluble, these requirements do not
apply.7 This exemption, however, FDA refers to does not apply to
a lot of new molecular entities (NMEs) which are highly permeable
but have low solubility.8 In addition to in vitro analysis, it is
equally important to check PK profiles of investigational drugs by
co-administration of BCRP specific inhibitors in animals.
Synthesis of new BCRP inhibitors: To explore the SAR of Ko143 ana-
logs, we prepared Ko134 (the demethoxy analog of Ko143) and its
analogs (Scheme 1). Pictet–Spengler condensation of commercially
available
followed by coupling of the resulting secondary amine 2 with
N-Fmoc-5-tert-butyl- -glutamic acid ester afforded 3. Fmoc-depro-
L-tryptophan methyl ester (1) with isovaleraldehyde
L
tection and subsequent intramolecular cyclization/cleavage of 3
yielded Ko134.6 Treatment of Ko134 with TFA to remove the
tert-butyl group gave acid 4.9 Amidation with tert-butylamine and
sulfamoyl chloride provided tert-butyl amide 5.10
To synthesize a new BCRP inhibitor with better PK profiles, we
decided to replace the long ester tail on the D ring of Ko143 with a
more stable group (Scheme 2). To explore the SAR of the analogs
with different electron-donating groups, we started with three
different indoles (6A, 6B, 6C). Coupling of 6-methoxy-indole 6 with
1-benzyl-2-methyl-(S)-1,2-aziridinedicarboxylate and ytterbium
⇑
Corresponding author. Tel.: +1 617 444 1275; fax: +1 617 444 1480.
E-mail address: yuexianfrank.li@takeda.com (Y. Li).
http://dx.doi.org/10.1016/j.bmcl.2015.11.077
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

552
Y. Li et al. / Bioorg. Med. Chem. Lett. 26 (2016) 551–555
O
NH
D
H
O
O
O
O
O
O
O
N
C
OH
N
H
O
H2N
O
O
A
B
OH
N
O
OH
H
O
Ko143
Novobiocin
O
O
N
O
O
O
O
N
H
O
O
NH
N
N
H
O
O
O
N
H
N
GF120918
XR9576
N
F
N
HN
O
Cl
N
N
H
N
H
O
O
N
N
N
N
N
O
Gefitinib
Imatinib
Figure 1. Structures of reported BCRP inhibitors.
Table 1
PK profiles of Ko143 in rats
10000
IV-2 mg/kg
PO-20 mg/kg
1000
100
10
1
0.1
0
4
8
12
16
20
24
Time (hrs)
Route
Dose (mg/kg)
T1/2 (h)
Tmax (h)
Cmax (nmol/L)
AUC0–24 h (h nmol/L)
CL (L/h/kg)
65.9
Vss (L/kg)
F (%)
iv-JVC
po-JVC
2
50
0.2
85.6
72.2
14.8
2
21.4
3
T1/2 = half life, Tmax = time of maximum drug concentration, Cmax = maximum drug concentration, AUC = area under the curve, CL = clearance, Vss = volume of distribution, F
(%) = bioavailability, iv = intravenous, JVC = jugular vein cannulation, po = per os (by mouth).
triflate yielded 6-methoxytryptophan derivative 7 followed by
deprotection gave amino acid 8.5,11 To explore the SAR of two
stereoisomers, we carried both isomers into the final step. More
cis product 9A (63.5% yield) than trans product 10B (15.1% yield)
was obtained during the Pictet–Spengler condensation by using
dry and pure starting material 8A and employing
a low

Y. Li et al. / Bioorg. Med. Chem. Lett. 26 (2016) 551–555
553
O
H
O
H
O
O
O
H
O
O
O
NH
NH2
ii
i
iii
HN
O
O
N
N
H
N
H
N
H
3
1
2
O
O
H
O
H
NH
O
O
H
NH
O
NH
O
O
OH
HN
N
iv
N
N
O
v
O
O
N
H
N
H
N
H
Ko134
4
5
Scheme 1. Reagents and conditions: (i) isovaleraldehyde, TFA, CH2Cl2, 0 °C, 2.5 h, 54.9%; (ii) N-Fmoc-5-tert-butyl-L-glutamic acid ester, diisopropylethylamine, 2-chloro-1,3-
dimethylimmidazolinium hexafluorophosphate, N-methylpyrrolidone, rt, 7 days, 37.8%; (iii) piperidine, THF, rt, 24 h, 60%; (iv) TFA, CH2Cl2, rt, 24 h, 89%; (v) tert-butylamine,
dimethylsulfamoyl chloride, DMAP, ACN, N,N-dimethyl-n-butylamine, 9%.
O
H
O
O
O
Cbz
NH2
iii
N
i
ii
R
O
R
O
R
O
H
H
N
H
N
H
N
H
6A: R = H
7A: R = H
8A: R = H
6B: R = OCH3
6C: R = CH3
7B: R = OCH3
7C: R = CH3
8B: R = OCH3
8C: R = CH3
O
O
O
NH
O
H
O
H
H
O
O
Boc
NH
R
O
NH
R
O
N
R
O
N
N
H
N
H
N
H
9A
9B
9C
13A: R = H, compound A
13B: R = OCH3
: R = H
: R = OCH
: R = CH
11A: R = H
11B: R = OCH3
11C: R = CH3
3
iv
v
13C: R = CH3
3
O
H
O
H
O
H
NH
O
O
O
O
Boc
NH
N
N
R
O
R
O
R
O
NH
N
H
N
H
N
H
10A: R = H
14A: R = H
12A: R = H
10B: R = OCH3
10C: R = CH3
14B: R = OCH3
14C: R = CH3
12B: R = OCH3
12C: R = CH3
Scheme 2. Reagents and conditions: (i) 1-benzyl-2-methyl-(S)-1,2-aziridinedicarboxylate, ytterbium triflate, CH2Cl2, rt, 24 h, 64.7% (7A), 48.6% (7B), 51.5% (7C); (ii) H2
balloon, MeOH, 10% Pd/C, 3 h; (iii) isovaleraldehyde, TFA, CH2Cl2, ꢀ25 to ꢀ30 °C, 1.5 h, 63.5% (9A, ii and iii), 14.9% (9B, ii and iii), 24.4% (9C, ii and iii), 21.2% (10A, ii and iii), 40%
(10B, ii and iii), 39.6% (10C, ii and iii); (iv) N-(tert-butoxycarbonyl)-L-alanine, HATU, DMF, N,N-diisopropylethylamine, rt, 4 days, 83.4% (11A), 90% (11B), 72.8% (11C), 81.3%
(12A), 90% (12B), 72.8% (12C); (v) (1) TFA, CH2Cl2, rt, 2 h, (2) saturated NaHCO3, CH2Cl2, rt, 2 h, 42.2% (13A, compound A), 27.4% (13B), 24% (13C), 50% (14A), 43.6% (14B), 25.2%
(14C).
temperature (ꢀ25 to ꢀ30 °C).12 Instead of using Fmoc protected
amino acid as in Scheme 1, more commonly used and less expen-
sive Boc-protected amino acids were utilized for the step iv. To
couple this secondary amine 9 (or 10), HATU (O-(7-azabenzotria-
zol-1-yl)-N,N,N0,N0-tetramethyluronium
was found to be better than BEP (2-bromo-1-ethyl-pyridinium
tetrafluoroborate), CIP (2-chloro-1,3-dimethylimmidazolinium
hexafluorophosphate), oxalyl chloride, or thionyl chloride.6,13,14
hexafluorophosphate)

554
Y. Li et al. / Bioorg. Med. Chem. Lett. 26 (2016) 551–555
Of the BOC protected amino acids subjected to the reaction condi-
tions (Ala, His, Asp, Ser, and Gln), only Boc-protected alanine pro-
vided practical yield of 83.4% (for 11A). The coupling was lengthy
(4 days) because of the less reactive secondary amines and steric
hindrance. Microwave technology was found to be of no advantage
for the coupling of 9 (or 10).15 De-protection of 11 (or 12) with TFA
followed by an intramolecular cyclization/cleavage of the resulting
amine gave final product 13 (or 14). The cyclization was completed
in 2 h by simply washing the CH2Cl2 solution with saturated
NaHCO3.16 The overall yield of compound A was 14% in five steps.
Potencies and ADME data of the synthesized BCRP inhibitors: In
order to determine the inhibitory potency of the new compounds,
the IC50s of these compounds on BCRP mediated E3S (Estrone sul-
fate) were measured on Caco-2 cells. Ko134 (a tert-butyl ester) and
the tert-butyl amide 5 showed similar potency on BCRP efflux
transporter, but the acid 4 failed to show inhibition up to 10 lM
(Table 2).17 This indicates that the ester portion is not necessary
for the potency and we could use more stable and simple groups,
such as a methyl group, to replace it. Compound A showed almost
the same potency as that of Ko143. But the trans-isomer, 14A,
Figure 2. Superimpositions of conformations of compound A (purple) and trans-
isomer 14A (green).
Table 2
failed to exhibit activity up to 10 lM (Table 2). This is because of
IC50 on BCRP efflux transporter
the obvious difference in conformations of two compounds with
a change at one stereo center (Fig. 2). By introducing a methoxy
group, Ko143 is about 14-fold more potent than Ko134 (Table 2).
But the potency disappears after adding the 2nd methoxy group.
The potency is similar to that of Ko143 after introducing a methyl
group. Like 14A, both trans isomers (14B, 14C) failed to exhibit
Compound
IC50
(
lM)
IC50 of Ko143 (lM)
Ko134
4
5
Compound A
14A
13B
14B
13C
3.25
>10
3.87
0.60
>10
>10
>10
1.98
>10
0.23
0.23
0.23
0.61
0.61
1.70
1.70
1.81
1.81
activity up to 10 lM (Table 2). The observed differences in IC50 val-
ues of Ko143 were caused by the inter-assay variability existed in
Caco-2 cell assays. Neither compound A nor Ko143 inhibits P-gp
transporter, and both were selective BCRP inhibitors (Table 3).17
Compared to Ko143 (Table 1), compound A exhibited much better
PK profiles in rats (more than 42-fold lower in clearance and more
than 100% of relative bio-availability) (Table 4).18
14C
Test system: Caco-2-cells, substrate: 3H-Estron-3-sulfate (26 nM).
Note: IC50 of Ko143 was determined along with each synthesized compound
because of the variability in actual determinations of IC50
.
In summary, we have described the design, synthesis, inhibitory
potencies, and PK data of Ko143 analogs. The new BCRP inhibitor
(compound A) was evaluated against Ko143, the commonly used
BCRP inhibitor. Compared to Ko143, compound A showed similar
potencies and a significantly improved PK profile (42-fold lower
clearance and 100% of relative bio-availability) in rats. Compound
A is thus a useful probe for delineating the role of BCRP in vivo.
Table 3
IC50 on P-gp transporter
Compound
IC50
Ko143 (
>30
lM)
Compound A (
>30
lM)
Test system: Caco-2-cells.
Table 4
PK profiles of compound A in rats
1000
100
10
IV-2 mg/kg
PO-50 mg/kg
1
0
0
2
4
6
Time (hr)
Route
Dose (mg/kg)
T1/2 (h)
Tmax (h)
Cmax (nmol/L)
AUCall (h nmol/L)
CL (L/h/kg)
1.54
Vss (L/kg)
F (%)
iv-JVC
po-JVC
2
20
0.9
2.0
4000
49,000
1.64
1.3
10,100
123
See Table 1 for abbreviations.

Y. Li et al. / Bioorg. Med. Chem. Lett. 26 (2016) 551–555
555
14. Wu, G.; Liu, J.; Bi, L.; Zhao, M.; Wang, C.; Baudy-Floc’h, M.; Ju, J.; Peng, S.
Tetrahedron 2007, 63, 5510.
Acknowledgements
15. Cagide, F.; Reis, J.; Gaspar, A.; Borges, F. Tetrahedron Lett. 2011, 52, 6446.
16. Hernadez, D.; Nielsen, L.; Lindsay, K. B.; Lopez-Garcia, M. A.; Bjerglund, K.;
Skrydstrup, T. Org. Lett. 2010, 12, 3528.
Thanks are due to Scott Rowland of Computation Chemistry
Department of Takeda Pharmaceuticals International Co. for
providing conformational superimpositions, and Mingkun Fu of
Pharmaceutical Profiling Group of Takeda Pharmaceuticals
International Co., for providing HRMS data.
17. Determination of IC50 on BCRP and P-gp transporters: To determine potency of
compounds on inhibition of transporters, bi-directional transport studies are
performed in Caco-2 cells (American Type Culture Collection, Manassas, VA) at
37 °C, according to Xia et al. (Expression, localization and functional
characteristics of breast cancer resistance protein in Caco-2 cells. Drug Met.
Disp. 2005, 33, 637). Prior to each experiment, the confluent cell monolayers on
TranswellTM inserts are washed and equilibrated for 30 min with transport
media (Hank’s balanced salt solution (HBSS) containing 10 mM of N-2-
hydroxyethyl-piperazine-N0-2-ethanesulfonic acid (HEPES) and 10 mM of
glucose, pH 7.4). The experiment is initiated by adding a solution containing
the 3H-digoxin (25 nM, substrate for P-gp), or 3H-E3S (Estrone sulfate, 17 nM,
substrate for BCRP) to either the apical (for A-to-B transport) or basolateral (for
B-to-A transport) compartment in the absence or presence of various
concentrations of compound A or Ko143. At 30, 60, 90, and 120 min, 0.05 mL
aliquot of receiving solutions are sampled from the basolateral side (for A-to-B
transport) or from the apical side (for B-to-A transport), and replaced
immediately with an equal amount of fresh transport media. The
Supplementary data
Supplementary data (experimental procedures of the synthe-
sized compounds) associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2015.11.
077.
References and notes
radioactivity in each sample is measured by 1450 MicroBeta TriLux,
a
1. Rosenberg, M. K.; Bikadi, Z.; Chan, J.; Liu, X.; Ni, Z.; Cai, X.; For, R. C.; Mao, Q.
Structure 2010, 18, 482.
2. Xia, C. Q.; Liu, N.; Yang, D.; Miwa, G.; Gan, L. Drug Metab. Dispos. 2005, 33, 637.
3. Giacomin, K. M.; Huang, S. M.; Tweedie, D. J.; Bent, L. Z.; Brouwer, K. L. Nat. Rev.
Drug Disc. 2010, 9, 215.
4. Allen, J. D.; van Loevezijn, A.; Lakhai, J. M.; van der Valk, M.; van Tellingen, O.;
Reid, G.; Schellens, J. H. M.; Koomen, G. J.; Schinkel, A. H. Mol. Cancer Ther. 2002,
1, 417.
5. Pick, A.; Muller, H.; Wiese, M. Bioorg. Med. Chem. Lett. 1999, 20, 180.
6. Li, Y.; Hayman, E.; Plesescu, M.; Prakash, S. R. Tetrahedron Lett. 2010, 49, 1480.
7. Food and Drug Administration, Guidance for Industry, Drug Interaction Studies—
Study Design, Data Analysis, Implications for Dosing, and Labeling
Recommendations February 2012.
microplate scintillation and luminescence counter (PerkinElmer Life
Sciences). Radioactivity (3H) of the dosing solution is measured and used to
calculate the initial donor concentration of the substrate.
18. Determination of in vivo pharmacokinetic profile: To determine pharmacokinetic
profile of compound A and Ko143 in vivo, Sprague-Dawley rats (Charles River
Labs, Worcester, MA) are administered 2.0 mg/kg or 20 mg/kg compound A or
2.0 mg/kg or 50 mg/kg Ko143, formulated in 0.5% HPMC/0.2% Tween80, via iv
or po, respectively. After administration of compound A or Ko143, blood is
obtained from all animals at predose and at 0.083, 0.25, 0.5, 1, 4, 8, and 24 h
postdose. Approximately 200 lL of whole blood is collected from the jugular
vein catheter of each animal into tubes containing the anticoagulant
dipotassium ethylenediaminetetraacetic acid (K2EDTA) and is further
processed into plasma at approximately 4 °C. The concentration of
8. Wu, C. Y.; Benet, L. Z. Pharm. Res. 2005, 22, 11.
compound
A or Ko143 in plasma is determined using LC/MS/MS.
9. Chen, Z.; Fan, J.; Kende, A. S. J. Org. Chem. 2004, 69, 79.
10. Wakasugi, K.; Nakamura, A.; Tanabe, Y. Tetrahedron Lett. 2001, 42, 7427.
11. Nishikawa, T.; Kajii, S.; Wada, K.; Ishikawa, M.; Isobe, M. Synthesis 2002, 1658.
12. Bailey, P. D.; mcLay, N. R. J. Chem. Perkin Trans. 1 1993, 441.
13. He, Y.; Wilkins, J. P.; Kiessling, L. L. Org. Lett. 2006, 8, 2483.
Pharmacokinetic analysis of the individual plasma concentration data is
performed using standard non-compartmental methods (PhoenixTM
WinNonlinÒ, Version 6.1 [Pharsight Corp (Mountain View, CA, USA)]).