Design and synthesis of a hybrid potentiator-corrector agonist of the cystic fibrosis mutant protein ΔF508-CFTR

Article abstract of DOI:10.1016/j.bmcl.2009.11.020

A developing therapy of cystic fibrosis caused by the ΔF508 mutation in CFTR employs correction of defective CFTR chloride channel gating by a 'potentiator' and of defective CFTR protein folding by a 'corrector'. Based on SAR data for phenylglycine-type potentiators and bithiazole correctors, we designed a hybrid molecule incorporating an enzymatic hydrolysable linker to deliver the potentiator (PG01) fragment 2 and the corrector (Corr-4a) fragment 13. The hybrid molecule 14 contained PG01-OH and Corr-4a-linker-CO₂H moieties, linked with an ethylene glycol spacer through an ester bond. The potentiator 2 and corrector 13 fragments (after cleavage) had low micromolar potency for restoration of ΔF508-CFTR channel gating and cellular processing, respectively. Cleavage of hybrid molecule 14 by intestinal enzymes under physiological conditions produced the active potentiator 2 and corrector fragments 13, providing proof-of-concept for small-molecule potentiator-corrector hybrids as a single drug therapy for CF caused by the ΔF508 mutation.

Full text of DOI:10.1016/j.bmcl.2009.11.020

Bioorganic & Medicinal Chemistry Letters 20 (2010) 87–91
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of a hybrid potentiator–corrector agonist
of the cystic fibrosis mutant protein
DF508-CFTR
Aaron D. Mills a,b,c, Choong Yoo a, Jeffrey D. Butler a, Baoxue Yang b, A. S. Verkman b, Mark J. Kurth a,
*
a Department of Chemistry, University of California, Davis, CA 95616, USA
b Departments of Medicine and Physiology, University of California, San Francisco, CA 94143, USA
c Department of Chemistry, University of Idaho, Moscow, ID 83844, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A developing therapy of cystic fibrosis caused by the
DF508 mutation in CFTR employs correction of
Received 6 October 2009
Accepted 9 November 2009
Available online 13 November 2009
defective CFTR chloride channel gating by a ‘potentiator’ and of defective CFTR protein folding by a ‘cor-
rector’. Based on SAR data for phenylglycine-type potentiators and bithiazole correctors, we designed a
hybrid molecule incorporating an enzymatic hydrolysable linker to deliver the potentiator (PG01) frag-
ment 2 and the corrector (Corr-4a) fragment 13. The hybrid molecule 14 contained PG01-OH and
Corr-4a–linker–CO2H moieties, linked with an ethylene glycol spacer through an ester bond. The poten-
tiator 2 and corrector 13 fragments (after cleavage) had low micromolar potency for restoration of
Dedicated to the memory of Professor Aaron
Mills (University of Idaho)
D
F508-CFTR channel gating and cellular processing, respectively. Cleavage of hybrid molecule 14 by
Keywords:
intestinal enzymes under physiological conditions produced the active potentiator 2 and corrector frag-
ments 13, providing proof-of-concept for small-molecule potentiator–corrector hybrids as a single drug
Cystic fibrosis transmembrane conductance
regulator protein (CFTR)
therapy for CF caused by the
DF508 mutation.
D
F508-CFTR
Potentiators
Correctors
Ó 2009 Elsevier Ltd. All rights reserved.
Hybrids
Multi-ligand drug
Introduction: The genetic disease cystic fibrosis (CF) is caused by
mutations in the cystic fibrosis transmembrane conductance regu-
lator (CFTR) gene (CFTR), which encodes a cAMP-regulated chlo-
of activators, bithiazoles (Fig. 1, Corr-4a), are believed to bind to
F508-CFTR at the endoplasmic reticulum and facilitate its folding
and plasma membrane targeting; these activators are referred to as
‘correctors’. These compounds (phenylglycines and bithiazoles)
were identified from screening a diverse, small-molecule collection
of 150,000 compounds.6,7
D
ride channel.1–3 Mutations in
DF508-CFTR lead to distinct defects
in channel gating and cellular processing.1 Cystic fibrosis results
in chronic lung infection, deterioration of lung function, and death.
D
F508-CFTR is misfolded, retained at the endoplasmic reticulum
Because both a potentiator and corrector are likely required to
(ER), and rapidly degraded.4 Small-molecule therapy will likely re-
quire compounds that correct the two major underlying problems
in CF: (a) CFTR misfolding and ER retention, and (b) defective chan-
nel gating.5–7 Herein, we report proof-of-principle data toward the
design, synthesis, and component conjugation-site tolerance of a
potentiator–corrector linked hybrid. This cleavable conjugate ap-
proach requires significant SAR data to determine tolerant sites
for individual ligand conjugation and is a key first step in the mul-
tiple ligand approach.8
treat cystic fibrosis9 caused by the
DF508 mutation, cystic fibrosis
is an attractive target for the development of a multi-ligand drug.
This approach to the treatment of complex diseases with single
compounds containing multiple drug ligands is an emerging para-
digm in drug discovery.9 Multi-ligand ‘hybrid’ drugs have advanta-
ges and disadvantages. One advantage is that the synthetic
chemistry development takes place in the less expensive, early
stages of drug development. Another advantage is that the multi-
ligand approach can avoid complex pharmacokinetic/pharmacody-
namic relationships, which require extensive clinical studies to
clarify drug–drug interactions in cocktail/multicomponent drugs.8
Disadvantages with this strategy involve issues in variable poten-
cies of the individual moieties as well as differences in their bio-
availability and in vivo processing.10
Previously, we identified several chemical classes of small-mol-
ecules that correct each of these two defects.5–7 One class of acti-
vators, phenylglycines (Fig. 1, PG01), are believed to bind to
D
F508-CFTR at the cell surface and increase chloride channel gat-
ing; these activators are referred to as ‘potentiators’. Another class
Results and discussion: To synthesize a molecule containing
potentiator and corrector fragments, possible linker connection
strategies were considered (Fig. 1). A review of recent work and
* Corresponding author. Tel.: +1 530 752 8192; fax: +1 530 752 8995.
E-mail address: mjkurth@ucdavis.edu (M.J. Kurth).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.11.020

88
A. D. Mills et al. / Bioorg. Med. Chem. Lett. 20 (2010) 87–91
A
CH3
OMe
Cl
O
CH3
N
H3C
H
N
H
N
S
N
N
H
NH
O
O
N
S
CH3
Corr-4a
PG01
O
B
linker
CH3
N
O
O
H
N
H
N
S
N
N
NH
H
O
O
N
S
CH3
Cl
potentiator fragment
corrector fragment
Figure 1. (A)
DF508-CFTR corrector Corr-4a and potentiator PG01. (B) Strategy for a potentiator–corrector hybrid molecule.
SAR data for bithiazole correctors9 suggested that connection of
the linker through the 2-methoxy group may be a suitable position
for installation of the linker to the corrector. Reviewing SAR data
for the phenylglycine potentiators pointed toward incorporation
of a benzyl alcohol functionality on the aniline ring in place of
the 4-isopropyl group.
Chemistry: The synthesis of the hybrid complex began with the
synthesis of the potentiator fragment 2 (Scheme 1). Starting from
commercially available 4-aminobenzyl alcohol and coupling with
N-methyl-Boc-phenylglycine under EDC coupling conditions gave
the Boc protected 2° amine 1. TFA mediated Boc deprotection
and EDC coupling of the resulting 2° amine with 3-indole acetic
acid gave the potentiator fragment 2 in 75% yield together with a
small amount of ester 3 (11%).
The synthesis of the corrector fragment began with the
synthesis of the linker (Scheme 2). Diethylene glycol was reacted
with t-butylbromoacetate to give ethylene glycol t-butyl ester 4.
Mitsunobu coupling11 with 2-nitro-4-chlorophenol gave 5 in excel-
lent yield. Reduction of the nitro group using stannous chloride
followed by treatment with thiophosgene gave isothiocyanate 7.
Subsequent treatment of isothiocyanate 7 with ammonia gas in
DCM afforded the thiourea 8. Commercially available benzoylis-
othiocyanate was reacted with ammonia gas to afford thiourea 9,
which was reacted with 3-chloro-2,4-pentanedione to yield thia-
zole 10. Bromination of 10 with pyridinium tribromide in HBr gave
a-bromoketone 11 in excellent yield. Thiazole cyclization of thio-
urea 8 and -bromoketone 11 gave Corr–linker-t-butylester 12,
a
which was deprotected with TFA affording Corr-4a–CO2H 13. The
hybrid molecule 14 (Scheme 3) was constructed via an EDC cou-
pling reaction of PG01-OH fragment 2 and Corr-4a–CO2H fragment
13 which gave the hybrid in 11% yield.
Activity measurements: The
DF508-CFTR potentiator and correc-
tor activities of fragment 2, fragment 13, and hybrid 14, were
measured, and compared with reference compounds PG01 and
Corr-4a. Activities were assayed by established methodology
utilizing FRT epithelial cells stably coexpressing human
DF508-
CFTR and the high-sensitivity halide-sensing green fluorescent
protein YFP-H148Q/I152L as described previously.5–7 For measure-
ment of potentiator activity, cells were grown at 27 °C for 24 h to
allow
DF508-CFTR trafficking to the cell plasma membrane.
D
F508-CFTR channel function was measured from the kinetics of
halide (iodide) influx after incubation with test compound for
10 min in the presence of the cAMP agonist forskolin. For measure-
ment of corrector activity, cells were incubated at 37 °C for 20 h in
the presence of test compound, washed with PBS, and assayed for
iodide influx in the presence of forskolin and the reference
OH
O
CH3
N
OH
O
CH3
N
HO
Boc
Ph
N
H
Boc
1. TFA, DCM
rt, 98%
EDC, DMF/DCM
0oC to rt. 84%
NH2
2. EDC, DMF/
DCM, 0ºC
to rt
NH
1
O
OH
NH
O
HO
NH
CH3
N
O
O
CH3
O
N
N
H
N
NH
H
O
O
3; 11%
2; 75%
Scheme 1. Synthesis of PG01-OH fragment 2.

A. D. Mills et al. / Bioorg. Med. Chem. Lett. 20 (2010) 87–91
89
NaHMDS, THF
0ºC, 2h;
tBuO2C
36%
O2N
Cl
HO
O
OH
tBuO2C
O
O
OH
Br
4
HO
PPh3, DIAD,
THF, 0ºC to rt
90%
Cl
Cl
tBuO2C
O
tBuO2C
O
H2N
O2N
SnCl2
S=CCl2
Et3N
H2O/DCM
0ºC to rt
95%
EtOH, 90ºC
1.5 h, 51%
O
O
O
6
O
O
5
Cl
S
Cl
tBuO2C
tBuO2C
O
H2N
O
SCN
O
HN
NH3
O
DCM, 0ºC to rt
3 h, 98%
O
7
8
O
O
O
O
S
NH3
NCS
N
NH2 Cl
H
EtOH, 80oC
12 h, 60%
DCM, 0oC to rt
4 h, 99%
9
O
N
O
N
S
H
10
O
8,
N
pyridinium
tribromide
HBr, 30 wt%
in AcOH, rt
12 h, 90%
O
EtOH
N
H
S
80ºC
12h, 71%
11
Br
O
O
N
N
S
S
Cl TFA/
DCM
Cl
N
H
N
H
S
S
N
N
1:1
HN
HN
rt, 3h
99%
t BuO2C
HO2C
O
O
O
O
O
O
12
13
Scheme 2. Synthesis of Corr–CO2H fragment 13.
EDC, DMAP
O
N
2
DCM/DMF
0ºC to rt, 24 h
11%
+
S
Cl
N
H
S
13
N
O
HN
O
carbonic anhydrase
or
O
O
O
CH3
N
intestinal contents
O
14
N
H
HN
O
Scheme 3. Synthesis of hybrid 14 and subsequent hydrolysis by carbonic anhydrase or intestinal contents of this hybrid to the active potentiator PG-OH 2 and corrector
Corr4a–CO2H 13 fragments.

90
A. D. Mills et al. / Bioorg. Med. Chem. Lett. 20 (2010) 87–91
potentiator genistein (50 lM). Measurements were made on a
20.32
A
fluorescence plate reader with an automated syringe pump, incor-
porating appropriate positive and negative controls in all measure-
ments, each of which was made in quadruplicate. The first graph
(Fig. 2) summarizes concentration-activity data (iodide influx
rates) for potentiator PG01, PG01-OH 2, and hybrid 14 in the
potentiator assay. In the second panel of Figure 2, Corr-4a is com-
pared with corrector fragment 13 and potentiator–corrector hybrid
14 in the corrector assay. Both the potentiator (2) and corrector
(13) fragments exhibit strong activity in their corresponding as-
says. This retained activity validates that the structural modifica-
tions necessary to synthesize hybrid 14. The lack of activity of
hybrid 14 is likely due to its low penetration into the cell interior.
LC/MS: An LC method to identify the potentiator, corrector, and
hybrid peaks was developed using a Waters 2695 LC and a Waters
PDA 996 detector coupled to an Alliance mass spectrometer,
ionization mode; electrospray (+), mass range 200–1200 Da, 23-V
cone voltage, column; XTerra MS C18 (Waters, 2.1 mm Â
17.29
Corr-CO2H
PG-OH
carbonic anhydrase
19.54
8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00
Corr-CO2H
PG-OH
14
hybrid
B
17.50
20.55
intestinal contents
22.93
6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00
Figure 3. (A) HPLC spectra of the supernate from the reaction of hybrid 14 and
carbonic anhydrase. (B) HPLC spectra of the supernate from the reaction of hybrid
14 and mouse intestinal contents.
50 mm  3.5
lm). Reversed-phase HPLC separations were carried
out using a C18 column connected to a solvent delivery system
(model 2690; Waters, Milford, MA). The solvent system consisted
of a linear gradient from 0% CH3CN/100% H2O, to 100% CH3CN/0%
H2O, over 28 min, followed by 4 min at 100% CH3CN/0% H2O, and
3 min at 0% CH3CN/100% H2O to equilibrate the column for the next
run (0.2 ml/min flow rate). The retention times for PG01-OH (2),
Corr-4a–CO2H (13), and hybrid 14 were: 17 min, 20 min, and
23 min, respectively, detected at 256 nm. Mass spectra were used
to confirm compound identity. Mass spectrometer specifications
are as follows: Waters Alliance (HT 2790 + ZQ) mass spectrometer
utilizing positive ion detection mode, scanning from 200 to 1200 Da.
Hydrolysis of the hybrid: The hybrid was first treated with car-
bonic anhydrase, a major enzyme present on the intestinal mucosa,
incubating for 10 min at rt (Fig. 3).12 Methanol was added to the
solution, cooled in an ice bath at 0 °C, and centrifuged at 16,000g
for 10 min. The supernate was analyzed by LC/MS. The hybrid
was completely hydrolyzed to the corresponding potentiator and
corrector fragments as identified by LC/MS (Fig. 3A). Hydrolysis
of the hybrid 14 was performed in parallel by incubation with
the intestinal contents from mice at 37 °C (90% humidity, 5%
CO2) for 4 h. Methanol was added to the solution, cooled to 0 °C,
and centrifuged at 16,000g for 10 min. The supernate was analyzed
by LC/MS. The hybrid was completely hydrolyzed to the corre-
sponding potentiator and corrector fragments identified by LC/
MS (Fig. 3B). Peaks were correlated to spectra of purified com-
pounds run under identical conditions. Disappearance of the hy-
brid 14 peak (retention time, 23 min) in both HPLC experiments
(A and B) and the appearance of the two corresponding active frag-
ments PG-OH 2 (retention time, 17 min), Corr–CO2H 13 (retention
time, 20 min) illustrate that the hybrid is being hydrolyzed to the
active fragments.
Conclusion: This work provides proof-of-concept for the design,
synthesis, and component conjugation-site tolerance of an ester-
0.16
PG 01
linked
DF508-CFTR potentiator–corrector hybrid molecule. Fur-
2
ther, this study provides insight into the structural requirements/
tolerances of a multiple ligand approach for CF treatment. The
resultant potentiator and corrector fragments following hydrolysis,
as would occur in the gastrointestinal tract, were active in the cor-
DF508-CFTR cellular processing and chloride
channel gating, respectively. Thus, a single compound can be engi-
neered to confer the distinct activities required to restore chloride
0.12
-
d[I ]
0.08
dt
rection of defective
(mM/s)
0.04
14
channel function in cystic fibrosis caused by the
DF508 mutation.
0
0.001 0.01
0.1
1
10
100
Acknowledgments
[compound] (mM)
The authors thank the Tara K. Telford Fund for Cystic Fibrosis
Research at the University of California, Davis, the National Insti-
tutes of Health (Grants DK072517 and GM076151), and the Na-
tional Science Foundation [Grants CHE-0910870, CHE-0443516,
CHE-0449845, and CHE-9808183 (NMR spectrometers)].
0.12
13
0.10
0.08
0.06
0.04
0.02
-
Corr-4a
d[I ]
dt
(mM/s)
Supplementary data
14
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.11.020.
0.1
1
10
100
References and notes
[compound] (mM)
1. (a) Bobadilla, J. L.; Macek, M.; Fine, J. P.; Farrell, P. M. Hum. Mutat. 2002, 19, 575;
(b) Gadsby, D. C.; Vergani, P.; Csanady, L. Nature 2006, 440, 477; (c) Dalemans,
W.; Barbry, P.; Champigny, G.; Jallat, S.; Dott, K.; Dreyer, D.; Crystal, R. G.;
Pavirani, A.; Lecocq, J. P.; Lazdunski, M. Nature 1991, 354, 526.
Figure 2. Dose–response data for potentiator assay of compounds PG01, 2, and 14
(top panel SEM; n = 4) and for corrector assay of compounds 13, Corr-4a, and 14
(bottom panel, SEM; n = 4).

A. D. Mills et al. / Bioorg. Med. Chem. Lett. 20 (2010) 87–91
91
2. (a) Pilewski, J. M.; Frizzell, R. Physiol. Rev. 1999, 79, S215; (b) Dave, C. G.; Shah,
R. D. Heterocycles 1999, 51, 1819; (c) Dave, C. G.; Joshipura, H. M. Indian J. Chem.,
Sect B 2002, 41B, 650.
3. (a) Sheppard, D. N.; Welsh, M. J. Physiol. Rev. 1999, 79, S23; (b) Bobadilla, J.;
Macek, M.; Fine, J. P.; Farrell, P. M. Hum. Mutat. 2002, 19, 575.
4. (a) Denning, G. M.; Anderson, M. P.; Amara, J. F.; Marshall, J.; Smith, A. E.;
Welsh, M. J. Nature 1992, 358, 761; (b) Lukacs, G. L.; Mohamed, A.; Kartner, N.;
Chang, X.-B.; Riordan, J. R.; Grinstein, S. EMBO J. 1994, 13, 6076; (c) Kopito, R. R.
Physiol. Rev. 1999, 79, S167; (d) Du, K.; Sharma, M.; Lukacs, G. L. Nat. Struct. Mol.
Biol. 2005, 12, 17.
5. Pedemonte, N.; Sonawane, N. D.; Taddei, A.; Hu, J.; Zegarra-Moran, O.; Suen, Y.
F.; Robins, L. I.; Dicus, C. W.; Willenbring, D.; Nantz, M. H.; Kurth, M. J.; Galietta,
L. J.; Verkman, A. S. Mol. Pharmacol. 2005, 67, 1797.
Robins, L. I.; Verkman, A. S.; Kurth, M. J. Bioorg. Med. Chem. Lett. 2008, 18, 2610;
(c) Yu, G. J.; Yoo, C. L.; Yang, B.; Lodewyk, M. W.; Meng, L.; El-Idreesy, T. T.;
Fettinger, J. C.; Tantillo, D. J.; Verkman, A. S.; Kurth, M. J. J. Med. Chem. 2008, 51,
6044.
7. Yang, H.; Shelat, A. A.; Guy, R. K.; Gopinath, V. S.; Ma, T.; Du, K.; Lukacs, G. L.;
Taddei, A.; Folli, C.; Pedemonte, N.; Galietta, L. J.; Verkman, A. S. J. Biol. Chem.
2003, 278, 35079.
8. Morphy, R.; Rankovic, Z. J. Med. Chem. 2005, 20, 6523.
9. Yu, G. J.; Yoo, C. L.; Yang, B.; Lodewyk, M. W.; Meng, L.; El-Idreesy, T. T.;
Fettinger, J. C.; Tantillo, D. J.; Verkman, A. S.; Kurth, M. J. J. Med. Chem. 2008, 51,
6044.
10. Meunier, B. Acc. Chem. Res. 2008, 41, 69.
11. Mitsunobu, O.; Yamada, Y. Bull. Chem. Soc. Jpn. 1967, 40, 2380.
12. Tu, C.; Thomas, H. G.; Wynns, G. C.; Silverman, D. N. J. Biol. Chem. 1986, 261,
10100.
6. (a) Pedemonte, N.; Lukacs, G. L.; Du, K.; Caci, E.; Zegarra-Moran, O.; Galietta, L.
J.; Verkman, A. S. J. Clin. Invest. 2005, 115, 2564; (b) Yoo, C. L.; Yu, G. J.; Yang, B.;