Synthesis of new steroidal inhibitors of P-glycoprotein-mediated multidrug resistance and biological evaluation on K562/R7 erythroleukemia cells

Article abstract of DOI:10.1021/jm501676v

A simple route for improving the potency of progesterone as a modulator of P-gp-mediated multidrug resistance was established by esterification or etherification of hydroxylated 5α/β-pregnane-3,20-dione or 5β-cholan-3-one precursors. X-ray crystallography of representative 7α-, 11α-, and 17α-(2′R/S)-O-tetrahydropyranyl ether diastereoisomers revealed different combinations of axial-equatorial configurations of the anomeric oxygen. Substantial stimulation of accumulation and chemosensitization was observed on K562/R7 erythroleukemia cells resistant to doxorubicin, especially using 7α,11α-O-disubstituted derivatives of 5α/β-pregnane-3,20-dione, among which the 5β-H-7α-benzoyloxy-11α-(2′R)-O-tetrahydropyranyl ether 22a revealed promising properties (accumulation index 2.9, IC₅₀ 0.5 μM versus 1.2 and 10.6 μM for progesterone), slightly overcoming those of verapamil and cyclosporin A. Several 7α,12α-O-disubstituted derivatives of 5β-cholan-3-one proved even more active, especially the 7α-O-methoxymethyl-12α-benzoate 56 (accumulation index 3.8, IC₅₀ 0.2 μM). The panel of modulating effects from different O-substitutions at a same position suggests a structural influence of the substituent completing a simple protection against stimulating effects of hydroxyl groups on P-gp-mediated transport.

Full text of DOI:10.1021/jm501676v

Subscriber access provided by UNIV LIBRE DE BRUXELLES
Article
Synthesis of New Steroidal Inhibitors of P-glycoprotein-mediated Multidrug
Resistance and Biological Evaluation on K562/R7 Erythroleukemia Cells
Marc Rolland de Ravel, Ghina Alameh, Maxime Melikian, Zahia Mahiout, Agnès Emptoz-
Bonneton, Eva-Laure Matera, Thierry Lomberget, Roland Barret, Luc Rocheblave, Nadia
Walchshofer, Sonia Beltran, Lucienne El Jawad, Elisabeth Mappus, Catherine Grenot,
Michel Marie PUGEAT, Charles Dumontet, Marc LeBorgne, and Claude Yves Cuilleron
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501676v • Publication Date (Web): 29 Jan 2015
Downloaded from http://pubs.acs.org on February 1, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
LeBorgne, Marc; Université de Lyon - UCB Lyon 1 - ISPB, EA 4446 B2C :
"
Biomolecules, Cancer and Chemoresistances",
Cuilleron, Claude; Université de Lyon, Université Lyon 1,Inserm U863,
ISPB-Faculté de Pharmacie, Laboratoire de Chimie Organique
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Synthesis of New Steroidal Inhibitors of Pꢀ
glycoproteinꢀmediated Multidrug Resistance and
Biological Evaluation on K562/R7 Erythroleukemia
Cells
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
§
§
§
§
Marc Rolland de Ravel, Ghina Alameh, Maxime Melikian, Zahia Mahiout, Agnès Emptozꢀ
§
†
‡§
‡§
‡§
Bonneton, EvaꢀLaure Matera, Thierry Lomberget, Roland Barret, Luc Rocheblave, Nadia
‡§
§
§
§
§
Walchshofer, Sonia Beltran, Lucienne El Jawad, Elisabeth Mappus, Catherine Grenot,
‡
†
‡§
*,§
Michel Pugeat, Charles Dumontet, Marc Le Borgne, and Claude Yves Cuilleron
†
Centre de Recherche en Cancérologie de Lyon ꢀ Université Claude Bernard Lyon 1, INSERM
U1052ꢀCNRS UMR5286, Centre Léon BérardꢀCheney D, 28 rue Laënnec, 69373, Lyon Cedex
08, France
§
INSERMꢀU863 Hormones stéroïdes et protéines de liaison, Université Claude Bernard Lyon 1,
Avenue Rockefeller, 69373, Lyon Cedex 08, France
8
ACS Paragon Plus Environment
1

Page 3 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
‡
Université de Lyon, Université Lyon 1, Faculté de Pharmacie, ISPB, EA 4446 Biomolécules
Cancer et Chimiorésistances, SFR Santé LyonꢀEst CNRS UMS3453ꢀINSERM US7, 8 Avenue
Rockefeller, Fꢀ69373 Lyon Cedex 8, France
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This paper is dedicated to the memories of Christiane Villard (1945ꢀ1993) and Daniel Royan
(1942ꢀ1998).
ABSTRACT
A simple route for improving the potency of progesterone as a modulator of Pꢀgpꢀmediated
multidrug resistance was established by esterification and/or etherification of hydroxylated 5α/βꢀ
pregnaneꢀ3,20ꢀdione or 5βꢀcholanꢀ3ꢀone precursors. Xꢀray crystallography of representative 7αꢀ,
11αꢀ and 17αꢀ(2’R/S)ꢀOꢀtetrahydropyranyl ether diastereoisomers revealed different
combinations of axialꢀequatorial configurations of the anomeric oxygen. Substantial stimulation
of accumulation and chemosensitization was observed on K562/R7 erythroleukemia cells
resistant to doxorubicin, especially using 7α,11αꢀOꢀdisubstituted derivatives of 5α/βꢀpregnaneꢀ
3,20ꢀdione, among which the 5βꢀHꢀ7αꢀbenzoyloxyꢀ11αꢀ(2’R)ꢀOꢀtetrahydropyranyl ether 22a
revealed promising properties (accumulation index: 2.9, IC : 0.5 µM versus 1.2 and 10.6 µM for
5
0
progesterone), slightly overcoming those of verapamil and cyclosporin A. Several 7α,12αꢀOꢀ
disubstituted derivatives of 5βꢀcholanꢀ3ꢀone proved even more active especially the 7αꢀOꢀ
methoxymethylꢀ12αꢀbenzoate 56 (accumulation index: 3.8, IC : 0.2 µM). The panel of
5
0
modulating effects from different Oꢀsubstitutions at a same position suggests a structural
ACS Paragon Plus Environment
2

Journal of Medicinal Chemistry
Page 4 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
influence of the substituent completing a simple protection against stimulating effects of
hydroxyl groups on Pꢀgpꢀmediated transport.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Multidrug resistance (MDR), either intrinsic or acquired, remains a major limit for cancer
chemotherapy. The predominant mechanism is a reduced drug accumulation in cells consecutive
to an active efflux by energyꢀdependent transporters belonging to the ATPꢀbinding cassette
1
(ABC) family. The main contribution comes from Pꢀglycoprotein (Pꢀgp/ABCB1) which can
export structurally unrelated cytotoxic drugs as well as a wide variety of compounds including
2
synthetic or natural hormonal steroids. Pꢀgp is often overexpressed in resistant tumors but is
also present in many normal cells. Pꢀgp expression was found to be stimulated under the
3
influence of the Pregnane Xenobiotic Receptor (PXR). Other MDR proteins such as the
Multidrug Resistance Protein 1 (MRP1/ABCC1) and the Breast Cancer Resistance Protein
(BCRP/ABCG2) can also contribute to drug efflux. Moreover, nonꢀPꢀgpꢀmediated MDR
4
mechanisms have been reported. The structure of human Pꢀgp consists of a singleꢀchain protein
of 1280 amino acids organized in a spatial arrangement extrapolated from the Xꢀray structure of
5ꢀ6
mouse Pꢀgp.
Considerable efforts have been devoted to the development of molecules able to inhibit the
efflux of cytotoxic drugs by ABC transporters, focused especially on Pꢀgp modulators for which
7ꢀ11
structureꢀactivity relationships have also been proposed
but most often revealed sideꢀeffects
1
2ꢀ13
not compatible for approval in therapy.
Among the wide number of candidate structures for
ACS Paragon Plus Environment
3

Page 5 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
developing inhibitors of Pꢀgpꢀmediated drug efflux, several natural and synthetic steroids have
been investigated.
14
The first report about the role of progesterone as a moderately active inhibitor of Pꢀgp efflux
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
stimulated numerous further studies. Investigations on the inhibiting effects of progesterone
derivatives and metabolites revealed that the nontoxic 5βꢀdihydroprogesterone but not 5αꢀ
1
5ꢀ16
dihydroprogesterone, was nearly as active as progesterone
while 17αꢀhydroxyꢀ5βꢀ
dihydroprogesterone was reported to enhance Pꢀgp inhibition without interacting with
1
7
glucocorticoid receptors. Progesterone has been the target of numerous structural modifications
to stimulate its Pꢀgp inhibiting effects. Improvements were reported after introduction of
1
8
optimized 7αꢀthiophenyl substituents also preventing interaction with progesterone receptors,
1
9ꢀ20
or of 11αꢀbenzoate and carbamate substituents.
Synthetic drugs structurally related to
21
progesterone and modified at the 17βꢀacetyl sideꢀchain were also employed. A few isolated
22ꢀ25
natural products with pregnane skeletons have been reported as Pꢀgp inhibitors.
Pꢀgp
inhibiting properties were also found for 21ꢀdimethylaminobenzoate derivatives of two
26
27ꢀ28
glucocorticoid receptor agonists as well as for several synthetic glucocorticoids.
Many
other compounds related to steroidal pharmaceuticals were also found able to overcome MDR
2
9
such as the antiprogestin drug mifepristone (RU486) bearing a 11ꢀdimethylaminophenyl
3
0
31
substituent and their related estradiol analogs, as well as progestins and the steroidal
3
2
32
antiestrogens ICI 182,780 and 164,384. A few reports have mentioned MDR inhibition
33
34,35
properties of bile acid or salt derivatives such as glycocholic acid, deoxycholate,
3
6
taurolithocholate, taurochenodeoxycholate and glycochenodeoxycholate and ursodeoxycholate.
Mechanisms associated with alterations in the fluidity of the plasma membrane have been
37
postulated for some of the bile salts effects.
ACS Paragon Plus Environment
4

Journal of Medicinal Chemistry
Page 6 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The importance of steroids either as inhibitors or transported substrates has justified structureꢀ
activity studies. Investigations on corticosteroid transport by murine mdr1ꢀPꢀgp variants have
identified mutations in the transmembrane domain 4ꢀ6 of the first half of the protein altering the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17,38
recognition of 17αꢀhydroxyl and/or 20ꢀketone groups of steroid inhibitors.
Comparative
studies using different synthetic glucocorticoids revealed structural features which significantly
3
9
influenced their transport by Pꢀgp, completing those established for other interacting steroid
1
5,16,40
41
42
structures.
3DꢀQSAR studies and homology modeling/docking approaches have been
developed. Fluorescenceꢀquenching experiments have suggested interactions of steroid
40,43
modulators in the proximity of the ATPꢀbinding site
potentially associated with ATPaseꢀ
4
4
dependent conformational changes controlling Pꢀgpꢀmediated drug transport. Therefore
attempts have been made to generate Pꢀgp inhibitors with mixed functions that are able to
45ꢀ48
interfere with both ATPase and modulator sites.
The mechanisms underlying modulation of
Pꢀgp transport as well as ATPase activity appear complex owing to the involvement of several
4
9ꢀ51
interacting binding sites according to ligand structures.
The present study was undertaken with the view to explore whether very simple chemical
modifications such as esterification and etherification of monoꢀ or dihydroxylated steroidal
precursors derived from progesterone, 5αꢀ or 5βꢀdihydroprogesterone as well as from cholic acid
series could provide structureꢀactivity criteria which are useful for improving modulation of Pꢀ
gpꢀmediated drug efflux. The use of hydroxylated precursors related either to known
physiological steroids or their metabolites and the versatility of chemical modifications of
hydroxyl groups also appeared particularly favorable for future structural optimizations aimed at
maintaining a strong inhibiting potency in vivo without stimulating unwanted interactions or
intrinsic toxicity beyond clinically acceptable limits.
ACS Paragon Plus Environment
5

Page 7 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CHEMISTRY
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The three monosubstituted 11αꢀ, 17αꢀ and 21ꢀOꢀtetrahydropyranyl (OTHP) ether derivatives of
progesterone (Scheme 1) were prepared by tetrahydropyranylation of their commercially
available hydroxyl precursors followed by preparative thinꢀlayer chromatography (TLC)
purification which was able to separate the 11αꢀ and 17αꢀ (2’R/S)OTHP diastereoisomers 1a,b
and 2a,b but not those of the 21ꢀ(2’R+S)OTHP ether 3.
The access to the monosubstituted 7αꢀ, 11αꢀ, 17αꢀOTHP ether and 7αꢀbenzoate derivatives of
5βꢀdihydroprogesterone (Scheme 2) required a preliminary synthesis of the corresponding
hydroxyl precursors. Catalytic hydrogenation of 6α,7αꢀepoxyꢀpregnꢀ4ꢀeneꢀ3,20ꢀdione 4 afforded
5
2
7
αꢀhydroxyꢀ5βꢀpregnaneꢀ3,20ꢀdione 5 according to a reported procedure. Benzoylation of this
precursor gave the 7αꢀbenzoate 6 while tetrahydropyranylation led to the 7αꢀ(R+S)OTHP ether 7
as a mixture of R/S isomers which could not be separated by TLC. Catalytic hydrogenation of
5
3
11αꢀhydroxyprogesterone led to 11αꢀhydroxyꢀ5βꢀpregnaneꢀ3,20ꢀdione
8 which, after
tetrahydropyranylation and TLC, provided the separated 11αꢀ(R/S)OTHP isomers 9a,b. A similar
3
8
protocol applied to 17αꢀhydroxyprogesterone led to 17αꢀhydroxyꢀ5βꢀpregnaneꢀ3,20ꢀdione 10
and to the separated 17αꢀ(R/S)OTHP isomers 11a,b.
A first synthetic scheme leading to 7α,11αꢀOꢀdisubstituted derivatives of both 5α/βꢀpregnaneꢀ
3,20ꢀdione series relied upon a 11αꢀOꢀtertꢀbutyldimethylsilyloxy (OTBDMS)ꢀ5ꢀenꢀ7ꢀone
intermediate (Scheme 3) prepared from 11αꢀhydroxyprogesterone. This enone was converted to
a 5ꢀeneꢀ3,20ꢀbisꢀketal 12, then protected as the 11αꢀOTBDMS ether 13. Allylic oxidation with
54
55
the CrO ꢀ(pyridine) complex led to the 5ꢀenꢀ7ꢀone 14. Catalytic hydrogenation of this enone
3
2
ACS Paragon Plus Environment
6

Journal of Medicinal Chemistry
Page 8 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
gave similar proportions of 5αꢀpregnanꢀ7ꢀone 15a and 5βꢀpregnanꢀ7ꢀone 15b which could not be
separated. Stereoselective reduction of these unseparated 7ꢀketones with lithium triꢀsecꢀ
5
6
butylborohydride (LꢀSelectride) afforded a mixture of 5βꢀ and 5αꢀpregnanꢀ7αꢀol derivatives 16
and 17 readily separated by flashꢀchromatography. Benzoylation of the 7αꢀol precursor 16 led to
the 7αꢀbenzoate 18 which, after cleavage of the 11ꢀsilyl group with tetraꢀnꢀbutylammonium
fluoride, gave the 11αꢀol derivative 19. Acidolysis of the ketal groups provided 7αꢀbenzoyloxyꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
11αꢀhydroxyꢀ3,20ꢀdione 20 which was benzoylated to give the 7α,11αꢀdibenzoate 21.
Tetrahydropyranylation of the 11αꢀol precursor 20 followed by flashꢀchromatography led to the
separated 11αꢀ(R/S)OTHP isomers 22a,b. The monobenzoate 20 was also saponified to the
7α,11αꢀdiol derivative 23 then converted to the 7α,11αꢀbisꢀOTHP ether which could be separated
by TLC as the two 11αꢀ(R/S) isomers 24a,b without separation of the 7αꢀ(R/S) isomers.
Benzoylation of the 7αꢀol precursor 17 led to the 7αꢀbenzoate 25 which, after desilylation, gave
the 7αꢀbenzoyloxyꢀ11αꢀol derivative 26. Acidolysis of the ketal group provided the 7α,11αꢀ
dihydroxyꢀ3,20ꢀdione 27 then converted either to the 7α,11αꢀdibenzoate 28 or to the separated
7αꢀbenzoyloxyꢀ11αꢀ(R/S)OTHP isomers 29a,b.
A second synthetic scheme was designed to lead exclusively to 7αꢀOTHPꢀ11αꢀbenzoate
derivatives of 5βꢀpregnanꢀ3,20ꢀdione (Scheme 4) according to the one employed for obtaining
the 7αꢀOTHP ether 7. The 11αꢀOTBDMS ether 30 prepared from 11αꢀhydroxyprogesterone was
dehydrogenated to the 4,6ꢀdienꢀ3ꢀone 31 then converted to the 6α,7αꢀepoxide 32. Catalytic
hydrogenation of epoxide 32 afforded the 5βꢀpregnanꢀ7αꢀol precursor 33. This compound was
converted to the 7αꢀ(R+S)OTHP ether 34 recovered as an unseparated mixture of R/S isomers.
Desilylation gave the 7αꢀ(R+S)OTHPꢀ11αꢀol derivative 35 which, after benzoylation and TLC
ACS Paragon Plus Environment
7

Page 9 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
separation, provided the two separated 11αꢀbenzoyloxyꢀ7αꢀ(R/S)OTHP isomers 36a,b.
Acidolysis of the unseparated mixture of isomers 36a,b led to the 7αꢀhydroxyꢀ11αꢀbenzoate 37.
The synthesis of 11α,17αꢀOꢀdisubstituted derivatives of 5βꢀpregnaneꢀ3,20ꢀdione (Scheme 5)
required an access to the 11α,17αꢀdiol precursor 41. Mild acidolysis of 3,20ꢀbisꢀ
ethylenedioxypregnꢀ5ꢀenꢀ11αꢀol derivative 12 led to the 3ꢀethylenedioxyꢀ11αꢀhydroxypregnꢀ5ꢀ
enꢀ20ꢀone 38. Oxygenation of this 20ꢀketone in the presence of NaH and triethylphosphite
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
7
afforded the 17αꢀhydroxyꢀ20ꢀketone 39. Acidolysis of the ketal group led to the 4ꢀenꢀ3ꢀone 40
which was reduced by catalytic hydrogenation exclusively to the 11α,17αꢀdihydroxyꢀ5βꢀ
pregnaneꢀ3,20ꢀdione 41. This 11α,17αꢀdiol precursor was converted by a first partial
tetrahydropyranylation step to a 11αꢀmonoꢀOTHP ether then separated by TLC as the 11αꢀ
(R/S)OTHP isomers 42a,b. A second tetrahydropyranylation step of the remaining 17α–hydroxyl
group afforded 11α,17αꢀbisꢀOTHP derivatives 43 showing a partial interconversion of the
previously separated 11αꢀ(R/S) isomers. Preparative TLC allowed the recovery of fractions
1
13
enriched in three of the four possible isomers, characterized by their H/ C NMR spectra and
liquid chromatographyꢀmass spectrometry (LCꢀMS) profiles (cf Supporting Information). A first
partially purified fraction, contained the 11αꢀ(S),17αꢀ(R) isomer 43a as the major product (c.a.
70%) mainly contaminated with the 11αꢀ(S),17αꢀ(S) isomer 43c. A second impure fraction
consisted in a mixture of the 11αꢀ(R),17αꢀ(R) isomer 43b as the major product (c.a. 35%)
contaminated with decreasing amounts of isomers 43a (c.a. 30%) and 43c (c.a. 25%). A third
nearly pure fraction contained only the isomer 43c (c.a. 95%). The 11αꢀ(R),17αꢀ(S) isomer 43d
could not be isolated as an independent fraction but was identified by NMR as the main byꢀ
product in the two fractions enriched in isomers 43a or 43b. Owing to difficulties encountered
for further purification (including attempts using reverseꢀphase HPLC hardly transposable at a
ACS Paragon Plus Environment
8

Journal of Medicinal Chemistry
Page 10 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
preparative scale), preliminary biological assays were restricted to the three available major
fractions respectively containing enriched isomers 43a, 43b and the pure isomer 43c.
The access to 7α,12αꢀdisubstituted derivatives of either methyl 3ꢀoxoꢀ5βꢀcholanꢀ24ꢀoate
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
series or their 24ꢀCH OMe analogs (Scheme 6) relied on 7α,12αꢀdihydroxyꢀ3ꢀone precursors.
2
5
8
Methyl cholate 44 was selectively oxidized with the Ag CO /Celite reagent to the 7α,12αꢀ
2
3
dihydroxyꢀ3ꢀketone 45. Benzoylation of this diol precursor gave the 7α,12αꢀdibenzoate 46 while
tetrahydropyranylation afforded the 7α,12αꢀbisꢀOTHP derivative 47 as an unseparated mixture
of R/S isomers. In order to avoid the formation of R/S diastereoisomers, an attempt was made to
59
use the 5,6ꢀdihydroꢀ4ꢀmethoxyꢀ2Hꢀpyran etherification reagent. Etherification of the 7α,12αꢀ
diol derivative 45 with an excess of this reagent remained incomplete even after 48 h at room
temperature and gave 7α,12αꢀbisꢀOꢀ(5’,6’ꢀdihydroꢀ2’Hꢀpyranꢀ4’ꢀyloxy)ether (ODHPE) 48 as the
major product. This ODHPE derivative was identified as an unsaturated dihydropyranyl ether
adduct homologous to the etherification reagent having lost the angular methoxy group of the
1
13
initial tetrahydropyranyl ether adduct, as established by H and C NMR and confirmed by mass
spectrometry. Selective monoꢀtetrahydropyranylation of the 7α,12αꢀdiol precursor 45 with a
limited amount of dihydropyran resulted in an incomplete reaction leading to a mixture of
compounds containing the 7αꢀhydroxyꢀ12αꢀmonoꢀOTHP ether 49 as the major product, the 7αꢀ
monoꢀOTHPꢀ12αꢀol regioisomer 50 as the minor product and trace amounts of 7α,12αꢀdiol
derivative 45 and 7α,12αꢀbisꢀOTHP ether 47. This mixture was benzoylated and the resulting
benzoates were purified by preparative TLC to give the two 12αꢀbenzoyloxyꢀ7αꢀ(R/S)OTHP
isomers 51a,b and the two 7αꢀbenzoyloxyꢀ12αꢀ(S/R)OTHP isomers 52a,b. An alternative access
to the 12αꢀbenzoyloxyꢀ7αꢀ(R/S)OTHP isomers 51a,b was obtained via selective benzoylation of
the 12αꢀhydroxyl group followed by tetrahydropyranylation of the 7αꢀhydroxyl group (Rolland
ACS Paragon Plus Environment
9

Page 11 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
de Ravel, unpublished results). Selective monoꢀOꢀmethoxymethylation of the 7α,12αꢀdiol
precursor 45 was undertaken using a stoichiometric amount of chloromethyl methyl ether which
provided a mixture of compounds containing the 7αꢀhydroxyꢀ12αꢀmonoꢀOꢀmethoxymethyl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
OMOM) ether 53 as the major product and its 7αꢀmonoꢀOMOMꢀ12αꢀol regioisomer 54 as the
minor product. These 7αꢀand 12αꢀol derivatives were separated by flashꢀchromatography and
benzoylated to give the 7αꢀbenzoyloxyꢀ12αꢀOMOM ether 55 and the 7αꢀOMOMꢀ12αꢀbenzoate
5
6.
The conversion of the 24ꢀCOOMe group of methyl cholate 44 to a 24ꢀCH OMe ether was
2
made in three steps by partial silylation to the 3ꢀOTBDMS ether 57, reduction with lithium
aluminum hydride to the 24ꢀol intermediate 58 and by selective etherification of this alcohol with
MeIꢀNaH to the 24ꢀOꢀmethyl ether 59. Desilylation afforded the 3α,7α,12αꢀtriol 60 which was
58
selectively oxidized at the 3ꢀposition with the Ag CO /Celite reagent to give the 7α,12αꢀ
2
3
dihydroxyꢀ3ꢀketone 61. Complete benzoylation of this substrate led to the 7α,12αꢀdibenzoate 62.
Selective monoꢀtetrahydropyranylation of the same diol derivative 61 afforded a mixture
containing the 7ꢀmonoꢀ(R+S)OTHPꢀ12αꢀol derivative 63, the 7αꢀhydroxyꢀ12ꢀmonoꢀ(R+S)OTHP
ether 64 (major product) and the 7α,12αꢀbisꢀOTHP ether 65 which could all be separated by
preparative TLC. Benzoylation of the 12αꢀol precursor 63 followed by chromatographic
separation provided the 12αꢀbenzoyloxyꢀ7αꢀ(R/S)OTHP isomers 66a,b. A similar treatment of
the 7αꢀol derivative 64 the 7αꢀbenzoyloxyꢀ12αꢀ(S/R)OTHP isomers 67a,b.
STRUCTURAL ASSIGNMENTS
ACS Paragon Plus Environment
10

Journal of Medicinal Chemistry
Page 12 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Optical rotations. A preliminary attempt for the assignment of (R) and (S) configurations of
the 11αꢀOTHP isomers of progesterone 1a and 1b was made by correlation of optical rotations
6
0
with those of (R/S)OTHP isomers derived either from adenosine or from (S)ꢀ(+)ꢀmethyl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
61
mandelate
for which absolute stereochemistries had been established from the
6
1
radiocristallographic structure of the less dextrorotatory (S) isomer. This suggested the same
(S) stereochemistry for the less dextrorotatory 11αꢀOTHP isomer 1b and therefore a similar trend
for the other pairs of OTHP isomers.
Xꢀray crystallography studies. The structural assignments concerning both the
tetrahydropyranyl ether R/S isomerism and the 5βꢀH stereochemistry of pregnane skeletons were
unambiguously assessed from radiocristallographic data established for three characteristic
representative pairs of 7αꢀ, 11αꢀ and 17αꢀ(R/S)OTHP isomers, two of which in 5βꢀpregnaneꢀ
3,20ꢀdione series and the third one in pregnꢀ4ꢀeneꢀ3ꢀone series (Figure 1). No adequate
crystalline material could be obtained for 5βꢀcholane derivatives. The 7αꢀOTHP isomer 36a
revealed a (R) configuration associated with an equatorial Oꢀlinkage to the THP ring. The
diastereoisomer 36b revealed a (S) configuration associated with an axial Oꢀlinkage linkage to
6
0
the THP ring as reported for the (2’S)OTHP derivative of adenosine. The 11αꢀOTHP
diastereoisomers 22a and 22b were confirmed respectively as (R) and (S) isomers, both with a
same equatorial Oꢀlinkage to the THP ring. Moreover, the 5βꢀH configuration was clearly
observable for these two pairs of isomers. The 17αꢀOTHPꢀpregnꢀ4ꢀeneꢀ3ꢀone diastereoisomers
2a and 2b were confirmed respectively as (R) and (S) isomers, both with the same axial Oꢀ
linkage to the THP ring, in keeping with previous conformational assignments on similar 17αꢀ
6
2
(
R/S)OTHP isomers. These structures revealed not only the absolute stereochemistries of 7αꢀ,
11αꢀ and 17αꢀ(R/S)OTHP diastereoisomers but also showed three different combinations of
ACS Paragon Plus Environment
11

Page 13 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
orientations of ether linkages between the steroid hydroxyl group and the tetrahydropyranyl ether
ring. These differences might result from steric constraints not predictable from NMR data on
6
3
anomeric effects reported for 2ꢀalkoxytetrahydropyrans.
An additional hypothesis of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
conformational differences between crystalline and solubilized forms cannot be excluded. These
differences of orientations of ether links with the THP rings may therefore not be safely
extrapolated to the other homologous structural analogs reported in this study.
1
13
Assignments for H and C NMR shift. NMR assignments were based mainly on
correlations with NMR data reported in literature and on correlations between the effects of
homologous structural modifications reported in this study. The mentioned structures take into
account the configurations of the 7αꢀ,11αꢀ and 17αꢀ(R/S)OTHP diastereoisomers deduced from
radiocristallographic data. In the absence of crystal structures for 12αꢀOTHP ethers, putative (R)
and (S) configurations opposite to those made for 11αꢀOTHP ethers were assigned on the
assumption that the characteristic higher deshielding of the vicinal 12ꢀmethylene group by the
11αꢀ(R) isomer might be exerted on the vicinal 11ꢀmethylene by the 12αꢀ(S) isomer for 12αꢀ
OTHP ethers by keeping with the change in the order of optical rotations.
Assignments for the carbon atom resonances of the steroid skeleton were made using data
64ꢀ70
reported for related hydroxysteroid analogs.
The stereochemistries of 7αꢀOꢀsubstitutedꢀ5αꢀ
dihydroprogesterone derivatives 28 and 29a,b were also confirmed by comparison with NMR
6
6
data for 5αꢀcholestanꢀ7α/βꢀol/acetate isomers according to the original article erroneously
13
64
transcribed for 7βꢀisomers in a reference review on C NMR of steroids.
Assignments for the carbon atom resonances of 7αꢀ, 11αꢀ, 12αꢀ, and 17αꢀOTHP ether rings
1
13
71,72
were made by comparison with reported H/ C NMR data on steroidal OTHP ethers.
A
ACS Paragon Plus Environment
12

Journal of Medicinal Chemistry
Page 14 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
strong characteristic difference between the (R) and (S) isomers was observed for the
dioxygenated Cꢀ2’ carbon atom resonance which was found either at ca. 100ꢀ103 ppm for 7αꢀ
(R)ꢀ, 11αꢀ(R)ꢀ and 12αꢀ(S)OTHP isomers or at ca. 95ꢀ98 ppm for the opposite isomers. Such a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
marked difference did not appear for 17αꢀOTHP ethers that showed Cꢀ2’ at ca. 96ꢀ97 ppm for
both R/S isomers with only a very slight downfield shift (< 1 ppm) for the (R) isomer.
1
3
The 7αꢀ, 11αꢀ, 12αꢀ and 17αꢀ(R/S)OTHP isomers were found to influence the C resonances
of carbon atoms of the steroid skeleton at short and long distance as well as of those of other
substituents. In all cases the 7αꢀ, 11αꢀ and 17αꢀOTHP (R) isomers and the 12αꢀOTHP (S) isomer
could be readily differentiated from the opposite isomers by a much stronger deshielding of the
steroidal carbon atom bearing the OTHP substituent (ca. 7ꢀ8 ppm for 11αꢀOTHP, ca. 5 ppm for
7αꢀ and 12αꢀOTHP ethers but only ca. 3 ppm for 17αꢀOTHP ether) as well as of the adjacent
steroidal methylene carbon atom (ca. 3ꢀ4 ppm).
1
H NMR data revealed few additional characteristic effects of (R/S)OTHP isomers. For the
7αꢀ, 11αꢀ, 12αꢀ and 17αꢀOTHP ethers, R/S isomerism induced slight shifts mainly on Hꢀ2’ of the
OTHP ether group, on the steroidal Hꢀ7, Hꢀ11 and Hꢀ12 protons, and on the 18ꢀ, 19ꢀ and 21ꢀMe
groups. Combining these effects helped to characterize the isomers present in the bisꢀOTHP
ether derivatives 43aꢀd, 24a,b and 47. For all 7αꢀOꢀsubstituted 3ꢀoxoꢀ5βꢀHꢀderivatives, the
1
52,53
reported characteristic deshielding of the H NMR signal of the axial 4αꢀH
was observed
between 2.9 and 3.5 ppm.
BIOLOGICAL RESULTS AND DISCUSSION
ACS Paragon Plus Environment
13

Page 15 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Protocols of biological evaluation. The evaluation of the relative potencies of the different
substituted steroid derivatives to modulate Pꢀgpꢀmediated multidrug resistance was made by
using K562/R7 erythroleukemia cells which are highly resistant to doxorubicin. The identity of
ABC transporters was analyzed by RTꢀqPCR which revealed that these cells contain a high level
of ABCB1 mRNA (308192 copies normalized to G3PDH gene) indicating a predominant
presence of Pꢀgp, confirmed by Western blot (cf Supporting Information). Conversely, the
amount of ABCC1 mRNA corresponding to MRP1 was very low (1664 copies) while the
mRNAs for ABCC2/MRP2) or ABCG2/BCRP remained hardly detectable (< 70 copies). The
screening protocol was based on four criteria: (i) shortꢀterm effects on intracellular accumulation
of daunorubicin measured by flow cytometry after 1 h incubation (expressed as an accumulation
index representing the ratio of daunorubicin fluorescence in the presence and absence of
steroidal modulator), (ii) longꢀterm effects on chemosensitization of doxorubicin cytotoxicity
measured with the classical 3ꢀ(4,5ꢀdimethylthiazolylꢀ2)ꢀ2,5ꢀdiphenyltetrazolium bromide reagent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(MTT) for the assay of cell survival (expressed as IC50 value for doxorubicin), including
comparisons with the reference modulators cyclosporin A and verapamil as well as the two
2
9,30
established steroid modulators, progesterone and RU486,
(iii) restoration of chemosensitivity
to doxorubicin of resistant K562/R7 cells versus sensitive K562 parental cells, (iv) evaluation of
intrinsic toxicity of modulators on parental K562 nonꢀresistant cells.
Accumulation assays were performed at high concentrations of both daunorubicin (10 µM)
and tested modulator (10 µM) in order to include the detection of molecules which have low or
very moderate activities. For the most efficient steroid modulators, the accumulation was also
measured using lower 5 µM concentrations, at which the reference modulator, cyclosporin A,
was found to restore a maximal accumulation of daunorubicin similar to the one observed in
ACS Paragon Plus Environment
14

Journal of Medicinal Chemistry
Page 16 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
sensitive cells. Chemosensitization assays were performed at a low 1 µM concentration of
steroid modulators using a concentration range of doxorubicin comprised between 0.1 and 100
µM. Intrinsic cytotoxicities of modulators were evaluated in sensitive K562 parental cells using
high concentrations (10 µM) of modulators (Tables 1 and 2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Accumulation assays (daunorubicin) and chemosensitization assays (doxorubicin
cytotoxicity). Pregnꢀ4ꢀeneꢀ3,20ꢀdione derivatives. A preliminary evaluation of the effects of
progesterone on daunorubicin accumulation in K562/R7 cells revealed only a very moderate
modulating activity on Pꢀgpꢀmediated drug efflux (Table 1) confirmed by a weak
chemosensitization of doxorubicin cytotoxicity whereas both 11αꢀ and 17αꢀhydroxyprogesterone
were inactive.
Evaluations of the monosubstituted 11αꢀ, 17αꢀ and 21ꢀOTHP ether derivatives of
progesterone showed 11αꢀ(S)OTHP isomer 1b and 17αꢀ(R/S)OTHP isomers 2a,b increased
2
9,30
accumulation (Table 1) similarly to the reference 11ꢀpꢀdimethylaminophenyl steroid RU486.
n the other hand, the weak effects on both accumulation and chemosensitization of the 21ꢀ
0
(R+S)OTHP ether 3 contrasted with the efficiency of 21ꢀpꢀdimethylaminobenzoate derivatives of
26
dichlorisone or 17ꢀdeoxydexamethasone.
5
βꢀPregnaneꢀ3,20ꢀdione derivatives. With a view to facilitate chemical modifications such as
the introduction of stable 7ꢀOꢀalkoxy/acyloxy substituents as well as to limit potential
17,18
interactions with hormone receptors,
5βꢀdihydroprogesterone skeleton.
the progesterone skeleton was replaced by the saturated
ACS Paragon Plus Environment
15

Page 17 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A preliminary comparison between unsubstituted 5βꢀdihydroprogesterone and progesterone
revealed similar activities on daunorubicin accumulation whereas both 11αꢀ and 17αꢀhydroxyꢀ
5βꢀdihydroprogesterone 8 and 10 were inactive (Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The monosubstituted 7αꢀ(R+S)ꢀ, 11αꢀ(R/S)ꢀ and 17αꢀ(R/S)OTHP derivatives of 5βꢀ
dihydroprogesterone 7, 9a,b and 11a,b, were all more efficient on accumulation than 5βꢀ
dihydroprogesterone (Table 1). Only minor differences of both accumulation and
chemosensitization were found according to 11αꢀ or 17αꢀsubstituent positions while
chemosensitization was improved for (S) isomers in both cases. The 7αꢀOTHP ether 7 proved to
be more efficient than the 7αꢀbenzoate 6 on both accumulation and chemosensitization. No
attempt was made to prepare the 21ꢀOTHP ether owing to the poor efficiency of the progesterone
analog 3.
These preliminary results on Oꢀmonosubstituted steroids stimulated a strategy focused
towards the access to 7α,11αꢀ or 11α,17αꢀOꢀdisubstituted derivatives of 5βꢀdihydroprogesterone
in order to evaluate possible beneficial additive/synergic effects, using simple benzoate ester and
(
R/S)OTHP ether substituents.
The 7α,11αꢀdibenzoate 21 induced a moderate increase of accumulation but a nearly twoꢀfold
increase of chemosensitization as compared with the effects of the 7αꢀbenzoate 6. The two
α,11αꢀbisꢀ[(R+S),(R/S)]OTHP isomers 24a,b induced similar moderate effects on
7
accumulation, comparable with those observed for the homologous 7α,11αꢀdibenzoate 21 as well
as for the corresponding 7αꢀ(R+S)ꢀ and 11αꢀ(R/S)OTHP isomers 7 and 9a,b. On the other hand,
both isomers 24a,b stimulated chemosensitization much more strongly than the dibenzoate 21
and the 11αꢀ(R/S)OTHP isomers 9a,b devoid of a 7αꢀOTHP substituent.
ACS Paragon Plus Environment
16

Journal of Medicinal Chemistry
Page 18 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The 11α,17αꢀbisꢀOTHP ether containing the four isomers 43aꢀc showed a significant activity
only for the TLC fraction enriched up to 70% of 11αꢀ(S),17αꢀ(R) isomer 43a. This fraction
produced a moderate effect on accumulation associated with a nearly fourꢀfold increase of
chemosensitization as compared with the 11αꢀ(S)ꢀ and 17αꢀ(R)OTHP isomers 9b and 11a,
despite similar accumulations. This improvement is in keeping with the strong potency of the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
0
1
1,17αꢀbisꢀpꢀdimethylaminophenyl steroid RU49953 as compared with the monoꢀ11ꢀpꢀ
dimethylaminophenyl steroid RU486.
A more refined strategy consisted to replace only one of the two ester groups of the 7α,11αꢀ
dibenzoate 21 by a tetrahydropyranyl ether. A highly significant improvement of modulating
properties was observed for the heteroꢀOꢀdisubstituted 7αꢀbenzoyloxyꢀ11αꢀ(R)OTHP ether 22a
which induced a threeꢀfold increase of both accumulation and chemosensitization (accumulation
index: 2.9, IC : 0.5 µM) as compared with the dibenzoate 21. The 11αꢀ(S)OTHP
5
0
diastereoisomer 22b proved to be much less active with an effect on accumulation close to that
of dibenzoate 21 but associated with a much lower efficiency on chemosensitization (Table 1).
On the other hand, changing the 7αꢀbenzoate of compound 21 to obtain the 7αꢀ(R)OTHPꢀ11αꢀ
benzoate 36a brought an increase of accumulation nearly reaching the high level observed for its
regioisomer 22a . A lower effect was found for the 7αꢀ(S) diastereoisomer 36b. However, the 7αꢀ
(R/S) isomers 36a,b showed only weak effects on chemosensitization.
5αꢀHꢀDihydroprogesterone derivatives. The 7α,11αꢀdibenzoate 28 induced a moderate
accumulation identical to that of its 5βꢀH analog 21 but was associated with a much more potent
chemosensitizing effect (IC : 0.4 µM). The replacement of the 11αꢀbenzoate group by the 11αꢀ
50
(R)OTHP ether in compound 29a decreased both accumulation and chemosensitization in
ACS Paragon Plus Environment
17

Page 19 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
contrast to the strong improvements found for the 5βꢀH analog 22a. The 11αꢀ(S)OTHP isomer
2
9b induced the same accumulation as its 5βꢀH analog 22b but associated with an improvement
of chemosensitization, although of moderate intensity. A comparison of the influence of the two
1αꢀ(R/S)OTHP isomers between the 5αꢀ and 5βꢀdihydroprogesterone series revealed a more
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
favorable role of the 11αꢀ(R)OTHP isomers for the isomeric pairs 29a,b and 22a,b, but not for
the 5βꢀHꢀisomers 24a,b.
5βꢀcholanꢀ3ꢀone derivatives. Similar evaluations were extended to methyl 7α,12αꢀOꢀ
disubstituted 3ꢀoxoꢀ5βꢀcholanꢀ24ꢀoate derivatives (Table 2). The 7α,12αꢀdibenzoate 46 was
found far less efficient in both accumulation and chemosensitization assays than its 5βꢀ
dihydroprogesterone analog 21. In contrast, the 7α,12αꢀbisꢀOTHP ether 47, despite the absence
of separation of R/S isomers, induced a marked increase of both accumulation and
chemosensitization (accumulation index : 3.5, IC : 0.7 µM). These effects differ from the
50
modest accumulation associated with a stronger chemosensitization observed for the 7α,11αꢀbisꢀ
OTHPꢀ5βꢀdihydroprogesterone 11α(R/S) analogs 24a,b. A parallel evaluation of the 7α,12αꢀbisꢀ
ODHPE analog 48 (accumulation index: 3.7, IC : 0.6 µM) revealed a slightly stronger
5
0
improvement of both accumulation and chemosensitization than observed with the 7α,12αꢀbisꢀ
OTHP ether 47 although a possibility remains that increased effects could occur for compound
47 after separation of OTHP isomers.
Replacing the 12αꢀbenzoate group of the 7α,12αꢀdibenzoate 46 by a 12αꢀOTHP ether
significantly increased both accumulation and chemosensitization for the 12ꢀ(R)OTHP ether 52b
whereas a stronger increase of accumulation was associated to a weak influence on
chemosensitization for the 12ꢀ(S) isomer 52a. Much more important effects resulted from
ACS Paragon Plus Environment
18

Journal of Medicinal Chemistry
Page 20 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
replacing the 7αꢀbenzoate of compound 46 by a 7αꢀOTHP ether, as shown by the strong
improvements of accumulation and chemosensitization found for either the 7ꢀ(R) isomer 51a
(
accumulation index: 2.9, IC : 0.4 µM) or the 7ꢀ(S) isomer 51b (accumulation index: 3.3, IC :
5
0
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.4 µM) leading to effects challenging those of the aboveꢀmentioned active compound, 7αꢀ
benzoyloxyꢀ11αꢀ(R)OTHPꢀ5βꢀdihydroprogesterone 22a.
A further extension was made by replacing one of the two 7α,12αꢀdibenzoate groups of
compound 46 by a OMOM ether having a more simple dioxygenated structure than an OTHP
ether. The 12αꢀOMOM ether 55 showed increased levels of accumulation and
chemosensitization slightly above those found for the corresponding 12αꢀ(S/R)OTHP isomers
52a,b. The 7αꢀOMOM ether 56 produced the most significant effects observed in this study on
both accumulation and chemosensitization (accumulation index: 3.8, IC : 0.2 µM).
5
0
A complementary study was undertaken using a more limited number of similarly Oꢀ
substituted 5βꢀcholanꢀ3ꢀone analogs in which the terminal 24ꢀCOOMe ester group of the side
chain was converted into a 24ꢀCH OMe ether group, potentially more resistant to hydrolysis,
2
thus providing a potential advantage for use in vivo. The 7α,12αꢀdibenzoate ether 62 proved to
be a poorly effective modulator as found for its homologous dibenzoate 46 of the 24ꢀCOOMe
series. The 7αꢀbenzoyloxyꢀ12αꢀ(S/R)OTHP isomers 67a,b, induced significant increases of
accumulation, as observed for 24ꢀCOOMe analogs. In contrast, the twoꢀfold increase on
chemosensitization previously found for 12ꢀ(R) isomer in the 24ꢀCOOMe series was observed in
this case for the 12ꢀ(S) isomer 67a.
The 7αꢀ(R/S)OTHPꢀ12αꢀbenzoate isomers 66a,b both induced a marked improvement of
accumulation as observed for their 24ꢀCOOMe analogs 51a,b but associated with an opposite
ACS Paragon Plus Environment
19

Page 21 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
order of effects of (R/S) isomerism and a much lower stimulation of chemosensitization which
remained comparable with the levels found for the 7αꢀ(R/S)OTHPꢀ11αꢀbenzoate isomers 36a,b.
A comparison was established with the modulating properties of the hydroxyl steroidal
precursors in the view to determine whether some positions were more favorable either to
increase activity after esterification or etherification or to limit losses of activity after hydrolysis.
Among the hydroxyl precursors evaluated in this study, a slight activity on accumulation was
observed for the 7α,12αꢀdiol derivative 45, the 7αꢀbenzoyloxyꢀ11αꢀol derivative 20 and the 7αꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OMOMꢀ12αꢀol derivative 54. In contrast, the 7αꢀhydroxyꢀ11αꢀbenzoate 37 and 7αꢀhydroxyꢀ
1
2αꢀOMOM ether 53 were able to stimulate a significant accumulation. Benzoylation of the
1αꢀ and 12αꢀhydroxyl groups leading to compounds 21 and 56 produced a much stronger effect
1
on accumulation than benzoylation of the 7αꢀhydroxyl group leading to compounds 21 and 55.
Such a difference suggests a lesser sensitivity to the presence of the more hindered 7αꢀhydroxyl
group. Blocking hydroxyl groups by ester or ether groups always increased accumulation except
for the 7α,12αꢀdibenzoates 46 and 62 for which accumulation remained low.
Restoration of chemosensitivity to doxorubicin of resistant R7 cells versus sensitive K562
parental cells. Four representative Oꢀdisubstituted steroidal modulators were tested, two of the
5βꢀdihydroprogesterone series: the 7αꢀbenzoyloxyꢀ11αꢀ(R)OTHP ether 22a and 7α,11αꢀbisꢀ
[
(R+S), (S)]OTHP ether 24b and two of the methyl 3ꢀoxoꢀ5βꢀcholanꢀ24ꢀoate series: the 7α,12αꢀ
bisꢀODHPE derivative 48 and 7αꢀOMOM,12αꢀbenzoate 56, all with a strong potency to increase
both accumulation and chemosensitization. The levels of cytotoxicity of doxorubicin on resistant
K562/R7 cells restored after chemosensitization with these four modulators were compared with
the cytotoxicity of doxorubicin alone on the sensitive parental K562 cells. As shown by the doseꢀ
ACS Paragon Plus Environment
20

Journal of Medicinal Chemistry
Page 22 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
response curves (Figure 2) a large proportion of sensitivity could be restored at a low 1 µM
modulator concentration, according to the respective chemosensitization potencies but without
reaching the level of sensitive parental K562 cells.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Intrinsic toxicity. The intrinsic toxicity of steroidal modulators, expressed as percentage of
surviving cells (Tables 1 and 2) was measured on the nonresistant parental K562 cell line after
incubation for 72 h at a high 10 µM concentration, leading to values ranging from 43.5 % (most
toxic compound 56) up to 100% (nontoxic compound 22b). The 5βꢀdihydroprogesterone
skeleton appeared more favorable (values above 60% for all tested derivatives) than the 5βꢀ
cholanꢀ3ꢀone skeleton. Among the most potent chemosensitizing 5βꢀdihydroprogesterone
derivatives, the 7α,11αꢀbisꢀ[(R+S),(R)]OTHP isomer 24a (86.7% cell survival) appeared as the
less toxic one, followed by the 7αꢀbenzoyloxyꢀ11αꢀ(R)OTHP ether 22a (80.8%). Among the Oꢀ
disubstituted methyl 3ꢀoxoꢀ5βꢀcholanꢀ24ꢀoate derivatives, a cell survival above 60% was
observed only for the two strongly efficient 7α,12αꢀbisꢀODHPE derivative 48 (76.3%) and 7αꢀ
(R+S),12αꢀ(R+S)ꢀbisꢀOTHP ether 47 (63.2%). A higher toxicity was observed for the two very
active 7αꢀ(R/S)OTHPꢀ12αꢀbenzoate isomers 51a,b (51.3 and 44.6%) and for 7αꢀOMOMꢀ12αꢀ
benzoate 56 (43.5%). Among the Oꢀdisubstituted 24ꢀmethoxyꢀ5βꢀcholane derivatives only the
7αꢀbenzoyloxyꢀ12αꢀ(S)OTHP ether 67a showed a low intrinsic toxicity (75.8% cell survival) but
had only a weak modulating potency. However, at the 1 µM concentration employed for the
evaluation of chemosensitization potency on resistant K562/R7 cells, none of the compounds
reported in this study showed any significant toxic effects in systematic controls made in the
absence of the cytotoxic agent doxorubicin.
ACS Paragon Plus Environment
21

Page 23 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CONCLUSION
The introduction of Oꢀtetrahydropyranylethers, Oꢀ(5’,6’ꢀdihydroꢀ2’Hꢀpyranꢀ4’ꢀyloxy)ether
and Oꢀmethoxymethyl ethers as well as of benzoate esters on appropriate steroidal hydroxyl
precursors, was found able to provide a simple access to potent modulators of Pꢀgpꢀmediated
drug efflux, much more efficient than progesterone on K562/R7 MDR cells, especially using
conveniently Oꢀdisubstituted derivatives. The use of OTHP ethers resulted in various efficiencies
according to the position, the R/S isomerism and the possible occurrence of different
axial/equatorial configurations of the anomeric oxygen as shown in this study by Xꢀray
crystallography of monosubstituted steroid models. This kind of modifications offers an
alternative or complementary approach to reported methods for obtaining steroid modulators
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
8
based either on 7αꢀpꢀthiophenylamino groups introduced on progesterone or on the favorable
7
pꢀdimethylamino phenyl group present in 7ꢀ or 11ꢀmonosubstituted derivatives of estradienꢀ3ꢀ
2
9,30
one or of estradiol, on a 11,17ꢀdisubstituted estradiol derivative
or on 21ꢀmonosubstituted
2
6
corticoid derivatives.
The classification of the efficiency of steroid derivatives was made according to three criteria
of cytotoxic drug accumulation, chemosensitization and intrinsic toxicity (Tables 1 and 2). The
Oꢀdisubstituted steroidal derivatives were found particularly apt to provide Pꢀgp modulators
more potent than the reference steroid RU486. The most efficient compound in 5βꢀ
dihydroprogesterone series was the 7αꢀbenzoyloxyꢀ11αꢀ(R)OTHP ether 22a. A stronger
efficiency was found in the 5βꢀcholanꢀ3ꢀone series for the 7αꢀOMOMꢀ12αꢀbenzoate 56,
followed by the 7α,12αꢀdisubstituted derivatives 48, 47 and 51b, but was associated with an
increased intrinsic toxicity.
ACS Paragon Plus Environment
22

Journal of Medicinal Chemistry
Page 24 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
For several other compounds, accumulation did not appear to be well correlated with
chemosensitization. Chemosensitization could be either relatively too strong, as found for the
5
αꢀHꢀ7α,11αꢀdibenzoate 28 and the 7α,11αꢀbisꢀ[(R+S),(R/S)]OTHP isomers 24a,b or relatively
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
too weak, as for the 7αꢀ(R/S)OTHPꢀ11αꢀbenzoate isomers 36a,b, the 7αꢀbenzoyloxyꢀ12αꢀ
(S)OTHP isomer 52a and the 7αꢀ(R/S)OTHPꢀ12αꢀbenzoate isomers 66a,b. This lack of
30
correlation, also reported for RU486, raises unresolved questions, especially concerning the
evolution of the intracellular doxorubicin concentration along the timeꢀscale of
chemosensitization effects, well beyond that of the shortꢀtime fluorescence measurements of
accumulation. The longꢀterm accumulation effects and/or the activation of mechanisms
contributing to chemosensitization might be influenced by the structural sensitivity of
modulators to metabolic transformations and to chemical or biochemical hydrolysis of ether or
ester groups according to their position and stereochemistry. Moreover, little is known on the
mechanisms of Pꢀgp inhibition which might differ according to the structures of modulators and
to the possible existence of different sites of interaction yet to be characterized.
For more detailed biological investigations in view of in vivo studies, the 5βꢀ
dihydroprogesterone series were chosen as a first priority, particularly the 5βꢀHꢀ7αꢀbenzoyloxyꢀ
11αꢀ(R)OTHP ether 22a which is more efficient on both accumulation and chemosensitization
than the three tested reference modulators RU486, verapamil and cyclosporin A. This lead
compound has a strong effect on accumulation, potentially representative of a strong inhibition
of Pꢀgp drugꢀtransport. Other advantages of compound 22a are its low intrinsic toxicity, its
absence of binding to progesterone receptors and its moderate activation of PXR (Alameh, G.
unpublished results). Other promising candidates from 5βꢀcholanꢀ3ꢀone series might benefit
ACS Paragon Plus Environment
23

Page 25 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
from further optimizations especially modifications of the C17ꢀsideꢀchain including conversion
to a 17βꢀacetyl group in the view to decrease intrinsic toxicity.
Most of these results indicate that inhibition of Pꢀgpꢀmediated drug efflux is very sensitive to
the structural features of steroidal modulators. Such a sensitivity suggests direct interactions with
privileged parts of Pꢀgp in keeping with numerous studies claiming either a competition at the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
drug efflux site(s) or an interaction at modulation site(s) inducing conformational transmission
7
2
effects. However, the assessment and identification of binding sites for modulators as well as
the validation of structureꢀactivity relationships might be complicated by site heterogeneity
17,38,49
according to the structures of the modulator and of the effluxed drug
and by the lack of
data concerning correlations between binding parameters and Pꢀgp inhibition potency. The
localization of interaction sites using affinity labeling probes derived from the most efficient
modulators identified in this study may facilitate molecular modeling experiments using the
6
recently reported 3D structure of mouse Pꢀgp so as to improve the design of further generations
of nontoxic adjuvants for cancer chemotherapies.
EXPERIMENTAL SECTION
1
13
Chemistry. The synthetic protocols and physicoꢀchemical data, including H/ C NMR and
HRMS characterizations of all steroid compounds, are available freeꢀofꢀcharge as Supporting
Information.
ACS Paragon Plus Environment
24

Journal of Medicinal Chemistry
Page 26 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Biology and biochemistry. Cell culture. The human erythroleukemic K562 cell line and the
74
resistant K562/R7 MDR cell line, obtained by treatment of K562 cells with doxorubicine. Cells
were cultured in RPMI 1640 supplemented with 10% foetal calf serum, Lꢀglutamine (2 mM),
glucose (0.3%), sodium pyruvate (1 mM), penicillin (200 U/mL), and streptomycin (100 µg/mL).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cells were maintained at 37 °C in a 5% CO atmosphere. Media and supplements were obtained
2
from InvitrogenꢀGibco (Paisley, UK), culture flasks from BDꢀFalcon (Meylan, France).
Isolation of RNA, and RTꢀqPCR. Total RNA was extracted from K562/R7 cells (RNeasy Mini
kit; Qiagen). Firstꢀstrand cDNAs were first synthesized from 5 µg of total RNA in the presence
TM
of 50 units of the Superscript
II reverse transcriptase using both random hexamers and oligo
(
dT) primers (Invitrogen Kit). qPCR was performed in a final volume of 20 µl containing 5 µl of
a 60ꢀfold dilution of the RT reaction medium, 15 µl of reaction buffer from the FastStart DNA
Master Plus SYBR Green Kit (Roche Diagnostics), and 10 pmol of the specific forward and
7
5
reverse primers (Operon Biotechnologies, Germany). Standard curves were prepared for each
target and reference gene. Each assay was performed in duplicate, and validation of the qPCR
runs was assessed by evaluation of the melting temperature of the products and by the slope and
error obtained with the standard curve. The analyses were performed using LightꢀCycler
software (Roche Diagnostics). Results are expressed as relative levels after normalization by
76
G3PDH mRNA.
Intracellular accumulation of daunorubicin measured by flow cytometry. The accumulation of
daunorubicin in the presence of steroid modulators in resistant K562/R7 cells was measured by
77
ꢀ6
flow cytometry using a reported procedure. Cells (1.10 cells/tube) were incubated for 1 h at
7 °C in 1 mL RPMI 1640 medium containing daunorubicin at 10 µM in the presence or absence
3
of steroids at 10 µM. The cells were then washed twice with iceꢀcold PBS and kept on ice until
ACS Paragon Plus Environment
25

Page 27 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
analysis by flow cytometry on a FACSꢀII (BectonꢀDickinson Corp., Mountain View, CA).
Assays were performed in duplicate in at least three separate experiments. Cyclosporin A, an
7
8
inhibitor of Pꢀgp, was used as a positive control.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cytotoxicity assay using MTT reagent. Cell viability was determined on exponentially
growing resistant K562/R7 cells using the MTT assay. This assay is based on the conversion by
metabolically active cells of the MTT reagent into formazan crystals. K562/R7 cells (8 000/well)
were seeded into 96ꢀwell plates and incubated with 100 µL of RPMI medium for 24 h. For each
assay, steroid or control drug (1 µM) were added in the absence or presence of eight
ꢀ7
ꢀ4
concentrations of doxorubicin (ranging from 10 to 10 M) in a final volume of 200 µL of
medium. After incubation for 72 h, 20 µL of MTT reagent (5 mg/mL in PBS buffer) were added
to each well and the plate was further incubated for 4 h at 37 °C, allowing viable cells to change
the yellow MTT into darkꢀblue formazan crystals. Supernatants were carefully discarded and 100
µL of isopropanol containing 10% HCl 1 M were added to dissolve the formazan crystals.
Absorbance in each well was determined at 570 nm using a microplate reader (MultisKan EX,
Thermo Electron). Assays and controls, using support medium only, were performed in
triplicates in at least three separate experiments. Results were expressed as the percentage of
7
8
surviving cells in each well compared to untreated cells. IC50 values, defined as the
concentration of doxorubicin inhibiting cell growth by 50%, were estimated using the Sigma Plot
11.0 software.
Intrinsic Cytotoxicity of Steroid Inhibitors. K562 cells (8 000/well) were seeded into 96ꢀwell
plates and incubated with 100 µL of RPMI medium for 24 h. Steroids (10 µM) were then added
in a final volume of 200 µL of medium. After 72 h incubation, cells were treated with MTT as
ACS Paragon Plus Environment
26

Journal of Medicinal Chemistry
Page 28 of 56
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
described above. Each assay was performed in triplicate and results were expressed as the
percentage of surviving cells in each well compared to untreated cells.
Statistical Analysis. Statistical analysis was performed using three replicates in at least three
separate experiments. The data, expressed as the mean ± SE, were analyzed by oneꢀway
ANOVA using the Sigma Plot 11.0 software. Differences for which p < 0.05 were considered
significant.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
27

Page 29 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 1. ORTEP views of diastereoisomeric 17αꢀ(R/S)OTHP ethers 2a/b, 11αꢀ(R/S)OTHP
ethers 22a /b and 7αꢀ(R/S)OTHP ethers 36a/b.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
28