Neurosteroid analogues. 18. Structure-activity studies of ent-steroid potentiators of γ-aminobutyric acid type a receptors and comparison of their activities with those of alphaxalone and allopregnanolone

Article

pubs.acs.org/jmc

Neurosteroid Analogues. 18. Structure−Activity Studies of ent-

Steroid Potentiators of γ‑Aminobutyric Acid Type A Receptors and

Comparison of Their Activities with Those of Alphaxalone and

Allopregnanolone

Mingxing Qian,^†Kathiresan Krishnan,^†Eva Kudova,^†,▽Ping Li,^‡Brad D. Manion,^‡Amanda Taylor,^§

George Elias,^§Gustav Akk,^‡,#Alex S. Evers,^†,‡,#Charles F. Zorumski,^§,⊥,#Steven Mennerick,^§,⊥,#

and Douglas F. Covey*^{,†,‡,§,#}

^†Departments of Developmental Biology, ^‡Anesthesiology, ^§Psychiatry, and ^⊥Anatomy and Neurobiology, Washington University in

St. Louis School of Medicine, St. Louis, Missouri, 63110, United States

^#The Taylor Family Institute for Innovative Psychiatric Research, Washington University in St. Louis, St. Louis, Missouri, 63110,

United States

S

* Supporting Information

ABSTRACT: A model of the alignment of neurosteroids and ent-neuro-

steroids at the same binding site on γ-aminobutyric acid type A (GABA_A)

receptors was evaluated for its ability to identify the structural features in ent-

neurosteroids that enhance their activity as positive allosteric modulators of this

receptor. Structural features that were identiﬁed included: (1) a ketone group

at position C-16, (2) an axial 4α-OMe group, and (3) a C-18 methyl group.

Two ent-steroids were identiﬁed that were more potent than the anesthetic

steroid alphaxalone in their threshold for and duration of loss of the righting

reﬂex in mice. In tadpoles, loss of righting reﬂex for these two ent-steroids

occurs with EC₅₀values similar to those found for allopregnanolone. The

results indicate that ent-steroids have considerable potential to be developed as anesthetic agents and as drugs to treat brain

disorders that are ameliorated by positive allosteric modulators of GABA_Areceptor function.

In a subsequent study of 7-OBn and 11-OBn substituted

analogues of compounds 1 and ent-1, we provided evidence in

support of the hypothesis that one enantiomer of androsterone

is inverted relative to the other in the neurosteroid modulation

site on the GABA_Areceptor.¹³We report herein a continuation

of these enantioselectivity studies with the immediate goal of

learning more about the SAR of ent-steroid modulators of

GABA_Areceptors and the long-term goal of identifying new ent-

steroids with therapeutic potential.

A heuristic model for the alignment of the enantiomers of

androsterone at the neurosteroid binding site in the GABA_A

receptor has been used to select the ent-steroids chosen for

synthesis and evaluation (Figure 1A). The model posits that the

A-rings of both enantiomers are identically located in the

neurosteroid binding site as inferred from our previous study.¹³

This model, when considered in the context of what has been

shown in previous studies of steroid modulators of GABA_A

receptors, was then used to make the following series of

predictions for the activity of new ent-steroid analogues.

First, previous structure−activity (SAR) studies of steroid

anesthetics acting at GABA_Areceptors have shown that

INTRODUCTION

■

Currently there is considerable interest in neurosteroid

physiology and in the development of neurosteroid analogues

for the treatment of a variety of disorders of central nervous

system (CNS) function and as new intravenous general

anesthetics.¹⁻⁴The vast majority of the medicinal chemistry

done in the ﬁeld of neuroactive steroids is focused on

developing new neurosteroid analogues having the absolute

conﬁguration of naturally occurring steroids.⁵The potential of

ent-neurosteroids, steroids with an absolute conﬁguration

opposite to that of naturally occurring steroids, to be developed

for these purposes has not been addressed even though ent-

steroids could have a diﬀerent spectrum of actions in the CNS

and diﬀerent routes of metabolism.

As an initial approach to evaluating the potential of ent-

steroids as new anesthetic agents and CNS drugs, we have been

engaged for a number of years in exploring the molecular basis

for the enantioselectivity of neurosteroid action at γ-amino-

butyric acid type A (GABA_A) receptors.⁶⁻¹⁰Androsterone (1),

a naturally occurring neurosteroid, is a weak potentiator of γ-

aminobutyric acid (GABA)-mediated chloride currents at

GABA_Areceptors (Chart 1).^11,12We reported previously that

its enantiomer, ent-androsterone (ent-1), is a more eﬀective

positive allosteric modulator of this ion channel than steroid 1.⁹

Received: October 9, 2013

Published: December 12, 2013

171

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

inverted orientation of compound ent-1 relative to 1, the model

predicts that introducing a 2α-OMe group as a modiﬁcation of

compound ent-1 will yield inactive analogue ent-2, whereas

modiﬁcation with a 4α-OMe group will give active analogue

ent-3.

Chart 1

Second, the model places the oxygens in the ketone groups at

C-17 in the androsterone enantiomers in nearly the same

location. However, results obtained with highly active

anesthetic steroid 4 suggest that the C-17 position may not

be the optimal location for the carbonyl group in ent-steroids.¹⁵

An alignment of the A-rings of steroid 4, a synthetic analogue of

endogenous neurosteroid allopregnanolone (5), with com-

pound ent-1 suggests that the carbonyl group would be better

placed at C-16 (Figure 1B). This prediction, along with that of

an expected loss of activity for the corresponding 16-

ketosteroids, was evaluated using the enantiomer pairs (7−9

and ent-7−ent-9).

Third, the model predicts that the C-18 and C-19 methyl

groups are similarly located in the neurosteroid binding site.

Either, both, or neither of these methyl groups may be

important for orientation/activity of the enantiomers in the

neurosteroid binding site. Previous studies have shown that the

C-18 Me group can aﬀect the number of components in TBPS

binding curves (one vs two) as well as the kinetic modes of

action.^16,17This predicts that the activity of analogues of ent-1

without the C-18 methyl group will have their activities altered

more than ent-1 analogues lacking the C-19 methyl group. This

prediction was tested with compounds ent-7 and ent-10−ent-

12.

Herein, we report results with novel steroid and ent-steroid

analogues that are consistent with all three of these predictions.

In addition, we have identiﬁed ent-steroids (ent-7 and ent-9)

with potencies comparable to those of the endogenous

neurosteroid 5 and higher than those of the anesthetic steroid

alphaxalone (6).

CHEMISTRY

■

Because the synthetic pathways to either enantiomer of the

compounds described in Schemes 1−3 are identical, only those

for the ent-steroids are discussed here. Physical and

a

Scheme 1

Figure 1. Molecular modeling. (A) Proposed model of the alignment

of enantiomers 1 and ent-1 at the neursosteroid modulation site on

GABA_Areceptors. (B) Proposed model of the alignment of

compounds 4 and ent-1 at the same site. Aligments were obtained

by superimposing O-3, C-3, C-13, and C-18 for each pair of

compounds. Additional views of the superimpositions from diﬀerent

perspectives are presented in the Supporting Information along with a

view showing the superimposition of steroid 5 and analogue ent-9

(Supporting Information Figure S1).

a

Reagents: (a) Li(t-OBu)₃AlH, THF, −40 °C (82%); (b) MeSO₂Cl,

Et₃N, CH₂Cl₂(100%); (c) LiBr, DMF, 125 °C (87%, Δ²and Δ³, 5:1);

(d) 30% H₂O₂, HCO₂H, CH₂Cl₂(100%, 2,3 and 3,4 epoxides); (e)

H₂SO₄, MeOH (62%).

analogues with 2β-substituents (e.g., 2), but not 4β-substituents

(e.g., 3), are highly active.¹⁴Thus, as a consequence of the

172

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

a

spectroscopic details for the corresponding steroids, when

prepared, are given immediately after these data are reported

for the corresponding ent-steroids in the Experimental Section.

The synthesis of compound ent-1 was described previously.⁹

The synthesis of ent-2 is shown in Scheme 1. Starting material

ent-13 was prepared as previously reported.⁹Regio- and

stereoselective reduction of this compound’s 3-ketone group

with Li(t-OBu)₃AlH yielded after workup predominately

product ent-14 (82%). Mesylation of the resultant 3α-hydroxyl

group gave the corresponding mesylate ent-15 (100%), and

elimination of the mesylate group gave predominately the Δ²-

oleﬁnic product ent-16 (87%). Epoxidation of the double bond

using HCOOH/H₂O₂gave epoxide ent-17 (100%), and

opening of the epoxide with H₂SO₄/MeOH gave ent-2 (62%).

The synthesis of ent-3 is shown in Scheme 2. Starting

material ent-18 was prepared as previously reported.⁷Treat-

Scheme 3

a

Scheme 2

a

Reagents: (a) (i) C₆H₅CHO, EtOH, KOH, (ii) NaBH₄, CeCl₃,

EtOH, 0 °C, (iii) Ac₂O, Et₃N, DMAP (91−97% overall); (b) (i) O₃,

MeOH/EtOAc, −78 °C, (ii) Me₂S (82−98% overall); (c) (i) SmI₂,

THF (under N₂), (ii) Jones reagent (70−94%); (d) K₂CO₃, MeOH/

H₂O (93−98%).

a

convert 19-nortestosterone into ent-19-nortestosterone.^19,20

Starting material ent-34 was prepared as described previously⁹

and converted to the 17α-tosylate ent-35 (91%). Stereoselective

reduction of the 3-ketone group of ent-35 was achieved using

K(s-Bu)₃BH to yield product ent-36 (71%) containing the 3β-

hydroxyl group. Rearrangement of compound ent-36 to

product ent-37 (91%) was achieved using MeMgBr in reﬂuxing

toluene. Compound ent-37 is not very soluble in the solvents

used for ozonolysis of the compound’s double bond, so it was

converted to the MOM protected compound ent-38 (90%) and

then converted by ozonolysis to the diketone ent-39 (65%).

Aldol condensation of compound ent-39 using NaOH/aqueous

MeOH gave enone ent-40 (75%). Li/liq NH₃of this enone

gave product ent-41 (72%).

At this point, the MOM protecting group at the 3-position,

which was chosen for it is stability in the Li/liq NH₃reduction

step, had to be exchanged for an acetate group because the

MOM group would not survive the harsh conditions of a

subsequent CrO₃oxidation step. The MOM group in

compound ent-41 was removed using 6 N HCl/MeOH, and

the resultant hydroxyl group in compound ent-42 (92%) was

acetylated using AcOAc/Et₃N/DMAP to yield the 3-acetate

ent-43 (95%).

CrO₃/AcOH oxidation of compound ent-43 cleaved the D-

ring to yield crude dicarboxylic acid ent-44, which was isolated

after esteriﬁcation using AcCl/MeOH as diester ent-45 (51%).

Dieckmann condensation using NaOMe/THF of compound

ent-45 yielded the D-ring β-ketoester ent-46 (71%) as the only

condensation product. An analogous oxidation, esteriﬁcation,

ring closure sequence has been reported previously for (5α)-3-

ketosteroids.²¹⁻²³In the cited study, the hydrogen at the ring

fusion and the carbomethoxy group in the product were found

to be syn to each other. For this reason, we assigned the

Reagents: (a) Zn dust, AcOH (43%); (b) 30% H₂O₂, HCO₂H,

CH₂Cl₂(84%); (c) H₂SO₄, MeOH (78%).

ment of the starting material with Zn powder in HOAc gave the

(5α)- and (5β)-Δ³-oleﬁnic products in about a 1:1 ratio as

determined by NMR. Fractional recrystallization gave the pure

product ent-19 (43%). Epoxidation of the double bond using

HCOOH/H₂O₂gave epoxide ent-20 (84%), and opening of

the epoxide with H₂SO₄/MeOH gave ent-3 (78%).

In Scheme 3, we adapted a route that we used previously to

convert a steroid 3-ketone group to a 2-ketone group¹⁸to

transform the17-ketone groups of ent-1−ent-3 and ent-21 to

the 16-ketone groups of ent-7−ent-9 and ent-11, respectively.

Each starting material was condensed with benzaldehyde,

reduced to a diol using NaBH₄/CeCl₃and acetylated using

AcOAc/Et₃N/DMAP to give products ent-22 (91%), ent-25

(95%), ent-28 (97%), and ent-31 (91%). The 16-ketone groups

were then formed by ozonolysis of the exocyclic double bonds

at the 16-position to give products ent-23 (82%), ent-26 (90%),

ent-29 (87%), and ent-32 (98%), respectively. In the next step,

the 17α-acetate groups were removed using SmI₂in THF. This

reaction also results in partial reduction of the 16-ketone

groups. The crude 16-ketone, 16-alcohol product mixtures were

treated with Jones reagent to give the isolated products ent-24

(94%), ent-27 (94%), ent-30 (84%), and ent-33 (70%),

respectively. Saponiﬁcation of the 3β-acetate groups gave the

target compounds ent-7 (93%), ent-8 (97%), ent-9 (94%), and

ent-11 (98%), respectively.

The synthetic routes to the 18-nor compound ent-10

(Scheme 4) and 18,19-dinor compound ent-12 (Scheme 5)

are largely derived from methods developed previously to

prepare chrysene enantiomers from 19-nortestosterone or to

173

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

a

Scheme 4

Scheme 5

a

Reagents: (a) p-TsCl, pyr, 40 °C (91%); (b) (i) K(s-Bu)₃BH, THF,

−78 °C, (ii) NaOH, 30% H₂O₂(71% overall); (c) MeMgBr, toluene,

reﬂux (91%); (d) ClCH₂OMe, (i-Pr)₂EtN, CH₂Cl₂, 0 °C (90%); (e)

O₃, CH₂Cl₂, −78 °C, (ii) Zn dust, AcOH (65% overall); (f) NaOH,

MeOH/H₂O (75%); (g) Li/liq. NH₃, THF (72%); (h) 6 N HCl,

MeOH (92%); (i) Ac₂O, Et₃N, DMAP (95%); (j) CrO₃, AcOH, 60

°C; (k) AcCl, MeOH (51%, steps j and k); (l) NaOMe, THF, reﬂux

(71%); (m) LiCl, DMF, 160 °C (86%).

a

Reagents: (a) CrO₃, AcOH, 60 °C; (b) AcCl, MeOH (61%, steps a

and b); (c) NaBH₄, EtOH, 0 °C (74%); (d) NaOMe, THF under N₂,

100 °C (81%); (e) LiCl, DMF, 160 °C (85%); (f) p-TsCl, pyr, DMAP,

65 °C (91%); (g) n-BuLi, MeP(Ph)₃Br, C₆H₆/THF under N₂(80%);

(h) MeMgBr, toluene under N₂, 100−115 °C (90%); (i) (i) O₃,

CH₂Cl₂, −78 °C, (ii) Zn dust, AcOH (51% overall); (j) NaOH,

MeOH/H₂O (82%); (k) Li/liq. NH₃, THF (63%); (l) (i) K(s-

Bu)₃BH, THF, −78 °C, (ii) NaOH, 30% H₂O₂(70%).

stereochemistry of the 17-carbomethoxy group in product ent-

46 the α-conﬁguration. Decarbomethoxylation using LiCl in

nondried DMF gave the desired analogue ent-10 (86%). Using

carefully dried DMF gave lower yields.

regenerates the ketone group from the vinylidene group and

cleaves the endocyclic double bond to form trione 56 (51%).

Aldol condensation of trione 56 completes the 19-norsteroid to

ent-18,19-dinorsteroid transformation, yielding product ent-57

(82%). Li/liq NH₃reduction of the double bond in enone ent-

57 yields product ent-58 (63%), and regio- and stereoselective

reduction of the 3-ketone group with K(s-Bu)₃BH yields target

compound ent-12 (70%).

For the synthesis of ent-12, steroid 47 was used as starting

material.²⁴CrO₃/AcOH oxidation of steroid 47 cleaved the A-

ring to yield crude 2,3-secosteroid 48 (81%), which was

puriﬁed and isolated after esteriﬁcation using AcCl/MeOH as

the diester 49 (61%). The 17-ketone group of 2,3-secosteroid

49 was then reduced using NaBH₄to yield product 50 (74%).

Dieckmann condensation of compound 50 using NaOMe/

THF yielded the A-norsteroid 51 (81%) as the only

condensation product and the stereochemistry of the 3-

carbomethoxy group was assigned, by literature analogy,²¹⁻²³

the α-conﬁguration. Decarbomethoxylation using LiCl in

nondried DMF gave A-norsteroid 52 (85%). This part of the

synthetic sequence has the eﬀect of converting the A-ring of a

steroid into what will become the D-ring of an ent-steroid.

The last part of the synthesis of compound ent-12 involves

converting the D-ring of A-norsteroid 52 into the A-ring of ent-

steroid ent-57. Compound 52 was converted into the tosylate

53 (91%), and the A-ring ketone ring of this tosylate was

converted into the vinylidene group of A-norsteroid 54 (80%).

Treatment of compound 54 with MeMgBr in reﬂuxing toluene

gave the dioleﬁn 55 (90%). Ozonolysis of compound 55

[³⁵S]TBPS DISPLACEMENT RESULTS

■

The compounds shown in Chart 1 were evaluated as

noncompetitive displacers of [³⁵S]TBPS from the picrotoxin

binding site on the heterogeneous GABA_Areceptors found in

rat brain membranes (Table 1; Supporting Information Figure

S2 shows each analogue in the alignment orientations shown in

Figure 1 along with the IC₅₀values). The results demonstrate

ﬁrst, for reference, that the endogeneous neurosteroid 5 (IC₅₀=

74 7 nM) and compounds ent-7 and ent-9 (IC₅₀= 81 8 and

83 11 nM, respectively) are the most potent displacers of

[³⁵S]TBPS. These three compounds are all more potent than

the clinically used anesthetic steroid 6 (IC₅₀= 226 24 nM).

174

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

Table 1. Inhibition of [³⁵S]TBPS Binding by Steroids 1−9

substituent has largely opposite eﬀects on displacement

potency. For the steroids, regardless of the position of the D-

ring ketone group, displacement potency is greatly reduced by

this substituent. For the corresponding ent-steroids, the axial 4-

OMe substituent has little, if any, eﬀect on displacement

potency.

a

and Steroid Enantiomers ent-1−ent-3 and ent-7−ent-12

compd

IC₅₀(nM)

nHill

b

1

414 127

311 36

0.89 0.20

1.00 0.10

1.26 0.24

1.10 0.33

b

ent-1

2

2190 420

6490 3330

Compounds ent-10−ent-12 were evaluated to assess the

relative importance of the C-18 and C-19 Me groups for the

high displacement potency found for compound ent-7.

Compound ent-10, which lacks the C-18 Me group, is about

a 19-fold weaker displacer of [³⁵S]TBPS than compound ent-7.

Compound ent-11, which lacks the C-19 Me group, is only

about a 3-fold weaker displacer of [³⁵S]TBPS than compound

ent-7. Compound ent-12, which lacks both C-18 and C-19 Me

groups, has an IC₅₀value that is essentially the same as that of

compound ent-10. These results indicate that in this series of

ent-steroids the C-18 Me group has a substantially larger eﬀect

on [³⁵S]TBPS displacement potency than does the C-19 Me

group. This diﬀerence was anticipated based on previous results

obtained from similar studies carried out on steroids and

discussed in the Introduction. The eﬀect of the C-18 Me group

on kinetic processes aﬀecting the magnitude of potentiation of

whole cell GABA currents is reported in the next section.

ent-2

3

c

>30000

ent-3

198 25

128 11

0.98 0.11

1.44 0.15

0.89 0.06

1.10 0.11

1.03 0.05

0.92 0.07

1.13 0.14

1.12 0.13

1.56 0.51

1.15 0.15

1.06 0.14

0.86 0.06

1.05 0.23

d

4

e

5

74

7

f

6

226 24

895 49

7

ent-7

8

81

8

1270 162

3030 430

6680 2,180

83 11

ent-8

g

9

ent-9

ent-10

ent-11

ent-12

1520 230

238 21

h

2020 550

a

The results presented are from duplicate experiments performed in

triplicate. Error limits are calculated as the standard error of the mean.

Unless noted otherwise, compounds inhibited binding of [³⁵S]TBPS

ELECTROPHYSIOLOGY RESULTS

b

c

by ≥95%. Literature values.⁹A binding curve could not be

■

d

The compounds shown in Chart 1 were initially evaluated for

their ability to potentiate chloride currents elicited by 2 μM

GABA (a concentration that on average gates ∼4% of maximal

GABA response) at rat α₁β₂γ_2Ltype GABA_Areceptors

expressed in Xenopus laevis oocytes (Table 2). The results

demonstrate that compounds with a low IC₅₀value for

[³⁵S]TBPS displacement tended to exhibit detectable potentia-

tion of GABA_Areceptor currents at 100 nM, the lowest

concentration tested. All active compounds exhibited concen-

tration dependent potentiation, and the weakest compounds

failed to exhibit potentiation even at the highest concentration

tested (10 μM), suggesting their qualitative inactivity The

results shown were obtained on diﬀerent batches of oocytes

and reported potentiation values do not account for minor

variations in GABA EC₅₀and other potential variables (e.g.,

diﬀerent extent of receptor phosphorylation) that may

quantitatively aﬀect GABA and neurosteroid responsiveness.

Therefore, only qualitative comparisons of results for the

diﬀerent analogues can be made for the results reported in

Table 2.

calculated. Maximum displacement at 30 μM was 45%. Literature

values.¹⁵Literature values.⁴⁶Literature values.⁴⁸Inhibition was

e

f

g

partial at 30 μM. Maximal displacement for the calculated binding

h

curve was 65%. Inhibition was partial at 30 μM. Maximal

displacement for the calculated binding curve was 88%.

Comparison of the IC₅₀values for the steroid pair 1 and 7

shows that moving the D-ring ketone group from position C-17

to C-16 decreases displacement potency by about 2-fold. By

contrast, this same structural modiﬁcation for the compound

pair ent-1 and ent-7 increases displacement potency by about 4-

fold.

Introduction of an axial 2-OMe group lowers displacement

potency for steroids with a 17-ketone group (compare steroids

1 and 2) by about 5-fold, but for steroids with a 16-ketone

group (compare steroids 7 and 8) the substituent lowers

displacement potency only slightly (<2-fold). An axial 2-OMe

substituent causes a much larger decrease in displacement

potency for the corresponding ent-steroid pairs. For the pair

ent-1 and ent-2, the eﬀect is about 21-fold. For the pair ent-7

and ent-8, the eﬀect is about 37-fold.

Addition of an axial 4-OMe group yields steroid analogues

that are weak displacers of [³⁵S]TBPS. Steroids 3 (17-ketone)

and 9 (16-ketone) displaced only 45% and 65% of bound

[³⁵S]TPBS at the highest concentration tested (3 μM),

respectively. An opposite eﬀect on displacement potency was

found for the axial 4-OMe substituent in the enantiomeric

compounds ent-3 and ent-9. For the 17-ketone pair ent-1 and

ent-3, the axial 4-OMe group increased displacement potency

by about 1.5-fold, and for the 16-ketone pair ent-7 and ent-9

there was no signiﬁcant eﬀect of the axial 4-OMe group on

displacement potency.

In summary, an axial 2-OMe substituent has a modest

negative eﬀect on displacement potency for steroids having a

17-ketone group and little, if any, eﬀect on steroids with a 16-

ketone group. A large negative eﬀect of this substituent on the

IC₅₀value is found for the corresponding ent-steroids having

the D-ring ketone group in either position. An axial 4-OMe

In general, results found in electrophysiological evaluations

presented in Table 2 are congruent with results from

[³⁵S]TBPS binding studies presented in Table 1. In Table 2,

steroids 1 (17-ketone) and 7 (16-ketone) are qualitatively

similar to each other, and both are modest enhancers of

currents. Results in Table 2 for the axial 2-OMe substituted

compounds suggest that this substituent has a small negative

eﬀect on the concentration dependent potentiation caused by

the steroids (compare steroids 1 and 2, 7 and 8) and a large

negative eﬀect on potentiation eﬀects of the corresponding ent-

steroids (compare compounds ent-1 and ent-2, ent-7 and ent-

8). For steroid analogues 3 and 9, which have the axial 4-OMe

substituent, the results in Table 2 indicate that this substituent

eliminates potentiation of currents by the compounds at

concentrations up to 10 μM. For analogues ent-3 and ent-9, this

substituent also qualitatively diminishes potentiation of current

(compare analogues ent-1 and ent-3, ent-7 and ent-9).

However, both of these ent-steroids still show a concentration

175

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

its C-17 vinylcyano analogue steroid 4 and had an activity very

similar to that of the more active endogenous neurosteroid 5.

In Figure 2D, a quantitative comparison of the series of ent-

16-ketosteroid analogues with the diﬀerent pattern of C-18 and

C-19 Me group substitutions is shown. The results from direct

comparisons of 5 μM compound agree with the conclusion

made from the qualitative results presented in Table 2. Absence

of the C-18 Me group yields compounds with lower activity

than those having this group whether or not the C-19 Me

group is present.

To verify that the eﬀects of the two most active ent-16-

ketosteroids (ent-7 and ent-9) and endogenous neurosteroid 5

were mediated by the same site and similar mechanisms, we

examined the eﬀects of mutations that in previous studies had

been shown to aﬀect receptor modulation by steroids. These

studies utilized mutated rat α₁β₂γ_2Lreceptors expressed in HEK

293 cells. Steroid 5 potentiation of rat α₁β₂γ_2Lreceptors was

previously shown to be drastically reduced in the α₁(Q241L)-

β₂γ_2Lmutant, possibly due to the Leu residue being unable to

form a hydrogen bond with the steroid molecule.^25,26

Potentiation by compounds ent-7 and ent-9 was also reduced

in this mutated receptor (Table 3). Receptors containing

α₁(S240L)β₂γ_2Lsubunits that abolish potentiation by some 5β-

reduced steroids, but in which potentiation by steroid 5 is

maintained,²⁵are also potentiated by both ent-steriods (Table

3). The third mutated receptor on which the compounds were

tested is α₁(W245L)β₂γ₂receptor. This mutated receptor is not

potentiated by steroids, and it has been proposed that this

mutation aﬀects a general transduction element required for

steroid-induced potentiation.²⁵Neither analogue ent-7 nor

analogue ent-9 potentiates receptors carrying this mutation

(Table 3). Thus, these studies with mutated receptors are

consistent with all three compounds (ent-7, ent-9, and 5) acting

at the same potentiation sites.

Table 2. Modulation of Rat α₁β₂γ_2LGABA_AReceptor

Function by Steroids 1−9 and Steroid Enantiomers ent-1−

ent-3 and ent-7−ent-12

a

oocyte electrophysiology

compd

0.1 μM

1 μM

10 μM

(gating) 10 μM

b

1

0.97 0.02

1.27 0.29

0.84 0.01

0.93 0.04

1.08 0.21

1.21 0.02

1.17 0.04

1.26 0.14

2.91 0.57

1.01 0.16

1.42 0.02

0.92 0.04

0.91 0.01

0.88 0.08

1.41 0.07

0.93 0.02

0.97 0.02

0.80 0.02

1.41. 0.01

3.66 0.89

1.04 0.04

0.93 0.10

1.01 0.14

3.67 0.33

5.06 0.60

3.89 1.34

4.70 1.11

1.38 0.13

7.72 0.39

1.23 0.09

1.03 0.03

0.72 0.02

6.48 0.40

1.13 0.02

2.43 0.02

0.76 0.06

5.44 0.19

18.87 2.38

3.39 0.32

1.53 0.11

0.91 0.07

12.21 1.26

21.51 7.07

9.65 3.87

19.64 4.04

9.98 3.09

34.70 1.69

4.48 0.90

2.95 0.24

0.69 0.01

16.00 2.21

3.56 0.15

10.52 0.61

1.65 0.13

0.02 0.01

0.03 0.21

0.08 0.07

−0.01 0.01

0.12 0.17

0.05 0.03

0.17 0.00

0.37 0.07

0.11 0.02

−0.02 0.02

0.15 0.04

0.01 0.00

0.01 0.01

0.02 0.01

0.08 0.03

0.01 0.02

0.02 0.01

b

ent-1

2

ent-2

3

ent-3

c

4

d

5

e

6

7

ent-7

8

ent-8

9

ent-9

ent-10

ent-11

ent-12

0

0.02

a

The GABA concentration used for the control response was 2 μM.

Each compound was evaluated on at least four diﬀerent oocytes at the

concentrations indicated, and the results reported are the ratio of

currents measured in the presence/absence of added compound.

Gating represents direct current gated by 10 μM compound in the

absence of GABA, and this current is reported as the ratio of

compound-only current/2 μM GABA current. Error limits are

b

calculated as standard error of the mean (N ≥ 4). Literature values.⁹

c

d

e

Literature values.¹⁵Literature values.⁴⁶Literature values.⁴⁸

Five ent-16-ketosteroids were examined in greater detail to

determine how single-channels kinetics are aﬀected by the axial

4-OMe substituent (ent-9) or by the diﬀerent C-18 and/or C-

19 Me substituents (ent-7 and ent-10−ent-12). These studies

were performed using rat wild-type α₁β₂γ_2Lreceptors expressed

in HEK 293 cells.

Previous single-channel studies have demonstrated that

receptors activated by high concentrations of GABA produce

clusters of single-channel openings, i.e., episodes of intense

activity separated by long quiescent periods.²⁷Coapplication of

neurosteroids with GABA elicits speciﬁc changes in the

intracluster open and closed time distributions. For example,

in the presence of steroid 5, the changes are: an increase in the

mean duration of the longest-lived open state, an increase in the

fraction of events in this open state, and a decrease in the

fraction of events in the longest intracluster closed state.²⁸

Table 4 reports a statistical analysis of these parameters in the

presence of the ﬁve novel ent-16-ketosteroids. Sample single-

channel traces along with the open and closed time histograms

for channel events in the corresponding patches are shown in

Figure 3.

dependent increase in potentiation, an eﬀect not seen with

steroids 3 and 9.

Comparison of results in Table 2 for ent-16-ketosteroid

analogues ent-7 and ent-10−ent-12 shows that absence of the

C-18 Me group (ent-10) has a larger negative impact on

potentiation than does absence of the C-19 Me group (ent-11).

The analogue without the C-18 and C-19 Me groups (ent-12)

has little potentiating activity at concentrations up to 10 μM.

For quantitative purposes, several direct comparisons of

potentiating eﬀects of selected compounds at a ﬁxed

concentration on the same oocyte were made (Figure 2A−

D). In Figure 2A, the potentiation caused by 17-ketosteroids

1−3 and 16-ketosteroids 8 and 9 at 5 μM are compared. The

minor eﬀects on potentiation caused by the axial 2-OMe

substituents (steroids 2 and 8) and the major negative eﬀects of

the axial 4-OMe groups (steroids 3 and 9) on potentiation are

readily apparent when the potentiating eﬀects of the four

analogues are compared with that of reference steroid 1.

In Figure 2B, quantitative comparisons for selected ent-

steroids (0.5 μM) are shown. Changing the position of the C-

17 ketone (ent-1) group to C-16 (ent-7) enhances potentiation.

The enhancement obtained by moving the position of the

ketone group is eliminated by the addition of an axial 2-OMe

group (ent-8) and further augmented by the addition of an axial

4-OMe group (ent-9).

Comparison of compounds ent-7 and ent-9, compounds

containing the C-18 and C-19 Me groups, reveals that the

major eﬀect of the axial 4-OMe substituent (ent-9) is to

increase the mean lifetime of the longest open state. Both

compounds increase the prevalence of long open events and

decrease the prevalence of long closed events.

In Figure 2C, a quantitative comparison of the potentiating

eﬀects of analogue ent-9 with those of the reference steroids 4−

6 on the same oocyte at 0.5 μM is shown. Analogue ent-9 had

higher activity than either the reference anesthetic steroid 6 or

Comparison of results for compounds ent-7 and ent-10

allows the eﬀect of the C-18 Me group to be determined. Both

176

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

Figure 2. Side-by-side, equimolar comparison of structurally related compounds in Xenopus oocytes expressing wild-type α₁β₂γ_LGABA_Areceptors.

The traces in the top panels of (A−D) represent current responses of oocytes to 0.5 μM GABA alone (smallest, black trace) and GABA plus the

color-indicated compound at 5 μM (A,D) or 0.5 μM (B,C). Bar graphs indicate normalized peak current relative to the response to GABA alone.

Table 4. Summary of Results on the Eﬀects of ent-9, ent-7,

ent-11, ent-10, and ent-12 on Single-Channel Currents from

Rat α₁β₂γ_2LGABA_AReceptors

Table 3. Modulation of Rat Wild-Type and Mutant α₁β₂γ_2L

GABA_AReceptor Function by Steroid Enantiomers ent-9 and

ent-7

a

compd

OT long (ms) fraction OT long fraction CT long

receptor

α₁β₂γ_2L

ent-9

ent-7

b

control (no

6.3 2.6

0.13 0.05

0.29 0.06

9.2 5.2**

7.8 3.9**

1.3 0.3*^,**

7.6 3.8*^{, #}

1.0 0.1^#,**

steroid)

α₁(Q241L)β₂γ_2L

α₁(S240L)β₂γ_2L

α₁(W245L)β₂γ_2L

1.3 0.1**^,**

9.4 2.9**^{, #}

0.9 0.1^#,**

ent-9

17.7 6.6***

12.4 3.4

9.8 3.0

0.53 0.03***

0.57 0.05***

0.20 0.08

0.06 0.03***

0.05 0.01***

0.11 0.05***

0.21 0.04

ent-7

ent-11

ent-10

ent-12

a

The data give the mean ( SD) eﬀect of 20 μM ent-9 or ent-7 on

16.1 3.9*

11.5 3.3

0.53 0.06***

0.49 0.04***

peak currents elicited by GABA at a concentration (5 μM for wild-

type, 10 μM for mutants) producing 11−17% of maximal current.

Number of cells was 4−7 at each receptor−compound combination.

Receptors were expressed in HEK 293 cells. Statistical analysis applies

to comparison with no eﬀect (using paired two-sample t test) and

comparison with potentiation observed at the wild-type receptor

(ANOVA with Bonferroni posthoc correction). (^#, not signiﬁcant; *, p

< 0.05; **, p < 0.01)

0.27 0.04

a

The rat wild-type α₁β₂γ_2LGABA_Areceptor was activated by 50 μM

GABA in the absence (control) or presence of 20 μM ent-9, ent-7, ent-

11, ent-10, or ent-12. Number of patches was 4−7. The intracluster

open time histograms were best-ﬁtted to the sums of three (control,

ent-11, ent-10, and ent-12) or two exponentials (ent-9 and ent-7). The

intracluster closed time histograms were ﬁtted to the sums of three

exponentials. The data give the mean duration and relative frequency

(fraction) of the longest-lived intracluster open time component and

the relative frequency (fraction) of the longest-lived intracluster

closed-time component. Previous work has shown that potentiation by

steroids is mediated by changes in these three parameters.^16,30

Statistical analysis was conducted using ANOVA with Bonferroni

posthoc correction. The signiﬁcance level applies to comparison to

control condition. *, P < 0.05; ***, P < 0.001. Nonsigniﬁcant changes

analogues increase the fraction of the longest open state. By

contrast, analogue ent-7, but not analogue ent-10 (the analogue

without the C-18 Me group), decreases the prevalence of the

longest closed state. Both compounds prolong the mean

duration of the longest state but the eﬀect reaches signiﬁcance

only in the presence of analogue ent-10.

are not marked. The control data are from our previous publication.⁴⁹

b

177

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

Figure 3. Exposure to compounds ent-9, ent-7, ent-11, ent-10, or ent-12 modulates single-channel currents from wild-type α1β2γ2 GABA_Areceptors.

Sample single-channel currents and the respective open and closed time histograms in the presence of 50 μM GABA (A), GABA + 20 μM ent-9 (B),

GABA + 20 μM ent-7 (C), GABA + 20 μM ent-11 (D), GABA + 20 mM ent-10 (E), or GABA + 20 μM ent-12 (F). Channel openings are shown as

downward deﬂections. In the presence of GABA alone, the open times were 0.27 ms (35%), 2.4 ms (53%), and 4.6 ms (11%), and the closed times

were 0.22 ms (53%), 1.4 ms (24%), and 8.9 ms (22%). In the presence of GABA + ent-9, the open times were 0.80 ms (52%) and 30.3 ms (48%),

and the closed times were 0.24 ms (67%), 1.4 ms (28%), and 14.3 ms (6%). In the presence of GABA + ent-7, the open times were 0.64 ms (39%)

and 10.9 ms (61%), and the closed times were 0.27 ms (71%), 2.1 ms (24%), and 25.0 ms (5%). In the presence of GABA + ent-11, the open times

were 0.30 ms (43%), 5.0 ms (24%), and 9.7 ms (33%), and the closed times were 0.43 ms (59%), 3.0 ms (24%), and 19.6 ms (17%). In the presence

of GABA + ent-10, the open times were 0.48 ms (33%), 6.6 ms (21%), and 13.2 ms (46%), and the closed times were 0.38 ms (52%), 2.8 ms (31%),

and 38.5 ms (16%). In the presence of GABA + ent-12, the open times were 0.29 ms (29%), 2.1 ms (23%), and 13.2 ms (48%), and the closed times

were 0.25 ms (54%), 2.0 ms (24%), and 27.8 ms (22%). The summary of averaged data from multiple patches is given in Table 4

Comparison of results for compounds ent-7 and ent-11

allows the eﬀect of the C-19 Me group on single-channel

properties to be determined. Both compounds decrease the

fraction of dwells in the longest closed state. However, analogue

ent-7 (the compound with the C-19 Me group), unlike

analogue ent-11 (the compound without the C-19 Me group),

increases the fraction of dwells in the longest open state.

Overall, the results suggest that single-channel properties are

aﬀected in diﬀerent ways by each Me group. The C-19 Me

group has its largest eﬀect on the prevalence of the longest

open event, and the C-18 Me group has its largest eﬀect on the

prevalence of the long-lived closed state.

When neither Me group is present (ent-12), the sole kinetic

eﬀect of the compound is an increase in the fraction of the

longest open state. This suggests that the increase in this kinetic

parameter is not controlled solely by the presence of a C-19 Me

substituent. In fact, previous studies have found that even

steroid analogues with other ring systems such as cyclopenta-

[b]phenanthrenes and cyclopenta[b]anthracenes can increase

the prevalence of long openings,²⁹thus indicating that this

single-channel parameter is sensitive to a wide range of

structural variations.

group from position C-17 to C-16 has no signiﬁcant eﬀect on

tadpole LRR but lowers LSR because only steroid 7 has an

EC₅₀value below 10 μM. This same structural modiﬁcation for

the compound pair ent-1 and ent-7 led to the latter having

lower EC₅₀values for both LLR and LSR (∼3-fold in both

cases), results consistent with the same trend found in binding

data reported in Table 1.

For the 17-ketosteroids, the axial 2-OMe group had a minor

negative eﬀect on potency for either LRR or LSR (compare

steroids 1 and 2). This substituent was without eﬀect on LRR

or LSR for steroids with a 16-ketone group (compare steroids 7

and 8). All four of these steroids have very similar LRR and

LSR activity proﬁles. For the ent-17-ketosteroid pair ent-1 and

ent-2, the latter compound has a higher LRR EC₅₀value (∼2-

fold) and the LSR EC₅₀value is raised above 10 μM. For the

enantiomer pair ent-7 and ent-8, the eﬀect of the axial 2-OMe

group is very large. This substituent raises both the LRR and

LSR EC₅₀values by ∼5-fold. No inconsistencies were found for

the axial 2-OMe substituent eﬀects between the binding (Table

1) and tadpole reﬂex results.

The steroids 3 (17-ketone) and 9 (16-ketone) with the axial

4-OMe substituent have low activity as anesthetics in tadpoles.

On the basis of the high IC₅₀values found in the [³⁵S]TBPS

displacement experiments (Table 1), the result that steroid 9

had any eﬀect on either LRR or LSR at concentrations below

30 μM is somewhat surprising. Yet again, the axial 4-OMe

substituent in the enantiomeric compounds ent-3 and ent-9

produced analogues with high activity. The EC₅₀values for

both LRR and LSR are the lowest values observed and correlate

TADPOLE LOSS OF RIGHTING REFLEX (LRR) AND

LOSS OF SWIMMING (LSR) RESULTS

■

The anesthetic eﬀects of the compounds in tadpoles are

summarized in Table 5. Comparison of EC₅₀values for the

steroid pair 1 and 7 shows that moving the D-ring 17-ketone

178

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

Table 5. Eﬀects of Steroids 1−9 and Steroid Enantiomers

ent-1−ent-3 and ent-7−ent-12 on Tadpole Righting and

a

Swimming Reﬂexes

tadpole LRR

EC₅₀(μM)

tadpole LRR

tadpole LSR

EC₅₀(μM)

tadpole LSR

b

compd

n_Hill

c

d

1

3.38 0.90

1.42 0.18

5.09 2.43

2.39 0.28

none

−2.83 2.66

−2.17 0.48

−1.29 1.22

−2.10 0.42

none

c

ent-1

5.48 0.12

>10

−33

0

2

ent-2

3

>10

none

Figure 4. Correlations between rank order LRR anesthetic eﬀects in

tadpoles and TBPS displacement potency. The highest potency

compound was assigned the rank of 1, and the least potent compound

was assigned the rank of 18. For the least potent anesthetic

compounds 3 and ent-12, their weak activity did not allow

determination of the LLR EC₅₀value. These compounds were

arbitrarily assigned the two highest LRR rankings of 18 and 17,

respectively. Red symbols represent enantiomers (“e”). Black symbols

represent natural compounds. Pearson r value of a linear regression

(solid line) through all points was 0.81 (p < 0.05).

ent-3

0.59 0.07

1.44 0.20

0.42 0.04

1.12 0.14

3.60 1.52

0.55 0.09

3.15 0.30

2.91 0.36

3.15 0.25

0.34 0.04

3.64 2.17

1.16 0.21

none

−1.84 0.32

−2.84 0.77

−1.83 0.32

−3.38 2.28

−2.74 3.25

−2.90 0.07

−2.55 0.78

−2.37 0.80

−2.60 0.70

−2.75 1.13

−4.32 11.8

−4.03 4.06

2.94

0

−20

−33

−7.5 1.1

−34

−30

−37

−15

−21

−74

−36

1

0

e

4

5.48 0.20

5.5 0.5

f

5

g

6

5.48 0.11

0

7

7.54

0

ent-7

8

1.73 0.04

9.29 0.03

1

0

ent-8

9

10.3

13.5

0

ent-9

ent-10

ent-11

ent-12

1.73 0.03

none

ANESTHESIA IN MICE RESULTS

■

5.48 0.12

none

−33

0

The potency, rate of onset, and rate of recovery for compounds

ent-7 and ent-9 relative to these parameters for anesthetic

steroid 6 were made using tail vein injections in mice. The

duration of anesthesia, deﬁned as loss of righting reﬂex,

observed for the compounds is shown in Figure 5. We

The methods are as reported previously.⁴⁶Error limits are calculated

as the standard error of the mean (N = 10 or more animals at each of

ﬁve or more diﬀerent concentrations). LSR typically has a very steep

a

b

concentration−response curve. The reported n_Hillvalues vary widely

depending on how many animals had LSR at the concentration

(typically, 3 μM) preceding the concentration (typically, 10 μM) that

produced LSR in all animals. No SAR conclusions were based on these

c

d

slope values. Literature values.⁹“None” indicates that all animals had

a swimming response at 10 μM test compound. Literature values.¹⁵

e

Literature values.⁴⁶Literature values.⁴⁸

f

g

with the low IC₅₀values for [³⁵S]TBPS displacement (Table 1).

Both EC₅₀values are lower than those found for anesthetic

steroid 6 and are comparable to those found for the potent

endogenous neurosteroid 5. Thus, eﬀects of the axial 2-OMe or

4-OMe groups in the LRR and LSR bioassay generally maintain

the pattern wherein these substituents have opposite eﬀects in

the steroid and ent-steroid series.

The contribution that the C-18 and C-19 Me groups have on

potency for LRR and LSR of compound ent-7 was determined

by comparing the EC₅₀values of this compound with those of

compounds ent-10−ent-12. Compound ent-10, which lacks the

C-18 Me group, has a higher LRR EC₅₀value than compound

ent-7 by about 7-fold and does not cause LSR at concentrations

up to 10 μM. Compound ent-11, which lacks the C-19 Me

group, is about 2-fold less potent at causing LRR and 3-fold less

potent at producing LSR than compound ent-7. Compound

ent-12, which lacks both the C-18 and C-19 Me groups,

produces neither LRR nor LSR at concentrations up to 10 μM.

Overall, these results indicate that in this series of ent-steroids

the C-18 Me group has a major eﬀect on LRR and LSR whereas

the C-19 Me group has a minor eﬀect.

Figure 5. Duration of anesthesia (LRR) induced by tail vein injection

of alphaxalone (6), and ent-7, and ent-9. The compounds were

dissolved in 22.5% aqueous 2-(hydroxypropyl)-β-cyclodextrin. For

compound ent-9, serial dilutions of the 8 mg/kg stock solution with

0.9% saline were made to obtain the lower doses tested.

previously found that the threshold dose for mice LRR in mice

by this route of administration was 8 mg/kg for steroid 6, which

caused LRR of less than 1 min duration.¹⁵A 16 mg/kg dose of

steroid 6 caused LRR of about 4 min duration. At a dose of 8

mg/kg, compound ent-7 caused LRR of about 3 min duration

and at a dose of 16 mg/kg duration of LRR was increased to

about 7 min. The threshold dose for compound ent-7 was not

determined. For compound ent-9, no LRR was observed at a

dose of 1 mg/kg and LRR of about 1 min was observed at a

threshold dose of 2 mg/kg. At 4 mg/kg, LRR induced by

compound ent-9 lasted about 5 min and at a dose of 8 mg/kg

LRR lasted for about 9 min. For all three compounds, the onset

of anesthesia was immediate (<30 s) and recovery was

characterized by a rapid progression over 1−2 min from an

Figure 4 shows a rank order correlation for the [³⁵S]TBPS

IC₅₀and tadpole LRR EC₅₀values for all 18 compounds. The

Pearson correlation coeﬃcient is 0.81, indicative of a strong

correlation between these parameters. The greatest outliers on

the plot are compounds ent-2, ent-8, and 9, all of which were

more potent at causing LRR than expected from their

[³⁵S]TBPS IC₅₀values.

179

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

initial return of leg movement followed by righting and

subsequent walking around the cage.

[³⁵S]TBPS displacement potency and single channel ki-

netics.^16,17Hence, we expected similar eﬀects for the C-18

and C-19 Me group substitution pattern in active ent-steroids.

The eﬀect of this Me group substitution pattern was evaluated

using compounds ent-7 and ent-10−ent-12.

DISCUSSION

■

This is the second SAR study to follow up on our initial

observation that ent-androsterone and ent-etiocholanolone are

modulators of GABA_Areceptors.^9,10In the ﬁrst follow up study,

we presented a model for how the pair of androsterone

enantiomers were aligned when bound to the site that causes

steroid potentiation of GABA_Areceptors.¹³In this study, we

further validate and reﬁne that model (Figure 1) and also show

that it is useful for identifying structural features that could lead

to development of novel clinically useful intravenous general

anesthetics. We identify two candidate ent-steroids that have

this potential.

Compound ent-11 lacks the C-19 Me group. Relative to

compound ent-7, which has both the C-18 and C-19 Me

groups, ent-11 has diminished activity in all three screening

bioassays. Removing the C-18 Me group, as in compound ent-

10, causes a larger loss of activity relative to compound ent-7

than does removal of the C-19 Me group (ent-11) in all three

screening bioassays. Compound ent-12, which lacks both the C-

18 and C-19 Me groups, is not an eﬀective potentiator in the

electrophysiology screen and is inactive in the tadpole screens

at or below 10 μM. This dinor ent-steroid is a very weak

displacer of [³⁵S]TBPS. Overall, the activities of compounds

ent-7 and ent-11, the compounds with the C-18 Me group, are

similar to each other, and compounds ent-10 and ent-12, the

compounds without the C-18 Me group, are similar to each

other. Consistent with expectations, those with the C-18 Me

group have the higher activity. We attribute the eﬀect of the C-

19 Me group to be explained in terms of its eﬀect on log P.

Removing this substituent, lowers log P (ent-7, log P 3.71; ent-

11, log P 3.24), thereby decreasing a compound’s accumulation

in the membrane where the steroid potentiation sites on the

GABA_Areceptor are located.^26,28As a consequence of its lower

eﬀective concentration in the membrane bilayer, compound

ent-11 has modestly decreased activity. With regard to the large

loss of activity found when the C-18 Me group is not present,

we propose that this is due to removal of a structural feature

that is required to meet pharmacophore requirements for high

activity. The calculated log P values for compounds ent-10 and

ent-11 are the same (3.24), indicating that log P alone does not

adequately explain why compounds without a C-18 Me group

have such low activity. The very low activity of compound ent-

12 (10 μM), which lacks both Me groups, is likely due to the

combination of the log P (2.77) and pharmacophore

requirement eﬀects. The correlation of log P with biological

activity is further addressed later in the Discussion.

We carried out electrophysiologcal experiments with three

diﬀerent mutated forms of GABA_Areceptors to verify that

potentiation eﬀects of ent-16-ketosteroids (ent-7, ent-9) were

aﬀected by these mutations in the same way as those of

endogeneous neurosteroid 5. The eﬀects of all three mutations

on the actions of the two ent-16-ketosteroids and steroid 5 were

the same. We infer that the actions of the ent-steroids examined

in this study are mediated by the previously identiﬁed steroid

modulation site.²⁶

We conducted single-channel measurements on the eﬀects of

analogues ent-7 and ent-9−ent-12 for two reasons. The ﬁrst

goal was to determine the kinetic components through which

these ent-steroids potentiate rat α₁β₂γ_2Lreceptors for

comparison with changes observed in the presence of steroid

5. The second goal was to determine whether changes in ent-

steroid structure correlate with the speciﬁc kinetic eﬀects.

Previous studies employing single-channel kinetic analysis have

revealed, for example, that the presence of the C-18 Me group

in a steroid as well as the conﬁguration at C-5 (α vs β) can

determine the ability of a steroid to aﬀect the mean duration of

long channel openings.¹⁶Data presented in this study

signiﬁcantly extend prior knowledge and satisfy both of these

goals.

Three predictions were made based on our original model

and prior SAR studies of steroids and ent-steroids that

potentiate GABA_Areceptors. The ﬁrst concerned the eﬀect of

an axial 4α-OMe group on the activity of compound ent-1. It

was shown about 40 years ago in the published SAR studies

that led to the clinical use of anesthetic steroid 6 that axial 2β-

alkoxy groups were compatible with high anesthetic activity and

it was stated that various axial 4β-groups were not.¹⁴We

conﬁrmed this SAR using steroids 2 (axial 2β-OMe substituent)

and 3 (axial 4β-OMe substituent). Our model predicted that as

a consequence of the inverted binding of ent-steroids relative to

steroids at the steroid potentiation site, an ent-steroid with an

axial 2α-OMe group would yield a compound with low activity

(ent-2), whereas an ent-steroid with an axial 4α-OMe group

would yield a compound (ent-3) with about the same

anesthetic activity as steroid 2. This was found to be correct

and consistently observed in all three of our screening bioassays

([³⁵S]-TBPS displacement, oocyte electrophysiology, and

tadpole LRR and LSR).

The second prediction was that for active ent-steroids, a more

favorable position for the D-ring hydrogen bond acceptor

ketone group found at position C-17 in steroid potentiators,

would be at position C-16 in ent-steroid potentiators. It also

implies that moving the ketone group to the C-16 position in

steroids would decrease the activity of steroid potentiators. The

prediction for the ent-steroids was evaluated by comparing the

actions of compounds ent-1 and ent-3 with those of ent-7 and

ent-9, respectively. In both cases, this structural change

produced compounds with much higher activity in all three

bioassays. The prediction for the steroids was evaluated with

the steroid pairs 1 and 7 as well as with steroid pairs 2 and 8.

The lower activity expected for steroids 7 and 8 was observed

in the [³⁵S]TBPS displacement and oocyte electrophysiology

screens which measure actions at GABA_Areceptors. However,

these two compounds were modestly more potent in the

tadpole LRR and LSR screen. This result was unexpected but

could be explained by steroid 7 and 8 modulation of other

receptors that contribute to LRR and LSR in tadpoles.

The third prediction was that the activities of the ent-steroids

would be more greatly aﬀected by removal of the C-18 Me

group than they would be by removal of the C-19 Me group. In

the model shown in Figure 1, the C-18 and C-19 Me groups of

both steroids and ent-steroids are in nearly the same place in

three-dimensional space. This implies that the eﬀects observed

for C-18 and C-19 Me groups in steroids and ent-steroids

should be the same. In steroids, their potentiation eﬀects are

little altered when the C-19 Me group is absent.²⁴By contrast,

removal of the C-18 Me group can have a major impact on

180

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

We found that compound ent-9 had the identical proﬁle of

single-channel kinetic eﬀects as the previously studied

endogenous neurosteroid 5.²⁸Removing the axial 4α-OMe

group (ent-7) removed the eﬀect of increasing the mean

lifetime of long open events. We then examined the eﬀect of C-

18, C-19 Me substituents on the proﬁle of kinetic eﬀects

observed with compound ent-7. We found that the C-19 Me

group’s eﬀect correlated with open time properties. By contrast,

the C-18 Me group’s eﬀect correlated with decreased

prevalence of the long closed state. The C-18 Me group

seems to be a speciﬁc structural eﬀect, whereas the C-19 Me

group does not because some cyclopenta[b]phenanthacene and

cyclopenta[b]anthracenes analogues without a Me group placed

in a position equivalent to that occupied by the steroid C-19

Me group, but containing the equivalent of a steroid C-18 Me

group, produce similar eﬀects on long open state events.²⁹For

this reason, we postulate that the C-19 Me group in both

steroids and ent-steroids having a trans A,B-ring fusion increases

analogue potentiation predominately by an eﬀect on log P.

Comparison of single channel data (Table 4) with whole-cell

recordings (Table 2) indicates that the kinetic component that

contributes most to enhancement of the whole-cell peak

response is the reduction in the fraction of long closed events.

We note that this is in agreement with our previous study

examining GABA_Areceptor potentiation by steroid 5.^28,30

The correlation of log P with the biological activity of

neurosteroids merits further discussion. The introduction of

polar substituents at the C-11 and/or the C-21 position of

neurosteroid 5 yields analogues with reduced activity at GABA_A

receptors. The decreased activity of the more polar analogues

strongly correlates with log P.^31,32However, the correlation

does not rule out the possibility that the polar substituents yield

analogues which do not satisfy the pharmacophore require-

ments for maximal activity and that the correlation with log P is

fortuitous. More generally, it implies that making structural

changes that aﬀect log P and pharmacophore requirements are

diﬃcult to interpret when both parameters change in the same

direction. There are two ways to circumvent this commonly

encountered interpretation diﬃculty in SAR studies. One is to

use enantiomers which have identical log P values. Thus,

changes in pharmacological activity (in bioassays where

pharmacokinetic and metabolism issues are not confounding

factors) can be unambiguously correlated with ﬁt to the

pharmacophore requirements. The second way is to identify

analogues wherein the log P and pharmacological activity move

in opposite directions. Both approaches are exempliﬁed in this

study.

Figure 6. Correlation between compound log P and log TBPS IC₅₀

values for compounds with reported IC₅₀values below 500 nM. Red

symbols represent enantiomers (“e”). Black symbols represent natural

compounds. There was no correlation between log P and log TBPS

IC₅₀(or between log P and TBPS IC₅₀, Supporting Information Figure

S3).

conclusion. Rather, it only indicates that receptor−ligand

interactions should not be viewed as a rigid lock and key ﬁt

but as an induced ﬁt between both molecules as is now widely

appreciated in SAR studies.

Finally, we examined the in vivo anesthetic eﬀects of

compounds ent-7 and ent-9 in mice (Figure 5). At equivalent

doses, both compounds induced LRR for a longer period of

time than did anesthetic steroid 6. Compound ent-9 was

particularly potent at causing LRR. It induced about twice the

length of LRR at a dose of 8 mg/kg as did anesthetic steroid 6

at the higher dose of 16 mg/kg. Our expectation is that the

lower log P value of compound ent-9 will increase its water

solubility and that ent-steroids, in general, may provide a

structural framework for the development of anesthetic steroid

analogues with improved water solubility.

Additional eﬀorts will be required to determine the full

potential of ent-steroids as clinically useful drugs. As noted

above, compound ent-9 has the potential to be a clinically

useful anesthetic. Other ent-steroids could be potentially useful

for treating conditions in which increasing neuronal inhibition

by potentiation of GABA_Areceptor function has proven useful

(e.g., seizures, anxiety, depression).^1,33Because enzymes and

nuclear hormone receptors that bind endogeneous steroids

have structurally deﬁned chiral binding sites, there is the

potential that ent-steroids will not be bound with high aﬃnity

to these proteins. If so, then ent-steroid drugs might not

strongly interfere with endogenous steroid hormone biosyn-

thesis or be agonists for steroid hormone receptors. Several

examples of the failure of ent-steroid hormones to act as

agonists at nuclear receptors are already reported in the

literature.³⁴⁻³⁶Results from those studies led to the suggestion

that ent-steroid drugs have the potential to treat brain disorders

or traumatic brain injury without acting as hormones that

stimulate the development and growth of hormone-dependent

cancers.^36,37

Overall, we failed to ﬁnd a correlation between [³⁵S]TBPS

IC₅₀values and log P for those compounds which strongly

displaced [³⁵S]TBPS (IC₅₀< 500 nM) as shown in Figure 6. As

a speciﬁc example, the endogeneous neurosteroid 5 and

analogue ent-9 have very diﬀerent log P values (5, log P 4.46;

ent-9, log P 3.19) but essentially the same IC₅₀values. In total,

ﬁve enantiomeric pairs of steroids were examined in this study

and in all cases the pairs of enantiomers had diﬀerent activities.

Because our mutagenesis data support the conclusion that the

pairs of enantiomers act at the same steroid potentiation site on

the GABA_Areceptor, this most likely reﬂects the ability of one

enantiomer in each pair, independent of log P, to meet

pharmacophore requirements better than the other enantiomer.

It remains possible that the neurosteroid potentiation site has a

diﬀerent induced ﬁt when bound to steroids than it does when

bound to ent-steroids, but this does not invalidate our

It is also possible that the half-lives of ent-steroid drugs may

be quite diﬀerent, and potentially longer, than those of steroid

drugs used as anticonvulsants, anxiolytics, and other neuro-

active agents. In this regard, liver metabolism of ent-steroids has

received almost no attention thus far. One reported study of the

glucuronidation of three ent-steroids showed that the

enantiomers are preferred substrates for diﬀerent glucuronosyl-

transferases than those that accept the naturally occurring

enantiomers as substrates.³⁸Other diﬀerences in liver

metabolism are also likely to occur. Thus, it is likely that any

eﬀort to develop ent-steroids as drugs will require a great deal of

181

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

compound ent-17 using the procedure described for the preparation

of compound ent-2. Flash column chromatography (silica gel eluted

eﬀort with the attendant generation of a lot of new knowledge

as so little is known about the in vivo actions and metabolism of

ent-steroids in humans.

23

with 20% EtOAc in hexanes) gave product 2; mp 151−153 °C; [α]_D

+92.4 (c 0.37, CHCl₃). IR ν_max3439, 1738 cm⁻¹. ¹H NMR (300 MHz,

CDCl₃) δ 3.84−3.83 (m, 1H), 3.22 (br s, 4H), 2.54 (br s, 1H), 2.38

(dd, 1H, J = 19.2 Hz, 8.8 Hz), 0.86 (s, 3H), 0.76 (s, 3H), 0.71−0.64

(m, 1H). ¹³C NMR (CDCl₃) δ 221.4, 80.5, 67.7, 56.4, 54.9, 51.3, 47.6,

38.7, 35.9, 35.6, 35.2, 34.3, 31.8, 31.4, 30.6, 27.6, 21.5, 19.9, 13.6, 12.9.

Anal. (C₂₀H₃₂O₃) C, H.

CONCLUSION

■

We conclude that a previously proposed model showing the

alignment of androsterone enantiomers at the steroid

potentiation site on GABA_Areceptors has predictive value.

The model accurately predicted: (1) that the potentiating

eﬀects of compound ent-1 would be increased in an analogue

with the 16-ketone group (ent-7), (2) that high activity would

be maintained by adding by adding an axial 4α-OMe

substituent (ent-9), but not an axial 2α-OMe group (ent-8),

and (3) that the C-18 Me group in ent-16-ketosteroids is more

important for enhancement of potentiation than the C-19 Me

group. We compared the length of loss of righting reﬂex in mice

induced by tail vein injection of anesthetic steroid 6 and

analogues ent-7 and ent-9. We found that both ent-steroids

induced a longer period of loss of righting reﬂex at a lower dose

than did steroid 6. We conclude that the high activity of these

two compounds results from their ability to satisfy

pharmacophore requirements, not because of an increase in

log P, because compound ent-9, has a lower log P than

anesthetic steroid 6 and both of these ent-neurosteroids have a

lower log P than endogeneous neurosteroid 5. We propose that

ent-steroids have high potential for development as both

intravenous anesthetics and drugs to treat brain disorders that

are alleviated by potentiation of GABA_Areceptor function.

(3β,4α,5β,8α,9β,10α,13α,14β)-3-Hydroxy-4-methoxyandrostan-

17-one (ent-3). Compound ent-3 (200 mg, 78%) was prepared from

compound ent-20 using the procedure described for the preparation of

ent-2. Flash column chromatography (silica gel eluted with 35%

EtOAc in hexanes) gave product ent-3 as a white solid; mp 215−218

°C; [α]_D²³−99 (c 0.06, CHCl₃). IR ν_max3510, 2917, 2838, 1735, 1594,

1

1443, 1375, 1242 cm⁻¹. H NMR (400 MHz, CDCl₃) δ 4.02 (br s,

1H), 3.35 (s, 3H), 3.04 (s, 1H), 2.44 (dd, 1H, J = 19.0 Hz, 9.0 Hz),

0.98 (s, 3H), 0.86 (s, 3H), 0.62 (m, 1H). ¹³C NMR (CDCl₃) δ 221.5,

85.4, 66.1, 59.0, 55.2, 51.5, 47.8, 44.0, 36.2, 35.8, 35.0, 31.8, 31.5, 31.1,

25.2, 25.0, 21.7, 19.6, 14.0, 13.8. Anal. (C₂₀H₃₂O₃) C, H.

(3α,4β,5α)-3-Hydroxy-4-methoxyandrostan-17-one (3). Steroid 3

(270 mg, 81%) was prepared from the natural enantiomer of

compound ent-20 using the procedure described for the preparation

of compound ent-2. Flash column chromatography (silica gel eluted

with 20% EtOAc in hexanes) gave product ent-3 as a white solid; mp

20

223−225 °C; [α]_D+95 (c 0.06, CHCl₃). IR ν_max3513, 1734 cm⁻¹.

¹H NMR (300 MHz, CDCl₃) δ 3.97 (br s, 1H), 3.30 (s, 3H), 2.95−

2.85 (m, 1H), 2.44 (dd, 1H, J = 19.3 Hz, 8.8 Hz), 0.93 (s, 3H), 0.81 (s,

3H). ¹³C NMR (CDCl₃) δ 221.7, 85.7, 66.3, 59.1, 55.4, 51.7, 47.9,

44.2, 36.3, 36.0, 35.2, 32.0, 31.7, 31.3, 25.4, 25.1, 21.9, 19.8, 14.1, 14.0.

Anal. (C₂₀H₃₂O₃) C, H.

[3α,5α,17(20)Z]-3-Hydroxy-11-oxopregn-17(20)-ene-21-nitrile

(4). The compound was prepared as previously described.¹⁵

(3β,5β,8α,9β,10α,13α,14β)-3-Hydroxyandrostan-16-one (ent-7).

Compound ent-24 (88 mg, 0.27 mmol) was dissolved in methanol

(10 mL), and water (0.5 mL) and K₂CO₃(5 mmol, 680 mg) were

added. The reaction was reﬂuxed for 2 h and cooled to room

temperature, water (30 mL) was added, and the product was extracted

into EtOAc (50 mL × 3). The combined extracts were dried, ﬁltered,

and the solvent removed. The residue was puriﬁed by ﬂash column

chromatography (silica gel eluted with 25−40% EtOAc in hexanes) to

give product ent-7 (72 mg, 93%) as a white solid; mp 152−153 °C;

EXPERIMENTAL SECTION

■

General Methods. Solvents were either used as purchased or dried

and puriﬁed by standard methodology. Extraction solvents were dried

with anhydrous Na₂SO₄and, after ﬁltration, removed on a rotary

evaporator. Flash column chromatography was performed using silica

gel (32−63 μm) purchased from Scientiﬁc Adsorbents (Atlanta, GA).

Melting points were determined on a Koﬂer micro hot stage and are

uncorrected. FT-IR spectra were recorded as ﬁlms on a NaCl plate.

NMR spectra were recorded in CDCl₃at ambient temperature at 300

or 400 MHz (¹H) or 74 or 100 MHz (¹³C). Purity of >95% was

determined for all evaluated compounds by combustion analysis for

C,H performed by M-H-W Laboratories (Phoenix, AZ). Steroids 1, 5,

6, 13, 14, and 18 were purchased from Steraloids (Newport, RI).

K(sec-Bu)₃BH (K-Selectride) was purchased from Aldrich Chemical

Co. (Milwaukee, WI). The log P values were calculated using

ChemDraw from CambridgeSoft (Cambridge, MA).

23

1

[α]_D+156.8 (c 0.25, CHCl₃). IR ν_max3435, 1737 cm⁻¹. H NMR

(300 MHz, CDCl₃) δ 4.06 (m, 1H), 0.87 (s, 3H), 0.81 (s, 3H). ¹³C

NMR (CDCl₃) δ 219.0, 66.4, 55.9, 54.2, 51.7, 39.3, 39.2, 39.0, 38.2,

36.3, 35.8, 34.9, 32.2, 31.9, 28.9, 28.3, 20.3, 18.1, 11.2. Anal.

(C₁₉H₃₀O₂) C, H.

(3α,5α)-3-Hydroxyandrostan-16-one (7). Steroid 7 (50 mg, 90%)

was prepared from the natural enantiomer of compound ent-24 using

the procedure described for the preparation of compound ent-7.

Product 7 was obtained as a white solid; mp 152−154 °C (lit mp³⁹

(5β,8α,9β,10α,13α,14β)-Androstane-3,17-dione (ent-1). The com-

pound was prepared as previously described.⁹

23

153−154 °C); [α]_D−154.0 (c 0.20, CHCl₃). IR ν_max3441, 1738

(2α,3β,5β,8α,9β,10α,13α,14β)-3-Hydroxy-2-methoxyandrostan-

17-one (ent-2). The 2β,3β-epoxide ent-17 (250 mg, 0.86 mmol),

containing minor amounts of the 2α,3α-epoxide, was dissolved in

MeOH (10 mL), a drop of conc H₂SO₄was added, and the reaction

was stirred at room temperature for 3 h. The reaction mixture was

made basic with aqueous NaHCO₃, and the MeOH was removed

under reduced pressure. The resulting residue was diluted with water

and extracted with EtOAc (3 × 80 mL). The combined organic

extracts were dried and concentrated to give a solid which was puriﬁed

by column chromatography (silica gel eluted with 20−35% EtOAc in

hexanes) to give ent-2 as a white solid (170 mg, 62%); mp 153−155

1

cm⁻¹. H NMR (CDCl₃) δ 4.05 (m, 1H), 0.87 (s, 3H), 0.81 (s, 3H).

¹³C NMR (CDCl₃) δ 219.0, 66.2, 55.8, 54.2, 51.6, 39.2, 39.1, 38.9,

38.1, 36.2, 35.7, 34.8, 32.2, 31.8, 28.8, 28.2, 20.2, 18.0, 11.2. Anal.

(C₁₉H₃₀O₂) C, H.

(2α,3β,5β,8α,9β,10α,13α,14β)-3-Hydroxy-2-methoxyandrostan-

16-one (ent-8). Compound ent-8 (120 mg, 97%) was prepared from

compound ent-27 using the procedure described for the preparation of

compound ent-7. Flash column chromatography (silica gel eluted with

25−40% EtOAc in hexanes) gave product ent-8 as a white solid; mp

209−211 °C; [α]_D²³+144 (c 0.25, CHCl₃). IR ν_max3447, 1729 cm⁻¹.

¹H NMR (300 MHz, CDCl₃) δ 3.95 (br s, 1H), 3.32 (s, 4H), 0.95 (s,

23

°C; [α]_D−98.3 (c 0.23, CHCl₃). IR ν_max3438, 2920, 1739, 1595,

3H), 0.85 (s, 3H). ¹³C NMR δ 218.9, 80.6, 68.0, 56.6, 55.8, 54.9, 51.7,

39.2, 39.1, 38.8, 38.2, 36.1, 35.3, 34.3, 32.2, 32.0, 27.8, 20.4, 18.1, 13.0.

Anal. (C₂₀H₃₂O₃) C, H.

1

1453, 1406, 1372, 1255, 1214 cm⁻¹. H NMR (400 MHz, CDCl₃) δ

3.95 (br s, 1H), 3.32 (br s, 4H), 2.45 (dd, 1H, J = 19.2 Hz, 8.7 Hz),

1.00 (m, 1H), 0.95 (s, 3H), 0.85 (s, 3H), 0.77 (m, 1H). ¹³C NMR δ

221.5, 80.6, 68.1, 56.6, 55.1, 51.4, 47.8, 39.0, 36.1, 35.8, 35.5, 34.5,

32.0, 31.5, 30.8, 27.8, 21.7, 20.1, 13.8, 13.1. Anal. (C₂₀H₃₂O₂) C, H.

(2β,3α,5α)-3-Hydroxy-2-methoxyandrostan-17-one (2). Steroid 2

(206 mg, 73%) was prepared from the natural enantiomer of

(2β,3α,5α)-3-Hydroxy-2-methoxyandrostan-16-one (8). Steroid 8

(68 mg, 94%) was prepared from the natural enantiomer of ent-27

using the procedure described for the preparation of compound ent-7.

23

Product 8 was obtained as a white solid; mp 205−207 °C; [α]_D

182

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

1

−143.5 (c 0.20, CHCl₃). IR ν_max3445, 1730 cm⁻¹. H NMR (400

MHz, CDCl₃) δ 3.92 (br s, 1H), 3.31 (s, 4H), 0.94 (s, 3H), 0.84 (s,

3H). ¹³C NMR (CDCl₃) δ 218.9, 80.6, 68.0, 56.6, 55.8, 54.9, 51.7,

39.2, 39.1, 38.8, 38.2, 36.1, 35.3, 34.3, 32.2, 32.0, 27.8, 20.4, 18.1, 13.0.

Anal. (C₂₀H₃₂O₃) C, H.

hydride reagent, and the mixture was stirred for a few minutes. Then 1

M HCl was added and the reaction was warmed to room temperature.

The product was extracted into EtOAc, and the combined organic

extracts were washed with brine, dried, and the solvent removed to

give an oil, which was puriﬁed by ﬂash column chromatography (silica

gel eluted with 30−40% EtOAc in hexanes) to give product ent-14 as a

(3β,4α,5β,8α,9β,10α,13α,14β)-3-Hydroxy-4-methoxyandrostan-

16-one (ent-9). Compound ent-9 (50 mg, 94%) was prepared from

compound ent-30 using the procedure described for the preparation of

compound ent-7. Flash column chromatography (silica gel eluted with

25−40% EtOAc in hexanes) gave product ent-9 as a white solid; mp

23

colorless solid (1.49 g, 82%); mp 171−173 °C; [α]_D−95.4 (c 0.13,

1

CHCl₃). IR ν_max3470, 2928, 2853, 1732, 1450, 1374 cm⁻¹. H NMR

(CDCl₃) δ 3.59 (m, 1H), 2.41 (dd, 1H, J = 19.2 Hz, 9.0 Hz), 2.05 (dd,

1H, J = 19.2 Hz, 9.0 Hz), 0.85 (s, 3H), 0.83 (s, 3H), 0.69 (m, 1H). ¹³C

NMR (CDCl₃) δ 221.4, 71.1, 54.4, 51.4, 47.8, 44.8, 38.0, 36.9, 35.8,

35.6, 35.0, 31.5, 31.4, 30.9, 28.4, 21.7, 20.5, 13.8, 12.3. Anal.

(C₂₀H₃₂O₃) C, H.

23

198−200 °C; [α]_D+152 (c 0.2, CHCl₃). IR ν_max3458, 1725 cm⁻¹.

¹H NMR (CDCl₃) δ 4.01 (br s, 1H), 3.34 (s, 3H), 3.02 (d, 1H, J = 1.1

Hz), 0.98 (s, 3H), 0.87 (s, 3H). ¹³C NMR δ 219.0, 85.4, 66.2, 59.0,

55.9, 55.1, 51.8, 43.9, 39.3, 39.3, 38.1, 36.2, 34.9, 32.6, 31.6, 25.2, 25.0,

19.9, 18.1, 14.0. Anal. (C₂₀H₃₂O₃) C, H.

(3α,5β,8α,9β,10α,13α,14β)-3-[(Methylsulfonyl)oxy]-androstan-

17-one (ent-15). MeSO₂Cl (0.58 mL, 7.5 mmol) was added to a cold

solution (0 °C) of compound ent-14 (1.45 g, 5 mmol) and Et₃N (1.4

mL, 10 mmol) in CH₂Cl₂(15 mL), and the mixture was slowly

warmed to room temperature (∼15 min) and stirred for 30 min. An

aqueous saturated NaHCO₃solution was added, and the product was

extracted into CH₂Cl₂. The organic extracts were washed with brine,

dried, and concentrated to give an oil, which was puriﬁed by ﬂash

column chromatography (silica gel eluted with 15−30% EtOAc in

hexanes) to give mesylate ent-15 as a white solid (1.84 g, 100%); mp

(3α,4β,5α)-3-Hydroxy-4-methoxyandrostan-16-one (9). Steroid 9

(62 mg, 90%) was prepared from the natural enantiomer of compound

ent-30 using the procedure described for the preparation of compound

23

ent-7. Steroid 9 was obtained as a white solid; mp 196−198 °C; [α]_D

1

−141.8 (c 0.17, CHCl₃). IR ν_max3445, 1725 cm⁻¹. H NMR (400

MHz, CDCl₃) δ 3.99 (br s, 1H), 3.33 (s, 3H), 3.00 (m, 1H), 0.97 (s,

3H), 0.85 (s, 3H). ¹³C NMR (CDCl₃) δ 219.0, 85.4, 66.2, 59.0, 55.9,

55.1, 51.8, 43.9, 39.3, 39.1, 38.1, 36.2, 34.9, 32.5, 31.6, 25.2, 25.0, 19.9,

18.1, 14.0. Anal. (C₂₀H₃₂O₃) C, H.

1

150−152 °C (dec). IR ν_max2919, 1737, 1452, 1352 cm⁻¹. H NMR

(CDCl₃) δ 4.43 (m, 1H), 2.86 (s, 3H), 2.27 (dd, 1H, J = 19.2 Hz, 8.8

Hz), 0.71 (s, 3H), 0.70 (s, 3H), 0.57 (m, 1H). ¹³C NMR (CDCl₃) δ

220.2, 81.3, 53.6, 50.7, 47.2, 44.2, 38.2, 36.2, 35.3, 34.9, 34.6, 34.4,

31.0, 30.2, 28.1, 27.7, 21.2, 20.0, 13.3, 11.7.

(3β,5α)-3-[(Methylsulfonyl)oxy]-androstan-17-one (15). The

known steroid 15⁴⁰(1.62 g, 99%) was prepared from the natural

enantiomer of compound ent-14 using the procedure described for the

preparation of compound ent-15.

(3β,5β,8α,9β,10α,13α,14β)-18-Norandrostan-16-one (ent-10). A

mixture of β-ketoester ent-46 (25 mg, 0.075 mmol), LiCl (100 mg)

and DMF was heated at 160 °C for 30 min under N₂. The reaction

mixture was cooled, diluted with water, and extracted with EtOAc. The

combined organic extracts were dried and concentrated to give an oﬀ-

white solid. The crude product was puriﬁed by ﬂash column

chromatography (silica gel eluted with 30−40% EtOAc in hexanes)

to give product ent-10 as a white solid (18 mg, 86%); mp 174−177

23

°C; [α]_D+171.9 (c 0.1, CHCl₃). IR ν_max3475, 2923, 2853, 1723

(5β,8α,9β,10α,13α,14β)-Androst-2-en-17-one (ent-16). Mesylate

ent-15 (1.60 g, 4.34 mmol) and LiBr (3.89 g, 44.7 mmol) in DMF (15

mL) were heated at 125 °C under N₂for 2 h. The reaction mixture

was cooled, made basic with aqueous NaHCO₃, and extracted with

EtOAc. The combined organic extracts were washed with brine, dried,

and concentrated to give an oil which was puriﬁed by ﬂash column

chromatography (silica gel eluted with 5−20% EtOAc in hexanes) to

give a 5:1 mixture (determined by NMR) of Δ²and Δ³oleﬁns as a

white solid (1.03 g, 87%). IR ν_max3021, 2933, 2886, 2842, 1744, 1655,

1467, 1377 cm⁻¹. ¹H NMR (CDCl₃) (major isomer) δ 5.52 (m, 2H),

2.38 (dd, 1H, J = 19.2 Hz, 8.8 Hz), 0.81 (s, 3H), 0.72 (s, 3H). ¹³C

NMR (CDCl₃) (major isomer) δ 220.8, 125.5, 125.5, 54.0, 51.2, 47.4,

41.2, 39.5, 35.6, 34.9, 34.5, 31.4, 30.4, 30.0, 28.2, 21.6, 20.0, 13.5, 11.5.

(5α)-Androst-2-en-17-one (16). The known steroid 16³⁹(1.1 g,

91%, 7:1 mixture of Δ²and Δ³oleﬁns) was prepared from the natural

enantiomer of compound ent-15 using the procedure described for the

preparation of compound ent-16.

(2β,3β,5β,8α,9β,10α,13α,14β)-2,3-Epoxyandrostan-17-one (ent-

17). Compound ent-16 (containing the unseparated Δ³oleﬁn, 0.97

g, 3.56 mmol) was dissolved in CH₂Cl₂(20 mL) and HCO₂H (1.34

mL), and then 30% H₂O₂(3.3 mL) was added and the reaction was

stirred for 4 h at room temperature. MeOH (20 mL) was added and

after 3 min, 10% aqueous NaOH (15 mL) was added, and the mixture

was stirred for 5 min. Then 10% HCl (20 mL) was added dropwise

and the resulting mixture was stirred for 3 min. The product was

extracted into CH₂Cl₂(3 × 150 mL), and the organic extracts were

dried and concentrated to give an oil which was puriﬁed by ﬂash

column chromatography (silica gel eluted with 20−30% EtOAc in

hexanes) to give the epoxide ent-17 (containing the 3,4-epoxide) as a

white solid (1.06 g, 100%). IR ν_max2918, 1740, cm⁻¹. ¹H NMR

(CDCl₃) (major isomer) δ 3.10−2.98 (m, 2H), 2.40 (dd, 1H, J = 19.6

Hz, 9.0 Hz), 0.72 (s, 3H), 0.66 (s, 3H), 0.56 (m, 1H). ¹³C NMR

(CDCl₃) (major isomer) δ 220.4, 53.5, 51.8, 50.9, 50.4, 47.2, 37.9,

35.9, 35.4, 34.8, 33.4, 31.2, 30.2, 28.7, 27.8, 21.4, 19.8, 13.3, 12.3.

(2α,3α,5α)-2,3-Epoxyandrostan-17-one (17). The known steroid

17⁴¹(1.05 g, 91%) was prepared from the natural enantiomer of

compound ent-16 (containing the unseparated Δ³oleﬁn) using the

procedure described for the preparation of compound ent-17.

1

cm⁻¹. H NMR (CDCl₃) δ 4.06 (br s, 1H), 2.36 (m, 2H), 0.78 (s,

3H). ¹³C NMR (CDCl₃) δ 218.5, 66.4, 53.3, 50.2, 45.9, 43.8, 43.3,

41.7, 38.8, 36.2, 35.8, 32.2, 32.0, 31.1, 29.0, 28.1, 24.8, 11.1. Anal.

(C₁₈H₂₈O₂) C, H.

(3β,5β,8α,9β,10α,13α,14β)-19-Norandrostan-16-one (ent-11).

Compound ent-11 (207 mg, 98%) was prepared from compound

ent-33 using the procedure described for the preparation of compound

ent-7. Flash column chromatography (silica gel eluted with 25−40%

EtOAc in hexanes) gave product ent-11 as a white solid; mp 160−162

23

°C; [α]_D+162.0 (c 0.22, CHCl₃). IR ν_max3401, 2915, 2861, 1743

1

cm⁻¹. H NMR (CDCl₃) δ 4.08 (br s, 1H), 0.87 (s, 3H). ¹³C NMR

(CDCl₃) δ 218.8, 66.3, 56.0, 50.9, 48.0, 47.0, 40.53, 40.47, 39.2, 39.1,

38.1, 35.8, 33.4, 32.9, 31.2, 25.1, 23.4, 18.1. Anal. (C₁₈H₂₈O₂) C, H.

(3β,5β,8α,9β,10α,13α,14β)-3-Hydroxygonan-16-one. (ent-12). K-

Selectride (1 M in THF, 0.45 mL) was added dropwise under N₂to a

cooled solution (−78 °C) of compound ent-58 (108 mg, 0.41 mmol)

in anhydrous THF (15 mL). After 2 h stirring at −78 °C, water (2

mL) was added and the mixture was allowed to reach room

temperature. Then aqueous NaOH (2 mL, 6 M) and 30% H₂O₂(2

mL) were added and the reaction was stirred for 30 min. The product

was extracted with CH₂Cl₂(2 × 50 mL), and the combined extracts

were washed with aqueous HCl (1 N), saturated aqueous NaHCO₃,

and brine. Solvent was dried and evaporated. Flash column

chromatography (silica gel eluted with 10% EtOAc in hexanes) gave

product ent-12 (76 mg, 70%) as a white solid; mp 168−169 °C

(EtOAc/hexanes); [α]_D²³+218.6 (c 0.17, CHCl₃). IR ν_max3466, 2917,

2846, 1724 cm⁻¹. ¹H NMR (CDCl₃) δ 4.08 (1H, m), 2.30−2.41 (2H,

m), 1.97−2.10 (2H, m). ¹³C NMR (CDCl₃) δ 218.5, 66.5, 49.3, 47.7,

47.3, 47.2, 46.1, 43.8, 43.6, 40.7, 35.9, 33.5, 33.1, 31.4, 31.1, 29.6, 23.6.

Anal. (C₁₇H₂₆O₂) C, H.

(5β,8α,9β,10α,13α,14β)-Androstane-3,17-dione (ent-13). This

compound was prepared as described previously.⁹

(3α,5β,8α,9β,10α,13α,14β)-3-Hydroxyandrostan-17-one (ent-14).

A 1 M THF solution of Li(t-OBu)₃AlH (6.86 mL, 1.1 equiv) was

added to a cold (−40 °C) solution of compound ent-13 (1.80 g, 6.24

mmol) in THF (15 mL), and the resulting mixture was stirred at −40

°C for 90 min. A few drops of acetone were added to consume excess

183

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

(8α,9β,10α,13α,14β)-Androst-4-ene-3,17-dione (ent-18). This

compound was prepared as previously described.⁷

column chromatography (silica gel, eluted with 25−35% EtOAc in

hexanes) to give product ent-22 as a colorless liquid containing benzyl

acetate. The benzyl acetate was removed by applying high vacuum at

60 °C for 10 h to give product ent-22 as a white solid (1.9 g, 91%); mp

162−164 °C. IR ν_max2932, 2855, 1735, 1492, 1447, 1371, 1237 cm⁻¹.

¹H NMR (CHCl₃) δ 7.37− 7.17 (m, 5H), 6.21 (d, 1H, J = 2.4 Hz),

5.37 (s, 1H), 5.03 (s, 1H), 2.68 (dd, 1H, J = 16.8 Hz, 6.6 Hz), 2.21 (s,

3H), 2.07 (s, 3H), 0.83 (s, 3H), 0.78 (s, 3H). ¹³C NMR (CHCl₃) δ

171.2, 170.7, 141.0, 137.6, 128.3 (2 × C), 128.2 (2 × C), 126.4, 123.5,

84.6, 70.0, 54.1, 48.9, 42.9, 40.0, 36.5, 35.9, 34.8, 32.8, 32.7, 31.5, 30.9,

28.1, 26.0, 21.5, 21.2, 20.3, 12.3, 11.3.

(3α,17β)-16-(Phenylmethylene)-androstane-3,17-diol Diacetate

(22). Steroid 22 (144 mg, 90%) was prepared from the natural

enantiomer of compound ent-1 using the procedure described for the

preparation of compound ent-21. Steroid 22 had: ¹H NMR (CDCl₃) δ

7.29−7.10 (m, 5H), 6.13 (d, 1H, J = 2.4 Hz), 5.29 (s, 1H), 4.94 (s,

1H), 2.62 (dd, 1H, J = 16.8 Hz, 6.6 Hz), 2.12 (s, 3H), 1.98 (s, 3H),

0.74 (s, 3H), 0.69 (s, 3H). ¹³C NMR (CDCl₃) δ 171.1, 170.6, 140.9,

137.5, 128.2 (2 × C), 128.1 (2 × C), 126.4, 123.5, 84.5, 69.9, 54.0,

48.8, 42.9, 39.9, 36.4, 35.8, 34.8, 32.8, 32.7, 31.5, 30.9, 28.1, 26.0, 21.5,

21.1, 20.2, 12.2, 11.3.

(5β,8α,9β,10α,13α,14β)-Androst-3-en-17-one (ent-19). To a boil-

ing solution of compound ent-18 (750 mg, 2.63 mmol) in glacial

AcOH (40 mL), Zn dust (4.5g) was added in several portions during a

period of 15 min and then heating was continued an additional 15 min.

The reaction was cooled and the Zn dust was ﬁltered and the ﬁltrate

was collected. The ﬁlter-cake was washed with AcOH and EtOAc.

Solvents were removed from the combined ﬁltrates and washings,

water was added to the residue, and the product was extracted into

EtOAc. The combined extracts were washed with aqueous NaHCO₃

and brine and dried, and solvents were evaporated to give a white solid

which is a mixture (2.3:1 by NMR) of product ent-19 and the epimeric

(5α)-3-ene product. The product mixture was crystallized from

hexanes to give pure product ent-19 (310 mg, 43%); mp 122−125 °C.

IR ν_max3016, 2968, 2940, 2837, 2807, 1742, 1470, 1443, 1376, 1251

cm⁻¹. ¹H NMR (CDCl₃) δ 5.60 (m, 1H), 5.29 (m, 1H), 2.44 (dd, 1H,

J = 19.0 Hz, 9.0 Hz), 0.88 (s, 3H), 0.80 (s, 3H). ¹³C NMR (CDCl₃) δ

221.3, 131.0 125.6, 53.5, 51.5, 47.9, 45.9, 35.8, 35.1, 35.0, 34.0, 31.6,

30.9, 27.1, 23.4, 21.8, 20.3, 13.9, 11.8.

(5α)-Androst-3-en-17-one (19). The known steroid 19⁴²(350 mg,

37%) was prepared from the natural enantiomer of compound ent-18

using the procedure described for the preparation of compound ent-

19.

(3β,4β,5β,8α,9β,10α,13α,14β)-2,3-Epoxyandrostan-17-one (ent-

20). Compound ent-20 (300 mg, 84%) was prepared from compound

ent-19 using the procedure described for the preparation of compound

ent-17. The crude product was obtained as an oil and after ﬂash

column chromatography (silica gel eluted with 20−30% EtOAc in

hexanes) yielded a white solid; mp 148−150 °C. IR ν_max2928, 2882,

2859, 1740, 1472, 1446, 1405, 1373, 1251 cm⁻¹. ¹H NMR (CDCl₃) δ

3.09 (s, 1H), 2.63 (d, 1H, J = 3.9 Hz), 2.36 (dd, 1H, J = 19.6 Hz, 9.0

Hz), 0.79 (s, 3H), 0.73 (s, 3H). ¹³C NMR (CDCl₃) δ 220.7, 55.4, 52.5,

51.9, 51.0, 47.6, 46.6, 35.6, 34.8, 34.0, 31.3, 30.5, 30.2, 26.3, 21.5, 21.1,

20.4, 13.7, 13.3.

(3β,5β,8α,9β,10α,13α,14β,17α)-3,17-Dihydroxyandrostan-16-one

Diacetate (ent-23). Compound ent-23 (110 mg, 82%, starting from

compound ent-1) was prepared from compound ent-22 using the

procedure described for the preparation of compound ent-26.

1

Compound ent-23 was a white solid and had: H NMR (CDCl₃) δ

4.96 (m, 2H), 2.11 (s, 3H), 2.01 (s, 3H), 0.77 (s, 6H). ¹³C NMR

(CDCl₃) δ 211.0, 170.5, 170.2, 85.5, 69.7, 53.8, 45.2, 41.6, 39.7, 36.2,

35.9, 35.8, 34.2, 32.6, 32.4, 31.4, 27.8, 25.8, 21.4, 20.5, 19.7, 12.3, 11.2.

(3α,5α,17β)-3,17-Dihydroxyandrostan-16-one Diacetate (23).

Steroid 23 (109 mg, 91%) was prepared from the natural enantiomer

of compound ent-22 using the procedure described for the preparation

1

of compound ent-26. Steroid 23 was a white solid and had: H NMR

(CDCl₃) δ 4.98−4.95 (m, 2H), 2.14 (s, 3H), 2.03 (s, 3H), 0.78 (s,

6H). ¹³C NMR (CDCl₃) δ 211.1, 170.6, 170.3, 85.6, 69.8, 53.9, 45.3,

41.7, 39.8, 36.2, 36.0, 35.9, 34.2, 32.6, 32.5, 31.5, 27.9, 25.9, 21.4, 20.6,

19.8, 12.4, 11.2.

(2α,3α)-2,3-Epoxyandrostan-17-one (20). The known steroid 20⁴³

(300 mg, 95%) was prepared from the natural enantiomer of

compound ent-19 using the procedure described for the preparation

of compound ent-16. Steroid 20 had: IR ν_max2926, 1738, 1403, 1250

cm⁻¹. ¹H NMR (CDCl₃) δ 3.05−3.04 (m, 1H), 2.61 (d, 1H, J = 3.9),

2.35 (dd, 1H, J = 18.7, 10.2), 0.76 (s, 3H), 0.70 (s, 3H). ¹³C NMR

(CDCl₃) δ 220.5, 55.4, 52.5, 51.8, 51.0, 47.5, 46.6, 35.5, 34.7, 34.0,

31.3, 30.5, 30.1, 26.3, 21.5, 21.0, 20.3, 13.7, 13.2.

(3β,5β,8α,9β,10α,13α,14β)-3-Hydroxyandrostan-16-one Acetate

(ent-24). Freshly prepared Sm ﬁlings (1.5 mmol, 225 mg) were

added to THF (10 mL) and I₂(1.0 mmol, 254 mg) in THF (5 mL)

was added at room temperature. The suspension was stirred under N₂.

After 30 min, the mixture became deep-blue indicating SmI₂

formation, and stirring was continued for another 30 min. Compound

ent-23 (0.28 mmol, 110 mg) in THF/methanol (10/1, 11 mL) was

added. After 2 h, 10% aqueous Na₂CO₃(60 mL) was added and the

product was extracted into EtOAc (3 × 50 mL). The combined

extracts were washed with brine (2 × 20 mL) and dried. Solvents were

removed, and the residue was puriﬁed by ﬂash column chromatog-

raphy (silica gel, eluted with 20% EtOAc in hexanes) to give product

ent-24 (48 mg, 51%) and a compound in which the 16-ketone group

had been reduced to a 16-hydroxyl group of undetermined

conﬁguration (48 mg). Jones oxidation of the 16-hydroxyl group

gave additional compound ent-24 (40 mg, 43%). ¹H NMR (CDCl₃) δ

5.00 (br s, 1H), 2.03 (s, 3H), 0.86 (s, 3H), 0.81 (s, 3H). ¹³C NMR

(CDCl₃) δ 218.7, 170.6, 69.9, 55.8, 54.1, 51.7, 39.9, 39.1, 38.1, 35.9,

34.8, 32.7, 32.5, 32.1, 28.0, 26.0, 21.5, 20.3, 18.0, 11.3.

(3β,5β,8α,9β,10α,13α,14β)-3-Hydroxyestran-17-one (ent-21).

Compound ent-21 was prepared from (8α,9β,10α,13α,14β,17α)-17-

hydroxyestr-4-en-3-one³⁷using the methods reported previously for

the preparation of compound ent-1 from ent-testosterone.⁹Compound

ent-21 had: mp 158−160 °C; [α]_D= −108.5 (c 0.24, CHCl₃). IR ν_max

1

3487, 2859, 1739, 1444, 1374, 1261 cm⁻¹. H NMR (CHCl₃) δ 4.01

(s, 1H), 2.42 (dd, 1H, J = 19.3 Hz, 9.8 Hz), 0.86 (s, 3H). ¹³C NMR

(CHCl₃) δ 221.6, 66.3, 50.6, 48.2, 47.9, 47.0, 40.7, 40.5, 35.9, 35.8,

33.4, 32.9, 31.4, 29.8, 24.9, 23.7, 21.6, 13.8.

(3α,5α)-3-Hydroxyestran-17-one (21). The compound was pre-

pared as previously described.⁴⁴

(3β,5β,8α,9β,10α,13α,14β,17α)-16-(Phenylmethylene)-andros-

tane-3,17-diol Diacetate (ent-22). Compound ent-1 (1.25 g, 4.5

mmol) and benzaldehyde (1.4 mL, 14 mmol) were added to KOH

(700 mg) dissolved in EtOH (60 mL), and the reaction was stirred at

room temperature for 16 h. The reaction mixture was cooled to 0 °C,

and CeCl₃·7 H₂O (7.45 g, 20 mmol) and NaBH₄(756 mg, 20 mmol)

were added, the reaction was allowed to warm to room temperature,

and stirring continued for 3 h. Glacial AcOH (7 mL) was added, and

the product extracted into EtOAc (3 × 150 mL). The combined

extracts were dried and the solvents removed to give an oil. The oil

was dissolved in CH₂Cl₂(30 mL) and AcOAc (1.4 mL, 15 mmol),

NEt₃(4.2 mL, 30 mmol) and DMAP (200 mg) were added, and

stirring was continued at room temperature for 4 h. Aqueous saturated

NaHCO₃was then added. After 1 h, the product was extracted into

CH₂Cl₂(3 × 125 mL). The combined extracts were dried and solvents

removed to give an oil. The crude product was puriﬁed by ﬂash

(3α,5α)-3-Hydroxyandrostan-16-one Acetate (24). Steroid 24 (60

mg, 91%) was prepared from the natural enantiomer of compound

ent-23 using the procedure described for the preparation of compound

1

ent-23 and had: H NMR (CDCl₃) δ 5.00 (br s, 1H), 2.03 (s, 3H),

0.85 (s, 3H), 0.80 (s, 3H). ¹³C NMR (CDCl₃) δ 218.7, 170.6, 69.9,

55.8, 54.1, 51.7, 39.9, 39.2, 38.1, 35.9, 34.8, 32.7, 32.5, 32.1, 28.0, 26.0,

21.5, 20.3, 18.0, 11.3.

(2α,3β,5β,8α,9β,10α,13α,14β,17α)-2-Methoxy-16-(phenylmethy-

lene)-androstane-3,17-diol Diacetate (ent-25). Compound ent-25

(441 mg, 95%) was prepared as a white solid from compound ent-2

using the procedure described for the preparation of compound ent-22

and had: mp 167−169 °C. IR ν_max2932, 1737, 1599, 1492, 1448, 1369,

1241 cm⁻¹. ¹H NMR (CDCl₃) δ 7.36−7.06 (m, 5H), 6.21 (d, 1H, J =

2.4 Hz), 5.37 (s, 1H), 5.02 (br s, 1H), 3.36 (br s, 4H), 2.67 (dd, 1H, J

184

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

= 16.8 Hz, 7.0 Hz), 2.21 (s, 3H), 2.08 (s, 3H), 0.96 (s, 3H), 0.78 (s,

3H). ¹³C NMR (CDCl₃) δ 171.2, 170.3, 141.0, 137.6, 128.3 (2 × C),

128.2 (2 × C), 126.4, 123.6, 84.6, 77.7, 69.9, 57.0, 54.8, 48.9, 43.0,

39.9, 37.0, 36.5, 35.7, 34.3, 31.5, 30.9, 29.0, 27.7, 21.4, 21.2, 20.3, 13.1,

12.3.

(3α,4β,17β)-4-Methoxy-16-(phenylmethylene)-androstane-3,17-

diol Diacetate (28). Steroid 28 (126 mg, 100%) was prepared as a

white semisolid from the natural enantiomer of compound ent-3 using

the procedure described for the preparation of compound ent-22.

Steroid 28 had: IR ν_max1733, 1241 cm⁻¹. ¹H NMR (CDCl₃) δ 7.28−

7.11 (m, 5H), 6.13 (d, 1H, J = 1.9 Hz), 5.28−5.26 (m, 1H), 4.97−4.96

(m, 1H), 3.29 (s, 3H), 2.62−2.57 (m, 1H), 2.11 (s, 3H), 2.00 (s, 3H),

0.90 (s, 3H), 0.69 (s, 3H). ¹³C NMR (CDCl₃) δ 171.1, 170.4, 141.0,

137.6, 128.3 (2 × C), 128.2 (2 × C), 126.4, 123.5, 84.6, 82.5, 68.6,

58.9, 55.0, 49.0, 45.0, 42.9, 36.4, 35.9, 34.8, 32.6, 31.8, 30.9, 25.0, 22.1,

21.4, 21.1, 19.9, 14.0, 12.3.

(2β,3α,17β)-2-Methoxy-16-(phenylmethylene)-androstane-3,17-

diol Diacetate (25). Steroid 25 (150 mg, 95%) was prepared from the

natural enantiomer of compound ent-2 using the procedure described

for the preparation of compound ent-22 and had: IR ν_max1737, 1240

1

cm⁻¹. H NMR (CDCl₃) δ 7.28−7.11 (m, 5H), 6.13 (d, 1H, J = 2.3

Hz), 5.28−5.27 (m, 1H), 4.94 (br s, 1H), 3.27 (br s, 4H), 2.58−2.54

(m, 1H), 2.12 (s, 3H), 1.99 (s, 3H), 0.88 (s, 3H), 0.69 (s, 3H). ¹³C

NMR (CDCl₃) δ 171.1, 170.2, 140.9, 137.5, 128.2 (2 × C), 128.1 (2 ×

C), 126.4, 123.5, 84.5, 77.6, 69.8, 56.9, 54.8, 48.8, 42.9, 39.9, 36.9, 36.5,

35.6, 34.2, 31.4, 30.9, 29.0, 27.6, 21.4, 21.4, 21.1, 20.3, 13.0, 12.2.

(2α,3β,5β,8α,9β,10α,13α,14β,17α)-3,17-Dihydroxy-2-methoxyan-

drostan-16-one Diacetate (ent-26). A stream of O₃was bubbled

through a solution of compound ent-25 (345 mg, 0.70 mmol)

dissolved in MeOH/EtOAc (2:1; 20 mL) at −78 °C until a blue color

persisted for 10 min. The excess O₃was purged from the reaction

using an O₂stream, as evidenced by the solution becoming colorless.

Me₂S was then added (2 mL), and the reaction was allowed to warm

to room temperature and then allowed to stir for 14 h. Solvents were

removed, and the residue was puriﬁed by ﬂash column chromatog-

raphy (silica gel, eluted with 30−40% EtOAc in hexanes) to yield

compound ent-26 as a foamy white solid (265 mg, 90%). IR ν_max2934,

1762, 1742, 1451, 1370, 1240 cm⁻¹. ¹H NMR (CDCl₃) δ 4.98 (d, 1H,

J = 2.0 Hz), 4.97 (s, 1H), 3.36 (br s, 4H), 2.13 (s, 3H), 2.04 (s, 3H),

0.93 (s, 3H), 0.79 (s, 3H). ¹³C NMR (CDCl₃) δ 210.9, 170.2, 170.1,

85.6, 77.5, 69.7, 56.9, 54.6, 45.2, 41.7, 39.8, 36.7, 36.2, 36.0, 35.7, 33.7,

31.5, 28.8, 27.4, 21.3, 20.6, 19.9, 13.0, 12.4.

(3β,4α,5β,8α,9β,10α,13α,14β,17α)-3,17-Dihydroxy-4-methoxyan-

drostan-16-one Diacetate (ent-29). Compound ent-29 (205 mg,

87%) was prepared as a white solid from compound ent-28 using the

procedure described for the preparation of compound ent-23.

Compound ent-29 had: mp 158−160 °C. IR ν_max2928, 1762, 1742,

1449, 1372, 1240 cm⁻¹. ¹H NMR (CDCl₃) δ 5.02 (d, 1H, J = 2.0 Hz),

4.97 (s, 1H), 3.34 (s, 3H), 3.01 (br s, 1H), 2.13 (s, 3H), 2.05 (s, 3H),

0.96 (s, 3H), 0.79 (s, 3H). ¹³C NMR (CDCl₃) δ 211.0, 170.3 (2 × C),

85.6, 82.2, 68.5, 58.8, 54.8, 45.3, 44.8, 41.6, 36.2, 36.0, 35.8, 34.3, 32.3,

31.8, 24.8, 21.9, 21.3, 20.6, 19.4, 13.9, 12.4.

(3α,4β,17β)-3,17-Dihydroxy-4-methoxyandrostan-16-one Diace-

tate (29). Steroid 29 (96 mg, 96%) was prepared as a foamy solid from

the natural enantiomer of compound ent-28 using the procedure

described for the preparation of compound ent-26. Steroid 29 had: ¹H

NMR (CDCl₃) δ 4.97−4.92 (m, 2H), 3.29 (s, 3H), 2.97 (br s, 1H),

2.08 (s, 3H), 2.00 (s, 3H), 0.90 (s, 3H), 0.74 (s, 3H). ¹³C NMR

(CDCl₃) δ 211.0, 170.3 (2 × C), 85.6, 82.2, 68.5, 58.8, 54.8, 45.3, 44.8,

41.6, 36.2, 36.1, 35.8, 34.3, 32.3, 31.9, 24.8, 21.9, 21.4, 20.6, 19.4, 14.0,

12.4.

(3β,4α,5β,8α,9β,10α,13α,14β)-3-Hydroxy-4-methoxyandrostan-

16-one Acetate (ent-30). Compound ent-30 (72 mg, 84%) was

prepared as a white solid from compound ent-29 using the procedure

described for the preparation of compound ent-24. Compound ent-30

had: mp 159−161 °C. IR ν_max2944, 2861, 1733, 1449, 1407, 1380,

1244, 1216 cm⁻¹. ¹H NMR (CDCl₃) δ 5.03 (d, 1H, J = 1.6 Hz), 3.36

(s, 3H), 3.03 (br s, 1H), 2.07 (s, 3H), 0.98 (s, 3H), 0.87 (s, 3H). ¹³C

NMR (CDCl₃) δ 218.8, 170.4, 82.4, 68.6, 58.9, 55.8, 55.0, 51.8, 44.9,

39.3, 39.1, 38.1, 35.9, 34.9, 32.4, 32.4, 24.9, 22.0, 21.4, 19.9, 18.1, 14.0.

(3α,4β,5α)-3-Hydroxy-4-methoxyandrostan-16-one Acetate (30).

Steroid 30 (64 mg, 78%) was prepared as a foamy solid from the

natural enantiomer of compound ent-29 using the procedure described

(2β,3α,17β)-3,17-Dihydroxy-2-methoxyandrostan-16-one Diace-

tate (26). Steroid 26 (120 mg, 94%) was prepared as a foamy solid

from the natural enantiomer of compound ent-25 using the procedure

described for the preparation of compound ent-26. Steroid 26 had: IR

ν_max1762, 1741, 1239 cm⁻¹. ¹H NMR (CDCl₃) δ 4.94−4.93 (m, 2H),

3.28 (br s, 4H), 2.09 (s, 3H), 2.00 (s, 3H), 0.89 (s, 3H), 0.75 (s, 3H).

¹³C NMR (CDCl₃) δ 211.0, 170.3, 170.1, 85.6, 77.5, 69.7, 56.9, 54.6,

45.2, 41.7, 39.8, 36.7, 36.3, 36.0, 35.7, 33.7, 31.5, 28.9, 27.4, 21.3, 20.6,

19.9, 13.0, 12.4.

(2α,3β,5β,8α,9β,10α,13α,14β)-3-Hydroxy-2-Methoxyandrostan-

16-one Acetate (ent-27). Compound ent-27 (170 mg, 94%) was

prepared as a foamy solid from compound ent-26 using the procedure

described for the preparation of compound ent-24. Compound ent-27

had: IR ν_max2932, 2850, 1742, 1453, 1408, 1369, 1243, 1214 cm⁻¹. ¹H

NMR (CDCl₃) δ 4.98 (s, 1H), 3.32 (s, 4H), 2.04 (s, 3H), 0.94 (s, 3H),

0.85 (s, 3H). ¹³C NMR (CDCl₃) δ 218.5, 170.1, 77.6, 69.8, 56.9, 55.8,

54.8, 51.6, 39.8, 39.2, 39.1, 38.1, 36.7, 35.7, 34.3, 32.1, 28.9, 27.6, 21.3,

20.3, 18.0, 13.0.

(2β,3α,5α)-3-Hydroxy-2-methoxyandrostan-16-one (27). Steroid

27 (85 mg, 83%) was prepared from the natural enantiomer of

compound ent-26 using the procedure reported for the preparation of

compound ent-24. Steroid 27 had: IR ν_max1741, 1243 cm⁻¹. ¹H NMR

(CDCl₃) δ 4.98−4.97 (m, 1H), 3.32 (s, 4H), 2.04 (s, 3H), 0.94 (s,

3H), 0.85 (s, 3H). ¹³C NMR (CDCl₃) δ 218.6, 170.1, 77.5, 69.8, 56.9,

55.8, 54.8, 51.7, 39.8, 39.2, 39.1, 38.2, 36.7, 35.7, 34.3, 32.1, 28.9, 27.6,

21.3, 20.3, 18.1, 13.1.

1

for the preparation of compound ent-24. Steroid 30 had: H NMR

(CDCl₃) δ 5.04−5.03 (m, 1H), 3.35 (s, 3H), 3.02 (br s, 1H), 2.06 (s,

3H), 0.98 (s, 3H), 0.86 (s, 3H). ¹³C NMR (CDCl₃) δ 218.8, 170.4,

82.4, 68.6, 58.9, 55.8, 55.0, 51.8, 44.9, 39.3, 39.1, 38.1, 35.9, 34.9, 32.5,

32.4, 25.0, 22.0, 21.4, 19.9, 18.1, 14.0.

(3β,5β,8α,9β,10α,13α,14β,17α)-19-Nor-16-(phenylmethylene)-

androstane-3,17-diol Diacetate (ent-31). Compound ent-31 (670

mg, 91%) was prepared as an oil from compound ent-21 using the

procedure described for the preparation of compound ent-22.

Compound ent-31 had: IR ν_max2919, 2862, 1738, 1600, 1492, 1446,

1370, 1240 cm⁻¹. ¹H NMR (CDCl₃) δ 7.40−7.10 (m, 5H), 6.24 (b s,

1H), 5.40 (b s 1H), 5.07 (b s, 1H), 2.69 (dd, 1H, J = 16.7 Hz, 6.3 Hz),

2.23 (s, 3H), 2.07 (s, 3H), 0.81 (s, 3H). ¹³C (CDCl₃) NMR δ 171.0,

170.6, 140.9, 137.5, 128.2 (4 × C), 126.4, 123.5, 84.5, 69.7, 48.0, 47.8,

46.5, 42.9, 40.4, 37.4, 36.6, 36.4, 33.2, 30.7, 30.4, 29.9, 25.0, 24.2, 21.4,

21.1, 12.1. Anal. (C₂₉H₃₈O₄) C, H.

(3β,4α,5β,8α,9β,10α,13α,14β)-4-Methoxy-16-(phenylmethylene)-

androstane-3,17-diol Diacetate (ent-28). Compound ent-28 (300

mg, 97%) was prepared as a white solid from compound ent-3 using

the procedure described for the preparation of compound ent-22.

Compound ent-28 had: mp 150−152 °C. IR ν_max2935, 2859, 1738,

(3β,5β,8α,9β,10α,13α,14β,17α)-19-Nor-3,17-dihydroxyandro-

stan-16-one Diacetate (ent-32). Compound ent-32 (530 mg, 98%)

was prepared as a white solid from compound ent-31 using the

procedure described for the preparation of compound ent-26.

Compound ent-32 had: mp 116−118 °C. IR ν_max2919, 2861, 1763,

1

1599, 1492, 1448, 1371, 1241 cm⁻¹. H NMR (CDCl₃) δ 7.30−7.08

1

1736, 1445, 1371, 1239, 1216 cm⁻¹. H NMR (CDCl₃) δ 4.91 (br s,

(m, 5H), 6.13 (d, 1H, J = 2.2 Hz), 5.28 (s, 1H), 4.97 (d, 1H, J = 2.3

Hz), 3.30 (s, 3H), 3.22 (s, 1H), 2.60 (m, 1H), 2.12 (s, 3H), 2.00 (s,

3H), 0.90 (s, 3H), 0.69 (s, 3H). ¹³C NMR (CDCl₃) δ 171.2, 170.4,

141.0, 137.6, 128.3 (2 × C), 128.2 (2 × C), 126.4, 123.5, 84.6, 82.5,

68.6, 58.9, 55.0, 49.0, 45.0, 42.9, 36.4, 35.9, 34.8, 32.5, 31.8, 30.9, 25.0,

22.0, 21.4, 21.2, 19.9, 14.0, 12.3.

1H), 4.89 (s, 1H), 2.19 (dd, 1H, J = 18.7 Hz, 7.7 Hz), 2.03 (s, 3H),

1.92 (s, 3H), 0.71 (s, 3H). ¹³C NMR (CDCl₃) δ 210.6, 170.1, 169.9,

85.4, 69.3, 47.4, 46.1, 44.1, 41.5, 39.5, 37.1, 36.2, 35.9, 35.6, 32.8, 30.3,

29.6, 24.3, 23.7, 21.1, 20.3, 12.0. Anal. (C₂₂H₃₂O₅) C, H.

(3β,5β,8α,9β,10α,13α,14β)-19-Nor-3-hydroxyandrostan-16-one

Acetate (ent-33). Compound ent-33 (290 mg, 70%) was prepared as a

185

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

colorless oil from compound ent-32 using the procedure described for

the preparation of compound ent-24. Compound ent-33 had: IR ν_max

2918, 2862, 1737, 1444, 1408, 1378, 1244, 1214 cm⁻¹. ¹H NMR

(CDCl₃) δ 4.97 (b s, 1H), 1.97 (s, 3H), 0.82 (s, 3H). ¹³C NMR

(CDCl₃) δ 218.1, 170.3, 69.6, 55.8, 50.8, 47.8, 46.5, 40.4, 39.0, 38.9,

38.0, 37.3, 36.5, 36.5, 33.1, 31.0, 29.8, 25.0, 23.3, 21.2, 17.9. Anal.

(C₂₀H₃₀O₃) C,H.

10−15% EtOAc in hexanes) to give product ent-38 as a colorless

liquid (960 mg, 90%). IR ν_max2924, 2857, 1445, 1368, 1302, 1213

1

cm⁻¹. H NMR (CDCl₃) δ 4.57 (s, 2H), 3.75 (s, 1H), 3.28 (s, 3H),

1.51 (s, 3H), 0.64 (s, 3H). ¹³C NMR (CDCl₃) δ 136.4, 127.1, 94.3,

71.4, 54.8, 53.8, 52.2, 45.3, 39.2, 37.0, 35.8, 33.5, 32.8, 32.2, 28.3, 28.0,

26.2, 25.6, 25.2, 13.2, 11.2.

(1R,4aR,4bR,7S,8aR,10aS)-7-(Methoxymethoxy)docecahydro-4b-

methyl-1(3-oxobutyl-2(1H)-phenanthrenone (ent-39). Compound

ent-38 (2.38 g, 7.5 mmol) was dissolved in CH₂Cl₂(100 mL) and

cooled to −78 °C. O₃was passed through the solution until it

remained blue for 20 min, and then O₂was passed through the

solution until it became colorless. The reaction was warmed to 0 °C,

AcOH (30 mL) was added, and the CH₂Cl₂was removed on a rotary

evaporator. Additional AcOH (15 mL) and Zn dust (6 g) were added,

and the reaction was stirred at room temperature for 3 h to decompose

the ozonide intermediate. The Zn was removed by ﬁltration through

Celite, and the Celite was washed with EtOAc. The combined ﬁltrate

and washings were combined and the volatile solvents removed to

leave the product in AcOH. The AcOH was carefully neutralized with

aqueous NaHCO₃and the product extracted into EtOAc. The

combined extracts were dried and solvent removed to leave the crude

product as an oil which was puriﬁed by ﬂash column chromatography

(silica gel eluted with 20−35% EtOAc in hexanes) to give product ent-

39 (1.7 g, 65%) as a liquid. IR ν_max2928, 1712, 1447, 1367, 1213 cm⁻¹.

¹H NMR (CDCl₃) δ 4.57 (s, 2H), 3.75 (s, 1H), 3.28 (s, 3H), 2.03 (s,

(5β,8α,9β,10α,13α,14β,17α)-17-Androstan-3-one (ent-34). This

compound was prepared as previously described.⁴⁵

(5β,8α,9β,10α,13α,14β,17α)-17-[[(4-Phenylmethyl)sulfonyl]oxy]-

androstan-3-one (ent-35). Compound ent-34 (1.8 g, 6.54 mmol), p-

TsCl (2.5 g, 13.08 mmol), and pyridine (60 mL) were heated at 40 °C

for 18 h. The reaction mixture was cooled and poured into a beaker

containing crushed ice. The mixture was extracted with CH₂Cl₂(3 ×

100 mL), and the combined extracts were washed with 1N HCl to

remove the pyridine, washed with brine, dried, and solvent removed.

The crude product was puriﬁed by ﬂash column chromatography

(silica gel eluted with 10% EtOAc in CH₂Cl₂) to give ent-35 as a white

crystalline solid (2.65 g, 91%); mp 180−182 °C. IR ν_max2917, 2851,

1710, 1598, 1449, 1415, 1361, 1253, 1217 cm⁻¹. ¹H NMR (CDCl₃) δ

7.77 (d, 2H, J = 7.8 Hz), 7.32 (d, 2H, J = 7.8 Hz), 4.24 (t, 1H, J = 8.2

Hz), 2.44 (s, 3H), 0.98 (s, 3H), 0.80 (s, 3H). ¹³C NMR (CDCl₃) δ

211.7, 144.4, 134.1, 129.6 (2 × C), 127.76 (2 × C), 89.8, 53.5, 49.7,

46.5, 44.5, 43.0, 38.4, 38.0, 36.0, 35.6, 35.0, 31.0, 28.6, 27.6, 23.2, 21.6,

21.6, 20.7, 11.8. 11.4.

(3β,5β,8α,9β,10α,13α,14β,17α)-17-[[(4-Phenylmethyl)sulfonyl]-

oxy]-androstan-3-ol (ent-36). Compound ent-35 (2.8 g, 6.3 mmol)

was dissolved in THF (50 mL) and cooled to −78 °C, and K-

selectride in THF (1 M, 7.2 mL, 7.2 mmol, 1.2 equiv) was added

dropwise. Stirring at −78 °C was continued for 2 h. Aqueous NaOH

(3 M, 20 mL) and then 30% H₂O₂(20 mL) were added, and the

reaction was warmed to room temperature and stirred for 2 h. Water

(150 mL) was added, and the product was extracted into EtOAc. The

combined EtOAc extracts were dried and the solvents removed to give

an oil, which was puriﬁed by ﬂash column chromatography (silica gel

eluted with 10% EtOAc in CH₂Cl₂) to give product ent-36 (2 g, 71%);

mp 186−188 °C. IR ν_max3400, 2918, 2850, 1599, 1446, 1355 cm⁻¹. ¹H

NMR (CDCl₃) δ 7.77 (d, 2H, J = 7.1 Hz), 7.32 (d, 2H, J = 7.1 Hz),

4.24 (t, 1H, J = 7.8 Hz), 4.03 (d, 1H, J = 2.3 Hz) 2.44 (s, 3H), 0.78 (s,

3H), 0.75 (s, 3H). ¹³C NMR (CDCl₃) δ 144.1, 134.2, 129.6 (2 × C),

127.8 (2 × C), 90.1, 66.4, 66.3, 54.1, 50.0, 43.0, 39.0, 36.1, 35.7, 35.1,

32.1, 31.3, 28.9, 28.2, 27.6, 23.2, 21.6, 20.0, 11.8, 11.1.

3H), 0.65 (s, 3H). ¹³C NMR (CDCl₃) δ 212.2, 208.7, 94.3, 71.2, 54.9,

54.6, 52.1, 42.3, 41.7, 40.8, 38.7, 36.0, 33.0, 32.6, 32.0, 29.6, 28.0, 26.2,

26.1, 19.2, 10.9.

(3β,5β,8α,9β,10α,14β)-3-(Methoxymethoxy)-^D-homo-18-noran-

drost-13(17a)-en-17-one Acetate (ent-40). Compound ent-39 (1.55

g, 4.43 mmol) was dissolved in MeOH (15 mL) and water (1.5 mL),

10% methanolic NaOH (12 mL) was added, and the reaction was

stirred at room temperature for 1 h. The reaction was then acidiﬁed

with 3N HCl and the product extracted into CH₂Cl₂. The combined

extracts were washed with aq NaHCO₃and brine, dried, and the

solvents removed. The crude product was puriﬁed by ﬂash column

chromatography (silica gel eluted with 30−40% ethyl acetate in

hexanes) to give product ent-40 as a white solid. (1.10 g, 75%); mp

98−100 °C. IR ν_max2926, 1674, 1622, 1449, 1364, 1259, 1209 cm⁻¹.

¹H NMR (CDCl₃) δ 5.68 (s, 1H), 4.55 (s, 2H), 3.74 (s, 1H), 3.27 (s,

3H), 0.63 (s, 3H). ¹³C NMR (CDCl₃) δ 199.5, 166.4, 123.7, 94.3,

71.1, 54.9, 52.0, 43.6, 43.0, 38.6, 36.2, 35.8, 35.8, 35.7, 33.1, 32.5, 31.1,

27.9, 26.1, 25.9, 25.3, 10.9.

(3β,5β,8α,9β,10α,14β)-17-Methylandrost-14(17)-en-3-ol (ent-37).

MeMgBr (3 M in Et₂O, 5 mL) was added to reﬂuxing toluene (80

mL) containing dissolved compound ent-36 (1.34 g, 3 mmol), and

reﬂuxing was continued for 1 h under N_2.The reaction was cooled,

acidiﬁed with 1N HCl, and extracted with CH₂Cl₂. The combined

extracts were washed with brine, dried, and concentrated to give an oil.

The oil was dissolved in MeOH (20 mL) and 5 M HCl (prepared

from concentrated HCl diluted with MeOH) was added and the

reaction was stirred at room temperature for 15 h. The reaction was

made basic by the addition of aqueous NaHCO₃, and the MeOH was

removed. The aqueous residue was diluted further with water and the

product extracted into EtOAc. The combined extracts were dried and

concentrated to give an oil, which was puriﬁed by ﬂash column

chromatography (silica gel eluted with 15−35% EtOAc in hexanes) to

give product ent-37 as a colorless oil (750 mg, 91%). IR ν_max3337,

(3β,5β,8α,9β,10α,13α,14β)-3-(Methoxymethoxy)-^D-homo-18-nor-

androstan-17-one (ent-41). Liquid NH₃(250 mL) was condensed in

a 3-necked 500 mL round-bottom ﬂask ﬁtted with a Dewar condenser

containing dry ice/acetone and an overhead stirrer and then placed in

a cold bath at −78 °C. Li (140 mg, 20 mmol) was added and stirring

continued for 30 min, during which time the solution became deep

blue. THF (80 mL) was added, and after 10 min, compound ent-40

(942 mg, 2.7 mmol) dissolved in THF (40 mL) was added and stirring

continued for 1 h. Solid NH₄Cl (5 g) was added, and the solution

became colorless. The reaction was allowed to warm to room

temperature, and the liquid NH₃was allowed to evaporate. Water was

added and the product extracted into EtOAc. The combined extracts

were washed with brine, dried, and solvent removed to give a yellow

solid which was puriﬁed by ﬂash column chromatography (silica gel

eluted with 30% EtOAc in hexanes) to give product ent-41 as a white

solid (650 mg, 72%); mp 85−87 °C. IR ν_max2919, 2859, 1717, 1445,

1366, 1322, 1206, 1209 cm⁻¹. ¹H NMR (CDCl₃) δ 4.65 (s, 2H), 3.82

(s, 1H), 3.36 (s, 3H), 0.72 (s, 3H). ¹³C NMR (CDCl₃) δ 211.9, 94.5,

71.4, 55.1, 52.9, 48.6, 47.4, 43.3, 41.3, 40.8, 39.1, 35.9, 34.3, 33.5. 32.6,

30.9, 30.3, 28.3, 26.2, 24.6, 11.2.

1

2920, 2856, 1444, 1359, 1255 cm⁻¹. H NMR (CDCl₃) δ 4.02 (br s,

1H), 1.58 (s, 3H), 0.68 (s, 3H). ¹³C NMR (CDCl₃) δ 136.5, 127.4,

66.4, 53.9, 52.3, 45.3, 38.7, 37.1, 36.1, 35.8, 32.3, 32.2, 28.9, 28.3, 28.1,

25.6, 25.2, 13.4, 11.1.

(3β,5β,8α,9β,10α,14β)-3-(Methoxymethoxy)-17-methylandrost-

14(17)-ene (ent-38). Compound ent-37 (823 mg, 3 mmol) was

dissolved CH₂Cl₂(80 mL) and cooled to 0 °C. (i-Pr)₂EtN (2.6 mL, 15

mmol) and ClCH₂OMe (0.76 mL, 10 mmol) were added, and the

reaction was stirred at room temperature for 6 h. The reaction mixture

was made basic by adding aqueous NaHCO₃solution and the product

extracted into CH₂Cl₂. The combined extracts were washed with brine,

dried, and solvent removed to give a viscous liquid. The crude product

was puriﬁed by ﬂash column chromatography (silica gel eluted with

(3β,5β,8α,9β,10α,13α,14β)-3-Hydroxy-^D-homo-18-norandrostan-

17-one (ent-42). Compound ent-41 (500 mg, 1.50 mmol) dissolved in

MeOH (20 mL) was stirred with 6 N HCl (6 mL) at room

temperature for 24 h. The HCl was neutralized by careful addition of

aqueous NaHCO₃solution and the MeOH removed. Water was

added, and the product was extracted into EtOAc. The combined

extracts were washed with brine, dried, and the solvents removed to

186

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

give the crude product as an oﬀ-white solid which was puriﬁed by ﬂash

column chromatography (silica gel eluted with 40−50% EtOAc in

hexanes). Compound ent-42 was obtained as a white solid (400 mg,

92%); mp 227−229 °C. IR ν_max3395, 2939, 2922, 2850, 1699, 1594,

1419, 1360, 1320, 1264, 1208 cm⁻¹. ¹H NMR (CDCl₃) δ 4.04 (s, 1H),

0.71 (s, 3H). ¹³C NMR (CDCl₃) δ 211.9, 66.2, 52.9, 48.6, 47.4, 43.2,

41.2, 40.8, 38.4, 36.1, 35.7, 34.3, 32.0, 31.0, 30.3, 28.9, 28.2, 24.6, 11.0.

(3β,5β,8α,9β,10α,13α,14β)-3-Hydroxy-^D-Homo-18-norandrostan-

17-one Acetate (ent-43). Compound ent-42 (370 mg, 1.27 mmol),

AcOAc (0.5 mL), and Et₃N (1 mL) in CH₂Cl₂(8 mL) were stirred at

room temperature for 24 h. Aqueous satd NaHCO₃(5 mL) was

carefully added, and the mixture was stirred for 1 h. The product was

extracted into CH₂Cl₂, and the combined extracts were washed with

brine, dried, and the solvent removed to give an oﬀ-white solid. The

crude product was puriﬁed by ﬂash column chromatography (silica gel

eluted with 20−30% EtOAc in hexanes) to give product ent-43 as a

white solid (400 mg, 95%); mp 141−143 °C. IR ν_max2920, 2860,

2733, 1717, 1446, 1361, 1246 cm⁻¹. ¹H NMR (CDCl₃) δ 4.96 (s, 1H),

2.00 (s, 3H), 0.69 (s, 3H). ¹³C NMR (CDCl₃) δ 211.6, 170.4, 69.8,

69.6, 52.7, 48.5, 47.3, 43.1, 41.1, 40.7, 39.2, 35.7, 34.2, 32.6, 32.5, 30.8,

30.2, 27.9, 25.9, 24.5, 21.4, 21.3, 11.1, 11.0.

precipitated and was ﬁltered, washed with water, and dried overnight

at room temperature to give pure dicarboxylic acid (5.53 g, 60%) as a

white solid. The aqueous phase was extracted with CH₂Cl₂(2 × 150

mL), and the combined extracts were washed with brine (2 × 100

mL), dried, and evaporated to aﬀord impure dicarboxylic acid 48 (2 g,

21%) as an oily residue. Pure steroid 48 had: ¹H NMR (CDCl₃) δ 0.88

(3H, s). ¹³C NMR (75 MHz, CDCl₃) δ 221.5, 179.4, 179.3, 50.6, 47.9,

47.7, 43.2, 40.6, 39.4, 38.8, 36.0, 35.3, 32.3, 31.6, 29.4, 26.0, 21.7, 13.9.

AcCl (6 mL, 23 mmol) was added to steroid 48 (7.53 g) dissolved

in MeOH (100 mL), and the reaction was stirred at room temperature

for 2 h. Then water (50 mL) was added and the product was extracted

into CH₂Cl₂(2 × 100 mL). The combined extracts were washed with

water, dried, and the solvent removed to yield an oily residue which

was puriﬁed by ﬂash column chromatography (silica gel eluted with

15% EtOAc in hexanes) to give steroid 49 (4.43 g, 52%) as an oil.

Unreacted steroid 48 was reclaimed by washing the silica gel with

MeOH, and the esteriﬁcation procedure was repeated to give

additional product 49 (780 mg, 9%) as an oil. IR ν_max2925, 2858,

1737 cm⁻¹. ¹H NMR (CDCl₃) δ 3.65 (6H, s), 0.86 (3H, s). ¹³C NMR

(CDCl₃) δ 220.9, 173.6 (2 × C), 51.7, 51.6, 50.6, 47.99, 47.90, 43.3,

40.6, 39.3, 39.1, 36.0, 35.3, 32.3, 31.7, 29.4, 25.9, 21.7, 13.9.

(3β,5α)-3-(Acetyloxy)-16,17-seco-^D-homo-18-norandrostane-

16,17-dioic Acid (ent-44) and (3β,5α)-3-(Hydroxy)-16,17-seco-^D-

homo-18-norandrostane-16,17-dioic Acid Dimethyl Ester (ent-45).

A mixture of compound ent-43 (96 mg) and a CrO₃solution (0.7 mL)

[prepared by mixing CrO₃(500 mg), water (0.7 mL), and acetic acid

(0.8 mL)] and a solution of aqueous methanolic acetic acid (1 mL)

[prepared by mixing acetic acid (30 mL), water (1 mL) and methanol

(0.05 mL)] was stirred at 60 °C for 4.5 h. The reaction was cooled to

room temperature, water (2.5 mL) was added, and the reaction was

stirred at room temperature for 13 h. Additional water (15 mL) was

added, and the product was extracted into CH₂Cl₂. The combined

extracts were dried and the solvent removed to give dicarboxylic acid

ent-44 as a white solid, which was immediately converted without

characterization to diester ent-45 upon stirring with dry HCl in MeOH

(10 mL) [prepared by adding AcCl (3 mL) to MeOH (8 mL)] at

room temperature for 15 h. The HCl was cautiously neutralized with

aqueous NaHCO₃, and the MeOH was removed. Water was added,

and the product was extracted into EtOAc. The combined extracts

were washed with brine, dried, and solvent removed to give an oil.

Puriﬁcation by ﬂash column chromatography (silica gel eluted with

40% EtOAc in hexanes) yielded compound ent-45 as a colorless oil

(54 mg, 51%). IR ν_max3446, 2924, 2858, 1738, 1436, 1361, 1258 cm⁻¹.

¹H NMR (CDCl₃) δ 4.03 (s, 1H), 3.65 (s, 6H), 0.71 (s, 3H). ¹³C

NMR (CDCl₃) δ 173.7 (2 × C), 66.3, 53.0, 51.5, 51.4, 45.0, 40.8, 39.3,

38.8, 38.3, 36.1, 35.7, 35.3, 32.7, 31.9, 31.3, 28.9, 28.4, 24.4, 11.0.

(3β,5β,8α,9β,10α,13α,14β,17α)-3-Hydroxy-16-oxo-18-norandros-

tane-17-carboxylic Acid Methyl Ester (ent-46). A mixture of

compound ent-45 (53 mg, 0.14 mmol), NaOMe (27 mg, 0.5

mmol), and THF (8 mL) was heated at reﬂux for 0.5 h. The reaction

was cooled, acidiﬁed with 1 N HCl to pH 3, and the product extracted

into CH₂Cl₂(3 × 75 mL). The combined extracts were washed with

brine, dried, and solvents removed to give a white solid. The crude

product was puriﬁed by ﬂash column chromatography (silica gel eluted

with 40% EtOAc in hexanes) gave pure product ent-46 as a white solid

(34 mg, 71%); mp 136−138 °C. IR ν_max3435, 2920, 2855, 1756, 1727,

1435, 1410, 1384, 1339, 1261 cm⁻¹. ¹H NMR (CDCl₃) δ 4.06 (t, 1H, J

= 2.5 Hz), 3.75 (s, 3H), 2.86 (d, 1H, J = 2.5 Hz), 2.50 (dd, 1H, J =

18.0 Hz, 5.8 Hz), 0.78 (s, 3H). ¹³C NMR (CDCl₃) δ 210.2, 169.5,

66.3, 61.9, 53.1, 52.3, 47.1, 46.6, 43.2, 41.6, 38.7, 36.2, 35.7, 32.2, 31.9,

30.0, 28.9, 28.0, 24.5, 11.0.

(5α,17β)-17-Hydroxy-3,4-secoestrane-2,3-dicarboxylic Acid Di-

methyl Ester (50). Steroid 49 (1.55 g, 31.4 mmol) was dissolved in

stirred EtOH (20 mL) and cooled in an ice-bath. NaBH₄(218 mg, 5.8

mmol) was added in small portions. After 1.5 h, a solution of water (50

mL) and AcOH (2 mL) was added and the solution was stirred for 1h.

The product was extracted into CH₂Cl₂(2 × 125 mL), washed with an

aqueous 1 N HCl, satd aqueous NaHCO₃, and brine, dried, and the

solvent removed. Flash column chromatography (silica gel eluted with

15% EtOAc in hexanes) gave secosteroid 50 (1.16 g, 74%) as an oily

product. IR ν_max3452, 2950, 2921, 2867, 1733 cm⁻¹. ¹H NMR

(CDCl₃) δ 3.60−3.65 (7H, m), 0.73 (3H, s). ¹³C NMR (CDCl₃) δ

173.7 (2 × C), 82.0, 51.68, 51.62, 50.2, 47.9, 43.5, 43.1, 41.2, 39.5,

39.2, 36.9, 35.4, 32.5, 30.7, 30.1, 26.3, 23.3, 11.2.

(3α,5α,17β)-17-Hydroxy-2-oxo-A-norestrane-3-carboxylic Acid

Methyl Ester (51). A NaOMe solution (5 mL) was prepared by

dissolving Na (1.14 g, 49 mmol) in MeOH (25 mL) followed by

MeOH evaporation on a rotary evaporator. Dry THF (30 mL) was

added, and the ﬂask was ﬁlled with N₂. Steroid 50 (5 g, 14.1 mmol)

dissolved in dry THF (50 mL) was slowly added. The reaction mixture

was heated at 100 °C for 1 h under N₂and then allowed to attain

room temperature. Then 1 N HCl (15 mL) was slowly added and the

product was extracted into CH₂Cl₂(2 × 100 mL). The combined

extracts were washed with brine, dried, and the solvent removed. Flash

column chromatography (silica gel eluted with 15% EtOAc in

hexanes) gave product 51 (3.7 g, 81%) as a white solid; mp 138−

142 °C (EtOAc/hexanes). IR ν_max3440, 2918, 2866, 1756, 1727 cm⁻¹.

¹H NMR (CDCl₃) δ 3.75 (3H, s), 3.67 (t, 1H, J = 8 Hz), 2.88 (d, 1H,

J = 12.6 Hz), 2.48 (dd, 1H, J = 6.6 Hz, 18 Hz), 0.77 (3H, s). ¹³C NMR

(CDCl₃) δ 210.1, 169.7, 81.9, 62.1, 52.4, 49.7, 48.7, 47.5, 45.7, 43.6,

43.3, 41.6, 36.5, 30.5, 30.2, 29.9, 27.1, 23.3, 11.4. Anal. (C₁₉H₂₈O₄) C,

H.

(5α,17β)-17-Hydroxy-A-norestran-2-one (52). LiCl (1.95 g, 46.1

mmol) was added to a solution of steroid 51 (3.7 g, 11.5 mmol)

dissolved in DMF (50 mL) and water (0.5 mL), and the reaction was

heated at 160 °C for 35 min. Then crushed ice and water were added

and the product was extracted into CH₂Cl₂(2 × 120 mL). The

combined extracts were washed with water and dried. After solvent

evaporation, the residue was puriﬁed by ﬂash column chromatography

(silica gel eluted with 10% EtOAc in hexanes) to give product 52 (2.6

g, 85%) as a white solid; mp 177−179 °C (EtOAc/hexanes). IR ν_max

1

3450, 2915, 2861, 1732 cm⁻¹. H NMR (CDCl₃) δ 0.77 (3H, s),

(5α)-Estrane-3,17-dione (47). This compound was prepared as

previously described.²⁰

2.41−2.31 (2H, m), 3.66 (t, 1H, J = 8 Hz). ¹³C NMR (CDCl₃) δ

218.3, 82.0, 49.9, 48.9, 48.6, 46.1, 44.2, 43.9, 43.6, 41.7, 36.6, 30.9,

30.64, 30.61, 27.1, 23.4, 11.4. Anal. (C₁₇H₂₆O₂) C, H.

(5α)-17-Oxo-3,4-secoestrane-2,3-dicarboxylic Acid (48) and (5α)-

17-Oxo-3,4-secoestrane-2,3-dicarboxylic Acid Dimethyl Ester (49).

A solution of CrO₃(10.9 g), H₂SO₄(98%, 16 mL), and water (76 mL)

was added dropwise to a stirred solution of compound 47 (7.9 g, 28.5

mmol) in AcOH (80 mL) at 65 °C. The reaction was heated at 70 °C

for 1 h. Then crushed ice and water (500 mL) were added and the

mixture was stirred at room temperature for 1 h. Steroid 48

(5α,17β)-17-[[(4-Methylphenyl)sulfonyl]oxy]-A-norestran-2-one

(53). A solution of steroid 52 (2.46 g, 9.37 mmol), DMAP (57 mg,

0.49 mmol), and p-TsCl (6.25 g, 32.8 mmol) in anhydrous pyridine

(50 mL) was heated at 65 °C overnight. The reaction mixture was

poured into ice−water and the product extracted into CH₂Cl₂(2 × 75

187

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

mL). The combined extracts were washed with aqueous HCl, aqueous

NaHCO₃, brine, and dried. After solvent evaporation, the oily residue

was puriﬁed by ﬂash column chromatography (silica gel eluted with

10% EtOAc in hexanes) to give steroid 53 (3.56 g, 91%) as a white

solid; mp 138−139 °C (ether/hexanes). IR ν_max2920, 2858, 1741,

IR ν_max2932, 2854, 1740, 1662, 1612 cm⁻¹. ¹H NMR (CDCl₃) δ 5.84

(1H, s). ¹³C NMR (CDCl₃) δ 217.4, 199.8, 165.8, 124.9, 48.7, 48.6,

46.5, 45.8, 43.6, 43.2, 42.6, 36.6, 35.3, 31.4, 30.7, 30.3, 26.5. Anal.

(C₁₇H₂₂O₂) C, H.

(5β,8α,9β,10α,13α,14β)-Gonane-3,16-dione (ent-58). Anhydrous

NH₃(10 mL) was condensed using a Dewar condenser into a three-

neck ﬂask containing Li metal (20 mg, 2.9 mmol) at −78 °C. Then

anhydrous THF (12 mL) was added and the resulting blue solution

was stirred for 0.5 h. A solution of compound ent-57 (150 mg, 0.58

mmol) in dry THF (6 mL) was added dropwise to the vigorously

stirred solution. After 2 h of stirring, the reaction color was discharged

by careful addition of solid NH₄Cl in portions and left overnight while

the NH₃evaporated. The reaction was then acidiﬁed with aqueous 1 N

HCl, and the product was extracted into EtOAc (2 × 50 mL). The

combined organic phases were washed with brine, dried, and the

solvent evaporated. The residue was dissolved in acetone (50 mL), and

Jones reagent was added dropwise until an orange color persisted. The

course of the reaction was checked by TLC. Then 2-propanol was

added dropwise until the reaction mixture turned green. After 30 min,

the reaction mixture was poured into water−ice. The product was

extracted into CH₂Cl₂(2 × 30 mL). The combined extracts were

washed with brine and dried, and the solvent was evaporated. Flash

column chromatography (silica gel eluted with 15% EtOAc in

hexanes) gave compound ent-58 (95 mg, 63%) as a white solid; mp

188−189 °C (EtOAc/hexanes). IR ν_max2914, 2867, 1740, 1705 cm⁻¹.

¹H NMR (CDCl₃) δ 0.90−2.42 (24H, m). ¹³C NMR (CDCl₃) δ

1

1598, 1357, 1188, 1176 cm⁻¹. H NMR (CDCl₃) δ 7.80−7.77 (2H,

m), 7.34−7.31 (2H, m), 4.28 (t, 1H, J = 8 Hz), 2.45 (3H, s), 2.33 (dd,

2H, J = 6.6 Hz, 18 Hz), 0.82 (3H, s). ¹³C NMR (CDCl₃) δ 217.9,

144.6, 134.6, 129.8 (2 × C), 128.0 (2 × C), 90.0, 48.9, 48.6, 48.4, 45.9,

44.1, 43.8, 43.6, 41.2, 36.0, 30.8, 30.4, 27.7, 26.8, 23.3, 21.8, 12.0. Anal.

(C₂₄H₃₂O₄S) C, H.

(5α,17β)-2-Methylene-17-[[4-(methylphenyl)sulfonyl]oxy]-A-

norestrane(54). A solution of n-BuLi (2.5 M, 24.1 mmol, 9.6 mL) was

added dropwise to a solution of MeP(Ph)₃Br (8.85 g, 24.7 mmol) in

dry benzene (70 mL) and dry THF (12 mL) under N₂, and the

mixture was stirred for 30 min. A solution of steroid 53 (2.58 g, 6.19

mmol) in dry benzene (35 mL) was added and stirred at room

temperature. After 6 h, the reaction was diluted with water (50 mL)

and extracted with EtOAc (2 × 75 mL). The combined extracts were

washed with brine and dried and the solvents evaporated. Flash

column chromatography on (silica gel eluted with 5% EtOAc in

hexanes) gave steroid 54 (2.05 g, 80%) as a white solid; mp 136−137

°C (EtOAc/hexanes). IR ν_max2920, 2855, 1657, 1598, 1359 cm⁻¹. ¹H

NMR (CDCl₃) δ 7.80−7.77 (2H, m), 7.34−7.31 (2H, m), 4.83 (2H,

m), 4.27 (t, 1H, J = 8 Hz), 2.45−2.36 (5H, m), 0.81 (3H, s). ¹³C NMR

(CDCl₃) δ 151.5, 144.5, 134.6, 129.8 (2 × C), 128.0 (2 × C), 105.9,

90.3, 51.0, 49.0, 48.9, 46.7, 43.7, 41.3, 39.6, 37.6, 36.2, 31.0, 30.8, 27.8,

26.7, 23.5, 21.8, 12.1. Anal. (C₂₅H₃₁O₃S) C, H.

218.1, 211.6, 49.1, 48.7, 47.3, 46.8, 46.0, 45.9, 43.7, 43.5, 43.4, 41.4,

33.8, 31.1, 31.0, 30.4, 30.0.

(5α)-17-Methyl-2-methylene-A-norgon-13(17)-ene (55). Steroid

54 (2 g, 4.8 mmol) was dissolved in anhydrous toluene (50 mL)

and heated to 100 °C. MeMgBr (3.0 M in Et₂O, 8 mL, 24 mmol) was

added dropwise to the stirring hot solution under an N₂atmosphere,

and a white precipitate appeared. The reaction was heated for 1 h at

115 °C. The ﬂask was cooled, a few pieces of crushed ice were added,

and the pH of the solution was adjusted to pH 2 by dropwise addition

of 2 N aqueous H₂SO₄. The toluene layer was separated, and the

aqueous layer was extracted with EtOAc (2 × 100 mL). The combined

extracts were washed with brine and dried and the solvents evaporated.

Flash column chromatography on (silica gel eluted with 1% EtOAc in

hexanes) gave steroid 55 (1.04 g, 90%) as an oily product. IR ν_max

[³⁵S]-TBPS Binding Methods. The methods used were as

described previously.⁴⁶

Xenopus Oocyte and HEK 293 Cell Electrophysiological

Methods. Receptor expression and whole-cell and single-channel

recordings were carried out as described previously.^27,46,47

Tadpole Behavioral Methods. The methods used were as

described previously.⁴⁶

Mouse Behavioral Methods. The methods used were as

described previously.⁴⁸

ASSOCIATED CONTENT

1

3067, 2921, 2851, 1657, 1441 cm⁻¹. H NMR (CDCl₃) δ 4.86 (2H,

■

m), 1.62 (3H, s). ¹³C NMR (CDCl₃) δ 151.9, 136.7, 128.3, 105.7,

100.2, 52.6, 51.5, 51.3, 47.1, 46.4, 39.7, 37.8, 37.4, 31.5, 31.2, 28.4,

25.6, 13.6.

S

* Supporting Information

Table of elemental analysis results, additional views of

superimposed analogues, and structures in Table 1 with

[³⁵S]TBPS IC₅₀values. This material is available free of charge

via the Internet at http://pubs.acs.org.

(3aS,5aR,6S,9aR,9bS)-Decahydro-6-(3-oxobutyl)-1H-benz[e]-

indene-2,7-dione (56). A solution of steroid 55 (760 mg, 3.13 mmol)

in CH₂Cl₂(55 mL) was treated with O₃at −78 °C until a blue color

persisted (ca. 1 h). O₂was passed through the solution for 30 min

until the blue color disappeared. AcOH (50 mL) was added, and the

CH₂Cl₂was evaporated under vacuum without heating. Then AcOH

(10 mL) and Zn dust (2.03 g, 31 mmol) were added, and the reaction

mixture was stirred at room temperature for 1.5 h. Zn dust was ﬁltered

oﬀ through cotton, washing with CH₂Cl₂(100 mL). The Zn dust was

stirred with EtOAc (50 mL) for 1 h. The solids were ﬁltered oﬀ, the

combined solvents were evaporated, and the product was puriﬁed by

ﬂash column chromatography (silica gel eluted with 20% EtOAc in

hexanes) to give compound 56 (390 mg, 51%) as an oil. IR ν_max3406,

AUTHOR INFORMATION

■

Corresponding Author

*Phone: 314-362-1726. Fax: 314-362-7058. E-mail: dcovey@

wustl.edu.

Present Address

^▽Department of Neuroprotectives; Institute of Organic

Chemistry and Biochemistry, Academy of Sciences of Czech

Republic, v.v.i., Flemingovo nam. 2, 166 10 Prague 6 Dejvice,

Czech Republic.

1

2919, 2859, 1742, 1709 cm⁻¹. H NMR (CDCl₃) δ 2.09 (3H, s). ¹³C

NMR (CDCl₃) δ 217.0, 211.6, 209.1, 53.9, 48.4, 47.4, 46.5, 45.6, 43.6,

43.2, 41.6, 40.9, 32.0, 31.1, 30.5, 30.0, 19.5.

Notes

(8α,9β,10α,13α,14β)-Gon-4-ene-3,16-dione (ent-57). A solution

of NaOH [10% w/v in MeOH/water (9:1), 2 mL] was added to a

solution of compound 56 (392 g, 1.41 mmol) in MeOH (20 mL), and

the mixture was stirred at room temperature. After 3 h, a few pieces of

crushed ice were added, the pH was adjusted to pH 2 by adding

aqueous 1 N HCl, and the product was extracted into CH₂Cl₂(2 × 70

mL). The combined extracts were washed with brine, dried, and the

solvent evaporated. Puriﬁcation by ﬂash column chromatography

(silica gel eluted with 20% EtOAc in hexanes) gave compound ent-57

(302 mg, 82%) as a white solid; mp 147−148 °C (EtOAc/hexanes).

Washington University receives income and equity based on a

license of this and related technology to Sage Therapeutics, Inc.

D.F.C, A.S.E. and C.F.Z have equity holdings in Sage

Therapeutics, Inc. C.F.Z. serves on the Scientiﬁc Advisory

Board of Sage Therapeutics, Inc. Sage Therapeutics, Inc. did

not support this research or have any other role in the research

or its conclusions.

The authors declare no competing ﬁnancial interest.

188

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

analogue with potent anesthetic activity. J. Med. Chem. 2011, 54,

3926−3934.

ACKNOWLEDGMENTS

■

This research was supported in part by NIH grant GM47969

(D.F.C., A.S.E., C.F.Z.) and the Bantly Foundation (C.F.Z.).

(16) Akk, G.; Bracamontes, J. R.; Covey, D. F.; Evers, A.; Dao, T.;

Steinbach, J. H. Neuroactive steroids have multiple actions to

potentiate GABA_Areceptors. J. Physiol. 2004, 558, 59−74.

(17) Evers, A. S.; Chen, Z. W.; Manion, B. D.; Han, M.; Jiang, X.;

Darbandi-Tonkabon, R.; Kable, T.; Bracamontes, J.; Zorumski, C. F.;

Mennerick, S.; Steinbach, J. H.; Covey, D. F. A synthetic 18-norsteroid

distinguishes between two neuroactive steroid binding sites on GABA_A

receptors. J. Pharmacol. Exp. Ther. 2010, 333, 404−413.

(18) Wang, C.; Rath, N. P.; Covey, D. F. Neurosteroid Analogues.

13. Synthetic methods for the preparation of 2β-hydroxygonane

derivatives as structural mimics of ent-3alpha-hydroxysteroid modu-

lators of GABA(A) receptors. Tetrahedron 2007, 63, 7977−7984.

(19) Teerhuis, M. H.; Huisman, I. A. M.; Groen, M. B. Synthesis of

ent-nortestosterone from its naturally occurring antipode. Tetrahedron

Lett. 2001, 42, 2869−2871.

ABBREVIATIONS USED

■

CNS, central nervous system; GABA, γ-aminobutyric acid;

GABA_A, γ-aminobutyric acid type A; [³⁵S]TBPS, [³⁵S]-tert-

butylbicyclophosphorothionate; SAR, structure−activity rela-

tionships; LRR, loss of righting reﬂex; LSR, loss of swimming

reﬂex

REFERENCES

■

(1) Zorumski, C. F.; Mennerick, S. Neurosteroids as therapeutic leads

in psychiatry. JAMA Psychiatry 2013, 70, 659−660.

(2) Brinton, R. D. Neurosteroids as regenerative agents in the brain:

therapeutic implications. Nature Rev. Endocrinol. 2013, 9, 241−250.

(3) Reddy, D. S. Neurosteroids: endogenous role in the human brain

and therapeutic potentials. Prog. Brain Res. 2010, 186, 113−137.

(4) Belelli, D.; Lambert, J. J. Neurosteroids: endogenous regulators of

the GABA_Areceptor. Nature Rev. Neurosci. 2005, 6, 565−575.

(5) Veleiro, A. S.; Burton, G. Structure−activity relationships of

neuroactive steroids acting on the GABA_Areceptor. Curr. Med. Chem.

2009, 16, 455−472.

(6) Wittmer, L. L.; Hu, Y.; Kalkbrenner, M.; Evers, A. S.; Zorumski,

C. F.; Covey, D. F. Enantioselectivity of steroid-induced γ-amino-

butyric acid_Areceptor modulation and anesthesia. Mol. Pharmacol.

1996, 50, 1581−1586.

(7) Nilsson, K. R.; Zorumski, C. F.; Covey, D. F. Neurosteroid

analogues. 6. The synthesis and GABA_Areceptor pharmacology of

enantiomers of dehydroepiandrosterone sulfate, pregnenolone sulfate,

and (3α,5β)-3-hydroxypregnan-20-one sulfate. J. Med. Chem. 1998, 41,

2604−2613.

(8) Covey, D. F.; Nathan, D.; Kalkbrenner, M.; Nilsson, K. R.; Hu,

Y.; Zorumski, C. F.; Evers, A. S. Enantioselectivity of pregnanolone-

induced γ-aminobutyric acid_Areceptor modulation and anesthesia. J.

Pharmacol. Exp. Ther. 2000, 293, 1009−1016.

(9) Katona, B. W.; Krishnan, K.; Cai, Z. Y.; Manion, B. D.; Benz, A.;

Taylor, A.; Evers, A. S.; Zorumski, C. F.; Mennerick, S.; Covey, D. F.

Neurosteroid analogues. 12. Potent enhancement of GABA-mediated

chloride currents at GABA_Areceptors by ent-androgens. Eur. J. Med.

Chem. 2008, 43, 107−113.

(10) Li, P.; Bracamontes, J.; Katona, B. W.; Covey, D. F.; Steinbach,

J. H.; Akk, G. Natural and enantiomeric etiocholanolone interact with

distinct sites on the rat α1β2γ2L GABA_Areceptor. Mol. Pharmacol.

2007, 71, 1582−1590.

(11) Turner, D. M.; Ransom, R. W.; Yang, J. S.; Olsen, R. W. Steroid

anesthetics and naturally occurring analogs modulate the γ-amino-

butyric acid receptor complex at a site distinct from barbiturates. J.

Pharmacol. Exp. Ther. 1989, 248, 960−966.

(12) Hawkinson, J. E.; Kimbrough, C. L.; Belelli, D.; Lambert, J. J.;

Purdy, R. H.; Lan, N. C. Correlation of neuroactive steroid modulation

of [³⁵S]t-butylbicyclophosphorothionate and [³H]flunitrazepam bind-

ing and γ-aminobutyric acid_Areceptor function. Mol. Pharmacol. 1994,

46, 977−985.

(20) Stastna, E.; Rath, N. P.; Covey, D. F. The use of symmetry in

enantioselective synthesis: four pairs of chrysene enantiomers prepared

from 19-nortestosterone. Org. Biomol. Chem. 2011, 9, 4685−4694.

(21) Nace, H. R.; Pyle, J. L. Nor steroids. IX. Synthesis of A-

norandrostanes via the Dieckmann cyclization. J. Org. Chem. 1971, 36,

81−83.

(22) Nace, H. R.; Smith, A. H. Nor steroids. X. Synthesis of A-nor

steroids via the Dieckmann condensation. J. Org. Chem. 1973, 38,

1941−1944.

(23) Kontonassios, D.; Sandris, C. A-nor steroids. 1. The

stereochemistry of a 2-carbomethoxy-A-nor-cholestan-3-one obtained

via the Dieckmann condensation. Steroids 1982, 39, 411−417.

(24) Han, M.; Zorumski, C. F.; Covey, D. F. Neurosteroid analogues.

4. The effect of methyl substitution at the C-5 and C-10 positions of

neurosteroids on electrophysiological activity at GABA_Areceptors. J.

Med. Chem. 1996, 39, 4218−32.

(25) Akk, G.; Li, P.; Bracamontes, J.; Reichert, D. E.; Covey, D. F.;

Steinbach, J. H. Mutations of the GABA-A receptor α1 subunit M1

domain reveal unexpected complexity for modulation by neuroactive

steroids. Mol. Pharmacol. 2008, 74, 614−627.

(26) Hosie, A. M.; Wilkins, M. E.; da Silva, H. M. A.; Smart, T. G.

Endogenous neurosteroids regulate GABA_Areceptors through two

discrete transmembrane sites. Nature 2006, 444, 486−489.

(27) Steinbach, J. H.; Akk, G. Modulation of GABA_Areceptor

channel gating by pentobarbital. J. Physiol. 2001, 537, 715−733.

(28) Akk, G.; Shu, H. J.; Wang, C.; Steinbach, J. H.; Zorumski, C. F.;

Covey, D. F.; Mennerick, S. Neurosteroid access to the GABA_A

receptor. J. Neurosci. 2005, 25, 11605−11613.

(29) Scaglione, J. B.; Jastrzebska, I.; Krishnan, K.; Li, P.; Akk, G.;

Manion, B. D.; Benz, A.; Taylor, A.; Rath, N. P.; Evers, A. S.;

Zorumski, C. F.; Mennerick, S.; Covey, D. F. Neurosteroid analogues.

14. Alternative ring system scaffolds: GABA modulatory and anesthetic

actions of cyclopenta[b]phenanthrenes and cyclopenta[b]anthracenes.

J. Med. Chem. 2008, 51, 1309−1318.

(30) Akk, G.; Covey, D. F.; Evers, A. S.; Mennerick, S.; Zorumski, C.

F.; Steinbach, J. H. Kinetic and structural determinants for GABA-A

receptor potentiation by neuroactive steroids. Curr. Neuropharmacol.

2010, 8, 18−25.

(31) Chisari, M.; Eisenman, L. N.; Covey, D. F.; Mennerick, S.;

Zorumski, C. F. The sticky issue of neurosteroids and GABA_A

receptors. Trends Neurosci. 2010, 33, 299−306.

(32) Chisari, M.; Eisenman, L. N.; Krishnan, K.; Bandyopadhyaya, A.

K.; Wang, C.; Taylor, A.; Benz, A.; Covey, D. F.; Zorumski, C. F.;

Mennerick, S. The Influence of Neuroactive Steroid Lipophilicity on

GABA_AReceptor Modulation: Evidence for a Low-Affinity Interaction.

J. Neurophysiol. 2009, 102, 1254−1264.

(33) Zorumski, C. F.; Paul, S. M.; Izumi, Y.; Covey, D. F.; Mennerick,

S. Neurosteroids, stress and depression: potential therapeutic

opportunities. Neurosci. Biobehav. Rev. 2013, 37, 109−122.

(34) Biellmann, J.-F. Enantiomeric steroids: synthesis, physical, and

biological properties. Chem. Rev. 2003, 103, 2019−2033.

(13) Krishnan, K.; Manion, B. D.; Taylor, A.; Bracamontes, J.;

Steinbach, J. H.; Reichert, D. E.; Evers, A. S.; Zorumski, C. F.;

Mennerick, S.; Covey, D. F. Neurosteroid analogues. 17. Inverted

binding orientations of androsterone enantiomers at the steroid

potentiation site on γ-aminobutyric acid type A receptors. J. Med.

Chem. 2012, 55, 1334−1345.

(14) Phillipps, G. H. Structure−activity Relationships in Steroidal

Anaesthetics; Churchill Livingstone: New York, 1974; p 32−47.

(15) Stastna, E.; Krishnan, K.; Manion, B. D.; Taylor, A.; Rath, N. P.;

Chen, Z. W.; Evers, A. S.; Zorumski, C. F.; Mennerick, S.; Covey, D. F.

Neurosteroid analogues. 16. A new explanation for the lack of

anesthetic effects of Δ¹⁶-alphaxalone and identification of a Δ¹⁷⁽²⁰⁾

189

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Journal of Medicinal Chemistry

Article

(35) Katona, B. W.; Cummins, C. L.; Ferguson, A. D.; Li, T.;

Schmidt, D. R.; Mangelsdorf, D. J.; Covey, D. F. Synthesis,

characterization, and receptor interaction profiles of enantiomeric

bile acids. J. Med. Chem. 2007, 50, 6048−58.

(36) Vanlandingham, J. W.; Cutler, S. M.; Virmani, S.; Hoffman, S.

W.; Covey, D. F.; Krishnan, K.; Hammes, S. R.; Jamnongjit, M.; Stein,

D. G. The enantiomer of progesterone acts as a molecular

neuroprotectant after traumatic brain injury. Neuropharmacology

2006, 51, 1078−1085.

(37) Green, P. S.; Yang, S. H.; Nilsson, K. R.; Kumar, A. S.; Covey, D.

F.; Simpkins, J. W. The nonfeminizing enantiomer of 17β-estradiol

exerts protective effects in neuronal cultures and a rat model of

cerebral ischemia. Endocrinology 2001, 142, 400−406.

(38) Sneitz, N.; Krishnan, K.; Covey, D. F.; Finel, M.

Glucuronidation of the steroid enantiomers ent-17β-estradiol, ent-

androsterone and ent-etiocholanolone by the human UDP-glucur-

onosyltransferases. J. Steroid Biochem. Mol. Biol. 2011, 127, 282−288.

(39) Fajkos, J.; Sorm, F. Steroids. IX. Synthesis of 3α-hydroxy-16-

oxoandrostane and 3,16-dioxo-4-androstane. Collect. Czech. Chem.

Commun. 1954, 19, 349−356.

(40) Kamijo, S.; Matsumura, S.; Inoue, M. CCl₃CN: a crucial

promoter of mCPBA-mediated direct ether oxidation. Org. Lett. 2010,

12, 4195−4197.

(41) Gardi, R.; Vitali, R.; Ercoli, A. 1. Ethers of 3α-hydroxy-5α-

androstane derivatives. Gazz. Chim. Ital. 1962, 92, 632−646.

(42) McKenna, J.; Norymberski, J. K.; Stubbe, R. D. Partial reduction

of steroid hormones and related substances. Part III. The reaction of

αβ-unsaturated ketones with zinc in acetic acid. J. Chem. Soc. 1959,

2502−2509.

(43) da Silva, E. J.; Roleira, F. M.; Sa e Melo, M. L.; Neves, A. S.;

Paixao, J. A.; de Almeida, M. J.; Silva, M. R.; Andrade, L. C. X-Ray and

deuterium labeling studies on the abnormal ring cleavages of a 5β-

epoxide precursor of formestane. Steroids 2002, 67, 311−319.

(44) Hu, Y.; Zorumski, C. F.; Covey, D. F. Neurosteroid analogues:

structure−activity studies of benz[e]indene modulators of GABA_A

receptor function. 1. The effect of 6-methyl substitution on the

electrophysiological activity of 7-substituted benz[e]indene-3-carbon-

itriles. J. Med. Chem. 1993, 36, 3956−3967.

(45) Hu, Y. F.; Wittmer, L. L.; Kalkbrenner, M.; Evers, A. S.;

Zorumski, C. F.; Covey, D. F. Neurosteroid analogues. Part 5.

Enantiomers of neuroactive steroids and benz[e]indenes: total

synthesis, electrophysiological effects on GABA_Areceptor function

and anesthetic actions in tadpoles. J. Chem. Soc., Perkin Trans. 1 1997,

3665−3671.

(46) Jiang, X.; Manion, B. D.; Benz, A.; Rath, N. P.; Evers, A. S.;

Zorumski, C. F.; Mennerick, S.; Covey, D. F. Neurosteroid analogues.

9. Conformationally constrained pregnanes: structure−activity studies

of 13,24-cyclo-18,21-dinorcholane analogues of the GABA modulatory

and anesthetic steroids (3α,5α)- and (3α,5β)-3-hydroxypregnan-20-

one. J. Med. Chem. 2003, 46, 5334−5348.

(47) Li, P.; Covey, D. F.; Steinbach, J. H.; Akk, G. Dual potentiating

and inhibitory actions of a benz[e]indene neurosteroid analog on

recombinant α1β2γ2 GABA_Areceptors. Mol. Pharmacol. 2006, 69,

2015−2026.

(48) Bandyopadhyaya, A. K.; Manion, B. D.; Benz, A.; Taylor, A.;

Rath, N. P.; Evers, A. S.; Zorumski, C. F.; Mennerick, S.; Covey, D. F.

Neurosteroid analogues. 15. A comparative study of the anesthetic and

GABAergic actions of alphaxalone, Δ¹⁶-alphaxalone and their

corresponding 17-carbonitrile analogues. Bioorg. Med. Chem. Lett.

2010, 20, 6680−6684.

(49) Bracamontes, J. R.; Li, P.; Akk, G.; Steinbach, J. H. A

neurosteroid potentiation site can be moved among GABA_Areceptor

subunits. J. Physiol. 2012, 590, 5739−47.

190

dx.doi.org/10.1021/jm401577c | J. Med. Chem. 2014, 57, 171−190

Article Doi

Neurosteroid analogues. 18. Structure-activity studies of ent-steroid potentiators of γ-aminobutyric acid type a receptors and comparison of their activities with those of alphaxalone and allopregnanolone

DOI: 10.1021/jm401577c

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm401577c

Full text of DOI:10.1021/jm401577c

Products guided by the article

R&D Labs maybe for 13094-19-4

Relevant to this article

Hot Product