roughly 30% more efficacious than tamoxifen. Estrogenic activity
for compounds 23 and 24 was reduced compared to tamoxifen and
they demonstrated partial agonist activity compared to tamoxifen,
which demonstrated antagonist activity.
curves are provided in the Supplementary Material. All four
analogs demonstrated superior potency and percent maximal
activation of GPN-stimulated calcium release compared to
tamoxifen. And while the EC50 values for activation of GPN-
stimulated calcium release were slightly lower in WT cells
compared to GD-patient cells, all four compounds showed
significantly greater efficacy (maximal response) in GD-patient
cells. Compounds 28 and 31 in particular demonstrated high
maximal activation in GD-patient cells (227% and 240%,
respectively) in comparison to the maximal activation in WT cells
(56% and 50%, respectively).
We then moved to test our hypothesis regarding the preference
of a non-conjugated distal phenyl ring in the diphenylethanone
scaffold by introducing 2-substitutents in this ring (compounds 25
– 31, Table 2).
As previously discussed, the n-butyl derivative 21 showed no
activity in the calcium release assay in GD patient-derived cells.
However, introduction of a 2-substituent into that scaffold resulted
in analogs that showed activity in that assay that was similar to
(compounds 25 and 26) or superior to (compounds 27 and 28) that
seen with tamoxifen. Greater potency was realized when the
planarity of the substituent was eliminated. This is seen by
comparing compound 25 to 26 (EC50 = 2.3 M and 0.65 M,
respectively) and compound 27 to 28 (EC50 = 2.2 M and 1.4 M,
respectively). A significant increase in efficacy was seen when
the isopropyl group was introduced into the 2-position of the distal
phenyl ring (compound 28, Emax = 401% of the basal response to
GPN). The presence of the branched alkyl group also had a
beneficial effect on estrogenic activity. The ethenyl derivative 25
showed significant estrogen receptor activity at a concentration of
10 M (100% of the EC100 response seen with -estradiol) whereas
the propenyl (27) and i-propyl (28) analogs were less potent in the
estrogen receptor assay (35% and 45%, respectively).
Interestingly, the ethyl derivative 26 showed similar potency in the
estrogen receptor assay but was found to be an antagonist rather
than an agonist like 25, 27 and 28. The reasons for this switch in
intrinsic activity are not clear, although it should be noted that the
ethyl group is present in a comparable position in the
tamoxifen/idoxifene scaffold so it is possible that the ethyl group
is a privileged substituent in this position.
To differentiate compounds 27, 28, 29 and 31 and prioritize
one to advance to in vivo studies, we examined their in vitro drug-
like properties in two assays, namely maximum aqueous solubility
in 2% DMSO/phosphate buffered saline and stability in mouse
liver microsomes in the presence of NADPH (an indicator of
liability to oxidative metabolism). The results are summarized in
Table 4.
Table 4. Drug-like properties of selected diphenylethanones.
Cd. #
Max. Aq.
Microsomal Stability (t1/2, min.)
Solubility (µM)
Mouse
9.1
Human
24.4
27
28
29
31
1
45.2
49.4
40.0
31.7
6.9*
7.9
21.6
22.3
7.7
23.7
26.7
8.9
> 60
* Value reported in reference 23.
All four phenylethanone derivatives showed superior aqueous
solubility compared to tamoxifen. All four phenylethanone
derivatives also showed moderate stability in mouse and human
liver microsomes. Compound 29 was somewhat more stable in
mouse liver microsomes than the other three. However, compound
31 stood out from the others because of its lack of any estrogenic
activity at 10 M (Table 2). For this reason we selected compound
31 for advancement.
Similar results were seen when the i-butyl group was
substituted for the n-butyl moiety. Compound 29, which
combined the i-butyl substitution on the ethanone chain and the i-
propyl group in the 2-position of the distal phenyl ring,
demonstrated reasonable potency, superior efficacy and reduced
estrogenic activity compared to tamoxifen. Moving the i-propyl
group to the 3-position of the distal phenyl ring (compound 30)
resulted in a compound with significantly reduced efficacy (80%
compared to 406% for compound 29). Constraining the i-propyl
group to give the cyclopropyl analog 31 resulted in a molecule that
showed similar potency and enhanced efficacy in the calcium
release assay compared to tamoxifen and was devoid of estrogenic
activity at a concentration of 10 M.
In conclusion, a phenotypic screen identified tamoxifen as a hit
with modest functional potency to reverse the lysosomal calcium
deficits in GD patient-derived cells. SAR aimed at reducing the
planar conformation of tamoxifen led to the identification of a
novel series of phenylethanone derivatives, ultimately resulting in
the selection of compound 31 for further evaluation. In the current
studies compound 31 was tested as a racemic mixture. Despite the
realatively high pKa predicted for the enolic proton of 31
(calculated pKa24 for the proton adjacent to the carbonyl = 15.66),
data demonstrate that drugs containing enolic hydrogens can
undergo some racemization at physiological pH, especially in the
presence of phosphate buffers.25,26 It will be important to
understand the role of chirality on in vitro and in vivo activity in
this series of compounds and understand the propensity for them
to undergo partial or full racemization at physiological pH. These
and other studies are currently in progress and will be the subject
of future publications.
Table 3. Activity of selected compounds in lysosomal calcium
assay.
Lysosomal Ca+2 Assay
Cd #
GD-patient cells
EC50, M Emax, %
WT cells
EC50, M
Emax, %
27
28
29
31
1
2.2
1.48
1.8
3.1
10
257%
227%
449%
240%
148%
1.1
0.7
1.1
1.1
2.5
109%
56%
148%
50%
28%
Declaration of Competing Interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
Activity for the four most potent compounds (27, 28, 29 and
31) and tamoxifen in the lysosomal calcium assay in GD patient-
derived cells and wild-type cells are summarized in Table 3. Full