Synthesis and evaluation of third generation vitamin D3 analogues as inhibitors of Hedgehog signaling

Article abstract of DOI:10.1016/j.ejmech.2018.11.028

The Hedgehog (Hh) pathway is a developmental pathway with therapeutic potential as a target for a variety of cancers. In recent years, several vitamin D-based compounds have been identified as potent inhibitors of Hh signaling. These analogues contain aromatic phenol A-ring mimics coupled to the CD-ring side chain of vitamin D3 through modified seco-B regions. To continue structure-activity relationship studies on this class of Hh pathway inhibitors, multiple series of vitamin D-based analogues that contain an amine-based seco-B tether and/or incorporate a hydroxyl moiety on C-25 were designed and synthesized. These compounds were evaluated in multiple cell lines for their anti-Hh activity, and we identify analogues 16, 21, 22 as potent vitamin D-based Hh inhibitors (IC₅₀ values of 110–340 nM). We also performed a series of mechanism of action studies in knockout cell lines to further explore whether these analogues inhibit the Hh pathway through a known Hh pathway component or the vitamin D receptor. While the specific cellular target that mediates these effects remains elusive, our studies suggest multiple cellular targets may mediate the anti-Hh activity of this scaffold.

Full text of DOI:10.1016/j.ejmech.2018.11.028

Accepted Manuscript
Synthesis and evaluation of third generation vitamin D3 analogues as inhibitors of
Hedgehog signaling
Chad A. Maschinot, Lianne Q. Chau, Robert J. Wechsler-Reya, M. Kyle Hadden
PII:
S0223-5234(18)30987-5
DOI:
https://doi.org/10.1016/j.ejmech.2018.11.028
EJMECH 10885
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 13 September 2018
Revised Date: 9 November 2018
Accepted Date: 9 November 2018
Please cite this article as: C.A. Maschinot, L.Q. Chau, R.J. Wechsler-Reya, M.K. Hadden, Synthesis
and evaluation of third generation vitamin D3 analogues as inhibitors of Hedgehog signaling, European
Journal of Medicinal Chemistry (2018), doi: https://doi.org/10.1016/j.ejmech.2018.11.028.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Synthesis and Evaluation of Third Generation
Vitamin D3 Analogues as Inhibitors of Hedgehog
Signaling
Chad A. Maschinot1†, Lianne Q. Chau2, Robert J. Wechsler-Reya2, M. Kyle Hadden1,*
1 Department of Pharmaceutical Sciences, University of Connecticut, 69 North Eagleville Road,
Unit 3092, Storrs, CT 06269, United States
2 Tumor Initiation and Maintenance Program, NCI-Designated Cancer Center, Sanford Burnham
Prebys Medical Discovery Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037,
United States
KEYWORDS: Hedgehog; Vitamin D; Basal Cell Carcinoma; Medulloblastoma; Gli.
1

ACCEPTED MANUSCRIPT
ABSTRACT
The Hedgehog (Hh) pathway is a developmental pathway with therapeutic potential as a target
for a variety of cancers. In recent years, several vitamin D-based compounds have been
identified as potent inhibitors of Hh signaling. These analogues contain aromatic phenol A-ring
mimics coupled to the CD-ring side chain of vitamin D3 through modified seco-B regions. To
continue structure-activity relationship studies on this class of Hh pathway inhibitors, multiple
series of vitamin D-based analogues that contain an amine-based seco-B tether and/or
incorporate a hydroxyl moiety on C-25 were designed and synthesized. These compounds were
evaluated in multiple cell lines for their anti-Hh activity, and we identify analogues 16, 21, 22 as
potent vitamin D-based Hh inhibitors (IC50 values of 110 - 340 nM). We also performed a series
of mechanism of action studies in knockout cell lines to further explore whether these analogues
inhibit the Hh pathway through a known Hh pathway component or the vitamin D receptor.
While the specific cellular target that mediates these effects remains elusive, our studies suggest
multiple cellular targets may mediate the anti-Hh activity of this scaffold.
2

ACCEPTED MANUSCRIPT
1. Introduction
Aberrant signaling of the Hedgehog (Hh) pathway has been associated with a variety of disease
states from osteoporosis to several forms of cancer, including basal cell carcinoma (BCC) and
medulloblastoma (MB) [1-3]. The pathway is composed of two transmembrane receptor
proteins, Patched (Ptch) and Smoothend (Smo), as well as multiple cytosolic proteins, including
Suppressor of Fused (SuFu) and the glioma associated oncogene (Gli) family of transcription
factors. Multiple small molecule inhibitors of Hh signaling are under preclinical and clinical
development, and two Smo antagonists, Vismodegib and Sonidegib, have been approved by the
FDA for metastatic BCC [3]. The development of acquired resistance and relapse to Vismodegib
and Sonidegib [4-5] has prompted the continued search for alternative Hh pathway inhibitory
scaffolds. Multiple approaches are currently being investigated that target both canonical (Smo
and Gli) and non-canonical signaling [6].
Fig 1. Vitamin D-based inhibitors of Hedgehog signaling.
Among the scaffolds under investigation for their anti-Hh activity is the vitamin D-based
family of pathway inhibitors (Fig. 1). As naturally occurring metabolites, the canonical vitamin
3

ACCEPTED MANUSCRIPT
D signaling pathway is well studied and is traditionally associated with calcium homeostasis via
the ‘active’ hormone calcitriol (3), which functions through binding to and activation of the
vitamin D receptor (VDR) [7].
Following the initial reports of Hh inhibition by vitamin D3 (VD3, 1) [8-10], we have
characterized several VD3-based small molecules that selectively inhibit Hh signaling [11-14].
Initial structure-activity relationship (SAR) studies for this scaffold focused on 1 as our lead
compound because it showed minimal VDR agonism compared to 2 and 3. The first series of
analogues incorporated an aromatic A-ring moiety coupled to the natural CD-ring of 1 (4)
through an ester linker (5), which greatly increased selectivity and potency compared to 1 [11].
When the seco-B region was investigated further, an increase in potency was observed with the
use of nitrogen containing (amine) tethers (6 – 8) [13]. Additionally, inversion of the C-8
stereocenter from the α- to β-orientation did not show a significant effect on inhibitory activity
[13]. Herein, we report further SAR for this scaffold with a particular focus on expanding our
amine-linked analogues while also incorporating the C-25 hydroxyl moiety present in calcitriol.
In addition, we have utilized several knockout cell lines to provide initial evidence of cross-talk
between VDR and the Hh pathway.
2. Results and discussion
2.1. Analogue design rationale
One of the most notable observations from our previous SAR was that reducing the seco-B linker
region from two atoms (5 and 6) to a single atom (7 and 8) altered the most potent A-ring
hydroxyl position from meta (7) to para (8). A second notable observation was that little
difference in activity was demonstrated between the α- (8) or β-orientations (10) of the linker
4

ACCEPTED MANUSCRIPT
region suggesting there may be no preference for a specific orientation of the C-8 stereoisomer.
With these differences in mind, we envisioned a new series of analogues to further evaluate (1)
whether other observations from our ester analogues [11] were applicable to compounds
containing an amine in the seco-B region (12 – 16) and (2) whether extension of the nitrogen-
containing linker back to two atoms by the addition of a methylene group would affect the
inhibitory activity (18 – 19).
Our initial evaluation of these nitrogen-based linker analogues indicated that the free A-ring
hydroxyl group was required and provided the foundation for a second series of analogues in
which a C-25 hydroxyl group was introduced (21 – 22, 27 – 28). Previous reports from our
research and others [9-14] demonstrated that both 2 and 3 are more potent inhibitors of Hh
signaling compared to 1; therefore, we synthesized a series of analogues that could both increase
potency of the scaffold while also providing potential probes to evaluate crosstalk between VDR
and Hh signaling [15]. Finally, a series of analogues that masked the C-25 hydroxyl as the
methyl ether (25, 31, 34) were synthesized and evaluated to determine whether the increased
anti-Hh activity associated with our initial C-25 hydroxylated compounds could be maintained
while reducing the VDR activation that results from addition of the hydroxyl group.
2.2. Analogue synthesis
Synthesis of the amine linker analogues is centered on the one pot reductive amination of the
desired amine with Grundmann’s ketone (11) or the extended aldehyde 17. Following previously
described protocols [11], 11 is obtained in two steps from VD3 via global ozonolysis and
oxidation of 4 (Scheme 1). Reductive amination of 11 with sodium triacetoxyborohydride in the
presence of the requisite amine yielded corresponding amines 12 – 16. To extend the linker,
5

ACCEPTED MANUSCRIPT
aldehyde 17 was obtained in high yield following hydrolysis of the enol ether obtained through a
Wittig reaction between 11 and methoxymethyl triphenylphosonium chloride. The same
reductive amination procedure was applied to 17 to provide extended amines 18 – 19 (Scheme
1).
a - b
c
1
H
H
12 - 16
O
HN
R
11, Grundmann’s Ketone
OMe
R:
O
d - e
O
15
12
13, 3-OMe
14, 4-OMe
OH
16
c
H
H
HN
18; 3-OH
19; 4-OH
17
H
O
OH
Scheme 1. Synthesis of amine seco-B analogues of vitamin D3. Reagents and Conditions: (a) i.
O3, pyridine, DCM:MeOH (3:1), 0 °C to RT; ii. NaBH4; 92% over two-steps (b) PDC, DCM; 96
% (c) NaBH(OAc)3, AcOH, R-amine, DCE; 20 – 65% (d) PPh3Cl(MeOMe), NatBuO, THF, 0 °C
to RT; 89% (e) 2N HCl, MeOH, 60 °C; 75%.
6

ACCEPTED MANUSCRIPT
Scheme 2. Synthesis of C-25 functionalized single atom nitrogen containing seco-B analogues.
Reagents and conditions: (a) RuCl3, NaIO4, pyridine, MeCN:H2O, 60 °C; 67% (b) NaBH(OAc)3,
AcOH, R-amine, DCE; 15 – 35 % (c) MeI, NaH, DMF, 0 °C; 17 – 25 %.
Introduction of the C-25 hydroxyl was performed using ruthenium (III) chloride and sodium
periodate for the in-situ generation of ruthenium tetroxide (Scheme 2) [16]. Due to steric
hindrance of 11, specifically interactions between C-18 and C-20, selective hydroxylation at the
C-25 position provided 20 in good yield (65%). Hydroxylated amines 21 and 22 were obtained
from 20 through the reductive amination procedure described above. Conversion of the C-25
hydroxyl to the methyl ether (23) was performed using methyl iodide and sodium hydride. Our
desired product was obtained in low yield due to concomitant α-methylation at C-9 (24). Most
likely, α-methylation at C-9 occurs following the desired etherification due to the presence of 23
and the ‘dimethylated’ byproduct 24 (the 25-OH, C-9 α-methyl compound was not observed).
Reductive amination of 23 provided 25 in modest yield.
Reductive amination of aldehyde 26 yielded 27 and 28, C-25 hydroxylated analogues of 18
and 19, respectively (Scheme 3). Since the C-9 methylated byproduct (23) was the major product
observed during the etherification of 20, addition of the methyl ether group to the aldehyde was
performed following its conversion to the methoxy enol ether to afford 29 (Scheme 3). The yield
of 29 was significantly increased through this route compared to our initial preparation of 23.
7

ACCEPTED MANUSCRIPT
Following hydrolysis of 29 to its corresponding aldehyde (30), reductive amination with 3-
aminophenol yielded amine 31.
Scheme 3. C-25 functionalization of two atom amine-linked analogues. Reagents and conditions:
(a) PPh3Cl(MeOMe), NatBuO, THF; 89 % (b) 2N HCl, MeOH, 60 °C; 60 - 75% (c)
NaBH(OAc)3, AcOH, R-amine, DCE; 32 – 59 % (d) MeI, NaH, DMF, 0 °C; 62%.
8

ACCEPTED MANUSCRIPT
Scheme 4. C-25 functionalized ester analogues. Reagents and conditions: (a)TES-Cl, imidazole,
DMF; 94% (b) NaBH4, MeOH:DCM; 88 - 92% (c) 3-OMOM-benzoic acid, DCC, DMAP,
DCM; 52 - 70% (d) CSA, MeOH; 60 - 75% (e) MeI, NaH, DMF, 0 °C; 35%.
Ketone 20 provided the starting material for the synthesis of C-25 hydroxylated ester 33.
Protection of the C-25 hydroxyl group as the triethyl silyl ether (TES) followed by sodium
borohydride reduction gave alcohol 32 (Scheme 4). Triethyl silane provided a useful protecting
group because following esterification with the MOM-protected 3-hydroxybenzoic acid, both
protecting groups could be removed simultaneously with camphor sulfonic acid (CSA) to afford
analogue 33. In a similar fashion, ester 34 was obtained in three steps from ketone 23.
2.3. Biological evaluation
2.3.1. SAR and Preliminary Pharmacokinetic Analysis.
The initial evaluation of VD3 analogues as Hh pathway inhibitors was performed in C3H10T1/2
cells, an Hh-dependent mouse embryonic fibroblast (MEF) cell line. In addition, we previously
reported significant upregulation of the VDR target gene Cyp24A1 following VDR activation in
these MEF cells, allowing for dual observations of potency and selectivity in this cellular model
9

ACCEPTED MANUSCRIPT
[11-14]. Following our previously utilized procedures, the Hh target gene Gli1 was upregulated
in the MEFs via endogenous administration of oxysterols (OHCs; 20(S)-OHC and 22(S)-OHC at
5 µM each), which are well-characterized agonists of Hh signaling. Inhibition of Hh signaling
(Gli1 downregulation) and activation of VDR (upregulation of Cyp24A1) were quantified with
qPCR following co-administration of OHCs and VD3 analogue (5 µM) (Tables 1-2). Analogues
that inhibited Hh signaling at 5 µM were evaluated for their ability to regulate Hh and VDR
signaling in a concentration-dependent fashion in both the MEFs and the Hh-dependent murine
BCC cell line, ASZ001.
Table 1. In Vitro Activity of Vitamin D3 Amine Linked Analogues.
C3H10T1/2
Cyp24A1a,c
ASZ001
Cyp24A1a,c
% Hh Activitya,b
IC50 (µM)d
IC50 (µM)d
DMSO
OHCs
1g
1.0
1.0
--
--
1.0
100
--
--
--
--
35.7 ± 0.3
9.2 ± 0.4e
30 ± 1
8,300 ± 40
>30,000e
>30,000
22 ± 5
4.1 ± 0.3
0.4 ± 0.1
3.0 ± 1.0h
0.7 ± 0.1
0.9 ± 0.1
1.0 ± 0.1
0.4 ± 0.1
3.9 ± 0.8
0.3 ± 0.1
---
1.1 ± 0.1
0.5 ± 0.1
89.1 ± 41.5 h
5.2 ± 0.2
1.8 ± 0.3
1.7 ± 0.1
0.7 ± 0.1
6.3 ± 1.0
0.6 ± 0.2
---
2,200 ± 200e
11,000 ± 400f
>30,000
1.9 ± 0.1
1.8 ± 0.2
1.8 ± 0.1
2.7 ± 0.7
1.6 ± 0.2
3.0 ± 0.8
---
2
3f
5g
1.9 ± 0.5
2.5 ± 0.1
11.6 ± 0.1
0.7 ± 0.2
31 ± 3
6 g
7.0 ± 0.9
17 ± 4
7 g
8 g
45 ± 20
3.0 ± 0.5
21 ± 3
9g
10 g
1.3 ± 0.6
98 ± 10
110 ± 10
73 ± 7
1.2 ± 0.1
1.9 ± 0.5
2.4 ± 0.9
6.1 ± 3.4
12
---
---
---
13
---
---
---
14
110 ± 1
---
---
---
15
10

ACCEPTED MANUSCRIPT
4.6 ± 0.1
13 ± 8
24 ± 1
3.1 ± 0.1
3.8 ± 0.7
0.6 ± 0.1
16
18
19
0.3 ± 0.1
1.2 ± 0.1
0.6 ± 0.1
1.1 ± 0.7
7.8 ± 3.3 e
5.8 ± 0.9 e
13 ± 5
1.9 ± 1.6
8.9 ± 1.8
aAll analogues evaluated at 5 µM unless otherwise indicated. Values represent % Gli1 mRNA expression
relative to OHC control (set to 100%; 20(S)-OHC and 22(S)-OHC at 5 µM each) after 24 h. cValues represent
Cyp24A1 mRNA expression compared to DMSO control (set to 1.0). dValues represent mean ± SEM for at
b
e
f
least two separate experiments performed in triplicate. Compound evaluated at 2.5 µM. Compound
evaluated at 1 µM. gPreviously published results [Ref 13]. hValue expressed in nanomolar (nM).
Similar to the SAR results with our previous ester-linked analogues [11], removal or masking
of the aromatic hydroxyl group (12 – 15) resulted in complete loss of anti-Hh activity (Table 1).
More intriguing results were observed when the linker region was extended two atoms.
Compared to 8, amine 16 demonstrated a moderate loss in anti-Hh activity and a slight increase
in selectivity. When the arrangement of the linker atoms was reversed (18 and 19), the location
of the hydroxyl group became more of a contributing factor as 19 (4-phenol) showed greater
potency compared to 18 (3-phenol), a trend similar to the ester series. A second noticeable result
was the comparable activities for 8 and 19. Comparing the activities of 8, 16, and 19 would
suggest that the aromatic A-ring aligns the amine in the linker for key intermolecular binding
interactions with the VD3 binding pocket.
Differing results were obtained by incorporating either a hydroxyl or methyl ether on the side
chain at C-25 (Table 2). Compared to the non-hydroxylated analogues 7 and 8, hydroxylation at
C-25 (21 and 22, respectively), demonstrated comparable anti-Hh activity in the MEFs, but were
significantly more active in the ASZ001 BCC cells (IC50 value range = 0.11 - 0.22 µM).
Hydroxylation at C-25 for these analogues also resulted in enhanced upregulation of Cyp24A1
(15,000 – 30,000 compared to 20- to 40-fold over DMSO), suggesting the hydroxyl moiety
significantly improves binding to and activation of VDR. Masking the C-25 hydroxyl as the
methyl ether (25) reduced both Hh inhibition and VDR activation compared to 21 in both cell
lines.
11

ACCEPTED MANUSCRIPT
Surprisingly, the addition of the C-25 hydroxyl group (27) to the extended linker analogue
(18) had little effect on Hh inhibition (IC50 values ~1 µM) and minimal Cyp24A1 upregulation
(10 – 30-fold) in either cell line. Similarly, minimal differences in potency or selectivity were
found following C-25 hydroxylation of 19 (analogue 28) except for a slight increase in Cyp24A1
upregulation in ASZ cells (6-fold vs. 140-fold). Even though no major differences were observed
following C-25 hydroxylation, etherification at the C-25 position of 27 (analogue 31) yielded
interesting results. Analogue 31 demonstrated increases in anti-Hh potency as well an increase in
upregulation of Cyp24A1 in ASZ cells.
Table 2. Evaluation of C-25 functionalized analogues.
C3H10T1/2a
ASZ001
% Hh
Cyp24A1a,c
Cyp24A1c, f
IC50 (µM)d
IC50 (µM)d
EC50 (µM)e
Activitya,b
GDC 0.9 ± 0.1
1.6 ± 0.1
8,300 ± 40
>30,000f
>30,000
>30,000
15,000 ± 2,000
89 ± 7
82 ± 7 h
40 ± 10h
1.1 ± 0.1
0.5 ± 0.1
89.1 ± 41.5 h
0.1 ± 0.1
0.2 ± 0.1
0.6 ± 0.1
1.3 ± 0.1
0.5 ± 0.1
0.7 ± 0.1
0.4 ± 0.1
1.0 ± 0.2
0.9 ± 0.2
--
35.7 ± 0.3
9.2 ± 0.4f
30 ± 1
4.1 ± 0.3
0.4 ± 0.1
3.0 ± 1.0h
1.3 ± 0.2
0.9 ± 0.1
2.3 ± 0.1
1.7 ± 0.1
0.4 ± 0.1
1.1 ± 0.1
1.6 ± 0.1
1.6 ± 0.1
2,200 ± 200
11,000 ± 400g
>30,000
0.9 ± 0.1
0.6 ± 0.1
0.1 ± 0.1
0.1 ± 0.1
0.6 ± 0.1
0.7 ± 0.1
> 10
1
2
3g
21
22
25
27
28
31
33
34
5.9 ± 0.9
11 ± 1
>30,000
>30,000
3.3 ± 0.7
4.1 ± 0.7
5.6 ± 2.2
3.7 ± 0.7
2.9 ± 0.5
0.7 ± 0.1
11,000 ± 2,000
8.4 ± 2.0
34 ± 1
6.1 ± 0.4
27 ± 4
140 ± 8
1.3 ± 0.7
0.6 ± 0.1
0.4 ± 0.6
>10
2,400 ± 800
>30,000
27,000 ± 800
14 ± 2
14 ± 3
aAll analogues evaluated at 5 µM unless otherwise indicated. bValues represent %Gli1 mRNA expression relative to
OHC control (set to 100) after 24 h. cValues represent Cyp24A1 mRNA expression compared to DMSO control (set
d
to 1.0). Gli1 IC50 values represent mean ± SEM for at least two separate experiments performed in triplicate.
12

ACCEPTED MANUSCRIPT
eCyp24A1 EC50 values represent mean ± SEM for at least two separate experiments performed in triplicate.
g
h
fCompound evaluated at 2.5 µM. Compound evaluated at 1 µM. Value expressed in nanomolar (nM). GDC =
Vismodegib or GDC-0449
With significant upregulation of Cyp24A1 observed in both cell lines at a single
concentration for several of the 25-OH and 25-OMe analogues, EC50 values were obtained in the
ASZ001 cells to further determine selectivity for Hh over VDR signaling (Table 2).
Interestingly, a strong correlation was observed between Gli1 IC50 and Cyp24A1 EC50 values for
multiple analogues (21, 25, 31, and 33). In addition, even analogues that did not correlate as
closely in regards to the absolute Gli1 IC50 and Cyp24A1 EC50 values (27 - 28, 34) followed the
overall trend that reduced Hh inhibition is closely associated with reduced VDR activation.
Taken together, these results suggested to us a potential role for VDR activation in the regulation
of Hh signaling for VD-based compounds, results that were explored further as described below.
Finally, several of the analogues that demonstrated potent Hh-inhibition in the immortalized
MEF and ASZ cell lines were evaluated for their anti-proliferative activity in a primary Hh-
dependent MB derived from a conditional Patched knockout (Math1-Cre:Ptchfl/fl, hereafter called
Ptch-CKO) mouse (Table 3) [17-19]. Interestingly, the GI50 value for each of these compounds
was greater than 1 µM, which is significantly higher than the IC50 values for Gli1 down-
regulation in the C3H10T1/2 and ASZ cells. Of particular note was the reduced activity of
calcitriol (3, GI50 = 1.4 µM) in the Ptch-CKO cells compared to the immortalized cell lines.
Table 3. In vitro activity in primary Ptch-CKO MB cells.
Analogue
GI50 (µM)a
5.8 ± 2.8b
1.4 ± 0.4
>10
1
3
5
2.1 ± 0.5
10
13

ACCEPTED MANUSCRIPT
2.1 ± 0.7
5.3 ± 3.5
2.2 ± 1.0
1.6 ± 0.5
>10
19
21
22
27
33
aAll analogues evaluated
following 48 hr incubation.
bAll values represent Mean ±
SEM of at least two separate
experiments performed in
triplicate.
Select analogues were also evaluated in a series of in vitro studies to determine preliminary
pharmacokinetic (PK) parameters for this scaffold (Table 4). The first take-away from this study
was the overall poor solubility of the compounds, which were essentially insoluble at pH 4.0 and
7.4. This is not a surprising result given the overall hydrophobic nature of the VD3 scaffold. The
most soluble compound at pH 4.0 was analogue 22 with a single-atom amine seco-B and C-25
hydroxylation, but it demonstrated a significant decrease in solubility at pH 7.4. Analogue 6 with
the amide seco-B region was the second most soluble compound at pH 4.0 and maintained
solubility at pH 7.4.
Table 4. Pharmacokinetic studies of select analogues.
Clearance
(mL/min/mg
protein)
ND
Solubility (mg/mL)
T1/2
Cmpd
(HLM, min)a
pH 4.0
pH 7.4
>120b
>231
ND
ND
3
< 3
1.6 x 10-4
2.7 x 10-4
6.2 x10-7
7.5 x 10-7
1.3 x 10-7
2.4 x 10-3
1.7 x 10-5
6.0 x 10-5
4.8 x 10-5
6.7 x 10-4
6.8 x 10-4
2.6 x 10-6
2.6 x 10-7
2.7 x 10-6
4.2 x 10-6
1.4 x 10-5
1.2 x 10-4
3.1 x 10-5
5
>231
< 3
6
57 ± 0.1
>231
12 ± 0.2
< 3
8
10
19
22
25
31
33
>231
< 3
31 ± 0.5
>231
22 ± 0.4
< 3
43 ± 7
47 ± 1
16 ± 3
15 ± 0.3
14

ACCEPTED MANUSCRIPT
49 ± 7
15 ± 0.3
2.7 x 10-6
2.3 x 10-6
34
aMetabolic stability in human liver microsomes (HLM), time until 50% of compound
b
disappeared. Metabolic stability in rat liver homogenate, time until 50% of compound
disappeared [Ref 20].
Comparing the results of metabolic stability assays for 8, 10, 19 and 22 can provide general
conclusions in regards to the overall stability of the amine series of analogues since each of these
compounds contain the 4-phenolic A-ring moiety. Looking at compounds 8 and 10, the only
structural difference is the orientation of the seco-B region at C-8, where 8 maintains the α-
orientation. While both compounds maintain similar solubility, the α-orientation is significantly
less stable in human liver microsomes (HLMs) and has a higher clearance rate compared to the
β-orientation of 10. Similarly, the extended amine seco-B analogue 19 maintains β-orientation at
C-8 and shows a very similar stability profile to the single atom seco-B analogue 10, while the
α-orientation of 22 significantly reduces metabolic stability. Addition of the C-25 hydroxyl to 8
(analogue 22) decreased the half-life and increased the clearance rate while increasing the
solubility at pH 4.0. Although 25 has the 3-phenol A-ring, masking of the C-25 hydroxyl group
as the methyl-ether increased the half-life significantly and reduced clearance to values similar to
the non-functionalized C-25 analogues.
2.3.3. Knockout cell line studies
Multiple reports have suggested both direct and indirect Hh inhibitory mechanisms for
vitamin D-based compounds [9-10, 15, 21-22]. Based on our repeated findings that inhibition of
Hh signaling correlates with upregulation of VDR activity, we hypothesized that vitamin D-
based compounds may inhibit Hh signaling through direct binding to an Hh pathway component,
but their anti-Hh activity might also be a downstream result of VDR activation. To further
investigate potential crosstalk between the two signaling pathways, a series of studies in several
15

ACCEPTED MANUSCRIPT
knockout MEFs and the primary Ptch-CKO cells to evaluate 1, 3, 8, and 22 were performed.
Compounds 8 and 22 were chosen because they have similar Gli1 IC50 values across cell lines
despite significantly different abilities to upregulate Cyp24A1 mRNA.
Initially, we studied the ability of these compounds to regulate Hh and VDR signaling at a
single concentration in VDR-/- and SuFu-/- cells (Figure 2). If the vitamin-D based analogues
down-regulate Gli1 independently of VDR, then they should maintain their potent Hh inhibition
in the VDR-/- MEFs. By contrast, if these analogues regulate Hh signaling at the level of Smo,
which has been suggested through several indirect studies [9-10, 15], their ability to down-
regulate Gli1 expression should be completely abolished in the SuFu-/- MEFs. Interestingly, the
ability of the steroidal compounds to inhibit Hh signaling was significantly reduced, but not
completely abolished, in the two knockout lines compared to the wild-type C3H10T1/2 MEFs.
By contrast, the known Smo antagonist Vismodegib retained full inhibition of Hh signaling in
the VDR-/- MEFs, but was completely inactive in the SuFu-/- MEFs. These results strongly
suggest that the vitamin-D based analogues are capable of regulating the Hh pathway through
both canonical Hh signaling and VDR. Not surprisingly, the ability of these analogues to up-
regulate Cyp24A1 was significantly reduced in the VDR-/- MEFs; however, we also noted
decreased up-regulation of Cyp24A1 in the Sufu-/- MEFs.
16

ACCEPTED MANUSCRIPT
Fig. 2. Effects of VD3-based analogues on VDR and Hh signaling in knockout MEFs.
Comparison of Gli1 down-regulation (A) and Cyp24A1 up-regulation (B) across cell lines.
Based on the results of the single concentration studies, we performed full dose-response
experiments in both VDR-/- and Sufu-/- MEFs for 1, 3, 8 and 22. Significant decreases in anti-Hh
potency and Cyp24A1 up-regulation were observed for 3, 8 and 22 (Table 5). Calcitriol (3)
showed the greatest decrease in potency with an IC50 of ~3 nM in C3H10T1/2 MEFs and IC50
values of ~ 6 – 8 µM in VDR-/- and Sufu-/- cells, respectively. Additionally, Cyp24A1 expression
17

ACCEPTED MANUSCRIPT
induced by 3 significantly decreased from the C3H10T1/2 MEFs to the knockout MEFs. Both
synthetic analogues demonstrated similar trends in the VDR-/- and Sufu-/- MEFs, i.e. a significant
reduction in Hh pathway inhibition in both cell lines compared to the wild-type MEFs.
Interestingly, the ability of compound 22 to upregulate Cyp24A1 in the VDR-/- cells provided
support for the Sufu/VDR cross-talk mechanism recently proposed [15].
Table 5. In vitro activity for select analogues in various MEFs.
C3H10T1/2
VDR-/-
SuFu-/-
Cyp24A1a,b
2.9 ± 1.0
590 ± 70d
1.2 ± 0.1
51 ± 5
Cmpd
Cyp24A1a,b
8,300 ± 40
>30,000d
Cyp24A1a,b
5.5 ± 0.1
530 ± 200d
8.1 ± 2
IC50 (µM)c
IC50 (µM)c
IC50 (µM)c
> 10
4.1 ± 0.3
4.5 ± 1.0
6.3 ± 1.0
3.9 ± 0.2
4.9 ± 0.01
1
0.0031 ± 0.001
0.40 ± 0.02
0.89 ± 0.03
8.5 ± 3
3
45 ± 20
3.1 ± 0.4
2.9 ± 0.3
8
15,000 ± 2,000
140 ± 30
22
aAll analogues evaluated at 5 µM unless otherwise indicated. Values represent Cyp24A1 mRNA expression
b
c
compared to DMSO control (set to 1.0). Gli1 IC50 values represent mean ± SEM for at least two separate
experiments performed in triplicate. dEvaluated at 1 µM.
To determine if the potential for crosstalk between VDR and Hh signaling was specific to the
steroidal scaffold or a broader effect of VDR activation, we evaluated the ability of a non-seco-
steroidal VDR agonist, LG190178 (35, Fig. 3), to modulate Hh and VDR signaling in several
cell lines (Table 6). In the initial disclosure of 35, the compound was determined to have lower
calcemic effects than 3 while being equipotent as a VDR agonist [23]. Similar to 3, the non-
steroidal agonist 35 significantly up-regulated Cyp24A1 expression in the C3H10T1/2 and ASZ
cell lines (Table 5). Unexpectedly, compound 35 was significantly more potent as an Hh
pathway inhibitor (sub-nanomolar IC50 values) than compound 3 in the MEF and ASZ cells.
Finally, a significant reduction in Hh inhibition for both compounds was seen in the VDR-/- cells
(low micromolar IC50 values).
As noted above, 3 was significantly less active in the Ptch-CKO cells compared to either the
C3H10T1/2 or ASZ cells. Similar anti-proliferative results were obtained for compound 35 in the
18

ACCEPTED MANUSCRIPT
Ptch-CKO cells (GI50 > 10 µM). The reduced anti-Hh activity in the primary cell line correlated
well with our results in the VDR-/-, which led us to evaluate VDR expression levels in the Ptch-
CKO cells. Expression of VDR mRNA was significantly lower in the Ptch-CKO cells compared
to normal granule neuron precursors and normal adult cerebellum directly isolated from the
mouse, as well as the Hh-dependent NIH3T3 MEF cell line (Supplemental Figure 1). Taken
together, the data from the knockout cell lines strongly suggest that inhibition of Hh signaling
can be partially attributed to a downstream effect of VDR activation and that these results are not
specific to the steroidal scaffold.
Fig. 3. Structure of Non-steroidal VDR agonist LG190178.
Table 6. Comparing the cellular activity of calcitriol (3) and LG190178 (35).
C3H10T1/2a
ASZ001e
Gli1 IC50
(nM)c
VDR-/- a
Cmpd
Gli1 IC50
Gli1 IC50
EC50
Cyp24A1a,b
Cyp24A1a,b
>30,000
Cyp24A1a,b
(nM)c
(nM)d
(µM)c
>30,000
>30,000
3.1 ± 1.0
89.1 ± 41.5
130 ± 1
530 ± 200
250 ± 100
6.3 ± 1.0
3
0.25 ± 0.04
13,000 ± 2,000 0.23 ± 0.11
7.7 ± 3.2
1.9 ± 0.2
35
aCompounds evaluated at 1 µM. Values represent Cyp24A1 mRNA expression compared to DMSO control (set
b
c
d
to 1.0). IC50 (Gli1, nM) and EC50 (Cyp24A1, nM) values represent mean ± SEM for at least two separate
experiments performed in triplicate.
3. Conclusion
Combining the results contained herein with our previous VD3 SAR studies [11-13], it is
clear that the free hydroxyl group of the A-ring is required to maintain anti-Hh activity for the
amine-linked analogues; however, the preferred position of the phenol shifted from the 3-
19

ACCEPTED MANUSCRIPT
position in the ester/amide series (5-6, 9) to the 4-position for the amine-linked analogues (7-8,
10). In addition, extension of the single atom amine tether by a single methylene (16, 18-19)
demonstrated the same preference for the para- over the meta-phenol. This shift may be due to
the slightly increased flexibility of the seco-B amine region, which may allow these compounds
to better orient the hydroxyl group within the binding site. The more rigid ester and amide seco-
B analogues are similar to the natural diene system of the vitamin D compounds. As expected,
the addition of a hydroxyl moiety at C-25 resulted in significant upregulation of Cyp24A1
mRNA expression across both cell lines for the majority of VD3 analogues, presumably as a
result of VDR activation. Taken together with the cellular data, our preliminary PK studies
provide additional insight into future analogue development. First, while a specific orientation of
the seco-B region is not required for Hh inhibition, the β-orientation is more stable. Second,
masking the C-25 hydroxyl group may provide a method to balance cellular potency/selectivity
and increased microsomal stability.
In addition to the SAR we determined for these vitamin D-based compounds, we noticed a
significant correlation between Hh inhibition and VDR activation for several analogues, leading
us to explore whether the anti-Hh properties of these compounds are dependent on VDR
activation. When select analogues (1, 3, 8, and 22) were analyzed in VDR-/- and Sufu-/- MEFs, a
significant decrease in anti-Hh was observed; however, inhibition of Hh signaling was not
completely abolished. The partial loss in both knockout lines suggests that these compounds may
downregulate Gli1 through multiple cellular targets. In addition, as a nuclear receptor, activation
of VDR may be able to overcome Sufu knockout by disrupting activation/transcription of Hh
target genes. A role for VDR in regulating the Hh pathway is further supported by our findings
that the non-secosteroidal VDR agonist 35 is a potent inhibitor of the Hh signaling pathway.
20

ACCEPTED MANUSCRIPT
Additional studies to further characterize potential cross-talk between these two cellular
pathways is currently underway.
4. Experimental Section
4.1. General information
VD3 used for chemical synthesis was purchased from HBC Chem, Inc. Solvents (anhydrous,
ACS or HPLC grade) were purchased from Fisher Scientific or Sigma Aldrich. Column
chromatography was performed using silica gel purchased from Sorbtech (Sorbent
Technologies). NMR data was collected on a Bruker AVANCE 500 MHz spectrometer using
deuterated chloroform (CDCl3, Cambridge Isotope Laboratories, INC), and analysis performed
using MestReNova 12.0.0. HRMS data was analyzed at the Mass Spectrometry Facility at the
University of Connecticut. Infrared (IR) analysis was performed on a Bruker Alpha Platinum
ATR instrument using OPUS software (v 7.2).
4.2. Purity analysis
All compounds analyzed in biological assays were greater than 95% pure based on the following
HPLC method. Analogues were dissolved in HPLC grade MeOH (1 mg/mL) and injected (20 –
50 µL) into an Agilent Manual FL-Injection Valve (600 bar) on an Agilent 1100/1200 Series
HPLC equipped with a Kinetex 5 µm C18 100Å (150 X 4.6 mm) column and Agilent 1100
Series Photodiode Array Detector. The mobile phase consisted of 95% MeOH/5% H2O for all
analogues with a run time of 10 – 20 min and a flow rate of 1.0 mL/min. Purity was assessed at
254 nm.
21

ACCEPTED MANUSCRIPT
4.3. Synthesis of amine seco-B analogues
4.3.1. General reductive amination procedure [13]. Ketone 11 (0.6 mmol) and the requisite
amine (1.2 mmol) were dissolved in 1,2-dichloroethane (8 mL). Sodium triacetoxyborohydride
(1.8 mmol) was added followed by acetic acid (50 µL) and titanium (IV) isopropoxide (1.2
mmol). The mixture was stirred at RT or reflux for 24 - 48 h. The reaction was quenched with
saturated sodium bicarbonate (20 mL) and diluted with diethyl ether (60 mL). The aqueous layer
was washed with diethyl ether (3 × 60 mL) and the organic layers combined, dried over
anhydrous MgSO4, filtered, and concentrated. Crude residue was purified by column
chromatography (SiO2, 100% Hex to 25% EtOAc in Hex) affording recovery of the starting
ketone 11 (20-60%), and the desired amine analogue (15 - 80%).
4.3.2. (1R,3aR,4S,7aR)-7a-methyl-1-((R)-6-methylheptan-2-yl)-N-phenyloctahydro-1H-inden-4-
amine (12). Reductive amination of 11 with aniline provided amine 12; 65%, 69 mg. HPLC:
95%, Rt = 7.408 min. 1H NMR (500 MHz, CDCl3)
δ 7.19 (t, J = 7.9 Hz, 2H), 6.68 (t, J = 7.3 Hz,
1H), 6.62 (d, J = 7.7 Hz, 2H), 3.77 (s, 1H), 2.04 (m, 2H), 1.9-1.5 (9H), 1.41 (m, 3H), 1.31 (m,
3H), 1.17 (m, 5H), 1.07 (m, 1H), 1.00 (m, 3H), 0.97 (d, J = 6.7 Hz, 3H), 0.92 (m, 6H). 13C NMR
(126 MHz, CDCl3)
δ 148.5, 129.2, 129.2, 116.4, 112.5, 56.9, 51.7, 51.1, 45.7, 42.2, 40.3, 39.5,
35.9, 35.4, 30.6, 28.0, 27.1, 23.8, 23.2, 22.8, 22.5, 18.5, 17.9, 13.6. IR (neat) vmax 2926, 1600,
1502, 1466, 1428, 1313, 1249, 863. DART-HRMS: m/z calcd. for C24H39NO: 342.3161 [MH]+.
Found 342.3132.
4.3.3. (1R,3aR,4S,7aR)-N-(3-methoxyphenyl)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-
1H-inden-4-amine (13). Reductive amination of 11 with m-anisidine provided amine 13; 45%, 23
22

ACCEPTED MANUSCRIPT
1
mg. HPLC: 99%, Rt = 4.062 min. H NMR (500 MHz, CDCl3)
δ 7.09 (m, 1H), 6.24 (m, 2H),
6.16 (m, 1H), 3.81 (s, 3H), 3.75 (s, 1H), 2.03 (m, 2H), 1.9-1.5 (8H),1.40 (m, 3H), 1.31 (m, 2H),
1.17 (m, 4H), 1.06 (m, 1H), 0.99 (m, 3H),0.96 (m, 3H), 0.92 (m, 6H). 13C NMR (126 MHz,
CDCl3)
δ 160.9, 149.9, 129.9, 129.8, 105.9, 101.5, 98.4, 56.9, 55.1, 51.7, 51.1, 45.6, 42.2, 40.2,
39.5, 39.4, 36.4, 35.9, 35.4, 35.2, 30.6, 28.04, 28.01, 27.7, 27.1, 24.5, 24.4, 23.8, 23.2, 22.8, 22.5,
20.2. IR(neat) vmax 2927, 1611, 1493, 1463, 1208, 1159, 1208, 1159, 1050, 829, 750, 686.
DART-HRMS: m/z calcd. for C25H41NO: 372.3266; [MH]+. Found 372.3233.
4.3.4. (1R,3aR,4S,7aR)-N-(4-methoxyphenyl)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-
1H-inden-4-amine (14). Reductive amination of 11 with p-anisidine provided amine 14; 42%, 32
1
mg. HPLC: 99%, Rt = 4.891 min. H NMR (500 MHz, CDCl3)
δ 6.81 (d, J = 8.9 Hz, 2H), 6.59
(d, J = 8.8 Hz, 2H), 3.79 (s, 3H), 3.68 (s, 1H), 2.04 (m, 2H), 1.92 (m, 1H), 1.9-1.5 (7H), 1.39 (m,
3H), 1.31 (m, 2H), 1.17 (m, 5H), 1.06 (m, 1H), 1.00 (s, 3H), 0.96 (m, 3H), 0.92 (m, 6H). 13C
NMR (126 MHz, CDCl3)
δ 151.5, 143.0, 115.0, 113.8, 56.9, 55.9, 52.2, 51.9, 42.2, 40.3, 39.5,
35.9, 35.4, 30.1, 28.0, 27.1, 23.8, 23.2, 22.8, 22.6, 18.5, 17.9, 13.7. IR(neat) vmax 2928, 2864,
1508, 1464, 1262, 1172, 1041, 814. DART-HRMS: m/z calcd. for C25H41NO: 372.3266; [MH]+.
Found 372.3249.
4.3.5.
N-((1R,3aR,4S,7aR)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-1H-inden-4-
yl)benzo[d][1,3]dioxol-5-amine (15). Reductive amination of 11 with (3,4-methylenedioxy)-
1
aniline provided amine 15; 20%, 24 mg. HPLC: 95%, Rt = 5.672 min. H NMR (500 MHz,
CDCl3)
δ 6.68 (d, J= 8.3 Hz, 1H), 6.26 (s, 1H), 6.05 (m, 1H), 5.87 (m, 2H), 3.65 (s, 1H), 2.03
(m, 2H), 1.90 (m, 1H), 1.8-1.5 (8H), 1.38 (m, 4H), 1.29 (m, 2H), 1.19 (m, 5H), 1.06 (m, 1H),
23

ACCEPTED MANUSCRIPT
0.98 (s, 3H), 0.95 (d, J = 6.5 Hz, 3H), 0.92 (m, 6H). 13C NMR (126 MHz, CDCl3)
δ 148.3, 144.5,
108.6, 100.4, 95.6, 56.9, 52.2, 51.8, 42.2, 40.3, 39.5, 35.9, 35.4, 30.6, 28.0, 27.1, 23.8, 23.2, 22.8,
22.5, 18.5, 17.8, 13.6. IR(neat) vmax 3378, 2920, 2860, 1714, 1635, 1600, 1504, 1465, 1377,
1094, 875. DART-HRMS: m/z calcd. for C25H39NO2: 386.3059 [MH]+. Found 386.3048.
4.3.6.
4-((((1R,3aR,4S,7aR)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-1H-inden-4-
yl)amino)methyl)phenol (16). Reductive amination of 11 with 4-hydroxybenzylamine provided
amine 16; 32%, 10 mg. HPLC: 97%, Rt = 3.399 min. 1H NMR (500 MHz, CDCl3)
δ
7.05 (d, J =
Hz, 2H), 6.78 (d, J = Hz, 2H), 5.17 (s, 1H), 3.40 (s, 1H), 3.02 (m, 1H), 2.54 (m, 2H), 2.37 (m,
1H), 2.1-1.9 (3H), 1.75 (m, 2H), 1.67 (m, 5H), 1.55 (m, 2H), 1.48 (m, 2H), 1.35 (m, 3H), 1.15
13
(m, 3H), 1.06 (s, 3H), 0.90 (m, 9H). C NMR (126 MHz, CDCl3)
δ 153.9, 131.6, 130.1, 115.2,
59.8, 55.0, 48.8, 39.4, 36.9, 36.7, 35.3, 34.2, 29,7, 28.3, 28.0, 26.2, 24.3, 22.8, 22.5, 22.4, 22.1,
19.1. IR(neat) vmax 2953, 2922, 2854, 1732, 1685, 1553, 1515, 1455, 1259, 1090, 1011, 799.
DART-HRMS: m/z calcd. for C25H42NO: 372.3266 [MH]+. Found 372.3248.
4.3.7.
(1R,3aS,4R,7aR)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-1H-indene-4-
carbaldehyde (17). A mixture of methoxymethyl triphenylphosphonium chloride (2.1 g, 6 mmol)
in THF (30 mL) was cooled to 0° C. With vigorous stirring, sodium t-butoxide (640 mg, 5.7
mmol) was added. The heterogeneous mixture immediately changed color to yellow and on to
deeper yellow and finally deep red at 0 ºC. After 30 min, a solution of 11 (400 mg, 1.5 mmol) in
THF (10 mL) was added to the above mixture. TLC indicated consumption of 11 after 20 min.
Water (80 mL) was added and the mixture was washed with ethyl acetate (3 x 80 mL). The
organic layers were combined, dried over anhydrous MgSO4, filtered, and concentrated. The
24

ACCEPTED MANUSCRIPT
crude material was purified using column chromatography (SiO2, 100% Hex to 2% EtOAc in
Hex) to yield the vinyl ether; 89%, 390 mg as a clear oil. 1H NMR (500 MHz, CDCl3) δ 5.84 (m,
1H), 3.43 (s, 3H), 1.97 (m, 3H), 1.77 (m, 4H), 1.52 (m, 3H), 1.42 (m, 3H), 1.33 (m, 3H), 1.25
(m, 1H), 1.19 (m, 5H), 1.01 (m, 1H), 0.92 (d, J = 6.5 Hz, 3H),0.87 (d, J = 6.6 Hz, 3H), 0.86 (d, J
13
= 6.6 Hz, 3H), 0.68 (s, 3H). C NMR (126 MHz, CDCl3) δ138.9, 118.2, 59.3, 56.0, 52.1, 44.4,
40.4, 39.5, 36.2, 28.0, 27.9, 25.1, 23.8, 22.8, 22.5, 22.4, 21.9, 18.8, 11.6. DART-HRMS: m/z
calcd. for C20H37O: 293.2844; [MH]+ Found: 293.2849.
The vinyl ether was refluxed in methanol (10 mL) and 3N HCl (5 mL) for 12 h. The reaction
was neutralized by the addition of 2 M NaOH, and the aqueous fraction washed with ethyl
acetate (3 x 50 mL). The organic layers were combined, dried over anhydrous MgSO4, filtered,
and concentrated. The crude material was purified using column chromatography (SiO2, 100%
Hex to 2% EtOAc in Hex) to yield 17 (75%, 280 mg) as a clear oil. 1H NMR (500 MHz, CDCl3)
δ 9.55 (m, 1H), 2.30 (m, 1H), 1.99 (m, 1H), 1.86 (m, 1H), 1.78 (m, 1H), 1.66 (m, 2H), 1.53 (m,
2H), 1.30 (m, 8H), 1.12 (m, 5H), 1.02 (m, 1H), 0.93 (d, J = 46.5 Hz,3H), 0.87(d, J = 2.1 Hz,
3H), 0.86 (d, J = 2.0 Hz, 3H), 0.71 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 205.1, 55.4, 50.6,
49.4, 42.7, 39.4, 39.4, 36.1, 35.6, 28.0, 27.8, 25.7, 24.2, 23.7, 22.8, 22.5, 20.7, 18.7, 11.6.
DART-HRMS: m/z calcd. for C19H35O: 279.2688; [MH]+ Found: 279.2684.
4.3.8.
3-((((1R,3aS,4R,7aR)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-1H-inden-4-
yl)methyl)amino)phenol (18). Reductive amination of 17 with 3-aminophenol provided 18; 46%,
1
8.3 mg. HPLC: 99%, Rt = 3.119 min. H NMR (500 MHz, CDCl3)
δ 7.05 (t, J = 8.0 Hz, 1H),
6.22 (m, 2H). 6.13 (s, 1H), 4.50 (bs, 1H), 3.17 (m, 1H), 2.79 (m, 1H), 2.03 (m, 1H), 1.92 (m,
2H), 1.72 (m, 2H), 1.60 (m, 3H), 1.42 (m, 3H), 1.32 (m, 1H), 1.19 (m 5H), (m, 2H), 0.99 (d, J =
25

ACCEPTED MANUSCRIPT
13
6.5 Hz, 3H), 0.94 (dd, J = 6.6, 2.1 Hz, 6H), 0.74 (s, 3H). C NMR (126 MHz, CDCl3)
δ 156.8,
150.2, 130.1, 105.9, 104.1, 99.6, 56.2, 53.7, 49.0, 43.2, 40.0, 39.6, 39.5, 36.5, 36.4, 36.2, 35.7,
31.2, 28.1, 28.0, 27.9, 24.6, 24.5, 23.8, 22.8, 22.6, 21.9, 18.7, 11.9. IR(neat) vmax 2948, 2923,
2864, 1614, 1508, 1467, 1438, 1333, 1186, 1157, 838. DART-HRMS: m/z calcd. for C25H42NO:
372.3266 [MH]+. Found 372.3270.
4.3.9.
4-((((1R,3aS,4R,7aR)-7a-methyl-1-((R)-6-methylheptan-2-yl)octahydro-1H-inden-4-
yl)methyl)amino)phenol (19). Reductive amination of 17 with 4-aminophenol provided 19; 38%,
6.2 mg. HPLC: 96%, Rt = 3.384 min. 1H NMR (500 MHz, CDCl3)
δ
6.72 (m, 2H), 6.59 (m, 2H),
4.33 (m, 2H), 3.20 (m, 1H), 2.71 (m, 1H), 1.90(m, 4H), 1.72 (m, 2H), 1.57 (m, 3H), 1.41 (m,
5H), 1.30 (m, 2H), 1.18 (m 6H), 1.06 (m, 1H), 0.99 (m, 3H), 0.93 (m, 6H), 0.73 (s, 3H). 13C
NMR (126 MHz, CDCl3) δ 116.2, 114.1, 56.2, 53.7, 52.1, 50.1, 45.5, 44.0, 43.1, 39.5, 36.4, 34.8,
31.2, 28.0, 27.6, 25.7, 24.4, 23.8, 22.8, 22.5, 21.9, 20.5, 19.5, 18.7, 11.8. IR(neat) vmax 2950,
2923, 2865, 1512, 1468, 1232, 1215, 818. DART-HRMS: m/z calcd. for C25H42NO: 372.3266
[MH]+. Found 372.3268.
4.4. Synthesis of C-25 functionalized single atom nitrogen containing seco-B analogues.
4.4.1.
(1R,3aR,7aR)-1-((R)-6-hydroxy-6-methylheptan-2-yl)-7a-methylhexahydro-1H-inden-
4(2H)-one (20). RuCl3 (100 mg, 0.48 mmol) and NaIO4 (1.5 g, 7.0 mmol) were suspended in
H2O (20 mL) with pyridine (40 µL, 0.49 mmol). A solution of 11 (600 mg, 2.3 mmol) in MeCN
(20 mL) was added to the reaction mixture and stirred at 60 ºC for 48 h. The mixture was diluted
with saturated sodium thiosulfate (20 mL) and filtered through celite. The filtrate was washed
26

ACCEPTED MANUSCRIPT
with EtOAc (3 x 50 mL), dried with anhydrous sodium sulfate, and concentrated to afford ketone
1
20, which was purified with column chromatography (25% EtOAc in Hex), 420 mg, 67%. H
(500MHz, CDCl3)
δ 2.48 (m, 1H), 2.35-2.23 (2H), 2.15 (m, 1H), 2.06 (m, 1H), 1.93 (m, 2H),
1.77 (m, 1H), 1.65-1.55 (2H), 1.51-1.35 (7H), 1.30 (m, 1H), 1.26 (s, 6H), 1.13 (m, 1H), 1.01 (d, J
13
= 6.0 Hz, 3H), 0.68 (s, 3H). C NMR (126 MHz, CDCl3)
δ 212.0, 71.0, 62.0, 56.6, 49.9, 44.3,
41.0, 39.0, 36.2, 35.5, 29.4, 29.2, 27.5, 24.0, 20.7, 19.1, 18.7, 12.5.
4.4.2.
3-(((1R,3aR,4S,7aR)-1-((R)-6-hydroxy-6-methylheptan-2-yl)-7a-methyloctahydro-1H-
inden-4-yl)amino)phenol (21). Reductive amination of 3-aminophenol with ketone 20 provided
amine 21; 34%, 12 mg. HPLC: 95%, Rt = 2.624 min. 1H NMR (500 MHz, CDCl3)
δ
7.01 (t, J =
8.0 Hz, 1H), 6.19 (m, 2H), 6.12 (m, 1H), 3.04 (m, 1H), 2.20 (m, 1H), 1.98 (m, 2H), 1.77 (m, 2H),
1.50 (m, 10H), 1.31 (m, 1H), 1.28 (s, 6H), 1.13 (m, 2H), 1.01 (d, J = 6.5 Hz, 3H), 0.98 (s, 1H),
13
0.96 (s, 3H). C NMR (126 MHz, CDCl3)
δ 156.9, 149.8, 130.1, 106.0, 103.5, 99.3, 71.4, 56.5,
53.4, 45.6, 45.3, 44.3, 36.7, 35.2, 35.2, 33.1, 29.4, 29.3, 29.2, 29.1, 27.7, 24.3, 24.2, 21.3, 20.2,
19.4. IR(neat) vmax 3492, 3252, 2916, 1510, 1404, 1245, 1210, 1184, 1155, 962, 816. DART-
HRMS: m/z calcd. for C24H39NO2: 374.3059 [MH]+. Found 374.3068.
4.4.3.
4-(((1R,3aR,4S,7aR)-1-((R)-6-hydroxy-6-methylheptan-2-yl)-7a-methyloctahydro-1H-
inden-4-yl)amino)phenol (22). Reductive amination of 4-aminophenol with ketone 20 provided
1
amine 22; 29%, 19 mg. HPLC: 99%, Rt = 2.340 min. H NMR (500 MHz, CDCl3)
δ 6.71 (m,
2H), 6.53 (m, 2H), 2.96 (m, 1H), 2.17 (m, 1H), 1.97 (m, 2H), 1.80 (m, 2H), 1.68 (m, 1H), 1.58
(m, 1H), 1.47 (m, 6H), 1.27 (s, 6H), 1.11 (m, 1H), 1.00 (d, J = 6.6 Hz, 3H), 0.95 (s, 3H). 13C
NMR (126 MHz, CDCl3)
δ 147.4, 116.2, 114.8, 109.9, 71.2, 56.4, 54.9, 45.7, 45.3, 44.3, 36.7,
27

ACCEPTED MANUSCRIPT
35.3, 35.2, 33.2, 29.7, 29.4, 29.2, 27.7, 24.5, 24.2, 21.4, 20.2, 19.4. IR(neat) vmax 3520, 3250,
2915, 1514, 1373,1241, 1189, 1162, 1105, 910, 822. DART-HRMS: m/z calcd. for C24H39NO2:
374.3059 [MH]+. Found 374.3058.
4.4.4.
(1R,3aR,7aR)-1-((R)-6-methoxy-6-methylheptan-2-yl)-7a-methylhexahydro-1H-inden-
23 and (1R,3aR,5R,7aR)-1-((R)-6-methoxy-6-methylheptan-2-yl)-5,7a-
4(2H)-one
(
)
dimethylhexahydro-1H-inden-4(2H)-one (24). Ketone 11 (100 mg, 0.38 mmol) was dissolved in
3 mL anhydrous DMF, capped with a rubber septum and cooled to 0º C. MeI (250 µL, 3.9 mmol)
was added followed by NaH (70 mg, 1.1 mmol). The reaction was stirred at 0º C for 2 hrs before
warming to RT for 16 hrs, at which time the reaction was quenched with H2O and extracted with
EtOAc (3 x 10 mL). The organic layers were concentrated and purified via column
chromatography (SiO2, 4% EtOAc in hexanes). The reaction yielded both starting material 11
(50 mg, 50 %) along with ketones 23 (27 mg, 25%) and 24 (20 mg, 17%).
1
Ketone 23. H NMR (500 MHz, CDCl3)
δ 3.21 (s, 3H), 2.49 (m, 1H), 2.30 (m, 2H), 2.14 (m,
1H), 2.03 (m, 1H), 1.95 (m, 2H), 1.73 (m, 2H), 1.6 – 1.2 (12H), 1.18 (s, 6H), 1.12 (m, 1H), 1.00
13
(d, J = 6.2 Hz, 3H), 0.67 (s, 3H). C NMR (126 MHz, CDCl3)
δ 212.1, 74.6, 62.0, 56.6, 49.9,
49.1, 40.9, 40.2, 39.0, 36.3, 35.5, 27.5, 25.1, 25.0, 24.0, 20.2, 19.1, 18.7, 12.5.
1
Ketone 24. H NMR (500 MHz, CDCl3)
δ
3.21 (s, 3H), 2.72 (m, 1H), 2.46 (m, 1H), 2.11 (m,
1H), 1.96 (m, 2H), 1.73 (m, 4H), 1.43 (m, 8H), 1.22 (d, J = 7.4 Hz, 3H), 1.18 (m, 6H),1.08 (m,
1H), 1.00 (d, J = 6.3 Hz, 3H), 0.68 (s, 3H). 13C NMR (126 MHz, CDCl3)
δ
216.0, 74.6, 57.3,
56.6, 49.6, 49.1, 43.7, 40.2, 36.2, 35.5, 35.1, 30.3, 29.7, 27.6, 25.0, 20.1, 19.0, 18.7, 18.5, 13.1.
28

ACCEPTED MANUSCRIPT
4.4.5.
3-(((1R,3aR,4S,7aR)-1-((R)-6-methoxy-6-methylheptan-2-yl)-7a-methyloctahydro-1H-
inden-4-yl)amino)phenol (25). Reductive amination of 3-aminophenol with ketone 23 provided
1
amine 25; 15%, 14 mg. HPLC: 95%, Rt = 2.801 min. H NMR (500 MHz, CDCl3)
δ 7.02 (m,
1H), 6.16 (m, 3H), 3.73 (s, 1H), 3.23 (m, 3H), 2.02 (m, 2H), 1.9-1.3 (20H), 1.25 (m, 3H), 1.19 (s,
6H), 1.09 (m, 5H), 0.97 (m, 6H), 0.89 (m, 2H). 13C NMR (126 MHz, CDCl3)
δ
156.8, 150.1,
130.0, 105.3, 103.3, 99.0, 74.7, 56.7, 51.6, 51.1, 49.1, 42.2, 40.3, 40.2, 36.2, 35.3, 30.6, 29.7,
27.1, 25.0, 23.2, 20.2, 18.5, 17.8, 13.6. IR(neat) vmax 2993, 2923, 2852, 1725, 1611, 1559, 1509,
1455, 1363, 1261, 1188, 1085, 814. DART-HRMS: m/z calcd. for C25H42NO2: 388.3216 [MH]+.
Found. 388.3243.
4.4.6. (1R,3aS,4R,7aR)-1-((R)-6-hydroxy-6-methylheptan-2-yl)-7a-methyloctahydro-1H-indene-
4-carbaldehyde (26). Aldehyde 26 was synthesized following the same procedure described
above for 17 starting with ketone 20; 70% over two steps, 390 mg. Although a mixture of
α
- and
β-aldehydes were formed (2:1 mixture based on the aldehyde proton peak), a single isomer was
obtained for each reductive amination procedure. 1H NMR (500 MHz, CDCl3)
δ 9.60 (m, 0.5H),
9.53 (m, 1H), 3.47 (m, 0.5H), 3.42 (m, 1H), 1.97 (m, 2H), 1.83 (m, 4H), 1.75 (m, 2H), 1.62 (m,
6H), 1.48 (m, 12H), 1.34 (m, 2H), 1.27 (m, 10H), 1.00 (m, 6H), 0.96 (s, 3H). 13C NMR
(126 MHz, CDCl3)
δ 205.3, 71.0, 55.4, 52.1, 50.7, 49.5, 47.6, 45.7, 44.3, 43.4, 39.4, 36.6, 35.6,
35.0, 29.7, 29.3, 29.2, 27.5, 25.9, 25.7, 24.2, 23.8, 21.2, 20.7, 19.7, 19.2, 18.7, 11.6.
4.4.7.
3-((((1R,3aS,4R,7aR)-1-((R)-6-hydroxy-6-methylheptan-2-yl)-7a-methyloctahydro-1H-
inden-4-yl)methyl)amino)phenol (27). Reductive amination of 3-aminophenol with aldehyde 26
1
provided amine 27; 48%, 15 mg. HPLC: 99%, Rt = 2.375 min. H NMR (500 MHz, CDCl3)
δ
29