Synthesis and biological activity of selenopsammaplin A and its analogues as antitumor agents with DOT1L inhibitory activity

Article abstract of DOI:10.1016/j.bmc.2021.116072

Disruptor of telomeric silencing-1 like (DOT1L) is a histone H3 methyltransferase which specifically catalyzes the methylation of histone H3 lysine-79 residue. Recent findings demonstrate that DOT1L is abnormally overexpressed and the upregulated DOT1L evokes the proliferation and metastasis in human breast cancer cells. Therefore, the DOT1L inhibitor is considered a promising strategy to treat breast cancers. Non-nucleoside DOT1L inhibitors, selenopsammaplin A and its analogues, were firstly reported in the present study. Selenopsammaplin A was newly designed and synthesized with 25% overall yield in 8 steps from 3-bromo-4-hydroxybenzaldahyde, and thirteen analogues of selenopsammaplin A were prepared for structure–activity relationship studies of their cytotoxicity against cancer cells and inhibitory activity toward DOT1L for antitumor potential. All synthetic selenopsammaplin A analogues exhibited the higher cytotoxicity compared to psammaplin A with up to 6 – 60 times depending on cancer cells, and most analogues showed significant inhibitory activities against DOT1L. Among the prepared analogues, the phenyl analogue (10) possessed the most potent activity with both cytotoxicity and inhibition of DOT1L. Compound 10 also exhibited the antitumor and antimetastatic activity in an orthotopic mouse metastasis model implanted with MDA-MB-231 human breast cancer cells. These biological findings suggest that analogue 10 is a promising candidate for development as a cancer chemotherapeutic agent in breast cancers.

Full text of DOI:10.1016/j.bmc.2021.116072

Bioorg. Med. Chem. 35 (2021) 116072
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and biological activity of selenopsammaplin A and its analogues
as antitumor agents with DOT1L inhibitory activity
,
,
Hae Ju Hana 1, Woong Sub Byunb 1, Gyu Ho Leea, Won Kyung Kimb, Kyungkuk Janga,
Sehun Yanga, Jewon Yanga, Min Woo Hac d, Suckchang Honga, Jeeyeon Leea, Jongheon Shinb,
,
Ki Bong Ohe, Sang Kook Leeb *, Hyeung-geun Parka
,
,*
a Research Institute of Pharmaceutical Science and College of Pharmacy, Seoul National University, Seoul 151-742, Republic of Korea
b Natural Products Research Institute, College of Pharmacy, Seoul National University, Seoul 151-742, Republic of Korea
c College of Pharmacy, Jeju National University, 102 Jejudaehak-ro, Jeju 63243, Republic of Korea
d Interdisciplinary Graduate Program in Advanced Convergence Technology & Science, Jeju National University, 102 Jejudaehak-ro, Jeju, 63243, Republic of Korea
e Department of Agricultural Biotechnology, College of Agriculture and Life Sciences, Seoul National University, Seoul 151-921, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Disruptor of telomeric silencing-1 like (DOT1L) is a histone H3 methyltransferase which specifically catalyzes the
methylation of histone H3 lysine-79 residue. Recent findings demonstrate that DOT1L is abnormally overex-
pressed and the upregulated DOT1L evokes the proliferation and metastasis in human breast cancer cells.
Therefore, the DOT1L inhibitor is considered a promising strategy to treat breast cancers. Non-nucleoside DOT1L
inhibitors, selenopsammaplin A and its analogues, were firstly reported in the present study. Selenopsammaplin
A was newly designed and synthesized with 25% overall yield in 8 steps from 3-bromo-4-hydroxybenzaldahyde,
and thirteen analogues of selenopsammaplin A were prepared for structure–activity relationship studies of their
cytotoxicity against cancer cells and inhibitory activity toward DOT1L for antitumor potential. All synthetic
selenopsammaplin A analogues exhibited the higher cytotoxicity compared to psammaplin A with up to 6 – 60
times depending on cancer cells, and most analogues showed significant inhibitory activities against DOT1L.
Among the prepared analogues, the phenyl analogue (10) possessed the most potent activity with both cyto-
toxicity and inhibition of DOT1L. Compound 10 also exhibited the antitumor and antimetastatic activity in an
orthotopic mouse metastasis model implanted with MDA-MB-231 human breast cancer cells. These biological
findings suggest that analogue 10 is a promising candidate for development as a cancer chemotherapeutic agent
in breast cancers.
Selenopsammaplin A
Structure-activity relationship
Cytotoxicity
DOT1L histone H3 methyltransferase
Antimetastasis
1. Introduction
checkpoint, and the DNA stability.1,2 Recent findings also suggested that
the overexpression of DOT1L and hyperactivation of DOT1L-mediated
The overall survival rate of cancer patients is affected not only by the
use of appropriate anticancer drugs but also by the inhibition of cancer
cell metastasis. Disruptor of telomeric silencing-1 like (DOT1L), a his-
tone H3 methyltransferase, specifically catalyzes the methylation of
histone H3 on the lysine-79 residue (H3K79). DOT1L-mediated H3K79
methylation is considered to be strongly associated with various bio-
logical processes, such as embryonic cell development, cell division
H3K79 methylation may play a crucial role in the initiation and main-
tenance of mixed lineage leukemia (MLL)-rearranged leukemia. There-
fore, DOT1L is considered a promising therapeutic target in the
development of anticancer agents for MLL-rearranged leukemia.3–5 In
addition, recent evidence has also suggested that DOT1L may solely
function as an oncogene in solid tumors including breast cancer without
MLL fusion.6–8 In particular, DOT1L transactivates epithelial-
Abbreviations: DNMT, DNA methyltransferase; FBS, Fetal bovine serum; HDAC, Histone deacetylase; IC50, Half maximal inhibitory concentration; HRP, Horse-
radish Peroxidase; PsA, Psammaplin A; PVDF, polyvinylidene fluoride; SDS, sodium dodecyl sulfate; SRB, Sulforhodamine B; TBS, Tris-buffered saline; TCA, tri-
chloroacetic acid; EMT, epithelial-mesenchymal transition; PAGE, polyacrylamide gel electrophoresis; H&E, hematoxylin and eosin.
* Corresponding authors.
E-mail addresses: hansy0574@snu.ac.kr (H. Ju Han), sky_magic@naver.com (W. Sub Byun), sklee61@snu.ac.kr (S. Kook Lee), hgpk@snu.ac.kr (H.-g. Park).
1
Co-first author
https://doi.org/10.1016/j.bmc.2021.116072
Received 19 January 2021; Received in revised form 3 February 2021; Accepted 4 February 2021
Available online 11 February 2021
0968-0896/© 2021 Elsevier Ltd. All rights reserved.

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
mesenchymal transition (EMT)-promoting genes, which is an essential
process for cancer cell invasion and metastasis9, suggesting that DOT1L
is a suitable molecular target for the development of antitumor and
antimetastasis therapeutics for aggressive breast cancer. Since the
DOT1L histone H3 methyltransferase has been recognized as a prom-
ising target for the development of novel anticancer agents, efforts to
identify effective DOT1L inhibitors have also become common.10,11 To
date, however, only a few nucleoside-type inhibitors have been reported
and no other classes of small molecules have been developed.12 (see Fig
1.).
As part of our ongoing search for novel antimetastatic cancer
chemotherapeutic agents targeting histone-modifying enzymes, we
recently developed very efficient synthetic methods for preparing
psammaplin A and its analogues, and evaluated their cytotoxicity.14
However, only a few compounds have exhibited antitumor activity
comparable to psammaplin A itself in both in vitro cell culture and in
vivo xenograft mouse models and poor antimetastatic activity (see Fig
2).
Selenium, the element with atomic number 34, belongs to the same
family as oxygen and sulfur in the periodic table. It is a trace element
present in the body as part of 21st amino acid, selenocysteine (SeC), but
it is essential for metabolism, and has been shown to have anticancer
effects.25 As oxidative stress is known to be associated with carcino-
genesis, the anticancer activity of selenium might be due to the anti-
oxidant properties of selenium-containing proteins (selenoproteins),
including proteins in the glutathione peroxidase family and selenopro-
tein P, which has been tested against bladder cancer, breast cancer,
prostate cancer, lung cancer and colon cancer.26–29 Selenium involved
organic compounds have also been reported to induce apoptosis of
cancer cells and to exhibit anticancer effects in prostate cancer, colo-
rectal cancer, liver cancer, and blood cancer.30 In addition, as selenium
has more lipophilic physicochemical properties than oxygen and sulfur,
its presence in organic molecules can result in better oral bioavailability
by better penetration across the cell membrane, including the blood–-
Figure 2. Structure of seleno-acyclovir (1), seleno-tenofovir (2), and seleno-
psammaplin A (9).
prepared compound 3 via the Knovenagel condensation followed by
reduction from ethyl acetoacetate and 3-bromo-4-hydroxybenzaldehyde
according to the previous method.13 The direct
α-nitrosation of 3 by
treatment with n-butylnitrite in basic condition of sodium ethoxide
under ethanol at 0 ◦C produced oxime 4 via
α-NO-substitution followed
by subsequent rearrangement.
Selective THP protection of 4 in the presence of catalytic amount of
p-toluenesulfonic acid in methylene chloride at room temperature
afforded 5 (95%). Then, the ethyl ester of 5 was hydrolyzed to corre-
sponding acid 6 with 1 N KOH in ethanol (99%). Activation of 6 by
coupling with N-hydroxyphthalimide using DCC in 1,4-dioxane afforded
7. Without purification, the addition of selenocystamine to 7 generated
dimer 8. Finally, selenopsammaplin A (9) could be obtained by THP
deprotection in acidic conditions (84%). Using the optimized synthetic
method, thirteen analogues of selenopsammaplin A (10–22) were pre-
pared from the corresponding aldehydes (Scheme 1). Before we per-
formed the evaluation their biological activities, we needed to confirm
the chemical stability of selenopsammaplin A and it’s analogues. Sele-
nopsammaplin A (9) was dissolved in a mixture of dimethylsulfoxide
and 0.1 M phosphate buffer (pH 7.0) (1:1). After 48 h at room temper-
ature, there was no significant changes, such as cleavage of diselenide
bond or hydrolysis of oxime groups.
brain barrier (BBB). Based on these findings, seleno-acyclovir (1)31
,
seleno-adefovir32, and seleno-tenofovir (2)33,34 have been successfully
reported as efficient anti-viral agents.
As part of our focus on the anticancer effects of organo-selenium, we
designed the novel selenopsammaplin A (9) with improved anticancer
activity by replacing the disulfide bond in psammaplin A with a dis-
elenide bond. Herein, we report a new synthetic method for accessing
selenopsammaplin A (9) and its analogues with a structure–activity
relationship by evaluating cytotoxicity and DOT1L inhibitory activity. In
addition, the in vivo antitumor activity and antimetastatic activity of a
selected compound was determined by employing an orthotopic mouse
metastasis models implanted with human breast cancer cells.
2. Results and discussion
2.2. Cytotoxic activity of selenopsammaplin a analogues and
structure–activity relationship
2.1. Chemistry
The cytotoxicity of selenopsammaplin A (9) and its analogues (10 –
22) were examined against a panel of human cancer cell lines. As shown
in Table 1, the analogues of selenopsammaplin A showed cytotoxicity up
to 6 – 60 times greater than that of psammaplin A itself depending on
cancer cells. As a result, the size of the aromatic group affects to cyto-
toxicity. Larger substituents result in lower cytotoxicity (entries 2, 3, 12,
10 > 11, 20; entries 9 – 10, 17 > 18). In the case of halides, there was no
significant difference in the cytotoxicity except for 4-fluoro analogue 12
(entries 4 – 8). However, the 3-fluoro-4-hydroxy analogue (entry 14, 22)
and the 3-chloro-4-hydroxy analogue (entry 13, 21) afforded the best
activities, and their activities were similar to that of selenopsammaplin
A (9). Notably, unsubstituted analogue 10 showed a higher cytotoxicity
than other substituted analogues except the 3-halide-4-hydroxy ana-
logues (9, 21, and 22). To further confirm whether the compounds
selectively suppress the cancer cells, the active compounds selenop-
sammaplin A (9) and analogue 10 were evaluated in MCF10A cells
First, the synthesis of selenopsammaplin A (9) was attempted by
modifying the synthetic route (Scheme 1) to psammplin A.13 We
Figure 1. Structure of psammaplin A15–24.
2

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
Scheme 1. Synthesis of selenopsammaplin A (9) and it’s analogues (10 – 22) from various aldehydes.
DOT1L enzyme activity, while psammaplin A (PsA) showed negligible
inhibition against DOT1L at a tested concentration (Figure 3A).
Table 1
Cytotoxicity of selenopsammaplin
A
(9) and its analogues (10
–
Since selenopsammaplin A (9) and some of its analogues exhibited
DOT1L inhibitory activity in cell-free biochemical assay, we further
assessed the expression level of H3K79 di-methylation (H3K79me2, the
biological target of DOT1L) by Western blotting analysis in MDA-MB-
231 cells. MDA-MB-231 cells are known to exhibit overexpressed
H3K79me2 due to the hyperactivation of DOT1L.35,36 Consistent with
the data for cell-free enzyme activity, selenopsammaplinA (9) and some
of its analogues (compound 10 and 17 – 22) significantly down-
regulated expression level of H3K79me2 (Figure 3B). The β-naphthyl
derivative (11) did not possess inhibitory activity, indicating that a
relatively larger R group decreases the DOT1L inhibitory activity.
Notably, analogues with only halide substituents (12 – 16) were less
active than analogues with other functional groups. The derivative with
a simple phenyl substituent at the R position (10) exhibited the most
potent inhibitory activity both in cell-free enzyme activity and DOT1L-
mediated H3K79me2 expression (Figure 3A and 3B). Moreover, to
confirm whether the inhibition of histone lysine residue methylation by
the compound 10 is selective on DOT1L-mediated H3K79 methylation,
the methylation status of other histone H3 lysine residues (K4, K27, K36)
were determined by Western blotting analysis. As shown in Figure 3C,
compound 10 exhibits remarkably selective inhibitory effects on H3
lysine 79 residues over other histone H3 lysine residues in MDA-MB-231
cells. However, the effect of psammaplin A (PsA) on the methylation of
histone H3 residues at a concentration of 200 nM was negligible
(Figure 3C). These findings suggest that selenopsammaplin A analogues
are considered promising non-nucleoside inhibitors of DOT1L.
22)a.
Entry
R
A
HCT
116c
MDA-
SK-
HEP-
1e
SNU
638f
549b
MB231d
1
3-Br-4-OH-Ph (9)
Ph (10)
0.03
0.08
0.38
0.10
0.33
0.25
0.28
0.31
0.05
0.14
0.09
0.10
0.02
0.02
1.76
0.30
0.01
0.09
0.12
0.11
0.18
0.28
0.52
0.17
0.07
0.10
0.09
0.13
0.01
0.01
0.61
1.20
0.16
0.06
0.19
0.13
0.42
0.33
0.25
0.39
0.20
0.14
0.09
0.10
0.17
0.19
1.31
8.70
0.21
0.12
0.17
0.14
0.27
0.33
0.32
0.31
0.18
0.28
0.17
0.19
0.11
0.13
1.29
0.40
0.02
0.05
0.27
0.07
0.05
0.22
0.18
0.15
0.03
0.41
0.06
0.28
0.01
0.02
0.56
0.20
2
3
β-naphthyl (11)
4-F-Ph (12)
4
5
3,4-F2-Ph (13)
4-Cl-Ph (14)
6
7
3,4-Cl2-Ph (15)
4-Br-Ph (16)
8
9
4-EtO-Ph (17)
4-BnO-Ph (18)
4-Nitro-Ph (19)
4-t-Bu-Ph (20)
3-Cl-4-OH-Ph (21)
3-F-4-OH-Ph (22)
Psammaplin A (PsA)
Etoposide
10
11
12
13
14
15
16
a
Results are expressed as the calculated half maximal inhibitory concentra-
tions (IC50) of test compounds (
μM). All values are the means of at least three
experiments.
b
c
Human lung cancer cells.
Human colon cancer cells.
d
Human breast cancer cells.
e
f
Human liver cancer cells.
Human stomach cancer cells.
2.4. Antimetastatic activity and regulation of EMT-associated biomarkers
by compound 10 in human breast cancer cells
(normal human breast epithelial cell line). The IC50 values of compound
9 and 10 were 16.14 μM and 18.11 μM, respectively, against MCF10A
It was reported that the hypermethylation of H3K79 is highly
correlated with the enhancement of breast cancer metastasis via mod-
ulation of EMT pathway.16,17 Therefore, the effect of 10 on metastatic
activity was evaluated by employing cell migration and invasion assays
in aggressive and highly metastatic MDA-MB-231 human breast cancer
cells. As depicted in Figures 4A and 4B, compared to vehicle-treated
control cells, compound 10 effectively suppressed wound closure (cell
migration) and invasion of cancer cell through Matrigel-coated invasion
chamber in a concentration-dependent manner, but psammaplin A (PsA)
was not capable to inhibit the migration and invasion of cancer cells at a
concentration of 400 nM. To further determine whether the mechanism
of the inhibitory activity of compound 10 toward cell migration and
cells which showed over 100-fold higher cytotoxicity against a panel of
cancer cells.
2.3. Histone H3 Lysine79 methyltransferase (DOT1L) inhibitory activity
of selenopsammaplin A analogues
A series of selenopsammaplin A (9) and its analogues were evaluated
in the DOT1L cell-free enzyme activity assay to identify non-nucleoside
small-molecule DOT1L inhibitors. Among the tested analogues, eight
compounds (9, 10, and 17 – 22) including selenopsammaplin A (9) at a
test concentration of 200 nM showed significant inhibitory effects on the
3

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
Figure 4. Inhibitory effects of compound 10 on the migration and invasion
potential of human breast cancer cells. (A) Monolayers of MDA-MB-231 cells
were scratched mechanically and treated with compound 10 for 24 h. Repre-
sentative images of wound closure obtained under a light microscope are shown
(left). The areas of the wounds were quantified using ImageJ (right). (B) The
cells were pretreated with compound 10 or PsA at the indicated concentrations
(50–400 nM) for 24 h, and cells were reseeded into the upper chambers of the
Transwell and incubated for 24 h. Invaded cells were fixed, stained, imaged
(left), and counted (right). (C) The MDA-MB-231 cells were treated with com-
pound 10 at the indicated concentrations (50–200 nM) for 24 h and the mRNA
levels of CDH1, CDH2, ZEB1, VIM were examined using real-time PCR. (D) The
cells were treated with 10 for 48 h, and then detected the expression levels of
EMT biomarkers by Western blotting analysis. β-Actin was used as an internal
control. The relative expression level of the proteins was semi-quantified using
NIH ImageJ software. Data are expressed as the mean values ± SD (n = 3) and
are representative of three separate experiments. *p < 0.05 and **p < 0.01
indicate significant differences relative to the vehicle-treated control group.
Figure 3. Effects of selenopsammaplin A analogues on DOT1L activity. (A)
DOT1L (500 ng/well) enzyme activity levels were analyzed after incubation
with selenopsammaplin A analogues (1 μM) for 2 h. EPZ-5676 (100 nM) was
used as a positive control. All data are expressed as mean values ± SD (n = 3)
and are representative of three separate experiments. *p < 0.05, **p < 0.01
indicates significant differences relative to the vehicle-treated control group.
(B) Inhibitory activity of selenopsammaplin A analogues toward the methyl-
ation of H3K79 in MDA-MB-231 breast cancer cells. (C) Selective inhibitory
effect on H3K79me2 by compound 10 in MDA-MB-231 cells. Data are expressed
as the mean values ± SD (n = 3) and are representative of three separate ex-
periments. *p < 0.05, **p < 0.01, and ***p < 0.001 indicates significant dif-
ferences relative to the vehicle-treated control group.
partly associated with the modulation of the EMT processes mediated by
DOT1L in breast cancer cells.
2.5. Docking of 10 and DOT1L
The binding interactions between the most effective analogue 10 and
DOT1L were evaluated by molecular docking simulations. As shown in
Figure 5, the majority of the surface area of the cofactor (S-adenosyl-L-
methionine, SAM) binding site of DOT1L is buried inside the protein.
The docked pose of 10 with the highest binding score was overlapped
well with the original ligand (TT8) in the active site (Figure 5A). One
phenyl group of 10 forms pi-pi and pi-alkyl interactions with Phe223,
Pro133, Leu224, and Val249 in the hydrophobic cavity of DOT1L. The
hydroxyl group (OH) on the adjacent oxime forms a hydrogen bond with
the NH of Gly163, whereas the NH of the neighboring amide group of
invasion is associated with the EMT gene regulation, the expression of
major EMT biomarkers was analyzed in MDA-MB-231 cells. Compound
10 effectively upregulates the expression of the epithelial marker CDH1
(E-cadherin) and downregulates the mesenchymal markers, CDH2 (N-
cadherin), ZEB1, and VIM (vimentin), at the mRNA level (Figure 4C) and
protein level (Figure 4D). Therefore, the present data indicate that the
inhibition of cancer cell migration and invasion by compound 10 is
4

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
Figure 5. Predicted binding mode of 10 from the docking analysis. (A) The
overlaid structures of 10 (pink) with the original ligand (TT8, green) docked
into the X-ray crystal structures of the human DOT1L (PDB code: 3SR4). (B)
Docked pose of 10 with the highest docking score, demonstrating hydrogen
bonds (green) and hydrophobic interactions (pink). (C) Interaction diagram of
10 with the DOT1L key residues.
the ligand forms a hydrogen bond with the carboxylate side chain of
Glu186. The NH of the other amide forms a hydrogen bond with the
carbonyl oxygen of Gly163, whereas the carbonyl oxygen of the amide is
hydrogen bonded with the NH of Gly137. The OH of the next oxime
forms a hydrogen bond with the carbonyl group in the side chain of
Gln168. The nearby phenyl group forms a pi-sigma hydrophobic inter-
action with Asn241 inside the pocket. Note that Phe223, Asp222,
Glu186, Lys187, Gln168, and Gly163 are critical residues for the specific
recognition between the original cofactor (SAM) and DOT1L.37
2.6. Effects of compound 10 on tumor growth and metastatic potential in
mouse metastasis model
The antitumor and antimetastatic activities of compound 10 were
determined using an in vivo orthotopic mouse metastasis model
implanted with luciferase-expressing MDA-MB-231 cells. The cells were
implanted onto the fourth fat pad of the mice, and after the tumor size
reached approximately 200 mm3, compound 10 (5 or 15 mg/kg),
psammaplin A (PsA, 30 mg/kg), or paclitaxel (5 mg/kg) was intraperi-
toneally administered to each mouse thrice per week for 38 days.
Compared to the control groups, compound 10 significantly inhibited
tumor growth, and at the end of the experiments, the inhibition rates
were 45.5% and 55.1% at 5 mg/kg and 15 mg/kg, respectively, for
compound 10 (Figure 6A). The excised tumor weights measured on the
termination day of the experiment were also significantly lower after
treatment with compound 10 compared to those of control groups
(Figure 6B). No body weight changes or overt toxicity were found in the
in vivo experiments with compound 10, while body weight loss was
moderately observed in the group of paclitaxel treatment (Figure 6C).
Tumor growth and incidence of metastasis were also determined by
using IVIS to confirm the antitumor and antimetastatic potential of
compound 10. It was found that the luciferase signals were decreased by
compound 10 treatment at the primary tumor sites (fat pad of the
mouse), and the signal intensity in excised lung tissues was also effec-
tively decreased by compound 10 treatment (Figure 6D and 6E).
Moreover, the incidence of metastatic cells in the lung was mani-
festly suppressed by compound 10 compared to the vehicle-treated
groups (Figure 7A). These findings indicate that the primary tumor
growth and lung metastasis was effectively inhibited by compound 10 in
in vivo animal models. To further confirm whether the antitumor and
antimetastatic activities of compound 10 are correlated with the inhi-
bition of DOT1L-mediated H3K79 methylation and the regulation of
Figure 6. Antitumor and antimetastatic activities of compound 10 in an MDA-
MB-231-bearing orthotopic mouse metastasis model. (A) Tumor volumes in the
MDA-MB-231 orthotopic mouse metastasis model. Vehicle, compound 10, PsA,
or paclitaxel were administered thrice per week for 38 days. Tumor volumes
were measured every 3 days with electronic caliper. (B) Primary tumors from
mouse fat pad were excised from mice at the termination of the experiment, and
tumor weights were measured. *p < 0.05, **p < 0.01, and ***p < 0.001 indi-
cate significant differences relative to the vehicle-treated control group. (C)
Body weights of the mice were measured once a week. (D) Luciferase imaging
(IVIS) of each group of the mice was detected on the last day of the experiment.
(E) Organs (lungs, livers, spleens, and kidneys) of mice were excised from mice
at the termination of experiment, and fluorescence intensity was imaged by
IVIS. Control data (vehicle- and paclitaxel-treated controls) were shared with
previously reported in Mol Ther Oncolytics 15 (2019) 140.35.
EMT markers, additional biochemical analyses of excised tumor tissues
were performed by Western blotting. Compound 10 remarkably down-
regulated the expressions of H3K79me2 and vimentin and upregulated
the expression of E-cadherin in excised tumor tissues in a dose-
dependent manner (Figure 7B). In addition, compound 10 also down-
regulated the expression levels of Ki67, a biomarker of cell proliferation,
vimentin, and H3K79me2, while compound 10 upregulated the
5

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
from the corresponding aldehydes by employing a Knoevenagel
condensation, an
α-nitrosation, and a dimerization with selenocyst-
amine as the key steps. A structure–activity relationship study with the
prepared selenopsammaplin A analogues was performed to evaluate
their cytotoxicity in a panel of cancer cells and the inhibition of DOT1L
in aggressive human breast cancer cells. All of the selenopsammaplin A
analogues showed cytotoxicities greater than that of psammaplin A, and
some of the analogues significantly inhibited DOT1L. Representative
compound 10 exhibited an excellent cytotoxicity and ability to inhibit
DOT1L in vitro. Antitumor and antimetastatic activities were also found
in an in vivo mouse metastasis model involving the implantation of
human breast cancer cells. The underlying molecular mechanism of the
antimetastatic activity of compound 10 might be through the regulation
of the EMT process by the inhibition of DOT1L-mediated H3K79
methylation. Taken together, these findings give an information with
the antitumor and antimetastatic activities of novel selenopsammaplin A
analogues via regulation of histone methylation in breast cancer cells,
and DOT1L may be a promising target in the development of anticancer
agents for metastatic breast cancers.
4. Experimental section
4.1. Chemistry
All reagents purchased from commercial sources were used without
further purification. TLC analyses were performed using pre-coated TLC
plate (silica gel 60 GF254, 0.25 mm). Flash column chromatography was
performed on flash silica gel 230 ~ 400 mesh size. Infrared analyses
(KBr pellet) were performed by FT-IR. 1H NMR spectra was recorded at
300 MHz, 400 MHz, 500 MHz or 800 MHz with reference to CDCl3 (δ
7.24), CD3OD (δ 3.31), or DMSO-d6 (δ 2.49). 13C NMR spectra was ob-
tained by 75 MHz, 100 MHz, 125 MHz or 200 MHz spectrometer relative
to the central CDCl3 (δ 77.0), CD3OD (δ 49.0) or DMSO-d6 (δ 39.5)
resonance. Coupling constants (J) in 1H NMR are in Hz. Low-resolution
mass spectra (LRMS) and high-resolution mass spectra (HRMS) were
measured on positive-ion FAB spectrometer. Melting points were
measured on melting point apparatus and were uncorrected.
4.2. General procedure for the synthesis of selenopsammaplin a (9) and
its analogues (10 ~ 22)
Figure 7. Compound 10 suppresses tumor growth and lung metastasis by
regulating DOT1L-mediated H3K79 methylation (A) Representative hematox-
ylin and eosin staining of excised lungs in the orthotopically implanted mouse
metastasis model; arrow, metastasized cells. (B) Small portions of tumors from
each group were homogenized in complete lysis buffer (active motif). The
expression levels of indicated proteins were determined by western blotting
analysis using antibodies against H3K79me2, E-cadherin, and vimentin. β-Actin
was used as an internal control. The relative intensity of indicated proteins was
semiquantified using the NIH ImageJ software. (C) Immunohistochemical
analysis was performed for the detection of Ki67, H3K79me2, E-cadherin, and
Vimentin. MDA-MB-231/Luc tumor sections were counterstained with hema-
toxylin and imaged with Vectra 3.0 Automated Quantitative Pathology Imaging
System (PerkinElmer, 200 × ). All data are representative of three separate
experiments. *p < 0.05, **p < 0.01, and ***p < 0.001 indicate significant
differences relative to the vehicle-treated control group.
(E)-Ethyl 3-(3-bromo-4-hydroxyphenyl)-2-hydroxyimino)pro panoate
(4). Sodium ethoxide (313 mg, 4.6 mmol) was added to solution of 3
(630 mg, 2 mmol) in ethanol (10 mL) at 0 ◦C. n-Butyl nitrite (257
μL, 2.2
mmol) was added and the mixture was stirred for 3 h at rt. After
completion of the reaction, the ethanol was evaporated and the residue
was diluted with ethyl acetate (200 mL), washed with 1 N-HCl (50 mL)
and brine (50 mL) in order, dried over anhydrous MgSO4, filtered, and
concentrated in vacuo. The residue was purified by column chromatog-
raphy (silica gel, hexane : ethyl acetate = 6 ~ 3 : 1) to afford 4 (435 mg,
72% yield) as a white solid. m.p. = 87 ◦C; 1H NMR (300 MHz, CD3OD) δ
= 7.35 (d, J = 2.01 Hz, 1H), 7.05 (dd, J1 = 8.42 Hz, J2 = 2.04 Hz, 1H),
6.77 (d, J = 8.25 Hz, 1H), 4.21 (q, J = 6.96 Hz, 2H), 3.80 (s, 2H), 1.25 (t,
J = 7.14 Hz, 3H) ppm; 13C NMR (100 MHz, CD3OD) δ = 166.1, 154.6,
152.7, 135.2, 131.1, 131.0, 117.9, 111.3, 63.4, 30.7, 15.2 ppm; FT/IR =
3347, 1720, 1493, 1420, 1288, 1203, 1023, 858, 802, 759, 725 cmꢀ 1
;
expression of E-cadherin (Figure 7C). These findings suggest that the
antitumor and antimetastatic activities of compound 10 in breast cancer
cells might be associated in part with the regulation of DOT1L-mediated
H3K79 methylation and the EMT processes.
HRMS (FAB) calcd for [C11H13BrNO4]+ 302.0028, found: 302.0035. The
spectral data were identical with previous results.13
(E)-Ethyl 3-(3-bromo-4-hydroxyphenyl)-2-{[(tetrahydro-2H-pyran-2-
yl)oxy]imino}propanoate (5); p-Toluenesulfonic acid monohydrate (25
mg, 0.1 mmol) and 3,4-dihydro-2H-pyran (DHP, 218 mg, 2.6 mmol)
were added to a solution of 4 (400 mg, 1.3 mmol) in dichloromethane (6
mL) at 0 ◦C. The mixture was stirred at rt for 6 h. After completion of the
reaction, a few drops of methanol was added to the reaction mixture and
stirred for 10 min. The reaction mixture was diluted with dichloro-
methane (150 mL), washed with aq.-NH4Cl (30 mL) and brine (50 mL),
3. Conclusion
The present study reports that we designed and synthetized the novel
selenopsammaplin A and its analogues by replacing the disulfide bond in
psammaplin A with a diselenide bond. We prepared fourteen analogues
6

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
dried over anhydrous MgSO4, filtered, and concentrated in vacuo. The
residue was purified by column chromatography (silica gel, hexane :
ethyl acetate = 4 ~ 2 : 1) to afford 5 (478 mg, 94% yield) as a yellow oil.
1H NMR (300 MHz, CD3OD) (E/Z form) : δ 7.33 (d, J = 2.01 Hz, 1H),
7.00 (dd, J = 8.34, 2.19 Hz, 1H), 6.76 (d, J = 8.22 Hz, 1H), 6.64 (s, 1H),
4.15 (m, 2H), 3.76 (q, J = 13.56 Hz, 2H), 3.53 (m, 2H), 1.73 (br, 3H),
1.51 (br, 3H), 1.16 (t, J = 7.14 Hz, 3H) ppm; 13C NMR (100 MHz,
CD3OD): δ 163.048, 151.388, 151.246, 132.713, 129.134, 128.512,
115.922, 109.390, 101.386, 62.152, 61.757, 30.072, 28.051, 24.509,
18.639, 13.621 ppm; FT-IR (KBr) 3412, 2946, 2871, 2738, 2565, 1892,
1718, 1609, 1579, 1508, 1494, 1466, 1454, 1441, 1420, 1393, 1373,
1354, 1334, 1288, 1260, 1203, 1154, 1112, 1062, 1040, 1020, 966, 927,
901, 859, 817, 802, 756, 729, 673, 627 cmꢀ 1; HRMS (FAB+) calcd for
[C16H21BrNO5]+ = 386.0603, found = 386.0598. The spectral data were
identical with previous results.13
ethyl acetate (50 mL), washed with water (10 mL × 3) and brine (10 mL)
in order, dried over anhydrous MgSO4, filtered, and concentrated in
vacuo. The residue was purified by column chromatography (silica gel,
hexane : ethyl acetate = 3 ~ 1 : 1) to afford selenopsammaplin A (9)
(103 mg, 84% yield) as a yellow solid. m.p. = 72 ◦C; 1H NMR(800 MHz,
DMSOd6) δ 11.84 (s, 2H), 10.02 (s, 2H), 8.10 (t, J = 5.88 Hz, 2H), 7.29
(s, 2H), 7.01 (dd, J = 8.28, 1.16 Hz, 2H), 6.83 (d, J = 8.32 Hz, 2H), 3.45
(q, J = 6.61 Hz, 4H), 3.01 (t, J = 7.08 Hz, 4H) ppm; 13C NMR(200 MHz,
DMSOd6) δ 163.1, 152.3, 151.8, 132.8, 129.1, 128.8, 116.1, 108.8, 39.8,
28.2, 27.7 ppm; FT/IR = 3368, 2925, 2853, 2378, 2310, 1712, 1657,
1530, 1492, 1423, 1362, 1280, 1205, 1044, 1007,967, 781, 763, 748,
670 cmꢀ 1
;
HRMS (FAB): m/z [M
+
Na]+ calcd for
[C22H24Br2N4O6Se2Na]+ = 780.8291, found = 780.8281.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[2-(hydroxyl
imino)-3-phenylpropanamide] (10). Following the general procedure of
selenopsammaplin A (9) from benzaldehyde, 10 was obtained as a yel-
low solid (overall 27% yield). m.p. = 139 ◦C; 1H NMR(800 MHz,
CD3OD) δ 7.24 (d, J = 7.68 Hz, 4H), 7.19 (t, J = 7.68 Hz, 4H), 7.12 (t, J
= 7.36 Hz, 2H), 3.90 (s, 4H), 3.53 (t, J = 6.92 Hz, 4H), 2.99 (t, J = 6.92
Hz, 4H) ppm; 13C NMR(200 MHz, CD3OD) δ 166.7, 154.0, 138.9, 130.9,
128.0, 41.9, 30.7, 30.1 ppm; FT/IR = 3841, 3734, 3680, 2973, 2381,
2359, 2349, 2308, 1658, 1648, 1525, 1491, 1424, 1362, 1219, 1054,
1033, 1012, 772, 671, 649 cmꢀ 1; HRMS (FAB): m/z [M + H]+ calcd for
[C22H27N4O4Se2]+ 571.0363, found: 571.0363.
(E)-3-(3-bromo-4-hydroxyphenyl)-2-{[(tetrahydro-2H-pyran-2-yl)oxy]
imino}propanoic acid (6). Compound 5 (450 mg, 1.2 mmol) was added to
1 M KOH solution (3.5 mL) in ethanol. The mixture was stirred for 4 h
until no more 5 was observed by TLC analysis. If the reaction was not
completed, the mixture was supplement by water (1 mL). After
completion of the reaction, the reaction mixture was diluted with ethyl
acetate (30 mL) and extracted with 0.2 N NaOH (15 mL × 3). The
aqueous phase was acidified until pH 4 using 1 N HCl in ice-water bath.
Then extracted by ethyl acetate (50 mL × 3), washed with brine, dried
over anhydrous MgSO4, filtered, and concentrated in vacuo. The residue
6 (413 mg, 99% yield) was obtained as a pale yellow solid. m.p. =
137 ◦C; 1H NMR (300 MHz, CDCl3) : δ 7.42 (d, J = 2.12 Hz, 1H), 7.12
(dd, J = 8.40, 2.12 Hz, 1H), 6.87 (d, J = 8.40 Hz, 1H), 5.45 (m, 1H), 3.83
(s, 2H), 3.60 (m, 2H), 1.85 (m, 3H), 1.63 (m, 3H), ppm; 13C NMR (100
MHz, CDCl3) : δ 163.53, 151.23, 150.79, 132.80, 129.93, 128.65,
116.15, 109.94, 101.80, 62.25, 29.49, 28.00, 24.73, 18.47 ppm; FT-IR
(KBr) 3343, 3018, 2949, 2874, 1916, 1818, 1728, 1606, 1580, 1506,
1494, 1454, 1421, 1373, 1319, 1287, 1259, 1207, 1185, 1111, 1070,
1035, 989, 967, 927, 899, 870, 835, 806, 755, 718, 672 cmꢀ 1; HRMS
(FAB+) calcd for [C14H17BrNO5]+ = 358.0290, found = 358.0272. The
spectral data were identical with previous results.13
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[2-(hydroxy imino)-
3-(naphthalen-2-yl)propanamide] (11). Following the general proced-
ure of selenopsammaplin A (9) from 4-β-naphthylbenzaldehyde, 11 was
obtained as a yellow solid (overall 29% yield). m.p. = 164 ◦C; 1H NMR
(800 MHz, DMSOd6) δ 11.93 (s, 1H), 8.15 (t, J = 5.84 Hz, 1H), 7.82 (m,
2H), 7.80 (m, 4H), 7.67 (s, 2H), 7.45 (m, 4H), 7.40 (d, J = 8.48 Hz, 2H),
3.98 (s, 4H), 3.46 (t, J = 6.68H, 4H), 3.01 (t, J = 7.08 Hz, 4H) ppm; 13
C
NMR(200 MHz, DMSOd6) δ 163.2, 151.7, 134.5, 133.0, 131.6, 127.7,
127.5, 127.4, 127.3, 126.8, 126.0, 125.4, 29.2, 28.2, 28.2 ppm; FT/IR =
3368, 2922, 2851, 2349, 1657, 1619, 1529, 1452, 1362, 1236, 1189,
1013, 965, 847, 776, 764, 750, 721, 645, 615 cmꢀ 1; HRMS (FAB): m/z
[M + H]+ calcd for [C30H31N4O4Se2]+ 671.0676, found: 671.0673.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(4-fluoro
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(3-bromo-4-
hydroxyphenyl)-2-{[(tetrahydro-2H-pyran-2-yl)oxy]imino} propanamide]
(8). N,N’-Dicyclohexylcarbodiimide (364 mg, 1.9 mmol) and N-
hydroxyphthalimide (346 mg, 2.1 mmol) were added to a solution of 6
(400 mg, 1.1 mmol) in 1,4-dioxane (5 mL). The mixture was stirred for 2
h until no more 6 was observed by TLC analysis, and diluted with ethyl
acetate (50 mL). The mixture was washed with water (25 mL) and brine
(25 mL) in order, dried over anhydrous MgSO4, filtered, and concen-
trated in vacuo. Without further purification, the residue (7) was dis-
solved in 1,4-dioxane (5 mL). Selenocystamine dihydrochloride (178
mg, 0.56 mmol) was dissolved in methanol (2.5 mL) with triethylamine
phenyl)-2-(hydroxyimino)propanamide] (12). Following the general
procedure of selenopsammaplin A (9) from 4-fluorobenzaldehyde, 12
was obtained as a yellow solid (overall 26% yield). m.p. = 147 ◦C; 1H
NMR(800 MHz, CD3OD) δ 7.26 (m, 4H), 6.92 (t, J = 8.8 Hz, 4H), 3.87 (s,
4H), 3.54 (t, J = 6.96 Hz, 4H), 3.01 (t, J = 6.96 Hz, 4H) ppm; 13C NMR
(200 MHz, CD3OD) δ 166.5, 164.3, 163.1, 153.9, 134.9, 132.6, 116.6,
41.9, 30.1 ppm; FT/IR = 3842, 3735, 3272, 2924, 2348, 2310, 1706,
1648, 1621, 1526, 1508, 1424, 1357, 1281, 1219, 1091, 1046, 1010,
995, 814, 772, 741, 689 cmꢀ 1; HRMS (FAB): m/z [M + H]+ calcd for
[C22H25F2N4O4Se2]+ 607.0174, found: 607.0167.
(311
μL, 2.23 mmol). This solution was added to the reaction mixture
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(3,4-di
and stirred for 4 h. After completion of the reaction, the solvent was
evaporated and the residue was diluted with ethyl acetate (100 mL),
washed with water (25 mL) and brine (25 mL) in order, dried over
anhydrous MgSO4, filtered, and concentrated in vacuo. The residue was
purified by column chromatography (silica gel, hexane : acetone = 3 : 1)
to afford compound 8 (377 mg, 74% yield) as a white solid. m.p. =
144 ◦C; 1H NMR (300 MHz, CD3OD) δ 7.42 (d, J = 2.22 Hz, 2H), 7.07 (m,
2H), 6.77 (d, J = 8.43 Hz, 2H), 5.37 (m, 2H), 3.82 (q, J = 12.64 Hz, 4H),
3.57 (m, 8H), 3.05 (t, J = 6.87 Hz, 4H) ppm; 13C NMR (200 MHz,
CD3OD) δ 165.6, 155.7, 154.8, 135.6, 131.2, 130.8, 117.9, 111.4, 103.2,
102.1, 63.9, 63.8, 56.1, 42.3, 32.4, 30.7, 30.2, 27.3, 26.9, 21.2, 20.5
ppm; FT/IR = 3371, 1658, 1534, 1493, 1423, 1358, 1285, 1209, 1017,
fluorophenyl)-2-(hydroxyimino)propanamide] (13). Following the gen-
eral procedure of selenopsammaplin A (9) from 3,4-difluorobenzalde-
hyde, 13 was obtained as a yellow solid (overall 20% yield). m.p. =
114 ◦C; 1H NMR(800 MHz, CD3OD) δ 7.15 (m, 2H), 7.09 (m, 2H), 7.05
(m, 2H), 3.86 (d, J = 5.36 Hz, 4H), 3.56 (t, J = 6.92 Hz, 4H), 3.02 (t, J =
6.96 Hz, 4H) ppm; 13C NMR(200 MHz, CD3OD) δ 166.3, 153.3, 152.6,
151.5, 150.4, 136.4, 127.4, 119.8, 118.7, 41.9, 30.0 ppm; FT/IR = 3840,
3735, 3269, 2923, 2348, 2310, 1648, 1608, 1516, 1425, 1283, 1219,
1115, 1048, 1010, 994, 875, 816, 772, 713, 630 cmꢀ 1; HRMS (FAB): m/z
[M + H]+ calcd for [C22H23F4N4O4Se2]+ 642.9986, found: 642.9969.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(4-chloro
phenyl)-2-(hydroxyimino)propanamide] (14). Following the general
procedure of selenopsammaplin A (9) from 4-chlorobenzaldehyde, 14
983, 801, 720 mꢀ 1
;
HRMS (FAB): m/z [M + H]+ calcd for
[C32H41Br2N4O8Se2]+ 925.9543, found: 925.9547.
was obtained as a yellow solid (overall 23% yield). m.p. = 61 C; H
NMR(800 MHz, CD3OD) δ 7.24 (m, 8H), 3.87(s, 4H), 3.54 (t, J = 6.96
Hz, 4H), 3.00 (t, J = 6.96 Hz, 4H) ppm; 13C NMR(200 MHz, CD3OD) δ
166.5, 153.6, 137.8, 133.8, 132.5, 130.1, 41.9, 30.1, 30.1 ppm; FT/IR =
3840, 3735, 3284, 2925, 2854, 2372, 2310, 1748, 1658, 1527, 1489,
1
◦
Synthesis of Selenopsammaplin A (9). p-Toluenesulfonic acid mono-
hydrate (6.2 mg, 0.03 mmol) was added to a solution of 8 (150 mg, 0.2
mmol) in methanol (2 mL). The mixture was refluxed for 19 h and
evaporated in vacuo to remove methanol. The residue was diluted with
7

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
1426, 1360, 1204, 1090, 1015, 968, 783, 762 cmꢀ 1; HRMS (FAB): m/z
[M + H]+ calcd for [C22H25Cl2N4O4Se2]+ 638.9583, found: 638.9586.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(3,4-di
NMR(800 MHz, DMSOd6) δ 11.79 (s, 2H), 8.07 (t, J = 5.88 Hz, 2H), 7.25
(d, J = 8.40H, 4H), 7.11 (d, J = 8.16 Hz, 4H), 3.77 (s, 4H), 3.45 (q, J =
6.69 Hz, 4H), 3.01 (t, J = 7.12 Hz, 4H) ppm; 13C NMR(200 MHz,
DMSOd6) δ 163.2, 151.9, 148.3, 133.6, 128.4, 125.0, 39.8, 34.0, 31.1,
28.4, 28.2 ppm; FT/IR = 3285, 3056, 2962, 2867, 1910, 1659, 1627,
1529, 1460, 1428, 1362, 1268, 1213, 1140, 1109, 1010, 968, 917, 859,
835, 811, 756, 707, 666 cmꢀ 1; HRMS (FAB): m/z [M + H]+ calcd for
[C30H43N4O4Se2]+ 683.1615, found: 683.1614.
chlorophenyl)-2-(hydroxyimino)propanamide] (15). Following the gen-
eral procedure of selenopsammaplin A (9) from 3,4-dichlorobenzalde-
hyde, 15 was obtained as a yellow solid (overall 17% yield). m.p. =
61 ◦C; 1H NMR(800 MHz, CD3OD) δ 7.41 (d, J = 2.00 Hz, 2H), 7.35 (d, J
= 8.24 Hz, 2H), 7.19 (dd, J = 8.32, 2.00, 2H), 3.87 (s, 4H), 3.55 (t, J =
6.92 Hz, 4H), 3.02 (t, J = 6.92 Hz, 4H) ppm; 13C NMR(800 MHz, CD3OD)
δ 166.2, 153.0, 139.8, 133.8, 132.9, 132.1, 131.9, 131.0, 41.9, 30.1,
30.0 ppm; FT/IR = 3294, 2924, 2853, 2349, 2319, 1748, 1659, 1528,
1470, 1426, 1397, 1360, 1274, 1202, 1132, 1031, 1009, 969, 876, 817,
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(3-chloro-4-
hydroxyphenyl)-2-(hydroxyimino)propanamide] (21). Following the
general procedure of selenopsammaplin A (9) from 3-chloro-4-hydroxy-
benzaldehyde, 21 was obtained as a pale yellow solid (overall 25%
yield). m.p. = 132 ◦C; 1H NMR(800 MHz, CD3OD) δ 7.19 (d, J = 2.16 Hz,
2H),7.02 (dd, J = 8.32, 2.08 Hz, 2H), 6.76 (d, J = 8.32 Hz, 2H), 3.78 (s,
4H), 3.54 (t, J = 6.92 Hz, 4H), 3.01 (t, J = 6.92 Hz, 4H) ppm; 13C NMR
(200 MHz, CD3OD) δ 166.6, 154.0, 153.4, 132.2, 131.1, 130.5, 122.1,
118.2, 41.9, 30.1, 29.6 ppm; FT/IR = 3286, 2923, 2852, 2320, 1658,
1526, 1424, 1345, 1279, 1227, 1211, 1038, 1027, 789, 781, 770, 761,
781, 719 cmꢀ 1
;
HRMS (FAB): m/z [M
+
H]+ calcd for
[C22H23Cl4N4O4Se2]+ 706.8804, found: 706.8806.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(4-bromo
phenyl)-2-(hydroxyimino)propanamide] (16). Following the general
procedure of selenopsammaplin A (9) from 4-bromobenzaldehyde, 16
1
◦
was obtained as a yellow solid (overall 16% yield). m.p. = 68 C; H
NMR(800 MHz, CD3OD) δ 7.35 (d, J = 10.8 Hz, 4H), 7.18 (d, J = 8.48
Hz, 4H), 3.86 (s, 4H), 3.54 (t, J = 6.96 Hz, 4H), 3.00 (t, J = 6.92 Hz, 4H)
ppm; 13C NMR(200 MHz, CD3OD) δ 166.4, 153.5, 138.3, 133.2, 132.9,
121.7, 41.9, 30.2, 30.1 ppm; FT/IR = 3284, 3054, 2925, 2854, 2310,
1657, 1527, 1486, 1427, 1403, 1357, 1292, 1208, 1138, 1098, 1071,
1011, 967, 917, 856, 800, 754 cmꢀ 1; HRMS (FAB): m/z [M + H]+ calcd
for [C22H25Br2N4O4Se2]+ 726.8573, found: 726.8575.
711, 665 cmꢀ 1
;
HRMS (FAB): m/z [M
+
H]+ calcd for
[C22H25Cl2N4O6Se2]+ 670.9482, found: 670.9482.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(3-fluoro-4-
hydroxyphenyl)-2-(hydroxyimino)propanamide] (22). Following the
general procedure of selenopsammaplin A (9) from 3-fluoro-4-hydroxy-
benzaldehyde, 22 was obtained as a yellow solid (overall 19% yield). m.
p. = 191 ◦C; 1H NMR(800 MHz, CD3OD) δ 6.95 (dd, J = 12.16, 2.00 Hz,
2H), 6.87 (dd, J = 8.20, 1.28 Hz, 2H), 6.75 (t, J = 8.72 Hz, 2H), 3.79 (s,
4H), 3.55 (t, J = 6.96 Hz, 4H), 3.01 (t, J = 6.96 Hz, 4H) ppm; 13C NMR
(200 MHz, CD3OD) δ 166.6, 154.0, 152.8, 145.2, 130.6, 126.9, 119.2,
118.4, 41.9, 30.1, 29.7 ppm; FT/IR = 3757, 3317, 2320, 1658, 1517,
1440, 1360, 1288, 1236, 1198, 1109, 1038, 1002, 970, 880, 800, 781,
773, 702, 666 cmꢀ 1; HRMS (FAB): m/z [M + H]+ calcd for
[C22H25F2N4O6Se2]+ 639.0073, found: 639.0074.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(4-ethoxy
phenyl)-2-(hydroxyimino)propanamide] (17). Following the general
procedure of selenopsammaplin A (9) from 4-ethoxybenzaldehyde, 17
was obtained as a yellow solid (overall 22% yield). m.p. = 146 ◦C; 1H
NMR(800 MHz, CD3OD) δ 11.77 (s, 2H), 8.06 (t, J = 5.88 Hz, 2H), 7.10
(d, J = 8.64 Hz, 4H), 6.78 (m, 4H), 3.95 (q, J = 6.99 Hz, 4H), 3.72 (s,
4H), 3.44 (q, J = 6.72 Hz, 4H), 3.00 (t, J = 7.12 Hz, 4H), 1.28 (t, J =
1.28 Hz, 6H) ppm; 13C NMR(200 MHz, CD3OD) δ 163.2, 156.8, 152.1,
129.8, 128.5, 114.1, 62.8, 28.2. 28.0, 14.6 ppm; FT/IR = 3388, 3280,
2923, 2320, 1649, 1621, 1528, 1509, 1477, 1427, 1288, 1250, 1182,
1114, 1050, 1010, 996, 808, 693 cmꢀ 1; HRMS (FAB): m/z [M + H]+
calcd for [C26H35N4O6Se2]+ 659.0887, found: 659.0886.
4.3. Cell culture
Human cancer cells (A549, HCT116, MDA-MB-231, SK-HEP-1, and
SNU-638) and human breast epithelial cells (MCF10A) were obtained
from the American Type Culture Collection (Manassas, VA, USA) or
Korean Cell Line Bank (Seoul, Korea). The cells were cultured in an
appropriate medium (Dulbecco’s modified Eagle’s medium for MDA-
MB-231 and SK-HEP-1; Roswell Park Memorial Institute 1640 for
A549, HCT116 and SNU-638 cells; Dulbecco’s modified Eagle’s me-
dium: Nutrient Mixture F-12 for MCF-10A) supplemented with
antibiotics-antimycotics (PSF: 100 units/mL sodium penicillin G, 100
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(4-(benzyl oxy)
phenyl)-2-(hydroxyimino)propanamide] (18). Following the general
procedure of selenopsammaplin A (9) from 4-benzyloxybenzaldehyde,
18 was obtained as a yellow solid (overall 27% yield). m.p. = 163 ◦C;
1H NMR(800 MHz, DMSOd6) δ 11.78 (s, 2H), 8.07 (t, J = 5.88 Hz, 2H),
7.41 (d, J = 7.20 Hz, 4H), 7.37 (t, J = 7.60 Hz, 4H), 7.31 (t, J = 7.32 Hz,
2H), 7.11 (d, J = 8.64 Hz, 4H), 6.89 (d, J = 11.36 Hz, 4H), 5.03 (s, 4H),
3.73 (s, 4H), 3.45 (q, J = 6.69 Hz, 4H), 3.00 (t, J = 7.12 Hz, 4H); 13
C
μ
g/mL streptomycin, and 250 ng/mL amphotericin B) and 10% fetal
◦
NMR(200 MHz, DMSOd6) δ 163.2, 156.7, 152.1, 137.2, 129.8, 128.9,
128.4, 127.7, 127.6, 114.6, 69.1, 28.2, 28.0 ppm; FT/IR = 3381, 3224,
3033, 2924, 2854, 1650, 1620, 1528, 1509, 1454, 1426, 1379, 1359,
bovine serum (FBS) in an incubator containing 5% CO2 at 37 C. All
reagents were purchased from Gibco® Invitrogen Corp. (Grand Island,
NY, USA).
1292, 1246, 1176, 1105, 1076, 1040, 1007, 795, 728, 693, 627 cmꢀ 1
;
HRMS (FAB): m/z [M + H]+ calcd for [C36H39N4O6Se2]+ 783.1200,
found: 783.1196.
4.4. Cell proliferation assay
Cell proliferation was measured by Sulforhodamine B (SRB) assay.38
Briefly, cells were seeded in 96-well plates and incubated for 30 min (for
zero day controls) or treated with test compounds for the indicated
times. After incubation, cells were fixed, dried and stained with 0.4%
SRB in 1% acetic acid solution. Unbound dye was washed off and stained
cells were dissolved in 10 mM Tris (pH 10.0). Absorbance was measured
at 515 nm, and cell proliferation was determined. IC50 values were
calculated by non-linear regression analysis using TableCurve 2D v5.01
software (Systant Software Inc., Richmond, CA, USA). All reagents were
purchased from Sigma-Aldrich.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[2-(hydroxy imino)-
3-(4-nitrophenyl)propanamide] (19). Following the general procedure
of selenopsammaplin A (9) from 4-nitrobenzaldehyde, 19 was obtained
as a yellow solid (overall 22% yield). m.p. = 70 ◦C; 1H NMR(800 MHz,
CD3OD) δ 8.09 (d, J = 8.72 Hz, 4H), 7.49 (d, J = 8.80 Hz, 4H), 4.02 (s,
4H), 3.55 (t, J = 6.92 Hz, 4H), 3.02 (t, J = 6.92 Hz, 4H) ppm; 13C NMR
(200 MHz, CD3OD) δ 166.2, 152.6, 148.7, 147.1, 132.0, 125.2, 41.9,
30.9, 30.1 ppm; FT/IR = 3383, 3058, 2925, 1656, 1604, 1517, 1429,
1345, 1234, 1205, 1108, 1038, 1014, 970, 860, 815, 788, 774, 768, 762,
710, 665 cmꢀ 1; HRMS (FAB): m/z [M + H]+ calcd for [C22H25N6O8Se2]+
661.0064, found: 661.0071.
(2E,2′E)-N,N’-[diselanediylbis(ethane-2,1-diyl)]bis[3-(4-(tert-butyl)
phenyl)-2-(hydroxyimino)propanamide] (20). Following the general
procedure of selenopsammaplin A (9) from 4-t-butylbenzaldehyde, 20
4.5. DOT1L enzyme activity assay
DOT1L enzyme activity was measured using S-adenosyl methionine
(SAM) as the methyl group donor and synthesized DOT1L as the
1
◦
was obtained as a yellow solid (overall 33% yield). m.p. = 88 C; H
8

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
substrate (BPS Bioscience, Cat. No. 52202; San Diego, CA, USA) ac-
G-3′ and 5′-CTT GGC TGA GAG GAT GGT GTA A-3′; CDH2, 5′-AGC CAA
CCT TAA CTG AGG AGT-3′ and 5′-GGC AAG TTG ATT GGA GGG ATG-3′;
ZEB1, 5′- GCA CCT GAA GAG GAC CAG AG-3′ and 5′-GTG TAA CTG CAC
AGG GAG CA-3′; VIM, 5′-AGA TGG CCC TTG ACA TTG AG-3′ and 5′-TGG
AAG AGG CAG AGA AAT CC; β-Actin, 5′-AGC ACA ATG AAG ATC AAG
AT-3′ and 5′-TGT AAC GCA ACT AAG TCA TA-3′.
cording to the manufacturer’s instructions.
4.6. Western blotting analysis
Total cell lysates were prepared in 2 × sample loading buffer [250
mM Tris-HCl (pH 6.8), 10% glycerol, 4% sodium dodecyl sulfate (SDS),
2% β-mercaptoethanol, 0.006% bromophenol blue, 5 mM sodium
orthovanadate, and 50 mM sodium fluoride, Bio-Rad]. The protein
concentrations of samples were measured using the bicinchoninic acid
4.10. In vivo orthotopic mouse tumor model
All animal experiments were conducted following the guidelines
approved by the Seoul National University Institutional Animal Care and
Use Committee (IACUC; permission number: SNU-170912–13). Female
nude mice (BALB/c-nu), 5–6 weeks old, were purchased from Central
Laboratory Animal, Inc. (Seoul, Korea) and housed under pathogen-free
conditions with a 12 h light–dark schedule. Luciferase expressing MDA-
MB-231 cells were inoculated orthotopically into the fourth fat pad of
method.39 Equal amounts of protein (5-20
μg) were separated by 8–15%
SDS-polyacrylamide gel electrophoresis (PAGE) and were transferred to
polyvinylidene fluoride membranes (Millipore, Bedford, MA, USA). The
membranes were blocked with 5% bovine serum albumin (BSA, Sigma-
Aldrich) and were then probed with anti-E-cadherin, anti-N-cadherin
(BD Bioscience), anti-H3K4me2, anti-H3K27me2, anti-H3K36me2, anti-
H3K79me2, anti-Histone H3 (Cell Signaling Technology, Beverly, MA,
USA) or anti-ZEB1, anti-Vimentin, anti-β-actin (Santa Cruz Biotech-
nology, Dallas, TX, USA) antibodies. The blots were detected with an
enhanced chemiluminescence (ECL) detection kit (GE Healthcare, Little
Chalfont, UK).40
mice (1 × 107 cells in 100
μL of 50:50 DMEM/Matrigel) using a 29-gauge
needle. Ten days after implantation, mice were randomized into vehicle
control and treatment groups of six animals per group and were
administered with vehicle (EtOH/Cremophor/Normal saline = 5:5:90),
10 (5 or 15 mg/kg body weight), Psammaplin A (30 mg/kg) or Paclitaxel
(5 mg/kg body weight), as a positive reference control. Compounds
were administered intraperitoneally three times per week for 38 days.
An additional week later, anesthetized mice were positioned in IVIS
(PerkinElmer, Waltham, MA, USA) and were imaged 15 min after in-
jection of D-luciferin (150 mg/kg, Gold Bio Technology) resuspended in
PBS. Mice were Euthanized and primary tumors and organs (lungs,
livers, spleens, kindeys) were excised, weighed, and frozen for further
biochemical analysis or fixed for immunohistochemistry. The length (L),
width (W), and height (H) of the tumors were also measured using a
digital slide caliper once a week, and tumor volumes (mm3) were esti-
mated by the formula LWH/2. Toxicity was assessed based on the
lethality and body weight loss exhibited by the nude mice.43
4.7. Wound healing assay
MDA-MB-231 cells were grown to 80–90% confluence in a 6-well
plate. A confluent monolayer of the cells was artificially wounded
with a 200 μL pipette tip, and the detached cells were washed with
phosphate-buffered saline (PBS, Invitrogen Corp.), followed by incuba-
tion with 1% FBS in medium containing various concentrations of
compound for 24 h. The wounds were photographed at 0 and 24 h on an
inverted microscope (Olympus, Tokyo, Japan). The wound area was
quantified using ImageJ Software (National Institutes of Health) and
presented as wound healing (%) in comparison with the area of the
wound at 0 h.
4.11. In vivo acute oral toxicity evaluation model
4.8. Cell invasion assay
All animal experiments were conducted following the guidelines
approved by the Seoul National University Institutional Animal Care and
Use Committee (IACUC; permission number: SNU-180409–2-1). ICR
mice (30 ± 2 g each) of both sexes were purchased from Central Labo-
ratory Animal, Inc. (Seoul, Korea) and housed under pathogen-free
conditions with a 12 h light–dark schedule. All the animals except
control group were administered a single oral dose of compound 10 at
300 mg/kg body weight. The control animals were received only vehicle
(EtOH/Cremophor/Normal saline = 5:5:90). Signs of toxicity and
mortality were observed daily for 7 days.
Twenty-four-well transwell membrane inserts with 6.5 mm di-
ameters and 8
coated with 10
μm pore sizes (Corning, Tewksbury, MA, USA) were
μ
L of type I collagen (0.5 mg/mL, BD Biosciences, San
Diego, CA, USA) and 20
μ
L of a 1:20 mixture of Matrigel (BD Bio-
sciences)/PBS. After treatment with compounds for 24 h, MDA-MB-231
cells were harvested, resuspended in serum-free medium, and plated
(0.5–1 × 106 cells/chamber) in the upper chamber of the Matrigel-
coated transwell insert. Medium containing 30% FBS was used as a
chemoattractant in the lower chambers. At 24 h of incubation, the cells
that had invaded the outer surface of the lower chambers were fixed and
stained using Diff-Quik Staining Kit (Sysmex, Kobe, Japan) and
imaged.41 Representative images from three separate experiments are
shown, and the number of invaded cells was counted in 5 randomly
selected microscopic fields (200 × magnification).
4.12. Ex vivo immunohistochemistry and biochemical analyses of tumors
Tumor tissues fixed in 4% PFA were embedded in paraffin, sectioned
mounted to slides, deparaffinized with xylene, rehydrated with ethanol
series, and subjected to antigen retrieval. The slides were incubated with
the indicated antibodies, which were detected using the LSAB + System-
HRP kit (Dako, Glostrup, Denmark), and counterstained with hema-
toxylin. Stained sections were observed and imaged using a Vectra 3.0
Automated Quantitative Pathology Imaging System (Perkin Elmer,
Waltham, MA, USA). A portion of frozen tumors excised from nude mice
was homogenized using a hand-held homogenizer in Complete Lysis
Buffer (Active Motif, Carlsbad, CA, USA) or TRI reagent (Invitrogen,
Carlsbad, CA, USA). The protein expression levels of the tumor lysates
were determined. The protein concentrations of tumor lysates were
measured using the Bradford assay.44
4.9. RNA isolation and real-time polymerase chain reaction (real-time
PCR)
The total RNA from the cells was extracted with TRI reagent (Invi-
trogen, Carlsbad, CA, USA), and 1
μg of total RNA was reverse-
transcribed using a Reverse Transcription System (Cat. No. A3500;
Promega, Madison, WI, USA) according to the manufacturer’s in-
structions. Real-time PCR was conducted using iQ SYBR Green Supermix
(Bio-Rad, Hercules, CA, USA) according to the manufacturer’s in-
structions. The threshold cycle (CT) was determined using Bio-Rad CFX
manager 3.1 software. Relative quantification between compounds and
untreated controls normalized to the levels of β-Actin mRNA was
calculated using the comparative CT method.42 The following primers
were used for real-time PCR: CDH1, 5′-GTT ATT CCT CTC CCA TCA GCT
4.13. Docking methods
Molecular docking simulations were conducted using the Surflex-
9

H. Ju Han et al.
Bioorganic & Medicinal Chemistry 35 (2021) 116072
2 Yao Y, Chen P, Diao J, et al. J Am Chem Soc. 2011;133:16746–16749.
Dock in Sybyl-X2.1.1 (Tripos Inc, St Louis, MO) with the crystal struc-
tures of human DOT1L (PDB: 3SR4). The 3D structures of 10 were
prepared by generating 3D Concord conformations from the 2D struc-
tures. The original ligand from the X-ray structure (PDB: 3SR4) was
removed and the prepared ligand 10 was docked into the crystal
structure of human DOT1L. The water molecules in the active site were
also removed. Staged minimization was performed with Powell’s
method until the gradient converged to a value of 0.5 kcal/mol∙Å.
Tripos force field option was applied with use current charges. Protomol
was generated based on the location of the original ligand, TT8, with a
threshold of 0.5 Å and bloat of 0 Å. The protein movement option was
applied to allow sufficient flexibility in the binding pocket of the protein.
Docking performance was validated by the docking scores along with
the visual inspections of the re-docked structures compared to the
original pose. Molecular interactions between the ligand and protein
were further analyzed using Discovery studio 4.5 Visualizer (Biovia,
3 Bernt KM, Zhu N, Sinha AU, et al. Cancer Cell. 2011;20:66–78.
4 Nguyen AT, Taranova O, He J, Zhang Y. Blood. 2011;117:6912–6922.
5 Daigle SR, Olhava EJ, Therkelsen CA, et al. Cancer Cell. 2011;20:53–65.
6 Zhang L, Deng L, Chen F, et al. Oncotarget. 2014;5:10665–10677.
7 Cho MH, Park JH, Choi HJ, et al. Nat Commun. 2015;6:7821–7834.
8 Michalak EM, Visvader JE. Mol Oncol. 2016;10:1497–1515.
9 Lee JY, Kong G. Oncotarget. 2015;6:30451–30452.
10 Anglin JL, Deng L, Yao Y, et al. J Med Chem. 2012;55:8066–8075.
11 Wang Y, Li L, Zhang B, et al. J Med Chem. 2017;60:2026–2036.
12 Campbell CT, Haladyna JN, Drubin DA, et al. Mol Cancer Ther. 2017;16:1669–1679.
13 Hong S, Lee M, Jung M, Park Y, Kin M-H, Park H-g. Tetrahedron Lett. 2012;53:
4209–4211.
14 Hong S, Shin Y, Jung M, et al. Eur J Med Chem. 2015;96:218–230.
ˇ
`
15 Quinoa E, Crew CP. Tetrahedron Lett. 1987;28:3229–3233.
16 Rodriguez ADA, Akee RK, Scheuer PJ. Tetrahedron Lett. 1987;28:4989–4992.
17 Arabshahi LS, Schmitz FJ. J Org Chem. 1987;52:3584–3586.
18 Kim D, Lee IS, Jung JH, Yang SI. Arch Pharm Res. 1999;22:25–29.
19 Jung JH, Sim CJ, Lee CO. J Nat Prod. 1995;58:1722–1726.
20 Kim D, Lee IS, Jung JH, Lee CO, Choi SU. Anticancer Res. 1999;19:4085–4090.
21 Shin J, Lee H-S, Seo W, Rho J-R, Cho KW, Paul VJ. Tetrahedron. 2000;26:9071–9077.
22 Pina IC, Gautschi JT, Wang GY, et al. J Org Chem. 2003;68:3866–3873.
23 Ahn MY, Jung JH, Na YJ, Kim HS. Gynecol Oncol. 2008;108:27–33.
24 Mora FD, Jones DK, Desai PV, et al. J Nat Prod. 2006;69:547–552.
25 L.C. Clark, G.F., Jr. Combs, B.W. Turnbull, E.H. Slate, D.K. Chalker, J. Chow, L.S.
Davis, R.A. Glover, G.F. Graham, E.G. Gross, A. Krongrad, J.L., Jr. Lesher, H.K. Park,
B.B., Jr. Sanders, C.L. Smith and J.R. Taylor, JAMA, 1996;276: 1957–1963.
26 Moscow JA, Schmidt L, Ingram DT, Gnarra J, Johnson B, Cowan KH. Carcinogenesis.
1994;15:2769–2773.
¨
SanDiego, CA, USA) and PyMOL-v1.0 (Schrodinger KK, Tokyo, Japan).
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.
27 Calvo A, Xiao N, Kang J, et al. Cancer Res. 2002;62:5325–5335.
28 Hu YJ, Diamond AM. Cancer Res. 2003;63:3347–3351.
29 Meplan C, Nicol F, Burtle BT, et al. Antioxid Redox Signal. 2009;11:2631–2640.
30 Sanmartin C, Plano D, Palop JA. Mini Rev Med Chem. 2008;8:1020–1031.
31 Sahu PK, Umme T, Yu J, et al. J Med Chem. 2015;58:8734.
Acknowledgements
32 Sahu PK, Kim G, Nayak A, et al. Asian J Org Chem. 2016;5:183–186.
33 Yu J, Zhao LX, Park J, et al. J Med Chem. 2017;60:3422–3437.
34 Yu J, Ahn S, Kim HJ, et al. J Med Chem. 2017;60:7459–7475.
35 Byun WS, Kim WK, Han HJ, et al. Mol Ther Oncolytics. 2019;15:140–152.
36 Byun WS, Kim WK, Yoon JS, Jarhad DB, Jeong LS, Lee SK. Biomolecules. 2020;10:
530–545.
This study was supported by a grant from the National Research
Foundation of Korea (NRF) funded by the Korean Government (NRF-
2016M3A9B6902986).
Appendix A. Supplementary material
37 Anglin JN, Song Y. J Med Chem. 2013;56:8972–8983.
38 Vichai V, Kirtikara K. Nat Protoc. 2006;1:1112–1116.
39 Smith PK, Krohn RI, Hermanson GT, et al. Anal Biochem. 1985;150:76–85.
40 Kim WK, Byun WS, Chung HJ, et al. Biochem Pharmacol. 2018;152:71–83.
41 Park SK, Byun WS, Lee S, et al. Biochem Pharmacol. 2020;178:114053–114067.
42 Byun WS, Jin M, Yu J, et al. Biochem Pharmacol. 2018;158:84–94.
43 Byun WS, Kim S, Shin YH, Kim WK, Oh D-C, Lee SK. J Nat Prod. 2020;83:118–126.
44 Bradford MM. Anal Biochem. 1976;72:248–254.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmc.2021.116072.
References
1 Nguyen AT, Zhang Y. Genes Dev. 2011;25:1345–1358.
10