Discovery of New H2S Releasing Phosphordithioates and 2,3-Dihydro-2-phenyl-2-sulfanylenebenzo[d][1,3,2]oxazaphospholes with Improved Antiproliferative Activity

Article abstract of DOI:10.1021/acs.jmedchem.5b00848

Hydrogen sulfide (H₂S) is now recognized as a physiologically important gasotransmitter. Compounds which release H₂S slowly are sought after for their potential in therapy. Herein the synthesis of a series of phosphordithioates based on 1 (GYY4137) are described. Their H₂S release profiles are characterized using 2,6-dansyl azide (2), an H₂S specific fluorescent probe. Most compounds have anticancer activity in several solid tumor cell lines and are less toxic in a normal human lung fibroblast, WI38. A preferred compound, 14, with 10-fold greater anticancer activity than 1, was shown to release H₂S in MCF7 cells using a cell active probe, 21. Both permeability and intracellular pH (pHi) were found to be significantly improved for 14 compared to 1. Furthermore, 14 was also negative in the AMES test for genotoxicity. Cyclization of these initial structures gave a series of 2,3-dihydro-2-phenyl-2-sulfanylenebenzo[d][1,3,2]oxazaphospholes, of which the simplest member, compound 22 (FW1256), was significantly more potent in cells. The improved therapeutic window of 22 in WI38 cells was compared with three other cell types. Potency of 22 was superior to 1 in MCF7 tumor spheroids and the mechanism of cell death was shown to be via apoptosis with an increase in cleaved PARP and activated caspase-7. Evidence of H₂S release in cells is also presented. This work provides a "toolbox" of slow-release H₂S donors useful for studies of H₂S in biology and as potential therapeutics in cancer, inflammation, and cardiovascular disease. (Chemical Equation Presented).

Full text of DOI:10.1021/acs.jmedchem.5b00848

Subscriber access provided by NEW YORK UNIV
Article
Discovery of new H2S releasing phosphordithioates
and 2,3-dihydro-2-phenyl-2-sulfanylenebenzo[d]
[1,3,2]oxazaphospholes with improved antiproliferative activity
Wei Feng, Xinyi Teo, Wisna Novera, Pondy Murugappan Ramanujulu, Liang
Dong, Dejian Huang, Philip Keith Moore, Lih-Wen Deng, and Brian W. Dymock
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00848 • Publication Date (Web): 06 Jul 2015
Downloaded from http://pubs.acs.org on July 8, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of new H S releasing
2
phosphordithioates and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2,3-dihydro-2-phenyl-2-sulfanylenebenzo[d]
[
1,3,2]oxazaphospholes with improved
antiproliferative activity
a,b
b
c
a,b
Wei Feng , Xinyi Teo , Wisna Novera , Pondy Murugappan Ramanujulu , Liang
d
d
b,e
c
a*
Dong , Dejian Huang , Philip K. Moore , Lih-Wen Deng , Brian W. Dymock
a
Department of Pharmacy, Block S4A Level 3, 18 Science Drive 4, National
b
University of Singapore, Singapore 117543; Life Sciences Institute, Centre for Life
Sciences Level 5, 28 Medical Drive, National University of Singapore, Singapore
c
117456; Department of Biochemistry, Block MD7 #04-06, 8 Medical Drive,
d
National University of Singapore, Singapore 117596; Department of Chemistry,
Block S8 Level 3, Science Drive 3, National University of Singapore, Singapore
e
117543; Department of Pharmacology, UHL Level 5-02R, Lee Kong Chian Wing, 21
Lower Kent Ridge Road, National University of Singapore, Singapore 119077.
Keywords: hydrogen sulfide, fluorescent probe, anti-proliferative, anti-cancer,
internal pH (pHi), confocal microscopy, tumor spheroid, apoptosis
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Hydrogen sulfide (H S) is now recognized as a physiologically important
2
gasotransmitter. Compounds which release H S slowly are sought after for their
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
potential in therapy. Herein the synthesis of a series of phosphordithioates based on 1
(GYY4137) are described. Their H S release profiles are characterized using
2
2
,6-dansyl azide (2), an H S specific fluorescent probe. Most compounds have
2
anticancer activity in several solid tumor cell lines and are less toxic in a normal
human lung fibroblast, WI38. A preferred compound, 14, with 10-fold greater
anticancer activity than 1, was shown to release H S in MCF7 cells using a cell active
2
probe, 21. Both permeability and internal pH (pHi) were found to be significantly
improved for 14 compared to 1. Furthermore 14 was also negative in the AMES test
for genotoxicity. Cyclisation of these initial structures gave a series of
2
,3-dihydro-2-phenyl-2-sulfanylenebenzo[d][1,3,2]oxazaphospholes of which the
simplest member, compound 22 (FW1256), was significantly more potent in cells.
The improved therapeutic window of 22 in WI38 cells was compared with 3 other cell
types. Potency of 22 was superior to 1 in MCF7 tumor spheroids and the mechanism
of cell death was shown to be via apoptosis with an increase in cleaved PARP and
activated caspase-7. Evidence of H S release in cells is also presented. This work
2
provides a ‘toolbox’ of slow-release H S donors useful for studies of H S in biology
2
2
and as potential therapeutics in cancer, inflammation and cardiovascular disease.
ACS Paragon Plus Environment

Page 3 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Hydrogen sulfide (H S) at high concentrations is a toxic, malodorous gas, however, at
2
low concentrations H S is increasingly recognized as an important cell-signaling
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
gasotransmitter, the so-called ‘third gasotransmitter’, after the extensively studied
1
, 2
carbon monoxide and nitric oxide. H S is produced endogenously via enzymatic
2
reactions of cysteine, homocysteine and cystathionine mediated by cystathionine-β
3
-
synthase (CBS) and cystathionine-γ-lyase (CSE). Furthermore, a third enzyme
specific to the mitochondria, 3-mercaptopyruvate sulfurtransferase (3-MST), has been
4
identified to produce H S.
2
In the past decade, H S has been shown to play significant roles in physiology and
2
5
-7
8-10
11,
pathology including cardioprotection,
inflammation,
atherosclerosis,
1
2
13
14
15
16, 17
neuromodulation, hypertension, hemorrhagic shock, pain perception
and
1
8
gastric mucosal integrity. In cancer, H S has been shown to trigger apoptosis via
2
1
9
20
activation of MAP kinase and caspase-3 , sulfhydration of NF-ΚB and H S donors
2
2
1, 22
have reported anti-angiogenic
effects with phosphorylation of the protein kinase
2
3
Akt being implicated. Furthermore, the PI3K-Akt pathway was also shown to be
activated in cardioprotective studies of H S. However, to date, the mechanisms of
2
action of the effects of H S have only been partially elucidated. Kimura has reported
2
that H S protects cells from oxidative stress via enhancing the production of
2
2
4, 25
glutathione (GSH) by increasing the transport of cysteine.
It has also been
24
suggested that H S may suppress oxidative stress directly as an antioxidant.
2
Recently it has been shown that mitochondrial targeting of H S has a powerful
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
6
anti-oxidant effect. H S has been reported to regulate protein function via protein
2
S-Sulfhydration (protein-S-SH), adding to the ‘persulfide pool’ of which some of the
20, 27-29
main players are S-cysteine and S-glutathione at levels up to 100 µM.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Conversely, there are also reports of the opposing effects of H S, where, for example,
2
3
0
inhibition of its production appears to be of benefit. Such a wide range of
physiological studies clearly implicate H S as an important factor in disease.
2
Therapeutically focused mechanistic studies are hampered by the limited availability
of ‘controllable’ and ‘drug-like’ H S donors, defined as compounds with a useful
2
range of known H S release rates under physiological conditions. Sodium sulfide
2
(
Na S) and sodium hydrogen sulfide (NaHS) salts, giving virtually instant generation
2
of H S upon dissolution and hydrolysis by water, are still frequently used in biological
2
studies. Compared to these inorganic salts, organic H S donors are able to release H S
2
2
slowly and continuously at low concentrations similar to endogenous levels. There are
several classes of such organic H S donors, based on diverse mechanisms of H S
2
2
2
6, 31-35
release, which have been developed to date (Figure 1).
S
O
O
H
AcHN
S
S
S
N
N
S
S
S
H
R
O
S S COR
HO
ADT-OH
(dithiolthione)
DATS
persulfide-based donors
Cys-Act
O
S
S
O
R
S
S
MeO
P S
N
R
P
SH
N
S
H
NH2
NO2
O N
2
Phenyldithiophos
phoramidate
Thioglycine (R=H)
Thiovaline (R=iPr)
GYY4137 (1)
Gem-dithio derivative
O
Figure 1. Representative H S donor compounds reported in the literature.
2
Among these compounds, GYY4137 (1) is one of the most extensively studied
ACS Paragon Plus Environment

Page 5 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
organic H S donors with a wide range of biological activities, both in oncology and
2
3
6, 37
non-oncology, having been reported in over 50 publications to date.
Therapeutic
38-46
applications of 1 range from cardiovascular to inflammation and aging to cancer.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In oncology, Deng and co-workers reported that 1 exhibits anti-cancer activity in vitro
3
9, 47
without significantly affecting normal cells.
In support of H S donors as potential
2
therapeutics, Deng also demonstrated the anti-cancer effect of 1 in a mouse model of
3
9
cancer. It was found that this anticancer activity was also H S-concentration
2
dependent. Lu and co-workers have shown that 1 has promising activity in
4
8
hepatocellular cancer through blocking STAT3. Recently Deng and co-workers have
reported that continuous exposure of cells to H S by dosing the short-lived NaHS salt
2
every 2 hours mimics the effects of the slow releasing 1, whereas a single dose of
NaHS has no effect. Prolonged exposure to H S by dosing cells with a single dose of
2
1
, led to intracellular acidification via increased glycolysis and deregulation of cell pH
4
7
control. All of these findings support the critical requirement for appropriate H S
2
concentrations to be delivered at a suitable rate to exert a physiologic effect without
toxicity. To further study the relationship between H S concentrations and anticancer
2
biological activity, we herein report the design and synthesis of a series of
dithiophosphorus H S donors, based on 1. These compounds display a wide range of
2
H S release profiles and anti-proliferative activities. By varying the nucleophiles
2
employed in the synthesis, it is possible to control the reactivity of the compounds,
thus tuning the H S release rate. Such a related series of tool compounds, within the
2
same molecular series, delivering a broad range of H S concentrations over hours to
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
days, have not been previously reported and may lead to deeper understanding of H2S
cellular mechanisms and potentially to new therapeutics for unmet medical needs. We
also show that a preferred compound, 14, has an order of magnitude improvement in
antiproliferative activity compared to 1 while retaining a therapeutic window over
normal cells, lowers intracellular pH (pHi) to a greater extent than 1 and increases
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
intracellular H S levels. We also describe further development of the series via
2
cyclisation
which
leads
to
a
new
series
A
of
2,3-dihydro-2-phenyl-2-sulfanylenebenzo[d][1,3,2]oxazaphospholes.
preferred
compound, 22, has increased potency towards solid tumor cells while being non-toxic
to WI38 cells at the highest concentration tested. Taken together, this work provides
new H S donor compounds, with a wide range of H S release rates, and takes a step
2
2
towards a potential H S therapy for the treatment of cancer.
2
RESULTS AND DISCUSSION
In Vitro Quantification of H S Released from 1 (GYY4137)
2
Application and optimization of a recently described fluorescent probe for
quantification of slow-release H S evolved directly from 1 and its novel analogues is
2
herein described. However, such probes are still only indirectly quantifying H2S
evolved from donor compounds following the necessary reduction reaction of the
probe by H S. Hence we also explored the direct detection of H S by headspace
2
2
GC-MS which does not involve any chemical transformations to achieve a readout.
Until the recent development of new H S selective probes, there have been few
2
practical methods suitable to detect the low levels of H S evolved from therapeutic
2
ACS Paragon Plus Environment

Page 7 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds. A detailed study of the H S release profile of 1 over a period of days, the
2
time course of many of the aforementioned therapeutic studies, is essential for a better
understanding of its activities. Current techniques for H S detection described in the
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
9,
50
51
literature, include colorimetric
and electrochemical assays,
gas
5
2
53
chromatography, metal-induced sulfide precipitation and the most sensitive and
54, 55
probably most accurate method yet reported using HPLC.
The latter method,
although complex to carry out, is being applied by Kevil to study H S levels in the
2
5
6, 57
clinic.
These techniques often require significant sample processing, hence
estimates of H S concentrations can be highly variable and potentially unreliable,
2
especially without expert guidance. For the quantification of H S released from
2
dithiophosphorus compounds, such as 1, we required a robust and easy to carry out
method for routine laboratory application. A methylene blue assay has been employed
for H S detection however, hydrochloric acid, which can significantly accelerate the
2
H S release rate, is required in the system, potentially resulting in errors in the
2
54
58-70
measured H S concentrations. Many fluorescent probes
have been developed
2
recently, some are very recently published and have not yet been evaluated by other
groups. Among these probes, 2,6-dansyl azide (2, Figure 2) reported by Wang in
6
2
2
014, reacts with H S in a few minutes to generate a fluorescent product with a
2
higher quantum yield compared to the previously published 1,5-dansyl azide probe
6
9
(3).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
N3
O
O
S
O
S
N3
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
,6-dansyl azide (2)
1,5-dansyl azide (3)
Figure 2. Dansyl azide probes reported for specific use in H S quantification.
2
Following extensive attempts to synthesise and apply a suitable probe for the
detection of H S specifically from 1, we were encouraged by promising initial results
2
with probe 2. The 1,5-dansyl azide alternative probe can also be used but 2 is more
sensitive with lower background. We proceeded to optimize simple cell-free in vitro
assay conditions for monitoring of H S release from 1 over a period of days with
2
probe 2. We prepared both 1 and 2 as per the reported procedures with one
modification in the synthesis of 2 (see Supporting Information).
Compound 1 is an ionic salt and its solubility is generally low in most organic
solvents. Probe 2 is a lipophilic organic compound and is not soluble under aqueous
conditions. Thus to completely dissolve both components as close as possible to
physiological conditions a balance of organic and aqueous solvent is required. A
mixture of acetonitrile (MeCN) and phosphate buffered saline (PBS) at pH 7.4 was
chosen for initial studies. We first investigated the stability of 2 with different volume
ratios of PBS:MeCN over 4 days, a time period of particular relevance to the study of
slow H S releasers. With more than 70% PBS, the probe is not soluble. When PBS is
2
less than 10%, the media is not clear due to precipitation of phosphate salts. Hence
probe stability over time was carried out with PBS:MeCN ratios ranging from 10% to
ACS Paragon Plus Environment

Page 9 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
0%. Background fluorescence response units increased with increases in PBS:MeCN
ratios (Figure 3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
500
2000
500
1000
10% PBS:MeCN
0% PBS:MeCN
30% PBS:MeCN
2
1
5
00
0
0
50
100
Time (h)
Figure 3. Fluorescence response of probe 2 at 10%, 20% and 30%
PBS:MeCN over time at 22 ºC.
Stability of 2 was also tested in pH 3.0 PBS, however, a clear solution is not possible
to obtain at this pH and inconsistent results were obtained. Hence we settled on 10%
PBS:MeCN as the media for further studies. Next we wanted to understand
application of this probe in media with additional components found in cells at their
physiological concentrations. Reliable quantification of H S inside living cells is still
2
very challenging. Nevertheless we wanted to understand if 2 could potentially be used
for assessment of intracellular levels of H S released from 1. It has been reported that
2
probe 2 is stable in 100 µM glutathione (GSH) or cysteine. However we found that 2
reacts with physiological (millimolar) levels of GSH or cysteine under the optimized
assay conditions. When 2 was incubated with GSH or cysteine, fluorescence response
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
increased significantly, particularly for added cysteine, indicating that 2 is not
selective for H S quantification with physiological concentrations of added biological
2
thiols. (Figure S88). Notwithstanding these limitations we have found that 2 is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
applicable in situations where lower levels of such thiols are present, such as in cell
7
1
media.
Having established the optimal solvent conditions for probe stability, we next
determined the H S release profile of standard solutions of sodium sulfide (Na S).
2
2
Salts such as Na S or NaHS, both commonly used in the H S literature, rapidly
2
2
hydrolyze in water releasing H S within seconds into the solution. Solutions of Na S
2
2
with 2 were freshly prepared from dried solid sodium sulfide salt (nonahydrate) added
directly to a solution of 2 (0.2 mM) in the optimized solvent 9:1 MeCN:PBS, at room
temperature. A direct correlation between Na S levels and fluorescence response units
2
(RU) was observed enabling a calibration equation to be established relating
fluorescence RU to concentration of H S, applicable to the analysis of H S releasing
2
2
drugs (Figure 4A).
ACS Paragon Plus Environment

Page 11 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A:
6
0000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
y = 243.07x
R² = 0.9968
50000
4
3
2
0000
0000
0000
10000
0
0
50
100
150
200
[
Na S] μM
2
B:
7
6
0000
0000
y = 65.411x
R² = 0.9764
50000
4
3
2
1
0000
0000
0000
0000
0
0
200
400
600
800
1000
[
Na S] µM
2
Figure 4. Calibration plot of fluorescence response of probe 2 (A:
.2 mM, B: 1 mM) in 10% PBS:MeCN versus sodium sulfide
concentration over time at 22 ºC.
0
This calibration applies only to experiments using 0.2 mM of 2. If a different
concentration of 2 is used a separate calibration should be determined, for example
with 1 mM of 2 (Figure 4B). With a calibration in hand we next determined the H2S
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
release profile of compound 1 at 22 ºC (Figure 5A). Three concentrations were
studied to reflect conditions in typical published pharmacological studies with up to
72
5
mM concentration of 1: 0.2, 1 and 5 mM, tested in triplicate, thrice (total N=9).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A:
60
50
40
30
20
10
0
0
.2 mM GYY 1 mM GYY 5mM GYY
1
2
3
4
5
Time (day)
B:
8
00
0
.2 mM GYY
1 mM GYY
5 mM GYY
700
6
5
4
3
00
00
00
00
200
1
00
0
1
2
3
4
5
Time (day)
Figure 5. Quantification of H S released from 1 over 5 days. A: at
2
2
1
2 ºC with 0.2 mM 2. B: at 37 ºC with 0.2 mM 2 (for 0.2 and
.0 mM 1) and 1 mM 2 (for 5 mM 1).
At the lower concentration of 0.2 mM of 1, lower amounts of H S was detected, near
2
ACS Paragon Plus Environment

Page 13 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to the limit of detection of the assay (approximately 1 μM). When the concentration
of 1 was increased to 1 mM, 7.5±0.6 µM of H S was detected after 4 days. Increasing
2
the concentration of 1 to 5 mM, 49.4±3.5 μM of H S was quantified after 5 days at
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
room temperature. About 10 μM of H S was released every day showing a steady
2
release profile suitable for an H S therapeutic. However given the lower
2
concentrations of 1 required to observe inhibition of cancer cell proliferation (IC50
approximately 350 µM in MCF7 cells) it is likely that (i) H S release rates may be
2
higher in cells due to enhanced release from cellular components; (ii) the higher
temperature in cells may increase H S release rates and; (iii) the probe sensitivity may
2
be too low.
To study the influence of temperature on H S rate, a similar analysis was carried out
2
at 37 ºC at the same three concentrations using 0.2 mM of probe 2 (Figure 5B). At
0
.2 mM, 1 again released trace amounts of H S. When the concentration of 1 was
2
increased to 1 mM, a 4.7 fold higher amount of H S was detected at 37 ºC than at
2
2
5
2 ºC, after 4 days incubation. When the concentration of 1 was further increased to
mM, a surprisingly large amount of H S (0.8 mM), 16 fold greater than at 22 °C,
2
was detected after a 4 day incubation at 37 ºC. Note that 1 mM of probe 2 was
required for quantification of the higher levels of evolved H S at 5 mM concentration
2
of 1. Regardless of any non-linear effects between probe responses at 0.2 and 1.0 mM
concentrations, this is a higher than expected evolution of H S at 37 °C which may
2
explain the reported cellular and in vivo pharmacology of 1, i.e. 1 may be releasing
more H S in living systems than expected from its in vitro behavior, potentially due to
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the higher temperature in cells compared to ‘room temperature’. Given the high
sensitivity of the H S release to temperature future experiments were conducted with
2
a sample of 1 as a control.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Having established the H S release profile of 1 with fluorescent probe 2, we wanted to
2
provide additional direct evidence for the release of H S from 1. Hence we employed
2
head space GC-MS (HS-GC-MS) to find evidence for the presence of H S gas, thus
2
7
3
proving that 1 is able to release H S. Under the assay conditions used for 2 (5 mM 1
2
in 9:1 MeCN:PBS) the mass of H S, 34, was indeed observed. Unfortunately, under
2
the conditions it was quite weak compared to strong signals for MeCN and oxygen.
Dissolving 1 in pure PBS at 5 mM solved the MeCN interference but oxygen was still
a problem. We were also concerned that H S, being extremely volatile, was escaping
2
from the solution and the headspace before a sample could be taken. Hence a solution
of 1 was dissolved in PBS and placed in a microwave tube, which can be sealed
gas-tight. After sealing, the tube was vacuumed to remove oxygen (see experimental
section). Thus, when H S was evolved from 1, its concentration in the headspace was
2
significantly higher so as to enable detection (Figure 6). Thus we confirmed
qualitatively the release of H S from 1.
2
ACS Paragon Plus Environment

Page 15 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
.
.
4
2
1
0
0
0
0
e
e
e
0
0
0
0
+
+
+
0
0
0
0
0
0
0
0
0
0
0
7
7
7
0
0
0
0
2
2
2
. 8 e + 0 7
. 6 e + 0 7
. 4 e + 0 7
8
0
CH2Cl2
6
4
2
0
0
0
2 . 2 e + 0 7
2 e + 0 7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
1
. 8 e + 0 7
. 6 e + 0 7
. 4 e + 0 7
1
3
.
3
5
1
.
4
0
1
.
4
5
1
.
5
0
1
.
5
5
1
.
6
c
H2S + O2
#
3 9 :
H
yd ro g e n su lfid e
S
a n 7 5 7 (1 .5 5 0 m in ): G Y Y E V A C U A T E D
4
H2S
3 2
O2
1 . 2 e + 0 7
e + 0 7
8
6
4
0 0 0
0 0 0
0 0 0
8 0 0 0
6 0 0 0
4 0 0 0
2 0 0 0
0
1
8
6
4
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
2 0 0 0
4 0
4 0
4 4
3
6
4 8
6 0
6 0
6 4
6 5
0
2
0
2
5
3
0
3
5
4
0
4
5
5
0
5
5
6
0
6
5
7
0
2 0
2 5
3 0
3 5
4 5
5 0
5 5
7 0
2 0 0 0 0 0 0
0
. 5 0
1 . 0 0
1 . 5 0
2 . 0 0
2 . 5 0
3 . 0 0
3 . 5 0
4 . 0 0
4 . 5 0
5 . 0 0
5 . 5 0
6 . 0 0
6 . 5 0
7 . 0 0
7 . 5 0
8 . 0 0
Time
Figure 6. Gas chromatography tandem mass spectrometry. Chromatogram showing peaks at
approximately 1.52 minutes for H S + oxygen and 1.84 minutes for dichloromethane. Upper
2
insert: peak at 1.52 minutes showing co-elution of oxygen and hydrogen sulfide. Lower
insert: Mass spectrometry chromatogram showing H S signal at 34 and oxygen signal at 32.
2
In summary, utilizing recently reported probe 2 we have optimised a straightforward
assay for H S quantification from the slow releasing H S donor drug, 1, over a period
2
2
of up to 4 or 5 days. Longer incubation times are also possible with this method.
Probe 2 is stable under neutral conditions with 10% aqueous phase enabling the H2S
release profile of 1 to be determined. Furthermore, we have proven that H S is indeed
2
released from 1 using HS-GC-MS technology. With the desired practical, robust assay
in hand we next prepared a number of direct analogues of 1 aiming to identify a series
of compounds with a range of H S release profiles and improved anti-proliferative
2
activity.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Design and Synthesis of Compounds
Until recently when a series of 5 O-substituted N-phenyl phosphordithioates, with
74
only oxygen and nitrogen phosphorus substituents (Figure 1), were reported , to our
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
knowledge morpholine-substituted 1 was the only reported phosphordithioate based
derivative
characterized
in
a
biological
system.
These
reported
O,N-phosphordithioates, with variations in their structures only in the substituents, not
directly on atoms bonded to phosphorus, have very similar H S release rates so we
2
hypothesized that variations distal to the phosphorus were unlikely to change the
hydrolysis reactivity of similar compounds. Given this and the wide range of
applications of compound 1 we wanted to further explore carbon-phosphorus based
phosphordithioates with the aim to understand the role of the substituents, both
directly attached and distal to the phosphorus, in controlling the release rates of H S,
2
3
7
assuming a predominantly water hydrolysis mechanism (SI Figure 89). We selected
1 as our lead compound due to its reported drug-like properties which gave us reason
to believe that this structural class holds particular promise for eventual development
of a H S donor clinical candidate.
2
Direct analogues of 1 were prepared by ring-opening of Lawesson's Reagent (LR) in
dichloromethane at room temperature, similar to the previously described
37
7
procedure. As well as amines used as per the synthesis of 1, phenols and alcohols
were employed as nucleophiles (Table 1).
ACS Paragon Plus Environment

Page 17 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
1
Table 1. Synthesis and P NMR chemical shifts of H S donor compounds.
2
S
morpholinea
Nucleophile
solvent
S
RO
P
S
S
P
S
RO
P
S H N
2
OR
Nu
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Lawesson's reagent
3
1
Compound
R
Nucleophile
Yield
P NMR δ
(%)
(ppm, DMSO-d6)
1
Me
67
89.00
N
O
4
5
Me
Me
OMe
OPh
31.1
65.5
107.75
106.7
O
6
7
Me
50
111.90
F
F
O
Me
Me
36.3
80
107.39
108.49
OMe
b
8
O
O
N
H
b
9
Me
Me
Me
Me
NH2
72.2
52
106.95
103.33
104.82
80.31
i
1
0
1
O Pr
n
1
O Bu
20.8
30.0
c
12
NHCH Ph
2
N
O
13
Ph
Ph
25.5
37
87.17
79.71
c
14
NHCH Ph
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
1
Compound
R
Nucleophile
Yield
%)
P NMR δ
(
(ppm, DMSO-d6)
106.83
1
5
6
Ph
Ph
OMe
25
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
i
1
O Pr
102.63
a
b
c
With the exception of compounds 7, 8 and 11; isolated as a zwitterion; isolated as a benzylamine
salt.
Products were precipitated from the reaction mixtures by adding morpholine, or
another amine, to form ionic salts which could be purified by filtration, washing and
drying. This process was not successful with all attempted reactions (see experimental
section). For some nucleophiles, like anilines, pyrazole and thiophenols, no
precipitation occurred. Carbon nucleophiles also did not give desired products. All of
1
13
31
the isolated compounds were characterized with H, C, PNMR, LRMS and/or HR
ESI mass spectroscopy analysis. Due to their unstable nature in aqueous systems we
were unable to obtain satisfactory HPLC data. To confirm our NMR assignments the
molecular structure of isopropoxy compound 10 was determined via single crystal
X-ray structure analysis. This structure clearly indicated a tetrahedral phosphorus with
attached isopropoxy and methoxyphenyl groups, with morpholine as the counter
ion (Figure7), confirming the structural assignments.
ACS Paragon Plus Environment

Page 19 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 7. Single crystal X-ray structure of 10 confirming the structure
with the phosphorus core atom bearing two sulfurs, an isopropoxy
group, methoxyphenyl and a morpholine counterion.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
With these compounds, H S is released most probably through hydrolysis by water.
2
Therefore it is expected that the electrophilicity of the phosphorus atom will greatly
affect the hydrolysis-mediated H S release rate. Hence, modifying the electron density
2
of the phosphorus atom by varying the nature of the nucleophile should lead to
changes in hydrolysis rate and hence rate of H S generation. Both electron-donating
2
(compound 5) and electron-withdrawing (compound 4) phenols were successfully
employed in the synthesis. Alkyl alcohols (compounds 4, 8 and 9) were used to
further broaden the electrophilicity range of the phosphorus atom. Compared to the
3
1
P NMR shift of 1 (c.90 ppm) and other amine nucleophiles, compounds derived
3
1
from alkyl alcohols and phenols had P NMR shifts in the range 102 -112 ppm.
Considering the expected large downfield shift when oxygen is replaced by sulfur,
8
these chemical shifts are in the expected range. Compared with 1, the further
downfield chemical shift of P-O compounds (compounds 4-11 and 15-16) is likely
due to the increased electronegativity of the oxygen atom increasing the positive
charge on phosphorus and in turn increasing back-bonding from the sulfur which is
9
associated with the upfield shift displayed. The electron-withdrawing (e.g.
111.90 ppm for 6) and electron-donating (e.g. 107.39 ppm for 7) substituted phenols
3
1
only slightly affect the P NMR chemical shift due to the strong proximity of the
electronegative oxygen compared to the rather distant placement of the substituent.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compared to phenols, alkyl alcohol nucleophiles led to up-field chemical shifts of
around 100 ppm. The effect of substituents of the aryl ring directly attached to the
phosphorus atom were also studied by changing the methoxy (OMe) group to a
phenoxy (OPh) group using Belleau's reagent. The range of available Lawesson’s
reagent (LR) derivatives is extremely limited hence we have not yet expanded our
studies of substituents of this aryl ring. However, using Belleau's reagent, four
phenoxyphenyl derivatives bearing amine and alkyl alcohol groups were prepared
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
31
(compounds 13-16).The phenoxy group did not significantly affect the P NMR
shifts for alcohols (compare 15 with 4 and 16 with 10) and only slightly for
morpholine (compare 13 with 1). In addition to the two tertiary amine morpholine
derivatives two secondary amines were also prepared with benzylamine, compounds
3
1
1
2 and 14, both with very similar P chemical shifts but significantly shifted well
3
1
below 90 ppm compared to the oxygen substituted derivatives. The range of
P
chemical shifts achieved with this group of compounds was expected to lead to
differences in their reactivity and H S release rates. These rates were next
2
characterized using the optimized assay with probe 2.
Quantification of Evolved H S In Vitro and Rate of Release
2
Having these compounds in hand, we turned our attention to the determination of their
H S release rates in the optimized PBS:MeCN buffer system. Generally, these
2
compounds release H S very slowly, ideal for biological applications, hence the
2
parent compounds must be studied at high concentrations. At concentrations of
0
.2 mM, H S is barely detectable in this system in most cases of slow release
2
ACS Paragon Plus Environment

Page 21 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds. At 1 mM compound concentrations, released H S is quantifiable with
2
many slow release compounds. However only when the compounds were tested at the
highest concentration of 5 mM could significant amounts of H S be detected and
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
quantified. Either derived from Lawesson's Reagent or Belleau's reagent, all
compounds bearing phenol nucleophiles (5, 6 and 7) released around 20 μM H S after
2
a four-day incubation at 22 ºC (Figure 8A and Table 2).
A:
25
20
15
10
5
0
2
4 h
48 h
72 h
96 h
4
5
6
7
9
10
11
15
16
blank
Compound
B:
2
2
1
1
50
00
50
00
2
4 h 48 h 72 h 96 h
50
0
1
12
13
14
Compound
Figure 8. Quantification of H S evolved from Table 1 compounds at
2
2
2 ºC over 4 days at 5 mM. A: slow releasing compounds (Compound 8
(not shown) same as blank). B: faster releasing compounds.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
Table 2. Summary of H S Quantification Data
2
Percentage of
theoretical
Percentage of
theoretical
2
H S
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Cpd
2
H S
Normalised Normalised
S Release S Release
Release
µM,
2 °C)b
H
2
S Release
Ratio of
Compound Concn
Release
maximum H
2
S
maximum H
released
(37 °C)
6.14%
2
S
H
2
H
2
(µM, 37 °C)b
(
37/22 °C
(
mM) Categorya
released
(22 °C)f
(37 °C)f
2
(
22 °C)
1
4
5
6
7
8
9
5
5
5
5
5
5
5
Medium
Low
143±10
5.9±0.24
21±0.74
22±1.7
1.43%
0.06%
0.21%
0.22%
0.19%
<0.01%
0.02%
614±82d
15±1.2
4.3
2.5
2.0
1.4
1.6
2.7e
2.7
1.0
<0.1
0.1
0.2
0.1
-
1.0
<0.1
0.1
0.15%
Low
41±2.2
0.41%
Low
31±2.4
0.31%
0.1
Low
19±0.85
<LOD
31±2.8
0.31%
0.1
Low
2.7±0.78
6.1±0.58
0.03%
<0.1
<0.1
Low
2.3±0.24
0.06%
<0.1
ACS Paragon Plus Environment

Page 23 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
0
1
2
3
4
5
5
Low
Low
4.0±0.21
3.8±0.18
219±5.8
168±7.4
128±8.7
6.0±0.25
5.8±0.33
0.04%
12±0.62
12±0.69
507±164d
554±120d
281±23d
16±2.0
0.12%
3.0
3.2
2.3
3.3
2.2
2.7
2.6
2.7
1.1
0.8
1.3
<0.1
<0.1
1.5
<0.1
<0.1
0.8
0.04%
2.19%
1.68%
1.28%
0.06%
0.06%
0.82%
67.2%
2.4%
0.12%
5.07%
5.54%
2.81%
0.16%
0.15%
21.8%
76.2%
1.9%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
5
Medium
Medium
Medium
Low
5
1.2
0.9
5
0.9
0.5
15
5
<0.1
<0.1
<0.1
<0.1
1
1
1
1
2
6
7
8
9
0
5
Low
15±1.3
0.5
0.5
0.5
0.5
High
82±14c
218±64c
6.8g
31.6h
672±19c
762±24c
56g
110.4h
High
Medium
High
24±1.4c
19±3.6c
2.0g
40g
2.8h
485±14c
48.5%
626±39c
62.6%
90.7h
.0±0.1b
1.9±0.23b
1
Blank
-
-
-
-
-
1
2±0.25c
4.9±0.43d
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
6
.9±0.32c
a
Release rates classified as low ( <50 µM H S over 96 h at 5 mM), medium (>50 µM H S over 96 h at 5 mM or <50 µM H S over 48 h at
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
b
c
0
.5 mM), high (>50 µM H S over 48 h at 0.5 mM),. assay at 5 mM compound using 0.2 mM probe at 96 h unless otherwise stated. 0.5 mM
2
d
e
f
compound with 1 mM probe after 48 hours. 5 mM compound using 1 mM probe at 96 h. taken as a ratio of the blank. normalized to
g
h
compound 1. Cpd 1 used as blank (12 µM). Cpd 1 used as blank (6.9 µM). Data for Cpd 1 was taken from assays conducted in the same
plates as the other compounds and on average was about 3-fold higher [H S] at room temperature compared to the earlier experiments during
2
assay development.
ACS Paragon Plus Environment

Page 25 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
When the temperature was increased to physiological levels, 37 ºC (Figure 9A),
approximately 31-41 μM H S was detected from 5, 6 and 7 after four days,
2
representing less than 0.5% of the theoretical maximum amount of available H S from
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
these compounds.
A:
45
40
35
30
25
20
15
10
5
0
24 h
48 h
72 h
96 h
4
5
6
7
8
9
10
11
15
16
blank
Compound
B:
7
6
00
00
2
7
4 h
2 h
48 h
96 h
500
4
3
2
00
00
00
100
0
1
12
13
14
blank
Compound
Figure 9. Quantification of H S evolved from Table 1 compounds at 37 ºC over 4 days at
2
5
mM. A: slow releasing compounds. B: faster releasing compounds at 5 mM.
However, compounds derived from alkyl alcohols (4, 9, 10, 11, 15 and 16) released
much less H S, only 2.3–6 μM at 22 ºC (Figure 8A and Table 2) and 6.1–16 μM at
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
7 ºC (Figure 9A and Table 2), only up to 0.16% of available sulfur. Derived from
Belleau's reagent, 15 and 16 released a similar amount of H S to their LR-derived
2
counterparts, 4 and 10. Compound 8, bearing an ethylaminoethoxy side chain did not
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
release any detectable H S after 4 days at 22 °C and was only just quantifiable at
2
37 °C. This may due to an electrostatic intramolecular interaction forming a six
membered ring between the positively charged nitrogen atom and negatively charged
sulfur atom, reducing the conformational freedom and reactivity of the phosphorus
centre. Compounds 13 and 14, which were derived from Belleau's reagent and amine
nucleophiles morpholine and benzylamine, respectively, and the corresponding
LR-derived 1 and 12 released about 5-6 fold more H S than the alcohol/phenol
2
derived compounds at 22ºC, equating to approximately 1.2-2.2% of the theoretical
maximum amount of H S (Figure 8B). When 1, 12, 13 and 14 were tested at 37 ºC, a
2
2
.2-3.3-fold increase in accumulated H S was detected at the end of the 4-day assay
2
period, equating to between 281 and 554 µM H S which is comparable with that of 1
2
(614 µM), and representing 2.8 to 5.5% of theoretical available sulfur (Figure 9B).
The higher increase for 1 going from 22 to 37 °C may be one reason why compound 1
has found so many applications in biological systems. The question as to what is the
‘ideal’ H S release profile for a H S donor compound will probably depend on a
2 2
multitude of factors, not just temperature. Intracellular permeability of the compound,
pH sensitivity and sub-cellular localization are all relevant factors to consider as well
as the underlying biological effects of the molecular scaffold as a whole. We were
interested in whether the compounds were entering cells or remaining in the
ACS Paragon Plus Environment

Page 27 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
extracellular matrix hence we measured permeability of a number of compounds
using the parallel artificial membrane permeability assay (PAMPA) (see below).
Encouraged by this range of H S release profiles from dithiophosphorus derivatives
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bearing one carbon and one heteroatom, we next explored increased combined
electronegativity of the atoms attached to phosphorus. Compounds bearing two
heteroatoms on phosphorus fulfilled this objective. A series of compounds were
synthesized from P S in a one step protocol (compounds 17-20, Table 2).
4
10
3
1
Table 3. Synthesis and P NMR chemical shifts of faster-releasing H S donor
2
compounds.
S
R1
R2
DCM
12 eq Amine
N P S H N
2
P S
+
10
4
R2
N
R1
R1 R2
31
Compound
N(R )R
Yield (%)
P NMR δ
1
2
(ppm, DMSO-d6)
NH
1
7
8
31.1
65.5
0.19
0.32
1
N
O
19
NHCH Ph
50
0.33
0.74
2
20
36.3
N
S
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 103
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Simply mixing P4S10 with amines, isolable products precipitate after stirring for a
period of time. Pure products can be obtained by filtration and washing. These
products are quite hydrophilic and degrade quite rapidly hence were kept under dry
conditions. For example, 20 will become sticky even when exposed to air for less than
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
31
one hour. As expected, these compounds have P NMR chemical shifts at around 0
ppm, significantly up-field of compounds in Table 1. When compounds 17-20were
incubated at one tenth the concentration of the less reactive compounds, 0.5 mM, at
2
2 and 37 ºC, a rapid release of H S was detected. In particular, tertiary amines 18 and
2
2
0 were very rapid releasers of H S producing 672 and 485 µM, respectively, at 22 °C
2
after only 2 days (Figure 10A and Table 2).
A:
800
0
.5 h
5 h
24 h
48 h
7
6
00
00
500
4
3
00
00
200
1
00
0
17
18
19
20
blank
Compound
B:
ACS Paragon Plus Environment

Page 29 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
9
00
0
.5 h
5 h
24 h
48 h
800
7
6
00
00
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
500
4
3
2
00
00
00
100
0
1
7
18
19
20
blank
Compound
C:
60
0
.5 hours
4 hours
24 hours
48 hours
5
4
3
2
0
0
0
0
10
0
1
7
18
19
20
blank
Compound
Figure 10. Quantification of H S evolved from faster releasing
2
compounds at 0.5 mM, using 1 mM probe 2, over 48 hours. A: 22 ºC.
B: 37 ºC. C: 0.1 mM compound, 0.2 mM 2.
At 37 °C, the amounts of H S produced were even larger representing 76.2 and 62.6%
2
of the theoretical maximums for 18 and 20, respectively (Figure 10B). At the lower
concentration of 0.1 mM, a significant amount of H S, around 20% and 5% of the
2
theoretical maxima, respectively, were detected for compounds 18 and 20 at 37 ºC
ACS Paragon Plus Environment