Discovery and optimization of 4,5-diarylisoxazoles as potent dual inhibitors of pyruvate dehydrogenase kinase and heat shock protein 90

Article abstract of DOI:10.1021/jm5010144

Upregulation of pyruvate dehydrogenase kinase (PDHK) has been observed in a variety of cancers. Inhibition of PDHK offers an attractive opportunity for the development of novel cancer therapies. To obtain novel PDHK inhibitors, we took advantage of the homology of the ATP-binding pocket between Heat Shock Protein 90 (HSP90) and PDHK, and utilized 4,5-diarylisoxazole based HSP90 inhibitor for structural design. Our efforts led to the identification of 5k that inhibited PDHK1 with an IC₅₀ value of 17 nM, which, however, showed marginal cellular activity. Further structural optimization resulted in compound 11a with improved cellular activity which could effectively modulate the metabolic profile of cancer cells and lead to the inhibition of cancer cell proliferation, evidenced by the increased oxidative phosphorylation and decreased glycolysis and associated oxidative stress. Our results suggested 11a as an excellent lead compound and a favorable biological tool to further evaluate the therapeutic potential of PDHK and HSP90 dual inhibitors in the treatment of cancer.

Full text of DOI:10.1021/jm5010144

Subscriber access provided by EASTERN KENTUCKY UNIVERSITY
Article
The Discovery and Optimization of 4,5-Diarylisoxazoles
as Potent Dual Inhibitors of Pyruvate Dehydrogenase
Kinase (PDHK) and Heat Shock Protein 90 (HSP90)
Tao Meng, Dadong Zhang, Zuoquan Xie, Ting Yu, Shuchao
Wu, Min Huang, Mei-Yu Geng, and Jingkang Shen
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm5010144 • Publication Date (Web): 10 Nov 2014
Downloaded from http://pubs.acs.org on November 23, 2014
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 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The Discovery and Optimization of
4,5ꢀDiarylisoxazoles as Potent Dual Inhibitors
of Pyruvate Dehydrogenase Kinase (PDHK)
and Heat Shock Protein 90 (HSP90)
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
Tao Meng1#, Dadong Zhang2#, Zuoquan Xie2, Ting Yu1, Shuchao Wu2, Min Huang2,*,
Meiyu Geng 2,*and Jingkang Shen1,*
1 Division of Medicinal Chemistry, State Key Laboratory of Drug Research, Shanghai
Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203,
People’s Republic of China
2 Division of Antitumor Pharmacology, State Key Laboratory of Drug Research,
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai
201203, People’s Republic of China
# These authors contributed equally to this work.
KEYWORDS: Pyruvate Dehydrogenase Kinases (PDHKs), Heat Shock Protein 90
(HSP90), 4,5ꢀDiarylisoxazoles, Dual Inhibitors
ABSTRACT: Upregulation of pyruvate dehydrogenase kinase (PDHK) has been
observed in a variety of cancers. Inhibition of PDHK offers an attractive opportunity
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 45
1
2
3
4
5
6
7
8
9
for the development of novel cancer therapies. To obtain novel PDHK inhibitors, we
took advantage of the homology of the ATPꢀbinding pocket between HSP90 and
PDHK, and utilized 4,5ꢀdiarylisoxazole based HSP90 inhibitor for structural design.
Our efforts led to the identification of 5k that inhibited PDHK1 with an IC50 value of
17 nM, which however, showed marginal cellular activity. Further structural
optimization resulted in compound 11a with improved cellular activity which could
effectively modulate the metabolic profile of cancer cells and led to the inhibition of
cancer cell proliferation, evidenced by the increased oxidative phosphorylation and
decreased glycolysis and associated oxidative stress. Our results suggested 11a as an
excellent lead compound and a favorable biological tool to further evaluate the
therapeutic potential of PDHK and HSP90 dual inhibitors in the treatment of 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
INTRODUCTION
Most cancer cells feature a switch in metabolism from mitochondriaꢀbased glucose
oxidation to cytoplasmꢀbased glycolysis even under normoxia, also known as
“Warburg effect”. Such a metabolic profile, which is characterized by increased
glycolysis and suppressed glucose oxidation, provides cancer cells with a proliferative
advantage. Thus, to disrupt metabolic dependencies of cancer cells may represent a
promising anticancer strategy. Indeed, increasing studies have suggested that targeting
cancerꢀspecific metabolic remodeling offer therapeutic opportunities in cancer
treatment.1
Pyruvate dehydrogenase kinase (PDHK) is a mitochondrial enzyme that inhibits
pyruvate dehydrogenase complex, a gatekeeper of glucose oxidation, and prevents
2
ACS Paragon Plus Environment

Page 3 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
converting pyruvate to acetylꢀCoA, the substrate for the Krebs’cycle. Activation of
PDHK results in the uncoupling of glycolysis to glucose oxidation. Four isoforms,
PDHK1, PDHK2, PDHK3 and PDHK4, have been identified in human cells, among
which, PDHK1 is closely associated with cancer malignancy. PDHK1 is activated in
various cancers such as lung cancer2 and head and neck squamous cancer.3 PDHK1 is
transcriptionally regulated by oncogenes cꢀMYC and hypoxiaꢀinducible factor1α
(HIF1α)4 to control metabolic and malignant phenotypes of cancer cells.5 A recent
study has also revealed that PDHK1 is commonly phosphorylated in human cancers
by diverse oncogenic tyrosine kinases, which is important for promoting Warburg
effect and tumor growth.6 These evidence collectively suggest therapeutic
opportunities arising from targeting PDHK in cancer therapy. In spite of the
increasing number of reported PDHK inhibitors, such as dichloroacetate (DCA),7
none of them has successfully entered clinical use.
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
Figure 1. Known PDK inhibitors: Radicicol, Dichloroacetate (DCA), AZD7545, and
Nov3r.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 45
1
2
3
4
5
6
7
8
9
So far three major classes of PDHK inhibitors have been reported, which inhibit
PDHK via different mechanisms. AZD7545 and Nov3r (Figure 1) contain the
trifluoromethylpropanamide group and represent a class of PDHK inhibitors binding
to the lipoylꢀbinding pocket of the enzymes.8 DCA (Figure 1) inactivates PDHK
kinase activity via binding to the helix bundle in the Nꢀterminal.9 Radicicol (Figure 1)
belongs to a class of ATPꢀcompetitive inhibitor and inhibits kinase activity by binding
to the ATPꢀbinding pocket of PDHK, also known as GHKL domain.10 Notably,
GHKL domain has also been found in several other ATPꢀbinding proteins, including
histidine kinase, DNA gyrase B, topoisomerases,11 and heat shock protein HSP90.12
The folding of these GHKL domains is similar and over 50% of residues from
PDHKs are superimposable with the corresponding residues in HSP90.12b This notion
suggests that structureꢀactivity relationship (SAR) information of HSP90 inhibitors
could possibly guide the design of ATPꢀcompetitive PDHK inhibitors.13
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 fact, two HSP90 inhibitors Radicicol and M77976 are known to bind weakly to
the ATP binding site of PDHK3 and PDHK4, respectively. M77976 showed an IC50
of 648 ꢁM against PDHK4,14 whereas radicicol inhibited PDHK with a similar low
potency of 400 ꢁM (IC50) against PDHK1 and 230 ꢁM against PDHK3.12b, 15
Structurally, M77976 belongs to the 4,5ꢀdiarylꢀsubtituted azole class. We speculated
that the SAR for HSP90 might be translated to the activity for PDHK. Therefore, we
examined AUY922, another HSP90 inhibitor belonging to 4,5ꢀdiaryl azole class but
20ꢀfold more potent than M77976. As expected, the isoxazole core of AUY922
enabled enhanced activity against PDHK1 compared to M77976, with an IC50 of 11.9
4
ACS Paragon Plus Environment

Page 5 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ꢁM. Meanwhile, we tested PDHK1 inhibitory activity of several known HSP90
inhibitors with different scaffolds, including purine analog (BIIB021), benzamide
analog (SNXꢀ2112) and resorcinol derivative (AT13387), but no active compounds
were found at a concentration up to 30 ꢁM.
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
Scheme 1. Design of PDHK inhibitors based on the combination of known
PDHK4 and HSP90 inhibitors
In spite of significantly improved potency against PDHK1, the potency of AUY922
against HSP90 appeared much higher (IC50 = 0.021 ꢁM),16 indicating that the affinity
of oxazoleꢀbased AUY922 for PDHK1 was at least 500ꢀfold lower than that for
HSP90. In order to further increase the potency of PDHK inhibition, we synthesized
two reported HSP90 inhibitors (compounds 1 and 2) derived from M77976 and
AUY922 (Scheme 1) and tested their activity in inhibiting PDHKs.16 The replacement
of iPr group (AUY922) with chloro (compound 1) resulted in about 10ꢀfold increase
of the activity for PDHK1. Replacing the 4ꢀmorpholinomethyl group with 4ꢀmethoxyl
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 45
1
2
3
4
5
6
7
8
9
(compound 2) slightly increased the PDHK1 potency and relative selectivity between
PDHK1 and HSP90. Further inspection of this structure indicated that the amide
group was suitable sites for modification. Thus, SAR studies were carried out by
systemic molecular modulation on the compound 2. The ethylamino group was
replaced by incorporating solubilizing groups and meanwhile to probe the possibility
of additional interaction with the amino acid side chains of PDHKs. Further
optimization was centered on 3ꢀacrylamide substituted isoxazole derivatives in order
to improve the cellular activity (Scheme 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
Scheme 2. Structural modification of compound 2
CHEMISTRY
The initial set of isoxazole derivatives 1 and 2, as well as the key intermediate 3
(Scheme 3) were synthesized according to known literature procedures.16 Hydrolyzing
compound 3 with sodium hydroxide gave acid 4. This acid was treated with oxalyl
chloride to afford the corresponding acid chloride, which was used for subsequent
acylations and boron trichlorideꢀmediated deprotection to yield the amides 5aꢀe. Acid
4 can also be coupled with 1ꢀNꢀBocꢀ4ꢀaminopiperidine to afford amide 12, which was
deprotected with 50% TFA to generate 13. With intermediate 13 in hand, compound
6
ACS Paragon Plus Environment

Page 7 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5fꢀm were prepared by either acylation (5f), sulfonation (5g), ureation (5h) or
reductive amination (5iꢀm) of the intermediate 13. 3ꢀAcrylamide substituted isoxazole
derivatives 10aꢀi were made using alternative procedures (Scheme 3) from the key
intermediate 3 as well. Reduction of compound 3 under LiAlH4 condition was utilized
to afford the alcohol 6, which was oxidized with activated MnO2 to give the aldehyde
7, followed by HornerꢀWadsworthꢀEmmons reaction giving α,βꢀunsaturated ester 8.
This ester can be hydrolyzed with aqueous sodium hydroxide to afford acid 9.
Amidation with various amines, followed by debenzylation with boron trichloride
provided the products 10aꢀi. Some compounds of this class were made by a modified
sequence, in which compounds 10bꢀi were demethylated with boron tribromide to
afford the phenolic compounds 11aꢀh (Scheme 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
Scheme 3. Conversion of Key Intermediate 3 to Isoxazole PDHK1 Inhibitorsa
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 45
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Reagents and conditions: (a) NaOH, ethanol; (b) (i) Oxalyl chloride, DCM, 0
oCꢀrt; (ii) triethylamine, DCM, 0 oCꢀrt; (c) 1M BCl3 in DCM, DCM; (d) (i) Oxalyl
o
chloride, DMF, DCM; (ii) DIPEA, DCM; (e) CF3COOH, 0 C; (f1) acetyl chloride,
triethylamine, DCM; (f2) methanesulfonyl chloride, triethylamine, DCM; (f3)
isocyanatoethane, triethylamine, DCM; (f4) RꢀCHO, NaBH(OAc)3, AcOH,
1,2ꢀdichloroethane, 0 oCꢀrt; (g) 1M BCl3 in DCM, DCM; (h) LiAlH4, THF, 0 oC; (i)
o
MnO2, DCM; (j) 60% NaH, THF, 0 C; (k) NaOH aq., ethanol; (l) RNH2, EDCI,
HOBt, triethylamine, DCM; (m) 1M BCl3 in DCM; (n) 1M BBr3 in DCM
RESULTS AND DISCUSSION
8
ACS Paragon Plus Environment

Page 9 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Therefore, a set of isoxazole amide substituted derivatives with different amino
group were firstly synthesized and evaluated for their activity against PDHK. PDHK1
kinase activity was assessed using an enzymeꢀlinked immunosorbent assay (ELISA)
using recombinant PDHK1, and radicicol was used as a reference compound. HSP90
activity was assessed by luorescence polarization (FP) assay. All the compounds were
evaluated at an initial concentration of 10 ꢁM against PDHK1. Compounds showing
over 50% inhibition rate at 10 ꢁM were further evaluated to measure IC50 values. The
PDHK1 inhibitory activities of the isoxazole amide derivatives (5aꢀm) were shown in
Table 1. It was observed that the variation of the amino group affected the biological
activity of the synthesized analogues. These compounds in general showed enhanced
PDHK1 inhibitory activity in comparison to compound 2. Four compounds were
synthesized with the Rꢀgroup selected as: 2,4ꢀdichlorobenzylamine (5a),
N,Nꢀdimethylaniline (5b), 4ꢀpyridyl (5c), 1ꢀbenzylꢀNꢀmethylpiperidinꢀ4ꢀamine (5d).
It seemed the three benzylamine (5a) and aniline (5bꢀc) showed similar activities
compare to 2 with ethylamino group. Interestingly, compound 5d showed dramatic
increase of PDHK1 inhibitory activity with IC50 of 25 nM, which suggested that the
basic nitrogen may contribute to the high potency. To test this hypothesis, 5e, a
compound lack of the basic nitrogen was prepared. Indeed, 5e lost the activity against
PDHK1. Furthermore, the introduction of acetyl (5f), methylsulfonyl (5g) and
ethylurea (5h) in the piperidine also dramatically decreased the activity compare to
their parent compound 5d, proving the importance of the basicity of the nitrogen in
the piperidine ring. Further modification of 4ꢀaminopiperidine revealed additional
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
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 45
1
2
3
4
5
6
7
8
9
gain of activity by replacing the benzyl group (5d) with various substituted phenyl
ring, heterocycles, or alkyl groups. It suggested that the electronic withdrawn group
(2,4ꢀdichloro, 5j) substituted phenyl ring decreased the activity, whereas the
electronic donating group (4ꢀmethoxybenzyl substitution) retained or slightly
increased the activity, with examples 5k giving IC50 of 17 nM. 4ꢀPyridylmethyl (5i)
or other alkyl substitutions (5l, 5m) retained or slightly decreased the activity.
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
Table 1. Effects of Amides of 4,5-Diarylisoxazole on Inhibition of PDHK1 in
ELISA assaya
cpds
5a
R Group
PDHK1 IC50 (µM)
HSP90 IC50 (µM)
N
0.61 ± 0.03
1.600 ± 0.028
H
Cl
Cl
5b
5c
0.38 ± 0.02
0.58 ± 0.15
0.091 ± 0.003
0.154±0.022
5d
5e
0.025 ± 0.005
> 10
0.036 ± 0.002
>10
10
ACS Paragon Plus Environment

Page 11 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5f
0.21 ± 0.02
0.13 ± 0.07
0.025 ± 0.003
0.045 ± 0.007
5g
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
5h
5i
0.15 ± 0.01
0.020 ± 0.006
0.015 ± 0.004
0.059 ± 0.018
5j
5k
5l
0.16 ± 0.07
0.017 ± 0.004
0.068 ± 0.023
0.127 ± 0.009
0.016 ± 0.001
0.023 ± 0.001
5m
0.028 ± 0.007
0.039 ± 0.002
a The IC50 was calculated from two independent experimental measurements.
Table 2. Effects of Amides of New Vinylogs on Inhibition of PDHK1 in ELISA
assaya
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 45
1
2
3
4
5
6
7
8
cpds
10a
X
R
PDHK1 IC50 (µM)
HSP90 IC50 (µM)
OCH3
1.916 ± 0.631
0.677 ± 0.081
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
48
49
50
51
52
53
54
55
56
57
58
59
60
10b
10c
10d
OCH3
OCH3
OCH3
0.845 ± 0.200
0.557 ± 0.027
0.491 ± 0.094
0.596 ± 0.114
0.700 ± 0.033
0.718 ± 0.095
10e
OCH3
1.050 ± 0.129
0.215 ± 0.027
10f
10g
10h
11a
11b
11c
OCH3
OCH3
OCH3
OH
1.070 ± 0.471
4.306 ± 2.362
2.369 ± 0.514
0.604 ± 0.054
0.952 ± 0.073
0.776 ± 0.168
0.208 ± 0.021
4.232 ± 0.747
2.255 ± 0.283
0.512 ± 0.044
0.576 ± 0.066
0.666 ± 0.021
OH
OH
12
ACS Paragon Plus Environment

Page 13 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
11d
11e
11f
11g
OH
OH
OH
OH
0.878 ± 0.313
0.521 ± 0.113
1.417 ± 0.436
2.949 ± 0.721
0.357 ± 0.066
0.338 ± 0.083
3.939 ± 0.378
1.859 ± 0.001
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
a The IC50 was calculated from two independent experimental measurements.
Our initial SAR exploration of the amide substituted isoxazoles led to the discovery
of 5d and 5k as PDHK1 inhibitors with great potency. We then proceeded to evaluate
the cellular activity of 5d and 5k in cancer cells. However, compounds 5d and 5k
failed to show activities in the cellular assay (Figure 2C). We suspected that the lack
of cellular activity was most likely attributed to the poor permeability of this structure
class. We therefore tested permeability coefficient (Papp) of compound 5d and 5k in
Cacoꢀ2 cells (Table 5), which suggested that permeability was indeed a concern (Papp
< 1× 10ꢀ6 cm/s). Next, we turned our attention to the vinylogues, which may change
the conformation of these compounds and improve permeability.17 Based on the
above SAR information, 4ꢀaminopiperidine was replaced with Nꢀacrylamide
substituted piperizines. Data presented in Table 2 showed that the new vinylogs
(10aꢀh and 11aꢀg) all exhibited moderate inhibitory activity against PDHK1. The data
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 45
1
2
3
4
5
6
7
8
9
indicated the substituents on the aromatic ring of the arylpiperizine had some effect
on their activity against PDHK1. In general, the introduction of EWGs on the phenyl
ring decreased activity compared with the EDGs substituted phenyl derivative or
electronic rich aromatic rings. The compounds 10bꢀd having 2ꢀ, 3ꢀ and
4ꢀmethoxypheyl at R position showed improved activity compared with the
compound 10a with nonꢀsubstituted phenyl ring, whereas, compounds 10g and 10h
bearing the EWGs such as 3,4ꢀdichloro, 3ꢀtrifluoromethyl group showed decreased
activity. Compounds 10eꢀg containing the heterocyclic substituted piperizine also
showed similar PDHK1 inhibitory activities. Demethylation of compounds 10bꢀi
afforded the corresponding compounds 11aꢀh. As shown in Table 2, most of the
phenolic derivatives retained the inhibitory activity and the SAR showed similar
pattern as their parent compounds. We then measured the cellular activity of these
compounds (compounds 11aꢀh, and compound 5k) using a high content analysis,
which examined the impact on the downstream PDH phosphorylation. As shown in
Figure 2AꢀB, most of these scaffolds showed good cellular activity. Compounds 10b,
10c, 10d, 10e, 10f, 11a, 11b, 11c, 11d and 11e effectively decreased phosphorylation
of both S293 and S232. The enzymatically less potent compounds 11g only slightly
decreased the phosphorylation of S232 but not that of S293. The cellular activity of
the three most potent compounds 10f, 11a and 11c were further verified using
immunoblotting analysis, where 10 ꢁM of each compound almost abolished PDH
phosphorylation at both sites (Figure 2C).
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
14
ACS Paragon Plus Environment

Page 15 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We also compared the activity of these compounds on PDHK1 and HSP90, which
overall showed a comparable inhibition against the two enzymes (Table 1 and Table
2). For instance, compounds 10f, 11a and 11c showed at least 10ꢀfold decrease than
compounds 5d, 5k and 5m in both HSP90 and PDHK1 inhibition at the molecular
level, which may be explained by the similar ATP binding affinity of PDHK and
HSP90. ATP binding affinity for HSP90 is at hundred micromolar level, with a Km
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
20
range from 770 to 840 µM 18, 19, comparable to ATP binding affinity to PDHK . At
the cellular level, both compound sets exhibited apparent cellular impacts on HSP90
activity, as indicated by the dramatic downꢀregulation of HSP90 client proteins such
as AKT, HER2 and cꢀMET, and some upꢀregulation of its client protein HSP70
(Figure 2C). Together, compounds 11a, 11c and 10f exhibited apparent inhibition on
PDHK1 besides its inhibition HSP90, in spite of a decrease of activity on HSP90
when compared to compounds 5d, 5k and 5m.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 45
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Cellular activity of compounds on PDHKs inhibition; A, B) NCIꢀH1299
cells were treated with indicated compounds for 12 h (10 µM); AZD7545 was a
16
ACS Paragon Plus Environment

Page 17 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
positive contro1. The decrease of PDH phosphorylation (S293 and S232) was a
reflection of PDHK inhibition, and was determined by high content analysis, in which
cells were defined as positive or negative according to their relative fluorescence
intensity compared to the average intensity; The data were expressed as the percent of
positive cells versus total cells; C) Activity of compounds were determined by
immunoblotting analysis. AZD7545 and radiciciol were used as positive controls;
*P<0.05, versus untreated control group.
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
We then asked whether the enhanced cellular potency of was caused by improved
cell membrane permeability. Papp of selected compounds 10f, 11a and 11c was
measured and compared with that of 5d and 5k. Consistent with their cellular activity,
compounds 5d and 5k showed low permeability (
ratio (Table 3), whereas compounds 10f, 11a and 11c exhibited moderate
permeability (
app > 5 × 10ꢀ6 cm/s) and low efflux ratio (Table 3). This result indicated
P
app < 1× 10ꢀ6 cm/s) and high efflux
P
that the compromised cellular activity of compound 5d and 5k may result from the
low cell membrane permeability and high efflux ratio. Presumably, compounds pass
through the additional barrier of mitochondria membranes may have much more
impacts on their activity against PDHK1 at cellular level, as PDHK1 is located in
mitochondrial which is different from HSP90 that localized at cytosol and easy to
reach for compounds. Therefore, compounds 10f, 11a and 11c were supposed to be
easier to reach mitochondrial PDHK, and would be considered in further
optimization.
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 45
1
2
3
4
Table 3. Permeability of selected compounds
5
6
7
8
9
Papp (× 10ꢀ6 cm/s)
(AP→BL)
Papp (× 10ꢀ6 cm/s)
(BL→AP)
cpds
Efflux ratio
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
5d
5k
0.936 ± 0.029
0.962 ± 0.012
6.90 ± 0.03
11.2 ± 0.4
9.41 ± 0.27
8.70 ± 0.21
5.09 ± 0.03
15.5 ± 0.9
15.0 ± 0.2
10.05
9.04
0.74
1.38
2.29
10f
11a
11c
6.54 ± 0.14
To gain a better understanding of PDHK inhibitory activity and associated
metabolic modulation efficacy of the compounds, compound 11a, the most potent
compound on both enzymatic and cellꢀbased assays, was selected as a representative.
First, we determined the mode of action of the compound. As a compound derived
from the ATP competitive inhibitors, the ATP competitiveness of the compound was
determined using LineweaverꢀBurk plot. Radicicol, an ATP competitive inhibitor of
PDHK, was selected as a positive control. The result showed that compound 11a
inhibited PDHK1 activity in an ATPꢀdependent manner, with curves crossing at the
yꢀintercept (Figure 3A), suggesting that compound 11a competed with ATP to access
the ATPꢀbinding pocket of PDHK. To examine the selectivity of compound 11a on
18
ACS Paragon Plus Environment

Page 19 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PDHK isozymes, the inhibitory activity against at PDHK2, PDHK3 and PDHK4 was
determined as well. The IC50 against PDHK2, PDHK3 and PDHK4 was 1.126 µM,
5.336 µM and 1.743 µM respectively (Figure 3B), showing that compound 11a is
panꢀPDHKs inhibitor, with a less potency on other isozymes in contrast to PDHK1.
The cellular potency of 11a was evaluated in NCIꢀH1299 nonꢀsmall lung cancer cell
cline. Cells were treated with various concentrations of 11a for 12 h. The PDH
phosphorylation at both S293 and S232 sites were significantly decreased in a
concentrationꢀdependent manner, as well as the increase of HSP70 and decrease of
AKT (Figure 3C). The cellular activity of compound 11a on phosphorylation of PDH
was further evaluated by high content analysis (Figure 3D, 3E). Compound 11a
showed a similar potency as that of AZD7545 and was nearly 1000 fold more potent
than DCA. These results demonstrated that compound 11a is a novel ATP
competitive inhibitor of PDHK, with potent activity at both molecular and cellular
levels.
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
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 45
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Characterization of inhibition of compound 11a on PDHK; A) ATP
competitiveness of compound 11a on PDHK1 that determined by LineweaverꢀBurk
plot, showing compound 11a is an ATP competitive inhibitor on PDHK1; B)
Inhibitory activity of 11a on PDHK isozymes, as determined by ELISA assay; C)
Inhibitory activity of compound 11a on PDH phosphorylation in NCIꢀH1299 cells as
20
ACS Paragon Plus Environment

Page 21 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
determined by western blot; D) Effect of 11a on PDH phosphorylation (S293 and
S232), HSP70 and AKT in NCIꢀH1299 cells as determined by high content analyzer,
AZD7545 and DCA were positive controls, the magnification is 200×; E)
Quantitative analysis of Figure 3D, cells were defined as positive cells and negative
cells according the fluorescence intensity with above or below the averaged intensity,
the data was expressed as the percent of positive cells verses total cells; *P<0.05, as
compared to control group.
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
Inhibition of PDHK activity results in the activation of PDH, leading to a switch of
pyruvate consumption from lactate production to oxidative phosphorylation. As
shown in Figure 4A, lactate production was significantly decreased by compound 11a
at 10 µM in NCIꢀH1299 cancer cells after 12 h of treatment, showing a similar
potency as 10 mM of DCA. The metabolic modulatory effect of compound 11a was
evaluated by measuring oxygen consumption rate (OCR) and extracellular
acidification (ECAR). ECAR was decreased by compound 11a, which is consistent
with the lactate production change (Figure 2B). Unexpectedly, OCR was decreased as
well (Figure 2C), which may resulted from the simultaneous Hsp90 inhibition that
overrides the metabolic effect. As such, we expanded to other wellꢀknown HSP90
inhibitors, including 17ꢀDMAG, PUꢀH71, BIIB 021, SNXꢀ2112 and STAꢀ9090. All
these compounds led to decreased OCR (Supplementary Figure S1A). We further
compared the ratio of OCR/ECAR. OCR/ECAR was dramatically increased by
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 45
1
2
3
4
5
6
7
8
9
compound 11a (Figure 4D, Supplementary Figure S2) but not by HSP90 specific
inhibitors (Figure 4D, Supplementary Figure S1B). Thus, compound 11a is distinct
from HSP90 inhibitors by increasing the ratio of OCR/ECAR, which may result from
concurrent PDHK and HSP90 inhibition.
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 line with accelerated oxidative phosphorylation, compound 11a increased the
production of reactive oxygen species (ROS), a byproduct and indicator of oxidative
phosphorylation. (Figure 4C). We also tested the antiꢀproliferative effect of compound
11a in NCIꢀH1299 cells using CCK8 assay. Compound 11a doseꢀdependently
inhibited NCIꢀH1299 cell proliferation with an IC50 of 9.21 ± 1.33 µM (Figure 4D).
As compound 11a also has inhibitory activity on HSP90, the antiꢀproliferation effect
may result from its impacts on both PDHK and HSP90.
22
ACS Paragon Plus Environment

Page 23 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Effect of compound 11a on cell metabolism and proliferation; A)
NCIꢀH1299 cells was treated with various concentration of compound 11a for 12 h,
lactate production was significantly decreased by compound 11a at 10 µM, DCA was
used as a positive control; B) ECAR was significantly increased by compound 11a at
10 µM (12 h), DCA (10 mM) and SNXꢀ2112; C) OCR was significantly increased by
DCA (10 mM), but decreased by compound 11a at 10 µM (12 h) as well as
SNXꢀ2112; B) Ratio of ECAR/OCR was significantly increased by compound 11a at
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 45
1
2
3
4
5
6
7
8
9
10 µM (12 h) as well as by DCA (10 mM), but not SNXꢀ2112; E) Reactive oxygen
species production was significantly increased by compound 11a at 10 µM; F)
compound 11a inhibited NCIꢀH1299 cell proliferation with a IC50 of 9.21±1.33 (µM),
as measured by CCK8;*P<0.05, as compared to control group.
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 addition to NCIꢀH1299 cells, we further expanded to other cancer cell lines
derived from different tissues to verify the anticancer activity of compound 11a via
metabolic modulation. Two lung cancer cell lines A549 and H460, and colon cancer
cell lines HCT116 and breast cancer cell lines MDAꢀMBꢀ231 were selected. The
results showed that compound 11a effectively decreased PDH phosphorylation at both
S293 and S232 sites in a concentrationꢀdependent manner and accordingly the lactate
production in parallel (Figure 5) in all tested cell lines.
24
ACS Paragon Plus Environment

Page 25 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 45
1
2
3
4
5
6
7
8
9
Figure 5. Activity of compound 11a in cancer cell lines; Inhibitory activity of
compound 11a on PDH phosphorylation (12 h) and lactate production (12 h) in A549
cells (A), H460 cells (B), HCT116 cells (C) and MDAꢀMBꢀ231 cells (D); *P<0.05, as
compared to control group.
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
CONCLUSIONS
Starting with the HSP90 inhibitors, we developed several 3ꢀamide substituted
isoxazole derivatives as potent inhibitors of PDHK and HSP90. To improve cellular
potency of this series, we designed their vinylogues aiming at achieving balanced
PDHK in vitro potency and physicochemical properties aiding cell penetration.
Among those, compound 11a was found to be an excellent lead compound as well as
a favorable biological tool to further evaluate the therapeutic potential of dual PDHK
and HSP90 inhibitor in the treatment of cancer.
EXPERIMENTAL SECTION
Chemistry: General Methods. 1H NMR (400 MHz) spectra were recorded by using
a Varian Mercuryꢀ400 high performance digital FTꢀNMR spectrometer with
13
tetramethylsilane (TMS) as an internal standard. C NMR (100 or 125 MHz) spectra
were recorded by using a Varian Mercuryꢀ400 high performance digital FTꢀNMR
spectrometer or Varian Mercuryꢀ500 high performance digital FTꢀNMR spectrometer.
Abbreviations for peak patterns in NMR spectra are the following: br = broad, s =
26
ACS Paragon Plus Environment

Page 27 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
singlet, d = doublet, and m = multiplet. Lowꢀresolution mass spectra were obtained
with a Finnigan LCQ Deca XP mass spectrometer using a CAPCELL PAK C18 (50
mm × 2.0 mm, 5 ꢁm) or an Agilent ZORBAX Eclipse XDB C18 (50 mm × 2.1 m, 5
ꢁm) in positive electrospray mode. Highꢀresolution mass spectra were recorded by
using a Finnigan MATꢀ95 mass spectrometer or an Agilent Technologies 6224 TOF
mass spectrometer. Purity of all compounds was determined by analytical Gilson
highꢀperformance liquid chromatography (HPLC) using a YMC ODS3 column (50
mm × 4.6 mm, 5 ꢁm). Conditions were as follows: CH3CN/H2O eluent at 2.5 mL
min−1 flow [containing 0.1% trifluoroacetic acid (TFA)] at 35 °C, 8 min, gradient 5%
CH3CN to 95% CH3CN, monitoring by UV absorption at 214 and 254 nm. The
purities of all biologically evaluated compounds are > 95%. All solvents and reagents
were used directly as obtained commercially unless otherwise noted. All air and
moisture sensitive reactions were carried out under an atmosphere of dry argon with
heatꢀdried glassware and standard syringe techniques. Melting points were determined
using a SGW Xꢀ4 hot stage microscope and are uncorrected.
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
(E)-3-(5-(5-chloro-2,4-dihydroxyphenyl)-4-(4-methoxyphenyl)isoxazol-3-yl)-1-(
4-phenylpiperazin-1-yl)prop-2-en-1-one (10a). 9 (150mg, 0.26mmol) in DCM was
added EDCI (60 mg, 0.32 mmol) and HOBt (43 mg, 0.32 mmol), then triethylamine
(110 µL, 0.79 mmol) and 1ꢀphenylpiperazine (43 mg, 0.26 mmol) was added, the
mixture was stirred at RT overnight. The resulting mixture was washed with water,
saturate NaHCO3, brine, and dried with Na2SO4. After removal of the solvent, the
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 45
1
2
3
4
5
6
7
8
9
brown solid was dissolved in DCM, then 1M BCl3 in DCM (1.3 mL, 1.3 mmol) was
added, the mixture was stirred at RT for 6 h. Removed the solvent and the product
was purified via silica gel chromatography(DCM/MeOH) to give a yellow solid (77.3
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
mg, 56% yield, HPLC purity: 99); mp 244 – 248 oC; 1H NMR (400 MHz, Acetoneꢀd6
7.40 (d, = 15.6 Hz, 1H), 7.29 ꢀ 7.21 (m, 5H), 7.18, (d, = 15.6 Hz, 1H), 7.04 ꢀ
6.98 (m, 4H), 6.84 (t, = 7.3 Hz, 1H), 6.69 (s, 1H), 3.83 (s, 3H), 3.81 ꢀ 3.68 (m, 4H),
)
δ
J
J
J
3.25 ꢀ 3.15 (m, 4H); HRMS calcd for C29H25ClN3O5 ([M ꢀ H]ꢀ): 530.1488, found:
530.1477.
(E)-3-(5-(5-chloro-2,4-dihydroxyphenyl)-4-(4-methoxyphenyl)isoxazol-3-yl)-1-(
4-(2-methoxyphenyl)piperazin-1-yl)prop-2-en-1-one (10b). Compound 10b was
prepared according the described procedure of 10a, to give 10b as yellow solid (81.7
mg, 56% yield, HPLC purity: 98); mp 195 – 197 oC; 1H NMR (400 MHz, Acetoneꢀd6
7.40 (d, = 15.6 Hz, 1H), 7.32 ꢀ 7.23 (m, 3H), 7.17 (d, = 15.6 Hz, 1H), 7.03 ꢀ 6.88
)
δ
J
J
(m, 6H), 6.68 (s, 1H), 3.87 (s, 3H), 3.83 (s, 3H), 3.79 ꢀ 3.74 (m, 2H), 3.70 ꢀ 3.65 (m,
2H), 3.07 ꢀ 3.01 (m, 4H); HRMS calcd for C30H27ClN3O6 ([M ꢀ H]ꢀ): 560.1594, found:
560.1588.
(E)-3-(5-(5-chloro-2,4-dihydroxyphenyl)-4-(4-methoxyphenyl)isoxazol-3-yl)-1-(
4-(4-methoxyphenyl)piperazin-1-yl)prop-2-en-1-one (10c). Compound 10c was
prepared according the described procedure of 10a, to give 10c as yellow amorphous
solid (87.5 mg, 60% yield, HPLC purity: 96); 1H NMR (400 MHz, Acetoneꢀd6
)
δ
7.39
28
ACS Paragon Plus Environment

Page 29 of 45
Journal of Medicinal Chemistry
1
2
3
4
(d,
J = 15.6 Hz, 1H), 7.29 ꢀ 7.20 (m, 4H), 7.17 (d, J = 15.6 Hz, 1H), 7.01 ꢀ 6.95 (m,
5
6
7
8
9
3H), 6.88 ꢀ 6.83 (m, 2H), 6.80 (s, 1H), 3.82 (s, 3H), 3.78 ꢀ 3.73 (m, 5H), 3.69 ꢀ 3.66
(m, 2H), 3.09 ꢀ 3.02 (m, 4H); C30H27ClN3O6 ([M ꢀ H]ꢀ): 560.1594, found: 560.1583.
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
(E)-3-(5-(5-chloro-2,4-dihydroxyphenyl)-4-(4-methoxyphenyl)isoxazol-3-yl)-1-(
4-(3-methoxyphenyl)piperazin-1-yl)prop-2-en-1-one (10d). Compound 10d was
prepared according the described procedure of 10a, to give 10d as yellow solid (84.6
o
1
mg, 58% yield, HPLC purity: 100); mp 229 – 230 C; H NMR (400 MHz,
Acetoneꢀd6 7.39 (d, = 15.6 Hz, 1H), 7.29 ꢀ 7.23 (m, 3H), 7.20 ꢀ 7.12 (m, 2H), 7.04
ꢀ 6.96 (m, 2H), 6.67 (s, 1H), 6.59 (dd, = 8.3, 2.3 Hz, 1H), 6.54 (t, = 2.3 Hz, 1H),
6.43 (dd, = 8.2, 2.3 Hz, 1H), 3.83 (s, 3H), 3.78 ꢀ 3.67 (m, 7H), 3.23 ꢀ 3.17 (m, 4H);
HRMS calcd for C30H27ClN3O6 ([M ꢀ H]ꢀ): 560.1594, found: 560.1588.
)
δ
J
J
J
J
(E)-3-(5-(5-chloro-2,4-dihydroxyphenyl)-4-(4-methoxyphenyl)isoxazol-3-yl)-1-(
4-(pyridin-2-yl)piperazin-1-yl)prop-2-en-1-one (10e). Compound 10e was prepared
according the described procedure of 10a, to give 10e as yellow solid (77.5 mg, 56%
o
yield, HPLC purity: 100); mp 251 – 255 C; 1H NMR (400 MHz, Acetoneꢀd6
)
δ
8.15
(m, 1H), 7.55 (m, 1H), 7.40 (d,
J
= 15.6 Hz, 1H), 7.28 ꢀ 7.25 (m, 3H), 7.19 (d,
J =
15.6 Hz, 1H), 7.03 ꢀ 6.99 (m, 2H), 6.84 (d,
J
= 8.6 Hz, 1H), 6.68 ꢀ 6.65 (m, 2H), 3.84
(s, 3H), 3.78 ꢀ 3.55 (m, 8H); HRMS calcd for C28H26ClN4O5 ([M+H]+): 533.1586,
found: 533.1592.
29
ACS Paragon Plus Environment