Selective Inhibitors of Human Liver Carboxylesterase Based on a β-Lapachone Scaffold: Novel Reagents for Reaction Profiling

Article abstract of DOI:10.1021/acs.jmedchem.6b01849

Carboxylesterases (CEs) are ubiquitous enzymes that are responsible for the metabolism of xenobiotics, including drugs such as irinotecan and oseltamivir. Inhibition of CEs significantly modulates the efficacy of such agents. We report here that β-lapachone is a potent, reversible CE inhibitor with K_i values in the nanomolar range. A series of amino and phenoxy analogues have been synthesized, and although the former are very poor inhibitors, the latter compounds are highly effective in modulating CE activity. Our data demonstrate that tautomerism of the amino derivatives to the imino forms likely accounts for their loss in biological activity. A series of N-methylated amino derivatives, which are unable to undergo such tautomerism, were equal in potency to the phenoxy analogues and demonstrated selectivity for the liver enzyme hCE1. These specific inhibitors, which are active in cell culture models, will be exceptionally useful reagents for reaction profiling of esterified drugs in complex biological samples.

Full text of DOI:10.1021/acs.jmedchem.6b01849

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Selective inhibitors of human liver carboxylesterase based upon
a #-lapachone scaffold: Novel reagents for reaction profiling
M. Jason Hatfield, Jingwen Chen, Ellie M Fratt, Liying Chi, John C. Bollinger, Randall J Binder,
John Bowling, Janice L. Hyatt, Jerrod Scarborough, Cynthia Jeffries, and Philip M Potter
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01849 • Publication Date (Web): 23 Jan 2017
Downloaded from http://pubs.acs.org on January 24, 2017
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 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Selective inhibitors of human liver carboxylesterase based upon a βꢀ
lapachone scaffold: Novel reagents for reaction profiling
M. Jason Hatfield†,1, Jingwen Chen†,1, Ellie M. Fratt1, Liying Chi1, John C.
Bollinger2, Randall J. Binder1, John Bowling1, Janice L. Hyatt1, Jerrod
Scarborough1, Cynthia Jeffries1, Philip M. Potter*,1
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 ꢀ Department of Chemical Biology and Therapeutics,
2 ꢀ Department of Structural Biology,
St. Jude Children's Research Hospital, Memphis, TN 38105, United States
† – These authors contributed equally to this work.
* Corresponding author.
Philip M. Potter
Department of Chemical Biology and Therapeutics,
St. Jude Children's Research Hospital,
262 Danny Thomas Place,
Memphis,
TN 38105, United States
Tel: 901ꢀ595ꢀ6045
Fax: 901ꢀ595ꢀ4293
Eꢀmail: phil.potter@stjude.org.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
ABSTRACT
4
5
6
7
8
Carboxylesterases (CE) are ubiquitous enzymes that are responsible for the
metabolism of xenobiotics, including drugs such as irinotecan and oseltamivir.
Inhibition of CEs significantly modulates the efficacy of such agents. We report
here that βꢀlapachone is a potent, reversible CE inhibitor with Ki values in the
nanomolar range. A series of amino and phenoxy analogues have been
synthesized and while the former are very poor inhibitors, the latter compounds
are highly effective in modulating CE activity. Our data demonstrate that
tautomerism of the amino derivatives to the imino forms likely accounts for their
loss in biological activity. A series of Nꢀmethylated amino derivatives, which are
unable to undergo such tautomerism, were equal in potency to the phenoxy
analogues, and demonstrated selectivity for the liver enzyme hCE1. These
specific inhibitors, which are active in cell culture models, will be exceptionally
useful reagents for reaction profiling of esterified drugs in complex biological
samples.
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
2
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
INTRODUCTION
4
5
6
7
8
Carboxylesterases (CE) are members of the esterase class of enzymes that
hydrolyze esters into their corresponding carboxylic acids and alcohols.1 These
enzymes are present in essentially all organisms and are thought to play a
protective role against xenobiotics that may contain these chemotypes.2 As such,
numerous biologically important small molecules contain ester groups including
xenobiotics (e.g., cocaine and heroin3) as well as many clinically used agents (for
example, irinotecan (CPTꢀ11; 7ꢀethylꢀ10ꢀ[4ꢀ(1ꢀpiperidino)ꢀ1ꢀ
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
piperidino]carbonyloxycamptothecin4ꢀ6), capecitabine7, oseltamivir8, lidocaine9
and meperidine10). The prevalence of the ester function in these latter molecules
is partly due to the fact that this chemotype improves water solubility and
bioavailability of these compounds. Hence for the development of new drugs
from chemical screens that are poorly water soluble, it is highly likely that
medicinal chemists would develop ester, carbamate or amide derivatives that
would result in compounds that may be substrates for CEs.
The impact of CEs on drug metabolism has been demonstrated both with regard
to toxicity and antitumor activity. For example, plasmaꢀesterase deficient mice
are more resistant to the toxic effects of irinotecan due to reduced conversion to
the toxic metabolite 7ꢀethylꢀ10ꢀhydroxycamptothecin (SNꢀ38).11 Conversely,
tumor cells that express CEs are more sensitive to this agent since enhanced
drug hydrolysis occurs, resulting in higher intracellular concentrations of 7ꢀethylꢀ
10ꢀhydroxycamptothecin and as a consequence, more cytotoxicity.12 Clearly
therefore, the levels of CE are important with respect to these sequelae. We
have determined that molecules containing the ethaneꢀ1,2ꢀdione moiety can be
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
5
6
7
8
9
potent inhibitors of human CEs and these compounds can significantly modulate
the hydrolysis of esterified agents.13 More recently, we have searched natural
product databases with a defined pharmacophore (containing the structural and
electronic components of the ethaneꢀ1,2ꢀdione chemotype) and identified the
tanshinones as highly effective CE inhibitors.14 These compounds modulate
enzyme activity in vitro and the cytotoxic effects of irinotecan in cell culture
models. The tanshinones are present in high quantities in Danshen root, a herbal
medicine widely used in China, and this is currently being used in clinical trials in
the US as Compound Danshen Dripping Pill. Potentially therefore, the
metabolism of esterified agents may be impacted in individuals using this
material.
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
The tanshinones are all pan CE inhibitors, modulating activity of both the human
liver isoform (hCE1; CES1) and the intestinal enzyme (hiCE; CES2). Hence while
these are important reagents for biochemical studies, they do not allow
discrimination of CEꢀmediated ester hydrolysis in complex samples.
Based upon our observations, we have furthered our initial database searches
and evaluated the presence of the ethaneꢀ1,2ꢀdione moiety in other natural
products. This identified βꢀlapachone (ARQ501; 1)15 as a candidate molecule
and results described here demonstrate that this compound, and selected
derivatives thereof, are potent human CE inhibitors, both in vitro, and in cell
culture models. Based upon this information, we have developed isozyme
selective compounds that specifically inhibit hCE1.16
4
ACS Paragon Plus Environment

Page 5 of 47
Journal of Medicinal Chemistry
1
2
3
RESULTS AND DISCUSSION
4
5
6
Identification of 1 as a carboxylesterase inhibitor
7
8
9
We have previously reported that the ethaneꢀ1,2ꢀdione moiety, when present
within a hydrophobic scaffold, can act as a potent inhibitor of CEs.13, 17ꢀ20 This
has been demonstrated for a series of tanshinones14 and we have confirmed that
Chinese herbal medicines that contain these molecules can modulate esterified
drug metabolism. Since such medicines are widely used, we sought to examine
whether other natural products might contain structurally similar compounds and
if so, whether they harbored biological activity towards human CEs. Therefore
using a esterase inhibitor pharmacophore that we have previously defined14, we
identified 1 (Figure 1) as a potential candidate molecule.
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
Since this compound is readily available, we determined the ability of 1 to inhibit
human CEs. As indicated in Table 1, when using oꢀnitrophenyl acetate (oꢀNPA)
as a substrate, this molecule was a reasonable inhibitor of hiCE (Ki = 109 nM),
although considerably less active (~10ꢀfold) towards hCE1. Indeed, the potency
of 1 was similar to that seen for other 1,2ꢀdiones (cf. 14 nM and 45 nM for benzil
with hiCE and hCE1, respectively13; 118 nM and 398 nM for dihydrotanshinone
with hiCE and hCE1, respectively14). However as these concentrations are lower
than that required for 1 to induce cytotoxicity in cultured cells in vitro21ꢀ23, this
suggests that modulation of CE activity likely occurs under these circumstances.
We advise caution therefore, when using this compound in conjunction with
esterified agents that might be activated (e.g., irinotecan) or inactivated by these
enzymes. Based upon this information, we sought to characterize the ability of
analogues of 1 to inhibit human CEs.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 47
1
2
3
Synthesis of 4ꢀsubstituted 4ꢀphenoxynaphthaleneꢀ, 4ꢀ
4
5
6
7
8
(phenylamino)naphthaleneꢀ, and 4ꢀphenyl(methyl)amino naphthaleneꢀ1,2ꢀ
diones
9
Using a simple one step synthetic approach24, we generated a library of phenoxy,
phenylamino and phenylꢀNꢀmethylamino analogues in reasonable yields, which
differed only in the substituent at the 4ꢀposition of the free phenyl ring (H, MeO,
Me, F, Cl, Br, I). In all cases, synthesis readily occurred to yield either yellow
(phenoxy) or dark red (phenyl amino and phenyl(methyl)amino) solids. Following
physical characterization and confirmation of identity (see Supporting
Information), the ability of these compounds to inhibit human CEs was assessed.
Inhibition of human CEs by 4ꢀsubstituted 4ꢀphenoxynaphthaleneꢀ1,2ꢀdiones
All phenoxynaphthaleneꢀ1,2ꢀdiones were potent inhibitors of both hCE1 and hiCE
when using oꢀNPA as a substrate, although the Ki values were 4ꢀ to 11ꢀfold lower
for the former enzyme (Table 2). This contrasts the activity of 1 which is ~10ꢀfold
more active towards hiCE. A trend was observed such that the more hydrophobic
compounds (i.e., those with greater logP values) were more potent enzyme
inhibitors (data not shown). It should be noted, that in general, all of the 4ꢀ
phenoxynaphthaleneꢀ1,2ꢀdiones were more active than 1, and that none of these
compounds demonstrated any appreciable activity towards human
acetylcholinesterase (AChE; Table 2). All compounds acted in a partially
competitive fashion with respect to the mode of enzyme inhibition, i.e., while they
act in a competitive manner, they are unable to completely inhibit product
formation at infinite inhibitor concentrations.25 This is similar to results obtained
using the benzils and the tanshinones.13, 14
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
6
ACS Paragon Plus Environment

Page 7 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
Inhibition of human CEs by 4ꢀsubstituted 4ꢀ(phenylamino)naphthaleneꢀ1,2ꢀ
diones
6
7
8
9
In contrast to that seen for 4ꢀphenoxynaphthaleneꢀ1,2ꢀdiones, essentially all of
the phenylamino derivatives (compounds 9ꢀ15) were inactive towards both
human enzymes (Table 2). Indeed, the maximal inhibition observed at an
inhibitor concentration of 10 ꢁM was 41% for 4ꢀ[(4ꢀ
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
iodophenyl)amino]naphthaleneꢀ1,2ꢀdione (15) with hCE1. To determine if there
were any structural or molecular interactions that might account for the lack of
enzyme inhibition by the amino derivatives, we docked 3 of each class of
compounds (molecules 2, 3 and 6, and 9, 10 and 13) into the active site of the
crystal structure of hCE1. This was achieved using Molsoft ICM software using
the default parameters for small ligand docking. As indicated in Figure 2, all of
the small molecules were juxtaposed adjacent to the active site serine in an
identical conformation to that seen for compound 1. In addition, virtually no
differences were observed in the distances from the Oγ atom of the amino acid to
the carbonyl carbon atoms of the docked compounds (distances ranged from
3.03Å to 3.26Å). In general, the ICM scores for docking of the small molecules
were comparable (see Figure 2 legend), although all of the values obtained for
the phenoxy derivatives (compounds 2, 3 and 6) were lower than that seen for
the respective aniline. Since the mechanism of enzyme inhibition is believed to
be due to a coordinated attack by the serine Oγ atom on one of the carbonyl
carbon atoms within the the 1,2ꢀdione moiety, it is unlikely that the loss of activity
of the phenylamino derivatives is due to poor access or localization within the
enzyme active site.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
1
2
3
4
5
6
7
8
9
Since previous reports have indicated that phenylaminonaphthaleneꢀ1,2ꢀdiones
can undergo tautomerism resulting in loss of the 1,2ꢀdione moiety (see Figure
3a)26ꢀ28, we assessed whether the imino forms would dock in a similar fashion as
the phenoxy analogues. As can be seen (Figure 3c), the carbon atom attached to
the hydroxyl group was proximal to the serine Oγ atom in compound 9, with the
carbonyl carbon up to 4.45Å from this residue. Similar results were obtained from
docking the imino tautomers of the other phenylaminonaphthaleneꢀ1,2ꢀdiones
(data not shown). Previous studies using a panel of benzoins and corresponding
benzils13, 17 have yielded similar results, but it has not been possible to undertake
simple chemical modifications of these molecules to specifically address the role
of the hydroxyl/carbonyl carbon atom configuration with respect to CE inhibition.
We postulated therefore that it is unlikely that the imino compounds would act as
inhibitors since the carbonyl carbon atom is not adjacent to the serine Oγ atom.
Inhibition of human CEs by 4ꢀsubstituted phenyl
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
(methyl)aminonaphthaleneꢀ1,2ꢀdiones
To limit the aminoꢀimino tautomerism of the 4ꢀ(phenylamino)naphthaleneꢀ1,2ꢀ
diones, we synthesized a similar panel of Nꢀmethylated aniline analogues
(compounds 12ꢀ22) using the same methods, and evaluated whether they would
act as CE inhibitors. We hypothesized that methylating the nitrogen atom would
prevent formation of the imino moiety due to the lack of the labile amine
hydrogen atom. As indicated in Table 2, all of the Nꢀmethyl substituted
compounds, with the exception of the iodo analogue (22), were potent inhibitors
of hCE1 with Ki values comparable to that seen for the phenoxy analogues
(Table 1). Indeed, the most potent compound, (the bromo derivative; 21)
8
ACS Paragon Plus Environment

Page 9 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
demonstrated an inhibition constant of ~16 nM. These data further support the
hypothesis that the loss of the 1,2ꢀdione in the phenylaminonaphthaleneꢀ1,2ꢀ
diones, via tautomerism to the imino form (Figures 3a and 3b), significantly
compromises enzyme inhibitory activity of the former.
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
When we docked compound 16 into the active site of hCE1, the docking pose
demonstrated a significant difference in the distance of the serine Oγ atom to the
closest carbonyl atom, as compared to the imino form of 9 (Figures 3c and d).
For the latter compound, the distance is 4.45Å, whereas for the Nꢀmethylated
analogue, this value is 3.08Å. It should also be noted that with the imino form, the
OH group is juxtaposed closest to the catalytic serine Oγ atom. Since esterases
do not attack alcohol bonded carbon atoms, these docking studies provide
further support for the lack of activity of the phenyl aniline derivatives.
Interestingly, the Nꢀmethylated aniline analogues with smaller, less hydrophobic
substituents demonstrated selectivity for hCE1, with little or no activity towards
hiCE at concentration up to 10ꢁM (Table 2). Compound 21 was weakly active
towards hiCE (53% inhibition at 10 ꢁM), whereas 22 was moderately potent
towards both human CEs. The mechanism to account for this selectivity is not
understood since we do not believe that any electronic effects could be
responsible for the loss of activity. Potentially, the architecture of the hiCE active
site may be sufficiently different from hCE1 such that steric hinderance may
influence CE inhibitor binding. However, since the crystal structure of hiCE has
not been determined, and this enzyme is very labile29, an analysis of the
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
5
interaction of the phenyl (methyl)aminonaphthaleneꢀ1,2ꢀdiones with this enzyme
may require other physical approaches (e.g., NMR, SPR, etc).
6
7
8
9
Compound 22, which contains an iodine atom at the 4ꢀposition of the free phenyl
ring, was considerably less active that the other analogues (Ki with hCE1 =
400nM; Table 2). However, docking of this molecule into the active site of this
protein demonstrated no difference in the alignment as compared to the other Nꢀ
methyl analogues that were potent inhibitors (data not shown). Indeed, the
distance between the serine Oγ and the carbonyl carbon atoms was measured at
3.01 Å and 3.28Å for 22. However, we have previously demonstrated that for
hCE1, the amino acids forming the entrance to the active site gorge are relatively
immobile, limiting substrate access to the catalytic residues.29 We hypothesize
therefore, that 22 is unable to freely enter the enzyme active site due to the
increased size of the iodo atom, relative to the substituents in the other
phenyl(methyl)amino derivatives. As a consequence, the Ki value for hCE1
inhibition by the iodo derivative is larger (i.e. the molecule is less potent). Since
the active site in hiCE is more plastic and can readily accept larger substrates
than hCE1,3, 29 22 demonstrates modest activity towards the former enzyme,
likely due to its ability to access the catalytic serine residue. In toto, these results
argue that the biological activity of 22 is modulated by the relative size of the
molecule and its ability to enter the active site and interact with the catalytic
amino acids within the human CEs.
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
10
ACS Paragon Plus Environment

Page 11 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Assessment of imine formation in phenylaminonaphthaleneꢀ1,2ꢀdiones
To confirm that imine formation was the likely reason for the loss of activity of the
phenylaminonaphthaleneꢀ1,2ꢀdiones, we undertook three different approaches to
assess the presence of this chemotype in these molecules. Firstly, the crystal
structures of compounds 2, 9 and 16, as well as 1,2ꢀnaphthoquinone (23) were
determined, and the lengths of relevant bonds were assessed. We also
compared these results with those obtained from structurally similar molecules
(4ꢀaminoꢀ1,2ꢀnaphthquinone (24) and 4ꢀ(9Hꢀcarbazolꢀ9ꢀyl)ꢀ1,2ꢀnaphthoquinone
(25)) that had previously been crystalized. We hypothesized that in compounds
that would could undergo the aminoꢀimino tautomerization, the CꢀN bond
between the 1,2ꢀdioneꢀcontaining ring and the heteroatom would be shorter in
length due to the formation of the double bond.
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
As can be seen in Figure 4, the CꢀN bond in 9 is shorter than that seen in the Nꢀ
methyl derivative 16 (1.352 Å versus 1.374 Å), and considerably smaller than
that observed in the carbazoyl analogue 25 (1.411 Å). In addition, this bond
length in 9, is longer than that observed in the unsubstituted 4ꢀaminoꢀ1,2ꢀ
naphthoquinone (24). Since compounds 24 and 25 represent molecules which
would be the most likely and unlikely to undergo imine formation, respectively,
these results suggest that the CꢀN bond in 9 likely has partial double bond
character.
Further support for this tautomerization was provided by examining the CꢀC and
the C=O bonds in the respective analogues (bonds 2, 3 and 4 in Figure 4). In 2,
16, 23, and 25, the CꢀC bond (bond 2) is shorter than that observed for
molecules 9 and 24, consistent with the hypothesis that this is likely to exist as a
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
1
2
3
4
5
6
7
8
9
single bond in the latter two molecules. In contrast, bond 3 is shorter in 9 and 24
as compared to the other compounds, suggesting that it exhibits more double
bond character in these molecules. Furthermore, the C=O bond (bond 4 in Figure
4) is shorter in the compounds that would be unlikely to undergo imine formation
(i.e., 2, 16, 23, and 25), but considerably longer in 9 and 24. Overall, these
structural studies argue that 9 and 24 maintain more imineꢀlike character than the
other molecules analyzed.
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
Since the crystallographic studies obviously evaluated bond distances in the solid
state which may not be entirely reflective of what occurs in solution, we used
NMR to determine the structures of naphthaleneꢀ1,2ꢀdione analogues under
different solvent conditions. While the1H and 13C spectra for compounds 2 and 16,
when dissolved in DMSO, were readily interpretable, the data obtained for 9
contained multiple overlapping peaks and were difficult to assign (see Supporting
Information). Since we presumed that this was due to the presence of the two
different tautomers in the sample, we opted to simplify the analyses and to
evaluate the 19F NMR signals using the fluorinated derivatives (compounds 5, 12
and 19 for the phenol, the aniline and the Nꢀmethylaniline analogues,
respectively). As indicated in Figure 5 (panels a and e), single peaks were
observed for 5 (115.77ppm) and 19 (116.32ppm), corresponding to the fluorine
atom in the para position of the appended benzene ring. However, the aniline
(12) yielded two signals at 114.97ppm and 119.80ppm (Figure 5c), likely due to
the presence of the amino and imino tautomers within the sample. Consistent
with this hypothesis, the inclusion of base (triethylamine) into the latter sample,
yielded a spectrum containing a broad single peak (116.93ppm), likely resulting
12
ACS Paragon Plus Environment

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
from an increase in the rate of interconversion of the imine to the amine (see
Figure 5d). Base would facilitate this process by abstraction of the labile proton in
this structure. Since the shift of the signal peak was very similar to the singlet
seen for 19 in DMSO, we believe that under basic conditions 12 exists primarily
in its amino form.
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
To confirm the results seen in organic solvent, we used UV spectroscopy to
assess the changes in absorption of the molecules in aqueous solutions (20%
DMSO:80% 200 mM sodium phosphate buffer) at different pH values. Little, if
any, change was observed in the spectra for compounds 2 and 16 over the pH3ꢀ
11 range (Figures 5g and 5j), consistent with the hypothesis that these molecules
do not change structure in these acidic/basic environments. In contrast,
compound 9 ((4ꢀphenylamino)naphthaleneꢀ1,2ꢀdione), demonstrated significant
changes in the UV spectrum over the pH range used (Figure 5h). Under neutral
or acidic conditions, the spectra were essentially identical with a peak maxima at
475 nm. However at pH9 and pH11 this maxima was shifted to lower
wavelengths of 461 and 453 nm, respectively. This hypsochromic shift is
consistent with a decrease in conjugation of 9 due the loss of the double bond
between the nitrogen atom and the aliphatic ring (bond 1 in Figure 4), i.e., by
conversion from the imine to the amine. Since no changes in the spectra were
observed for compounds 2 and 16 over these pH ranges, we conclude that this
tautomerization does not occur under these conditions for these molecules.
Overall, these crystallographic, NMR and UV spectral analyses indicate that
compound 9 likely exists in tautomeric forms that include the imine and amine.
This is not observed with either the phenoxy (2), fluorophenoxy (5), Nꢀ
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
1
2
3
4
5
6
7
8
9
methylamino (16) or 4ꢀfluorophenyl Nꢀmethylamino (19) derivatives. Since the
imino tautomers would be very unlikely to act as CE inhibitors (due to the loss of
the 1,2ꢀdione moiety), our data is strongly supportive of the hypothesis that the
lack of biological activity of the amines is due to the presence of this
tautomerization.
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 oseltamivir and irinotecan hydrolysis by βꢀlapachone
analogues
Having demonstrated inhibition of CEs in vitro by the naphthaleneꢀ1,2ꢀdiones, we
assessed whether these molecules were membrane permeable and could
modulate drug metabolism in cultured cells. Two different drugs were used as
substrates since they are selectively hydrolyzed by different human CEs: hCE1
for oseltamivir; and hiCE for irinotecan. As indicated in Figure 6a, 1 and the
phenoxy analogues (2ꢀ7) were reasonable inhibitors of hCE1, yielding anywhere
from a 12ꢀ47% reduction of metabolite formation. However, the
phenyl(methyl)amino derivatives (16ꢀ21) were considerably more potent (Figure
6b) resulting in significant loss of oseltamivir carboxylate production (up to 99%
with compound 19), confirming that these molecules can inhibit hCE1
intracellularly.
Interestingly, while 1 was an excellent inhibitor of hiCE in cells (as indicated by
the greater than 90% loss in 7ꢀethylꢀ10ꢀhydroxycamptothecin production; Figure
6c), the 4ꢀphenoxynaphthaleneꢀ1,2ꢀdiones (2ꢀ7) were essentially inactive in this
assay. This contrasts with the in vitro results, although it should be noted that
these compounds are 5ꢀ to 10ꢀfold less active towards hiCE as compared to
hCE1 (Table 1). Consistent with the in vitro assays, compounds 16ꢀ21 yielded
14
ACS Paragon Plus Environment

Page 15 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
very little inhibition of hiCE (Figure 6d). Overall, these results confirm the in vitro
selectivity of the Nꢀmethyl amino derivatives towards hCE1, and demonstrate the
cell permeability of these compounds.
9
To date, the identification of the specific CEs responsible for drug activation in
humans has been problematic, in part due to lack of specific reagents to assess
such reactions. For example, there are no specific hCE1 inhibitors and/or simple
chromogenic substrates that can be used for rapid biochemical analyses. We
have generated benzene sulfonamides as selective hiCE modulators30 and have
tried to use these in combination with pan CE inhibitors (e.g., benzils13) to assess
esterified substrate hydrolysis in complex samples.31 While partially successful,
due to the numerous esterases present within mammalian tissues and
microsomes, definitive results require the use of timeꢀconsuming and laborꢀ
intensive chromatographic purification. However, based upon the data presented
here, we propose that either compound 17 or 18 would be the best choice for use
in reaction profiling of complex biological samples with respect to CEꢀmediated
drug hydrolysis. These molecules exhibit potency towards hCE1 (Ki values of
~80 and 11 nM, respectively), yet at 10 ꢁM, only demonstrate 7% and 25%
inhibition of hiCE (Table 2). In cell based studies, both compounds were potent
and selective, yielding greater than 94% inhibition of hCE1 and less than 23%
inhibition of hiCE (Figure 6). Used in combination with benzene sulfonamides32,
benzils13, and total esterase inhibitors (e.g. bisꢀ(4ꢀnitrophenyl) phosphate), these
compounds will allow studies to document hCE1ꢀspecific substrate metabolism in
complex enzyme mixtures.
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
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
4
5
6
7
8
9
Since oꢀquinones are known to be cytotoxic, we determined U373MG cell
viabilities after 30 min exposure (the time period used for the cellular enzyme
inhibition studies), and growth inhibition after 72 hr dosing, for selected Nꢀ
methylaniline analogues. Short term assays indicated no obvious effects, with
cell viabilities of 99 ± 6%, 100 ± 14% and 96 ± 5% being recorded for 16, 18 and
19, respectively. In longer term assays (72 hr), these molecules yielded IC50
values of 12.8 ± 5.8 ꢁM, 4.6 ± 1.1 ꢁM, and 3.9 ± 1.0 ꢁM, respectively (see
Supporting Information). The latter data suggest that these compounds do exert
cytotoxic effects, but these only occur over an extended time frame. Since
enzyme inhibition studies typically involve incubations of less than 1 hr, the
effects of these inhibitors with regard to cell toxicity would be minimal and can
essentially be discounted.
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
In summary, we have determined that 1 is a potent inhibitor of human CEs and
this occurs at concentrations lower than that required to induce cytotoxicity.21ꢀ23
Typically 1ꢀ5 ꢁM is required for the latter, whereas the Ki values for 1 with
purified hCE1 or hiCE are 1.2 and 0.1 ꢁM respectively. Since 11 different clinical
trials with 1 are noted in Clinicaltrials.gov, studies must be designed cautiously to
ensure that this compound is not combined with agents that might be activated
(e.g., irinotecan) or inactivated by these enzymes. Additionally, our studies have
identified potent, selective hCE1 inhibitors based upon the βꢀlapachone scaffold.
This is the first report of such compounds and these molecules will now allow for
a detailed evaluation of drug metabolism by samples containing multiple human
16
ACS Paragon Plus Environment

Page 17 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CEs. Hence, reaction profiling of complex enzyme mixtures (e.g., tissue samples,
microsomes, cell extracts, etc.) using a combination of specific CE inhibitors will
allow for rapid identification of target enzymes. Such studies are currently
underway.
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
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
EXPERIMENTAL SECTION
4
5
6
Materials and general procedures
7
8
9
1 was purchased from Sigma Aldrich (St. Louis, MO) and was used without
purification. Pure hCE1 and hiCE were generated by expressing the respective
cDNAs in Spodoptera frugiperda Sf9 cells using baculoviral vectors. Secreted,
active enzyme was purified from serumꢀfree culture media using preparative
isoelectric focusing and chromatography.3, 33 Human AChE was purchased from
Sigma Biochemicals (St. Louis, MO). U373MG cells expressing hCE1 or hiCE
have been previously described.34
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
Reagents for chemical synthesis were purchased from Sigma Aldrich (St. Louis,
MO) or Oakwood Products (Estill, SC) and were used without further purification.
Reactions were typically carried out at a 10mmole scale and analyzed using a
Waters Acquity UPLC with an SQ Detector 2 (UV, ELSD, MS). Purification of
products was achieved using a Biotage ACI flash chromatography system with
Biotage SNAP Ultra HPꢀSphere 25 ꢁM 10g columns or a Waters Prep LC 4000
with Waters symmetry C18 column 19 X 300 mm. Overall yields were in the 20ꢀ
40% range. Typically, crude material was dissolved in a minimal amount of
DMSO and applied to the column. Compounds were then eluted using a 10ꢀ
100% hexane/ethyl acetate gradient over 20 min at a flow rate of 35 ml/min.
Solvent from purified fractions was then removed under reduced pressure and
the identity of isolated molecules confirmed using 1H and 13C NMR, and HRMS.
All compounds were deemed to be greater than 95% pure using analytical
UPLCꢀMS analysis and NMR.
18
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NMR spectra were obtained in either DMSOꢀd6 or CDCl3 using a 500 MHz Bruker
Avance spectrometer equipped with XH dual probe. Chemical shifts are reported
based upon TMS or fluorobenzene as a standard. HRMS were determined with a
Waters Acquity UPLC – Xevo G2 QTOF spectrometer.
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
The atom coordinates for compounds 24 and 25, AMNPQH10 and XANRUB,
respectively, were obtained from the CSD database and analyzed using Mercury
software.
Chemical synthesis
General methods for 4ꢀsubstituted 4ꢀphenoxynaphthaleneꢀ1,2ꢀdiones (2ꢀ8)
4ꢀSubstituted 4ꢀphenoxynaphthaleneꢀ1,2ꢀdiones were synthesized by
condensation of the corresponding phenol (5.5 mmol) with sodium 3,4ꢀdioxoꢀ3,4ꢀ
dihydronaphthaleneꢀ1ꢀsulfonate (5 mmol).24 Reactions were conducted in 150ml
of water containing KOH (5 mmol). After solvent removal, the dried residue was
subjected to flash chromatography, preparative HPLC, and in some cases,
recrystallization from ethanol, to afford yellow solids. The synthesis of
compounds 2, 3 and 6 have been previously reported and their chemical
parameters are included in the Supporting Information.24, 35 The physical and
chemical properties of novel molecules are listed here.
4ꢀ(4ꢀMethylphenoxy)naphthaleneꢀ1,2ꢀdione (4)
Compound 4 was synthesized according to the general method using 4ꢀ
methylphenol as the starting material. This yielded a yellow solid (53 mg, 28%).
Mp 165 ºC; UVꢀVis λmax: 250 nm, 280 nm, 326 nm, and 403 nm in methanol. 1H
NMR (400 MHz, CDCl3) δ 8.16 (dd, J = 1.4, 7.7 Hz, 1H), 8.09 (dd, J = 1.2, 7.8 Hz,
1H), 7.77 (td, J = 1.5, 7.7 Hz, 1H), 7.65 (td, J = 1.3, 7.6 Hz, 1H), 7.31 – 7.23 (m,
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 47
1
2
3
4
5
6
7
8
9
2H), 7.09 – 6.99 (m, 2H), 5.67 (s, 1H), 2.40 (s, 3H). 13C NMR (101 MHz, CDCl3)
δ 179.55, 179.38, 169.02, 150.12, 136.78, 135.10, 131.85, 131.72, 130.81 (2C),
130.50, 129.34, 124.92, 120.93 (2C), 106.32, 20.91. HRMS (ESI): m/z 265.0873
(M + H) (calcd for C17H12O3, 265.0864) (M + H).
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
4ꢀ(4ꢀFluorophenoxy)naphthaleneꢀ1,2ꢀdione (5)
Compound 5 was synthesized according to the general method using 4ꢀ
fluorophenol as the starting material. This yielded a yellow solid (80 mg, 37%).
Mp 190 ºC; UVꢀVis λmax: 250 nm, 320 nm, and 398 nm in methanol. 1H NMR (500
MHz, CDCl3) δ 8.18 (dd, J = 1.3, 7.7 Hz, 1H), 8.08 (dd, J = 1.1, 7.9 Hz, 1H), 7.78
(td, J = 1.4, 7.7 Hz, 1H), 7.67 (td, J = 1.2, 7.6 Hz, 1H), 7.23 – 7.12 (m, 4H), 5.64
(s, 1H). 13C NMR (126 MHz, CDCl3) δ 179.46, 179.11, 168.78, 161.77, 159.81,
148.11, 135.17, 132.02, 131.42, 130.46, 129.51, 124.87, 122.87, 117.28, 117.10,
106.33. 19F NMR (471 MHz, DMSO) δ ꢀ115.77. HRMS (ESI): m/z 269.0614 (M +
H) (calcd for C16H9FO3, 269.0611) (M + H).
4ꢀ(4ꢀBromophenoxy)naphthaleneꢀ1,2ꢀdione (7)
Compound 7 was synthesized according to the general method using 4ꢀ
bromophenol as the starting material. This yielded a yellow solid (63 mg, 24%).
Mp 210 ºC; UVꢀVis λmax: 250 nm, 278 nm, 328 nm, and 397 nm in methanol. 1H
NMR (500 MHz, CDCl3) δ 8.21 – 8.15 (m, 1H), 8.07 (dd, J = 1.2, 7.8 Hz, 1H),
7.78 (td, J = 1.4, 7.6 Hz, 1H), 7.67 (td, J = 1.2, 7.6 Hz, 1H), 7.65 – 7.58 (m, 2H),
7.11 – 7.04 (m, 2H), 5.65 (s, 1H). 13C NMR (126 MHz, CDCl3) δ 179.40, 179.02,
168.36, 151.34, 135.18, 133.54 (2C), 132.07, 131.34, 130.46, 129.55, 124.87,
123.14 (2C), 120.21, 106.44. HRMS (ESI): m/z 328.9810 (M + H) (calcd for
C16H9BrO3, 328.9813) (M + H).
20
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4ꢀ(4ꢀIodophenoxy)naphthaleneꢀ1,2ꢀdione (8)
4
5
6
7
8
Compound 8 was synthesized according to the general method using 4ꢀ
Iodophenol as the starting material. This yielded an orange solid (132 mg, 44%).
Mp 220 ºC; UVꢀVis λmax: 247 nm, 274 nm, 329 nm, and 397 nm in methanol. 1H
NMR (500 MHz, CDCl3) δ 8.21 (dd, J = 1.3, 7.6 Hz, 1H), 8.08 (dd, J = 1.1, 7.8 Hz,
1H), 7.87 – 7.76 (m, 3H), 7.69 (td, J = 1.2, 7.6 Hz, 1H), 7.00 – 6.93 (m, 2H), 5.68
(s, 1H). 13C NMR (126 MHz, CDCl3) δ 179.40, 179.04, 168.34, 152.19, 139.55
(2C), 135.18, 132.07, 131.36, 130.47, 129.56, 124.88, 123.48 (2C), 106.47,
91.07. HRMS (ESI): m/z 376.9671 (M + H) (calcd for C16H9IO3, 376.9675) (M +
H).
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
General methods for 4ꢀsubstituted (4ꢀphenylamine)naphthaleneꢀ1,2ꢀdiones (9ꢀ
15)
4ꢀSubstituted (4ꢀphenylamine)naphthaleneꢀ1,2ꢀdiones were synthesized by
condensation of the corresponding aniline (5.5 mmol) with sodium 3,4ꢀdioxoꢀ3,4ꢀ
dihydronaphthaleneꢀ1ꢀsulfonate (5 mmol).24 Syntheses were conducted in 150 ml
of water and after solvent removal, the products were purified using flash
chromatography, to generate red solids. The synthesis of compounds 9ꢀ14 have
been previously reported and their physical parameters are reported in the
Supporting Information.24, 27, 28, 36, 37
4ꢀ((4ꢀIodophenyl)amino)naphthaleneꢀ1,2ꢀdione (15)
Compound 15 was synthesized according to the general method using 4ꢀ
iodoaniline as the starting material. This yielded a red solid (148 mg, 78%). Mp
262 ºC; UVꢀVis λmax: 280 nm, 340 nm and 456 nm in methanol. 1H NMR (500
MHz, DMSO) δ 8.32 (d, J = 7.9 Hz, 1H), 8.03 (dd, J = 1.4, 7.8 Hz, 1H), 7.82 (dd,
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
1
2
3
4
5
6
7
8
9
J = 7.0, 8.4 Hz, 1H), 7.74 (m, 4H), 6.94 (d, J = 8.1 Hz, 2H), 5.88 (s, 1H). 13C
NMR (126 MHz, DMSO) δ 182.24, 155.71, 148.96, 138.28, 137.53, 134.08,
133.48, 131.66, 131.59, 127.31, 124.86, 116.96, 102.14, 76.15. HRMS (ESI):
m/z 375.9838 (M + H) (calcd for C17H12NIO2, 375.9834) (M + H).
General methods for 4ꢀsubstituted (4ꢀphenyl(methyl)amine)naphthaleneꢀ1,2ꢀ
diones (16ꢀ22)
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
4ꢀSubstituted (4ꢀphenyl(methyl)amine)naphthaleneꢀ1,2ꢀdiones were synthesized
in an identical fashion to that described for the (4ꢀphenylamine)naphthaleneꢀ1,2ꢀ
diones (see above) to afford red solids. The synthesis of compound 16 has been
previously reported and its physical parameters are reported in the Supporting
Information.28
4ꢀ((4ꢀMethoxyphenyl)(methyl)amino)naphthaleneꢀ1,2ꢀdione (17)
Compound 17 was synthesized according to the general method using 4ꢀ
methoxyꢀNꢀmethylaniline as the starting material. This yielded a red solid (59 mg,
40%). Mp 167 ºC; UVꢀVis λmax: 280 nm, 383 nm, and 483 nm in methanol. 1H
NMR (500 MHz, CDCl3) δ 8.07 (dd, J = 1.5, 7.6 Hz, 1H), 7.35 (td, J = 1.0, 7.5 Hz,
1H), 7.24 (td, J = 1.6, 7.8 Hz, 1H), 7.08 – 7.01 (m, 3H), 6.90 – 6.83 (m, 2H), 6.25
(s, 1H), 3.80 (s, 3H), 3.46 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 181.04, 178.29,
159.36, 158.12, 140.75, 133.25, 132.69, 132.24, 129.90, 129.01, 128.89, 126.72
(2C), 115.18 (2C), 109.83, 55.54, 44.62. HRMS (ESI): m/z 294.1129 (M + H)
(calcd for C18H15NO3, 294.1130) (M + H).
4ꢀ((4ꢀMethylphenyl)(methyl)amino)naphthaleneꢀ1,2ꢀdione (18)
Compound 18 was synthesized according to the general method using 4ꢀmethylꢀ
Nꢀmethylaniline as the starting material. This yielded a red solid (35 mg, 25%).
22
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Mp 140 ºC; UVꢀVis λmax: 274 nm, 320 nm, 376 nm, and 467 nm in methanol. 1H
NMR (500 MHz, CDCl3) δ 8.08 (dd, J = 1.5, 7.8 Hz, 1H), 7.35 (td, J = 1.1, 7.5 Hz,
1H), 7.29 – 7.19 (m, 1H), 7.18 – 7.10 (m, 2H), 7.09 – 6.96 (m, 3H), 6.27 (s, 1H),
3.47 (s, 3H), 2.34 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 180.96, 178.43, 159.54,
145.34, 136.57, 133.25, 132.74, 132.19, 130.58 (2C), 129.95, 129.06, 128.90,
125.32 (2C), 110.39, 44.40, 20.99. HRMS (ESI): m/z 278.1180 (M + H) (calcd for
C18H15NO2, 278.1181) (M + H).
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
4ꢀ((4ꢀFluorophenyl)(methyl)amino)naphthaleneꢀ1,2ꢀdione (19)
Compound 19 was synthesized according to the general method using 4ꢀfluoroꢀ
Nꢀmethylaniline as the starting material. This yielded a red solid (44 mg, 31%).
Mp 190 ºC; UVꢀVis λmax: 272 nm, 320 nm, 370 nm, and 461 nm in methanol. 1H
NMR (500 MHz, CDCl3) δ 8.09 (dd, J = 1.5, 7.7 Hz, 1H), 7.38 (td, J = 1.1, 7.6 Hz,
1H), 7.27 (td, J = 1.5, 7.8 Hz, 1H), 7.15 – 7.05 (m, 2H), 7.09 – 7.00 (m, 3H), 6.28
(s, 1H), 3.47 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 180.59, 178.57, 160.67 (d, J
= 248.0 Hz), 159.41, 143.98, 133.43, 132.45, 132.14, 130.17, 129.31, 128.73,
126.95 (d, J = 8.5 Hz)(2C), 116.96 (d, J = 23.1 Hz)(2C), 111.00, 44.48. HRMS
(ESI): m/z 282.0934 (M + H) (calcd for C17H12FNO2, 282.0930) (M + H).
4ꢀ((4ꢀChlorophenyl)(methyl)amino)naphthaleneꢀ1,2ꢀdione (20)
Compound 20 was synthesized according to the general method using 4ꢀchloroꢀ
Nꢀmethylaniline as the starting material. This yielded a red solid (88 mg, 59%).
Mp 182 ºC; UVꢀVis λmax: 245 nm, 275 nm, 320 nm, 373 nm, and 462 nm in
1
methanol. H NMR (500 MHz, CDCl3) δ 8.08 (dd, J = 1.5, 7.6 Hz, 1H), 7.39 (td, J
= 1.1, 7.5 Hz, 1H), 7.31 (m, 3H), 7.06 (m, 3H), 6.29 (s, 1H), 3.48 (s, 3H). 13C
NMR (126 MHz, CDCl3) δ 180.44, 178.67, 159.27, 146.38, 133.60, 132.40,
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
7
8
132.02, 131.76, 130.24, 130.08 (2C), 129.32, 128.70, 126.33 (2C), 111.76, 44.06.
HRMS (ESI): m/z 298.0636 (M + H) (calcd for C17H12ClNO2, 298.0635) (M + H).
4ꢀ((4ꢀBromophenyl)(methyl)amino)naphthaleneꢀ1,2ꢀdione (21)
9
Compound 21 was synthesized according to the general method using 4ꢀbromoꢀ
Nꢀmethylaniline as the starting material. This yielded a red solid (62 mg, 36%).
Mp 182 ºC; UVꢀVis λmax: 246 nm, 319 nm, 365 nm and 463 nm in methanol. 1H
NMR (500 MHz, CDCl3) δ 8.07 (dd, J = 1.4, 7.7 Hz, 1H), 7.46 (m, 2H), 7.39 (td, J
= 1.0, 7.5 Hz, 1H), 7.30 (m, 1H), 7.04 (m, 3H), 6.29 (s, 1H), 3.48 (s, 3H). 13C
NMR (126 MHz, CDCl3) δ 180.40, 178.67, 159.22, 146.85, 133.63, 133.02 (2C),
132.36, 131.99, 130.24, 129.31, 128.68, 126.61 (2C), 119.48, 111.90, 43.96.
HRMS (ESI): m/z 342.0127 (M + H) (calcd for C17H12BrNO2, 342.0129) (M + H).
4ꢀ((4ꢀIodophenyl)(methyl)amino)naphthaleneꢀ1,2ꢀdione (22)
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
Compound 22 was synthesized according to the general method using 4ꢀiodoꢀNꢀ
methylaniline as the starting material. This yielded a red solid (74 mg, 38%). Mp
184 ºC; UVꢀVis λmax: 270 nm, 316 nm, and 452 nm in methanol. 1H NMR (500
MHz, CDCl3) δ 8.10 (dd, J = 1.4, 7.6 Hz, 1H), 7.65 (m, 2H), 7.40 (td, J = 1.1, 7.5
Hz, 1H), 7.31 (td, J = 1.5, 7.8 Hz, 1H), 7.05 (d, J = 8.1 Hz, 1H), 6.86 (m, 2H),
6.30 (s, 1H), 3.47 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 180.38, 178.76, 159.20,
147.58, 139.00 (2C), 133.66, 132.40, 132.03, 130.29, 129.42, 128.65, 126.82
(2C), 112.18, 90.35, 43.88. HRMS (ESI): m/z 389.9990 (M + H) (calcd for
C17H12INO2, 389.9991) (M + H).
1,2ꢀNaphthoquinone (23)
24
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Compound 23 was synthesized as previously described16, and purified by
sublimation to yield orange plateꢀlike crystals. Physical parameters for this
molecule are provided in the Supporting Information.
9
Enzyme inhibition assays
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
The inhibition of CEꢀmediated oꢀnitrophenyl acetate (oꢀNPA) and oseltamivir
hydrolysis by the 1,2ꢀdiones was performed as previously described.32 Routinely
all data points were performed in quadruplicate, with at least 8 different
concentrations of inhibitor. Analysis of results was undertaken by fitting data to
the following equation to determine the mode of enzyme inhibition:24
[I]{[s](1−
β
) + Ks(
α
−
β
α
)}
Ks}
i =
[I]{[s] + Ks}+ Ki{
α
α
[s] +
where i = fractional inhibition, [I] = inhibitor concentration, [s] = substrate
concentration, α = change in affinity of substrate for enzyme, = change in the
β
rate of enzyme substrate complex decomposition, Ks is the dissociation constant
for the enzyme substrate complex and Ki is the inhibitor constant. GraphPad
Prism software was used to evaluate the curve fits where α ranged from 0 to ∞
and
β
from 0 to 1. Those generating the highest r2 values were then analyzed
using Akaike’s information criteria38, 39 to identify the best model for enzyme
inhibition. Ki values for enzyme inhibition were then calculated from the equation
predicted by Prism to be the best fit for the experimental data.32
Inhibition of human AChE was assessed using a spectrophotometric assay with
acetylthiocholine (ATCh) as a substrate.40
Small molecule docking
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
1
2
3
4
5
6
7
8
9
Small molecules were drawn, optimized and docked into the hCE1 enzyme
active site (RCSB code 1MX1) using ICM Pro software (Molsoft LLC, San Diego,
CA). Default program parameters were used for docking, except that the
‘Thoroughness’ value was set at 10. This allows the program to undertake a
more detailed analysis of the different poses adopted by the compound during
the docking procedure. ICM scores obtained from these analyses provided an
estimate of the likelihood of small molecule binding. This score (lower is better) is
based upon the following parameters: hydrogen bond interactions between the
small molecule and the active site; the internal force field energy of the ligand;
differences in solvation energies upon ligand binding; loss of entropy from
unbound and bound conditions; hydrophobic and electrostatic energies; and Hꢀ
bond acceptor and donor desolvation. Images of docking poses were then
generated by displaying the three catalytic amino acids (S221, E353, H464) and
removing the protein ribbon. Distances from the serine Oγ atom to the carbonyl
or hydroxyl carbon atoms within the small molecules were determined from the
respective docking file.
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
Small molecule xꢀray crystallography
Selected crystals were affixed to MiTeGen sample supports with lowꢀviscosity
cryogenic oil, and flash cooled to 100K for xꢀray analysis on a Bruker X8 κ
diffractometer. The crystal was illuminated with the Xꢀray beam from a Bruker
IꢂSCu microfocus sealed tube with the resulting images being integrated using
SAINT software (Bruker) using a narrowꢀframe algorithm. A multiꢀscan
absorption correction was applied, and data were corrected for interꢀframe
26
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
scaling differences with SADABS. The structure was solved via the SHELXTL
software package and refined via SHELXL 2014 (Bruker). Hydrogen atoms
were placed in idealized positions, and refined according to a riding model.
Coordinates for all structures have been deposited at the Cambridge
Crystallographic Data Centre (CCDC).
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
The atom coordinates for compounds 24 and 25, AMNPQH10 and XANRUB,
respectively, were obtained from the Cambridge Structural Database and
analyzed using Mercury software.
In vivo inhibition of oseltamivir and irinotecan metabolism
To assess inhibition of oseltamivir, 1 x 105 U373MGhCE1 cells were plated into
individual wells of a 96 well plated and allowed to attach overnight. The next day,
inhibitor (5 ꢁM) was added to cells and after 30 min, the media was removed and
replaced with fresh media containing 20 ꢁM oseltamivir and 5 ꢁM inhibitor. After
30 min, the media was removed, an equal volume of acetonitrile was added, and
the concentrations of metabolites in these samples were determined by UPLCꢀ
MS. Assays for the inhibition of hydrolysis of irinotecan were performed as
previously described.34
All assays were run in duplicate, and DMSO was used as a control. In addition,
blank samples (i.e., media lacking cells) were analyzed and these concentration
values were subtracted from all results.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
1
2
3
Ancillary Information
4
5
6
7
8
Supporting Information. The synthetic methods and physical parameters for
previously described compounds; the UPLC traces and 1H, 13C and 19F (as
appropriate) spectra for all molecules; the crystallographic data and coordinates
for compounds 2, 9, 16, and 23; and, cell growth inhibition curves for compounds
16, 18 and 19.
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
The CCDC codes for molecules 2, 9, 16 and 23 are 1526906, 1526905, 1526907
and 1526904, respectively. Authors will release the atomic coordinates and
experimental data upon article publication.
Molecular Formula Strings information for all compounds are also included as
Supporting Information.
Corresponding author: Philip M. Potter, Department of Chemical Biology and
Therapeutics, St. Jude Children's Research Hospital, 262 Danny Thomas Place,
Memphis, TN 38105, United States
Tel: 901ꢀ595ꢀ6045; Fax: 901ꢀ595ꢀ4293; Eꢀmail: phil.potter@stjude.org.
Acknowledgements
This work was supported in part by NIH grants CA108775 and AT007531, a
Cancer Center Core grant CA21765, and by the American Lebanese Syrian
Associated Charities (ALSAC) and St. Jude Children’s Research Hospital
28
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
(SJCRH). We thank William Pomerantz for helpful discussions with interpretation
of the 19F NMR spectra.
6
7
8
9
Abbreviations Used: AChE – acetylcholinesterase ; ATCh – acetylthiocholine;
CCDC – Cambridge Crystallographic Data Centre; CE – carboxylesterase;
irinotecan – 7ꢀethylꢀ10ꢀ[4ꢀ(1ꢀpiperidino)ꢀ1ꢀpiperidino]carbonyloxycamptothecin;
hCE1 – human liver CE, CES1; hiCE – human intestinal CE, CES2; IC50 –
concentration of drug required to inhibit cell growth by 50%; oꢀNPA – oꢀ
nitrophenyl acetate.
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
29
ACS Paragon Plus Environment