Diverse amide analogs of sulindac for cancer treatment and prevention

Article abstract of DOI:10.1016/j.bmcl.2017.09.022

Sulindac is a non-steroidal anti-inflammatory drug (NSAID) that has shown significant anticancer activity. Sulindac sulfide amide (1) possessing greatly reduced COX-related inhibition relative to sulindac displayed in vivo antitumor activity that was comparable to sulindac in a human colon tumor xenograft model. Inspired by these observations, a panel of diverse sulindac amide derivatives have been synthesized and their activity probed against three cancer cell lines (prostate, colon and breast). A neutral analog, compound 79 was identified with comparable potency relative to lead 1 and activity against a panel of lymphoblastic leukemia cell lines. Several new series also show good activity relative to the parent (1), including five analogs that also possess nanomolar inhibitory potencies against acute lymphoblastic leukemia cells. Several new analogs identified may serve as anticancer lead candidates for further development.

Full text of DOI:10.1016/j.bmcl.2017.09.022

Accepted Manuscript
Diverse amide analogs of sulindac for cancer treatment and prevention
Bini Mathew, Judith V. Hobrath, Michele C. Connelly, R. Kiplin Guy, Robert
C. Reynolds
PII:
S0960-894X(17)30918-6
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.09.022
BMCL 25286
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
2 August 2017
4 September 2017
11 September 2017
Please cite this article as: Mathew, B., Hobrath, J.V., Connelly, M.C., Kiplin Guy, R., Reynolds, R.C., Diverse
amide analogs of sulindac for cancer treatment and prevention, Bioorganic & Medicinal Chemistry Letters (2017),
doi: http://dx.doi.org/10.1016/j.bmcl.2017.09.022
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Diverse amide analogs of sulindac for cancer treatment and prevention
Bini Mathewa, Judith V. Hobrathb, Michele C. Connellyc, R. Kiplin Guyd, Robert C. Reynoldse
aDrug Discovery Division, Southern Research Institute, 2000 Ninth Avenue South, Birmingham, AL 35205, USA
bDrug Discovery Unit, College of Life Sciences, University of Dundee, Dundee DD1 5EH, United Kingdom
cDepartment of Chemical Biology & Therapeutics, St Jude Children's Research Hospital, 262 Danny Thomas
Place, Mailstop 1000, Memphis, TN 38105-3678 USA
dThe University of Kentucky College of Pharmacy, 214H BioPharm Complex, Lexington, KY 40536-0596, USA
eDivision of Hematology and Oncology, The University of Alabama at Birmingham, Birmingham, Alabama
35294, USA
Abstract: Sulindac is a non-steroidal anti-inflammatory drug (NSAID) that has shown significant
anticancer activity. Sulindac sulfide amide (1) possessing greatly reduced COX-related
inhibition relative to sulindac displayed in vivo antitumor activity that was comparable to
sulindac in a human colon tumor xenograft model. Inspired by these observations, a panel of
diverse sulindac amide derivatives have been synthesized and their activity probed against three
cancer cell lines (prostate, colon and breast). A neutral analog, compound 79 was identified
with comparable potency relative to lead 1 and activity against a panel of lymphoblastic
leukemia cell lines. Several new series also show good activity relative to the parent (1),
including five analogs that also possess nanomolar inhibitory potencies against acute
lymphoblastic leukemia cells. Several new analogs identified may serve as anticancer lead
candidates for further development.
Key words: NSAIDs; sulindac; amides; cancer
Nonsteroidal anti-inflammatory drugs (NSAIDs) are a chemically diverse family of drugs commonly used
clinically to treat a variety of inflammatory conditions including pain associated with arthritis. A number of these
drugs possess antipyretic activity in addition to having analgesic and anti-inflammatory action, and thus have
utility in the treatment of pain and fever. These are widely known to be cyclooxygenase (COX) inhibitors. The
COX enzymes participate in the metabolism of prostaglandins. For example, COX-1 enzymes produce
prostaglandins that are important for the stomach lining and kidney function. COX-2 enzymes are crucial to anti-
inflammatory reactions in the body.1 Many currently marketed NSAIDs have both COX-1 and COX-2 inhibitory
effects. Although the class is highly used both clinically and over the counter, chronic COX inhibition is
________________
*Corresponding author. Tel.: +1-(205) 934-8276; Fax: +1-(205) 934-2543. E-mail: rcr12lkt@uab.edu
Abbreviations: NSAID, non-steroidal anti-inflammatory drug; COX, cyclooxygenase; SSA, sulindac sulfide amide; SAR,
structure-activity relationship; AA, amino acid; ALL, acute lymphoblastic leukemia; CPC, choroid plexus carcinoma; TPSA,
topological polar surface area; LipE, lipophilic efficiency.

associated with gastrointestinal, renal, and cardiovascular side effects.2-6 Epidemiological, preclinical, and clinical
studies suggest that NSAIDs and COX-2 inhibitors display striking cancer chemopreventive efficacy by reducing
cancer incidence in the general population.1,7,8 Although the mechanism responsible for the anticancer activity of
NSAIDs is still unknown, it is commonly attributed to COX-2 mediation.9-11 Studies have shown that several
human tumors contain increased levels of COX-2 as compared to normal tissues. It has been determined that
excess expression of COX-2 in cancer cells correlates to increased tumorigenic potential. Experiments have
demonstrated that COX-2 specific NSAIDs restore apoptotic responses, or programmed cell death, in tumor cells.
By selective enzyme inhibition and restoration of normal apoptotic responses, research data support the
hypothesis that the anti-cancer mechanism is COX-2 dependent. Several COX-2 independent mechanisms have
also been postulated for the antineoplastic properties of NSAIDs including activation of apoptosis, inhibition of
angiogenesis, or direct inhibition of cancer cell growth by blocking signal transduction pathways responsible for
cell proliferation.12-15 A number of molecular targets have been implicated in the cyclooxygenase independent
pathways of anti-cancer effects of NSAIDs, as reviewed recently,16,17 and it was our intention to develop potential
probes for these activities through ready variations in the scaffold.
Sulindac, or Clinoril, is a substituted indene-3-acetic acid, and is a clinically used NSAID with established
chemopreventive activity.13 Sulindac contains a chiral center (the sulfoxide), although the commercial drug is sold
as a racemic mixture. In the body, the drug exists primarily in three active forms (see Figure 1) due to oxidation-
reduction cycling. This cycling scrambles chirality at the SO bond and oxidation to the sulfone is irreversible.13
The sulfide metabolite generated through bioreduction in vivo is a potent, non-selective COX enzyme inhibitor
and shows compelling anticancer activity as compared with other NSAIDS.
Figure 1. In vivo metabolism (oxidation/reduction cycling) of the –SOCH3 center

Based on our and others’ data, it is clear that the NSAIDs have multifarious actions ranging from anti-
inflammatory and anti-proliferative effects to selective anticancer cytotoxic activity.1,18 The fact that NSAIDs are
relatively safe and orally effective suggests that there is an opportunity through simple, specific alterations of the
diverse scaffolds to modify the basic structure to probe these various activities, and, at the same time, produce
derivations that are more likely to yield good biological probes, animal activity, and drug leads. As an example,
the preparation of neutral carboxamides is known to bias the NSAID scaffold towards selective COX-2 activity.
As well, and supported by our preliminary results that replacement of the carboxylate functionality in sulindac in
the form of sulindac sulfide amide (SSA) (1) can virtually abolish COX-related activity and toxicity while
potentiating anticancer activity in vitro and yielding in vivo xenograft activity.18 These results are promising in
that sulindac derivatives lacking the carboxylate functionality may show attenuated toxicity (gastric, renal, and
cardiovascular) associated with COX inhibition. The presented chemical modifications have begun to explore,
and possibly improve on, other NSAID activity space relating to selective cancer prevention and therapy.
Herein, we describe the synthesis and anti-cancer activity of a series of sulindac analogs. Compound 1 (SSA)
was considered the lead compound, and for this study two modification sites were chosen, the methylthiophenyl
ring, and the acetamide linker (Figure 2). Various aryl groups were introduced in the indene core ring and a series
of amides and sulfonamides were synthesized. All compounds were screened against three cancer cell lines
(prostate, colon and breast) and selected compounds against a panel of additional cancer cell lines (five leukemia
cancer cell lines and choroid plexus carcinoma cell line).
Figure 2. Outline of structural modifications on the lead compound 1
Chemistry
Compounds 1-86 were synthesized by the coupling of different sulindac derivatives with various amines
using either HBTU19 or HATU20 as the coupling agent (Scheme 1). Sulindac derivatives, if not commercially

available, were synthesized from 2-(5-fluoro-2-methyl-1H-inden-3-yl)acetic acid and the corresponding aryl
aldehydes in the presence of NaOMe.21
Scheme 1. Synthetic pathway to analogs 1-86. Reagents: (a) HBTU or HATU, TEA or DIEA, MeCN
N,N’-Dimethylaminoethyl derivatives of various sulindac analogs (10, 22, 31) were treated with Boc-L-
valine under the same coupling reaction conditions followed by hydrolysis for the removal of Boc protecting
group to produce 87-89 in good yields. Sulfonamides 90-103 were also synthesized from amides, 10 and 22 by
reacting with suitable sulfonyl chlorides in pyridine and N-methylimidazole (Scheme 2).
Scheme 2. 87-103. Reagents and conditions: (a) HBTU, TEA, Boc-L-valine, MeCN (b) H+ (c) RSO2Cl, NMI, Pyridine, 0 oC
Screening Results
Using quantitative high-throughput screening all reported compounds were evaluated against PC3
prostate, HT29 colorectal carcinoma and MDA-MB-231 breast cancer cell lines. Description of the assays and
screening method is provided in Appendix A, supplementary materials. SSA (1) was identified earlier as a lead

compound for our structure-activity relationship (SAR) study with modification at the methylthiophenyl ring, and
the acetamide linker. Amides shown in Scheme 1 were categorized as acyclic basic amides (Table 1), cyclic basic
amides (Table 2), aromatic basic amides (Table 3) and neutral amides (Table 4). Screening data for the acyclic
basic amides with modifications at the 4-methylthiobenzylidene ring at the C-1 position are shown in Table 1.
First, the effect of various acyclic basic amide linkers at the C-3 position was explored (2-14) while retaining the
4-thiomethylbenzylidene group at the C-1 position.
Table 1. Screening data for control 1 and basic acyclic amides 2-37.
Ar
Cmpd
R1 & R2
CC50 (µM)
PC3
HT29
MDA-MB-231
2.67±0.08
5.18±0.46
4.20±0.17
3.19±0.08
3.98±0.45
5.41±0.42
5.22±0.20
3.77±0.35
3.96±1.46
5.84±0.35
7.66±0.49
6.65±0.22
3.03±0.19
3.92±0.39
>50.00
R1 = H, R2 = -CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2N(CH2CH3)2
R1 = H, R2 = -CH2CH2CH2N(CH2CH3)2
R1 = CH3, R2 = -CH2CH2N(CH3)2
R1 = CH3, R2 = -CH2CH2CH2N(CH3)2
R1 = CH2CH3, R2 = -CH2CH2N(CH2CH3)2
R1 = CH3, R2 = -CH2CH2N(CH2CH3)2
R1 = CH2CH3, R2 = -CH2CH2N(CH3)2
R1 = CH3, R2 = -CH2CH2N(CH3)H
0.65±0.03
2.46±0.14
1.54±0.05
1.53±3.33
1.83±0.04
2.18±0.03
1.83±0.04
1.49±0.02
1.36±0.05
2.80±0.07
2.90±0.05
2.54±0.10
1.38±0.05
3.12±0.15
6.99±0.19
5.45±0.35
4.33±0.22
5.63±0.26
6.98±0.27
6.21±0.44
5.65±0.26
5.07±0.31
6.75±0.24
7.65±0.29
6.56±0.43
3.92±0.29
4.90±0.23
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
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
R1 = CH3, R2 = -CH2CH2CH2N(CH3)H
R1 = CH2CH3, R2 = -CH2CH2N(CH2CH3)H
R1 = CH2CH3, R2 = -CH2CH2CH2N(CH2CH3)H
R1 = CH(CH3)2, R2 = -CH2CH2CH2N(CH(CH3)2)H 2.30±0.07
R1 = H, R2 = -CH2CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2N(CH2CH3)2
33.27±2.92
19.78±0.77
27.12±2.0
38.92±7.98
18.40±0.78
23.97±1.51
18.15±0.94
34.53±2.45
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
R1 = H, R2 = -CH2CH2CH2N(CH2CH3)2
R1 = CH3, R2 = -CH2CH2CH2N(CH3)2
R1 = CH2CH3, R2 = -CH2CH2N(CH2CH3)2
R1 = CH3, R2 = -CH2CH2N(CH2CH3)2
R1 = CH2CH3, R2 = -CH2CH2N(CH3)2
R1 = CH3, R2 = -CH2CH2N(CH3)H
R1 = CH3, R2 = -CH2CH2CH2N(CH3)H
R1 = CH2CH3, R2 = -CH2CH2N(CH2CH3)H
R1 = CH2CH3, R2 = -CH2CH2CH2N(CH2CH3)H
18.18±1.40
17.48±1.14
>50.00
>50.00
>50.00
>50.00
>50.00
R1 = CH(CH3)2, R2 = -CH2CH2CH2N(CH(CH3)2)H 16.28±0.97 48.28±8.05
R1 = H, R2 = -CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2N(CH2CH3)2
R1 = CH3, R2 = -CH2CH2N(CH3)2
R1 = CH3, R2 = -CH2CH2N(CH3)H
R1 = H, R2 = -CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2CH2N(CH3)2
R1 = H, R2 = -CH2CH2N(CH2CH3)2
R1 = CH3, R2 = -CH2CH2N(CH3)2
6.22±0.44 18.27±1.34
6.08±0.30 13.03±0.74
5.42±0.38 13.88±0.78
4.68±0.23 11.44±0.52
8.05±0.31 23.06±1.31
1.57±0.07
1.05±0.05
1.12±0.05
2.02±0.06
15.05±0.88
10.79±0.45
12.26±0.34
10.78±0.65
16.96±1.04
3.63±0.27
1.93±0.13
3.54±0.24
3.80±0.51
4.88±0.53
3.25±0.13
4.48±0.32
5.88±0.32
R1 = H, R2 = -CH2CH2N(CH3)2
6.89±0.58 28.90±2.05
18.06±1.07
36
R1 = H, R2 = -CH2CH2N(CH3)2
12.67±0.73
>50.00
37.46±3.45
37
Among these 13 compounds, none was found to be more active compared to the lead agent 1 although many
showed relatively comparable cell growth inhibition. Simple modifications of the N,N-dimethylaminoethylamino

group of 1 reduced the activity by 2-5 fold. Compound 2, with a N,N-dimethylpropyl substitution at the
acetamide linker, reduced the activity by 4-fold compared to the lead compound, SSA (1) in HT29 cells, and by 2-
fold in PC3 and MDA-MB-231. Compounds 3 and 4 with a N,N-diethylamino group at the ethylene or propylene
acetamide groups were more active than 2, but less potent than 1 in all three cell lines. Either an additional
substituent at the acetamide nitrogen or removal of a substituent from the terminal nitrogen did not improve the
activity (5-14) dramatically. While maintaining the 4-methylsulfinylbenzylidene at the C-1 position, we explored
the SAR of selected basic acyclic acetamide linkers (15-26). All these analogs led to reduced, or a complete loss
of potency. Selected basic acyclic acetamide linkers with a 3,4,5-trimethoxybenzylidene group at the C-1 position
(27-31) showed better activity than their corresponding 4-methylsulfinylbenzylidene analogs, but were less active
than corresponding 4-thiobenzylidene analogs. These compounds were less active than 1, but still were highly
active in vitro in all three assays. We then expanded the SAR with selected basic acyclic acetamide linkers, while
keeping the benzothiazol-2-ylmethylene group as a fixed unit (32-35). These analogs demonstrated significant
activity similar to their 4-methylthiobenzylidene counterparts. We also explored the anticancer activity of pyridin-
4-ylmethylene (36) and 4-sulfonylbenzylidene (37) analogs of 1. These compounds were modestly active against
all three cell lines, but less potent than 1.
Screening results for basic cyclic amides 38-66 are summarized in Table 2. Compounds 38-51, with a 4-
methylthiobenzylidene ring at the C-1 position and various basic cyclic acetamide linkers at the C-3 position
showed significant activity against cancer similar to acyclic basic amide analogs 1-14. We also explored the SAR
of selected aryl modifications at the C-1 position compared to selected basic cyclic acetamides (52-66). However,
none of these series, except the benzothiazolyl analogs (61-65), demonstrated significant activity.
Table 2. Screening data for basic cyclic amides 38-66.
CC50 (µM)
PC3
Ar
Cmpd
R1 & R2
HT29
MDA-MB-231
1.81±0.06
5.27±0.32
3.78±0.26
38
39
40
R1 = H, R2 =
R1 = H, R2 =
1.65±0.14
5.36±0.18
5.42±0.24
7.66±0.52
2.98±18.68 6.66±0.43
R1 = H, R2 =
1.41±0.04
1.16±0.08
4.33±0.23
4.64±0.23
3.75±0.23
3.72±0.23
R1 = H, R2 =
41
42
R1 = H, R2 =
1.55±0.06
5.10±0.36
4.16±0.29
R1, R2 =
43
1.69±0.08
2.17±0.06
1.13±0.07
1.06±0.05
1.64±0.07
4.28±0.16
6.23±0.37
2.76±0.19
4.06±0.27
4.45±0.16
3.77±1.18
4.05±0.23
3.08±0.07
2.61±0.09
4.10±0.19
R1, R2 = -(-CH2CH2-)2NCH2CH2CH2N(CH3)2
R1, R2 = -(-CH2CH2-)2NCH2CH2N(CH3)2
R1, R2 = -(-CH2CH2-)2N(pyridin-4-yl)
R1, R2 = -(-CH2CH2-)2CH(pipyridin-1-yl)
44
45
46
47
48
49
R1 = H, R2 =
R1, R2 =
4.29±0.19 13.42±0.49
11.50±0.53
7.21±0.22
2.94±0.16
7.35±0.32
50
51
R1, R2 =
R1, R2 =
2.47±0.67
6.50±0.84
3.69±0.17

R1, R2 = -(-CH2CH2-)2NCH2CH2CH2N(CH3)2
R1, R2 = -(-CH2CH2-)2N(Pyridin-4-yl)
R1, R2 = -(-CH2CH2-)2CH(pipyridin-1-yl)
23.75±1.48
9.20±0.39 21.93±2.16
>50.00
19.03±1.26
>50.00
46.18±2.63
22.74±2.29
>50.00
52
53
54
>50.00
>50.00
>50.00
R1 = H, R2 =
R1, R2 =
55
56
20.16±1.06
>50.00
>50.00
13.39±0.71 46.49±5.98
41.09±3.78
R1 = H, R2 =
57
4.18±0.19
2.79±0.13
9.62±0.66
5.45±0.68
10.38±0.59
6.50±0.77
11.14±0.56
3.56±0.28
R1 = H, R2 =
58
59
R1, R2 = -(-CH2CH2-)2N(Pyridin-4-yl)
4.22±0.15 12.28±0.72
60
R1, R2 =
1.21±0.04
1.62±0.06
5.25±0.20 12.28±0.37
1.66±0.04
4.94±0.28
4.21±0.23
R1 = H, R2 =
R1, R2 = -(-CH2CH2-)2NCH2CH2CH2N(CH3)2
61
62
63
2.90±0.07
7.71±1.5
5.58±0.33
R1, R2 = -(-CH2CH2-)2N(Pyridin-4-yl)
7.72±0.45
R1 = H, R2 =
64
1.58±0.04
4.25±0.15
3.61±0.23
7.19±0.31
R1, R2 =
65
66
2.68±0.12 10.19±0.41
R1 = H, R2 =
While maintaining the 4-methylthiobenzylidene at the C-1 position, we also examined the effects of basic
aromatic acetamide linkers at the C-3 position (Table 3). Compounds 67 and 68 with a N,N-dimethylamino group
at the phenyl group were inactive in all three assays. Compounds with 2, 3, and 4-pyridylmethyl acetamide (69,
70 and 71) showed inhibition but all with lesser potency than lead 1.
Table 3. Screening data for basic aromatic amides 67-71.
CC50 (µM)
PC3
>50.00
>50.00
5.83±1.18
18.52±7.78
Ar
Cmpd
R1 & R2
HT29
>50.00
>50.00
4.88±0.18
4.83±1.1
MDA-MB-231
>50.00
30.60±15.76
6.46±0.93
7.36±1.94
13.69±1.85
R1 = H, R2 = 4-(N,N-dimethylamino)benzyl
R1 = H, R2 = 4-(N,N-dimethylamino)phenyl
R1 = H, R2 = 4-pyridylmethyl
67
68
69
70
71
R1 = H, R2 = 3-pyridylmethyl
R1 = H, R2 = 2-pyridylmethyl
10.31±0.79 13.64±2.10
We next explored the anticancer activity of neutral amides (Table 4).
Table 4. Screening data for neutral amides 72-86.
CC50 (µM)
PC3
16.36±8.55
12.98±6.83
11.03±5.95
>50.00
Ar
Cmpd
R1 & R2
HT29
MDA-MB-231
3.61±1.87
3.40±1.07
5.66±2.69
>50.00
R1 = H, R2 = benzyl
1.52±0.39
2.08±0.82
2.16±0.71
>50.00
72
73
74
75
76
77
R1 = H, R2 = furan-2-ylmethyl
R1 = H, R2 = thiophen-2-ylmethyl
R1 = H, R2 = -CH2CH2CH(CH3)2
R1 = H, R2 = D-Alanine methyl ester 16.22±2.61 44.18±7.38
R1 = H, R2 = L-Alanine methyl ester 23.92±1.92 >50.00
31.99±2.33
38.86±6.04

R1 = H, R2 = L-Histidine methyl ester 23.01±3.09
>50.00
2.64±1.19
>50.00
>50.00
78
79
80
R1 = H, R2 = benzyl
0.50±0.23
1.00±0.50
0.78±0.32
2.73±1.21
R1 = H, R2 = furan-2-ylmethyl
R1 = H, R2 = -CH2CH2CH(CH3)2
17.65±1.11 20.23±2.29
27.72±3.37
81
R1 = H, R2 = -CH2CH2CH(CH3)2
5.24±1.18
8.20±1.92
>50.00
8.73±1.97
82
R1 = H, R2 = benzyl
23.56±2.04
19.63±2.98
6.70±2.23
24.20±5.35
83
84
85
R1 = H, R2 = furan-2-ylmethyl
R1 = H, R2 = -CH2CH2CH(CH3)2
5.48±1.70 26.44±15.80
>50.00
>50.00
R1 = H, R2 = benzyl
2.14±0.46
5.71±1.75
3.68±1.10
86
Compounds 72-78 have a 4-methylthiobenzylidene ring at the C-1 position. Compound 72, with a benzyl
group at the acetamide linker, reduced the activity by 4-fold as compared to lead compound in HT29 and PC3
cells and in MDA-MB-231 cells by 1.5-fold. The aromatic heterocyclic compounds such as furan-2-yl analog 73
and thiophene-2-yl analog 74 had activity very similar to the corresponding phenyl analog 72 in all screening
assays, but were less active than lead 1. Compound 75, where the terminal dimethylamino group of 1 was
replaced by an isopropyl group, was not active. The amino acid analogs 76-78 were >25-fold less potent than the
lead compound. By maintaining the 4-methylsulfinylbenzylidene at the C-1 position, we explored the SAR of
selected acetamide linkers. Among these three compounds (79-81), compound 79 with a benzyl group at the
acetamide linker showed excellent activity relative to lead 1. Furan-2-yl analog 80 also showed significant
activity as compared to its methylthiobenzylidene analog 73, but was less active than 1. Selected acetamide
linkers with a 3,4,5-trimethoxybenzylidene group, benzothiazol-2-ylmethylene group or a pyridin-4-ylmethylene
group at the C-1 position (82-86) did not show improved activity.
Intestinal cells, cancer cells, and other cell types are known to have specific amino acid (AA) uptake
mechanisms that have been utilized to increase drug uptake upon AA conjugation. For example, conjugation of
the active HSV agent acyclovir and penciclovir as the valyl ester through the AA-COOH group results in blood
levels 3-5 times that of the parent drug due to specific valine uptake mechanisms.22,23 Inspired by this concept,
compounds 87-89 were synthesized from N-methyl-N-(2-(methylamino)ethyl)acetamide analogs 10, 22 and 31.
All the three compounds, however, were less active than their corresponding parent compounds (Table 5) and
potential gastric uptake remains untested.
Table 5. Screening data for compounds 87-89.
CC50 (µM)
Cmpd
HT29
PC3
MDA-MB-231
7.33±0.85
>50.00
3.72±1.50
30.20±0.59
7.08±0.77
>50.00
87
88
89
10.73±0.71 17.65±1.89
18.84±1.61

We also explored a sulfonamide linkage due to its inherent biostability and the opportunity to take advantage
of further diversity generation using commercially available sulfonyl chlorides as well as the ease and typical high
yields of sulfonamide formation. Further, certain sulfonamides have been reported to demonstrate anticancer
activity.24-26 Hence, compounds 90-103 were synthesized from 10 and 22 and screened in the three cancer cell
panel. A few analogs showed modest activity in these screens (Table 6).
Table 6. Screening data for compounds 90-103.
CC50 (µM)
Cmpd
HT29
PC3
MDA-MB-231
11.57±1.22
10.09±1.00
>50.00
4.04±0.52 15.54±1.39
4.20±0.43 14.34±1.97
90
91
8.82±1.16
12.92±1.24
16.20±3.80
35.28±7.75
>50.00
>50.00
>50.00
>50.00
>50.00
>50.00
92
93
94
95
>50.00
>50.00
>50.00
>50.00
96
12.36±0.98 42.59±7.47
12.54±0.80
>50.00
97
98
99
100
101
102
103
42.64±3.16
6.09±0.30
>50.00
8.20±0.57
12.17±1.37
17.77±2.46
29.08±7.05
32.08±8.17
>50.00
6.18±0.39 10.95±0.96
3.99±0.49
6.35±1.06
8.96±1.79
7.52±1.16
10.60±1.91 9.20±1.46
A selected set of analogs were additionally screened against a panel of six cancer cell lines at St. Jude
Children’s Research Hospital consisting of four acute lymphoblastic leukemia (ALL) cell lines, one lymphoma
line and a cell line derived from a mouse model of choroid plexus carcinoma (CPC), as described in Appendix A,
supplementary materials. The selected compounds included the most potent series, namely, compounds 1-14, 32-
35, 38-51, 59, 61-65 and 79. Out of the analog series presented in Table 4 – 5, representative analogs 72-74, 79,
82, 84, 87 and 89 were also evaluated against this panel. Table 7 lists screening results for lead 1 and the analogs
that show activity against at least one of these cancer cells. Compounds 5, 33, 35, 50, 62 and 89 stand out as
nanomolar growth inhibitors of at least one type of acute lymphoblastic leukemia cancer cell line on this
screening panel. Further, compound 79 shows low micromolar growth inhibition against all lymphoblastic
leukemia cell lines on the screening panel while very weak activity against the choroid plexus carcinoma cell line.
These analogs with the exception of 89 are also good inhibitors of the HT29, PC3 and MDA cancer cell lines.
Interestingly, a related scaffold containing a 3,4,5-trimethoxyphenyl aryl moiety, the indomethacin analog of
compound 27 was previously shown to inhibit potently proliferation of multiple cancer cell lines through
interfering with tubulin polymerization.27 Thus, we identified a number of analogs potently inhibiting multiple
types of cancer cell lines. These compounds may be pursued in future work as potential anticancer leads. Of note,
the parent hit sulindac sulfide amide (SSA, 1) showed no activity against cancer cell lines in this panel (Table 7).

Table 7. Cancer cell line screening data and cytotoxicity counter screen against BJ cells. Confidence intervals are
shown in parentheses; NA: not active; ND: undetermined due to questionable curve fit.
EC50 (µM)
Cmpds
CPC300
NA
NA
NA
65 (53-96)
NA
NA
NA
27 (26-29)
50 (48-52)
NA
NALM16
NA
RAJI
NA
21 (16-27)
JURKT
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
REH
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
27 (21-34)
18 (15-29)
NA
697
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
BJ
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
>7.57
1
4
5
ND
0.071 (0.059-0.1) 0.83 (0.26-0.79)
0.35 (0.27-0.53)
0.3 (0.16-0.74)
NA
NA
76 (36-72)
76 (48-74)
NA
33
35
43
44
45
46
50
59
61
62
63
79
89
ND
ND
1.9 (1.4-2.5)
NA
0.49 (0.13-1.3)
NA
NA
ND
NA
NA
NA
NA
NA
47 (26-73)
NA
NA
2.6 (2.4-2.7)
NA
0.097 (0.067-0.19)
NA
72 (56-130)
120 (95-150)
NA
NA
1.3 (1.2-1.3)
NA
1.3 (1.2-1.4) 3.3 (3.2-3.5)
NA ND
4.1 (4.0-4.1)
0.96 (0.26-1.2) >7.57
Cytotoxicity considerations and computed physicochemical properties.
Lipophilicity (in addition to molecular weight and other molecular characteristics) is a crucial
physicochemical property to consider during lead optimization.28-30 Like the parent compound, SSA, most analogs
presented contain a positively charged amine, which coupled with high lipophilicity raises concern that such
compounds may accumulate in cell membranes interacting with non-polar lipid chains while the positively
charged amine associates with negatively charged head groups of fatty acids. This may lead to cytotoxicity
through membrane destabilization. Selected series and representative analogs have been counter-screened for
cytotoxicity against BJ cells, a normal human foreskin fibroblast cell line31 where cytotoxicity was determined at
10 µM drug concentration. The following compounds were evaluated for cytotoxicity in this assay: compounds 1-
14, 32-35, 38-51, 59, 61-66, 72-74, 79, 82, 84, 87 and 89 (Table 7). All compounds screened showed EC50 > 7.57
µM in the BJ cytotoxicity assay suggesting that these compounds are not overtly toxic to this “normal” fibroblast
cell line (Table 7). Compound 79 lacks a charged amine, yet it has the highest efficacy against all three cancer cell
lines among the presented analogs while also potently inhibiting five lymphoblastic leukemia cell lines.
Replacement of the thiomethyl (SSA) with the polar methylsulfinyl in 79 decreases lipophilicity of this analog.
A study examining correlations between physicochemical properties and toxicological outcomes has shown
that topological polar surface area (TPSA) and cLogP properties correlate with in vivo toxicity in a set of 245
preclinical Pfizer compounds for which animal toxicity and exposure data has been collected.32 The ‘least toxic’
group was reported to have TPSA > 75 Å2 and cLogP < 3. We computed physicochemical properties relevant to

lipophilicity for selected representative analogs in the most potent series, including LogP, LogD, TPSA and
lipophilic efficiency (LipE), as shown in Table A.1, Appendix A. While listed compounds have LogP > 3 and
show only modest variations compared to lead 1 (LogP 5.03), TPSA values are higher for most analogs relative to
SSA which may be considered an improvement compared to the parent compound. Analogs 32-34, 61 and 80
have the highest (close to 75 Å2) TPSA values compared to lead 1 (TPSA 57.64 Å2). Compounds 40, 41, 44, 61,
62 have LogD values less than 3, compared to 4.41 of lead 1.
Lipophilic efficiency (LipE) relates potency to lipophilicity and may be used to assess and optimize binding
efficiency in the context of lipophilicity. LipE listed in Table A.1 (Appendix A) was calculated as LipE = pCC50
(HT29) – logD (at pH 7.4). While the optimal target range for LipE is considered between 5 and 7, these analogs
show a trend of slight improvement compared to SSA (LipE 1.78), compounds 41, 61 and 62 possessing LipE >
3. It is important to note that the CC50 values utilized for computing LipE results for these compounds were
determined using whole cell screening assays. Generally, we would expect more potent in vitro dose-response
activities against a specific protein target compared to whole cell activities which then would reflect in calculated
LipE values. Thus, it is not surprising that LipE values obtained for these compounds are lower than the accepted
optimal norms assessed in enzymatic, target-based assays.
A number of analogs in the reported series were designed to improve metabolic stability, based on metabolic
site liability predictions using the CYP3A4 metabolism model implemented in the StarDrop software. For
example, introducing a benzothiazolyl group as in 32, 61 is predicted to remove metabolically labile sites
associated with the 4-methylthiophenyl group in SSA (as shown in Figures A.2, A.4, compared to SSA in Figure
A.1, Appendix A). In addition, metabolic vulnerability may be also reduced by cyclization of the amine as
exemplified for compounds 40 and 61 (Figures A.3 and A.4 in Appendix A, respectively).
In summary, we report herein an extensive SAR study of sulindac amide analogs that show significant
anticancer activity in HT29, PC3 and MDA-MB-231 proliferation assays. Our primary goal in this work was to
expand diversity around positions of the basic indene structure for probing anticancer activities of diverse amide
analogs of sulindac sulfide amide (SSA). The most potent compound identified, compound 79, shows significant
activity and potency compared to the lead compound SSA (1). This analog also has potent growth inhibitory
activity against the five lymphoblastic lecukemia cell lines screened, against which SSA is inactive. Compound
79 with a benzyl group at the acetamide linker and a 4-methylsulfinylbenzylidene group at the C-1 position is a
neutral analog lacking the basic amine moiety of SSA. Introducing various heterocycles or substituents in place of
the 4-methylthiobenzylidine and the dimethylethylamine in lead compound 1 resulted in analog series that
maintain growth inhibition at low micromolar potency in all three cancer cell lines suggesting that there is fair
tolerance for alterations in these regions while maintaining good anticancer activity. Screening against the panel
of six additional cancer cell lines identified five new analogs, compounds 5, 33, 35, 50, 62, that also show highly

potent growth inhibitory activity against acute lymphoblastic leukemia cell line(s) in which the parent compound
1 was inactive. While possessing comparable anticancer activity to lead 1, a number of analogs within the
presented series show improved computed physicochemical properties compared to compound 1, namely,
improvement of lipophilic efficiency (LipE) and/or topological polar surface area (TPSA) and/or LogD properties.
Among potent analogs that show improvement of physicochemical properties are compounds 32-34 with a
benzothiazol-2-ylmethylene at the C-1 position and acyclic basic amide linkers, the series 40, 41, 44 containing
various basic cyclic amide groups, analogs 61, 62 combining benzothiazol-2-ylmethylene at the C-1 position with
basic cyclic amides at C-3, while compounds 79, 80 containing a benzyl and furanylmethyl amide group are
neutral analogs. Analogs screened in a BJ cell cytotoxicity assay showed no toxicity against this fibroblast cell
line. Our original design parameters were based on generation of structural diversity and ease of synthesis while,
for the most part, maintaining a basic group near the carboxamide generated during coupling. Our presumption,
based on earlier modeling and testing of our lead agent SSA, has been that the basic amine moiety replacing
carboxylate in sulindac would attenuate or abrogate COX binding due to charge-charge interactions of the
protonated amine with positively charged amino acid side chains in the COX 1-2 active sites.13,18 Whether
compound 79, a neutral analog of sulindac lacking a basic amine, has any COX activity is presently unknown.
This hypothesis and the mechanism of action for the new compounds presented herein will require experimental
confirmation, and such experiments will be the focus of future work. Selected active compounds may also serve
as potential candidates for formulation and in vivo assessment for bioavailability and efficacy.
Acknowledgements
This work was supported by grants from the National Institutes of Health NCI 1R01CA131378 (RCR PI)
and DOD Era of Hope award W81XWH-07-1-0463 (RCR PI). We thank Lucile White, Lynn Rasmussen and
Melinda Ingram for HTS support services. We would also like to acknowledge contributions by the High
Throughput Biology Center in the Department of Chemical Biology and Therapeutics at St. Jude Children’s
Research Hospital in Memphis, TN for BJ cell line data and additional anticancer cell line screens. The work at
St. Jude was supported by American Lebanese Syrian Associated Charities (ALSAC).
Supplementary material
Appendix A. Biological studies: experimental methods, computed physicochemical properties and
metabolic stability predictions.
Appendix B. Synthetic experimental methods and analytical data.
References

1.
2.
3.
Chan TA Nonsteroidal anti-inflammatory drugs, apoptosis, and colon-cancer chemoprevention. Lancet
Oncol. 2002; 3: 166-174.
Cannon CP, Cannon PJ Physiology. COX-2 inhibitors and cardiovascular risk. Science. 2012; 336: 1386-
1387.
Yu Y, Ricciotti E, Scalia R, Tang SY, Grant G, et al. Vascular COX-2 modulates blood pressure and
thrombosis in mice. Sci Transl Med. 2012; 4: 132ra154.
4.
5.
Vane JR, Botting RM Mechanism of action of antiinflammatory drugs. Int J Tissue React. 1998; 20: 3-15.
Vane JR, Bakhle YS, Botting RM Cyclooxygenases 1 and 2. Annu Rev Pharmacol Toxicol. 1998; 38: 97-
120.
6.
Mukherjee D Selective cyclooxygenase-2 (COX-2) inhibitors and potential risk of cardiovascular events.
Biochem Pharmacol. 2002; 63: 817-821.
7.
Marx J Cancer research. Anti-inflammatories inhibit cancer growth--but how? Science. 2001; 291: 581-
582.
8.
Thun MJ, Henley SJ, Patrono C Nonsteroidal anti-inflammatory drugs as anticancer agents: mechanistic,
pharmacologic, and clinical issues. J Natl Cancer Inst. 2002; 94: 252-266.
Brown JR, DuBois RN COX-2: a molecular target for colorectal cancer prevention. J Clin Oncol. 2005;
23: 2840-2855.
9.
10.
Eberhart CE, Coffey RJ, Radhika A, Giardiello FM, Ferrenbach S, et al. Up-regulation of cyclooxygenase
2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology. 1994; 107:
1183-1188.
11.
12.
Husain SS, Szabo IL, Tamawski AS NSAID inhibition of GI cancer growth: clinical implications and
molecular mechanisms of action. Am J Gastroenterol. 2002; 97: 542-553.
Alberts DS, Hixson L, Ahnen D, Bogert C, Einspahr J, et al. Do NSAIDs exert their colon cancer
chemoprevention activities through the inhibition of mucosal prostaglandin synthetase? J Cell Biochem.
1995; Suppl 22: 18-23.
13.
14.
Piazza GA, Keeton AB, Tinsley HN, Whitt JD, Gary BD, et al. NSAIDs: old drugs reveal new anticancer
targets. Pharmaceuticals. 2010; 3: 1652-1667.
Hanif R, Pittas A, Feng Y, Koutsos MI, Qiao L, et al. Effects of nonsteroidal anti-inflammatory drugs on
proliferation and on induction of apoptosis in colon cancer cells by a prostaglandin-independent pathway.
Biochemical Pharmacology. 1996; 52: 237-245.
15.
Elder DJ, Halton DE, Hague A, Paraskeva C Induction of apoptotic cell death in human colorectal
carcinoma cell lines by a cyclooxygenase-2 (COX-2)-selective nonsteroidal anti-inflammatory drug:
independence from COX-2 protein expression. Clin Cancer Res. 1997; 3: 1679-1683.

16.
17.
18.
Liggett JL, Zhang X, Eling TE, Baek SJ Anti-tumor activity of non-steroidal anti-inflammatory drugs:
Cyclooxygenase-independent targets. Cancer Letters (NY, US). 2014; 346: 217-224.
Gurpinar, E, Grizzle, WE, Piazza, GA NSAIDs Inhibit Tumorigenesis, but How? Clinical Cancer
Research. 2013; 20: 1104-1113.
Piazza GA, Keeton AB, Tinsley HN, Gary BD, Whitt JD, et al. A novel sulindac derivative that does not
inhibit cyclooxygenases but potently inhibits colon tumor cell growth and induces apoptosis with
antitumor activity. Cancer Prev Res (Phila). 2009; 2: 572-580.
19.
Dourtoglou V, Gross B, Lambropoulou V, Ziodrou C O-Benzotriazolyl-N,N,N',N'-tetramethyluronium
hexafluorophosphate as coupling reagent for the synthesis of peptides of biological interest. Synthesis.
1984; 7: 572-574.
20.
21.
22.
Carpino LA 1-Hydroxy-7-Azabenzotriazole - an Efficient Peptide Coupling Additive. Journal of the
American Chemical Society. 1993; 115: 4397-4398.
Sperl G, Gross P, Brendel K, Pamukcu R, Piazza G Indenyl hydroxamic acids, (hydroxy) ureas and
urethanes for treating patients with precancerous lesions US Patent US6300346 B1, 2000.
Kim DK, Lee N, Kim YW, Chang K, Im GJ, et al. Synthesis and evaluation of amino acid esters of 6-
deoxypenciclovir as potential prodrugs of penciclovir. Bioorg Med Chem. 1999; 7: 419-424.
23.
24.
Han HK, Oh DM, Amidon GL Cellular uptake mechanism of amino acid ester prodrugs in Caco-
2/hPEPT1 cells overexpressing a human peptide transporter. Pharm Res. 1998; 15: 1382-1386.
Abbate F, Casini A, Owa T, Scozzafava A, Supuran CT Carbonic anhydrase inhibitors: E7070, a
sulfonamide anticancer agent, potently inhibits cytosolic isozymes I and II, and transmembrane, tumor-
associated isozyme IX. Bioorg Med Chem Lett. 2004; 14: 217-223.
25.
Rostom SA Synthesis and in vitro antitumor evaluation of some indeno[1,2-c]pyrazol(in)es substituted
with sulfonamide, sulfonylurea(-thiourea) pharmacophores, and some derived thiazole ring systems.
Bioorg Med Chem. 2006; 14: 6475-6485.
26.
Lee SK, Park SM, Im C Cytotoxicities and Quantitative Structure Activity Relationships of B13
Sulfonamides in HT-29 and A549 Cells. Korean J Physiol Pharmacol. 2011; 15: 423-429.
27.
Chennamaneni S, Zhong B, Lama R, Su B COX inhibitors Indomethacin and Sulindac derivatives as
antiproliferative agents: Synthesis, biological evaluation, and mechanism investigation. Eur J Med Chem.
2012; 56: 17-29.
28.
Waring MJ Defining optimum lipophilicity and molecular weight ranges for drug candidates-Molecular
weight dependent lower logD limits based on permeability. Bioorg Med Chem Lett. 2009; 19: 2844-2851.
29.
Johnson TW, Dress KR, Edwards M Using the Golden Triangle to optimize clearance and oral
absorption. Bioorg Med Chem Lett. 2009; 19: 5560-5564.

30.
31.
Meanwell NA Improving drug candidates by design: a focus on physicochemical properties as a means of
improving compound disposition and safety. Chem Res Toxicol. 2011; 24: 1420-1456.
Mathew B, Hobrath JV, Ross L, Connelly MC, Lofton H, Rajagopalan M, Guy RK, Reynolds RC
Screening and development of new inhibitors of FtsZ from M. tuberculosis. PLoS ONE. 2016; 11(10):
e0164100.
32.
Hughes JD, Blagg J, Price DA, Bailey S, Decrescenzo GA, et al. Physiochemical drug properties
associated with in vivo toxicological outcomes. Bioorg Med Chem Lett. 2008; 18: 4872-4875.