Targeting Heterogeneous Tumors Using a Multifunctional Molecular Prodrug

Article abstract of DOI:10.1021/jacs.9b07171

Reported here is a molecular construct (K1) designed to overcome hurdles associated with delivering active drugs to heterogeneous tumor environments. Construct K1 relies on two cancer environment triggers (GSH and H₂O₂) to induce pro

Full text of DOI:10.1021/jacs.9b07171

Subscriber access provided by the University of Exeter
Article
Targeting Heterogeneous Tumors Using a Multifunctional Molecular Prodrug
Amit Sharma, Min-Goo Lee, Miae Won, Seyoung Koo, Jonathan F.
Arambula, Jonathan L. Sessler, Sung-Gil Chi, and Jong Seung Kim
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b07171 • Publication Date (Web): 11 Sep 2019
Downloaded from pubs.acs.org on September 11, 2019
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 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 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.
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 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Targeting Heterogeneous Tumors Using a Multifunctional Molecular
Prodrug
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
Amit Sharma,†,‡ Min-Goo Lee,§,‡ Miae Won,† Seyoung Koo,† Jonathan F. Arambula,Ψ Jonathan L. Sess-
ler*,⊥, Sung-Gil Chi,*,§ and Jong Seung Kim*,†
Ψ
† Department of Chemistry, Korea University, Seoul 02841 Korea.
§ Department of Life Sciences, Korea University, Seoul 02841, Korea.
⊥ Institute for Supramolecular Chemistry and Catalysis, Shanghai University, Shanghai 200444, China.
Y Department of Chemistry, University of Texas at Austin, Austin, TX 78712-1224, USA.
‡ These authors contributed equally.
ABSTRACT: Reported here is a molecular construct (K1) designed to overcome hurdles associated with delivering active drugs to
heterogeneous tumor environments. Construct K1 relies on two cancer environment triggers (GSH & H2O2) to induce prodrug acti-
vation. It releases an active drug form (SN-38) under conditions of both oxidative and reductive stress in vitro. Specific uptake of K1
in COX-2 positive aggressive colon cancer cells (SW620 and LoVo) was seen, along with enhanced anticancer activity compared
with the control agent SN-38. These findings are attributed to environmentally triggered drug release, as well as simultaneous scav-
enging of species giving rise to intracellular redox stress. K1 serves to downregulate various cancer survival signaling pathways
(AKT, p38, IL-6, VEGF, and TNF-a) and upregulate an anti-inflammatory response (IL-10). Compared with SN-38 and DMSO as
controls, K1 also displayed an improved in vivo therapeutic efficacy in a xenograft tumor regrowth model with no noticeable system-
atic toxicity at the administrated dose. We believe that the strategy described here presents an attractive approach to addressing solid
tumors characterized by intratumoral heterogeneity.
date most carrier-drug conjugates targeting cancer have ex-
■ INTRODUCTION
ploited stable but conditionally labile linkages that are sensitive
to a single biological input, such as subphysiological pH, tem-
perature, protease, glutathione, or hydrogen peroxide.7,8 How-
ever, the biomarkers in question are rarely unique to cancerous
sites, resulting in suboptimal selectivity and reduced overall
therapeutic efficacy.9 The concurrent use of different biological
triggers may allow these limitations to be overcome.10 How-
ever, care must be taken that the therapeutic agent being subject
to masking/demasking for delivery is released ultimately in its
unmodified active form so as to avoid challenges from the reg-
ulatory bodies.11,12
Colorectal cancer (CRC) is the third most frequently diag-
nosed cancer in the United States. Its global burden is expected
to increase by 60% through 2030 in spite of large-scale initial
screening efforts.13,14 Complex pro-tumor inflammatory cas-
cades and immune-mediated processes play roles in both CRC
initiation and metastasis.15 Irinotecan (also referred to as CPT)
is a front-line clinically approved chemotherapy drug for colon
cancer. It is converted to the active metabolite SN-38 by car-
boxylesterase-mediated hydrolysis of the bispiperidine moiety
(present as a solubilizing functional group). This active form
induces toxicity in cancer cells by forming a topoisomerase
Owing to the high cost, long development times, and limited
success rates associated with the development of new cancer
therapeutics, alternative strategies involving modification of
anticancer drugs approved by US Food and Drug Administra-
tion (FDA) are attractive.1 Such modification efforts could min-
imize the recognized side-effects of the anticancer agents in
question or improve their clinical efficacy. Ultimately, this
could translate into improved outcomes, including prolonged
cancer patient survival.2 Particularly effective may be modifi-
cation strategies that enhance the delivery of chemotherapeutic
drugs to a target site of interest. To date, remarkable advances
in understanding the intricate pathways responsible for malig-
nant progression, as well a greater appreciation of cancer devel-
opment, have translated into a number of innovative strategies
for targeting cytotoxins to specific sites of interest.1,3,4 Typi-
cally, this is done in a phenotype-specific fashion by conjugat-
ing a drug or prodrug to a carrier that takes advantage of dis-
ease-related biomarkers to effect targeting and release. This can
minimize adverse side-effects, reduce required dose levels or
evade some multidrug resistance mechanisms, such as mem-
brane-associated p-glycoprotein mediated drug efflux.5,6 To
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Design strategy underlying the proposed inflammation guided redox-responsive cancer prodrugs K1 and K2 that are the subject
of the present study.
inhibitor-DNA complex, thereby affecting DNA functions.
However, owing to its slow hydrolysis rate, CPT displays a 100-
to 1000- fold lower cell-based toxicity than SN-38 with a cor-
respondingly reduced level of therapeutic efficacy in humans.16
Resistance to apoptotic cell death mechanisms has also been put
forward as a reason for poor clinical outcomes in the case of
CPT-based therapies.17 It is our belief that these clinical defi-
ciencies might be overcome in part by more effective targeting
of the active SN-38 drug form to tumors.
One targeting strategy that is appealing involves the use of a
COX-2 inhibitor. COX-2 is a key enzyme involved in the syn-
thesis of tumor cell-derived prostaglandins (PGs). PGs and
COX-2 serve both as pro-inflammatory mediators and immune
suppressors of anticancer immunity.18,19 Increased expression of
COX-2 mRNA and PGs has been reported in various cancers,
including breast, colorectal, lung, stomach and pancreatic can-
cers with these levels being correlated with poor survival out-
comes in CRC patients.20 Recently, COX-2 inhibitors have
shown promise as antiangiogenetic therapies, either alone or in
combination with anticancer drugs; they have also been seen to
inhibit VEGF-independent PG-induced tumor angiogenesis in
preclinical models.21,22 We, and others believe that COX-2 in-
hibitors could play a useful role as targeting agents, particularly
in the case of heterogeneous tumors.23-26 The present study was
designed to test further this supposition and, in doing so, ad-
dress through chemical means the problem of tumor heteroge-
neity.
be present in the tumor microenvironment. This allows active
delivery of an FDA approved drug, SN-38. Construct K1 also
incorporates indomethacin, a COX-2 inhibitor that has been uti-
lized successfully to achieve tumor targeting and knockdown of
inflammatory regulated immune-suppressive genes. The molec-
ular combination embodied in K1 (i.e., SN-38 and indometha-
cin) was thus expected to be advantageous in addressing the
problem of aggressive CRC where tumor heterogeneity, COX-
2 related inflammation, and effective cancer-selective prodrug
activation constitute recognized hurdles. Since K1 incorporates
multiple potentially useful subunits into a single construct, it
might also offer benefits in terms of reduced off-target toxicity
and control over drug pharmacokinetics and biodistribution ef-
fects. The present study was designed to test these hypotheses.
■ RESULTS AND DISCUSSION
The motivation underlying the present study was a desire to de-
velop a prodrug conjugate that would allow for i) cancer-selec-
tive drug delivery in vivo and ii) active agent release through
more than one mechanism of action. Conjugate K1 was de-
signed with such objectives in mind. This system contains both
an NSAID COX-2 inhibitor and an SN-38 active payload de-
signed to effect tumor targeting and growth inhibition, respec-
tively. It also contains a specific thioether that was expected to
undergo cleavage when exposed to either GSH or H2O2.34
The synthesis of prodrug K1 is outlined in Scheme S1 (Sup-
porting Information, SI). First, intermediate 3 was synthesized
via a Michael addition reaction between 2-mercaptopropionic
acid and tert-butyl acrylate. Intermediate 3 was coupled with
the topoisomerase I inhibitor, 7-ethyl-10-hydroxyl-camptothe-
cin (SN-38, active metabolite of irinotecan) through esterifica-
tion using EDC/DMAP (EDC = 1-ethyl-3-(3-dimethyla-
minopropyl)carbodiimide; DMAP, 4-dimethylamino-pyridine)
to furnish intermediate K2 (cf. Scheme 1). Subsequent tert-bu-
tyl deprotection with TFA (trifluoroacetic acid) in methylene
chloride and subsequent acid-amine coupling of intermediates
4 and 2 resulted in the formation of conjugate K1 in good yield.
All the intermediates and products were characterized by stand-
ard analytical means (cf. Supporting Information).
Cancer cells are remarkably heterogeneous.27,28 As compared
to normal cells, many cancer cells exist in a strongly reducing
environment due to elevated levels of intracellular glutathione
(GSH). This feature is often exacerbated under conditions of
chemotherapy.29 In seeming contradiction, a number of cancers
are characterized by oxidative stress due to the overproduction
of various reactive oxygen species (ROS).30 Elevated reductive
and oxidative stress may exist in different tumors (inter-tumoral)
and even co-exist at different progression stages within the
same tumor (intra-tumoral) or at the sub-organelle level within
the same cancer cell.31 This heterogeneity can be manifest in
terms of intra-tumoral redox potential differences. A few redox-
based nano-strategies have been reported in an effort to address
the problems associated with targeting heterogeneous tumor en-
vironments.32,33 However, to our knowledge, a combined ap-
proach involving selective tumor targeting followed by activa-
tion of a therapeutic drug by specific molecular mechanisms has
yet to be reported.
Here we detail the synthesis and study of a new molecularly
targeted system (K1, Scheme 1) that is subject to activation us-
ing either reducing (GSH) or oxidative (H2O2) inputs likely to
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Proposed activation modes for prodrug K1 expected to be operative under conditions of reductive and oxidative stress. (A) Under
reductive stress, glutathione (GSH) undergoes thiolysis at the SN-38 linked ester functionality to form an intermediate with simultaneous
release of active drug SN-38. Under oxidative stress, the sulfide functionality in the chemical linker is oxidized to a sulfone/sulfoxide moiety
resulting in conversion of a hydrophobic site to a hydrophilic one that upon hydrolysis releases the active drug SN-38. Time-dependent
fluorescence intensity change corresponding to SN-38 release upon incubation of K1 (10 µM) with (B) GSH (5 mM, 30 min) and (C) H2O2
(0.125 mM, 10 min) in phosphate saline buffered (37 °C, λex = 365 nm, Slit width 3/3). High-performance liquid chromatogram of K1 (10
µM) recorded at different time intervals following treatment with (D) GSH (5 mM), (E) H2O2 (0.125 mM) and SN-38 in PBS (37 °C).
To investigate the drug release behavior of K1 under model
reductive (glutathione, GSH) and oxidative (hydrogen peroxide,
H2O2) stress conditions, UV-Vis, fluorescence and high-perfor-
mance liquid chromatography (HPLC) studies were carried out
the -SH group of GSH) of the phenolic ester moiety to release
free SN-38. Unlike other GSH responsive prodrugs, K1 allows
for direct release of an active drug form as the result of the con-
comitant consumption of GSH.7,36 Separately, incubation of K1
with H2O2 (0.125 mM, 4 h) results in oxidation of the thioether
moiety to the corresponding sulfone or sulfoxide. These latter
putative intermediates are relatively hydrophilic, which is
thought to abet hydrolysis. This proposed hydrolysis cleaves the
phenolic ester present in K1 and releases SN-38 in its free (ac-
tive) form. A combination of UV-Vis, fluorescence, HPLC
(Figure 1C & 1D) and ESI-MS analyses (Figures S4 S5) pro-
vide support for the proposed mechanism of K1 activation
shown in Figure 1A. The activation behavior of K2 was also
investigated (Figure S6). Finally, the stability of prodrugs K1
and K2 stability in human serum was examined. Briefly, we
found that K1 is essentially stable up to 24 h when incubated in
at 37 ºC in PBS in the presence of 30% human serum, whereas
K2 undergoes substantially hydrolysis to SN-38 over the same
24 h period when exposed to 10% human serum (Figure S7).
Accordingly, the focus of the present study was on K1 with K2
being used as a control.
As noted above, K1 is comprised of SN-38 (topoisomerase I
inhibitor) and a indomethacin (COX-2 inhibitor) subunit that is
proposed to provide targeting through interactions with COX-2
proteins upregulated in many tumor types. To assess this, we
performed cell viability assays using human cell lines with var-
iable expression of COX-2 protein so as to determine whether
K1 induces COX-2-dependent cytotoxic activity. In cells with
low COX-2 expression (i.e., Caco-2, DLD-1, Figure S8), treat-
ment with K1 or the simpler control system K2 resulted in a
slight decrease in proliferation relative to LoVo and SW620
in phosphate-buffered saline (PBS, pH = 7.4) at 37 °C. K1
proved largely stable in PBS until exposed to either H2O2 or
GSH as confirmed by fluorescence and HPLC analyses (Figure
1 & S1-3). An enhancement in the intensity of the UV-Vis ab-
sorbance band of K1 at 365 nm was seen upon incubation with
GSH (5 mM) and H2O2 (0.125 mM) (Figure S2). Further, as
shown in Figure 1B & 1C, K1 gives rise to a weak emission
band at 540 nm (λex = 365 nm). When tested under model re-
ductive stress prodrug activation conditions mimicking those
expected under physiological conditions,29,30 i.e., incubation of
K1 with GSH (5 mM), an enhancement in the fluorescence in-
tensity at 540 nm was observed. The fluorescence band at 540
nm is characteristic of free SN-38. Incubation with H2O2 at lev-
els deemed physiologically relevant29,35 produced a similar en-
hancement in the overall fluorescence intensity (Figure 1C). We
also tested the activation behavior of K1 by monitoring the
change in the fluorescence emission features upon exposure to
different concentrations of GSH and H2O2 (Figure S3).
Additional evidence for K1 activation upon exposure to GSH
and H2O2 came from the combined high-performance liquid
chromatography (HPLC) (Figure 1D&E) and electronic spray
ionization mass spectrometry (ESI-MS) studies (Figures S4 and
S5). K1 is characterized by a retention peak at 24.5 min in the
HPLC chromatogram under our conditions of analysis; how-
ever, incubation with GSH (5 mM) for 10 h, leads to formation
of new peaks with retention times of 8.0 min (active drug SN-
38) and 16 min (byproduct), respectively. These results are con-
sistent with K1 undergoing thiolysis (presumably mediated by
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Effect of K1 and K2 on tumor cell viability and correlations with the expression status of COX-2. (A) Effect of K1 and K2 on
cancer cell cytotoxicity. Cells were treated with K1 and K2 (0, 1, 5 and 10 nM) for 48 h. Cell viability was assessed by means of a WST-1
assay (mean ± SD, n = 3, * * p < 0.01, * p < 0.05, n.s., not significant). (B) Effect of K1 and K2 on cancer cell growth. Cells were treated
with 1 nM K1 or 1 nM K2 for 5 days and were stained with crystal violet solution. (C) Effect of K1 on cell viability in human normal and
cancer cells. Cells were treated with the indicated doses of K1 for 48 h. Cell viability was determined using a WST-1 assay (mean ± SD, n
= 3, * p < 0.05). (D & E) Comparative analysis of the extent of intracellular uptake of K1 in human normal and cancer cells. Cells were
treated for the indicated times and with the indicated doses of K1. Fluorescence intensities were determined using SpectraMAX I3.
cells characterized by increased COX-2 expression levels (Fig-
ure 2A & 2B).37
drug release from K1 and that this would occur more effectively
in the cancer cells due to differences in their concentrations rel-
ative to normal cells (Figure 1B & 1C). We thus tested whether
K1 could release the active drug in cancer cells under intracel-
lular redox condition. For this study, LoVo cells were pretreated
with N-acetylcysteine (NAC, a GSH precursor) and H2O2 to in-
duce changes in the cellular redox environment. The cells were
then treated with K1 and the release of SN-38 was monitored
by monitoring the fluorescence emission. As expected, pretreat-
ment with both H2O2 and NAC increased significantly the fluo-
rescence intensity, as expected for the release of SN-38 within
these K1-treated cells (Figure 3A).
We further examined whether alterations in intracellular redox
status would influence cell growth and the viability of cancer
cells treated with K1. In the absence of redox stress, the effect
of K1 on cancer cell cytotoxicity was similar to that of SN-38.
However, in NAC-pretreated cells, K1 reduced cancer cell
growth significantly as compared to treatment with SN-38 (Fig-
ures 3B, S10A). Similar results were seen in GSH (cellular an-
tioxidant)-siRNA knock-downed cells in which treatment with
K1 reduced significantly cancer cell growth and cytotoxicity
relative to SN-38 (Figure 3C, and Figure S10A). Similar results
To assess further the COX-2 selectivity and anticancer poten-
tial of K1, additional cell viability assays were performed.
These involved comparing COX-2 positive cancer cells (i.e.,
LoVo, SW620) to healthy human cell lines NHDFs (normal hu-
man dermal fibroblasts) and MCF10A (human breast epithelial
cell line) that have low COX-2 expression levels (Figure 2C,
Figure S8). A positive correlation between the in vitro potency
of K1 and the COX-2 expression levels was seen. In addition,
knockdown of endogenous COX-2 expression by means of
siRNA COX-2 transfection in SW620 cells resulted in a reduc-
tion in the K1-induced cytotoxicity (Figure S9).
To assess whether K1 displayed relatively enhanced cancer
cell targeting, we examined the intracellular accumulation of
K1 in cancer cells (SW620, LoVo) and normal cells (NHDF &
MCF10A). Treatment with K1 gave rise to both a dose- and
time-dependent induction in the cancer cells possessing higher
COX-2 expression but not in the normal cells or cancer cells
characterized by a low level of COX-2 expression (Figure 2D
& 2E). These results provide support for the notion that the ob-
served cytotoxic effects, as well as the fluorescence induction
seen with K1, were highly dependent on COX-2 protein expres-
sion levels.
were obtained for K1 in 3D-tumor spheroid models (LoVo cells)
(Figure S11).
Previous studies have reported that high levels of antioxi-
dants or free radical species alter the tumor redox status, result-
ing in drug resistance during cancer treatment.38 The finding
that treatment with H2O2 and GSH under model conditions ef-
fected drug release from K1 led us to infer that under in vitro
conditions these or related redox-active species would induce
p38 Mitogen-activated protein kinase (MAPK) and protein ki-
nase B (AKT) signaling are key pathways regulating cancer cell
proliferation and apoptosis in response to redox status.39 To un-
derstand the role that K1 presumably has on cancer cell growth
and survival under redox conditions considered relevant to
those present in many cancerous locales, GSH knock down
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
SN-38 and analyzed via western blot (Figure S10B). As shown
1
2
3
4
5
6
7
8
in Figure 3D, treatment of control cells with K1 or SN-38 re-
sulted in a similar induction of cleavage of poly (ADP-ribose)
polymerase (PARP) and caspase-3, and reduced phosphoryla-
tion of both p38 and AKT. In contrast, after GSH knockdown,
K1 treatment enhanced the induction of PARP and caspase-3
cleavage and strongly reduced phosphorylation of AKT and p38,
whereas cells treated with SN-38 were no longer responsive and
thus considered resistant to SN-38. These results provide sup-
port for the design expectation that K1 would produce an anti-
cancer therapeutic response under conditions of both oxidative
and reductive stress, with simultaneous reduction of intracellu-
lar oxidants/antioxidant levels as outlined in Figures 1A and 3E.
We also compared the cytotoxicity of K1 with other controls
(DMSO, SN-38 only, indomethacin only, SN-38 + indometha-
cin) in SW620 cancer cells (Figure S12). It was found that K1
exhibited improved toxicity relative other tested controls. Spe-
cifically, it proved 5-fold more potent than SN-38 alone and 3-
fold more active than the combination of SN-38 and indometh-
acin (Table S1).
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
To determine whether the anti-inflammatory effect from the
indomethacin moiety in K1 was mediated by COX-2 inhibition,
we evaluated the change in COX-2 activity by detecting PGE2
levels in K1-treated cancer cells having variable COX-2 expres-
sion by means of an ELISA assay.40 As shown in Figure 4A, K1
effects strong attenuation of PGE2 expression only in cancer
cells having high COX-2 expression levels (LoVo and SW620
cells) but not in cells characterized by low COX-2 expression
levels (Caco-2 and DLD1 Cells). We further observed that treat-
ment with K1, but not SN-38, effectively decreased the PGE2
level in SW620 cells in response to external lipopolysaccharide
(LPS) mediated stimulation (Figure 4B).
To escape from immune surveillance, cancer cells often coa-
lesce with cancer-associated immune cells and pro-inflamma-
tory cytokines, such as interleukin-6 (IL-6) and tumor necrosis
factor-a (TNF-a), which are induced by tumor-associated mac-
rophages.41 This inflammatory microenvironment provides
growth and survival advantages for cancer cells and enhances
tumor cell resistance towards chemotherapies through activa-
tion of the NF-κB signaling pathway.42 We thus sought to de-
termine whether K1 inhibited NF-κB activation and its target
gene expression in SW620 cancer cells and RAW264.7 macro-
phages. Semi-quantitative RT-PCR analyses revealed that K1
treatment suppressed the expression of NF-κB target genes,
such as IL-6, TNF-a, and vascular endothelial growth factor
(VEGF) in both a time- and dose-dependent manner (Figure
4C). In particular, K1 exerted a strong inhibitory effect on LPS
activation of these NF-κB target genes in both cancer cells and
macrophages (Figure 4D). Moreover, immunoblot analysis of a
major NF-κB component p65 and its regulator IκBα revealed
that K1 treatment inhibited the phosphorylation of both p65 and
IκBα, again in a time- and dose-associated manner (Figure 4E).
We also examined the effect of K1 on inflammatory cytokine
production using the enzyme-linked immunosorbent assay
(ELISA) in RAW264.7 macrophages. As shown in Figure 4F,
K1, but not SN-38, decreased in a statistically significant man-
ner the production of the pro-inflammatory cytokines IL-6 and
TNF-a while increasing production of the anti-inflammatory
cytokine, IL-10.
Figure 3. Effect of K1 upon intracellular oxidative or reductive
stress in LoVo cells. (A) Fluorescence intensity of K1 in response
to H2O2 (50 µM) and NAC (50 µM) pretreatment (mean ± SD, n =
3, * p < 0.01). (B) Effect of K1 and SN-38 on cancer cell growth.
LoVo cells were transfected with GSH siRNA or control
siRNA and treated with NAC (50 µM) and then treated with the
indicated dose of K1 and SN-38. Cells were stained with a solution
of crystal violet. (C) Effect of K1 and SN-38 on cancer cell cyto-
toxicity. LoVo cells were treated with K1 and SN-38 (0, 5 and 10
nM) for 48 h. Cell viability was assessed by means of a WST assay
(mean ± SD, n = 3, * p < 0.01). (D) Effect of K1 on cancer cell
survival and apoptosis upon oxidative stress. Control or GSH
knock-downed LoVo cells were treated with the indicated doses of
K1 and SN-38 for 48 h. The expression level of cancer survival
proteins (phospho AKT and phospho p38) were determined by
western blot. Cell apoptosis was measured by Western blot analysis
of cleaved PARP and cleaved caspase-3. Tubulin was used as a
protein loading control. (E) Schematic illustration of the cytotoxi-
city induced by K1 in response to redox stress.
studies were conducted in LoVo cells so as to mimic cellular
oxidative stress. These cells were then treated with either K1 or
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Anti-inflammatory effects of K1. (A) Comparison of PGE2 levels in COX-2 low- or high-expressing cancer cells upon K1 treat-
ment. The levels of secreted PGE2 were determined using an ELISA assay (mean ± SD, n = 3, * p < 0.01, n.s., not significant). (B) Effect
of K1 on LPS-induced PGE2 secretion in SW620 colon cancer cells. (C) Dose and time-associated effects of K1 on IL-6, TNF-a, and VEGF
mRNA expression levels in SW620 cells. (D) Effect of K1 on LPS-induced IL-6, TNF-a, and VEGF mRNA expression levels in SW620
cancer cells and RAW264.7 mouse macrophage cells. (E) Time and dose-dependent effects of K1 on the NF-kB signaling pathway. (F)
ELISA detection of IL-6, TNF-a, and IL-10 in RAW264.7 cells in response to SN-38, K1 or DMSO as vehicle control (mean ± SD of
triplicate assays). *p < 0.01 (Student’s t-test).
In order to evaluate the tumor target specificity of the drug
released from K1 in vivo, nude mice bearing SW620 colon can-
cer subcutaneous xenografts were subjected to in vivo fluores-
cence image analysis 3 hours after administration of the agent.
As shown in Figure 5A, the fluorescence signal of drug released
from K1 was observed in the xenograft tumor of K1-treated
mice, while the fluorescence images of the excised organs dis-
played no such fluorescence signal (Figure 5A). To correlate
the biological therapeutic effect with that of the in vitro cancer
cell-based studies discussed above, we further assessed the ef-
ficacy of K1 and SN-38 in nude mice bearing SW620 xeno-
grafts. Compared with DMSO, a mild response to SN-38 treat-
ment was seen (20% tumor reduction), while K1-treated tumors
displayed improved suppression of tumor growth (41% reduc-
tion) (Figure 5B-D, Figure S13). No significant weight loss was
observed during any course of treatment (Figure S14). Also, as-
partate transaminase (AST), alanine transaminase (ALT) and
serum creatine levels in the blood serum of mice treated with
K1 remained in the normal range over the course of treatment
(Figures S15 and S16). In contrast, test groups treated with SN-
38 showed higher serum levels of AST and ALT. Additionally,
we examined the effects of K1 on tumor survival cytokine gene
expression using xenograft tumor tissues. K1 induced a signif-
icant reduction in the TNF-a, IL-6, and VEGF gene expression
levels as compared to DMSO or SN-38 (Figure 5E). Taken in
concert, these findings provide support for the conclusion that
treatment with K1 provides for improved antitumor efficacy
relative to controls, as evidenced by both tumor regrowth anal-
yses and the observed selective suppression of cancer survival
cytokine expression.
In summary, we have developed K1, a molecular-based system
that targets cancers with COX-2 overexpression and which is
activated intracellularly by reductive and oxidative stress ex-
pected to exist in various cancerous microenvironment. This ac-
tivation serves to release the active drug SN-38. Redox medi-
ated activation of the prodrug not only allows for release of the
active drug SN-38, it also scavenges the corresponding biomol-
ecules (GSH, H2O2) so as to downregulate alternative cell sur-
vival signaling pathways. The indomethacin moiety incorpo-
rated into K1 serves to knockdown inflammatory responses me-
diated by COX-2 and is thought to contribute to the improved
efficacy seen for this agent relative to controls. Operationally,
K1 proved capable of inhibiting tumor growth to greater extent
than the FDA approved drug SN-38, with no significant side
effects as inferred from liver function assays, analyses of blood
markers. and body weight measurements. Detailed toxicologi-
cal analyses of K1 are ongoing and will be reported in due
course.
The current investigation provides support for the notion that
small, molecular-based drug delivery systems that have the po-
tential to release an active drug form under conditions of both
oxidative and reductive stress may lead to improvements in ef-
ficacy as determined from in vitro analyses and mouse model
studies. It thus sets the stage for the further development of tar-
geted therapies for use in addressing the heterogeneous tumor
microenvironments that make the development of improved
cancer therapeutics challenging.
■ CONCLUSIONS
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
*jongskim@korea.ac.kr (J.S.K.)
1
2
3
Author Contributions
4
5
‡
These authors contributed equally.
6
7
Notes
The authors declare no competing financial interest.
8
9
■ ACKNOWLEDGMENTS
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
This work was supported in part by a Creative Research Initiatives
project (grant no. 2018R1A3B1052702, J.S.K.) and a grant from
the National Research Foundation of Korea (NRF-
2018R1A2A1A05020236, S.-G.C.) and 2017R1D1A1B03030062,
M.W.). Funding from the National Cancer Institute (RO1 CA68682,
J.L.S. and R15 CA232765, J.F.A.) and the Robert A. Welch Foun-
dation (F-0018, J.L.S.) is also acknowledged.
■ REFERENCES
(1)
Hay, M.; Thomas, D. W.; Craighead, J. L.; Economides, C.;
Rosenthal, J. Clinical development success rates for investigational
drugs. Nat. Biotechnol. 2014, 32, 40-51.
(2)
Hanahan, D.; Weinberg, R. A. Hallmarks of cancer: the next
generation. Cell 2011, 144, 646-674.
(3)
Allen, T. M., Ligand-targeted therapeutics in anticancer
therapy. Nat. Rev. Cancer 2002, 2, 750-63.
(4)
Sharma, A.; Arambula, J. F.; Koo, S.; Kumar, R.; Singh, H.;
Sessler, J. L.; Kim, J. S. Hypoxia-targeted drug delivery. Chem. Soc.
Rev. 2019, 48, 771-813.
(5)
cer with diamonds. Sci. Transl. Med. 2011, 3, 73ps8.
(6) Sharma, A.; Lee, M. G.; Shi, H.; Won, M.; Arambula, J. F.;
Merkel, T. J.; DeSimone, J. M., Dodging drug-resistant can-
Sessler, J. L.; Lee, J. Y.; Chi, S. G.; Kim, J. S. Overcoming Drug Re-
sistance by Targeting Cancer Bioenergetics with an Activatable Pro-
drug. Chem-Us 2018, 4, 2370-2383.
(7)
Lee, M. H.; Sharma, A.; Chang, M. J.; Lee, J.; Son, S.; Sess-
ler, J. L.; Kang, C.; Kim, J. S. Fluorogenic reaction-based prodrug con-
jugates as targeted cancer theranostics. Chem. Soc. Rev. 2018, 47, 28-
52.
Figure 5. In vivo therapeutic efficacy of K1 in tumor xenografts.
(A) Representative fluorescence images of mice bearing SW620
tumors and ex vivo images of dissected organs. Li = liver, Lu =
lung, He = heart, Sp = spleen, Ki = kidney, Tu = tumor. (B) Repre-
sentative images of SW620 xenograft tumors treated with control
(DMSO), SN-38, and K1. (C) K1-induced regression of tumors (n
= 5 per group, mean ± SD, * p < 0.01, ** p < 0.05). (D) Tumor
inhibition levels determined from sacrificed mice following treat-
ment with DMSO, SN-38, or K1, * p <0.05 and ** p <0.01, n =
5/group. Tumor inhibition rate (%) = 100-(tumor weight of treat-
ment/tumor weight of DMSO)´100 (E) Anti-inflammatory cyto-
kine mRNA levels of the mice in the DMSO control, SN-38, and
K1 treated tissue (mean ± SD), * p < 0.05, ** p < 0.01).
(8)
Gnaim, S.; Scomparin, A.; Das, S.; Blau, R.; Satchi-Fainaro,
R.; Shabat, D. Direct real-time monitoring of prodrug activation by
chemiluminescence. Angew. Chem. Int. Ed. 2018, 57, 9033-9037.
(9)
McCawley, L. J.; Matrisian, L. M. Matrix metalloprotein-
ases: multifunctional contributors to tumor progression. Mol. Med. To-
day 2000, 6, 149-156.
(10)
Dong, Z.; Feng, L.; Chao, Y.; Hao, Y.; Chen, M.; Gong,
F.; Han, X.; Zhang, R.; Cheng, L.; Liu, Z. Amplification of Tumor Ox-
idative Stresses with Liposomal Fenton Catalyst and Glutathione In-
hibitor for Enhanced Cancer Chemotherapy and Radiotherapy. Nano
Lett., 2019, 19, 805-815.
(11)
regulation of products of nanotechnology. Science 2012, 336, 299-300.
(12) Zhang, Y.; Chan, H. F.; Leong, K. W. Advanced materials
Hamburg, M. A. Science and regulation. FDA's approach to
and processing for drug delivery: the past and the future. Adv. Drug
Deliv. Rev. 2013, 65, 104-120.
(13)
Siegel, R. L.; Miller, K. D.; Fedewa, S. A.; Ahnen, D. J.;
■ ASSOCIATED CONTENT
Meester, R. G. S.; Barzi, A.; Jemal, A. Colorectal cancer statistics, 2017.
CA: Cancer J. Clin. 2017, 67, 177-193.
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website. Detailed experiment procedures and support-
ing figures (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
(14)
van der Geest, L. G.; Lam-Boer, J.; Koopman, M.; Verhoef,
C.; Elferink, M. A.; de Wilt, J. H. Nationwide trends in incidence, treat-
ment and survival of colorectal cancer patients with synchronous me-
tastases. Clin. Exp. Metastasis 2015, 32, 457-465.
(15)
Markman, J. L.; Shiao, S. L. Impact of the immune system
and immunotherapy in colorectal cancer. J. Gastrointest. Oncol. 2015,
6, 208-223.
■ AUTHOR INFORMATION
(16)
Slatter, J. G.; Schaaf, L. J.; Sams, J. P.; Feenstra, K. L.; John-
Corresponding Author
*chi6302@korea.ac.kr (S.-G.C.)
*sessler@utexas.edu (J.L.S.)
son, M. G.; Bombardt, P. A.; Cathcart, K. S.; Verburg, M. T.; Pearson,
L. K.; Compton, L. D.; Miller, L. L.; Baker, D. S.; Pesheck, C. V.; Lord,
R. S. 3rd, Pharmacokinetics, metabolism, and excretion of irinotecan
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
(CPT-11) following I.V. infusion of [(14)C]CPT-11 in cancer patients.
Drug Metab. Dispos. 2000, 28, 423-433.
(29)
Traverso, N.; Ricciarelli, R.; Nitti, M.; Marengo, B.; Furfaro,
A. L.; Pronzato, M. A.; Marinari, U. M.; Domenicotti, C. Role of glu-
tathione in cancer progression and chemoresistance. Oxid. Med. Cell
Longev. 2013, 2013, 972913.
1
2
3
4
5
6
7
8
(17)
de Man, F. M.; Goey, A. K. L.; van Schaik, R. H. N.;
Mathijssen, R. H. J.; Bins, S. Individualization of Irinotecan Treatment:
A Review of Pharmacokinetics, Pharmacodynamics, and Pharmaco-
genetics. Clin. Pharmacokinet. 2018, 57, 1229-1254.
(30)
Szatrowski, T. P.; Nathan, C. F., Production of large
amounts of hydrogen peroxide by human tumor cells. Cancer Res. 1991,
51, 794-798.
(31)
Oxidative stress, inflammation, and cancer: how are they linked? Free
Radic. Biol. Med. 2010, 49, 1603-1616.
(32)
mers for drug delivery: from molecular design to applications. Polym.
Chem.-Uk 2014, 5, 1519-1528.
(33)
(18)
Nakanishi, M.; Rosenberg, D. W. Multifaceted roles of
PGE2 in inflammation and cancer. Semin. Immunopathol. 2013, 35,
123-137.
Reuter, S.; Gupta, S. C.; Chaturvedi, M. M.; Aggarwal, B. B.
(19)
Zelenay, S.; van der Veen, A. G.; Bottcher, J. P.; Snelgrove,
K. J.; Rogers, N.; Acton, S. E.; Chakravarty, P.; Girotti, M. R.; Marais,
R.; Quezada, S. A.; Sahai, E.; Sousa, C. R. E. Cyclooxygenase-Depend-
ent Tumor Growth through Evasion of Immunity. Cell 2015, 162,
1257-1270.
Huo, M.; Yuan, J.; Tao, L.; Wei, Y. Redox-responsive poly-
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
Lyu, Y.; Tian, J.; Li, J.; Chen, P.; Pu, K. Semiconducting
(20)
Ogino, S.; Kirkner, G. J.; Nosho, K.; Irahara, N.; Kure, S.;
polymer nanobiocatalysts for photoactivation of intracellular redox re-
actions. Angew. Chem. Int. Ed. 2018, 57, 13484-13488.
Shima, K.; Hazra, A.; Chan, A. T.; Dehari, R.; Giovannucci, E. L.;
Fuchs, C. S. Cyclooxygenase-2 Expression Is an Independent Predictor
of Poor Prognosis in Colon Cancer. Clin. Cancer Res. 2008, 14, 8221-
8227.
(34)
Napoli, A.; Valentini, M.; Tirelli, N.; Muller, M.; Hubbell, J.
A. Oxidation-responsive polymeric vesicles. Nat. Mater. 2004, 3, 183-
189.
(21)
Xu, L.; Stevens, J.; Hilton, M. B.; Seaman, S.; Conrads, T.
(35)
Yang, W.; Gao, X.; Wang, B. Boronic Acids: Biological and
P.; Veenstra, T. D.; Logsdon, D.; Morris, H.; Swing, D. A.; Patel, N.
L.; Kalen, J.; Haines, D. C.; Zudaire, E.; St Croix, B. COX-2 inhibition
potentiates antiangiogenic cancer therapy and prevents metastasis in
preclinical models. Sci. Transl. Med. 2014, 6, 242ra84.
Medical Applications of Boronic Acids; Hall, D. G., Ed.; Wiley-VCH
Verlag GmbH & Co. KGaA: Weinheim, 2005, p 481-512.
(36)
Austin, C. D.; Wen, X.; Gazzard, L.; Nelson, C.; Scheller, R.
H.; Scales, S. J. Oxidizing potential of endosomes and lysosomes limits
intracellular cleavage of disulfide-based antibody-drug conjugates.
Proc. Natl. Acad. Sci. U S A 2005, 102, 17987-17992.
(22)
Kim, H. S.; Sharma, A.; Ren, W. X.; Han, J.; Kim, J. S.
COX-2 Inhibition mediated anti-angiogenic activatable prodrug poten-
tiates cancer therapy in preclinical models. Biomaterials 2018, 185, 63-
72.
(23)
Peng, X. J., An Off-On COX-2-Specific Fluorescent Probe: Targeting
the Golgi Apparatus of Cancer Cells. J. Am. Chem. Soc. 2013, 135,
11663-11669.
(37)
Che, X. H.; Chen, C. L.; Ye, X. L.; Weng, G. B.; Guo, X. Z.;
Yu, W. Y.; Tao, J.; Chen, Y. C.; Chen, X., Dual inhibition of COX-2/5-
LOX blocks colon cancer proliferation, migration and invasion in vitro.
Oncol. Rep. 2016, 35, 1680-1688.
Zhang, H.; Fan, J. L.; Wang, J. Y.; Zhang, S. Z.; Dou, B. R.;
(38)
Liu, Y.; Li, Q.; Zhou, L.; Xie, N.; Nice, E. C.; Zhang, H.;
Huang, C.; Lei, Y., Cancer drug resistance: redox resetting renders a
way. Oncotarget 2016, 7, 42740-42761.
(24)
Zhang, H.; Fan, J.; Wang, J.; Dou, B.; Zhou, F.; Cao, J.; Qu,
J.; Cao, Z.; Zhao, W.; Peng, X., Fluorescence discrimination of cancer
from inflammation by molecular response to COX-2 enzymes. J. Am.
Chem. Soc. 2013, 135, 17469-17475.
(39)
Moldogazieva, N. T.; Lutsenko, S. V.; Terentiev, A. A. Re-
active Oxygen and Nitrogen Species-Induced Protein Modifications:
Implication in Carcinogenesis and Anticancer Therapy. Cancer Res.
2018, 78, 6040-6047.
(25)
Uddin, M. J.; Crews, B. C.; Ghebreselasie, K.; Marnett, L.
J., Design, synthesis, and structure-activity relationship studies of flu-
orescent inhibitors of cycloxygenase-2 as targeted optical imaging
agents. Bioconjug. Chem. 2013, 24, 712-723.
(40)
Cuendet, M.; Mesecar, A. D.; DeWitt, D. L.; Pezzuto, J. M.,
An ELISA method to measure inhibition of the COX enzymes. Nat.
Protoc. 2006, 1, 1915-1921.
(26)
Uddin, M. J.; Crews, B. C.; Ghebreselasie, K.; Daniel, C. K.;
(41)
Gonzalez, H.; Hagerling, C.; Werb, Z. Roles of the immune
Kingsley, P. J.; Xu, S.; Marnett, L. J., Targeted imaging of cancer by
fluorocoxib C, a near-infrared cyclooxygenase-2 probe. J. Biomed. Opt.
2015, 20, 50502.
(27)
314-322.
(28)
system in cancer: from tumor initiation to metastatic progression.
Genes Dev. 2018, 32, 1267-1284.
(42)
Ratnam, N. M.; Peterson, J. M.; Talbert, E. E.; Ladner, K. J.;
Visvader, J. E. Cells of origin in cancer. Nature 2011, 469,
Rajasekera, P. V.; Schmidt, C. R.; Dillhoff, M. E.; Swanson, B. J.; Ha-
verick, E.; Kladney, R. D.; Williams, T. M.; Leone, G. W.; Wang, D.
J.; Guttridge, D. C. NF-κB regulates GDF-15 to suppress macrophage
surveillance during early tumor development. J. Clin. Invest. 2017, 127,
3796-3809.
Marusyk, A.; Polyak, K. Tumor heterogeneity: causes and
consequences. Biochim. Biophys. Acta 2010, 1805, 105-117.
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
Table of Contents
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment