Profluorogenic reductase substrate for rapid, selective, and sensitive visualization and detection of human cancer cells that overexpress NQO1

Article abstract of DOI:10.1021/ja309346f

Achieving the vision of identifying and quantifying cancer-related events and targets for future personalized oncology is predicated on the existence of synthetically accessible and economically viable probe molecules fully able to report the presence of these events and targets in a rapid and highly selective and sensitive fashion. Delineated here are the design and evaluation of a newly synthesized turn-on probe whose intense fluorescent reporter signature is revealed only through probe activation by a specific intracellular enzyme present in tumor cells of multiple origins. Quenching of molecular probe fluorescence is achieved through unique photoinduced electron transfer between the naphthalimide dye reporter and a covalently attached, quinone-based enzyme substrate. Fluorescence of the reporter dye is turned on by rapid removal of the quinone quencher, an event that immediately occurs only after highly selective, two-electron reduction of the sterically and conformationally restricted quinone substrate by the cancer-associated human NAD(P)H:quinone oxidoreductase isozyme 1 (hNQO1). Successes of the approach include rapid differentiation of NQO1-expressing and -nonexpressing cancer cell lines via the unaided eye, flow cytometry, fluorescence imaging, and two-photon microscopy. The potential for use of the turn-on probe in longer-term cellular studies is indicated by its lack of influence on cell viability and its in vitro stability.

Full text of DOI:10.1021/ja309346f

Subscriber access provided by FORDHAM UNIVERSITY
Article
Pro-fluorogenic Reductase Substrate for Rapid, Selective, and Sensitive
Visualization and Detection of Human Cancer Cells that Overexpress NQO1
William C Silvers, Bijeta Prasai, David H Burk, Matthew L Brown, and Robin Lindsey McCarley
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja309346f • Publication Date (Web): 30 Nov 2012
Downloaded from http://pubs.acs.org on December 10, 2012
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 the American Chemical Society 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 7
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Pro-fluorogenic Reductase Substrate for Rapid, Selective,
and Sensitive Visualization and Detection of Human Cancer
Cells that Overexpress NQO1
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
William C. Silvers,† Bijeta Prasai,† David H. Burk,‡ Matthew L. Brown,∫ and Robin L. McCarley†,*
†Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803ꢀ1804
‡Pennington Biomedical Research Center, Baton Rouge, LA 70808ꢀ4124
∫Department of Biological Sciences, Louisiana State University, Baton Rouge, LA 70803ꢀ1804
ABSTRACT: Achieving the vision of identifying and quantifying cancerꢀrelated events and targets for future personalized oncology is
predicated on the existence of synthetically accessible and economically viable probe molecules fully able to report the presence of these
events and targets in a rapid, and highly selective and sensitive fashion. Delineated here are the design and evaluation of a newly syntheꢀ
sized turn-on probe whose intense fluorescent reporter signature is revealed only through probe activation by a specific intracellular enꢀ
zyme present in tumor cells of multiple origins. Quenching of molecular probe fluorescence is achieved through unique photoꢀinduced
electron transfer (PeT) between the naphthalimide dye reporter and a covalently attached, quinoneꢀbased enzyme substrate. Fluorescence
of the reporter dye is turned on by rapid removal of the quinone quencher, an event that immediately occurs only after highly selective,
twoꢀelectron reduction of the sterically and conformationally restricted quinone substrate by the cancerꢀassociated human
NAD(P)H:quinone oxidoreductase isozyme 1 (hNQO1). Successes of the approach include rapid differentiation of NQO1ꢀexpressing and
nonꢀexpressing cancer cell lines via the unaided eye, flow cytometry, fluorescence imaging, and twoꢀphoton microscopy. The potential for
use of the turnꢀon probe in longerꢀterm cellular studies is indicated by its lack of influence on cell viability and its in vitro stability.
pressors16 and is present in a diverse group of human tumor cells
■
INTRODUCTION
(e.g., pancreas, colon, breast, lung, liver, stomach, kidney,
head/neck, and ovaries) at levels 2ꢀ to 50ꢀfold greater than in
normal tissue.17 Furthermore, NQO1 content/activity in tumor
cells is strongly affected by cell life cycle and therapeutic apꢀ
proaches.18ꢀ20 Importantly, NQO1 is found in the cytosol and
catalyzes the strict twoꢀelectron reduction of quinones to hydroꢀ
quinones, making this flavoenzyme an ideal target for proꢀdrug
therapies21 and proꢀfluorophores.
Molecular probes whose fluorescent reporter signal is generatꢀ
ed (turnꢀon probes) by enzyme activation1 hold great potential for
identification, enumeration, and study of living cancer cells—
outcomes invaluable for accurate and early diagnoses and optimiꢀ
zation of surgical and personalized chemotherapeutic treatꢀ
ments.2,3 In particular, the successful development of enzymeꢀ
activatable probes that can yield rapid, and highly sensitive and
selective, reporting of species or events associated with cancer
cells4ꢀ6 will allow for definition of diseased and healthy tissue
borders during fluorescenceꢀassisted surgical resection of cancerꢀ
ous tissues and collection of realꢀtime information on tumor cell
microenvironment or the pharmacodynamic effect of drugs on
specific tumor cells.3,5ꢀ13 To date, live cancer cell detection with
varying degrees of selectivity and sensitivity has been limited to
routes employing extracellular or cellꢀsurface protein recognition
of a covalently attached component of the probe or reporter.3,12,14
However, successful identification of living cancer cells and their
type differentiation can be achieved by targeting endogenous,
intracellular, cancerꢀassociated enzymes with hydrophobic small-
molecule (<1000 Da), turnꢀon probes, an outcome not possible
with largeꢀmolecule, recognitionꢀbased probes.
A potentially significant and unexplored route with turnꢀon
probes for diagnostics is that in which selectivity and sensitivity
of the target reporting process are controlled by enzymatically
stimulated removal of fluorescence quencher from the probe. In
particular, this route employs probes whose reporter fluorescence
is quenched by photoꢀinduced electron transfer (PeT)15 from a
covalently attached enzyme substrate. Fluorescence ensues from
the reporter upon removal of the quencher substrate by an endogꢀ
enous, cytosolic, diseaseꢀassociated enzyme found in diseased
cells from a wide range of origins.
Herein we report the design, properties, and NQO1ꢀspecific
cellular activation of a firstꢀgeneration, PeTꢀquenched fluoresꢀ
cence probe. The turnꢀon sensor probe readily penetrates the
membrane of human cancer cells, because it is hydrophobic, small
in molecular weight, and charge neutral. Upon rapid and preferꢀ
ential, twoꢀelectron reduction of the quinone quencher subunit of
the turnꢀon probe, an intensely lightꢀemissive reporter results
from autonomous removal of the activated quinone subunit. Effiꢀ
cacious PeT quenching of fluorescence prior to subunit self cleavꢀ
age is ensured by careful selection of the electronic properties of
the naphthalimide reporter and quinone quencher. The resulting
highly fluorescent, cationic reporter is retained by cells, and it
exhibits a strong Stokes shift between its absorption and fluoresꢀ
cence emission maxima due to the pushꢀpull internal charge transꢀ
fer mechanism associated with the naphthalimide scaffold. Overꢀ
all, these probe characteristics allow for rapid and enhanced sigꢀ
nalꢀtoꢀbackground imaging and detection of living cancer cells
without the typical requirement of unactivated probe removal
from the environment. In total, the proꢀfluorogenic probe proꢀ
vides for realꢀtime, highly sensitive and selective human tumor
cell analysis and differentiation based on NQO1 content.
■
RESULTS AND DISCUSSION
Q3NI Probe Fluorescence is Controlled by Photo-
NAD(P)H:quinone oxidoreductase isozyme 1 (NQO1)16ꢀ19 is
intimately involved with cancer, as it is a gatekeeper for the 20S
proteasomal degradation of the p53, p73α, and p33 tumor supꢀ
induced Electron Transfer (PeT) Quenching. We deꢀ
signed the Q3NI probe to have the fluorescence signal of its naphꢀ
thalimide reporter22,23 quenched via oxidative electronꢀtransfer
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 7
(OeT) by the covalently attached quinone propionic acid motif,
Figure 1A. It was posited that it would be possible to achieve this
rescence enhancement for NI versus Q3NI and very large Stokes
shift of 116 nm (λmax, abs = 374 nm, max, em = 490 nm) bode well
1
λ
2
3
4
for use of Q3NI as a multifunctional turnꢀon probe for sensing
and imaging applications that utilize reductive stimuli capable of
initiating removal of the reduced quinone group.28ꢀ31
A.
OeT
Quenching
ReT
Quenching
Reductive
Stimulus
Cyclization
Fluorescence Dequenching of Q3NI is Achieved by
Reduction-initiated Removal of Quinone. We then wished
to determine if it is possible to produce the NI reporter from the
Q3NI probe by the expected cyclizative cleavage reaction of the
hydroquinone via the gemꢀdialkyl effect32 that occurs subsequent
to twoꢀelectron reduction of the quinone.28 Thus, the strong reꢀ
ducing agent sodium dithionite was added to aqueous solutions of
Q3NI. Under these conditions, it was found that NI is rapidly
released as indicated by the increase in timeꢀdependent fluoresꢀ
5
6
7
8
HQ3NI
Q3NI
= 0.007
NI
Φ
Φ
= 0.23
B.
C.
800
700
600
500
400
300
200
100
0
800
700
600
500
400
300
200
100
0
9
λex = 375 nm
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
λex = 375 nm
Q3NI
NI
cence intensity (Figure 2) at 470 nm (
λ
ex = 370 nm). To provide
1000
400 450 500 550 600 650 700
400 450 500 550 600 650 700
Wavelength (nm)
Wavelength (nm)
NI
Q NI
3
800
Figure 1. Schematic representation of the unique utilization
of PeT quenching in the hNQO1 probe Q3NI and the emission
properties associated with Q3NI probe and NI:
(A) proposed quenching mechanisms for Q3NI probe prior to
and after chemical/hNQO1ꢀcatalyzed reduction, and subseꢀ
quent production of the fluorescent NI reporter. Fluorescence
spectra and optical images of cuvettes containing
2.0 × 10−6 Μ solutions of (B) Q3NI probe and (C) NI reportꢀ
600
400
200
0
NI
er in pH 7.4, 0.1 M PBS that result from excitation at λex
375 nm (spectra) and ex = 365 nm (images).
=
λ
Q NI
1
novel mechanism of reporter quenching by carefully tuning the
electronic and optical properties of the quinone OeT quencher and
the naphthalimide reporter NI, due to extant naphthalimide reꢀ
porters that are quenched by reductive electronꢀtransfer (ReT).22
0 10 20 30 40
200300400500600
time (min)
Figure 2. Comparing the dithioniteꢀinitiated formation of NI
reporter from Q3NI probe and Q1NI. The fluorescence (λex
The
RehmꢀWeller
equation,
Equation
1,
=
e2
ꢀGPeT = ED − EA − ꢀG00
−
370 nm, λem = 470 nm) from 3ꢀmL solutions of
1.0 × 10−5 Μ Q3NI probe and Q1NI in pH 7.4 , 0.1 M PBS
was monitored after their reduction by addition of 2.75 mg
sodium dithionite (denoted by arrow).
εd
(1)
was used to examine possible quinone propionic acid quenchers24
and 1,8ꢀnaphthalimide reporters, as well as linkers between the
OeT quencher and the NI reporter, so as to ensure that quenching
is thermodynamically feasible and efficient.15 In this equation, ED
is the redox potential of the donor and EA that of the acceptor,
ꢀG00 is the energy of the first excited singlet state of the reporter,
and e2/εd is the Coulombic interaction energy of the ion pair,
known to be 0.06 eV.15 The energy of the first excited singlet
state of NI was measured to be 3.06 eV.15 From voltammetric
measurements, ED of NI was determined to be 1.74 V, and EA for
the quinone propionic acid group of Q3NI was found to be −1.01
V (Figure S2). From these values, the energy change for this OeT
process, ꢁGPeT, is calculated to be −0.37 eV, indicating that elecꢀ
tron transfer from the excited dye to the electronꢀpoor quinone is
thermodynamically favorable. The quinone was attached to the
naphthalimide via an Nꢀmethylethanolamine linker through a
carbamate to the amine of the naphthalimide ring. This linker
imparts three crucial properties on Q3NI: the linker is sufficiently
short to allow for a high probability of electron transfer, the elecꢀ
tronꢀwithdrawing carbamate yields a favorable ꢁGPeT, and the
presence of the tertiary amide provides enhanced environmental
stability.23,25 As a result, the fluorescence of Q3NI in pH 7.4, 0.1
M PBS is effectively quenched in comparison to that of the free
NI reporter, as noted by their spectra in Figure 1B and 1C and
proof positive that the increase in fluorescence for reduced Q3NI
results from cyclizative cleavage of the hydroquinone reduction
product as the lactone, a second probe (Q1NI) was synthesized
such that the rate of lactone formation from its hydroquinone form
is >103 times slower than in the case of reduced Q3NI, due to the
lack of the two methyl groups on the geminal carbon.28 Dithionite
reduction of the Q1 group is known to be equally fast as the Q3
group.28 As seen in Figure 2, reduced Q3NI exhibits exceedingly
rapid NI reporter production in comparison to reduced Q1NI; at
37 min (the maximum recorded for Q3NI), the signal for reduced
Q3NI is 20ꢀfold higher than that for Q1NI. Although it could be
argued the signal for reduced Q3NI is attributable to the absence
of photoꢀinduced electron transfer quenching of the reporter by
the hydroquinone of Q3NI, a significantly favorable ꢁGPeT of
−3.37 eV for the reductive quenching process of the NI reporter
by the hydroquinone clearly supports its occurrence. That is,
once the quinone is reduced to its hydroquinone (HQ3NI, Figure
1A), quenching of reporter fluorescence occurs by reductive elecꢀ
tron transfer (ReT), a common feature with naphthalimide dyes.22
Furthermore, we have isolated and identified the NI reporter from
Q3NI solutions treated with dithionite (Figure Sꢀ6). Thus, the
rapid increase in fluorescence for reduced Q3NI results from
naphthalimide dequenching caused by reductionꢀinitiated removal
of the quinone propionic acid group by lactonization. Due to the
unique quenching mechanism of the Q3NI turnꢀon probe sensor,
pronounced fluorescence signal enhancement upon revealing the
NI reporter and its large Stokes shift, and the known resistance of
respective fluorescence quantum yields (Φ) of 0.007 and 0.23
obtained using quinine sulfate as standard.26 The quantum yield
for NI is superior or comparable to that of other dyes applied to
cancer detection and localization, such as
Φ = 0.0028 for indocyꢀ
anine green and
Φ
= 0.21 for Cy5.5 dyes.13,27 The 33ꢀfold fluoꢀ
ACS Paragon Plus Environment

Page 3 of 7
Journal of the American Chemical Society
the quinone propionic acid trigger group to reduction by other
biological species,31 we investigated Q3NI as a sensor probe of
human NAD(P)H:quinone oxidoreductase isozyme 1 (hNQO1)
activity in realꢀtime biological applications.
the unquenched NI reporter (
Φ
= 0.23) and quenched Q3NI probe
1
2
3
NQO1-positive
NQO1-negative
A
B
C
Q3NI is Activated by hNQO1 at a High Rate. We deterꢀ
mined if Q3NI is a substrate capable of activation by hNQO1 to
yield the NI reporter at a significant rate. To do so, we obtained
the apparent kinetic parameters from MichaelisꢀMenten kinetic
treatment33 of the timeꢀdependent NI reporter production, namely
the Michaelis constant (Km), maximum velocity (Vmax), catalytic
constant (kcat), and substrate specificity (kcat/Km). The high rate of
NI reporter production under in vitro conditions is readily apparꢀ
ent in the inset of Figure 3; after 5 min, a fluorescent signal has
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
HT-29
A549
H596
Figure 4. Visual differentiation of (A) HTꢀ29, (B) A549, and
(C) H596 cancer cells cultured on 2.2 × 2.2 cm glass subꢀ
strates after 10ꢀmin incubation with 2 × 10−5 M Q3NI probe.
Images of cells were captured with a digital camera upon
0.03
300
250
200
150
100
50
NI
irradiation at
represents 1 cm.
λex = 365 nm with a handꢀheld lamp. Scale bar
0.02
0.01
(
Φ =0.007) and the large Stokes shift of the NI reporter that reꢀ
Q NI + NQO1
3
sults in fluorescence emission in the visible spectrum (400 nm to
~600 nm). These outcomes point to potential use of the Q3NI
probe sensor in the realꢀtime, visible (without the aid of imaging
equipment) and accurate determination of tumor/healthy tissue
borders so as to allow for surgical resection of tumors with small
foci.
ꢁ
ꢁ
ꢁ
ꢁ
0
0
1
2
3
4
5
time (min)
30
0
10
20
[Q3NI] x 10-6 M
Flow cytometry assays (Figure 5) were used to assess the apꢀ
plicability of the Q3NI probe to rapidly detect and quantify tumor
Figure 3. Kinetics plot of hNQO1 (2.0 × 10−5 g) toward
Q3NI Probe. Inset is an example of an assay observing fluoꢀ
rescent NI production from 1 × 10−6 M Q3NI, relative to sigꢀ
nal for complete NI formation (1 × 10−6 M). Values (n = 3)
are the average ± 1 standard deviation.
100
H446
H596
80
HT-29
been attained that is 22% of the maximum achievable, yielding
2.2 × 10−7 M of released reporter. From the plot shown in Figure
A549
60
3, we obtained Km = 3.86 ± 0.79 ꢁM, Vmax = 0.037 ± 0.002 ꢁmol
min−1mgꢂNQO1−1, kcat = 0.019 ± 0.001 s−1, and kcat/Km = 4.94 ±
0.33 × 103 M−1s−1. Due to the presence of the nonꢀbulky ethanoꢀ
lamine linker and the need of only a single activation step to reꢀ
veal reporter fluorescence, the kinetic constants are significantly
higher than those of other NQO1 activatable fluorophores,31,34,35
thereby ensuring sufficient signal enhancement for fast detection
of hNQO1 activity.
40
20
Cancer Cells can be Rapidly Visualized and Differen-
tiated Based on hNQO1 Activation of Q3NI. Our interest in
the potential use of the Q3NI probe in microfluidicꢀbased cell
sorting/isolation of circulating tumor cells36 and fluorescenceꢀ
guided resection of tumors3 led us to evaluate if the Q3NI probe
can be used to discriminate between cancer cells of different oriꢀ
gins and hNQO1 content, with and without the aid of sophisticatꢀ
ed instrumentation. The colorectal carcinoma cell line HTꢀ29 and
the nonꢀsmall cell lung cancer (NSCLC) A549 cell line are known
to possess significant hNQO1 activity, while the NSCLC H596
cell line has been reported to have undetectable hNQO1 activiꢀ
ty.37,38 After a 10ꢀmin incubation period in a cell culture solution
containing 2 × 10−5 M Q3NI, it was possible to differentiate beꢀ
tween the various substrateꢀcultured cells (4.84 cm2) using only a
handꢀheld fluorescent lamp emitting at 365 nm and the unꢀaided
eye (Figure 4). Both HTꢀ29 (3.69 × 106 total cells) and A549
(5.72 × 106 total cells) appeared fluorescent blue, while H596
(3.96 × 106 total cells) exhibited no apparent emission. The abilꢀ
ity to visually determine the presence of hNQO1 in a small numꢀ
ber of cells is due to the marked difference in fluorescence from
0
101
102
Intensity (
103
104
λ
= 457/60 nm)
em
Figure 5. Flow cytometry assays of Q3NI activation by
hNQO1ꢀpositive (A549 and HTꢀ29) and hNQO1ꢀnegative
(H596 and H446) cancer cells. Assays were performed by
counting 10,000 cells that had been exposed to 2.0 × 10−5
M
Q3NI probe for 10 min; λex = 405 nm, λem = 457/60 nm.
cells containing hNQO1. The probe was incubated in a suspenꢀ
sion of HTꢀ29, A549, H596, and H446 (a small cell lung carciꢀ
noma known to be devoid of hNQO1 activity37) cells for either 10
or 60 min, and a flow cytometer was used to measure the fluoresꢀ
cence (λex = 405 nm, λem = 457/60 nm) in 10,000 individual cells.
In Figure 5 are shown the histograms for each cancer cell line. It
is clear that a highꢀintensity, unimodal distribution of signals is
obtained for Q3NI activation in each of the two hNQO1ꢀpositive
cell lines (HTꢀ29 and A549), while the negative cell lines H596
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 7
and H446 produced minimal fluorescence. It was also found that
toꢀbackground value achieved for activation of Q3NI to NI reꢀ
1
2
3
4
5
6
7
8
there was little change in the cell count or intensity of the histoꢀ
grams for a longer probe incubation time (60 min vs. 10 min,
Figure S3), demonstrating the rapid and substantial activation of
Q3NI in A549 and HTꢀ29 cells. Importantly, the sustained low
fluorescence observed with the H446 cells (Figure S3) points to
the intracellular stability of Q3NI (lack of nonꢀspecific activaꢀ
tion). Thus, it is indicated that Q3NI is a highly sensitive and
selective probe capable of being used to rapidly discern different
types of tumor cells in fluidic streams.
porter (9ꢀ to 23ꢀfold) in target versus nonꢀtarget cells, relative to
that of other exogenously introduced sensor probes for whole
tumor analysis (2.5ꢀ to 5ꢀfold),40ꢀ42 points to the potential of Q3NI
to provide highly selective tumor cell analyses with low limits of
detection, even in the face of possible background fluorescence
from hemoglobin and other species.1 In addition, our results indiꢀ
cate that dye quantum yield is affected little, and there is no apꢀ
parent efflux of NI reporter from cells during paraformaldehyde
fixing, as noted by sustained fluorescence in fixed samples stored
for 10 months in the laboratory ambient. Collectively, there is
great potential for use of our probe/reporter system for ex vivo
quantitative analysis of excised tumor cells and longꢀterm in vivo
and in vitro imaging.
Multiphoton (MP) microscopy imaging of cells and tissues is
more advantageous when compared to traditional fluorescence
microscopy, because use of the characteristic longꢀwavelength
photons offers higher fluorophore and cellular photostability that
provides extended imaging duration, less background signal from
outꢀofꢀfocus excitation and scattering events, and deeper penetraꢀ
tion depth. In addition, MP imaging is ideal for direct observaꢀ
tion of targets in their physiological environment and ex vivo
thickꢀspecimen sampling where 2D and 3D maps can be generatꢀ
ed.13 During incubation with complete growth medium containꢀ
ing 2.0 × 10−5 M Q3NI, 2ꢀphoton microscopy revealed significant
fluorescence signal from NI in living, hNQO1ꢀpositive HTꢀ29
and A549 cells (Figure 6) and minimal signal in two living,
hNQO1ꢀnegative cell lines, H596 and H446. The average fluoꢀ
rescence signal was determined to be 13ꢀfold higher in A549 cells
compared to H596 cells, and 3.66 × 104ꢀfold higher versus H446
cells. Similar results were obtained with the HTꢀ29 cell line, with
the cytosolic intensity being 15ꢀ and 4.51 × 104ꢀfold higher comꢀ
pared to H596 and H446 cells. As before, the signal appears
somewhat heterogeneous throughout the cytosolic space of the
HTꢀ29 and A549 cells due to NI accumulation in acidic orgaꢀ
nelles. To ensure Q3NI and NI had little effect on cell health,
cells were incubated in a 2.0 × 10−5 M Q3NI solution in complete
Live and Fixed hNQO1-containing Cancer Cells can
be Selectively Imaged via Q3NI Activation. We desired to
know if cancer cell types can be differentiated in conventional
fluorescence microscopy images resulting from Q3NI activation
by hNQO1, and if it is possible to obtain images that allow collecꢀ
tion of information regarding the fate of the released NI reporter.
Furthermore, we wished to learn if the NI reporter can be successꢀ
fully used in upconversion fluorescence (multiꢀphoton) imaging
of live cells.13 In agreement with flow cytometry data, wideꢀfield
imaging of fixed hNQO1ꢀpositive cells exposed to the Q3NI
probe for 10 min revealed significant probe uptake and activation
that leads to intracellular NI fluorescence for the A549 and HTꢀ29
cell lines; however, minimal signal was observed in the hNQO1ꢀ
negative H596 cells (Figure S4). There was no indication of reꢀ
porter in the nucleus, pointing to the lack of NQO1 there; this is
in clear contrast to previous work using immunohistochemical
staining of permeabilized, fixed cells.39 The average cytosolic
signal was 9 times higher in A549 cells versus NQO1ꢀnegative
H596 cells, while it was 23 times higher in HTꢀ29 cells compared
to the NQO1ꢀnegative H596 cells. After incubating live HTꢀ29
cells for 20 min with Q3NI followed by exposure to acidic orgaꢀ
nelleꢀspecific Lysotracker Red in the media in the imaging dish, it
was found that the majority of the NI signal originated from the
cytosolic region. Not surprisingly, accumulation of the basic
(secondary amine pKa ~11) NI occurred in acidic late endosomal
and lysosomal vesicles (Figure S5), a beneficial outcome that
leads to enhanced intracellular retention of NI. The higher signalꢀ
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
NQO1-positive
NQO1-negative
A
HT-29 B
A549 C
H596 D
H446
G
E
HT-29 F
A549
H596 H
H446
Figure 6. Twoꢀphoton confocal microscopy imaging of living HTꢀ29 (A,E), A549 (B,F), H596 (C,G), and H446 (D,H) cells in the
presence of 2.0 × 10−5 M Q3NI probe. Fluorescence images of each cell line are on the top row with their respective differential interꢀ
ference contrast (DIC) image on the bottom row. Images were acquired at
λex = 750 nm (3% laser power) without any washing steps
between Q3NI probe addition and imaging. Scale bars represent 2.5 × 10−5 m.
ACS Paragon Plus Environment

Page 5 of 7
Journal of the American Chemical Society
growth medium for one hour and one day, and then cell viability
with the PBS buffer; this NADH solution was subsequently used
to prepare all other solutions so as to have a final concentration of
1 × 10−4 M
ꢀNADH in each assay. Solutions consisting of 2 ×
10−6 to 6 × 10−5 M Q3NI were made using the NADH stock. A
1.33 g/mL stock solution of recombinant human NQO1 (Sigmaꢀ
Aldrich) was prepared using the same buffer as above so as to
give 2.0 × 10−5 g hNQO1 per assay. Each assay was performed in
a quartz fluorescence cuvette containing 1.5 mL Q3NI solution
and initiated by the addition of 1.5 mL hNQO1 solution. Measꢀ
urements were collected every 30 s for 5 min. Fluorescence units
were converted to concentration by relating the signal increase to
a fluorescence signal derived from a known concentration of NI.
Plots of velocity versus Q3NI concentration were used to obtain
apparent Km and Vmax values from nonꢀlinear least squares analyꢀ
sis employing algorithms developed by Cleland for Michaelisꢀ
Menten kinetics.33
Cell Viability. Cells were cultured in 5 mL of complete
growth medium in a sterile tube. To the medium was added Q3NI
(from a 1.8 × 10−3 M stock solution) in DMSO, to give a 2.0 ×
10−5 M Q3NI solution, and the cells were incubated at 37 °C, 5%
CO2 for 1 h or 24 h, after which 1 mL of the cell suspension was
removed and then 0.1 mL trypan blue was added. Cells were
immediately counted using a hemocytometer and microscope.
Optical Differentiation. HTꢀ29, A549, and H596 cells were
cultured overnight on 2.2 × 2.2 cm glass cover slips. Old growth
medium was removed and replaced with a 2.0 × 10−5 M Q3NI
solution in fresh growth medium. Cells were incubated at 37 °C
with Q3NI for 10 min and then rinsed with 0.1 M PBS. Cover
slips of each cell line were immediately placed upside down on
glass slides and visualized with a Kodak digital camera in conꢀ
junction with a Mineralight Model UVGLꢀ25 lamp (365 nm, 0.16
amps, 60 Hz).
1
2
3
4
5
6
7
8
was assessed with a trypan blue assay. After 1 h, cell viability for
HTꢀ29, A549, and H596 was 97.7%, 98.8%, and 100%, while it
was 97.7%, 98.7%, and 98.4% after 24 h. Of particular imꢀ
portance is the realꢀtime nature of the Q3NI probe in imaging, as
this system does not require the timeꢀconsuming wash steps charꢀ
acteristic of alwaysꢀon reporters.1,4
β
ꢁ
■
CONCLUDING REMARKS
In summary, we have presented the successful development and
implementation of a quenched turnꢀon probe whose fluorescent
reporter signal is selectively and quickly generated and mainꢀ
tained in human cancer cells upon activation by a cancerꢀ
associatedꢀenzyme substrate that is subsequently removed from
the probe to yield the free deꢀquenched reporter dye. Specifically,
intracellular reduction of the quinone substrate by
NAD(P)H:quinone oxidoreductase and successive selfꢀremoval of
the reduced substrate from the reporter leads to elimination of the
photoꢀinduced electronꢀtransfer quenching event, thereby yielding
the naphthalimide reporter possessing a highly intense and
Stokesꢀshifted emission. As a result, we have achieved rapid and
highly selective/sensitive detection, differentiation, and enumeraꢀ
tion of cancer cells in flowing streams and under traditional miꢀ
croscopy conditions without the use of recognition agents, an
outcome that has in turn allowed for imaging of live cancer cells
by multiꢀphoton fluorescence microscopy. The latter outcome
bodes well for in vivo and ex vivo thickꢀspecimen imaging, where
the efficacy of treatment protocols and the cellular impact of
pharmaceutical targets can be evaluated for personalized oncoloꢀ
gy. Finally, control over the emission characteristics of the reꢀ
porter dye22 coupled to our ability to visibly see, by the unꢀaided
eye, relatively small numbers of cancer cells, reveals the potential
of this new class of probes for application to fluorescenceꢀguided
surgical resection of cancerous tissues.
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
Confocal Colocalization. Confocal fluorescence images
were acquired with a Leica TCS SP5 tandem scanning multiphoꢀ
ton laser scanning microscope using a 40x oil immersion objecꢀ
tive lens (1.25 NA). Imaging of Lysotracker Redꢀ and Q3NIꢀ
exposed cells was accomplished using a sequential scanning
■
EXPERIMENTAL
Full details regarding materials, synthesis of compounds, and
method with
λ
ex = 405 nm at 10% output and collecting emission
their characterization are described in the Supporting Information.
Cell Culture. HTꢀ29 (human colorectal adenocarcinoma),
A549 (human nonꢀsmall cell lung cancer, NSCLC), H596 (human
NSCLC), and H446 (human smallꢀcell lung cancer) were all purꢀ
chased from American Type Cell Culture. HTꢀ29 cells were culꢀ
tured in McCoy’s 5A medium supplemented with 10% fetal boꢀ
vine serum (FBS) and 100 IU/mL penicillinꢀstreptomycin, A549
cells were cultured in Fꢀ12K medium with 10% FBS and 100
IU/mL penicillinꢀstreptomycin, and H596/H446 were cultured in
RPMIꢀ1640 with 10% FBS and 100 IU/mL penicillinꢀ
streptomycin. Cells were incubated at 37 °C in a humidified inꢀ
cubator containing 5% wt/vol CO2.
between 417 and 467 nm (Q3NI) or
λ
ex = 561 nm at 10% output,
and the emitted light was collected between 574 and 621 nm
(Lysotracker Red). Differential interference contrast (DIC) imagꢀ
es were obtained using a PMT detector and 633ꢀnm light at 3%
output as an illumination source. All images were collected at 37
°C using the Leica TCS SP5 in resonant scanning mode (16 KHz)
at a zoom setting of 3.6. Images were line averaged 64 times.
Cells were cultured overnight in black 35 x 10 mm, 22ꢀmm well,
glassꢀbottom dishes (Chemglass Life Sciences). Prior to imaging,
the medium was removed and replaced with medium (containing
no phenol red) and maintained at 37 °C. A concentrated solution
of Q3NI in DMSO was added directly to the dish to give a conꢀ
centration of 2.0 × 10−5 M, with [DMSO] ≤ 1%. Cells were incuꢀ
bated with Q3NI for 20 min. Five min prior to imaging,
Lysotracker RedꢀDND 99 in DMSO was added to give a concenꢀ
tration of 1 × 10−7 M.
In vitro 2-Photon Imaging. Twoꢀphoton confocal fluoresꢀ
cence images were acquired as above using a MaiTai twoꢀphoton
laser tuned to 750 nm (3% laser power, modelocked Ti:sapphire
laser; Tsunami Spectra Physics), and emission was collected using
a shortꢀpass 680ꢀnm filter. DIC images were collected as above.
Cells were imaged, with no washing or medium removal steps,
after a 10ꢀmin incubation with Q3NI. Image analysis was perꢀ
formed in ImageJ.
Flow Cytometry. To a 1ꢀmL suspension of cells (1 × 106) in
complete growth medium was added Q3NI from a stock dimethylꢀ
sulfoxide (DMSO) solution to give a final [Q3NI] = 2 × 10−5 M
and [DMSO] ~1%. After 10ꢀ or 60ꢀmin incubations at 37 °C, the
cells were fixed with 30 mL of 4% paraformaldehyde in 0.1 M
PBS for 1 h, followed by washing twice with 0.1 M PBS, and
then resuspending in 1 mL PBS. Data acquisition was carried out
on an iCyt Reflection flow cytometer using 405ꢀnm excitation and
a 457/60 nm emission filter; logarithmic amplification was used.
Winlist software (Verity Software House, Topshame, ME) was
used to count 10,000 cells per sample; FlowJo software was used
to construct histograms.
Enzyme Kinetics. All fluorescence measurements (
nm, em = 470 nm) were obtained at room temperature, with soluꢀ
tions composed of pH 7.4, 0.1 M PBS/0.1 M KCl/0.007% BSA.
A stock solution of 1 × 10−4 M
ꢀnicotinamide adenine dinucleoꢀ
tide, reduced disodium salt (NADH, SigmaꢀAldrich), was made
λex = 390
λ
β
■
ASSOCIATED CONTENT
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 7
(23) Qian, X.; Xiao, Y.; Xu, Y.; Guo, X.; Qian, J.; Zhu, W. Chem.
Commun. 2010, 46, 6418ꢀ6436.
(24) Mendoza, M. F.; Carrier, N. H.; Hettiarachchi, S. U.; McCarley, R.
L. Biochemistry 2012, 51, 8014ꢀ8026.
(25) Nicolaou, M. G.; Wolfe, J. L.; Schowen, R. L.; Borchardt, R. T. The
Journal of Organic Chemistry 1996, 61, 6633ꢀ6638.
(26) FeryꢀForgues, S.; Lavabre, D. J. Chem. Educ. 1999, 76, 1260ꢀ1264.
(27) Gioux, S.; Choi, H. S.; Frangioni, J. V. Mol. Imaging 2010, 9, 237ꢀ
55.
(28) Ong, W.; Yang, Y.; Cruciano, A. C.; McCarley, R. L. J. Am. Chem.
Soc. 2008, 130, 14739ꢀ44.
(29) Ong, W.; McCarley, R. L. Macromolecules 2006, 39, 7295ꢀ7301.
(30) Ong, W.; McCarley, R. L. Chem Commun (Camb) 2005, 4699ꢀ701.
(31) Silvers, W. C.; Payne, A. S.; McCarley, R. L. Chem. Commun.
2011, 47, 11264ꢀ11266.
(32) Milstien, S.; Cohen, L. A. Proceedings of the National Academy of
Sciences of the United States of America 1970, 67, 1143ꢀ1147.
(33) Cleland, W. W. Methods Enzymol. 1979, 63, 103ꢀ138.
(34) Huang, S. T.; Lin, Y. L. Org. Lett. 2006, 8, 265ꢀ268.
(35) Huang, S.ꢀT.; Peng, Y.ꢀX.; Wang, K.ꢀL. Biosensors & Bioelectron-
ics 2008, 23, 1793ꢀ1798.
(36) Adams, A. A.; Okagbare, P. I.; Feng, J.; Hupert, M. L.; Patterson,
D.; Gottert, J.; McCarley, R. L.; Nikitopoulos, D.; Murphy, M. C.; Soper,
S. A. J. Am. Chem. Soc. 2008, 130, 8633ꢀ8641.
(37) Beall, H. D.; Murphy, A. M.; Siegel, D.; Hargreaves, R. H.; Butler,
J.; Ross, D. Mol. Pharmacol. 1995, 48, 499ꢀ504.
(38) SmitskampꢀWilms, E.; Hendriks, H. R.; Peters, G. J. General
pharmacology 1996, 27, 421ꢀ9.
(39) Winski, S. L.; Koutalos, Y.; Bentley, D. L.; Ross, D. Cancer Res
2002, 62, 1420ꢀ4.
(40) Andreev, O. A.; Dupuy, A. D.; Segala, M.; Sandugu, S.; Serra, D.
A.; Chichester, C. O.; Engelman, D. M.; Reshetnyak, Y. K. Proceedings
of the National Academy of Sciences of the United States of America
2007, 104, 7893ꢀ7898.
Supporting Information
1
2
3
4
5
6
7
8
Synthetic protocol and characterization data for Q3NI, Q1NI, and
NI; flow cytometry results; wideꢀfield and confocal images of
fixed cells exposed to Q3NI. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
9
tunnel@LSU.edu
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
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
This material is based upon work supported by the US National
Institutes of Health (5R21CA135585) and the US National Sciꢀ
ence Foundation under grant number CHEꢀ0910845. This work
utilized the facilities of the Cell Biology and Bioimaging Core
that are supported in part by COBRE (NIH 8 P20ꢀGM103528)
and NORC (NIH 2P30ꢀDK072476) center grants from the Naꢀ
tional Institutes of Health. We thank Drury Ingram and Marilyn
Dietrich for help with flow cytometry experiments.
■
REFERENCES
(1) Razgulin, A.; Ma, N.; Rao, J. Chem. Soc. Rev. 2011, 40, 4186ꢀ216.
(2) Siegel, R.; Naishadham, D.; Jemal, A. CA Cancer J Clin 2012, 62,
10ꢀ29.
(3) Nguyen, Q. T.; Olson, E. S.; Aguilera, T. A.; Jiang, T.; Scadeng, M.;
Ellies, L. G.; Tsien, R. Y. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 4317ꢀ
22.
(4) Kobayashi, H.; Choyke, P. L. Acc. Chem. Res. 2010, 44, 83ꢀ90.
(5) Luo, S.; Zhang, E.; Su, Y.; Cheng, T.; Shi, C. Biomaterials 2011, 32,
7127ꢀ38.
(6) Shi, H.; He, X.; Wang, K.; Wu, X.; Ye, X.; Guo, Q.; Tan, W.; Qing,
Z.; Yang, X.; Zhou, B. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 3900ꢀ5.
(7) Blum, G.; von Degenfeld, G.; Merchant, M. J.; Blau, H. M.; Bogyo,
M. Nat Chem Biol 2007, 3, 668ꢀ677.
(8) Bremer, C.; Tung, C.ꢀH.; Weissleder, R. Nat Med 2001, 7, 743ꢀ748.
(9) Jaffer, F. A.; Kim, D. E.; Quinti, L.; Tung, C. H.; Aikawa, E.; Pande,
A. N.; Kohler, R. H.; Shi, G. P.; Libby, P.; Weissleder, R. Circulation
2007, 115, 2292ꢀ8.
(10) Simon, E. Meas. Sci. Technol. 2010, 21, 112002ꢀ112026.
(11) Tung, C.ꢀH. Peptide Science 2004, 76, 391ꢀ403.
(12) Urano, Y.; Sakabe, M.; Kosaka, N.; Ogawa, M.; Mitsunaga, M.;
Asanuma, D.; Kamiya, M.; Young, M. R.; Nagano, T.; Choyke, P. L.;
Kobayashi, H. Sci. Transl. Med. 2011, 3, 110ra119.
(13) van den Berg, N. S.; van Leeuwen, F. W.; van der Poel, H. G.
Current Opinion in Urology 2012, 22, 109ꢀ20.
(14) Ogawa, M.; Kosaka, N.; Choyke, P. L.; Kobayashi, H. ACS Chem.
Biol. 2009, 4, 535ꢀ546.
(15) Lakowicz, J. R. In Principles of Fluorescence Spectroscopy; Third
ed.; Springer: 2006, p 331ꢀ351.
(16) DinkovaꢀKostova, A. T.; Talalay, P. Arch. Biochem. Biophys. 2010,
501, 116ꢀ23.
(17) Danson, S.; Ward, T. H.; Butler, J.; Ranson, M. Cancer Treat. Rev.
2004, 30, 437ꢀ449.
(18) Schlager, J. J.; Hoerl, B. J.; Riebow, J.; Scott, D. P.; Gasdaska, P.;
Scott, R. E.; Powis, G. Cancer Research 1993, 53, 1338ꢀ1342.
(19) Dong, G.ꢀZ.; Youn, H.; Park, M.ꢀT.; Oh, E.ꢀT.; Park, K. H.; Song,
C. W.; Kyung Choi, E.; Park, H. J. International Journal of Hyperther-
mia 2009, 25, 477ꢀ487.
(20) Choi, E. K.; Terai, K.; Ji, I. M.; Kook, Y. H.; Park, K. H.; Oh, E. T.;
Griffin, R. J.; Lim, B. U.; Kim, J. S.; Lee, D. S.; Boothman, D. A.; Loren,
M.; Song, C. W.; Park, H. J. Neoplasia 2007, 9, 634ꢀ642.
(21) Chen, Y.; Hu, L. Medicinal Research Reviews 2009, 29, 29ꢀ64.
(22) Duke, R. M.; Veale, E. B.; Pfeffer, F. M.; Kruger, P. E.; Gunnlaugsꢀ
son, T. Chem. Soc. Rev. 2010, 39, 3936ꢀ3953.
(41) Shi, C. Sci. Transl. Med. 2012, 4, 121le1.
(42) Jiang, T.; Olson, E. S.; Nguyen, Q. T.; Roy, M.; Jennings, P. A.;
Tsien, R. Y. Proceedings of the National Academy of Sciences of the
United States of America 2004, 101, 17867ꢀ17872.
ACS Paragon Plus Environment

Page 7 of 7
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
Enzyme Substrate-
Quenched Reporter
Cleaved
Substrate
Fluorescent
Push-pull
Reporter
hν
h
ν
Q3NI
NI
Activation
Cell Entry
NQO1
Enzyme
Table of Contents artwork
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