Detection of DT-diaphorase Enzyme with a ParaCEST MRI Contrast Agent

Article abstract of DOI:10.1002/chem.201700721

A responsive magnetic resonance (MRI) contrast agent has been developed that can detect the enzyme activity of DT-diaphorase. The agent produced different chemical exchange saturation transfer (CEST) MRI signals before and after incubation with the enzyme

Full text of DOI:10.1002/chem.201700721

DOI: 10.1002/chem.201700721
Communication
&
Imaging Agents
Detection of DT-diaphorase Enzyme with a ParaCEST
MRI Contrast Agent
Iman Daryaei,[b] Kyle M. Jones,[c] and Mark D. Pagel*[a]
stratifying patients for chemotherapy with quinone-based
drugs is a notable challenge.
Abstract: A responsive magnetic resonance (MRI) contrast
agent has been developed that can detect the enzyme ac-
tivity of DT-diaphorase. The agent produced different
chemical exchange saturation transfer (CEST) MRI signals
before and after incubation with the enzyme, NADH, and
GSH at different pH values whereas it showed good stabil-
ity in a reducing environment without enzyme.
The reductive lactonization of quinone propionic amide was
first introduced as a novel strategy for controlled drug re-
lease.[10] This class of organic molecules uses a trimethyl lock
(TML), which benefits from a steric effect between a methyl
group on the quinone ring and two methyl groups on the pro-
pionic acid side chain that accelerates the rate of lactonization
by a factor of ꢀ1016 M.[11] This feature made TML an ideal tool
for developing a trigger for molecular release with applications
in chemistry, biology, and pharmacology.[12] The introduction of
latent fluorescent agents based on TML provided new applica-
tions of this compound for imaging studies.[13] More recently,
several fluorescence TML-based quinone agents were intro-
duced for detecting DTD in cancer[14] and as theranostics for
imaging as well as therapeutic applications.[15] Although these
agents provide novel tools for detecting enzyme activity, the
limited depth of view of fluorescence imaging within in vivo
tissues demands another class of agents that detect enzyme
activity in deep tissues.
DT-diaphorase (DTD), also known as NAD(P)H-Quinone Oxyre-
ductase 1 (NQO1), is a flavoenzyme that catalyzes the two-
electron reduction of quinone, quinone imines, and azo dyes
in the presence of NAD(P)H,[1] which results in detoxification of
cells from xenobiotics.[2] The enzyme is upregulated in non-
small cell lung carcinoma, colorectal carcinoma, liver cancers,
as well as in breast carcinomas in various cell organelles, cyto-
sol, and extracellular regions. High expression of DTD is corre-
lated with a lower disease free survival and overall survival.[3]
For comparison, normal tissues have low expression levels of
the enzyme.[4]
Chemical exchange saturation transfer (CEST) is a relatively
new magnetic resonance imaging (MRI) technique with prom-
ising applications that detect enzymes, redox state, metabo-
lites, and measurement of pH and temperature, with some of
these applications advancing to preclinical and clinical stud-
ies.[16] The CEST technique offers several advantages relative to
conventional MRI techniques including selectively turning the
signal on and off, the capability to detect multiple CEST signals
from various labile protons on the same agent or from differ-
ent agents, and more flexibility in designing responsive MRI
agents for a particular target by simply changing functional
groups. To generate CEST, selective saturation occurs by using
a radiofrequency pulse that matches the resonance frequency
of exchangeable protons on a contrast agent, which suppress-
es the MR signal from that proton. The exchange of the satu-
rated proton between the contrast agent and surrounding
water molecules can transfer saturation from the contrast
agent to the water and consequently some water molecules
lose MR signal. The measurement of water signals while satu-
rating pulses are iteratively applied over a range of MR fre-
quencies provides a CEST spectrum, in which exchangeable
protons produce signals with unique chemical shifts.[17]
DTD can bioreductively activate quinone-based prodrugs to
facilitate selective chemotherapy.[5] Resistance to chemothera-
py with quinone-bearing drugs has been associated with
downregulation of DTD,[6] and the determination of DTD activi-
ty has been suggested as a measure of quinone-bearing anti-
tumor-drug resistance.[7] Mitomycin C and its derivatives such
as EO9 (apaziquone) have been extensively investigated as
prodrugs activated by DTD in various types of cancers[8] and
several ongoing clinical trials (such as NCT02563561) are based
on the therapeutic effect of these compounds.[5a,9] However,
[a] Dr. M. D. Pagel
Department of Medical Imaging, University of Arizona
1501 N. Campbell, P.O. Box 245067
Tucson, Arizona 85724 (USA)
E-mail: mpagel@u.arizona.edu
[b] I. Daryaei
Department of Chemistry and Biochemistry
University of Arizona, 1306 E. University Blvd., Room 221
Tucson, Arizona 85721-0041 (USA)
[c] Dr. K. M. Jones
Department of Biomedical Engineering
University of Arizona, 1127 E James E. Rogers Way P.O. Box 210020
Tucson, AZ 85721-0020 (USA)
We previously reported Yb-DO3A-oAA as a CEST MRI agent
that produces a CEST signal from a proton of the amide func-
tional group connecting the aniline ring to the DO3A chelate
and another CEST signal from the protons of the amine group
on the aniline ring.[18] We have shown that adding a relatively
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under
https://doi.org/10.1002/chem.201700721.
Chem. Eur. J. 2017, 23, 6514 –6517
6514
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 1. The quinone functional group in Yb-DO3A-oAA-TML-Q was reduced to hydroquinone in the presence of DT-diaphorase enzyme, NADH, and GSH.
The newly formed hydroxy group attacked and dissociated the amide bond as a result of the steric effect between the methyl group on the benzene ring
and the two adjacent methyl groups, consequently resulting in the release of the CEST agent.
large ligand to the aniline group causes a loss of one or both
CEST signals.[19] Therefore we decided to attach a quinone
moiety to the aniline ring of the molecule to eliminate the
CEST signal from the protons of the amine group of the aniline
ring to develop a responsive agent that detects DTD. We hy-
pothesized that reduction of the quinone functional group to
a hydroquinone group followed by intramolecular reaction of
the newly formed hydroxyl functional group and the amide
bond would release the CEST agent, which consequently
would produce two CEST signals (Scheme 1).[10a]
After incubation of the cloaked CEST agent (12 mm) with 10
units (100 mg) of DTD enzyme, NADH (71 mm), and GSH
(47 mm), the solution generated 12.5% CEST signal at 9 ppm
(red line in Figure 1B). The solution also showed 0.9% CEST
signal at À9 ppm (dashed red line in Figure 1B) indicating that
some unreacted agent remained in the sample. In addition,
a 44% CEST signal at 2.5 ppm was observed (dotted red line in
Figure 1B), which was attributed to the amide protons of
NADH and GSH as shown experimentally (Figure S1 in the Sup-
porting Information). Formation of the product was also vali-
dated with mass spectroscopy which confirmed that Yb-DO3A-
oAA was released from the cloaked CEST agent (Figure S2) only
with DTD, NADH, and GSH.
The agent was tested in several buffer systems, and PBS of-
fered the best activation of the cloaked CEST agent in the pres-
ence of DTD enzyme, nicotinamide adenine dinucleotide
(NADH), and glutathione (GSH). The cloaked CEST agent at
a concentration of 12 mm in PBS showed 3.2% CEST signal at
À9 ppm (dashed red line in Figure 1A) arising from the proton
of the amide functional group connecting the aromatic ring to
the Yb-DO3A chelate (blue amide in Scheme 1). No signal at
9 ppm was observed, at which the free amine would show
a CEST signal. However, the presence of a broad peak with
4.1% CEST signal at 3.5 ppm (dotted red line in Figure 1A) vali-
dated that the second amide group was present in the agent.
We evaluated the stability of the CEST MRI contrast agent to
confirm the specificity toward the reductase enzyme, NADH,
and GSH for activation. The cloaked CEST agent was incubated
with DTD and NADH without GSH, and with DTD and GSH
without NADH. Analysis of the solution with MRI showed 7.2%
CEST at À9 ppm for the solution with NADH and the enzyme
(Figure 1C), and 4.6% CEST at À9 ppm for the solution with
GSH and the enzyme (Figure 1D) with no CEST signal at
9 ppm, indicating that all reaction components were needed
for the activation of the agent.
Similar optical agents can be activated with the enzyme and
NADH without requiring GSH, because the optical agents can
be detected at nm-to-mm concentrations and thus a high ratio
of NADH relative to the optical agents can be used to com-
plete the reaction in a short time. Using a higher ratio of
NADH relative to our CEST agent was impractical because
a large concentration of NADH causes a line-broadening effect
that overlaps with the signal from the activated CEST agent.
However, this would not create a limitation in vivo considering
that NADH is in a relatively low concentration and is produced
through cyclic reduction of NAD+. Therefore continuous pro-
duction of NADH would provide enough concentration of the
co-enzyme for the enzymatic reaction. A smaller ratio of NADH
relative to the CEST agent in combination with instability of
the reduced form of NADH in solutions under oxidative condi-
tions will result in a reaction with low yield or longer experi-
ment time. A solution to this problem is addition of GSH to
prevent oxidation of NADH.
Figure 1. Yb-DO3A-oAA-TML-Q was reacted with DT-diaphorase enzyme at
pH 8.1Æ0.1 and 378C for 15 h. CEST spectra were acquired with 6 mT, satura-
tion power applied for 4 sec. (A) No reaction was detected when no
enzyme, GSH, and NADH were added to the solution. (B) A new CEST signal
at 9 ppm (solid red line) was detected when the CEST MRI agent was react-
ed with the enzyme, GSH, and NADH. The absence of NADH (C) or GSH (D)
did not allow formation of the desired product whereas the CEST signal of
unactivated CEST MRI agent (dashed lines in C and D) was clearly detected.
The same reactions were performed in an anaerobic cham-
ber to investigate the effect of a reducing environment on the
activation of the agent. No signal indicating activation of the
agent was detected in the solution of the cloaked CEST agent
Chem. Eur. J. 2017, 23, 6514 –6517
www.chemeurj.org
6515
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
major absorbance peaks at 260 and 339 nm (teal line in Fig-
ure S3 in the Supporting Information). The pure solution of the
cloaked agent with or without GSH showed only one peak at
260 nm (purple line and green line in Figure S3). The solutions
containing the cloaked CEST agent incubated with the enzyme
and NADH, with or without GSH under anaerobic conditions,
showed an appearance of a shoulder at 292 nm (blue and red
lines in Figure S3).
The magnitude of the CEST signal is directly correlated with
the saturation power level applied during a CEST MRI experi-
ment.[19b] In our studies, increasing the power level is favorable
for producing a larger CEST signal, which facilitates signal de-
tection. However, a relatively low power level is preferred in
a clinical setting due to concerns for patient safety. Our results
suggested that a power level as low as 3.5 mT was sufficient to
produce detectable signals, which is a safe level for future in
vivo studies.
Figure 2. Yb-DO3A-oAA-TML-Q showed stability under anaerobic conditions
(A) whereas the presence of DT-diaphorase enzyme and NADH in presence
(B) or absence (C) of GSH activated the agent and produced a CEST signal at
9 ppm at pH 7.0Æ0.25, 378C for 15 h, power level of 6 mT, and saturation
time of 4 sec. The agent showed good stability in a reducing environment
even in the presence of the enzyme and GSH in an anaerobic chamber (D).
In conclusion, we have introduced a new responsive CEST
MRI agent for the detection of DT-diaphorase enzyme. The
cloaked CEST agent showed only one signal at À9 ppm before
activation. In the presence of the enzyme, NADH, and GSH, the
agent showed a different signal at 9 ppm after activation. The
agent exhibited good stability in a reducing environment
where no activation was observed. This cloaked agent is
a promising template for the design of molecules for in vivo
studies that investigate reductase activity within tissues.
without NADH or DTD enzyme, and the 2.8% CEST signal at
À9.75 ppm indicated that the cloaked CEST agent was com-
pletely stable in anaerobic conditions (Figure 2A). The cloaked
CEST agent incubated with the enzyme and NADH with GSH
under anaerobic conditions produced 2.0% and 8.9% CEST sig-
nals at 9 ppm and À9.5 ppm, respectively, indicating that
some of the cloaked CEST agent was activated (Figure 2B). In-
terestingly, the solution of the cloaked CEST agent with the
enzyme and NADH without GSH produced 3.5% CEST at
9 ppm and 8.5% CEST at À9.25 ppm (Figure 2B) indicating
that anaerobic conditions could substitute for GSH in activat-
ing the cloaked agent. Therefore the reducing environment of
tumors could potentially be sufficient for activation of the
cloaked agent. However, the activation of the agent did not
occur without NADH (Figure 2D) showing that NADH was still
required. Furthermore, in reducing environments, NADH re-
mains in its active (reduced) form for a longer time that accel-
erates activation of the cloaked CEST agent. Interestingly, the
agent is sufficiently stable to tolerate such a reducing environ-
ment even in the presence of GSH. Such stability makes the
CEST agent specific for the detection of the enzyme.
Experimental Section
The cloaked CEST agent was synthesized in 8 steps (Scheme 2) ac-
cording to previously reported methodologies with some modifica-
tions.[10a,19a] We conjugated the carboxylic acid form of the quinone
4 to tBu protected DO3A-oAA compound (10) taking advantage of
the moderate reaction and flexibility of the reaction with low boil-
ing point solvents such as dichloromethane. In addition, the prod-
uct of this step was easily purified with gravity column chromatog-
raphy. The tBu groups were removed upon purification. The qui-
none moiety was sufficiently stable and tolerated harsh acidic con-
ditions during deprotection of the DO3A-oAA macrocycle. Finally,
YbIII was chelated with DO3A-oAA in water at 408C. The final prod-
uct was synthesized with an overall isolated yield of 6% and was
purified using HPLC. Different combinations of the cloaked agent,
DT-diaphorase enzyme, b-nicotinamide adenine dinucleotide, and
glutathione were incubated at 378C and their CEST signals were
measured using a 7 T Bruker Biospec MRI scanner.
We adjusted the pH of the solutions to 8.0 because previous
reports showed that this pH was suitable for increasing the
stability of NADH.[20] However, the basic pH caused a decrease
in the CEST signal of the amide group in the activated form of
the agent. A neutral pH (7.0Æ0.25) was suitable to observe
both CEST signals of the activated agent. Small 0.75 ppm shifts
in saturation frequencies of CEST signals were observed that
were due to differences in pH 8.1 and 7.2 between the sam-
ples. The pH-dependent changes in protonation of functional
groups can alter resonance frequencies of labile protons.[21] In
addition, changes in the exchange rates of labile protons affect
their resonance frequencies.
Acknowledgements
The authors would like to thank Dr. Vahe Bandarian and Micah
T. Nelp for access to their laboratory instruments and active in-
volvement in setting up enzyme reactions. The authors also
thank Dr. Indraneel Ghosh, Dr. Karla Camacho-Soto, and Dr.
Luca Ogunleye for access to their laboratory instruments and
assistance with HPLC analysis and purifications. This research
was supported by the National Institutes of Health (NIH)
through grants R01 CA169774 and P01 CA95060. K.M.J. was
supported through NIH training grants T32HL007955 and
T32HL066988.
Further evidence of activation of the cloaked CEST agent
was obtained by measuring UV/Vis absorbance of solutions
before and after reaction. A solution of NADH showed two
Chem. Eur. J. 2017, 23, 6514 –6517
www.chemeurj.org
6516
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[7] a) C. Glorieux, J. M. Sandoval, N. Dejeans, G. Ameye, H. A. Poirel, J.
Verrax, P. B. Calderon, Life Sci. 2016, 145, 57–65; b) D. Siegel, C. Yan, D.
Ross, Biochem. Pharmacol. 2012, 83, 1033–1040.
[8] a) R. M. Phillips, Cancer Chemother. Pharmacol. 2016, 77, 441–457;
b) R. M. Phillips, H. R. Hendriks, G. J. Peters, Br. J. Pharmacol. 2013, 168,
11 –18; c) S. Noura, M. Ohue, T. Shingai, S. Kano, H. Ohigashi, M. Yano,
O. Ishikawa, A. Takenaka, K. Murata, M. Kameyama, Ann. Surg. Oncol.
2011, 18, 396–404; d) R. M. Phillips, M. A. Naylor, M. Jaffar, S. W. Dough-
ty, S. A. Everett, A. G. Breen, G. A. Choudry, I. J. Stratford, J. Med. Chem.
1999, 42, 4071–4080; e) D. Siegel, H. Beall, C. Senekwitsch, M. Kasai, H.
Arai, N. W. Gibson, D. Ross, Biochemistry 1992, 31, 7879–7885; f) D.
Siegel, N. W. Gibson, P. C. Preusch, D. Ross, Cancer Res. 1990, 50, 7483–
7489.
[9] a) T. Shimizu, S. Murata, H. Sonoda, E. Mekata, H. Ohta, K. Takebayashi, T.
Miyake, T. Tani, Mol. Clin. Oncol. 2014, 2, 399–404; b) R. Glynne-Jones, L.
Kadalayil, H. M. Meadows, D. Cunningham, L. Samuel, J. I. Geh, C. Low-
dell, R. J. Beare, R. Begum, J. A. Ledermann, D. Sebag-Montefiore, Ann.
Oncol. 2014, 25, 1616–1622; c) I. Giang, E. L. Erin, L. Boland, G. M. K.
Poon, AAPS J. 2014, 16, 899–916; d) C. P. Guise, A. M. Mowday, A.
Ashoorzadeh, R. Yuan, W.-H. Lin, d.-H. Wu, J. B. Smaill, A. V. Patterson, K.
Ding, Chin. J. Cancer Res. 2014, 33, 80–86.
[10] a) L. A. Carpino, S. A. Triolo, R. A. Berglund, J. Org. Chem. 1989, 54,
3303–3310; b) K. L. Amsberry, R. T. Borchardt, Pharm. Res. 1991, 8, 323–
330; c) B. Wang, S. Liu, R. T. Borchardt, J. Org. Chem. 1995, 60, 539–543.
[11] a) C. Danforth, A. W. Nicholson, J. C. James, G. M. Loundon, J. Am. Chem.
Soc. 1976, 98, 4275–4281; b) K. L. Amsberry, R. T. Borchardt, J. Org.
Chem. 1990, 55, 5867–5877.
[12] M. N. Levine, R. T. Raines, Chem. Sci. 2012, 3, 2412–2420.
[13] S. S. Chandran, K. A. Raines, R. T. Dickson, J. Am. Chem. Soc. 2005, 127,
1652–1653.
Scheme 2. Synthesis of Yb-DO3A-oAA-TML-Q. The caboxylic acid form of
quinone was prepared in 2 steps. Then it was reacted with tert-butyl protect-
ed DO3A-oAA that was prepared in 3 steps. The resulting product was de-
protected and then used to chelate YbIII.
[14] a) S. U. Hettiarachchi, B. Prasai, R. L. McCarley, J. Am. Chem. Soc. 2014,
136, 7575–7578; b) W. C. Silvers, A. S. Payne, R. L. McCarley, Chem.
Commun. 2011, 47, 11264–11266; c) W. C. Silvers, B. Prasai, D. H. Burk,
M. L. Brown, R. L. McCarley, J. Am. Chem. Soc. 2013, 135, 309–314.
[15] P. Liu, J. Xu, D. Yan, P. Zhang, F. Zeng, B. Li, S. Wu, Chem. Commun.
2015, 51, 9567–9570.
Conflict of interest
The authors declare no conflict of interest.
[16] I. Daryaei, M. D. Pagel, Res. Rep. Nucl. Med. 2015, 5, 19–32.
[17] M. Woods, D. E. Woessner, A. D. Sherry, Chem. Soc. Rev. 2006, 35, 500–
511.
[18] a) V. R. Sheth, Y. Li, L. Q. Chen, C. M. Howison, C. A. Flask, M. D. Pagel,
Magn. Reson. Med. 2012, 67, 760–768; b) V. R. Sheth, G. Liu, Y. Li, M. D.
Pagel, Contrast Media Mol. Imaging 2012, 7, 26–34; c) G. Liu, Y. Li, V. R.
Sheth, M. D. Pagel, Mol. Imaging 2012, 11, 47–57.
[19] a) Y. Li, V. R. Sheth, G. Liu, M. D. Pagel, Contrast Media Mol. Imaging
2011, 6, 219–228; b) G. Liu, Y. Li, M. D. Pagel, Magn. Reson. Med. 2007,
58, 1249–1256.
[20] L. Rover Jr. , J. C. B. Fernandes, G. de Oliveira Neto, L. T. Kubota, E. Kate-
kawa, S. H. P. Serrano, Anal. Biochem. 1998, 260, 50–55.
Keywords: CEST MRI · dt-diaphorase · molecular imaging ·
NQO1 · trimethyl lock
[1] M. Belinsky, A. K. Jaiswal, Cancer Metastasis Rev. 1993, 12, 103–117.
[2] a) S. Chen, K. Wu, R. Knox, Free Radical Biol. Med. 2000, 29, 276–284;
b) G. Tedeschi, S. Chen, V. Massey, J. Biol. Chem. 1995, 270, 1198–1204.
[3] Y. Ma, J. Kong, G. Yan, X. Ren, D. Jin, T. Jin, L. Lin, Z. Lin, BMC Cancer
2014, 14, 414–423.
[4] a) S. Dansona, T. H. Ward, J. Butler, M. Ranson, Cancer Treat. Rev. 2004,
30, 437–449; b) S. L. Winski, Y. Koutalos, D. L. Bentley, D. Ross, Cancer
Res. 2002, 62, 1420–1424.
[21] Y. Wu, T. C. Soesbe, G. E. Kiefer, P. Zhao, A. D. Sherry, J. Am. Chem. Soc.
2010, 132, 14002–14003.
[5] a) P. L. Gutierrez, Free Radical Biol. Med. 2000, 29, 263–275; b) L. Gharat,
R. Taneja, N. Weerapreeyakul, B. Rege, J. Polli, P. J. Chikhale, Int. J.
Pharm. 2001, 219, 1–10.
[6] a) N. Sagara, M. Katoh, Cancer Res. 2000, 60, 5959–5962; b) R. J. Riley, P.
Workman, Biochem. Pharmacol. 1992, 43, 1657–1669.
Manuscript received: February 15, 2017
Accepted Article published: March 30, 2017
Final Article published: April 20, 2017
Chem. Eur. J. 2017, 23, 6514 –6517
www.chemeurj.org
6517
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim