Near-infrared photoactivatable nitric oxide donors with photoacoustic readout

Article abstract of DOI:10.1016/bs.mie.2020.05.003

In this chapter, we motivate the need for photoactivatable NO donor molecules and give a brief survey of the existing chemical tools in the field. We then provide detailed protocols for the synthesis and validation of a near-infrared light-activated NO do

Full text of DOI:10.1016/bs.mie.2020.05.003

CHAPTER SIX
Near-infrared photoactivatable
nitric oxide donors with
photoacoustic readout
Effie Y. Zhou, Hailey J. Knox, Christopher J. Reinhardt, Gina Partipilo,
Jefferson Chan*
Department of Chemistry and Beckman Institute for Advanced Science and Technology, University of Illinois
at Urbana-Champaign, Urbana, IL, United States
*
Corresponding author: e-mail address: jeffchan@illinois.edu
Contents
1
2
. Introduction
114
115
116
117
117
118
123
125
126
128
132
134
138
140
140
. Selection of photoactivatable nitric oxide releasing molecules
2
2
.1 Photolabile NO releasing moieties
.2 Principles of photoacoustic tomography
3
4
5
. Design of photoNOD-1
. Synthesis of photoNOD-1
. In vitro characterization of photoNOD-1 and rNOD-1
5
5
5
5
5
5
.1 Photophysical characterization of photoNOD-1 and rNOD-1
.2 Tissue phantom preparation
.3 Optical and photoacoustic validation of rNOD-1 release upon irradiation
.4 Validation of NO release upon irradiation of photoNOD-1 by EPR
.5 Assessing stability of photoNOD-1
.6 Validation that photoNOD-1 and rNOD-1 are non-cytotoxic
6
7
. Characterization of photoNOD-1 in vivo
6
6
.1 Identification of optimal wavelengths for monitoring photoNOD-1 in vivo
.2 Confirmation of light-dependent conversion of photoNOD-1 to rNOD-1
in vivo
141
143
146
146
6
.3 Application of photoNOD-1 to a murine tumor model
. Concluding remarks
References
Abstract
In this chapter, we motivate the need for photoactivatable NO donor molecules and
give a brief survey of the existing chemical tools in the field. We then provide detailed
protocols for the synthesis and validation of a near-infrared light-activated NO donor
molecule, photoNOD-1, developed in our research group. With this tool, NO can be
released in vivo in a radiation-dependent manner that can be monitored using photo-
acoustic imaging.
Methods in Enzymology, Volume 641
ISSN 0076-6879
https://doi.org/10.1016/bs.mie.2020.05.003
# 2020 Elsevier Inc.
All rights reserved.
1
13

114
Effie Y. Zhou et al.
1. Introduction
Nitric oxide (NO) is an endogenously produced free radical
gasotransmitter that regulates a staggering collection of physiological func-
tions. In mammals, NO is biosynthesized by various nitric oxide synthase
(NOS) enzymes as a product of arginine metabolism or generated by nitrite
reduction under hypoxic conditions (Gladwin, Grubina, & Doyle, 2009).
NO exerts diverse downstream effects through soluble guanylate cyclase
(sGC)-dependent and sGC-independent pathways. At nanomolar concen-
trations, NO can bind to the heme cofactor of sGC, which elicits a confor-
mational change and an increase catalytic activity. This in turn enhances the
conversion of GTP to cGMP. cGMP then modulates the activity of phos-
phodiesterase, cGMP-dependent protein kinases and ion-gated channels.
Together this affects platelet aggregation, smooth muscle relaxation, leuko-
cyte infiltration, neurotransmission, and a variety of other functions in a
tissue-dependent manner (Derbyshire & Marletta, 2012). NO can also react
with other reactive oxygen species (such as hydrogen peroxide and super-
oxide) or exert other sGC-independent functions via the S-nitrosylation
of cysteine residues and nitration of tyrosine residues in certain target pro-
teins. These processes are still active areas of research (Bruckdorfer, 2005;
Denninger & Marletta, 1999; Smith & Marletta, 2012). In light of its diverse
physiological roles, NO must be tightly regulated to maintain homeostasis.
Dysregulation of NO has been associated with pathologies including cardio-
vascular disease (Naseem, 2005), neurodegeneration (Duncan & Heales,
2
005), inflammation (Bogdan, 2001), reperfusion injury (Schulz, Kelm, &
Heusch, 2004), and cancer (Xu, Liu, Loizidou, Ahmed, & Charles, 2002).
The role of NO in cancer biology is especially interesting. While some
studies show NOS or NO levels correlate with tumor suppression, other
research shows NO is correlated with metastatic potential and tumor
growth. These seemingly contradictory results, which have been reviewed
in detail, are in part due to the dependence of NO biology on the concen-
tration, duration, tissue type and local microenvironment (Vanini, Kashfi, &
Nath, 2015). As a result, NO has earned a reputation as a double-edged
sword, and researchers are highly motivated to understand and precisely
manipulate NO levels in models of disease.
Several common strategies exist for the modulation of NO levels in vivo.
Many of these methods, namely knocking out or silencing NOS enzymes,
treating subjects with NOS inhibitors, or applying NO gas or NO-releasing

Near-infrared photoactivatable nitric oxide donors
115
molecules, alter systemic NO levels. While these methods can be informa-
tive in certain contexts, they cannot easily distinguish specific effects within
a given pathology from global changes in physiological NO signaling.
Therefore, there is a rising interest in the use of activatable donors that
can release NO in a controlled manner, without eliciting global side effects.
To date, activatable NO releasing molecules (NORMs) have been
developed to respond to a variety of stimuli (Wang et al., 2002). In principle,
these molecules can be administered systemically, circulate through the
subject, and release NO only upon engagement of the target stimulus.
The stimulus (for instance, a dramatic difference in pH or enzymatic activity)
must be unique to the pathological condition to avoid global side effects.
This presents researchers with the challenge of identifying a unique and suf-
ficiently abundant stimulus for each pathology of interest, then developing
NO releasing moieties that can selectivity respond to that stimulus. This
challenge motivated our group’s interest in photoactivatable NO donors.
Because light can be applied to a subject in a controlled, user-defined man-
ner, photoactivatable NO releasing molecules (photoNORMs) have great
potential to provide both temporal and spatial control over NO release.
In this chapter, we highlight the criteria for selecting a photoNORM
for a given study. We then provide a brief survey of photoactivatable
NO donor classes with a focus on their activation wavelengths, their
utility for biological experiments, and applications in fluorescence imaging.
Finally, we provide a brief primer on photoacoustic (PA) tomography and
outline a detailed guide to our group’s development of photoNOD-1, a
molecule that can release NO in the presence of near-IR light in a process
that can be monitored using PA imaging (Zhou et al., 2018).
2
. Selection of photoactivatable nitric oxide
releasing molecules
When designing or selecting a photoNORM for a given biological
application, several criteria should be considered. First, a photoNORM
must be activated by an appropriate irradiation wavelength, depending upon
the endpoint application of the photoNORM. For instance, if the NO is to
be delivered to two-dimensional cell cultures, a NORM that is labile to UV
or visible wavelengths may be sufficient. However, if an experiment requires
that a sample be irradiated for extended periods of time, or if NO needs to be
delivered to deep tissues, a NORM that is labile to longer-wavelength light

116
Effie Y. Zhou et al.
may be required to minimize phototoxicity or maximize tissue penetrance,
respectively. Second, the photoNORM should be stable to physiological
conditions such that NO is selectively released upon irradiation, avoiding
undesirable off-target effects. Third, a mechanism for validating NO release
and/or monitoring the progress of NO delivery should be incorporated.
This could be accomplished by amperometry, an NO-selective probe, or
by incorporating a release readout directly into the photoNORM.
2
.1 Photolabile NO releasing moieties
A variety of photoNORMs have been developed that feature a metal-
nitrosyl moiety as the NO source. The majority of these are transition-metal
based, especially iron, manganese, and ruthenium, which tightly bind to
NO. These metal nitrosyls are moderately labile to UV light, but tuning
of ligands or coordination of chromophores to the metal center can
enable NO release with wavelengths in the visible to near-IR. However,
many demonstrate poor stability under physiological conditions and require
nanodelivery systems to avoid off-target effects (Sortino, 2010; Xiang,
Guo, & Liu, 2017). Moreover, the examples that have successfully delivered
NO directly without encapsulation still present questions regarding the
toxicity of the metal complexes, and the complexes do not always provide
a mechanism by which NO release can be observed.
Organic photoNORMs utilize a variety of functional groups to
release NO under UV irradiation, including caged diazeniumdiolates, aryl
N-nitrosamines, and nitrobenzenes (Wang et al., 2002). Recently, these
strategies have been revamped to incorporate a fluorescent readout of NO
delivery upon activation with single photon visible light (He, Xia, et al.,
2
018; He, Ye, et al., 2018) or two-photon near-IR light (Xie et al., 2017;
Zhang et al., 2016). The incorporation of a fluorescence readout makes
these molecules excellent tools for studying NO biology in cell culture,
without the need of an NO electrode or NO-sensitive fluorescent dye to
confirm NO release. These tool compounds are therefore excellent for the
delivery of NO to cell cultures and shallow tissues. On the other hand,
the poor tissue penetrance of visible light and even two-photon fluorescence
methods (which can image tissues at a maximum depth of 1.5mm) (Wang &
Yao, 2016) limit their utility in complex, three-dimensional in vivo settings.
This motivated our group’s interest in developing a photoNORM that can
effectively deliver NO to deeper tissues while still maintaining a readout
of release.

Near-infrared photoactivatable nitric oxide donors
117
2.2 Principles of photoacoustic tomography
Photoacoustic (PA) tomography is an imaging technology that interrogates
a specimen with innocuous near-IR excitation light and elucidates the fea-
tures of the specimen via ultrasonic detection. Unlike fluorescence methods,
which rely upon emitted photons and are therefore depth-limited by both
the excitation and emission wavelengths, PA imaging relies on heat from
non-radiative relaxation pathways to afford thermoelastic expansions which
propagate through the specimen as ultrasound. Since ultrasound is signifi-
cantly more tissue penetrant than light, this means that the depth of PA
imaging is primarily limited by the tissue penetrance of the near-IR excita-
tion light. Practically, this means that PA imaging can yield high resolution
results at depths up to 8–10cm in biological tissues (Reinhardt & Chan,
2018; Wang & Yao, 2016).
3
. Design of photoNOD-1
photoNOD-1 consists of an NO-releasing moiety (an aryl N-nitroso)
appended to an aza-BODIPY core. The aza-BODIPY dye, which exhibits
4
ꢀ1
ꢀ1
a large extinction coefficient (>10 M cm ) at near-IR wavelengths,
absorbs energy upon irradiation to facilitate homolytic cleavage of the
N-nitroso bond. This yields an aniline-containing dye product (rNOD-1)
and the desired NO radical as well as a significant shift in the absorbance
wavelength of the aza-BODIPY dye. This wavelength shift enables simul-
taneous PA detection of photoNOD-1 and rNOD-1 and enables monitor-
ing of NO release (Scheme 1).
Scheme 1 photoNOD-1 exhibits a strong PA signal upon excitation at λblue, with neg-
ligible PA signal upon excitation at λred. Continuous irradiation at λblue drives the release
of NO, generating rNOD-1 as a byproduct. rNOD-1 exhibits a small PA signal upon
excitation at λblue and a large PA signal at λred. Therefore, the release of NO can be
monitored by the changes in PA signal at λblue and λred
.

118
Effie Y. Zhou et al.
4. Synthesis of photoNOD-1
rNOD-1 can be synthesized in six steps from 4-aminoacetophenone
and 4-methoxyacetophenone. One additional nitrosation step is required to
generate photoNOD-1 from rNOD-1 (Scheme 2). The following synthetic
Scheme 2 Synthetic route to rNOD-1 and photoNOD-1.

Near-infrared photoactivatable nitric oxide donors
119
procedures require understanding of standard organic chemistry techniques
and the availability of basic synthetic laboratory equipment. All running
solvents are defined as percent volume to volume.
tert-Butyl (4-acetylphenyl)carbamate (1)
1
. Preheat an oil bath to 100°C.
. Add 4 -aminoacetophenone (6.7g, 50mmol, 1.0equiv.) and di-tert-
0
2
butyl dicarbonate (13g, 60mmol, 1.2equiv.) to a 250mL round
bottom flask equipped with a magnetic stir bar and a rubber septum.
. Flush the reaction vessel with nitrogen.
3
4
. Measure 1,4-dioxane (60mL) using a graduated cylinder. Briefly
remove the rubber septum from the reaction vessel, pour the solvent
directly onto the combined solids with continuous magnetic stirring,
and quickly cap the reaction mixture with the septum.
. Heat the reaction vessel in an oil bath for 8.5h, or until the reaction is
complete by thin layer chromatography (TLC) (25% ethyl acetate in
hexanes).
5
6
. Concentrate the reaction mixture by rotary evaporation to isolate the
crude product as a solid. If the crude product does not readily solidify,
allow the oil to stand on the benchtop until it crystallizes to a solid. This
process can be accelerated by gently scratching the sides of the flask with
a spatula or glass rod.
7
8
. NOTE: Steps 6–9 are optional. Alternatively, the crude product can be
used directly in the next reaction.
. Suspend the solids in a small volume of 20–25% ethyl acetate in hexanes
and isolate the first batch of solid product via vacuum filtration.
Confirm the purity of the product by proton NMR (CDCl3).
. Isolate the filtrate from Step 8. Concentrate the filtrate by rotary
evaporation.
9
1
0. Repeat steps 8–9 until no more pure batches of product (>95% purity
by NMR) can be obtained. NOTE: We obtained 86% yield in two
batches.
tert-butyl (4-acetylphenyl)(methyl)carbamate (2)
1
. Add tert-butyl (4-acetylphenyl)carbamate (1, 3.0g, 12.8mmol, 1.0equiv.)
to a 250mL round bottom flask equipped with a magnetic stir bar and a
rubber septum.
2
. Flush the reaction vessel with nitrogen. Maintain a gentle flow of nitro-
gen through the reaction vessel using a nitrogen line and a vent needle.
. Add anhydrous THF (40mL) and stir continuously at room tempera-
ture until the solids are completely dissolved.
3

120
Effie Y. Zhou et al.
4
5
. Cool the reaction mixture to 0°C on an ice bath.
. Add sodium hydride (60wt% dispersion in mineral oil) (0.61g,
1
1
5.3mmol, 1.2equiv.) in five equal portions over the course of
0min, briefly removing the rubber septum for each addition.
6
7
8
. Continue stirring the suspension for an additional 20min at this
temperature.
. Add methyl iodide (1.57mL, 31.9mmol, 2.5equiv.) in one portion via
syringe.
. Warm the reaction mixture to room temperature. Continue stirring for
an additional 40min, or until reaction is complete by TLC (15% ethyl
acetate in hexanes).
9
. Completely remove the solvent by rotary evaporation to isolate the
crude alkylated product.
1
1
0. Transfer the crude residue to a separatory funnel with ethyl acetate and
water. Allow the layers to separate. Isolate the organic layer and wash
vigorously with an equal volume of brine (3ꢁ).
1. Dry the organic layer over sodium sulfate and concentrate by rotary
evaporation to isolate the alkylated product as a yellow oil in quantita-
tive yield. Use directly without purification.
tert-Butyl methyl (4-(4-Nitro-3-phenylbutanoyl)phenyl)
carbamate (3)
1
2
3
. Add tert-butyl (4-acetylphenyl)(methyl)carbamate (2, 0.59g,
.38mmol, 1.0equiv.) and ethanol (10mL) to a 50mL round bottom
2
flask equipped with a magnetic stir bar.
. Add benzaldehyde (0.27mL, 2.62mmol, 1.1equiv.) and potassium
hydroxide (10M aqueous solution, 0.72mL, 7.15mmol, 3.0equiv.)
via syringe. Stir at room temperature for 2.5h.
. Add nitromethane via syringe (1.90mL, 35.7mmol, 15equiv.) and
stir the reaction mixture at room temperature for 10.5h. NOTE:
Over the course of the reaction, the solution should turn dark brown
in color.
4
. Concentrate the reaction mixture by rotary evaporation to remove
most of the ethanol. Dilute the crude mixture in brine and extract with
dichloromethane (3ꢁ).
5
6
. Dry the combined organic extracts over sodium sulfate and concentrate
to isolate the crude product.
. Dry load the crude product onto celite and purify the product by flash
column chromatography (25% ethyl acetate in hexanes). This will
afford the product as a yellow oil (0.47g, 1.4mmol, 59% yield).
NOTE: Compounds 3 and 4 can be stored under ambient conditions.

Near-infrared photoactivatable nitric oxide donors
121
However, the products may slowly decompose over time, generating a
dark brown color. While older batches may be used, repurification may
be necessary if low yields are observed in later synthetic steps.
-(4-Methoxyphenyl)-4-nitro-3-phenylbutan-1-one (4)
1
1
. Add 4-methoxyacetophenone (10g, 67mmol, 1.0equiv.) and ethanol
130mL) to a 500mL round bottom flask equipped with a magnetic stir
(
bar. Stir until the solids are fully dissolved.
2
3
. Add benzaldehyde (6.8mL, 62mmol, 0.9equiv.) via syringe.
. Cool the reaction mixture to 0°C and add potassium hydroxide
(10M aqueous solution, 21mL, 207mmol, 3equiv.).
4
. Allow the reaction mixture to warm to room temperature and stir
for 1.5h, yielding a white precipitate, until the reaction is complete
by TLC.
5
6
7
. Isolate the white solids by vacuum filtration to obtain the intermediate
chalcone in 87% yield.
. Resuspend the intermediate (4.8g, 20mmol, 1equiv.) in ethanol
(35mL) in a 100mL round bottom flask equipped with a stir bar.
. Add nitromethane (13.6mL, 254mmol, 13equiv.) and potassium
hydroxide (10M aqueous solution, 0.34mL, 3.4mmol, 0.2equiv.)
and stir the reaction at room temperature for 3h, or until the reaction
is complete by TLC.
8
9
. Concentrate the reaction mixture by rotary evaporation to remove
most of the ethanol.
. Dilute the crude mixture in saturated aqueous ammonium chloride and
extract with ethyl acetate (3ꢁ).
1
0. Dry the organic layer over sodium sulfate and concentrate to
isolate the crude product as a white solid in 91% yield (79% from
4
-methoxyacetophenone). Use without purification.
tert-Butyl (Z)-(4-(5-((5-(4-Methoxyphenyl)-3-phenyl-2H-pyrrol-
-ylidene)amino)-4-phenyl-1H-pyrrol-2-yl)phenyl)(methyl)-
carbamate (5)
2
1
2
. Preheat an oil bath to 60°C.
. Add3(475mg,1.2mmol, 1.0equiv.)and4(718mg,2.4mmol,2.0equiv.)
to a 50mL round bottom flask equipped with a magnetic stir bar.
. Add ethanol (36mL) and heat the reaction mixture at 60°C until the
solids are completely dissolved. NOTE: This reaction generates a
statistical mixture of the desired heterodimeric product with the two
undesired homodimeric products. Therefore, it is critical to dissolve
3
3
and 4 in the solvent prior to adding ammonium acetate to maximize
the potential for heterodimer formation.

122
Effie Y. Zhou et al.
4
5
6
7
. Add ammonium acetate (1.38g, 18mmol, 15equiv.) to the reaction
mixture and attach a reflux condenser to the reaction vessel.
. Increase the temperature of the oil bath to 90°C. Reflux the reaction
mixture for 15h.
. Cool the reaction mixture to room temperature and isolate the precip-
itated solids by gravity filtration.
. Dry load the crude product onto celite and purify the crude product
by flash column chromatography, gradually increasing the polarity of
the running solvent from 100% dichloromethane to 0.5% methanol
in dichloromethane to isolate a blue-green solid in 25% yield.
rNOD-1
1
2
. Add 5 (100mg, 0.166mmol, 1.0equiv.) to a 100mL round bottom flask
equipped with a magnetic stir bar.
. Add dichloromethane (10mL) and triethylamine (0.333mL, 2.48mmol,
15equiv.) to the reaction vessel to dissolve 5. Cap the reaction vessel with
a rubber septum and flush the reaction vessel with nitrogen.
. Add boron trifluoride diethyl etherate (0.30mL, 2.48mmol, 15equiv.)
dropwise to the rapidly stirring reaction mixture using a syringe. Stir for
an additional 15h. NOTE: All materials that encounter this reagent
should be quenched with bicarbonate solution before disposal.
. Dilute the reaction mixture with dichloromethane and carefully add a
saturated aqueous solution of sodium bicarbonate to quench the resid-
ual boron trifluoride. NOTE: Great care should be taken to avoid
splashing upon quenching the excess boron trifluoride.
3
4
5
. Pour the quenched reaction into a separatory funnel and allow the
phases to separate. Isolate the organic layer, then wash with additional
sodium bicarbonate solution (2ꢁ).
6
7
. Isolate the organic layer, dry over sodium sulfate, and concentrate to a
dark residue.
. Dry load the crude product onto celite and purify by flash column
chromatography, gradually increasing the polarity of the running sol-
vent from 30% to 50% ethyl acetate in hexanes to afford the product
as a red-brown solid (79% yield).
photoNOD-1
1
2
. Add rNOD-1 (72.7mg, 0.130mmol, 1.0equiv.) to a 25mL round
bottom flask equipped with a magnetic stir bar.
. Add tetrahydrofuran (8.7mL) and dichloromethane (4.35mL) to
dissolve the solid residue.

Near-infrared photoactivatable nitric oxide donors
123
3
4
. Add acetic acid (4.35mL) and sodium nitrite (44.8mg, 0.653mmol,
.0equiv.) and stir the reaction mixture at room temperature for 2h,
5
or until the reaction appears complete by TLC (80% dichloromethane
in hexanes).
. Dilutethereactionmixturewithadditionaldichloromethane,thenquench
the acetic acid by slowly adding a saturated solution of sodium bicarbonate
until gas evolution ceases. NOTE: Quenching the reaction too quickly
may result in rapid bubbling, causing loss of product if the reaction vessel
overflows. To avoid this, first transfer the reaction mixture to a larger flask,
or place the reaction vessel in a secondary container before quenching.
. Transfer the quenched reaction mixture to a separatory funnel to isolate
the organic layer. Dry the organic layer over sodium sulfate and con-
centrate to isolate the crude product.
5
6
. Purify the crude product by flash column chromatography using 80%
dichloromethane in hexanes as the running solvent. This will yield the
product as a green solid (48% yield). NOTE: Low yields are primarily
due to the decomposition of photoNOD-1 to rNOD-1 on silica.
However, rNOD-1 can isolated from the column and resubjected
to the reaction conditions if needed. photoNOD-1 can be stored as
a dry solid at or below ꢀ20°C in the dark for several months without
noticeable decomposition.
5
. In vitro characterization of photoNOD-1 and rNOD-1
Upon completion of photoNORM synthesis, it is necessary to
i) assess the photophysical properties of both the photoNORM and its
(
proposed turnover product, (ii) demonstrate that the compound generates
the proposed turnover product upon irradiation with near-IR light with
concomitant release of NO, and (iii) confirm that the photoNORM is stable
in the absence of irradiation. To expedite our experiments and minimize the
error involved in weighing small amounts of compound for each experiment,
we recommend aliquoting compounds in advance, as described below.
Equipment
1
3
3
1
3
—Microbalance
—Dram vials per compound (1mL capacity)
6—PCR tubes per compound (0.2mL capacity)
—Rack for storing PCR tubes
—Micropipettes and tips (P1000, P200, and P10)

124
Effie Y. Zhou et al.
Reagents
Compound of interest (ꢂ3mg)
CHCl , or other organic solvent in which the compound is soluble
3
Procedure (makes three batches of 12 aliquots)
1
2
. Label the dram vials (A, B, and C).
. Place three dozen PCR tubes in the tube rack, labeling each set of a
dozen tubes with (A, B, or C)
3
4
. Zero the microbalance and acquire the tare weight of each dram vial.
. Add a small amount of solid compound to each dram vial (1–2mg
each). Note the exact mass of solid added to vials A, B, and C.
. Based on the molecular weight of the compound, calculate the volume
of solvent required to generate a 5mM solution.
5
6
7
8
. Transfer the dram vials to an ice bath (to prevent solvent evaporation)
and add the appropriate amount of CHCl to each.
3
. Cap each dram vial and shake vigorously or sonicate until the solids are
fully dissolved. Return to the ice bath to prevent solvent evaporation.
. Quickly pipet 20μL of solution into each PCR tube. Be careful to
match the dram vial label with the PCR tube labels, to obtain a dozen
aliquots from each vial (A, B, and C).
9
. Remove the solvent completely by storing the racks of PCR tubes in a
dark vacuum dessicator. NOTE: If the organic solvent is chloroform,
the aliquots should dry within 2–3h under these conditions. However,
removal of higher boiling solvents such as DMF and DMSO may take
several days.
1
1
0. Reserve an aliquot each from batches A, B, and C, and determine the
extinction coefficient of the compound. This process is described in
Section 5.1.
1. Store the remaining dried aliquots in a foil-covered vessel at or below
ꢀ
20°C.
When preparing for any experiments that require compound of interest,
warm the required number of compound aliquots to room temperature.
Redissolve in DMF or DMSO (20μL of solvent should yield approximately
5
mM compound), then check the true concentration of the aliquot(s) using
the following procedure:
1
2
. Add 999μL of chloroform to a quartz cuvette and baseline the
spectrophotometer.
. Add 1μL of the compound stock solution. Pipet up and down to mix
thoroughly.

Near-infrared photoactivatable nitric oxide donors
125
3
4
. Scan the sample in the spectrophotometer to identify the absorbance
value at 681nm (the λmax of the dye).
. Using the Beer-Lambert law, determine the true concentration of the
stock solution based upon the extinction coefficient of the dye deter-
mined in Section 5.1. NOTE: Using the extinction coefficient to
back-calculate the true concentration ensures that slight differences in
compound concentration between batches (A, B, and C) will not affect
the experimental outcomes. Moreover, the concentration of dye can be
precisely reproduced for every experiment.
5
. Adjust the solution to the desired concentration by diluting in additional
solvent, or by reconstituting additional aliquots, and verifying the
concentration by repeating steps 1–4.
5.1 Photophysical characterization of photoNOD-1 and rNOD-1
This experiment identifies each compound’s wavelength of maximum
absorption and the compound’s extinction coefficient at that wavelength.
This information is vital, as it allows the user to reliably check the con-
centration of a compound stock solution prior to all future experiments.
Below, we describe our characterization of photoNOD-1 and rNOD-1
as an example. The properties of these two compounds are described in
Table 1.
Equipment
1
1
3
—UV–visible range spectrophotometer
—Quartz cuvettes with corresponding cuvette caps
—Micropipettes and tips (P10, P200, P1000)
Reagents
3
1
aliquots—photoNOD-1
aliquot—rNOD-1
Organic solvents: CHCl , DMF
3
Aqueous buffer: 50% ethanol, 20mM HEPES, adjusted to pH 7.4
Table 1 Summary of photophysical properties.
2
1
21
Compound
ε (M cm ) (CHCl
3
)
λ
max (CHCl
3
) (nm)
λmax (aqueous) (nm)
4
photoNOD-1 5.59ꢁ10
681
733
677
746
4
rNOD-1
6.34ꢁ10

126
Effie Y. Zhou et al.
Procedure
1
. Warm up the spectrophotometer at least 15min in advance of the
experiment.
2
. Setup the experiment to collect absorbance spectra from 400 to
900nm, at 0.5nm increments.
3
4
. Meanwhile, prepare a stock solution of dye.
. Add a known volume of CHCl or aqueous buffer to a quartz cuvette
3
(0.5–1mL) and cap the cuvette to minimize solvent evaporation.
5
6
. Blank the spectrophotometer using the baseline function.
. Briefly remove the cuvette cap and titrate in the compound stock solu-
tion, taking note of the volume added so that the dye concentration can
be calculated.
7
. Mix thoroughly by pipetting up and down 10 times. Re-cap the
cuvette.
8
9
. Collect an absorbance spectrum.
. Continue steps 5–7 until at least five scans have been collected in which
the peak absorbance value falls between 0.1 and 1.0a.u.
1
1
0. Repeat steps 2–8 for each batch of compound (A, B, and C) to obtain a
total of 15 data points.
1. Identify the wavelength of maximum absorbance for the selected com-
pound. Generate a linear regression for absorbance vs. concentration
and use the Beer-Lambert Law to determine the extinction coefficient
of the compound of interest.
5
.2 Tissue phantom preparation
The purpose of a tissue phantom is to provide a medium through which
incident irradiation light will be scattered. This phantom benefits in vitro
PA studies in two ways. First, it mimics the light scattering that occurs in
tissues. Second, it enables irradiation of a larger region of the sample, rather
than irradiating just the portion of the sample that is aligned with the laser.
Equipment
1
1
1
1
1
—Microwave (standard food-grade)
—Glass beaker (250mL capacity)
—Spatula
pair—Heat resistant gloves (optional)
—Tissue phantom mold (custom) equipped with two pieces of
hypodermic steel tubing.

Near-infrared photoactivatable nitric oxide donors
127
NOTE: In the absence of a custom tissue phantom mold, 50mL falcon
tube equipped with the hypodermic steel tubing can serve as an adequate
substitute.
Reagents
Deionized water (78mL)
Agarose LE (Molecular Biology Grade, Gold Bio) (4g)
2% fat milk (2mL)
Procedure
1
2
. Weigh out agarose LE (4g) into a 250mL beaker.
. Add deionized water (78mL) and 2% milk (2mL) and stir vigorously
with a spatula to suspend the solids.
3
. Place the beaker into a standard microwave and heat for 45s or until a
viscous, translucent gel is formed, pausing briefly at 30s to stir. NOTE:
Microwave strengths vary, so watch carefully and pause the heating as
needed to prevent excessive bubbling or overflow.
4
5
. Remove the beaker from the microwave and stir vigorously to ensure
an even consistency. NOTE: Beaker will be very hot; use heat resistant
gloves if needed.
. Immediately pour the hot gel into the custom Teflon mold, completely
submerging the steel tubing. Tap the mold vigorously on the benchtop
to displace any bubbles in the gel.
6
7
. Cool the gel in the Teflon mold for at least 2h at 4°C.
. Carefully remove the steel tubes, then turn the Teflon mold upside-
down and gently tap on the benchtop to remove the solidified phantom
(see Fig. 1). NOTE: Phantoms can be used immediately, or stored sub-
merged in deionized water for up to 5 days at 4°C. The phantom may
be prepared up to 5 days in advance, but at the minimum the phantom
should be poured at least 2h prior to the start of this experiment.
Alternatively, to make a phantom using a 50mL falcon tube, follow the
following procedure (Scheme 3):
1
. Drill two parallel 0.08in. channels through the diameter of the
falcon tube.
2
3
4
5
6
. Thread the channels with hypodermic steel tubing (or FEP tubing).
. Pour the hot gel into the mold, completely submerging the tubing.
. Tap the mold vigorously to remove bubbles.
. Cool the gel for at least 4h at 4°C.
. Carefully remove the tubing, then invert the falcon tube and gently
squeeze the sides to remove the tissue phantom from the tube.

128
Effie Y. Zhou et al.
Fig. 1 (Left) Tissue phantom after cooling in the mold. (Center) Phantom after removing
hypodermic tubing and separating from the mold. (Right) Representative images phan-
toms loaded with FEP tubes containing a sample. Reproduced with permission from ACS
(
Fathi, P., Knox H. J., Sar, D., Tripathi, I., Ostadhossein, F., Misra, S. K., Esch, M. B., Chan, J.,
Pan, D. (2019). Biodegradable biliverdin nanoparticles for efficient photoacoustic imaging.
ACS Nano, 13 (7), 7690–7704).
Scheme 3 Process for making a tissue phantom without a custom mold. (A) Drill 0.08in.
diameter parallel holes through a 50mL falcon tube, approximately 1cm apart. Insert
hypodermic steel tubing or FEP tubing. (B) Pour in prepared tissue mimicking phantom
solution and allow to set at 4°C. (C) Remove the tubing from the mold, then remove the
tissue phantoms from the tubes. Cut the phantom into a rectangular prism, such that
1
cm of phantom remains above and below the channels in the phantom.
7
. Cut the phantom into a rectangular prism using a razor blade. The final
phantom should be a rectangular prism with 1cm of phantom material
on either side of the channels that will contain the FEP sample tubes.
5
.3 Optical and photoacoustic validation of rNOD-1 release
upon irradiation
In order to monitor the release of NO from a photoNORM using an
imaging modality, it is necessary to detect a spectral change during release.

Near-infrared photoactivatable nitric oxide donors
129
Here, we monitor the loss of photoNOD-1 and the formation of rNOD-1
by both absorbance and PA imaging.
Equipment
1
1
1
2
4
2
—Photoacoustic tomographer (Nexus 128, Endra Life Sciences)
—UV–visible spectrophotometer
—Tissue-mimicking phantom (prepared as in Section 5.2)
—FEP tubes: FEP tubing (0.08in. diameter, 5cm length)
—FEP caps: FEP tubing (0.12in. diameter, 0.25cm length)
—Quartz cuvettes (0.5mL capacity)
Reagents
Organic solvents: CHCl , DMF
3
photoNOD-1 (5mM in DMF)
milli-Q water
Procedure
1
2
3
4
5
. Warm up the photoacoustic tomographer at least 30min in advance of
the experiment.
. Warm up the spectrophotometer at least 15min in advance of the
experiment.
. Prepare a 5mM stock solution of photoNOD-1 in DMF. Verify the
concentration.
. Using the DMF stock, prepare a 10μM solution of photoNOD-1
in CHCl and acquire an initial absorbance spectrum (400–900nm).
3
. Load the photoNOD-1 solution into the tissue mimicking phantom:
a. Using a P200 micropipette, carefully transfer 200μL of the
photoNOD-1 solution into an FEP tube.
b. Seal the tube on end by folding the edge of the tube upon itself and
securing the crease with an FEP cap.
c. Thread the uncapped edge of the FEP tube through the phantom.
d. Seal the uncapped edge as described in step 5b.
e. Repeat steps 5a–5d with the second FEP tube so that the phantom
is loaded with two identical solutions of dye (see Fig. 1).
. Place the phantom into the bowl of the photoacoustic tomographer.
Add enough deionized (Milli-Q) water to the bowl to just cover the
phantom. Optional: Place a small weight on top of the phantom to
prevent it from floating in the bowl.
6
7
. Acquire PA measurements using the schedule shown in Table 2.
a. Recall that in chloroform, photoNOD-1 can be detected at
λblue ¼680nm and rNOD-1 can be detected at λ ¼730nm.
red
b. Recall that irradiating photoNOD-1 at λ
NO and the formation of rNOD-1.
drives the release of
blue

130
Effie Y. Zhou et al.
Table 2 Experimental schedule for in vitro irradiations using the photoacoustic
tomographer
Rotation
time (s)
Excitation
OR [angles/ Irradiation
Cumulative
Timepoint Scan wavelengths pulses per
time at 680nm irradiation time
(
min)
mode (nm)
angle]
(min)
+0.2
+0.3
+0.2
+0.3
+0.2
+1.8
+0.2
+1.8
+0.2
+4.8
+0.2
+4.8
+0.2
+4.8
+0.2
+4.8
+0.2
+4.8
+0.2
+4.8
+0.2
+4.8
+0.2
at 680nm
0
0
1
3
5
1
1
2
2
3
3
4
C
730, 680
6
0.2
C
680
6
0.5
.5
C
730, 680
680
6
0.7
C
6
1.0
C
730, 680
680
6
1.2
C
9
3.0
C
730, 680
680
6
3.2
C
9
5.0
C
730, 680
680
6
5.2
SS
C
[120/39]
10.0
10.2
15
0
5
0
5
0
5
0
730, 680
680
6
SS
C
[120/39]
730, 680
680
6
15.2
20
SS
C
[120/39]
730, 680
680
6
20.2
25
SS
C
[120/39]
730, 680
680
6
25.2
30
SS
C
[120/39]
730, 680
680
6
30.2
35
SS
C
[120/39]
730, 680
680
6
35.2
40
SS
C
[120/39]
6
730, 680
40.2
C, continuous mode; SS, Step-and-Shoot mode. Gray rows indicate short scans which are used to
collect PA images of the samples contained in the tissue phantoms. Collect images at λred 730nm first
(
corresponding to the detection of rNOD-1) followed by 680nm (corresponding to the detection
and irradiation of photoNOD-1). White rows indicate scans used to irradiate the samples for extended
periods of time.

Near-infrared photoactivatable nitric oxide donors
131
8
. After the 40min timepoint is complete, remove the samples from the
tissue phantom:
a. Remove one FEP cap from each FEP tube.
b. Gently unfold the exposed crease in the FEP tube and carefully slide
the tube from the phantom. Take care not to damage the phantom
in the process. NOTE: The creased end of the FEP tube can also
be cut off with a pair of sharp scissors or a razor blade, if more
convenient.
9
. Combine the contents of both FEP tubes in an empty quartz cuvette
and acquire a final absorbance spectrum (400–900nm).
1
1
0. Approximate the quantification of NO release upon irradiation by
matching the absorbance spectrum from step 10 with a linear combi-
nation of the spectra for photoNOD-1 and rNOD-1 (see Fig. 2).
1. Repeat steps 5–10 twice more in order to collect three datapoints for
each timepoint.
Fig. 2 (Left) Representative quantification of NO release after irradiation of photoNOD-1
in chloroform. A linear combination of the spectra of pure rNOD-1 (red) and pure
photoNOD-1 (blue) yield a calculated spectrum (black) that matches the spectrum of
the irradiated photoNOD-1 sample (purple). The concentration of rNOD can be app-
roximated using the intensity of the red line with the extinction coefficient of rNOD.
(
Right) The conversion of photoNOD-1 to rNOD-1 is dependent on the irradiation time
at 680nm. Reproduced with permission from ACS (Zhou, E. Y., Knox, H. J., Reinhardt, C. J.,
Partipilo, G., Nilges M. J., & Chan, J. (2018). Near-infrared photoactivatable nitric oxide
donors with integrated photoacoustic monitoring. Journal of the American Chemical
Society, 140, 11686–11697. https: // doi.org/10.1021/jacs.8b05514).

132
Effie Y. Zhou et al.
1
2. Analyze the PA data:
a. Reconstruct the PA images for each timepoint.
b. Quantify the average PA signal of both FEP tubes at each wave-
length for each timepoint.
c. Plot the relative PA signal (PA at λred divided by PA at λblue) as a
function of irradiation time (see Fig. 2).
5
.4 Validation of NO release upon irradiation of photoNOD-1
by EPR
For a given photoNORM, it is critical to confirm that NO is in fact
+
the species released upon irradiation (vs. nitrosonium [NO ] or nitroxyl
[HNO]). One of the most effective techniques for this determination is
EPR spectroscopy, which enables the detection of paramagnetic centers.
Here we describe the trapping and EPR detection of NO upon photorelease
from photoNOD-1.
Equipment
1
1
1
—Photoacoustic tomographer (Nexus 128, Endra Life Sciences)
—X-band Continuous Wave EPR spectrometer
—Schlenk line equipped with nitrogen gas and high vacuum
capabilities
2
1
1
1
1
1—Hematocrit capillary tubes
—Capillary tube sealing tray with wax plate
—Quartz EPR tube
2—Septa capped dram vials (4mL capacity)
—Microsyringe (50μL)
Reagents
photoNOD-1 (5mM in DMF)
MAHMA-NONOate (a commercially available NO donor that releases
equiv. NO)
2
Potassium hydroxide (10mM in water)
FeSO ꢃ7H O
4
2
N-dithiocarbamoyl-N-methyl-D-glucamine (MGD)
Deionized water
DMF
Procedure
1
. Warm up the photoacoustic tomographer for at least 30min prior to
starting the experiment. This ensures adequate heating of the water
bath and optical system.
2
. Degas septum-capped vials of water, DMF, and 10mM KOH solution
by bubbling nitrogen vigorously through each solution for at least

Near-infrared photoactivatable nitric oxide donors
133
30min at room temperature. NOTE: Great care must be taken to elim-
inate all sources of oxygen from this experiment until the irradiation
of the samples has been completed. Samples should also be protected
from light throughout this experiment.
3
4
. Prepare a 5mM stock solution of photoNOD-1 in degassed DMF.
. Prepare solutions of iron sulfate heptahydrate (8mM) and MGD
(40mM):
a. Transfer solid iron sulfate heptahydrate or solid MGD to a septum-
capped dram vial.
b. Evacuate the vial under high vacuum.
c. Flush the vial with nitrogen gas to remove any remaining oxygen.
d. Add degassed water through the septum
5
6
. Evacuate three 4mL septum capped dram vials under high vacuum and
flush with nitrogen.
. In the indicated order, add following volumes of degassed water,
degassed DMF, aqueous iron sulfate heptahydrate, aqueous MGD
solution, and photoNOD-1 solution to each of three dram vials.
This yields a 1:1 solution of water and DMF containing 0.8mM
Fe(II), 4.0mM MGD, and 0.2mM photoNOD-1. NOTE: The iron
sulfate heptahydrate and MGD generate the NO spin-trapping
complex, Fe-(MGD) , in situ.
2
a. Water, 120μL
b. DMF, 184μL
c. 8mM Fe(II), 40μL
d. 40mM MGD, 40μL
e. 5mM photoNOD-1, 16μL
7
. Irradiate each vial containing photoNOD-1 and Fe-(MGD) in the
2
photoacoustic tomographer at λblue (680nm) for 0, 5, or 40min using
step-and-shoot mode. NOTE: We use the photoacoustic tomographer
as the source of excitation light. In principle, alternative sources of
continuous 680nm excitation light may be used as a substitute.
. After the irradiations are complete, open the vial and place a hematocrit
capillary tube inside. Allow the tube to fill by capillary action.
. Seal the capillary tube on each side with the wax tube sealant.
8
9
1
1
0. Place the capillary tube into the quartz EPR tube to acquire the EPR
spectra.
1. Acquire the EPR spectra at room temperature on the EPR spectrom-
eter using the following parameters:
a. Modulation frequency: 100kHz
b. Modulation amplitude: 2.5G

134
Effie Y. Zhou et al.
c. Gain: 4000
d. Sweep width: 100G
e. Scan time: 0.5min per scan, six scans
f. Microwave power: 12dB
g. Microwave frequency: 9.30GHz
1
2. Generate a calibration curve to quantify NO release using MAHMA-
NONOate:
a. Prepare a range of MAHMA-NONOate stock solutions (0, 2, 4,
and 8mM) in degassed 10mM KOH.
b. Repeat steps 5–12, substituting the various MAHMA-NONOate
solutions in the place of photoNOD-1, to generate 0, 50, 100, and
2
00μM solutions of MAHMA-NONOate in the presence of
Fe-(MGD)2.
c. Incubate the samples in the dark at room temperature for an hour
(
>20 half-lives, t1/2 ¼3min at room temperature) to ensure com-
plete NO release.
d. Acquire the EPR spectrum of each sample as described in step 11.
e. Plot the EPR intensity (peak height) against the concentration of
NO released (see Fig. 3).
1
3. Determine the concentration of NO released by the irradiation of
photoNOD-1 using the calibration curve generated in step 12.
5
.5 Assessing stability of photoNOD-1
To ensure that NO will not be released from a photoNORM in the absence
of irradiation, it is necessary to verify the stability of the photoNORM under
conditions that could encourage transnitrosylation (for instance, to glutathi-
one, forming S-nitrosoglutathione), denitrosylation, and denitrosation. In
this experiment, we treat 5μM photoNOD-1 with 1mM reduced glutathi-
one, 20μM metal ions (including copper (I) and (II), iron (II) and (III)),
rat liver microsomes with high CYP450 activity, or 20% human plasma
in aqueous buffers at 37°C and show that photoNOD-1 is stable to these
conditions. This experimental procedure can be modified to screen for a
variety of other analytes.
Equipment
1
2
1
1
—UV–visible spectrophotometer
1–1.6mL Eppendorf tubes
—floating foam tube rack
—Nitrogen tank or Schlenk line equipped with nitrogen gas

Fig. 3 (Left) EPR spectra of spin-trapped NO released from MAHMA-NONOate. (Center) Calibration curve correlating EPR intensity to con-
centration of NO released, assuming each molecule of MAHMA-NONOate releases 2 equiv. of NO. (Right) EPR spectra of spin-trapped NO
released from photoNOD-1 upon irradiation for 0, 5, or 40min at 680nm. Reproduced with permission from ACS (Zhou, E. Y., Knox, H. J.,
Reinhardt, C. J., Partipilo, G., Nilges M. J., & Chan, J. (2018). Near-infrared photoactivatable nitric oxide donors with integrated photoacoustic mon-
itoring. Journal of the American Chemical Society, 140, 11686–11697. https:/doi.org/10.1021/jacs.8b05514).

136
Effie Y. Zhou et al.
1
4
4
3
2
—Water bath, preheated to 37°C
—Quartz cuvettes, 1.5mL capacity
—Small rubber septa to cap the quartz cuvettes
—Micropipettes and tips (P10, P200, and P1000)
—Microsyringes (10, 50μL)
Reagents
Assay buffer (50% ethanol, 20mM HEPES, pH 7.4)
Phosphate buffered saline (PBS)
photoNOD-1 (50μL, 5mM in DMF)
Rat liver microsomes (pooled from male Sprague-Dawley rats, Sigma
Aldrich) (50μL)
NADPH (50μL, 2mM in water)
Human plasma
Reduced glutathione (50μL, 100mM in water)
CuCl (50μL, 2mM in water)
2
FeCl (50μL, 2mM in water)
3
[
(CH CN) Cu]PF (50μL, 2mM in degassed water)
3
4
6
FeSO 7H O (50μL, 2mM in degassed water)
4
2
Procedure
1
2
3
4
5
. Warm up the spectrophotometer at least 15min in advance of the
experiment.
. Degas a portion of assay buffer by gently bubbling nitrogen through the
solution for 30min.
. Degas water by gently bubbling nitrogen through the solution for
30min.
. Prepare stock solutions of the relevant compounds at the concentrations
described above in the reagents section.
. Assess the stability of photoNOD-1 to CYP450 enzymes:
a. To each of three quartz cuvettes, add 986.5μL of PBS.
b. Collect a baseline spectrum on the UV–visible spectrophotometer.
c. Add 10μL of rat liver microsomes followed by 1μL of 5mM
photoNOD-1 stock solution to each cuvette, carefully pipetting
up and down to mix.
d. Cap the cuvettes with rubber septa and incubate the cuvettes in the
water bath at 37°C for at least 10min. NOTE: While photoNOD-1
is generally poorly soluble in aqueous solutions without organic cos-
olvents, the rat liver microsomes help to solubilize photoNOD-1 in
this experiment during this preincubation period.

Near-infrared photoactivatable nitric oxide donors
137
e. Add 2.5μL of 2mM NADPH to each cuvette, mix the solution,
and immediately acquire initial absorbance spectra (400–900nm).
f. Cap the cuvettes with rubber septa and incubate the cuvettes at 37°
C for 1h in a water bath.
g. Mix the solutions and acquire final absorbance spectra.
. Assess the stability of photoNOD-1 to human plasma:
a. To each of four quartz cuvettes, add the following reagents:
i. 200μL of human plasma
6
ii. 799μL of PBS
b. Carefully pipet up and down to mix, then cap the cuvettes with
rubber septa and incubate the cuvettes at 37°C for at least 10min
in a water bath.
c. Using one cuvette, collect a baseline spectrum on the UV–visible
spectrophotometer.
d. Add 1μL of 5mM photoNOD-1 stock solution to each of the
remaining three cuvettes, carefully pipetting up and down to mix.
e. Acquire initial absorbance spectra.
f. Cap the cuvettes with rubber septa and incubate the cuvettes in the
water bath at 37°C for 1h.
g. Acquire final absorbance spectra (400–900nm).
. Assess the stability of photoNOD-1 to glutathione and the reactive
metal species. NOTE: For experiments involving Cu(I) or Fe(II),
avoid introducing air into the cuvettes by evacuating cuvettes under
high vacuum, then flushing the cuvettes with nitrogen. Degassed
solutions should be injected through the septum with a micro-
syringe. This will minimize oxidation to Cu(II) and Fe(III),
respectively.
7
a. To each of three quartz cuvettes, add 989μL of assay buffer. Use
one of the three cuvettes to acquire a baseline spectrum.
b. Add 1μL of 5mM photoNOD-1 stock solution to each of the three
cuvettes.
c. Cap the cuvettes with rubber septa and incubate the cuvettes in the
water bath at 37°C for at least 10min.
d. Add 10μL of stock solution corresponding to the species of interest.
e. Acquire initial absorbance spectra.
f. Cap the cuvettes with rubber septa and incubate the cuvettes in the
water bath at 37°C for 1h.
g. Acquire final absorbance spectra.

138
Effie Y. Zhou et al.
6
4
2
0
0
0
0
RLM
Buffer Cu(I) Cu(II)Fe(II )F e(III)
GSH
PA irradiation
0% Plasma / PBS
2
Fig. 4 Ratiometric fold turn-onof photoNOD-1 upon incubation with various analytes,
as determined from the absorbance spectra before and after stability tests. Data is plot-
ted as the meanꢄSD (n¼3). Reproduced with permission from ACS (Zhou, E. Y., Knox, H. J.,
Reinhardt, C. J., Partipilo, G., Nilges M. J., & Chan, J. (2018). Near-infrared photoactivatable
nitric oxide donors with integrated photoacoustic monitoring. Journal of the American
Chemical Society, 140, 11686–11697. https:/doi.org/10.1021/jacs.8b05514).
8
. To calculate the ratiometric fold turn-on for each experiment (Fig. 4),
use the following equation.
Ratiometricfoldturnꢀon ¼ ðAbs730 nm=Abs
Þ
=ðAbs730 nm=Abs
Þ
680 nm final
680 nm initial
5
.6 Validation that photoNOD-1 and rNOD-1 are non-cytotoxic
Because a photoNORM is designed to be used as an activatable NO source
in biological experiments, it is important that we confirm that neither
the photoNORM nor its product after irradiation is inherently cytotoxic
to cellular systems. Confirmation of this fact suggests that any toxicity
observed in future experiments can be attributed to the effects of NO, rather
than to the donor or dye.
Equipment
9
1
1
6-well plate, clear-bottomed, coated for cell culture
cell culture hood
incubator suitable for mammalian cell culture, with 5% CO2
atmosphere
epifluorescence microscope equipped with a DAPI filter cube and an
RFP filter cube
1

Near-infrared photoactivatable nitric oxide donors
139
Reagents
4T1 cells
RPMI 1640 Medium supplemented with 10% FBS and 1% penicillin/
streptomycin
photoNOD-1 and rNOD-1 (0, 2.7, 5.4, 13.5, and 27mM in DMSO)
Propidium iodide (0.5mg/mL)
Hoechst 33342 stain (0.1mg/mL)
Procedure
1
2
3
. Warm the modified DMEM media in a 37°C water bath.
4
. Seed 30 wells with 160μL (8.8ꢁ10 cells/mL of media) of 4T1 cells.
. Incubate the plate for 12h at 37°C, or until the wells approach 50–60%
confluence.
4
5
6
. Meanwhile, prepare the stock solutions of photoNOD-1 and rNOD-1
in DMSO as described in the Reagents list.
. Dilute each stock solution 1:9 in modified DMEM to generate solu-
tions of 270μM, 540μM, 1.35mM, and 2.70mM in media.
. Add 20μL of the appropriate dye solution to each well such that
the cells will be incubated with 0, 30, 60, 150, or 300μM of dye
(see Fig. 5).
7
8
. Incubate the plate, in the dark, for 24h.
. Add 10μL of propidium iodide and 10μL of Hoechst 33342 to
each well.
9
. Incubate the plate in the dark for an additional 15min.
1
1
0. Image each of the wells using an epifluorescence microscope:
a. Count Hoechst-stained cells using a DAPI filter cube
b. Count Propidium iodide-stained cells using the RFP filter cube
1. Calculate percent viability by dividing the number of cells unstained by
propidium iodide by the total number of Hoechst-stained cells.
Fig. 5 Layout of 96-well plate. Highlighted wells are plated with 4T1 cells. Wells
highlighted in blue are incubated with various concentrations of photoNOD-1; wells
highlighted in orange are incubated with various concentrations of r NOD-1.

140
Effie Y. Zhou et al.
6. Characterization of photoNOD-1 in vivo
Once the release of NO from a given photoNORM has been dem-
onstrated in vitro, validation of the release must also be confirmed in vivo. In
this section, we validate that irradiation-mediated release of NO is operative
in vivo. We first use authentic photoNOD-1 and rNOD-1 to identify the
wavelengths (λblue ¼700nm, λ ¼830nm) that will maximize the PA rat-
red
iometric signal change upon release of NO from photoNOD-1 in vivo. We
then validate that photoNOD-1 turns over in shallow tissues upon exposure
to near-IR light. Finally, we demonstrate that photoNOD-1 can be used as
an effective tool to modulate tumor progression in an irradiation-dependent
manner, and we show that repression of tumor growth is likely due to
NO-induced apoptosis. These protocols may be adapted for other near-
IR activatable photoNORMs to assess their in vivo release efficacy.
The following experiments require training in the ethics and practice
of animal handling and may not be conducted without institutional
approval. All in vivo experiments were performed with the approval of
the Institutional Animal Care and Use Committee of the University of
Illinois at UrbanaꢀChampaign, following the principles outlined by the
American Physiological Society on research animal use. Female BALB/c
mice (5–7 weeks old) were acquired from The Jackson Laboratory.
6
.1 Identification of optimal wavelengths for monitoring
photoNOD-1 in vivo
Because photoNOD-1 and rNOD-1 may solubilize differently in tissue than
in our in vitro solvent systems, we conduct this experiment to identify the
wavelengths at which the dyes can be most easily detected in vivo.
Equipment
1
1
2
2
1
1
—Photoacoustic tomographer (Nexus 128, Endra Life Sciences)
—Vaporizer equipped with isoflurane and oxygen (for anesthesia)
—sterile Eppendorf tubes
—sterile insulin syringes
—electric shaver, for hair removal
—bottle depilatory cream, for hair removal
Animals
1—Female BALB/c mouse, 5–7 weeks old
Reagents
photo NOD-1
rNOD-1
Sterile saline containing 2% DMSO by volume

Near-infrared photoactivatable nitric oxide donors
141
Procedure
1
. Warm up the photoacoustic tomographer for at least 30min prior to
imaging.
2
. Anesthetize the mouse in an isoflurane/oxygen chamber, then main-
tain an anesthetized state using a stream of isoflurane through a nose
cone for the remainder of the experiment.
3
4
5
6
7
. Shave both flanks of the mouse. Remove any residual hair with
depilatory cream.
. After hair removal is complete, gently wash away the residue from the
skin with warm damp gauze.
. Formulate 30μM photoNOD-1 in the saline/DMSO solution in a
sterile Eppendorf tube.
. Using an insulin syringe, inject 25μL of the 30μM photoNOD-1
solution subcutaneously into one flank.
. Place the mouse in the tomographer bowl.
a. Acquire PA spectra for the flank (continuous mode, 6s rotation
time), scanning from 850 to 680nm at 10nm intervals. NOTE:
It is important to scan from red wavelengths to blue wavelengths,
as scanning near λblue first will cause NO release and rNOD-1
generation that will be detected in the spectrum.
. Repeat steps 4–7 for rNOD-1, injecting the dye into the
opposite flank.
8
9
. Process the data by generating three-dimensional reconstructed images,
selecting equal-volume regions of interest (ROIs) from each image,
and quantifying the mean PA signal in each ROI, using the Horos
DICOM viewer or equivalent software.
10. Plot the PA signal versus excitation wavelength for each compound to
determine the optimal wavelengths of interest (see Fig. 6).
6
.2 Confirmation of light-dependent conversion
of photoNOD-1 to rNOD-1 in vivo
The goal of this experiment is to verify that near-IR irradiation can drive the
conversion of photoNOD-1 to rNOD-1 (and thus the release of NO) in live
tissues that contain diverse species which may absorb or scatter the incident
light. In our hands, irradiation of photoNOD-1 for 5min yields a 1.37-fold
ratiometric PA turn-on relative to a theoretical maximum of 2.09-fold,
suggesting that NO release can indeed be achieved in live animals.
Equipment
1—Photoacoustic tomographer (Nexus 128, Endra Life Sciences)
Vaporizer equipped with isoflurane and oxygen (for anesthesia)
Insulin syringes

142
Effie Y. Zhou et al.
Fig. 6 PA spectra of photoNOD-1 and rNOD-1 in vivo. We select λblue ¼700nm and
λred ¼830nm as the ideal wavelengths to track photoNOD-1 and rNOD-1, respectively.
Reproduced with permission from ACS (Zhou, E. Y., Knox, H. J., Reinhardt, C. J., Partipilo, G.,
Nilges M. J., & Chan, J. (2018). Near-infrared photoactivatable nitric oxide donors with inte-
grated photoacoustic monitoring. Journal of the American Chemical Society, 140,
1
1686–11697. https: // doi.org/10.1021/jacs.8b05514).
Animals
3
—Female BALB/c mice, 5–7 weeks old
Reagents
photo NOD-1
Sterile saline with 2% DMSO
Ethanol
Procedure
1
2
3
4
5
. Follow steps 1–6 from Section 6.1 for all three animals, injecting
photoNOD-1 into both flanks.
. Acquire initial PA images at λred (830nm) and λblue (700nm) (contin-
uous mode, 6s rotation time, 0.2min irradiation at λblue).
. Irradiate one flank at λblue (step-and-shoot mode, 120 angles, 39 pulses
per angle, 4.8min). Do not irradiate the second flank.
. Acquire final PA images at λred and λblue (continuous mode, 6s rotation
time, 0.2min irradiation at λblue).
. Process the data by generating three-dimensional reconstructed images,
selecting equal-volume regions of interest (ROIs) from each image, and
quantifying the mean PA signal in each ROI at both wavelengths, at the
initial and final timepoints, using the Horos DICOM viewer or equiv-
alent software.
6
. Calculate the ratiometric fold turn-on of photoNOD-1 to rNOD-1
with and without irradiation (Fig. 7):
Ratiometricfoldturnꢀon ¼ ðPA830 nm=PA
Þ
=ðPA830 nm=PA
Þ
700 nm final
700 nm initial