In vitro oxidations of low-density lipoprotein and RAW 264.7 cells with lipophilic O(3P)-precursors

Article abstract of DOI:10.1039/d0ra01517b

A beneficial property of photogenerated reactive oxygen species (ROS) is the capability of oxidant generation within a specific location or organelle inside a cell. Dibenzothiophene S-oxide (DBTO), which is known to undergo a photodeoxygenation reaction to generate ground state atomic oxygen [O(3P)] upon irradiation, was functionalized to afford localization within the plasma membrane of cells. The photochemistry, as it relates to oxidant generation, was studied and demonstrated that the functionalized DBTO derivatives generated O(3P). Irradiation of these lipophilic O(3P)-precursors in the presence of LDL and within RAW 264.7 cells afforded several oxidized lipid products (oxLP) in the form of aldehydes. The generation of a 2-hexadecenal (2-HDEA) was markedly increased in irradiations where O(3P) was putatively produced. The substantial generation of 2-HDEA is not known to accompany the production of other ROS. These cellular irradiation experiments demonstrate the potential of inducing oxidation with O(3P) in cells. This journal is

Full text of DOI:10.1039/d0ra01517b

RSC Advances
View Article Online
View Journal | View Issue
PAPER
In vitro oxidations of low-density lipoprotein and
RAW 264.7 cells with lipophilic O(3P)-precursors†
Cite this: RSC Adv., 2020, 10, 26553
a
John T. Petroff, II,a Ankita Isor,a Satyanarayana M. Chintala,
Carolyn J. Albert,b
Jacob D. Franke,b David Weinstein,a Sara M. Omlid,a Christopher K. Arnatt,a
David A. Fordb and Ryan D. McCulla
*
a
A beneficial property of photogenerated reactive oxygen species (ROS) is the capability of oxidant
generation within a specific location or organelle inside a cell. Dibenzothiophene S-oxide (DBTO), which
is known to undergo a photodeoxygenation reaction to generate ground state atomic oxygen [O(3P)]
upon irradiation, was functionalized to afford localization within the plasma membrane of cells. The
photochemistry, as it relates to oxidant generation, was studied and demonstrated that the
functionalized DBTO derivatives generated O(3P). Irradiation of these lipophilic O(3P)-precursors in the
presence of LDL and within RAW 264.7 cells afforded several oxidized lipid products (oxLP) in the form
of aldehydes. The generation of a 2-hexadecenal (2-HDEA) was markedly increased in irradiations where
O(3P) was putatively produced. The substantial generation of 2-HDEA is not known to accompany the
production of other ROS. These cellular irradiation experiments demonstrate the potential of inducing
oxidation with O(3P) in cells.
Received 17th February 2020
Accepted 26th June 2020
DOI: 10.1039/d0ra01517b
rsc.li/rsc-advances
where an entire cell may be exposed to the exogenous ROS, and
there exist cases where this ‘shotgun blast’ approach is not
favorable.3,5,19 Ground state atomic oxygen [O(3P)] is an oxidant
Introduction
Reactive oxygen species (ROS) are widely utilized and studied in
their endogenous and exogenous forms through the lens of
biology.1–5 These species come in the form of singlet oxygen
(1O2), ozone (O3), hydroxy radicals (cOH), superoxide (O2ꢀ),
peroxides (ROOR), and others. Some ROS, such as O2ꢀ and cOH,
are particularly biologically relevant since they are produced
endogenously.6,7 However, others, especially 1O2, have been
used in various applications to study the role of ROS in biology
relating to oxidative stress.3,8–12 The role of oxidative stress in
biology and medicine is of great importance as endogenous
ROS are involved in normal cell signaling and diseases
alike.10,13–17 With oxidative stress underpinning several of the
global leading causes of death, expanding the capacity to
generate ROS in cells to further elucidate the pathophysiology of
oxidative stress is important.18
which has shown selectivity towards the oxidation of specic
functional groups in biomolecules, and thus, O(3P) may be
useful in biological oxidations.12,20,21 O(3P) is a frequent topic of
research in atmospheric chemistry, but has only been studied in
solution for the last few decades.22–25 An additional feature of
photogeneration of O(3P) is the precursors carry the oxidant and
are not reliant on endogenous molecular oxygen to create ROS,
as is the case for photosensitizers. This freely diffusing oxidant
is typically generated through irradiation of heterocycle
oxides.26,27 For example, dibenzothiophene S-oxide (DBTO)
generates O(3P) and the corresponding deoxygenated product
dibenzothiophene (DBT) as shown in Fig. 1.23,25,26,28,29
Recently, it has been conrmed through uorescent
microscopy that functionalizing dibenzothiophene S,S-oxides
(DBTOOs) allows for non-toxic organelle-specic localization in
Currently, there exists a variety of means to generate or
introduce ROS and oxidative stress in cells, with photosensi-
tizers and peroxides being leading methods.4,5 Photosensitizer
upon irradiation typically excite oxygen to singlet oxygen or
undergo electron transfer generating superoxide. One of the
downsides to many of these methods is a lack of selectivity
aDepartment of Chemistry, Saint Louis University, St. Louis, MO, USA. E-mail: ryan.
mcculla@slu.edu
bDepartment of Biochemistry and Molecular Biology, Saint Louis University School of
Medicine, St. Louis, MO, USA
Fig. 1 Generation of DBT and O(3P) through irradiation of DBTO.
RSC Adv., 2020, 10, 26553–26565 | 26553
† Electronic supplementary information (ESI) available. See DOI:
10.1039/d0ra01517b
This journal is © The Royal Society of Chemistry 2020

View Article Online
RSC Advances
Paper
Fig. 2 Compounds 1-SO2, 1-SO, 2-SO2, 2-SO, and 3-SO.
HeLa cells.30,31 This was achieved with 4-(8-octyl-5,5-dioxidodi-
Previous studies generating O(3P) with 2,8-bis(hydrox-
benzo[b,d]thiophen-2-yl)butanoic acid (1-SO2) whose analogous ymethyl)dibenzo[b,d]thiophene 5-oxide (3-SO) in the presence
sulfoxide is
4-(8-octyl-5-oxidodibenzo[b,d]thiophen-2-yl) of 1-O-hexadec-10-enyl-sn-glycero-3-phosphocholine (pLPC), low-
butanoic acid (1-SO) and with 4-(5,5-dioxidodibenzo[b,d] density lipoprotein (LDL), and RAW 264.7 cells produced
thiophen-2-yl)butanoic acid (2-SO2) whose sulfoxide analog is evidence that O(3P) generates unique products unseen with
4-(5-oxidodibenzo[b,d]thiophen-2-yl)butanoic acid (2-SO) as other ROS.20 The formation of these unique oxidized lipid
shown in Fig. 2. With conrmation that the DBT backbone can products (oxLP) was seen when pLPC and LDL, but not RAW
be localized in organelles, the plasma membrane microscopy 264.7 cells, were exposed to O(3P). In light of these results, it was
dyes (1-SO2 and 2-SO2) were chosen as the model systems to hypothesized that O(3P)-precursors with increased lipophilicity
mimic their sulfoxide analogs. The dyes, 1-SO2 and 2-SO2, have would afford more oxLP in these systems. To test this hypoth-
good overlap with known plasma membrane dyes.31 As the esis, sulfoxide analogs (1-SO & 2-SO) of the sulfone plasma
sulfone (1-SO2 and 2-SO2) and sulfoxide (1-SO and 2-SO) are membrane dyes were synthesized.
structurally similar, it was posited 1-SO and 2-SO would localize
within plasma membranes of the cell.
To complete the primary objective of testing the hypothesis
stated above, the photochemistry and oxidation of LDL and
Scheme 1 Synthesis of 1-SO.
26554 | RSC Adv., 2020, 10, 26553–26565
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
Scheme 2 Synthesis of 2-SO.
Results and discussion
It has been shown, with uorescent microscopy, that func-
tionalization of dibenzothiophene sulfones (DBTOOs) with
particular substituents will direct the small molecule to localize
in specic cellular sites and organelles.30,31 The addition of
a butanoic acid to the 2 position of a DBTOO with or without an
octyl group at the 8 position (1-SO2 and 2-SO2) provided staining
of the plasma membrane in cells.31 Extrapolating from the
capacity for 1-SO2 and 2-SO2 to localize in the plasma
membrane, it was posited their sulfoxide analogs 1-SO and 2-SO
would be similarly lipophilic. Thus, 1-SO and 2-SO were
synthesized to determine if increased lipophilicity of a O(3P)-
precursor enhanced the formation of lipid oxidation products
upon irradiation.
The synthesis of 1-SO and 2-SO mirrors the previously re-
ported synthesis of 1-SO2 and 2-SO2 until each molecule's
respective last step.31 However, to produce 1-SO as shown in
Scheme 1, a Kumada coupling was conducted with 2-bromodi-
benzothiophene (4) and n-octyl magnesium bromide in dry
tetrahydrofuran (THF) with a nickel catalyst.31 Following
workup and purication with normal phase chromatography, 2-
octyldibenzo[b,d]thiophene (5) was isolated in a 64% yield.32
The 2-octyldibenzo[b,d]thiophene (5) was added to a solution of
2 : 1 dichloroethane (DCE)/nitrobenzene with succinic
Fig. 3 Toluene common intermediate experiment with the irradiation
of 1-SO and 2-SO.
RAW 264.7 cells by 1-SO, 2-SO, and 3-SO was investigated. Their
oxidative photochemistry was vetted and their ability to gener-
ated O(3P) was veried. These O(3P) precursors were exposed to
LDL and then, separately, RAW 264.7 cells in the presence of
UVA light. The LDL and cells were subsequently analyzed by GC-
MS to determine if unique oxidized lipid products could be
found. Following analysis of the MS data, it was determined that
irradiation of photolabile lipophilic O(3P) precursors in the
presence of LDL or RAW 264.7 cells generate 2 hexadecenal (2-
HDEA) in amounts signicantly greater than control. This data
suggests that unique, and previously unachievable, oxidation
chemistry is achieved when O(3P) precursors are irradiated
while inside of RAW 264.7 cell membranes.
Table 1 Common intermediate test in toluene of 1-SO and 2-SO and their quantum yields of sulfide formation
Toluene oxidation product yieldsa
m/p-Cresolb
fsulde
c
Compound
Benzaldehyde
Benzyl alcohol
o-Cresol
DBTOd
DBTOf
DBTOg
1-SOg
5 ꢁ 3
17 ꢁ 3
6 ꢁ 3
13 ꢁ 4
17.6 ꢁ 0.9
26 ꢁ 5
13 ꢁ 1
0.0026 ꢁ 0.0004e
0.0046 ꢁ 0.0007f
—
22 ꢁ 5
1.8 ꢁ 1.2
2.3 ꢁ 1.0
14.7 ꢁ 4.1
2.0 ꢁ 0.1
3.1 ꢁ 0.1
21.4 ꢁ 1.2
11 ꢁ 2
8.6 ꢁ 3.5
4.1 ꢁ 0.1
4.8 ꢁ 1.1
7.8 ꢁ 0.5
7.5 ꢁ 0.7
0.0012 ꢁ 0.0002g
2-SOg
0.0020 ꢁ 0.0003g
a
Yields of toluene oxidation products were calculated relative to the corresponding sulde produced during photodeoxygenation. Error was
b
c
determined as a 95% condence interval. Measured as single peak. Deoxygenation quantum yields of sulde formation in acetonitrile.
d
e
f
g
Data from Satyanarayana et al.35 Data from Gregory et al.28 Data from ref Rockafellow et al.27 Data from this work.
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 26553–26565 | 26555

View Article Online
RSC Advances
Paper
anhydride in the presence of aluminum trichloride (AlCl3) and in these experiments. Conversely, O(3P), when unadulterated by
the reaction was kept under an inert atmosphere. The addition residual O2, favors cresols formation.26
of the oxobutanoic acid to form 4-(8-octyldibenzo[b,d]thiophen-
The results of common intermediate experiments of 1-SO
2-yl)-4-oxobutanoic acid (6) was followed by reduction of the and 2-SO with toluene are listed in Table 1. Since benzylic
carbonyl to form 4-(8-octyldibenzo[b,d]thiophen-2-yl)butanoic oxidation is an unreliable indicator in this common interme-
acid (1-S). The deviation from the previously reported diate experiment due to its sensitivity to dissolved molecular
synthesis involves the formation of the sulfoxide, rather than oxygen, the ratios of o-cresol and m/p-cresol formation were
a sulfone, which was performed with 1.1 equivalents of mCPBA compared. Both 1-SO and 2-SO generated more o-cresol than m/
at ꢀ30 ꢂC which afforded 4-(8-octyl-5-oxidodibenzo[b,d] p-cresols in a nearly 2 : 1 ratio. Likewise, DBTO generated more
thiophen-2-yl)butanoic acid (1-SO) in a 69% yield.
o-cresol; however, the ratio was closer to 1 : 1. While there may
To synthesize 2-SO, a nearly identical approach was taken be other processes involved for 1-SO and 2-SO that increased the
with 1-SO sans the use of 4 as a reactant. Dibenzothiophene (7) observed ratio, cresol formation in and of itself can indicate
was treated with succinic anhydride in the presence of a catalyst O(3P) is being produced during photo-deoxygenation.33 Addi-
to produce 4-(dibenzo[b,d]thiophen-2-yl)-4-oxobutanoic acid (8). tionally, the total yield of the oxidized products for 1-SO was half
The carbonyl of 8 is subsequently reduced using zinc amalgam of that observed for DBTO. For 2-SO, nearly identical yields of
in acidic conditions to form 4-(dibenzo[b,d]thiophen-2-yl)- cresols were observed compared to 1-SO. However, the yields of
butanoic acid (2-S). This sulde is oxidized to the sulfoxide benzaldehyde and benzyl alcohol increased to 14.7% and
with mCPBA. This reaction produces 4-(5-oxidodibenzo[b,d] 21.4%, respectively. The difference in yields among 1-SO, 2-SO,
thiophen-2-yl)-butanoic acid (2-SO) in 52% yield (Scheme 2).
and DBTO, despite the similarity of the chromophores, under-
To examine the oxidation proles of 1-SO and 2-SO, a solu- scores the sensitivity of this reaction to the reaction conditions.
tion of the respective sulfoxide in toluene was degassed by However, the formation of cresols in a similar ratio for 1-SO and
argon sparging and then irradiated with 14 broadly emitting 2-SO indicate that their photodeoxygenation yields some O(3P)
LZC-UVA bulbs for four hours. In previous reports, the standard even if other background processes cannot be ruled out.
generator of O(3P), DBTO, produced benzaldehyde, benzyl
To examine the efficiency of photodeoxygenation for 1-SO
alcohol, o-cresols, and m/p-cresols as shown in Fig. 3.26,33 and 2-SO, the quantum yield of sulde formation (fsulde) for 1-
However, the ratio of the oxidized products for toluene has been S and 2-S, respectively, was determined. The fsulde was deter-
found to be very sensitive to degassing techniques and the mined by irradiating a degassed sample of a sulfoxide in an
particular sulfoxide undergoing deoxygenation.34 Incremental inert solvent, such as acetonitrile, at a given wavelength, typi-
changes in residual molecular oxygen appear to affect benzylic cally 320 nm.26,27 The quantum yield of deoxygenation for DBTO
oxidation in toluene as photo-generated O(3P) is speculated to was approximately 0.003.28 Saturated solutions of 1-SO and 2-SO
react with residual O2 to form ozone that increases the forma- were prepared in acetonitrile, argon sparged, and irradiated at
tion of benzaldehyde and benzyl alcohol.34 For example, larger 320 nm. The irradiations were halted prior to exceeding 10%
headspaces or longer irradiation periods led to increased sulfoxide conversation to ensure that the sulfoxide absorbed the
benzaldehyde and benzyl alcohol even aer argon sparging.34 majority of the light. Aer the corresponding sulde formation
Also, aer performing freeze–pump–thaw (FPT) cycles,28 which was determined and ux was measured by chemical actinom-
is a superior degassing technique, irradiation of DBTO in etry, the fsulde for 1-SO and 2-SO were found to be 0.0012 and
toluene afforded no benzaldehyde. Functionalization or modi- 0.0020, respectively. These values are lower than that of DBTO,
cation of the DBTO chromophore with aromatic substituents with 1-SO preforming at nearly half the efficiency of 2-SO.
has been observed to decrease the yield of oxidized products.33,34
Low-density lipoprotein (LDL) is comprised of surface-
These observations with DBTO functionalization suggest that containing phospholipids, which are known to react with
other additional processes, which do not involve O(3P), may O(3P), and a central core of other lipids including glycerides and
produce benzaldehyde and enhance benzyl alcohol production cholesterol esters.20 A substantial portion of the esteried fatty
acids of LDL are known to be polyunsaturated, providing
a population of likely targets for O(3P), which has an affinity for
unsaturated hydrocarbons.36,37 Previously, 3-SO has been used
to generate O(3P) in solution with LDL and isolated phospho-
lipids.20 In this previous work, UV-irradiation of LDL in the
presence of 3-SO resulted in a substantial increase of four
aldehyde products compared to UV-irradiation alone. These
four aldehydes were: tetradecanal (TDA), pentadecanal (PDA),
hexadecenal (HDA), and 2-hexadecenal (2-HDEA). In this work,
the formation of these four aldehydes plus octadecanal (ODA)
as shown in Fig. 4, were monitored as evidence for lipid
oxidation induced by photodeoxygenation of 1-SO, 2-SO, and 3-
SO.
LDL also serves as a more complex system than an isolated
phospholipid in solution, allowing one to determine if the
Fig.
4
Potential
lipid
oxidation
products
arising
from
photodeoxygenation.
26556 | RSC Adv., 2020, 10, 26553–26565
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
lipophilicity of the O(3P)-precursor enhances oxLP. To measure
oxLPs, 1 mL solutions of either 200 mM 1-SO, 2-SO, 3-SO, 1-S, 2-
S, and dibenzo[b,d]thiophene-2,8-diyldimethanol (3-S) with
2 mg mLꢀ1 of LDL were irradiated with broadly emitting UVA
bulbs for 2 hours. Additional control experiments included 1-
SO, 2-SO, 3-SO, 1-S, 2-S, and 3-S being incubated in the dark
with LDL and irradiating LDL alone. Aer irradiation, the
samples were subject to Bligh–Dyer extractions and derivatized
with O-(2,3,4,5,6-pentauorobenzyl)hydroxylamine hydrochlo-
ride (PFB) and then analyzed on GC-MS, where the molecular
ions for PFB-derivatized TDA, PDA, HDA, 2-HDEA, and ODA
were monitored. The results of these experiments are shown in
Fig. 5.
The oxidation of LDL by all three sulfoxides (1-SO, 2-SO, and
3-SO) upon UV irradiation gave similar results (Fig. 5A: 1-SO +
light, Fig. 5B: 2-SO + light, Fig. 5C: 3-SO + light). For all three
sulfoxides, the amount of 2-HDEA had the largest increase
compared to control experiments, and the total amounts of
lipid aldehydes detected generally increased. In the control
experiments with only irradiation and no compounds (Fig. 5,
light) or no light (Fig. 5A: 1-SO + dark, Fig. 5B: 2-SO + dark,
Fig. 5C: 3-SO + dark), less than 10 pmol per mg LDL for any one
aldehyde was detected. To verify that photodeoxygenation of 1-
SO, 2-SO, and 3-SO was the sole cause of the observed increase
of aldehydes, the corresponding suldes 1-S, 2-S, and 3-S were
tested (Fig. 5A: 1-S + light, Fig. 5B: 2-S + dark, Fig. 5C: 3-S + light).
For all three suldes, a signicant increase in TDA, PDA, and 2-
HDEA compared to the other control experiments was observed.
Compared to their corresponding sulfoxides, the suldes yiel-
ded approximately the same, if not more, TDA and PDA. Addi-
tionally, 2-HDEA was the dominant aldehyde product for both 2-
S and 3-S. However, the three suldes all produced at least 60
pmol less 2-HDEA per mg of LDL than their corresponding
sulfoxides. Compared to all three controls, only the increase in
2-HDEA for 1-SO, 2-SO, and 3-SO and light had p-values of less
than 0.05 indicating signicance. These results led to the
conclusion that photodeoxygenation has the most substantial
effect on the formation of 2-HDEA.
The rate of photodeoxygenation for 1-SO, 2-SO, and 3-SO was
not the same as seen in Table 1 and previous work.38 This
discrepancy stems from quantum yield measurements being
done under anaerobic conditions in organic solvent, which is
the standard approach.27,34 However, these irradiations reported
here occurred in aqueous media under aerobic conditions,
38
which is known to inuence quantum fsulde. Thus, to quan-
tify the extent to which photodeoxygenation was affecting 2-
HDEA formation, the amount of 2-HDEA formed compared to
sulde formation (i.e. 1-S, 2-S, and 3-S) was determined. To
determine the extent of photodeoxygenation, 1-SO, 2-SO, and 3-
Fig. 5 Amount of oxidized lipid product arising from LDL after irra-
diation of O(3P)-precursors and additional control experiments. (A) SO were irradiated in the same conditions as were used to
LDL oxidized by the photodeoxygenation of 1-SO (1-SO + light) and
controls, which were the corresponding sulfide in light (1S + light), 1-
SO without irradiation (1-SO dark), and irradiation of cells without
addition of compounds (light). (B) LDL oxidized by the photo-
oxidize LDL except without LDL present. The results of these
experiments are used to accurately account for 2-HDEA forma-
tion and are shown in Fig. 6. Both 2-SO and 3-SO produced
similar amounts of 2-HDEA relative to the amount of deoxy-
deoxygenation of 2-SO (2-SO + light) and controls similar to (A) with
2-SO or 2-S. (C) LDL oxidized by the photodeoxygenation of 3-SO (3- genation. This was consistent with the similar quantum yields
for 2-SO and 3-SO, 0.0012 and 0.002,38 respectively, and the
SO + light) and controls similar to (A) with 3-SO or 3-S.
nearly identical amounts of 2-HDEA formed as shown in Fig. 5.
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 26553–26565 | 26557

View Article Online
RSC Advances
Paper
(FBS), 1% penicillin–streptomycin (Thermo Fisher 10 000 U
mLꢀ1), and 1% GlutaMAX. Cells were split and then seeded into
96-well plates with 10 000 cells per well. The plate was then
incubated for 24 hours. Vehicle control was added to the wells
to reach a nal concentration of 0.2% DMSO in DPBS. The
plates were then incubated for 10 minutes and transferred to
the photoreactor where they were irradiated with 14 broadly
emitting UVA bulbs. Cell viability was then determined using
a MTS assay. While the cells tolerated 5 minutes UV exposure,
cell viability decreased to 23% and ꢀ3% at 1 and 2 hours,
respectively, as shown in Fig. 7.
To examine the morphology of the cells at 2 hours, the cells
were plated on a cell culture dish (10 ꢃ 200 mm) with 88 000
cells per mL density. The cell culture dish was incubated for 24
hours and then placed in a photoreactor with 14 broadly emit-
ting UV-A bulbs. The cell culture dish was then examined on
Olympus BX60 microscope (Fig. S2†). As expected, there
morphological changes were consistent with cell death;
however, the cell membrane remained largely intact. Thus,
while cell viability was signicantly decreased, the ability more
lipophilic O(3P)-precursors to generate lipid oxidation products
in cells could still be examined.
As described above, RAW 264.7 cells were grown in media as
described above before being split and seeded into 6-well plates
with 1 ꢃ 106 cells per well. The plate was incubated for 24 hours,
allowing the cells to multiply to conuency. Aer the incubation
media was removed, the given compound (1-SO, 2-SO, 3-SO, 1-S,
2-S, or 3-S) was added to the well in a nal concentration of 200
mM in 1 mL of PBS. The 6-well plate was then incubated for an
additional 10 minutes. Aer incubation, the plate was trans-
ferred to the photoreactor where it was irradiated with broadly
emitting UVA bulbs for 2 hours. Once irradiation was complete,
Bligh–Dyer extractions were performed. Once Bligh–Dyer
extractions were complete, the sample was puried over
a SupelCo SupelClean LC-Si column using chloroform as the
eluant. The isolated eluent was evaporated, reconstituted, and
PFB derivatized. The crude derivatization solution was then
puried by liquid–liquid extraction. The extract was then
prepared for analysis by GCMS using the same method as was
used with LDL.
Fig. 6 Amount of 2-HDEA produced in LDL relative to the extent of
photodeoxygenation of 1-SO, 2-SO, and 3-SO upon irradiation.
The most lipophilic sulfoxide, 1-SO, produced over double the
amount of 2-HDEA per sulde generated compared to 2-SO and
3-SO. Thus, the smaller quantum yield for 1-SO compared to 2-
SO and 3-SO is likely the reason for the decreased amount of 2-
HDEA for 1-SO in Fig. 5. The log P values of 1-SO, 2-SO, and 3-SO
were calculated or measured at 6.26, 2.86, and 1.88, respec-
tively.21 Overall, these results indicate the increased lipophilicity
of 1-SO makes it more efficient at producing 2-HDEA in LDL.
In previous oxidations of LDL with 3-SO, 2-HDEA was sus-
pected of arising from oxidation of the vinyl ether group of
plasmalogens, which is known to react with various ROS.20,39,40
While 2-HDEA was the major product when isolated plasmal-
ogens and LDL were exposed to O(3P), 2-HDEA has not been
observed to form with other ROS. Singlet oxygen (1O2) formed
long-chain aldehydes when generated in Chinese hamster
ovarian cells.41 Hydroxyl radical (cOH) formed a-hydrox-
yaldehydes when reacted with LDL, and ozone (O3) forms long-
chain saturated aldehydes when generated in the presence of
plasmalogens.20,40–42 This indicates the 2-HDEA was produced
from a reaction between O(3P) and plasmalogens.
The irradiation of RAW 264.7 cells with (1-SO, 2-SO, 3-SO, 1-
S, 2-S, or 3-S), the associated dark controls, and the untreated
While several other ROS have been generated in cells, O(3P)
has not.4,5,20,43–45 The previously reported efficacy of 1-SO2 and 2-
SO2 as microscopy dyes suggested that 1-SO and 2-SO would
also incorporate into cells.31 To examine the capacity of 1-SO, 2-
SO, and 3-SO to produce oxLP, RAW 264.7 cells were chosen as
the target cell line, since plasmenylethanolamine comprises
approximately 36% of the ethanolamine glycerolipid pool.46 An
analog of plasmenylethanolamines in the form of pLPC has
previously been shown to react readily with O(3P) in solution.20
As with many cell lines, RAW 264.7 cells are sensitive to UV
light, and previous work had found that 5 minutes of irradiation
was enough to induce cell death.47 Thus, cell viability was
examined to determine the amount of cell death under the
irradiation conditions used in this study. RAW 264.7 cells were
grown in RPMI supplemented with 10% fetal bovine serum
Fig. 7 Cell viability of RAW 264.7 cells determined by MTS assay after
irradiation with UVA bulbs over time. Error bars are 95% confidence
intervals.
26558 | RSC Adv., 2020, 10, 26553–26565
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
control afforded oxLPs in the form of TDA, PDA, HDA, 2-HDEA,
and ODA for all conditions. These results are shown in Fig. 7.
Unlike LDL, the control experiments of irradiation without
compounds (Fig. 8, light) and no light (Fig. 8A, 1-SO dark,
Fig. 8B, 2-SO dark) demonstrated an overall increase in oxLP,
with the exception of 3-SO in the dark. In fact, 3-SO in the dark
gave nearly identical amounts of the OLPs as untreated cell as
shown in the ESI (Fig. S3†). Compared to 3-SO dark, the expo-
sure of RAW 264.7 cells to UV irradiation resulted in signicant
increases of TDA, PDA, HDA, and ODA. This was not surprising
since UV exposure is known to generate aldehydes in cells.48,49
Additionally, substantial increases in TDA, PDA, HDA, and ODA
were observed for 1-SO and 2-SO compared to 3-SO when
incubated in the dark. This is consistent with the previous
studies that have shown destabilization of the cellular plasma
membrane make it more susceptible to oxidation.50,51 Thus, the
addition of lipophilic small molecules and UV-irradiation
appears to increase the rate of oxidation leading to TDA, PDA,
HDA, and ODA; however, these processes have no effect on 2-
HDEA formation.
Since small lipophilic molecules and UV-irradiation natu-
rally resulted in an increase in TDA, PDA, HDA, and ODA, it was
not surprising that the irradiation of RAW 264.7 cells with 1-SO,
2-SO, 3-SO, 1-S, 2-S, or 3-S resulted in increase of these four
aldehydes. However, unlike the controls which formed little 2-
HDEA, irradiation of RAW 264.7 cells with 1-SO, 2-SO, 3-SO, 1-S,
2-S, or 3-S all showed an increase in 2-HDEA. All the sulfoxides
yield more 2-HDEA than the corresponding sulde, which was
consistent with the results observed for LDL. For 1-SO and 2-SO,
they produced approximately double the amount of 2-HDEA
compared to 1-S and 2-S, respectively. This increase in 2-HDEA
for 1-SO and 2-SO compared to their corresponding suldes was
found to correspond to p-values of less than 0.01 indicating
signicance. Interestingly, 3-SO only produced 30% more 2-
HDEA than 3-S, which could not be used to support the
hypothesis that photodeoxygenation of 3-SO had any effect on 2-
HDEA formation. This was consistent with previous reports
where no noticeable quantity of oxLP could be measured
compared to the controls for 3-SO.20 These results indicated that
increased lipophilicity of the O(3P)-precursors 1-SO and 2-SO
assists in producing more 2-HDEA upon irradiation and
concomitant photodeoxygenation.
The increase in TDA, PDA, HDA, and ODA in the dark when
RAW 264.7 cells were exposed to the sulfoxides suggested that 1-
SO, 2-SO, and 3-SO may have some toxicity. The cell viability of
RAW 264.7 cells upon incubation of 1-SO and 1-S without the
subsequent irradiation was examined. While 1-S showed no
dark toxicy, a decrease of cell viability to 24% was observed at
200 mM for 1-SO (Fig. S1†). Lower concentrations had no
Fig. 8 Amount of oxidized lipid product arising from RAW 264.7
cells after irradiation of O(3P)-precursors and additional control
experiments. (A) RAW 264.7 cells oxidized by the photo-
deoxygenation of 1-SO (1-SO + light) and controls, which were the
corresponding sulfide in light (1S + light), 1-SO without irradiation signicant effect on cell viability.
(1-SO dark), and irradiation of cells without additional compounds
As with LDL, the amount of deoxygenation for 1-SO, 2-SO,
(light). (B) RAW 264.7 cells oxidized by the photodeoxygenation of
2-SO (2-SO + light) and controls similar to (A) with different
compounds. (C) RAW 264.7 cells oxidized by the photo-
deoxygenation of 3-SO (3-SO + light) and controls similar to (A)
with different compounds.
and 3-SO was not expected to be the same. Thus, the amount of
2-HDEA formed compared to sulde formation was determined
to quantify the effect of photodeoxygenation on 2-HDEA
formation. The sulde formation was quantied in the same
manner as how the LDL sulde formation was measured, except
the samples were housed in a 6-well plate. The results are shown
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 26553–26565 | 26559

View Article Online
RSC Advances
Paper
formation. This work with RAW 264.7 cells is the rst report of
O(3P)-precursors being used to generate oxidized products in
cells. This opens the possibility that O(3P)-precursors may be
functionalized to direct their photocleaved oxidant in some
specied manner in cells.
Experimental
Materials
2,8-Bis(hydroxymethyl)dibenzo[b,d]thiophene 5-oxide (3-SO)
and the corresponding sulde (3-S) were prepared from diben-
zothiophene as previously reported.38 2-[15,15,16,16,16-d5]-
hexadecenal (2-[d5]-hexadecenal) was also prepared by a previ-
ously reported method as well.52 Low-density lipoprotein was
purchased from Lee Biosolutions. Compounds 1-S, 2-S, 4, 5, 6,
7, and 8 were all prepared following previously reported
procedures.31 All other chemicals were purchased from Alfa
Aesar, Oakwood Chemical, Ark Pharm, Thermo Fisher Scien-
tic, and Sigma-Aldrich and were used without further puri-
cation. Flash chromatography was performed using a Biotage
Horizon with silica gel (SiliaFlash P60) purchased from Sili-
Cycle. 1H and 13C NMR spectra were gathered on a Bruker DRX-
400 in either DMSO-d6 or CDCl3. log P calculations were
completed using ChemDraw Ultra 12.0. For gas chromatograph
analysis, a Shimadzu GC-2010 Plus with an AOC-201 Auto
Injector on a Restek Rxi-5ms column was used. An Agilent 1200
Series HPLC tted with a quaternary pump and diode array
detector was used for HPLC chromatographs run on an Agilent
Eclipse XDB-C18 column (5 mm, 150 ꢃ 4.6 mm). High-
resolution mass spectra were measured using a Thermo Scien-
tic Q Exactive Orbitrap equipped with a Nano ESI ionization
source. Absorption spectra were recorded using a Shimadzu UV-
1800 UV-Vis spectrophotometer using samples dissolved in
acetonitrile contained in a 10 ꢃ 10 mm quartz cell at
a concentration of 0.025 and 0.035 mM for 2-SO and 1-SO,
Fig. 9 Amount of 2-HDEA produced in RAW 264.7 cells relative to the
extent of photodeoxygenation of 1-SO, 2-SO, and 3-SO upon
irradiation.
in Fig. 9. When the extent of photodeoxygenation is considered,
1-SO and 2-SO had nearly the same yield of 2-HDEA and nearly
double of what was observed for 3-SO. This was unlike LDL,
where only 1-SO had a higher yield of 2-HDEA. A potential
reason for this difference would be if 1-SO and 2-SO both
incorporated into the plasma membrane of the RAW 264.7 cells
to a similar extent. This supposition is supported by the similar
amounts of oxLP observed for 1-SO and 2-SO when incubated
with RAW 264.7 cells in the dark (Fig. 8A 1-SO dark, Fig. 8B 2-SO
dark).
Conclusion
Two lipophilic DBTO derivatives (1-SO and 2-SO) were respectively.
synthesized, and these molecules were concluded to generate
Synthesis of 4-(8-octyl-5-oxidodibenzo[b,d]thiophen-2-yl)
O(3P) by common intermediate experiments with toluene, butanoic acid (1-SO).31 4-(8-Octyldibenzo[b,d]thiophen-2-yl)
where they both generated cresols upon photodeoxygenation. butanoic acid (1-S) (140 mg, 0.37 mmol) was combined with
The O(3P) precursors, 1-SO and 2-SO, have fsulde of 0.0012 DCM (10 mL). The solution was cooled to approximately
and 0.002, respectively. These sulfoxides, as well as 3-SO and ꢀ30 ꢂC. To this cooled solution, meta-chloroperoxybenzoic
the control experiments, generated OLPs in the form of TDA, acid (mCPBA) (77%) (105 mg, 0.39 mmol) was added. The
PDA, HDA, 2-HDEA, and ODA when irradiated in the presence solution was stirred for four hours, maintaining the temper-
of LDL. The principal product for the LDL irradiations was 2- ature at approximately ꢀ30 ꢂC, then allowed to stir overnight,
HDEA which agrees with previous reports and has only been eventually reaching room temperature. Aer stirring, the
attributed to O(3P) formation. When controlling for total solution was poured over 50 mL of saturated sodium bicar-
sulde formation for the respective sulfoxide and LDL irradi- bonate water. Additional DCM (20 mL) was added. The organic
ations, it becomes apparent that the greater the lipophilicity of layer was washed with saturated sodium bicarbonate water (4
the molecule, as measured by log P, the more oxLP formed. ꢃ 30 mL). The organic layer was separated, dried with anhy-
RAW 264.7 cells were treated with 1-SO, 2-SO, 3-SO and asso- drous MgSO4, and ltered. The dry organic layer was evapo-
ciated controls then irradiated. The sulfoxides and suldes rated under reduced pressure to reach a volume of about 25
afforded meaningful amounts oxLP in the form of TDA, PDA, mL. The resulting solution was puried by ve separate
HDA, and ODA; however, 2-HDEA, the oxLP attributed to O(3P), preparative thin layer chromatography plates using 1 : 1
was only generated in appreciable amounts with the sulfoxides EtOAc : hexanes with 0.1% acetic acid. The silica associated
1-SO, 2-SO, and 3-SO. When this data was corrected for sulde with the product band was washed with 95% EtOAc and 5%
formation it again became apparent that an increase of lip- MeOH. The ltrate was evaporated under reduced pressure to
ophilicity of the O(3P)-precursor led to increased oxLP yield an off-white solid (101.3 mg, 69%). 1H NMR (chloroform-
26560 | RSC Adv., 2020, 10, 26553–26565
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
d,400 MHz): d (ppm) 7.89 (t, J ¼ 7.6 Hz, 2H), 7.62 (d, J ¼ 6.8 Hz, cuvette. The solution, which had an absorption of 2 or greater at
2H), 7.30 (d, J ¼ 7.8 Hz, 2H), 2.80 (t, J ¼ 7.7 Hz, 2H), 2.72 (t, J ¼ 320 nm, was degassed by purging it with argon gas for 10
7.7 Hz, 2H), 2.42 (t, J ¼ 7.3 Hz, 2H), 1.97–2.08 (m, 2H), 1.68 minutes. The solution was then irradiated with 320 ꢁ 6 nm light
(quin, J ¼ 7.3 Hz, 2H), 1.21–1.41 (m, 10H), 0.82–0.94 (m, 3H) using a 75 W xenon arc lamp focused on a monochromator. The
13C NMR (chloroform-d, 101 MHz): d (ppm) 177.3, 148.4, 146.6, reactions were carried out for less than 10% conversion. The
143.2, 143.2, 142.6, 142.6, 137.7, 137.3, 129.9, 129.7, 127.6, concentration increase of the sulde was obtained using HPLC
127.4, 36.2, 35.2, 32.9, 31.8, 31.3, 29.4, 29.2, 26.1, 22.6, 14.1, analysis. Photolysis of azoxybenzene to yield the rearranged
14.1, 26.0 MS (ESI) calculated for C24H29O3S+ m/z: 399.1988, product, o-hydroxyazobenzene, was used as
found m/z: 399.1982.
a chemical
actinometer.34,53
Synthesis of 4-(5-oxidodibenzo[b,d]thiophen-2-yl)butanoic
acid (2-SO).31 4-(Dibenzo[b,d]thiophen-2-yl)butanoic acid (2-S)
(150 mg, 0.55 mmol) was combined with DCM (10 mL). The
solution was cooled to approximately ꢀ30 C. To this cooled
Irradiation of LDL with O(3P) precursors, their respective
suldes, and other conditions
ꢂ
2 mg (2 mL) of LDL was added to a 40 mm (length) quartz test
tube. To the test tube was added 974 mL of Milli-Q treated water,
along with 22.2 mL of 0.009 M sulfoxide (1-SO, 2-SO, or 3-SO) or
suldes (1-S, 2-S, and 3-S) in 10% DMSO PBS solution. For
untreated controls, 22.2 mL of a 10% DMSO PBS solution was
added in lieu of the compound solution. The test tube was
sealed with a SubaSeal septa tting 11 mm internal diameter.
Samples which were to not be exposed to light were then
wrapped in aluminum foil. These samples were then placed in
a rotating carousel designed to suspend test tubes and the
carousel with the samples was then placed into Luzchem LZC-
4X with 14 Hitachi FL8BL-B broadly emitting uorescent
bulbs centered at 352 nm (UVA) then irradiated for 2 hours.
Immediately following irradiation samples were placed on dry
ice prior to analysis.
solution, meta-chloroperoxybenzoic acid (mCPBA) (77%)
(105 mg, 0.61 mmol) was added. The solution was stirred for
four hours, maintaining the temperature at approximately
ꢀ30 ꢂC, then allowed to stir overnight, eventually reaching
room temperature. Aer stirring, the solution was poured over
50 mL of saturated sodium bicarbonate water. Additional
DCM (20 mL) was added. The organic layer was washed with
saturated sodium bicarbonate solution (4 ꢃ 30 mL). The
organic layer was separated, dried with anhydrous MgSO4, and
ltered. The dry organic layer was evaporated under reduced
pressure to reach a volume of about 25 mL. The resulting
solution was puried by ve separate preparative thin layer
chromatography plates using 1 : 1 EtOAc : hexanes with 0.1%
acetic acid. The silica associated with the product band was
washed with 95% EtOAc and 5% MeOH. The ltrate was
evaporated under reduced pressure to yield an off white solid
(81.5 mg, 52%). 1H NMR (chloroform-d,400 MHz): d (ppm) 7.98
Cell line culture
(d, J ¼ 7.5 Hz, 1H), 7.90 (d, J ¼ 7.6 Hz, 1H), 7.79 (d, J ¼ 7.5 Hz, RAW 264.7 cell line was purchased from ATCC (Manassas, VA).
1H), 7.55–7.65 (m, 2H), 7.46–7.53 (m, 1H), 7.31 (d, J ¼ 7.9 Hz, RAW 264.7 cells were incubated at 37 ꢂC in 95% relative
1H), 2.72–2.84 (m, 2H), 2.34–2.44 (m, 2H), 1.96–2.05 (m, 2H) humidity with 5% CO2. The cells were cultured in RPMI media
13C NMR (DMSO-d6,101 MHz): d (ppm) 174.3, 147.3, 145.2, containing 10% FBS, 1% glutamine, and 1% penicillin and
142.4, 136.8, 136.5, 132.7, 129.9, 129.7, 127.5, 127.4, 122.6, streptomycin.
122.4, 34.5, 33.4, 26.1 MS (ESI) calculated for C16H15O3S+ m/z:
287.0742, found m/z: 287.0729.
RAW 264.7 cell lipid oxidation studies
RAW 264.7 cells (1 ꢃ 106) were seeded on a six-well cell culture
Common intermediate experiments with toluene
plate in 2 mL of RPMI supplemented with 10% FBS, 1% peni-
2.0–3.5 mM solutions of the sulfoxides (DBTO, 1-SO and 2-SO) cillin–streptomycin (Thermo Fisher 10 000 U mLꢀ1), and 1%
were prepared in toluene with 0.125 mM or 1 mM dodecane, GlutaMAX. The cells were then incubated for 24 hours at 37 ꢂC
which was used as an internal standard. The undissolved in a humidied atmosphere of 5% CO2. Prior to experimental
sulfoxide was ltered off, and 4.0 mL of the ltered solution conditions, the cell culture media was removed and 978 mL of
was transferred into a 5 mL quartz cuvette. The solution was PBS was added. To the well was added 22.2 mL of 0.009 M
degassed by argon sparging for 15–30 min and the solution sulfoxide (1-SO, 2-SO, or 3-SO) or suldes (1-S, 2-S, or 3-S) in 10%
was irradiated using 14 Luzchem UVA bulbs centered at DMSO PBS solution. For untreated controls, 22.2 mL of a 10%
350 nm (LZC-UVA). The concentrations of toluene oxidation DMSO PBS solution was added in lieu of the compound solu-
products were obtained by GC-FID analysis, and the increase tion. The plate was incubated for 10 minutes in the aforemen-
in the concentration of sulde was obtained through HPLC tioned conditions. Following the incubation, the plates were set
analysis.
on a rotating at carousel inside a Luzchem LZC-4C with 14
Hitachi FL8BL-B broadly emitting uorescent bulbs centered at
352 nm (UVA) and irradiated for 2 hours.
Quantum yield of deoxygenation as monitored by sulde
formation
Lipid extraction for LDL and RAW 264.7 cells
2.7–3.5 mM solutions of the sulfoxides (2-SO and 1-SO) were
prepared in acetonitrile. The undissolved sulfoxide was ltered, The 1 mL volume from the either the LDL test tubes or the
and 4.0 mL of the ltered solution was transferred into a quartz RAW 264.7 plates was collected for analysis. Cells were
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 26553–26565 | 26561

View Article Online
RSC Advances
Paper
scraped prior to removing the PBS which they were in. To the test tube, 22.2 mL of 0.009 M sulfoxide (1-SO, 2-SO, or 3-SO) in
respective vessel, 1 mL of 154 mM saline solution was added. 10% DMSO PBS solution was also added. Each condition was
The LDL test tubes were vortexed and the RAW 264.7 plates produced in duplicate. These samples were then placed in (or
were again scrapped. The saline was then combined with the on for 6-well plates) a rotating carousel which was placed into
original collection. From the RAW 264.7 combined suspen- Luzchem LZC-4C with 14 Hitachi FL8BL-B broadly emitting
sion, 200 mL was removed for protein quantication. The cells uorescent bulbs centered at 352 nm (UVA) and irradiated for
and the LDL samples were extracted using the Bligh and Dyer 2 hours. Following irradiation, samples were immediately
lipid extraction technique in the presence of 10 pmol of 2-[d5]- analyzed on HPLC to measure sulde formation. The sulde
hexadecenal.54 Extraction was conducted twice per sample. formation was subsequently quantied using standard
Aer lipid extraction, samples underwent silica extraction on curves.
LC-Si (Supelco) columns using chloroform to separate plas-
malogens from free fatty aldehydes.55 The samples were then
prepared for GC-MS analysis by derivatization with penta-
RAW 264.7 cell toxicity studies
uorobenzyl (PFB) hydroxylamine. Derivatization with PFB RAW 264.7 cells were incubated at 37 ꢂC in 95% relative humidity
hydroxylamine was performed by resuspending the reaction with 5% CO2. The cells were cultured in RPMI media containing
products in 300 mL of ethanol and 300 mL of PFB hydroxyl- 10% FBS, 1% glutamine, and 1% penicillin and streptomycin.
amine (6 mg mLꢀ1) in water. The ethanol–water mixture was The cells were split 1 : 20 on reaching ꢄ90% conuency. RAW
vortexed for 5 min at room temperature, with further incu- 264.7 cells were plated at 10 000 cells per well on six 96-well plates
bation at room temperature for 25 min. Following the incu- (CellStar – Greiner bio-one clear) with a cell suspension volume of
bation, 1.2 mL of water was added to the reaction products, 100 mL per well. Three sets of two plates (UV and No-UV) were
and the reaction products were extracted with 4 : 1 v : v marked for 5 min UV-A irradiation, 1 hour UV-A irradiation, and
cyclohexane/diethyl ether followed by resuspension in petro- 2 hour UV-A irradiation each. The plates were then incubated for
leum ether prior to GC-MS analysis.
24 hours. Aer incubation, 50 mL of 0.6% DMSO in DPBS was
introduced in the wells (nal concentration – 0.2% DMSO in
DPBS). The plates were then incubated for 10 minutes. Three
plates were irradiated with UV-A light in the photoreactor for
5 min, 1 hour, and 2 hours, respectively. Correspondingly,
thermal controls (No-UV) were placed next to the photoreactor
wrapped in aluminum foil. Aer UV-A irradiation, the UV-A along
with No-UV plates were placed back into the incubator. The
plates were then analyzed the next day using a MTS assay.
To examine the toxicity of 1-SO and 1-S, RAW 264.7 cells
were plated at 10 000 cells per well on six 96-well plates
(CellStar – Greiner bio-one clear) with a cell suspension
volume of 100 mL per well. The plates were then incubated for
24 hours. 1 mM stock solutions of 1-SO and 1-S were prepared
in 1% DMSO in DPBS. The solutions were vortexed and soni-
cated for ꢄ5 minutes. The stock solutions were then used to
prepare 3ꢃ concentrations (600 mM, 60 mM, 6 mM, 600 nM,
60 nM, 6 nM) in DPBS. Additionally, VC of 0.6% DMSO in
DPBS was prepared. Aer incubation, 50 mL of the 3ꢃ solu-
tions (nal concentrations – 200 mM, 20 mM, 2 mM, 200 nM, 20
nM, 2 nM) and VC (nal concentrations 0.2% DMSO in PBS)
were added to the wells and incubated for 10 minutes. The
plates were wrapped in aluminum foil and placed outside of
the incubator for 2 hours. The plates were then incubated for
24 hours aer the experiment and MTS assay was performed
the next day.
Aldehyde quantication for LDL and RAW-264.7 cells
Aer derivatization with PFB hydroxylamine, the samples were
analyzed by capillary gas chromatography-mass spectrometry
(GC-MS). GC-MS analysis was performed in the negative ion
chemical ionization mode (NICI) with methane as the reagent gas
using an Agilent J&W DB-1 column (12 m, 0.2 mm inner diam-
eter, 0.33 mm lm thickness) with a 6890 gas chromatograph-
5973 mass spectrometer (Agilent). The source temperature was
set at 150 ꢂC. The electron energy was 193.3 eV and the emission
current was 49.4 A. The injector and transfer lines were main-
tained at 250 ꢂC and 280 ꢂC, respectively. The GC oven was
maintained at 150 ꢂꢂC for 3.5 min, increased at the rate of
ꢀ1
ꢂ
25 C min to 310 C and held at 310 ꢂC for another 5 min.
Products quantication was performed using selected ion
monitoring (SIM) by comparing integrated area of the respective
product to the integrated area produced by the internal standard
2-[d5]-hexadecenal at m/z ¼ 418. The ve products monitored
were tetradecanal (m/z ¼ 387), pentadecanal (m/z ¼ 401), 2-hex-
adecenal (m/z ¼ 413), hexadecanal (415) and octadecanal (m/z
443). All statistical comparisons between samples were made
using a one-way ANOVA with post-hoc Tukey HSD test.
Quantication of protein from RAW 264.7 cell studies
Utilizing the 200 mL aliquot from the RAW 264.7 cell studies,
a Bradford assay was conducted using 25 mL of the 200 mL
sample, in duplicate, following the standard procedure.56
Cell viability was determined by an MTS assay. A fresh solution
of 2 mg mLꢀ1 MTS solution in DPBS and 0.92 mg mLꢀ1 PMS
solution in DPBS was prepared. 25 mL of MTS solution (100 m;L of
PMS solution for every 2 mL MTS solution) was added to each
well. The plates were then placed back in the incubator for ꢄ1.5 h
and analyzed using Flexstation 3 multimode plate reader and
Quantication of sulde formation in 6-well plates and quartz
test tubes
To either the well of a 6-well plate (for RAW 264.7 cells) or percent viability was calculated. Statistical analysis was performed
a 40 mm quartz test tube (for LDL) 978 mL of PBS (RAW 264.7 ad-hoc t-tests were performed with Welch correction using
cells) or Milli-Q treated water (LDL) was added. To the well or GraphPad Prism. Cell viability for No-UV (control) plates were
26562 | RSC Adv., 2020, 10, 26553–26565
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
calculated using (Abssample ꢀ Absblank) ꢃ 100/(AbsVC ꢀ Absblank
)
9 M. Schieber and S. Chandel Navdeep, ROS Function in Redox
Signaling and Oxidative Stress, Curr. Biol., 2014, 24(10),
R453–R462.
and
for
UV
plates
were
calculated
using
(Abssample ꢀ Absblank) ꢃ 100/(Abs(VC-NoUV) ꢀ Absblank(NoUV)
)
10 Y. Posen, V. Kalchenko, R. Seger, A. Brandis, A. Scherz and
Y. Salomon, Manipulation of redox signaling in
mammalian cells enabled by controlled photogeneration of
reactive oxygen species, J. Cell Sci., 2005, 118(9), 1957–1969.
11 J. Korang, I. Emahi, W. R. Grither, S. M. Baumann,
D. A. Baum and R. D. McCulla, Photoinduced DNA
cleavage by atomic oxygen precursors in aqueous
solutions, RSC Adv., 2013, 3(30), 12390–12397.
12 M. Zhang, G. E. Ravilious, L. M. Hicks, J. M. Jez and
R. D. McCulla, Redox Switching of Adenosine-50-
phosphosulfate Kinase with Photoactivatable Atomic
Oxygen Precursors, J. Am. Chem. Soc., 2012, 134(41), 16979–
16982.
Visualization of RAW 264.7 cells pre- and post-UV treatment
Cells were plated on two 10 ꢃ 200 mm cell culture dishes
(Cellstar by Greiner bio-one) in 5 mL media (cell density 88 000
cells per mL) and incubated for 24 hours. One culture dish was
irradiated in a Luzchem photoreactor with 14 UV-A LZC bulbs
for 2 hours and a control dish was wrapped in foil and kept
adjacent to the photoreactor. The dishes were then visualized
using a Olympus BX60 Microscope and images were captured at
200ꢃ magnication.
Conflicts of interest
13 T. Finkel, Signal transduction by reactive oxygen species, J.
Cell Biol., 2011, 194(1), 7–15.
There are no conicts to declare.
14 K. Venkataraman, S. Khurana and T. C. Tai, Oxidative stress
in aging–matters of the heart and mind, Int. J. Mol. Sci., 2013,
14(9), 17897–17925.
15 V. J. Thannickal and B. L. Fanburg, Reactive oxygen species
in cell signaling, Am. J. Physiol.: Lung Cell. Mol. Physiol., 2000,
279(6), L1005–L1028.
16 G. Testa, F. Cacciatore, G. Galizia, D. Della-Morte,
F. Mazzella, A. Langellotto, et al., Waist Circumference but
Not Body Mass Index Predicts Long-Term Mortality in
Elderly Subjects with Chronic Heart Failure, J. Am. Geriatr.
Soc., 2010, 58(8), 1433–1440.
Acknowledgements
This work was supported by the National Science Foundation
under CHE-1900417. A portion of this study was supported by
research funding from the National Institutes of Health R01
GM-115553 to D. A. F.
References
1 R. W. Redmond and J. N. Gamlin, A Compilation of Singlet 17 J. K. Paik, J. S. Chae, R. Kang, N. Kwon, S. H. Lee and
Oxygen Yields from Biologically Relevant Molecules,
Photochem. Photobiol., 1999, 70(4), 391–475.
2 M. C. DeRosa and R. J. Crutchley, Photosensitized singlet
J. H. Lee, Effect of age on atherogenicity of LDL and
inammatory markers in healthy women, Nutr., Metab.
Cardiovasc. Dis., 2013, 23(10), 967–972.
oxygen and its applications, Coord. Chem. Rev., 2002, 233– 18 N. J. Pagidipati and T. A. Gaziano, Estimating deaths from
234, 351–371.
cardiovascular disease: a review of global methodologies of
3 B. V. Chernyak, D. S. Izyumov, K. G. Lyamzaev,
mortality measurement, Circulation, 2013, 127(6), 749–756.
A. A. Pashkovskaya, O. Y. Pletjushkina, Y. N. Antonenko, 19 X. Zhang, Y. Yuan, L. Jiang, J. Zhang, J. Gao, Z. Shen, et al.,
et al., Production of reactive oxygen species in
mitochondria of HeLa cells under oxidative stress, Biochim.
Biophys. Acta, Bioenerg., 2006, 1757(5), 525–534.
Endoplasmic reticulum stress induced by tunicamycin and
thapsigargin protects against transient ischemic brain
injury: involvement of PARK2-dependent mitophagy,
Autophagy, 2014, 10(10), 1801–1813.
4 J. J. P. Gille and H. Joenje, Cell culture models for oxidative
stress: superoxide and hydrogen peroxide versus 20 M. T. Bourdillon, B. A. Ford, A. T. Knulty, C. N. Gray,
normobaric hyperoxia, Mutat. Res., DNAging: Genet. Instab.
Aging, 1992, 275(3), 405–414.
5 A. Pinto, Y. Mace, F. Drouet, E. Bony, R. Boidot, N. Draoui,
et al., A new ER-specic photosensitizer unravels 1O2-
M. Zhang, D. A. Ford, et al., Oxidation of Plasmalogen,
Low-Density Lipoprotein and RAW 264.7 Cells by
Photoactivatable Atomic Oxygen Precursors, Photochem.
Photobiol., 2014, 90(2), 386–393.
driven protein oxidation and inhibition of deubiquitinases 21 J. Korang. Photochemical Generation Of Atomic Oxygen
as a generic mechanism for cancer PDT, Oncogene, 2015,
35, 3976.
6 R. J. Aitken and S. D. Roman, Antioxidant systems and
O(3P) in Aqueous Medium and Its Biological Application.
Doctor of Philosphy thesis, Saint Louis University, Ann
Arbor, MI, 2012.
oxidative stress in the testes, Oxid. Med. Cell. Longevity, 22 E. Lucien and A. Greer, Electrophilic Oxidant Produced in
2008, 1(1), 15–24.
the
Photodeoxygenation
of
1,2-Benzodiphenylene
7 M. P. Murphy, How mitochondria produce reactive oxygen
species, Biochem. J., 2009, 417(1), 1–13.
8 I. Liguori, G. Russo, F. Curcio, G. Bulli, L. Aran, D. Della-
Morte, et al., Oxidative stress, aging, and diseases, Clin.
Interventions Aging, 2018, 13, 757–772.
Sulfoxide, J. Org. Chem., 2001, 66(13), 4576–4579.
23 K. B. Thomas and A. Greer, Gauging the Signicance of
Atomic Oxygen [O(3P)] in Sulfoxide Photochemistry. A
Method for Hydrocarbon Oxidation, J. Org. Chem., 2003,
68(5), 1886–1891.
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 26553–26565 | 26563

View Article Online
RSC Advances
Paper
24 O. R. Wauchope, S. Shakya, N. Sawwan, J. F. Liebman and 38 J. Korang, W. R. Grither and R. D. McCulla,
A. Greer, Photocleavage of plasmid DNA by Photodeoxygenation of Dibenzothiophene S-oxide
dibenzothiophene S-oxide under anaerobic conditions, J.
Sulfur Chem., 2007, 28(1), 7.
Derivatives in Aqueous Media, J. Am. Chem. Soc., 2010,
132(12), 4466–4476.
25 G. Bucher and J. C. Scaiano, Laser Flash Photolysis of 39 S.
Pyridine N-Oxide: Kinetic Studies of Atomic Oxygen [O(3P)]
in Solution, J. Phys. Chem., 1994, 98(48), 12471–12473.
26 R. D. McCulla and W. S. Jenks, Deoxygenation and Other
Photochemical Reactions of Aromatic Selenoxides1, J. Am.
Chem. Soc., 2004, 126(49), 16058–16065.
Stadelmann-Ingrand,
R.
Pontcharraud
and
B. Fauconneau, Evidence for the reactivity of fatty
aldehydes released from oxidized plasmalogens with
phosphatidylethanolamine to form Schiff base adducts in
rat brain homogenates, Chem. Phys. Lipids, 2004, 131(1),
93–105.
27 E. M. Rockafellow, R. D. McCulla and W. S. Jenks, 40 K. M. Wynalda and R. C. Murphy, Low-concentration ozone
Deoxygenation
of
dibenzothiophene-S-oxide
and
reacts with plasmalogen glycerophosphoethanolamine
lipids in lung surfactant, Chem. Res. Toxicol., 2010, 23(1),
108–117.
dibenzoselenophene-Se-oxide: a comparison of direct and
sensitized photolysis, J. Photochem. Photobiol., A, 2008,
198(1), 45–51.
41 O. H. Morand, R. A. Zoeller and C. R. Raetz, Disappearance
of plasmalogens from membranes of animal cells
subjected to photosensitized oxidation, J. Biol. Chem.,
1988, 263(23), 11597–11606.
42 W. Jira and G. Spiteller, Plasmalogens and their oxidative
degradation products in low and high density lipoprotein,
Chem. Phys. Lipids, 1996, 79(2), 95–100.
28 D. D. Gregory, Z. Wan and W. S. Jenks, Photodeoxygenation
of Dibenzothiophene Sulfoxide: Evidence for a Unimolecular
SꢀO Cleavage Mechanism 1, J. Am. Chem. Soc., 1997, 119(1),
94–102.
29 S. M. Omlid, S. A. Dergunov, A. Isor, K. L. Sulkowski,
J. T. Petroff, E. Pinkhassik and R. D. McCulla, Evidence for
diffusing atomic oxygen uncovered by separating reactants 43 W. M. Sharman, C. M. Allen and J. E. van Lier, Photodynamic
with
a
semi-permeable nanocapsule barrier, Chem.
therapeutics: basic principles and clinical applications, Drug
Discovery Today, 1999, 4(11), 507–517.
Commun., 2019, 55(12), 1671–1846.
30 J. T. Petroff II, K. N. Skubic, C. K. Arnatt and R. D. McCulla, 44 R. L. Lipson, E. J. Baldes and M. J. Gray, Hematoporphyrin
Asymmetric Dibenzothiophene Sulfones as Fluorescent
derivative for detection and management of cancer,
Nuclear Stains, J. Org. Chem., 2018, 83(22), 14063–14068.
Cancer, 1967, 20(12), 2255–2257.
31 J. T. Petroff 2nd, S. Grady, C. K. Arnatt and R. D. McCulla, 45 T. Wei, C. Chen, J. Hou, W. Xin and A. Mori, Nitric oxide
Dibenzothiophene Sulfone Derivatives as Plasma
induces oxidative stress and apoptosis in neuronal cells,
Membrane Dyes, Photochem. Photobiol., 2020, 96(1), 67–73.
Biochim. Biophys. Acta, Mol. Cell Res., 2000, 1498(1), 72–79.
32 J. Liu, S. Hu, W. Zhao, Q. Zou, W. Luo, W. Yang, et al., Novel 46 R. A. Zoeller, S. Rangaswamy, H. Herscovitz, W. B. Rizzo,
Spectrally Stable Saturated Blue-Light-Emitting Poly
[(uorene)-co-(dioctyldibenzothiophene-S,S-dioxide)]s,
Macromol. Rapid Commun., 2010, 31(5), 496–501.
A. K. Hajra, A. K. Das, et al., Mutants in a macrophage-like
cell line are defective in plasmalogen biosynthesis, but
contain functional peroxisomes, J. Biol. Chem., 1992,
267(12), 8299–8306.
33 X. Zheng, S. M. Baumann, S. M. Chintala, K. D. Galloway,
J. B. Slaughter and R. D. McCulla, Photodeoxygenation of 47 T. Ikehara, M. Nakahashi, S. Zehong, M. Akutagawa,
dinaphthothiophene, benzophenanthrothiophene, and
benzonaphthothiophene S-oxides, Photochem. Photobiol.
Sci., 2016, 15(6), 791–800.
K. Tsuchiya, M. Kitamura, et al., Effects of UV-A LED light
irradiation on growth of cultured RAW 264.7 cells, Toxicol.
Environ. Chem., 2015, 97(2), 243–255.
34 J. T. Petroff II, S. M. Omlid, S. M. Chintala and R. D. McCulla, 48 S. A. Marchitti, Y. Chen, D. C. Thompson and V. Vasiliou,
Wavelength dependent photochemistry of expanded
chromophore and asymmetric dibenzothiophene S-oxide
derivatives, J. Photochem. Photobiol., A, 2018, 358, 130–137.
Ultraviolet radiation: cellular antioxidant response and the
role of ocular aldehyde dehydrogenase enzymes, Eye
Contact Lens, 2011, 37(4), 206–213.
35 S. M. Chintala, J. T. Petroff Ii, A. Barnes and R. D. McCulla, 49 M. N. Jones, Surfactants in membrane solubilisation, Int. J.
Photodeoxygenation of phenanthro[4,5-bcd]thiophene S- Pharm., 1999, 177(2), 137–159.
oxide, triphenyleno[1,12-bcd]thiophene S-oxide and perylo 50 K. Parsi, Interaction of detergent sclerosants with cell
[1,12-bcd]thiophene S-oxide, J. Sulfur Chem., 2019, 1–13. membranes, Phlebology, 2015, 30(5), 306–315.
36 Z. Wan and W. S. Jenks, Photochemistry and Photophysics of 51 M. Nazari, M. Kurdi and H. Heerklotz, Classifying
Aromatic Sulfoxides. 2. Oxenoid Reactivity Observed in the
Photolysis of Certain Aromatic Sulfoxides, J. Am. Chem.
Soc., 1995, 117(9), 2667–2668.
surfactants with respect to their effect on lipid membrane
order, Biophys. J., 2012, 102(3), 498–506.
52 V. V. Brahmbhatt, F.-F. Hsu, J. L. F. Kao, E. C. Frank and
D. A. Ford, Novel carbonyl and nitrile products from
reactive chlorinating species attack of lysosphingolipid,
Chem. Phys. Lipids, 2007, 145(2), 72–84.
37 L. Viens, A. Athias, G. Lizard, G. Simard, S. Gueldry,
S. Braschi, et al., Effect of lipid transfer activity and
lipolysis on low density lipoprotein (LDL) oxidizability:
evidence for lipolysis-generated non-esteried fatty acids 53 N. J. Bunce, J. Lamarre and S. P. Vaish, Photorearrangement
as inhibitors of LDL oxidation, J. Lipid Res., 1996, 37(10),
2179–2192.
of azoxybenzene to 2-hydroxyazobenzene: a convenient
26564 | RSC Adv., 2020, 10, 26553–26565
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
RSC Advances
chemical actinometer*, Photochem. Photobiol., 1984, 39(4),
531–533.
Long Chain Aldehydes by GC-MS Analysis, Lipids, 2008,
43(3), 275–280.
54 E. G. Bligh and W. J. Dyer, A rapid method of total lipid 56 M. M. Bradford, A rapid and sensitive method for the
extraction and purication, Can. J. Biochem. Physiol., 1959,
37(8), 911–917.
55 V. V. Brahmbhatt, C. Nold, C. J. Albert and D. A. Ford,
Quantication of Pentauorobenzyl Oxime Derivatives of
quantitation of microgram quantities of protein utilizing
the principle of protein-dye binding, Anal. Biochem., 1976,
72(1), 248–254.
This journal is © The Royal Society of Chemistry 2020
RSC Adv., 2020, 10, 26553–26565 | 26565