Design, synthesis and biological evaluation of hybrid nitroxide-based non-steroidal anti-inflammatory drugs

Article abstract of DOI:10.1016/j.ejmech.2018.01.077

Dual-acting hybrid anti-oxidant/anti-inflammatory agents were developed employing the principle of pharmacophore hybridization. Hybrid agents were synthesized by combining stable anti-oxidant nitroxides with conventional non-steroidal anti-inflammatory drugs (NSAIDs). Several of the hybrid nitroxide-NSAID conjugates displayed promising anti-oxidant and anti-inflammatory effects on two Non-Small Cell Lung Cancer (NSCLC) cells (A549 and NCI-H1299) and in ameliorating oxidative stress induced in 661 W retinal cells. One ester-linked nitroxide-aspirin analogue (27) delivered better anti-inflammatory effects (cyclooxygenase inhibition) than the parent compound (aspirin), and also showed similar reactive oxygen scavenging activity to the anti-oxidant, Tempol. In addition, a nitroxide linked to the anti-inflammatory drug indomethacin (39) significantly ameliorated the effects of oxidative stress on 661 W retinal neurons at efficacies greater or equal to the anti-oxidant Lutein. Other examples of the hybrid conjugates displayed promising anti-cancer activity, as demonstrated by their inhibitory effects on the proliferation of A549 NSCLC cells.

Full text of DOI:10.1016/j.ejmech.2018.01.077

Accepted Manuscript
Design, synthesis and biological evaluation of hybrid nitroxide-based non-steroidal
anti-inflammatory drugs
Komba Thomas, Terry W. Moody, Robert T. Jensen, Jason Tong, Cassie L. Rayner,
Nigel L. Barnett, Kathryn E. Fairfull-Smith, Lisa A. Ridnour, David A. Wink, Steven E.
Bottle
PII:
S0223-5234(18)30096-5
10.1016/j.ejmech.2018.01.077
EJMECH 10154
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 30 November 2017
Revised Date: 15 January 2018
Accepted Date: 23 January 2018
Please cite this article as: K. Thomas, T.W. Moody, R.T. Jensen, J. Tong, C.L. Rayner, N.L. Barnett,
K.E. Fairfull-Smith, L.A. Ridnour, D.A. Wink, S.E. Bottle, Design, synthesis and biological evaluation of
hybrid nitroxide-based non-steroidal anti-inflammatory drugs, European Journal of Medicinal Chemistry
(2018), doi: 10.1016/j.ejmech.2018.01.077.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Design, Synthesis and Biological Evaluation of
Hybrid Nitroxide-Based Non-Steroidal Anti-
inflammatory Drugs
Komba Thomas1, Terry W. Moody2, Robert T. Jensen2, Jason Tong3, Cassie L Rayner3, Nigel
L Barnett3,4, Kathryn E. Fairfull-Smith1, Lisa A. Ridnour2, David A. Wink2 and Steven E.
Bottle1
1School of Chemistry, Physics and Mechanical Engineering, Queensland University of
Technology, (QUT) GPO Box 2434, Brisbane, QLD 4001, Australia
2Center for Cancer Research, National Cancer Institute, Cancer and Inflammation Program,
Frederick, MD 21702-1201, USA
3Queensland Eye Institute, South Brisbane, Queensland, Australia
4The University of Queensland, UQ Centre for Clinical Research, Herston, Queensland,
Australia
Keywords: Antioxidants/ Inflammation/ Cyclooxygenase/ Dual-Action/ Nitroxides/ A549
Non-Small Cell Lung Cancer/ NSAID/ Oxidative Stress/ Retina/ Photoreceptors/
Pharmacophore Hybridization/ Reactive Oxygen Species
1

ACCEPTED MANUSCRIPT
ABSTRACT: Dual-acting hybrid anti-oxidant/anti-inflammatory agents were developed
employing the principle of pharmacophore hybridization. Hybrid agents were synthesised by
combining stable anti-oxidant nitroxides with conventional non-steroidal anti-inflammatory
drugs (NSAIDs). Several of the hybrid nitroxide-NSAID conjugates displayed promising
anti-oxidant and anti-inflammatory effects on two Non-Small Cell Lung Cancer (NSCLC)
cells (A549 and NCI-H1299) and in ameliorating oxidative stress induced in 661W retinal
cells. One ester-linked nitroxide-aspirin analogue (27) delivered better anti-inflammatory
effects (cyclooxygenase inhibition) than the parent compound (aspirin), and also showed
similar reactive oxygen scavenging activity to the anti-oxidant, Tempol. In addition, a
nitroxide linked to the anti-inflammatory drug indomethacin (39) significantly ameliorated
the effects of oxidative stress on 661W retinal neurons at efficacies greater or equal to the
anti-oxidant Lutein. Other examples of the hybrid conjugates displayed promising anti-cancer
activity, as demonstrated by their inhibitory effects on the proliferation of A549 NSCLC
cells.
INTRODUCTION
Chronic inflammation is a key contributing factor in the mechanisms underlying the
pathogenesis of numerous inflammatory and neurodegenerative diseases such as rheumatoid
arthritis, atherosclerosis, Parkinson’s disease, and some cancers [1-10]. Chronic inflammation
is mainly characterized by oxidative stress and prolonged and excessive inflammation that
may translate into an irreparable damage to host tissues and, in severe situations, organ
malfunction or death [11, 12]. In general, non-steroidal anti-inflammatory drugs (NSAIDs)
are the most common therapeutic agents used to manage inflammatory symptoms [9, 13-15].
NSAIDs are a structurally diverse group of drugs with a similar mode of action. They exert
their therapeutic action (as anti-pyretics, anti-inflammatories and analgesics) mainly by
2

ACCEPTED MANUSCRIPT
inhibiting the cyclooxygenase (COX) enzyme [6, 14-24]. COX has two main isoforms: COX-
1 is the constitutionally expressed isoform that, under physiological conditions, is involved in
basic cytoprotective functions such as maintaining the gastrointestinal mucosal integrity.
COX-2 is inducibly expressed, mainly in response to inflammatory stimuli from infections or
injuries [6,9,13].
Most traditional NSAIDs, such as indomethacin and aspirin, inhibit both COX-1 and COX-
2 enzymes. The non-selectivity of conventional NSAID therapy can lead to adverse side
effects, notably gastrointestinal ulceration and bleeding, platelet dysfunction and renal
complications, as a result of decreased levels of cytoprotective prostaglandins [25]. Notably,
oxidative stress is recognized as a major contributor to NSAID-induced gastric mucosa
ulceration [26].
Thus, to effectively manage chronic inflammatory diseases and limit the associated
NSAID-induced damage, there is a clear need for an effective anti-oxidant intervention. Our
approach to this [27] was to exploit the anti-oxidant capacity of stable nitroxide compounds -
which is mainly attributed to the redox cycle that involves the nitroxide (A), and its
hydroxylamine (B) and oxoammonium ion (C) derivatives (Scheme 1). This redox cycle
enables nitroxides to protect biological tissues against oxidative stress, potentially via
superoxide dismutase-mimetic activity, via direct scavenging of radicals and reaction with
reactive oxygen species (ROS), and/or via the inhibition of lipid peroxidation processes and
enzymes that produce ROS such as myeloperoxidase [1, 28, 29].
Scheme 1. Reversible redox cycle of nitroxides.
3

ACCEPTED MANUSCRIPT
O
N
R
R
A
1
x
e
-
O
O
x
e
1
Redox
Cycle
O
HO
N
N
2 e- Red
2 e- Ox
R
R
R
C
R
B
Our aim in this work was to employ the pharmacophore hybridization strategy [31, 31] to
synthetically combine anti-oxidant nitroxides with a series of NSAIDs to produce novel
hybrid dual-acting, nitroxide-based NSAID agents. The hybrid agents were constructed by
either merging the two structural subunits or via cleavable (ester and amide bonds) and non-
cleavable (amine bond) linkages (Scheme 2). We anticipated that the hybrid agents would
retain the anti-inflammatory therapeutic benefits of the parent templates (anti-oxidant and
anti-inflammatory effects) and at the same time, the presence of the nitroxide unit would
minimize the drug-induced oxidative stress-related side effects. To this end, we report herein
the synthesis and some properties of NSAID pharmacophores (32 examples including aspirin,
salicylic acid, indomethacin, 5-aminosalicylic acid 5-ASA and 2-hydroxy-5-[2-(4-
trifluoromethylphenyl)-ethylaminobenzoic acid) linked with various nitroxide compounds
and the therapeutic evaluation of representative lead compounds on 3 well studied cell lines
linked to oxidative stress.
Scheme 2. The design of novel nitroxide-NSAID agents employing pharmacophore
hybridization strategies a
4

ACCEPTED MANUSCRIPT
aNitroxides (A) and NSAIDs (B) are chemically combined to produce merged, cleavable and non-cleavable
hybrid agents.
RESULTS AND DISCUSSION
Chemistry. The salicylate class of NSAIDs was first incorporated with anti-oxidant
nitroxides by taking advantage of the structural similarities of the parent templates.
Specifically, pyrroline nitroxide 2 was merged with salicylic acid 1 and acetylsalicylic acid 3
to produce new hybrid nitroxide-salicylate molecules 5-carboxy-6-hydroxy-1,1,3,3-
tetramethylisoindolin-2yloxyl 4 (salicylic acid-TMIO) and 5-carboxy-6-acetoxy-1,1,3,3-
tetramethylisoin-2-yloxyl 5 (aspirin-TMIO) as shown in Figure 1.
Figure 1. Parent templates and merged-hybrid nitroxide-salicylate target compounds
The synthesis of the merged-hybrid target compounds 4 and 5 is outlined in Scheme 3. The
5-bromo-1,1,3,3-tetramethylisoindoline precursor 6 was synthesized in three steps from
5

ACCEPTED MANUSCRIPT
commercially available phthalic anhydride by following previous established literature
procedures [32].
Scheme 3. Synthesis of merged salicylic acid TMIO 4 and aspirin TMIO 5a
aReagents and conditions: a. NaOMe/ MeOH, DMF, reflux, Ar, 85%; b. Br2, AlCl3, DCM, 0 °C, 1 h, Ar, 87%;
c. H2O2, NaHCO3, DCM/MeOH, 5 min, 98%; d. K4[Fe(CN)6], nBulmi, CuI, o-xylene, 180 °C, 3 d, 78%; e.
mCPBA, DCM, 0 °C, 90%; f. NaOH, EtOH/H2O reflux, 16 h, 89%; g. BBr3, DCM, -78 °C-RT, 18 h, 40%; h.
H2, Pd/C, THF, RT, Ar, 15 min, then i. AcCl, TEA, 0 °C, 1.5 h, 96%; j. BBr3, DCM, -78 °C-RT, 18 h, 85%; k.
LiOH, H2O/MeOH, 0 °C, RT 20 min, PbO2, 92%; l. TEA, AcCl, THF, 0 °C-RT, 3 h, Ar, 91%.
When bromoamine 6 was subjected to a copper (I) catalyzed methanolysis, in the presence
of dimethylformamide as co-solvent, the 5-methoxyamine derivative 7 was obtained in 88%
yield. Selective ring mono-bromination of 7, achieved with bromine in the presence of
anhydrous aluminium chloride, yielded the 2,5-dibromo compound 8 in good yield (78%)
which was subsequently reduced to the corresponding secondary amine 9 in excellent yield
(98%). The cyanation of 9 was achieved using potassium hexacyanoferrate(II) (K4[Fe(CN)6]),
as the cyanide source. In this case, the aminonitrile 10 was obtained in high yield by heating a
reaction mixture of 9 and K4[Fe(CN)6] at reflux in the presence of catalytic CuI with N-
butylimidazole as a co-solvent. The amino nitrile 10 was oxidized to the corresponding
6

ACCEPTED MANUSCRIPT
nitroxide derivative 11 (94%) under mild oxidation conditions using m-chloroperoxybenzoic
acid (mCPBA). Basic hydrolysis of 11 furnished the corresponding carboxylic acid 12 in 89%
yield. The target salicylic acid TMIO 4 was obtained initially by direct de-methylation of
compound 12 using boron tribromide. Using this reagent however only gave compound 4 in a
modest yield (40%). The low yield was attributed to the potential formation of a complex
between BBr3 and the nitroxide moiety. Such nitroxide-BBr3 complex formation could
initiate multiple degradation pathways for both the starting material and the desired nitroxide
targets. Alternatively, the nitroxide moiety of 12 was first protected with an acetyl protecting
group prior to the de-methylation. This was achieved by first reducing compound 12 to its
corresponding hydroxylamine 13 via palladium-catalyzed hydrogenation. The in situ
generated hydroxylamine 13 was then allowed to react with acetyl chloride in the presence
triethylamine to give the N-acetyl protected compound 14. Compound 14 was then de-
methylated using boron tribromide to afford N-acetoxy salicylic acid 15 which was
subsequently hydrolyzed to the desired salicylic acid TMIO 4 in 75% overall yield over the
three steps from 12. The target aspirin-TMIO 5 was obtained almost quantitatively by
acetylating salicylic acid-TMIO 4 with acetyl chloride in the presence of triethylamine.
Figure 2. Chemical structures of parent nitroxides and NSAIDs.
7

ACCEPTED MANUSCRIPT
Figure 2 shows the structures of the parent templates (stable nitroxides and NSAIDs) used
for the synthesis of the cleavable and non-cleavable nitroxide-NSAID hybrid conjugates. The
ester and amide-linked nitroxide-aspirin conjugates (27-29) were synthesized by reacting
aspirin with the respective nitroxide precursors (17, 18 and 20) under carbodiimide coupling
conditions (Scheme 4, Series I).
Scheme 4. Synthesis of two series of salicylate-nitroxide conjugates a,b
Series I
Series II
a Series I: Reagents and conditions: a. Nitroxide (17, 18 or 20), EDC, DMAP, DCM, RT, 1 d, 53-90%.
b Series II: Reagents and conditions: a. Nitroxide (16, 19 or 21), EDC, DMAP, DCM, RT, overnight, 83-91%; b.
NaClO2, NaH2PO4, H2O2, MeCN, 0 °C-RT, 80-88%.
With the next set of target nitroxide-salicylate conjugates (34-36), the o-formyl phenyl ester
nitroxide intermediates (31-33) were obtained following the carbodiimide coupling of
carboxylic acid nitroxides (16, 19 and 21) with salicylaldehyde 30 (Scheme 4, Series II). The
o-formyl derivatives (31-33) were then oxidized to the corresponding salicylates (34-36) in
high yields (80-88%) under Pinnick oxidation conditions. Similar structural modifications
were carried out with indomethacin 24 and benzyl 2-hydroxybenzoate 25 to give novel
indomethacin and benzyl salicylate-nitroxide cleavable conjugates (37-43, Figure 3).
8

ACCEPTED MANUSCRIPT
Figure 3. Hybrid indomethacin and benzyl salicylate-nitroxide cleavable conjugates
In addition to the salicylate and indomethacin hybrids, a number of stable nitroxide
compounds were incorporated into the 5-ASA framework as depicted in Scheme 5. Methyl 2-
hydroxy-5-nitrobenzoate 44 was first protected with an acetyl group to furnish the nitro-
diester 45 which was reduced to the amine derivative 46 under Pd/C hydrogenolysis. The
amino ester 46 was then coupled to various carboxylic acid nitroxides (16, 19, 21 and 22) to
give the amide derivatives (47-50). A mild basic hydrolysis of the amide-diesters (47-50)
afforded the amide salicylates (53-56) in high yields (83-91%). To compare the therapeutic
efficacy of the novel nitroxides conjugates to known pharmacophores, the 5-ASA conjugates
of known anti-oxidants Trolox and cinnamic acid were also prepared (51 and 52).
Scheme 5. Synthesis of amide-linked nitroxide-5-ASA conjugatesa
9

ACCEPTED MANUSCRIPT
aReagents and conditions: a. AcCl, TEA, 0 ºC, 30 min. to RT, 1 h, 91%; b. H2, Pd/C, EtOAc, 50 psi, 5 h, 90%;
c. Appropriate carboxylic acid, EDC, DMAP, DCM, RT, 1 d, 83-91%; d. 1 M NaOH/MeOH, overnight, 83-
91%.
In addition to the 5-ASA amide conjugates, the ethylamino-linked nitroxide-5-ASA non-
cleavable conjugate 63 was also synthesized (Scheme 6). The bromomethoxyamine precursor
57 was generated in two steps from bromoamine 6 following literature procedures [33]. The
methyl ester 58 was obtained following an oxidative decarboxylative coupling protocol that
involved refluxing a degassed reaction mixture of bromomethoxyamine 57 and potassium
malonate in the presence of catalytic amounts of BINAP, and DMAP [34]. Methyl ester 58
was readily converted to the nitroxide 59 using mCPBA and the carboxylic acid derivative 23
was obtained in almost quantitative yield following basic hydrolysis of methyl ester 59. EDC-
mediated coupling of carboxylic acid nitroxide 23 with the amino ester 46 furnished the
amide derivative 60. Subsequent hydrolysis of compound 60 afforded the amide-linked
salicylic acid nitroxide conjugate 61. Selective reduction of the amide group of 60 to furnish
the corresponding amine 62 was achieved using a pre-formed solution of sodium
acyloxyborohydride in refluxing dioxane. Final hydrolysis of 62 afforded the amine-linked
salicylic acid nitroxide 63 in 72% yield.
Scheme 6. Synthesis of amine-linked nitroxide-5-ASA conjugatea
10

ACCEPTED MANUSCRIPT
aReagents and conditions: a. Methyl potassium malonate, [Pd(Allyl)Cl]2, BINAP, DMAP, mesitylene, reflux,
Ar, 1d, 54%; b. mCPBA, DCM, 0 °C, 3 h, 89%; c. MeOH/NaOH, 60 °C, 2 h, 94%; d. 46, EDC, DMAP, DCM,
RT, 1 d, 95%; e. NaBH4, AcOH, dioxane, reflux, Ar, 30 min, 68%; f. 0.5 M NaOH/MeOH, overnight, 72%; g. 1
M NaOH/MeOH, overnight, 72%.
Biological evaluation. A range of novel nitroxide-NSAID hybrids were investigated
for their in vitro anti-oxidant, anti-inflammatory and anti-cancer effects. The efficacy of two
lead compounds (27 and 39) on ROS generation was tested on three different ROS-sensitive
cell types, two Non-Small Cell Lung Cancer (NSCLC) cell lines, A549 and NIH-H1299, as
well as a mouse photoreceptor cone cell line (661W retinal photoreceptor cells). The A549
NSCLC cells are a type of epithelial lung cancer that is relatively insensitive to chemotherapy
and radiation therapy, and which accounts for over 80% of lung cancers [35]. The 661W
photoreceptor cells are also highly valuable for investigating ROS injury, in this case, derived
from the high flux of oxygen in the retina that is linked to dysfunction and eventual loss of
vision.
In vitro anti-oxidant action. The anti-oxidant capacity of the nitroxide-NSAID conjugates
was determined by evaluating their ability to scavenge ROS generated in A549 NSCLC cells
via the addition of hydrogen peroxide (H2O2). Noting the limitations of the methodology, an
11

ACCEPTED MANUSCRIPT
indication of the H2O2-induced ROS produced by A549 cells was obtained through
fluorescence generated from 2,7-dichlorofluorescein diacetate (DCFH-DA) [36]. Since the
radical scavenging effect of the new hybrid compounds would be expected to arise primarily
from the nitroxide moiety, the studies were carried out by comparing Tempol, probably the
most widely studied anti-oxidant nitroxide, to the structurally-analogous hybrid compound 27
(Table 1). Both Tempol and the conjugate drug 27 lowered the increase in ROS caused by
H2O2, but had no effect on basal ROS levels. Notably, only 10 µM of the hybrid compound
27 was needed to generate similar ROS scavenging delivered by Tempol used at 10-times the
concentration (100 µM).
Table 1. ROS scavenging action of nitroxide-NSAID-conjugates on NSCLC A549 cells
Compound
Concentration
(µM)
Relative
fluorescence (%)
None
H2O2
100 + 5aa
194 + 13**
144 + 12*a
155 + 15*a
103 + 6aa
H2O2 + 27
H2O2 + Tempol
Tempol
10
100
100
10
98 + 4aa
27
* The mean value + S.D. of 8 determinations is indicated; *, p < 0.05; ** p < 0.01 relative to control; a p <
0.05; aa p < 0.01 using the Student’s t-test (relative to H2O2). This experiment is representative of 2 others.
12

ACCEPTED MANUSCRIPT
In vitro anti-inflammatory and anti-cancer effects. Epidermal growth factor receptor
(EGFR) was exploited as the target protein for evaluating the anti-inflammatory capacity of
the novel nitroxide-NSAID hybrids. As a common NSCLC drug target, the EGFR signaling
pathway is responsible for COX-2 prostaglandin (PGs) production [37-40]. The COX-
induced prostaglandin E2 (PGE2) produced by A549 cells was quantified using the enzyme-
linked immunosorbent assay. As shown in Table 2, the ester-linked conjugates (27, 34, 41
and 43) displayed strong inhibition of the COX-induced PGE2 production. In contrast, the
amide-linked conjugates (28, 29 and 39) showed moderate inhibitory action.
Table 2. Inhibitory effects of nitroxide-NSAID conjugates on COX-induced PGE2 in
NSCLC A549 cells
Compound
Concentration (µM)
PGE2, pg.mL
None
4
57 + 4
50 + 5
52 + 5
31 + 4*
54 + 6
45 + 9
34 + 6*
58 + 3
23 + 5*
30
30
30
30
30
30
30
30
5
27
28
29
34
39
41
13

ACCEPTED MANUSCRIPT
30
21 + 6*
42
The mean value + S.D. or 3 determinations each repeated in duplicate is indicated using A549 cells; p < 0.05,
*; p < 0.01, ** using the Student’s t-test.
Further COX inhibition experiments were conducted at lower concentrations (4 µM) of
selected conjugates (27 and 28) along with Tempol and the parent aspirin (Table 3). The
ester-linked conjugate 27 significantly inhibited PGE2 productions even at the lower
concentration of 4 µM. The amide-linked conjugate 28 on the other hand showed only
moderate inhibition at 4 µM. The most interesting result was the inhibitory action of the
conjugates in comparison to the aspirin parent. Notably, compound 27 was approximately an
order of magnitude more effective at inhibiting COX-induced PGE2 production in NSCLC
cells than aspirin.
Table 3. Comparison of individual components with nitroxide-NSAID conjugates on
inhibiting of COX-induced PGE2 in NSCLC A549 cells
Compound
Concentration (µM) PGE2, pg.mL
126 + 17
None
Aspirin
90
90
90
4
79 + 11
105 + 11
75 + 14*
87 + 9*
99 + 13
Tempol
Aspirin + Tempol
27
4
28
14

ACCEPTED MANUSCRIPT
The mean value + S.D. or 3 determinations each repeated in duplicate is indicated using A549 cells; p < 0.05,
*; p < 0.01, ** using the Student’s t-test.
The nitroxide-NSAID conjugates were further tested for their inhibitory action on NSCLC
proliferation using the MMT assay. The ester-linked conjugates (27, 41 and 42) inhibited
A549 cell proliferation with IC50 values in the range of 118-151 µM (Table 4). In contrast,
their amide-linked counterparts (28, 29 and 37) displayed moderate cell inhibitory potency
with IC50 values 300 µM).
Table 4. Inhibitory effects of nitroxide-NSAID conjugates on NSCLC A549 cell growth.
Compound
IC50 (µM)
130 + 23
300
27
28
29
37
41
42
300
151 + 8
121 + 12
118 + 16
The mean value + S.D. of 3 experiments each repeated in duplicate is shown using NSCLC A549 cells; p <
0.05, * using the Student’s t-test.
Further cell growth inhibitory studies using NCI-H1299 cells were conducted with
compound 27, along with Tempol and aspirin, at different concentrations (Table 5). At a
concentration of 180 µM, compound 27 displayed strong inhibitory capacity against the
15

ACCEPTED MANUSCRIPT
growth of these cells. However, no significant inhibition was observed for compound 27 at a
lower concentration (18 µM). In contrast, Tempol at 90 µM or 900 µM concentrations had
little effect on NCI-H1299 proliferation. Only a moderate inhibitory action was observed for
aspirin. However, this was only observed at higher aspirin concentrations (900 µM).
Table 5. Inhibitory effects of nitroxide-NSAID conjugates on NCI-H1299 cell growth.
Compound
Aspirin
Concentration (µM) % Growth
90
900
90
103 + 4
49 + 2*
100 + 4
98 + 5
Aspirin
Tempol
Tempol
900
90
Aspirin + Tempol
Aspirin + Tempol
27
102 + 6
53 + 4*
97 + 3
900
18
180
13 + 1*
27
The mean value + S.D. of 3 experiments each repeated in duplicate is shown using NCI-H1299 cells; p <
0.05, * using the Student’s t-test.
Cell Culture of Retinal Cells
16

ACCEPTED MANUSCRIPT
The 661W retinal photoreceptor cell line was established in T75 tissue culture flasks in
Dulbecco Modified-Eagles Medium (DMEM), supplemented with 10% fetal bovine serum
(FBS), 2 mM L-glutamine and 50 U/ml penicillin, 50 µg/ml streptomycin. Cells were
harvested for experimental use upon reaching 80-85% confluence.
The prospects for therapeutic intervention to reduce or prevent cellular damage induced by
oxidative stress was evaluated by a fluorescence-based response for cell populations using
flow cytometry and exploiting a fluorescent probe that responds to changes in the cellular
redox environment under both pro- and anti-oxidant conditions [41]. Flow cytometry
provides a convenient and rapid screening method to measure the biological efficacy of novel
anti-oxidant compounds and we have previously exploited it to monitor the overall changes
to the cellular redox environment of cells with varying metabolic activity [42].
Two representative dual-acting anti-inflammatory, anti-oxidant compounds, the aspirin-
nitroxide hybrid 27 and the indomethacin-nitroxide 39, were tested for their efficacy for
alleviating antimycin (AMC) derived fluorescence response with lutein used as a comparison.
Lutein is an effective, widely adopted anti-oxidant and free radical scavenger that displays
anti-inflammatory properties, and affords cellular protection, particularly to retinal neurons,
from oxidative injury [43, 44].
Stimulation of mitochondrial ROS production in 661W retinal cells with AMC (1 µM)
resulted in a 50% reduction in mean fluorescence intensity using a fluorescent probe that
responds to cellular oxidative status [40]. Anti-oxidant data were normalized to the maximal
effect of AMC and expressed as the % amelioration of the AMC-induced change in mean
fluorescence i.e. (100*(AMC+antioxidant – AMC) / (Control – AMC)). Lutein significantly
reduced the effects of AMC on probe fluorescence by 89.7±12.4%; p = 0.0002 (NB. Larger
% represents better response). The indomethacin-nitroxide hybrid 39 significantly
17

ACCEPTED MANUSCRIPT
ameliorated the effects of AMC on probe fluorescence at each concentration tested;
(97.6±12.1% for 10 µM; p = 0.0003; 82.3 ± 5.5% for 50 µM; p = 0.0015; 88.4 ± 14.0% for
100 µM; p = 0.0007), while compound 27 also produced a dose dependent effect, it
significantly lowered the impact of AMC only at 100 µM (31.1 ± 15.0% for 10 µM; p =
0.4318; 46.0 ± 20.7% for 50 µM; p = 0.1099; 92.8 ± 11.9% for 100 µM; p = 0.0004) (Figure
4). Notably the indomethacin-nitroxide hybrid 39 provided greater protection against AMC at
10 µM, than that provided by the same concentration of lutein, suggesting, especially at lower
concentrations, this hybrid compound may provide greater protection against oxidative stress
than current state-of-the-art anti-oxidants.
Figure 4. Hybrid indomethacin and aspirin nitroxides and their impact on antimycin-
induced mitochondrial ROS productiona
a
Data were obtained from 3-5 independent experiments and expressed as the mean ± SEM. Statistical
comparisons were made between the effects of each experimental compound and the effect of AMC alone using
a one-way ANOVA with Dunnett’s multiple comparison test. p < 0.05 was taken to be statistically significant
for this assay.
18

ACCEPTED MANUSCRIPT
CONCLUSION
A pharmacophore hybridization strategy was successfully employed to design and
synthesise a series of 30 novel potential dual-acting (anti-oxidant and anti-inflammatory)
nitroxide-NSAID conjugates. Selected novel hybrids were evaluated for their antioxidant,
anti-inflammatory and anticancer effects on A549 Non-Small Cell Lung Cancer cells. Several
nitroxide conjugates displayed significant antioxidant effects by inhibiting ROS generated by
A549 cells. While the ester-linked conjugates inhibited of the COX enzyme, the amide-linked
counterparts delivered only moderate inhibition. Notably, the nitroxide conjugate 27 provides
better inhibition of the COX enzyme than parent aspirin. Another nitroxide hybrid (39), a
structural combination with the anti-inflammatory drug indomethacin, significantly
ameliorated the effects of oxidative stress on 661W retinal neurons at efficacies greater or
equal to the recognized anti-oxidant Lutein. Importantly, the hybrid conjugates also possess
promising anticancer effects in inhibiting the proliferation of NIH-H1299 NSCLC cells. This
work demonstrates that merged/cleavable/non-cleavable hybrid agents can deliver enhanced
therapeutic efficacy with multiple modes of action over the individual parent species.
ASSOCIATED CONTENT
Supporting Information. Characterization data for final and representative compounds in
each series.
AUTHOR INFORMATION
Corresponding Author
*(S.E.B.) Email: s.bottle@qut.edu.au.
ACKNOWLEDGEMENT
19

ACCEPTED MANUSCRIPT
This work was supported by the Australian Research Council Centre of Excellence for Free
Radical Chemistry and Biotechnology (CE 0561607) and Queensland University of
Technology (KT, KEF, SEB). The research was also supported by the Intramural Research
Program of the NIH, Cancer and Inflammation Program (TWM, LAR, DAW).
ABBREVIATIONS USED
AcCl acetyl chloride; AMC antimycin; ASA amino salicylic acid; BINAP (±)-2,2′-
Bis(diphenylphosphino)-1,1′-binaphthalene; br broad; COX cyclooxygenase; DCM
dichloromethane; dd doublet of doublet; DMAP 4-Dimethylaminopyridine; EDC N-(3-
dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride; EtOAc ethyl acetate; EtOH
ethanol; FTIR fourier transform infrared spectroscopy; MeCN acetonitrile; MeOH methanol;
MTT [3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide]; nBulmi N-
butylimidazole; O/N overnight; PG, PGE2 Prostaglandins E2; QTOF LC-MS Quadrupole
Time of Flight Mass Spectrometer Liquid Chromatography-Mass Chromatography; RT room
temperature; TEA trimethylamine; TEMPO 2,2,6,6-tetramethylpiperidine-1-yloxyl; Tempol
4-hydroxy-2,2,6,6-tetramethylpiperidin-1-yloxyl; THF tetrahydrofuran; TMIO 1,1,3-
tetramethylisoindolin-2-yloxyl; w weak.
EXPERIMENTAL SECTION
General Procedures. All air-sensitive reactions were carried out under ultra-high pure
argon. Diethyl ether and toluene were dried by storing with sodium wire. All other solvents
were dried using the Pure Solv Micro 4 L solvent purification system (PSM-13-672).
Solvents used for extractions and silica gel column chromatography were AR grade.
Crystalline K4[Fe(CN)6]·3H2O was ground to a fine powder and then dried at 80 °C at 0.5
Torr for 10 h. All other reagents were purchased from commercial suppliers and used without
20

ACCEPTED MANUSCRIPT
1
13
further purification H and C NMR spectra were recorded with Bruker Avance 600 MHz,
400 MHz or Varian 400 MHz spectrometers and referenced to the relevant solvent peak.
HPLC was performed with a HP Agilent 1100 HPLC instrument. HRMS was performed with
an Agilent accurate Quadrupole Time of Flight Mass Spectrometer Liquid Chromatography-
Mass Chromatography (QTOF LC-MS) mass spectrometer. Formulations were calculated by
elemental analysis using a Mass Lynx 4.0 or Micromass Opus 3.6 instrument. FTIR spectra
were recorded with a Nicolet 870 Nexus Fourier Transform Infrared Spectrometer equipped
with a DTGS TEC detector and an ATR objective. All melting points were measured on a
Buchi Melting Point M-565 apparatus. The EPR spectra were recorded on a Magnettech
MiniScope MS400 spectrometer. Whereas the pyrrolidine and piperidines nitroxides (16-19)
are commercially available, the isoindoline nitroxides (20-22) were synthesized by following
previously published literature procedures [45-50]. An earlier report on the synthesis of
compound 28 provided only limited characterisation data [51] and we have previously
described the synthesis of 27 and 37 [52]. Others have also reported further data on these
compounds [53, 54].
5-Methoxy-1,1,3,3-tetramethylisoindoline 7. CuI (56.25 mg, 0.2 equiv.) was added to a
solution of 5-bromo-1,1,3,3-tetramethylisoindoline 6 (500 mg, 1.967 mmol, 1 equiv.) in DMF
(3 mL) and NaOMe (5 M in MeOH, 12 mL) under Ar. The reaction mixture was heated at
reflux for 15 h and allowed to return to RT. It was then diluted with H2O and extracted with
Et2O. The combined Et2O (4 x 40 mL) extracts were washed with brine (50 mL), dried over
anhydrous Na2SO4 and concentrated under reduced pressure. The resulting solid residue was
purified by silica column chromatography (CHCl3/EtOH, 10:0.5) to give 5-methoxy-1,1,3,3-
tetramethylisoindoline 7 as a pale white semi-solid (343.3 mg, 85%). Mp. 57-58 °C. HPLC
purity (>95%). 1H NMR (CDCl3, 400 MHz): δ = 1.45 (s, 6 H, CH3), 1.47 (s, 6 H, CH3), 3.87
(s, 3 H, OCH3), 6.65-7.39 (m, 3 H, Ar-H). 13C NMR (CDCl3, 100 MHz): δ = 31.9 ( C-CH3),
21

ACCEPTED MANUSCRIPT
32.0 (C-CH3), 55.4 (OCH3), 62.5 (CCH3), 62.8 (CCH3), 112.8 (Ar-C), 123.3 (Ar-C), 125.7
(Ar-C), 140.0 (Ar-C), 147.2 (Ar-C), 157.4 (Ar-C). HRMS (ES): m/z (%) = calcd. for
C13H20NO [M + H]+ 206.1539; found 206.1574. ATR-FTIR: υmax= 3415 (s, N-H), 1154 (s, C-
N), 1042 (C-O) cm-1.
2,5-Dibromo-6-methoxy-1,1,3,3-tetramethylisoindoline 8. A solution of 5-methoxy-
1,1,3,3-tetramethylisoindoline 7 (1.50 g, 731 µmol, 1 equiv.) in DCM (25 mL) under Ar was
cooled to 0 °C. A solution of bromine (942 µL, 1.83 mmol, 2.5 equiv.) in DCM (10 mL) was
added dropwise followed by addition of anhydrous aluminium trichloride (3.48 g, 2.56 mmol,
3.5 equiv.). The reaction mixture was stirred for 1 hr at 0 °C, then poured onto ice (40 mL)
and stirred vigorously for further 20 min. The solution was then basified to or above pH 12
with aqueous NaOH (10 M) solution and stirred for 10 min. The mixture was extracted with
DCM (4 x 50 mL), the combined DCM extracts were washed with brine (50 mL) and the
solvent removed under reduced pressure to give light yellow oil. The oil was triturated with
methanol to give 2,5-dibromo-6-methoxy-1,1,3,3-tetramethylisoindoline 8 as a yellow solid
(2.307 g, 87%). Mp. 97-98 °C. HPLC purity (93%). 1H NMR (CDCl3, 400 MHz): δ = 1.41 (s,
6 H, CH3), 1.44 (s, 6 H, CH3), 3.91 (s, 3 H, OCH3), 6.66 (s, 1 H, Ar-H), 7.3 (s, 1 H, Ar-H).
13C NMR (CDCl3, 100 MHz): δ = 28.1 (C-CH3), 28.3 (C-CH3), 56.5 (OCH3), 69.2 (C-CH3),
69.7 (C-CH3), 105.4 (Ar-C), 110.6 (Ar-C), 126.6 (Ar-C), 137.7 (Ar-C), 144.8 (Ar-C), 155.3
(Ar-C). ATR-FTIR: υmax= 3000 (m, Ar C-H), 1232 (s, C-N), 1034 (C-O) cm-1.
5-Bromo-6-methoxy-1,1,3,3-tetramethylisoindoline 9. To a suspension of 2,5-dibromo-
6-methoxy-1,1,3,3-tetramethylisoindoline 8 (900 mg, 2.479 mmol, 1 equiv.) and NaHCO3
(208 mg, 2.479 mmol, 1 equiv.) in MeOH/DCM (10:5 mL) was added dropwise aqueous
H2O2 (30%) until the observed effervescence ceased. The reaction mixture was stirred for 5
min followed by the addition of NaOH (5 M). The resulting solution was extracted with
DCM (4 x 40 mL) and the combined DCM extracts were washed with brine (50 mL), dried
22

ACCEPTED MANUSCRIPT
over anhydrous Na2SO4, and concentrated under reduced pressure to give 5-bromo-6-
methoxy-1,1,3,3-tetramethylisoindoline 9 as a beige solid (688 mg, 98%). Mp. 59-60 °C.
1
HPLC purity (>99%). H NMR (CDCl3, 600 MHz): δ = 1.42 (s, 6 H, CH3), 1.45 (s, 6 H,
13
CH3), 3.91 (s, 3 H, OCH3), 6.62 (s, 1 H, Ar-H), 7.26 (s, 1 H, Ar-H). C NMR (CDCl3, 100
MHz): δ = 31.8 (C-CH3), 32.0 (C-CH3), 56.4 (OCH3), 62.4 (C-CH3), 62.8 (C-CH3), 105.1
(Ar-C), 110.5 (Ar-C), 126.2 (Ar-C), 142.3 (Ar-C), 149.6 (Ar-C), 155.3 (Ar-C). HRMS (ES):
m/z (%) = calcd. for C13H19BrNO [M + H]+ 284.0645; found 284.0723. ATR-FTIR: υmax
=
3307 (w, N-H), 2961 (m, Ar C-H), 1307 (s, C-N), 1038 (C-O) 699 (s, C-Br) cm-1.
5-Cyano-6-methoxy-1,1,3,3-tetramethylisoindoline 10. A Schlenk vessel that contained a
mixture of 5-bromo-6-methoxy-1,1,3,3-tetramethylisoindoline 9 (2.76 g, 9.72 mmol, 1
equiv.), K4[Fe(CN)6] (837 mg, 1.94 mmol, 0.2 equiv.), CuI (223 mg, 1.17 mmol, 0.12 equiv.)
N-butylimidazole (2.5 mL, 19.45 mmol, 2 equiv.) in o-xylene (20 mL) was degassed and then
heated at refluxed at 180 °C for 3 d. The resulting mixture was allowed to return to RT before
it was diluted with water and then extracted with Et2O (4 x 60 mL). The combined Et2O
extracts were washed with brine (50 mL), dried over anhydrous Na2SO4 and concentrated
under reduced pressure. The crude residue was purified by silica gel column chromatography
(EtOAc) and recrystallized from cyclohexane to give 5-cyano-6-methoxy-1,1,3,3-
tetramethylisoindoline 10 as an off-white solid (1.75 g, 78%). Mp. 138-139 °C. HPLC purity
(>99%). 1H NMR (CDCl3, 400 MHz): δ = 1.42 (s, 6 H, CH3), 1.45 (s, 6 H, CH3), 1.73 (s, 1 H,
13
N-H), 3.94 (s, 3 H, OCH3), 6.66 (s, 1 H, Ar-H), 7.27 (s, 1 H, Ar-H). C NMR (CDCl3, 100
MHz): δ = 31.6 (C-CH3), 31.9 (C CH3), 56.2 (OCH3), 62.4 (C-CH3), 63.2 (C-CH3), 100.8
(C=N), 104.5 (Ar-C), 117.0 (Ar-C), 126.8 (Ar-C), 141.4 (Ar-C), 156.2 (Ar-C), 161.5 (Ar-C).
HRMS (ES): m/z (%) = calcd. for C14H19N2O [M + H]+ 231.1492; found 231.1560. ATR-
23

ACCEPTED MANUSCRIPT
FTIR: υmax= 3326 (w, N-H), 2968 (m, Ar C-H), 2221 (m, C N), 1155 (s, C-N), 1042 (C-O)
cm-1.
5-Cyano-6-methoxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 11. m-Chloroperoxybenzoic
acid (1.78 g, 6.43 mmol, 1.3 equiv.) was added to a solution of 5-cyano-6-methoxy-1,1,3,3-
tetramethylisoindoline 10 (1.14 g, 4.95 mmol, 1 equiv.) in DCM (100 mL) at 0 °C. The
cooling bath was removed after 30 min and the reaction stirred at RT for a further 1.5 h. The
DCM layer was washed with HCl (2 M), NaOH (5 M), and brine solutions (50 mL) and
before being dried over anhydrous Na2SO4. The DCM was removed under reduced pressure
and the solid residue obtained was recrystallized from EtOH to give bright yellow needles of
5-cyano-6-methoxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 11 (1.09 g, 90%). Mp. 200-201
°C. HPLC purity (>99%). HRMS (ES): m/z (%) = calcd. for C14H17N2NaO2 [M + Na]+
•
268.1182; found 268.1230. ATR-FTIR: υmax= 3048 (w, Ar C-H), 2231 (m, C=N), 1472 (N-
O), 1161 (s, C-N), 1041 (C-O) cm-1; EPR: g = 2.0009, aN = 1.81 mT.
5-Carboxy-6-methoxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 12. A suspension of 5-
cyano-6-methoxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 11 (760 mg, 3.1 mmol, 1.00 equiv.)
in NaOH (5 M, 10 mL)/EtOH (5 mL) was heated at reflux for 16 h. The reaction mixture was
cooled to RT, then diluted with H2O and washed with Et2O (2 x 40 mL). The Et2O layer was
discarded. The aqueous layer was cooled in ice bath and acidified with HCl (2 M) before it
was extracted with Et2O (50 mL x 4). The combined Et2O extracts were washed with brine
(50 mL) and dried over anhydrous Na2SO4 and evaporated under reduced pressure. The
residue was recrystallized from H2O/EtOH to give 5-carboxy-6-methoxy-1,1,3,3-
tetramethylisoindolin-2-yloxyl 12 as yellow solid (729 mg, 89%). Mp. 244-245 °C (dec.).
•
HPLC purity (>99%). HRMS (ES): m/z (%) = calcd. for C14H18NNaO4 [M + Na]+ 287.1128;
24

ACCEPTED MANUSCRIPT
found 287.1714. ATR-FTIR: υmax= 3400-2450 (m, br, OH), 2973 (m, Ar C-H), 1675 (s,
C=O), 1360 (s, C-N), 1202 (C-O) cm-1; EPR: g = 2.0009, aN = 1.83 mT.
5-Carboxy-6-hydroxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 4 from 12. BBr3 (1.9 mL,
1.89 mmol, 1 M solution in DCM, 2.5 equiv.) was added dropwise to a solution of 5-carboxy-
6-methoxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 12 (200 mg, 757 µmol, 1 equiv.) in DCM
(15 mL) at -78 °C under Ar atmosphere. The reaction was allowed to return to RT and left to
stir for 18 h. H2O was added to the resulting mixture to quench excess any BBr3 reagent. The
crude product was extracted with EtOAc (50 mL x 4) and the combined EtOAc extracts were
washed with brine (50 mL), dried over anhydrous Na2SO4, and concentrated under reduced
pressure. The crude residue was purified by silica gel column chromatography (DCM/MeOH,
6:0.4) and recrystallized from H2O/EtOH to give 5-carboxy-6-hydroxy-1,1,3,3-
tetramethylisoindolin-2-yloxyl 4 as yellow crystals (76 mg, 40%). Mp. 207-208 °C (dec.).
•
HPLC purity (>99%). HRMS (ES): m/z (%) = calcd. for C13H18NO4 [M + 2H]+ 252.1230;
found 252.1186. ATR-FTIR: υmax= 3400-2500 (m, br, OH), 2972 (m, Ar C-H), 1674 (s,
C=O), 1201 (C-O) cm-1; EPR: g = 2.0017, aN = 1.80 mT.
2-Acetoxy-5-carboxy-6-methoxy-1,1,3,3-tetramethylisoindoline 14. A reaction mixture
of 5-carboxy-6-methoxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 12 (660 mg, 2.5 mmol, 1
equiv.) and Pd/C (66 mg, 10%, 62.5 µmol, 0.025 equiv.) in THF (15 mL) was flushed with
Ar for 10 min. Then, a balloon of H2 was connected and the reaction mixture stirred for 15
min and then cooled in ice/H2O bath. TEA (697 µL, 5 mmol, 2 equiv.) and AcCl (355 µL, 5
mmol, 2 equiv.) were added dropwise and the resulting mixture was stirred for 30 min. The
cooling bath was removed and stirring was continued for a further 1 h. The reaction mixture
was filtered through Celite and concentrated in vacuo. The crude residue was stirred in
aqueous MeOH (10 mL, 2 mL H2O) for 1 h, then diluted with H2O, and extracted with
EtOAc (4 x 50 mL). The EtOAc extracts were washed with brine (40 mL), dried over
25

ACCEPTED MANUSCRIPT
anhydrous Na2SO4, and concentrated in vacuo to give 2-acetoxy-5-carboxy-6-methoxy-
1,1,3,3-tetramethylisoindoline 14 as a clear solid (738 mg, 96%). Although compound 14 was
pure enough to be used subsequent step, it was further purified by silica gel flash column
chromatography (EtOAc/CHCl3, 2:1) and recrystallized from cyclohexane. Mp. 149-150 °C.
1
HPLC purity (>99%). H NMR (CDCl3, 400 MHz): δ = 1.42 (d, 6 H, CH3), 1.48 (d, 6 H,
CH3), 2.2 (s, 3 H, C=OCH3), 4.1 (s, 3 H, OCH3), 6.78 (s, 1 H, Ar-H), 7.98 (s, 1 H, Ar-H),
13
10.81 (s, br, 1 H, CO2H). C NMR (CDCl3, 100 MHz): δ = 19.2 (CO-CH3) 25.0, 25.2 (C-
CH3), 28.6, 28.8 (C-CH3), 57.0 (OCH3), 67.9 (C-CH3), 68.4 (C-CH3), 105.0 (Ar-C), 117.4
(Ar-C), 127.5 (Ar-C), 138.1 (Ar-C), 151.7 (Ar-C), 158.3 (Ar-C), 165.2 (CO2H), 171.2
(C=OCH3). HRMS (ES): m/z (%) = calcd. for C16H21NNaO5 [M + Na]+ 330.1312; found
330.1401. ATR-FTIR: υmax= 3267 (m, br, OH), 2973 (m, Ar C-H), 1772 (s, Ac C=O), 1709
(carboxylic acid C=O) 1194 (C-O) cm-1.
2-Acetoxy-5-carboxy-6-hydroxy-1,1,3,3-tetramethylisoindoline 15. BBr3 (1.4 mL, 1.4
mmol, 1 M solution in DCM, 2.5 equiv.) was added dropwise to a solution of 2-acetoxy-5-
carboxy-6-methoxy-1,1,3,3-tetramethylisoindoline 14 (170 mg, 559 µmol, 1 equiv.) in DCM
(10 mL) at -78 °C under an Ar atmosphere. The reaction was allowed to return to RT and left
to stir for 18 h. H2O was added to the resulting mixture to quench any excess BBr3 reagent.
The crude product was extracted with EtOAc (4 x 20 mL) and the combined EtOAc extracts
were washed with brine (20 mL), dried over anhydrous Na2SO4, and concentrated under
reduced pressure. The crude residue was purified by silica gel column chromatography
(DCM/MeOH, 6:0.4) and recrystallized from cyclohexane to give 2-acetoxy-5-carboxy-6-
hydroxy-1,1,3,3-tetramethylisoindoline 15 as a white solid (153 mg, 85%). Mp. 168-169 °C.
HPLC purity (>98%).1H NMR (CDCl3, 400 MHz): δ = 1.4 (s, 6 H, CH3), 1.48 (s, 6 H, CH3),
2.2 (s, 3 H, C=OCH3), 6.77 (s, 1 H, Ar-H), 7.64 (s, 1 H, Ar-H), 10.6 (s, br, 1 H, CO2H). 13C
NMR (CDCl3, 100 MHz): δ = 19.2 (CO-CH3) 25, 25.2 ( C-CH3), 28.6, 28.8 ( C-CH3), 57.0
26

ACCEPTED MANUSCRIPT
(OCH3), 67.9 (C-CH3), 68.4 (C-CH3), 105.0 (Ar-C), 117.4 (Ar-C), 127.5 (Ar-C), 138.07 (Ar-
C), 151.7 (Ar-C), 158.3 (Ar-C), 165.2 (CO2H), 171.2 (C=OCH3). HRMS (ES): m/z (%) =
calcd. for C15H20NO5 [M + H]+ 294.1336; found 294.2269. ATR-FTIR: υmax= 3095 (m, br,
OH), 2973 (m, Ar C-H), 1737 (s, Ac C=O), 1677 (carboxylic acid C=O), 1160 (C-O) cm-1.
5-Carboxy-6-hydroxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 4 from 15. A suspension
of 2-acetoxy-5-carboxy-6-hydroxy-1,1,3,3-tetramethylisoindoline 15 (136 mg, 464 µmol, 1
equiv.) in H2O/MeOH (2 mL/2 mL) was cooled to 0 °C. LiOH (56 mg, 2.3 mmol, 5 equiv.)
was added and the reaction mixture stirred overnight while allowed to warm to RT. The
resulting solution was washed with Et2O and the Et2O layer discarded. The aqueous layer was
cooled in ice bath, acidified with HCl (2 M) and then extracted with Et2O (4 x 30 mL). The
combined Et2O extracts were stirred over PbO2 (28 mg, 116 µmol, 0.25 equiv.) for 20 min,
dried over anhydrous Na2SO4, filtered and concentrated in vacuo. The crude residue was
recrystallized from H2O/EtOH to give 5-carboxy-6-hydroxy-1,1,3,3-tetramethylisoindolin-2-
yloxyl 4 as yellow crystals (107 mg, 92%). Mp. 207-208 °C (dec.). HPLC purity (>99%).
HRMS (ES): m/z (%) = calcd. for C13H18NO4 [M + 2H]+ 252.1230; found 252.1186. ATR-
•
FTIR: υmax= 3400-2500 (m, br, OH), 2972 (m, Ar C-H), 1674 (s, C=O), 1201 (C-O) cm-1;
EPR: g = 2.0017, aN = 1.80 mT.
6-Acetoxy-5-carboxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 5. A solution of 5-carboxy-
6-hydroxy-1,1,3,3-tetramethylisoindolin-2-yloxyl 4 (89 mg, 354.4 µmol, 1 equiv.) in THF
under Ar was cooled to 0 °C. TEA (99 µL, 709 µmol, 2 equiv.) and AcCl (50.4 µL, 709
µmol, 2 equiv.) were added dropwise and the resulting mixture stirred for 3 h while allowing
to return to RT. Water was added to the reaction mixture and it was then extracted with DCM
(3 x 30 mL). The DCM extracts were washed with HCl (1 M, 15 mL), brine (20 mL) and
dried over anhydrous Na2SO4. The combined DCM was removed under reduced pressure and
the crude residue obtained was recrystallized from H2O/EtOH to give 6-acetoxy-5-carboxy-
27

ACCEPTED MANUSCRIPT
1,1,3,3-tetramethylisoindolin-2-yloxyl 5 as a yellow solid (94 mg, 91%). Mp. 206-207 °C.
•
HPLC purity (>99%). HRMS (ES): m/z (%) = calcd. for C15H19NO5 [M + H]+ 293.1258;
found 293.1221. ATR-FTIR: υmax= 3400-2600 (m, br, OH), 2972 (m, Ar C-H), 1765 (s, Ac
C=O), 1695 (carboxylic acid C=O) 1184 (C-O) cm-1; EPR: g = 2.0016, aN = 1.80 mT.
Benzyl 2-hydroxybenzoate 25. NaHCO3 (1.46 g, 17.4 mmol, 1.2 equiv.) was added to a
solution salicylic acid 1 (2 g, 14.5 mmol, 1 equiv.) in DMF (20 mL) and the resulting mixture
was stirred at 70 °C for 10 min. The temperature was reduced to 50 °C followed by the
addition of benzylbromide (1.81 mL, 15.2 mmol, 1.05 equiv.). The reaction mixture was
stirred for 4 h and then allowed to cool to RT. H2O (50 mL) was added and the crude product
was extracted with EtOAc (4 x 60 mL). The combined EtOAc extracts were washed with
brine (50 mL), dried over anhydrous Na2SO4 and evaporated under reduced pressure.
Purification of the crude residue by silica gel chromatography (Hexane/EtOAc, 5:0.2)
1
afforded compound 25 as clear oil (3.11 g, 94%). HPLC purity (>99%). H NMR (CDCl3,
400 MHz): δ = 5.42 (s, 2 H, CH2), 6.9 (m, 1 H, Ar-H), 7.4 (m, 6 H, Ar-H), 7.91 (q, 1 H, Ar-
13
H), 10.78 (s, 1 H, O-H). C NMR (CDCl3, 100 MHz): δ = 67.0 (CH2), 112.4 (Ar-C), 117.6
(Ar-C), 119.2 (Ar-C), 128.3 (Ar-C), 128.6 (Ar-C), 128.7 (Ar-C), 130.0 (Ar-C), 135.3 (Ar-C),
135.8 (Ar-C), 161.8 (Ar-C), 170.0 (C=O). ATR-FTIR: υmax= 3188 (m, br, O-H), 3000 (m, Ar
C-H), 1670 (s, C=O), 1086 and 1133 (s, C-O) cm-1.
General procedure for the synthesis of salicylic acid derivatives 27-29. A solution of
aspirin 3 (150 mg, 833 µmol, 1 equiv.), appropriate nitroxide 17, 18 or 20 (1 mmol, 1.2
equiv.), EDC (191.5 mg, 1 mmol, 1.2 equiv.), and DMAP (13 mg, 104 µmol, 0.125 equiv.) in
DCM (10 mL) was stirred under Ar for 1 d. The resulting reaction mixture was diluted
(DCM, 150 mL), washed HCl (2 M, 30 mL) and brine (30 mL) solutions, dried over
anhydrous Na2SO4 and evaporated under reduced pressure. The crude residue was purified by
28

ACCEPTED MANUSCRIPT
silica gel column chromatography (CHCl3) and then recrystallization from
cyclohexane/EtOAc (except 28) to give the corresponding salicylate-nitroxide.
2,2,6,6-tetramethylpiperidin-1-yloxyl-4-yl 2-acetoxybenzoate 27. Reddish brown solid
(251 mg, 90%). Mp. 52-53 °C. HPLC purity (>95%). HRMS (ES): m/z (%) = calcd. for
•
C18H25NO5 [M + H]+ 335.1727; found 335.1688. ATR-FTIR: υmax= 2977 (w, ArC-H), 1724
(s, C=O), 1255 (s, C-O) cm-1; EPR: g = 2.0012, aN = 1.97 mT.
2-((2,2,6,6-tetramethylpiperidin-1-yloxyl-4-yl)carbamoyl)phenyl acetate 28. Reddish
brown solid (153 mg, 55%). Mp. 50-52 °C. HPLC purity (>95%). HRMS (ES): m/z (%) =
•
calcd. for C18H26N2O4 [M + H]+ 334.1887; found 334.1850. ATR-FTIR: υmax= 3314 (m, N-
H), 2976 (w, ArC-H), 1640 (s, C=O), 1546 (s, C=O), 1229 (s, C-O) cm-1; g = 2.0012, aN =
1.96 mT.
2-((1,1,3,3-tetramethylisoindolin-2-yloxyl-5-yl)carbamoyl)phenyl acetate 29. Yellow
solid (162 mg, 53%). Mp. 101-102 °C. HPLC purity (>95%). HRMS (ES): m/z (%) = calcd.
for C21H24N2O4 [M + H]+ 368.1731; found 368.1689. ATR-FTIR: υmax= 3340 (m, N-H),
•
2971 (w, ArC-H), 1681 (s, C=O), 1543 (s, C=O), 1255 (s, C-O) cm-1; g = 2.0010, aN = 1.80
mT.
General procedure for the synthesis of formyl-nitroxides 31-33. Following similar
procedure as for 27-29, compounds 31-33 were obtained from 30 (1.31 mmol, 1.05 equiv.)
and the appropriate carboxy-nitroxide 16, 19 and 21 (1.25 mmol, 1 equiv.).
2-Formylphenyl 2,2,5,5-tetramethylpyrrolidin-1-yloxyl-3-carboxylate 31. Yellow
crystalline solid (319 mg, 88%). HPLC purity (>95%). 1H NMR (CDCl3, 600 MHz): δ = 7.19
(s, br, 1 H, Ar-H), 7.47 (d, br, 1 H, Ar-H), 7.7 (s, br, 1 H, Ar-H), 7.9 (s, br, 1 H, Ar-H), 10.1
•
(s, br, HC=O). HRMS (ES): m/z (%) = calcd. for C16H21NO4 [M + H]+ 291.1465; found
291.1428. ATR-FTIR: υmax= 3350-2400 (m, br, O-H), 1734 (s, C=O), 1688 (s, C=O), 1288 (s,
C-N), 1196 (s, C-O) cm-1; g = 2.0009, aN = 1.84 mT.
29