Polyamine-Conjugated Nitroxides Are Efficacious Inhibitors of Oxidative Reactions Catalyzed by Endothelial-Localized Myeloperoxidase

Sophie Maiocchi, Jacqueline Ku, Tom Hawtrey, Irene De Silvestro, Ernst Malle, Martin Rees,

Article

Polyamine-Conjugated Nitroxides Are Eﬃcacious Inhibitors of

Oxidative Reactions Catalyzed by Endothelial-Localized

Myeloperoxidase

Shane R. Thomas,* and Jonathan C. Morris*

Cite This: Chem. Res. Toxicol. 2021, 34, 1681 −1692

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: The heme enzyme myeloperoxidase (MPO) is a key mediator of endothelial

dysfunction and a therapeutic target in cardiovascular disease. During inﬂammation, MPO

released by circulating leukocytes is internalized by endothelial cells and transcytosed into

the subendothelial extracellular matrix of diseased vessels. At this site, MPO mediates

endothelial dysfunction by catalytically consuming nitric oxide (NO) and producing reactive

•

oxidants, hypochlorous acid (HOCl) and the nitrogen dioxide radical ( NO ). Accordingly,

there is interest in developing MPO inhibitors that eﬀectively target endothelial-localized

MPO. Here we studied a series of piperidine nitroxides conjugated to polyamine moieties as

novel endothelial-targeted MPO inhibitors. Electron paramagnetic resonance analysis of cell

lysates showed that polyamine conjugated nitroxides were eﬃciently internalized into

endothelial cells in a heparan sulfate dependent manner. Nitroxides eﬀectively inhibited the

consumption of MPO’s substrate hydrogen peroxide (H O ) and formation of HOCl

catalyzed by endothelial-localized MPO, with their eﬃcacy dependent on both nitroxide and

conjugated-polyamine structure. Nitroxides also diﬀerentially inhibited protein nitration

•

catalyzed by both puriﬁed and endothelial-localized MPO, which was dependent on NO scavenging rather than MPO inhibition.

Finally, nitroxides uniformly inhibited the catalytic consumption of NO by MPO in human plasma. These studies show for the ﬁrst

time that nitroxides eﬀectively inhibit local oxidative reactions catalyzed by endothelial-localized MPO. Novel polyamine-conjugated

nitroxides, ethylenediamine-TEMPO and putrescine-TEMPO, emerged as eﬃcacious nitroxides uniquely exhibiting high endothelial

cell uptake and eﬃcient inhibition of MPO-catalyzed HOCl production, protein nitration, and NO oxidation. Polyamine-conjugated

nitroxides represent a versatile class of antioxidant drugs capable of targeting endothelial-localized MPO during vascular

inﬂammation.

INTRODUCTION

hypochlorous acid (HOCl), (ii) the oxidation of nitrite

■

−

•

(

NO ) into the nitrogen dioxide radical ( NO ) that mediates

Considerable experimental and clinical evidence has estab-

lished the innate immune heme enzyme, myeloperoxidase

MPO), as a key mediator of endothelial dysfunction. This is

protein tyrosine nitration, and (iii) the catalytic consumption

of the homeostatic signaling biomolecule, nitric oxide

(

,2,13

(

NO).

As MPO catalyzes these deleterious oxidative

a major pathogenic event in cardiovascular disease that

increases the adverse clinical event risk of coronary artery

reactions in a localized manner in endothelial and

7,13

disease patients. Multiple clinical studies show an inverse

subendothelial compartments,

there is a need to discover

relationship between circulating MPO levels and clinical

eﬃcacious MPO inhibitors that are capable of accumulating in

these compartments at suﬃcient concentrations.

−6

indices of endothelial dysfunction.

MPO, which is released

extracellularly by activated circulating leukocytes during

inﬂammation, transcytoses across the endothelium, accumulat-

ing within endothelial cells and the subendothelial extracellular

Piperidine nitroxides are a class of cell-permeable small

molecule antioxidants, which can attenuate oxidative stress via

multiple mechanisms, including (i) superoxide dismutase

matrix of the vascular wall. Thus, MPO and biomarkers of its

oxidative reactions are enriched in the endothelial and

subendothelial compartments of vessels of cardiovascular

Received: March 4, 2021

Published: June 4, 2021

,8−12

disease patients.

At this site, and in the presence of its

substrate hydrogen peroxide (H O ), MPO catalyzes several

deleterious oxidative reactions that mediate endothelial

dysfunction, including the following: (i) the conversion of

−

chloride ions (Cl ) into the potent tissue-damaging oxidant

2021 American Chemical Society

681

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

SOD) mimetic activity (Scheme 1A), (ii) direct radical

scavenging, and (iii) participating in radical−radical termi-

Article

(

Additionally, the use of nitroxides as therapeutics is limited

by poor in vivo bioavailability. High doses (up to millimolar

concentrations) of the widely used piperidine nitroxide 4-

hydroxy-TEMPO (tempol) are typically employed to achieve

Scheme 1. SOD Mimetic Activity and MPO Inhibition by

Nitroxides

beneﬁcial eﬀects in animal models of inﬂammation. We

hypothesized that functionalizing nitroxides with cellular-

targeting moieties enhances endothelial internalization and

improves their ability to inhibit oxidative reactions catalyzed by

endothelial-localized MPO. Accordingly, we report on the

chemical synthesis of a novel series of piperidine nitroxides

conjugated to cationic polyamines with the aim of increasing

their endothelial accumulation. This approach is rationalized

ﬁrst by reports that conjugation of enzymes or small molecules

to polyamines enhances endothelial cell uptake and transport

2,23

across the blood−brain barrier

and second that cellular

uptake of both polyamines and MPO is reportedly mediated by

heparan sulfate proteoglycans (HSPGs).

,24

Thus, we examine

whether polyamine conjugation confers enhanced cellular

uptake relative to the widely employed unconjugated nitroxide

tempol and its dependency on HSPGs. We further examined

the inﬂuence of polyamine conjugation on inhibition of MPO-

catalyzed oxidative reactions in both isolated and endothelial

cell systems.

Our studies establish for the ﬁrst time that nitroxides

eﬀectively inhibit oxidative reactions catalyzed by endothelial-

localized MPO and we identify novel polyamine-conjugated

nitroxides, ethylenediamine-TEMPO and putrescine-TEMPO,

as eﬃcacious nitroxides uniquely exhibiting high endothelial

uptake and eﬃcient inhibition of MPO-catalyzed HOCl

production, protein nitration, and NO oxidation.

(

A) Nitroxides are superoxide dismutase mimetics whereby they

•−

catalytically dismutate superoxide anion radical (O₂). (B) MPO

consumes H O₂to form MPO compound I (reaction 1), which

mediates a two-electron oxidation of halides into the respective

hypohalous acid (e.g., Cl⁻into HOCl) via the halogenation cycle

(

reaction 2). MPO also oxidizes small molecule substrates via its

peroxidase cycle through two successive one-electron reactions

reaction 3 followed by reaction 5, e.g., one-electron oxidation of

(

−

•

−

NO₂into the nitrogen dioxide radical [ NO ] or NO into NO ).

Nitroxides (R−NO ) are competitive substrates for MPO compound

I, resulting in the formation of MPO compound II and the

•

EXPERIMENTAL METHODS

■

corresponding oxoammonium cation (RN O) (reaction 4).

General Chemical Synthesis Information. Chemical reagents

were commercially purchased and used without further puriﬁcation.

Solvents were puriﬁed according to well-established procedures.

Eﬃcacious nitroxide MPO inhibitors are poor substrates for MPO

compound II such that reaction 6 is slowed, and in the absence of

other peroxidase substrates, compound II accumulates, which does

not participate in the halogenation cycle. Nitroxides inhibit in a

catalytic fashion as the oxoammonium cation can be recycled back to

Methanol was distilled from magnesium and stored over 3 Å

molecular sieves under an atmosphere of nitrogen or argon.

Chloroform was passed through basic alumina (70% w/v) directly

before use. Dichloromethane (DCM) was obtained from a Pure Solv

dry solvent system (Innovative Technology, Inc. model no. PS-MD-

7). All reactions were followed by analytical thin layer chromatog-

raphy (TLC) carried out with the use of Fluka PET-foils Silica Gel 60

and compounds visualized by irradiation with short-wavelength UV

light and stained with Ninhydrin Spray dip (3.3 g of ninhydrin/100

mL of ethanol). Flash chromatography was performed by using

chromatographic silica media LC60A, 40−63 μm, with appropriate

solvent systems. Solvent was eluted with a Thomson SINGLE StEP

pump at the manufacturer’s recommended ﬂow rate (Thomson

Instrument Co., Oceanside, CA, USA). Deactivated silica gel was

prepared by mixing silica gel with 5% (v/w) of triethylamine. Nuclear

magnetic resonance (NMR) spectra were collected on a 300 and 75

MHz Bruker DPX 300 or Bruker Avance III 300 and are reported in

parts per million (ppm). NMR spectra of nitroxyl radicals were

obtained by adding a drop of phenylhydrazine to form the

corresponding nonradical hydroxylamine. Compound purity was

>95% as judged by HPLC.

the nitroxide radical through a reaction with H O (reaction 7).

nation reactions. Nitroxides were also recently identiﬁed as a

new class of competitive substrate inhibitors of MPO (Scheme

B, reactions 4 and 6), where they were reported to inhibit

8,15−18 15

HOCl production.

Rees et al. established that

piperidine nitroxides competitively react with MPO compound

I (Scheme 1B, reaction 4 versus reaction 2) to yield the

corresponding nitroxide oxoammonium cation and MPO

compound II, which does not participate in MPO’s

halogenation cycle (Scheme 1B). Kinetic studies conﬁrmed

that nitroxides are poor substrates for MPO compound II;

thus, in an isolated system with no peroxidase substrates, MPO

compound II accumulates and HOCl production is limited.

However, in complex biological systems, several physiological

substrates (e.g., tyrosine, urate, NO , O ) can reduce MPO

−

•−

compound II to regenerate native MPO, which is capable of

participating in HOCl production (Scheme 1B, reaction 5

General Synthetic Procedure for Compounds 11a−11e and

2. Ti(OiPr) (1.2 equiv) was added dropwise to solid 4-oxo-

17,20

TEMPO (1 equiv), and the mixture was stirred for 20 min. Boc

protected polyamines (2 equiv; 9a−9e, 10) were added portionwise,

and the reaction mixture was stirred for 3 h. The mixture was diluted

versus reaction 6).

As such, the ability of nitroxides to

inhibit MPO-catalyzed oxidative reactions in biological

environments where competing physiological substrates are

with methanol (1.5 M) prior to the addition of NaBH CN (1 equiv)

available is thought to be substantially limited. For the ﬁrst

and stirring for 24 h. The reaction was stopped by addition of water,

and the inorganic white precipitate was removed by ﬁltration and

washing with ethyl acetate. This mixture was evaporated in vacuo to

remove ethyl acetate. The aqueous layer was extracted with ethyl

time, in this study we address the ability of nitroxides to

directly inhibit oxidative reactions catalyzed by MPO that is

sequestered within endothelial cells.

682

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

Article

acetate, and the combined organic layers were washed with brine,

× TBST) and incubated in TBST-milk with the relevant anti-mouse

or anti-rabbit horseradish peroxidase (HRP) conjugated secondary

antibody (1:3000, Abcam) for 1 h at room temperature. Membranes

were washed (3 × TBST) and protein bands were detected with the

use of ECL Western blotting detection reagents (Amersham

Biosciences) and developed with hyperﬁlm (GE healthcare) or

visualized by using ImageQuant LAS4000 (GE Healthcare Life

Sciences). Tubulin was used to index equal protein loading.

Densitometric analyses of Western blot bands of multiple

independent experiments were carried out with ImageJ software.

Measuring Endothelial Cell Nitroxide Uptake by EPR. Conﬂuent

endothelial cells were incubated in Hanks Balanced Salt Solution (1 ×

HBSS, calcium, magnesium, no Phenol Red, Gibco, Life Technolo-

gies) and treated with nitroxides (10 μM) for 20 min at 37 or 4 °C.

Cells were then washed with ice-cold PBS and harvested into cell lysis

buﬀer (ice-cold 0.1% Triton X-100 in 1 × PBS, 350 μL) and stored at

−20 °C until EPR analysis. In some experiments, cells were

preincubated with heparin (1250 μg/mL, Sigma) for 45 min prior

to addition of nitroxides and incubation for a further 20 min.

Measurements of nitroxide concentration and protein levels in cell

lysates were carried out with EPR and BCA protein assay (Pierce,

Thermo Scientiﬁc), respectively. Cellular nitroxide concentration was

expressed as nanomoles of nitroxide per microgram of cell protein.

For EPR analysis, samples were thawed and diluted by 10% with

potassium ferricyanide (11 mM; ﬁnal concentration ∼1 mM) to

oxidize all available nitroxide to its radical and hence EPR detectable

form. Samples (250 μL) were added to a ﬂattened aqueous sample

cell (WG-814-Q; Wilmad, Buena, NJ, USA), and spectra were

acquired on a Bruker X-band EPR coupled to Xenon software (cavity,

4119HS; magnet, X-band ER073; microwave bridge, Bruker EMXplus

premium Xbridge). EPR parameters were conversion time (1.3 s),

time constant (5.24 s), sweep width (80 G), receiver gain (60 dB),

attenuation (10 dB), modulation amplitude (2 G), and sweep time

(520 s). The low-ﬁeld nitroxide spectrum was used to quantify

nitroxide concentrations alongside a standard curve using known

concentrations of the nitroxide tempol (4HT; 0−1 μM).

dried over Na SO , and concentrated in vacuo. The resulting orange

oil was adsorbed onto silica and puriﬁed by ﬂash chromatography

(

yield 30−74%).

General Synthetic Procedure for Compounds 1−6. Solutions

of puriﬁed Boc protected nitroxyl radicals in DCM (0.01 M) were

stirred with excess triﬂuoroacetic acid (10 equiv) for 4 h, and the

mixture was evaporated under reduced pressure. The resulting dark

oil was taken up in methanol/DCM (1:400), treated with anhydrous

potassium carbonate (2 g for 200 mL of DCM), and stirred overnight.

The mixture was ﬁltered via a Celite plug eluting with DCM and

concentrated in vacuo, yielding free nitroxyl radicals 1−6 as orange

oils. No further puriﬁcation was necessary.

Biological Materials and Methods. General Biological

Materials. Puriﬁed human neutrophil MPO and H O (30% w/v)

were obtained from Merck Millipore. MPO was reconstituted (0.6−2

μM) in Milli Q water and stored at 4 °C. The concentrations of H O

−

stocks were determined by spectrophotometry (H O ε = 43.6 M

240

−

cm ). H O solutions were diluted in Milli Q water and used fresh

each time. The NO donor NOC-9 (t₁_/₂2.7 min, 22 °C) was from

Santa Cruz Biotechnology. NOC-9 solutions were prepared in ice-

cold 10 mM NaOH and used fresh. Unless otherwise indicated, all

other materials were purchased from Sigma-Aldrich, were of the

highest purity available, and were used without further puriﬁcation.

For experiments, Chelex-treated phosphate buﬀer (0.1 M, pH 7.4;

“phosphate buﬀer”) was used. The nitroxides, 4-amino-TEMPO

(4AT), 4-carboxy-TEMPO (4CT), 4-hydroxy-TEMPO (4HT) (all

from Sigma-Aldrich), and mito-TEMPO (MT) (Santa Cruz Biotech),

were of the highest purity available. Novel nitroxides 1−6 were

synthesized as described above. The identity and purity (>95%) of the

synthesized novel nitroxides were veriﬁed via NMR, HRMS, and

HPLC.

Preparation of Human Plasma. Plasma was obtained after

centrifugation (5000 rpm, 10 min at 4 °C) of freshly isolated

heparinized blood donated by healthy consenting adult volunteers as

per a protocol approved by the UNSW Human Ethics Review

Committee. Aliquots of the isolated plasma were immediately frozen

and stored at −80 °C. Plasma aliquots were thawed immediately

before experiments and used within 1 h of thawing for MPO NO

oxidase experiments.

Measurement of H O Consumption by Isolated MPO. MPO-

catalyzed H O consumption was measured with an H O -speciﬁc

electrode (ISO-HPO-2) interfaced to an Apollo 4000 free radical

analyzer (World Precision Instruments) and quantiﬁed via LabScribe

3 software. Reactions were run at room temperature (∼22 °C) in

stirred air-saturated Chelex-treated 0.1 M phosphate buﬀer solutions

containing NaCl (100 mM). Reactions contained H O (25 μM) and

Endothelial Cell Culture. Primary bovine aortic endothelial cells

(

ECs; Gelantis or Lonza) were cultured on gelatin-coated (Sigma;

.05% w/v in PBS, 20 min, 22 °C) tissue culture ﬂasks (75 cm ,

Nunc, Thermo Scientiﬁc) in endothelial basal medium (EBM; Lonza)

with all supplements added except hydrocortisone (endothelial cell

growth medium microvascular, EGM-MV; Lonza). Endothelial cells

methionine (300 μM; added to scavenge HOCl) in the absence or

presence of nitroxides (1 μM) and were initiated by the addition of

MPO (20 nM). H O consumption was recorded for up to 6 min

were maintained at 37 °C in a 5% CO humidiﬁed atmosphere and

used for experiments between passages 3 and 9. For experiments, ECs

were plated on gelatin-coated (0.05% w/v in PBS) 12-well plates

following MPO addition.

Measurement of HOCl Production by Isolated MPO. HOCl

production catalyzed by isolated MPO was measured by the 2-nitro-5-

(Nunc, Thermo Scientiﬁc) and 60 mm Petri dishes (Corning or

thiobenzoic acid (TNB) and iodide-catalyzed tetramethylbenzidine

Nunc, Thermo Scientiﬁc) and grown until conﬂuent.

(TMB) assays. For the TNB assay, the TNB reagent was prepared

by dissolving DTNB in aqueous 0.05 M NaOH solution to make a

ﬁnal concentration of 1 mM DTNB (TNB is formed via hydrolysis of

DTNB in NaOH) and then diluting 1:40 into phosphate buﬀer.

Reactions were run in phosphate buﬀer and contained MPO (100

Western Blotting. Puriﬁed proteins or cell lysates in 1 × SDS

sample buﬀer were boiled for 5 min at 100 °C and centrifuged for 5

min (12 000 rpm), and proteins were separated by SDS-PAGE using

either 3−8% Tris acetate gels or 10% Bis-Tris gels (NuPAGE,

Thermo Fisher). Proteins were resolved at 165−180 V over 55−65

min with either NuPAGE SDS Tris acetate running buﬀer or MOPS

buﬀer, respectively. Resolved proteins were transferred onto nitro-

cellulose membranes by using the iBlot Gel transfer system according

to the manufacturer’s instructions (Invitrogen). Membranes were

blocked with 5% nonfat dry milk in Tris buﬀered saline with Tween-

−

nM), taurine (20 mM), and Cl (100 mM) in the absence or

presence of nitroxides (10 μM) and/or heparin (1 mg/mL), and they

were started by H O (50 μM) addition. After 5 min the reaction was

−1

terminated by the addition of catalase (50 μg mL ), and an aliquot of

reaction mixture was added to TNB reagent in a 96-well microtiter

plate. The extent of TNB oxidation to DTNB arising from HOCl-

mediated oxidation of taurine into taurine chloramine was quantiﬁed

after 5 min at 412 nm with a FLUOstar Omega microplate reader

(BMG Labtech). Data were expressed as a percent of the level of

TauNHCl formed by MPO in the absence of nitroxides. For the TMB

assay, reactions were performed in PBS (pH 7.4) containing 140 mM

0 (TBST) for at least 30 min. Membranes were then replaced with

% nonfat dry milk in TBST (TBST-milk) containing the required

primary antibody (i.e., mouse monoclonal anti-HOCl oxidized

protein, clone 2D10G9 (that does not cross-react with epitopes

generated by oxidative reactions involving nitrating species, transition

−

metals, or lipid peroxidation), 1:20; mouse monoclonal antitubulin

Sigma), 1:10000; mouse monoclonal anti-3-nitrotyrosine (Merck

Millipore), 1:3000; rabbit polyclonal anti-3-nitrotyrosine (Abcam),

:3000) and incubated overnight at 4 °C. Membranes were washed (3

Cl , MPO (10 nM), and taurine (5 mM) in the absence or presence

(

of nitroxides (0.1−10 μM). Reactions were initiated by the addition

of H O (50 μM) and terminated after 6 min by the addition of

−

catalase (20 μg mL ). An aliquot of reaction mixture was added to

683

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

Article

the TMB reagent (2 mM; in 400 mM acetate buﬀer, pH 5.4,

containing 10% dimethylformamide and 100 μM sodium iodide) with

the extent of TMB oxidation due to HOCl-mediated oxidation of

taurine into taurine chloramine quantiﬁed by measuring the

absorbance at 650 nm.

RESULTS

■

Synthesis of Polyamine-Conjugated Nitroxides. Piper-

idine nitroxides were conjugated to cationic polyamines to

enhance their bioavailability at endothelial and subendothelial

sites where MPO resides. TEMPO derivatives 1−6 were

synthesized as per the modular synthetic approach in Scheme

Measurement of Tyrosine Nitration by Isolated MPO. MPO-

catalyzed conversion of NO₂⁻into NO and the resultant nitration of

•

. Diamines were mono-Boc protected by addition of di-tert-

protein tyrosines was measured as the formation of 3-nitrotyrosine on

BSA by Western blotting using a rabbit polyclonal anti-3-nitrotyrosine

antibody. Reactions contained MPO (50 nM), BSA (100 μg/mL),

Scheme 2. Synthesis of Polyamine-Conjugated Nitroxides

−

and NO₂(100 μM), in the absence and presence of nitroxides (0−5

μM) in phosphate buﬀer, and were initiated by the addition of H O

(

50 μM), followed by incubation at 37 °C for 30 min. The reaction

was stopped by the addition of sample loading buﬀer containing SDS

for subsequent Western blot analysis.

Measurement of NO Consumption by MPO. The NO donor

NOC-9 (t₁_/₂2.7 min, 22 °C) was added to human plasma diluted into

.1 M phosphate buﬀer (ratio 1:5; pH 7.4, ∼22 °C, air saturated) with

rapid mixing to achieve a concentration of ∼500 nM NO (ﬁnal NOC-

concentration ∼2 μM), at which time MPO-catalyzed reactions

were initiated by the sequential addition of H O (10 μM) and MPO

(15 nM) at 3.5 and 4 min, respectively, after the addition of NOC-9

in the absence and presence nitroxides (50 μM). Changes in NO

concentration were measured continuously with an NO-speciﬁc

electrode (ISO-NOP), interfaced to a one-channel free radical

analyzer (TBR1025) and LabScribe 3 software (World Precision

Instruments). The initial rate of NO consumption was measured by

performing a linear regression of the slope over a 5 s time period

beginning 5 s following the addition of MPO. An NO standard curve

was performed as per the manufacturer using sulfuric acid and known

−

concentrations of NO .

Measurement of H O Consumption by Endothelial-Localized

MPO (Amplex Red Assay). H O consumption by endothelial cells

(i) Diamine (10 equiv), Boc O (1 equiv), CHCl , 16 h, 0 °C to RT;

was measured by the Amplex Red assay (Life Technologies). For this,

conﬂuent endothelial cells in HBSS (containing 0.2% BSA) at 37 °C

were incubated in the absence or presence of MPO (20 nM) for 2 h

and washed to remove unincorporated MPO. Cells in HBSS were

then nontreated or treated with nitroxides (10 μM) and incubated for

(ii) CF COOCH CH (1 equiv), MeOH, 1 h, −78 °C; (iii) Boc O (4

equiv), MeOH, 20 h, 0 °C to RT; (iv) saturated aqueous NH ,

MeOH, 24 h, RT; (v) Ti(OiPr)

10 (2 equiv), 3 h, RT; (vii) NaBH

viii) TFA (10 equiv), DCM, 4 h, RT, then K CO , MeOH/DCM, 12

(1.2 equiv), 20 min, RT; (vi) 9a−9e,

CN (1 equiv), MeOH, 24 h, RT;

(

h, RT.

5 min prior to H O (30 μM) addition and removal of supernatant

2 2

(50 μL) at 5, 10, 15, 30, and 60 min, which was mixed in a black-

bottomed 96-well microplate (Thermo Fisher) with 50 μL of Amplex

Red/HRP solution prepared as per the manufacturer’s instructions.

The microplate was incubated in the absence of light for 30 min at

room temperature, and H O levels were quantiﬁed by measurement

butyl dicarbonate to yield compounds 9a−9e in yields of 71%

quantitative (quant.), while tri-Boc-spermine (10) was

synthesized in a one-pot reaction, as previously reported

(52% yield). Compounds 11a−11e and 12 were generated

of the ﬂuorescence intensity (excitation 545 nm (530−560 nm)/

emission 590 nm) with a FLUOstar Omega plate reader (BMG

Labtech). H O concentration was determined by using a standard

by reductive amination of 4-oxo-TEMPO with partially

protected polyamines (9a−9e, 10) as described (Scheme

2v−vii); 4-oxo-TEMPO was stirred with 1.2 equiv of titanium

isopropoxide and 1 equiv of mono-Boc protected diamines or

tri-Boc-spermine, followed by reduction with sodium cyano-

borohydride to yield 11a−11e and 12 in yields of 30−74%.

Final compounds 1−6 (Table 1, >95% purity) were obtained

after removal of Boc protecting groups with triﬂuoroacetic acid

and subsequent formation of the free amine by stirring in

methanol/DCM with potassium carbonate.

curve generated with known concentrations of H O (0−50 μM).

Measurement of Chlorination and Nitration Reactions Cata-

lyzed by Endothelial-Localized MPO. Conﬂuent endothelial cells in

HBSS (containing 0.2% BSA) at 37 °C were loaded with MPO (20

nM) for 2 h and then washed with HBSS to remove unincorporated

MPO. Cells were nontreated or treated with nitroxides (10 μM) for

5 min prior to the addition of H O (50 μM). To study MPO-

−

catalyzed nitration, NO₂(100 μM) was added 15 min prior to H O .

Endothelial Cell Uptake of Nitroxides. The cellular

accumulation of nitroxides was examined by incubating

conﬂuent cultures of aortic endothelial cells with nitroxides

followed by subsequent determination of nitroxide content in

cell lysates with electron paramagnetic resonance (EPR)

spectroscopy by measuring the signal amplitude of the low-

ﬁeld spectrum (Figure 1A). Figure 1B shows that 4AT and

novel nitroxides 1 and 3 exhibited the greatest extent of

endothelial cell uptake. Novel nitroxides 4, 5, and 6 were

internalized to a lesser degree, exhibiting ∼32, 42, and 36%,

respectively, of the uptake apparent for novel nitroxide 1 and

Cells were incubated at 37 °C for 60 min and then lysed in 1 × SDS-

loading buﬀer for Western blotting analysis of 3-nitrotyrosine or

HOCl-oxidized proteins.

Statistical Analyses. Data are the mean ± SEM of three or more

independent experiments. Statistical diﬀerences in column graphs

between two treatments were assessed with a Student t test and one-

way ANOVA with Bonferroni’s or Dunnett’s post hoc test for more

than two treatments and for curves with two-way ANOVA. Statistical

diﬀerences in IC₅₀values were assessed by using a one-way ANOVA

with the Dunnett’s post hoc test. Analyses were performed with Prism

software, and a P-value of <0.05 was considered signiﬁcant.

684

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

Article

Table 1. Compounds Studied in This Work

4AT. Additionally, 4HT, 4CT, or MT showed comparatively

poor uptake exhibiting only ∼3, 7, and 14%, respectively, of the

uptake of 1 and 4AT. Overall, the eﬃciency of nitroxide

endothelial cell uptake was 4AT = 1 = 3 > 2 > 4 = 5 = 6 > MT

Figure 1. Endothelial cell uptake of nitroxides. Conﬂuent aortic

endothelial cells were incubated with nitroxides (10 μM) for 20 min,

and the cell lysate content of nitroxides was measured by EPR. (A)

Representative low-ﬁeld EPR spectrum of nitroxides; the signal

amplitude was used for quantitation. (B) Nitroxide content in

endothelial cells. Data are the mean ± SEM of three independent

experiments performed in triplicate for all conditions. ∗∗∗, P < 0.001;

ns, nonsigniﬁcant; compared to 4AT. (C) Relative contents of 3 and 6

in endothelial cells incubated at either 37 or 4 °C. Data are expressed

as a percent of nitroxide uptake at 37 °C and represent the mean ±

SEM of three independent experiments for all conditions. ∗∗, P <

4CT = 4HT.

We further conﬁrmed that the majority of nitroxide cellular

uptake required an energy-dependent process as endothelial

uptake of novel nitroxides 3 and 6 was inhibited by 70−80% in

endothelial cells cultured at 4 °C versus cells cultured at 37 °C

(Figure 1C). The role of endothelial-expressed HSPGs in

facilitating endothelial uptake of positively charged nitroxides

was next assessed by performing a competitive binding study

testing the ability of the heparan sulfate analogue, heparin, to

limit cellular nitroxide uptake. Heparin, at an ∼10-fold molar

excess, inhibited the uptake of 4AT, 3, or 6 by ∼60% (Figure

.01, relative to treatment of cells at 37 °C. (D) Prior addition of

heparin (1250 μg/mL or ∼100 μM; assuming an MW for heparin of

3 kDa) inhibits nitroxide uptake by endothelial cells. Data are

D), indicating a signiﬁcant role for HSPGs in the endothelial

expressed as a percent of nitroxide uptake in the absence of heparin

and represent the mean ± SEM of three independent experiments for

all conditions. ∗∗∗, P < 0.001, relative to treatment of cells with

nitroxides in the absence of heparin; ctrl, maximal nitroxide uptake in

the absence of heparin (i.e., 100%).

uptake of these nitroxides.

Inhibition of MPO Chlorination Activity by Nitro-

xides. We next tested the ability of nitroxides to inhibit the

chlorination activity of isolated MPO. Initially, an H O

electrode was used to assess nitroxide inhibition of isolated

MPO-catalyzed H O consumption in the presence of

,27

space.

Figure 3A shows that H O consumption was

2 2

−

physiological Cl levels; inhibition of H O consumption in

markedly enhanced in MPO-containing endothelial cells.

While nitroxides did not inhibit basal H O consumption by

control endothelial cells, nitroxides 1, 3, 4AT, and 4HT

signiﬁcantly inhibited the MPO-accelerated H O loss in

endothelial cells by 50−60% (Figure 3). In contrast, 2, 4, 5, 6,

4CT, and MT did not signiﬁcantly aﬀect H O consumption in

the presence of Cl⁻(and absence of other competing

substrates) is a direct measure of inhibition of MPO’s catalytic

activity and an indirect index of inhibition of MPO’s

chlorination activity (Scheme 1B, reactions 1 and 2).

Structure−function activity was observed regarding the eﬃcacy

of commercial nitroxides to inhibit MPO-catalyzed H O

MPO-containing endothelial cells (Figure 3B).

consumption with the order of eﬃciency 4AT > 4HT > MT

4CT (Figure 2A). For novel nitroxides, while 1−5 eﬀectively

inhibited MPO-catalyzed H O consumption, 6 showed a

The eﬀect of nitroxides on HOCl production catalyzed by

isolated or endothelial-localized MPO was next assessed. We

quantiﬁed the individual IC₅₀values for inhibition of MPO

chlorination for both the commercial and novel nitroxides

using the sensitive iodide-catalyzed TMB assay (Figure 4). We

observed structure−function activity for nitroxide-mediated

inhibition of HOCl production catalyzed by isolated MPO,

with 4AT being the most eﬃcient, 4HT being intermediate,

and MT or 4CT exhibiting reduced or no inhibition,

lower inhibitory capacity (Figure 2B).

To study the eﬃcacy of nitroxides to inhibit H O

catabolism by endothelial-localized MPO, we employed the

Amplex Red assay to detect H O and performed experiments

in endothelial cells containing transcytosed MPO, which

concentrates within endothelial cells and the subendothelial

685

Chem. Res. Toxicol. 2021, 34, 1681−1692

independent measurements.

Article

Figure 2. Nitroxides inhibit H O consumption by isolated MPO.

Eﬀect of (A) commercial or (B) novel nitroxides (1 μM) on H O

consumption by puriﬁed MPO. Reactions were performed in 0.1 M

−

phosphate buﬀer (pH 7.4) and contained H O (25 μM), Cl (100

mM), and the HOCl-scavenger methionine (300 μM). Reactions

were initiated by the addition of MPO (20 nM) as indicated. The

H O -consumption traces shown are representative of three

Figure 4. Nitroxide-mediated inhibition of MPO’s chlorination

activity. Reactions were performed at room temperature in phosphate

−

buﬀered saline (pH 7.4) containing 140 mM Cl , MPO (10 nM), and

taurine (5 mM) in the absence or presence of nitroxides (0.1−10

μM), initiated by the addition of H O (50 μM) and terminated after

−

min by the addition of catalase (20 μg mL ). MPO-catalyzed

HOCl production (chlorination activity) was then assessed by the

sensitive iodide-catalyzed TMB assay. (A) Dose−response curves

showing the inhibition of MPO-catalyzed HOCl production by 4HT,

CT, and novel nitroxides 1 and 3. Results are expressed as the

percent chlorination activity of MPO in the absence of nitroxides

Figure 3. Inhibition of H O consumption by endothelial-localized

(

^c^oⁿ^t^r^o^l⁾^,^w^hⁱ^c^h^w^a^s^a^s^sⁱ^gⁿ^e^d^a^v^a^l^u^e^o^f¹⁰⁰^%^.⁽^B⁾^T^a^b^u^l^a^t^e^d^I^C50

MPO. Conﬂuent endothelial cells were incubated in the absence or

presence of MPO (20 nM) for 2 h and the cells washed to remove

unbound MPO. Endothelial cells were then incubated in the absence

or presence of the relevant nitroxides (10 μM) before H O (30 μM)

values of nitroxides (i.e., concentration necessary to inhibit MPO-

catalyzed HOCl production by 50%), which were determined from

the dose−response curves using nonlinear regression. ∗, P < 0.05,

relative to the IC of 4HT using one-way ANOVA with post hoc

was added, and the time-dependent loss of H O in the media was

Dunnett’s test. Data represents the mean ± SEM, n = 4−12 (n = 12

for 4HT, n = 4 for all other nitroxides).

measured at the indicated times using the Amplex Red assay. (A)

Inhibition of MPO-dependent H O consumption in endothelial cells

by 4AT or novel nitroxide 3 over a 30 min time period. (B) Inhibition

of H O consumption by endothelial-localized MPO by nitroxides at

min post H O addition. Data are expressed as a percent of maximal

2 2

apparent for nitroxide-mediated inhibition of MPO-catalyzed

H O consumption by MPO-containing endothelial cells in the

H O consumption (Figure 2).

absence of nitroxides (MPO ctrl) at 5 min (100%) and represent the

mean ± SEM of three independent experiments. ∗∗∗, P < 0.001; ns,

nonsigniﬁcant; compared to MPO ctrl.

Positively charged and polyamine-based nitroxides bind with

high aﬃnity to negatively charged heparan sulfates, which are

abundant in the subendothelial space where MPO is

sequestered. As heparan sulfates may interfere with the

access of positively charged and novel nitroxides into the MPO

active site, the eﬀect of a molar excess of heparin (a heparan

sulfate analogue) on the inhibitory capacity of these nitroxides

toward MPO was tested using the TNB assay (Figure 5).

While heparin addition did not aﬀect the ability of novel

nitroxide 1 to inhibit MPO-catalyzed HOCl production, it

abolished the MPO inhibitory capacity of 4 and 6 and partially

respectively (representative curves are shown in Figure 4A). All

novel nitroxides 1−6 eﬀectively inhibited MPO-catalyzed

HOCl production, exhibiting signiﬁcantly greater or similar

eﬃcacy as 4HT (Figure 4), with the most eﬃcacious being 3

(

putrescine-TEMPO) and 2 (propanediamine-TEMPO). The

order of eﬃcacy of nitroxide inhibition of HOCl production

catalyzed by isolated MPO (Figure 4) generally mirrored that

686

Chem. Res. Toxicol. 2021, 34, 1681−1692

Article

Figure 5. Eﬀect of heparin on nitroxide-mediated inhibition of MPO’s

chlorination activity. Reactions contained MPO (100 nM), taurine

(

20 mM), and Cl⁻(100 mM) in the absence (ctrl) or presence of

nitroxides (10 μM) and the absence (black bars) or presence (gray

bars) of heparin (1 mg/mL; 77 μM based on average MW of 13.000

kDa) and were initiated by the addition of H O (50 μM). Following

min incubation reactions were terminated by the addition of

−

catalase (50 μg mL ), and MPO-catalyzed HOCl production was

assessed via the TNB assay. Results are expressed as a percent of

taurine chloramine production formed by MPO in the absence of

nitroxides (ctrl) that was assigned a value of 100%. The results

represent the mean ± SEM of three independent experiments. ∗∗∗, P

Figure 6. Eﬀect of nitroxides on HOCl production by endothelial-

localized MPO. Conﬂuent endothelial cells were incubated in the

absence or presence of MPO (20 nM) for 2 h and cells washed to

remove unbound MPO. Endothelial cells were then incubated in the

absence or presence of nitroxides (10 μM) for 15 min before H O

0.001; ∗, P < 0.05; ns, nonsigniﬁcant; compared to values obtained

in the absence of heparin for each treatment.

(

50 μM) was added and endothelial cells were incubated for 1 h. Cell

attenuated the inhibitory activity of 2, 3, 5, and 4AT by ∼2.4-,

lysates were then assessed for HOCl production using a monoclonal

anti-HOCl oxidized protein antibody (clone 2D10G9) and Western

blotting. (A) Representative Western blot of three independent blots,

which was cropped to show the high molecular weight protein bands

detected above 460 kDa. (B) Densitometric analyses of three

independent experiments/Western blots were performed with ImageJ

software. Data are expressed as a percent of the intensity of the

HOCl-oxidized protein bands formed in MPO-containing cells

exposed to H O in the absence of nitroxides (MPO). Data represent

∼

2.3-, ∼1.8-, and ∼3.3-fold, respectively (Figure 5).

27 7

Prior works by us and others have established that

incubation of conﬂuent aortic endothelial cells with MPO

results in the transcytosis and accumulation of the enzyme

primarily in the subendothelial space, where it catalyzes the

chlorination and nitration of extracellular matrix proteins,

including ﬁbronectin. To study the eﬃcacy of nitroxides to

inhibit HOCl production by endothelial-localized MPO, the

formation of HOCl-oxidized proteins in MPO-containing

endothelial cells was measured via Western blotting using a

well-characterized anti-HOCl-oxidized protein antibody

the mean ± SEM of three independent experiments. ∗∗, P < 0.01; ∗,

P < 0.05; ns, nonsigniﬁcant; compared to 4AT.

(

2D10G9; it does not cross-react with epitopes generated by

Prior work shows that endothelial-localized MPO catalyzes

3-nitrotyrosine formation in the subendothelial matrix of

diseased arteries in vivo and of endothelial cells in vitro.

oxidative reactions involving nitrating species, transition

metals, or lipid peroxidation reactions ). Consistent with

our prior data, exposure of MPO-containing endothelial cells

to low micromolar levels of H O (50 μM) aﬀorded the

detection of high molecular weight HOCl-oxidized proteins,

derived from cross-linked aggregates of oxidized extracellular

matrix proteins formed within the subendothelium where

transcytosed MPO accumulates (Figure 6). Nitroxides

diﬀerentially inhibited the formation of HOCl-oxidized protein

by endothelial-localized MPO; novel nitroxides 1, 2, 3, and 5,

as well as 4AT and 4HT, all signiﬁcantly inhibited HOCl

production, while 4, 6, 4CT, and MT showed reduced

inhibition (Figure 6).

7,12,27

We therefore next tested the ability of nitroxides to inhibit 3-

nitrotyrosine formation catalyzed by endothelial-localized

MPO using Western blot and a 3-nitrotyrosine antibody.

Treatment of MPO-containing endothelial cells with H O in

−

the presence of NO₂resulted in the formation of high

molecular weight proteins immunoreactive for 3-nitrotyrosine

(Figure 8). Marked diﬀerences were noted regarding the ability

of nitroxides to inhibit 3-nitrotyrosine formation in MPO-

containing endothelial cells; while 4AT did not inhibit 3-

nitrotyrosine levels, 4HT, 4CT and MT inhibited 3-nitro-

tyrosine formation by ∼70, 88, and 45%, respectively (Figure

8). For the novel nitroxides, 1 and 3 inhibited protein nitration

by ≥80%, 4 and 6 inhibited protein nitration by 60−75%, and

2 inhibited protein nitration by ∼30%. In contrast, 5 did not

signiﬁcantly inhibit protein nitration catalyzed by endothelial-

localized MPO.

Inhibition of MPO-Catalyzed Protein Nitration by

Nitroxides. A signiﬁcant reaction catalyzed by MPO in vivo is

the conversion of NO₂⁻into NO that reacts with protein

•

tyrosines to form 3-nitrotyrosine, a hallmark of oxidative or

nitrosative tissue damage during inﬂammation and cardiovas-

cular disease.

inhibition by nitroxides of tyrosine nitration of bovine serum

albumin (BSA) catalyzed by isolated MPO and determined by

Western blotting with an anti-3-nitrotyrosine antibody. All

nitroxides eﬀectively inhibited BSA tyrosine nitration catalyzed

by MPO, with the novel nitroxides and 4AT exhibiting

signiﬁcantly greater eﬃcacy compared to 4HT and 4CT

,31

We initially examined the dose-dependent

Inhibition of the NO Oxidase Activity of MPO by

Nitroxides. MPO’s NO oxidase activity is a key oxidative

mechanism impairing NO bioactivity in inﬂammation, e.g.,

endotoxin-challenged rodents, human coronary artery

32,33

disease patients,

and myocardial infarction patients. We

tested the ability of novel nitroxides to inhibit MPO-catalyzed

NO consumption in diluted human plasma in the presence of

physiological steady-state levels of NO (∼500 nM) delivered

(Figure 7).

687

Chem. Res. Toxicol. 2021, 34, 1681−1692

Article

Figure 8. Nitroxides inhibit protein nitration by endothelial-localized

MPO. Conﬂuent endothelial cells were incubated in the absence or

presence of MPO (20 nM) for 2 h and cells washed to remove

unbound MPO. Cells were treated with nitroxides (10 μM) and

−

NO₂(100 μM) for 15 min prior to H O (50 μM) addition and

incubation for 1 h. 3-Nitrotyrosine levels in cell lysates were assessed

by Western blotting using an anti-3-nitrotyrosine antibody. (A)

Representative Western blot of three independent blots, which has

been cropped to show the high molecular weight 3-nitrotyrosine

bands detected above 460 kDa. (B) Densitometric analysis of Western

blots was performed with ImageJ software. Data are expressed as a

percent of the intensity of the 3-nitrotyrosine bands formed in MPO-

−

containing endothelial cells exposed to H O in the presence of NO

and absence of nitroxides (ctrl). Data represent the mean ± SEM of

three independent experiments. ∗∗∗, P < 0.001; ∗∗, P < 0.01; ∗, P <

.05; ns, nonsigniﬁcant; compared to MPO-containing endothelial

−

cells exposed to H O in the presence of NO and absence of

nitroxides (ctrl).

by the NO donor NOC-9. Novel nitroxides 1−6 and 4AT all

signiﬁcantly inhibited the initial rates of MPO-catalyzed NO

consumption in human plasma by 50−60% (Figure 9).

DISCUSSION

■

In this study we characterized several piperidine nitroxides

including a novel series of polyamine-conjugated piperidine

nitroxides (Table 1) for their propensity to (i) accumulate into

endothelial cells and (ii) inhibit several oxidative reactions

catalyzed by both isolated and endothelial-localized MPO that

are mechanistically linked as causes of endothelial dysfunction

during cardiovascular disease including MPO-catalyzed H O

Figure 7. Eﬀect of nitroxides on MPO-catalyzed BSA nitration.

−

Reactions contained MPO (50 nM), BSA (100 μg/mL) and NO₂

(

100 μM) in the absence (ctrl) or presence of nitroxides (0.1−5 μM)

and were initiated by H O addition (50 μM). After incubation at 37

°C for 30 min the reactions were terminated and the level of 3-

−

•

metabolism, conversion of Cl into HOCl or NO into NO ,

nitrotyrosine on BSA determined by Western blot. (A) The Western

blot shown is representative of three to four independent blots. 3-

Nitrotyrosine levels were quantiﬁed by densitometric analysis of the

and MPO’s NO oxidase activity. We provide the ﬁrst evidence

that the presence of cationic amino (−NH ) groups para to the

8 kDa band by ImageJ software. No immunoreactivity was present

nitroxyl radical signiﬁcantly enhance cellular internalization of

piperidine nitroxides in an HSPG-dependent manner and that

nitroxides eﬀectively inhibit MPO-catalyzed oxidative and

nitrosative damage in the complex biological environment of

endothelial cells. This work also reveals important structure−

function information useful for the further development of

polyamine-conjugated nitroxides. Collectively, this work

provides evidence supporting that one mechanism by which

piperidine nitroxides preserve endothelial function in vivo is

MPO inhibition. Moreover, our studies provide the foundation

to support further development of endothelial-targeted MPO

−

for BSA + H O + NO in the absence of MPO (data not shown).

(B) Representative dose−response curves showing the inhibition of

MPO-catalyzed BSA nitration by 4HT and 3. Results are expressed as

percent nitration by MPO in the absence of nitroxides (ctrl), which

were assigned a value of 100%. (C) Tabulated IC₅₀values (nitroxide

concentration required to inhibit MPO-catalyzed BSA nitration by

0%) of nitroxides. ∗, P < 0.05 relative to IC of 4HT using one-way

ANOVA with a post hoc Dunnett’s test. Data represents the mean ±

SEM of three to four independent experiments for all conditions. IC₅₀

values were determined from dose−response curves using nonlinear

regression.

688

Chem. Res. Toxicol. 2021, 34, 1681−1692

Article

internalization of these nitroxides by ∼60% supports a

signiﬁcant role for HSPGs. For polyamine-conjugated nitro-

xides this likely reﬂects the reported role for HSPGs in

polyamine cellular uptake. The remaining 40% of endothelial

nitroxide uptake that was refractory to heparin may reﬂect

HSPG-independent routes, e.g., polyamine import via cationic

amino acid transporters. Although HSPGs are important for

polyamine-conjugated nitroxide and MPO endothelial uptake,

the extent to which nitroxides and MPO follow analogous or

diﬀerent intracellular transport routes is unknown and requires

further investigation.

With regard to inhibition of chlorination activity by isolated

MPO, our data showed that positively charged (4AT, 1−6) or

neutral (4HT) nitroxides eﬃciently inhibited MPO-catalyzed

H O metabolism and HOCl production, while negatively

charged (4CT) nitroxides were less eﬃcient, ﬁndings that were

consistent with studies by Rees et al. Previous studies

indicate that MPO poorly metabolizes negatively charged

substrates due to the presence of a carboxy group in the

peroxidase substrate channel (most likely from the glutamine

Figure 9. Nitroxides inhibit MPO-catalyzed NO oxidation in human

plasma. Eﬀect of novel nitroxides 1−6 or 4AT (50 μM) on the initial

rate of NO consumption catalyzed by puriﬁed MPO (15 nM) in

diluted human plasma. (A) Representative traces of NO consumption

for 4AT and 3. The NO donor NOC-9 (∼500 nM) was added to

plasma (diluted 1:5 in 0.1 M phosphate buﬀer, pH 7.4) followed by

addition of H O (10 μM) and MPO (15 nM) at 3.5 and 4 min post

addition of NOC-9. (B) Initial rates of MPO-catalyzed NO

consumption in the absence (ctrl) or presence of nitroxides. Data

are the mean ± SEM of at least three independent experiments for all

conditions. ∗∗∗, P < 0.001 relative to the absence of nitroxides (ctrl).

242 residue). This may explain the poor inhibitory activity of

4CT relative to 4HT and 4AT. On the other hand, the eﬃcacy

of nitroxides to inhibit oxidative reactions catalyzed by

endothelial-localized MPO appears to be inﬂuenced by the

interaction of a number of factors including nitroxide

bioavailability (i.e., endothelial cell uptake) and capacity to

inhibit MPO enzyme activity, as well as competitive binding to

other cellular components such as HSPGs. This was most

evident with the novel nitroxides. Despite exhibiting similar

inhibition of isolated MPO chlorination activity, they showed

structure−function activity for the inhibition of endothelial-

localized MPO. Novel nitroxides 1 (ethylenediamine-

TEMPO) and 3 (putrescine-TEMPO) (i) avidly accumulated

into cultured endothelial cells (Figure 1), (ii) eﬃciently

inhibitors useful for combating endothelial dysfunction during

vascular inﬂammation.

Our study aimed to enhance the bioavailability of nitroxides

at endothelial sites via conjugation to polyamines. We found

that nitroxides exhibited diﬀerences in endothelial cell uptake,

which related to the lipo/hydrophilicity and/or charge of the

inhibited H O consumption and HOCl production by

2 2

endothelial-localized MPO (Figures 3−6), and (iii) largely

maintained their inhibitory capacity toward MPO in the

presence of exogenous heparan sulfates (Figure 5). In contrast,

novel nitroxides 4 (cadaverine-TEMPO) and 6 (spermine-

compounds. As such, hydrophilic nitroxides with a net

positive charge (1−6, 4AT) were more eﬃciently internalized

in comparison to either hydrophilic neutral nitroxides (4HT),

negatively charged nitroxides (4CT), or lipophilic positively

charged nitroxides (MT). Tempol (4HT) is the most widely

studied nitroxide, and commonly, high doses of 4HT are

TEMPO) poorly inhibited H O consumption and HOCl

2 2

production by endothelial-localized MPO (Figures 3 and 6),

which diverged from their capacity to eﬀectively inhibit

isolated MPO (Figures 2 and 4). This could be accounted

for by the fact that endothelial cell uptake of 4 and 6 was lower

than that of 1 or 3 (Figure 1B). Moreover, exogenous addition

of heparan sulfates abrogated the inhibitory activity of 4 or 6

toward isolated MPO (Figure 5). This indicates that the

HSPG-binding aﬃnity of certain polyamine-conjugated nitro-

xides is a key determinant of their ability to interact with and

inhibit MPO within endothelial cells and subendothelial matrix

that is enriched with HSPGs. Consistent with this, putrescine

21,35

necessary to provide in vivo protection in animal models,

which is attributed to its low cellular uptake and poor in vivo

1,35,36

bioavailability.

Our EPR data are in line with these

reports, as endothelial cell uptake of 4HT was <5% of

positively charged nitroxides 1, 3, or 4AT (Figure 1B). We also

examined endothelial uptake of the mitochondrial-targeted

nitroxide, MT, which Dikalova et al. report is internalized ∼5-

fold > 4HT. Our data similarly show that MT is internalized

into endothelial cells ∼4-fold > 4HT. We also show that the

novel polyamine-conjugated nitroxides 1−6 were internalized

exhibits a lower heparin binding aﬃnity than spermine and

exogenous heparin abolished the ability of 6 (spermine-

TEMPO) to inhibit MPO’s chlorination activity, but to a lesser

extent for 3 (putrescine-TEMPO) (Figure 5). Taken together,

our data show for the ﬁrst time that low micromolar

concentrations of certain novel polyamine-conjugated (1 and

3) and commercial nitroxides (4AT and 4HT) can eﬀectively

access and inhibit HOCl production catalyzed by endothelial-

localized MPO. We further reveal that the structure−function

relationships reﬂect diﬀerences in endothelial internalization,

access to MPO’s active site (i.e., eﬃcacy of inhibition of

isolated MPO), and HSPG-binding aﬃnity.

−7-fold > MT. Overall, our studies show that conjugation of

TEMPO nitroxides to polyamines, and the presence of a

positive charge in the para position to the nitroxyl radical,

increases endothelial internalization above that of the

commonly studied 4HT.

As HSPGs are important for endothelial binding and

,24

transcytosis of MPO and polyamine cellular uptake,

examined the inﬂuence of HSPGs on the internalization of

AT and certain polyamine-conjugated nitroxides. Our ﬁnding

that the heparan sulfate analogue heparin inhibited endothelial

689

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

The Supporting Information is available free of charge at

Article

While nitroxides were recently identiﬁed as a novel class of

competitive, “compound II trapping” MPO inhibitors,

reservations exist regarding their eﬃcacy as MPO inhibitors

in complex biological environments replete with MPO

compound II reductants capable of reforming native MPO

contrasted with a lead agent, 4AT, that strongly inhibited

MPO-catalyzed HOCl production but not protein nitration in

endothelial cells. Also, the enhanced cellular internalization of

1 and 3 may allow them to be used at lower concentrations in

vivo than the commonly used 4HT to treat endothelial

dysfunction during inﬂammatory vascular disease. Finally,

unlike other polyamine-conjugated nitroxides (4 and 6),

heparan sulfate binding to 1 and 3 exerted reduced eﬀects

on their MPO inhibitory capacity, meaning they retain their

eﬃcacy of action in the subendothelium that is rich in HSPGs

and where MPO is known to accumulate. Our ﬁndings

therefore advocate further testing of the protective actions of

ethylenediamine-TEMPO (1) and putrescine-TEMPO (3) in

preclinical models of cardiovascular disease where MPO plays

a pathogenic role.

(Scheme 1 B, reaction 5), which is available to re-enter the

halogenation cycle (Scheme 1 B, reactions 1 and 2). Despite

this, our data establish that low micromolar concentrations of

nitroxides eﬃciently inhibited MPO within cultured endothe-

lial cells in which endogenous MPO compound II substrates

•

−

(

e.g., ascorbate, O₂) are available. The eﬃcacy of nitroxides

within the complex endothelial environment may reﬂect the

unique catalytic nature of nitroxides as reversible MPO

inhibitors and their SOD mimetic activity and hence ability

•

−

to scavenge O₂, which can otherwise reduce MPO

compound II to limit nitroxides’ inhibitory action toward

MPO. This advocates that nitroxides have the capacity to

eﬀectively inhibit the chlorination activity by endothelial-

localized MPO in vivo.

ASSOCIATED CONTENT

Supporting Information

https://pubs.acs.org/doi/10.1021/acs.chemrestox.1c00094.

■

sı

MPO is a versatile enzyme that also impacts endothelial

function through its nitration reactions that proceed via the

,12

Complete syntheses and characterization of all com-

pounds and compound purity statements (PDF)

enzyme’s peroxidase cycles.

Our studies showed that

nitroxides’ capacity to inhibit MPO-mediated protein nitration

was not related to their ability to inhibit MPO activity and was

•

AUTHOR INFORMATION

Corresponding Authors

instead due to direct NO scavenging, a conclusion consistent

■

6,41

with Vaz et al.

With respect to inhibition of protein

nitration by endothelial-localized MPO, we found conﬂicting

results in particular for 4AT, which did not inhibit protein

nitration by endothelial-localized MPO. While the reasons for

this are unclear, the eﬃcacy of nitroxides to prevent protein

Jonathan C. Morris − School of Chemistry, University of New

South Wales, Sydney, New South Wales 2052, Australia;

orcid.org/0000-0002-5109-9069 ;

Email: jonathan.morris@unsw.edu.au

nitration catalyzed by endothelial-localized MPO likely relates

Shane R. Thomas − School of Medical Sciences, University of

New South Wales, Sydney, New South Wales 2052,

Australia; orcid.org/0000-0003-1080-8954;

Email: shane.thomas@unsw.edu.au

•

to the rate of reaction of NO with the relevant nitroxide

versus alternative biomolecule targets present in endothelial

cells, as well as the rate of reaction of nitroxides with other

locally produced MPO-derived substrate radicals (e.g., urate

radicals).

Authors

Sophie Maiocchi − School of Medical Sciences and School of

Chemistry, University of New South Wales, Sydney, New

South Wales 2052, Australia; Center for Nanotechnology in

Drug Delivery, University of North Carolina at Chapel Hill,

Chapel Hill, North Carolina 27599, United States

Jacqueline Ku − School of Medical Sciences, University of New

South Wales, Sydney, New South Wales 2052, Australia

Tom Hawtrey − School of Chemistry, University of New South

Wales, Sydney, New South Wales 2052, Australia

Irene De Silvestro − School of Chemistry, University of New

South Wales, Sydney, New South Wales 2052, Australia

Ernst Malle − Gottfried Schatz Research Center, Molecular

Biology & Biochemistry, Medical University of Graz, 8036

Graz, Austria

Finally, MPO NO oxidase activity is regarded as an

important oxidative reaction impairing endothelial function.

Thus, we examined whether the novel nitroxides inhibited this

important reaction. All the nitroxides tested inhibited the NO

oxidase activity of MPO in human plasma with similar

eﬃcacies. The lack of structure−function activity shown by

the various nitroxides supports the conclusion of our recent

work indicating that inhibition of MPO’s NO oxidase activity

by nitroxides in plasma occurs independent of their eﬀects on

the turnover of MPO’s peroxidase cycle and instead reﬂects

nitroxide scavenging of MPO-derived substrate radicals that

otherwise consume NO.

CONCLUSIONS

■

Martin Rees − School of Medical Sciences, University of New

South Wales, Sydney, New South Wales 2052, Australia

Piperidine nitroxides possess multiple antioxidant functions. In

this study, we show for the ﬁrst time that functionalized

nitroxides can access endothelial-sequestered MPO and

eﬃciently inhibit local oxidative reactions (HOCl production,

protein nitration) causally linked to endothelial dysfunction in

vascular disease. Nitroxides also inhibit MPO’s NO oxidase

activity in human plasma. Of the novel nitroxides, ethylenedi-

amine-TEMPO (1) and putrescine-TEMPO (3) emerged as

promising MPO-targeted nitroxides as conjugation of these

polyamines to TEMPO uniquely conferred both a high

endothelial cell uptake and eﬃcient inhibition HOCl

production and protein nitration catalyzed by endothelial-

localized MPO, as well as NO oxidation in plasma. This

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.chemrestox.1c00094

Author Contributions

The manuscript was written through contributions of all

authors. All authors have given approval to the ﬁnal version of

the manuscript.

Funding

This work was supported by a Diabetes Australia Research

Trust Grant (S.R.T.), National Health & Medical Research

Council Project Grants APP1058508 and APP1125392

690

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

S.R.T.), and a UNSW Goldstar award (M.R., S.R.T.). S.M.

received support from an Australia Postgraduate Award (APA).

Notes

Involvement of Myeloperoxidase-Mediated Oxidant in Plaque Erosion

Article

(

and Thrombogenesis. Arterioscler., Thromb., Vasc. Biol. 24, 1309−

(

314.

12) Baldus, S., Eiserich, J. P., Brennan, M. L., Jackson, R. M.,

The authors declare no competing ﬁnancial interest.

Alexander, C. B., and Freeman, B. A. (2002) Spatial mapping of

pulmonary and vascular nitrotyrosine reveals the pivotal role of

myeloperoxidase as a catalyst for nitrotyrosine nitration in

inflammatory diseases. Free Radical Biol. Med. 33, 1010−1019.

(13) Eiserich, J.P. B. S., Brennan, M. L., Ma, W., Zhang, C., Tousson,

A., Castro, L., Lusis, A. J., Nauseef, W. M., White, C. R., and Freeman,

B. A. (2002) Myeloperoxidase, a Leukocyte-derived vascular NO

oxidase. Science 296, 2391−2394.

ABBREVIATIONS

AT, 4-amino-TEMPO; 4CT, 4-carboxy-TEMPO; 4HT, 4-

hydroxy-TEMPO; 3-NO Tyr, 3-nitrotyrosine; Boc, tert-butox-

ycarbonyl; BSA, bovine serum albumin; ctrl, control; Cl ,

■

−

chloride; DCM, dichloromethane; EPR, electron paramagnetic

resonance; HOCl, hypochlorous acid; HSPGs, heparan sulfate

proteoglycans; H O , hydrogen peroxide; MPO, myeloperox-

(

14) Soule, B. P., Hyodo, F., Matsumoto, K. i., Simone, N. L., Cook,

•

J. A., Krishna, M. C., and Mitchell, J. B. (2007) The chemistry and

biology of nitroxide compounds. Free Radical Biol. Med. 42, 1632−

1650.

idase; MT, mito-TEMPO; NO , nitrogen dioxide radical;

NO , nitrite; NO, nitric oxide; O , superoxide; SOD,

superoxide dismutase; TLC, thin layer chromatography; Tyr ,

tyrosyl radical; TMB, tetramethylbenzidine; TNB, 2-nitro-5-

thiobenzoic acid; TLC, thin layer chromatography

−

•−

•

(

15) Rees, M. D., Bottle, S. E., Fairfull-Smith, K. E., Malle, E.,

Whitelock, J. M., and Davies, M. J. (2009) Inhibition of

myeloperoxidase-mediated hypochlorous acid production by nitro-

xides. Biochem. J. 421, 79−86.

(16) Vaz, S. M., and Augusto, O. (2008) Inhibition of

Myeloperoxidase-mediated protein nitration by tempol: Kinetics,

mechanism, and implications. Proc. Natl. Acad. Sci. U. S. A. 105,

191−8196.

17) Maiocchi, S. L., Morris, J. C., Rees, M. D., and Thomas, S. R.

2017) Regulation of the nitric oxide oxidase activity of myeloper-

oxidase by pharmacological agents. Biochem. Pharmacol. 135, 90−115.

18) Malle, E., Hazell, L., Stocker, R., Sattler, W., Esterbauer, H., and

REFERENCES

■

(

1) Nussbaum, C., Klinke, A., Adam, M., Baldus, S., and Sperandio,

M. (2013) Myeloperoxidase: A leukocyte-derived protagonist of

inflammation and cardiovascular disease. Antioxid. Redox Signaling 18,

(

92−713.

2) Maiocchi, S. L., Ku, J., Thai, T., Chan, E., Rees, M. D., and

Thomas, S. R. (2021) Myeloperoxidase: A versatile mediator of

endothelial dysfunction and therapeutic target during cardiovascular

disease. Pharmacol. Ther. 221, 107711.

(

Waeg, G. (1995) Immunologic detection and measurement of

hypochlorite-modified LDL with specific monoclonal antibodies.

Arterioscler., Thromb., Vasc. Biol. 15, 982−989.

(3) Thomas, S. R., Witting, P. K., and Drummond, G. R. (2008)

Redox control of Endothelial function and dysfunction. Molecular

Mechanisms and Therapeutic opportunities. Antioxid. Redox Signaling

19) Queiroz, R. F., Vaz, S. M., and Augusto, O. (2011) Inhibition of

(

0, 1713−1766.

4) Baldus, S., Heitzer, T., Eiserich, J. P., Lau, D., Mollnau, H.,

Ortak, M., Petri, S., Goldmann, B., Duchstein, H. J., Berger, J.,

Helmchen, U., Freeman, B. A., Meinertz, T., and Munzel, T. (2004)

the chlorinating activity of myeloperoxidase by tempol: Revisiting the

kinetics and mechanisms. Biochem. J. 439, 423−431.

20) Rees, M. D., Maiocchi, S. L., Kettle, A. J., and Thomas, S. R.

(

2014) Mechanism and regulation of peroxidase-catalyzed nitric oxide

Myeloperoxidase enhances nitric oxide catabolism during myocardial

consumption in physiological fluids: Critical protective actions of

ascorbate and thiocyanate. Free Radical Biol. Med. 72 , 91 −103.

(21) Wilcox, C. S. (2010) Effects of tempol and redox-cycling

nitroxides in models of oxidative stress. Pharmacol. Ther. 126, 119−

145.

ischemia and reperfusion. Free Radical Biol. Med. 37, 902−911.

(5) Vita, J. A., Brennan, M. L., Gokce, N., Mann, S. A., Goormastic,

M., Shishehbor, M. H., Penn, M. S., Keaney, J. F., Jr., and Hazen, S. L.

2004) Serum myeloperoxidase levels independently predict

endothelial dysfunction in humans. Circulation 110, 1134−1139.

6) van der Zwan, L. P., Teerlink, T., Dekker, J. M., Henry, R. M. A.,

Stehouwer, C. D. A., Jakobs, C., Heine, R. J., and Scheffer, P. G.

2010) Plasma myeloperoxidase is inversely associated with

(22) Wengenack, T. M., Curran, G. L., Olson, E. E., and Poduslo, J.

F. (1997) Putrescine-modified catalase with preserved enzymatic

activity exhibits increased permeability at the blood-nerve and blood-

brain barriers. Brain Res. 767, 128−135.

endothelium-dependent vasodilation in elderly subjects with

abnormal glucose metabolism. Metab., Clin. Exp. 59, 1723−1729.

(23) Simoni, E., Bergamini, C., Fato, R., Tarozzi, A., Bains, S.,

(7) Baldus, S., Eiserich, J. P., Mani, A., Castro, L., Figueroa, M.,

Motterlini, R., Cavalli, A., Bolognesi, M. L., Minarini, A., Hrelia, P.,

Lenaz, G., Rosini, M., and Melchiorre, C. (2010) Polyamine

conjugation of curcumin analogues toward the discovery of

mitochondria-directed neuroprotective agents. J. Med. Chem. 53,

Chumley, P., Ma, W., Tousson, A., White, C. R., Bullard, D. C.,

Brennan, M. L., Lusis, A. J., Moore, K. P., and Freeman, B. A. (2001)

Endothelial transcytosis of myeloperoxidase confers specificity to

vascular ECM proteins as targets of tyrosine nitration. J. Clin. Invest.

Stocker, R. (1996) Presence of Hypochlorite-modified Proteins in

264−7268.

24) Belting, M., Persson, S., and Fransson, L. Å. (1999)

Proteoglycan involvement in polyamine uptake. Biochem. J. 338,

17−323.

25) Leonard, J., Lygo, B., and Procter, G. (1995) Advanced Practical

Organic Chemistry, 2nd ed., Blackie Academic & Professional.

26) Dypbukt, J. M., Bishop, C., Brooks, W. M., Thong, B., Eriksson,

(

08, 1759−1770.

8) Hazell, L. J., Arnold, L., Flowers, D., Waeg, G., Malle, E., and

(

Human Atherosclerotic Lesions. J. Clin. Invest. 97, 1535−1544.

(9) Malle, E., Waeg, G., Schreiber, R., Gröne, E. F., Sattler, W., and

Gröne, H.-J. (2000) Immunohistochemical evidence for the

myeloperoxidase/H2O2/halide system in human atherosclerotic

lesions Colocalization of myeloperoxidase and hypochlorite-modified

proteins. Eur. J. Biochem. 267, 4495−4503.

(

H., and Kettle, A. J. (2005) A sensitive and selective assay for

chloramine production by myeloperoxidase. Free Radical Biol. Med.

(

9, 1468−1477.

(10) Sugiyama, S., Okada, Y., Sukhova, G. K., Virmani, R., Heinecke,

27) Rees, M. D., Dang, L., Thai, T., Owen, D. M., Malle, E., and

J. W., and Libby, P. (2001) Macrophage Myeloperoxidase Regulation

by Granulocyte Macrophage Colony-Stimulating Factor in human

atherosclerosis and implications in Acute Coronary Syndromes. Am. J.

Pathol. 158, 879−891.

Thomas, S. R. (2012) Targeted subendothelial matrix oxidation by

myeloperoxidase triggers myosin II-dependent de-adhesion and alters

signaling in endothelial cells. Free Radical Biol. Med. 53 , 2344 −2356.

(28) Geall, A. J., and Blagbrough, I. S. (2000) Homologation of

polyamines in the rapid synthesis of lipospermine conjugates and

related lipoplexes. Tetrahedron 56, 2449−2460.

(11) Sugiyama, S., Kugiyama, K., Aikawa, M., Nakamura, S., Ogawa,

H., and Libby, P. (2004) Hypochlorous Acid, a Macrophage Product,

Induces Endothelial Apoptosis and Tissue Factor Expression.

691

Chem. Res. Toxicol. 2021, 34, 1681−1692

Chemical Research in Toxicology

(29) Siegenthaler, K. O., Schäfer, A., and Studer, A. (2007)

Article

Chemical surface modification via radical C-C bond-forming

reactions. J. Am. Chem. Soc. 129, 5826−5827.

(30) Rudolph, T. K., Rudolph, V., Witte, A., Klinke, A., Szoecs, K.,

Lau, D., Heitzer, T., Meinertz, T., and Baldus, S. (2010) Liberation of

vessel adherent myeloperoxidase by enoxaparin improves endothelial

function. Int. J. Cardiol. 140 , 42 −47.

(31) Radi, R. (2004) Nitric oxide, oxidants, and protein tyrosine

nitration. Proc. Natl. Acad. Sci. U. S. A. 101, 4003−4008.

(32) Rudolph, T. K., Wipper, S., Reiter, B., Rudolph, V., Coym, A.,

Detter, C., Lau, D., Klinke, A., Friedrichs, K., Rau, T., Pekarova, M.,

Russ, D., Knöll, K., Kolk, M., Schroeder, B., Wegscheider, K.,

Andresen, H., Schwedhelm, E., Boeger, R., Ehmke, H., and Baldus, S.

2012) Myeloperoxidase deficiency preserves vasomotor function in

humans. Eur. Heart J. 33, 1625−1634.

33) Baldus, S., Rudolph, V., Roiss, M., Ito, W. D., Rudolph, T. K.,

(

Eiserich, J. P., Sydow, K., Lau, D., Szocs, K., Klinke, A., Kubala, L.,

Berglund, L., Schrepfer, S., Deuse, T., Haddad, M., Risius, T., Klemm,

H., Reichenspurner, H. C., Meinertz, T., and Heitzer, T. (2006)

Heparins increase endothelial nitric oxide bioavailability by liberating

vessel-immobilized myeloperoxidase. Circulation 113, 1871−1878.

35) Wilcox, C. S., and Pearlman, A. (2008) Chemistry and

34) Hervé, F., Ghinea, N., and Scherrmann, J. M. (2008) CNS

delivery via adsorptive transcytosis. AAPS J. 10, 455 −472.

Antihypertensive Effects of Tempol and Other Nitroxides. Pharmacol.

Rev. 60, 418−469.

36) Thiemermann, C. (2003) Membrane-permeable radical

scavengers (tempol) for shock, ischemia-reperfusion injury, and

inflammation. Crit. Care Med. 31, S76−S84.

(37) Dikalova, A. E., Kirilyuk, I. A., and Dikalov, S. I. (2015)

Antihypertensive effect of mitochondria-targeted proxyl nitroxides.

Redox Biol. 4, 355−362.

(38) Poulin, R., Casero, R. A., and Soulet, D. (2012) Recent

advances in the molecular biology of metazoan polyamine transport.

Amino Acids 42, 711−723.

(39) Burner, U., Jantschko, W., and Obinger, C. (1999) Kinetics of

oxidation of aliphatic and aromatic thiols by myeloperoxidase

compounds I and II. FEBS Lett. 443, 290−296.

(40) Belting, M., Haysmark, B., Jönsson, M., Persson, S., and

Fransson, L. Å. (1996) Heparan sulphate/heparin glycosaminoglycans

with strong affinity for the growth-promoter spermine have high

antiproliferative activity. Glycobiology 6, 121−129.

(41) Goldstein, S., Merenyi, G., Russo, A., and Samuni, A. (2003)

The role of oxoammonium cation in the SOD-mimic activity of cyclic

nitroxides. J. Am. Chem. Soc. 125, 789−795.

692