Carnosine protects cardiac myocytes against lipid peroxidation products

Article abstract of DOI:10.1007/s00726-018-2676-6

Endogenous histidyl dipeptides such as carnosine (β-alanine-l-histidine) form conjugates with lipid peroxidation products such as 4-hydroxy-trans-2-nonenal (HNE and acrolein), chelate metals, and protect against myocardial ischemic injury. Nevertheless, it is unclear whether these peptides protect against cardiac injury by directly reacting with lipid peroxidation products. Hence, to examine whether changes in the structure of carnosine could affect its aldehyde reactivity and metal chelating ability, we synthesized methylated analogs of carnosine, balenine (β-alanine-N^τ-methylhistidine) and dimethyl balenine (DMB), and measured their aldehyde reactivity and metal chelating properties. We found that methylation of N^τ residue of imidazole ring (balenine) or trimethylation of carnosine backbone at N^τ residue of imidazole ring and terminal amine group dimethyl balenine (DMB) abolishes the ability of these peptides to react with HNE. Incubation of balenine with acrolein resulted in the formation of single product (m/z 297), whereas DMB did not react with acrolein. In comparison with carnosine, balenine exhibited moderate acrolein quenching capacity. The Fe²⁺ chelating ability of balenine was higher than that of carnosine, whereas DMB lacked chelating capacity. Pretreatment of cardiac myocytes with carnosine increased the mean lifetime of myocytes superfused with HNE or acrolein compared with balenine or DMB. Collectively, these results suggest that carnosine protects cardiac myocytes against HNE and acrolein toxicity by directly reacting with these aldehydes. This reaction involves both the amino group of β-alanyl residue and the imidazole residue of l-histidine. Methylation of these sites prevents or abolishes the aldehyde reactivity of carnosine, alters its metal-chelating property, and diminishes its ability to prevent electrophilic injury.

Full text of DOI:10.1007/s00726-018-2676-6

Amino Acids
https://doi.org/10.1007/s00726-018-2676-6
ORIGINAL ARTICLE
Carnosine protects cardiac myocytes against lipid peroxidation
products
Jingjing Zhao1,2 · Dheeraj Kumar Posa1,2 · Vijay Kumar3 · David Hoetker1,2 · Amit Kumar3 · Smirthy Ganesan1,2
Daniel W. Riggs1,2 · Aruni Bhatnagar1,2 · Michael F. Wempe3 · Shahid P. Baba1
·
Received: 13 July 2018 / Accepted: 30 October 2018
© Springer-Verlag GmbH Austria, part of Springer Nature 2018
Abstract
Endogenous histidyl dipeptides such as carnosine (β-alanine-l-histidine) form conjugates with lipid peroxidation products
such as 4-hydroxy-trans-2-nonenal (HNE and acrolein), chelate metals, and protect against myocardial ischemic injury.
Nevertheless, it is unclear whether these peptides protect against cardiac injury by directly reacting with lipid peroxidation
products. Hence, to examine whether changes in the structure of carnosine could afect its aldehyde reactivity and metal
chelating ability, we synthesized methylated analogs of carnosine, balenine (β-alanine-Nτ-methylhistidine) and dimethyl
balenine (DMB), and measured their aldehyde reactivity and metal chelating properties. We found that methylation of Nτ
residue of imidazole ring (balenine) or trimethylation of carnosine backbone at Nτ residue of imidazole ring and terminal
amine group dimethyl balenine (DMB) abolishes the ability of these peptides to react with HNE. Incubation of balenine with
acrolein resulted in the formation of single product (m/z 297), whereas DMB did not react with acrolein. In comparison with
carnosine, balenine exhibited moderate acrolein quenching capacity. The Fe2+ chelating ability of balenine was higher than
that of carnosine, whereas DMB lacked chelating capacity. Pretreatment of cardiac myocytes with carnosine increased the
mean lifetime of myocytes superfused with HNE or acrolein compared with balenine or DMB. Collectively, these results
suggest that carnosine protects cardiac myocytes against HNE and acrolein toxicity by directly reacting with these aldehydes.
This reaction involves both the amino group of β-alanyl residue and the imidazole residue of l-histidine. Methylation of
these sites prevents or abolishes the aldehyde reactivity of carnosine, alters its metal-chelating property, and diminishes its
ability to prevent electrophilic injury.
Keywords Acrolein · Cardiac myocytes · Histidyl dipeptides · 4-Hydroxy-trans-2-nonenal
Introduction
and aging (Giordano 2005; Sawyer et al. 2002). However
long-term studies with humans that targeted reactive oxygen
species with antioxidants such as beta carotene (Hennek-
ens et al. 1996) and vitamin E (Yusuf et al. 2000) failed to
show any benefcial efects on cardiovascular disease. The
inability of antioxidants to decrease cardiovascular disease
risk is likely due to the complex pathophysiological role
of ROS, which at low concentrations participate in signal-
ing, whereas at high concentration they cause tissue dam-
age by reacting with proteins, DNA and phospholipids (Hal-
liwell 2015). Oxidation of phospholipids in the fatty acid
membranes results in the generation of several highly toxic
lipid peroxidation products such as 4-hydroxy-trans-2-non-
enal (HNE) and acrolein (Porter et al. 1995; Niki 2009).
Increased levels of HNE or HNE-modifed protein adducts
have been detected in the hearts of dilated cardiomyopathic
hamsters (Kato et al. 2010), hypertensive rats (Benderdour
Extensive evidence suggests that oxidative stress is a sig-
nifcant feature of myocardial injury due to heart failure,
ischemia–reperfusion (I/R) injury, pathological hypertrophy
Handling Editor: W. Derave.
* Shahid P. Baba
spbaba01@louisville.edu
1
Department of Medicine, Diabetes and Obesity Center,
University of Louisville, 580 South Preston Street, Delia
Baxter Building, Room 421A, Louisville, KY 40202, USA
2
Department of Medicine, Envirome Institute, University
of Louisville, Louisville, KY, USA
3
Skaggs School of Pharmacy and Pharmaceutical Sciences,
University of Colorado, Aurora, CO, USA
Vol.:(0123456789)
1 3

J. Zhao et al.
et al. 2003, 2004), rats infused with isoproterenol (Srivas-
tava et al. 2007), mice subjected to transaortic constriction
(Zhang et al. 2007), heart failure (Liu et al. 2005), rabbits
subjected to myocardial infarction-induced heart failure
(Qin et al. 2007), and dogs subjected to tachycardia-induced
heart failure (Zhang et al. 2009). Lipid peroxidation products
such as HNE are metabolically removed by several enzymes
including aldose reductase (AR) (Baba et al. 2018; Srivas-
tava et al. 1998, 1999), aldehyde dehydrogenase (ALDH2)
(Chen et al. 2008; Gomes et al. 2014), and glutathione-
S-transferases (GSTs) (Conklin et al. 2015). Although
extensive investigations have delineated the pathological
role of lipid peroxidation products and identifed metabolic
pathways that detoxify these toxic aldehydes, non-enzymatic
pathways, which can sequester these aldehydes and protect
against myocardial injury, have been less well studied.
Most animals and humans synthesize a wide variety of
histidyl dipeptides, such as carnosine anserine (β-alanine-
Nπ-methylhistidine), balenine (β-alanine-Nτ-methylhistidine)
and homocarnosine (ϒ-amino-butyryl-histidine) (Boldyrev
et al. 2013). Among these dipeptides, carnosine is most
abundant in the humans, whereas its methylated analogue,
anserine, is found mostly in birds (Boldyrev et al. 2013).
Balenine is present primarily in whales (Abe 2000). Car-
nosine is synthesized from β-alanine and histidine by the
enzyme carnosine synthase (ATPGD1) (Drozak et al. 2010),
which belongs to the ATP grasp family of enzymes (Fawaz
et al. 2011). It is accumulated to high abundance in tissues
with high metabolic activity, such as the heart, the skeletal
muscle, and the brain (Boldyrev et al. 2013). Because these
peptides contain the imidazole ring of l-histidine, they can
bufer intracellular pH. This bufering activity of histidyl
dipeptides is believed to facilitate glycolysis by bufering
intracellular pH under anaerobic conditions (Hipkiss 2009).
In addition to the imidazole ring, these peptides also con-
tain highly reactive nucleophilic amines, which impart them
the ability to bind with a wide variety of lipid peroxida-
tion products such as HNE and acrolein (Aldini et al. 2002;
Carini et al. 2003). The reaction products between carnosine
HNE and acrolein are extruded out in human urine (Baba
et al. 2013). Previous work has shown that in obese diabetic
patients, carnosine supplementation increases the extrusion
of advanced glycation end products and improves the mark-
ers of metabolic syndrome (de Courten et al. 2016).
oxidative stress. Nevertheless, the mechanism of this pro-
tection remains unclear. During oxidative stress, carnosine
can protect the heart either by bufering intracellular pH,
chelating metals, quenching singlet oxygen or reacting
directly with the electrophilic products of lipid peroxidation.
To distinguish between these possibilities, we synthesized
methylated analogs of carnosine, balenine and di-methyl
balenine (DMB), in which the imidazole ring of l-histidine
and the amino group of β-alanine are methylated. We rea-
soned that if carnosine prevents electrophilic injury due to
HNE and acrolein by directly reacting with these aldehydes,
then methylation of its reactive groups (the amino group of
β-alanine and the imidazole ring of histidyl residue) should
diminish its reactivity with HNE and acrolein, and thereby
prevent or abolish its protective efects.
Materials and methods
d-Histidine was purchased from TCI America. Hydro-
chloric acid gas, 1,1′-carbonyldiimidazole, methyl iodide,
3-(dimethylamino) propanoic acid hydrochloride, eth-
ylenediaminetetraacetic acid (EDTA), ferrozine, EDC
(N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide), and
N-methylmorpholine, carnosine, were purchased from
Sigma-Aldrich and tyrosine-histidine (internal standard:
IS) was purchased from Bachem. Dry dicholoro methane
(DCM) and dry N, N dimethylformamide (DMF) were pur-
chased from Acros Organics. Adult male C57BL/6 mice
were obtained from The Jackson Laboratory (Bar Harbor,
ME).
NMR spectroscopy and thin layer chromatography
1
The H and 13C NMR spectra were recorded in NMR
1
spectrometers operating at 400.00 MHz for H NMR and
100.6 MHz for 13C NMR. The chemical shifts were cali-
brated from the residual peaks observed for the deuterated
solvents such as chloroform (CDCl3), dimethyl sulfoxide
(DMSO-d6), and D2O at δ 7.26, 2.50, and 4.79 ppm for 1H
NMR, respectively. Chemical shifts were calibrated from the
residual peaks observed for the deuterated solvents such as
chloroform (CDCl3), and dimethyl sulfoxide (DMSO-d6) at
δ 77.0 and 39.5 ppm for 13C NMR, respectively. Multiplici-
ties of 1H NMR spin couplings are: s for singlet, bs for broad
singlet, or m for multiplet and overlapping spin systems. Val-
ues for the coupling constants (J) are reported in hertz (Hz).
Thin layer chromatography (TLC) was performed using pre-
coated silica gel plates and visualized by UV light (254 nm).
Recent work from our laboratory has shown that car-
nosine levels are depleted in hypertrophic and failing
hearts (Sansbury et al. 2014). Moreover, we have found
that post-ischemic recovery of contractile function after
global ischemia is signifcantly improved in mouse hearts
perfused with carnosine (Baba et al. 2013), suggesting that
carnosine prevents myocardial injury during conditions of
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
(S)-1-carboxy-2-(1-methyl-1H-imidazol-4-yl) ethana-
minium chloride (4):
Synthesis of balenine
(S)-methyl 5-oxo-5,6,7,8-tetrahydroimidazo[1,5-c] pyrimi-
dine-7-carboxylate (2):
In a round-bottomed flask, (S)-methyl 2-amino-
3-(1H-imidazol-5-yl) propanoate dihydrochloride (5.0 g,
20.65 mmol) and 1,1′-carbonyldiimidazole (3.52 g,
21.7 mmol) were added to DMF (100 mL) at room tem-
perature (RT). The reaction mixture was heated at 60 °C
for 6 h and evaporated to take of all the DMF. The crude
material was purifed with silica gel column chromatog-
raphy to get the desired compound, (S)-methyl 5-oxo-
5,6,7,8-tetrahydroimidazo[1,5-c] pyrimidine-7-carboxy-
late, (2, 2.9 g, 69%) as a white solid compound.
In a round-bottom fask, (S)-7-(methoxycarbonyl)-2-
methyl-5-oxo-5,6,7,8-tetrahydroimidazo[1,5-c] pyrimidin-
2-ium iodide (3, 4.00 g, 11.9 mmol) was added to 4 N
HCl (100 mL). The reaction mixture was refuxed for 6 h,
evaporated to dryness, and the solid extract was washed
with hexanes and dried to get the desired compound, (S)-
1-carboxy-2-(1-methyl-1H-imidazol-4-yl) ethanamin-
ium chloride, (4, 2.10 g, yield = 87%) as a yellow solid
compound.
1H NMR (400 MHz, D2O) δ 8.69 (bs, 1H), 7.44 (bs,
1H), 4.36 (t, J = 6.60 Hz, 1H), 3.90 (s, 3H), 3.32–3.19 (m,
2H) ppm.
1H NMR (400 MHz, DMSO d6) δ 8.55–8.53 (m, 1H),
8.10–8.09 (m, 1H), 6.82–6.80 (m, 1H), 4.45–4.42 (m, 1H),
3.62 (s, 3H), 3.22–3.20 (m, 2H) ppm.
(S)-methyl 2-amino-3-(1-methyl-1H-imidazol-4-yl) pro-
panoate hydrochloride (5):
S)-7-(methoxycarbonyl)-2-methyl-5-oxo-5,6,7,8-
tetrahydroimidazo[1,5-c] pyrimidin-2-ium iodide (3):
In a round-bottomed fask, (S)-methyl 5-oxo-5,6,7,8-
tetrahydroimidazo[1,5-c] pyrimidine-7-carboxylate
(2, 3.00 g, 15.4 mmol) and methyl iodide (3.83 mL,
102.5 mmol) were added to acetonitrile (100 mL) at
In 300 mL of methanolic hydrochloric acid, (S)-
2-amino-3-(1-methyl-1H-imidazol-4-yl) propanoic acid
hydrochloride (4, 2.60 g, 12.64 mmol) was added at RT.
The reaction mixture was then refluxed for 48 h, and
then evaporated to dryness to get more than 99% pure of
(S)-methyl 2-amino-3-(1-methyl-1H-imidazol-4-yl) pro-
panoate hydrochloride, (5, 2.60 g, 94%) as a white solid.
1H NMR (400 MHz, D2O) δ 8.70 (bs, 1H), 7.44 (bs,
1H), 4.51 (t, J = 6.80 Hz, 1H), 3.91 (s, 3H), 3.87 (s, 3H)
3.34–3.20 (m, 2H) ppm.
o
RT. The reaction mixture was heated at 80 C for 16 h,
and evaporated to dryness to get the desired compound,
(S)-7-(methoxycarbonyl)-2-methyl-5-oxo-5,6,7,8-
tetrahydroimidazo[1,5-c] pyrimidin-2-ium iodide, (3,
4.9 g, yield = 94%) as an of-white solid compound.
1H NMR (400 MHz, D2O) δ 9.39 (s, 1H), 7.44 (bs, 1H),
4.77–4.75 (m, 1H), 3.99 (s, 3H), 3.80 (s, 3H), 3.53–3.51
(m, 2H) ppm.
1 3

J. Zhao et al.
(S)-methyl 2-(3-(tert-butoxycarbonylamino)
propanamido)-3-(1-methyl-1H-imidazol-4-yl)propanoate
(6):
(S)-2-(3-aminopropanamido)-3-(1-methyl-1H-imidazol-
4-yl) propanoic acid, (7, 2.30 g, yield = 76%) as a solid
compound.
1H NMR (400 MHz, D2O) δ 7.54 (bs, 1H), 6.94 (bs,
1H), 4.49–4.45 (m, 1H), 3.69 (s, 3H), 3.63–3.57 (m, 1H),
3.14–2.87 (m, 3H), 2.53–2.50 (m, 1H), 2.26–2.23 (m, 1H)
ppm.
Synthesis of dimethyl‑balenine
Steps 1–5 for dimethyl-balenine synthesis were similar to
balenine
In a reaction flask under nitrogen, we sequentially
added (S)-methyl 2-amino-3-(1-methyl-1H-imidazol-
4-yl) propanoate hydrochloride (5, 5.00 g, 22.78 mmol),
Boc-b-alanine (4.70 g, 25.04 mmol), EDC (N-(3-
Dimethylaminopropyl)-N′-ethylcarbodiimide) (4.80 g,
25.04 mmol) followed by DCM (100 mL) and N-meth-
ylmorpholine (10.0 mL, 91.04 mmol via syringe). The
contents were stirred at RT under nitrogen and the sol-
ids were gradually dissolved. The contents were stirred
at RT for 24 h, and slowly diluted into iced water and
extracted with DCM. The organic phase was dried, evapo-
rated and chromatographed using DCM and methanol as
eluents to get (S)-methyl 2-(3-(tert-butoxycarbonylamino)
propanamido)-3-(1-methyl-1H-imidazol-4-yl) propanoate
(6, 3.90 g, yield = 49%) as a white solid compound.
1H NMR (400 MHz, CDCl3) δ 7.46–7.43 (m, 1H),
7.38–7.34 (m, 1H), 6.74 (s, 1H), 5.78 (bs, 1H), 4.91–4.87
(m, 1H), 3.81 (s, 3H), 3.72 (s, 3H) 3.56–3.51 (m, 2H),
3.20–3.08 (m, 2H), 2.58–2.46 (m. 2H), 1.54 (s, 9H) ppm.
(S)-2-(3-aminopropanamido)-3-(1-methyl-1H-imidazol-
4-yl) propanoic acid (7):
Synthesis of (R)-methyl 2-(3-(dimethylamino)
propanamido)-3-(1-methyl-1H-imidazol-4-yl)propanoate
(7): (R)-1-methoxy-3-(1-methyl-1H-imidazol-4-yl)-1-ox-
opropan-2-aminium chloride (2.5 g, 8.9 mmol), 3-(dimeth-
ylamino) propanoic acid hydrochloride (1.5 g, 9.8 mmol),
EDC (N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide)
(1.9 g, 9.8 mmol) and DCM (100 mL) were added sequen-
tially to a reaction fask under nitrogen. The solution was
cooled in an ice water bath under nitrogen. N-methyl-
morpholine (3.9 mL, 35.8 mmol via syringe) was added
slowly. The ice water bath was removed 5–10 min after
the completion of addition. The contents were stirred
at RT under nitrogen and the solids were gradually dis-
solved. The reaction mixture was stirred at RT for 48 h.
The reaction mixture was dried on reduced pressure and
chromatographed using DCM and methanol as eluents to
get desired compound, (R)-methyl 2-(3-(dimethylamino)
propanamido)-3-(1-methyl-1H-imidazol-4-yl) propanoate
(7, 1.1 g, 34%) as a solid compound. 1H NMR (400 MHz,
CDCl3) δ 8.89 (bs, 1H), 8.81 (d, J = 7.99 Hz, 1H), 7.13
(bs, 1H), 4.85–4.78 (m, 1H), 3.96 (s, 3H), 3.75 (s, 3H)
3.60–3.52 (m, 1H), 3.37–3.28 (m, 3H), 3.23–3.15 (m.
1H), 2.94–2.88 (m. 1H), 2.87 (s, 6H) ppm. 13C NMR
(100.6 MHz, CDCl3) δ 170.62, 169.36, 135.06, 130.75,
120.49, 53.65, 52.87, 52.30, 50.79, 43.49, 36.23, 30.70,
26.04 ppm.
To (S)-methyl 2-(3-(tert-butoxycarbonylamino)
propanamido)-3-(1-methyl-1H-imidazol-4-yl) pro-
panoate (6, 4.50 g, 12.7 mmol), 2 N NaOH (38.1 mL)
in THF (70 mL) and methanol (70 mL) was added. The
reaction mixture was stirred at RT for 24 h. The sol-
vent was evaporated and the reaction mixture was neu-
tralized by 1 N HCl to pH 3.0 and stirred for 1 h. The
reaction was neutralized by saturated sodium bicar-
bonate. Water was evaporated to dryness and the solid
was washed with DCM and dried again. The obtained
solid was dissolved in hot ethanol and passed through
the Celite plug, twice, to get the desired compound,
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
Metal chelating capacity
The iron (Fe2+) chelating ability of different histidyl
dipeptides was monitored by measuring the formation
of ferrous–ferrozine complex as described previously by
(Canabady-Rochelle et al. 2015). Briefy, 2 mM ferrous
chloride (FeCl2) solution was added to diferent dipep-
tide solutions in pure water. After 3 min of incubation at
room temperature, the chelation reaction was inhibited by
the addition of 5 mM ferrozine solution. The absorbance
was measured at 560 nm after 10 min. EDTA was used
as a reference chelator and the metal chelating capacity
of dipeptides was calculated in comparison to EDTA.
Metal quenching capacity was measured as decrease in
the absorbance of ferrous–ferrozine complex and reported
as follows:
Synthesis of (R)-2-(3-(dimethylamino) propanamido)-
3-(1-methyl-1H-imidazol-4-yl) propanoic acid (8): To (R)-
methyl 2-(3-(dimethylamino) propanamido)-3-(1-methyl-
1H-imidazol-4-yl) propanoate (0.5 g, 1.8 mmol), 2 N NaOH
(3.54 mL) in THF (10 mL) and methanol (10 mL) was added
and the reaction was stirred for 16 h at RT. Later, the sol-
vent was evaporated under reduced pressure and the pH of
the reaction mixture was adjusted to 7 with 1 N HCl solu-
tion. Water was evaporated to dryness and the solid product
was washed with DCM (2 × 20 mL) and dried again. The
obtained solid was dissolved in ethanol (6.0 mL) and passed
through the Celite plug (5.0 g), twice, to get the desired
compound, (R)-2-(3-(dimethylamino) propanamido)-3-(1-
methyl-1H-imidazol-4-yl) propanoic acid (8, 360 mg, 76%
yield) as an of-white solid compound. 1H NMR (400 MHz,
D2O) δ 7.90 (bs, 1H), 6.98 (bs, 1H), 4.40–4.34 (m, 1H), 3.67
(s, 3H), 3.37–3.25 (m, 2H), 3.07–3.00 (m, 1H), 2.94–3.15
(m, 1H), 2.94–2.86 (m, 1H), 2.80 (s, 6H), 2.74–2.63 (m,
2H) ppm.13C NMR (100.6 MHz, D2O) δ 177.25, 170.99,
136.58, 133.92, 119.61, 54.93, 53.63, 42.81, 34.10, 29.73,
28.66 ppm. LC/MS–MS (18C column), tR =5.4 min, 269.1
→ 223.8 m/z.
(
)
Iron chelating capacity (%) = A0 − As∕A0 × 100,
(1)
A0 is the blank absorbance and As is the absorbance
of the sample dipeptide. The line slope obtained for each
dipeptide was compared with EDTA to determine the
EDTA equivalent chelating capacity expressed as μmol
EDTA equivalent. EDTA equivalent chelating capacity for
each dipeptide was calculated as follows:
EDTA equivalent chelating capacity = as∕a0,
(2)
where as is the line slope of dipeptide for chelating activity,
plotted as iron chelating capacity (percent) vs concentration
(μM), a0 is the line slope of EDTA for chelating activity,
plotted iron chelating capacity (percent) vs concentration
(μM).
Synthesis of histidyl‑dipeptide‑aldehyde conjugates
Rate constants
Histidyl dipeptide–aldehyde conjugates were prepared
as described previously (Baba et al. 2013). Briefy, acr-
olein was synthesized by acid hydrolysis of diethyl acetal
acrolein (Sigma) in HCl (0.1 M) at room temperature for
30 min. Carnosine–propanal and balenine–propanal conju-
gates were synthesized by incubating 100 mM acrolein (20
µL) with 100 mM carnosine (20 µL) or 100 mM balenine
(20 µL) in 1 mM phosphate bufer pH 7.0 at 37°C for 2 h.
Carnosine–HNE conjugate was synthesized by incubating
HNE (0.2 mM; Cayman Chemical) with 2 mM carnosine
in 1 mM phosphate bufer pH 7.0 at 37 °C for 2 h at room
temperature. For mass spectrometry, the dipeptides and
histidyl dipeptide–aldehyde conjugates were individu-
ally infused into a stream of 0.55 mL/min 50:50 A: B
UPLC solvent going into a Waters Quattro Premier XE
triple quadrupole mass spectrometer (Milford, MA). We
manually adjusted all voltage and collision energy setting
to determine the optimal ionization voltages, collision
The rate constants for the reactions of peptides with alde-
hydes were measured as described previously (Barski et al.
2013). Briefy, HNE and acrolein were incubated with car-
nosine, balenine and DMB (50μM-2 mM) in 0.15 M potas-
sium phosphate, pH 7.4 at 37°C. Disappearance of acrolein
and HNE was monitored by UV–Vis spectrophotometer at
approximately 215 and 223 nm, respectively. A mono-expo-
nential equation was used to ft the time course of each reac-
tion and the exponential constant, kobs, was plotted against
the histidyl dipeptide concentrations. Plots of kobs vs histidyl
dipeptides were straight lines, and the bimolecular rate con-
stants were calculated by their slopes.
1 3

J. Zhao et al.
energies and product ions. Maximum ionization condi-
tions and optimal daughter ions were used for program
sensitive MRMs.
cardiac myocytes/feld) were acquired at 5 min intervals to
quantify numbers of hypercontracted and non-hypercon-
tracted cells.
Isolation of cardiac myocytes histidyl dipeptide
treatment
Statistical analysis
Diference in survival time between the diferent treatments
was analyzed by Student’s t test. Statistical signifcance was
accepted at p<0.05. Group data are mean standard devia-
tion (SD). Survival data for cardiac myocytes were modeled
using the Weibull survival distribution (Castro and Bhatna-
gar 1993). In this distribution, the probability density func-
tion of time t is:
Cardiac myocytes were isolated from adult C57/BL6 mice
by using a modifed Langendorf perfusion system with
collagenase digestion, as described previously (Keith et al.
2009). Briefy, the mouse hearts were excised, rinsed with
physiological saline, and perfused with oxygenated (95%
O2–5% CO2), Ca2+-free modifed Tyrode bicarbonate bufer
(bufer A in mM: NaCl 126, KCl 4.4, MgCl2 1.0, NaHCO3
18, glucose 11, HEPES 4, 2,3-butanedione monoxime 10,
taurine 30, pH 7.35) at 37 °C for 5 min. The extracellu-
lar matrix was digested by bufer containing bufer A with
Liberase Blendzyme type 1 0.25 mg/mL, 1 mg/mL albu-
min, 0.015 mg/mL DNase, 0.015 mg/mL proteinase in the
50 mL of recirculating digestion bufer (bufer A with 25 µM
CaCl2) for 12–15 min. Hearts were separated in mincing
bufer (10 mL digestion bufer with 9 mg/mL albumin, and
Liberase) and cells were allowed to settle. CaCl2 was reintro-
duced in a graded fashion at 5-min intervals (fve total steps)
to sequentially increase the Ca2+ concentration to 500 µM.
f(t) = ꢀꢁ(ꢀt)ꢁ−t exp[−(ꢀt)ꢁ ].
(3)
γ is the shape parameter, λ is the scale parameter: which
are the scale and shape factors, respectively. The normalized
survival function (at t=0, S=1) is:
S(t) = exp [(−ꢀt)]ꢁ .
(4)
Mean life times were calculated as: 1/λ Γ (1+1/γ), where
Γ represents the Gamma function. Curve ftting was per-
formed using the Proc NLIN procedure in SAS version 9.4
software (SAS Institute, Inc., Cary, North Carolina).
Uptake of histidyl dipeptides in myocytes
and superfusion with HNE and acrolein
Results
Synthesis of balenine and N, N‑dimethyl d‑balenine
Uptake of histidyl dipeptides in myocytes was measured
using Waters Quattro Premier XE mass spectrometer (Blanc-
quaert et al. 2016). Briefy, myocytes isolated from adult
mice were pretreated with histidyl dipeptides 1 mM: carno-
sine, balenine and DMB for 12-16 h. Lysates were collected
by washing the cells in PBS and collected in 10 mM HCl
containing internal standard tyrosine-histidine, sonicated for
10 s, centrifuged at 16,000×g for 10 min and the superna-
tant was diluted with three volumes of ice-cold acetonitrile.
The samples were thoroughly vortexed to precipitate pro-
teins and kept on ice for 15 min and centrifuged at 16,000×g
for 10 min at 4 °C. Carnosine, balenine, and DMB were
separated by using a Waters Acquity BEH HILIC column
(1.7 µm, 2.1×50 mm) column and analyzed by mass spec-
trometry in the positive ion mode. Chromatograms of histi-
dyl dipeptides were acquired using the transitions carnosine:
227 → 110 m/z, balenine: 241 → 124 m/z, dimethyl bale-
nine 269 → 124 m/z and tyrosine-histidine 319 → 110 m/z
in multiple reaction monitoring (MRM) mode. To examine
whether histidyl dipeptides protect against HNE and acrolein
induced hypercontracture, myocytes were pretreated with
1 mM each of carnosine, balenine and DMB for 12–16 h.
The cells were then superfused with either 50–60 μM HNE
or 5 μM acrolein for 60 min, and digital images (80–100
Histidine has two tautomeric forms N-1 and N-3, which pos-
sess both imine and amine character. To selectively alkylate
the N-3 position of the imidazole ring, we temporarily pro-
tected the N-1 position of (S)-methyl 2-amino-3-(1H-imida-
zol-5-yl) propanoate dihydrochloride (1) by treating it with
1,1′-carbonyldiimidazole in DMF, which aforded the cyclic
urea intermediate 2 as a crystalline solid in a good yield.
Methylation of urea intermediate 2 with methyl iodide in
acetonitrile resulted in the synthesis of N-3 methylated salt 3
in good yield. The cyclic urea 2 was opened with hydrolysis
of methyl ester by refuxing in 4 N HCl to give the amino
acid hydrochloride 4. Acid group of amino acid hydrochlo-
ride 4 was protected by refuxing it in methanolic HCl solu-
tion. Compound 5 was coupled with Boc-β-alanine using
EDC.HCl as the coupling agent and N-methylmorpholine as
a base to aford the desired dipeptide 6. Complete deprotec-
tion of the protecting groups aforded the l-balenine (7) in
good yield (Fig. 1).
The N, N-dimethyl-d-Balenine (DMB) was synthe-
sized following the steps 1–5 described for balenine. The
acid group of amino acid hydrochloride 5 was protected
by refuxing it in methanolic HCl solution. Compound 6
was coupled with 3-(dimethylamino) propanoic acid using
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
Step-2
Step-3
Step-1
Step-4
Step-5
Step-6
Fig. 1 Preparation of balenine. Reagents and conditions for synthesis of naturally occurring histidyl dipeptide balenine steps 1–6
EDC.HCl as the coupling agent and N-methylmorpholine
as a base to aford the desired dipeptide 7. Complete de-
protection of protecting groups aforded the N, N-dimethyl-
d-Balenine (8) in good yield (Fig. 2).
species (m/z 265, 283, 303). Both balenine, which is meth-
ylated at Nτ atom of imidazole ring (Fig. 3c), and DMB, in
which three CH3 groups are attached on the carnosine back-
bone (Fig. 3e), were unable to form conjugates with HNE
(results not shown). Incubation of balenine with acrolein
(10:1 molar ratio) led to the formation of one major ionic
species with m/z of 297 (Fig. 3d), which is likely a Michael
adduct between the carbonyl atom of acrolein and the amino-
terminal group of β-alanine or Nτ of histidine.
Methylation of carnosine diminishes aldehyde
binding capacity
Because Nτ atom of imidazole ring and the tertiary amino
group of β-alanine are requisite for binding with reactive
aldehydes (Carini et al. 2003), we evaluated the efects
of methylation on these residues on aldehyde binding. To
optimize the conditions for characterization of aldehyde-
histidyl dipeptide conjugate formation, carnosine was used
as a reference dipeptide. As reported previously, incuba-
tion of carnosine with HNE (10:1 molar ratio) led to the
formation of one reaction product (m/z 383), which could
be ascribed to a Michael adduct between the C3 of HNE
and Nτ atom of imidazole ring (Fig. 3a) (Baba et al. 2013;
Aldini et al. 2002). Incubation of acrolein with carnosine
(10:1 molar ratio) resulted in the formation of multiple ionic
To examine the relative reactivity of the peptides, we
compared the second-order rate constants for the reaction
of HNE and acrolein with carnosine, balenine, and DMB.
In agreement with our previous observations, incubation of
carnosine with acrolein led to a time-dependent decrease
in the acrolein concentration (Fig. 4a) (Barski et al. 2013),
suggesting rapid reactivity between acrolein and carno-
sine. In contrast, acrolein concentration was only slightly
decreased in the balenine–acrolein mixture compared with
acrolein alone after 5 h of incubation (Fig. 4c). No change
in HNE or acrolein concentrations was observed when the
aldehydes were incubated with DMB, compared with the
1 3

J. Zhao et al.
Fig. 2 Preparation of dimethyl balenine (DMB). Reagents and conditions for synthesis of synthetic carnosine analog DMB steps 1–7
A
B
C
D
E
241
297
100
50
100
100
50
100
50
100
50
269
383
265
50
283
303
-1e-1
m/z
250
300
350
230
240
250
360
390
420
296
297
298
299
260
270
280
Fig. 3 LC/MS spectra of diferent histidyl dipeptides with lipid per-
oxidation products. Histidyl dipeptides: carnosine, balenine, and
dimethyl balenine (DMB) were incubated with HNE or acrolein (10:1
molar ratio) in 1 mM phosphate bufer (pH 7.4). Ionic species were
monitored by ESI/MS: a carnosine–HNE (m/z 383), b carnosine–pro-
panal (m/z 265, 283, 303), and c balenine (m/z 241). Inset shows the
chemical structure of d balenine–propanal (m/z 297); the Michael
adduct of balenine with acrolein; and e dimethyl balenine (m/z 269)
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
Fig. 4 Methylation of carno-
sine backbone diminishes its
aldehyde quenching potential.
Rate of disappearance of
0µM
50µM
150µM
250µM
A
B
100µM
200µM
120
80
40
0
900x10-3
600x10-3
300x10-3
acrolein (100 μM) in a reaction
mixture containing diferent
concentrations of a carnosine, c
balenine, and e dimethyl bale-
nine (50–250 μM) in 0.15 M
potassium phosphate bufer, pH
7.4. f Rate of disappearance of
HNE incubated with DMB. The
reaction mixtures were incu-
bated at 37 °C and the decrease
in absorbance of acrolein and
HNE was monitored at 215 and
223 nm, respectively. b, d Data
are shown as discrete points and
curves were best ft of a single
exponential equation to the data
[y=Ae−kobs.t]. Concentration
dependence of kobs. Second-
order rate constants were
0
50 100 150 200 250
Carnosine (µM)
0
1
2
3
4
5
6
C
D
120
80
40
0
420x10-3
360x10-3
300x10-3
0
1
2
3
4
5
6
0
50 100 150 200 250 300
Balenine (µM)
calculated from the best fts of
the linear dependence and are
shown in Table 1
E
F
60
40
20
0
180
120
60
0
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Time (h)
Time (h)
Table 1 Bimolecular rate constant of histidyl dipeptides with HNE
and HNE with these dipeptides. Among diferent histidyl
dipeptides, carnosine exhibited highest reactivity with HNE
and acrolein (Table 1). The second-order rate constant for
the reaction between acrolein and balenine was 19% of the
rate constant of the reaction between carnosine and acrolein.
These observations suggest that methylation of Nτ atom of
imidazole ring or terminal amino group decreases or dimin-
ishes the aldehyde binding ability of carnosine.
and acrolein
Aldehydes Carnosine (M−1 s−1
)
Balenine (M−1 s−1
)
Dimethyl
balenine
(M−1 s−1
)
4HNE
Acrolein
0.035 0.003
0.288 0.013
nd
nd
nd
0.055 0.005
nd not detected
Metal quenching capacity of histidyl dipeptides
HNE and acrolein incubated alone for 5 h (Fig. 4e, f). The
rate of decrease in HNE and acrolein concentration could
be described by a single exponential: At =A0e−kobs.t, where
At is the concentration of aldehyde at time t, A0 is the initial
aldehyde concentration, and kobs is the apparent rate constant
of the reaction. The rate constants were calculated using this
equation at several diferent concentrations of dipeptides. As
shown in Fig. 4b, d and Table 1 the values of kobs showed
a linear dependence on carnosine and balenine concentra-
tion with a slope equal to the second-order rate constant,
which were used to calculate the rate of reaction for acrolein
Given that carnosine chelates numerous frst transition met-
als (Baran 2000), we next determined whether methylation
at Nτ atom of imidazole ring (balenine) or trimethylation
(DMB) could afect its metal quenching capacity. Using
EDTA as a reference chemical chelator, we compared the
Fe2+ chelating ability of these dipeptides and calculated the
EDTA equivalent chelating capacity. EDTA as a reference
chelator showed a linearization of metal quenching in the
range 0–50 μM (Fig. 5a, d), carnosine and balenine showed
linearization of metal quenching in the range between 0 and
1 3

J. Zhao et al.
A
B
C
D
Peptides
EDTA eq. chelating
90
60
30
0
capacity (µmole)
90
60
30
0
90
60
30
0
0.084 0.005
Carnosine
Balenine
0.179 0.001*
Not detectable
Dimethylbalenine
µM
0
10 20 30 40 50
0
200 400 600 800 1000
Carnosine
0
50 100 150 200 250
Balenine
EDTA
Fig. 5 Metal chelating capacity of diferent histidyl dipeptides. Iron (Fe2+) chelating capacity of a EDTA b carnosine and c balenine. d EDTA
equivalent chelating activity of diferent histidyl dipeptides
500 μM (Fig. 5b, d) and 0–250 μM (Fig. 5c, d) respectively.
In reference to EDTA, balenine exhibited the highest chelat-
ing capacity (0.179 0.002 μmol of EDTA equivalent)
compared with carnosine (0.087 0.005 μmol of EDTA
equivalent; p<0.05), whereas trimethylation of carnosine
backbone abolished the metal chelating property (Fig. 5d).
Taken together, these results suggest that single methyla-
tion enhances and trimethylation diminished metal chelating
property of carnosine.
carnosine levels in the isolated adult cardiac myocytes using
LC/MS/MS and detected~0.249 0.035 nmole/mg protein
of carnosine in these cells. To ensure that the dipeptides
are taken up by the cardiac myocytes, we incubated these
cells with 1 mM of each carnosine, balenine and DMB for
12–16 h and dipeptide concentration was measured by LC/
MS/MS. Our results showed that under similar incubation
conditions, the intracellular concentrations of the dipep-
tides in cardiac myocytes were comparable, indicating that
all three dipeptides are taken up by isolated myocytes (car-
nosine: 2.77 1.032 nmole/mg protein; balenine: 3.759
1.388 nmole/mg protein; and DMB: 6.243 1.482 nmole/
mg protein; Fig. 6a). To test whether these dipeptides protect
against HNE-induced hypercontracture, myocytes were pre-
treated with 1 mM each of carnosine, balenine, and DMB for
12–16 h and superfused with HNE (50–60 μM) for 60 min.
The cells were then continuously monitored under the light
microscope and survival was computed by comparing their
survival using the Weibull distribution. We found that during
Carnosine protects against HNE‑induced
hypercontracture in myocytes
Because carnosine binds with reactive aldehydes (Aldini
et al. 2002) and excessive HNE generation is a hallmark
of cardiac dysfunction (Baba et al. 2018), we next tested
whether enhancing carnosine levels in the isolated adult
mouse cardiac myocytes would prevent against HNE-
induced myocyte injury. For this, we first measured
55 min
0 min
HNE
+Carnosine +Balenine +DMB
A
B
C
#
1.2
0.8
0.4
0.0
*
70
*
6
3
0
*
*
HNE
35
Car
Bal
DMB
0
X
Data
HNE Car HNE DMB HNE Bal
X
Data
0
10 20 30 40 50 60 70
NT Car NT Bal NT DMB
Time (min)
Fig. 6 Carnosine pretreatment prevents HNE-induced hypercontrac-
ture. a Levels of histidyl dipeptides in myocytes non-treated (NT) or
treated with 1 mM each of carnosine (car), balenine (bal), and dime-
thyl balenine (DMB). Inset shows representative images of myocytes
at 0 min and after 55 min of superfusion with HNE (50–60 μM)
with and without pretreatment with the indicated histidyl dipeptides.
Hypercontracture was monitored at 5 min intervals for 60 min. b
Mean lifetime of myocytes was calculated using the Weibull Sur-
vival distribution function (S(t)=exp–(λt)γ). c Data for mean life
time are presented as mean SD. *p<0.05 vs HNE treated myocytes,
#p<0.05 vs carnosine and balenine treated myocytes n=3–5 mice in
each group
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
the initial phase of superfusion, there was a progressive
increase in the hypercontracture; however, after 40-45 min
of superfusion, the number of hypercontracted cells in the
presence of HNE increased abruptly (Fig. 6b). The mean
lifetime of the myocytes superfused with HNE was 42
6 min, which was signifcantly increased to 53 8 min
(p<0.05) in myocytes preloaded with carnosine. In contrast,
balenine and DMB pretreatments were unable to diminish
HNE toxicity and thus afect the mean lifetime of superfused
myocytes (balenine: 41 9 min; DMB: 45 8 min; Fig. 6c).
Discussion
The major fnding of this study is that carnosine protects
against aldehyde-induced hypercontracture in isolated car-
diac myocytes and that this protection is related to the ability
of this dipeptide to directly scavenge electrophilic aldehydes
by forming covalent adducts. This notion is supported by
the observation showing that the decrease in aldehyde bind-
ing capacity by methylation of Nτ of imidazole ring, but
preserved metal chelating property and trimethylation of
histidyl dipeptides was accompanied by decreased ability
of the dipeptide to protect against aldehyde induced hyper-
contracture. Collectively, these results add a new facet to
the biology of carnosine and corroborate the critical role of
its direct reactive ability in preventing electrophilic injury.
Previous work has shown that carnosine forms a con-
jugate with HNE through a mechanism that involves the
formation of reversible α, β unsaturated imine between
the amino group on the β-alanyl residue and the aldehyde
carbonyl of HNE. This intermediate imine acts as a cata-
lyst for the Michael addition between the C-3 of HNE and
the Nτ atom of imidazole ring (Aldini et al. 2002). It has
been shown that HNE also binds with anserine, which has
a methyl residue attached to Nπ atom of imidazole (Aldini
et al. 2002), indicating that Nτ atom of imidazole is neces-
sary for HNE binding. However, less is known about the
aldehyde binding ability of balenine, which is methylated at
Nτ atom of imidazole ring and is largely present in whales
(Abe 2000). Our LC/MS results showed that balenine was
unable to react with HNE, which further highlights the role
of Nτ atom in imidazole ring for HNE binding. Although it
was expected that an intermediate imine derivative between
the amino group of β-alanine and the aldehyde carbonyl of
HNE might be formed, our results suggest that the suppres-
sion of Michael type addition between the C-3 of HNE and
Carnosine pretreatment prevents acrolein‑induced
hypercontracture in cardiomyocyte
Because a wide variety of reactive aldehydes are generated
in diseased myocardium (Baba et al. 2018) and carnosine has
the ability to bind several diferent aldehydes (Carini et al.
2003), we further tested whether carnosine could also pre-
vent acrolein-induced hypercontracture. As before, adult car-
diac myocytes were pretreated with 1 mM each of carnosine,
DMB and balenine for 12–16 h and then superfused with 5
μM acrolein. Hypercontracture in the superfused myocytes
was continuously monitored for 60 min. As shown in Fig. 7,
the mean lifetime of myocytes superfused with acrolein was
41 3 min which was signifcantly increased to 50 4 min
in myocyte pretreated with carnosine. In contrast, balenine
and DMB pretreatments did not reduce acrolein-induced
toxicity or afect myocyte survival (balenine: 43 3 min;
DMB: 41 3 min). Taken together, these results suggest
that methylation of carnosine either at the amino group of
β-alanine or the imidazole ring of l-histidine abolishes its
ability to prevent electrophilic injury to cardiac myocytes.
This observation is consistent with the notion that carnos-
ine prevents HNE and acrolein-induced injury by forming
conjugates with these aldehydes.
Fig. 7 Acrolein-induced hyper-
50 min
contracture is attenuated by car-
0min
Acrolein +Carnosine +Balenine
+DMB
nosine pretreatment. a Hyper-
contracture was monitored
at 5 min interval for 60 min.
Mean lifetime was calculated
using the Weibull distribution
A
B
function. Inset shows the rep-
1.2
resentative images of myocytes
at 0 min and after 50 min of
70
35
0
*
acrolein (5 μM) superfusion. b
0.8
Car
Data for mean lifetime are pre-
Bal
Acr
sented as mean SD. *p<0.05
vs acrolein treated myocytes,
0.4
DMB
n=4–5 mice in each group
0.0
X
Data
Acr Car
Acr DMB
Acr Bal
0
10 20 30 40 50 60 70
Time (min)
1 3

J. Zhao et al.
the Nτ atom of imidazole ring in balenine prevents the for-
mation of this imine derivative.
Canabady-Rochelle et al. (2015), using EDTA as a reference
metal chelator reported that carnosine chelates Fe2+, equiva-
lent to 0.2 μmole of EDTA equivalent chelating capacity.
In line with these observations we found that carnosine has
0.08 μmole of EDTA equivalent chelating capacity and the
methylation of carnosine at Nτ of histidine enhanced it to 0.2
μmole of EDTA equivalent chelating capacity, whereas tri-
methylation completely abolished this property. The reason,
why methylation of Nτ enhanced the metal chelating ability
is not clear. Previous studies suggest that the metal chelat-
ing capacity of carnosine is dependent on pH (Dobbie and
Kermack 1955) (Lenz and Martell 1964) and carnosine has a
pKa of 7.1, whereas balenine has a pKa of 6.9 (Okuma 1991).
It seems likely that these diferences in pKa might regulate
the formation of metal-carnosine complexes. However,
future studies are needed to delineate the mechanism by
which methylation at Nτ enhances the metal chelating capac-
ity and whether this methylation prefers the monomeric or
the dimeric complexes. Our results showing that DMB was
unable to chelate metals confrms earlier fndings that NH2
terminal groups of the carnosine backbone are essential for
metal binding (Dobbie and Kermack 1955). Although we
did not test the metal quenching ability of anserine, which
has a pKa 7.15 (Okuma 1991), future studies are needed to
compare the metal quenching ability of this dipeptide and
determine how methylation at Nπ of histidine efects metal
chelating capacity. Taken together, these fndings suggest
that the single methylation of carnosine on Nτ of imida-
zole group does not abolish its metal chelating property, but
diminishes its aldehyde quenching ability, whereas trimeth-
ylation of carnosine abolishes both the metal chelating and
the aldehyde binding ability of peptide.
Incubation of carnosine with acrolein yielded several
ionic species (m/z 265, 283, 321, 339) which are formed due
to the binding of either one, two or three acrolein molecules
with Nτ, Nπ atom of the imidazole ring or terminal amino
group of β-alanine (Carini et al. 2003). Our results show-
ing that incubation of balenine with acrolein furnishes only
one main products at m/z 297, suggest that only one mole
of acrolein binds to one mole of balenine through Michael
addition to either Nπ atom of imidazole ring or to the amino
group of β-alanine. The inability of DMB to react with HNE
or acrolein further confrms the role of Nτ of imidazolic
ring and terminal amine group in aldehyde conjugation.
Although it was expected that acrolein may form a conjugate
with the Nπ atom of imidazole ring in DMB, our ESI/MS
did not detect any ionic species conjugated with acrolein,
suggesting that methylation at three residue of carnosine
blocks its accessibility to small, unsaturated aldehydes. Fur-
thermore, the second-order rate constants for the reaction of
HNE and acrolein with diferent histidyl dipeptides showed
that carnosine reacts faster with acrolein than balenine or
DMB. Previously, we had reported that anserine had either
equal or high, acrolein and HNE binding rate (Barski et al.
2013), which further strengthens the rationale that Nτ of
imidazole moiety is essential for aldehyde binding. Collec-
tively these results suggest that carnosine displays the most
efcient reaction with reactive aldehydes than balenine and
that Nτ of imidazole ring of histidyl dipeptides is essential
for conjugation with the bulkier aldehydes such as HNE.
Several analytical studies have suggested that carno-
sine forms complexes with the transition metals such as
Cu2+, Co2+, Ni2+ (Baran 2000). Among these complexes,
Cu2+–carnosine metal complex is the most well studied
model for metal binding (Dobbie and Kermack 1955) (Lenz
and Martell 1964). It was frst proposed by Dobbie and
Kermack that the binding of carnosine with Cu2+ occurs
as a monomer through the Nτ of imidazole group, terminal
NH2 and the deprotonated peptide nitrogen of carnosine.
Later structural studies using NMR, X-ray crystallographic
studies revealed that Cu2+–carnosine complex exists as a
dimer, in which Cu2+ ion binds with the terminal amino
group, amide nitrogen, and the carboxylate oxygen of one
dipeptide molecule and the Nπ of imidazole group of the
other dipeptide (Freeman and Szymanski 1967). However,
subsequent studies at low or room temperature revealed that
the Cu2+–carnosine complexes could exist as a dimer and
monomer in equilibrium (Brown et al. 1980, 1982). In addi-
tion to copper, carnosine may also chelate iron. Carnosine is
capable of inhibiting iron-catalyzed oxidation of lipid phos-
phohatidycholine liposomes and lipid peroxides, however
the inability of carnosine to form complex with Fe2+ has
also been reported (Decker et al. 1992). Recent study by
Of the several lipid peroxidation products, HNE and
acrolein are some of the most toxic species that have been
implicated as signifcant mediators of cardiac injury under
a variety of conditions, including heart failure (Ismahil et al.
2011), pathological hypertrophy (Baba et al. 2018), I/R
injury (Conklin et al. 2015) and aging (Moreau et al. 2003).
Normal cellular levels of HNE range from 0.8 to 2.8 μM
(Esterbauer et al. 1991) although up to 10–100 mM HNE
has been measured in oxidizing microsomes (Benedetti et al.
1982, 1984). Exposure of isolated myocytes to HNE and acr-
olein causes prolongation of the action potential, increases
arrhythmogenesis, elevation in intracellular sodium, and
cell death (Conklin et al. 2015; Bhatnagar 1995). Recent
reports from our laboratory demonstrated that superfusion
with HNE increases autophagy in cardiac myocytes (Baba
et al. 2018). Perfusion of isolated hearts with HNE leads
to coronary vasodilation (van der Kraaij et al. 1990) and
progressive decrease in the peak systolic pressure (Ishikawa
et al. 1986). Similarly, it has been shown that exposure to
acrolein increases, heart rate variability, left ventricular
developed pressure and arrhythmia (Kurhanewicz et al.
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
2017). In agreement with these observations, we found that
exposure of isolated myocytes to HNE or acrolein causes a
time-dependent hypercontracture and cell death, indicating
that both these aldehydes are highly toxic to cardiac myo-
cytes. Previous studies have shown that overexpression or
activation of enzymes that detoxify aldehydes protect against
cardiac injury mediated by reactive aldehydes. Overexpres-
sion of ALDH2 in the cardiac tissue diminishes the alde-
hydic load in cardiac tissues and ameliorates doxorubicin-
induced myocardial dysfunction (Sun et al. 2014a), improves
contractile function in heart failure mice (Sun et al. 2014b)
and protects the heart against I/R injury (Endo et al. 2009).
Activation of ALDH2 also reduces ischemic damage to the
heart (Chen et al. 2008) and improves ventricular remod-
eling through detoxifcation of HNE (Gomes et al. 2014).
Similarly, activation of AR during late phase of ischemic
preconditioning diminishes the myocardial injury by metab-
olizing toxic aldehydes (Shinmura et al. 2002). Reports from
our laboratory and others have shown that the deletion of
enzymes that detoxify reactive aldehydes exacerbates car-
diac injury via excessive accumulation of lipid peroxidation
products. Genetic deletion of GSTP increases the levels of
acrolein and acrolein-protein adducts in the ischemic hearts
and sensitizes the hearts to I/R injury (Conklin et al. 2015).
Similarly, deletion of AR or ALDH2 enhances accumula-
tion of HNE protein adducts in the cardiac tissue, promotes
pathological remodeling (Baba et al. 2018) and aggravates
doxorubicin-induced myocardial dysfunction, respectively
(Sun et al. 2014a). However, even though the pathological
role of reactive aldehydes and the cardioprotective role of
aldehyde-metabolizing enzymes have been established, gene
therapy with these enzymes to protect myocardium against
aldehyde induced injury is complicated and not a viable
therapeutic option, and therapy with small cardioprotec-
tive peptides may be more feasible. Indeed, our previous
work has shown that perfusion of isolated hearts with car-
nosine promotes post-ischemic recovery (Baba et al. 2013),
suggesting that carnosine could be a therapeutically viable
option for preventing myocardial injury.
of diferent histidyl dipeptides against HNE and acrolein.
Although, several assays have been used to measure cytotox-
icity in cardiac myocytes, our previous work has shown that
measurements of hypercontracture are refective of calcium
overload secondary to sodium infux (Conklin et al. 2015).
This assay is therefore a robust and sensitive indicator of
myocyte injury induced by oxidants (Castro and Bhatnagar
1993). Before inducing hypercontracture, aldehydes such as
HNE alter the inactivation properties of the sodium chan-
nel (Castro and Bhatnagar 1993), which leads to sodium
infux and a resultant increase in intracellular calcium via the
sodium–calcium exchanger (Conklin et al. 2015). Although
the specifc molecular events of HNE and acrolein toxicity
were not reexamined in this study, we believe that carnosine
prevents HNE and acrolein-induced hypercontracture, not by
interfering with molecular events of injury, but by directly
quenching these aldehydes, and thereby diminishing their
efective cytotoxic concentrations. Our results showed that
the cardiac myocytes pretreated with carnosine imparted
protection, whereas pretreatment with balenine, which had
preserved metal chelating property, is taken up by cardio-
myocytes to the similar extent as carnosine, but had dimin-
ished aldehyde binding capacity, lacked the ability to pro-
tect cardiac myocytes from aldehyde induced toxicity. These
observations defne more clearly, the role of carnosine as an
efective aldehyde sequestering agent. Further, the D-analog
of trimethylated carnosine, which is suggested to be resist-
ant (Orioli et al. 2011) exhibits enhanced bioavailablity,
gives the leverage to test the aldehyde quenching ability of
these dipeptides, lacked the ability to scavenge reactive alde-
hydes, was unable to protect the aldehyde induced cell death.
These results further validate our hypothesis, that aldehyde
sequestering by these dipeptides could be one of the major
routes for cardioprotection. Nevertheless, in future, both
in vivo and ex vivo experiments are required to determine
the relative efcacy of carnosine and its methylated analogs
to determine whether the protective efects of carnosine are
due to direct binding to lipid peroxidation products.
In summary, this is the frst report to compare the alde-
hyde quenching and metal chelating efcacy of naturally
occurring histidyl dipeptides balenine and synthetic dipep-
tide DMB with carnosine. By synthesizing and studying the
analogues in which the Nτ of histidine was methylated (bale-
nine) or the Nτ of histidine and amino group of β-alanine
were methylated (DMB), we could examine, for the frst
time, the role of Nτ of imidazole ring in aldehyde binding
and metal chelating, and further defne the adduction chem-
istry of carnosine. Our results showing that balenine, which
was unable to form a conjugate with HNE or react avidly
with acrolein, suggest that Nτ of histidine is necessary to
complete the conjugation reaction with bulkier aldehydes
and to maintain high reactivity with small unsaturated alde-
hydes. Thus, both balenine and DMB represent a new series
Carnosine is a multifaceted molecule that could scavenge
ROS, quench singlet oxygen, sequester reactive aldehydes,
chelate metals and bufer intracellular pH (Boldyrev et al.
2013). Thus, during ischemia, it could preserve myocardial
integrity by targeting multiple mediators of cardiac injury.
However, our results showing that carnosine increased
the survival of cardiac myocytes superfused with reactive
aldehydes in comparison with histidyl dipeptides that have
diminished ability to bind these aldehydes, suggest that
direct sequestering of aldehydes by carnosine might be one
of the underlying mechanisms to protect against myocardial
injury.
In our experiment, we used isolated adult cardiac myo-
cytes, which enabled us to individually test the efcacy
1 3

J. Zhao et al.
a marker of oxidative stress in hypertrophy development. Am
J Physiol Heart Circ Physiol 287(5):H2122–2131. https://doi.
org/10.1152/ajpheart.00378.2004
of natural and synthetic carnosine analogues that could be
used to investigate the biological relevance of carnosine,
especially under conditions where reactive aldehydes or
metal toxicity is suspected to be an important mediator of
pathological stress. Our data showing that the quenching of
reactive aldehydes by carnosine in myocytes lay down the
foundation for investigating the role of enzyme ATPGD1
or carnosine in clinical outcomes such as acute myocardial
infarction or heart failure.
Benedetti A, Fulceri R, Ferrali M, Ciccoli L, Esterbauer H, Comporti
M (1982) Detection of carbonyl functions in phospholipids of
liver microsomes in CCl4- and BrCCl3-poisoned rats. Biochim
Biophys Acta 712(3):628–638
Benedetti A, Comporti M, Fulceri R, Esterbauer H (1984) Cytotoxic
aldehydes originating from the peroxidation of liver microsomal
lipids. Identifcation of 4,5-dihydroxydecenal. Biochim Biophys
Acta 792(2):172–181
Bhatnagar A (1995) Electrophysiological efects of 4-hydroxynon-
enal, an aldehydic product of lipid peroxidation, on isolated rat
ventricular myocytes. Circ Res 76(2):293–304
Acknowledgements We would like to thank Bioanalytical Core of the
Diabetes and Obesity Center for biochemical analysis.
Blancquaert L, Baba SP, Kwiatkowski S, Stautemas J, Stegen S,
Barbaresi S, Chung W, Boakye AA, Hoetker JD, Bhatnagar A,
Delanghe J, Vanheel B, Veiga-da-Cunha M, Derave W, Ever-
aert I (2016) Carnosine and anserine homeostasis in skeletal
muscle and heart is controlled by beta-alanine transamination.
J Physiol. https://doi.org/10.1113/jp272050
Funding This work was supported by grants from the National
Institutes of Health, R01HL122581-01 (SPB), R01HL55477 and
GM103492 (AB).
Boldyrev AA, Aldini G, Derave W (2013) Physiology and patho-
physiology of carnosine. Physiol Rev 93(4):1803–1845. https
://doi.org/10.1152/physrev.00039.2012
Compliance with ethical standards
Conflict of Interest All authors declare that no competing fnancial
Brown CE, Antholine WE, Froncisz W (1980) Multiple forms of
the copper(II)-carnosine complex. J Chem Soc Dalton Trans
4:590–596
interest exists.
Ethical approval All treatments and protocols were approved by the
University of Louisville, Institutional Animal Care and Use Commit-
tee. The ethical approval number is 15387.
Brown CE, Vidrine DW, Julian RL, Froncisz W (1982) Copper (II)
dimers in solution: evidence for motional averaging of coupling
tensors without chemical dissociation. J Chem Soc Dalton Trans
12:2371–2377
Canabady-Rochelle LL, Harscoat-Schiavo C, Kessler V, Aymes
A, Fournier F, Girardet JM (2015) Determination of reduc-
ing power and metal chelating ability of antioxidant peptides:
revisited methods. Food Chem 183:129–135. https://doi.
org/10.1016/j.foodchem.2015.02.147
References
Abe H (2000) Role of histidine-related compounds as intracellular
proton bufering constituents in vertebrate muscle. Biochemistry
(Mosc) 65(7):757–765
Carini M, Aldini G, Beretta G, Arlandini E, Facino RM (2003)
Acrolein-sequestering ability of endogenous dipeptides: char-
acterization of carnosine and homocarnosine/acrolein adducts
by electrospray ionization tandem mass spectrometry. J Mass
Spectrom 38(9):996–1006. https://doi.org/10.1002/jms.517
Castro GJ, Bhatnagar A (1993) Efect of extracellular ions and
modulators of calcium transport on survival of tert-butyl
hydroperoxide exposed cardiac myocytes. Cardiovasc Res
27(10):1873–1881
Aldini G, Carini M, Beretta G, Bradamante S, Facino RM (2002) Car-
nosine is a quencher of 4-hydroxy-nonenal: through what mecha-
nism of reaction? Biochem Biophys Res Commun 298(5):699–706
Baba SP, Hoetker JD, Merchant M, Klein JB, Cai J, Barski OA,
Conklin DJ, Bhatnagar A (2013) Role of aldose reductase in the
metabolism and detoxifcation of carnosine–acrolein conjugates.
J Biol Chem 288(39):28163–28179. https://doi.org/10.1074/jbc.
M113.504753
Chen CH, Budas GR, Churchill EN, Disatnik MH, Hurley TD,
Mochly-Rosen D (2008) Activation of aldehyde dehydro-
genase-2 reduces ischemic damage to the heart. Science
321(5895):1493–1495. https://doi.org/10.1126/science.1158554
Conklin DJ, Guo Y, Jagatheesan G, Kilfoil PJ, Haberzettl P, Hill
BG, Baba SP, Guo L, Wetzelberger K, Obal D, Rokosh DG,
Prough RA, Prabhu SD, Velayutham M, Zweier JL, Hoetker JD,
Riggs DW, Srivastava S, Bolli R, Bhatnagar A (2015) Genetic
defciency of glutathione S-Transferase P increases myocardial
sensitivity to ischemia-reperfusion injury. Circ Res 117(5):437–
449. https://doi.org/10.1161/CIRCRESAHA.114.305518
de Courten B, Jakubova M, de Courten MP, Kukurova IJ, Vallova
S, Krumpolec P, Valkovic L, Kurdiova T, Garzon D, Barba-
resi S, Teede HJ, Derave W, Krssak M, Aldini G, Ukropec J,
Ukropcova B (2016) Efects of carnosine supplementation on
glucose metabolism: pilot clinical trial. Obesity (Silver Spring)
24(5):1027–1034. https://doi.org/10.1002/oby.21434
Decker EA, Crum AD, Calvert JT (1992) Diferences in the antioxi-
dant mechanism of carnosine in the presence of copper and iron.
J Agric Food Chem 40(5):756–759
Baba SP, Zhang D, Singh M, Dassanayaka S, Xie Z, Jagatheesan G,
Zhao J, Schmidtke VK, Brittian KR, Merchant ML, Conklin DJ,
Jones SP, Bhatnagar A (2018) Defciency of aldose reductase
exacerbates early pressure overload-induced cardiac dysfunction
and autophagy in mice. J Mol Cell Cardiol 118:183–192. https://
doi.org/10.1016/j.yjmcc.2018.04.002
Baran EJ (2000) Metal complexes of carnosine. Biochemistry (Mosc)
65(7):789–797
Barski OA, Xie Z, Baba SP, Sithu SD, Agarwal A, Cai J, Bhatnagar A,
Srivastava S (2013) Dietary carnosine prevents early atheroscle-
rotic lesion formation in apolipoprotein E-null mice. Arterioscler
Thromb Vasc Biol 33(6):1162–1170. https://doi.org/10.1161/
ATVBAHA.112.300572
Benderdour M, Charron G, DeBlois D, Comte B, Des Rosiers C (2003)
Cardiac mitochondrial NADP+-isocitrate dehydrogenase is inac-
tivated through 4-hydroxynonenal adduct formation: an event that
precedes hypertrophy development. J Biol Chem 278(46):45154–
45159. https://doi.org/10.1074/jbc.M306285200
Benderdour M, Charron G, Comte B, Ayoub R, Beaudry D, Foisy S,
Deblois D, Des Rosiers C (2004) Decreased cardiac mitochon-
drial NADP+-isocitrate dehydrogenase activity and expression:
1 3

Carnosine protects cardiac myocytes against lipid peroxidation products
to acrolein. Toxicol Appl Pharmacol 324:51–60. https://doi.
org/10.1016/j.taap.2016.10.008
Dobbie H, Kermack WO (1955) Complex-formation between polypep-
tides and metals. 2. The reaction between cupric ions and some
dipeptides. Biochem J 59(2):246–257
Lenz GR, Martell AE (1964) Metal complexes of carnosine. Biochem-
istry 3:750–753
Drozak J, Veiga-da-Cunha M, Vertommen D, Stroobant V, Van Schaft-
ingen E (2010) Molecular identifcation of carnosine synthase as
ATP-grasp domain-containing protein 1 (ATPGD1). J Biol Chem
285(13):9346–9356. https://doi.org/10.1074/jbc.M109.095505
Endo J, Sano M, Katayama T, Hishiki T, Shinmura K, Morizane
S, Matsuhashi T, Katsumata Y, Zhang Y, Ito H, Nagahata Y,
Marchitti S, Nishimaki K, Wolf AM, Nakanishi H, Hattori F,
Vasiliou V, Adachi T, Ohsawa I, Taguchi R, Hirabayashi Y, Ohta
S, Suematsu M, Ogawa S, Fukuda K (2009) Metabolic remodeling
induced by mitochondrial aldehyde stress stimulates tolerance to
oxidative stress in the heart. Circ Res 105(11):1118–1127. https
://doi.org/10.1161/CIRCRESAHA.109.206607
Liu YH, Carretero OA, Cingolani OH, Liao TD, Sun Y, Xu J, Li LY,
Pagano PJ, Yang JJ, Yang XP (2005) Role of inducible nitric
oxide synthase in cardiac function and remodeling in mice with
heart failure due to myocardial infarction. Am J Physiol Heart
Circ Physiol 289(6):H2616–2623. https://doi.org/10.1152/ajphe
art.00546.2005
Moreau R, Heath SH, Doneanu CE, Lindsay JG, Hagen TM (2003)
Age-related increase in 4-hydroxynonenal adduction to rat heart
alpha-ketoglutarate dehydrogenase does not cause loss of its cata-
lytic activity. Antioxid Redox Signal 5(5):517–527. https://doi.
org/10.1089/152308603770310167
Niki E (2009) Lipid peroxidation: physiological levels and dual bio-
logical efects. Free Radic Biol Med 47(5):469–484. https://doi.
org/10.1016/j.freeradbiomed.2009.05.032
Esterbauer H, Schaur RJ, Zollner H (1991) Chemistry and biochemistry
of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free
Radic Biol Med 11(1):81–128
Okuma HAAE (1991) Efect of Temperature on the bufering capaci-
ties of histidine-related compounds and fsh skeletal muscle. Nip-
pon Suisan Gakkaishi 57(11):2101–2107
Fawaz MV, Topper ME, Firestine SM (2011) The ATP-grasp enzymes.
Bioorg Chem 39(5–6):185–191. https://doi.org/10.1016/j.bioor
g.2011.08.004
Orioli M, Vistoli G, Regazzoni L, Pedretti A, Lapolla A, Rossoni
G, Canevotti R, Gamberoni L, Previtali M, Carini M, Aldini G
(2011) Design, synthesis, ADME properties, and pharmacological
activities of beta-alanyl-d-histidine (d-carnosine) prodrugs with
improved bioavailability. Chem Med Chem 6(7):1269–1282. https
://doi.org/10.1002/cmdc.201100042
Freeman HC, Szymanski JT (1967) Crystallographic studies of metal-
peptide complexes. V. (Beta-alanyl-l-histidinato)copper(II)dihy-
drate. Acta Crystallogr 22(3):406–417
Giordano FJ (2005) Oxygen, oxidative stress, hypoxia, and heart fail-
ure. J Clin Invest 115(3):500–508. https://doi.org/10.1172/JCI24
408
Gomes KM, Campos JC, Bechara LR, Queliconi B, Lima VM, Disatnik
MH, Magno P, Chen CH, Brum PC, Kowaltowski AJ, Mochly-
Rosen D, Ferreira JC (2014) Aldehyde dehydrogenase 2 activa-
tion in heart failure restores mitochondrial function and improves
ventricular function and remodelling. Cardiovasc Res 103(4):498–
508. https://doi.org/10.1093/cvr/cvu125
Porter NA, Caldwell SE, Mills KA (1995) Mechanisms of free radical
oxidation of unsaturated lipids. Lipids 30(4):277–290
Qin F, Simeone M, Patel R (2007) Inhibition of NADPH oxidase
reduces myocardial oxidative stress and apoptosis and improves
cardiac function in heart failure after myocardial infarction. Free
Radic Biol Med 43(2):271–281. https://doi.org/10.1016/j.freer
adbiomed.2007.04.021
Halliwell BGJ (2015) Free radical in biology and medicine. Oxford
University Press, Oxford
Sansbury BE, DeMartino AM, Xie Z, Brooks AC, Brainard RE, Wat-
son LJ, DeFilippis AP, Cummins TD, Harbeson MA, Brittian KR,
Prabhu SD, Bhatnagar A, Jones SP, Hill BG (2014) Metabolomic
analysis of pressure-overloaded and infarcted mouse hearts. Circ
Heart Fail 7(4):634–642. https://doi.org/10.1161/CIRCHEARTF
AILURE.114.001151
Hennekens CH, Buring JE, Manson JE, Stampfer M, Rosner B, Cook
NR, Belanger C, LaMotte F, Gaziano JM, Ridker PM, Willett
W, Peto R (1996) Lack of efect of long-term supplementation
with beta carotene on the incidence of malignant neoplasms and
cardiovascular disease. N Engl J Med 334(18):1145–1149. https
://doi.org/10.1056/NEJM199605023341801
Sawyer DB, Siwik DA, Xiao L, Pimentel DR, Singh K, Colucci WS
(2002) Role of oxidative stress in myocardial hypertrophy and fail-
ure. J Mol Cell Cardiol 34(4):379–388. https://doi.org/10.1006/
jmcc.2002.1526
Hipkiss AR (2009) Carnosine and its possible roles in nutrition and
health. Adv Food Nutr Res 57:87–154. https://doi.org/10.1016/
S1043-4526(09)57003-9
Ishikawa T, Esterbauer H, Sies H (1986) Role of cardiac glutathione
transferase and of the glutathione S-conjugate export system in
biotransformation of 4-hydroxynonenal in the heart. J Biol Chem
261(4):1576–1581
Shinmura K, Bolli R, Liu SQ, Tang XL, Kodani E, Xuan YT, Sriv-
astava S, Bhatnagar A (2002) Aldose reductase is an obligatory
mediator of the late phase of ischemic preconditioning. Circ Res
91(3):240–246
Ismahil MA, Hamid T, Haberzettl P, Gu Y, Chandrasekar B, Srivas-
tava S, Bhatnagar A, Prabhu SD (2011) Chronic oral exposure to
the aldehyde pollutant acrolein induces dilated cardiomyopathy.
Am J Physiol Heart Circ Physiol 301(5):H2050–2060. https://doi.
org/10.1152/ajpheart.00120.2011
Srivastava S, Harter TM, Chandra A, Bhatnagar A, Srivastava SK, Pet-
rash JM (1998) Kinetic studies of FR-1, a growth factor-inducible
aldo-keto reductase. Biochemistry 37(37):12909–12917. https://
doi.org/10.1021/bi9804333
Srivastava S, Watowich SJ, Petrash JM, Srivastava SK, Bhatnagar A
(1999) Structural and kinetic determinants of aldehyde reduc-
tion by aldose reductase. Biochemistry 38(1):42–54. https://doi.
org/10.1021/bi981794l
Kato Y, Iwase M, Ichihara S, Kanazawa H, Hashimoto K, Noda A,
Nagata K, Koike Y, Yokota M (2010) Benefcial efects of growth
hormone-releasing peptide on myocardial oxidative stress and left
ventricular dysfunction in dilated cardiomyopathic hamsters. Circ
J 74(1):163–170
Srivastava S, Chandrasekar B, Gu Y, Luo J, Hamid T, Hill BG, Prabhu
SD (2007) Downregulation of CuZn-superoxide dismutase
contributes to beta-adrenergic receptor-mediated oxidative
stress in the heart. Cardiovasc Res 74(3):445–455. https://doi.
org/10.1016/j.cardiores.2007.02.016
Keith RJ, Haberzettl P, Vladykovskaya E, Hill BG, Kaiserova K,
Srivastava S, Barski O, Bhatnagar A (2009) Aldose reductase
decreases endoplasmic reticulum stress in ischemic hearts.
Chem Biol Interact 178(1–3):242–249. https://doi.org/10.1016/j.
cbi.2008.10.055
Sun A, Cheng Y, Zhang Y, Zhang Q, Wang S, Tian S, Zou Y, Hu
K, Ren J, Ge J (2014a) Aldehyde dehydrogenase 2 ameliorates
doxorubicin-induced myocardial dysfunction through detoxifca-
tion of 4-HNE and suppression of autophagy. J Mol Cell Cardiol
71:92–104. https://doi.org/10.1016/j.yjmcc.2014.01.002
Kurhanewicz N, McIntosh-Kastrinsky R, Tong H, Ledbetter A, Walsh
L, Farraj A, Hazari M (2017) TRPA1 mediates changes in heart
rate variability and cardiac mechanical function in mice exposed
1 3

J. Zhao et al.
Sun A, Zou Y, Wang P, Xu D, Gong H, Wang S, Qin Y, Zhang P, Chen
Y, Harada M, Isse T, Kawamoto T, Fan H, Yang P, Akazawa H,
Nagai T, Takano H, Ping P, Komuro I, Ge J (2014b) Mitochon-
drial aldehyde dehydrogenase 2 plays protective roles in heart
failure after myocardial infarction via suppression of the cytosolic
JNK/p53 pathway in mice. J Am Heart Assoc 3(5):e000779. https
://doi.org/10.1161/JAHA.113.000779
The Heart Outcomes Prevention Evaluation Study Investigators.
N Engl J Med 342(3):154–160. https://doi.org/10.1056/nejm2
00001203420302
Zhang P, Xu X, Hu X, van Deel ED, Zhu G, Chen Y (2007) Inducible
nitric oxide synthase defciency protects the heart from systolic
overload-induced ventricular hypertrophy and congestive heart
failure. Circ Res 100(7):1089–1098. https://doi.org/10.1161/01.
RES.0000264081.78659.45
van der Kraaij AM, de Jonge HR, Esterbauer H, de Vente J, Steinbusch
HW, Koster JF (1990) Cumene hydroperoxide, an agent induc-
ing lipid peroxidation, and 4-hydroxy-2,3-nonenal, a peroxida-
tion product, cause coronary vasodilatation in perfused rat hearts
by a cyclic nucleotide independent mechanism. Cardiovasc Res
24(2):144–150
Zhang P, Hou M, Li Y, Xu X, Barsoum M, Chen Y, Bache RJ (2009)
NADPH oxidase contributes to coronary endothelial dys-
function in the failing heart. Am J Physiol Heart Circ Physiol
296(3):H840–846. https://doi.org/10.1152/ajpheart.00519.2008
Yusuf S, Dagenais G, Pogue J, Bosch J, Sleight P (2000) Vitamin E
supplementation and cardiovascular events in high-risk patients.
1 3