NADP+-dependent dehydrogenase activity of carbonyl reductase on glutathionylhydroxynonanal as a new pathway for hydroxynonenal detoxification

Article abstract of DOI:10.1016/j.freeradbiomed.2015.02.001

An NADP⁺-dependent dehydrogenase activity on 3-glutathionyl-4-hydroxynonanal (GSHNE) was purified to electrophoretic homogeneity from a line of human astrocytoma cells (ADF). Proteomic analysis identified this enzymatic activity as associated with carbonyl reductase 1 (EC 1.1.1.184). The enzyme is highly efficient at catalyzing the oxidation of GSHNE (K_M 33 μM, k_cat 405 min^-1), as it is practically inactive toward trans-4-hydroxy-2-nonenal (HNE) and other HNE-adducted thiol-containing amino acid derivatives. Combined mass spectrometry and nuclear magnetic resonance spectroscopy analysis of the reaction products revealed that carbonyl reductase oxidizes the hydroxyl group of GSHNE in its hemiacetal form, with the formation of the corresponding 3-glutathionylnonanoic-δ-lactone. The relevance of this new reaction catalyzed by carbonyl reductase 1 is discussed in terms of HNE detoxification and the recovery of reducing power.

Full text of DOI:10.1016/j.freeradbiomed.2015.02.001

Author's Accepted Manuscript
Nadpþ-Dependent Dehydrogenase Activity Of
Carbonyl
Reductase
On
Glutathionyl-
Hydroxynonanal As A New Pathway For
Hydroxynonenal Detoxification
Roberta Moschini, Eleonora Peroni, Rossella
Rotondo, Giovanni Renzone, Dominique Melck,
Mario Cappiello, Massimo Srebot, Elio Napoli-
tano, Andrea Motta, Andrea Scaloni, Umberto
Mura, Antonella Del-Corso
www.elsevier.com/locate/freerad-
biomed
PII:
S0891-5849(15)00041-6
http://dx.doi.org/10.1016/j.freeradbiomed.2015.02.001
FRB12302
DOI:
Reference:
To appear in: Free Radical Biology and Medicine
Received date: 13 November 2014
Revised date: 15 January 2015
Accepted date: 2 February 2015
Cite this article as: Roberta Moschini, Eleonora Peroni, Rossella Rotondo,
Giovanni Renzone, Dominique Melck, Mario Cappiello, Massimo Srebot, Elio
Napolitano, Andrea Motta, Andrea Scaloni, Umberto Mura, Antonella Del-
Corso, Nadpþ-Dependent Dehydrogenase Activity Of Carbonyl Reductase On
Glutathionyl-Hydroxynonanal As A New Pathway For Hydroxynonenal
Detoxification, Free Radical Biology and Medicine, http://dx.doi.org/10.1016/j.
freeradbiomed.2015.02.001
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal
pertain.

NADP+-DEPENDENT DEHYDROGENASE ACTIVITY OF CARBONYL REDUCTASE
ON GLUTATHIONYL-HYDROXYNONANAL AS
HYDROXYNONENAL DETOXIFICATION.
A
NEW PATHWAY FOR
Roberta Moschinia, Eleonora Peronia#, Rossella Rotondoa#, Giovanni Renzoneb, Dominique Melckc,
Mario Cappielloa, Massimo Srebotd, Elio Napolitanoe, Andrea Mottae, Andrea Scalonib, Umberto
Muraa and Antonella Del-Corsoa*.
aUniversity of Pisa, Department of Biology, Biochemistry Unit, via S. Zeno, 51, 56123, Pisa, Italy,
bProteomics & Mass Spectrometry Laboratory, ISPAAM-CNR, via Argine 1085, I-80147, Napoli,
Italy.
cInstitute of Biomolecular Chemistry, ICB-CNR, Comprensorio Olivetti, Edificio A, Via Campi
Flegrei 34, I-80078 Pozzuoli (Naples), Italy.
dHealth Unit 5 Pisa, Pontedera Hospital, Ginecology and Obstetric Unit, Via Roma 180, 56025
Pontedera, Italy.
eScuola Normale Superiore, Piazza dei Cavalieri, Pisa, Italy.
# These authors contributed equally to the work.
*Corresponding author: Antonella Del-Corso, University of Pisa, Department of Biology,
Biochemistry Unit, via S. Zeno, 51, 56123, Pisa, Italy.
Phone: +39.050.2211454; Fax: +39.050.2211460
E-mail: antonella.delcorso@unipi.it

Abstract
A NADP+ dependent dehydrogenase activity on 3-glutathionyl-4-hydroxynonanal (GSHNE) was
purified to electrophoretic homogeneity from a line of human astrocytoma cells (ADF). Proteomic
analysis identified this enzymatic activity as associated with carbonyl reductase 1 (E.C. 1.1.1.184).
The enzyme is highly efficient at catalyzing the oxidation of GSHNE (KM33 µM,kcat.405 min-1), as
it is practically inactive towards trans-4-hydroxy-2-nonenal (HNE) and other HNE-addicted thiol-
containing amino acid derivatives. Combined mass spectrometry and nuclear magnetic resonance
spectroscopy analysis of the reaction products revealed that carbonyl reductase oxidizes the
hydroxyl group of GSHNE in its hemiacetal form, with the formation of the corresponding 3-
glutathionyl-nonanoic-ꢀ-lactone. The relevance of this new reaction catalyzed by carbonyl
reductase 1 is discussed in terms of HNE detoxification and the recovery of reducing power.
Keywords: 4-hydroxy-2-nonenal; 3-glutathionyl-4-hydroxynonanal; carbonyl reductase;
hydroxynonenal detoxification;

Introduction
The pathophysiological effects of oxidative stress on cells and tissues are assumed to be related, at
least in part, to endogenous lipid peroxidation and to the highly reactive molecules consequently
produced, of which the most abundant is trans-4-hydroxy-2-nonenal (HNE) [1-3]. HNE is
generated as a racemic mixture of 4R- and 4S- enantiomers primarily from the peroxidation of 2-6
polyunsatured fatty acids [4].It may account for more than 95% of the enals produced [5,6]. Basal
HNE concentrations in tissue and plasma range from 0.8 to 2.8 µM [7], and there is evidence of
increased levels up to 4.5 mM in peroxidizing membranes under oxidative stress conditions [8].
HNE plays an important role in the pathogenesis of several diseases, including atherosclerosis and
Alzheimer's disease [6,9,10]. The toxicity of HNE depends on its high chemical reactivity, which is
associated with the presence of an aldehydic group, a double bond and a secondary alcohol at the
chiral centre C4 [11]. HNE can react with nucleophiles via 1,2 and 1,4 addition [7,12,13]. The main
protein modification associated with HNE occurs via 1,4-Michael addition with Cys, Lys and His
residues [7,12,14]. A number of proteins, including albumin, amyloid beta peptide, alpha-synuclein,
aldose reductase and several glutathione S-transferase (GST) isoforms, have been reported to be
covalently modified in vitro by HNE [15-25]. On the other hand, the 1,2-addition of the carbonyl
group to the primary amine of Lys has also been reported, which generates the corresponding Schiff
base [13,24].
As a signaling molecule, HNE is involved in modulating in a concentration-dependent manner
several cellular processes, such as proliferation, differentiation and apoptosis. High concentrations
of HNE have thus been reported to cause cell cycle arrest [26], differentiation [27]and apoptosis
[28,29].Lower concentrations, at least in some cell types, appear to induce proliferation [30].
Intracellular control of HNE levels is thus required in order to modulate the cell cycle.
HNE metabolism involves reducing aldehyde to the corresponding 1,4 dihydroxynonene (DHN) by
aldose reductase [31,32] and alcohol dehydrogenase [33,34] or oxidating aldehyde to the 4-

hydroxynonenoic acid (HNA) by aldehyde dehydrogenase [35-37]. However, the formation of the
3-glutathionyl-4-hydroxynonanal (GSHNE) is considered as the main pathway of the HNE
metabolism [38-40]. GSHNE, which is present in solution essentially (95%) in its cyclic hemiacetal
form [7,41], can be produced both in a spontaneous reaction between HNE and glutathione (GSH),
and in a reaction efficiently catalyzed by GST [7,42]. Of the isoforms of GST in mammalian
tissues, a subgroup of the ꢁ-class isozymes shows the highest catalytic efficiency for HNE [43-46].
As GSHNE is the most significant HNE derivative, its metabolism represents the main pathway for
the control of HNE levels. Depletion of intracellular GSH by buthionine sulfoximine or by
oxidative insult reduces GSHNE levels with a consequent increase in free HNE amounts and in
cellular toxicity [47,48]. However, since GSHNE is an inhibitor of GST activity, the adduct needs
to be removed from the cell by extrusion and/or intracellular metabolism [49,50].
As with HNE, GSHNE is susceptible to both oxidative and reductive transformations. Thus, aldose
reductase reduces GSHNE to 1,4-dihydroxynonane-glutathione (GSDHN) [32], which has been
shown to be involved in the proliferation and inflammation cell signaling cascades [51-53]. On the
other hand, GSHNE can be oxidized by NAD-dependent aldehyde dehydrogenase to 4-
hydroxynonanoic acid-glutathione (GSHNA) [54] or its lactone form [55]. These oxidized
derivatives can be further metabolized by cytochrome P450 4A, with the subsequent production of
2-hydroxylated and carboxylated metabolites [38,55-57]. Finally, the glutathione moiety of these
adducts can be transformed into mercapturic acid in the kidney and the resulting metabolites are
excreted into urine [58-60].
The present study on human astrocytoma ADF cells highlights a highly efficient and specific
dehydrogenase activity of carbonyl reductase 1 (CBR1) on GSHNE but not on HNE, , which was
never recognized and reported before. This novel activity of CBR1 is here discussed in terms of
detoxification and recovery of cell reducing power.

Materials and Methods
Materials
GSH, Cys, CysGly, "GluCys, BSA, SDS, 5,5’-dithiobis- (2-nitrobenzoic acid) (DTNB), trans-2-
nonenal, NADP+, NADPH, 9,10-PQ, and sequencing-grade trypsinwere purchased from Sigma
Aldrich (St. Louis, MO, USA). 4-Oxo-nonenal (ONE), prostaglandin B1 (PGB1) and 4-hydroxy-2-
nonenal mercapturic acid were purchased from Cayman Chemicals (Ann Arbor, MI, USA)
Whatman DEAE-cellulose (DE-52) and Sephacryl S200 were purchased from GE Healthcare (Little
Chalfont, UK). Blue Sepharose and Bradford reagent were purchased from Bio-Rad (Hercules, CA,
USA). YM10 membranes (10kDa cut-off) were purchased from Amicon Millipore (Darmstadt,
Germany). RPMI 1640, penicillin, streptomycin, glutamine and foetal bovine serum (FBS) were
obtained from Lonza (Basel, Switzerland). Dialysis tubing (10kDa cut-off) was purchased from
Spectrum Laboratories Inc (Rancho Dominguez, CA, USA). All inorganic chemicals were of
reagent grade from BDH (VWR international ltd, Poole Dorset, UK). All solvents were HPLC-
grade from J.T. Baker Chemicals (USA).
Cell cultures
ADF cells were kindly provided by Dr. W. Malorni, Istituto Superiore di Sanità, Rome, Italy.ADF
cells were cultured on 10 cm diameter dishes in RPMI1640 medium supplemented with 10% (v/v)
fetal bovine serum, 50 mU/mL penicillin/streptomycin and 2 mM glutamine at 37 °C, in a
humidified 5% CO2 atmosphere. Cell plates were washed with a phosphate buffer solution (PBS)
(0.08% w/v NaCl, 0.002% w/v KCl, 0.002% w/v KH2PO4, 0.006% Na2HPO4), cells were harvested
with a scraper and stored as suspensions in PBS (8 x 107 cells/mL) at -80°C until use.
Assay of carbonyl reductase

The activity of CBR1 was determined at 37°C using GSHNE as the substrate to be oxidized. The
reaction was monitored following the increase in absorbance at 340 nm due to the reduction of
NADP+(ꢂ340 = 6.22 mM-1.cm-1). The reaction mixture contained 0.1 mM GSHNE, 0.18 mM NADP+
in 50 mM sodium phosphate buffer pH 8.4. One unit of enzyme activity is the amount that catalyzes
the conversion of 1 µmol of substrate/min in the above assay conditions.
The dehydrogenase activity of CBR1 was evaluated as described above on different substrates at the
reported concentrations (see Results). In the case of PGB1, 2% ethanol was present in the assay
mixture. The NADPH-dependent reductase activity of CBR1was determined in the conditions
described above on different substrates at the reported concentrations, following the decrease in
absorbance at 340 nm due to the oxidation of 0.18 mM NADPH. In the case of 9,10-PQ and
menadione, the assay mixture contained DMSO at a fixed final concentration of 2% and 1.25%
(v/v), respectively. DMSO and ethanol (2% final concentration) had no effect on the activity of
CBR1, as measured in standard conditions. KM and kcat values were determined by linear regression
analysis of kinetic data of double reciprocal plots using GraphPad software. The kcat values were
calculated on the basis of a MW of 31 kDa.
Purification of the GSHNE NADP+ dependent dehydrogenase activity.
A GSHNE-NADP+-dependent dehydrogenase (GSHNE-DH) activity was purified from ADF cells.
All procedures were carried out at 4°C. Cell lysates were obtained through a freezing and thawing
protocol, followed by a centrifugation at 10,000 x g for 30 min. The supernatant was diluted 1:1
with4 mM DTT in10 mM sodium phosphate pH 7, and referred to as the “crude extract “.The latter
was applied onto a DE-52 column (2.5x 9 cm) which was eluted with2 mM DTT in 10 mM sodium
phosphate pH 7 buffer (SB) at a flow rate of 17 mL/h, collecting 3 mL fractions. SB containing 0.2
M NaCl was then applied. Fractions displaying activity were pooled and concentrated using an
Amicon YM10 membrane. Thus, the concentrated protein solution was applied on a Sephacryl
S200 column (1.6 x 80 cm) which was eluted with SB containing 0.1 M NaCl, at a flow rate of 20

mL/h, collecting 1.7 mL fractions. The active fractions were pooled and applied on a Blue-
Sepharose column (1.2 x 8 cm), equilibrated with SB at 10 mL/h; column elution was performed
with SB supplemented with 0.1 mM NADP+ and NaCl at different concentrations (see details on
Supplementary Materials) collecting 1.5 mL fractions. Fractions displaying GSHNE-DH activity
were pooled, concentrated to a protein concentration of 0.01 mg/mL using an Amicon YM10
ultrafiltration membrane, and stored at 4°C, until used.
Synthesis of aldehydes.
Diethyl acetals of HNE and 4-hydroxy-2-noninal were prepared according to a previous procedure
[61]. 4-hydroxy-2-nonanal diethyl acetal was prepared by catalytic hydrogenation (PtO in ethyl
acetate) of 4-hydroxy-2-noninal diethyl acetal [62]. Free aldehydes were prepared by acid
hydrolysis (pH 3.0) of the diethylacetals for 1 h, at 4°C.The concentration of HNE and trans-2-
nonenal was determined by measuring the absorbance at 224 nm using an extinction coefficient of
13.7 mM-1.cm-1 [61].
Preparation of thiol-adducts.
GSHNE, CysHNE, CysGlyHNE, and "GluCysHNE were prepared by incubating 1mM HNE in the
presence of 1.5mM of each thiol containing compound in 50 mM sodium phosphate buffer pH 7.4,
at 37 °C, for 1h, followed by an overnight incubation at 4 °C. The reaction was monitored by
following the consumption of both HNE, by checking the absorbance at 224 nm, and of the
corresponding reduced thiols by the Ellman assay [63]. Taking into account the spontaneous
oxidation of the SH moiety, a stoichiometric consumption of thiols and HNE was observed in all
cases. In addition, the residual HNE accounted for no more than 10% of the initial value. GS-
nonanal was prepared by incubating trans-2-nonenal with GSH in the same conditions described
above. In all cases, resulting derivatives were stored at -80°C, until used.

Proteomic identification of the GSHNE NADP+ dependent dehydrogenase activity.
A protein sample exhibiting GSHNE-DH activity was analyzed by SDS-PAGE; protein bands
present within the gel were excised, minced, in gel-reduced with DTT, S-alkylated with
iodoacetamide, and digested with trypsin [64]. Resulting peptide mixtures were desalted with )Zip-
TipC18 (Millipore) devices, and analyzed for protein identification by nLC-ESI-LIT-MS/MS, using
an LTQ XL mass spectrometer (Thermo, USA) equipped with a Proxeon nanospray source
connected to an Easy-nanoLC (Thermo, USA) [65].Peptides were separated on an Easy C18 column
(100 mm × 0.075 mm, 3 )m) (Thermo, USA);mobile phases were 0.1% v/v formic acid (solvent A)
and 0.1% v/v formic acid in acetonitrile (solvent B), running at total flow rate of 300 nL/min.
Linear gradient was initiated 20 min after sample loading; solvent B ramped from 5% to 35% over
45 min, from 35% to 60% over 10 min, and from 60% to 95% over 20 min. Spectra were acquired
in the range m/z 400-2000. Peptide samples were analyzed under collision-induced dissociation-
MS/MS data-dependent product ion scanning procedure, enabling dynamic exclusion (repeat count
1 and exclusion duration 60 s) over the three most abundant ions. Mass isolation window and
collision energy were set to m/z 3 and 35%, respectively.
Raw data from nLC-ESI-LIT-MS/MS analysis were searched with a MASCOT search engine
(version 2.2.06, Matrix Science, U.K.) against an updated UniProtKB non-redundant (2014/04/16)
database. Database searching was performed by using a mass tolerance value of 2.0 Da for
precursor ion and 0.8 Da for MS/MS fragments, trypsin as proteolytic enzyme, a missed cleavage
maximum value of 2, and Cys carbamidomethylation and Met oxidation as fixed and variable
modifications, respectively. Other MASCOT parameters were kept as default. Candidates with at
least 2 unique assigned peptides with an individual peptide expectation value <0.05 (corresponding
to a confidence level for peptide identification
>95%) were considered confidently identified.
Definitive peptide assignment was always associated with manual spectra verification.

Mass spectrometry analysisof the reaction products deriving from the CBR1-catalyzed oxidation of
GSHNE.
A sample containing 102 µM GSHNE, 180 µM NADP+, 62 µM GSH, 8.9 µM HNE in 30 mM
phosphate buffer, pH 8.4, was incubated with 3.6 mU/mL CBR1, at 37 °C. The extent of reaction
was monitored on a time-course basis by sampling the reaction mixture at different reaction times.
Samples 1 to 4 were withdrawn at 0, 10, 40 and 80 min of incubation, respectively; sample 5
corresponded to 75 min of further incubation of the mixture after supplementation with additional
enzyme (2.5 mU/mL). In parallel, identical samples containing isolated components in 30 mM
phosphate buffer, pH 8.4, i.e. 102 µM GSHNE, 180 µM NADP+, 62 µM GSH, 8.9 µM HNE, 3.6
mU/mL CBR1, or mixtures of that were incubated at 37 °C, as reported above. After sampling,
samples were immediately frozen in dry ice, lyophilized and stored at -80 °C, until used. Samples
were then solubilized in 0.1% formic acid and analyzed by nLC-ESI-LIT-MS/MS, as described
above. In this case, solvent B ramped from 5% to 35% over 10 min, from 35% to 95% over 2 min.
Raw data from nLC-ESI-LIT-MS/MS analyses were manually interpreted and assigned to specific
glutathione derivatives.
A semi-quantitative measurement of the reaction products was obtained by analyzing the same
samples with nLC-ESI-LIT-MS as described above, with the unique exception that
chromatographic runs were acquired only in MS scan mode, without ion fragmentation [66, 67]. In
this case, extraction and integration of the LC-MS peak areas corresponding to the reagent GSHNE
and to the observed reaction products were performed in the same total ion chromatogram; the
Genesis algorithm in the Xcalibur software (Thermo, USA) was used to this purpose. Since
different ionization properties were expected for the different molecules, a measurement of their
relative amount was obtained by extracting and integrating peak areas corresponding to these
compounds and to Met-Arg-Phe-Ala (Sigma, USA); the latter peptide was used as internal reference
and spiked in all samples at the same concentration before MS analysis. Semi-quantitative
evaluation of the modification extent was then determined by calculating the ratio: Peak Areaspecies of

interest/Peak Areainternal reference peptide. This procedure was applied to all samples reported above by
performing experiments in quintuplicate.
NMR spectroscopy measurements.
For acquisition of NMR spectra, reaction samples reported in the previous section were dissolved in
1
2
30 mM sodium phosphate buffer, pH 8.4, and95% H2O, 5% H2O (Cortecnet, Voisins-Le-
Bretonneux, France). All spectra were recorded on a 600 MHz Bruker Avance III spectrometer
(Bruker BioSpin GmbH, Rheinstetten, Germany) equipped with a CryoProbe™. One-dimensional
1
(1D) H-NMR spectra were collected at 300 K with the excitation sculpting pulse sequence [68] to
suppress the water resonance. We used a double-pulsed field gradient echo, with a soft square pulse
of 4 ms at the water resonance frequency, and the gradient pulses of 1 ms each in duration, adding
128 transients of 64k complex points, with an acquisition time of 4 s/transient. Time-domain data
were all zero-filled to 128k complex points, and prior to Fourier transformation, an exponential
multiplication of 0.6 Hz was applied. Two-dimensional (2D) clean total correlation spectroscopy
(TOCSY) spectra [69] were recorded using a standard pulse sequence and incorporating the
excitation sculpting sequence for water suppression. In general, 360 equally spaced evolution-time
period t1 values were acquired, averaging 16 transients of 2048 points with 6024 Hz of spectral
width. Time-domain data matrices were all zero-filled to 4096 points in both dimensions, and prior
to Fourier transformation a Lorentz-to-Gauss window was applied for both t1 and t2 dimensions for
all the experiments. Both homonuclear 1D and 2D spectra were referred to 0.10 mM TSP, assumed
to resonate at ꢀ = 0.00 ppm in both dimensions.
For the natural abundance of 2D 1H-13C Heteronuclear Single Quantum Coherence (HSQC) spectra,
we used an echo-antiecho phase sensitive pulse sequence using adiabatic pulses for
decoupling[70].One hundred-twenty eight equally spaced evolution time period t1 values were
acquired, averaging 64 transients of 2048 points and using GARP4 for decoupling. The final data
matrix was zero-filled to 4096 in both dimensions, and apodized before Fourier transformation by a

shifted cosine window function in t2 and in t1. Linear prediction was also applied to extend the data
to twice their length in t1. Spectra were referred to internal TSP, assumed to resonate at ꢀ = 0.00
ppm in both dimensions.
Other methods.
Protein determination was performed according to Bradford [71]. SDS-PAGE was performed
according to Laemmli [72]; gels were stained by the silver stain technique [73]. The activity of
aldose reductase, glucose-6-phosphate dehydrogenase, isocitrate dehydrogenase and 6-P gluconate
dehydrogenase were determined at 37°C, according to [74-77], respectively.
RESULTS
Enzyme Purification and characterization. A marked NADP+ reduction was observed when ADF
cell crude extracts were incubated in the presence of GSHNE. This dehydrogenase
activity(GSHNE-DH) was purified approximately 250-fold with an overall yield of 26%. The
purification procedure included ionic exchange, exclusion and affinity chromatography. During the
purification, attention was focused on discriminating between known NADP(H) oxidoreductase
activities and GSHNE-DH activity. The ionic exchange step was able to separate both glucose 6-
phosphate dehydrogenase and isocitric dehydrogenase from GSHNE-DH activity. The gel filtration
chromatographic step separated the 6-phosphogluconate dehydrogenase from the GSHNE-DH
which was significantly smaller in size (approximately 30 kDa). Finally, GSHNE-DH was separated
from aldose reductase by the affinity chromatography step, giving rise to a pure GSHNE-DH
preparation characterized by a protein doublet in SDS-PAGE, with an apparent molecular weight of
31-32 kDa (see the Supplementary Figures 1S-4S for chromatographic elution profiles and SDS-
PAGE). The pure enzyme preparation (11 U/mg protein) was stable for at least two weeks in SB
containing 0.1 mM NADP+ and 1.5 M NaCl, at 4°C. Before kinetic analysis, the enzyme was

subjected to an overnight dialysis on dialysis tubing (cut off 10 kDa) against 10 mM sodium
phosphate pH 7.0.
Both bands present in SDS-PAGE of the GSHNE-DH preparation (Supplementary Figure 4S) were
subjected to independent proteomic analyses. Both gave the same identification results, proving
their nature as 15-hydroxyprostaglandin dehydrogenase (E.C. 1.1.1.184), also referred to as
carbonyl reductase 1 (CBR1). nLC-ESI-LIT-MS/MS mapping experiments provided identical
results for both bands, with the exception of a peptide component uniquely observed in the case of
the high mass-migrating protein component, which was associated with a CBR1 form containing a
carboxyethyl mojety at Lys238 (see the Supplementary Fig. 5S for the MS/MS spectrum of the
modified peptide). This component was absent in the low mass-migrating protein component, which
conversely showed the presence of the corresponding non-modified counterpart (data not shown).
NADP(H) dependent activity of CBR1fromADF cells. The NADPH-dependent reductase activity
of the purified CBR1 was tested on a number of molecules, including classical substrates of this
enzyme (Table 1). Of these, only 9,10-PQ, menadione and ONE, in the order of decreasing
effectiveness, were reduced by the enzyme, while different alkenals and their glutathionyl
derivatives (including GSHNE) were completely unaffected. The kinetic parameters evaluated for
the NADPH-dependent reduction of 9,10-PQ (1.8 0.2 µM and 101 4 min-1, for KM and kcat,
respectively, Fig. 1) were in accordance with the literature data reported for human CBR1 [78].
Also the NADP+-dependent dehydrogenase activity of the enzyme was tested on a number of
molecules (Table 2), including PGB1, a distinctive substrate for the dehydrogenase action of CBR1
[79] and GSHNE, which was used in this study to monitor enzyme purification. Of the tested
substrates, only GSHNE and PGB1 were transformed. No effect on the enzyme activity was
observed when either N-acetyl cysteine or buthionine sulfoximine were present in the assay mixture
at the final concentration of 0.5 mM.

The reaction was strictly dependent on NADP+ since NAD+ was completely ineffective as a
cofactor for all the tested substrates. The same pattern of dehydrogenase activity towards pyridine
cofactors and different substrates was observed with a recombinant CBR1 obtained from RNA
isolated from ADF cells. It is worth noting that the cDNA sequence obtained for the ADF enzyme
completely matched with the human wild type sequence.
NADP+-dependent GSHNE-DH activity of CBR1. A comparative kinetic analysis between
GSHNE and PGB1 revealed that the former was a far better substrate for the enzyme than the one
considered as its distinctive substrate (Fig. 2). In the case of GSHNE oxidation, a KM of 33 2 µM
and a kcat of 405 7 min-1 were measured; for PGB1 a KM of 91 8 µM and a kcat of 60 2 min-1
were obtained. Thus, the specificity constant (ks) for GSHNE (12.3 x 103 0.7 x 103 mM-1 min-1)
was approximately 18 fold higher than for PGB1 (0.7 x103 0.1 x 103 mM-1 min-1). In addition,
while a 40% decrease in activity was observed with PGB1 when the pH of the reaction buffer was
decreased from 8.4 to 7.4, no changes in activity (a less than 10% decrease) were observed when
GSHNE was used as substrate in the pH range 8.4-6.4.
The effectiveness of NADP+ as a cofactor for the GSHNE oxidation is reported in Fig. 3. CBR1
displayed a KM value for the cofactor of 11 2µM and a kcat value of 367 17 min-1, which
accounted for a ks value equal to 33.4 x 103 5.9 x 103 mM-1min-1. Figure 3 also reports the effect
of both NAD+ and NADPH on the GSHNE oxidation. There was an apparent competitive inhibitory
effect of NADPH on the dehydrogenase reaction (Ki value approximately equal to 6 µM). In
addition, NAD+, which cannot be used as a cofactor by the enzyme, failed to interfere with the
NADP+ dependent dehydrogenase reaction.
Identification of the reaction products by nLC-ESI-LIT-MS/MS. Analysis of the reaction products
of the GSHNE oxidation catalyzed by CBR1 was performed on samples taken from a
transformation time curve of the substrate at 37°C, as deriving from a mixture containing 102 µM

GSHNE, 180 µM NADP+ and 3.6 mU/mL of purified CBR1in 30 mM sodium phosphate pH 8.4.
The mixture also contained GSH and HNE (62 µM and 8.9 µM, respectively) as residual reagents
of the GSHNE preparation; both compounds had been previously verified not to affect the enzyme
activity. The increase in absorbance at 340 nm (linked to NADP+ reduction) was monitored;
samples at 0, 10, 40 and 80 min of incubation (samples 1 to 4, respectively) were withdrawn as
reported in the experimental section. Under these conditions, the extent of NADP+ reduction
accounted for approximately 40 µM, which increased to 52 µM after 75 min of further incubation
of the mixture after supplementation with additional 2.5 mU/mL of the enzyme (sample 5). All
samples were then analyzed by nLC-ESI-LIT-MS/MS and NMR.
nLC-ESI-LIT-MS/MS analysis of standard compounds and the reaction mixtures mentioned above
(samples 1 to 5) identified initial reagents and glutathione derivatives that progressively appeared
after the addition of CBR1. A number of peaks showing the same mass value were observed in the
nLC-ESI-LIT-MS/MS profile that, considering the occurrence of various chiral centers in these
HNE-addicted glutathione derivatives, were associated with pairs of unresolved diastereomers (see
the Supplementary Fig. 6S for the total ion current and ion-extracted chromatograms of sample 4),
as already reported in the case of GSHNE [80,81] and other HNE-thiol adducts [82,83]. The most
representative peaks referred to GSHNE (exp. and theor. MH+ values at m/z464.18 and 464.21,
respectively) (Fig. 4A), which was originally present in the mixture before CBR1 addition and
decreased during the incubation with the enzyme (Fig. 5), and to glutathione derivatives showing an
experimental MH+ signal at about m/z462.23, namely glutathionyl-4-ketononanal or glutathionyl-4-
hydroxynonanoic acid (GSHNA) ꢀ-lactone (theor. MH+ value at m/z462.19) (Fig. 4B), which
increased on a time-course basis (Fig. 5). Less significant was the generation of GSHNA in its
acyclic form (exp. and theor. MH+ values at m/z480.25 and 480.20, respectively) (Fig. 4C), and
even less the generation of the double-centered oxidative product glutathionyl-4-ketononanoic acid
(exp. and theor. MH+ values at m/z478.19 and 478.19, respectively) (Fig. 5). Semi-quantitative
measurements of the reaction mixtures assumed that all HNE-addicted glutathione derivatives

reported in this study might present the same ionization tendency. According to the nature of the
products observed during time-course nLC-ESI-LIT-MS/MS analysis of the incubation mixtures, a
reaction pathway catalyzed by CBR1 was proposed (Fig. 6).
Characterization of the reaction products by NMR analysis. The nature of the initial reagents and
of the resulting product(s) was finally demonstrated by NMR analysis. We firstly analyzed single
products and their mixtures without CBR1. Resonances were assigned using 2D homo- and
heteronuclear experiments. In the GSHNE/GSH/HNE and GSHNE-GSH-HNE-NADP+ spectra
(Fig. 7), the envelope centered at 5.57 ppm was assigned to 1-H of GSHNE and was consistent with
the occurrence of the cyclic hemiacetal form of HNE adducts [81] under the experimental
conditions we used. Four peaks (I at 5.59 ppm, II at 5.53 ppm, III at 5.50 ppm, and IV at 5.47 ppm)
were identified. They stem from the four major diastereomers of the spontaneous reaction of
racemic HNE with GSH. Considering the chiral centers, there are eight potential GSHNE
diastereomers in total, and it is possible that each of the four peaks constitute a pair of unresolved
diastereomers [81]. The remaining proton signals of GSHNE were also identified and are indicated
in Fig. 7 (middle and upper spectra).
Analysis of the low field region of the reaction time course in phosphate buffer (Fig. 8) indicated
that, in the presence of CBR1, peaks I-IV originating from proton 1 of GSHNE started to be
modified after 10 min, with all NADPH peaks appearing and the corresponding reduction of
NADP+ signals. After 80 min, peaks II, III and IV almost disappeared, while peak I was still
partially present, with the doublet at 5.50 ppm originating from NADPH. After 155 min (top
spectrum),peak I was barely visible. Analysis of the high-field region between 2.5 and 2.3 ppm
clarified the presence of the GSHNA δ-lactone (Fig. 9). In fact, formation of lactone requires
disappearance of proton 1(Fig.6); this can be followed in the evolution of the GSHNE signals
labeled 2 in Fig. 9. The ongoing reaction from 10 to 155 min indicated that the multiplicity of

proton 2 changed from a multiplet (at 10 min) to become a double doublet (at 80 min). This
multiplicity was only due to the coupling with proton 3, while coupling with proton 1 was removed
by the lactone formation. The combined characterization through MS and NMR approaches of the
initial reagent, in its hemiacetal form, and of the product(s) resulting as result of the CBR1-
catalyzed process clarified the reaction pathway as mainly yielding glutathionyl-4-hydroxynonanoic
acid ꢀ-lactone.
DISCUSSION
An assessment of the rational basis for the multiplicity of the effects exerted by HNE [26-30]entails
defining all its metabolic pathways and understanding their importance. Although various enzymes
are known to directly intervene on HNE, the main step of its detoxification (both in the reductive
and in the oxidative direction) appears to be mediated by the conjugation with GSH [3], which can
occur spontaneously or catalyzed by glutathione transferase (GST) isoforms [7,42].
Since GSHNE is the most abundant HNE derivative, its metabolism represents the main pathway
for the control of HNE levels. In this context, GSHNE can be reduced to GS-DHN by aldose
reductase or oxidized to GSHNA and GSHNA lactone by NAD-dependent aldehyde dehydrogenase
[54,55]. The data reported in this study add a novel element to this scenario, and demonstrate that
GSHNE can also be efficiently oxidized to GSHNA lactone by carbonyl reductase 1 (CBR1), a
well-represented NADP(H)-dependent enzyme in various human cells [78,85]. This enzyme has
been already reported to catalyze the NADPH-dependent reduction of different endogenous and
xenobiotic carbonyl compounds [86], including 4-oxononenal (ONE) and its glutathione adduct
(GS-ONE) [87], thus playing an active role against oxidative stress, neurodegeneration and
apoptosis [88-92]. However, except for the glutathionyl-NO adduct, endogenous substrate(s) of
CBR1 have not been successfully identified [93].
The protocol optimized for the purification of this GSHNE NADP+ dependent dehydrogenase
activity from ADF cells led to the isolation of two protein species, migrating with apparent mass of

31-32 kDa in SDS-PAGE (see Supplementary Fig. 4S). Proteomic analysis identified both proteins
as human CBR1, also known as NADP+ 15-hydroxyprostaglandine dehydrogenase, and proved
their migration differences due to the occurrence of a carboxyethyl mojety at Lys238 (see
Supplementary Fig. 5S).This modification has been already described for the human enzyme and
has been demonstrated to have no effect on the protein activity and specificity [94,95].
The enzyme isolated from ADF cells acts on classical substrates of CBR1, as it is able to reduce
9,10-PQ, menadione and ONE (Table 1, Fig. 1) and also to oxidize PGB1 (Table 2). A kinetic
characterization of the ADF GSHNE-DH activity using both PGB1 and GSHNE as substrates
(Fig.2) revealed a significantly higher efficiency of the enzyme for GSHNE, for which a specificity
constant was found (12.3 x 103 mM-1 min-1) that was approximately 18-fold higher than PGB1. The
reduction of prostaglandin A1 by CBR1 has already been reported to be positively affected by the
presence of a glutathionyl moiety on the molecule [85], which may explain the marked difference
reported here between GSHNE and PGB1 as substrates to be oxidized. In any case, GSHNE was
readily and efficiently oxidized by CRB1 in a wide range of pH values (6.5-8.4).
The time-course evolution of the oxidative products of GSHNE was revealed by a combined MS
and NMR approach (Figures 4, 5, 7, 8 and 9). The problem of attributing the most abundant
reaction product with MH+signal at m/z462.23 (Fig. 4B) at the end of the reaction to either
glutathionyl-4-ketononanal or glutathionyl-4-hydroxynonanoic acid ꢀ-lactone was solved by NMR,
which indicated the latter as the main oxidation derivative. In this context, time-course analysis of
peaks I-IV in NMR spectra (Fig. 8) suggests that II-IV disappeared at the end of the incubation,
while peak I was still visible. This could indicate that three diastereomers react faster than the one
corresponding to peak I. According to the chemical shift values previously reported by Balogh et al.
[81], the lowest-field signal resonating at 5.59 ppm by analogy should correspond to 3S,4R-
GSHNE. Accordingly, we can speculate that the latter species does not properly fit into the CBR1
active site. A similar condition of diastereoisomer-oriented GSHNE oxidation was also observed
during nLC-ESI-LIT-MS/MS analysis of the reaction products (see Supplementary Fig. 6S).

To our knowledge, only one study has reported evidences of a NADP+-dependent oxidation of
GSHNE as catalyzed by recombinant ALDH3A1 [54]. This enzyme, however, is mainly active on
HNE and, only to a lesser extent, on GSHNE. On the other hand, the specificity displayed by CRB1
towards GSHNE with respect to HNE is worth noting, according to the data reported in this study,
as the latter compound is not processed at all by the enzyme. The identification of GSHNA-ꢀ-
lactone as the main reaction product gives a first conceivable rationale basis for the selective
recognition of GSHNE by CBR1. Product analysis, in fact, indicates the hydroxyl group of the
hemiacetal form of the glutathionyl-derivative as a reactive center. This hydroxyl group is derived
from a structural rearrangement that HNE fails to undergo because of the trans disposition of C4-
hydroxyl and C1-carbonyl groups. However, the occurrence of a hemiacetal hydroxyl group seems
not a sufficient requirement for enzyme action, since 4-hydroxynonanal, a molecule competent for
cyclic hemiacetal formation, was not recognized as a substrate (Table 2). These results confirmed
the importance of the glutathionyl-moiety in tailoring good substrates for CBR1 [85], as it occurs
for GSHNE. The occurrence of structural restrictions exerted by the glutathionyl moiety present on
substrates is underlined by the complete lack of action of CBR1 on other HNE-addicted thiol-
containing amino acid and dipeptide substrates, which were tested for the enzyme dehydrogenase
activity (Table 2). Furthermore, the inability of the enzyme to oxidize GS-nonanal (Table 2)
indicates that CBR1 does not recognize a substrate containing a free carbonyl moiety, but the
occurrence of a hemiacetal structure is also required.
The formation of GSHNA lactone was reported also to occur when GSHNE was incubated with
cytosolic fractions of murine liver and kidney [55]. On the basis of an experimental inhibition of
GSHNE oxidation by disulfiram, the oxidative metabolism of GSHNE was associated with a NAD+
dependent aldehyde dehydrogenase activity [55]. However, this observation was misleading since
disulfiram is not a selective inhibitor of ALDHs, but also inhibits CBR1 [96]. Indeed, we have
measured a 32% inhibition of the purified ADF GSHNE DH activity in the presence of 20 µM
disulfiram, when the assay was performed in standard assay conditions (data not shown). In

addition, in dialyzed rat liver extracts GSHNE appears to be oxidized through both a NAD+ and a
NADP+-dependent reaction (R. Moschini and R. Rotondo, 2014, unpublished results). Finally, a
NADP+ dehydrogenase activity on GSHNE has recently been found also in human placental, bovine
kidney and rat testis dialyzed crude extracts (R. Moschini and R. Rotondo, 2014, unpublished
results).
As generally occurs with other oxidoreductases, CBR1 efficiently and specifically recognizes
NADP+ as a cofactor with a KM of 11 µM (Fig. 3). However, the apparent inhibitory effect exerted
by NADPH on GSHNE oxidation, especially at the NADP+/NADPH ratios present in vivo, raises
the question on the functional involvement of the GSHNE-DH activity of CBR1 in the GSHNE
antioxidant metabolic pathways. In fact, the low NADP+/NADPH ratio values generally present in
normal healthy cells would suggest that CBR1 does not act as a dehydrogenase. However, this
condition does not concur in cancer cells, where high (or very high) NADP+/NADPH ratios have
generally been measured. For example, PC3 human prostate cancer cells present a NADP+/NADPH
value of 2.2, which reaches the apex of 6.5 under glucose deprivation conditions [97]. Similarly,
breast and colon cancer cells show NADP+/NADPH ratios equal to 5 and 1, respectively, which
dramatically increase under glucose deprivation conditions in which NADPH became practically
undetectable [98]. In addition, mutated forms of both the mitochondrial and cytosolic isoforms of
NADP-dependent isocitrate dehydrogenase, diffusely occurring in glioma and acute myeloid
leukaemia cells, channel the ꢁ–ketoglutarate (formed with the concomitant generation of NADPH)
towards the formation of 2-hydroxyglutarate with NADPH consumption [99].
All these evidences support our proposal for the possible entrance of CBR1 in the enzyme pattern
(i.e. the pentose phosphate pathway enzymes, isocitrate and malate dehydrogenases) aimed at
rebuilding the cell-reducing conditions through the NADPH recovery. This role, which seems to
take place in cancer cells, may also more generally apply to biological systems undergoing
oxidative stress, where lipid peroxidation takes place. In this context, CBR1 should play a dual role
in HNE detoxification and in the recovery of the cellular reducing power. In conclusion, the

efficient and specific NADP+-dependent oxidation of GSHNE to GSHNA-ꢀ-lactone catalyzed by
CBR1, as shown in this study, suggests another way in HNE breakdown, thus leading to the
possibility of identifying new strategies for the modulation of the HNE metabolism.
Acknowledgements. This work was supported in part by MIUR, PRIN Project 2009M8FKBB_003
and in part by the University of Pisa.
References
[1] Parola, M.; Bellomo, G.; Robino, G.; Barrera, G.; Dianzani, M.U. 4-Hydroxynonenal as a
biological signal: molecular basis and pathophysiological implications. Antioxi. Redox Signal. 1:
255-284; 1999.
[2] Dianzani, M.U. 4-Hydroxynonenal from pathology to physiology. Mol. Aspects Med. 24: 263-
272; 2003.
[3] Dalleau, S.; Baradat, M.; Guéraud, F.; Huc, L. Cell death and diseases related to oxidative
stress: 4-hydroxynonenal in the balance. Cell Death Differ. 20: 1615-1630; 2013.
[4] Schneider, C.; Tallman, K.A.; Porter, N.A.; Brash, A.R. Two distinct pathways of formation of
4-hydroxynonenal. Mechanism of nonenzymatic transformation of the 9- and 13- hydroperoxides of
linoleic acid to 4-hydroxyalkenals. J. Biol. Chem. 276: 20831-20838; 2001.
[5] Benedetti, A.; Comporti, M.; Esterbauer, H. Identification of 4-hydroxynonenal as a cytotoxic
product originating from the peroxidation of liver microsomal lipids. Biochim. Biophys. Acta 620:
281-296; 1980.
[6] Uchida, K. 4-Hydroxy-2-nonenal: a product and mediator of oxidative stress. Prog. Lipid Res.
42: 318-343; 2003.
[7] Esterbauer, H.; Schaur, R.J.; Zollner, H. Chemistry and biochemistry of 4-hydroxynonenal,
malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11: 81-128; 1991.
[8] Benedetti, A.; Comporti, M.; Fulceri, R.; Esterbauer, H. Cytotoxic aldehydes originating from
the peroxidation of liver microsomal lipids. Identification of 4,5-dihydroxydecenal. Biochim.
Biophys. Acta 792: 172-181; 1984.
[9] Liu, Q.; Raina, A.K.; Smith, M.A.; Sayre, L.M.; Perry, G. Hydroxynonenal, toxic carbonyls, and
Alzheimer disease. Mol. Aspects Med. 24: 305-313; 2003.

[10] Picklo, M.J.; Montine, T.J.; Amarnath, V.; Neely, M.D. Carbonyl toxicology and Azheimer’s
disease. Toxicol. Appl. Pharmacol. 184: 187-197; 2002.
[11] Schaur, R.J. Basic aspects of the biochemical reactivity of 4-hydroxynonenal. Mol. Aspects
Med. 24: 149-159; 2003.
[12] Nadkarni, D.V.; Sayre, L.M. Structural definition of early lysine and histidine adduction
chemistry of 4-hydroxynonenal. Chem. Res. Toxicol. 8: 284-291; 1995.
[13] Wade, L.G., Jr. Organic chemistry, 5th ed. Upper Saddle River, NJ: Prentice-Hall; 2003: 1042-
1045.
[14] Doorn, J.A.; Petersen, D.R. Covalent modification of amino acid nucleophiles by the lipid
peroxidation products 4-hydroxy-2-nonenal and 4-oxo-2-nonenal. Chem. Res. Toxicol. 15: 1445-
1450; 2002.
[15] Aldini, G.; Gamberoni, L.; Orioli, M.; Beretta, G.; Regazzoni, L.; Maffei Facino, R.; Carini,
M. Mass spectrometric characterization of covalent modification of human serum albumin by 4-
hydroxy-trans-2-nonenal. J. Mass. Spectrom. 41: 1149-1161; 2006.
[16] Szapacs, M.E.; Riggins, J.N.; Zimmerman, L.J.; Liebler, D.C. Covalent adduction of human
serum albumin by 4-hydroxy-2-nonenal: kinetic analysis of competing alkylation reactions.
Biochemistry 45: 10521-10528; 2006.
[17] Magni, F.; Galbusera, C.; Tremolada, L.; Ferrarese, C.; Kienle, M.G. Characterisation of
adducts of the lipid peroxidation product 4-hydroxy-2-nonenal and amyloid beta-peptides by liquid
chromatography/electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 16:
1485-1493; 2002.
[18] Siegel, S.J.; Bieschke, J.; Powers, E.T.; Kelly, J.W. The oxidative stress metabolite 4-
hydroxynonenal promotes Alzheimer protofibril formation. Biochemistry 46: 1503-1510; 2007.
[19] Qin, Z.; Hu, D.; Han, S.; Reaney, S.H.; Di Monte, D.A.; Fink, A.L. Effect of 4-hydroxy-2-
nonenal modification on alpha-synuclein aggregation. J.Biol.Chem. 282: 5862-5870; 2006.
[20] Del Corso, A.; Dal Monte, M.; Vilardo, P.G.; Cecconi, I.; Moschini, R.; Banditelli, S.;
Cappiello, M.; Tsai, L.; Mura, U. Site-specific inactivation of aldose reductase by 4-
hydroxynonenal. Arch. Biochem. Biophys.350: 245-248; 1998.
[21] Mitchell, A.E.; Morin, D.; Lamé, M.W.; Jones, A.D. Purification, mass spectrometric
characterization, and covalent modification of murine glutathione S-transferases. Chem. Res.
Toxicol. 8: 1054-1062; 1995.
[22] van Iersel, M.L.; Ploemen, J.P.; Lo Bello, M.; Federici, G.; van Bladeren, P.J. Interactions of
alpha, beta-unsatured aldehydes and ketones with human glutathione S-transferase P1-1. Chem.
Biol. Interact. 108: 67-78; 1997.

[23] Shireman, L.M.; Kripps, K.A.; Balogh, L.M.; Conner, K.P.; Whittington, D.; Atkins, W.M.
Glutatione transferase A4-4 resists adduction by 4-hydroxynonenal. Arch. Biochem. Biophys. 504:
182-189; 2010.
[24] Bachi, A.; Dalle Donne, A.; Scaloni, A. Redox proteomics chemical principles, methodological
approaches and biological/biomedical promises. Chem. Rev. 113: 596-698; 2013.
[25] Dalle-Donne I.; Scaloni, A.; Giustarini, D.; Cavarra, E.; Tell, G.; Lungarella, G.; Colombo, R.;
Rossi, R.; Milzani, A. Proteins as biomarkers of oxidative/nitrosative stress in diseases: the
contribution of redox proteomics. Mass Spectrom Rev. 24: 55-99; 2005.
ꢀ
[26] Wonisch, W.; Kohlwein, S.D.; Schaur J.; Tatzber, F.; Guttenberger, H.; Zarkovic, N.; Winklen,
R. Treatment of the budding yeast Saccharomyces cerevisiae with the lipid peroxidation product 4-
HNE provokes a temporary cell cycle arrest in G1 phase. Free Rad. Biol. Med. 25: 682-687; 1998.
[27] Cheng, J.Z.; Singhal, S.S.; Saini, M.; Singhal, J.; Piper, J.T.; Van KuijK, F.J.; Zimniak, P.;
Awasthi, Y.C.; Awasthi, S. Effects of mGST A4 transfection on 4-hydroxynonenal-mediated
apoptosis and differentiation of K562 human erythroleukemia cells. Arch. Biochem. Biophys. 372:
29-36; 1999.
[28] Camandola, S.; Poli, G.; Mattson, M.P. The lipid peroxidation product 4-hydroxy-2,3-nonenal
increases AP-1-binding activity through caspase activation in neurons. J. Neurochem. 74: 159-168;
2000.
[29]Ji, C.; Amarnath, V.; Pietenpol, J.A.; Marnett, L.J. 4-Hydroxynonenal induces apoptosis via
caspase-3 activation and cytochrome c release. Chem. Res. Toxicol. 14: 1090-1096; 2001.
[30] Ruef, J.; Rao, G.N.; Li, F.; Bode, C.; Patterson, C.; Bhatnagar, A.; Runge, M.S. Induction of
rat aortic smooth muscle cell growth by the lipid peroxidation product 4-hydroxy-2-nonenal.
Circulation 97:1071-1078¸1998.
[31] Vander Jagt, D.L.; Kolb, N.S.; Vander Jagt, T.J.; Chino, J.; Martinez, F.J.; Hunsaker, L.A.;
Royer, R.E. Substrate specificity of human aldose reductase: identification of 4-hydroxynonenal as
an endogenous substrate. Biochim. Biophys. Acta 1249; 117-126; 1995.
[32] Srivastava, S.; Chandra, A.; Bhatnagar, A.; Srivastava, S.K.; Ansari, N.H. Lipid peroxidation
product, 4-hydroxynonenal and its conjugate with GSH are excellent substrates of bovine lens
aldose reductase. Biochem. Biophys. Res. Commun. 217: 741-746; 1995.
[33] Canuto, R.A.; Muzio, G.; Maggiora, M.; Poli, G.; Biasi, F.; Dianzani, M.U.; Ferro, M.; Bassi,
A.M.; Penco, S.; Marinari, U.M. Ability of different hepatoma cells to metabolize 4-
hydroxynonenal. Cell Biochem. Funct. 11: 79-86; 1993.

[34] Hartley, D.P.; Ruth, J.A.; Petersen, D.R. The hepatocellular metabolism of 4-hydroxynonenal
by alcohol dehydrogenase, aldehyde dehydrogenase, and glutatione S-transferase. Arch. Biochem.
Biophys. 316: 197-205.
[35] Pappa, A.; Estey, T.; Manzer, R.; Brown, D.; Vasiliou, V. Human aldehyde dehydrogenase
3A1 (ALDH3A1): biochemical characterization and immunohistochemical localization in the
cornea. Biochem. J. 376: 615-623: 2003.
[36] Murphy, T.C.; Amarnath, V.; Gibson, K.M.; Picklo, M.J. Sr. Oxidation of 4-hydroxy-2-
nonenal by succinic semialdehyde dehydrogenase (ALDH5A). J. Neurochem. 86: 298-305; 2003.
[37] Kong, D.; Kotraiah, V. Modulation of aldehyde dehydrogenase activity affects ( )-4-hydroxy-
2E-nonenal (HNE) toxicity and HNE-protein adducts levels in PC12 cells. J. Mol. Neurosci. 47:
595-603; 2012.
[38] Alary, J.; Guéraud, F.; Cravedi, J.P. Fate of 4-hydroxynonenal in vivo: disposition and
metabolic pathways. Mol. Aspects Med. 24: 177-187; 2003.
[39] Awasthi, Y.C.; Ansari, G.A.; Awasthi, S. Regulation of 4-hydroxynonenal mediated signaling
by glutathione S-transferases. Methods Enzymol. 401: 379-407; 2005.
[40] Boon, P.J.; Marinho, H.S.; Oosting, R.; Mulder, G.J. Glutathione conjugation of 4-hydroxy-
trans-2,3-nonenal in the rat in vivo, the isolated perfused liver and erythrocytes. Toxicol. Appl.
Pharmacol. 159: 214-223; 1999.
[41] Esterbauer, H.; Zollner, H.; Scholz, N. Reaction of glutathione with conjugated carbonyls. Z.
Naturforsch C. 30: 466-473; 1975.
[42] Guéraud, F.; Atalay, M.; Bresgen, N.; Cipak, A.; Eckl, P.M.; Huc, L.; Jouanin, I.; Siems, W.;
Uchida, K. Chemistry and biochemistry of lipid peroxidation products. Free Radic. Res. 44: 1098-
1124; 2010.
[43] Zimniak, P.; Singhal, S.S.; Srivastava, S.K.; Awasthi, S.; Hayden, J.B.; Awasthi, Y.C.
Estimation of genomic complexity, heterologous expression, and enzymatic characterization of
mouse glutathione S-transferase mGSTA4-4 (GST 5.7). J. Biol. Chem. 269: 992-1000; 1994.
[44] Singhal, S.S.; Awasthi, S.; Srivastava. S.K.; Zimniak, P.; Ansari, N.H.; Awasthi, Y.C. Novel
human ocular glutathione S-transferases with high activity toward 4-hydroxynonenal. Invest.
Ophathalmol. Vis. Sci. 36: 142-150; 1995.
[45] Hubatsch, I.; Ridderström, M.; Mannervik, B. Human glutathione transferase A4-4: an alpha
class enzyme with high catalytic efficiency in the conjugation of 4-hydroxynonenal and other
genotoxic products of lipid peroxidation. Biochem. J. 330: 175-179; 1998.

[46] Balogh, L.M.; Le Trong, I.; Kripps, K.A.; Shireman, L.M.; Stenkamp, R.E.; Zhang, W.;
Mannervik, B.; Atkins, W.M. Substrate specificity combined with stereopromiscuity in glutathione
transferase A4-4-dependent metabolism of 4-hydroxynonenal. Biochemistry 49: 1541-1548; 2010.
[47] Yadav, U.C.; Ramana, K.V.; Awasthi, Y.C.; Srivastava, S.K. Glutathione level regulates HNE-
induced genotoxicity in human erythroleukemia cells. Toxicol. Appl. Pharmacol. 227: 257-264;
2008.
[48] Aldini, G.; Granata, P.; Marinello, C.; Beretta, G.; Carini, M.; Facin, R.M. Effects of UVB
radiation on 4-hydroxy-2-trans-nonenal metabolism and toxicity in human keratinocytes. Chem.
Res. Toxicol. 20: 416-423; 2007.
[49] Ishikawa, T.; Esterbauer, H.; Sies, H. 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: 1576-1581; 1986.
[50] Sharma, R.; Yang, Y.; Sharma, A.; Dwivedi, P.; Popov, V.L.; Boor, P.J.; Singhal, S.S.;
Awasthi, S.; Awasthi, Y.C. Mechanisms and physiological significance of the transport of the
glutathione conjugate of 4-hydroxynonenal in human lens ephitelial cells. Invest. Ophathalmol. Vis.
Sci. 44: 3438-3449; 2003.
[51] Ramana, K.V.; Bhatnagar, A.; Srivastava, S.; Yadav, U.C.; Awasthi, S.; Awasthi, Y.C.;
Srivastava, S.K. Mitogenic responses of vascular smooth muscle cells to lipid peroxidation-derived
aldehyde 4-hydroxy-trans-2-nonenal (HNE): role of aldose reductase-catalyzed reduction of the
HNE-glutathione conjugates in regulating cell growth. J. Biol. Chem. 281: 17652-17660; 2006.
[52] Srivastava, S.; Ramana, K.V.; Bhatnagar, A.; Srivastava, S.K. Synthesis, quantification,
characterization, and signaling properties of glutathionyl conjugates of enals. Methods Enzymol.
474: 297-313; 2010.
[53] Ramana, K.V.; Fadl, A.A.; Tammali, R.; Reddy, A.B.; Chopra, A.K.; Srivastava, S.K. Aldose
reductase mediates the lipopolysaccharide-induced release of inflammatory mediators in
RAW264.7 murine macrophages. J. Biol. Chem. 281: 33019-33029; 2006.
[54] Black, W.; Chen, Y.; Matsumoto, A.; Thompson, D.C.; Lassen, N.; Pappa, A.; Vasiliou, V.
Molecular mechanisms of ALDH3A1-mediated cellular protection against 4-hydroxy-2-nonenal.
Free Rad. Biol. Med. 52: 1937-1944; 2012.
[55] Alary, J.; Fernandez, Y.; Debrauwer, L.; Perdu, E.; Guéraud, F. Identification of intermediate
pathways of 4-hydroxynonenal metabolism in the rat. Chem. Res. Toxicol. 16: 320-327; 2003.
[56] Alary, J.; Debrauwer, L.; Fernandez, Y.; Paris, A:; Cravedi, J.P.; Dolo, L.; Rao, D.; Bories, G.
Identification of novel urinary metabolites of the lipid peroxidation product 4-hydroxy-2-nonenal in
rats. Chem. Res. Toxicol. 11: 1368-1376; 1998.

[57] Guéraud, F.; Alary, J.; Costet, P.; Debrauwer, L.; Dolo, L.; Pineau, T.; Paris, A. In vivo
involvement of cytochrome P450 4A family in the oxidative metabolism of the lipid peroxidation
product trans-4-hydroxy-2-nonenal, using PPARalpha-deficient mice. J. Lipid. Res. 40: 152-159;
1999.
[58] Alary, J.; Bravais, F.; Cravedi, J.P.; Debrauwer, L.; Rao, D.; Bories, G. Mercapturic acid
conjugates as urinary end metabolites of the lipid peroxidation product 4-hydroxy-2-nonenal in the
rat. Chem. Res. Toxicol. 8: 34-39; 1995.
[59] de Zwart, L.L.; Hermanns, R.C.; Meerman, J.H.; Commandeur, J.N.; Vermeulen, N.P.
Disposition in rat of [2-3H]-trans-4-hydroxy-2,3-nonenal a product of lipid peroxidation.
Xenobiotica 26: 1087-1100; 1996.
[60] Stevens, J.F.; Maier, C.S. Acrolein: sources, metabolism, and biomolecular interactions
relevant to human health and disease. Mol. Nutr. Food Res. 52: 7-25; 2008.
[61] Esterbauer, H.; Weger, W. Über die wirkungen yon aldehyden auf gesunde und maligne zellen.
3. Mitt. Monatshefte für Chemie 98: 1995-2000; 1967.
[62] 5ori4, I.; Müller, S.; List, B. Kinetic resolution of homoaldols via catalytic asymmetric
transacetalization. J. Am. Chem. Soc. 132: 17370-17373; 2010.
[63] Ellman, G.L. Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82: 70-77; 1959.
[64] Talamo, F.; D'Ambrosio, C.; Arena, S.; Del Vecchio, P.; Ledda, L.; Zehender, G.; Ferrara, L.;
Scaloni, A. Proteins from bovine tissues and biological fluids: defining a reference electrophoresis
map for liver, kidney, muscle, plasma and red blood cells. Proteomics 3: 440-460; 2003.
ꢀ
[65] Salzano, A.M.; Novi, G.; Arioli, S.; Corona, S.; Mora, D.;Scaloni, A. Mono-dimensional blue
native-PAGE and bi-dimensional blue native/urea-PAGE or/SDS-PAGE combined with nLC-ESI-
LIT-MS/MS unveil membrane protein heteromeric and homomeric complexes in Streptococcus
thermophilus.
ꢀJ. Proteomics 94:240-261; 2013.
[66] Salzano, A.M. ; Renzone, G.; Scaloni, A.; Torreggiani, A.; Ferreri, C.; Chatgilialoglu,
C.Human serum albumin modifications associated with reductive radical stress.
889-898; 2011.
ꢀMol. Biosyst. 7:
[67] Lirussi,L.; Antoniali, G.; Vascotto, C.; D'Ambrosio, C.; Poletto, M.; Romanello, M.; Marasco,
D.; Leone, M.; Quadrifoglio, F.; Bhakat, K.K.; Scaloni, A.; Tell, G. Nucleolar accumulation of
APE1 depends on charged lysine residues that undergo acetylation upon genotoxic stress and
modulate its BER activity in cells.Mol Biol Cell. 23:4079-4096; 2012.
[68] Hwang, T.L.; Shaka, A.J. Water suppression that works: excitation sculpting using arbitrary
waveforms and pulse field gradients. J. Magn. Reson. 112: 275-279; 1995.

[69] Griesinger, C.; Otting, G.; Wüthrich, K.; Ernst, R.R. Clean TOCSY for proton spin system
identification in macromolecules. J. Am. Chem. Soc. 110: 7870-7872; 1988.
[70] Kay, L. E.; Keifer, P.; Saarinen, T. Pure absorption gradient enhanced heteronuclear single
quantum correlation spectroscopy with improved sensitivity. J. Am. Chem. Soc. 114: 10663-10665;
1992.
[71] Bradford, M.M. A rapid and sensitive method for the quantitation of microgram quantities of
protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248-254; 1976.
[72] Laemmli, U.K. Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227: 680-685; 1970.
[73] Wray, W.; Boulikas, T.; Wray, V.P.; Hancock, R. Silver staining of proteins in polyacrylamide
gels. Anal. Biochem. 118: 197-203; 1981.
[74] Del Corso, A.; Barsacchi, D.; Giannessi, M.; Tozzi, M.G.; Camici, M.; Houben, J.L.;
Zandomeneghi, M.; Mura, U. Arch. Biochem. Biophys. 283: 512-518; 1990.
[75] Noltmann, E.A.; Gubler, C.J.; Kuby, S.A. Glucose 6-phosphate dehydrogenase
(Zwischenferment). I. Isolation of the crystalline enzyme from yeast. J. Biol. Chem. 236: 1225-
1230; 1961.
[76] Kim, T.K.; Lee, P.; Colman, R.F. Critical role of Lys212 and Tyr140 in porcine NADP-
dependent isocitrate dehydrogenase. J. Biol. Chem. 278: 49323-49331; 2003.
[77] Betts, S.A.; Mayer, R.J. Purification and properties of 6-phosphogluconate dehydrogenase
from rabbit mammary gland. Biochem. J. 151: 263-270; 1975.
[78] Schieber, A.; Rainer, W.F.; Ghisla, S. Purification and properties of prostaglandin 9-
ketoreductase from pig and human kidney. Eur. J. Biochem. 206: 491-502; 1992.
[79] Jarabak, J. Early steps in prostaglandin metabolism in the human placenta. Am. J. Obstet.
Gynecol. 138: 534-540; 1980.
[80] Ji, B.; Ito, K.; Suzuki, H.; Sugiyama, Y.; Horie, T. Multidrug resistance-associated protein2
(MRP2) plays an important role in the biliary excretion of glutathione conjugates of 4-
hydroxynonenal. Free Radic Biol Med. 33: 370-378; 2002.
ꢀ
[81] Balogh, L.M.; Roberts, A. G.; Shireman, L. M.; Greene, R.J.; Atkins, W.M. The
stereochemical course of 4-hydroxy-2-nonenal metabolism by glutathione S-transferase. J. Biol.
Chem. 283: 16702-16710; 2008.
[82] Aldini, G.; Carini, M.;Beretta, G.; Bradamante, S.; Facino, R. M. Carnosine is a quencher of 4-
hydroxy-nonenal: through what mechanism of reaction?Biochem. Biophys. Res. Commun. 298: 699-
706; 2002.

[83] Beretta, G.; Artali, R.; Regazzoni, L.; Panigati, M.;Facino, R.M. Glycyl-histidyl-lysine (GHK)
is a quencher of alpha,beta-4-hydroxy-trans-2-nonenal: a comparison with carnosine. Insights into
the mechanism of reaction by electrospray ionization mass spectrometry, 1H NMR, and
computational techniques. Chem. Res. Toxicol. 20:1309-1314; 2007.
ꢀ
[84] Xie, C.; Zhong, D.; Chen, X. A fragmentation-based method for the differentiation of
glutathione conjugates by high-resolution mass spectrometry with electrospray ionization. Anal.
Chim. Acta 788: 89-98; 2013.
[85] Wermuth, B. Purification and properties of an NADPH-dependent carbonyl reductase from
human brain. Relationship to prostaglandin 9-ketoreductase and xenobiotic ketone reductase. J.
Biol. Chem. 256: 1206-1213; 1981.
[86] Hoffmann, F.; Maser, E. Carbonyl reductases and pluripotent hydroxysteroid dehydrogenases
of the short-chai dehydrogenase/reductase superfamily. Drug Metab. Rev. 39: 87-144; 2007.
[87] Doorn, J.A.; Maser, E.; Blum, A.; Claffey, D.J.; Petersen, D.R. Human carbonyl reductase
catalyzes reduction of 4-oxonon-2-enal. Biochemistry 43: 13106-13114; 2004.
[88] Forrest, G.L.; Gonzalez, B. Carbonyl reductase. Chem. Biol. Interact. 129: 21-40; 2000.
[89] Kelner, M.J.; Estes, L.; Rutherford, M.; Uglik, S.F.; Peitzke, J.A. Heterologous expression of
carbonyl reductase: demonstration of prostaglandin 9-ketoreductase activity and paraquat
resistence. Life Sci. 61: 2317-2322; 1997.
[90] Maser, E. Neuroprotective role for carbonyl reductase? Biochem. Biophys. Res. Commun. 340:
1019-1022; 2006.
[91] Kim, Y.N.; Jung, H.Y.; Eum, W.S.; Kim, D.W.; Shin, M.J.; Ahn, E.H.; Kim, S.J.; Lee, C.H.;
Yong, J.I.; Ryu, E.J.; Park, J.; Choi, J.H.; Hwang, I.K.; Choi, S.Y. Neuroprotective effects of PEP-
1-carbonyl reductase 1 against oxidative-stress-induced ischemic neuronal cell damage. Free Rad.
Biol. Med. 69: 181-196; 2014.
[92] Rashid, M.A.; Lee, S.; Tak, E.; Lee, J., Choi, T.G.; Lee, J.W.; Kim, J.B.; Youn, J.H.; Kang, I.;
Ha, J.; Kim, S.S. Carbonyl reductase 1 protects pancreatic !-cells against oxidative stress-induced
apoptosis in glucotoxicity and glucolipotoxicity. Free Rad. Biol. Med. 49: 1522-1533; 2010.
[93] Bateman, R.L.; Rauh, D.; Tavshanjian, B.; Shokat, K.M. Human carbonyl reductase 1 is an S-
nitrosoglutathione reductase. J. Biol. Chem. 283: 35756-35762; 2008.
[94] Krook, M.; Ghosh, D.; Strömberg, R.; Carlquist, M.; Jörnvall, H.; Carboxyethyllysine in a
protein: native carbonyl reductase/NADP(+)-dependent prostaglandin dehydrogenase. Proc. Natl.
Acad. Sci. 90: 502-506; 1993.
[95] Sciotti, M.A.; Nakajin, S.; Wermuth, B.; Baker, M.E. Mutation of threonine-241 to proline
eliminates autocatalytic modification of human carbonyl reductase. Biochem. J. 350: 89-92; 2000.
ꢀ

[96] Iwata, N.; Inazu, N.; Satoh, T. The purification and properties of NADPH-dependent carbonyl
reductases from rat ovary. J. Biochem. 105: 556-564; 1989.ꢀ
[97] Ahmad, I.M.; Aykin-Burns, N.; Sim, J.E.; Walsh, S.A.; Higashikubo, R.; Buettner, G.R.;
Venkataraman, S.; Mackey, M.A.; Flanagan, S.W.; Oberley, L.W.; Spitz, D.R. Mitochondrial O2-
and H2O2 mediate glucose deprivation-induced stress in human cancer cells. J. Biol. Chem. 280:
4254-4263; 2005.
[98] Aykin-Burns, N.; Ahmad, I.M.; Zhu, Y.; Oberley, L.W.; Spitz, D.R. Increased levels of
superoxide and H2O2 mediate the differential susceptibility of cancer cells versus normal cells to
glucose deprivation. Biochem. J. 418: 29-37; 2009.
[99] Cairns, R.A.; Harris, I.S.; Mak, T.W. Regulation of cancer cell metabolism. Nat. Rev. Cancer
11: 85-95; 2011.
Figure captions
Figure 1. Double reciprocal plot for the reduction of 9,10-PQ by CBR1.
The initial rate measurements of the NADPH-dependent reduction of 9,10-PQ are reported as
double reciprocal plot. The assays were performed in standard conditions using 1.2 mU/mL of
purified CBR1. Error bars represent the standard deviations of the mean from at least three
independent measurements.
Figure 2.Kinetic analysis of the NADP+ dependent dehydrogenase activity of CBR1.
The initial rate measurements of the NADP+-dependent oxidation of GSHNE (panel A) and PGB1
(panel B) are reported as double reciprocal plot. The assays were performed in standard conditions
using 2.3 mU/mL and 4.6 mU/mL of purified CBR1, respectively. Error bars represent the standard
deviations of the mean from at least three independent measurements.
Figure 3. Cofactor specificity of CBR1.
The initial rate of GSHNE oxidation measured at the indicated concentrations of NADP+ alone
(closed circles), or in the presence of either 10 µM NADPH (squares) or10 µM of NAD+(triangles)
are reported as double reciprocal plot. The assays were performed in standard conditions using2.3
mU/mL of purified CBR1. Error bars represent the standard deviations of the mean from at least
three independent measurements.

Figure 4. Fragmentation spectra of the GSH derivatives observed in this study after their
collision induced dissociation in positive polarity.
The fragment nomenclature is identical to that described in Xie et al. [84]. In particular, a=P-
75.032, b=145.061, c=P-146.069, d=P-273.096 or 275.112, e=P-129.043, f=130.050, g=P-232.070,
h= P-249.095, I =308.091, j=P-307.084 or 305.068, k=274.103, where P represents the [M+H]+ ion
of the specific GSH conjugate. Panels A, B, C and D show the fragmentation spectrum of the
species showing a [M+H]+ ion at the experimental m/z value 464.18, 462.23, 480.25 and 478.19,
respectively. R1, R2, R3 and R4 represent the corresponding neutral losses of C9H16O2 (156.12 Da),
C9H14O2 (154.10 Da), C9H16O3 (172.11 Da) and C9H14O3 (170.09 Da), respectively. Apart from the
most intense peak in each presented spectrum, the other regions in the spectra were magnified by a
factor of 10. To obtain a more certain assignation of the MS2 spectra, several experiments of MS3
fragmentation were also performed. As an example, Supplementary Fig. 7S shows the MS3
experiment of the peak at m/z 446.07 in the MS2 spectrum of the original species at m/z 464.18,
which was assigned to the neutral loss of a single water molecule from the parent ion.
Figure 5. Time course of GSHNE oxidation catalyzed by CBR1.
Plots of peak areas of the species detected at m/z values 464.18, 462.23, 480.25 and 478.19 at
different times of incubation with CBR1 (panels A to D, respectively), as obtained by the
integration of the corresponding extracted ion currents in LC-MS runs.
Figure 6. Proposed reaction pathway catalyzed by CBR1.
The scheme reports possible reaction pathways as deriving from the nature of the products observed
during time-course nLC-ESI-LIT-MS/MS and NMR analyses of the incubation mixtures.
Figure 7. NMR spectra of single products and their mixtures in phosphate buffer.