Oxidation of cholesterol and O-protected derivatives by the environmental pollutant NO2?

Article abstract of DOI:10.1039/c5cc09663d

Exposure of O-protected and free cholesterol 1 to NO₂· under exclusion of water leads to nitroimine nitrates 2 through a non-radical mechanism, which reveals the high susceptibility of the π system to oxidative damage. In the presence of moisture the reaction leads to 6-nitrocholesterols 3, which result from hydrolysis and oxidation of 2.

Full text of DOI:10.1039/c5cc09663d

ChemComm
View Article Online
View Journal
COMMUNICATION
Oxidation of cholesterol and O-protected
derivatives by the environmental pollutant NO2 †
ꢀ
Cite this:DOI: 10.1039/c5cc09663d
A. N. Zalewski, J. G. Nathanael, J. M. White and U. Wille*
Received 23rd November 2015,
Accepted 16th February 2016
DOI: 10.1039/c5cc09663d
www.rsc.org/chemcomm
ꢀ
Exposure of O-protected and free cholesterol 1 to NO2 under moiety with concurrent joining of the remaining peptide ends to
exclusion of water leads to nitroimine nitrates 2 through a non- form a shortened peptide.4
ꢀ
radical mechanism, which reveals the high susceptibility of the p
system to oxidative damage. In the presence of moisture the (NO2ꢀ) = 1.2 Â 10À2 M atmÀ1, at 298 K],6 the majority of inhaled
Because of the only moderate water solubility of NO2 [KH
ꢀ
reaction leads to 6-nitrocholesterols 3, which result from hydrolysis NO2 is believed to reach the lower respiratory tract. In the
and oxidation of 2.
lower airways cholesterol (1a) is the most abundant neutral
lipid in the epithelial lining fluid,7 which is directly exposed to
Nitrogen dioxide, NO2ꢀ, is an important gaseous air pollutant, oxidizing environmental pollutants.8 Cholesterol is an integral
which is formed through combustion processes, for example component of the pulmonary surfactant (content about 5–10%)9a
car exhaust. This highly poisonous free radical is a promoter of and essential for the integrity of cellular membranes by maintaining
oxidative stress,1 and in vivo studies showed that NO2ꢀ exposure membrane fluidity and function.9b Interestingly, elevated cholesterol
significantly reduces the levels of antioxidants (e.g., glutathione, concentrations were found in the airway epithelial cells in
ascorbic acid, uric acid, etc.) in the airway surface fluids (ASF).2 mouse models of cystic fibrosis.10
Such a weakened defence shield could provide a pathway for
environmental oxidants to directly attack proteins and lipids
present on cell surfaces or in the ASF. The resulting highly
reactive oxidation products may subsequently damage the
underlying epithelial cells and cause inflammation. However,
despite the known adverse health effects, it is surprising that
details of the processes that take place during the encounter of
ꢀ
NO2 with ASF constituents are still not well understood at the
molecular level.
We recently discovered an unexpected ‘dual’ reactivity of
ꢀ
ꢀ
NO2 with oligopeptides. Thus, NO2 reacts as a radical oxidant
Reaction of cholesterol with the non-radical air pollutant
[E0(NO2ꢀ/NO2ꢀ) = 1.03 V]3 with peptides possessing readily ozone has been shown to occur at the alkene moiety and leads
oxidisable side chains, such as in tyrosine or tryptophan, which to formation of bioactive oxysterols, which, if not eliminated,
leads to formation of nitrotyrosines or pyrroloindolines, respectively.4 induce apoptosis and cytotoxicity.7,11 A high reactivity towards
ꢀ
Oligopeptides with non-oxidisable side chains, on the other NO2 has also been reported. In vivo exposure studies showed a
ꢀ
hand, react with NO2 via its dimer N2O4 through non-radical significant reduction of the cholesterol content in the brains
N-nitrosation of peptide bonds. This triggers a rearrangement of of guinea pigs.12 Likewise, loss of cholesterol was also observed
the peptide backbone through formal excision of an amino acid in cholesterol monomolecular films upon exposure to gaseous
13
ꢀ
NO2
.
However, only few studies were concerned with the
identification of the reaction products and mechanism of their
formation. Thus, when 1a was exposed to NO2 in tetrachloro-
ꢀ
School of Chemistry, Bio21 Institute, The University of Melbourne, Parkville,
Victoria 3010, Australia. E-mail: uwille@unimelb.edu.au
methane14 or in hexane under varying conditions,15 formation
of cholesteryl nitrite (1 with R = NO) was found as major
pathway, in addition to by-products that result from reactions
involving the CQC moiety.14
† Electronic supplementary information (ESI) available: General experimental
procedures, spectroscopic data for 1–3b,c, crystallographic data for 2b, Gaussian
archive entries for the reaction 10 - 11. CCDC 1436865. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c5cc09663d
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
Because of the two reactive sites in cholesterol, i.e., the protonated molecular ion [M + H]+ at m/z 522.35370, which
+
electron rich p system and the hydroxyl group, knowledge of is consistent with the molecular formula C28H48N3O6 (calcd
ꢀ
the products formed in reactions with NO2 under simulated 522.35373) and confirms addition of three nitrogen and four
1
environmental conditions is essential for gaining mechanistic oxygen atoms to the molecular framework. The H NMR data
understanding and assessing the potential biological implications. revealed only loss of the vinyl proton at C-6, whereas the 13C
We have therefore performed a product study of the reaction of NMR spectrum showed the carbon atoms of the former alkene
ꢀ
NO2 with the ‘isolated’ p system using the O-protected cholesterols moiety at d = 92.99 (C-5) and 176.01 ppm (C-6), respectively.
1b and 1c, and with free cholesterol (1a). Different experimental Fortunately, 2b crystallised upon standing. X-ray analysis
conditions were employed: (i) the heterogeneous reaction of gaseous (Fig. S14, ESI†) revealed a nitroimine group at C-6 and a nitrate
ꢀ
NO2 with solid 1b,c in the absence and presence of moisture, and ester at C-5, which sits trans to the angular methyl group and
(ii) the homogeneous reaction of NO2ꢀ with 1a–c in organic solvents, the oxygen substituent at C-3. Likewise, HRMS analysis for 2c
e.g., dichloromethane or acetonitrile.
For the gas/solid exposure studies, solid 1b or 1c was placed which is consistent with the molecular formula C29H48N3O7
in a reaction flask sealed with a silicon septum. Liquid NO2
showed the protonated molecular ion [M + H]+ at m/z 550.34955,
+
ꢀ
(calcd 550.34868). Although 2c did not crystallise, the comparable
(ca. 20 equiv.) was injected into an open vial standing upright in spectroscopic data of 2b and 2c strongly support the presence of a
the reaction flask to avoid direct contact ꢀwith the substrate. similar nitroimine nitrate motif in 2c (see ESI† for details).
Evaporation created an atmosphere of NO2 gas, which reacted
with 1b,c at the gas/solid interface.‡ In order to simulate the exposure to gaseous NO2 occurred in the presence of air. This
The nitroimine nitrate ꢀester was also exclusively formed, when
conditions in airway surfaces, where the aqueous lining fluid suggests an ionic mechanism, similar to our findings in the
containing cholesterol is exposed to gaseous NO2ꢀ, we also reaction of NO2 with peptides,4,5 since intermediately formed
ꢀ
performed experiments where the substrate was moistened with radicals would be expected to react, at least in part, with oxygen
ꢀ
40 equivalents of water prior to exposure to gaseous NO2 .§ The in air, leading to different products.
reactions in organic solvents were carried out under exclusion
Interestingly, when the reaction was carried out as homo-
ꢀ
of moisture by injecting a measured excess amount of liquid geneous reaction by dissolving 1b,c and NO2 in dichloromethane
ꢀ
NO2 into the solution containing the substrate, as described or acetonitrile, the same outcome was obtained. Under these
previously.4,5 For experimental details see ESI.† Product identifi- conditions, however, consumption of the starting material was
cation was performed by NMR, HRMS (ESI), IR and X-ray complete after 20 minutes. Similar to the heterogeneous gas
analysis (where possible).
exposure studies, elongation of the reaction time did not lead to
When solid O-protected cholesterols 1b or 1c were exposed decomposition or further conversion of the reaction product 2.
ꢀ
to gaseous NO2 under exclusion of moisture a change of the This demonstrates that aprotic organic solvents provide a suitable
ꢀ
substrate from a white solid to a green-blue oil occurred within and timesaving model environment for NO2 gas phase exposure
ꢀ
few hours of contact. The reaction was very clean, and the studies under non-aqueous conditions. Finally, the role of NO2
1H NMR spectra of the respective raw reaction mixtures showed as the sole species responsible for the transformation was
formation of the nitroimine nitrates 2b and 2c as only product undoubtedly confirmed by in situ 1H NMR analysis of the reaction
(1H NMR purity 494%), which did not undergo further reaction, of 1b, which revealed 2b as the only product after 20 minutes of
even after an extended contact time of up to three days (Scheme 1). reaction time (see Fig. S12, ESI†).
ꢀ
Due to some decomposition on the silica column, the isolated
The reaction of cholesterol (1a) with NO2 in dichloromethane
yield after purification was lower, although, once isolated, both 2b gave a complex mixture of products, undoubtedly due to reactions
and 2c are stable compounds. HRMS analysis for 2b revealed the involving the hydroxyl group. Fortunately, we were able to isolate
the major product in 25% yield, which was unambiguously
identified from the NMR, IR and HRMS data as the nitroimine
nitrate 2a (Scheme 1, see ESIꢀ† for details). This clearly shows
that selective reaction of NO2 with the p system in 1a occurs
even in the presence of the hydroxyl group to a significant
extent, which demonstrates the high susceptibility of the alkene
moiety in cholesterol to oxidative damage. Gas/solid exposure
studies with 1a were also attempted, but the resulting product
mixture was too complex to be analysed.
A mechanistic proposal for formation of 2 is outlined in
Scheme 2a (picturing only the structural motif relevant to this
reaction). We believe that the dimer N2O4 is the actual key-
species in this process,4,5 which is a non-radical nitrosating
agent as revealed by its isomeric ionic form [NO+NO3À].16 The
reaction occurs by electrophilic addition of NO+ to the CQC
bond, which is immediately followed by recombination of the
resulting cationic intermediate 4 with NO3À from the sterically
ꢀ
Scheme 1 Reaction of cholesterols 1a–c with NO2
.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016

View Article Online
ChemComm
Communication
ꢀ
Scheme 2 Proposed mechanism for formation of the nitroimine nitrate 2 and nitrocholesterol 3 in the reaction of cholesterols 1a–c with NO2 in the
absence (a) and presence of water (b). M06-2X/6-311++G** energies for the model reaction 10 - 11 in kJ molÀ1
.
less hindered side to give the nitroso nitrate ester 5. Tautomerisation
At first sight, formation of 3c might suggest a straightforward
ꢀ
of the nitroso group yields oxime 6, which is subsequently converted radical process, where addition of NO2 to the sterically less
to the nitroimine through reÀaction with NO+.17 The high efficiency of hindered site of the alkene moiety in 1c is followed by hydrogen
20
ꢀ
the trapping of 4 by NO3 was revealed ꢀthrough competition abstraction, for example by NO2 , to restore the p system
experiments performed with 1b and NO2 in dichloromethane (mechanism not shown). However, because of the unquestionable
in the presence of excess acetic acid. 1H NMR analysis of role of water in this transformation, we explored the mechanism in
the product mixture indicated formation of the corresponding more detail. Thus, exposure of a 1 : 1 mixture of nitroimine nitrate
nitroimine acetate (not shown)18 together with nitrate 2b in a 2c and nitrocholesterol 3c to gaseous NO2 in the presence of
ꢀ
1 : 10 ratio, which suggests that NO+ and NO3 exist as contact moisture resulted in complete conversion of 2c into 3c (see
À
ion pair with very limited solvent separation. It should be noted Fig. S13 in the ESI†). This provides strong support for 2c being
ꢀ
that the ionic character of N2O4 is a feature largely known for an intermediate in the NO2 mediated transformation 1c - 3c.
the condensed phase.16 Our finding that 2 is formed in both the No reaction took place when 2c was treated with either water or
ꢀ
gas/solid reaction as well as in aprotic organic solvents, indicates with NO2 in the absence of moisture (see above). However, 3c
that [NO+NO3À] should also be the active species in our gas phase was also formed, when the reaction of NO2 was performed with
ꢀ
exposure studies. It could be speculated that dissociation into the 2c that was dampened with methanol (not shown), illustrating
ꢀ
ion pair occurs when gaseous NO2 /N2O4 ‘dissolves’ in the solid the importance of a proton source for this transformation. On
ꢀ
substrate 1, which could explain the colour change observed the other hand, although the system NO2 /N2O4/water provides
during the exposure (see above).
a considerably acidic environment, no reaction occurred when 2c
Interestingly, a completely different outcome was obtained was treated with concentrated nitric acid in dichloromethane. This
when solid cholesterol acetate 1c, which was moistened with shows that an acidic environment alone is not responsible for the
ꢀ
water (40 equivalents), was exposed to gaseous NO2 .¶ According transformation 2 - 3. Based on these observations a mechanism
1
to H NMR analysis of the crude reaction mixture, practically is proposed in Scheme 2b. Due to the electron-withdrawing nature
quantitative conversion of 1c to 6-nitro cholesterol 3c occurred, of the nitro group, the imine nitrogen atom in 2 is positively
which is stable under the reaction conditions for several days polarised and could be attacked by water in a nucleophilic allylic
(up to 36 hours were explored).8 The lack of a vinyl proton in the substitution, where nitric acid acts as leaving group. Protonation of
1H NMR spectrum of 3c and the molecular ion of the dimer the nitro group in 7 enables a subsequent heterolytic N–N bond
[2M + Na]+ at m/z 969.68402 in the HRMS, which corresponds cleavage in 8 that yields nitroso compound 9 and nitrous acid.
to the molecular formula C58H94N2O8Na+ (calcd 969.69024), Density functional theory calculations for the simplified model
suggests substitution of the hydrogen atom at C-6 by NO2.** system 10 - 11 at the M06-2X/6-311++G** level of theory showed
Since the product did not crystallise, the structure of 3c was that this process is not only associated with a modest activation
assigned by comparing the NMR data with those of a reference barrier, Ea, of 51 kJ molÀ1, but is also considerably exothermic
sample, which was prepared by nitration of 1c with sodium (DE = À109 kJ molÀ1).‡‡ The resulting 6-nitroso cholesterol 9 is
nitrite and nitric acid.19 It should be noted that exposure of subsequently oxidised to the final product by NO2 , which acts
ꢀ
moistened cholesterylmethyl ether 1b to gaseous NO2 resulted as one-electron oxidant in this step.21
ꢀ
in ether cleavage and formation of various hydrophilic products,
which could not be identified.††
In conclusion, the cholesterol alkene moiety is readily attacked
by NO2 in both O-protected and free cholesterol. This reaction
ꢀ
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
The latter could be produced through collision induced dissociation
as [M + Na]+ at m/z 496.33850 in the HRMS (ESI), which is in accordance
with the molecular formula C29H48NO4Na+ (calcd 496.34756).
†† Ether cleavage is likely a result of the acidic environment of this
reaction system.
involves N2O4 as key species, which acts as a non-radical nitro-
sating agent and leads to formation of nitroimine nitrate 2. In
the presence of moisture, 2 undergoes rapid hydrolysis and
ꢀ
further NO2 mediated oxidation to give nitrocholesterol 3. The
‡‡ The reaction energy is calculated for the product association
latter would be the expected product in airway surface fluids,
complex [11–H2O–NO+].
ꢀ
where it could serve as biomarker for NO2 pollution. In fact,
1 R. M. Lodovici and E. Bigagli, J. Toxicol., 2011, 2011, 487074;
nitrated lipids have been detected in vivo, where they play an
important role in cell signalling.22 Recent research revealed also
beneficial effects in some nitrated lipids through their decom-
position to release NOꢀ, which causes vasorelaxation and has
anti-inflammatory effects.23
¨
U. Poschl and M. Shiraiwa, Chem. Rev., 2015, 115, 4440.
2 F. J. Kelly, Occup. Environ. Med., 2003, 60, 612.
3 R. E. Huie and P. Neta, J. Phys. Chem., 1986, 90, 1193.
4 L. F. Gamon, J. M. White and U. Wille, Org. Biomol. Chem., 2014, 12, 8280.
5 L. F. Gamon, J. G. Nathanael, B. I. Taggert, F. A. Henry, J. Bogena
and U. Wille, Chem. – Eur. J., 2015, 21, 14924.
6 W. L. Chameides, J. Geophys. Res., 1984, 89, 4739.
7 M. K. Pulfer, K. Harrison and R. C. Murphy, J. Am. Soc. Mass
Spectrom., 2004, 15, 194.
ꢀ
The mechanism by which NO2 was found to react with
the cholesterol p system in this work is considerably different
from the radical addition/allylic hydrogen abstraction pathway
8 K. M. Gowdy and M. B. Fessler, Pulm. Pharmacol. Ther., 2013, 26, 430.
ꢀ
proposed by Pryor et al. for the reaction of NO2 with cyclohexene,
´
9 (a) J. Perez-Gil, Biochim. Biophys. Acta, 2008, 1778, 1676; (b) S. Orgeig and
C. B. Daniels, Comp. Biochem. Physiol., Part A: Mol. Integr. Physiol., 2001,
129, 75.
which was used as model system to study exposure of unsaturated
fatty acids to air pollution.20 The findings presented here, in
particular the mechanistic sꢀtudies, confirm our previous obser-
vations that exposure to NO2 leads to oxidative damage not only
through radical pathways, but that ionic chemistry also contributes
crucially to the reactivity of this important environmental pollutant
10 D. Jiang, D. Fang, T. J. Kelley and J. D. Burgess, Anal. Chem., 2008, 80, 1235.
11 M. Dahl, A. K. Bauer, M. Arredouani, R. Soininen, K. Tryggvason, S. R.
Kleeberger and L. Kobzik, J. Clin. Invest., 2007, 117, 757; M. K. Pulfer and
R. C. Murphy, J. Biol. Chem., 2004, 279, 26331; A. Sevanian, J. Berliner and
H. Peterson, J. Lipid Res., 1991, 32, 147; Y. C. O’Callaghan, J. A. Woods
and N. M. O’Brien, Cell Biol. Toxicol., 2001, 17, 127.
12 H. Farahani and M. Hasan, Pharmacol. Toxicol., 1990, 66, 146.
13 A. M. Kamel, A. Felmeister and N. D. Weiner, J. Pharm. Sci., 1970,
59, 1807.
and needs to be considered when assessing NO2 toxicity.4,5
ꢀ
This conclusion is further ꢀsupported by a recent suggestion that
antioxidants mediate NO2 absorption in the ASF by catalysing
14 T. Kobayashi and K. Kubota, Chemosphere, 1980, 9, 777.
15 S. S. Mirvish, D. M. Babcook, A. D. Deshpande and D. L. Nagel,
Cancer Lett., 1986, 31, 97.
its hydrolytic disproportionation to NO+ and NO3À 24 This could
.
be the key to our understanding why NO2ꢀ has such pronounced
biological effects in vivo, despite its low water solubility.
CAUTION: N-Nitroimines are known carcinogens. Proper
precautions against inhalation of the vapours and contact with
skin should always be maintained.
16 M. Shiri, M. A. Zolfigol, H. G. Kruger and Z. Tanbakouchian, Tetrahedron,
2010, 66, 9077; P. Gray and A. D. Yoffe, Chem. Rev., 1955, 55, 1069;
C. L. Lv, Y. D. Liu and R. Zhong, J. Phys. Chem. A, 2008, 112, 7098.
17 J. J. Lie, N-Nitroimines and N-Nitrosoimines, in Science of Synthesis, 27:
Category 4. Compounds with Two Carbon Heteroatom Bonds, ed. A. Padwa
and D. Bellus, Georg Thieme Verlag KG, 2005; S. Adamopoulos, A. J.
Boulton, R. Tadayoni and G. A. Webb, J. Chem. Soc., Perkin Trans. 1,
1987, 2073.
This work was supported by The University of Melbourne
and the National Computational Infrastructure (NCI).
´
´
´
´
18 Y. Lopez, K. M. Ruız-Perez, R. Yepez, R. Santillan, M. Flores-Alamo
and M. A. Iglesias-Arteaga, Steroids, 2008, 73, 657; A. B. Landge and
C. R. Narayanan, Indian J. Chem., Sect. B: Org. Chem. Incl. Med. Chem.,
1995, 34, 389; M. Onda, Y. Konda and R. Yabuki, Chem. Pharm. Bull.,
1975, 23, 611; M. Onda, R. Yabuki, K. Takeuchi and Y. Konda, Chem.
Pharm. Bull., 1976, 24, 1795; A. G. Gonzalez, R. Freire, M. G. Garcia-
Estrada, J. A. Salazar and E. Suarez, An. Quim., 1972, 68, 1145.
19 Shamsuzzaman, A. Mashrai, H. Khanam, Y. N. Mabkhot and W. Frey,
J. Mol. Struct., 2014, 1063, 219.
Notes and references
‡ The reaction was quenched by removing unreacted NO2 through a
ꢀ
stream of nitrogen, followed by dissolving the reaction mixture in
dichloromethane and concentration in vacuo. Aqueous work-up, where
the reaction was first quenched with saturated sodium bicarbonate
solution, followed by extraction with dichloromethane, gave the same
outcome (see ESI†).
20 W. A. Pryor and J. W. Lightsey, Science, 1981, 214, 435.
21 B. G. Gowenlock, J. Pfab and V. M. Young, J. Chem. Soc., Perkin Trans.
2, 1997, 1793.
22 B. Kalyanaraman, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 11527;
H. Rubbo and R. Radi, Biochim. Biophys. Acta, 2008, 1780, 1318.
23 E. S. Lima, M. G. Bonini, O. Augusto, H. V. Barbeiro, H. P. Souza and
D. S. P. Abdalla, Free Radical Biol. Med., 2005, 39, 532.
24 S. Enami, M. R. Hoffmann and A. J. Colussi, J. Phys. Chem. B, 2009,
113, 7977.
§ Because of the low soluꢀbility of 1b/c in water, it was not possible to
study the reaction of NO2 in the aqueous phase.
¶ Due to the complexity of the reaction involving 1a under anhydrous
conditions, the reactions in presence of moisture were only performed
with the O-protected cholesterols.
ꢀ
8 It was found that reaction of 1c with NO2 to 3c requires only trace
amounts of moisture.
** The ion of the sodium adduct of the dimer was significantly more
abundant in the mass spectrum of 3c than that of the monomer.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016