San Antonio, TX
Sunrise Free Radical School
Oxy 2002
Oxygen 2002
Sunrise Free Radical School
November 21-24, 2002
San Antonio, TX
What do •NO, O=NOO-, NO2, N2O3, etc. do,
and how do they do it: Basics
James K. Hurst, Ph.D.
Department of Chemistry
FULMER 170
Washington State University
PULLMAN, WA 99164-4630
Tel: 509-335-7848/5-5334
Fax: 509-335-8867
E-mail: [email protected]
Oxygen Society Annual Meeting,
San Antonio, TX Nov 21-24, 2002
James Hurst
Lecture Notes-Biologically Relevant Chemistry of ONOOH
and Its Reaction Products
James K. Hurst
Department of Chemistry
Washington State University
Pullman, WA 99164
Goal: To provide the chemical background required to productively contemplate
origins and consequences of peroxynitrite formation in biological environments
Challenges: Critical errors exist in the chemical literature on reactive nitrogen
species (RNS) and these have been inadvertently incorporated into discussions of
biological reaction mechanisms.
“An expert seldom gives an objective view; he gives his own view”-Morarji Desai
Caveats: The pathophysiology of oxidative stress may bear no simple relationship
to the major reaction pathways of RNS and reactive oxygen species (ROS).* This
seems particularly likely for c hronic diseases whose progression involves long
periods of insult, but could also be an issue under conditions of acute exposure to
oxidants.
*e.g., toxicity of OH• toward bacteria:
OH• is not very toxic in biological environments because it is too
reactive to reach vulnerable targets.
Premise 1:
100
H2O
OH + H+ + eses + N2O + H+
N2 + OH
% viable E.coli
75
OH
.
50
Net: H2O + N2O
radical
scavengers
25
CO3
0
0
5
10
.-
15
irradiation time (min)
20
25
N2 + 2OH
Toxin
(E. coli)
LD90
(molecules/bacterium)
rel. toxicity
108
3×1011
1010
2×1011
1.0
0.003
0.01
0.005
(however, killing by OH• is intracellular!)
OH•(internal)*
105
1000
*internal volume of 106 cells/mL M 5×10-7 cm3
HOCl
H2O2
CO3•OH•
Conclusion: intracellularly generated OH• is highly toxic!
Premise 2:
OH• formed by ONOOH bond homolysis cannot be a significant
component of the oxidative load because bimolecular reactions of ONOO/ONOOH (e.g., with CO2, thiols (RSH), metalloproteins) will predominate
in biological environments.
NO2 + OH
k1
k2[Mn+]
-
ONOO /ONOOH
k2''[RSH]
k2'[CO2]
At pH 7.4,
k1 M 0.34 s-1
k2[Mn+] 300 s-1
k2'[CO2] M 20 s-1
k2'’[RSH] 3 s-1
Simple competition kinetics predicts:
[OH•]/[ONOO-/ONOOH] = k1/*(ki[Xi]) 0.015
i.e., less than 2% of the peroxynitrite formed will generate OH•. Nonetheless, if
intracellular OH• is 103 times more toxic than other oxidants, it could still be a significant
component of oxidative stress.
Possible Biological Origins of ONOO-/ONOOH
A.
NO• + O2•-
ONOO- (radical coupling, k 1010 M-1s-1)
Primary sources of NO• and O2•- in respiring tissues:
NO•:
H2N
NH2
H2N
NH
NH
NADPH
O2
H3N
H2N
NOH
NH
1/2 NADPH
O2
H2O
COO
O
H3N
COO
+ NO
H2O
H3N
COO
catalyzed by NOS isozymes (nNOS(I), iNOS(II), eNOS(III))
O2•-:
i) O2-binding heme proteins--
N
N
H
O
Fe
II
(Hb,Mb)
Fe
+ O2
X
O
[e-]
O
III
X
FeIV
H2O
π
+
X
(cyt P450, NOS)
FeIII
+ O2 -
X
e.g., hemoglobin (Hb); myoglobin (Mb);
cytochrome P450; L-arg or tetrahydrobiopterin (H4B)-depleted NOS.
(O-O) = 1105 cm-1 (Hb-O2, Mb-O2)
1140 cm-1 (P450-O2)
cf. 1556 cm-1 (O=O)
1145 cm-1 (KO2)
836 cm-1 (NH4O2H)
ii) endogenous semiquinones (e.g., mitochondrial bc complex (III))--
Q•-/QH + O2 Q(+ H+) + O2•iii) stimulated phagocytes (e.g., neutrophil)--
NADP+
C6H12O6
O2 -
[1]
[2]
CO2
NADPH
O2
[1] pentose phosphate pathway
[2] NADPH oxidase
iv) molybdoflavoenzymes (e.g., xanthine oxidase)
O
N
HN
H2O2, O2 -
N
N
H
xanthine
O
hypoxanthine
O
H
N
HN
N
H
O
N
H
uric acid
O
N
H
A conceptual problem (?)-optimal formation of ONOO-/ONOOH requires approximately equal fluxes
of NO• and O2•e.g.,
ONOOH
k1( slow)
{int}
DHR
Rh (hv) + H2O + NO2-
0.8
0.3
Rhodamine (µM/min)
0.6
Rhodamine (µM/min)
H2O + O2
HN
N
H
N
O
H2O2, O2 -
H2O + O2
0.4
0.2
0.2
0.1
-
R(O2 ) = 0.73 µ M/min
0
1
2
3
-
R(NO)/R(O2 )
R(NO) = 0.64 µM/min
0.0
0.0
4
0
4
8
-
R(O2 )/R(NO)
data from: [Jourd”heuil, et al. J.Biol Chem. 2001, 276, 28799-28805]
12
Interpretation:
O2-
NO
ONOO {H+, CO2}
DHR
{OH , NO2 , CO3 -}
NO 2- + OH -
O2
NO
O2-
O2 + OH -
Rhodamine
-
-
NO
O2
NO
{N2O3} O2NOO -
O2 + HCO3CO2 + NO2-
H2O
NO 2-
O2 + NO 2-
Is NOS capable of simultaneous generation of NO• and O2•-?
FeII
[e-]
O2*
X
FeIII
O
FeIII
X
X
NO + O2 H4B
-
or [e ]
H3 B
NO-
O
O22-
NOFeII
FeIII
X
X
O*
SO*
S
FeIV
H2O*
π
+
X
[Stuehr et al. J. Biol. Chem. 2001, 276, 14533-14536]
B. “Nitroxyl” reactions-NO + O2 +[e-]
ONOO-
-[e-]
Eo7 (NO,H+/1HNO) = -0.55 V
1
ONOOH
HNO + O2
Potential biological sources of HNO--
NOS (H4B-depleted environments)
RS-NO, e.g.:
RSH + N2O3
{NO+-NO2-}
RSH + RS-NO
RS-NO + NO2- + H+
RSSR + HNO
Spin restrictions and kinetics—
1
1
NO- + 3O2
HNO + 3O2
3
1
ONOO-
1
ONOOH
NO- + 3O2
1
HNO + 3O2
[2NO,2HO2
1
spin-forbidden
(i.e., slow
ONOO-
2
NO,2HO2]
1
ONOOH
spin-allowed,
(can be fast)
ONOO-/ONOOH formation from “nitroxyl” depends upon its physical
properties (e.g.,spin state, acidity, redox potential).
Previous description--
HNO œ H+ + 1NO-, pKa w 4.7
(weak acid, present as singlet anion in physiological milieu)
reaction with O2 is spin-forbidden
Reassessment-“There is only as much truth in any natural science as there is mathematics in it”-as translated by Sergei Lymar from obscure Russian origins
[Shafirovich & Lymar, Proc. Natl. Acad. Sci, USA 2002, 99, 7340-7345]
1
ûGo = * ûfGoproducts - * ûfGoreactants
NO-
o
∆fG (kJ/mol)
240
210
acidity of HNO-1
HNO œ H+ + 3NOûfGo: 115 0 180
ûGo = -RT lnKa = 65 kJ/mol; pKa = 11.4
3
HNO
3
NO-
180
pKa = 11.4
1
HNO
110
NO reduction potential-H+ + NO + e- 1HNO
o
ûfG : 0 102 0 115
ûGo = -nFEo = 13 kJ/mol; Eo = -0.14V (pH 0)
-0.55V (pH 7)
Eo = -0.14 V
NO
100
Photoreaction Dynamics (reactions with O2 and NO•):
hv
HN2O3-
NO- (25%) + 1HNO (75%)
3
1
pKa =9.7
N2O32+
hv
H+
NO
ONOO-
NO2- + 1HNO
3
O2
slow
3
NO- + H+
(spin-forbidden
+
H -transfer)
NO
N3O3-
HNO
N2O + NO2-
Biological implications--
“nitroxyl” in physiological environments is 1HNO (the acid) and is
strongly reducing (Eo7 M -0.55V)
1
HNO and NO• are nitroxyl “sinks”-can ONOO-/ONOOH be formed from 1HNO in physiological environments?
PN/[1HNO] (103×[O2] + 107×[OH-])/{107×([NO•] + [1HNO])}
where PN = [ONOO-] + [ONOOH]
If [O2] > [NO•], [O2] = 10 µM, [NO•] + [1HNO] = 10 nM, then at pH 7.4,
PN/[1HNO] 0.3 (or 30% of the 1HNO decays to PN)
However, if either [NO•] or [1HNO] >> 10 nM, the PN yield is negligible.
(Experimentally--HN2O3- decomposition causes O2 to decrease, implying reaction
with 1HNO)
Low reduction potential of NO• accounts for ability of “nitroxyl” to reduce redoxactive metalloproteins (e.g., (Cu-Zn)SOD, (Fe3+)cyt c, (Fe3+)Hb). What accounts
for its ability to oxidize RSH and Hb-O2? 1HNO is also a moderately strong
oxidant, i.e.,
1
HNO + 2H+ + 2e-
e.g.,
NH2OH (Eo7 M 0.3 V)
RSH + 1HNO
H+ + Hb-O2 + 1HNO
3+
-
(Fe -O2 )
Hb-O2 + NO
RS-NHOH
RSSR + NH2OH
(Fe3+)Hb + NO + H2O2
(Fe3+)Hb + NO3-
Since 1HNO is both strongly oxidizing and reducing, it is unstable to
disproportionation:
3 1HNO 2 NO• + NH2OH, ûEo7 M 0.85 V
Thus, many redox-active metalloproteins are potentially capable of catalyzing
1
HNO decay.
C. “Nitrosyl” reactions-NO+ + H2O2
+[e-]
H+ + ONOOH
Eo(NO+/O) = 1.2 V
-[e-]
NO + O2 -
+[e ]
ONOO -
-
-[e ]
1
ONOOH
HNO + O2
NO+ + H2O œ NO2- + 2H+; K = [NO2-][H+]2/ [NO+] = 103 M2
at pH 7.4, [NO+]/[NO2-] 10-18!
ONOOH
H2O2
NO+
[ONOOH]/[NO2-] = kH2O2[H2O2]/kH2O[H2O]
H2O
= 65[H2O2] M-1
NO2-
at [H2O2] 100 µM, [ONOOH]/[NO2-] 0.007, i.e.,
less than 1% of the generated NO+ will form ONOOH.
Conclusion-free NO+ is not involved in biological ONOO-/ONOOH formation
— too difficult to oxidize NO•
— at physiological [ONOOH], reaction cannot compete with
hydration by H2O.
“Stabilized” NO+--i.e., N2O3 (or NO+-NO2-)--as a nitrosating agent?
NO2- + R
R(asc,urate)
2NO + O2
NO2
NO2NO
RS-
RS + NO2-
N2O3
H2O2
ONOOH + NO2-
RS-(cys,glutathione)
RS-NO + NO2-
N2O3 reactions-at pH 7.4: kRSH M 106 M-1s-1; kH2O2 105 M-1s-1; kH2O[H2O] M103 s-1
if [H2O2] = 0.1 mM; [RSH] = 5 mM,
[ONOOH]/[N2O3] kH2O2[H2O2]/ (ki[Xi]) = 0.007 (0.7%)
NO2 reactions-at pH 7.4: kR M kRSH M 3×107 M-1s-1; kNO = 109 M-1s-1
if [NO•] M 1 µM; [urate] M 0.3 mM (plasma); [asc] M 0.5 mM (cytosol);
[RSH] M 5 mM (cytosol) or 0.1 mM (plasma),
[N2O3]/[NO2•] = kN2O3[N2O3]/ (ki[Xi]) M 0.006 (cytosol)
0.08 (plasma)
Overall-[ONOO]/[NO2•] = [N2O3]/[NO2•]×[ONOOH]/[N2O3] 0.08×0.007 = 6×10-4
(< 0.1%)
Primary pathway for ONOO-/ONOOH formation under physiological conditions
is yet to be identified!
What might change this viewpoint?
Predictions are based upon rate ratios measured for aqueous solution. However,
• Gases are more soluble in hydrocarbons than H2O;
• In lipidic regions of tissues, reaction
dynamics will be very different--s.p.,
— reactions forming neutral products will be
favored;
— intermediates susceptible to hydrolytic
disproportionation by H2O will be
protected (t1/2 will increase).
NO
NO 2
O2
Binding to metalloproteins can activate and stabilize ONOOH precursors:
O2
O2
O2 -
Mn+
M(n+1)+
NO
NO -
Mn+
M(n+1)+
Mn+
NO
• Eo of bound O2, NO• increased
• t1/2 of reactive intermediates
increased
• spin restrictions removed
(Biologically relevant chemical properties)
What is peroxynitrite?
O
O
a structural isomer of nitrate
N
O
N
that contains a peroxo bond:
O
O
O
X-ray structure {(CH3)4N+/OONO-}:
[Worle et al., Chem. Res. Toxicol. 1999, 12, 305-307]
N-O(peroxo) bond length = 1.35 Å
N-O single bond length = 1.41 Å
N=O double bond length = 1.20 Å
or
significant multiple bond character in the N-O peroxo
bond indicated by bond length; delocalization of N
lone pair electrons over nuclear framework stabilizes cis
and trans planar geometries.
from ab initio calculations:
∆Εo = 22-27 kcal/mol
(~12 kcal/mol for ONOOH)
∆Εo = 3-4 kcal/mol
O
N
O
a weak acid:
ONOOH
cf, HOOH
N
O
O
O
O
ONOO- + H+
-
+
HOO + H
pKa = 6.8
pKa = 11.6
O-O bond
dissociation energies:
ONO
HO
OH
OH
NO2 + OH
OH + OH
BDE = 20 kcal/mol
BDE = 52 kcal/mol
A little peroxynitrite history:
First prepared at the turm of the century [Baeyer & Villiger, Berichte1901, 34,
755], but not characterized until much later (1930's).
Regarded as little more than a laboratory curiosity--e.g., the classic 1st text on
mechanistic inorganic chemistry by Yost & Russell (“Systematic Inorganic
Chemistry”, 1946) gives it just one line.
Some interesting research on aromatic hydroxylations and nitrations appeared in
the 1950's [e.g., Halfpenny, E., Robinson, P. L. J. Chem. Soc. 1952, 928-936, ibid.
939-946.]:
NO2
NO2
OH
ONOOH
+
(major)
+
+
NO2
(minor)
OH +
NO2
NO2
but was subsequently largely forgotten. Reactivity was attributed to formation of
radical intermediates because diaryl coupling was observed among the minor
products.
An early mechanistic study providing evidence for formation of OH• and NO2•
during ONOOH decomposition was published [Mahoney, L. R. J. Am. Chem. Soc.
1970, 92,1562-1563], but also apparently overlooked by later researchers.
Chemical reactivity:
direct reactions
indirect reactions
heme proteins
peroxidases
SOD
oxidations
H+
oxidations
RSH, RSeH
nucleophiles
ONOO
-
CO2
S
-
ONOOCO2
H+
NO3- + CO2
*S
ONOOH
NO3- + H+
oxidations
*
direct reactions:
— 1e- and 2e- oxidations
— stoichiometric (%100%) yields
— R = k[ONOOH][S];
(i.e., bimolecular)
indirect reactions:
— 1e- oxidations
— substoichiometric ( 40%) yields†
— R = k(ONOOH) or k[ONOO-][CO2]
(i.e., independent of S!)
†:
yield
(40%)
[S]
What is “=”? ( i.e., the reactive intermediate):
“One of the symptoms of an approaching nervous breakdown is the belief
that one’s work is terribly important.”--Bertrand Russell
Two alternative viewpoints:
(1)
Unreactive forms are geminate radical pairs; “=” represents radicals that escape
the cage:
{NO2, OH }
ONOOH
NO2 + OH
S
-H
+
NO3- + H+
+
oxidations
+H
ONOO-
CO2
{NO2, CO3 -}
ONOOCO2-
NO2 + CO3 S
-
NO3 + CO2
support:
ONO-OH
ONO-OCO2cf. HO-OH
H3C-CH3
BDE(calc.)
BDE(calc.)
BDE (exptl.)
BDE (exptl.)
20 kcal/mol
9 kcal/mol
52 kcal/mol
88 kcal/mol
peroxo O-O bonds are very weak!
oxidations
(2)
Unreactive and reactive forms are geometrical isomers:
oxidations
cis-ONOOH
"transoid"
trans-ONOOH
-
+
NO3 + H
support:
thermodynamic calculations show rate of O-O bond homolysis in ONOOH is too
slow to account for observed reaction rates (?)
Thermodynamic estimates of ONOOH homolysis rates:
ONOOH
k1
k-1
NO2 + OH
[Koppenol et al., Chem. Res. Toxicol. 1992, 5, 834-842]: k1(calc.) = 10-4-10-6 s-1
conclusion: since the experimentally determined 1st order rate constant for ONOOH
decay is kdecay = 0.8 s-1 (4-6 orders of magnitude larger than the estimated
k1, for bond homolysis), the reaction cannot occur by this pathway!
[Mereyni &Lind, Chem. Res. Toxicol. 1997, 10, 1216-1220]: k1(calc.) = 2.5 s-1
conclusion: the estimated O-O homolysis rate constant (k1) is totally consistent with the
measured kdecay.
[Koppenol & Kissner, Chem. Res. Toxicol. 1998, 11, 1216-1220]: k1(calc.) = 0.01 s-1
conclusion: perhaps a few % decay by this pathway, but not much.
What’s the problem? Why can’t our experts agree?
(1)
(2)
(3)
K = [NO2•][OH•]/[ONOOH] = k1/k-1
(k-1 w 5×109 M-1s-1)
-RTlnK = ûGo = ûfGo(NO2•)aq + ûfGo(OH•)aq - ûfGo(ONOOH)aq
ûfGo = ûfHo - TûfSo
Enthalpies:
ûfHo(ONOO-)aq = ûfHo(NO3-)aq - ûH1
ûfHo(ONOOH)aq = ûfHo(ONOO-)aq - ûH2
(ûH1 = measured heat of
isomerization, i.e.,
ONOO- NO3-)
(ûH2 = measured heat of ionization, i.e.,
ONOOH œ ONOO- + H+)
Entropies:
ONOOH(g) ONOOH(aq)
or
for which So(aq) = So(g) + ûSo, where ûSo = solvation entropy
ONOO (g) ONOO (aq)
ûSo is very difficult to estimate!
A kinetic approach:
O2 + 2NO2- + 2H+
CO2
NO3-
ONOO+H+
ONOO -
ONOOH
-H+
NO3- + H+
radical pathways for ONOOH decay (note: these exist!--the issue is whether or not they can
account for the observed product yields):
ii) NO2--mediated isomerization:
i) direct decomposition:
.
.
.
OH- + NO2
OH + NO2
ONOOH
O2 + NO2-
ONOOH+ + NO2- + .O2Net: ONOOH + ONOO-
NO2-
-
O2 + 2NO2 + H
+
ONOO
.
.
NO +
iv) N2O3-catalyzed decomposition:
.
.
ONOO-
2H+ + 2NO2O2 + NO2-
NO2 + OH-
NO2ONOOH
-
+
H + NO3
O2-
H2O
.
.
OH + NO2
Net: ONOOH
iii) NO2--mediated decomposition:
-
.
ONOOH
2H+ + NO2- + NO3-
H2O
.
NO + O2O2 + NO2-
N2O3
.
ONOO-
.
NO2 + NO2 + NO2-
.
OH + NO2
Net: ONOOH + ONOO-
-
O2 + 2NO2 + H
+
Net: 2ONOO-
40
predicted yield (----)
experimental data (O)
% O2 yield
30
20
10
0
4
6
8
10
pH
Conclusion: the radical pathway accurately predicts decomposition yields
O2 + 2NO2-
More predictions of the radical model:
NO2- inhibits O2 formation in acidic media; not in alkaline;
inhibition by organic radical scavengers will be reversed by NO2- in alkaline media;
inhibition by Fe(CN)64- will not be reversed by NO2-.
(i)
(ii)
(iii)
Prediction (i):
% O2 yield
In alkaline media, NO2• formed from reaction of
NO2- with OH• with NO• formed by ONOOdissociation to give same products as direct
decomposition; in acid NO2• reacts with another
NO2• to give net isomerization.
a
40
experimental
calculated
20
0
0
0
2
4
1 40 400
_
(mM NO2 )
pH 9.0
06
0.2 28 40
_
(mM NO2 )
pH 6.8
10
Predictions (ii) and (iii):
Organic radicals formed in
reactions with OH• decay by pathways
that do not form O2; the more reactive
NO2- effectively scavenges OH• in the
presence of these compounds,
generating O2 by pathway iii. Fe(CN)64reacts rapidly with both NO2• and OH•,
effectively quenching any further
reaction, e.g.,
% O2 yield
+ scavenger
+ scavenger + NO2
no added
scavenger
40
DMSO
benzoate
EtOH
b
t-BuOH
20
4-
Fe(CN)6
a
0
b
c
0
v) Fe(CN)64- inhibition of O2 formation:
(NO2- absent):
ONOOH
.
.
OH + NO2
(NO2--mediated):
OH- + .NO2
NO2ONOOH
Net: ONOOH + 2 Fe(CN)64-
.
.
OH + NO2
-
-
2 Fe(CN)64Fe(CN)64-
NO2- + Fe(CN)63-
Fe(CN)64-
NO2 + OH + 2 Fe(CN)6
OH- + NO2- + 2 Fe(CN)63-
NO2- + Fe(CN)63-
3-
Conclusion: OH• and NO2• radical formation during ONOOH decay is extensive. At present
no undisputed evidence exists that is inconsistent with this reaction model--[Coddington, J.W. et
al. J. Am. Chem. Soc. 1999, 121, 2438-2443].
Reactive intermediates in bicarbonate media:
How do we know that ONOO- and CO2 are the reactants?
O2 + 2NO2- + 2H+
CO2
NO3-
ONOO+
+H
ONOO-
-H+
ONOOH
NO3- + H+
-d[PN]/dt = kapparent[PN][HCO3]T
-1 -1
kapparent (M s )
3000
[PN] = [ONOO-] + [ONOOH]
[HCO3]T = [CO2] + [HCO3-]
2000
kapparent = k2/(1 + [H+]/Ka)(1 + Ka’/[H+])
or
k2'/(1+ Ka/[H+])(1 + [H+]/Ka’)
1000
0
5
6
pH
7
8
pH-rate profile indicates that reactant pairs are either (ONOOH, HCO3-) or (ONOO-, CO2).
These can be distinguished by a stopped-flow pH-jump experiment because the equilibrium:
HCO3- + H+ œ {H2CO3} H2O + CO2
is slow in neutral solutions:
[CO2]o = 0 mM
hν
HCO 3-, CO 2
acidic
[CO2]o = 0.1 mM
pH 7.4
0.8
ONOOH
alkaline
Since the reaction rate depends upon [CO2],
the isomerization reaction involves ONOOand CO2 (k2 = 3×104 M-1s-1 @ 23 oC)
[PN]t/[PN]o
ONOO -,
0.4
[CO2]o = 2.3 mM
0.0
0.00
0.04
0.08
time (s)
0.12
=
What is “ ” (the reactive intermediate) formed in the indirect reactions of
ONOOCO2-?
direct flow-EPR detection of CO3•- from reaction of ONOO- with CO2:
[Bonini et al. J. Biol. Chem. 1999, 274, 10802-10806]
Conclusion:
“Indirect” reactions occur by O-O bond homolysis in
peroxynitrite adducts (ONOOH, ONOOCO2-) to generate reactive
radicals (OH•, CO3•-, NO2•) as secondary oxidants