Redox buffering of hypochlorous acid by thiocyanate in physiologic fluids

Article abstract of DOI:10.1021/ja0438361

The major antimicrobial products of neutrophilic myeloperoxidase (MPO) in physiologic fluids are hypochlorous acid (HOCl) and hypothiocyanite (OSCN^-), and the former is generally believed to be the killing agent. However, we have determined that HOCl oxidizes SCN^- in a facile nonenzymic reaction. The observed kinetics and computational models substantiate the hypothesis that SCN^- serves to moderate the potential autotoxicity of HOCl by restricting its lifetime in physiologic fluids. Furthermore, the oxidizing equivalents of HOCl are preserved in OSCN^-, a more discriminate biocide that is not lethal to mammalian cells. Copyright

Full text of DOI:10.1021/ja0438361

Published on Web 11/19/2004
Redox Buffering of Hypochlorous Acid by Thiocyanate in Physiologic Fluids
Michael T. Ashby,* Amy C. Carlson, and M. Jared Scott
Department of Chemistry and Biochemistry, UniVersity of Oklahoma, 620 Parrington OVal,
Norman, Oklahoma 73019
Received October 9, 2004; E-mail: mashby@ou.edu
Neutrophils, which comprise 33-75% of all leukocytes in
humans, possess an impressive armory of oxidative and non-
oxidative mechanisms for combating microorganisms. The oxidative
defense mechanism of neutrophils is based upon NADPH oxidase
and myeloperoxidase (MPO). The NADPH oxidase system reduces
molecular oxygen to generate reactive oxygen species (ROS),
SCN- (i.e., pseudo-first-order conditions) yield traces at 300 nm
-
(near λmax for OCl ) that fit single-exponential kinetics models
-
between 500 µM < [OH ] < 1.03 M, thus suggesting first-order
-
dependence on [OCl ]. The reaction rates were also first-order with
-
+
respect to 1/[OH ]
0
(i.e., first-order with respect to [H ]) over three
-
decades of change in [OH ] (Figure 1) and first-order with respect
-
-
including H
2
O
2
. In the presence of H
2
O
2
, MPO is capable of
to [SCN ] (Supporting Information). Production of OSCN as the
primary oxidation product was indicated by an increase in absor-
bance at 240 nm and further confirmed by employing a double-
-
oxidizing all of the halides (except F ), as well as the pseudohalide
-
thiocyanate (SCN ), to produce hypohalites that are many orders
of magnitude more cytotoxic than H
2
O
2
itself. Hypohalites, and in
mixing stopped-flow sequence to convert the product to (SCN)
2
-
-
particular hypochlorous acid (HOCl), appear to play a pivotal role
via comproportionation of OSCN and SCN (Supporting Informa-
1
15
in inflammation. When unchecked, neutrophil-induced inflamma-
tion). These data are consistent with a facile proton equilibrium
tion results in local tissue damage, and MPO has been linked to
to generate HOCl (eq 3), followed by a rate-limiting reaction (eq
2
3
-
-
7
-1
numerous chronic diseases, such as atherosclerosis, cystic fibrosis,
4) of HOCl with SCN to yield OSCN (k
3
) 2.34(9) × 10 M
4
-1
and periodontitis. Since HOCl is an indiscriminant oxidant, whereas
s ). Considering only this reaction, we estimate at mean concentra-
-
5-9
-
hypothiocyanite (OSCN ) is not lethal to mammalian cells,
considerable attention has focused on the substrate selectivity of
tions of SCN that are found in saliva (1-3 mM) and plasma (20-
120 µM), the half-life of HOCl cannot be more than ca. 15 µs and
400 µs, respectively.
MPO.1
0,11
We demonstrate here that HOCl is capable of rapidly
oxidizing SCN to give OSCN . Thus, the nonenzymic transfer of
oxidizing equivalents from HOCl to SCN substantiates the
hypothesis that SCN can serve the role of a redox buffer, thereby
-
-
While the preceding discussion of the half-lives of HOCl in a
-
-
medium that contains SCN is illuminative, it does not address
-
the issue of steady-state concentrations of HOCl in physiologic
fluids, nor the potential physiological consequences of eq 4. To
broach these topics, it is necessary to consider dynamic models.
governing the lifetime of the more powerful oxidant HOCl and its
potential for host self-destruction.
MPO is in the ferric form in its resting state. Rapid reaction
1
1
Such models can be developed because all of the rate constants
for eqs 1-4 have been measured by us and others.1 Although it
6,17
2 2
with H O produces compound I (MPO-I), which is two oxidizing
equivalents above MPO:
is not our intention in this Communication to provide a complete
model for the redox buffering of HOCl by SCN , Figure 2 illustrates
-
k1
the consequence of the k
of MPO (1 µM) and a high concentration of H
3
pathway for a relatively low concentration
(1 mM),
MPO + H O y z MPO-I
(1)
(2)
2
2
k-1
2
O
2
conditions that are conceivable for extracellular MPO during a
k2
-
-
-
respiratory burst. The initial rates of reaction of MPO-I with X
MPO-I + X
98 MPO + OX
determine the relative amount of the two hypohalites that are
The substrate specificities of MPO are determined by the rates of
reaction of MPO-I with X . Because of the relative abundance
initially formed (not illustrated on the time-scale of Figure 2):
-
12
-
x
-
of halides, it is often suggested that the chief substrate of MPO in
k [X ]
-
2
-
blood plasma is Cl , and that the HOCl which is produced is
%X )
(5)
-
-
Cl
2
-
SCN
2
[SCN-]
responsible for the cytocidal capacity of the MPO system.13
However, we observe that the nonenzymic rate of oxidation of
k
[Cl ] + k
-
-
-
-
SCN by HOCl to give hypothiocyanite (OSCN ) is nearly
diffusion-controlled,14 thus suggesting that the major hypohalite
which equals 72% OCl and 28% SCN for typical blood plasma
-
-
18
concentrations ([Cl ]
0
) 100 mM and [SCN ] ) 100 µM).
0
-
produced by MPO in the absence of efficient reductants is OSCN :
However, the relative amounts of these hypo(pseudo)halites change
-
as the HOCl that is produced in eq 2 reacts with SCN via eq 4.
k
-a
-
+
The concentration-time traces of Figure 2 were computed using
rate equations that were programmed into Mathematica. The
simultaneous differential equations were solved by numerical
methods since the complexity of the equations forbids the develop-
ment of closed solutions (Supporting Information). No assumptions
were made to simplify the rate laws except that the Brønsted acid-
base chemistry is facile with respect to all other kinetic processes,
OCl + H y z HOCl
(3)
ka
k3
-
-
-
+
HOCl + SCN
98 Cl + OSCN + H
(4)
The rate of the reaction of HOCl and SCN- (eq 4) is too fast at
physiologic pH to measure by stopped-flow. However, the equili-
-
-
brium of eq 3 is driven to OCl under basic conditions (pK
a
(HOCl)
and that the concentration of Cl was assumed to remain constant
)
7.4), thus sufficiently slowing reaction 4 to facilitate measure-
by virtue of its high concentration. Dynamic changes in concentra-
tion were taken into consideration for all other species. Note that
ment. Single-mixing stopped-flow reactions of HOCl with excess
15976
9
J. AM. CHEM. SOC. 2004, 126, 15976-15977
10.1021/ja0438361 CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. . Predicted Partitioning of HOCl in Human Blood Plasma
by the Components of the Low MW Fraction that are Expected to
Exhibit a Significant Rate of Reaction
(M- s-1)a
1
% HOCl
component
concn (M)
k
3
′
-
-
-
-
-
-
-
7
7
Cys
Met
His
3 × 10
5 × 10
2 × 10
4 × 10
2 × 10
7 × 10
7 × 10
1 × 10
3 × 10
4 × 10
1 × 10
5 × 10
1 × 10
2 × 10
5 × 10
2 × 10
0.05%
0.01%
8
5
5
5
8
8
7
5
3
4
6
5
7
0.01%
Lys
<0.01%
<0.01%
<0.01%
<0.01%
99.93%
Trp
AA
taurine
SCN
-
-3
a
-
Value k3′ is defined in eq 4 where SCN is replaced with the
component.
of 35-55 g/L (about 750 µM). HSA bears HOCl reactive groups
(e.g., 12 Met, 2 Cys, 67 S-S).
-
-
Figure 1. Plot of keff ([SCN ]0 ) 10 mM, [OCl ]0 ) 0.1 mM, µ ) 1) as
-
-1
Experiments are underway to explore the competitiveness of the
a function of [OH ] illustrating first-order dependence. The data are fit
to the linear function: keff ) (9.85 × 10 M s )/[OH ] + 1.27 × 10
-
2
-1
-
-2
-
various components of blood plasma with respect to SCN .
-
1
s
.
-
It has been previously suggested that the SCN reaction pathway
of MPO might serve as a means of ameliorating self-destruction
by the more powerful reactant HOCl by acting as a redox buffer,
but no specific theories were advanced that explained how the
enzyme might regulate the two substrate pathways.1 The nonen-
6
-
zymic transfer of the oxidizing equivalents of HOCl to SCN that
are described here serves such a role by governing the lifetime of
the more powerful oxidant, thereby moderating the potential
autotoxicity of HOCl. Furthermore, the oxidizing equivalents of
-
22
HOCl are preserved in OSCN , a more discriminate biocide that
is not lethal to mammalian cells.
Acknowledgment. We are grateful to the PRF (35088-AC3)
and the OCAST (HR02-019) for their financial support.
Supporting Information Available: Experimental details. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Hereafter, “HOCl” refers to the equilibrium mixture of HOCl and OCl-,
-
Figure 2. Concentration-time curves for HOCl (solid line) and OSCN
dashed line), with and without consideration of the k3 pathway.
about 1:1 at pH ) 7.4.
(
(
(
(
2) Podrez, E. A.; Abu-Soud, H. M.; Hazen, S. L. Free Radical Biol. Med.
2000, 28, 1717-1725.
2 2
in reality, such high concentrations of H O would deactivate MPO
3) Conese, M.; Copreni, E.; Di Gioia, S.; De Rinaldis, P.; Fumarulo, R. J.
Cystic Fibrosis 2003, 2, 129-135.
on the millisecond time scale by formation of halide-inactive
compound II.19 Nonetheless, Figure 2 clearly illustrates the redox-
4) Altman, L. C.; Baker, C.; Fleckman, P.; Luchtel, D.; Oda, D. J. Periodontal
Res. 1992, 27, 70-79.
-
buffering effect of SCN ; substantial quantities of HOCl are not
(5) Carlsson, J.; Edlund, M. B.; Haenstroem, L. Infect. Immun. 1984, 44,
581-586.
-
produced until the complete oxidation of SCN . Figure 2 also
(
6) White, W. E., Jr.; Pruitt, K. M.; Mansson-Rahemtulla, B. Antimicrob.
illustrates another significant aspect of the k
in the rate of production of OSCN over that from the k
alone. Whether this increased rate via the k
3
pathway, an increase
pathway
pathway has physi-
ological consequence remains to be demonstrated.
Agents Chemother. 1983, 23, 267-272.
-
(7) Bjoerck, L.; Claesson, O. J. Dairy Sci. 1980, 63, 919-922.
2
(
8) Marshall, V. M.; Reiter, B. J. Gen. Microbiol. 1980, 120, 513-516.
3
(9) Carlsson, J. J. Oral Path. 1987, 16, 412-416.
10) van Dalen, C. J.; Whitehouse, M. W.; Winterbourn, C. C.; Kettle, A. J.
Biochem. J. 1997, 327, 487-492.
(
-
7
-1 -1
, k ) 58 s-1, k
-1
-
4
-1 -1
We propose that the relative amounts of HOCl and OSCN that
are produced in vivo are not dictated by enzyme substrate
selectivities alone but rather by kinetic competition between the
(11) k
1
k
) 2.3 × 10
M
s
2
(Cl ) ) 2.5 × 10
M
s ,
7
(SCN ) ) 9.7 × 10 M s-1, K
-
6
-1
) 4.0 × 10 M, k ) 2.3 × 10 M
-8
-1
2
a
3
-
1
s
.
-
(12) Rate ) k [MPO-I][X
2
].
(13) Roos, D.; Winterbourn, C. C. Science 2002, 296, 669-671.
14) The bimolecular rate constant for a diffusion-controlled reaction of two
uncharged particles in water is about 8 × 10
(15) Barnett, J. J.; Stanbury, D. M. Inorg. Chem. 2002, 41, 164-166.
16) Furtmueller, P. G.; Burner, U.; Obinger, C. Biochemistry 1998, 37, 17923-
7930.
-
rate of reaction of HOCl with SCN and other reductants. While
(
physiologic fluids are very complicated and we cannot expect to
9
M
-1
s .
-1
-
accurately predict a priori whether the reaction of SCN with HOCl
(
is competitive in vivo, we can begin to probe this issue because
normal reference values are known for plasma components,20 and
1
(17) Furtmuller, P. G.; Obinger, C.; Hsuanyu, Y.; Dunford, H. B. Eur. J.
Biochem. (FEBS) 2000, 267, 5858-5864 and references therein.
the rate constants for reaction of many of these species with HOCl
are available.21 When the low molecular weight components of
human blood plasma are taken into consideration, assuming these
species are in excess with respect to HOCl and partitioning is
dictated by initial rates, we arrive at the remarkable conclusion that
(
18) This result is consistent with previous separate rate measurements of MPO-
catalyzed oxidation of Cl and SCN- and the observation that combined
-
solutions of Cl- and SCN- exhibit rates for loss of H
2
2
O that are simply
-
-
the weighted average of the individual rates for Cl and SCN for similar
concentrations (ref 10).
(
19) Jantschko, W.; Furtmuller, P. G.; Zederbauer, M.; Lanz, M.; Jakopitsch,
C.; Obinger, C. Biochem. Biophys. Res. Commun. 2003, 312, 292-298.
-
SCN would consume nearly all available HOCl (Table 1 and
(20) Duh, S.-H.; Cook, J. D. In Laboratory Reference Values, 27th ed.;
McDonough, J. T., Ed.; Williams & Wilkins: Baltimore, MD, 2000; pp
Supporting Information). However, there are many macromolecular
components of plasma that are likely to compete effectively for
HOCl. For example, human serum albumin (HSA) is the main
protein component of plasma with a normal concentration range
2074-2090.
(21) Pattison, D. I.; Davies, M. J. Chem. Res. Toxicol. 2001, 14, 1453-1464.
(22) Ashby, M. T.; Aneetha, H. J. Am. Chem. Soc. 2004, 126, 10216-10217.
JA0438361
J. AM. CHEM. SOC.
9
VOL. 126, NO. 49, 2004 15977