Evidence for isoniazid oxidation by oxyferrous mycobacterial catalase-peroxidase

Full text of DOI:10.1021/ja962047j

J. Am. Chem. Soc. 1996, 118, 11303-11304
11303
Evidence for Isoniazid Oxidation by Oxyferrous
Mycobacterial Catalase-Peroxidase
ments, a small increase in absorbance at 428 nm was observed
immediately after mixing the enzyme (0.3 µM) with excess
isoniazid (1 mM) under CO gas, consistent with reduction of
approximately 10% of the enzyme to the ferrous CO form
Richard S. Magliozzo* and Jovita A. Marcinkeviciene
8
previously characterized (λmax ) 428 nm ). When this experi-
ment was repeated using a solution of isoniazid freshly prepared
from drug recrystallized from methanol, no increase in absor-
bance at 428 nm was observed. This result suggested that a
contaminant of commercial isoniazid was able to reduce ferric
catalase-peroxidase. To confirm this result and to determine
if the reductant was derived from the drug molecule itself, a
buffered solution (pH ) 7) of isoniazid was stored for several
days at room temperature. Anaerobic treatment (under CO) of
resting catalase-peroxidase with this isoniazid solution yielded
quantitative conversion of all the ferric enzyme to the ferrous
CO form.
The aged isoniazid solution was found by HPLC to contain
isonicotinic acid. The suspected presence of hydrazine was
demonstrated spectrophotometrically by the formation of a
chromophore equivalent to that formed from authentic hydrazine
(λmax ) 458 nm) using a modified Ehrlich reagent.16 The
possibility that hydrazine reduces ferric catalase-peroxidase was
then directly confirmed. Excess hydrazine (hydrazine mono-
hydrate) added anaerobically to the ferric enzyme under CO
immediately and quantitatively produced the optical spectrum
of the ferrous CO form.
Departments of Physiology and Biophysics
and Biochemistry, Albert Einstein College
of Medicine of YeshiVa UniVersity
1
300 Morris Park AVenue, Bronx, New York 10461
ReceiVed June 17, 1996
Isoniazid (isonicotinic acid hydrazide) has been used as a
first-line antibiotic for the treatment of tuberculosis for decades,
yet the details of its mechanism of action are still incomplete.
Studies revealed many years ago that isonicotinic acid and
-pyridylmethanol were produced from the drug1 though the
,2
4
mycobacterial enzyme or enzymes responsible were not defined.
The accumulation of evidence that isoniazid-resistant strains of
mycobacteria have reduced activity of a hemoprotein hydro-
peroxidase (catalase-peroxidase)3-5 suggested that isoniazid is
6,7
a prodrug converted by this enzyme into a bacteriocidal agent.
Susceptibility to isoniazid could be produced in isoniazid-
4
resistant Escherichia coli or Mycobacteria smegmatis upon
introduction and expression of the M. tuberculosis katG gene
encoding the catalase-peroxidase. The purified catalase-
8
9-12
Experiments were designed to test the hypothesis that the
reduction of ferric catalase-peroxidase under aerobic conditions
leads to production of the oxyferrous enzyme, which could
function as an oxidant of isoniazid. Incubations of catalase-
peroxidase were prepared with isoniazid, with and without
hydrazine under aerobic and anaerobic conditions. The addition
of hydrazine (15 µM) to resting catalase-peroxidase (3 µM)
in the presence of 150 µM isoniazid led to the disappearance
of 44 µM isoniazid in 2.5 h (Table 1). Isonicotinic acid and
isonicotinamide accounted nearly quantitatively for the con-
peroxidase from M. smegmatis (and other bacteria
) contains
ferric heme according to optical and/or EPR spectrosopic
analyses and catalyzes classical peroxidative reactions. The M.
tuberculosis katG enzyme produces radicals in the presence of
13
isoniazid and H2O2. It is therefore unusual that M. tuberculosis
catalase-peroxidase was reported to catalyze isoniazid oxidation
in Vitro without peroxide activation of the enzyme.14 Experi-
mental results reported here suggest that the isoniazid oxidation
mechanism involves oxyferrous catalase-peroxidase.
In the first experimental approach, optical spectroscopy was
used to follow the reaction between isoniazid and resting M.
smegmatis catalase-peroxidase. In the second, HPLC was used
to quantitate the majority products in reactions catalyzed by
the enzyme in the presence of hydrogen peroxide or hydrazine,
a reducing agent.
1
4
verted drug. No reaction occurred in an identical reaction
mixture incubated under CO, demonstrating the requirement for
oxygen and suggesting that the oxyferrous form of the enzyme
was the oxidant formed under aerobic conditions. No reaction
occurred using freshly dissolved, recrystallized isoniazid under
aerobic conditions in the absence of hydrazine, consistent with
the spectrophotometric results. No oxidation of the drug
occurred in aerobic incubations with hydrazine in the absence
of catalase-peroxidase.
The formation of a catalytically active enzyme from resting
(
ferric) M. smegmatis catalase-peroxidase in the absence of
exogenous peroxide may occur as a result of iron reduction and
binding of O2 to give the oxyferrous form. In initial experi-
8
In an earlier report it was demonstrated that o-dianisidine is
oxidized peroxidatively by M. smegmatis catalase-peroxidase,
(
(
(
1) Youatt, J. Am. ReV. Tuberc. 1958, 78, 806-812.
-
1
-1
2) Youatt, J. Aust. J. Chem. 1961, 14, 308-314.
yielding a red chromophore (ꢀ460 nm ) 11 mM cm ). This
reaction was used to check for the production of hydrogen
peroxide in situ (from autooxidation of hydrazine, as suggested
by a reviewer) that could potentiate peroxidation of isoniazid
by the enzyme. No peroxidation was detected optically during
3) Sriprakash, K. S.; Ramakrishnan, T. J. Gen. Microbiol. 1970, 60,
25-132.
1
3
1
(
4) Zhang, Y.; Heym, B.; Allen, B.; Young, D.; Cole, S. Nature 1992,
58, 591-593.
(
5) Heym, B.; Alzari, P. M.; Honore, N.; Cole, S. T. Mol. Microbiol.
995, 15, 235-245.
6) Shoeb, H. A.; Bowman, B. U., Jr.; Ottolenghi, A. C.; Merola, A. J.
Antimicrob. Agents Chemother. 1985, 27, 399-403.
7) Shoeb, H. A.; Bowman, B. U., Jr.; Ottolenghi, A. C.; Merola, A. J.
Antimicrob. Agents Chemother. 1985, 27, 404-407.
8) Marcinkeviciene, J. A.; Magliozzo, R. S.; Blanchard, J. S. J. Biol.
2
h incubations of 3 µM enzyme with 140 µM o-dianisidine
(
and 14 or 28 µM hydrazine under conditions in which as little
as 0.05 µM product would have been detected. Complete
peroxidation of the substrate could be demonstrated in this
reaction mixture upon addition of excess peroxide after the 2 h
incubation, demonstrating that the enzyme remained active.
Experiments were designed to elicit peroxidative breakdown
of isoniazid by catalase-peroxidase by adding hydrogen
peroxide to the enzyme (3 µM) in solutions containing 150 µM
isoniazid in one or several aliquots, to final concentrations in
molar excesses ranging from 100-fold to greater than 5000-
fold over that of enzyme. Although low yields were found
(
(
Chem. 1995, 270, 22290-22295.
(
(
9) Hicks, D. B. Biochim. Biophys. Acta 1995, 1229, 347-355.
10) Yumoto, I.; Fukumori, Y.; Yamanaka, T. J. Biochem. 1990, 108,
5
2
83-587.
1
7
(
11) Hochman, A.; Goldberg, I. Biochim. Biophys. Acta 1991, 1077,
99-307.
(
12) Youn, H.-D.; Yim, Y.-I.; Kim, K.; Hah, Y. C.; Kang, S.-O. J. Biol.
Chem. 1995, 270, 13740-13747.
(
13) Hillar, A.; Loewen, P. C. Arch. Biochem. Biophys. 1995, 323, 438-
4
7
46.
(
14) Johnsson, K.; Schultz, P. G. J. Am. Chem. Soc. 1994, 116, 7425-
426.
(16) Toida, I. Am. ReV. Respir. Dis. 1962, 85, 720-726.
(17) No optical evidence for compound I formation was found on addition
of either very small or large excesses of hydrogen or alkyl peroxide to
catalase-peroxidase.
8
(
15) A modification of a previously published procedure was used to
purify the enzyme, which had an optical ratio A408/A280 ) 0.56 and specific
activity greater than 500 units/mg.
S0002-7863(96)02047-1 CCC: $12.00 © 1996 American Chemical Society

11304 J. Am. Chem. Soc., Vol. 118, No. 45, 1996
Communications to the Editor
Table 1. Isoniazid Breakdown into Isonicotinic Acid and
Isonicotinamide Catalyzed by Mycobacteria smegmatis
Catalase-Peroxidasea
Scheme 1
isonicotinic isonicotin-
isoniazid
acid, µM
amide, µM reacted, µM
0
15
0
0
17
5
5
44
(air)b
(CO)
4
c
4
0
0
0
30
15
25/70
15
0
30/70
100
0
0
f
0/15
20
h
in extensive decomposition of the drug in the absence of
catalase-peroxidase (Table 1). Isoniazid was nearly quanti-
tatively converted into isonicotinic acid and isonicotinamide after
an additional 2.5 h incubation in the presence of catalase-
peroxidase. These observations indicate that nonenzymatic
114
a
Reactions initially contained 3 µM heme plus 150 µM isoniazid
added from a freshly prepared stock solution (10 mM in water) in 15
mM potassium phosphate buffer, pH ) 7. Products were analyzed by
reversed phase HPLC chromatography after 2.5 h incubation (or at the
indicated times). Enzyme (E) was removed at the end of the incubations
by ultrafiltration (Centricon 30), and the HPLC analysis was performed
on the retentate. No other reagents were added and compounds were
quantitated on the basis of the areas of peaks measured in chromato-
grams of freshly prepared standard solutions of isonicotinic acid,
isonicotinamide and isoniazid in 15 mM potassium phosphate buffer.
The unreacted isoniazid and the two breakdown products quantitated
here were resolved by elution with the aqueous phase (50 mM
ammonium acetate, pH ) 7). An acetonitrile gradient eluted additional
material that was not identified or quantitated since most of the lost
isoniazid could be accounted for by the yield of isonicotinic acid and
isonicotinamide. Some loss of isoniazid occurs during manipulation
of samples and due to binding of isoniazid to the enzyme. This fraction
was subtracted from the starting concentration to calculate amounts
2
+
Mn -catalyzed reactions of isoniazid produce good yields of
the same products as the enzyme-catalyzed reactions.
The evidence that oxyferrous M. smegmatis catalase-
peroxidase may be catalytically competent as an oxidant of
isoniazid, as outlined in Scheme 1, may be related to oxidations
25-27
catalyzed by oxyperoxidase
and to the catalytic mechanism
28
of the enzyme tryptophan 2,3-dioxygenase. Analogy may also
exist with the mechanism of reduction of methemoglobin by
phenylhydrazine and its reaction with oxyhemoglobin.
2
9
30,31
The antimycobacteriocidal action of isoniazid has a known
3
2,33
requirement for oxygen in cell cultures.
That observation,
b
reacted. Determinations are within (10 µM. Hydrazine concentration,
along with the present results, may be especially relevant to
the mechanism of action of the drug since the level of peroxide
in cultured mycobacteria may never be appropriate for initiation
of peroxidative reactions involving catalase-peroxidase and
isoniazid. The presence of a hydrazidase activity in mycobac-
c
1
5 µM. The solution of enzyme plus isoniazid was degassed and
equilibrated under CO before the anaerobic addition of 15 µM
d
hydrazine. Conditions: 15 µM hydrazine, 150 µM isoniazid, no
e
enzyme. Hydrogen peroxide (16 mM) was added in one aliquot, and
f
the reaction was incubated for 1.5 h. Hydrogen peroxide was added
g
16
in 5 aliquots over a 2 h period. Isoniazid (150 µM) was incubated
teria leads to the possibility that, in ViVo, the bona fide oxidant
2
+
h
with 2 µM Mn ; entries are for analyses after 4 and 6.5 h. After 4
of isoniazid is oxyferrous catalase-peroxidase generated upon
hydrazinolysis of the drug and reaction of hydrazine with the
ferric enzyme under aerobic conditions.
2
+
h preincubation of isoniazid with Mn , enzyme (3 µM) was added to
the reaction mixture followed by incubation for 2.5 h.
based on the concentration of peroxide added (Table 1, selected
results) the cleavage of the drug may suggest that a complex
Acknowledgment. This work was supported by NIH grants
GM40168, RR02583, GM33449, and AI33696. The authors thank Drs.
Jack Peisach and John Blanchard for helpful discussions.
18,19
analogous to horseradish peroxidase compound I
is active
as an oxidant. However, the oxyferrous form of catalase-
peroxidase assumed to be the catalytically competent form of
the enzyme generated in the presence of hydrazine, may also
be generated in the presence of high concentrations of peroxide.
In support of this suggestion is the evidence that large excesses
of H2O2 favor the formation of compound III (oxyperoxidase)
JA962047J
(
22) Quemard, A.; Dessen, A.; Sugantino, M.; Jacobs, W. R., Jr.;
Sacchettini, J. C.; Blanchard, J. S. J. Am. Chem. Soc. 1996, 118, 1561-
562.
(23) Quemard, A.; Sacchettini, J. C.; Dessen, A.; Vilcheze, C.; Bittman,
1
1
8-20
from resting (ferric) horseradish peroxidase.
R.; Jacobs, W. R.; Blanchard, J. B. Biochemistry 1995, 34, 8235-8241.
(24) Ito, K.; Yamamoto, K.; Kawanishi, S. Biochemistry 1992, 31,
11606-11613.
2
+
The use of Mn has been reported in experiments demon-
strating inhibition of the mycobacterial fatty acyl-ACP (acyl
(
25) Tamura, M.; Yamazaki, I. J. Biochem. 1972, 71, 311-319.
carrier protein) enoyl reductase (inhA gene product) by isoniazid
(
26) Makino, R.; Yamada, H.; Yamazaki, I. Arch. Biochem. Biophys.
with catalase-peroxidase,2
1,22
and inhibition of the reductase
1
976, 173, 66-70.
has been implicated in the bacteriocidal mechanism of the
Mn has also been reported to catalyze the aerobic
decomposition of isoniazid in a radical-mediated mechanism.
(27) Yokota, K.; Yamazaki, I. Biochem. Biophys. Res. Commun. 1965,
2
2,23
2+
18, 48-53.
(
drug.
28) Ishimura, Y.; Nozaki, M.; Hayishi, O.; Nakamura, T.; Tamura, M.;
2
4
Yamazaki, I. J. Biol. Chem. 1970, 245, 3593-3602.
2+
Here, incubation of isoniazid (150 µM) with 2 µM Mn resulted
(29) Rostorfer, H. H.; Totter, J. R. J. Biol. Chem. 1956, 221, 1047-
1
055.
(
18) Chance, B. Arch. Biochem. Biophys. 1952, 41, 404.
(30) Goldberg, B.; Stern, A.; Peisach, J. J. Biol. Chem. 1976, 251, 3045-
(
19) Blumberg, W. E.; Peisach, J.; Wittenberg, B. A.; Wittenberg, J. B.
3051.
J. Biol. Chem. 1968, 243, 1854-1862.
(31) Augusto, O.; Kunze, K. L.; Ortiz de Montellano, P. R. J. Biol. Chem.
(
(
20) Nakajima, R.; Yamazaki, I. J. Biol. Chem. 1987, 262, 2576-2581.
21) Johnsson, K.; King, D. S.; Schultz, P. G. J. Am. Chem. Soc. 1995,
1982, 257, 6231-6241.
(32) Mitchison, D. A.; Selkon, J. B. Am. ReV. Tuberc. 1956, 74, 109.
(33) Youatt, J. Am. ReV. Respir. Dis. 1969, 99, 729-749.
1
17, 5009-5011.