The biomimetic oxidation of dieldrin using polyhalogenated metalloporphyrins

Article abstract of DOI:10.1039/a809080g

Biomimetic oxidation of dieldrin produces the same metabolites as generated in vivo, which suggest a 'radical oxygen-rebound' mechanism.

Full text of DOI:10.1039/a809080g

The biomimetic oxidation of dieldrin using polyhalogenated metalloporphyrins
Fumio Hino and David Dolphin*
Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, B.C., Canada V6T 1Z1.
E-mail: ddolphin@qlt-pdt.com
Received (in Corvallis, OR, USA) 19th November 1998, Accepted 4th January 1999
Biomimetic oxidation of dieldrin produces the same metabo-
lites as generated in vivo, which suggest a ‘radical oxygen-
rebound’ mechanism.
The general concensus is that the majority of P-450 mediated
hydroxylations proceed via radical rearrangements. Radical
rearrangements have been well-documented in P-450 bio-
6
chemistry and ‘radical clocks’ have been used to measure the
Dieldrin 1, like other insecticides, accumulates in the food chain
and its use, as with other organochlorinated materials such as
DDT, has been banned in many countries. Dieldrin accumulates
principally in adipose tissue and is only slowly metabolized.
Even though it has not been used in North America for more
rate of the rebound process.12 Thus bicyclo[2.1.0]pentane gave
a 7:1 mixture of endo-2-hydroxybicyclo[2.1.0]pentane and
1
3
cyclopent-3-en-1-ol when oxidized by rat liver microsomes.
The bicyclo[2.1.0]pentan-2-yl radical rearranges to the cyclo-
9
21 14
pent-3-en-1-yl radical with a rate constant of 2.4 3 10 s
.
1
15
than two decades it is, to this day, still found in the food chain.
In addition, Groves et al. have shown that when a chiral
iron porphyrin hydroxylated ethylbenzene, the differences
observed between the pro-R and pro-S hydrogen atoms were
best explained via radical intermediates. Nevertheless, New-
comb, Hollenberg and their colleagues16 have recently con-
cluded, from their studies on a hypersensitive cyclopropyl
radical probe, that ring opening of a cyclopropylmethyl radical
is unreasonable in their system and that a cationic species is
In the laboratory it was found that one year after injection of
dieldrin only ~ 50% of the material had been metabolized and
2
excreted, while half the dose was still retained by rabbits. Some
of the less interesting dieldrin metabolites arose from opening
of the epoxide ring, but one particular metabolite, the intra-
molecularly bridged pentachloro ketone 6, allowed for some
very interesting speculation as to its mechanism of formation.
3
McKinney et al. suggested that the other principle mammalian
metabolite, syn-9-hydroxydieldrin 3, might be an intermediate
in the formation of 6, but they eventually considered this
unlikely since feeding of 3 to a rat produced no pentachloro
ketone in its liver, where it is known to accumulate. It was
Cl
H
O
O
Cl
4
suggested, in 1975, that a common intermediate, the species
Cl
Cl
containing a radical at C-9 (2, Scheme 1), could account for the
formation of both 3 and 6; this suggestion is consistent with
Groves’ oxygen-rebound mechanism for the hydroxylation of
Cl
6
[
O]
alkanes by cytochrome P-450 which was formulated in
O
FeIV–Por+•
Fe –Por
III
,6
1
976.5
'
Compound I'
+
We have been unsuccessful in oxidizing dieldrin using
Cl
3
H
H
8
Cl
H
isolated liver microsomes, due principally to its solubility
properties and the destruction of microsomes under forcing
conditions. We also failed in oxidizing dieldrin using metal-
loporphyrins derived from tetrakis(2,6-dichlorophenyl)por-
phyrin as biomimetic catalysts since dieldrin is quite recalcitrant
towards oxidation and the porphyrin catalysts were destroyed
before sufficient substrate oxidation could occur.
9
Cl
Cl
2
5
4
6
HO
O
Cl
O
1
Cl
Cl
Cl
8
Cl
Cl
Cl
7
Cl
1
5
Recently, however, new highly halogenated porphyrins
containing eight halogen atoms on the porphyrin periphery have
been reported and these third generation catalysts (10, 11) are
showing exciting potential as biomimetic and industrial cata-
III
Fe –Por
Cl
H
•
OH Cl
H
•
Cl
Cl
•
7
lysts. The oxidizing power of these catalysts and the turnover
O
Cl
O
numbers that they exhibit are extraordinary, and both the iron
and manganese complexes 10 and 11 effectively oxidize
dieldrin to give the same products (3, 6) as those previously
identified from mammalian metabolism studies.
Cl
Cl
Cl
Cl
Cl
Cl
Cl
2
4
••
•
III
+•
•
III
HO–Fe –Por
HO–Fe –Por
•
••
We have examined a number of cooxidants in addition to
9
those shown in Table 1 and found that Oxone (KHSO
5
) and
Cl
O
H
OH
Cl
H
O
2 2
caused considerable destruction of catalyst and, at best,
III
Cl
+
Fe –Por
gave only a trace of oxidized dieldrin. On the other hand,
surprisingly good yields of 3 and 6 were observed using
hypochlorite, iodosylbenzene (PhIO) and alkyl hydroperoxides
in mixed aqueous/organic solvents; small amounts of the ketone
Cl
O
Cl
O
Cl
Cl
Cl
Cl
Cl
Cl
Cl
7
were also seen when the reaction was carried out in only
3
7
organic solvents. Ketone 7 arises from the over-oxidation of 3;
when 3 was treated with 10a and PhIO, 7 was isolated from the
reaction in 24% yield. Similar overoxidation of an initially
formed alcohol regularly occurs with these P-450 mimics, but
the extent of overoxidation can be often controlled by choice of
catalyst, cooxidant and reaction conditions.11
H+
6
Scheme 1
Chem. Commun., 1999, 629–630
629

catalytic cycle) or rearrange to the more stable tertiary radical at
C-2 (4) where transfer of an hydroxyl radical from iron followed
by loss of HCl would generate 6.
The fact that we have been unable to bring about the acid
catalyzed conversion of 3 ? 6 does not preclude cationic
intermediates in the P-450 metalated hydroxylation of dieldrin.
Nevertheless, we conclude that the present study favors radical
intermediates in the biomimetic oxidation of dieldrin and by
analogy in the enzymatic reaction as well. Perhaps we should
not be too surprised in either this case or that of Newcomb and
Hollenberg et al. that electronically stacking a particular
substrate may impose different reaction pathways for P-
4
50-mediated hydroxylations.
As shown previously7 these highly chlorinated metal-
loporphyrins can accurately mimic the cytochromes P-450. Not
only can the production of natural metabolites be mimicked
(
and presumably predicted) but milligram (and even larger if
required) amounts of such metabolites can be produced at the
bench.
We thank the Canadian Natural Sciences and Engineering
Research Council of Canada and Meiji Pharmaceutical Uni-
versity, Tokyo, Japan.
Table 1 Oxidation of dieldrin (ref. 8)
Product (%)b
Notes and references
Ratio
6/(3 + 7)
1
G. D. Bellward, R. J. Norstrom, P. W. Whitehead, J. E. Elliot, S. M.
Bandiera, C. Dwonschak, T. Chang, S. Forbes, B. Cadario, L. E. Hart
and K. M. Cheng. J. Toxicol. Environ. Health, 1990, 30, 33.
F. Korte and H. Arent. Life Sci., 1965, 4, 2017.
Catalyst Co-oxidant Solventa
6
3
7
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
0
NaOCl
NaOCl
C
D
C
D
A
B
B
A
B
A
B
A
B
A
B
3
7
5
25
19
5
2
65
0.33
2.4
0.64
1.9
19
10
2
3
12
16
36
24
31
6
38
43
12
15
8
trace 73
47
trace 30
—
—
J. D. McKinney, H. B. Mathews and N. K. Wilson, Tetrahedron Lett.,,
t
Bu OOH
4
1
973, 21, 1895; J. D. McKinney, S. M. DePaul Palaszek, H. B. Mathews
t
Bu OOH
and M. H. Behandale, in Pesticides in the Environment. A Continuing
Controversy, ed. W. B. Deichmann, Intercontinental Medical Book
Corp., New York, 1973, p. 103.
Foreign Compound Metabolism Vol. 3, A Specialist Periodical Report,
Senior Reporter D. E. Hathway, The Chemical Society, London, 1975,
p. 404.
J. T. Groves and G. A. McClusky, J. Am. Chem. Soc., 1976, 98, 859;
J. T. Groves, G. A. McClusky, R. E. White and M. J. Coon, Biochem.
Biophys. Res. Commun., 1988, 81, 154.
1a
1a
1a
1a
1a
1b
1b
1b
1b
1b
1b
NaOCl
NaOCl
PhIO
63
51
3
trace trace 70
4
5
6
7
t
Bu OOH
14
12
3
—
—
—
—
—
—
—
—
—
30
32
77
60
81
80
40
36
18
2.7
3.6
4
t
Bu OOH
NaOCl
NaOCl
PhIO
8
1.9
trace
trace
10
8
PhIO
11
31
41
8
J. T. Groves and Y.-Z. Han, Models and Mechanisms of Cytochrome P-
t
Bu OOH
3.1
5.1
0.6
4
50 Action, in Cytochrome P-450, 2nd edn., ed. P. Ortiz de Montellano,
t
Bu OOH
1
995, pp. 1–45.
c
Fenton
H O
2 2
MeCN
13
D. Dolphin, T. Traylor and L.Y. Xie, Acc. Chem. Res., 1997, 30, 251.
a
8 The cooxidant (0.66 mmol) was added with stirring, at room
temperature, in ten equal portions every 30 min to a solution of the
porphyrin catalyst (0.032 mmol) in 10 ml of solvent (A, B, or C). An
hour after the last addition of cooxidant the mixture was poured onto
A: H
2
O–CH
2
Cl
2
–MeOH = 2:10:30. B: pH 2 buffer–CH
2
Cl
2
–MeOH =
–MeOH = 1:1. Ref. 9. c Ref. 10.
b
2
2 2 2 2
:10:30. C: CH Cl . D: CH Cl
intermediate in the rearrangements. In order to best accom-
modate their observations they suggest that the active oxidant in
P-450 hydroxylations might be an iron–H O complex and that
2 2
water (50 ml) and CH
was extracted with CH
were dried over MgSO
applied to a thick layer silica TLC plate and developed with light
petroleum–EtOAc (5:1).
2
Cl
Cl
, filtered and concentrated to 1 ml which was
2
(50 ml). After separation the aqueous phase
2
2
(2 3 20 ml). The combined organic phases
4
the first species formed in the hydroxylation reaction is a
+
protonated alcohol formed by ‘OH ’ insertion into a C–H bond
9
The isolated products had identical NMR and IR spectra to authentic
samples.
and that any subsequent rearrangements occur via solvolysis.
The question then arises as to how the principal oxidation
product 6 is generated. Metabolic studies suggest that 6 does not
derive from 3 and we have found no acid catalyzed reactions of
1
0 A solution of dieldrin (0.26 mmol) and Fe(ClO
5 ml) was flushed with argon, H (0.1 mmol) in MeCN (0.5 ml) was
slowly added over 15 min and the solution was then allowed to stand at
4 2
) (0.1 mmol) in MeCN
(
2 2
O
3
which generated 6. Moreover, carbonium ion intermediates on
2 2
room temperature for 1 h. The mixture was then extracted with CH Cl
(2 3 20 ml) and the products separated by TLC.
1 D. Mansuy, Coord. Chem. Rev., 1993, 125, 129.
2 P. R. Ortiz de Montellano, Oxygen Activation and Reactivity in
Cytochrome P-450, 2nd edn., ed. P. R. Ortiz de Montellano, Plenum,
New York, 1995.
the highly chlorinated ‘perchlorocyclopentadiene ring’ seem
unlikely. Radical intermediates, however, would be consistent
with the oxygen-rebound mechanism proposed by Groves and a
radical at C-2 would be stabilized by the chlorine atom. Indeed,
the P-450 mediated oxidation of dieldrin and the parallel
biomimetic studies described here add further support to the
oxygen-rebound mechanism and the intermediacy of substrate
radicals where, as shown in Scheme 1, an initial hydrogen atom
1
1
1
3 P. R. Ortiz de Montellano and R. A. Stearns, J. Am. Chem. Soc., 1987,
1
09, 3415.
1
4 V. W. Bowry and K. V. Ingold, J. Am. Chem. Soc., 1991, 113, 5699.
15 J. T. Groves and P. Viski, J. Am. Chem. Soc., 1989, 111, 8537.
16 M. Newcomb, M.-H. Le Tadic-Biadatti, D. L. Chestney, E. S. Roberts
and P. F. Hollenberg, J. Am. Chem. Soc., 1995, 117, 12 085; P. H. Toy,
B. Dhanabalasingham, M. Newcomb, I. H. Hanna and P. F. Hollenberg,
J. Org. Chem., 1997, 62, 9114; M. Newcomb, M.-H. Le Tadic-Biadatti,
D. A. Putt and P. F. Hollenberg. J. Am. Chem. Soc., 1995, 117, 3312.
abstraction from C-9 by ‘Compound I’ (the common inter-
IV+·
mediate in these oxidations, ONFe
8) would generate the
substrate radical 2 and a heme intermediate 9 which can be
formulated as an hydroxyl radical coordinated to (and stabilized
by) the ferric iron. Intermediate 2 can then react at C-9 to give
the alcohol 3 (and the ferric porphyrin ready for another
Communication 8/09080G
630
Chem. Commun., 1999, 629–630