Improvement of a biomimetic porphyrin catalytic system by addition of acids.

Article abstract of DOI:10.1248/cpb.50.744

The conditions of the use of the manganese/porphyrin/imidazole system needed to be improved in order to obtain larger amounts of models of metabolites. An increase of the oxidation yields and a better preservation of this catalytic system have been obtained on the examples of various alkanes, by an acid addition in the reaction mixture. Three manganoporphyrins were checked for evaluation of the reaction. These results were extended to molecules of therapeutical interest such as ibuprofen and phenylbutazone.

Full text of DOI:10.1248/cpb.50.744

7
44
Chem. Pharm. Bull. 50(6) 744—748 (2002)
Vol. 50, No. 6
Improvement of a Biomimetic Porphyrin Catalytic System by Addition
of Acids
Jérôme SEGRESTAA, Philippe VÉRITÉ, François ESTOUR, Sabine MÉNAGER, and Olivier LAFONT*
Laboratoire de Pharmacochimie, Département de Chimie Organique, Faculté de Médecine et de Pharmacie de Rouen; 22
Boulevard Gambetta, 76183 Rouen Cedex 1, France. Received November 16, 2001; accepted March 11, 2002
The conditions of the use of the manganese/porphyrin/imidazole system needed to be improved in order to
obtain larger amounts of models of metabolites. An increase of the oxidation yields and a better preservation of
this catalytic system have been obtained on the examples of various alkanes, by an acid addition in the reaction
mixture. Three manganoporphyrins were checked for evaluation of the reaction. These results were extended to
molecules of therapeutical interest such as ibuprofen and phenylbutazone.
Key words porphyrin catalyst; oxidation; metabolite; imidazole cocatalyst; acidic medium
Biomimetic catalytic systems of cytochrome P 450 en- when greater amounts of imidazole (44 mmol) were used
zymes present at least two interests: they allow the catalysis (85%, experiment C), but the reaction time had to be dou-
of selective oxidations in organic synthesis, and the predic- bled. We observed a significant destruction of imidazole in
tion and the preparation of models of metabolites of biologi- the presence of an excess of oxidizing agent (200 mmol),
cal interest. Cytochromes P 450 catalyse both alkane hydrox- (experiments A—C): less than 25% of the starting imidazole
ylation and alkene epoxidation in the presence of an oxygen were then recovered at the end of the reactions. Oxidation
atom donor. Unfortunately, in the case of alkane oxidation by yields depended upon the amount of imidazole in the
biomimetic systems, yields remained medium and it was medium. When imidazole (24 mmol) was added dropwise
1
)
necessary to improve the reaction conditions.
during 1 h (experiment B), or when greater amounts of imi-
Amongst the oxygen atom donors, hydrogen peroxide is dazole (44 mmol) were used (experiment C), oxidation yields
particularly convenient for hydrocarbon mono oxygenation: increased, respectively to 60 and 85%. The increase of imida-
reaction yields are good and the technique is simple to im- zole amounts introduced in the reaction medium was related
plement. As the use of hydrogen peroxide is only compatible to an increase (experiment C compared to experiment B) of
2
)
with second-generation metalloporphyrins, we chose the the degradation of both catalyst (21%) and cocatalyst (89%
most robust and efficient one: Mn(TDCPP)Cl; [TDCPP in). Meanwhile, the ratio between alcohol and ketone re-
stands for tetrakis-(2,6-dichlorophenyl)porphyrin] (Chart 1; mained the same (about 0.8).
R ϭR ϭCl; R ϭH).
These results showed clearly that imidazole degradation
1
3
2
A nitrogen base cocatalyst being necessary, imidazole ap- was a limiting factor of alkane oxidation. Besides, the solu-
3
)
peared to give the best oxidation yields. Under these condi- tion consisting in adding imidazole, as soon as its destruction
tions, there was a competition between alkane oxidation and
imidazole oxidation. These effects have been shown in the
4
,5)
case of alkene oxidation. Destructive oxidation of imida-
zole became negligible in presence of oxidant or alkene ex-
cesses. These conditions have been transposed to alkane hy-
droxylation without description of their effects on the cocata-
7
)
lyst.
Oxidation of Ethylbenzene in the Presence of an Excess
of Hydrogen Peroxide In order to establish the behaviour
of the cocatalyst and to avoid its destruction, we decided to
study what occured under these reaction conditions (excess
of oxidative agent: hydrogen peroxide, 200 mmol) on an
example of saturated hydrocarbon: ethylbenzene (40 mmol).
All experiments were realized in the presence of 1 mmol of
Mn(TDCPP)Cl (Chart 2).
The results are summarized in Table 1 (experiments A—
C). Oxidation yields have been determined on isolated oxida-
tion products, and not by gas or liquid chromatography
analysis, in order to finalize a method for the preparation of
large amounts of models of metabolites.
Chart 1
When imidazole (24 mmol) was added dropwise in the re-
action mixture silmutaneously to hydrogen peroxide in order
to prevent any decrease of the cocatalyst amount in the
medium, the reaction yield increased significantly (40 to
6
0%, experiments A and B). The results were even better
Chart 2
∗
To whom correspondence should be addressed. e-mail: Olivier.Lafont@univ-rouen.fr
© 2002 Pharmaceutical Society of Japan

June 2002
745
went on, was limited by the concomitant degradation of the of Oxidative Agent and of Formic Acid The use of weak
porphyrin catalyst. We meant to study operative conditions acids appeared to be the most efficient process to use. These
allowing to maintain a good stability of imidazole in order to results were still valid when an oxidant excess was used. In
prevent both its oxidation, and the degradation of the cata- presence of 7 equivalents of formic acid, oxidative yield im-
lyst.
proved without excessive catalyst degradation (18%, Table 1,
Oxidation of Ethylbenzene under Stoichiometric Con- experiment D). This allowed to obtain a reaction yield (80%,
ditions The reaction was performed in stoichiometric con- Table 1, experiment D) similar to the rate obtained by adding
ditions, so far as alkane or hydrogen peroxide were con- an extra quantity of imidazole during the reaction (85%,
cerned, in order to minimize eventual interactions of an ex- Table 1, experiment C). The improvement of experiment D
cess of these compounds with the oxidative process. Never- consisted in the preservation of the catalyst from its degrada-
theless under these stoichiometric conditions, the yields tion (82% Mn(TDCPP)Cl regenerated in experiment D ver-
(
15%) were much lower than those observed in the presence sus 63% in experiment C). The reaction time had also an in-
of an excess of hydrogen peroxide. In order to avoid N-oxide fluence on the reaction yield, which in acid presence, in-
formation from imidazole, we decided to add various acids to creased until 4 h (80%, Table 1, experiment D), and then re-
the medium: HCl and H SO as strong acids (Fig. 1), formic mained unchanged.
2
4
acid and acetic acid as weak acids (Fig. 2). The reaction evo-
lution was followed on the basis of the oxidation yield and of catalyst or cocatalyst, but in the presence of an excess of
Mn(TDCPP)Cl destruction. weak acid, the yields were very low (Table 1, experiments E
Role of Acid When the reaction was performed without
In the case of weak acids (Fig. 2), the best results were ob- and F). This observation proved that the acid was not alone a
tained when 7 to 8 acid equivalents were added. The results catalyst of the reaction, but that it helped the catalysts in their
were similar wether formic acid or acetic acid were used. The action. As regards imidazole, its destruction was complete,
yields (28%) were then higher than in the case of strong when it was treated in the absence of substrate, by hydrogen
acids (Fig. 1; 20%) and varied more gradually about this peroxide and of Mn(TDCPP)Cl. On the other hand, when
value.
formic acid was added in the reaction medium, 73% of imi-
Oxidation of Ethylbenzene in the Presence of an Excess dazole was intact at the end of reaction time. This imidazole
protection by a weak acid could be explained by a reversible
protonation of the nitrogen atom making less available the
electron pair.
In one case, the nature of an oxidized imidazole derivative
could be isolated, when the experiment was realized in the
presence of both formic and phenylacetic acids. It resulted
from the oxidation of the three carbon atoms into carbonyl
groups, leading to parabanic acid (Chart 3). The low value of
the yield (9%) showed that the protection of imidazole nitro-
gen atoms by acids was effective.
Fig. 1. Influence of the Presence of a Strong Acid upon the Reaction
Extension to Other Hydrocarbons In order to test the
Fig. 2. Influence of the Presence of a Weak Acid Upon the Reaction
Table 1. Experiments in the Presence of an Excess of Oxidative Agent
Conditions
Imidazole (mmol)
Formic acid (mmol)
Time (h)
A
B
C
D
E
F
a)
a)
24
0
2
24
0
2
44
0
4
24
168
4
0
168
4
24
168
4
Yield (%)
40
60
85
80
3
0
Ratio (alkohol/ketone)
Regenerated Mn(TDCPP)Cl (mmol)
Degraded imidazole (mmol (%))
0.9
0.78
20.4 (85)
0.8
0.80
18.5 (77)
0.8
0.63
34.8 (79)
0.8
0.82
nc
nc
0.95
—
—
—
0 (0)
Conditions A to E, Mn(TDCPP)Cl (1 mmol)/H O (200 mmol)/ethylbenzene (40 mmol); condition F, without Mn(TDCPP)Cl. a) Added dropwise. nc: not calculated.
2
2

7
46
Vol. 50, No. 6
Regenerated
Table 2. Evaluation of Other Catalysts
Oxidation
yields without
formic acid
Oxidation
Regenerated
yields with catalyst without catalyst with
formic acid
(%)
formic acid
(%)
formic acid
(%)
(%)
Chart 3
Mn(TPP)Cl
Mn(TMP)Cl
Ͻ1
2
7
0
81
76
11
47
Conditions: Mn(TPP)Cl or Mn(TMP)Cl (1 mmol)/H O2 (200 mmol)/ethylbenzene
2
(40 mmol)/imidazole (24 mmol).
Table 3. Extension to Other Hydrocarbons
Oxidation
yields without
formic acid
Oxidation
Regenerated
Regenerated
yields with catalyst without catalyst with
formic acid
(%)
formic acid
(%)
formic acid
(%)
(%)
Tetraline
81
59
50
92
75
58
75
72
79
93
91
85
Decaline
Chart 4
Chart 5
Chart 6
Cyclohexane
Chart 8
tested: Mn(TPP)Cl and Mn(TMP)Cl. Both of them were dif-
ferent from Mn(TDCPP)Cl so far as the substituents of the
phenyl group were concerned. For Mn(TPP)Cl, the nucleus
was unsubstituted (Chart 1; R ϭR ϭR ϭH) [TPP stands for
1
2
3
tetrakisphenylporphyrin], and for Mn(TMP)Cl the nucleus
was substituted by three methyl groups (Chart 1; R ϭR ϭ
1
2
R ϭCH ) [TMP stands for tetrakismesitylporphyrin]. In the
3
3
case of alkenes oxidation, these porphyrins had already
proved to be less efficient than Mn(TDCPP)Cl.
In both cases, addition of formic acid succeeded in pro-
tecting the manganoporphyrin from degradation (Table 2).
For Mn(TPP)Cl the percentage of regenarated catalyst in-
creased in acidic medium from 0 to 81 and for Mn(TMP)Cl,
it increased from 47 to 76.
Unfortunately, the weak efficience of these two catalysts
did not allow to increase oxidation yields (Table 2).
Application to Antiflammatory Agents In order to
Chart 7
validity of the process with other hydrocarbons, we chose apply this methodology to medicinal compounds, we chose
various compounds: tetraline (Chart 4) which presents meth- two antiflammatory agents as example of structures of thera-
ylene groups in a-position to the benzene ring like ethylben- peutical interest, ibuprofen and phenylbutazone. Concerning
zene, decaline (Chart 5) for its structural relation to tetraline, ibuprofen, hydroxylation occurred on the tertiary isobutyl
cyclohexane (Chart 6) which can be considered as a part of carbon (Chart 8). The hydroxy metabolite was obtained in
decaline and diphenylmethane (Chart 7) which is a more 15% yield. In vivo, this regioselective oxidation was the
8
,9)
bulky substrate.
major biodegradation pathway.
For phenylbutazone, the
1
0)
In the cases of the most reactive compounds (tetraline, de- oxidation of this molecule had already been achieved. Ac-
caline, cyclohexane), addition of formic acid significatively cording to the metabolic or the degradation pathways, several
improved oxidation yields and the protection of the por- products of oxidized phenylbutazone has been isolated.
1
0,11)
phyrinic catalyst (Table 3). In spite of steric hindrance, in the Some of them were of particular interest because of their
case of diphenylmethane, addition of formic acid allowed to deleterious effects. When phenylbutazone was oxidized by
obtain oxidized compound with weak yield (15%), the cata- the biomimetic manganese/porphyrine/imidazole catalytic
lyst being protected from degradation (90%).
Evaluation of Other Porphyrin Catalytic Systems To pound (Chart 9) which may be involved in allergic reac-
system under acidic conditions, only the 4-hydroxy com-
1
2,13)
extend these results, two other manganoporphyrins were tions
was isolated with a very good yield (90%).

June 2002
747
(
30% in water, 200 mmol) was diluted in acetonitrile (1/9, v/v), and was
added dropwise (during 2 h) at room temperature, to a solution of
Mn(TDCPP)Cl (1 mmol), imidazole (24 mmol), substrate (40 mmol) and
formic acid (168 mmol) in 400 ml of dichloromethane/acetonitrile (1/1, v/v).
The mixture was stirred at room temperature for 4 h. Evaporation of the sol-
vent under reduced pressure afforded a residual oil which was chro-
matographed (SiO ). In the case of ibuprofen and phenylbutazone, the quan-
tities of all reagents and solvents were divided by ten.
2
Chart 9
1
,2,3,4-Tetrahydronaphtalen-1-ol, 4-hydroxy-3,4-dihydronaphtalen-1(2H)-
one and 3,4-dihydronaphtalen-1(2H)-one were respectively obtained in 61%
4-hydroxy-3,4-dihydronaphtalen-1(2H)-one) and 31% (3,4-dihydronaph-
talen-1(2H)-one) yields under acidic conditions (versus 48% of hydroxy-
compounds and 33% of 3,4-dihydronaphtalen-1(2H)-one) yields without
(
In conclusion, oxidative conditions applied to alkenes
were not ideally transposable to alkanes or alkyl groups. The
destructive oxidation of the cocatalyst, which was a limiting acid). The solvent for the chromatography was cyclohexane/acetone (8/2,
factor of the reaction, did not allow to obtain good yields v/v). A mixture of oxidized compounds of decaline (separating by GC/MS
but not isolated) was obtained in 75% yield. Cyclohexanol and cyclohexa-
none were respectively obtained in 44 and 14% yields under acidic condi-
tions (versus 38 and 12% yields without acid). The solvent for the chro-
matography was chloroform. Diphenylmethanol and benzophenone were re-
while maintaining an appreciable catalyst regeneration.
Addition of formic acid improved the efficiency of the oxi-
dation of alcanes or alkyl chains using Mn(TDCPP)Cl/hydro-
gen peroxide/imidazole biomimetic system as catalyst. Under spectively obtained in 8 and 7% yields under acidic conditions. The solvent
acidic conditions, the cocatalyst appeared to be more stable for the chromatography was cyclohexane/acetone (9/1, v/v). 2-[4-(2-Hydroxy-
2
-methylpropyl)phenyl]propanoic acid was obtained in 15% yield under
and the catalyst loss remained negligible. The hydroxylation
yields have been improved by acid addition, observation
which tends to impair that imidazole acts as a basic catalyst.
acidic conditions. The solvent for the chromatography was dichloromethane/
diethyl ether (1/1, v/v). mp 122 °C. NMR (CDCl ) d: 1.15 (6H, s, 2ϫCH ),
1
3
3
.43 (3H, d, CH ), 2.67 (2H, s, CH ), 3.64 (1H, q, CH ), 7.15 (4H, m,
3 2 2
ϩ
.
With these improvements, the reaction became quantitatively phenyl). MS m/z (relative intensity): 222 (M , 3), 207 (5), 177 (5), 164
(
(
74), 162 (38), 159 (8), 119 (100), 115 (14), 105(11), 91 (92), 77 (8), 59
67). 4-Butyl-4-hydroxy-1,2-diphenylpyrazolidine-3,5-dione was obtained in
efficient to prepare sufficient amounts of models of metabo-
lites.
The use of other porphyrin catalysts confirmed the interest
of addition of formic acid but showed that Mn(TDCPP)Cl
was a better choice.
9
0% yield under acidic conditions. The solvent for the chromatography was
dichloromethane/cyclohexane (1/1, v/v). mp 128 °C. NMR (CDCl ) d: 0.77
3
(
3H, t, CH ), 1.29 (4H, m, 2ϫCH ), 1.95 (2H, t, CH ), 4.67 (1H, s, OH),
3 2 2
ϩ
.
7.15 (10H, m, 2ϫPhenyl). MS m/z (relative intensity): 324 (M , 95), 183
(
100), 120 (17), 93 (20), 77 (86), 57 (29).
Experimental
Oxidation Using Mn(TPP)Cl or Mn(TMP)Cl/Imidazole Catalytic
Mn(TDCPP)Cl and Mn(TPP)Cl were prepared according to the respective System Hydrogen peroxide (30% in water, 200 mmol) was diluted in ace-
6
,14—16)
cited procedures.
Mn(TMP)Cl was obtained as previously de- tonitrile (1/9, v/v), and was added dropwise (during 2 h) at room tempera-
1
6)
1
scribed. All oxidized compounds were identified by H-NMR and mass ture, to a solution of Mn(TPP)Cl or Mn(TMP)Cl (1 mmol), imidazole (24
spectrometry. Oxidized products of ethylbenzene, tetraline, cyclohexane and mmol), ethylbenzene (40 mmol) and formic acid (168 mmol) in 400 ml of
diphenylmethane were known compounds.
Oxidation Using Mn(TDCPP)Cl/Imidazole Catalytic System. Gen- temperature for 4 h. The solvent was evaporated under reduced pressure, and
eral Procedure for Oxidation of Ehtylbenzene in the Presence of an Ex-
the residual oil was chromatographed (solid phase: SiO , solvents: cyclo-
dichloromethane/acetonitrile (1/1, v/v). The mixture was stirred at room
2
cess of Hydrogen Peroxide Hydrogen peroxide (30% in water, 200 mmol) hexane/acetone (9/1, v/v)) to give 1-phenylethanol and acetophenone. Yields
was diluted in acetonitrile (1/9, v/v), and was added dropwise (during half of are mentionned in Table 2 for every experiment.
the total time of the reaction) at room temperature, to a solution of
Regeneration of Manganoporphyrin Catalyst When the oxidated
Mn(TDCPP)Cl (1 mmol for experiments A—C), imidazole (24 mmol for ex- compounds were isolated, the solid phase of chromatography was washed by
periments A and B, 44 mmol for C), ethylbenzene (40 mmol) in 400 ml of acetone. Then, the catalyst was regenerated by eluting with ethanol. Ethanol
dichloromethane/acetonitrile (1/1, v/v). The mixture was stirred at room solution was concentrated under reduced pressure to 50 ml. Fifty milliliters
temperature for 2 h (experiments A and B) or 4 h (experiment C). The sol- of water was added, and the residue was filtered to recover manganopor-
vent was evaporated under reduced pressure, and the residual oil was chro- phyrin which was dried and weighed.
matographed (solid phase: SiO , solvents: cyclohexane/acetone (9/1, v/v)) to
Regeneration of Imidazole Isolation of oxidized compounds and re-
give 1-phenylethanol and acetophenone. Yields are mentionned in Table 1 generation of imidazole was not performed in the same experiment because
for every experiment (A—C).
imidazole could not be separated in the conditions of isolation of oxidized
General Procedure for Oxidation of Ehtylbenzene in Stoichiometric compounds.
Conditions The same general procedure was used, but hydrogen peroxide
When the oxidative reaction was complete, the medium was neutralized
100 mmol) and ethylbenzene (100 mmol) were substituted to the previous with sodium carbonate, and the solvent was evaporated under reduced pres-
values. Strong acids volumes used for experimental results indicated in Fig. sure. The mixture was diluted in dichloromethane and extracted with water.
were 1.05 ml (0.5 eq), 2.10 ml (1.0 eq), 3.20 ml (1.5 eq) in the case of hy- The aqueous solution was dried in a dessiccator, the residue was diluted in
drochloric acid (35% water solution), and 0.35 ml (0.5 eq), 0.70 ml (1.0 eq), dichloromethane and was filtered. The organic solution was evaporated
.05 ml (1.5 eq) in the case of sulfuric acid (93% water solution). For weak under reduced pressure to recover imidazole which was idendified and quan-
acids, results indicated in Fig. 2 were obtained with 0.90 ml (1 eq), 3.60 ml tified.
2
(
1
1
(
(
(
4 eq), 5.40 ml (6 eq), 6.30 ml (7 eq), 7.20 ml (8 eq), 8.10 ml (9 eq), 14.40 ml
16 eq) in the case of formic acid, and 1.40 ml (1 eq), 7.60 ml (4 eq), 8.20 ml
6 eq), 9.60 ml (7 eq), 10.10 ml (8 eq), 12.30 ml (9 eq), 21.90 ml (16 eq) in
Acknowledgments We thank Laurent Petit for his contribution to this
work.
the case of acetic acid. Reaction time was 4 h. Oxidation yields are respec-
tively mentionned in Figs. 1 and 2.
References
General Procedure for Oxidation of Ehtylbenzene in the Presence of
an Excess of Hydrogen Peroxide and of Formic Acid The same proce-
dure for oxidation of ehtylbenzene in the presence of an excess of hydrogen
peroxide (vide supra) was used, but eventually (experiments D and F) the
amount of imidazole (24 mmol) was substituted to the previous values and
formic acid (168 mmol) was added in the medium. Yields are also men-
tioned in Table 1 for every experiment (D—F).
1) Meunier B., Bull. Soc. Chim. Fr., 4, 578—594 (1986).
2) Mansuy D., Battioni P., “Metalloporphyrins Catalyzed Oxidations,” ed.
by Sheldon R. A. Dekker, New York, 17, 99—132 (1994).
3) Battioni P., Bartoli J.-F., Leduc P., Fontecave M., Mansuy D., J. Chem.
Soc., Chem. Commun., 1987, 791—792.
4) Renaud J. P., Battioni P., Bartoli J. F., Mansuy D., J. Chem. Soc., Chem.
Commun., 1985, 888—889.
General Procedure for Oxidation of Tetraline, Decaline, Cyclohexane,
Diphenylmethane, Ibuprofen and Phenylbutazone Hydrogen peroxide
5) Rocha Gonsalves A., Johnstone R. A. W., Pereira M., Shaw J., J.
Chem. Soc. Perkin Trans. 1, 1991, 645—649.

7
48
Vol. 50, No. 6
6)
7)
8)
Lindsey J. S., Hsu M. C., Schreiman I. C., Tetrahedron Lett., 27, 11) Fabre H., Hussam-Eddine N., Mandrou N., J. Pharm. Sci., 73, 1706—
4
969—4970 (1986).
Battioni P., Renaud J. F., Bartoli J. F., Reina-Artiles M., Fort M., Man- 12) Alexander D. M., Mathew G. E. A., Wilson J. B., Xenobiotica, 15,
suy D., J. Am. Chem. Soc., 110, 8462—8470 (1988). 123—128 (1985).
Baillie T. A., Adams W. J., Kaiser D. G., Olanoff L. S., Halstead G. W., 13) Bundgaard H., Arch. Pharm. Chem., 6, 223—225 (1978).
1709 (1984).
Harpoothian H., Van Giessen G. J., J. Pharm. Exp. Ther., 249, 517—
23 (1989).
Sanins S. M., Adams W. J., Kaiser D. G., Halstead G. W., Baillie T. W.,
14) Kinh-Botulinsky M., Meunier B., Inorg. Chem., 27, 209—210 (1988).
15) Van Der Made A. W., Hoppenbrower E. J. H., Nolte R. J. M., Drenth
W., Rech. Trav. Chim. Pays Bas, 107, 15—16 (1988).
5
9
)
Drug. Metab. Dispos., 18, 523—527 (1990).
16) Lindsey J. S., Wagner R. W., J. Org. Chem., 54, 827—836 (1989).
1
0) Gupta R. N., J. Chromatogr., 530, 160—163 (1990).