Oxidation of carbidopa by tyrosinase and its effect on murine melanoma

Article abstract of DOI:10.1016/j.bmcl.2009.05.002

Oxidation of the anti-Parkinsonian agent carbidopa by tyrosinase was investigated. The products of this reaction were identified as 3-(3,4-dihydroxyphenyl)-2-methylpropanoic acid and 6,7-dihydroxy-3-methylcinnoline. These results demonstrate that after ox

Full text of DOI:10.1016/j.bmcl.2009.05.002

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3507–3510
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Oxidation of carbidopa by tyrosinase and its effect on murine melanoma
a
a
a
b
_
´
Beata Ga˛sowska-Bajger , Bozena Fra˛ckowiak-Wojtasek , Sabina Koj , Tomasz Cichon ,
Ryszard Smolarczyk b, Stanisław Szala b, Hubert Wojtasek a,
a Faculty of Chemistry, University of Opole, Ul. Oleska 48, 45-052 Opole, Poland
*
b
_
Department of Molecular Biology, Center of Oncology, Maria Skłodowska-Curie Memorial Institute, Wybrzeze Armii Krajowej 15, 44-101 Gliwice, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Oxidation of the anti-Parkinsonian agent carbidopa by tyrosinase was investigated. The products of this
reaction were identified as 3-(3,4-dihydroxyphenyl)-2-methylpropanoic acid and 6,7-dihydroxy-3-meth-
ylcinnoline. These results demonstrate that after oxidation of the catechol moiety to an o-quinone either
a redox exchange with the hydrazine group or a cyclization reaction occur. The cyclization product
underwent additional oxidation reactions leading to aromatization. The cyclization reaction is undesired
in the case of hydrazine-containing anti-melanoma prodrugs and will have to be taken into account in
designing such compounds. Carbidopa was tested against B16(F10) melanoma cells in culture and
Received 17 March 2009
Revised 1 May 2009
Accepted 3 May 2009
Available online 7 May 2009
Keywords:
Tyrosinase
Carbidopa
Hydrazine
Prodrug
showed cytotoxicity significantly higher than either of its oxidation products and
however, was not specific to this cell line.
L-dopa. This effect,
Ó 2009 Elsevier Ltd. All rights reserved.
Melanoma
Carbidopa
((2S)-3-(3,4-dihydroxyphenyl)-2-hydrazinyl-2-
We have recently shown that the hydrazine group in amino acid
phenylhydrazides9 and in the antitumor drug procarbazine10 can
be oxidized by o-quinones and therefore indirectly by the action
of tyrosinase. Based on these results we have postulated that this
redox exchange reaction can be utilized in activation of anti-mela-
noma prodrugs with a hydrazine linker. Carbidopa, an approved
drug containing both the catechol and hydrazine moieties was an
obvious choice to test the reaction in an intramolecular setting be-
fore designing target compounds.
methylpropanoic acid) is an aromatic amino acid decarboxylase
inhibitor1 which was introduced as a co-drug with
L-dopa in the
treatment of Parkinson’s disease in the early 1970s.2–4 Its metabo-
lism was studied primarily in vivo and several metabolites were
identified in plasma and urine of a number of mammalian species,
including humans, in free form or as glucuronate conjugates: 3-
(3,4-dihydroxyphenyl)-2-methylpropanoic acid, 3-(4-hydroxy-3-
methoxyphenyl)-2-methylpropanoic acid, 3-(3-hydroxyphenyl)-
2-methylpropanoic
acid,
3-(4-hydroxy-3-methoxyphenyl)-2-
We first performed the reaction of 0.1 mM carbidopa (Sochinaz
S.A., Vionnaz, Switzerland) with mushroom tyrosinase11 (specific
activity 2956 U/mg) and monitored it spectrophotometrically.
The reaction was carried out in 2.6 mL of 100 mM sodium phos-
phate buffer pH 6.8. We observed a spectrum very similar to that
obtained in the reaction catalyzed by sweet potato polyphenol oxi-
dase8 with a maximum at 370 nm (Fig. 1). The second maximum at
410 nm was transient and remained for prolonged periods of time
only if the substrate concentration was 0.25 mM or higher (data
not shown). We next carried out oxygen consumption measure-
ments at 0.1 mM concentration of substrates (12 mL of 100 mM so-
dium phosphate buffer, pH 6.8), which demonstrated that the
methyllactic acid, 3-(3-hydroxyphenyl)-2-methyllactic acid, and
3,4-dihydroxyphenylacetone.5–7 These products suggest participa-
tion of oxidases, catechol O-methyltransferase and dehydroxylat-
ing systems in the metabolism of this drug. However,
involvement of any particular enzyme in the degradation of carbi-
dopa has not been studied so far. Sweet potato polyphenol oxidase
was applied for flow injection-spectrophotometric determination
of L
-dopa and carbidopa in pharmaceutical formulations,8 but no
attempts were made to identify the products of the oxidation of
carbidopa by this enzyme. It was even speculated that the product
of this reaction was dopachrome, although the UV–vis spectrum of
the chromophore differed markedly from dopachrome, which can-
not be formed in this reaction. The spectrum obtained after oxida-
tion of carbidopa by polyphenol oxidase shows two maxima: at ca.
360 and 410 nm8 and cannot be attributed to any of the metabo-
lites identified so far, which do not absorb in this region.
reaction rate for carbidopa was slightly lower than for
(0.19 0.030 M/s and 0.27 0.027 M/s, respectively) and less
oxygen was consumed per mole of carbidopa (80 M and 91 lM
L-dopa
l
l
l
after 60 min incubation for dopa and carbidopa, respectively,
Fig. 2). The reaction was then performed at 0.25 mM or 0.4 mM
concentration of carbidopa, under conditions where oxygen was
the limiting agent, to prevent secondary oxidations and polymeri-
zation, and the mixture was analyzed by HPLC (Altech Alltima
* Corresponding author. Tel.: +48 77 452 7122; fax: +48 77 452 7101.
E-mail address: Hubert.Wojtasek@uni.opole.pl (H. Wojtasek).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.05.002

3508
B. Ga˛sowska-Bajger et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3507–3510
1.2
1
0.45
0.4
11.5
0.35
0.3
0.8
0.6
0.4
0.2
0
6.6
4.9
0.25
0.2
0.15
0.1
0.05
220
270
320
370
420
470
520
Wavelength (nm)
0
0
5
10
t (min.)
15
20
Figure 1. Spectral changes during oxidation of 0.1 mM carbidopa by
5 lg of
tyrosinase in 2.6 mL of 100 mM sodium phosphate buffer, pH 6.8. The spectra
displayed were recorded immediately after addition of the enzyme and then after 2,
4, 6, 10, 20, and 30 min. The arrow shows the direction of spectral changes (increase
of absorbance).
Figure 3. Chromatographic analysis of a mixture of 0.25 mM carbidopa with 5 lg of
tyrosinase in 1 mL of 100 mM sodium phosphate buffer, pH 6.8, after 30 min
incubation (detection at 220 nm). Retention times determined by analysis of
standards were: 4.9 min for carbidopa, 6.6 min for 6,7-dihydroxy-3-methylcinno-
line, 11.5 min for 3-(3,4-dihydroxyphenyl)-2-methylpropanoic acid, and 16.4 min
for 3,4-dihydroxyphenylacetone (not seen in the presented chromatogram).
0.26
0.24
0.22
0.2
time of 6.6 min in HPLC analysis gave an NMR spectrum character-
istic for a bicyclic aromatic compound: 1H (D2O)—2.625 (3H, s),
6.815 (1H, s), 7,298 (1H, s), 7.549 (1H, s); 13C (D2O)—20.43,
103.67, 105,99, 117,93, 125,43, 145.82, 148.70, 153.36, 157.55).
The structure was determined from an 1H, 13C HMBC spectrum
and LC/ESI-MS analysis (Bruker MicrOTOF-Q, 177.1 a.m.u. in a po-
sitive ion mode, 175.1 a.m.u. in a negative ion mode) as 6,7-dihy-
droxy-3-methylcinnoline (2, Scheme 1). The UV–vis spectrum of
the isolated compound was very similar to that of the reaction
mixture (Fig. 4). These results demonstrate that after oxidation of
the catechol moiety to an o-quinone by tyrosinase either the redox
exchange (intramolecular or intermolecular) with the hydrazine
group or the nucleophilic attack of the latter on the former take
place. Possible pathways leading to the formation of these two
products are presented in Scheme 1. The mechanisms proposed
are consistent with the results of in vivo studies of the metabolism
of carbidopa, which did not detect hydrazine in the urine or plasma
of experimental animals.5,6 The pathway leading to the formation
of 3-(3,4-dihydroxyphenyl)-2-methylpropanoic acid, postulated
by the authors, included oxidation of the hydrazine group and its
loss as a nitrogen molecule.6 6,7-Dihydroxy-3-methylcinnoline is
formed by cyclization of the o-quinone (nucleophilic attack of
the hydrazine nitrogen atom) and a subsequent 4-electron oxida-
tion. However, steps and factors participating in the formation of
this bicyclic aromatic product remain unclear.
The cyclization of o-quinones with a hydrazine group in the
side-chain is an undesired side-reaction from the point of view of
designing anti-melanoma prodrugs.10 It may reduce the yield of
effector release in the case of dialkyl hydrazines. Cyclization should
not occur, however, in the case of hydrazides, carbazates or semic-
arbazides. Acylation of the hydrazine moiety at the distal nitrogen
atom should make it insufficiently nucleophilic, as it has been re-
cently demonstrated for dopamine derivatives.12
Small amounts of 3,4-dihydroxyphenylacetone were also iso-
lated from our reaction mixtures: 1H (DMSO-d6)—2.047 (3H, s),
3.502 (2H, s), 6.428 (1H, dd), 6.557 (1H, d), 6.651 (1H, d) (3, Scheme
1). This compound was not detected in the HPLC analysis of the ori-
ginal reaction mixtures (retention time of a synthetic reference
was 16.4 min), which confirms previous suggestions that it is pro-
duced from carbidopa during sample manipulation.6 3,4-
Dihydroxyphenylacetone was the major product obtained previ-
ously after electrochemical and chemical oxidation of carbidopa.13
Two oxidation reactions were detected by cyclic voltammetry and
1
0.18
0.16
2
0.14
0.12
0.1
0
300
600
900
t (s)
1200
1500
1800
Figure 2. Oxygen consumption measurements during oxidation of 0.1 mM carbi-
dopa (1) and 0.1 mM -dopa (2) by 46 g of tyrosinase in 12 mL of 100 mM sodium
phosphate buffer, pH 6.8.
L
l
C18, 150 Â 3 mm column connected to a Beckman System Gold
instrument with a diode array detector and a 20 L sample loop).
l
Separation was performed with 0.1% TFA in water and acetonitrile
as the mobile phase (10% acetonitrile for 2 min, then 10–60% ace-
tonitrile gradient in 18 min), at a flow rate of 0.4 mL/min. Chro-
matograms were recorded at 220, 280, and 340 nm. Two
products were detected in the reaction mixture in addition to the
unreacted substrate (Fig. 3). To identify them, a preparative reac-
tion was performed at 45 mg scale (0.25 mM carbidopa concentra-
tion, 800 mL of 10 mM sodium phosphate buffer, pH 6.8). After the
UV–vis spectra remained unchanged, the enzyme was removed
from the reaction mixture by ultrafiltration (Amicon Ultra-15,
10,000 MWCO, Millipore), the filtrate was concentrated by evapo-
ration to ca. 1/10 of the initial volume and loaded on a 16 mL C18
column (Bakerbond Octadecyl). Elution was performed with a
stepwise gradient of acetonitrile in water. Fractions were analyzed
spectrophotometrically and by TLC. Three products were detected,
isolated and identified. The major product with a retention time of
11.5 min in HPLC analysis was identified by NMR analysis as 3-
(3,4-dihydroxyphenyl)-2-methylpropanoic acid (1, Scheme 1): 1H
(CD3OD, Bruker Ultrashield 400 MHz)—1.065 (3H, d), 2.433 (1H,
dd), 2.549 (1H, m), 2.840 (1H, dd), 6.516 (1H, dd), 6.624, (1H, d),
6.640 (1H, d). The spectrum of this product showed the character-
istic signals of nonequivalent methylene protons also present in
the substrate (2.433 and 2.840 ppm for this compound, 2.675
and 2.784 ppm for carbidopa). The second product with a retention

B. Ga˛sowska-Bajger et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3507–3510
3509
COOH
O
O
COOH
NH2
COOH
NH2
HO
HO
HO
HO
Tyrosinase
O2
O2
H+
N
N
HN
HN
NH
Tyrosinase
O2
CO2
O
O
COOH
COOH
HO
HO
HO
HO
H+
N
NH
NH
H2N
NH
NH2
- N2
+ H2O - NH2NH2
HO
HO
HO
HO
COOH
COOH
HO
HO
O
N
H
3
1
"O2"
CO2
HO
HO
N
N
2
Scheme 1. Postulated reactions occurring after oxidation of carbidopa by tyrosinase: o-quinone may undergo a redox exchange reaction with the hydrazine group leading to
the azo derivative and regenerating the catechol; loss of a nitrogen molecule gives 3-(3,4-dihydroxyphenyl)-2-methylpropanoic acid (1); nucleophilic attack of the hydrazine
distal nitrogen atom on the o-quinone generates a bicyclic product, whose oxidation leads to 6,7-dihydroxy-3-methylcinnoline (2); non-enzymatic oxidation of the hydrazine
group followed by decarboxylation leads to a hydrazone, which upon hydrolysis gives 3,4-dihydroxyphenylacetone (3).
carcinoma cells,14 which was later shown to be due to hydrogen
0.7
peroxide generated by the autooxidation of this compound.15 The
0.6
generation of H2O2 was attributed to the catechol moiety, whereas
the hydrazine group was not considered in this process. A mixture
of L-dopa and carbidopa was also applied in earlier studies of the
0.5
0.4
0.3
0.2
0.1
0
anti-melanoma effect of 4-S-cysteaminylphenol and its N-acetyl
derivative.16 This mixture significantly improved the cytotoxicity
of the tested compounds. A combination of L-dopa and carbidopa
was also tested in patients with metastatic malignant melanoma.17
However, it was ineffective even at the highest tolerated doses and
severe side effects were observed, which included gastrointestinal
toxicity and postural hypotension. Levodopa–carbidopa mixture
also inhibited the transformation of lymphocytes derived from pa-
tients with malignant melanoma.18 We have, however, not found
any reports describing the effect of carbidopa alone against mela-
noma. We have therefore tested it in cell cultures and compared
the effect with that of dopa.
220
270
320
370
420
470
520
Wavelength (nm)
Figure 4. The UV–vis spectrum of the isolated 6,7-dihydroxy-3-methylcinnoline.
We used B16(F10) murine melanoma cells (from the American
Type Culture Collection, ATCC) and two control cell lines: NIH3T3
(ATCC) and HECa10 (lymph node-derived and provided by Dr. D.
differential-pulse voltammetry. One of them was assigned to the
catechol group and the other one to the carboxylate. Carbon diox-
ide and hydrazine were detected as side products.13 However, the
postulated pathway leading to 3,4-dihydroxyphenylacetone in-
cluded the formation of a carbocation connected to a hydrazine
group13, which seems extremely unlikely. We believe that the sec-
ond oxidation process with Ep = +980 mV in differential-pulse vol-
tammetry corresponds to the oxidation of the hydrazine group and
not the carboxylate, as proposed by the authors,13 and the reac-
tions proceed by a pathway presented in Scheme 1.
´
Dus, Institute of Immunology and Experimental Therapy, Wrocław,
Poland). Cells (2 Â 103 cells per well) were cultured in RMPI 1640
TM
medium (100 lL) supplemented with 10% FBS using NUNCLON
Surface (NUNCTM) 96-well plates. Cultures were kept in a humidi-
fied standard incubator (37 °C and 5% CO2). After 24 h, the tested
compounds were added in quadruplicate to the culture media at
eleven different concentrations (10–2000 lM). The culture vessels
were incubated for further 24 or 48 h, the culture media were re-
placed with MTT (3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltet-
razolium bromide) solution (0.5 mg MTT/1 mL PBSꢀ) and the cells
Recently, the cytotoxic effect of carbidopa has been demon-
strated against human pulmonary carcinoid and small cell lung

3510
B. Ga˛sowska-Bajger et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3507–3510
were incubated for 3 h at 37 °C. The formed formazan crystals were
dissolved in acid–isopropanol solution. Spectrophotometric mea-
surements were performed using ELISA ELx800 reader (Bio-Tek
Instruments Inc.) at 570 nm. Percentage of surviving cells was esti-
mated as: (absorbance at time t/initial absorbance) Â 100%. Carbi-
140
120
100
80
A
dopa showed cytotoxicity significantly greater than
L-dopa. It was
lethal at 250
lM concentration after 48 h incubation (Fig. 5). How-
ever, this effect was not specific for the melanoma cells—similar
toxicity was also observed for the control NIH3T3 and HECa10 cell
lines (data not shown).
60
40
We have also tested the biological activity of the oxidation
products of carbidopa by tyrosinase: 3-(3,4-dihydroxyphenyl)-2-
methylpropanoic acid and 6,7-dihydroxy-3-methylcinnoline. As
expected for catechols, both compounds showed substantial cyto-
toxicity, although neither was as effective as carbidopa. The dose-
dependence profile for 3-(3,4-dihydroxyphenyl)-2-methylpropa-
noic acid was similar after 24 and 48 h. 6,7-Dihydroxy-3-methylc-
innoline was not effective after 24 h but was more potent after
48 h (Fig. 6). Again, however, the effect was not specific to
B16(F10) melanoma cells.
20
0
0
500
1000
1500
2000
120
100
80
60
40
20
0
B
Greater cytotoxicity of carbidopa than any of the 3 other tested
compounds (L-dopa, 3-(3,4-dihydroxyphenyl)-2-methylpropanoic
acid, and 6,7-dihydroxy-3-methylcinnoline) demonstrates that
the hydrazine group plays an important role in this process and
the effect observed previously in human pulmonary carcinoid
and small cell lung carcinoma cells14,15 is not caused by the cate-
chol moiety alone.
The strong side effects observed during the clinical trials of the
L
-dopa17 can most likely be attrib-
0
500
1000
Concentration [µM]
1500
2000
combination of carbidopa and
uted to dopamine formed by decarboxylation of
L
-dopa. Since the
metabolism of carbidopa does not lead to any physiologically sig-
nificant compounds, it may be worth trying to reevaluate its effect
Figure 6. Toxicity of 3-(3,4-dihydroxyphenyl)-2-methylpropanoic acid (-j-) and
6,7-dihydroxy-3-methylcinnoline (-d-) against B16(F10) cells in vitro: A—after
24 h, B—after 48 h.
160
A
140
120
100
80
on human melanoma. Modifications of its structure may also lead
to compounds with improved efficiency and selectivity.
Acknowledgments
We thank Sochinaz S. A., Vionnaz, Switzerland, for the generous
gift of a sample of carbidopa and Jerzy Tarnawski for performing
the HPLC analysis. This work was supported by a Grant form the
Polish Ministry of Science and Higher Education No. 2 P05F 00330.
60
40
20
References and notes
0
0
500
1000
1500
2000
1. Porter, C. C.; Watson, L. S.; Titus, D. C.; Totaro, J. A.; Byer, S. S. Biochem.
Pharmacol. 1962, 11, 1067.
140
120
100
80
2. Chase, T. N.; Watanabe, A. M. Neurology 1972, 22, 384.
3. Papavasiliou, P. S.; Cotzias, G. C.; Duby, S. E.; Steck, A. J.; Fehling, C.; Bell, M. A.
N. Eng. J. Med. 1972, 286, 8.
4. Fermaglich, J.; Chase, T. N. Lancet 1973, 1, 1261.
5. Vickers, S.; Stuart, E. K.; Bianchine, J. R.; Hucker, H. B.; Jaffe, M. E.; Rhodes, R. E.;
Vandenheuvel, W. J. Drug Metab. Dispos. 1974, 2, 9.
6. Vickers, S.; Stuart, E. K.; Hucker, H. B. J. Med. Chem. 1975, 18, 134.
7. Vickers, S.; Stuart, E. K. J. Pharm. Sci. 1973, 62, 1550.
8. Fatibello-Filho, O.; da Cruz Vieira, I. Analyst 1997, 122, 345.
9. Gasowska, B.; Frackowiak, B.; Wojtasek, H. Biochim. Biophys. Acta 2006, 1760,
1373.
B
60
10. Gasowska-Bajger, B.; Wojtasek, H. Bioorg. Med. Chem. Lett. 2008, 18, 3296.
11. Gasowska, B.; Kafarski, P.; Wojtasek, H. Biochim. Biophys. Acta 2004, 1673, 170.
12. Borovansky, J.; Edge, R.; Land, E. J.; Navaratnam, S.; Pavel, S.; Ramsden, C. A.;
Riley, P. A.; Smit, N. P. Pigment Cell Res. 2006, 19, 170.
40
20
13. Tompe, P.; Halbauer, A. N.; Ladanyi, L. Anal. Chim. Acta 1993, 273, 391.
14. Gilbert, J. A.; Frederick, L. M.; Ames, M. M. Clin. Cancer Res. 2000, 6, 4365.
15. Gilbert, J. A.; Frederick, L. M.; Pobst, L. J.; Ames, M. M. Biochem. Pharmacol. 2005,
69, 1159.
0
0
500
1000
Concentration [µM]
1500
2000
16. Miura, T.; Jimbow, K.; Ito, S. Int. J. Cancer 1990, 46, 931.
17. Gurney, H.; Coates, A.; Kefford, R. J. Invest. Dermatol. 1991, 96, 85.
18. Wick, M. M. J. Invest. Dermatol. 1987, 88, 535.
Figure 5. Toxicity of carbidopa (-j-) and L-dopa (-d-) against B16(F10) cells
in vitro: A—after 24 h, B—after 48 h.