Indirect oxidation of the antitumor agent procarbazine by tyrosinase-Possible application in designing anti-melanoma prodrugs

Article abstract of DOI:10.1016/j.bmcl.2008.04.041

The interaction of tyrosinase with the anticancer drug procarbazine has been investigated. In the presence of the enzyme alone no oxidation of this dialkylhydrazine above the background level was observed. However, when phenolic substrates (4-tert-butylcatechol or N-acetyl-l-tyrosine) were included in the reaction mixture, procarbazine was rapidly degraded. Oxygen consumption measurements showed that in a mixture both the phenolic substrate and the drug were oxidized. The major product of procarbazine degradation was isolated and identified as azoprocarbazine, the first active metabolite of this drug detected in previous in vivo and in vitro studies. This indirect oxidation of the hydrazine group in this anticancer agent indicates possible application of a hydrazine linker in construction of tyrosinase-activated anti-melanoma prodrugs.

Full text of DOI:10.1016/j.bmcl.2008.04.041

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3296–3300
Indirect oxidation of the antitumor agent procarbazine
by tyrosinase—Possible application in designing
anti-melanoma prodrugs
Beata Ga˛sowska-Bajger and Hubert Wojtasek*
Institute of Chemistry, University of Opole, Ul. Oleska 48, 45-052 Opole, Poland
Received 2 March 2008; revised 16 April 2008; accepted 16 April 2008
Available online 22 April 2008
Abstract—The interaction of tyrosinase with the anticancer drug procarbazine has been investigated. In the presence of the enzyme
alone no oxidation of this dialkylhydrazine above the background level was observed. However, when phenolic substrates (4-tert-
butylcatechol or N-acetyl-L-tyrosine) were included in the reaction mixture, procarbazine was rapidly degraded. Oxygen consump-
tion measurements showed that in a mixture both the phenolic substrate and the drug were oxidized. The major product of procar-
bazine degradation was isolated and identified as azoprocarbazine, the first active metabolite of this drug detected in previous in vivo
and in vitro studies. This indirect oxidation of the hydrazine group in this anticancer agent indicates possible application of a hydra-
zine linker in construction of tyrosinase-activated anti-melanoma prodrugs.
ꢀ 2008 Elsevier Ltd. All rights reserved.
Malignant melanoma is the most aggressive skin tumor
resulting from neoplastic transformation of melano-
cytes, which shows a steadily increasing incidence.1 De-
spite over 40 years of development of chemotherapy
there is currently no effective drug against this cancer
and new therapeutic agents are urgently needed. In the
early stages of the disease surgical intervention is usually
sufficient and gives good prognosis, but in cases of
metastasis there is little hope of survival. Dacarbazine
(DCIT) is the primary agent applied in chemotherapy,
frequently combined with other cytostatic agents
(tamoxifen, cisplatin, nitrozoureas, and vinca alkaloids),
with the CVD and Dartmouth regimens appearing to be
the most effective.1 However, positive responses are only
observed in about 30–40% of cases and complete remis-
sions are rare.2
steps: hydroxylation of L-tyrosine to L-Dopa and its
subsequent oxidation to dopaquinone, which undergoes
a series of non-enzymatic and enzymatic reactions lead-
ing to melanins.3 Tyrosinase can also convert other
monophenols and o-diphenols to o-quinones, which
are inherently cytotoxic. Utilization of tyrosinase in
the treatment of melanoma was therefore considered
long time ago. The first tyrosinase’s substrate displaying
promising cytotoxic effects against melanoma was 4-
hydroxyanisole.4,5 Later examples include 2,4-dihydrox-
yphenylalanine6 and 4-S-cysteaminylphenol and its
derivatives.7–10 In the past decades numerous analogues
of these compounds have been synthesized and tested
(reviewed in Ref. 11). However, only 4-hydroxyanisole
reached clinical trials and was withdrawn because of sig-
nificant renal and hepatotoxicity,12 later attributed to its
epoxidation by liver cytochromes P450, leading to the
formation of p-quinone.13
In mammals the metabolic pathway unique to melano-
cytes is melanogenesis, which offers the possibility of
developing a targeted chemotherapy specific to this tu-
mor. Tyrosinase (EC 1.14.18.1) is the key enzyme in this
pathway catalyzing the two initial and rate limiting
The possibility of selectively releasing cytotoxic agents
in melanocytes from tyrosinase-activated prodrugs has
recently also drawn substantial attention.14–17 This con-
cept has been named Melanocyte-directed Enzyme Pro-
drug Therapy.14 Initially, the cytotoxic agents were
attached to the primary amino group of tyrosinase sub-
strates (dopamine, tyrosine, and related compounds) by
carbamate or urea linkers.14–16 However, recent studies
on such dopamine derivatives have shown that the con-
Keywords: Tyrosinase; Hydrazine; Redox exchange; Melanoma;
Prodrug.
*
Corresponding author. Tel.: +48 77 452 7122; fax: +48 77 452
7101; e-mail: Hubert.Wojtasek@uni.opole.pl
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.04.041

B. Ga˛sowska-Bajger, H. Wojtasek / Bioorg. Med. Chem. Lett. 18 (2008) 3296–3300
3297
version of the amino group to amide, carbamate, or urea
derivatives makes the nitrogen atom insufficiently nucle-
ophilic for the cyclization of the corresponding o-qui-
nones to dihydroindoles and therefore the release of
the active group is unlikely.18 In the meantime, new
derivatives have been developed, where the effector part
has been connected to the aromatic amino group of 4-
aminophenol or 6-aminodopamine via a urea or thio-
urea linker.17
We first analyzed the reaction of procarbazine (Sigma-
Tau, Gaithersburg, MD, USA) with 4-tert-butyl-o-
benzoquinone generated by the oxidation of 4-tert-
butylcatechol (Fluka) with sodium periodate (POCh,
Gliwice, Poland) monitoring it spectrophotometrically.
Reactions were carried out as described previously for
the amino acid phenylhydrazides.24 4-tert-butylcatechol
(final concentration 0.25 or 0.5 mM, always in excess
over the oxidizing agent) was mixed with a solution of
sodium periodate (final concentration 0.1, 0.2 or
0.25 mM), the volume was adjusted to 2.6 ml with the
buffer (100 mM sodium phosphate, pH 6.8) and the
UV/vis spectrum from 300 to 600 nm was recorded after
1 min. In a parallel reaction, the same components were
mixed and after 1 min a solution of procarbazine (final
concentration 0.1 or 0.25 mM) was added to a final vol-
ume of 2.6 ml. UV/vis spectra were recorded immedi-
ately after adding procarbazine and then in 2 min
intervals until no changes were detected. This procedure
was applied to avoid dilution of the first reaction with
the solution of procarbazine. The addition of procarba-
zine led to instantaneous disappearance of 4-tert-butyl-
o-quinone reflected by a loss of absorbance in the
400 nm region (Fig. 1). This contrasted markedly with
the reaction of amino acid phenylhydrazides, which re-
quired 30–60 min for completion.24
Tyrosinase is also capable of oxidizing compounds other
than phenols and o-diphenols. Aromatic amines, o-ami-
nophenols, and o-diamines were described as substrates
for the enzyme from Neurospora crassa19 and Agaricus
bisporus,20 although the reaction rates were orders of
magnitude smaller than for the corresponding phenols
or catechols. Phenylhydrazine can also be oxidized by
this enzyme21 and the reaction leads to its irreversible
inactivation.22
Hydrazine derivatives are commonly found in nature,
either as natural compounds, industrial chemicals, or
pharmaceuticals. Drugs containing a hydrazine group
include the anti-parkinsonians carbidopa and bensera-
zide, the primary antituberculosis agent isoniazid, older
generation antidepressants iproniazid, nialamide, phe-
nelzine, and isocarboxazid, the antihypertensive hydral-
azine, and the antitumor agent procarbazine.23 Most of
these compounds are actually prodrugs, which are con-
verted to their active forms in the organism by oxidative
enzymes, primarily cytochrome P450 monooxygenases
and peroxidases. However, participation of tyrosinase
in the oxidation of any of these compounds has not been
taken into consideration.
We next carried out enzymatic oxidations of 4-tert-
butylcatechol (0.1 or 0.25 mM final concentration) by
5 or 20 lg of mushroom tyrosinase (specific activity
3216 U/mg, Sigma–Aldrich) without or with procarba-
zine (0.1, 0.2 or 0.3 mM final concentration). UV/vis
spectra recorded in the 300–600 nm range showed that
the final concentration of 4-tert-butyl-o-benzoquinone
(determined from the absorbance at 406 nm) was not
significantly affected by the presence of procarbazine,
but its formation was delayed (Fig. 2).
We have recently shown that amino acid phenylhydraz-
ides are oxidized by o-quinones, and therefore indirectly
by tyrosinase in the presence of phenols or o-diphe-
nols.24 Since the oxidation of these compounds to azo
derivatives leads to their rapid cleavage and release of
amino acids, we concluded that a hydrazine linker
may be utilized in construction of tyrosinase-activated
anti-melanoma prodrugs. To prove the feasibility of
this approach we decided to test the reaction with a
hydrazine containing anticancer drug used in medical
practice—procarbazine (N-isopropyl-4-[(N0-methylhyd-
razino)methyl]benzamide). Procarbazine was developed
as an antitumor agent in the 1960s and used extensively
against lymphoid malignancies, particularly Hodgkin’s
lymphomas and certain brain tumors. However, serious
side effects, including secondary leukemias and sterility
later led to its withdrawal, although it has been recently
reintroduced in combination therapies and yielded
impressive results.25 Activation pathway of this drug
starts with its oxidation to azoprocarbazine attributed
primarily to cytochrome P450 monooxygenases and per-
oxidases. Later reactions include tautomerization to
methylhydrazone, which can hydrolyze to p-formyl-N-
isopropylbenzamide and methylhydrazine, or undergo
further oxidation to azoxy derivatives.26–29 All these
compounds may decompose generating free radicals,
which are thought to be responsible for the cytotoxic
effects.30,31
Reactions of procarbazine with tyrosinase were then
monitored by oxygen consumption measurements.24
Reaction mixtures contained procarbazine (0.025, 0.05,
0.1, 0.2 or 0.3 mM), 4-tert-butylcatechol or N-acetyl-L-
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
300
350
400
450
500
550
600
Wavelength (nm)
Figure 1. Spectral changes after the addition of 0.25 mM procarbazine
to 4-tert-butyl-o-benzoquinone generated by oxidation of 0.5 mM 4-
tert-butylcatechol with 0.25 mM NaIO4. The spectra were recorded at
2 min intervals. The arrow indicates the direction of spectral changes
(decrease of 4-tert-butyl-o-benzoquinone absorbance).

3298
B. Ga˛sowska-Bajger, H. Wojtasek / Bioorg. Med. Chem. Lett. 18 (2008) 3296–3300
of oxygen consumption increased greatly and it was
0.3
0.25
0.2
A
approximately proportional to the concentration of the
phenolic substrate (Fig. 3)—44.1, 89.5, 140, and
260 nM/s for 0.01, 0.025, 0.05 and 0.1 mM of 4-tert-
butylcatechol, respectively. Total oxygen consumption
corresponded well to the concentration change required
for the oxidation of both compounds (e.g., 0.18 mM for
0.1 mM of catechol, whereas the theoretical value equals
0.2 mM) and far exceed that required for the oxidation
of 4-tert-butylcatechol alone (0.05 mM).
0.15
0.1
0.05
0
Similar experiments were performed with N-acetyl-L-
tyrosine. Oxidation of this monophenolic, non-cyclizing
substrate displays a characteristic lag phase. The pres-
ence of procarbazine in the reaction mixture reduced
the lag period significantly, indicating that it acted as a
reducing agent (Fig. 4). After the lag phase the reaction
rates depended solely on the concentration of the pheno-
lic substrate and not on the concentration of procarba-
zine. When procarbazine was the major substrate and
the concentration of N-acetyl-L-tyrosine varied, the
reaction rates increased after the lag period, proportion-
ally to the concentration of the phenolic substrate (Fig.
5).
300
350
400
450
500
550
600
Wavelength (nm)
0.3
0.25
0.2
B
0.15
0.1
0.05
0
300
350
400
450
500
550
600
0.3
Wavelength (nm)
1
2
3
0.25
0.2
0.3
0.25
0.2
C
0.15
0.1
4
5
1
0.05
0
0.15
0.1
0
300
600
900
t (s)
1200
1500
1800
0.05
0
2
Figure 3. Oxygen consumption measurements during the oxidation of
0.3 mM procarbazine (1) and its mixtures with 0.01 mM (2), 0.025 mM
(3), 0.05 mM (4) and 0.1 mM (5) 4-tert-butylcatechol by 42 lg of
tyrosinase.
0
5
10
15
20
25
30
t (min.)
Figure 2. Spectral changes during the oxidation of 0.25 mM 4-tert-
butylcatechol (A), and a mixture of 0.25 mM 4-tert-butylcatechol with
0.1 mM procarbazine (B) by 5 lg of tyrosinase. Spectra were recorded
at 4 min intervals. Changes of absorbance at 406 nm during oxidation
0.25 mM 4-tert-butylcatechol (1), and a mixture of 0.25 mM 4-tert-
butylcatechol with 0.1 mM procarbazine (2) by 5 lg of tyrosinase (C).
The arrow in (A) and (B) indicates the direction of spectral changes
(increase of 4-tert-butyl-o-benzoquinone absorbance).
0.3
0.25
1
0.2
tyrosine (0, 0.01, 0.025, 0.05 or 0.1 mM) and 42, 84, 126
or 168 lg of tyrosinase (to keep the enzyme concentra-
tion the same as in the spectrophotometric measure-
ments). Two sets of experiments were performed: the
concentration of procarbazine was kept constant and
the concentration of the phenolic substrate was varied
or vice versa. All the reactions were monitored for
60 min. Oxygen consumption in mixtures of procarba-
zine with the enzyme alone was very slow (3.17 nM/s).
However, in the presence of 4-tert-butylcatechol the rate
2
0.15
3
4
1800
0.1
0
300
600
900
t (s)
1200
1500
Figure 4. Oxygen consumption measurements during the oxidation of
0.2 mM N-acetyl-L-tyrosine (1) and its mixtures with 0.025 mM (2),
0.05 mM (3) and 0.1 mM (4) procarbazine by 84 lg of tyrosinase.

B. Ga˛sowska-Bajger, H. Wojtasek / Bioorg. Med. Chem. Lett. 18 (2008) 3296–3300
3299
0.3
0.25
0.2
400
A
350
1
300
250
200
150
100
50
2
3
4
0.15
0.1
5
0.05
0
0
0
300
600
900
t (s)
1200
1500
1800
0
5
10
15
20
25
30
t (min.)
Figure 5. Oxygen consumption measurements during the oxidation of
0.3 mM procarbazine (1) and its mixtures with 0.01 mM (2), 0.025 mM
(3) 0.05 mM (4) and 0.1 mM (5) N-acetyl-L-tyrosine by 168 lg of
tyrosinase.
600
500
400
300
200
100
0
B
To identify the products formed in these reactions from
procarbazine we performed HPLC analysis of a mixture
of procarbazine and 4-tert-butylcatechol with tyrosi-
nase. Reactions were carried in 1 ml of 100 mM sodium
phosphate buffer, pH 6.8 with 5 lg of tyrosinase at var-
ious concentrations of the substrates (0.1–0.5 mM).
Samples were analyzed immediately after the addition
of the enzyme and then after 15, 30, and 60 min. Sepa-
rations were performed on a ChromSep Inertsil 5
ODS-2 column (250 · 3 mm) connected to a Varian Po-
laris instrument with a dual wavelength UV/vis detector
and a 20 ll sample loop with 10 mM sodium phosphate
buffer, pH 6.8 and acetonitrile as a mobile phase (20–
80% acetonitrile gradient), at a flow rate of 0.4 ml/min.
Chromatograms were recorded at 230 and 280 nm.
Again, in the presence of the phenolic substrate rapid
disappearance of procarbazine was observed. It was re-
placed by one major (Rt = 16 min) and one minor prod-
uct (Rt = 14.5 min) (Fig. 6B). The major product was
isolated from a preparative reaction (0.4 mM procarba-
zine—12.8 mg, 0.1 mM 4-tert-butylchatechol and 1 mg
of tyrosinase in 125 ml of 10 mM sodium phosphate
buffer, pH 6.8). The reaction was monitored spectropho-
tometrically by taking the UV/vis spectra in the 300–
600 nm region until no change in the absorption range
of 4-tert-butyl-o-benzoquinone was observed. After
completion, the reaction mixture was frozen, lyophi-
lized, dissolved in 20 ml of water and separated on a
7 ml Bakerbond octadecyl column. After washing the
column with water, products were eluted with 40% ace-
tonitrile in water, 4 ml fractions were collected, analyzed
by TLC and their UV/vis spectra were taken. The frac-
tion containing the major product was lyophilized, dis-
0
5
10
15
20
25
30
t (min.)
Figure 6. Chromatographic analysis of a mixture of 0.5 mM procar-
bazine with 5 lg of tyrosinase (A) and a mixture of 0.25 mM 4-tert-
butylcatechol, 0.25 mM procarbazine and 5 lg of the enzyme (B) after
30 min incubation. Retention times determined by analysis of stan-
dards were 9 min for procarbazine, 20.5 min for 4-tert-butyl-o-benzo-
quinone, and 21.5 min for 4-tert-butylcatechol (not seen in the
presented chromatogram). The major product with a retention time
of 16 min has been identified as azoprocarbazine. The minor product
with a retention time of 14.5 min has been tentatively identified as p-
formyl-N-isopropylbenzamide methylhydrazone.
NMR (a singlet at 3.00 ppm) spectra of its mixture with
the major product indicated that it was the hydrazone
tautomer of azoprocarbazine (p-formyl-N-isopropylben-
zamide methylhydrazone) decomposing slowly to the
aldehyde (p-formyl-N-isopropylbenzamide—a small
NMR signal at 10.08 ppm was also observed in these
samples). As already indicated by oxygen consumption
measurements, very little conversion of procarbazine
was observed when the compound was incubated with
tyrosinase alone (Fig. 6A)—the amount of azoprocarb-
azine was similar to that obtained as a result of autoox-
idation in the absence of the enzyme (data not shown).
These results clearly demonstrate that in a mixture with
monophenols or o-diphenols procarbazine is oxidized by
tyrosinase indirectly through o-quinones (Scheme 1).
1
solved in CDCl3 and its H NMR spectrum was taken
(Bruker Avance 300 MHz). The product was unequivo-
cally identified as azoprocarbazine. The signals of the
methyl and methylene groups showed characteristic
shifts from 2.81 to 3.80 and from 4.19 to 4.95 ppm,
respectively.29 The minor product could not be purified
completely by standard gravitational chromatography.
However, its chromatographic behavior (elution from
a reverse-phase column slightly before azoprocarb-
azine29), and the UV/vis (maximum at 300 nm26) and
We believe that this reaction offers an alternative ap-
proach for the preparation of anti-melanoma prodrugs.
Combination of a phenolic substrate with a hydrazine
linker would create compounds from which the release
of the active component after activation by tyrosinase
would not occur as a result of a nucleophilic attack on

3300
B. Ga˛sowska-Bajger, H. Wojtasek / Bioorg. Med. Chem. Lett. 18 (2008) 3296–3300
6. Morrison, M. E.; Yagi, M. J.; Cohen, G. Proc. Natl. Acad.
Sci. U.S.A. 1985, 82, 2960.
7. Miura, S.; Jimbow, K.; Ito, Y.; Ueda, T.; Ito, S. J. Invest.
Dermatol. 1985, 84, 320.
8. Miura, S.; Ueda, T.; Jimbow, K.; Ito, S.; Fujita, K. Arch.
Dermatol. Res. 1987, 279, 219.
O
H
N
H
N
O
N
H
+
O
R
9. Yamada, I.; Seki, S.; Matsubara, O.; Ito, S.; Suzuki, S.;
Kasuga, T. J. Invest. Dermatol. 1987, 88, 538.
10. Ito, S.; Kato, T.; Ishikawa, K.; Kasuga, T.; Jimbow, K.
Biochem. Pharmacol. 1987, 36, 2007.
11. Riley, P. A. Pigment Cell Res. 2003, 16, 548.
12. Rustin, G. J.; Stratford, M. R.; Lamont, A.; Bleehen, N.;
Philip, P. A.; Howells, N.; Watfa, R. R.; Slack, J. A. Eur.
J. Cancer 1992, 28A, 1362.
OH
H
N
N
N
HO
+
O
R
13. Moridani, M. Y.; Cheon, S. S.; Khan, S.; O’Brien, P. J.
Drug Metab. Dispos. 2002, 30, 1063.
Scheme 1. Redox reaction occurring between procarbazine and o-
quinones.
14. Jordan, A. M.; Khan, T. H.; Osborn, H. M.; Photiou, A.;
Riley, P. A. Bioorg. Med. Chem. 1999, 7, 1775.
15. Jordan, A. M.; Khan, T. H.; Malkin, H.; Osborn, H. M.;
Photiou, A.; Riley, P. A. Bioorg. Med. Chem. 2001, 9,
1549.
the o-quinone moiety but its redox exchange with the
linker. Acyl hydrazine derivatives may be used for rapid
release of a variety of cytotoxic agents, whereas alkyl
hydrazines, for example, based on the structure of pro-
carbazine, may serve as sources of alkyl radicals or carb-
ocations. Preparation of such compounds is now under
way.
16. Jordan, A. M.; Khan, T. H.; Malkin, H.; Osborn, H. M.
Bioorg. Med. Chem. 2002, 10, 2625.
17. Knaggs, S.; Malkin, H.; Osborn, H. M.; Williams, N. A.;
Yaqoob, P. Org. Biomol. Chem. 2005, 3, 4002.
18. 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.
19. Toussaint, O.; Lerch, K. Biochemistry 1987, 26, 8567.
20. Gasowska, B.; Kafarski, P.; Wojtasek, H. Biochim.
Biophys. Acta 2004, 1673, 170.
Acknowledgments
21. Lerner, H. R.; Mayer, A. M.; Harel, E. Phytochemistry
1974, 13, 397.
22. Lerner, H. R.; Harel, E.; Lehman, E.; Mayer, A. M.
Phytochemistry 1971, 10, 2637.
23. Toth, C. In vivo 2000, 14, 299.
We thank Sigma-Tau, Gaithersburg, USA, for the gen-
erous gift of a sample of procarbazine. This work was
supported in part by a Grant form the Polish Ministry
of Science and Higher Education No. 2 P05F 00330.
24. Gasowska, B.; Frackowiak, B.; Wojtasek, H. Biochim.
Biophys. Acta 2006, 1760, 1373.
25. Massoud, M.; Armand, J. P.; Ribrag, V. Eur. J. Cancer
2004, 40, 1924.
References and notes
26. Dunn, D. L.; Lubet, R. A.; Prough, R. A. Cancer Res.
1979, 39, 4555.
27. Moloney, S. J.; Prough, R. A. Arch. Biochem. Biophys.
1983, 221, 577.
1. Buzaid, A. C. Crit. Rev. Oncol. Hematol. 2002, 44, 103.
2. O’Day, S. J.; Kim, C. J.; Reintgen, D. S. Cancer Control
2002, 9, 31.
3. Sanchez-Ferrer, A.; Rodriguez-Lopez, J. N.; Garcia-
Canovas, F.; Garcia-Carmona, F. Biochim. Biophys. Acta
1995, 1247, 1.
4. Riley, P. A. J. Pathol. 1969, 97, 185.
5. Riley, P. A. J. Pathol. 1970, 101, 163.
28. Moloney, S. J.; Wiebkin, P.; Cummings, S. W.; Prough, R.
A. Carcinogenesis 1985, 6, 397.
29. Weinkam, R. J.; Shiba, D. A. Life Sci. 1978, 22, 937.
30. Sinha, B. K. Free Radic. Res. Commun. 1991, 15, 189.
31. Sinha, B. K. Biochem. Pharmacol. 1984, 33, 2777.