New L-dopa codrugs as potential antiparkinson agents

Article abstract of DOI:10.1002/ardp.200700228

This paper reports the synthesis and preliminary evaluation of new L-dopa (LD) conjugates (1 and 2) obtained by joining LD with two different natural antioxidants, caffeic acid and carnosine, respectively. The antioxidant efficacy of compounds 1 and 2 was

Full text of DOI:10.1002/ardp.200700228

412
Arch. Pharm. Chem. Life Sci. 2008, 341, 412–417
Full Paper
New L-Dopa Codrugs as Potential Antiparkinson Agents
Piera Sozio1, Antonio Iannitelli1, Laura Serafina Cerasa1, Ivana Cacciatore1, Catia Cornacchia1,
Gianfabio Giorgioni2, Massimo Ricciutelli2, Cinzia Nasuti3, Franco Cantalamessa3, and
Antonio Di Stefano1
1 Dipartimento di Scienze del Farmaco, Universitꢀ ”G. D'Annunzio”, Chieti, Italy
2 Dipartimento di Scienze Chimiche, Universitꢀ di Camerino, Camerino (MC), Italy
3 Dipartimento di Scienze Farmacologiche e Medicina Sperimentale, Universitꢀ di Camerino, Camerino (MC),
Italy
This paper reports the synthesis and preliminary evaluation of new L-dopa (LD) conjugates (1 and
2) obtained by joining LD with two different natural antioxidants, caffeic acid and carnosine,
respectively. The antioxidant efficacy of compounds 1 and 2 was assessed by evaluating plas-
matic activities of superoxide dismutase (SOD) and glutathione peroxidase (GPx) in the rat. Rat
striatal concentration of LD and dopamine (DA), and central nervous effects were evaluated after
oral administration of the codrugs 1 and 2. The results suggest that, though our codrugs are
devoid of significant antioxidant activity, they are able to induce sustained delivery of DA in rat
striatum and can improve LD and DA release in the brain.
Keywords: Caffeic acid / Carnosine / L-dopa / L-dopa codrugs / Parkinson's disease /
Received: November 8, 2007; accepted: February 1, 2008
DOI 10.1002/ardp.200700228
scavenging activity of antioxidant molecules covalently
Introduction
linked with LD, we have focused our attention on caffeic
acid and carnosine. Caffeic acid and its derivative clova-
mide (N-caffeoyl-L-dopa) have been reported to com-
pletely block the production of ROS [4]. Their radical scav-
enging and antioxidative activities are mainly due to the
presence of two phenolic alcohol groups at ortho posi-
tions. Carnosine, synthesized in the skeletal muscle or
brain tissue by carnosine synthetases, prevents ROS-
induced peroxidation of lipids and cell death [5, 6].
Recent reports suggest that carnosine has potential ther-
apeutic applications in many diseases of the central nerv-
ous system such as Alzheimer's disease, PD, and cerebral
ischemic diseases [7–10]. Our studies focused on provid-
ing two novel codrugs (1 and 2) obtained by joining the
3,4-diacetyloxy-L-dopa methyl ester with caffeic acid or
carnosine (Fig. 1). These compounds were evaluated as
potential antiparkinson drugs with antioxidant proper-
ties. The present paper reports the physicochemical (lipo-
philicity and solubility), pharmacokinetic (rates of chem-
ical and enzymatic hydrolysis), antioxidant (activities of
SOD and GPx), and dopaminergic (effects on spontaneous
locomotor activity of rats) properties of both codrugs.
L-Dopa (LD) is a direct precursor of dopamine (DA), which
is deficient in the brains of patients suffering from Par-
kinson's disease (PD). Although LD is the treatment of
choice, its metabolism generates a variety of cytotoxic
reactive oxygen species (ROS) that contribute to the pro-
gression of the disease, increasing the loss of nigrostria-
tal dopaminergic neurons [1]. Therefore, employing anti-
oxidant molecules to scavenge ROS is an important strat-
egy for preventing or slowing PD. We have recently dem-
onstrated that multifunctional codrugs containing LD
and (R)-a-lipoic acid or glutathione seem to protect parti-
ally against the oxidative stress deriving from auto-oxida-
tion and monoamine oxidase (MAO)-mediated metabo-
lism of DA [2, 3]. In order to further explore the radical-
Correspondence: Antonio Di Stefano, Dipartimento di Scienze del Farm-
aco, Universitꢀ ”G. D'Annunzio”, Via dei Vestini 31, 66100 Chieti, Italy.
E-mail: adistefano@unich.it
Fax: +39 871 355-4706
Abbreviations: dopamine (DA); L-dopa (LD); glutathione peroxidase
(GPx); monoamine oxidase (MAO); Parkinson's disease (PD); reactive
oxygen species (ROS); superoxide dismutase (SOD)
i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Chem. Life Sci. 2008, 341, 412–417
Antiparkinson Codrugs
413
Reagents: a) DMF, DCM, Et3N, PyBOP.
Scheme 1. Synthesis of codrug 1.
aqueous solutions was also evaluated, and the results are
listed in Table 1. Codrug 2 showed good water solubility
(more than 9 mg/mL at pH 7.4) necessary for good oral
absorption, while the log P was high only for compound
1 (more than 2).
Figure 1. Chemical structures of multifunctional codrugs 1 and
2.
Biological testing
Chemistry/Synthesis
The new compounds were also investigated in vitro to
evaluate their chemical and enzymatic stability. The reac-
tivities to chemical hydrolysis in buffer solutions at
pH 1.3, 5.0, and 7.4, were evaluated by pseudo-first-order
rate constants. The rate data showed that the two
codrugs are stable in aqueous solutions; in particular, at
pH 1.3 all compounds are stable (tꢀ A 2.94 h) and at
pH 5.0 and 7.4 they are stable enough (tꢀ A 0.44 h) to be
absorbed intact from the intestine (see Table 2). In rat
plasma, catechol esters and amide bonds of the studied
Codrug 1 was synthesized through the interaction of 3,4-
diacetyloxy-L-dopa methyl ester hydrochloride 3 with caf-
feic acid 4 (Scheme 1) [11–13]. Compound 2 was prepared
by condensation of N-acetyl-3,4-diacetyloxy-L-dopa 5 and
carnosine methyl ester hydrochloride 6 (Scheme 2 and
Section 4, Experimental).
Coefficient partition (log P) of conjugates 1 and 2 was
determined in n-octanol/phosphate buffer, pH 7.4, by sat-
uration of the shake-flask method [13, 14]. Solubility in
Table 1. Physicochemical properties of codrugs 1 and 2.
Compound
logPa)
Solubility in watera)
(mg/mL)
Solubility at pH 1.3a) Solubility at pH 7.4a) Solubility at pH 5.0a)
(mg/mL)
(mg/mL)
(mg/mL)
1
2.29 (l 0.11)
–1.12 (l 0.06)
0.26610– 3 (l 1610– 4
29.00 (l 1.45)
)
2.08610– 3
(l 0.10610– 3
35.53 (l 1.68)
0.10610– 3
(l 0.01610– 3
9.10 (l 0.04)
1.17610– 3
(l 0.06610– 3
27.55 (l 1.38)
)
)
)
2
a)
Values are means of three experiments; standard deviation is given in parentheses.
Table 2. Kinetic data for chemical hydrolysis of codrugs 1 and 2 at 378C.
Compound
pH 1.3a)
Kobs (h– 1
pH 5.0a)
Kobs (h– 1
pH 7.4a)
tꢀ (h)
)
tꢀ (h)
)
tꢀ (h)
Kobs (h– 1
)
1
2
2.94 (l 0.10)
147.00 (l 7.39)
0.24 (l 0.01)
1.77 (l 0.07)
18.81 (l 0.94)
0.40 (l 0.02)
0.44 (l 0.02)
10.15 (l 0.50)
1.57 (l 0.07)
0.005 (l 0.2610– 3
)
0.037 (l 0.2610– 3
)
0.068 (l 3.3610– 3
)
a)
Values are means of three experiments; standard deviation is given in parentheses.
i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

414
P. Sozio et al.
Arch. Pharm. Chem. Life Sci. 2008, 341, 412–417
Table 3. Rate constants for the hydrolysis of codrugs 1 and 2 in
80% rat plasma and 80% human plasma at 378C.
Compound
Rat plasmaa)
Human plasmaa)
Kobs (min – 1
)
tꢀ
Kobs
tꢀ (min)
1
2
immediate hydrolysis
immediate hydrolysis
/
/
6.53 (l 0.23) 0.11 (l 0.4610 – 2
22.29 (l 1.07) 0.031 (l 0.5610 – 3
)
)
a)
Values are means of three experiments; standard deviation is
given in parentheses.
can use specific transporters to enter the CNS and act as
antioxidant agent [15, 16]. Figure 2 shows LD plasma con-
centration trends obtained in rats over time following
administration of codrugs 1, 2, and LD. The values of the
LD-plasma concentration at 3 h post-dose were about
85 lg/mL for compound 2 and 80 lg/mL for compound 1.
A rapid decrease of concentration levels was observed 2 h
after LD administration. The results indicate that
codrugs 1 and 2 are able to prolong plasma LD levels and
could be beneficial in the treatment of motor dysfunc-
Reagents: a) DMF, Et3N, DCC.
Scheme 2. Synthesis of codrug 2.
derivatives were cleaved and LD was released in one step tions directly related to LD plasma-level fluctuations
with an observed immediate conversion. Hydrolysis in [17, 18].
human plasma proceeds more slowly with formation of
LD. These derivatives were identified by LC/MS and NMR
analysis and the data are listed in Table 3.
The new LD codrugs were also evaluated comparing
neostriatal LD and DA levels after administration of com-
pounds 1, 2, and LD. A previously reported high perform-
The degradation process was found to be correlated ance liquid chromatography (HPLC) method with electro-
with first-order kinetics, and LD was released in quantita- chemical detection (EC) was utilized [19]. The changes in
striatal levels of LD and DA are shown in Fig. 3. The stud-
tive amounts. The high water solubility of codrug 2 meets
the requirements for gastro-intestinal absorption, while ied compounds seem able to induce sustained delivery of
the low lipophilicity (log P = –1.12) probably prevents both LD and DA in rat striatum with respect to an equi-
molar dose of LD; the striatal levels of DA were still ele-
the cross of the blood-brain barrier (BBB). Codrug 2, by
vated after 10 h (about 50 pmol/mg for compound 1).
peripheral hydrolysis, produces LD and carnosine which
Figure 2. Plasma-concentration profile
of LD after administration of LD, 1, and
2 in rats. Data are expressed as mean
l SE. Each experiment was performed
in triplicate.
i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2008, 341, 412–417
Antiparkinson Codrugs
415
Figure 3. Rat striatal levels of LD and DA after administration of LD, 1 and 2. Data are expressed as mean l SE. Each experiment
was performed in triplicate.
Figure 4. Locomotion, rearing, and
grooming, 1.5 h after administration of
LD, 1, and 2 in rats. Data are expressed
as mean l SE of five experiments. * P a
a
0.05 compared to control group; P a
0.05 compared to LD-treated group.
erature, caffeic acid and carnosine act as natural antioxi-
dants with hydroxyl-radical-scavenging and lipid-peroxi-
Conclusion
In this study, we assessed the antioxidant efficacy of the
codrugs 1 and 2, evaluating, in the plasma of rats, the
activity of superoxide dismutase (SOD) and glutathione
peroxidase (GPx), two enzymes critical for protection
against oxidative stress, produced by LD [20, 21]. Table 4
shows the activities of these antioxidant enzymes that
form the primary defense system against ROS in the
plasma 1.5 h after drug administration in the rats. No sig-
nificant modifications were revealed in the SOD and GPx
activities among the four groups treated with vehicle,
LD, 1, and 2. For these reasons, the studied codrugs do
not seem to protect against the oxidative stress deriving
from the MAO-mediated metabolism of DA. From the lit-
dase activities and can modulate the SOD and GPx activ-
ities when there are strong oxidative stress conditions. In
our experimental conditions (after only one drug admin-
istration), we suppose that the antioxidant efficacy does
not appear because the ROS levels are too low [9, 22, 23].
To evaluate the dopaminergic activity, we studied the
effects of 1 and 2 on the spontaneous locomotor activity
of rats in comparison with LD-treated animals. As can be
observed in Fig. 4, drug administration 1.5 h after gavage
led to a pattern of behavioral depression in the treated
groups compared to the control group, which received
vehicle only. The results obtained after treatment with 2
showed less behavioral depression on locomotion
i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

416
P. Sozio et al.
Arch. Pharm. Chem. Life Sci. 2008, 341, 412–417
Table 4. Values of superoxide dismutase (SOD) and glutathione
peroxidase (GPx) activities, 1.5 h after administration of LD, 1,
and 2 in rats.
SOD (U/g N Hb)a)
GPx (U/g N Hb)a)
Control
LD
1
143 (l12)
171 (l16)
149 (l3)
821 (l25)
760 (l11)
795 (l3)
2
167 (l14)
801 (l16)
a)
Data are expressed as mean l SE of five experiments.
group. As observed for locomotor activity, treatment
with LD, 1, and 2 also led to a pattern of behavioral
depression in anxiety-like behavior (Fig. 5). In addition,
treatment with 2 induced a significant (P a 0.05) increase
of entries and time spent in the center of the open field
compared with the LD-treated group, as shown in Fig. 5.
The data obtained allow us to distinguish the different
pharmacokinetics of the codrugs used at the same molar
doses as LD. It seems that, compared to LD treatment, the
two codrugs induce fewer sedative effects impairing loco-
motion and anxiety-related behavior. This may be due to
the slower biodistribution of codrug 2 in the CNS: the
lower values of plasma concentration of LD, 1.5 h after
gavage, could induce a slower DA delivery to the CNS
and, thus, there would be fewer effects on behavioral
activities.
On the whole, these results suggest that our codrugs
are devoid of significant antioxidant activity but can
improve the release of LD and DA in the brain. Neverthe-
less, to better investigate the bio-availability of com-
pounds 1 and 2, ”in-vivo” microdialysis studies in rat stria-
tum are in progress.
Financial support from the Ministero dell' Istruzione, dell'Uni-
versitꢀ e della Ricerca (MIUR) is gratefully acknowledged.
The authors have declared no conflict of interest.
Figure 5. Anxiety test 1.5 h after administration of LD, 1, and 2
in rats. Data are expressed as mean l SE of five experiments. *
P a 0.05 compared to control group; a P a 0.05 compared to LD-
treated group.
Experimental
The purity of all new compounds was checked by HPLC using the
column Waters X-Terra RP18 (5 lm, 3.0615 mm) with MeOH /
H2O (60 : 40) as eluent. The microanalyses results were within
l 0.4%. 1H-NMR spectra were recorded at 300 MHz in DMSO-d6 as
solvent.
Compound 1: d ppm 9.41 (s, 1H, Ar-OH), 9.15 (s, 1H, Ar-OH),
8.48 (d, 1H, J = 270 Hz, NHCO), 7.34–6.60 (m, 6H, ArH), 6.52 (d,
1H, J = 270 Hz, =CH-Ar), 6.34 (d, 1H, J = 360 Hz, CHCO), 4.55 (q,
1H, aCH), 3.58 (s, 3H, COOCH3), 2.85 (m, 2H, CH2), 2.23 (s, 6H,
26CH3CO).
(975.56 l 2.85) compared to the LD-treated group
(555.5 l 97.92) (P a 0.05).
A similar pattern can be observed in the anxiety-like
behavior 1.5 h after drug administration (Fig. 5). As
shown, treatment with LD and codrugs 1, 2 induced a sig-
nificant decrease of entries (P a 0.05) and time spent in
the center of the open field compared with the control
Compound 2: d ppm 1.91 (s, 3H, CH3CON), 2.1 (s, 6H, CH3CO),
2.21–2.41 (m, 2H, bCH2), 2.60–2.79 (m, 2H, aCH2), 2.8–3.19 (m,
i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2008, 341, 412–417
Antiparkinson Codrugs
417
4H, 26CH2), 3.66 (s, 3H, OCH3), 4.38–4.44 (t, 1H, aCH), 4.6–4.72
(t, 1H, aCH), 6.42–6.71 (m, 3H, ArH), 6.87 (s, 1H, CH), 7.62 (s, 1H,
CH).
[12] S. H. Park, S. H. Kang, S. H. Lim, H. S. Oh, K. H. Lee, Bioorg.
Med. Chem. Lett. 2003, 13, 3455–3459.
[13] C. Chavis, F. Grodenic, J.-L. Imbach, Eur. J. Med. Chem. 1981,
16, 219–227.
[14] D. J. W. Grant, T. Higuchi in Solubility behaviour of organic
compounds, Wiley, New York, 1990, pp. 164–175.
References
[15] W. H. Streng in Characterization of Compounds in Solution –
Theory and practice, Kluver Academic Plenum Publisher,
New York, 2001, pp.161–180.
[1] D. G. Graham, Mol. Pharmacol. 1978, 14, 633–643.
[2] A. Di Stefano, P. Sozio, A. Cocco, A. Iannitelli, et al., J. Med.
Chem. 2006, 49, 1486–1493.
[16] N. S. Teuscher, R. F. Keep, D. E. Smith, Pharmaceut. Res.
2001, 18, 807–813.
[3] F. Pinnen, I. Cacciatore, C. Cornacchia, P. Sozio, et al., J.
Med. Chem. 2007, 50, 2506–2515.
[17] R. Kohen, Y. Yamamoto, K. C. Cundy, B. N. Ames, Proc.
[4] J. P. Ley, H. J. Bertram, J. Agric. Food Chem. 2003, 51, 4596–
Natl. Acad. Sci. USA 1988, 85, 3175–3179.
4602.
[18] U. Gundert-Remy, R. Hildebrand, A. Stiehl, E. Weber, et al.,
Eur. J. Clin. Pharmacol. 1983, 25, 69–72.
[5] R. Kohen, Y. Yamamoto, K. C. Cundy, B. N. Ames, Proc. Natl.
Acad. Sci. USA 1988, 85, 31753179.
[19] M. Da Prada, H. H. Keller, L. Pieri, L. Kettler, W. E. Haefely,
[6] S. D. Marchis, C. Modena, P. Peretto, A. Migheli, et al., Bio-
chemistry 2000, 65, 824–833.
Experientia 1984, 40, 1165–1172.
[20] C. Nasuti, R. Gabbianelli, M. L. Falcioni, A. Di Stefano, et
al., Toxycology 2007, 229, 194–205.
[7] A. N. Fonteh, R. J. Harrington, A. Tsai, P. Liao, M. G. Har-
rington, Amino Acids 2007, 32, 213–224.
[21] C. Michiels, M. Raes, O. Topussaint, J. Remacle, Free Radic.
Biol. Med. 1994, 17, 235–248.
[8] J. E. Preston, A. R. Hipkiss, D. T. Himsworth, I. A. Romero, J.
N. Abbott, Neurosci. Lett. 1998, 242, 105–108.
[22] A. J. Nappi, E. Vass, Biochim. Biophys. Acta 1997, 1336, 295–
[9] J. H. Kang, K. S. Kim, S. Y. Choi, H. Y. Kwon, et al., Biochim.
Biophys. Acta 2002, 1570, 89–96.
301.
[23] U. J. Jung, M. K. Lee, Y. B. Park, S. M. Jeon, M. S. Choi, J.
[10] A. Boldyrev, R. Song, D. Lawrence, D. O. Carpenter, Neuro-
science 1999, 94, 571–577.
Pharmacol. Exp. Ther. 2006, 318, 476–483.
[11] N. Bodor, K. B. Sloan, T. Higuchi, K. Sasahara, J. Med. Chem.
1977, 20, 1435–1445.
[24] M. A. Babizhayev, M. C. Seguin, J. Gueyne, R. P.
Evstigneeva, et al., Biochem. J. 1994, 304, 509–516.
i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com