Archives of Biochemistry and Biophysics 546 (2014) 1–7

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi

Mammalian dopa decarboxylase: Structure, catalytic activity and inhibition q Mariarita Bertoldi ⇑ Section of Biochemistry, Department of Life and Reproduction Sciences, University of Verona, Strada Le Grazie 8, 37134 Verona, Italy

a r t i c l e

i n f o

Article history: Received 13 October 2013 and in revised form 21 December 2013 Available online 6 January 2014 Keywords: Mammalian Dopa decarboxylase Mechanism Kinetic analysis Decarboxylation Oxidative deamination Inhibitors

a b s t r a c t Mammalian Dopa decarboxylase catalyzes the conversion of L-Dopa and L-5-hydroxytryptophan to dopamine and serotonin, respectively. Both of them are biologically active neurotransmitters whose levels should be finely tuned. In fact, an altered concentration of dopamine is the cause of neurodegenerative diseases, such as Parkinson’s disease. The chemistry of the enzyme is based on the features of its coenzyme pyridoxal 50 -phosphate (PLP). The cofactor is highly reactive and able to perform multiple reactions, besides decarboxylation, such as oxidative deamination, half-transamination and Pictet–Spengler cyclization. The structure resolution shows that the enzyme has a dimeric arrangement and provides a molecular basis to identify the residues involved in each catalytic activity. This information has been combined with kinetic studies under steady-state and pre-steady-state conditions as a function of pH to shed light on residues important for catalysis. A great effort in DDC research is devoted to design efficient and specific inhibitors in addition to those already used in therapy that are not highly specific and are responsible for the side effects exerted by clinical approach to either Parkinson’s disease or aromatic amino acid decarboxylase deficiency. Ó 2014 Elsevier Inc. All rights reserved.

Background Dopa decarboxylase (DDC,1 EC 4.1.1.28) was first identified in 1938 in mammalian kidney tissue, revealing that it catalyzes an essential step in epinephrine biosynthesis. Subsequently, it was reported that the reaction specificity of the enzyme is broader, since, in addition to decarboxylate L-Dopa to dopamine, it is able to transform L-5-hydroxytryptophan to serotonin and, although much less efficiently, also other aromatic amino acids such as p-tyrosine, tryptophan and phenylalanine to the corresponding amines (trace aromatic amines). For this reason, DDC is also described, more correctly, as aromatic amino acid decarboxylase (AADC). Its role is to supply organism with essential neurotransmitters. Despite its low substrate specificity towards different aromatic amino acids, DDC plays a key role in controlling aromatic amines level. Dopamine and serotonin act as neuro-modulators and can influence mood reg-

q This article was originally intended for publication in ABB Special Issue: Cofactor Assisted Enzymatic Catalysis. 544, (15 February 2014) http://www. sciencedirect.com/science/journal/00039861/544. ⇑ Fax: +39 045 8027170. E-mail address: [email protected] 1 Abbreviations used: DDC, dopa decarboxylase; AADC, aromatic amino acid decarboxylase; PLP, pyridoxal 50 -phosphate; UPR, unfolded protein response; PD, Parkinson’s disease.

http://dx.doi.org/10.1016/j.abb.2013.12.020 0003-9861/Ó 2014 Elsevier Inc. All rights reserved.

ulation, cognitive and physiological homeostasis as well as motor coordination [1–4]. Not only their synthesis, but also their metabolic and uptake pathways are strictly connected. Serotonin depletion is responsible for some forms of major depression, anxiety and aggression. Dopamine depletion is responsible for lack of motivation and for anhedonia. When there is an imbalance with either one, there is an increased chance of the other becoming unbalanced as a result of their interdependent nature. Balancing serotonin, dopamine, and other biogenic amines such as norepinephrine is the focus of research into treating many mood disorders. A deficiency in DDC synthesis, due to non-sense or frameshift mutations, or an alteration in its activity, due to single amino acid substitutions generating an aberrant gene product, leads to pathological conditions related to impaired cognitive and physiological homeostasis as well as motor coordination [1–4] and/or neuropsychiatric disorders [5,6]. DDC deficiency is a rare recessive disorder due to mutations in the DDC gene and to the related inability to synthesize dopamine and serotonin. This leads to severe developmental delay [7]. A comprehensive review of known variants involved and their possible structure to function relationship is given in [8]. Moreover, DDC is not considered to be rate-limiting in physiological catecholamines or indoleamines synthesis, but it becomes rate-limiting in several pathological states related to aberrant dopamine production, such as Parkinson disease (PD) or the

2

M. Bertoldi / Archives of Biochemistry and Biophysics 546 (2014) 1–7

bipolar syndrome [9–11]. In particular, PD is a chronic progressive neurological disorder characterized by tremor, bradykinesia, rigidity and postural instability. These symptons are caused by the low levels of dopamine resulting from the degeneration of dopamine-producing cells in the substantia nigra of the brain [12]. DDC is widespread in mammals, insects and plants. In mammals, the enzyme is found in neuronal and non-neuronal tissues. While its presence in neuronal cells is in line with its activity in neuro-modulators synthesis, it is more difficult to explain its abundance in other tissues, for example the kidney, where the enzyme is highly expressed. In insects, in addition to tyramine and octopamine, the major insect hormones, also dopamine and serotonin synthesized by DDC play crucial roles in the behaviour and development of insects [13], and are precursors of compounds contributing to sclerotization of cuticle [14]. Finally, in plants there exist two forms of substrate-specific DDC: a tyrosine decarboxylase and a tryptophan decarboxylase whose peculiar role is related to plant hormone synthesis and plant maturation [15]. Moreover vegetal DDC is also able to produce aromatic alkaloids useful as precursors of pharmaceutically active molecules [15]. Mammalian DDC has been purified from different sources (pig [16], rat [17,18] and human [19–21], although the best studied is the pig kidney enzyme. All three DDCs were cloned and expressed in Escherichia coli [19–23] and kinetic, mechanistic and inhibition studies were further performed. The human versus pig DDC shares 90% sequence identity with 95% positives (this percentage represents the fraction of residues that are either identical or similar), the human versus rat enzyme shares 89% sequence identity with 96% positives. Thus, the three enzymes are closely related and the essential amino acid residues are conserved. DDC belongs to the a-family of aminotransferases (Fold-Type I), in particular to the subgroup II of a-decarboxylases to which also glutamate decarboxylase and histidine decarboxylase belong. Their chemistry and reaction mechanism bases on the chemical features of the coenzyme: pyridoxal 50 -phosphate (PLP) [24]. Moreover, they share common features, such as similar active site architectures and conserved essential amino acidic residues. The resolution of the structure of histidine decarboxylase in the presence of the inhibitor histidine methyl ester [25] revealed an active site conformation similar to that of DDC in the complex with carbiDopa [26] except for one residue, i.e. Ser-354, of histidine decarboxylase replaced by glycine in the same position in DDC. Interestingly, the S354G variant of histidine decarboxylase presents a decreased activity towards histidine and an increased activity towards L-Dopa [25]. In addition, both decarboxylases possess a mobile loop considered essential for catalysis containing a tyrosine residue (Tyr-332 in DDC and Tyr-334 in histidine decarboxylase) playing a critical role in catalytic activity [25,27]. Until now, much work has been reported on DDC, although reviews found in the literature are mainly focused on its role in PD or on other pathological states related to enzyme deficiency. Several papers, in fact, deal with the presence or the abundance of DDC in various areas of central nervous system and other neuronal cell types. Another line of research is focused on the search for inhibitors with the aim of slowing down, at least, the progression of PD. Finally, a few reviews are involved in structure to function relationship of DDC, while many research articles on catalysis and mechanism exist and are widespread in literature. The study of the biochemistry of DDC pointing to the molecular approach is worthy since it could aid in filling the gap between enzymatic features and cause-to-effect disease relationship. In this review, an update on DDC kinetic and mechanistic studies will be provided as well as a possible correlation of the features of mammalian DDCs with their functions.

Structural determinants of mammalian DDC Up to now, the structures of native holo pig kidney DDC, ligandbound holo pig kidney DDC and apo human DDC have been solved [26,28]. Overall these structures contribute to provide a picture of active site organization. In all three cases, the enzyme is a structural and functional dimer with the typical arrangement of FoldType I PLP-enzymes. Previous structural studies of superimposition of the overall sequence and structure of DDC with aspartate aminotransferase [29], bacterial and human glutamate decarboxylase [30,31] and histidine decarboxylase [25] revealed which residues are required to assure fundamental catalytic activities and which ones are specific for each enzymatic species. The knowledge of this molecular basis is fundamental in aimed pharmacological treatment oriented to a specific enzyme. The information obtained reveals that: (i) the organization of the PLP binding cleft (Fig. 1) is mediated by some essential conserved residues: Lys-303 is the Schiff base lysine residue which covalently attaches the formyl group of PLP, Asp-271 makes a salt bridge interaction with the protonated pyridine nitrogen of PLP, His-192 is the pyridine stacking residue that sandwiches the aromatic ring of PLP, an extensive network of residues, namely Ser147, Ser-149, Asn-300, His-302 and Phe-309, contributes to the hydrogen binding interactions with PLP and to hold in place the phosphate group of the cofactor; (ii) the localization of the catechol ring of the ligand (carbiDopa) bound to DDC reveals that residues Ile-1010 and Phe-1030 (from the neighboring subunit), and Trp71, Tyr-79, Phe-80, Thr-82 are involved in substrate binding and cofactor stabilization, finally Thr-246, which faces the opposite face of the PLP ring, is suggested to play a role in the catalytic mechanism [32]; (iii) a stretch comprising residues 327–354 and 328–339 is missing in both human apo and pig holo structure, respectively, highlighting the presence of a flexible loop containing an important conserved residue, Tyr-332, possibly related to catalytic quinonoid protonation step (Scheme 1, see below) [26] and a protease sensitive consensus [33,34]. The relevance of most of the residues mentioned above arises also from an analysis of mutated variants leading to AADC deficiency (OMIM#608643), an inherited rare neuro-metabolic disease. Point mutations in residues interacting directly or indirectly with PLP or its microenvironment (G102S, S147R, S250F, A275T F309L) lead to modified PLP binding [21,35] as expected by 3D inspection. Moreover, many reports on site-directed mutagenesis studies [32,36] corroborate the proposal advanced on the role of the various PLP-binding residues. Interestingly, it can be observed that the organization of the active site is different in the apo compared to the holo form, the former being in an ‘‘open’’ conformation with respect to the ‘‘closed’’ arrangement of the latter [28]. When PLP is added, an apo to holo transition takes place, determining the rearrangement of a region containing residues 66–84 [28], important for substrate aromatic side chain location, and leading to a movement of the PLP-binding Lys-303 by 6 Å. The dimeric subunits approach one another by moving 20 Å and closing the active site [28]. This movement involves the whole structure making Trp-304, Phe-80 and Tyr-274 stabilized by aromatic stacking interactions among them. Notably, Tyr-274 in the holo form is hydrogen bonded to His-302 that, in turn, interacts with Tyr-79 (see above). This network is present in the holo form prone to substrate binding, while it is absent in the apo form [28]. The biological consequences of this remarkable open-closed conformational change could be multiple: on the one hand, the selective access of PLP to apoDDC could be driven directly by pyridoxal kinase since it was reported that its steric constraints are complementary to the cleft of the open apoDDC [28], on the other, the open enzyme conformation is easily degraded by proteases, suggesting that apoenzyme concentration levels

3

M. Bertoldi / Archives of Biochemistry and Biophysics 546 (2014) 1–7

Fig. 1. Active site of carbiDopa-complexed DDC [26]. The cofactor-ligand complex is ball-and-stick in CPK color, the two chains of DDC are green and yellow. Labels refer to position of a-carbon of each residue. The image was rendered with PyMol. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Lys-303

H C

NH+ L-aromatic

NH+

amino acid

C

R NH+

CO2

+

N H I

H

R CH

COO-

H

NH+ H+

+

+

N H

N H

II

N H

III

IV

R = CH2-catechol or indole moiety Scheme 1. Decarboxylation of L-aromatic amino acids catalyzed by DDC.

could be regulated by UPR basing on the availability of the PLP cofactor (Fig. 2). Reaction mechanisms It is known that PLP enzymes are able to catalyze, besides their main reaction, multiple side reactions occurring as ‘‘mistakes’’ due to the chemical reactivity of the cofactor. The structural requirements of the protein moiety are organized for directing the reaction through only one of the possible pathways [24] but have no

precise control on it, since PLP is highly reactive. It follows that catalyzing side reactions is a common character of PLP enzymes and these could sometimes be physiologically significant. DDC is not an exception, since in addition to the decarboxylation reaction, it is able to catalyze multiple side reactions. Decarboxylation of L-aromatic amino acids The main reaction catalyzed by DDC is decarboxylation of the aromatic amino acids L-Dopa and L-5-hydroxytryptophan. Its

4

M. Bertoldi / Archives of Biochemistry and Biophysics 546 (2014) 1–7 Table 2 Kinetic parameters of the reactions catalyzed by DDC: decarboxyation with L-5hydroxytryptophan and oxidative deamination with serotonin.

Decarboxylation-Pig Rat Human Oxidative deamination-Pig

kcat

Km

1.98 s1 [45,54] n.d. n.d. 1.26 min1 [41]

0.055 mM [54] n.d. n.d. 0.47 mM [41]

n.d., not determined.

Fig. 2. Proposed pathways of different conformations of DDC as a function of coenzyme and/or substrate binding. If apoenzyme is not complexed with cofactor it could be degraded by proteases.

chemistry is related to PLP reactivity (Scheme 1). The native form of the enzyme exists as an internal aldimine of PLP with Lys-303 (I). Upon substrate binding, the internal aldimine converts into an external aldimine (II) that, according to Dunathan hypopthesis [37] places the carboxy group in a position energetically favorable to decarboxylation. The subsequent quinonoid intermediate (III) is then protonated at the a-carbon to give the corresponding amine (IV) that leaves the active site and another catalytic cycle could start again. Kinetic analyses on this mechanism were undertaken with the pig [38], the rat [23,39] and the human enzyme [20]. The aim of these investigations was to identify the residue(s) responsible for catalysis and/or binding. A mechanistic paper was published in 1983 on the pig kidney enzyme [38] using L- and Damino acids as ligands useful to dissect the properties of the external aldimine (it should be remembered that according to Dunathan hypothesis [37] D-amino acids place an hydrogen atom perpendicular to the coenzyme-imine system, instead of a carboxylate as the L-isomers, thus the external aldimines with D-analogs are theoretically ‘‘inactive’’, that is unable to be further processed, and represent useful models for dissecting external Schiff base properties) and to measure catalytic parameters (Tables 1 and 2). Overall, this study proposes the existence of two forms of external aldimines, one absorbing at 420 nm and the other at 390 nm. These were attributed to the 40 N-protonated and 40 N-unprotonated form of the Schiff base between substrate and PLP [38]. A similar spectroscopic and kinetic analysis of the rat enzyme [23] allowed measurements of the rat enzyme kinetic parameters and proposed that the internal aldimine exists as a mixture of ketoenamine and enolimine tautomers. Later, in 1999, Hayashi et al. performed a detailed pH kinetic analysis of reaction of the rat enzyme with LDopa [39] and suggested that: (i) the 430 and 380 nm bands, shown in the spectra at various pH values, should be attributed to a Michaelis complex and to a 4N0 -protonated external aldimine, respectively, and (ii) the substrate L-Dopa preferentially binds unprotonated to the 40 N-protonated internal Schiff base. The kinetic parameters of rat DDC are reported in Table 1. A recent investigation [20] on spectroscopic and kinetic studies of the human enzyme as a function of pH led to the following results. Upon substrate binding a 420 nm band shifts to 390 nm at increasing pH values. This transition is characterized by a pKa value of 6.4 that is

similar to those observed in both kcat and kcat/Km profiles. This suggests that the ionization is related to a catalytic event. The two absorbing species could thus be attributed to the 40 N-protonated and 40 N-unprotonated Schiff bases, respectively [20]. Another pKa value obtained on the basic side of the kcat profile and on the Kd(PLP) profile could suggest the involvement of some cofactor group, for example the 50 -phosphate ester, as reported elsewhere [40]. This assignment, however, is tentative, since it cannot ruled out that a residue involved in catalysis is responsible for this pKa given that this is found also on the kcat profile. A good candidate was proposed to be Tyr-332. In fact, an useful information for identifying some catalytically essential residues in decarboxylation comes from a mutational study regarding Tyr-332 [27]. In that paper, it was proposed that this residue is responsible for reprotonation of the quinonoid intermediate generated upon CO2 release from the enzyme–substrate complex. Notably, the Y332F variant switches its reaction specificity from amine generation to aldehyde generation performing an oxidative deamination of L-Dopa (see below).

Oxidative deamination of aromatic amines In 1996 it was reported that pig DDC is able to catalyze oxidative deamination of serotonin [41], (Scheme 2). Later, it was demonstrated that this catalytic activity is also performed by the enzyme in the presence of dopamine and a-methylDopa as substrates [42]. Interestingly, the generated carbonyl compounds act as suicide or mechanism-based inhibitors of the enzyme [41,42]. The catalytic mechanism proposed is the following: at first, the internal aldimine converts to external aldimine with the amine (I) which is subsequently either deprotonated or decarboxylated to give a quinonoid intermediate (II). This is attacked by a molecule of dioxygen to form a hydroperoxy intermediate (III) leading, through still unknown steps, to dissociation of the hydroperoxy amino acid and nonenzymatic decomposition into peroxide and the imine, which hydrolyzes producing the carbonyl aldehydic/ketonic product and ammonia. The stoichiometry of dioxygen consumed with respect to carbonyl compound and ammonia formed as well as amine oxidized is 1:2 [41,42]. Recent work revealed the formation of a ketimine and superoxide as reaction intermediates [43]. The aromatic compounds produced possess similar biological activities as the aromatic amines and thus are strong biologically active signals. In recent years, studies with an analogue of serotonin undergoing oxidative deamination with DDC, i.e. Dtryptophan methyl ester [43,44], shows the accumulation of the quinonoid intermediate of this peculiar reaction.

Table 1 Kinetic parameters of the reactions catalyzed by DDC: decarboxyation with L-Dopa and oxidative deamination with dopamine/a-methylDopa.

Decarboxylation-Pig Rat Human Oxidative deamination-Pig

kcat

Km

5.8 s1 [54] 6.3 s1 [39] 5.1 s1 [20] 3.03 min1[42]/5.68 min1[42]

0.070 mM [54] 0.086 mM [23] 0.028 mM [20] 2.48 mM/0.045 mM [42]

5

M. Bertoldi / Archives of Biochemistry and Biophysics 546 (2014) 1–7

OOH Lys-303

H C

R

NH+ aromatic amine

NH+

NH+

NH+ H+

+

N H I

RC H

R CH

H

O2

+

+

N H

N H

II

N H

III

IV H2O OOH

R = CH2-catechol or indole moiety RCH

+ PLP

NH3+ H2O2 aromatic aldehyde and ammonia Scheme 2. Oxidative deamination reaction catalyzed by DDC.

Surprisingly, under anaerobic conditions DDC catalyzes the half-transamination of its substrates of oxidative deamination. The investigation on this topic led to the proposal of an active site model for DDC flexible to the various substrates accommodated, dioxygen included [45]. The Tyr-332 to Phe variant was found to catalyze oxidative deamination of L-Dopa, changing the reaction specificity of DDC from decarboxylation to oxidative deamination [27]. On this basis, this residue was appointed as performing the quinonoid protonation step during decarboxylation reaction (step 3 in Scheme 1). When Tyr-332 is absent, the quinonoid reacts with dioxygen leading to Dopa oxidation. A pH investigation of the oxidative deamination reaction revealed the presence of a pKa value of 7.8 on the kcat profile [46], the same value found in L-Dopa decarboxylation profile, thus reinforcing its attribution to the coenzyme phosphate ester group or to a residue essential for both activities. A pKa of about 6.4 was found for the conversion of the 420–384 nm bands arising after addition of serotonin to DDC at increasing pH values. The value is similar to that found in the same transition with L-Dopa (see above) and this corroborates the attribution of the pKa to ionization of the 40 N-protonated to 40 N-unprotonated external aldimine. The reactivity towards dioxygen is not confined to DDC since other decarboxylases like glutamate and ornithine decarboxylase are able to perform an oxidative deamination reaction of their amethyl substrates [36]. In plants, it has been reported the existence of a PLP-enzyme capable of decarboxylating phenylalanine [47,48] to phenylacetaldehyde via an oxidative deamination reaction. This aromatic aldehyde is then reduced to phenylethylalcohol; both compounds are volatile molecules with typical rose-like/flowery odor used in cosmetic and aroma industries [49]. In Drosophila a direct pathway from L-Dopa to phenylacetaldehyde has been recently reported and the enzyme responsible for this activity was characterized [14]. It shares high sequence identity with human DDC and is able to perform oxidative deamination of L-Dopa. The aldehydic product was suggested to form cross-linking bridges among free amino groups of proteins to form the flexible insect cuticle [14]. Thus, the reaction described in mammal DDC could be a more common activity among PLP-enzymes. Notably, the plant oxidative decarboxylases have the Tyr-332 (according to DDC numbering) to Phe substitution that prevents reprotonation after decarboxylation.

Half-transamination Finally, DDC is able to perform the half-transamination of D-aromatic amino acids [34] as well as of aromatic amines, the latter under anaerobic conditions. This reaction is well reviewed in [45] and no further progresses were made in recent years. Mechanistically (Scheme 3a), a substrate of half-transamination forms the external aldimine (I) which is then deprotonated to give the quinonoid intermediate (II). This is reprotonated on the 40 -carbon to give the corresponding ketimine which hydrolyzes into a-keto acid and the PLP cofactor converts into the PMP form that dissociates from the enzyme, leaving it inactive in the apo form. The measured rate constant of this reaction is 0.124 min1 and 0.019 [34] min1 for D-5-hydroxytryptophan and D-Dopa, respectively. The physiological role of this reaction is unknown. Concomitant with half-transamination, a Pictet–Spengler reaction occurs leading to a cyclic substrate–coenzyme adduct that irreversibly inactivates the enzyme [34]. This arises, possibly, by an inefficient control of PLP reactivity (Scheme 3b) or an unproductive closure of the active site. In many variants of residues belonging to the PLP-binding cleft, in fact, this cyclization activity is high, since the coenzyme microenvironment is altered and does not assure complete reaction control.

Inhibition studies The search for an inhibitor of DDC started a long time ago. DDC is implicated in Parkinson’s disease since dopamine level is low in the disease due to degeneration of neurons producing it. Thus, the common therapy adopted is based on L-Dopa administration and implies also an inhibitor of peripheral DDC that would otherwise deplete the given drug. This inhibitor should not cross the blood– brain barrier. CarbiDopa or benserazide [26] (and references therein) were the molecules associated to the treatment with L-Dopa. Their chemistry is based on a hydrazine group that forms a hydrazone derivative with PLP, thus blocking it and inactivating the enzyme. Thus, a greater amount of L-Dopa could reach the brain where it could be transformed to dopamine ameliorating disease symptoms. The structure of the complex DDC-carbiDopa was solved revealing that Thr-82 is implicated in 40 -hydroxyl catechol ring binding [26].

6

M. Bertoldi / Archives of Biochemistry and Biophysics 546 (2014) 1–7

COO-

a Lys-303

R

C

NH+ D-aromatic amino acid

R C COO-

H

NH+

NH+

NH+ H+

+

H

H+

+

N H I

RCH

+

N H

N H

II

N H

III

IV H2O

b

aromatic ketoacid and PMP

R = CH2-catechol or indole moiety

Scheme 3. (a) Half-transamination of D-aromatic amino acid and aromatic amines under anaerobic conditions catalyzed by DDC. (b) Pictet–Spengler cyclization reaction occurring at the active site of DDC.

Various other compounds were reported to inhibit DDC: (i) those acting via a suicide mechanism by alkylating the enzyme: a-chloromethyl and a-fluoromethyl derivatives of Dopa (reviewed in [50]), a-vinylDopa and a-acetylenic Dopa [50]; (ii) the phosphopyridoxyl aromatic amino acids Schiff base analogs [51] and (iii) substrate analogs, like green tea polyphenols [52]. Their Kd values are in the range of 102 to 106 M. Moreover, 3,4 dihydroxyphenylacetone, the product of oxidative deamination of a-methylDopa, behaves as a site directed affinity label [42], while serotonin (and/or its aldehyde) as a mechanism-based inhibitor [41]. The latter are the carbonyl compounds generated by the oxidative deamination catalyzed by DDC, thus suggesting a modulator role of this side reaction. The common negative feature of all these compounds is that they target the PLP cofactor and thus are not selective for DDC. Rather, they irreversibly bind to free PLP and other PLP-enzymes.

Recently, a new approach for efficient drug design was applied: a virtual screening of putative DDC inhibitors was performed. This analysis was supported by an in vitro analysis [53]. The most active compound was used as core to develop other chemical modifications to enhance inhibitory properties. With this combined investigation, an interesting reversible competitive inhibitor, Amb2470350, was found with Ki of 500 nM (Fig. 3). This could represent a new molecule for the development of a specific DDC inhibitor [53]. The compound was predicted not to cross the blood–brain barrier [53] and presents some interesting features that could be a starting point for hit-to-lead development of effective drugs. In fact, it does not contain a hydrazinic group, does not bind free PLP and is a reversible inhibitor of DDC [35]. Overall, DDC provides an excellent example of PLP high reactivity only partially controlled or directed by the protein moiety of the enzyme. However, its wide substrate and reaction specificity could represent an occasion for exerting a sort of regulation of substrates/products levels. Much work is still to be performed on this ‘‘pleiotropic’’ interesting protein.

O

Acknowledgment OH N

N

N

The research reviewed in this paper is partly supported by FUR2012, University of Verona.

OH

N SH Fig. 3. Structure of the most promising DDC-inhibitor, Amb2470350 [53]. Its chemistry bases on a core with a cathecol ring with a meta- and para-hydroxylic groups linked to 3-mercapto-1,2,4-triazole heterocycle through a rigid methylidene-amino moiety. The triazole ring is further substituted with a benzene ring decorated with a methoxy substituent in ortho position.

References [1] [2] [3] [4]

K.M. Crisp, K.A. Mesce, J. Exp. Biol. 207 (2004) 4535–4542. O. Kiehn, O. Kjaerulff, J. Neurophysiol. 75 (1996) 1472–1482. E. Marder, J.S. Eisen, J. Neurophysiol. 51 (1984) 1362–1374. J. Schotland, O. Shupliakov, M. Wikstrom, L. Brodin, M. Srinivasan, Z.B. You, M. Herrera-Marschitz, W. Zhang, T. Hokfelt, S. Grillner, Nature 374 (1995) 266– 268.

M. Bertoldi / Archives of Biochemistry and Biophysics 546 (2014) 1–7 [5] J. Reith, C. Benkelfat, A. Sherwin, Y. Yasuhara, H. Kuwabara, F. Andermann, S. Bachneff, P. Cumming, M. Diksic, S.E. Dyve, P. Etienne, A.C. Evans, S. Lal, M. Shevell, G. Savard, D.F. Wong, G. Chouinard, A. Gjedde, Proc. Natl. Acad. Sci. USA 91 (1994) 11651–11654. [6] A.D. Borglum, T.G. Bruun, T.E. Kjeldsen, H. Ewald, O. Mors, G. Kirov, C. Russ, B. Freeman, D.A. Collier, T.A. Kruse, Mol. Psychiatry 4 (1999) 545–551. [7] D.F. Shih, C.D. Hsiao, M.Y. Min, W.S. Lai, C.W. Yang, W.T. Lee, S.J. Lee, PLoS One 8 (2013) e71741. [8] R. Montioli, B. Cellini, C. Borri Voltattorni, J. Inherit. Metab. Dis. 34 (2011) 1213–1224. [9] M.B. Feany, W.W. Bender, Nature 404 (2000) 394–398. [10] E. Masliah, E. Rockenstein, I. Veinbergs, M. Mallory, M. Hashimoto, A. Takeda, Y. Sagara, A. Sisk, L. Mucke, Science 287 (2000) 1265–1269. [11] M.J. Aminoff, West J. Med. 161 (1994) 303–308. [12] R.A. Hauser, Eur. Neurol. 62 (2009) 1–8. [13] R.B. Hodgetts, S.L. O’Keefe, Annu. Rev. Entomol. 51 (2006) 259–284. [14] C. Vavricka, Q. Han, Y. Huang, S.M. Erickson, K. Harich, B.M. Christensen, J. Li, PLoS One 6 (2011) e16124. [15] P.J. Facchini, K.L. Huber-Allanach, L.W. Tari, Phytochemistry 54 (2000) 121– 138. [16] C.B. Voltattorni, A. Minelli, C. Turano, FEBS Lett. 17 (1971) 231–235. [17] P. Dominici, B. Tancini, D. Barra, C.B. Voltattorni, Eur. J. Biochem. 169 (1987) 209–213. [18] M. Ando-Yamamoto, H. Hayashi, T. Sugiyama, H. Fukui, T. Watanabe, H. Wada, J. Biochem. 101 (1987) 405–414. [19] C.S. Hoefig, K. Renko, S. Piehl, T.S. Scanlan, M. Bertoldi, T. Opladen, G.F. Hoffmann, J. Klein, O. Blankenstein, U. Schweizer, J. Kohrle, Mol. Cell. Endocrinol. 349 (2012) 195–201. [20] R. Montioli, B. Cellini, M. Dindo, E. Oppici, C.B. Voltattorni, Biomed. Res. Int. 2013 (2013) 161456. [21] R. Montioli, E. Oppici, B. Cellini, A. Roncador, M. Dindo, C.B. Voltattorni, Hum. Mol. Genet. 22 (2013) 1615–1624. [22] P.S. Moore, P. Dominici, C. Borri, Biochem. J. 315 (Pt. 1) (1996) 249–256. [23] H. Hayashi, H. Mizuguchi, H. Kagamiyama, Biochemistry 32 (1993) 812–818. [24] R.A. John, Biochim. Biophys. Acta 1248 (1995) 81–96. [25] H. Komori, Y. Nitta, H. Ueno, Y. Higuchi, J. Biol. Chem. 287 (2012) 29175– 29183. [26] P. Burkhard, P. Dominici, C. Borri-Voltattorni, J.N. Jansonius, V.N. Malashkevich, Nat. Struct. Biol. 8 (2001) 963–967. [27] M. Bertoldi, M. Gonsalvi, R. Contestabile, C.B. Voltattorni, J. Biol. Chem. 277 (2002) 36357–36362. [28] G. Giardina, R. Montioli, S. Gianni, B. Cellini, A. Paiardini, C.B. Voltattorni, F. Cutruzzola, Proc. Natl. Acad. Sci. USA 108 (2011) 20514–20519. [29] C. Momany, R. Ghosh, M.L. Hackert, Protein Sci. 4 (1995) 849–854.

7

[30] G. Capitani, D. De Biase, C. Aurizi, H. Gut, F. Bossa, M.G. Grutter, EMBO J. 22 (2003) 4027–4037. [31] G. Fenalti, R.H. Law, A.M. Buckle, C. Langendorf, K. Tuck, C.J. Rosado, N.G. Faux, K. Mahmood, C.S. Hampe, J.P. Banga, M. Wilce, J. Schmidberger, J. Rossjohn, O. El-Kabbani, R.N. Pike, A.I. Smith, I.R. Mackay, M.J. Rowley, J.C. Whisstock, Nat. Struct. Mol. Biol. 14 (2007) 280–286. [32] M. Bertoldi, C.B. Voltattorni, Arch. Biochem. Biophys. 488 (2009) 130–139. [33] B. Tancini, P. Dominici, M. Simmaco, M.E. Schinina, D. Barra, C.B. Voltattorni, Arch. Biochem. Biophys. 260 (1988) 569–576. [34] M. Bertoldi, P. Frigeri, M. Paci, C.B. Voltattorni, J. Biol. Chem. 274 (1999) 5514– 5521. [35] B. Cellini, R. Montioli, E. Oppici, C.B. Voltattorni, Open Biochem. J. 6 (2012) 131–138. [36] M. Bertoldi, S. Castellani, C. Bori Voltattorni, Eur. J. Biochem. 268 (2001) 2975– 2981. [37] H.C. Dunathan, Proc. Natl. Acad. Sci. USA 55 (1966) 712–716. [38] C.B. Voltattorni, A. Minelli, P. Dominici, Biochemistry 22 (1983) 2249–2254. [39] H. Hayashi, F. Tsukiyama, S. Ishii, H. Mizuguchi, H. Kagamiyama, Biochemistry 38 (1999) 15615–15622. [40] B. Cellini, M. Bertoldi, R. Montioli, C. Borri Voltattorni, Biochemistry 44 (2005) 13970–13980. [41] M. Bertoldi, P.S. Moore, B. Maras, P. Dominici, C.B. Voltattorni, J. Biol. Chem. 271 (1996) 23954–23959. [42] M. Bertoldi, P. Dominici, P.S. Moore, B. Maras, C.B. Voltattorni, Biochemistry 37 (1998) 6552–6561. [43] M. Bertoldi, B. Cellini, R. Montioli, C. Borri Voltattorni, Biochemistry 47 (2008) 7187–7195. [44] M. Bertoldi, B. Cellini, B. Maras, C.B. Voltattorni, FEBS Lett. 579 (2005) 5175–5180. [45] M. Bertoldi, C. Borri Voltattorni, Biochim. Biophys. Acta 1647 (2003) 42–47. [46] M. Bertoldi, C.B. Voltattorni, Protein Sci. 10 (2001) 1178–1186. [47] X.M. Chen, H. Kobayashi, M. Sakai, H. Hirata, T. Asai, T. Ohnishi, S. Baldermann, N. Watanabe, J. Plant Physiol. 168 (2011) 88–95. [48] Y. Kaminaga, J. Schnepp, G. Peel, C.M. Kish, G. Ben-Nissan, D. Weiss, I. Orlova, O. Lavie, D. Rhodes, K. Wood, D.M. Porterfield, A.J. Cooper, J.V. Schloss, E. Pichersky, A. Vainstein, N. Dudareva, J. Biol. Chem. 281 (2006) 23357–23366. [49] E. Pichersky, J.P. Noel, N. Dudareva, Science 311 (2006) 808–811. [50] C. Borri Voltattorni, M. Bertoldi, Dopa decarboxylase, in: T. Creighton (Ed.), Encyclopedia of Molecular Medicine, John Wiley and Sons, New York, 2001. [51] A. Orlacchio, C. Borri-Voltattorni, C. Turano, Biochem. J. 185 (1980) 41–46. [52] M. Bertoldi, M. Gonsalvi, C.B. Voltattorni, Biochem. Biophys. Res. Commun. 284 (2001) 90–93. [53] F. Daidone, R. Montioli, A. Paiardini, B. Cellini, A. Macchiarulo, G. Giardina, F. Bossa, C. Borri, PLoS One 7 (2012) e31610. [54] M. Bertoldi, C. Borri, Biochem. J. 352 (Pt 2) (2000) 533–538.