Letter

Cite This: ACS Med. Chem. Lett. 2019, 10, 431−436 pubs.acs.org/acsmedchemlett

L‑DOPA-quinone Mediated Recovery from GIRK Channel Firing

Inhibition in Dopaminergic Neurons
Bruno M. Bizzarri,† Lorenzo Botta,*,† Daniela Aversa,‡ Nicola B. Mercuri,‡,§ Giulio Poli,∥
Alessandro Barbieri,∥ Nicola Berretta,*,‡ and Raffaele Saladino*,†
†
Dipartimento di Scienze Ecologiche e Biologiche, Università della Tuscia, Via S. C. De Lellis 44, 01100 Viterbo, Italy
‡
IRCCS Fondazione Santa Lucia, Via Ardeatina, 306/354, 00142 Rome, Italy
§
Dipartimento di Medicina dei Sistemi, Università di Roma Tor Vergata, Via Montpellier, 1, 00133 Rome, Italy
∥
Dipartimento di Biotecnologie, Chimica e Farmacia, Università degli studi di Siena, Via A. Moro 2, 53100 Siena, Italy
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

*
S Supporting Information
Downloaded via UNIV DE SANTIAGO DE CHILE on October 19, 2020 at 14:44:04 (UTC).

ABSTRACT: The oxidative degeneration of dopamine-releas-

ing (DAergic) neurons in the substantia nigra pars compacta
(SNc) has attracted much interest in preclinical research, due
to its involvement in Parkinson’s disease manifestations.
Evidence exists on the participation of quinone derivatives in
mitochondrial dysfunction, alpha synuclein protein aggregation,
and protein degradation. With the aim to investigate the role of
L-DOPA-quinone in DAergic neuron functions, we synthesized
L-DOPA-quinone by use of 2-iodoxybenzoic acid and
measured its activity in recovery from dopamine-mediated
firing inhibition of SNc neurons. Noteworthy, L-DOPA-
quinone counteracts firing inhibition in SNc DAergic neurons
caused by GIRK opening. A possible mechanism to explain the
effect of L-DOPA-quinone on GIRK channel has been
proposed by computational models. Overall, the study showed the possibility that L-DOPA-quinone stabilizes GIRK in a
preopen conformation through formation of a covalent adduct with cysteine-65 on the GIRK2 subunit of the protein.
KEYWORDS: L-DOPA-quinone synthesis, biological activity, dopaminergic neurons firing, Parkinson, GIRK channel

T he ventral midbrain is the area where most of the dopamine

(DA) released in the central nervous system originates,
being synthesized and released by a local population of
recognized mechanism leading to SNc DAergic neurons
degeneration in PD. Indeed, it is DA itself, locally released in
the SNc area, that has been indicated as a main actor of this
dopamine-releasing (DAergic) neurons arranged into two oxidative process, particularly through its oxidation into
main subregions, the substantia nigra pars compacta (SNc) quinones.7,8 Evidence exists on the participation of quinones
and the ventral tegmental area. It is known that the release of DA in mitochondrial dysfunction, alpha synuclein protein aggrega-
from these neuronal populations is not limited to the relative tion, and protein degradation;8 however, little is known of the
areas of projection, but occurs also locally, through a somato- role of these reactive molecules in regulating SNc DAergic
dendritic DA release taking place within the ventral midbrain. neurons.
The functional role of this locally released DA is well established, L-DOPA (L-3,4-dihydroxyphenylalanine) still remains the
as it acts on somato-dendritic D2 receptors (D2Rs) located onto best and most used DA pro-drug in treatment of PD.9 After
the DAergic neurons, also called D2 autoreceptors, causing the administration, L-DOPA interacts with specific carriers and
opening of an inwardly rectifying potassium conductance crosses the blood−brain barrier (BBB), after which it is
through interposition of a G protein (GIRK channel). Thus, converted to DA within the brain.10 Data are available on the
the overall effect of DA in the ventral midbrain is hyper- relationships between the metabolic transformation of DOPA to
polarization of DAergic neurons’ membrane potential, resulting corresponding DOPA-quinone and the emergence of side-
in inhibition of their firing rate and alteration in DA effects during the treatment of PD.11 Moreover, we demon-
transmission.1−6
SNc DAergic neurons have attracted much interest in
Special Issue: Highlighting Medicinal Chemistry in Italy
preclinical research, due to their involvement in motor control,
such that their degeneration is the main hallmark of Parkinson’s Received: October 11, 2018
disease (PD) motor symptoms. Although many pathogenic Accepted: January 9, 2019
factors have been suggested, oxidative stress is a widely Published: January 9, 2019

© 2019 American Chemical Society 431 DOI: 10.1021/acsmedchemlett.8b00477

ACS Med. Chem. Lett. 2019, 10, 431−436
ACS Medicinal Chemistry Letters Letter

strated that unnatural DOPA-containing peptides obtained from recorded a total of 44 active neurons (5 slices) with a regular
DOPA-quinone intermediates12,13 were characterized by anti- basal firing rate (1.71 ± 0.12 Hz), all of which responded with
PD activity on individual DAergic neurons in rat ex vivo inhibition of their firing rate, as expected from typical DAergic
midbrain slices.14 neurons following stimulation of their somato-dendritic
With the aim to investigate the role of quinone intermediates D2Rs.1−3,5,6,19,20 The firing remained inhibited under continu-
in DAergic neurons function, we focused our attention on the ous perfusion in quinpirole, such that the firing rate of the
synthesis of L-DOPA-quinone and on its possible interactions DAergic neurons was inhibited to 4.07 ± 3.56% of control after
with the GIRK channels activated in response to D 2 10 to 15 min in 100 nM quinpirole. This perfusion time was
autoreceptor stimulation. L-DOPA-quinone was synthesized selected to make sure that a clear steady-state level had been
by selective oxidation of protected L-tyrosine, by using 2- reached.21 However, when DOPA-quinone 3 (300 μM) was
iodoxybenzoic acid (IBX). Molecular dynamics (MD) simu- added to the medium, still in the continuous presence of
lations have been performed to rationalize the biological data, quinpirole, a partial recovery from firing inhibition was
suggesting an interaction mechanism between L-DOPA- observed, attaining 81.85 ± 19.08% of control after 15 min, a
quinone and a specific cysteine residue of GIRK2 that value that was significantly different from that in quipirole alone
encompasses the transmembrane protein.

■
(P < 0.01; F(2,12) = 15.99, n = 5; Figure 1). The possible
desensitization of D2Rs by DOPA-quinone was ruled-out by
RESULTS AND DISCUSSION looking at its effect on the GIRK-mediated response by GABAB
L-DOPA-quinone has never been isolated in the past as a receptor stimulation.
consequence of its high reactivity to yield DOPAchrome,
followed by polymerization to melanin.15 The use of L-DOPA-
quinone that we made as intermediate for the preparation of
unnatural L-DOPA-containing peptides14 suggested its in-
creased stability when the α-amino functionality was protected.
For this reason, N-Boc-L-Tyr tert-butyl ester 1 was subjected to
IBX oxidation avoiding reducing conditions (Scheme 1). IBX

Scheme 1. Synthesis of L-DOPA-quinone

promotes the ortho-hydroxylation of phenol to catechol (ortho-

diphenol) in the presence of a reducing medium. The reactivity
of IBX is similar to that of polyphenol oxidases, passing through Figure 1. DOPA-quinone 3 counteracts firing inhibition of SNc
a relatively stable quinone intermediate. The high regioselectiv- DAergic neurons due to GIRK opening in response to D2R stimulation.
ity of the oxidation is a consequence of the initial intramolecular The raster plots are single units of four sample neurons recorded from
the same slice, during drugs exposure indicated on top. The figure
oxygen transfer from iodine(V) in a λ5-iodanyl intermediate to
below shows the firing frequency normalized to that measured before
the ortho-position of the phenolic group.16 The oxidation of 1 drug perfusion. D 2 R stimulation with quinpirole (100 nM)
was performed according to a slightly modified general progressively reduced the firing rate, as indicated by the partial
procedure previously applied.13,17 Briefly, compound 1 (0.068 inhibition at 2 min (white column) and attaining a maximal effect at
mmol) was treated with IBX (1.1 equiv) in propan-2-ol (12 mL) later stage (black column). When DOPA-quinone 3 (300 μM) was
at 25 °C for 2 h. Hindered propan-2-ol was selected with the aim added, in the continuous presence of quinpirole, a significant firing rate
to inhibit solvolysis processes. Under these experimental recovery was observed (gray column; P < 0.01).
conditions N-Boc-L-DOPA-quinone tertbutyl ester 2 was
obtained in quantitative conversion of substrate and 65% yield It is known that, similarly to D2Rs, also GABAB receptor
of product. The low mass balance was most probably due to stimulation leads to inhibition of SNc DAergic neurons firing
formation of overoxidation and polymerization side-products. rate through GIRK conductance opening.3,22 We thus took
Compound 2 (0.01 mmol) was then deprotected by mild advance of this converging mechanism and tested whether
treatment with TFA (2.2 equiv) in CH2Cl2 (1 mL) at 25 °C for 2 DOPA-quinone could inhibit a GIRK channel-mediated firing
h, to yield L-DOPA-quinone 3 as trifluoroacetate salt in 45% inhibition, triggered by this different receptor pathway,
yield (Scheme 1).18 independently from D2Rs (Figure 2).
The spontaneous firing of Substantia nigra pars compacta We recorded with the MEA device the firing rate of 31
(SNc) neurons in slices was detected as extracellular single unit neurons (in 4 slices), with a regular basal firing rate (1.89 ± 0.25
spikes with a MEA device, as previously reported.19 With the aim Hz), all of which were previously identified as DAergic by the
to trigger a GIRK channel-mediated functional response, we transient inhibition of their firing rate in response to a brief (1−2
exposed the slice to the D2R agonist quinpirole (100 nM). We min) challenge with 100 μM DA. The slices were then treated
432 DOI: 10.1021/acsmedchemlett.8b00477
ACS Med. Chem. Lett. 2019, 10, 431−436
ACS Medicinal Chemistry Letters Letter

The mechanism by which DOPA-quinone could promote the

closure, and alternatively inhibit the opening of the GIRK
channel, was evaluated through a computational approach. We
focused our attention on GIRK2 channel, which is the most
prevalent GIRK isoform in both substantia nigra and ventral
tegmental area.24 DOPA-quinone is an electrophilic molecule
able to react with sulfhydryl nucleophilic moieties. As an
example, the formation of the 5-cysteinylDOPA (5-cDOPA)
adduct is a key step in the synthesis of pheomelanin.25 Covalent
adducts, such as 5-cDOPA, can produce irreversible alterations
of enzymes or cellular proteins activity due to the alkylation of
residues in the catalytic site or in key portions of functional
domains. Particularly, catecholamine-quinones were suggested
to be responsible for the reduction of DA uptake by alkylating
specific cysteine residues of the dopamine transporter.23 They
are also able to block the catalytic activity of brain tyrosine
hydroxylase,26 showing neurotoxic effects.27 On the basis of this
Figure 2. DOPA-quinone 3 counteracts firing inhibition of SNc data, we developed a model for the formation of covalent
DAergic neurons due to GIRK opening in response to GABAB receptor DOPA-quinone adducts with cysteine residues of GIRK2,
stimulation. The raster plots are single units of four sample neurons evaluating their effect on the concerted opening/closing
recorded from the same slice, during drugs exposure indicated on top.
The figure below shows the firing frequency normalized to the firing
mechanism of the GIRK2 channel.28
rate measured before drug perfusion. The cells were first challenged The alkylation by sulfhydryl-modifying reagents of N-
with a brief (1−2 min) exposure in DA (100 μM), in order to confirm terminal cysteine 65 (C65) and C-terminal cysteine 321
that they were DAergic. Indeed, their firing rate was inhibited (vertically (C321) residues, both placed within the cytoplasmic domain
striped column). The slice was then perfused with the D2R antagonist of GIRK2 channel, determine the inhibition of GIRK mediated
sulpiride (10 μM), and a new control level of firing rate was established current in GIRK1/GIRK2 heteromeric channels.29 In particular,
(horizontally striped column). In the continuous presence of sulpiride, the alkylation of C65 was shown to have a high impact in current
GABAB receptor stimulation with baclofen (300 nM) progressively modulation, and it was found to be considerably more accessible
reduced the firing rate, as indicated by the partial inhibition at 2 min than C321 to sulfhydryl modifying reagents, as is also
(white column) and attaining a maximal effect at later stage (black
demonstrated by the X-ray crystallographic structure of murine
column). However, when DOPA-quinone 3 (300 μM) was added, in
the continuous presence of baclofen and sulpiride, a significant firing GIRK2 channel (98.3% identity with human isoform) in the
rate recovery was observed (gray column; P < 0.05). preopen conformation, once complexed with β−γ G-protein
subunits, dioctanoyl-l-alpha-phosphatidyl-D-myo-inositol-4,5-
diphosphate (PIO), and Na+ (PDB code 4KFM). For this
reason, we assumed C65 as the specific cysteine residue to which
with the specific D2R antagonist sulpiride (10 μM), kept in the DOPA-quinone would bind. C65 is placed in the proximity of
medium thereafter, in order to completely exclude any D2R specific loops of the protein, i.e., G-loop, CD-loop, and C-linker,
contribution. Following perfusion with the GABAB receptor which have shown an essential role in the concerted opening/
agonist baclofen (300 nM), DAergic neuron iring was inhibited closing mechanism of the channel, together with the N-terminal
and remained inhibited after more than 10 min of continuous loop to which C65 belongs (Figure 3). In fact, mutations located
baclofen perfusion to 24.94 ± 14.46% of control. However, in these regions can significantly affect the function of GIRK
when we added DOPA-quinone 3 (300 μM), in the continuous channels.30 In particular, the CD-loop is implied in the
presence of baclofen and sulpiride, a recovery of the firing was regulation of GIRK channel gating through conformational
observed, reaching 73.64 ± 14.42% of control. The level of firing movements involving a network of interactions with the C-linker
inhibition after 15 min in 300 μM DOPA-quinone 3 and and the G-loop, this latter constituting the second ion gate of the
baclofen was significantly lower than that in baclofen alone (P < channel.30−32 The CD-loop also constitutes part of the binding
0.05; F(4,15) = 14.27, n = 4). site for cellular Na+ (Figure 3), which demonstrated to amplify
These data indicated a specific role of DOPA-quinone 3 in the the opening of the channel in the presence of G-proteins.28
recovery from firing inhibition caused by GIRK channel opening The hypothesis that DOPA-quinone can disrupt the opening
in response to D2Rs or GABAB receptors stimulation. Recently, process of GIRK2 by alkylating C65 is also supported by an X-
our research team has reported that a similar form of firing ray structure of GIRK2 cytoplasmic domain complexed with
recovery may be obtained in response to DA, through an cadmium (Cd2+) ions, which suggests a mechanism for the
opposing depolarizing current generated by activation of the concentration-dependent inhibitory effect of Cd2+ on GIRK2
dopamine transporter (DAT).21 According to literature, the activity (PDB code 3AUW).33 In the crystal structure, the Cd2+
possibility that the same mechanism may also underlie our ion is located in a high-affinity site, placed at the entrance of Na+
present results is very unlikely. Indeed, for it to occur, DOPA- binding site, where it directly interacts with C65 and with
quinone should activate and be uptaken by DAT, while products residues of both CD-loop and N-terminus in octahedral
of DA oxidation-like quinones have been shown to inhibit rather coordination geometry. Through this network of interactions,
than stimulate DAT.23 Although future investigation may Cd2+ would block GIRK2 opening by trapping the channel in its
unravel presently unexpected effects on DAT, the most closed conformation as a sort of inverse agonist, tethering
conceivable hypothesis is that DOPA-quinone reverts firing between N-terminus and CD loop.33
inhibition by acting directly onto previously opened GIRK- In order to calculate a reliable model of the 5-cDOPA adduct
channels. with C65 (PDB code 4KFM; Figure 3), we employed the
433 DOI: 10.1021/acsmedchemlett.8b00477
ACS Med. Chem. Lett. 2019, 10, 431−436
ACS Medicinal Chemistry Letters Letter

Figure 4. Predicted binding mode of 5-cDOPA outside Na+ binding

site.

Figure 3. Crystal structure of GIRK2 channel (PDB code 4KFM). The its opening, revealing a concerted mechanism of gating in which
protein loops implied in channel gating, i.e., CD-loop, G-loop, and N- all protein monomers cooperatively work to allow the necessary
terminus, are shown in green, blue, and orange, respectively, together
global conformational change.28 Thus, C65 was covalently
with the sodium ion. Residues D228, H233, and E315, involved in
channel gating, are displayed together with C65. modified in all channel monomers prior to MD simulations.
The MD simulations confirmed the presence of the
interactions among 5-cDOPA and the two lysine residues K64
covalent docking protocol of AutoDock4.2,34 using a modified and K90, and revealed that the amino group of 5-cDOPA was
set of parameters (see Materials and Methods for details).35,36 able to form a strong salt bridge with the carboxylate group of
Since C65 is placed in front of the binding site of Na+ ions, we E315 (Figure 4), maintained for more than 85% of the
hypothesized that the catechol moiety of 5-cDOPA could even simulation, on average. The interaction with E315, which
occupy such pocket, thus interfering with the agonist-like effect belongs to the G-loop and is crucially involved in the channel
of Na+ in the modulation of GIRK2 gating. In fact, the gating mechanism, might have a key role in determining the
amplification of GIRK2-mediated current is concentration- effect of 5-cDOPA on GIRK-mediated current. In fact,
dependent since Na+ constitutes a mechanism of negative mutations of this particular residue, which is extremely
feedback on excessive electrical excitability, which occurs when conserved among all GIRK channels, can substantially alter
cellular Na+ concentration rise enough above normal levels.37 the protein functionality and/or abolish the channel-induced
Therefore Na+ is not supposed to be a constitutive element of current.29
GIRK2 channel, and we removed it from the protein structure Moreover, in the MD simulation of native GIRK2 we
prior to docking calculations. Interestingly, the calculation observed an interaction between the carboxylate group of
suggested that 5-cDOPA was more likely to be placed outside E315 and the side chain of the residue H233 in CD-loop (Figure
Na+ binding site rather than within the pocket (Figure 4). This is 3). During the movements of the CD-loop associated to channel
probably due to the presence of the carboxylic group in the gating, H233 should undergo a profound conformational
ligand, which is able to form strong interactions with the lysine rearrangement, inverting its orientation and climbing over the
residues surrounding the binding site (K64 and K90). CD-loop itself (Figure S8).37 Interestingly, the formation of the
To assess the reliability of the docking solution and to evaluate salt bridge between 5-cDOPA and E315 determined the
if 5-cDOPA could effectively interfere with GIRK2 opening, disruption of the interaction between E315 and H233; this
molecular dynamics (MD) simulations were also performed. produced a modification of H233 orientation totally opposite
The MD protocol was set up using the X-ray structure of wild- with respect to the conformational rearrangement that should be
type GIRK2 channel in complex with Na+ ions and PIO associated to channel gating. In fact, the imidazole ring of H233
cofactors as a template. After embedding the protein in a lipid moved within Na+ binding site after few nanoseconds of
bilayer and solvating the system on the “extracellular” and simulation and remained trapped between CD-loop and V276,
“intracellular” side (see Materials and Methods for details), 50 locked inside the pocket by the aromatic ring of 5-cDOPA
ns of MD simulations were performed. As shown in Figure S7, (Figure 4). Thus, the reciprocal disposition of 5-cDOPA and
the total energy of the native GIRK2 rapidly (after 1 ns) reached H233 would prevent this latter residue to move outside Na+
the equilibrium. Moreover, after about 2.5 ns, the root-mean binding site, blocking the conformational freedom of the whole
squared deviation of the protein α carbons with respect to the CD-loop and impeding the rearrangement required for channel
crystallographic structure showed an overall constant value opening.
around 2.0 Å. The same MD protocol was applied to the C65- In addition, 5-cDOPA forms hydrogen bonds with CD-loop
covalently modified protein, and the interactions observed residues like N231 or H233 itself that reinforce the conforma-
during the simulation between 5-cDOPA and the surrounding tional restraint of the CD-loop determined by the presence of
protein residues were analyzed. According to Wang and co- the covalent ligand. Through this network of interactions, 5-
workers, four Gβγ subunits must bind GIRK2 channel to allow cDOPA is anchored to and interconnects with different N-
434 DOI: 10.1021/acsmedchemlett.8b00477
ACS Med. Chem. Lett. 2019, 10, 431−436
ACS Medicinal Chemistry Letters Letter

terminus and CD-loop residues. This predicted tethering, which

showed to be detrimental for channel gating when induced by
Cd2+ ions,33 could stabilize the pre-open conformation of the
channel. The tethering could thus block the flickering process,
i.e., the rapid transition between open (conductive) and pre-
open (nonconductive) conformations, and channel gating.30 In
fact, during the opening process, CD-loop and N-terminus
undergo a large rearrangement, as shown in the crystal structure
of GIRK2 cytoplasmic domain (CTD) in a partially open state
(PDB code 3SYQ).37 This double effect might constitute the
mechanism through which DOPA-quinone, after the formation
of the 5-cDOPA adduct, stabilizes GIRK2 preopen conforma-
tion thus inhibiting the channel-mediated current.
The same covalent docking protocol applied on C65 was also
applied on C321 as a control. By just looking at the localization
of the covalent adduct (Figure S6), it is possible to understand
that C321 would be hard to reach by DOPA-quinone. In fact,
C321 is placed in the part of the protein that stays between the
outer surface (where C65 is located) and the inner surface (the
channel pore). Moreover, polar and very flexible residues like
K199, K200, K64, R324, Q322, and E315 narrow the access to Figure 5. Predicted binding pose for 5-cDA within Na+ binding site of
GIRK2. H-bonds among 5-cDA and the surrounding protein residue
C321 and limit the space around this residue. In agreement with
are shown as dashed lines.
these considerations, the covalent docking protocol generated
only a single possible solution associated to an unfavorable
ligand−protein binding affinity, probably due to both the steric
hindrance and the electrostatic repulsions between 5c-DOPA gating. Overall, the computational studies suggested that S-
and the surrounding residues. These results suggest that C321 is cysteinylcathecol derived from DOPA-quinone and DA-
not likely to be alkylated by DOPA-quinone. However, MD quinone could inhibit GIRK2 channel opening by promoting
simulation studies performed as control suggested that a the tethering of G-loop, CD-loop, and N-terminus. Additionally,
concurrent alkylation of both C65 and C321 could have a DOPA-quinone can also inhibit the GIRK2 gating machinery by
strong impact on GIRK-mediated current and thus on the firing interfering with the CD-loop movements involving H233.
of dopaminergic neurons (see Supporting Information S9 for
details).
The computational model was then extended to dopamine
■
*
ASSOCIATED CONTENT
S Supporting Information

quinone (DA-quinone), a compound that cannot be isolated The Supporting Information is available free of charge on the
due to its very high reactivity. In particular, we evaluated ACS Publications website at DOI: 10.1021/acsmedchem-
whether the alkylation of C65 by DA-quinone to afford 5-S- lett.8b00477.
cysteinyldopamine (5-cDA) might produce an effect on GIRK2 Slice preparation for electrophysiology; multielectrode
channel gating similar to that previously predicted for 5-cDOPA. array recordings; drugs used for electrophysiology;
In this latter case, covalent docking suggested that 5-cDA would covalent docking; molecular dynamics simulations;
preferentially occupy the Na+ binding site, interacting with the covalent docking results for DOPA-quinone bound to
protein loops implied in the channel gating. MD simulations C321; analysis of the MD simulation; superimposed
highlighted that the amino group of 5-cDA is able to form structures of GIRK2; MD simulation results of GIRK2
interactions with the residues that can also coordinate the Na+ with C65 and C321 concurrently alkylated by DOPA-
when this ion is present in the binding site. Particularly, a stable quinone; general information for reagents and synthetic
interaction with the carboxylate group of D228 is observed, and procedures; general information for the HPLC system;
it is maintained for about 80% of the simulation, on average. The HPLC spectrum of L-DOPA-quinone 3 (PDF)

■
amino group of 5-cDA can also arrange stable hydrogen bonds
with the backbone carbonyls of D228 and other residues such as AUTHOR INFORMATION
R230, thus demonstrating to well fit the pocket constituting the Corresponding Authors
Na+ binding site to which it is strongly anchored (Figure 5). The *E-mail: [email protected].
orientation of 5-cDA results in stabilization by additional H- *E-mail: [email protected].
bonds between its hydroxy groups and residues N231 and V67. *E-mail: [email protected].
Particularly, the H-bond interaction with the backbone carbonyl
ORCID
of V67 was observed for more than 80% of the whole MD
simulation, on average. Bruno M. Bizzarri: 0000-0001-7085-5432
It is worth noting that the interaction between E315 and Raffaele Saladino: 0000-0002-4420-9063
H233 was maintained as in the reference simulation of wild-type Notes
The authors declare no competing financial interest.

■
GIRK2 (Figure 3) and that no conformational rearrangement of
H233 is observed. However, 5-cDA is able to form a wide
network of interactions with CD-loop residues, and thus, a deep ACKNOWLEDGMENTS
tethering between N-terminus and CD-loop could stabilize the The Filas project MIGLIORA from Regione Lazio and the
channel in the pre-open conformation and prevent channel project PRONAT from CNCCS SCARL are acknowledged.
435 DOI: 10.1021/acsmedchemlett.8b00477
ACS Med. Chem. Lett. 2019, 10, 431−436
ACS Medicinal Chemistry Letters

■
Letter

REFERENCES G-protein-gated potassium channels. J. Neurosci. 2005, 25, 11468−

11478.
(1) Lacey, M. G.; Mercuri, N. B.; North, R. A. Dopamine acts on D2 (21) Aversa, D.; Martini, A.; Guatteo, E.; Pisani, A.; Mercuri, N. B.;
receptors to increase potassium conductance in neurons of the rat Berretta, N. Reversal of dopamine-mediated firing inhibition through
substantia nigra zona compacta. J. Physiol. 1987, 392, 397−416. activation of the dopamine transporter in substantia nigra pars
(2) Kim, K. M.; Nakajim, Y.; Nakajima, S. G protein-coupled inward compacta neurons. Br. J. Pharmacol. 2018, 175, 3534−3547.
rectifier modulated by dopamine agonists in cultured substantia nigra (22) Cruz, H. G.; Ivanova, T.; Lunn, M. L.; Stoffel, M.; Slesinger, P. A.;
neurons. Neuroscience 1995, 69, 1145−1158. Lüscher, C. Bi-directional effects of GABA(B) receptor agonists on the
(3) Beckstead, M. J.; Grandy, D. K.; Wickman, K.; Williams, J. T. mesolimbic dopamine system. Nat. Neurosci. 2004, 7, 153−159.
Vesicular dopamine release elicits an inhibitory postsynaptic current in (23) Whitehead, R. E.; Ferrer, J. V.; Javitch, J. A.; Justice, J. B. Reaction
midbrain dopamine neurons. Neuron 2004, 42, 939−946. of oxidized dopamine with endogenous cysteine residues in the human
(4) Ford, C. P. The role of D2-autoreceptors in regulating dopamine dopamine transporter. J. Neurochem. 2001, 76, 1242−1251.
neuron activity and transmission. Neuroscience 2014, 282, 13−22. (24) Nishizawa, D.; Gajya, N.; Ikeda, K. Identification of Selective
(5) Luscher, C.; Slesinger, P. A. Emerging roles for G protein-gated Agonists and Antagonists to G Protein-Activated Inwardly Rectifying
inwardly rectifying potassium (GIRK) channels in health and disease. Potassium Channels: Candidate Medicines for Drug Dependence and
Nat. Rev. Neurosci. 2010, 11, 301−315. Pain. Curr. Neuropharmacol. 2011, 9, 113−117.
(6) Krashia, P.; Martini, A.; Nobili, A.; Aversa, D.; D’Amelio, M.; (25) Slominski, A.; Tobin, D. J.; Shibahara, S.; Wortsman, J. Melanin
Berretta, N.; Guatteo, E.; Mercuri, N. B. On the properties of identified pigmentation in mammalian skin and its hormonal regulation. Physiol.
dopaminergic neurons in the mouse substantia nigra and ventral Rev. 2004, 84, 1155−228.
tegmental area. Eur. J. Neurosci. 2017, 45, 92−105. (26) Kuhn, D. M.; Arthur, R. E., Jr; Thomas, D. M.; Elferink, L. A.
(7) Lipski, J.; Nistico, R.; Berretta, N.; Guatteo, E.; Bernardi, G.; Tyrosine hydroxylase is inactivated by catechol-quinones and
Mercuri, N. B. L-DOPA: a scapegoat for accelerated neurodegeneration converted to a redox-cycling quinoprotein: possible relevance to
in Parkinson’s disease? Prog. Neurobiol. 2011, 94, 389−407. Parkinson’s disease. J. Neurochem. 1999, 73, 1309−1317.
(8) Segura-Aguilar, J.; Paris, I.; Muñoz, P.; Ferrari, E.; Zecca, L.; (27) Miyazaki, I.; Asanuma, M. Dopaminergic neuron-specific
Zucca, F. A. Protective and toxic roles of dopamine in Parkinson’s oxidative stress caused by dopamine itself. Acta Med. Okayama 2008,
disease. J. Neurochem. 2014, 129, 898−915. 62, 141−50.
(9) Di Stefano, A.; Sozio, P.; Cerasa, L. S.; Iannitelli, A. L-Dopa (28) Wang, W.; Touhara, K. K.; Weir, K.; Bean, B. P.; MacKinnon, R.
prodrugs: an overview of trends for improving Parkinson’s disease Cooperative regulation by G proteins and Na(+) of neuronal GIRK2
treatment. Curr. Pharm. Des. 2011, 17, 3482−93. K(+) channels. eLife 2016, 5, No. e15751.
(10) Bizzarri, B. M.; Rotelli, L.; Botta, G.; Saladino, R. Current (29) Guo, Y.; Waldron, G. J.; Murrell-Lagnado, R. A role for the
advances in DOPA and DOPA-peptidomimetics: chemistry, applica- middle C terminus of G-protein-activated inward rectifier potassium
tions and biological activity. Curr. Med. Chem. 2015, 22, 4138−4165. channels in regulating gating. J. Biol. Chem. 2002, 277, 48289−48294.
(11) Hälbig, T. D.; Koller, W. C. Levodopa, Parkinson’s Disease and (30) Whorton, M. R.; MacKinnon, R. X-ray structure of the
Related Disorders Part II. In Handbook of Clinical Neurology, 2nd ed.; mammalian GIRK2-βγ G-protein complex. Nature 2013, 498, 190−
Koller, W. C., Melamed, E., Eds.; Elsevier B.V.: Amsterdam, 2007; pp 197.
31−72. (31) Lü, S.; An, H.; Zhang, H.; Long, M. Structural Basis for
(12) Botta, G.; Delfino, M.; Guazzaroni, M.; Crestini, C.; Onofri, S.; Differences in Dynamics Induced by Leu Versus Ile Residues in the CD
Saladino, R. Selective Synthesis of DOPA and DOPA Peptides by Loop of Kir Channels. Mol. Neurobiol. 2016, 53, 5948−5961.
Native and Immobilized Tyrosinase in Organic Solvent. ChemPlu- (32) Meng, X. Y.; Zhang, H. X.; Logothetis, D. E.; Cui, M. The
sChem 2013, 78, 325−330. molecular mechanism by which PIP(2) opens the intracellular G-loop
(13) Bizzarri, B. M.; Pieri, C.; Botta, G.; Arabuli, L.; Mosesso, P.; gate of a Kir3.1 channel. Biophys. J. 2012, 102, 2049−59.
Cinelli, S.; Schinoppi, A.; Saladino, R. Synthesis and antioxidant activity (33) Inanobe, A.; Matsuura, T.; Nakagawa, A.; Kurachi, Y. Inverse
of DOPA peptidomimetics by a novel IBX mediated aromatic oxidative agonist-like action of cadmium on G-protein-gated inward-rectifier K(+)
functionalization. RSC Adv. 2015, 74, 60354−60364. channels. Biochem. Biophys. Res. Commun. 2011, 407, 366−71.
(14) Bizzarri, B. M.; Martini, A.; Serafini, F.; Aversa, D.; Piccinino, D.; (34) Bianco, G.; Forli, S.; Goodsell, D. S.; Olson, A. J. Covalent
Botta, L.; Berretta, N.; Guatteo, E.; Saladino, R. Tyrosinase mediated docking using autodock: Two-point attractor and flexible side chain
oxidative functionalization in the synthesis of DOPA-derived methods. Protein Sci. 2016, 25, 295−301.
peptidomimetics with anti-Parkinson activity. RSC Adv. 2017, 7, (35) Poli, G.; Gelain, A.; Porta, F.; Asai, A.; Martinelli, A.; Tuccinardi,
20502−20509. T. Identification of a new STAT3 dimerization inhibitor through a
(15) Mondal, S.; Thampi, A.; Puranik, M. Kinetics of Melanin pharmacophore-based virtual screening approach. J. Enzyme Inhib. Med.
Polymerization during Enzymatic and Nonenzymatic Oxidation. J. Chem. 2016, 31, 1011−1017.
Phys. Chem. B 2018, 122, 2047−2063. (36) Dal Piaz, F.; Vera Saltos, M. B.; Franceschelli, S.; Forte, G.;
(16) Magdziak, D.; Rodriguez, A. A.; Van De Water, R. W.; Pettus, T. Marzocco, S.; Tuccinardi, T.; Poli, G.; Ebrahimi, S. N.; Hamburger, M.;
R. R. Regioselective Oxidation of Phenols to o-Quinones with o- De Tommasi, N.; Braca, A. Drug Affinity Responsive Target Stability
Iodoxybenzoic Acid (IBX). Org. Lett. 2002, 4, 285−288. (DARTS) Identifies Laurifolioside as a New Clathrin Heavy Chain
(17) Bizzarri, B. M.; Botta, L.; Capecchi, E.; Celestino, I.; Checconi, Modulator. J. Nat. Prod. 2016, 79, 2681−2692.
P.; Palamara, A. T.; Nencioni, L.; Saladino, R. Regioselective IBX- (37) Whorton, M. R.; MacKinnon, R. Crystal structure of the
Mediated Synthesis of Coumarin Derivatives with Antioxydant and mammalian GIRK2 K+ channel and gating regulation by G proteins,
Anti-influenza Activities. J. Nat. Prod. 2017, 80, 3247−3254. PIP2, and sodium. Cell 2011, 147, 199−208.
(18) Suzuki, Y.; Takagi, N.; Sano, T.; Chimuro, T. Design and
synthesis of a novel fluorescent protein probe for easy and rapid
electrophoretic gel staining by using a commonly available UV-based
fluorescent imaging system. Electrophoresis 2013, 34, 2464−2472.
(19) Berretta, N.; Bernardi, G.; Mercuri, N. B. Firing properties and
functional connectivity of substantia nigra pars compacta neurones
recorded with a multi-electrode array in vitro. J. Physiol. 2010, 588,
1719−1735.
(20) Koyrakh, L.; Luján, R.; Colón, J.; Karschin, C.; Kurachi, Y.;
Karschin, A.; Wickman, K. Molecular and cellular diversity of neuronal

436 DOI: 10.1021/acsmedchemlett.8b00477

ACS Med. Chem. Lett. 2019, 10, 431−436