Conformational ensembles of non-peptide -conotoxin mimetics and Ca+2 ion binding to human voltage-gated N-type calcium channel Cav2.2

Abstract

Chronic neuropathic pain is the most complex and challenging clinical problem of a population that sets a major physical and economic burden at the global level. Ca2+-permeable channels functionally orchestrate the processing of pain signals. Among them, N-type voltage-gated calcium channels (VGCC) hold prominent contribution in the pain signal transduction and serve as prime targets for synaptic transmission block and attenuation of neuropathic pain. Here, we present detailed in silico analysis to comprehend the underlying conformational changes upon Ca2+ ion passage through Cav2.2 to differentially correlate subtle transitions induced via binding of a conopeptide-mimetic alkylphenyl ether-based analogue MVIIA. Interestingly, pronounced conformational changes were witnessed at the proximal carboxyl-terminus of Cav2.2 that attained an upright orientation upon Ca+2 ion permeability. Moreover, remarkable changes were observed in the architecture of channel tunnel. These findings illustrate that inhibitor binding to Cav2.2 may induce more narrowing in the pore size as compared to Ca2+ binding through modulating the hydrophilicity pattern at the selectivity region. A significant reduction in the tunnel volume at the selectivity filter and its enhancement at the activation gate of Ca+2-bound Cav2.2 suggests that ion binding modulates the outward splaying of pore-lining S6 helices to open the voltage gate. Overall, current study delineates dynamic conformational ensembles in terms of Ca+2 ion and MVIIA-associated structural implications in the Cav2.2 that may help in better therapeutic intervention to chronic and neuropathic pain management.

Full text

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)
This is an open access article which appeared in a journal published
by Elsevier. This article is free for everyone to access, download and
read.
Any restrictions on use, including any restrictions on further
reproduction and distribution, selling or licensing copies, or posting
to personal, institutional or third party websites are defined by the
user license specified on the article.
For more information regarding Elsevier's open access licenses
please visit:
http://www.elsevier.com/openaccesslicenses

Conformational ensembles of non-peptide
x
-conotoxin mimetics and
Ca
+2
ion binding to human voltage-gated N-type calcium channel Ca
v
2.2
Sameera
a
, Fawad Ali Shah
b
, Sajid Rashid
a,
⇑
a
National Center for Bioinformatics, Quaid-i-Azam University, Islamabad, Pakistan
b
Riphah Institute of Pharmaceutical Sciences, Riphah International University, Islamabad, Pakistan
article info
Article history:
Received 29 May 2020
Received in revised form 24 August 2020
Accepted 26 August 2020
Available online 3 September 2020
Keywords:
Chronic neuropathic pain
N-type voltage-gated calcium channel
(Ca
v
2.2)
Non-peptide
x
-conotoxin mimetics
inhibitor
Synaptic transmission
abstract
Chronic neuropathic pain is the most complex and challenging clinical problem of a population that sets a
major physical and economic burden at the global level. Ca
2+
-permeable channels functionally orches-
trate the processing of pain signals. Among them, N-type voltage-gated calcium channels (VGCC) hold
prominent contribution in the pain signal transduction and serve as prime targets for synaptic transmis-
sion block and attenuation of neuropathic pain. Here, we present detailed in silico analysis to comprehend
the underlying conformational changes upon Ca
2+
ion passage through Ca
v
2.2 to differentially correlate
subtle transitions induced via binding of a conopeptide-mimetic alkylphenyl ether-based analogue
MVIIA. Interestingly, pronounced conformational changes were witnessed at the proximal carboxyl-
terminus of Ca
v
2.2 that attained an upright orientation upon Ca
+2
ion permeability. Moreover, remarkable
changes were observed in the architecture of channel tunnel. These findings illustrate that inhibitor bind-
ing to Ca
v
2.2 may induce more narrowing in the pore size as compared to Ca
2+
binding through modu-
lating the hydrophilicity pattern at the selectivity region. A significant reduction in the tunnel volume
at the selectivity filter and its enhancement at the activation gate of Ca
+2
-bound Ca
v
2.2 suggests that
ion binding modulates the outward splaying of pore-lining S6 helices to open the voltage gate. Overall,
current study delineates dynamic conformational ensembles in terms of Ca
+2
ion and MVIIA-associated
structural implications in the Ca
v
2.2 that may help in better therapeutic intervention to chronic and neu-
ropathic pain management.
Ó2020 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and
Structural Biotechnology. This is an open access article under the CC BY-NC-ND license (http://creative-
commons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Chronic pain is a serious health issue that affects more than 25%
of the world’s population, and is increasing with the population
ages [1]. Approximately, 8% of general population is affected by
neuropathic pain [2] that is abnormal signaling due to injury or
malfunction in the peripheral or central nervous system resulting
in exaggerated pain sensations [3–6]. It originates from plastic
changes at peripheral, spinal, or supraspinal sites known as periph-
eral neuropathies (postherpetic neuralgia, toxic neuropathies and
focal traumatic neuropathies), central neuropathies (ischemic
cerebrovascular injury, spinal cord injury and Parkinson’s disease)
and mixed neuropathies (diabetic neuropathies, sympathetically
maintained pain) [4]. Neuropathic pain has challenged the biomed-
ical research to develop more effective drugs. Many drugs that
treat inflammatory pain, such as Nonsteroidal anti-inflammatory
drugs (NSAIDs) or opioids are in general much less effective in
relieving neuropathic pain and exhibit efficacy only at high doses
or when administered by more invasive (e.g., intraspinal) routes.
Currently, neuropathic pain is treated mostly by tricyclic antide-
pressants, specific serotonin and norepinephrine modulators, and
sodium and calcium channel modulators [7]. Currently, only 3
Food and Drug Administration (FDA) approved drugs, gabapentin,
pregabalin and ziconotide target voltage-gated calcium channels
(VGCCs); however, pregabalin and gabapentin have serious side
effects and low efficacy issues [8]. The
x
-conotoxin MVIIA or Zico-
notide (SNX-11; Prialt
Ò
) is a novel non-opioid analgesic drug that
is a synthetic version from a large class of marine cone snail pep-
tides [9–11]. This drug has been approved for the symptomatic
management of severe chronic pain. It is administered intrathe-
cally to attain optimal analgesic efficacy along with less serious
side-effects as it has limited ability to cross blood–brain barrier
[11]. This has prompted the development of dozens of small-
molecule blockers that are effective in the animal models [12–13].
https://doi.org/10.1016/j.csbj.2020.08.027
2001-0370/Ó2020 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
⇑
Corresponding author.
E-mail address: sajidrwp@yahoo.co.uk (S. Rashid).
Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
journal homepage: www.elsevier.com/locate/csbj
Author's Personal Copy

Animal studies demonstrate that due to peripheral tissue
inflammation or nerve injury, Ca
v
2.2 (N-type Ca
2+
channel subunit
alpha-1B) expression is enhanced in the dorsal horn [14–15].
Ca
v
2.2 deficient mice exhibit hyposensitivity to inflammatory
and neuropathic pain [16–18]. Hence, this channel is considered
as an important target for treating nociceptive and neuropathic
pain [19–22].Ca
v
2.2 is a transmembrane VGCC that is involved
in various cell signaling responses and membrane potential alter-
ations that induce calcium influx. As a result, intracellular calcium
level elevates to micro molar range [23] that stimulates various
calcium channel processes, i.e., gene transcription, membrane
excitability, neurotransmitter release, neurite outgrowth and acti-
vation of calcium-dependent enzymes such as calmodulin-
dependent protein kinase II (CaMKII) and protein kinase C (PKC)
[24]. The prolonged elevation of intracellular calcium levels is cyto-
toxic [25] as the intrinsic gating processes and cell signaling path-
ways tightly regulate the channel activity and trafficking to and
from membrane [26]. The blockage of synaptic transmission
through Ca
v
2.2 serves as a prime mechanism for the reduction of
pain signals to the central nervous system [27].Ca
v
2.2 is a complex
ion channel protein that is comprised of 4 or 5 different subunits
encoded by multiple genes. The largest and most important sub-
unit is
a
1 subunit of 262 kDa molecular weight that is encoded
by CACNA1B gene in human [28]. All Ca
v
a
1 subunits of VGCC pos-
sess a common transmembrane topology in all transmembrane
domains. Their important functions include pore conduction, volt-
age sensing, gating apparatus and channel regulation [29].Ca
v
a
1
subunits have four homologous domains (D
I
-D
IV
), each having six
transmembrane segments (S1-S6). S4 helix is voltage-sensing seg-
ment, while the loop between S5 and S6 segments in each domain
is involved in pore formation [29]. It involves in ion conductance
and selectivity. In addition to
a
1-subunit, VGCCs are composed
of other subunits that are b-subunits,
a
2d-subunits and
c
-
subunits. Among them, 4 b- and
a
2d-subunits, while 8
c
-
subunits are involved in enhancing
a
1-subunit cell surface expres-
sion and its interaction with diverse intracellular signaling mole-
cules that modulate the gating properties [30–32].
Animal venoms have high efficacy and selectivity against a wide
range of biological targets such as membrane proteins i.e., ion
channels, receptors and transporters [33–36]. To date, about
500–700 Conus species have been identified [37] containing valu-
able neuroactive peptides. Multiple attempts have been made to
develop conopeptide mimetics and small molecule inhibitors
against VGCCs. Among 21
x
-conotoxin peptides identified so far
[38–39], the most well characterized
x
-conotoxins are Ca
v
2.2
blockers: MVIIA, CVID and GVIA [40]. Here we characterized the
published conopeptide-mimetic inhibitors that are based on
diverse scaffolds i.e., dendritic scaffold [41], benzothiazole deriva-
tive [42], anthranilamide derivative [43], anthranilamide scaffold
[44] and anthranilamide compounds with diphenylmethylpiper-
azine moiety [40] to block Ca
v
2.2 (Table S1). Our in-silico analysis
provides significant information about the binding pattern and
its impact on the Ca
v
2.2 channel conformation and activity. By
exploring the association of Ca
v
2.2 and small-molecule
conopeptide-mimetic inhibitors, our study may provide invaluable
insights into the potential anesthetic and analgesic impact on pain
by multi-targeted inhibition.
2. Methodology
2.1. Molecular modeling of Cav2.2
The amino acid sequence of Homo sapiens
a
1-subunit of Ca
v
2.2
was retrieved through UniProtKB (www.uniprot.org) having an
accession number: Q00975. Due to lack of crystal structure, Oryc-
tolagus cuniculus voltage-gated calcium channel, Ca
v
1.1 (PDB ID:
5GJW; resolution: 3.6 Å; sequence coverage: 63% (70-1873AA);
sequence identity: 50%;) structure [45] was used as a template to
model
a
1-subunit of Homo sapiens Ca
v
2.2 through SWISS-MODEL
(swissmodel.expasy.org). The modeled structure of Human Ca
v
2.2
(contains 1771 residues (Phe76-Pro1846)) exhibits 0.45 global
quality estimation score (GMQE) and 44% sequence similarity.
The predicted structure was visualized by UCSF Chimera 1.11.2
[46]. The poor rotamers and outliers in predicted model was
refined by WinCoot [47] and subsequently validated by multiple
evaluation tools. MolProbity server (http://molprobity.manchester.
ac.uk/) was used to analyze the Ramachandran score, Ramachan-
dran outliers, poor and favored rotamers, bad angles and bonds.
ERRAT server (https://servicesn.mbi.ucla.edu/ERRAT/) was utilized
to calculate the overall quality factor of modeled structure.
2.2. Selection of inhibitors
Initially, 30 inhibitors were retrieved against Ca
v
2.2 through
extensive literature survey (Table S1). These inhibitors were non-
peptide mimetics of
x
-conotoxins (MVIIA, CVID and GVIA) isolated
from cone snail. From these 30 inhibitors, 7 compounds were fil-
tered out based on their pore binding positions (Table S2). In C1,
three important amino acids mimetics (R10, L11 and Y13) of
x
-
conotoxin MVIIA were attached to dendritic scaffold [41]. In C2-4
benzothiazole scaffold [48,38] and in C5 and C6 contained
anthranilamide scaffold that projected the side chain mimetics of
the key residues (K2, Y13 and R17) in
x
-conotoxin GVIA [43,48].
C7 shared a similar pattern to that of C6, except bearing an
anthranilamide scaffold that was modified by replacing phe-
noxyaniline substituent with a diphenylmethylpiperazine moiety
[40]. 2D structures of these inhibitors were drawn by ChemDraw
Pro 12.0 [49] and converted into 3D coordinates that were further
energy minimized using Avogadro
Ò
[50] tool through Merck
Molecular Force Field (MMFF94) and steepest descent algorithm
[51].
2.3. Molecular docking analysis
Molecular docking analysis was performed through PatchDock
[52] to evaluate the interaction pattern of Ca
v
2.2 with Ca
2+
ion
and selected compounds. For Ca
v
2.2 and Ca
2+
ion docking, Ca
2+
ion selective and permeable residues of Ca
v
2.2 were obtained
through UniProtKB (www.uniprot.org/uniprot/Q00975), were pro-
vided. For all docking runs, modeled Ca
v
2.2 and compounds were
subjected to rigid docking though small-scale flexibility, implicitly
by allowing few steric clashes and intermolecular penetrations.
PatchDock identifies geometric patches through segmentation
algorithm, surface matching, filtering and scoring [52]. After
detailed comparative analysis of the binding sites of inhibitors,
suitable candidate solutions were chosen on the basis of Root Mean
Square Deviation (RMSD) clustering. Based on these findings, out of
30 docked compounds (Table S1,Fig. S1) only 7 were shortlisted on
the basis of their binding abilities at the ion selectivity and perme-
ability region of Ca
v
2.2 pore (Table S2). These compounds were fur-
ther scrutinized by published IC
50
values [40–44]. The interactions
were carefully evaluated through UCSF Chimera 1.11.2 [46].
Graphical visualization of hydrogen bonding, hydrophobic and
electrostatic interactions were analyzed by LigPlus [53] and Dis-
covery Studio 4.5 [54].
2.4. Molecular dynamics simulation analysis
C1 was selected for detailed analysis owing to its binding pref-
erence at the SF ring of pore region by interacting with three resi-
dues of EEEE motif (E314, E663, and E1365) (Table S2). In order to
2358 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

evaluate comparative conformational readjustments, apo-Ca
v
2.2,
Ca
2+
ion and C1-bound modeled Ca
v
2.2 (1771 residues; Phe76-
Pro1846) complexes were subjected to MD simulation assay
through Visual Molecular Dynamics (VMD) [55] and Nanoscale
Molecular Dynamics (NAMD) tools [56]. For all three systems, psf-
gen plugin of VMD was utilized to generate protein structure files
(PSF). These structures were solvated by Heimut Grubmuller’s SOL-
VATE program [57] that fills empty space inside the pores by sur-
rounding it with water. The undesired water molecules were
removed. The solvated structures were embedded in the remaining
system by deleting unwanted water molecules in the hydrophobic
region of Ca
v
2.2 using VMD.
An automated system of VMD generated lipid bilayer by
Membrane Builder by specifying 1-palmityol 2-oleoyl phos-
phatidylcholine (POPC) membrane patch of length 128 Å in X
and Y coordinates and CHARMM27 force field [58]. VMD was
utilized to align partially solvated Ca
v
2.2 tetramer with POPC
lipid membrane. Ca
v
2.2 was settled in the membrane with the
hydrophobic residues; while, the orientation of protein was
adjusted to avoid its overlap with lipid molecules. Subsequently,
the overlapping water molecules were removed. VMD Solvate
plugin was used for the solvation of entire system by placing
it in the specified size water box. Ionization of all systems was
performed by VMD Autoionize plugin that generates specific ionic
concentration of NaCl (0.4 mmol/L) and transmutes water mole-
cules into ions. The membrane patches were initially equili-
brated by short simulation runs (0.5 ns) while keeping the
system static, except lipid tails to disorder in fluid-like bilayer.
The temperature was set to 300 K by employing Langevin
dynamics to maintain constant temperature and 1 atm pressure
(by Langevin piston method) throughout production simulation.
Minimizations were carried out for 1000 steps. NAMD 2.9 was
used to perform the equilibrium MD simulations for all systems.
Instead of standard CHARMM parameter file, a modified param-
eter file containing ‘‘NBFIX” with correction for carbonyl oxygen-
Ca
2+
ion interaction (distance restrained to 2.85 Å) was used.
This correction is essential for the selectivity filter (SF) in
Ca
v
2.2. SwissParam [59] was utilized for the topology file gener-
ation of compounds. The periodic boundary conditions (PBC) and
Particle Mesh Ewald (PME) grid size was also set for all systems
in the NAMD configuration files. The outputs were analyzed hav-
ing disordered lipid tails. MD simulations were run with full
dynamics that are essential for the setting of unusual atomic
positions in the systems, followed by minimization and equili-
bration. Harmonic constraints were enforced on Ca
v
2.2. The min-
imization and equilibration steps were executed for 0.5 ns by
NAMD2 [56]. The full systems were then equilibrated by keeping
protein channel unconstrained for 0.5 ns. Fully equilibrated sys-
tems were analyzed by computing Root Mean Square Deviation
(RMSD) and Root mean square fluctuation (RMSF) plots through
VMD. The final production runs were carried out for 100 ns. The
resulting outputs were analyzed by VMD tool [55] and UCSF Chi-
mera 1.11.2 [46].
3. Results
3.1. Model analysis of Ca
v
2.2
The crystal structure of Oryctolagus cuniculus Ca
v
1.1 [60] (PDB
ID: 5GJW; resolution: 3.6 Å; sequence coverage: 63%; sequence
identity: 50%) was selected as a template to model 3D structure
of Ca
v
2.2 through SWISS-MODEL (swissmodel.expasy.org). An RMSD
value of 0.196 Å was observed upon superimposition of template
and target structures. Ramachandran plot designated the presence
of 90.23% residues of Ca
v
2.2-modeled structure in the favored
region, 97.60% residues as favored rotamers, 0.17% residues as poor
rotamers and 1.46% residues as Ramachandran outliers. The
observed ERRAT quality factor value was 81.46%.
Human Ca
v
2.2 also known as N-type calcium channel consists
of 2,339 residues with a molecular mass of ~262 kDa comprises
4 domains (D
I
-D
IV
), each contains 6 transmembrane (S1-S6)
a
-
helices (Fig. 1A) [45]. Two helices (S5 and S6) from each domain
constitute a pore that is essential for the conduction of Ca
2+
ions
(Fig. 1)[45]. The voltage sensor domain (VSD) consists of 4 helices
(S1-S4) in each domain (Movie S1). Any voltage change across the
cell membrane is sensed by VSD that surrounds the pore region. In
between the S5 and S6 helices, a P-loop is present that acts as a
selectivity filter for Ca
2+
ion permeability [45]. The N- and C-
termini of channel are localized in the cytoplasmic environment.
In addition to these four domains, there are two additional
domains named as EF-hand (helix-loop-helix domain) and IQ
(isoleucine-glutamine motif) that are crucial for Ca
2+
binding. EF-
hand is a helical domain that is flanked by a 12-residue loop from
both sides [61]. It is involved in binding with Ca
2+
ions through
undergoing multiple conformational changes [61]. IQ motif in IQ
domain interacts with Ca
2+
/calmodulin.
3.2. Ca
2+
ion coordination
As voltage-gated calcium channels (VGCCs) allow Ca
2+
ions to
pass through the pore region, Ca
2+
ions were coordinated with
the modeled Ca
v
2.2 structure. Among surrounding residues,
Glu314 and Glu1365 residues potentially coordinated with Ca
2+
ion having bond lengths of 3.12 Å and 3.25 Å, respectively (Fig. 2).
3.3. Molecular docking analysis
Multiple reports suggested that non-peptide analogues of
x
-
conotoxins may target Ca
v
2.2 channels [38,41,43,70,83–85]. These
non-peptide conotoxin mimetics mimic the scaffolds of
x
-
conotoxin MVIIA, CVID and GVIA [62]. In total, out of 30 docked
compounds (Fig. S1;Table S1), 7 [40–44] were selected on the
basis of their binding pattern at the pore region of Ca
v
2.2 shared
by S5 and S6 helices along with P-loop of each domain (Fig. 4,
S2). These compounds contain dendritic, benzothiazole or
anthranilamide scaffolds, attached with Y, L/K and R residues
(Fig. 3,Table S2). The IC
50
values of 7 selected compounds against
Ca
v
2.2 are listed in Table S2. Compound C1 exhibits a dendritic
scaffold. C2-4 contain benzothiazole, C5-6 have anthranilamide
scaffold, whereas C7 holds a modified anthranilamide scaffold with
a phenoxyaniline group that is replaced into diphenylmethylpiper-
azine moiety. These compounds may be used for the treatment of
chronic and neuropathic pain by blocking Ca
v
2.2 VGCCs.
Compound C1 interacted with the residues of pore and S6 seg-
ments of Ca
v
2.2. It exhibited favorable associations with 3 of the
selectivity filter (SF) ring residues (Glu314, Glu663 and Glu1365)
(Fig. 4I). Compounds C2 and C3 binding was prompted through
hydrophobic association with residues of D
I
(pore), D
III
(S5, pore
and S6) and D
IV
(pore and S6). Ala1652, Thr1653, Ser1696 residues
of Ca
v
2.2 were involved in hydrogen bonding with both com-
pounds (Fig. 4II, 4III). C4 exhibited association with D
III
(S5, pore
and S6) and D
IV
(pore and S6), while a single hydrogen bond was
observed with O-atom of Thr1653 residue (Fig. 4IV). Contrary to
C2-C4, compounds C5-C6 showed interactions with the residues
of D
I
(S5, pore and S6) and D
II
(S6). Moreover, a single H-bond
was observed via Asn697 and Met313 residues, respectively
(Fig. 4V, 4VI). Compound C7 exhibited a favorable number of
hydrophobic associations with the residues of D
I
(S5, pore and
S6), D
II
(S6), D
III
(pore) and D
IV
(pore). Furthermore, Met347,
Asn697, Thr1653 and Thr1363 residues exhibited hydrogen bond-
ing (Fig. 4VII). The schematic binding details are indicated in
Fig. S2. These findings illustrate that only compound C1 exhibited
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2359
Author's Personal Copy

bonding with three residues (E314, E663, and E1365) of the Ca
2+
ion SF ring, therefore, compound C1 may be a good choice for
MD simulation assay.
3.4. Molecular dynamics simulation analysis
Compound C1 was selected for detailed binding pattern and
conformational ensemble study based on its localization at the SF
ring and central cavity of pore. In order to explore the overall sta-
bility and time-dependent conformational transitions in Ca
v
2.2
upon binding to C1 and Ca
+2
ions, molecular dynamics (MD) simu-
lations were performed by utilizing Visual Molecular Dynamics
(VMD) and Nanoscale Molecular Dynamics (NAMD) tools. Systems
preparation (apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1) for MD
simulations were carried out by wrapping the Ca
v
2.2 structural
model in 1-palmityol 2-oleoyl phosphatidylcholine (POPC) mem-
brane to create a native-like surrounding (Fig. S5). All systems
were minimized and equilibrated to analyze the passage of Ca
2+
ions through embedded ion channel in the lipid membrane.
The overall protein stability and time-dependent interactions
were monitored by generating PDB files at different time intervals.
Subsequently, underlying conformational and structural changes
in the protein channel in association with Ca
2+
ion and C1 were
analyzed. Root mean square deviation (RMSD), root mean square
fluctuation (RMSF), radius of gyration (Rg), radial pair distribution
function g(r) and number of hydrogen bonds were calculated
throughout 100 ns time scale.
To examine the convergence in all three systems, RMSD values
computed through C
a
-atoms were plotted against time (Fig. 5A).
RMSD scores observed over 100 ns for apo-Ca
v
2.2 and both com-
plexes displayed an abrupt rise up to 10 nm. All simulated struc-
tures acquired a stable state at 40 ns. The apo-Ca
v
2.2 and Ca
2+
-
Fig. 1. N-type Ca
2+
channel subunit alpha-1B (Cav2.2) structural and topological representation. (A) Ca
v
2.2 specific domains DI-DIV, represented by corresponding colors.
Each domain consists of six transmembrane segments (S1-S6), connected by P-loops. The starting and ending residues of segments are labeled in dark blue color along with
the glutamate residue in P-loop between S5 and S6 that is important for Ca
2+
ion permeability and selectivity (shown in red color). The EF-hand domain is indicated in sky
blue color, while IQ domain is shown in orchid color. (B) Homology model of Ca
v
2.2 indicating the membrane organization and helical arrangement in the respective color.
The loop regions between the domains of modeled structure is hidden to avoid ambiguity. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
2360 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

bound Ca
v
2.2 attained a stable state within a range of 11–15 nm
whereas, C1-bound Ca
v
2.2 demonstrated an increasing trend in
the RMSD profile (18–19 nm). RMSF plot revealed certain fluctuat-
ing residues at the loop regions of Ca
v
2.2 between D
II
and D
III
(Fig. 5B, Table 1). Glu314, Glu663, Glu1365 and Glu1655 residues
of SF ring involved in Ca
2+
ion selective permeability exhibited
lower fluctuations during simulation assays; however, these resi-
dues were more stable in Ca
v
2.2-Ca
2+
complex as compared to
apo-Ca
v
2.2 and Ca
v
2.2-C1 complex. In addition to SF ring, Gly353,
Ala706, Ala1413 and Ala1705 residues (the G/A/A/A ring) at inner
gate of channel and interacting residues lying at S4-S5 linker were
quite stable in both complexes. Major fluctuations were observed
in the loop regions, while
a
-helices and ion selectivity residues
remained stable during the course of simulation.
Radius of gyration (Rg) values were calculated for both com-
plexes and apo-Ca
v
2.2 to get insight into compactness of protein
Fig. 2. Ca
2+
ion coordination at the selectivity filter (SF) ring of Ca
v
2.2. (A) Top view of Ca
v
2.2. The four homologous repeats exhibit a clockwise arrangement, when viewed
from extracellular side. VSD of each domain is labeled by specific colors. Glu residues responsible for the Ca
2+
ion selectivity and permeability are shown in red sticks. (B)
Interactions are analyzed by LigPlus [53]. (C) The positions of Glu residues are labeled in respective colors. The distances between them are illustrated by dotted lines. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 3. 2D structures of selected mimetics. C1, MVIIA (dendritic scaffold attached to Y13, L11 and R10 residues); C2, CVID (benzothiazole scaffold attached to K2, Y13 and R17
residues); C3, GVIA (benzothiazole scaffold attach to K2, Y13 and R17 residues); C4, GVIA (benzothiazole scaffold attached to K2, Y13 and R17 residues); C5, GVIA
(anthranilamide scaffold attached to K2, Y13 and R17 residues); C6, GVIA (anthranilamide scaffold attached to K2, Y13 and R17 residues) and C7, GVIA (modified
anthranilamide scaffold attached to K2 and R17 residues) [41,48,38,43,40]. The scaffolds are indicated in grey color.
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2361
Author's Personal Copy

throughout MD simulation. Rg analysis provides quantitative
information about changes in the tertiary structures linked to the
compactness of the simulated system. Higher Rg values from the
original depict the protein structure expansion. The apo-Ca
v
2.2
exhibited a decreasing trend in Rg values from 56.74 Å to 52.5 Å,
highlighting the compactness of system at 40 ns onwards. The
structural equilibrium for Ca
2+
-bound Ca
v
2.2 was observed
throughout the simulation assay with the Rg values ranging
between 55.79 Å and 57.17 Å, indicating the consistent system
compactness (Fig. 5C). The highest values of Rg for C1-bound
Ca
v
2.2 complex was detected as 58.96 Å during 1 ns of MD simu-
lation. Afterward, Rg values gradually decreased up to 52 Å at
70 ns of time scale and remained stable up to 100 ns. Conse-
quently, Ca
v
2.2-C1 complex exhibited a tight packing as compared
to Ca
+2
-bound Ca
v
2.2 suggesting more firmness in channel upon
binding to C1. MD simulation trajectory files for both complexes
were analyzed for hydrogen bond interactions, which remained
stable during the entire simulation time (Fig. 5D). The highest
numbers of hydrogen bonds in Ca
2+
-bound and C1-bound Ca
v
2.2
were observed at 45.8 ns (403) and 89.4 ns (417), respectively.
Radial pair distribution function g(r) provides the probability of
finding particles at a certain distance r. G(r) values were calculated
for four residues (Glu314, Glu663, Glu1365 and Glu1655) that
form SF ring to explore their distances with Ca
2+
ion (Fig. 6A,B).
The atom OE1 of Glu314 exhibited interactions with Ca
2+
ion.
OE1 atoms of Glu314 and Glu1365 residues more likely preferred
to localize (11832.6 and 10041.5 times) in the vicinity of Ca
2+
ion
at a distance of 2.15 Å (Table 2). Similarly, OE1 atoms of Glu663
and Glu1655 residues exhibited highest peaks at distances of
2.25 Å and 4.85 Å, respectively. Interatomic distance calculation
among Glu314, Glu663, Glu1365 and Glu1655 residues suggested
that Glu1365-Glu1655 residues were closer to Ca
+2
ion at a dis-
tance of 3.75 Å whereas, Glu663-Glu1655 residues exhibited a dis-
tance of 8.65 Å.
Fig. 4. Binding analysis of compounds C1-7 and Ca
v
2.2. C1-7 are shown in dark green, dark pink, dark sky blue, chartreuse, orange, dark magenta and firebrick colors in circles
I-VII, respectively. The channel interacting residues are shown in grey wires and labeled in black color. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
2362 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

Furthermore, g(r) plot results for Ca
2+
ion and OE1 atoms of 4 glu-
tamate residues (Glu314, Glu663, Glu1365 and Glu1655) were val-
idated by distance calculation analysis through UCSF Chimera
1.11.2 for the PDB files generated at 10 ns and 99 ns of simulation
time. Evidently, distances for Glu314, Glu663 and Glu1365 residues
to Ca
2+
ion were slightly reduced, while Glu1655 residue moved
away (4.86 Å to 6.04 Å) from Ca
2+
ion, indicating the movement of
Ca
2+
ion towards Glu314, Glu663 and Glu1365 residues (Fig. 6C,D).
Fig. 5. Time-dependent analysis of apo-Ca
v
2.2 (yellow), Ca
v
2.2-Ca
2+
ion (pink) and Ca
v
2.2-C1 (green) complex at 100 ns. (A) RMSD versus time plot. (B) RMSF plot with
labeled momentous fluctuating residues in respective colors. (C) Radius of gyration (Rg) plot with respect to time. (D) Hydrogen bonds analysis. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)
Table 1
Root Mean Square Fluctuation (RMSF) for apo- Ca
v
2.2, Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1 complex during 100 ns MD simulation.
Residues Apo-Ca
v
2.2 (nm) Ca
v
2.2-Ca
2+
ion (nm) Ca
v
2.2-C1 (nm)
Selectivity filter (SF) ring Glu314 2.39 1.18 2.24
Glu663 2.26 1.35 2.14
Glu1365 2.27 1.22 2.06
Glu1655 2.26 1.12 2.69
G/A/A/A ring Gly353 2.42 1.57 1.28
Ala706 2.24 1.59 1.22
Ala1413 2.06 1.60 1.76
Ala1705 2.22 1.48 2.22
S4-S5 linker residues Pro222 2.36 1.84 1.51
Leu223 2.41 1.68 1.31
Ser608 2.50 1.92 1.83
Ile609 2.28 1.76 1.80
Asn1281 2.13 1.50 2.07
Val1282 2.09 1.50 2.07
Lys1297 2.42 1.82 2.44
Ala1598 2.46 2.13 2.41
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2363
Author's Personal Copy

3.5. Conformational change analysis
Through comparative MD simulation analysis for Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1 complexes in comparison to apo-Ca
v
2.2, details
of critical residues including gating charge residues of VSDs, ion
and inhibitor interacting residues at the pore region, pore lining
residues, hydrophobic residues in the inner gate, S4-S5 linker
interacting residues to S6 segments of pore domain were observed.
Moreover, the influence of these residues in the opening and clos-
ing of Ca
v
2.2 inner gate was explored in both systems, in particu-
larly at SF and ion permeable regions of Ca
v
2.2.
For Ca
v
2.2-Ca
2+
ion complex, all 4 domains (D
I
-D
IV
) were super-
imposed at 99 ns time scale (Fig. 7A). Though multiple structural
variations were observed in all 4 domains (D
I
-D
IV
)ofCa
v
2.2, diver-
gence was more prominent at the loop and linker regions that
allow the movement of individual segments during voltage sensing
and inner gate opening/closing. These structural variations exhib-
ited by D
I
-D
IV
domains may lead to asymmetric tetrameric confor-
mation of Ca
v
2.2. In this conformation, S1-S4 VSDs of Ca
v
2.2 bear
similar but non-identical structural features. Each Ca
v
2.2 VSD con-
tains four transmembrane
a
-helices (S1-S4). Upon superimposi-
tion, these transmembrane segments (S1-S4 VSDs) revealed an
RMSD value of ~1.3 Å (Fig. 7A). Positively charged residues (Arg
and Lys) of S4 segments at every third and fourth position were
labeled as R1-R6. The human Ca
v
2.2 S4 segment exhibits R1-R5
on VSD
I
, R2-R6 on VSD
II
, R1-R6 on VSD
III
and R2-R6 residues on
VSD
IV
(Fig. 7A). In each VSD, S4 segment-specific gating charged
residues were aligned at one side (Fig. 7B). R1-R4 residues were
positioned above the conserved occluding Phe residue in CTC
(charge transfer center), while R5 and R6 resides were lined below
(Fig. 7C). Extracellular pointing of 4 out of 6 conserved R residues
allow forming a depolarized or upstate generation. The upstate
Fig. 6. Radial pair distribution function g(r) analysis. (A) g(r) plot for oxygen atoms (OE1) of 4 glutamate residues (Glu314, Glu663, Glu1365 and Glu1655) and Ca
2+
ion. (B) g
(r) plot for interatomic distances (OE1 atoms) of Glu314, Glu663, Glu1365 and Glu1655 residues. (C) and (D) Time-dependent distances measured by UCSF Chimera 1.11.2
among OE1 atoms of Glu314, Glu663, Glu1365 and Glu1655 residues and Ca
2+
ion using PDB files generated at 10 ns and 99 ns of MD simulation run, respectively.
Table 2
Radial pair distribution function g(r) for Ca
2+
ion and its selectivity filter (SF) ring
residues in Ca
v
2.2-Ca
2+
ion complex.
Interaction between Glu- Ca
2+
ion
and Glu-Glu
Distances
(Å)
Radial pair distribution
function g(r)
Glu314 (OE1)-Ca
2+
ion 2.15 11832.6
Glu663 (OE1)-Ca
2+
ion 2.25 1837.5
Glu1365 (OE1)-Ca
2+
ion 2.15 10041.5
Glu1655 (OE1)-Ca
2+
ion 4.85 969.6
Glu314-Glu1365 6.95 255.4
Glu314-Glu1655 5.55 427.14
Glu314-Glu663 5.55 249.2
Glu663-Glu1365 5.55 223
Glu663-Glu1655 8.65 69.7
Glu1365-Glu1655 3.75 283.4
2364 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

levels may vary among different structures or within the domains
of same structure. These levels are determined by gating charges
due to residues above the bulky hydrophobic residue (Phe) of
CTC [63]. The CTC is composed of conserved negative or polar resi-
dues including a highly conserved occluding bulky hydrophobic
residue localized on S2 segment plus an invariant Asp residue,
localized on S3 segment. These residues along with another polar
or negatively charged residue (An1 and An2) of S2 segment con-
tribute in the transmembrane movement by interacting with gat-
ing charged residues on S4 segment. In VSD tetramer,
Glu134/518 residue localized on S2 of VSD
I, II
, Asp1187 of VSD
III
and Asn1507 of VSD
IV
interact with R2 and R3, while R5 interacts
with Glu144/528/1197/1517 of S2 segment. As evident in rabbit
Ca
v
1.1 [60], in human Ca
v
2.2, Asp residues of VSD
II
and VSD
IV
S3
segments were observed in the formation of CTC (Fig. 7).
To observe structural conformations, VSDs were superimposed
for both complexes (Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1). These results
elucidated noticeable structural divergence where VSD
II
and VSD
III
exhibited prominent conformational changes at S3 and S4 seg-
ments (Fig. 7D, E). Remarkably, S4 and S3 helices of Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1 complexes displayed predominant kinks (Fig. 7-
C-F). In contrast, S2
IV
and S3
IV
segments exhibited better structural
alignment compared to other VSDs. Major conformational varia-
tions were observed in the S4-S5 linker region, revealing that dur-
ing voltage sensing, the flexibility of linker region may have an
influential role in the S4 segment movement.
3.6. Outer vestibule of Ca
v
2.2
The primary structures of outer vestibule of all four domains of
Ca
v
2.2 exhibited well aligned SF residues (Fig. 8A). The simulated
structures of both Ca
2+
ion and C1-bound Ca
v
2.2 revealed two rings
of charged residues at the outer vestibule. The inner ring is com-
posed of four negatively charge residues (Glu314
I
, Glu663
II
,
Glu1365
III
and Glu1655
IV
; the EEEE motif) neighbored by Asp664
II
and Arg1650
IV
[64]. The outer ring is composed of Asp325
I
,
Arg1634
IV
and Glu1651
IV
. The hydrophobic ring was composed of
Ile319
I
, Val668
II
, Val1370
III
and Ile1660
IV
residues that parted the
outer and inner rings (Fig. 8B-C). Recent studies showed that
x
-
conotoxin GVIA forms salt bridges with the outer ring of Ca
v
2.2;
however, the apolar residues of hydrophobic ring sterically hinder
the binding of toxin with the outer ring [64]. In our study, Ca
2+
-
bound Ca
v
2.2 exhibited a wide groove at the pore region sour-
rounded by negatively charged residues that may allow Ca
2+
ion
binding at inner ring. Similarly, compound C1 interaction was
mediated by SF ring in the pore. Such conformation may allow
the appropriate binding of C1 with the pore residues of Ca
v
2.2.
3.7. Analysis of Ca
v
2.2 pore domain region
VSDs and pore structure are coupled in such a manner that S4
segment of VSD is connected to S5 of the pore domain via S4-S5
linker. Subsequently, a helical structure surrounds the pore domain
and runs parallel to the intracellular side of membrane. Each VSD is
localized adjacent to the neighboring pore domain and influences
the gate of its own domain as well as of neighboring domain. Chan-
nel pore region and VSDs are coupled together due to direct inter-
actions of S4-S5 linkers and S6 segments via Gly353 (S6
I
), Ala706
(S6
II
), Ala1413 (S6
III
) and Ala1705 (S6
IV
). Sequence alignment
(Figs. S3, S4) and structural comparison of Ca
v
1.1 and Ca
v
1.2 chan-
nels indicate that instead of Gly residue in the S6
III
segments of
Ca
v
1.1 and Ca
v
1.2 (G/A/G/A ring), Ca
v
2.2 contains Ala residue that
Fig. 7. Superimposition of four protomers. (A) Comparison of VSD
I-IV
S2 and S4 segments in Ca
v
2.2-Ca
2+
ion complex. Individual VSDs of Ca
v
2.2 are colored blue, golden, lime
green and violet red colors, respectively. The gating charge residues (R1-R6) in S4 and the shifted residue in S4 segment of VSD
III
(cyan color) along with anion1 (An1), anion2
(An2) and occluding Phe residue in CTC (Charge Transfer Center) localized on S2 segment are shown as sticks, while S2 and S4 segments are indicated in wire form. The S2 and
S4 segments of 4 VSDs are quite similar; however, they have high conformational divergence. (B) Side view of S4 segments. All the gating charge residues on S4 segments are
aligned at one side of helix. (C-F) S2-S4 segments of 4 VSDs of Ca
v
2.2-Ca
2+
ion complex are structurally aligned relative to VSDs of Ca
v
2.2-C1 complex. VSDs of Ca
2+
ion-bound
Ca
v
2.2 are colored blue, golden, green and violet red colors, respectively. The inhibitor-bound Ca
v
2.2 is shown in spring green color. (For interpretation of the references to
color in this figure legend, the reader is referred to the web version of this article.)
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2365
Author's Personal Copy

makes ‘G/A/A/A ring’ [65]. Recent studies potentiate that mutations
of G/A/G/A residues impact the movement of VSDs [66]. The G/A/A/
A position in the S6 is faced toward the S4-S5 linker to induce a
direct interaction between them for tight packing. In apo-Ca
v
2.2,
the G/A/A/A ring residues lacked close interactions with loop resi-
dues localized between the S4-S5 linker and the S5 segment
(Fig. 9B). In Ca
2+
-bound Ca
v
2.2, Gly353, Ala706 and Ala1413 resi-
dues were involved in interaction with the neighboring residues
of loop region to facilitate the opening of channel pore at the entry
gate for Ca
2+
ions (Fig. 9D). In contrast, in Ca
v
2.2-C1 complex,
Ala1705 with Lys1597 and Ala1598 residues of the similar loop
were involved in channel activity (Fig. 9F). Due to lack of Gly353,
Ala706 and Ala1413 involvements in interaction, this loop moved
towards S6 segment that prevented its interaction with S4-S5 lin-
ker residues and narrowed the tunnel at the inner gate to close the
passage for ion flow.
Lys218-Leu223 (S4
I
-S5
I
) and Val1814-His1844 (IQ domain)
regions of apo-Ca
v
2.2 and Ca
v
2.2-Ca
2+
ion complex attained loop
conformations (Fig. S7), while in Ca
v
2.2-C1 complex; these regions
adopted helical conformations (Fig. S8). The helical break of S4
III
was involved in kinking or bending the segment. S3
III
was posi-
tioned in parallel with the membrane as compared to other VSDs.
The position of S4-S5 linker was also varied in all VSDs. In all sys-
tems, loop regions exhibited more fluctuations, except SF region of
channel pore. Similarly, more conformational readjustments were
visible at the EF-hand and IQ domains in all systems. The IQ
domain was more inclined to inner gate of channel in Ca
v
2.2-
Ca
2+
ion complex.
The pore domain of Ca
v
2.2 exhibits a pseudo-four-fold symme-
try similar to rabbit Ca
v
1.1 [60]. At sequence level, the extracellular
pore loop in the pore domain adjacent to SF of all 4 domains exhib-
ited clear differences. In all systems, a close comparison among
Fig. 8. Outer vestibule of four domains of Ca
v
2.2. (A) Sequence alignment of Ca
v
2.2 outer vestibule region in all four domains. The highlighted region signifies the selectivity
filter lining residues. (B) Extracellular and cytosolic views of Ca
v
2.2-Ca
2+
ion complex outer vestibule. (C) Extracellular and cytosolic views of Ca
v
2.2-C1 complex outer
vestibule. Coloring scheme for residues are: hydrophobic, gray; polar, green; negatively charge, red and positively charge, blue. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of this article.)
2366 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

Fig. 9. Conformational transition analysis in apo-Ca
v
2.2, Ca
v
2.2 complex with Ca
2+
ion and C1. (A), (C) and (E) Membrane topology and side views of apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1 complexes. D
I
-D
IV
domains are indicated in blue, gold, green and violet red colors, respectively. EF-hand domain is shown in sky blue, IQ domain is in
orchid, Ca
2+
ion in gray and C1 is indicated in dark green color. (B), (D) and (F) Cytosolic view of apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1 complexes. The dotted lines at the
pore region indicate a cooperative unit. Residues of G/A/A/A ring on S6 and their interacting residues on S4-S5 linkers are labeled in black color. The spheres indicate their side
chains. (G), (H) and (I) Side views of channel having superimposed 4 domains of apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1 complexes, respectively embedded in membrane at
67 ns time-scale. The domain segments are labeled in black color. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of
this article.)
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2367
Author's Personal Copy

Fig. 10. Architecture and tunnel comparison in apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
and Ca
v
2.2-C1 complexes. The VSDs have been omitted for better illustration. (A), (B) and (C) Side
views of permeation path of pore domains for apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
ion and Ca
v
2.2-C1structures, respectively. The ion conducting passage, calculated by MoleOnline
(https://mole.upol.cz/online/) is illustrated by purple-colored tunnel. D
I-IV
are indicated in blue, gold, green and violet red colors, respectively. Each segment of domain is
labeled in the respective domain color. The Ca
2+
ion is indicated in gray and C1 in dark green color. Pore radii along with the pore distances is displayed for each structure. Top
red arrows indicate the ion selective and permeable regions localized at the narrowest tunnel (radius close to 0). Furthermore, the closest part of hydrophobic area (position
close to 0) formed by Val351, Phe704, Phe1411 and Phe1703 residues is indicated by red arrows (bottom). The blue to yellow colored plot demonstrates the hydrophobic
strength of pore forming residues. (D), (E) and (F) Circle displays extracellular view of SF residues of the pore region are indicated by red sticks for apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
ion
and Ca
v
2.2-C1 complexes, respectively. Cytoplasmic view of the occluding residues of inner gate (Val351, Phe704, Phe1411 and Phe1703) is shown in stick representation.
(G), (H) and (I) Tunnel forming residues along with their distances measured by UCSF Chimera 1.11.2 for apo-Ca
v
2.2, Ca
v
2.2-Ca
2+
and Ca
v
2.2-C1 structures, respectively.
Dotted lines indicate distances between adjacent residues of tunnel. Adjacent to tunnel, only two domains were shown to avoid ambiguity. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)
2368 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

repeats demonstrated that P-loop regions of D
I
and D
III
were longer
than the P-loops of D
II
and D
IV
(Fig. 9). In addition, P-loops of D
I
and
D
III
exhibited 1–2 pairs of antiparallel b-strands, respectively
(Fig. 9). P-loop of D
III
exhibited an extended conformation and
more protrusion into the extracellular space. Through the P-loop
region,
a
1 and
a
2 subunits were associated together. Upon super-
imposition of all 4 domains at 10, 60, 67 and 99 ns time scales,
varying structural preferences were observed for individual VSD
segments (Figs. S6, S7 and S8). Pronounced variations were visible
in S3 and S4 segments (VSD
II
) of apo-Ca
v
2.2 and Ca
v
2.2-Ca
2+
ion
complex, while in Ca
v
2.2-C1 complex, VSD
III
exhibited more bends
within membrane (Fig. 9G-I, S6, S7 and S8). P1 and P2 helices of S3
and S4 segments were structurally similar in Ca
v
2.2-Ca
2+
ion com-
plex than that of Ca
v
2.2-C1 complex and apo-Ca
v
2.2. S5 and S6 seg-
ments were conformationally dissimilar in both complexes. In case
of Ca
v
2.2-Ca
2+
ion complex, S4-S5 linker helices of D
III
were more
divergent than others.
At the pore region, Glu residues (Glu314, Glu663, Glu1365 and
Glu1655) localized at the periphery of ion selectivity region are
clustered to regulate the tunnel passage (Fig. 10). The apo-Ca
v
2.2
displayed a wide tunnel radius at SF region as compared to other
two complexes (Fig. 10A-C). Underneath the selectivity filter vesti-
bule, a typical hydrophobic cavity passes along with the side por-
tals penetrated by transverse membrane lipids, a feature similar
to rabbit Ca
v
1.1 [45] and bacterial Na
v
channels [67]. The asym-
metric S6 bundle of Ca
v
2.2 is tightly screwed at the inner activation
gate. The apo-Ca
v
2.2 and Ca
v
2.2-C1 complex exhibited narrow tun-
nels in comparison to Ca
v
2.2-Ca
2+
ion complex, indicating a closed
pore conformation through channel (Fig. 10A-C,G-I). Three aro-
matic residues (Phe704, Phe1411 and Phe1703), localized at the
corresponding positions at S6
II-IV
along with Val351 of S6
I
(pore-
occluding S6 residues) mediated the pore seal formation at the
cytosolic region (Fig. 10D-F). Below the aromatic ring, hydrophobic
residues of S6
I-II
(Leu and Gly/Ala) and S6
III-IV
(Val and Ala) facili-
tated in the channel closure. Furthermore, hydrophobic residues
(Val351, Phe704, Phe1411 and Phe1703) lined at the narrowest
point along the pore enclosed the entrance to the SF vestibule.
3.8. Structural comparison of human Ca
v
2.2
In order to explore the open or close state of channel, Ca
v
2.2-
Ca
2+
ion and Ca
v
2.2-C1complexes were superimposed with apo-
Ca
v
2.2 simulated structure, relative to the pore domains. These
structures revealed an RMSD value of ~1.24 Å (Fig. 11). The VSDs
of apo-Ca
v
2.2 possessed a depolarized or ‘up’ conformation and a
closed inner gate, suggesting a potentially inactive state. All 4
homologous repeats/domains exhibit a counterclockwise arrange-
ment when viewed from intracellular side and remain conserved in
all eukaryotic calcium and sodium channels [60]. In our analysis,
there was a subtle clockwise rotation of VSDs in Ca
2+
-bound
Ca
v
2.2 VSD
III
relative to apo-Ca
v
2.2, while a counterclockwise rota-
tion was observed in C1-bound Ca
v
2.2 (Fig. 11). Moreover, a coun-
terclockwise movement was detected in VSD
II
of Ca
2+
-bound
Ca
v
2.2 relative to apo-Ca
v
2.2 (cytosolic view). The observations of
inner gate suggested more distances among C
a
atoms of S6
III
for
Ca
2+
-bound Ca
v
2.2 in comparison to apo-Ca
v
2.2 and C1-bound
Ca
v
2.2, whose inner gate is closed, while shorter distances were
observed for S6
I
,S6
II
and S6
IV
segments. Due to a kink in S6
IV
seg-
ment of C1 bound Ca
v
2.2, apo-Ca
v
2.2 exhibited more distance. Sim-
ilarly, S6
II
segments of both apo and C1-bound Ca
v
2.2 were
Fig. 11. Structural comparison of apo-Ca
v
2.2with Ca
v
2.2, Ca
2+
- and C1-bound human Ca
v
2.2. The pore domain superimposition demonstrates a clockwise rotation of VSDs in
Ca
v
2.2-Ca
2+
ion complex (periplasmic view along the channel axis) relative to the apo-Ca
v
2.2, with the exception of VSD
II
that moves counterclockwise. Overall, C1-bound
Ca
v
2.2 reveals a subtle counterclockwise rotation.
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2369
Author's Personal Copy

different than that ofCav2.2 -Ca
2+
- complex Overall, the inner gates
of both apo-Ca
v
2.2 and C1-bound Ca
v
2.2 shared a similar confor-
mational pattern. An inward movement of S6 segments in
Ca
v
2.2-C1 complex and a reduced pore size at the activation gate
suggests that the gate is potentially closed. Thus the ‘‘up” (depolar-
ized) conformation of VSDs and closed pore dimension may signify
a potentially inactivated state of Ca
v
2.2 upon C1 inhibitor binding.
4. Discussion
Chronic pain is considered as a major issue that affects the over-
all life quality of patient. Approximately, ~1.5 billion people are
suffering with this condition globally [68]. Recently, a mouse
model expressing N-type Ca
v
2.2 VGCCs at the plasma membrane
of peripheral somatosensory neurons via
a
2d-1 accessory subunit
revealed the role of these channels in pain modulation [69], sug-
gesting that Ca
v
2.2 Ca
2+
channel may prove to be a potential drug
target for novel analgesic therapies. Purposefully, the mimicry of
venom peptides may hold a considerable promise due to lack of
detailed structural information of neuronal ion channel. The non-
peptide mimetics of MVIIA and GVIA
x
-conotoxins isolated from
marine cone snail efficiently bind to neuronal Ca
2+
channels at
low concentrations [9,62]. Here in this study, out of 30 known
non-peptide analogues of
x
-conotoxins (Table S1,Fig. S1), 7 pep-
tides (Fig. 3;Table S2) were shortlisted on the basis of their binding
abilities at the Ca
v
2.2 pore region. These blockers bearing dendritic,
benzothiazole or anthranilamide scaffolds (Fig. 3) may be used for
the effective management of chronic and neuropathic pain by
blocking Ca
v
2.2 VGCCs.
Subsequently, through in silico analysis, we compared confor-
mational ensemble of Ca
v
2.2 bound to conopeptide-mimetic alkyl-
phenyl ether-based analogue MVIIA (C1) [41] and Ca
2+
ion to
demonstrate potential anesthetic and analgesic impact of inhibitor.
C1, initially proposed by Horwell group [70] contains a dendritic
scaffold that mimics three key residues (Arg10, Leu11 and Tyr13)
of MVIIA conotoxin [41]. In our analysis, the dendritic scaffold
was found to be hydrophobically associated with Met313,
Phe346 and Met347 residues of Ca
v
2.2, while Glu1365 and
Thr1363 residues were linked to Arg10 and Tyr13 of C1 via hydro-
gen bonding. In Ca
v
2.2-C1 complex, RMSF profile comparison indi-
cated significant transitions in Ala260, Ala420, D
II
-D
III
liker region
(Ala985-Met1067, except Glu998 residue that exhibited more fluc-
tuation in the Ca
2+
-bound Ca
v
2.2) and Glu1332 residue of C1-
bound Ca
v
2.2. Interestingly, upon Ca
2+
ion permeability, D
II
-D
III
lin-
ker region acquired more stability than that of C1-bound Ca
v
2.2.
This linker region has been suggested as an important docking site
for the binding of synaptic proteins [71]. The inhibitor-bound
channel (52 Å) structure was more compact than Ca
2+
-bound
Ca
v
2.2 (57 Å) indicating that Ca
2+
-dependent structural destabiliz-
ing effects may lead in the channel opening (Fig. 5C). Possibly, Ca
2+
ion-dependent opening of voltage-gated channel may create a
functional binding site for synaptic vesicles in the N-type channels
as reported in the recent studies [72,73]. Another evidence sug-
gests that channel lacking D
II
-D
III
linker region is dramatically less
sensitive to MVIIA and GVIA conotoxins than the full-length con-
struct [74]. The presence of more structural rearrangements in
the D
II
-D
III
linker region of C1-bound Ca
v
2.2 may perhaps facilitate
key residues for toxin binding. Our simulations delineate signifi-
cant conformational changes at the proximal carboxyl-terminus
(EF-hand and IQ domain helix) of Ca
v
2.2 that attains an upright ori-
entation upon Ca
+2
ion permeability, (Fig. S7;Movie S2). In con-
trast, movement of C1-bound Ca
v
2.2 IQ domain helix was totally
different (Fig. S8;Movie S3). Possibly, such movements may be
linked with the extent of channel opening at the activation gate.
At IQ domain, a competitive binding of calmodulin (CaM) and
Ca
+2
binding proteins (CaBPs) influences the Ca
+2
-dependent facil-
itation and inactivation of the Voltage-gated channels [75–78],
while EF-hand motifs serve as Ca
+2
ion sensors. Indeed, following
the ion selectivity by Glu residues, such oscillations in the IQ helix
and EF-hand orientation may provide a docking site for CaM asso-
ciation to induce auto-inhibitory mechanism by Ca
+2
binding [79].
Evidently, SF ring is localized at the TM pore region with a large
external vestibule or central cavity lined by the S6 segments and
the intracellular activation gate formed due to intersection of S6
helices that exhibited more variations. Though tunnel lengths were
quite comparable (41.2 Å, 40 Å and 40.2 Å, respectively), in con-
trast to apo-Ca
v
2.2, both complexes (Ca
v
2.2-Ca
2+
and Ca
v
2.2-C1)
displayed remarkable changes in the architecture of channel tun-
nels (Fig. 10;Movie S4, S5). These findings illustrate that C1 bind-
ing to Ca
v
2.2 induces more narrowing of the pore size at the
activation gate as compared to Ca
2+
-bound Ca
v
2.2 due to differ-
ences in the hydrophilicity pattern at the selectivity region
(Fig. 10F and 10G). Generally, the inner lining of tetrameric channel
pore remains in the hydrophilic environment to allow the ion pas-
sage through the hydrophobic gate [80]. Subsequent occluding
residues (Val351, Phe704, Phe1411 and Phe1703) facing the intra-
cellular side contribute in the enhancement of hydrophobicity.
Such dynamic structural rearrangements within the selectivity fil-
ter are crucial for channel gating [81]. In our analysis, a pro-
nounced reduction in the tunnel volume at the selectivity filter
and its enhancement at the activation gate of Ca
+2
-bound Ca
v
2.2
suggests that ion binding allows an outward splaying of pore-
lining S6 helices to open the voltage gate. Recent evidence supports
the movement of S6 helices in both sodium and potassium chan-
nels that is initiated at the hinge-point in the middle of the helix
[82].
Collectively, given the analgesic efficacy and minimal side
effects of
x
-conotoxins [40], our study reveals MVIIA-associated
structural implications and subtle changes in the Ca
v
2.2 for the
development of better therapeutic intervention for chronic and
neuropathic pain. Clearly, MD-based conformational analysis of
pain blocking
x
-conotoxins may largely help in the potent inhibi-
tion of human presynaptic ion channels for devising promising
therapeutic strategy.
CRediT authorship contribution statement
Sameera: Formal analysis, Investigation, Writing - original
draft. Fawad Ali Shah: Formal analysis, Visualization. Sajid
Rashid: Conceptualization, Project administration, Resources,
Supervision, Validation, Writing - review & editing.
Acknowledgments
We acknowledge all members of Functional Informatics Lab,
National Centre for Bioinformatics especially Saima Younis and
Maryam Rozi for their indispensable help, support and
encouragement.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Declarations of interest
None.
2370 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy

Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.csbj.2020.08.027.
References
[1] Torrance N, Ferguson JA, Afolabi E, Bennett MI, Serpell MG, et al. Neuropathic
pain in the community: more under-treated than refractory?. PAIN
Ò
2013;154
(5):690–9.
[2] Mogil JS. Sex differences in pain and pain inhibition: multiple explanations of a
controversial phenomenon. Nat Rev Neurosci 2012;13(12):859–66.
[3] Scholz J, Woolf CJ. Can we conquer pain?. Nat Neurosci 2002;5(11):1062–7.
[4] Jensen TS, Gottrup H, Sindrup SH, Bach FW. The clinical picture of neuropathic
pain. Eur J Pharmacol 2001;429:1–3.
[5] Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science
2000;288(5472):1765–8.
[6] Millan MJ. The induction of pain: an integrative review. Prog Neurobiol
1999;57(1):1–64.
[7] Attal N, Cruccu G, Haanpää M, Hansson P, Jensen TS, et al. EFNS guidelines on
pharmacological treatment of neuropathic pain. Eur J Neurol 2006;13
(11):1153–69.
[8] Lee S. Pharmacological inhibition of voltage-gated Ca
2+
channels for chronic
pain relief. Curr Neuropharmacol 2013;11(6):606–20.
[9] Schroeder CI, Lewis RJ.
x
-conotoxins GVIA, MVIIA and CVID: SAR and clinical
potential. Mar Drugs 2006;4(3):193–214.
[10] Williams JA, Day M, Heavner JE. Ziconotide: an update and review. Expert Opin
Pharmacother 2008;9(9):1575–83.
[11] McGivern JG. Ziconotide: a review of its pharmacology and use in the
treatment of pain. Neuropsychiatr Dis Treat 2007;3(1):69.
[12] Bear B, Asgian J, Termin A, Zimmermann N. Small molecules targeting sodium
and calcium channels for neuropathic pain. Curr Opin Drug Discov Devel
2009;12(4):543–61.
[13] Yamamoto T, Takahara A. Recent updates of N-type calcium channel blockers
with therapeutic potential for neuropathic pain and stroke. Curr Top Med
Chem 2009;9(4):377–95.
[14] Cizkova D, Marsala J, Lukacova N, Marsala M, Jergova S, et al. Localization of N-
type Ca 2+ channels in the rat spinal cord following chronic constrictive nerve
injury. Exp Brain Res 2002 Dec 1;147(4):456–63.
[15] Yokoyama K, Kurihara T, Makita K, Tanabe T. Plastic change of N-type Ca
channel expression after preconditioning is responsible for prostaglandin E2-
induced long-lasting allodynia. Anesthesiology 2003;99(6):1364–70.
[16] Hatakeyama S, Wakamori M, Ino M, Miyamoto N, Takahashi E, et al.
Differential nociceptive responses in mice lacking the
a
1B subunit of N-type
Ca
2+
channels. NeuroReport 2001;12(11):2423–7.
[17] Kim C, Jun K, Lee T, Kim SS, McEnery MW, et al. Altered nociceptive response in
mice deficient in the
a
1B subunit of the voltage-dependent calcium channel.
Mol Cell Neurosci 2001;18(2):235–45.
[18] Saegusa H, Kurihara T, Zong S, Kazuno AA, Matsuda Y, et al. Suppression of
inflammatory and neuropathic pain symptoms in mice lacking the N-type Ca
2+
channel. EMBO J 2001;20(10):2349–56.
[19] Yaksh TL. Calcium channels as therapeutic targets in neuropathic pain. The
Journal of Pain. 2006;7(1):S13–30.
[20] Altier C, Zamponi GW. Targeting Ca
2+
channels to treat pain: T-type versus N-
type. Trends Pharmacol Sci 2004;25(9):465–70.
[21] Sabido-David C, Faravelli L, Salvati P. The therapeutic potential of Na
+
and Ca
2+
channel blockers in pain management. Expert Opin Invest Drugs 2004;13
(10):1249–61.
[22] Winquist RJ, Pan JQ, Gribkoff VK. Use-dependent blockade of Ca
v
2.2 voltage-
gated calcium channels for neuropathic pain. Biochem Pharmacol 2005;70
(4):489–99.
[23] Wadel K, Neher E, Sakaba T. The coupling between synaptic vesicles and Ca
2+
channels determines fast neurotransmitter release. Neuron 2007;53(4):563–75.
[24] Wheeler DG, Groth RD, Ma H, Barrett CF, Owen SF, et al. Ca(V)
1
and Ca(V)
2
channels engage distinct modes of Ca(2+) signaling to control CREB-dependent
gene expression. Cell 2012;149(5):1112–24.
[25] Stanika RI, Villanueva I, Kazanina G, Andrews SB, Pivovarova NB. Comparative
impact of voltage-gated calcium channels and NMDA receptors on
mitochondria-mediated neuronal injury. J Neurosci 2012;32(19):6642–50.
[26] Simms BA, Zamponi GW. Trafficking and stability of voltage-gated calcium
channels. Cell Mol Life Sci 2012;69(6):843–56.
[27] Park J, Luo ZD. Calcium channel functions in pain processing. Channels. 2010;4
(6):510–7.
[28] Catterall WA. Structure and regulation of voltage-gated Ca
2+
channels. Annu
Rev Cell Dev Biol 2000;16(1):521–55.
[29] Hofmann F, Biel M, Flockerzi V. Molecular basis for Ca
2+
channel diversity.
Annu Rev Neurosci 1994;17(1):399–418.
[30] Flynn R, Zamponi G. Calcium channel regulation by RGK proteins. Channels.
2010 Nov;4(6):434–9.
[31] Hidalgo P, Neely A. Multiplicity of protein interactions and functions of the
voltage-gated calcium channel b-subunit. Cell Calcium 2007;42(4–5):389–96.
[32] Karunasekara Y, Dulhunty AF, Casarotto MG. The voltage-gated calcium-
channel bsubunit: more than just an accessory. Eur Biophys J 2009 Dec 1;39
(1):75–81.
[33] Carstens BB, Clark JR, Daly LN, Harvey JP, Kaas Q, et al. Engineering of
conotoxins for the treatment of pain. Curr Pharm Des 2011;17(38):4242–53.
[34] Fry BG, Roelants K, Champagne DE, Scheib H, Tyndall JD, et al. The
toxicogenomic multiverse: convergent recruitment of proteins into animal
venoms. Annu Rev Genomics Hum Genet 2009;22(10):483–511.
[35] Lewis RJ, Garcia ML. Therapeutic potential of venom peptides. Nat Rev Drug
Discov 2003;2(10):790–802.
[36] Tedford HW, Sollod BL, Maggio F, King GF. Australian funnel-web spiders:
master insecticide chemists. Toxicon 2004;43(5):601–18.
[37] Olivera BM. Conus snail venom peptides. In: Handbook of Biologically Active
peptides. Academic Press; 2006. p. 381–8.
[38] Baell JB, Duggan PJ, Forsyth SA, Lewis RJ, Lok YP, et al. Synthesis and biological
evaluation of nonpeptide mimetics of
x
-conotoxin GVIA. Bioorg Med Chem
2004;12(15):4025–37.
[39] Nielsen KJ, Schroeder T, Lewis R. Structure–activity relationships of
x
-
conotoxins at N-type voltage-sensitive calcium channels. J Mol Recognit
2000;13(2):55–70.
[40] Tranberg CE, Yang A, Vetter I, McArthur JR, Baell JB, et al.
x
-Conotoxin GVIA
mimetics that bind and inhibit neuronal Cav2. 2 ion channels. Mar Drugs 2012
Oct;10(10):2349–68.
[41] Menzler S, Bikker JA, Suman-Chauhan N, Horwell DC. Design and biological
evaluation of non-peptide analogues of omega-conotoxin MVIIA. Bioorg Med
Chem Lett 2000;10(4):345–7.
[42] Baell JB, Duggan PJ, Lok YP.
x
-Conotoxins and approaches to their non-peptide
mimetics. Aust J Chem 2004;57(3):179–85.
[43] Baell JB, Duggan PJ, Forsyth SA, Lewis RJ, Lok YP, et al. Synthesis and biological
evaluation of anthranilamide-based non-peptide mimetics of
x
-conotoxin
GVIA. Tetrahedron 2006;62(31):7284–92.
[44] Andersson A, Baell JB, Duggan PJ, Graham JE, Lewis RJ, et al.
x
-Conotoxin GVIA
mimetics based on an anthranilamide core: Effect of variation in ammonium
side chain lengths and incorporation of fluorine. Bioorg Med Chem 2009;17
(18):6659–70.
[45] Wu J, Yan Z, Li Z, Qian X, Lu S, et al. Structure of the voltage-gated calcium
channel Ca v 1.1 at 3.6 Å resolution. Nature 2016;537(7619):191–6.
[46] Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, et al. UCSF
Chimera—a visualization system for exploratory research and analysis. J
Comput Chem 2004;25(13):1605–12.
[47] Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot.
Acta Crystallogr D Biol Crystallogr 2010;66(4):486–501.
[48] Duggan PJ, Lewis RJ, Lok YP, Lumsden NG, Tuck KL, et al. Low molecular weight
non-peptide mimics of
x
-conotoxin GVIA. Bioorg Med Chem Lett 2009;19
(10):2763–5.
[49] Mills N. ChemDraw Ultra 10.0 CambridgeSoft, 100 CambridgePark Drive,
Cambridge, MA 02140. www.cambridgesoft.com.
[50] Hanwell MD, Curtis DE, Lonie DC, Vandermeersch T, Zurek E, et al. Avogadro:
an advanced semantic chemical editor, visualization, and analysis platform. J
Cheminf 2012;4(1):17.
[51] Halgren TA. Merck molecular force field. I. Basis, form, scope,
parameterization, and performance of MMFF94. J Comput Chem 1996;17(5–
6):490–519.
[52] Schneidman-Duhovny D, Inbar Y, Nussinov R, Wolfson HJ. PatchDock and
SymmDock: servers for rigid and symmetric docking. Nucleic Acids Res
2005;33(suppl_2):W363–7.
[53] Wallace AC, Laskowski RA, Thornton JM. LIGPLOT: a program to generate
schematic diagrams of protein-ligand interactions. Protein Eng Des Sel 1995;8
(2):127–34.
[54] BIOVIA DS. BIOVIA Discovery Studio 2017 R2: A comprehensive predictive
science application for the Life Sciences. San Diego, CA, USA http://accelrys.
com/products/collaborative-science/biovia-discovery-studio. 2017.
[55] Humphrey W, Dalke A, Schulten K. VMD: visual molecular dynamics. J Mol
Graph 1996;14(1):33–8.
[56] Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, et al. Scalable
molecular dynamics with NAMD. J Comput Chem 2005;26(16):1781–802.
[57] Grubmüller H. SOLVATE v. 1.0. Theoretical Biophysics Group. Institute for
Medical Optics, Ludwig-Maximilians University, Munich. 1996.
[58] MacKerell Jr AD, Bashford D, Bellott ML, Dunbrack Jr RL, Evanseck JD, et al. All-
atom empirical potential for molecular modeling and dynamics studies of
proteins. J Phys Chem B 1998;102(18):3586–616.
[59] Zoete V, Cuendet MA, Grosdidier A, Michielin O. SwissParam: a fast force field
generation tool for small organic molecules. J Comput Chem 2011;32
(11):2359–68.
[60] Wu J, Yan Z, Li Z, Yan C, Lu S, et al. Structure of the voltage-gated calcium
channel Cav1. 1 complex. Science. 2015 Dec 18;350(6267):aad2395.
[61] Finn BE, Evenäs J, Drakenberg T, Waltho JP, Thulin E, et al. Calcium-induced
structural changes and domain autonomy in calmodulin. Nat Struct Biol
1995;2(9):777–83.
[62] Duggan PJ, Tuck KL. Bioactive mimetics of conotoxins and other venom
peptides. Toxins. 2015;7(10):4175–98.
[63] Bezanilla F. The voltage sensor in voltage-dependent ion channels. Physiol Rev
2000;80(2):555–92.
[64] Chen R, Chung SH. Complex structures between the N-type calcium channel
(CaV2.2) and
x
-conotoxin GVIA predicted via molecular dynamics.
Biochemistry 2013;52(21):3765–72.
[65] Hering S, Zangerl-Plessl EM, Beyl S, Hohaus A, Andranovits S, et al. Calcium
channel gating. Pflügers Archiv-European Journal of Physiology. 2018;470
(9):1291–309.
Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372 2371
Author's Personal Copy

[66] Beyl S, Depil K, Hohaus A, Stary-Weinzinger A, Linder T, et al. Neutralisation of
a single voltage sensor affects gating determinants in all four pore-forming S6
segments of Ca V 1.2: a cooperative gating model. Pflügers Archiv-Eur J Physiol
2012;464(4):391–401.
[67] Zhang X, Ren W, DeCaen P, Yan C, Tao X, et al. Crystal structure of an
orthologue of the NaChBac voltage-gated sodium channel. Nature 2012;486
(7401):130–4.
[68] Stevens EB, Stephens GJ. Recent advances in targeting ion channels to treat
chronic pain. Br J Pharmacol 2018;175(12):2133.
[69] Nieto-Rostro M, Ramgoolam K, Pratt WS, Kulik A, Dolphin AC. Ablation of
a
2d-
1 inhibits cell-surface trafficking of endogenous N-type calcium channels in
the pain pathway in vivo. Proc Natl Acad Sci 2018 Dec 18;115(51):E12043–52.
[70] Menzler S, Bikker JA, Horwell DC. Synthesis of a non-peptide analogue of
omega-conotoxin MVIIA. Tetrahedron Lett 1998;39(41):7619–22.
[71] Sheng ZH, Rettig J, Takahashi M, Catterall WA. Identification of a syntaxin-
binding site on N-type calcium channels. Neuron 1994;13(6):1303–13.
[72] Kiyonaka S, Wakamori M, Miki T, Uriu Y, Nonaka M, et al. RIM1 confers
sustained activity and neurotransmitter vesicle anchoring to presynaptic Ca 2+
channels. Nat Neurosci 2007;10(6):691–701.
[73] Han Y, Kaeser PS, Südhof TC, Schneggenburger R. RIM determines Ca
2+
channel
density and vesicle docking at the presynaptic active zone. Neuron 2011;69
(2):304–16.
[74] Kaneko S, Cooper CB, Nishioka N, Yamasaki H, Suzuki A, et al. Identification
and characterization of novel human Cav2. 2 (
a
1B) calcium channel variants
lacking the synaptic protein interaction site. J Neurosci 2002;22(1):82–92.
[75] Qin N, Olcese R, Bransby M, Lin T, Birnbaumer L. Ca
2+
-induced inhibition of the
cardiac Ca
2+
channel depends on calmodulin. Proc Natl Acad Sci 1999;96
(5):2435–8.
[76] Pitt GS, Zühlke RD, Hudmon A, Schulman H, Reuter H, et al. Molecular basis of
calmodulin tethering and Ca
2+
-dependent inactivation of L-type Ca
2+
channels.
J Biol Chem 2001;276(33):30794–802.
[77] Zhou H, Yu K, McCoy KL, Lee A. Molecular mechanism for divergent regulation
of Cav1. 2 Ca
2+
channels by calmodulin and Ca
2+
-binding protein-1. J Biol
Chem 2005;280(33):29612–9.
[78] Cui G, Meyer AC, Calin-Jageman I, Neef J, Haeseleer F, et al. Ca
2+
-binding
proteins tune Ca
2+
-feedback to Cav1. 3 channels in mouse auditory hair cells. J
Physiol 2007;585(3):791–803.
[79] Mori MX, Vander Kooi CW, Leahy DJ, Yue DT. Crystal structure of the CaV2 IQ
domain in complex with Ca
2+
/calmodulin: high-resolution mechanistic
implications for channel regulation by Ca
2+
. Structure 2008;16(4):607–20.
[80] Yonkunas M, Kurnikova M. The hydrophobic effect contributes to the closed
state of a simplified ion channel through a conserved hydrophobic patch at the
pore-helix crossing. Front Pharmacol 2015;27(6):284.
[81] McCoy JG, Nimigean CM. Structural correlates of selectivity and inactivation in
potassium channels. Biochim Biophys Acta (BBA)-Biomembr 2012;1818(2):272–85.
[82] Domene C, Doyle DA, Vénien-Bryan C. Modeling of an ion channel in its open
conformation. Biophys J 2005;89(1):L01–L3.
[83] Schroeder CI, Smythe ML, Lewis RJ. Development of small molecules that
mimic the binding of
x
-conotoxins at the N-type voltage-gated calcium
channel. Mol Diversity 2004;8(2):127–34.
[84] Baell JB, Duggan PJ, Forsyth SA, Lewis RJ, Lok YP, Schroeder CI, et al. Synthesis
and biological evaluation of anthranilamide-based non-peptide mimetics of
x
-conotoxin GVIA. Tetrahedron 2006;62(31):7284–92.
[85] Gleeson EC, Graham JE, Spiller S, Vetter I, Lewis RJ, Duggan PJ, et al. Inhibition
of N-type calcium channels by fluorophenoxyanilide derivatives. Mar Drugs
2015;13(4):2030–45.
2372 Sameera et al. / Computational and Structural Biotechnology Journal 18 (2020) 2357–2372
Author's Personal Copy