International Journal of Molecular Sciences

Review Pharmacological Approaches for the Modulation of the KV4.x and KChIPs

Pilar Cercós 1, Diego A. Peraza 2,3 , Angela de Benito-Bueno 2,3 , Paula G. Socuéllamos 2,3, Abdoul Aziz-Nignan 4, Dariel Arrechaga-Estévez 4, Escarle Beato 4, Emilio Peña-Acevedo 4, Armando Albert 5 , Juan A. González-Vera 6 , Yoel Rodríguez 4 , Mercedes Martín-Martínez 1 , Carmen Valenzuela 2,3,* and Marta Gutiérrez-Rodríguez 1,*

1 Instituto de Química Médica (IQM-CSIC), 28006 Madrid, Spain; [email protected] (P.C.); [email protected] (M.M.-M.) 2 Instituto de Investigaciones Biomédicas Alberto Sols (IIBM), CSIC-UAM, 28029 Madrid, Spain; [email protected] (D.A.P.); [email protected] (A.d.B.-B.); [email protected] (P.G.S.) 3 Spanish Network for Biomedical Research in Cardiovascular Research (CIBERCV), Instituto de Salud Carlos III, 28029 Madrid, Spain 4 Department of Natural Sciences, Hostos Community College of CUNY, New York, NY 10451, USA; [email protected] (A.A.-N.); [email protected] (D.A.-E.); [email protected] (E.B.); [email protected] (E.P.-A.); [email protected] (Y.R.) 5 Instituto de Química Física Rocasolano (IQFR-CSIC), 28006 Madrid, Spain; [email protected] 6 Departamento de Físicoquímica, Facultad de Farmacia, Universidad de Granada, 18071 Granada, Spain; [email protected] * Correspondence: [email protected]; (C.V.); [email protected] (M.G.-R.);  Tel.: +34-91-585-4493 (C.V.); +34-91-258-7493 (M.-G.R.) 

Citation: Cercós, P.; Peraza, D.A.; Abstract: Ion channels are macromolecular complexes present in the plasma membrane and in- Benito-Bueno, A.d.; Socuéllamos, P.G.; tracellular organelles of cells. Dysfunction of ion channels results in a group of disorders named Aziz-Nignan, A.; Arrechaga-Estévez, channelopathies, which represent an extraordinary challenge for study and treatment. In this review, D.; Beato, E.; Peña-Acevedo, E.; we will focus on voltage-gated potassium channels (KV), specifically on the KV4-family. The activa- Albert, A.; González-Vera, J.A.; et al. tion of these channels generates outward currents operating at subthreshold membrane potentials as Pharmacological Approaches for the recorded from myocardial cells (ITO, transient outward current) and from the somata of hippocampal Modulation of the Potassium Channel neurons (ISA). In the heart, KV4 dysfunctions are related to Brugada syndrome, atrial fibrillation, KV4.x and KChIPs. Int. J. Mol. Sci. hypertrophy, and heart failure. In hippocampus, K 4.x channelopathies are linked to schizophrenia, 2021, 22, 1419. https://doi.org/ V 10.3390/ijms22031419 epilepsy, and Alzheimer’s disease. KV4.x channels need to assemble with other accessory subunits (β) to fully reproduce the ITO and ISA currents. β Subunits affect channel gating and/or the traffic to Academic Editor: the plasma membrane, and their dysfunctions may influence channel pharmacology. Among KV4 Antonio Ferrer-Montiel regulatory subunits, this review aims to analyze the KV4/KChIPs interaction and the effect of small Received: 16 January 2021 molecule KChIP ligands in the A-type currents generated by the modulation of the KV4/KChIP Accepted: 28 January 2021 channel complex. Knowledge gained from structural and functional studies using activators or Published: 31 January 2021 inhibitors of the potassium current mediated by KV4/KChIPs will better help understand the under- lying mechanism involving KV4-mediated-channelopathies, establishing the foundations for drug Publisher’s Note: MDPI stays neutral discovery, and hence their treatments. with regard to jurisdictional claims in published maps and institutional affil- Keywords: voltage-gated potassium channels KV4; transient outward current; A-type current; potas- iations. sium channel interacting (KChIPs); –protein interactions; KV4/KChIPs modulators

Copyright: © 2021 by the authors. 1. Introduction Licensee MDPI, Basel, Switzerland. Ion channels are macromolecular complexes present in the plasma membrane and in- This article is an open access article tracellular organelles of all cells. They play an important role in the maintenance of cellular distributed under the terms and integrity, smooth muscle contraction, secretion of hormones, and neurotransmitters, among conditions of the Creative Commons Attribution (CC BY) license (https:// others. Dysfunction of ion channels results in a group of disorders named channelopathies, creativecommons.org/licenses/by/ which represent an extraordinary challenge for study and treatment [1]. 4.0/).

Int. J. Mol. Sci. 2021, 22, 1419. https://doi.org/10.3390/ijms22031419 https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 2 of 21

cellular integrity, smooth muscle contraction, secretion of hormones, and neurotransmit- Int. J. Mol. Sci. 2021,ters,22, 1419 among others. Dysfunction of ion channels results in a group of disorders named 2 of 21 channelopathies, which represent an extraordinary challenge for study and treatment [1]. In this review, we will focus on voltage-gated potassium channels (KV), specifically on the KV4-family or Shal-related subfamily. Four members compose this family: KV4.1 In this review, we will focus on voltage-gated potassium channels (KV), specifically (KCND1), KV4.2 (KCND2), and two splice variants of KV4.3 (KCND3), which have an over- on the KV4-family or Shal-related subfamily. Four members compose this family: KV4.1 all sequence identity of 60% [2,3]. As with other eukaryotic KV channels, KV4.x are sym- (KCND1), K 4.2 (KCND2), and two splice variants of K 4.3 (KCND3), which have an metrically assembled as homoV - or heterotetramers of pore-forming subunitsV (α subunits) over-all sequence identity of 60% [2,3]. As with other eukaryotic K channels, K 4.x around a central pore [4,5]. Each of these α subunits have six transmembrane helices V(S1– V are symmetrically assembled as homo- or heterotetramers of pore-forming subunits (α S6), with its N- and C-terminus pointing to the cytoplasm. At the N-terminus, the princi- subunits) around a central pore [4,5]. Each of these α subunits have six transmembrane pal motif is the so-called T1 domain, which governs the specificity of assembly and is helices (S1–S6), with its N- and C-terminus pointing to the cytoplasm. At the N-terminus, involved in channel tetramerization and its gating. The first four helices (S1–S4) form the the principal motif is the so-called T1 domain, which governs the specificity of assembly voltage-sensing domain, whereas S5, S6, and the P-loop that link them, form the pore do- and is involved in channel tetramerization and its gating. The first four helices (S1–S4) main. There is homogeneity within the KV4.x transmembrane region, except in the S6- form the voltage-sensing domain, whereas S5, S6, and the P-loop that link them, form the proximal cytosolic C-terminus, where KV4.3 isoform 1 has an additional exon that encodes pore domain. There is homogeneity within the KV4.x transmembrane region, except in 19 amino acids (residues 488–506) (Figure 1). Owing to of this addi- the S6-proximal cytosolic C-terminus, where KV4.3 isoform 1 has an additional exon that tional exon, KVencodes4.3 has two 19 amino variants acids in human (residues tissues: 488–506) KV4.3L (Figure (long1). form) Owing and to K alternativeV4.3S (short splicing of this form). Both splice variants are identical except for the additional sequence of 19 amino additional exon, KV4.3 has two variants in human tissues: KV4.3L (long form) and KV4.3S acids present in(short the long form). form Both KV4.3L splice [6,7]. variants are identical except for the additional sequence of 19 KV4 channels are highly expressed in smooth muscles, heart, and brain. The activa- amino acids present in the long form KV4.3L [6,7]. tion of these channels generates outward currents operating at subthreshold membrane KV4 channels are highly expressed in smooth muscles, heart, and brain. The activa- potentials, as tionrecorded of these from channels myocardial generates cells (I outwardTO, transient currents outward operating current) at subthreshold and from membrane the somata of hippocampal neurons (ISA, somato-dendritic subthreshold-activating A-type potentials, as recorded from myocardial cells (ITO, transient outward current) and from the + V V V + K current) [3,8,9]somata. Among of hippocampal K 4 channels, neurons K 4.2 (IandSA, K somato-dendritic4.3 are the most subthreshold-activatingthoroughly studied A-type K members of thecurrent) family. [ 3Concerning,8,9]. Among the K Vmyocardium,4 channels, K KVV4.24.2 and and K KVV4.34.3 arepresent the most a comple- thoroughly studied mentary expression,members with of K theV4.2 family. being Concerningmore abundant the myocardium,in the ventricle K Vand4.2 andKV4.3 K Vin4.3 the present a com- atria [8,9]. In theplementary heart, KV4 expression, dysfunctions with are K involvedV4.2 being in moreBrugada abundant syndrome, in the atrial ventricle fibri- and KV4.3 in llation, hypertrophythe atria, and [8, 9heart]. In failure the heart, [10]. K V4 dysfunctions are involved in Brugada syndrome, atrial In the brain,fibri-llation, the KV4 channel hypertrophy, expression and heartdepends failure on the [10 ].neuron type and region, and continuous efforts areIn themade brain, to unravel the KV4 the channel organization expression of these depends channels on the in neuron different type sub- and region, and cellular compartmentscontinuous and efforts neuron are madetypes to [11,12]. unravel In the hippocampus, organization of K theseV4 channelopathies channels in different subcellu- are linked to schizophrenia,lar compartments epilepsy and neuron, and Alzheimer’s types [11,12]. disease In hippocampus, [13–15]. Hence, KV4 channelopathies the phar- are linked macological modulationto schizophrenia, of ITO and epilepsy, ISA may and have Alzheimer’s therapeutic disease value in [13 the–15 treatment]. Hence, of the these pharmacological pathologies. modulation of ITO and ISA may have therapeutic value in the treatment of these pathologies.

Figure 1. Alignment of human KV4. Alignment with Clustal within Jalview. Residues that are identical in the four sequences are boxed [16].

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 3 of 21

Int. J. Mol. Sci. 2021Figure, 22, 1419 1. Alignment of human KV4. Alignment with Clustal within Jalview. Residues that are 3 of 21 identical in the four sequences are boxed [16].

The α subunits of KV4 channels are sufficient for the formation of functional K+ chan- + nels. However, onlyThe the αexpressionsubunits of of K KVV44 α channelssubunits does are sufficient not fully reproduce for the formation the ITO and of functional K ISA currents [17].channels. KV4 need However, to assemble only with the expressionother accessory of K Vsubunits4 α subunits, forming does macromo- not fully reproduce the lecular complexesITO andwith I SAdifferentcurrents regulatory [17]. KV 4(or need β) subunits, to assemble including with other KChIPs accessory (potassium subunits, forming channel interactingmacromolecular proteins), DPPscomplexes (dipeptidyl with different-peptidase regulatory-like proteins), (or β) subunits, KCNEs (also including KChIPs termed MinK-related(potassium peptides, channel or MiRPs), interacting KChAPs proteins),, and DPPsKvβx subunits (dipeptidyl-peptidase-like [17–20]. Likewise, proteins), KC- these channelsNEs are (alsosusceptible termed to MinK-related be phosphorylated peptides, by or PKA, MiRPs), PKC, KChAPs, PI-4K-beta, and MAPKs, Kvβx subunits [17–20]. SRC kinases, andLikewise, so on these[21–23]. channels Moreover, are susceptibleit has been to described be phosphorylated that KV4.3 byand PKA, NaV1.5 PKC, PI-4K-beta, influence the functionMAPKs, of SRC each kinases, other and[24,25]; so onthe [21 existence–23]. Moreover, of a channelosome it has been describedformed by that KV4.3 and KV4.3, KChIP3Na, andV1.5 Ca influenceV3.1 channels the function has als ofo been each described other [24,25 [26,27].]; the existence The expression of a channelosome of β formed subunits variesby both KV4.3, among KChIP3, organs and and/or CaV3.1 among channels different has also regions been describedof the same [26 ,organ27]. The expression [3,12,28–33]. Therefore,of β subunits the function varies both of these among channels organs depends and/or among not only different on the regions tissue of the same where they areorgan expressed, [3,12, 28but–33 also]. Therefore,on the signaling the function context, ofinvolving these channels accessory depends subunits, not only on the kinases, or eventissue other where ion channels. they are These expressed, β subunits but alsoaffect on channel the signaling gating and/or context, the involving traf- accessory β fic to the plasmasubunits, membrane, kinases, and or their even dysfunctions other ion channels. may affect These channelsubunits pharmacology affect channel gating [20,34,35]. and/or the traffic to the plasma membrane, and their dysfunctions may affect channel pharmacology [20,34,35]. This review aims to analyze the KV4/KChIPs interaction and the effect of small mol- This review aims to analyze the KV4/KChIPs interaction and the effect of small ecule KChIP ligands in the A-type currents generated by the modulation of the KV4/KChIP molecule KChIP ligands in the A-type currents generated by the modulation of the channel complexes. Knowledge gained from structural and functional studies using acti- KV4/KChIP channel complexes. Knowledge gained from structural and functional studies vators or inhibitors of the potassium current mediated by KV4/KChIPs will contribute to using activators or inhibitors of the potassium current mediated by KV4/KChIPs will con- understanding the mechanisms involved in the KV4-mediated-channelopathies and to es- tribute to understanding the mechanisms involved in the K 4-mediated-channelopathies tablish the bases of drug discovery to treat them. V and to establish the bases of drug discovery to treat them.

2. Potassium Channel2. Potassium Interacting Channel Proteins Interacting (KChIPs) Proteins (KChIPs) KChIPs belongKChIPs to the calcium belong tobinding the calcium protein binding superfamily. protein In superfamily. mammals, the In mammals,KChIP the KChIP subfamily is composedsubfamily isof composed four members, of four members, KChIP1, KChIP1,KChIP2, KChIP2, KChIP3 KChIP3 (also known (also known as as DREAM DREAM or )or calsenilin),, and andKChIP4 KChIP4 (also (also known known as calsenilin as calsenilin-like-like protein), protein), encoded encoded by by four four genes (KC(NIP1KCNIP1-4-4), all), of all them of them with with several several spliced spliced variants variants [36 [36–39].–39 KChIP1,]. KChIP1, KChIP3 KChIP3,, and KChIP4 and KChIP4 areare brain brain-predominant,-predominant, while while KChIP2 KChIP2 is is predominantly predominantly expressed expressed in in both the heart and the heart and brain brain [ [40].40]. The The four four KChIP KChIP isoforms isoforms have have a variablea variable N-terminal N-terminal region, region where, the main where the maivariationsn variations in in the the sequence sequence as a wells well as as a conserveda conserved C-terminal C-terminal domain domain can can be observed, in be observed, inwhich which four four EF-hand-like EF-hand-like domains domains are are located located (Figure (Figure2)[ 2)41 ].[41]. Apart Apart from from KChIP’s activity KChIP’s activityon on the the Kv4 Kv4 modulation, modulation, they they also also show show a plethoraa plethora of of functions functions including including the modulation the modulationof of cardiac cardiac and and neuronal neuronal excitability, excitability, -binding presenilin-binding proteins, proteins and, and calcium-dependent cal- cium-dependenttranscriptional transcriptional mediators mediators [42 [42–44–].44].

Figure 2. Alignment of human KChIPs. Alignment with ClustalO within Jalview. Residues that are identical in the four sequences are boxed [16].

Int. J. Mol. Sci. 2021, 22, 1419 4 of 21

2.1. Three-Dimensional Structures of Free KChIPs As of today, few three-dimensional (3D) structures of KChIPs have been determined (PDB IDs 1S1E, 2E6W, 2JUL, 3DD4) using either nuclear magnetic resonance (NMR) or X-ray diffraction techniques. In 2004, one of the first KChIP structures was solved at 2.30 Å resolution by Scannevin et al., corresponding to KChIP1 (PDB ID 1S1E) [45]. Scannevin et al.’s study aimed to better understand the interaction of KChIP1 with KV4 subunits. The KChIP1 structure has mainly two regions; the N-terminal one comprising residues 12–95 and the C-terminal one comprising residues 96–192. The N-terminal region houses five α helices (H1 through H5), while the C-terminal houses four α helices (H6 through H9). As has been shown in other calcium binding proteins, KChIP1 exhibits the classical EF-hand, consisting of a helix-turn-helix motif [46–49]. The pairs of EF-hands are connected by linker loops in a U shape. Specifically, for KChIP1, the first and second EF-hands in the N-terminal region are formed by helices H2 and H3 and helices H4 and H5, respectively. In the C-terminal region of KChIP1, helices H6 and H7 and helices H8 and H9 form the third and fourth EF-hands, respectively (see Figure3A). H10 is not involved in any EF-hand motif and corresponds to the C-terminal helix and engages in a network of hydrophobic interactions with a crevice formed between Leu127, Thr130, Tyr134, and Ile154 of EF-hand 3 and Phe178, Met182, Phe195, and Cys199 of EF-hand 4 (see Figure3A). In addition, the X-ray crystal structure of KChIP1 showed only two bound Ca2+ ions located in EF-hands 3 and 4. EF-hands 1 and 2 lack the canonical sequence for Ca2+ coordination, explaining the lack of Ca2+ binding. A few years later, in 2007, the solution NMR structure of KChIP3 EF3-4 (PDB ID 2E6W) became available, consisting of five α helices and a short two- stranded antiparallel β-sheet [50]. EF-hand 3 is formed by α helix H1, β1-strand, and α helix H2, and EF-hand 4 is formed by α helix H3, β2-strand, and α helix H4. Two bound Ca2+ ions were also located in EF-hands 3 and 4. H5 is the C-terminal helix. Different from the KChIP1 structure (PDB ID 1S1E), the KChIP3 EF3-4 structure (PDB ID 2E6W) showed a disordered long loop connecting EF-hands 3 and 4, which, due to its flexibility, is thought to function as fingers during or other molecules’ binding and recognition. In 2008, Lusin et al. determined another solution NMR structure of KChIP3 (PDB ID 2JUL) [51]. This NMR KChIP3 structure unveiled a similar overall folding as KChIP1 (PDB ID 1S1E). Within the C-terminal region of KChIP3, there are two domains: one comprising EF-hand 1 (residues 90–119) and EF-hand 2 (residues 128–157); and the second one comprising EF-hand 3 (residues 163–192) and EF-hand 4 (residues 211–240). The two domains’ interface forms a cleft through the interactions between EF-hand 2 (Tyr130, Phe133, Leu134, and Ala137) and EF-hand 3 (Leu173, Ile190, and Met197) (see Figure3A). Likewise, the two bound Ca 2+ ions are located in EF-hands 3 and 4. In KChIP3, EF-1 is not functional, as its loop’s sequence is not the canonical one for Ca2+ binding, and EF-2 bounds Mg. The C-terminal helix (residues 243–254), positioned after EF-hand 4, engages in a network of intramolecular and hydrophobic interactions with a crevice formed between Leu167, Ala170, Met191, and Leu194 of EF-hand 3 and Phe218, Met222, and Cys239 of EF-hand 4. Later on, in 2009, the X-ray crystal structure of KChIP4a was solved at 3.0 Å resolution by Liang et al. (PDB ID 3DD4) [52]. The KChIP4a structure also revealed four EF-hand motifs: six α-helices (H0–H5) comprising the N-terminal region and five α-helices (H6–H10) comprising the C-terminal region (see Figure3B). Contrary to other KChIPs, the KChIP4a arrangement of the N-terminal α-helices (H0–H2) causes the α-helix H10 to swing outward at ~45◦ from the hydrophobic pocket, also presented in other KChIPs (see Figure3A,B ). Overall, KChIPs structures have in common the four EF-hand motifs and the two bound Ca2+ ions located in EF-hands 3 and 4, but could differ in the arrangement of α-helix H10 with respect to the hydrophobic pocket. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 5 of 21 Int. J. Mol. Sci. 2021, 22, 1419 5 of 21

Figure 3. Three-dimensional structures of free KChIPs. The four EF-hand motifs are shown, as rep- Figure 3. Three-dimensional structures of free KChIPs. The four EF-hand motifs are shown, as resentative examples: (A) KChIP3 (PDB ID 2JUL) [51] and (B) KChIP4a (PDB ID 3DD4) [52]. Differ- entrepresentative from KChIP1- examples:3, KChIP4a C (A-terminal) KChIP3 α-helix (PDB H10 ID (red) 2JUL) swings [51] outward and (B) at KChIP4a ~45° from (PDB the hydro- ID 3DD4) [52]. phobicDifferent pocket. from Ca KChIP1-3,2+ ions are shown KChIP4a as yellow C-terminal spheres.α -helix H10 (red) swings outward at ~45◦ from the hydrophobic pocket. Ca2+ ions are shown as yellow spheres. 2.2. Three-Dimensional Structures of KChIPs-KV4.2/KV4.3 Channels

2.2.To Three-Dimensional better understand Structures the structural of KChIPs-K basis of Vthe4.2/K recognitionV4.3 Channels and modulation of KV4 by KChIPs,To better few X understand-ray crystal structure the structurals of the basis KChIP1 of- theKV4 recognition complex have and been modulation determined of KV4 by (PDBKChIPs, IDs 1S6C, few X-ray 2I2R, 2NZO) crystal. structuresIn 2004, the offirst the X- KChIP1-Kray crystal structureV4 complex of the have core been domain determined of(PDB KChIP1(residues IDs 1S6C, 2I2R, 34–216), 2NZO). lacking In the 2004, N-terminal the first variable X-ray crystal domain, structure in complex of thewith core N- domain terminalof KChIP1(residues fragment of homoK 34–216),V4.2 lacking(Kv4.2N30, the residues N-terminal 1–30) variable was solved domain, at 2.0 Å in resolution complex with N- (PDB ID 1S6C) [53]. This complex is a fusion construct made of rat (Rattus norvegicus) terminal fragment of homoKV4.2 (Kv4.2N30, residues 1–30) was solved at 2.0 Å resolution KChIP1 (residues 34–216) to the KV4.2N30, such that the first residue Met1 of KV4.2N30 (PDB ID 1S6C) [53]. This complex is a fusion construct made of rat (Rattus norvegicus) was linked directly to the last residue Met216 of KChIP1 (residues 34–216). This complex exhibitsKChIP1 a (residuesclam-shaped 34–216) dimeric to the assembly. KV4.2N30, Four such EF-hands that formthe first a shell residue-shaped Met1 confor- of KV4.2N30 mationwas linked and create directly a hydrophobic to the last residuegroove inside Met216 the of shell. KChIP1 This structure (residues also 34–216). shows Thisthat complex theexhibits first 20 a clam-shapedresidues of KV dimeric4.2 form assembly.an α-helix ( Fourα1), which EF-hands is kinked form around a shell-shaped Pro10, a conformationcon- servedand create residue a hydrophobic among KV4 channels. groove α inside1 is connected the shell. approximately This structure collinearly also shows with thatthe the first C20-terminal residues α-helix of K H10V4.2 of form KChIP1 an inα-helix head-to (α-tail1), fashion, which running is kinked from around the EF- Pro10,hand 4 end a conserved ofresidue the shell among to the EF KV-hand4 channels. 1 end (seeα Figure1 is connected 4A). By around approximately the same time collinearly the previous with the C- strterminalucture (PDBα-helix ID 1S6C) H10 ofwas KChIP1 reported in in head-to-tail 2004, a second fashion, structure running of KChIP2 from in the complex EF-hand 4 end withof the a variant shell to of the KV4.2 EF-hand (KV4.2*) 1 was end reconstructed (see Figure4 A).at 21 By Å aroundusing negative the same stain time electron the previous microscopystructure (PDB and single ID 1S6C) particle was averaging reported [54]. in 2004, Single a secondpurified structureKChIP2-K ofV4.2* KChIP2 channel ins complex were visualized in this study. The structure of these channels unveiled that KChIP2 does with a variant of KV4.2 (KV4.2*) was reconstructed at 21 Å using negative stain electron not interact directly with the membrane domain of the channel to affect function. How- microscopy and single particle averaging [54]. Single purified KChIP2-K 4.2* channels ever, the results indicated that KChIP2 acts through the four internal and four externalV were visualized in this study. The structure of these channels unveiled that KChIP2 does columns of the KV4.2*-KChIP2 channels [54]. not interact directly with the membrane domain of the channel to affect function. However, the results indicated that KChIP2 acts through the four internal and four external columns of the KV4.2*-KChIP2 channels [54].

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 7 of 21

helix of Kv4.2, as shown in the complex structure of KChIP1 (residues 34–216) to the KV4.2N30 (PDB ID 1S6C), in which the first residue Met1 of KV4.2N30 was linked directly to the last residue Met216 of KChIP1 (residues 34–216) [53]; alternatively, it could move Int. J. Mol. Sci. 2021, 22, 1419 6 of 21 outward from the hydrophobic pocket, being replaced by α1 helix of KV4.3, as shown in the KChIP1-KV4.3 structures (PDB IDs 2I2R and 2NZO) [55,58].

Figure 4. Three-dimensional structures of KChIPs–KV4 complexes. (A) KChIP1(34–216)–KV4.2N30 Figurecomplex 4. structureThree-dimensional (PDB ID 1S6C) structures showing ofthe KChIPs–K four EF-handV4 motifs complexes. (helices (A H2) KChIP1(34–216)–K–H9) and the central V4.2N30 complexhydrophobic structure pocket (PDB occupied ID 1S6C) by H10 showing (red) and the K fourV4.2N30 EF-hand (light blue) motifs helices (helices [53]. H2–H9) (B) Leftand panel: the central hydrophobicOverall structure pocket of the occupied KChIP1– byKV4.3 H10 T1 (red) complex and as K Vseen4.2N30 from (lightthe cytosolic blue) helices side (PDB [53 ].ID(B 2I2R)) Left panel: [55]. KChIP1s are shown in dark blue and KV4.3 members are shown in white. Ca2+ ions are shown Overall structure of the KChIP1–KV4.3 T1 complex as seen from the cytosolic side (PDB ID 2I2R) [55]. as yellow spheres and Zn2+ ions are shown in orange. Boxes represent interaction site 1 and interac- KChIP1s are shown in dark blue and K 4.3 members are shown in white. Ca2+ ions are shown as tion site 2. Top right panel: Interaction siteV 1 showing KChIP1 in dark blue and KV4.3 T1N helix, 2+ yellowT1N linker spheres, and and T1 domain Zn ions in white are shown. Ca2+ ions in orange. are shown Boxes as representyellow spheres. interaction Bottom site right 1 and panel interaction: siteInteraction 2. Top site right 2 showing panel: the Interaction main interface site 1KChIP1(34 showing–216) KChIP1-KV4.2N30 in dark residues. blue KChIP1 and K Vin4.3 dark T1N helix, T1Nblue linker,and KV4.2N30 and T1 residues domain in in white. white. Ca2+ ions are shown as yellow spheres. Bottom right panel: Interaction site 2 showing the main interface KChIP1(34–216)-K 4.2N30 residues. KChIP1 in dark 3. KChIP Ligands V blue and KV4.2N30 residues in white. After the assembly of KV4.x channels with KChIP subunits, these regulatory subunits induce an increase in the traffic of KV4.x channels to the plasma membrane, a delay in the Later on, in 2006, a third structure of the human KChIP1-rat KV4.3 T1 complex was macroscopic inactivation kinetics, and an acceleration of both the activation and the re- solved with a 3.35 Å resolution (PDB ID 2I2R) [55]. KV4.3 T1 comprises the N-terminal covery kinetics from inactivation of KV4.x channels (Figure 5) [3,36]. cytoplasmic domain (residues 1–143), while KChIP1 encompasses residues 37–216 contain- ing the core conserved functional of KChIPs. Different from the X-ray crystal structure described above (PDB ID 1S6C), this structure revealed that the assembly between KChIP1 and KV4.3 T1 forms a cross-shaped octamer involving the T1 tetramer at the center (see Figure4B left panel) and individual KChIP1s extending radially (see Figure4B left panel). In addition, this latter structure (PDB ID 2I2R) displays two main interaction sites. Inter- 2 action site 1, with a surface area of ~2100 Å , is located between residues 3 and 21 of a conserved hydrophobic segment on the N-terminal side of T1 (called T1N; Figure4B left and top right panels) and a large hydrophobic pocket formed by KChIP1. Site 1, different from the complex described above (PDB ID 1S6C), is also formed by a large conforma- tional change in helix H10, which rotates toward the edge of the hydrophobic pocket (see Figure4B top right panel). T1N is comprised of the following: the T1N helix; an α-helix comprising residues 4–17; and the T1N linker, a loop containing residues 18–39. Both portions contain the essential residues for recognition of KV4 by KChIP [36,56]. Conserved T1N residues 7–21 engage in most of the interactions with KChIP1 hydrophobic pocket. Int. J. Mol. Sci. 2021, 22, 1419 7 of 21

KV4.3 T1N residues Trp8, Phe11, and Trp19, which are important for KChIP-KV4 recog- nition and current modulation, fit into hydrophobic crevices within the interaction site 1 pocket [53,57]. Trp8 in T1N interacts with Phe60, Val69, Phe74, Ile77, and Phe111 in KChIP1. Phe11 in T1N interacts with Phe98, Phe111, Ala114, Leu115, Leu118, and Trp129 in KChIP1. Trp19 in T1N interacts with Tyr134, Ile150, Ile154, and Tyr155 as well as T1N Met20 in KChIP1. On the other hand, interaction site 2, with a surface area of ~900 Å2, encompasses the interaction between the KChIP1 helix H2 and the KV4.3 T1 α2-helix and the KV4.3 T1 domain “docking loop” (Glu70, Phe73 and Asp78) on the cytoplasmic face of the T1 complex. Phe73 in KV4.3 is engaged in an aromatic stacking interaction with Tyr57 in KChIP1. In addition, the side chain of Asp78 in KV4.3 forms a salt bridge with Arg51 of KChIP1, while backbone atoms of Glu72 and Phe75 appear to engage in hydrogen bonding with Gln54 in KChIP1. Each KChIP1 interacts with T1N from one KV4.3 via interaction site 1, however, KChIP1 also interacts with the adjacent T1 KV4.3 monomer via interaction site 2 (see Figure4B left and bottom right panels). Biochemical and functional studies revealed that site 1 is involved in channel trafficking to plasma membrane, whereas site 2 is involved in channel gating. Afterwards, in 2007, a similar X-ray crystal structure to PDB ID 2I2R of the complex of human KChIP1 and KV4.3 N-terminus was solved at a resolution of 3.2 Å (PDB ID 2NZO) [58]. This structure shares similar characteristics to the one described above (PDB ID 2I2R) [55]. The KChIP1-KV4.3 complex exhibits the above two interfaces’ interaction, site 1 and site 2. In site 1, the KV4.3 N-terminal peptide binds to the KChIP1 deep hydrophobic pocket, replacing helix H10 of KChIP1, which modified the inactivation kinetics of KV4. As this binding occurs, a notable conformation change in helix H10 of KChIP1 takes place. Helix H10 swings outward ~40◦ from the hydrophobic groove. In addition, helices H2 and H8 translate outward by ~2.5 Å and ~1.5 Å, respectively, facilitating the configuration of the peptide binding pocket on the KChIP1 surface. In site 2, KChIP1 interacts with a neighboring KV4.3N, different from the one in site 1, via KChIP1 helix H2 and residues 70–78 of KV4.3. Phe73, conserved among the KV4 members, but not among the other KV channels, fits tightly into a hydrophobic cavity formed by Leu39, Leu42, Leu53, Tyr57, and Phe108 in KChIP1. Additionally, salt bridges seem to be formed between Glu77 and Asp78 of Kv4.3N and Lys50 and Arg51 of KChIP1, respectively. Another conserved residue among the KV4 members, Glu70, is engaged via a salt bridge with Lys61 in KChIP1. As Lys61 is conserved among KChIPs, but not in other EF-hand proteins including frequenin and , which suggests that salt bridges could play important roles in the recognition of KV4 by KChIPs (PDB ID 2NZO) [58]. Taken together, the KChIP1-KV4.3 structure reveals a clamping mode of the complex, where a single KChIP1monomer sideways clamps two neighboring KV4.3 N-termini [55,58]. In addition, the α-helix H10 of KChIPs could remain in the hydrophobic pocket and connect with α1 helix of Kv4.2, as shown in the complex structure of KChIP1 (residues 34–216) to the KV4.2N30 (PDB ID 1S6C), in which the first residue Met1 of KV4.2N30 was linked directly to the last residue Met216 of KChIP1 (residues 34–216) [53]; alternatively, it could move outward from the hydrophobic pocket, being replaced by α1 helix of KV4.3, as shown in the KChIP1-KV4.3 structures (PDB IDs 2I2R and 2NZO) [55,58].

3. KChIP Ligands

After the assembly of KV4.x channels with KChIP subunits, these regulatory subunits induce an increase in the traffic of KV4.x channels to the plasma membrane, a delay in the macroscopic inactivation kinetics, and an acceleration of both the activation and the recovery kinetics from inactivation of KV4.x channels (Figure5)[3,36]. Int. J. Mol. Sci. 2021, 22, 1419 8 of 21 Int. J. Mol. Sci. 2021Int.Int.Int., 22 J. J.J. ,Mol. Mol. Mol.x FOR Sci. Sci.Sci. PEER 2021 20212021, , REVIEW, 22 2222,, , x xx FOR FORFOR PEER PEERPEER REVIEW REVIEWREVIEW 8 of 21 888 of ofof 21 2121

Figure 5. Original KV4.3 (A) and KV4.3+KChIP3 (B) records generated after applying the pulse pro- Figure 5. OriginalFigureFigure KV4.3 5.5. (OriginalOriginalA) and K KKVV4.3+KChIP3V4.34.3 ((AA)) andand (KKBVV)4.3+KChIP3 4.3+KChIP3records generated ((BB)) recordsrecords after generatedgeneratedapplying the afterafter pulse applyingapplying pro- thethe pulsepulse pro-pro- Figure 5. A B Original KV4.3 ( ) andtocol K shownV4.3+KChIP3 in the upper ( ) records part of generated the Figure. after Note applying that K theV4.3+KChIP3 pulse protocol channels shown exhibit in the a upper slower tocol shown in tocolthetocol upper shownshown part inin ofthethe the upperupper Figure. partpart Note ofof thethe that Figure.Figure. KV4.3+KChIP3 NoteNote thatthat channelsKKVV4.3+KChIP34.3+KChIP3 exhibit channelschannels a slower exhibitexhibit aa slowerslower part of the Figure. Note thatinactivat KV4.3+KChIP3ion rate channels than KV4.3 exhibit ones a. slower inactivation rate than KV4.3 ones. inactivation rateinactivatinactivat than KVionion4.3 rateonesrate thanthan . KKVV4.34.3 onesones .. KChIP ligands modify this general trend, and the knowledge gained from the modula- KChIP ligandsKChIPKChIPKChIP modify ligands ligandsligands this general modify modifymodify trend, this thisthis general generalgeneral and the trend, trend,trend, knowledge and andand the thethe gained knowledge knowledgeknowledge from the gained gainedgained mod- from fromfrom the thethe mod- mod-mod- tion of the K 4.x/KChIP complex by small molecules could open new avenues to modulate V V ulationulationulationV of ofof the thethe K KKVV4.x/KChIP4.x/KChIP4.x/KChIP complex complexcomplex by byby small smallsmall molecules moleculesmolecules could couldcould open openopen new newnew avenues avenuesavenues to toto mod- mod-mod- ulation of thethe K potassium4.x/KChIP complex outward by current small through molecules Kv4 could channel open complexes. new avenues At present,to mod- eight KChIP ulate the potassiumulateulateulate thethethe outward potassiumpotassiumpotassium current outwardoutwardoutward through currentcurrentcurrent Kv4 throughchannelthroughthrough complexes.Kv4Kv4Kv4 channelchannelchannel A complexes.tcomplexes.complexes. present, eight AAAttt present,present, present, eighteighteight ligands have been described as KV4.x/KChIP modulators together with structural and/or KChIPKChIP ligandsligands havehave beenbeen described described asas KKVV4.x/KChIP4.x/KChIP modulatorsmodulators togethertogether withwith structural structural KChIP ligandselectrophysiologicalKChIP have beenligands described have studies been as K describedV (see4.x/KChIP Table as1). modulatorsKV4.x/KChIP together modulators with togetherstructural with structural and/or electrophysiologicaland/orand/orand/or electrophysiological electrophysiologicalelectrophysiological studies (see Table studies studiesstudies 1). ( ((seeseesee Table TableTable 1). 1).1). Table 1. Known small-molecule K 4.x/KChIP modulators. Table 1. Known small-molecule KVV4.x/KChIP modulators. Table 1. KnownTableTable small 1.1.-molecule KnownKnown small smallKV4.x/KChIP--moleculemolecule modulators. KKVV4.x/KChIP4.x/KChIP modulators.modulators.

LigandV KV4.x Channel KChIPx Effect Ref. Ligand LigandLigandLigandKV4.x ChannelKKK VV4.x4.x4.x KChIPx Channel ChannelChannel KChIPxKChIPxKChIPxEffect EffectEffectEffectRef. Ref.Ref.Ref.

VK 4.2 KV4.2 KKKKChIP1VV4.24.24.2V KChIP1KChIP1KChIP1 KChIP1 KChIP3InhibitionInhibition Inhibition [44,59] [[44,59]44,59] VKV4.3 Inhibition Inhibition[44,59] [44,59] KV4.3 KKKKChIP3VV4.34.34.3 KChIP3KChIP3KChIP3

CL-888 (1) CLCLCLCL-888---888888888 ( (1(111))))

V KV4.3 KKKKChIP3KVV4.34.34.3V4.3 KChIP3KChIP3KChIP3Inhibition KChIP3 InhibitionInhibitionInhibition Inhibition[44] [44][44] [[44]44]

Repaglinide (Repaglinide2RepaglinideRepaglinideRepaglinide) ( ( 2((22)2))) PolyunsaturatedPolyunsaturatedPolyunsaturatedPolyunsaturatedPolyunsaturated fatty Kv4.3 fatty fatty fattyfatty Kv4.3Kv4.3Kv4.3Kv4.3 KChIP1 KChIP1KChIP1KChIP1InhibitionKChIP1 InhibitionInhibitionInhibition Inhibition[60–62] [60[60 [[6060–––62]6262]62]] acids (PUFAs)acidsacidsacidsacids (PUFAs) (PUFAs)(PUFAs)Kv4.2 Kv4.2Kv4.2Kv4.2Kv4.2

--- KChIP3------— KChIP3KChIP3KChIP3Inhibition KChIP3 InhibitionInhibitionInhibition Inhibition[62] [62][62] [[62]62]

1,8-ANS (3) 1,81,81,81,8-ANS---ANSANSANS (( (3(333))))

KChIP2KChIP2 Inhibition/Acti-Inhibition/Acti- KChIP2V Inhibition/Acti-KChIP2 Inhibition/Acti- KV4.3 KKKVV4.34.34.3 [63–67] [63[63[63–––67]67]67] KChIP3 KChIP3KChIP3KChIP3vation vationvationvation

NS5806 (4) NS5806NS5806NS5806 ( ((444)))

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 8 of 21

Figure 5. Original KV4.3 (A) and KV4.3+KChIP3 (B) records generated after applying the pulse pro- tocol shown in the upper part of the Figure. Note that KV4.3+KChIP3 channels exhibit a slower inactivation rate than KV4.3 ones .

KChIP ligands modify this general trend, and the knowledge gained from the mod- ulation of the KV4.x/KChIP complex by small molecules could open new avenues to mod- ulate the potassium outward current through Kv4 channel complexes. At present, eight KChIP ligands have been described as KV4.x/KChIP modulators together with structural and/or electrophysiological studies (see Table 1).

Table 1. Known small-molecule KV4.x/KChIP modulators.

Ligand KV4.x Channel KChIPx Effect Ref.

KV4.2 KChIP1 Inhibition [44,59] KV4.3 KChIP3

CL-888 (1)

KV4.3 KChIP3 Inhibition [44]

Repaglinide (2) Polyunsaturated fatty Kv4.3 KChIP1 Inhibition [60–62] acids (PUFAs) Kv4.2 Int. J. Mol. Sci. 2021, 22, 1419 9 of 21

--- KChIP3 Inhibition [62] Table 1. Cont.

1,8Ligand-ANS (3) KV4.x Channel KChIPx Effect Ref.

KChIP2 Inhibition/Acti- KKVV4.34.3 KChIP2KChIP3 Inhibition/Activation[63[63–67]–67 ] Int.Int. J. J. Mol. Mol. Sci. Sci. 2021 2021, ,22 22, ,x x FOR FOR PEER PEER REVIEW REVIEW KChIP3 vation 99 of of 21 21 Int.Int. J.J. Mol.Mol. Sci.Sci. 20212021,, 2222,, xx FORFOR PEERPEER REVIEWREVIEW 99 ofof 2121

NS5806 ( 4()4)

Inhibition/Acti-Inhibition/Acti- KVK4.3V4.3 KChIP2 Inhibition/Acti-Inhibition/Acti-Inhibition/Activation [68][68] KKV4.3V4.3 KChIP2KChIP2 vation [68][68] KV4.3 KChIP2 vationvation [68] vation NS3623 (5) NS3623NS3623NS3623NS3623 (((55 5)()5) )

K K4.3V4.3 KChIP3 Inhibition Inhibition [69] [69] KKV4.3V4.3 KChIP3KChIP3 InhibitionInhibition [69][69] KV4.3 KChIP3 Inhibition [69] IQM-PC330 (6) IQMIQMIQM-PC330IQM--PC330PC330-PC330 ( (6(6 6)()6) )

V KKV4.34.3 KChIP3KChIP3 InhibitionInhibition [69][69] KKVV4.34.3 KChIP3KChIP3 KChIP3 InhibitionInhibition Inhibition [69][69] [69 ]

IQMIQM--PC332PC332 (7)(7) IQMIQMIQM-PC332--PC332PC332 (7) (7)(7)

Inhibition/Acti- V Inhibition/Acti-Inhibition/Acti- KKV4.34.3 KChIP3KChIP3 Inhibition/Acti- [70][70] KKVV4.34.3 KChIP3KChIP3 KChIP3 Inhibition/Activationvation [70][70][70 ] V vationvationvation IQM-266 (8) IQMIQM-266-266 (8) (8) IQMIQM-266-266 (8)(8)

CL-888 is a diaryl-urea compound identified by a yeast two-hybrid assay that binds CLCLCL--888888-888 isis is aa a diaryldiaryl diaryl--ureaurea-urea compoundcompound compound identifiedidentified identified byby by aa a yeastyeast yeast twotwo two--hybridhybrid-hybrid assayassay assay thatthat that bindsbinds binds CL-888 is a diaryl-urea compound identifiedV by a yeast two-hybrid assay that binds to toto KChIP1 KChIP1 and and modulates modulates KChIP1 KChIP1 activity activity on on K KV4.34.3 channel channel function. function. In In a a surface surface plasmon plasmon toto KChIP1KChIP1 andand modulatesmodulates KChIP1KChIP1 activityactivity onon KKVV4.34.3 channelchannel function.function. InIn aa surfacesurface plasmonplasmon KChIP1 and modulates KChIP1 activity on KV4.3 channel function. In aV surface plasmon resonanceresonance assay,assay, CLCL--888888 modifiesmodifies thethe apparentapparent affinityaffinity ofof KChIP1KChIP1 toto KKV4.34.3--NN,, butbut itit isis notnot resonanceresonance assay,assay, CLCL--888888 modifiesmodifies thethe apparentapparent affinityaffinity ofof KChIP1KChIP1 toto KKVV4.34.3--NN,, butbut itit isis notnot resonance assay, CL-888 modifiesV the apparent affinity of KChIP1 to KV4.3-N, but it is ableable toto dissociatdissociatee thethe KChIP1/KKChIP1/KV4.34.3--NN complex,complex, suggestingsuggesting aa conformationalconformational changechange trig-trig- ableable toto dissociatdissociatee thethe KChIP1/KKChIP1/KVV4.34.3--NN complex,complex, suggestingsuggesting aa conformationalconformational changechange trig-trig- not able to dissociate the KChIP1/KV4.3-N complex, suggesting a conformationalV change geredgered byby thethe compoundcompound binding.binding. ElectrophysiologicalElectrophysiological studiesstudies withwith KKV4.3+KChIP14.3+KChIP1 chan-chan- geredgered byby thethe compoundcompound binding.binding. ElectrophysiologicalElectrophysiological studiesstudies withwith KKVV4.3+KChIP14.3+KChIP1 chan-chan- triggerednels showed by the a much compound greater binding. inhibition Electrophysiological of the current generated studies after with the KV 4.3+KChIP1 activation of nelsnelsnels showed showed showed aa a much much much greater greater greater inhibition inhibition inhibition ofof of the the the current current current generated generated generated afterafter after the the the activation activation activation of of of channelsV showed a much greater inhibition of the current generatedV after the activation KKV4.3+KChIP14.3+KChIP1 channchannelsels thanthan thatthat producedproduced byby thethe activationactivation ofof KKV4.34.3 channelschannels [59].[59]. An-An- KKVV4.3+KChIP14.3+KChIP1 channchannelsels thanthan thatthat producedproduced byby thethe activationactivation ofof KKVV4.34.3 channelschannels [59].[59]. An-An- ofother KV4.3+KChIP1 study analyzed channels the effects than of that CL produced-888 and repaglinide by the activation on KV4.3+KChIP3 of KV4.3 channels versus K [59V4.3]. otherother study study analyzed analyzed the the effects effects of of CL CL-888-888 and and repaglinide repaglinide on on K KV4.3+KChIP3V4.3+KChIP3 versus versus K KV4.3V4.3 otherAnotherexpressed study study analyzedin a analyzedmammalian the effects the cell effects of line CL of[44].-888 CL-888 In and both repaglinide and studies, repaglinide it on was KV onconcluded4.3+KChIP3 KV4.3+KChIP3 that versus CL- versus888 KV4.3 and expressedexpressed in in a amammalian mammalian cell cell line line [44]. [44]. In In both both studies, studies, it itwas was concluded concluded that that CL CL-888-888 and and expressedKrepaglinidV4.3 expressed ine a counteracted mammalian in a mammalian cellKChIPs line [44].effects cell lineIn byboth [ 44reducing ].studies, In both the it was studies,maximum concluded it waspeak that concluded current CL-888 and and that ac- repaglinidrepaglinide ecounteracted counteracted KChIPs KChIPs effects effects by by reducing reducing the the maximum maximum peak peak current current and and ac- ac- repaglinidCL-888celerating ande thecounteracted repaglinide decay of the KChIPs counteracted current. effects Docking KChIPs by reducing studies effects ofthe byCL maximum- reducing888 into thepeak the crystal maximumcurrent structure and peak ac- of celeratingcelerating the the decay decay of of the the current. current. Docking Docking studies studies of of CL CL-888-888 into into the the crystal crystal structure structure of of celeratingcurrentKChIP1 and (PDBthe accelerating decay ID 2S1E) of the thesuggested current. decay ofDocking that the the current. bindingstudies Docking ofof CLthis- studies888 compound into of the CL-888 crystalis located into structure the in crystala depth of KChIP1KChIP1 (PDB (PDB ID ID 2S1E) 2S1E) suggested suggested that that the the binding binding of of this this compound compound is is located located in in a adepth depth KChIP1structurerelatively (PDB of hydrophobic KChIP1 ID 2S1E) (PDB suggested cavityID 2S1E) ofthat suggested KChIP1, the binding that surrounded theof this binding compound by of Tyr93, this is compound Ala97,located Trp129 in is a located depth, and relrelativelyatively hydrophobic hydrophobic cavity cavity of of KChIP1, KChIP1, surrounded surrounded by by Tyr93, Tyr93, Ala97, Ala97, Trp129 Trp129, ,and and relinLeu133atively a depth (Figure hydrophobic relatively 6) [59]. hydrophobic cavityTo accommodate of cavity KChIP1, of CL KChIP1, surrounded-888, flexibility surrounded by for Tyr93, Lys138 by Tyr93, Ala97, and Ala97, Trp129 Trp129,, isand nec- Leu133Leu133 (Figure (Figure 6 )6 [59].) [59]. To To accommodate accommodate CL CL-888-888, flexibility, flexibility for for Lys138 Lys138 and and Trp129 Trp129 is is nec- nec- Leu133andessary Leu133 (Figure (the (Figureonly 6) [59].Trp6)[ inTo59 KChIP1). ].accommodate To accommodate The presence CL-888 CL-888,, flexibilityof this flexibility Trp for residue Lys138 for Lys138 near and theTrp129 and binding Trp129 is nec- site is essarynecessaryessary (the (the (theonly only only Trp Trp Trpin in KChIP1). inKChIP1). KChIP1). The The The presence presence presence of of ofthis this this Trp Trp Trp residue residue residue near near the thethe binding bindingbinding site sitesite essaryopens (the the onlyopportunity Trp in KChIP1). for fluorescence The presence spectroscopy of this Trpto analyze residue the near binding the binding of different site opeopensopensns the thethe opportunity opportunityopportunity for forfor fluorescence fluorescencefluorescence spectroscopy spectroscopyspectroscopy to toto analyze analyzeanalyze the thethe binding bindingbinding of ofof different differentdifferent opemoleculesns the opportunity [59]. for fluorescence spectroscopy to analyze the binding of different moleculesmoleculesmoleculesmolecules [59].[59]. [[59].59].

Int.Int. J. Mol. J. Mol. Sci. Sci.2021 2021, 22, 22,, 1419 x FOR PEER REVIEW 10 of 21 10 of 21

Figure 6. Cartoon and surface representation of (A) KChIP1 (PDB ID 2I2R) residues suggested by Figure 6. Cartoon and surface representation of (A) KChIP1 (PDB ID 2I2R) residues suggested by molecularmolecular modeling modeling to be to involved be involved in CL- in888 CL-888 binding binding[59]. (B) KChIP3 [59]. (B (PDB) KChIP3 ID 2JUL) (PDB residues ID 2JUL) sug- residues gestedsuggested by molecular by molecular modeling modeling to participate to participate in 8-anilino in-1- 8-anilino-1-naphthalenenaphthalene sulfonate (1,8- sulfonateANS) bind- (1,8- ANS) ingbinding [62]. [62].

TwoTwo putative putative hydrophobic hydrophobic binding binding sites sites on KChIP3 on KChIP3 surface surface were identified were identified by fluo- by fluores- rescencecence studies studies using using thethe fluorescentfluorescent probes probes 8- 8-anilino-1-naphthaleneanilino-1-naphthalene sulfonate sulfonate (1,8-ANS) (1,8-ANS) and and6-anilino 6-anilino 2-naphthalene 2-naphthalene sulfonate sulfonate (2,6-ANS), (2,6-ANS), whichwhich are highly sensitive sensitive to to their their im- immediate mediateenvironment environment and widely and widely used touse studyd to study many many biological biological systems systems (Figure (Figure6)[ 662) [62].]. Additional Additional docking studies using the KChIP3 NMR structure of PDB ID 2JUL provided docking studies using the KChIP3 NMR structure of PDB ID 2JUL provided information information about the binding mode of 1,8-ANS and 2,6-ANS toward KChIP3. In these about the binding mode of 1,8-ANS and 2,6-ANS toward KChIP3. In these studies, possible studies, possible binding sites were analyzed and three of them were selected based on thebinding lowest sites energy were andanalyzed the presence and of threeresidues of adequate them were for selectedANS interaction. based onTwo the pockets lowest energy wereand thelocated presence at the ofC- residuesterminal domain adequate (site for 1 ANSand site interaction. 2), and one Two at the pockets EF-1–EF were-2 inter- located at the face.C-terminal Site 1, between, domain EF (site-3 and 1 and EF-4, site involves 2), and Val163, one at Lys166 the EF-1–EF-2 (participating interface. in a hydrogen Site 1, between, bondEF-3 with and the EF-4, sulfonate involves group Val163, of ANS), Lys166 Leu167, (participating Cys239, Asn247, in a hydrogen and Gln250. bond Site with2 is the sul- mainlyfonate hydrophobic, group of ANS), and Leu167, it is located Cys239, near Asn247,EF-4. In andthis Gln250.pocket ANS Site is 2 surrounded is mainly hydrophobic, by Tyr174,and it isIle182, located Met187, near Ile EF-4.190, Met191, In this Ile194, pocket His214, ANS Val215, is surrounded Phe218, Met249 by Tyr174,, and Phe252 Ile182, Met187, (FigureIle190, 6 Met191,B). The ability Ile194, of His214,1,8-ANS Val215,to interact Phe218, with the Met249, C-terminal and domain Phe252 of (Figure KChIP36B (161). The– ability 256)of 1,8-ANS agrees with to interactits binding with pocket the being C-terminal located domain at this domain. of KChIP3 Moreover, (161–256) as the agrees emis- with its sionbinding spectrum pocket of Trp169 being located and absorption at this domain. spectrum Moreover, of 1,8-ANS as overlap, the emission it was determined spectrum of Trp169 thatand the absorption distance between spectrum the of ANS 1,8-ANS molecule overlap, and Trp169 it was was determined 8–16 Å , which that the was distance similar between in the apo and Ca2+ bound KChIP3 and is adequate for the two identified C-terminal sites the ANS molecule and Trp169 was 8–16 Å, which was similar in the apo and Ca2+ bound [62]. KChIP3 and is adequate for the two identified C-terminal sites [62]. In heterologous systems, arachidonic acid modulates the KV4.x/KChIPs current [60,61]. In these studies, it was shown that arachidonic acid (AA) decreases the maximum peak amplitude of the currents generated by neuronal, KV4.3, and KV4.3+KChIP1 channels, with the greater effect being on KV4.2/KChIP1. Both analyses showed that AA produced this ef- fect in parallel with an acceleration of the decay of the current. Indeed, one of these studies Int. J. Mol. Sci. 2021, 22, 1419 11 of 21

demonstrated that the inhibitory effects produced by AA are not dependent on the kinetics of the macroscopic speed of the inactivation current. The authors also observed an acceler- ation of the KV4.2 current generated by the activation of KV4.2 channels with a deletion in the region of the channel necessary to assembly with KChIP1, whereas this celerity in the macroscopic inactivation is not observed in KV4.2 WT channels [61]. Thus, the authors infer that these results are due to the greater degree of inactivation through open channels observed in the presence of KChIP1 rather than in its absence [61,70]. Additionally, it was demonstrated that arachidonic acid directly binds to KChIP3 into site 2 located near EF-4 at the C-terminal domain by 1,8-ANS displacement studies on KChIP3 [62]. NS5806 is another diaryl-urea compound identified as a potassium current activator. NS5806 has been used to study the central role of the transient outward potassium current (ITO) on the pathogenesis of the Brugada syndrome (BrS) and to address the molecular composition of cardiac ITO [64,71]. Furthermore, studies with heterologous expressed putative ITO channel subunits indicated that the presence of KChIP2 is mandatory in the modulation of the KV channel produced by NS5806 [63]. Later on, the role of NS5806 on neuronal A-type channel function was also studied, showing that NS5806 is a valuable tool to unravel the subunit composition of native A-type channels in neurons [65]. In 2014, the NS5806 binding to KChIP3 was described [66]. This interaction was first studied by following the fact that the fluorescence emission of the unique tryptophan residue presents in the peptide sequence of KChIP3 upon binding to NS5806. The presence of NS5806 led to a significant diminution in the tryptophan fluorescence emission through a resonance energy transfer process, which correlates well with a dynamic quenching of the excited state of tryptophan. Titration of KChIP3 with increasing amounts of NS5806 showed that this interaction takes place in a Ca2+-dependent manner, with a dissociation constant of 2–5 µM in the Ca2+-bound form. NS5806/KChIP3 interaction was also characterized by displacement studies in the presence of 1,8-ANS, which confirmed that NS5806 binds to a hydrophobic cavity on KChIP3 near EF-hand 4. Displacement of 1,8-ANS bound to that cavity also allowed to determine the affinities for NS5806 binding to KChIP3, which were very similar to those previously obtained with the quenching data. Fluorescence lifetime measurements showed that the displacement of 1,8-ANS by NS5806 resulted in a significant decrease in the pre-exponential parameter related to 1,8-ANS bound at the solvent restricted hydrophobic cavity. This correlates well with the idea that the KChIP3 binding site is placed in a hydrophobic site and that KChIP3 calcium binding increases availability of the NS5806 to this cavity. The interaction between NS5806 and KChIP3 was also characterized by isothermal calorimetry, indicating that the association process is enthalpy-driven. On the other hand, in the calcium-bound form, F218A showed similar affinity for the KV4.3 (2–22) peptide as the wild-type, however, in the presence of the activator, there is a decrease in affinity. These data showed that Tyr174 and Phe218 are important residues for the NS5806 binding, and also have an influence in the interaction with the KV4.3-peptide (residues 2–22). Based on these studies, the authors suggested that NS5806 binding led to an increase in H10 flexibility. A 10 ns molecular dynamics simulation showed that, in the presence of NS5806, there is a reorganization of H10 helix and Phe252 to allow the binding of the activator, and that this cavity at the C-terminus can accommodate both the activator and the KV4.3 helix [66]. Molecular docking studies using KChIP3 structures (PDB IDs 2JUL and 2E6W) allowed analyzing the binding of NS5806 to KChIP3 [66]. The energetically most favorable binding site was located near EF-3, EF-4, and the H10 helix, with the fluorinated phenyl ring pointing towards the solvent. This binding pocket is similar to the one identified for 1,8-ANS at the C-terminus of KChIP3, which is in agreement with the displacement of 1,8-ANS by NS5806 (Figure6B) [ 62]. In the binding pocket, the ligand is surrounded by hydrophobic residues, namely, Tyr174, Ile194, Phe218, and Phe252. NS5806 establishes stacking interactions with Tyr174, Phe218, and Phe252. To ascertain the involvement of Try174, Phe218, and Phe252, site-directed mutagenesis studies were carried out. The replacement of Phe252 by Ala led to a mutant that aggregates, which suggested a role in Int. J. Mol. Sci. 2021, 22, 1419 12 of 21

KChIP3 stability. KChIP3 mutants Y174A or F218A, in the Ca2+ bound state, led to two- to threefold decreases in NS5806 affinity, whereas in the apo-state, either there is no change in affinity, as for F218A, or there is a twofold increase, as for Y174A. Y174A mutant also affected the binding affinity of KV4.3 (2–22) peptide, leading to a threefold decrease in affinity in the Ca2+ bound state, whereas in the presence of NS5806, there is an increase in the affinity of Y174A mutant for this peptide. On the other hand, in the calcium-bond form F218A showed similar affinity for KV4.3 (2–22) peptide as wild-type, however, in the presence of the activator, there is a decrease in affinity. These data show that Tyr174 and Phe218 are important residues for NS5806 binding, and also have an influence in the interaction with KV4.3-peptide (residues 2–22). Based on these studies, the authors suggested that NS5806 binding led to an increase in H10 flexibility. A 10 ns molecular dynamics simulation showed that, in the presence of NS5806, there is a reorganization of H10 helix and Phe252 to allow the binding of the activator, and that this cavity at the C-terminus can accommodate both the activator and the KV4.3 helix [66]. NS3623, a diaryl-urea analogue of NS5806, was first described as a inhibitor and later on as a hERG activator [72,73]. Next, in canine preparations, it was described that NS3623 is a dual activator of IKr and ITO [74]. Although the effects of NS3623 on the KV4.3 and HERG channels activation were not mediated by the presence of different β subunits, it seems that KChIP2 and DPP6 modulate the compound’s effects on inactivation kinetics. However, little is known regarding the structural features of the NS3623 binding site on KChIP2 [68]. In addition, IQM-PC330 and IQM-PC332 were identified as KChIP3 ligands by a multidisciplinary approach that involved structure-based drug design, organic synthesis, single-site directed mutagenesis, and surface plasmon resonance (SPR) experiments [69]. Starting from CL-888, the urea linker was replaced by an amide linker, and based on computational studies, a virtual compound library was designed, in which an additional substituted-phenyl ring was incorporated to provide further features for a more efficient interaction with KChIP3. Surface plasmon resonance experiments showed improved affinity for KChIP3. To study the binding mode of IQM-PC330 and IQM-PC332 derivatives, molecular docking studies were carried out using homology models of human KChIP3. A search for possible binding pockets was performed, and a site around Tyr118 and Tyr130 was selected for the docking of these two derivatives, which was subsequently refined by molecular dynamics studies (Figure7A). These studies showed that the n-butyl-biphenyl group is located within a hydrophobic pocket, interacting with Leu96, Phe100, Ile117, Tyr118, Phe121, Phe151, Leu155, Leu159, and the hydrophobic part of Glu103. A greater variability is observed for the binding site of the di-chloro-phenyl moiety between IQM- PC330 and IQM-PC332, although some interactions are shared by them, as the ones with Tyr130 and Leu158. To ascertain the feasibility of this binding pocket, two mutants were prepared, Y118A and Y130A. The binding of both compounds to Y118A and Y130A was evaluated by SPR experiments, showing a decrease in the binding affinity compared with the wild type KChIP3 (Figure7B). Int. J. Mol. Sci. 2021, 22, 1419 13 of 21

Figure 7. Binding site of IQM-PC330 and IQM-PC332 [69]. (A) Cartoon representation of the molecular docking complex of KChIP3 bound to IQM-PC330 (right) and IQM-PC332 (left). (B) Direct binding of IQM-PC330 and IQM-PC332 in surface plasmon resonance (SPR) assays to immobilized wtKChIP3 compared with Tyr118Ala and Tyr130Ala KChIP3 mutants.

(C) Electrophysiological effects of IQM-PC330 and IQM-PC332 on KV4.3/KChIP3 wt and KV4.3/KChIP3 mutant channels. Data are shown as mean ± SEM. * p < 0.05 vs. block produced in KV4.3/DREAM wt channels; ** p < 0.01 vs. block produced in KV4.3/DREAM wt channels; *** p < 0.001 vs. block produced in KV4.3/DREAM wt channels.

In order to evaluate the electrophysiological effects of IQM-PC330 and IQM-PC332 on these potassium channels, experiments on KV4.3/KChIP3 channels expressed in a heterologous system were performed (Figure8). Both compounds inhibited the current with IC50 of 1.6 µM and 6.8 µM, respectively. IQM-PC330 accelerated the macroscopic inactivation and, therefore, the inhibition of the charge was greater than that observed when measuring the maximum peak current. On the other hand, IQM-PC332 accelerated this inactivation at low concentrations (between 0.01 and 0.1 µM), producing similar effects to those observed by IQM-PC330. However, at greater concentrations than 1 µM, IQM-PC332 produced opposite effects over the inactivation process and, therefore, its inhibitory effects on the current were greater when measured at the maximum peak current than when measured at the charge. Both compounds slowed the recovery process towards values very close to those recorded on KV4.3 in the absence of KChIP3 [69]. In this study, the effect of the Tyr to Ala mutations in the KV4.3/KChIP3 channel complex was also analyzed. Expression of KChIP3 Tyr118Ala and KChIP3 Tyr130Ala increased the amplitude of KV4.3 current similarly to KChIP3 WT, and accelerated the kinetics of the recovery from inactivation of KV4.3 channels. In KV4.3/KChIP3-Tyr118Ala and KV4.3/KChIP3 Tyr130Ala channels, two concentrations (1 and 3 µM) of each IQM-PC compound were tested. In these experiments, both concentrations produced a negligible inhibition of the current measured both at the charge or at the peak current (Figure7C,D). Thus, these site-directed mutagenesis results are in agreement with the binding pocket derived from the theoretical studies [69]. Int. J. Mol. Sci. 2021Int., 22, J.x Mol.FOR Sci.PEER2021 REVIEW, 22, 1419 14 of 21 14 of 21

Figure 8. Electrophysiological effects of IQM-PC330 on KV4.3 and KV4.3/KChIP3 channels [69]. (A) Figure 8. Electrophysiological effects of IQM-PC330 on KV4.3 and KV4.3/KChIP3 channels [69]. (A) Bar chart comparing Bar chart comparing the inhibition of KV4.3/KChIP3 currents produced by IQM-PC330, measured the inhibition of K 4.3/KChIP3 currents produced by IQM-PC330, measured at the maximum peak current and at the at the maximumV peak current and at the charge. Inset shows original current recordings. (B) Origi- B charge. Insetnal showsnormalized original current current records: recordings. control ((black) Original line) and normalized with IQM current-PC330 records: (blue line). control (C) Left (black panel line) and with IQM-PC330shows (blue current line). (C recordings) Left panel elicited shows by current the recovery recordings protocol elicited shown by the in recoverythe top. Right protocol panel shown shows in the the top. Right panel showseffects the effects of IQM of- IQM-PC330PC330 on the on recovery the recovery process. process. Dashed Dashed lines linesrepresent represent the recovery the recovery process process of K ofV4.3 KV 4.3 current without KChIP3.current Note without that KChIP3. IQM-PC330 Note slows that theIQM recovery-PC330 process,slows the reverting recovery KChIP3 process, effects. reverting Data K areChIP3 shown ef- as mean ± SEM. * p

IQM-266 was alsoIQM-266 identified was alsoin our identified Medicinal in ourChemistry Medicinal Program Chemistry oriented Program to iden- oriented to identify tify KChIP3 ligandsKChIP3 [70]. ligands Surface [70 plasmon]. Surface resonance plasmon resonanceexperiments experiments showed that showed IQM- that266 IQM-266 binds binds to KChIP3to with KChIP3 a K withD = 4.6 a KD 0.7= 4.6μM.± To0.7 evaluateµM. To evaluate the electrophysiological the electrophysiological effects effectsof of IQM-266, IQM-266, its effectsits effects on KV on4.3/KChIP3 KV4.3/KChIP3 channels channels were evaluated were evaluated and compared and compared with those with those induced induced on KV4.3on KchannelsV4.3 channels alone (Figure alone (Figure 9). In 9these). In theseexperiments experiments,, it was it observed was observed that that IQM-266 IQM-266 inhibitedinhibited the KV the4.3/KChIP3 KV4.3/KChIP3 current current in a concentration in a concentration-,-, voltage voltage-,-, and time and-de- time-dependent- pendent-manner.manner. It diminished It diminished the maximum the maximum peak current peak current at all concentrations at all concentrations tested, tested, but, at a but, at a certaincertain range rangeof concentrations, of concentrations, it slowed it slowed the inactivation the inactivation kinetics kinetics and IQM and- IQM-266266 produced produced an increasean increase in the in transmembrane the transmembrane charge. charge.

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 15 of 21 Int. J. Mol. Sci. 2021, 22, 1419 15 of 21

Figure 9. Concentration–dependenceFigure 9. Concentration–dependence of inhibition of and/or inhibition increase and/or of the increase current of or the the current charge throughor the charge KV4.3/KChIP3 channels producedthrough byKV IQM-2664.3/KChIP3 [70 channels]. (A) Concentration–response produced by IQM-266 curve[70]. ( ofA) theConcentration effects of IQM-266–response (continuous curve of line) on KV4.3/KChIP3the channels.effects of TheIQM dashed-266 (continuous line represents line) theon fitKV of4.3/KChIP3 the data to channels. a Hill equation The dashed with nHline = represents 1 (n = 34). the (B) Effects on the current inducedfit of the bydata IQM-266 to a Hill in equation KV4.3/KChIP3 with nH channels = 1 (n = 34). at the (B) peak Effects current on the and current in the induced charge (measuredby IQM-266 as the area V of the currentin afterK 4.3/KChIP3 applying channels 250 ms pulses at the topeak +60 current mV; n and= 34). in the (C) charge Current (measured records of as K thV4.3/KChIP3e area of the current to +60 mV in the after applying 250 ms pulses to +60 mV; n = 34). (C) Current records of KV4.3/KChIP3 to +60 mV in absence and in the presence of IQM-266 (3 µM). (D) Current records of KV4.3/KChIP3 to +60 mV in the absence and in the absence and in the presence of IQM-266 (3 μM). (D) Current records of KV4.3/KChIP3 to +60 mV the presence of IQM-266 (10 µM). * p < 0.05 when comparing the effect of IQM-266 on the peak current and on the charge in the absence and in the presence of IQM-266 (10 μM). * p < 0.05 when comparing the effect of IQM- through KV4.3/KChIP3 channels. 266 on the peak current and on the charge through KV4.3/KChIP3 channels. Similarly to IQM-PC330 and IQM-PC332, IQM-266 slowed the recovery process. Similarly Moreover, to IQM-PC330 and as and observed IQM-PC332, with polyunsaturated IQM-266 slowed fatty the acids recovery (PUFAs), process. IQM-266 increased Moreover, andthe as observed closed-state with inactivation polyunsaturated degree, fatty which acids is ( consistentPUFAs), IQM with-266 a increased preferential binding of the closed-state inactivation degree, which is consistent with a preferential binding of IQM-266 to a pre-activated closed state of KV4.3/KChIP3 channels (Figure 10). The effects IQM-266 to a preof this-activated ligand wereclosed also state explored of KV4.3/KChIP3 in rat dorsal channels root ganglion (Figure neurons. 10). The In effects these cells, IQM-266 of this ligand were also explored in rat dorsal root ganglion neurons. In these cells, IQM- inhibited the peak amplitude and slowed the inactivation of IA [70]. Thus, all these results 266 inhibited identifythe peak IQM-266 amplitude as aand new slowed chemical the tool inactivation that might of allow IA [70]. a better Thus, understanding all these of KChIP3 results identifyphysiological IQM-266 as function.a new chemical tool that might allow a better understanding of KChIP3 physiological function.

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 16 of 21 Int. J. Mol. Sci. 2021, 22, 1419 16 of 21

Figure 10. FigureIQM-266 10. increases IQM-266 the increases closed-state the inactivationclosed-state ofinactivation the KV4.3/KChIP3 of the K channelsV4.3/KChIP3 [70]. channels (A) Current [70]. records (A) generated V by KV4.3/KChIP3Current channelsrecords generated in the absence by K and4.3/KChIP3 in the presence channels of IQM-266in the absence after applying and in the the presence pulse protocols of IQM shown- in the 266 after applying the pulse protocols shown in the upper part of the figure. (B) Time course of upper part of the figure. (B) Time course of closed-state inactivation of KV4.3/KChIP3 channels at a membrane potential closed-state inactivation of KV4.3/KChIP3 channels at a membrane potential (−40 mV) at which this (−40 mV) at which this current inactivates, but does not conduct. Note how IQM-266 increases the closed-state inactivation. current inactivates, but does not conduct. Note how IQM-266 increases the closed-state inactivation.

4. Concluding Remarks4. Concluding Remarks

There is a continuousThere iseffort a continuous to unravel effort the K toV4 unravel channelosome the KV4 in channelosome human physiology, in human physiology, its involvementits in involvement numerous pathologies, in numerous and pathologies, the search andof chemical the search modulators. of chemical The modulators. The study of KV4 channelosome,study of KV4 a channelosome,s well as their regulation as well as theirby other regulation proteins by using other novel proteins and using novel and selective small selectivemolecule smallligands, molecule represents ligands, pharmacological represents pharmacological and molecular challenges. and molecular challenges. This ongoing effortThis will ongoing assist effortwith the will validation assist with of K theV4 validationchannelosome of K asV4 therapeutic channelosome tar- as therapeutic gets, and the searchtargets, for and better the and search effective for better drug and candidates effective for drug the candidates treatment forof neuro- the treatment of neu- degenerative androdegenerative cardiovascular and diseases cardiovascular in which diseasesthey are ininvolved. which they Here, are we involved. have re- Here, we have viewed KV4/KChIPsreviewed modulators KV4/KChIPs acting modulators as inhibitors acting or activators as inhibitors of the or activatorsA-type currents of the A-type currents

Int. J. Mol. Sci. 2021, 22, 1419 17 of 21

as pharmacological tools for a better comprehension of the pathophysiological processes in which this channelosome participates. KChIPs modulators have been mainly described in the last decade and used as chemical tools contributing to a better understanding of the KV/KChIP interaction. To study the complex interaction of KV4 channelosome and the development of mod- ulators, the interplay of different techniques is central to advancement. In this regard, fluorescent biosensors, a potent tool for probing biomolecules in their native environment and for visualizing dynamic processes, have provided relevant clues regarding the binding site of KChIPs modulators. These tools are well suited for highlighting molecular alter- ations associated with pathological disorders, thereby offering means of implementation of sensitive and alternative technologies for diagnostic purposes. Therefore, the development of new fluorescent probes with improved photophysical properties and a high signal-to- noise ratio will constitute invaluable tools for the study of the KV4 channelosomes with high spatial and temporal resolution. The size and complexity of protein–protein interfaces make the design of ligand modulators of these interfaces into a challenging task. The knowledge of 3D structures of KV accessory subunits, in particular KChIPs, has provided the identification of regions of the protein structures that can be targeted for the discovery of new molecules that modulate channel activity. This information, together with molecular docking studies, led to the development of new small molecules as KV/KChIPs modulators. These promising results will encourage further structural and theoretical studies to gain insights into the molecular determinants of activators and inhibitors of the A-type currents, as well as to design novel compounds as chemical tools to study the protein–protein interaction network of the KV4 channelosome. Therefore, they will allow the identification of therapeutically relevant targets and the design of new potential drugs able to modulate them. The location, expression, or morphological changes of the KChIPs in different tissues, such as the brain [28] or the heart [33], open new opportunities for the treatment of diseases related to these systems. For example, KChIP2 (at the RNA level) is downregulated in sev- eral cardiac pathologies (atrial fibrillation, cardiac hypertrophy, or heart failure) [18,40,75]. Thus, the search for new KChIP2 modulators capable of increasing the A-type currents would represent a new family of drugs useful in the treatment of such cardiac pathologies. Altogether, the knowledge acquired from structural and functional studies using the recently discovered KV4/KChIPs modulators and those likely to come soon will help under- stand the underlying mechanism involving KV4-mediated-channelopathies, establishing the foundations for drug discovery, and hence their treatments.

Author Contributions: Conceptualization A.A.; J.A.G.-V.; Y.R., M.M.-M., C.V. and M.G.-R.; funding acquisition, Y.R., M.M.-M., C.V., and M.G.-R.; writing original draft P.C., D.A.P., A.d.B.-B., P.G.S., A.A.-N., D.A.-E., E.B. and E.P.-A.; writing-review and editing, A.A.; J.A.G.-V.; Y.R., M.M.-M., C.V. and M.G.-R.; project administration C.V. and M.G.-R. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Ministerio de Ciencia e Innovación (MICIU) Spain SAF2016- 75021-R and PID2019-104366RB-C21 (to CV), RTI2018-097189-B-C22 (to MMM), PID2019-104366RB- C22 (to MG-R); the Instituto de Salud Carlos III CIBERCV program CB/11/00222 (to CV); and the Consejo Superior de Investigaciones Científicas grants: PIE 201820E104, 2019AEP148 (to CV), and 201880E109 (to MMM and MG-R). The cost of this publication was paid in part by funds from the European Fund for Economic and Regional Development (FEDER). PC held a postgraduate FPI fellowship from the Spanish Ministry of Economy, Industry, and Competitivity (MINECO). DAP held CSIC contracts. AdB-B holds a MICIU predoctoral contract (BES-2017-080184). PGS holds a MICIU predoctoral contract FPU2017/02731. AAN, DAE, EB, and EPA were sponsored by CUNY Research Scholars Program (CRSP) and Collegiate Science Technology Entry Program (CSTEP). YR is grateful to the Hostos Office of Academic Affairs for providing travel awards and the OpenEye Scientific Software for providing free academic licenses. We acknowledge support of the publication fee by the CSIC Open Access Publication Support Initiative through its Unit of Information Resources for Research (URICI). Int. J. Mol. Sci. 2021, 22, 1419 18 of 21

Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations

1,8-ANS 8-anilino-1-naphthalene sulfonate DPPs Dipeptidyl-peptidase-like proteins ISA Somato-dendritic subthreshold-activating A-type K+ current ITO Transient outward current KV Voltage-gated potassium channels KChIPs Potassium channel interacting proteins MAPKs Mitogen-activated protein kinase NMR Nuclear magnetic resonance PI4K-beta Phosphatidylinositol 4-kinase beta PKA Protein kinase A PKC Protein kinase C SPR Surface plasmon resonance

References 1. June-Bum, K. Channelopathies. Korean J. Pediatr. 2014, 57, 18. [CrossRef] 2. Isbrandt, D.; Leicher, T.; Waldschütz, R.; Zhu, X.; Luhmann, U.; Michel, U.; Sauter, K.; Pongs, O. structures and expression profiles of three human KCND (Kv4) potassium channels mediating A-type currents I(TO) and I(SA). Genomics 2000, 64, 144–154. [CrossRef][PubMed] 3. Birnbaum, S.G.; Varga, A.W.; Yuan, L.L.; Anderson, A.E.; Sweatt, J.D.; Schrader, L.A. Structure and function of Kv4-family transient potassium channels. Physiol. Rev. 2004, 84, 803–833. [CrossRef][PubMed] 4. Nanao, M.H.; Zhou, W.; Pfaffinger, P.J.; Choe, S. Determining the basis of channel-tetramerization specificity by X-ray crystallog- raphy and a sequence-comparison algorithm: Family values (FamVal). Proc. Natl. Acad. Sci. USA 2003, 100, 8670–8675. [CrossRef] [PubMed] 5. Strang, C.; Cushman, S.J.; DeRubeis, D.; Peterson, D.; Pfaffinger, P.J. A Central Role for the T1 Domain in Voltage-gated Potassium Channel Formation and Function. J. Biol. Chem. 2001, 276, 28493–28502. [CrossRef] 6. Ohya, S.; Tanaka, M.; Oku, T.; Asai, Y.; Watanabe, M.; Giles, W.R.; Imaizumi, Y. Molecular cloning and tissue distribution of an alternatively spliced variant of an A-type K+ channel α-subunit, Kv4.3 in the rat. FEBS Lett. 1997, 420, 47–53. [CrossRef] 7. Kong, W.; Po, S.; Yamagishi, T.; Ashen, M.D.; Stetten, G.; Tomaselli, G.F. Isolation and characterization of the human gene encoding I(to): Further diversity by alternative mRNA splicing. Am. J. Physiol. Heart Circ. Physiol. 1998, 275, 1963–1970. [CrossRef] 8. Serodio, P.; Kentros, C.; Rudy, B. Identification of molecular components of A-type channels activating at subthreshold potentials. J. Neurophysiol. 1994, 72, 1516–1529. [CrossRef] 9. Serôdio, P.; Vega-Saenz De Miera, E.; Rudy, B. Cloning of a novel component of A-type K+ channels operating at subthreshold potentials with unique expression in heart and brain. J. Neurophysiol. 1996, 75, 2174–2179. [CrossRef] 10. Yang, K.C.; Nerbonne, J.M. Mechanisms contributing to myocardial potassium channel diversity, regulation and remodeling. Trends Cardiovasc. Med. 2016, 26, 209–218. [CrossRef] 11. Dabrowska, J.; Rainnie, D.G. Expression and distribution of Kv4 potassium channel subunits and potassium channel interacting proteins in subpopulations of interneurons in the basolateral amygdala. Neuroscience 2010, 171, 721–733. [CrossRef][PubMed] 12. Alfaro-Ruíz, R.; Aguado, C.; Martín-Belmonte, A.; Moreno-Martínez, A.E.; Luján, R. Expression, cellular and subcellular localisation of Kv4.2 and Kv4.3 channels in the rodent hippocampus. Int. J. Mol. Sci. 2019, 20, 246. [CrossRef][PubMed] 13. MacDonald, J.F.; Belrose, J.C.; Xie, Y.-F.; Jackson, M.F. Nonselective cation channels and links to hippocampal ischemia, aging, and dementia. Adv. Exp. Med. Biol. 2013, 961, 433–447. [CrossRef][PubMed] 14. Bernard, C.; Anderson, A.; Becker, A.; Poolos, N.P.; Deck, H.; Johnston, D. Acquired dendritic channelopathy in temporal lobe epilepsy. Science 2004, 305, 532–535. [CrossRef][PubMed] 15. Breen, M.S.; Dobbyn, A.; Li, Q.; Roussos, P.; Hoffman, G.E.; Stahl, E.; Chess, A.; Sklar, P.; Li, J.B.; Devlin, B.; et al. Global landscape and genetic regulation of RNA editing in cortical samples from individuals with schizophrenia. Nat. Neurosci. 2019, 22, 1402–1412. [CrossRef] 16. Waterhouse, A.M.; Procter, J.B.; Martin, D.M.A.; Clamp, M.; Barton, G.J. Jalview Version 2—A multiple sequence alignment editor and analysis workbench. Bioinformatics 2009, 25, 1189–1191. [CrossRef][PubMed] 17. Nerbonne, J.M. Molecular basis of functional voltage-gated K+ channel diversity in the mammalian myocardium. J. Physiol. 2000, 525, 285–298. [CrossRef] Int. J. Mol. Sci. 2021, 22, 1419 19 of 21

18. Radicke, S.; Cotella, D.; Graf, E.M.; Banse, U.; Jost, N.; Varró, A.; Tseng, G.N.; Ravens, U.; Wettwer, E. Functional modulation of the transient outward current Ito by KCNE β-subunits and regional distribution in human non-failing and failing hearts. Cardiovasc. Res. 2006, 71, 695–703. [CrossRef] 19. Radicke, S.; Cotella, D.; Graf, E.M.; Ravens, U.; Wettwer, E. Expression and function of dipeptidyl-aminopeptidase-like protein 6 as a putative β-subunit of human cardiac transient outward current encoded by Kv4.3. J. Physiol. 2005, 565, 751–756. [CrossRef] 20. Capera, J.; Serrano-Novillo, C.; Navarro-Pérez, M.; Cassinelli, S.; Felipe, A. The potassium channel odyssey: Mechanisms of traffic and membrane arrangement. Int. J. Mol. Sci. 2019, 20, 734. [CrossRef] 21. Mao, L.M.; Wang, J.Q. Synaptically Localized Mitogen-Activated Protein Kinases: Local Substrates and Regulation. Mol. Neurobiol. 2016, 53, 6309–6315. [CrossRef][PubMed] 22. Nakamura, T.Y.; Nakao, S.; Wakabayashi, S. Emerging roles of neuronal Ca2+ sensor-1 in cardiac and neuronal tissues: A mini review. Front. Mol. Neurosci. 2019, 12, 56. [CrossRef][PubMed] 23. Van der Heyden, M.A.G.; Wijnhoven, T.J.M.; Opthof, T. Molecular aspects of adrenergic modulation of the transient outward current. Cardiovasc. Res. 2006, 71, 430–442. [CrossRef][PubMed] 24. Portero, V.; Wilders, R.; Casini, S.; Charpentier, F.; Verkerk, A.O.; Remme, C.A. KV4.3 Expression Modulates NaV1.5 Sodium Current. Front. Physiol. 2018, 9, 178. [CrossRef][PubMed] 25. Clatot, J.; Neyroud, N.; Cox, R.; Souil, C.; Huang, J.; Guicheney, P.; Antzelevitch, C. Inter-regulation of kv 4.3 and voltage-gated sodium channels underlies predisposition to cardiac and neuronal channelopathies. Int. J. Mol. Sci. 2020, 21, 5057. [CrossRef] [PubMed] 26. Turner, R.W.; Zamponi, G.W. T-type channels buddy up. Pflugers Arch. Eur. J. Physiol. 2014, 466, 661–675. [CrossRef] 27. Anderson, D.; Mehaffey, W.H.; Iftinca, M.; Rehak, R.; Engbers, J.D.T.; Hameed, S.; Zamponi, G.W.; Turner, R.W. Regulation of neuronal activity by Cav3-Kv4 channel signaling complexes. Nat. Neurosci. 2010, 13, 333–337. [CrossRef] 28. Alfaro-Ruíz, R.; Aguado, C.; Martín-Belmonte, A.; Moreno-Martínez, A.E.; Luján, R. Cellular and subcellular localisation of kv4-associated kchip proteins in the rat cerebellum. Int. J. Mol. Sci. 2020, 21, 6403. [CrossRef] 29. Abbott, G.W. β subunits functionally differentiate human Kv4.3 potassium channel splice variants. Front. Physiol. 2017, 8, 1–15. [CrossRef] 30. Melnyk, P.; Ehrlich, J.R.; Pourrier, M.; Villeneuve, L.; Cha, T.J.; Nattel, S. Comparison of distribution and expression in cardiomyocytes of canine pulmonary veins versus left atrium. Cardiovasc. Res. 2005, 65, 104–116. [CrossRef] 31. McCrossan, Z.A.; Abbott, G.W. The MinK-related peptides. Neuropharmacology 2004, 47, 787–821. [CrossRef][PubMed] 32. Isom, L.L.; De Jongh, K.S.; Catterall, W.A. Auxiliary subunits of voltage-gated ion channels. Neuron 1994, 12, 1183–1194. [CrossRef] 33. Decher, N.; Barth, A.S.; Gonzalez, T.; Steinmeyer, K.; Sanguinetti, M.C. Novel KChlP2 isoforms increase functional diversity of transient outward potassium currents. J. Physiol. 2004, 557, 761–772. [CrossRef][PubMed] 34. Pongs, O.; Schwarz, J.R. Ancillary subunits associated with voltage-dependent K+ channels. Physiol. Rev. 2010, 90, 755–796. [CrossRef][PubMed] 35. Bett, G.C.L.; Rasmusson, R.L. Modification of K+ channel-drug interactions by ancillary subunits. J. Physiol. 2008, 586, 929–950. [CrossRef][PubMed] 36. An, W.F.; Bowlby, M.R.; Betty, M.; Cao, J.; Ling, H.P.; Mendoza, G.; Hinson, J.W.; Mattsson, K.I.; Strassle, B.W.; Trimmer, J.S.; et al. Modulation of A-type potassium channels by a family of calcium sensors. Nature 2000, 403, 553–556. [CrossRef] 37. Buxbaum, J.D.; Choi, E.K.; Luo, Y.; Lilliehook, C.; Crowley, A.C.; Merriam, D.E.; Wasco, W. Calsenilin: A calcium-binding protein that interacts with the and regulates the levels of a presenilin fragment. Nat. Med. 1998, 4, 1177–1181. [CrossRef] 38. Carrión, A.M.; Link, W.A.; Ledo, F.; Mellström, B.; Naranjo, J.R. DREAM is a Ca2+-regulated transcriptional repressor. Nature 1999, 398, 80–84. [CrossRef] 39. Morohashi, Y.; Hatano, N.; Ohya, S.; Takikawa, R.; Watabiki, T.; Takasugi, N.; Imaizumi, Y.; Tomita, T.; Iwatsubo, T. Molecular cloning and characterization of CALP/KChIP4, a novel EF-hand protein interacting with presenilin 2 and voltage-gated potassium channel subunit Kv4. J. Biol. Chem. 2002, 277, 14965–14975. [CrossRef] 40. Kuo, H.C.; Cheng, C.F.; Clark, R.B.; Lin, J.J.C.; Lin, J.L.C.; Hoshijima, M.; Nguye-Trân, V.T.B.; Gu, Y.; Ikeda, Y.; Chu, P.H.; et al. A defect in the Kv channel-interacting protein 2 (KChIP2) gene leads to a complete loss of Ito and confers susceptibility to ventricular tachycardia. Cell 2001, 107, 801–813. [CrossRef] 41. Ames, J.B.; Lim, S. Molecular structure and target recognition of proteins. Biochim. Biophys. Acta Gen. Subj. 2012, 1820, 1205–1213. [CrossRef][PubMed] 42. Burgoyne, R.D. Neuronal calcium sensor proteins: Generating diversity in neuronal Ca2+ signalling. Nat. Rev. Neurosci. 2007, 8, 182–193. [CrossRef][PubMed] 43. Bähring, R. Kv channel-interacting proteins as neuronal and non-neuronal calcium sensors. Channels 2018, 12, 187–200. [CrossRef] [PubMed] 44. Naranjo, J.R.; Zhang, H.; Villar, D.; González, P.; Dopazo, X.M.; Morón-Oset, J.; Higueras, E.; Oliveros, J.C.; Arrabal, M.D.; Prieto, A.; et al. Activating transcription factor 6 derepression mediates neuroprotection in Huntington disease. J. Clin. Investig. 2016, 126, 627–638. [CrossRef] 45. Scannevin, R.H.; Wang, K.W.; Jow, F.; Megules, J.; Kopsco, D.C.; Edris, W.; Carroll, K.C.; Lü, Q.; Xu, W.; Xu, Z.; et al. Two N-Terminal Domains of Kv4 K+ Channels Regulate Binding to and Modulation by KChIP1. Neuron 2004, 41, 587–598. [CrossRef] Int. J. Mol. Sci. 2021, 22, 1419 20 of 21

46. Bourne, Y.; Dannenberg, J.; Pollmann, V.; Marchot, P.; Pongs, O. Immunocytochemical Localization and Crystal Structure of Human Frequenin (Neuronal Calcium Sensor 1). J. Biol. Chem. 2001, 276, 11949–11955. [CrossRef] 47. Flaherty, K.M.; Zozulya, S.; Stryer, L.; McKay, D.B. Three-dimensional structure of recoverin, a calcium sensor in vision. Cell 1993, 75, 709–716. [CrossRef] 48. Vijay-Kumar, S.; Kumar, V.D. Crystal structure of recombinant bovine . Nat. Struct. Biol. 1999, 6, 80–88. [CrossRef] 49. Babu, Y.S.; Bugg, C.E.; Cook, W.J. Structure of refined at 2.2 Å resolution. J. Mol. Biol. 1988, 204, 191–204. [CrossRef] 50. Yu, L.; Sun, C.; Mendoza, R.; Wang, J.; Matayoshi, E.D.; Hebert, E.; Pereda-Lopez, A.; Hajduk, P.J.; Olejniczak, E.T. Solution structure and calcium-binding properties of EF-hands 3 and 4 of calsenilin. Protein Sci. 2007, 16, 2502–2509. [CrossRef] 51. Lusin, J.D.; Vanarotti, M.; Dace, A.; Li, C.; Valiveti, A.; Ames, J.B. NMR structure of DREAM: Implications for Ca2+-dependent DNA binding and protein dimerization (Biochemistry (2008) 47, 8, (2252–2265)). Biochemistry 2008, 47, 6088. [CrossRef] 52. Liang, P.; Wang, H.; Chen, H.; Cui, Y.; Gu, L.; Chai, J.; Wang, K.W. Structural insights into KChIP4a modulation of Kv4.3 inactivation. J. Biol. Chem. 2009, 284, 4960–4967. [CrossRef][PubMed] 53. Zhou, W.; Qian, Y.; Kunjilwar, K.; Pfaffinger, P.J.; Choe, S. Structural Insights into the Functional Interaction of KChIP1 with Shal-Type K+ Channels. Neuron 2004, 41, 573–586. [CrossRef] 54. Kim, L.A.; Furst, J.; Gutierrez, D.; Butler, M.H.; Xu, S.; Goldstein, S.A.N.; Grigorieff, N. Three-Dimensional Structure of Ito: Kv4.2-KChIP2 Ion Channels by Electron Microscopy at 21 Å Resolution. Neuron 2004, 41, 513–519. [CrossRef] 55. Pioletti, M.; Findeisen, F.; Hura, G.L.; Minor, D.L. Three-dimensional structure of the KChIP1-Kv4.3 T1 complex reveals a cross-shaped octamer. Nat. Struct. Mol. Biol. 2006, 13, 987–995. [CrossRef][PubMed] 56. Bähring, R.; Dannenberg, J.; Peters, H.C.; Leicher, T.; Pongs, O.; Isbrandt, D. Conserved Kv4 N-terminal Domain Critical for Effects of Kv Channel-interacting Protein 2.2 on Channel Expression and Gating. J. Biol. Chem. 2001, 276, 23888–23894. [CrossRef] 57. Callsen, B.; Isbrandt, D.; Sauter, K.; Hartmann, L.S.; Pongs, O.; Bähring, R. Contribution of N- and C-terminal channel domains to Kv channel interacting proteins in a mammalian cell line. J. Physiol. 2005, 568, 397–412. [CrossRef][PubMed] 58. Wang, H.; Yan, Y.; Liu, Q.; Huang, Y.; Shen, Y.; Chen, L.; Chen, Y.; Yang, Q.; Hao, Q.; Wang, K.W.; et al. Structural basis for modulation of Kv4 K+ channels by auxiliary KChIP subunits. Nat. Neurosci. 2007, 10, 32–39. [CrossRef] 59. Bowlby, M.R.; Chanda, P.; Edris, W.; Hinson, J.; Jow, F.; Katz, A.H.; Kennedy, J.; Krishnamurthy, G.; Pitts, K.; Ryan, K.; et al. Identification and characterization of small molecule modulators of KChIP/Kv4 function. Bioorg. Med. Chem. 2005, 13, 6112–6119. [CrossRef] 60. Holmqvist, M.H.; Cao, J.; Knoppers, M.H.; Jurman, M.E.; Distefano, P.S.; Rhodes, K.J.; Xie, Y.; An, W.F. Kinetic modulation of Kv4-mediated A-current by arachidonic acid is dependent on potassium channel interacting proteins. J. Neurosci. 2001, 21, 4154–4161. [CrossRef] 61. Boland, L.M.; Drzewiecki, M.M.; Timoney, G.; Casey, E. Inhibitory effects of polyunsaturated fatty acids on Kv4/KChIP potassium channels. Am. J. Physiol. Cell Physiol. 2009, 296.[CrossRef][PubMed] 62. Gonzalez, W.G.; Miksovska, J. Application of ANS fluorescent probes to identify hydrophobic sites on the surface of DREAM. Biochim. Biophys. Acta Proteins Proteom. 2014, 1844, 1472–1480. [CrossRef][PubMed] 63. Lundby, A.; Jespersen, T.; Schmitt, N.; Grunnet, M.; Olesen, S.-P.; Cordeiro, J.M.; Calloe, K. Effect of the Ito activator NS5806 on cloned Kv4 channels depends on the accessory protein KChIP2. Br. J. Pharmacol. 2010, 160, 2028–2044. [CrossRef][PubMed] 64. Calloe, K.; Soltysinska, E.; Jespersen, T.; Lundby, A.; Antzelevitch, C.; Olesen, S.P.; Cordeiro, J.M. Differential effects of the transient outward K+ current activator NS5806 in the canine left ventricle. J. Mol. Cell. Cardiol. 2010, 48, 191–200. [CrossRef] [PubMed] 65. Witzel, K.; Fischer, P.; Bähring, R. Hippocampal A-type current and Kv4.2 channel modulation by the sulfonylurea compound NS5806. Neuropharmacology 2012, 63, 1389–1403. [CrossRef] 66. Gonzalez, W.G.; Pham, K.; Miksovska, J. Modulation of the voltage-gated potassium channel (Kv4.3) and the auxiliary protein (KChIP3) interactions by the current activator NS5806. J. Biol. Chem. 2014.[CrossRef] 67. Zhang, H.; Zhang, H.; Wang, C.; Wang, Y.; Zou, R.; Shi, C.; Guan, B.; Gamper, N.; Xu, Y. Auxiliary subunits control biophysical properties and response to compound NS5806 of the Kv4 potassium channel complex. FASEB J. 2020, 34, 807–821. [CrossRef] 68. Lainez, S.; Doray, A.; Hancox, J.C.; Cannell, M.B. Regulation of Kv4.3 and hERG potassium channels by KChIP2 isoforms and DPP6 and response to the dual K+ channel activator NS3623. Biochem. Pharmacol. 2018, 150, 120–130. [CrossRef] 69. Lopez-Hurtado, A.; Peraza, D.A.; Cercos, P.; Lagartera, L.; Gonzalez, P.; Dopazo, X.M.; Herranz, R.; Gonzalez, T.; Martin-Martinez, M.; Mellström, B.; et al. Targeting the neuronal calcium sensor DREAM with small-molecules for Huntington’s disease treatment. Sci. Rep. 2019, 9, 7260. [CrossRef] 70. Peraza, D.A.; Cercós, P.; Miaja, P.; Merinero, Y.G.; Lagartera, L.; Socuéllamos, P.G.; Izquierdo García, C.; Sánchez, S.A.; López- Hurtado, A.; Martín-Martínez, M.; et al. Identification of IQM-266, a Novel DREAM Ligand That Modulates KV4 Currents. Front. Mol. Neurosci. 2019, 12, 11. [CrossRef] 71. Calloe, K.; Cordeiro, J.M.; Di Diego, J.M.; Hansen, R.S.; Grunnet, M.; Olesen, S.P.; Antzelevitch, C. A transient outward potassium current activator recapitulates the electrocardiographic manifestations of Brugada syndrome. Cardiovasc. Res. 2009, 81, 686–694. [CrossRef][PubMed] 72. Bennekou, P.; De Franceschi, L.; Pedersen, O.; Lian, L.; Asakura, T.; Evans, G.; Brugnara, C.; Christophersen, P. Treatment with NS3623, a novel Cl-conductance blocker, ameliorates erythrocyte dehydration in transgenic SAD mice: A possible new therapeutic approach for sickle cell disease. Blood 2001, 97, 1451–1457. [CrossRef][PubMed] Int. J. Mol. Sci. 2021, 22, 1419 21 of 21

73. Hansen, R.S.; Diness, T.G.; Christ, T.; Wettwer, E.; Ravens, U.; Olesen, S.P.; Grunnet, M. Biophysical characterization of the new human ether-a-go-go-related gene channel opener NS3623 [N-(4-Bromo-2-(1H-tetrazol-5-yl)-phenyl)-N0-(30- trifluoromethylphenyl)urea]. Mol. Pharmacol. 2006, 70, 1319–1329. [CrossRef][PubMed] 74. Calloe, K.; Di Diego, J.M.; Hansen, R.S.; Nagle, S.A.; Treat, J.A.; Cordeiro, J.M. A dual potassium channel activator improves repolarization reserve and normalizes ventricular action potentials. Biochem. Pharmacol. 2016, 108, 36–46. [CrossRef] 75. Caballero, R.; de la Fuente, M.G.; Gómez, R.; Barana, A.; Amorós, I.; Dolz-Gaitón, P.; Osuna, L.; Almendral, J.; Atienza, F.; Fernández-Avilés, F.; et al. In Humans, Chronic Atrial Fibrillation Decreases the Transient Outward Current and Ultrarapid Component of the Delayed Rectifier Current Differentially on Each Atria and Increases the Slow Component of the Delayed Rectifier Current in Both. J. Am. Coll. Cardiol. 2010, 55, 2346–2354. [CrossRef]