Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

The venom of scorpions contains a rich mixture of various compounds including many peptide components with a wide range of molecular weights. The biologically active constituents are often small peptide toxins that modulate the ion channels in the plasma membrane of a variety of cells. Some of these toxins alter the operation of Na+-, Cl−-, or ryanodine-sensitive Ca2+ channels, but the largest and best-studied group consists of toxins that block K+ channels (KTx).

Potassium channels represent the largest and most diverse ion channel type in the human organism with very wide tissue distribution and functional roles (Shieh et al. 2000). Peptide toxins that bind to specific K+ channels have proven to be very valuable for two reasons:

  1. 1.

    They can be used as efficient molecular tools to learn about ion channel structure and function. The ability to examine how mutations in the toxin and/or channel affect the interaction offers great flexibility in the use of these peptides. Docking simulations with toxins of known structure make it possible to pinpoint minor structural differences in the topology of closely related channels, which may then explain observed functional differences between them. Blocking a certain subset of K+ channels by high selectivity toxins can distinguish between the functions of similar channels expressed by the same cell.

  2. 2.

    Considering the enormous variety of the physiological and pathophysiological roles of K+ channels and their cell-/tissue-specific expression distribution, they are attractive pharmacological targets in the therapy of several diseases. Successful experiments in animal disease models with potassium channel-blocking toxins have provided proof of concept for the feasibility of these efforts.

Primary Structure of KTxs

A large number of scorpion toxins have been identified by the isolation of mRNA from the venom gland. The reverse-transcribed cDNA sequences can be used to predict the amino acid sequences of the peptides expressed in the venom gland. These sequences are generally 50–60 amino acid-long precursor sequences containing signal peptides and other residues that are removed after translation. Many of such KTx peptides have not been tested on K+ channels yet; however, all long-chain peptides have putative or confirmed mature, short-chain derivatives in the α-KTx group. Other toxins were directly isolated from scorpion venoms and purified with high-performance liquid chromatography (HPLC) method, and their amino acid sequences were obtained by Edman degradation or MS/MS. In most of the cases, the amount of peptide isolated from the venom is not sufficient to determine the sequence, disulfide pairing , 3D structure, and the receptor specificity of the toxin. Knowing the amino acid sequences, toxins can be synthesized with recombinant techniques or chemical synthesis methods which allow the production of the peptides in large amount.

Secondary and Tertiary Structures and Classification of KTxs

The group of scorpion toxins targeting K+ channels comprises short-chain (23–43 residues) and long-chain (42–84 residues) peptides, whose structure is stabilized by three or four disulfide bridges . At present, KTx toxins are classified into four families, α, β, γ, and κ, based on structural similarities and their specificities for various K+ channels. Except for the κ-KTx s, all members of the other three families share a characteristic structural motif, called cysteine-stabilized α/β motif (CS-αβ) , in which the α-helix is connected to a strand of the β-sheet (consisting of at least two strands, i.e., an αββ topology) by two disulfide bridges in Ci–Cj and Ci+4–Cj+2 configuration. Although the CS-αβ -fold is a dominant structural feature among KTxs, it is not exclusive for this class of molecules, as peptides with different functions also share this motif (Dimarcq et al. 1998; Thomma et al. 2002; Caldwell et al. 1998; Zhao et al. 2002). Thus, the K+ channel-blocking property should not be assumed solely based on the presence of this fold.

A highly conserved pair of residues, dubbed “the functional dyad ” in many KTxs, was found to be important for high-affinity block of various K+ channels. It consists of a lysine, whose positively charged side chain protrudes into the negatively charged environment of the selectivity filter of the potassium-conducting pore, and a hydrophobic (often aromatic) residue often situated nine positions downstream in the sequence, sterically separated by about 7 Å from the lysine. The dyad is found on the β-sheet side of the toxins. The dyad performs the same function even in toxins from sea anemone that have folds different from the signature CS-αβ. Besides the dyad, other residues also play crucial roles in forming the contact surface with the channel, thus determining selectivity (see below). Moreover, KTxs without the dyad that still block K+ channels with high affinity have also been described, suggesting that the dyad is not an essential element for blockade (Batista et al. 2002).

Members of the KTx group of scorpion toxins that were discovered the earliest and shared high sequence and structural homology were classified into the α family, and the nomenclature α-KTxm .n was introduced to denote the nth member of the mth subfamily among the α-KTx toxins. The toxins included in the original classification were short (<40 residues) and contained six conserved cysteines. Since then the α-KTx family has vastly expanded, now including 133 members, and ranging from 23 to 43 residues in size. Most of them have 3 disulfide bonds ; however, all members of families 6, 12 (except 12.5 and 12.7), and 23 are stabilized by 4 disulfide bonds. α-KTx toxins are generally known to block Shaker-type Kv channels and Ca2+-activated potassium channels.

Toxins of the β-KTx family are longer than the α-KTx toxins (47–84 residues stabilized by 3 disulfide bonds) and originate from the Buthidae , Caraboctonidae , and Scorpionidae families. These toxins contain two functionally different domains: a freely moving, possibly α-helical N-terminal segment, and a more compact cysteine-rich C-terminal segment that contains the signature CS-αβ structural motif. The N-terminal segment confers cytolytic activity to the toxin, while the C-terminal domain is responsible for K+ channel-blocking ability. These toxins are further divided into three classes based on sequence similarity.

A separate family, γ-KTx , has been devoted to KTx toxins interacting with K+ channels of the ether-á-go-go-related gene (ERG) family. γ-KTxs are all stabilized by four disulfide bonds except members 2.1 and 2.2. Their length ranges from 36 to 47 amino acids. The topography of the outer pore/turret region of the ERG family of channels is quite different from that of most other Kv channels ; therefore, it is not surprising that a particular toxin is not likely to block both groups of channels. However, there has been one toxin, BmTx3, found, which belongs to subfamily α-KTx15 and yet blocks hERG channels (Huys et al. 2004a). A more detailed study revealed that two basic residues on the α-helix side of the toxin interact with the hERG channel , most likely with residues in the turret region, distant from the selectivity filter . On the other hand, BmTx3 also possesses the Ki-Yi+9 functional dyad on the β-sheet side, via which it was shown to block A-type K+ currents , and thus was suggested to have two interaction surfaces, each acting on different channels. Interestingly, despite the presence of the dyad, the toxin does not block Shaker-type channels. Some of the γ-KTxs were found to be selective among human and rat ERG1 , ERG2, and ERG3 channels, such as γ-KTx1.1 , γ-KTx1.7, γ-KTx1.8, and γ-KTx2.1, which blocked these channel subtypes with varying affinities and therefore can be employed for the discrimination of these channels (Restano-Cassulini et al. 2006, 2008). There are currently 29 toxins that belong in the γ-KTx family.

The newest family of K+ channel-blocking toxins from scorpion venom is the κ-KTx family of peptides consisting of 22–28 residues. They originate from scorpions in the Scorpionidae and Liochelidae families. Unlike members of the other KTx families, which are based on the CS-αβ scaffold, κ-KTx toxins adopt a structure that is formed by two parallel α-helices linked by two disulfide bridges . Although the presence of the functional dyad in hefutoxins , the first κ-KTxs to be described, and Om-toxins suggested that their targets would be K+ channels, their affinities were found to be very low on the assayed channels. At present 18 κ-KTx s are listed in UniProt.

The systemic and common names of all currently known KTx toxins are listed in Table 1.

Table 1 List of KTx toxins isolated from scorpion venoms (main reference: http://www.uniprot.org/docs/scorpktx). The columns contain the Swiss-Prot accession number, the systemic and other names of the toxins, and the number of amino acids and disulfide bridges. The biological activity if determined is represented by the inhibition of a given channel (IC50 or Kd in brackets) if not stated otherwise (activation, cytotoxic effect, etc.). The abbreviated terms in the Method column indicate the principle of the measurement of the toxin-channel interaction (electrophysiology: voltage-clamp on insect or vertebrate cells; radio ligand: radioactive ligand binding assay; Xenopus: voltage-clamped Xenopus laevis oocytes; Rb+ flux: measurement of radioactive Rb+ efflux through K+ channels).
Full size table

Characterization of Toxin-Channel Interactions, Mechanism of Block

To test the affinity of a given peptide to its receptors, several methods are available. At the time of the isolation of the first scorpion toxins (noxiustoxin in 1982 (Carbone et al. 1982)), the availability of cloned ion channel genes was limited. The first ion channel gene cloned in 1982 was the nicotinic acetylcholine receptor (nAChR) of the torpedo ray (Noda et al. 1982) followed by the voltage-gated sodium channel of electric eel in 1984 (Noda et al. 1984). Therefore, the general way to test the efficiency of a peptide in inhibiting K+ channels was to isolate excitable cells generally from rat nervous system and measure the effect of the test substance on the endogenously expressed channels. Toxin-K+ channel interactions can be tested on the fast-inactivating A-type current of these cells, which is generated by Kv1.4, Kv3.4, Kv4.1, Kv4.2, and Kv4.3 α-subunits (Vacher et al. 2004; Song et al. 1998; Song 2002) or on delayed-rectifier currents of Kv1.1, Kv1.2, Kv1.5, Kv1.6, Kv2.1, Kv3.1, and Kv3.2 channels (Song 2002). These cells also express Ca2+-activated K+ channels which makes them suitable to test the inhibitory effect of the toxins on small conductance (SK) channels (Legros et al. 1996; Jouirou et al. 2004). Other primary cell cultures were also used for testing, such as bovine aortic endothelial cells (Nieto et al. 1996) or neurons from snail (Laraba-Djebari et al. 1994) or rabbit (Crest et al. 1992).

After the cloning of individual ion channel genes and the application of heterologous expression systems in Xenopus oocytes, insect, or mammalian cells, more precise methods became available to determine the receptors of the toxins (Schwartz et al. 2013; Varga et al. 2012; Lebrun et al. 1997). Measurements can be done by radiography or with electrophysiological methods. Radiography methods can be direct or indirect. Direct measurements require the radioactive labeling (in most of the cases, 125I) of the toxin which may alter the receptor specificity of the labeled toxin compared to the unlabeled form (Koch et al. 1997). Indirect assays are based on the competition of the test substance with a well-characterized radioactive labeled ligand (such as 125I apamin or 125I noxiustoxin) for the binding site (Legros et al. 1996; Pedarzani et al. 2002). The disadvantage of such measurements is that they measure the association and dissociation of the peptides to the targeted receptors at any contact surface. Kd (dissociation constant ) values in such measurements do not necessarily represent the pore-blocking ability or the dose dependence of the inhibition of the ionic flux through the channels (IC50 ). Determination of the radioactive 86Rb+ flux is another general tool to test the K+ channel inhibiting ability. Cells expressing voltage-gated K+ channels are loaded with 86Rb+ and then depolarized by high K+-containing extracellular solution. 86Rb+ flows through open K+ channels and amount of extracellular 86Rb+ can be determined with scintillation counter. Inhibitors of potassium channels decrease the 86Rb+ flux in a dose-dependent manner; therefore, the half-inhibiting concentration (IC50) with such method can be determined (Bartschat and Blaustein 1985; Koschak et al. 1998). Electrophysiological methods permit the direct measurement of ionic currents through voltage-clamped membranes. For these measurements a variety of different cells or membrane preparations can be used. Primary cell lines (neurons, lymphocytes, etc.) expressing specific ion channels endogenously are widely used for the measurements (Schwartz et al. 2013; Varga et al. 2012; Vacher et al. 2001). Recombinant techniques allow the expression of specific ion channels in various cell types (Xenopus oocytes, mammalian cells, etc.) which has the advantage of measuring specific inhibitory effect of a toxin on a given ion channel with very low probability of aspecific effect due to the absence of endogenously expressed channels (Schwartz et al. 2013; Bagdany et al. 2005; Romi-Lebrun et al. 1997a).

The receptor site for KTxs is the K+ channel pore; competition experiments confirmed that the toxins bind to a region that overlaps with the tetraethylammonium (TEA) binding site at the external entrance of the pore and that only a single peptide molecule is able to occupy the binding site at a given time (Varga et al. 2012; Miller 1988).

The relatively small size of KTxs enables them to deeply enter the vestibule of the channels allowing for multiple contact points and also exposes the majority of their residues, which results in highly variable interaction surfaces even due to minor changes in the sequence. These features enable the toxins to bind to channel surfaces in various orientations. There have been three major modes of interaction described between K+ channels and KTxs. The most frequently identified and best-characterized interaction is via the functional dyad described above (Fig. 1a). In these cases the β-sheet side of the toxin faces the entrance of the channel pore and the lysine side chain in the selectivity filter, and the hydrophobic interaction of the other dyad residue mostly accounts for the high-affinity binding.

Fig. 1
figure 1

(a) Typical blocking scheme of an α-KTx in the pore of a Shaker-related Kv channel. The interaction surface is on the β-sheet side of the toxin forming several close contacts with the bottom of the vestibule, and the side chain of the critical lysine protrudes deeply into the selectivity filter. (b) Block of the KCa2.2 (SK) channel by α-KTx5.3 occurs by an inverted orientation of the toxin compared to the typical α-KTx mechanism; the main interaction surface is on the α-helical side of the toxin. Channel residues involved in the interaction are localized in the turret region and the bottom of the vestibule. Brownian dynamics of the recognition of the scorpion toxin P05 with small-conductance calcium-activated potassium channels. (c) Interaction of γ-KTx2.1 with the HERG channel occurs mainly between the α-helix of the toxin and the large turret region of the channel. The toxin does not enter very deeply and does not fully block permeation

Full size image

A different mode of interaction was described between KCa2.x channels and α-KTx4.2 and members of the α-KTx5 subfamily (Rodriguez de la Vega et al. 2003) (Fig. 1b). In these instances influential residues were localized on the α-helix side of the toxins. Two arginines (for TSκ (α-KTx4.2 )) and three arginines (for P05 (α-KTx5.3 )) were identified as critical for binding to small conductance calcium-activated potassium channels that made contacts with channel residues on the bottom of the vestibule and the turret region. Thus, compared to typical α-KTx-Kv channel interactions, the contact region is on the opposite side of the toxins and farther away from the selectivity filter.

Members of the γ-KTx family seem to employ yet another way to bind to hERG channels. As described above for BmTx3, γ-KTxs most likely bind to extracellular segments of the extended S5–S6 linker in ERG channels , which may form an extra amphipathic α-helix (Fig. 1c). As exemplified by ErgTx (Pardo-Lopez et al. 2002), this mode of block differs in several respects from the typical α-KTx mode of block, whose characteristics are mainly defined by the critical lysine’s interaction with the selectivity filter. Due to the deep penetration of the lysine side chain, it not only interacts with potassium ions in the pore but also senses the electric field, which makes this mode of block by α-KTX sensitive to external K+ concentration and to the applied membrane voltage. Since γ-KTxs lack the equivalent of the lysine and thus do not interact directly with the pore, the block is insensitive to K+ ext, but not the membrane potential. This was explained by structural rearrangements in the S5-P linker brought on by strong depolarization, which destabilizes ErgTx binding. Although the overlap of the ErgTx binding site with that of TEA places it at the outer mouth of hERG, the inability of ErgTx to produce total current block suggests an off-center binding position rather than a complete plugging of the pore as known for α-KTx s. γ-KTx s are assumed to bind with their α-helix side in an orientation different from the two previous modes and interact with residues even farther from the selectivity filter (Rodriguez de la Vega et al. 2003; Pardo-Lopez et al. 2002).

The block mechanism of the KTxs has been studied by several methods, the earliest ones involving a large number of mutations both in the toxin and channel sequences. The structure of the toxins was fairly well known from NMR studies (Bontems et al. 1991, 1992), and based on geometric constraints, useful conclusions could be drawn about the topology of the outer pore region of the channels. Using conservative and nonconservative mutations and measuring the binding affinities, the most influential residues were identified (Goldstein et al. 1994). Most KTxs carry a high net positive charge and thus are likely to be attracted toward the negatively charged environment of the selectivity filter by long-range interactions. This involvement of electrostatic interactions is supported by the ionic strength dependence of toxin binding (MacKinnon et al. 1989). However, even charge-neutralizing mutations of toxin residues that drastically affected binding affinity had little effect on association rates, implying that toxin affinity is mostly determined by pairwise close contact interactions with channel residues. Residues forming “close contact” were defined as those whose conservative mutations resulted in great changes in binding affinity and in which the affinity change mostly arose from the dissociation rate of the toxin.

A very influential residue, the mutation of which changed binding affinity by several orders of magnitude, was identified in the “wall” of the vestibule or turret region of the Shaker channel (F425G mutation), whose role was confirmed for the corresponding residue in Kv1.3 as well (position 380) (Aiyar et al. 1995). Strikingly, this residue is very far from the cluster of other critical residues surrounding the entryway of the pore. It was shown that this residue does not contribute to normal binding of the toxin, but can greatly reduce accessibility to the pore by steric hindrance if a bulky residue is situated here. This finding underlines the fact that even residues that are not located on the typical interaction site of the toxin or the channel can have an effect on the formation of a specific channel-toxin complex, which may be a determining factor in the channel selectivity of a toxin.

Applying thermodynamic mutant cycle analysis, the closely interacting residue pairs could be pinpointed with even higher accuracy (Ranganathan et al. 1996). In this technique residues of the toxin and the channel are mutated individually and then simultaneously, and based on the binding affinities of the various combinations, a pairwise coupling energy is calculated, which characterizes the tightness of the interaction of the pair. These early studies established the critical role of the central (dyad ) lysine and recognized that it must interact with residues forming K+ binding sites in the pore based on the external K+ concentration dependence of the binding. In contrast to the pore-blocking mechanism discussed above, some spider toxins bind to the voltage sensors and modify channel gating instead of plugging the conduction pore (hanatoxin ). Chimeric toxins constructed of two other toxins active on different channels were also used to learn about the relevance of various peptide regions in the binding to different K+ channel subtypes (Regaya et al. 2004). Then, the calculated and hypothesized interaction topology can be further refined by docking simulations that use homology models of the target channel based on known X-ray crystallographic structures of a related channel and typically NMR-derived structures of the toxins. Comparison of the results of docking calculations with different channels can provide clues about which channel residues may allow or prevent high-affinity binding of the toxin.

The identified receptors of KTx toxins can be found in Table 1.

Binding and Selectivity of α-KTxs at the Molecular Level: Docking Simulations and NMR Structure Determinations of the Complexes

From the results obtained using a variety of techniques listed above, the picture of a general blocking mechanism has emerged that is employed by the majority of confirmed high-affinity K+ channel-blocking toxins. Most toxins carry a high net positive charge and thus are likely to be attracted toward the negatively charged environment of the selectivity filter by long-range interactions. This involvement of electrostatic interactions is supported by the ionic strength dependence of toxin binding (MacKinnon et al. 1989). As described above, many toxins feature the conserved functional dyad that superimposes spatially even in toxins of various lengths and structures (Menez 1998; Dauplais et al.1997) and is a good indicator of high-affinity K+ channel blockade. However, as sequence comparisons and docking simulations reveal, the hydrophobic residue of the dyad may have a major influence on the selectivity of a toxin such that the often present tyrosine shows preference for Kv1.2 channels over Kv1.3, while a threonine at that position directs toxin preference toward Kv1.3 . Recent studies confirmed these expectations with toxins in which the hydrophobic dyad residue was mutated (Bartok et al. 2013).

A similar strategy was used to convert charybdotoxin (ChTx) , which blocks several Kv channels and KCa3.1 into a more selective toxin (Rauer et al. 2000). Docking simulations aided by thermodynamic mutant cycle analyses revealed minor structural differences in the otherwise very similar topology of the external vestibules of Kv and KCa channels . A cluster of negatively charged residues was found in the turret of Kv1.3, not present in KCa3.1. A lysine residue of ChTx, which lies close to this cluster in the bound state, was mutated to negatively charged residues, which significantly reduced the affinity for Kv1.3 and therefore improved selectivity for KCa3.1.

Most models of toxin binding assume rigid topological structures for both the channel and toxin surfaces that must be complementary to a certain extent for the formation of the contact points that establish tight binding. However, recent NMR studies challenged this view and suggested that both structures are capable of flexible rearrangements during the formation of the channel-toxin complex (Lange et al. 2006). Using solid-state NMR spectroscopy (ssNMR), which is performed in a medium with limited mobility compared to the classical liquid-state NMR, the docking of kaliotoxin (KTX , α-KTx3.1) to a KcsA-Kv1.3 chimeric channel was studied. The pore region of Kv1.3, which contains the binding site for KTX, was inserted into KcsA, a bacterial K+ channel with known crystal structure at the time, and structural changes were investigated upon KTX binding.

The authors observed significant ssNMR chemical shift changes for several KTX residues that are found on one side of the KTX three-dimensional structure bound to the channel and confirmed the general layout of the interaction surface from previous models describing KTx-Kv channel complexes. The results indicated that the structure of the outer and inner helices of KcsA-Kv1.3 was mostly unaffected by KTX binding, but changes were detected in both the pore helix and the selectivity filter, which were quite significant for the GYG signature selectivity filter residues. Their data suggests that the critical lysine side chain is inserted more deeply into the selectivity filter than previous models had assumed and that its methylene groups replace water molecules in the entry region of the pore. This insertion induces a new conformational state of the filter with characteristics of both the conducting and collapsed conformation that was described for KcsA. This reorientation, along with small changes in the toxin itself, is thought to strengthen the binding by allowing a more intimate contact between the toxin and the pore.

A follow-up study by the same group further investigated this phenomenon and found similarities between the structural changes associated with toxin binding and C-type inactivation, a process which makes Kv channels nonconducting during prolonged depolarizations via rearrangement of the external pore region (Zachariae et al. 2008). Using molecular dynamics simulations, ssNMR, and electrophysiological measurements, they showed that upon toxin binding, rotation of external pore residues widens the pore and increases the number of contacts with the toxin, which both contribute to increased affinity. Thus, the original “lock and key” model of toxin binding was modified to a “hand and glove” or “induced fit” model to account for the mutual flexibility and adaptation of the two partners .

Therapeutic Applications

Many of the toxins of various venomous species are known to exert their harmful effects through interactions with the ion channels expressed by the cells of the prey. With detailed knowledge of the role of an ion channel in a cell’s functions and the effects of peptide toxins on the channel, the behavior of cells or even organs can be manipulated in a desired way to achieve therapeutic goals. The high number of potassium channel genes expressed in the human body and the variety of cellular functions that they perform present many potential targets for such medical goals. Although the pharmacological properties of small molecule channel modulators are generally better suited for therapeutic applications, peptide toxins still have some advantages that make them attractive as drug candidates. One important aspect of these is that the greater contact area of the peptides compared to small molecules with the target channel allows a higher-affinity binding; thus, a lower concentration of the blocker is required. The other aspect again arises as a result of the higher number of contact points with the channel, which enables the toxin to differentiate among channels with similar, but still slightly differing structures. As described in previous sections, even minute differences in the topology of the interaction surfaces can lead to great changes in binding affinity. The resulting selectivity is a critical characteristic of drug molecules as this prevents unwanted side effects by avoiding interactions with off-target channels.

Several in vivo experiments in animal disease models have proven the efficacy and applicability of small K+ channel-blocking peptides (Varga et al. 2012; Koshy et al. 2014). Although some of these experiments were performed with toxins originating from other species (ShK toxins from Stichodactyla helianthus), the similar size, structure, and mechanism of action assure that KTxs from scorpions would be just as effective in these applications (Dauplais et al. 1997).

The best-studied target of therapeutic application is the voltage-gated Kv1.3 channel expressed by lymphocytes. In patients with autoimmune diseases, the disease-associated autoantigen-specific T cells were identified as co-stimulation-independent effector-memory T cells, which express a high number of Kv1.3 channels. This was confirmed in multiple sclerosis , type 1 diabetes mellitus, and rheumatoid arthritis patients (Markovic-Plese et al. 2001; Wulff et al. 2003). As the activation and proliferation of the effector-memory T cells responsible for most of the tissue damage can be suppressed by selective Kv1.3 blockers, major improvements can be achieved by the use of such peptides. This concept has been elegantly proven in experiments, in which disease development or progression was prevented in rat models of multiple sclerosis, type 1 diabetes mellitus, rheumatoid arthritis, contact dermatitis, and delayed-type hypersensitivity . An advantage of this approach is that it specifically suppresses effector-memory T cell activation without compromising the protective immune response. Experiments have shown that at therapeutically relevant concentrations, the toxins did not cause toxicity in the animals (Beeton et al. 2001, 2006) and did not suppress the protective immune response to acute viral and bacterial infections.

Several naturally highly Kv1.3 -selective KTxs have been identified, for example, the recently characterized Vm24 (α-KTx 21.1) from the venom of Vaejovis mexicanus smithi, with very high affinity (Kd = 2.9 pM) and exceptionally high (>1,500-fold) selectivity over several other ion channels assayed, including the closest relatives of Kv1.3. It was also shown to reduce delayed-type hypersensitivity in rats; thus, it promises to be a valuable tool for applications requiring selective Kv1.3 blockade (Varga et al. 2012).

A Kv1.3-specific peptide was also found effective in counteracting the negative effects of elevated caloric intake by mice that were fed a diet rich in fat and fructose. It produced effects similar to the effects of Kv1.3 gene deletion, which included a reduction of blood levels of cholesterol, sugar, and insulin and enhanced insulin sensitivity. Overall toxin application resulted in decreased weight gain, adiposity, and fatty liver (Upadhyay et al. 2013).

Another disease where selective KTxs have potential therapeutic value is myotonic dystrophy type 1 (DM1), , because voltage-gated K+ channels are responsible for myoblast proliferation and differentiation.

Comparison of the functional potassium channel expression in myoblasts from healthy individuals to myoblasts from patients with DM1 revealed a switch from KCa1.1 to Kv1 channels. Specifically, Kv1.2 and Kv1.5 channel expression increased, along with a decrease in KCa1.1 expression in DM1 myoblasts. Pharmacological block of Kv1 channels in DM1 myoblasts was found to normalize proliferation and improve other factors of myotube production. In contrast, wound healing and myotube formation were impaired by selective inhibition of KCa1.1 channels in normal myoblasts. Thus, detrimental effects of the switch in K+ channel expression associated with the early stage of myogenesis in DM1 may be counteracted by selective KTxs (Tajhya et al. 2014).

Besides effector-memory T cells in the synovial fluid, resident joint cells known as fibroblast-like synoviocytes (FLS) are also responsible for many of the pathogenic features of rheumatoid arthritis (RA). FLS in RA (RA-FLS) become invasive and cause joint damage by releasing proteases and proangiogenic and proinflammatory growth factors. RA-FLS were shown to upregulate KCa1.1 channels, which localize on the leading edge of the plasma membrane. Blockade of KCa1.1 inhibited cellular migration and invasion, along with the production of pathogenic factors by interfering with cytoskeletal rearrangements. Pharmacological inhibition of KCa1.1 also improved the clinical symptoms in rat models of RA (Tanner et al. 2014). As in the cases above, the use of a selective KTx inhibitor may render general immunosuppression unnecessary during RA treatment in the future.

Recent results indicate that K+ channel inhibition may also be a beneficial tool in enhancing antitumor immunity (Koshy et al. 2013). Blockade of KCa3.1 channels was found to increase the degranulation and cytotoxicity of adherent natural killer cells and to increase the ability of these cells to reduce in vivo tumor growth.

Conclusion and Future Directions

The examples above illustrate the wide spectrum of potential applications, in which K+ channel-specific scorpion toxins of high affinity and selectivity may be used to accomplish therapeutic goals. With the number of identified KTxs growing by the day and the expansion of the body of knowledge on K+ channel distributions and functions along with details of the toxin-channel interactions, this spectrum is likely to broaden even more, and routine clinical use of these peptides may soon become reality.

Cross-References