The ‘Patch-Clamp’ Technique and its Application in Studies on Voltage-Gated Potassium Channels Kv1.3 in Normal and Cancer Cells

Abstract

The introduction of this chapter presents a historical outline of experimental methods applied in electrophysiology until development of the ‘patch-clamp’ technique. The first section briefly describes the ‘patch-clamp’ technique and its configurations, with areas of practical applications of the configurations. The second section of the chapter presents an application of the ‘whole-cell’ configuration in studying activity of voltage-gated potassium channels Kv1.3. It is pointed out that an application of this configuration enabled the discovery of these channels in human T lymphocytes in 1984. Studies performed later applying the ‘whole-cell’ configuration have shown that these channels are also expressed in many other cell types, both normal and cancer, both in the plasma membrane and in the inner mitochondrial membrane. It was also shown that the channels may be blocked by many chemically unrelated compounds. Finally, results obtained upon a combination of the ‘whole-cell’ recording with non-electrophysiological techniques provided evidence that some inhibitors of these channels may find a clinical application in therapy of many diseases, including T-cell mediated autoimmune diseases, chronic inflammatory diseases, severe cases of COVID-19 and some cancer disorders.

1. Introduction

History of electrophysiology began at the beginning of the XX-th century. In 1902, Julius Bernstein formulated the ‘potassium electrode hypothesis’. According to this hypothesis, the resting cell membrane is selectively permeable for potassium ions [1, 2, 3]. Thus, the resting membrane potential can be calculated applying the Nernst’s equilibrium potential for potassium ions. Taking into account that the intracellular potassium concentration is much higher than the extracellular one, the resting membrane potential is negative and equal to ca. -90 mV. Bernstein proposed that upon an excitation, the membrane is undergoing a transient ‘breakdown’ becoming equally permeant for all ions. Under these circumstances, the membrane potential is changing from the resting negative value to ca. 0 mV – a membrane is being ‘depolarised’. Three years later, Herrman proposed the ‘cable model’ of an axon having a cylindrical conducting core surrounded by a membrane with a high electrical resistance and capacitance, similarly to a submarine cable [4, 5]. A depolarisation of one point of the cable is propagated as a longitudinal electric current flowing from region being depolarised to regions not yet depolarised [4, 5].

The main problem for electrophysiologists at that time was lack of an experimental technique that could be applied to verify theoretical hypotheses. The first electronic device, a Wheatstone bridge, was applied in electrophysiology only in 1920s. In 1923, Cole and Curtis applied a Wheatstone bridge to measure cell membrane resistance and capacitance. Their results showed that a cell membrane has a high resistance (in the gigaohm range) and capacitance, whereas the cytoplasm has conductance of 30–60% of the conductance of the bathing saline [1]. Results obtained by Curtis and Cole confirmed that the Herrman’s ‘cable model’ was correct.

Electrophysiological studies on electrical properties of cell membranes were pushed forward after a rediscovery of a squid giant axon by Young in 1936 [6]. It has become a valuable model system for further electrophysiological studies. Results published by Hodgkin in 1937 showed that a depolarisation wave during propagation of action potentials in axons was a longitudinal electric current. These results were in accordance to the Herrman’s ‘cable model’ [7]. In 1939, changes of membrane conductance upon firing action potentials in a squid giant axon were measured by Cole and Curtis [8]. The same year, Hodgkin and Huxley measured full action potentials in squid giant axons [9]. Results obtained by Hodgkin and Huxley were confirmed in studies performed independently by Curtis and Cole [10, 11]. These results provided evidence that resting membrane potential in an axon was close to the Nernst’s equilibrium potential for potassium ions. These results were in accordance to the ‘potassium electrode hypothesis’. However, upon an excitation, no ‘breakdown’ of the cell membrane was observed [8, 9, 10, 11]. Instead, the cell membrane was transiently being depolarised to ca. +40 mV [9, 10, 11]. Thus, the membrane potential changed not only its value but also its sign from negative to positive. Such a change of the sign was an ‘overshoot’ [9, 10, 11]. The mystery of this ‘overshoot’ remained unexplained for several years.

Studies on the nature of action potentials in squid giant axons were continued in the late 1940s by Hodgkin, Huxley and Katz. Results of their experiments provided evidence that the ‘overshoot’ was due to a transiently increased membrane permeability for sodium ions [12]. Because the extracellular sodium concentration is much higher than the intracellular one, the Nernst’s equilibrium potential for sodium ions is positive (ca. +50 mV). Thus, increase in the membrane permeability for sodium ions caused a massive influx of these ions changing the membrane potential to a value close to the Nernst equilibrium potential for sodium ions. This ‘sodium hypothesis’ provided a correct explanation for the depolarisation phase of action potentials in excitable cells [12].

The next important step in electrophysiology was a development of a new experimental technique. The ‘voltage-clamp’ was developed independently by Marmont [13], Cole [14] and then by Hodgkin, Huxley and Katz (1949–1952) [15, 16, 17, 18]. During ‘voltage-clamp’ experiments, ion currents flowing through the selected cell membrane are measured under a controlled (‘clamped’) voltage. The voltage can be kept constant or changed during an experiment, according to the experimental protocol applied by a researcher. In other words, the ‘voltage-clamp’ enables measurements of ion currents flowing in response to a given voltage stimulus.

The idea of the ‘voltage-clamp’ measurement is briefly presented on Figure 1A. The membrane potential E is measured by the voltage microelectrode – E`. The electrode is connected to the high-impedance follower circuit and to the input of the amplifier working in a negative feedback loop. The command voltage (V<sub>comm</sub>) is applied to the amplifier, which compares it to the membrane potential E. If the E is different from the V<sub>comm</sub>, the amplifier compensates for the difference immediately by applying the error signal. The error signal is the current I, which is applied to the examined cell by the current microelectrode I`. The current I changes the membrane potential to a value of the V_comm. Thus, the current recorded in a ‘voltage-clamp’ measurement is the error signal. An electrical circuit is closed by an additional grounded current microelectrode I″.

Figure 1.

(A) Scheme of the electric circuit applied in the ‘voltage-clamp’ technique, and (B) Equivalent circuit of the cell membrane. A detailed description in the text.

An equivalent electric circuit of the cell membrane is shown in Figure 1B. The circuit contains voltage-dependent sodium and potassium conductance, g_Na and g_K, respectively; voltage-independent leak conductance g_L and membrane capacitance, C_M.

Currents recorded upon an application of a given voltage stimulus were a sum of sodium and potassium currents, leak current and capacitance current. Both the leak and the capacitance components were irrelevant for electrophysiological studies and were subtracted from the final record. Ion currents recorded from the giant axon of Loligo upon the membrane depolarisation contained a transient inward sodium current followed by a sustained outward potassium current. After having applied an extracellular sodium-free solution, only a sustained outward potassium current was recorded. The sodium current was obtained as the difference current [15].

Hodgkin and Huxley also built a theoretical model, which correctly describes changes in ion permeability when firing action potentials in electrically excitable cells. This ‘Hodgkin-Huxley Model’ is still used for computer simulations of action potentials, in both research studies and teaching activities.

According to the model, the membrane permeability for sodium and potassium ions is represented by the sodium and potassium conductance, respectively [19]. Both conductances depend on the membrane potential and time. When the action potential is being generated, the sodium conductance is rapidly rising to the maximum, due to activation of voltage-gated sodium channels. After a short time, it is decaying to zero, due to the channel inactivation. Inactivated channels are unable to conduct ions even while being open. The channels have to recover from inactivation in order to be able to work again. The potassium conductance is rising slower, due to a delayed activation of ‘delayed rectifier’ voltage-gated potassium channels. According to the Hodgkin-Huxley Model, the ‘delayed rectifier’ voltage-gated potassium channels do not inactivate. Therefore, the potassium channels remain active for a longer time and the potassium conductance remains high until the channels finally close at the end of the hyperpolarisation phase of action potential. A rapid activation of voltage-gated sodium channels (working in a positive feedback loop) is responsible for the depolarisation phase of action potential. A combination of inactivation of voltage-gated sodium channels and delayed activation of voltage-gated potassium channels (working in a negative feedback loop) is responsible for the repolarisation phase, when the membrane potential is going back to the resting value. Long-lasting activity of ‘delayed rectifier’ voltage-gated potassium channels is responsible for the hyperpolarisation phase, which follows the repolarisation. During hyperpolarisation, the membrane potential is below the resting value being almost equal to the Nernst’s equilibrium potential for potassium ions. The membrane hyperpolarisation is necessary to make voltage-gated sodium channels recover from inactivation. Recovery from inactivation requires some time. During that time known as time of the absolute refraction, no action potential can be generated even when a strong stimulatory pulse is applied. The period of absolute refraction is followed by a period of relative refraction when a strong stimulatory pulse is able to generate next action potential, whereas a standard pulse still cannot.

The activation and inactivation of voltage-gated sodium channels, as well as the activation and closure of voltage-gated potassium channels, were described applying a first-order chemical reaction formalism. An ion current flowing through a membrane is a sum of currents flowing through single ion channels.

The Hodgkin-Huxley Model correctly described opening of single ion channels, which is a stochastic process. The probability of channel opening depends on membrane potential and time. The channels are open by so-called ‘gating particles’. There are three ‘gating particles’ in case of voltage-gated sodium channels and four gating particles in case of ‘delayed rectifier’ voltage-gated potassium channels. The channels are open only if all the gating particles are in the ‘open’ position. If the probability of opening of one gating particle of a voltage-gated sodium channel is equal to m, the probability of opening of the channel is equal to m³. If the probability of opening of one gating particle of a voltage-gated potassium channel is equal to n, the probability of opening of the channel is equal to n⁴. The inactivation of a voltage-gated sodium channel is controlled by a different ‘inactivating particle’ working independently to the ‘gating particles’. If the probability that a sodium channel is not inactivated is equal to h, the probability that the channel is open and not inactivated is equal to m³h. The values of m, n and h depend on membrane potential and time.

Hodgkin and Huxley did not have any opportunity to measure ion currents flowing through single ion channels, since such measurements were impossible by applying the ‘voltage clamp’ technique. Single-channel recording has become possible only after having developed the ‘patch-clamp’ technique, more than 20 years later.

2. The ‘patch-clamp’ technique: historical outline

The ‘patch-clamp’ technique was firstly applied by Sakmann and Neher, who wanted to record sodium currents in denervated frog muscle fibres [20]. Their idea was to isolate a ‘patch’ of the cell membrane. The area of the ‘patch’ had to be small enough to contain single ion channels. The ‘patch’ was isolated by applying a glass pipette (with the recording electrode inside) having a narrow tip of a diameter 3–5 μm. The tip was obtained by applying a pipette puller. The tip was pressed against the surface of the examined cell, making a high-resistance (ca. 50 MΩ) electrical contact (‘seal’) with the ‘patch’ (the rest of the examined cell remains untouched). Under these experimental conditions, currents flowing through single ion channels were recorded at a given membrane potential. However, these currents were contaminated by a large background noise caused by the ‘seal’ [20, 21].

The ‘patch-clamp’ technique was significantly improved when the gigaohm-seal (‘giga-seal’) was discovered in 1980 by Sigworth and Neher (Figure 2) [22]. The researchers showed that upon a transient application of a gentle suction to the pipette, a tight contact between the tip and the ‘patch’ was obtained. It was revealed by a rapid increase of the membrane resistance from megaohm to gigaohm range (‘giga-seal’). When a ‘giga-seal’ was obtained, the ‘patch’ formed an Ω-shaped protrusion inside the tip. The distance between the rim of the tip and the contacting ‘patch’ membrane was ca. 1 Å [22]. Such a short distance occurs upon a transfer of lipid monolayers onto glass substrates [23].

Figure 2.

Above: a scheme of the ‘patch-clamp’ technique, based on a formation of a ‘gigaohm seal’, between the tip of the recording pipette and the patch of membrane of the examined cell. Below: an example of an ion current flowing through a single voltage-gated sodium channel (downward deflections).

An application of ‘giga-seals’ reduced the background noise in the ‘patch’ by an order of magnitude [24]. As a result, both quality of records and time resolution of the method were significantly improved [24]. Due to a high resistance of ‘giga-seals’, a single microelectrode is applied in the ‘patch-clamp’, to both ‘clamp’ the membrane potential and record ion currents [22, 24]. Moreover, ‘giga-seals’ share a high mechanical stability. Therefore, the ‘patch’ can be excised from the examined cell. Alternatively, it is possible to disrupt the ‘patch’ and obtain a direct contact with the whole examined cell (see below). Similarl to the ‘voltage-clamp’ technique, the examined cell must be placed in a bathing solution, with a grounded reference microelectrode, to close the electrical circuit.

3. The ‘patch-clamp’ recording configurations

3.1 The ‘cell-attached’

The improved ‘patch-clamp’ technique and its basic recording configurations were described in detail in a classical work by Hamill and co-workers [24]. Single-channel recording was described in detail later by Sakmann and Neher [25]. A schematic diagram with the configurations is depicted in Figure 3.

Figure 3.

Recording configurations of the ‘patch-clamp’ technique based on a formation of a ‘giga-seal’.

As it is shown, the basic ‘cell attached’ configuration is established after having obtained the ‘giga-seal’. Under ‘cell attached’ conditions, single-channel currents from a membrane patch on an intact cell are recorded.

The ‘cell attached’ is the least invasive configuration of the patch-clamp technique. The voltage on the examined ‘patch’ is equal to the cell resting membrane potential minus the pipette potential. The voltage is ‘clamped’ provided that the resting cell membrane potential remains stable, while ion currents are flowing through the ‘patch’. This is the case when the ion conductance of the ‘patch’ is much lower than the conductance of the rest of examined cell.

The ability to record single-channel currents from the ‘patch’ is limited by the background noise. This noise is produced by the ‘giga-seal’, by leak currents flowing through the ‘patch’, by the recording pipette and by the electronic circuit [24]. The background noise limits suitability of the ‘cell-attached’ configuration to ion channels having the single-channel conductance more than a few pS [24].

3.2 The ‘inside-out’

The ‘patch’ can be excised from the rest of the examined cell by withdrawal the recording pipette (Figure 3). After the excision, a tight vesicle is formed on the pipette tip [24]. The vesicle contains two membranes: the inner and the outer one [24]. In order to record ion currents from the ‘patch’, the outer membrane has to be disrupted. This can be done by withdrawal the recording pipette from the bathing solution for a short time, or by applying a calcium-free bathing solution [24].

The intracellular side of the ‘patch’ is directly exposed to the bathing solution. This configuration is useful for studying ionic channels activated by intracellular factors, such as intracellular calcium. The ‘inside-out’ configuration is widely applied in studying properties of calcium-activated potassium channels – K(Ca). These channels are open not by changes of the membrane potential, but by intracellular calcium ions. The probability of opening of K(Ca) channels strongly depends on the intracellular calcium concentration. Usually, it is rising from 0 to 100 percent when the intracellular calcium concentration is rising from 100 nM to 1 μM.

In the ‘inside-out’ configuration, the voltage on the examined ‘patch’ is equal to the pipette potential with an opposite sign. The ‘patch’ membrane potential is more stable than in the ‘cell-attached’ configuration. On the other hand, the ‘inside-out’ configuration is much more invasive. This may cause a wash-out of important intracellular factors controlling the activity of examined channels. Therefore, single-channel characteristics in the ‘inside-out’ recording may be different than those obtained on intact cells.

3.3 The ‘whole-cell’

As it was mentioned above, the ‘patch’ can be disrupted. This can be done by applying a pipette solution containing 150 mM KCl and low calcium (less than 1 μM), a gentle suction or by a short voltage pulse (ca. 200 mV) (Figure 3). After the disruption, a direct contact between the pipette tip and the intracellular milieu is obtained. Importantly, the disruption does not destroy the seal between the pipette rim and the cell membrane [24].

In the ‘whole-cell’ configuration, the membrane potential of the examined cell is equal to the pipette potential. Area of application of this configuration is limited to small cells (less than 30 μm in diameter) making ion currents not larger than a few nA [24]. The ‘whole-cell’ recording is irreplaceable for studying ion channels in small cells, such as lymphocytes or lymphoid cell lines (see below). Another important advantage of the ‘whole-cell’ configuration is ability to achieve the single-channel resolution [24, 26]. By applying the fluctuation analysis, it is possible to calculate the single-channel conductance. Such an approach is irreplaceable for studying ion channels with a conductance less than a few pS. The main disadvantage of the ‘whole-cell’ recording is its invasiveness, which may cause a wash-out of important intracellular factors controlling the activity of examined channels.

3.4 The ‘outside-out’

The ‘outside-out’ (known also as the ‘excised patch’) can be obtained from the ‘whole-cell’ configuration by withdrawal the recording pipette from the examined cell (Figure 3). In such a case, a small ‘patch’ is formed on the pipette tip [24]. The ‘patch’ is exposed to the bathing solution with its extracellular side. The membrane potential of the ‘patch’ cell is equal to the pipette potential.

The ‘outside-out’ recording often follows the ‘whole-cell’ recording [24]. It is useful for studying channels activated by extracellular ligands, such as neurotransmitters.

Similar, to the ‘whole-cell’, the main disadvantage of this configuration is its invasiveness, which may cause a wash-out of important intracellular factors controlling the activity of examined channels.

4. Voltage-gated potassium channels Kv1.3 in normal and cancer cells: studies with an application of the ‘patch-clamp’ technique in a combination with non-electrophysiological approaches

Voltage-gated potassium channels (Kv channels) are integral membrane proteins that share selectivity for potassium ions and are activated upon a change of the membrane potential to more positive values (membrane depolarisation). Opening of Kv channels provides transportation of potassium ions across the cell membrane down their electrochemical gradient. Movement of potassium ions through the channels makes ‘potassium current’, which can directly be recorded applying the ‘patch-clamp’ technique. The direction of these currents depends on the membrane potential. If the potential is higher than the ‘reversal potential’, the current is outward (positive). If the potential is lower than the ‘reversal potential’, the current is inward (negative). If the potential is equal to the ‘reversal potential’, the current is equal to zero. The ‘reversal potential’ for Kv1.3 channels in the plasma membrane is about −80 mV under physiological conditions. The membrane depolarisation always leads to more positive potentials. Therefore, currents flowing through Kv1.3 channels in the plasma membrane are outward (positive). Kv1.3 channels were firstly described as the ‘n’ (‘normal’) channels in 1984 in the plasma membrane in human T lymphocytes [27]. The channels were characterised in detail one year later by Cahalan and co-workers [28]. The authors applied the ‘whole-cell’ recording to record the channel activity on both the whole-cell and the single-channels level [28]. The channels are ‘delayed rectifier’ Kv channels, which activate upon the membrane depolarisation and then undergo an inactivation. An ‘inactivated’ channel is unable to conduct ions even while being open. The channel has to ‘recover’ from inactivation to become able to work again. It was shown that Kv1.3 channels undergo a complex inactivation, with the ‘fast’ and the ‘slow’ component [28, 29, 30]. The ‘slow’ component, originally named as a ‘cumulative inactivation’, is a characteristic feature of Kv1.3 channels [28]. Recovery from inactivation is also complex, having the ‘fast’ and the ‘slow’ component [28]. It was shown that a small number of Kv1.3 channels in the plasma membrane are active under a resting membrane potential. Activation of the channels provides stabilisation of the resting membrane potential [28, 29, 30]. Kv1.3 channels work in a negative feedback loop, since opening of the channels counteracts the membrane depolarisation, which leads to the opening. Studies applying non-electrophysiological techniques showed that Kv1.3 channel is one of mammalian Shaker Kv channels. Kv1.3 channels are encoded by KCNA3 gene [29, 30]. Kv1.3 channels are homotetramers containing four α subunits and four intracellular regulatory β subunits [29]. Kv1.3 channels are expressed abundantly in many different cell types [29, 30]. The channels are also expressed in the inner mitochondrial membrane (mito Kv1.3 channels) [30, 31, 32]. Single-channel mito Kv1.3 channel currents were recorded applying the ‘inside-out’ configuration on the channels expressed in so-called mitoplasts [31]. Mitoplasts were obtained by applying an osmotic shock to mitochondria isolated from examined cells [31]. During an osmotic shock, mitochondria are swelling. The swelling makes the outer mitochondrial membrane collapse, thereby exposing the stretched inner mitochondrial membrane to the ‘patch-clamp’ micropipette [31]. Mito Kv1.3 channels are slightly voltage-dependent. The channels are active at the resting potential of the inner mitochondrial potential (IMM), which is equal to ca. -180 mV. This potential is lower than the ‘reversal potential’ for mitoKv1.3 channels (ca. 0 mV). Therefore, opening of mito Kv1.3 channels induces an influx of potassium ions, thereby depolarising the inner mitochondrial membrane [30, 31, 32].

Activity of Kv1.3 channels is required in a proliferation of Kv1.3 channel-expressing cells [30, 33, 34, 35, 36, 37, 38]. It was shown that a blockage of Kv1.3 channels in the plasma membrane inhibits Kv1.3 channel-expressing cell proliferation at the checkpoint between the G1 and the S phase of the cell cycle [30, 33, 34, 35, 36, 37, 38]. An involvement of Kv1.3 channels in a regulation of the cell proliferation can be described applying two distinct models: the ‘membrane potential model’ and the ‘voltage sensor model’ [30, 38].

Kv1.3 channels also participate in induction of apoptosis of normal and cancer cells that express this channel type [30, 31, 32, 34, 36, 37, 38].

Apoptosis is a process of a programmed cell death, which occurs in all normal cells. Apoptosis occurs via two major pathways: the death receptor and the mitochondrial pathway [30, 31, 32, 34, 36, 37, 38].

It was shown that Kv1.3 channels are required for apoptosis of Kv1.3 channel-expressing cells. The plasma membrane channels are upregulated upon activation of membrane death receptors by Fas ligands [30, 38]. Activation of the channels provides a sustained efflux of potassium ions and cell shrinkage, which is indispensable for induction of apoptosis [30]. On the other hand, it was shown that an inhibition of mito Kv1.3 channels by the pro-apoptotic protein Bax is required to induce the mitochondrial pathway of apoptosis of Kv1.3 channel-expressing cells [30, 34, 36, 37, 38]. The mechanism of involvement of mitoKv1.3 channels in the cell apoptosis was described in detail elsewhere [30, 32, 38].

It was shown that some membrane-permeant small-molecule inhibitors of mitoKv1.3 channels may mimic the pro-apoptotic effect of the Bax protein, thereby leading to the cell apoptosis. These compounds may selectively eliminate Kv1.3 channel-expressing cancer cells while sparing the normal ones. The mechanism is described in detail elsewhere [30, 34, 36, 37, 38]. The most promising candidates for a putative clinical application in cancer therapy are ‘mitochondriotropic’ compounds, which combine a high lipophilicity with a positive charge. These compounds preferentially inhibit mitoKv1.3 channels, without significantly affecting the channels in the plasma membrane [30, 34, 36, 37, 38].

It is known that the expression of Kv1.3 may significantly be changed (downregulated or upregulated) both under physiological and pathological conditions (Table 1) [30, 33, 34, 36, 37, 38, 39, 40, 41, 42, 43, 44].

Cell type	Changes in expression and its putative clinical significance	References
Activated naïve and central memory human T lymphocytes (TCM)	Upregulation (ca. 300 channels per cell)	[30, 33, 38]
Activated effector memory human T lymphocytes (TEM)	Strong upregulation (ca. 1500 channels per cell)	[30, 33, 38]
CD8+ T lymphocytes in patients with ovarian cancer	Upregulation. Putatively valuable diagnostic and prognostic biomarker of a malignant ovarian cancer	[39]
Microglia, ‘classically’ activated by lipopolysaccharide (LPS)	Upregulation	[40]
Mature neoplastic B cells in chronic lymphocytic leukaemia (B-CLL)	Upregulation	[30, 41, 42, 43, 44]
Human leukaemic Jurkat T cells	Upregulation	[30, 38]
Colon cancer, smooth muscle cancer, skeletal muscle cancer, lymph node cancer	Upregulation	[30, 34, 38, 41, 42, 43, 44]
Breast cancer, pancreas adenocarcinoma, colorectal cancer	Downregulation due to methylation of the promoter region of Kv1.3-encoding gene. Putative diagnostic and prognostic marker	[30, 34, 38, 41, 42, 43, 44]

Table 1.

Examples of changes of Kv1.3 channel expression under physiological and pathological conditions.

It is known that an inhibition of Kv1.3 channels by may putatively be beneficial in therapy of various diseases (Figure 4) [30, 34, 36, 37, 38, 41, 42, 43, 44].

Figure 4.

An overview of diseases, which may putatively be treated by inhibitors of Kv1.3 channels.

Kv1.3 channels are inhibited not only by ‘classical’ blockers of potassium channels, such as tethraethylammonium (TEA) or 4-ammino-pyridine (4-AP), but also by many other compounds, both inorganic and organic [29, 30, 31, 32, 33, 34, 35, 36, 37, 38] (see below).

4.1 Patch-clamp studies on the inhibition of Kv1.3 channels in normal and cancer cells: examples of results obtained on model systems

In order to study inhibitory effects of various drugs on Kv1.3 channels, it is useful to apply a model system, in which the channels are expressed endogenously and predominantly. In case of normal cells, such a system are human T lymphocytes isolated from peripheral blood of healthy donors. In case of cancer cells, a useful model system can be lymphoblastic human T cell line Jurkat [38].

Because of a size of both T lymphocytes and Jurkat T cells, it is recommended to apply the whole-cell recording. Two types of experimental protocols can be applied in patch-clamp studies on Kv1.3 channels: sequences of depolarising voltage steps and ‘voltage ramps’.

Figure 5 depicts Kv1.3 whole-cell currents recorded in a human T lymphocyte applying a sequence of depolarising voltage steps under control conditions, upon an application of genistein at 40 μM concentration and after wash-out of the drug [45].

Figure 5.

Top: scheme of a sequence of depolarising voltage steps applied in the experiment. Below: examples of the whole-cell Kv1.3 currents recorded in a normal T lymphocyte under control conditions: (A), upon application, (B) after wash-out; and (C) of genistein at 40 μM concentration.

The experimental protocol contained 7 depolarising voltage steps from the holding potential of −90 mV, to the potentials from −60 mV to +60 mV, with an increment of 20 mV (Figure 5, upper part) [45]. Duration time of each step was 50 ms, the time interval between steps – 30 s [45]. Relatively long time interval between the steps was necessary to provide recovery of the channels from inactivation. Leak current was subtracted online, applying the P/4 procedure [45]. The idea of this procedure is to apply, before each step, four identical pre-pulses of a value equal to ¼ of the applied step and an opposite sign. The currents recorded upon an application of these pre-pulses are then summarised and subtracted from the current recorded upon an application of the step. This method enables an online subtraction of the leak current, because this current rises linearly with the voltage, according to the Ohm’s law. On the other hand, the Kv1.3 current remains unaffected, since this current is not ‘ohmic’ (the current-to-voltage relationship is not linear).

Apparently, an application of genistein significantly reduced the currents and slowed down their activation rate. The inhibitory effect was reversible [45].

An alternative approach is an application of the ‘voltage ramps’, where the cell membrane was gradually depolarised from the holding potential of −90 mV to the potentials from −100 mV (start of the ‘ramp’) to +40 mV (end of the ‘ramp’) [46]. Time duration of each ‘voltage ramp’ was 340 ms, time interval between the ‘ramps’ - 30 s. Relatively long time interval between the ‘ramps’ was necessary to provide recovery of the channels from inactivation. Figure 6 depicts Kv1.3 whole-cell currents recorded in a cancer Jurkat T cell applying ‘voltage ramps’ under control conditions, upon an application of isobavachalcone (IBC) at 3 μM concentration and after wash-out of the drug [46]. The recorded ‘ramp current’ contained both the Kv1.3 currents and the unspecific leak current (subtracted during a subsequent offline analysis) (Figure 6) [46].

Figure 6.

Whole-cell currents recorded in a Jurkat T cell (lower panel) applying the ‘voltage ramp’ (upper panel), under control conditions (blue trace), upon an application of 3 μM isobavachalcone (IBC - orange trace) and during wash-out of the drug (grey trace).

Apparently, an application of isobavachalcone significantly reduced the Kv1.3 current [46]. Of note, the Kv1.3 channel component did not recover after wash-out of the drug. This indicates that in this case, the inhibitory effect on Kv1.3 channels was partially irreversible [46].

5. Inhibitors of Kv1.3 channels as putative therapeutic agents

Several groups of Kv1.3 channel inhibitors may putatively be applied in a medicinal practice (Table 2). Among them are both peptide inhibitors and small-molecule organic compounds, such as calcium channel blockers, Nonsteroidal Anti-Inflammatory Drugs (NSAIDs), macrolide antibiotics, psoralens and their ‘mitochondriotropic’ derivatives, naturally occurring polycyclic compounds (Table 2). Naturally occurring compounds, which share the ability to inhibit Kv1.3 channels, are flavonoids 8-prenyl- and 6-prenyl-naringenin, isoxanthohumol, genistein, acacetin, chrysin and polyphenol – resveratrol; chalcones xanthohumol, isobavachalcone and licochalcone A and statins pravastatin, lovastatin, mevastatin and simvastatin [38, 55]. Importantly, the inhibitory effect on the channels is additive or synergistic when the flavonoids 8-prenylnaringenin, 6-prenylnaringenin, acacetin and chrysin and the chalcones xanthohumol and isobavachalcone are co-applied with the statins mevastatin and simvastatin [38, 55]. To the group of small-molecule organic compounds also belong newly designed thiophene-based inhibitors, recently designed by Gubic and co-workers (Table 2) [56].

Class of inhibitors	Names of compounds	Concentration required for channel block	Putative medicinal applications	References
Peptide inhibitors	ShK, BgK, MgTX, BmPO2, BmKTX, OSK1, KTX, ChTX, MTX, NTX, Pi1, Vm24, BF9	From picomolar to nanomolar	T-cell mediated autoimmune diseases, neuroinflam-matory diseases, liver diseases	[34, 35, 38, 47]
Calcium channel blockers	Verapamil, Nifedipine, Diltiazem, Benidipine	Up to 100 μM	Chronic inflammatory diseases	[29, 38, 48, 49]
Nonsteroidal Anti-Inflammatory Drugs, macrolide antibiotics	Diclofenac sodium, salicylate, indomethacin, clarithromycin, chloroquine	Up to 100 μM	Chronic inflammatory diseases, severe cases of COVID-19	[38, 48, 49, 50]
Psoralens and clofazimine	Psora-4, PAP-1, clofazimine	Nanomolar	Cancer disorders characterised by an over-expression of Kv1.3 channels	[30, 34, 36, 38]
‘Mitochon- Driotropic’ derivatives of PAP-1	PAPTP, PCARBTP, PCTP	Low nanomolar, low micromolar	As mentioned above	[30, 36, 37, 38, 51, 52, 53, 54]
Naturally occurring polycyclic compounds	Flavonoids, polyphenols, chalcones, statins	From low micromolar up to 100 μM	As mentioned above	[30, 38, 55]
Thiophene-based compounds	Series of compounds obtained by structural optimisation of benzamide-based hit compounds	Nanomolar	As mentioned above	[56]

Table 2.

Overview of inhibitors of Kv1.3 channels that may putatively be applied in a medicinal practice.

6. More research required

As it was shown above, an application of the ‘patch-clamp’ technique, mainly in the whole-cell recording mode, enabled researchers to discover many various inhibitors of Kv1.3 channels, in both normal and cancer cells. An inhibition of the channels may be beneficial in treatment of a wide spectrum of diseases. However, in order to apply various inhibitors of Kv1.3 channels in medicinal practice, more research studies combining the ‘patch-clamp’ technique and non-electrophysiological approaches are required.

Studies on new compounds have to be performed. One of them is genistein, a plant-derived isoflavone known as a potent inhibitor of the protein tyrosine kinase (PTK). Genistein is a biologically active compound, which exerts pleiotropic effects including anti-proliferative and pro-apoptotic activities on Kv1.3 channel-expressing cancer cells [55]. It was shown that genistein is an inhibitor of Kv1.3 channels in normal human T lymphocytes [45]. Recent preliminary studies provide evidence that genistein also inhibits Kv1.3 channels expressed in human T cell line Jurkat [Teisseyre, preliminary results]. The inhibitory effect of genistein on the channels occurred in a concentration-dependent manner. It was accompanied by a significant slowing of the channel activation rate without significantly affecting the inactivation. The inhibitory effect of genistein on the channels was reversible for all applied concentrations. Obtained results may suggest that genistein molecules directly bind to the channel protein and stabilise the channel in its closed state, leading to a delay of its opening upon an application of an activating (opening) stimulus. The delayed opening of the channels may be responsible for an apparent reduction of the whole-cell currents upon an application of genistein. Genistein molecules probably dissociate from the channel protein upon its opening and re-associate with the channel after its closure [Teisseyre, preliminary results]. The inhibitory effect of genistein on Kv1.3 channels in Jurkat T cells is significantly augmented upon a co-application with mevastatin [Teisseyre, preliminary results]. In order to study the mechanism of the channel inhibition by genistein, more studies will have to be done, including the docking analysis. A combination of electrophysiological and non-electrophysiological techniques is necessary to study a putative relationship between the channel inhibition and anti-cancer activities of genistein.

Other naturally occurring compounds to be tested are flavonoids morin, hesperetin, quercetin and kaempferol; chalcones xanthohumol and isobavachalcone and a polyphenol resveratrol. All the flavonoids listed above are inhibitors of voltage-gated potassium channels Kv11.1 expressed in Xenopus oocytes [57, 58]. The potency of the Kv11.1 channel inhibition is the highest for morin and the lowest for quercetin and kaempferol [57, 58]. Moreover, hesperetin is an inhibitor of Kv1.5 channels expressed in HEK293 cells [57]. Quercetin is also an activator of Kv1.5 channels expressed in Xenopus oocytes and in rats ‘in vivo’ and an inhibitor of these channels in HEK293 cells [57]. Xanthohumol and isobavachalcone are already known as inhibitors of Kv1.3 channels in cancer Jurkat T cells [30, 38, 46, 55]. More studies combining the ‘patch-clamp’ and non-electrophysiological techniques are necessary to elucidate putative relationship between inhibition of the channels and anti-cancer activity of these compounds [55]. Resveratrol is known as an inhibitor of K1.3 channels in normal human T lymphocytes [30, 38]. A putative inhibitory effect of this compound on the channels expressed in cancer cells remains to be studied in detail, as well as the relationship between the channel inhibition and anti-cancer activity. Among statins, the best candidates for further studies are cerivastatin, fluvastatin and atorvastatin. All these compounds share anti-cancer activities against Kv1.3 channel-expressing cancer cells [55]. The selected flavonoids, chalcones and polyphenols will be applied alone and in combination with the statins [55].

7. Conclusions

The ‘patch-clamp’ technique is a useful electrophysiological technique that can be applied to record activity of ion channels on both the single-channel and the whole-cell level. An application of the later configuration is necessary in case of small cells, which cannot be examined applying the classical two-microelectrode voltage-clamp technique. One of types of voltage-gated potassium channels, which were discovered in small cells, such as human T lymphocytes, applying the whole-cell ‘patch-clamp’ technique, are Kv1.3 channels. These channels are widely expressed among tissues, both normal and cancer. The channels play an important role in regulation of the cell life and death. Inhibition of the channels may be beneficial in treatment of numerous diseases. Inhibitors of Kv1.3 channels may putatively find an application in a medicinal practice. In order to accomplish this goal, more combined studies with an application of the ‘patch-clamp’ and non-electrophysiological techniques will have to be performed.

Acknowledgments

The authors want to express the best thanks to Mrs. Anna Uryga for a successful research co-operation.

Conflict of interest

The authors have no conflict of interest.