Potent Suppression of Kv1.3 Potassium Channel and IL-2 Secretion by Diphenyl Phosphine Oxide-1 in Human T Cells

Diphenyl phosphine oxide-1 (DPO-1) is a potent Kv1.5 channel inhibitor that has therapeutic potential for the treatment of atrial fibrillation. Many other Kv1.5 channel blockers also potently inhibit the Kv1.3 channel, but whether DPO-1 blocks Kv1.3 channels has not been investigated. The Kv1.3 channel is highly expressed in activated T cells, which is considered a favorable target for immunomodulation. Accordingly, we hypothesized that DPO-1 may exert immunosuppressive and anti-inflammatory effects by inhibiting Kv1.3 channel activity. In this study, DPO-1 blocked Kv1.3 current in a voltage-dependent and concentration-dependent manner, with IC50 values of 2.58 µM in Jurkat cells and 3.11 µM in human peripheral blood T cells. DPO-1 also accelerated the inactivation rate and negatively shifted steady-state inactivation. Moreover, DPO-1 at 3 µM had no apparent effect on the Ca2+ activated potassium channel (KCa) current in both Jurkat cells and human peripheral blood T cells. In Jurkat cells, pre-treatment with DPO-1 for 24 h decreased Kv1.3 current density, and protein expression by 48±6% and 60±9%, at 3 and 10 µM, respectively (both p<0.05). In addition, Ca2+ influx to Ca2+-depleted cells was blunted and IL-2 production was also reduced in activated Jurkat cells. IL-2 secretion was also inhibited by the Kv1.3 inhibitors margatoxin and charybdotoxin. Our results demonstrate for the first time that that DPO-1, at clinically relevant concentrations, blocks Kv1.3 channels, decreases Kv1.3 channel expression and suppresses IL-2 secretion. Therefore, DPO-1 may be a useful treatment strategy for immunologic disorders.


Introduction
Diphenyl phosphine oxide-1 (DPO-1) represents a novel class of Kv1.5 blocker that has been documented to selectively prolong atrial action potential duration (APD) without exerting significant effects on ventricular APD [1], [2]. In vivo, DPO-1 has been shown to increase the atrial effective refractory period and terminate atrial arrhythmias within 60 s in dog [3], [4]. However, in cholinergic atrial fibrillation (AF) models in pig, DPO-1 at 0.5 mM did not terminate AF in a majority of experiments [5]. Recently, we have found that DPO-1 preferentially blocks the open state of the Kv1.5 channel by binding with several key residues in the S5pore loop-S6 domains [1]. Interestingly, all of these Kv1.5 key residues are also highly conserved in the Kv1.3 channel [6], [7], which provides a molecular structural basis for why many Kv1.5 blockers (e.g., AVE0118 and S0100176) also potently inhibit the Kv1.3 channel [8], [9]. Therefore, the structural homology among these channel subtypes suggested that DPO-1 may block Kv1.3 channels as well.
To date, no studies have evaluated the ability of DPO-1 to block Kv1.3 channels. Xenopus oocytes or mammalian cell lines (human embryonic kidney 293 and Chinese hamster ovary cells) have been widely used to characterize the electropharmacological properties of Kv1.3 channel blockers [19], [20]. However, use of these cell types does not reflect the true physiological environment of the Kv1.3 channel in human T cells and does not allow the immunomodulatory effects of Kv1.3 channel blockers to be evaluated. In the present study, we compared the effects of DPO-1 on Kv1.3 channels in the Jurkat cell line and in human peripheral blood T cells, and further investigated the effects of DPO-1 on Ca 2+ influx, and IL-2 secretion in Jurkat cells. In the current study we demonstrate for the first time that DPO-1 blocks Kv1.3 currents, decreases Kv1.3 channel expression, attenuates Ca 2+ influx, and inhibits IL-2 production.

Ethics statement
In this study, the study protocol with human blood samples was approved by the Ethics Committee of Tongji Medical College of Huazhong University of Science and Technology. Human blood samples were taken from healthy blood donors, who were provided written informed consent for the collection of blood and subsequent T cells isolation and analysis.

Cell preparation and culture condition
The human leukemia T-cell line, Jurkat E6-1, was obtained from the American Tissue Culture Collection (ATCC, Rockville, MD, USA). Human peripheral blood T cells were separated from whole blood samples using Ficoll gradients and purified by negative selection using CD4 + T Cell Isolation Kit (Miltenyi Biotec, Bergisch-Gladbach, Germany). All cells were grown in culture medium consisting of RPMI 1640 supplemented with 10% heat-inactivated FBS, 10 mM HEPES, 2 mM glutamate, 100 U/ mL penicillin/streptomycin and were maintained at 37uC in a humidified 95% air and 5% CO 2 atmosphere. Jurkat cells were stimulated with 50 ng/mL phorbol ester (PMA, Sigma-Aldrich, St Louis, MO) and 5 mg/mL phytohematogglutinin (PHA, Sigma-Aldrich). DPO-1 (Tocris Biosciences, Bristol, UK) at 3 mM or 10 mM was added at the onset of stimulation or 30 minutes prior to stimulation. Margatoxin (MgTX; 10 nM) and charybdotoxin (ChTX; 100 nM) were used as positive controls to block Kv1.3 channels (both from Alomone Laboratories Ltd, Jerusalem, Israel).

Electrophysiological recordings
All currents were recorded using a whole-cell patch configuration at room temperature. The external Ringer solution was (in mM): 137 NaCl, 4 KCl, 1.8 CaCl 2 , 1 MgCl 2 , 10 glucose, and 10 HEPES, adjusted with NaOH to pH 7.4. For Kv1.3 current measurements, the pipette solution (in mM) consisted of 130 KCl, 1 MgCl 2 , 5 EGTA, 5 Mg-ATP, 10 HEPES, adjusted to pH 7.2 with KOH. Kv1.3 currents were elicited by depolarizing pulses ranging from 280 mV to +60 mV in 10 mV increments from the holding potential of 280 mV. The membrane potential was measured in zero current (I = 0) mode using whole-cell patch technique. K Ca currents were elicited by using 200 ms voltage ramps from 2120 mV to +60 mV, applied every 30 s from the holding potential of 240 mV. For K Ca measurements, the pipette solution contained (in mM): 130 potassium aspartate, 10 EGTA, 8.55 CaCl 2 , 2.08 MgCl 2 , and 10 HEPES, adjusting to pH 7.2 using KOH, with a calculated free [Ca 2+ ] of approximately 1 mM. For some experiments recording K Ca , the bath solution was replaced by a K + Ringer solution, which featured identical ingredients, except that all NaCl was substituted by KCl.
The pCLAMP 9.0 and Origin 8.5 (OriginLab Corporation, Northampton, MA, USA) software were used for data acquisition and analysis. Membrane conductance (G) was defined as I/ (V2E rev ), where I is the peak amplitude, V is the test voltage, and E rev is the reversal potential (290 mV) of Kv1.3 currents [21]. The steady-state inactivation curve was studied by using a double-pulse protocol, in which the test voltage was stepped to +40 mV, 100 ms long, and preceded by 30 s preconditioning pulses from 280 to 0 mV in 10 mV steps. The data for activation and steady-state inactivation were fitted with the Boltzmann equation to get the half-maximum voltage of activation or inactivation voltage (V 1/2 ) and slope factor. The percent inhibition (% inhibition) of the current was defined as [(I control 2I drug )/I control ]*100%. The concentration required for 50% inhibition of the current (IC 50 ) was determined by fitting the data with the Hill equation.

Western blotting
Jurkat cell total lysate was prepared according to standard procedures. Cells were washed twice in cold PBS, homogenized with PMSF buffer (Roche, Basel, Switzerland), and centrifuged at 12,000 rpm for 20 min. Protein concentrations of the supernatants were measured using a BCA kit (Pierce, Rockford, IL, USA). Samples containing 40 mg of total protein were boiled in SDS loading buffer, separated on a 10% SDS-PAGE gel, and transferred to nitrocellulose membranes. The membranes were blocked in 5% dry milk, supplemented with 0.3% Tween-20 TBS, for 1 h at room temperature and probed with anti-Kv1.3 (diluted 1:500, Abcam, MA, USA) or anti-b-actin (diluted 1:1000, Santa Cruz Biotechnology, Inc., CA, USA) overnight at 4uC. The membranes were washed three times and incubated with antirabbit or anti-mouse peroxidase-conjugated IgG (diluted 1:3000, Antigene, Wuhan, China) for 1 h at room temperature. Bands were visualized using West Pico Chemiluminescent Substrate (Thermo Scientific Inc., Bremen, Germany) and quantified by densitometric analysis using Quantity One software (Bio-Rad Laboratories, Hercules, CA).

Single cell Ca 2+ measurements
Jurkat cells were loaded with 2 mM fluo-3 AM (Invitrogen, Carlsbad, CA) at room temperature for 30 min and washed twice with Ca 2+ free Ringer's solution. Cells were allowed to adhere to laminin-coated glass coverslip chambers (Sigma-Aldrich) on the stage of a Nikon (Tokyo, Japan) A1SI laser confocal microscope equipped with a 406 (Uplan/Apo, N.A. 1.0) objective and illuminated at 488 nm. Fluorescence emissions at 530 nm were captured with a charge-coupled device camera, digitized, and analyzed using the NIS-Elements microscope imaging software program. Images were recorded at intervals of 10 s. Ca 2+ release was induced by 1 mM thapsigargin (TG, Alomone) in Ca 2+ free Ringer solution. After 5 min, normal Ringer solution (2 mM CaCl 2 ) was reintroduced. The fluorescent index DF/F 0 was calculated as the average Ca 2+ response (DF = F-F 0 , F 0 is the mean value of background fluorescence).

IL-2 secretion measurements
IL-2 production in Jurkat cells was measured using an ELISA kit (eBioscience, San Diego, CA, USA) following the manufacturer's instructions. Cells were centrifuged at 1500 rpm for 10 min, and the supernatants were collected to measure IL-2 concentrations. Reactions were performed in 96-well plates coated with the capture antibody and stopped with phosphoric acid (1 M). Absorbance was measured with an automatic plate reader at 450 nm. Each experiment was repeated at least three times in duplicate.

Statistical analysis
Results are reported as mean6SEM. The significance of any differences before and after drug treatment was evaluated using a paired t-test. Comparisons between groups were carried out by analysis of variance with Turkey's post-test. Significance was set at P,0.05.

DPO-1 blocks Kv1.3 channel currents in human T cells
We first defined the Kv channels expressed in Jurkat cells and human peripheral blood T cells. Kv currents were elicited by 300 ms pulses at test potentials ranging from 280 to +60 mV with 10 mV steps, from a holding potential of 280 mV. The outward currents at potentials positive to 240 mV were characterized by rapid activation and slow inactivation as reported previously ( Figure 1A and 1B) [22], [23]. After application of 1 nM MgTX, a selective Kv1.3 channel blocker [24], the currents were nearly completely abolished ( Figure 1C), confirming that the Kv1.3 current was the predominant component of the outward current in Jurkat cells and human peripheral blood T cells.
We then examined the concentration-dependent inhibitory effects of DPO-1 on Kv1.3 currents in Jurkat cells and human peripheral blood T cells. Cells were perfused externally with different concentrations of DPO-1 from 0.3 to 30 mM. The peak current amplitude generated by stimulation at +40 mV was used as the index of inhibition (Figure 2A Figure 2C and 2D). The Hill coefficient was very close to 1, which suggests that DPO-1 blocks the channel in a 1:1 stoichiometric ratio.
We further evaluated the biophysical effects of DPO-1 on Kv1.3 in Jurkat cells. Figure 3A shows the current density-voltage relationships, before and after the addition of DPO-1 and after the washout of DPO-1 in Jurkat cells. DPO-1 at 3 mM reduced the Kv1.3 current for the entire voltages range over which this current was activated, and the blocking effect of DPO-1 was completely reversible after washout. Further, the membrane potential was depolarized from 241.0761.93 mV to 225.9962.91 mV (P,0.01) and recovered to 241.5461.60 mV after washout.
To investigate the voltage-dependent block of DPO-1, the percent inhibition of DPO-1 was plotted as a function of membrane potential together with the activation curve. In the presence of DPO-1, inhibition increased steeply between 240 and 0 mV, from 767% to 5465%, corresponding to the voltage range for channel opening. This inhibition reached a steady state between +10 and +60 mV, when the Kv1.3 channels were fully activated ( Figure 3B). These results suggested that blocking Kv1.3 by DPO-1 treatment is dependent on the voltage of the depolarizing pulse in a manner that indicates an open channel block.
To further determine whether blocking Kv1.3 channels with DPO-1 requires channel opening, we recorded currents in Jurkat cells before and after a pulse-free period of incubation with the compound. Control currents were elicited by repetitive application of depolarizing step pulses at +40 mV. DPO-1 (3 mM) was applied to the bath while the membrane potential was held at 280 mV without any depolarization pulses. After an interval of 8 min to assure a steady-state condition, consecutive 250 ms pulses were applied every 10 s. Figure 3C and 3D show that DPO-1 did not affect the amplitude peak current density in the pulse-free conditions, where the amplitude peak Kv1.3 current density at pulse 1 was 62.73 vs. 66.35 pA/pF under control conditions, but significantly accelerated the inactivation of Kv1.3 channels (the time constants of inactivation were 19662 ms under control and 13261 ms at pulse 1). The onset of blockade occurred rapidly once the channel was opened with depolarization pulses (the amplitude peak current at pulse 17 was 35.98 vs. 66.35 pA/pF at pulse 1). Moreover, the current recovered quickly upon washout even in pulse-free conditions. These results indicate that DPO-1 preferentially binds to the open state of the Kv1.3 channel.
A double-pulse protocol was used to investigate the effects of DPO-1 on voltage-dependent of steady-state inactivation of Kv1.3 currents in Jurkat cells. DPO-1 at 3 mM caused a negative shift of the steady-state inactivation curve ( Figure 3F). The half inactivation voltage V 1/2 were 24361 mV and 25361 mV for control and 3 mM DPO-1, respectively (P,0.001 vs. control). The decay phase of the Kv1.3 current at +40 mV was well fitted to a single exponential equation, yielding a time constant of 234628 ms in control, and 167617, 95612, and 6667 ms in 1, 3, and 10 mM DPO-1, respectively (P,0.001 vs. control for all of the concentrations) ( Figure 3E). Thus, DPO-1 accelerated the channel inactivation rate in a concentration dependent manner.

DPO-1 did not block K Ca channel currents in human T cells
T cells also express the calcium activated potassium channel K Ca , which plays a major role in regulating the membrane  [25]. K Ca 2.2 is the main K Ca channel involved in the Ca 2+ signaling of Jurkat cells [26], whereas K Ca 3.1 is the main K Ca channel in peripheral T cells [25]. Therefore, we investigated whether DPO-1 could also block K Ca currents in both these two kinds of human T cells. Figure 4A shows ramp currents in normal Ringer solution before and after application of DPO-1 in Jurkat cells. To avoid any confounding effects of the Kv currents, we measured the K Ca slope conductance between 2120 mV and 260 mV and fitted with the results as a linear function. DPO-1 at 3 mM had no apparent effects on the slope factor for K Ca (P = 0.709). To increase the slope conductance for K Ca and make it more accurate to measure, we used the high K + Ringer solution, which was also widely used in many studies [26], [27][28][29] Application of 1 mM TRAM-34, a selective K Ca channel blocker [29], robustly decreased the K Ca slope conductance ( Figure 4A and 4B). In contrast, DPO-1 at 3 mM had no effects on K Ca in both Jurkat cells and human peripheral blood T cells ( Figure 4C and  4D), which was also confirmed after the blockade Kv1.3 currents in the presence of MgTX ( Figure S1). The slope conductance for Jurkat cells was 1.3960.07 nS in control group, and 1.3960.09 nS in the DPO-1 group (P = 0.884). Likewise, slope conductance for human peripheral blood T cells was 1.6260.28 nS in the control group and 1.6260.29 nS in cells treated with DPO-1 (P = 1.000).The results for the control T cells were similar to previous reports [28], [30]. Therefore, our results suggest that DPO-1 had no effect on K Ca either in physiological or high K + external solution.

Kv1.3 channel current density and protein expression was inhibited by pre-treatment with DPO-1 in Jurkat cells
We further investigated the effects of 24 h DPO-1 pre-treatment on Kv1.3 channel currents and protein expression levels in Jurkat cells. Similar to acute administration, pre-incubation with DPO-1 also blocked the Kv1.3 channel current in a concentrationdependent manner. The current density at +40 mV decreased from 48.5262.31 pA/pF in control cells to 24.3762.21, 21.7762.21 and 18.5462.10 pA/pF in 1, 3, and 10 mM DPO-1 treated cells, respectively (P,0.001 vs. control for all concentrations) ( Figure 5A).

The Ca 2+ influx to Ca 2+ -depletion Jurkat cells was blunted by DPO-1
The Kv1.3 channel controls Ca 2+ homeostasis in T cells, and Kv1.3 channel inhibition significantly reduces Ca 2+ influx [10]. To confirm that DPO-1 inhibited Ca 2+ influx, we measured intracellular Ca 2+ concentration by laser confocal microscopy. Depletion of Ca 2+ from the endoplasmic reticulum was induced by 1 mM TG in Ca 2+ -free Ringer's solution. In the absence of extracellular Ca 2+ , TG induced a very small, transient intracellular Ca 2+ concentration rise in Jurkat T cells. When 2 mM CaCl 2 was applied to the Ca 2+ -free Ringer's solution, Ca 2+ flowed into the cell through the Ca 2+ release-activated Ca 2+ channel (CRAC) channel to significantly increase intracellular Ca 2+ , as illustrated in Figure 6A. Pre-treatment with DPO-1 for 24 h led to a concentration-dependent inhibition of Ca 2+ influx through the CRAC channel in Ca 2+ -depleted Jurkat cells,but did not alter TGinduced intracellular store Ca 2+ release ( Figure 6). The peak DF/ F 0 value representing Ca 2+ influx decreased from 4.17 in control to 2.4, 1.2, and 0.2, in 0.3, 1 and 3 mM DPO-1, respectively (P,0.01 vs. control for 0.3 and P,0.001 for 1 and 3 mM). Similar results were also found in the pre-treatment of DPO-1 for 30 min ( Figure S4).

DPO-1 inhibited the secretion of IL-2 in activated Jurkat cells
To determine whether Kv1.3 channel and Ca 2+ influx inhibition by DPO-1 could result in functional immunosuppression, we measured IL-2 secretion in the growth media by ELISA. As shown in Figure 7, IL-2 secretion was increased robustly upon stimulation with PHA+PMA (P,0.001), which served as the positive control. IL-2 levels were significantly reduced in the presence of 3 (P,0.05) and 10 (P,0.01) mM DPO-1 pre-treatment for 24 h. IL-2 secretion was also reduced with 10 nM MgTX and 100 nM ChTX, two known Kv1.3 blockers [31].

Discussion
This is the first study to investigate the effect of DPO-1, a novel Kv1.5 blocker, on potassium channels in human T cells. Our results demonstrated that DPO-1 blocked the Kv1.3 channel without any significant effect on the K Ca channel. Similar to our previous observation on the Kv1.5 channel, DPO-1 blocked the Kv1.3 channel in a voltage-and concentration-dependent manner, preferentially blocking the open channel state [1]. Pretreatment with DPO-1 for 24 h significantly reduced Kv1.3 current density and protein expression. Considering the importance of Kv1.3 in triggering Ca 2+ influx and IL-2 synthesis, we further investigated the effects of DPO-1 on Ca 2+ influx and IL-2 production. We found that DPO-1 inhibited Ca 2+ influx as well as IL-2 secretion. Therefore, our results suggested that in addition to inhibiting Kv1.5 channels, DPO-1 has marked inhibitory effects on Kv1.3 currents and its expression, Ca 2+ influx and IL-2 production.
DPO-1 blocked the Kv1.3 channel in a voltage-dependent manner, and this inhibitory effect increased steeply from voltages of 240 mV to 0 mV. Moreover, DPO-1 did not produce any obvious effect until the channel was first opened by a depolarization pulse, indicating that DPO-1 did not affect the Kv1.3 channel in the closed state. Combined, these effects suggest that DPO-1 interacts with the open state of Kv1.3 channels, which is the same mechanism previously observed for the Kv1.5 channel [1]. Of note, DPO-1 failed to cause a crossover phenomenon of tail current traces, which is in contrast with the properties of other open-channel blockers of the Kv1.3 channel [20], [32]. One possible explanation is that the functional dissociation of DPO-1 from the Kv1.3 channel at repolarization may not interfere with the transition of the channel from an open to a closed state. In addition, our data show that DPO-1 not only accelerated the time course of inactivation in a concentration-dependent manner but also shifted the voltage-dependent steady-state inactivation in the negative direction, indicating that DPO-1 also binds to the inactivated-state. Previous studies have shown that FK-506 [33] and verapamil [34], binding to both the open and inactivation states, caused significant negative shift of the voltage-dependence of steady-state inactivation. Unlike DPO-1, staurosporine [35] and fluoxetine [32], the pure Kv1.3 open channel blockers, had no effects on the voltage-dependence of steady-state inactivation. However, the interactions between DPO-1 and the inactivation state require more sophisticated kinetic analysis.
In addition to the Kv1.3 channel, human T cells express K Ca channels that help set up the membrane potential to drive Ca 2+ influx necessary for IL-2 synthesis [10], [25]. Our electrophysio-logical data shows that acute application of 3 mM DPO-1 blocked the Kv1.3 channel, but had little effect on the K Ca channel. Meanwhile, the membrane potential was depolarized from 241.0761.93 mV to 225.9962.91 mV. In human T cells, a small fraction of Kv1.3 channel is active at rest to maintain the membrane potential [25], [36]. Therefore, we assumed that DPO-1 could block Kv1.3 channel even in the rest potential range. In addition, membrane depolarization by blocking Kv1.3 has been confirmed to be an effective method to prevent T-cell activation and therefore has applications in many autoimmune conditions [11]. DPO-1 also significantly decreased Ca 2+ influx and IL-2 secretion in activated Jurkat cells, leading to the attenuation of the immune response. Consistent with other reports, both MgTX and ChTX blocked Kv1.3 to inhibit IL-2 secretion [31]. Furthermore, we also demonstrated that pre-treatment with DPO-1 for 24 h but not 30 min significantly reduced the channel protein expression in T cells. Unlike DPO-1, pre-treatment with MgTX for 24 h had no effect on Kv1.3 protein expression. In addition, pre-treatment with DPO-1 for 24 h had no effects on Kv1.3 mRNA expression ( Figure S5). These results suggested thatdirect and acute blockade of Kv1.3 channel by DPO-1 can lead to obvious inhibition of IL-2 secretion. Meanwhile, the inhibitory effect of DPO-1 on Kv1.3 protein expression may strengthen its immunosuppression potential. Overall, our data indicate that DPO-1, by blocking the Kv1.3 currents, down-regulating the Kv1.3 protein expression, and depolarizing membrane potential, blunts Ca 2+ influx and results in decreased production of IL-2.  Both Kv1.3 and K Ca channels play crucial roles in T cell activation. However, the relative contribution varies in certain T cell subtypes. The activation of T EM cells (the major pathogenic T cells in tissue sites of inflammation [37]) primarily depends on Kv1.3 channels, while K Ca channels have a major role in the activation of naïve and central memory T cells [38]. Importantly, potent and selective Kv1.3 inhibition preferentially prevents T EM function, whereas naïve and central memory T cells (necessary for physiological immune responses) escape inhibition by augmenting K Ca expression [27]. In animal experiments (rats and monkeys), effective dose of DPO-1 as a Kv1.3 blocker showed no apparent cardiac side effects, except for prolonging atrial refractory period [39]. Therefore, the Kv1.3 channel may be considered an attractive target for immunomodulatory therapies [18], [40].
In summary, our study provides the first evidence that Kv1.5 blocker DPO-1 suppresses Kv1.3 channels and exerts immunomodulatory effects by the inhibition of calcium influx and stimulation induced IL-2 production. Our results suggest that DPO-1 may be effective in inflammatory autoimmune diseases and benefits for the development of new drug on immunoregulatory therapy.