2023 年 46 巻 10 号 p. 1394-1402
Dimenhydrinate, an H1 receptor antagonist, is generally used for the prevention and treatment of nausea and vomiting. However, cardiac arrhythmias have been reported to be associated with the overdose of histamine H1 receptor antagonists, indicating the probable effect of antihistamines on ion channels. By using a two-microelectrode voltage clamp, we have herein studied the electrophysiological effects of dimenhydrinate on the human Kv1.5 channel in the Xenopus oocyte expression system. Dimenhydrinate acutely and reversibly suppressed the amplitudes of the peak and the steady-state current, within 6 min. The inhibitory effect of dimenhydrinate on the peak and the steady-state Kv1.5 currents increased progressively from −10 to +50 mV. At each test voltage, the drug suppressed both the peak and the steady-state currents to a similar extent. When the oocytes were stimulated at the rates of 5- and 30-s intervals, dimenhydrinate-induced a use-dependent blockade of the human Kv1.5 channel. Dimenhydrinate expedited the timecourse of the Kv1.5 channel activation more effectively than the timecourse of its inactivation. However, the activation and inactivation curves of the channel were not altered by the H1 receptor antagonist. In conclusion, we found that dimenhydrinate inhibits the human Kv1.5 channel by changing the channel’s activation mode, thereby possibly increasing the possibility of triggering cardiac arrhythmias and affecting atrial fibrillation.
Dimenhydrinate (Fig. 1) is an ethanolamine antihistamine used for the prevention and treatment of motion sickness, nausea, and vomiting.1) Dimenhydrinate is a combination of two drugs: diphenhydramine and 8-chlorotheophylline (a xanthine derivative that decreases the sedating properties of diphenhydramine) in a 1 : 1 ratio.2) Diphenhydramine is a first-generation antihistamine drug and an inverse agonist that binds to and stabilizes the inactive conformation of the histamine H1 receptor.3) It is known to interact with many neurotransmitters,4) while an overdose of diphenhydramine induces anticholinergic effects.1) Tachycardia and deaths as a result of cardiac arrhythmias and heart attacks have been reported in association with diphenhydramine.1,5) Diphenhydramine can inhibit Na+ channels and the rapidly-activating delayed rectifier K+ current (IKr), thereby causing prolongation of the action potential duration (APD) and the QT interval.6) In contrast, to our knowledge, there have been no reports of 8-chlorotheophylline changing any ion channels in human or animals. Considering the significant number of reports regarding the effects of misused OTC dimenhydrinate for medical and non-medical purposes1,4) as well as the fact that diphenhydramine is the main functional constituent of dimenhydrinate, the mechanisms underlying the risks associated with the overdose, poisoning, and abuse of dimenhydrinate should be investigated.
Voltage-dependent potassium channels (Kv) are a superfamily of 12 subfamilies that control the resting membrane potentials and APDs.7) The membrane’s depolarization usually opens the Kv channels that are closed at resting potentials.8) These channels may serve as targets for neurotransmitters, hormones, and class III antiarrhythmic drugs that prolong the APD and refractoriness.9) Kv1.5 channels are part of the Shaker family (Kv1), and the Kv1.5 currents facilitate the repolarization of action potentials via the ultra-rapid delayed rectifier K+ current, IKur, in the heart.7) The Kv1.5 channel consists of a homotetramer of the α-subunit (that can form a pore), and is abundantly present in arterial smooth muscle cells,10) and the human atrial myocytes.11) Olson et al. have reported that both in vitro and in vivo, a Kv1.5 deficiency can cause atrial fibrillation, defective atrial electrical instability, and arrhythmias.12) On the other hand, gain-of-function in Kv1.5 channel enhance atrial fibrillation susceptibility.13,14)
We have herein, assessed dimenhydrinate for its acute effects on the human Kv1.5 channel currents, and for the underlying biophysical mechanisms of the drug’s action by employing the Xenopus oocyte expression system. After considering the pharmacological risk of cardiac arrhythmias associated with the use of dimenhydrinate, the present study has clinical implications by demonstrating that dimenhydrinate might be able to cause cardiac arrhythmias and affecting atrial fibrillation, possibly through the inhibition of the human Kv1.5 channel.
Message Machine T7 kits (Ambion, Austin, TX, U.S.A.) were used to prepare human Kv1.5 (hKv1.5, GenBank: BC099665.3) cRNA in vitro transcription to be stored later in a nuclease-free water medium (−80 °C). Oocytes (Stages V and VI) were extracted surgically from female Xenopus laevis (Nasco, Modesto, CA, U.S.A.) after being anesthetized with ice for 30 min at 10 min intervals. Theca and follicle layers were eliminated from oocytes with fine forceps. After two days, 20 nL cRNA (0.4 µg/µL) was injected into Xenopus oocytes. All previous steps were conducted under the Medical Research Guidelines of Kangwon National University IACUC. The injected Xenopus oocytes were preserved under the following conditions, temperature equal to 17 °C, modified Barth’s solution, consisting of (mM): 1 KCl, 88 NaCl, 10 N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES), 0.4 CaCl2, 2.4 NaHCO3, 1 MgSO4, 0.33 Ca(NO3)2 (pH 7.4), and 50 µg/mL gentamicin sulfate. Later, the current was measured 4 to 5 d post-injection.
Ethics StatementAll animal procedures were conducted under the Medical Research Guidelines of Kangwon National University IACUC.
Voltage-Clamp Recordings from Oocytes and SolutionsND96 solution was composed of (mM): 10HEPES, 2KCl, 1.8CaCl2, 1MgCl2, and 96NaCl (pH 7.4). The chamber, where the oocytes were preserved, was constantly perfused with ND96 solution as the interchanges were completed within 3 or 4 min. After 6–12 min of solution exchange, we measured the currents at room temperature (20–23 °C) using OC-725C two-microelectrode voltage clamp amplifier (Warner Instruments, Hamden, CT, U.S.A.). Regarding electrode preparation, they were filled with 3M KCl with a resistance of 2.5–4, and 2–2.5MΩ for voltage-recording and current-passing electrodes, respectively. Conversely, AD–DA converter (Digidata 1200, Axon Instruments) and pCLAMP software (v5.1, Axon Instruments) were used for stimulation and data acquisition. A dimenhydrinate stock solution (Fig. 1) was produced in dimethyl sulfoxide (DMSO) and added to the external solution at a predetermined concentration shortly before each experiment. On the other hand, Sigma provided the dimenhydrinate and other reagents.
Data AnalysisThe software used for data analysis was Origin 8.0 (OriginLab Corporation, Northampton, MA, U.S.A.). We fitted the current trace of activation and inactivation phases in a single exponential function, considering it as the dominant time constant. On the other hand, the data related to concentration-dependent current inhibition were fitted to a Hill equation:
 |
IC50 is the concentration at which half-maximal peak currents were inhibited and [D] is the dimenhydrinate concentration. In contrast, fitting the data to a Boltzmann function yielded steady-state activation curves. The equation was as follows:
 |
V is the test potential, V1/2 is the half-activation potential (voltage at which the conductance was half-activated) and k is the slope factor.
Means ± standard error of the mean (S.E.M.) were used to present the data. “n” Represents the number of experiments. Statistical significance was tested with one-way ANOVA followed by Tukey’s post hoc tests. p < 0.05 was considered a statistically significant value.
By using the expression system of the Xenopus oocyte, dimenhydrinate was tested for its effect on Kv1.5 currents. Each concentration of dimenhydrinate was tested on the same individual oocyte. Figure 2A presents the steady-state and peak currents of the human Kv1.5 channel, as measured in Xenopus oocytes under control conditions and an exposure to 100 µM of dimenhydrinate for 6 and 12 min. Our results indicate that the drug suppressed both the peak and the steady-state currents within 6 min. The concentration-dependent reduction of the Kv1.5 channel currents was observed after an exposure to a dimenhydrinate concentration ranging between 100 and 1000 µM (Fig. 2B). The peak current at a +50-mV depolarizing pulse was 64.7 ± 2.2 and 61.5 ± 1.9% of that of the control after a treatment with 100 µM of dimenhydrinate for 6 and 12 min, respectively (n = 5–9; Figs. 2B, C). As far as the steady-state current is concerned, dimenhydrinate (at the same concentration) reduced the current to 62.0 ± 2.7 and 58.1 ± 2.0% of that of the control after a 6- and a 12-min treatment, respectively (n = 5–9; Figs. 2D, E). These findings indicate that a similar decline in the peak current and the steady-state current can be achieved within 6 min of an exposure to dimenhydrinate. Moreover, when different concentrations of dimenhydrinate were applied, we found that the currents were inhibited in a concentration-dependent manner (Figs. 2F, G). The human Kv1.5 channel peak currents were inhibited by dimenhydrinate for 6 and 12 min with IC50 values of 133.8 ± 7.4 and 125.2 ± 7.9 µM, respectively (n = 5–9; p > 0.05 when comparing the valuses obtained at 6 and 12 min after drug treatment; Fig. 2F). The IC50 values for the dimenhydrinate-induced blockade of the human Kv1.5 steady-state currents for 6 and 12 min were 109.3 ± 5.4 and 100.2 ± 4.7 µM, respectively (n = 5–9; p > 0.05 when comparing the valuses obtained at 6 and 12 min after drug treatment; Fig. 2G). Hence, dimenhydrinate blocked human Kv1.5 channels concentration-dependently, irrespective of the time period for which the treatment lasted (6 or 12 min).
(A) Superimposed current traces were compared before (control) and after an exposure to 200 µM of dimenhydrinate for 6 and 12 min, with voltage pulses (of 2-s duration) between −50 to +50 mV, and 10-mV increments every 10 s starting from −60 mV (holding potential). (B–E) Current-voltage (I–V) relationship of the peak and the steady-state currents associated with the Kv1.5 channel, under control conditions and after an exposure to different concentrations of dimenhydrinate for 6 min (B, D) or 12 min (C, E). At the end of the depolarizing pulses, the peak currents were measured at peak and steady-state currents were determined at the end of depolarizing pulses. Then, the peak and steady-state currents were normalized to their respective values at +50 mV before applying the drug. (F, G) Analyses of the concentration–response inhibition curves of the peak current (F) and the steady-state current (G) for 6 and 12 min. Symbols with error bars represent the mean ± standard error of the mean (S.E.M.) of the recorded data (n = 5–9).
The dimenhydrinate-induced blockade of the Kv1.5 currents has been examined for reversibility. As shown in Figs. 3A and B, the Kv1.5 channel late current (Ilate) was reduced immediately after administration of 200 µM dimenhydrinate containing ND96 and decreased progressively. After the reduction of the late current to 48.6% by the drug, a washing for 25 min managed to reverse the inhibition to 87.1% of the value observed just before the drug treatment. A similar result was observed in five experiments. For the same time-period, we also measured the current without applying the drugs (see the plot in Fig. 3). The results of a 45-min continuous recording have demonstrated that the late current at the beginning decreased to 96.4%, thereby indicating that the current inhibition due to the drug cannot be attributed to the observed run-down of the currents.
(A) Recording example of a late current before and after washing out the 200 µM of dimenhydrinate. Starting from the holding potential (−60 mV), the currents were induced by a 200-ms depolarizing pulse at +30 mV with 10-s intervals. (B) The altered late currents after the exposure to dimenhydrinate were normalized to their respective values in the control state before the exposure. Black scatters represent the control late current at control conditions (45 min of current recordings without an exposure to the drug). All values were normalized to the first late current (n = 5).
With a single exponential function, the activation and the inactivation phases of the Kv1.5 channel currents were fitted so as to determine the effects of dimenhydrinate (Fig. 4). An exposure to 100, 200, and 500 µM dimenhydrinate for 12 min decreased the activation time constant (τ) at +50 mV to 57.7 ± 5.6, 41.9 ± 9.0, and 41.1 ± 1.7% of control value, which were significant changes when compared with control (n = 5–9; p = 0.04026, 0.00994, 0.00193, respectively; Figs. 4A, B). These results indicated that dimenhydrinate can progressively accelerate the channel activation rate in all concentration range tested. On the contrary, only the 500 µM concentration (applied for 12 min) of dimenhydrinate significantly decreased the τ of the inactivation by the test pulse of +50 mV to 67.2 ± 7.6% of control value, which was a significant change when compared with control (n = 5–9; p = 0.00138; Figs. 4C, D), thereby indicating that the inactivation timecourse was made faster only by the highest concentration we tested. As a result, dimenhydrinate was able to shorten the Kv1.5 channel activation timecourse more efficiently than the inactivation timecourse.
The time constants of the current activation and inactivation were estimated by fitting single exponential functions to traces induced by a single +50 mV pulse with 2 s duration from a holding potential of −60 mV. (A, C) Representative normalized current traces of the activation (A) and the inactivation (C) phase in the absence of dimenhydrinate (dark) and in the presence of 100, 200, and 500 µM of dimenhydrinate for 12 min (colored). Each current trace was normalized to its peak value. (B, D) Summary of the normalized activation (B) and the inactivation (D) time constants. Data presented as mean ± S.E.M. (n = 5–9 oocytes per concentration). *: p < 0.05 compared with control.
In order to determine whether the antihistamine effects are voltage-dependent, the observed dimenhydrinate-induced suppression of the Kv1.5 currents was assessed at different test voltages (Fig. 5). Figure 5A presents the superimposed traces for the 100-µM dimenhydrinate-induced Kv1.5 currents recorded at depolarizing pulses of −10, 0, +10, and +50 mV. During 12 min of a 100-µM dimenhydrinate treatment, the Kv1.5 peak currents were inhibited by 11.9 ± 4.8, 23.8 ± 1.8, 28.8 ± 1.7, 32.1 ± 1.9, 34.7 ± 1.9, 36.7 ± 1.9, and 38.5 ± 1.9% in the case of the peak currents (n = 5–9; Fig. 5B), showing that the blockade increases more sharply during the channel opening voltage range from −10 to 0 mV, and follows a gentle slope when the voltage range exceeds 0 mV. The dimenhydrinate-induced reduction in steady-state current exhibited similar characteristics (n = 5–9; Fig. 5C). The depolarization-dependent dimenhydrinate-induced inhibition of the peak and the steady-state currents increased biphasically across a range of test voltages from −10 to +50 mV (n = 5–9; Figs. 5B, C), thereby demonstrating that dimenhydrinate has attenuated the Kv1.5 channel currents in a voltage-dependent manner. However, the voltage-dependent suppression did not differ significantly between the peak and the steady-state currents at drug concentrations ranging 100–1000 µM (n = 5–9; p > 0.05 when comparing the valuses of the peak and the steady-state currents; 200 µM shown in Fig. 5D; data regarding the 100-, 500-, and 1000-µM concentrations are not shown).
(A) Top: Currents were induced by a 2-s depolarizing pulse (at −10 to 50 mV), starting from a holding potential equal to −60 mV. (A) Bottom: Superimposed current traces under control conditions (before exposure) and after an exposure to 100 µM of dimenhydrinate and different voltage values (namely, −10, 0, +10, and +50 mV). (B, C) Dimenhydrinate-induced Kv1.5 peak (B) and steady-state current (C) inhibitions at various voltages. At each voltage value, the currents under the influence of different dimenhydrinate concentrations (namely, 100, 200, 500, and 1000 µM) were normalized to the one of the control state. Current inhibition (%) = 100 × (control current - drug condition current)/control current. (B, C) Asterisk (*) and daggers (†) represent the statistical significance of the difference between the current inhibition at each test voltage and the inhibition at −10 and 0 mV, respectively. (D) Comparison of the 200-µM dimenhydrinate-induced Kv1.5 peak and steady-state current inhibition at various voltages. Symbols with error bars represent the mean ± S.E.M. of the recorded data (n = 5–9). *,†: p < 0.05.
Subsequently, two-pulse protocols were used in order to evoke tail currents and to determine whether dimenhydrinate would affect the activation and inactivation gatings. By fitting two different Boltzmann equations to normalized tail currents, the steady-state activation and inactivation curves were obtained (Fig. 6). The V1/2 values of the activation curves were −7.75 ± 1.53, −8.39 ± 1.33, and −10.0 ± 1.39 mV, with k values of 9.06 ± 1.44, 8.59 ± 0.88, and 6.85 ± 0.92 after an exposure to 0, 100 and 200 µM of dimenhydrinate, respectively (n = 6 − 7; p > 0.05 when comparing the V1/2 values and k valuses obtained at 0, 100, and 200 µM of dimenhydrinate; Figs. 6A, B). These findings suggest that dimenhydrinate did not shifted the steady-state activation curves. In addition, the V1/2 values in the inactivation curve were −15.2 ± 0.41, −16.7 ± 0.47, and −18.0 ± 0.57 mV, with k values of 3.60 ± 0.28, 3.73 ± 0.37, and 3.41 ± 0.54 after an exposure to 0, 100, and 200 µM of dimenhydrinate, respectively (n = 5–8; p > 0.05 when comparing the V1/2 values and k valuses obtained at 0, 100, and 200 µM of dimenhydrinate; Figs. 6C, D). These findings suggest that dimenhydrinate has failed to change the Kv1.5 channel’s steady-state inactivation curves.
(A) Representative steady-state activation tail currents were recorded at −40 mV (indicated by the dotted arrow under the voltage protocol) after 100 ms of depolarizing pulses (−60 to +50 mV) in the control state (before the exposure to the drug) and after an exposure to 100 and 200 µM of dimenhydrinate for 12 min. (B) Steady-state activation curves were generated by normalizing to the tail amplitude to the maximum value when depolarized, and by fitting the data with a Boltzmann equation. Symbols with error bars represent the mean ± S.E.M. of the recorded data (n = 6–7). (C) Representative tail current traces were induced by using 250-ms depolarizing pulses of +50 mV (indicated by the dotted arrow under the voltage protocol); 30-s preconditioning pulses were produced at −60 to +20 mV in the control state (before the drug exposure) as well as after an exposure to 100 and 200 µM of dimenhydrinate for 12 min. (D) Steady-state inactivation curves were generated by normalizing to peak amplitude when depolarized to +50 mV, by fitting the data with a Boltzmann equation. Symbols with error bars represent the mean ± S.E.M. of the recorded data (n = 8).
Finally, a depolarizing step of 0.2 s to +50 mV at 5- or 30-s intervals was used in order to test whether dimenhydrinate would inhibit the Kv1.5 channels in a frequency-dependent manner (Fig. 7). We performed control experiments to determine the use-dependent effect on Kv1.5 current in oocytes without dimenhydrinate. As shown in Fig. 7A, the Kv1.5 channel peak currents at 180 s after the beginning of depolarizing step in control conditions did not change significantly by changes in stimulation rates of 5- and 30-s intervals (n = 5; p > 0.05). However, the application of 200 µM dimenhydrinate for 180 s significantly changed the peak currents from 96.8 ± 0.8% of control at stimulation rates of 30-s intervals to 87.0 ± 0.9% of control at stimulation rates of 5-s intervals, respectively (n = 4; p = 0.00018 when comparing the values obtained at 180 s after drug treatment at stimulation rates of 5- and 30-s intervals). Then we subtracted the values of the data in the presence of dimenhydrinate from the values of the control data in which dimenhydrinate was absent in perfusing solution (Fig. 7B). Dimenhydrinate induced the human Kv1.5 channel block more strongly at higher stimulation frequencies than at lower stimulation frequencies at the same time after drug treatment; dimenhydrinate inhibited the Kv1.5 current 21.9 ± 0.9 and 11.9 ± 0.8% at 180 s after the drug treatment at stimulation rates of 5- and 30-s intervals, respectively (n = 4–5; p = 0.00016 when comparing the values obtained at 180 s after drug treatment at stimulation rates of 5- and 30-s intervals). Theses results indicated that dimenhydrinate can bind to the Kv1.5 channel more strongly at higher stimulation frequencies, showing a strong use-specific dependency associated with the dimenhydrinate-induced inhibition of the Kv1.5 channels.
(A) Forty, and twenty repeated 200-ms depolarizing pulses (at +50 mV) starting from a holding potential (at −80 mV) at 5 and 30 s in the control state (before the exposure to the drug) and after an exposure to 200 µM of dimenhydrinate for 2 min (when the test solution is completely changed into the chamber). All the currents were normalized to control current obtained at the first depolarizing pulse in the absence of dimenhydrinate. (B) Dimenhydrinate-induced inhibition of Kv1.5 peak current obtained from panel (A). Current inhibition (%) = 100 × (control current-drug condition current)/control current. Symbols with error bars represent the mean ± S.E.M. of the recorded data (n = 4–5).
To clarify whether the human Kv1.5 channel blocking the action of dimenhydrinate is due to diphenhydramine or chlorotheophylline in the mixture, we have performed the additional experiments with the two components of dimenhydrinate (Fig. 8). Treatment of diphenhydramine at 100 and 500 µM for 6 min decreased the Kv1.5 peak current at a +50-mV depolarizing pulse concentration-dependently to 81.7 ± 1.8 and to 44.8 ± 2.8% of that of the control, respectively, as shown in Fig. 8B (n = 5; p = 3.89415E-6 when comparing the valuses obtained at 100 and 500 µM; Fig. 8B). However, chlorotheophylline at the same concentrations did not change the peak current significantly (n = 4; p > 0.05 when comparing the valuses obtained at control with the value obtained at either 100 or 500 µM; Fig. 8C). Similar results were obtained in steady-state currents. Diphenhydramine at 100 and 500 µM for 6 min decreased the Kv1.5 steady-state current at a +50-mV depolarizing pulse concentration-dependently to 80.3 ± 2.9 and to 50.2 ± 5.3% of that of the control, respectively (n = 5; p = 0.00102 when comparing the valuses obtained at 100 and 500 µM; Fig. 8D). However, chlorotheophylline at the same concentrations failed to change the steady-state current significantly (n = 4; p > 0.05 when comparing the valuses obtained at control with the value obtained at either 100 or 500 µM; Fig. 8E). Therefore, the Kv1.5 channel blocking the action of dimenhydrinate would be due to diphenhydramine rather than chlorotheophylline in the mixture.
(A) Superimposed current traces were compared before (control) and after an exposure to 100 µM of either diphenhydramine or 8-chlrorotheophyline for 6 min, with voltage pulses (of 2-s duration) between −50 to +50 mV, and 10-mV increments every 10 s starting from −60 mV (holding potential). (B) Current-voltage (I–V) relationship of the Kv1.5 channel peak currents, under control conditions and after an exposure to 100 and 500 µM of diphenhydramine for 6 min. (C) Current-voltage (I–V) relationship of the Kv1.5 channel peak currents, under control conditions and after an exposure to 100 and 500 µM of 8-chlorotheophylline for 6 min. (D) Current-voltage (I–V) relationship of the Kv1.5 channel steady-state currents, under control conditions and after an exposure to 100 and 500 µM of diphenhydramine for 6 min. (E): Current-voltage (I–V) relationship of the Kv1.5 channel steady-state currents, under control conditions and after an exposure to 100 and 500 µM of 8-chlorotheophylline for 6 min. Symbols with error bars represent the mean ± S.E.M. of the recorded data (n = 4–5).
Dimenhydrinate, one of the ethanolamine antihistamines, is a widely used OTC drug. It can cause frequent unifocal ventricular ectopic beats, and can increase the heart rate responses to baroreceptor unloading.15,16) Moreover, diphenhydramine, one of the components of dimenhydrinate,17) can seriously affect the QT interval.5) Therefore, dimenhydrinate may change ion currents involved in cardiac rhythm. In this study, dimenhydrinate was electrophysiologically assessed for its effects on human Kv1.5 currents; an important determinant for the regulation of the cardiac action potential.
By employing the Xenopus oocyte’s expression system, we found that dimenhydrinate inhibits the human Kv1.5 channel currents quickly and reversibly, depending on its use and its concentration (Figs. 2, 3, 7). The depolarization-dependent dimenhydrinate inhibition of the peak and steady-state currents increased biphasically across a range of test voltages (from −10 to +50 mV; Fig. 5), thereby demonstrating that dimenhydrinate was able to attenuate the Kv1.5 channel currents in a voltage-dependent manner. A voltage-dependent blockade of the Kv1.5 by dimenhydrinate did not vary when compared between the peak and the steady-state currents, thereby confirming that the drug exerted blocking effects on both the open-state and the closed-state of the Kv1.5 channel (Fig. 5). However, the blockade of the Kv1.5 by dimenhydrinate exhibited a use-specific dependency, with a higher inhibition observed at the stimulation frequency of 5 s rather than at 30 s (Fig. 7); this finding indicates that dimenhydrinate can exert a more inhibitory effect upon the open-state than the closed-state of the channel. As far as the biophysical mechanisms underlying the dimenhydrinate-induced blockade of the Kv1.5 channel are concerned, dimenhydrinate progressively expedited the rates of the channel activation more effectively than those of the inactivation (Fig. 4). However, dimenhydrinate failed to affect its steady-state activation and inactivation curves (Fig. 6). These results suggest that dimenhydrinate could affect the gating process of the channel opening rather than its closing. The dimenhydrinate-induced suppression of the human Kv1.5 channel occurred acutely, and the inhibition was reversible (Fig. 3); a finding indicating the drug’s direct action on the human Kv1.5 channel. Finally, it is more likely that diphenhydramine rather than 8-chlorotheophylline in the dimenhydrinate mixture mainly reduced the human Kv1.5 currents (Fig. 8).
The plasma concentration of diphenhydramine ranged 25–50 ng/mL, equivalent to 0.09–0.17 µM diphenhydramine, which would correspond to 0.17–0.34 µM dimenhydrinate.18) Moreover, when 50 mg of diphenhydramine—the appropriate adult oral dose—was orally administered, peak plasma concentrations were 80–100 ng/mL corresponding to 1.1–1.4 µM dimenhydrinate.19,20) The lethal serum concentration of diphenhydramine has been defined as 5 µg/mL, which corresponds to 137.5 µM dimenhydrinate.21) This concentration is comparable with the IC50 of 100 µM approximately for the human Kv1.5 blockade by dimenhydrinate in the present study (Fig. 2). In addition, owing to the accessibility of dimenhydrinate, tolerance and dependency on the drug have been reported along with potential euphoric effects accompanying its heavy use (>100 mg/d).22) Nishino et al. found 18.7 µg/mL diphenhydramine concentration in the sreum of a 45-year-old man who died after an overdose of the drug, which corresponds to approx. 514 µM of dimenhydrinate.23) Therefore, the present study could suggest that dimenhydrinate could induce the blockade of human Kv1.5 channel in cases of an overdose through poisoning and resistance due to the drug addiction and misuse.
Terfenadine and rupatadine, second-generation antihistamines, have been shown to block the human Kv1.5 channels in a concentration-dependent manner.24,25) In fact, a biphasic voltage-dependent blockade of the channel by these drugs occurred: the drugs inhibited the channel steeply over a range of channel opening voltages (−30 to 0 mV), and gradually at positive to 0 mV voltages (where the channel is fully open).24,25) Similarly, Fig. 5 shows that the voltage dependency for the dimenhydrinate-induced blockade increases more sharply during the channel opening voltage range (−10 to 0 mV), and follows a gentle slope when the voltage range exceeds 0 mV. These dual-phase properties can be observed during the inhibitory action of various classes of open channel blockers, such as anorexinogen agents, lipoxygenase inhibitors, vasodilators, and angiotensin II receptor type 1 antagonists.26–29) The putative binding sites for the rupatadine and the lipoxygenase inhibitor appeared to be located in the external entryway of the Kv1.5 channel, as mutated forms of the channel (namely, the R485A and the R487V) exhibited a non-biphasic voltage-dependent blockade and a significantly reduced inhibition when compared with the wild-type channel, respectively.25,29) Therefore, the present study suggests that dimenhydrinate, a first-generation antihistamine, can act as an open channel blocker of the human Kv1.5 channel, contributing to various cardiac side effects (such as frequent unifocal ventricular ectopic beats and increased heart rate responses to baroreceptor unloading).1,5) The hKv1.5 blocking effect of dimenhydrinate could synergistically increase the APD since diphenhydramine has been shown to inhibit human ether-a-go-go related gene (hERG) channel.6,30) In addition, a potential treatment strategy for atrial fibrillation is the development of drugs targeting human Kv1.5, which inhibits IKur,31) because IKur inhibition can be used to treat atrial fibrillation by prolonging the APD of the patient.32,33) Therefore, the present study can provide structural information on a possible therapeutic drug in selected subgroups of patients with atrial fibrillation.
In the present study, dimenhydrinate acutely and reversibly blocked the human Kv1.5 channel within 6 min. Dimenhydrinate accelerated the velocity of the channel’s activation and inactivation, however, the drug did not affect the activation and inactivation gating. The voltage and the frequency dependency of the dimenhydrinate-induced Kv1.5 channel blockade indicated that dimenhydrinate could act as an open channel blocker. In conclusion, when considering the serious risk associated with the frequent abuse of dimenhydrinate, the present study has clinical implications by demonstrating that dimenhydrinate, particularly in overdose, may induce cardiovascular toxicity through the inhibition of the human Kv1.5 channel.
The authors wish to thank Prof. Han Choe (Department of Physiology, Bio-Medical Institute of Technology, University of Ulsan College of Medicine, Seoul, Korea) for providing the human Kv1.5 gene.
This work was supported by the National Research Foundation of Korea (NRF) Grant funded by the Korea Government (MSIT) (No. RS-2023-00250981) and by the Research Grant from Institute of Medical Sciences, Kangwon National University 2023.
Seo-In Park: Investigation, Formal analysis, Writing—Original Draft. Soobeen Hwang: Formal analysis, Writing—Original Draft. Jong-Hui Kim: Formal analysis, Writing—Original Draft. Se-Ran Yang: Interpretation of data. Su-Hyun Jo: Conceptualization, Visualization, Writing—Review and Editing, Supervision
The authors declare no conflict of interest.
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.
