Effects of dalfampridine and its metabolites on cloned human potassium channels Kv 1.1, Kv 1.2, and Kv 1.4 expressed in human embryonic kidney cells

Background Dalfampridine (4-aminopyridine; 4-AP) is a potassium channel blocker that has been available in the United States as a treatment to improve walking in patients with multiple sclerosis. 4-AP is well-characterized in vitro with regard to inhibition of neuronal potassium channels, but the potential contribution of its metabolites to clinical activity has not been determined. This study evaluated the concentration–response of 4-AP and its two primary metabolites, 3-hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine sulfate, for inhibition of the potassium channels Kv 1.1, Kv 1.2, and Kv 1.4, which are considered candidates for mediating effects of 4-AP on action potential conduction because of their presence in axonal membranes. Methods Stable transfection of cDNA for Kv 1.1, Kv 1.2, and Kv 1.4 was performed into HEK293 cells, and colonies of cells containing each channel were selected and maintained under appropriate cell culture conditions. Electrophysiological measurements were performed using a patch-clamp technique in at least three cells for each concentration (50, 500, 5000, and 50,000 μM) of 4-AP and the two metabolites, with each cell acting as its own control. Concentration–response curves were constructed for 4-AP and each metabolite. Data were analyzed using nonlinear least-squares fit, and concentrations inhibiting the channels by 50% (IC50) were estimated. Results 4-AP induced similar concentration-dependent inhibition profiles of all three potassium channels, resulting in a narrow range of IC50 values across channels (242 µM to 399 µM). Across the three channels, the IC50 values of 3-hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine sulfate were 1–2 orders of magnitude higher (less potent) than those of 4-AP. Conclusions 3-Hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine sulfate demonstrated low in vitro potency for Kv 1.1, Kv 1.2, and Kv 1.4 inhibition, suggesting that these metabolites are unlikely to contribute to the positive pharmacodynamic effects of 4-AP. A limitation of this study is that while the metabolites were substantially less active at these representative potassium channels in vitro, the untested possibility exists that they may be active at one or more of the many other channel types that occur in vivo.


Introduction
Dalfampridine (4-aminopyridine; 4-AP) is a potassium channel blocker that has been studied extensively in the laboratory and in the clinic. An extended-release formulation of the drug, dalfampridine extended release, is available in the United States for the treatment of walking impairment in patients with multiple sclerosis (MS) 1 . This formulation is also approved for this use in several other countries where it is known as prolonged-, sustained-, or modified release fampridine. 4-AP has been well-characterized in vitro with regard to inhibition of a wide range of neuronal potassium channels, although the mechanism of action of 4-AP in MS has not been clearly established. While concentrations that result in 50% inhibition (IC 50 ) of these channels have been determined to be mostly in the millimolar range [2][3][4] , the average plasma concentration obtained with therapeutic dosing in clinical trials of the approved formulation ranged from 0.29 to 0.32 mM (27.6-30.2 ng/mL) 5,6 , which is 3-4 orders of magnitude lower than the typical concentrations used in in vitro laboratory studies to block potassium currents. However, the putative mechanism of action for its clinical effects is the relief of conduction block in demyelinated axons 7 , although it may also act at presynaptic sites, potentially enhancing neurotransmission through delay of repolarization and increased influx of calcium 8,9 . Several studies have described the pharmacokinetic characteristics of 4-AP in healthy volunteers and in the target population of people with MS [10][11][12][13][14] . In an excretion balance study in healthy volunteers using 14 C-radiolabeled 4-AP, elimination was almost exclusively (96%) by the renal route 10 . Although recovery in urine was mainly as unchanged parent compound, two primary metabolites were initially identified as 2-hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine based on approximate retention times using high performance liquid chromatography. Further characterization using established reference standards showed that the human metabolites of 4-AP were 3-hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine sulfate, the latter as a result of sulfate conjugation, and that they accounted for 510% of urinary excretion 15 . However, the potential contribution of these metabolites to the clinical activity of dalfampridine has not been determined.
Whereas K v 1.1 and K v 1.2 are voltage-gated potassium channels of the delayed rectifier type, K v 1.4 is a fast inactivating channel of the A-type (Table 1) 16 . These channels were considered suitable for evaluating the relative potency of the parent drug and its metabolites, and they are considered candidates for mediating effects of 4-AP on action potential conduction because of their presence in axonal membranes; K v 1.1, K v 1.2 and K v 1.4 are relevant components of axonal membrane heterotetrameric channels, and K v 1.4 is also a homotetrameric synaptic membrane channel. However, the IC 50 for these channels in vitro is much higher than the effective plasma concentration achieved with dalfampridine treatment. Therefore, the purpose of this study was to evaluate the concentration-response of 4-AP and its two primary metabolites for inhibition of three of the most common K v channels, K v 1.1, K v 1.2, and K v 1.4.

Methods
This study was conducted in accordance with good laboratory practice standards. All chemicals were obtained from

Cell cultures
Stable transfection of cDNA for each of the potassium channels was performed into HEK293 cells (American Type Culture Collection, Manassas, VA, USA), a human embryonic kidney cell line, as previously described 17 . After colony selection, the separate colonies of transfected HEK293 cells were cultured at 37 C in Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12 (D-MEM/F-12) supplemented with 10% fetal bovine serum, 100 U/mL penicillin G sodium, 100 mg/mL streptomycin sulfate and 500 mg/mL G418. Cells used for electrophysiology were plated in plastic culture dishes. Before testing, the cells were washed twice with Hank's balanced salt solution, treated with trypsin and re-suspended in culture medium at $1-1.5 Â 10 6 cells in 20 mL. Cells in suspension were allowed to recover for 1-3 hours with incubation at 37 C in a humidified 95% air/5% CO 2 atmosphere. Immediately before electrophysiological measurement, the cells were washed in 4-[2-hydroxyethyl]-1-piperazineethanesulfonic acid (HEPES)-buffered physiological saline (HB-PS).
Test samples were applied at 5-minute intervals via disposable polyethylene micropipette tips to naïve cells. Each solution exchange was performed in quadruplicate, and consisted of aspiration and replacement of 45 mL of the total 50-mL volume. The duration of exposure to each sample concentration was 5 minutes.

Electrophysiological measurements
Electrophysiological measurements were performed in at least three cells for each concentration using the PatchXpress system (Model 7000A, Molecular Devices LLC, Sunnyvale, CA, USA), at ambient temperature, with each cell acting as its own control. The methods for applying this commercially available patch-clamp system in the assessment of ion channels have been previously described 18 . In preparation for the recording, an intracellular solution consisting of 130 mM K-Asp, 5 mM MgCl 2 , 5 mM EGTA (ethylene glycol tetraacetic acid), 4 mM ATP (adenosine-5 0 -triphosphate), and 10 mM HEPES (pH to 7.2 with KOH) was loaded into the intracellular compartments of the Sealchip 16 planar electrode (Aviva Biosciences, San Diego, CA, USA). Membrane currents were recorded using dual-channel patch clamp amplifiers, and before digitization, signals were low-pass filtered at one-fifth of the sampling frequency.

Data analysis
Data acquisition and analysis were performed using pCLAMP software (Axon Instruments, Union City, CA, USA). Steady state was defined by the limiting constant rate of change with time (linear time dependence). The steady state before and after each application was used to calculate the percentage of current inhibited at each concentration. Results obtained from different cells were averaged and plotted as the mean AE standard deviation.
Concentration-response curves were constructed for each test compound, and data were fitted to the following equation: where % Inhibition represents the proportion of ion channel current inhibited at each concentration, IC 50 is the concentration resulting in 50% inhibition, and n is the Hill coefficient. Nonlinear least-squares fits were performed with the Solver add-in for Excel 2000 (Microsoft Corporation, Redmond, WA, USA), and the IC 50 was calculated. The kinetics of K v 1.4 channel inactivation were analyzed with the Clampfit 9.2 program (Molecular Devices LLC).

Results
Concentration-response curves for inhibition of the three evaluated potassium channels are presented in Figures 1-3 for 4-AP and its metabolites 3-hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine sulfate, respectively, and the calculated IC 50 concentrations are summarized in Table 2.
As shown in Figure 1, 4-AP induced similar concentration-dependent inhibition profiles of all three potassium channels. The percent inhibition at each concentration was generally comparable among the channels, and ranged from 20.2% to 26.9% at 50 mM, 54.5-63.0% at 500 mM, 74.4-83.1% at 5000 mM, and 86.5-94.4% at 50,000 mM. This similarity resulted in a narrow range of estimated IC 50 values across channels: 242 mM for K v 1.1 and 399 mM for both K v 1.2 and K v 1.4 (Table 2).
In contrast, 3-hydroxy-4-aminopyridine demonstrated a greater potency for inhibition of K v 1.1 than the other two channels (Figure 2). However, the IC 50 values of 3hydroxy-4-aminopyridine for K v 1.1 (7886 mM), K v 1.2 (23,652 mM) and K v 1.4 (23,191 mM) were greater than one order of magnitude higher than those of 4-AP   ( Table 2), indicating less potency. Similarly, the IC 50 values of 3-hydroxy-4-aminopyridine sulfate were approximately two orders of magnitude greater than those of 4-AP for K v 1.1 and K v 1.4 (Table 2). While the highest concentration of 3-hydroxy-4-aminopyridine sulfate (50,000 mM) resulted in 58.5% and 66.0% inhibition of K v 1.1 and K v 1.2, respectively (Figure 3), K v 1.4 was only inhibited by 27%. Overall inhibition of K v 1.4 by this metabolite was sufficiently low that a concentration-response curve was not derived.
Neither of the 4-AP metabolites demonstrated substantial effects on the kinetics of K v 1.4 inactivation ( Table 3). In contrast, inactivation was accelerated with 4-AP in a concentration-dependent manner, with enhancement of inactivation by 62.1% at 5000 mM; the percent change in inactivation was not determined at the highest 4-AP concentration of 50,000 mM.

Discussion
This is the first study to evaluate the pharmacodynamic properties of the two major 4-AP metabolites against these potassium channels. Although the inhibitory potential of 4-AP for various potassium channels has previously been assessed, this study re-evaluated 4-AP for specific activity using stably transfected K v 1.1, K v 1.2, and K v 1.4 channels.
The relevancy of the channels tested in this study was based on their presence in axonal membranes, with all three channels considered potential candidates for mediating the effects of 4-AP on action potential conduction since they are components of the juxtaparanodal heterotetrameric K v channels. In particular, K v 1.1 and K v 1.2 have a distribution in the internodal membrane that is consistent with changes in 4-AP sensitivity following demyelination 3,19 . However, in this as well as in previous studies, the sensitivity of these channels to 4-AP appears to be too low to correspond to the clinically relevant, submicromolar concentrations achieved with therapeutic dosing in patients with MS 3 , and the maximization of clinical benefit at plasma concentrations less than 0.5 mM 20 , suggesting the participation of other potassium channels in these effects.
The IC 50 values of 4-AP reported here for K v 1.1 (242 mM) and K v 1.2 (399 mM) were consistent with other studies reviewed by Judge et al. 3 Therefore, given the higher concentrations needed to block these channels in vitro compared with the upper range of plasma concentrations of approximately 0.92 mM (87.3 ng/mL) reported with dalfampridine at the recommended daily dose of 10 mg twice daily in the clinical trials 5,6 , these channels do not appear to be likely candidates for the observed clinical effects.
The IC 50 of 4-AP for K v 1.4, 399 mM (37,506 ng/mL), was similar to that for K v 1.1 and K v 1.2. This is lower than has previously been reported (IC 50 range of 647-13,000 mM) 3 . The reason for these differences is not clear but may be due to the source of the channels or the  expression systems that were used in the previous studies.
In particular, the study by Stühmer et al. 21 , which reported an IC 50 of 13,000 mM, used cloned and sequenced cDNAs that were isolated from a rat cortex cDNA library, with channel function evaluated by expressing the cloned channels in Xenopus laevis oocytes. They then determined the concentration dependence of inhibition on outward current using whole cell recordings at 20 mV. In contrast, the study reported here relied on stable transfection of cDNA into HEK293 cells, and membrane currents were recorded using dual channel patch clamp amplifiers. There are several potential reasons for the difference between therapeutic plasma concentrations and IC 50 values for in vitro potassium-channel inhibition. The first is that it is possible that 4-AP is a potent inhibitor of one or more potassium channels that have not yet been identified. Given that 4-AP is clinically effective at systemic concentrations that are well below the IC 50 values of the potassium channels tested so far 2 , it seems likely that this could be the case. Furthermore, how 4-AP concentrates in the microenvironment of a given K v channel pore is not known, and thus plasma levels may not reflect what a transmembrane channel may experience. An alternative possibility is that in vitro assessment of inhibition of a cloned channel might not accurately represent what occurs in the native environment in vivo. In particular, given that potassium channel multimers are found in vivo that have not been tested using in vitro cloned channels 2 , this is also feasible.
It should also be considered that, given the ubiquity of potassium channels throughout the tissues of the body, and particularly the nervous system, if 4-AP were not highly selective for the particular channels in the target tissue (presumably the demyelinated axonal membrane), it would be unlikely to have any clinical utility, given all the additional potential effects it might have on other parts of the neuromuscular system and cell types. There are data indicating that the sensitivity of potassium channels increases by at least 10-fold in chronically damaged central nerve fibers in the context of experimental spinal cord injury 22 , and it is possible that channel expression in demyelinated plaques is highly specific and uniquely sensitive to the low concentrations of 4-AP achieved with the therapeutic dose.
For the three representative potassium channels tested, the IC 50 values of the two primary metabolites of 4-AP (3hydroxy-4-aminopyridine and 3-hydroxy-4-aminopyridine sulfate) were in every case more than 30-fold higher (less potent) than the parent compound. Thus, these metabolites appear to be unlikely to contribute significantly to the primary pharmacodynamic effect of 4-AP, at least through the potassium channels evaluated in this study.
It is important to recognize an important limitation of this study that warrants conservative interpretation of the results. While the metabolites were substantially less active at these representative potassium channels in vitro, there remains the possibility that they may be active at one or more of the many other channel types that have been described to occur in vivo.

Conclusion
Compared with the parent drug, the two primary metabolites of 4-AP demonstrate low potency for blockade of the representative potassium channels K v 1.1, K v 1.2, and K v 1.4 in vitro. Thus, these metabolites are unlikely to contribute to the positive pharmacodynamic effects of 4-AP that may be observed during treatment in patients with MS.

Transparency
Declaration of funding This study was funded by Acorda Therapeutics, Inc, Ardsley, NY, USA.
Declaration of financial/other relationship A.C., A.B., and T.P. are employees and stockholders of Acorda Therapeutics, Inc, Ardsley, NY, USA.