Regulation of Kv4.2 A-Type Potassium Channels in HEK-293 Cells by Hypoxia

We previously observed that A-type potassium currents were decreased and membrane excitability increased in hippocampal dentate granule cells after neonatal global hypoxia associated with seizures. Here, we studied the effects of hypoxia on the function and expression of Kv4.2 and Kv4.3 α subunit channels, which encode rapidly inactivating A-type K currents, in transfected HEK-293 cells to determine if hypoxia alone could regulate IA in vitro. Global hypoxia in neonatal rat pups resulted in early decreased hippocampal expression of Kv4.2 mRNA and protein with 6 or 12 h post-hypoxia. Whole-cell voltage-clamp recordings revealed that similar times after hypoxia (1%) in vitro decreased peak currents mediated by recombinant Kv4.2 but not Kv4.3 channels. Hypoxia had no significant effect on the voltage-dependencies of activation and inactivation of Kv4.2 channels, but increased the time constant of activation. The same result was observed when Kv4.2 and Kv4.3 channels were co-expressed in a 1:1 ratio. These data suggested that hypoxia directly modulates A-type potassium channels of the subfamily typically expressed in principal hippocampal neurons, and does so in a manner to decrease function. Given the role of IA to slow action potential firing, these data are consistent with a direct effect of hypoxia to decrease IA as a mechanism of increased neuronal excitability and promotion of seizures.


INTRODUCTION
Voltage-dependent potassium (Kv) channels have a role in many developmental nervous system diseases, such as learning and cognitive impairment and epilepsy (Lawson, 2000;Wulff et al., 2009). A-type potassium channels are abundantly expressed in neurons and serve a number of functions, including regulation of excitability, fast-spiking, neurotransmitter release, and control of neural networks in physiological and pathophysiological processes (Birnbaum et al., 2004). A-type currents (I A ) exhibit rapid activation, inactivation, and rapid recovery from inactivation at hyperpolarized membrane potentials (Jerng et al., 2004).

Frontiers in Cellular Neuroscience
www.frontiersin.org We previously reported that A-type K + currents were significantly decreased and membrane excitability was increased in hippocampal dentate granule cells after seizure-inducing hypoxia in vivo (Peng et al., 2013). However, the mechanisms that mediated this hypoxic inhibition remain to be elucidated. In the current study, we therefore investigated the mRNA and protein expression levels of the Kv4.2 and Kv4.3 after hypoxia treatment in rats. We further expressed Kv4.2 and co-expressed Kv4.2/Kv4.3 in HEK-293 cells to investigate a potential direct influence of hypoxia on the function of A-type channels.

IN VIVO HYPOXIA TREATMENT
Animals were obtained from the Animal Biosafety Level 3 Laboratory (ABSL-3) of Wuhan University. All experimental protocols were approved by the Committee on the Ethics of Animal Experiments of the Wuhan University (China). Rat pups on postnatal day 10 were removed from the litter and placed on a heating pad in a custom-made acrylic glass chamber with an O 2 sensor mounted inside and three gas inlets for N 2 infusion (Peng et al., 2013). For the treated pup, the chamber O 2 concentration was lowered by N 2 infusion to 7% within 30-40 s, maintained at 6-7% for an additional 4 min, lowered to 5-6% for 8 min, and then lowered by 1% per minute until the pup became apneic for 30 s, at which time the chamber lid was removed for exposure to room air. The total time of hypoxia exposure was 14-16 min. Using this approach, hypoxia-treated pups typically exhibited spontaneous convulsive seizures lasting 10-60 s beginning 2-4 min after hypoxia onset, which recurred throughout hypoxia exposure and continued for several minutes after return to room air. Littermate controls were kept at room air and maintained body temperature by heating pad. All rats were immediately returned to their dam after the experiment. Animals were sacrificed for the brain sample at 6, 12, and 24 h after hypoxia treatment.

QUANTITATIVE REAL TIME PCR ANALYSIS
Total RNA was isolated from the whole hippocampus of rat using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) followed by chloroform extraction and isopropanol precipitation. RNA was quantified by spectrophotometric absorbency at 260 nm, purity confirmed by A260/A280 ratio. RNA samples were stored at −80°C.

WESTERN BLOT
The samples were extracted from the whole hippocampus, homogenized with 2% SDS, 100 mM dithiothreitol, and 10% glycerol and stored at 4°C. Equal amount of proteins were loaded and separated on 10% SDS-PAGE. Then the proteins were transferred to a polyvinylidene fluoride (PVDF) membrane and blocked by immersion for 2 h in Tris-buffered saline (TBS) containing 5% dry milk. The membranes were then incubated at 4°C overnight with rabbit polyclonal anti-Kv4.2 (Sigma, USA; used at 1:1000) and Kv4.3 receptor antibodies (Sigma, USA; used at 1:1000) and βactin rat monoclonal antibody (Tianjin Sungene Biotech Co., Ltd, Tianjin, China;used at 1:2000). After incubation, the membranes were washed three-times with TBST (TBS containing 0.2% Tween-20) and were then incubated again for 2 h with the goat anti-rat IgG (diluted 1:20,000) at room temperature. The membranes were washed again with TBST three-times. Finally, the reaction was developed using a chemiluminescent reagent (ECL; Ecl Advantage Inc., Menlo Park, CA, USA) and exposed to Hyper film (GE Healthcare Life Sciences, Pittsburgh, PA, USA).

CELL TRANSFECTIONS AND TREATMENT WITH HYPOXIA
Cell culture was carried out using standard procedures. HEK-293 cells were maintained in DMEM supplemented with 10% fetal calf serum (GIBCO BRL) and incubated in a humidified atmosphere containing 5% CO 2 in air at 37°C. For transfections, 1 µg of pcDNA3.0-Kv4.2 or pcDNA3.0-Kv4.2/Kv4.3 DNA into HEK-293 cells was performed using Lipofectamine 2000 (Invitrogen) according to the instruction of the manufacturer. All experiments were recording HEK-293 cells within 20 passages. To study the effect of hypoxia on K + currents, cells after transfected were maintained in a modular incubator chamber (Billups-Rothenberg, Del Mar, CA, USA) at 37°C with 1% O 2 for 6 or 12 h. Cells transfected for 24 h total were used for electrophysiological recordings, with hypoxia exposure carried during the final 6 or 12 h of incubation.

ELECTROPHYSIOLOGICAL RECORDINGS
K + currents were studied using the whole-cell patch-clamp technique and voltage-clamp recordings were carried out at room temperature (22-25°C) by using a MultiClamp-700B amplifier and Digidata-1440A A/D converter (Axon Instruments, CA, USA). Recording electrodes were pulled using a Sutter P-97 puller (Sutter Instruments, CA, USA). Electrodes were filled with intracellular solution containing (in millimolar): 125 KCl, 4 MgCl 2 , 10 HEPES, 10 EGTA, and 5 MgATP, adjusted to 310 mOsm/L with sucrose, pH 7.2 with KOH. The external solution contained (in millimolar): 141 NaCl, 4.7 KCl, 1.2 MgCl 2 , 1.8 CaCl 2 , 10 glucose, and 10 HEPES, adjusted to 330 mOsm/L with sucrose, pH 7.4 with NaOH. The resistance of the recording pipette filled with internal solution was 2-6 MΩ. The adjustment of capacitance compensation Frontiers in Cellular Neuroscience www.frontiersin.org and series resistance compensation was done before recording the membrane currents. The holding potential was −80 mV. Isolated HEK-293 cells were studied 24 h after transfection. Membrane currents were filtered at 2 kHz and digitized at 10 kHz, and the data were stored in compatible PC computer for off-online analysis using the pCLAMP 10 acquisition software (Axon Instruments, CA, USA). The voltage-dependence of activation was studied by a 400 ms constant depolarizing pulse from −50 mV to +40 mV in 10 mV steps (see Figure 2A inset). To determine the voltage-dependence of steady-state inactivation of Kv4.2, the voltage-dependence of inactivation was assessed by measuring the peak amplitude of current responses evoked by test depolarization to +40 mV after 1.5 s pre-pulse to potentials between −110 and 0 mV with 10 mV increment (see Figure 2B inset).

DATA ANALYSIS
All data were analyzed by Igor Pro 6.10 (Wavemetrics, Lake Oswego, OR, USA), pCLAMP 10 (Axon Instruments, CA, USA) and Origin 7.5 (Microcal Software, USA). To construct steadystate activation curves, peak currents (I ) were converted into conductance (G) using the formula: where V m is the membrane potential, and V rev is the reversal potential (−90 mV in our experiments). The normalized conductance was fitted by a Boltzmann Equation: where V 1/2 is the half-maximal membrane potential for activation and k is the slope factor. The steadystate inactivation curves were established using a Boltzmann equation: where V 1/2 is the half-maximal membrane potential for inactivation and k is the slope factor. Mono-or bi-exponential functions were used to fit current rise time and decay time, respectively, using one of the following equations: where t is time, A 1 and A 2 are the amplitudes, and τ 1 and τ 2 are the time constants of decay.
All data reported in this study are expressed as means ± SEM, Statistical significance was verified using one-way ANOVA test and Student's t -test. P < 0.05 was the evaluation criterion for statistical significance.

HYPOXICSUPPRESSION OF Kv4.2 CHANNEL-MEDIATED CURRENT
Whole-cell patch-clamp recording was performed on HEK-293 cells transfected with Kv4.2. As shown in Figures 2A-C, Kv4.2 channel currents exhibited rapid activation and inactivation typical of A-type potassium currents. Since there was no significant difference in mRNA and protein between 12 and 24 h after hypoxia treatment in vivo, we examined the properties of Kv4.2 channel currents at only 6 and 12 h after hypoxia treatment. Peak Kv4.2 channel-mediated current amplitudes were significantly decreased at 6 and 12 h after hypoxia treatment. The peak current amplitude was 2.55 ± 0.54 nA (n = 10) at +40 mV in the control group and 1.70 ± 0.52 nA (n = 10, P < 0.01) and 1.46 ± 0.17 nA (n = 13, P < 0.01) at 6 and 12 h after hypoxia treatment, respectively ( Figure 2C).
To characterize the rate of activation of the Kv4.2-induced currents in HEK-293 cells, the rising phases of Kv4.2 currents evoked at test potentials between 0 and +40 mV were fitted by single exponential function after the onset of the voltage step to the peak of the outward current. As is evident, the Kv4.2-induced currents activate rapidly, and the activation time constants vary with voltage, decreasing with increasing membrane depolarization ( Figure 2F). However, the Kv4.2-induced currents activate more slowly after hypoxia treatment. Figure 2G showed that the activation time constant evoked at +40 mV in control was increased at 6 and 12 h after hypoxia treatment from 1.23 ± 0.17 ms (n = 6) to 1.68 ± 0.13 ms (n = 7, p < 0.05 vs. control) and 2.19 ± 0.22 ms (n = 10, p < 0.01 vs. control), respectively (Table 1). Next, we investigated the decay kinetics of Kv4.2 currents evoked at +40 mV. The decay phase of the currents could be well fitted by a bi-exponential function, yielding fast and slow-decay time constants. As shown in Figure 2H, neither τ fast nor τ slow displayed a significant change after hypoxia treatment.

Frontiers in Cellular Neuroscience
www.frontiersin.org To determine the rates of Kv4.2 recovery from steady-state inactivation, cells were first depolarized to +40 mV for 1 s (to inactivate the currents), then hyperpolarized to −110 mV for varying times (ranging from 10 to 200 ms) prior to test depolarizations to +40 mV (to assess recovery). Recovery curves were generated by plotting normalized current amplitudes (normalized to maximum) of activation as a function of pre-pulse time and fitting with a single exponential. At −110 mV, the recovery time constant in control was 94.84 ± 9.94 ms (n = 8) ( Figure 3A). The recovery time constant at 6, 12 h after hypoxia treatment were not significantly changed (91.24 ± 12.90 ms, n = 10, p > 0.05, 6 h vs. control; 86.32 ± 8.36 ms, n = 10, p > 0.05, 12 h vs. control) ( Table 1). Therefore, except for the time to activation, most biophysical features of Kv4.2 currents remained unchanged after hypoxia.
In addition, the rising and decay phases of the Kv4.2/Kv4.3 currents were described by single or double exponentials. Similar to the results for Kv4.2 (Figure 2F), the Kv4.2/Kv4.3-induced currents activate rapidly, and the activation time constants vary with voltage, decreasing with increasing membrane depolarization ( Figure 4F). However, the Kv4.2-induced currents activate more slowly after hypoxia treatment. At +40 mV, the activation time constant in the control group was significantly increased at 6 and 12 h after hypoxia treatment from 0.97 ± 0.04 ms (n = 10) to 2.53 ± 0.23 ms (n = 8, p < 0.001 vs. control) and 1.99 ± 0.23 ms (n = 7, p < 0.001 vs. control), respectively ( Figure 4G) (Table 1). However, hypoxia had no significant effect on the fast and slow-decay time constants (Figure 4H).

DISCUSSION
The current study showed that hypoxia down-regulated I A in vitro. In addition, hypoxia increased the time constant of activation without significantly altering the voltage-dependence of activation and inactivation. Furthermore, hypoxia had similar effects on co-expressed Kv4.2 and Kv4.3 channels. Importantly, we also observed that hypoxia in vivo decreased the expression of Kv4.2 and Kv4.3 in the hippocampus, providing a possible mechanism of I A inhibition observed previously after global hypoxia [see Patel et al. (1997)].
As expected, heterologously expressed Kv4.2 and Kv4.3 α subunits showed properties of I A (rapid activation and inactivation) (Birnbaum et al., 2004). Previous studies have shown that I A can be regulated by hypoxia/ischemia, but the functional significance of this change is speculative (Cummins et al., 1991;Hyllienmark and Brismar, 1996;Gebhardt and Heinemann, 1999;Chi and Xu, 2000). I A and other K + currents were transiently increased after ischemia/reperfusion in vivo (Chi and Xu, 2000), yet brain slices exposed to hypoxia in vitro resulted in inhibition of K + currents (Cummins et al., 1991;Hyllienmark and Brismar, 1996;Gebhardt and Heinemann, 1999). Consistent with the latter, our data showed that I A was decreased after hypoxia treatment in an in vitro expression system. One possible mechanism underlying the reduction of potassium currents is that the number of K + channels decreased after hypoxia treatment. Kv1.3 protein levels were observed to decrease in Jurkat T cells after hypoxia, with the hypoxic down-regulation of Kv1.3 expression depending on hypoxia severity, indicating that the number of Kv channels could be altered by hypoxia (Conforti et al., 2003a). Moreover, hypoxia down-regulated the transcription and expression of Kv1.2 and Kv1.5 in rat pulmonary arterial smooth muscle cells (PASMC) (Wang et al., 1997). The human embryonic kidney cells (HEK-293) have been widely used as one mammalian expression system in the study of Kv channels. It's important to acknowledge the possibility that endogenous K channels could contribute to currents measured through the heterologously expressed Kv channels in these cells. However, HEK-293 do not endogenously express Frontiers in Cellular Neuroscience www.frontiersin.org  (Jiang et al., 2002;Thomas and Smart, 2005). Also, as a blank control, we recorded from HEK-293 cells that were transfected with the reporter gene GFP without the Kv subunits (same protocol as shown in Figure 2A). The average amplitude of the depolarizationactivated current was <500 pA, with a slow-activation phase and slow-decay phase, very different from A-type potassium channel (as shown in Figure S1 in Supplementary Material). Thus, in the current study, we demonstrated that Kv4.2 and Kv4.3 mRNA and protein were decreased by hypoxia, supporting decreased channels as a mechanism of I A inhibition. Another possibility is that channel properties, such as the open probability and opening time, were altered after hypoxia treatment. The redox state of amino acid residues in channel proteins Frontiers in Cellular Neuroscience www.frontiersin.org  is a potent regulator of channel activity for multiple ion channel types (Bahring et al., 2001;Su et al., 2007). Furthermore, Kvβ1.2 could regulate the redox and oxygen sensitivity of the Kv4.2 currents (Perez-Garcia et al., 1999). Thus, redox modulation resulting from hypoxia could be a mechanism of altered channel function.
Frontiers in Cellular Neuroscience www.frontiersin.org However, the decrease of I A after hypoxia treatment in the current study is different from previous studies on I A during hypoxia. Deng et al. (2011) reported that I A was up-regulated in ischemia-resistant large aspiny neurons and other studies have shown increased potassium currents during hypoxia (Leblond and Krnjevic, 1989). Variation in the results of previous reports and our observations could be attributed to multiple experimental variables that include the preparation (brain slice vs. cultured cells), the time and severity of hypoxia, and the comparison of native versus recombinant channels. Nonetheless, our data provide evidence that Kv4 channel function can be directly modulated by hypoxia in a manner to decrease I A and promote membrane excitability.
A-type potassium currents play a critical role in action potential repolarization and inhibition of burst firing in many types of neurons and cardiac muscle (Castro et al., 2001;Kim et al., 2005). I A in neuronal dendrites also contributes to the resting membrane potential and regulates firing frequency (Birnbaum et al., 2004). The reduction of I A by hypoxia could increase action potential half-widths and shorten the latency to the first action potential compared to physiological conditions (Peng et al., 2013). This shortened latency also increases the probability of action potential generation in response to synaptic excitation and can change a repetitive firing pattern to burst firing, thus increasing the excitability of the neuron. Further studies have indicated that the recovery from inactivation of I A can be altered by hypoxia (Ruschenschmidt et al., 2006). Although this could functionally inhibit I A , our previous study indicated that the speed of recovery from I A inactivation was not altered by hypoxia (Peng et al., 2013).
The signaling mechanisms for I A inhibition by hypoxia in the current study are speculative. Activation of PKA or PKC can shift the activation curve of I A in a depolarizing direction in hippocampal area CA1 dendrites (Birnbaum et al., 2004). Furthermore, the inhibition of I A induced by glial cell line-derived neurotrophic factor (GDNF) is mediated by ERK/MAPK (Yang et al., 2001). Kv4.2 is expressed on dendrites in CA1 pyramidal neurons and likely forms the pore-forming subunits of I A channels in this area (Jerng et al., 2004). It is worth noting that Kv4.2 is a substrate for ERK in vitro and in hippocampal pyramidal neurons (Adams et al., 2000), so Kv4.2 is a target for regulation by ERK/MAPK. It is likely that kinase phosphorylation of Kv4.2 decreases the probability of channel opening or the number of channels in the dendritic membrane, and thus, this could have been a mechanism of decreased I A .
In summary, the current study showed a down-regulation of Kv4.2 mRNA and protein and Kv4.3 channels mRNA after hypoxia treatment in vitro, consistent with decreased channel expression after seizure-inducing hypoxia in neonatal rat in vivo. Brief hypoxia exposure induced the inhibition of peak I A . In addition, the time constant of activation of I A was increased by hypoxia. The down-regulation of A-type channels by hypoxia may play a role in the development of neonatal seizures that result from an hypoxic episode, and our data raised possible mechanisms of direct Kv4 channel regulation by hypoxia.

ACKNOWLEDGMENTS
This work was supported by the National Natural Sciences Foundation of China (No. 81100970, 81370737, and 81371422), the National Basic Research Program of China (2010CB529803) and the Fundamental Research Funds for the Central Universities. Part of this manuscript is the result of work supported with resources and the use of facilities at the Central Texas Veterans Health Care System, Temple, TX, USA.