4-Aminopyridine induced hyperpolarizing oscillations in pediatric human epileptic tissue are network-driven potassium currents that are abolished by activation of KCNQ2–5 (Kv7.2-Kv7.5) channels

Figures

• 

  Fig. 1.. 4-AP induces synchronized bursting and causes HypOs. A) Traces of paired current clamp recordings of L2/3 PNs indicating synchronized bursting after 20-min of 4-AP (100 μM) wash-on (inset: expanded individual burst). B) Traces of a paired recording of L2/3 PNs showing synchronized HypOs after 4-AP wash-on (n = 5) (inset: expanded individual HypO). C) The amplitude of HypOs is dependent on membrane potential, and this dependence is similar in control (left) and epileptic (right) tissue. D) Reversal potential for HypOs calculated by simple linear regression is not different between control (blue) and epileptic (purple) L2/3 PNs(pslope = 0.325, pintercept = 0.756). E) When L2/3 PNs were held at −55 mV, epileptic HypOs had significantly larger amplitudes compared to control HypOs (Welch’s t-test: p = 0.0020, ncontrol = 24, nepilepsy = 128). F) When maintained in 4-AP alone, epileptic HypO amplitude increased over 30-min. G) Epileptic HypO frequency decreased over 30-min.

• 

  Fig. 2.. HypOs are not regulated by GABA or glutamate receptor activation. A) Example traces of 4-AP + GABAA receptor blocker, gabazine (GBZ, 10 μM), GABAB receptor blocker, CGP (20 μM), and Na+ channel blocker, tetrodotoxin (TTX, 1 μM) influence on HypOs. Ai) HypO amplitude was not significantly reduced by GBZ and CGP (Wilcoxon test: p = 0.250, n = 8), but HypOs were eliminated by consecutive addition of TTX (n = 3). Aii) HypO frequency was significantly reduced with wash-on of GBZ and CGP (Wilcoxon test: p = 0.004, n = 8). B) Example traces showing the effects of a GABAA receptor blocker, bicuculline methiodide (BMI, 20 μM), and a GABAB receptor blocker, Phaclofen (PH, 20 μM). These traces also demonstrate biphasic HypO with depolarizing followed by hyperpolarizing phases. BMI eliminated the depolarizing phase of the oscillations. Bi-ii) GABAergic receptor blockade with BMI and PH had no significant influence on HypO amplitude or frequency (paired t-test: amplitude: p = 0.181, frequency: p = 0.169, n = 5,). C) Example traces of HypOs in 4-AP alone or in the presence of glutamate receptor blockers (AMPA receptor blocker DNQX (20 μM) and NMDA receptor blocker APV (50 μM)). Ci-ii) Blockade…

• 

  Fig. 3.. The reversal potential for HypOs is mediated by potassium. A) Schematic of paired whole-cell recordings from L2/3 PNs with high K+ internal in one neuron (dark green) and high Cl− internal in one neuron (light green). B) Paired recording with 4-AP (100 μM) wash-on and different internal solutions (high K+ vs high Cl−) indicates that when chloride has a depolarized reversal potential, this ion contributes to bursting activity that is synchronized with HypOs. C) Paired recording demonstrates that the addition of GABAA and GABAB receptor blockers (GBZ, 10 μM and CGP, 20 μM, respectively) to a cell recorded with a high Cl− internal causes bursting to terminate and HypOs to become unmasked. D, Top) Schematic of whole-cell recording with a Low K+ internal solution. D, Bottom) Example trace of L2/3 PN HypOs when a low K+ internal solution was used demonstrates that HypO is depolarizing when the K+ reversal potential is positive. E) HypO amplitudes (ΔmV) are dependent on the reversal potentials of both potassium and chloride (nHighK+ = 4, nHighCl- = 5, nLowK+ = 3). F) HypO frequencies are not controlled by differences in chloride or potassium driving force (nHighK+ = 4, nHighCl- = 5,…

• 

  Fig. 4.. Potassium channel modulators perturb HypOs to varying degrees. A-H) Example traces of 4-AP induced HypOs before (left) and after (right) K+ channel modulator wash-on. I) Normalized HypO amplitudes before and after K+ channel blocker wash-on indicates retigabine (50 μM, n = 5) and to varying degrees TEA (30 mM, n = 3/5) are the only K+ channel blockers capable of eliminating HypOs. J) Normalized HypO frequencies before and after K+ channel blocker wash-on indicates reduction in HypO frequency due to K+ channel blockers, substantial frequency reduction with TEA (n = 5), and elimination of HypOs with retigabine (n = 5). Acronyms: Bumetanide = BUM, Furosemide = FUR, Tertiapin-Q = Tert-Q, Charybdotoxin = CTX, Tetraethylammonium = TEA, Retigabine = RTG.

• 

  Fig. 5.. Voltage-gated Ca2+ channel blockage and chelation of internal Ca2+ diminish HypOs. A) Example traces of HypOs in the presence of 4-AP (100 μM) and 4-AP with CdCl2 (200 μM) demonstrates that Cd2+ can eliminate HypOs by blocking voltage-gated calcium channels. B, C) HypOs are completely abolished and thus both amplitude and frequency are reduced to zero in the presence of CdCl2 (paired t-test: amplitude: p = 0.007, frequency: p = 0.0014, n = 5). D) Example traces of 4-AP + internal BAPTA salt internal (5 mM, 10 mM or 20 mM) influence on HypOs. E-F) Calcium chelation with 20 mM of intracellular BAPTA leads to a significant reduction in HypO amplitude and frequency (MWU test: amplitude: p = 0.009, frequency: p = 0.017, n = 5). Lower concentrations of intracellular BAPTA lead to smaller effects.

• 

  Fig. 6.. Gap junctions influence HypOs. Ai) Example traces of 4-AP (100 μM) with NFA (300 μM) wash-on causing a reduction in HypO amplitude. Aii) Example traces showing NFA leading to a reversal of polarity in 4-AP induced HypOs, and the termination of HypOs with consecutive addition of BMI (20 μM). B, C) NFA caused a significant reduction in HypO amplitude and frequency (Wilcoxon test: amplitude: p = 0.016, frequency: p = 0.016, n = 7). D) Example traces of the effects of T16Ainh-A01 (25 μM) on HypOs. E, F) T16Ainh-A01 caused a significant reduction in HypO amplitude but did not significantly affect frequency (paired t-test: amplitude: p = 0.048, frequency: p = 0.083, n = 4). G) Example traces of 4-AP showing the effects of MFA (100 μM) alone or in combination with bicuculline methiodide (BMI, 20 μM), and Phaclofen (PH, 20 μM). Note that the changes in MFA are progressive over time. H, I) MFA (100 μM) demonstrates similar effects to NFA and significantly decreases HypO amplitude and frequency (Wilcoxon test: amplitude: p = 0.016, frequency: p = 0.016, n = 7). Additionally, MFA can cause polarity switch to depolarizing potentials (n = 4/7). J) Example traces of the influence of…

• 

  Fig. 7.. Schematic of the proposed mechanism of how KCNQ2–5 channel activation and gap junction blockade mediate 4-AP induced HypOs. A-B) KCNQ channels are not open at −55 mV and would not suppress 4-AP induced neuronal bursting or synchronization. Activation of KCNQ2–5 channels allows for slow, m-current to attenuate neuronal bursting, network synchronization, and 4-AP induced HypOS. C–D) Functional gap junctions in astrocytes allow effective buffering of [K+]o and allow interneurons to synchronize electrically. With blockade of gap junctions, inhibitory interneurons are less likely to form synchronous network activity. Moreover, gap junction blockade inhibits astrocytic [K+]o buffering leading to an overall increase in [K+]o. Overall, this would lead to reduction in 4-AP induced HypOs.