Figure 2. Loss of KCNQ2/3 activity leads to increased excitability of PV⁺interneurons.

(a) Top, representative voltage responses from a +150 pA current injection step (1 s) in PV- and SST-like interneurons in either the CA1 region of the hippocampus (P12–P17) or L2/3 of the somatosensory cortex (P8–P11). For L2/3 recordings, cells were also confirmed by immunoreactivity against SST antibodies. Bottom, summary graphs showing the effect of deleting KCNQ2 and KCNQ3 channels on action potential number from CA1 PV-like (control n = 8/6; IN:Kcnq2/3 null n = 8/5), SST-like (control n = 19/8; IN:Kcnq2/3 null n = 8/4), and L2/3 (PV-like: control n = 10/7; IN:Kcnq2/3 null n = 8/4; SST-like: control n = 10/6; IN:Kcnq2/3 null n = 5/4) interneurons (Vh=-75 to −77 mV). For CA1 PV-like cells (P16–P25), F_(9,126)=2.849, p=0.0043; for L2/3 PV-like cells, F_(9,144)=3.845, p=0.0002); for CA1 SST-like cells (P15–P19), F_(9,225)=0.601, p=0.7955; and for L2/3 SST-like cells, F_(9,117)=0.326, p=0.965. Significance was determined using a two-factor mixed ANOVA. See Figure 2—figure supplement 1 showing that indeed SST cells express KCNQ2 and KCNQ3 mRNA. (b) Top, representative voltage responses to a series of current injection steps (1 s) in PV⁺ and SST⁺ interneurons in the CA1 region of the hippocampus (Vh=-75 to −77 mV). Bottom left, summary graph showing the effect of deleting KCNQ2 and KCNQ3 channels on action potential number from CA1 PV⁺ cells (control n = 15/8; PV:Kcnq2/3 null n = 14/7; F_(9,243)=3.558 with p=0.0004). Middle left, summary graph showing that loss of KCNQ2/3 channels decreases PV⁺ input resistance (control, n = 15/8; PV:Kcnq2/3 null, n = 14/7; df = 27 t=−2.54 p=0.017 unpaired Student’s t-test). See also Figure 2—figure supplement 2 regarding PV⁺ Kcnq2/3 null neurons diversity of intrinsic properties. Middle right, summary graph showing the effect of deleting KCNQ2 and KCNQ3 channels on action potential number from CA1 SST⁺ cells (control n = 6/2; SST:Kcnq2/3 null n = 8/4; F_(9,108)=0.729 with p=0.6814). Bottom right, summary graph showing loss of KCNQ2/3 channels did not decrease SST⁺ input resistance (control n = 6/2; SST:Kcnq2/3 null, n = 8/4; df = 12 t=−0.42 p=0.68 unpaired Student’s t-test). ‘n’ designates number of cells followed by number of animals. Each data point represents recording from one neuron. Data in summary graphs are represented as mean and s.e.m.

Figure 2—figure supplement 1. FISH shows presence of KCNQ2 and KCNQ3 in SST⁺interneurons.

Confocal micrographs of coronal sections of CA1 region of wild-type mice. Micrographs shows co-localization of mRNA for somatostatin, KCNQ2 and KCNQ3. All micrographs have been counterstained with DAPi (blue).

Figure 2—figure supplement 2. PV:Kcnq2/3 null interneurons could differ in their intrinsic excitability properties.

Top, representative voltage responses to a series of current injection steps (1 s) in PV⁺ interneurons in the CA1 region of the hippocampus (P16–P25). We found that Kcnq2/3 null PV⁺ interneurons could exhibit two intrinsic excitability behaviors. Type I that has a sag ratio of 0.9 and greater, similar to control neurons, and type II that has a sag ratio of 0.8 or less. Excluding type II cells (red; as in Figure 2b) or including all types of cells (blue) as shown here, it did not change the conclusions of this work. The lower sag ratio might indicate immature PV⁺ cells, however, other possibilities cannot be excluded at this point.

Figure 2—figure supplement 3. The pan-KCNQ blocker XE991 increases PV⁺interneuron excitability.

Left, representative voltage responses from a + 200 pA current injection step (1 s; Vh=-75 to −77 mV) in PV⁺ interneurons (Pvalb-Cre;Kcnq2^+/+;Ai9) from the CA1 region of the hippocampus (P23–P25) before and after application of 20 μM XE-991 (15 min application). Right, summary graphs showing the of XE-991 on action potential number (n = 6/3; F_(1,5)=38.379, p=0.0016) and input resistance (control 169 ± 20 MΩ,+XE-991 143 ± 20 MΩ, n = 6/3; t = 1.355 df = 5 p=0.2335). Significance was determined using two-way factor repeated measures ANOVA and two-tailed paired Student’s t-test for action potential number and input resistance, respectively.