Report

IKCa Channels Are a Critical Determinant of the Slow

AHP in CA1 Pyramidal Neurons
Graphical Abstract

Authors
Brian King, Arsalan P. Rizwan, ...,
Gerald W. Zamponi, Ray W. Turner

Correspondence
rwturner@ucalgary.ca

In Brief
The molecular identity of a Ca2+dependent slow afterhyperpolarization
(sAHP) that controls cortical neuronal
excitability has gone unsolved for over 30
years. King et al. now show that IKCa
(KCa3.1) channels underlie the sAHP in
CA1 pyramidal cells to suppress temporal
summation of EPSPs and mediate spike
accommodation.

Highlights
d

CA1 pyramidal cells express intermediate-conductance

Ca2+-activated K+ channels

An sAHP exhibits a pharmacological profile specific to IKCa

channels

IKCa channels reduce temporal summation of EPSPs and

mediate spike accommodation

IKCa channels are a key determinant of the sAHP in CA1

pyramidal cells

King et al., 2015, Cell Reports 11, 175182

April 14, 2015 2015 The Authors
http://dx.doi.org/10.1016/j.celrep.2015.03.026

Cell Reports

Report
IKCa Channels Are a Critical Determinant
of the Slow AHP in CA1 Pyramidal Neurons
Brian King,1,2 Arsalan P. Rizwan,1,2 Hadhimulya Asmara,1 Norman C. Heath,1 Jordan D.T. Engbers,1 Steven Dykstra,1
Theodore M. Bartoletti,1 Shahid Hameed,1 Gerald W. Zamponi,1 and Ray W. Turner1,*
1Hotchkiss

Brain Institute, University of Calgary, Calgary, AB T2N 4N1, Canada

author
*Correspondence: rwturner@ucalgary.ca
http://dx.doi.org/10.1016/j.celrep.2015.03.026
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
2Co-first

SUMMARY

Control over the frequency and pattern of neuronal

spike discharge depends on Ca2+-gated K+ channels
that reduce cell excitability by hyperpolarizing
the membrane potential. The Ca2+-dependent slow
afterhyperpolarization (sAHP) is one of the most
prominent inhibitory responses in the brain, with
sAHP amplitude linked to a host of circuit and
behavioral functions, yet the channel that underlies
the sAHP has defied identification for decades.
Here, we show that intermediate-conductance
Ca2+-dependent K+ (IKCa) channels underlie the
sAHP generated by trains of synaptic input or postsynaptic stimuli in CA1 hippocampal pyramidal cells.
These findings are significant in providing a molecular identity for the sAHP of central neurons that will
identify pharmacological tools capable of potentially
modifying the several behavioral or disease states
associated with the sAHP.
INTRODUCTION
Of the many classes of K+ channels recognized in central neurons, few have as key a role in regulating the frequency and
pattern of spike discharge as Ca2+-gated K+ channels. Despite
this, only two classes of Ca2+-gated K+ channels were recognized to control signal processing and spike output in central
neurons: big-conductance (BK) and small-conductance (SK)
channels (Adelman et al., 2012; Berkefeld et al., 2010). The properties of BK and SK channels differ in key respects that mediate
repolarization over relatively short time frames of activity, with
BK channels typically generating a fast afterhyperpolarization
of
10 ms and SK channels a medium afterhyperpolarization
of
100 ms. However, other Ca2+-gated K+ channels exert
even greater control over membrane excitability than BK or SK
channels. One of the longest-standing examples of this is the
Ca2+-gated slow afterhyperpolarization (sAHP) of seconds duration (Andrade et al., 2012). The sAHP is unique in being highly
amenable to modulation by neurotransmitters, with sAHP amplitude linked to circuit functions that include synaptic plasticity,

electroencephalography, aging, and several disease states (Disterhoft et al., 2004; Haug and Storm, 2000; Madison and Nicoll,
1986; Martn et al., 2001; Zhang et al., 2013).
Despite the wealth of information on sAHP properties, the molecular identity of the Ca2+-dependent K+ channel(s) underlying
the sAHP has defied explanation. A third possible Ca2+-dependent channel is the intermediate-conductance Ca2+-gated K+
channel (SK4, IKCa, KCa3.1) (Ishii et al., 1997; Joiner et al.,
1997; Logsdon et al., 1997). While these were not originally
thought to be expressed in CNS neurons (reviewed in Wulff
et al., 2007), recent work in cerebellar Purkinje cells and on
IKCa protein distribution suggests a wide potential expression
pattern in central neurons (Engbers et al., 2012; Turner et al.,
2015). The current study tested the hypothesis that IKCa channels contribute to the Ca2+-dependent component of the sAHP
in CA1 pyramidal cells. We report that K+ channels of intermediate conductance with the unique pharmacological profile of IKCa
channels indeed underlie the sAHP, identifying the molecular basis for one of the largest inhibitory responses in central neurons.
RESULTS
The sAHP of CA1 Pyramidal Cells
The sAHP in CA1 hippocampal pyramidal cells is known to
mediate spike accommodation, in which spike frequency is progressively reduced during an injected current pulse and followed
by a post-stimulus afterhyperpolarization (Madison and Nicoll,
1986). The sAHP can also be evoked by synaptic inputs in stratum radiatum (SR) using short stimulus bursts (530 pulses,
50 Hz), with the sAHP apparent during and immediately after
the stimulus train (Figures 1 and 2). To test the functional role
of IKCa channels, we first blocked SK channels using 100 nM
apamin and Kv7 channels with 10 mM XE-991, given a role for
Kv7 channels in other hippocampal neurons (Tzingounis et al.,
2010). It is known that IKCa channels are apamin insensitive
(Wulff et al., 2007), and we confirmed that 10 mM XE-991 had
no effect on IKCa channels (Figure S1A) and did not impede
SR-evoked synaptic transmission (Vervaeke et al., 2006). Any
contribution from the Na-K pump was minimized by recording
at 32 C (Gulledge et al., 2013). To focus on excitatory synaptic
potentials, we applied 50 mM picrotoxin to block GABAergic
transmission. The combination of apamin, XE-991, and picrotoxin produced little qualitative change in the firing response to
Cell Reports 11, 175182, April 14, 2015 2015 The Authors 175

Figure 1. TRAM-34-Sensitive Mechanisms

Contribute to Spike Accommodation and
the sAHP

10
mV

10
mV

50 ms

200 ms

SR Stimulation, 30 pulses, 50 Hz

last sec

(7)

500 1000

500 ms

* *

4
2
0

500 1000

Time frame (ms)

1st 200 ms
Fold Frequency

*
8
6
4
2
0

Fold Number

Fold Frequency

1st 500 ms

20
mV

10
mV

last 400 ms

*
(7)

1
0
200 400

Fold Number

TRAM-34

100 ms

6
4
2
0

200 400

Time frame (ms)

(A) Spike accommodation during depolarizing

current injection is reduced by 1 mM TRAM-34 to
increase spike number and frequency. Mean bar
plots indicate the fold change in spike frequency or
spike number over the times indicated by horizontal bars following TRAM-34.
(B) Repetitive SR stimulation (30 pulses, 50 Hz) in
the same cell as in (A) is associated with prominent
EPSP suppression and spike accommodation that
is reduced by infusion of 1 mM TRAM-34. Open
arrows and inset illustrate the effects of TRAM-34
on the sAHP that follows a synaptic train. Mean bar
plots indicate the fold change in spike frequency or
spike number over the times indicated by horizontal bars following TRAM-34.
(C) Repetitive SR stimulation (30 pulses, 50 Hz) at
sub- or supra-threshold intensity in the absence of
picrotoxin to preserve inhibitory inputs reveals a
significant effect by TRAM-34 (1 mM) on EPSP
summation and spike output.
All recordings in (A) and (B) were obtained in
100 nM apamin, 10 mM XE-991, and 50 mM
picrotoxin, while those in (C) did not include
picrotoxin. In all cases, TRAM-34 was internally
perfused through the electrode. Values are mean
SEM; *p < 0.05. See also Figures S1S3.

GABAergic inhibition intact

TRAM-34

5
mV
200 ms

SR Stimulation, 30 pulses, 50 Hz

current pulse injection or SR stimulus trains (see the Supplemental Experimental Procedures and Figures S2A and S2B).
The role of IKCa channels can be distinguished using the selective blockers TRAM-34, Senicapoc, or NS-6180 (Stocker
et al., 2003; Strbk et al., 2013; Wulff et al., 2001). The selectivity of TRAM-34 has been thoroughly established, with IKCa
channels exhibiting far greater sensitivity to 1 mM TRAM-34
than SK or BK channels (Wulff et al., 2000). Control tests
confirmed that 1 mM TRAM-34 had no significant effect on BK,
Kv7.3, or TMEM16B channels expressed in tsA-201 cells or
the SR-evoked excitatory postsynaptic current paired-pulse ratio in tissue slices (Figures S1BS1E). Since TRAM-34 must first
be internalized to block IKCa channels, we found that the fastest
and most reliable block was obtained through internal perfusion
of 1 mM TRAM-34 through the electrode (see the Experimental
Procedures).
IKCa Channels Contribute to Spike Accommodation and
Temporal Summation
Internal perfusion of 1 mM TRAM-34 reduced spike accommodation and the sAHP in rat CA1 pyramidal cells evoked by current
injection or a 30-pulse SR stimulus train (Figures 1A and 1B).
The initial suppression of excitatory postsynaptic potential
(EPSP) summation by the sAHP during a 50-Hz SR stimulus train
176 Cell Reports 11, 175182, April 14, 2015 2015 The Authors

5
mV

could be extensive, with the membrane

potential often approaching or even falling below the resting potential (Figure 1B).
Block of the sAHP by TRAM-34 led to
a maintained depolarization during SRevoked synaptic trains and a loss of spike
accommodation evident in an increase in the frequency and
number of spikes (Figures 1A and 1B). The magnitude of a
post-stimulus TRAM-34-sensitive sAHP was more prevalent
for SR-evoked trains compared to just threshold current-evoked
spike discharge (Figures 1A and 1B). This is expected, given a
known relation between sAHP amplitude and spike frequency
and number (Madison and Nicoll, 1984), with threshold current
injection evoking spike output at 12.7 2.3 Hz (n = 7) compared
to 36.1 5.3 Hz for 50-Hz SR stimuli. Each of these results was
obtained in rats of age postnatal day 18 (P18) to P23 (n = 8) or
P90 (n = 2) and in mice ranging from P25 to P50 (n = 8) or P60
(n = 2). We further delivered SR stimulus trains when picrotoxin
was excluded (n = 3) to determine if the effects of TRAM-34 could
be detected when inhibitory GABAergic inputs were intact.
These tests confirmed that TRAM-34-sensitive currents are
effective when GABAergic inputs are intact by modulating
temporal summation and spike accommodation for either subor supra-threshold SR stimulation (Figure 1C).
200 ms

The Synaptically Evoked sAHP Is Reduced in KCa3.1/

Mice
We next examined the ability to record the sAHP in KCa3.1/
mice compared to wild-type (WT). We found that in response
to depolarizing current injection, pyramidal cells in KCa3.1/

KCa3.1-/-

ns

TRAM-34

20
mV

6
4
2
0

sAHP Area (mV*s)

TRAM-34

10
mV
1 sec

Baseline Voltage (mV)

KCa3.1-/-

(4)

Cont TRAM

12

ns

10
8
6
4
2

(7)

(7)

Cont

TRAM

(6)

(5)

Cont TRAM

KCa3.1-/WT Control (7)

WT TRAM (7)
KCa3.1-/- Control (6)

1.5
1.4
1.3
1.2
1.1
1.0
0

50 Hz, 30 SR stimuli

10

15

20

25

30

Stimulus Number

Wild-type
Senicapoc

wild type

10
0

Control

TRAM-34

15

14

-65
mV

20

Wild-type

-66
mV

(4)

25

Cont TRAM

500 ms

ns

Spike number

Control

Frequency (Hz)

F
Control

KCa3.1-/-

Control

Control
2
mV
2
mV
1 sec

Control

DC-EBIO

SKA-31

1 sec

DC-EBIO
DC-EBIO+ SKA-31

Figure 2. The sAHP Is Reduced in KCa3.1/ Animals

(A) Current-evoked firing in CA1 pyramidal cells of KCa3.1/ mice lacks
sensitivity to internal infusion of 1 mM TRAM-34.
(B) Responses in mouse CA1 pyramidal cells to SR stimulus trains (30 pulses,
50 Hz) reveal an sAHP in WT, but not KCa3.1/, animals that is reduced by
1 mM TRAM-34 in WT animals.
(C) Mean sAHP area in WT and KCa3.1/ mice before and after TRAM-34 for
tests as in (B).
(D) Mean baseline membrane voltage normalized to the voltage at the start of
repetitive SR stimulation (as in (B) reveal a suppression of EPSP summation in
WT animals that is reduced to that of KCa3.1/ animals by 1 mM TRAM-34.
(E) The sAHP following five-pulse 50 Hz SR stimulus trains in rat pyramidal cells
is reduced by the IKCa blocker Senicapoc (100 nM) and enhanced by the IKCa
agonists DC-EBIO (0.1 mM) or SKA-31 (1 mM).
(F) Superimposed recordings of five-pulse 50-Hz SR stimulus trains in pyramidal cells of KCa3.1/ mice in control conditions and after sequential
perfusion of DC-EBIO (0.1 mM) and SKA-31 (1 mM) show no effect on the
afterpotential.
All recordings were obtained in 100 nM apamin, 10 mM XE-991, and 50 mM
picrotoxin. TRAM-34 was internally perfused through the electrode. Values are
mean SEM; *p < 0.05, **p < 0.01, ns, not significant, paired Students t test
(A), two-way ANOVA (C), and repeated-measures ANOVA (F).

mice exhibited a characteristic initial high-frequency burst attributable to the actions of BK channels (Gu et al., 2007) but no
prominent spike accommodation (Figure 2A). Infusion of 1 mM
TRAM-34 through the electrode had no significant effect on
the frequency or number of spikes evoked (Figure 2A). Trains
of SR synaptic input in WT mice (30 pulses, 50 Hz) were followed

by a pronounced sAHP that peaked immediately following the

train and lasted for several seconds (Figure 2B). By comparison,
recordings in KCa3.1/ mice revealed at most a small-amplitude sAHP following a SR stimulus train (Figure 2B), presumably
carried by the Na-K pump (Gulledge et al., 2013). Moreover, infusion of 1 mM TRAM-34 in the electrode rapidly reduced the sAHP
in WT, but not KCa3.1/, mice (Figures 2B and 2C). Closer examination of the baseline membrane voltage (EPSPs) during the
SR stimulus train revealed a suppression of EPSP summation
that was greater in WT mice than in KCa3.1/ animals (Figure 2D). Infusion of TRAM-34 during recordings from WT mice
also promoted temporal summation during the SR stimulus train
and raised the baseline voltage to within the range recorded in
KCa3.1/ animals (Figure 2D).
Further evidence for a role for IKCa channels was obtained
when the sAHP following a five-pulse 50 Hz SR stimulus train
was reduced by the alternate IKCa channel blocker Senicapoc
(100 nM) (sAHP area reduced by 61% 17%, n = 6, p < 0.05)
(Figure 2E). Conversely, sAHP area was enhanced by DC-EBIO
(0.1 mM) (241% 47%, n = 5, p < 0.05) or SKA-31 (1 mM)
(363% 108%, n = 6, p < 0.01), two agonists that increase the
sensitivity of IKCa channels to [Ca]i (Figure 2E) (Wulff et al.,
2007). Finally, sequential perfusion of DC-EBIO and then SKA31 had no effect on the SR-evoked afterpotential in KCa3.1/
mice (n = 5) (repeated-measures ANOVA, p > 0.05) (Figure 2F).
Since all of these recordings were conducted in 100 nM apamin,
the effects of DC-EBIO and SKA-31 are consistent with an action
on IKCa-mediated current. Together, these data provide strong
evidence that IKCa channels contribute to the synaptically
evoked sAHP of CA1 pyramidal cells.
CA1 Pyramidal Cells Exhibit Outward Current with a
Pharmacology Consistent with IKCa Channels
To further test the membrane expression of IKCa channels,
we obtained outside-out recordings from CA1 pyramidal cell
somata. To provide a consistent source of Ca2+ the internal
solution for recording was buffered to 1 mM [Ca]i and HVA
and LVA Ca2+ channels were blocked using 30 mM Cd2+ and
300 mM Ni2+ in the artificial cerebrospinal fluid (aCSF). IKCa
current was further isolated using a set of drugs to eliminate
other voltage- or Ca2+-gated ion channels (see the Supplemental
Experimental Procedures and Table S1). A ramp command
was then applied under voltage clamp from a holding potential
of 110 mV to +60 mV (500 ms) every 30 s over 5 min to ensure
stable recordings. Internal perfusion of the electrode with 1 mM
TRAM-34 reduced the total outward current by 38% 5.4% in
seven out of eight recordings (measured at +40 mV, p < 0.01)
(Figure 3A). The TRAM-34-sensitive current gave rise to a linear
current-voltage (I-V) plot that reversed at 90.1 4.0 mV (n = 7),
a value that approached the predicted value of 99 mV for EK
(Figure 3A). The scorpion toxins charybdotoxin (ChTx) and maurotoxin (MTx) are also known to block IKCa channels (Table S1)
(Castle et al., 2003; Wulff et al., 2007). In the presence of 5 mM
tetraethylammonium chloride (TEA) to block BK and Kv1 channels, we tested the effects of pressure ejecting 100 nM ChTx
or MTx. ChTx reduced total current by 33% 7.5% (n = 7,
measured at +40 mV, p < 0.01), with ChTx-sensitive current
reversing on mean I-V plots at 81.3 8.8 mV (n = 7) (Figure 3B).
Cell Reports 11, 175182, April 14, 2015 2015 The Authors 177

Control
TRAM-34 (n = 7)
30

Subtraction

Control
ChTx (n = 7)

mV
-80 -60 -40 -20

50
pA
100 ms

20 40 60

D
50

20
pA
100 ms

60 mV
-110 mV

-80 -60 -40 -20

-10

mV
20 40 60

30
20

Subtraction

20
10

-80 -60 -40 -20

-20

Control
PKAcat (n = 5)

40
30

Subtraction

20

pA

40

10

pA

50
pA
100 ms

80
60

20

Subtraction

Figure 3. CA1 Pyramidal Cells Exhibit an

Outward Current with a Pharmacological
Profile Consistent with IKCa Channels

Control
MTx (n = 5)

pA

pA

50
pA
100 ms

mV

60 mV

20 40 60

-110 mV

MTx reduced outward current by 40% 7.6% (n = 5, p < 0.01)

and reversed at 94.1 6.7 mV (n = 5) (Figure 3C). Finally, internal perfusion of the electrode with the catalytic subunit of protein
kinase A (PKAcat, 100 U/ml) reduced outward current by 35%
4.3% in five out of six cells (p < 0.001) that reversed at 97.6
4.7 mV (n = 5) (Figure 3D). A comparison of these results revealed
that the degree of block by each of the compounds (TRAM-34,
ChTx, MTx, and PKAcat) was not significantly different (oneway ANOVA, p > 0.05). The results again reveal a K+ current in
CA1 pyramidal cells with the unique pharmacological profile of
IKCa channels.
CA1 Pyramidal Cells Exhibit Ca2+-Dependent
Intermediate Conductance K+ Channels
We next determined if Ca2+-activated K+ channels of intermediate conductance could be recorded in pyramidal cells. For this,
we obtained somatic on-cell patch recordings using a HEPESbuffered aCSF-based electrode solution to preserve intracellular
contents and native Ca2+ buffering mechanisms. IKCa current
was isolated using a set of blockers of voltage- or Ca2+-gated
ion channels in both the electrode and external aCSF (see the
Supplemental Experimental Procedures and Table S1). Calcium
channels were not blocked in order to promote Ca2+ conductance across the patch by delivering a set of spike-like depolarizing commands through the electrode (50 Hz, 20 pulses, 5 ms,
80 mV). The membrane potential was subsequently stepped to
a set of steady-state potentials from 60 to +30 mV with respect
to the resting state to examine evoked currents. When recording
under conditions of 3.25 mM [K]o in the on-cell electrode, we
assumed that the resting potential under the patch approached
the value obtained in a separate set of perforated-patch recordings of 64 1.5 mV (n = 12). To simplify interpretation, outward
current is presented with respect to the cell interior as an upward
deflection in all on-cell recordings.
We found that the train of spike-like commands was followed
by activation of non-inactivating single channels (n = 4) or even
macroscopic current (n = 6) (Figures 4A and 4B). Single-channel
amplitude revealed a mean conductance of 30.4 5.8 pS (n = 4)
(Figure 4A), a value within the range for IKCa channels reported
178 Cell Reports 11, 175182, April 14, 2015 2015 The Authors

10

mV
-80 -60 -40 -20-5

20 40 60

Shown are outside-out recordings from pyramidal

cell somata with 1 mM [Ca]i in response to a
500-ms ramp command from 110 mV to +60 mV.
Mean I-V plots reflect currents blocked by the
indicated agents following subtraction of test
from control recordings. IKCa current was isolated
using blockers described in the Supplemental
Experimental Procedures.
(AD) In each case, outward current is voltage independent and reduced by (A) TRAM-34 (1 mM),
(B) ChTx (100 nM), (C) MTx (100 nM), or (D) PKAcat
(100 U/ml). ChTx and MTx were focally pressure
ejected, while TRAM-34 and PKAcat were internally perfused in the patch electrode.
SEMs for mean values in I-V plots are indicated by
the shaded area.

in other cell types (Wulff et al., 2007). Both single channels

(n = 4) and macroscopic currents (n = 6) were rapidly reduced
by bath perfusion of 1 mM TRAM-34, which will cross the cell
membrane in regions outside of the on-cell patch (Figures 4A
and 4B). The I-V plot for the TRAM-34-sensitive currents evoked
by the spike-like stimulus train reversed
10 mV more negative
than the native resting potential (Figure 4A) and could exhibit
slight inward rectification at more hyperpolarized potentials
(n = 4/6).
To determine the ion selectivity of macroscopic currents, we
obtained on-cell recordings using an electrode solution containing 140 mM KCl to establish equimolar [K] across the membrane
patch, thus setting EK to 0 mV. By comparison, ECl under these
conditions was predicted to rest at 90 mV across the patch.
All other blockers in the electrode and bath are described in
the Supplemental Experimental Procedures. To apply the series
of spike-like depolarizing commands over a range that would
promote Ca2+ influx through the patch, the holding potential
was set to 65 mV and the membrane subsequently stepped
to potentials over a range of 40 mV to +120 mV (D mV) with
respect to 65 mV. As before, we recorded non-inactivating
macroscopic currents that were reduced by 1 mM TRAM-34
(n = 6) (Figure 4B). Importantly, the TRAM-34-sensitive currents
were confirmed as representing a K+ conductance by reversing
through 0 mV (EK) on the I-V plot (Figure 4B).
The Ca2+ sensitivity of these currents was established when
0.1 mM DC-EBIO increased peak macrocurrent in six out of
nine cells (Figure 4C). Subsequent perfusion of the membrane
permeable BAPTA-AM (10 mM) fully blocked evoked currents
(n = 3) (Figure 4C). PKA-mediated phosphorylation is well known
to block the sAHP of CA1 pyramidal cells (Lancaster et al., 1991;
Pedarzani et al., 1998) as well as IKCa channels (Wong and
Schlichter, 2014). We found that 8-bromo-cyclic AMP (100 mM)
rapidly reduced channel activity or macroscopic outward current
in on-cell recordings (n = 5) (Figure 4D). These data reveal that a
Ca2+-dependent intermediate-conductance K+ channel can be
recorded that is sensitive to TRAM-34, an internal Ca2+ chelator,
or elevation of cyclic AMP, all properties consistent with IKCa
channels.

[K]o = 3.25 mM

A
mV

[K]o = 140 mM

Control

O3
O2
O1
C

+30

mV

Control

-60

DC-EBIO

+30

Current (pA)

+20

Current (pA)

Control

mV
+30

+120

0
-20
-40
C
O1
O2

-60

30 mV

-40
-60 mV

TRAM-34

-60

BAPTA-AM

Subtraction

+120

50
pA

80 mV

30 mV

1 sec

RMP
-60 mV

TRAM-34
+30

2
pA
200 msec

80 mV
30 mV
RMP

-40

80 mV
-65 mV

Current (pA)

1.0

= 30 pS

0.5
RMP
-0.5
-1.0
-1.5

(-65)

-60

-40

-20

RMP

20

Applied voltage from rest ( mV)

6
4
2
0
-2

ECl

D
mV
+30

-40 mV

-60 mV

1.5

Current (pA)

5
pA
2 sec

120 mV

EK

-60
+30

-4
-6
-8

Control

8-bromo-cAMP

-60

n=6

-100 -80 -60 -40 -20 0 20 40 60

Predicted transmembrane
potential (mV)

80 mV

5
pA
30 mV

1 sec

RMP
-60 mV

Figure 4. IKCa Channels Are Expressed in CA1 Pyramidal Cells

Shown are on-cell somatic recordings using a HEPES-buffered aCSF internal solution, with Ca2+ influx evoked by a repetitive spike-like command (50 Hz,
20 pulses, 5 ms, 80 mV) followed by steps to different steady-state potentials. The resting membrane potential (RMP) across the patch is presumed to be
approximately 65 mV for records in (A), (C), and (D) with 3.25 mM [K]o, while that for 140 mM [K]o in (B) was set at 65 mV. Voltage commands reflect the step
applied to the electrode and displayed with net depolarizing commands upward. Current is displayed with respect to the cell interior with outward current as an
upward deflection. The mean I-V plot in (B) reflects TRAM-34-sensitive currents calculated by subtraction of test from control records.
(A) On-cell recording using 3.25 mM [K]o in the electrode reveals single channel activity following the pulse train that is rapidly reduced or blocked by 1 mM TRAM34. Current reversed approximately 10 mV from the resting state.
(B) On-cell recordings of macroscopic current using equimolar [K]o exhibits reversal of TRAM-34-sensitive current through 0 mV (EK).
(C) On-cell recorded macroscopic current is enhanced by the IKCa agonist DC-EBIO (0.1 mM) and blocked by 10 mM BAPTA-AM.
(D) On-cell macroscopic current is reduced by 100 mM 8-bromo-cyclic AMP. Records in (B) show every second record for clarity.
The electrode and bath perfusate in all cases contained blockers as described in the Supplemental Experimental Procedures. Traces were filtered at 400500 Hz
(8-pole Bessel). Values are mean SEM.

Synaptic Activation of IKCa Channels and the sAHP

To test for synaptic activation of IKCa channels in relation to the
sAHP, we again used on-cell recordings and short trains of
supra-threshold SR stimuli (five pulses, 50 Hz). We further
applied a 60-mV steady-state holding potential above the resting
membrane potential to the patch to increase the driving force for
K+ current, and we applied the same blockers as for single channels above (Figure 4).
In 5 of 27 cases, on-cell recordings yielded single channels that
were activated during or by the end of a five-pulse SR stimulus
train (Figure 5A). A high rate of opening was apparent immediately
following the train followed by a persistent flickering state for
several seconds (Figure 5A). Alternatively, a macroscopic outward current was recorded that had a rapid onset following the
SR stimulus train and remained open for a variable duration of
25 s before spontaneously terminating (n = 5) (Figure 5B). Calcu-

lating the ensemble average of SR-evoked macroscopic currents

revealed an outward current that peaked immediately following
the SR stimulus train and then dissipated over 45 s (Figure 5B).
Bath application of 1 mM TRAM-34 rapidly reduced or blocked
SR-evoked single channels (n = 5) or macrocurrents (n = 5) (Figures 5A and 5B). These results are important in providing the first
evidence for synaptic activation of an identified K+ channel with
activity that correlates to the sAHP of CA1 pyramidal cells.
The Ca2+-activated outward current underlying sAHP current
(IsAHP) is most often recorded under voltage clamp following a
depolarizing step command of
1 s. We directly compared the
sensitivity of the step-evoked and SR-evoked IsAHP to TRAM34 application. Perforated-patch whole-cell recordings were
used to preserve intracellular Ca2+ buffering and outward currents evoked by five SR stimuli (50 Hz) or by a step command
to +60 mV (500 ms) (Figures 5C and 5D). An outward current
Cell Reports 11, 175182, April 14, 2015 2015 The Authors 179

Control

TRAM-34

2
pA
200 msec

o
c
+60 mV

SR stimulation, 5 pulses, 50 Hz

o
c
SR stimulation

100 msec

B
Control

2
pA

Control

TRAM-34
SR
-65 mV

TRAM-34

1 sec

20
pA

Control

60 mV

TRAM-34
1 sec

10
pA

-65 mV

5
pA
1 sec

Control

ChTx

1 sec

4
pA

SR
-65 mV

Control
5
pA

n=9

60 mV

SR stimulation, 5 pulses, 50 Hz

10
pA

ChTx

1 sec
-65 mV

was recorded following the SR stimulus train in seven out of eight

cells with an amplitude of 47 5 pA and duration of 6.8 1.5 s
(n = 7). In the same cells a step command to +60 mV evoked
an IsAHP of 28 11 pA and 6.3 1.2 s (n = 7). Perfusion of
1 mM TRAM-34 reduced both the SR- and step-evoked outward
current (n = 7, p < 0.01) (Figure 5C). If IKCa channels underlie
IsAHP, then the outward current should also be blocked by
ChTx. We thus repeated these tests in the additional presence
of 5 mM TEA to first block BK channels and found that pressure
ejection of 100 nM ChTx reduced the IsAHP evoked by either SR
stimulation (n = 6, p < 0.05) or a step command (n = 7, p < 0.05)
(Figures 5C and 5D).
DISCUSSION
The current study reveals that IKCa channels contribute substantially to the Ca2+-dependent sAHP in CA1 hippocampal pyramidal cells. Support for this was obtained through recordings
of intermediate-conductance K+ channels evoked in relation to
the sAHP with the unique pharmacological profile for IKCa channels. These data are thus important in resolving the molecular
basis for one of the largest Ca2+-dependent inhibitory postsynaptic responses in central neurons.
180 Cell Reports 11, 175182, April 14, 2015 2015 The Authors

1 sec

Figure 5. IKCa Channels Are Evoked in CA1

Pyramidal Cells by Synaptic Stimulation
(A and B) On-cell recordings with a HEPES-buffered aCSF electrode solution (3.25 mM [K]). Currents are illustrated with respect to the cell interior
(outward current upward). Currents were evoked
using a 50 Hz, five-pulse SR stimulus train with a
net 60 mV depolarized holding potential to increase driving force for K+ across the patch. The
electrode and bath perfusate contained blockers
as described in the Supplemental Experimental
Procedures. (A) A single channel is activated during and following a five-pulse SR stimulus train and
is blocked by bath perfusion of 1 mM TRAM-34.
Dashed lines depict open (o) and closed (c) states,
with the same example expanded below. (B) Oncell recordings of an outward macroscopic current
that opens for prolonged but variable periods of
time following SR stimulus trains (arrows) and is
reduced or blocked by TRAM-34. An ensemble
average from 9 SR stimulus trains (lowest trace)
reveals an average time course equivalent to
an sAHP. Transients in (B) reflect capacitive
transients from spontaneous spike discharge in
the cell.
(C and D) Comparison of outward current evoked
by SR stimulation to the IsAHP evoked by a step
command in perforated-patch recordings in the
presence of 100 nM apamin, 10 mM XE-991, and
50 mM picrotoxin (C and D), with 5 mM TEA further
included in (D) to block BK channels for tests with
ChTx. Currents were evoked by five suprathreshold SR stimuli (50 Hz) or a 500-ms step to
60 mV to evoke IsAHP and bath apply 1 mM TRAM34 or pressure eject 100 nM ChTx. Results in (C)
and (D) are from separate cells.
Traces in (A) and (B) were filtered at 500 Hz and the
expanded trace in (A) at 1 kHz (8-pole Bessel).
Values are mean SEM.

Ionic Contributions to the sAHP

Studies on the ionic basis for the sAHP of CA1 pyramidal cells
have considered several channel subtypes. Fluctuation analysis
implicated a K+ channel of 25 pS (Sah and Isaacson, 1995).
The SK1 isoform was thus considered to underlie the sAHP,
but this was later discounted (Bond et al., 2004). A role for
TMEM16B channels in mediating various aspects of repolarization has been proposed for CA1 cells using the drugs NFA and
NPPB (Huang et al., 2012), but these drugs also block IKCa
channels (Fioretti et al., 2004; Wulff et al., 2007). We further
found no block of TMEM16B channels by TRAM-34 when expressed in tsA-201 cells, and the current isolated here reversed
through EK and not ECl. A role for voltage-gated Kv7 channels
(IM) in generating an sAHP has been reported, but these are
less involved in CA1 cells (Tzingounis et al., 2010) and we
consistently recorded a TRAM-34-sensitive sAHP in XE-991.
While sodium-dependent K+ channels can contribute to an
sAHP in other cells (Zhang et al., 2010), the sAHP and IsAHP
recorded here is Ca2+ dependent (Figures S4AS4C). A role
for the Na-K pump was proposed based on the effects of
ouabain at 35 C using high-frequency current pulses (Gulledge
et al., 2013). We found that the majority of the sAHP evoked by
this protocol at 35 C was blocked by internal infusion of 1 mM

TRAM-34, with a small remaining TRAM-34-insensitive component subsequently blocked by 20 mM ouabain (Figures S4D
and S4E).
Interestingly, Ca2+-dependent single channels of
19 pS
conductance were earlier reported in both tissue slices (Lima
and Marrion, 2007) and cultured neurons (Lancaster et al.,
1991). However, in neither case was the molecular identity of
these channels attained. The sAHP in cultured hippocampal
neurons was further reduced by clotrimazole (Shah et al.,
2001), a drug related to TRAM-34 but later recognized as
non-specific. The sAHP and spike accommodation was also
reported to be insensitive to ChTx (Lancaster and Nicoll,
1987; Shah and Haylett, 2000). By comparison, we found
ChTx-sensitive current in outside-out and perforated-patch recordings. The reason for this difference is unknown but might
reflect differences in internal Ca2+ buffering. Indeed, our most
stable recordings of IKCa channels and IsAHP were obtained
using on-cell or perforated-patch recordings that minimize a
disruption of [Ca]i.
IKCa Channels and the sAHP
Our current interpretation for the ionic basis of the sAHP in CA1
pyramidal cells is that the majority of the large-amplitude, Ca2+dependent early phase (45 s) of the sAHP is mediated by IKCa
channels. An additional smaller contribution can be made by Kv7
channels or by the Na-K pump, although the Na-K pump is expected to be most active following intense activation. Delineating IKCa channels as a key contributor to the sAHP in a
cortical pyramidal neuron is significant in that IKCa channels
were believed to be restricted to endothelial cells and activated
microglia in central regions (Wulff et al., 2007). The reasons why
early probes for in situ hybridization did not detect KCa3.1 signal
in brain is unknown, since a signal from at least endothelial
KCa3.1 would be expected. However, recent work has shown
that IKCa channels can be recorded in cerebellar Purkinje cells
(Engbers et al., 2012) and GFP expression driven by the
KCa3.1 promoter suggested an even wider expression pattern
of IKCa (Turner et al., 2015). Indeed, numerous central neurons
express an sAHP that shares many properties with the sAHP in
CA1 pyramidal cells, including Ca2+ dependence, insensitivity
to BK or SK channel blockers, and rapid block by kinase pathways, with a very close parallel found in enteric and myenteric
neurons (Neylon et al., 2004; Vogalis et al., 2002). The finding
that IKCa-mediated suppression of temporal summation is still
active under conditions of intact GABAergic inhibition further ensures that the influence of IKCa channels will be present under
physiological conditions. We thus expect the role for IKCa channels in generating an sAHP to be widely applicable to other classes of central neurons.
EXPERIMENTAL PROCEDURES
Animal Care and Slice Preparation
Experiments were conducted on P18P24 male Sprague-Dawley rats (Charles
River) raised from timed-pregnant dams or on breeding colonies of P25P50
C57BL/6 WT mice or P25P60 KCa3.1/ mice (UC Davis) (as described
in Turner et al., 2015) according to approved procedures by the Canadian
Council of Animal Care. Details on in vitro slice preparation are provided in
the Supplemental Experimental Procedures.

Electrophysiology
A suite of patch-clamp recording techniques was used to record from rat or
mouse hippocampal CA1 pyramidal cells under current- or voltage-clamp
conditions in the in vitro slice preparation (3234 C) and from tsA-201 cells
(22 C). Details on recording equipment and patch recording configurations
can be found in the Supplemental Experimental Procedures.
Solutions and Drug Applications
The lipophilic drugs TRAM-34, Senicapoc, and NS-6180 must be internalized
to target an internal binding site (Stocker et al., 2003; Strbk et al., 2013;
Wulff et al., 2001). Using bath perfusion of these drugs, channel activity was
reduced in on-cell recordings in 1015 min while whole-cell recordings could
take 2030 min to achieve full block. The most reliable block was obtained with
TRAM-34 and Senicapoc, with infusion of 1 mM TRAM-34 by exchanging the
electrode solution (ALA Scientific Instruments) achieving a block of IKCa channels in <10 min. The majority of tests in whole-cell recording configuration thus
used internal perfusion of 1 mM TRAM-34. Further details on electrode and
external solutions are given in the Supplemental Experimental Procedures.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures,
four figures, and one table and can be found with this article online at http://
dx.doi.org/10.1016/j.celrep.2015.03.026.
AUTHOR CONTRIBUTIONS
R.W.T., B.K., A.P.R., H.A., N.C.H., and G.W.Z. designed the study or provided
reagents. B.K., A.P.R., H.A., N.C.H., S.H., J.D.T.E., and T.M.B. performed the
experiments or analyzed the data. R.W.T., B.K., and A.P.R. wrote the
manuscript.
ACKNOWLEDGMENTS
We gratefully acknowledge a gift of Senicapoc from H. Wulff (University of California, Davis) and thank H. Wulff and J. Adelman (Vollum Institute, OR) for
helpful comments on the manuscript. M. Kruskic and L. Chen provided expert
technical assistance. This work was supported by the Canadian Institutes of
Health Research (CIHR; R.W.T. and G.W.Z.), studentships from the Alberta
Heritage Foundation for Medical Research (AHFMR; J.D.T.E), a T. Chen
Fong award (J.D.T.E.), the Killam Foundation (J.D.T.E.), a CIHR-CGS PhD studentship (J.D.T.E.), and CIHR and Alberta Innovates-Health Solutions (AI-HS)
postdoctoral fellowships (T.M.B.). G.W.Z. and R.W.T. are AI-HS Scientists.
Received: November 27, 2014
Revised: January 30, 2015
Accepted: March 10, 2015
Published: April 9, 2015
REFERENCES
Adelman, J.P., Maylie, J., and Sah, P. (2012). Small-conductance Ca2+-activated K+ channels: form and function. Annu. Rev. Physiol. 74, 245269.
Andrade, R., Foehring, R.C., and Tzingounis, A.V. (2012). The calciumactivated slow AHP: cutting through the Gordian knot. Front. Cell. Neurosci.
6, 47.
Berkefeld, H., Fakler, B., and Schulte, U. (2010). Ca2+-activated K+ channels:
from protein complexes to function. Physiol. Rev. 90, 14371459.
Bond, C.T., Herson, P.S., Strassmaier, T., Hammond, R., Stackman, R.,
Maylie, J., and Adelman, J.P. (2004). Small conductance Ca2+-activated K+
channel knock-out mice reveal the identity of calcium-dependent afterhyperpolarization currents. J. Neurosci. 24, 53015306.
Castle, N.A., London, D.O., Creech, C., Fajloun, Z., Stocker, J.W., and Sabatier, J.M. (2003). Maurotoxin: a potent inhibitor of intermediate conductance
Ca2+-activated potassium channels. Mol. Pharmacol. 63, 409418.

Cell Reports 11, 175182, April 14, 2015 2015 The Authors 181

Disterhoft, J.F., Wu, W.W., and Ohno, M. (2004). Biophysical alterations of hippocampal pyramidal neurons in learning, ageing and Alzheimers disease.
Ageing Res. Rev. 3, 383406.

Sah, P., and Isaacson, J.S. (1995). Channels underlying the slow afterhyperpolarization in hippocampal pyramidal neurons: neurotransmitters modulate the
open probability. Neuron 15, 435441.

Engbers, J.D.T., Anderson, D., Asmara, H., Rehak, R., Mehaffey, W.H.,
Hameed, S., McKay, B.E., Kruskic, M., Zamponi, G.W., and Turner, R.W.
(2012). Intermediate conductance calcium-activated potassium channels
modulate summation of parallel fiber input in cerebellar Purkinje cells. Proc.
Natl. Acad. Sci. USA 109, 26012606.

Shah, M., and Haylett, D.G. (2000). Ca(2+) channels involved in the generation
of the slow afterhyperpolarization in cultured rat hippocampal pyramidal neurons. J. Neurophysiol. 83, 25542561.

Fioretti, B., Castigli, E., Calzuola, I., Harper, A.A., Franciolini, F., and Catacuzzeno, L. (2004). NPPB block of the intermediate-conductance Ca2+-activated
K+ channel. Eur. J. Pharmacol. 497, 16.
Gu, N., Vervaeke, K., and Storm, J.F. (2007). BK potassium channels facilitate
high-frequency firing and cause early spike frequency adaptation in rat CA1
hippocampal pyramidal cells. J. Physiol. 580, 859882.
Gulledge, A.T., Dasari, S., Onoue, K., Stephens, E.K., Hasse, J.M., and Avesar,
D. (2013). A sodium-pump-mediated afterhyperpolarization in pyramidal neurons. J. Neurosci. 33, 1302513041.
Haug, T., and Storm, J.F. (2000). Protein kinase A mediates the modulation of
the slow Ca(2+)-dependent K(+) current, I(sAHP), by the neuropeptides CRF,
VIP, and CGRP in hippocampal pyramidal neurons. J. Neurophysiol. 83,
20712079.
Huang, W.C., Xiao, S., Huang, F., Harfe, B.D., Jan, Y.N., and Jan, L.Y. (2012).
Calcium-activated chloride channels (CaCCs) regulate action potential and
synaptic response in hippocampal neurons. Neuron 74, 179192.
Ishii, T.M., Silvia, C., Hirschberg, B., Bond, C.T., Adelman, J.P., and Maylie, J.
(1997). A human intermediate conductance calcium-activated potassium
channel. Proc. Natl. Acad. Sci. USA 94, 1165111656.
Joiner, W.J., Wang, L.Y., Tang, M.D., and Kaczmarek, L.K. (1997). hSK4, a
member of a novel subfamily of calcium-activated potassium channels.
Proc. Natl. Acad. Sci. USA 94, 1101311018.
Lancaster, B., and Nicoll, R.A. (1987). Properties of two calcium-activated
hyperpolarizations in rat hippocampal neurones. J. Physiol. 389, 187203.
Lancaster, B., Nicoll, R.A., and Perkel, D.J. (1991). Calcium activates two types
of potassium channels in rat hippocampal neurons in culture. J. Neurosci. 11,
2330.
Lima, P.A., and Marrion, N.V. (2007). Mechanisms underlying activation of the
slow AHP in rat hippocampal neurons. Brain Res. 1150, 7482.
Logsdon, N.J., Kang, J., Togo, J.A., Christian, E.P., and Aiyar, J. (1997).
A novel gene, hKCa4, encodes the calcium-activated potassium channel in
human T lymphocytes. J. Biol. Chem. 272, 3272332726.
Madison, D.V., and Nicoll, R.A. (1984). Control of the repetitive discharge of rat
CA 1 pyramidal neurones in vitro. J. Physiol. 354, 319331.
Madison, D.V., and Nicoll, R.A. (1986). Cyclic adenosine 30 ,50 -monophosphate
mediates beta-receptor actions of noradrenaline in rat hippocampal pyramidal
cells. J. Physiol. 372, 245259.
Martn, E.D., Araque, A., and Buno, W. (2001). Synaptic regulation of the slow
Ca2+-activated K+ current in hippocampal CA1 pyramidal neurons: implication in epileptogenesis. J. Neurophysiol. 86, 28782886.
Neylon, C.B., Nurgali, K., Hunne, B., Robbins, H.L., Moore, S., Chen, M.X., and
Furness, J.B. (2004). Intermediate-conductance calcium-activated potassium
channels in enteric neurones of the mouse: pharmacological, molecular and
immunochemical evidence for their role in mediating the slow afterhyperpolarization. J. Neurochem. 90, 14141422.
Pedarzani, P., Krause, M., Haug, T., Storm, J.F., and Stuhmer, W. (1998). Modulation of the Ca2+-activated K+ current sIAHP by a phosphatase-kinase balance under basal conditions in rat CA1 pyramidal neurons. J. Neurophysiol.
79, 32523256.

182 Cell Reports 11, 175182, April 14, 2015 2015 The Authors

Shah, M.M., Miscony, Z., Javadzadeh-Tabatabaie, M., Ganellin, C.R., and

Haylett, D.G. (2001). Clotrimazole analogues: effective blockers of the slow
afterhyperpolarization in cultured rat hippocampal pyramidal neurones. Br.
J. Pharmacol. 132, 889898.
Stocker, J.W., De Franceschi, L., McNaughton-Smith, G.A., Corrocher, R.,
Beuzard, Y., and Brugnara, C. (2003). ICA-17043, a novel Gardos channel
blocker, prevents sickled red blood cell dehydration in vitro and in vivo in
SAD mice. Blood 101, 24122418.
Strbk, D., Brown, D.T., Jenkins, D.P., Chen, Y.J., Coleman, N., Ando, Y.,
Chiu, P., Jrgensen, S., Demnitz, J., Wulff, H., and Christophersen, P.
(2013). NS6180, a new K(Ca) 3.1 channel inhibitor prevents T-cell activation
and inflammation in a rat model of inflammatory bowel disease. Br. J. Pharmacol. 168, 432444.
Turner, R.W., Kruskic, M., Teves, M., Scheidl-Yee, T., Hameed, S., and Zamponi, G.W. (2015). Neuronal expression of the intermediate conductance calcium-activated potassium channel KCa3.1 in the mammalian central nervous
system. Pflugers Arch. 467, 311328.
Tzingounis, A.V., Heidenreich, M., Kharkovets, T., Spitzmaul, G., Jensen, H.S.,
Nicoll, R.A., and Jentsch, T.J. (2010). The KCNQ5 potassium channel mediates a component of the afterhyperpolarization current in mouse hippocampus. Proc. Natl. Acad. Sci. USA 107, 1023210237.
Vervaeke, K., Gu, N., Agdestein, C., Hu, H., and Storm, J.F. (2006). Kv7/KCNQ/
M-channels in rat glutamatergic hippocampal axons and their role in regulation
of excitability and transmitter release. J. Physiol. 576, 235256.
Vogalis, F., Harvey, J.R., and Furness, J.B. (2002). TEA- and apaminresistant K(Ca) channels in guinea-pig myenteric neurons: slow AHP channels.
J. Physiol. 538, 421433.
Wong, R., and Schlichter, L.C. (2014). PKA reduces the rat and human KCa3.1
current, CaM binding, and Ca2+ signaling, which requires Ser332/334 in the
CaM-binding C terminus. J. Neurosci. 34, 1337113383.
Wulff, H., Miller, M.J., Hansel, W., Grissmer, S., Cahalan, M.D., and Chandy,
K.G. (2000). Design of a potent and selective inhibitor of the intermediateconductance Ca2+-activated K+ channel, IKCa1: a potential immunosuppressant. Proc. Natl. Acad. Sci. USA 97, 81518156.
Wulff, H., Gutman, G.A., Cahalan, M.D., and Chandy, K.G. (2001). Delineation
of the clotrimazole/TRAM-34 binding site on the intermediate conductance
calcium-activated potassium channel, IKCa1. J. Biol. Chem. 276, 32040
32045.
Wulff, H., Kolski-Andreaco, A., Sankaranarayanan, A., Sabatier, J.M., and
Shakkottai, V. (2007). Modulators of small- and intermediate-conductance
calcium-activated potassium channels and their therapeutic indications.
Curr. Med. Chem. 14, 14371457.
Zhang, L., Kolaj, M., and Renaud, L.P. (2010). Ca2+-dependent and Na+dependent K+ conductances contribute to a slow AHP in thalamic paraventricular nucleus neurons: a novel target for orexin receptors. J. Neurophysiol. 104,
20522062.
Zhang, L., Ouyang, M., Ganellin, C.R., and Thomas, S.A. (2013). The slow
afterhyperpolarization: a target of b1-adrenergic signaling in hippocampusdependent memory retrieval. J. Neurosci. 33, 50065016.

Cell Reports
Supplemental Information

IKCa Channels Are a Critical Determinant

of the Slow AHP in CA1 Pyramidal Neurons
Brian King, Arsalan P. Rizwan, Hadhimulya Asmara, Norman C. Heath, Jordan D.T.
Engbers, Steven Dykstra, Theodore M. Bartoletti, Shahid Hameed, Gerald W. Zamponi,
and Ray W. Turner

Figure S1, related to Figure 1. Pharmacological isolation of IKCa channels

Whole-cell recordings obtained with alpha subunits of the indicated proteins transiently expressed in tsA201 cells and recorded at room temperature.
(A) The Kv7 channel blocker XE-991 (10 M) has no significant effect on IKCa channels. IKCa was
coexpressed with calmodulin and currents recorded in 1 M [Ca]i in response to a ramp command.
(B - D) TRAM-34 (1 M) has no significant effect on BK channels (B), Kv7.3 channels (C), or
TMEM16B calcium-gated chloride channels recorded with 100 nM [Ca]i (D). TMEM16B currents are
rapidly blocked by 300 M niflumic acid (NFA) in (D).
(E) TRAM-34 (1 M) does not significantly affect the mean SR-evoked EPSC paired pulse ratio (20 ms
interstimulus interval).
Values are mean SEM with sample values shown in brackets.
1

Figure S2, related to Figures 1 and 2. Background channel blockers have minimal effects on the
sAHP and IsAHP
(A) Perfusion of 100 nM apamin, 10 M XE-991, and 50 M picrotoxin to block SK and Kv7 channels
and GABAergic transmission slightly reduce but do not eliminate spike accommodation during current
pulse injection.
(B) The early phase of the sAHP following a train of SR input is blocked (arrow) and the sAHP is more
evident after perfusion of apamin, XE-991, and picrotoxin (inset).
(C) The sAHP evoked by a train of SR stimuli is progressively reduced by block of NMDA receptors (DLAP5, 25 M) and AMPA receptors (DNQX, 10 M).
(D, E) Recordings of IsAHP evoked as a tail current under whole-cell voltage clamp using a depolarizing
step command (D) or following a SR stimulus train (E). Perfusion of apamin, XE-991, and picrotoxin
selectively reduces the early component of outward current consistent with a block of SK channels. Spikes
and stimulus artefacts in (B, C, E) and currents during the command pulse in (D) are truncated.

Figure S3, related to Figures 1 and 2. The sAHP in whole-cell or perforated patch configurations
resists wash out
(A) Whole-cell current clamp (left) and perforated patch voltage clamp (right) recordings from P18-22 rat
CA1 pyramidal cells. Applying a stimulus protocol consisting of either 5 suprathreshold SR stimuli (50 Hz)
(left) or a 500 ms step to 60 mV to evoke IsAHP (right) revealed sAHP currents that were stable for over
30 min. Recordings were started after 5 min stabilization time post-breakin or perforation, and shown here
comparing initial recordings (0 min, black) and after 30 min (red).
(B) Summary plots of mean sAHP or IsAHP areas and peak magnitudes after the end of the stimulus (n = 7
for both conditions).
Values are mean SEM.

Figure S4, related to the Discussion. The majority of the sAHP is calcium-dependent and TRAM-34sensitive
(A, B) Representative whole-cell recordings of IsAHP evoked by a step command to +60 mV at 34C is
blocked by perfusion of low extracellular calcium (a), as is an inward tail current recorded in another cell
in the presence of internal 1 M TRAM-34 (B).
(C) A perforated patch whole-cell recording at the soma records an IsAHP of 6 sec duration following a
step command to +60 mV at 22 C. The IsAHP is amplified upon generation of a calcium current during
the step command, visible as an all-or-none unclamped calcium spike (inset). Recordings in (B) were
conducted using a KMeSO4-based internal solution with 0.1 EGTA, ATP, GTP, creatine, and 1 M
TRAM-34.
4

(D, E) Representative whole-cell recording from a rat CA1 pyramidal cell applying a stimulus protocol
consisting of 1 nA, 2 ms current pulses delivered at 50 Hz for 3 sec to test the role for the Na-K pump in
generating the sAHP at 35C (as in Gulledge et al. 2013). Internal infusion of TRAM-34 (1 M) blocks a
large component of the sAHP. Subsequent addition of bath perfused 20 M ouabain blocked the small
remaining TRAM-insensitive component of the sAHP. All recordings in (D, E) were conducted in the
presence of 100 nM apamin, 10 M XE-991, and 50 M picrotoxin.
(E) Bar plot of mean sAHP area following the end of the current pulse train for records such as shown in
(D).
Values are mean SEM. *, p < 0.05. Sample numbers for mean values are shown in brackets.

Table S1, related to Figures 3, 4 and 5. Values of Kd/IC50/EC50 and the actual applied concentrations of
agonists and antagonists used.

Inhibitors
Tetrodotoxin
Apamin
TEA

4-AP

CsCl
TRAM-34

Senicapoc
ChTx

Maurotoxin
Ni2+
Cd2+
8-bromo-cAMP

PKACat

XE-991
Activators
DC-EBIO
SKA-31

Actions
Nav1.1-1.4,
1.6-1.7
SK1-3
BK
Kv1.1
Kv1.6
Kv3.x
Kv1.x
Kv2.x
Kv3.x
Kv4.2
HCN1-4
KCa3.1

KCa3.1
KCa3.1
Kv1.2
Kv1.3
Kv1.6
KCa3.1
Kv1.2
Cav3.x
Cav1.x
Cav2.3
PKA-I andII

Kd

20-25 nM

IC50/EC50

Applied
External

1-24 nM

1 M

2-12 nM
80-330 M
0.5 mM
1.7-7 mM
0.09-0.3 mM
0.1-1.5 mM
0.5-4.5 mM
0.02-1.2mM
2-5 mM
0.16-0.20 mM
20 nM

100 nM
5 mM

11 nM
10 nM
1.7-17 nM
0.5-2.6 nM
1 nM
1 nM
0.1 nM
50-300 M

100 nM
100 nM

2.14 M

5 mM

2 mM

2 mM
1 M

1 M

(Coetzee et al., 1999)

(Stieber et al., 2005)

(Jenkins et al., 2013; Wulff
et al., 2001; Wulff et al.,
2000)
(Maezawa et al., 2012)
(Van Renterghem et al.,
1995)
(Coetzee et al., 1999)

100 nM

(Castle et al., 2003)

300 M
30 M

(Lee et al., 1999)

(Hobai et al., 1997)
(Catterall et al., 2005b)
(Schwede et al., 2000)
(Hoffman and Johnston,
1998)
(Wong and Schlichter, 2014)

100 M

0.0150.019 nM

References
(Catterall et al., 2005a;
Zimmer, 2010)
(Sah and Faber, 2002)
See (Coetzee et al., 1999)
for pharmacology of TEA

0.8 M

100 U/ml

KCa3.1
CaMbinding
domain
Kv7.x

2.2 M

10 M

(Schwarz et al., 2006)

KCa3.1

1 M
260 nM

0.1 M
1 M

(Wulff and Castle, 2010)

(Sankaranarayanan et al.,
2009)

Applied
Internal

Supplemental Experimental Procedures:

Slice preparation:

Animals were anaesthetized by isoflurane inhalation and 270 m dorsal hippocampal slices cut in ice-cold
solution (mM): 215 sucrose, 25 NaHCO3, 20 D-glucose, 2.5 KCl, 0.5 CaCl2, 1.25 NaH2PO4 and 3 MgCl2
bubbled with carbogen gas. The slices were incubated for 10-15 min (34C) in artificial cerebrospinal fluid
(aCSF) composed of (mM): 125 NaCl, 3.25 KCl, 1.5 CaCl2, 1.5 MgCl2, 25 NaHCO3, and 25 D-glucose
bubbled with carbogen gas. Slices were stored at room temperature until recordings conducted at 32-34 C
as a submerged preparation.
Evoking the sAHP

As earlier reported (Wu et al., 2004), the amplitude of the evoked sAHP could differ between cells, with
up to 20% exhibiting little or no detectable sAHP. The current study focused on those cells exhibiting a
detectable sAHP. The combination of apamin, XE-991, and picrotoxin slightly reduced but did not block
spike accommodation evoked during current pulse injection (Figure S2A; n = 8). There was also little
qualitative change in the firing response to SR stimulus trains, although the early phase of the subsequent
AHP (SK-mediated) was reduced and the sAHP was slightly more prominent, likely due to an increase in
calcium influx in the absence of SK channels (Figure S2B; n = 8). The ability to detect both AMPA- and
NMDA-mediated components of the SR-evoked sAHP was also maintained in the presence of apamin, XE991, and picrotoxin (Figure S2C; n = 8). Recording the IsAHP following a step command or a burst of SR
stimuli established that these compounds reduced primarily an early component of outward current
consistent with an SK-mediated response (Figures S2D and S2E; n = 8). Recordings of IsAHP using
whole-cell or perforated patch configurations were stable over 30 min time (Figure S3, n = 7), revealing
minimal influence of washout or a change in access resistance that could account for the actions of applied
drugs.
Patch recordings

Patch recordings were made using Multiclamp 700B amplifiers and Digidata 1440A with DC-10 kHz
bandpass filter and pClamp software. Pipettes were constructed from 1.5 mm O.D. fiber-filled glass (A-M
Systems) with resistance of 4-8 M. Series resistance was compensated with bridge balance circuitry for
current clamp recordings and during voltage clamp recordings by up to 80% compensation. Negative bias
current of < 200 pA was applied during current clamp recordings to maintain a subthreshold resting
potential at ~-65 mV. For whole-cell experiments, control recordings were made >5 min after break in to
promote full stabilization with the internal solution. During whole-cell recording, cells were rejected for
any drift in access resistance of > 20%. On-cell recordings were obtained with a 10 kHz cutoff filter and
processed offline by filtering at 240-500 Hz (Bessel 8-pole).
Solutions and drug applications

Chemicals were obtained from Sigma unless noted. The following drugs were used to block the identified
ion channels to isolate IKCa: BK (TEA, 5 mM; IbTx, 100 nM), SK (apamin, 100 nM), Kv7 (XE-991, 10
M), Kv4.x (4-AP, 5 mM external, 2 mM internal), Kv1.x (TEA, 5 mM; ChTx, 100 nM; MTx, 100 nM),
sodium (TTX, 200 nM - 1 M), HVA calcium (CdCl2, 30 M), LVA calcium (NiCl2, 300 M), HCN
(external CsCl, 2 mM). TRAM-34 (1 M) was applied most often by internal perfusion of the electrode
and by bath perfusion when indicated. Synaptic responses were blocked by: GABA-A (picrotoxin, 50 M),
7

GABA-B (CGP55485, 1 M, Tocris), NMDA (DL-AP5, 25 M, Ascent Scientific), and AMPA/KA

(DNQX, 10 M, Tocris Scientific). NiCl2, CdCl2, CsCl, TEA, 4-AP, and picrotoxin were prepared daily
from stock solutions. DNQX, DC-EBIO, and SKA-31 were first dissolved in dimethylsulfoxide (DMSO)
(final DMSO < 0.1%). Senicapoc was a gift of H. Wulff (UC Davis). Pressure ejection of toxins were
carried out in a HEPES (10)-buffered aCSF of (mM): 150 NaCl, 3.25 KCl, 1.5 CaCl2, 1.5 MgCl2, 10
HEPES, 20 D-glucose, pH 7.3. Pressure electrodes included BSA (0.1%) to reduce drug binding, and food
coloring (1:100) to visualize the region of drug ejection.
Patch recording configurations and media
External solution for all patch configurations: External solution (aCSF) was composed of (mM): 125
NaCl, 3.25 KCl, 1.5 CaCl2, 1.5 MgCl2, 25 NaHCO3, and 25 D-glucose bubbled with carbogen. All slice
recordings were conducted at 32 -34 C.
Whole-cell internal: Current-clamp whole-cell recording solution consisted of (mM): 130 KMeSO3, 0.1
EGTA, 10 HEPES, 7 NaCl, 0.3 MgCl2, pH 7.3 with KOH, providing an ECl of -75 mV and EK of -97 mV; .
5 di-tris-creatine phosphate , 2 Tris-ATP and 0.5 Na-GTP were added daily from frozen stocks to wholecell recording solutions.
Perforated-Patch internal: Gramicidin-perforated patch recordings solution contained (mM): 10 KCl, 135
K-Gluconate, 10 HEPES, 1 MgCl2, 75 g/ml gramicidin prepared in DMSO (< 0.01% DMSO in the
internal).
Outside-out internal: Outside-out voltage clamp recordings to isolate IKCa currents used an electrode
solution of (mM): 140 KCl, 2.83 MgCl2, 10 HEPES, 5 EGTA, 2 4-AP, 4.25 CaCl2 (1 M [Ca]i,
Maxchelator Ca/Mg/ATP/EGTA Calculator, 0.165 ionic strength), pH 7.3 (EK -99 mV); 5 di-tris-creatine
phosphate, 2 Tris-ATP and 0.5 Na-GTP were added daily from frozen stocks to outside-out internal
solution. The external solution further contained synaptic blockers and TTX, apamin, XE-991, TEA, 4-AP,
and CsCl to block SK, BK, Na, Kv1, Kv4, Kv7, LVA and HVA Ca2+ channels, and HCN channels (as
above, and detailed in Table S1).
On-cell internal: On-cell recording electrodes to isolate IKCa channels contained HEPES-buffered aCSF
with (mM): 150 NaCl, 3.25 KCl, 1.5 CaCl2, 1.5 MgCl2, 10 HEPES and 20 D-glucose, pH 7.3. The electrode
solution further contained synaptic blockers and TTX, apamin, XE-991, TEA, 4-AP, and CsCl to block SK,
BK, Na, Kv1, Kv4, Kv7 channels (as above, and detailed in Table S1). The external bath solution for oncell recordings contained the same drugs as the electrode solution but did not contain excitatory synaptic
blockers (DL-AP5 and DNQX) or TTX to allow for synaptic excitation.
tsA-201 cells

tsA-201 cells were maintained as previously described (Engbers et al., 2012; Rehak et al., 2013) and
transiently transfected with cDNA (5 g/l) of Kv7.3, BK, TMEM16B or IKCa and calmodulin together
with cDNA for GFP to identify cells successfully transfected. Currents were recorded at room temperature
(22 C) in aCSF consisting of (mM): 120 NaCl, 3 NaHCO3, 4.2 KCl, 1.2 KH2PO4, 1.5 MgCl2, 10 DGlucose, 10 HEPES and 1.5 CaCl2 , pH adjusted to 7.3 with NaOH. The following internal solutions were
used for voltage-clamp recordings in tsA-201 cells. For recording IKCa channels, electrodes were filled
8

with (mM): 140 KCl, 5 EGTA, 5 HEPES, 2.83 MgCl2, 4.25 CaCl2 (1.1 M free [Ca]i, Maxchelator
Ca/Mg/ATP/EGTA Calculator, http://maxchelator.stanford.edu/CaMgATPEGTA-TS.htm, 0.165 ionic
strength), pH adjusted to 7.3 with KOH. For recording BK currents, electrodes were filled with (mM): 140
KCl, 5 HEPES, 0.1 EGTA, 0.5 MgCl, 5 ATP, 1 GTP, pH adjusted to 7.3. For recording Kv7.3 currents,
electrodes were filled with (mM): 140 KCl, 0.1 EGTA, 2.5 MgCl2, 10 HEPES, pH adjusted to 7.3. For
recording TMEM16B currents, electrodes were filled with (mM): 15.22 CsCl, 124.78 CsMeSO3, 5 EGTA,
10 HEPES, 1.73 MgCl2, 1.68 CaCl2 (100 nM free [Ca]i, Maxchelator Ca/Mg/ATP/EGTA Calculator,
http://maxchelator.stanford.edu/CaMgATPEGTA-TS.htm), with ECl = -45 mV, and pH adjusted to 7.3.
TMEM16B currents were activated by giving 200 ms voltage steps from a holding potential of -40 mV.
Stimulation

Synaptic input was evoked using a concentric bipolar electrode (Frederick Haer, CBCMX75(JL2))
positioned in the mid SR and driven by a stimulus isolation unit (Digitimer, 0.1-0.2 ms pulse width).
Data Analysis and Statistical Methods

sAHP areas were measured as the area under (in current clamp mode) or over (in voltage clamp mode)
baseline, defined as the mean voltage current or voltage level preceding the current or voltage protocol,
from 200 ms after the protocol to 7 secs after the protocol. Amplitude of the sAHP was measured as the
voltage level 200 ms after the current or stimulus protocol. Mean single channel conductance was
calculated for hyperpolarizing potentials where channel amplitude was best delineated. Data were analyzed
using Clampfit 10 software and custom Matlab R2007B scripts, and statistical analysis in OriginPro 8.
Paired-sample Student t-tests were used unless otherwise indicated.

Supplemental References
Castle, N.A., London, D.O., Creech, C., Fajloun, Z., Stocker, J.W., and Sabatier, J.M. (2003). Maurotoxin: a potent
inhibitor of intermediate conductance Ca2+-activated potassium channels. Mol. Pharmacol. 63, 409-418.
Catterall, W.A., Goldin, A.L., and Waxman, S.G. (2005a). International Union of Pharmacology. XLVII.
Nomenclature and structure-function relationships of voltage-gated sodium channels. Pharmacol. Rev. 57, 397-409.
Catterall, W.A., Perez-Reyes, E., Snutch, T.P., and Striessnig, J. (2005b). International Union of Pharmacology.
XLVIII. Nomenclature and structure-function relationships of voltage-gated calcium channels. Pharmacol. Rev. 57,
411-425.
Coetzee, W.A., Amarillo, Y., Chiu, J., Chow, A., Lau, D., McCormack, T., Moreno, H., Nadal, M.S., Ozaita, A.,
Pountney, D., et al. (1999). Molecular diversity of K+ channels. Ann. N. Y. Acad. Sci. 868, 233-285.
Engbers, J.D.T., Anderson, D., Asmara, H., Rehak, R., Mehaffey, W.H., Hameed, S., McKay, B.E., Kruskic, M.,
Zamponi, G.W., and Turner, R.W. (2012). Intermediate conductance calcium-activated potassium channels modulate
summation of parallel fiber input in cerebellar Purkinje cells. Proc. Natl. Acad. Sci. U S A 109, 2601-2606.
Hobai, I.A., Bates, J.A., Howarth, F.C., and Levi, A.J. (1997). Inhibition by external Cd2+ of Na/Ca exchange and
L-type Ca channel in rabbit ventricular myocytes. Am. J. Physiol. 272, H2164-2172.
Hoffman, D.A., and Johnston, D. (1998). Downregulation of transient K+ channels in dendrites of hippocampal CA1
pyramidal neurons by activation of PKA and PKC. J. Neurosci. 18, 3521-3528.
Jenkins, D.P., Yu, W., Brown, B.M., Lojkner, L.D., and Wulff, H. (2013). Development of a QPatch Automated
Electrophysiology Assay for Identifying KCa3.1 Inhibitors and Activators. Assay Drug Dev. Technol. 11, 551-560.
Lee, J.H., Gomora, J.C., Cribbs, L.L., and Perez-Reyes, E. (1999). Nickel block of three cloned T-type calcium
channels: low concentrations selectively block alpha1H. Biophys. J. 77, 3034-3042.
Maezawa, I., Jenkins, D.P., Jin, B.E., and Wulff, H. (2012). Microglial KCa3.1 Channels as a Potential Therapeutic
Target for Alzheimer's Disease. Int. J. Alzheimers Dis. 2012, 868972.
Rehak, R., Bartoletti, T.M., Engbers, J.D.T., Berecki, G., Turner, R.W., and Zamponi, G.W. (2013). Low voltage
activation of KCa1.1 current by Cav3-KCa1.1 complexes. PLoS One 8, e61844.
Sah, P., and Faber, E.S. (2002). Channels underlying neuronal calcium-activated potassium currents. Prog.
Neurobiol. 66, 345-353.
Sankaranarayanan, A., Raman, G., Busch, C., Schultz, T., Zimin, P.I., Hoyer, J., Kohler, R., and Wulff, H. (2009).
Naphtho[1,2-d]thiazol-2-ylamine (SKA-31), a new activator of KCa2 and KCa3.1 potassium channels, potentiates
the endothelium-derived hyperpolarizing factor response and lowers blood pressure. Mol. Pharmacol. 75, 281-295.
Schwarz, J.R., Glassmeier, G., Cooper, E.C., Kao, T.C., Nodera, H., Tabuena, D., Kaji, R., and Bostock, H. (2006).
KCNQ channels mediate IKs, a slow K+ current regulating excitability in the rat node of Ranvier. J. Physiol. 573,
17-34.
Schwede, F., Maronde, E., Genieser, H., and Jastorff, B. (2000). Cyclic nucleotide analogs as biochemical tools and
prospective drugs. Pharmacol. Ther. 87, 199-226.
Stieber, J., Stockl, G., Herrmann, S., Hassfurth, B., and Hofmann, F. (2005). Functional expression of the human
HCN3 channel. J. Biol. Chem. 280, 34635-34643.
Van Renterghem, C., Vigne, P., and Frelin, C. (1995). A charybdotoxin-sensitive, Ca(2+)-activated K+ channel with
inward rectifying properties in brain microvascular endothelial cells: properties and activation by endothelins. J.
Neurochem. 65, 1274-1281.
Wong, R., and Schlichter, L.C. (2014). PKA reduces the rat and human KCa3.1 current, CaM binding, and Ca2+
signaling, which requires Ser332/334 in the CaM-binding C terminus. J. Neurosci. 34, 13371-13383.
Wu, W.W., Chan, C.S., and Disterhoft, J.F. (2004). Slow afterhyperpolarization governs the development of NMDA
receptor-dependent afterdepolarization in CA1 pyramidal neurons during synaptic stimulation. J. Neurophysiol. 92,
2346-2356.
Wulff, H., and Castle, N.A. (2010). Therapeutic potential of KCa3.1 blockers: recent advances and promising trends.
Expert Rev. Clin. Pharmacol. 3, 385-396.
10

Wulff, H., Gutman, G.A., Cahalan, M.D., and Chandy, K.G. (2001). Delineation of the clotrimazole/TRAM-34
binding site on the intermediate conductance calcium-activated potassium channel, IKCa1. J. Biol. Chem. 276,
32040-32045.
Wulff, H., Miller, M.J., Hansel, W., Grissmer, S., Cahalan, M.D., and Chandy, K.G. (2000). Design of a potent and
selective inhibitor of the intermediate-conductance Ca2+-activated K+ channel, IKCa1: a potential
immunosuppressant. Proc. Natl. Acad. Sci. U S A 97, 8151-8156.
Zimmer, T. (2010). Effects of tetrodotoxin on the mammalian cardiovascular system. Mar. Drugs 8, 741-762.

11