KCNE3 is an inhibitory subunit of the Kv4.3 potassium channel

BBRC Biochemical and Biophysical Research Communications 346 (2006) 958–967 www.elsevier.com/locate/ybbrc

KCNE3 is an inhibitory subunit of the Kv4.3 potassium channel Alicia Lundby, Søren-Peter Olesen

*

The Danish National Research Foundation Centre for Cardiac Arrhythmia, Department of Medical Physiology, University of Copenhagen, The Panum Institute, 3 Blegdamsvej, DK-2200 Copenhagen N, Denmark Received 27 May 2006 Available online 9 June 2006

Abstract The mammalian Kv4.3 potassium channel is a fast activating and inactivating K+ channel widely distributed in mammalian tissues. Kv4.3 is the major component of various physiologically important currents ranging from A-type currents in the CNS to the transient outward potassium conductance in the heart (I to ). Here we show that the KCNE3 b-subunit has a strong inhibitory effect on current conducted by heterologously expressed Kv4.3 channels. KCNE3 reduces the Kv4.3 current amplitude, and it slows down the channel activation and inactivation as well as the recovery from inactivation. KCNE3 also inhibits currents generated by Kv4.3 in complex with the accessory subunit KChIP2. We find the inhibitory effect of KCNE3 to be specific for Kv4.3 within the Kv4 channel family. Kv4.3 has previously been shown to interact with a number of b-subunits, but none of the described subunit-interactions exert an inhibitory effect on the Kv4.3 current.  2006 Elsevier Inc. All rights reserved. Keywords: Voltage-gated potassium channel; I to current; Inhibitory b-subunit; MIRP2; Kv4.3; KCND3; KCNE3

Voltage-gated potassium (Kv) channels are key attenuating control elements of cellular excitability. Kv channels are found in almost all cell types, and their importance in cellular physiology is evidenced by the number of diseases caused by their dysfunction; CNS disorders, cardiac arrhythmias, and epilepsy [1]. Kv channels comprise a large family of proteins; thus far twelve Kv channel a-subunit families (Kv1–12) have been identified in mammals [2]. Within the Kv4 subfamily three a-subunits (Kv4.1–4.3) have been described. These are all fast activating and inactivating channels, and they recover from inactivation faster than any other Kv channels characterized. Kv4 channels are mainly expressed in the brain and the heart [3–7]. Although not always necessary, many K+ channels have associated subunits. Members of different families of b-subunits have been reported to interact with Kv4.3, namely KCNE1-2, Kvb1-3, KChIP2, and DPPX-Y [8–12]. Currents recorded from native cells have in several cases been found to be similar to currents conducted by complexes *

Corresponding author. Fax: +45 35327449. E-mail address: spo@mfi.ku.dk (S.-P. Olesen).

0006-291X/$ - see front matter  2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.06.004

of K+ channel a-subunits and KCNE b-subunits. Thus, I Ks like currents can be generated by KCNQ1 + KCNE1 [13], I Kr -like currents by hERG + KCNE2 [14], and A-type currents in skeletal muscle by Kv3.4 + KCNE3 [15]. The KCNE family consists of five members (KCNE1–5) that are all found to modulate the KCNQ1 channel in vitro [13,16–19]. KCNE1–3 have additionally been shown to alter the currents of other KCNQ channels, ERG, HCN, and members of the Kv1-4 channel families [17,20–23]. Previously, no inhibitory effect of any b-subunit on the Kv4.3 current has been reported. In the present study, we demonstrate a drastic inhibition of Kv4.3 currents by the KCNE3 subunit. The inhibition is observed on heterologously expressed channels in Xenopus laevis oocytes as well as in HEK-293 cells. The inhibition is also observed for Kv4.3 channels in complex with the accessory subunit KChIP2. We find the inhibitory effect of KCNE3 to be specific for Kv4.3 within the Kv4 channel family as KCNE3 does not affect currents carried by Kv4.1 or Kv4.2. Given that Kv4.3 and KCNE3 are both found in human heart and brain [15,20,24], the interaction described in this study might have interesting physiological relevance.

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

Materials and methods

959

(140 mM NaCl, 4 mM KCl, 2 mM CaCl2, 1 mM MgCl2, and 10 mM Hepes, pH 7.4). All experiments were performed at room temperature.

Molecular biology Analysis of electrophysiological data cDNAs coding for hKv4.1 (AF166003) and hKv4.2 (NM012281) were a kind gift from D. Isbrandt and cDNA coding for hKv4.3 (short isoform, NM172198) was kindly provided by O. Pongs. cDNAs encoding hKv4.3, hKCNE3 (AF302494), and hKChIP2.2 (AF199598) were subcloned into the expression vector pXOOM and cDNAs encoding hKv4.1 and hKv4.2 were subcloned into the expression vector pGEM. After linearization cDNA was purified by High Pure PCR Product Purification Kit (Roche A/S Diagnostics, Hvidovre, Denmark). Synthesis of capped cRNA was performed by in vitro transcription using the mMESSAGE mMACHINE T7 kit (Ambion, Austin, TX). cRNA was extracted and dissolved using the MegaClear kit (Ambion, Austin, TX) to a final concentration between 0.2 and 1 lg/ll. The concentration was determined photometrically. The integrity of the transcripts was confirmed by agarose gel electrophoresis and cRNA was stored at 80 C until injection. Transient expression in X. laevis oocytes and HEK-293 cells Xenopus oocytes. Female X. laevis frogs were anaesthetized for 15– 20 min in 2 g/L Tricain (3-aminobenzoic acid ethyl ester, Sigma A-5040) before an ovarian lobe was removed from the abdominal cavity through a small insertion. Oocytes were defolliculated enzymatically by incubation in 1% collagenase (Boehringer Mannheim, 1088831) and 0.1% trypsin inhibitor (Sigma T-2011) in Kulori medium (90 mM NaCl, 1 mM KCl, 1 mM MgCl2, 1 mM CaCl2, and 5 mM Hepes, pH 7.4) for 1 h followed by five washes in Kulori containing 0.1% BSA (Sigma A-6003), and incubation for 1 h in a hypertonic phosphate buffer (100 mM K2HPO4, pH 6.5). Subsequently, stage V and VI oocytes were selected, and kept in Kulori medium for 24 h at 19 C before injection of 50 nl cRNA. cRNA was injected using a Nanoject microinjector (Drummond Scientific, Broomall, PA). Oocytes were kept at 19 C in Kulori medium for 1–4 days before measurements were performed. HEK-293 cells. HEK-293 cells were cultured in DMEM (Life Technologies, Carlsbad, CA) supplemented with 10% FCS (Life Technologies, Carlsbad, CA) and grown in T25 flasks (Nunc) at 37 C in 5% CO2. The day of transfection 50% of the T25 flask surface was covered by cells; cells were transfected with 0.5 lg of each DNA sample using Lipofectamine (Life Technologies, Carlsbad, CA) according to the manufacturer’s instructions. In all cases GFP was cotransfected as a reporter. Forty-eight hours post transfection the cells were trypsinized and transferred to coverslips for electrophysiological experiments. Electrophysiology Two-electrode voltage-clamp experiments. Currents were recorded using a two-electrode voltage-clamp amplifier (CA-1B, Dagan, Minneapolis, MN). Electrodes were pulled from borosilicate glass capillaries on a horizontal patch electrode puller and had a tip resistance between 0.3 and 2.0 MX when filled with 1 M KCl. During experiments oocytes were placed in a small chamber (volume: 200 ll) connected to a continuous flow system, and channel activity was measured in Kulori medium. All experiments were performed at room temperature 48 h after injection unless indicated otherwise. Patch-clamp experiments. All experiments were performed in the whole-cell configuration. Currents were recorded using an EPC-9 amplifier (HEKA electronics, Germany), data were sampled with Pulse software (HEKA electronics, Germany), and analyzed with IGOR software (Wavemetrics, Lake Oswego, OR). Rs was compensated 80% and did not exceed 7 MX. Electrodes were pulled from borosilicate glass capillaries on a horizontal patch electrode puller and had a tip resistance between 1.8 and 2.4 MX when filled with intracellular Ringer solution (110 mM KCl, 31/10 mM KOH/EDTA, 5.17 mM CaCl2, 1.42 mM MgCl2, and 10 mM Hepes, pH 7.2). During recordings transfected HEK-293 cells were exposed to a continuous flow of extracellular NaCl Ringer solution

Data analysis was performed using IGOR software. All values are shown as means ± SEM. Inactivation time-constants as well as the timeconstant for recovery from inactivation were found by fitting the data to a single exponential, i.e., I(t) = yo + A · exp (t/s). The difference between the peak current amplitude and the current at the end of a test pulses in absolute current values is referred to as the transient outward current. Current– voltage relations were obtained by plotting the transient outward current as a function of the test potential. The inactivation as well as the activation kinetics were analyzed by fitting the current–voltage relations to a two-state Boltzmann distribution of the form I(V) = 1/(1 + exp((V1/2-V)/a)), where V1/2 is the potential for half-maximal inactivation or activation, respectively, and a is the slope factor. Data shown for inactivation time-constants, time to peak and mean current levels are found from current traces recorded at +60 mV. n indicates the number of independent experiments. Comparison of the biophysical properties in the presence and absence of accessory subunits was performed using an unpaired t-test for cases where two groups were compared and using a one-way ANOVA followed by Dunnet’s multiple comparison tests when more than two groups were compared. Data were considered significant at P < 0.05.

Results KCNE3 inhibits the Kv4.3 current and affects its gating properties To investigate if KCNE3 b-subunits affect currents carried by Kv4.3 potassium channels, we expressed Kv4.3 alone or together with the KCNE3 b-subunit in X. laevis oocytes and performed two-electrode voltage-clamp recordings. Fig. 1A shows representative current traces of Kv4.3 recorded from oocytes subjected to a standard activation protocol. Oocytes were held at 80 mV before recording currents for 2 s at potentials ranging from 140 mV to +100 mV in 20 mV increments. Tail currents were measured by 1.5 s pulses to +40 mV. Oocytes expressing Kv4.3 channels showed fast activating and inactivating voltage-dependent currents at potentials positive to 40 mV. The steady-state activation curve obtained by plotting the normalized transient outward current as a function of clamp potential revealed half-maximal activation of 24 ± 2 mV (n = 10, Fig. 1). Likewise, the steadystate inactivation curve derived from the tail currents revealed a half-maximal inactivation of 61 ± 2 mV (n = 10). Oocytes co-expressing Kv4.3 and KCNE3 were subject to a slightly modified clamp protocol. In this case, we only tested the Kv4.3 + KCNE3 channel’s response between 100 mV and +60 mV (Fig. 2A). Oocytes co-expressing Kv4.3 and KCNE3 were not clamped at potentials negative to 100 mV and positive to +60 mV, because we observed slowly activating endogenous currents in the oocytes at these potentials of considerable magnitude compared to the small magnitude of Kv4.3 + KCNE3 currents. For oocytes expressing Kv4.3 alone this was not a consideration since the Kv4.3 current amplitudes were 10-fold larg-

960

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

A

B 12

τinac = 109 ± 8 ms

20

15

I (nA)

I (μA)

10

10

8 6 4

5

2 0

0 0

1

2

0.0

3

0.2

0.4

C

D

1.0

V½ = 24 ± 2 mV

0.4

1.0

0.4 0.2

0.0

0.0 -100

-60

-20

20

60

100

V½ = -61 ± 2 mV

0.6

0.2

-140

0.8

0.8

I / Imax

I / Imax

0.8 0.6

0.6

time (s)

time (s)

-140

-100

-60

-20

E

20

60

100

V (mV)

V (mV) 1.0

Imax2 / Imax1

0.8

τrec = 490 ± 24 ms

0.6 0.4

1

+40 mV

2

Δt

0.2 -80 mV 250 ms

250 ms

0.0 0

500

1000

1500

2000

2500

3000

time (ms) Fig. 1. Characterization of Kv4.3 currents. (A) Representative current traces recorded from Xenopus laevis oocytes expressing Kv4.3. Currents were elicited from a holding potential of 80 mV in 2 s voltage steps to potentials ranging from 140 mV to +100 mV in 20 mV increments, and tail currents were recorded at +40 mV. (B) Representative current traces recorded from HEK-293 cells transfected with Kv4.3. Currents were elicited from a holding potential of 80 mV in 500 ms voltage steps to potentials ranging from 140 mV to +100 mV in 20 mV increments. Tail currents were recorded at +40 mV. (C) Steady-state activation curve of Kv4.3 measured from oocytes. (D) Steady-state inactivation curve for Kv4.3 measured from oocytes. The normalized transient outward tail current recorded at +40 mV is shown as a function of the pre-pulse potential. (E) Kv4.3 recovery from inactivation. Oocytes expressing Kv4.3 were clamped by two 250 ms pulses at +40 mV with variable inter-pulse interval. The holding potential and the potential during the recovery step was 80 mV. The fraction of the transient outward current recorded at the second pulse compared to the one recorded at the first pulse is shown as a function of the inter-pulse time.

er. For comparisons between Kv4.3 and Kv4.3 + KCNE3 currents data recorded at +60 mV were used. Co-expression of Kv4.3 and KCNE3 resulted in a drastic reduction of the characteristic transient outward Kv4.3 current (Fig. 2). The average transient outward current at +60 mV for Kv4.3 was 12 ± 2 lA (n = 10) and for Kv4.3 + KCNE3 was 0.5 ± 0.1 lA (n = 12) (P < 0.0001). KCNE3 also affected the gating properties of Kv4.3

(Table 1). Inactivation of the Kv4.3 channel is twice as slow when the channel is co-expressed with KCNE3 (109 ± 8 ms versus 220 ± 16 ms, P < 0.0001), and it takes the channel twice as long to recover from inactivation when KCNE3 is present (490 ± 24 ms versus 979 ± 67 ms, P < 0.0001). Also, the time from the onset of a pulse to the maximum current is recorded (time to peak), is increased in the presence of KCNE3 (13 ± 1 ms versus

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

τinac = 220 ± 16 ms

B

2

1.5

I (nA)

I (μA)

A 3

1 0

1.0 0.5 0.0

0

1

2

0.0

3

0.2

0.4

0.6

0.8

time (s)

time (s)

C

D 1.0

1.0

0.8

0.8

I / Imax

I / Imax

961

V½ = 6 ± 2 mV

0.6 0.4

0.4

0.2

0.2

0.0

0.0 -100

-60

-20

20

V½ = -72 ± 2 mV

0.6

-100

60

-60

-20

V (mV)

20

60

V (mV)

E 1.2

Imax2 / Imax1

1.0 0.8

τrec = 979 ± 67 ms

0.6 +40 mV

0.4 0.2

2

1 Δt

-80 mV 250 ms

250 ms

0.0 0

500

1000

1500

2000

2500

3000

time (ms) Fig. 2. Characterization of currents conducted by co-expressed Kv4.3 and KCNE3. (A) Representative current traces obtained from Xenopus laevis oocytes co-expressing Kv4.3 and KCNE3 in a molar 1:1 ratio. Currents were elicited from a holding potential of 80 mV by 2 s voltage steps to potentials ranging from 100 mV to +60 mV in 20 mV increments, and tail currents were recorded at +40 mV. (B) Representative current traces obtained from HEK-293 cells transfected with Kv4.3 and KCNE3. Currents were elicited from a holding potential of 80 mV to clamp potentials between 100 mV and +60 mV in 20 mV increments for 500 ms followed by 300 ms tail current recording at +40 mV. (C) Steady-state activation of Kv4.3 + KCNE3 measured from oocytes. (D) Steady-state inactivation of Kv4.3 + KCNE3 measured from oocytes. (E) Recovery from inactivation of Kv4.3 + KCNE3 measured from oocytes using the same pulse protocol as in Fig. 1.

22 ± 2 ms, P < 0.001). Finally, the inactivation as well as the activation of Kv4.3 appears to be slightly left-shifted in the presence of KCNE3 ( 61 ± 2 mV versus 72 ± 2 mV P = 0.001 and 24 ± 2 mV versus 6 ± 2 mV P < 0.0001). However, since the activation and inactivation curves for Kv4.3 + KCNE3 are not fully saturated these data have to be interpreted with caution. We verified that KCNE3 modulates Kv4.3 currents independently of expression system by performing patch-clamp experiments on transfected HEK-293 cells. Current traces from transfected HEK-293 cells were elicited from a holding potential of 80 mV to clamp

potentials ranging from 140 mV to +100 mV for Kv4.3 and from 100 mV to +60 mV for Kv4.3 + KCNE3 for 500 ms in 20 mV increments. Tail currents were recorded at +40 mV for 300 ms (Figs. 1B and 2B). We measured the transient outward current at +60 mV and found that the co-expression of Kv4.3 and KCNE3 also inhibited the transient outward Kv4.3 current in HEK-293 cells. The current densities were 411 ± 50 pA/pF for Kv4.3 (n = 9) versus 122 ± 35 pA/ pF for Kv4.3 + KCNE3 (n = 6) (P < 0.001). We conclude that KCNE3 can act as an inhibitory subunit of Kv4.3 independently of the expression system.

962

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

Table 1 Functional effects of KCNE3 on Kv4.3 and Kv4.3 + KChIP2 currents expressed in Xenopus laevis oocytes Parameter

Kv4.3

4 Kv4.3 4 KCNE3

4 Kv4.3 4 KChIP2

4 Kv4.3 4 KChIP2 1 KCNE3

4 Kv4.3 4 KChIP2 2 KCNE3

4 Kv4.3 4 KChIP2 4 KCNE3

Mean current at 60 mV (lA) Recovery from inactivation (srec in ms) Steady-state activation (V1/2 in mV) Steady-state inactivation (V1/2 in mV) Inactivation time-constant (sinac in ms) Time-to-peak at 60 mV (speak in ms) Number of cells (n)

12 ± 2 490 ± 24 24 ± 2 61 ± 2 109 ± 8 13 ± 1 10

0.5 ± 0.1  979 ± 67  6 ± 2  72 ± 2  220 ± 16  22 ± 2  12

42 ± 2 43 ± 1 24 ± 2 53 ± 1 150 ± 1 10 ± 1 10

6 ± 2§ 76 ± 4§ 14 ± 2§ 54 ± 2 222 ± 11§ 25 ± 2§ 19

4 ± 1§ 110 ± 8§ 16 ± 1§ 56 ± 2 302 ± 7§ 36 ± 2§ 10

6 ± 1§ 118 ± 3§ 5 ± 3§ 56 ± 1 251 ± 14§ 31 ± 2§ 9

  §

Significantly different from Kv4.3 (P < 0.05). Significantly different from 4 Kv4.3 + 4 KChIP2 (P < 0.05).

KCNE3 amount determines level of Kv4.3 inhibition We next investigated whether the level of Kv4.3 current inhibition was dependent on the amount of KCNE3. Fig. 3 shows the time-course of current expression at different molar ratios. We co-injected Kv4.3 and KCNE3 in molar ratios ranging from 1:1 to 10:1 and followed the amplitude of the transient outward current for 4 days. We did not test oocytes injected with an excess of KCNE3 due to the nearly complete inhibition by KCNE3 at a 1:1 molar ratio. Ninety-two hours after injection we observed that the transient outward current was inhibited by more than 90% when injecting Kv4.3 and KCNE3 cRNA in a molar 1:1 ratio. When injecting 10 times more Kv4.3 than KCNE3 the transient outward current was only inhibited by approximately 50%. That is, the amount of KCNE3 determines the degree of Kv4.3 inhibition. The Kv4.3 current is modified by delayed injection of KCNE3 cRNA To investigate whether KCNE3 can modulate Kv4.3 currents in a system where Kv4.3 protein is previously expressed, we performed experiments with delayed injection of KCNE3 cRNA into oocytes already expressing Kv4.3 (Fig. 4). Fig. 4A illustrates the average transient outward current amplitudes measured at +60 mV for oocytes injected with Kv4.3 cRNA. Forty-eight hours after injection, when a current level of 12 ± 2 lA was reached, half of the oocytes were injected with KCNE3 cRNA and the other half was injected with water. Forty-eight hours later the Kv4.3 current was almost completely blocked in KCNE3 injected oocytes. Oocytes that had not been re-injected with KCNE3 maintained high Kv4.3 current levels. The experiment shows a significant reduction in the Kv4.3 current level the day after delayed injection of KCNE3, and a close to complete inhibition after 48 h. These data suggest that the density of Kv4.3 current can be modulated by a later expression of KCNE3 b-subunit. KCNE3 does not inhibit Kv4.1 and Kv4.2 We decided to investigate if the inhibitory effect of KCNE3 is a general feature of the Kv4 channel family

encompassing Kv4.1-3. We expressed Kv4.1 and Kv4.2 in X. laevis oocytes alone as well as co-injected with KCNE3 and performed two-electrode voltage-clamp experiments using the same protocol as in Fig. 1. Fig. 5 shows that Kv4.1 and Kv4.2 are fast activating and inactivating channels with kinetic properties similar to Kv4.3. Co-expression of Kv4.1 and KCNE3 neither affected the current amplitude nor the biophysical parameters characterizing Kv4.1 (Table 2). Co-expression of Kv4.2 and KCNE3 resulted in a small reduction in the Kv4.2 current amplitude (P = 0.046), but all kinetic parameters tested were unaffected by the co-expression of KCNE3. The data summarized in Table 2 are recorded from oocytes injected with cRNA for a- and b-subunits in a molar 1:1 ratio. To investigate if the small current reduction observed between Kv4.2 and Kv4.2 + KCNE3 is real, we injected oocytes with cRNA encoding the channel protein and KCNE3 in a molar 1:4 ratio. With the excess of KCNE3 we did not observe a decrease in Kv4.2 current amplitude (Mean current at 60 mV = 7 ± 1 lA, n = 4, P = 0.32), and we conclude that Kv4.2 and Kv4.2 + KCNE3 currents do not differ. Based on the findings that Kv4.1 and Kv4.2 currents are unaffected by the co-expression with KCNE3, it leads us to the conclusion that KCNE3 does not interact with either of these channels. We thus conclude that KCNE3 inhibition of Kv4.3 currents is specific within the Kv4 channel family. Modulation of the Kv4.3 + KChIP2 current by KCNE3 Previous studies have shown that KChIP2 immunoprecipitates with Kv4 subunits isolated both from the brain and the heart [9,25]. We therefore decided to investigate if the KCNE3 subunit can also inhibit Kv4.3 current in the presence of KChIP2. We expressed Kv4.3, KChIP2, and varying amounts of KCNE3 in X. laevis oocytes and performed two-electrode voltage-clamp experiments using the same protocol as in Fig. 1. We co-injected Kv4.3 and KChIP2 in a molar ratio of 1:1 as stoichiometric studies have shown that 4 KChIP subunits interact with four Kv4 subunits in vivo [7]. Since we do not know how many KCNE3 subunits potentially interact with one Kv4.3 + KChIP2 complex, we injected KCNE3 in ratios corre-

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

963

Kv4.3 : KCNE3

A 14

1:0 12

I (μA)

10

8

6

10 : 1

4

2

4:1 1:1

0 0

20

40

60

80

time (h)

B

C 14

8

12

I (μA)

I (μA)

8

0

0 6 0 4 2

time (s)

4:1

I (μA)

1 Kv4.3 : 1 KCNE3

4 Kv4.3 : 1 KCNE3

10 Kv4.3 : 1 KCNE3

time (s)

2

1.5

3 1 Kv4.3 : 0 KCNE3

0

2

1:1

I (μA)

I (μA)

10

10 : 1

1:0

12

0

0 0

time (s)

2

0

time (s)

2

Fig. 3. Kv4.3 inhibition depends on the amount of KCNE3. Oocytes co-injected with Kv4.3 and varying amounts of KCNE3 (molar Kv4.3:KCNE3 ratios of 1:0, 10:1, 4:1, and 1:1) were voltage-clamped by step protocols as described in Fig. 2. All data points are averaged from a number of oocytes ranging from 4 to 19. (A) The transient outward current amplitude at +60 mV was recorded and plotted at different time points after injection. The ability of KCNE3 to inhibit the Kv4.3 current is evident at all molar ratios tested. (B) The transient outward current amplitudes at +60 mV measured 96 h after injection are shown for the molar ratios tested. (C) Representative current traces elicited at +60 mV for oocytes co-expressing Kv4.3 and KCNE3 in the molar ratios indicated. Please observe different scales on the ordinates.

sponding to one, two, and four KCNE3 subunits per channel complex. As expected, co-expression of Kv4.3 with KChIP2 greatly enhanced the current amplitude (Fig. 6) [9]. When co-expressing Kv4.3 and KChIP2 with KCNE3 the current amplitude was reduced by more than 80% for all molar ratios tested. That is, KCNE3 also has an inhibitory effect on the Kv4.3 current amplitude in the presence of KChIP2 (One-way ANOVA for all KCNE3 ratios: P < 0.0001, Dunnet’s comparison test for individual KCNE3 ratios: P = 0.01). KCNE3 also affects the gating properties of the Kv4.3 + KChIP2 complex (Table 1). We found the trend in the effect of KCNE3 on Kv4.3 to be

independent of the presence of KChIP2 for all parameters tested. Discussion In the present study, we showed that KCNE3 can drastically inhibit Kv4.3 currents expressed in X. laevis oocytes and in HEK-293 cells. In oocytes the reduction observed was more than 10-fold for the average transient outward currents measured at +60 mV. We verified this severe inhibition of Kv4.3 currents by KCNE3 from five different batches of oocytes. The data shown are a subset of the

964

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

A 14 Kv4.3 12

10

a

I (μA)

b 8

6

4 c 2 Kv4.3 and d re-injected KCNE3 0 0

20

40

60

80

100

time (h) a t = 48 h

b t = 52 h

14

I (µA)

I (µA) 0

time (s)

2

0

time (s)

2

d t = 96 h

2.5

0

0

0

0

c t = 72 h

5

I (µA)

14

I (µA)

B

0

time (s)

2

0

time (s)

2

Fig. 4. Inhibition of the transient outward Kv4.3 current by delayed injection of KCNE3. (A) The transient outward current amplitude recorded at +60 mV is shown as a function of time after injection. All oocytes were injected with cRNA encoding Kv4.3 at time t = 0 and incubated at 19 C. At time t = 48 h half the oocytes were injected with KCNE3 cRNA in a molar ratio of 1:1 compared to Kv4.3 and the other half was injected with water. Oocytes that were not injected with KCNE3 maintained a stable transient outward current amplitude for the following 48 h, whereas the transient outward current amplitude in oocytes injected with KCNE3 rapidly decreased. (B) Representative current traces elicited from oocytes injected with Kv4.3 at t = 0 and reinjected with KCNE3 at t = 48 h. a is recorded just prior to KCNE3 injection, b is 4 h after KCNE3 injection, and c and d are 24 and 48 h after KCNE3 injection, respectively. All data points are averaged from a number of oocytes ranging from 6 to 19.

experiments performed, as we have only included data from oocytes subjected to all test-protocols used. In addition to decreasing the current density, KCNE3 was found to slow down the Kv4.3 gating properties: the time-constant of recovery from inactivation, the inactivation timeconstant and the time-to-peak were all increased in the presence of KCNE3. To our knowledge, the inhibitory effect of KCNE3 on Kv4.3 is the first inhibitory effect on Kv4.3 by a b-subunit ever reported. We further showed that delayed injection of KCNE3 can almost completely inhibit Kv4.3 currents. That is, transcription of KCNE3 in a system already expressing Kv4.3 channels can modulate the Kv4.3 current density, suggesting that KCNE3 transcription can act as a regulatory mechanism of the Kv4.3 current density. As we do not know the turn-over rate for Kv4.3 channels we cannot conclude whether the regulation occurs at the level of channel trafficking or by reduction of the Kv4.3 open probability or single channel conductance. By co-expressing Kv4.1 or Kv4.2 channels with KCNE3 in X. laevis oocytes we found that KCNE3 does not alter the function of these channels. This leads us to the conclusion that KCNE3 specifically modulates Kv4.3 function

within the Kv4 family. From previous studies it is known that KCNE1 and KCNE2 increase the Kv4.3 current density and that KCNE4 does not have any effect on Kv4.3 [10,23]. The effect of KCNE5 on Kv4.3 has not been investigated. From the KCNE subunits investigated it thus appears that the inhibitory effect of KCNE3 on Kv4.3 is specific within the KCNE family. Given that modulated expression of KCNE3 may act as a regulatory mechanism for Kv4.3 current density it is possible that KCNE3 down-regulates the current amplitude of Kv4.3 channels already present in the cell membrane in certain tissues. Several studies have co-localized Kv4.3 and KCNE3 to the same organs and tissue. Northern blot analysis has revealed KCNE3 protein expression in the brain, the heart, and in skeletal muscle [15,20]. Similarly, by Northern blot analysis Kv4.3 protein expression has been detected in the brain and in the heart [26]. Expression of Kv4.3 and KCNE3 in the human heart (left atrium and ventricle) has further been confirmed by RT-PCR studies [24]. The fact that expression of Kv4.3 and KCNE3 has been reported in the same tissues opens for the possibility that the channel and subunit interact in vivo. Since Kv4.3 is expressed at high levels both in the brain and in the

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

A

B 24

24

Kv4.1+KCNE3

16

I (µA)

I (µA)

Kv4.1 16

8

8

0

0

0

time (s)

2

C

0

2

time (s)

D 12

12

Kv4.2 I (µA)

I (µA)

965

8

8

4

4

0

0 0

time (s)

Kv4.2+KCNE3

0

2

time (s)

2

Fig. 5. KCNE3 does not affect Kv4.1 or Kv4.2 currents. Representative current traces recorded from Xenopus laevis oocytes expressing Kv4.1 (A), oocytes co-expressing Kv4.1 + KCNE3 in a molar 1:1 ratio (B), oocytes expressing Kv4.2 (C), and oocytes co-expressing Kv4.2 + KCNE3 in a molar 1:1 ratio (D). Currents were elicited from a voltage-clamp protocol identical to the one shown in Fig. 1.

Table 2 Functional effects of KCNE3 on Kv4.1 and Kv4.2 currents expressed in Xenopus laevis oocytes Parameter

Kv4.1

Kv4.1 KCNE3

Kv4.2

Kv4.2 KCNE3

Mean current at 60 mV (lA) Recovery from inactivation (srec in ms) Steady-state activation (V1/2 in mV) Steady-state inactivation (V1/2 in mV) Inactivation time-constant (sinac in ms) Time-to-peak at 60 mV (speak in ms) Number of cells (n)

12 ± 1 274 ± 40 6±3 61 ± 2 154 ± 7 11 ± 1 19

12 ± 1 273 ± 29 3±1 61 ± 2 161 ± 7 11 ± 1 10

9±1 444 ± 12 8±1 59 ± 3 107 ± 8 8±1 14

6 ± 1  502 ± 30 10 ± 1 56 ± 3 130 ± 11 10 ± 1 12

 

Significantly different from Kv4.2 (P < 0.05).

heart, a subunit with inhibitory properties such as KCNE3 would potentially have a significant impact on membrane potential in several cell types. The Kv4.3 channel has been identified as the primary channel underlying the human cardiac I to current [27,28], and the b-subunit KChIP2 has been shown to be an important accessory subunit in the generation of this current [29]. Co-expression of the Kv4.3 channel and KChIP2 subunit in a heterologous system generates currents alike I to but does not replicate the native cardiac currents with complete fidelity. It is speculated that additional accessory subunits may play a role in forming the native I to [30]. For instance, in the left ventricular wall I to has been measured to be 4fold smaller and to recover from inactivation slower in subendocardial myocytes than in epicardial myocytes [31].

This difference cannot be explained by Kv4.3 and KChIP2 alone. In this study we have shown that KCNE3 has an inhibitory effect on Kv4.3 + KChIP2, and that KCNE3 slows the recovery from inactivation of the complex. Based on these results we propose that KCNE3 may play a role in the generation of I to . If this is the case our data indicate that KCNE3 would play a more prominent role in endocardial tissue than in epicardial tissue of the left ventricular wall. Previous studies have reported a diverse set of interactions between KCNE3 and different Kv channels. KCNE3 has been reported to form a constitutively open channel in assembly with KCNQ1 (Kv7.1), and to increase the current density of Kv3.4. Further, KCNE3 is found to inhibit KCNQ4 (Kv7.4), hERG (Kv11), Kv2.1, and Kv3.1

966

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967

Transient outward current at 60 mV (µA)

40

30

20

10

4 Kv4.3 + 4 KChIP2

4 Kv4.3 + 4 KChIP2 + 2 KCNE3

4 Kv4.3 + 4 KChIP2 + 1 KCNE3

4 Kv4.3 + 4 KChIP2 + 4 KCNE3

60 12

0

0 0

time (s) 2

12 I (µA)

I (µA)

I (µA)

I (µA)

12

0 0 time (s) 2

0 0 time (s) 2

0 time (s) 2

Fig. 6. KCNE3 inhibits currents generated by Kv4.3 + KChIP2. Oocytes were co-expressed with Kv4.3 + KChIP2 and Kv4.3 + KChIP2 + KCNE3 in the molar ratios indicated, and currents were elicited by a voltage-clamp protocol similar to the one shown in Fig. 1. The mean transient outward current at +60 mV and representative current traces are shown for each injection tested.

[15,17,20]. Although, KCNE3 interacts with a number of Kv channels, there is no straight forward correlation between the channel type and the way KCNE3 affects the kinetic properties of the channel it interacts with. Similar mutations within the KCNE1-3 are found to result in equivalent modulations of KCNQ1, hERG, and Kv3.4, indicating that KCNE1-3 interact by a similar mechanism with different a-subunits [32]. Since the transmembrane segment of KCNE1 is found to interact with the pore of KCNQ1 [33], it is speculated that it is the transmembrane segment of KCNE3 that interacts with the Kv channels. This correlates with the finding that residues 71–73 in the transmembrane segment of KCNE3 are crucial for the interaction of KCNE3 with KCNQ1 [34]. The results presented in this paper demonstrate the ability of KCNE3 to severely inhibit the function of the Kv4.3 channel. The presence of KCNE3 drastically reduces the Kv4.3 current amplitude, and it doubles the time-constant for recovery from inactivation and the inactivation timeconstant. This effect of KCNE3 could potentially cause an increased cellular excitability. However, given that we do not know if the proteins interact in vivo the exact physiological significance of the Kv4.3 and KCNE3 interaction has to be elaborated in future studies. Acknowledgments Dr. Nicole Schmitt is thanked for valuable help on molecular biology and discussions throughout the work. The work was supported by The Danish National Research

Foundation and the Danish Medical Research Council (to SPO). References [1] F.M. Ashcroft, Ion Channels and Disease, Academic Press, San Diego, CA, USA, 2000. [2] B. Hille, Ion Channels of Excitable Membranes, Sinauer Associates, Inc., Sunderland, MA, USA, 2001. [3] T.J. Baldwin, M.L. Tsaur, G.A. Lopez, Y.N. Jan, L.Y. Jan, Characterization of a mammalian cDNA for an inactivating voltage-sensitive K+ channel, Neuron 7 (1991) 471–483. [4] M.D. Pak, K. Baker, M. Covarrubias, A. Butler, A. Ratcliffe, L. Salkoff, mShal, a subfamily of A-type K+ channel cloned from mammalian brain, Proc. Natl. Acad. Sci. USA 88 (1991) 4386–4390. [5] P. Serodio, B. Rudy, Differential expression of Kv4 K+ channel subunits mediating subthreshold transient K+ (A-type) currents in rat brain, J. Neurophysiol. 79 (1998) 1081–1091. [6] P. Serodio, E. Vega-Saenz de Miera, B. Rudy, Cloning of a novel component of A-type K+ channels operating at subthreshold potentials with unique expression in heart and brain, J. Neurophys. 75 (1996) 2174–2179. [7] J.E. Dixon, W. Shi, H.S. Wang, C. McDonald, H. Yu, R.S. Wymore, I.S. Cohen, D. McKinnon, Role of the Kv4.3 K+ channel in ventricular muscle. A molecular correlate for the transient outward current, Circ. Res. 79 (1996) 659–668. [8] E.K. Yang, M.R. Alvira, E.S. Levitan, K. Takimoto, Kvbeta subunits increase expression of Kv4.3 channels by interacting with their C termini, J. Biol. Chem. 276 (2001) 4839–4844. [9] W.F. An, M.R. Bowlby, M. Betty, J. Cao, H.P. Ling, G. Mendoza, J.W. Hinson, K.I. Mattsson, B.W. Strassle, J.S. Trimmer, K.J. Rhodes, Modulation of A-type potassium channels by a family of calcium sensors, Nature 403 (2000) 553–556. [10] I. Deschenes, G.F. Tomaselli, Modulation of Kv4.3 current by accessory subunits, FEBS Lett. 528 (2002) 183–188.

A. Lundby, S.-P. Olesen / Biochemical and Biophysical Research Communications 346 (2006) 958–967 [11] M.S. Nadal, A. Ozaita, Y. Amarillo, E. Vega-Saenz de Miera, Y. Ma, W. Mo, E.M. Goldberg, Y. Misumi, Y. Ikehara, T.A. Neubert, B. Rudy, The CD26-related dipeptidyl aminopeptidase-like protein DPPX is a critical component of neuronal A-type K+ channel, Neuron 37 (2003) 449–461. [12] X. Ren, Y. Hayashi, N. Yoshimura, K. Takimoto, Transmembrane interaction mediates complex formation between peptidase homologues and Kv4 channels, Mol. Cell Neurosci. 29 (2005) 320–332. [13] M.C. Sanguinetti, M.E. Curran, A. Zou, J. Shen, P.S. Spector, D.L. Atkinson, M.T. Keating, Coassembly of K(V)LQT1 and minK (IsK) proteins to form cardiac I(Ks) potassium channel, Nature 384 (1996) 24–25. [14] G.W. Abbott, F. Sesti, I. Splawski, M.E. Buck, M.H. Lehmann, K.W. Timothy, M.T. Keating, S.A. Goldstein, MiRP1 forms IKr potassium channels with HERG and is associated with cardiac arrhythmia, Cell 97 (1999) 175–187. [15] G.W. Abbott, M.H. Butler, S. Bendahhou, M.C. Dalakas, L.J. Ptacek, S.A. Goldstein, MiRP2 forms potassium channels in skeletal muscle with Kv3.4 and is associated with periodic paralysis, Cell 104 (2001) 217–231. [16] N. Tinel, S. Diochot, M. Borsotto, M. Lazdunski, J. Barhanin, KCNE2 confers background current characteristics to the cardiac KCNQ1 potassium channel, EMBO J. 19 (2000) 6326–6330. [17] B.C. Schroeder, S. Waldegger, S. Fehr, M. Bleich, R. Warth, R. Greger, T.J. Jentsch, A constitutively open potassium channel formed by KCNQ1 and KCNE3, Nature 403 (2000) 196–199. [18] M. Grunnet, T. Jespersen, H.B. Rasmussen, T. Ljungstrom, N.K. Jorgensen, S.P. Olesen, D.A. Klaerke, KCNE4 is an inhibitory subunit to the KCNQ1 channel, J. Physiol. 542 (2002) 119–130. [19] K. Angelo, T. Jespersen, M. Grunnet, M.S. Nielsen, D.A. Klaerke, S.P. Olesen, KCNE5 induces time- and voltage-dependent modulation of the KCNQ1 current, Biophys. J. 83 (2002) 1997–2003. [20] Z.A. McCrossan, A. Lewis, G. Panaghie, P.N. Jordan, D.J. Christini, D.J. Lerner, G.W. Abbott, MinK-related peptide 2 modulates Kv2.1 and Kv3.1 potassium channels in mammalian brain, J. Neurosci. 23 (2003) 8077–8091. [21] M. Zhang, M. Jiang, G.N. Tseng, MinK-related peptide 1 associates with Kv4.2 and modulates its gating function: potential role as beta subunit of cardiac transient outward channel? Circ. Res. 88 (2001) 1012–1019. [22] N. Decher, F. Bundis, R. Vajna, K. Steinmeyer, KCNE2 modulates current amplitudes and activation kinetics of HCN4: influence of KCNE family members on HCN4 currents, Pflugers Arch. 446 (2003) 633–640. [23] M. Grunnet, H.B. Rasmussen, A. Hay-Schmidt, M. Rosenstierne, D.A. Klaerke, S.P. Olesen, T. Jespersen, KCNE4 is an inhibitory

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

967

subunit to Kv1.1 and Kv1.3 potassium channels, Biophys. J. 85 (2003) 1525–1537. B. Ordog, E. Brutyo, L.G. Puskas, J.G. Papp, A. Varro, J. Szabad, Z. Boldogkoi, Gene expression profiling of human cardiac potassium and sodium channels, Int. J. Cardiol. (2005), Epub ahead of print. W. Guo, H. Li, F. Aimond, D.C. Johns, K.J. Rhodes, J.S. Trimmer, J.M. Nerbonne, Role of heteromultimers in the generation of myocardial transient outward K+ currents, Circ. Res. 90 (2002) 586–593. W. Kong, S. Po, T. Yamagishi, M.D. Ashen, G. Stetten, G.F. Tomaselli, Isolation and characterization of the human gene encoding Ito: further diversity by alternative mRNA splicing, Am. J. Physiol. 275 (1998) 1963–1970. L.A. Kim, J. Furst, M.H. Butler, S. Xu, N. Grigorieff, S.A. Goldstein, Ito channels are octomeric complexes with four subunits of each Kv4.2 and K+ channel-interacting protein 2, J. Biol. Chem. 279 (2004) 5549–5554. S. Kaab, J. Dixon, J. Duc, D. Ashen, M. Nabauer, D.J. Beuckelmann, G. Steinbeck, D. McKinnon, G.F. Tomaselli, Molecular basis of transient outward potassium current downregulation in human heart failure: a decrease in Kv4.3 mRNA correlates with a reduction in current density, Circulation 98 (1998) 1383–1393. H.C. Kuo, C.F. Cheng, R.B. Clark, J.J. Lin, J.L. Lin, M. Hoshijima, V.T. Nguyen-Tran, Y. Gu, Y. Ikeda, P.H. Chu, J. Ross, W.R. Giles, K.R. Chien, A defect in the Kv channelinteracting protein 2 (KChIP2) gene leads to a complete loss of I(to) and confers susceptibility to ventricular tachycardia, Cell 107 (2001) 801–813. I. Deschenes, D. DiSilvestre, G.J. Juang, R.C. Wu, W.F. An, G.F. Tomaselli, Regulation of Kv4.3 current by KChIP2 splice variants: a component of native cardiac I(to)? Circulation 106 (2002) 423–429. M. Nabauer, D.J. Beuckelmann, P. Uberfuhr, G. Steinbeck, Regional differences in current density and rate-dependent properties of the transient outward current in subepicardial and subendocardial myocytes of human left ventricle, Circulation 93 (1996) 168–177. G.W. Abbott, S.A. Goldstein, Disease-associated mutations in KCNE potassium channel subunits (MiRPs) reveal promiscuous disruption of multiple currents and conservation of mechanism, FASEB J. 16 (2002) 390–400. Y.F. Melman, S.Y. Um, A. Krumerman, A. Kagan, T.V. McDonald, KCNE1 binds to the KCNQ1 pore to regulate potassium channel activity, Neuron 42 (2004) 927–937. Y.F. Melman, A. Domenech, S. de la Luna, T.V. McDonald, Structural determinants of KvLQT1 control by the KCNE family of proteins, J. Biol. Chem. 276 (2001) 6439–6444.