Transforming growth factor-α changes firing properties of developing neocortical GABAergic neurons by down-regulation of voltage-gated potassium currents

Neuroscience 122 (2003) 637– 646

TRANSFORMING GROWTH FACTOR-␣ CHANGES FIRING PROPERTIES OF DEVELOPING NEOCORTICAL GABAERGIC NEURONS BY DOWN-REGULATION OF VOLTAGE-GATED POTASSIUM CURRENTS H. NAMBA,* N. TAKEI AND H. NAWA

Pollock et al., 1990; Sherwood et al., 1997; Lesser and Lo, 1995; Lesser et al., 1997). Regulation of the voltagegated currents promotes the development of firing phenotypes, such as repetitive firing to differentiate the function of a given neuron. The epidermal growth factor family has neurotrophic actions on postmitotic neurons in the CNS (for reviews, see: Junier, 2000; Yamada et al., 1997; Xian and Zhou, 1999). In particular, transforming growth factor ␣ (TGF␣) is widely expressed in the forebrain during development and is the main endogenous ligand of the epidermal growth factor receptor, ErbB1 (Seroogy et al., 1993; Kornblum et al., 1997). ErbB1 is expressed in GABAergic neurons in the developing and adult rat neocortex (Go´mez-Pinilla et al., 1988; Kornblum et al., 1995), which suggests direct actions of ErbB1 ligands on GABAergic neurons. The effects of ErbB1 stimulation on the development of firing properties and expression of voltage-gated currents of the GABAergic neurons are largely unknown. The repetitive firing of action potentials is determined by the combination of voltage-gated sodium and potassium currents (Llina´s, 1988; Baxter and Byrne, 1991; Hille, 2001; Turrigiano et al., 1995; Rothe et al., 1999). Cortical GABAergic neurons have three types of voltage-gated potassium current components, including the transient (IA) current, slow delayed rectifier potassium current (IK), and fast delayed rectifier potassium current (I4AP), which is blocked by low concentrations of 4-aminopyridine (4-AP) (Massengill et al., 1997). These currents are mediated by Kv4, Kv2 and Kv3 channels, respectively (Song, 2002). All these currents affect the repetitive firing pattern (Malin and Nerbonne, 2000; Malin and Nerbonne, 2002; McBain and Fisahn, 2001; Rudy and McBain, 2001) and waveform of an action potential, such as duration and fast after-hyperpolarization (Storm, 1990; Du et al., 2000; McBain and Fisahn, 2001; Rudy and McBain, 2001). Here, neocortical primary cultures were chronically treated with TGF␣ and alterations in firing properties and voltage-gated currents of morphologically identified GABAergic neurons were determined using whole-cell patch-clamp techniques. Furthermore, to examine whether the TGF␣-induced electrophysiological alterations are related to the mRNA level, Kv channel mRNAs from total culture and the bipolar neurons were

Department of Molecular Neurobiology, Brain Research Institute, Niigata University, 1-757 Asahimachi, Niigata 951-8585, Japan

Abstract—Transforming growth factor-␣ (TGF␣), a member of the epidermal growth factor family, has neurotrophic actions on postmitotic neurons. We examined the chronic effects of TGF␣ on the electrophysiological properties of one type of GABAergic neuron, identified by its bipolar morphology, in neocortical primary culture. Approximately 85% of the bipolar neurons were GABA-immunoreactive. In response to depolarizing current injection, the bipolar neurons usually showed tonic firing of action potential under control conditions. After treatment with TGF␣ (20 ng/ml) for 2 days, these neurons failed to generate trains of action potentials. Furthermore, the treatment altered the action potential waveforms of the bipolar neurons, including the duration and amplitude of the fast after-hyperpolarization, which implies a reduction in voltage-gated potassium currents. In contrast, TGF␣ did not affect the firing properties of pyramidal-shaped non-GABAergic neurons. Voltageclamp recordings from the bipolar neurons indicated that chronic treatment with TGF␣ markedly decreased the current densities of slow delayed rectifier (IK) and transient voltage-gated potassium currents, whereas the treatment had no effect on voltage-gated sodium current and fast delayed rectifier potassium current densities. Reverse transcription-polymerase chain reaction analysis of potassium channel mRNA in the bipolar neurons revealed that the reduction in the IK current density was caused by Kv2.2 mRNA down-regulation. Thus, chronic treatment with TGF␣ down-regulated slow delayed rectifier and transient voltage-gated potassium currents, and in parallel, suppressed repetitive generation of action potentials in the cortical GABAergic neurons. © 2003 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: TGF␣, repetitive firing, IK, IA, Kv2.2, ErbB1.

Many neurotrophins and growth factors, which promote neuronal survival and differentiation, regulate voltagegated currents in neurons as well as in neural cell lines (review: Lo, 1998; neurons: Adamson et al., 2002; Gonzalez and Collins, 1997; McFarlane and Cooper, 1993; Rothe et al., 1999; cell lines: Dichter et al., 1977; *Corresponding author. Tel: ⫹81-25-227-0614; fax: ⫹81-25-2270814. E-mail address: [email protected] (H. Namba). Abbreviations: ANOVA, analysis of variance; RT-PCR, reverse transcription-polymerase chain reaction; TEA-CL, tetraethylammonium chloride; TGF␣, transforming growth factor ␣; TTX, tetrodotoxin; 4-AP, 4-aminopyridine.

0306-4522/03$30.00⫹0.00 © 2003 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2003.08.013

637

638

H. Namba et al. / Neuroscience 122 (2003) 637– 646

quantified by reverse transcription-polymerase chain reaction (RT-PCR).

EXPERIMENTAL PROCEDURES All procedures employing experimental animals were performed in compliance with our institutional guidelines for the care and use of laboratory animals. Efforts were made to minimize the number of animals used and their suffering.

Neocortical culture Whole cerebral neocortices of embryonic day 19 rats were mechanically dissociated and plated onto poly-D-lysine-coated dishes at a density of 1000 –1500 cells/mm2. Cortical neurons were grown with Dulbecco’s Modified Eagle medium containing 1 mM glutamine (Ajinomoto, Tokyo, Japan), 2% fetal bovine serum, nutrient mixture N2, and 10 mM HEPES (pH 7.3; Narisawa-Saito et al., 1999). The following day, the original medium was replaced with serum-free medium (serum-free N2 medium). Human recombinant TGF␣ (20 ng/ml; Sigma Chemical Co., St. Louis, MO, USA) was applied daily to serum-free N2 medium for 2 day from 6 or 7 day in vitro.

Immunocytochemistry To confirm that morphologically identified bipolar neurons were GABA-immunoreactive, as reported by Turrigiano and colleagues (Desai et al., 1999b; Rutherford et al., 1997, 1998), bipolarshaped neurons were identified by their morphology and photographed in culture fixed with 4% paraformaldehyde, 0.5% glutaraldehyde and 2% sucrose in 0.15 M phosphate buffer. After washing with Tris-buffered saline, cells were treated with 0.3% Triton X-100 in NSB (containing with 0.15 M NaCl, 25 mM Tris– HCl (pH 7.4), 3% normal goat serum and 0.3% bovine serum albumin) for 30 min and incubated with rabbit anti-GABA antiserum (1:2000 in NSB; Immunotech, Prague, Czech Republic) overnight at 4 °C. After washing with Tris-buffered saline, cells were incubated with biotinylated goat anti-rabbit IgG (1:200 in NSB; Vector Laboratories, CA, USA) for 60 min. The avidin-biotin system (Vector Laboratories) was used with 3,3⬘-diaminobenzidine tetrahydrochloride (0.05%) and H2O2 (0.003%) in 0.1 M sodium acetate buffer (pH 7.0) containing with 1% NiCl2 to visualize the immunoreactive cells.

Electrophysiology Electrophysiological recordings were performed at 8 or 9 day in vitro using whole-cell patch-clamp techniques. The growth medium was replaced by recording medium containing 150 mM NaCl, 4 mM KCl, 2 mM CaCl2, 10 mM HEPES, 10 mM glucose, 1 mM MgSO4, and 10 mM sucrose (320 mOsm; pH⫽7.4). Wholecell patch-clamp recording was performed using borosilicate glass capillaries (4 –5 M⍀). The signals were amplified with an Axopatch 200B amplifier (Axon Instruments, Union City, CA, USA). Data acquisition was performed using either Clampex 6 or 7 via a digital converter, DigiData 1200 (Axon Instruments). The acquired data were analyzed using Clampfit 6 (Axon Instruments). For current-clamp recordings and voltage-clamp recordings of voltage-gated potassium currents, the patch pipette was filled with internal solution containing 140 mM KMeSO3, 10 mM KCl, 10 mM HEPES, 0.2 mM EGTA, and 3 mM Mg-ATP, with the osmolarity was adjusted to 310 mOsm with sucrose (pH⫽7.4). To analyze neuronal firing properties, membrane potentials were held at ⫺60 or ⫺70 mV, then constant depolarizing current pulses of 400 ms duration were injected. Isolation of voltage-gated potassium currents was performed according to Massengill et al. (1997) (see Fig. 5A). Briefly, total outward potassium currents were evoked in a single cell by a

series of depolarizing voltage steps in a normal external solution containing 0.2 ␮M tetrodotoxin (TTX; Wako Chemicals, Osaka, Japan) to block voltage-gated sodium currents, and 2 mM Cd2⫹ to block calcium and calcium-activated potassium currents. Because of the external Cd2⫹, rather large depolarizing voltage steps were necessary to inactivate the IA currents (Song et al., 1998; Andreasen and Hablitz, 1992). IA current was obtained by digital subtraction of traces in which the IA current was inactivated (⫺20 mV prepulse) from the total outward potassium currents. IK current was defined as the current activated after a prepulse to ⫺20 mV in the presence of 100 ␮M 4-AP (Sigma Chemical Co.). Fast delayed rectifier potassium currents (I4AP), which are sensitive to lower concentrations of 4-AP (Martina et al., 1998; Massengill et al., 1997), were obtained by digital subtraction of current traces of IK from outward currents with a ⫺20 mV prepulse. To measure sodium currents, capacitive transients and leak conductance were subtracted using a P/4 protocol (Benzanilla and Armstrong, 1977). The external solution for whole-cell sodium current experiments contained 33 mM NaCl, 107 mM choline chloride, 4 mM KCl, 1.6 mM CaCl2, 0.4 mM CdCl2, 10 mM HEPES, 10 mM tetraethylammonium chloride (TEA-Cl), 5 mM 4-AP, 10 mM glucose, 1 mM MgSO4, and 5 mM sucrose (320 mOsm; pH⫽7.4). The internal solution for the sodium currents contained 5 mM NaCl, 140 mM CsMeSO3, 10 mM KCl, 10 mM HEPES, 0.5 mM EGTA, 3 mM Mg-ATP, 5 mM TEA-Cl, and 10 mM sucrose (310 mOsm; pH⫽7.4). The decrease in [Na⫹]o decreased the amplitude of the sodium currents, enabling us to clamp the voltage of recorded neurons (Desai et al., 1999a; Huguenard et al., 1988). To record these voltage-gated currents, series resistance was compensated to 80%. Currents were analyzed only if leak currents remained unchanged and series resistance changed by less than 20%. Each current was measured at peak amplitude and divided by cell membrane capacitance to obtain current density.

Quantitative RT-PCR Total cellular RNA from neocortical culture was extracted with 1 ml Isogen RNA extraction reagent according to the manufacturer’s instructions (Nippon Gene, Tokyo, Japan). cDNA was prepared from 5 ␮g extracted RNA with the SuperScript™ First-Strand Synthesis System containing random hexamers (Invitrogen, Tokyo, Japan). The cDNA was subjected to PCR with TaKaRa Ex Taq™ (TaKaRa, Otsu, Japan) to detect the expression of mRNAs. All the PCR primers were made to order by Invitrogen according to our sequence design. Sequences and locations of the following primers refer to sequences and locations published in GenBank (National Center of Biotechnology Information, available at http://www.ncbi.nlm.nih.gov). Primers for Kv2.1 were described previously and gave a 229 bp PCR product (Baranauskas et al., 1999). Kv2.2 mRNA (GenBank accession M77482) was detected with a pair of primers of 5⬘-AGAACCGGAGTGAGGGATGT (position 8) and 5⬘-GAACTGGATGACAGAGCCACAGA (position 304), which gave PCR product of 316 bp. Primers for Kv4.2 and Kv 4.3 were described previously (Song et al., 1998). In Kv4.3, two PCR products of 217 bp and 274 bp were obtained because of a 19-amino acid insertion (Ohya et al., 1997). The primers for ␤-actin mRNA were described previously (Ming et al., 1999). The cycle conditions for all PCR reactions were: 94 °C for 5 min, followed by 20 –26 cycles (DNA denaturation at 94 °C for 30 s, primer annealing at 54 –59 °C for 30 s, and DNA extension at 72 °C for 40 s), and 72 °C for 10 min. Annealing temperatures were set to 55, 54, 55, 59, and 56 °C for Kv2.1, Kv2.2, Kv4.2, Kv4.3 and ␤-actin primers, respectively. To quantify mRNA levels, the appropriate number of PCR cycles that produced linear amplifications was determined (Xiong et al., 1999). cDNA amplification from Kv2.1, Kv2.2, Kv4.2, Kv4.3, and ␤-actin mRNA lost linearity after 30,

H. Namba et al. / Neuroscience 122 (2003) 637– 646

A

C

639

B

D

control

14 control

12

TGFα

10 8

TGFα

20mV 100ms

6

∗

4 2 0

100pA

0

40 60 80 100 120

20

injected current (pA) Fig. 1. TGF␣ reduced repetitive generation of action potentials in GABA-immunoreactive bipolar neurons. (A) Hoffman-Modulation Contrast image of a cultured bipolar-shaped neuron (arrow) before immunostaining. (B) The same photographic field after immunostaining shows that the bipolarshaped neuron has GABA immunoreactivity (arrow). Note pyramidal-shaped neuron (arrowhead) and another morphologically unidentified neuron (double arrowhead) have no GABA-immunoreactivity. (C) Spike firing in response to constant depolarizing current of 100 pA. Membrane potentials of the cells were held at ⫺70 mV. In control cultures, the bipolar neurons had repetitive spike firing at a frequency of approximately 60 Hz, in contrast to TGF␣-treated cultures, where bipolar neurons generated only a few spikes. (D) Number of spikes was significantly reduced by TGF␣ treatment (control, n⫽8; TGF␣, n⫽13). Numbers indicate spikes that were generated during the 200 ms after the onset of current injection. Scale bar⫽15 ␮m.

26, 28, 28, and 20 cycles, respectively. Therefore, the PCR cycles were set to 18 for ␤-actin, 24 for Kv2.2, and 26 for Kv2.1, Kv4.2 and Kv4.3. For quantification, the PCR product was analyzed by 2% agarose gel electrophoresis and stained with SYBR-Green I (Molecular Probe, Eugene, OR, USA). The fluorescence strength for PCR products was quantified using a fluoroimage analyzer Fujifilm FLA-2000 (Fujifilm, Japan). To quantify the expression of Kv channels in bipolar neurons, mRNA was collected from 10 cells using patch pipettes (Baranauskas et al., 1999). Cultures were washed with normal recording medium. Neurons were patched in the cell-attached mode and aspirated into a pipette containing approximately 5 ␮l of sterile water. After aspiration of the neuronal contents, the pipette was removed from the holder and broken in a 1.5 ml siliconized Eppendorf tube containing ice-cooled 100 ␮l Isogen. Yeast transfer RNA (5 ␮g; Sigma Chemical Co.) was added as a carrier for ethanol precipitation and incubated for 30 min at 4 °C. After extraction of total RNA, cDNA was prepared from 3 ␮g extracted RNA with SuperScript™ containing random hexamers. As amplification from Kv2.2 and ␤-actin mRNA lost linearity after 35 and 31 cycles respectively, the PCR cycles were set to 33 for Kv2.2 and 28 for ␤-actin.

Statistics All the data are presented as mean⫾S.E.M. Statistical significance was evaluated with either Student’s t-test or two-way repeated measures analysis of variance (ANOVA).

RESULTS Effects of TGF␣ on firing properties of GABAergic neurons and pyramidal neurons To evaluate the effect of TGF␣ on GABAergic neurons, we identified one type of GABAergic neuron, the bipolarshaped neuron, based on the morphology reported by Turrigiano and colleagues (Desai et al., 1999b; Rutherford et al., 1997, 1998). These neurons were readily identified by their elongated cell bodies and two or three emerging dendrites (Sekirnjak et al., 1997). In the 20 cells identified with bipolar morphology (Fig. 1A), 17 (85%) were immunoreactive to the anti-GABA antibody (Fig. 1B). Among those, four cells showed strong, six

640

H. Namba et al. / Neuroscience 122 (2003) 637– 646

Table 1. TGF␣ did not change the passive properties of GABAergic bipolar neurons and pyramidal neurons Bipolar neuron

Vm (mV) Rs (M⍀) Rm (M⍀) Cm (pF)

Pyramidal neuron

Control

TGF␣

Control

TGF␣

⫺36.7⫾2.7 (7) 24.4⫾1.1 (14) 1459⫾134 (14) 19.5⫾1.5 (14)

⫺33.1⫾2.0 (13) 27.1⫾1.8 (11) 1351⫾250 (11) 19.5⫾2.7 (11)

⫺46.7⫾0.9 (7) 21.5⫾1.3 (7) 1381⫾191 (7) 22.2⫾1.2 (7)

⫺43.2⫾1.7 (8) 22.7⫾3.0 (8) 1383⫾136 (8) 25.7⫾2.3 (8)

In bipolar neurons, resting potential, Vm, was measured in current clamp mode on the cells used to determine firing properties. Series resistance, Rs, input resistance, Rm, and whole-cell capacitance, Cm, were measured in voltage-clamp mode on cells used to measure sodium and potassium currents. In pyramidal neurons, all parameters were measured from the same cells used to determine firing properties. Numbers of cells are indicated in parentheses. No differences were significant.

The action potential waveform and the latency of the first action potential of the bipolar neurons were compared between control and TGF␣-treated cells (Fig. 2). Control neurons generated shorter duration action potentials of approximately 2 ms followed by a larger amplitude fast after-hyperpolarization. In contrast, in TGF␣-treated cells, the duration increased significantly (P⫽0.037, t-test; Fig. 2B) and the amplitude of the fast after-hyperpolarization diminished significantly (P⫽0.012, t-test; Fig. 2C). In addition, the latency to produce the first action potential during depolarizing current steps was decreased in TGF␣-treated cells to some extent, but the decrease was not statistically significant (P⫽0.13, t-test; Fig. 2D, E). These changes in

showed medium, and seven showed weak GABA immunoreactivity. Under control conditions, these neurons tonically generated action potentials at approximately 60 Hz with constant intervals and exhibited a prominent fast after-hyperpolarization between spikes in the train (Fig. 1C, upper trace). In TGF␣-treated cells, however, the numbers of spikes induced by current injections was significantly reduced (Fig. 1C, D; ANOVA P⫽0.0045). Measurement using whole-cell current- and voltage-clamp modes indicated that there were no significant differences in the passive properties between control and TGF␣-treated cells (Table 1).

A

C

B 5

∗

4 TGFα

control 25mV

0 -10

3

-20

2

-30

1

-40 -50

0

10ms

control

D

E

control

∗

TGFα

control

TGFα

0.2 0.15 0.1

TGFα

0.05 25mV

0 control

TGFα

100ms Fig. 2. Effects of TGF␣ on the firing properties of bipolar neurons. (A) Typical examples of action potentials in bipolar neurons induced by depolarizing current injection. (B, C) TGF␣ significantly changed the duration of action potentials (AP) (B) and amplitude in fast after-hyperpolarization (AHP) (C) (control; n⫽7; TGF␣; n⫽13). (D) Typical traces of the first action potentials of control and TGF␣-treated neurons injected by 20 pA and 40 pA depolarizing currents, respectively. Arrow indicates the onset of depolarizing current step. (E) The latencies of the generation of the first action potentials tended to increase in TGF␣-treated neurons (control; n⫽7; TGF␣; n⫽13).

H. Namba et al. / Neuroscience 122 (2003) 637– 646

A

641

B

control

15 control TGFα

10

TGFα

20mV 100ms

5

0 0 10 20 30 40 50 60 70

40pA

injected current (pA) Fig. 3. TGF␣ had no effect on spike firing of pyramidal neurons. (A) Spike firing in response to constant depolarizing current of 40 pA. Membrane potentials of the cells were held at ⫺60 mV. These neurons had repetitive spike firing at a frequency of approximately 25 Hz. (B) Number of spikes was not changed by TGF␣ treatment (control, n⫽7; TGF␣, n⫽8). Numbers indicate spikes that were generated during the 400 ms current steps.

gated sodium currents between control and TGF␣-treated bipolar neurons. To insure that the membrane potential was clamped during activation of sodium currents, the external solution contained a lower concentration of sodium (see Experimental Procedures). In response to a voltage-step to ⫺30 mV from a holding potential of ⫺70 mV, voltage-gated sodium currents were activated, and a voltage step to 0 mV induced maximum amplitudes (Fig. 4A). There was no significant difference in peak current densities between control and TGF␣-treated cells (P⫽0.757, t-test; Fig. 4B).

firing properties suggest a decrease in voltage-gated potassium currents in TGF␣-treated cells. To address the issue of cell-type specificity in the effects of TGF␣, we analyzed firing properties of pyramidal-shaped non-GABAergic neurons (see Fig. 1A, arrowhead) (Rutherford et al., 1998). These neurons generated action potentials at approximately 20 Hz of maximum frequency with constant or increasing inter-spike intervals (Fig. 3A). We did not detect any significant differences in the passive properties (Table 1) or numbers of action potentials evoked by depolarizing current injections between control and TGF␣-treated neurons (Fig. 3A, B; ANOVA P⫽0.91). In this type of neuron, TGF␣ had no effect on the action potential duration or the amplitude of the fast after-hyperpolarization (data not shown).

Reduction of voltage-gated potassium currents by TGF␣ in the bipolar neurons To examine the types of potassium currents that were affected by alterations in the firing properties, the amplitudes of these currents were compared between TGF␣treated and non-treated bipolar cells. Certain cortical GABAergic neurons have three types of voltage-gated potassium currents, including slow and fast delayed rectifier K⫹ currents components (IK and I4AP, respectively) and

Effects of TGF␣ on voltage-gated sodium currents in the bipolar neurons Because both inward and outward currents affect neuronal firing patterns, we first compared the amplitude of voltage-

voltage (mV)

100 pA 5 msec

current density (pA/pF)

B

A

-40

-20

0

20

40

0 -5 -10 -15 -20 -25

control

TGFα

Fig. 4. Effects of TGF␣ on sodium current in bipolar neurons. (A) Voltage-gated sodium currents were evoked by a series of 30 ms voltage steps to ⫺30, ⫺20, ⫺10, 0, ⫹10, ⫹20, and ⫹30 mV from a holding potential of ⫺70 mV. (B) Maximum current densities induced by a 0 mV test pulse were not significantly different between control and TGF␣-treated neurons (control, n⫽8; TGF␣, n⫽7).

642

H. Namba et al. / Neuroscience 122 (2003) 637– 646

A a

b

IA+4AP+K

I4AP+K prepulse -20mV

a-b

C

IK

50

cont

40 TGFα

30 20

∗

10 0 -40 -20

0

20

40

Q$

b-c

IA

current density (pA/pF)

B

I4AP

50

cont

40

I4AP

PVHF

D

IA cont

100

TGFα

30 20 10 0 -40 -20

voltage (mV)

IK

prepulse -20mV

current density (pA/pF)

prepulse -100mV

current density (pA/pF)

c

0

20

40

voltage (mV)

TGFα

50

∗ 0 -40 -20

0

20

40

voltage (mV)

Fig. 5. TGF␣ decreased voltage-gated potassium currents of bipolar neurons. (A) Isolation of voltage-gated potassium currents. (a) Total outward voltage-gated potassium currents evoked in external solution with 0.2 ␮M TTX and 2 mM Cd2⫹. (b) Inactivation of IA with prepulse of ⫺20 mV. (a– b) IA was obtained by subtraction of traces from the total outward currents to currents with a ⫺20 mV prepulse. (c) IK was defined as the current activated after a prepulse to ⫺20 mV in the presence of 100 ␮M 4-AP. (b– c) I4AP currents were isolated by subtraction from the current responses with a prepulse of ⫺20 mV to the responses of IK currents. All the current traces were evoked by 400 ms voltage steps to ⫺20, ⫺10, 0, ⫹10, ⫹20, and ⫹30 mV with a 250 ms prepulse from a holding potential of ⫺80 mV. (B) Treatment with TGF␣ significantly reduced the current density of IK. (C) Treatment with TGF␣ had no effect on the I4AP current density. (D) Treatment with TGF␣ significantly reduced the IA current density (control, n⫽6; TGF␣, n⫽4).

a rapidly inactivating A-type K⫹ current (IA) component (Massengill et al., 1997). In the bipolar neurons, therefore, these types of potassium currents were isolated according to Massengill et al. (1997). The IA currents (Fig. 5A, a– b) were isolated by the subtraction from total outward current responses evoked by a series of voltage steps with a prepulse of ⫺100 mV (Fig. 5A, a) to that with a prepulse of ⫺20 mV, which inactivated IA currents (Fig. 5A, b). The IK currents were evoked by a series of voltage steps with the prepulse in 100 ␮M 4-AP, which inactivated fast delayed rectifier potassium currents, I4AP (Fig. 5A, c). This current began to activate near ⫺10 mV and was characterized by slow activation kinetics (10 –90% rise time for ⫹40 mV test pulse: 17.7⫾4.4 ms) and slowly inactivating components at the more depolarized voltages. The I4AP currents (Fig. 5A, b, c) were isolated by the subtraction from the current

responses of a series of voltage steps with the prepulse (Fig. 5A, b) to the responses of IK currents (Fig. 5A, c). These I4AP currents were characterized by relatively rapid activation kinetics (10 –90% rise time for ⫹40 mV test pulse: 5.6⫾0.6 ms). Current densities of the IK currents were significantly reduced by TGF␣ from 38.5⫾6.4 pA/pF to 19.0⫾1.1 pA/pF for the largest voltage step of ⫹40 mV (P⫽0.03, t-test; Fig. 5B). In contrast, in the I4AP currents we could not detect a significant difference in the current densities between control and TGF␣-treated bipolar neurons (P⫽0.83, t-test; Fig. 5C). The IA currents began to activate near ⫺10 mV both in control and TGF␣-treated bipolar neurons. The current densities of IA currents were significantly reduced by TGF␣ from 86.8⫾20.8 pA/pF to 27.2⫾2.8 pA/pF for the largest voltage step (P⫽0.04, t-test; Fig. 5D).

H. Namba et al. / Neuroscience 122 (2003) 637– 646

643

B

A control

TGFα

control

1

TGFα

Kv2.1 0.75

Kv2.2

∗

0.5

Kv4.2

∗

Kv4.3c 0.25 Kv4.3a actin 0

Kv2.1 Kv2.2 Kv4.2 Kv4.3a Kv4.3c

C

bipolar total cells

D

control

TGFα

Kv2.2

Kv2.1

actin

Kv2.2

0.8 actin

0.6 0.4

∗

0.2 0 control

TGFα

Fig. 6. Expression of Kv channels mRNA in total culture and bipolar neurons. (A) Typical examples of the PCR products from Kv2 and Kv4 channels, and ␤-actin in culture are shown with SYBR-Green I staining. Because of splice variants of Kv4.3 (Ohya et al., 2001), two bands were detected, which originated from Kv4.3c (upper bands) and Kv4.3a (lower bands). (B) TGF␣ significantly decreased expression of Kv2.2 and Kv4.3a and had no effect on those of Kv2.1, Kv4.2, and Kv4.3c (control: n⫽5, TGF␣: n⫽5). The values were normalized with those of PCR products amplified from ␤-actin mRNA in parallel and the ratios of these PCR products are shown. (C) PCR products from Kv2.1, Kv2.2, and ␤-actin in total culture (left lane) and 10 bipolar neurons (right lane) are shown with SYBR Green I staining. No Kv2.1 mRNA expression was detected in mRNA from bipolar neurons. Each product from 10 bipolar neurons was amplified by 35 cycles of PCR. (D) TGF␣ markedly reduced Kv2.2 channel expression in mRNA from 10 bipolar neurons (control: n⫽8, TGF␣: n⫽8). The value was normalized with those of PCR products amplified to ␤-actin mRNA in parallel and ratios of PCR products were shown. Inset: Products from Kv2.2 and ␤-actin are shown with SYBR-Green I staining, which were amplified by 33 cycles and 28 cycles of PCR, respectively.

TGF␣ decreased Kv2.2 channel expression in the bipolar neuron To test whether these reductions in potassium currents result from the down-regulation of Kv channels, we analyzed mRNA expression of Kv channels in TGF␣-treated cultures using RT-PCR. The IK current is mediated by two subtypes of Kv2 channels, Kv2.1 and Kv2.2 (Coetzee et al., 1999; Song, 2002). To quantify the differences in mRNA expression of these Kv channels, RT-PCR was performed and amounts of PCR products were normalized to those of ␤-actin mRNA. The chronic treatment with TGF␣ did not change the relative level of Kv2.1 expression (P⫽0.62, t-test), whereas the expression of Kv2.2 mRNA was significantly decreased in TGF␣-treated cultures to 58% that of controls (P⫽0.013, t-test; Fig. 6B).

To examine whether the reduced Kv2.2 channel expression in total culture reflects down-regulation in the bipolar neurons, we extracted the mRNA from 10 bipolar neurons using patch pipettes, and quantified Kv2 channel expression. In the bipolar population, we efficiently detected only Kv2.2 channel mRNA and failed to detect Kv2.1 mRNA under our PCR conditions. Kv2.2 channel mRNA expression was compared between control and TGF␣-treated bipolar neurons. To standardize the amounts of input mRNA, PCR product levels were all normalized to those of ␤-actin. As a result, Kv2.2 channel expression was significantly reduced by TGF␣ to 26% that of controls (P⫽0.002, t-test; Fig. 6C). The IA current is mediated by the Kv4 family, Kv4.2 and Kv4.3 (Song, 2002), and both subtypes are ex-

644

H. Namba et al. / Neuroscience 122 (2003) 637– 646

pressed in the neocortex (Ohya et al., 2001; Serodio and Rudy, 1998). There was no significant difference in Kv4.2 channel mRNA expression between control and TGF␣-treated cultures (P⫽0.26, t-test). Both the Kv4.3a mRNA and a longer splice variant, Kv4.3c, were expressed in neocortex (Ohya et al., 2001), and were detected as two PCR products of 217 bp and 274 bp, respectively (Fig. 6A). Kv4.3a expression was significantly decreased to 73% that of controls in TGF␣treated cultures (P⫽0.003, t-test), whereas Kv4.3c mRNA expression did not change significantly (P⫽0.34, t-test; Fig. 6B). In mRNA extracted from the bipolar neurons, neither Kv4.2 nor Kv4.3 channels were detected in most mRNA samples extracted from the bipolar neurons (data not shown). Thus, changes in firing properties in TGF␣-treated bipolar neurons, including suppression of repetitive firing and alterations in action potential waveforms result from the down-regulation of both in IK and IA currents, which at least in part, occur at Kv mRNA level.

DISCUSSION We demonstrated that TGF␣ changed the firing properties of morphologically identified bipolar neurons in neocortical culture. In culture, 85% of the morphologically identified neurons were GABA-immunoreactive. In agreement, their firing frequency is consistent with that of cultured GABAergic neurons identified previously (Rutherford et al., 1997; Desai et al., 1999b). TGF␣ down-regulated the transient and slow delayed rectifier potassium currents in this neuronal population, without changing the amplitude of voltage-gated sodium currents. Further, the down-regulation of the slow delayed rectifier currents was caused by a reduction in the mRNA expression of Kv2.2 channels in the bipolar neurons. This finding represents a novel effect of TGF␣ on neocortical GABAergic neurons, and demonstrates target specificity of action of TGF␣, as TGF␣ had no effects on the electrical properties of pyramidal neurons. Alterations in the action potential waveform, including the duration and the amplitude in fast after-hyperpolarization, suggest down-regulation of voltage-gated potassium currents in the TGF␣-treated bipolar neurons (Fig. 2A–C). In the bipolar neurons, we discriminated three types of voltage-gated potassium currents, a transient current, IA, a slow delayed rectifier current, IK and, a fast delayed rectifier current, I4AP, according to Massengill et al. (1997). These currents strongly affect action potential waveforms, including the duration and fast after-hyperpolarization (Blaine and Ribera, 2001; Erisir et al., 1999; Schwindt et al., 1988; Storm, 1990; Viana et al., 1993). In addition, the tendency of decreased latency to produce the first action potential (Fig. 2D, E) suggests a reduction in IA current (Song et al., 1998; Shibata et al., 2000; Yarom et al., 1985). Although we could not rule out the possibility that Ca2⫹-activated K⫹ currents contributed to action potential duration and after-hyperpolarization (Zhang and McBain, 1995), voltage-clamp recordings revealed marked de-

crease in IK and IA currents in the TGF␣-treated bipolar neurons. The slow delayed rectifier IK current is mediated by potassium channels in the Kv2 family, Kv2.1 and Kv2.2 (Song, 2002). In mRNA sampled from the bipolar neurons, we detected PCR products of Kv2.2 mRNA, whereas Kv2.1 mRNA was under detection limit, suggesting a higher abundance of Kv2.2 mRNA. This is consistent with previous immunohistochemical studies of the neocortex and developing hippocampus that reported Kv2.1 channels are dominantly expressed in pyramidal neurons, whereas the Kv2.2 channel distribution is more restricted to GABAergic interneurons (Hwang et al., 1993; MaleticSavatic et al., 1995). The marked decrease in Kv2.2 mRNA in bipolar neuron RNA suggests that down-regulation of IK currents in the bipolar neurons results from decreased Kv2.2 channel mRNA expression. The transient IA current is mediated by potassium channels in the Kv4 family, Kv4.2 and Kv4.3 (Song, 2002). Both subtypes are expressed in the neocortex (Ohya et al., 2001; Serodio and Rudy, 1998). Because mRNA levels for both channels were too low to detect in mRNA extracted from bipolar neurons, we did not determine whether TGF␣ treatment down-regulated IA currents by decreasing the mRNA levels. The significant decrease in Kv4.3a mRNA expression only in total culture RNA suggests that chronic treatment with TGF␣ down-regulates in the level of Kv4.3a mRNAs in unidentified neurons. In certain hippocampal GABAergic neurons, Kv4.3 channels are expressed more dominantly than Kv4.2 (Lien et al., 2002). Further single cell RT-PCR analyses are necessary to determine which types of channels are enriched in the bipolar neurons (Martina et al., 1998; Tkatch et al., 2000). In cultured bipolar neurons, the fast delayed rectifier I4AP current was detected as suggested previously in other GABAergic neurons, such as fast spiking neurons (Massengill et al., 1997; Martina et al., 1998; Fig. 5A). This type of current contributes to high frequency repetitive firing in fast spiking neurons and is mediated by Kv3 family members (Lau et al., 2000; Massengill et al., 1997; McBain and Fisahn, 2001; Rudy and McBain, 2001). The current I4AP density, however, was not influenced by treatment with TGF␣. Alternatively, suppression of repetitive spike firing in TGF␣-treated bipolar neurons would result from the marked decrease in the amplitude of IK and IA currents. Recent studies with dominant negative expression and overexpression of Kv channels demonstrated the role of slow delayed and transient potassium currents in producing repetitive firing patterns (Malin and Nerbonne, 2000, 2002). Suppression of IK currents by the dominant negative expression of Kv2 decreased the percentage of tonic firing cells and increased those in adapting firing cells, whereas the overexpression of Kv2 channel increases IK density and the percentage of tonic firing cells. Similar results were obtained in the case of Kv4 channels which mediate the IA current (Malin and Nerbonne, 2000). Though the voltage-gated sodium currents also affect repetitive spike firing (Rothe et al., 1999), TGF␣ had no effect on their amplitude in the bipolar neurons. Thus, the

H. Namba et al. / Neuroscience 122 (2003) 637– 646

reductions in both types of potassium currents, IK and IA, might lead to the suppression of repetitive firing in the TGF␣-treated bipolar neurons. Further analyses of neurochemical phenotypes are required to determine if the impairment in the repetitive spike firing might reflect a dedifferentiating action of TGF␣ on the development of the cortical GABAergic neurons. Acknowledgements—This work was supported by the Tsukada Grant for Niigata University Medical Research to H. Namba, the Ministry of Health, Labor and Welfare to N. Takei, and Grant-inAid for Creative Scientific Research to H. Nawa. We thank Miss K. Ishii for technical assistance.

REFERENCES Adamson CL, Reid MA, Davis RL (2002) Opposite actions of brainderived neurotrophic factor and neurotrophin-3 on firing features and ion channel composition of murine spiral ganglion neurons. J Neurosci 22:1385–1396. Andreasen M, Hablitz JJ (1992) Kinetic properties of a transient outward current in rat neocortical neurons. J Neurophysiol 68:1133– 1142. Baranauskas G, Tkatch T, Surmeier DJ (1999) Delayed rectifier currents in rat globus pallidus neurons are attributable to Kv2.1 and Kv3.1/3.2 K⫹ channels. J Neurosci 19:6394 –6404. Baxter DA, Byrne JH (1991) Ionic conductance mechanisms contributing to the electrophysiological properties of neurons. Curr Opin Neurobiol 1:105–112. Benzanilla F, Armstrong CM (1977) Inactivation of the sodium channel: I. Sodium current experiments. J Gen Physiol 70:549 – 566. Blaine JT, Ribera AB (2001) Kv2 channels form delayed-rectifier potassium channels in situ. J Neurosci 21:1473–1480. Coetzee WA, Amarillo Y, Chiu J, Chow A, Lau D, McCormack T, Moreno H, Nadal MS, Ozaita A, Pountney D, Saganich M, VegaSaenz de Miera E, Rudy B (1999) Molecular diversity of K⫹ channels. Ann NY Acad Sci 868:233–285. Desai NS, Rutherford LC, Turrigiano GG (1999a) Plasticity in the intrinsic excitability of cortical pyramidal neurons. Nat Neurosci 2:515–520. Desai NS, Rutherford LC, Turrigiano GG (1999b) BDNF regulates the intrinsic excitability of cortical neurons. Learn Mem 6:284 –291. Dichter MA, Tischler AS, Greene LA (1977) Nerve growth factorinduced increase in electrical excitability and acetylcholine sensitivity of a rat pheochromocytoma cell line. Nature 268:501–504. Du J, Haak LL, Phillips-Tansey E, Russell JT, McBain CJ (2000) Frequency-dependent regulation of rat hippocampal somato-dendritic excitability by the K⫹ channel subunit Kv2.1. J Physiol 522: 19 –31. Erisir A, Lau D, Rudy B, Leonard CS (1999) Function of specific K⫹ channels in sustained high-frequency firing of fast-spiking neocortical interneurons. J Neurophysiol 82:2476 –2489. Go´mez-Pinilla F, Knauer DJ, Nieto-Sampedro M (1988) Epidermal growth factor receptor immunoreactivity in rat brain: development and cellular localization. Brain Res 438:385–390. Gonzalez M, Collins WF 3rd (1997) Modulation of motoneuron excitability by brain-derived neurotrophic factor. J Neurophysiol 77:502– 506. Hille B (2001) Ion channels of excitable membranes. Sunderland, MA: Sinauer. Huguenard JR, Hamill OP, Prince DA (1988) Developmental changes in Na⫹ conductances in rat neocortical neurons: appearance of a slowly inactivating component. J Neurophysiol 59:778 –795. Hwang PM, Fotuhi M, Bredt DS, Cunningham AM, Snyder SH (1993) Contrasting immunohistochemical localizations in rat brain of two

645

novel K⫹ channels of the Shab subfamily. J Neurosci 13:1569 – 1576. Juniper MP (2000) What role(s) for TGF␣ in the central nervous system? Prog Neurobiol 62:443–473. Kornblum HI, Hussain RJ, Bronstein JM, Gall CM, Lee DC, Seroogy KB (1997) Prenatal ontogeny of the epidermal growth factor receptor and its ligand, transforming growth factor alpha, in the rat brain. J Comp Neurol 380:243–261. Kornblum HI, Gall CM, Seroogy KB, Lauterborn JC (1995) A subpopulation of striatal GABAergic neurons expresses the epidermal growth factor receptor. Neuroscience 69:1025–1029. Lau D, Vega-Saenz de Miera EC, Contreras D, Ozaita A, Harvey M, Chow A, Noebels JL, Paylor R, Morgan JI, Leonard CS, Rudy B (2000) Impaired fast-spiking, suppressed cortical inhibition, and increased susceptibility to seizures in mice lacking Kv3.2 K⫹ channel proteins. J Neurosci 20:9071–9085. Lesser SS, Lo DC (1995) Regulation of voltage-gated ion channels by NGF and ciliary neurotrophic factor in SK-N-SH neuroblastoma cells. J Neurosci 15:253–261. Lesser SS, Sherwood NT, Lo DC (1997) Neurotrophins differentially regulate voltage-gated ion channels. Mol Cell Neurosci 10:173– 183. Lien CC, Martina M, Schultz JH, Ehmke H, Jonas P (2002) Gating, modulation and subunit composition of voltage-gated K⫹ channels in dendritic inhibitory interneurones of rat hippocampus. J Physiol 538:405–419. Llina´s RR (1988) The intrinsic electrophysiological properties of mammalian neurons: insights into central nervous system function. Science 242:1654 –1664. Lo DC (1998) Instructive roles of neurotrophins in synaptic plasticity. Prog Brain Res 117:65–70. McBain CJ, Fisahn A (2001) Interneurons unbound. Nat Rev Neurosci 2:11–23. McFarlane S, Cooper E (1993) Extrinsic factors influence the expression of voltage-gated K currents on neonatal rat sympathetic neurons. J Neurosci 13:2591–2600. Maletic-Savatic M, Lenn NJ, Trimmer JS (1995) Differential spatiotemporal expression of K⫹ channel polypeptides in rat hippocampal neurons developing in situ and in vitro. J Neurosci 15:3840 –3851. Malin SA, Nerbonne JM (2000) Elimination of the fast transient in superior cervical ganglion neurons with expression of KV4.2W362F: molecular dissection of IA. J Neurosci 20:5191– 5199. Malin SA, Nerbonne JM (2002) Delayed rectifier K⫹ currents, IK, are encoded by Kv2 alpha-subunits and regulate tonic firing in mammalian sympathetic neurons. J Neurosci 22:10094 –10105. Martina M, Schultz JH, Ehmke H, Monyer H, Jonas P (1998) Functional and molecular differences between voltage-gated K⫹ channels of fast-spiking interneurons and pyramidal neurons of rat hippocampus. J Neurosci 18:8111–8125. Massengill JL, Smith MA, Son DI, O’Dowd DK (1997) Differential expression of K4-AP currents and Kv3.1 potassium channel transcripts in cortical neurons that develop distinct firing phenotypes. J Neurosci 17:3136 –3147. Ming Y, Bergman E, Edstrom E, Ulfhake B (1999) Reciprocal changes in the expression of neurotrophin mRNAs in target tissues and peripheral nerves of aged rats. Neurosci Lett 273:187–190. Narisawa-Saito M, Carnahan J, Araki K, Yamaguchi T, Nawa H (1999) Brain-derived neurotrophic factor regulates the expression of AMPA receptor proteins in neocortical neurons. Neuroscience 88: 1009 –1014. Ohya S, Tanaka M, Oku T, Asai Y, Watanabe M, Giles WR, Imaizumi Y (1997) Molecular cloning and tissue distribution of an alternatively spliced variant of an A-type K⫹ channel alpha-subunit, Kv4.3 in the rat. FEBS Lett 420:47–53. Ohya S, Tanaka M, Oku T, Furuyama T, Mori N, Giles WR, Watanabe

646

H. Namba et al. / Neuroscience 122 (2003) 637– 646

M, Imaizumi Y (2001) Regional expression of the splice variants of Kv4.3 in rat brain and effects of C-terminus deletion on expressed K⫹ currents. Life Sci 68:1703–1716. Pollock JD, Krempin M, Rudy B (1990) Differential effects of NGF, FGF, EGF, cAMP, and dexamethasone on neurite outgrowth and sodium channel expression in PC12 cells. J Neurosci 10:2626 – 2637. Rothe T, Bahring R, Carroll P, Grantyn R (1999) Repetitive firing deficits and reduced sodium current density in retinal ganglion cells developing in the absence of BDNF. J Neurobiol 40:407–419. Rudy B, McBain CJ (2001) Kv3 channels: voltage-gated K⫹ channels designed for high-frequency repetitive firing. Trends Neurosci 24: 517–516. Rutherford LC, DeWan A, Lauer HM, Turrigiano GG (1997) Brainderived neurotrophic factor mediates the activity-dependent regulation of inhibition in neocortical cultures. J Neurosci 17:4527–4535. Rutherford LC, Nelson SB, Turrigiano GG (1998) BDNF has opposite effects on the quantal amplitude of pyramidal neuron and interneuron excitatory synapses. Neuron 21:521–530. Schwindt PC, Spain WJ, Foehring RC, Stafstrom CE, Chubb MC, Crill WE (1988) Multiple potassium conductances and their functions in neurons from cat sensorimotor cortex in vitro. J Neurophysiol 59: 424 –449. Sekirnjak C, Martone ME, Weiser M, Deerinck T, Bueno E, Rudy B, Ellisman M (1997) Subcellular localization of the K⫹ channel subunit Kv3.1b in selected rat CNS neurons. Brain Res 766:173–187. Serodio P, Rudy B (1998) Differential expression of Kv4 K⫹ channel subunits mediating subthreshold transient K⫹ (A-type) currents in rat brain. J Neurophysiol 79:1081–1091. Seroogy KB, Lundgren KH, Lee DC, Guthrie KM, Gall CM (1993) Cellular localization of transforming growth factor-alpha mRNA in rat forebrain. J Neurochem 60:1777–1782. Sherwood NT, Lesser SS, Lo DC (1997) Neurotrophin regulation of ionic currents and cell size depends on cell context. Proc Natl Acad Sci USA 94:5917–5922. Shibata R, Nakahira K, Shibasaki K, Wakazono Y, Imoto K, Ikenaka K

(2000) A-type K⫹ current mediated by the Kv4 channel regulates the generation of action potential in developing cerebellar granule cells. J Neurosci 20:4145–4155. Song WJ, Tkatch T, Baranauskas G, Ichinohe N, Kitai ST, Surmeier DJ (1998) Somatodendritic depolarization-activated potassium currents in rat neostriatal cholinergic interneurons are predominantly of the A type and attributable to coexpression of Kv4.2 and Kv4.1 subunits. J Neurosci 18:3124 –3137. Song WJ (2002) Genes responsible for native depolarization-activated K⫹ currents in neurons. Neurosci Res 42:7–14. Storm JF (1990) Potassium currents in hippocampal pyramidal cells. Prog Brain Res 83:161–187. Tkatch T, Baranauskas G, Surmeier DJ (2000) Kv4.2 mRNA abundance and A-type K⫹ current amplitude are linearly related in basal ganglia and basal forebrain neurons. J Neurosci 20:579 –588. Turrigiano G, LeMasson G, Marder E (1995) Selective regulation of current densities underlies spontaneous changes in the activity of cultured neurons. J Neurosci 15:3640 –3652. Viana F, Bayliss DA, Berger AJ (1993) Multiple potassium conductances and their role in action potential repolarization and repetitive firing behavior of neonatal rat hypoglossal motoneurons. J Neurophysiol 69:2150 –2163. Yamada M, Ikeuchi T, Hatanaka H (1997) The neurotrophic action and signalling of epidermal growth factor. Prog Neurobiol 51:19 –37. Yarom Y, Sugimori M, Llinas R (1985) Ionic currents and firing patterns of mammalian vagal motoneurons in vitro. Neuroscience 16:719 – 737. Xian CJ, Zhou XF (1999) Roles of transforming growth factor-alpha and related molecules in the nervous system. Mol Neurobiol 20: 157–183. Xiong H, Yamada K, Han D, Nabeshima T, Enikolopov G, Carnahan J, Nawa H (1999) Mutual regulation between the intercellular messengers nitric oxide and brain-derived neurotrophic factor in rodent neocortical neurons. Eur J Neurosci 11:1567–1576. Zhang L, McBain CJ (1995) Potassium conductances underlying repolarization and after-hyperpolarization in rat CA1 hippocampal interneurones. J Physiol 488:661–672.

(Accepted 13 August 2003)