3.1 Brain slices 



Experiments with brain slices were carried out on 14 to 21 days old male or female Wistar rats in compliance with international guiding principals for the use of animals in research. The number of animals was limited to the necessary number permitting reliable statistical analysis. According to the animal rights guidelines special procedures that minimize discomfort, pain and distress were applied. The animals were kept under standardized conditions and had access to water and dry food ad libitum. Light/darkness regime of 12:12 hours at a controlled temperature of 24 ± 1 ˚C was not inverted.

Brain slice preparation


Most of the experiments in this work were performed on horizontal hippocampus-entorhinal cortex slices (Fig.9). Prior to decapitation the animals were deeply anesthetized with ether. After decapitation the skin was cut with a scalpel from the caudal (posterior) to rostral (anterior) part, so that the calvarium could be uncovered. After careful opening of the skull with sharp scissors cutting along the still soft sutura sagitalis the two cranial pieces were peeled laterally. After careful removing of the Dura mater and disrupting the optic nerves the brain was rapidly (within half a minute) freed from the skull with spatula together with the Cerebellum and washed immediately with ice-cold ACSF, bubbled with 95% O2 and 5% CO2. Slicing was done at 4-8ºC in order to keep the tissue firm and to reduce metabolic activity. The preparation was done under submersion in ice-cold oxygenated ACSF. The Cerebellum as well as the frontal third of the brain was excised and both hemispheres were separated. The caudal third of each of the hemispheres was glued with the dorsal plane to the vibratome chamber using cyanoacrylate glue. Slice preparation was done from basal to dorsal direction. Brain pieces and slices were soaked in oxyenated ACSF (or high sucrose ACSF) during the whole process of cutting. The angle and vibration frequency of the blade were adjusted to prevent the tissue being pushed while cutting the slices. Combined hippocampal-entorhinal cortex slices of 300 µm thickness were prepared using a vibrating microtome (Dosaka Microslicer DTK-1000, Kyoto, Japan). Slices were immediately and very carefully moved (using a brush) to an incubation chamber, containing oxygenated ACSF at room temperature (20-22˚C). There, the slices were allowed to recover for at least 1 hour before being transferred to the recording chamber.

Fig. 9 Preparation of horizontal hippocampal slices [Witter, 00].



As a nutrient solution for the brain slices during preparation as well as during the experimental phase artificial cerebro-spinal fluid (ACSF) was used (Table 1). This solution was continuously bubbled with carbogen gas (5% CO2, 95% O2) to maintain a pH of 7.4. While recording, ACSF was superfused at 3 ml/min at room temperature (~22oC).

Table 1 Content of acute brain slice preparation and perfusion solutions.


ACSF (Ringer) /normal/

Sucrose cutting solution


129 mM

87 mM

Stock I


1,25 mM

1,25 mM


10 mM

25 mM

MgSO 4

1,8 mM


CaCl 2

1,6 mM

0,5 mM


3 mM

2,5 mM

MgCl 2


7 mM



75 mM


26 mM

26 mM

Stock II

Both Stock I and Stock II were mixed with H2O daily in a ratio of 1:8:1 and bubbled with carbogen minimum 1 hour before use, to adjust the pH of 7.4.


Although sucrose substitution for sodium or the use of 10 Mg2+: 1 Ca2+ instead of 1 Mg2+:1 Ca2+ is reported to amend the viability of slices, the modified procedure that incorporates both of these ideas did not result in significant improvement of the slice quality.

3.2 HEK 293 cells

Heterologous expression of potassium channel cDNAs was done in human embryonic kidney cells, transformed with Adenovirus Type 5, American Type Culture Collection, 12301, Parklawn Drive, Rockville, Maryland 20852, USA; code designation HEK 293 or 293 cells.

Maintaining HEK 293 cells in culture


HEK 293 cells were grown in a monolayer in plastic petri dishes of 35 or 99mm. Under optimal growth conditions (37˚C, 5% CO2), 293 cells doubled about every 36 hours. To maintain consistency and successful transfection procedures low passage 293 cells were employed. Untransfected HEK cells were grown on big culture plates (99 mm) in 10 ml complete growth medium (Table 2) and were split every 2 days (after reaching 80-90 % confluence). 24 hours before transfection 293 cells were harvested and replated onto small petri dishes (35 mm) in 2 ml complete medium at a density to reach around 30 % on the next day. The cultures were incubated for 24 hours at 37˚C in humidified cell-culture incubator (Heraeus, Hanau, Germany) in an atmosphere of 5% CO2. The medium was changed 1-2 hours prior to transfection.

Transient transfection of HEK- 93 cell lines

The standard calcium phosphate-mediated transfection of adherent cells [Graham, 73] was used for transient transfection of HEK 293 cell lines with potassium channel cDNA. The cells were transfected with Kv1.4 and Kv4.2 potassium channel subunit cDNA.


To prevent contamination sterile media, vials, instruments and cells in a vertical laminar flow hood were used. All solutions were sterilized by filtration trough 0.22 µm-pore filtre and were brought to room temperature before use. For each transfection plate two separate 0.5 ml polystyrene tubes were prepared:

While vortexing the HBS solution, the CaCl2/DNA mixture were added dropwise to the tube, incubated for 30-40 min at room temperature, during which CaPO4-crystals precipitate with the adherent DNA. The solution was permitted to turn opaque, but no or only very few visible precipitates were allowed to form. After resuspending the CaPO4/DNA precipitate it was quickly and gently added dropwise onto the cells and rocked back and forth few times to evenly distribute the precipitate on the cells. By observing the cells under the microscope one could see very small black particles evenly distributed over the cells. The transfected 293 cells were incubated 4-5 hours at 5% CO2 and afterwards washed (3x) with sterile PBS (containing no Ca2+ or Mg2+). Thereafter PBS was replaced with pre-warmed complete medium. Incubation of 48-60 hours was usually adequate for high expression rates of the protein of interest.


The transfection efficiency drastically fell when HEK cells were of late passage or reached high confluence. Therefore it was of great importance to time the passage procedures well. Aliquots of early passages of the HEK cell line were stored in liquid nitrogen to ensure a renewable source of cells. To prepare HEK cells for freezing, cells with 80% confluence were washed once with PBS (containing no Ca2+ or Mg2+), trypsinized and harvested. Afterwards the cells were counted, collected by centrifugation (500 x g for 10 min.) and resuspended in 4˚C Cell Freezing Medium at 1-2x106 cells/ml. 1ml aliquots were then dispended into labelled freezing vials and incubated at –20˚C for 1-2 hours, -80˚C overnight and subsequently transferred to liquid nitrogen (-196˚C). There the cells could be preserved for months to years. Nevertheless the viability of the frozen stocks must be confirmed two or more weeks later by starting a fresh culture. To defrost prepared aliquots, the vials were taken from liquid N2 and placed directly into a water bath (37˚C). After short time the cells were centrifuged and the pellet was resuspended in fresh warm MEM by trituration.

Culturing media

Table 2 Content of HEK-293 cell line culturing media

Complete medium

Freezing medium

Washing medium

Storage of compound/media

supplied by


88   %

10   %





10   %

80   %





1   %






1   %




Fa.Seromed/ Biochrom KG




100   %





10   %


18˚C /hood/


MEM, minimum essential medium, FCS-fetal calf serum, L-Glu, L-glutamine; PBS-phosphate buffer saline, DMSO, dimethyl sulfoxide.


Transfection buffers

Table 3 Content of HEK-293 transfection solutions and buffers.

2xHBS /HEPES buffered saline/





280 mM



100µl 2xHBS


50 mM




1,5 mM





2 mM



cDNA /Kv1.4, Kv4.2 or EGFP/








to 100µl



1 ml/ -20˚C

1 ml/ -20˚C



HBS, HEPES//buffered saline; ddH 2 O, double-distilled water; L-Glu, L-glutamine; NaHPO 4 , sodium hydrogen phosphate;

In pyramidal neurons channels responsible for transient potassium currents are known to be comprised by two distinct principal (α) subunit genes: Kv1.4 and Kv4.2. In order to investigate probable differential modulation of Kv1.4 or Kv4.2 by AA or its derivatives and by radical oxygen species, Kv α-subunit cDNAs were separately transfected into HEK 293 cells. 


One part of the recorded HEK 293-Kvα-expressing cells was co-transfected with EGFP-cDNA as a reporter gene to identify transfected cells by fluorescence microscopy. However, this reduced transfection efficiency.

Superfusion solutions

In HEK 293 cells patch-clamp experiments a saline with the following composition (in mM) was used as a perfusion solution:


140 NaCl, 4 KCl, 2 MgCl 2 , 1 CaCl 2 , 10 HEPES, 5 g lucose, pH 7.4. All solutions were always prewarmed to 37oC before use in experiments to exclude a temperature shock of the HEK-293 cells.

3.3 Electrophysiological recordings

3.3.1 Patch pipettes

Patch pipettes were freshly pulled from borosilicate glass (GB 150F-8P, Science Products, Hofheim, Germany) using a horizontal micropipette puller (P-97 Flaming/Brown Micropipette Puller, Sutter Instrument, USA) and were then fire-polished (Microforge, Narishige, Japan). The tip of the pipette was filled with control intrapipette solution (IPS) (1-2µl) and backfilled with IPS, containing the substance of interest. It is of great importance that no air bubbles or particles block the pipette tip. Patch pipettes had initial open-pipette resistances of 3-10 MΩ. Higher resistances lead to poorer electrical access to the cell, but also less washout of interior cell contents.

3.3.2 Intrapipette solution (IPS)

The recording solution had the following composition in mM: 120 KCl, 1 CaCl2, 10 HEPES, 11 EGTA, 2 MgCl2, 20 Glucose; pH 7.3. The substances of interest were then dissolved in IPS (see Table 4). The filling solution was filtered with 0.22 µm filters (Nalgene 4-mm syringe filters with nylon membrane), filled into new 1 ml luer syringes and freshly–pulled plastic microtubes. For patch-clamp osmolarity and pH are critical, therefore they were adjusted for both perfusion (310 ± 10 mOsm/ kg, pH 7.4) and intrapipette (300 ± 10 mOsm/ kg, pH 7.3) solutions. All solutions were freshly prepared.

3.3.3 Pharmaka 


Most of the substances, used in this work, were prepared in stock solutions, if possible. Prior to experiment these compounds were diluted to required concentrations in perfusion or intrapipette solutions, depending on the application site and brought to the proper osmolarity and pH values. Unstable substances demanded preparation of fresh solutions. The following table gives an overview of the substances used, their concentrations and application sites.

Table 4 Pharmaka


Application site



Arachidonic acid /AA/

1 pM


cellular second messenger, precursor of eicosanoids

Light- and oxygen sensitive

Eicosatetraynoic acid /ETYA/

100 pM


non-metabolizable analogue of AA

Light-, and oxygen sensitive

Ascorbic acid /asc.acid/

20   µM



Light-, temperature and oxygen sensitive

0.4 mM


Glutathione /GSH/

2-20 mM



20 mM


N-acetyl-cystein /NAC/

20 mM

incubation (extracellular)


precursor of GSH


10 µM


antioxidant, water soluble derivative of tocopherol

100 µM


Hydrogen peroxide /H2O2/

80 µM


oxidizing agent

Tetrodotoxin /TTX/

1 µM


fast sodium channel blocker

AA and ETYA were stored as 1M stock solutions in DMSO at –20˚C. DMSO itself, diluted to pM concentrations had no effect on IA or IK (v). Ascorbic acid was used from freshly made stock, when needed and kept in dark container up to 5 days at –20˚C. GSH was stored as 20 mM stock in IPS, when used for intracellular application. During experimentation intrapipette solutions were kept on ice in darkness.

3.3.4 Voltage-clamp and discontinuous amplifier 


The main requirement of voltage-clamp procedure is to prevent changes in cell membrane potentials. That is why in voltage-clamp is generated compensation current, which is exactly as big as the current, which flows through the membrane, but in opposite direction. This is carried out by a negative feedback mechanism, by which the membrane potential is measured and compared with a reference value – the reference voltage. Every difference of the reference value from the actual measured membrane voltage activates a controller, which injects current with reversed direction to the cell. This compensation current is measured in patch-clamp experiments. It provides direct insight into membrane conductivity and the function of ion channels and ion transporters, by which it is determined. Simplifying, one could imagine the cell membrane as an Ohm resistor. In other words, the voltage is linearly dependent on the current (Ohm’s Law: U=R.I). Instead of the resistance R, one uses its reciprocal value, the conductance (1/R=g). The Ohm’s Law in this case is U=I/g or I= U*g. The voltage U is in this case the offset between reference voltage Uref and the reversal potential Urev for a given ion species, that carry the current: with I= (Uref- Urev).g.

The current I is the most important measurement category for all voltage-clamp amplifiers and every current change is directly proportional to the changes of the membrane conductivity.

3.3.5 Recording electrode

Chlorided silver wire, connected with the probe end of the amplifier is inserted into the glass patch-pipette filled with IPS and serves as a recording electrode.

3.3.6 Experimental setup 


In the recording chamber brain slices were superfused with carbogenated ACSF with a flow rate of 3 ml/min or HEK cells media was replaced with pre-warmed HEK perfusing saline. The electrophysiological experiments were performed at room temperature (~20°C). ECLII stellate cells and ECLIII and CA1 pyramidal neurons from acute brain slices, as well as transfected HEK 293 cells were visualized and identified under an upright microscope (Olympus BX 51WI, Germany), equipped with IR-DIC optics. In the entorhinal cortex, L III pyramidal neurons were classified by their triangular somata exhibiting one principal apical and two to three minor basal dendrites. L II stellate neurons were identified by their irregular polygonal somata. They were larger than pyramidal neurons and gave rise to four or five roughly equivalent dendrites, forming a stellate morphology. CA1 pyramidal neurons were easy to distinguish, because they are organized in clear-cut subfield within the hippocampus.

HEK cells grew in monolayer, but 2-4 days after plating there was a tendency of forming confluent cumuli. In this case the measurements were performed on the remaining single cells.

For electrophysiological recordings, only healthy cells were selected. These are cells with defined round borders, bright, with a really good contrast and a smooth appearance. Cells with rough, spotty surface or dark colour, even if they had good contrast were discarded, because those are usually seriously damaged, about to become apoptotic and do not allow formation of seals. In some cases swollen cells allowed seal formation, but no longer whole-cell recordings.


When the pipette is brought across the air-saline interface, positive pressure (0.1-0.5 psi) was applied to prevent tip contamination and to clean cell surface from tissue debris (in slices). As the HEK cells grow in monolayer high pressure would be destructive for them. After forming a bleb on the cell surface, tight seal (> 1 GΩ) was achieved by application of low negative pressure, using a 1 ml-syringe. Whole-cell break-in was accomplished by giving a very brief, sharp pulse of suction, considerably smaller in the case of HEK 293 cells. Access resistance (Ra) ranged from 10–14 MΩ, and was compensated by up to 60%; data were discarded if Ra > 15 MΩ. Input resistance was measured at a holding membrane potential level close to resting membrane potential (RMP) from the voltage response elicited by a small current pulse (-60 pA). Only cells that showed stable RMP negative to -60 mV and Ri > 120 MΩ (in current-clamp mode) were selected for study. Pharmacological tests of so selected cells were completed within 15 min. and corrected offline for eventual run down effects. Membrane potential of the cells was held at – 80 mV. All experiments were conducted at room temperature (20 - 22 ºC).

3.3.7 Experimental protocols

In whole-cell patch-clamp mode, total outward potassium currents were activated after 800ms hyperpolarizing prepulse at –110 mV (to remove inactivation) by subsequent depolarization of the membrane to potentials between –80 and +50 mV in increments of 10 mV (Fig.10 A1). When followed by 50 ms interval at –20 mV IA inactivated completely (Fig.10 A2) permitting later isolation of IA from the mixed currents.

Fig. 10 Test potential protocols for activation and inactivation of total outward and delayed rectifier potassium currents.


Delayed rectifier potassium currents were also elicited from different potentials (-110 to +20mV, 10 mV increments) and recorded at +30 mV (Fig.10 B). Ik(v) were non-inactivating in the investigated range of durations of voltage pulses. Whole-cell patch-clamp recordings were maintained in voltage-clamp mode, using an EPC-9 amplifier with built-in ITC-16 interface board, controlled by Pulse/PulseFit software (both HEKA Electronik, Lambrecht, Germany). The current signals were filtered at 3 kHz and 10 kHz and stored on a disk. The P/n method incorporated in Pulse/PulseFit software was used for leak correction (leak holding at –80 mV).

3.4 Data and statistical analysis

By depolarization of the membrane potassium permeability through a voltage-gated channel rises rapidly and then decays slower than it activates. These two processes could be described with the terms activation and inactivation. When the channels open, a mechanism is put into action to close the channel again. The special closed state in this situation is referred to as the inactivated state. The channels indeed possess two different types of gates. Closing of the inactivation gate is time dependent, whereas opening it is voltage dependent. The inactivation state can be considered time independent and is therefore termed steady-state inactivation. 

On single channel level activation and inactivation corresponds to a change in probability of given channel to open or close upon depolarization. Inactivated channels could not be activated to conducting state until their inactivation is removed. Whole-cell recording makes possible to follow the changes in this probability and the duration of these states in multiple channels on the same time. The voltage-dependence of open state probability could be measured as followed: by a negative holding potential all channels of interest are brought to a state, from which activation is possible (closed state). Then the membrane potential is manipulated to more positive values (depolarization) and one part from the channels is opening. The recorded amplitude of the ionic current is a measure of the probability of the channels to open at certain test potential at the same time. The experiment is then repeated for different test potentials and voltage-conductance relation curve is built. The conductance at each potential is normalized to the maximal conductance for the cell and plotted vs. test potentials. The resulting curve is sigmoidal and can be fitted with Boltzmann function (see below). Different pulse protocols and similar procedure of calculation is applied to describe steady-state inactivation of the channels. Both curves describe the voltage dependence of closed and opened states of the channels. These curves could be influenced by pharmaka, etc. If there is an overlap between the activation curve and the inactivation curve, then there will be a small, constant current in the area of overlap, called window current, that contribute to the setting of the resting membrane potential. A characteristic value that could be derived from voltage-conductance relation is the voltage (V50), at which exactly the half of the channels are activated (open or closed). Another essential parameter that describes the dependence of the channel conductivity from membrane potential changes is the slope of the resulting curve k. Other important parameters that could be derived at each test potential are the time constants of activation and inactivation and the maximal amplitude of the current at each voltage steps.

3.4.1 Separation of fast and persistent K+-currents using prepulse inactivation


Prepulse inactivation takes advantage of the differences in the inactivation properties of the fast (transient) and sustained (delayed rectifier) K+ currents [Klee, 95]. TTX at 1 µM was included in the bath solution to isolate transient and persistent potassium currents from fast TTX-sensitive Na+ currents (which are completely blocked by this TTX concentration). K+ currents were evoked from a holding potential of -110 mV to test pulses ranging from -80 to +50 mV in 10 mV steps. Transient K+-current (Fig.11, B) was obtained by subtracting the current obtained after a -20 mV prepulse, which elicits only the delayed rectifier K+ current (Fig.11, C), from the current obtained after more hyperpolarized prepulse (-110 mV), which elicits both fast and persistent K+ currents (total outward potassium current, Fig.11, A).

Fig. 11 Procedure for potassium current separation

The current traces were analysed “offline” using combination of Pulse/PulseFit, IgorPro and Prism software.

3.4.2 Data analysis


Current subtraction, rescaling and curve fitting were performed with IgorPro (WaveMetrics Inc.). Currents at each test potential were converted to conductances, using:

g =I/(Vi-Ek), where I is the peak outward current, E K is the reversal potential, and V i is the test pulse voltage.

Voltage-conductance relations for activation and steady-state inactivation were plotted and fitted with the Boltzmann single exponential equations for


where V 0.5 is the voltage for half-maximal activation in millivolts, V is the test pulse voltage, k is the corresponding slope factor, and g / g max - the normalized conductance. 

For effects on the maximum amplitude, currents at each recording time, measured at +30 mV, were normalized to the current at time=0 (immediately after obtaining whole-cell configuration, ‘control’) and plotted against time. Data are presented as means ± SEM (Standard Error of Means) or % of control. Statistical analyses of the data were performed with Prism (GraphPad Software Inc.) using Student’s unpaired t-test. Probability of p<0.05 is considered to be significant.

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML-Version erstellt am: