we have been designing basic and advanced instruments in our institute ranging from simple physiographs to advanced analytical devices and detectors and, at the same time, have benefited from the experience to repair, upgrade, and modify the already available equipment used by faculty members, support staff, and students.

Patch-clamp electrophysiology was pioneered in specialist biophysical laboratories by biologists who were strongly orientated toward physics and collaborated closely with engineers and physicists. Eventually, equipment and software became commercially available and “user friendly.” The unfortunate side effect of this has been that it has become easier to work with the equipment, obtain data, and have the data analyzed automatically without the novice experimenter evaluating the validity of the results or analyses (8). In modern amplifiers, many features, including the pipette offset, series resistance compensation, and whole cell capacitance, which on traditional amplifiers must be adjusted manually, are automatically determined by the amplifier, communicated to the software, and stored. Previous experience on traditional amplifiers provides the users of automated amplifiers with the experience to determine the reliability and validity of these values (17).

The basic designs and theories of voltage-clamp circuitries have been described in detail by Strickholm (18–20), Smith et al. (16), and Sigworth (15); however, no report can be found of construction and reproduction of these amplifiers in other laboratories and/or any successful recording performed directly with them outside the originator's laboratory. The published designs are not easy for a novice researcher to reproduce due to extensive theoretical detail, while some sections require further explanation and clarification.

The aim of this project was to construct a simple patch-clamp amplifier for basic research and training purposes that is easy to reproduce in every laboratory with minimal technical experience. With the recent advances in the computer and software technologies, many of the amplifier functions described in older designs are now available on software platforms. For this reason, the presented design has been simplified compared with the original design to reduce the construction cost and complexity. Moreover, the modular design of this amplifier makes it easy to modify each component of the amplifier with new features and more advanced designs.

It is believed that the experience obtained during the construction and setup of this amplifier will enable many scientists to design their own amplifiers, obtain the core expertise to modify equipment to meet their specific needs for their research projects, and, at the same time, provide cheap, feasible, and practical equipment for research and training purposes.

MATERIALS AND METHODS

Printed circuit board design and development.

The schematic circuits were designed and virtually tested using CircuitMaker 5.0 software (MicroCode Engineering). The printed circuit boards (PCBs) were designed using PCB Maker 1.0 software (Houston Systems) and printed on transparent single-sided FR4 PCBs using regular lithographic procedures. After the process had been completed, the PCBs were washed in chloroform and then acetone to remove any residue from the boards. The PCB bottom layer, top silkscreen, and front panel designs presented as Supplementary Material in this work are of actual size.¹

¹Supplemental Material for this article is available online at the Advance in Physiology Eduction website.

They can be easily used to reproduce the PCBs used in the construction of the amplifier.

Power supply.

The amplifier and its accessories are powered by a regulated ±15-V power supply (Fig. 1) with a ripple of <0.2 mV.

Input circuit.

The command voltage, generated by the digital-to-analog interface of a computer or pulse generator, is supplied to the amplifier via the input stage (Fig. 2). In this circuit, two separate inputs go through op-amps (U10 and U11) and along with an optional holding potential (V_H), which is calibrated to read the voltage on the V HOLD (P1) dial, add up in U12. The V_H is switched on or off and its polarity is selected via the polarity selector (S3). All operational amplifiers used in the amplifier are OP27 (Analogue Devices) except in the headstage. The command voltage (V_c) supplied to inputs can be inverted by S1 and S2 switches. The bath electrode is connected to ground voltage (V_g) out and its potential is adjusted by the junction zero (P2) potentiometer and polarity selector (S4). Henceforth, all trimmers (TRM) are of the multiturn onboard type, and all potentiometers (P) are of the 10-turn type coupled to a 10-turn dial on the panel.

Signal generator.

An optional internal pulse generator is shown in Fig. 3. It was found during the construction and calibration of amplifier that it was easier to use the internal pulse generator. The pulse generator provides square pulses with adjustable durations, intervals, and amplitudes. Pulses with durations of 5–200 ms, intervals of 20–1,000 ms, and amplitudes of 0–1,500 mV can be generated with the provided capacitors. The HEX Schmitt Trigger gate (CD4584) generates a square pulse with an interval that is defined by C2 and the interval potentiometer (P1). The leading edge of this pulse triggers monostable multivibrator 4098, which, in turn, produces a pulse with a duration determined by C4 and the duration potentiometer (P2). The latter pulse triggers the CMOS analog switch DG300. The input of the CMOS switch is connected to the positive voltage via TRM1 and the amplitude potentiometer (P3), which adjust the amplitude of the pulse. The circuit may easily be used as a pulse generator and pulse trainer for other experiments such as isolated organs setups, for which input 2/output 2 is provided.

The headstage.

The headstage acts as a current-to-voltage converter (Fig. 4) based on AD549 (Analog Devices), an ultralow input-bias current operational amplifier. Alternatively, it can be replaced by OPA128 (Burr Brown), which has a lower input bias and has a similar pin out. This op-amp operates in unity gain follower configuration and records the potential at the tip of the electrode through its noninverting lead. The feedback resistor (R_f) is usually 100 MΩ for whole cell experiments, but higher values may be used for higher resolution or lower values for the injection of higher currents. To minimize the input current leak, the noninverting lead of AD549 is directly soldered to the input Bayonet-Neill-Concelman (BNC) connector of the headstage rather than going through the PCB holes (7). The R_f (R1) and capacitance compensation capacitor (C1) connect to the BNC connector in the same manner. In this configuration, the input lead is isolated from the surrounding voltages. The casing of the AD549 is driven by the output voltage (V_e) to minimize the capacitance. A 25-Ω resistor decouples the output of op-amp from the cable capacitance to avoid oscillations. Like all other op-amps in this project, TRM1 is used to null the offset voltage of the amplifier, as explained later. The original designer claims that the headstage can be used for single-channel recordings by replacing the R_f (R1) with a high-value resistor (i.e., 10 GΩ) and recalibrating the amplifier (19).

The main amplifier.

Most patch-clamp measurements are performed in voltage-clamp mode, where the membrane potential (V_m) is held at a given voltage while the current passing through membrane channels is measured (13). Voltage clamp usually occurs through an electronic feedback system where the measured potential (V_e) is compared with the potential set by the experimenter (V_c). Any deviation of the recorded potential from the V_c is instantly corrected by compensatory current injection (I_f). This current is an accurate representation of the ionic current under investigation (8) and is proportional to the number of open channels in the membrane at that specific V_c. Such small currents are usually measured by recording the voltage drop across a large R_f (14). The voltage recorded at the electrode tip (V_e) when using patch electrodes is very close to V_m due to the low resistance of the electrode (usually <10 MΩ). The change in V_e has been shown to be (19) as follows:

where R_m is the membrane resistance and R_e is the electrode series resistance. If R_e is considerably smaller than R_m, like in patch-clamp electrodes, then θ is very close to unity and V_e is very close to V_m. For this reason, series resistance compensation is not generally necessary with patch electrodes, which have a low resistance (13, 19). The original design includes a complete series resistance compensation used in the intracellular recording, which has been omitted for simplicity in this circuit (19).

Voltage-clamp mode.

The voltage at the tip of the electrode (V_e), which is very close to V_m in patch-clamp experiments, is measured by U0 in the headstage and sent to the U1 op-amp via the 3-pole/3-stage “mode” switch (S1/3 in Fig. 5). S1/3, along with S2/3 and S3/3, put the amplifier in voltage-clamp mode (VC mode), electrometer (V_m) mode, or current-clamp mode (CC mode). When a command potential (V_c) is applied in voltage-clamp mode, the required current (I_f) is injected by U1, resulting in a shift of V_e toward V_c. Changes in the output of U1 (V_o) in response to the change in V_c (V_e = −V_c) have been shown to be as follows:

to measure the current passing through the R_f (Fig. 6), V_e is subtracted from V_o in unity gain configuration in op-amp U2b, giving the index of current (V_i) passing through the R_f as follows: V_i is supplied to the nulling circuit for further calculations. U4 and U5 provide amplified versions of V_e and V_c to the capacitor connected to the electrode input to compensate for the input capacity. The capacitance compensation is discussed in detail below.

In current-clamp configuration, the mode switch is placed in the current-clamp position. The actual measurement is seen in V_e so the U2b op-amp becomes unused. With a command signal V_c, constant current is passed through the R_f and through any attached input to V_e (microelectrode, cell, ground, etc.). The current (I_c) has been shown to be as follows:

Output and nulling circuit.

S1(3/3) connects V_e to U8 in current-clamp mode and V_i to U8 in voltage-clamp mode through calibration trimmers TRM7 and TRM8, which are explained in Calibration and operation (Fig. 7). U7 provides a calibrated proportion of V_c to U8 to determine the membrane and/or electrode resistance (P1) by nulling the signal at the amplifier output (V_io) in current-clamp mode. In voltage-clamp mode, P1 can be used for leak subtraction. V_m is nulled using P2 and its value is shown on its dial. U8 has a switchable gain, and its output is passed through U9 to be filtered at four different frequencies. V_io is the final output of the system, which is connected to the oscilloscope and/or data-acquisition interface.

Capacitance compensation.

Currents across a cell membrane consist of two components: ionic current flowing through ion channels of interest and capacitive current, which charges the membrane. The capacitive current masks early events such the activation of Na⁺ channels. As under ideal conditions, capacitive currents are linear and not voltage dependent, and they can be omitted from the final output by injecting an equal amount of current to the pipette via the capacitor in the headstage. Any remaining transient can be subtracted from the signal of interest through a process called leak subtraction (17).

For the first method, the amplifier features two controls for capacitance compensation: NEGATIVE CAPACITANCE and STEP FUNCTION, driven by U4 and U5 op-amps (Fig. 5). U4 is driven by V_e and is used for the compensation of pipette capacitance (P2), which appears as narrow spikes with the same polarity at the beginning and end of the command pulse. U5 is driven by V_c and is used for the cancellation of slower transients (P1) in conjunction with negative capacitance. The ratio of applied capacitance compensation can be adjusted using the BALANCE knob (P3). The provided circuitry can effectively remove pipette capacitance transients as well as most of the cell membrane transient if small cells are being recorded.

The removal of the remaining transients, especially from larger cells, is best done using a divided pulse protocol such as P/N subtraction, which is available on most commercial software as well as the free software listed below in this work. This subtraction can be done either online or offline. P/N leak subtraction was first proposed by Bezanilla and Armstrong (2). In this subtraction scheme, each test voltage step is preceded by a series of N (typically 4) leak voltage steps of 1/N (1/4 or −1/4, depending on polarity) amplitude of the test pulse activated from a potential at which no voltage-activated currents are activated. In a P/4 protocol, these 4 × 1/4 amplitude traces are summed together and are subtracted from the actual current trace of interest to isolate the ionic current of interest. Note that subtraction of capacitive and leakage currents is purely cosmetic. It does not actually improve the signal recorded. However, it may reveal small current components that would otherwise be difficult to identify (17).

Another technique for speeding up the response to a command step and elimination of capacitance transients is “supercharging,” as introduced by Armstrong and Chow in 1987 (1). Supercharging is an open-loop method, meaning there is no feedback and little risk of oscillations. Supercharging is accomplished by adding a brief “charging” pulse at the start and end of the command voltage pulse. This means that, initially, the membrane is charging toward a larger final value than expected. In its crudest form, the charging pulse amplitudes or durations are adjusted empirically by the investigator so that the V_m does not overshoot. Since the V_m is not directly observable by the user, this is accomplished by adjusting the controls until the current transient is as brief as possible (14). The simple yet effective design provided by Strickholm (20) is an add on to the command signal pathway, which is easy to adjust and automatically provides the appropriate supercharging potentials for step commands of different amplitudes.

It is important to know that when recording from cells with complex geometries, only the fastest component of the whole cell charging transient can be nulled and no effort should be made to remove the portion of the slower components associated with process charging as this would result in overcompensation of the cell body (11).

Assembly.

The power supply is housed in a separate metal casing and connected to other circuits via a cable. Any 3-point connector may be used to connect the power supply to the main amplifier; however, caution must be taken to avoid connection of wrong polarities when connecting/disconnecting power cables between the power supply and amplifier. The headstage circuit is placed in an aluminum casing of the proper size and connected to the main amplifier via a multicore shielded cable not longer than 2.5 m. R_f, V_e, and capacitance compensation cables must be connected using the individually shielded cores of the cable (Fig. 4). A BNC connector is mounted on the headstage aluminum casing, which connects to the noninverting input of the U0 op-amp along with the capacitance compensation capacitor and R_f, without making contact or being close to the PCB or casing. This formation results in the complete insulation of the input-BNC junction. A DB-9 serial connector is used to connect the headstage cable to the main amplifier. It is very important to wash the BNC connector, aluminum casing, and headstage PCB in chloroform and then acetone to remove any possible residue that may cause a current leak between the holder connector and other grounded parts.

The command input circuit, signal generator, main amplifier, and nulling circuit are placed in a suitable metal casing. Power is provided to all four boards, and the output of the signal generator is connected to input 1 of the command input stage. Input 2 (EXT. INPUT) and V_g OUT are connected to BNC connectors mounted on the front panel of the casing. The output of the command input circuit is connected to V_c inputs on the input stage and nulling circuit. V_e and V_i outputs of the main amplifier are connected to the respective inputs on the nulling circuit. V_io is connected to the front panel mounted BNC (OUT). Headstage connector (DB9) is mounted on the front or rear panel and is connected to the main board. All the selectors, switches, and multiturn potentiometers are mounted on the front panel and connected to the board (Fig. 8). It must be noted that connections to any input/output devices outside the metal casing must be done through high-quality shielded cables. A ground connection is provided for each input/output connection for possible monitoring purposes (e.g., multimeter or oscilloscope). Shielded cables are necessary for connecting V_g, input 2 (Fig. 2), and V_io BNCs to the PCBs. On the top silkscreen, the switches and selectors are shown in a way that the common/center pin is clearly shown and potentiometer connectors should be connected in the same manner as shown on the top silkscreen to provide the correct rotation on the panel.

Grounding.

Proper grounding is an essential part of a patch-clamp recording setup. All metal casings, ground connections, BNCs, and wire shielding must be properly grounded using the star formation to avoid ground loops. In this formation, a large metallic mass (i.e., an iron block) is connected to the ground cable of the laboratory, and all the cables coming from the instruments are connected to the metal mass. A proper and clean ground connection is essential to the quality of the recording (8).

Model cell.

A model cell is provided that can be used for initial and periodic calibration of the system (Fig. 9). The provided model cell resembles a 100 MΩ electrode connected to a cell with access resistance of 100 MΩ and a membrane capacitance (C_m) of 50 pF. A switch is provided to select “electrode” or “cell” modes.

Software and data acquisition.

Data acquisition can be performed by a simple connection of the amplifier output (V_io) to an oscilloscope and recorded using a video camera and tape recorder. In this experiment, the output of the amplifier was connected to analog input of an Advantech PCI-1716 data-acquisition board (Advantech). The pulse-generating software and pulse-recording software were compiled in Labview 7 Express software (National Instruments). Alternatively, the authors recommend one of the free software available online and listed below. The DAQ cards that are supported by each software are listed in their respective webpages [Strathclyde Institute of Pharmacy and Biomedical Sciences, http://spider.science.strath.ac.uk/sipbs/page.php?show=software_ses; and the Department of Physiology, University of McMaster (10), http://www-fhs.mcmaster.ca/huizinga/labpatch.htm].

The above free software packages offer a wide variety of advanced features such as pulse protocol generators, digital leak subtraction, pipette resistance measurement, and many other features.

Calibration and operation.

The recommended R_f in the headstage is 100 MΩ for patch-clamp experiments. Higher resistor values can be used for higher resolution, and lower values may be used for larger current injections. To start a routine calibration procedure, the amplifier is set to current-clamp mode, and current injections through the command inputs are set to zero by shorting the inputs to the ground. The headstage input is also shorted to the ground. All op-amps should be balanced for the input offset voltage starting from U10 to U12 and then U0, U1, U2b, U4, U5, U7, U8, and U9 in sequence by adjusting the offset TRMs while reading the output of each op-amp by a multimeter or oscilloscope. P2 and P3 in Fig. 7 must be adjusted to provide zero voltage to the U8 input. Next, to simulate a microelectrode, a 100-MΩ resistor is connected from the headstage input to the ground (Fig. 9), and a −100-mV square wave command (V_c) is applied to the amplifier through either the pulse generator or EXT. INPUT.

This gives V_e = [(R_e/R_f)(−V_c)] or 100 mV = [(100 MΩ/100 MΩ)(−100 mV)]. TRM7 in Fig. 7 is adjusted so that the value is observed at V_io = V_e.

The K1 (pipette resistance) potentiometer, which is attached to a multiturn dial on the panel, is set to read a multiple of R_e (e.g., 1.00 is indicative of 100 MΩ), and B7 and K7 are adjusted to null the signal observed at V_io. The K1 dial now reads the actual input resistance. The provided model cell is switched to the “cell” position; thus, a complete cell is constructed consisting of a 100 MΩ resistor as R_m, a 100-MΩ resistor as Re, and a 50 pF capacitor representing C_m. At the null of the steady state, K1 reads R_e + R_m, which should read 2 at this point, indicating a resistance of 200 MΩ at the headstage input.

The model cell is switched back to “electrode” mode, and the amplifier is set to voltage-clamp mode.

A −100-mV step command is applied. The current passing through the electrode here corresponds to I_f = V_i/R_f = −V_c/R_e. Using TRM8 in Fig. 7, V_io is calibrated to read in pA/mV in different gains according to Table 1.

The V HOLD potentiometer (P1 in Fig. 2) is adjusted using TRM6 to provide a range of ±100 mV and so that the value read on the dial corresponds to the actual value measured at the output. The same procedure is repeated for JUNCTION ZERO (P2) using TRM4 and TRM5 in Fig. 2 to provide ±100 mV. To measure the V_m in CC, K_m (V_m, P2) is adjusted using TRM6 in Fig. 7 to provide ±100 mV.

Junction potential measurement.

The bath saline solution is grounded by a reference electrode. The ground offset adjustment is used to compensate for imbalances in electrode inputs. R_g is calibrated to provide a range of ±100 mV. With similar saline solutions in the bath and pipette and proper AgCl coating of the silver electrode, the offset should be near zero (19).

Validation of linearity.

To ensure that the amplifier responds in a linear manner in all V_cs without any rectification, the system was tested using the model cell (Fig. 9) in “electrode” mode by applying increasing pulses from a V_c of −100 mV stepped to +100 mV and from a V_c of +100 mV stepped down to −100 mV. The test revealed that the amplifier responded in full linearity throughout its full operational range (R² = 1 for both positive and negative V_c; Fig. 10).

Electrode holder.

To make the replacement of electrodes simple and fast, the electrode holder was connected to the headstage using a BNC connector. In this design, which has been described in detail by Hamill et al. (6), it is possible to disconnect the holder from the headstage and replace the electrode away from the patch-clamp rig. Also, regular regeneration of the AgCl coating on the silver wire of the holder is easy to perform (6). The silver BNC connector has a Teflon insulator, which has a volume resistivity of 10¹⁷–10¹⁸ Ω.

Recoding procedure for whole cell voltage clamp.

The device is switched on in CC mode with zero current injection and capacitance compensation controls, and experimental parameters (V_m and pipette resistance) are set to zero. The headstage input is shorted to ground. The device is left on for 30 min, and the baseline is adjusted by means of Output Zero. A pulled and heat-polished electrode is connected to the holder. Moderate air pressure is applied to the electrode holder to prevent contamination of the electrode tip while it is being immersed in the bath solution. Possible capacitance spikes due to pipette resistance are compensated using negative capacitance. The observed response at V_io is shown in Fig. 11A. The junction potential is zeroed using the JUNC. ZERO dial (Fig. 11B). In patch-clamp experiments, the junction potential should not be more than a few millivolts with a similar chloride ion concentration in the pipette and external solution (19). To measure electrode resistance, a square pulse of −10 mV with a duration of 200 ms is applied, and the signal at V_io is nulled using the pipette resistance dial (K1), which shows the electrode resistance on the dial (Fig. 11C). The amplifier is now ready for CC experiments. To proceed to a voltage-clamp experiment, the pipette and membrane knobs are returned to the zero position, and the amplifier is switched to voltage-clamp mode. A response that is indicative of electrode resistance is seen at V_io and the current passing through the feedback resistor is I_f = V_i/R_f = −V_c/R_e, with I_f = 1,000 pA for a 10-MΩ electrode with the gain switch set to ×10 (Fig. 11D).

Formation of a gigaseal is indicated by the disappearance of the response in V_io (Fig. 11E). Perforation can be performed by applying gentle suction at the tubing, which is indicated by a slight increase in the response seen at V_io and the appearance of a larger transient (Fig. 11F). Next, the amplifier is placed in V_m mode and the V_m is measured by zeroing the baseline using the V_m dial (K_m). K_m is returned to zero, and the V_c is set to read the V_m in either the software or using the V HOLD dial. The amplifier is switched back to voltage-clamp mode, and the membrane baseline current should read zero.

To cancel/minimize the capacitive transients seen at V_io, short hyperpolarizing pulses (10 mV) are applied, and the transient is cancelled using the STP FUNC and NEG CAP dials. To cancel the linear leakage currents, the current observed after the application of 10 mV hyperpolarizing pulses is zeroed using the pipette resistance (K_m) potentiometer. Alternatively, positive/negative subtraction function of the software can be used for this purpose. The preparation is now ready for the pulse protocol through the EXT CMD input.

Actual recording from cells.

The amplifier was used to record inward Ca²⁺ currents from PC12 cells. Rat pheochromocytoma (PC12) cells were cultured in DMEM with 5% FBS, 5% horse serum, 100 mg/ml streptomycin, and 100 IU/ml penicillin at 37°C in a humidified atmosphere of 95% air-5% CO₂. Cells were differentiated in the presence of 50 ng/ml nerve growth factor for 4 days. To perform whole cell recordings, cells were plated on poly-l-lysine-coated glass coverslips. Except where noted, all chemicals and reagents were purchased from Sigma (St. Louis, MO). Whole cell recordings were made according to standard techniques under an inverted microscope (Olympus) along with a manual 3-axis water hydraulic micromanipulator (MHW-3, Narishige).

Data were low pass filtered at 2 kHz and digitized at 5 kHz with 16-bit resolution via an Advantech PCI-1716 board (Advantech). Stepping pulses were produced with the same card. Recorded data were analyzed offline using Labview 7 Express software (National Instruments). After whole mode had been established, capacitive transients were cancelled at 5 mV depolarizing voltage pulses with analog compensation circuitry of the amplifier. Linear leakage was compensated electronically (K₁ control) at −10-mV hyperpolarizing pulses from the V_c.

Voltage-activated Ca²⁺ currents were evoked by 200-ms depolarizing steps from a V_c of −80 mV with a 10-mV increase in pulse amplitude per step.

The following standard whole cell bath solution was used for PC12 experiments: 10 mM BaCl₂, 135 mM tetraethylamonium-Cl, 1 mM MgCl₂, and 10 mM HEPES (pH 7.3). Pressure-polished pipettes had a final resistance of 2–5 MΩ and were filled with 125 mM CsCl, 10 mM tetraethylamonium-Cl, 5 mM EGTA, 10 mM HEPES, 1 mM MgCl₂, 10 mM glucose, 4 mM ATP, and 0.5 mM GTP (pH 7.2). During the recording, cells were continuously perfused with the test solution at a rate of 1 ml/min. To facilitate drainage of the perfusion bath, the drain pipe was connected to a sealed plastic container (50-ml volume) inside the Faraday cage, which functioned as a trap. The container was connected to a peristaltic pump placed out of the Faraday cage via a polyethylene tube (3 mm inner diameter) to produce negative pressure inside the container, resulting in continuous and consistent flow of the solution outside the perfusion bath.

RESULTS

Recorded whole cell Ba²⁺ currents are shown in Fig. 12, and the respective current-voltage curve is shown. The response of the amplifier was consistent with the expected results during different recording steps and throughout the experiments.

DISCUSSION

This article is the first published work that constructs the circuitry described by Strickholm (18–20) and performs actual recordings with it. The described design provides proper performance for routine whole cell patch clamp. Step-by-step validation using a model cell yielded precise, linear, and reproducible results, while actual experiments performed on cells produced acceptable outcomes. The constructed amplifier was successfully used in our laboratory to evaluate the inhibitory effects of two new Ca²⁺ channel blockers (12).

The construction of a patch-clamp amplifier and setting up a patch-clamp rig gives scientists and postgraduate students the unique opportunity to understand the basic and advanced concepts of electrophysiology in the aspects of instrumentation. This experience is also vital for neuroscientists who plan to set up a new laboratory on their own. Moreover, this work also aims at providing a simple guideline for those investigators with lower budgets and experience. Application of the principles enunciated here should result both in increased control over the experimental environment and offer some level of budget savings. The cost of electronic parts used in the construction of the amplifier presented in this work is estimated to be less than $400 (United States currency) based on retail prices. A further reduction in the cost of setting up a patch-clamp rig for training and educational purposes is possible by in-house construction of a Faraday cage, invert microscope, and vibration isolation table. An article by Bustamante (5) has provided details for the construction of a relatively inexpensive inverted microscope that meets the specifications required for patch-clamp and other electrophysiological investigations at the cellular level. The principal parts needed for the construction of the microscope may be obtained either from an old microscope or purchased from a variety of manufacturers (5). The vibration isolation table described by O'Reilly and Richardson (10) combines an ergonomic design with a cost-effective vertical vibration isolation system. A similar table was developed in our laboratory at a cost of approximately $800 including labor, materials, and the purchase of a commercial tabletop (9).

The original design claims that the amplifier can be adopted to record single-channel current using a high-value R_f (10GΩ) in the headstage and recalibrating the amplifier (19). However, the aforementioned claim is based on experiments performed on model cells. It should be noted that noise and response time are major limiting factors in single-channel recording. Further experiments in low-noise environments using high-quality GΩ resistors with minimal thermal noise are needed to prove this claim.

The authors would like to emphasize that a patch-clamp amplifier is a complex and sensitive piece of electronic equipment, and the main goal of this article is to provide a simplified instruction for beginners to experience the step-by-step construction, calibration, and setup of a basic patch-clamp rig to maintain and eventually modify their equipment in the future. The organization and details in this work are provided in such a way that once all the needed parts are available, an amateur with minimal experience in hobby electronics and basic understanding of electricity concepts can start building the amplifier immediately and roughly finish the project in 2 wk. However, the authors would like to suggest that beginners take a few extra days to read one of the widely available books on the basics of electronics and electricity (3, 4).

The experience obtained during the construction and setting up of a patch-clamp amplifier also provides the basic expertise to upgrade, modify, and design many other laboratory tools for teaching and research purposes and overcome the cost barrier of setting up and expanding neurophysiology laboratories.

GRANTS

This work was supported in part by research project no. 10-M-T as the PhD thesis of A. Rouzrokh.