Electrical detection of the contact between a microinjection pipette and cells

Author(s) Lukkari, M.; Karjalainen, M.; Sarkanen, R.; Linne, M.-L.; Jalonen, T.; Kallio, P.

Title Electrical detection of the contact between a microinjection pipette and cells

Citation Lukkari, M.; Karjalainen, M.; Sarkanen, R.; Linne, M.-L.; Jalonen, T.; Kallio, P. 2004.

Electrical detection of the contact between a microinjection pipette and cells. 26th International Conference of the IEEE Engineering in Medicine and Biology Society, 1.- 5.9.2004, San Francisco, USA 2557-2560.

Year 2004

DOI http://dx.doi.org/10.1109/IEMBS.2004.1403735 Version Post-print

URN http://URN.fi/URN:NBN:fi:tty-201410081494

Copyright © 2004 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

All material supplied via TUT DPub is protected by copyright and other intellectual property rights, and duplication or sale of all or part of any of the repository collections is not permitted, except that material may be duplicated by you for your research use or educational purposes in electronic or print form. You must obtain permission for any other use. Electronic or print copies may not be offered, whether for sale or otherwise to anyone who is not an authorized user.

Abstract—This paper presents the basic design and tests of a device designed for detecting the contact between a microinjection pipette and cell membrane. The device facilitates the automation of the microinjection procedure of living adherent cells. The measurement of the contact is based on measuring the resistance of the pipette. Breakage and clogging of the pipette can be detected with the same measurement.

Keywords

—

Current state-of-the-art injection micromanipulators are typically joystick controlled and therefore still slow due to the manual or semi-automatic operation. Automation of the injection system is needed for making a faster and more reliable research instrument. In this work, a device for measuring the contact between a microinjection pipette and a cell has been developed in order to automate intracellular injection of adherent cells. Also the breakage and clogging of the pipette can be detected with the same device and from the same measurement signal.

The function of the device is based on measuring the resistance of the pipette in a similar way as in a patch clamp current/voltage recording device. In a patch clamp device, a ground reference electrode is placed in a recording chamber and a stimulus pulse (a square pulse) is given via a recording electrode inside the glass pipette. The pipette current is measured with a sensitive current-to-voltage converter. The pipette has a specific resistance which determines the voltage of the measured signal. When a patch clamp pipette is brought in contact with the cell, it forms a Giga Ohm resistance seal (Giga-seal), and the increase in the resistance reduces the current, which can be seen as smaller or completely disappearing pulses.

1Test measurements with a patch clamp device and injection pipettes were made before building the contact detection device. Similar effects were seen with injection pipettes as with conventional patch clamp pipettes. The size reduction of the pulse, when using the commercially available microinjection pipettes is not as large as with conventional patch-clamp pipettes, because the tip of an injection pipette is considerably smaller. However, the test measurements showed that it is possible to use a similar

The research work has been carried out as part of the project Integration of Automatic Intracellular Microinjection and Bioelectrical Recordings supported by the Academy of Finland (Grant number 102342).

measurement also with injection pipettes. When a pipette breaks, the resistance between the pipette electrode and ground reference electrode is reduced, and the pulse gets larger. Clogging of the pipette increases the resistance of the pipette and the pulse gets smaller, similarly as in contact with a cell.

Electrical Detection of the Contact between a Microinjection Pipette and Cells M. J. Lukkari

, M.I. Karjalainen

, R. Sarkanen

, M-L. Linne

, T.O. Jalonen

, P.J Kallio

1Institute of Automation and Control ²Institute of Signal Processing, Tampere University of Technology

3Institute of Medical Technology and ⁴Cell Research Center, University of Tampere

5Department of Biological and Environmental Science, University of Jyväskylä, Finland

For automated microinjection purposes, such a complex device as for the patch clamp measurements is not needed.

The purpose of this work is to make a simplified design for measuring only the here mentioned changes in pulses, facilitating the automation of intracellular microinjections.

Furthermore, the implementation of the contact detection device into a micromanipulator must be taken into account in design, which requires the miniaturization of the whole contact detection device.

II.SYSTEM DESIGN

The system design of the contact detection device is based on the general principle of a current/voltage clamp circuit (such as presented e.g. in [1], [2] and [3]). A simplified system design block diagram of the contact detection device is presented in Fig 1. The stimulus signal processing and scaling block includes adjustable scaling of the stimulus signal and buffering of the input signal. The current measurement block consists of a sensitive current- to-voltage converter and of a differential amplifier. The measurement electrode is placed inside the injection pipette filled with conductive solution and the ground reference (signal ground) electrode is in the recording chamber containing the cells in recording medium. The signal conditioning and amplification block includes a 1 kHz low- pass filter and an output amplifier having different amplifications from 2 to 20.

CURRENT MONITOR

STIMULUS INPUT STIMULUS

PROCESSING AND SCALING CURRENT

MEASUREMENT PIPETTE

SIGNAL CONDITIONING AND

AMPLIFICATION

Fig 1: A simplified block diagram of the contact detection device.

The stimulus input signal was decided to be generated with computer software using a D/A-board and the current was decided to be measured with an A/D-board. In this way, it is also possible to implement offset compensation for the measurements.

III.IMPLEMENTATION

As any generally used patch clamp device, the contact detection device was built to have two different parts: a head stage and a control unit. The head stage includes the measurement current-to-voltage converter circuit and the control unit includes all the other functions of the device.

The smallest signal levels used with the contact detection device are usually 1-10mV, and especially the smallest signal levels set special demands for the design of the head stage and the materials used in it. The head stage includes the most sensitive parts of the contact detection device and because of the low signal levels the components and also the materials of the head stage have to be carefully selected.

The case of the head stage was made of a tight aluminum closure for low noise recording (Fig. 2). A pipette is attached to the head stage by using a pipette holder. The recording electrode is connected to the pipette holder and goes into the pipette. To minimize the noise, the pipette holder is attached directly to the head stage. The measurement electrode of the contact detection device is connected to the pipette holder by placing it on the metal groove in the middle of the holder and a half circle piece of the holder is used to lock the electrode on its place with a rubber o-ring (Fig. 2). The metal groove is soldered to a wire that is connected to the measurement circuit inside the head stage.

Two features of the pipette holder are important: the holder must be mechanically stable and must not introduce electrical noise [2]. The insulating parts of the holder should be of a low-loss material, and should have a hydrophobic surface to prevent the formation of conducting water films.

Teflon best fulfils these criteria [3]. An unshielded type of pipette holder made of Teflon was built for the contact detection device for comparing the signals with a commercially available patch clamp device currently used in the laboratory (EPC-7, List-Electronic, Germany). However, a shielded holder together with the grounding of the nearby surfaces could provide sufficient shielding for 50-60 Hz interference even without the use of a Faraday cage [3]. For the implementation of the contact detection device in the injection micromanipulator, the signal will be further tested also with a shielded pipette holder, because it will be implemented near other electrical equipment in the micromanipulator where the use of Faraday cage may not be possible. However, the shielded holder introduces more of the random noise than the unshielded one, because surrounding the holder with a driven or grounded shield adds to the input capacitance and thus increases the noise.

Fig 2: A) Head stage of the contact detection device. B) Exploded view of the head stage.

The cable connecting the head stage to the control circuit carries the power supply voltages, the stimulus signal, the measurement signal, power ground and signal ground to the head stage. The contact detection device uses for this a six-conductor shielded cable. The flexibility and lightness of the cable are important, since the cable should not disturb the precise and delicate movements of the micromanipulator. A stiff and heavy cable also causes strain for the connections of the head stage circuit and the solder joints.

The signal ground is connected to the ground reference electrode (chloridized silver electrode) in the recording chamber.

The pipettes used in this work are ready-made commercially available FemtotipII pipettes made by Brinkmann Instruments Inc. (Westbury NY, USA) The self- pulled patch pipettes used for comparison are from Harvard Apparatus Limited (0.86/1.5mm diameter, borosilicate glass, Holliston MA, USA) and made using Kopf Model 750 needle/pipette puller (Tujunga CA, USA).

IV.TESTING

For test measurements, the head stage was mounted onto a Narishige MV-3 hydraulic 3-D micromanipulator.

The other devices used in the test measurements were: Grass Instruments S48 stimulator, Tektronix TDS1002

oscilloscope connected to a computer via RS-232 cable and used with Tektronix Wavestar software, Olympus CK-40 microscope, TMC anti-vibration table and a self made Faraday cage. The cells used in recordings were cultured human neuroblastoma cells (SH-SY5Y).

Based on the test measurements made with the List EPC-7 patch clamp device and using FemtotipII pipettes, the expected changes in the recorded signal were as follows:

When the pipette is not in the medium there are no pulses in the signal because the connection via ions in the solution does not exist. When the pipette is placed in the medium, the ground electrode and the measurement electrode are connected via ions in the solution and the pulses appear.

When the pipette is moved close to the cell and it touches the cell membrane, the pulses become smaller in amplitude and the signal noisier. The actual Giga-seal (pulses vanish totally) are very rare with FemtotipII pipettes, though with self-pulled patch pipettes seals are obtained. When the tip of the pipette breaks, the signal becomes larger, because the resistance of the pipette decreases.

In the test measurements of the contact detection device, the stimulus pulse sizes of 1 mV, 5 mV and 10 mV (final scaled pulse) were used. The most common changes in the signal are presented in this paper using 5 mV stimulus pulse (Fig. 3 - Fig. 5).

In Fig. 3 - Fig. 5, the presented signals are: the upper signal is the measured current output of the contact detection device and the lower signal is a non-scaled stimulus pulse recorded directly from the stimulator output.

The setting for the stimulus pulse scaling used in the presented measurements was 0.01 and the smallest amplification of the output amplifier, gain of 2, was used.

The measured data was modified so that each current measurement signal has an offset of 300mV and each stimulus -300mV. Moreover, with the Wavestar software, the stimulus pulse and the current monitor pulse cannot be recorded at the same time but in turns, and therefore the measurement data have been modified so that the pulses in the stimulus and current monitor signals start at the same time, as they appear on the oscilloscope screen.

The size of the pulse is not constant with the used FemtotipII injection pipettes, but varies a little between different pipettes. The filling of the pipettes was difficult and small air bubbles caused problems in measurements, sometimes reducing the signal. Only the most typical signal sizes and changes are presented in Fig. 3 - Fig. 5.

Fig. 3 illustrates how the pulse appears when the pipette tip is moved into the solution of the recording chamber. Fig.

4 shows the changes in the pulse (approximately 50%

smaller pulses) when the pipette touches the cell membrane.

Fig. 5 shows the change in the signal (approximately 50%

increase in the pulses) when the tip of the pipette was chipped. The breakage depicted in Fig. 5 was barely visible with a microscope. When the pipette breaks more the pulses become clearly larger.

0 50 100 150 200 250 300 350 400

-400 -200 0 200 400 600

A)

Time/ms

Voltage/mV

0 50 100 150 200 250 300 350 400

-400 -200 0 200 400 600

B)

Time/ms

Voltage/mV

Fig 3: The change in the signal when the pipette is moved into the recording chamber (5mV stimulus pulse). A) The pipette is in the air, B)

The pipette is moved down to the conducting recording solution.

0 50 100 150 200 250 300 350 400

-400 -200 0 200 400 600

A)

Time/ms

Voltage/mV

0 50 100 150 200 250 300 350 400

-400 -200 0 200 400 600

B)

Time/ms

Voltage/mV

0 50 100 150 200 250 300 350 400

-400 -200 0 200 400 600

C)

Time/ms

Voltage/mV

Fig 4: Changes in the signal when the pipette touches the cell membrane (5mV stimulus pulse). A) Pipette in solution, B) Pipette touches the cell membrane C) Pipette pulled away from the cell. This was the most typical

change in the signal when pipette was in contact with a cell.

0 50 100 150 200 250 300 350 400 -500

0 500 1000

A)

Time/ms

Voltage/mV

0 50 100 150 200 250 300 350 400

-500 0 500 1000

B)

Time/ms

Voltage/mV

Fig 5: Breaking of the pipette (stimulus pulse 5mV). A) Pipette intact, B) Signal when the tip of the pipette breaks.

Usually in patch clamp recordings, it is recommended to make only one recording with one patch clamp pipette because the tip of the pipette gets dirty and the formation of a proper Giga-seal becomes more difficult. With FemtotipII pipettes we were able to make many contact measurements with the same pipette and up to 15 contacts were recorded with one pipette.

V.DISCUSSION

The measurement results were quite similar between the new contact detection device and the tested EPC-7 patch clamp device. The desired changes were detected with the contact detection device and the proven repeatability of the contact measurement using the one and same pipette was an important result.

There were some differences in the changes of the pulse between the patch clamp device and the contact detection device, when making a contact with the cell. The changes in the pulse seemed to be “slower” and pulses more “flat” with contact detection device and there was more of the rapid random noise in the measurements with the patch clamp device. This might be due the different operational amplifiers used in the devices (the type of the operational amplifier of the patch clamp device is unknown) and different filtering used. The filtering of the contact detection device was done with the 1 kHz low pass filter and the filtering used in the measurements with the EPC-7 patch clamp device was 3 kHz. Also the changes in the pulse size seemed bigger in the measurements with the contact detection device. A less noisy signal is better for microinjection and for detecting the changes in the pulses when the pipette is in contact with cell.

Current monitor signal of the contact detection device showed small capacitive peaks, especially with 5mV signal, that were probably caused by the input capacitance of the

current measurement circuit and can also be a characteristic of the measurement operational amplifier. The capacitive peaks were actually expected to appear, because the capacitive compensation circuit that exists in the patch clamp amplifier was not built for the contact detection circuit. These small peaks should not be a problem in detecting changes in the microinjection pipette, as the peaks are known. Also more “rounded” (not as fast rising) pulses can be used to avoid these capacitive peaks.

VI.CONCLUSIONS AND FUTURE WORK

A miniaturized contact detection device which can be integrated with a microinjection manipulator has here been described. The contact detection device is able to detect the contact between a cell and a microinjection pipette, as well as the breakage and clogging of the pipette. This will make it possible to develop a fully automatic microinjection system for intracellular injection of single adherent cells.

Future work will include the implementation of the contact detection device to the microinjection micromanipulator and addition of the offset compensation function. Combining the whole system with visual detection system will also be research topics of the future.

ACKNOWLEDGMENT

The authors would like to thank professors H. Koivo and K. Saksela for their support in the research work.

R

[1] Single-Channel Recording, Editors: Bert Sakmann and Erwin Neher, 2nd edition, August 1995.

[2] The Axon Guide for Electrophysiology & Biophysics Laboratory Techniques, Axon Instruments Inc., 1993.

[3] F.J. Sigworth, EPC-7 User’s Manual, List-Electronic, Germany [4] http://www.brinkmann.com, Femtotip II pipettes

[5] Linne M.-L., An Electrophysiological Patch-Clamp System, M.Sc.E.E. Thesis, Tampere University of Technology, Department of Electrical Engineering, 1993.

[6] Practical Electrophysiological Methods: A Guide for In Vitro Studies in Vertebrae Neurobiology, Editors: Helmut Kettenmann and Rosemarie Grantyn, Wiley-Liss Inc., New York, 1992 [7] Daniel Johnston and Samuel Miao-Sin Wu, Foundations of

Cellular Neurophysiology, 1997

[8] Myllymaa S., Noise in Electrophysiological Patch-Clamp Recordings, M.Sc. Thesis, Tampere University of Technology, Department of Electrical Engineering, 2000.

[9] Dempster J., Computer Analysis of Electrophysiological Signal, Academic Press, London, 1993.