Viigipuu.qxp

ATLA 32, 417–423, 2004 Twenty-first Scandinavian Cell Toxicology Workshop 417
Microinjection of Living Adherent Cells by Using a Institute of Automation and Control, Tampere University of Technology, Tampere, Finland Summary — Testing in vitro is an alternative to animal experimentation. The capillary pressure microinjec-
tion technique is a supporting technology for efficient in vitro testing. The main benefit of the technique is the possibility of injecting large molecules into a single living cell. The ultimate goal of the research dis- cussed in this paper is to increase the cell survival rate in capillary pressure microinjection. A method to reli- ably evaluate cell survival rate is therefore needed. A three-phase evaluation process is presented in this paper. The first phase determines the success rate of the injection capillary to penetrate the cell membrane.
The second phase studies the success rate of delivering the injection substance inside the cell, while the third phase studies cell survival after the microinjection. In addition to the three-phase evaluation process, this paper describes the initial results of penetration and injection tests performed by using a semi-auto- matic capillary pressure microinjection system developed by the research group. Three adherent cell lines, namely, retinal pigment epithelial cells, MCF-7 human breast cancer cells and SH-SY5Y neuroblastoma cells, were used in the experiments. The results of the penetration tests show that the average success rate of penetrating the cell membrane using the micromanipulator was 87%. The goal of the injection tests was to demonstrate the successful microinjection of living cells and to study the injection success rate.
Fluorescein dextran was injected into MCF-7 cells, and preliminary results showed an injection success rate of 49%. In the survival tests, the neuronal cells were microinjected with KCl. During long-term observation after the microinjection, the microinjected cells first decreased their adhesion to the plate, but later adhered to the bottom of the plate and even grew some dendrites. In the next phase of the study, more tests will be performed in order to obtain a statistically reliable value for the survival rate. Key words: adherent cells, capillary pressure microinjection, cell curvival rate, injection success rate,
micromanipulator, penetration success rate, semi-automatic microinjection. Address for correspondence: K. Viigipuu, Institute of Automation and Control, Tampere University of
Technology, P.O. Box 692, 33101 Tampere, Finland. microinjection has been the introduction of com- Living cells are increasingly used in drug discovery, puter control in the micromanipulation and functional genomics, toxicology and many other microinjection processes, as well as the control and fields. One of the reasons is the trend toward reduc- standardisation of experimental conditions such as ing the need for laboratory animals in experiments.
the preparation of the cells and the reproducible Therefore, there is an increasing need for alterna- production of injection microcapillaries (1, 2) tive methods to animal experimentation.
The best injection rates — a successful injection Capillary pressure microinjection (CPM) is one of of a substance into a cell — are said to be as high as the supporting technologies for in vitro testing. The 70–80% (3), while the average cell survival rates main benefit of this technique is the possibility of achieved with the CPM technique are around 50% injecting macromolecules into a single living cell.
(4). However, as far as we know, the success rates The technique is already used in many application areas. Microinjection experiments are performed in areas such as cytology, physiology, genetic engi- The limited number of cells that can be injected neering, molecular biology, virology, tumour biol- in a certain time is a disadvantage of the tech- ogy, developmental biology, pharmacology and nique (5). Thus, there is a demand for automation to increase the injection rate. Another aspect to be Substances are typically injected into a cell to considered is the mechanical damage caused to manipulate and monitor the fundamental biochem- the cell when penetrating the membrane. The istry of the specific living cell. Cellular organelles, speed must be sufficiently high to penetrate the proteins, enzymes, antibodies, genes, metabolites, cell successfully, without damaging the cell. The ions, DNA and RNA, and various markers, for velocity at which the tip of the microcapillary enters the cell can be in the order of 700µm/s (6).
This paper provides a systematic approach for Dulbecco’s modified Eagle’s medium (DMEM) sup- evaluation of the CPM technique, which includes plemented with 10% fetal bovine serum (FBS), three phases for the determination of success rates 5mM glutamine, and an antibiotic-antimycotic solu- for penetration, injection and cell survival. The tion (AB-AM; 100 units/ml penicillin, 100g/ml survival rate related to the technique itself is streptomycin, 250ng/ml amphotericin B). MCF-7 needed as a reference: when a potentially harmful cells were cultured in phenol red-free DMEM with substance is injected into a cell, the probability F12 (DMEM/F12), supplemented with 5% dextran- that the cell dies due to the procedure itself must coated, charcoal-stripped treated, FBS, plus peni- be known and taken into account. In order to reli- cillin-streptomycin, 10ng/ml insulin, and 1nM ably determine cell survival rates for various 17β-oestradiol. SH-SY5Y cells were grown in 1:1 microinjections, the injection success and penetra- minimum essential medium/Ham’s F-12 medium tion success rates must be known. The injection (Kaighn’s modification), supplemented with AB- success rate is important when the effect of the AM, 10% FBS, 2mM L-glutamine and 0.1mM non- injected substance is considered, since the proba- essential amino acids. For transportation and tests bility with which the substance is delivered into outside the incubator, the medium was replaced the cell must be known. The penetration success with Leibowitz L-15 medium (Sigma-Aldrich, rate is important when the effect of the procedure Munich, Germany; L5520), which requires no pH itself is considered, as the probability with which adjustment with carbon dioxide, supplemented with the microcapillary penetrates the cell membrane 5% dextran-coated, charcoal-stripped treated, FBS, must be known, in order to evaluate the adverse penicillin-streptomycin, 10ng/ml insulin, 1nM 17β- Microcapillaries
Microcapillaries (Femtotip II; Eppendorf, Hamburg, Germany), with an outer diameter of 0.7µm and an inner diameter of 0.5µm (± 0.1µm), were used. Experiments were performed on adherent cell lines.
Retinal pigment epithelial (RPE) cells were used in the experiments, mainly because of their availabil- Injection substances
ity and favourable characteristics. These cells are not too sensitive to environmental changes and KCl and fluorescein dextran (fluorescein isothio- mechanical effects. RPE cells were originally iso- cyanate) were used as injection substances.
lated from fresh pig eyes obtained from a local slaughterhouse. The cells were passaged for a max- imum of four times before they were used for the Semi-automatic microinjection system
tests. At this point, RPE cells still resemble their in vivo counterparts and are not dedifferentiated. RPE The system consists of a micromanipulator, a pres- cells were grown in 92mm × 17mm Petri dishes sure injector, a vision system, a cell incubation sys- (Nunc, Roskilde, Denmark). A second adherent cell tem and software. The system uses the CPM line used was the human breast cancer cell line, MCF-7, provided by Professor Pirkko Härkönen (University of Turku, Finland). SH-SY5Y human neuroblastoma cells were used to demonstrate that Micromanipulator
the CPM method also works with a more sensitive adherent cell line. The SH-SY5Y cells were obtained The MANiPEN micromanipulator developed by the from the American Type Culture Collection research group was used for the high-precision posi- (Rockville, MD, USA; cat no. CLR-2266). They grew tioning of the microcapillary. Positioning tasks as a mixture of floating and adherent cells and include: a) locating the tip of the microcapillary in exhibited fine cell processes (neurites). Both MCF-7 the field of view of the microscope; b) moving the tip and SH-SY5Y cells were grown in 12-well multi- toward the cells; and c) penetrating the cell mem- brane. The MANiPEN micromanipulator is a joy- stick-controlled, semi-automatic device. After the microcapillary is positioned on a chosen cell by using the joystick, the microinjection is triggered by pressing a button on the joystick. The micromanip- All cell culture media and supplements were from ulator is fixed to a stand at such an angle that the Gibco Invitrogen Life Sciences (Paisley, UK). The tip of the microcapillary can be located in a well of maintenance medium for RPE cells contained Microinjection of cells by using a semi-automatic system 419 Pressure injector
1. Lower the microcapillary by using the joystick- controlled micromanipulator until the tip of the The micromanipulator is connected with a pres- sure injector (MPPI-2; Applied Scientific Instruments, Eugene, OR, USA). The pressure 2. Press the injection button on the joystick for injector has one channel, and its pressure range is microinjection. The injection pressure used for 0–700kPa. Settings for the applied pressure, such neuronal cells was around 0–40kPa in the as the amount and duration of the pressure pulse, experiments described in this paper. This corre- and the amount of balance pressure, are changed manually by using dials. The pressure pulse is 3. Continue injecting cells in the chosen field of Vision system
4. Move to the next field of view and repeat the microinjection procedure, or start cell monitor- The vision system includes an inverted optical microscope, a charge coupled device camera, a frame grabber and vision algorithms. The vision system makes it possible to record the event of microinjection, as well as long-term monitoring of The ultimate goal of the research is to increase the cell survival rate in CPM procedures, so a method for evaluating cell survival rate is needed. The first Cell incubation system
steps in the evaluation of the cell survival rate are evaluations of penetration and injection rates. The The system includes an incubation chamber (Chip- following sections will propose methods for a reli- Man Technologies, Tampere, Finland), which mim- able evaluation of the penetration success, injection ics the environment of the cells in vivo by success and cell survival rates. Furthermore, initial results for the semi-automatic CPM system devel- oped by the group are given. In the course of the experiments described in this paper, the site of Software
microinjections (nucleus/cytoplasm) has not been the focus of interest. The injections were made in The system consists of two computers, one to con- the highest part of the cell, near the nuclear area.
trol the micromanipulator, and one to control the Penetration tests
Microinjection procedure
For the successful microinjection of living cells, pre- cise penetration of the cell membrane is needed.
To start a microinjection procedure, the cell The cell membrane has elastic properties and could medium is changed to Leibowitz L-15, as the cells deform under the microcapillary, preventing pene- are open to the air during the experiment. It is tration. In this study, penetration tests were per- preferable to choose an area where the cells are well formed to evaluate the success rate of penetration.
spread, as this makes it easier to inject a single cell The visual validation of membrane penetration is at a time. In cases where special dyes are not used, challenging, due to limited resolution of an optical the microinjection area should be marked to facili- microscope and the high speed of the penetration tate finding the same area for subsequent analysis movement. Therefore, the penetration rate was determined by injecting an excessive amount of During the experiment, the magnification is substance into a cell with a high pressure pulse, changed several times. It is recommended to first resulting in an eruption of the cell membrane. The focus on cells with the lowest magnification, for eruption was easily visualised by using the vision example, 40× the magnification. The microcapil- lary can be centred roughly in the field of view by The tests were performed on both RPE (Figure 1) using the naked eye. When the tip is in the centre of and neuronal cells by using the semi-automatic the field of view, the magnification can be microinjection system. The amount of pressure increased. For the microinjection of adherent cells, applied was gradually increased until a visible erup- tion of the membrane occurred. Pressure values of 7–9)To summarise, the suggested microinjection pro- 280–365kPa were used in the experiments. A total of 55 cells were penetrated, of which 48 were suc- Figure 1: Illustration of the penetration test A retinal pigmented epithelial cell is shown a) before and b) after penetration.
Scale bar = 20µm. cessfully penetrated. This resulted in an average of 49%. The primary reason for the relatively low success rate of 87% and proved that the tip of the injection rate was clogging of the capillary.
microcapillary efficiently penetrated the cell mem- Automatic detection of a clogged microcapillary is brane when the MANiPEN micromanipulator was therefore one of the main issues for future study.
Survival tests
Injection tests
The goals of the survival tests were to prove that Another important parameter in microinjection is the CPM method itself does not harm the cells, successful delivery of the injection substance into and to study the survival rate of the microinjected the chosen single cell. The main challenges here are the injection of a correct amount of the injection The survival tests were performed on neuronal substance, the detection of the contact between the cells by using the semi-automatic microinjection cell and the microcapillary, and the tendency of the system. The amount and the injected substance microcapillary to clog. The goals of the injection itself have to be harmless to the cell, so minute tests were to demonstrate successful injections of amounts of KCl (with an injection pressure of living cells and to study injection success rate. The 40kPa) were microinjected. KCl is considered a injection tests were performed with MCF-7 cells by safe substance, due to the fact that it is found using the semi-automatic microinjection system.
inside the cell itself, and does not change the The amount of pressure applied was gradually low- potential of the cell (as, for instance, NaCl would ered to find its proper value. For cells which are do, i.e. resulting in an eruption of the cell). To 20µm in diameter, the injected volume should be in encourage cell survival, temperature and pH the range of several hundreds of femtolitres. It is were maintained by the incubation system. The difficult to recognise the injection event without a temperature was set at 37°C, and the pH at 7.4.
special dye. Therefore, in the injection tests, fluo- The injected cells were followed by the vision sys- rescein dextran was microinjected into MCF-7 cells tem for several hours after the injection. The (Figure 2). In addition to the inverted microscope, a most visible effect, which proved the survival of microscope with a fluorescence option was used to the neuronal cell, happened during the first 6 hours (Figure 3). Microinjected cells changed A total of 82 cells were injected, of which 40 were their shapes to become spherical, adhered back successfully injected, giving an average success rate to the bottom of the well, and even grew some Microinjection of cells by using a semi-automatic system 421 Figure 2: Illustration of the injection test A human breast cancer cell line (MCF-7) is shown a) before and b) 15 minutes after fluorescein dextran microinjection.
Scale bar = 20µm. dendrites. In future study, more tests will be per- clogging of the microcapillary and the lack of a reli- formed to determine statistically reliable sur- able detection method for the contact between the cell and the capillary. A higher survival rate could be gained by having automatic detection of the cap- illary clogging and contact between the capillary and the cell, and this will be one of the major issues Because of the elastic property of the cell mem- In general, clogging of the microcapillary is one of brane, penetrating the membrane without damag- the main problems connected with the CPM tech- ing the cell can be challenging (8, 10). When nique. A clogged microcapillary reduces the flow penetration occurs, the membrane must snap through the tip, or blocks the flow of an injection back over the tip of the microcapillary. With a suf- substance. The result is either a loss of repro- ficiently high acceleration, penetration will occur.
ducibility or an unsuccessful injection. The micro- However, too high an acceleration tends to cause capillary becomes clogged, for example, by particles the tip to vibrate, which can damage the mem- flowing in the cell culture medium or by a part of brane. In particular, after-vibrations must be the cell membrane. In the experiments described in eliminated or minimised (11). The results of the this paper, the best results for unclogging the penetration tests performed in this study showed microcapillary were obtained by applying a high that the average success rate of penetrating the pressure or by raising the microcapillary rapidly out cell membrane was 87%. This high penetration rate proves that the microinjection system pro- The goal of the survival tests was to evaluate vided a correct penetration speed and that the the overall success of the microinjection. A crite- elasticity of the cell membrane did not raise prob- rion for a living cell in good condition is that it lems. However, the damage caused by mechanical continues its active life after the microinjection. If penetration was not studied in the penetration the cell changes its morphology in 5–10 minutes, tests. Evaluation of cell survival rate could also be it is considered, if not dead, at least to be suffer- used for improving the membrane penetration ing significantly as a result of the microinjection.
On the other hand, as seen in the survival tests, Although the penetration success rate is high, at cell recovery might take several hours. Therefore, 87%, the fluorescein dextran injections showed a the vision system should be able to provide infor- relatively low microinjection success rate of 49%.
mation about the cell culture for several hours The primary reasons for a low success rate are the Figure 3: Illustration of the survival test A human neuroblastoma cell (SH-SY5Y) is shown a) before and b) 6 hours after KCl microinjection.
Scale bar = 20µm. A cell inevitably responds to the microinjection partners have expertise in different areas. While event. Even when the microinjection has been suc- the Medical School at the University of Tampere cessful, there might be changes in shape or adhe- focused on the cell cultures and the Technical sion. Therefore, in microinjection experiments Research Centre of Finland on the machine vision where the injection substance itself is expected to and incubation systems, the tasks of the Tampere have an influence on the cell, both the injection sub- University of Technology included the improve- stance and microinjection can affect the cell’s ment of the injection methods and the development behaviour. In this case, it is especially important to of a pen-shaped micromanipulator, MANiPEN.
have a proper visualisation tool, which provides This paper proposes a three-phase evaluation process for capillary pressure microinjection sys- tems, and evaluated a semi-automatic system devel- Adams, S.R., Bacskai, B.J., Taylor, S.S., Tsen, R., oped by the research group. The unique shape and Lemos, J. & Podesta, E.J. (1993). Optical Probes for small size of the MANiPEN manipulator will per- Cyclic AMP: Fluorescent Probes for Biological mit the simultaneous use of several micromanipu- Activity in Living Cells (ed. W.T. Mason), pp.
lators in the future. Each micromanipulator could 166–172. San Diego, CA, USA: Academic Press.
Ansorge, W. (1988). Performance of an automated have its own task, such as microinjection of sub- system for capillary microinjection into living cells.
stances, electrophysiological recording of a cell, or Journal of Biochemical and Biophysical Methods 16,
Ansorge, W. (1982). Improved system for micro- pipette microinjection into living cells. Experimental Cell Research 140, 31–37.
Davia, B.R., Yannariello-Brown, J., Prokopishyn, N.L., Luo, Z., Smith, M.R. & Wang, J. (2000). Glass This paper is part of the work performed in a proj- needle-mediated microinjection of macromolecules ect SOLOMANDA (Manipulation, Detection and and transgenes into primary human blood stem/pro- Analysis of Single Living Cells), funded by the genitor cells. Blood 95, 437–444.
National Technology Agency of Finland (TEKES) Pepperkok, R. & Saffrich, R. (2001). Microinjection in 2000–2002, and supported by the Academy of and Detection of Probes in Cells. Website http://www.
Finland. The aim of the project was to develop ansorge-group.embl.de/inject2003/. EMBO Practical methods and devices which facilitate the replace- Laffaffian, I. & Hallett, M.B. (1998). Lipid-assisted ment of laboratory animals by human cell lines.
microinjection: introducing material into the cytosol New devices and systems that are capable of treat- and membranes of small cells. Biophysical Journal ing individual cells were developed. The project 75, 2558–2563.
Microinjection of cells by using a semi-automatic system 423 Brinkman Instruments Inc. (1994). Sample Prepar- microfilament organization in living nonmuscle ation for Microinjection. Website http://www.
cells by microinjection of plasma vitamin D-binding brinkman. com/ECETappl2.asp. Dr Rainer Pepperkok, protein or DNase I. Proceedings of the National Université de Genéve, Département de Biologie Academy of Sciences, USA 87, 5774–5478.
Cellulaire Sciences (Accessed 4.9.03).
10. Kishimoto, T. (1986). Microinjection and cytoplas- Jaffe, L. (1999). Rindy Jaffe Microinjection Manual.
mic transfer in starfish oocytes. Methods in Cell Website http://egg.edu/panda/injection. Mark Terasaki Biology 27, 379–394.
Lab, Department of Physiology, University of 11. Brown, K.T. & Flaming, D.G. (1995). Methods in the Connecticut Health Center (Accessed 4.9.03).
Neurosciences, Vol. 9, Advanced Micropipette Tech- Sanger, J.M., Dabiri, G., Mittal, B., Kowalski, M.A., niques for Cell Physiology (ed. A.D. Smith), 296pp.
Haddad, J.G. & Sanger, J.W. (1990). Disruption of Chichester, Guildford and King’s Lynn, UK: Biddles.

Source: http://ae.tut.fi/research/MST/files/our/abstracts/2003_viigipuu2.pdf