An automated system for intracellular and intranuclear injection

https://doi.org/10.1016/j.jneumeth.2007.11.028Get rights and content

Abstract

The Xenopus oocyte expression system has played an important role in the study of cellular proteins, particularly in the field of membrane physiology; expression of transporters and ion channels has significantly advanced our knowledge of these membrane proteins and the rapid and easy expression of mutants has been crucial in many structure–function studies. Xenopus oocytes are an expression system in many ligand-binding assays and in functional screening for ion channel modulators. Several commercially available automated technologies use this system, generating a demand for large numbers of oocytes injected with ion channel genes. Injection of oocytes with genetic material is generally carried out manually. Here we describe an automated system capable of injecting up to 600 oocytes per hour. Oocytes are contained in microplates with conical wells, a simple calibration procedure by the operator is required and pipette filling and oocyte injection are carried out automatically. Following intracellular injection of mRNA coding for ligand-gated ion channels close to 100% of oocytes tested positive for expression, and intranuclear injection of cDNA gave a rate of expression >50%. Moreover, we demonstrate that this method can also be successfully applied to inject zebrafish embryos and could be extended to other cell types.

Introduction

Investigating the function of a particular cellular protein such as a membrane transporter, ion channel, metabotropic receptor or kinase in a native cell is often difficult due to the large number of interfering factors, such as similar proteins, present in the native cell. The identification and cloning of genes coding for many of these proteins has been an important step that has enabled researchers to express the resulting proteins in host cell expression systems. One of the principal advantages of expressing cloned genes in a host cell system is that it enables the study of the resulting proteins in relative isolation. Thus, host cell systems are chosen not only for their ability to express the genes of interest, but also for their lack of endogenous proteins which could make studies of the expressed protein difficult.

Oocytes from the south-African clawed frog Xenopus laevis were first used as an expression system for cellular proteins in 1971, where it was demonstrated that they were able to synthesise haemoglobin following intracellular injection of the corresponding mRNA (Gurdon et al., 1971). Subsequently it was demonstrated that they were capable of expressing a number of different proteins such as globin, interferon and viral proteins (Gurdon et al., 1973, Gurdon et al., 1974, Laskey and Gurdon, 1974, Laskey et al., 1972, Woodland et al., 1974) and in 1982 it was reported that, following injection of mRNA coding for acetylcholine- and GABA-sensitive receptors, X. laevis oocytes expressed fully functional channels with properties similar to those in native cells (Miledi et al., 1982a, Miledi et al., 1982b, Miledi et al., 1983). Although early studies involved the injection of mRNA, Wickens and colleagues demonstrated that intranuclear injection of cDNA in X. laevis oocytes also resulted in protein expression (Wickens et al., 1980). It was sufficient that the gene of interest was in the correct orientation between a eukaryotic promoter such as SV40 and the SV40 transcription termination sequence, and transcription, capping, polyadenylation and export of mRNA from the nucleus were carried out by the oocyte. Moreover, cDNAs containing introns were correctly spliced and expressed by the oocyte (Bertrand et al., 1991). Although the injection of cDNA into the nucleus of the oocyte requires more precision than cytoplasmic injection, this technique eliminates the need for in vitro transcription of cRNA. An additional advantage is that cDNA is highly stable and does not require the careful storage and handling necessary for cRNA.

As an expression system for cellular proteins, Xenopus oocytes have a number of advantages over mammalian cell lines. The frogs cost relatively little and are easily maintained and reproduce in captivity. Oocytes can be harvested many times from the same frog by partial ovarectomy and can survive outside the animal for up to a month if conserved at 4 °C and due to their large size; handling of the oocytes is easy. Xenopus oocytes have the necessary enzymes for the expression of a wide range of mammalian proteins and the oocytes are compatible with many bioassays, making them particularly useful in scanning mutagenesis studies that require the expression and functional testing of large series’ of mutant proteins, as well as the pharmacogenomic application of screening existing drugs or drug candidates across proteins resulting from genetic variations, such as single nucleotide polymorphisms. Despite its numerous advantages, the Xenopus oocyte expression system also has several disadvantages. The most important of these is whether ion channels expressed in amphibian cells are assembled and behave in an identical fashion to those expressed in mammalian cells. Single channel experiments comparing the properties of rat α3β4 nAChRs expressed in oocytes with a stably expressing mammalian cell line (L-α3β4) reported that α3β4 receptors expressed in the mammalian cell line more closely resembled native channels from rat superior cervical ganglion than those in oocytes (Lewis et al., 1997, Sivilotti et al., 1997). It is not clear if these differences are due to a different stoichiometry of the channel subunits in oocytes or due to different posttranslational modification or functional modulation, such as phosphorylation, by the amphibian cell machinery. Additional challenges are posed by their large size, which gives a relatively slow fluid exchange time around the oocyte compared to a mammalian cell, this can be a problem for recording fast-desensitizing ligand-gated channels in oocytes. During long incubations with lipophilic compounds these can accumulate in the oocyte and slowly leak out after the compound has been removed preventing washout. Another disadvantage is that presently, each oocyte must be individually injected with genetic material, which is slower than the simultaneous transfection of large numbers of mammalian cells.

One area of research where the Xenopus oocyte expression system is widely used is in the study of ion channels. A main advantage is that most cDNAs and cRNAs are readily expressed without the need to develop a cell line, whereas there is often a significant delay from the cloning of a new receptor until it can be expressed in a mammalian cell line. Studying the function of an ion channel in a native cell often requires substitution of the ions present in the physiological recording solutions with impermeant ions to suppress endogenous currents which can mask the current of interest. This can result in significant effects on the properties of the ion channel being studied and is often not sufficient to remove the interference of other channels that are permeable to the same ions. In some cases interferance from endogenous channels and receptors can be minimised by having selective blockers of these ion channels in the recording solution; however, this is dependent on the availability of such compounds and for many ion channels selective blockers are not available. Xenopus oocytes contain few endogenous ion channels and transporters that can interfere with measurements of the protein of interest. Following injection of genetic material into the oocyte, proteins are expressed after 1–2 days and functional studies of ion channels and receptors are easily performed using conventional two-electrode voltage clamp techniques. The only consideration, is that due to the large size of the oocyte, the amplifier must be able to deliver sufficient current to clamp the transmembrane voltage in the voltage clamp configuration, which is now achieved with commercially available amplifiers.

The X. laevis expression system is used routinely for screening applications in drug discovery using expressed ion channel targets. Presently, there are several fully and semi-automated oocyte electrophysiology recording systems available commercially: Roboocyte (http://www.multichannelsystems.com) is fully automated and records from oocytes in a 96-well microplate; OpusXpress 6000A (http://www.moleculardevices.com) can record up to eight oocytes in parallel and is designed primarily for drug screening applications; and ScreenTool (http://www.npielectronic.com) is also designed for screening of compounds and combines electrophysiological recording and automated liquid handling. These machines require the injection of large numbers of oocytes with genetic material coding for target proteins, which can be an arduous task if carried out manually. The principal aim of this work was to design an injection system to automate this procedure in Xenopus oocytes and which could be extended to other cell types, such as the injection of genetic material or sperm in embryos from Zebrafish, Drosophila and ultimately mammals. Embryos of zebrafish and Drosophila contain a large number of identical genes and processes to those found in higher mammals making them widely used in studies of development and in disease. In addition genetically altered zebrafish are now used in a number of toxicological and drug screening assays, generating a need for new genetic mutants. The challenge of extending this method to other cell types will be in part dictated by the cell size, zebrafish embryos have a diameter of around 1.2 mm, which is similar to that of Xenopus oocytes, Drosophila embryos are 500 μm in length, whereas a human egg has a diameter of 100 μm. Thus we will commence by the largest cell types.

This work describes the design and validation of a fully automated system suitable for intracellular and intranuclear injection of cDNA and cRNA in Xenopus oocytes and zebrafish embryos.

Section snippets

Preparation of Xenopus oocytes

X. laevis were sacrificed according to Swiss National guidelines. Ovaries were removed and placed in Barth's medium, which contained 88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO3, 1 mM Hepes, 0.8 mM MgSO4, 0.33 mM Ca(NO3)2, 0.4 mM CaCl2, kanamycin 20 units/ml and penicillin-streptomycin 100 units/ml adjusted to pH 7.4 with NaOH. Ovaries were kept at 4 °C and the bathing solution was changed weekly. Under these conditions oocytes remained viable for up to 4 weeks. Prior to injection, oocytes were defolliculated by

cRNA injection

As a first validation of the injector we started with the injection protocol having the least stringent injection conditions. Many researchers using the X. laevis expression system inject cRNA coding for a protein of interest. Although cRNA requires more careful storage and handling than cDNA, it can be injected anywhere in the cytoplasm of the oocyte and thus requires less precision than for intranuclear injection of cDNA. Moreover, cRNA injection generally gives a higher percentage of oocytes

Discussion

In this work we demonstrate the feasibility of automated intranuclear and intracellular injection of Xenopus oocytes and zebrafish embryos. There are several points to consider for an automatic versus manual injection, such as the number of oocytes to be injected, the number oocytes per hour that can be injected with the machine versus manually, the success rate of the injection and the relative cost of the equipment. Techniques such as expression cloning and in-cell NMR spectroscopy and

Applications

For the present validation of the automatic injector we have concentrated mainly on the injection of cDNA and cRNA coding for ligand-gated ion channels; however, injection of cRNA and cDNA coding for a variety of membrane and intracellular proteins, such as voltage-gated ion channels, metabotropic receptors, 7TM receptors, steroid receptors, transporters, aquaporins, and whole RNA injection is possible. Moreover, the simultaneous injection of several cDNAs can be used to investigate the effects

Acknowledgments

Thanks to Franck Bontems for the GFP clone. This work was supported by a Swiss National Science Foundation award to R.C.H.

References (31)

  • A.V. Gamarnik et al.

    Exploring RNA virus replication in Xenopus oocytes

    Methods Mol Biol

    (2006)
  • A.V. Gamarnik et al.

    Translation and replication of human rhinovirus type 14 and mengovirus in Xenopus oocytes

    J Virol

    (2000)
  • J.B. Gurdon et al.

    Use of frog eggs and oocytes for the study of messenger RNA and its translation in living cells

    Nature

    (1971)
  • C.B. Kimmel et al.

    Stages of embryonic development of the zebrafish

    Dev Dyn

    (1995)
  • R.A. Laskey et al.

    The translation of viral RNAs in frog oocytes

    Hamatol Bluttransfus

    (1974)
  • Cited by (58)

    • Robotic orientation control and enucleation of cells

      2023, Robotics for Cell Manipulation and Characterization
    • Toxicity testing of natural products using the zebrafish model system

      2022, Biomedical Product and Materials Evaluation: Standards and Ethics
    • Robotic and microfluidic systems for single cell injection

      2021, Micro and Nano Systems for Biophysical Studies of Cells and Small Organisms
    • Antagonizing α7 nicotinic receptors with methyllycaconitine (MLA) potentiates receptor activity and memory acquisition

      2019, Cellular Signalling
      Citation Excerpt :

      During testing the test room was dimly lit by a small lamp (25 W), located in the corner of the room. Oocytes were harvested from mature Xenopus laevis females and were prepared, injected with cDNA, and recorded using standard procedures [26,28]. On the day after dissociation, oocytes were injected with 0.2 ng of cDNA expression vector, coding for the human α7nAChR subunit, in their nucleus using an automatic injection system (RoboInject, Multi Channel Systems) [28].

    View all citing articles on Scopus
    View full text