Misexpression of genes in brain vesicles by in ovo electroporation


  • Harukazu Nakamura,

    1. Department of Molecular Neurobiology, Institute of Development, Aging and Cancer, Tohoku University, Seiryo-machi 4-1, Aoba-ku, Sendai 980-8575, Japan.
    Search for more papers by this author
    • *Author to whom all correspondence should be addressed.

  • Yuji Watanabe,

    1. Department of Molecular Neurobiology, Institute of Development, Aging and Cancer, Tohoku University, Seiryo-machi 4-1, Aoba-ku, Sendai 980-8575, Japan.
    Search for more papers by this author
  • Jun-ichi Funahashi

    1. Department of Molecular Neurobiology, Institute of Development, Aging and Cancer, Tohoku University, Seiryo-machi 4-1, Aoba-ku, Sendai 980-8575, Japan.
    Search for more papers by this author


Transfection to living chick embryos in ovo by electroporation has been recently developed. In this mini-review, misexpression in brain vesicles is introduced. To transfect, expression plasmid is inserted in the brain vesicle, and the square pulse of 25 V, 50 ms was charged five times. The translation product of the transfected gene is detected 2 h after electroporation, and reaches the peak at 24 h after electroporation. Transfection is so effective that this method is contributing greatly to the study of the molecular mechanisms of morphogenesis.


In the field of experimental embryology, chick embryos are good materials because it is easy to access the embryo and manipulate it. But these days, the importance of chick embryos as experimental material for embryology has decreased because: (i) it is very difficult to grow inbred strains; and (ii) the embryos stay in the uterus for the first 24 h of the embryonic period, which makes it difficult to produce transgenic chickens or to knock out certain genes. Recently, we improved the electroporation method to transfer certain genes to chick embryos in ovo ( Muramatsu et al. 1997 ; Funahashi et al. 1999 ). By this method, very rapid and effective expression of the introduced gene can be obtained. In this minireview, we briefly describe the method used for gene transfer into brain vesicles.

Procedure for in ovo electroporation

Plasmids Usually, cDNA was inserted into the expression vector, pMiwSV, which had RSV (Rous Sarcoma virus) enhancer and β-actin promoter ( Suemori et al. 1990; Wakamatsu et al. 1997 ). pRc/CMV (Invitrogen Co., Carlsbad, CA, USA), which had CMV (cytomegalovirus) enhancer, can also be used as an expression vector. Both work well as expression vectors in chick embryos.

Injection of DNA solution Embryos were incubated until they reached the desired stage. For transfection to the mesencephalon, stage 10 ( Hamburger & Hamilton 1951; 36 h incubation) embryos are usually used. A window of about 2 cm in diameter should be opened on the top of the egg after removing 4 mL of albumen. Injection of Indian ink underneath the embryo facilitated visualization of it ( Fig. 1C). The head region of an embryo was exposed by cutting the vitelline membrane with a microscalpel. To ensure that the injected DNA stays in the lumen of the neural tube, it is recommended to cut the most anterior part of the neural tube with a microscalpel. Then, DNA solution was injected with a micropipette into the lumen of the neural tube ( Fig. 1D). If the anterior tip of the central canal is closed, injected DNA solution comes out when one pulls the pipette out of the neural tube. To transfect the mesencephalic region, a micropipette was inserted from the metencephalon anteriorly ( Fig. 1D), and the DNA solution was injected. For this process, the micropipette was put in the hematocrit tube, and injection of the solution was controlled by mouth. Micropipettes for the injection of DNA into the embryo were made of glass capillary tubes 1 mm in diameter (GD-1; Narishige, Tokyo, Japan). About 0.1–0.2 μL of 1 μg/μL DNA solution in TE (Tris-EDTA) buffer or in phosphate-buffered saline (PBS) was injected.

Figure 1.

In ovo electroporation. A pair of electrodes held by a manipulator (A) was inserted from a window opened on the shell (B). The electrode was put on the vitelline membrane overlying the embryo (C), and a 25 V, 50 ms pulse was charged five times. Plasmid solution was injected into embryonic stage (E2; HH stage 10) chick neural tube (D) prior to pulse charge. Dimensions of the electrode are shown schematically in (E). Most of the electrode was insulated (black in figure) so that only the tip was exposed (white area).The cells in the neural tube look normal 1 h after electroporation (F). Deposits on the right-hand side of the neural tube are a complex of plasmids and the color substrate that was not removed by washing in dimethylformamide. Twenty-four hours after electroporation, the development of yolk sac plexus, vitelline veins and vitelline arteries was retarded in the area in contact with the electrodes (arrows in (G)). Bar, 2 mm (C), 50 μm (F), 4 mm (G).

Electroporation Electrodes (0.5 mm in diameter with an exposed length of 1 mm) set on a micromanipulator (MN-151; Narishige ( Fig. 1A,B)) were put on the vitelline membrane at a distance of 4 mm apart ( Fig. 1C), and a small volume of Hanks’ solution was placed between the electrodes. Then the square pulse (25 V, 50 ms) was charged five times. Pulses were generated every 1 s so that a pulse of 50 ms was followed by a 950 ms rest phase. Pulses were generated by Electroporator T820 and OptimizerTM (BTX, San Diego, CA, USA), or by CYU 21 (Tokiwa Science, Tsukushino-city, Fukuoka, Japan).

Resistance between the electrodes was dependent on the volume of Hanks’ solution, but the current through the embryo may be constant if the distance between the electrodes is fixed. After electroporation, the window was sealed with Scotch tape, and the embryos reincubated at 38°C.

As DNA is negatively charged, DNA moved toward the anode in the electric field so that the anode side of the tissue was transfected. In this sense, the neural tube is a very convenient tissue to use for applying in ovo electroporation. The DNA solution was easily injected into the canal of the neural tube, and stayed there. The cathode side served as a control so that we could compare the effects of transfection in the same embryo. The electric charge in our experimental conditions did not injure the cells of the neural tube ( Figs 1F,2E). Only places where the electrodes were directly attached were damaged so that vascularization was affected to some extent, which may be the main cause of embryonic death ( Fig. 1G). We found the above conditions optimum for transfection to the mesencephalon at around stage 10 for chick embryos. If the distance between the electrodes was shorter, a lower voltage can produce similar transfection efficiency. To do so, however, we had to put the electrodes near the embryo, which interfered with subsequent vascular development, and caused more embryonic death. Widening of the distance between the electrodes reduced transfection efficiency.

Figure 2.

Efficiency of electroporation. Efficiency of in ovo electroporation was assessed by injecting lacZ expression vector (pMiwZ), or by greenfluorescent protein (GFP) expression vector (pEGFP-N1) at stage 10. The lacZ signal was already recognizable 2 h after electroporation (A), and became strong 3 h after electroporation (B). Efficiency of the transfection of the introduced gene can be assessed in ovo by coelectroporation with GFP vector ((C); 9 h after electroporation). Twenty-four hours after electroporation (D,E), more than half of the cells expressed lacZ at the transfection zone. The expression was transient, but the lacZ signal was still strong 72 h after electroporation (F). The lacZ transfection exerted no morphological effects. Arrow in (D) indicates the section in (E). Bars: 200 μm (A–D,F), 50 μm (E).

Expression of introduced DNA

Expression of introduced genes was checked by pMiwZ, which encoded lacZ reporter. The lacZ translation product was detectable 2 h after electroporation ( Fig. 2A). More cells express lacZ 3 h after electroporation ( Fig. 2B). The lacZ signal reached peak intensity around 20 h after electroporation ( Fig. 2D; Momose et al. 1999 ). More than 50% of cells were transfected ( Fig. 2E) in the transfection zone. The En-2 translation product was also detected by 2 h after electroporation, and repression of Pax6 by introduced En-2 was already detectable by 3 h after electroporation in the diencephalic region ( Araki & Nakamura 1999).

The plasmid we used was not integrated into the host chromosome so that expression was transient. Expression level decreased from about 20 h after electroporation. As the lacZ translation product is rather stable, we detected the product 72 h after electroporation.

It was shown that coelectroporation of lacZ and GFP vector results in the overlapped patterns of the two genes ( Momose et al. 1999 ). Co-injection of greenfluorescent protein (GFP) expression plasmids with the DNA of interest made it possible to monitor in ovo the transfection efficiency under a fluorescence dissection microscope ( Fig. 2C). We used commercially available pEGFP-N1 (Clonetech; Palo Alto, CA, USA) for this purpose.

Application for tissue-specific introduction is described by Momose et al. (1999) . This technique is extensively applied to study the function of certain genes and the gene expression cascade in development ( Ogino & Yasuda 1998; Araki & Nakamura 1999; Funahashi et al. 1999 ; Momose et al. 1999 ; Okafuji et al. 1999 ; Takeuchi et al. 1999 ; Katahira et al. 2000 ).