|Home | About | Journals | Submit | Contact Us | Français|
The functional characterization of genes expressed during mammalian retinal development remains a significant challenge. Gene targeting to generate constitutive or conditional loss of function knockouts remains cost and labor intensive, as well as time consuming. Adding to these challenges, retina expressed genes may have essential roles outside the retina leading to unintended confounds when using a knockout approach. Furthermore, the ability to ectopically express a gene in a gain of function experiment can be extremely valuable when attempting to identify a role in cell fate specification and/or terminal differentiation.
We present a method for the rapid and efficient incorporation of DNA plasmids into the neonatal mouse retina by electroporation. The application of short electrical impulses above a certain field strength results in a transient increase in plasma membrane permeability, facilitating the transfer of material across the membrane 1,2,3,4. Groundbreaking work demonstrated that electroporation could be utilized as a method of gene transfer into mammalian cells by inducing the formation of hydrophilic plasma membrane pores allowing the passage of highly charged DNA through the lipid bilayer 5. Continuous technical development has resulted in the viability of electroporation as a method for in vivo gene transfer in multiple mouse tissues including the retina, the method for which is described herein 6, 7, 8, 9, 10.
DNA solution is injected into the subretinal space so that DNA is placed between the retinal pigmented epithelium and retina of the neonatal (P0) mouse and electrical pulses are applied using a tweezer electrode. The lateral placement of the eyes in the mouse allows for the easy orientation of the tweezer electrode to the necessary negative pole-DNA-retina-positive pole alignment. Extensive incorporation and expression of transferred genes can be identified by postnatal day 2 (P2). Due to the lack of significant lateral migration of cells in the retina, electroporated and non-electroporated regions are generated. Non-electroporated regions may serve as internal histological controls where appropriate.
Retinal electroporation can be used to express a gene under a ubiquitous promoter, such as CAG, or to disrupt gene function using shRNA constructs or Cre-recombinase. More targeted expression can be achieved by designing constructs with cell specific gene promoters. Visualization of electroporated cells is achieved using bicistronic constructs expressing GFP or by co-electroporating a GFP expression construct. Furthermore, multiple constructs may be electroporated for the study of combinatorial gene effects or simultaneous gain and loss of function of different genes. Retinal electroporation may also be utilized for the analysis of genomic cis-regulatory elements by generating appropriate expression constructs and deletion mutants. Such experiments can be used to identify cis-regulatory regions sufficient or required for cell specific gene expression 11. Potential experiments are limited only by construct availability.
The DNA concentration required for electroporation is 5μg/μl. This typically requires the desired plasmids to be amplified using a Maxi-prep (Qiagen) or equivalent method followed by a purification and concentration of the DNA to the working amount. The following steps describe the preparation of DNA to the working amount.
The following steps are performed with the assistance of a stereomicroscope. Once practiced the process of eye opening, incision, and injection takes less than 1.5 minutes which is not enough time for a properly anesthetized pup to recover. Using a sharp 30-gauge needle, carefully open the eye by cutting along the fused junctional epithelium (Fig. 1C). Do not apply excessive force as this may result in cutting of the underlying eye. Avoid cutting beyond the range of the eyelid junction as this will result in bleeding which can obscure the injection.
Examples of P0 neonatal retinas electroporated with a pCAG-EGFP expression plasmid are presented at 3 days following electroporation (P3) and 14 days following electroporation (P14) in figure 4. The area of retinal tissue electroporated is variable from experiment to experiment depending on the extent of tissue damage incurred during the injection and evenness of the spread of the DNA solution into the subretinal space. Typically 90–100% of electroporated retinas will feature cells that have successfully incorporated and expressed the introduced plasmid. However, when factoring in tissue morphology and electroporation efficiency only 40–60 % of electroporated retinas will be suitable for comparative analysis. Rosette formation and retinal detachment at the site of needle penetration is nearly always observed and this region should not be used for analysis. Successful DNA spread during the microinjection results in efficiently electroporated tissue lateral to the site of needle penetration free from rosettes and detachment. Gliosis is also an important experimental concern and nearly always occurs at the site of needle penetration. However, as with rosette formation and retinal detachment, gliosis is typically not significant in laterally electroporated tissue. Immunohistochemical markers such as glial fibrillary acidic protein (GFAP) can be utilized to determine whether levels of reactive gliosis in analyzed regions fall within acceptable thresholds. Sectioned retinas demonstrating that the morphology of the electroporated retina remains intact and that EGFP expression labels individual electroporated cells are presented. Examples 3 days following a P0 electroporation (Fig. 4A–C) and 14 days following a P0 electroporation (Fig. 4D–F) are presented.
In P3 retinas the majority of electroporated cells are located in the neuroblastic layer (NBL) of the retina, and do not demonstrate distinct morphological characteristics of any of the differentiated neural or glial cells of the retina (Fig. 4B, C). By P14 electroporated cells can be found in the outer nuclear layer (ONL) and the inner nuclear layer (INL) of the retina. Furthermore the various labeled cells now display characteristic morphological hallmarks of differentiated neurons including interneurons of the INL, photoreceptors, and Müller glia (Fig. 4E, F).
In vivo electroporation represents a rapid and efficient method for the transformation of retinal cells with DNA expression plasmids. This method allows the experimenter to perform gain of function studies by ectopically introducing a gene of interest under the control of a ubiquitously expressed promoter or to perform loss of function studies by using shRNA constructs targeting genes of interest. Furthermore, multiple DNA plasmids can be electroporated simultaneously, allowing the experimenter to analyze multiple gene effects or to knock down one gene while introducing a second gene of interest. Electroporation of plasmids driving expression of Cre recombinase can also be utilized to eliminate gene expression in a mosaic fashion where animals containing a floxed gene of interest are available. This method is particularly valuable for the determination of intrinsic vs. extrinsic gene function. Conversely electroporation of a gene back into the retinas of conventional knockout mice allows the experimenter to perform rescue studies attempting to reverse phenotypes identified in knockouts. Mice of any background strain can be utilized for electroporation subject to the requirements of the study being performed. However, it is best, where possible, to use strains which demonstrate strong maternal behavior, i.e. Swiss Webster or CD1, in order to minimize the possibility of rejection of electroporated pups. Albino laboratory strains are easier to work with since the spread of the DNA solution during microinjection can be easily visualized. Strains with darkly pigmented eyes such as C57BL/6J are more challenging since the Fast Green tracer can not be discriminated against the pigment of the retinal pigmented epithelium.
The number of cells electroporated is extensive however the ability to electroporate progenitor populations competent to generate the earliest-born cell types is limited. Late-born cells including photoreceptors, bipolar cells, Müller glia, and amacrine cells are easily electroporated using this method. Conversely, early-born cells including retinal ganglion cells, horizontal cells, and subsets of amacrine cells are typically not electroporated using this method. A modification to this protocol in which whole embryonic retinas are dissected, electroporated in vitro and then cultured as explants can be utilized to identify genes regulating specification of early-born cell classes7.
This work was funded by NIH R01EY020560-01 and by a W.M. Keck Young Scholar in Medical Research Award. The authors would like to thank Joseph Bedont for his assistance during the imaging of retinal preparations and injections.
No conflicts of interest declared.