Recent improvements in organotypic brain slice preparations have permitted their exploitation for biotechnological applications. Organotypic slices maintain local structural characteristics of in vivo biology, including functional synaptic connections. Here we present a regioselective biolistic delivery method to label and genetically manipulate these slices.
Transfection of DNA has been invaluable for biological sciences and with recent advances to organotypic brain slice preparations, the effect of various heterologous genes could thus be investigated easily while maintaining many aspects of in vivo biology. There has been increasing interest to transfect terminally differentiated neurons for which conventional transfection methods have been fraught with difficulties such as low yields and significant losses in viability. Biolistic transfection can circumvent many of these difficulties yet only recently has this technique been modified so that it is amenable for use in mammalian tissues.
New modifications to the accelerator chamber have enhanced the gene gun's firing accuracy and increased its depths of penetration while also allowing the use of lower gas pressure (50 psi) without loss of transfection efficiency as well as permitting a focused regioselective spread of the particles to within 3 mm. In addition, this technique is straight forward and faster to perform than tedious microinjections. Both transient and stable expression are possible with nanoparticle bombardment where episomal expression can be detected within 24 hr and the cell survival was shown to be better than, or at least equal to, conventional methods. This technique has however one crucial advantage: it permits the transfection to be localized within a single restrained radius thus enabling the user to anatomically isolate the heterologous gene's effects. Here we present an in-depth protocol to prepare viable adult organotypic slices and submit them to regioselective transfection using an improved gene gun.
Originally the biolistic technique, a turn-of-phrase for “biological ballistics”, was established for particle-mediated gene transfer into plant cells1. This physical method of cell transformation accelerates micro- or nanoparticles at high velocity to overcome the physical barriers of the impermeable cell membranes in order to deliver cargos such as DNA or dyes. Because it does not depend on specific ligand-receptors and/or the biochemical properties at the cell surface membranes, particle-mediated gene transfer can be readily applied to a variety of biological systems such as organotypic brain slices.
Using organotypic slices have advantages over other in vitro platforms since they maintain many anatomical and biochemical properties that are pertinent to in vivo biology2-4. The slices mostly conserve the local architectural characteristics from where they have originated and preserve neurochemical activity and connectivity of the synapses. The use of brain slices for basic research, and in pharmaceutical endeavors, has concomitantly increased with the number of possible biotechnological manipulations to measure and monitor the neurobiological behaviors of the brain in an in vivo like context3,5-7. The major advantages for using organotypic slice-based assay systems is that it provides easy experimental control and allows precise manipulations of extracellular environments.
Fruitfully, organotypic slice culture systems have been established from a variety of brain regions such as, but not restricted to, the cortex, spinal cord, and cerebellum8-10. Furthermore, a number of cocultures have been demonstrated, which allow the assessment of intercellular communication across distal brain regions as well as between neurons and pathological cells11,12. Many protocols have already been established to successfully culture organotypic slices and can maintain long-term viability and many recent studies now utilize the membrane interface methods and various modifications to it13. This principle maintains the organotypic slices at the interface between the medium and the incubator's humidified atmosphere by placing the slices on a porous membrane filter. The medium can thus provide sufficient nutrients to feed the organotypic slices via capillary motion. Typically slices have been prepared from early postnatal animals (3 – 9 days old; P3 – 9). However, brain tissues from these slices display a high level of cellular plasticity and have an inherent resistance to mechanical stresses, which is helpful to obtain viable cultures, yet mature synapses and neuroanatomical circuitry have not fully developed in vivo until 2 to 3 weeks of age14. For example, previous observations had shown that hippocampal slices obtained from P0 – 1 neonates, although highly viable following preparation, gradually lost some morphological characteristics. Essentially, they were shown to be unsuitable for long-term cultures suggesting immature cells were more likely to de-differentiate compared to organotypic cultures from older animals15,16. For this reason our method has been optimized for adult organotypic brain slices at which maturation and architectural development have reached their terminal stages13,17-21. Nevertheless, this method is also suitable for neonate and juvenile organotypic slices.
Once the viable organotypic slices have been produced the entire plate containing the slices can be brought to the biolistic mount and submitted to regioselective delivery and transfection. Proper mounting of the gene gun (as described in Figure 1), oriented 90° at a distance of 10 mm directly over the slices (from the aperture to the tissue), permits the rapid biolistic delivery of the 40 nm gold particle coated cargos. These cargos such as dyes and fluorescent DNA vectors, as well as any gene of interest, are readily delivered into the viable slices with ease. This method thus describes the protocol necessary to deliver, in a regioselective manner, cargo coated gold nanoparticles into viable organotypic brain slices. The gene gun barrel and optimal mounting can be seen in Figure 1. For further information regarding the improved barrel and nanoparticle ballistics, please see O'Brien, et al.17.
Subsequent refinements are also indicated for the user to optimize conditions such as the proper amount of gold delivered per target area, defined as Microcarrier Loading Quantity (MLQ), and to determine the amount of DNA loaded per mg of gold, defined as DNA Loading Ratio (DLR). Prior to precipitating DNA onto gold particles and loading them into the Tefzel tubing that comprises the actual gene gun 'cartridges', it is necessary to calculate the amount of DNA and gold required for each transfection which can differ slightly among tissues and conditions (ratio should be maintained between 1 mg gold and 1 µg DNA to 1:5). It is vital to prepare the proper proportion of MLQ to DLR as otherwise DNA coated gold particles could adhere together forming larger than expected agglomeration, which could reduce the overall homogeneity of the biolistic spread as well as increase tissue damage and cytotoxicity.
This method shows the recent improvements in biolistic delivery, which have been tested and determined to be amenable for neonatal, juvenile and adult organotypic brain slices. Furthermore, by exploiting the gene gun barrel's reduced biolistic spread, the user is now able to selectively transfect a punctual region in the brain. Following the appropriate incubation time, the expression of the fluorescent proteins, which reaches its maximum rates between 24 and 48 hr, can be visualized by epifluorescence and confocal microscopy. These fluorophores permit the morphological analysis and localization of individual cells within biologically relevant structures. However, the regioselective modulation of the organotypic slices by other heterologous genes can also be monitored by any other tests amenable to these preparations. Possible working strategies for the localized gene delivery are presented in Figure 2.
The use of animals and animal tissues should strictly adhere to ethical committee approval under local rules and regulation. All tissues obtained during this study adhered to the MRC-LMB's animal experimentation guidelines.
1. Preparation of Materials & Culture Media
2. Preparation of Organotypic Slices
3. Preparation of DNA-coated Micro & Nano-projectiles
4. Biolistic Transfection on Organotypic Slices
5. Fixing and Visualization of Brain Slices
6. Confocal Applications
Organotypic slice viability can be monitored with lactate dehydrogenase activity, propidium iodide labeling, and dUTP staining as was previously reported13,22. Evidently, the viability and integrity of the slices are important for the long-term sustained expression of the delivered genetic cargo. Following biolistic delivery into the organotypic slices, the expression of the fluorescent heterologous gene was monitored by confocal microscopy. The tightness of the biolistic spread using the modified barrel can be seen in Figure 3A. Figure 3B shows a representative image of the transfected hippocampus region using DNA coated gold nanoparticles. Figure 4 shows representative pictures of transfected adult mouse organotypic slices. As can be seen in Figure 4A, a Purkinje neuron with a remarkable dendritic harbor found at the interface of the Purkinje layer and the molecular layer of the cerebellum shows pronounced labeling following transient biolistic transfection with pEGFP-N1. Figure 4B shows a higher magnification of these dendrites where spines can be observed. Figure 4C shows a pyramidal cell found in the CA1 region of the hippocampus following biolistic delivery of pDSredFP coated gold nanoparticles.
Dyes instead of fluorescent encoded DNA cargos can also be used in a similar manner to visualize morphological characteristics in organotypic slices. DiO, which labels the plasma membranes, can more rapidly delineate individual cell morphologies or can be used in conjunction with a DNA coated biolistic delivered to the same region to confirm the regioselection. As can be seen in Figure 5A a neuron stained with DiO in the hippocampus also shows long dendrites with numerous spines. An arrow indicates the axon which is identified by the presence of synaptic boutons in comparison to the dentritic spines on the other neurites. Figure 5B shows another example of the CA1, which highlights numerous hippocampal projections with a DAPI counter stain to label the nucleus. Figure 5C shows a higher magnification of these types of parallel dendrites with numerous spines.
Evidently, these images illustrate morphological characteristics following cellular labeling. Nothing precludes the same methodology from being applied to deliver any number of exogenous genes coated onto the gold nanoparticles. To permit a visual confirmation of successful transfection, a single plasmid can be constructed to co-express both the gene of interest and a fluorescent reporter protein on the same vector.
Figure 1. Gene gun barrel and mounting. (A) The improved barrel developed by Dr. O'Brien at the MRC-LMB that has a restricted biolistic spread that minimizes tissue damage by lowering the required pressure for particle acceleration. (B) The side view of an optimized gene gun mounting developed by Dr. Henderson and Dr. Andras Nagy at the University of Toronto. The mounting holds the gene gun at precisely 90˚ perpendicular to the organotypic slices while permitting the user to accurately control the distance of biolistic delivery. For optimal biolistic conditions the barrel aperture should be placed directly over the region of interest at a 10 mm distance from the organotypic slices. (C) Back view of the gene gun mounting. The region of interest to be transfected needs to be placed directly underneath the barrel aperture. Targeting can be confirmed with dye (diolistic) or fluorescent protein tranfection (biolistic). Please click here to view a larger version of this figure.
Figure 2. Organotypic slice preparation and regioselective targeting. (A) A DMEM embedded adult mouse brain ready for slicing. (B) The mould was glued onto the vibroslicer mounting with the blade ready to cut a slice. (C) Regioselective heterologous gene expression strategies for biolistic delivery into coronal organotypic brain slices. Images prepared from pictures obtained from the Allen Brain Atlas anatomical reference atlas23. Left panel shows two different brain regions targeted as an example. To restrict the gene expression (1) to the cortical area of the left hemisphere. To restrict the heterologous gene expression (2) to the hypothalamus region of the brain. Middle panel shows that the expression of two heterologous genes (3 and 4) can be isolated and compares the hippocampus of two hemispheres in the same organotypic slice, essentially eliminating any bias from using different sections and/or granting the possibility to address distal crosstalk from the effects of those genes. Right panel shows the possible overlap of transgenic delivery to express the first gene (5) in a precise cortical region while expressing another gene (6) in a disparate yet overlapping cortical area. This permits to analyze the effect of genetic modulation of each heterologous gene alone and also in combination within the same slice. Please click here to view a larger version of this figure.
Figure 3. Biolistic scatter pattern seen on filter paper and low magnification image. (A) The improved barrel was fired onto filter paper to determine the spread of gold particles. Left side shows the distribution of the improved barrel while the right panel shows the normal barrel. Black bar: 1 cm. Image taken from O'Brien, et al.17. (B) Low magnification image showing the stochastic transfection patter within the 3 mm biolistic spread within the hippocampal region of 6 week old mice. White bar: 100 µm. Please click here to view a larger version of this figure.
Figure 4. Biolistic delivery of fluorescent protein encoded DNA into organotypic slices of 6 week old mice. (A) Inverted confocal images of a fixed organotypic slice of the cerebellum of a pEGFP-N1 coated gold particles transfected Purkinje cell. Image was taken at 40X magnification. Black bar: 30 µm. (B) Shows a 60x magnification of another Purkinje cell's dentritic harbor to highlight the spines. Black bar: 5 µm. (C) Shows a confocal image of fixed slices taken from the hippocampal region at 60x magnification of four DSredFP transfected pyramidal cells. Black bar: 10 µm. Please click here to view a larger version of this figure.
Figure 5. Diolistic delivery of membrane labeling fluorescent dyes into organotypic slices. (A) shows the live imaging of a hippocampal neuron labeled with DiO under 20X magnification. The dentritic spines can clearly be observed as well as the axon as identified by the absence of dentritic spines and the presence of synaptic boutons indicated by a white arrow. White bar: 30 µm. (B) Pronounced staining of the hippocampal region labeled with DiO and counterstained with DAPI. White bar: 30 µm. (C) Higher magnification (60x) of the dentritic spines found in the CA1 region of the hippocampus. White bar: 5 µm. Please click here to view a larger version of this figure.
The protocol describes approaches to deliver dyes and genetic materials into adult organotypic slices by using an enhanced gene gun. Essentially, the types of dyes and the variety of possible genes make this method multifaceted and amenable to answer a wide range of biological questions. The regioselective delivery method used, in this case to specific brain region, opens new avenues for experimentation as heterologous gene modulation of localized areas within a biologically relevant architecture hasn't been easily feasible before. We have previously determined that this method can be used on adult organotypic slices, where mature synapses are fully formed, as it showed improved cell survival and heterologous gene expression for many weeks13. With the advent of new and improved culture conditions for mammalian brains as well as other tissues, the use of regioselective genetic modulations will become increasingly useful for a wide variety of biochemical analyses.
As we shall highlight here and previously mentioned by others24, many different factors could readily affect the accuracy and reliability of biolistic delivery. Firstly, the preparation of the gold particle linked to dyes or DNA must be adequately balanced to prevent aggregation and to facilitate the cartridge cargo expulsion. Calcium and spermidine in the DNA-gold micro particles solution can aid the binding of DNA molecules to the gold nanoparticles. The ability for the nanoparticles to penetrate the cells under reduced pressure was previously determined17. Secondly, the gene gun must be mounted as steady as possible with the correct angle and distance. A reliable pressure gauge is also necessary for reproducibility, to maintain tissue integrity, and to properly accelerate the gold particles to their intended targets. Indeed, the distance, orientation, and stability of the equipment are also a crucial factors for the targeting and precision of regioselective delivery. As these brain structures are very small, minor variations in the firing angle could easily render the entire procedure less reliable. And finally, the preparation and maintenance of viable organotypic slices have been fraught with difficulties for many decades yet abundant and reliable protocols for different animals, brain regions, ages and cocultures are currently available2,4,25,26. These procedures should be user-optimized before submitting the tissues to biolistic procedures. This will more readily permit the user to unbiasedly ascertain the impact of the biolistic particles and the effect of the heterologous gene on their own cultures. To this end, the use of fluorescent genes such as EYFP and EGFP can both permit a novel user to test the targeting precision and also follow the cellular viability through heterologous gene expression over a sustained period of time13. Many of the critical steps for reliable and reproducible use of this protocol are highlighted in the following three paragraphs.
For efficient transfection, the DNA concentration must be appropriate. Concentrations that are too low would hinder the transfection yield while concentrations that are too high can cause agglomeration of the nanoparticles. Aggregates of gold can dramatically reduce transfection efficiency, cause uneven distributions, and can lead to increased cytotoxicity and oxidative stress. Problems with coating efficiency are likely due to the expiration of the spermidine solution (should be replaced every 2 to 3 months). For an adequate biolistic spread, the labeling must be even. A non-homogeneous labeling of the gold nanoparticle/DNA suspension may be due to the PVP solution. It should also be replaced every 2 to 3 months. Coating of the gold nanoparticles can be seen on agarose gel electrophoresis as a higher MW band27.
Each biological system under investigation as well as each different instrument used might require small changes in gas pressure to reach optimized levels of transfection and biolistic spread. The critical parameter is the pressure of the helium pulse required to strip the micro- or nano-carriers from the plastic cartridge and propel them into the tissue. A high gas pressure will affect cell survival, because it will create shock waves across the target and disrupt or detach the tissue from the matrix17. Aiming the gene gun barrel and having the apparatus exactly perpendicular to the tissue is important for accurate biolistic delivery. The selected area should be placed in the direct center of the barrel at precisely 10 mm from the outer aperture. This results in a 3 mm diameter of gold particle delivery around the epicenter. The gene gun barrel should be cleaned with 70% ethanol after each use. To protect the tissue from large particle aggregates, the barrel also contains a nylon mesh that should be replaced regularly to maintain optimum performance. In order to replace the mesh unscrew the cap then insert a new nylon mesh between the cap and the O-ring.
Fluorescently labeled cells should be visible at least 1 – 2 days after transfection around the epicenter. Absence or misalignment of labeling can result from inadequate preparation and instability of the gene gun mounting. The observations of cells using confocal microscopy depends on a number of factors: the wavelength of the excitation/emission light, pinhole size, NA of the objective lens, refractive index of components in the light path, depth of the tissue, and the alignment of the instrument. These factors should be optimized for each system under investigation.
Recent advances in biolistics were exploited for the success of this method. The gene gun barrel used in this study was modified in order to reduce the pressure necessary as well as funnel the biolistic scatter pattern into a more constrained and specific area17,19. Other barrel modifications have tried to restrict the biolistic spread and reduce the outflow pressure28 yet this custom designed gene gun barrel designed by Dr. O'Brien is a single component fitted to the standard Helios Gene gun. Subsequently, sub-micrometer gold particles were used to reduce the invasiveness of micrometer particles that we had previously determined caused cellular damage and ultimately, loss of sustained heterologous gene expression. These changes to the biolistic procedure improved the overall feasibility and reproducibility of the method13. Other improvements on slice preparation such as in-depth troubleshooting of the slicing procedure, using a DMEM-Agarose matrix to preserve the brain, and numerous other useful advances in organotypic brain slice preparation previously reported13 enabled us to explore the use of adult slices that have developed mature synaptic networks.
In our study we mainly used dyes and fluorescent heterologous genes to visualize cell morphology as a readout on transfection efficiency and an ability to image the morphological characteristics of neurons in mouse brain tissues. The signal to noise ratio of the stochastic delivery within a defined radius enabled us to completely isolate a single cell's dentritic harbor within its native context. As numerous efforts are used to investigate these morphological characteristics, mapping 'connectomics' or to label single cells for various analyses, this method could easily be combined with existing protocols such as electrophysiology and neurite measurements29,30. Evidently, combinations of fluorescent heterologous genes could enable a much more robust delineation of single cells as the stochastic distribution of fluorescent proteins, and their combination would determine the color of the individual cells31,32. The use of cell specific promoters such as synapsin and glial fibrillary acid protein upstream of the heterologous gene exon can also restrict the expression to subpopulations33. Indeed, the unending variety of genetic material could also be delivered in a regioselective manner by gold nanoparticles using this same procedure. The total transfected regions could thus be analyzed with traditional methods, such as dissection and submitting that region to western immunoblotting to analyze the localized effect of the heterologous gene.
Improvements in organotypic slice procedures will also permit wider use of brain tissues for long-term experimentation as well as permit the use of other tissues and cocultures and would facilitate transfection procedures. Lipid and polymer transfection have been previously reported to affect membrane homeostasis, internalization, protein permeability34 and frequently show very low efficiency to transfect terminally differentiated neurons. Cell specific cargo targeting is also amenable only to cells that express the cell surface receptors needed for internalization, and although this method can be highly selective, it can lack targeted regioselectivity35. Viral vectors otherwise are often difficult and costly to produce, require stringent regulations and have on occasion demonstrated safety concerns. Biolistic delivery thus has an unparalleled ability to regioselectively transfect neurons and other cell types within viable tissues. As gold is non toxic, further advances in the use of biolistics could pave the way for biomedical uses such as transdermal pharmaceutical delivery, noninvasive in vivo gene delivery, as well as localized nonviral gene therapies.
The authors have nothing to disclose.
The authors would like to thank the MRC-LMB workshop for the production of the enhanced gene gun barrel. We would also like to thank Dr. David R. Hampson at the University of Toronto for the use of equipment and resources.
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
Helios gene gun system | Bio Rad | 165-2431 | Gene gun and tube prep station |
HEPES | Sigma | 83264-100ML-F | |
NaCl | Sigma | S7653-250G | |
KCl | Sigma | P9333-500G | |
Na2HPO4 | Sigma | S7907-100G | |
KH2PO4 | Sigma | P9791-100G | |
0.2 µm sterile filter unit | Millipore | SCGPU05RE | to filter media |
DMEM | Gibco | 21885-025 | |
Fetal calf serum | Gibco | 10270-098 | |
D-Glucose | Sigma | G8270-100G | |
N2 | Life Technologies | 17502-048 | |
Penicillin/Streptomycin | Sigma | p4333 | |
Polyvinylpyrrolidone | Sigma | 856568-100G | |
Ethanol | Sigma | 277649-100ML | |
Spermidine | Sigma | S2626 | |
Agarose | Bio Rad | 161-3100EDU | |
Vibroslicer | Laica | VT1200 | We used a custom design |
Gene gun mount | Built in house at U of T | ||
Humidified cell culture incubator | |||
Gold nanoparticles | cytodiagnostics | G-40-20 | |
pEYFP-N1 | Clontech | ||
Tubing cutter | Bio Rad | 165-2422 | |
Tefzel tubing | Bio Rad | 165-2441 | |
Barrel | Custom made by the LMB | ||
Cell culture insert | Millipore | PICM0RG50 | |
Paraformaldehyde | Sigma | 76240 Fluka | |
Dissecting microscope | for convenience | ||
Confocal microscope | user defined for usage | ||
Glass slide | VWR | 2588-CA48323-185 | |
Microcover glass | VWR | 2448-CA48366-067-1 | |
Vectashield mounting media | Vector laboratories | H-1400 |