Biolistic transfection and expression analysis of acute cortical slices

Abstract

Background: Biolistic gene gun transfection has been used to transfect organotypic cultures (OTCs) or dissociated cultures in vitro. Here, we modified this technique to allow successful transfection of acute brain slices, followed by measurement of neuronal activity within a few hours.

New method: We established biolistic transfection of murine acute cortical slices to measure calcium signals. Acute slices are mounted on plasma/thrombin coagulate and transfected with a calcium sensor. Imaging can be performed within 4 h post transfection without affecting cell viability.

Results: Four hours after GCaMP6s transfection, acute slices display remarkable fluorescent protein expression level allowing to study spontaneous activity and receptor pharmacology. While optimal gas pressure (150 psi) and gold particle size used (1 μm) confirm previously published protocols, the amount of 5 μg DNA was found to be optimal for particle coating.

Comparison with existing methods: The major advantage of this technique is the rapid disposition of acute slices for calcium imaging. No transgenic GECI expressing animals or OTC for long periods are required. In acute slices, network interaction and connectivity are preserved. The method allows to obtain physiological readouts within 4 h, before functional tissue modifications might come into effect. Limitations of this technique are random transfection, low expression efficiency when using specific promotors, and preclusion or genetic manipulations that require a prolonged time before physiological changes become measurable, such as expression of recombinant proteins that require transport to distant subcellular localizations.

Conclusion: The method is optimal for short-time investigation of calcium signals in acute slices.

Keywords: Acute slices; Calcium imaging; Gene-gun transfection; Genetically encoded calcium indicators; Neocortex.

Figure 1.. Experimental procedure.

(A) Shortly after decapitation, the cortex was removed and placed on the chopper plate. The cortex was cut into 350 μm thick slices and the slices were transferred into GBSS solution (in the presence of glutamate receptors blockers) and allowed to recover for 30 minutes (B). (C) After recovery, to immobilize the free-floating slices, the slices were mounted on a coverslip containing a mixture of (1:2) plasma and GBSS/thrombin and allowed to coagulate for 15 minutes. (D) The slices were then transferred into roller tubes filled with 750 μl oxygenated ACSF containing the glutamate receptor blockers, for 15 minutes. (E) The slices were transferred into the gene-gun blasting setup (The detailed description of this accessory setup can be found in (Wirth and Wahle, 2003)). The acute slices were then transfected with the desired plasmid. The slices were brought back into the roller tubes and kept in a roller incubator at 37°C (F). The blockers were removed 30 min after transfection and the ACSF solution was repeatedly exchanged every hour. (G) Four hours post transfection, acute slices were transferred into the recording chamber and calcium activity was recorded. See Material and Methods section for detailed description.

Figure 2.. Effect of gene-gun tissue bombardment on cell survival.

(A–F) Confocal images (10× magnification) of representative acute slices stained with propidium iodide (PI). (A) PI staining of a non-transfected acute slice. (B) PI staining of a transfected acute slice. (C-E) PI staining of acute slices challenged with glutamate receptor agonists: 50 μM NMDA (C), 100 μM AMPA (D), and 25 μM kainate (E) for 15 min and stained with PI. (F) PI staining of a non-transfected acute control slice stimulated with vehicle (H₂O). Scale bars: 30 μm. (H) The graph indicates the relative change in the number of stained PI cells (mean ± S.E.M.) in a selected ROI compared to non-transfected acute control slices. ***P < 0.001 indicates significant differences between treated acute slices and non-transfected acute control slices. One-way ANOVA followed by Holm-Sidak Multiple Comparison Test. The number of acute slices analysed is indicated above the graph in (H). The data is obtained from 2 independent acute slice preparations.

Fig. 3.. Live-imaging of EGFP expression in acute slices.

Confocal laser microscope images exemplifying the time course of EGFP fluorescence intensity after transfection (A-F). Note that the dendritic processes can be visualized 4 hours post-transfection. (G) The graph indicates average fluorescence pixel intensity (mean ± S.E.M.) from different neurons selected at different time points (20 neurons/time point). One-way ANOVA followed by Holm-Sidak Multiple Comparison Test, ***p < 0.001, **p < 0.01 and *p < 0.05. Scale bars: 30 μm. The data is obtained from 2 independent acute slice preparations.

Fig. 4.. Biolistic transfection in acute slices parameters.

Acute slices were transfected with EGFP and subsequently incubated for 4 hours to allow for EGFP expression. Six arbitrary areas of interest (AOI) of 1 mm² were selected for cell counting from each acute slice. The number of acute slices analysed is indicated within the graphs in (A-C). (A) The graph indicates the number of transfected cells per mm² (mean ± S.E.M.) at different applied gas pressures. The maximal transfection efficiency is achieved when acute slices were blasted at 150 psi. (B) The graph indicates the number of transfected cells per mm² (mean ± S.E.M.) for different gold particle sizes which vary from 0.5 μm to 5 μm. (C) The graph indicates the number of transfected cells per mm² (mean ± S.E.M.) using different amounts of DNA plasmid for coating. One-way ANOVA on Ranks followed Tukey’s Test, ***p < 0.001 and **p < 0.01. The data is obtained from 2 independent acute slice preparations. (D) Single acute slice detail analysis showing which cell types underwent successful transfection in percentage (12 acute slices used, two independent preparations). (E-H) Examples of confocal laser microscope transfectant images at 20x magnification. (E) A pyramidal cell example and, in the zoomed area in (F), dendritic spines are shown (red arrows). (G) An example of an interneuron (red arrow) and a group of glial cells are shown in (H). Scale bars in E, G and H: 30 μm. Scale bar in F: 10 μm.

Fig. 5.. Recording of network activity in GECI-transfected acute slices.

(A and B) Examples of spinning-disc confocal images of a pyramidal cell at 20× magnification exemplifying the change of GCaMP6s fluorescence intensity during resting activity (F⁰) and peak amplitude during spontaneous activity. Scale bar: 30 μm. The ΔF/F⁰ trace for this cell is shown in (C). The arrows indicate F⁰ during resting activity and any peak amplitude during spontaneous activity. (D) The graph shows the average increase in calcium signal amplitude in control (spontaneously active) and after bath application of 50 μM APV expressed as ΔF/F⁰ (mean ± S.E.M.) for Reln^wt and Reln^cKO acute slices and in (E) shows the frequency of calcium events per 3 minutes. (F) The graph shows the calcium signal amplitude expressed as ΔF/F⁰ (mean ± S.E.M.) in control (spontaneously active) and after bath application of 10 μM NBQX for Reln^wt and Reln^cKO acute slices and (G) the frequency of calcium events per 3 minutes. (H) The graph shows the calcium signal amplitude expressed as ΔF/F⁰ (mean ± S.E.M.) in control (spontaneously active) and after bath application of 10 μM muscimol for Reln^wt and Reln^cKO acute slices and (I) the frequency of calcium events per 3 minutes. One-way ANOVA followed by Holm-Sidak Multiple Comparison Test, ***p < 0.001. The number of recorded cells is indicated above the graphs. The data is obtained from 2 independent acute slice preparations.